ENVtRONMENTAL SURVEiLLANCE AROUND NUCLEAR )NS1ALLAT)0NS
ENVIRONM ENTAL SURVEILLANCE AROUND NUCLEAR INSTALLATIONS
VOL. I The following States are Members of the International Atomic Energy Agency:
AFGHANISTAN HAITI PAKISTAN ALBANIA HOLY SEE PANAMA ALGERIA HUNGARY PARAGUAY ARGENTINA ICELAND PERU AUSTRALIA INDIA PHILIPPINES AUSTRIA INDONESIA POLAND BANGLADESH IRAN PORTUGAL BELGIUM IRAQ ROMANIA BOLIVIA IRELAND SAUDI ARABIA BRAZIL ISRAEL SENEGAL BULGARIA ITALY SIERRA LEONE BURMA IVORY COAST SINGAPORE BYELORUSSIAN SOVIET JAMAICA SOUTH AFRICA SOCIALIST REPUBLIC JAPAN SPAIN CAMEROON JORDAN SRI LANKA CANADA KENYA SUDAN CHILE KHMER REPUBLIC SWEDEN COLOMBIA KOREA, REPUBLIC OF SWITZERLAND COSTA RICA KUWAIT SYRIAN ARAB REPUBLIC CUBA LEBANON THAILAND CYPRUS LIBERIA TUNISIA CZECHOSLOVAK SOCIALIST LIBYAN ARAB REPUBLIC TURKEY REPUBLIC LIECHTENSTEIN UGANDA DENMARK LUXEMBOURG UKRAINIAN SOVIET SOCIALIST DOMINICAN REPUBLIC MADAGASCAR REPUBLIC ECUADOR MALAYSIA UNION OF SOVIET SOCIALIST EGYPT, ARAB REPUBLIC OF MALI REPUBLICS EL SALVADOR MEXICO UNITED KINGDOM OF GREAT ETHIOPIA MONACO BRITAIN AND NORTHERN FINLAND MONGOLIA IRELAND FRANCE MOROCCO UNITED STATES OF AMERICA GABON NETHERLANDS URUGUAY GERMAN DEMOCRATIC REPUBLIC NEW ZEALAND VENEZUELA GERMANY, FEDERAL REPUBLIC OF NIGER VIET-NAM GHANA NIGERIA YUGOSLAVIA GREECE NORWAY ZAIRE, REPUBLIC OF GUATEMALA ZAMBIA
The Agency's Statute was approved on 23 October 1956 by the Conference on the Statute of the IAEA held at United Nations Headquarters, New York; it entered into force on 29 July 1957. The Headquarters of the Agency are situated in Vienna. Its principal objective is "to accelerate and enlarge the contribution of atomic energy to peace, health and prosperity throughout the world".
Printed by the IAEA in Austria August 1974 PROCEEDINGS SERIES
ENVIRONM ENTAL SURVEILLANCE
AROUND NUCLEAR INSTALLATIONS
PROCEEDINGS OF A SYMPOSIUM HELD BY THE INTERNATIONAL ATOMIC ENERGY AGENCY IN WARSAW, 5 - 9 NOVEMBER 1973
/и fwo vo/Mfwes
V O L I
INTERNATIONAL ATOMIC ENERGY AGENCY VIENNA, 1974 ENVIRONMENTAL SURVEILLANCE AROUND NUCLEAR INSTALLATIONS IAEA, VIENNA, 1974 STI/PUB/353 FOREWORD
Adequate environmental surveillance is an essential part of the m easures taken by the operators of nuclear installations and by the national competent authorities to ensure that such installations do not produce any pollution of the environment through the release of radioactive contaminants under normal operating conditions, and to provide information on which protective action can be taken in the event of an accidental release. There has been argument and uncertainty about the nature and extent of the environmental surveillance programmes required for different types of nuclear installations; much of this is attributable to a lack of clear definition of the aims of such surveillance. As part of its programme to assist Member States in controlling and minimizing the environmental effects of nuclear programmes, the Inter national Atomic Energy Agency, in co-operation with the Government of Poland, convened a Symposium on Environmental Surveillance Around Nuclear Installations in Warsaw on 5-9 November 1973. It was attended by 200 participants from 2 6 Member States and by representatives of 8 international organizations. Sixty-one papers were presented in eight sessions covering the objectives of environmental surveillance, pre- operational investigations, environmental monitoring procedures innormal and emergency situations, the interpretation of results, research and supportive studies, and examples of the environmental surveillance programmes conducted at specific installations. A small number of the papers dealt with non-radioactive contaminants. In the final session three short papers by invited speakers reviewed the problems arising in the establishment of standards and derived working limits and the operation of adequate environmental surveillance systems for both radioactive and non-radioactive contaminants thatmightbe released to the environment in the nuclear industry. A panel of selected participants then discussed these topics and replied to questions submitted in written form and orally by the other participants. It was emphasized that clear objectives should be set for any environ mental surveillance programme in order to avoid wastage of resources in manpower and equipment. It is obviously more efficient in some cases to rely on the monitoring of gaseous and liquid effluents in order to check the compliance with the authorized lim its on releases to the environment. How ever, as was pointed out by many speakers, it is not sufficient, especially in populated areas, to rely entirely oneffluentmonitoring. Properlydesigned environmental monitoring programmes are also needed to confirm that there are no unsuspected releases, or unsuspected pathways of exposure, and to reassure the public that adequate care is takenfor their protection. Environ mental measurements can also contribute to increased scientific knowledge of the behaviour of radionuclides in the environment. The present book contains all the papers and discussions of the Symposium. The Agency gratefully acknowledges the assistance and co operation of the Polish authorities, which helped greatly towards the success of the meeting. Æ DJTORML NOTÆ
The papers and discussions incorporated in the proceeding's published by the Jniemaiionai atom ic .Energy Agency are edited by tbe Agency's edi- toriai staff io tbe extent considered necessary for the reader's assistance. Tbe views expressed and tbe générai styie adopted remain, however, tbe responsibility of tbe named authors or participants. For tbe saife of speed of pubiication tbe present Proceedings bave been printed by composition typing and photo-offset iithography. Within tbeiimi- tations imposed by this method, every effort bas been made to maintain a bigb editoriai standard; in particuiar. tbe units and symbois empioyed are to tbe fuiiest practicable extent tbose standardized or recommended by tbe competent international scientific bodies. Tbe affiliations of authors are tbose given at tbe time of nomination. Tbe use in tbese Proceedings of particuiar designations of countries or territories does not impiy any judgement by tbe Agency as to the iegai status of sucb countries or territories, of tbeir authorities and institutions or of tbe deiimitation of their boundaries. The mention of specific companies or of their products or brand-names does not impiy any endorsement or recommendation on the part of tbe international atom ic Ænergy Agency. CONTENTS
INTRODUCTORY PAPER
Surveillance de l'environnement des installations nucléaires: L 'h eu re du ré a lism e (IA E A -S M -1 8 0 /7 6 )...... 3 P. Pellerin D isc u ssio n ...... 14
OBJECTIVES OF ENVIRONMENTAL SURVEILLANCE (Sessions I and II)
The objectives and requirements for environmental surveillance at USAEC facilities (IAEA-SM-180/33)...... 19 M .B . Biles, M.W. Tiernan, A. Schoen, C .G . Welty D isc u ssio n ...... 32 Analyse des principes directeurs des programmes de surveillance de l'environnem ent (IA E A -S M -1 8 0 /4 6 )...... 35 G. Bresson, R. Coulon Principles and practice of environmental monitoring in the United Kingdom (IA E A -S M -1 8 0 /8 )...... 47 N .T . Mitchell, H .J. Dunster, A .W . Kenny, E . A. B. Bir se Objectifs de la surveillance d'une installation nucléaire (IA EA -SM -180/5) ...... 61 E . Nagel D iscu ssio n ...... 64 The necessity for environmental surveillance in the evaluation of nuclear power plant site s (IA E A -SM -1 8 0 /2 9 ) ...... 65 P . Handge, F .O . Hoffman D iscu ssio n ...... 85
PRE-OPERATIONAL INVESTIGATIONS (Session II cont. )
Environmental gamma radiation measurements in nuclear power station siting stu d ies in Poland (IA EA -SM -180/22) ...... 89 B. Gwiazdowski, J. Pensko, J. Jagielak, M. Biernacka, K. Mamont-Ciesla D iscu ssio n ...... 105 Study of the background radiation around the construction site of a nuclear power station in Bulgaria with a view to controlling the radiological health of the population in the future (IA E A -SM -180/27)...... 107 G. Vasilev, M. Yotov, D. Keslev D iscu ssio n ...... 110 Movement of radionuclides in the ground in relation to environmental safety of nuclear power plants (IAEA-SM-180/74).. . Ill Z. Dlouhÿ, O. Safâr D iscu ssio n ...... 118 Natural gamma background radiation in the environs of nuclear facilities: Pre-operational analysis of dose to the neighbouring population (IA E A -SM -180/23) ...... 121 J . PeAsko D iscu ssio n ...... 130 О деятельности органов Совета экономической взаимопомощи в области охраны и улучшения окружающей среды по проблеме обеспечения радиационной безопасности (IA E A -S M -1 8 0 /8 1 )...... 131 В .Х а к е D is c u s s io n ...... 135
ENVIRONMENTAL MONITORING PROCEDURES - NORMAL AND EMERGENCY SITUATIONS (Sessions III and IV)
Environmental radioactivity surveillance methods for a nuclear fac ility (IA E A -SM -180/42) ...... 139 J . Sedlet D iscu ssio n ...... 153 Analytical system s applied to monitoring the aquatic environment in the control of radioactive waste disposal (IAEA-SM-180/7) .... 155 J .W .R . Dutton, N .T . Mitchell, E. Reynolds, L. Woolner D iscu ssio n ...... 166 Detailed measurement of ^ 1 in air, vegetation and milk around three operating reactor sites (IAEA-SM-180/44) ...... 169 B .H . Weiss, P .G . Voillequé, J .H . Keller, B. Kahn, H .L . Krieger, A. Martin, C .R . Phillips D iscu ssio n ...... 189 Measurement of low concentrations of radioactive noble gases in effluents and in the atm osphere (IA EA -SM -1 8 0 /6 6 )...... 191 A. Botlino, G. Lembo, F. Scacco, G. Sciocchetti, J. Lasa An analytical study of the environmental survey methods in use around nuclear installations in the countries of the European C o m m u n ity in l9 7 1 (IA E A -SM -180/4)...... 199 P . Recht, J. Smeets, R. Amavis, A .A . Cigna Permanent gas measurements as part of an environmental su rveillan ce program m e (IA EA -SM -180/38) ...... 225 J .M . Matuszek, C .J . Paperiello, C .O . Kunz, J .A . Hutchinson, J .C . Daly D iscu ssio n ...... 233 Сбор проб и определение трития в приземном слое воздуха (IAEA-SM-180/78) ...... 235 Л.И. Гедеонов, В.А. Блинов, A.B. Степанов, В.М. Гаврилов, B.П.Тишков, А.М. Максимова D iscu ssio n ...... 241 Monitoring low-level radioactive aqueous discharges from a nuclear power station in a seawater environment (IA E A -SM -180/41) ...... 243 D .M . Montgomery, H .L . Krieger, В. Kahn D iscu ssio n ...... 256 Practical reference levels for radioactive contamination in environmental surveillance (IAEA-SM-180/57) ...... 257 G. Boeri, C. Brofferio Rapid pore water analysis for sediments adjacent to reactor d isc h arg e s (IA E A -SM -180/37) ...... 269 E .K . Kalil Integral dosimeter study of gamma-ray dose in the vicinity of a nuclear reactor (IAEA-SM-180/26)...... 285 G. Vasilev, G. Filev D iscu ssio n ...... 288 Early surveillance around coastal nuclear installations (IA E A -SM -180/36)...... 289 T .R . Folsom, V .F . Hodge D iscu ssio n ...... 2 99 Emergency surveillance around a nuclear reactor site (IA EA -SM -180/15) ...... 301 G .H . Palmer, K .E .G . Perry D iscu ssio n ...... 310 Intercalibration of methods for radionuclide measurements on a m arin e sedim ent sam ple (IA E A -S M -1 8 0 /9 )...... 313 R. Fukai, G .A . Statham, S. Ballestra, K. Asari Problems in calibrating instruments for environmental gamma exp osure dose m easu rem en ts (IA EA -SM -180/21) ...... 337 J . Jagielak, B. Gwiazdowski, J. Peftsko, A. Zak A new low-background counting facility for instrument calibration (IA EA -SM -180/65) ...... 351 S. A b e D iscu ssio n ...... 358 Rapid determination of radionuclides in milk: Results of an inter comparison organized jointly by the IAEA and CEC (IA EA -SM -180/2) ...... 359 O. Suschny, J. Heinonen, D. Merten, J. Smeets, R. Amavis, A. Bonini D iscu ssio n ...... 392
INTERPRETATION OF RESULTS (Session V) Some considerations of the effects of the accidental release of fissio n products from re a c to rs (IA E A -SM -180/6) ...... 395 J .R . Beattie D iscu ssio n ...... 399 Radiation hazards to the population resulting from conventional and n u clear e le c tric pow er production (IA E A -S M -1 8 0 /2 0 )...... 403 Z . Jaworowski, J. Bilkiewicz, ElibietaDobosz, Danuta Grzybow ska, Ludwika Kownacka, Z. WroAski D iscu ssio n ...... 412 Interpretation of aerosol release measurements at nuclear power p lan ts (IA E A -S M -1 8 0 /3 0 )...... 413 R . Fritze, G. Herrmann D iscu ssio n ...... 422 Application of system analysis methodology to the determination of the limiting radiological capacity of the area surrounding a n u clear fac ility (IA E A -SM -180/60)...... 425 L . Frittelli Radioactive gas and aerosol production by the CERN high energy accelerators and evaluation of their influences on environmental p ro b lem s (IA E A -SM -1 8 0 /10) ...... 433 A. Peetermans, J. Baarli D iscu ssio n ...... 448 Analytical quality control and information value of results in environm ental a n a ly sis (IA EA -SM -180/14) ...... 449 D. Beyer, D. Merten, J. Heinonen Developments in the United Kingdom in the derivation of emergency reference levels in environmental materials (IAEA-SM-180/12) . . . 457 P a m e la M. Bryant D iscu ssio n ...... 467
L is t of C h a irm e n ...... 471 S e c re ta ria t of the S y m p o siu m ...... 471 INTRODUCTORY PAPER
IAEA-SM-180/76
SURVEILLANCE DE L'ENVIRONNEM ENT DES INSTALLATIONS NUCLEAIRES: L'HEURE DU REALISM E
P. PELLERIN Service central de protection contre les rayonnements ionisants, M inistère de la santé publique et de la sécurité sociale,
Le Vésinet, France
Abstract-Résumé
ENVIRONMENTAL SURVEILLANCE AROUND NUCLEARINSTALLATIONS: THE HOUR FOR REALISM. In the last 25 years innumerable studies have been carried out on the various aspects of possible radio active contamination of the environment by nuclear installations. Apart from the Windscale incident (the consequences of which were in any case very minor) no really serious situation has ever been recorded. On the basis of the most pessimistic assumptions, the anticipated development of the nuclear industry up to the year 2020 could perhaps result in an additional dose to the population of the order of a hundredth of that from natural radiation. Despite a general increase in plant capacity, technological developments have increased the safety of nuclear installations to a level rarely equalled in other industries, which makes a serious accident highly improbable. During this period medical radiodiagnosis, despite progress in radio- protection, has continued practically to double natural radiation, and its use is increasing on average by 10% per year in the advanced countries. It would be unreasonable in these circumstances - with the exception of course of purely scientific research- to continue to expend efforts and vast sums of money on further developing surveillance around nuclear installations; it is already fully adequate and should be confined to routine operations which are now well defined for each type of installation. On the other hand, everything necessary should be done to ensure that these operations are carried out effectively and systematically in all cases. A return to realism is therefore essential for the following reasons: (a) so as not to create artificially a more and more paradoxical situation, disquieting to the public, which would be bound ultimately to be exploited against the interests of nuclear power, and (b) so as not to retard the necessary development of nuclear energy, and to avoid unjustified squandering of resources on the protection side.
SURVEILLANCE DE L'ENVIRONNEMENT DESINSTALLATIONS NUCLEAIRES: L'HEURE DU REALISME. Depuis 25 ans, des études innombrables ont été effectuées sur les divers aspects de l'éventuelle contamination radioactive de l'environnement par les installations nucléaires. Si l'on excepte l'incident de Windscale (dont les conséquences furent d'ailleurs très modestes), aucune situation réellement préoccupante n'a jamais pu être mise en évidence. Dans les hypothèses les plus pessimistes, le développement prévu de l'industrie nucléaire, jusqu'en l'année 2020, pourrait peut-être entramer pour la population une irradiation supplémentaire de l'ordre du centième de l'irradiation naturelle. Malgré une augmentation générale de
continue pratiquement à doubler l'irradiation naturelle, et son usage s'accroît en moyenne de 10% par an dans les pays développés. 11 serait déraisonnable, dans ces conditions, et a l'exception bien entendu des études à but purement scientifique, de persister à investir de nouveaux efforts et des crédits considérables pour développer plus encore la surveillance autour des installations nucléaires: celle-ci est très satisfaisante et doit se limiter aux seules opérations de routine désormais bien définies pour chaque type d'installation.
réalisme est donc désormais indispensable: a) pour ne pas créer artificiellement une situation de plus en plus paradoxale, inquiétante pour le public, et qui ne manquerait pas, à terme, d'être exploitée à l'encontre de l'énergie nucléaire; b) pour ne pas freiner le développement nécessaire de l'énergie nucléaire, et éviter un gaspillage de crédits injustifié sur le plan sanitaire.
3 4 PELLERIN
Dans le rapport que le Comité Scientifique des Nations Unies pour les effets des rayonnements ionisants a adressé ä 1'Assemblée Générale en )972, il est dit au chapitre intitulé "récapitulation des doses engagées provenant de l'industrie de l'énergie nucléaire" [1]:
"Aux taux de 1970 et aux taux envisagés pour l'an 2000, une année "de production d'énergie électrique impliquerait des doses à la population "correspondant respectivement à environ cinq minutes, et ä un jour de "l'irradiation par le fond naturel de rayonnement".
C'est-à-dire qu'en l'an 2000, l'irradiation supplémentaire liée aux centrales électro-nucléaires ne représenterait, dans les hypothèses les plus pessimistes, que le quatre centième, environ, de l'irradiation naturelle inévitable.
Dans ce même rapport, le Comité Scientifique estime que la dose aux gonades délivrée en moyenne par le radiodiagnostic médical dans certains pays, dépasse actuellement la moitié de la dose distribuée par le fond natu rel de rayonnement [2]. Cependant, cette dernière évaluation doit être consi dérée comme très optimiste, car elle surestime l'importance de la dose aux gonades au détriment de la dose somatique correspondante, qui doit être prise en considération notamment pour d'éventuels effets cancérogènes. Cette dose somatique venant du radiodiagnostic médical est de plusieurs centaines de millirads par an, donc beaucoup plus élevée que la dose aux gonades [3].
Si l'on ajoute que les évaluations les plus récentes et les plus sérieuses montrent que, dans les pays développés, l'accroissement du nombre des installations de radiodiagnostic médical s'échelonne de quelques pourcents jusqu'à !5% par an (avec une moyenne de l'ordre de )0% par an) [4], l'on voit qu'il n'est pas impossible, dans certaines circonstances, que, dans dix ans, la dose annuelle somatique due au radiodiagnostic médical puisse dépasser six fois la valeur de l'irradiation naturelle. Autrement dit, dans dix ans, si le rythme actuel du développement du radiodiagnostic se poursuit, il n'est pas exclu que, dans certains cas, l'irradiation annuelle résultante puisse attein dre six années d'exposition au fond naturel par an (soit plus de deux mille ans d'exposition due à la production nucléaire d'énergie électrique!).
Dans ces conditions, je pose la question suivante: Est-il raison nable de continuer à persécuter les quelques millirads hypothétiques attribués aux centrales électro-nucléaires de l'an 2000 pendant qu'on laisserait sans contrSle le radiodiagnostic médical distribuer des doses approchant le millier de millirads?
Bien entendu, de nombreuses commissions internationales et organismes nationaux s'efforcent, en bonne coopération avec les médecins utilisateurs et les constructeurs, de ramener l'irradiation médicale au strict nécessaire sans restreindre les informations indispensables que le radiodiagnostic fournit au médecin.
Je crois cependant que le moment est venu, pour les médecins, les radiobiologistes, et les physiciens de santé des installations atomiques de toutes natures, de prendre conscience de la situation paradoxale dans laquelle nous nous trouvons et qui ne peut que s'aggraver si l'on n'y remédie. Cette situation paradoxale a pour effet de provoquer l'inquiétude d'une partie du public, qui en déduit le raisonnement suivant: "Si l'on développe des con trôles aussi sévères et des études de plus en plus coûteuses au sujet de la seule protection de l'environnement des centrales nucléaires, et si l'on fait tant de communications et de colloques à leur sujet, pendant qu'on laisse une IAEA-SM-180/76 5 quasi-totale liberté au diagnostic radiologique médical dont on ne parle pra tiquement jamais, c'est que l'irradiation par l'énergie nucléaire est beaucoup plus dangereuse que celle du radiodiagnostic médical qui distribue pourtant des doses cent à mille fois plus élevées".
Ce raisonnement est évidemment absurde, car il n'y a pas deux sortes d'irradiation, la bonne et la mauvaise. Mais il faut bien reconnaître, comme j'ai eu l'occasion de le rappeler récemment [5], que la prolifération des réunions et des communications en radioprotection nucléaire favorise cette interprétation erronée.
Certains groupes entretiennent, d'autre part, dans l'esprit du public une confusion permanente entre les rejets normaux autorisés et contrS- lés des centrales nucléaires, et les accidents que l'on peut envisager jusque dans les hypothèses les plus invraisemblables. Bien entendu, la description de ces "accidents" est dramatisée au maximum, et elle finit par susciter l'idée absolument fausse que les centrales nucléaires fonctionneraient en permanence à la limite de la catastrophe. D'ailleurs, ici encore, l'inflation verbale et le jargon technique des spécialistes ont dénaturé la situation et involontairement apporté à la contestation des arguments d'inquiétude, notam ment par une exploitation tendancieuse du terme "accident maximum croyable" qui n'a jamais eu, dans l'esprit de ceux qui l'ont proposé, la dimension apocalyptique que certains s'acharnent à lui donner. Les mêmes réserves doivent être formulées à l'égard de l'expression "homme-rad" qui favorise les exploitations les plus tendancieuses, et devrait être abandonnée.
Car, c'est bien en tant qu'hygiéniste que je tiens à affirmer que le développement de l'énergie nucléaire représentera un immense bénéfice pour la santé de l'homme:
a) parce que seul l'accroissement de l'énergie aussi bien dans les pays industrialisés que dans les pays en voie de développement, et sa disponibilité quelles que soient les circonstances, permettra d'élever le niveau sanitaire qui, même dans les pays avancés, est loin d'être satisfaisant;
b) parce que le développement des centrales électro-nucléaires mettra fin à la production d'énergie par les procédés actuels très pol luants, les pollutions radioactives résultantes étant, comme l'a confirmé le Comité Scientifique des Nations Unies, très faibles et sans commune mesure avec les précédentes. Même dans les évaluations les plus pessimistes, l'éventuelle action nocive de ces faibles niveaux de la radioactivité n'a jamais pu dépasser le stade des hypothèses.
Pour mémoire, je rappelle simplement en effet:
- qu'une centrale électrique au charbon de )000 mégawatts rejette chaque jour dans l'atmosphère 250 tonnes de dioxyde de soufre, t20 tonnes d'oxydes d'azote et 20 tonnes de cendres [6];
- que, pour un pays comme la France, en 1968, il a été rejeté dans l'atmosphère un total de ! 500 000 tonnes de dioxyde de soufre, 500 000 tonnes d'oxydes d'azote et 33 000 tonnes de cendres par suite du fonctionnement des foyers domestiques et industriels. On a pu montrer que des pollutions de cet ordre augmentent, entre autres, le taux des bronchites chroniques graves jusqu'à 7X [7]; 6 PELLERIN
- que l'utilisation, jusqu'en l'an 2000, des combustibles fossiles au rythme de croissance actuel déterminerait une augmentation allant jusqu'à 18X de la teneur en gaz carbonique de l'air du globe, et un réchauffement moyen de 0,5° pour l'ensemble du globe [7];
- que, si l'on considère la pollution radioactive naturelle, indépen dante de l'industrie nucléaire, la combustion de certains charbons peut libérer dans l'atmosphère, par gramme de cendre, une radio activité naturelle de 5 à plus de !00 picocuries d'émetteurs alpha (particulièrement radiotoxiques) [8];
- que, dans une centrale électrique classique de puissance moyenne, les combustibles fossiles libèrent de même de l'uranium, du thorium et du radium en quantités suffisantes pour que les popula tions avoisinantes puissent recevoir des doses atteignant, à l'os, jusqu'à 50 millirads par an [9];
- que, si l'on considère la seule irradiation naturelle dans le Massif Central français, dans les Vosges, en Forêt Noire, dans les Alpes Suisses et le Jura, dans certaines régions du Brésil et des Indes, celle-ci passe fréquemment des ]00 millirads relevés au niveau de la mer à plus de 1500 millirads [10], sans que l'on ait pu mettre en évidence de conséquences sur la mortalité et la morbi dité (bien au contraire, certaines populations exposées en perma nence à un rayonnement ionisant naturel d'intensité jusqu'à dix fois plus élevée qu'au niveau de la mer, semblent présenter une exceptionnelle longévité);
- qu'enfin le seul fait de vivre à une altitude plus élevée de 25 mètres (la hauteur d'un immeuble de sept étages) augmente l'irradiation naturelle d'un millirad par an.
On n'en finirait pas d'allonger cette liste d'exemples qui montre, lorsqu'on replace l'irradiation par les rayonnements ionisants à son juste rang dans l'échelle des "dangers" présentés par les pollutions de tous ordres, à quel point la polarisation sur le seul risque nucléaire est absurde.
D'ailleurs pour que les arguments de la contestation de l'énergie nucléaire soient fondés, il faudrait admettre l'existence d'une invraisemblable conspiration à l'échelle mondiale de tous les organismes qui, d'une façon ou d'une autre, en contrôlent le développement: ministères de l'industrie ou de l'équipement, services de santé publique qui les contrSlent, organismes natio naux responsables des programmes d'énergie électrique, commissions internatio nales s'occupant de radioprotection, de dosimétrie, d'effets des rayonnements ionisants, etc.
L'organisation de la lutte contre l'irradiation et la contamination radioactive est bien au contraire certainement l'aspect de l'hygiène publique pour lequel, au cours des vingt dernières années, ont été consentis les plus grands efforts technologiques et financiers. Le résultat en est que la pollution radioactive ne pose aucun réel problème malgré un important développement de l'énergie électro-nucléaire, et que, jusqu'à présent, jamais aucune preuve sérieuse n'a pu être apportée qu'elle ait entraîné la maladie ou la mort d'un seul individu dans ses applications pacifiques, en dehors de quelques accidents internes aux premières installations de recherche, inévitables dans toute indus trie et, dans son cas, particulièrement rares et limités, à tel point qu'elle devrait servir de modèle en matière de sécurité industrielle. IAEA-SM-180/76 7
Le véritable danger serait d'abord que l'opposition à l'énergie nucléaire se généralise et mette alors réellement en cause le développement indispensable de cette source d'énergie particulièrement saine. Il s'agirait lâ, en fait, d'une agression délibérée contre la santé de l'homme pour laquelle l'assainissement nécessaire des sources d'énergie en même temps que leur déve loppement doit entraîner un immense progrès.
Mais les conséquences seraient, à terme, encore plus graves dans la mesure où cette opposition se retournerait finalement contre les buts-mêmes qu'elle prétend atteindre: en effet, bien que la défense de l'environnement présente parfois des aspects excessifs (qui ne sont d'ailleurs le fait que d'une faible minorité dont les motivations sont loin d'être claires), elle est justifiée par la défense de la santé de l'homme, et cette attitude récente est hautement louable. Mais, comme toute orientation nouvelle, elle n'est actuel lement l'objet d'une approbation générale qu'à cause même de sa nouveauté. Il serait, dès lors, très regrettable qu'elle ne se présente que comme une mode passagère, et qu'à force d'en entendre parler quotidiennement le public finisse par en oublier l'importance. Comme pour le Guillot de la fable, crier au loup de façon injustifiée c'est prendre le risque tragique de se retrouver seul face à lui le jour où il apparaîtra réellement! Si on lasse l'opinion publique en créant des inquiétudes dont elle ne peut manquer, plus ou moins rapidement, de constater le peu de fondement et le ridicule, la faveur dont bénéficie actuel lement l'environnement passera et l'on se retrouvera, devant sa dégradation, sans l'arme salutaire que représente cette opinion publique.
C'est particulièrement vrai pour l'énergie nucléaire et plus encore depuis que, récemment, les menaces de pénurie énergétique se sont brusquement aggravées: il est absolument certain que les impératifs de la protection de l'environnement seraient non seulement aussitôt oubliés, mais même violemment rejetés par le public lui-même s'il commençait à ressentir, à l'échelon indivi duel, les effets réels d'une restriction de l'énergie sur ses habitudes, son confort, sa liberté, et même sa santé.
Je n'en veux que deux exemples: sans même parler de la circulation automobile, croit-on que les considérations sur la protection de l'environnement pèseraient lourd devant l'obligation, pour un pays privé de pétrole, de revenir partout au chauffage individuel au charbon dont on sait pourtant qu'il repré sente la cause de la pollution la plus importante de l'atmosphère? Et dans quel pays a-t-on interdit l'usage du tabac alors que la preuve, indiscutable ici, est faite qu'il tue chaque année, de par le monde, des centaines de milliers de personnes par cancer ou maladies de coeur, dont une forte proportion de non- fumeurs qui en subissent malgré eux les effets?
Notons que les moyens d'information, et la presse en particulier, ne sont pas toujours étrangers à la dramatisation excessive des problèmes nuclé aires. Un important représentant de la presse américaine ne déclarait-il pas franchement au récent congrès de radioprotection à Washington [11]: "Rendre "compte du décollage et de l'atterrissage quotidien sans incidents de 300 avions "par jour sur un aérodrome international ne présenterait, pour nous journalis mes, aucun intérêt. Aussi n'en parlons-nous jamais. Far contre, ce qui nous "intéresse c'est l'écrasement au sol d'un seul de ces avions, car c'est cela "qui attire le public!"
C'est inconsciemment, j'ose l'espérer, que les journalistes déforment donc la juste appréciation de la réalité par le public en ne relevant que les corrélations positives, celles qui, il faut bien le reconnaître, correspondent aux tendances morbides du lecteur. Cette attitude très discutable n'est pas différente de celle que quelques spécialistes excessifs de la radioprotection 8 PELLERIN
ont adoptée pour démontrer, pensent-ils, que l'énergie nucléaire pourrait être à l'origine de centaines de milliers de cancers et de leucémies. Ils oublient simplement qu'à l'inverse, en sélectionnant les seules corrélations favorables, on démontrerait tout aussi facilement qu'elle détermine une régression specta culaire de toutes ces maladies!
Parmi les critiques qui s'adressent à l'énergie atomique, l'une des plus fréquentes met en avant l'absence de preuve de l'existence d'un seuil d'action des rayonnements. Or cette absence de seuil n'a été retenue à l'ori gine, par les radiobiologistes, que comme une hypothèse de travail volontaire ment pessimiste, afin de définir des critères de sécurité particulièrement sévères, à une époque où l'on ne disposait que de données encore fragiles en la matière. Mais la validité de cette hypothèse n'a jamais été confirmée.
En fait, il est probable que cette conception sera à reconsidérer à la lumière des récentes découvertes sur l'importance des phénomènes de répara tion des molécules biologiques fondamentales [12]. Dans le domaine des nui sances chimiques qui, elles, provoquent quotidiennement des intoxications réelles, l'hypothèse de l'absence de seuil n'a d'ailleurs pas été retenue en matière de protection sanitaire [13]. Il n'est donc pas impossible que, dans l'avenir, une attitude identique soit envisagée pour l'action des rayonnements, mais l'on ne peut que se réjouir des précautions sages qui ont été prises dans ce domaine jusqu'ici.
Quelles sont donc les dispositions d'un contrôle réaliste de l'énergie nucléaire?
J'ai déjà eu l'occasion, à plusieurs reprises, dans le cadre des activités de l'Agence Internationale de l'Energie Atomique [5,14,15] de détailler les principes sur lesquels reposent ces dispositions. Je ne les rappelle donc ici que de façon succincte: -le premier principe est que le pouvoir de décision pour l'implan tation d'une centrale nucléaire, et le contrôle de son environnement doivent être exercés, indépendamment de l'exploitant, par les auto rités de Santé Publique. En effet, aux yeux des populations intéres sées, seuls les organismes compétents en radiobiologie, qui ne sont pas impliqués dans la production, peuvent lui apporter les apaisements nécessaires et lui prouver que les mesures prises par l'exploitant sont efficaces: il ne s'agit pas de mettre en doute la compétence ni les mesures de ce dernier, mais il faut bien reconnaître que l'expé rience a maintes fois montré qu'il ne pouvait être son propre avocat; -le second principe est la confirmation de la responsabilité de l'exploitant des centrales nucléaires. Il doit être astreint à prendre toutes les mesures, tant au niveau de la prévision que du fonctionnement, pour ne porter atteinte à la santé de quiconque du fait de ses activités. Cette responsabilité implique qu'il soit parfaitement conscient de la situation qu'il crée, donc qu'il véri fie en permanence lui-même les conséquences de son activité pour 1'environnement ; -il faut en contre-partie prendre conscience du très faible niveau actuel et à venir de la pollution radioactive et faire preuve de bon sens lorsqu'on la compare avec les autres risques. Dans ce sens, par exemple, le congrès des spécialistes de radioprotection en lan gue allemande [16], qui s'est tenu à Berlin en 1971, a estimé à l'unanimité que, dans l'ensemble, la contamination par la radioacti vité artificielle, quelle qu'en soit l'origine, restait actuellement nettement inférieure aux fluctuations de la radioactivité naturelle. IAEA-SM-180/76 9
Sur le plan pratique, ces principes impliquent un certain nombre d'obligations :
)°) L'exploitant d'une centrale nucléaire doit effectuer, indépen damment des services de Santé Publique, le contrSle de toutes les pollutions qu'il est susceptible de provoquer dans l'environnement (air, eau, chaîne ali mentaire, etc.). Aucun rejet ne doit être effectué sans une analyse préalable des activités en cause afin d'en calculer la dilution et de vérifier que les conditions imposées seront effectivement respectées. En ce qui concerne plus particulièrement les rejets atmosphériques, leur caractère instantané impose qu'au moins une évaluation préalable de l'activité totale soit effectuée avant chaque rejet. Un enregistrement à la cheminée pendant tout le rejet devra confirmer cette évaluation.
Cette disposition ne doit pas être transposée telle quelle aux effluents liquides et l'utilisation pour ces derniers d'appareils dits "à seuil d'alarme" ne peut en aucun cas remplacer les contrSles préalables qui doivent être effectués en laboratoire sur des prélèvements dans les cuves ou réservoirs de stockage transitoire des effluents.
2°) Dans tous les cas, les dispositions doivent être prises pour étaler dans le temps les rejets liquides ou gazeux, en vue de leur dilution la plus grande, les concentrations maximales admissibles ne devant être con sidérées que comme une lim ite extrême en deçà de laquelle tout sera fait pour se tenir aussi bas que possible.
3°) Chaque centrale doit disposer du matériel de laboratoire lui permettant localement d'effectuer correctement les analyses qui viennent d'être définies, et au moins la mesure des activités alpha totales, bêta et gamma totales (avec mesure séparée du tritium ). Un étalonnage correct de ces méthodes et de l'appareillage doit être vérifié périodiquement par les services de Santé P u b liq u e .
4°) Un registre officiel de l'ensemble de ces opérations doit être rigoureusement tenu à jour par l'exploitant.
5°) Tous les incidents susceptibles d'influer directement ou indi rectement sur le milieu doivent, même s'ils sont sans conséquence, être portés sans délai à la connaissance du service de Santé Publique et notés sur le r e g i s t r e .
A titre d'exemple, ces dispositions sont d'ores et déjà appliquées en France dans le cadre de conventions et elles vont être incessamment confir mées par des textes réglem entaires. Ainsi, pour une centrale de 900 MW élec triques, située sur un fleuve d'un débit de l'ordre de )000 m3 par seconde, les dispositions suivantes ont été prises:
a) Les normes à respecter:
Compte tenu de la capacité radiologique du bassin et du plan d'ensem ble des installations à venir, l'activité volumique maximale surajoutée par l'installation, calculée après dilution dans le fleuve, ne devra pas dépas ser 20 picocuries par litre pour les émetteurs a, ß (tritium mesuré séparé ment), ßy et y purs, et ]000 picocuries par litre pour le tritium.
En ce qui concerne les mesures immédiates des effluents gazeux, les activités volumiques moyennes totales, calculées après dilution au niveau du sol, au-delà d'une zone de )000 mètres au maximum autour de la cheminée 10 PELLERÍN
de rejet, ne devront pas, dans les conditions les plus défavorables, dépas ser 40 000 picocuries par mètre cube pour les gaz (activité y totale) et 100 picocuries par mètre cube pour les aérosols (activité B totale). b) Les contrôles à effectuer par l'exploitant:
- analyse de composition physico-chimique de tous les effluents liquides avant rejet, et mesure de leurs activités a, 6 et y totales, avec mesure séparée du tritium ;
- vérification de la dilution en activité 6 totale dans le fleuve, à mi- durée de tout rejet, sur prélèvement effectué à un kilomètre en aval;
- contrôles trim estriels, en trois forages, de la nappe phréatique avec recherche des radioéléments éventuellement en cause;
- enregistrement continu, à un mètre du sol, du rayonnement y dans l'air ambiant en trois points du site à une distance maximale de ¡000 mètres de la cheminée. L'un de ces points doit être situé sous le vent dominant de la cheminée et doit comporter, de plus, la mesure de l'activité volu mique g totale d'un prélèvement quotidien par aspiration de poussières sur filtre fixe, de l'activité volumique 6 totale d'un prélèvement mensuel de précipitations et de l'activité massique B totale (potassium 40 exclu) de deux prélèvements mensuels de végétaux;
- activité volumique 6 totale (potassium 40 exclu) de deux prélèvements mensuels de lait provenant d'une ferme située entre deux et cinq kilomètres sous le vent dominant de la cheminée.
c) Les vérifications par la Santé Publique:
La Santé Publique est destinataire de tous les résultats correspon dants. Elle effectue notamment, de son côté, au moins les contrôles suivants:
- analyse détaillée d'un échantillon aliquote moyen mensuel des effluents dits de basse activité prélevés avant rejet (activité totale inférieure à 0,t microcurie par litre);
- analyse détaillée sur un prélèvement, avant rejet, de tout effluent d'acti vité totale supérieure à 0,] microcurie par litre;
- analyse et contrôle de la dilution, sur prélèvement continu ou périodique, de l'eau du fleuve en aval du site;
- contrôle sur une station d'aspiration extérieure au site de l'activité de l ' a i r ;
- tous contrôles supplémentaires à l'intérieur comme à l'extérieur du site, qu'elle juge nécessaire, pour faire face à ses responsabilités réglemen t a i r e s .
L'ensemble de ces dispositions, déjà sévères, constitue, pour la protection effective de la santé des populations, une garantie très sûre, con firmée par l'expérience de vingt ans. Il n'est donc pas question de faire preuve, dans ce domaine, d'un perfectionnisme supplémentaire qui se traduirait par un gaspillage en crédits et en spécialistes qu'il vaut bien mieux consacrer à d'autres nuisances, réellement dangereuses celles-là. IAEA-SM-180/76 11
Seules des considérations de recherche scientifique peuvent encore ju stifier que, dans quelques cas, les investigations soient poussées au-delà de ce qui est d'ores et déjà plus que satisfaisant pour la protection de la s a n t é .
Il est certain que les différents organismes, responsables à l'o ri gine de la promotion de l'énergie atomique, ont autrefois investi d'importants efforts en crédits, en matériel et en spécialistes pour tenter de prévoir, aussi loin que possible, les éventuelles conséquences à long terme de ce déve loppement sur la santé de l'homme. D'où la floraison à laquelle on a assisté, au cours des vingt-cinq dernières années, de toutes sortes d'études, notamment écologiques, destinées à mettre en évidence le cheminement des radioéléments dans le milieu et les voies qu'ils pouvaient, fortuitement parfois, emprunter pour atteindre l'homme.
Il faut bien reconnaître que, dans l'ensemble, peu de résultats vraiment capitaux sont sortis de ces études (dont certaines pourtant fort sérieuses) menées dans les différents pays possédant une industrie nucléaire. D'une façon générale, elles se caractérisent par une extrême diversité en même temps que chacune ne traite que d'un problème très particulier, voire étroitement local: reconcentration de tel radioélément dans tel organisme animal ou végétal, progression de tel autre dans tel type de sol, étude écolo gique limitée autour de tel centre, etc. Mais il y a beaucoup de radioélé ments, d'organismes, de types de sol et de centres nucléaires, et il est, finalement, très d ifficile d'extraire de tout cela des conclusions présentant une réelle portée générale.
On peut dire, en tous cas, que l'incroyable nombre de ces programmes apporte au moins une certitude réconfortante: en vingt-cinq ans, les études aussi bien métaboliques qu'écologiques menées dans le monde entier n'ont jamais permis de relever de fait ou de situation réellement préoccupants autour des centrales nucléaires, et le seul incident important, celui de Windscale, n'a finalement eu aucune conséquence décelable pour la santé des populations.
Bien sûr, la sensibilité des appareillages et des méthodes s'est affinée, et l'expérience des chercheurs est devenue considérable. D'une façon générale, l'on peut dire que le niveau de la détection des radioéléments dans le milieu s'est, depuis l'origine, abaissé à peu près de deux ordres de gran deur, en même temps que les progrès de la métrologie radioactive et les nom breuses intercomparaisons permettaient de donner aux résultats un caractère beaucoup plus fiable. On retrouve ainsi, aujourd'hui, dans le milieu des radio éléments à des niveaux que l'on n'aurait pas soupçonnés il y a deux décennies, mais ceci à des taux qui s'échelonnent, selon les radioéléments, entre un cen tième et un millionième des concentrations maximales adm issibles correspondantes.
Il ne serait donc pas raisonnable d'augmenter encore les divers investissements pour tenter d'obtenir des performances toujours plus grandes: ce que l'on fait contre la nuisance radioactive doit, à ce stade, être comparé avec ce que l'on fait contre les autres nuisances et, désormais, c'est avant tout sur ces dernières que doivent porter les efforts et que l'on doit obtenir une am élioration.
Il n'est donc pas impossible que l'on en arrive, dans les organismes responsables de l'énergie nucléaire, comme cela s'est déjà produit dans quelques pays, à des révisions déchirantes dans 1'organisation-même de certains services et dans la définition de leur mission, particulièrement à la lumière des études "coût-bénéfice". 12 PELLEMN
Mais il ne faudrait pas que ces révisions altèrent, au sein-même de ces organismes, la qualité de la sécurité et la répartition des responsabilités entre deux domaines complémentaires, mais qui doivent rester distincts: la sûreté nucléaire et la radioprotection. Il n'est donc pas inutile, pour terminer, de rappeler ici les définitions précises de chacune de ces activités:
a) La Sûreté Nucléaire concerne la sécurité que doit présenter l'installation elle-même, la "machine" en quelque sorte. Cette vérification permanente de la sécurité intervient dès la conception des plans du réacteur, de ses annexes et des bâtiments. Elle intervient ensuite pendant toute la période de construction et d'organisation dont toutes les étapes doivent être sanctionnées par ses vérifications, et se poursuit pendant toute la durée du fonctionnement de l'in stallation dont elle vérifie constamment la main tenance. Elle intervient aussi lors de tout projet de modification quelcon que de cette installation et, enfin, lors de sa liquidation. C'est donc essentiellem ent une responaaM Mtà
b) La Radioprotection: sa vocation est essentiellement sanitaire. Il lui appar tient donc, aussi bien à l'intérieur qu'à l'extérieur des installations, d'étudier et de prévoir les conséquences pour l'homme de leur fonctionnement et, par conséquent, de veiller à ce que les lim ites maximales fixées ne soient jamais atteintes. Lui reviennent, par conséquent, les calculs de doses effectives au niveau de l'homme et des organes critiques, à travers la recherche des cheminements des radioéléments dans le milieu (elle peut, à cet effet, rechercher ce que l'on appelle les "groupes critiques", encore que leur importance et leur u tilité réelle semblent avoir été sérieusement surestimées). Il s'agit donc essentiellement d'une ¿e mádgcína et ¿e M oioytgtgs.
Où se situe la frontière entre ces deux champs de responsabilités? Très exactement au point de sortie des effluents radioactifs dans le milieu (cheminée ou canalisation de rejet). Mais il est bien évident que la santé de l'homme étant le but prim ordial, les dispositions de construction auxquelles doit veiller la sûreté nucléaire sont étudiées par elle à partir des limites proposées par la radioprotection à qui revient donc l'in itiative. Celà implique donc que, tout en conservant la totale responsabilité dans son domaine propre, chacune des parties fournisse à l'autre toutes les informations qui peuvent lui être utiles.
C 'est enfin bien évidemment au niveau des services responsables de Santé Publique que se font la coordination dernière et éventuellement les arbitrages entre radioprotection et sûreté nucléaire, puisque l'activité de ces deux disciplines vise finalement à réaliser des conditions optimales pour la santé de l'homme.
CONCLUSION
Il peut paraître paradoxal que le responsable d'un service national de Santé Publique, chargé du contrôle de la pollution radioactive, se fasse l'avocat du développement des centrales électro-nucléaires.
Ce paradoxe n'est qu'apparent. En effet, appartenant à la Santé Publique, et exclusivement préoccupé de sa sauvegarde, je considère les centra les électro-nucléaires comme l'un des moyens d'amélioration les plus efficaces de la protection sanitaire des populations, car leur développement ne peut IAEA-SM-180/76 13 manquer de faire disparaître, à terme, des pollutions traditionnelles extrême ment graves auxquelles, malheureusement, le public ne prête par habitude que peu d'attention. Les conditions dans lesquelles l'autorisation et le fonction nement des centrales électro-nucléaires sont actuellement contrôlés par la Santé Publique, dans la plupart des pays, présentent des garanties exception nelles, pour un risque minime. Ce qui est regrettable, c'est que des garanties équivalentes n'aient pas été instaurées pour la plupart des pollutions tradi tionnelles .
Comme, en matière de contrôle des conséquences du développement de l'énergie nucléaire, le dernier mot appartient à la Santé Publique, il est logique que, connaissant l'immense bénéfice que l'on est en droit d'attendre de cette forme de l'énergie, notamment sur le plan de la protection sanitaire prise dans son ensemble, elle en appuie la promotion sous la réserve, bien entendu, que les meilleures conditions de sécurité soient assurées.
Je voudrais donc terminer en encourageant les ingénieurs responsa bles du développement de l'énergie nucléaire à ne pas se laisser influencer par les querelles ridicules et affligeantes de la contestation. Celle-ci u tilise notamment la confusion systématique entre le seuil de danger et le seuil de détection des rayonnements qui est extraordinairement bas, beaucoup plus bas que celui de la plupart des autres techniques de mesure des nuisances. A partir de cette confusion, on essaie d'entraîner le public dans la dialectique spécieuse suivante: "si on détecte si bas, c'est parce que c'est très dangereux"!
En réalité c'est exactement l'inverse: il se trouve qu'on a la chance de pouvoir détecter les rayonnements à des niveaux très faibles et c'est la raison pour laquelle on a pu conduire les études de leurs effets comme on n'a malheureusement jamais pu le faire pour aucune des autres nuisances, dont cer taines restent toujours, elles, réellement préoccupantes.
REFERENCES
[1] NATIONS UN1ES-UNSCEAR, Ionizing radiations: levels and effects 1, UN, New York (1972) 73. [2] NATIONS UNIES-UNSCEAR, Ionizing radiations: levels and effects 1, UN, New York (1972) 133-172. [3] PRETRE, S., We are being fooled with these genetically significant doses from diagnostic X-rays (Notes to the Editor) Health Physics 25 2 (1973). [4] NATIONS UNIES-UNSCEAR, Ionizing radiations: levels and effects 1, UN, New York (1972) 150. [5] PELIERIN, P., MORONI, J-Р., «Principes de la surveillance de l'environnement d'une centrale nucléaire», Environmental behaviour of radionuclides released in the nuclear industry (Compt. Rend. Coll. Aix-en-Provence, .1973) AIEA, Vienne (1973) 111. [6] CHANTEUR. )., PELLERIN, P ., Pollution nucléaire et rayonnements ionisants, Rapport au Comité National pour l'Aménagement du Territoire (Oct. 1971). [7] Energie et Environnement, Rapport n° 8 du Ministère de la Protection de la Nature et de l'Environnement, La Documentation Française, Paris (1972). [8] NISHIWAKI, Y ., et coll., «Atmospheric contamination of industrial areas including fossil-fuel power stations, and a method of evaluating possible effects on inhabitants», Environmental aspects of nuclear power stations (Compt. Rend. Coll. New York, 1970) AIEA, Vienne (1971) 247. [9] MARTIN, J-E., et coll., «Radioactivity from fossil-fuel and nuclear power plants», Environmental aspects of nuclear power stations (Compt. Rend. Coll. New York, 1970) AIEA, Vienne (1971) 325. [10] NATIONS UNIES-UNSCEAR, Ionizing radiations: levels and effects 1, UN, New York (1972) 3 et 23-39. [11] CAREY, F., SHAPLEY, D ., SIMMONS, H. T ., in Panel discussion: «Science writers evaluation of the congress», 33 Congr. Intern. IRPA, Washington (9-14 sept. 1973). [12] LATARJET, R., Interaction of radiation energy with nucleic acids, Curr. Top. Radiat. Res., Quarterly 8 1 (1972) 1-38. [ 13] Réunion du Comité d'experts de la planification et de l'administration des programmes nationaux de lutte contre les effets adverses des polluants, OMS, Genève (Oct. 1973). 14 PELLERIN
[14] PELLERIN, P., « Surveillance de l'environnement des centrales nucléaires^, Environmental aspects of nuclear power stations (Compt. Rend. Coll. New York. 1970) AIEA, Vienne (1971) 387. [15] PELLERIN, P., Centrales nucléoélectriques: surveillance de la radioactivité de l'environnement, Bulletin AIEA 14 3 (1972). [16] Huitième Séminaire européen de radioprotection, en langue allemande, Berlin, (30 juin-2 juillet 1971).
DISCUSSION
J. S E D L E T : D o you believe that the p r o b l e m of safe storage of radioactive wastes from nuclear power plants has been adequately solved? P. P E LLERIN: No, not entirely but I would say it is less of a problem than storing certain non-radioactive industrial waste (e. g. chemical). Very satisfactory solutions are beginning to emerge for dealing with very high level waste, notably vitrification and incorporation in c e r a m i c s as well as storage in salt mines. The main problem with these methods is the cost but that can be ov e r c o m e . M. D E L P L A : I should like to c o m e backto the question of accidents in nuclear p o w e r stations. T h e s e h appen quite often but they are m e r e l y operational incidents without any risk to the public. Thanks to the pre cautions taken no worker inside a station has ever been irradiated severely e nough to affect his health a n d no one outside a station has ever been irradi ated except by amounts barely detectable with the most sensitive instruments. I a m referring here only to nuclear plants producing electricity and not to nuclear research centres, in which a number of people have been acciden tally irradiated, s o m e with fatal results. N . G . G U S E V : Y o u said that in F r a n c e the permitted concentration of beta-gamma-radioactive aerosols (apart from ^H) in the atmosphere around nuclear facilities is 100 pCi/m^. D o these regulations take into account the possibility of the isotope mixture containing long-lived gamma-emitting isotopes like 137(^ 60(2o^ 5i(^ ^ which after settling on the ground, m a y , in the equilibrium state, create high doses of g a m m a radiation in the area. P. PELLERIN: The 100 pCi/m^ and the other figures I mentioned are operational limits for activity measurements close to nuclear installations performed with the mini m u m stipulated amount of monitoring equipment. At distances of less than 1000 m from the point of release the figures are necessarily high, as dilution is far from complete. These limits are purely provisional and should not be confused with the regular standards. L. I. G E D E O N O V : Professor Pellerin, you referred to exaggeration of radiation danger on the part of the public and the press and I can confirm this. Journalists once changed the title of one of our articles from "The ocean will be clean" to " B e w a r e - radioactivity;" H o w e v e r , in protesting against such exaggeration, you maintain that the object of our work is to protect man. Does our task not also include the protection of the environ ment as such? Should we not also concern ourselves with future generations? What is your opinion of the concept of m a x i m u m permissible concentrations? Would it not be better to adopt the principle of restricting the amount of radioactive material released? P. PELLERIN: I quite agree with you. When I speak of protecting man, I certainly m e a n future generations as well, and to that end we must protect the environment. Therefore all unnecessary pollution should be avoided. IAEA-SM-180/76 15
В. L I N D E L L : I should like to r e m i n d you of the existence of the recent I CRP Publication 22 which explains paragraph 52 in ICRP Publication 9, namely that all doses should be kept as low as readily achievable, taking into account e c o n o m i c and social considerations. O n e consequence of this relates to the question asked by Mr . Gedeonov, n a m e l y that it is less useful to refer to concentrations of active materials than to refer to the a m o u n t s or, even better, to estimates of dose commitments and collective doses from given releases of active material into the environment. G. BRESSON: As has often been pointed out, it is necessary to distinguish between different types of installations (power stations, laboratories, etc. ) without losing sight of the fact that environmental surveillance must comply with the same general criteria. Accidents must also be well defined with respect to the environment, in which m a n occupies the predominant place. One should not overemphasize man's needs at the expense of the environment or vice-versa but the paramount consideration is, of course, the health of mankind. Speakers have mentioned the psychological aspects and the difficulties that can be encountered in the nuclear field as c o m p a r e d with other sectors of the e c o n o m y . Evidently, the more these problems are highlighted, the more the discussion escalates. I r e m e m b e r M r . Pellerin saying at a recent conference that in 1971 there were m o r e than fifty meetings on radiation protection involving about 2500 papers, twenty of those meetings being concerned with radioactive contamination of the environment, and that in 1972 the report of the Scientific Committee of the United Nations contained more than 50 000 bibliographic references on radiation protection, yet during that s ame period there was only one meeting on a non-radioactive pollutant (lead). All this provides food for thought. P. PELLERIN: Yes, we should not forget that the need to protect the environment arises from the need to safeguard man's health. Protecting the environment for its own sake would be pointless and taken to the extreme wo u l d m e a n eliminating m a n to prevent h i m causing pollution! M . D E L P L A : Yes, m a n must be protected above all else but there is no justification for blindly applying the law of proportionality between dose and effect and blaming nuclear power stations for cancer and genetic deaths, as s o m e people are wont to do.
OBJECTIVES OF ENVIRONMENTAL SURVEILLANCE Chairmen В. LINDELL (Sweden) P. PELLERIN (France) LAEA-SM-180/33
THE OBJECTIVES AND REQUIREMENTS FOR ENVIRONMENTAL SURVEILLANCE AT USAEC FACILITIES
M.B. BILES, M.W. TIERNAN. A. SCHOEN, C .G . WELTY Division of Operational Safety, United States Atomic Energy Commission, Washington, D .C ., United States of America
Abstract
THE OBJECTIVES AND REQUIREMENTS FOR ENVIRONMENTAL SURVEILLANCE AT USAEC FACILITIES. All USAEC facilities which have a significant potential for affecting the environment or which release significant quantities of pollutants are required to conduct a routine environmental surveillance program. practices intended to minimize the release of radioactivity and non-radioactive pollutants to the environment are functioning as planned. The basic objectives of the program are to determine the adequacy and effectiveness of effluent treatment and control, to determine compliance with applicable standards for the control of effluents and protection of the environment, and to establish a normal operational base-line against which the impact of non-routine releases can be monitored and assessed. New facilities are required to conduct surveys prior to start-up to characterize background levels of radioactivity, non-radioactive pollutants, toxic materials and other environmental parameters, and to identify potential pathways for human exposure and/or environmental stress as a basis for determining the nature and extent of the routine environmental surveillance program to be implemented subsequently. Existing facilities are required to conduct a routine program to determine the levels of radioactivity and non-radioactive pollutants in the environment as a result of plant operations, and to evaluate the environmental impact of any releases in relation to applicable standards. A summary of the results of the environmental surveillance programs in the vicinity of major USAEC installations throughout the United States of America is also discussed, including assessments of environmental impact, and the experience obtained with these programs.
1. tNTRODUCTiON
Environmental monitoring programs are conducted at all U.S. Atomic Energy Commission (AEC) facilities which could have a significant effect on the environment, or which release, process, or dispose of significant quantities of radioactivity or nonradioactive poHutants. The AEC is engaged in a wide variety of nuclear activities, for example, uranium enrichment, plutonium production, accelerator and reactor operations, and fue] fabrication and reprocessing, at many sites located throughout the U.S. (Figure 1). The kinds and amounts of radioactivity and nonradioactive pollutants of interest in the environment of each site are, of course, a consequence of the activities carried out at the site. The sites vary considerably in size, ranging from industria! sized tracts of a few square kilometers to very targe tracts of several thousand square kiiometers. The sites also differ markedly with respect to the nature and use of the surrounding environment and population distribution.
Though the monitoring programs conducted at the various sites differ in detail due to the dissimilarities among the sites, each has been established to satisfy the same objectives and requirements which are presently included in an AEC Management Directive (AECM 0513) entitled, "Effluent and Environmental Monitoring and Reporting." This directive prescribes policy for environmental monitoring, describes the objectives of the monitoring programs, and establishes uniform monitoring and reporting requirements for all AEC operations, thereby providing for a consistent approach to environmental monitoring regardless of site location.
19 20 BILES et al.
FIG. 1. Major USAEC sites.
The purpose of this paper is to outline and discuss these objectives and requirements and to describe typical monitoring practices and results. Historically the AEC has required environmental monitoring at each of its sites and observance of guidance from the International Commission on Radiological Protection (ICRP), the National Council on Radiation Protection and Measurements (NCRP), and the Federal Radiation Council (FRC) with respect to radiation protection of the public.
It should be noted here that, in addition to its operational responsibilities, the AEC is also responsible for licensing and regulating the commercial sector of the nuclear industry in the U.S. This entails parallel responsibilities to those described herein in connection with environmental and effluent monitoring and reporting for radioactivity from nuclear steam electric power plants, fuel reprocessing facilities and other commercial nuclear activities. However, this paper does not cover these activities, nor does it cover the activities of the U.S. Environmental Protection Agency (EPA) which also has some responsibilities for radioactivity in the environment or the activities of certain states which conduct independent monitoring of nuclear facilities through agreements with the AEC. It deals only with the operation of AEC-owned facilities.
2. OBJECTIVES OF ENVtRONMENTAL MON!TOR!NG
The overall objective of the environmental monitoring program is to determine if effluent controls are performing as expected and if the public and environment are being adequately protected with respect to any materials released from AEC operations. In general, the programs are expected to provide the information necessary to determine compliance with environmental radiation protection standards, and to assess environmental impact and the adequacy and effectiveness of waste treatment and control for both radioactivity and nonradioactive pollutants.
In connection with protection of the general public from radiation, the objective is to identify all significant sources of exposure, and to accurately assess the dose, if any, that the public is receiving as a result of the operations being conducted at the site. Judgments in this IAEA-SM-180/33 21 regard are made against the radiation protection standards in an AEC Management Directive (AECM 0524) entitled, "Radiation Protection Standards." These standards are consistent with the guidance and recommendations of the ICRP, NCRP, and FRC relative to public exposure, and incorporate the policy of holding public exposure to the lowest practicable level. There is a parallel objective to assess compliance with air and water quality standards for nonradioactive poUutants prescribed by Federa] ]aws and regulations or by the state in which the site is located. Attainment of these objectives requires maintaining surveillance of site operations as regards any potential impact of such operations on the environment.
With respect to waste treatment and control, the processes and operations involved in the handling of radioactive materials are designed and operated to maximize containment and control of effluent radioactivity. The introduction of radioactivity into liquid and airborne effluents is controlled at the source, and the effluents are monitored prior to discharge. The environmental monitoring program is intended to provide information on the efñcacy of effluent treatment and control, and is looked upon as yet another check on the adequacy and effectiveness of operating practices and procedures and the control of wastes.
3. REOUtREMENTS FOR ENVIRONMENTAL MONITORING
3.1 Environmental Monitoring and Reporting
3.1.1 Monitoring
A preoperational environment survey is required at each new site or facility which could have a significant potential for adverse environmental impact, or which will process, release, or dispose of significant quantities of radioactivity or nonradioactive pollutants, unless it is under the umbrella of an existing monitoring program. This survey provides the baseline data for the subsequent evaluation of the impact of site operations on the environment.
An environmental monitoring program which fulfills the objectives discussed earlier is required at all existing sites and facilities. It is expected that judgment will be exercised on the extent of the program, but that all signiñcant pathways of human exposure will be monitored and consideration will be given to such things as hazard potential, quantities and concentrations of materials released, local public interest or concern, and utilization of land and water resources and other environmental activities in the vicinity of the site. The dissimilarities in the operational and environmental situation among the sites make it inappropriate for the same monitoring practices and procedures to be followed at each site.
Because they are directly relatable to an assessment of human exposure, the environmental parameters most commonly and frequently sampled for radioactivity are air, water, and food stuffs, although soil, sediment, and vegetation are frequently sampled, usually as trend indicators. Ambient radiation is also measured at many sites. The nuclides most commonly analyzed for are strontium-90, cesium-137, iodine-131, tritium, and plutonium-239. Table 1 indicates the types of environmental media routinely sampled for radioactivity and the number of sampling locations for six types of sites.
The frequency with which various environmental media are sampled and analyzed is dependent on the nature of the media and on the characteristics of the radionuclides of interest. The dynamic systems, such as streams, ambient air, milk, aquatic biota and wildlife, are continuously or frequently monitored or sampled regardless of the physical decay rate of the radionuclides of interest. However, compositing of samples over a long period of time is often practiced when long-lived radionuclides are measured. The more static systems, such as lake water, soil and sediment, are less frequently sampled and analyzed unless the radionuclides of interest have a short half-life. 22 BILES et a l.
TABLE)
ENVIRONMENTAL SAMPLES AND NUMBER OF SAMPLING LOCATIONS AT SIX TYPES OF AEC SITES 1972
TYPE OF SAMPLE AND NUMBER OF SAMPLtNG LOCAT)ONS Z О t- < Û < 2 < 5 b < g Ш S и О i E Û z TYPE OF StTE < g < < g м к о U- < > ЙS
PLUTONtUM PROCESS)NG AND RESEARCH 33 46 60 90
URANtUMPROCESStNG 9 11 4 16
REACTOR AND FUEL REPROCESStNG 15 16 14 6 19 14
MULTtPURPOSE LABORATORY 36 10 9 12 1 5
ACCELERATOR 18 6 4
NUCLEAR EXPLOStVES TESTtNG 104 91 360 35 88 60
An annual evaluation of the radiation dose received by the public as a result of operations is made at all sites where a routine environmental monitoring program is conducted. Every effort is made to assure that the dose assessments are representative of the actual exposure to individual members of the public and to population groups. The method of choice used in assessing dose is the calculation of dose based on environmental monitoring data, for example, concentrations of radioactivity in ambient air, drinking water, and milk. In those instances in which the concentrations of radioactivity in environmental media are too low to be measured, doses are calculated based on the quantities of radioactivity discharged from the site, using meteorological and other parameters affecting the levels of human exposure.
Pollution control in the U.S. with respect to nonradioactive pollutants is now being achieved with effluent standards as well as with generally applicable environmental standards. Potential sources of pollution, including AEC facilities, are generally required to monitor effluents rather than environmental media for these materials. Monitoring of the environment is generally the responsibility of the pollution control authorities and is required of a facility only if control is expressed in terms of environmental levels. AEC facilities are subject to and must meet all effluent standards established pursuant to Federal and state anti-pollution laws.
In point of fact the environmental monitoring programs at many sites go far beyond the minimum requirements and encompass a wide variety of parameters. At the sites which have the potential to influence the water quality of a nearby river or stream, the water above and below the plant is often monitored for such parameters as dissolved oxygen, pH, temperature, turbidity, conductance, biological oxygen demand, and coliform bacteria, or for chemicals discharged from the sites, even though the chemicals are monitored at the point of discharge. IAEA-SM-180/33 23
Groundwater is atso often monitored for nonradioactive constituents, as is ambient air quality. Most commonly measured are those parameters for which ambient air standards have been established, for example, sulfur oxides, nitrogen oxides, and particulates. Again, the objective is to assess the impact of site operations on the environment.
3.1.2 Reporting A report on the results of the environmental monitoring program is prepared annually for each site. The report is intended to serve several purposes: to describe the program, to document the monitoring results, and to summarize and interpret how the levels of radioactivity and nönradioactive pollutants in the environs of the sites resulting from site operations compare to applicable standards or relevant parameters such as background or natural radioactivity. Copies of this report are routinely sent to Federal, state, and local public health and pollution control agencies and other interested parties and persons. Onsite environmental monitoring data are not included in this report unless the data are directly relatable to applicable standards or the characterization of offsite conditions.
An environmental monitoring report is not required for those sites which do not have a significant potential for environmental impact. Each of these sites however, must submit an annual summary which provides a concise statement of the limited potential effect of its operations on the offsite environment and the public. Supporting data, including effluent data as appropriate, are included.
3.2 Effluent Monitoring and Reporting
3.2.1 Monitoring
Monitoring and reporting of radioactivity in effluents is done in conjunction with environmental monitoring. Effluent radioactivity data are used as a management tool to evaluate the effectiveness of waste treatment and control, and provide the source term for assessing potential environmental impact. The data are also used to compile an annual inventory of the quantities of radioactivity released from AEC facilities to the environment, and to onsite waste treatment or disposal systems. Effluent monitoring data for nonradioactive pollutants are used to determine compliance with standards. Standard sampling and analytical techniques are encouraged whenever possible. Measurements of volume, rate of discharge, content, etc., are made at the point which most closely represents what is being released after all treatment and control has been implemented. Continuous sampling is used where there is wide variation in the concentration or mixture of potential pollutants in the effluent stream. Periodic sampling suffices when the concentrations and mixtures are reasonably constant and there is little likelihood of unusual variations. Similarly, proportional sampling may be necessary when effluent flow rates vary widely, whereas a single representative sample will suffice for batch discharges. Normally, analyses are performed for specific radionuclides expected or known to be present in effluent streams in measurable concentrations in lieu of gross alpha or beta measurements.
3.2.2 Reporting
An effluent monitoring and reporting system which provides an annual inventory of the quantities and concentrations of radioactivity released to the environment is in operation at each site. The inventory includes all significant releases, including accidental releases. A computer-based system facilitates uniform reporting from all sites. The system consists of a series of computer programs which maintain information on past and current releases from each significant discharge point at each site. The system can provide summary information, both before and after radionuclide decay, by discharge point, facility, or site for airborne and liquid releases. It has been determined that, at present, a similar system for compiling information on nonradioactive pollutants is neither necessary nor practical. ] j ARMS-LARGE AREA SURVEY^ SPEOAL-ARMS SURVEY
FIG.2. Surveys performed by USAEC's aerial radiological measuring system (ARMS.). IAEA-SM-180/33 25
3.3 Special Environmenta! Monitoring Capabilities
There is an occasional need to augment the routine effluent and environmental monitoring programs at the sites with special monitoring capabilities. The services of special AEC groups and independent laboratories within Federal and state agencies are frequency used for this purpose.
3.3.1 Mobile Laboratories
Mobile radiological monitoring capabilities are maintained at most sites for special and emergency monitoring purposes and, in some instances, as a part of the routine monitoring programs. These mobile units vary from small vehicles equipped with portable air sampling and radiation monitoring instruments, to large vans equipped with multichannel analyzers, analytical laboratory facilities, and shadow shielded whole body counters. Although environmental monitoring in the vicinity of the sites is usually accomplished by using fixed sampling and monitoring equipment, mobile units are frequently used to supplement fixed station monitoring and to perform special studies. For example, Mound Laboratory has studied the levels of tritium oxide and tritium gas in the ambient air using a van equipped with special sampling and counting facilities. During the period when the large piutonium production reactors at the Hanford site were operating and releasing cooling water containing activation products to the Coiumbia River, a mobile whole body counter was used in a study to determine the uptake of radioactivity in the nearby populace.
3.3.2 Aeria) Radiological Measuring System
The AEC is increasingly using the Aerial Radiological Measuring System (ARMS) for performing routine radiation surveys of large land areas around AEC and private nuclear installations. Although ARMS was developed and is maintained primarily as an emergency response system, it is a valuable tool for performing routine surveys which assist in maintaining the system in a high state of proficiency.
The ARMS is operated for the AEC by EG&G, Inc., and is capable of making rapid surveys and identifying unknown gamma emitting contamination. To date, land area surveys covering more than 10 percent of the conterminous United States have been performed (Figure 2). ARMS is being used to make periodic surveys at all AEC sites to map the distribution and levels of naturally occurring and man-made gamma radiation in onsite and offsite areas, and to detect any signiñcant buildup or movement of contaminants.
The system incorporates on-board instrumentation for gross gamma radiation measurements and spectral analyses, air sampling, meteorological measurements, airplane positioning, black and white photography, and infrared scanning. The ARMS instrumentation is capable of making spectral analyses which permit the detection and measurement of specific gamma emitters, including americium-241 which is a convenient indicator of the presence and distribution of plutonium-239. Figures 3, 4 and 5 show the results of an ARMS survey for gross activity, cesium-137 and americium-241 over a waste management area at the Hanford site. The photographic and infrared sensing equipment on board are used to investigate other environmental conditions such as soil erosion, geological formations, and thermal conditions. For example, the results of an ARMS infrared scan of the thermal plume produced by cooling water discharged from a conventional steam-electric plant are illustrated in Figure 6. The ARMS instrumentation can sense as little as a quarter of a degree centigrade change in temperature, and has the capability of plotting temperature contours.
3.4 Environmental Protection Agency and State Agencies
EPA and a number of state agencies have established radioactivity monitoring programs in the vicinity of AEC sites as a part of their environmental surveillance activities. Results 26 BILES et al.
METRES SCALE
FIG.3. Hanford waste-handling area: Gross gamma contamination.
METRES SCALE
FIG.4. Hanford waste-handling area: Cesium-137 contamination. IAEA-SM-180/33 27
SCALE
PIG.5. Hanford waste-handling area: Americium-241 contamination.
from these programs not only supplement, but provide confirmation of AEC monitoring data. Although the AEC is responsibte for regulating the releases of radioactivity from nuclear installations, the EPA is responsible for establishing and verifying compliance with the generally applicable environmental radiation standards which were formerly established by the Federal Radiation Council. EFA is therefore interested in the concentrations of radionuclides in the vicinity of all nuclear installations. The EFA has sampled environmental media in the vicinity of AEC sites to make an independent assessment of radioactivity levels. On a number of occasions the AEC has requested EPA to conduct special independent radiation surveys in the vicinity of an AEC site. There are also numerous programs for exchange of samples and cross calibration of radiometric techniques to assure comparability of analytical results.
4. RESULTS OF MONtTORtNG PROGRAMS AT AEC StTES
4.1 Radioactivity
As indicated earlier the potential radiation dose that a hypothetical member of the public could receive as a result of operations is estimated at all sites which have the potential for contributing significantly to radiation exposure. These dose estimates, which are summarized here, provide an indication of the generally small impact that AEC operations have had on the environment. In 1972 dose estimates were made for 32 sites. In most cases, they were not more than a few percent of the radiation protection standards, regardless of whether the dose was estimated for an individual or for a suitable sample of the population. This confirms the adequacy and effectiveness of effluent treatment and control, and the vigorous implementation of the program undertaken in the past few years to significantly reduce the quantities and concentrations of radioactivity, and nonradioactive pollutants as well, in effluents released from the sites. IE e al, et BILES
FIG.6. Thermal plume by 'ARMS' infra red scan at a conventional power plant at Pittsburgh, California. IAEA-SM-180/33 2 9
4.1.1 Plutonium Processing and Research Sites
At sites involved in plutonium handling a concerted effort has been made to develop a monitoring capability for very low levels of airborne plutonium because of the concerns in connection with this mode of exposure. Airborne plutonium is currently routinely measured at fallout levels, i.e., 10**^ pCi/cm^; therefore, the contribution of any site to exposure from this nuclide is readily estimated. In 1972, airborne releases ranged from about 10 to less than 1 millicurie of plutonium. The resulting radiation exposures were estimated to be very low. However, the major concern with respect to plutonium is its buildup in the environment, particularly in soils. Therefore, despite the very low levels of exposure around these sites, an active program is underway to reduce the amounts of plutonium being released. Significant reductions have already been made, principally through the installation of high efficiency filters and by testing their efficiency before and after installation. The current goal is to reduce plutonium discharges from all facilities to non-detectable levels in the effluent streams.
One of these sites has measurable levels of plutonium in offsite areas due to activities or practices which have been discontinued. At the Rocky Flats Plant in Colorado, plutonium contaminated machining oil, which was stored in drums on an unsheltered storage pad, leaked onto the soil from whence the plutonium was transported offsite by wind action. The drums have been removed and the storage area stabilized, but low levels of residual soil contamination extend outward from the site as shown by the computer plotted contour lines in Figure 7. The contours were derived from an extensive soil sampling program. Soil contamination levels range up to 2 ^Ci/m^ near the plant perimeter with a total quantity offsite estimated to be 3-5 curies. Air concentrations averaged over a year ranged from 1.7 to 3.6x 10"* 6 juCi/cm^ o f air, compared to the radioactivity concentration guide (RCG) of 2 x11""' ^Ci/cm3 .
FIG .". Plutonium in soil and air in the vicinity of the USAEC Rocky Flats Plant. 30 BILES et a l.
4.1.2 Uranium Processing Sites
Uranium releases occur at sites where the metal is refined, enriched, or fabricated into reactor fuel elements. In 1972 the estimated intake of uranium by an individual at the boundary of these sites, from both airborne and liquid sources combined, ranged up to about 1% of the standard, which is based primarily on the chemical toxicity of uranium in the kidney.
4.1.3 Reactor and Fuel Reprocessing Sites
Major reactor and fuel reprocessing operations are conducted at large, isolated sites such as the National Reactor Testing Station and the Savannah River Plant. In 1972 the estimated whole body exposure at the boundary of such sites from atmospheric releases of radioactivity ranged from 1 to 2 mrem, less than 1 % of the radiation protection standard for an individual. Tritium and argon-41 contributed nearly all of the dose, although some was also due to krypton, xenon, and iodine-131. Some radioactivity from liquid effluents reaches river water, which subsequently becomes an important pathway of exposure through irrigated crops, drinking water supply, consumption of fish, and recreation. The estimated dose from this pathway is well below 1 mrem at all of the sites. It should be noted that although these sites manage all of the AEC high level wastes, this operation does not contribute significantly to offsite exposure.
4.1.4 Multipurpose Laboratory Sites
A wide variety of nuclear research and development activities are conducted at a number of relatively small sites located near population centers. The effluents from these sites are somewhat different in composition from those from the other sites discussed above. The major contributors to public exposure are direct gamma sources, for example, a field irradiation source, and airborne tritium and argon. In 1972 estimated doses ranged up to about 18 mrem from gamma irradiation, and up to a few mrem from tritium and noble gases. Little exposure occurs from radioactivity in liquid effluents which enter nearby streams because the water is not used for drinking or irrigation. In some cases the ingestion of fish from such streams could result in some dose, although it is estimated to be very small, less than 0.1 mrem annually.
4.1.5 Accelerator Sites
The sources of population exposure near AEC accelerator facilities are the neutrons and gamma rays emitted from the machines, and airborne activation products, namely, oxygen-15, nitrogen-13, carbon-11, and argon-41. Combined neutron and gamma radiation exposure rates in offsite areas range from less than one mrem per year at most sites to about 50 mrem per year at one site. Normally, airborne activation products are not released while the accelerators are operating, and there is a waiting period before exhaust fans are turned on in beam areas to permit decay of the short-lived activation products. The dose to individuals offsite due to the short-lived activation products, principally argon-41, is less than approximately 1.5 mrem annually at all sites.
4.1.6 Nuclear Explosives Test Site
Nuclear explosives are tested at the Nevada Test Site (NTS). The testing program is not a significant source of radiation exposure because all tests are conducted underground. Consequently little if any radioactivity from such tests escapes to the atmosphere. In 1972 the offsite dose to a suitable sample of the population was estimated to be less than 0.1 mrem from airborne effluents.
At the NTS, nuclear weapons and related testing has resulted in soil contamination onsite and in adjacent areas. The resulting cumulative levels of plutonium in soils on and off IAEA-SM-180/33 31 the site are of particular interest to the AEC and are being evaluated by the Nevada Applied Ecology Group which consists of AEC, AEC contractor, and EPA personnel. The plutonium levels in soil outside government-controlled lands range from less than 1 (approximately worldwide background) to about 140 nCi/m^. Air sampling data gathered around the site, and in areas downwind of the most contaminated areas, indicate annua! average concentrations due to resuspension to be less than one percent of the RCG of 2 x lO**** ^Ci/ctrP.
The distribution and levels of plutonium in excess of worldwide background in the vicinity of all AEC plutonium handling facilities and sites are being assessed through soil sampling and americium-241 gamma surveys. Americium-241 is found in conjunction with plutonium-239 and the relative concentrations of the two nuclides can be determined by alpha spectrometry. With further refinements in monitoring and computer techniques for plotting soil concentration contours, it is anticipated that the environmental levels and distribution of plutonium in excess of worldwide background can be reliably determined.
4.2 Nonradioactive Pollutants
A number of nonradioactive materials, for example, beryllium, have received special attention at AEC sites. With few exceptions beryllium is continuously monitored in the exhaust stacks of the facilities in which it is processed. The Environmental Protection Agency recently established an emission rate of 10g per day per facility for the entire beryllium industry based on a long established AEC ambient air standard of 0.01 g per cubic meter of air. AEC sites, using high efficiency particulate air filters, have maintained beryllium discharges at less than one-tenth of the EPA emission standard as a matter of practice. Most AEC sites which process beryllium also routinely monitor for this material in ambient air.
tn the last few years the largest expenditures for pollution control at AEC sites have been made in connection with reducing sulfur oxides and particulate emissions from fossil fuel steam plants to comply with applicable state pollution control standards. Control of particulates is being accomplished by converting boiler units to lower residual fuels, improving burning efficiency, and/or installing electrostatic precipitators. Sulfur releases have been reduced, thus far, by using low sulfur content fuels, although such fuels are now in heavy demand and in some locales in short supply. Environmental monitoring for pollutants from fossil fuel plants at AEC sites, often performed under unfavorable meteorological conditions, has demonstrated that emissions are low enough to be within the ambient air standards of the U.S. Environmental Protection Agency.
The most troublesome chemical pollutant in liquid effluents has been the chromate additive used for corrosion inhibition in cooling towers and heat exchangers. The U.S. Public Health Service drinking water standard of 0.05 mg of hexavalent chromium ion per liter of water is often imposed on effluent streams prior to discharge to a receiving water body. The most feasible techniques available for meeting the chromate standard are improved control of chromate levels in cooling waters, water recycle, and substitution of other corrosion inhibitors. AEC experience in controlling chemical pollutants released to surface and ground waters has generally been good, although additional controls on releases have been required in recent years with the adoption of more stringent EPA and state standards for the protection of drinking water sources and aquatic life.
The Hanford site is probably the best known of all AEC sites in connection with thermal discharges and their effects on the environment. In 1964, nine plutonium production reactors were in operation on the Columbia River at Hanford. Eight of the reactors used once-through-cooling; that is, water for cooling was drawn from the river, passed through the reactor core, and returned to the river. The heated discharges from all nine reactors raised the average river temperature about 1.5 °C, and were of concern because the Columbia River and 32 BILES et al.
TA BLE 4 SUMMARY OF DOSE ESTIMATES FOR HYPOTHETICAL INDIVIDUALS NEAR AEC SITES 1 9 7 2
DOSE PERCENT OF CRtHCAL MODE OF PRtNCtPAL TYPE OF SITE MREM STANDARD ORGAN EXPOSURE NUCLiDES Plutonium Processing and Research <1-3 < 0 .6 Bone Air Ptutonium
Uranium Processing 1 Kidney Air-Water Uranium
Reactor and Fuel Rep— , <1-2 < 0 .4 Whote Body Air Water Zinc-65
Multipurpose Laboratory < 1 - 18 <1-4 Who!e Body Gamma Air
Accelerators <1-50 10 Whote Body Direct Neutron
Nuclear Explosi.es Testing <0.1 < 0 Л 2 Whote Body Air Noble Gases
its tributaries are major salmon spawning sites. Much research has been conducted by the AEC and others to determine the effects of the temperature increase on these and other species of fish. The results indicate that adult ñsh are not harmed, and that there is only a remote probability of harming juvenile ñsh. Between 1965 and 1971 the eight single-pass reactors were shut down, and thermal effluents are no longer a major concern at that site. Thermal discharges at other AEC sites have generally not been a concern with respect to possible adverse effects on fish life because cooling waters are not discharged directly to a receiving stream or because cooling water volume is too small to be of consequence.
5. SUMMARY
Major AEC sites are required to conduct an environmental monitoring program to provide assurance that the public and the environment are being protected with respect to radioactivity and nonradioactive pollutants from site operations, and as a check on the adequacy and effectiveness of waste treatment and control. In 1972 the estimated doses to hypothetical individuals living near AEC sites were generally well below the radiation protection standards. The estimated doses are summarized in Table II for six different types of AEC sites. The results of a site's environmental monitoring program are documented annually in an environmental monitoring report. The purpose of the report is to summarize and interpret how environmental levels of site contributed radioactivity and nonradioactive pollutants compare to applicable standards or relevant parameters. Copies of these reports are routinely sent to Federal, state and local public health and pollution control agencies, and are available to the public.
DISCUSSION
P. R. K A M A T H : Would you please indicate the non-radioactive pollutants associated with the different nuclear units? M. B. BILES: The non-radioactive pollutants to which I referred in the paper are those that are of greatest c oncern at U S A E C - o w n e d facilities. S o m e of these pollutants are produced b y non-nuclear support facilities and include particulate matter and oxides of sulphur and nitrogen released to the atmosphere from fossil-fuel-burning steam generating plants. The most IAEA-SM-180/33 33 troublesome pollutants of this type released f r o m nuclear facilities include corrosion inhibitors such as chromâtes in liquids discharged from reactor and uranium diffusion plant cooling towers, waste thermal energy in single pass coolants from production reactors, nitrates in liquid wastes from fuel preparation and reprocessing facilities, and oxides of nitrogen discharged to the a t m o s p h e r e at fuel reprocessing facilities. Although, in s o m e instances, such pollutants, at the point of discharge, have exceeded standards applicable to waters classed as aquatic life systems or drinkingwater sources, they have not resulted in concentrations in excess of applicable standards for air or water beyond small mixing zones. P. R. K A M A T H : Could you describe the follow-up studies on thermal pollution and indicate the detection sensitivity of the infra-red camera? M. B.. BILES: The infra-red measuring equipment of the U S A E C 's A R M S system was only recently installed and, therefore, has had limited use up to now. It is anticipated, however, that this equipment will enable us to define m o r e clearly the distribution and dissipation of thermal energy in fresh water and marine systems and provide a firmer basis for thermal pollution studies conducted in these systems. Up to now, very intensive thermal pollution studies have been conducted on the Columbia River at the Hanford Reservation and on the S a vannah River at the S a vannah River plant, to determine whether, and to what extent, th e r m a l discharges have affected these aquatic systems. All levels of aquatic life have been studied in these ecosystems, with particular emphasis on the life cycle of salmon and trout in the cold waters of the Columbia River and the quality of algae in the w a r m and highly nutrient waters of the Savannah River. Although past and present operations have resulted in the temperature of the Columbia and Savannah Rivers being raised by 1 to 2°C, the results of these studies indicate no observable effects on the aquatic life of the two rivers due to past and present thermal discharges. The A R M S system is capable of detecting a quarter of a degree Celsius variation in surface temperature and, using existing computer programs, can plot surface-temperature contours at one quarter of a degree Celsius or greater intervals. Surface level temperature measurements are made for calibration purposes and subsurface measurements can be made as necessary from a boat, to provide three-dimensional temperature profiles.
IAEA-SM-180/46
ANALYSE DES PRINCIPES DIRECTEURS DES PROGRAMMES DE SURVEILLANCE DE L' ENVIRONNEMENT
G. BRESSON, R. COULON Commissariat à l'énergie atomique, Département de protection, Fontenay-aux-Roses, France
Abstract-Risumé
ANALYSIS OF GUIDING PRINCIPLES FOR ENVIRONMENTAL SURVEILLANCE PROGRAMMES. on statutory, technical, economic and social criteria. Highly elaborate rules have been established in most countries, as well as at the international level. Designed to meet the specific problems that are likely to arise in each case, these rules must nevertheless fit into a more general framework, a framework which in some cases can be found ready-made in existing legislation. As regards the technical foundations, it seems essential to get away from the fixed idea of radiological harm and to consider as a whole all the consequences likely to result from the operations of an installation, including in one's study only the most characteristic elements. In an analysis of this kind one will necessarily consider problems relating to the installation itself to the site, and more generally to the environment around the installation. The economic
ANALYSE DES PRINCIPES DIRECTEURS DES PROGRAMMES DE SURVEILLANCE DE L'ENVIRONNEMENT.
INTRODUCTION
Contrairement à d'autres activités de type industriel ou économique, le domaine nucléaire s' est développé avec un souci constant de la connais sance, la maftrise et, partant, la possibilité de limitation, de son éventuelle action sur le milieu et sur 1' ho m m e . C' est ainsi que 1' on a très rapidement admis qu' il convenait d'assortir toute installation de ce type d'un controle, plus ou moins élaboré, de son environnement.
35 36 BRESSON et COULON
Il appartenait, et il appartient toujours bien entendu, aux autorités nationales de se prononcer quant à la nécessité d'établir une surveillance et d'en définir les modalités, les contraintes pouvant s' avérer différentes d'un pays à 1' autre. Cependant, la plupart des organismes internationaux (CIPR, AIEA, E U R A T O M , etc.) ont pensé devoir intervenir au niveau de 1' établissement des principes directeurs gouvernant 1' organisation de tels programmes, dans le but, louable, d'assurer une certaine cohérence dans un domaine qui dépasse très souvent les frontières nationales. Mais aussi, 1' experience acquise au cours des deux dernières décades justifie un réexamen des principes acquis. D' une façon générale, il semble qu' au delà de considérations réalistes sur le rôle des programmes de surveillance, 1' établissement de ces derniers ait été fortement influencé par 1' aspect psychologique du problème, et, pour les responsables, le sentiment de crainte que leur inspirait l'incerti tude relative dans laquelle ils se trouvaient. Cet état de fait fort compréhensible dans la «phase de jeunesse» doit céder le pas à une conception plus objective au fur et à mesure que le rythme de croisière sera atteint. Il importe de s' y préparer dés maintenant. L e s activités de type nucléaire ne sont pas les p r e m i è r e s et les seules qui puissent présenter un risque potentiel pour 1' h o m m e et son milieu; ceci veut dire que les responsables se sont déjà, dans le passé, trouvés confrontés à des problèmes de cet ordre et que par conséquent il existe déjà une base sur laquelle il est possible de se fonder. On peut arguer que le développement considérable des activités conventionnelles comportant des risques pour 1' environnement et la forte prise de conscience de 1' opinion publique font que ces bases sont m a l adaptées et qu' il conviendrait plutôt de les revoir; en fait, qu' il s' agisse de traiter les autres d o m a i n e s c o m m e le domaine nucléaire ou le domaine nucléaire c o m m e les autres domaines, le véritable problème est de ne pas faire vis-à-vis de cette nouvelle forme d'utilisation de 1' énergie une individualisation qui ne pourrait que lui être préjudiciable.
LES PRINCIPES
L e s principes relatifs à 1' établissement des p r o g r a m m e s de s u r veillance sont étroitement liés à la notion d'objectif. La mise en place d'une surveillance ne doit pas relever de critères autres que ceux de répondre à une nécessité et la conception du programme être fonction de la réponse à donner. Pour ce qui est des problèmes posés par les installations lors de leur fonctionnement normal, on a proposé un trës grand nombre d'objectifs pour les programmes de surveillance. En les examinant, on s' aperçoit qu' ils peuvent être classés en quatre catégories:
1) Les objectifs d'ordre réglementaire: - vérification du respect de la réglementation en vigueur, - protection de 1' exploitant contre d'éventuels recours. 2) Les objectifs d'ordre sanitaire ou technique: - estimation de 1' exposition réelle résultante, - vérification des contrôles effectués au niveau des rejets, - détection des rejets non prévus, 1AEA-SM-180/46 37
- amélioration des connaissances sur le comportement des polluants dans le milieu et sur leur transfert. 3) Les objectifs d'ordre économique: - possibilité de rejet évitant 1' utilisation de procédés techniques d'épuration coûteuse. 4) L e s objectifs d'ordre psychologique et social: - possibilité de rassurer 1' opinion publique et d'éviter toute perturbation sociale.
Il convient donc d'examiner successivement chacun d'entre eux et d'en discuter le bien fondé et les limites dans le contexte général de 1' ensemble des activités susceptibles d'apporter certaines nuisances à des individus, à des groupes d'individus ou à une population.
1. OBJECTIFS D' ORDRE REGLEMENTAIRE
Il s' agit là d'un d o m a i n e pour lequel les principes directeurs ne peuvent guère avoir de portée générale dans la mesure où la législation diffère considérablement en la matière d'un pays à 1' autre. Il existe toutefois un élément quasi c o m m u n : c' est 1' adoption dans la plupart des règlementations nationales des recommendations émises par la C I P R en ce qui concerne les n o r m e s de base, en particulier les limites d'exposition pour les personnes du public; toutefois, les contraintes imposées à un exploitant pour lui permettre de respecter ces limites peuvent varier notablement en fonction de la réglementation générale, dite de droit commun, ou des textes de portée particulière.
1.1. Réglementation générale en France
La mise en oeuvre d'établissements dont les activités sont susceptibles d'apporter à un certain nombre d'individus une gène, voir des risques, ne constitue pas un problème nouveau; par la loi du 19 décembre 1917 relative aux établissements dangereux, insalubres ou incommodes, les législateurs ont voulu que les pouvoirs publics aient un droit de regard sur leur création et leur fonctionnement par le biais d'un régime de déclaration ou d'autorisation. C' est une loi de portée générale qui couvre tous les secteurs, y compris celui de la radioactivité, en particulier par les modifications qui lui ont été apportées ultérieurement. L' insuffisance de cette loi, lorsque sont apparues des installations nucléaires importantes, a entraîné la promulgation d'un nouveau décret, celui du 11 d é c e m b r e 1963 relatif aux «installations nucléaires de base», décret lui-même modifié récemment (avril 1973). Ce décret modifié poursuitdansl' espritde la loi de 1917 le régime des autorisations de création, mais, c o m m e elle, reste très discret sur les mesures qui doivent être prises par 1' exploitant; il est simplement indiqué que <
1.2. Les textes particuliers
E n matière de nuisance radioactive, le texte le plus important de la réglementation française est sans doute le décret du 20 juin 1966. Fondé sur les recommendations de la Commission Internationale de Protection Radiologique et les directives émanant de la Communauté Européenne, ce texte définit les «équivalents de dose m a x i m u m admissibles» pour les personnes professionnellement exposées et pour les individus du public et précise qu' il appartient à 1' exploitant de prendre toute m e s u r e pour assurer le respect de ces limites. L e décret définit aussi les concentrations m a x i m a l e s admissibles dans 1' air et dans 1' eau, pour les travailleurs et pour la population, concentrations dont le respect garantit le non dépassement des équivalents de dose imposés. Malgré 1' apport important que rèprésentait ce décret, il est assez vite apparu qu' il manquait encore de précision quant aux modalités d'applica tion et qu' il subsistait certaines lacunes eu égard aux perspectives de déve loppement du domaine de 1' électro-nucleaire. Le prem ier point a été partiellem ent résolu par un systèm e de con ventions anticipantd' ailleurs sur le décret de 1966 passées entre exploitant et autorités de santé publique qui fixent à partir des CMAair et des CMAeau la quantité d'effluents radioactifs pouvant être libérés dans le m ilieu ainsi que les m odalités de contrôle. La nécessité d'établir un système généralisé plus cohérent a cepen dant conduit à la préparation de deux projets de décret relatifs 1' un aux effluents gazeux, l'autre aux effluents liquides. Aux termes de ceux-ci, tout exploitant serait conduit d'effectuer préalablement à la mise en service d'une installation nucléaire, une demande d'autorisation de rejet accompagnée d'une étude faisant appa raître les équivalents de dose auxquels les individus du public seraient soumis du fait de ces rejets. L' autorisation lui serait accordée conjointe ment par les ministères concernés; 1' arrêté d'autorisation à durée limitée pourrait imposer à 1' exploitant d'effectuer certains contrôles. IAEA-SM-180/46 39
1.3. Obligations réglementaires
D u point de vue de 1' exploitant, le problème se simplifie dans la mesure où sa seule obligation consiste à respecter les limites de rejets qui lui ont été fixées, donc à fortiori les limites d'exposition du décret de 1966, ainsi qu' à effectuer les contrôles de rejets qui lui seront éventuel lement imposés. Il ne s e m b l e donc pas qu' il soit réglementairement tenu, sauf si notification expresse lui en est faite, d'exercer une surveillance autre que le contrôle de rejet. Cependant il peut y être contraint pour une double raison. D' une part la compléxitè et la haute technicité de certaines installa tions et les incertitudes qui subsistent encore dans les méthodes d'éva luation de 1' exposition des individus ne le mettent peut-être pas totalement à 1' abri de certaines failles; en clair il peut involontairement se mettre en infraction avec la législation s' il y a dépassement des limites d'exposi tion tolérées. D' autre part, dans le contexte brûlant actuel de 1' opinion publique, il n' est pas inconcevable qu' il puisse être m i s en accusation et il doit avoir le souci de se donner les moyens d'assurer sa défense, c' est-à-dire de faire la preuve qu' il n' a pas enfreint la législation. Dans 1' un et 1' autre cas il ne peut trouver de certitude qu' à travers une surveillance appropriée qui lui permette d'évaluer 1' exposition réelle résultante. U ne s' agit donc pas là d'une obligation m a i s d'une garantie dont 1' exploitant peut juger bon de s' entourer. U lui appartient également de décider la nature du programme à développer; on peut penser à priori qu' il ne nécessitera pas de très grands développements compte-tenu, d'une part, de la base annuelle sur laquelle sont fondées les limites d'exposition, d'autre part, de la possibilité de ne s' intéresser qu' au groupe critique dans la m e s u r e où il sera nettement défini. Il convient n é a n m o i n s de souligner deux difficultés: L' exploitant n' est responsable que des conséquences du fonctionne m e n t de sa propre installation. Il doit donc être en m e s u r e d'évaluer 1' augmentation de nuisance correspondante, ce qui impose une bonne connaissance de celles qui préexistaient: d'où 1' importance particulière des «études de point zéro», importance qui peut être parfois négligée en raison de la multiplicité des p r o b l è m e s qui se posent au m o m e n t où elles doivent être réalisées. Pour la m ê m e raison, 1' exploitant doit pouvoir distinguer la nuisance dont il est responsable de celle qui peut é m a n e r d'autres installations. C' est là un problème purement technique, mais qui amène à poser celui de la responsabilité qui pèse, non plus sur 1' exploitant, mais sur les autorités publiques responsables. Il parait en effet logique d'admettre qu' ayant autorisé des rejets d'effluents dans certaines limites, ces m ê m e s autorités portent la responsabilité d'une mésestimation de leurs conséquences éventuelles. Ayant par ailleurs seules la connaissance des dossiers relatifs à tous les établissements nucléaires et celle de 1' évolution de leur fonctionne ment, il semble également qu' elles aient obligation de prendre en charge le contrôle du respect des dispositions réglementaires lorsqu' il y a super position des effets. 40 BRESSON et COULON
On peut imaginer qu' au plan local un partage des tâches permette en évitant un double emploi d'obtenir les renseignements indispensables à chacune des deux parties; au plan national apparaît'déjà la nécessité de 1' existence d'un p rogramme général de surveillance dont la responsabilité incomberait aux autorités publiques.
2. OBJECTIFS D' ORDRE SANITAIRE ET TECHNIQUE
Il est permis de se demander si, dès lors que les contrôles effectués au niveau de 1' installation, particulièrement des effluents qu' elle libère dans le milieu, suffisent à démontrer le respect des dispositions légales imposées à 1' exploitant tant par la réglementation générale que par les prescriptions particulières accompagnant éventuellement 1' autorisation délivrée, il est indispensable de développer un programme de surveillance. A u cours du chapitre précèdent il est apparu que, pour des raisons de responsabilité et de préservation de ses intérêts, 1' exploitant peut-être amené à mettre en place un tel programme. D' autres considérations peuvent également intervenir, qui relèvent du seul plan sanitaire et technique. Pour cela, il convient d'envisager successivement ce qui se rapporte à 1' installation, au site, et enfin au milieu pris d'une façon générale.
2.1. L' installation
L' installation n' apporte en elle-même aucune nuisance pour 1' environne ment tant qu' il s' agit de son fonctionnement normal, hormis celles qui relèveraient de considérations d'esthétique, occupation de terrain, etc., dont 1' évaluation ne rentre pas dans le cadre de cet exposé. Son seul impact direct sur le milieu environnant se situe dans le cadre de circonstances accidentelles. Dans ce domaine il semble que, tout au moins pour les installations françaises, le problème ait été parfaitement résolu. Considérée c o m m e un établissement présentant un caractère dangereux au m ê m e titre que d'autres activités, 1' établissement nucléaire, ou 1' installa tion, est répertoiriêe auprès des services de sécurité locaux. Tout accident dont 1' am p l e u r le justifierait, entraînerait le déclenche ment d'un plan d'organisation de secours (appelé plan «ORSECRAD») mettant en jeu les moyens de secours conventionnel auxquels s' ajouteraient les moyens spécifiques à la détection et 1' évaluation des nuisances radiologiques. Le plan O R S E C R A D implique la mise en œuvre d'un programme de surveillance qui répond parfaitement aux principes généralement admis: localisation des zones et des vecteurs dangereux ou potentiellement dangereux, exécution des mesures appropriées, transmission rapide de 1' information aux responsables. Aux termes d'un texte à paraître, chaque programme de surveillance est adressé à un Comité National d'Experts Médicaux qui a par ailleurs un rôle de conseil quant aux actions éventuelles à entreprendre. Il n' en reste pas moins que la notion de risque dû à 1' installation relève essentiellement des études de sûreté qui ont été développées depuis sa conception jusqu' à son fonctionnement en passant par sa mise en oeuvre. IAEA-SM-180/46 41
La surveillance se situe donc au niveau du bon fonctionnement de tous les dispositifs techniques de filtration, piêgeage, confinement, en particulier en ce qui concerne d'éventuelles fuites.
2.2. L e site
La conception de ce que 1' on appelle «le site» variant selon les inter locuteurs, il n' est sans doute pas inutile de donner la définition qui lui sera attribuée ici. Au plan géographique le site concerne «la configura tion propre du lieu occupé par une ville et qui lui fournit les éléments locaux de la vie matérielle et les possibilités d'extension». En transposant de la ville à 1' installation nucléaire, on peut interpréter cette définition c o m m e caractérisant la zone appropriée qui reçoit et entoure 1' installation et où se situent des interférences directes entre 1' une et 1' autre. C' est bien sûr là que pourront se manifester les effets i m m é d i a t s les plus importants puisque d'une façon générale, les phénomènes de dilution qui interviennent au fur et à m e s u r e que 1' on s' éloigne de 1' installation tendent à en diminuer 1' importance. L e site a pour cette raison été 1' objet essentiel des préoccupations relatives à la protection. Il faut noter que ces préoccupations ont jusqu' à présent été orientées quasi exclusivement vers le risque radiologique, alors que dans le futur elles devront tenir compte également des autres nuisances possibles: la pollution thermique notamment, qui pourra nécessiter une surveillance aussi bien du milieu physique (climatologie) que de la biosphère, et la pollution chimique qui est de moins en moins négligée. Sur le plan du site, les objectifs d'ordre sanitaire ou technique qui sont fixés aux programmes de surveillance sont généralement les suivants:
2.2.1. Vérification des hypothèses de base
Les critères qui permettent aux autorités de fixer à 1' exploitant les limites de rejet auxquelles il doit se conformer sont généralement le fruit d'hypothèses, basées sur les données que 1' on possède sur le milieu et son utilisation, ainsi que des connaissances générales relatives au comportement des radioéléments dans les différents compartiments de ce milieu. M ê m e si dans un certain nombre de cas il a été procédé à des études reproduisant les conditions réelles propres au site concerné, voir à des études in situ, il s' agit le plus souvent d'observations à court terme qui n'ont pas un caractère parfaitement absolu. Il est donc évident qu' il convient de mettre en place, dés le démarrage de 1' installation un p r o g r a m m e de surveillance dont 1' objet est de vérifier le bien fondé des hypothèses de base qui ont été utilisées pour la fixation des limites. P o u r répondre à cet objectif, le p r o g r a m m e devra nécessairement être très développé, couvrant non seulement les maillons des voies de transfert possibles, mais aussi ceux qui auraient pu a priori ne pas être retenus. A u plan géographique, il peut d'ailleurs déborder le cadre du site, dans la mesure où il existe des voies de transfert importantes susceptibles d'affecter des populations très éloignées de la zone d'activité 42 BRESSON et COULON de 1' installation. P a r contre, il doit être limité dans le t e m p s puisque sans objet dès lors que les vérifications jugées nécessaires auraient été faites. Encore faudra-t-il ne jamais perdre de vue que certaines des condi tions ayant été utilisées pour la fixation des limites peuvent changer au cours du temps: modification du réseau hydrographique, modification des techniques de culture et d'élevage, etc. L e p r o g r a m m e de surveillance tel qu' il vient d'être évoqué et que 1' on peut appeler « p r o g r a m m e initial» peut donc avoir besoin d'être repris en fonction de la nécessité de révision des bases d'estimation. Il appartient bien évidemment à 1' exploitant de le réaliser, étant donné qu' il concerne la vérification des hypothèses sur lesquelles il a fondé sa demande d'autorisation: mais les résultats obtenus intéressent de la m ê m e façon les autorités publiques.
2.2.2. Evaluation de 1' exposition des individus
Il s' agit là d'un point important. C o m m e il a été indiqué, sachant que les rejets sont effectués à 1' intérieur des limites permises, que ces dernières ont été fixées de façon à ce que 1' exposition correspondante de la population soit c o n f o r m e aux r e c o m m e n d a t i o n s ou prescriptions d'ordre réglementaire et enfin, ayant vérifié éventuellement par un p r o g r a m m e initial qu' il en était bien ainsi, on peut se d e m a n d e r s' il est justifié, au seul plan sanitaire ou technique d'évaluer 1' exposition résultante réelle. La réponse à une telle question ne peut sans doute être formulée de façon très tranchée et dépend essentiellement des conditions propres à chaque cas. Il a notamment été proposé déjà que 1' établissement d'un programme de surveillance n' apparaissait pas nécessaire dans la m e s u r e où les estimations des doses résultantes ne représentaient qu' une faible fraction des limites acceptables. Par contre, on peut raisonnablement penser que lorsqu' elles sont susceptibles d'être non négligeables par rapport à ces limites, il était bon d'en avoir une estimation aussi exacte que possible. Toute augmentation imprévue, quelle que soit sa cause étant d'autant moins souhaitable qu' elle rapproche 1' exposition réelle de la limite admissible, il est nécessaire d'être en m e s u r e de la déceler et de 1' evaluer. C o m m e en general le nombre d'individus susceptibles de subir une irradiation significative est limité et caractérisé le groupe critique, le programme de surveillance peut fort bien n' être concu que pour lui: il portera donc essentiellement sur les paramétres critiques (radionucléides, vecteurs) et plus spéciale ment sur ceux qui permettent de calculer les doses d'irradiation avec un minimum d'hypothèses complémentaires. Un autre cas où il peut être nécessaire d'estimer les doses d'expo sition est celui de 1' installation située à proximité d'un centre de popula tion important; m ê m e si les doses individuelles demeurent faibles, la dose globale reçue par cette population peut être relativement élevée. Il est cependant généralement très difficile d'en faire une estimation réelle du fait de la multiplicité des situations particulières (déplacements, habitudes alimentaires, origine des produits consommés, etc.). IAEA-SM-180/46 43
2.2.3. Verifications du contrôle des rejets et détection des rejets non prévus
Bien que des contrôles de 1' activité des effluents soient systématique m e n t effectués au niveau du rejet, on ne peut exclure la possibilité d'une erreur matérielle ou humaine entraînant la libération dans le milieu d'une activité plus importante que prévue. D a n s le m ê m e ordre d'idée, on peut concevoir 1' existence de fuites non détectées. Le contrôle a postériori peut dans une certaine mesure fournir des indications permettant la m i s e en évidence de tels «incidents» et, m ê m e si ceux-ci n' ont qu' une probabilité faible de se produire, constituer une garantie supplémentaire pour 1' exploitant. Il permet également aux autorités responsables de la Santé Publique des vérifications des contrôles qu' elles ne peuvent réaliser d'une façon systématique. Il semble donc utile, sinon nécessaire, de prévoir que cet objectif entre dans le programme de surveillance. Celui-ci sera nécessairement limité au vecteur initial, c' est à dire 1' air ou 1' eau, la qualité de la réponse se détériorant au fur et à mesure que 1' on s' en éloigne, en raison de la mise en œuvre de tous les paramètres de transfert. Si le problème parait relativement simple dans le cas d'un rejet liquide, puisqu' il s' agit simplement de contrôler la concentration des effluents dans 1' eau du réceptacle en aval du point d'émission, il est, semble-t-il, plus complexe au niveau des rejets atmosphériques; la dis persion des gaz ou des aérosols est tributaire des conditions météorolo giques instantanées ce qui implique un choix judicieux des points de mesure. 11 est par ailleurs évident que si le p r o g r a m m e est limité quant au choix des vecteurs surveillés et de la zone couverte, il doit par contre mettre l'accent sur la fréquence des prélèvements et des mesures.
2.2.4. Amélioration des connaissances sur le comportement des radioéléments dans le milieu
Aussi indéniable que soit 1' importance des observations in situ pour 1' obtention d'informations réalistes sur les processus de transfert des polluants lorsqu' ils sont introduits dans le milieu, il ne semble pas qu' il puisse y avoir compatibilité entre un programme de surveillance et un p r o g r a m m e qui réponde à un tel objectif. La différence fondamentale qui existe entre eux se situe au niveau du choix des prélèvements, de la fréquence des échantillonnages et des types de mesures à effectuer, le programme type «écologique» étant nécessairement beaucoup plus complet que le programme de surveillance. S' il est convenable qu' un p r o g r a m m e de type écologique puisse aussi servir de prog r a m m e de surveillance, il n' est, par contre, pas très réaliste d'envisager la réciproque. Cela signifie qu' il doit y avoir une définition préalable très nette du rôle que doit jouer le programme à mettre en oeuvre; il est bien entendu possible que dans certains cas, par exemple celui de centres de recherches, 1' aspect scientifique prime sur 1' aspect opérationnel et que le prog r a m m e soit étudié de façon à couvrir les deux aspects. Mais il est clair aussi que les centres de production ne se sentiront pas concernés par cet objectif indirect. 44 BRESSON et COULON
2.3. L'environnement
L' élargissement du problème conduit tout naturellement après avoir envisagé le cas de 1' installation, puis celui du site tel qu' il a été défini, à considérer 1' environnement dans un contexte plus général; il n' est cette fois plus question de traiter d'une installation particulière mais de 1' en s e m b l e des sources d'exposition quelle que soit leur nature. L a part prise p ar les installations nucléaires est, bien entendu, importante et le sera de plus en plus en fonction de leur nombre croissant et des zones de recouvrement possibles. On peut là aussi, selon les circonstances, envi sager 1' établissement d'un programme de surveillance à différents niveaux: le niveau régional (cas d'un bassin fluvial par exemple), le niveau national, et m ê m e le niveau international. Il est clair que le problème dépasse, à ce stade, celui 3e 1' exploitant d'une installation nucléaire, à moins qu' il ne s' agisse d'un exploitant unique portant à lui seul la responsabilité des conséquences du fonctionne m e n t de toutes les installations. Plus généralement, c' est 1' affaire des autorités responsables et particulièrement des autorités de santé publique. Il leur appartient à ce niveau de veiller à ce que les interférences entre diverses sources ne conduisent à des doses individuelles inacceptables pour certains groupes d'individus. Elles doivent également disposer pour 1' ensemble de la population des éléments permettant d'estimer la dose globale qui lui est délivrée et, partant, le détriment correspondant. L' organisation et la réalisation d'un programme de surveillance à 1' échelon national constitue une tâche considérable; sa conception peut être variable selon les conditions propres à chaque pays (agricoles, économiques, humaines, etc.) de la m ê m e façon que sa mise en œuvre, laquelle dépend en outre des attributions de responsabilité et de 1' existence des moyens appropriés. Pour ces raisons, et aussi du fait que le caractère plus général d'un tel p r o g r a m m e 1' exclut quelque peu du cadre de cet exposé, il n' en sera pas traité ici, si ce n' est pour en souligner 1' importance.
3. OBJECTIFS D'ORDRE ECONOMIQUE
Il est deux façons de les considérer. La première consiste à accorder au programme de surveillance un rôle économique actif, dans la mesure où, garant de la sauvegarde des impératifs sanitaires, il rend possible 1' évacuation concertée dans le milieu d'un certain nombre de polluants. Il peut dans ce cas être concu sur la base d'un objectif d'ordre économique dans la mesure où 1' on établit u n bilan entre le coût que représente sa m i s e en oeuvre et le gain qu' il apporte en évitant d'avoir à utiliser des dispositifs techniques d'épuration et de stockage. Sur le plan pratique, une telle conception serait sans doute trop pragmatique et ne pourrait s' accorder avec des principes plus nuancés, notamment celui qui veut que 1' exposition de la population soit aussi réduite que possible; par conséquent, il est difficile d'accorder une telle importance au programme de surveillance. L' autre conception est de ne lui reconnaître qu' un caractère passif c o m m e chapitre de dépenses à inscrire dans les frais généraux; établi IAEA-SM-180/46 45 pour répondre à certaines nécessités il n' est qu' un élément comptable de la rubrique «coût». Ceci i m p o s e pa r contre qu' il soit en aucune façon lié à une notion de rentabilité, et en particulier, qu' il ne soit justiciable d'aucun aménagement destiné à améliorer un équilibre budgétaire. Il ne semble donc pas qu'un objectif économique puisse influer sur 1' établissement du programme de surveillance.
4. OBJECTIF D' ORDRE PSYCHOLOGIQUE ET SOCIAL
Il s' agit là encore d'un aspect très important du problème puisque ainsi qu' il 1' a été dit plus avant, il est manifeste qu' il a largement pesé sur les actions de surveillance entreprises jusqu' à présent. L' idée de base est que le fait d'exercer une surveillance très rigou reuse permet de calmer les inquiétudes, peut-être légitimes, d'une opinion publique particulièrement sensibilisée et d'éviter que celle-ci ne soit à m ê m e de contraindre les autorités gouvernementales à prendre des mesures susceptibles d'entraver le développement du domaine nucléaire. Aux tenants de cette conception s' opposent ceux qui pensent que la sévérité de la surveillance peut, dans 1' esprit de certaines personnes peu averties, être associée à la notion d'importance du risque et qu' ainsi pratiquer 1' une revient à avouer 1' autre. La vérité doit certainement être plus nuancée. Il est vrai que le fait d'attirer trop nettement 1' attention sur le contrôle, donc indirectement sur le risque, peut aller à 1' encontre du but recherché. Il est vrai aussi que «la contestation» prend bien souvent des chemins que la présentation d'une multitude des résultats d'obstruera pas. Mais il ne faut pas cependant pousser cette thèse trop loin et admettre que le public peut exiger que sa protection soit assurée non seulement par des études préalables, mais aussi par des contrôles ultérieurs. Se fonder sur le manque d'intérêt qu' il peut y attacher lorsque la présence d'installa tions nucléaires a été acceptée serait sans doute imprudent car cet intérêt se réveillerait brusquement à 1' occasion du moindre incident; des exemples de cet ordre peuvent sans aucun doute être présentés par de nombreuses nations. Le poids de 1' aspect psychologique ne doit malgré tout pas prévaloir sur les autres objectifs dans 1' organisation de la surveillance à un m o m e n t où 1' utilisation de 1' énergie nucléaire doit pouvoir se m o n t r e r compétitive par rapport aux autres sources d'énergie et il serait regrettable qu' elle se trouve pénalisée injustement aussi bien financièrement que psycholo giquement; la société toute entière risquerait d'en subir le préjudice. Il importe donc que les programmes de surveillance éventuellement nécessaires répondent aux objectifs premiers, qu'ils soient d'ordre réglementaire, technique ou sanitaire, et que l'information du public accrédite 1' idée que tout ce qui doit être fait est réalisé avec rigueur.
CONCLUSION
Les principes relatifs à 1' établissement des programmes de surveil lance au voisinage des installations nucléaires ont été définis dans un cadre très général. Leur application pratique doit tenir compte des réalités 46 BRESSON et COULON et contraintes particulières: type de réglementation, politique suivie en la matière p ar les autorités, caractéristiques de 1' installation, du site, des effluents, contexte psychologique, etc. Il importe surtout que pour des raisons marginales une importance particulière ne soit accordée au développement de la surveillance, laquelle risquerait d'apparaître c o m m e une pénalisation vis-à-vis des autres sources d'énergie et, à la limite, pourrait influer sur 1' avenir du domaine nucléaire. Par contre, il est tout aussi nécessaire que des critères d'ordre économique ne conduisent à négliger les mesures propres à garantir la sécurité de la population. D e tout ce qui précède, il apparaît que les points essentiels pourraient être les suivants:
1) Intérêt certain d'un programme de surveillance préalable, très complet mais à durée limitée, pour 1' établissement du «point zéro». 2) Importance d'un programme de surveillance initial destiné à vérifier le bien fondé des études relatives aux conséquences sanitaires des rejets et la conformité des prévisions faites.
Très élaboré quant au choix des échantillons concernés et aux mesures à effectuer, il est par contre également à durée limitée.
3) Conception du programme de surveillance en routine en fonction, d'une part, des conditions propres à 1' installation et au site, d'autre part, des éléments appropriés par le programme initial.
Le p rog r a m m e de routine devrait porter sur le vecteur initial, air ou eau, afin de vérifier les activités réellement libérées et, le cas échéant, détecter les rejets imprévus. Il devrait également, si l'importance des conséquences estimées le justifie, concerner ceux des éléments à partir desquels pourrait être évaluées de la façon la plus directe les doses réelles reçues par les individus les plus exposés ou exposés de façon significative.
4) Nécessité de disposer d'un programme de surveillance plus vaste, à 1' échelle du pays par exemple, destiné à évaluer 1' exposition globale de la population dans son ensemble.
Adaptées aux conditions réelles, ces bases devraient permettre d ' assurer à la population une garantie nécessaire et suffisante. IAEA-SM-180/8
PRINCIPLES AND PRACTICE OF ENVIRONMENTAL MONITORING IN THE UNITED KINGDOM
N.T. MITCHELL Ministry of Agriculture. Fisheries and Food, Fisheries Radiobiological Laboratory, Lowestoft, Suffolk H.J. DUNSTER National Radiological Protection Board, Harwell, Didcot, Berks
A . W . K E N N Y Department of the Environment, London
E.A.B. BIRSE Scottish Development Department, H M Industrial Pollution Inspectorate for Scotland, Edinburgh, United Kingdom
Abstract
PRINCIPLES AND PRACTICE OF ENVIRONMENTAL MONITORING IN THE UNITED KINGDOM. Environmental monitoring is an essential feature of the control system for radioactive waste disposal in the United Kingdom. Current procedures, the result of more than 25 years' experience, relate primarily to the nuclear power programme as the major source of radioactive waste requiring disposal. The paper outlines the basis of environmental monitoring and the relative roles of central government departments who set the controls, and operators of the sites irom which the disposals are made. It goes on to discuss the objectives of programmes.
part is to provide a check on the adequacy of control measurements. A further and important objective is research, much of it aimed at providing a feedback of information into the control system, and a third objective is the maintenance of good public relations. The paper shows how environmental monitoring programmes are
in both the design and the interpretation of programmes. The functions and value of derived working limits and investigation levels are indicated. The paper concludes with an account of how the principles work out in practice with examples drawn from the nuclear industry. The extent to which purposeful monitoring is required varies widely between sites and between types of waste, the main emphasis being on liquid wastes with little on surveillance of gaseous releases and almost none in respect of solid materials.
1. INTRODUCTION
Environmental monitoring is an essential part of the control system for radio active waste disposait 1J in the United Kingdom and is principally concerned with the nuclear power programme as the primary source of waste requiring disposal. This paper discusses current practices, the culmination of mo r e than two decades of experience.
47 48 MITCHELL et al.
The nature and sources of radioactive waste have been discussed extensively^2, 3,4]. Those in the nuclear power pr o g r a m m e m a y be summarized as falling into four groups according to source: research and development into new and existing reactor systems, fuel element manufacture, nuclear-powered electricity-generating stations and fuel-reprocessing plants. Outside these cate gories radioisotopes are used in medicine, industry and research, but relatively little becomes waste requiring disposal, for mu c h is either of very short half-life or is in the form of solid sources and does not therefore become waste in normal circumstances. Some waste does accrue in the course of the manufacturing pro cesses, but both the quantities involved and their consequences are insignificant compared with those from the nuclear power programme.
Radioactive wastes are disposed of in gaseous, liquid and solid forms but those which are liquid feature most prominently as sources of exposure of indivi dual me m b e r s of the public, and a large proportion of monitoring in the U K is therefore related to this category of waste.
2. THE BASIS OF ENVIRONMENTAL MONITORING
The fundamental characteristic of environmental monitoring programmes, as practised in the UK, is that they are designed with so m e specific object in mi n d and carried out to serve that purpose. U K radioactive waste disposal po l i c y ^ puts the emphasis on public radiation exposure, so that monitoring applied to the control of radioactive waste disposal is directed primarily to assessment of public radiation exposure. The potential risk to environmental resources will be minor if the pub lic radiation exposure limitation is maintained!-6,7 J
The administrative structure of the control system depends on departments of central government, independent of those which are responsible for promoting the use of nuclear energy. In England and Wales there is a joint responsibility for the major nuclear sites - the Ministry of Agriculture, Fisheries and Food, acting with the Department of the Environment in England and with the Welsh Office in Wales - whilst elsewhere a single Department is responsible - the Scottish Office in Scotland and the Ministry of D e v e l o p m e n t in No r t h e r n Ireland. M u c h of the environmental monitoring done to confirm the adequacy of control measures falls to these departments to organize. This is not to say that all such monitoring is done by these departments; the operators of the major sites have their own pro g r a m m e s required as a condition of their authorizations to dispose of waste.
3. OBJECTIVES AND DESIGN OF ENVIRONMENTAL MONITORING PROGRAMMES
Although most of the U K environmental monitoring programmes have a co m m o n basis, in so m e wa y or another being related to waste disposal control, their objectives are mo r e complex, serving a number of purposes. The primary objec tive, related directly to control, m a y be divided into two parts, the first being to provide a basis for the assessment of public radiation exposure, the second a further me a n s of checking the adequacy of control me a s u r e s . A further objec tive is research, m u c h of which is aimed at providing a feedback of information into the control system, whilst another purpose is the provision of public information. IAEA-SM-180/8 49
3.1. The control system
3.1.1. Monitoring to assess public radiation exposure
This is the most fundamental of all the objectives of environmental monitoring, for opportunities for direct measurement of radiation exposure of the public occur so rarely that for all practical purposes they can be ignored. Instead exposure must be assessed from environmental measurements in conjunction with data on the habits of the population concerned.
All environmental monitoring programmes depend to some extent on a process of modelling of the environment, though it is in the assessment of public radiation exposure that it becomes most necessary. The system practised in the U K is the critical path approach, also used as the basis for the control s y s t e m M in setting authorized limits, and this depends on achieving a basic understanding of the wa y in which radionuclides behave after release to the environment, in particular the routes or "pathways" through which contamination of environmental materials and h u m a n radiation exposure occurs. Identification of pathways is the first stage of habits surveys in which the characteristics of exposed populations are studied; this leads to identification of the pathway(s), material(s) and the individual(s) or group(s) of the public which are "critical" and in relation to wh o m control limits on discharges are set. The emphasis to date has been on exposure of individuals since this, rather than the collective dose to large groups of the public, has been found to be limiting in almost every case; nonetheless, analysis along exposure pathways provides the model for estimation of collective dose to the whole of the population exposed to the effects of a specific disposal - which in an extreme case would be the world-wide population.
Whilst the environmental model describes a number of stages in the pathway, monitoring for the purpose of estimating public radiation exposure will almost invariably concentrate on the penultimate stage in the chain, sampling the critical foodstuff for radionuclide analysis - when the pathway involves internal exposure - or measuring the ambient radiation dose-rate in the case of an external exposure pathway. The pr o g r a m m e of measurement can usually be restricted to a small number of points in one, or at most only a few, pathways; further details will be found in the examples quoted in section 5.
3.1.2 Monitoring for other control purposes
All important discharges - those sufficient to generate significant public radia tion exposure - are under constant surveillance as a result of monitoring of critical pathways. However, this is not the only role of environmental monitoring in a waste disposal control system which aims not merely at limiting exposure within specified limits but also at minimizing exposure as far as is reasonably practi cable. Monitoring also provides a me a n s of detecting avoidable releases and though in the main this is achieved by careful sampling and analysis of radioactive discharges, releases ma d e unwittingly can be discovered by environmental mo n i toring procedures, by sampling a non-critical indicator material. Well-known examples are the algae used in the aquatic environment, inedible seaweeds and freshwater mosses, and the "tacky shade" system used to monitor deposition from the atmosphere. 50 MITCHELL et al.
3.2. Research
The existence of radioactivity in the environment, both natural and artificial, has provided wide opportunities for research, though in the context of this paper discussion will be confined to work which has a direct impact on the conduct of the control system.
In depending on the prior assessment of environmental capacity, the U K control system depends heavily on data being available of the environmental behav iour of a range of radionuclides in a variety of materials. Particularly important are data on concentration factors, defined as [concentration in the material/ concentration in water] at equilibrium, and while they ma y be of no direct value in controlling the discharges from the site from which they were derived, they do play an invaluable role in making possible better estimates of environmental capa city elsewhere. In a similar way data on dispersion built up from experience with actual discharges help to verify the hydraulic models used in making pre- operational estimates of discharges.
One of the main objects of the authorizing departments, once a discharge has been under wa y long enough to create measurable contamination, is to reassess the environmental capacity. This requires a mo r e extensive monitoring pr o g r a m m e than that necessary to assess the degree of radiation exposure resulting from the discharges and must be organized in conjunction with mo r e detailed analysis of the effluent than is usually specified for control purposes. Measurements will usually be of critical nuclide(s) and critical material(s), though occasionally so m e value can be drawn from use of indicator materials, in which case the nuclide need not necessarily have any radiological significance.
Research monitoring based on indicator materials has further uses, espe cially in the stage of development of a site when discharges have no detectable effect on critical materials. A s well as providing a mo r e sensitive means of esti mating public radiation exposure, indicator monitoring can give useful advance warning of impending changes which ma y have s o m e bearing on public radiation exposure.
3. 3. Provision of information to the public
Gaining the confidence of the public in the operational safety of nuclear sites is an essential prerequisite to their acceptance of nuclear power. In the context of safety of radioactive waste disposal environmental monitoring pr o g r a m m e s occupy an important role, though it is not U K policy to run special surveys solely for this purpose but rather to utilize existing monitoring programmes. Even where there is adequate scientific evidence to rely solely on effluent analysis - where dis charges are insufficient to generate measurable contamination and a significant level of public radiation exposure - it is usually found worth while to continue a modest programme of monitoring to provide unequivocal evidence of the safety of discharges. Such monitoring continues to be designed on the s a m e scientific principles as adhered to elsewhere, and surveys are based on the potentially criti cal exposure pathways.
It has become standard policy, especially by the operators of nuclear power stations, to maintain contact with the local public through groups known as local IAEA-SM-180/8 51 liaison committees, composed of the elected representatives and officials of local government. One wa y in which confidence has been established is by furnishing these representatives of the public with information from monitoring programmes of the environmental and public health consequences of discharges; meetings have also become a forum for explaining impending changes in, for instance, monitoring procedures, for in isolated rural areas typical of ma n y of the nuclear sites in the UK, changes in the established monitoring pattern are soon noted and discussed by the indigenous population.
4. THE INTERPRETATION OF MONITORING PROGRAMMES
For monitoring pro g r a m m e s destined to be used to estimate public radiation exposure the method of interpretation is inherent in their design. Data from the habits survey which identified the critical pathway (and hence the critical material being sampled) are used to convert levels of contamination or ambient dose-rates into rates of exposure. This method applies equally to individuals, to critical groups and, by extension of the environmental model, to large populations - the main difference being the value adopted for consumption rate or occupancy, although the geographical distribution of large populations is likely to be so wide that their exposure cannot be based on one single value of contamination level in a material or the dose-rate at a single location.
In this way the environmental "exposure pathway" model which has been the basis for control, including the design of the monitoring programme,can be utilized to evaluate the significance of any series of measurements, comparing the calcula ted rate of exposure against the primary standards, the IC R P-recommended dose limits. Use of this method means going back to first principles, whilst in practice m o r e frequent use is ma d e of a secondary standard known as the derived working limit (DWL), calculated from IC R P dataL 9J. The D W L for internal exposure is, strictly, a tertiary standard since it is calculated from permissible daily rates of intake of radioactivity (already a secondary standard) which would deliver exposure at a rate equal to the ICRP-reco m m e n d e d dose limit.
The D W L is a very convenient working tool and has been used widely through out the UK. For the purposá of assessing the significance of monitoring data it m ay be compared against individual measurements, though a more common and meaningful practice is to use annual averages of contamination level or dose-rate, so reducing inevitable fluctuations to manageable proportions and computing a value - the fraction of the I C R P-recommended dose limit - which is consistent with the mi n i m u m period (1 year) to which the recommendations themselves refer.
More complicated environmental models are needed to deal with situations after isolated events, such as nuclear explosions or reactor accidents. It is necessary for such models to represent the dynamic behaviour of the environment and in complex cases this would require the mathematical techniques of systems analysis proposed in so m e countries, though it has not yet been necessary to use these techniques in the UK. If sufficient data about the environment can be obtained or postulated, it will still be possible to relate the transient concentration in a compartment of the model to the dose commitment to me m b e r s of the critical group. For accidental releases of materials that mo v e rapidly through the environment, the ma x i m u m concentration in a suitable compartment, such as milk, m a y be the 52 MITCHELL et al. most convenient quantity to relate to exposure, an approach which has been used in f environmental monitoring pro g r a m m e s following accidental
In principle the use of the D W L can be applied to measurements of an indi cator material, i.e. to a material not in the critical pathway. In practice, however, this requires a high degree of realism in the environmental model. A n alternative approach is analogous to that used by the I C R P in relation to individual monitoring for internal contamination!-HJ. In the environmental situation, this method involves selecting a fraction, perhaps a few per cent, of the dose limit to the critical group and calling this an Investigation Level. The environmental model is then used to estimate the value for the indicator measurement corresponding to this investigation level. Since this value depends not only on the selection of the investigation level but also on the environmental model, it ma y be termed a Derived Investigation Level. A s long as results in the indicator material remain below the derived investigation level, the simple environmental pr o g r a m m e can be continued. If, however, results rise above the derived investigation level, the reason must be sought and, if necessary, a mo r e conventional monitoring pro g r a m m e aimed at assessing doses must be brought into effect. Although little deliberate use of this concept has been ma d e in the UK, the emphasis having been on the D W L , intuitive use is made, with rather arbitrarily chosen values, to dis card trivial results.
5. MONITORING IN PRACTICE
5.1. Waterborne discharges
Discharges of liquid waste are ma d e to several sectors of the aquatic environ ment - to lakes, rivers and estuaries and to coastal areas of the sea. Discharges from the major sites - those associated with the nuclear power industry - are usually ma d e to coastal and estuarine waters, to take advantage of the large degree of dispersion available; particular exceptions are the nuclear power station on the shore of Lake Trawsfynydd and the establishments such as Harwell in the Thames Valley which release treated waste into the River Thames. The following examples indicate the range of situations that exists in the U K and the environmental moni toring procedures which have been devised to cater for them. Detailed discussion is contained in reports by MitchellL 12 J.
5.1.1. Freshwater rivers and lakes
Although both these types of environment are liable to be sources of drinking water, this is not often the critical pathway for exposure of the public. The River Thames, receiving discharges such as that from Harwell, provides an example where drinking water is the critical pathway. A s a part of the management of waste on site, the effluents are monitored before discharge, and as a part of the legal control, monthly limits are set to which the establishments must conform. Govern ment control comprises spot checks and ma y include random sampling of effluents. The authorizations permitting the discharges of wastes from these establishments to the River Th a m e s were designed to limit the concentration of activity in the river immediately below the discharge points to levels, based on Medical Research Council recommendations, at which the river water would be radiologically accep table for drinking. Monitoring consists in (1) the check monitoring referred to ÍAEA-SM-180/8 53 above, (2) quarterly determinations of radiostrontium and radiocaesium on bulk daily samples of London's drinking water drawn from the Thames, (3) full radio chemical analysis of a sample of this water about once a year, and (4) occasional analyses of river mu d just downstream of the outfalls. Interpretation of monitor ing results is complicated by the presence of fallout from testing of nuclear weapons. Despite the fact that radiostrontium in the drinking water wa s well below the recommended ma x i m u m levels suggested by the ICRP, treatment of the waste was modified to remove this element and thus reduce the amount ingested. This action was in accordance with the general principle of reducing radiation exposure of the public wherever reasonably practicable.
In sharp contrast, the critical pathway from discharges to Lake Trawsfynydd from the ma g n o x power station on its shores is fish - notably trout, as a source of which the lake is well-known. This is a result of the combination of two particular factors - the presence of caesium-137 and -134, as major constituents of effluents, and the very low hardness of the lake water; in these circumstances radiocaesium concentrates in fish to a high degree. In fact drinking water is not abstracted from the lake or from the outflow, but even if it were it would still be only a minor path way compared with fish consumption.
Monitoring for basic radiological control purposes consists of regular s a m p ling of trout and perch. Samples are analysed for total beta activity, following which specific radionuclides are estimated as necessary. These currently consist of the critical radionuclides, caesium-137 and -134, which are analysed by g a m m a spectrometry, and strontium-90, a minor contaminant, whose main source is cur rently fallout. In addition, a wider-ranging pr o g r a m m e of research monitoring is undertaken involving lake water and lake bed, important stages in the pathway, illustrating the type of studiesL13j undertaken to provide data for reassessment of environmental capacity.
5.1. 2. River estuaries and the sea
It is appropriate to pair these environments, for the river estuaries used for disposal of liquid wastes have mu c h in co m m o n with the sea, being largely saline waters, and monitoring procedures follow a similar pattern for each. There are two types of major establishment sited on estuaries in the U K - concerned respec tively with fuel fabrication and nuclear power generation. A s a near-typical example of the latter, Bradwell power station releases low-level wastes after treatment into the Blackwater Estuary. Effluent composition is complex but a single pathway is unequivocally the critical route to public radiation exposure, the critical material being oysters, for which the estuary is well-known. In early years zinc-65 was the critical nuclide, later to be replaced in this role by silver-110m. The monitoring pr o g r a m m e consists primarily of regular sampling of oysters which are analysed for gross beta radioactivity and the gamma-emitting nuclides. Other contaminants are found as well as 65zn and H O m ^ g but make no significant contribution to public radiation exposure. In the sa m e wa y as for Trawsfynydd, extensive r e s e a r c h ! - 14 J hag ma d e possible the reassessment of discharge limits as well as providing a useful data bank for application elsewhere.
None of the power stations on open coasts produce any measurable contami nation, for although environmental capacities are large the needs of the establish ments are quite small. Nevertheless, monitoring programmes are based on the 54 MITCHELL et al.
(potentially) critical pathways and are modest in scale - typically involving sampling of fish or shellfish - and measurement of the ambient g a m m a radiation dose-rate on beaches at no mo r e frequent intervals than 3 months. Occasionally, such pro grammes are supplemented by adding an indicator seaweed which can improve the sensitivity of detection of any contamination which ma y be present.
By far the largest sources of liquid radioactive waste to the marine environ ment, or indeed elsewhere in the UK, are the two sites where fuel is reprocessed. The largest is at Windscale, Cumberland, which receives irradiated fuel from the nuclear power programme and from which low-level liquid waste is discharged to the Irish Sea. The other is at Dounreay, Caithness, which takes oxide fuel from materials-testing reactors and the U K fast breeder reactor development programme and for which the Pentland Firth is the receiving environment. Both show some interesting features of exposure pathways which are unique in U K experience to date. There are three important pathways for Windscale discharges. The first and longest-standing is the result of harvesting of the locally-grown seaweed Porphyra which is manufactured into the foodstuff laverbread, critical nuclides being ruthenium-106 and cerium-144. Monitoring is complex and has been discussed, along with the implications for public radiation exposure, by Preston and JefferiesLlS, 16]. Samples are collected monthly from a range of points on the nearby coastline, and analysed primarily for total beta radioactivity, and for the more important gamma-emitters, ruthenium-106 and cerium-144, which are criti cal for exposure of the GI Tract. Strontium-90, plutonium-239/240 and americium-241 are also analysed because of their contribution to bone exposure. This takes care of control monitoring, though for assessment of actual exposure the laverbread product is sampled at the time of sale on the retail market, because dilution with uncontaminated weed from other areas occurs during manufacture. The degree to which this occurs cannot be predicted with complete confidence; occasionally laverbread manufactured from undiluted Cumberland weed has appeared for sale, so that control has been largely on the local seaweed.
A second important pathway involves external exposure due to contamination of the foreshore, especially areas where m u d and fine silt collect, the critical nuc lides being the gamma-emitters zirconium-95/niobium-95 and ruthenium-106. Basic radiological control monitoring consists of regular g a m m a dose-rate surveys, especially concentrating on those areas where the dose-rate is highest (a nearby river estuaiy), and like the previous pathway, this pathway has been the subject of extensive research.
A third pathway for Windscale discharges of increasing importance in recent years involves fish, the critical nuclides being caesium-137 and -134. Representa tive sampling is difficult because of the size of the fishery, and though the highest contamination is found in samples in the immediate vicinity of the discharge point, the main emphasis has been on commercial landings supplemented by special sur veys. Analysis is usually required only for caesium-137 and -134, because other radionuclides rarely reach detectable levels of contamination. This is a pathway for which total population dose is no w estimated, as it has become the largest source of such exposure from marine waste disposals; for this purpose m e a n levels of contamination in commercial sources of fish are combined with statistics on total landings. IAEA-SM-180/8 55
The two critical pathways for Dounreay also show interesting features not found elsewhere. Both pathways are external, the first being the result of contami nation of fishing gear by particulate material which concentrates fission products, especially cerium-144, generating mainly beta radiation exposure of the fishermen's hands. Monitoring consists of measurement of the dose-rate from the nets, supple mented by research into correlation of net dose-rates and discharge rates of the m o r e important radionuclides.
The second pathway is the result of suspended matter carried into rocky clefts in the coastline and left behind by the receding tide, generating a field of g a m m a - radiation exposure to those very few wh o use these beaches. Monitoring takes the form of regular dose-rate surveys at selected points, with less frequent g a m m a spectrometric analyses. Interpretation is difficult because of uncertainties in the occupancy rate, and the significance of contamination is assessed on somewhat pessimistic assumptions as to the ma x i m u m conceivable value.
5. 2. Airborne discharges
The great majority of users of open sources of radioactive materials require small amounts of fairly innocuous radionuclides in operations where only a small fraction could be dispersed to the environment as an airborne discharge. Thus, although in practice each case must be considered individually, there is usually no need to carry out any kind of environmental survey in relation to possible airborne releases. There are, however, a number of establishments where large amounts of radioactive material exist. These are varied in character and include (1) estab lishments at which nuclear reactors are operated for power production or research purposes, (2) establishments at which large amounts of radioactive material are handled (e.g. fuel-reprocessing plants, laboratories for the commercial production of radioactive materials, plants for the production of nuclear fuels and fuel ele ments for reactor systems), and (3) establishments at which large amounts of radioactive materials are used for research purposes.
Environmental monitoring programmes have been developed for specific establishments and are related to the radioactive materials used and the possible routes by which they m a y result in radiation exposure of the public. These pro g r a m m e s are based on an identification of the critical radionuclides involved and the critical pathways of exposure. T w o examples illustrate the principles involved.
The first example is from fuel-reprocessing - the plants at Windscale and Dounreay - which both process irradiated fuel containing radioiodine. The con trolling Departments require that the irradiated fuel be stored until the amount of radioiodine is reduced to an acceptably low level, since radioiodine release would contaminate pasture that ma y be grazed by dairy cattle and cause contamination of milk and irradiation of the thyroid glands of consumers. T o evaluate the radiation dose by this particular route, fortnightly milk samples are taken from farms in the vicinity of the establishments and analysed for iodine-131.
The major constituent of releases to atmosphere from these plants is krypton-85 though, like noble gases elsewhere (e. g. argon-41 from nuclear power stations), it has not produced any significant radiation exposure of the public and no routine monitoring is undertaken. Indeed, an analysis [l?l of the consequences of release of krypton-85, both locally and on a world-wide basis, indicates that 56 MITCHELL et al. present levels of radiation exposure - including that from fallout which is currently the largest source - are only a minute fraction of IC R P-recommended dose limits. However, a modest programme of monitoring is contemplated to verify existing levels and because of its world-wide interest.
The second example is the fuel-element manufacturing plant at Springfields in Lancashire, which handles large amounts of uranium. Various operations result in airborne releases of uranium that can potentially contaminate pasture in the vicinity of the establishment. A n analysis of the situation revealed that grazing animals are at greater risk than the public, the hazard being the chemical toxicity of uranium, m u c h greater than that presented to the hu m a n population by the con tamination of air, food or water. The first monitoring programme consisted in sampling herbage, to be replaced by twice-yearly sampling of bovine faeces to per mit a mu c h better assessment of the intake of grazing animals.
The advantages of such pr o grammes are that they are logical and scientifi cally satisfying and that they permit estimates of the radiation dose to which members of vulnerable groups may be exposed. Nevertheless, the emphasis they place on certain radionuclides and pathways can easily divert attention from unexpected discharges of radionuclides that could be significant in giving early warning of maloperation or deterioration of so m e part of an installation.
Monitoring programmes that permitted estimates of radiation doses were developed and used in Great Britain in the early stages of operation of ne w instal lations or establishments for which no previous experience existed. However, with one or two minor exceptions, no radioactive contaminants were found in the environment which could be attributed to the operation of these establishments, and no appreciable increase in the radiation exposure of any me m b e r of the public occurred as a result of airborne discharges. In these circumstances, elaborate and costly environment-survey pr o grammes designed to establish the exposure of local population groups are clearly unjustified. Instead, sampling systems are being installed, designed to indicate that no change in the situation is occurring and consisting of muslin impregnated with "tacky" substances to act as precipitation collectors located at convenient sites around each major establishment^ 18,19], The indications are that such systems, and particularly the tacky muslin, are cheaper and simpler to operate and that they are mu c h mo r e sensitive detectors of releases of a wide range of radionuclides than are the m o r e usual biological samples. If these detectors indicate that airborne discharges are occurring, a full quantitative assessment of the significance of the discharge can follow. If there is a possibility that the discharge could cause an appreciable increase in the radiation exposure of the local population over natural background, suitable monitoring pro g r a m m e s aimed at determining the radiation exposure can be developed. More probably, the source of the discharge can be identified and eliminated.
5. 3. Solid waste disposals
The amounts of radioactive materials associated with solid wastes vary over a very wide range and various me a n s are used for their management and disposal.
Where practicable, solid waste containing small quantities of radioactive material is disposed of with ordinary refuse on to public tips, achieving some dis persion in the ordinary refuse of the establishment where practicable. Government IAEA-SM-180/8 57 inspectors check refuse during unannounced visits, and samples are taken for analy sis to ensure compliance with the standards. Occasionally g a m m a dose-rate surveys of refuse tips are carried out and samples collected of drainage from them.
Wh e r e disposals on refuse tips are not practicable or where the levels of radioactivity are too high, the waste is disposed of at moderate cost through government land-disposal facilities. Sites are selected with considerable care to ensure that (1) any water draining from them will not affect drinking-water supplies and (2) the public will not be exposed to radiation in any way. Very few sites of this kind exist in Great Britain. They are securely fenced, and the drainage from them is monitored regularly. Radiation-dose measurements in adjacent areas accessible to the public are also ma d e occasionally.
S o m e operators do not dispose of radioactive waste but, instead, store it in specially constructed pits or silos, the essential feature of which is that the waste is contained in a structure of concrete or other suitable material. These pits or silos are within the site boundaries, so that the public has no access to them, and are so organized that the waste could be recovered for disposal should this ever become necessary.
Waste not suitable for disposal in any of the above ways is packaged in suit able containers and disposed of at sea. Disposals are ma d e in areas selected to meet criteria designed primarily to minimize the accidental recovery of the con tainers, especially by commercial fishing operations. To this end all dumping areas are in water at least 2000 m deep - beyond the edge of the continental shelf - and are also away from cable routes. Wastes are packed in containers designed to reach the sea bed intact without rupturing, and manufacturing processes are subject to inspection to ensure that the rigid standards set are observed. Simple critical pathway-type hazard assessments! 2o] carried out for these disposal areas show that the rate of return of radioactivity to m a n through the food chains excludes any possi bility of the detection of any radionuclides, let alone any risk to the population, so that no environmental monitoring is attempted.
6. SUMMARY AND CONCLUSIONS ,
Environmental monitoring related to radioactive waste disposal has, since its inception, been based on the principles of the critical pathway approach, so that sampling pr o g r a m m e s have concentrated on the critical material, with the prime emphasis in analysis being on the critical nuclide(s). This is reinforced by other monitoring of the environment to ensure stringent control of radioactive waste dis posal, including efñuent analysis, though not, in general, in great detail except in conjunction with research.
The emphasis has therefore been on the environment, because control which has included monitoring of the environment has proved to be the most effective and reliable wa y of showing that the U K policy objectives are met. Indeed, the approach that relies on ma x i m u m permissible concentrations related to drinking water would, if applied in the UK, have led to unacceptable levels of contamination in so m e path ways. Control by monitoring the critical pathways has also proved to be very economical, especially so on a value-for-effort-expended basis, but, even mo r e important, it has enabled us to maintain very high standards of radiation protection of the public from radioactive waste disposal. 58 MITCHELL et al.
REFERENCES
[ l] KENNY, A. W . , MITCHELL, N. T., United Kingdom waste management policy, Management of Low- and Intermediate-Level Radioactive Wastes (Proc. Conf. Vienna 1970), IAEA, Vienna (1970) 69. [ 2] PRESTON, A. etaL, Int. Conf. peaceful Uses Atom. Energy (Proc. Conf. Geneva 1971) 11, UN, N e w York (1972) 415. [ з1 D O U G A L L , I., N E W E L L , J. E . , Management of low- and intermediate-level radioactive wastes at the nuclear power stations of the United Kingdom Central Electricity Generating Board, Management of Low- and Intermediate- Level Radioactive Wastes (Proc. Conf. Vienna 1970), IAEA, Vienna (1970) 275. [ 4] NEA, Radioactive Waste Management Practices in Western Europe, OECD, Paris (1972). [ 5] The Control of Radioactive Wastes, Cmnd 884 (1959) HMSO, London. [ б] W O O D H E A D , D. S., The biological effects of radioactive waste (Proc. Roy. Soc. Lond. (B)) 177 (1971) 423. [ 7] W O O D H E A D , D. S., Levels of radioactivity in the marine environment and the dose commitment to marine organisms, Interaction of Radioactive Contami nants with Constituents of the Marine Environment (Proc. Symp. Seattle 1973), IAEA, Vienna (1973) 499. [ 8] P R E S T O N , A . , M I T C H E L L , N. T . , The evaluation of public radiation expo sure from the controlled marine disposal of radioactive waste (with special reference to UK ) , Interaction of Radioactive Contaminants with Constituents of the Marine Environment (Proc. Symp. Seattle 1973), IAEA, Vienna (1973) 575. [ 9] P R E S T O N , A . , The United Kingdom approach to the application of IC R P stan dards to the controlled disposal of radioactive waste resulting from the nuclear power programme, Environmental Aspects of Nuclear Power Stations (Proc. Symp. N e w York 1970), IAEA, Vienna (1971) 147. [10] B R Y A N T , P. М., The derivation and application of limits and reference levels for environmental radioactivity in the United Kingdom, Health Physics Aspects of Nuclear Facility Siting (Proc. Midyear Topical Symp. Hlth Phys. Soc. ) Щ (1971) 634. [11] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Report of Committee IV on Evaluation of Radiation Doses to Body Tissues from Internal Contamination due to Occupational Exposure, I C R P Publication 10, Pergamon Press, Oxford (1968). [12 ] M I T C H E L L , N. T . , Radioactivity in surface and coastal waters of the British Isles 1971, M A F F Fisheries Radiobiological Laboratory, Lowestoft, Tech. Rep. F R L 9 (1973). [13] PRESTON, A. , JEFFERIES, D. F. , The radioecology of Lake Trawsfynydd (In press). [l4l P R E S T O N , A . , The control of radioactive pollution in a North Sea oyster fishery, Helgoländer wiss. Meeresunters Г7 (1968) 269. [1 5 ] P R E S T O N , A., JEF F E R I E S , D. F., The assessment of the principal public radiation exposure from, and the resulting control of, discharges of aqueous radioactive waste from the United Kingdom Atomic Energy Authority factory atWindscale, Cumberland, Hlth Phys. 13 5 (1967) 477. [l6l PRESTON, A . , JEFFERIES, D. F., The ICRP critical group concept in relation to the Windscale sea discharges, Hlth Phys. 16 1 (1969) 33. [17 ] DUNSTER, H. J., W A R N E R , B. F., The disposal of noble gas fission pro ducts from the reprocessing of nuclear fuel, U K A E A Hlth and Saf. Branch Rep. AHSB(RP)R101 (1970) HMSO, London. IAEA-SM-180/8 59
[l8l GARLAND, P. J., LOVETT, M. B., WILSON, R. B., A simple system for the rapid determination of airborne radioactivity, Rapid Methods for Me a s u r ing Radioactivity in the Environment (Proc. Symp. Neuherbergl971), IAEA, Vienna (1971) 723. [19 ] JONES, J. K. et al., The experience of the Central Electricity Generating Board in monitoring the environments of its nuclear power stations, Hlth Phys. 24 6 (1973) 619. [20] W E B B , G. A. M., M O R L E Y , F., Assessment of limits for deep ocean dis posal of radioactive waste, Nat. Rad. Prot. Board Rep. N R P B - R 1 4 (1973) HMSO, London.
IAEA-SM-180/5
OBJECTIFS DE LA SURVEILLANCE D'UNE INSTALLATION NUCLEAIRE
E . NAGEL IFR, Institut fédéral de recherches du génie nucléaire, Würenlingen, Suisse
Abstract-Résumi
THE OBJECTIVES OF MONITORING A NUCLEAR INSTALLATION. The regulations for the protection of population in the vicinity of a nuclear installation are becoming increasingly stringent. The personal dose rates permitted during normal operation are very low, so much so that direct measurements are practically impossible. For surveillance purposes one can rely, in the case of more or less continuous gas releases, on average annual radioactivity concentrations at the source itself; these measurements are relatively easy to make, but the calculation models used to derive individual doses received in the vicinity provide only approximate values. Temporary build-ups of emission cannot be reflected in the calculation of annual averages unless they are relatively frequent. Otherwise they have to be considered separately, because during emissions of short duration the dispersion factor at a particular site can differ by one or two orders of magnitude from the .average annual factor. This can, moreover, substantially alter the critical population dose rate. Moreover, there is a growing trend towards the construction of nuclear power stations in the vicinity of densely populated areas, which could result in a steady increase in the total population dose as well (man.rem). This will also have to be taken into account in the future.
OBJECTIFS DE LA SURVEILLANCE D'UNE INSTALLATION NUCLEAIRE. Les directives de protection de la population dans les environs d'une installation nucléaire sont de plus en plus sévères. Les débits de doses individuels que l'on veut encore admettre en temps d'exploitation normale sont très faibles, si bien qu'il est pratiquement impossible de les mesurer directement. Pour assurer la surveillance on peut se référer, pour les effluents gazeux et en cas d'émission plus ou moins constante, aux valeurs moyennes annuelles de la concentration de la radioactivité à la source même: la mesure en est plus facile. Malgré cela, les modèles de calcul devant permettre d'en déduire les doses individuelles reçues dans les environs ne peuvent fournir que des valeurs approximatives. Des émissions accrues passagères ne peuvent être comprises dans le calcul des moyennes annuelles que si elles sont relativement fréquentes. Dans le cas contraire, il faudra les considérer séparément car, lors d'émissions de courte durée, le facteur de dispersion peut, dans un site déterminé, différer de un à deux ordres de grandeur par rapport au facteur annuel moyen. Il peut, en outre, provoquer une modification importante du débit de dose de la population critique. D'autre part, on constate une tendance toujours plus grande à construire des centrales nucléaires à proximité de zones à forte densité de population, ce qui aura pour effet que la dose de la population totale (m an-rem) augmentera, elle aussi, toujours davantage. Il faudra donc en tenir compte à l'avenir.
Les directives de protection de la population dans les environs d'une installation nucléaire se font de plus en plus sévères. Les débits de dose individuels que l'on veut encore admettre en temps d'exploitation normale sont très faibles et ne comportent que quelques millirems par an, si bien qu'il devient très difficile de les m e s u r e r directement. Des progrès techniques ont permis ces derniers temps de fabriquer des instruments suffisamment sensibles avec lesquels on peut détecter des débits de dose y allant jusqu'à quelques ^rems par heure. Ces appareils — ce sont des chambres à ionisation pressurisées — peuvent être utilisés avec succès notamment pour la surveillance des environs d'une source d'émission d'effluents gazeux. On est en outre à m ê m e de mesurer les doses dues aux rayonnements d'origine naturelle et de les déterminer séparément des doses
61 62 NAGEL provenant des sources de radioactivité artificielle. L o r s de l'émission plus ou moins constante d'effluents gazeux, certains types de dosimêtres permettent eux aussi de déterminer des débits de dose assez bas mais seulement après un temps d'exposition prolongé. C'est le cas des dosimêtres à thermoluminescence au fluorure de calcium en particulier; ils ne sont pourtant pas assez sensibles pour enregistrer, par exemple, des doses annuelles de 5 m r e m s seulement, attendu qu'il ne serait pas possible de les séparer des effets de la radioactivité d'origine naturelle. L e s débits de dose provenant de cette dernière doivent dans tous les cas être déterminés par des mesures parallèles à des points qui ne sont pas touchés par des émissions de radioactivité artificielle. Lorsqu'on utilise les deux moyens de détection qui ont été décrits, on peut assurer une surveillance assez poussée de toute une région aux environs d'une source de radiation importante, d'un réacteur nucléaire par exemple. Pour évaluer à l'avance les débits de dose dans des zones qui seront très probablement touchées par des effluents radioactifs, de préférence s'il s'agit d'une source d'émission continue, on peut recourir à une méthode indirecte, en calculant les doses probables en utilisant des m o d è l e s de calcul appropriés. Certains modèles ont été publiés, en particulier, par le Ministère de l'éducation et de la science de la République fédérale d'Allemagne. P o u r obtenir par ce m o y e n des résultats fiables, on devra connaître les valeurs moyennes de la concentration des effluents radioactifs à la source, portant sur une période suffisamment longue. D'autre part, il faudra disposer de données assez précises des conditions météorologiques régnant au site et concernant spécialement la fréquence des directions du vent. A défaut, on pourra se référer sans grand d o m m a g e aux résultats d'une station voisine si elle se trouve dans la m ê m e région climatologique et si on a tenu c o m p t e des particularités locales (vents locaux, topographie) qui, dans un pays c o m m e la Suisse, peuvent jouer un rôle qui n'est pas négligeable. A l'Institut fédéral de Würenlingen on est, depuis peu, en train de comparer les différentes méthodes d'évaluation. D'une part, on mesure avec une chambre à ionisation très sensible et avec des dosimêtres à thermo luminescence (CaF 2 ) les «doses ambiantes» â divers points aux environs d'une source de ^ A r qui est, en l'occurrence, le réacteur à eau lourde de l'Institut et qui est partiellement refroidi à l'air. Les points de mesures ont été choisis en fonction des vents dominants de la région. D'autre part, on a calculé les doses ambiantes auxquelles il faut s'attendre aux m ê m e s points des alentours pendant des conditions normales d'exploitation du réacteur, en se référant à la fréquence moyenne des vents et à la concentration de ^ A r dans la c h e m i n é e d'évacuation de l'air. C e faisant, on ne peut pas tenir compte d'émissions accrues passagères et irrégulières. Leurs effets doivent être considérés séparément. Les doses ambiantes, qu'elles soient mesurées ou calculées, sont en s o m m e des valeurs théoriques. Elles supposent que des personnes seront continuellement s o u m i s e s aux effets d'un effluent radioactif, c'est-à-dire qu'elles se tiendront constamment au m ê m e emplacement, à la m ê m e distance dans la m ê m e direction de la source. D e plus, la dose ambiante n'est valable qu'à l'extérieur de toute protection, en plein air, à tout endroit librement accessible. Il serait peut-être plus réaliste de remplacer la dose ambiante par une conception de « m a n - r e m réel» qui tiendrait davantage compte du cours de la vie de la population en général que de situations extrêmes et théoriques, m ê m e si ces dernières sont plus simples à définir. IAEA-SM-180/5 63
A ce propos on pourrait citer, c o m m e exemple, le cas d'un petit groupe de personnes habitant à 300 m du réacteur de l'Institut de W ü r e nlingen que nous avons cité plus haut, qui rejette continuellement, en t e m p s d'exploita tion, une certaine quantité de ^ A r dans l'atmosphère. Les maisons dans lesquelles ces personnes habitent sont situées dans l'axe d'un des vents dominants du site. Les mesures ainsi que les calculs montrent que la dose ambiante due au rayonnement y de ^ A r au point critique doit être de l'ordre de plusieurs dizaines de millirems par année. E n examinant les choses de plus prés on s'aperçoit, pourtant, que la dose provenant de ^ A r qui est réellement reçue par les habitants est plus faible. E n voici les raisons: tout d'abord, le vent qui a m è n e la radioactivité dans la direction critique souffle la majeure partie du temps pendant la nuit, entre 22 h et 7 h (fréquence: 60%). Or, á ce moment, les gens sont normale ment à l'intérieur des maisons. O n a mesuré le facteur de protection de celles-ci et on a trouvé que le débit de dose y total (y c o m p r i s les effets des radiations d'origine naturelle) était environ quatre fois moins élevé à l'inté rieur qu'à l'extérieur des maisons. D'autre part, on a constaté que les effets des rayonnements y naturels étaient environ de 20 á 40% plus élevés dans les habitations qu'en plein air. C e fait avait déjà été observé dans d'autres circonstances, dans des maisons de constructions diverses qui n'étaient pas situées dans les environs du réacteur^. Si l'on tient compte de cette augmentation de la dose naturelle, le débit de dose provenant uniquement de l'argon s'abaisse à environ un neuvième de la valeur mesurée á l'extérieur des maisons. Tout semble donc indiquer que, pendant la nuit, les personnes considérées ne reçoivent en réalité qu'une petite fraction de la dose ambiante totale (en m o y e n n e au m a x i m u m 8%). Pendant la journée les conditions d'exposition aux rayons y de ^ A r ne sont pas plus m a u v a i s e s puisque, d'une part, le vent critique est m o i n s fréquent et que, d'autre part, les gens vaquent àleursoccupations, s'éloignent vers leurs places de travail, à l'école; en m o y e n n e , ils passent environ 6 h loin du site, ou à l'abri des constructions de l'institut dans lequel ils travaillent en partie. Entre 7 et 21 h, ils recevront au m a x i m u m 15% de la dose ambiante totale. Le débit de dose moyen reçu par les individus de ce groupe critique s'élève en réalité au total à m o i n s d'un quart de la valeur de la dose ambiante. Cette différence mérite d'être relevée, m ê m e si l'exemple cité ne représente pas la majorité des cas possibles. Il est certain que la conception, que l'on pourrait appeler de «man - rem réel» et les résultats objectifs qu'elle peut fournir, présente des avantages par rapport à la «dose ambiante» qui est de rigueur, surtout depuis que la tendance se fait sentir d'ériger des centrales nucléaires toujours plus prés de zones à forte densité de population. Certes, il faudra l'évaluer à nouveau lors de chaque nouveau projet si l'on veut calculer à l'avance et au plus prés les débits de dose auxquels il faut réellement s'attendre aux alentours d'une installation projetée. Tandis que les facteurs de protection des habitations et des places de travail peuvent
** Les mesures ont été faites avec la chambre à ionisation qui a été mentionnée plus haut. On a remarqué que dans ces maisons on trouvait les débits de dose les plus élevés aux étages supérieurs, voire sous les toits. Ce fait semble indiquer que cet effet n'est pas uniquement dû aux matériaux de construction mais peut-être, en grande partie, aux rayons cosmiques. 64 NAGEL
être étudiées indépendamment, les informations météorologiques, elles, devront faire l'objet de recherches plus spécifiques. Il faudra porter une attention particulière à la fréquence des vents et à leur distribution journa lière et si possible connaître dans quelles conditions de diffusion les vents critiques soufflent le plus souvent. On l'a déjà mentionné plus haut, ces données ne devront pas forcément faire l'objet d'une longue campagne de mesure au site même. Pratiquement, l'évaluation des doses réelles n'impliquerait pas d'études supplémentaires dans ce domaine à celles que l'on fait actuellement pour calculer les doses ambiantes.
DISCUSSION
P. PELLERIN (Chairman): I found this paper extremely interesting because it is very factual. However, I have serious reservations about the use ofthe term'man-rem'except in very specialized circles, because it lends itself to the use of very high figures (thousands or even millions of man - rems) which are often used to alarm the public. M. DELPLA: Yes, one must not forget that the manrem concept is based on the hypothesis of proportionality between dose and effect, whatever the value of the dose. Thus doses of millirems multiplied by millions of people give thousands of man-rems. One should not, moreover, attempt to interpret man-rems in terms of cancer or genetic deaths and this also applies of course to the individual dose accumulated by a worker. For example, it is quite inadmissible to apply the proportionality hypothesis in order to translate such a dose into a cancerogenic effect. IAEA-SM-180/29
THE NECESSITY FOR ENVIRONM ENTAL SURVEILLANCE IN THE EVALUATION OF NUCLEAR POWER PLANT SITES
P. HANDGE, F.O. HOFFMAN Institut für Reaktorsicherheit der Technischen Überwachungsvereine e. V ., Cologne, Federal Republic of Germany
Abstract
THE NECESSITY FOR ENVIRONMENTAL SURVEILLANCE IN THE EVALUATION OF NUCLEAR POWER PLANT SITES. In order to determine the suitability of a particular site for a nuclear power plant, the potential radiological burden to the surrounding population must be predicted to ensure that the radiological protection guidelines w ill not be exceeded when the plant is in operation. In the Federal Republic of Germany, however, the necessary information for such predictions is often known only on a 'global' basis. Although conservative values have generally been assumed for prediction purposes, their use does not guarantee adherence to the guidelines in all situations, especially where environmental transport of *^*1 over the pasture cow milk pathway is concerned. This problem is, therefore, studied for typical topographical conditions found in the Federal Republic of Germany. The influences of the meteorological, topographical and ecological conditions on the potential radiological burden are studied in detail, and critical conditions that have to be considered when determining the need for environmental surveillance are classified accordingly. In addition, the locations where pre-operational and operational environmental monitoring are necessary are discussed, referring also to the use of biological samples as indicators of potential radiation burdens and environmental contamination. Considering the restrictive
future growth of the nuclear industry, environmental surveillance would prove useful for a ll sites. Finally, to illustrate the need for environmental surveillance when site-specific information is not available, an example of a site located in a mountain-valley complex has been included.
1. INTRODUCTION
An earlier study has shown that, on the basis of the expected releases of radioactivity during normal operation of nuclear power plants, no siting limitations should, in general, occur in the Federal Republic of Germany, even when there is a relatively large number of such plants in a small area [1 ]. A ssum ing stack heights of between 100 and 150 m,with a level topography and average European weather conditions, external )3 and y-doses ranging from 0.01 to 0.5 mrem/a were predicted. These values lie far below the actual radiation protection guideline of 30 mrem/a for the whole body. Nevertheless, the necessity for special surveillance of internal exposures must be recognized, because the factors involved can often not be accurately predicted. At present, in the Federal Republic of Germany, general estimates of routes of internal exposure involve uncertainties stemming from the use of empirical data that have not been checked against the environmental conditions actually found at a particular site. Particular topographic situations, for example mountains lining a river valley, can influence the local weather pattern such that it differs significantly from that considered typical for the surrounding
65 66 HANDGE and HOFFMAN region. When several nuclear facilities are involved, it is also possible that peculiar environmental conditions could lead to multiple radiation burdens which would not be recognized in present calculations. Conversely, where general assessments made at present suggest that the radiation burden exceeds present guidelines, it is equally possible that the calculations have been overly conservative; hence expensive safety measures would be taken when, in reality, no such action would be necessary. An attempt, therefore, is made in this paper to investigate the circum stances in which a specific environmental surveillance programme would be necessary, paying special attention to the environmental transport of-*^1. A situation in which simultaneous monitoring of the stack release, atmospheric dispersion and critical pathways should be required are described, with emphasis upon: (a) monitoring of release of individual nuclides; (b) investigation of meteorological conditions specific to a given site; (c) evaluation of the ecological factors influencing transfer through a food chain; (d) delimitation of areas where, under certain circumstances, radiological protection standards may be exceeded; (e) specification of possible biological indicators; and (f) choosing suitable measurement programmes. In conclusion, the need for special controls as determined from environ mental surveillance data are demonstrated using an example of a particular site having topographical features that could be representative of many mountain-valley situations.
2. EXPECTED RELEASES OF RADIOACTIVITY INTO THE ATMOSPHERE
The release of radioactive substances into the air from existing nuclear installations in the Federal Republic of Germany has been determined from operating experience [2-4]. Schmitz and Siitterlin have estimated the gaseous effluents [5] for future power plants operating within the range of 1000 to 1300 MW(e). These studies suggest that the probable rates of release from LWRs lie in the range of 5000 to 200 000 Ci/а for noble gases, 0.5 to 30 C i/а for aeroso ls, and 0.05 to 1 C i/а for -^1. The upper levels are representative of release rates from the earlier designs of nuclear installa tions, while the lower levels are more representative of modern facilities. Although a detailed determination of the individual isotopes in releases to the atmosphere have not been established, such data are available for certain nuclides. Aerosols, for example, are primarily composed of ^Rb, and I38(2s, The quantities of and ^C s, and ^Sr and 9°Sr, have been estimated, though not measured directly. However, based on existing data, 1311 can be singled out as being an isotope in the effluent that is of extreme radiological importance, even when considering the high rates of noble gas release from earlier LWRs. This conclusion has been confirmed by investi gations [1] that, at the same time, have shown that under certain conditions the environmental transport of 13^ over the pasture-*cow-*milk pathway could lead to doses in excess of the existing guideline value of 90 mrem/a to the thyroid of small children. IAEA-SM-180/29 67
3. THE TRANSPORT OF RADIOACTIVE SUBSTANCES IN THE ATMOSPHERE
The atmospheric dispersion of radioactive substances is dependent upon the meteorological conditions prevailing at the time of release. At specific sites, especially when local topographic features affect the meteorological conditions, the parameters influencing atmospheric dispersion may vary considerably from the regional average [6]. For example, deviations from regional weather patterns may be brought about by such phenomena as mountain-valley winds or the diversion of regional winds by large mountain ranges. Similar effects can occur near large bodies of water, where the differences in land and water surface temperatures are often large enough to create air circulation that favours certain directions. Even such factors as the proximity of buildings and variations in vegetative cover may influence the dispersion of radioactive materials. Each site situation, therefore, may lead to very different values for the dispersion factors of both continuous and short-term releases. The following factors are most often considered in the assessment of atmospheric dispersion for short-term emissions of radionuclides: (a) atm ospheric stability; (b) wind speed; (c) the effective stack height. For continuous releases, further parameters must be considered, such as the average annual wind frequency, the relative frequency of atmospheric stability classes in each wind direction, and the occurrence of inversion, fumigation or fog conditions. In addition to these factors, average annual precipitation data must be employed when calculating the wash-out of radio active materials from the atmosphere and the resulting deposition on the ground. Typical topography encountered in the Federal Republic of Germany can be:
A. Plains (or relatively flat lands) — dominant in northern Germany; B. River valleys bordered by mountains or foot-hills (mountain-valley complexes) — typical for southern Germany; C. Plains bordered on only one side by mountains or foot-hills. (The transitional area between plains and mountains is often cut by rivers.)
Dispersion factors for these situations (estimated by using Pasquill diffusion categories) are listed in Table I, which shows, for continuous releases, that the maximum values of the average ground-level air concen tration per unit emission may lie within the range of 4 X 10'S to 3 X 10*4 s/m3, depending upon local site conditions and effective stack heights. It should be noted that, even with equal stack heights, differences in dispersion ranging over one-half to a whole order of magnitude between different sites are no rarity, as shown in Fig.l [7-9]. The ground-level concentration maxima for continuous releases, however, will always be within a radius of 4.0 km of the emission source, independent of conditions prevailing at the individual site. The corresponding dilution factors for short-term releases lie within a range of 5 X 10*^ to about 1 X 10*2 s/m^ (see Table II). At ground-level, air concentration maxima may occur up to a distance of 40 km from the source, depending upon weather conditions and effective stack height. The 68 HANDGE and HOFFMAN
TABLE I. DISPERSION FACTORS (MINIMUM VALUES) FOR TYPICAL TOPO GRAPHIC SITUATIONS IN THE FEDERAL REPUBLIC OF GERMANY
Topography
(m) (km) (m) (m) (s/m*) (s/m*)
100 100 1x10-' to 4x10^ 1.5x10*" (0.5 to 2.0 km) (0.5 to 3.0 km) Plains 150 150 4x10*' to ZX10-' 8X10** (0.5 to 2.0 km) (0.5 to 3.0 km)
100 -120 0 .2 100 0 lxl0'< to 3x10"* io*s River (0.2 km) (0.2 km) valley 150 30 5X10"* to 2X10*' 1X10-* to 3x10"' (0.3 km) (0.2 km) Plain 200 0.5 150 0 3x10^ to 2x10*' 2xl0'3 (0.5 km) (0.5 km) 150 for drift ditec 4x10*' to 2x10*' 8x10** (0.5 to 2.0 km) (0.5 to 3.0 km)
EFFECTIVE STACK HEIGHT (m)
FtG .l. Minimum average dispersion factors, x (s/m^), as a function of the effective stack height. IAEA-SM-180/29 69
TABLE II. DISPERSION FACTORS (MINIMUM VALUES) FOR SHORT TERM RELEASES
Pasquill Dispersion factor (s/m^) for effective^ stack height heff of: diffusion category о(Ь) 20 m 50 m 100 m 150 m
A 9 x 10*" 4 x 10"* 7 xlO 'S 2.4 x 10*5 1.3 X 10*5
В 1.5 x 10'S 4 x 10*" 7 x 10-s 2.2 x 10*5 1.2 x 10-5
С 3 x 10*' 4 x 10"* 7 xl0 *s 1.8 x 10*5 8 x 10*5
D 8 Х М '" 3.6 x 10"* 5.5 x 10*5 1.3x 10*5 6 x 10*6
E 1 x 10-" 3.2 x 10*" 4 .4 X 10's 8.5 x 10'" 2.6 x 10"'
F 1 x 10'" 2.7 x îor" 3.0 x 10*5 3.6 x 10*6 7 x 10*'
3 fot wind speed, u = 1 m/s. b fot heff = 0 and a distance from the emission source of 100 m.
highest concentration maxima are, however, independent of the stack height for Pasquill's diffusion categories A to D. For these categories, concen tration maxima can be expected to occur within a distance of 6 km of the emission source. For effective stack heights that are less than 50 m, concentration maxima for all diffusion categories will occur within this distance [ 10 ]. Thus, during normal operation, the maximum ground-level air concen trations and subsequent deposition on ground should occur within 6 km of the emission source, even during precipitation, since wash-out is also independent of stack height. Only under accident conditions should consideration of distances greater than 6 km be necessary. As can be seen from the above discussion, dispersion factors may vary significantly from site to site. Such differences can be expected to occur not only for sites having extreme topographic features, but also for sites whose topographic characteristics are reasonably similar. Even in the latter case, differences of an order of magnitude cannot be excluded. There fore, the direct transfer of data on the atmospheric dispersion characteristics of one site to another apparently similar site, or the derivation of dispersion factors from the regional meteorological data is only valid under very special circumstances; in most cases there are large uncertainties involved. In order to assess the site-specific conditions of atmospheric dispersion adequately, it is necessary to obtain information about the relationship and frequency of: diffusion categories and wind distribution; inversion weather conditions and the corresponding heights of barrier layers; and fumigation conditions. Unfortunately, in mountain-valley complexes, the actual behaviour of wind currents is, in most cases, not known. Since meteoro logical data are usually derived from information averaged over periods of several years, such data are suitable for the estimation of dispersion factors for prolonged releases, but their use in deriving short-term factors may not be appropriate. Adoption of regional meteorological data for a specific site is, therefore, only possible if either the monitoring stations are in the close vicinity of 70 HANDGE and HOFFMAN the nuclear installation, or if the calculated radiation burden is at least one order of magnitude lower than the relevant radiation protection guideline. Such adoption, however, will not be acceptable when large uncertainties exist in the prediction of the transport of radionuclides through food chains, or when sites are located in complex terrains. In mountain-valley complexes, for example, the necessary meteorological information should always be obtained from special site-specific pre-operational monitoring programmes.
4. THE TRANSPORT OF RADIOACTIVE SUBSTANCES FROM AIR THROUGH CRITICAL FOOD CHAINS
In order to understand the magnitude of the uncertainties involved in the transport of radioactive substances through food chains, the factors affecting contamination of vegetation, animal uptake and the consumption of contaminated animal or vegetable products by man must be thoroughly investigated. B e c a u s e *^1 is an especially critical nuclide, it is used here to illustrate the difficulties involved with accurately estimating radiological burdens without adequate site-specific information. Although 13^ may be concentrated in aquatic as well as terrestrial ecosystems, its impact due to deposition on vegetation and its subsequent concentration along the pasture** cow^milk pathway appears, at present, to be the most important.
4.1. The deposition of radioactive substances on vegetation
In all food chains the primary and crucial link is vegetation. While contamination of vegetation does occur through root uptake from the soil, the principal mode of contamination is through dry deposition or in rainfall on p arts in direct contact with the air. The chemical form, reactivity and particle size of the radionuclide appear to be the most significant factors controlling deposition on vegetation, but the effects of wind, humidity and temperature, and of the density, height, growth form and leaf characteristics of vegetation cannot be ignored. Obviously, the contamination of vegetation is dependent not only upon the quantity and type of radioactive effluents, but also on the prevailing environmental conditions.
4.1.1. Dry deposition The rate of transfer of radionuclides from an above-ground concentration in air to a given surface under dry conditions is most commonly referred to as the deposition velocity (Vg). For the dry deposition of ^*1 on grass land, Vg-values as high as *^-4X 10*2 m /s have been reported when the isotope has been in the form of 131^ vapour. The Vg for submicron particles of I3ii is lower by about an order of magnitude, and values for organic compounds of I3li are lower still, of the order of 10*3 m /s or less. Despite the wide range of values, a Vg-value of 10'2 m /s is frequently used and is often recommended for general assessment calculations under the conservative assumption that all of the I3ij in air exists as elemental iodine vapour. Such an assumption is considered prudent because of the uncertainties involved in determinations of the composition of reactor effluents containing 13li and their behaviour in the atmosphere. IAEA-SM-180/29 71
Unfortunately, an additional uncertainty arises when attempting to distinguish between a which represents the transfer from air to a surface area comprising soil and vegetation, and a Vg representing a transfer directly to the vegetation growing on a given surface area of ground. For example, general assessment calculations made in the USA [ 11 ] and inthe UK [12, 13] employ a Vg of 10'^ m /s. However, in the UK the amount of I31i that is actually deposited on vegetation is calculated using an'initial retention factor'of 0.25 [12-14]. No such retention factor is used by the USAEC [11]. Therefore, one can conclude that the Vg used in the UK is intended to represent the surface deposition on both soil and vegetation, while the (identical) Vg used in the USA refers to the deposition of on the vegetation itself (per surface area of ground).
4.1.2. Wet deposition
During periods of precipitation, rain and snow will wash out some of the radionuclides released from the plume. Of importance in determining the consequence of wash-out is a knowledge of the chemical reactivity of the nuclide, its particle or droplet size, and the precipitation rate. For dry iodine gas, predicted wash-out values are between 3 X 10*6 to 2 X 10"? s'^ during rainfall. In rainfall of 0.5 to 3.0 mm/h, 30 to 40 micron droplets are reported to range from 2 X 10*4 to 8 X 10*4 s*i [ 13] . Thus, under conditions of heavy precipitation, the leaching of soluble radionuclides like 13^1 from the atmosphere will be very significant; however, it is also likely that vegetation retention and grazing intensity will decrease [ 15 ].
4.2. Field loss of activity
Followingdeposition on vegetation, radionuclides maybe removed by weathering, grazing, mechanical clipping, sublimation, and the natural pro cesses ofvegetation decomposition following death and loss of leaves. Under extreme conditions,the effective half-life of ^ I on vegetation has ranged from about 3 to 7 d (corresponding to a 'field loss' half-life of 6 to >35 d); an average value of 5 d (corresponding to a field loss half-life of 13 to 14 d) has generally been used for assessment calculations [16]. The use of an average field loss half-life for the long-lived isotopes may, however, be misleading without the additional knowledge of the prevailing environmental conditions around a given site [15].
4.3. The uptake of radioactivity by animals and man
The uptake of radioactivity by animals is not only dependent upon the feeding habits of the animal, but also upon its physiological demands, the nutritional quality of its food, the availability of its food and the indirect influences of weather. In general, the organism consuming the largest quantity of vegetation will initially collect the largest amount of deposited radioactivity. However, with certain long-lived isotopes, a further concen tration may occur along special predator/prey links in food chains. In arctic food chains, for example, the highest l37Qs concentrations were found in mountain lions and wolves, which depended almost entirely upon large 72 HANDGE and HOFFMAN herbivores for food [15]. For short-lived isotopes such as ^I, however, the herbivorous animal whose diet is entirely composed of fresh vegetation will be the source of the highest concentration of deposited radioactivity. Since ofthelargeherbivoresthe cow is usually the most important within man's food chain, cows grazing on contaminated pastures are often critical vectors of radioactivity. For special groups, such as farmers or vegetarians, contaminated vegetation may be an important factor, but fresh whole-milk from a "family cow" is considered to be the critical food for ^ I ingestion by small children (the ' critical group' ) [ 16- 19 ]. Meat from either domestic or wild herbivores is not considered to be an important factor in ^ 1 ingestion, but it may well be the critical food pathway for other isotopes [ 15 ].
4.3.1. The uptake of contaminated vegetation by dairy cows
The daily intake of fresh forage, and hence any freshly deposited radio active materials, by dairy cows is influenced by the weight, breed and metabolism of the animal, and the type of forage, yield and nutrient content of pasture vegetation. Seasonal variation is also important. However, the most influential factor is probably the management practice employed by the farmer. Only when the cow is grazing on open range, without any restriction, is the effect of human control minimal [ 16, 20]. In the Federal Republic of Germany the most common method of grazing is a controlled type where cows are either grazed on different portions of a single pasture once or twice a day (strip grazing) or are grazed in rotation on two or more pastures per season (rotational grazing). The intake of long-lived radionuclides may occur throughout the entire year if the winter feed is contaminated, but for ^ 1 the effective period of intake is restricted to the time when the cows are on pasture, including the time when cows are fed forage freshly cut from exposed fields (green chop). The feeding of concentrates and special high-protein foods (Kraftfutter) will generally reduce the intake of contaminated vegetation by reducing the fresh green forage intake. In the Federal Republic of Germany the grazing season is about six months. This varies from year to year and is directly dependent upon weather (e.g. especially as it affects the die-off of grasses, or the covering of pastures by snow). Thus, in the country bordering on the Alps (the Alpenvorland) the grazing season may be as short as 80 days, while in the lowlands of the Federal Republic of Germany this season may exceed 200 days. During mild days in winter, however, it is not uncommon to find cows on pasture, though most of their diet would probably still be composed of dried and stored feed and forage supplements [16]. The forage consumption of cows in the Federal Republic of Germany varies on the average from 8 to ^-15 kg (dry weight) per cow per day [16, 21]. Extremes of forage density at times when dairy cows enter a given pasture have been reported to range from about 50 g/m^ to 400 g/m2 [21]. Therefore, the range of an effective grazing area or 'utilized area factor' (UAF) would be expected to lie within 300 to 20 m^/d per cow. For general assessment calculations, 160 m^/d-cow has been assumed for conditions in the UK [14], and 45 m^/d-cow has been recommended as an average value for the USA [20]. In reality, however, the UAF will be dependent upon actual grazing and management conditions occurring at the time of deposition on vegetation. Critical conditions may be expected when IAEA-SM-180/29 73 the forage density is low, such as occurs in late season or under poor soil conditions or conditions of overgrazing, and when management practices require the animal to obtain all of its food from pasture herbage.
4. 3.2. The transfer of radioactive substances to milk
The transfer of ingested radioactive substances to milk by dairy cows depends on a number of interrelated factors. Those factors which are intrinsic to the cow are milk yield, breed and stage of lactation. Other factors that are important are the amounts of stable elements having a similar chemical behaviour in the diet, and the biological availability of the radio isotope. It should be noted that considerable variation has been observed in laboratory studies in the fractions of administered material that have appear ed in milk. Such variations were found not only between different animals but also in a given animal at different times [15]. For 13Ц, seasonal changes are reported to affect the percentage of the amount of 13^1 ingested daily that appears in each litre of milk. This not only varies with time, but also with geographical location and the yearly variations in the climate. In the UK for instance, an average milk transfer factor (fgj of 0.55% d* cow/1 for prolonged exposure and 0.35% d-cow/1 for short-term exposure has been used in assessment calculations [14], while in the USA, an f^ of l%d-cow/1 is standard [15]; for some unexplained reason cows in the USA are capable of secreting more ^ 1 in milk than English cows [15]. No known values are available for the secretion of radioiodine by cows in the Federal Republic of Germany. It has been reported in Ref. [22], however, that it is possible for cows to have a substantially higher f^ (3 to 5% d-cow/1) if they are put on freshly contaminated pasture after being fed goitrogens. Perhaps it should also be mentioned that the consumption of milk from ewes and goats might in individual cases lead to radiation burdens comparable to those following the consumption of cow milk. Although these animals would not be expected to eat as much vegetation per day as would a cow, their secretion of isij is significantly greater [ 15, 23].
4.3.3. Vegetation to milk transfer factors
The transfer of deposited radioactive materials from vegetation to milk has often been described by a single factor expressed as a ratio of (Ci/1) : (Ci/m^). These factors have usually derived from actual field measurements, and their use in general assessment work includes the advantage of by-passing a detailed investigation of all of the factors influencing the transport of radio nuclides from vegetation to the animal and from the animal into milk. These factors, however, involve considerable uncertainties since the environmental and grazing conditions prevalent when many of the original measurements were taken were not reported. It is also difficult to ascertain whether these factors represent a milk concentration that would result from the total areal deposition on soil and vegetation, or whether they represent a milk concentration resulting from a given vegetation contamination reported in curies per square metre. For 131^ the majority of these factors appear to lie around 0.10 (Ci/1) (Ci/m^)*^ [15], but vegetation to milk transfer factors have been found ranging from 74 HANDGE and HOFFMAN about 0.75 [24] to less than 0.02(Ci/l)(Ci/m^)'^. Although there is a tendency to use the factor 0. 10 as an empirical value, its use under certain circum stances could lead to an underestimate of milk contamination. For example, reported averages of 0.2, 0.4 and 0.5(Ci/l)(Ci/m^)'^ were measured in three different locations around the Studsvik Research Station in Sweden during the later part of the sum m er of 1967 [24]. IntheUK[12, 14], afactorof 0.2(Ci/l)(Ci/m^)'^ for the transfer of 13S from a total groun^ area deposition ( vegetation and soil) into milk has been used for assessment purposes. This ratio is derived assuming a grazing area of 160 m^/cow-d, an 'initial vegetation retention factor' of 0.25, and a milk secretion factor of 0.55% d. cow/1. It can therefore be deduced that a factor representing the transfer from vegetation to milk in the UK would approach 0.9 (Ci/l) (C i/т'^)Г*. In the USA 0.09 (Ci/l) (C i/m^)*^ is currently in use as a guiding value for estimating the transfer of from vegetation to milk [11]. This means that the assessment value used in the USA is a factor of ten lower than the values used in the UK. Even more confusion arises when one considers the evidence for a more efficient secretion into milk by cows in the USA. In the Federal Republic of Germany a value of 0.65 (Ci/1) (Ci/m^)'^ has been recommended for use in calculating the ^ 1 transfer from vegetation to milk [16] (see below).
4.4. The assessment of the transport of from air to man
As can be seen in Table III, the parameters commonly used to assess the ^1 concentration in milk resulting from a given concentration in air are very variable. Even if one assumes that all of the iodine in air is inor ganic, the values for each parameter may vary over an order of magnitude. Fortunately, the uncertainty in the field half-life is not too important since the physical half-life of 13Ц is relatively short.
T A B L E III. CRITICAL PA RA M ETERS USED IN THE ASSESSM ENT OF THE TRANSPORT OF FROM AIR TO MILK
Factor Reported range General assessment values Ref.
Deposition velocity (m/s) 10^ to 10*5 1 C [11-24]
"Initial retention factor" 1.0 ^ to 0.10 1.0^ to 0.25
Field half-life (d) 35 to 6 14 to 12
Utilized-area factor (mVcowd) 300 to 12 160 to 45
Secretion into m ilk (d* cow/1) 3 xlO*^ to 2 x 10*3 1 xl0"2 to 3.5 x 10'S Milk:ground ratiob, (Ci/l):(Ci/m^) 0. 75 to 0. 02 0.26 to 0. 09b
R When the deposition velocity pertains to the direct transfer of radioactive materials from air to the the edible parts of vegetation growing within a given area of ground, an "initial retention factor" of 1.0 may be assumed. b These ratios are assumed to represent a milk concentration resulting from a total areal deposition (soil and vegetation), except for the general assessment value of 0.09 used by the USAEC [11] to represent the transfer from vegetation to milk. IAEA-SM-180/29 75
Because of the lack of experimental data derived for conditions represen tative of the environmental factors affecting pasture conditions, grazing habits and animal physiology in the Federal Republic of Germany, assessment calculations depend on information obtained from the literature [16, 17, 26]. For our assessment calculations, we have chosen a V^ representative of a total areal deposition of 10*2 m /s, a Teff on pasture forage of 5 days, and a fm of 1% d - cow/1. An initial retention factor of 0.4 and UAF of 65 m^/cow-d have been derived from estimates of average forage conditions and grazing habits corresponding to an assumed pasture vegetation density of 170 g/m^ (dry weight basis) and a grazing intake of 11.0 kg (dry matter)/ cow - d [ 16, 27 ]. Under conditions of prolonged release, our calculations predict that an above-ground air concentration of 1 C i/m3 will deposit approximately 6.2 X 103 Ci/m2 on pasture land, of which approximately 2.5 X 10^ Ci/m^ will be retained on forage vegetation. Of this, 1.6 X 10^ Ci will be consumed per day by a grazing cow, and 1. 6 X 103 Ci will be secreted into each litre of milk. Using age-dependent values for the various parameters involved in the calculation of the ^ 1 dose to the thyroid and assuming a milk con sumption of 0.8 1/d by 'critical children' during a 183 day grazing season, a maximum annual dose of 3.4X 1012 rems can be expected [16, 17]. Thus an air concentration of 4. 3 X 10*13 Ci/m3 would lead to a dose commitment of 1. 5 rem/a to the ' critical group'. These estimates are compared to the predictions of general assessment calculations made in the USA and the UK in Table IV.
TABLE IV. THE COMPARISON OF GENERAL ASSESSMENT PREDICTIONS FOR THE USA, THE UK AND BY 1RS FOR THE TRANSPORT OF ^ 1 FROM AIR TO MAN DURING PROLONGED RELEASES
Above ground air concentration 1 1 1 (u n ite s )
Total ground area deposition - 6.2 xio" 6.2 x 10" (unit^n^)
Vegetation contamination 6.9 X 10" 1.6 x 10" 2.5 x 10" (unit/m^)
M ilk contamination 0.57x103 1.4 x 1()3 1.6 x 10" (unit^itre)
Annual dose commitment 3.6 xlO*" 4.8x10*" 3.4 xlO*" ' critical group* (footnote a) (footnote a) (footnote b) (rem'Ci/jnit)
Above ground air concentration 4.2 Х10-И 2.8 X 1 0 *" 4.3 x 10*" required to deliver an annual (footnote a) (footnote B) dose commitment of 1.5 rems (Ci/hiS)
Ref. [111 [12.14] [16,17]
^ Assumes continuous grazing by cows during the entire year, b Assumes age-dependent dose parameters and a 183 day grazing period. 76 HANDGE and HOFFMAN
Obviously, before the accuracy of these calculations can be tested, data must be collected in actual field experiments. Unfortunately, such field studies have not been conducted in the Federal Republic of Germany.
5. BIOLOGICAL INDICATORS OF RADIOACTIVE MATERIALS
In implementing an environmental monitoring programme, important biological indicators for radioactive materials will have to be determined. A biological indicator may be specified as such because of its ability to effectively concentrate radionuclides and/or because of its importance in the human diet. Food items are the best indicators of the radiological burden of a ' critical group', but with low-level releases other biological samples, more sensitive as indicators for some given radionuclide, may have to be used in order to detect the presence of this nuclide in the ecosystem. The most meaningful indicator for I3li would be the milk of grazing animals. In circumstances where milk samples cannot be obtained, samples could be taken from vegetation growing on actual pastures, potential pastures or in private gardens. Vegetation monitoring could be instrumented if vegetation to milk transfer ratios applicable to the various grazing conditions found in the Federal Republic of Germany were known. In this case, it would be essential to specify the amount of available edible pasture vegetation per unit surface area of ground covered by edible vegetation (edible kg per edible m2). When concentrations in vegetation or milk cannot be detected, sampling the thyroids of grazing animals may be useful for determining the presence of^I. Bovine thyroids have been examined with a view to monitoring iodine in various establishments in the USA [28-32], but thyroids from other animals could prove to be equally suitable. In India, for example, one gram of goat's thyroid was reported to contain 13.5 times more 1^1 than was found per litre of cow's milk and 8 times more than that found per kilogram of grass [33]. In Portugal, the thyroids of sheep were found to be more sensi tive as indicators of *^1 than were those of cattle [ 34]. For the purpose of estimating the contamination of milk, Stigall et al. have used a ratio of 12 Ci/g cow thyroid t 1 Ci/1 milk [ 32]. The reliability of thyroid-to-milk estimates, however, cannot be assessed at present; future experiments might be able to provide the data necessary to enable these correlations to be made. Under normal conditions, when slaughtering domestic animals for the purpose of sampling ^ 1 is not feasible, thyroids may be collected from game. Among the wild herbivores which may be found around nuclear sites in the Federal Republic of Germany, the roe deer (Capreolus capreolus) would probably be the most suitable indicator for I3ii. it can be hunted alm ost the entire year, and although it would not be expected to graze as great an area as would the larger red deer (Cervus elaphus) it has a better defined home range. Thus, thyroids of the roe deer would act as indicators for I3il contamination in the general vicinity of where the animal was collected. Unfortunately, quantitative relationships between the 134 content of the thyroids of roe deer and the I3ij content of cow milk are not known. At present, thyroid samples from wild herbivores could be best used as quali tative indicators of unusual amounts of environmental 13li in situations where IAEA-SM-180/29 77
TABLE V. A LIST OF POSSIBLE BIOLOGICAL INDICATORS FOR THE PRESENCE OF 13Ц
FRESH M ILK C. THYROIDS OF DOMESTIC HERBIVORES
1) Dairy cows 1) dairy cows
2) goats 2) beef cattle
3 )ew es 3) sheep
(a) when measurements are taken, the 4) goats amount of stored feed consumption should benoted
B. FRESH VEGETATION D. THYROIDS OF WILD HERBIVORES
1) leafy table vegetation 1) roe deet (Capreolus capreolus) [ Reh]
2) pasture vegetation 2) red deer (Cervus elaphus) [Rothirsch]
3) vegetation of potential pastures 3) fallow deer (Dama dama) [Damhirsch]
(a) should be recorded in both fresh and 4) European hare (Lepus europaeus) [Hase] dry weight of edible substance per ground area covered by such substances (kg/m^)
neither milk samples, nor thyroids of domestic herbivores, nor pasture vegetation can be obtained. A list of possible biological indicators of I3ii in the Federal Republic of Germany is given in Table V.
6. CONSIDERATIONS OF THE NECESSITY FOR AND REQUIREMENTS OF AN ENVIRONMENTAL SURVEILLANCE PROGRAMME
A review of the preceding sections reveals that general assessment calculations that are to be used to predict the environmental transport and ensuing doses resulting from the release of radioactive substances involve uncertainties due to the lack of site-specific information. Consequently,it has been the practice to use conservative assumptions for general assess ment purposes, in order to be on the ' safe side' . This practice, however, may not be appropriate in all cases since these conservative calculations could deceptively indicate that levels set as radiation protection guidelines have been exceeded. In particular, this could occur for sites situated: (a) in narrow valleys (see Section 7); (b) near large bodies of water; (c) on plains when installations are equipped with low stacks. An environmental surveillance programme is would be absolutely necessary in these situations in order to determine the potential radiological burden to the surrounding population accurately and to improve the sensitivity of predictive estimates. However, for monitoring an eventual build-up in the environment of long-lived radionuclides such as 89Sr, 90g^ 129^ 134çg and ^ C s, a surveillance programme would be useful at all sites. 78 HANDGE and HOFFMAN
TABLE VI. CONSIDERATIONS INVOLVED IN THE EVALUATION OF THE NECESSITY FOR ENVIRONMENTAL SURVEILLANCE
A. THE CONDITIONS OF RELEASE
B. ATMOSPHERIC DISPERSION
1) prevailing meteorological and topographical conditions
2) nearby buildings
3) stack height
C. WATER CONTAMINATION
D. CRITICAL ROUTES OF EXPOSURE
1) location, life and dietary habits of 'critical groups'
2) location of 'edible vegetation'
3) deposition or absorption of radionuclides on or by vegetation (aquatic and terrestrial)
4) 'critical pathways' through which radionuclides may enter 'critical foods'
5) factors affecting the transport of nuclides along these 'critical pathways'
When evaluating the necessity for environmental surveillance, it is necessary to delineate in which areas and under which circumstances radia tion protection guidelines could be exceeded. This requires comprehensive knowledge of any peculiar environmental factors liable to play dominant roles in the region surrounding a nuclear installation, and involves the determination of 'critical nuclides', 'critical pathways' and 'critical groups or individuals of the general population' (Table VI). For example, critical conditions for would occur if it were released primarily as elemental vapour, and if pasture land exists within 6 km of the release point, and if small children consume milk coming from cows grazing on such pastures. The worst possible conditions could occur if these cows depended entirely upon fresh pasture vegetation for food in a local environ ment showing a chronic lack of natural iodine, and if such pastures were such that large areas had to be grazed. Unless knowledge of such conditions is already available, the required information must be obtained from an environmental surveillance programme conducted before the operation of the installation. IAEA-SM-180/29 79
TABLE VII. METEOROLOGICAL PARAMETERS WHICH MUST BE ESTIMATED DURING A PRE-OPERATIONAL ENVIRONMENTAL SURVEILLANCE PROGRAMME
A. WIND SPEED
B. WtND DIRECTION
C. THE OCCURRENCE OF ATMOSPHERIC STABILITY CATEGORIES
D. THE FREQUENCY AND EXTENT OF SUCH WEATHER PHENOMENA AS:
1) inversion
2) fumigation conditions
3) fog
4) precipitation
A pre-operation surveillance programme should investigate all local factors that could influence environmental transport of radionuclides and the subsequent radiological burden to the surrounding inhabitants. Such a programme would involve the collection of site-specific meteorological data (Table VII), and the assessment of critical routes of radiation exposure. One meteorological station located in the immediate vicinity of the site will usually suffice. More than one monitoring location, however, should be used in complex terrains if considerable variations in local wind currents and atmospheric stability are likely to occur. In such areas the precise number and siting of meteorological stations will have to be determined according to the specific conditions, especially with a view to monitoring at a distance from the site in order to assess the dispersion in case of accidental releases. For the derivation of long-term dilution factors, meteorological data should be representative of a sampling period of at least two years. Momentary and seasonal deviations from annual averages should be considered in order to assess the dispersion conditions that might prevail during short-term releases and during critical periods when food-chain transport is of importance; for example, the meteorological information essential for the assessment of the pasture-*cow-rnilk transport of must be obtained from data collected during the grazing season. In addition to the above measurements, the pre-operational monitoring programme should include measurements of the initial radiation background. These measurements, which should be conducted in both the aquatic and the terrestrial environments, are needed to determine the presence of radio activity from other sources. The information obtained from measurements 80 HANDGE and HOFFMAN performed before the operation of a nuclear facility would, therefore, enable 'critical areas' to be defined by determining the location of maximum ground- level air concentrations and by delineating the potentially critical terrestrial and aquatic food chains. By defining these 'critical areas', monitoring locations can be chosen at which special measurements would be required during subsequent operation of the nuclear facility. During the operation of the nuclear installation, detailed monitoring of the 'critical nuclides' must be made in addition to gross measurements of radioactive emissions. Environmental monitoring for those 'critical nuclides' conducted during the operation of the installation would help to predict radiation burdens accurately and to test the sensitivity of general assessment calculations. Measurements taken for these purposes should be performed at regular intervals in air, water and 'critical foods' of the 'critical areas' which have been determined during the pre-operational programme. Furthermore, the information obtained from these measure ments would help increase the knowledge of the behaviour of radionuclides under specific environmental conditions and would make it possible to assess the suitability of a given site for additional nuclear installations more confidently. For the detection of the presence of a given nuclide in the ecological complex, non-food biological indicators may serve as suitable samples (see Section 5). However, for the prediction of radiation burdens resulting from short-term releases that exceed the expected annual average release rates, ' critical areas' would have to be determined from the existing meteorological monitoring system, and the 'critical foods' sampled accordingly, Therefore, it is absolutely necessary that at least one meteorological monitoring station continues to function during the operational life of the nuclear installation.
7. THE NECESSITY FOR ENVIRONMENTAL SURVEILLANCE OF SITES LOCATED IN MOUNTAIN VALLEY COMPLEXES
The previous considerations and requirements for the implementation of an environmental surveillance programme will now be applied to a specific situation. For this purpose, a hypothetical site located in a mountain-valley complex has been chosen as an especially interesting example because of the special topographical and meteorological problems. The topography surrounding this hypothetical site is shown in Fig. 2. The site is located on the banks of a river, in a valley that is approximately 2 km wide. The hills on each side rise above the site to about 100 to 120 m,with the highest peaks reaching 140 m. The direction of the valley is WSW to ENE. In such a situation, a stack height of 150 m would be practically insigni ficant in comparison to the surrounding peaks. Even when calculating the effective stack height by considering the efflux speed of effluent gases (as well as the effects of near-by buildings and vegetation), the effective height above the hills would be from 20 m to a maximum of 50 m. As already mentioned, the wind distribution of sites located in mountain- valley complexes is influenced considerably by the topography. Whereas the general wind direction in the Federal Republic of Germany is basically westerly, varying mainly in the quadrant from SW to NW, the actual behaviour of wind in this valley would have to be assessed in a site-specific meteorological monitoring programme. IAEA-SM-180/29 81
100 to ELEVATION a )20m KEY )20 to 140 to 140 m )60m
RELATIVE FREQUENCY OF WIND DIRECTION AND MINIMUM DISPERSION FACTORS FOR THE HYPOTHETICAL SITE
Rel. frequency Effective Min. dispersion f.tctor (s/m^) Wind of stack height (m) short-term
150 1 x lO '' 8 xl0*s NE 0.18 30 5 x 10*6 2 x lO '"
150 2 x lO *' 8 xlO^^ SW 0.35 30 1 xlO-s 2 x IO-'
W 0.15 30 4 xl0*s 2 x 10'"
150 5 xlO** 8 x 10'S Others < 0.07 еа 30 2 x lO '" 2 x lO '" 82 HANDGE and HOFFMAN
Based on information from sites with similar topographical features, however, one can roughly estimate that the winds in this valley would be channelled to fit a pattern:
NE 18% SW 35% W 15% other directions ^ 7% each Since site-specific atmospheric stability conditions are also unknown, they must be assumed from data pertaining to average German weather conditions. The dispersion factors derived from these assumed meteorological conditions are listed in the table in the caption to Fig. 2. For the valley floor, the estimated minimum annual average dispersion factor would be 2 X 10"? s/щЗ. For the hillsides bordering the valley, how ever, annual average dispersion factors of 4 X 10*6 to 1 X 10'S s/m^ have been estimated. Assuming a nuclear power plant of some 1200 MW(e), a release rate of about 0.5 Ci/а can be expected, unless special equipment for its reduction is installed [5]. Therefore, if small children were to drink milk from cows grazing on pastures in the surrounding hills, a thyroid dose of 500 mrem/a may be calculated [ 16 ] (see subsection 4. 4). More conservative assumptions [ 7] would lead to a calculated thyroid dose of ^-800 mrem/a. These results would indicate that additional filters or other equipment n ecessary for reducing the release of I3ij from 0. 5 to 0.05 Ci/а would need to be installed in order to comply with the current radiological protection guideline of 90 mrem/a for the thyroids of the ' critical group' . The reduction required might be still greater if one were to consider the effects of a short-term release which exceeded the average I3ii release rate, or if one were to consider the additional effects of wash-out created by precipitation from the plume of a wet cooling tower. Without monitoring the specific local conditions, the validity of these requirements would be extremely difficult to assess with the given topography. Because of the uncertainties inherent in the general assessment calcu lations of transport through the food chain of 131] (see Section 4) and the derivation of dispersion factors without specific environmental information (see Section 3), an accurate assessment of the actual radiological impact of any 13Ц releases from this hypothetical site can, obviously, only be obtained from a surveillance programme conducted in and around the vicinity of the nuclear installation. In order to improve initial predictions, a pre operation surveillance programme should include both meteorological monitoring (as described in Section6) and an investigation of local factors that could influence the I3ii uptake by the ' critical group'. In particular, such an investigation should consider: (a) The location of actual and potential pastures and fields of vegetables; (b) The dominant grazing habits, agricultural management practices, and general conditions of vegetable growth; (c) The effective length of the grazing season, and the season of food-crop production; (d) The dietary habits of local children, especially infants likely to consume whole cow milk; (e) The native iodine content of the diets of grazing animals and the local population. IAEA-SM-180/29 83
The information that would be obtained from this programme would enable more realistic values to be derived for use in general assessment calculations. Improved assessments may prove valuable as early-warning indicators of situations where restrictive guideline values (such as ourpresent90mrem/a value for isii emissions) might be exceeded during the later operation of a nuclear installation. Monitoring the milk of cows grazing on 'critical pastures' during the operation of the installation would, of course, be necessary to check the results obtained from pre-operational assessments. The direct monitoring of air, vegetation and milk, however, would be the only way of accurately predicting the doses that might result following an accidental or short-term excessive release.
8. FINAL REMARKS
In considering the ISij emission from modern nuclear plants and the current 90 m rem /a to the thyroid limit set by the guideline in the F ed eral Republic of Germany, this study has demonstrated the absolute necessity of carrying out environmental surveillance for nuclear installations to be sited in a complex topographical situation. Environmental monitoring, how ever, would also be useful under all circumstances for anticipating the environmental accumulation of long-lived isotopes that might result with the future growth of the nuclear industry. Interestingly, requirements similar to those recommended in this report have been recommended for nuclear power plants with light water reactors in the USA by the USAEC [ 11 ]. With the information gained from environmental monitoring, the evaluation of the suitability of nuclear sites will be more precise, thus enabling the need for special safety equipment to be appropriately assessed. The monitoring requirements described within this report should be adequate for site situations likely to occur in the Federal Republic of Germany.
REFERENCES
[1 ] HANDGE, P . , SCHWARZER, W ., 7th IRS-Fachgespräch "Betriebliche Ableitung radioaktiver Stoffe aus kerntechnischen Anlagen", Rep. IRS-T-23 (July 1972) 111. [2 ] ENGEL. H ., Betriebserfahrungen mit Brennelementen und Aktivitäten in SWR-Kemkraftwerken, Atom wirtschaft 11 (1970) 523. [3 ] WECKESSER, A . , et a l . , Bau und Betriebserfahrungen mit Leichtwasserreaktoren in der BRD, Atom- wirtschaft 9-11 (1972) 479. [4 ] HERRMANN, G ., Abgabe radioaktiver Stoffe bei Normalbetrieb aus Leichtwasserreaktoren in der BRD, Rep. IRS-W-1 (March 1972). [5 ] SCHM ITZ, G ., SÛTTERLIN, L ., 7th IRS-Fachgespräch "Betriebliche Ableitung radioaktiver Stoffe aus kemtechnischen Anlagen", Rep. IRS-T-23 (July 1972) 43. [6 ] Meteorology and Atomic Energy 1968, USAEC Rep. ТЮ-24190. [7 ] Emissionsquellstärke von Kernkraftwerken, Schriftenreihe Kernforschung Gersbach & Sohn Verlag (1972). [8 ] VOGT, K . I . , Umweltkontamination und Strahlenbelastung durch radioaktive Abluft aus kerntechnischen Anlagen, KFA Jülich Rep. JÜL-637-ST (1970). [9 ] NESTER, K . , Statistische Auswertung der Windmessungen im Kernforschungszentrum Karlsruhe aus den Jahren 1968/69, KFZ Karlsruhe Rep. KFK 1606. [10] PASQUILL, F ., Atmospheric Diffusion, Van Nostrand, London (1962). [11] UNITED STATES ATOMIC ENERGY COMMISSION, Regulatory Guide 1.42, Interim Licensing Policy on as Low as Practicable for Gaseous Radioiodine Releases from Light-Water-Cooled Nuclear Power Reactors, USAEC, Washington, DC (June 1973). 84 HANDGE and HOFFMAN
[12] BRYANT, P .M ., Derivation of working limits for continuous release ratios of iodine-131 to the atmos phere in a m ilk producing area, Health Phys. 10 (1964) 249. [13] BEATTIE, J . R . , BRYANT, P .M ., Assessment of environmental hazards from reactor fission product release, UKAEA Rep. AHSB (S) R-135 (1970). [14] MORLEY, F ., BRYANT, Pamela M ., "Basic and derived radiological standards for the evaluation of environ mental contamination", Environmental Contamination by Radioactive Materials (Proc. Seminar Vienna, 1969), IAEA, Vienna (1969) 255.
Critic. Rev.Environ.Cont. (Sep.1971) 337. [16] HOFFMAN, F. O ., Environmental variables involved with the estimation of the amount of m i in milk and the subsequent dose to the thyroid, Rep. IRS-W-6 (June 1973). [17] HOFFMAN, F. O ., Parameters to be Considered When Calculating the Age-Dependent ^*1 dose to the Thyroid, Rep. IRS-W-5 (April 1973). [18] BAYER, A . , Die ortsabhängige spezifische Dosis d von J^*, K FZ Karlsruhe Rep. KFK 1661 (Aug. 1972). [19] GARNER, R. J . , RUSSEL, R .S ., "Isotopes of iodine", Radioactivity and Human Diet (RUSSEL, R .S ., Ed.), Pergamon Press, Oxford (1966) 297. [20] KORANDA, J.J., Agricultural factors affectingthedailyintakeoffreshfalloutbydairycows, University of California Rep. UCRL-12479 (1965). [21] VOIGTLÄNDER, G ., Institut für Grünlandlehre, Technische Universität München, private communication. [22] TAMPLIN, A .R ., 1-131, 1-133 and C ow M ilk, University of California Rep. UCRL-14146 (1965). [23] COMAR, C . L . , "Radioactivity in animals * entry and metabolism", Radioactivity and Human Diet (RUSSEL, R .C ., Ed.), Pergamon Press (1966) 127. [24] BERGSTRÖM, S. О. W ., GYLLANDER, C ., "Iodine-131 in grass and m ilk correlated with releases from
Vienna, 1969), IAEA, Vienna (1969)403. [25] PETERSO N .H .T., J r . , SMITH, J.M ., "Guides for predicting thyroid dose from environmental measure ments following radioiodine releases", Environmental Surveillance in the Vicinity of Nuclear Facilities. (REINIG, W .C ., Ed.), Charles С. Thomas (1970). [26] BAYER, A ., Die altersabhängigen Ingestions-Dosisfaktorengg und gg, vonJod-131, K FZ Karlsruhe Rep. K FK -1582 (1972).
4 (1970) 57. [28] ZIMBRICK, J. D . , VOILLEQUÉ, Eds., Controlled Environmental Radioiodine Tests at the National Reactor Testing Station, Progr. Rep. No. 4, USAEC Rep. 1DO-12065 (Jan. 1969). [29] PELLETIER, C. A ., ZIMBRICK, J. D ., "Kinetics of environmental radioiodine transport through the milk-food chain", Environmental Surveillance in the Vicinity of Nuclear Facilities (REINIG, W .C ., Ed.), Charles C. Thomas (1970) 257. [30] HARVEY, R .S., "Biological indicators of environmental radioactivity", Environmental Surveillance in the Vicinity of Nuclear Facilities (REINIG, W .C ., Ed.), Charles С. Thomas (1970) 136. [31] KAHN, B., BLANCHARD, R.L. , KRIEGER, H .L., KOLDE, H .E., SMITH, D.B., MARTIN, A., GOLD, S., AVERETT, W. J . , BRINCK, W. L ., KARCHES, G .J., "Radiological surveillance studies at a boiling water nuclear power reactor", Environmental Aspects of Nuclear Power Stations (Proc. Symp. New York, 1970), IAEA, Vienna (1971) 535. [32] STIGALL, G .E ., FOWLER, T . W . , KRIEGER, H. L. , ^*1 discharge from an operating boiling water reactor nuclear power station, Health Phys. 20 (1971) 593. [33] BHAT, I. S ., KAM ATH, P. R., Goat thyroids as indicators for routine environmental monitoring of radioiodine, Health Phys. 16 (1969) 65. [34] SIMOES, J . H . , TEIXEIRA, M .R * , ^ 1 in thyroids of cattle and sheep of Portugal, Zentralbl. Veterinaermed. Beih. ¿1 (1970) 101.
DISCUSSION
Pamela BRYANT: I should like to comment on a particular point made by Mr. Hoffman in his interesting paper. He said, rightly, that there is very little information to test the validity of the calculations of transfer of 1311 to milk. We had one such opportunity in the United Kingdom, when short* cooled reactor fuel was reprocessed at Windscale over a period of IAEA-SM-180/29 85 about three weeks. Measurements were made of discharges, concentrations in milk and meteorological parameters. The concentrations in milk were about three times the calculated values. This is considered to be a satis factory agreement when the difficulties of representative stack monitoring and of selection of appropriate values for the various parameters are taken into account. F.O. HOFFMAN: I should just like to add that, although an uncertainty factor of three may be satisfactory for general estimations, it may lead to problems when the results of these estimations approach guideline values. In this case, the general assumptions must be refined with site-specific information, and later the improved estimations should be tested against the monitoring of milk. J. SCHWIBACH: If a guideline were exceeded by a factor of three in a certain calculation, you would not obtain an operating licence. However, in questions of public health, too great an importance should not be attached to such factors, which are really only orders of magnitude for use in discussions. F.O. HOFFMAN: I should like to emphasize that environmental surveillance around the site could show that these calculations are so overly conservative as to be completely unrealistic. On the other hand, an un certainty factor of three, such as Miss Bryant has mentioned, could lead to a case where calculations are deceptively under-conservative. Therefore, I believe that site-specific information should be obtained in order to refine the results of hypothetical calculations, before a critical licensing decision is made. E. NAGEL: You referred to winds blowing across valleys which are deeper than the height of a stack. I do not believe that such winds can blow from one side to the other, sweeping the floor of the valley en route, unless the valley is fairly shallow (a few hundred metres at the most). Moreover, the moving air m asses go up to the top of the valley side and down again behind it, so a contaminant will not be shot against the opposite side in a straight line like an arrow, for example. F. O. HOFFMAN: However, if this effect has not been confirmed by local meteorological monitoring, the more pessimistic assumption would be applied in general assessment calculations. I would definetly like to see more site-specific information in these situations in order to avoid the adoption of assumptions that may be overly pessimistic.
PRE-OPERATIONAL INVESTIGATIONS Chaitman P. PELLERIN (France) IAEA-SM-180/22
ENVIRONMENTAL GAMMA RADIATION MEASUREMENTS IN NUCLEAR POWER STATION SITING STUDIES IN POLAND
B. GWIAZDOWSKI, J. PEÑSKO, J. JAGIELAK, M. BIERNACKA. K. MAMONT-CIESLA Central Laboratory for Radiation Protection, Warsaw, Poland
Abstract
ENVIRONMENTAL GAMMA RADIATION MEASUREMENTS IN NUCLEAR POWER STATION SITING STUDIES IN POLAND. A survey of environmental gamma background radiation in a region surrounding one of the proposed nuclear power station sites in Poland, in the vicinity of Zarnowieckie Lake, has been described. Over a
surface measurements comprised the survey of gamma background exposure dose rate at one metre above the ground at 100 measurement points distributed over a region of 40 km radius. The airborne monitor survey was made by recording the dose rate during flights on parallel paths 2 km apart. The dose rate was determined for ground level over a square net of points spaced at 2 km sides, which was the base used for drawing up the
1. INTRODUCTION
Good knowledge of the true conditions pertaining in the natural environment in a region considered for a nuclear power station site is an important requirement in siting procedures. Investigations and proper recognition of the 'initial status of the environment' before the nuclear power station commences operation are necessary for later estimating any influence of long-lived activity due to the existence of the station on the environment and the neighbourhood population [ 1-3]. Data obtained from such environmental surveillance, whose lack cannot be made good once the power station has entered service, form one of the basic elements needed for ensuing control of the environment during normal operation and in the event of an accident [ 4-6]. The natural gamma background radiation of a particular region of a country is one of the distinctive factors defining the status of the environment. Monitoring of the gamma background dose-rate level permits of control and evaluation of population exposure over wide areas, using comparatively simple measurement techniques [ 7, 8].
2. PROGRAMME OF WORK
The gamma background radiation emanating from the ground proceeds from naturally occurring radioactive elements included in different parts of the environment and from certain artificial radioactive elements, whose sources are radioactive fall-out and industrial activity. The wide ranges of
89 p A L 7 ' С SEA (О о WADWK e al. et GWIAZDOWSKI
F IG .l. The distribution of field measurement points and the airborne monitor measurement track in the environs of Zainowieckie Lake. IAEA-SM-180/22 91 energy of the gamma radiation emitted and the low levels of the activities of these radiation sources require the use of extremely sensitive instruments and appropriate techniques and methods of measurement. The investigations should be extended to cover a wide area around a prospective site, with the aim of supplying reliable information on the level of the gamma background in that area. F or the proposed site on the southeastern bank of the Zarnowieckie Lake, we took as the area in which measurements were to be made an area having a radius of about 40 km centred on the site. The requirem ent of high accuracy of measurement over a large area made a more complex method of survey necessary, that included three separate methods of measurement: (a) Field measurements, using dose meters based on high-pressure ionization chambers; (b) Soil sample analysis, using a low-background scintillation spectro m eter; (c) Airborne monitor measurements. The programme of work covered a three-year period, with 2 to 5 series of field measurements of the exposure dose rate, with soil samples being taken at 100 measurement points in the specified region at the same time, these being subsequently analysed. The programme included a single airborne radiometer survey. The distribution of field measurement points and the air measurement track are shown in Fig. 1.
3. METHODS AND INSTRUMENTATION
To fulfil the requirements of the survey programme, methods of measuring very low exposure dose rates and spectrometric analysis of low- activity soil samples [ 9-11] worked out in the Central Laboratory for Radiological Protection in Warsaw were used.
3.1. High-pressure ionization chamber method
To be able to measure the gamma background dose rate accurately under field conditions, two portable current-type high-pressure ionization chambers were specially made. Such a chamber consists of a 5 litre high-pressure steel bottle of 4 mm wall thickness. It is filled with argon at a pressure of about 35 atm. The chamber is placed within a 1. 5 mm thick aluminium shield mounted on a folding stand which ensures that the chamber is at a fixed height from the surface of the ground. The shield of the chamber can be earthed to prevent disturbances inthe m easuring system (Fig. 2). To minimize the leakage, the main insulator was made of Teflon; a guard ring was also used. The leakage current proved to be as low as 10 ^ A when the current corresponding to the exposure dose rate of the natural background was of the order of 10"^ A. A dynamic capacitor-electrometer (type VA-J-51) was used with the chamber, supplied under field conditions from a 12 V battery. The chambers were calibrated using and ^1 reference sources, following our own proce dure [12]. For calibration purposes, the directional response of the chamber, the effect of scattered radiation and attenuation in the air were taken into account. Details of the calibration method and technique are 92 GWIAZDOWSKI et al.
F IG .2. Portable high-pressure ionization ehambet dose meter for gamma background measurements.
given in a separate paper [13]. The values obtained for the calibration factors for these ionization chamber dose meters lie in the range 2. 5 to 4.0 (nR/h) per (mV/s); this indicates the good sensitivity of the instruments.
3.2. Soil sample analysis
Spectral analysis of the gamma radiation emitted from radioisotopes contained in a soil sample make it possible to evaluate the concentrations of various isotopes in the soil. Such data also enabled us to calculate exposure dose rate values above the surface of the earth at the point of interest [ 9]. Measurements were made using a scintillation probe having a 5 in dia. x 2 in Nal(Tl) crystal in a steel shield having 15 cm thick walls. IAEA-SM-180/22 93
FIG .3. Block diagram of the low-background scintillation spectrometer for soil sample analysis (1. Photomulti plier tube; 2. N a l(T l) crystal; 3. Soil sample container; 4. Low-background measurement castle).
The detector was coupled to a multichannel analyser (Fig. 3). The sample to be investigated was placed in an aluminium container whose shape ensured a good measuring geometry. The weight of a soil sample, well dried and broken up, was about 3 kg. The concentrations of radioisotopes in the sample were obtained by solving a set of equations describing the relations between the count rates from the gamma radiation of particular isotopes or radioisotope families. The ranges were chosen to include the y photopeaks of the isotopes in the soil. The number of ranges and equations in the set must at least be equal to the number of isotopes whose concentrations were to be determined in a sample examined. During the survey, concentrations in the soil of natural potassium (containing 4°K), as well as those of the uranium and thorium families were determined with the aid of specially prepared standard samples. The concentrations of other isotopes, from fall-out, etc., were not estimated because their activities were negligible in comparison with the total activity of the sample, as confirmed by previous investigations [9, 12]. Calculation of isotope concentrations in the samples and the correspond ing contributions to the exposure dose rate of gamma background were made by using a program in Algol for a GIER computer.
3.3. Airborne monitor measurements
An airborne scintillation monitor (type ARS-2; made in the USSR) was used. The radiometer is intended for use in an AN-2 biplane designed for geological prospecting. We adapted the instrument to give measurements of gamma background emanating from the earth. After a careful calibration using ^ R a and ^ 1 reference sources [10], the instrument will give a continuous record of gamma exposure dose rate (inpR/h) with automatic correction for flight altitude changes in the range between 20 and 100 m, with an error of less than 2%. This enables all the measurements to be normalized to the results obtained at a height of 1 m above the earth's surface. 94 GWIAZDOWSKI et al.
F IG .4. Block diagram of the ARS-2 airborne radiometer.
The ARS-2 radiometer incorporates scintillation detectors comprising a block of four 90 mm dia. x 60 mm Nal(Tl) crystals and one photomultiplier. The signal from the detector, after amplification (Fig. 4), passes through a shaping stage and is fed into a pulse ratemeter. The output from the rate- meter is connected through a d.c. amplifier to one channel of a recorder, the other channel being coupled to the plane's altimeter. The radiometer has two ranges, for which full-scale deflection of the recorder was calibrated as being equivalent to 2. 5 pR/h and 50 pR/h. The error in the gamma background radiation measurements with the airborne radiometer under stationary conditions is estimated to be ± 15%. In flight this error may be increased due to mechanical vibration. The measurements of gamma exposure dose rate by means of the air borne radiometer make a rapid estimation possible of the gamma radiation field over large areas of country [ 5, 6, 14]. The method enables the gamma background dose rate distribution to be obtained in the form of a map, which might also prove useful for rapid localization of contaminated areas in the event of an accident.
4. MEASUREMENTS AND RESULTS
The environmental survey encompassed a region of 40 km radius around the proposed site of the nuclear power station to be built at the southeastern side of Zarnowieckie Lake. The surface measuring points were distributed non-uniformly, dividing the whole area into three zones: Zone I, an area of 2 km radius around the lake; Zone II, a circle of 13 km radius around the site but beyond the Zone I area; Zone III, the annular area from 13 to 40 km radius. ад ¿ПС SEA
FÍG .5. Results of airborne radiometer measurements in the Zarnowieckie Lake area. CD О) TABLE I. RESULTS OF ENVIRONMENTAL GAMMA BACKGROUND DOSE RATE MEASUREMENTS (in Ionization chamber Soil sample analysis Airborne No. L o c a l i t y VI V IX VI IX Mean VI V IX VI IX IX 70 - 71 71 72 72 value 70 71 71 72 72 ^llue 70 1 Zarnowieckie Lake 3.3 3.8 3.5 3.0 3.9 3.5 4.8 3.8 6.1 2.9 4.7 4.5 4 - 6 2 Zarnowieckie Lake 2.3 2.2 1.7 3.0 1.9 2.2 2.1 2.4 3.1 1.7 2.3 2.3 4 - 6 3 Zarnowieckie Lake 4.5 4.1 4.3 4.2 3.5 4.1 3.9 5.2 4.4 3.6 4.3 4.3 4 - 6 4 Zarnowieckie Lake 3.7 3.7 3.5 3.2 2.7 3.4 4.2 4.4 4.3 4.2 5.3 4.5 4 - 6 5 Zarnowieckie Lake 3.2 3.0 3.3 4.1 3.7 3.5 3.6 3.6 3.2 3.6 4.6 3.7 4 - 6 6 Kartoszyno 3.5 3.1 3.2 3.2 3.7 3.3 3.2 4.0 3.4 2.4 3.6 3.3 4 - 6 al. et GWIAZDOWSKI 7 Czymanowo 3.7 3.0 3.0 3.4 3.0 3.2 2.9 3.2 4.6 2.6 4.8 3.6 4 - 6 8 Nadóle 4.3 3.7 3.6 3.2 3.5 3.7 4.4 5.6 5.2 4.2 6.2 5.1 4 - 6 8a Nadóle 3.6 2.7 2.6 3.6 2.8 3.1 3.2 3.6 3.2 2.7 3.5 3.2 4 - 6 9 Brzym 3.4 2.6 3.1 3.5 4.3 3.4 4.5 4.8 5.5 5.2 6.7 5.3 4 - 6 10 Zamowiec 3.6 3.3 3.0 2.3 3.0 3.0 2.8 2.8 3.2 2.4 3.4 2.9 4 - 6 11 Dçbek 2.0 -- 2.1 1.8 2.0 1.9 1.9 - 1.5 2.5 2.0 4 - 6 12 Karwieñskie Mota 1.9 1.2 1.5 1.9 1.6 1.6 1.7 2.0 2.1 1.6 - 1.9 4 - 6 13 Szary Dwór 2.7 1.5 2.0 3.0 2.2 2.3 3.2 2.9 3.1 2.2 2.3 2.7 4 - 6 14 Karwia-Sulicice 3.3 2.1 2.3 1.9 2.6 2.4 4.6 5.9 3.0 3.5 7.7 4.9 4 - 6 15 Odargowo 4.3 3.7 2.8 3.5 4.0 3.7 3.4 4.8 4.5 3.6 5.1 4.3 6 - 8 16 Sulicice 5.7 4.3 4.2 4.0 4.9 4.6 6.6 6.4 6.5 6.9 7.1 6.7 6 - 8 17 Krokowa 5.7 4.0 4.3 4.5 4.5 4.6 5.2 6.6 7.3 5.4 5.0 5.9 6 - 8 18 Jeldzino 5.1 4.2 3.2 4.1 3.7 4.1 5.6 5.9 6.5 5.0 6.2 5.8 6 - 8 19 Lisewo 5.9 5.1 5.2 5.1 4.5 5.2 5.7 7.8 5.9 4.3 5.8 5.9 6 - 8 20 Kianino 4.6 4.9 - - 4.9 2.8 4,3 6.2 -- 8.3 7.9 7.5 6 - 8 21 Dobre Lake 4.9 4.1 4.5 4.8 4.8 4.6 4.1 4.5 5.8 - 3.2 4.4 4 - 6 22 Warszkowo- Opalino 4.7 3.9 - 3.0 3.1 3.7 3.6 4.1 - 2.3 5.0 3.8 6-r 8 23 Piasnica 3.5 2.6 2.1 2.7 2.6 2.7 3.1 4.6 3.1 - 3.8 3.7 6 - 8 24 Warszkowo 5.8 5.4 4.4 4.4 4.8 5.0 6.3 9.3 6.7 5.7 7.3 7.1 6 - 8 25 Lesniewo-Darzlubie 4.7 3.7 2.8 4.2 3.9 3.9 5.1 5.4 5.3 4.3 5.3 5.1 6 - 8 T A B L E I. (cont. ) Ionization chamber Soil sample analysis Airborne monitor No. VI V IX VI IX Mean VI V IX VI IX Mean IX 70 71 71 72 72 value 70 71 71 72 72 value 70 26 Mechowo 4.4 4.4 3.8 4.2 4.2 4.2 4.4 6.6 5.8 3.2 7.7 5.5 6 - 8 27 Domatowo- Starzyno 3.8 3.2 3.2 2.8 - 3.3 4.0 4.3 4.5 4.3 - 4.3 6 - 8 28 Lisewo-Kiani.no 1.4 1.0 --- 1.2 1.4 1.6 - - - 1.5 6 - 8 29 Wejherowo- Piasnica 4.6 3.0 2.6 3.0 2.2 3.1 4.7 2.8 2.5 2.2 4.6 3.4 4 - 6 30 Oile 3.5 2.0 - 2.1 3.2 2.7 4.5 3.6 - 2.7 5.3 4.0 6 - 8 31 Rybno Kaszubskie 3.8 3.2 2.8 2.5 3.4 3.1 3.1 3.8 4.1 3.2 4.3 3.7 6 - 8 32 Kniewo 3.6 2.2 2.4 2.2 3.4 2.8 3.8 5.0 5.7 4.3 4.7 4.7 6 - 8 33 Kochanowo Lesne 4.7 4.1 2.8 3.7 - 3.8 5.4 5.7 4.5 4.6 - 5.1 6 - 8 IAEA-SM-180/22 34 Chynowie 3.6 2.9 2.7 2.5 3.1 3.0 2.5 3.7 2.3 3.1 3.6 3.0 6 - 8 35 Tadzino 4.6 4.0 3.5 4.0 3.5 3.9 4.6 5.4 5.8 4.3 5.6 5.1 6 - 8 36 D^browka Ш . 4.6 4.3 - 4.2 4.4 4.4 4.4 5.4 - 2.5 5.1 4.4 4 - 6 37 Wysokie 5.1 4.1 - - - 4.6 5.1 5.1 5.1 - - 5.1 6 - 8 38 Salino 4.4 4.7 --- 4.6 4.9 4.0 --- 4.5 4 - 6 39 Mierzyno 4.5 3.1 2.9 3.3 3.3 3.4 3.4 4.9 5.0 3.5 5.0 4.4 6 - 8 40 Koükowo 5.2 3.7 3.3 4.7 3.9 4.2 4.4 6.0 5.0 4.4 5.6 5.1 8 - 10 41 Gniewino 5.8 4.4 4.6 4.0 4.3 4.6 6.2 6.4 7.9 5.3 5.9 6.3 6 - 8 42 Lublewko 5.1 4.4 3.6 4.0 3.8 4.2 5.9 6.2 5.4 4.0 6.3 5.6 6 - 8 43 Choczewskie Lake 3.1 2.5 2.0 1.2 - 2.2 2.5 3. 0 1.7 2.3 - 2.4 4 - 6 44 Osieki l^borskie 5.1 4.4 3.6 4.1 3.8 4.2 4.4 5.1 4.5 4.1 5.5 4.7 6 - 8 45 Salinka 4.9 3.9 3.6 3.4 3.7 3.9 5.2 5.7 5.1 4.2 5.5 5.1 6 - 8 46 Bychowo 6.7 5.3 4.9 3.7 3.3 4.8 8.3 7.9 9.2 3.6 6.8 7.2 6 - 8 47 Siuchowo 5.7 5.0 4.0 4.5 4.5 4.7 5.1 6.3 5.5 4.8 6.5 5.6 6 - 8 48 Bia^ogora 2.4 1.8 1.4 2.6 1.7 2.0 2.8 2.2 2.0 2.0 2.3 2.3 4 - 6 49 Opalino 5.3 3.6 3.2 4.2 1.3 3.5 5.9 5.5 4.2 5.3 4.1 5.0 4 - 6 50 Wierzchucino 3.1 2.7 2.7 2.3 2.8 2.7 3.3 4. 0 4.3 2.7 4.0 3.7 4 - 6 (О -J со со T A B L E I. (cont. ) Ionization chamber Soil sample analysis Airborne monitor No. L o c a l i t y VI V IX VI IX VI V IX VI IX Mean ' IX 70 71 71 72 72 tt!ue 70 71 71 72 72 value 70 51 Pobibcie Wierzch. 3.6 2.8 - 3.2 2.4 3.1 -- 2.8 6 - 8 52 Wicko 4.9 4.2 - 4.6 4.5 4.7 - - 4.6 8 - 1 0 53 Roszczyce 4.5 2.5 - 3.5 2.9 3.4 - - 3.2 6 - 8 54 Zwartowo 7.1 5.1 - 6.1 6.6 7.3 - - 7.0 6 - 8 55 Szczenurze 2.7 2.1 - 2.4 1.3 2.1 -- 1.7 4 - 6 - -- 56 Ulinia 2.6 2.3 2.5 1.9 2.2 2.1 4 - 6 al. et GWIAZDOWSKI 57 Ciekocino-Sasino 6.2 4.7 5.5 5.6 6.1 -- 5.9 6 - 8 58 -Eeba 2.8 1.6 -- 2.2 2.1 2.2 - - 2.2 4 - 6 59 Kopalino 3.2 2.7 -- 3.0 2.6 3.1 - - 2.9 4 - 6 60 Lubiatowo 3.4 2.9 2.1 - - 2.8 2.1 2.8 2.2 - 2.4 4 - 6 61 Jastrz^bia Góra 3.4 1.7 3.7 - 2.9 2.6 2.1 - 1.9 2.2 4 - 6 62 Jastamia 1.9 1.7 0.9 - 1.5 1.9 2.1 - 2.0 2.0 4 - 6 63 Swarzewo 4.4 3.5 3.5 - 3.8 5.8 6.9 - 5 . 6 . 6.1 4 - 6 64 Cha^upy 1.4 3.1 2.6 - 2.4 1.2 - - 1.2 1.2 4 - 6 65 Rozewie 4.2 4.5 4.5 - 4.4 3.9 5.6 - 3.3 4.3 4 - 6 66 Zelistrzewo 4.7 4.3 --- 4.5 4.7 5.0 -- 4.9 6 - 8 67 ^ebez 3.6 2.8 - - 3.2 4.7 4.8 - - 4.8 8 - 1 0 68 Wejherowo 3.7 2.9 - - 3.3 2.8 3.4 - - 3.1 4 - 6 69 Celbowo 4.5 - - 4.5 7.6 6.8 -- 7.2 6 - 8 70 Mrzezino 2.9 1.8 - - 2.4 3.7 4.0 - - 3.9 4 - 6 71 Kosakowo 3.9 2.8 - 3.4 2.9 3.2 -- 3.1 6 - 8 72 Rewa 2.4 1.6 - 2.0 2.4 2.1 -- 2.3 4 - 6 73 Reda 3.9 4.0 - 4.0 4.9 5. 5 - - 5.2 4 - 6 74 Naniec 4.7 3.9 - 4.3 3.5 3.8 - - 3.7 4 - 6 75 Reda 3.7 3.3 3.5 4.6 4.2 4.4 6 - 8 ...... T A B L E I. (cont. ) Ionization chamber Soil sample analysis No. VI V IX VI IX VI V IX VI IX Mean IX 70 71 71 72 72 ^ u e 70 71 71 72 72 value 70 76 Gdynia-Chy Ionia 3.4 --- 3.4 4.0 3.8 - 3.9 4 - 6 77 Chwaszczyno 3.1 3.0 -- 3.1 4.8 4.9 -- 4.9 * 78 Karnien 3.1 3.0 -- 3.1 3.3 6.1 - - 4.7 - 79 Nowy Dwor Wejh. 4.7 4.0 - - 4.4 4.8 5.2 -- 5.0 4 - 6 80 Wejherowo-Biafa 3.8 3.9 -- 3.9 4.3 4.8 -- 4.6 4 - 6 81 tebno 2.5 2.4 -- 2.5 4.2 5.0 -- 4.6 - 82 Sianowo Lesne 4.4 3.4 - - 3.9 4.4 5.4 - - 4.9 - 83 Goscicino 4.6 3.9 - - 4.3 4.9 5.6 - - 5.3 4 - 6 84 Czçstkowo 5.0 4.8 - - 4.9 5.6 5.5 - - 5.6 4 - 6 85 Sczepcz 4.7 4.7 - - 4.7 6.0 6.0 - - 6.0 - 86 Wyszecino 3.8 4.8 - 4.3 3.9 --- 3.9 4 - 6 87 T&czewo 5.0 3.3 -- 4.2 5.1 3.7 -- 4.4 - 88 Mirachowo 4.2 3.9 - - 4.1 3.8 4.2 - - 4.0 - 89 Bukowina 4.6 3.8 - - 4.2 3.7 4.0 -- 3.9 - 90 Xebunia 4.5 4.0 -- 4.3 5.5 5.7 -- 5.6 - 91 D§browa Góra 4.9 4.2 -- 4.6 4.9 5.0 - - 5.0 6 - 8 92 Dziçcieiec 4.5 4.0 -- 4.3 4.5 5.6 - - 5.1 4 - 6 93 LubowidzLake 2.3 1.5 - - 1.9 2.7 2.1 - - 2.4 4 - 6 94 Osowo- Maszewo 4.2 3.9 - - 4.1 3.5 6.0 - - 4.8 - 95 Pogorzelica 4.2 3.1 -- 3.7 3.2 3.4 -- 3.3 4 - 6 96 Zçczyce 3.4 2.6 -- 3.0 3.2 3.4 - - 3.3 4 - 6 97 Swichówko 5.2 4.0 -- 4.6 4.9 5.2 - - 5.1 6 - 8 98 Garczegorze 5.7 6.0 -- 5.9 7.4 7.3 -- 7.4 , 6 - 8 99 Redkowice 3.6 2.8 - - 3.2 2.9 4.1 -- 3.5 4 - 6 100 ¿iebieñ 6.3 4.4 *- 5.4 4.9 3.1 - - 4.0 6 - 8 GWÍAZDOWSKI et al. II. CONCENTRATIONS OF NATURAL POTASSIUM, URANIUM :ORIUM IN SOIL SAMPLES AND APPROPRIATE EXPOSURE DOSE 5 FOR THE ZARNOWIECKIE LAKE AREA (Sep. 1972) Concentrations Exposure dose С к C-rh ?U ?Th ^tot. (g/g) (Mg/g) (^g/g) (pR/h) (fR/h) (MR/h) 1 0.017 1.05 2.91 2.8 0.8 1.1 4.7 2 0.008 0.57 1.40 1.3 0.4 0.5 2.3 3 0.014 1.04 2.94 2.4 0.8 1.1 4.3 4 0.016 1.88 2.96 2.7 1.5 1.1 5.3 5 0.015 1.15 3.21 2.5 0.9 1.2 4.6 6 0.012 0.79 2.50 2.1 0.6 0.9 3.6 7 0.014 1.72 3.06 2.3 1.3 1.1 4.8 8 0.019 2.17 3.88 3.1 1.7 1.5 6.2 8í 0.010 1.21 2.51 1.6 0.9 0.9 3.5 9 0.020 2.10 4.71 3.3 1.6 1.8 6.7 10 0.011 0.85 2.28 1.8 0.7 0.9 3.4 11 0.011 0.38 0.75 1.9 0.3 0.3 2.5 13 0.006 0.79 1.71 1.0 0.6 0.6 2.3 14 0.022 2.43 5.96 3.6 1.9 2.2 7.7 15 0.016 1.45 3.41 2.7 1.1 1.3 5.1 16 0.022 2.03 4.94 3.6 1.6 1.9 7.1 17 0.017 1.26 2.96 2.9 1.0 1.1 5.0 18 0.020 1.81 4.04 3.3 1.4 1.5 6.2 19 0.016 2.02 3.91 2.7 1.6 1.5 5.8 20 0.021 2.99 5.33 3.6 2.3 2.0 7.9 21 0.010 1.00 2.08 1.7 0.8 0.8 3.2 22 0.014 1.53 3.86 2.3 1.2 1.4 5.0 23 0.012 1.00 2.47 2.1 0.8 0.9 3.8 24 0.023 1.79 5.42 3.9 1.4 2.0 7.3 25 0.017 1.51 3.28 2.9 1.2 1.2 5.3 26 0.023 2.25 5.79 3.8 1.8 2.2 7.7 29 0.013 2.23 1.88 2.2 1.7 0.7 4.6 30 0.015 1.81 3.77 2.5 1.4 1.4 5.3 31 0.013 1.37 2.86 2.1 1.1 1.1 4.3 32 0.014 1.66 2.93 2.3 1.3 1.1 4.7 34 0.013 0.86 2.30 2.1 0.7 0.9 3.6 0.016 1.93 3.85 2.7 1.5 1.4 5.6 IAEA-SM-180/22 1 0 1 TABLE II. (cont. ) No. Ск Си C-rh ?K ?Th Ptot. (g/g) (Mg/g) C g/g) (pR/h) (pR/h) (pR/h) (MR/h) 36 0.015 1.36 3.94 2.6 1.1 1.5 5.1 39 0.016 1.37 3.21 2.7 1.1 1.2 5.0 40 0.018 1.41 3.70 3.1 1.1 1.4 5.6 41 0.017 1.85 4.01 2.9 1.4 1.5 5.9 42 0.020 1.72 4.44 3.3 1.3 1.7 6.3 44 0.018 1.62 3.43 3.0 1.3 1.3 5.5 45 0.017 1.77 3.38 2.8 1.4 1.3 5.5 46 0.019 2.44 4.75 3.1 1.9 1.8 6.8 47 0.020 1.92 4.10 3.4 1.5 1.5 6.5 48 0.008 0.53 1.54 1.3 0.4 0.6 2.3 49 0.012 1.22 2.89 2.0 1.0 1.1 4.1 50 0.012 1.30 2.72 2.0 1.0 1.0 4.0 Taking into account the technical measurement possibilities, we decided to place 10 points in Zone I, 40 points in Zone II, and 50 points in Zone III (F ig .l). A ll field m easurem ents were made in the y ears 1970, 1971 and 1972, taking as a rule five series of measurements in Zones I and II, and two series in the Zone III. Some exceptions were made for the points at the sea coast. During the survey accurate descriptions were filed of the location of each point, including photographic records, and details included descrip tion soil samples taken. The measurements from the aeroplane were made in September of 1970. The area of the survey was covered by parallel flight paths spaced 2 km apart. The results obtained are shown in Fig. 5. A sum m ary of the survey resu lts are given in Table I. It com prises all results of the ionization chamber measurements of exposure dose rate at the 100 points, and the mean value of the results for each point. The results of dose rate calculations from soil sample analyses are also shown in the table. An example of results of this analysis, for one series of measurements, is presented in Table II. This table gives the concentrations of uranium, thorium and potassium in the soil sample, appropriate dose rate values and calculated total gamma background exposure dose rate in pR/h. A gamma background dose rate distribution chart for the Zarnowieckie Lake area was prepared from the airborne radiometer measurements, as shown in Fig. 6. The range of dose rate of this map, corresponding to each surface measuring point, is also given for comparison in the last column of Table I. 102 WADWK et l. ta e GWIAZDOWSKI FIG. 6. Gamma background distribution in the Zarnowieckie Lake area, obtained from the airborne radiometer measurements, September 1970. IAEA-SM-180/22 103 5. DISCUSSION OF RESULTS The experience gained in this survey shows that the airborne radiometer survey is a quick method that enables a rapid assessment of the radiation field distribution over large areas of the country to be made, while ensuring continuity of the measurement to a great extent. Portable high-pressure ionization chambers give great accuracy when making radiation exposure dose measurement under field conditions. Spectrometric analysis of soil samples identifies the important isotopes and enables one to apportion the total radiation dose rate between them. Joint application of these methods ensures a good accuracy of measure ment, as confirmed by the good correlation between the various results, as shown in Fig. 7: such correlation is difficult to obtain in long-term studies in the lowest ranges of dose rate measurements. It also makes it possible to perform large-scale environmental studies in a comparatively short period of time. HIGH PRESSURE tONIZATION CHAMBER: DOSE RATE („R/h) F IG .7. Regression line calculated using the method of least-squares and the standard deviation of the means of the gamma background exposure dose rate obtained from ionization chamber measurements in the field and from spectrometric analysis of the soil samples, each taken for 100 measurement points during the period 1970-72. 104 GWIAZDOWSKI et al. The resu lts obtained from the survey in the Zarnowieckie Lake area give a good picture of the level and distribution of the gamma background radiation field. As one can see from Table I, the value of the dose rate is in general quite low - lying in the range from about 2 pR/h to about lO^R/h. One can observe the ch aracteristic decrease of the dose-rate value in the vicinity of and above the lakes and the sea. The comparison of results of measurements at the same points over the period of three years shows good consistency. As can be seen from Table II, there is a clear correlation between total dose rate and the potassium content of the soil - also a characteristic of regions with a low natural gamma background. In our opinion the data that have been gathered provide a sufficient basis for future control of the environment both in the vicinity of the nuclear power station and in the surrounding region. This will make it possible to estimate any changes in the dose to the population living in the area. REFERENCES [1] RAMEY, J . T . , "Environmental considerations in the regulatory process for nuclear power plants in the USA: The role of the public and public understanding", Environmental Aspects of Nuclear Power Stations (Proc. Symp. New York, 1971), IAEA, Vienna (1971) 627. [2] SAIKI, M ., KOYANAGI, Y ., OHMOMO, Y ., UCHIYAM A, M ., "Environmental considerations and public acceptance of nuclear plant site selection in coastal areas in Japan", Environmental Aspects of Nuclear Power Stations (Proc. Symp. New York, 1971), IAEA, Vienna (1971) 821. [3] DUNSTER, H .J., "Environmental monitoring: British policy and procedures", Environmental Aspects of Nuclear Power Stations (Proc. Symp. New York, 1971), IAEA, Vienna (1971) 427. [4] EISENBUD, M ., "Review of USA power reactor operating experience", Environmental Aspects of Nuclear Power Stations (Proc. Symp. New York, 1971), IAEA, Vienna (1971) 861. [5] WACHSMANN, F ., SCHWIBACH, J . , "Considerations for nuclear power plants in areas with high population density in the Federal Republic of Germany", Environmental Aspects of Nuclear Power Stations (Ptoc. Symp. New York, 1971), IAEA, Vienna (1971) 791. [6] NASIM, M ., "Environmental aspects of Karachi nuclear power plant", Environmental Aspects of Nuclear Power Stations (Proc. Symp. New York, 1971), IAEA, Vienna (1971) 781. [71 PENSKO, J . , "Natural gamma background radiation in the enviions of nuclear facilities: Pre-operationai analysis of dose to the neighbouring population", these Proceedings, paper IAEA-SM -180/23. [8] BECK, H.L., LOWDER, W .M ., McLAUGHLIN, J.E., "In situ external environmental gamma-ray measurements utilizing Ge(Li) and Nal(Tl) spectrometry and pressurized ionization chambers", Rapid Methods for Measuring Radioactivity in the Environment (Proc. Symp. Neuherberg, 1971), IAEA, Vienna (1971) 499. [9] PENSKO, J . , JAGIELAK, J . , MAMONT, K ., Spectrometric methods of measurement of the terrestrial gamma background radiation and concentration of radionuclides in the soil, Nukleonika 6 15 (1970) 573. [10] PENSKO, J., BIERNACKA, M., GWIAZDOWSKI, B., NIEDZIOtKA, E., MAMONT-CIESLA, K., monitor, Rep. CL0R-87/D (1971). [11] PENSKO, J., GWIAZDOWSKI, B., JAGIELAK, J., BIERNACKA, M., MAMONT, K., "Combined environmental radioactivity measurements for rapid estimation of the gamma radiation field", Rapid Methods for Measuring Radioactivity in the Environment (Proc. Symp. Neuherberg, 1971), IAEA, Vienna (1971) 443. [12] PENSKO, J . , JAGIELAK, J . , GWIAZDOWSKI, B ., ZAK, A ., Automatic methods of continuous measure ment of earth gamma background exposure dose, Rep. CL0R-92/D (1972). [13] JAGIELAK, J., GWIAZDOWSKI, B., PENSKO, J. , ZAK, A., "Problems in calibrating instruments for environmental gamma exposure dose measurements", these Proceedings, paper IAEA-SM -180/21. [14] BECK, C . K . , "Environmental monitoring for USA licensed power reactors", Environmental Aspects of Nuclear Power Stations (Proc. Symp. New York, 1971), IAEA, Vienna (1971) 401. IAEA-SM-180/22 105 DISCUSSION О. VANDERBORGHT: Have you done any measurements in Poland on the effect of type of housing on the radiation dose to the population, that is to say the effect of building materials such as bricks, concrete and wood or the effect of high buildings shielding against cosmic rays, etc. ? B. GWIAZDOWSKI: We have carried out measurements inside buildings and the resu lts are reported in Nukleonika^. In the near future we plan to measure the radioactivity of building materials and investigate the community dose due to the use of different building materials. L.I. GEDEONOV: You described some very interesting and sophisticated methods of determining gamma ray doses due to the natural radioactivity of the soil. What levels of surface contamination from artificial radionuclides can you detect with these methods? B. GWIAZDOWSKI: As we have shown in an earlier paper (Ref. [11] of the paper), an activity of 1. 5 pCi/m ^ of ^^1 deposited on the ground gives a true dose-rate value of about 13. 0 prad/h. The value of 1. 5 pCi/m^ is equivalent to 1. 5 Ci/km^. Using the ionization chamber method, dose rates of about 1 prad/h are easily detectable, so we can assume that a contamination of about 100 mCi of ^ I/k m ^ can easily be detected. With the spectrom etric soil analysis technique, the limit of detectability is of course much lower. * PENSKO, J.. MAMONT, K ., WARDASZKO, T ., Measurements of ionizing radiation inside blocks of flats in Poland, Nukleonikal4 4 (1969) 93: Eng. trans. byUSAECasAEC-tr-7027/4 UC-34 TT 69-50012/4. IAEA-SM-180/27 STUDY OF THE BACKGROUND RADIATION AROUND THE CONSTRUCTION SITE OF A NUCLEAR POW ER STATION IN BULGARIA W ITH A VIEW TO CONTROLLING THE RADIOLOGICAL HEALTH OF THE POPULATION IN THE FUTURE G. VASILEV, M. YOTOV, D. KESLEV Institute of Radiobiology and Radiation Hygiene, Sofia, Bulgaria Abstract STUDY OF THE BACKGROUND RADIATION AROUND THE CONSTRUCTION SITE OF A NUCLEAR POWER STATION IN BULGARIA WITH A VIEW TO CONTROLLING THE RADIOLOGICAL HEALTH OF THE POPULATION IN THE FUTURE. The first nuclear power station in Bulgaria is being constructed near the town of Kosloduy. Since 1968 a study of the different factors that characterize the natural background radioactivity in the region around the station has been in progress. With a view to the future, when the power station is operational, a special control system has been created for checking the radiological status of the health of the population in the surrounding region. 1. INTRODUCTION The First Nuclear Power Station in Bulgaria is being constructed near Koslöduy, a town in northern B u lgaria lying on the Danube. The first reactor is due to enter into operation during 1974. Over the next ten years, the capacity of the nuclear power station will be increased by adding five further reactors to give a total station output of 2 600 MW(e). All reactors will be of water-water type, manufactured in the USSR. The site for the nuclear power station has been chosen on the basis of the following considerations [1]: (a) There is a demand for electricity in this area of the country; (b) The nuclear power station is sited on the banks of the Danube and river water will be used in the secondary coolant circuit; (c) The area is comparatively sparsely populated; (d) The nearest town downwind of the plume of any radioactive products emanating from the stacks is situated about 80 km away. 2. RADIATION M EASUREM ENTS The region of radius 3 km around the nuclear power station is defined as a protected zone for public health purposes and contains no populated sites. There is additionally a controlled zone having a radius of 12 km. The protected zone will be used for intensive checking of the radiation situation around the reactors. 107 108 VASILEV et al. Since 1968, that is five years before the nuclear power station was due to come into operation, a special programme was set up for studying the natural background radiation. Forty control points were defined around the nuclear power station in the controlled zone and in the direction of the main plume to a distance of 80 km. 2.1. Atmosphere Atmospheric samples were taken four times monthly, using continuous sampling over a period of 24 hours. Monitoring points were situated in populated places, along roads carrying heavy traffic and in places with green vegetation. The total beta and alpha activities for the different seasons were deter mined. A sedimentation method was used at some points to determine the total beta activity and ^°Sr and ^ C s contents due to fall-out. 2.2. The Danube Water samples and bottom sediment are taken in each season from the Danube for checking the total beta radioactivity, and the ^°Sr an d ^ C s contents. Once in each three months hydrophytes, plankton and fish are studied, and a number of factors are determined. The points chosen for control of radioactivity in the Danube are situated both upstream and downstream of the nuclear power station. The control point furthest from the site is placed at distance 250 km downstream. 2.3. Soil and agricultural products Soil and agricultural samples were examined every six months. (Snow was checked once each year, when the thaw had set in. ) The most typical agricultural products for the region were chosen, namely, wheat, grapes and certain common vegetables. Samples were taken once a year, during harvesting. Samples of milk were examined every month, being pooled from milkings extending over a whole week. Meat and bones were checked twice during the year, together with soil and plantation samples. The above samples were studied using standard radiochemical and radiometric methods [2]; the total beta activity and the 9°Sr and ^ C s contents were determined. 2. 4. Natural y -radiation background The natural background is measured by network of measuring points in the controlled zone and in direction of the plume. This has made it possible to determine the gamma exposure dose rate 0.5 m from the ground surface and the integral dose for three-month periods. The values of background gamma radiation were lower than those found in other regions of the country. This is explained by the fact that Koslöduy is situated in the Danube valley, which is of alluvial type with a low content of heavy metals. Values of the fall-out radiation are similar to those found in other countries having a similar geographic disposition [3]. [AEA-SM-180/27 109 TABLE I. INTEGRAL DOSES OF RADIATION ABOVE THE BACKGROUND EXPECTED WHEN THE KOSLÓDUY STATION IS IN OPERATION integral dose Zone (rem/a) Protected zone for public health purposes 5 x 1 0'ho 40 x 10*3 (3 km radius around site) Controlled zone 0.1 x 10*з to 5. 0 x 10*з (12 km radius around site) Natural background in this region 80 x 10*3 This has enabled us to obtain a satisfactorily full picture of the natural radioactive background in the region surrounding the Koslöduy Nuclear Power Station. The basic criterion for defining the radioactive safety in the region surrounding the nuclear power station is to be the integral dose of irradi ation received by the population living in the regions. Calculations were made to obtain an estimate of the integral doses of irradiation of individuals living in the region with the power station assumed as operating at normal full load. The integral doses expected at the zone boundaries are shown in Table I. Systems of radiation protection at the nuclear power station must ensure the minimum possible contamination of the environment and, in particular, of the Danube. A special system for keeping a control on the radiation situation in the region surrounding the nuclear power station when it begins operation has been created. It uses stationary dosimetry posts situated in 'protected' and 'controlled' zones. These posts are part of the power station's own system for interior and exterior dosimetric control. In parallel with this first system is a second dosimetric control system placed in the 'controlled' zone and in the plume. This second system is operated for the Ministry of Health by the State public health authority. Both systems, independently of one another will record radioactivity in the atm osphere, the gam m a background, and the radioactivity in the Danube, the soil and agricultural products. Similar methods and timing will be used. The method used to determine the mean discharges of radioactivity from the stacks will be sensitive enough to define the quantity of radioactive iodine released to the atmosphere. The information about radioactivity in the environment gathered up to the present (1973) before the Koslöduy Station goes on stream is sufficient to allow estimates to be made of the actual irradiation dose to the population and to estimate any contamination of environment in the future. These data will also be used as a criteria when choosing sites for other nuclear power stations which are to be built in Bulgaria in future years. 1 1 0 VAStLEV et al. REFERENCES [1] Siting ofReactors and Nuclear Research Centres (Proc. Symp. Bombay, 1963), IAEA, Vienna (1963). [2] Methods of Radiochemical Analysis, World Health Organization, Geneva (1966). [3 ] RUSSELL, R. S. , Ed., Radioactivity and Human Diet, Pergamon Press, London, New York (1967). DISCU SSION N. G. GUSEV: What b ase s did you use for calculating the doses to the population from the nuclear power station? G. VASILEV: We have considered only the external radiation dose to the human body for normal full-load operation of the station. W. M. BURKHARDT: How many m easuring points around the nuclear power station are operated by the station itself, and how many by the Ministry of Health? G. VASILEV: The external radiation monitoring system run by the station itself consists of 25 stationary dosimetric points located in the 'protected' and in the 'controlled' zones. The second system, run by the Ministry of Health, consists of 15 measuring points located in the controlled zone and outside it — along the direction of the plume for a distance of up to 80 km. P. PELLERIN (Chairman): Does the reactor at the Koslöduy station have a double coolant circuit with heat exchanger or does the water cooling the core p a ss straight into the turbine? G. VASILEV: The reactor is cooled by a double-circuit system and the water in the second circuit is not radioactive. P. PELLERIN: How has the local population reacted to the prospect of living near a nuclear power station? G. VASILEV: There is of course some apprehension amongst the general public and we plan to launch an information campaign via the press, radio and television. IAEA-SM-180/74 M OVEM ENT OF RADIONUCLIDES IN THE GROUND IN RELATION TO THE ENVIRONM ENTAL SAFETY OF NUCLEAR POWER PLANTS Z. DLOUHY, O. áÍAFÁR Institute of Nuclear Research, Йе2, near Prague, Czechoslovakia Abstract MOVEMENT OF RADIONUCLIDES IN THE GROUND IN RELATION TO THE ENVIRONMENTAL SAFETY OF NUCLEAR POWER PLANTS. During the operation of nuclear power plants, damage to any of the systems in which liquid radioactive wastes are transported or stored might occur; this could lead to contamination of the ground. To estimate the size of the hazard of population exposure in the environs of the plant, it is necessary to estimate the consequences of such an accident in advance. This requires a detailed knowledge of the underground water flow as well as the sorption properties of the soils which form the geological structure of the zone at and atound the site. To this end, the authors have evaluated a site on which the building of a 1460 M W(e) nuclear power plant is proposed for the period 1978-83. 1. INTRODUCTION Problems of storage of radioactive wastes produced during the operation of nuclear power plants can be solved by applying the principle known as double containment. Thus, the probability of radioactive substances escaping into the environment is essentially decreased. Storage tanks for individual waste types are usually constructed throughout of stainless steel, and they are situated in concrete cells, the bottoms and walls of which are lined either with stainless or carbon steels. If a tank fails, the wastes are trapped in the cell, leakage of any tank being signalled. The cell content and the waste remaining in the failed tank are then pumped to a stand*by tank. Radioactive liquid could escape into ground only in the event of the simultaneous failure of both containments, an event of very low probability, but even this possibility cannot be completely excluded. It must be borne in mind that radioactive wastes will have to be stored in the tanks for periods of at least 25 years and that the composition of the contents cannot be deter mined accurately at present, especially any corrosive components therein. However, a more probable event is a leakage of a valve in the lines carrying the liquid waste to the storage tanks, due perhaps to wear over a long period of use. In such a case, however, the amounts of waste which could escape into the ground are much lower than in the case of failure of a large-capacity tank. Finally, a third case that might result in contamination of the ground could occur during temporary storage of 'fixed' radioactive wastes. After some time, operation of a nuclear plant will result in wastes that must be solidified in some way, probably in a cheap and efficient plant in which the waste is reduced in volume and 'fixed' in concrete, bituminous-concrete, etc., to form blocks. These blocks must then be stored temporarily on the 1 1 1 1 1 2 DLOUHŸ and SÍAFÁR site of the nuclear plant before they are transported to a final storage. The contamination might arise due to failure of the coating of these blocks, and contact of the solidified waste with rain water, with subsequent leaching of some of the mass might occur. The aim of our work is, therefore, to analyse the consequences of a possible contamination of the ground in the environs of a nuclear power plant site and to estimate the effect on environmental safety. The results of research work carried out partly inthe field (taking soil samples, determining water table levels, etc. ) and partly in the laboratory (determination of sorption and dispersion properties of soils, have served to establish the range and probable direction of the penetration of radioactive substances, in addition to determining the effects on drinking water sources in the vicinity of the site. One of the localities evaluated with this problem in mind is the proposed site in South Moravia for a nuclear power plant of the New Voronezh type, to have a power output of 1760 MW(e). 2. GEOLOGICAL AND HYDROGEOLOGICAL CONDITIONS OF THE SITE The site is in the southern part of the Namësï region and is situated at an altitude of 385 to 395 m above sea level. The boundaries of this moderately modelled terrain are formed by the valley of the Jihlava River and to the south by the O lesna brook. The site itself lies on the top of a flat terrain wave. The orographic watershed of the streams mentioned passes through this rise. The bedrock of the region is formed by typical moldanubic metamorphed rocks, which may be found in the overlying layer in all stages of trans formation induced by weathering processes. The decomposed, weathered or partly weathered, and even compact, rocks retain their layer structure in the covering mantle, the upper part of which is formed by quaternary loams to sandy clays with a surface, not-too-thick humus layer. Some neogene sediments (clayey sands, sands to sandy gravels) are also present, pre dominantly at higher sites. These sediments are local neogene basins which remained from previous extended sediments of neogene (miocene) age and which were denuded due to erosion processes. The argillaceous cover is formed predominantly by sandy clays. The terrain surface is formed by humus clays (top soil) 0.2 to 0.4 m in thickness, and partially humus loams (subsoil layer) attaining thicknesses of 0.4 to 0.9m. The thickness of argillaceous cover after subtracting the humus layer varies between 0.2 and 3.5 m. The metamorphed moldanubic rocks correspond stratigraphically with the so-called many-coloured formation, composed exclusively of orthoslates. Their boundary passes through the plant site itself. The rock types comprise among others: (a) amphibolites, forming concordant inclusions in other rocks; (b) granullites or granullitic gneisses, which form the whole eastern part of the plant site but are not extended to the area of more particular interest; (c) magmaticbiotitic gneisses, which are subjected to weathering processes, often to considerable depths with respect to their minéralogie composition; (d) granites, which form the bedrock of the western part of the plant site, i.e. in the area of particular interest. During our work, we made test boreholes in the range of the upper layers of the covering formations, i.e. in the argillaceous layers, and decomposed and weathered rocks. Other data were taken from existing sources [ 1 ]. IAEA-SM-180/74 113 0 4 2 BOREHOLE No О----- DIRECTION OF GROUND WATER FLOW FIG. 1. Site of proposed nuclear power plant. The decomposed rocks, eluvia in the locality studied, represent the rock, initially hard rock, which has taken on the character of predominantly incohesive earths. Coarse sands prevail among them, while to a lesser extent they contain argillaceous sandy gravel and sandy clay. The thickness of weathered rock varies from 0.6 to 8.7 m. The hydrogeologic situation of the plant site may be characterized in the metamorphed moldanubic rocks predominantly by waters of the rock crack type, which are already communicated in the weathered rocks, in the vicinity of the neogene small basins and in eluvia. In addition to the situation of the water table, which was established by making a number of bore holes, the conditions of the groundwater were checked in more detail in small shafts, 114 DLOUHŸ and ÍAFÁR which confirmed the coincidence of the orographic and hydrologie watersheds and determined the groundwater flow (Fig. 1). The level of the underground w ater on the whole plant site must be assum ed to lie at 3.5 to 7 m below the surface, in edge regions at depths of about 1.0 m. The velocity of under ground water measured corresponds to a filtration coefficient, k, of (2.2-3.5) X 10"^ cm/s. 3. RESEARCH METHODS AND RESULTS The soil samples in the nuclear power plant area were taken by means of the drilling rig RNH-6. Additional test pits were made on the ground delimited for the purposes of this work to establish the water table and to serve for sampling [2]. A section of the area of interest is shown in F ig. 2. The soil samples taken may be divided essentially into four groups: (I) Argillaceous sandy quaternary clays; (II) Argillaceous sands having varying content of weathered rocks; (III) Fine to medium dark sands (decomposed amphibolites and biotitic gn eisses); (IV) Many-coloured weathered rocks of the bedrock. For these samples the coefficients of permeability were determined in the first range, which served as an estimate of the flow rate of under ground water through the given medium, then specific mass and porosity to calculate relative rates of radionuclide migration through the soils. The physico-chemical properties investigated included the distribution coefficients, Kg, of the I37(^g and ^Sr. The determination of the relative movement of radiostrontium comprised the measurement of the dispersion 0____ too ___ 200 300 400 ' 500 600 [m) LOAMS WEATH'D GNEISS l * A * I COMPACT GNEISS I S S 3 WEATHERED I K ^ I PRIM. CHANGED GNEISS ROCKS IAEA-SM-180/74 115 coefficients D^(Sr) in the direction of the underground water flow. The methods used in individual determinations are described in the authors' earlier papers in this field [ 3, 4]. The values of the permeability coefficients varied in the range of 10"^ to 10"^ cm/s; they were obtained partly by calculation from grain-size distribution in the materials studied and partly experimentally. The results obtained indicate that the soils are all slightly permeable or permeable, which has also been confirmed by the behaviour of the characteristic granularity curves. Mean values of specific masses for individual material groups differed only little. They lay in the range of 1.8 to 2.0 g/crn^. Sample porosities varied in the range of 32% to 44%. The values of the distribution coefficients lay in the ranges (0.6 to 3.0)X1()4 cm^/g for caesium and (0.7 to 1.3)X102cm3/g for strontium. The distribution coefficients reveal the ability of rocks to sorb radionuclides from the groundwater and thus to contribute to the slowing-down of their movement in the direction of underground water flow. Relative strontium migration rates with respect to water movement during the flow of contaminating substances through the ground varied over the range of 0.008 - 0.09 for loamy clays, 0.006 * 0.05 for argillaceous sands, 0.01 - 0.09 for fine to medium sands and 0.015 - 0.08 for weathered bedrocks. The dispersion coefficients lay in the range of 1.2 to 2.2 cm^/d. The evaluation method is described in more detail in previously published work [5]. It follows from the determination of the sorption characteristics of these soils that they possess good retention ability, which may considerably decrease the radionuclide migration both in the unsaturated zone and in the water-bearing formation. The relative migration rate was never higher than one tenth of the water flow rate for strontium; caesium migrated through the soils at a rate which was less by one or two orders of magnitude. The values of dispersion coefficients indicate that the radionuclides will not disperse in the medium given at a rate higher than that of the ground water in the region. 4. DISCUSSION If considerable amounts of radioactive liquid penetrate the ground, the contaminating liquid seeps through the upper unsaturated zone until it reaches water-bearing layers, mixes with groundwater and moves in the direction of the greatest piezometric gradient to the places accessible to man. These places may be wells or streams communicating with the groundwater in the given region. The nuclides contained in radioactive liquid never attain man by these routes immediately after having entered the ground, and their quantities and concentrations in the places accessible to man also differ greatly from the quantities which initially penetrated the ground. Most radionuclides are sorbed efficiently on most of the various"rocks and, besides, it must be rem em bered that, in a number of ca se s, the concentrations of the radioactive substances decrease to a negligible value even before reaching man due to radioactive decay if the path travelled from the place of con tamination is sufficiently great and the flow rate of the groundwater correspondingly low. 116 DLOUHŸ and 3ÀFÀR From these points of view, both the operational and the natural con ditions in this locality may be considered as favourable. The storage of liquid and solid radioactive wastes is very well secured from the technical point of view, and only in the unlikely event of a complicated failure of the storage system can the ground become contaminated with radioactive sub stances. The hydrogeological conditions are also favourable; the results of tests in the field show that the groundwater flows at less than 0.1 cm/d, while the filtration coefficient of most of the materials lies in the range of 10'S to 3 X 10*4 cm/s, which corresponds roughly to the values of 1 to 30 cm/d. If groundwater flows continuously at a rate of 30 cm/d, which is the rate corresponding to the most unfavourable conditions considered, the contaminating radionuclide can travel a distance of 100 m in a year. This means that the radionuclide cannot reach a point away from the plant site where underground water appears on the surface or communicates with surface waters (about 500 m in the former case, and 700 m in the latter), in less than 5 and 7 years,respectively. A longer time interval is, however, much more probable. With respect to the direction of groundwater flow, a hydrogeologic watershed exists in the region of the plant site; the directions of under ground flow are shown on Fig. 1. The final plan of the buildings of the nuclear power plant will show whether the water-bearing zone would be affected on one or the other side of the watershed by any contamination that might reach the ground. The conditions will be still influenced, without any doubt, by excavation work and building foundations on the plant site. Essentially, one of two possible cases are to be expected, the contaminant will move with groundwater either in a northerly or a southerly direction. The extent of the region affected will depend first of all on the retention ability of the rocks form ing the unsaturated zone on one hand and the w ater-bearing layer on the other; and to a le sse r extent on the rate of the underground flow. From the results of this work and our previous experience, and from similar results obtained by other authors, it may be estimated that most radioactive substances will remain concentrated, even after a longer period has elapsed, in the immediate vicinity of the initial site of con tamination. To clarify the situation, the consequences of two model accidents resulting in ground contamination with the conditions prevailing at the plant site are given. 4.1. Leakage of 10 Ci of a one-year-old fission product mixture After 10 years radiostrontium and radiocaesium will represent 70% of the total amount of radioactive substance contained in the mixture; after 20 y ears practically only these two radionuclides will be present in the mixture. Their relative migration with respect to the flow rate of ground water is, however, very small. A quicker movement may be assumed for radiostrontium, the relative migration rate of which may attain even 0.1 of the water velocity. Corres ponding distances which ^°Sr reaches after one year, 10 years and 50 years are 10 m, 100 m and 500 m, respectively. In this way we arrive at the con clusion that it will take at least some 50 years after the accident before any 90Sr reaches places where this nuclide may come into direct contact with man. During the travel in the direction of underground flow radiostrontium is dispersed in the water-bearing formation and is thus diluted by the ground :AEA-SM-180/74 117 water. The region affected by the contaminant is usually of an oblong shape, it is elongated in the direction of the underground flow, and vertically it is bounded by the thickness of the water-bearing formation; its width is approximately one third to one half of its length. If the initial amount of radiostrontium at the moment of ground contamination was 0.38 Ci, the average concentration of this nuclide in the groundwater 10 years later will be roughly 5 X 10*^ Ci/1, and 50 years later approximately 10'^°Ci/ 1 (assuming that no 90sr will be sorbed on clays and sands present in ground); in fact, it is probable that the concentrations would be lower by one or two orders of magnitude. In conclusion, it may be said that in neither case, i.e. whether the northern or the southern basin is affected, will radiostrontium attain the surface waters sooner than in 50 years, and its concentration is most unlikely to exceed 10'H Ci/1, values which are well below the limit recommended by the ICRP for this radionuclide in drinking water. It must be noted here that there would always be enough time available to nullify successfully the consequences of an accident. First of all it would be necessary to confine the affected region by making several test bore holes and providing the so-called chemical or physico-chemical barriers, involving injections of convenient substances to retain strontium in place in the water-bearing zone Í 6, 7] or to remove the vital part of the con taminant and thus to repair the consequences of the accident. 4.2. Escape of 10 m3 of a concentrate from the evaporation residues containing a 10 Ci mixture of activated corrosion products 5 years old Such a mixture would contain, predominantly, ^Fe, 60^^ 54м^ which are the nuclides having the longest half-lives, though none of these exceeds the half-life of ^°Sr. The highest permissible concentrations of these radionuclides in drinking water vary over the range of 10*s to 10"^ Ci/1, which is well above the ICRP values for ^3r. At low relative migration rates of these radionuclides in groundwater, these being one thousandth to one hundredth of the groundwater velocity, it is clear that neither of the two surface water areas could be contaminated significantly before multiple dilution and radioactivity decay to negligible concentration had taken place in this case. Almost 95% of radionuclides are retained in the first decimetres of ground [8] which is first contacted by the radioactive liquid. When the radionuclide reaches the water-bearing formation it would disperse uniformly, as an analysis of samples did not show the presence of any minerals and rocks having extraordinarily high retention characteristics. 5. CONCLUSIONS In the event of a complex failure of the liquid storage facility, liquid radioactive wastes could contaminate firstly the lower areas in the immediate vicinity of the leakage source, dependent on the depth of the foundation of the store. The wastes will eventually reach the groundwater (after the saturation of soil pores) by gravity flow. The radioactive liquid is then transported in the direction of the groundwater flow. Agricultural plants and other vegetation may be contaminated where the water table comes to the surface or where it attains the level of the roots. The surface water and drainage water could also become contaminated. 118 DLOUHŸ and 3AFÁg It may be said that the environment on a broader scale is not threatened by immediate danger of contamination and its consequences since: (a) The nearest village is situated in the 1st protection zone of the nuclear power plant and will be abandoned, therefore, before the reactor commences operation; (b) Irrigation and sprays need probably not be considered; (c) Underground water will not be used for drinking in thearea of the1st protection zone; (d) The penetration time into the region mentioned via transport with the groundwater is considerable, being approximately 5 years (for к = 3.4 X 10"^ cm /s, for a distance of 500 m); (e) Finally the retardation effect resulting from radionuclide sorption by soils must be respected; However, prerequisites for the facility itself are that a reliable design of the storage facility be chosen, proper and sufficient safety measures be provided, and systematic operational and dosimetric control of the storage system be ensured. REFERENCES [1 ] POLÁSKOVA, M . , Geological Research Work Carried out on the Dukovany Nuclear Power Plant Site, Internal Rep., Geological Survey, Ostrava, CSSR (April 1971). [2 ] DLOUHY, Z . , SAFAR, О . , Movement of Radionuclides in the Ground and Environmental Safety from the Point of View of Possible Groundwater Contamination on the Dukovany Site; Rep. ÚJV 2832. Ch (June 1972). [3 ] BENE^, J . , DLOUHY, Z ., "Behaviour of some critical radionuclides in soils", Proc. lRPA2ndEur. Congr. on Radiation Protection, Akadémiai Kiadó, Budapest, Hungary (1973) 431. [4 ] DLOUHY, Z. , ^AFAR, O ., "Study of groundwater flow in site selection forradioactive waste storage", paper Int. Conf. Indication Methods for the Study of Groundwater Flow, Piesfany, ÔSSR, 30 Nov. -2 Dec. 1971. [5 ] DLOUHY, Z . , áAFÁÉí, О ., "Movement of radionuclides in geological formations", paper at Nat. Conf. Use of Radionuclides in Water Industry, Prague, ÓSSR, 15 - 18 Oct. 1973. [6 ] BAETSLE, L .H ., SOUFFRIAU, J,, "Installation of chemical barriers in aquifers and their significance in accidental contamination", Disposal of Radioactive Wastes into the Ground (Proc. Symp. Vienna, 1967). IA EA , Vienna (1967) 229. [7 ] DLOUHY, Z . , SÍAFÁFÍ, О ., Storage of Radioactive Wastes in Geological Formations, Internal Rep. ÚJV - RO - 2/73. [8 ] DLOUHY, Z. , "Movement of radionuclides in the aerated zone", Disposal of Radioactive Wastes into the Ground (Proc. Symp. Vienna, 1967), IAEA, Vienna (1967) 241. DISCUSSION J.R. BEATTIE: I believe you said that you have found in your studies that caesium moves perhaps twenty times more slowly than strontium in the soil. In England we have evidence from experiments that a fraction of caesium-137 becomes chemically combined with clay minerals present in the soil. Do you have any evidence from your work as to what fraction of caesium becomes chemically fixed in soil and will therefore not move? O. SAFÁR: No, I'm afraid we haven't studied this question. J. SCHWIBACH: In the event of groundwater contamination it will be of great advantage to know that the groundwater will take, say, five years to reach the next well. However, after that time, the water will arrive and IAEA-SM-180/74 119 could still give rise to problems if it contained substantial amounts of long- lived radionuclides. Thus the time lag gives only short-term relief; the long-term problem still remains. J. SEDLET: Mr. Safar, have you made any tritium measurements in groundw ater? O. SAFAR: No. J. SEDLET: How do you measure the water flow rate, both vertically and horizontally? O. SAFAR: The water flow rate was measured in the horizontal direction only, using the borehole dilution technique. P. PELLERIN (Chairman): To enable us to better appreciate the order of magnitude of the water migration rate quoted in your paper, could you tell us whether the water table is subject to industrial pumping and, if so, what is the approximate extraction rate? O. SAFAR: The water is not subject to any industrial pumping. IAEA-SM-180/23 NATURAL GAMMA BACKGROUND RADIATION IN THE ENVIRONS OF NUCLEAR FACILITIES Pre-operational analysis of dose to the neighbouring population J. PEÑSKO Radiation Protection Department, Institute for Nuclear Research, Swierk, Otwock, Poland Abstract NATURAL GAMMA BACKGROUND RADIATION IN THE ENVIRONS OF NUCLEAR FACILITIES: PRE-OPERATIONAL ANALYSIS OF DOSE TO THE NEIGHBOURING POPULATION. The paper presents the method of calculating and calculations of the mean values of the natural background dose-rate equivalent for gonads and bone marrow of an individual for the inhabitants of one district in Poland taken at random. These values, weighted to allow for some of the most important environmental factors, were 78.0 mrem/a for gonads and 72.1 mrem/a for bone marrow. The collective dose from this source of radiation for the district under investigation does not exceed 1.2 x 10** m a n -re m /a . The calculations were carried out based mainly upon measurements made using a scintillation radiometer installed in an aeroplane. Numerous measurements of the exposure dose rate at a height of 1 m above the ground level taken using a high-pressure ionization chamber, in addition to the results of gamma spectro- metric analyses of soil samples, were also taken into account. The body absorbed dose conversion factors were considered, as was time spent by the inhabitants indoors. The use of an appropriate weighting factor for making allowances for the radiological properties of building materials is proposed. The method can be used for estimating mean population dose due to external radiation originated from natural and artificial radioactive sources present in the environs of nuclear facilities. 1. PRELIMINARY CONSIDERATIONS The human body, and hence certain vital organs, receives doses of ionizing radiation resulting from our living within fields of both terrestrial and cosmic radiation during the whole of a life-time. The dose itself varies with geographical location and altitude above sea level of the locality and is subject to some minor variations in time. A part of this dose is due to beta radiation from naturally-occurring radioactive substances in the environment; however, the main contribution comes from natural gamma radiation sources. While the latter affects the whole body, the ß-radiation affects almost only the superficial tissue of the human body. The contribution of cosmic radiation and terrestrial gamma radiation to the dose of the population of nearly the whole world accounts for more than 40% of the total dose received from natural sources. The sources of the external gamma radiation background are naturally-occurring or artificial radioactive substances dispersed in the air, in soil, in rocks or on the surface of earth. The mean individual dose from this external gamma radiation background as accumulated in the critical organs during any period of an individual's 121 122 PEÑSKO life-time can be calculated with sufficient accuracy from exposure dose-rate values, measured directly in the air at a certain elevation above ground level (usually 1 m), or by using individual integral dosim eters, which would be carried by a representative group of the population for appropriately long periods of time. In the first case,the measured instantaneous values should be suitably representative for the group of the population under surveillance. Hence, the number of measured values should be great enough to give a meaningful average value with all important factors, such as demographic distribution of population, mean tim es spent in outdoor and indoor activities, anthropo metric conditions, etc. taken into account. The final mean dose for a given group of the population is generally calculated for a one-year period and only for the important (from a radiation hygiene point of view) organs of human body. The selection of these organs should respect predicted or observed biological effects of sm all radiation doses, which are important because of their impact on social diseases. Hence, if such diseases as bone tumours, leukaemia and congenital anomalies among the exposed population are caused by ionizing radiation, these observed effects result from the doses received by osteocites, bone marrow and gonads, respectively. Therefore these organs are considered to be the most important when evaluating collective doses from natural sources of terrestrial and cosmic radiation. In order to estimate the mean individual dose due to the natural back ground it is necessary, apart from determining the exposure doses in non- built*-up areas, to consider the measured exposure dose rates inside houses in the area under investigation. This can be carried out using a method described in an earlier paper [1] . It is also necessary to consider con version factors of the measured exposure dose rate in the air into absorbed dose rate in the critical organs. Accurate calculations of the foregoing factors for the population exposed to different natural gamma radiation fields and to artificial radioactive fall-out were recently presented by Bennett [2]. These calculations were then experimentally confirmed by phantom measurements. The starting point for calculations of y-ray transm ission in superficial tissue was set by Beck and Planque [3] by calculating gamma radiation spectra in the air at different elevations for the radionuclides зз&ц, 232^ 40^ formly dispersed in soil. The results of their work proved that,in the field, the angular distribution of the exposure dose rate of terrestrial gamma radiation background at sm all elevations (approx. 1 m) had its maximum in the direction which was close to horizontal (60° to 85°). With this in mind it was assumed by Bennett, in contradistinction to other authors [4-6], that taking a normal incidence of radiation onto a phantom gives the best approxi mation of the real situation existing in non-built-up areas. Corrections for the lim ited thicknesses and attitude of the human body were introduced during calculations. In order to estimate the dose absorbed by gonads with a uniformly irradiated human body in a horizontal plane, anatomic sketches of human body cross-sections were used. Thus the calculated values of the conversion factors for gonads were obtained, being 0. 86 for testicles and 0. 77 for ovaries. These values were much bigger than the value of 0. 6 which had been recommended by the UN Scientific Committee on the Effects of Atomic Radiation in 1962 and than those given by Spiers and Overton [6, 7]. IAEA-SM-180/23 123 Sim ilar work on the irradiation of bone marrow by the natural back ground radiation was carried out recently by Clifford and Facey [8]. The value of the conversion factor for the absorbed dose in that type of tissue was 0. 71 for normal incidence of radiation (of 0. 66 MeV energy) on the vertical axis of a standing phantom. Clifford and Facey proved that the value of the conversion factor depends much more on the angle of incidence of the radiation than on the radiation energy. Thus, the values of mean absorbed dose received by certain important organs of an individual, representative for a given group of the population, were multiplied by the number of its members (i. e. number of inhabitants of a country or of a given region, with possible subdivision according to their sex). The products result in numerical values of the mean collective dose, a knowledge of which is of great importance in evaluating different aspects of radiation protection. The foregoing measurements and calculations can be used in estimating environmental hazards to workers or when designing large nuclear facilities (i. e. power reactors or fuel fabrication plants),during normal operation and during postulated accidents. Comparison of the collective doses received by groups of population as a result of the operation of nuclear facilities with those due to the natural background is one of the factors which enable dose lim its to be set for protecting the population and concentration lim its determined for artificial radionuclides in the natural environment. 2. ANALYSES OF MEASUREMENT DATA To carry out such an evaluation for the area around the site of the first nuclear power plant in Poland, calculations of mean individual dose for gonads and bone marrow were made for inhabitants of the Zarnowieckie Lake area in the district of Gdansk using field data for the external gamma dose resulting from the natural background radiation. The measurement results obtained in earlier work [9] and presented on Fig. 1 were used. In Fig. 1, the central point, E, denotes the nuclear power plant site, which is surrounded by 80 radially located sectors at different distances from E. The sectors thus selected are bounded by 8 main directions and by 10 distances from the central point up to a radius of 30 km (1, 2, 3, 5, 7, 10, 15, 20, 25 and 30 km). The division of the region under investigation into sectors was arbitrary and was made only to take into account such factors as distribution of population and geographic variations of the natural back ground of gamma radiation when calculating the collective dose. The division adopted in this paper has an additional advantage: it enables one to use the calculational data to evaluate possible variations of gamma radia tion background in the vicinity of the nuclear power plant. These variations might be caused by a reactor accident or by normal operation, m atters that, to a large extent, depend upon the wind directions. The numbers of the population in each of the 80 sectors and the values of the yearly dose from the terrestrial gamma radiation background absorbed in the air were defined afterwards. Demographic relations were based upon official evaluations made by the Regional Planning Office in GdaAsk for 1970, based on statistical data from 31 December 1964. 124 PEÑSKO Л4ЯЛЮИ%ЮйГ ¿Æ f f?7P ------е^/я#а<з FIG. 1. Division into sectors of the ¿amowieckie Lake region for mean population dose estimation. The map shows also the airborne radiometric values. IAEA-SM-180/23 125 TABLE I. CONVERSION FACTORS FOR THE DOSE ABSORBED IN AIR AND IN TISSUES Conversion Medium absorbing factor Ref. radiation energy (ra d /R ) Air 0.869 [16] Soft tissue 0 .9 6 [6 ] Gonads 0 .8 2 Г2] Bone marrow 0.71 [8] TABLE II. MEAN VALUES OF COLLECTIVE DOSE AND DOSE RATE ABSORBED IN CRITICAL ORGANS AS WEIGHTED OVER POPULATION DENSITY AND EXPOSURE DOSE RATE Mean collective Mean individual absorbed C ritical Number in dose rate dose rate organ population (man* rad/ a) (m rad /a ) Gonads 6349 142 282 44.6 Bone marrow 5417 142 282 38.1 To set values of annual exposure doses, the results of the following measurements were used: (a) Scintillation radiometer measurements made with equipment installed in an aeroplane; (b) Dose meter using a high-pressure ionization chamber placed at near ground level; (c) Spectrometric analyses of soil samples. The most probable values were introduced into the calculations. These data were used to calculate the collective dose absorbed in air, appropriate to the demographic distribution of population in the area being considered. The result was about 6760 m an-rad/a. The factors shown in Table I for conversion of exposure dose into absorbed dose in air and in different tissues were used. Table II gives calculated values of the total annual collective dose due to the terrestrial gamma background for gonads and bone marrow among the inhabitants of the Zarnowieckie Lake area, as well as the calculated mean values of individual dose for a statistically representative single inhabitant of that region. The foregoing values correspond to an average exposure dose rate of P = 6.2 pR/h in air at an 126 TABLE III. CALCULATIONAL RESULTS OF COLLECTIVE DOSE RATES (man-rad/a) AND MEAN ABSORBED DOSE RATES IN CRITICAL ORGANS (mrad/a) DUE TO THE NATURAL GAMMA BACKGROUND WITH TIME SPENT INDOORS AND OUTDOORS CONSIDERED Building Collective dose rate (man* rad/a) Mean individual absorbed m aterial C ritical dose rate shielding In wooden houses In brick houses organ PEÑSKO Outdoors (m rad /a ) coefficient (50% of people, (50% of people, T otal (25% of time) (weighting factor) 75% of time) 75% of time) Gonads 1587 1786 2381 5754 4 0 .4 0.9 0 6 Bone marrow 1354 1524 2031 4909 3 4 .5 0 .9 0 6 IAEA-SM-180/23 127 elevation of 1 m over a plane area. The mean values of absorbed dose rate, also given in Table II, do not consider the impact of time spent indoors nor doses due to cosmic radiation on the radiation burden of the critical organs. 3. IMPACT OF INDOOR ACTIVITIES The impact of time spent indoors on the irradiation of the population group can be evaluated best by random measurements of individual exposure dose rate in dwelling houses in the area being investigated. These m easure ments could in principle be made using dose m eters equipped with high- pressure ionization chambers,or thermoluminescent dosimeters. Because of the lack of such measurements for the Zarnowieckie Lake area, the evaluation was based on experimental data from other regions of Poland and from other countries. From the various publications [10-13] and the author's own observa tions [1], for most of the sites investigated the ratio of exposure dose m easured in dwelling houses to that measured outside is 0. 75 : 1 for wooden houses and 1.00 : 1 for brick or concrete houses. In some cases, when particular building m aterials are used (i.e. granite, blast-furnace cement, boiler slag) in areas of average gamma background level, the ratio can exceed 1 : 1. Assuming that one half of the population in the area investigated lives in wooden houses and the other half in brick or sim ilar houses, and assuming the probable time spent in outdoor activities is 6 hours a day [10], the mean values of collective dose and absorbed dose rates in critical organs of interest can be calculated. The results of calculations are shown in Table III. The table also gives the values of the building m aterial shielding coefficient which, as a weighting factor, may be helpful in the assessm ent of indoor activities in the collective dose estimation resulting from the natural gamma background. This weighting factor was calculated as the ratio of the mean individual absorbed dose rate during a temporary indoor stay (see Table III) to the absorbed dose rate during a continuous outdoor stay (Table II). 4. CONTRIBUTION OF COSMIC RADIATION Because of the lack Of experimental data for evaluating the cosmic radiation impact on the dose-rate equivalent in the area under investigation, data given in the literature were used. For calculational purposes, a constant value of cosmic radiation intensity for the ionizing component at sea level for the latitude range fitting Poland was adopted. According to recent work [14], the value for geomagnetic latitudes higher than 55° is 2. 20 ion pairs/cm 3 - s. This value corresponds well with that obtained by Shamos and Liboff [15], based on literature data published before 1966. Assuming the following conversion factors: 1 ion pair/cm3 - s = 1. 73 pR/h and 0. 96 rad/R for soft tissue, an absorbed dose rate from that component of cosmic radiation can be calculated. The calculated value equals 3. 65 prad/h or 32 m r a d /a . 128 PENSKO TABLE IV. MEAN INDIVIDUAL DOSE RATES FOR GONADS AND BONE MARROW DUE TO NATURAL GAMMA RADIATION AND COSMIC RADIATION FOR THE INHABITANTS OF THE ZARNOWIECKIE LAKE AREA Mean individual dose rate outside buildings Building equivalent Type of (m rad /a ) m aterial (m rem /a ) QF radiation w eighting Bone factor Bone Gonads Gonads marrow marrow Terrestrial gam m a 44.6 38.1 0.906 1 4 0 .4 3 4 .5 background C osm ic radiation (a) ionizing com ponent 32.0 32.0 footnote a 1 3 2 .0 3 2 .0 (b) neutron component 0.7 0.7 footnote a 8 5 .6 5 .6 T otal 7 8 .0 7 2 .1 ^ Building material weighting factors for cosmic radiation depend on the construction of a building and on the storey on which a given flat is located. The values of this coefficient were not estimated in this work and therefore the dose-rate equivalents obtained for cosmic radiation can be slightly increased. The doses due to the neutron component of cosmic radiation can, however, only be evaluated at a much lower accuracy. This is due to the much greater variations of intensity of that radiation with time, and with the altitude and latitude of a site under investigation. During the last decade, evaluation of such data was a topic of interest to only a few authors [17-23], who published significantly different values, ranging from 0.33 m rem /a [22] to 6. 8 m rem /a [19]. Because the differences are so manifest, it appears justifiable for this calculation to adopt the value recommended by ICRP [18] and adopted by the UNSCEAR [17] for mean latitudes at sea level as 0.7 m rad/a. Hence, using the quality factor QF = 8 as recommended by ICRP [18], the dose-rate equivalent of 5.6 m rem /a for cosmic neutrons at sea level is obtained. IAEA-SM-180/23 129 5. CONCLUSIONS To summarize, a complete estimation of mean exposure for gonads and bone marrow from both external gamma radiation and cosmic radiation for the inhabitants of Zarnowieckie Lake area was presented. The calcula- tional results are shown in Table IV. The collective dose rate from this source of radiation and for the district under investigation does not exceed 1.2 X 1()4 m an-rem /a. The foregoing calculations could be used for any other region or for the whole country. These mean values received by the population of a given region represent important information for risk evaluation to the given group of population due to artificial gamma radiation sources dispersed in the environment during operation of nuclear facilities. A sim ilar evalua tion of collective dose from natural ionizing radiation in the environment should be made, particularly for areas around proposed sites for large nuclear facilities. REFEREN CES [1] PEÑSKO, J., MAMONT, K ., WARDASZKO, T ., Measurements of ionizing radiation inside blocks of flats in Poland, Nukleonika 14 4 (1969) 93-103; Eng. translation by USAEC,AEC-tr-7027/4 UC-34 TT 69-50012/4. [2] BENNETT, B. G ., Estimation of gonadal absorbed dose due to environmental gamma radiation, Health Phys. 19 (1970) 757. [3] BECK, H ., de PLANQUE, G ., The Radiation Field in Air Due to Distributed Gamma-Ray Sources in the Ground, Rep. HASL-195 (1968). [4] SPIERS, F.W ., Radioactivity in man and his environment, Br. J. Radiol. 29 344(1956) 409. [5] O'BRIEN, K ., LOWDER, W .M ., SOLON, L. R ., Beta and gamma dose rates from terestrially distributed sources, Rad. Res. 9 (1958) 216. [6] SPIERS, F.W ., OVERTON, T .R., Attenuation factors for certain tissues when the body is exposed to nearly omni-directional gamma-radiation, Phys. Med. Biol. 7 1 (1962) 35. [7] UNSCEAR, Report of the UN Scientific Committee on the Effects of Atomic Radiation, UN, New York (1 9 6 2 )2 1 1 . [8] CLIFFORD, C .E., FACEY, R. A ., Changes in acute radiation hazards associated with changes in exposure geometry, Health Phys. 18 (1970) 217. [9] GWIAZDOWSKI, B., PEÑSKO, J. , JAGIELAK, J ., BIERNACKA, M ., CIESLA, К .. Environmental paper IAEA-SM-180/22. [10] SPIERS, F.W ., "Gamma-ray dose-rates to human tissues from external sources in Great Britain", Appendix D in The Hazards to Man of Nuclear and Allied Radiations, Second Report of the Medical Research Council, HMSO Cmnd 1225 (1960) 66-70. [11] LOWDER, W .M ., CONDON, W. J ., Measurement of the exposure of human population to environ mental radiation, Nature (London) 206 (1965) 658. [12] OHLSEN, H ., Bestimmung der mittleren BevSlkerungsbelastung durch natürliche äussere Strahlung auf dem Gebiet der DDR, Rep. No.SZS-14/69, Berlin(1969). [13] YEATES, D .B ., GOLDIN, A. S., MOELLER, D.W ., Radiation irom natural sources in the urban environment, Nuclear Safety (1971). [14] O'BRIEN, K ., Calculated cosmic-ray ionization in the lower atmosphere, J. Geophys. Res. [15] SHAMOS, M .H ., LIBOFF, A .R ., A new measurement of the intensity of cosmic ray ionization at sea level, J. Geophys. Res. J1 (1966) 4651. [16] INTERNATIONAL COMMISSION ON RADIOLOGICAL UNITS AND MEASUREMENTS (ICRU), Report 10b, NBS Handbook 85, US Dept, of Commerce, Washington, DC (1964). [17] Report of the United Nations Scientific Committee on the Effects of Atomic Radiation, Twenty-first Session, Supplement No. 14 (A/6314), UN, New York (1966). [18] UPTON, A .C ., Radiobiological aspects of the supersonic transport: A report of the ICRP Task Group on the biological effects of high-energy radiations, Health Phys. 12 (1966) 209. 130 PEÑSKO [19] WATT. D .E ., Dose equivalent rate from cosmic ray neutrons, Health Phys. J13 (1967) 501. [20] SCHAEFER, H .J., Public health aspects of galactic radiation exposure at supersonic transport altitudes, Aero. Med. 39 (1968) 1298. [21] SCHAEFER, H. J ., Radiation measurement at supersonic transport altitude with balloon-borne nuclear emulsions, NASA Joint Report NAMI-1068 (May 1969). [22] O'BRIEN, K. , McLAUGHLIN, J.E. , Calculation of Dose and Dose-equivalent Rates to Man in the Atmosphere from Galactic Cosmic Rays, USAEC Doc. HASL-228 (1970). [23] HAJNAL, F ., McLAUGHLIN, J.E ., WEINSTEIN, M .S., O'BRIEN, K ., 1970 Sea Level Cosmic RayNeutronMeasurements, USAEC Doc. HASL-241(1971). DISCUSSION W. M. BURKHARDT: What is the accuracy of the results shown in T a b le III? J. PE^SKO: The accuracy of our airborne measurements in static conditions is of the order of + 15%. However, for the calculation of the collective dose we also used the results of measurements with high-pressure ionization chambers at ground level and the results of spectrometric ana lysis of soil samples, which are more accurate. The over-all error in the estimation of the exposure dose rates in the area under investigation is approximately + 6%. L.I. GEDEONOV: What is the minimum surface density of gamma- emitting radionuclides that can be detected with your methods? J. PENSKO: We can quite easily detect 0 .2 jnCi/m.2 of isii deposited on the soil surface and also 2.2 pCi/m^ of I37cg^ representing about 0.1 of the Emergency Reference Levels recommended by the Medical Research Council in the United Kingdom. We confirmed this experimentally using аНЗпцд gamma radiation source with quite a large surface. We have also used the more sensitive gamma spectrometric technique to estimate the exposure dose rate for different components of natural (U, Th series and 4°K) and artificial (radioactive fall-out) gamma background radiation. With this method we were able to detect and estimate the exposure dose rates due to radionuclides such as 95Zr-95Nb^ I37çg and ^Mn, which were of the order of 0.1 ^R/h. P. PELLERIN (Chairman): With regard to the reference made in your paper to bone tumours and leukaemia I should like to point out that, except for a very few cases of exceptionally high professional exposure, no certain cause-and-effect relationship has been established between dose received and these diseases. LAEA-SM-180/81 О ДЕЯТЕЛЬНОСТИ ОРГАНОВ СОВЕТА ЭКОНОМИЧЕСКОЙ ВЗАИМОПОМОЩИ В ОБЛАСТИ ОХРАНЫ И УЛУЧШЕНИЯ ОКРУЖАЮЩЕЙ СРЕДЫ ПО ПРОБЛЕМЕ ОБЕСПЕЧЕНИЯ РАДИАЦИОННОЙ БЕЗОПАСНОСТИ В.ХАКЕ Секретариат Совета Экономической Взаимопомощи, М о с к в а * Abstract- Аннотация ACTIVITIES OF THE СМЕА AGENCIES IN THE FIELD OF ENVIRONMENTAL PROTECTION AND IMPROVEMENT IN CONNECTION WITH RADIATION SAFETY PROBLEMS. In che recent years cooperation among che CMEA member countries in the field of environmental protection and improvement in the context of ensuring radiation safety has been steadily expanding. This paper comprises decisions of corresponding CMEA organs in carrying out their extensive work in the above field, information on the setting up of the Scientific and Technological Coordination Council on Radiation Safety, and a summary of the CMEA scientific and technological cooperation programme aimed at securing radiation О ДЕЯТЕЛЬНОСТИ ОРГАНОВ СОВЕТА ЭКОНОМИЧЕСКОЙ ВЗАИМОПОМОЩИ В ОБЛАС ТИ ОХРАНЫ И УЛУЧШЕНИЯ ОКРУЖАЮЩЕЙ СРЕДЫ ПО ПРОБЛЕМЕ ОБЕСПЕЧЕНИЯ РА ДИАЦИОННОЙ БЕЗОПАСНОСТИ. Страны-члены СЭВ за последние годы все шире развивают сотруд ничество в области охраны и улучшения окружающей среды в соответст вии с "Комплексной программой дальнейшего углубления и совершенст вования сотрудничества и развития социалистической экономической интеграции стран-членов СЭВ", принятой ХХУ сессией Совета Эконо мической Взаимопомощи (июнь 1971 г.). Согласно Комплексной про грамме разработка мероприятий по охране природы представлена как одна из основных научно-технических проблем, подлежащих совмест ной разработке с применением наиболее эффективных форм сотрудни ч е с т в а . Работы,проведенные органами СЭВ с момента принятия Комплекс ной программы по развитию этой проблемы, нашли свое отражение в докладе Комитета СЭВ по научно-техническому сотрудничеству XXVII * СЭВ, пр-т Калинина 56, Москва, СССР 131 132 ХАКЕ сессии СЭВ (июнь 1973 г.) "О мероприятиях по расширению сотрудни чества стран-членов СЭВ и СФРЮ в области охраны и улучшения окру жающей среды и связанного с этим рационального использования при родных ресурсов". Наряду с другими важными для охраны и улучше ния окружающей среды проблемами в докладе указывается на необхо димость уделять серьезное внимание проблеме обеспечения радиацион ной безопасности,включающей в комплексе все основные вопросы обес печения чистоты окружающей среды в связи с внедрением в народное хозяйство стран-членов СЭВ атомной энергии в промышленных мас ш та б а х . Эта проблема связана, с одной стороны, с развитием атомной элек троэнергетики, и, с другой стороны, с расширением применения радио активных изотопов и ионизирующих излучений в различных областях науки и техники, в промышленности, сельском хозяйстве и медицине. Наиболее важное значение, однако, в настоящее время имеют работы, проводимые странами по защите окружающей среды в связи с разви тием атомной электроэнергетики. Тем более,что в соответствии с Комплексной программой страны-члены СЭВ большое внимание уделя ют мероприятиям, направленным на создание совместными усилиями научно-технических, производственных и организационных предпосы лок для ускорения развития и эффективного внедрения в народное хо зяйство атомной энергии в промышленных масштабах. Определенное место при этом отводится работам по обеспечению радиационной без опасности атомных электростанций на стадиях проектирования, строи тельства и эксплуатации. До 1971 г. деятельность органов СЭВ в области радиационной безопасности была направлена на разработку таких частных проблем, к ак : - Рекомендуемые исходные условия и критерии при выборе пло щадок для строительства атомных электростанций; - Общие принципы обеспечения безопасности атомных электро станций при проектировании, строительстве и эксплуатации; - Проведение научно-технических конференций по проблеме обра ботки и захоронения радиоактивных отходов; - Общие указания к осуществлению контроля за соблюдением норм радиационной безопасности; - Рекомендации по дезактивации технологического оборудования, применяемого в атомной энергетике; - Методики отбора проб для оценки загрязнения воды радионукли дам и ; и ряда других нормативных документов и рекомендаций по стандарти зации, принятых в последующем странами-членами СЭВ за основу при проведении соответствующих работ . Применение этих документов, а также имеющиеся, главным обра зом, в Советском Союзе инструкции и руководства способствовали строительству ряда АЭС, эксплуатация которых доказывает их относи тельную безопасность для окружающей среды . Проведенные в районе месторасположения Нововоронежской АЭС (СССР) исследования выявили, что концентрация радиоактивных аэро золей в атмосферном воздухе в радиусе 40 км остается практически постоянной и находится на уровне фоновых значений и после пуска АЭС и во время ее работы. На уровне фоновых значений находятся также IAEA -SM-180/81 1 содержание радиоактивных веществ в сельскохозяйственных растениях (от 9,9*10'^до 1,7-10 Ки/кг сырого веса продукта) и содержание ра диоактивных веществ в воде реки Дона (1,7-10*^- 1,52'10'^Ки/л ) по суммарной (3-активности [1]. Аналогичные результаты были получены специалистами ГДР в районе АЭС Райнсберг, согласно которым не было обнаружено никако го влияния атомной электростанции на уровень активности окружающей м е с т н о с т и . Более широкое развитие атомной электроэнергетики в странах- членах СЭВ, однако, ставит новые задачи перед органами СЭВ, имею щие цель обеспечить защиту населения и окружающей среды от возмож ного неблагоприятного влияния АЭС на стадии промышленного освое ния А Э С . Постоянная Комиссия СЭВ по использованию атомной энергии в мирных целях непосредственно после принятия Комплексной программы приступила к решению вопроса более планомерного развития работ в области радиационной безопасности, как частной проблемы по разработ ке мероприятий по охране природы. Комиссией была разработана и в ноябре 1971 г. принята "Программа сотрудничества стран-членов СЭВ в области радиационной безопасности" . Эта программа охватывает все основные вопросы, требующие решения в рамках научно-технического сотрудничества стран-членов СЭВ и разрабатываемые также в рамках Постоянной Комиссии СЭВ по электроэнергии. Таким образом были созданы условия для более тесного сотрудничества всех ведомств стран, занимающихся вопросами обеспечения радиационной безопаснос ти . С целью обеспечения более эффективной организации научно-тех нического сотрудничества в этой области Постоянная Комиссия СЭВ по использованию атомной энергии в мирных целях создала Координа ционный научно-технический совет по радиационной безопасности (КНТС-РБ), первое заседание которого состоялось в сентябре 1972 г. в г. Потсдаме (ГДР). Председателем этого КНТС назначен проф. В.Буркхардт (Государственное управление по атомной безопасности и защите от излучений ГДР). В работе КНТС-РБ участвуют предста вители всех государственных органов стран-членов СЭВ, занимающих ся вопросами обеспечения радиационной безопасности. В свете соответствующих решений Исполнительного Комитета СЭВ Постоянная Комиссия СЭВ по использованию атомной энергии в мирных целях приступила к организации координации сотрудничества стран-членов СЭВ по проблеме обеспечения радиационной безопасности в целом в рамках СЭВ . В связи с этим на втором заседании КНТС-РБ (сентябрь 1973 г., г.Дрезден, ГДР) были рассмотрены проект программы научно-техни ческого сотрудничества стран-членов СЭВ в области охраны и улучше ния окружающей среды по проблеме обеспечения радиационной безопас ности до 1980 года и основные направления научно-технического сот рудничества в этой области до 1990 года. Органы СЭВ ставят себе задачу разработать такой комплекс нор мативной, методической и технической документации, который позво лит развивать атомную электроэнергетику в широком масштабе с необ ходимой степенью безопасности для окружающей среды . 134 ХАКЕ Программа научно-технического сотрудничества стран-членов СЭВ предусматривает проведение работ по следующим девяти главным направлен иям : - общие вопросы обеспечения радиационной безопасности; * радиационный контроль; - обеспечение радиационной безопасности в атомной электроэнергетике ; " обеспечение радиационной безопасности при использовании атомной энергии в науке и технике; обеспечение радиационной безопасности в связи с применением источников ионизирующих излучений в медицине; - обеспечение радиационной безопасности в горнодобывающей промышленности; - обеспечение радиационной безопасности при обезвреживании жидких, твердых и газообразных отходов; - обеспечение радиационной безопасности при переработке облученного ядерного топлива; - обучение кадров в области радиационной безопасности. Основное внимание при этом уделяется вопросам нормативно методической и технической документации, а также вопросам стандар тизации и унификации, связанным с атомной электроэнергетикой. Намечается разработать прогноз развития работ в области радиа ционной безопасности в связи с прогрессом использования атомной энергии, общие положения обеспечения радиационной безопасности АЭС, санитарные правила, требования к защите окружающей среды, требования к конструкционным и строительным элементам АЭС, мето дики оценки надежности оборудования, перечни нарушений устройств в нормальной эксплуатации, классификацию аварий, методики оценки вероятности радиационных аварий, указания по защите персонала и населения, анализ радиационной обстановки на АЭС и облучаемости персонала, методы исредстваснижения облучаемости персонала АЭС и др. Все эти работы заканчиваются принятием органами СЭВ соот ветствующих рекомендаций странам-членам СЭВ. Хотя основное место в программе занимает разработка инженерно- технических мероприятий по предотвращению загрязнения окружающей среды, изучаются также вопросы наблюдения за состоянием окружающей среды, которые являются не менее важными. Предполагается разра ботка анализа систем индивидуальной дозиметрии с выявленим наиболее оптимальной системы контроля,разработка и совершенствование унифи цированных методик по определению выбросов радиоактивных продуктов в атмосферу, методик по отбору проб, по радиометрии газов и аэрозо лей, разработка оптимизированных систем внешнего контроля. В нас тоящее время подготавливается международное сравнение дозиметров и счетчиков общего тела . С целью обеспечения более эффективного подхода к организации сотрудничества и решению проблем органы СЭВ проведут в 1975 г. научно-техническую конференцию стран-членов СЭВ по обеспечению радиационной безопасности в связи с эксплуатацией АЭС, на которой все эти вопросы будут рассмотрены в комплексе с тем, чтобы в после дующем были выработаны соответствующие рекомендации странам-чле нам СЭВ по дальнейшим работам в этой области, в увязке с мероприя- IAEA-SM-180/81 135 тиями, проводимыми странами на общеевропейской основе,особенно, с мероприятиями МАГАТЭ. Кроме КНТС-РБ Постоянной Комиссией СЭВ по использованию атомной энергии в мирных целях создан ряд других координационных научно-технических советов, которые разрабатывают и отдельные воп- росы обеспечения чистоты окружающей среды . К числу таких советов относятся : - КНТС по водным режимам АЭС, занимающийся вопросом сниже ния радиационной обстановки на АЭС путем улучшения водного режима первого контура АЭС; - КНТС 1-3, занимающийся вопросами обезвреживания жидких, твердых и газообразных радиоактивных отходов и дезактивации заг рязненных поверхностей в целях решения одной из основных проблем обеспечения чистоты окружающей среды в связи с развитием атомной энергии в промышленных масштабах; - КНТС 1 -6 ,занимающийся вопросами радиохимической перера ботки ядерного топлива,а также вопросами обеспечения радиационной безопасности, т.е. исключения возможности попадания в окружающую среду радиоактивных изотопов с концентрацией, превышающей допус тимую . За последнее время КНТС по радиоактивным отходам и дезакти вации подготовил ряд методических материалов, таких, как: "Методика геологических, гидрогеологических и физико-химичес ких исследований при поисках,разведке и обосновании структур,при годных для безопасного захоронения радиоактивных отходов". "Методика выбора условий захоронения отвержденных отходов в зависимости от свойств и удельной активности". "Критерии выбора методов отверждения радиоактивных отходов атомных электростанций в зависимости от свойств отходов и природ ных условий захоронения". Завершены работы по ряду важных вопросов: исследование вклю чения в битум нитрата натрия, конструирование и испытание опытных образцов устройств и оборудования для дезактивации аппаратуры и загрязненных помещений, определение рациональной облицовки поме щений АЭС путем применения новых конструкционных материалов и Др. [2]. ЛИТЕРАТУРА [1] ПЕТРОСЬЯНЦ, А .М ., От Научного Поиска к Атомной Промышленности, Атомиздат, М., 1972. [21 Information about the activities of the Council for Mutual Economic Assistance in peaceful uses of atomic energy in 1972-1973, 17th IAEA General Conference, Vienna, September 1973. DISCUSSION G. BRESSON: What do you understand by 'radiological safety'? W. HAKE: The concept of radiological safety covers a whole range of m easures designed to protect the biosphere from the effects of artificial sources of ionizing radiation. In the first place radiation safety is based on the engineering design and operational practice of a nuclear power station 136 ХАКЕ which must be checked and approbed by radiation safety specialists. Any rediation from such sources must be in strict compliance with the standards laid down by the ICRP, IAEA or national bodies. In the second place radiation safety is based on monitoring of the environment and the population (including, in particular, the staff of nuclear plants). P. PELLERIN (Chairman): Are you implying that both aspects of radia tion safety should be treated as a whole? W. HAKE: Yes. Our programme covers both aspects. If we can help to optimize the engineering from the radiation safety point of view, then our subsequent task will be that much easier. J.M. MATUSZEK: I would go along with this philosophy. Particularly in the case of fuel reprocessing plants, we have found that engineering safety system s do not always function as well as calculated. Therefore I feel that a radiation protection service should also have engineering staff and not be limited to doctors and biologists. J.R. BEATTIE: I was very interested to hear you say that the 'proba bility of accidents' was among the subject to be considered by your organiza tion. Does this mean that you will be carrying out reliability analyses of reactor and plant system s, that is to say, the application of probability theory to the analysis of possible failure of sm all components leading on to deriva tions of the probability of failure of larger components and system s? W. HAKE: Yes, we are planning just such a programme. N.G. GUSEV: In the Soviet Union we use probability methods to fore cast the maximum em ission levels of inert radioactive gases by relating changes in the discharge rate of these gases, recorded continuously by oscillograph in nuclear power stations, to variations in the operating para meters of those stations. The forecasting of serious accidents by means of probability methods involves certain difficulties, as we all know. One possible approach to this problem was described at the Soviet-Swedish Symposium on nuclear power station safety at Studsvik, Sweden in March 19 73 in the paper by B.G. Pologikh et al. , entitled "Nuclear power station safety from the point of view of the consequences of serious accidents" and the paper by A. M. Bukrinsky et al. , entitled "Safety procedure for accidents involving primary circuit leakage". I.A. TERNOVSKY: At the Institute of Atomic Energy in Moscow we have been studying all available information on previous nuclear accidents (in particular, BILES, M.B., in Handling of Radiation Accidents (Proc. Symp. Vienna, 1969) IAEA, Vienna (1969)2 andCATLIN, R.J., op. cit. page 437), because we believe that such experience will enable us to improve the radiation safety of new plant. R. KIRCHMANN: Public opposition to nuclear energy hasbeenmentioned several times today and would appear to be a world-wide phenomenon. Is there currently any open opposition in CMEA countries and, if so, what form does it take? W. HAKE: It is very important to inform the public properly about radiation, radiation sources and radiation protection, the more especially as their ideas are still coloured by the bitter memory of two atomic bombs being dropped on a civilian population. However, the nuclear power programme of the CMEA countries indicates the positive attitude of the public to nuclear energy and no open opposition to the peaceful uses of nuclear energy has been observed in CMEA countries. ENVIRONMENTAL MONITORING PROCEDURES - NORMAL AND EM ERGENCY SITUATIONS Chairmen J. SCHWIBACH (Federal Republic of Germany) Rebeca M. de NULMAN (Mexico) IAEA-SM-180/42 ENVIRONMENTAL RADIOACTIVITY SURVEILLANCE METHODS FOR A NUCLEAR FACILITY* J. SEDLET Occupational Health and Safety Division, Argonne National Laboratory, Argonne, 111., United States of America Abstract ENVIRONMENTAL RADIOACTIVITY SURVEILLANCE METHODS FOR A NUCLEAR FACILITY. the programme it was necessary to distinguish between environmental radioactivity from nuclear test fall-out and from Argonne. It is now necessary to distinguish between radioactivity from several sources and nuclear installations, and at locations distant from Argonne, for dose evaluation purposes. This is a more difficult environment and will indicate future information needs and research problems. This work was performed under the auspices of the USAEC. INTRODUCTION Argonne National Laboratory, a research and development laboratory of the U.S. Atomic Energy Commission, has conducted a radioactivity monitoring program since 1948. The original purpose oí the program was to measure the effect of Laboratory operations on the radioactive content of the area. This remains a principal purpose, although the increased interest in envi ronmental health has changed the emphasis from only measuring radioactivity to attempts to evaluate very low radiation doses. Experience indicates the goal should be to measure existing radioactivity levels, regardless of current standards, since it is not possible to predict the direction standards w ill take in the future, nor which environmental materials will be important. 139 140 S ED LET The effect of Argonne operations is evaluated by measuring the radio activity in materials collected on and off the Laboratory site, and by measurements of penetrating radiation dose. The varied nature of the work at the Laboratory has required monitoring for a wide spectrum of radio nuclides at very low levels. Among the nuclear facilities at the Laboratory are uranium fueled reactors; several critical assemblies fueled with plutonium or uranium; a number of accelerators; a plutonium facility; and several multi-Curie hot cells and laboratories. 2 The Laboratory grounds themselves are located on a 15 km site, 43 km southwest of Chicago. The facilities are located in the central 6.9 km^ of of the site, which is surrounded by a security fence. The principal stream on the site is Sawmill Creek, into which Argonne waste water is discharged, and which enters the Des Plaines River about 2 km south of the site. The Des Plaines River joins the Kankakee River to form the Illinois River about 48 km southwest of the site. The first use of this waterway for drinking is on the Illin ois River about 225 km downstream from Argonne. The most common wind directions are from the west and southwest, but the winds are sufficiently variable so that monitoring for airborne releases must be carried out in all directions from the site. Figures 1 and 2 are maps of the site and the surrounding area and indicate the principal water and air sampling stations. SAMPLE COLLECTION AND MEASUREMENT A ir Airborne particulate material in outside air is sampled at 10 locations on the site and at 7 locations off the site. Samples are routinely measured for total alpha and beta activities, by gamma-ray spectrometry, and by specific radiochemical analyses. This is the only case where total counting is still considered necessary, since it provides a rapid measure of radon, thoron, and abnormal radioactivity. 41 Air near reactors is sampled and counted for Ar in an evacuated 7 liter steel container that fits over a Nal(Tl) crystal. While the sensi tivity is only 10 nCi/щЗ, the unit has the advantages of low cost, porta bility, and simple operation. Water vapor is collected on silica gel for Зн analysis. Water The principal water sampling is conducted in Sawmill Creek, which receives Argonne waste water at location 7M (Fig. 1), and the Des Plaines River. Other streams and lakes, including Lake Michigan and the Illinois River, are also sampled. In addition to specific radionuclide analyses, water samples continue to be measured for total activity by counting the residue obtained on evaporation. This method has the advantages that results can be obtained economically and compared with many years of sim ilar measurements. However, volatile materials are lost, and self-absorption is large, so the results can be misleading, particularly for beta emitters. The self-absorption corrections are those measured for 239рц and 2*%T1, and the results repre sent only the disintegration rates of these nuclides that would produce the observed counting rates. To obtain smooth mounts, it was found necessary to slurry the solids on a planchet with a liquid. With water, the alpha, IAEA-SM-180/42 141 ------S!TE BOUNDARY FIG. 1. Sampling locations on the site of Argonne National Laboratory. 142 SEDLET and to a lesser extent the beta, activity tended to concentrate in the lower layers, but satisfactory results were obtained with carbon tetra c h lo r id e . Since the results give no information on the identity of the radio nuclides, they cannot be used for dose assessment and pathway analysis. This serious lim itation would remain even if true disintegration rates were obtainable. It is doubtful whether this measurement would be introduced now. However, when the program was begun nearly all radionuclide analyses and identifications required tedious chemical separations and absorption curves. Total activity measurements were then a useful and important part of the program. IAEA-SM-180/42 143 Benthic Materials. Soil, and Plants Benthos (bottom sediment) is collected from the same bodies sampled for water. The results indicate where conditions are appropriate for concentra ting activity, and thus where radiation doses may be higher than expected. The bed may show low-level stream contamination when water analyses do not and can indicate water contamination that was undetected in the past. Soil and plants are collected both on and off the Laboratory site. After a suspected release, grass and soil sampling is used to delineate the path of airborne contamination, particularly in areas where there are no air samplers. Soil samples are collected by cutting cores from 5 to 30 cm deep, depending on the purposes of the sampling. A commercial golf-hole cutter has proven to be an inexpensive and effective sampling device. For measurement of long term deposition, 30 cm deep cores are collected using published criteria for site selection [1]. To study a suspected recent release, the top 5 to 10 cm of soil is collected to increase the detection sensitivity. Grass is collected from a 1 m^ area, and washed before analy sis to remove surface soil, except when very recent deposition is being s tu d ie d . Until recently, all bottom sediment, soils, and plants were counted for total activity. Samples are now measured by gamma-ray spectrometry and, after dissolution, by specific radiochemical analyses for nuclides not adequately determined by gamma-ray spectrometry. Other Samples Before 1973, each precipitation on the site was collected and analyzed separately. Precipitation remains a very sensitive, but discontinuous, collector of airborne radioactivity and is particularly useful for nuclear test debris. However, the need for measuring each individual precipitation in a facilities monitoring program has diminished, and only a total monthly sample is now collected, primarily for plutonium analysis. Milk samples are collected from dairy farms near the Laboratory to have a background of information on activity levels and analytical methods in the event of a release by Argonne. Radiochemical Analysis Tritium in water is measured by liquid scintillation counting. Some of the samples are first enriched by electrolysis. Pure beta emitters and electron-capture nuclides are separated by appropriate chemical methods. An ion-exchange separation on silver chloride and solvent extraction is used for small amounts of radioiodine in milk and water. Uranium is analyzed fluorometrically by the usual sodium fluoride fusion method and radiometrically by solvent extraction. Thorium, neptunium, and plutonium are separated by a coprecipitation-anion-exchange procedure [2], and counted by alpha spectrometry. Appropriate tracers are used for chemical yield determination. Solid samples are dissolved for analysis of nonvolatile elements by vigorous treatment with mineral acids, including hydrofluoric acid. The procedure has been found to dissolve plutonium in soil that has been ignited at 1000°C. The detection lim its in water and air are summarized in Table I. At these lim its the error at the 95% confidence level is equal to the result. 144 SEDLFT TABLE I. Detection Limits for Environmental Measurements 3 Nuclide or Activity Water (pCi/liter) Air (pCi/m ) (a ) Gamma-ray em itters 1-10 c a . 0.0 0 1 -4 -5 Beta-ray emitters (b) 0 .1 - 2 10 - 10 " H ydrogen-3 200, 15^ 0.1 Neptunium, Plutonium, 5 x 1 0 * " ^ ) 10*6 Thorium (alpha) -5 Uranium (natural) 0 .2 2 x 10 --4 Total Alpha 0 .2 2 x 10 „ -4 Total Beta 1 5 x 10 (a) Measured by gamma-ray spectrometry. Sensitivity varies with decay scheme and sample size. ^Measured by beta counting of separated activity. Sensitivity varies with sample size and beta energy. (c) The first figure is by direct counting of 10 ml, the second by electrolytic enrichment of 250 ml. ^^For 10 liter samples. The relative errors at 2 and 10 times the limit are about 50% and 10%, respectively. The detection lim its for solid samples are more difficult to express briefly, but may be inferred from those for air and water. The attempt has been to use the most sensitive and selective methods within the technical and economic grasp of the Laboratory. For measurement at existing levels or at levels desired by regulatory agencies, we have found it neces sary to increase sample size and counting sensitivity markedly, at a corresponding increase in manpower and equipment costs. RESULTS A ir Continuous air sampling is considered essential to the program. Dif ferences between on- and off-site results are used as evidence of activity from Argonne. Natural and fallout nuclides have been quite uniform on and off the site, and for such nuclides small differences between two larger numbers must be evaluated. The following types of information have been useful for this purpose: directional differences in activity correlated with wind directions; ^Be activity (this cosmogenic nuclide shows a seasonal variation that is useful in interpreting stratospheric fallout); and dif ferences between site boundary and central site samples. The only airborne activity from Argonne detected off the site has been 131i; 140ва, 60gQ^ Зц^ and ^A r have been detected on the site only. tAEA-SM-180/42 145 YEAR FIG. 3. Average monthly beta activity in air filters. ANL, 1953-73. The average monthly beta activity since 1953 (Fig. 3) is a concise way to summarize fallout data. Correlation with periods of atmospheric testing and seasonal fluctuations are apparent. The exponential changes are an interesting curiosity but of little significance in understanding atmospheric t r a n s p o r t . 41 Sampling for Ar is conducted in the vicinity of operating reactors, usually downwind at a point favorable for detection. Consequently, the average concentration is less than that measured. Results are summarized in Table II for two reactors. The ^Ar concentration in the exhaust air from both reactors is about 2.5 x 105 nCi/m^. The irradiated air from CP-5 is diluted with a large volume of inactive air before discharge. As a result, the ^lAr concentrations in outside air averaged only 5 times more near CP-5 although its ^Ar production rate is about 20 times greater. Concentrations of tritiated water vapor in air are measured near the CP-5 reactor, the principal source of at Argonne, at several other on site locations, and off-site. Results for several years are summarized in Table III. The on-site locations were chosen to determine the distance required to reduce the concentration to ambient levels. In two directions from the reactor, west and south, the 500-600 m distance is at the site security fence. The results at this distance, as well as at 2000 m north east, show good correlation with wind direction and the concentration near 146 SEDLET TABLE II. Argon-41 Concentrations Near CP-5 and Juggernaut Reactors, 1968-72 3 No. of No. g Concentration (nCi/m ) Samples 10 nCi/m Avg. Max. CP-5 620 207 116 1400 * Juggernaut 323 87 23 330 Results through April, 1970, when the reactor was decommissioned. TABLE III. Hydrogen-3 Concentrations in Air, 1968-72 3 3 L o c a tio n Avg. (pCi/m ) Range (pCi/m ) CP-5 (50 m east) 270 0 .9 - 5750 500-600 m from CP-5 18 0 .3 - 178 2000 m NE of CP-5 7 0 .4 - 37 O ff-site, 10 km NW 4 .8 0 .2 - 15 CP-5. Dilution to ambient levels occurs within 500-600 m in directions other than that from which the wind is blowing. The dose due to breathing tritiated water vapor continuously at the security fence concentration is only about 0.05 mrem/yr. Water Surface water in the area contains 1-4 pCi/liter of natural alpha activity and 5-10 pCi/liter of natural beta activity. Uranium and are the principal activities. The radionuclides found at various times in Sawmill Creek due to Argonne waste water have been ^H, other fission pro ducts, 58cQ^ 60(-Q^ ang concentrations have always been very low compared to the applicable standards. The activity added to the Creek by waste water is determined from the difference in activity upstream and downstream from the waste-water outfall. When fallout was high, signi ficant amounts were added to the Creek directly from the air, and the difference method required modification by including this source. The contribution of Sawmill Creek to Des Plaines River activity can be determined directly from the concentrations above and below the mouth of the Creek, since the Creek contributes only about 0.1% of the River water. Because of the large dilution, activity from Argonne waste water has infre quently been detected in the Des Plaines River, and has been limited to ^H and 5 8 co . IAEA-SM-180/42 147 Tritium has been the most abundant nuclide discharged into the Creek. In 1972, the average concentration was 600 pCi/liter below and 270 pCi/liter above the outfall. The concentration in the Des Plaines River attributable to Sawmill Creek is quite variable, rarely greater than 100 pCi/liter, and frequently difficult to distinguish statistically from fallout and natural Зц, Although only total concentration is dosimetrically important, this does not solve the problem for the surveillance program whose function is to determine in the environment activity from a specific f a c i l i t y . 3 In a 1972 study of H concentrations in the area, the average concen tration of 16 Lake Michigan samples collected from 4 locations at depths of 5 to 260 m was 280 pCi/liter with a standard deviation of 20 pC i/liter. No significant differences as a function of depth or distance from shore were observed, indicating nearly uniform mixing. Seventeen samples from other bodies of water averaged 283 ± 30 pC i/liter, and rainfall during the same period contained about 240 pC i/liter, again indicating a near-equilibrium condition. During 1971, the content of surface water was higher, averaging 360 ± 30 pC i/liter, while precipitation averaged about 550 pCi/ l i t e r . Benthic M aterials. Soil, and Plants The normal range of total activities in bottom sediment in the area is 2-35 pCiCC/g and 2-90 pCiß/g. About 90% of normal samples contain 15- 25 pCia/g and 50-75 pCiß/g, although activity in any one stream bed can vary greatly with time and location. All of the long-lived nuclides found in Sawmill Creek water below the outfall have also been found in the Creek bed. Activity from Argonne has not been detected in Des Plaines River bottom sed im en t. Natural soil activities in this area range from 10-30 pCia/g and 40- 80 pCiß/g. The natural uranium and thorium content of soil is 1-4 pCi/g. Localized areas on the site have occasionally become contaminated with 50- 1600 pCia/g and 200-2600 pCiß/g, prim arily from uranium, but including some plutonium and 60co. Normal nonvolatile activities in plant samples range from about 0.2-3 pCia/g and 15-40 pCig/g. Most of the beta activity is due to ^0ц. Grass has been a very sensitive indicator of recent fallout acti v i t y . Radioiodine 131 A release of I in February and March, 1961, illustrates some of the detection methods we have found useful. Iodine accidentally released from heated, irradiated fuel elements coated the exhaust ductwork and leaked off over a 5 week period. The results are summarized in Table IV. Radioiodine was detected only in off-site air samples south of the site. Some of this 131l was also found in rain, and consequently appeared in Sawmill Creek, both above and below the waste-water outfall. It is thus possible to draw erroneous conclusions if only the upstream and downstream differences are used to determine the origin of activity. Soil, and particularly grass, were quite useful in delineating the extent of the release. The concentrations in grass were about 100 times greater than in soil from the same location. The surface deposition was consistent with the concentration pattern found in the air samples. 148 S ED LET TABLE IV. Summary of Iodine-131 Concentrations, February- March, 1961 3 Air on-site - 3.3 pCi/m (avg.) 16.7 pCi/m3 (max.) 3 off-site - 0.20 pCi/m (avg.) 0.41 pCi/m3 (max.) Precipitation - 500 pCi/liter 3 Grass - < 1 to 15 x 10 pCi/g - detectable up to 2400 m from release - at south and west site boundaries only Soil - < 0.3 to 10^ pCi/g - detectable up to 150 m from release Creek Water - northeast of release, 2.5 pCi/liter - southeast of release, 55 pCi/liter Plutonium Plutonium measurements were begun in 1956. The separation technique was adequate, but sample sizes were small, no chemical yield tracer was used, and counter backgrounds were higher than in modern spectrometers. Plutonium in Sawmill Creek water, attributable to Argonne operations, aver aged about 7 pC i/liter in 1956, and decreased to below the detection limit (0.05 pCi/liter) in recent years. The plutonium content of the Creek bed decreased in a parallel manner. Plutonium from nuclear tests was found in most grass samples since this analysis was begun in 1959. The concentrations ranged from < 0.005 pCi/g to 0.15 pCi/g. Positive results (0.03-0.05 pCi/g) were obtained for a few soil samples on and off the site during this period. 239 l The Pu concentrations in Sawmill Creek below the outfall in 1971-2 averaged 2.3 fC i/liter. Above the outfall the concentrations were less than 0.5 fC i/liter. The average 239p^ content of the Des Plaines River was 0.72 + 0.20 fC i/liter above and 0.55 ± 0.20 fC i/liter below the mouth of the Creek. The sim ilarity at the two locations implies that the plutonium originated in fallout. In the Illinois River at two locations below the Dresden Nuclear Power Stations, *^^Pu concentrations were 0.53 and 0.22 fC i/liter. Lake Michigan concentrations were sim ilar, from 0.49 to 0.90 fC i/liter. Corresponding concentrations have been found in ocean water. For example, 0.72 fC i/liter was reported for a June, 1971, sample of Pacific Ocean surface water [3]. Apparently the removal mechanisms operate to a sim ilar extent in water bodies of whatever size. ^ These results were obtained by alpha spectrometry, which cannot distinguish between ^P u and ^°Pu. Consequently, when ^ P u concentrations are given it should be understood that both isotopes are included. IAEA-SM-180/42 149 Our chemical separation includes Np in the plutonium fraction, and this isotope can be distinguished by its alpha particle energy. An unex pected result of the Sawmill Creek analyses was the recent detection of 237цр at an average concentration of 0.11 pC i/liter, about 5 times greater than 239pu. The 237^ identification has also been confirmed chemically. This nuclide has not been found in water from other locations. A survey of the bed of Sawmill Creek below the waste-water outfall was made in 1972. The 239рц content ranged from about 0.010 to 0.030 pCi/g; the 238рц content was about 10 times less. The concentration at each loca tion could be correlated qualitatively with the absorptive capacity of the sediment and the flow rate of the water. Samples from other stream beds gave similar results for both isotopes so it is believed that most of the plutonium in the Creek bed was from fallout. 239 238 The Pu and Pu concentrations in air on the Argonne site during 1972 averaged 27 and 2.2 aCi/nP, respectively. The monthly concentrations showed the same variations as the fission products and 7ge, and indicated a stratospheric origin. The concentrations were sim ilar to those reported by other investigators for samples collected at similar latitudes, but away from nuclear installations [4,5]. Tables V and VI summarize the plutonium measurements that have been made on soil and grass during the past two years. The error limits are twice the standard deviation of the average. Fallout deposition values found by other laboratories [6] are in the same range as those reported here, about 2 nCi/m^ for 239рц, Thte distribution as a function of depth is apparent from the table. It appears that the 238рц/239рц ratios were higher at 15-30 cm than at 0-15 cm, but the large 238рц counting errors make this conclusion speculative. Large amounts of 238рц w ill be produced in power reactors, so monitoring for this nuclide w ill continue to be im p o rta n t. 239 The Pu content of grass has been more variable than soil. The soil/grass ratio, for pairs of samples collected at the same spot, varied greatly in 1971, but included are some grass samples that were not washed and some soil samples from 0-5 cm. The ratios for 1972 were more uniform, as were the sampling and pre-treatment methods. TABLE V. Plutonium Concentrations in Soil, 1971-2 (concentrations in nCi/m2) No. o f Depth L o c a tio n Sam ples (cm) Plutonium-239 Plutonium-238 on-site (1971) 24 0-5 1 .2 + 0 .1 0.06 ± 0.04 on-site (1971) 12 0-15 2.3 ± 0.4 0.11 ± 0.05 on-site (1972) 24 0-30 1.83 + 0.24 0.22 + 0.03 off-site (1971) 9 0-5 1 .2 i 0 .5 0.08 ± 0.04 off-site (1972) 6 0-30 1.63 ± 0.29 0.19 + 0.07 150 TABLE VI. Plutonium-239 Concentrations in Grass, 1971-2 No. o f fC i/g pCi/m^ (Soil/Grass) x 10^ L o c a tio n Sam ples Range Avg. Range Avg. Range Avg. SEDLET on-site (1971) 5 1 .9 - 5 .2 3 .5 ± 1 .2 0 .1 2 - 1 .1 0.61 ± 0.32 1.6-10.8 3.5 on-site (1972) 8 1 .5 - 8 .5 3 .7 ± 1 .6 0.21-0.67 0.35 ± 0.10 3 .5 - 7 .1 5 .1 off-site (1971) 2 2 .2 - 5 .5 - 0 .7 7 - 2 .1 - 0 .3 - 0 .5 - off-site (1972) 2 0 .8 7 - 2 .8 - 0 .1 7 -0 .2 9 - 5 .4 - 6 .5 - IAEA-SM-180/42 151 239 The Pu content of monthly precipitation samples in 1972 ranged from 0.12 to 0.75 and averaged 0.41 pCi/m^. The annual 1972 deposition by pre cipitation was only 0.3% of the average soil plutonium content, and implies that in the absence of further atmospheric testing, the soil content will not increase greatly. Penetrating Radiation Measurements of gamma-ray dose have been made since 1971 with CaF2 (Dy) thermoluminescent dosimeters. The off-site measurements averaged 104 mrem/yr in 1971 and 105 mrem/yr in 1972. In 1972, 95% of the individual results were within two standard deviations (11 mrem/yr) of the average. If it is assumed that the off-site readings accurately sample the radiation back ground of the area, a result between 94 and 116 mrem/yr may be considered normal with a 95% probability. Only 2.5% of the measurements of a back ground radiation field should be higher and 2.5% lower than this range. Several values outside of this range were obtained on the site. An abnormally low reading, 81 mrem/yr at 13D (Fig. 1), was probably due to the large amount of subsurface gravel in that area, containing below average concen trations of natural activities. At only one location was it difficult to determine the origin of an above-normal value. This occurred on the fence line at location 8H, where the dose averaged 23 mrem/yr above the off-site average. At this location, several sources could have been responsible. By placing dosimeters at corresponding distances and in various directions from each of the possible sources, it was found that the above-normal dose at 8H resulted principally from a temporary radioactive waste storage area at location 71. Thermoluminescent dosimetry is a relatively simple method for obtaining sensitive dose measurements. There is little question regarding its ability to measure background radiation with good accuracy, but there is a question as to the minimum dose it can distinguish above the natural background. Considering the system in routine use at our Laboratory, with a standard deviation of 5.5 mrem/yr, readings in the range of 120-125 mrem/yr have a very high probability of being abnormal, and one might conclude that the system can measure 10-15 mrem/yr above background. However, considering the statistical aspects of the measurements, and localized areas of abnorm ally low dose rates, it is possible that 100-110 mrem/yr could also be 10 mrem/yr above the natural background. One of the difficulties in relying on thermoluminescent dosimetry alone to measure very low additions to the natural background dose is that, in some respects, this type of dosimetry is similar to total activity measurements. L ittle or no information on the energy spectrum and dose rate are obtained. Other types of dosimeters, recording ion-chambers and energy-dependent detectors, should be used in conjunction with thermolumi nescent dosimetry if reliable measurements at the 10 mrem/yr level are d e s i r e d . SUMMARY AND CONCLUSIONS The environmental monitoring program has shown that the amounts of radioactive m aterials and radiation released to the environment by Argonne operations have been small and well within the guidelines specified in the past by regulatory agencies and standards commissions. It has, however, been necessary to measure environmental radioactivity with increasing sen sitivity and specificity as the concern over smaller doses has grown. 152 SEDLET L ittle work has been done at Argonne, thus far, on inferring dose to man, except to apply the basic ICRP uptakes and internal dose models for radioactivity in air and water. This has been due, in part, to the very low concentrations found in the environment, although it may become neces sary to estimate doses at these low levels. It has been our practice to measure radioactivity close to its discharge, where the detection proba bility is more favorable. At these points, however, population exposure is slight. For example, Зц from Argonne can often be found in Sawmill Creek and the Des Plaines River, but there is little human exposure at these locations. Downstream where the water is used, dilution and other possible Зц sources make it impossible to specify the Argonne contribution, and a very approximate estimate must be made that can at best be only an admini strative value without real radiobiological significance. While the doses at such large distances are insignificant by present standards, doses that are now considered important were considered negligible several years ago. It is difficult for one measurement on a sample to answer the questions currently being asked of surveillance programs. Plutonium in soil presents this type of difficulty. To obtain the total inventory and to compare depo sition between areas, sampling down to at least 15 cm is needed. Total deposition values are also important dosimetrically since this is the amount in the environment, not under control. However, it appears that the prin cipal hazard to man from this plutonium is by inhalation of resuspended material [7]. For this pathway only the top layer is significant, and an activity profile for a soil sample, requiring several radioactivity measure ments is needed. Other information of dosimetric significance, including various properties of the radionuclides as they exist in the environment, and of the environmental media, cannot be obtained by radioactivity measure ments alone. For example, there is some evidence that the specific activity of a nuclide can affect its dissolution rate, and thus affect its environ mental and biological behavior. Such an effect has been reported for 238рц and 239pu [8]. Even after the complete characterization of the environmental forms of trace radionuclides, the behavior of these forms in the body may not be known, and additional radiobiological studies may be necessary to evaluate their dosimetric significance. There are, consequently, many more problems facing those who conduct environmental surveillance programs than simply more sophisticated and sensitive radioactivity measurements. ACKNOWLEDGEMENT The environmental monitoring program was conceived in its present form by Dr. Andrew F. Stehney, who conducted the program in the early 1950's, with valuable assistance from Henry F. Lucas. In more recent years, very significant contributions have been made by several individuals, principally Dr. Norbert W. Golchert and Fred S. Iwami. Their assistance is gratefully acknowledged. REFERENCES [1] HARLEY, J. H., HASL Procedures Manual, USAEC Rep. HASL-300 (1972). [2] GOLCHERT, N. W., SEDLET, J., Radiochemical determination of plutonium in environmental water samples, Radiochem. Radioanal. Letters 12 (1972) 215. IAEA-SM-180/42 153 [3] WONG, K. М., HODGE, V. F ., FOLSOM, T. R ., "Concentrations of plutonium, cobalt, and silver radionuclides in selected Pacific seaweeds", page 93, Proceedings of Environmental Plutonium Symposium, USAEC Rep. LA-4756 (1971) (FOWLER, E. B., HENDERSON, R. W., MILLIGAN, M. F ., E ds.). [4] HARDY, E. P ., Jr ., Fallout Program, Quarterly Summary Report (Appendix), USAEC Rep. HASL-268 (1973). [ 5 ] ENVIRONMENTAL PROTECTION AGENCY, P lutonium in a ir b o rn e p a r t i c l e s , Jan.-March, 1972, Radiation Data and Reports JL3 (1972) 703. [6] KREY, P. W., HARDY, E. P., Jr., Plutonium in Soil Around the Rocky Flats Plant, USAEC Rep. HASL-235 (1970). [7] HEALY, J. W., "Some thoughts on plutonium in soils", page 113, Proceedings of Environmental Plutonium Symposium, USAEC Rep. LA-4756 (1971) (FOWLER, E. B., HENDERSON, R. W., MILLIGAN, M. F ., E ds.). [ 8 ] PARK, J . F . , CRAIG, D. K ., SMITH, V . H ., " S o l u b i l i t y changes o f plutonium oxides in suspension and effect on biological behavior after inhalation by beagle dogs", Paper 113, Int. Rad. Protection Assoc., 9-14 Sept., 1973, Washington, D.C., Abstracts of Papers. DISCUSSION F.F. LUYKX: You mentioned that concentrations of radioiodine in grass were found to be 100 times larger than in soil from the same location. What depth of soil were you considering? J. SEDLET: The soil depth was from zero to 1 or 2 cm, and the comparison was made on the basis of activity per unit weight of sample. N.G. GUSEV: Have you measured the dilution factors of gaseous em issions in the atmosphere? By that I mean the ratio of the concentration in curies per cubic metre to the emission in curies per unit time. J. SEDLET: The emission rates have been measured for the principal sources of ^Ar and at Argonne, but I have not included this data in the paper, and I do not recall the dilution factors. N.G. GUSEV: Have you measured the fall-out velocity of aerosols? J. SEDLET: We do have this data for precipitation but, again, I have not included them in the paper. J. SCHWIBACH (Chairman): Your paper indicates the amount of environmental monitoring needed for nuclear research centres with many different sources of potential activity releases. It would be of interest to compare predicted values with actual discharge rates of the more important radionuclides. J. SEDLET: In my oral presentation I did compare the measured tritium doses with the calculated doses based on discharge rates of tritium from the CP-5 reactor and the agreement was fairly good. J. SCHWIBACH (Chairman): You mention that the main area of deposition of ^ 1 is only some 2^ km away from the facility. In a densely populated country, this might well be an area of high utilization with crops, cattle or even a large community. This is the sort of range where our main environmental protection effort should be applied. J. SEDLET: I quite agree. IAEA-SM-180/7 ANALYTICAL SYSTEM S APPLIED TO MONITORING THE AQUATIC ENVIRONMENT IN THE CONTROL OF RADIOACTIVE W ASTE DISPOSAL J.W .R . DUTTON, N .T . MITCHELL, E. REYNOLDS. L. WOOLNER Ministry of Agriculture, Fisheries and Food, Fisheries Radiobiological Laboratory, Lowestoft, Suffolk, United Kingdom Abstract ANALYTICAL SYSTEMS APPLIED TO MONITORING THE AQUATIC ENVIRONMENT IN THE CONTROL OF RADIOACTIVE WASTE DISPOSAL. Laboratory, Lowestoft, in setting controls on liquid radioactive waste disposal. This has created a considerable demand for analytical systems covering a wide range of radionuclides and environmental materials which, influenced also by the nature of discharges, has led to the adoption of a flexible approach designed to ensure comprehensive surveillance of all the radiologically important contaminants of the environment. The analytical systems in current use are discussed, together with the way in which they are applied in routine operation. In addition to the counting of radioactivity, sample preparation and the processing of counting data, largely achieved by use of digital computers, are also discussed. The range of nuclides requiring analysis covers alpha, beta, gamma and X-ray emitters. Some analysis can be done on a gross basis, especially beta, but most of the samples received are ultimately analysed for specific constituents. Specific analysis is achieved either utilizing radiometric techniques without further oreparation or following separative radiochemistry. Experience has shown that the most valuable tool is gamma spectrometry using Nal(Tl) or Ge(Li) detectors. Most of the radiochemical separations are aimed at beta emitters, but also occasionally at alpha-emitting nuclides and others at low concentrations. This group includes strontium-90, iron-55, caesium-134 and -137 and technetium-99; examples of the methods which have been developed are discussed together with their 1. INTRODUCTION The Fisheries Radiobiological Laboratory, Lowestoft is concerned with the control of radioactive waste disposal from the major nuclear sites in the UK. Monitoring of the environment occupies an important role in this work and has been discussed in another paper at this symposium! ll in the context of policy and prac tice within the UK as a whole. The principal work done at Lowestoft is related to liquid wastes, and it has created a considerable demand for analytical systems over a wide range of radionuclides and environmental materials. Just as monitoring fulfils a number of objectives, so too do the analytical systems employed to estimate radioactivity. Analytical requirements can be broadly divided into two categories - one connected with assessm ent of internal exposure pathways, the second with research. Analyses for assessment of exter nal exposure pathways are considered to be outside the scope of this paper, since most of them consist of in situ measurements of radiation dose rate; associated radionuclide analysis is undertaken in a research context. 155 156 DUTTON et al. This paper sets out to discuss the more important of these systems; it includes those depending on instrumental analysis, with little initial sample preparation, and also selected examples of analytical systems in which prior radiochemical separation is necessary before counting. 2. SENSITIVITY REQUIREMENTS Any discussion of analytical technique must be based on the sensitivities required, which are usually considered in relation to derived working limits (DWLs) for the radionuclide(s). The derived working limit is the concentration of a radionuclide in a critical material which would result in exposure of the critical group equal to the ICRP-recommended dose limit. Individual values are therefore quoted for the critical organ of exposure, which will vary between radionuclides. A suitable limit of detection for routine monitoring is considered to be 1% of the DWL, although it may often be possible to attain a much greater sensitivity and this may be necessary for research purposes. Table I shows the values for 1% of DWLs for materials such as fish, edible seaweed and shellfish, calculated on the basis of daily consumption rates of 1000, 500 and 100 g. These consumption rates are quite arbitrary and are rather larger than those usually observed in practice for each of these materials. In con sequence, required sensitivities based on them will err on the high side and if met by a particular analytical method will indicate with confidence that the method is satisfactory. The radionuclides listed as examples are representative of those for which analysis has been required in aquatic environmental materials following waste disposal from the UK nuclear power programme. Examination of the data in Table I indicates that, using 1% of the DWL as the required analytical criterion, gammarspectrometric analysis is applicable to 12 out of the 19 radionuclides: these are gamma emitters with a reasonably high decay-scheme factor (> 10%), viz. 60Co, 65zn, 95zr, 95кь 106цц llOm^g 125gt, 141(¡e, 144ce, ^ C s and 137cs. For the rest chemical separation is essential, four being pure beta emitters (89sr, 90gp 99-pc and ^H), two alpha emitters (239p^ and 241 Am) whilst the last (55pe) decays by electron capture, emitting a 6 keV X-ray. The compara bility of method sensitivity and the 1% level of the DWL is discussed further for each analytical technique. 3. INITIAL PROCESSING Most of the environments from which materials are collected are remote from the laboratory and it will rarely be possible to analyse samples immediately following collection; some initial storage is therefore usually necessary. To maintain samples in good condition they are often deep-frozen soon after collection, especially perishable biota and sediments. This is achieved on land by the use of solid carbon dioxide, and at sea by the use of the ship's cold store. On return to the laboratory, samples can be kept in good condition deep-frozen in either the laboratory deep-freezers or a commercial store. The length of storage time is kept to a minimum because some loss of moisture can occur through the plastic wrappings, and as far as possible material is processed soon after return to the laboratory. Some materials are analysed in the original wet state, for instance when an urgent result is required or if some of the contaminating radioactivity is IAEA-SM-180/7 157 TABLE I Required limits of detection, based on 1% of DWL Radionuclide Permissible daily intake, pCi/day* 1000 g/day 500 g/day 100 g/day 'н 6.6 x 10^ 66 132 660 6 ''F e 1.8 x 10 18 36 180 6(1 Co 1.1 x lo ' 1.1 2.2 11.0 S ' z . 2.2 x lo ' 2.2 4.4 22.0 S °sr 2.2 x 10* 0.2 0.4 2.2 °°S r 8.8 x 10^ 0.009 0.02 0. 9 S 'z r 1. 3xlo' 1.3 2.6 13.0 2.2 x lo ' 2.2 4.4 22.0 99^ Tc 6. 6 x lo ' 6.6 13.2 66.0 103^ Ru 1.8 x 10^ 1.8 3.6 18.0 106^ Ru 2. 2 x 10* 0.2 0.4 2.2 " " ^ A g 6.6 x 1 0 * 0.66 1. 3 6.6 ^ S b 2.2 x lo ' 2.2 4.4 22.0 2.0 x lo ' 2.0 4.0 20.0 ^ C e 2.2 x 10* 0.2 0.4 2.2 ' " c s 2.0 x 10* 0.2 0.4 2.0 4.4 x 10* 0.4 0.9 4.4 239- Pu 1.1 x 10* 0.1 0.2 1.1 2 4 1 ^ 8.8 x 10^ 0. 09 0.2 0.9 *To produce exposure of the critical organ equal to the ICRP-recommended dose limit for members of the general public; values drawn from ICRP Publication 2 L11J. 158 DUTTON et al. volatile. For most samples, however, initial treatment prior to analysis consists of removing water by oven- or freeze-drying. The preferred technique for sedi ments is freeze-drying, using a commercially-available machine which can handle up to 6 x 1 kg samples. The main advantage of this technique is that it produces a friable sample which can be easily handled during further processing; in addition, the grain size is not significantly altered from that in the wet state. For biological materials there is no significant advantage in freeze-drying, and oven-drying at 105°C is the general procedure. After drying, most biological materials are homogenized, the whole sample being rough ground using a rotary hammer mill. This procedure is quite satisfactory for gamma-spectrometric analysis, but for counting of total activity further grinding is required to produce material which will pass through a 60 ¡j,m sieve. 4. METHODS OF DIRECT RADIOMETRIC ASSAY 4.1. Total activity measurements The use - and misuse - of "total activity" methods for alpha, beta and gamma measurement has been discussed in d e tail L 2 J. There is little need for any total - gamma assays within the FRL monitoring programmes since it is so easy to do full gamma-spectrometric analysis. In the case of alpha activity there is little demand for total counting and in any case little is discharged into the environment. Levels are too low for total-alpha counting methods to be of general value, and in the few materials where alpha activity can be detected a separation procedure is usually adopted. In contrast, total-beta counting fulfils a useful role and most samples are analysed in this way, though usually only as a preliminary to further analysis. In this procedureL 3J, 200 mg of the dried sample are spread evenly on a 100 cm2 planchette, and the activity is measured using a large anti-coincidence proportional counter with a 500 ¡j,g/cm2 end-window; the activity of the sample is compared with a KC1 standard, prepared in a similar way, and expressed as pCi-equivalent 4% . Self-absorption by the source and absorption of beta particles by the detector win dow result in a poor response at very low beta energies, but the efficiency relative to 40к rises to about 50% at 0. 3 MeV, so that even ^^Co and 95zr (ßmax 0- 31 and 0. 36/0.40 MeV respectively) can be detected with ease. The equipment used incorporates a 60-position automatic sample changer. Counting of individual samples is usually terminated by pre-selected count levels, though there is an overriding time control to avoid excessively long counting times when the count is near to background level. Optional two-stage counting modes may be employed for each sample - with or without absorber, which can be inserted automatically. The same equipment can be used for total-alpha counting, utilizing the same detector but under different plateau conditions. Such samples are counted against a natural uranium standard. The natural level of 40ц in biological materials varies between about 1. 5 and 6 pCi/g (wet) ; the geometry of the system is such that samples containing 3 pCi 40K/g (wet) will generate 3 cpm from 200 mg of dried material, over and above a background of 5-6 cpm. An increase of 1 pCi/g (wet) can therefore be measured with a standard deviation of about 30%. Reference to Table I shows that this sensi tivity is inadequate for many of the more toxic radionuclides in relation to the MEA-SM-180/7 159 higher foodstuff consumption rates, though there are many situations where due to lower radiotoxicity or consumption rate this sensitivity is adequate. Nonetheless, total-beta counting operated in this way is often sufficient to act as a warning of abnormal activity in a sample, out of the range of the contamination level usually experienced, and is a useful adjunct to more sophisticated analysis. 4.2. Gamma spectrometry 4.2.1. Nal(Tl) detectors As indicated in section 2, gamma spectrometry is the most useful analytical technique applied at the laboratory to monitoring in support of the control of radio active waste disposal. There are several types of thallium-activated sodium iodide detector systems in use, the most common being a cylindrical crystal, 75 mm diameter by 75 mm deep, coupled to a 200 channel pulse height analyser with punch tape output. The detector geometry has been described elsewhereM , the most common type consisting essentially of 25 cm3 polyethylene containers surrounding the crystal; in this way one calibration suffices for varying amounts of sample (ranging from contents of 1 to 14 containers). The use of pulse height analysers and associated equipment is conventional and well-known and need not be described here. Systems in use in the laboratory have been developed into a routine for large numbers of samples, the spectra being resolved by digital computer techniques so that standardization of operation has become essential, placing special demands on instrumental stability. The most common source of error is drift of peak position with respect to channel number, which is largely due to a combination of two factors: gain in amplification stages, especially the photomultiplier which is extremely sensitive to temperature varia tions, and changes in the applied high voltage to the assembly. By u se of high stability high voltage units (no m ore than a 0. 002% change in 24 hours), careful control of the counting room temperature (20°C ± 0. 5), and the incorporation of constant voltage transformers between the installation and the mains supply point, long-term peak positional errors have been reduced to less than 0. 5% per day - i. e. less than one channel drift daily when using a 200 channel pulse height analyser. The transformers also provide increased protection against spurious mains-borne electrical noise. The spectrum is interpreted by an iterative least-squares fit technique which is described in section 6. 3. 2. Quasi-logarithmic display of the gamma energy axis is employed, which has the advantage of increasing the energy span without loss of peak profile at lower energies. A typical result is shown in Table П, which is the final print-out of a Porphyra sample. 8 ± 0. 3 pCi/g Мбцц can be detected at a 95% confidence limit; in the presence of this amount of ^^Ru, as little as 0. 3 pCi/g 144cg can just be detected. 4.2.2. Ge(Li) detectors The Ge(Li) detectors in use at the laboratory have nominal volumes of 40 and 80 cm3, and have efficiencies of between 1 and 10% respectively of typical 75 x 75 mm Nal(Tl) crystals. However, because of their superior resolution, complex spectra can be resolved more easily than by the use of the Nal(Tl) detectors and 160 DUTTON et al. TABLE П Sample No. 2831 Detector No. 2 Windscale Station No. 15 Monthly Porphyra Miscellaneous 1 September 1972 0.1878 Wt counted = 48.0 g Counting date - 10 October 1972 Counting time 2xST ERR F-ratio 40-K 2.622 0.513 104.516 95-Zr/Nb 0.360 0.112 41.589 106-Ru 7.956 0.329 2343.701 110-Ag 0.091 0.060 9.202 134-Cs 0.071 0.050 7.829 137-Cs 0.437 0.086 102.760 144-Ce 0.290 0.261 5.199 238-U 0.978 0.285 46.985 this offsets their lower detection efficiency. The Ge(Li) detectors are generally used for samples taken close to the points of discharge, where concentrations are higher and more radionuclides are often present. Unexpected nuclides can be recognized from a visual inspection of the spectra, which often creates (but not necessarily justifies) greater confidence in the quality of the data when compared with the results of the complex mathematical procedure of the Nal(Tl) least- squares-fit technique. The operation of Ge(Li) detectors is less demanding than Nal(Tl) assemblies. The detectors are maintained at a constant temperature (-196°C) by use of liquid nitrogen, and bias voltage has comparatively little effect on peak position. How ever, if the higher resolution of Ge(Li) detectors compared with Nal(Tl) is to be fully exploited, the standards required of associated electronic equipment must be correspondingly higher. In particular, variations in the ambient temperature can cause appreciable changes in peak positions, the analogue to digital converter and associated amplifier being the sensitive elements of the assembly. In consequence temperature control of the counting room is still essential for long-term stability. For adequate representation of spectra at detector resolutions currently available (about 3 keV FWHM at 1. 33 MeV) and to fully exploit this degree of resolution, it is necessary to provide pulse height analysis equivalent to about 1 keV/channel. This means that for most fission product analyses, for which 0.1-1. 0 MeV is the usual spectral range, about 1000 channels are required, IAEA-SM-180/7 161 though for more complete coverage of spectra, when such neutron activation products as 60co, 65gn, etc. are present, a 0.1-2. 0 MeV scan is employed, which needs 2000 channels. 5. RADIOCHEMICAL METHODS 5.1. Beta emitters Strontium-90 is the most important member of the group of beta-emitting radionuclides listed in Table I which cannot be measured by gamma spectrometry, and the methods employed for this radionuclide (with 89sr) are discussed below. In addition, there are occasions when beta/gamma emitters which are normally analysed by gamma spectrometry cannot be estimated in this way, either because of low concentration (relative to others present) or because of the general com plexity of the gamma spectrum, and initial chemical separation is then necessary. For example, Юбци and 144ce fall into this category and chemical procedures for them are also discussed. 5.1.1. Strontium-90 (and -89) Although the classical procedure for radiostrontium is still in use at the laboratory, the unpleasant nature and hazard of the fuming nitric acid used in the calcium/strontium separation cycle has led to the development of an alternative technique^] especially suited for biological materials. This involves the precipita tion (after initial wet ashing and dissolution) of the alkaline and rare earths as phosphates, which are then redissolved in citric acid. The solution is adjusted to 0. 5% citric acid and 1% EDTA at pH 5; under these conditions the difference in strength of complexes with Mg, Ca and Sr is utilized to effect their separation by ion exchange (Dowex-50). Strontium is retained on the column with barium and radium and is then separated from them by preferential elution with 2M HC1. If 90sr alone is required, ^ S r can be used as the yield tracer: the pure strontium fraction is precipitated as the carbonate for the gamma counting of 85gp to measure the recovery, and the 90y is separated after it has grown in to equilibrium with the parent 90sr and measured by beta counting. If analysis for both ^^Sr and ^ S r is required, stable strontium is used to measure the chemical yield. The carbonate precipitate is counted through an absorber immediately after removal of yttrium-90 to cut out 90sr particles and so determine only the 89sr, the ^ S r being measured by its 90y daughter as before. A high 90gp.89gy ratio poses difficulties in the measurement of ^ S r because 90y grows-in during the period of count with absorber, the energy of its beta par ticle being too high to be filtered out. However, the converse does not occur and 89gr does not interfere with the 90gy count even when 89gy. 90g¡- patios are very high. The quality of the 90sr determination is limited only by the blank level, counter characteristics and the quantity of material used. With detectors at FRL which have a background of 1 cpm and a detection efficiency of about 40%, 1 pCi of 90sr can be measured with ease. 100 g of biological materials are therefore sufficient to achieve a sensitivity of 0.01 pCi/g (wet). 5.1. 2. Ruthenium-106 and technetium-99 A chemical separation is used for the determination of 106Ru ^ the complexity of the gamma spectrum makes it impossible to attain the required accuracy or pre cision. The method employed is a nitric/perchloric acid destruction of 2 g dried 162 DUTTON et al. material in a modified Bethge apparatus in the presence of ruthenium carrier, fol lowed by distillation into 5N NaOH. The alkaline perruthenate solution is reduced with methanol and the precipitated RuOg reduced further to metallic Ru with Mg powder. The Ru is weighed to determine the chemical yield and counted through an absorber to cut out the beta emissions from ЮЗди, Technetium-99 also distils with the ruthenium, but, under the reducing condi tions used (1 ml of 1% methanol in water), it is not coprecipitated with the Ru. A fter rem oval of the RuC<2 precipitate by centrifuging, the alkaline supernate con tains the 99%, which is electrodeposited on to a stainless steel or copper disc and counted under a thin (80 pg/cm^) end-window detector. Recovery of the 99тс ¡g virtually quantitative)-6 J (98 ± 2%), and 1 pCi of ЭЭтс can easily be measured, so that high sensitivity can be achieved, dependent only on the quantity of biological material taken. 5 .1 . 3. Cerium -144 This radionuclide has a relatively low permissible daily intake - it is of fairly high radiotoxicity - so that a low limit of detection is required; it also has a low gamma decay-scheme factor (the most abundant gamma photon, at 0.133 MeV, fol lows only 11% of the disintegrations), so that its detection by direct gamma spectrometry is relatively poor. This is often exacerbated by the presence of 106Ru and 103Ru in many of the samples creating a large Compton continuum on the edge of which the l^ C e peak lies. The removal of radioruthenium is accomplished by wet-ashing the sample with perchloric acid in the presence of ruthenium carrier, and the ashed residue is then analysed by gamma spectrometry to obtain the 144ce content. 5.2. Alpha emitters The radiological importance of alpha-emitting radionuclides discharged in liquid wastes has, so far, been low and analysis has only been necessary for those of two elements, plutonium and americium. Analytical systems have been developed at the laboratory!^, using solvent extraction. The solvent is trioctyl phosphine oxide (TOPO) in n-heptane, into which the actinides are extracted from a solution of sodium nitrate and nitric acid. After back extraction into ammonium carbonate, sources are electrodeposited for count ing by alpha spectrometry using a silicon barrier layer detector. Conditions for the extraction are normally so chosen that both plutonium (most of which is 239рц and 240pu) and americium (only 241дт) are extracted together. The exception is when plutonium-238 and americium-241 are both present; the alpha particle ener gies of these two nuclides are so nearly identical (5.452-5.495 and 5.433-5.476MeV respectively) that resolution cannot then be achieved by spectrometry, and indivi dual chemical separation is used. Yield is determined by using spikes of nuclides not found in environmental material - for instance 236рц and The same basic method has also been used for curium, for which 244çm can be used as yield tracer though it has been detected only rarely and at only very low concentration. 6. DATA PROCESSING Most of the processing of data from radioactive counting procedures is accomplished by use of digital computers. The laboratory has an 8K PDP-8/L MEA-SM-180/7 163 minicomputer on site and has access via a remote terminal to an ICL 1907 situated at the MAFF Computer Centre at Guildford. The 1907, with its 96K memory and wide range of peripherals, is therefore suitable for complex data processing, espe cially where no high degree of urgency is required. Conversely, the PDP-8/L is ideal for simple processing and also when rapid results are required. The data processing requirements of the laboratory are well satisfied by these two instru ments. A continuous demand is met on the PDP-8/L for immediate processing of quality control data and output from equipment counting samples which contain only a single nuclide; speed may be of the essence, for instance when the results are required before further operations can be carried out. On the other hand, the com plex calculations on spectra from mixtures of nuclides counted by Nal(Tl) gamma spectrometry are ideally handled by the ICL 1907. 6.1. Alpha counting Total alpha counting data are handled in the same way as those for total beta assays discussed below. The other alpha counting data are from spectrometry when resolution of spectral peaks is required, not only for the nuclides originally in the sample but also those deliberately added as yield tracers. Due to the high degree of resolution of the solid state detectors employed, a matrix inversion technique is sufficient, for which the PDP-8/L is used. 6.2. Beta counting Output from beta counters (both gas counters and those using liquid scintil lation detectors) is made on to standard 8-hole punched tape and programs have been written for the processing of these data using the PDP-8/L system. This is a routine application of the computer, without any unusual features other than those of background, decay, wet/dry ratios, etc. mentioned in 6. 3.1 below. 6. 3. Gamma counting Procedures adopted depend on the type of detector used (i. e. Nal(Tl) or Ge(Li)) and the nature of the sample, particularly whether or not it is a complex mixture of nuclides which have not been separated; the computer facilities used for data processing provide the most convenient basis for discussion. 6.3.1. PDP-8/L Processing of Nal(Tl) data by this computer is confined to simple situations - mainly samples containing only a single nuclide - though data from samples contain ing more than one nuclide counted on the high resolution Ge(Li) detectors can also be handled on this equipment. In addition to the basic calculations on specific parts of the spectra against standards, the computer is programmed to apply corrections for background and other factors such as dry/wet ratio of the original sample, sample weight, and time between sampling and counting where radioactive decay might significantly alter the result. Except for sediments, analyses of which are quoted in pCi/g (dry), the final result is the concentration of the radionuclide in the sample in its original form at the time of collection. 164 DUTTON et al. 6.3.2. ICL 1907 The potentiality of this system, and particularly its large memory, is exploited to the full in processing data from Nal(Tl) counting systems handling samples which contain several radionuclides. Before this computer became available, a matrix inversion system was employed, utilizing other computer facilities; analyses are now carried out using a Multiple Stepwise Regression technique developed by Efroym sont-SJ, which is essentially an iterative least - squares approach. The program, which is in Fortran, is run on the ICL 1907 via a remote terminal. In the basic least-squares approach^] it is assumed that the composition of the spectra is known, which obviously imposes difficulties in dealing with environmental samples, whereas the Stepwise Regression method searches for the radionuclides. Spectra of standard sources, corrected to "counts per minute minus back ground" (cpm -bgd) are held on a magnetic tape library, together with their specific activities and radioactive decay coefficients and a background spectrum for the detector. A sample spectrum is read in, converted to (cpm - bgd), and a search made through the library to find which nuclide gives the best fit to the spectrum. The process is then repeated trying the additional effect of each of the remaining nuclides in the library in turn, again to find which of the remaining nuc lides gives the best increase in fit. This process is continued until no further improvement in fit can be made. 7. QUALITY CONTROL The production of analytical data is of little value without a knowledge of its quality. Most of the discussion on precision which accompanied the measuring techniques described earlier has been based on the statistical variations of counting procedures. These must of course be evaluated, but other factors also affect the accuracy and precision of the technique: (a) the correct functioning of the counting equipment, (b) the quality of the standards used, (c) the assessment of the "blank" value of the method, and (d) systematic errors caused by the presence of unexpected impurities. The more complex quality control tests are those applied to the functioning of the gamma-spectrometric equipment, for which the PDP-8/L is used. It is neces sary to ensure that there has been no shift in calibration, and a standard source is counted daily on each instrument and the peak positions calculated using the Zimmerman technique. This involves obtaining a log/linear relationship of the ratio A(i+i)/A(i-i) against I, where is the counts in channel (1+1), Ац-i) is the counts in channel (1-1), and I is the channel number. Thus 1п^А ^^/А ц_^ j = al + b, and the peak position is given by -b/a. The peak positions are then checked to see whether they are within prescribed limits, an appropriate message being typed out. As a check on efficiency the peak area counts are calculated and correc ted back to a fixed date, again being tested to see whether they are within the set limit. Background checks, in which selected peak area values are calculated, are carried out weekly. IAEA-SM-180/7 165 Similar but simpler tests are applied to the alpha and beta counting equipment, again using the PDP-8/L. These involve the regular counting of standard and back ground sources, which are themselves cross-checked. The standards are produced by dilution from solutions obtained from The Radiochemical Centre Ltd., Amer sham; each new standard obtained is compared with the preceding one, and replicate dilutions (following the procedures recommended by the supplier) are always made. The presence of systematic errors due to impurities is often difficult to detect, but measurement of the half-life of some radionuclides is a valuable guide; recounting at suitable intervals establishes whether the value obtained is acceptable, and the level of measured activity at which deviations from the expected half-life occur is indicative of the degree of interference from other nuclides. Systematic errors are also investigated by analysis using more than one technique; for instance gamma-emitting nuclides are regularly measured by both Nal(Tl) and Ge(Li) spectrometry - it is unlikely that interpretative errors would lead to identi cal answers. A further comparison is occasionally possible for some radionuclides, e.g. lO^Ru, by subjecting the sample to both radiochemical separation and direct gamma-spectrometric techniques. Finally, participation in intercomparison exercises is considered of vital importance in sustaining high data quality, and the recent IAEA exercises!- Ю J have proved invaluable in confirming the quality of data produced by the laboratory. 8. CONCLUSIONS An analytical service for monitoring the aquatic environment in control of radioactive waste disposal is adequately provided by the use of Nal(Tl) gamma- spectrometric systems, together with radiochemical separations for those beta-emitting radionuclides which cannot be measured by this technique. Ge(Li) gamma spectrometry provides added information to ensure the quality of the data. In this context a quality control programme is essential. REFERENCES [ l] MITCHELL, N. T., DUNSTER, H. J ., KENNY, A. W., BIRSE, E. A. B., Principles and practice of environmental monitoring in the United Kingdom. This symposium, Paper SM-180/8. [ 2] PRESTON, A ., FUKAI, R ., VOLCHOK, H. L ., YAMAGATA, N., DUTTON, J. W. R. , "Radioactivity", Ch.7, A Guide to Marine Pollution (GOLDBERG, E. G., Ed.), Gordon and Breach Science Publishers Inc., New York (1972). [ З] DUTTON, J. W. R. , Gross beta counting of environmental materials, MAFF Fisheries Radiobiological' Laboratory, Lowestoft, Tech. Rep. FRL 3 (1968). [ 4] DUTTON, J. W. R ., Gamma spectrometric analysis of environmental materials. MAFF Fisheries Radiobiological Laboratory, Lowestoft, Tech. Rep. FRL 4 (1969). [ 5] IBBETT, R. D ., The determination of strontium-90 in environmental mate rials, using ion exchange and preferential chelation techniques, The Analyst 92 (1967) 417. 166 DUTTON et al. [ б] DUTTON, J. W. R ., IBBETT, R. D. , "The determination of technetium-99 in marine biological m aterials", Symposium on the Determination of Radio nuclides in Environmental and Biological Materials, April 1973. [ 7] HAMPSON, B. L ., TENNANT, D., Simultaneous determination of actinide nuclides in environmental materials using solvent extraction and alpha spectrometry, The Analyst (In press). [ 8] EFROYMSON, M. A ., "Multiple regression analysis", p. 191, Mathematical Methods for Digital Computers (RALSTON, A., WILF, H. S., Eds.), Wiley, New York (1967). [ э1 SALMON, L ., Computer analysis of gamma-ray spectra from mixtures of known nuclides by the method of least squares, UKAEA Report AERE-M1140 (1963). [io] FUKAI, R., BALLESTRA, S., MURRAY, C. N., "Intercalibration of methods for measuring fission products in seawater samples", pp. 3-27, Radioactive Contamination of the Marine Environment, IAEA, Vienna (197 3). [ ll] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Report of Committee П on Permissible Dose for Internal Radiation (1959), ICRP Publi cation No. 2, Pergamon Press, Oxford (1960). DISCUSSION J. SCHWIBACH (Chairman): Could you comment on the radionuclides ^°Ba, 14°La, 3^s ami 32p^ which are not included in your tables? N. T. MITCHELL: The tables were not intended to be exhaustive, though they include most of the radionuclides of environmental importance. i4°Ba and ^°La are not included because they do not appear in our power station effluents, as complementary effluent studies have confirmed. Sulphur-35 and, to a much sm aller extent, phosphorus-32 are present in these effluents but neither have any high degree of environmental significance; indeed we have never seen sulphur-35 in the environment. J. SCHWIBACH (Chairman): Do you have any more information on the release rates and environmental behaviour of ^Tc? N. T. MITCHELL: Our experience with technetium has been most interesting and provided no sm all challenge to the analytical group. It was a short release which led to the contamination which we detected and the quantity is not known. The environmental behaviour of technetium-99 is very interesting, particularly in respect of the sharp contrast between its conservative behaviour in sea-w ater and its high degree of concentration in fucus sea weeds. We found it in no other biota except for a trace in shellfish. As none was detected in the major exposure pathways, it was considered to be of no significance for public health. J.M. MATUSZEK: What is the attitude of engineers intheUnitedKingdom to the use of total-beta measurements for screening? Our experience with total-beta measurements is that the engineers who are involved with regula tion want total-beta quantified into pCi or pCi rather than counts/min. Once the screening values leave the laboratory in ^Ci units, we find that the engineers start using the values for strict regulatory control despite the ambiguity of such measurements. N. T. MITCHELL: An important distinction between the United States of America and the United Kingdom is that in the latter engineers are IAEA-SM-180/7 167 rarely involved with environmental monitoring data, which is largely in the hands of health physicists and others fam iliar with their significance and shortcomings. The only regulatory use of total activity measurements which I accept as valid is for effluents. However, I have no objection to total activity measurements on environmental m aterials being quoted, provided that the radionuclide composition of the contaminating and natural activity is also specified, so that its significance can be assessed without ambiguity. G. BOERI: Could you tell us what measuring technique you use for s^Fe, what the detection limits are, and whether you have ever found ^Fe in environmental sam ples? N. T. MITCHELL: We have two techniques in use, both requiring initial separation from inactive substrate and other radionuclides. Each makes use of the 5. 9 keV X-ray, one using proportional gas counting, the other gel scintillation of a white iron phosphate, to avoid colour quenching. There is little to choose between their sensitivity, each being sufficient to permit detection of the order of 0. 01 pCi/g (wet), or lower levels if more starting m aterial is used. Final sample preparation favours the scintillation method, which also perm its of detection of 59уе on the same counting sample. We have found low levels of ^Fe of nuclear power station origin in shellfish and algae, and of fall-out origin in fish. G. BOERI: I should also like to ask a question about ^P. As phosphorus has high concentration factors and 32p is discharged by gas-cooled reactors, do you m easure 32p in environmental sam ples? N. T. MITCHELL: 32p has been found intermittently at low concentration in Magnox power station liquid effluents. Although we have searched for this radionuclide in the environment, we have only found it on a few occasions — in oyster flesh in the Blackwater estuary — and conclude that the high concentration factor apparent from stable element data is not realized by the radionuclides, perhaps because of the absence of a link in the pathway. IAEA-SM-180/44 DETAILED MEASUREMENT OF ш AIR, VEGETATION AND MILK AROUND THREE OPERATING REACTOR SITES B.H. WEISS, P.G . VOILLEQUÉ United States Atomic Energy Commission, Washington J.H . KELLER Allied Chemical Corporation B. KAHN, H. L. KRIEGER, A. MARTIN, C .R . PHILLIPS United States Environmental Protection Agency, Washington, United States of America Abstract DETAILED MEASUREMENTS OF '" I IN AIR, VEGETATION AND MILK AROUND THREE OPERATING REACTOR SITES. Recent USAEC studies at several boiling water reactors showed that an average of only 5% of the ^ 1 in the off-gas was in elemental form, while the ^ 1 in ventilation exhaust air averaged 55% elemental. The effect and fate of organic iodide species in the environment around an operating reactor have not been extensively studied. A field programme was organized by the USAEC and the Environmental Protection Agency at three reactor sites to study the air-* grass-* cow-* milk portion of the food chain. Three utility companies were involved, and several laboratories participated in sample analyses. Methods were utilized which achieved minimum detectable concentrations of 1x10*^ pCi/cm^ in air, 0.1 pCi/litre in milk, and 50 pCi/kg(diy) in grass. The general approach was to determine the relationships of concentrations among the various media and to relate these to the releases from the nuclear power plants. Of particular importance were related to rainfall at the pasture locations. Maximum air-to-vegetation transfer velocities of 2x10**^ to 3 x 10*3 m /s were calculated using measured air concentrations and the upper limit vegetation conceit- 1. Introduction In the United States, federal jurisdiction over the environmental impact of nuclear fac ilities is shared by two agencies: the Atomic Energy Commission (AEC), through the licensing of such facilitie s and the Environmental Protection Agency (EPA) by virtue of its mandate to promulgate generally applicable environmental standards. Both agencies are actively involved in developing criteria to keep radiation exposures to the public from nuclear fac ilitie s "as low as practicable". The AEC formally adopted this philosophy by amendment of the AEC regulations in 1971 [1]. At the time of adoption of this rule, the Commission noted the desirability of developing more definitive guidance in connection with the "as low as practicable" concept and indicated that it was initiating 169 170 WEISS et al. discussions with the nuclear power industry and other competent groups in order to achieve that goal. As a result, the Commission proposed an amendment to this regulation [2] which would supplement the regulation by providing numerical guides for design objectives and technical specification requirements for limiting conditions for operation of light-water-cooled nuclear power reactors (LMPR) to keep radiation exposures as low as practicable. The proposed numerical guides were based on the light-water-cooled nuclear power reactor operating experience and state of technology at that time [3]. The guides for design objectives for these reactors are intended to provide reasonable assurance that resultant increases in radiation dose equivalent to individual members of the public living at the site boundary will generally be less than 5 percent of dose equiv alent due to natural background radiation (5 millirem per year) and average dose equivalent received by sizeable population groups will Generally be less than 1 percent of that due to natural background (1 millirem per year). These proposed amendments are the subject of an AEC rule making hearing which is still in progress. Experience with the licensing and operation of LWPRs has shown that generally the limiting dose is that delivered to the thyroid glands of young children following the release of radioactive isotopes of iodine, principally i3ii. The critical exposure pathway involves the release of radioiodine in elemental form in the reactor effluents and its subsequent transport and dispersion in the atmosphere. A fraction of the airborne radioiodine is deposited on vegetation which is consumed by dairy cattle. Direct ingestion by young children of fresh milk produced by these cattle completes the sequence of events which result in the exposure of concern. The AEC has undertaken two measurement programs to better understand the in-plant and environmental behavior of radioiodine released at LWPR installations. The first was its own evaluation of radioactivity behavior in and releases from LMPR installations, including determination of the radioiodine species released in stack and exhaust gases. Pelletier [6] has compiled the data available for five boiling water reactors (BMRs) in which radioiodine species measurements were made. A significant find ing was that an average of only 5% and 55% of the total radioiodine released from the stack and ventilation exhaust ducts, respectively, was in elemental form. The second program of independent measurements is the subject of the present report. The AEC asked the EPA to participate in a cooperative evaluation of radioiodine behavior in the environs of LWPRs. Joint AEC-EPA studies of air-vegetation and vegetation-milk transfer processes were conducted at two sites. The AEC conducted air and vegetation monitoring at the third BMR site. Relevant data for the three sites and the measurement programs are given in the next section. Monitoring of the quantity of iodine released and the radioiodine concentration in air, rainfall, vegetation, and milk was conducted during the 1973 pasture season to obtain relationships between (1) the activity released and air concentrations near the site boundary, (2) air concentrations, precipitation, and vegetation concentrations, and (3) vegetation concentrations and milk concentrations. The species fractionation of radioiodine being released from each facility was re-evaluated for comparison with the results previously reported in Reference [6]. IAEA-SM-180/44 171 Descriptions of the Sites and Sampling Locations Three of the BUR sites which had been studied by Pelletier [6] were selected for the environmental radioiodine evaluation. (A) The Dresden Nuclear Power Station site in central Illinois houses three BWRs; their thermal power rating are 700 MW for Unit 1 and 2527 MM each for Units 2 and 3. Radioactive gases from the steam je t air ejector (SJAE) and all ventilation air in Unit 1 are discharged through a 90-m stack. SJAE gases from Units 2 and 3 are combined prior to release through a 94-m stack. Ventilation air streams for Units 2 and 3 are also combined and these gases are discharged through a short stack (50-m) adjacent to a 43-m high turbine and reactor building complex with a cross-sectional area of 1500 m^. (В) The Monticello Nuclear Generating Plant, north of Minneapolis, Minnesota, is a 1670 MM(th) BWR. Its reactor off- gas releases are via a 100-m stack. Three ventilation exhaust air ducts discharge gases at the roof level of the reactor building which has a height of 42-m and a cross-sectional area of 2150 m2. (c) The Oyster Creek Nuclear Power Plant located near the Atlantic Ocean in New Jersey is a 1930 №(th) BUR whose gaseous effluents are all discharged via a 112-m stack. The nearby turbine building is 20-m high and has a cross-sectional area of 1070 m^ and the radwaste building has a height of 3.4 m and an area of 166 m^. All three sites utilize delay lines for the SJAE gases to reduce the activity discharged, to the environment. In addition all employ high efficiency particulate (HEPA) filters to remove radioactive particulates from their airborne effluents. Dresden Units 2 and 3 and the Monticello plant were installing additional airborne radioactivity treatment fa c ilitie s prior to and during the period when the studies were underway, but the new systems were not operated during this period. The initial effort was concentrated at the Dresden site (Figure 1) where three sampling locations were established in late April 1973. Sampling of the facility releases continued until early July when fresh fission products from the detonation of a nuclear device were in itially detected. Sampling of all media continued for an additional two weeks to follow the ^ 1 from fallout in the environment, Table I. The 30,000-41^ pasture and milk cows were provided by the station operator. The milk cows were four Holsteins, all approximately 2.5 years old and in their fir st milking cycle. The cows were milked twice daily.and were otherwise free to graze. Each cow was fed approximately 2 kg grain plus 0.5 kg protein supplement daily and had a non-iodized salt-lick available. A shallow spring-fed pond provided drinking water. Milk samples were collected daily at the pasture and approximately once per month at a "background" farm 11 km SSW of the station. An 11 litre aliquot from the morning milking and an 11 litre aliquot from the evening milking on the previous day were passed at initial flow rates of 100 ml/min. through an 80 cm3 anion-exchange resin (Dowex 1 x 8, 20-50 mesh, Cl'form). Air samples were collected usually weekly, using a constant flow (0.85 m^/min.) squirrel cage blower. The two-stage sampler was composed of a 23 cm x 18 cm charcoal impregnated filte r paper (AC-1) followed by a 2.5-cm bed of tetraethylenediamine (TEDA) - impregnated charcoal. Based on laboratory evaluations at higher mass loadings and higher relative 172 WEISS et al. IAEA-SM-180/44 173 TABLE I. DRESDEN STATION SAMPLING INFORMATION Distance Sampling From Location Media Sampled Stack Di stance D-l Milk, Air, Vegetation, Precipitation 1.3 km W D-2 Air, Vegetation 1.2 km HNW D-3 Air, Vegetation 2.6 km s humidities [4], we estimate that the ^ 1 collection efficiency of the combined sampling stages is greater than 95%. Grass samples of approximately 1 kg wet weight were collected from measured areas of 1 m2 or more cut to approximately 1 cm above ground. Rainwater was collected in two plastic lined trays with a total collection area of 0.35 щ2. After the end of every rain, but no more than once daily, the collected rainwater was poured into bottles and sent with the plastic liner to the laboratory where the radioiodine was chemically converted to the iodide and precipitated as Agi for counting. Field studies were begun at the Monticello site (Figure 2) in mid- June 1973. The program was essentially the same as that outlined above, except that rainwater samples were passed through 40 ml of the above mentioned anion resin which was then counted in a large Nal (TI) well counter. Table II contains data on the Monticello sampling program. Environmental measurements around the Oyster Creek site (Figure 3) were started in mid-May. Because that plant is located in an area of poor pastures and without a significant dairy farming industry, the principal emphasis was placed on measuring air concentrations and evaluation of atmospheric dispersion characteristics for the single plant release point. Data for the Oyster Creek sampling program are contained in Table III. Analytical Methods Because of the large numbers of samples collected and the long counting times required for each analysis, several analytical laboratories were enlisted to analyze the samples for ^зц . EPA's National Environmental Research Center, Cincinnati, Ohio and Environmental Radiation Facility, Montgomery, Alabama, performed analyses of the vegetation, milk, and precipitation samples report ed for the Dresden and Monticello pasture locations, respectively. AEC and AEC contractor laboratories_(Health Services Laboratory, Battelle-Northwest Laboratory, Oak Ridge National Laboratory, Allied Chemical Corporation, and Aerojet Nuclear Company) and the Naval Ordinance Laboratory performed analyses of all air sampler components and of additional vegetation samples obtained at the ten sampling locations. All the grass, air filte r and charcoal samples were analyzed by gamma spectrometry using Ge(Li) detectors. Each sample was normally counted for 1000 minutes. Milk and precipitation samples were counted by various methods, i.e ., Ge(Li), Nal(Tl) and low background beta detectors, depending on the manner of preparation. IAEA-SM-180/44 TABLE I I. MONTICELLO PLANT SAMPLING INFORMATION Sampling Distance Locations Media Sampled from Stack Oirection M-1 Milk, Air, Vegetation 2.7 km NM Precipitation M-2 Air, Vegetation 2.4 km ESE M-3 Air, Vegetation 1.2 km SSE M-4 Air, Vegetation 1.2 km USM Details of the procedures for the analysis of milk and vegetation samples have already been presented by the EPA-Cincinnati group [5,7] Similar procedures were utilized by the EPA-Montgomery laboratory. The AC-1 filte r and the charcoal bed contents were not treated chemically; both were counted directly in standard geometries. 13Ц standards prepared by the U.S. National Bureau of Standards were distributed by the AEC to each of the participating laboratories for calibration purposes. Typical detection limits for the analysis of ^ i i ^ the various media sampled are shown in Table IV. It should be pointed out that the detection limits listed were not the lowest reported but rather those that essentially all the laboratories could obtain. The variations in the detection limits were due to: 1) the large number of laboratories participating in the study and the differences in equipment utilized by those laboratories, 2) the variations in background activities, particularly ''Be, from the beginning of the study until its conclusion, 3) the delay in counting samples due to transport or backlog of samples; and 4) in the case of grass, an occasional inadequate quantity of sample. Plant Release Rates and Radioiodine Species Station operators are required to measure the amounts of radio activity present in effluent gases. For the radioiodine this requirement is usually fu lfilled by analyzing impregnated charcoal cartridges through which known fractions of effluent streams have passed. In order to independently evaluate the is iî release rates, the cartridges were forwarded by the licensee to the AEC for analysis. Except as indicated below, the release rates reported here were obtained from the independent evaluation of in the sampling cartridges and calculated using independently obtained exhaust flow data [6]. Figures 4 - 6 contain the release rates (Q,^Ci/sec) measured during the study period. As indicated in Figure 4, the Dresden Station release rates during the first few weeks are those measured by Commonwealth Edison. For the period following May 24, the average ratio of the AEC measured stack release rates shown in Figure 4 to the stack release rates reported by Commonwealth Edison was 0.88 which indicates that the two sets of measurements are in reasonable agreement. As indicated previously, the three stations were selected from those for which independent radioiodine species measurements had been previously obtained. Reference [6] describes the sampling 176 WEISS et al. IAEA-SM-180/44 177 TABLE I I I . OYSTER CREEK PLANT SAMPLING INFORMATION Sampling Location Media Sampled Distance from Stack Direction 0C-1 Air, Vegetation 2.4 km NNE 0C-2 Air, Vegetation 2.4 km SE OC-3 Air, Vegetation 2.0 km W TABLE IV.. 13Ц DETECTION LIMITS(a) AT TIME OF COUNTING Medium Detection Limit Air-Charcoal Impregnated Paper 5 x 10-" pCi/m3 Charcoal 1 x 10'^ pCi/щЗ Vegetation 50 pCi/kg (dry weight) Milk 0.1 pCi/litre Precipitation 0.1 pCi/litre (a) In this report, those values reported as less than a detectable level are less than 2o of the interference. The errors associated with measured values are quoted as + la. WEISS et al. FIG. 5. Monticello plant release rates. train for the species differentiation measurements. Table V presents the data on radioiodine species observed together with that obtained previously. The data show that the species frac tionation of the radioiodine released does not remain constant; however, the general pattern observed previously has not changed dramatically. The species fractionation, particularly for the ventilation exhausts is probably related to in-plant conditions (such as the existence of surface contamination) but the available data are not adequate to permit analysis of this hypothesis. Observed Air Concentrations Tables VI - VIII contain the observed air concentrations, X (pCi/m3), of 13^ for the ten sampling locations. The tabled values show the relative distribution of the 13Ц on the two sampling media and the total ^зц concentration. Evaluation of these data to obtain partial information on the radioiodine species in IAEA-SM-180/44 179 the environment is underway but has not been completed. In addition, meterological data from the three plants are being used to calculate atmospheric dispersion parameters (X/Q,m*2) for the ten sampling sites. These results will be compared with the observed values of X/Q and reported later. 6. Radioiodine Measured in Precipitation Collectors and Pasture Vegetation Table IX contains the observed precipitation amounts, the activity found in precipitation collectors, and the areal concentration of is ij ^ pasture vegetation for the Dresden Station and Monticello Plant surveys. Using the air concentration data from the period prior to fallout arrivai given in the previous section and the frequently observed 5-day effective half-life for on pasture vegetation [8], we computed daily sequences of predicted vegetation concentrations 180 WEISS et al. TABLE V. COMPARISONS OF RELEASED RADIOIODINE SPECIES MEASUREMENTS Monticello Plant Radioiodine Radioiodine Species (%) at the End Species (%) in of the SJAE Delay Line Ventilation Exhaust Air ¡3¡l Species Ref. 6 Present Study Ref. 6 Present Study Particulate (a) <0.1 ^0.04 12.0 10.3 l2 22.0 21.7 57.9 29.3 HOI 21.8 43.1 12.6 21.7 Organic 56.2 35.2 17.5 38.7 Oyster Creek Plant Radioiodine Species (%) at the End of the SJAE Delav Line 1 Species Reference 6 Present Study Particulate (a) <0.1 <0.2 l2 0.06 - 0.1 10.5 HOI 12.5 22.9 Organic 87.5 66.4 Dresden Plants (Units 2 and 3) Radioiodine Species (%) at the Radioiodine Species (%) End of the SJAE Delay Line in Ventilation Exhaust Air 131i Species Ref. 6 Present Study Ref. 6 Present Study 2 3 Unit 2/Unit 3 2 3 Unit 2 Unit 3 Particulate (a) <0.1 <0.1 <0.1 8 <0.1 24 <4 12 1.7 1.9 1.7 45 71 56 54-58 HOI 21.2 27.4 26.3 15 18 7 <4 Organic 77.1 70.5 72.0 32 11 13 38-42 Analysis of sequential particulate filters showed [6] that much of the radioiodine collected by these filters is actually I2. The same detailed analysis was not made in this study; it is possible therefore that the tabled values for I2 for the present study are lower than the true values by as much as the amount reported as "particulate". for various air-to-vegetation transfer velocities. The upper and lower limit air concentrations for each period were used to compute minimum and maximum transfer rates for each transfer velocity; the air concentrations were assumed to be constant throughout the sampling period. For the Dresden pasture location an average transfer velocity of Vd4^ x 10*3 m/sec was found to be consistent with the measurements; i.e ., if this value of Vd were assumed, none of the predicted vegetation concentrations would be greater than the detection limit for i3ij in vegetation during the period. Although definitive data for the Dresden Pasture (D-l) are lacking, the May 22 and July 4 measurements indicate <23% and 42% of the available isií transferred to the surface by wet processes was effectively retained by the vegetation. For the Monticello IAEA-SM-lSO/44 181 TABLE VI. OBSERVED AIR CONCENTRATIONS - DRESDEN STATION 1973 Filter Air Concentrations X (pCi/m3) Dates Type D-l D-2 D-3 4-24 AC-1 <2x10*2 <2x10-2 <2x10-2 to Charcoal <1 x 10-2 <8x10-3 <8x10-3 4-26 Total <3x10-2 <2.8 x 10*2 <2.8 x 10-2 4-26 AC-1 <7 x 10-3 <9x10*3 <4x10-3 to Charcoal <5 x 10-3 <7x10*3 4.0 ± 1.1 x 10-3 5-3 Total <1.2x10-2 <1.6x10*2 4 - 8 x 10-3 5-3 AC-1 <4x10-3 4.0±1.6x!0"3 <4x10*3 to Charcoal 3.7 ± 1.2 x 10-3 <5 x 10*3 <4x10*3 5-9 Total 3.7 - 7.7 x 10-3 4 - 9 x 10*3 <8x10-3 5-9 AC-1 <1 x 10*3 <1 x 10*3 <9x10*4 to Charcoal 3.3 ± .8 x 10-3 <3x10-3 <2x10*3 5-16 Total 3.3 - 4.3 x 10*3 <4x10-3 <2.9 x 10*3 5-16 AC-1 <1.2x10-4 <1.4x10*4 <1.5x10-4 to Charcoal <2.1 x 10-3 <1.4x10-3 <1.6x10-3 6-1 Total <2.2 x 10-3 <1.5x10*3 <1.8x10-3 6-1 AC-1 <2.5 x 10*4 <3.1 x 10*4 <2.8 x 10-4 to Charcoal 6.2 ± 1.2 x 10-3 <2.8 x 10-3 <2.8 x 10-3 6-8 Total 6.2 - 6.5 x 10-3 <3.1 x 10*3 <3.1 x 10-3 6-8 AC-1 2.7 ±.5x10-4 9.3 ±.6x10*4 <2.2 x 10-4 to Charcoal 6.9 i 1.1 x 10-3 4.7+1.2x10*3 <2.5 x 10-3 6-15 Total 6.9 - 7.2 x 10-3 5.6+1.2x10*3 <2.7 x 10-3 6-15 AC-1 6.7 ± 2.5 x 10-5 <1.0x10*3 <1.4x10-4 to Charcoal <1.7x10-3 <1.7x10-3 1.2 ± .3 x 10-3 6-22 Total 0.07 - 1.7 x 10-3 <2.7 x 10-3 1.2 - 1.3 x 10-3 6-22 AC-1 1.4 x 10-4 <4.7 x 10-4 to Charcoal 8.9 i .3 x 10-3 <1.9x10-3 <1.5x10-3 6-29 Total 8.9 - 9.0 x 10-3 <2.0 x 10*3 6-29 AC-1 5.1 ± .3 x 10-4 5.0 ±1.2x10*3 2.7 + .8 x 10-3 to Charcoal <2.4 x 10-3 3.9 ±1.2x10*3 3.4 ± 1.2 x 10-3 7-4 Total 0.51 - 2.9 x 10-3 8.9 ±1.7x10*3 6.1 + 1.4 x 10-3 7-4 AC-1 3.1 i .5 x 10'4 -3 2.4 ± .6 x 10*3 to Charcoal <1.1 x 10-2 2.0 +.3x10^ 2.0 ±.3x10-3 7-7 Total 0.031 - 1.2 x 10-2 4.4 ±.7x10*3 7-7 AC-1 2.3 + .2 x 10-3 (a) 2.1 ±.5x10-3 to Charcoal 1.68 ± .06 x 10*2 1.70 + .05 x 10-2 8.2 ±.2x10*3 7-19 Total 1.91 ± .06 x 10*2 10.3 ±.5x10*3 (a)Analysis not available. pasture (M-l) the June 17 and 25 data indicate <9% of the airborne ^ 4 transferred to a unit surface area was retained by vegetation Significant quantities of ^^1 were carried down by rain at the time of arrival of nuclear weapons test fallout at the Monticello site. Met deposition estimated for the July 6 and July 9 storms and vegetation samples collected on July 6 and July 10 (the July 9 vegetation sample was taken before the rainstorm) indicate areal concentrations (pCi/m^) for vegetation which are 52% and 18% of those found for surface deposition collectors. All those values are greater than the 5.9% initial retention observed during an 182 WEISS et al. TABLE VII. OBSERVED AIR CONCENTRATIONS - MONTICELLO PLANT Air Concentrations X (pCi/m3) 1973 Filter Dates Type M-l W-2 M-3 M-4 6-12 AC-1 2.4±.lxl0'3 <5.5x10*3 <1x10*3 to Charcoal 2.5±.3x10*3 <1.3xl0'2 <1.7x10-2 <3x10-3 6-15 Total 4.9±.3x10-3 <1.8x10*^ <1.8x10-2 6-15 AC-1 5.5±.8x10-4 <1.9x10-3 <7x10-4 to Charcoal 4.9±.7x10-4 <2x10*3 <2x10-3 9.5+2.4x10*4 6-22 Total 10.4+1.1x10-4 <3.9x10-3 <2.7x10*3 6-22 AC-1 3.1+.4x10-4 (a). 1.4+.4x10*3 <1x10-4 to Charcoal <5x10-4 <2x10*3 3.3+.9x10*3 <5x10-4 6-29 Total 3.1-8.1x10-4 4.7+1.0x10*3 <6x10*4 6-29 AC-1 1.0+.2x10-3 4.5+1.7x10-3 3.6+.5x10*3 to Charcoal 9.0±1.5x10-4 <3x10*3 <3x10*3 2.2+.2x10-3(a) , 7-6 Total 1.9±.3xl0-3 4.5-7.5x10*3 3.6-6.6x10*3 7-6 AC-1 S.3±.2x10-3 <9.6x10-3 1.0±.2xl0-2 to Charcoal 8.3±.3x10-3 9.1+1.5x10*3 3.9±.1x10-2 10.1+.4x10-3 7-13 Total 13.6±.4x10*3 9.1-18.7x10*3 4.9+.2x10*2 7-13 AC-1 (a) , <3.4x10*3 2.li.3x10*3 to Charcoal 1.1±.lxl0'2 <4.4x10-3 <3.0x10*3 <1.0x10-3 7-18 Total <7.8x10-3 2.1-5.1x10-3 (a)Analysis not available. extended study of fission product deposition on pasture vegetation [9]. In the cases where fallout was present, dry deposition may have contributed to the apparent elevated retention of ^ i i by vegetation. At the Monticello site, the observed effective half-time for this i3ii transferred to vegetation by wet processes was about 2 days; this corresponds to a removal half-time of 2.7 days. A 2-day effective half-life is uncommonly short; for example: the range of effective half-times observed in the entire CERT series was 3.5 - 6.5 days. [10] The observed weekly air concentrations were also used to obtain time sequences of predicted 13^ concentrations on vegetation at the Monticello pasture for various air-to-vegetation transfer velocities. In a manner similar to the Dresden Pasture comparison, the 5-day effective half-life was assumed for the period prior to the arrival of weapons test fallout. A constant transfer velocity of 3 x 10*3 m/sec is consistent with the measured upper limits for pasture vegetation concentrations. 7. Observed Radioiodine Concentrations in Milk Table IX also contains the data on i 3ij concentrations in milk from the four cows maintained at each of the two pasture locations. Half of the sample for a particular data came from the milk collected the previous evening. IAEA-SM-180/44 1 TABLE V III. OBSERVED AIR CONCENTRATIONS - OYSTER CREEK PLANT Air Concentrations X fpCi/m^) 19/Л n ¡ter Dates Type 0C-1 0C-2 0C-3 5-15 AC-1 <1.8 x 10-2 <1.6 x IO*? <1.4 x 10*3 to Charcoal 3.3 ± 1.2 x 10*3 <2.3 x 10-3 <2.5 x 10*3 5-22 Total 0.3 - 2.1 x 10*2 <1.8 x 10-2 <3.9 x 10*3 5-22 AC-1 (a) , (a) . (a) , to Charcoal <2 x 10*3 <1.6 x 10-3 <1.6 x 10*3 5-28 Total - -- 5-28 AC-1 (a) (a) , (a) , to Charcoal 5.5 i .8 x 10-3 <1.8 x 10*3 <1.8 x 10*3 6-5 Total - - - 6-5 AC-1 (a) (a) , (a) , to Charcoal 5.3 i 1.1 x 10-3 2.6 i .1 x 10*2 <2.2 x 10*3 6-12 Total - - - 6-12 AC-1 1.1 i .3 x 10-3 <2.2 x 10-3 (a) to Charcoal 4.7 ± .5 x 10*3 <2.1 x 10*3 Mo Analysis 6-19 Total 5.8 i .6 x 10*3 <4.3 x 10-3 6-19 AC-1 <6.3 x 10-5 <9 x 10-4 (a) to Charcoal <5.7 x 10-4 2.2 ± .8 x 10-3 2.3 ± .2 x 10-2 6-26 Total <6.3 x 10-4 2.2 - 3.1 x 10-3 6-26 AC-1 <2.0 x 10-4 3.1 i 1.1 x 10-3 (a) . to Charcoal 7.7 i 1.5 x 10-4 1.8 i .1 x 10-2 <1.2 x 10*^ 7-3 Total 7.7 - 9.7 x 10-4 2.1 i .1 x 10-2 7-3 AC-1 3.3 ± .3 x 10-3 4.1 i .9 x 10*3 (a) to Charcoal 1.37 ± .07 x 10-2 2.7 i .9 x 10-3 2.6 + 1.0 x 10'' 7-10 Total 1.70 ± .08 x 10*2 6.8 i 1.3 x 10-3 7-10 AC-1 2.7 ± .1 x 10*3 4.2 t .9 x 10*3 to Charcoal 6.0 ± .8 x 10-3 4.0 i .8 x 10-3 7-17 Total 8.7 + .8 x 10-3 8.2 ± 1.2 x 10*3 (a)Analysis not available. Prior to the arrival of fallout in early July, concentrations of i3ii in milk were very close to the nominal detection limits for both monitoring programs. All of the appearances of i3ij -¡n the milk at Monticello followed rainstorms which resulted in measurable transfer of i 3ij to the precipitation collectors even though the amount of ^3ii on vegetation remained undetectable. At the Dresden Pasture the only positive milk concentration not immediately preceded by measurable wet deposition occurred on June 30. The highest observed air concentration (excepting fallout periods) occurred during the week of June 22 - 29; dry transfer processes apparently increased vegetation concentrations sufficiently to cause the observed i 3ij concentration even though the increased concen tration in vegetation was not measurable. Of considerable interest are the ratios of the concentration of i3ij in milk (Сщ, pC i/litre), to (a) the 13^ concentration on vegetation covering a unit surface area (Cy, pCi/m^) and (b) the 184 WEISS et al. TABLE IX. !31i TRANSFERRED TO PRECIPITATION COLLECTORS, PASTURE VEGETATION & MILK ______Dresden Pasture (D-l)______Monticello Pasture (M-l)_____ 1311 Concentrations______1311 Concentrations______Precip. Pasture Precip. Pasture 1973 Rainfall Collector Vegetation Milk Rainfall Collector Vegetation Milk Dates fpCi/m?) (pCi/m2) (pCi/1itre) (mm) (pCi/m2) (pCi/m2) (pCi/litre) May 1 30 <0.4 <0.1 2 3 <0.4 <6 <0.1 3 <0.2 Mo Samples Taken Prior to June 13. 4 <7 <0.1 6 <0.1 6 <7 <0.1 7 <0.1 8 <7 <0.1 9 10 <0.4 <0.1 10 <6 <0.1 11 <0.1 12 <11 <0.1 13 <0.1 14 <7 <0.1 15 <0.1 16 <6 <0.1 17 <0.1 18 <8 <0.1 19 12 <0.3 <0.1 20 <7 <0.1 21 <0.1 22 10 22 + 3 <5 <0.1 23 <0.1 24 <7 <0.1 25 24 9 ± 2 <0.1 26 <9 <0.1 27 29 8 + 2 <9 <0.1 28 20 8 ± 2 <4 <0.1 29 5 <0.4 0.17 i .05 30 20 <0.4 <5 0.13 ± .03 31 0.26 ± .06 June 1 <8 <0.1 2 0.17 + .06 3 3 5 ± 1 <6 <0.1 4 5 <0.1 5 5 <0.7 <30 6 7 <0.8 <0.1 7 <9 <0.1 8 <0.1 9 <15 <0.1 10 <0.1 11 <10 <0.1 12 <0.1 Start of Monitoring Program <0.1 13 <0.1 <3 <0.1 14 <0.1 <0.1 15 10 <0.3 <12 <0.1 <3 <0.1 IAEA-SM-180/44 185 TABLE IX. 131j TRANSFERRED TO PRECIPITATION COLLECTORS, PASTURE VEGETATION & MILK (Cont.) Dresden Pasture (D-1)______Monticello Pasture (M-1) _____ 1311 Concentration______T31i Concentration Precip. Pasture Precip. Pasture 1973 Rainfall Collector Vegetation Milk Rainfall Col lector Vegetation Milk Dates (mm) (pCi/m2) (pCi/m2) (pCi/litre) (mm) (pCi/m?) (pCi/m?) (pCi/litre) June 16 <0.1 25 <20 <0.1 17 45 <0.3 <0.1 22 36 + 10 <3 <0.1 18 <0.1 <0.1 19 6 <0.5 <0.1 <3 0.21 + .05 20 10 <0.5 <0.1 <0.1 21 <0.1 0.17 + .05 22 <0.1 <0.1 23 <0.1 3 <0.1 24 10 <0.5 <0.1 <0.1 25 <8 <0.1 29 35 + 11 <3 <0.1 26 0.24 + .05 27 6 <0.5 <9 <0.1 <3 0.26 ± .05 28 <0.1 0.26 ± .1 29 <10 <0.1 <4 <0.2 30 0.27 ± .08 0.31 ± .05 July 1 <7 <0.2 <5 <0.2 2 <0.1 15 <30 0.32 + .05 3 10 <0.1 0.23 + .05 4 10 35 i 1 <0.1 <0.2 5 14 + 5 <0.1 <15 <0.1 6 6 + 2 3.7 ± .1(b) 18 214(a) 111 ±15 <0.1 7 <19 2.6 ± .05 82 ±6 22.6 ± .3(b) 8 1.40 ± .04 51 ±3 58.2 ± .5 9 <6 1.31 ± .04 10 120(a) 40 + 3 52.8 ± .4 10 - 60 ±2 91 ± 1 11 9 ± 4 2.8 i .05 161 ±6 79.7 + .5 12 3.3 i .06 85 ±6 73.0 ± .4 13 7 ± 2 3.2 ± .05 39 ±3 71.0 ± .4 14 3.2 + .1 59 ±5 77.0 ± .4 15 <11 3.8 + .06 32 ±3 64.2 + .4 16 4.1 ± .1 28 + 3 54.1 i .5 17 <7 3.6 + .1 41 ±4 45.1 i .3 18 <6 3.6 i .05 41 ±4 26.2 ± .2 19 20 2 5 ± .7 2.5 ± .05 - (a)computed from the measured value of 285 ± 25 pCi/m^ for July 9 which covers the period July 2-9. The depositions on July 6 and 9 were assumed to be proportional to the rainfall on those dates. The estimate for July 6 includes the correction for radioactive decay. (b)All 131i concentrations in milk from this date forwarded are due to fallout from detonation of a nuclear device. Confirmation was made by identification of fresh fission products on arass and in air. ^ 186 WEISS et al. TABLE X. MILK-TO-VEGETATION TRANSFER PARAMETERS, DRESDEN PASTURE ______Vegetation Data______Milk Data Transfer Parameters 1973 Cv Kv 1973 Cm Cm/Cv lOOCm/1 Date (pCi/m2) (pCi/kg)^) pate (pCi/litre) (m2/litre) (%/litre) 5-27 <9 <47 5-21 0.17 ±.05 >0.019 >0.036 5-28 <4 <20 5-30 0.,13 ±.03 >0.033 >0.065 5-30 <5 <20 5-31 0.,26 ±.06 >0.052 >0.13 6-1 <8 <40 6-2 0..17 ±.06 >0.021 >0.042 6-29 <10 <47 6-30 0..27 ±.08 >0.027 >0.057 7-5 14 + 5 130 + 30 7-6 3..7 i .1 0.26 ±..09 0.28 i -07 7-5 14 ±5 70 ±25 7-7 2..6 ±.05 0.19 ±..07 0..37 ± .13 7-6 6 ±2 30 ±15 7-8 1..40 ±.04 0.23 ±.,08 0..47 + .23 7-7 <19 <60 7-9 1..31 ±.05 >0.089 >0.22 7-10 9 ±4 80 + 35 7-12 3,.3 ±.06 0.37 ±..16 0..41 ± .18 7-11 9 i 4 80 ±35 7-13 3..2 ±.05 0.36 ±..15 0..40 ± .18 7-13 7 i 3 70 ±25 7-14 3..2 ±.1 0.45 ±,.20 0..46 i .16 7-13 7 ±3 70 ±25 7-15 3..8 ±.06 0.54 ±,.23 0..54 ± .19 7-15 <11 <67 7-16 4..1 i .1 >0.37 >0.61 7-15 <11 <67 7-17 3.,6 i .1 >0.33 >0.53 7-17 <7 <60 7-18 3 .6 ±.05 >0.51 >0.60 7-18 <6 <47 7-20 2..5 ±.05 >0.42 >0.53 Mean of Positive Values 0.34 0.42 Variation About the Mean(lo) 0.13 0.08 t^)Dry weight of vegetation; determined by drying to constant weight at 110° C. TABLE XI. MILK-TO-VEGETATION TRANSFER PARAMETERS, MONTICELLO PASTURE Vegetation Data Milk Data Transfer Parameters 1973 Cv Kv,- 1973 Cm Cm/Cv 100 Cm/I Date (pCi/m2) (pCi/kg)(a) Date (pCi/litre) (m^/litre) (%/litre) 6-17 <3 <25 6-19 0.21 ±.05 >0.07 >0.08 6-17 <3 <35 6-21 0.17 ±.05 >0.05 >0.05 6-25 <3 <19 6-26 0.24 ±.05 *0.08 >0.13 6-25 <3 <19 6-27 0.26 ±.05 >0.08 >0.14 6-27 <3 <21 6-28 0.26±.l >0.08 >0.12 6-29 <4 <34 6-30 0.31 ±.05 >0.07 >0.09 7-1 <5 <38 7-2 0.32 +.05 >0.06 >0.08 7-1 <5 <34 7-3 0.23 +.05 >0.04 >0.07 7-5 <16 <125 7-7 22.6 ±.3 >1.4 >1.8 7-6 111+15 650+88 7-8 58.2 ±.5 0.52 ±.07 0.90 ±.12 7-7 82 ±6 530 ±40 7-9 52.8 ±.4 0.64 ±.05 1.00 ±.07 7-8 51+3 750+45 7-10 91 ±.5 1.78 ±.10 1.21 ±.07 7-9 40 + 3 550 ±38 7-11 79.7 ±.5 1.99 ±.15 1.45 ±.10 7-10 60 ±2 540 ±20 7-12 73.0 ±.4 1.22 ±.04 1.36 ±.05 7-11 161 + 6 990+35 7-13 71.0 +.4 0.44 +.02 0.72 ±.02 7-12 85 + 6 400+25 7-14 77.0 ±.4 0.91 ±.06 1.92 +.12 7-13 39 + 3 470 ±29 7-15 64.2 ±.4 1.65 +.13 1.37 +.08 7-14 59 + 5 400 ±30 7-16 54.1 ±.3 0.92 ±.08 1.36 ±.10 7-15 32 ± 3 320 ±50 ли 45.1 +.3 1.41 +.13 1.40 +.13 7-16 28 ±3 430 ±30 7-18 26.2 +.2 0.94 ±.10 0.61 ±.05 Mean of 11 Positive Values L U 1.21 Variation About the Mean (lc) 0.52 0.38 (a)Dry weight of vegetation; determined by dryingto constant weight at 110° C. IAEA-SM-180/44 187 daily intake (I,pCi) of ^^1. Tables X and XI contain the computed values of these ratios for the Dresden and Monticello pasture locations respectively. Wherever possible, the grass concentrations used in the computation was that measured two calendar days before the date of the milk sample. This corresponds approximately to a 36-hour interval between consumption and the appearance in the milk. A daily ingestion of 10 kg (dry weight) of pasture vegetation was assumed in computing the daily ^ 1 intake; this value may be to some extent higher than that normally assumed, but is considered reasonable for cattle in the circumstances prevailing. The average value of (100 Cm/I) computed using the Monticello data is somewhat higher than the expected value of 1 %/litre. The Dresden mean is markedly lower, but owing the great variability among the individual values, the two are not statistically different at the 95% confidence level. Hoffman [8] has summarized information on the ratio Qn/Cv (m^/litre): his "high extreme", "average" and "low extreme" values are 0.87, 0.26, and 0.07 m^/litre respectively. The mean for Monticello is greater than, but not statistically different from, his high extreme value. During the July 6-16 period at Monticello the mean value of Cy was 68 ± 40 рС1/щ2 (sample standard deviation is indicated) and the average vegetation density was 123 + 50 g/m^. Assuming the 10-kg/day consumption rate and 1 %/litre for the transfer parameter one computes an expected average milk concentration of 55 pCi/1itre and а Сщ/Су ratio of 0.81. An average C^ of 63 pCi/litre was observed; the average ratio of C^/Cy was 1.13 m^/litre. For the corresponding period of positive vegetation and milk concentrations at Dresden, Cy was 9.4 ± 3.7 pCi/m^ and the average vegetation density was 138 ± 57 g/m^. If the same consumption rate and transfer parameter are assumed, the expected average milk concentration is 6.7 pCi/1itre and the expected С /Cy ratio is 0.72 m^/litre, as compared with an observed average C^ of 2.8 ± 1.0 pCi/litre and the average Cm/Cy ratio of 0.34 + .14 shown above. Conclusions and Comments Analysis of the data at the two pasture locations indicates that the most significant mechanism in transferring radioiodine from the atmosphere to vegetation was via washout during rainfall. The appearance of detectable levels of radioiodine in milk was almost always preceded by a rainfall. Periods of dry weather with similar concentrations of radioiodine in air did not result in a measurable amount of ^ 4 -¡n milk. During the periods when the is ij activities in both vegetation and milk were both above the detection lim its, ratios of milk concen trations to vegetation concentrations were generally greater than the values reported by Hoffman [8]. The Dresden average of 0.34 is greater than Hoffman's "average" value and the Monticello average of 1.13, although more variable, is greater than his "high extreme" value. As evidenced by the low deposition velocities found in this study, iodine species play a significant part in the processes of trans fer of radioiodine from air to vegetation. The air sampling data 188 WEISS et al. will continue to be evaluated with regard to possible conclusions as to the species fractionation in the environment during this study. The meterological data obtained during the study are being analyzed and air concentration values will be calculated and compared with observed values. Preliminary calculations indicate that small releases, i_.e^, from the vents, may have a significant effect on locations close to the facility. Better measurements of these smaller releases should be made by the operators in order to adequately estimate their contribution to the airborne activity near the facility boundary. A basic conclusion from this study is that measurements of very small quantities of radioiodine in the environment can be accom- lished in the field during periods of normal nuclear facility operation. As a result of this study and other recent work improved field studies can be designed and more sensitive analytical measurements can be developed to obtain definitive information regarding radioiodine behaviour under varying environmental con ditions. Mork by various laboratories is now proceeding on the development of better environmental species differentiating air samplers and improved analytical methods that will remove and concentrate the radioiodine collected in sampling media. It is our belief that the answers to the numerous questions raised regarding the fate of low concentrations of radioiodine in the environment can be best developed by field studies which utilize the sensitive procedures currently available. 9. Acknowledgments Me would like to acknowledge the cooperation and participation of Commonwealth Edison Company; Northern States Power Company; Jersey Central Power and Light Company; Division of Radiological Health, Illinois Department of Public Health; Division of Environ mental Health, Minnesota Department of Public Health; and Bureau of Radiation Protection, New Jersey Department of Environmental Protection. Our appreciation is also expressed to the various additional laboratories which performed the various measurements: Health Services Laboratory (National Reactor Testing Station- NRTS); Oak Ridge National Laboratory; Battelle Pacific Northwest Laboratories; Allied Chemical Company (NRTS); Aerojet Nuclear Corporation (NRTS); and Naval Ordinance Laboratory. REFERENCES [1] Code of Federal Regulations, Title 10, Part 20, Section 20.1(c), Part 50 Sections 50.34(a) and 50.36(a), effective January 2, 1971. [2] Licensing of Production and Utilization Facilities, Federal Register 36 June 9, 1971, 11113. [3] U.S. Atomic Energy Commission, Final Environmental Statement Con cerning Proposed Rule Making: Numerical Guides for Design Objectives and Limiting Conditions for Operation to Meet the Criterion "As Low As Practicable" for Radioactive Material in Light-Mater-Cooled Nuclear Power Reactor Effluents, MASH-1258 (1973). IAEA-SM-180/44 189 [4] Pence, D. T. (Allied Chemical Corporation) personal communication with J. H. Keppler. [5] Kahn, B., et al^., Radiological Surveillance Studies at a Boiling Mater Nuclear Power Station, BRH/DER 70-1 (1970) 76. [6] Pelletier, C. A., Results of Independent Measurements of Radioactivity in Process Systems and Effluents at Boiling Mater Reactors, May, 1973, 78 pgs. + appendices, interim report available from Directorate of Regulatory Operations, USAEC. [7] Krieger, H. L. and Martin, A., Measurement of 13Ц Through the Air-Grass-Milk Pathway in the Vicinity of a Boiling Mater Reactor, Presented at Conference on Bioassay, Environmental and Analytical Chemistry, September 4 and 5, 1973, at Moran, Wy. [8] Hoffman, F. 0., Environmental Variables Involved with the Estimation of the Amount of isil ^ Milk and the Subsequent Dose to the Thyroid, IRS-M-6 (June 1973). [9] Pelletier, C. A. and Voillequá, P. G., The Behavior of i37Qs and Other Fallout Radionuclides on a Michigan Dairy Farm, Health Phvsics. 21(1971) 777-792. [10] Zimbrick, J. D. and Voillequë, P. G. (editors), Controlled Environ mental Radioiodine Test at the National Reactor Testing Station, Progress Report No. 4, USAEC, ID0-12065, January 1969. DISCUSSION J. S C H W I B A C H (Chairman): Could you give some information on the technical background as to why you discuss not only stack but also vent releases. In general, vent release is unknown in Europe. All gaseous effluents should be discharged via a high stack. B. H. WEISS: At the Dresden and Monticello plants, the turbine building and reactor building ventilation air is exhausted through vent stacks which rise about five to ten m e t r e s above the reactor building. Although the release rate for the short ventilation stack is m u c h smaller than the higher stack discharging condenser oif-gases, at the Dresden pasture the calculated dis persion factor, x/Q, was m u c h higher for the ventilation exhaust and this then became the m o r e significant source of iodine at the pasture location. F.O. H O F F M A N : I have found your results to be of considerable value. It is especially interesting to k n o w that low concentrations of in the environment can be monitored with enough sensitivity to test the general assessment calculations that have been proposed to estimate radiation burdens resulting from the normal operation of a nuclear power facility. H o w e v e r , I m u s t point out that it is necessary to analyse critically all the parameters involved in the assessment of the environmental transport of radionuclides before comparing experimental results with prediction models. For example, you have quoted m y predicted Cm/Cv ratios which are supposed to be representative for grazing conditions within the Federal Republic of Germany (Ref. [8] of your paper). These ratios, however, are in reality the predicted ratio of a concentration in a litre of milk to a concentration in a total ground surface area (m^) c o m p o s e d of soil covered by "edible" 190 WEISS et al. pasture forage. If you were to correct these ratios to represent the trans fer of 131l from vegetation to milk, then these figures would be: a "high" of 1.8, an "average" of 0. 65 and a "low" of 0. 28 (Ci/1 milk per Ci/m^ vegetation). I should also like to ask you about the methodology of your grass sampling. I believe that the most effective samples are those cut from only that surface area of ground that is completely covered with edible forage. This would be the surface area most important for a grazing animal. If you had patchy vegetation conditions, milk predictions made from g/m2 vegetation samples taken from an actual ground area composed of bare or trampled areas and edible forage could be subject to an error that could lead to an under-estimation. Did you consider this effect? В. H. WEISS: Yes. Areas which were to be cut for grass samples were selected adjacent to the fenced pasture so that w e could s a m p l e areas that were not disturbed by the cows. In addition, field personnel were instructed to cut vegetation which they felt was representative of the cows' feeding habits and to avoid significant patchy areas. H o w e v e r , in practice it w a s difficult to obtain several square metres of vegetation which met fully with our requirements. J. R. BEATTIE: I was very interested to hear you say that you cor relate the appearance of iodine in milk with rainfall. Should you not there fore be m e a s u r i n g the wash-out coefficient, Л s'l, rather than the deposition velocity, Vg. D o you have any information about the values of the wash-out coefficient which might be deduced from the observations in your paper? В. H. WEISS: W e were not able to obtain specific values for a wash-out coefficient but rather m a d e a general observation from the data obtained. There has been some previous work which indicates that methyl iodide is not washed out of the atmosphere as readily as elemental iodine. If this study is repeated, w e hope to investigate the effects of precipitation more closely. In view of the wash-out and dry deposition of radioiodine, an evaluation of its fractionation in the environment is very important. IAEA-SM-180/66 MEASUREMENT OF LOW CONCENTRATIONS OF RADIOACTIVE NOBLE GASES IN EFFLUENTS AND IN THE ATMOSPHERE A. BOTLINO, G. LEMBO, F. SCACCO, G. SCIOCCHETTI Servizio Física Sanitaria, C.S.N., Casaccia, Italy J. L A S A Institute of Nuclear Physics, Cracow, Poland Abstract MEASUREMENT OF LOW CONCENTRATIONS OF RADIOACTIVE NOBLE GASES IN EFFLUENTS AND IN THE ATMOSPHERE. 1. INTRODUCTION At present krypton-85, a radioactive noble gas produced in nuclear reactors, is the most abundant man-made radioisotope in the troposphere. Its concentration is, because of continuous growth of the nuclear industry, expected to increase further. This explains the great interest in developing and improving analytical methods and techniques of measuring the concentrations of radioactive gases, especially the noble gases. The measurement of atmospheric ^^Kr, because of its low concentration, necessitates enrichment of the sample and purification before liquid scintillator measurements can be made. In the past, separation of krypton was done using chemical and combustion purification processes. It is only recently that techniques of cryogenic adsorption and gas chromato graphic separation have been used. All the previously used methods for separating ^^Kr from other air constituents suffer from the disadvantage that numerous operations are necessary and that pre-enrichment of the s a m p l e s is needed. In this paper we present a new continuous method of separation based on a special thermodynamic gas chromatographic technique with air as the carrier gas. 2. PHYSICAL PRINCIPLES OF THE THERMODYNAMIC GAS CHROMATOGRAPHY METHOD The essential point of this method is that, in place of customary carrier gas, ambient air flows continuously through the column. No sample 191 192 BOTLINO et al. injector is needed. Heating the chromatographic column by means of a moveable heater causes desorption of mixture components along the column, and as a result the trace components are greatly enriched by continuous addition of substances. Every component of the gas mixture is separated and accumulates in bands. Each band moves together with the heater in its own zone of desorption temperatures. Under steady conditions, speeds of the bands and of the heater are equal. If the column is constantly supplied with air w e can obtain mixture com p o n e n t s in a cyclic w a y at its end and at larger concentrations. Because different mixture components are separated as a result of differences in their desorption heats, and these differences A Q are proportional to temperature differences A T on the heater axis, then the temperature gradient in the heater defines a distance between the components leaving the column. T h e given c o m p o n e n t m o v e s along the c o l u m n in this heater zone in which the temperature exeeds its desorption temperature. This so-called migration temperature depends on the kind of substance, the column packing, the rate of travel of the heater and the flow rate of the carrier gas. The characteristic migration temperature is expressed as: Q - _ w — r— and n = — R ln Ar) a where A is a constant, a is the linear flow rate of gas in the column, w is the speed of heater, Q is the heat of desorption, and R is the gas constant. As m a y be seen from the above equations, a simultaneous and pro portional change in the speed of heater motion w and gas flow rate a, does not influence the characteristic temperature. However, changing only w or æ causes a variation of the characteristic temperature of respective components, hence the effect of separation. A proper choice of analytical conditions allows a good separation of interesting component to be achieved. 3. DESCRIPTION OF MEASURING APPARATUS A diagram of the device for separating krypton from air is shown in Fig. 1. Air under examination is supplied continuously by a compressor 1. Air flows through a silica gel filter 2 and an analytical column, 3, around which the heater, 4, is moved. The movement is realized by means of a motor, 14. Separated gases flow through a cross-section type detector 13, and electromagnetic valves, 5. Electromagnetic valves, 5, are controlled by switches, 20,commutated by a lever joined with the heater, their opening time being programmed by two electronic timers. A cross-section type detector is used for calibrating the apparatus and its signal is measured by means of an electrometer, 11, and recorded, 12. T h e separated krypton is stored in a charcoal trap, 5, i m m e r s e d in liquid nitrogen, 7. The column is cooled by a thermally-connected dual column, 18, through which flows liquid nitrogen from a Dewar, 17, and pump, 16. A ventilator, 19, causes an auxiliary flow of cooling air around the column,,3. Temperature inside the heater, 4, is measured and controlled by means of an electronic regulator, 15. IAEA-SM-180/66 193 й FIG. 2. Cross-section type of detector. 194 BOTLINO et al. M*concef)fraf/onf%J FIG.3. Dependence of cross-section type of detector signal on krypton concentration. Curvea - for hydrogen as carrier gas; b — for air as carrier gas. W h e n a sufficient concentration of krypton is stored in the trap, 6, it is transported into the cell of an anticoincidence scintillation counter, 8, or alternatively, it is stored in a liquid scintillator vial. T h e scintillation counter is supplied from a H T supply, 9, and pulses are counted by a scaler, 10. The cross-section type of detector (shown in Fig. 2) has been constructed specially for this purpose and w o r k s with air as the carrier gas. Its sensitivity is not high but sufficient when the calibration mixture contains about 0.1% of krypton. The dependence of the signal of this detector on the krypton concentration with hydrogen or with air as the carrier gas is presented in Fig. 3. F r o m an observation of the detector signal, when the heater is at the end of the column, it is possible to fix lever positions and p r o g r a m m e timers for proper commutation of the switches steering the electromagnetic valves, 5. Hence all the gases leaving the column flow to air and only the interesting component, in this case krypton, flows to the trap, 6. 4. DESCRIPTION OF THE MEASURING APPARATUS A plan of the construction of the experimental device is presented in Fig. 4. It consists of a panel where a temperature regulator, manual ÍAEA-SM-180/66 195 FIG.4. Construction of the krypton separator. 196 BOTLINO et al. switches, "mater" of cycles and potentiometer for controlling the heater speed are assembled. Above the panel, between two heat insulator screens, 5 and 6, are fixed the cooling system, 4, and analytical column, 7 Inside the inner screen, 6, the drive mechanism, 3, the electric contact system, the transmission system with the electric motor and switches are assembled. Electromagnetic valves, the cross-section type of detector and charcoal trap are a s s e m b l e d on the shelf. 5. R E S U L T S Data have been collected during continuous cycles of heating and refrigerating of the column. The apparatus has been tested using various krypton and xenon concentrations above the detection limit of the cross- section type of detector. The chromatogram in Fig. 5 shows the separation of krypton from air with xenon as a reference in the krypton-xenon enriched mixture. The parameters of the experimental run can be summarized as follows (a) The time of one cycle for a good separation of krypton and xenon from air is about 6 min; (b) The carrier gas (air) flow was 200 cm^/min; (c) The maximum heat temperature was 400°C. F r o m our initial results it can be seen that bands can be collected in an auxiliary charcoal trap at the end of the column in an automatic fashion using electromagnetic valves switched by two programmed electronic t i m e r s . FIG. 5. General view of the krypton separator. tAEA-SM-180/66 197 6. CONCLUSIONS The apparatus described is now under examination and some technical problems remain to be solved. Actually, the device has been tested under various operating conditions and it has been possible to achieve good separation and enrichment of a krypton-xenon mixture from air. The first results confirm that the device can be operated as a special type of preparative apparatus to produce analytical s a m p l e s of pure krypton and xenon. A s a m p l e prepared in this m a n n e r can be counted in the internal counter in liquid scintillation vials at a higher level of sensitivity. REFEREN CES [1] ZUCHOWSKY.A.A., TURKELTAUB, N.M ., Gas Chromatography Moscow (1962). [2] LASA, J . , LESNIEWSKA, B ., Chemia Analityczna И (1968) 1247. [3] LASA, J..Nukleonika^2 3 (1967). IAEA-SM-180/4 AN ANALYTICAL STUDY OF THE ENVIRONMENTAL SURVEY METHODS IN USE AROUND NUCLEAR INSTALLATIONS IN THE COUNTRIES OF THE EUROPEAN COMMUNITY IN 1971 P. RECHT, J. SMEETS, R. AMAVIS Public Health Directorate, European Energy Community, Luxembourg A.A. C I G N A Comitato Nazionale per 1 "Energía Nucleare, Laboratorio Radioattivitä Ambiéntale, Rome, Italy Abstract AN ANALYTICAL STUDY OF THE ENVIRONMENTAL SURVEY METHODS IN USE AROUND NUCLEAR INSTALLATIONS IN THE COUNTRIES OF THE EUROPEAN COMMUNITY IN 1971. In 1971 the Public Health Directorate of the European Commission made an investigation in the six member states of the European Communities concerning the environmental survey methods in use around nuclear installations and nuclear power plants, in addition to research centres and others. The aim of this study was to establish (a) the state of the art in this field (the release of radioactive effluents in relation to the factual environmental monitoring programmes) from the qualitative and the quantitative points of view, and (b) the principles on which such programmes are based. The results of this study are to serve for mutual exchange of information in order to promote the rationalization of methods on the basis of the experience gained in the different countries. Besides such a rationalization, a harmonization of the monitoring programmes at the Community level could be considered in the light of the generally accepted Community norms for the protection of man and his environment. Depending on the type of installation and the site, information 1. INTRODUCTION 1.1. A i m of the inv e stigation In 1971 the Public Health Directorate of the European Commission made an investigation in the six m e m b e r states of the European Communities concerning the environmental survey methods in use around nuclear instal lations and nuclear power plants, in addition to research centres and others. T h e a i m of this study w a s to establish: (a) T h e state of the art in this field (the releases of radioactive effluents in relation to the factual environmental monitoring programme) from the qualitative and the quantitative points of view; and (b) The principles on which such programmes are based. 199 200 RECHT et al. TABLE I. PLANTS FROM WHICH DATA WERE RECEIVED Type С = Centre P = Power plant {R = Reprocessing plantsÍ BELGIUM С Centre d'Etude de l'Energie Nucléaire, Mol R EUROCHEM1C, Mol GERMANY, FED. REP. С Forschungsreaktor München С Gesellschaft für Kernforschung mbH, Karlsruhe С Kernforschungsanlage Jülich GmbH p Kernkraftwerk Lingen p Kernkraftwerk Obrigheim GmbH p Kernkraftwerk RWE-Bayernwerk GmbH, Gundremmingen p Kernkraftwerk Stade p Reaktorstation Geesthacht p Versuchsatomkraftwerk Kahl GmbH ITALY c Centro Studi Nucleari della Casaccia c Lab. Fabr. e Controllo Elementi di Combustibile, Saluggia c Reattore RTS1 Galileo G alilei, Pisa c Societa Ricerche Impianti Nucleari, Saluggia p Centro Elettronucleare del Garigliano p Centrale Elettronucleare di Latina p Centrale Elettronucleare di Trino R Impianto EUREX, Saluggia R Centro ricerche nucleari della Trisaia THE NETHERLANDS C Instituut voor Toepassing van Atoomenergie in de Landbouw, Wageningen C Interuniversitair Reaktor Instituut, Delft C Reaktor Centrum Nederland, Petten C Technische Hogeschool Eindhoven P ï ; I e Ni^gT*'^ Nederland, IAEA-SM-180/4 201 T h e results of this study are to serve for mutual exchange of information in order to promote the rationalization of methods on the basis of the experience gained in the different countries. Besides such a rationalization, a harmonization of the monitoring pro g r a m m e s at the C o m m u n i t y level could be considered in the light of the generally accepted Community norms for the protection of ma n and his environment. 1.2. Procedures A questionnaire was distributed through the national authorities. The questionnaire was divided into four sections. In the first, a general description of the site was requested, including s o m e information on the nature of the nuclear plant. The second part concerned the characteristics of any effluents (their nature, composition, total flow-rate and concentration). The third asked for a s u m m a r y of the conditions of release with reference to the rate of the discharge, the specific devices and procedures available for each case and the monitoring programme at the point of release. The last section concerned environmental monitoring programmes for the various kinds of s a m p l e s (water, air and foodstuffs). 1.3. Results An answer to the questionnaire was obtained from the nuclear plants listed in Table I. The data concern 12 research centres or laboratories, 10 nuclear power plants and 3 reprocessing plants. Therefore a general view of the actual trend in the environmental monitoring programmes, representative for the European Community, can be obtained. 2. CHARACTERISTICS OF THE EFFLUENTS AND OF THE RELEASES The characteristics of the effluents released by the nuclear plants con sidered here are summarized^ in Tables II, III and IV. (The identification numbers of the plants do not correspond with the order of the list in Table I. ) 2.1. Composition The liquid and gaseous releases of research centres and laboratories vary widely due to the different natures of the sources. In m a n y cases the radionuclides discharged into the e nvironment are not reported because their presence depends upon the research being carried out at any given time. The gaseous fractions consist mainly of argon-41, xenon-133 and -135 and the isotopes of iodine. The liquid releases from the nuclear power plants have a more uniform composition (in general, activation products) but also for these no detailed composition was reported in the questionnaires and only an indicative answer was obtained. The radionuclides in the gaseous releases from these plants are mainly argon-41 and hydrogen-3. * Tables H-X ate placed together on pages 204-223. 202 RECHT et al. The effluents discharged by the reprocessing plants have a more uniform composition: krypton-85 in the gaseous fraction, and activation and fission products in the liquid one. 2.2. Rates of discharge The volumes of the effluents discharged also range between rather wide limits, particularly for the research centres and laboratories. In some instances the rates of discharge are very high because the low-level effluents are diluted with water used for cooling purposes. T h e nuclear p o w e r plants s e e m to be rather similar f r o m this point of view, probably because the technological requirements are rather similar, also. Some differences can be observed for the discharge rates from the reprocessing plants. 2. 3. Total activity and concentration The activities discharged per month range over many orders of magnitude because of the nature and sizes of the different plants. The values have tended to be grouped for all the types of plant around values of som e tens of millicuries per month, with a min i m u m at some hundreds of picocuries per month and a m a x i m u m at one curie per month. These values do not include the contribution of hydrogen-3 which, in some instances, m a y reach some tens of curies per month. T h e concentration of radionuclides in the effluents released into the environment is also peculiar to each plant considered. In fact, besides the inherent nature and size of each plant, the characteristics of the discharge also play an important role in determining the concentrations of nuclides in the effluent. In s o m e cases dilution with cooling water occurs before release into the environment, this notably lowering the concentration values. T h e r e fore a comparison between the various plants cannot be made without taking into account details of the plants themselves in order to avoid meaningless comparisons. 2.4. Pro c e d u r e of the release Nearly all the plants considered here have a continuous release of the gaseous effluents into the atmosphere; the one exception is one of the nuclear power stations. In another case, only a fraction of the gaseous effluents is discharged discontinuously. In any case the effluents are filtered (absolute filter, charcoal filter, etc. ) and monitored before discharge. In contradistinction, the release of the liquid effluents is m a d e on a discontinuous basis, after storage or treatment of the wastes according to the particular requirements. W h e n available, the cooling water of the plant is us e d to further dilute the waste. 3. THE SURVEILLANCE PROGRAMMES The surveillance programmes are based on sample collection and measurements. Therefore such programmes can be of two kinds. One IAEA-SM-180/4 203 type of p r o g r a m m e concerns itself with the effluent s a m p l e d before dis charge (survey pr o g r a m m e s ), and the other with the control of radiocontaminants after discharge into the environment (environmental monitoring programmes). 3.1. Effluent monitoring programmes The main characteristics of these programmes are reported in Tables V to VII. In general the gaseous effluents are monitored continuously while the liquid ones are monitored using a sample representative of the whole batch. In fact, while gases and aerosols are discharged through a chimney and the samples are obtained via a by-pass, the liquid low-level wastes are stored in pools or containers which are emptied at intervals. Notwithstanding the great variations in the plants and situations, the procedures and sensitivities of the monitoring surveys show a rather limited spread. Such uniformity is due to the similarity of the instruments employed. In general a higher sensitivity is obtained when individual radionuclides are measured, either by g a m m a spectrometry or following a chemical separation. Gross alpha and beta measurements are still very frequently used to obtain an idea of the trend of the activity discharged. W h e n the measured values exceed a pre-determined threshold,specific and more sophisticated monitoring procedures are initiated. 3.2. Environmental monitoring programmes In. general environmental monitoring represents the m a i n effort in the field of radioprotection for operation of nuclear facilities. The information obtained from the questionnaire transmitted by the nuclear plants of the European Economic Community is reported in Tables VIII-X. A m o n g the 12 nuclear research centres, two have no environmental monitoring programme, probably because of the low discharge rates in comparison with the local environmental capacity. The other centres have a monitoring programme based, in general, on the following substances: air, water, soil, vegetables and milk. Sometimes fish are also sampled. The frequency of sample collection varies between continuous collection (air, water) and one sample per year (fodder, cereals). The frequency of measurements, which is daily for many air samples collected on a continuous basis, decreases to a few times per year in many cases. W h e n types of m e a s u r e m e n t are considered, it is seen that gross alpha and beta determinations are rather common. The evaluation of individual radionuclides is less c o m m o n and is generally used for samples collected a few times per year. 90gr^ i37Qg^ I3lj g^d Зц radionuclides more frequently determined. The same considerations apply to nuclear power plants and reprocessing plants. Nearly ail the plants of these types provide continuous monitoring of the air (filtered dust and aerosols), s o m e t i m e s including sampling points situated d o w n w i n d of the plant. G a m m a spectrometry is often used, in addition to the gross beta measurements, to identify specific radionuclides. The minimum detectable amount of radioactivity which can be determined in each case were not reported in the tables since it was very often absent from the answers given in the questionnaires. 204 TABLE II. CHARACTERISTICS OF THE EFFLUENTS AND THE RELEASES OF THE RESEARCH CENTRES Gaseous Effluents Liquid Effluents О s Release Rate Release Rate Concentr. s Radionuclide C=cont. Radionuclide C=cont. Volume Activity й D^discont. Ci/sec D=discont. m3/month mCi/month yUCi/m . 1 Several D 1000 20 3-100 2 - - - Several D 180 0.2 0.1-1 EH e aï. et RECHT 3 - - - Several D 2200-30000 450 200 4 - -- Several D 10 2-(-3) 0.02 5 A 41 С 3.5-(-4) Several D 300 6 20 N 16 С 1.3'(-12) 6 Several С UD 40 4-(-6)-4-(-7) (-4)-(-5) 7 A 41 с 2.4-(-4) Several D 33 3-(-3) 0.1 8 1125,131 с 5-(-5) P 32, 1 125, D 170 250 150 1 131,Hg 197 Hg203 9 - -- Cr 57,Zr 95, D 140 2 0.1-100 Nb 95,Ru 103, Sb 124,3b 125 10 A 41 4.7-(-3) Several ЗОООО 3 0.1 H 3 5.1-(-5) H 3 " 37 1200 N13,0 15 1.6-(-7) T A B L E II (cont. ) Gaseous Effluents Liquid Effluents 0 Release Rate Release Rate Concentr. s Radionuclide C=cont. Radionuclide C=cont. Volume Activity D=discont. Ci/sec D=discont. m3/month mCi/month yuCi/m^ 1 1.6-(-9) 11 A 41 C,D 1.4-(-5) Several D 28000 18 0.7 H 3 C,D 1.8-(-6) H 3 D" 190 7000 Kr 85 C,D 1.6-(-7) Xe 133 C,D 8.3'(-7) 4 / 0 8 1 - M S - A E A I 1 131 C,D 5-(-9) 12 Xe 133, 135 С 9-(-4) Several D 19ЗОО 710 37 1 131 С 5.1-(-6) Rn 222 С 3.7-(-7) 205 оM О) TABLE III. CHARACTERISTICS OF THE EFFLUENTS AND THE RELEASES OF THE NUCLEAR POWER PLANTS о Gaseous Effluents Liquid Effluents Rate Release Rate Release Concentr. Radionuclide C=cont. Radionuclide C=cont. Volume Activity <—tg Рч D=discont. Ci/sec D=discont. m /month mCi/month ^uCi/m^ 13 Several D 460 0.2 0.5 EH e al. et RECHT 14 Several С (-2) Several D 420 840 2000 15 A 41 С (-4) Several D 840 1260 1500 H 3 D " 8400 10,000 16 Kr85,Xe133,135 D (-5) Several D 750 580 780 17 Several С 3-(-6) Several D 420 84 200 18 Several С 1.6-(-4) Several D 4-(7)* 500 (-2)* 19 Several С 2.8 (-4) Several D 840 84 100 20 Several С Several D 30000 21 - - - Several D 840 840 1000 22 Several С (-4) Several D 180 6 32 H 3 D " 110 610 * Includes the cooling water (n) = 10" TABLE IV. CHARACTERISTICS OF THE EFFLUENTS AND THE RELEASES OF THE REPROCESSING PLANTS о Gaseous Effluents Liquid Effluents 5: j Rate Release Rate Release Concentr. Radionuclide C=cont. Radionuclide C=cont. Volume Activity 4 / 0 8 1 - M S - A E A I Plant D=discont. Ci/sec D=discont. ¡дЗ/month mCi/month ^,uCi/m^ 23 Kr 85 С 3-(-8) Several D 12500 13 1 24 Kr 85 С Very low Several D 2700 3 1 25 С Several D 2300 90 37 (n) = 207 208 TABLE V. EFFLUENT MONITORING PROGRAMMES IN RESEARCH CENTRES Caseous Effluents Liquid Effluents О E5 Emission m m Detector, Treatment -H -w 43 и Min.Detec t.Amount m Min.Detect.Amount á гЧ (pCi/1) w (pCi/1) Evap. Separated radionuclides; Min. Detsct. Amount d a (pCi/1) 3 1 beta G GC ; 3000 x 2 alpha G GC x EH e al. et RECHT beta G GC X gamma S GC X з alpha G ZnS; 100 X beta G GH; 100 X gamma S X 4 alpha G M ; 1 X beta G GC; 1 X 5 alpha G ZnS; 20 X Pu beta G GM,LS; 50 X Sr90:2 gamma s 100 S X Cs137:2 6 alpha G ZnS; 5 X 7 gamma G S 20 X 8 beta G G GM gamma G 1000 S 1125,131 T A B L E V (cont.) Gaseous Effluents Liquid Effluents О s Emission to Ю Detector, Treatment 'ft -<4 43 Ю Min.Detect.Amount Ю Min. Detect.Amount >3 g r4 (pCi/1) r4 (pCi/1) Evap. Separated radionuclides; Min.Detect.Amount a (6 d (pCi/1) 3 9 alpha G GC; 10 x beta G GC; 10 X gamma S X alpha G 50 10 IAEA-SM-180/4 beta G 100 gamma S 11 alpha G GC; 5 beta G GC,LS; 10 H3 : 1600 gamma G gamma S 12 alpha G X beta G X H3, C14, Sr90 С = Gross S = Spectrometry GC = Gas Counter GH = Geiger-Mueller LS = Liquid scintillation counter SS = Solid state detector ZnS = ZnS scintillation counter 209 PS = Plastic scintillator 210 TABLE VI. EFFLUENT MONITORING PROGRAMMES IN NUCLEAR POWER PLANTS Gaseous Effluents Liquid Effluents о 3 Treatment Emission Detector, -n 0) Min.De t ec t.Amount и Min.Detec.Amount Evap. Separated radionuclides;Min.Detect.Amount g >3t f—t i—i3 !—Й (pCi/l) 6 (pCi/l) (pCi/l) РЧ 3 3 13 alpha G GC; 100 beta G с ы ; ю о Sr 90 EH et l. ta e RECHT gamma S 14 beta Sr 89,90 : 10 gamma С G gamma s S 1000 15 alpha G SS; 100 x beta G PS ; 100 X H3 : 2000; P 32 : 30; Ca45 : 3; Sr90 : 3 gamma S 16 beta H3 : 3000; gamma s 10 S Activation products: 0.1 - 0.01 17 gamma G G 1000 gamma SS 100 18 Beta G GC gamma GG 19 gamma G G T A B L E VI (cont. ) Gaseous Effluents Liquid Effluents О 3 Treatment Emission m m Detector, +3 g >3 Hin.Detec t. Amount и Min.Detec.Amount Evap. Separated radionuclides ; Min.Detect.Amount f—t3 r4 (pCi/1) гЧ (pCi/1) (pCi/1) PL) Й 3 3 20 alpha GG beta GG 21 beta G GC x 22 alpha G beta G 1 x gamma G 1000 G 5000 X gamma S See footnote Table 4 212 TABLE VII. EFFLUENT MONITORING PROGRAMMES IN REPROCESSING PLANTS Gaseous Effluents Liquid Effluents !C Emission Ю M Detector, Treatment to Min.De tec t.Amount m Min.Detect.Amount Evap. Separated radionuclids3;Min. Detect. Amount s гЧ (pCi/1) гЧ (pCi/1) (pCi/l) Й 3 23 alpha G beta G EH e at. et RECHT gamma S G 1000 gamma S 24 alpha G ZnS; 5 x beta G GM; 50 x gamma S 100 (1 131) G 100 gamma S 25 alpha G G X beta G G X H3, CS14, Sr90 gamma S X See footnote Table 4 T A B L E VIII. E N V I R O N M E N T A L M O N I T O R I N G P R O G R A M M E S I N R E S E A R C H C E N T R E S Sampling Analysis Frequency Determinations Frequency ) 3 & Material Sampled m о rn O' о others 5 'ф m Plant Plant No. No. of No. points of b m Cross alpha Cross beta Gross per per year per per week I per month continuous per year per per day per per day per week month per Ю о M ж 1 2 Surface water 3 1 X 1 Drinjking water 1 2 XX 2 Sediment 2 1 X 1 3 Seaweeds 3 2 X X 2 Mussels 3 2 X X 2 Shrimps and 1 4-6 XX 4-6 fishes Soil 1 4-6 XX 4-6 4 - 5 Air 3 XX 1 " 3 XX 1 Sediment 2 1 X X Ru 106 1 " 2 4 X X Ru 106 4 T A B L E VIII (cont. ) Sampling Analysis Frequency Determinations Frequency ë ,3 &CO -R Material Sampled f- Д оСГ- <**1 others g о*3* b Plant Plant No. No. of No. points of b ta m Gross alpha Gross beta Gross per month per per year continuous per day j week per Ю w с M ж per day weekper p< per year -g Fodder and hay 3 1 XX Ru 106 1 Л Milk 3 2 X X 2 " 3 2 X X 4 Vegetables 3 4 X X X Ru 106 4 Cereals 3 1 X X Ru 106 1 6 Air 1 X XX 1 " 1 X X 1 Surface mater 5 X XX 1 " 4 1 X X 1 " 1 1 X X 1 Milk 7 1 X XXXX 1 Fish 2 1 X 1 " 2 1 X X 4 7 Air 2 XX 1 Fallout 1 X x: X 1 Surface water 1 1 X X 1 Watertable 1 1 X X 1 Sediment 3 4 X X 4 T A B L E VIII (cont. ) Sampling Analysis Frequency Determinations Frequency . 3 3 & я >3 Material Sampled о f - (6 S о others 3 Ь ь Plant Plant No No. of No. points of m m Ф Gross alpha Gross beta Gross Sr 90 per per year continuous per per day per week P. per year о M ж A per week per month 8 Air 1 X X 1 Surface water (same as No.6) Fish 1 4 XX 4 9 Surface water 4 1 X X alpha and 1 beta spectu Watertable 6 1 X X " 1 10 Air 1 X X X 1 " 2 X X X 2 - Fallout 3 X 2 X X X 2 " 3 X 1 Pu 1 " 4 X 1 XX 1 Surface water 1 X X X X 1 Drinking water 1 1 X X X X 1 " 4 4 X X X X 4 " 1 1 x; X XX 1 Sediment, 1 1 XX 1 plankton " 2 4 XX 4 T A B L E VIII (cont. ) Sampling Analysis Frequency Determinations Frequency Material Sampled r'lt- о others "3 * Plant Plant No. No. of No. points of to Cross alpha Cross Gross beta Gross Gamma spectr. Sr 90 continuous per year per per day week per month per per per month w о M ж per day weekper per year Fish,aquatic 1 2 XX 2 8 plânts **" Vegetables 3 2 X X X 2 " 7 4 Pu 4 11 Soil 11 2 XXX XX Gross gamma 2 Vegetables 11 1-2 XXXXX " t-2 Milk 3 4 XXXX " A 12 Air 6 XXX 1 " 5 XXXX 1 " 5 X XX Sr 89 1 Fallout 5 X XX 1 " 5 XX X Sr 89 1 Surface water 6 2 XXX Ra 226 2 Fodder 3 1 XX 1 Milk 6 1 X 1 6 1 XX 1 . I. MONITORING IN POWER S T N A L P R E W O P R A E L C U N N I S E M M A R G O R P G N I R O T I N O M L A T N E M N O R I V N E IX. E L B A T 14 13 Plant No. Milk Air Vegetables water Drinking Milk Grass Air Sediment Fish Grass Material Sampled Material Surfacewater Surface water Surface i g lin p m a S Frequency S a m p lin g A n a l y s i s F r e q u e n c y Determ inations F r e q u e n c y .3 b +3 tú M aterial Sampled я <¡) E^ >3 О § C\ rn ь ь b m o t h e r s p. p< и 15 A i r Gross gamma S e a w a t e r Drinking mater Fish, m olluscs V egetables S e d i m e n t S a n d A i r Surface water Drinking water G r a s s M i l k C e r e a l s V e g e t a b l e s T A B L E IX (cont. ) Sampling Analysis Frequency Determinations Frequency A r-t & Material Sampled СЙ Ю 6 .IS06 CO r n Ю о Г-1 others No. of No. points of b m continuous per per day weekper Gross beta Gross per month per per year о 1 131 К 40 К per per day per weekper monthper per per year Air 5 XX 1 Surface water 23 1 XX 1 Soil 10 6 X 6 Vegetables 10 6 X 6 Milk 4 1 X X XXX 1 Air 2 X X X X 1 Fallout 2 X XX X 4 Drinking water 2 4 XX X 4 Soil 3 2 X X 2 Vegetables 7 1 XX X 1 Milk 1 1 XX X 1 Sediments 2 1 X X X 1 Aquatic plants 2 1 XX X 1 Fish 1 1 X XX 1 T A B L E IX (cont. ) Sampling Analysis Frequency Determinations Frequency Material Sampled others 19 Air Fallout Surface water Sediment Sr 89 Plankton,aquatic Sr 89 plants Fish Soil Vegetables Milk S r 89 Cereals 20 Surface water 21 Air Surface water Soil,vegetables T A B L E IX (cont. ) Sampling Analysis Frequency Determinations Frequency Material Sampled 22 Air Fallout Surface water Watertable Sediment Soil Seaweedj^plankton Milk Bees and honpy M M M TABLE X. ENVIRONMENTAL MONITORING PROGRAMMES IN REPROCESSING PLANTS Sampling Analysis ------' Frequency Determinations Frequency -- я 8 43 & 0) s Material sampled -is06 <*4Г- Ф о о others & a 5 b ь Plant Plant No. m Ф Ф No. of No. points of Gross Gross alpha beta Gross per per day monthper per per week per year ^3 per per day о M ж P* A per years ! ! ' continuous ) EH e al. et RECHT 23 Air 1 X X X 1 " 1 X X 1 Surface water 4 1 X X 1 " 4 1 X X 4 Watertable 6 X 1 X 1 " 6 X 1 XX X 1 " 1 1 X X 1 " 1 1 X 1 Drinking water 1 X XX X 1 Sediment 3 4 X X 4 3 4 X alpha 1 spectr. Soil 3 4 X X 4 " 3 4 X X 1 T A B L E X (cont. ) Sampling Analysis m Frequency Determinations Frequency о -WЙ я Й о й p< о§ +3 gА 8 ю 3 ь гЧ Ф А áf ь +3 tu § & ф й фй й .О 03 Ф G й Material sampled о -H *3 & § >) 03 03 о г-гп Й Ф § +s 03 03 CTt о others тЗ & § >3 о О ф ф ф ь ь 03 ф ф ф e¡ О A р. А А ё С! ю Ц о м ж А А А А Milk 4 1 X XX X 1 ё Vegetables 1 1 XX 1 3 1 1 X 4 Fish 2 1 XX 4 24 Air 8 X X 1 Surface water 3 1 X X 1 1 2 X X 2 1 1 X X 1 Soil 2 1 X X 1 Vegetables 2 1 X X 1 Milk 2 2 X X 2 25 Air 1 t X X ' X 1 Surface water 2 X Milk 1 1 X X 1 ьэ M о? 224 RECHT et al. 4. CONCLUDING REMARKS A general view of the situation for a number of the nuclear facilities of the European Community can easily be obtained by studying the data reported in the Tables 11-X. In 1965, ICRP Publication No. 7 defined a new policy for the establishment of surveillance programmes. The previous procedure of collecting as much data as possible around a nuclear plant was to change into a more Specific programme of monitoring particular points of the environment. It must be said that the picture obtained after this investigation, pro moted by the Public Health Directorate, seem s to confirm that the general situation is rather far from that suggested by the ICRP. Some nuclear research facilities have no environmental monitoring programme, or their programme is reduced to a simple monitoring of the surface water downstream the plant itself. Nevertheless compliance with the law is ensured because their low-level releases do not exceed the lim its of discharge fixed by the local authorities. Other plants have an environmental monitoring programme reflecting somewhat more the need for studying the behaviour of radionuclides in the environment. But too often the samples are merely analysed for their gross beta activity. Gamma-ray spectrometry, which is a powerful tool in radioecological studies, is more widely used by the industrial plants such as nuclear power and reprocessing plants. In order to support research on the ecocycles of radionuclides and to gain a better knowledge of the problems concerning environmental capacity, it seem s pertinent to aim at having more detailed programmes for environ mental monitoring, particularly with regard to the type of analysis. Although for some facilities there is no real necessity for such pro grammes on account of the low quantities of radionuclides released in com parison with the environmental capacity, it must be remembered that the task of research centres and laboratories should also include research on the characteristics of the environment. A suitable set of programmes based on this argument is fundamental to keeping in the future the high standard of safety for which the nuclear industry is noted. ACKNOWLEDGEMENTS The authors are very grateful to those who have collaborated in providing the data used in this report. IAEA-SM-180/38 PERMANENT GAS MEASUREMENTS AS PART OF AN ENVIRONMENTAL SURVEILLANCE PROGRAMME* J.M . MATUSZEK, C.J. PAPERIELLO, C.O. KUNZ, J.A. HUTCHINSON. J.C. DALY Division of Laboratories and Research, New York State Department of Health, Albany, N.Y., United States of America Abstract PERMANENT GAS MEASUREMENTS AS PART OF AN ENVIRONMENTAL SURVEILLANCE PROGRAMME. In addition to conventional monitoring of the krypton and xenon radioactivity released from nuclear facilities, the Radiological Sciences Laboratory of the New York State Department of Health has been conducting a study of the several permanent gases which are also released. The gaseous effluents from a boiling-water reactor (BWR), two pressurized light-water reactors (PWR), a high-temperature gas-cooled reactor (HTGR) and a pressurized heavy-water research reactor (PHWR) have been analysed for radioactive and stable gas constituents. In the case of the BWR, direct stack sampling has ensured representative sampling. The PWRs, HTGR and PHWR do not provide for direct stack sampling, so analysis of hold-up tank gas, cover gas, primary-coolant strip-gas, and containment air was required to derive characteristic patterns of gas releases from these reactors. Gamma-emitting gaseous species are measured spectrometrically using a Ge(Li) detector. Chromatographic separation of various gas fractions on a series of molecular sieve columns permits the use of internal gas-proportional counting tubes for spectrometric resolution of beta-emitting gas species. In addition to measurement of gamma-emitting noble gases, ^H (as hydrogen gas and methane), **C (as methane and carbon dioxide), ^Ar and ^Ar have been quantitatively identified. Measurements of radio active high hydrocarbons, carbon monoxide and methyl iodide in the gaseous efñuents are in progress. The Laboratory's experience in collection, handling and analysis of samples is discussed. Estimates of total annual release of the several species are provided. Relative release rates are compared to control requirements. 1 . INTRODUCTION As the result of a recent agreement between the U. S. Atomic Energy Com mission and the State of New York, a variety of reactor effluent samples have become available for detailed analysis. The Radiological Sciences Laboratory has obtained gas samples from a boiling-water reactor (BWR), two pressurized light-water reactors (PWR), one high-temperature gas-cooled reactor (HTGR) and a pressurized heavy-water research reactor (PHWR) in order to determine the impact of long-lived noble gases, tritium , 14c arte! other radionuclides nor mally not measured by routine reactor-monitoring techniques. The original goals of the program included a concerted effort to study the gaseous effluents from a nuclear fuels reprocessing plant. Unfortunately, fuel reprocessing gases were not available due to a temporary shutdown of the only operating commercial fuel reprocessing plant in the U. S. Two charcoal cartridges were obtained for analyses of iodine radioactivity. The facility operating staffs have been very helpful in providing samples and information concerning plant operations. * Supported in part by USAEC contract AT (11-1) 2222 and by USEPA contracts 68-01-0522 and 68-01-LA -0505. 225 2 26 MATUSZEK et al. 2 . EXPERIMENTAL PROGRAM 2.1 Sample Collection Gas samples are collected in a 16-liter stainless steel vessel (least active gaseous radionuclide <10*6 ^Ci/cm3), l-liter stainless steel vessels (<10'4 p.Ci/cm3), 125-cn)3 and 30-cm3 glass gas-collection bulbs (>10'^ nCi/cm3) and 125-cm3 or 14-cm3 septum bottles (>10'3 p.Ci/cm3). To permit measurement of even minor constituents, the largest possible sample is collected, limited only by the activity level that the facility health physics staff will permit off- site. Collection in the rubber-capped septum vials is avoided as much as pos sible because leakage often causes large sample losses. 2.2 Sample Processing and Counting Immediately upon receipt at the Laboratory, an aliquot of the sample is counted on a Ge(Li) spectrometry system for identification and measurement of gamma-emitting gaseous radionuclides. Other aliquots are then processed by the procedures outlined in Fig. 1 Two systems have been constructed to minimize the possibility of cross-contam ination when sample activities differ by four orders of magnitude. The inter- m ediate-activity rack is used for samples containing gases with activities H IGH-AC TIVITY6AS INTERMEOIATE-ACTIVITYGAS FÎG.1. Processing procedure for high-activity and intermediate-activity gas samples. IAEA-SM-180/38 227 < 10*6 p,Ci/cm? In order to avoid contamination of low-activity fractions, only those sample aliquots with total activity < 10*^ ^Ci are injected into the in- termediate-activity gas chromatograph. As a result, different gases from any given sample aliquot may require processing on different racks. For example, a relatively fresh BWR sample may be very high in 133xe or 85^ but have more than seven orders of magnitude less ^HH or ^HCHg. This sample would be pro cessed for xenon and krypton on the high-activity gas chromatograph, and the interm ediate-activity radionuclides would be reprocessed on the intermediate- activity gas chromatograph after initial separation on the high-activity rack. As another example, after a few liters of an aged holdup-tank sample have been bled down, the total krypton and xenon activity is so high that contamination of the interm ediate-activity gas chromatograph may occur. The high activity aliquots would therefore be transferred to the high-activity gas chromatograph, while the H2, CH^ an d CO2 fractions would be processed on the intermediate- activity gas chromatograph. A sample is injected into the appropriate rack, the volume is measured, and carriers are added. A rough separation from air constituents is made on molecular sieve 5A, which also collects the sample for injection into the gas chromatograph with a helium sweep. The sample is passed through a 2 0 f t - l o n g column of molecular sieve 5A, and separate aliquots are collected at liquid nitrogen temperature in U-tubes filled with molecular sieve 5A. The separated aliquots are driven off the molecular sieve traps into storage bulbs for re tention and subsequent transfer into the counting tubes. Low-activity fractions from the high-activity rack are repurified before counting by making a second separation on the interm ediate-activity chromatograph. Decontamination factors between fractions are 3 x 1 0 ^ f o r S 5 ](r in t o > 1 0 ^ for other combinations. The recovered volume of each aliquot is measured on a sample loading rack just prior to transfer into the internal gas-proportional counting tubes. Activity levels range from > 1 0 ^ counts/min-cm^ of original sample, in which case diluted fractions are taken for counting, to as low as 1 0 * 3 counts/mincm 3 in some of the lowest-activity fractions. Low-activity aliquots are counted in gas-proportional tubes mounted with plastic scintillator anticoincidence guards. A low-activity rack for analysis of environmental gas samples (activity l e v e l s < 10" i 0 p.Ci/cm3) is nearing completion. Detailed descriptions of the separation method and counting techniques have been accepted for publication [ 1 ,2 ] . 3. RESULTS AND DISCUSSION 3.1 Activity Levels Sample activities are initially calculated in nCi/cm3 but are then reported as ratios of one of the noble gases present. In Table I all values are normalized on 85кг. The values in Table I for the PWRs and HTGR are weighted averages of the values for the several types of samples collected. All ratios have been corrected for decay from time of sample collection to time of release. Note that the range of activities can be as much as 10^ between different radionuclides in a particular sample. The "less than" values (<) in Table I indicate activity levels below the limit of detection of the system. The detection limit is the level at which the error at the 95% confidence level equals the numerical value obtained. The CO2 ratios for the PWR samples are listed as minimum values (>) because the standard collection procedure at the PWRs for cover-gas, strip-gas 228 MATUSZEK et al. TABLE I. Average activity ratios (nuclide/^Kr) of long-lived permanent gas reactor effluents at time of release. Reactor 3?Ar 39 Ar 83кг i31""xe 133m^g 133xe Знн Знсн^ BWR 8.5(-2)3- l . K - 4 ) 1.0 9 9.5(1) 2.1(3) 8 (- 3 ) < 2 (-5 ) < 5 (- 3 ) 1 .2 (- 2 ) PWR(I) 3 .K - 3 ) < 4 (- 6 ) 1 .0 4 ( - l ) 1 3.2(1) 7(-4) <3(-4) 3(-3) > 2 (- 4 ) PWR(II)- 3.H-3) < l ( - 6 ) 1 .0 6 < 7 (- 6 ) <2(-5) 1.7(-4) > 4 (-5 ) PWR(II)& 8 (- 2 ) 6 (- 5 ) 1 .0 8(-2) l.K-2) 5 5 (- 4 ) < 5 (- 4 ) K - 2 ) < > 6 (-5 ) HTGR 4 .9 ( 1 ) 4 (- 6 ) 1 .0 < 3 (- 2 ) > 1 (- 1 ) 4 ( - l ) < 9 (-4 ) 9 (-4 ) PHWR 1 .9 (4 ) 3 ( - l ) 1.0 1.4(4) 2.2(4) 2 .8 1 .4 8 .6 a In this and all subsequent tables, the number in parenthesis is the power of 10 to which preceding numbers is raised . For example, 8.5(-2) equals 0.085. b Old core. and decay tank samples involves contact with water. The 14С0^ values for con tainment air, which is sampled directly, were too low to obtain real values, but maximum levels produced 14C02/85Kr ratios which are consistent with those for other PWR samples. The I^CH^/S^Kr ratio in PWR(I) containment air is similar to ratios for cover and strip gas. The containment air ratio for the new core in PWR(II) is approximately 10 times greater than the ratio for decay tank samples. Though containment air release contributes a small fraction of the total gaseous ac tivity released, the high l^CH^/S^Kr ratio in the PWR(II) containment air makes the average activity ratio in Table I significantly higher for that re a c to r. The Зцн v alu es fo r the HTGR are a lso thought to be minimum v a lu e s. In the HTGR the helium primary coolant is purified by passage through two charcoal beds, one at and one slightly above liquid nitrogen temperature. Charcoal under these conditions has poor ^HH retention, so a scaling factor of at least 10 may be required for representative values of the ^HH released from the HTGR. 3.2 Annual Release Levels To estimate the annual release of each permanent gas species, it is only necessary to multiply its activity ratio by the reported release (Table II) of the noble gas on which the ratios are normalized. These estimates are given in Table III. Since the PHWR operator does not measure any noble gases, those releases were calculated from pressure vessel volume and the concentration values obtained by the Radiological Sciences Laboratory. The PWR(II) was one of the reactors which experienced cladding cracks due to fuel densification [3]. Installation of new fuel elements apparently de creased the amounts of 85[(г, 133xe and other noble gases released by more than IAEA-SM-180/38 229 TABLE II. Reported annual releases of long-lived noble gases. releases (Cl) Reactor Year Total S5Kr 133xe BWR 1970 9 . 5 ( 3 ) [1538 MW(th)] 1971 2 . 5 (5 ) 4.3(4) 1972 5 . 2 ( 5 ) 6.5(4) PWR(1) 1970 1.7(3) [615 MW(th)] 1971 3.6(2) 1972 5.4(2) PWR(II) 19703. 1.0(4) [1300 MW(th)] 19713 3.2(4) 19723. 1.2(4) 1.2(3) 1.0(4) 1973& 5.6(2) 9(1) 4.2(2) HTGR 1970 5.7 [H6MW(th)] 1971 1.2(2) 1972 5.8(1) PHWR 1970 ^ [40MW(thP 1971 ^ NA- 1972 J an order of magnitude; the decrease approached a factor of 60 when compared to 1971, the highest year. The amounts of permanent gases apparently produced by activation outside the fuel elements do not appear to be appreciably reduced, however. The highest single-year releases of ^H, and *^Ar are listed in Table IV. The and ^C values are the sum of the ^HH and ^HCHg or and 14CÛ2 respectively. Although the HTGR operator has reported a maximum annual release of only 14 Ci/yr of as tritiated water in liquid effluents, the total *^H released may be well over 60 Ci/yr, a value more consistent with the 700 Ci/yr available for release from an HTGR of this size [41. Projections are provided in Table V of permanent gas releases for large reactors of similar thermal power presently under construction or on order. Because of the difficulties in insuring against l^CQg losses during sample collection, total gaseous 14c releases from a PWR may exceed the value in the table. Releases from a large HTGR could exceed 1,300 Ci/yr of ^HH and could approach 2 x 10^ Ci/yr of З^Аг if holdup times are less than 120 days. 3.3 Boundary Dose Levels Doses at a boundary (1 km from the reactor) were also estimated on the basis of effluent projections for new reactors. These were used to fa cilitate direct comparison of dose without the complication of differing power le v e ls . TABLE III. Estimated annual release of long-lived permanent gas. Estimated releases (Ci) 131m^e 133m^ 3 Reactor Year 3?Ar ^ K r HH ЗнСИд ^CH^ *-^C02 BWR 1970 6(-2) 8(-5) 7 (-l) 6 7.0(1) 1.5(3) 6(-3) <1.4(-5) <4(-5) 8(-3) 1971 2 2(-3) 2.0(1) 1.7(2) 1.9(3) 4.3(4) 1.6(-1) <4(-4) <1.0(-3) 2 (-l) 1972 3 3(-3) 3.1(1) 2.6(2) 2.9(3) 6.5(4) 3 (-D <6(-4) <1.6(-3) 4 (- l) PWR(I) 1970 1.7(-1) <2(-4) 5.3(1) 2.0(1) 5.0(1) 1.7(3) 4(-2) <1.6(-2) 1.6(-1) > l.l ( - 2 ) 1971 4(-2) <5(-5) 1.1(1) 4 2.0(1) 3.5(2) 8(-3) <3(-3) 3(-2) >2(-3) 1972 5(-2) <7(-5) 1.7(1) 7 3.5(1) 5.4(2) 1.2(-2) <5(-3) 5(-2) >4(-3) PWR(II) 19703. 4 l-2(-3) 1.2(3) 7.7(3) <8(-3) <2(-2) 2 (-l) >5(-2) 19713. 1.2(1) <4(-3) 3.7(3) 2.3(4) <3(-2) <7(-2) 6 (-l) >1.5(-1) 1972^ 5 <1.4(-3) 1.4(3) 9.0(3) HTGR 1970 2.9(2) 2(-5) 5.7 < 2 (-l) > 6 (-l) 2 <5(-3) 5(-3) 1971 6.0(3) 5(-4) 1.22(2) <4 >1.2(1) 5.0(1) <1.1(-1) 1.И-1) 1972 2.9(3) 2(-4) 5.8(1) <2 >6 2.0(1) <5(-2) 5(-2) PWR 1970^1 1971 ^ 1.9(-1) 3(-6) 9(-6) l.O (-l) 2(-l) 3(-5) 1.3(-4) 8(-5) 1972 J IAEA-SM-lSO/38 231 TABLE IV. Highest single-year releases of permanent gas radioactivity. Зн 14c 3?Ar Reactor Year [MW(th)3 (C i) (C i' (C i) BWR 1971 1538 2 ( - l ) 2 .4 ( - l ) 2 PWR(I) 1970 615 < 6 (- 2 ) > 1 .8 ( - 1 ) 1 .7 ( - 1 ) PW R(II)^ 1971 1300 < 1 (- D >8(-l) 1.2(1) PWR(II)^ 1973 1300 < 1 (- 1 ) 1 7 HTGR 1971 116 > 6 (1 ) < 2 ( - l ) 6 .0 (3 ) PHWR 40 2 ( - l ) 2 (- 4 ) 1 . 9 ( - i ) TABLE V. Projected annual release and boundary dose (1 km) of permanent gas radioactivity from new reactors. Reactor Power level T o ta l r e le a s e (Ci/vr) Boundary dose fmra d /v r) type [MW(th)] M e 3?Ar Зн i " c 3?Ar BWR 2760 4 ( - l ) 4 ( - l ) 4 4(-3) 6(-2) 4(-3) PWR3- 2440 2 ( - l ) >2 1 .3 (1 ) 2 (- 3 ) > 4 ( - l ) K - 2 ) HTGR 2500 >1.3(3) <5 1.3(5)^ 1.2(2) <1 1 .4 (2 ) ^ PHWR 2600 1 .3 (1 ) 1 .3 ( - 2 ) 1 .2 (1 ) K - l ) 2 (- 3 ) K - 2 ) Projected from PWR(II), new core, b For short holdup, 37др could approach 2(6) Ci/yr and 2(3) mrad/yr. The submersion dose for 3*?Ar was calculated using USAEC regulatory guides [5,6]. The 6,000 Ci of 3*7Ar released from the HTGR represents 20% of the total submersion dose. Holdup times less than the 120 days used at this HTGR could result in 37 being dose limiting, approaching 3 times that from S5Kr. Dose estimates for 14c require consideration of local exchange and uptake as well as submersion dose. Pasquill diffusion graphs С 5,6] were used to calculate concentrations at 1 km. Dose estimates were then made by direct comparison to doses from atmospheric 14c concentrations [72. Similar calculations were performed for Зн. The Pasquill graphs provided concen tration estimates. Cosmic ray production rates from Lai and Peters [8] were used with UNSCEAR values [7] to make the dose calculations. 232 MATUSZEK et al. The boundary dose of effluents not normally measured by the reactor operators will apparently be 1 mrad/yr or less, with the exception of ^A r and Зн from the HTGR. The source of 37дг in the gas effluents is suspected to be the re action 40Са(п,ч)37дг, which may take place on calcium impurities in some of the reactor materials, especially in the carbon moderator used in the HTGR [9]. Future reactor design appears oriented toward the development of breeder re actors, either liquid-metal cooled (LMFBR) or gas-cooled (GCBR). Since 37дг is most likely produced outside the fuel elements, emissions from the LMFBR or GCBR are difficult to predict. If calcium impurity concentrations are high, 37Ar releases may be as large as those from the HTGR, again producing a sig nificant immersion dose. Boundary dose from l^C in LMFBR and GCBR gas effluents should be low as well. The breeder reactors should present less nitrogen or carbon for ac tivation to 14c. Gaseous Зн, however, may be produced in the LMFBR at twice the HTGR production rate, and its retention by fuel cladding is expected to be poorer than in light-water reactors [4]. Unless specific control steps are taken, the size of the breeders and their siting density may create sig nificant dose levels, particularly in more heavily populated areas. 3.4 Adequacy of Utility Dose Estimates The question of total boundary dose requires consideration of a number of other factors which have come to light during this study. The oper ating staff at each reactor collects gaseous samples for gamma-spectrometric analysis by filling rubber-capped septum-vials through a hypodermic needle. The resulting samples leak at varying rates, some losing as much as half their activity in an hour or so. As a result, the fission noble-gas releases (on which dose estimates are based) may actually be higher than those reported. The adequacy of iodine release estimates is also suspect. The USAEC has found that a major portion of the radioiodine released from BWRs is organic iodine [10] . Organic radioiodine in the PWRs may also be abundant. Charcoal cartridges used to collect radioiodine at the reactors are of different types, ranging from standard respirator cartridges to 1-mm-thick charcoal-impregnated filters. Collection times vary from a few minutes to one week per filter. Retention efficiency, especially for organic iodine,can be expected to be very poor in some cases. The gas separation system at the Radiological Sciences Laboratory is therefore being modified to permit measurement of all iodine species in gas samples without the perturbing influence of varying or unknown charcoal retention efficiency. 4. SUMMARY Studies of gaseous reactor effluents indicate that fission noble gases released by water-cooled reactors account for most of the dose at the site boundary. Зц and 37 releases may produce significant boundary doses around HTGRs. The otherwise unmeasured permanent gases released by reactors produce boundary dose levels which are relatively low when compared to ICRP [11] and previous USAEC [12] recommendations but which can be significant in comparison to more recent USAEC requirements [13]. Dose levels for radionuclides such as 14C and Зн, which may exchange with the local environment, can be significant even though the amounts released annually may be small in comparison to total natural global production [14]. The local impact may be compounded if local persistence of and is high [15]. IAEA-SM-180/38 233 REFERENCES [1] KUNZ, С .О ., Separation techniques for reactor-produced noble gases, Noble Gases Symposium, US Environmental Protection Agency, Las Vegas, 1973 (to be published). [2] PAPERIELLO, C .J., Internal gas-proportional beta spectrometry for measurement of radioactive noble gases in reactor effluents, Noble Gases Symposium, US Environmental Protection Agency, Las Vegas, 1973 (to be published). [31 GILLETTE, R., Nuclear Safety: Damaged fuel ignites a new debate in AEC, Science 177 (1973) 330. [4] KOUTS, H., LONG, J., "Tritium production in nuclear reactors", Tritium (MOGHISSI, A.A., CARTER, M.W ., Eds), Messenger Graphics, Phoenix (1973) 38-45. [5] UNITED STATES ATOMIC ENERGY COMMISSION REGULATORY GUIDE 1.3, Assumptions Used for Evaluating the Potential Radiological Consequences of a Loss of Coolant Accident for Boiling Water Reactors, USAEC, Washington, D.C. (1973). [6] UNITED STATES ATOMIC ENERGY COMMISSION REGULATORY GUIDE 1.4, Assumptions Used for Reactors, USAEC, Washington. D.C. (1973). [7] REPORT OF THE UNITED NATIONS SCIENTIFIC COMMITTEE ON THE EFFECTS OF ATOMIC RADIATION; Ionizing Radiation: Levels and Effects, V ol.l: Levels, UN, New York (1972) 29. [8] LAL, D., PETERS, B., Cosmic ray produced radioactivity on the earth, Handb.Phys. 46 2 (1967) 551. [9] MATUSZEK, J.M ., PAPERIELLO, C .J., KUNZ, C.O., Reactor contributions to atmospheric noble gas radioactivity levels, Noble Gases Symposium, US Environmental Protection Agency, Las Vegas, 1973, (to be published). [10] PELLETIER, C .A ., Results of Independent Measurements of Radioactivity in Process Systems and Effluents at Boiling Water Reactors , unpublished USAEC Rep. (May, 1973). [11] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION (ICRP). Report of Committee II on Permissible Dose for Internal Radiation (1959), Pergamon Press, London (1959). [12] UNITED STATES ATOMIC ENERGY COMMISSION RULES AND REGULATIONS: Title 10 Code of Federal Regulations, Part 20, Standards for Protection Against Radiation, USAEC, Washington, D.C. (1970). [13] UNITED STATES ATOMIC ENERGY COMMISSION RULES AND REGULATIONS : Title 10 Code of Federal Regulations Part 50, Appendix I, Numerical Guides for Design Objectives and Limiting Conditions for Operation to Meet the Criterion 'As Low as Practicable* for Radioactive Material in Light-Water Cooled Nuclear Power Reactor Effluents, USAEC, Washington, D .C. (11 June 1971). [14] JACOBS, D .G ., Sources of Tritium and Its Behavior upon Release to the Environment, USAEC Rep. TBD-24635, available from US Department of Commerce, Springfield, Va. (1968). [15] KORANDA, J.J., Preliminary studies of the persistence of tritium and ^C in the Pacific Proving Ground, Health Phys. 11 (1965) 1445. DISCUSSION J. SCHWIBACH (Chairman): You have provided figures for releases of tritium as hydrogen. Do you also measure the release of tritium as water vapour in the gaseous effluents? J.M . MATUSZEK: Since the facility operators continuously monitor the amount of tritiated water vapour released, we have not performed many НТО analyses on gas samples. Water vapour is not a well-behaved gas in the sample containers. Total sample transfer with added heating is required to inject water vapour quantitatively into the separation system. Since this is inconvenient, we accept the facility operators' published НТО values. D.M. MONTGOMERY: Can your gas separation technique differentiate between ^C present in chemical form s other than СЩ or COg? J.M . MATUSZEK: Yes. We are currently performing analyses for CgHg and CgHg. The levels are 1% to 10% of the СЩ values. IAEA-SM-180/78 СБОР ПРОБ И ОПРЕДЕЛЕНИЕ ТРИТИЯ В ПРИЗЕМНОМ СЛОЕ ВОЗДУХА Л.И.ГЕДЕОНОВ, В.А.БЛИНОВ, A. В.СТЕПАНОВ,В.М.ГАВРИЛОВ, B.П.ТИШКОВ, А.М.МАКСИМОВА Радиевый институт им.В.Г.Хлопина, Л енинград, Союз Советских Социалистических Республик Abstract- Аннотация COLLECTION OF SAMPLES AND DETERMINATION OF TRITIUM IN THE LAYER OF AIR NEAR THE EARTH'S SURFACE. The development of atomic power requires, along with other types of radiological monitoring, tritium surveillance both in the vicinity of nuclear facilities and at considerable distances from them. At the V.G. Khlopin Radium Institute a system has been developed for sampling atmospheric moisture or unoxidized tritium, converting it to a counter gas for proportional counter measurements and for measuring tritium samples in concentrations of & l(f tritium units (TU). For measurements in the concentration range of l(f to 10^ TU, a proportional counter has been built with setting adjustment based on X-radiation from ^Fe. For the deter mination of higher concentrations, a liquid-scintillator type of counter suitable for work under field conditions has been proposed. СБОР ПРОБ И ОПРЕДЕЛЕНИЕ ТРИТИЯ В ПРИЗЕМНОМ СЛОЕ ВОЗДУХА. Развитие атомной энергетики привело к созданию дополнительного источника загрязнения биосферы тритием. По данным различных авто ров количество образующегося трития при работе реакторов составляет Ки О Т 15 Д О 3 0 — — ;----г------М В т(эл )го д В связи с этим возникает задача определения трития в контурных водах реакторов, сбросных водах и в приземном слое воздуха. Предпринятая работа ставила своей целью создание лабораторной аппаратуры, предназначенной для измерения содержания трития в пробах воды в пределах 10^-10^ ТЕ. О Т Б О Р П РО Б Как известно, тритий в атмосфере находится преимущественно в окисленной форме в виде НТО. Доля трития в молекулярной форме невелика и составляет значительную величину лишь в верхних слоях 235 236 ГЕДЕОНОВ и др. атмосферы. Улавливание паров воды из атмосферы осуществляется нами путем адсорбции на предварительно обезвоженном силикагеле, имеющем высокоразвитую гидрофильную поверхность. Конструкция пробоотборника представляет собой алюминиевый патрон-насадку с двумя отверстиями, вмещающими 2 кг гранулированного силикагеля. Воздух просасывается через поглотитель со скоростью 0,5 м^/мин. Продолжительность отбора определяется температурой и влажностью воздуха и выбирается такой, чтобы силикагель поглотил приблизитель но 200 г воды. По окончанию отбора силикагель, содержащий атмос ферную влагу, поступает в установку для извлечения воды и регенерации поглотителя. В том случае, если предполагается длительное хранение пробы, силикагель помещается в герметичную упаковку. Извлечение воды из силикагеля, насыщенного атмосферной влагой, и его регенерация производится в динамических условиях потоком горячего воздуха, циркулирующего в замкнутой системе. Воздух, на гретый в трубчатой электрической печи до 170-180°,поступает в алюминиевый патрон, содержащий влажный силикагель, и затем проходит через трубчатый холодильник производительностью 1C* ккал/ч. При этом воздух охлаждается до 35-40°С и теряет большую часть паров воды, конденсирующихся на стенках холодильника и сборника конденсата. За ходом процесса регенерации следят по изменению температуры воз духа на выходе из патрона с силикагелем . Вследствие большой тепло ты испарения воды температура воздуха, выходящего из поглотителя, не превышает 100°С и остается постоянной в течение всего времени десорбции воды из зерен силикагеля. По окончании процесса темпера тура воздуха резко поднимается и становится равной температуре на входе в патрон с поглотителем. В тех случаях,когда в анализируемом воздухе содержится заметное количество трития в неокисленной форме, производится предварительное окисление до воды. Окисление водорода производится в токе азота на гранулированной окиси меди, нагретой до 600°С. Образовавшиеся пары воды собираются в U-образной трубке, опущенной в жидкий азот. ИЗМЕРЕНИЕ АКТИВНОСТИ ПРОПОРЦИОНАЛЬНЫМ СЧЕТЧИКОМ Была выбрана традиционная цилиндрическая конструкция счетчика, в котором катодом служит цилиндр из нержавеющей стали с полирован ными внутри стенками, а анодом - вольфрамовая проволока диаметром 20 мкм (рис.1). Боковые стенки,пробки и корректирующие трубки изготовлены из плексигласа. Рабочий объем счетчика 2480 см^ . Электрическая емкость 30 пФ. На одной из боковых крышек имеется бериллиевое окно, свободно пропускающее Х-лучи s^Fe (Е = 5,9 кэВ). Пропорциональный счетчик помещается внутри кольцевой защиты из гейгеровских счетчиков МС*6,включенных с рабочим счетчиком в схему антисовпадений. Все счетчики заключены в пассивную защиту из свинца толщиной 50 мм . Предельные газообразные углеводороды — метан; этан, пропан, бутан обладают хорошими счетными свойствами, даже если к ним добавляется до 10% водорода. Преимуществом исполь зования углеводородов является то обстоятельство, что в процессе их синтеза из непредельных соединений и водорода, полученного из изме ряемого образца, в одну молекулу счетного газа может быть включено IAEA-SM-180/78 237 более чем два атома меченного тритием водорода,что позволяет увеличить концентрацию трития в рабочем объеме счетчика при неизмен ном давлении. Кроме того,ионизационные потери космического фона при использовании пропана или бутана при давлении больше 0,5 ат лежат далеко за пределами верхнего тритиевого энергетического ка нала. Поэтому в качестве счетного газа используется пропан с до бавкой до 10% меченного тритием водорода,а при измерении малоак тивных образцов — бутан, синтезированный из бутадиена. Процесс получения счетного газа складывается из нескольких стадий (рис.2). Первая заключается в разложении воды и получении эквивалентного количества молекулярного водорода по реакции: Mg (порошок) + HgO ^ * M gO + Hg . 238 ГЕДЕОНОВ и др. Следующая стадия — синтез бутана путем каталитического гидри рования бутадиена водородом,полученным из анализируемой пробы воды. Реакция гидрирования осуществляется на палладиевом ката лизаторе АК-62. Для получения бутана смесь бутадиена и водорода в стехиометрическом соотношении 1 : 2 пропускается через металличес кую трубку, заполненную слоем катализатора высотой 4-5 см. Полу ченный бутан вымораживается жидким азотом, а затем идет на за полнение счетчика. Отрицательные импульсы на выходе пропорционального счетчика в результате регистрации мягкого бета-излучения трития имеют ампли туду в пределах нескольких милливольт. Наряду с ними, в результате регистрации фонового излучения присутствуют импульсы, имеющие широкий диапазон амплитуд. Поэтому система усиления и регистрации должна удовлетворять следующим требованиям: 1. Устойчивостью к амплитудным перегрузкам; 2. Низким уровнем собственного шума; 3. Коэффициентом усиления 1000; 4. Стабильностью усиления и динамическим диапазоном, обеспечивающим линейность в области энергий бета-частиц тр и ти я . Усилительный тракт, удовлетворяющий этим требованиям, состоит из предусилителя, смонтированного на крышке пропорционального счетчика, и основного усилителя. С целью защиты от амплитудных перегрузок использованы дифференциальные усилительные каскады с гальванической связью(рис.З). имп мерительной установки. и к в о н а т с у й о н ь л е т и р е зм и а м е х с - к о л Б . З . с и Р IAEA-SM-180/78 КАНАЛАN 239 240 ГЕДЕОНОВ и др. Для отбора импульсов, соответствующих бета-частицам трития, производится дискриминация по верхнему и нижнему уровням . Нижний соответствует энергии бета-частиц 0,5 кэВ, верхний — 20 кэВ. Схема отбора антисовпадений имеет три входа: на два из них поступают сигналы от дискриминаторов нижнего и верхнего уровней, на третий — от активной защиты. Схема выдает выходной сигнал лишь в том случае, если при поступлении сигнала с дискриминатора нижнего уровня отсутствуют сигналы на двух других входах. Для настройки и контроля системы используются генератор, осциллограф и анализатор импульсов. Процесс настройки и проверки состоит из наблюдения и регистрации спектра Х*лучей железа-55 на выходе усилительного тракта и установки порогов дискриминации (рис.4). Регистрация импульсов на выходе системы отбора антисовпадений осуществляется пересчетным прибором ПП-15, с блоком печати БАП-2. Система в целом имеет следующие характеристики. Разрешающая способность по энергии,определенная с помощью Х-лучей железа-55, составляет 19%. Спектр бета-частиц трития, снятый с помощью ана лизатора АИ-128-3, хорошо совпадает с рассчитанным спектром и спектрами, приводимыми в литературе. Интегральный фон, снятый без активной защиты, составляет 289 ± 6 имп/мин. Фон в энергетическом канале трития — 16,0 ± 1,2 имп/мин. Чувствительность счетчика на рабочей смеси пропан + водород — 2,4-103 ТЕ, а на синтезированном бу тан е — 1 ' 10^ Т Е . СЦИНТИЛЛЯЦИОННЫЙ МЕТОД Измерение активности воды, содержащей тритий, с помощью жидкого сцинтиллятора обладает рядом достоинств. К ним относится простота приготовления образца, возможность автоматизации измерений и из готовления небольшой установки для использования в полевых условиях. С целью создания прототипа такого прибора был сконструирован и изго товлен одноканальный сцинтилляционный счетчик. Конструирование варианта для полевых работ заставило нас отказаться от некоторых схемных и конструктивных решений, позволяющих улучшить параметры счетчика, но делающих конструкцию громоздкой и тяжелой. Так, счетчик используется без активной и пассивной защиты, не применяется диф ференциальная дискриминация. Датчик счетчика установлен в холодиль нике и работает при температуре +10°С. Основой датчика является фотоумножитель ФЭУ-81. Для согласования высокого выходного сопротивления фотоумножителя со входом последующего блока БД -9 служит эмиттерный повторитель. Камера, в которой установлен ба рабан, позволяет производить установку и извлечение кювет без за светки фотоумножителя и подводить к фотокатоду ФЭУ поочередно одну из шести кювет. Выходной сигнал с усилителя дискриминатора БД-9 поступает на пересчетный прибор ПП-15 и линейный интенсиметр БИ-9 с регистри рующим потенциометром ЭПП-09. В качестве жидкого сцинтиллятора используется промышленный сцинтиллятор ЖС-8. Кювета имеет полезный объем 18 мл. Наимень ший регистрируемый уровень активности составляет 1 ' 10^ Ки/л. IAEA-SM-180/78 241 DISCUSSION J. SCHWIBACH (Chairman): What is the main application of the tritium m easuring techniques you describe — is it for nuclear power plant monitoring or environmental surveys? L. I. GEDEONOV: The less sensitive (scintillation) part of the system is designed for fairly high tritium concentrations, e. g. for measuring tritium in analytical control sam ples from nuclear power station cooling circuits and samples from protection and ventilation system s, and discharge pipes. The more sensitive part of the system, based on the use of a proportional counter, is for monitoring the tritium content in the air of working prem ises and in various environmental objects around nuclear power stations. Pro vided that the tritium is sufficiently concentrated in the sam ples, the system may be used for carrying out hydrological and meteorological surveys. P. PELLERIN: I should just like to point out that tritium, while not being terribly dangerous, is a 'diabolical' element for laboratories to deal with. It spreads everywhere and passes easily from a contaminated bottle to other bottles nearby which were not originally contaminated. One of the first rules, if you watit to make accurate measurements, is to keep the amounts of tritium in the laboratory low, and that is not always easy. J. M. MATUSZEK: We have experienced this problem with the plastic vials used for liquid scintillation counting. At laboratories making routine liquid scintillation measurements, appreciable losses of НТО can occur if sam ples are counted a day or more after preparation. IAEA-SM-180/41 MONITORING LOW-LEVEL RADIOACTIVE AQUEOUS DISCHARGES FROM A NUCLEAR POWER STATION IN A SEAWATER ENVIRONMENT D .M . MONTGOMERY, H .L. KRIEGER. В. KAHN US Environmental Protection Agency, National Environmental Research Center, Cincinnati, Ohio, United States of America Abstract MONITORING LOW-LEVEL RADIOACTIVE AQUEOUS DISCHARGES FROM A NUCLEAR POWER STATION IN A SEAWATER ENVIRONMENT. Methods were developed and tested for measuring ^Mn, ^Co, s°Co, ^Sr, ^°Sr, ^ 1. M4cs and '" C s in relatively large volumes of sea-water. These isotopes were found by analysis of low-level radioactive wastes to be the most radiologically significant among those discharged into sea-water from a BWR nuclear power station. The results of laboratory tracer experiments and field studies are presented. From a variety of concentration media that were considered for samples of 16 to 400 litres, the following system was selected: 1. Cartridge filter (prefilter and 0.45-^m membrane) operated at a flow rate of 15 1/hiin for collecting radio nuclides in suspended material; 2. Ion exchange system (300 cm^ of chelating resin for collecting ^Mn, ^Co and ^°Co followed by 200 cm^ of an inorganic exchanger for ^ C s and *^Cs) operated at a flow rate of 12 1/h; 3. Precipitation of Agi and then SrCOg from a volume of 16 litres. The filter and ion exchange sections were counted by gamma-ray spectrometry with a Ge(Li) detector plus 2000-channel analyser. The detection limit was about 0.01 pCi/1 for a 400 litre sample. The Agi and SrCOg precipitates were chemically purified and determined by beta-particle counting. For samples of 16 litres or most radionuclides in a 16 litre sample was 0.1 pCi/1. Data are presented regarding the distribution between particulate and dissolved fractions for these radionuclides, as well as some others (^Cr, ^Fe, 9 5^ ^Nb, ^Mo, ^B a, ^*Ce and ***Ce) that were found in the waste or in fall-out. INTRODUCTION The need for measuring radionuclides in the aquatic environment, both freshwater and seawater, has been well established and accepted. It is useful to differentiate between monitoring for control purposes and mea suring for radioecological investigations [1]. Control monitoring usually requires samples of 1 liter or less to ensure that maximum permissible concentrations have not been exceeded. Monitoring for radioecological studies may require special low level detection methods and samples of 100 liters or more in order to attain the necessary sensitivity. The discharge of low level radioactive waste by nuclear power stations into the aquatic environment sometimes results in concentrations much lower than would be detected by normal control monitoring even at the point of discharge [2,3]. Observations of radionuclides from low level wastes after dilution in the aqueous environment are important in determining the concentration of radionuclides by various forms of marine life and identi fying the critical radionuclides and pathways leading to population radiation exposure. Periodic measurement in the environment also serves to detect any long-term accumulation of environmental radioactivity. 243 244 MONTGOMERY et al. The general approach to radionuclide measurements in seawater has been concentration by coprecipitation [4-6], absorption [7,8], ion exchange [9, 10], or solvent extraction [4]. After concentration, the sample can be counted directly by gamma-ray spectrometry or treated radiochemically to isolate the various radionuclides and then counted by gamma-ray spectrom etry or beta-particle detectors. Methods for determining specific radionuclides in seawater ate referenced in two recent publications [1,11]. Other methods are available for sequential analysis of a mixture of radio nuclides from 1 to 200 liters of seawater [12-15]. A recent method by Silker et al^. [8] utilizes a sampler to concentrate particulate and dissolved radionuclides by filtration and adsorption from several thousand liters of seawater. The present work is part of a study initiated in 1971 at the Oyster Creek Nuclear Generating Station to provide information on the radiological impact of the station on the environment. The station is located near the Atlantic Coast, 3.2 km inland from Barnegat Bay, and discharges low-level radioactive liquid effluent into its once-through cooling system of circulating water from Barnegat Bay. Methods were developed and tested for determining radionuclides from low-level radioactive wastes after dilution in seawater. The salinity in the cooling water was between 16 and 25 °/oo and the pH, between 7.2 and 8.0. Because salinity and pH exhibited season al variation, all procedures were designed for salinity as high as 34 °/oo and a pH range from 7.0 to 8 .0 . In order to identify and quantify the radionuclides to be concentrated from canal water, samples itom the liquid waste tank were analyzed. Manganese-54, ^Co, 60(^ °°Sr, ^Sr, 131i, 134(^ 137(-g were found to be the most radiologically significant among the complex mixture. Based on the measured concentrations and reported dilution factors of 10,000 to 170,000, the collection system was designed for samples from 16 to 400 l i t e r s . In itial experiments on waste tank solutions indicated that many of the radionuclides are discharged as particles that can be collected by filtration even after dilution. The collection system was designed to retain particulate radionuclides by filtration through a 0.45-pm membrane filter. The ionic or filterable radionuclides of Mn, ^Co, 60co, ^ Cs, and 137cs were concentrated on an ion-exchange column suggested by Boni [16] consisting of the chelating ion-exchange resin Chelex-100 * and the inorganic ion exchanger ammonium hexacyanocobalt fe r r a te (NCFC). R adioiodine and radiostrontium were determined by precipitating Agi and SrCO^ from a 16-liter sample. The concentration system was developed and tested by: 1. Laboratory tracer studies to determine column dimensions and operating conditions for concentrating radionuclides by ion exchange; and condi tions and yields for determining radiostrontium and radioiodine by precipitation. 2. Field tracer studies by addition of reactor wastes to seawater samples to verify collection efficiencies by ion exchange and precipitation. The system was then used in the field for monitoring radionuclides in environmental samples during discharge of radioactive liquid wastes. * Mention of commercial products does not constitute endorsement by the US Environmental Protection A gency. IAEA-SM-180/41 245 LABORATORY TRACER STUDIES Chelex-lOO has been reported [10] to be quantitative for collecting manganese and cobalt from seawater at pH 7.6 and 9.0, respectively. Their retention on Chelex-100 between pH 7.0 and 8.0 was checked by spiking synthetic seawater with 5^Mn and ^Co and passing the samples through Chelex-100 columns. The results on synthetic seawater (shown in Table I) indicate > 98 percent retention of %Mn and 60Co on Chelex-100 and > 99 percent recovery of 134cg on NCFC. To demonstrate the feasibility of the ion exchange system for concen trating Mn, Co, and Cs from larger volumes. 15 liters of synthetic seawater spiked with 54цп, 57co, 85gp^ and 134cg were passed through a column of 30 cn)3 C helex-100, 30 cm^ NCFC on s i l i c a g e l, and 80 cm3 Dowex-1 x 8 (50-100 mesh, Cl" cycle). Table II shows that > 95 percent of 34цп, ^^Co, and 134cg were retained on the Chelex-100 and NCFC. Strontium-85 passed through the column, as expected. The 56 percent recovery of ^*^1 onthe Dowex-1 anion exchange resin suggested that quantitative recovery of 131i from seawater would require unreasonably large resin columns. To confirm the tracer studies with synthetic seawater, 8 li^ rs of Oyster Creek discharge canal water were spiked with 5^Mn, 57co^ and 134cg. The sample was passed through a column consisting of 8 cm3 of Chelex-100 and 4 cm3 of NCFC on silica gel. The results are given in Table III. Recoveries were > 97 percent for 5^Mn, 57^Q^ and 134cg^ but 0 percent for ^ S r . The field ion exchange column (for 400 liter seawater samples), con sisting of 300 cm3 of Chelex-100 resin and 200 cm3 of NCFC on silica gel, was designed on the basis of these retention studies. These dimensions are consistent with previous studies [10,17], which show the capacity of the field ion exchange column for Mn, Co, and Cs in seawater should be in excess of 600 liters. The Chelex-100 is also reported [10] to be effective for collecting other trace elements from seawater. Among those expected to be completely retained are Bi, Cd, Cu, In, Mo, Fd, Sc, Th, Y, and Zn. Table I Retention of Co and Mn on Chelex-100 and Cs on NCFC ______from Synthetic Seawater______ ______X Retained______Chelex-100b NCFCс pH of feed solution^_____^C o *^Mn______^*^Cs 7.0 99.4 98.0 — 7.5 99.7 99.2 99.9 8.0 99.8 99.7 — ^Feed solution was 100 ml of synthetic seawater spiked with 60co, -^Mh, and 134cg^ salinity 34°/oo ^8cm^of 50-100 mesh Chelex-100, Ca***2 cycle (1.2 cm diam. x 7.1 cm-height), flow rate: 5 ml/cm^-min. ^4 cm3of NCFC on 20-80 mesh silica gel (1.2-cm-diam. x 4.0-cm-height), flow rate: 5 ml/cm -min. 246 MONTGOMERY et al. Table I I Recovery of Radionuclides from 15 liters ______o f Syn th etic Seawater______ Recovery (X) Radionuclide Chelex-100 NCFC Dowex-1 Column E fflu en t ^Mn 101 + 5 < 0 .1 < 0 .1 < 0.1 5?Co 96 + 5 < 0.1 < 0.1 < 0.1 S^Sr < 0.1 < 0.4 < 0.1 100 + 1 131^ 8.0 + 0.4 < 0.1 56 + 3 36 + 3 ^ c s < 0.6 105 + 5 < 0.6 < 0.6 Column dimensions: Chelex-100 (50-100 mesh, Ca^ cycle) and NCFC beds, 2.5 cm diam. x 6.5 cm height, Dowex-1 resin bed, 3.1 cm diam. x 10.3 cm height. Feed solution: pH 7.6, volume 15 liters, salinity 34°/oo,flow rate 5 ml/cm^-min. Table I I I Recovery of Radionuclides from 8 liters of Oyster Creek Discharge Canal Water Recovery (X) Radionuclide Chelex-100 NCFC E fflu e n t ^Sln 99+ 5 < 0.1 < 0.1 5?Co 97+ 5 < 0.1 < 0.1 S 's r < 0 .1 < 0.1 100 4 ' c s < 0.1 100 + 5 < 0.4 Column dim ensions: Chelex-100 (50-100 mesh, Ca*^ cycle) 1.2 cm diam. x 7.2 cm height; NCFC, 1.2 cm diam. x 3.5 height. Feed solution: 8 liters of Oyster Creek discharge canal water; pH 7.6, salinity 20°/oo, flow rate 5 ml/cm^-min. METHODS The field measurements were as follows: 1. An aqueous sample of 200-400 liters was filtered through cartridge filters at a flow rate of 15 liters/min. 2. The filtrate was passed through an ion exchange column at a flow rate of 12 liters/hour. IAEA-SM-180/41 247 3. 16 liters of the column effluent or the filtrate were analyzed by precipitating iodine and strontium. 4. Samples were measured for radionuclide content by gamma-ray spectrom etry or beta-particle counting. F ilt r a t io n E arly f ie ld measurements u tiliz e d an 8-pm and a 0.45-pm membrane filter disc (142-mm diameter Millipore filters) in series. The flow rate was limited to approximately 1 liter/min and the prefilter and filter had to be changed every 4 to 8 liters, respectively. Filtration of 200 to 400 liters was tedious and required several hours. Membrane f i l t e r c a r tr id g e s were found to be more convenient fo r la rg e volumes. The filter cartridges^ were 67 mm in diameter by 233 пип in length with an effective filter area of 4600 cm^. The cartridge consists of an outer cellulose paper prefilter and an inner 0.45-pm membrane filter. The cartridge filter is either on a plastic core, which collapsed under 3-4 atm pressure, or on a stainless steel core, which withstood pressures to 7 atm. The samples were filtered by pumping water from the coolant canal through the cartridge filter with a self-priming centrifugal pump (Homelite XLS 1-1/2, total head 40 m at 42 liters/min with a 9-m suction lift). The outlet of the pump was connected by a 3/4-in-diam. plastic garden hose to a 3/8-in-diam. disc-type water meter with a hand valve for controlling the flow rate through the filter. The outlet from the water meter was connected to the cartridge housing by a short section of 3/4-in-diam. plastic garden hose. The maximum recommended flow r a te i s 19 lite r s /m in . The t o t a l volume that could be filtered through a cartridge filter varied from 70 liters for seawater with a solid content of 46 mg/liter to 380 liters at a solid content of 9 mg/liter. After filtering the samples, the cartridges were dried at 60°C for 24-48 hours. The end caps were removed w ith a power saw and the f i l t e r s - separated from the stainless steel core. The prefilter and filter were compressed with a hydraulic press under 200 atm pressure into a routinely- used 450-cm^ counting container. Three cartridges could be compressed into this container. Ion Exchange System The ion exchange system consisted of a 7-cm-diam. plexiglas tube filled with 300 cm-3 of 50-100 mesh Chelex-100 in the Ca+2 cycle and 200 cnP of NCFC coated on 20-80 mesh s i l i c a g e l. The two se c tio n s were separated by a perforated plastic disc. The inorganic ion exchanger, NCFC on silica gel, was prepared by the procedure of Terada et, al^ [17]. The filtrate was passed through the ion exchange column by gravity flow or pumped from a reservoir through the column with a tubing pump. Flow rates were limited to 12 liters/hour because of the relatively slow exchange rate of the chelating resin [10]. The resin sections were then separated, transferred to containers, and thoroughly mixed to ensure a homogeneous matrix for gamma-ray counting. 2 Gelman Instrument Co. Acroflow # 12505.1. 248 MONTGOMERY et al. Precipitation Iodine was determined by precipitation from a 16-liter sample. The procedure is a modification of a method for the determination of stable iodine in a 500-ml seawater sample [18]. Immediately after collecting the sample, 40 mg of iodine carrier as K10ß were added. The sample was acidified to pH 1 with concentrated HCI and the IOß reduced to I* by adding hydrazine sulfate. The sample was then taken to the laboratory for further treatment. The radioiodine was precipitated as Agl-AgCl by adding silver citrate and stirred vigorously for 2 hours. The mixed silver halide precipitate was allowed to settle overnight and collected by decanting and centrifuging. The iodine (I ) was oxidized with B^-ï^O to iodate (IOß), and then reduced to I 2 with hydroxylamine hydrochloride. It was extracted as I2 into toluene, back-extracted as I" into H2O con tain in g NagSOß, and finally precipitated as Pd^'^O for gravimetric yield determination. Chemical yields were 55 + 15 percent. For ^ Sr and 9^Sr determination, the supernate from the mixed silver halide precipitation was neutralized with NH^OH and buffered with NH^Cl; SrCOß was then precipitated by adding Na2C0ß [13]. The SrCOß, containing large amounts of CaCOß, was dried and then purified by repeated treatment with fuming HNOß and scavenging with Fe(0H)ß and BaCrO^. The sample was precipitated as SrCOß, weighed for chemical yield determination and counted immediately for ^Sr plus °^Sr content. After 2 weeks for ingrowth of ^^Y, the sample was dissolved and yttrium carrier was added. The Y was separated as yttrium oxalate and counted for Y to compute the Sr con tent. The chemical yield was determined by comparing the final Sr yield with the initial stable Sr concentration in the seawater sample ("8 mg/liter for seawater with salinity of 34 °/oo) as determined by atomic absorption spectrophotometry. The yield of SrCOß can also be determined by atomic absorption spectrophotometry. When the Sr yield was determined by weighing SrCOß, the SrCOß was checked to confirm that all CaCOß had been removed. A se q u e n tia l a n a ly sis fo r -^Mn, 58co, 60co, 1 3 4 ( ^ and ^ ^ C s may be included with the Sr and I procedure by adding 20 mg each of Co, Cs, and Mn carriers to the sample. After precipitating Agi, Cs is separated by addition of ammonium phosphomolybdate (AMP) [1]. The AMP is dissolved and cesium is chemically purified and precipitated as cesium chloroplatinate for chemical yield determination and counting [19]. The supernate is treated with KOH to coprecipitate Co and Mn as hydroxides [4] and separated as MnÛ2 and CoS. The Mn02 and CoS can be p u rifie d by ro u tin e methods and finally precipitated in suitable forms for gravimetric yield determination and counting [20]. Counting The cartridge filters and ion exchange sections were counted on a 60-cm^ Ge(Li) detector with a 2048-channel analyzer. The Ge(Li) detector was calibrated with U.S. NBS certified solutions. For various counting arrangements, the lowest photon efficiency was 0.21 percent (for the 1.332 MeV ^Co gamma-ray in 450 cm^). After chemical purification, ^Sr, ^Sr, and ^^1 were determined by beta counting in a low-background (1 count/min) gas flow counter. The beta-particle counting efficiency was approximately 30 percent. FIELD TRACER STUDIES Because the physico-chemical properties of the radionuclides from the reactor wastes in seawater were not known, the concentration system was IAEA-SM-180/41 249 tested in the field to verify collection efficiencies for radionuclides as they exist in the environment. In two tracer experiments, 400 to 1000 ml of waste tank solution were mixed with 190 to 208 liters of coolant water in a 220-liter plastic-lined drum. The solutions were circulated with a pump for 1 hour to simulate conditions in the discharge canal, and then passed through the collection system. In the first experiment, the filtrate from the cartridge filters was also filtered through a pair of Millipore filter discs (8 pm followed by 0.45 pm) to check the efficiency of the cartridges. On both occasions, a sample of liquid from the waste tank was analyzed to determine the concentration of radionuclides added to the water samples. Because radionuclides were found to deposit on the walls of the polyethyl ene sample bottle even when the sample was acidified to 10 percent with conc. HNOß, the bottle itself had to be analyzed. As an alternative, the samples were filtered at the time of collection to minimize these losses to container walls, and the filter and filtrate were then analyzed. The full scale testing of the concentration system with reactor wastes demonstrated the validity of the techniques for the radionuclides of interest. The observed distribution of radionuclides between the filter and filtrate indicated the presence of both particulate and dissolved forms. The overall quantitative recovery of these radionuclides showed that the dissolved species--those passing through 0.45-pm filters--were effectively recovered by ion exchange or precipitation. The results of these experiments are given in Tables IV and V. Recoveries of 5^Mn, 58co, ^Co, 134cg^ and 137cg by filtration and ion exchange were > 95 percent. Precipitation of strontium and iodine from 16 liters of seawater (see Table V) recovered 100 + 10, 120 + 30, and 96+ 7 percent of 89gx^ 9Ógp^ and 1 3 1 l, r e sp e c tiv e ly . The second tra c e r experim ent, in which a more complex m ixture of rad io n u c lid e s was observed, a ls o showed th at a number of other radionuclides were quantitatively removed by filtration. For ^3^Np, 35+5 percent was recovered on the filter and 59 + 10 percent on the Chelex ion exchange resin. ENVIRONMENTAL MEASUREMENTS On 5 occasions, the sampling system was used for monitoring in the coolant canal and Barnegat Bay during liquid waste releases. Sampling in the discharge section of the coolant-water canal was 0.8 km downstream from the point of release. The sampling in the intake section was 1 km upstream from the point of release. The undiluted liquid wastes were analyzed to predict concentrations in the discharge canal based on the release rate of waste tank and flow rate of the canal. The wastes dis charged into the bay can recirculate, hence sampling in the intake canal is necessary for relating discharge canal to effluent measurements. In the absence of recirculation during waste discharge, the discharge canal should be sampled prior or subsequent to a waste release to determine "background" radionuclides from fallout or recirculation of previous releases. Because of scheduling problems, it was not always possible to measure the recircu lation and background concentrations at the appropriate times. The results of two sets of measurements in the coolant canal (in Tables VI and VIl)show a relatively large number of radionuclides being discharged. Shorter-lived radionuclides were not detected because of the time lapse between sampling and analysis. As observed in tracer experiments with reactor wastes, the radionuclides were mostly particulate. The large particulate fraction for l-^Cs and ^^^Cs (Table VI) was not typical of T ab le IV Recovery oí Radionuclides on Concentration System, September 1972 % Recovery Radioactivity C artrid ge Membrane Radionuclide added. pCi/liter f i l t e r filter discs C helex-100 NCFC Dowex-1 T otal ^ C r 129 + 6 102 + 5 1 .8 + 0.1 < 1 < 1 < 1 104 + 5 ^Sin 61+3 69 + 3 0.10+0.01 29 + 2 < 0.1 <0.1 9 8 + 4 '"C o 330 + 15 55 +3 0.7 + 0.1 39 + 2 < 0.1 < 0.1 9 5 + 4 131^ 400 + 18 1.8 + 0.1 < 0.1 6.7 + 0.3 < 0.1 21 + 1 30 + 1 4 * C s 1900 + 100 0.8 + 0.1 < 0.1 8.0 + 0.4 8 9 + 4 < 0.1 9 8 + 4 4?Cs 3400 + 200 0.8 + 0.1 < 0.1 8.1 + 0.4 90 + 4 < 0.1 9 9 + 4 1. 400 ml from Waste Sample Tank В on September 25, 1972, was added to 190 liters ofcoolant canal water. 2. Coolant canal water - pH 7.5, salinity 24.8 °/oo. 3. Total volume passed through the collection system was 170 liters. 4. The + values are based on 2cr counting errors or a minimumof 5 percent; < values are 3a counting error. IAEA-SM-180/41 251 T able V Recovery of Radionuclides on Concentration System, July 1973 Radioactivity added, %, Recovered Radionuclide pCi/liter Miter Ion Exchange Total ^ C r 303 + 15 99 + 5 1..2 + 0.2 100 + 5 ^Mn 166 + 9 96 + 5 0..50+ 0.04 96 + 5 ^ C o 11 + 0.5 91 + 4 6..5 + 0.5 98 + 4 ^ F e 26 + 1 95 + 4 0,,4 + 0.2 95 + 4 6°Co 193 + 9 95 + 5 2..2 + 0.1 97 + 5 ^ Z n 3.2 + 0.5 116 + 16 7 + 2 120 + 16 S'sr 13 + 1 NA^ 100 + 103 100 + 10 9 °S r 0.5 + 0.1 NA 120 + зоз 120 + 30 ^ N b 18.0 + 0.5 5 6 + 3 < 1 56 + 3 95zr 5 .3 + 0.5 121 + 11 < 1 121 + 11 Mo 61 + 4 84 + 6 < 1 84 + 6 103^ Ru 3.7 + 0.5 93 + 13 < 1 93 + 13 0.6 + 0.3 0.1 67 + 16 70 + 20 ^^^Sb 3.7 + 0.5 100 + 14 < 1 100 + 14 131^. 28 + 3 57 + 6 39 + 4^ 96 + 7 " ^ C s 2.8 + 0 .2 25 + 12 91 + 9 116 + 15 i3 ? c s 7.4 + 0.5 26 + 6 73 + 5 99 + 8 ^ ° B a 11.1 + 1.6 22 + 3 < 1 22 + 3 ^ ^ C e 10.0 + 0.5 82 + 4 4 + 1 86 + 4 ^ ^ C e 22 + 3 57 + 8 < 1 37 + 8 58 + 8 35 + 5 59 + 10 94 + 11 131 ' ^Sr, 90^^ 1 determined by precipitation on 16-- l i t e r ;sample. NA - not analyzed N otes: 1. 1000 ml of waste sample tank on July 17 , 1973,,was added to 208 liters of coolant canal water.. 2. Coolant canal water: pH 7.2 , salinity 16.4*''/oo. 3. Total volume of solution passed through collection system was 190 liters. 4. The + v a lu e s a re based on a 2o counting e rro r or a minimum o f 5%; < values are 3a counting errors. 5 MONTGOMERY et al. 252 T a b le VI Radionuclides— ------in Discharge------o-r Canal.------: May 16. 1972 During Discharge. pCi/liter Before Discharg ^Calculated from waste analysis and dilution factor of 10,000. ^NA - not analyzed Notes : 1. Sample volumes: during discharge - 200 liters; before discharge - 330 liters. 2. Water analyses: during discharge - pH 7.3, salinity 17.1 °/oo, solids 32 mg/liter; before discharge - pH 7.3, salinity 17.1 °/oo, solids 26 mg/liter. T ab le V II Radionuclides in Coolant Canal July 17-18, 1973 Discharge canal July 17, 1973 Intakecanal during discharge Ju ly 18 . 1973 Measured Radionuclide Predicted^ Filter Filtrate Filter Filtrate*^ Predicted 3 'c r 0 .9 + 0.1 1.1 +0.3 NA^ ND^ NA 1.2 + 0.3 ^Mn 2.4 + 0.1 1.9 + 0.1 0.02 + 0.01 0.04 + 0.01 < 0.1 0 .8 +0.1 ''F e 0.42 + 0.03 0.3 + 0.1 ND ND NA 0 .7 + 0.3 ^ C o 0.14 + 0.01 0.12 + 0.03 ND ND < 0.1 0 .9 + 0.2 6°Co 2 .9 + 0.1 2.0 + 0.1 0.06 + 0.01 0.10 + 0.02 < 0.1 0 .7 + 0.1 S 's r 0.015 + 0.001 NA < 0.04 NA < 0.03 ... IAEA-SM-180/41 IAEA-SM-180/41 9 °Sr 0.003 + 0.001 NA 0.91 + 0.04 NA 0.34 + 0.04 --- ^N b 0.17 + 0.01 0.18 + 0.03 NA ND NA 1.1 + 0.2 93zr 0.10 + 0.01 0.12 + 0.04 NA ND NA 1.2 + 0.4 '°'R u 0.04 + 0.02 ND NA 0.06 + 0.03 NA — 131i 0.16 + 0.01 0.1 + 0.1 < 0.1 ND NA — ' ^ C s 0.016 + 0.005 < 0.01 0.08 + 0.01 ND < 0.1 ... ' ^ C s 0.02 + 0.01 < 0.01 0.46 + 0.01 0.02 + 0.01 0.31 + 0.06 ... ^ C e 0.06 + 0.01 ND 0.06 + 0.02 0.01 + 0.01 NA ... ^Calculated from waste analysis and dilution factor of 38,000 ^NA - not analyzed; ND - not detected, generally < 0.01 pCi/liter ^Filtrate was not passed through ion exchange system but 16 liters were analyzed by sequential analysis for 54цп, 60co^ 89sr, 90sr, 134cg, ^ C s. Notes : 1. Sample volumes: discharge canal water - 209 liters; intake water - 152 liters 2 5 3 2. Water analyses: discharge canal: pH 7.2, salinity 16.4 °/oo, solids 33 mg/liter intake canal: pH 7.2, salinity 18.1 °/oo, solids 46 mg/liter. 254 MONTGOMERY et al. other measurements where they were mostly in the dissolved fraction (> 95 percent). The ratios of measured concentration to the predicted concentra tion during discharge in Tables VI and VII have been corrected for the contribution from recirculation (intake) or background (discharge canal before discharge). The ratios range from 0.7 to 1.2 for those radionuclides measured with sufficient precision. Among those radionuclides of primary interest, the ratios were near unity for ^Mn, 134^g^ and 137^g less than unity for 58(¡o and 60co; they could not be determined for 90sr, and 131i. The ratios for 51cp and 59pg were also near unity, suggesting that these radionuclides may also be determined with this system. In view of the uncertainties associated with these measurements, the agreement is believed to be reasonable. Factors leading to uncertainties in this type of measurement include: 1. Dilution factors are based on nominal values of waste release and canal flow rates. 2. Homogeneous m ixing of w aste with can al w ater i s assumed. 3. Radionuclides may deposit from canal water prior to sampling or may be resuspended from canal sediment. 4. Recirculation effects were not measured (Table VI) or were not measured simultaneously with discharge canal sampling (Table VII). 5. Measurements of radionuclides in the liquid waste before dilution in seawater and after dilution both have considerable associated uncertainties because of the low radioactivity. It is apparent from the different distribution of radionuclides between particulate and dissolved species that concentration methods should be tested under actual conditions. Among desirable improvements in the demonstrated system would be collection at a faster flow rate on the columns and development of such columns for radiostrontium and radioiodine. SUMMARY The con cen tration system developed fo r stu d ie s a t a n u clear power station is useful for measuring ^^n, 58co^ °^Co, ^34çg^ gnd 137(-.^^^.^ 400 liters of seawater with a detection limit of 0.01 pCi/1; and Sr, 90sr, and 131i from 16 liters with a detection limit of 0.1 pCi/1. The high degree to which many radionuclides are retained on filters affords a convenient technique for concentrating thëse radionuclides from large volumes of seawater. This technique might prove useful in following dilution patterns of waste discharges when the distribution between particu late and dissolved species is known. In the course of this study, the relationshipsbetween radionuclide levels in liquid waste, seawater, and marine organisms were found to be extremely complex. Simple analysis or prediction of the extent of radio active contamination in the marine environment from monitoring liquid waste discharges is complicated by factors such as: 1. Lack of knowledge regarding the chemical and physical properties of radionuclides discharged to the seawater environment. 2. Variations of liquid waste discharges in frequency, duration, as well as radionuclide composition and concentration. IAEA-SM-180/41 255 3. Recirculation of discharged radioactivity. 4. Differences in dispersion patterns and dilution factors due to tidal and wind conditions. As a result, calculations of radionuclide levels in marine organisms from concentrations in water based on steady state assumptions are approxi mations at best. ACKNOWLEDGMENTS We thank Dr. David McCurdy and staff of the New Jersey Depart ment of Environmental Protection, and the staff of the Oyster Creek Nuclear Generating Station for assistance in making field measurements. The technical assistance of Betty Jacobs, Eleanor Martin, and George Frishkorn from our laboratory is acknowledged. REFERENCES [1] Reference Methods for Marine Radioactivity Studies, Technical Report Series No. 118, International Atomic Energy Agency, Vienna (1970). [2] KAHN, B-, et al., Radiological Surveillance Studies at a Boiling Water Nuclear Power Reactor, USPHS Rept. BRH/DER 70-1 (1970). [3] KAHN, B-, et al., Radiological Surveillance Studies at a Pressurized Water Nuclear Power Reactor, USEPA Rept. RD 71-1 (1971). [4] JOYNER, T ., HEALY, M. L . , CHAKRAVARTI, D ., KOYANAS, T. P reconcentration for Trace Analysis of Sea Waters, Env. Sei. Technol. 1^ (1967) 417-24. [5] SLOWEY, J . F ., HAYES, D ., DIXON, B ., HOOD, D ., D istrib u tio n of Gamma- emitting Radionuclides in the Gulf of Mexico, Proc. Symp. Marine Geochem istry, Univ. Rhode Islan d Occ. Publ. 3- 1965 (1965) 109-129. [6] CHAKRAVARTI, D ., LEWIS, G. B ., PALUMBO, R. F . , SEYMOUR, A. H ., Analyses of Radionuclides of Biological Interest in Pacific Waters, Nature, Lond. 203 (1964) 571-573. [7] YAMAGATA, N., IWASHIMA, K., Monitoring of Seawater for Important Radioisotopes Released by Nuclear Reactors, Nature, Lond. 200 (1963) 52. [8] SILKER, W. B ., PERKINS, R. W., RIECK, H. G ., A Sampler fo r Concentra ting Radionuclides from Natural Waters, Ocean. Engng. 2^ (1971) 49-55. [9] GOYA, H. A., LAI, M. G., Adsorption of Trace Elements from Seawater by C helex-100, USAEC Rept. USNRDL-TR-67-129 (1967). [10] RILEY, J. P., TAYLOR, D., Chelating Resins for the Concentration of Trace Elements from Seawater and their Analytical Use in Conjunction with Atomic Absorption Spectrophotometry, Anal. Chim. Acta 40 (1968) 479-485. [11] Panel on Radioactivity in the Marine Environment, Radioactivity in the Marine Environment. N ation al Academy of Scien ces (1971) 77-79. [12] LAI, M. G ., GOYA, H. A ., A Compendium of Radiochem ical Procedures for the Determination of Selected Fission Products in Seawater, USAEC Rept. USNRDL-TR-912 (1965). 256 MONTGOMERY et al. [13] SUGIHARA, T. T ., JAMES, H. I . , TROIANELLO, E. J . , BOWEN, V. T ., Radiochemical Separation of Fission Products from Large Volumes of Seawater:Strontium, Cesium, Cerium, Promethium, Anal. Chem. 31 (1959) 44-49. [14] MIYAKE, Y., A Sequential Procedure for Radiochemical Analysis of Marine Material, Unpublished working paper to Methods of Surveying and M onitoring Marine R a d io a c tiv ity , IAEA, S afety S e r ie s No. 11. (See Reference 1, page 205). [15] SODD, V. J., GOLDIN, A. S., VELTEN, R. J., Determination of Radio activity in Saline Waters, Anal. Chem. 32^ (1960) 25. [16] BONI, A. L., Rapid Ion Exchange Analysis of Radiocesium in Milk, Urine, Seawater and Environmental Samples, Anal. Chem. 38 (1966) 89-92. [17] TERADA, K., HAYAKAWA, H., SAWADA, K., RIVA, T., Silica Gel as a Support for Inorganic Ion Exchangers for Determination of Cesium in Natural Waters, Talanta 17 (1970) 955-963. [18] MATTHEWS, A. D., RILEY, J. P., A Study of Sugawara's Method for the Determination of Iodine in Seawater, Anal. Chim. Acta 51^ (1970) 299-301. [19] MORGAN, A., ARKELL, G. M., A Method for the Determination of Cesium-137 in Seawater, Health Physics 9^ (1963) 857. [20] KRIEGER, H. L., GOLD, S., Procedures for Radiochemical Analysis of Nuclear Reactor Aqueous Solutions, USEPA Rept. R4-73-013 (1973). DISCUSSION N. T. MITCHELL: Like you, I was surprised at the amount of radio- caesium you found in particulate form, which is also contrary to our general experience in sea-water. I wonder whether you have any explanation for this phenomenon, in particular whether there was a high sediment load in the discharge canal at the time? D.M. MONTGOMERY: The resuspension of sediment in the canal can increase the concentration of particulate radiocaesium. However, in this case a relatively large fraction of radiocaesium was particulate at the time of discharge and was not from the sediment. I have no explanation as to why radiocaesium was particulate. We have found many of the radionuclides from reactor wastes to be extremely reactive. Radionuclides were found to deposit on plastic sample bottles even when acidified to 10% with nitric acid. We have little information on the physico-chemical properties of the radio nuclides from these wastes, but it is an area of interest to us. IAEA-SM-180/57 PRACTICAL REFERENCE LEVELS FOR RADIOACTIVE CONTAMINATION IN ENVIRONMENTAL SURVEILLANCE G. BOERI, C.BROFFERIO Divisione di Protezione Sanitaria e Controlli, Comitato Nazionale per l'Energia Nucleare, Rome, Italy Abstract PRACTICAL REFERENCE LEVELS FOR RADIOACTIVE CONTAMINATION IN ENVIRONMENTAL SURVEILLANCE. The authors describe a method of defining practical reference levels of radioactive contamination in environmental samples collected for surveillance purposes around nuclear sites. The method makes use of the environmental radiological capacity concept and therefore takes into account the site characteristics but, for practical use, it is suggested that only the most meaningful values be retained. It is shown that the derivation of these levels can be made taking into account, for a particular nuclide, both the environmental capacity and the sensitivity of the method used to measure the contamination level. A series of contamination levels can thus be defined for each site; they may act as a reference of radioactive contamination for that particular environment. In the last part of the paper some applications of this method to Italian sites are discussed. 1. INTRODUCTION At the start of nuclear energy exploitation in Italy, industrial and research installations were operated with discharge lim its for radioactive effluents based upon the MPCs of radioactivity in air and water for members of the population set out by Italian law. In more recent years, in accordance with ICRP Report No. 7, a revision of these lim its was started by the Regulatory Authorities, taking into con sideration critical groups and making use of the concept of the so-called "limiting radiological capacity". The radiological capacities as assessed for the various nuclear sites in Italy range over a certain spread of values because the installations are generally located on sites displaying different natural and man-made features. Furthermore the installations are not all of the same general type, so that their authorized discharge lim its and their programmes for environmental surveillance are different. Therefore the Regulatory Authorities are faced with the problems of evaluating the available surveillance data with a view to determining the actual impact of discharges on the environment and in term s of actual or potential radiation exposure of the public. To cope with these problems we found it convenient to derive some practical reference levels for radioactivity in various segments of the environment; the present paper is devoted to the discussion of these le v e ls . 257 258 BOERI and BROFFERIO 2. ENVIRONMENTAL SURVEILLANCE ORGANIZATION IN ITALY 2.1. General considerations Initially discharge lim its in Italy were issued on the basis of MPCs in the effluents from nuclear installations and the licensees monitored total gross activity in the effluents at the discharge point and measured radio activity concentrations in some environmental sam ples (generally water, air and milk) collected around the installation. The CNEN (the Italian Regulatory Authority) by means of inspections controlled that radioactive efñuents were correctly monitored and that no appreciable increases of radioactive contamination occurred in the environment. In more recent years new discharge lim its have been issued on the basis of the critical pathway and critical group approach; the permitted discharge lim its generally represent only a stipulated fraction of the quantity of radioactivity determined to be the limiting radiological capacity for each site. Discharge limits are expressed as a discharge formula, containing certain term s that individually correspond to the nuclides significant from the viewpoints of both radiotoxicity or the quantities that might be discharged. Control and monitoring procedures have consequently to be suited to this approach. The licensee must evaluate, for each term of the formula, the amount released at the discharge point in order to prove compliance with the authorized lim its; he must also operate an environmental sur veillance programme so that the main human exposure pathways are kept under control. CNEN analyses surveillance data supplied by the licensees and seeks for an interpretation of them in term s of the exposure of the population from radioactive wastes releases. In addition to the aim of supplying to the public correct information concerning radioactive waste management, a selection of environmental data and dose evaluations will be published in a revised form of the national Bulletin that is circulated to central and local authorities (from 1974). 2. 2. The reference level approach Nuclear installations in Italy have, in general, different constructional and site characteristics. Of the three operating power plants, the Latina IAEA-SM-180/57 259 (GCR) station is on the sea-coast and the others are located on rivers — the Po river (a PWR) and the Garigliano river (a BWR). A large BWR is under construction on the Po river, also. In addition to the nuclear power stations, there are two pilot reprocessing plants, two relatively large research centres, some fuel element fabrication plants, and some sm all research facilities scattered over the country. When the new discharge lim its were issued, the environmental investi gations carried out for each site in order to find potential critical groups of the population having particular living habits, in accordance with the principles set out in ICRP Report No. 7, showed that the dietary habits of Italians are different in the various regions of the country. Furthermore, the Italian diet consists largely of locally produced foods (such as leaf vegetables and fruit) and, because the contribution to the total dose coming from the various food-stuffs are often of the same order of magnitude, it is not generally possible to define a particular critical pathway for radionuclides taken in with food. From these considerations, the environmental radioactivity m easure ments can be interpreted in term s of population exposure, only taking into account local conditions. It is, therefore, customary to correlate concen trations of radioactivity in environmental sam ples with doses received by man using 'derived working lim its' (DWL) or 'reference levels' or guides. We, too, think it is useful and practical to have some levels directly related to population exposure and to effluent discharges. Considering the Italian situation, we found that reference levels cannot be adopted on a nation wide basis, but rather they must be derived taking into consideration living and dietary habits of local populations and natural and man-made charac teristics of the local environment. Figure 1 shows the pattern of the approach adopted to derive the inter national DWLs (quoted in Table II), while Fig. 2 shows the method adopted by us to obtain our ELCs and ECCs (see later). It is readily recognized that the DWL for a food-stuff is directly related to the dose lim its for the population coming only from the ingestion of that particular food. The ELCs instead imply the attainment of dose lim its for population taking into account the contributions from all possible exposure pathways (diet, external exposure, internal contamination by inhalation, etc.). TABLE I. LATINA POWER STATION 260 Measurement N uclide group 1/100 ELC 1/1 0 ECC sensitivities (Ci/a) (Ci/a) (p C i/l or p C i/k g ) (pCi/l or pCi/kg) r 'H 5 x 10^ 10* water 10* 2 x 10^ 2 x 10^ 32p water 0 .2 1 5 8 0 .5 fish 1 .6 X 10^ 10^ 10 shellfish 1 .6 x 10^ 10' 10 OR ad BROFFERIO andBOERI '"Sr water 6 2 1 3 x l ( f 1 x 10* fish 6 2 5 shellfish 6 2 5 Z '" C s water 5 4 1 x 10* Eg 1 .5 x 10^ 1.3 x 10^ 1 x 10* EH ^ 2 .4 x 10 2 x 10 shellfish 5 x lo' 4 x 10* 1 x 10* Crustacea 5 x 10^ 4 x 10" 1 x 10* ^M n water 2 x 1 0 *' 1 x 1 0 *' 1 x 10* 1 x lo ' 3 fish 6 x lo" 2 x 10^ 8 x 10* shellfish 1 x 10" 3 x 10* 8 x 10* " C a water 2 x 10' 2.5 x 10' 2 x 10* 12 x IO" 1 x 10^ fish 2 x 10^ 2 .5 x 10^ 1 x 10^ shellfish 2 x 10^ 2 .5 x 10^ 1 x 10^ " 'P u 4 x 10* 1 x 1 0 *' water (2 x 10*^) (8 x 10*') 1 x 10° IAEA-SM-180/57 261 For these reasons we prepared for each site a set of reference levels, expressed as concentrations in the various segments of the environment, that correspond to the lim its of dose-commitment or to the allowed discharge lim its . Since our purpose is to verify the behaviour of effluents in the environ ment and to assess related radiation doses to the population, we suggest two kinds of reference level: the first, the ECC (Environment Compliance Concentration) corresponds to the concentration of a given nuclide expected in an environmental segment if the authorized limit is discharged; it serves the purpose of directly verifying the plant's compliance with the authorized lim its. The second level, the ELC (Environment Limit Concentration) represents the concentration of a given nuclide expected in an environmental segment in correspondence with the limiting radiological capacity; it allows of a prompt evaluation of the exposure of the public. The practical reason for establishing two separate levels lies in the fact that for a given radio nuclide the authorized lim its are not always the sam e fraction of the limiting radiological capacity, as these lim its are a compromise between operational needs and the necessity of keeping exposure of the public as low as possible (the "as low as readily achievable" principle). Since for the radiological capacity evaluation we have considered all the possible contributions from dietary and living habits, using a computer co de [1], for each radionuclide we shall have a series of reference levels each related to a specific segment of an exposure pathway. To avoid too many figures, we chose some values corresponding to either the most contaminated sam ples or the easiest samples to collect and then measure the corresponding environmental m aterials. These levels might find a useful application in acting as a means of checking for the radioecological studies which describe the behaviour of the effluents in the environment. 3. PRACTICAL REFERENCE LEVELS 3.1. D e riv a tio n On the basis of the preceding discussion, we derived the ELC and the ECC for each installation. For practical use, however, we employ only a fraction of the ELC and ECC. We adopted 1/100 of the ELC and 1/10 of the ECC. We chose these values taking into account considerations mainly of a practical character. As a matter of fact it is not customary in regulatory procedures to make reference to the whole limiting capacity; on the one hand, therefore, we take 1/100 of the ELC because, to our mind, this figure is low enough to assure an adequate degree of control before the assessed exposure of public reaches an appreciable fraction of the dose lim its. On the other hand this same ELC /100 value is high enough to allow of measurement without excessive effort, by the methods generally being used by the licensees. This is quite evident from an examination, for example, of columns 6 and 7 of Table I. In the sam e way in the monitoring of radioactive discharges, we think it is preferable to make a comparison with a fraction of the environment compliance concentration (ECC); for the same reasons as above with the 262 ВОЕИ and BROFFERIO ELC, we chose 1/10 of the ECC, this value being generally of the same order of magnitude as 1/100 of the ELC. It must be noted that only liquid effluents are considered in this paper, but obviously sim ilar methods can be applied to gaseous and airborne e fflu e n ts. 3.2. A case study: Latina Power Station We shall describe in some detail the method followed to derive the reference levels for the Latina Power Station. The Latina plant is a 210 MW(e) power reactor (GCR) located on the Thyrrenean sea coast, about 70 km south of Rome; it has been operating for about 10 years. Liquid effluents reach the sea by means of a 1 km long canal, having a 15 m^/s flow-rate. The radioecological investigations showed that the critical group consists of fishermen living on the coast; these people eat fish and shell fish caught locally in the sea (100 kg/а fish, 25 kg/а shrimps; 25 kg/a shellfish); they are also exposed to irradiation from water and sediments and due to handling of fishing gear. As a consequence of these living habits and considering the expected plant discharges, the following discharge lim its (discharge formula) were issued: Зн 31p 90g^_ 134^ + 137^ (P ) (P) ( where (ß--y), (ß) and (и) term s represent the activity of (ß--y), (8) and (c) emitters expressed as curie-equivalent of ^Mn, ^C a and 239Pu, respectively. The curie-equivalent is the activity (Ci) of a given nuclide that may be discharged from the installation to give an exposure corresponding to 1 Ci of the reference nuclide. The equivalent activity of a nuclide is obtained as the product of its actual activity (Ci) times a 'risk factor', which is the ratio of the radiological capacity for the reference nuclide to the one under consideration: Curie-equivalent = A - f where A is the activity (Ci) of the nuclide under consideration and Table I shows, for each term of the formula, the values of ELC/100 and ECC/10. Other items listed in the table are the limiting radiological capacity itself, the allowed discharge lim its, the critical group and the environmental segment of interest. The sensitivities of the measurement methods complete the information. Below, the criteria followed for choosing for each nuclide in the table the levels retained as the most conspicuous and useful for practical use are briefly discussed. IAEA-SM-180/57 263 T ritiu m Since tritium released to the environment does not undergo any concentration process, the only control point will be in the water. 32 p The phosphorus shows a high concentration factor in fish and shellfish, but as shellfish are generally more easily collected and do not migrate, their levels were considered more significant. 90 S r Like ^H, 9°Sr does not appreciably accumulate in marine organisms, so that ELC/100 and ECC/10 in water were chosen. Considering the ratio between fish concentration and measurement sensitivity as compared to levels for water and shellfish, we chose the fish as the 'segment' from which to evaluate the reference le v e ls . ¡3-т, 8 ander emitters It may be argued from Table I that, for mixtures of these em itters, we have derived the reference levels only for ^Mn, 239p^ This is due to the fact that, generally, the concentrations attained in the different segments are very low as a consequence of the low discharge rates from the installation. The licensee therefore makes only gross beta and beta-gamma determinations, and subtracts from the result the activity of those nuclides which appear explicitly in the formula; the value so obtained is assumed to be entirely due to the reference nuclide. Of course the relative ELC and ECC can be derived for any nuclide following the sam e method as for the term s given in the above-mentioned formula. The usefulness of such a practice would only be given if the discharge activity of any particular nuclide should increase so much that it represents a notable fraction of the allowed lim its. The same is true if the environmental behaviour of some particular nuclides are to be investigated. 3.3. Summarized data for some Italian plants In Table II are listed the reference levels, the ELCs and the ECCs for the three operating power stations and the Eurex pilot reprocessing p la n t. The table also contains some values drawn from the literature [2-4] (e.g. UK DWLs and USA RPGLs) that can be directly compared with the ELCs because both correspond to dose lim its, though they have been derived following different assumptions. An analytical examination of the data allows the following conclusions to be drawn and certain comments to be made. As expected, the reference levels show some variability from instal lation to installation, in the nuclides or the environmental segment concerned,as well as, naturally, in their relative values. This is par ticularly true in the case of the Latina station, that alone is located on the sea coast and discharges its effluents into the sea; the other plants are on r i v e r s . 264 TABLE II. SUMMARIZED DATA OF SOME ITALIAN PLANTS OR ad BROFFERIO andBOERI T A B L E П. (cont.) Latina E n v iro n > -^ ELC EGC ELC ECC ELC ECC ELC ECC DWL segm ent (pCi/l or pCi/kg) (pCi/l or pCi/kg) (pCi/l or pCi/kg) (pCi/l or pCi/kg) (pCi/t or pCi/kg) todine-131 1.6x10^ 1.5x10^ 1.5 xlO^ 1.3 xlO^ 4x10^ 8x10"' 4 x 10^ [2 ] M ilk - -180/57 IAEA-SM Fish 6 x lo " 6 x 10* 1.5 xlO^ 1.3 xlO^ 3 x 10' 6 x 1 0 *' 2 x 10^ [ 4] Shellfish Fish 1.5 xlo" 1.2 xlO^ 4.5 x lo' 3 x 10" 3 x l o ' 1.8 x l o ' 3 x 10^ 6 x 10^ 4x10^ 2.2 xlO^ M ilk (1 x 10^) (5 x 10^) (2x10) (4x10-') 3 x 10" [2 ] Plutonium -239 Water 8x10 2x10*' 2.3x10^ 2.3x10° Fish 2.3x10' 2.3x10^ 2 x l o ' 6 x 10^ 5 x 1 0 ' [3 ] 265 266 BOERI and BROFFERIO The 9°Sr levels might be expected to be lower in sea foodstuffs, since this nuclide concentrates to a greater extent in fresh-water organisms; due to dietary habits, as sea fish and shellfish are consumed in greater quantities than fresh-water fish, no significant difference in the levels can be observed. The value is quoted only for the Latina station, because it is an activation product of impurities in the graphite and accumulates notably in the marine environment. The licensee had to devise ad hoc measurements that, in general, confirmed the assumption about the environmental behaviour of ^P . in order to refine the knowledge of these phenomena and to follow a mainexposurepathway, sampling points for ^P were introduced in the surveillance programme. For some nuclides (caesium, strontium and iodine), the Trino and Eurex installations, as local conditions are much the same, have sim ilar reference levels since the exposure pathways are sim ilar. An interesting item is that in some instances the ^ 1 contribution to the dose due to fish consumption is of the same order as that from milk. This observation received direct confirmation in a recent survey on fishes caught in the Po river, downstream of the Trino station. Consequently it was decided to include a ^ 1 bimonthly measurement on fish samples in that particular surveillance programme. The comparison of the ELCs with the DWLs firstly shows a general agreement; in some instances, however, there are notable differences, reflecting the different assumptions and methods followed in deriving our reference levels. 4. CONCLUSIONS The interpretation of environmental surveillance data by means of a set of reference levels offers, in our opinion, some features of interest for a Regulatory Authority. In fact, as has been pointed out, the use of reference levels allows of an initial control of a plant discharge behaviour and a prompt evaluation of the consequent radiation doses to the public even if a more complete analysis could and should follow later. Of course, the reference levels are not, in our mind, fixed standard values nor regulatory ones, and they have to be subjected to review on the basis of surveillance data, in order to attain always more realistic figures relating to the sites under control. During the operating life of the plant it is possible, in fact, to follow the correspondence between the preliminary studies and the picture resulting from later environmental surveillance data. Another feature of these levels which we should like to emphasize is that, whenanew installation is submitted to the Regulatory Authority for licensing, the derivation of reference levels and comparison of these with those established for the plants already operating contributes towards ascertaining whether the pre-operational surveys are adequate, and makes it easier to design the subsequent environmental studies and surveillance programme correctly, and to adopt the most convenient measurement m e th o d s. IAEA-SM-180/57 267 ACKNOWLEDGEMENTS The authors should like to thank the following persons for their suggestions and helpful assistance: L. Bramati, F. Breuer, A. Cigna, G. De Cecco, O. Ilari, A. Nardi, C. Polvani. REFERENCES [1 ] BRAMATI, L., MARZULLO, T ., ROSA, I . , ZARA, G ., "Vadosca: a simple code for the evaluation of population exposure due to radioactive discharges", Paper at 3rd Int. Cong. IRPA, Washington, Sep. 1973. [2 ] DUNSTER, H .J., The Application and Interpretation of ICRP Recommendations in the United Kingdom Atomic Energy Authority, UKAEARep. AHSB(RP)R78 (1968). Idaho, Nov. 1970 (1971) 634. [4 ] WEAVER, C .L ., "A proposed radioactivity concentration guide for shellfish", Rad. Health Data and Reports (Sep. 1967) 491. IAEA-SM-180/37 RAPID PORE WATER ANALYSIS FOR SEDIMENTS ADJACENT TO REACTOR DISCHARGES E.K. KALIL Department of Geology, University of California, Los Angeles, Calif., United States of America Abstract RAPID PORE WATER ANALYSIS FOR SEDIMENTS ADJACENT TO REACTOR DISCHARGES. may result in environmental changes. Bacterial sulphate reduction w ill occur in organic-rich sediment near the sediment-water interface and w ill result in production of toxic hydrogen sulphide (HgS), which may suppress benthonic life on the ocean floor. An integrated approach is presented for collection, handling and analysis of sediments adjacent to reactor cooling water discharge. Samples are collected by gravity cores or grab samplers and transferred with minimal air contact to a squeezing chamber. The pore (interstitial) waters are squeezed from the sediment on board ship, filtered and collected in air-tight plastic syringes. Separation of pore waters allows of analysis of dissolved elements that enter the sediments from the overlying water or by release from the solid phase. Radioactivity can be readily assayed without interference from the solid detrital material of the sediment. The squeezer designed for this work is constructed from unreactive materials, and is simple to operate, allowing of rapid processing in the field. For this study, a flow-through electrode cell ' was connected to the squeezer in order to measure pH-value, total H^S and the oxidation-reduction potential (Eh) from the pore water effluent. A unique calibration procedure was developed for use of an AggSI sulphide electrode over a continuous pH range. The system is sensitive to at least 0.1 micromolar or 0.003 ppm Hg S. The HgS, Eh and pH profiles in sediment samples w ill rapidly show any microbiological effect caused by anomalous bottom conditions resulting from reactor discharge or other perturbations in the aquatic environment. Water exhausted from the flow-through cell is collected and can be examined for other chemical parameters of interest. 1. INTRODUCTION Nuclear reactors often use ocean water as the cooling fluid to dissipate excess heat. The heated discharge can indirectly poison the sediments without ever releasing any toxic chemicals. A slight increase in temperature causes a decrease in dissolved oxygen, an increase in bacterial res piration and an increase in the rate of diagenetic chemical reactions in the sediments. Zooplankton entrained in the cooling water suffer approximately a 20-30% mortality loss, and an undetermined phytoplankton loss may also occur. Even though this loss has no dramatic effect on their population [1], the decaying organic matter does place a high biological oxygen demand (BOD) on the adjacent sediment. In an area surrounding warm water discharge an ideal setting for bacterial growth exists, with warm temperature and high organic substrate. In some cases the warm water may never touch the sediments because of its lighter density and a resultant thermocline. 269 270 KAHL Probably the most important form of anaerobic respira tion in marine sediments is bacterial sulfate reduction [2]. Organic matter is oxidized anaerobically using SO^ (28 mM concn in sea water) as the terminal electron acceptor. A typical metabolic pathway for the sulfate reducing bacteria, Desulfovibri о desulfi ri cans is [3] SO^ + 2СН^СН0НС00Н ^ HpS + 2HC0g + 2CH^C00H. Hydrogen sulfide is the final product, which is toxic to most marine respiratory organisms. Aerobic metabolism is the nor mal mode of respiration in the surface of most sediments. There is sufficient dissolved oxygen to allow the bacteria to combust the available organic matter to CO2 . At some depth in the sediment, oxygen is depleted, beyond a depth where it cannot diffuse in, and anaerobic respiration and fermentation ensue. Several physiological and chemical effects are associated with sulfate reduction occurring at the surface of the sedi ments. For example, the pH can fall below 7, the concentra tion of trace metals can increase [4], and bottom fish can display nonbacterial fin rot [5]. The variety of benthic fauna can be reduced to two species of polychaete worms [6]. Rapid production of dissolved H^S in surface sediments thus effectively poisons the sediments to respiratory organisms. In order to monitor the process of sulfate reduction and related chemical process, it is convenient to analyze the interstitial water, or pore water of the sediment, where the reactions take place. Pore waters show large variation in the concentration of sulfate, sulfide, and bicarbonate and are a sensitive indicator for changes [7, 8]. Pore waters can easily be removed by squeezing and then analyzed, in this case, by electrodes which are quick and require only a small sample. The pH and redox potential are also possible indicators of sulfate reduction. In the pore waters, pH is inversely proportional to sulfide concentration, and falls in the narrow range of 6.9 to 8.2. Theoretical calculations on anoxic pore waters by BEN-YAAKOV [9] show that if all the sulfide is not removed as metal sulfides and stays in solu tion, the pH will be lowered to 6.9. Bacterial sulfate reduc tion results in an alkalinity increase by as much as 30 times [10]. Thus low pH values may be indicative of sulfate reduc tion. The redox potential, Eh, is defined by the equation rt. - rO . RT i__ (oxidants) Eh - E + ¡¡y log (reactants) where E = standard state electrode potential R = gas constant T = absolute temperature F = Faraday's constant n = number of electrons transferred tAEA-SM-180/37 271 (as measured with a platinum electrode relative to a standard hydrogen electrode). The meaning of Eh values and their re lationship to natural marine processes has been discussed by EMERY and RITTENBERG [11] and by WHITFIELD [12]. They con clude, that despite problems relating Eh to equilibrium thermodynamic properties of a given solution, it may be used to qualitatively describe the redox state of natural waters. It has been argued, for example, that positive Eh values, expressed in millivolts relative to the hydrogen electrode, indicate an aerobic environment, whereas negative values respresent anaerobic conditions. Because sulfate-reducing bacteria are obligate anaerobes, they grow only in environ- . ments with Eh values below about +30 mV [11]. After growth commences, the Eh drops further, even before the appearance of free dissolved H2S [13]. Where H2S concentrations are significant, BERNER [14] postulates an empirical relation between Eh and the logarithm of the activity of the sulfide ion (pS=), Eh = -0.485 + 0.0295pS=. The Eh serves as a qualitative measure of the degree of oxidation-reduction of the sediment and can be controlled by the different microbiological populations. Chemical monitoring of the sediments by means of the pore water analysis is capable of following the progress of bacterial activity. A drop in pH and Eh is associated with sulfate reduction and the presence of hydrogen sulfide. The following techniques describe the removal of pore water from various types of sediment and the subsequent electrochemical analysis. The technique is readily adaptable for other applications. 2. METHODS Few previous techniques for dissolved sulfide in sedi ments have given satisfactory results. BERNER [14] utilized a silver-silver sulfide electrode, reporting values in the form of pS", where pS" = -log a$=. This, however, does not give total H2S in a straight forward manner. WHITFIELD [12] used a combination probe, to measure pH, Eh, and pS" on a slurry of sediment and water. He used a silver-silver sul fide electrode, prepared according to BERNER, to measure pS". SUTHERLAND [15] measured dissolved sulfide in lake waters. He used a silver sulfide electrode, first adding a sodium hydroxide, sodium salicylate and ascorbic acid buffer to re tard loss of sulfide by oxidation. Other M et chemical tech niques, such as the zinc acetate-methylene blue method [16], have a serious drawback because the analysis may also include some acid soluble iron sulfides in addition to the water sol uble su!fi des. For this project, where rapid analysis of a large number of small volume samples was required, an improved integrated approach was developed that would also be compatible with other field studies. 272 KAHL Pore waters are separated by squeezing the sediment under low hydraulic pressure [17]. It is important to minimize air contact of the sample since HgS oxidation and/or COg exchange may alter these parameters from their i_n si tu values. Once the pore water has been removed, it must be quickly analyzed. This is done by an electro-chemical technique especially developed for this project and capable of measuring pH, Eh, and E$=. The squeezer, Figure 1, is comprised of four narts; two made of polyvinyl chloride (PVC), one of plexiglas^) (methyl methacrylate), and the other of polyethylene. All materials are unreactive with H2S and will not contaminate the sample. The individual parts are expendable and inexpensive to build. The squeezer was designed to remove enough water over a short time and with little effort so that it is suitable for rou tine work-on board ship. The squeezing jacket is made of plexiglasPPof pvc. The size of the barrel was designed to fit snuggly over the 1.5 in. core tube of a "Phleger Corer," but can be made larger for other core barrels. PLEXIGLASS PVC PISTON The squeezer is assem BARREL bled by inserting a base with a porous polyethylene filter disk into a squeez ing barrel. A piece of filter paper is smoothed down on top of the porous polyethylene. Hardened filter paper (Watman No. 50) was chosen as it was found to have the least effect on pH. The three ton press is I— I.5IO --- 1 constructed of ^ in. alumi num plates, two 5/8 in. steel rods and a 3 ton hy POROUS DISK draulic jack. . '/4 Samples can be taken IJ aboard a ship using a grab PVC B A SE sampler or a corer. Scuba PVC CORE REMOVER ! in divers can also take hand cores in shallow water. Surface samples were of most interest because of FIG.l. Squeezer. their direct effect on the IAEA-SM-180/37 273 ecology of the area. Grab samples from a "Shipek" sampler were immediately subsampled with a hand held piston corer, perpendicular to the surface of the sediment. The sediment is then extruded until only the top 5 cm remain in the barrel, and then transferred to the squeezer. The subsample corer fits snuggly within the squeezer barrel, such that sample transfer is made with minimal air contact as the mud slides into place. The squeezer is then tapped on the bottom to dislodge any air pockets. The plunger is inserted until the О-ring is seated by hand. Then hydraulic pressure is slowly applied. Excess sample and air is bled through an accurately placed vent hole in the barrel. Mater leaves the squeezer from a port in the base and passes through an inline milli- pore filter (0.45 micron pore size) and collected in a 12 ml disposable plastic syringe [17]. Applying continuous pres sure, about 10 ml of pore water were collected in 5-10 min. With long core samples, 5 cm sections are extruded directly into the squeezing chamber. Disassembly is rapid. The base is usually twisted off and then using the core remover, the piston and squeezed mud cake are pressed out [17]. If it cannot be dismantled by hand, compressed air is applied at the base outlet and the sediment and plunger blown out (a bicycle pump is sufficient). The plunger is of such a length that it can be inserted back wards to knock the base off. The collected pore water was analyzed immediately on board ship. If the sample is allowed to sit, CO2 will outgas and СаСОз will precipitate, the temperature will increase and a general deterioration of the investigated properties occurs. Commercial flow-through cells are available to measure pH, but not to measure sulfide and Eh. A cell was built to measure all these parameters on water samples of 3-5 mis. The cell, Figure 2, consists of a plexiglás^ manifold and four electrodes; a glass electrode (Sensorex Corp., Newport Beach, Calif.) a double-junction reference electrode (Orion Research Inc., Cambridge, Mass.) with KC1 as the outer filling solution, and&sulfide and platinum electrode. The platinum electrode is a ^ in. disc epoxy cemented into a fitted holder. The sulfide electrode is a new design obtained from Mr. John Kater of Sensorex Corp., and is made of Ag3SI which is fitted to a holder that plugs into the cell. 274 KALIL The electrodes were connected to a junction box and readings were made with a digital pH meter (model BYK, Martek Corp., Newport pH GLASS REFERENCE Beach, Calif.). Bubbles were easily exhausted by tapping and tilting the cell with the outflow upwards. EXHAUST IN FLOW A 3-way plastic stopcock was an effective inlet valve. Exhausted waters were either collected in a plastic syringe or vented with a length of tygon tubing. The silver sulfide io dide electrode was compared Eh to a silver sulfide (Ag2S) electrode of similar con PLATINUM SULFIDE struction, in the same cell, Aq^SI by substitution for the lin , platinum electrode. Response time for the Ag^SI was rapid, 5-10 sec, whereas, the AgpS electrode was still drifting after 30 sec. Data points FIG. 2. Flow-thtough electrode cell. from eleven samples analyzed with the AgqSI electrode fall more closely on the calibration line than those analyzed with the Ag2$ electrode. Sulfide concentration was determined spectrophotometrically on an aliquot of the sample by the methylene blue technique [18]. A calibration plot was established which related total dissolved HyS with millivolt output of the electrode at con stant pH ana constant temperature. The mV output of the sul fide electrode, Eg--, at in situ pH's were corrected to values at a constant pH of 8 , by use of Figure 3. The shape of the curve is set by the first and second dissociation constants of H2S and does not vary with concentration. The curves were generated by Dr. М. B. Goldhaber, by titrating an H2S standard sea water solution [18] in a closed "Edmond" type cell with dilute HC1 [19]. As these curves were generated for different concentrations, the millivolt output for each concentration is determined at pH 8 and then plotted. Fig ure 4 shows a calibration of total H2S vs Eg-- at constant pH and constant temperature (15°C). IAEA-SM-180/37 275 FIG .3. Curves used to correct Eg- values to constant pH-value. Eg= (mV at pH 8. 15° C) FIG. 4. Calibration of sulfide electrode. At pH 8 , the electrode exhibits a slope of -1/41, or a 41 mV change per order of magnitude of sulfide concentration, which is equivalent to 1.4 electron transfer per mole. The equation of the curve is l o g " 2 ^ " ' rr("Es""mv) + b where HgS is expressed in micro moles (10*6) per liter, Es-- is the mV potential of the electrode at pH 8 , and b is the intercept. The intercept is set daily by running a standard or sample, which is independently calibrated. This same electrode has a normal 2 electron transfer, or -1/27 slope in one normal sodium hydroxide solution at 15°C. 276 KALIL Using the above curves, sample readings were adjusted to pH 8 and the concentration determined. When the electrode response indicated the presence of dissolved sulfide, a 5 ml aliquot was saved for methylene blue analysis. Sulfide elec trode results agree to within ±5% of the methylene blue results. Met chemical methods for sulfide are sensitive to a threshold of about 1 pM, whereas the electrode gave stable readings to one and possibly two orders of magnitude lower. I2I°W 40° ¡19° 48° 47° lt6° 45° 44° W 3. RESULTS 3 5 ° N The system was first 3 4 ° N tested off the southern California coast in the area of the Los Angeles County sewage discharge 33°N at Whites Point, Fig ure 5. The surface sed iments are rich in sul 3 2 ° N fides and convenient for sampling. One hundred and sixty samples at 3 I ° N forty stations were taken with a "Shipek" grab sampler aboard the L.A. County research vessel, the Sea-S-Oee during February 1973. Results are listed in Table I for the top 5 cm of sediment. The sediments range from clay to coarse silt, and display a pH range from 6.9 to 8.0, the lowest pH values were measured in areas of highest sulfide concentration. Sulfide concen trations reach a maximum of 4500 pM directly at the outfall and fall off as shown in Figure 6 . Eh values range from -300 mV to about -50 mV, Figure 7, and closely follow the sulfide contours, indicating possible control by the dis- solved sulf i de 114]. In a second study a limited number of samples were collected from the area off the San Onofre Nuclear Generating Station (SONGS), Figure 5. On 5 July 1973, four short piston cores were collected by scuba divers from a small private boat. SONGS was shut down for refueling during this period, and there had been no warm water discharge for 34 days prior to sampling. Two cores, I and II, were taken in 8 m of water within 10 m of the discharge. At 10 and 20 cm below the sur face, respectively, there were black zones of iron sulfides overlain by tan beach sands. The sediment consists of coarse sand and coarse silt. In such coarse sediments it is impos sible to squeeze by conventional methods because the water drains out immediately upon transfer. IAEA-SM-180/37 277 TABLE 1. Pore water data from Whites Point, Each stati on is an average of four analysis. STATION DEPTH pH Eh EH^S 's = (Ft.) (mV, pH 8 ) (mV) (^M) 1A 1000 7.62 -32 0 IB 500 7-72 -51 -- 0 1C 200 7.92 -45 0 )D 100 7-87 -51 -- 0 2A 1000 7.44 -401 0 0.3 2B 500 7.46 -385 -30 0. 1 2C 200 7.46 -388 -15 0. 1 2D 100 7-55 -331 -30 <0.01 ЗА 1000 7-59 -303 -4 1 0 3B 500 7-65 -307 -38 0 3C 2 00 7.70 -297 -22 0 3D 100 7.73 -297 -58 0 4A 1000 7-42 -288 -65 0 4B 500 7.46 -267 - !5 0 4C 2 00 7.46 -372 -144 0.01 4D 100 7-59 -309 -94 0 5A 1000 7.42 -265 -56 0 5B 500 7.45 -237 -28 0 5C 200 7.34 -477 - 180 50 5D 100 7.53 -225 -54 0 5.5C 200 7.25 -577 -- 1350 6A 1000 7.77 -72 0 6B 500 7.73 - !96 0 6C 200 7.09 -496 3400 6D 100 7.71 -328 -- 0 7A 1000 7.47 -383 -127 0 . 1 7B 500 7.40 -343 . -91 <0.01 7C 200 6.94 -567 -300 4500 7D 100 7-53 -346 -196 <0.01 8A 1000 7-58 -324 -118 0 8 B 500 7-57 -297 -93 0 8C 200 7.46 -457 -2 12 7 8D !00 7.60 -364 -145 <0.01 9A 1000 7.3 -272 -30 0 9B 500 7.3 -276 -62 0 9C 200 7.3 -325 -124 0 9D 100 7.2 -378 -158 0 . 1 10A 1000 7.3 -279 -66 0 )OB 500 7.46 -268 +63 0 IOC 2 00 7.52 -261 +39 0 10D )00 7.46 -3!3 -85 0 278 KALIL FIG. 6. Total Hg S concentration in top 5 cm of sediment at Whites Point.Contours at 10* spacing. To collect pore water from such samples it is necessary to plug the end of the core under water with a squeezer end cap as described by KALIL and GOLDHABER [17]. When the core is retrieved on deck, the pore waters can be collected by allowing the pore waters to drain through the end cap. Two additional cores were taken as controls, approxi mately 1 km north and south of SONGS in 10 m of water. These sediments were fine grain sand and coarse silt, more cohesive and capable of being squeezed. Results of pore water analyses for these two cores are listed in Table II. IAEA-SM-180/37 279 FIG. 7. Eh in top 5 cm of sediment at Whites Point. Contours at 50 mV spacing. The cores are shown schematically in Figure 8. The black zones in cores I and II are iron sulfides, which indicates a region of sulfate reduction. The North and South cores are light to dark gray, with no smell of hydrogen sulfide. They display normal pore water pH and low but positive Eh values, Table II. Measuring dissolved hydrogen sulfide in these cores is important because it is this form of H2S which is toxic to bottom fauna. H2S in pore waters, however, is only a tran sient stage in the process of sedimentary pyrite (FeS2) formation. Sulfate is bacterially reduced to dissolved sulficb, 280 KAUL TABLE II. Data from cores at San Onofre Nuclear Generating Station Depth %C S" pH Eh EH,S (cm) (ppm)* (mV) (mV) (№) NORTH CORE 0- 7 0.115 1.8 7.68 + 22 -145 0 7-13 0. 193 1.4 - 7.68 + !2 -143 0 13-20 0.237 2 1.5 7-76 + 41 -142 0 20-25 0 . 127 26.3 7.68 + 34 -141 0 25-30 0.3 19 75.4 7-69 + 21 -140 0 30-35 0.429 94.2 7-64 + 58 -133 0 35-40 0.415 51-3 7-62 + 48 -135 0 40-45 0. 195 9.5 7.62 + 100 -127 0 45-50 0.070 8.1 7.62 + 122 -124 0 SOUTH CORE 0- 7 0. 125 27-2 7-07 + 51 - 81 0 7-14 0 . 156 66.9 7- 13 + 49 - 93 0 14-19 0. 168 113.8 7.07 + 36 -121 0 19-24 0 . 109 36.9 7 . 14 + 136 - 48 0 24-49 0. 12 1 60.8 7. 18 + 141 - 42 0 29-34 0 . 116 38.9 7. '2 + 151 - 37 0 34-39 0. 166 44.9 7. 13 + 168 - 33 0 39-44 0. 182 43.9 7.22 + 168 - 38 0 CORE 1 CORE 11 Depth S Depth %C s" %C (cm) (ppm)" (cm) (ppm)а 0- 7 0.640 5-00 0- 7 0.728 14.6 10-15 0.3 18 7.64 7-15 0.728 0.0 15-20 0.3 19 37-2 15-25 1.5 25-30 0.286 103.6 25-35 0 . 194 13.9 33-40 0.092 13.7 35-45 110.8 45-50 0.2 10 1 13. 1 50-54 0.069 4.8 ^ acid sol ubte sulfide IAEA-SM-180/37 281 II NORTH SOUTH which is subsequently removed from solution as metastable iron sulfides or "acid soluble" sulfides, which are nontoxic. The iron sulfides further react to form the stable mineral pyrite [20]. This removal of dissolved sulfide from solution is an inorganic process. Recent studies indicate that a high rate of sulfate reduction (sulfide production rate) is asso ciated with rapidly deposited near shore organic rich sediments. Under these conditions dissolved sulfide is generally produced faster than it is consumed and therefore accum ulates [21]. When the rate of sul fate reduction is slow, the sulfide produced is taken out of solution F IG .8. Cores from SONGS. as iron sulfides. The equilibrium concentration value of dissolved sulfide for reaction with the abundant iron oxide of sediments is below the level of detection [20]. In the cores north and south of the discharge, there was no dissolved sulfide in the pore waters based on the flow cell measurements. Pore waters were not quantitatively mea sured at the SONGS discharge, but there was a confirmed pre sence of dissolved H2S by smell in cores I and II. This pre sence of dissolved H2S is consistent with a rapid generation of dissolved sulfide. Sulfur isotopes from core I, for the acid soluble sul fide fraction, show the s32/s34 ratHt) be enriched by 40%. in the light isotope relative to sea water sulfate. This demonstrates the bacterial origin of the sulfide [20]. Metabolizable organic carbon can be argued to be the main rate limiting nutrient for sulfate reducing bacteria [20]. The cores north and south of the discharge are low, 0 . 1-0 .4%, in organic carbon, hence a slow rate of sulfate reduction. Cores I and II have 0.6-0.7% carbon at the sur face and show a carbon decrease with depth. This relatively high carbon content at the surface is sufficient to account for an increase in the rate of sulfate reduction, whereas the decrease with depth may indicate decomposition of the organic matter during microbiological activity [2], or just pre reactor levels of the organic content of the sediment. The origin of the excess organic matter at the discharge is most probably from heat susceptible organisms, such as Zooplankton, which are killed during entrainment. This particulate matter then settles to the sediment surface and becomes substrate for bacteria. The heated water does not 282 KALIL act directly on the sediment. The thermal discharge at SONGS is more buoyant than the cooler bottom waters and never reaches the bottom. Sulfate reduction poses no serious threat at SONGS be cause of the rapid current flow along the coast. Surface produced H2S will probably be rapidly oxidized. Reactors situated in bays and areas of restricted water circulation could pose a serious pollution problem to life in the sedi ments. 4. CONCLUSIONS A technique was developed for monitoring HgS, pH, and Eh, which has advantages over previous approaches. The concept of pore water analysis may also be applied for assay of dissolved radionuclides without interference from the solid phase, or for other uses. The above technique was applied in two environments po tentially subject to the influence of man. The effects noted are consistent with an increased concentration of dissolved sulfide in pore water resulting from man's activity on land. A program, as described here, should be instigated to monitor the sediments around reactor discharges for hydrogen sulfide production and other biochemical processes in the sediment. The methods presented above for removal of pore waters with subsequent electrochemical analysis are an effec tive and simple means for studying the biogeochemical per turbations of the sediments. ACKNOWLEDGMENTS I would like to thank Drs. I. R. Kaplan and M. B. Goldhaber for their discussion and guidance. Mr. D. Hotchkiss and J. Meistrell of the L. A. County Sanitation District were responsible for collecting samples off Whites Point. Drs. B. Méchalas and K. Muench of Southern California Edison arranged for sampling at SONGS. Support for this project was provided by the U. S. A. E. C., contract no. AT(04-3)-34 PA 134. REFERENCES [1] SOUTH. CALIF. EDISON CO., San Onofre Nucfear Generating Station Unit I, Applicant's Environmental Report, Oper ating License Stage, 1973. [2] BERNER, R. A., Principles of Chemical Sedimentology, McGraw-Hill, New York (1971) 240. [3] STANIER, R. V., D0UD0R0FF, M., ADELBERG, E. A., The Micro bial World, Prentis Hall, New Jersey (1970) 214. [4] GALLOWAY, J. N., Man's Alteration of the Natural Geochemical Cycle of Selected Trace Metals, Thesis, Chem. Dept., UCSD (1972) 143. IAEA-SM-180/37 283 [5] KIONTZ, G. H., BENDELE, R. A., Histopathological Analysis of Fin Erosion in Southern California Marine Fishes, South. Calif. Coastal Mater Research Project (1973). [6] ISACCS, J. D., The Ecology of the Southern California Bight: Implications for Mater Quality Management, SCCMRP (1973) 531. [7] BERNER, R. A., Chemical diagenesis of some modern car bonate sediments. J. Sei. 264 (l966) 1. [8] PRESLEY, B. J., KAPLAN, I. R., Changes in dissolved sul fate, calcium and carbonate from interstitial water of near-shore sediments, Geochim. et Cosmochim. Acta, 32 (1968) 1037. [9] BEN-YAAKOV, S., pH buffering of pore water of recent anoxic marine sediments, Limnol. and Oceanog. 18 1 (1973) 86. [10] BERNER, R. A., SCOTT, M. R., THOMLINSON, C., Carbonate alkalinity in the pore waters of anoxic marine sediments, Limnol. and Oceanog. 1_5^ 4 (1970) 544. [11] EMERY, K. 0., RITTENBERG, S. C., Early diagenesis of California basin sediments in relation to origin of oil, Bull. Amer. Assoc. Petrol. Geol. 36^5 (1952) 735. [12] WHITFIELD, M., Eh as an operational parameter in estuarine studies, M 4 (1969) 547. [13] NEDWELL, D. B., FLOODGATE, G. D., Temperature-induced changes in the formation of sulfide in a marine sediment, Mar. Biol . 1_4 1 (1972) 18. [14] BERNER, R. A., Electrode studies of hydrogen sulfide in marine sediments, Geochim. et Cosmochim. Acta, 2_7 (1 963) 563. [15] SUTHERLAND, J. C., Inst. Environ. Sei. Tech. Meet. Proc. (1970) 298. [16] ORLAND, H. P. Standard Methods for the Examination of Mater and Wastewater, American Publich Health Associa tion, Inc., New York (1965) 429. [17] KALIL, E. K., GOLDHABER, М. В., A sediment squeezer for removal of pore waters without air contact, J. of Sed. Pet. 43 2 (1973) 553. [18] CLINE, J. D., Spectra photometric determination of hydro gen sulfide in natural waters. Limnol. and Oceanog. 1_4 3 (1969) 454. [19] EDMOND, J. M., High precision determination of titration alkalinity and total carbon dioxide content of sea water by potentiometric titration, Deep-Sea Res. 1_7 (1970) 737. [20] GOLDHABER, M. B., KAPLAN, I. R., The sedimentary sulfur cycle, Ch. The Sea 5^, Wiley Publishing Co., (in press). [21] GOLDHABER, M. B., Ph.D. Thesis, University of California, Los Angeles (1973). IAEA-SM-180/26 INTEGRAL DOSIMETER STUDY OF GAMMA-RAY DOSE IN THE VICINITY OF A NUCLEAR REACTOR G. VASILEV, G. FILEV Institute of Radiobiology and Radiation Hygiene, Sofia, Bulgaria Abstract INTEGRAL DOSIMETER STUDY OF GAMMA-RAY DOSE IN THE VICINITY OF A NUCLEAR REACTOR. The authors describe a method of measuring gamma-ray dosç due to the release of radioactive inert gases in the vicinity of a nuclear reactor. The method is based on the use of high-sensitivity integral gamma-ray dosimeters. These instruments were developed by combining gamma-ray scintillators with dosimetric photo emulsions. The authors describe the working principle of the dosimeter and discuss such performance characte ristics as sensitivity, relationship between blackening and exposure dose, efficiency as a function ot the gamma- ray energy etc. The integral dosimeter was used to study gamma dose distribution in the vicinity of the IRT-2000 research reactor. Radiation conditions around the reactor are analysed on the basis of gamma-ray background counts made over a number of years. It is well known that nuclear reactors em it sm all amounts of radio active rare gases into the atm osphere. Among these are *'Ar, from acti vation of "A r, and the radioactive gas products generated by splitting krypton and xenon isotopes [1, 2] . The dose due to m ost radioactive inert gases in the atm osphere is calcu lated as an external dose, with negligible contribution to internal irradiation; the only exception is ^Kr. For m easurem ent purposes, a man is considered as being placed in a boundless hem ispherical radioactive cloud [3] . Consequently the radioactive gases surrounding a nuclear reactor can be controlled by m easuring the gam m a-ray dose at definite intervals of time with the help of appropriate integral dosim eters [4], rather than by determ ining gas concentrations. Studies of rare gas concentrations as well as gam m a-ray doses sur rounding experim ental and power nuclear reactors have been carried out by many authors. The data have shown that the magnitude of such gam m a- ray exposure is, as a rule, very low, and is com parable in order of m agni tude with that of the natural background gam m a radiation [ 5]. This, therefore, defines the fundam ental requirem ent for an integral dosim eter — a sufficiently high sensitivity. Our dosim eters were made by combining Nal(Tl) gam m a scintillators, having the dim ensions of 25 mm dia. X 12 mm, 25 mm dia. X 25 mm, 25 mm dia. X 40 mm or 50 mm dia. X 50 mm, and photodosim etry film s m anufactured by Gevaert. A fter ensuring that there w as good optical contact between the scintil lator and the film , the dosim eters were placed in light-tight and humidity- proof cassettes. For calibration purposes, the dosim eters were placed on open ground and irradiated using standard gamma sources containing "°Tm , ^Co, *^Cs, "M n or "C o (Table I). This enabled us to obtain the relationship 285 286 VASILEV and FILEV TABLE I. GAMMA-RAY ENERGIES OF NUCLIDES USED FOR CALIBRATION Radionuclide (MeV) Thulium-110 0.084 f 0.123 Cobalt-57 [0 .1 3 7 Caesium* 13*7 0.662 Manganese-54 0.842 f 1.172 Cobalt-60 [1.333 Y-RAY ENERGY (MeV) between the effectiveness of registering the gam m a rays and the energy of irradiation. A typical plot of such a relationship for the integral dosim eter equipped with a scintillator of 25 m m dia. X 25 mm is shown in Fig. 1. The effectiveness of the integral dosim eters for registering a compound energy spectrum was checked by using standard radium -266 sources. To determ ine any dependence between the degree of blackening of the film , S, and the exposure dose, D, the dosim eters were irradiated with doses of up to 100 mR. To ensure the typical dose rate due to the natural gam m a background (15 X 10"^ m R/h) fell in the middle of the dose-rate range studied, the dose rates, P, were varied between 25 X 10*^ and 100 X 10"' m R/h. IAEA-SM-180/26 287 DOSE.DÍmR) FIG. 2. Relationship between degree of blackening, S, of the film and the exposure dose, D, at a dose rate of 30X10*3 mR/h using a standard ^ R a source. It was established that in the dose-rate range considered, there was no dependence between degree of blackening of the film and dose rate, i. e. : S = S(P) = constant The dependence between the degree of blackening and exposure dose for the different scintillators using a Ra irradiation source giving a dose rate of 30X 10*^ m R/h is shown in Fig. 2. The data allow one to conclude that, in order to obtain a distinct degree of blackening of 0. 3 to 0. 5 for a 25 mm dia. X 25 mm scintillator for a dose rate of approxim ately twice that of the natural gam m a background, the du ration of irradiation m ust be 30 to 90 (24 h) days. This type of integral dosim eter was used to study the gam m a-ray dose distribution in the vicinity of the 2 MW(th) N uclear R esearch Reactor UPT-2000, a w ater-w ater heterogeneous reactor of reservoir type. A total of 120 dosim eters were placed in pairs at 60 points around the reactor. The points were chosen to be up to som e 5 km away from the reactor. D osim eters were placed in the shade at heights of 1 to 5 m above the ground. The irradiation periods were 30 to 90 days. A number of control dosim eters were irradiated at the sam e tim e and for sim ilar periods in regions having only a natural gam m a background. The whole experim ent extended over a period of three years. We were able to establish that there was a very slight increase of gam m a activity above that of the natural background in the direction of the prevailing winds in the region downwind of the reactor. The m aximum dose 288 VASILEV and FILEV rate w as found som e 2 50 m from the reactor: the figure w as (32 ± 4) X 10*" m R/h of which the natural gam m a background accounted for (15. 7 ± 0. 4) X 1 0 * " m R / h . We consider that the method of m easuring gam m a-ray dose described in this report is well suited for m onitoring in the regions surrounding a nuclear reactor, and it is planned that this method w ill be used for studies around the nuclear power station that is under construction in Bulgaria. REFERENCES [1] Siting of Reactors and Nuclear Research Centres ( Proc. Symp. Bombay, 1963), IA E A , V ienn a (1963). [2 ] Techniques for Controlling A ir Pollution from the Operation of Nuclear Facilities, Safety Series No. 17, IAEA, Vienna (1966). [3 ] Basic Safety Standards for Radiation Protection - 1967 Edition, Safety SeriesNo. 9, IA E A , V ienn a (1967). [4 ] Nocken, J., Alderhout, J., Health Phys. 9^6(1963)655. [5] БОЧВАР, И.А. и др. В сб. Вопросы Дозиметрии и Защиты от Излучения, Атомиз- дат, Москва ( 1966) 90 . DISCUSSION С. К. FITZSIM M ONS: In the USA we have experienced therm al fogging of our gam m a-sensitive film , especially in the case of long-term m easure m ents, and I wonder if you have had the sam e difficulty with your Nal(Tl) film system ? G. VASILEV: Yes, we used to have this problem. We tried out various kinds of film and found that Gevaert photodosim etry em ulsions gave the best results. The choice of scintillator is also im portant and the dosim eters should be kept in the shade. IAEA-SM-180/36 EARLY SURVEILLANCE AROUND COASTAL NUCLEAR INSTALLATIONS* T.R. FOLSOM, V.F. HODGE Scripps Institution of Oceanography, La Jolla, Calif., United States öf America Abstract EARLY SURVEILLANCE AROUND COASTAL NUCLEAR INSTALLATIONS. Several biological systems have been used to detect extremely small amounts of radioactivity emanating from nuclear reactors that discharge directly into marine environments. Special attention has been given to plutonium. Inorganic absorbing systems, including certain ferrocyanides, provide convenient and sensitive INTRODUCTION Many large nuclear installations are expected to appear along the sea coasts where large populations are growing. Planning disposal systems that w ill minimize hazards and economic losses requires thorough under standing of the probable behaviors of several classes of radionuclides that may find their way into coastal w aters. Fortunately, some useful informa tion concerning environmental responses can be learned through careful studies made long before pollutions have reached levels of any reasonable concern to human w elfare. Early information of this sort, however, requires the use of highly specialized equipment and techniques not generally employ ed by those interested only in satisfying existing legal restrictions con cerning the use of the environment. Moreover, coordinated efforts of biol ogists, chemists and physicists fam iliar with oceanographic problems are r e q u i r e d . This report w ill describe some sm all changes that have been observed following installation about 5 years ago of a 400 MW coastal nuclear power plant at San Onofre, California, about 7 0 Km north of Scripps Institution of Oceanography. Nuclide levels of four different radioactive elements observed at San Onofre w ill be compared with contemporary coastal and oceanic background levels which largely have depended, in a more or less * This work was supported by the USAEC. contract No. AT(04-3)-34, P.A. 71-17, and the US Office of Naval Research, contract No. USN N00014-69-A-0200-6011. 289 290 FOLSOM and HODGE complex way, on global fallou t. A few comparisons with levels reported for other coastal installations w ill be made; however, most attention at Scripps has been given to determining which nuclides associated with nuclear power production are accumulating in which marine species, and in developing extremely sensitive methods for follow ing radioactive changes in the several marine niches along coasts that are relatively openly- exposed to the ocean. EXPERIENCES WITH RADIOACTIVE SILVER NUCLIDES The changing concentrations of two silver nuclides attributable to global fallout have, since 1964 [l-5], been studied at Scripps in numerous oceanic and coastal sam ples. Because of these background records, it has been possible to identify extremely sm all silver nuclide anomalies in marine samples collected near the San Onofre Power Plant. Anomalous silver nuclide ratios have been found near this plant, and also higher absolute concentrations of silver-110m relative to concentrations found in comparable biological samples collected 70 Km away from the plant, i.e ., near La Jo lla. Table I illustrates that at least three different marine species, that accumulate silver nuclides readily, can be collected in the vicinity of this power plant, and also at points some distance away, so as to be useful for searching for local anom alies. The more common brown algae were found to be not useful for silver studies; however, a red alga, a m ollusc, and a surf grass were found to be effective and convenient. Figure 1 and Table II illu strate how much more sensitive and definitive are comparisons made between silver nuclide ratios rather than between ab solute concentrations. It should be noticed in Figure 1, that, in several types of reference samples collected between 1970 a n d 1973? silver-llQ m / silver-108m ratios between 1 and 3 were measured. However, biological specimens collected near San Onofre exhibited a ratio of about 200. It should be noted in Table II that extremely high absolute concentrations of both silver nuclides sometimes may be accumulated in organisms that appar ently have experienced no other artificial nuclide source but that of global fallout. Yellowfin tuna caught in the central N. Pacific sometimes contain nuclide ratios near unity but absolute concentrations many times higher than other oceanic fish. These sm all traces of silver generally were determined by simply drying selected tissu es and use of a summing-coincidence (Nal) gamma spectrom eter [6Ц, or a two-dimensional gamma spectrometer [4], or both. It was later found that radiosilver traces may be rapidly and quantitatively electroplated from plant or animal tissues after a silver carrier has been added to a slurry ground in a Waring Blendor [ 7 ]. This form of concentrating might be especially advantageous for those wishing to concentrate silver nuclides to the sm all volumes that are most suitable for germanium-lithium type of gamma spectrom e t e r s . COBALT NUCLIDES Several gamma em itting cobalt radionuclides also are useful for distin guishing the very earliest traces of power plant nuclide waste from fallout backgrounds. This may be seen from Table I where cobalt-60 and cobalt-58 concentrations in samples collected near San Onofre and others collected IAEA-SM-180/36 291 Table I. Small coastal disposals of artificial radioactivities from the San Onofre Nuclear Power Plant evidenced by relatively high ^ concentrations in selected marine organisms collected locally. pCi/Kg wet weight^ 6 0 110m. D a t e C o 5 ' c o A g S a m p le C o l l e c t e d ( 5.3 y r ) (72 d ) ( 2 5 3 d ) Sea hare (Aplysia I ß / 12/70 8 5 2260 1 1 4 californica) a 10/ 3/71 63 734 78 sh ell-less mollusc 22/ 6/ 7I 16 87 16 collected near 4/IO/ 7I 4 5 125 4 9 S a n O n o f r e 7 / I /72 7 12 12 18/ 12/72 34 6 4 1 9 5 collected near 22/ 6/71 2 < 1 < 1 L a J o l l a Agar Agar (Gelidium I 2/I 2/ 7O 2 4 313 3 5 sp.) a red alga 10/ 3/71 5 4 4 6 collected near 2/II/ 7I 4 11 13 S a n O n o f r e collected near 22/ 6/ 7I 5 < 1 < 1 L a J o l l a S u r f g r a s s 10/ 3/71 3 1 4 2 6 4 2 (Phyllospadix sp.) a 2/II /71 21 4 2 25 flow ering plant 1 7 / I /72 9 1 5 1 4 collected near 18/ 12/72 5 8 160 85 S a n O n o f r e collected near 10/ 3/71 9 < 1 < 1 L a J o l l a Ълэг comparison, concentrations are given for the same species collected .at La Jolla, Ca., 70 Km away. Counting errors of measurements are 10% or better. H alf-lives of these three artificial nuclides are given in parentheses. Other nuclides also have been detected. 70 Km away at La Jo lla (Scripps Institution) are compared. In the case of cobalt, as with silver, red algae and green surf grasses were found to accu mulate higher concentrations than do brown algae; and in general, more intense cobalt concentrations may be found in m olluscs, but these are more difficult to collect routinely. Coincidence-gamma spectrom eters sim ilar to those used for radiosilver determ inations were also used for sm all traces of cobalt. No chemical separation is needed but drying or ashing adds much to convenience and sensi tivity. Where larger fish are available, it is often profitable to dissect 292 FOLSOM and HODGE YEAR FIG. 1. Distinctive silver nuclide ratios found near a coastal nuclear power plant at San Onofre, California. The much different ratios in fallout for different years and in different biological samples are shown (see Table II). Table II. Silver-UQm/Silver-108m ratios at San Onofre and elsewhere. 110m, , pCi/Kg wet Ag/ 110m. 108m, 108m, Location Date Species Ag Ag Ag San Onofre 12/70 se a hare 3 150 0.6 200 10/71 Aplysia calif. 3 49 (- 0 .23) 213 12/72 (whole) 3 195 0.9 217 San Diego 7/64 albacore tuna 26 81 4 .6 17.6 (100 m iles 7/65 Thunnus alalu n g a 26 100 7-5 I 3.3 w est) 7/68 ( liv e r s ) 30 8.7 5-9 1.5 7/70 59 5.8 4 .6 1-3 7/71 76 4 .9 4.2 1.2 7/72 60 10.1 4.8 2 .1 Hawaii 9/70 yellowfin tuna 2 123 113 1 .1 Thunnus a lb acares ( l i v e r s ) La J olla 10/70 m ussel 30 0 .7 0.5 1 .4 1/73 Mytilus calif. 30 1 .9 0.6 3 .2 2/73 (whole) 30 1.5 0.7 2 .1 IAEA-SM-180/36 293 and count activity only in the liver tissue. The half-life of cobalt-58 (72 d) is a convenient one; when normalized against cobalt- 60, the cobalt-58 activity provides a sensitive indicator ratio. It is necessary, of course, to collect control samples from other areas exposed only to global fallout. Moreover, it is necessary to repeat collections of control samples frequent ly enough to keep up to date the record of the global fallout background. Experience has shown that the oceanic biosphere readily detects new inputs of nuclear fallout, especially silver and cobalt nuclides, and since 1968, relatively large new fallout inputs of zinc-65 have been detected in the ocean by inspection of biological concentrators [8]. CAESIUM NUCLIDES Consideration of caesium nuclides can hardly be avoided; however, surveillance techniques different from those that succeed with the transi tion metals must be adopted. The long life of caesium-I37 and its persist ence in upper layers in the ocean [8] provide for large-scale normalization against the changing global fallout background, especially its large scale and long range aspects. On the other hand, the short life of caesium-134 (2.1 yr) and the presence of this nuclide in power reactor wastes may give evidence of a change in coastal disposal rates. Table III illustrates the wide range of caesium-137/caesium-134 ratios that have been observed, varying from about 1.4 in samples collected at San Onofre to greater than 100 in recent oceanic waters exposed only to f a l l o u t . No biological system has been found that accumulates caesium as effec tively as those that may be exploited for collecting large samples of silver, cobalt and plutonium nuclides. Biological accumulations of caesium seldom are more than 100 times the concentration in sea water. Moreover, one must normalize the surveillance samples with care, because the alkalies vary considerably within apparently identical biological samples. Normalization against potassium rather than against sample weight appears most appropriate; for example, Figure 2 illustrates the parallel behavior of caesium and potassium in different tissues of an oceanic tuna. Table III. Caesium-137/Caesium - 134 ratios at San Onofre and elsewhere. Location Date Sample ^ c s / ^ c s P a c i f i c Ocean"*" 1 9 7 1 open ocean sea water > 1 0 0 San Onofre, Ca. 7 /7 1 discharge water 1.4 Indian Point, N.Y. I 9 7 O f i s h 1 . 0 United Kingdom3 II / 7 0 coastal sea water 7 . 0 ^ T y p ical I 97I sea water near California had about 0.3 pCi/i ^Cs. Power re a c to r on Hudson R iver estu ary [91- Measured at Scripps for the IAEA sea water reference sample intercali b r a tio n s , 1971 [ЮЦ. 294 FOLSOM and HODGE '^Cs(pCi/w eï kg) Several inorganic agents are available for selectively concentrating caesium nuclides from sea water to levels > 5 0 000 times the levels in sea w a te r . When r a t i o s o f gamma n u c lid e s a re d e s ir e d , th e d ir e c t gamma m eas urements introduced by Boni [ll] of caesium absorbed from sea water onto granular ferrocyanides are effective. When absolute levels of environmental caesium concentration are required, however, normalization against natural caesium collected on the absorber [12] can be used and is especially effec tive for detecting extremely small traces of radiocaesium. Large samples of caesium nuclides can be collected routinely by pumping large volumes of water through granular absorber beds. This monitoring can be done automatically at points near a nuclear installation. tAEA-SM-180/36 295 PLUTONIUM MARINE POLLUTION STUDIES The first measurements of plutonium in sea water and marine organisms made by Pillai at Scripps in 1964, disclosed the fact that many marine plants had strong affinities toward this element [ 13 ]. This suggested the surveillance of an environment by analysis of samples of its plants. In fact, the levels of the minute traces of plutonium -238 in local waters.were established by measuring the plutonium-238/plutonium -239 ratio in the ashes of giant brown algae. This was much simpler than precipitating plutonium directly from $00-litre samples of sea water as was done by Miyake in 1968 [l4]. However, Miyake did not have available in the open Western Pacific suitable algae for conveniently concentrating his samples. Interest in possible coastal plutonium contaminations increased when it became clear just how much of this nuclear material was being used and how much larger would be the amounts to be transported when its use for fuel was fully exploited. Inspections were made of several convenient marine organisms [15,l6] toward selecting effective plutonium monitors. Plutonium concentrations were found in numerous marine species, but the concentrations varied widely and were difficult to describe and compare because of the many morphological, chemical, and physical différencies amongst marine species. Neither sample normalization by wet weight, by dry weight, nor by ash weight was satisfactory. Some extreme cases were recognized however. The lowest concentrations 0.001 pCi/wet Kg) were generally found in muscle tissues, especially in large fish; much higher concentrations were found on large algae and surf grasses especially on the thinner blades, 0.2 to 2 pCi/wet Kg for example, in bulk samples of brown algae [15,16]. When it became evident that plutonium accumulated upon surfaces of many different organisms, the concentrations at several depths below the surface were quantitatively compared in typical specimens of some of the giant brown algae that are convenient to collect along the California coast [l6] at 10-60 meter depths beyond the surf zone. It was found that thin layers (0.2 mm) scraped from outer plant surfaces often had plutonium concentrations more than 200 times higher than what was found deep inside the plant. Clearly then samples must somehow be normalized with reference to surface areas. Fortunately, in many plant species, major synthesizing blades of fairly uniform thickness may be collected, so that areas may be estimated from sample weights. Moreover, this normalization problem largely disappears whenever only the nuclide ratios in the environment must be determined by biological samplings. Even when only nuclide ratios are required, there are several choices th a t may be made toward ob tain in g the most se n s itiv e estim ate s o f the more recent environmental conditions through sampling plant tissues. Amongst the factors that are involved are the age and growth rate of the tissues. Growth rates of few species have been studied, however, the giant brown alga Macrocystis pyrifera is of commercial importance and some aspects of its growth therefore have been recorded. It grows near the San Onofre power plant and also grows several hundred miles to the north and south. The plants grow in "forests" beyond the surf line, sending up numerous slender stipes 15-30 meters long tying together a hundred or more blades, each buoyed by a small bulb. The distal blades grow extremely rapidly, and the oldest, deepest, blades are seldom more than six months old. In a recent study at Scripps [17], radioactive concentrations on numerous blades along stipes were compared. At distances of about l/2 meter from the growing (distal) end of the stipe to about 10 meters nearer to the older (holdfast) end, the surface concentrations of plutonium -239 were found to increase at a fairly uniform rate from 1.0 x 10*17 curies/cm^ to 4.0 x 10*17 curies/cmS. Since all of the leaves, on the portions of stipe that were studied, had 296 FOLSOM and HODGE developed within a period estimated at about 120 d ay s, i t may be concluded that plutonium-239 deposited at a fairly uniform rate of about 3.0 x 10*17/120 or 2.5 x 10*^9 curies/day-cm2 over the blade surfaces. Since typical blades are 0.4 mm thick, there are 50,000 cm2 of area/wet Kg of wet blade sample; therefore, 2.5 x 10*19 x 5 x 10^ =1-3 x 10*1^ curies are accumulated per day per Kg of wet sample. Young (ten-day old) blades having 1.0 x 10*17 curies/cm^ or 0.5 pCi/Kg have commonly been observed. Moreover, o ld er b lad e s with roughly 4 tim es higher concentrations might have been selected where the most recent environmental conditions were not the main objective. Recent clean sea water samples collected near Scripps may be expected to contain about 6 x 10'^ pCi plutonium-239/litre and about 3 x 10*5 pCi plutonium -238/litre. Thus sampling 1 Kg of mature Macrocystis kelp blades corresponds to sampling 4 x O.5/6 x 10*4 or about 33OO litres of sea water. For this reason, only about 100 g of wet leaves are now used at Scripps for monitoring coastal plutonium-239, and larger samples are needed only when the less abundant plutonium-238 nuclide is to be determined with great p r e c isio n . Table IV. Plutonium-239/Plutonium-238 ratios at San Onofre and elsewhere. pCi/Kg wet"** 239pu/ Location Date Sample < ^ P u 3°Pu 238^ San Onofre З/71 su r f g ra ss 0 .27±0.02 o.oi5±o.oo5 18 ±6 Phyllo spad ix sp. San Onofre 5/73 brown algae i . 33± o .o 6 0.074±0.021 18+5 Macrocystis sp. La JollaS/73 brown algae i . 50±0.05 0. 068+0.016 22±5 Pelagophycus sp. La J o l l a II/ 7I brown algae 2 .l6 ± 0 .1 5 0.12±0.04 18+6 2 Pelagophycus sp. La J o l l a 7/ 7I Calif, sea water 0 .0006+0.0002 - - La Jo lla ^ 4/64 brown algae 0.45±0.02 o.oi5±o.oo5 30± i0 Eisenia sp. La J o lla ^ . Calif, sea water o.ooo4±o.oooi - - Irish Sea ',7 ^ " s e a weed" 2330+50 330+10 6 .8± о .з 40°N5 12/64 HASL soil, before 38 SNAP acciden t 40°N^ 1970/71 HASL soil, after 27+4 SNAP acciden t 40°N 1971 HASL surface air 1 to 13 sam plings ^One counting error (± one standard deviation). ^See Ref. [l 6]. ¡¡See Ref. [ 13]. Measured at Scripps for the IAEA sea weed reference sample intercalibrations -1972 (technical report). ?See Ref. [18]. ySee Ref. [19]. 'Assuming 10:1 wet weight/dry weight. IAïA-SM-180/36 297 Table IV lists a few plutonium-239/plutonium-238 ratios determined, near the San Onofre power plant and elsewhere for comparison. Some useful comparisons are afforded by this data were collected during early procedural development studies. However, demonstration of an anomalous plutonium ratio near the power plant has not yet been made. Nevertheless, very recently developed techniques now appear to make tests of this sort possible and also relatively simple to carry out. DISCUSSION Relatively few people who must develop coastal pollution surveillance methods w ill be afforded the convenience of a giant alga for their studies. However, there are various other effective marine concentrators although few have been studied in any detail as indicator organisms for specific elements. A great deal more must be learned even about the algae and coastal grasses, how fast they grow, which nuclides collect selectively in which species, and at what periods of their life cycles. It was of interest to find, soon after it was noted that plutonium deposited from sea water rigorously proportional to areas of interfacial surfaces that were exposed, that another alpha emitting radionuclide accumu lated proportionally in many of the same surfaces, but at a much higher rate. This is, on one hand, a nuisance requiring careful chemical purifi cation of the plutonium before counting is done. On the other hand, this second nuclide, natural polonium-210, is so abundant and so easy to purify and analyze that its trends and its distributions in the marine environment can suggest some of the probable behavior of the much-harder-to-measure plutonium nuclides. For example, the activity of polonium-210 has, in studies of hundreds of brown algal blades, been found to deposit upon a unit are a 230±50 times as intensely as plutonium-239 deposits, whether the blade is 10 days or 120 days old. Similar parallel deposition rates of polonium and plutonium are evident, although still not fully studied, in connection with several other sea water/biota interfaces common in the marine world. Why two, so unlike, elements should behave this way in the ocean is philo sophically exciting to some of us. Other people too may discover that there is a convenient way for studying growth and deposition rates; measuring polonium-210 is a relatively simple and easy way of learning something about the behavior of extremely small traces of plutonium. A warning should be given, however, that no further simplicity in the polonium-210/plutonium -239 ratio has been observed further up in the trophic web. Plutonium usually concentrates very little in tissues of higher organisms. Edible muscular portions of fish concentrate plutonium activities only 2 or 3 times the levels in the sea [13 ] while on thin algal blades the wet weight concentration may be a thousand-fold higher [15,16]. On the other hand, many tissues of higher organisms build up exceedingly high concentra tions of the natural alpha emitter polonium-210. Fortunately, most of these tissues are seldom eaten by humans. F o r example, an extreme concentration of polonium-210 has been observed in tissues of the little studied and almost never eaten "pyloric caecal mass" of one of the large oceanic fish [20]. Here the polonium-210 activity has been found to accumulate 2.2 million times the typical sea water level; and the radioactive dosage burden (from 5 million volt alpha particles released inside these tissues) is suspected to be the highest ever encountered in bulk tissues of an organism living in a truly natural environment. 298 FOLSOM and HODGE CONCLUSIONS Extremely early signs of radioactive pollutions associated with coastal nuclear installations can he detected and identified at levels just above the contemporary backgrounds caused by global fallout. Special techniques are required, however. Careful studies of the biological concentrators present in the environ ment as to growth habits and as to behavior toward specific elements at trace levels may uncover factors leading to extremely sensitive--and sometimes also convenient--surveillance procedures. Many of the environmental factors needed for developing early surveil lance methods are of a type much needed for understanding the ultimate capacities of the coastal environments that may be used for disposing much of the nuclear wastes of the future. ACKNOWLEDGEMENTS We thank T. Otsu of the National Marine Fisheries Center, Honolulu, for yellowfin tunas. The collections of samples from La Jolla and San Onofre by J. Grander and W. Nichols of Scripps Institution of Oceanography are grate fully acknowledged. REFERENCES [1] FOLSOM, T. R ., YOUNG, D . R ., Silver-llOm and cobalt-60 in oceanic and coastal organisms, Nature 206 ( 1965) 8 O3 . [2] FOLSOM, T. R., GRISMORE, R., YOUNG, D. R ., Long-lived gamma-ray em itting nuclide silver-108m found in Pacific marine organisms and used for dating, Nature 227 (1970) 94l. [ 3 ] FOLSOM, T. R ., YOUNG, D. R ., HODGE, V. F ., GRISMORE, R., "V ariatio n s o f 5^Mn, °0co^ °5 z n , HOm.Ag^ and 1^8m^g in tu n a s," Symposium on Radioecology, Third National Symposium, Oak Ridge, Tennessee, C0 NF-7 1 0 5 0 1 -P2 ( 1 9 7 1 ) 7 2 1 . [4 1 GRISMORE, R ., FOLSOM, T. R ., HODGE, V. F ., YOUNG, D. R ., A s t u d y of the radiosilver signature of the 1 9 6 1 -6 2 nuclear weapons testing period, Trans. New York Acad. Sei. ß4 (1972) 392. [5] HODGE, V. F., FOLSOM, T. R ., Estimate of the world budget of fallout silver nuclides, Nature 237 (1972) 93. [ 6 ] FOLSOM, T. R ., YOUNG, D . R ., FINNIN, L. E ., "Sum-coincidence gamma- ray spectrometry in tracing cobalt-60 and silver-llOm in marine organisms," Radioisotope Sample Measurement Techniques in Medicine and Biology, In tl. Atm. Energy Agency, Vienna ( 1 9 6 5 ) 57* [ 7 ] HODGE, V. F ., FOLSOM, T. R ., Cyanide extraction and electrodisposition of trace amounts of radioactive silver from large biological samples, Analyt. Chem. 44 ( 1 9 7 2 ) 3 8 1 . [8] HODGE, V. F ., FOLSOM, T. R., YOUNG, D. R., "Retention of fall-out constituents in upper layers of the Pacific Ocean as estimated from studies of a tuna population," Radioactive Contamination of the Marine Environment, In tl. Atm. Energy Agency, Vienna (1 9 7 3 ) 2 6 3 . [9Ü KAHN, BERND, "Radionuclides in the environment at nuclear power stations," Symposium on Radioecology, Third National Symposium, Oak Ridge, Tennessee, CONF-7 IO 5 OI-PI (l97l) 30- IAEA-SM-180/36 299 [10] FUKAI, R., BALLESTRA, S., MURRAY, C. N., "Intercalibration of methods for measuring fission products in seawater samples," Radioactive Contamination of the Marine Environment, In tl. Atm. Energy Agency, Vienna ( 1973) 3- Ell] BONI, A. L., Rapid ion exchange analysis of radiocesium in milk, urine, sea water, and environmental samples, Analyt. Chem. ß8 (1966) 89. [ 12 ] FOLSOM, T. R., SREEKUMARAN, C., "Same reference methods for determining radioactive and natural cesium for marine studies," Reference Methods for Marine Radioactivity Studies, Intl. Atm. Energy Agency, Vienna ( 1 9 7 0 ) 1 2 9 . [13 ] PILLAI, K. C., SMITH, R. C., FOLSOM, T. R., Plutonium in the marine environment, Nature 203 (1964) 568. [14] MIYAKE, Y., SUGIMURA, Y., Plutonium content in the western North Pacific waters, Pap. Meteorol. Geophys. (Tokyo) 19 ( 1968) 4 8 l. [15] WONG, K. M., HODGE, V. F ., FOLSOM, T. R ., "Concentrations of plutonium, cobalt, and silver radionuclides in selected Pacific seaweeds," Rroc. of Environmental Plutonium Symposium, Los Alamos, New Mexico, LA-47% ( 1 9 7 1 ) 9 3 . [l61 WONG, K. M., HODGE, V. F ., FOLSOM, T. R., Plutonium and polonium inside giant brown algae, Nature 237 (1972) 460. [ 1 7 ] HODGE, V. F ., HOFTMAN, F. L ., FOLSOM, T. R., Rapid accumulation of plutonium and polonium on giant brown algae, manuscript 1973- [18] HARDY, E. P., KREY, P. W., VOLCHOK, H. L ., "Global inventory and distribution of Pu-239 from SNAP-9A," USAEC Rep. HASL-250 (1972) 32 p. [1 9 ] VOLCHOK, H. L ., TOONKEL, L ., SCHONBERG, M., "R adion uclid es and lead in surface air," USAEC Ref. HASL-274 (App.) ( 1973) B98, B106. [20] FOLSOM, T. R., WONG, K. M., HODGE, V. F ., "Some extreme accumulations of natural polonium radioactivity observed in certain oceanic organisms," Proc. The Natural Radiation Environment II Symposium, Houston, Texas (in p re ss 1972). DISCUSSION G. VERGNAUD: You may be interested to know that our measurements on the species of m ollusc Patella Vulgata found on the coast north-west of La Hague in France indicated specific^°"A g activities ranging from 100 to 200 pCi/wet kg. The discharges into the sea from the centre of La Hague introduce 50-150 mCi of ^°"A g activity per month into the m arine environment, as m easured by sampling liquid effluents. T. R. FOLSOM: I am not surprised at this figure for molluscs. We shall be seeing these levels frequently in the future. P. PELLERIN: I should just like to point out that, for an equivalent amount of radioactive contamination, the reconcentrations observed in fresh water organism s are for the most part much greater than in organism s living in the sea, due undoubtedly to the carrier role played by all the m inerals in the sea. IAEA-SM-180/15 EMERGENCY SURVEILLANCE AROUND A NUCLEAR REACTOR SITE G.H. PALMER, K. E.G. PERRY United Kingdom Atomic Energy Authority, Atom ic Energy Establishment, Winfrith, Dorset, United Kingdom Abstract EMERGENCY SURVEILLANCE AROUND A NUCLEAR REACTOR SITE. Using filtration and delay beds the airborne radioactive discharges from a reactor system can. in normal operation, be reduced to a very low level. However there remains a low, but finite probability that in an accident situation a large release of airborne radioactivity could occur which might be of concern to nearby populations living in the district around the plant. The paper outlines the arrangements at a typical United Kingdom reactor site aimed at safeguarding the environment and the nearby population from the effects of sensitive indication of the wellbeing of the plant in normal operation, it can give a rapid early warning of a large release which might occur in an accident situation. Together with a supplementary emergency monitoring system, an early indication would be obtained of any release in order to initiate the emergency procedures. will be described, where the relative responsibilities and control measures for safeguarding the nearby populations are outlined. These arrangements include those for rapid district monitoring, the monitoring and control of m ilk supplies, the possible temporary evacuation of people and, if need be, the medication of people who INTRODUCTION Front the point of view of environmental pollution as a result of the discharge of gaseous waste products, nuclear power plants present far less severe problems than the equivalent fossil fuelled power station. However, as a result of operating a nuclear power reactor a very large quantity of radioactive chemical materials is produced as a by-product and if these materials were allowed to escape they could have a deleterious effect on the environment and in particular on the people living in it. The design of a modern reactor plant is such as to reduce to a very low level the release of any gaseous radioactive materials during the normal operation, and to provide means for reducing to acceptable levels the release in a large scale plant emergency. Although the design of the plant should be adequate to deal with any situation which may arise, it is still necessary to demonstrate the adequacy of the design and method of control by suitable monitoring of any potential airborne releases. It is established practice to monitor any likely gaseous discharge routes and, as a further safeguard, to monitor the environment as an additional check that no unacceptable radioactivity has escaped in any way to the environment as the result of the normal operation of the plant. Although every precaution is taken to reduce to a minimum the probability of an accident occurring, the possibility can never be eliminated, and an emergency plan is necessary to handle the situation should it arise. The equipment installed on the plant and the operational procedures used for the environmental surveillance, provides the basis for detecting and assessing the implications of a major accident and the initiation of remedial action to protect the general public. 301 302 PALMER and PERRY CONTROL OF GASEOUS WASTES The UK Atomic Energy Authority has been responsible for the basic research and development of reactor technology in the United Kingdom and operates at the Atomic Energy Establishment at Winfrith a prototype 100 MW(e) Steam G e n e ra tin g Heavy Water R e ac to r (SGHWR), th e OECD 20 MW(th) experimental DRAGON reactor and a number of zero energy experimental reactors carrying out a programme of basic reactor physics measurements. The Authority is responsible for controlling the discharge of gaseous effluents from its reactors, subject to the overall jurisdiction of two Government Departments, namely the Department of the Environment through its Alkali and Radiochemical Inspectorates, and the Ministry of Agriculture, Fisheries and Food through its Atomic Energy Section. An authorisation for the discharge of gaseous wastes is issued to the Authority for each of its nuclear sites by the two Government Departments. The terms of the Authorisation are general and require that the gaseous discharges are regulated so that "public health is not endangered" and in addition that "the best practical means" are used to keep the discharges to a minimum. The first of these requirements which used the recommendations of the International Commission for Radiological Protection as the guiding criteria must be met irrespective of costs, while the second becomes a compromise reached by agreement of the operators and controlling authorities, taking into account all economic and other social aspects of the problem. In support of the authorisation any recognised routes for the discharge of gaseous wastes, for example the ventilation stack from a reactor building, must be monitored and the results sent each month to the two Government Department Inspectors. As a further safeguard a programme of monitoring of certain agricultural indicators and food products in the neighbourhood of the reactor plant is carried out by the plant operators in collaboration with the technical staff of the controlling authorities. The type of materials monitored will vary from site to site, but generally consists of grass and milk sampling, the latter being an excellent indicator for radio iodine, strontium and caesium. The arrangements for the routine monitoring of gaseous discharges, supplemented by equipment and other monitoring surveys mentioned later, together with the routine environmental survey work, serve as the basis for the surveillance programme for dealing with an emergency situation. PLANT DESIGN AND INSTALLED EQUIPMENT Instrumentation for monitoring any radioactivity in the gaseous effluents from a reactor has been developed and installed on the Dragon and Steam Generating Heavy Water Reactors (SGHWR) at AEE, W infrith. The prime purpose of the equipment is to monitor on a continuous basis the discharges during normal operation and demonstrate they are very low and well below the design guide safety lim its agreed with the controlling authorities. With some additional features incorporated into the design, the equipment also provides a means of monitoring and assessing any large gaseous discharge which could possibly arise in a reactor accident situation. A typical gaseous effluent monitoring system as installed on the SGHW Reactor is described below. Similar systems are used on the Dragon Reactor at Winfrith and w ill be used on the Authority's prototype fast reactor now nearing completion at Dounreay in Scotland. The SGHW reactor is a direct cycle, pressure tube reactor, using light water as the coolant and heavy water as the principal moderator. The IAEA-SM-180/15 303 coolant is pumped up through the pressure tubes and partially boils as it passes over the fuel: the mixture of water and steam passing to a conventional steam drum where the steam and water are separated. The steam and any gases present pass through driers into the turbo-generator and from there to a water cooled condenser maintained under partial vacuum by a pump. The steam condenses and the pump extracts any non-condensable gases and any air which leaks into the condenser system and feeds them into an "off gas system" consisting of a 150 m long pipe and a time delay filtration bed (6 tonnes of graphite) before discharge through a ventilation stack. Air from other areas of the reactor is also discharged through the stack and Fig. 1 shows a schematic drawing of the ventilation system. The off-gases from the condenser amount to about 250-400 litres/m in of air which has leaked into the system, about 1 5 0 litres/m in of hydrogen and oxygen from hydrolysis of the coolant water and some radioactive gases. In normal operation the radioactive gases are mainly radioactive isotopes of Nitrogen (13и and anct Oxygen (1%) from the water. Should any fuel can failures occur during reactor operation, radioactive isotopes of the inert gases Xenon and Krypton also appear in the off gas system. The hydrogen and oxygen are recombined to prevent the accumulation of an explosive mixture and the residual gases pass through the 150 metre long duct to a cooler and demister before entering the graphite delay bed. The overall system introduces a delay time of about 4 hours for the Krypton isotopes and 60 hours for those of Xenon. The delay is sufficiently long to remove completely all radioactive materials with half lives up to tens of minutes. Various sampling points, shown in Fig. 1, along the "off-gas" route provide means for checking periodically the efficiency of the various filter components. Some of these sampling points are used for equipment which provides a continuous indication of the gas activity as it is being discharged, while others allow a continuous sample of the gas to be taken through a composite filte r ("MAY" pack) [1] which is removed at regular intervals and assessed in a counting laboratory. Hence while the con tinuous monitors give an immediate indication of the activity level and some indication of the composition of the discharged gases, the composite sample filter provides in retrospect, an accurate and very sensitive measure of 131l and any particulate activities discharged during a given sampling p e r io d . Figure 2 illustrates the equipment used for the continuous monitoring of the inert gas activity and 131i content of the condenser off-gas discharge to the stack. The gas flows through the equipment at a rate of 10 litres/m in, and first passes through a charcoal granule filter which collects any '3^1. The filter is monitored continuously by two sodium iodide scintillation detectors. One of these detectors is mounted very close to the filter to obtain maximum sensitivity while the other views the filter through a collimator designed to reduce the sensitivity by a factor of 100 for emergency purposes. The signal output from the detectors is taken to a single channel pulse amplitude analyser set to detect the 364 KeV emission from '31i. The output of the analyser is taken to a ratemeter, which has been calibrated to read directly the ^31i collected on the charcoal from the time the sampler was switched on or from the time the charcoal was last changed. The need to change the charcoal granules is determined by the amount of activity collected and can be carried out remotely from the Reactor Control Room. A charcoal storage hopper contains sufficient material for about 35 complete changes of the charcoal. The sensitivity of detection for l34i is determined by the contribution to the 3 6 4 KeV energy channel from other gamma emitting substances in the sampled 304 AMR n PERRY andPALMER FIG. 1. Winfrith SGHWR clean-up and gas monitoring plant. IAEA-SM-180/15 305 FIG. 2. SGHWR gas monitor * typical installation. 306 PALMER and PERRY gas and the gamma radiation background of the environment in which the equipment is located. Where the gamma radiation background is less than 30 pR/h and where there is negligible interference at 364 KeV from other gas activities, a level of about 1 pCi '3'j can be detected on the charcoal filter. To determine the total amount of activity discharged the ratio of sampling rate to stack discharge rate must be known. In the case illustrated (Fig. 2) with a sampling rate of 10 litres/m in and a total gas discharge rate to the stack of about 10,000 litres per minute, 1 pCi of 131i detected on the charcoal corresponds to a release of about 1 mCi. After passing through the charcoal filter, a defined volume of the gas (10 ml) is monitored by two further scintillation counters (identical to those used in the 131i monitor) to assess the level of gamma activity. The signal outputs of the detectors are taken to second pulse counting equipment set to detect x radiation of energy greater than 50 KeV and the output presented on a ratemeter. Alternatively the outputs of the detectors can be taken to a pulse amplitude analyser and the isotopic composition of the discharged activity established. The sensitivity of detection is determined by the volume of the sample and the efficiency of the detectors. The efficiency of detection w ill depend on the gamma ray energy but at a mean energy of 0.5 MeV a minimum detectable level of about 5 x 10'^ pCi/cm3 is obtained. At a stack gas discharge rate of 10,000 litre/min this corresponds to a discharge of about 10 curies per day, i.e . 5 Ci-MeV/d. Following the gamma monitor the gas passes into a second container (200 ml) fitted with a thin anthracene screen scintillation counter. This arrangement provides a rough estimate of the beta activity above energies of 1 5 0 KeV. The signal output is presented on a ratemeter. The beta counter is not significantly energy dependent and has a sensitivity of about 10*5 pCi/cm3 corresponding to a stack discharge level of about 200 mCi per day in the sampling arrangement shown in Fig. 2. Each of the detectors used in the system (i.e. '31l, gamma and beta) is fitted with a small switchable test source which allows a rapid check to be made, at any time, of the sensitivity and operation of the system. In addition to this system a direct reading gamma monitor, calibrated in Ci-MeV units, comprising two energy compensated Geiger counters is mounted in the base of the stack to give an instantaneous reading on a recorder in the reactor control room. The sensitivity of these detectors is such that a discharge of 50 Ci MeV per day can be estimated. The continuously sampling filte r packs ("MAY" packs), which are removed for measurement in the laboratory, allow a very sensitive measurement of any 1 3 I 1 released. The limit of detection of 131i released through the stack is estimated at about 0.1 pCi. In operation no radioiodine has been detected in the off gas line. This results mainly from the fact that any 131j which might appear in the primary coolant water, for example following a fuel can failure, is preferentially retained in the water phase and does not appear in a measurable amount in the steam to the turbine and condenser. In addition to the installed stack monitoring equipment described, a second installation using a system of gamma radiation detectors located a short distance from the reactor building is available for detecting any significant release of inert gases in an accident condition. The system was described in an earlier paper presented at the symposium on Rapid Methods for Measuring Radioactivity in the Environment, held in Neuherberg, Munich in 1 9 7 1 [2]. Essentially a number of gamma compensated Geiger IAEA-SM-180/15 307 counters are located in the field at about 150-200 metres distance from the reactor buildings. The system will detect the overhead passage of a cloud of radioactive inert gases. It is supplementary to the normal installed stack monitors and provides a means of detecting and roughly assessing a release of gas which possibly, as a result of the accident, is emitted from the reactor installation by a route other than the normal designed ventila tion stack. Consequently, the first line of defence in protecting the environment is the ventilation stack monitors which detect and help to assess the release of the radioactive gases in normal operation. The same system gives an immediate indication of any significant release in an accident situation. Supplementary installed field monitors give an indication of a release following an accident and can be arranged around the installation to give an immediate indication of the general direction in which the released activity has travelled. ENVIRONMENTAL SURVEILLANCE In support of these installed monitoring devices, a further environ mental surveillance programme is carried out which satisfies two require ments. The first is to demonstrate that the operation of the plant has had no detrimental effect on the environment, that is, it confirms the results of the installed plant monitoring systems. The second is that a detailed knowledge is obtained of the background conditions so that should an accident occur where radioactivity is dispersed in the environment and subsequently deposits on to the land, the extent of the contamination on the countryside can be evaluated. The first part of the environment surveillance programme consists of a measurement on a routine basis of the background Y radiation level arising from the natural radioactivity in the earth and any deposited radioactive materials (for example nuclear bomb test debris) on the surface. At AEE Winfrith a vehicle with a scintillation detector mounted on the roof is driven around a number of selected "ring roads" at radii of 1 mile, 3 miles, 5 miles, and 15 miles from the reactor site. The radioactive profile of the road surface and land bordering the road is recorded on equipment in the vehicle. Provision is made for measuring the y radiation level with portable equipment at selected points on the roadside. The results from such surveys carried out over the last 12 years are shown in Fig. 3 where the average radiation level over the area within 15 miles of the site is plotted against time. The variations in the radiation level as the result of nuclear bomb test debris is very evident. In addition to this survey, which is carried out twice a year as an emergency exercise, radiochemical analysis of milk is undertaken. Milk is collected every two weeks from eight farms surrounding the site, and analysed for ^3^1 content. Part of the sample is retained and bulked over a period of 3 months, when the bulked sample is analysed for 137cs and 90Sr. In addition to this analysis carried out at the nuclear site, a duplicate sample is supplied to the laboratories of the Ministry of Agriculture, Fisheries and Food, who carry out an independent check analysis. EMERGENCY PLANNING Although the arrangements for dealing with an emergency at an Atomic Energy Authority site vary in detail depending on the nature of the work, the overall pattern is reasonably uniform and is illustrated by a descrip tion of the main features of that existing at AEE, Winfrith. 308 PALMER and PERRY RADIATION LEVEL MEASURED ¡ METRE ABOVE THE GROUND IN THE COUNTRYSIDE AROUND WINFRITH. THE RELATIVELY LARGE INCREASE IN THE YEARS 1ЧЫ-19М RESULTED fROM RADIOACTIVE FALL OUT DEPOSITED ON THE EARTHS SURFACE FROM THE RUSSIAN AND AMERICAN NU C L E A R 6 0 M B TESTS IN THAT PERIOD. IN 1967 THE RADIATION LEVEL HAD RETURNED TO THE NATURAL B A C K G R O U N D LEVEL. AEE, Winfrith. The Director of the nuclear reactor site is responsible for setting up the necessary arrangements for dealing with any emergency which might arise. These arrangements must include measures for containing any accident and those necessary for the protection of the general public as well as the employees of the site. The responsibilities for the detailed emergency arrangements on site fall into two main areas. Firstly the manager of a particular reactor is responsible to the site Director for the arrangements in his reactor area for dealing with an accident. These include ensuring alarm devices are installed and regularly tested, detailed plans are made for the evacuation of the reactor staff and the assembly and control of an "incident team" for taking any rapid measures required in the area to shut down the plant and generally control the situation. Secondly a central site organisation is set up which is responsible for taking control of all operations outside the immediate incident area. This organisation arranges for the reception of any evacuees from the accident area, the control of all other staff on the site, the co-ordination of all service organisations, including fire brigade, heavy rescue, health physics monitoring, medical, and general administration services, canteens, transport, etc., and provides any services required by the reactor incident control team. The central site control organisation operates from a site emergency control centre. The centre is equipped with all the necessary communication facilities such as telephone, radio-telephones, tape recorders; detailed up IAEA-SM-180/15 309 to date information of the status of plant including the location of all radioactive m aterials; plans supported by photographs and photographic slides of major plant and the internal layouts of all technical buildings; continuously recording meteorological equipment and detailed information related to electricity supplies, water supplies, waste disposal facilities, etc. A store of heavy rescue equipment is available in the centre and a small personnel decontamination room maintained for the use of controlling p e r so n n e l. All automatic alarms, including those for fire, high radiation, criticality and any manually operated local building alarms, are relayed to and register in the control room, and the arrangements ensure that if an alarm operates, the situation is immediately brought to the attention of a duty engineer (a reactor shift manager from one of the reactor areas) who would, on the evidence available to him, decide whether or not to alert the central site emergency organisation personnel. Once alerted, the central site control centre is manned by a team of senior scientific and technical staff who have experience both of the operations on site and the emergency procedures. The responsibility for ensuring that adequate arrangements are made to protect the environment and in particular to safeguard the general public rests with the Director of the reactor establishment. To bring about the necessary arrangements a Local Liaison Committee is set up with the Director of the Establishment as Chairman and the members consist of representatives of the local and national Government organisations, other organisations with a direct interest such as River Authorities and farming communities, and senior staff from the atomic energy site. The scheme is designed around both the resources available on the atomic energy site and those available in the local community and the overall controlling responsibility rests with the Director of the site. Where off site interests are involved, the atomic energy site operators are responsible for declaring the emergency, and for carrying out all monitoring for radioactive airborne and deposited contamination, both on site and in the neighbouring countryside. The Local County Police play a large part in the off site arrangements and are alerted by the responsible officer at the atomic energy site if there is any possibility that off site interests or the general public will be involved. Detailed arrangements operated by the Police alert all outside organisations who have an interest or some function to carry out. Among the most important of these are - (a) The County Civil Emergency organisation, which exists for dealing with any public emergency such as a road, rail or aircraft a c c id e n t. (b) The local representative and central organisation of the Ministry of Agriculture, Fisheries and Food who would send a liaison officer to the reactor site and carry out their detailed arrange ments for the control and replacement, where necessary, of milk and food stuffs considered unsuitable for human consumption. (c) The County Medical Officer for Health who is responsible for any medical matters relevant to the health of the general public. 310 PALMER and PERRY Public Relations aspects of the emergency would be handled and controlled by the Atomic Energy Authority, making full use of the radio, television and news media, as considered necessary. The Atomic Energy site Public Relations Officer working closely with the London Office of the Authority, would deal with local enquiries from the press. CONCLUSION Since considerable effort is expended in the design and safety appraisal of any nuclear plant before it is allowed to operate, and regular inspections are carried out on all operating procedures, the chances of a large scale emergency and the need to operate the emergency plans are reduced to a low level. For this reason it becomes necessary to test the arrange ments regularly in order to reduce to a minimum the chances of a malfunction of any alarm equipment and to ensure that existing and new staff are kept and made fam iliar with the emergency procedures. Functional testing of the automatic alarm devices, for example those associated with high radiation and criticality accident warnings, fire alarms, high airborne contamination devices and building evacuation alarms, must be arranged at regular times known to all staff. Our experience shows that once a month in normal working hours is a satisfactory arrangement. The interval between tests is long enough to make staff stop momentarily and realise it is a warning but only a test, and avoids the situation where too frequent or irregular testing of a number of different alarms soon builds up a reaction, so that any genuine warning is dismissed as yet another test. In addition to the functional testing of equipment, exercises following simulated accident situations eure most important. In the early days of a site, tests are necessary of the detailed arrangements for the evacuation of staff, the assembly of incident teams, the mobilising of large scale monitoring effort, the deployment of transport services, the manning of decontamination centres for personnel, etc. At a later stage the exercises are necessary to give staff some practice in operating in an emergency situation and adjusting themselves into thinking of emergency radiation doses, rather than the very low radiation levels met in normal operations. Such large scale simulated accidents need careful planning to avoid mishaps and misunderstandings arising. They interfere with the normal work of the establishment and are costly in lost effort, hence too many such exercises are undesirable. One major exercise a year involving all site personnel is usually acceptable, with smaller supplementary exercises with limited aims being held on individual plants. REFERENCES [1] MEGAW, J., MAY, R .J., Reactor Science and Technology Nuclear Energy Parts A/В (1962) 427-436. [2] PALMER, G .H ., JOHNS, T .F . , PERRY, K .E .G ., IAEA-SM -148/49. R apid Detection of an Accidental Discharge of Radioactive Material to the Atmosphere. DISCUSSION E. HLADKY: How would the radioiodine detection sensitivity of the instrument shown in your Fig. 2 be affected, if a large amount of gaseous fission fragments were released, considering that these would also be ab sorbed by the activated charcoal? IAEA-SM-180/15 3 1 1 G. H. PALMER: One detector is arranged to be a factor of 100 less sesitive than the others, and this is already installed and thus does not need to be switched in after an accident. The graphite would trap a great deal of activity if a m ajor release occurred and it may be difficult to identify the iodine precisely. However, if the instrument is "off-scale" on all energy channels, it at least shows that a very high discharge of activity has occurred and this would be an early indication of an emergency situation. B. M. MICHAUD: Who bears the cost of a nuclear accident affecting the public — the operating authority or the government? G. H. PALMER: The operating authority is responsible for any costs arising from an emergency. W indscale paidcompensationto farm ers for milk which had to be dumped after the accident in 1957, but of course Wind scale at that time was operated by the UKAEA, which belongs to the Govern ment, so in fact the Government paid. B. M. MICHAUD: What is the public attitude to safety m easures designed to protect it in the event of an emergency? G. H. PALMER: It is difficult to predict accurately the reaction of the public but, because of the existence of our Local Liaison Committees, we have gone a long way towards reassuring the public that they would be well cared for should an accident occur. The public's main complaint after the W indscale accident was that they had not been told beforehand that accidents might affect them. The Local Liaison Committee system has remedied that situation. J. PEÑSKO: You showed a slide in which radiation detectors were placed some distance away from the reactor building in one direction only. What happens when the wind blows in the other direction? G. H. PALMER: This particular arrangement is designed to protect a sm all population living a short distance away in that direction. Sim ilar detection system s could be placed all the way round, if necessary. P. PELLERIN: I should like to congratulate you on your paper, which shows just how much realism and efficiency the British display — as always - in their practical approach. I should like to ask you a number of questions. First of all, have you had any '^1 leaks with this type of reactor? G. H. PALMER: Thank you for your kind rem arks. No iodine has ever been detected in the condenser off-gas system for the reasons outlined in my paper. However, we have, on occasions, had some iodine appearing in the prim ary containment of the reactor after closing down and depres- surizing the reactor following operation with some fuel defects and accom panying leakages of water (coolant) out of the circuit as the result of slight mechanical defects. When we clear out the iodine airborne in the prim ary containment prior to entry for maintenance work, we discharge about 1 mCi of '"'I in 24 hours. P. PELLERIN: How many people would be affected if the stable iodide tablets ever had to be issued? G. H. PALMER: Our plans for the distribution of iodide tablets cater for those living within a range of about 1. 5 to 2 km of the reactor. This amounts to about 200 people,which is probably as large a population as one can deal with in reasonable time. P. PELLERIN: Do you think you would be able to distribute these tablets quickly enough for them to be effective? G. H. PALMER: Our experimental inhalation work shows that stable iodide tablets taken even six hours after the exposure give a 50% protection. 312 PALMER and PERRY We feel that the arrangem ents we have made with the M edical Officer for Health will enable us to reach all those 200 people within that time. P. R. KAMATH; My question arises from your statement that Nuclear Installations should compensate for damage and deprivation. This has far- reaching implications, for example compensation for effects on food and fisheries. Is this position acceptable to the United Kingdom Government? G. H. PALMER: All I said was that Nuclear Installations would pay for any compensation following an accident situation. You are asking whether they would pay for claim s arising from effects caused by normal operation. I cannot speak for the Government but, should effects arise which were considered unacceptable by the controlling authorities, e. g. the M inistry of Agriculture, Fisheries and Food, the Department of the Environment or the Nuclear Installation Inspectorate, then the Government would stop the operations. In accident conditions things are quite different and my earlier rem arks on compensation stand. W. M. BURKHARDT: In the event of an emergency, who orders the distribution of the iodide tablets, the M edical Officer or the head of the em er gency committee, who is presumably an engineer? G. H. PALMER: The Medical Officer for Health is responsible for the health of the general public, so the ultimate decision must rest with him. The Head of the Emergency (Local Liaison) Committee is the Director of the Establishment and he advises the M edical Officer for Health. The reason for this is that the Director has the information on any assessm ent of the released activity and any monitoring results. It is unlikely that the advice of the Director would be rejected, since the arrangem ents are worked out with the M edical Officer for Health and in practice he really becomes part of the site's emergency organization. IAEA-SM-180/9 INTERCALIBRATION OF METHODS FOR RADIONUCLIDE MEASUREMENTS ON A MARINE SEDIMENT SAMPLE R. F U K A I , G.A. STATHAM, S. BALLESTRA, K . A S A R I International Laboratory of M arine Radioactivity, M o n a c o Preyenfec! йу V. Яе/ионеи Abstract INTERCALIBRATION OF METHODS FOR RADIONUCLIDE MEASUREMENTS ON A MARINE SEDIMENT SAMPLE. The results reported in the intercalibration of methods for radionuclide measurements on a marine sediment sample are surveyed. The preparation of the marine sediment sample, its homogeneity and the procedures for in te rc a la tio n are also described. Forty-four laboratories from 20 countries participated in this intercalibration exercise during 1972-73. The survey of the reported results shows that, although the scatter of the data for the sediment measurements is still considerably large, especially for ruthenium-106, the comparability of the results obtained by different laboratories has been improved, as a whole, when compared with that for the seaweed sample. Nevertheless, a comparison of the results obtained by Ge(Li) y-spectrometry with other methods indicates.that there are still problems of calibration of Ge(Li) y-spectrometers in some laboratories due, perhaps, to the radionuclide standards used and/or the procedures of instrument calibration. The probable concentrations estimated by various statistical treatments of the reported data are presented for major radionuclides reported, such as strontium-90, ruthenium-106, caesium-134, caesium-137 and cerium-144. 1. INTRODUCTION Adequate environmental monitoring is essential to ensure the safe opera tion of nuclear installations. As the development of the nuclear industry is expected to accelerate considerably in the coming decade, more monitoring efforts will be required. An effective monitoring system depends not only on a proper design of the program m e, which is one of the m ajor topics of the present symposium, but also on the quality of monitoring data on which rigorous hazard assessm ents are based. The latter aspect is under particular consideration by the IAEA's analytical quality control services. Since the sea receives, directly or indirectly, radionuclides released into the environ ment from nuclear operations and, at the sam e time, interconnects the continents of the globe, it is particularly important that the monitoring data of the marine environment be comparable on an international basis. It has been demonstrated, however, by the results of recent intercalibration exer cises on seawater [1] and seaweed [2] that the comparability of the data produced by different laboratories on homogeneous marine sam ples has not always been satisfactory. As marine sediment represents a sink for many radionuclides, it is considered an important monitoring m aterial. It is also an important source of the radiation dose received by marine biota and, in 313 314 FUKAI et al. some cases, by the human body, in the vicinity of nuclear installations. The present paper deals with the results of a survey carried out on the m easure ments of radionuclides in a homogeneous sediment sample distributed to different laboratories. It should be emphasized that the m aterial presented in this paper is the product of all the participating laboratories. The results indicate the starting point for achieving better comparability of radionuclide m easure ments in marine sediments in the future. 2. COLLECTION AND PREPARATION OF THE SEDIMENT SAMPLE The field collection of the sediment contaminated with radionuclides at monitoring levels was carried out by the Bhabha Atomic Research Centre, India.! About 50 kg of wet sediment was collected from m ussel beds in Bombay harbour during the winter of 1972 into a polyethylene barrel. Large pieces of gravel and shells were removed from the thick slurry. The slurry was frozen at - 40°C and freeze-dried. After freeze-drying, about 10 kg of dried sediment was obtained. The dried sediment was sent to the Monaco Laboratory, where it was homogenized mechanically in a vertically rotating polyethylene bottle. Sieving was not carried out. At this stage the sample contained 3. 2% m oisture, which was subsequently removed by drying at 105°C. The sediment sample was code numbered as SD-B-1. 3. HOMOGENEITY OF THE SAMPLE Prelim inary homogeneity tests were carried out by semi-quantitative -/-spectrometry on m ajor peak areas. Ten gram s of the sediment were put into a plastic counting tube and counted 10 tim es for 90 min using a 7.5 cm dia. x 7.5 cm Nal(Tl) well-type crystal coupled to a 200-channel analyser, in order to check the variations of counter functioning. Eight different sub-sam ples were counted under sim ilar counting conditions. Mean values of these countings with their standard deviations are presented in Table I. While the standard deviations of repeated counting did not exceed ±1% for different peak areas, the standard deviations of different sub-sam ples for major peak areas ranged around ±3%. The latter is considered to be a result of sample inhomogeneity. From these values it can be estimated that levels of m ajor у-em itters in 10 g of sample vary less than ±3% (one standard deviation) due to sam ple inhomogeneity. The results of homogeneity tests by determining individual radionuclides are given in Table II. The analytical methods adopted at Monaco were those ' Comment received from A. K. Ganguly and B. Patel (Bhabha Atomic Research Centre, India): The sediment sample discussed in the paper had been chosen from a large number of grab samples collected during our regular radio-ecological monitoring of Bombay Harbour. The sample for intercalibration exercise has been singled out to provide 'a sample' with reasonable radioactivity that could be easily detected. The activity detected in the sample, therefore, represents the level at a specific point only and should not be confused as a general level of activity. The distribution of various nuclides both horizontal and in vertical profile at various spots in the harbour has been discussed in our paper on 'Radioecology of Bombay Harbour — a tidal estuary'. Comprehensive papers on distribution of other radionuclides are under preparation. The only scientific relevance of this sample from our viewpoint is that the radionuclides have been adsorbed on sedimentary particles under natural environmental conditions. IAEA-SM-180/9 315 TABLE I. RESULTS OF PRELIMINARY HOMOGENEITY TESTS ON SD-B-1 BY COMPARING REPRESENTATIVE PEAK AREAS OF y-SPECTRA Channei range 9 - 21 55 - 62 64 - 90 144 - Ce '° 6Ru-Rh 'Sics 0.134 MeV 0. 51 MeV 0. 66 MeV 6520 ± 40(0.6%) 1100 ± 10 (0. 9%) 12 190 ± 60 (0.5%) 1 sub-sample ^ (counts/100 min per g) Mean of countings 6740 ± 210 (3.1%) 1150 ± 40 (3. 5%) 12 430 ± 350 (2.8%) on 8 different sub-samples ^ (counts/100 min per g) a Standard deviations are those for single measurement. described by Bowen [3] for strontium-90, by FAO/IAEA/WHO [4] for ruthenium-106, by Fukai et al. [ 5 ] for caesium -134 and caesium -137, and by HASL[6] for cerium-144. For the measurements of total caesium isotopes j3-counting procedures with radiochemical separationbyHASL[7] were also applied. At Woods Hole, a method sim ilar to that adopted in Monaco was used for strontium -90, while other radionuclides were measured by direct -y-spectrometry on the sediment using a Ge(Li) detector. The sample size used at Monaco was between 0. 5 and 5 g sediment and at Woods Hole 5 g for strontium-90 and 35 g for y-spectrom etry. The standard deviations for single determinations of these results presented in Table II confirm, in general, the homogeneity of the sam ple to be around ±3% standard deviation (It?) at these sample sizes when the errors introduced during chemical separation procedures are taken into account. The variations of the results for strontium-90 and caesium -134 at Monaco are larger due to the low levels of these radionuclides present and those for caesium -134 and cerium -144 at Woods Hole are larger due, perhaps, to counting statistics. The difference found between Monaco and Woods Hole for caesium -137 values is probably attributable to system atic errors involved in either or both of the methods employed. In conclusion, the concentrations of m ajor radionuclides in the distributed sample of 10 g size or less does not vary more than ±3% standard deviation (la) due to the difference in the composition. 4. PROCEDURE FOR INTERCALIBRATION Polyethylene bottles containing around 100 g of the prepared sediment were distributed to 55 institutions (including IAEA's Seibersdorf Laboratory) from 2 5 Member States of the IAEA. The participants were informed of the m ajor radionuclides present in the sam ple, approximate homogeneity of the sam ple, granulometric composition, clay mineral composition, organic carbon and nitrogen content, etc. For reporting the results, standard 316 FUKAI et al. TABLE II. RESULTS OF HOMOGENEITY TESTS ON SD-B-1 BY DETERMINATIONS OF INDIVIDUAL RADIONUCLIDES (pCi/g-dried m atter, on 1 Jan. 1973) Sample Quantities found Laboratory bottle 137- 144- No. '°S r ""R u *^Cs Cs Ce A 15. 6 80.4 11.2 363 142 В 14.7 78.8 10.2 367 135 С 18.0 72.4 8.7 360 128 60 13.4 74.4 10.9 355 135 Monaco ^ 65 15.1 78. 5 14. 5 364 132 Mean 15 77 11 362 134 ± 2 ± 3 ± 2 i 5 ± 5 i f (Single det. ) J L (13%) (3. 9%) (18%) (1.4%) (3.7%) 9 14.4 --- - 9 14. 9 - - -- 9 15.2 --- - 9 14.9 -- -- 9 14. 9 - - - - 9 13.6 - - - - WHOt*^ 9 - 74. 5 9.63 407 143 9 - 7 2 .6 9 .9 5 425 158 9 - 6 9 .8 1 0 .6 441 167 Mean 1 4 .7 7 2.3 1 0 .1 424 156 ± 0 .5 ± 2.4 ± 0. 5 ± 17 ± 12 (Single det. ) NJ L (3.4% ) (3.3% ) (5. 0%) (4 . 0%) (7.7% ) form s which included a report on instrumentation, calibration, calculation and analytical procedures as well as the analytical data, were distributed to the participants. The reference date for reporting the radioactivity was set at 1 January 1973. Finally, 42 institutions from 20 Member States and two laboratories of the IAEA reported the analytical results. A list of the institutions which participated in this intercalibration exercise is given as an Annex with the names of the investigators involved in this work. IAEA-SM-180/9 317 5. S U R V E Y O F T H E R E S U L T S The reported results were compiled for five frequently reported radio nuclides (strontium-90, ruthenium-106, caesium-134, caesium-137 and cerium -144) in Table III and for other less frequently reported radionuclides' in Table IV. Although some laboratories reported the results on plutonium-238 and -239 or americium-241, these results will be reported e lse w h e re . 5. 1. Data treatment All values given in Tables III and IV were taken from the original reports by the participants, being changed as little as possible; significant figures were maintained as they were reported; when the reported radio activity had not been normalized to the reference date of 1 January 1973, the decay was corrected; in cases where two or more results had been reported for one radionuclide by one laboratory, an arithmetic average for the laboratory was calculated and presented in the tables. These average values are underlined in the tables. The errors reported seem to represent, in most cases, standard deviations (la) based on counting statistics; in Tables III and IV the errors for the averaged values were computed from individual errors for two or three results reported; where four or more results were averaged, standard deviations (1er) were calculated from indi vidual results. 5.2. Scatter of the data The overall average values for five major radionuclides (strontium-90, ruthenium-106, caesium-134, caesium-137 and cerium -144) are given in Table V with maximum and minimum values reported for each radionuclide. In order to obtain overall averages, all available results were arithmetically averaged, regardless of their wide variation. Therefore, these average values given in Table V have little statistical significance as no criterion was applied for rejecting widely deviating data. Nevertheless, the standard deviations computed for these average values give ideas of the scatter of data for individual radionuclides. For the purpose of comparison, percentage standard deviations of overall averages for corresponding radionuclides analysed on the seawater [1] and seaweed [2] are also given in Table V. Except for ruthenium-106, the percentage standard deviations do not exceed ±10% for the sediment sample, while the factor between maximum and mini mum values for ruthenium-106 is around 10 000. When compared with data for seawater and seaweed sam ples, the scatter for the sediment data is definitely sm aller than that for SW-1-1 (lower level seawater) and AG-1-1 (seaweed), except for strontium-90, while only caesium-134 and cerium-144 data of the sediment showed sm aller scatter than that for SW-1-2 (higher level seawater). The comparison of sediment data with those of seaweed are especially warranted since the majority of the measurements for у-em itters in both sediment and seaweed were carried out by direct y-spectrometry with Ge(Li) detectors. Although the scatter of the individual data for sediment measurements is still large, especially for ruthenium-106, the narrower standard deviations of the data for у-emitting radionuclides in sediment compared to those obtained for seaweed indicate an improvement FUKAI et al. E III. RESULTS OF MEASUREMENTS ON MAJOR )NUCLIDES IN SEDIMENT SAMPLE SD-B-1 BY !RENT LABORATORIES dried matter, on 1 Jan. 1973 ode '"Sr ""Ru 144- 'Jo. '"C s '3?Cs Ce 1 - 71 ± 6 12.7 ± 0.5 367 ±3 146 ± 10 2 - - 10.5 ±0.8 361 ± 4 - 3 15.6 ± 1.6 89 ± 15 10.3 ± 1 384 ± 15 140 ± 15 4 - 76 * 7 present 403 ± 5. 0 54 ± 4 5 13.62±0.29 73 ± 1.4 10.2 ± 0.2 390 ±3 131 ± 1 6 - 46.7 ± 1.1 8 .5 * 0 .2 357. 5 ± 3.6 139.1 ±1.1 7 15.8 ±0.6 67 ± 5 8 ± 1 316 ± 10 117 ± 7 8 14.7 ±0.5 72.3 ±1.7 10.1 ± 0.4 424 ± 12 156 ± 9 9 13.7 ± 0.4 98.8 ±4.6 - 363.3 ±5.4 153.0 ±4.6 10 - 0.0667 ± 0.0055 0.399 ± 0.022 0.122±0.011 11 - 49 ± 3 8.5±0.3 340 ± 2 120 ± 2 12 24.1 ± 0.9 56.9 ±5.2 - 274.3 ±3.1^ 107.9 ±1.6 13 - 128 ± 13 15. 6 ± 1.3 726 ± 41 242 ± 26 14 13.05 ± 0.94 20.44 ± 0.69 10.00 ± 0.70 470.41 ± 7.06 162.77 ±2.44 15 15.7 ± 1.9 72. 5 ± 8.7 10.3 ± 1.4 416.1 ± 28.1 176.2±12.3 16 - 154±1 - - - 17 13.3 ±0.2 109±2 - 240 ± 1 a 115 ± 1 18 15.3 ±0.5 68 ± 3 9 ± 1 372 ± 3 117 ± 2 19 - 60 ± 10 16 ± 1.1 360 ± 11 130 ± 7 20 21 15. 5 ± 2. 5 -- 406 ± 33.4^ - 22 16.0 ± 0.2 34 ± 2. 5 4.6 ± 0.3 174 ± 5 72 ± 2. 5 23 13±2 N. D. - 370 ± 20 120 ± 10 24 16. 7 ± 2.4 76. 7 ± 5.2 - 360 ±30^ 121 ± 8 25 - 72 ± 4 10.3 ±0.5 370 ± 20 120 ± 10 26 9.2 ± 1.9 80 ± 3 11 ± 1.7 385 ±2.1 142 ± 2 27 3. 70 ± 0.02 - 10.0 ± 0.4 263 ± 2 - 28 1.83 ±0.2 43.0 ±5.0 6.2 ± 0.5 269 ±3.0 85.1 ±2.6 1 5 .4± 1.5 68. 5 ± 7. 0 7.5± 1.2 334 ± 26 167 ± 17.0 - 48.7 ±2.1 7.2 ± 0.3 358 ±1.4 94. 9 ± 2.1 IAEA-SM-180/9 319 T A B L E III (cont.) 3 Value represents '^Cs+ "'C s . in the comparability of the results obtained by different laboratories. This may be due to an increased awareness of the pitfalls in the calibration pro cedure of they-spectrom eter used, as well as to the sim pler radionuclide composition of the sediment sample (SD-B-1) as compared with the seaweed sam ple (AG-1-1). Although overall averages were not computed for the less frequently reported radionuclides presented in Table IV, the average values for sodium-22, potassium-40, cobalt-60, antimony-125, europium-154 and-155 may indicate approximate levels of these radionuclides present in this sedi ment sam ple. Considering the relatively short half-life of strontium-89 and the length of time elapsed between sample collection and dispatch, the reported data for strontium -89 seem to be questionable. Although the total number of reported results for each radionuclide is not quite sufficient to give significant statistical distributions of the data against concentration ranges, the frequency appearance of data over chosen concentration ranges are illustrated by histogram s for five m ajor radio nuclides, strontium-90 and ruthenium-106 in Fig. 1, and caesium -134, caesium -137 and cerium -144 in Fig. 2. The unit concentration ranges have been chosen to represent around 10% of the overall average values and very high values are excluded in the illustration. From these figures it can be said that strontium-90, ruthenium-106 and caesium-137 have sharper peaks 320 FUKAI et al. TABLE IV. RESULTS OF MEASUREMENTS ON LESS FREQUENTLY REPORTED RADIONUCLIDES IN THE SEDIMENT SAMPLE SD-B-1 Concentration reported Radionuclide Half-life Code No. of laboratory (pCi/g-dried matter)^ Na-22 2. 58 a 0.43 ± 0. 07 5 0. 5 ± 0.06 37 K-40 1.28 x 109 a 13 ± 2 43 20.7 ±4.5 15 9.3 ± 5.0 30 11.2 ±1.5 38 9.5 ± 0.5 Co-60 5.26 a 0.2 ± 0.1 43 0.13 ± 0.06 5 0.091 ± 0.007 17 Sr-89 50.4 d 1.87 ± 0.92 5 4.7 ± 0.8 28 Zr-Nb-95 65 d; 35 d 1.7 ± 0.1 20 Sb-125 2.7 a 3 ± 1 43 4.4 ± 0.6 26 2.80 ± 0.46 38 13.8 ± 7.0 39 Eu-154 16 a 0.94± 0.15 38 0.6 ± 0.2 43 Eu-155 1.7 a 1.8 ± 0.2 22 4.44 ± 0.21 38 4.0 ± 0.5 43 3. 0 ± 0.4 44 ^ Activity un 1 Jan. 1973. tAEA-SM-180/9 321 TABLE V. OVERALL AVERAGE VALUES AND RANGES OF THE REPORTED VALUES FOR THE MAJOR RADIONUCLIDES IN THE SEDIMENT SAMPLE SD-B-1 l° n 134- 137 _ 144 Radionuclides Sr 6Ru Cs Cs Ce No. of reported results 23 36 29 39 34 Max. value (pCi/g) 24.1 734 16 726 279 Min. value (pCi/g) 1.83 0. 0667 4.6 0.399 0.122 Overall average (pCi/g) 13.8 87 9.9 356 126 ±0.9 ± 19 ±0.5 ± 17 ± 9 o (m ean) (6.5%) (22%) (5.0%) (4. 8%) (7.1%) о (%, mean) for seawater sample ± 8.9% ± 39% ± 19% ± 13% ± 38% SW -I-lR ± 4.8% ± 11% ± 6.4% ± 4. 0% ± 28% SW-1-2 a for seaweed sample ± 5.0% ± 46% ± 71% ± 64% ± 27% AG-1-1 b ^ SeeRef. [1]. b SeeRef. [2]. <06 R u s o - e - ^ <5.0 22.5 30.0 0 50 <00 <50 2 0 0 )06 Sr fpC FIG. 1. Per cent frequency distribution of the reported results against concentration ranges for strontium-90 and ruthenium-106 in the sediment sample SD-B-1. 322 FUKA1 et al. <34^ 5 0 C s ' ^ s . ? S D -Я-1 SD - 8 - 1 Á340 3 c - $ 30 3 o- ^ 2 0 -- - Ю .] -- Ш м Jit i 11 A t ЛМЖ М д i 11 i A 5.0 tO.0 )5.0 20.0 0 200 400 600 8000 75 < 50 225 300 Cs (pCt/g) t s (pCí/gJ *Ce (pCi/g ^ FIG. 2. Per cent frequency distribution of the reported results against concentration ranges for caesium-134, 137 and cerium-144 in the sediment sample SD-B-1. of distribution than those of caesium -134 and cerium -144. In the case of ruthenium-106, however, the distribution shows the occurrence of another definite sub-peak approximately 30% below the highest peak. This sub-peak may be attributable to some system atic errors involved in y-spectrom etry, since all of these lower results were obtained by using either Ge(Li) detectors or Nal(Tl) crystals. 5. 3. Probable concentration values In order to arrive at the probable radionuclide concentrations in the sample, it is necessary to reject some widely deviating results. For the rejection of data various methods were used and the average concentrations obtained by excluding outlying data for strontium -90, ruthenium-106, caesium -134, caesium-137 and cerium -144 are given in Table VI. The first method used for rejecting outlying data is the application of Chauvenet's criterion [8,9] and the average values computed after applying this criterion are presented in Table VI. It was pointed out, however, by Natrella [10] that the application of the Chauvenet's criterion to the rejection of data tends to reject statistically valid data when the number of observations is sm all, so that it may sometimes give a biased view. He recommended the use of Dixon's criterion instead [10]. The average values for radio nuclide concentrations obtained after applying Dixon's criterion to rejection of the data are also given in Table VI. A comparison of these average values with those obtained by applying Chauvenet's criterion shows that they are practically identical. Another method utilized for estim ations of probable concentrations is the so-called 'method of higher frequency ranges'. This method has no theoretical statistical grounds, but gives a reasonable estim ate when a part of group of data is biased due to system atic errors involved. As was pointed IAEA-SM-180/9 323 TABLE VI. PROBABLE RADIONUCLIDE CONCENTRATIONS ESTIMATED FORSTRONTIUM-90, RUTHENIUM-106, CAESIUM-134, CAESIUM-137 AND CERIUM-144 IN SD-B-1 BY VARIOUS METHODS " Radionuclide " S r "'R u '" C s ^ Cs ^ C e No. of reported results 23 36 29 39 34 11.9 - 18.1 0 - 132 6.2 - 13.0 218 - 504 67 - 187 (pCi/g) 18 34 24 36 29 15.0 66 9.6 361 127 range (pCi/g) ± 1.4 ± 4 ± 0.3 ± 10 ± 5 о (mean) (9.3%) (6.1%) (3.1%) (2.8%) (3.9%) No. of results retained 21 . 35 29 37 33 after applying Dixon's criterion Average after rejection 13.9 69 9.9 356 122 of data by Dixon's criterion (pCi/g) ± 0 .7 ± 5 ± 0. 5 ± 11 ± 7 о (mean) ^ (5. 0%) (17.2%) (5. 0%) (3.1%) (5. 7%) Higher frequency 12.1 - 18.0 50 - 100 7.1 - 12.0 321 - 480 76 - 180 ranges chosen (pCi/g) No. of results in higher 18 22 22 29 28 frequency ranges Average in higher 15.0 73 9.6 377 129 frequency ranges (pCi/g) ± 0 .3 ± 2 ± 0.2 ± 6 ± 4 о (mean) { (2.0%) (2.7%) (2.1%) (1.6%) (3.1%) ^ On 1 Jan. 1973. out, the distribution of the data for ruthenium-106 is suspected to have a sub-peak due to system atic errors involved. In this case the average obtained by excluding the data being suspected to be biased due to system atic errors may be more near to the true value. In order to reduce the ranges to be considered for computing average values, the range of the highest frequency and neighbouring ranges were chosen and the averages were com puted for the data existing inside these ranges. The ranges chosen and average values computed for each radionuclide are given in Table VI. The average values computed by this method do not differ greatly from those 3 2 4 FUKAI et al. TABLE VII. AVERAGE VALUES OF THE RESULTS OBTAINED BY DIFFERENT METHODS OF MEASUREMENT '°"Ru 137 Cs I'M Ce Method used No. of No. of No. of PCi/g pC i/g pC i/g Ge(Li) 21 94 ± 33 (35%) 24 359 ± 26 (7.2%) 22 124 i 11 (8.9%) y-spectrometry Nal(Tl) 8 68 ± 6 (8. 8?.) 10 365 i6(1.6% ) 8 135±24(18%) y-spectrometry Radiochemical 7 88 ± 13 (15%) 5 328 ± 31 (9. 5%) 5 119±6(5.0%) methods TABLE VIII. RESIDUAL ACTIVITIES IN SEDIMENT SD-B-1 OF RUTHENIUM-106, CAESIUM-137 AND CERIUM-144 AFTER AQUA REGIA LEACHING AS ESTIMATED BY REPRESENTATIVE PEAK-AREA COUNTING Residual activity (%) ^ C e ^R u 's? Cs Drying status (55-65 channels) (116-140 channels) (66-85 channels) Freeze-dried 1st < 1 .9 2.4 1.4 <2.4 2nd 0.4 1.0 1.6 0.4 Oven-dried 1st < 5 .2 <6.7 5.5 <8.3 2nd 0.4 0.4 1.3 0.4 obtained by applying Chauvenet's or Dixon's criterion, although the narrowing down of the range considered brought the corresponding standard deviations closer. In the case of ruthenium-106 the method of higher frequency ranges brought up the average value about 10% and for caesium -137 about 5%. These values are probably closer to the true values than those approached purely statistically. 6. ANALYTICAL METHODS ADOPTED One of the characteristic features of this intercalibration exercise is, as has been the case for seaweed sam ples [2], the wide use of direct у-spectrom etry with Ge(Li) detectors. Except for the determination of strontium -90, the m ajority of the laboratories employed non-destructive -y-spectrometry by using Ge(Li) detectors for the determination of m ajor radionuclides, ruthenium-106, caesium-134, caesium-137 and cerium-144. IAEA-SM-180/9 32 5 TABLE IX. ANALYTICAL METHODS USED BY DIFFERENT LABORATORIES FOR THE DETERMINATION OF STRONTIUM-90 IN THE SEDIMENT SAMPLE SD-B-1 Method No. Brief description of method Code No. of laboratory Sr-1 Cone, acid leaching, Sr sep. withfum . HNOg, 8, 9, 14, 18, 22, 27, soy-milking as hydroxide or oxalate. 28 Sr-11 5, 42 9°Y-milking. Sr-IH HF-treat., alkali treat., Srsep. withfum. HNO3, 24 Y-m ilking. Sr-IV 3 fum. HNOg, ^Y -m ilking. Sr-V Ignition (450-600°C), acid leaching, Srsep. withfum. HNO3, 15, 26, 23 9°Y-milking as hydroxide or oxalate ppt. or with solvent extr. with HTTA-TBP/MIBK. Sr-VI HC1 + HF leaching, scavenging with Fe(OH)g, ^°Y-milking. 12 Sr-VII Cone, acid leaching, ^Y extr. withHDEHP, ß-counting. 21,23,29 sr-vm Cone, acid leaching, s°Y extr. with TBP, ß-counting. 35 Sr-IX Cone, acid leaching, Srsep. with ion-exchange, ^°Y-milking 17,43 Sr-X NaCOa fusion, dissol. with HC1, Sr sep. with ion-exchange, 7, 36 ß-counting on SrCOg or soy-m ilking by TBP-extr. The results obtained with this procedure are compared below with those produced by other methods. Another important aspect of sediment analysis is the effectiveness of leaching of radionuclides from sediments before m easurements are made, when radiochemical methods are employed. The results of leaching experi ments carried out at the Monaco Laboratory are also presented below, in connection with radiochem ical methods used by different laboratories for individual radionuclides. Due to the fact that a specific method was used by only one laboratory in many cases, it is difficult to correlate exactly the quality of the results with the methods adopted. Nevertheless, an attempt is made to evaluate the suitability of the methods for monitoring purposes as far as possible. 6.1. -/-spectrometry In Table VII the average values of the results obtained for ruthenium-106, caesium-137 and cerium-144 by different categories of the measurement 326 FUKA1 et al. TABLE X. ANALYTICAL METHODS USED BY DIFFERENT LABORATORIES FOR THE DETERMINATION OF RUTHENIUM-106 IN THE SEDIMENT SAMPLE SD-B-1 Method No. Brief description of method Code No. of laboratory Ru-1 Direct Ge(Li) y-spec. 3, 5, 6, 7, 8, 10, 11, 13, 14, 15, 18, 22, 26, 28, 31, 34, 37, 38, 41, 43, 44 Ru-11 Direct NaI(Tl) y-spec. 1, 4, 9, 19, 25, 30, 36, 39 Ru-111 Ignition at 400 - 500°C, fusion with KOH + KNO3, RuO^ 32, 42 extr. withCCl^, red. toRuOg, 6-counting. Ru-IV Conc. acid leaching, RuO^ distil, with KMnO^, red. 12, 17, 29 to RuOa or Ru-metal, ß-counting. Ru-V Fusion with KOH + KNO3 + Kg CO3, leaching with HgO + NaCIO, 16 RuOi distil., withPbO^, red. toRuOg, 6-counting. Ru-VI HF-treat., alkali leaching, RuO^ distil. withHClO^, red. 24 methods, that is Ge(Li) y-spectrom etry, Nal(Tl) y-spectrom etry and radio chemical methods, are given together with their standard deviations. Judging from the per cent standard deviations given in the table, it can be said that the scatter of the data obtained by Ge(Li) y-spectrom etry is larger than either that obtained by Nal(Tl) y-spectrom etry or that by radiochem ical methods for the three radionuclides concerned. In the case of ruthenium-106, the scatter is especially great. This indicates that there is a problem in calibrating a Ge(Li) y-spectrom eter in some laboratories due, perhaps, to the radionuclide standards used and/or procedures of instrument calibration. The spread of the results of Nal(Tl) y-spectrom etry is narrower than those of the other two categories of methods, except in the case of cerium-144. The cause for this large spread for cerium -144 should be interferences due to Compton scattering in Nal y-spectra. The results obtained by various radiochemical methods varied less than ±15% for all three radionuclides considered. It is strange, however, that the average value for caesium -137 by radiochemical methods is considerably lower than those of other categories, in spite of the fact that radiochem ical methods followed by ß-counting cannot differentiate between caesium -134 and caesium -137 and the activity reported should be the sum of these two radionuclides, expressed in term s of picocuries of caesium-137. The lower values obtained by radiochemical procedures may be due to insufficient leaching or inaccurate chemical yield determination. 6.2. Leaching procedures When radiochem ical procedures are applied to radionuclide determinations in sedim ents, it is essential to dissolve all radioactive atoms to be m easured IAEA-SM-180/9 327 TABLE XI. ANALYTICAL METHODS USED BY DIFFERENT LABORATORIES FOR THE DETERMINATION OF CAESIUM-134 AND -137 IN THE SEDIMENT SAMPLE SD-B-1 Method No. Brief description of method Code No. of laboratory Cs-1 Direct Ge(Li) y-spec. I,3,5,6,7,8,10, II, 13, 14, 15. 18, 22, 26, 27, 28, 29, 31, 34, 37, 38, 41, 43, 44 Cs-11 Direct NaI(Tl) y-spec. 1, 2, 4, 9, 19, 23, 25, 29, 30, 36, 39, 40 Cs-IH Cone, acid leaching, AMPabsorp. ofCs, CsgPtCl^ ppt., 12, 42 у-spec. and/or ß-counting. Cs-IV HF-treat., alkali leaching, AMP absorp. of Cs, Cs sep. 24 with ion-exchange Bio-Rex40, TPB-Cs, ß-counting. Cs-V HCI + NaCIO leaching, ion-exchange sep., AMPabsorp. 21 of Cs, ion-exchange purif. of Cs, CsgPtClg ppt., ß-counting. Cs-VI Cs extr. with phenol-nitrobenzene mix. in dodecyl benzene 24 sulphonatemed., CSgPtClg ppt., ß-counting. in order to ensure complete isotopic exchange with carrier atoms added. In order to examine the effectiveness of acid leaching, a series of experi ments were carried out at Monaco by m easuring residual activities with a Nal(Tl) y-spectrom eter. The results of these experiments are presented in Table VIII. The leaching was performed twice on 5 g of sediment with 100 ml aqua regia for 15 hours at room temperature and 5 hours at 90°C. The residual activities in peak areas corresponding to cerium -144, ruthenium-106 and caesium -137 were compared with those m easured before the leaching. The values in Table VIII represent the averages of three experimental runs. The results in the table show that two leachings with aqua regia dissolved 99% or more of ruthenium-106, caesium-137 and cerium -144 into solution, while 5% to 10% of these radionuclides stay in oven-dried sediments after one leaching only. Although ruthenium-106 can be effectively dissolved by two leachings with aqua regia, the large amount of iron dissolved at the sam e time from the sediment made the following radiochem ical procedures difficult. As for strontium -90, two aqua regia leachings gave constantly lower results than those by alkali attack with NaOH and successive acid leachings. The average value obtained by the form er procedure is 13.4 ±0.2 pCi sogp/g, being 13% lower when compared with 15.4 ±0.8 pCi 90Sr/g of the latter. 6. 3. Radiochemical procedures for individual radionuclides The analytical methods used by different laboratories for the m ajor radionuclides, strontium-90, ruthenium-106, caesium-134 and -137, and 3 28 FUKAJ et al. TABLE XII. ANALYTICAL METHODS USED BY DIFFERENT LABORATORIES FOR THE DETERMINATION OF CERIUM-144 IN THE SEDIMENT SAMPLE SD-B-1 Method No. Brief description of method Code No. of laboratory Ce-1 Direct Ge(Li) y-spec. I, 3, 5, 6, 7, 8, 10, I I , 13, 14, 15, 18, 22, 26, 28, 29, 34, 37, 38, 41, 43, 44 Ce-11 Direct Nal(TI) y-spec. 1, 4, 9 19, 23, 25, 29, 30, 36, 39 Ce-IH 17, 42 with MIBK after oxid., Ce ppt. as oxalate, у-spec. and/or ß-counting. Ce-IV Cone, acid leaching, hydroxide ppt., Ce extr. with TBP 12 after oxid., ion-exchange purif. of Ce, Ce ppt as iodate, 6- counting Ce-V HF-treatment, hydroxide ppt., mixed fluoride sep., ion- 24 exchange purif. of Ce, Ce ppt as iodate, 0-counting after purif. of iodate ppt. cerium -144, are listed in Tables IX-XII, respectively. These methods were classified depending on principal analytical schem es, regardless of slight differences in detailed procedures. As shown in Table IX, radiochemical methods used for strontium -90 are classified into ten m ajor schemes. Various dissolution steps of strontium-90 from sediment, such as concentrated acid leaching, alkali treatment, hydrofluoric acid treatment, carbonate fusion, etc. were employed in combination with various separation steps. Most of the methods listed in Table IX gave comparable results, except for the method Sr-VI which gave a very high result and the method Sr-VIII which produced con siderably lower results. This high result seem s to be due to insufficient purification steps for strontium -90 to separate it from other radionuclides contained in the sediment. The analytical methods used for ruthenium-106 which include four 'radiochemical schem es are given in Table X. Although the radiochemical procedures listed seem to give, in general, satisfactory results, the varia tion of the results obtained by distillation schem es of Ru-IV and Ru-V is large. Whether this variation is inherent to distillation schem es or some other causes cannot be judged from these results. As to caesium-134 plus caesium-137, four radiochemical schemes are listed in Table XI in addition to the methods by y-spectrom etry. As has been shown in Table VII these procedures tend to give lower results than those obtained by y-spectrom etry and the variations of the results are also l a r g e . tAEA-SM-180/9 3 2 9 Satisfactory results for cerium -144 determinations were obtained by four radiochem ical procedures given in Table XII, even though the results tend to be slightly lower than those by y-spectrom etry. The spread of the results is less for radiochemical methods than for y-spectrom etry. Although the general tendency is to use y-spectrom etry more and more in monitoring of the marine environment, the quality of the performance for radiochem ical methods should be maintained in the laboratories engaged in the monitoring work, since their higher sensitivity of quantitative deter mination is essential for the levels of certain m ajor radionuclides expected to occur in marine environmental sam ples. 7. GENERAL REMARKS During the survey of the reported results, the following general observa tions were made: 7. 1. As has been pointed out for seawater and seaweed intercalibration, the concept of the 'significant figures' of the m easurem ents was not taken into account in several reported results, bearing in mind their estim ated errors. This indicates the lack of a critical viewing by some investigators of their m easurem ents and possibly shows up a tendency to rely mechanically on computer read-outs of measurements. 7.2. The results of strontium-90 analysis on the sediment sample by different laboratories compare favourably, although a few results deviate considerably from the average value due to the unsuitable methods used. 7. 3. The comparability of the results reported on major y-emitting radionuclides, ruthenium-106, caesium -134, caesium-137 and cerium-144, in the sediment is, as a whole, better than that for the seaweed sam ple, although in both cases the m ajority of the m easurem ents were carried out by y-spectrom etry with Ge(Li) detectors. 7.4. For ruthenium-106 the scatter of the results obtained by Ge(Li) y-spectrom etry is much larger than that of the results obtained either by Nal(Tl) y-spectrom etry or by radiochemical methods. This indicates that there are still problems with the calibration of a Ge(Li) y-spectrom eter in some laboratories, probably due to the radionuclide standards used and/or procedures for instrument calibration. 7. 5. For cerium -144 results derived by Nal(Tl) y-spectrom etric m easurements varied more than those derived either by Ge(Li) y-spectrom etry or by radiochemical methods. This may be due to interference of Compton scattering in Nal y-spectra. 7. 6. The radiochemical methods generally provide results comparable to, or sometimes better than, those derived by y-spectrom etry, even for у-emitting radionuclides, when proper precautions in the performance of the measurement are followed. 3 3 0 FUKAI et al. ACKNOWLEDGEMENTS The authors wish to express their gratitude for the collaboration received from all the participants during the execution of the present intercalibration. They are especially grateful to Dr. A. K. Ganguly and Dr. B. Patel from Bhabha Atomic Research Centre, India, for their efforts spent in the collec tion of the marine sediment sam ple, to Dr. V. T. Bowen from Woods Hole Oceanographic Institution for his extensive collaboration in the homogeneity tests of the sample and to Dr. J. Heinonen from the Seibersdorf Laboratory (IAEA) who consented to present this paper at the Symposium. The authors' thanks are also due to Dr. E. K. Duursma and Mr. P. Parsi for the assistance received in the homogenization and homogeneity tests of the sam ple and to M rs. M. Blessington for the manuscript preparation. The International Laboratory of Marine Radioactivity operates under a tri-partite agreement between the International Atomic Energy Agency, the Government of the Principality of Monaco and the Oceanographic Institute at Monaco. The present work is also financially supported by UNESCO. This support is gratefully acknowledged. REFERENCES [1] FUKAI, R., BALLESTRA, S ., MURRAY, C .N ., "Intercalibration of methods for measuring fission products in seawater samples", Radioactive Contamination of the Marine Environment (Proc. Symp. Seattle, 1972), IAEA, Vienna (1973) 3. [2] FUKAI, R., BALLESTRA, S., "Intercalibration of methods for radionuclide measurements on a seaweed sample", presented at the Symp. on the Determination of Radionuclides in Environment and Biological Materials, London, April 1973. [3] BOWEN, V. T . , "Analyses of sea-water for strontium and strontium-90", Reference Methods for the Marine Radioactivity Studies, Technical Reports Series No. 118, IAEA, Vienna (1970) 93. [4] FAO/IAEA/WHO, "Methods of Radiochemical Analysis','WHO, Geneva (1966) 84. [5] FUKAI, R.. BALLESTRA, S., RAP AIRE, J.-L ., "A simple application of least-squares fitting to gamma spectrometry of marine environmental samples: The case of cesium radionuclides", Rapid Methods for Measuring Radioactivity in the Environment (Proc. Symp. Neuherberg, 1971), IAEA, Vienna (1971) 301. [ 6] HEALTH AND SAFETY LABORATORY, "HASL Procedures Manual (HARLEY, J .H ., Ed.)" HASL, USAEC. New York (1972) p. E-Ce-03-01. [7] HEALTH AND SAFETY LABORATORY, "HASL Procedures Manual (HARLEY, J .H ., Ed.)" HASL, USAEC, New York (1972) p. E-Cs-01-01. [ 8] OVERMANN, R. T . , CLARK, H .M .. Radioisotope Techniques. McGraw-Hill, New York-Toronto- London (1960). [9] CHASE, G. D ., RABINOWITZ, J . L . , Principles of Radioisotope Methodology, Burgess Publ. Com p., Minneapolis (1962). [10] NATRELLA, M. G ., Experimental Statistics, NBS Handbook 91, US National Bureau of Standards, Washington, D .C . (1963) 17-1. IAEA-SM-180/9 3 3 1 ANNEX LIST OF PARTICIPATING INSTITUTIONS * COUNTRY AND INSTITUTION INVESTIGATOR ARGENTINA Gerencia de Protección Radiológica y Seguridad, Comisión Nacional de Energfa Atómica, Buenos Aires AUSTRALIA Research Establishment, C. J. Hardy Australian Atomic Energy Commission, W.W. Flynn Sutherland BELGIUM Centre d'Etude de l'Energie Nucléaire, J. Colard МЫ E. Blok Institut Royal des Sciences Naturelles de Belgique, E. Peeters Laboratoire d'Océanographie Physique, J.M . Das Brussels DENMARK Health Physics Department, Research Establishment Ris%, Danish Atomic Energy Commission, Roskilde FINLAND Institute of Radiation Physics, A. Salo Helsinki L. Blomqvist FRANCE Service de Protection et des Etudes d Environnement, S. Haddad Centre d'Etudes Nucléaires de Grenoble, S. Descours Commissariat à l'Energie Atomique, M. Guitton Grenoble Service Central de Protection contre les Rayonnements Ionisants, P. Pellerin LeVesinet M .L. Remy J. P. Moroni 332 FU KAI et al. COUNTRY AND INSTITUTION INVESTIGATOR GERMANY, FEDERAL REPUBLIC OF Kernforschungsanlage Julich GmbH, G. Erdtmann Zentrallabor fur Chemische Analyse, Jülich Deutsches Hydrographisches Institut, H. Kausky Hamburg M. Kies INDIA Health Physics Division, B. Patel Bhabha Atomic Research Centre, C. Rangarajan Bombay ISRAEL Department of Nuclear Science, N.H. Shafrir Technion-Israel Institute of Technology, J. Laichtet Haifa ITALY Laboratorio Radioattività Ambiéntale, A. Cigna CNEN, Centro di Studi Nucleari della Casaccia, G. Bagliano Rome Centro Informazioni Studi Esperienze, C. Triulzi Milan G. Bonfanti C. Giacoletto JAPAN Japan Analytical Chemistry Research Institute, T. Asari Tokyo M. Chiba Tokyo Kyoiku University, N. Ike da Department of Chemistry, F. Seki Tokyo H. Kirita National Institute of Radiological Sciences, M. Saiki Chiba Maritime Safety Agency, D. Shoji Hydrographic Division, M. Shiozaki Tokyo Tokai Regional Fisheries Research Laboratory, H. Tsuruga Tokyo T. Umezu N. Kawasaki IAEA-SM-180/9 333 COUNTRY AND INSTITUTION INVESTIGATOR JAPAN (cont. ) K. Tsutsumi Power Reactor and Nuclear Fuel Development Co. K. Akutsu Tokai Branch, Y. Imakuma Tokaimura K. Yamada H. Ejiri Institute of Public Health, Tokyo NEW ZEALAND National Radiation Laboratory, Department of Health, J.E . Dobbs Christchurch POLAND MarineStationP.A.N., Department of Geophysics, Polish Academy of Sciences, Sopot PORTUGAL Laboratorio de Fisica e Engenharia Nucleares, С . Cacho Junta de Energía Nuclear, A. О. de Bettencourt Sacavem M. De Moura ROMANIA SOUTH AFRICA Isotope and Radiation Division, D. van As Atomic Energy Board, C . M. Vleggaar Pretoria SWEDEN Section for Health and Safety, P.O. Agnedal AB Atomenergi Studsvik, O. Sundgren Nykoping UNITED KINGDOM British Nuclear Fuels Limited, H. Howells Windscale and Calder Works, Т.Н . Boyd Sellafield A.W. Dunn 3 3 4 FUKAI et al. COUNTRY AND INSTITUTION INVESTIGATOR UNITED KINGDOM (cont. ) Fisheries Radiobiological Laboratory, N.T. Mitchell Ministry of Agriculture, Fisheries and Food, J.W .R. Dutton Lowestoft UNITED STATES OF AMERICA Woods Hole Oceanographic Institution, V. T. Bowen Woods Hole H. D. Livingston R. Mann Oregon State University, N. Cutshall School of Oceanography, Corvallis Eastern Environmental Radiation Laboratory, D. G. Easterly Environmental Protection Agency, Montgomery Scripps Institution of Oceanography, T. R. Folsom La Jolla V.T. Hodge T .J. Tatum Institute of Environmental Medicine, T. Kneip New York University Medical Center, New York S. Jinks Radiological Sciences Laboratories, J. M. Matuszek New York State Department of Health, J.C . Daly Albany J. Hutchinson C. Paperiello Radiochemistry and Nuclear Engineering Research Laboratory, D.M. Montgomery National Environmental Research Center, J.W. Kearney Environmental Protection Agency, Cincinnati Bio-Medical Division, V. Noshkin Lawrence Livermore Laboratory, K .M . Wong University of California, Livermore Atlantic Richfield Hanford Company, G .C . Oberg Richland H. E. Smith C.P. McLoughlin Laboratory of Radiation Ecology, W.R. Schell College of Fisheries, R. King University of Washington, Seattle Argonne National Laboratory, J. Sedlet Occupational Health and Safety, F. S. Iwami Argonne N.W. Golchert IAEA-SM-180/9 3 35 COUNTRY AND INSTITUTION INVESTIGATOR UNITED STATES OF AMERICA (com. ) E. D. Wood YUGOSLAVIA Institute for Medical Research and Occupational Health, Yugoslav Academy of Sciences and Arts, B. D. Alica Zagreb J. Marica INTERNATIONAL ATOMIC ENERGY AGENCY A. Woein R. Fukai G. Statham S. Ballestra K. Asari IAEA-SM-180/21 PROBLEMS IN CALIBRATING INSTRUMENTS FOR ENVIRONMENTAL GAMMA EXPOSURE DOSE MEASUREMENTS J. JAGIELAK, B. GWIAZDOWSKI, J. PENSKO, A. 2.AK Central Laboratory for Radiological Protection, Warsaw, Poland Abstract PROBLEMS IN CALIBRATING INSTRUMENTS FOR ENVIRONMENTAL GAMMA EXPOSURE DOSE MEASUREMENTS. three types of measurement geometry have been assumed: (1) isotropic distribution of the radiation source; (2) infinite plane radiation source; (3) combination of planar and isotropic distributions. The comparison of calibration factors obtained theoretically for the various types of instrument are discussed and confirmed experimentally. 1. INTRODUCTION M easurements of gamma background exposure dose rate coming from natural and artificial radioactive isotopes are made under very different and variable environmental conditions. To obtain the true value of the Earth'sgam m abackground dose rate one has to apply several corrections which depend on the conditions of m easurem ent and type of the instrument used [1*3]. Calibration of the m easuring equipment must take into account the complex character of gamma background energy spectrum, i. e. photon energy up to 3 MeV, in addition to any directional dependence of the instrument response. The gamma radiation reference source used for calibration must be properly chosen, so that it sim ulates as well as possible the spectrum of the radiation to be m easured; the position of the source with respect to the detector should be also as sim ilar to the m easuring conditions as possible. This latter requirement is difficult to execute as the measuring geometry under field conditions varies greatly, depending on the 'shape' of the immediate surroundings. Over the past years, studies of the natural gamma background have been made by the Central Laboratory for Radiological Protection at W arsaw, using several different types of portable and stationary instruments designed for field measurement and continuous monitoring of the exposure dose rate [4, 5]. This paper presents the results of work done in deriving the proper calibration factors for these instruments. 337 3 38 JAGIELAK et al. 2. INVESTIGATIONS OF THE EARTH'S GAMMA BACKGROUND DOSE RATE, DISTINGUISHING BETWEEN RADIATION INCIDENCE DIRECTIONS TO AND FROM THE EARTH'S SURFACE To estim ate the calibration factor of an instrument used for m easure ments of the Earth's gamma background radiation emitted by the whole of the surroundings, it is necessary: (a) To know the directional dependence of the instrument response, presenting in some form details of reading versus the angle of radiation incidence; and (b) To choose an appropriate model of the directional distribution of the radiation incidence, representing in the best possible way the m easurem ent geometry found under field conditions. FIG . 1. Construction of the steel shielding containing two scintillation detector probes for stationary gamma background measurements from the Earth. (Styropian: an insulating material. ) IAEA rSM-180/21 339 In most cases calibration factor values for measurements of the Earth's gamma background are calculated either assuming an isotropic angular distribution [1, 3], or an infinite plane*source distribution [2, 6]. Investigations of the variation of the Earth's gamma background dose rate with radiation directions both to and from the surface of the Earth have already been reported by Burch et al. [7]. The static gamma-ray monitor they used in this work consisted of a set of three ionization chambers placed vertically one above the other. The chambers had appropriate outer shielding. The dose rate measured by the upper chamber, which was exposed to the radiation coming in the direction of the Earth's surface, was from a few to some 20 per cent of the dose rate m easured by the lowest chamber which monitored the radiation coming from the Earth's surface. The variations in the ratio measured by the chambers were assigned by the authors to passing radioactive clouds originating from nuclear weapons testing which was in progress at that time in the atmosphere. Investigations of this kind were also being made by our laboratory at Warsaw [8]. The stationary scintillation monitor designed for these m easurem ents is shown in Fig. 1. It consists of two identical scintillation probes with 5 in dia. X 2 in Nal(Tl) crystals. One of them is directed upwards and the other in the direction of the Earth. Both probes are placed in steel shielding of 150 mm in thickness. The distance between the lower crystal and the Earth's surface is 180 cm. The stationary monitor makes possible independent measurement of gamma radiation coming from the direction of the atmosphere and the radiation originating from the Earth's surface over an angle of approxi* mately 2?r. The dependence of the responses of the upper and lower detectors on the angle of incidence of the radiation, m easured for а ззбцд reference source, is shown in Fig. 2. Measurements made at our station in W arsaw have shown that the dose rate indicated by the upper detector amounts to 20% to 25% of the Earth's total gamma background dose rate. 3. DIRECTIONAL DEPENDENCE OF INSTRUMENTS USED FOR GAMMA BACKGROUND STUDIES Gamma background radiation measurements are made by our labora tory using several different types of instruments, including GM counters, scintillation crystal probes and high-pressure ionization chambers. The calibration of the instruments used for measurements of gamma background dose rate was done using point reference sources of дгШ is^I. The choice of these radioactive substances resulted from the accepted and confirmed assumption that the energy spectrum of the natural gamma background is equivalent to one of which 70% is made up of the gamma spectrum of ^ & R a (Q . g pt) and 30% of that of ^1 [9]. 3. 1. GM counter probes Two types of GM probes have been developed in our laboratory. Each of them uses six counters of the STS-6 type, placed in an aluminium casing. The cylindrical type of probe is 130 mm in diameter and 310 mm in height. The GM counters are placed parallel on a pitch circle 90 mm in 340 JAGIELAK et al. FIG. 2. Directional response of scintillation detector probes placed in the apparatus shown in Fig. 1. diameter. The other type of probe is of discus shape, 520 mm in diameter and 90 mm in height. It has the GM counters distributed radially and sym m etrically in one plane. The constructional details of the two GM probes are shown in Figs 3 and 4. The dependence of the responses on the direction of radiation incidence for ^ R a and ^ 1 sources are shown in F i g s 5 an d 6. 3. 2. Scintillation probes Investigations of the gamma energy spectrum of the Earth's gamma background were made by our laboratory using scintillation spectrom eters equipped with Nal(Tl) crystals. These instruments make it possible, after proper calibration, to estim ate the total exposure dose rate, as well as to determine the contributions of particular isotopes and isotope fam ilies to the total gamma background dose rate. A stationary monitor for continuous measurement of background radiation, constructed in our laboratory, is shown in Fig. 1; it contains two crystal scintillator probes. M easurements of the Earth's gamma background were also made using portable scintillation spectrom eters. The dependence of the response on direction of incident radiation for the scintillation probe containing a 44 mm dia. X 32 mm Nal(Tl) crystal, obtained using 226Ra and ^^1 reference sources, is shown in Fig. 7. IAEA-SM-180/21 341 /7/í//W/7¿¿//7? Pi/Zf/* FIG.4. Construction of the flat type of probe with GM counters. 342 JAGIELAK et al. 226Ra and 1311. IAEA-SM-180/21 343 FIG.7. Directional response of the probe with 44 mm dia. X32 mm Nal(Tl) crystal scintillator for primary gamma rays of ^Ra and ^4. 3. 3. High-pressure ionization chambers One of the high-pressure ionization chambers made in our laboratory is shown in Fig. 8. The high-pressure steel bottle, having a wall thickness of 4 mm, forms the outer electrode of the chamber. The volume is about 5 litres. It is filled with argon at a pressure of 35 atm. This type of chamber is used in the construction of both stationary and portable instruments. The dependence of the response of a stationary high-pressure ionization chamber on angle of incidence of the prim ary radiation from 226Ra an d ^ 1 sources is shown in Fig. 9. 4. DEPENDENCE OF CALIBRATION FACTOR VALUE ON THE MODEL OF RADIATION INCIDENCE ASSUMED AND THE MEASUREMENT GEOMETRY 4. 1. Isotropic distribution of the Earth's gamma background If isotropic distribution is assum ed, the mean values of the instrument indication are calculated from the directional response curves obtained 344 JAGIELAK et al. IAEA-SM-180/21 345 FIG.9. Directional response of the stationary high-pressure ionization chamber for primary gamma radiation o f !!6Ra and ИЦ. for each monitor, from measurements made in 47r geometry (Figs 2, 5, 6, 7, 9). The mean values of the response, N, were derived from the e q u a tio n : N g S in 8 de о where Ng is the instrument reading when the angle between the principal axis of the detector and the direction of the incident radiation equals 6. Calibration factors for the instruments used in our laboratory, obtained using *3*1 and 226Ra calibration sources, are shown in Table I. 4.2. Infinite plane radiation source If an infinite plane radiation source representation is chosen, this assum es that the main source of radiation consists of the radioisotopes contained in the ground and on the ground surface. The curve showing the fraction of total prim ary gamma radiation flux (Аф/ф) incident on the TABLE I. CALIBRATION FACTORS OF INSTRUMENTS FOR MEASURING THE EARTH'S GAMMA BACKGROUND OBTAINED USING AND s^Ra REFERENCE SOURCES mi reference source 226Ra reference source Is o t r o p ic ! ^ ï f a o n r a d ia t io n r a d ia t io n m d ^ ! o n distribution distribution distribution distribution " о и г с Г 4 ' u J c e " AILK t al. et JAGIELAK ^ R / h ) 4.85x10*2 4.75x10*2 4 . 7 7 x 1 0 * 2 4 . 8 6 x 1 0 * 2 4 . 7 2 x 1 0 * 2 4 . 7 5 x 1 0 * 2 \.counts/m in У i y p Ï Ô Î ^ b e ) / \ ^ p R / h ^ 4 . 9 8 x 1 0 * 2 4 . 8 7 x l 0 ' 2 4 . 8 9 x 1 0 * 2 5 . 1 1 x 1 0 * 2 4 . 8 8 x 1 0 - 2 4 . 9 3 x 1 0 - 2 (flat type of probe) ^ p R / h ^ N al(Tl) scintillation 3 . 4 x l 0 ' 3 3.07X10*3 3.14x10*3 8.5 xlO-3 7.37 xlO-3 7 . 6 0 x 1 0 * 3 counts/min / crystal probe / \ ( p R / h ) 4 . 4 8 4 . 4 0 4 . 4 2 3 . 2 9 3 . 2 7 3 . 2 7 \ m V / s / ^i^iization^ c h a m b e r IAEA-SM-180/21 347 FIG. 10. Fraction of total primary gamma ray flux (Ap/p) incident on the detector at angles of 0± 5° to the crystal axis from half the ground space. Curve A is for a photon energy o f 3 MeV; Curve В for a photon energy of 0.5 MeV. detector versus the angle of incidence 8 for half the ground space is given in Fig. 10 for two incident photon energies [2]. The shapes of the curve show that response is largely independent of the incident photon energy. It can be seen that m ost of the photons arrive at the detector at large angles. From this dependence, and from the angular response charac teristics for and 226Ra radiation, calibration factors for the measuring equipment have been calculated. The results are shown in Table I. 4. 3. Complex model of the Earth's gamma background (directional distribution Taking into account the results of investigations of the directional distribution of the Earth's gamma background and the observations made during the field m easurem ents, we have taken a new complex model for the directional distribution of the gamma background radiation. The calibration factors were calculated with the assumption that 20% of the background radiation is directed towards the Earth's surface and has an isotropic distribution, while the other part (80%) is directed upwards from the Earth's surface and that this follows an infinite plane radiation source type of distribution. The values of the calibration factors calculated on this basis for and ^&Ra radiation are also included in Table I. 348 MGIELAK et at. TABLE II. CALIBRATION FACTORS FOR GAMMA BACKGROUND MEASURING INSTRUMENTS, ASSUMING THAT THE BACKGROUND SPECTRUM IS EQUIVALENT TO A RADIATION CONSISTING OF 30% Mil AND 70% 226Ra EM ITTED GAMMA RADIATIONS Infinite Complex plane radiation radiation radiation distribution distribution ^ pR/h 4 .86x10*2 4.7 3 x1 0 *2 4 .76x10*2 \counts/min / ^ Ï c y l ^ c a ï type of probe) / pR/h ^ GM counters 5 .0 7 x l0 *z 4 .88x10*2 4 .92x10*2 \counts/min / (flat type o f probe) ^ pR/h ^ Nal(Tl) scintillation 6.97X10*3 6.08x10*3 6.26x10*3 crystal probe fR /h ^ High-pressure 3.65 3.61 3.62 \ mV/s / ionization chamber 5. DISCUSSION The comparison of calibration factors obtained from measurements w ith 226Ra and ^ I sources and the subsequent calculations for GM counter probes and ionization chambers show that their dependence on the m easure* ment geometry under field conditions is sm all, not exceeding 5%. The difference in the calibration factors of scintillation probes is much more significant, being 10% to 15%. The values of calibration factors for natural gamma background measurements are given in Table II. They were calculated in the assumption that the m easured radiation gamma spectrum is equivalent to one consisting of 30% of an ^ 1 and of 70% of a ^ R a gamma radiation spectrum. The choice for calibration factor calculations of the proper model for the directional distribution of the gamma background radiation is of great importance when the instrument used has a considerably anisotropic response. An example of such a case is the scintillation probe discussed previously, containing an Nal(Tl) crystal. If the environment was to become contaminated with artificial radioactive products, the changes in the gamma radiation spectrum would m ost probably show an increase in the low energy region. That may cause an increase of the anisotropy in the instrument response. The choice of a proper model for the measurement geometry, as close as possible to the existing conditions for calculating the calibration factors, is insuch a case much more important than in the situation normally pertaining when m easuring the natural gamma background radiation. IAEA-SM-180/21 3 4 9 R E F E R E N C ES Handl. Ser. 4 6 3 (1956). [2] BECK, H .L., CONDON, W .J., LOWDER, M. W., Spectrometric techniques for measuring environmental gamma radiation, USAEC Rep. HASL-150 (1964). [3] PEÑSKO, J . , Earth Gamma Background Measurements in Poland, PTJ 45 (1968). [4] PEÑSKO, J . , JAGIELAK, J . , Methods of earth gamma background monitoring used in Central Laboratory for Radiological Protection, Acta Geophys. Pol. 18 2 (1970). [5] PEÑSKO, J . , JAGIELAK, J ., GWIAZDOWSKI, B., ZAK, A ., Automatic methods of continuous measure ment of earth gamma background exposure dose, Rep. CLOR-92/D (1972). [6] BECK, H .L ., LOWDER, W .M ., BENNETT, B .G ., CONDON, W .J., Further studies o f external environmental radiation, USAEC Rep. HASL-170 (1966). [7] BURCH, P .R .J., DUGGLEBY, J .C ., OLDROYD, B., SPIERS, F .W ., "Studiesof environmental radiation at a particular site with a static y -ray monitor", The Natural Radiation Environment (ADAMS, J. A. S ., LOWDER, W .M ., Eds), Univ. Chicago Press (1964) 767. [8] PEÑSKO, J . , JAGIELAK, J ., Static gamma background monitor for identification of contamination and dose-rate in the case of dispersal radionuclides to the atmosphere, Nukleonika 14 7 (1969) 831. [9] SPIERS, F.W ., McHUGH, M .J., APPLEBY, D .B ., "Environmental y-ray dose topopulations: surveys made with a portable meter", The Natural Radiation Environment (ADAMS, J.A .S ., LOWDER, W.M., Eds), Univ. Chicago Press (1964) 885. IAEA-SM-180/65 A NEW LOW-BACKGROUND COUNTING FACILITY FOR INSTRUMENT CALIBRATION S. ABE National Institute of Radiological Sciences, Chiba-shi, Japan Abstract A NEW LOW-BACKGROUND COUNTING FACILITY FOR INSTRUMENT CALIBRATION. Various types of radiation instruments are used for measuring low levels of radiation in the field and in environmental surveillance programmes. These instruments are generally calibrated at radiation levels as high as 100 times the natural background in order to reduce the effect of background radiations. However, radiation levels normally measured by these instruments differ little from natural background levels. The discrepancy between calibration and measurement levels produces difficulties in assessing and assigning a level of uncertainty to the measurements. A low-background facility has been planned to overcome these difficulties, and details of its constructional philosophy and practical requirements are presented. The room, which is underground, will also prove useful for studying and checking the low level measurement behaviour of various instruments. The aim is to achieve an overall background in the low-background room of one hundredth of the natural level at the ground surface. 1. INTRODUCTION Instruments that can m easure low levels of radiation are very much in demand for practical monitoring, and various types have been used for field measurements or for environmental surveillance program m es. Often, these instruments are calibrated or inter compared rather sim ply and indifferently in a norm al environment. However, background radiations which originates from natural sources and from artificial substances are present to some degree everywhere and it is inevitable that measurem ents are affected by them; this usually means that calibrating radiation sources must be strong enough to render the effect of the environmental radiation negligible. In practice, therefore, instruments for measuring low levels of radiation are calibrated at radiation levels of 100 tim es as high as the level in the norm al radiation environment. In fact, the norm al radiation levels to which I refer differ little from natural background levels, and may be two orders of magnitude lower than the calibrated level. The dis crepancy between levels to be m easured and levels to be calibrated gives rise to an uncertainty in the assessm ent of measured results. A first step to removing the discrepancy requires that the instruments be calibrated at normal radiation levels. I describe in this paper a new low-background counting facility that can be used for such calibration and other procedures that require a low background. 2. CHARACTERISTICS REQUIRED OF THE ROOM It is necessary that the radiation background in such a room is kept very low, that the level of scattered radiation is low and that its spectrum shows no significant peaks. 351 352 ABE The required radiation level in the room should, ideally, be lower than 1/100 of the norm al radiation level. It is difficult to obtain such low radiation levels throughout a large volume, so that the hundredth-part requirem ent will be thought of as the limiting one. The low-background character is also required because measurements at natural radiation levels are affected by the inherent noise in the instru ment used. Knowledge of noise or zero-point error are rarely available, and are a cause of difficulty in such studies. The requirem ents on scattering are important when calibration is conducted at a natural and/or lower radiation level from a weak radiation source. If the scattering intensity is high, or if peculiar peaks appear in its energy spectrum, the calibration may give rise to an ambiguous r e s u l t . The indispensability of such a low-background room in radiation studies is sim ilar to the need for an anechoic room in acoustics work and for a shielded room in electromagnetic wave studies. In each case, the 'waves' from the surroundings are screened out and scattering in the room is reduced to levels that are as low as possible. 3. PLAN OF THE PROPOSED FACILITY This facility consists of three parts; they are, the low-background room, the low-level counting room and the main building. The outlines of them are shown in Table I, Fig. 1 and Fig. 2. TABLE I. DETAILS OF THE LOW-BACKGROUND ROOM FACILITY Low-background Low-level room counting room Main building (LBKG room) (LLC room) Depth below ground (m) deeper than 20 deeper than 3 0 Shielding (wall thickness): Lead equivalent (cm) 10 10 0 Plastics (m) 3 2 0 Cleanliness of ventilated air very high high normal Ventilation rates (changes/h) 1/10 10 10 Floor space Rooms (m^) 20 24 1843 Corridor (m^ ) 53 21 16 Elevator (m^ ) - 4 - 4 - 4 ^ Contains air-conditioning equipment, electronics room, preparation room, changing room and monitoring room. IAEA-SM-180/65 ),Up main building Л 2 0 0 m * Г7 V /"7 / / / 7*7*77*7 77*77*1 V 7*1 *77*71 *T7" 3m LLCroom LLCroom lead equivalent 10 cm ptastics 2 m room space 24m* r— !L 20m ^-77 7 7 77 LBKG room lead equivalent 10cm plastics 3m LBKG room room space 20m* 11m 15m elevator -plastics v777Z^Jalead or others ^concrete FIG. 1. Vertical cross-section of the facility. FIG. 2. Horizontal cross-section of each(LLC and LBKG) room. 3.1. The low-background room The low-background (LBKG) room, in which radiation instruments can be calibrated and other studies to be mentioned later can be perform ed, is the principal room in the facility. The room is placed well below ground to reduce the effects of cosm ic radiation. The feature of the room is that both requirem ents (low background and weak scattering) are well satisfied. In particular, extreme care has been taken to obtain good scattering reduction properties. It is worth rem arking that a low-level counting room in which only precise counting of very low-activity sam ples is to be undertaken does not require the scattering reduction features described for the low-background room. However, neither an exact calibration nor a study of inherent noise can be undertaken in a volume containing scattered radiation. Since the radiation is at a level corresponding to the natural one or below it, it is a key point that the LBKG room be arranged to satisfy both requirem ents. The space units that make up the area (Fig. 2) are the laboratory room, the access corridor and the elevator. These space units are listed in T a b le I. The ventilation system of the room com prises two system s, each with two parallel units. One system provides the ventilation and cleaning of the room air. The second system provides the ventilation and flushing through the wall grooving. The form er (Fig. 3) com prises a prefilter, a cooling unit for precounting, a cold trap for trapping the radioactive rare gases, 3 5 4 ABE .separate, 1st unit systems ** 2nd unit outdoor air ) J/ . separate . ¡ prefitter ] 1st unit systems 2nd unit i ! jcooting unit ) ------cooling*1= ------trap ¡г ) A t 1 HEPA fitter ) A ! 1 air condition] )_____ ^____ J/ ______!____ ] fan ) 3 : HEPA fitter tow background room FIG. 3. Room ventilation system of the LBKG FIG. 4. Wall-grooving ventilation system of room, including a stand-by unit. the LBKG room, including a stand-by unit. radon and thoron, a high efficiency particle filter in the air flow (HEPAfilter) for filtering out radioactive particles, an air conditioner, a fan for providing positive room pressure, and a wide HEPA filter for adjusting both the air pressure and flow in the room. Two parallel units are provided to ensure non-stop operation of the room ventilation and air cleaning system . The radioactivity in the air in the room as well as that drawn in from outside are cleaned by the system . The wall-grooving ventillation^ system (Fig. 4) is designed to prevent accumulation of radon, thoron and their daughters originating from the wall m aterials. In this system , the cleaned air flows through a multitude of grooves cut into the plastic on the surface boundary between the plastics surface and the lead lining. The system also contains a parallel stand-by u n it. 3.2. - The low -level counting room The low-level counting (LLC) room has the same function as the common low-level counting facilities which are used for the counting of low-activity sam ples. Low-background detectors are also provided for counting low-activity sam ples, screening of sam ples andto enable prelim inary studies to be made. IAEA-SM-180/65 3 5 5 separate 1st unit ^ systems 2nd unit Г— ^ ------1 [ prefitter [ - Г * ' I I HEPA filter [ t_ i ^ ------, air condition T I fan " 1 fan HEPA fitter towteve) counting . .-watl i '"FI?"! FIG. S. Ventilation system of the LLC room, including a stand-by unit. The area ( see Fig. 2, Table I) is placed below ground level and com prises the laboratory room, a corridor and the elevator. The air cleaning and ventilation system of the room (see Fig. 5) is sim ilar to that in the LBKG room, differences being that the LLC room has no cold trap and only one unified system (i. e. no wall grooving ventilation system ), though also here a parallel stand-by unit is provided. 3.3. Main building The main building is above-ground, and com prises a room housing the air-conditioning equipment, an electronics room, a preparation room, a changing room, a monitoring room and the elevator hall. The ventilation and air conditioning system of the main building is an ordinary one. Work carried out in the building involves sample preparation, etc., supplying ventilating air, and checking access to the LLC and LBKG rooms. 4. SPECIAL CARE TAKEN WITH THE STRUCTURE OF THE FACILITY 4. 1. Reduction of background radiation The m eans used to reduce background radiation are listed in Tables 11a and b. There are two categories of background radiation. Into the first category fall the direct and indirect radiations originating from space, 3 5 6 ABE TABLE Ha. REDUCTION OF TABLE 11b. REDUCTION OF BACKGROUND RADIATION IN THE RADIOACTIVITY IN THE FACILITY - DIRECT FACILITY - RADIATIONS FROM RADIATIONS RADIOACTIVITY IN THE FACILITY 1. Underground rooms 1. The supply of cleanr air, i.e. activity- (Reduction of contribution of cosmic rays) free air and particulate-free air (Reduction of radiations from radon, thoron 2. Shielding with lead or iron and their daughters in the room air) (Shielding radiation from both surrounding soil and concrete walls) 2. Ventilation through wall, ceiling and Яоог grooving with clean air 3. Lining inside with plastics materials (Preventing accumulation of radioactive gas (Reduction of radiation from outer walls originating in building materials) and soil, and improvement of scattering properties) 3. Waterproof outer walls (Preventing radiation from radioactive 4. Maze substances in drainage water from entering) (Shielding of radiations from constructional materials) 4. Positive room-pressure (Prevention of the in-leaking of air from 5. Use of low-activity materials for wiring and the outside) plumbing systems 6. No doors the Earth's atmosphere, lithosphere, hydrosphere, and the walls and equipment. Into the second fall radiations originating from the radio activity that is carried in the room air, and that is accumulated naturally through and/or in the wall m aterials. Reduction of the radiations (Table II) is limited to shielding, reduction of scattering by keeping a large physical separation between instruments, sources and the walls, and the use of the least active constructional m aterials. In the room, adequate plastic and metal shielding is provided made of low-activity m aterials. Reduction of cosm ic ray background is provided by the underground construction. To reduce the effects of scattered radiation components, a maze-type of entry is used (i. e. no doors), in addition to keeping the volume of the room large. The reduction of the radioactivity (Table 11b) is ensured by supplying clean air (radioactivity and particle free) for ventillation, and providing a positive room pressure, etc. All these efforts reduce the total radiation background level in the LBKG room (in comparison with the natural background at ground- level) to the values corresponding to those shown in Table III. 4. 2. Reduction of scattered radiation To reduce the scattered radiation, it would be theoretically possible to reduce the air pressure. This is not a practical possibility and hence the only reduction can be achieved by wall design (Table IV). IAEA-SM-180/65 357 TABLE III. EXPECTED RADIATION L E V E L I N T H E R O O M 1. Contribution of cosmic radiation (ionizing component) < O.lpR/h 2. Contribution of radiations from soil and wall materials y-ray < 0.05 pR/h 3. Radon and thoron concentration in the room air < 1 x 1 0 " ^ Ci/cm^ air TABLE V. PREVENTION OF TABLE IV. IMPROVEMENT OF RADIOACTIVE CONTAMINATION SCATTERING PROPERTIES OF AND DECONTAMINATION T H E ROOM S 1. Use of plastics (low-Z) for the inside wall lining 1. Use of a positive-room pressure system (Prevention of interfering peaks from scattered radiation) 2. Care in choosing wall-lining materials 2. Keeping the room volume as large as possible 3. Prohibiting the taking of unnecessary or (Reduction of scattered radiation) open sources in the room 3. Use of special shaped surfaces on inner walls 4. Monitoring of persons and objects before (Reduction of scattered radiation) entering 5. Non-stop operation of the ventilation and cleaning system (Use of two parallel systems) (Stand-by electricity supply) TABLE VI. ELECTRICITY TABLE VII. SAFETY SUPPLY CONSIDERATIONS 1. Reduction of electrical noise l. Fire safety (1. ^Electrical shielding of the room) (1. Use of incombustible materials for (2. Perfect earth connections) building construction, lining, plumbing, (3. Use of shielded cables) wire e tc.) (2. No flam es or open lights in the room) 2. Use of stabilized and noiseless electric (3. No smoking) 2. Maze-type exit (Stand-by system, i.e. batteries + diesel generator) 358 ABE TABLE VIII. STUDIES TO BE TABLE IX. FUTURE APPLICATIONS MADE IN THIS FACILITY OF THIS R O O M 1. Instrument calibration at low radiation 1. Low-level counting room levels using weak sources 2. Study of cosmic rays 2. Study of the inherent background of instru- instruments 3. Room for whole-body counting in a low level radiation field 3. Study of stabilities of instruments which affect results measured at normal 4. Processing and preparation of some materials radiation levels in a low-background radiation environment 4. Intercomparison of various instruments 5. Biological studies in a low-level radiation regarding their detection properties in a low- environment level radiation field 4. 3. Other planning Other requirem ents are set out in Tables V to VII. 5. CONCLUDING REMARKS The studies that will be a feature of work in the facility are listed in Tables VIII and IX. I believe such facilities should be available throughout the world, making for better measurement accuracy and for more useful intercom parisons of low-level radiation results. DISCUSSION J.E. GUTHRIE: Have you already built the room which you describe? S. ABE: No, but we plan to in the near future. J. E. GUTHRIE: What background level do you expect to have inside th e r o o m ? S. ABE: About l/lOO of the natural background level. IAEA-SM-180/2 RAPID DETERMINATION OF RADIONUCLIDES IN MILK Results of an intercomparison organized jointly by the IAEA and CEC O. SUSCHNY, J. HEINONEN, D. MERTEN International Atomic Energy Agency, Vienna J. SMEETS, R. AMA VIS, A. BONINI Commission of the European Communities, Luxembourg Abstract RAPID DETERMINATION OF RADIONUCLIDES IN MILK. Results of an Intercomparison organized jointly by the IAEA and CEC. An intercomparison of the rapid determination of radionuclides in milk at concentrations which might be expected in the vicinity of a nuclear installation in the case of an unplanned or accidental release of radioactive products was organized jointly by the IAEA and CEC. The purpose of the project was to enable analytical work and to help administrative departments who are responsible for the protection of the public from undue exposure to radiation and radioactive materials to assess the reliability of the data which, in any real case, would form the basis for administrative action. The radionuclides employed were strontium-89 and 90, iodine-131, caesium-137 and barium-140. Two solutions containing mixtures of these nuclides at two different concentration levels (known to the organizers but unknown to the participants) were analysed at 34 laboratories. Preliminary results were to be obtained and reported after four hours time for analysis: this done, ample time was provided to check the results and report final values. In the case of the three gamma emitters this time schedule was adhered to by most analysts, the strontium isotopes, however, proved more difficult and caused delays at some laboratories. Of the 354 results reported for all the gamma emitters, 9 had to be rejected as ' outliers', while the standard deviation of the accepted values from the mean varied from 6.7% to 18.7% for the individual nuclides. The corresponding figures for the strontium were: 126 results received, 8 excluded, standard deviation 11.3% to 45.3%. 1. INTRODUCTION The safety record of nuclear industry is satisfactory and the probability of unplanned releases of radionuclides in significant amounts is extremely sm all. Nevertheless, it is the duty of the national and international authorities responsible for public safety in this field to cater for all possibilities, including that of serious releases. Despite this very low probability of emergencies occurring, which would necessitate rapid availability of analytical results concerning the extent of radioactive contamination, both the CEC and IAEA have felt for some time that it would be worth while to check on the rapidity and accuracy with which the laboratories concerned could provide such data. On 28 and 29 October 1969, the Commission of the European Communities held a meeting of experts from its Member States to study the establishment of a comparison programme on the rapid measurement of radioactive conta mination of milk (see Annex III). 3 5 9 3 6 0 SU SCHN Y et al. A representative of the International Atomic Energy Agency, who had been invited to this meeting, drew attention to an existing programme prepared by the Agency. The collaboration between the two organizations, resulting in an intercomparison organized in January and February of 1972, was based on this programme. At a further meeting organized by the Health Protection Directorate of the CEC (General Directorate V) which took place on 3 and 4 July 1972, the delegates of the laboratories of the Member States of the European Community, together with a representative from the IAEA, drew up a list of data obtained, and compared their results and the analytical methods adopted for measuring the radioactive pollution of milk (see Annex III). This report sets out the results of this first joint programme and gives a critical analysis of it. 2. AIM OF THE PROGRAMME The purpose of this project was to enable laboratories in Member States of one or both of the organizations to check the speed and accuracy of their analytical work. In addition, the intercomparison was considered to be useful also to those administrative departments of national or international scope in whose responsibilities lie the protection of the public from undue exposure to radioactive contamination in the event of a nuclear incident. The results of the intercomparison would serve to show them what level of reliability they could attach to reported results of contamination measurement which, in any real case, would form the basis for their administrative action. Milk was chosen because it is an important vehicle for radioactive contamination when an accident occurs. The nuclides selected for use in the exercise were those normally analysed for in milk when contamination with radioactive products (including fresh fission products) is suspected, namely, ^Sr, 90g^ 131^ I37^g For the choice of radionuclide concentrations we considered, on the one hand, the values of DWL and of ERL proposed by Bryant [ 1] and, on the other hand, the existing rules for the transport of radioactive m aterials. The activities of the two sam ples (in nCi/litre) obtained at source, were as follows: Sample Sample low level medium level (tiCi/1) (nCi/1) "S r 1 20 *Sr 1 1 " h 10 100 4*Cs 10 100 *°B a 2 20 These are rough values. Altogether, 34 laboratories took part in the exercise, two of these submitting two different sets of results each, obtained by different analysts, making a total of 36 sets of results. IAEA-SM-180/2 3 6 1 18 laboratories submitted their results through CEC, the rest reported directly to the IAEA. A form had been drafted for reporting (see Annex I). The list of laboratories and the person responsible (usually either the head of the laboratory or the responsible analyst) is given in Annex II. 3. PREPARATION OF SAMPLES For technical reasons (difficulty of transportation and preservation of spiked milk, and also considering costs), the spike solutions were provided in aqueous form in ampoules, and laboratories were asked to prepare their spiked milk by pouring the contents of these ampoules into fresh milk. Two levels of activity were provided, one to give 2 to 10 nCi of gamma emitters per litre of milk and one to give ten times as much, with ^°Sr equal to 1 nCi/l in each solution. Standard solutions of the gamma emitters and of ^°Sr were provided at the same time to allow laboratories to calibrate their counting equipment and so to eliminate differences due to calibration. A time schedule was provided which required preliminary calibration, followed by four hours of analysis and measurement; after this time preliminary results were to be transmitted by cable, telex or telephone. More accurate final results were to be sent off later. One difficulty experienced in the preparation of the exercise was the reluctance of isotope suppliers to supply ^°Ba and ^Sr for calibration at the same time. None of the big suppliers would oblige and, when finally a sm all company agreed to supply the two isotopes separated from a fission product solution, the S^Sr solution on delivery was found to be quite empty. Much later, when laboratories had already been informed that they would receive none of this nuclide, the latter was discovered to be present in great abundance as a contaminant in the ^°Ba solution and, therefore, also in both spikes. This caused some difficulty to analysts who had been told not to expect ^Sr and, probably, led to a somewhat larger error than expected, as well as to some delay in the determination of both strontium iso to p e s. 4. PRESENTATION OF RESULTS Tables I - XII show the results provided by the individual laboratories for the different radionuclides at "low" and "medium" spike level Tables I - VI give the results for the following radionuclides: ^1, ^ C s, n ° B a . Some laboratories analysed ^Sr and ^°Sr separately, see Tables VII - X. Others determined the sum of the two; their results are shown separately in Tables XI and XII. The accuracy quoted in columns 3 and 6 of Tables I to XII for the preliminary and the final results, respectively, is that claimed by the laboratories themselves, who were asked also to state the confidence level of their estimated accuracy. The last two columns in each table give information on the time taken for the analysis. Most laboratories succeeded in measuring the gamma activities in much less than the four hours allowed for the total analysis of all nuclides; a few also finished the determination of ^^Sr in this time. ^ Names of laboratories are replaced in these tables by a numerical code which does not correspond to the sequence of these laboratories in the Annex. 3 6 2 SU SCHN Y et al. Tables XIII and XIV summarize the results for low and medium spiked solutions and all the radionuclides analysed. They show the range of results obtained, the (generally small) number of results excluded, the range of those results which were used in the calculation of the mean and the deviation of this mean from the calibration value. For more clarity, results are shown also in graphical form: Figs 1-8 show the frequency distribution of results for the gamma emitters and for 9°Sr. The number of results obtained for ^Sr was considered too small to warrant a separate figure. Figure 15, Administrative data, shows for each laboratory of the Member States of the European Community the time elapsed between receiving the samples, sending the preliminary results and sending the definitive results. This figure shows the length of time needed by each laboratory to obtain results. Where there is a considerable time lapse, it is due to problems within an institute (administrative and organizational problems). 5. INTERPRETATION OF RESULTS The mean value shown by the dotted line in Figs 1-8 was calculated from the results obtained after screening of extreme values according to their statistical probability. The method used for this purpose was that of Graf and Henning [2] as reported byDoerffel [3], which considers permissible scattering lim its as a function of the experimental standard deviation s, the number of observations N and the required probability P. A value x^+i may be regarded as outside perm issible limits and discarded if it differs from the mean value x of all accepted results by more than a multiple of s defined by the statistics g(P, N): XN+i ^ x ± g (P, N) - s The factor g(P, N) for a probability of 95% (the value used in our test) and a number of 20 to 30 results is about 3. 8, which means that in most cases only values falling outside the range x ± 3. 8 s were excluded from the evaluation. In some cases in which the number of results available was sm aller, rather larger deviations had to be tolerated. F ig u r e s 9-14 are based on still another approach used to screen results: these figures are "control charts" in which results are classified according to whether they fall within the "warning lim its": x ± 2 s outside of these, but still within "control lim its": x ± 3 s or even outside the control lim its. When examining the accuracy figures given by the laboratories and reported in columns 3 and 6 of Tables I to XII, it turned out, however, that different laboratories used different methods in making these estimates, IAEA-SM-180/2 3 6 3 some only providing their statistical counting error, so that the figures in these columns cannot really be compared with each other. They are left in this document merely for comparison with the actual, experimentally determined deviations from the calibration values which are given in columns 4 and 7. There is little apparent correlation between these two sets of columns. In the case of ^^Sr and of S9+90g^ which no calibration value was available, these columns are used to show the deviation of the individual values from the experimentally determined mean value. With regard to the information provided in Tables I to XII regarding the time taken for analysis, it must again be considered with some caution, since here also not all laboratories used the same criteria, some only counting actual working time, others including time for the establishment of the required partial equilibria of daughters, some timing individual radionuclide determinations, others measuring the total time required for the separation and counting of all estimated radionuclides. It can only be said that the number of laboratories to finish the measurement in the allotted time would most probably have been much higher had the possible presence of ^Sr been indicated in the information given to them. Many laboratories, however, were not in a position to estim ate 9°Sr in the presence of both ^Sr and ^°La (daughter of ^°Ba) in the short time allowed. Perhaps the most important information given in Tables XIII and XIV is the standard deviation for each nuclide, given in (nCi/l) and also in (%). This standard deviation is shown to change very little from preliminary to final results. The difference between the observed mean and the calibration value which is shown in the next line is sm all and generally within the expected scattering lim its. All this proves that the use of more time- consuming methods in place of the rapid ones adds very little precision to the analysis and that the use of the rapid methods in emergency situations does not lead to undue error. There is also little difference between standard deviations obtained on the intermediate level as compared to the low-level solution but it should be remembered that the "low-level" solution used in our experiment still contained enough radioactivity to permit measurements to be carried out without being much influenced by reagent contamination or by radiation background fluctuations. All values within warning lim its (see Tables XV and XVI) are considered completely acceptable, those between warning and control lim its are considered as still tolerable but probably indicative of some analytical shortcomings which should be put right soon; finally, those outside the control lim its are considered as unacceptable. Laboratories producing unacceptable values should immediately check and, if necessary, improve their techniques. Until such improvement is documented (or the reason for an individual unacceptable value explained) results from the laboratory concerned may have to be regarded with caution, particularly if administrative decisions are to be based upon them. The position of values in the control charts is shown in Table XV. It can be seen from this table as well as from the figures that the number of results falling outside the control lim its is rather sm all, which indicates that most of the laboratories which took part in this intercomparison were per forming well in the analysis of gamma emitters. Also, the overall analytical situation seem s to be well in hand, at least for caesium-137 and iodine-131, which give standard deviations of 6% to 8%. The situation is less favourable in the case of ^°Ba, where standard deviations of 13% to 18% were found. 3 6 4 SUSCHNY et al. Results obtained on the two strontium isotopes are more difficult to evaluate. A rather arbitrary classification upon which Table XVI is based, distinguishes between results which fall within 30% of the calibration value and those which fall outside these lim its. In the consideration of the figures in this table we must bear in mind that no warning was given to laboratories to expect 39Sr in their sam ples and no standard was provided for this nuclide. The comparatively large margin of error carried by the analytical results may at least partially be explained by the difficulties caused by these facts, which also caused delays and increased errors in the preliminary deter mination of 90Sr. However, as is shown by the final results for 9°Sr, which were obtained by most laboratories after receiving information on the presence of S9Sr and which still carry a standard deviation of between 20% and 26%, there are other factors inherent in the analysis for the strontium isotopes which lead to a larger scatter of results. In contrast to the three gamma emitters, which were analysed by direct gamma spectrometry by most laboratories, or after chemical procedures involving a minimum of manipulation, the pure beta em itters ^Sr and^^Sr had to be isolated for measurement, distinguished from each other by radiation absorption, a method in which sm all errors in counting or in calibration may lead to large errors in 9°Sr values if that nuclide is present in small quantities compared to or by separation and meaáurement of 30Y, complicated by the presence of the barium-daughter ^°La. The final beta measurement is also more influenced by absorption, self-absorption and backscatter than the more reproducible gamma measurement. All these factors would be expected to lead to larger scatter of results for the strontium isotopes than those obtained with the other nuclides. The actual figures obtained nevertheless seem rather high. 6. CONCLUSIONS On the whole, the intercomparison of rapid radionuclide determinations in milk organized by the IAEA and CEC in 1972 has shown that in the case of a possible emergency situation the laboratories are able to provide the responsible authorities with rapid and reliable results when the gamma emitters ^1 and ^ C s are to be determined. Somewhat larger scatter must be expected in the analysis of ^°Ba. The pure beta emitter 9°Sr when present together with relatively large excess of ^gr may be difficult to determine separately and either a relatively large error, depending upon the S9gr/90gr ratio (when the two nuclides are determined in the same precipitate and distinguished only by use of their different maximum beta energies) or a considerable delay in the analysis, when this is carried out, after separation of strontium from barium and lanthanum, by partial equilibration of 9°Sr with its daughter 9^Y and measurement of the latter, may have to be tolerated. ACKNOWLEDGEMENTS The help provided by Dr. H. Houtermans of the Physics Section of the IAEA, who was responsible for the calibration by absolute measurement of all standards and spiked solutions, is gratefully acknowledged. IAEA-SM-180/2 365 TABLE I. RESULTS OF ^ 1 ANALYSES BY DIFFERENT INSTITUTES ON SPIKE SOLUTION A, LOW LEVEL Calibration value 10. 14 nCi/l: P = preliminary; F = final ts F Met iod Total s á E ^ d usee fot Lab. (he No. .2 -S > Д (n C i/1 ) (tiC i/1 ) ( * %) d (%) ( * %) d(%) P F P F 1 8 .8 0 5 - 1 3 . 2 9 .4 0 5 - 7 .3 2 2 4 * 1 9 .0 * 2 9 .5 0 5 - 6 .3 1 7 .5 * 3 10.00 20 - 1 .4 1 8. 0* 4 9 .4 2 5 - 7.1 9.42 5 - 7 . 1 2 2 2.8 2.8 5 9 .5 0 5 - 6 .3 9 .6 0 5 - 5 .3 3 3 4 . 0 6 1 0 .6 0 3 + 4 . 5 3 1 .4 * 7 1 0 .3 0 5 + 1.6 1 2 . 5 8 1 2 .6 0 5 + 2 4 .2 1 1.2 9 8.20 10 - 1 9 . 1 2 0. 6* 10 8 .7 0 10 - 1 4 . 2 2 0. 6* 11 9 .9 0 10 - 2 .4 1 0.6 12 11.20 7 + 1 0 .5 1 1 .1 9 5 + 1 0 .4 2 2 0.2 1 .3 13 -1 0 .5 0 5 + 3 . 5 3 1 0 .7 14 9 .7 0 10 - 4 .3 9 .7 0 10 - 4 .3 1 2 1.2 1.2 15 8 .6 7 20 - 1 4 . 5 8 .6 5 20 - 1 4 . 7 2 2 1. 0* 2 .5 * 16 1 0 .4 0 20 + 2 . 5 1 0 .9 17 1 0 .4 2 + 2 .5 1 0 .3 2 + 1.6 1 2 1 .3 * 1. 3а 18 9 .8 0 2 - 3 .4 10.10 1 - 0 .4 2 2 0.6 2.1 19 9 .7 5 1 - 3 . 8 9 .7 5 1 - 3 . 8 1 1 1. 0* 1. 0* 20 9 .8 0 2 - 3 . 8 9 .8 0 2 - 3 . 4 4 4 0.8 0.8 21 10.00 20 - 1 .4 4 1. 0 22 9 .1 9 5 - 9 .4 9 .5 0 4 - 6 .3 1 1 0 .5 * 1. 6* 23 10 12 - 1 .4 10.00 6 - 1 .4 2 2 1. 8* 3 . 5 * 24 1 0 .4 9 15 + 3 .4 1 0 .4 0 15 + 2 . 5 4 4 1.0 120. 0* 25 1 0 .6 0 10 + 4 . 5 1 0 .6 0 10 + 4 . 5 1 1 2 .5 2 . 5 26 8 .6 0 15 -15.2 9.00 10 - 11.2 3 3 3 . 6 * 1 4 .5 * 27 1 0 .9 6 10 + 8.1 1 0 .8 0 10 + 6 . 5 3 3 2.0 2.0 2 8 1 0 .3 8 1 + 2 . 4 1 0 .3 8 1 + 2 . 4 1 1 1 .3 1 .3 29 10.69 + 5.4 10.69 + 5.4 4 4 1.0 1.0 3 0 1 0 .2 7 9 + 1 .3 1 0 .2 7 9 + 1 .3 3 3 2 . 1* 2 . 1* 3 1 1 0 .3 0 10 + 1.6 1 0 .3 0 10 + 1.6 2 2 1 .5 1 .5 32 1 0 .8 0 3 + 6 .5 1 0 .8 0 3 + 6 . 5 2 2 1 .5 1 .5 33 9 .7 0 4 - 4 . 3 9 .7 0 4 - 4 . 3 1 1 1 4 .6 * 1 4 .6 * 3 4 10.00 15 - 1 .4 1 3 . 5 * 3 5 9 .1 9 10 - 9 .4 3 9 .0 * 36 366 SUSCHNY et al. TABLE II. RESULTS OF ^1 ANALYSES BY DIFFERENT INSTITUTES ON SPIKE SOLUTION В, MEDIUM LEVEL Calibration value 108.9 nCi isii/i- P = preliminary; F = final '- r e s u lt s f Method Total time used used for e - E -3 Lab. g No. 3 .5 -S *3 3 > д (nCi/1) (± <%) d(%) (nCi/1) (± %) d (%) P F PF 1 89.6 5 -17.7 100.2 5 - 8.0 2 2 4.0^ 19.0* 2 95.2 1 -12.6 1 5 '3 110.0 20 + 1.0 1 9-9* 4 96.4 5 -11.5 96.3 5 -115 2 2 2.7 2.7 5 103.0 5 - 5.4 103.0 5 - 5.4 3 3 1.3 6 111.0 3 + 1.9 3 0.2* 7 110.9 5 + 1.8 1 1.7 8 112.0 4 + 2.8 - 1 1.2 9 88.5 9 -18.7 2 0.6* 10 89.0 9 -18.2 2 0.6* 11 104.0 6 - 4.5 1 0.6 12 118.0 6 + 8.4 118.8 5 + 9.1 2 2 0.1 0.8 13 99.8 5 - 8.4 3 2.7 14 110.0 10 + 1.0 110.0 10 + 1.0 1 2 1.2 1.2 15 93.7 20 -13.9 108.6 20 - 0.3 2 3 0.8* 1.5* 16 98.8 5 - 9.2 1 0.9 17 107.7 0.5 - 1.1 108.8 0.5 - 0.1 1 2 1.3* 1.3* 18 106.0 1 - 2.7 106.0 1 - 2.6 2 2 0.6 0.9 19 104.4 1 - 4.1 104.4 1 - 4.1 1 1 0.5* 0.5* 20 107.8 2 - 1.0 107.8 2 - 1.0 4 4 0.7 0.7 21 116.2 20 + 6.6 4 0.8 22 101.6 3.5 - 6.7 101.1 1 - 7.2 1 1 0.2* 5.0* 23 110. 8 + 1.0 110.0 6 + 1.0 2 2 1.8* 3.5* 24 113.1 15 + 3.8 113.0 15 + 3.8 4 4 1.0 48.0* 25 112.0 10 + 2.8 112.0 10 + 2.8 1 1 2.5 2.5 26 98.0 10 -10.0 99.0 10 - 9.1 3 3 2.3* 3.9* 27 105.9 10 - 2.8 107.3 10 - 1.4 3 3 2.0 2.0 28 107.4 1 - 1.4 107.4 1 - 1.4 1 1 0.9 0.9 29 114.7 + 5.3 114.7 + 5.3 4 4 0.8 0.8 30 111.0 4 + 1.9 111.0 4 + 1.9 3 3 l.ia 1.1* 31 110.3 10 + 1.3 110.3 10 + 1.3 2 2 1.5 1.5 32 106.0 2 - 2.7 106.0 2 - 2.7 2 2 1.2 1.2 33 95.8 4 -12.0 95.8 4 -12.0 1 1 2.9* 2.9* 34 112.0 10+ 2.8 1 3.5* 35 98.1 10 - 9.8 3 9.0* 36 IAEA-SM-180/2 367 TABLE III. RESULTS OF ^ C s ANALYSES BY DIFFERENT INSTITUTES ON SPIKE SOLUTION A, LOW LEVEL Calibration value 9.11 nCi/l: P = preliminary; F = final 1 F Method Total time used S ^ ¿ > Lab. S' 'S *5 7ÏÜ No. § >S 3-S -3 g -ë 0 (nCi/1) (± %) d(%) (nCi/l) (± %) d (%) P F p F 1 8.90 5 - 2.3 9.10 5 - 0.1 2 2 4* 19.0* 2 9.40 5 + 3.2 1 7.5* 3 8.50 20 - 6.7 1 8. 0* 4 8.83 5 - 3.1 8.84 5 - 3.0 2 2 2.8 2.8 5 8.00 5 - 12.2 8.20 5 - 10.0 3 3 4.0 6 9.10 5 - 0.1 3 1.4* 7 9.18 5 + 0.8 1 2.5 8 8.00 5 - 12.2 1 1.2 9 8.90 10 - 2.3 2 0. 6* 10 8.80 10 - 3.4 2 0. 6^ 11 9.10 10 - 0.1 1 0.6 12 10.30 8 ^ 2.1 10.12 6 +12.0 2 2 0.2 1.3 13 9.42 5 + 3.4 3 10.7 14 10.00 10 +10.0 10.00 10 +10.0 1 2 1.2 1.2 15 8.96 20 - 1.6 8.89 20 - 2.4 2 2 1. 0* 2.5* 16 8.90 6 - 2.3 1 0.9 17 9.6 2 + 5.4 9.40 2 + 3.2 1 2 1. 3a 1.3* 18 10.10 2 +11.0 9.80 1 + 7.6 2 2 0.6 2.1 19 8.99 1 - 1.3 8.99 1 - 1.3 1 1 1.0* 1.0* 20 9.64 3 + 5.8 9.64 3 + 5.8 4 4 1.0 1.0 21 9.55 20 + 4.8 4 0.8 22 9.13 6 + 0.2 9.09 4 - 0.2 1 1 0.5* 1.6* 23 8.4 12 - 7.8 8.50 6 - 6.7 2 2 1.8* 3.5* 24 8.32 15 - 8.7 11.30 15 +24.2 4 4 1.5 120.0* 25 9.80 10 + 7.6 9.80 10 + 7.6 1 1 2.5 2.5 26 8.20 15 -10.0 9.00 10 - 1.2 3 3 3.6* 14.5* 27 10.43 10 +14.5 10.30 10 +13.0 3 3 2.0 2.0 28 9.51 1 + 4.4 9.51 1 + 4.4 1 1 1.3 1.3 29 9.78 + 7.4 9.78 + 7.4 4 4 1.3 1.3 30 9.13 4 + 0.2 9.13 4 + 0.2 3 3 2.1* 2.1* 31 9.10 10 - 0.1 9.10 10 - 0.1 2 2 1.5 1.5 32 10.20 6 +12.0 10.20 6 +12.0 2 2 1.5 1.5 33 9.20 3 + 1.0 9.20 3 + 1.0 1 1 14.6* 14.6* 34 10.00 15 +10.0 1 3.5* 35 8.38 10 - 8.0 3 9.0* 36 4.57b 5 -49.8 4. 57b 5 -49.8 1 1 1.5 1.5 * Total time used for analyses of three to five nuclides simultaneously, b Result not used in calculation of mean. 368 SUSCHNY et al. TABLE IV. RESULTS OF ^ C s ANALYSES BY DIFFERENT INSTITUTES ON SPIKE SOLUTION В, MEDIUM LEVEL Calibration value 98. 0 nCi/l: P - preliminary; F = final F - r e s u l t s F - r e s u l t s Method Total time used used for E -3 E ^ S > (Z? Lab . о .3 § 2 ё N o . g 3 > -3 -Ë g s (nCi/1) (± %) d (%) (nCi/l) ( i % ) d ( % ) PF p F 1 9 3 .5 5 - 4 . 6 9 9 .5 5 - 1 .5 2 2 4 . 0 * 1 9 .0 * 2 9 7 .2 1 - 0.8 1 7 .5 * 3 100.0 20 + 2.0 1 9 .9 * 4 9 0 .5 5 - 7 .6 9 0 .4 5 - 7 .7 2 2 2.8 2.8 5 9 4 .0 5 - 4 . 1 9 0 .0 5 - 8.2 3 3 1 .3 6 9 8 .0 5 0.0 3 0. 2* 7 9 8 .4 5 + 0 .4 1 1 .7 8 9 2 .0 4 - 6.1 1 1.2 9 9 1 .5 10 - 6.6 2 0. 6* Ю 9 1 .0 10 - 7 .1 2 0. 6* 11 9 6 .0 6 - 2.0 1 0.6 12 1 0 8 .0 8 +10.2 1 0 7 .9 6 +10.1 2 2 0.1 0.8 13 8 5 .6 5 - 12.6 3 2 .7 14 111.0 10 + 13.3 111.0 10 + 13.3 1 2 1.2 1.2 15 9 6 .5 20 - 1 .5 1 0 9 .2 20 +11.4 2 3 0. 8* 1 .5 * 16 9 4 .7 3 - 4 . 0 1 0 .9 17 9 7 .0 0 .5 - 1.0 9 7 .8 0 .5 - 0.2 1 2 1 .3 * 1 .3 * 18 1 0 2 .9 0.6 + 5 .0 1 0 3 .5 0 .5 + 5 .6 2 2 0.6 0 .9 19 9 2 .3 0 .7 - 5 .8 9 2 .3 0 .7 - 5 .8 1 1 0 .5 * 0. 5 * 20 1 0 4 .5 3 + 3 .6 1 0 4 .5 3 + 3 .6 4 4 1.0 1.0 21 1 0 3 .0 20 + 5 .1 4 0.8 22 9 5 .5 4 - 2 . 5 88.2 0.8 - 10.0 1 1 0. 2* 5 .0 * 23 91 8 - 7 .1 9 5 .0 6 - 3 .1 2 2 1. 8* 3 . 5 * 24 9 1 .8 15 - 6 .3 9 1 .8 15 - 6 .3 4 4 1 .5 4 8 . 0 * 25 9 8 .9 10 + 0 .9 9 8 .9 10 + 0 .9 1 1 2 . 5 2 .5 26 9 2 .0 10 - 6.1 9 4 .0 10 - 4 . 1 3 3 2 . 3 * 3 . 9 * 27 9 7 .1 10 - 0 .9 9 6 .9 10 - 1.1 3 3 2.0 2.0 28 101.2 1 + 3 .3 101.2 1 + 3 .3 1 1 0 .9 0 .9 2 9 1 0 6 .0 + 8.2 1 0 6 .0 + 8.2 4 4 1.1 1.1 3 0 9 8 .3 2 + 0 .3 9 8 .3 2 + 0 .3 3 3 1. 1* 1. 1* 31 9 6 .9 10 - 1.1 9 6 .9 10 - 1.1 2 2 1 .5 1 .5 32 1 1 4 .0 3 +16.3 114.0 3 +16.3 2 2 1.2 1.2 33 91.6 3 - 6.5 91.6 3 - 6 .5 1 1 2 . 9 * 2 . 9 * 34 100.0 10 + 2.0 1 3 .5 * 35 9 0 .5 10 - 7 .7 3 9 .0 * 36 4 9 .6 b 5 -49.4 49. бЬ 5 - 4 9 . 4 1 1 1 1 .5 1 .5 IAEA-SM-180/2 369 TABLE V. RESULTS OF ^ ° B a ANALYSES BY DIFFERENT INSTITUTES ON SPIKE SOLUTION A, LOW LEVEL Calibration value 1.98 nCi/1: P = preliminary; F = final P - r e s u l t s F - r e s u l t s M ethod T o t a l tim e used used for S -=j 0 g a n a lysis Lab . (hours) g 'S N o. g .3 ¿ ¡3 s 0 (n C i/1 ) (± %) d ( % ) (n C i/1 ) %) d ( % ) PF PF 1 2 .8 0 10 + 4 1 .5 2 1 9 .0 * 2 1 .8 0 10 - 9 .1 1 7 . 5a 3 2.00 20 + 1.0 1 8 .0 * 4 1 .9 3 5 - 2 .5 1 .9 2 5 - 3 . 0 2 2 2.8 2.8 5 2.20 10 +11.1 2.20 8 +11.1 3 3 4 . 0 6 2.10 10 + 6.0 3 1 .4 * 7 1 .9 8 5 0.0 1 2 .5 8 1 .6 0 7 - 1 9 . 2 1 0 .7 9 1 .8 0 10 - 9 .1 2 0 . 6* 10 1 .9 0 10 - 4 . 0 2 0. 6* 11 2.00 7 + 1.0 1 0.6 12 2.00 17 + 1.0 2.11 10 + 6.6 2 2 0.2 1 .3 13 2 .7 8 5 + 4 0 .5 3 1 0 .7 14 2 .7 0 15 + 3 6 .4 2 .7 0 15 + 3 6 .4 1 2 1.2 1.2 15 1 .2 6 25 - 3 6 . 4 1.66 20 - 1 6 . 2 2 2 1.0 16 2.10 35 + 6.0 1 0 .9 17 1 .3 2 - 3 4 . 4 1 .5 2 - 2 4 .2 1 2 1. 3a 1. 3a 17a 1 .9 - 4 . 0 5 18 1 .9 0 10 - 4 . 0 1 .8 9 3 - 4 . 5 2 2 0.6 2.1 19 2.01 4 + 1 .5 2.01 4 + 1 .5 1 1 1. 0* 1. 0* 20 2 .3 8 4 +10.1 2 .3 8 8 +10.1 4 4 2 .4 2 .4 21 22 1.66 20 - 1 6 .2 1 .5 6 20 - 21.2 1 1 0 .5 * 1. 6^ 23 1 .9 20 - 4 . 0 2.00 10 + 1.0 2 2 1. 8* 3 . 5a 24 1.69 15 -14.6 2 .0 6 15 + 4 . 0 4 4 3 . 0 120. 0* 25 2.10 10 + 6.0 2.10 10 + 6.0 1 1 2 . 5 2 .5 26 < 3 .5 0 b < + 7 6 .7 2 .7 0 25 + 3 6 .4 3 3 3 . 6 * 1 4 .5 ^ 2 6 a 2 .9 0 + 4 6 .5 2 .2 8 - 3 5 . 3 27 1 .9 0 25 - 4 . 0 2.10 25 + 6.0 3 3 2.0 2.0 2 8 2.20 1 +11.1 2.20 1 +11.1 1 1 1 .3 1 .3 2 9 1 .5 3 - 22.8 1 .5 3 - 22.8 5 5 1 .7 1 .7 3 0 1 .8 4 10 - 7 .0 1 .8 4 10 - 7 .0 3 3 2 . 1* 2 . 1* 31 2.10 10 + 6.0 2.10 10 + 6.0 2 2 1 .5 1 .5 32 2.02 13 + 2.0 2.00 13 + 1.0 2 2 1 .5 1 .5 33 1 .7 3 18 - 12.6 1 .7 3 18 - 12.6 1 1 1 4 .6 * 1 4 .6 * 34 1 .7 0 25 - 1 4 . 1 1 3 . 5 * 35 1 .7 6 10 - 11.1 3 9 .0 * 36 6.32b 5 +219.0 5.2бЬ 5 +165.0 4 4 11.0 11.0 37 0 SU SCHN Y et al. TABLE VI. RESULTS OF ^ ° B a ANALYSES BY DIFFERENT INSTITUTES ON SPIKE SOLUTION B, MEDIUM LEVEL Calibration value 21.8 nCi/l: P = preliminary; F = final f M et lod T o ta l ust d used for 0 4 0 g L a b . (ho j b N o . 3 ( n C i/ l ) (± %) d ( % ) ( n C i/ l ) (± %) d(%>) P F P F 1 1 5 .7 5 - 2 8 . 0 21.1 10 - 3 .2 2 2 4 * 1 9 .0 * 2 2 1 .7 5 - 0 .5 1 7 . 5 * 3 20.0 20 - 8 .3 1 8 . 0* 4 2 3 .2 5 + 6 .4 2 2 .4 5 + 2 .7 2 2 2.8 2.8 5 22.0 10 + 0 . 9 21.0 8 - 3 .7 3 3 1 .3 6 2 4 .0 13 +10.2 3 0 . 2* 7 2 1 .7 5 - 0 .5 1 1 .7 8 1 5 .3 5 - 2 9 . 8 1 0 .7 9 1 9 .7 10 - 9 .5 2 0 . 6* 10 1 9 .0 10 - 12.8 2 0 . 6* 11 20.0 4 - 8 .3 1 0.6 12 2 4 .0 8 +10.1 2 3 .6 6 + 8 .3 2 2 0.1 0.8 13 2 3 .1 5 + 5 .9 3 2 .7 14 2 7 .0 10 + 2 3 .8 2 7 .0 10 + 2 3 .8 1 2 1.2 1.2 15 1 7 .9 25 - 1 7 . 9 1 7 .0 25 - 22.0 2 3 0. 8* 1 .5 16 1 8 .2 10 - 1 6 . 5 1 0 .9 17 20.6 0 .5 - 5 .5 1 9 .5 0 .5 - 1 0 . 5 1 2 1 .3 * 1 .3 * 17a 21.1 - 3 .2 5 18 1 9 .5 2 - 10.6 1 9 .4 2 - 11.0 2 2 0.6 0 .9 19 21.0 2 - 3 .7 21.0 2 - 3 .7 1 1 0. 5 * 0 .5 * 20 2 2 .7 3 + 4 . 1 2 2 .7 8 + 4 . 1 4 4 2 .4 2 .4 21 1 9 .8 14 - 9 .2 4 0 .7 22 1 9 .2 10 - 1 1 . 9 1 9 .0 2 - 1 2 . 9 1 1 0. 2* 5 . 0 * 2 3 20 12 - 8 .3 21.0 7 - 3 .7 2 2 1. 8* 3 . 5 * 24 12.8 15 - 4 1 . 3 2 2 .5 15 + 3 .2 4 4 3 . 0 * 4 8 . 0 * 2 5 2 1 .7 10 - 0 .5 2 1 .7 10 - 0 .5 1 1 2 . 5 2 .5 2 6 2 9 .0 25 +33.1 26.0 20 + 19.3 3 3 2 . 3 * 3 . 9 * 2 6 a 2 5 .9 + 1 8 .8 2 2 . 9 + 5 .0 27 22.1 17 + 1 .4 2 2 .5 17 + 3 . 2 3 3 2.0 2.0 2 8 21.0 1 - 3 .7 21.0 1 - 3 .7 1 1 0 .9 0 .9 29 14.8 -32.2 14.8 -32.2 5 5 1 .7 1 .7 3 0 2 0 .4 12 - 6 .4 2 0 .4 12 - 6 .4 3 3 1. 1* 1. 1* 3 1 2 2 .9 10 + 5 .0 2 2 .9 10 + 5 .0 2 2 1 .5 1 .5 32 2 0 .5 6 - 5 .7 2 0 .5 6 - 5 .7 2 2 1.2 1.2 33 1 7 .9 19 -17.9 17.9 19 -17.9 1 1 2 . 9 * 2 . 9 * 34 22.0 10 + 0 .9 1 3 . 5 * 3 5 1 9 .5 10 - 10.6 3 9 .0 * 36 3 5 .5 ^ 5 3 0 .3 ^ 5 + 3 9 .0 4 4 11.0 11.0 IAEA-SM-180/2 371 TABLE VII. RESULTS OF ^° Sr ANALYSES BY DIFFERENT INSTITUTES ON SPIKE SOLUTION A, LOW LEVEL Calibration value 1.11 nCi/l: P = preliminary; F = final F - r e su lts Method Total time used E -§ о 13 L a b . (hours) N o. ,2 .B о о <3 -S 3 <3 -S 3 (nCi/l) (* %) d (%) (nCi/1) (± % ) d{%) PF p F 1 2 1 .0 5 8 - 5 .4 2 7 .5 * 3 1 .0 0 100 - 1 0 . 0 2 8 .0 4 0.26 6 - 2 9 . 7 1 .0 0 50 - 1 0 . 0 3 9 .1 7 1 .0 4 3 0 - 6 .3 1 .0 4 20 - 6 .3 3 . 6 3 . 6 3 . 0 4 .4 8 1.80 20 +62.1 2 3 .2 9 10 11 12 13 14 1 .0 8 10 - 2 .7 2 .6 15 16 17 1 .1 4 2 . 5 + 2 .7 1 .1 0 2 . 5 - 0 .9 1 .6 1 .6 9. 3b 9 . 3С 18 1 .0 0 33 - 1 0 . 0 1 6 .1 19 1 .6 1 17 + 4 5 .0 1 .6 2 . 8 * 2 0 1 .1 0 15 - 0 .9 1 .6 2 1 1 .0 5 14 - 5 .4 22 23 1 .1 0 25 - 0 .9 4 . 5 4 . 8 * 2 4 0.08*1 - 9 1 . 5 0 .0 7 * 15 - 9 2 . 0 1 .6 1 .6 5 .5 1 2 0 .0 25 0 .9 9 91 -10.8 1.19 10 + 7 .2 1 .6 1 .6 9 .6 1 2 .5 26 1 .1 0 - 0 .9 1 3 .7 27 1 .1 2 + 0 .9 1.14 3.9 + 2.7 1 1 7 .3 * 8 .5 28 29 30 0 .9 9 - 1 0 . 8 0 .9 9 -10.8 3.2.6 3.2.6 3.4 3 .4 31 1 .0 0 20 - 1 0 . 0 2 . 6 32 1.30 11 +17.1 1 .3 0 11 + 1 7 .1 2 2 1 1 .8 1 1 .8 33 0 .9 2 4 - 1 7 . 1 0 .9 2 4 - 1 7 . 1 3.6 3.6 5.9 5.9 34 2.00<1 25 + 8 0 .2 1 7 .0 35 1.06 10 - 4.5 3 . 2 . 6 36 0.89 5 -19.8 1 0.89 - 1 9 . 8 4 . 6 4 . 6 1 2 .5 1 2 .5 * Estimated overall accuracy given by the laboratory. * Total time used for analyses of **Sr and *°Sr simultaneously. b Not including 14 days' waiting time for the growing in of *Y . 372 SUSCHNY et al. T A B L E VIII. R E S U L T S O F ^° Sr ANALYSES BY DIFFERENT INSTITUTES ON SPIKE SOLUTION В, MEDIUM LEVEL Calibration value 1. 63 nCi/l: P = preliminary; F = final ' - r e s u l t s F - r e s u l t s M ethod T o t a l tim e = used used for 2 5 L a b . (ho N o . .а я 3 g 3 (n C i/1 ) ( * %) d ( % ) (n C i/1 ) ( i %) d ( % ) P F P F 1 2 0 .9 9 8 - 3 9 . 4 2 7 . 5 * 3 4 1.69 5 + 3.7 4.3.6 3 .6 ^ 5 1 .2 0 5 - 2 6 . 4 1 .3 4 5 - 1 7 . 8 1 .5 1 .6 3 . 6 6 2 .0 0 100 + 1 6 .6 3 9 .1 7 1 .5 0 3 0 - 8 .0 1 .3 5 20 - 1 7 . 2 3 . 6 3 .6 4 . 4 8 2.20 30 +35.0 2 3 .7 9 10 11 12 13 14 1 .5 0 10 - 8 .0 2 .6 15 16 17 2 .2 2 2 + 3 6 .3 1 .8 8 2 + 15.3 1 .6 1 .6 9 .3 ^ 9. 3С 18 2 .0 0 50 + 22.7 1 6 .1 19 2 .1 4 13 +31.3 1.6 2.8* 20 1 .2 3 18 - 2 4 . 6 1 .6 2 . 9 21 22 23 2.30 30 +41.1 4.5 4.2* 24 0 .9 6 - 4 1 . 1 0 .9 6 15 -41.1 1.6 1.6 5.5 4 8 . 0 2 5 2 .5 0 + 5 3 .4 1 .4 4 10 -11.7 1.6 1.6 9.6 1 4 .3 26 6 .0 0 * + 2 6 8 .0 1 3 .7 27 0 .9 4 - 4 3 . 0 2.08 4.8 +27.6 1 1 7 .3 * 8 . 5a 28 2 9 30 1 .4 6 - 1 0 . 4 1 .4 6 -10.4 3.2.6 3.2.6 3.4 3 .4 3 1 32 1 .7 0 11 + 4 .3 1 .7 0 11 + 4 .3 2 2 1 1 .8 1 1 .8 33 2 .4 0 4 + 4 7 .4 1 .1 0 10 -32.6 3.6 3.6 5.9 5 .9 3 4 3 5 1 .5 9 10 - 2 .4 3 . 2 . 6 36 2 .8 4 5 +74.4 3.14*3 5 +92.2 4.6 4.6 12.5 12.5 IAEA-SM-180/2 373 TABLE IX. RESULTS OF ^Sr ANALYSES BY DIFFERENT INSTITUTES ON SPIKE SOLUTION A, LOW LEVEL Mean value 2. 70 nCi/l for P-results and 3.15 nCi/l for F-results: P = preliminary; F = final P -r e su lts F -re su its Method Total time E ^ S g + * ^ o t s ) ' Lab. ë *§ No. g g 3 .3 -g g g (nCi/l) (± %) d(4b) (nCi/l) (± %) d (%) P FPF 1 2 2.90 8 - 7.9 2 7.5* 3 4 5 2.90 10 - 7.4 2.60 -17.5 1.5 1.6 4.3 6 7 3.20 + 1.6 8 9 10 11 12 13 14 2.60 10 -17.5 2.6 15 16 17 3.05 - 3.2 18 3.00 10 - 4.8 6.1 19 6.60b 3 +109.5 1.6 2.8" 20 21 1.3 14 -51.8 22 23 2.90 25 - 7.9 4.5 4.9" 24 25 3.28 6 +21.5 3.28 6 + 4.1 1.6 1.6 9.6 18.0 26 2.40 -11.2 1 3.7 27 3.64 +34.8 3.90 1.1 +23.8 1 1 7.3* 8.5" 28 29 30 31 3.40 15 + 7.9 2.6 32 33 34 35 36 37 3.79 20 +20.3 2 15 . * Estimated overall accuracy given by the laboratory. * Total time used for analysis of *^Sr and *S r simultaneously, b Result not used in calculation of mean. 3 7 4 SUSCHNY et al. TABLE X. RESULTS OF ^Sr ANALYSES BY DIFFERENT INSTITUTES ON SPIKE SOLUTION В, MEDIUM LEVEL Mean value 33. 4 nCi/l for P-results and 35. 0 nCi/l for F-results: P = preliminary; F = final - r e su lts F - r e su lts Method Total time § ^ Lab. No. Er .3 ¿2 о ÎÜ о Й -o S 2 Й -о о (nCi/1) (± 7°) d(%) (nCi/1) (± %) d(%.) P FPF 1 2 29.0 8 -17.1 2 7.5" 3 4 5 32.0 10 - 4.2 29.0 5 -17.1 1.5 1.6 3.6 6 7 34.4 - 1.7 8 9 10 11 12 13 14 24.0 10 -31.4 2.6 15 16 17 33.7 - 3.7 18 30.0 3 -14.3 1 6.1 19 58.0 1 +68.5 1.6 2.8^ 20 21 22 23 33.0 20 - 5.7 4.5 4.2* 24 25 32.0 3.3 - 4.2 32.0 3.3 - 8.6 1.6 1.6 9.6 18.0 26 30.0 -10.4 1 3.7 27 39.4 +18.0 40.4 0.3 +15.4 1 1 7.3* 8.5* 28 - 29 30 31 32 33 34 35 36 37 41.8 20 +19.4 2 1 ^ IAEA-SM-180/2 3 7 5 TABLE XI. RESULTS OF 89+90 g^. A N A L Y S E S B Y D IF F E R E N T INSTITUTES ON SPIKE SOLUTION A, LOW LEVEL Mean value 6. 37 nCi/1 for P-results and 5. 31 nCi/1 for F-results: P = preliminary; F = final f -results F-results Method Total time E -3 used uæd for 2 > Lab. o' S' § Й No. Ü .2 ¿ g Ж -ë S 2 Й "O 3 (nCi/1) (± %) d(%) (nCi/l) (± %) d(Tb) P F PF 1 5.19 20 + 2.3 1 33.0 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 4.50 -29.4 2 5.8 17 18 7.00 3 + 9.9 1 4.8 19 9.26 3 +45.4 1 2.8 20 8.52 6 +33.8 1 2.9 21 22 3.09 20 -51.5 3.62 15 -30.3 1 1 5.4 5.4 23 4.8 25 -24.6 4 4.0 24 25 26 27 28 7.40 1 +16.2 7.40 1 +42.7 1 1 24.4 24.4 29 5.05 1 -20.7 5.05 - 2.7 1 1 1.9 1.9 30 31 7.70 10 +20.8 2 3.5 32 33 34 35 36 376 SUSCHNY et al. TABLE XII. RESULTS OF S9+90gp ANALYSES BY DIFFERENT INSTITUTES ON SPIKE SOLUTION В, MEDIUM LEVEL Mean value 51. 5 nCi/l for P-results and 48. 2 nCi/l for F-results: P = preliminary; F = final P-re suit s F Method Total time E ^ used for ^ > s 5 Lab. о Й No. g Э Й -o S 2 Й -o S (nCi/l) (± %) d(%) (nCi/l) (i %) d(%) P FP F 1 45.6 20 - 5.4 1 33.0 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 38.0 -26.2 2 5.8 17 18 55.8 10 + 8.3 1 4.8 19 87.0 1 +69.0 1 2.8 20 80.4 6 +56.1 1 2.9 21 22 27.5 20 -46.6 30.4 15 -37.0 1 1 5.4 5.4 23 44 20 -14.5 4 3.9 24 25 26 27 28 43.8 1 -15.0 43.8 1 - 9.1 1 1 24.4 24.4 29 44.1 -14.4 44.1 - 8.5 1 1 1.9 1.9 30 31 77.3 10 +50.1 77.3 10 +60.5 2 2 3.0 3.0 32 33 34 17.0 25 -67.0 1 7.0 35 36 IAEA-SM-180/2 377 TABLE XIII. AVERAGE VALUES AND RANGES OF THE RESULTS OBTAINED ON SPIKE SOLUTION A; LOW LEVEL 13^ '*Cs "°Ba '°Sr "Sr P j F P j F P j F P j F P 1 F P j F 10.14 9.11 1.98 1.11 i 3% (nCi/1)* i 3% i 3% i 3% 27 30 28 31 27 33 13 22 5 12 9 4 27 30 27 30 25 32 10 20 5 11 9 4 lowered - - 1 1 2 1 3 2 - 1 - - deluded 11.20 12.60 10.43 11.30 6.32 5.26 2.00 1.80 3.64 6.60 9.26 7.40 (nCi/1) 8.60 8.20 4.57 4.57 1.26 1.50 0.08 0.07 1.30 2.60 3.09 3.62 (nCi/1) 2.60 4.40 5.86 6.73 5.06 3.76 1.92 1.73 2.34 4.00 6.17 3.78 (nJ/l) 2.60 4.40 2.43 3.30 1.64 1.30 0.41 0.91 2.34 1.30 6.17 3.78 (nCi/1) 9.99 9.99 9.30 9.28 1.96 2.03 1.05 1.12 2.70 3.15 6.37 5.31 (nCi/1) (± nCi/1) 0.67 0.86 0.65 0.70 0.37 0.35 0.12 0.22 0.91 0.42 2.08 1.56 Í* 6.7 8.6 7.0 7.5 18.7 17.1 11.3 19.8 33.6 13.5 32.7 29.4 (nCi/1) -0.15 -0.15 +0.19 +0.17 -0.02 +0.05 -0.06 +0.01 no calibration value (%) -1.5 -1.5 +2.1 +1.9 -1.0 +2.5 -5.4 +0.9 3 7 8 SUSCHNY et al. TABLE XIV. AVERAGE VALUES AND RANGES OF THE RESULTS OBTAINED ON SPIKE SOLUTION В; MEDIUM LEVEL "'t **Cs "°Ba *St "Sr " + *Sr P j F P j F P j F P j F P j F P j F 108.9 98.0 21.8 1.63 (nCi/l)* * 3% ± 3% ± 3% i 3% 27 30 28 31 29 33 11 20 4 11 10 5 27 30 27 30 28 32 10 19 4 11 10 5 -- 1 1 1 1 1 1 ---- 118.0 118.8 114.0 114.0 35.5 30.3 6.00 3.14 39.4 58.0 87.0 77.3 (nCi/l) 89.6 88.5 49.6 49.6 12.8 14.8 0.94 0.96 30.0 24.0 17.0 30.4 (nCi/l) 28.4 30.3 64.4 64.4 22.7 15.5 5.06 2.18 9.4 34.0 70.0 46.9 (nCi/l) 28.4 30.3 23.0 28.4 16.2 12.2 1.90 1.34 9.4 34.0 70.0 46.9 (nCi/1) 106.2 105.2 98.4 97.5 20.8 20.9 1.77 1.63 33.4 35.0 51.5 48.2 (nCi/1) (i nCi/l) 7.0 7.4 6.2 7.2 3.4 2.6 0.68 0.42 4.1 9.2 23.3 17.3 (i Tb) 6.6 7.1 6.3 7.4 16.2 12.5 38.1 25.6 12.4 26.1 45.3 36.0 (oCi/1) -2.7 -3.7 +0.4 -0.5 -1.0 -0.9 40.14 iO.OO (Tb) -2.5 -3.4 +0.4 -0.5 -4.6 -4.1 +8.6 iO.O IAEA-SM-180/2 3 7 9 TABLE XV. DISTRIBUTION OF RESULTS IN CONTROL CHARTS NUMBER OF LABORATORIES IN EACH CATEGORY Numbei of Results Nuclide Between Outside Between Total and control and of warning activity warning control limits (nCi/1) limits limits ^ Ю. 14 P 25 2 - 27 ? F 27 2 1 30 ^ 108.9 P 25 2 - 27 R F 28 2 - 30 g 9.11 P 26 1 1 28 É F 29 - 2 31 g 98.0 P 25 2 1 28 U F 27 3 1 31 3 1.98 P 21 4 2 27 F 28 4 1 33 S ¡3 ^ * 8 P 24 3 2 29 " F 30 2 1 33 TABLE XVI. DISTRIBUTION OF STRONTIUM RESULTS WITHIN THE ERROR LIMITS OF 30% Nuclide Number of results Outside Total and between 30% of number of activity limits limits results (nC i/l) X ± 30% 3 2 5 11 1 12 4 4 9 2 11 Strontium- Strontium- 89 ? i . i l p 10 3 13 § F 18 4 22 § 1.63 P 4 7 11 й F 14 6 20 á 6.37 P 6 3 9 ^ 5.31 F 2 2 4 g 51.5 P 5 5 10 g 4 8 .2 F 3 2 5 (/) nC/// !тм п о/ Л-геди/fs 705.2nC/// . . со//Ьго?йп 99 707 JO SUSCHNYal. et [ mean mean oí [ o.-resu^s /06.2nC/// FIG. 2. Distribution of results of ^*1 analyses at 108.9 nCi/1. 108.9 at analyses ^*1 of results of 2. Distribution FIG. FIG. 1. Distribution of results of *^I analyses at 10.14 nCi/1. 3 8 0 у. % % % /repuency F I G . 4 .Di s t r i b u t i o n o f r e s u l t s o f * ^ C s a n a l y s e s a t 9 8 . 0 n C i / 1 . I 3 Dsrbto frslso ^ nlss t .1n i/l. nC 9.11 at analyses s *^C of results of Distribution .3. FIG 7 / / / C n 77 0 7 ^ IABA-SM-180/2 JO Д? o'rfo yo/mp______co/'brof/on 7 Í / / C o 2 7 77 0 7 /o rs /fs r?su //no/ 381 SUSCHNY et al. JO Ь20 M 2.5 ¿9 nC'/i 7J 7.7 2.7 2.5 2.9 nC/// co//ôrof/on mcon of f.-r*su /fs 2.C8nC/// i /пм я o/ A -fPSV ^s 7.96лС/// co//ùro//on ______ FIG.5. Distribution of results of **°Ba analyses at 1.98 nCi/1. 40 jcreZ/m/nory resu/fs JO 20 J u æ 29 n o // 7J 25 2 9 n C /// ! jco/zbrof/on ! тм я o/ p.-ffstíífs 20FnC/// t meon of f. - r e sv/fs 2ÛFnC/// FIG.6. Distribution of results of *^*Ba analyses at 21.8nCi/l. IAEA-SM-180/2 //no/ resu/fs 6 0 60 40 40 20 20 Л. о 'д 7.'¿) 2.2 nC/// О б 7.0 22 nC/// mcon of p**r?su/fs M25nC//í coßbroMon ^ o / u ^ ______ FIG. 7. Distribution of results of Sr analyses at 1.11 nCi/l. 20 - //no/ resu /fs 7 0 f 0.9 7.J ^ ^ ¿5 ¿9nC/Y/ mcon of p -r^so^fs ?.77nC/// т м п of f.-rM u/^s ^.63nC//^ co//brof/on уо/ие_____ co//óro?/on ^o/u^ FIG.8. Distribution of results of ^°Sr analyses at 1.63 nCi/l. 384 SUSCHNY et al. 72.J Jr+J3¿/C¿ 77.6 ' ° *з<к9ь 70.74 О же e * so 6 .7 0 о \ G C6.M í4 SN o . 70 20 JO ^ = pre//nvnor/ reso/fs 0=f/f!3ÎffSU/fS FIG.9. Control chart for individual determinations^!: 10.14 nCi/1; S =7%). X+JS uc¿ 732 724 ОфО eo ФОeo 70C.9 ^ r - y - ФЭ e o — e ------Ф $ o o X-JSÍCÍ No. 70 20 JO FIG.10. Control chart for individual determinations (^^1; 108.9 nCi/1; S = 7%). I.1. oto hr o idvda dtriain C 91 ni1 S=7^b). nCi/1; 9.11 Cs determinations individual for chart Control 11. FIG. F I G . 1 2Co . n t r o l c h a r t f o r i n d i v i d u a l d e t e r m i n a t i o n s ( * ^ C s 9 8 . 0 n C i /S 1 ; = 6 ^ ) . n e ;/; /76 96 0 / 7 66 ao 40 .4 0 7 M M 720 77 .7 9 zao vo. 70 . o /v е ^ ¿ = Лпо/ = гми//б О prc//m/nor/ fpsu/fs = $ ^e ^o. e i^ О ? //no/ r^su/fs ? О r pfZ/oy гми/Уз prf/Zm/nory r Ф (8P о о -"ее -1 0 20 ?0 ------Ф<Э О ф G IAEA-SM-180/2 О Ф Ф ! ------О Ф 2 *" *20 ФЭ 30 ! ------<80 О Ф Г _&ДЁ_ Jf-J5 +5 L/И^А7+25 Л-25 5 2 + X a фО ФО JO gO fJ ¿c¿ Jf-JS JO c¿ u О Ф 6 . 9 4 О Ф OO ¿tVÍ 4.57 ao 385 386 SUSCHNY et al. 296 X + J5 UCÍ $ о о we О 2.65 / + 2S ÍWÍ wo О д о WO о О ф ж е 0*0 ° Й а а wo Л96 ф о / 9 o ф 06& w w 8 0 о о ° о . ° о W * о о о я о X-2S W Í о ** с 09Н /-J 6ÍCÍ G =Лпо/ г^ди/fs ¿ДВ No. !0 20 J0 FIG. 13. Control chart for individual determinations (*^Ва 1.98nCi/l; S = 17%). J5.5 WC JO.J W 25 а е фО О wo WO о 0 а л W w ; 0 о G wo 0 wo W w o ^ ° ^ ^ о G З э W WO о / - 2 5 ¿W¿. ^ о WO F-JS ¿ 4 S Л/о. Ю 20 JO ? 72.0 FIG. 14. Control chart for individual determinations (*^Ba 21.8nCi/l; S =13%). IAEA-SM-180/2 387 LABORATOR!ES \ ------Number of days between the date when the samples to be measured were received and the date when the first results were sent off (X— - — X). ------Number of days between the date when the samples to be measured were received and the date when the final results were sent off (0------0). (for all the radionuclides with the exception of ^Sr and ^Sr) ANNEX I QUESTIONNAIRE Luxembourg, EURATOM In tercom parison Ргоягатте RAPID MEASUREMENTS Participating Institute Name A ddress Country Report of Results: Time schedule (minutes) R adio nCi/l^ Estimated Methods of . XXX ) D ata n u c lid e s ------\ Analysis Sample processing Counting time T o tal tim e accu racy xx ) " o f r e s u lt s Sample Chemical preparation se p a ra tio n I -131 Sr-89 S r-9 0 Ca-137 Ba-140 Reference date : Date and hour of sample reception (laboratory) Date and hour of data transmission telegr. Dr. J. SMEETS, 29 rue Aldringen, Luxembourg. Telex EUROPE UJX. 423 and 446 x) activity at reference date and hour xx) state confidence level xxx) give a short description of method on separate sheet. IAEA-SM-180/2 389 ANNEX II LIST OF PARTICIPATING INSTITUTES BELGIUM Centre d'Etudes de l'Energie Nucléaire, Mol (Mrs. J. Colard and G. Koch) Institut d'Hygiène et d'Epidémiologie, Bruxelles (Dr. A. Lafontaine) DENMARK Danish Atomic Energy Commission, Health Physics Department, Ris% (Dr. A. Aarkrog) GERMANY, FED. REP. Chem. Landesuntersuchungsanstalt, Stuttgart (Dr. Stoll) Badische Anilin-Soda-Fabrik AG, Isotopenlaboratorium, Ludwigshafen (Rhein) (Dr. H. Guenzler) Materialprüfungsamt der Landesgewerbeanstalt Bayern, Abt. Strahlenschutz, Nürnberg (Dr. Dechert) Institut für Biophysik der Universität des Saarlandes, "Boris Rajewsky-Institut", Homburg/Saar (Dr. R. Kunkel) Chem. Landesuntersuchungsanstalt, Münster (Dr. H. Baumann) Chem. - und Lebensm. Unters., Messtelle für Radioaktivität, Hamburg (Dr. K. Boek) Bundesforschungsanstalt für Lebensmittelfrischhaltung, Karlsruhe (Dr. R. Schelenz) Bundesforschungsanstalt für Milchwirtschaft, Kiel (Dr. E. Knoop) Isotopenlaboratorium, Hamburg-Suelldorf (Dr. W. Feldt) Kernforschungsanlage Jülich, Jülich (Dr. H. Jacobs) Gesellschaft für Strahlenforschung, München (Dr. B. Sansoni) FRANCE S.C.P.R .I., Ministère de la Santé, LeVésinet(Dr. P. Pellerin) Ministère de l'Agriculture, Laboratoire de Radiobiologie, Paris (Dr. J. Morre) Commissariat à l'Energie Atomique - D. P. S ., 92 Fontenay-aux-Roses (Mr. L. Jeanmaire) ITALY Commission des Communautés Européennes C.C.R./Euratom, Ispra (Dr. A. Malvicini) Istituto di Igiene dell'Università di Pavia, Pavia (Dr. E. Lanzóla) Ministerodeglilnterni, Centro Studie Esperienze, Roma (Dr. F. Mazzini) Comitato Nazionale per l'Energia Nucleare, Casaccia - Roma (Dr. A. Cigna) KOREA Atomic Energy Research Institute, Chongryangri, Seoul (Dr. Young KuYoon) MEXICO Instituto de Energía Nuclear, Mexico 18, D.F. (Mrs. R. M. de Nulman) NETHERLANDS Rijksinstituut te Leiden, Leiden (Dr. W.G. deRuig) RijksinstituutvoordeVolksgezondheid, Bilthoven(Dr. F.C.M . Mattem) NORWAY Institutt for Atomenergi, Kjeller (Dr. E. Steinnes) SOUTH AFRICA Atomic Energy Board, Isotopes and Radiation, Pretoria (Dr. J.K. Basson) SWITZERLAND 390 SUSCHNY et al. SWEDEN National Institute of Radiation Protection, Special Labs, for Environmental Research, Stockholm (Dr. Swedjemark) UNITED KINGDOM Central Radiochemical Laboratory, Central Electricity Generating Board, West Farm place, Cockfosters, Barnet, Herts (Mr. T.W . Evett) South of Scotland Electricity Board, Hunterston Nuclear Power Station, Ayshire, Scotland UNITED STATES Dept, of Nuclear Sciences, Eberline Instrument Corp., Santa Fe, New Mexico OF AMERICA (Dr. E.A . Sanchez) Radiological Sciences Lab., New York State Dept. Health, Albany N.Y. (Dr. J.M. Matuszek) Radiochemistry D iv ., U. S. Atomic Energy Commission, Health and Safety L ab ., New York N .Y . (Dr. G .A . Welford) ANNEX III LIST OF PARTICIPANTS AT CEC MEETINGS a BELGIUM Mrs. J. Baruh Institut d'Hygiène et d'Epidemiologie, (1-2) 14 rue Juliette Wytsman, B-1050 Bruxelles Dr. G. Cantillon Ministère de la Santé Publique, (2) 14 rue Juliette Wytsman, B-1050 Bruxelles Mr. J. Colard Centre d'Etude de l'Energie Nucléaire, (1-2) Mol Mrs. de Clercq Institut d*Hygiène et d'Epidémiologie, (2) 14 rue Juliette Wytsman, B-1050 Bruxelles Mr. G. Koch Centre d'Etude de l'Energie Nucléaire, (1-2) Mol FRANCE Mr. L. Farges (1) Mr. L. Jeanm aire (2) Dr. J. Morre Ministère de l'Agriculture, (1-2) Laboratoire de Radiobiologie, 39 rue de Dantzig, Paris 15e № . F. Patti Commissariat à l'Energie Atomique - D.P.S. (2) B.P. n ° 6, 92 Fontenay-aux-Roses Mrs. L. Remy S.C.P.R .I., Ministère de la Santé Publique, (1-2) B.P. n" 35, 78 Le Vésinet a (i) refers to the meeting of 28/29 October 1969; (2) refers to the meeting of 3/4 July 1972; (1-2) refers to both meetings. IAEA-SM-180/2 391 GERMANY, FED. REP. Dr. K. Boek Chem. und Lebensm. Unters. (2) Messtelle für Radioaktivität, D-2 Hamburg Dr. H. Jacobs Kernforschungsanlage Jülich GmbH, (2) Postfach 365, D-517 Jülich Dr. D. Merten Kernforschungsanlage Jülich GmbH, (2) Zentralabteilung Strahlenschutz, Postfach 365, D-517 Jülich Dr. B. Sansoni Gesellschaft für Strahlenforschung, (2) D-8042 Neuherberg - Munich Dr. R. Schelenz Bundesforschungsanstalt für Lebensmittelfrischhaltung, (2) Engesserstrasse 20, D-75 Karlsruhe I Dr.A . Wiechen Bundesforschungsanstalt für Milchforschung, (2) Institut für Physik, Hermann Weigmann Strasse 1/27, D-23 Kiel ITALY Dr. G. Bagliano Comitato Nazionale per l'Energia Nucleare, (1-2) C.P. 2400 Casaccia, Rome Dr. A. Cigna Comitato Nazionale per l'Energia Nucleare, (1) C P. 2400, Casaccia, Rome Mr. C. Faloci Comitato Nazionale per l'Energia Nucleare, (1) Viale Regina Margherita 125, 1-00100 Rome Dr. P. Gaglione Commission des Communautés Européennes, (2) C. C.R. /Euratom, 1-2 1 020 Ispra - Varese Dr. E. Lanzóla Istituto di Igiene dell'Università di Pavia, (2) 1-27100 Pavia Dr. F. Mazzini Ministerodeglilnterni, (1-2) Centro Studi ed Esperienze, Capanelle, Rome Dr. A. Nardi Comitato Nazionale per l'Energia Nucleare, (1) Viale Regina Margherita 125, 1-00100 Rome LUXEMBOURG Dr. p. Kay ser Direction de la Santé Publique, (1) Luxembourg NETHERLANDS № . F.C.M. Mattem Rijksinstituut voor de Volksgezondheit, (1-2) Sterrenbos 1, Utrecht Dr. deRuig Rijkszuivelstation, (1-2) Vreewijkstraat 12B, Leiden ORGANIZATIONS INTERNATIONAL ATOMIC ENERGY AGENCY Dr. D. Merten Division of Research and Laboratories, (1) IAEA, PO Box 590, A -1011 Vienna, Austria 392 SUSCHNY et al. Dr. O. Suschny Division of Research and Laboratories, (2) IAEA, PO Box 590, A -1011 Vienna, Austria COMMISSION OF THE EUROPEAN COMMUNITIES Dr. J. Smeets D. G. Affaires Sociales, (1-2) (Chairman) Division Protection Sanitaire, 29 rue Aldringen, Luxembourg Mr. R. Amavis D.G. Affaires Sociales, (2) Division Protection Sanitaire, 29 rue Aldringen, Luxembourg Mr. A. Bonini D. G. Affaires Sociales, (2) Division Protection Sanitaire, 29 rue Aldringen, Luxembourg Mr. E. van der Stricht D.G. Affaires Sociales, (1) Division Protection Sanitaire, 29 rue Aldringen, Luxembourg REFERENCES [1] BRYANT, Pamela M. , Healrh Phys. 17 1 (1969) 51; Health Physics is the official journal of the Health Physics Society, UK. [2] GRAF, U ., HENNING, H .J ., Mitteilungsbl. Mathem. Stat. 4 (1952) 1. [3] DOERFFEL, K ., Beurteilung von Analyseverfahren und-ergebnissen, Springer, Berlin, Gottingen, Heidel berg, (1962): also in: Z. Anal. Chemie 185 (1962) 1-98. DISCUSSION Rebeca M. de N U L M A N (Chairman): I think that the Agency is performing a very valuable service with its intercomparison programme, since it enables the results obtained by different laboratories to be compared and, for countries like M e x i c o w h e r e there is only one laboratoryperforming analyses of environmental samples, it provides the only possible means of evaluating the analytical methods employed. G. H. P A L M E R : Were all the radionuclides added to the milk at the same time or was only one nuclide added to one sample of milk? J. H E I N O N E N : All the five radionuclides w e r e added to the s a m e s a m p l e of milk simultaneously. It is true that the participitants w e r e provided with two spike solutions in two separate ampoules for making two different samples for analysis. However, both of these solutions/samples contained all the five radionuclides and differed from each other only in activity, thus simulating two different levels of contamination. INTERPRETATION OF RESULTS Chairman MB. BILES (United States of America) IAEA-SM-180/6 SOME CONSIDERATIONS OF THE EFFECTS OF THE ACCIDENTAL RELEASE OF FISSION PRODUCTS FROM REACTORS J.R. BEATTIE Safety and Reliability Directorate, United Kingdom Atomic Energy Authority, Culcheth, Nr. Warrington, Lancs., United Kingdom Abstract SOME CONSIDERATIONS OF THE EFFECTS OF THE ACCIDENTAL RELEASE OF FISSION PRODUCTS FROM REACTORS. Nuclear power reactors and their radioactivity confinement systems are designed, constructed and operated so as to ensure, by the application of human ingenuity, knowledge and experience, that no uncon trolled release of radioactivity will occur. In normal operational conditions, radioactivity is discharged in gaseous and liquid effluents under controls which ensure that designated dose limits are not exceeded and that dose-rates are reduced as far below these limits as is reasonably practicable, bearing in mind costs and the importance of nuclear power. When all such requirements are fulfilled, there remains a finite if remote possibility of an accidental release of radioactivity to the atmosphere, which could adversely affect the health and safety of some of the public. Informed public opinion confirms this cautious view and is unlikely to be reassured except by reasoned argument. Plans for remedial action against the emergency following an unplanned radioactive release have been formulated in many countries, but may need revision as reactor development continues. This cannot be done without some fore-knowledge of the likely consequences of a release. Yet experience even from the few past accidental releases is limited, and relates to reactor types and accidents which may not be typical of present-day or future power reactors. However, such experience as has so far accrued may be combined with basic data such as fission yields, the physical and chemical properties of fission products and transuranic elements, evidence from experimental fuel tests and information regarding the radiological significance of the isotopes likely to be released. In this way forecasts can be made about radiation hazards to be countered and the relative levels of risk, should a release occur as a result of an accident, such as is conventionally assumed for a given reactor type. This evidence is reviewed in the paper, and deductions are made, for example about the radiological significance of isotopes other than iodine in future systems. The paper concludes by discussing with future systems mainly in mind, emergency surveillance and some remedial actions which might be put into effect. 1. INTRODUCTION Nuclear power reactors must be designed, constructed and operated in such a manner that uncontrolled releases of radioactivity do not occur. Of course, during normal operations, sm all quantities of radioactivity must be discharged in gaseous and liquid effluents, under controlled conditions that ensure that designated dose-lim its are not exceeded, if the reactor station is to continue to operate in a manner which will minimize radiation doses received by operators and members of the general public. It is considered important that dose-rates be reduced as far below ICRP dose lim its as is "readily achievable, economic and social considerations being taken into account" [1], and to this end many national regulatory bodies and reactor station ow ner-operatorssetlow er lim its which are some sm all fraction of those recommended by the International Com ission on Radio logical Protection. 395 396 BEATTIE In spite of every care which is taken, reactors, being designed, cons tructed and operated by men, are fallible and accidents can happen. There exists always a finite but sm all probability that an accidental and uncontrolled release of radioactivity to the atmosphere could occur, which might adversely affect the health and safety of some mem bers of the general public. Plans for rem edial actions to protect people during the period following such a release need to be formulated, and such plans may need to be revised from time to time as reactor development continues. But this cannot be done without some speculative calculations about the nature of future possible accidental releases of activity and their probable consequences to people and the environment. 2. THERMAL NEUTRON REACTORS For most present purposes we need not distinguish gas-cooled reactors from water-cooled reactors; we may consider in one all-em bracing section of the paper magnox-uranium gas-cooled reactors, and gas-cooled reactors fuelled with enriched uranium dioxide fuel in stainless steel cans, and all the present types of pressurized water reactors and boiling water reactors whether they use light water or heavy water or both. Regardless of probabilities, what would be the radiological consequences of an accidental release of fission products? There is only a little evidence on which tobase our predictions — one or two accidents in the not-so-recent past to reactors now obsolete, and some laboratory experiments in which fission product release was m easured from irradiated fuel which was deliberately over heated. To this we can add an extrapolation to higher temperature from our knowledge of fission product activity released from fuel in norm al operation. So one foresees that xenon, krypton and iodine isotopes and caesium -137 are likely to predominate in the release. I and a colleague have previously painted a picture of the likely effects of such a release [2]. To consider for sim plicity one example only — about 10 ^ curies of gaseous and volatile activity released (of which about 2x10^ Ci are xenon and krypton), in which the m ost important components would be 10^ Ci of iodine-131 and about 10^ Ci of caesium-137. If we consider the release to have occurred in Pasquill С or D conditions, so-called 'average weather' which occurs about 60% of the time in many tem perate regions, the effects would be as follows: (a) Whole-body external radiation from gamma rays from the cloud of gaseous and volatile fission products would be a few rads on the reactor site, but very sm all and insignificant off site. The inhalation of iodine would deliver a dose of 25 rads to a young child's thyroid at 1.6 km (1 mile). Remedial action is possible; a tablet containing 100 mg stable potassium iodide or iodate, if taken by mouth within two hours of inhalation of the radioactive iodine, will reduce the thyroid uptake of iodine and the thyroid dose by a factor of 10 or more. (b) Iodine would be deposited downwind. The possibility that young children could be drinking milk contaminated with iodine-131 would necessiate a ban on the sale and consumption of milk over an area about 10 km long downwind and some 1 to 3 km wide. The ban would last for several weeks and the cost of compensating milk suppliers would be of the order of £10 000. People would be exposed to significant levels of IAEA-SM-180/6 397 external gamma radiation from iodine isotopes deposited on the ground and this radiation would persist for some 3 to 4 weeks after the release; the distance affected would be only 150 m or so downwind and, since this distance is so short, temporary evacuation of only a few local residents at m ost is likely to be required. (c) After the gamma radiation from iodine deposited on the ground had decayed to insignificant levels, longer term low levels of gamma radia tion, from the barium-137m daughter of caesium-137 deposited, would become more apparent. At 1 km downwind the level of deposited caesium -137 would be about 10*^ Ci/rn^ and the gamma dose-rate would be 0.5 rad/a initially. This dose-rate would decline by about half in three years, and thereafter would decline more slowly with an effective half-life of 10 to 20 years [3]. Because 0.5 rad/a is the ICRP dose lim it for whole-body radiation to the general public, it is difficult to dism iss such long-term exposure as unimportant. It might well inter fere with the occupation and use of valuable land unless the land could be decontaminated in some way. Caesium-137 has a radioactive half-life of 30 years (93.5% of disintegrations result in barium-137m which has a half-life of 2.55 minutes and emits a 0. 662 MeV gamma ray). Caesium becomes entrapped in the crystal lattices of clay m inerals [4]. The observations made using a number of different soils at Harwell appear to show that this process occurs in one to three years, by which time caesium -137 will have moved downwards in the soil profile to a depth of a few centimetres [3]. What chemical remedies could be applied to soil so contaminated are not clear from the studies which have been made of radioactive contaminants in soil [4]; possibly some treatment could be devised by those who are expert in this field, sim ilar to the reclamation techniques used in England and the Low Countries to hasten the removal of sea salt from flooded coastal lands. A more drastic remedy, less taxing in ingenuity in soil chem istry, would be rem oval of 2 to 3 cm of top-soil at the earliest stage possible. Good land with deep top-soil would probably recover at once from this operation. Poorer land with only a thin top-soil could be rendered unproductive for many years, unless replacement top-soil were provided/ As to contamination of. buildings, paved roadways and open spaces, some rudimentary research on the problem of removing caesium contamination, possibly with weak acid washes, would seem a desirable precaution now. High temperature gas-cooled reactors, using one of various form s of graphite-based fuel and high pressure helium coolant, seem at present to be one of the more prom ising lines of development for thermal reactors. They open up possibilities of using steam conditions equal to or better than present-day advanced gas-cooled reactors, and prospects of direct-cycle gas turbines on the one hand, and nuclear steel making for example on the other. Generally speaking this type of reactor has many important character istics tending towards improved safety. Nevertheless possibilities for core over-heating can be envisaged, remote though these possibilities may seem to many proponents of the system ; it behoves one to consider at least the nature of the fission product release which might be possible in such circum stances. Iodine appears to be of minor significance, but caesium -13 7 could be of importance. If so, the rem arks concerning caesium-137 deposition, 398 BEATTIE which were made in the previous paragraph, again assum e importance. Strontium-90 is another long-lived fission product which could be released from this type of gas-cooled reactor in these extreme accidental circum stances. This is a beta-em itter with a radioactive half-life of 28 years, and its radio biological hazard-potential is well known. However recent studies have shown that it is less dangerous than was at one time thought, and probably caesium -137 is of much greater hazard potential. 3. FAST NEUTRON REACTORS It seem s natural to lead on from the previous paragraph to a brief mention of gas-cooled fast breeder reactors. Admittedly these are but 'paper' reactors at the present tim e, but their good breeding potential seem s likely to favour their introduction eventually, and some speculation about their potential for accidental release of activity may not be out of place. Assuming, as seem s essential for any reactor, that fuel vaporization caused by rapid reactivity addition will be excluded, one feels that the conclusions of the preceding paragraph are valid — perhaps doubly so. The higher gas pressure required of the helium coolant would appear to increase the degree of accidental fuel heating possible should a loss of pressure occur. Thus iodine-131 and caesium-137 may m erit our attention for this type of reactor, to o . But, of course, it is the sodium-cooled uranium dioxide fuelled fast reactor which is the focus of world interest in fast reactors at the present time, and it is this type of fast reactor which is being built today. Excluding for the moment, as again seem s essential to the author, fuel vaporization due to rapid reactivity addition, the sodium-cooled fast reactor appears to have particularly favourable characteristics abrogating fission product release. If limited localized damage occurred to fuel in a part of the core, this would release noble gases and possibly traces of iodine and caesium to the cover gas where they would be contained without significant change in cover gas pressure; the m ajor part of fission products released from fuel would be retained in the sodium coolant. Irradiated fuel is, in the course of normal operation, removed from under the sodium coolant, possibly after an interval to allow for reduction of after-heat. If cooling were then lost in the defuelling machine, fuel could melt and iodine-131 and caesium -137 could be released, but would be contained within the building. Because liquid sodium is at atmospheric pressure (thus loss of coolant accidents can be eliminated by design) and is an excellent medium for absorbing fission products, sodium-cooled fast reactors have inherent safety from m ost points of view. The one serious and still outstanding area of doubt is the possibility that widespread fuel movements of sufficient rapidity and co-ordination could result in rapid vaporization of some of the fuel. Either such an event must be prevented or the potential release suppressed. An elementary study of the consequences of such a release make this amply clear. If one considers the vaporization and escape of only some 100 to 2 00 gram s of this fuel, one is considering the release of all the activity referred to in Section 2 above (i.e. 103 Ci iodine-131, etc. ), and also the release of some 2x104 Ci of long-lived solid fission products (^Zr, Мбрд^ ^ IAEA-SM-180/6 399 of some 2 5 to 50 gram s of plutonium and other hazardous transuranic iso topes. The consequences, in brief, would include a total lung dose of about 100 rem to a child inhaling the air downwind at 1. 6 km (1 mile) and, from activity deposited on the ground at 1 km downwind, an initial external gamma radiation dose-rate of about 100 rad/a, declining in two to three years to the ICRP dose limit of 0. 5 rad/a. The same area would be contaminated with transuranic elements at a level several tim es greater than is normally considered acceptable, and would no doubt require decontamination. That such releases from fast reactors must be prevented and/or contained is o b v io u s. 4. REFLECTIONS ON ENVIRONMENTAL SURVEILLANCE AFTER AN ACCIDENT What lessons about environmental surveillance after an accidental release from a power reactor may be learned from the foregoing? Simply that more attention may need to be paid to monitoring ground contamination, particularly with caesium -137/barium -137m , and to devising ways of decon taminating areas so contaminated. In recent years too much attention may have been paid to methods for monitoring iodine and controlling iodine inhalation; alternatively it might be said that enough work has been done in that area. As to fast reactors, some time in the future, when they are as commonplace as thermal neutron reactors are today, routine environ mental monitoring in their vicinity will perhaps include a search for plutonium in order to re-assure the more nervous mem bers of the public. But monitoring instrum ents, other than those already considered adequate for the aftermath of an accidental release from a therm al neutron reactor, should not be required. The reason to my mind is that fast reactors will of necessity have been designed so as to preclude accidents in which even sm all quantities of fuel could be vaporized and released. REFERENCES [ 1] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION. Recommendations of the International Commission on Radiological Protection, ICRP Publication 9, Pergamon Press, London, New York (1969) 10. [2] BEATTIE, J. R., BELL, G. D., "A possible standard of risk for large accidental releases", Principles and Standards of Reactor Safety (Proc. Symp. Jülich, 1973), IAEA, Vienna (1973) 11. [3] GALE, H .J., etal., TheWeathering of Caesium-137in Soil, UKAEARep. AERE-R4241 (1963). [4] SCOTT RUSSELL, R., Ed., Ch. 14, Radioactivity and Human Diet, Pergamon Press, London, New York (1966). DISCUSSION M. B. BILES (Chairman): You made a very important point when you said we should begin studying ways of cleaning up caesium from soils. This also applies to plutonium and other persistent isotopes. On the basis of our experience in the United States of Am erica, soil rem oval should be avoided if possible, as this would involve disposing of large quantities of contaminated s o i l . 400 BEATTIE T. R. FOLSOM: In the United Kingdom you have adopted the practice of diluting radiostrontium and radioiodine with non-radioactive iodine and strontium and it would appear equally feasible to use a sim ilar technique for soil contaminated with caesium -137. In view of the cost of caesium , however, and since potassium and caesium appear to travel together through the biosphere, it would seem appropriate to use potash fertilizer, adding it directly to the soil and tilling it down as deeply as possible. J.R . BEATTIE: That is an excellent suggestion. One would have to study the relative uptakes and movements of, and chemical exchange between, potassium and caesium in order to obtain the necessary basic information for practical application. M. P. BUTAYE: In making your dose risk forecasts as a function of distance, you must have made some assumptions regarding meteorological conditions. Could you please say what these assumptions were? J.R . BEATTIE: The dose-rate calculations in my paper were performed assuming Pasquill С or D weather conditions, which corresponds approxi mately to zero lapse or sm all temperature gradient in the atmosphere. The wind speed assumed was 5 m /s, which is typical of this weather category. As to calculating risks and probabilities (which are not discussed in this particular paper, but are discussed in other papers we have written on other occasions) we take fully into account all Pasquill categories and all possible wind speeds with their probabilities of occurrence derived from recorded m eteorological observations and statistics. N.G. GUSEV: We have calculated that, in the event of accidental release of 24Na from a fast reactor, the risk to the population would be represented not by internal irradiation but by external gamma radiation from a plume containing sodium aerosols, as well as from sodium -24 deposited on the ground. Have you made sim ilar calculations ? J.R . BEATTIE: I agree that gamma radiation from deposited sodium-24 has to be added to the gamma radiation from deposited fission products considered in my paper. One difficulty in the analysis is to estim ate the quantity of sodium which would be released, which would vary with the particular reactor design. Fortunately sodium-24 has a relatively short half-life compared to the long-lived fission products I discussed in my paper. One purpose of the paper was to draw attention to the importance of these long-lived fission products in the case of hypothetical vaporization of fuel. N. G. GUSEV: You described hypothetical releases in the event of a fast reactor accident. Did you take into account the inevitable process of fractionation or are you considering a natural mixture. J.R . BEATTIE: The fission product mixture which I assumed for the hypothetical release from a fast reactor was what you have termed a "natural m ixture", that is to say all the fission products in the arbitrary quantity of 100-200 g of fuel were assum ed to have been volatilized. If the fuel tem pera ture exceeded about 5000°K, studies show that all fission products would be volatilized; however, if the fuel temperature were about 4000 К or less, I believe fractionation could take place and many of the long-lived fission products would tend not to be released. P. R. KAMATH: In the light of your calculations would you suggest one mile as the safe exclusion distance for therm al and fast reactors. J.R . BEATTIE: No, that would not be my philosophy. The size of release (103 curies of iodine-131, and various amounts of other radionuclides which I mentioned) which I used as the basis of the calculations, is one which IAEA-SM-180/6 401 I chose quite arbitrarily. It is a purely nominal figure of release, and the distance of one mile which appeared as a result of the calculations is also therefore nominal or arbitrary in character. Whether the exclusion distance should be 100 m etres, 1 kilometre or 1 m ile, is in my view a political question, and cannot therefore be answered solely on the basis of the kind of technical calculations which I have discussed here. В. M. MICHAUD: You took as an example for thermal reactors a dis charge of 105 Ci of gaseous and volatile effluents. For the purpose of establishing safety m easures to protect the public, what amount of discharge should one assum e at the present time for a maximum credible accident, in C i/G W (th )? J.R . BEATTIE: This is not a question which can be answered solely on the basis of the calculations presented in my paper. How is one to say what the size of release may be, without also investigating and attempting to state the probability of occurrence of this release? I took an arbitrary figure of 103 curies of iodine-131 in my paper. The iodine-131 inventory of a 2000 MW(th) reactor is about 6x107 curies, and I suppose some figure less than this might be considered the absolute upper limit of release; all the same the probability of such a release occurring must be very sm all indeed. But should one attempt to base emergency control m easures for protection of the public on a hypothetical accident which is so improbable that it might be assigned a probability of, for example, less than once in a million years — in other words in practical term s it will never happen. It would seem reasonable to me that one should take into account the kind and size of release which is more probable — say once in a hundred or a thousand years, and this, I believe, would be a sm all release. I know some people have considered 100 or 1000 curies of iodine-131 in this context, but that is their decision, not mine. IAEA-SM-180/20 RADIATION HAZARDS TO THE POPULATION RESULTING FROM CONVENTIONAL AND NUCLEAR ELECTRIC POWER PRODUCTION* Z. JAWOROWSKI, J. BILKIEWICZ, ElibietaDOBOSZ, Danuta GRZYBOWSKA, Ludwika KOWNACKA, Z. WRONSKI Central Laboratory for Radiation Protection, Warsaw, Poland Abstract RADIATION HAZARDS TO THE POPULATION RESULTING FROM CONVENTIONAL AND NUCLEAR ELECTRIC POWERPRODUCTION. contaminants are presented in relation to sources of emission and meteorological factors. The possible increase of the radiation dose to the population from these nuclides is discussed, comparing it to the exposure resulting from the operation of nuclear power plants. The results o f the study on the release to the 1. INTRODUCTION The aim of this work is to compare the relative hazards to populations caused by radioactive pollutants emitted as a result of fossil-fuel and nuclear power production. There is a scarcity of information on the dispersion of radionuclides from fossil-fuel power stations as compared with nuclear ones. In addition, the conventional production of energy also introduces into the environment agents that are toxic, though not radioactive. This m akes the adverse effects on population from the two types of pollution sources difficult to compare. The basis for assessm ent of radiation hazards from these sources should be information on concentrations of relevant radionuclides in the bodies of exposed m em bers of the population and in their immediate environment. In the case of artificial radionuclides emitted by nuclear power stations the situation is clearer than with conventional ones, as in the form er the levels of the m ajority of the radioisotopes in the population and environment originate from this source and the amounts dispersed are rather well known, whereas the natural radioisotopes dispersed from fossil-fuel power stations are added to their existing natural levels. It seem s, therefore, thatit is a prerequisite to determine the natural levels of the natural radionuclides in the environment and in the human body and to assess the possible influence of industrial activity on these levels. Anderson et al. [1] and Eisenbud et al. [2] were among the first to estim ate the radioactive pollution of the atm osphere due to burning fossil-fuel. * This work has been performed under the auspices of the US Environmental Protection Agency, under Research Contract No. 5-531-1. 403 404 JAWOROWSKI et al. Martin et al. [3,4] and Bedrosian et al. [ 5], calculated the relative radiation doses from both types of em ission, based (i) on measurements of the radioactivity of dust collected during air sampling near three coal-fired stations and (ii) on releases of noble and activation gases from several power reactors. They compared the local radiation hazards from both types of power plants, assum ing that, for the radionuclides emitted by fossil-fuel plants, the main route into the body is through inhalation. The results of measurements of releases of radium-226 from conventional power plants have already been published by our laboratory [6, 7], in addition to calculations of global em ission of this nuclide and le a d - 2 1 0 . In this paper, new data are presented on the temporal changes of ^ R a , uranium, thorium, stable lead, vanadium and cadmium in the environment and on their geographical distribution. Also the results of measurements of the concentrations of these nuclides in soil as a function of the distance to em ission sites are reported. These data, and also the concentrations of radium-226 in contemporary and old human bones, are used for making an assessm ent of the radiation hazards arising from operation of fossil-fuel p l a n t s . 2. TEMPORAL CHANGES The natural content of radium-226 in the atmosphere originates from inorganic m atter introduced into it by volcanic fumes, forest fires and m eteoritic dust, and by transfer of sea water and soil to the atmosphere by the action of the wind. It is now difficult to establish the true level of this natural content of radium-226 in the contemporary atmosphere and in precipitations. The study of tem poral variations in concentrations of pollutants in the environment may be used in making an assessm ent of this content. To this end, we used a collection of glacier ice sam ples ranging in age from 800 years old (12th Century) to new sam ples (1973). These sam ples were collected from Storbreen, in the Jotunheimen Mountains in Norway, a location remote from industrial centres. As may be seen from Fig. 1, the concentrations of radium-226 in ice increased from the level of 0. 0006 pCi per kilogram of ice in 12th Century to 0.0032 pCi/kg in 1973. This approximately five-fold increase in ^ R a concentration may be compared with the approximately fifty-fold increase observed during the past 80 years in the Tatra Mountains, Poland, which are located close to centres of industrial activity [6] . The concentrations of radium-226 in the same year also differ in the two localities; in 1970, for example, the concen trations were two orders of magnitude higher in Central Europe than in Norway. The concentration of 0. 0006 pCi ^R a/kg found in the layers dating from thel2th Century in the Norwegian glacier is the lowest value found by us in precipitation. If it is supposed that this is the natural concentration for precipitation in this region, and using the basic expression of Slade [ 10] : D = V j T X where D is the deposition (pCi/m^), V is deposition velocity (m /s), T is total duration of deposition (seconds) and X the average concentration in air IAEA-SM-180/20 405 1870 1Б70 1200 at the deposition point (pCi/m^), and assum ing 800 mm of precipitation per annum, the average natural air concentration of radium-226 over Norway in the 12th Century may be deduced to be about 3.0 X 10"^ pCi/m^. This may be compared with the recent concentrations of radium-226 in the atmosphere, ranging from 0.8X10*4 pCi/m^ in New York [ 2 ] to 2. 4 X 10*3 pCi/m^ near a fossil-fuel power station in Widdows Creek, USA [5], figures which are three to four orders of magnitude higher than the natural concentration. Sim ilar trends in the tem poral variations of concentrations were observed in the cases of uranium, stable lead, vanadium and cadmium. The increases in the concentrations of uranium and vanadium have proceeded rather slowly, though the increases have been more pronounced in recent years. Concentrations of cadmium increased rapidly during the last decade, being three tim es higher in 1973 than in 1960. Stable lead concentrations in Norwegian glacier ice increased about 10 times in contemporary sam ples as compared with those from the 12th Century (Fig. 2). 406 JAWOROWSKI et at. It is interesting to note that the atmospheric increase of radium-226 was accompanied by a sim ilar increase in the concentrations of this nuclide in tim ber sam ples (Pinus silvestris) from the 20th Century as compared with thosefrom the 18th Century (Tablel) [7 ]. This suggests that sim ilar tem poral trends might also occur in other form s of life, including the human population. IAEA-SM-180/20 407 TABLE I. CONCENTRATIONS OF ^ R a ipj pi^E T R E E S [ 7] '"Ra (pCi/g ash) 18th Century: 0.03 В 0. 01 с 0.007 Contemporary: 0.14 - 0.27 0.65 - 0.85 3. GEOGRAPHICAL DISTRIBUTION It is interesting to see that the concentrations of lead in precipitation in Norway are about a factor of ten lower than in precipitation in Central Europe, whereas in the case of radium -226 this difference reached a factor of 100. This may be caused by the greater volatility of lead as opposed to that of radium at the tem perature of fuel combustion, which leads to the formation of sm aller particles of lead, travelling to greater distances from the source than the larger particles containing the greater proportion of the radium -226. The existence of a fractionation effect on fly ash composition that is related to the size of the dust particles was reported in earlier work [7], in which we also found that, on a local scale, the greater fall-out of radium -226 was not in the vicinity of the source of em ission but many kilom etres downwind. We found that the concentrations of radium -226, uranium, thorium and lead in the soil were related to the distance from the emission sources. The soil sam ples were collected at two depths, one between 0 and 5 cm from the surface and the second between 5 and 10 cm. As may be seen in Table II, the upper soil layers contain greater amounts of these pollutants than the lower ones in industrial regions. Assum ing that the differences between the concentrations of radium-226 in the upper and lower soil layers represent the long-term deposition of industrial fall-out, these differences were used in calculating the average radium-226 concentration in the ground-level air over the particular sampling points. Using the generalized equation of Slade [10, 11 ] : where X is the average concentration in air at the deposition point (pCi/m^), 8 the sector width (radians), Dx the deposition (pCi/m^), x the distance to the point of deposition from the em ission source (m etres), &z the standard deviation of the vertical concentration distribution (m etres) (see Ref. [ 10] ), 408 JAWOROWSK1 et al. TABLE II. CONCENTRATIONS OF ^R a TWO LAYERS OF SOIL M'Ra Th U Pb ^catio n (PCi/g) (ppm) (ppm) (pg/g) 0-5 5-10 0-5 5-10 0-5 5-10 0-5 5-10 Depth cm cm cm cm cm cm cm cm POWER STATIONS 1 km 0.100 0.328 1.91 1.94 0.57 0.57 39.5 29.5 15 km 2. 255 0.491 4. 02 1.63 4. 02 0.40 171.6 37.0 60 km 1. 093 0.902 3.76 2.73 0.88 1.17 35.7 39.4 120 km 0.256 0.554 2.48 2.68 0.64 0.63 54.2 54.5 Siekierki-W arszawa 300 m 0.673 0.489 5.21 4. 55 0.93 0.62 -- 1 km W 0. 580 0.420 2.45 2.08 0.93 0.2 -- 15 km 4. 086 0.721 2.24 1. 82 0.63 0.44 63.9 33.4 Adamów 0. 5 km 0. 326 0. 385 2.40 2.28 0.96 0.78 43.4 32.8 4 km 1.147 0. 804 1.71 2.78 1.14 0.38 36.4 31.9 10 km 0.496 0.560 1.64 1.75 0.47 0.73 35.9 39.5 20 km W 0. 692 0.238 2.35 1.70 0.70 0.48 36.1 69.5 300 m 0.366 0.404 2.43 2. 55 0. 58 0.63 32.4 34.7 20 km 0.183 0.718 1.36 1. 37 0. 39 0. 39 30.8 71.9 30 km W 0.180 0.993 1.53 2.36 0.57 0.25 28.7 24.8 RU R A L AFLEAS A 0.856 1.038 4.02 3.60 1.61 2.33 36.5 31.9 В 0.358 0.928 1.63 1.45 0.96 1.16 36.5 28.0 с 1. 626 1.000 2.81 2. 84 0.54 1.14 53.6 - D 1.090 1. 053 10.38 7.18 1.32 1.34 - - E 1.071 0.669 2.81 5.05 1.17 0.81 -- F 1.451 0.758 4. 99 6.41 1.19 1.44 -- G 0.424 0.585 3.49 2. 85 1.12 1.42 48.2 48.4 К the dispersion factor (m*2) (see Ref. [10]), the deposition velocity, taken as 0.004 m /s, T the total duration of deposition (seconds), and h the effective release height, we found that, depending on the weather category, the radium-226 concentrations in ground level air under the plume may vary from 4. 31 X 10"^ to 5. 77 pCi/m^ (Table III), if the initial weather category remained constant throughout the deposition time. The content of radium -226 in the fly ash from three coal-fired power stations in Poland was found to range from 0.4 to 4.2 pCi/g. In the fly ash sam ples with high concentrations of radium-226, the solubilities of this nuclide in 1M ammonium acetate and IN HC1 were 5.0% and 36.8%, respectively, i. e. higher than usually found in soils [ 12] . IAEA-SM-180/20 409 TABLE III. ^ R a CONCENTRATIONS IN GROUND-LEVEL AIR, CALCULATED FROM SOIL DEPOSITION '"R a " H i ; (pCi/m3) ZeraA - 15 km F 2.39 D 4.55 x 10'2 Siekierki * 15 km F 5.77 D 1. 92 X 10"i Adamów - 20 km F 5.68 X 10'i D 4. 31 X 10"^ In vegetables and other plants the concentrations of radium -226, uranium and thorium were found to be related to the distance from the em ission source, and higher in industrial areas than in the agricultural ones. This indicates that a part of the natural radionuclides dispersed by fuel consumption might enter the human body through the gastro-intestinal tract. Our plant data, however, are too sparse to permit of their being used for calculating their contribution to the body burden of these nuclides in the population. 4. RADIATION DOSES To calculate radiation doses to the population the air concentration values were therefore used. The highest air concentration of radium-226, calculated for weather category F from radium -226 deposition in soil, was 5.77 pCi/m3. The radiation dose in bone for people living at this deposition point was calculated assum ing that the weather diffusion category F [ 13] existed for 10% of the deposition time, taken as 12.5 years, and that the winds blew from the em ission source to the deposition point for 50% of t h is t im e . As the F category in this area is not a frequent weather category, and as only a minute fraction of the population lives in the area burdened by so large an industrial fall-out, the use of this value of radium-226 air concen tration for body-burden calculations may be regarded as conservative. With the ICRP particle inhalation model and the expression describing the relations between tissue concentrations of a nuclide, C, and its daily intake. I, [14]: ? b * I 0 .6 9 3 X m where T^ = 16 000 days, m = 7000 g, we calculated that the radium-226 deposited in bones by inhalation under the conditions described above is 0.425 pCi per gram of ash, assum ing 5% radium solubility in the fly ash. This corresponds to 365 m rem /a to the osteocytes. 410 JAWOROWSKI et at. At the sam e deposition point, the radium-226 air concentration for the weather category D, which exists at this point for 35% of the deposition time, was 1. 92 X 10"^ pCi/m^. This corresponds to a bone concentration o f 0.05 pCi per gram of ash; the corresponding radiation dose rate to the osteocytes is 43 m rem /a. These two values are about five times higher than the average ones found by Stahlhofen for the German population, but not much different from his maxima, which reached 0.03 pC i^R a per gram of bone ash and 25.8 m rem /a for osteocytes. Even higher radium-226 concentrations in bone have been reported for the Am erican (USA) population, reaching 0.19 pCi/g of ash [16]. It seems, therefore, that such a large contribution from industrial sources to the radium-226 body burden, as calculated for high-deposition areas, are not unrealistic ones. It might be supposed that besides inhalation, industrial radium-226 may enter the body by ingestion with vegetables, fruits, etc., since we found the plants to be contaminated with this nuclide. We have taken this contribution to be negligible, as in 15 sam ples of human bones from past centuries we have found radium-226 concentrations ranging from 0.010 to 0.04 pCi per gram of ash, i.e. sim ilar to contemporary levels. The contribution of 43 m rem /a to the radiation dose was calculated for rather extreme conditions and for people living in the vicinity of coal- fired power stations. The average concentration of radium-226 in air m easured in Warsaw was 1. 6 X 10*5 pCi/m^. Inhalation of this air would result in a contribution to radium-226 level in bones of only 2.3 X 10'^ pCi per gram of ash, corresponding to a radiation dose of 0. 02 m rem /a; this may be neglected as compared with the natural one (Table IV). The osteocyte dose caused by inhalation of radium-226 dispersed by fossil-fuelled power stations may be compared with the doses from much larger nuclear power stations in the USA: Dresden I (BWR) near Chicago, Illinois, and Yankee (PWR) at Rowe, M assachusetts [ 17] . At Dresden I, the inhalation dose at the exclusion boundary was found to be 7 X 10'^ m rem /a for the thyroid from and at Yankee, 0.15 m rem /a for the whole body fr o m Зн. TABLE IV. CALCULATED ^ R a CONCENTRATIONS IN BONE И 'Ra Method of evaluation ï a t l ï y (pCi/g ash) " E r Calculated from soil F 0.43 365 concentrations D 0. 05 43 Calculated from air concentration in Warsaw (average) 0. 000023 0. 02 Measured in bone (average) 0.012 1 0 .3 [1 5 ] IAEA-SM-180/20 411 TABLE V. RADIATION DOSE RATES TO CRITICAL ORGANS FOR THREE POWER STATIONS Power station (mrem/a) (mrlm)fperMW(e)) Siekierki, 43toosteocytes 0.11 fossil fuel, 15 km Dresden 1, 7 к 10"^ to thyroid (*^ 1) 10*7 nuclear (BWR), exclusion boundary Yankee, 0.15 to whole body 2 x 10"* nuclear (PWR), exclusion boundary Expressed in relation to the MW(e) outputs of the power stations, the radiation dose received by neighbouring populations by inhalation is several orders of magnitude higher for a fossil fuel power station than for a nuclear one (Table V). It has previously been suggested that a part of lead-210 and its daughters in the atmosphere originates from industrial sources [7, 18]. Therefore it may be supposed that the radiation dose from radium -226 emitted by fossil-fuel power stations is accompanied by doses from other naturally- occurring radionuclides, such as lead-210, bismuth-210 and polonium-210, and also thorium, radon-222 and others. Besides radioactive pollutants the combustion of fossil fuel introduces into the environment vast amounts of non-radioactive w astes, among them being toxic heavy m etals. Lead, cadmium and vanadium levels have increased in atmospheric precipitation in the past century, as indicated by our glacier studies. At the present state of knowledge, it is difficult to compare the radiation burden of the population with the effects of these toxic pollutants. It can be stated, however, that the radiation doses to the general population, which are, relatively, higher with fossil-fuel power stations and sm aller with nuclear ones, are minute enough to be neglected as a health hazard. On the other hand, there is no evidence that the non-radioactive pollutants dispersed in the course of conventional power production can be sim ilarly neglected. LITERATURE [1] ANDERSON, W ., MAYNEORD, W .V ., TURNER, R .C ., Nature (London) 174 (1954) 424. [2] EISENBUD. M ., PETROW, H.G. , Science 144 (1964) 288. [3 ] MARTIN, J. E. , HARVARD, E. D. , OAKLEY, D .T ., "Radiation doses from fossil-fuel and nuclear power plants", Ch. 9, Power Generation and Environmental Change (BERKOWITZ, D .A ., SQUIRES, A .M ., Eds) , Symp. Committee on Environmental Alteration, Am. Ass. Adv. Sei. , 28 Dec. 1969. [4 ] MARTIN, J . E . , HARVARD, E .D ., OAKLEY, D .T . , SMITH, J.M . , "Radioactivity from fossil-fuel and nuclear power plants", Environmental Aspects o f Nuclear Power Stations (Proc. Symp. New York, 1910), IAEA. Vienna (1 9 4 ) 325. 412 JAWOROWSKI et al. [5] BEDROSIAN, P. H ., EASTERLY, D. G ., CUMMINGS, S. C ., Radiological Survey Around Power Plants Using Fossil Fuel, USEPA Rep. EERL 71-3 (1910). [ 6] JAWOROWSKI. Z . . BILKIEWICZ, J. , ZYLICZ, E. , Health Phys.20 (1 9 4 ) 449. [7 ] JAWOROWSKI, Z ., BILKIEWICZ, J. , KOWNACKA, L., WLODEK, S ., "A rtificial sources of natural 7-11 Aug. 1972. [ 8] JAWOROWSKI, Z .. Rep. NEIC-RR-29 (1967). [9 ] JAWOROWSKI, Z., At. Energy Rev. 7_(1969) 3. [10] SLADE, D.H., Ed.. US AEC Rep. TID-24190 (1968). PHS (USA) Rep. BRH/DER 70-1 (1971). [12] RUSSANOVA, G. V ., inMaterialyRadioekologicheskikhlssledovanijvprirodnykhBiogeotsenozakh, KOMI branch of the USSR Academy of Sciences, Institute of Biology, Syktyvkar, (1971) 50. [13] PASQUILL, F ., Atmospheric Diffusion, Van Nostrand, London (1962). [ 14] INTERNATIONAL COMMISION ON RADIOLOGICAL PROTECTION, ICRP Publication 2, Report of (1959); also Health Phys. ^ (1960) 1. [15] STAHLHOFEN, W ., THESIS, J.W ., Goethe University, Frankfurt am Main (1964). [16] HOLTZMAN, R.B., USAEC Rep. ANL-6199 (1960). [17] BLANCHARD, R. L ., KAHN, B. , "Pathways for the transfer (of radionuclides) from nuclear power reactors through the environment to m an", Proc. Symp. Radioecology Applied to the Protection o f Man and his Environment, Rome (1971) 175. [18] PEIRSON, D .H ., CAMBRAY, R .S ., SPICER, G .S ., Tellus 18 (1966) 427. DISCUSSION O. J. A. TIAINEN: If we compare the hazards associated with conven tional and nuclear power stations, natural gas-fired power stations should also be taken into account. These are economically attractive in some countries and the hazards are sm all. IAEA-SM-180/30 INTERPRETATION OF AEROSOL RELEASE MEASUREMENTS AT NUCLEAR POWER PLANTS R. FRITZE Technischer Überwachungs-Verein Norddeutschland eV, Hamburg G. HERRMANN Technischer Überwachungs-Verein Hannover eV, Hannover, Federal Republic of Germany Abstract INTERPRETATION OF AEROSOL RELEASE MEASUREMENTS AT NUCLEAR POWER PLANTS. Five nuclear power plants had been put into operation in the Federal Republic of Germany by 1972, and 15 more are expected to be in operation by 1980. Increasing attention is, therefore, being given to the interpretation and standardization of release measurements, since they provide the most effective means for immediate surveillance of the impact on the environment. Problems mainly connected with aerosol radio activity are discussed in this paper. The licensed release rates for aerosol activity for the five plants lie between 1.2m Ci/handl.8Ci/h. How the release is to be measured is, however, not specified. Commonly, aerosols are collected on a filter in a by-pass to the stack and their activity (natural, corrosion and fission products) is measured during collection and after certain decay periods with a 6-sensitive counter. For this paper, the Lingen plant (BWR) and the Stade plant (PWR) have been taken as reference plants and release rates of radio active aerosols and nuclide compositions are reported. The releases from BWRs contain a relatively large fraction of short-lived radionuclides, and weighting the values of the activities released on the basis of the half- lives is possible, if revised release rates are not granted, and a balance is drawn accordingly. For PWRs, it is suggested that the weekly release rates be determined from a sample collected over the period of one week, measuring the 0-activity at the end of the collecting period. The reasons for and against either y-monitoring or 6-monitoring systems for the measurements made during collection are discussed on the basis of the nuclides which determine the licensed release rate. Both methods can in principle be used. In PWRs it may be advantageous for plant control to use y-spectrometry, as this may give an indication of unexpected releases inside the plant. As a part of the aerosols is lost in the pipe leading from the stack to the monitor, a 'loss factor* has been determined and is discussed. 1. INTRODUCTION By 1980 about 20 nuclear power stations are expected tobe in operation in the Federal Republic of Germany, producing 20 000 MW(e), representing about 20% of the electrical energy production. It can be anticipated that components (e. g. the fuel canning) and system s to contain radioactivity (e. g. leakage control) will be further improved, so that discharges to the environment during norm al operation can be kept "as low as possible". And it is hoped that the reliability of passive and active system s to contain radioactivity during accident conditions will be such as to make m ajor releases very unlikely. Direct determinations of radiation doses to the population in the environs of the plant will remain difficult to make and relatively expensive. Never theless, techniques for m easuring doses to representative individuals in the 413 414 FRITZE and HERRMANN population are improving (cf. Ref. [1] ). Monitoring of radioactivity in the environs of nuclear power stations is, of course, already carried out on a routine basis. This mainly serves the purpose of keeping artificial activity releases into the human environment under control, a factor which has been so severely neglected for chemically toxic agents. The first link in the chain of m easures to control the impact of nuclear technology, namely the determination of releases from an individual plant, has to be seen in the context of direct dose determinations and environmental monitoring. Its importance will increase significantly as the density of nuclear power stations rises. Standardization of release measurements is being furthered. The contribution to radiation exposure of the population from nuclear energy will remain sm all in comparison with the exposure from other man- made sources, in particular with those resulting from medical uses of radiation. M oreover, the fraction of the dose contribution resulting from releases of radioactive aerosols is sm all in comparison with the total dose due to nuclear energy. It is argued, therefore, that great precision in m easuring aerosols and efforts aimed at providing continuous monitoring are not justified. The problems involved in m easuring aerosol releases from light water reactors are discussed in this paper and some conclusions a r e d ra w n . 2. DOSES AND DOSE LIMITS A short summary concerning the present status of perm issible doses resulting from the operation of nuclear power stations in the Federal Republic of Germany (FRG) will serve to introduce the considerations regarding the measurement of aerosol releases and their interpretation. In 1969, a competent body in the Federal Republic of Germany recom mended (see, for example, Ref. [2] ), that the 5 rem s per 30 years (5 rem s per generation) recommended by ICRP for the genetically signifi cant dose shall be divided into three parts that were categorized as follows: (i) for medical applications; (ii) for nuclear technology; (iii) for other artificial radiation exposures. Of the 5 rem s per 30 years, 2 rem s were allotted to nuclear technology. (It should be noted that, in contrast with the ICRP recommendation, m edical exposures have been included in this 5 rem s per 30 years figure. ) The 2 rem s for nuclear energy are to be further divided into 1 rem allotted to exposure caused by activity discharged with liquid waste and 1 rem allotted to exposure due to activity released into the atmosphere. Only the latter figure, leading to a perm issible dose per year of 30 mrem, is of relevance to the subject of this paper. Since the density of nuclear power stations will be increasing, the above-mentioned body recommended that the dose limit for the population at large should also be applied to persons working or living in the vicinity of a nuclear power plant. This limit should, furthermore, be the general limit for whole-body exposure, that is also with regard to somatic effects. It must be mentioned here, that a dose lim it of 90 m rem /a has been recommended IAEA-SM-180/30 415 for exposure of the thyroid. No special limits have been recommended for radionuclides carried by aerosols so far, to our knowledge. Radioactivity in form of aerosols released to the environment delivers radiation doses to the population in three ways: (a) By external irradiation while it is carried in the atmosphere; (b) Due to incorporation by inhalation with the air; (c) Due to incorporation by ingestion via the food chain. Methods and models for calculating doses for these three form s of irradiation are known. They are, however, not extensively used to determine lim its for aerosol discharges into the atmosphere. A simplified method is usually used: it is assum ed that a certain concentration of an unidentified mixture corresponds to the equivalent whole-body dose. By defining the concentration limit in the atmosphere and then applying an appropriate atmospheric dilution factor, a discharge limit is obtained. Rather arbitrarily som etim es, actual pulse rates as directly observed or corrected by a factor derived from the effective half-life, or even pulse rates after one week of decay are used to determine discharge activities, which values are then compared with the discharge lim its. 3. DETECTED NUCLIDES All radionuclides which are released in aerosols derive from the prim ary coolant or are decay products of such nuclides. The fate of the nuclides in the period between their formation and their path to the release point is different in BWRs and PWRs. A quantitative model for the transfer of nuclides which become attached to aerosols has not yet, to our knowledge, been established. Nuclides that have been detected in prim ary coolant and discharged air are described in the following paragraphs. The nuclides that have been detected in the prim ary coolant of the Stade plant (PWR) and of the Lingen plant (BWR) are listed in Table I. The activity of the nuclides ^^Sr and 90Sr have to be determined by radiochem ical analysis; in the coolant of the Stade plant the activity of these two nuclides is estimated to be less than 10'S Ci/m ^. Due to leakage from the prim ary system the nuclides listed in Table I are released into the air of the plant buildings; they can only be determined in this air following their accumu lation on a filter. At the Stade plant a flow-rate of 20 m^/h of air is drawn through an aerosol filter in a by-pass to the stack. The nuclides that have been identi fied on this filter after a collection time of two hours are listed in Table II. If the analysis is carried out immediately at the end of the collection period, l3^Cs is found to be the nuclide with the highest activity on the filter; after a decay time of 1.5 hours the predominant activity is due to ^Co. The activity of the nuclides has not been determined as the total activity collected on the filter is too sm all. Only 88Rb and ^ C s have been detected during operating periods in aerosol sam ples from the Lingen plant, which are collected on filters impregnated with charcoal and analysed with a Nal(Tl) crystal. During shut-down, when maintenance work is carried out, s°Co has also been found. All the nuclides listed in Table II and produced by activation will, of course, be present, too. 416 FRITZE and HERRMANN TABLE I. LIST OF NUCLIDES WHICH HAVE BEEN DETECTED IN THE P R I M A R Y C O O L A N T O F T H E S T A D E P L A N T ( P W R ) IN J A N U A R Y 1973 AND IN THE COOLANT OF THE LINGEN PLANT (BWR) IN JUNE 1970 The nuclides of the noble gases and of iodine are omitted. Stade plant Lingen plant Nuclide H alf-life activity activity ( C i ^ ) (Ci/m") Na-24 15 h 6 .5 x 10*^ 1.1 x 10*3 Cr-51 28 d 2.3 x 1 0 *' - Mn-54 303 d 8.9 x 10*s - Mn-56 2.6 h 1 . 8x 10 "< - Co-58 70 d 2.1 x 10'' 8. 2 x 10*< Co-60 5.3 a 2 .7 x 10*5 9.6 x 10"* Fe-59 45 d 2 .5 x 10*5 - Sr-89 52 d - 2 .5 x 10*3 Sr-90 28 a - 1.1 x 10"* Sr-91 9.7 h - 1.1 x 10'* Sr-92 2.7 h - 1.4 x 10"' Mo-99 66 h 1.1 X 10*2 4 .6 x IO*" Tc-99 m 6.0 h 3 .4 x 10*^ 1.1 x 10"' Tc-101 15 min - 9.6 x 10*' Sb-122 2.8 d 8 .7 x 10*3 - Sb-124 60 d 5.6 x 10*3 - Cs-134 2.1 a 5.4 x 10"* 3 .6 x 10"* Cs-136 13 d 1. 8x 10"* 1.4 x 10"* Cs-137 30 a 1 .9 x 10*з 3 .4 x 10"* Cs-138 32 min - 7 .6 x 10"' Ba-139 84 min - 2 .5 x 1 0*' Ba-140 13 d - 6.0 x 10*3 Ba-141 18 min - 4 .7 x 10"' Ba-142 11 min - 3 .4 x 10"' Np-239 2.3 d 6 .7 x 10*3 2.6 x 10*' IAEA-SM-180/30 417 TABLE II. LIST OF NUCLIDES WHICH HAVE BEEN IDENTIFIED ON THE AEROSOL FILTER IN THE BY-PASS TO THE STACK OF THE STADE PLANT (PWR) IN JANUARY 1973 y-emission Nuclide H alf-life max probability [MeV] [MeV] Mn-54 303 d EC 0.84 1.00 Co-58 71 d EC, 6+ 0.81 0.99 0.51 0.30 Co-60 5 .2 a 0.3 1.33 1.00 1.17 1.00 Rb-88 18 min 5.2 1.86 0.21 0.90 0.13 Sb-124 60 d 2.3 0.60 0.97 Xe-133 5.3 d 0.35 0.08 0.37 Cs-137 30 a 0.51 0.66 0.92 Cs-138 32 min 3.40 1.42 0.73 1.01 0.25 0.46 0.23 4. MEASUREMENT OF AEROSOL ACTIVITY Identical types of aerosol monitor are in use in all the five nuclear power plants mentioned above. The length of the pipe leading from the stack to the monitor is between 20 m (at the Stade plant) and 55 m (at the Gundrem- mingen plant), the flow-rate of the by-pass is between 1 m^/h (at the Lingen plant) and 20 m^/h (at the Stade and Obrigheim plants). The activity collected on the filter is m easured with an end-window tube. After a collecting time of 2, 3 or 4 hours the filter is moved to a second GM tube and after a time of 40 hours or more to a third GM tube. By this means it is possible to m easure the activity during sampling and after two different decay periods; the fractions in the aerosols with short and long half-lives can, therefore, be estimated. As, commonly, the aerosol release rate is measured in a by-pass to the stack, it must be borne in mind that a fraction of the aerosols is lost in the pipe leading from the stack to the monitor. The loss factors, i.e. the fraction which passes through the pipe and is collected on the filter of the monitor, has been determined at the Stade plant. One filter-set, consisting of an aerosol filter preceding a charcoal filter, was placed directly at the stack, while a second one was placed at the end of a pipe 20 m in length and 5 cm in diam eter. Collection extended 418 FRITZE and HERRMANN over a period of 7 days, at a flow-rate of 1.7 m^/h of air pumped through the filter-sets; the activities collected on the two filter sets were compared. The loss factors are: *with reference to the ß-activity collected on the aerosol filter: Lg = 0.46 ± 0.01 Ж with reference to the activity of ^Co (Ey= 810 keV) collected on the aerosol filter: L"Co-58" 0.40 ±0-13 tw ith reference to the activity of ^ 1 (Ey = 364 keV) collected on the charcoal filter: L i -131 = 0-74 ± 0.08 Certainly the system atic errors are larger than the statistical errors noted a b o v e . The results of these m easurements show that the loss factors may influence the accuracy of the release determinations. When monitoring the activity collected on the filter, quick changes in the counting rate have been observed at the Stade plant. These changes are caused by noble gases which are observed on the filter; Table II shows that xenon-133 has been identified on the aerosol filter. As the sudden occurrence of noble gases on the filter may simulate an increase of aerosols, the continuous monitoring of the aerosol activity with a y-sensitive detector has advantages. The discrimination level of this detector should be set at 500 keV so that the detector is insensitive to the radiations of ^ X e and ^K r and nearly insensitive to ^X e. This discrimination level does not change the detection efficiency of the other nuclides listed in Table II. Sim ilar considerations are valid for a BWR if the activities of ^Sr and 9°Sr are negligible for the purpose of plant surveillance. It should be pointed out here that some attention m ust be paid to the calibration of the monitor. The calibration may be done taking the nuclide 9°Sr as reference for j3-monitoring and the nuclide ^ C s for т-monitoring. 5. INTERPRETATION OF MEASUREMENTS The measurements of aerosol activity serve two purposes. Firstly, an alarm signal should be given at a certain specified discharge rate; for this purpose continuous measurement is required, which does, however, not require high precision. Nevertheless it poses some problem s, particularly for BW Rs, where the fraction of short-lived nuclides can be expected to be higher than for PWRs. For normal reactor operation, a continuous m easure ment of released aerosol activity for alarm purposes may not be required, because increased activity releases will always be accompanied by increased gaseous activity. Since, however, in nuclear power stations waste collection and treatment facilities are provided which may become a source of aerosol IAEA-SM-180/30 419 FIG. 1. DCLs for different mixtures of nuclides. activity, in particular when malfunctioning, we believe that a continuous measurement of released aerosol activity is necessary and should be made, possibly before dilution with the entire exhaust air. Secondly, the measurement of aerosol activity serves the purpose of recording the discharged activity, which has to be compared with the discharge limit. As described briefly in Section 2, licensed release rates are at present usually determined by assum ing an unidentified mixture of radionuclides in air. It is too restrictive, however, if this is used for a nuclide mixture containing a large fraction of short-lived nuclides, the release rate, for example in Ci/h, measured by methods described in pre vious sections being compared with a perm issible release rate on the basis of 'an unidentified m ixture'. It is not feasible, however, to licence a release rate limit for each nuclide. Therefore it would be necessary either to interpret the measurements in an appropriate way, or to state discharge lim its for groups of nuclides; the group characteristics that could be used for this purpose are the radioactive half-lives. The calculation of radiation doses to body organs due to inhalation of radioactive aerosols would,with present models (see Ref. [3] ), require knowledge of the nuclides, their abundance, their decay properties, their metabolic properties, the atmospheric dilution and the particle size distribu tion of the aerosols inhaled. This is a task which, to our knowledge, has not yet been carried out (1973). In order to obtain an estim ate of the radi ation hazard due to the inhalation of a nuclide m ixture, we have chosen to com pare the results obtained when changing the relative fraction of 420 FRITZE and HERRMANN DCL FIG.2. DCL plotted against half-life. nuclides in a mixture at its derived concentration limit by using the summa tion form ula recommended by ICRP in the following form: where С is the derived concentration limit for the m ixture, К the reference concentration lim it, c¡ the relative activity of the ith nuclide, and k¡ a factor by which the DCL of the ith nuclide differs from К (k ^ 1). As an illustration, we shall assum e that K= 10*i"Ci/m 3, that for ^ the DCL is К and that c¡ =f"ci. The condition Ec¡f" = 1 then exists, from which f can be calculated for a given c^ For Ci= 0.001, 0.01 and 0.1 the DCLs and the concentrations for each group are shown in Fig. 1. As was to be expected, the DCLs differ considerably, in this case by about a factor of ten. It is not practicable to carry out a continuous measurement of each individual nuclide carried by an aerosol; it would, however, be required if their distribution was to be determined according to the DCL, from which in turn the DCL for the mixture could be obtained. The DCL of a nuclide depends on the physical half-life, as does the counting rate also when m easuring the activity during sampling or afterwards. This is qualitatively used when the released activity is determined by m easuring the activity on the the filter after some period of sampling or even after some period of decay, thus giving lessw eightingto nuclides with short half-lives. The m ajority of the nuclides have DCLs below 10"^ Ci/m^, and those with half-lives of less than ten days below lO'^Ci/m^ (except iodine isotopes). IAEA-SM-180/30 421 A plot of the DCLs versus physical half-life (Fig. 2) reveals that all beta emitting nuclides of interest bar 9°Sr can be found above a limit line. By knowing the half-life of the m easured activity, one could determine the lowest DCL for nuclides with these half-lives. When m easuring the activity of a nuclide mixture during decay, one can determine only the effective physical half-lives at certain tim es. For the distribution of nuclides that are of interest here, and according to their half-lives, the effective physical half-life at certain tim es could be used to determine the corresponding minimum DCL. It would be thus possible to find correction factors from a decay curve for the activities measured at these tim es. A correction of the activity at the beginning of the decay due to the decrease of short-lived nuclides during sampling is not required; this follows from the previous discussion. A method based on sim ilar considerations is described in Ref. [4]. In order to avoid determination of effective half-lives, Bindewald [5] has proposed that the activities be m easured at different times on one or more sampling filters after collection, that an appropriately chosen factor be applied to each measurement and that the values obtained be added, which leads to a weighted concentration, K: where A^ is the activity of the kth measurement, the kth volume of air sampled, t^ the corresponding sampling tim es, 7^ the decay tim es, and a^the weighting factors. The param eters in this formula, in particular a^, can be selected in such a way that the maximum of the DCLs, according to the curve in Fig. 2, is taken into account. Any residual activity on the filter would have to be interpreted as 9°Sr, unless it is analysed for. The above considerations were based on the radiation hazard due to inhalation. A m ajor part of the longer-lived nuclides, in particular 9°Sr, 89g^ i37Qg^ released into the atmosphere will fall-out and in some part enter the food chain. Intake lim its have already been suggested by the Federal Radiation Council in the USA (see Ref. [6] ). The radiation hazard due to ingestion has not been taken into account in the Federal Republic of Germany so far when setting discharge lim its for aerosol activity. A sim ilar approach as is used for iodine-131, namely the reduction of the activity concentration in the air in the environ ment by a certain factor, as is applied in regulations in other countries, is being discussed in the Federal Republic of Germany. 6. SUMMARY AND CONCLUSIONS Measurement of the aerosols in air discharged from nuclear power stations should be carried out on a continuous basis. The results obtained would serve mainly for plant surveillance. Since the m ajority of the nuclides are gamma em itters, a gamma sensitive detector could be used. This would have the advantage that the background due to natural activity would not interfere and that the equipment could be set up so that 'bursts' of noble gases would not simulate an increased release of aerosol activity. The alarm level of the monitor installed at PWRs should be based on a DCL of 422 FRITZE and HERRMANN 10'9 Ci/m.3. At BWRs the alarm level could take into account nuclides with higher DCLs. A satisfactory accounting of the release of longer-lived nuclides can only be accomplished if a larger volume is sampled, say, over a periof of one week. The nuclides collected on the filter could be m easured by beta or gamma detection — in the latter case possibly by gamma spectrometry. In order to make an account for ^°Sr, the filters for a certain extended period (e. g. a quarter of a year) should be pooled and analysed for this n u c lid e . ACKNOWLEDGEMENT We are indebted to Kernkraftwerk Lingen GmbH and Kernkraftwerk Stade GmbH which gave us perm ission to publish the activity data. REFERENCES [1] MORGAN, K .Z ., "Health physics measures to implement new USAEC regulations relating to radiation exposure of the general public", European Congress on Radiation Protection, Budapest, May 1912. [2] SCHWIBACH, J ., Strahlenschutzrichtwerte für die Genehmigung der Ableitung radioaktiver Stoffe, "atw" 5 (1973) 280. [3] Inhalation Risks from Radioactive Contaminants, IAEA Technical Reports Series No. 142, IAEA, Vienna, 1973. [4] VOHRA, K.G., NAIR, P.V.N., SINGH, A.N., "New concepts in air monitoring and control of inhalation dose in nuclear operations", Assessment of Airborne Radioactivity (Proc.Symp. Vienna, 1967), IAEA, Vienna (1967) 23. [5] BINDEWALD, H .. private communication. [6] MORGAN, K .Z., STRUXNESS, E.G., "Criteria for the control of radioactive effluents". Environmental Aspects of Nuclear Power Stations (Proc. Symp. New York, 1970), IAEA, Vienna (1971) 211. DISCUSSION R. BÖDEGE: You said the loss factor on the Stade monitoring system was 0.4. Does this mean that, if 100% of the activity enters the tube, then 40% will leave it? G. HERRMANN: Yes, that is correct. R. BÜDEGE: We measured the loss factors at Lingen four years ago and found them to be of the sam e order of magnitude, i. e. 0.4 to 0.5, for a tube 50 m long and of 1.25 inches diam eter. G. HERRMANN: But that was just for iodine, wasn't it? R. BÖDEGE: No, it was for iodine and for aerosols. W. STEPHAN: If you carry out continuous beta measurement of the air released from the stack of a pressurized water reactor, then, apart from noble gas activity, you will also be m easuring nätural aerosol activity which is higher by a factor of 10 or more than the artificial aerosol activity. The artificial aerosol activity released from a pressurized water reactor (predominantly and 60Co) is so low that continuous gamma measurement would appear to be unnecessary, as people living in the area can receive absolutely no dose from these releases. IAEA-SM-180/30 423 G. HERRMANN: As I have said, measurement of aerosol releases serves two purposes, the first being plant surveillance, for which con tinuous monitoring is required. If this is done with gamma detectors, back ground from natural activity can be avoided. The other purpose is to record the irradiation dose to the environment which is mainly due to the long-lived nuclides and, for balanced results, the aerosols should be collected from a large air volume, as is done at Obrigheim. J. R. BEATTIE: I noticed from one of your slides that you have found antimony-124 in the gaseous effluent of the Stade plant. Is this a fission product or an activation product, and if the latter, what is the source of the antimony-124 in the reactor? R. FRITZE: We think that the antimony nuclides derive from the neutron sources installed in the core of the Stade plant. H. EDELHAUSER: I should like to emphasize that the perm issible whole-body dose limit of 30 mrem mentioned by Mr. Herrmann is actually applied to critical groups in the neighbourhood of the plant. Furtherm ore, for the purpose of licensing nuclear power plants, where very sm all dis charge rates can be expected, the value of 30 m rem /a for radioactive m aterial discharged into the air is applied more conservatively to the most unfavourable point outside the boundary of the plant. IAEA-SM-180/60 APPLICATION OF SYSTEM ANALYSIS METHODOLOGY TO THE DETERMINATION OF THE LIMITING RADIOLOGICAL CAPACITY OF THE AREA SURROUNDING A NUCLEAR FACILITY L. FRITTELLI Comitato Nazionale per 1'Energía Nucleate, CSN Casaccia, Rome, Italy Abstract APPLICATION OF SYSTEM ANALYSIS METHODOLOGY TO THE DETERMINATION OF THE LIMITING RADIOLOGICAL CAPACITY OF THE AREA SURROUNDING A NUCLEAR FACILITY. of each radionuclide and each exposure pathway to the dose received by the public in an area surrounding a nuclear facility owing to the release of radionuclides into the atmosphere and into surface waters. This makes it possible to identify the 'criticalradionuclides', the'critical pathways' and the 'critical population group', which are then used in planning a radiological monitoring programme in the environs of the nuclear installation. The rate constants for the transfer of radionuclides between the various components of the system are calculated from the literature data or from data on the agricultural and meteorological characteristics of the site and the 1. INTRODUCTION Environmental System Analysis is a methodology which offers con siderable prom ise as an approach for predicting exposure dose from radio nuclides that might be transferred to man through a variety of environmental pathways [ 1 - 3]. The dynamic behaviour of radionuclides in the environment along the various exposure pathways can be modelled by means of a system of coupled compartments through which the radionuclide movement can be predicted on the basis of mathematical equations. In this paper a compartimentai model is set up for simulating the environ mental behaviour of the low-level liquid w astes released from a nuclear research centre into the surrounding environment. The model predictions are used for evaluating the Environmental Radiological Capacity (ERC) of the receiving environment by comparing the contribution of each radionuclide in each exposure pathway to the total dose received by the individuals living in the area surrounding the centre. The method of using the m odel's predictions about the 'critical' exposure pathways for developing and operating a program m e of environmental monitoring is discussed. 2. THE CONCEPT OF ENVIRONMENTAL RADIOLOGICAL CAPACITY In this paper the ERC for a radionuclide of a given sector of the receiving environment (e.g. a watercourse, a lake, the atmosphere) is defined as: "the annual input, for given release conditions, of the radionuclide to the 425 426 FRITTELLI given sector of the local receiving environment which will result in a dose commitment (DC), via the critical exposure pathways, equal to the recom mended dose lim it (DL) for individual m em bers of the public." The DC concept is very useful in assessing the consequences of environ mental contamination caused by a nuclear plant per 'unit practice', i.e. on e year of operation, when a significant contribution to the total dose received by the m em bers of the public is caused by long-lived nuclides. The DC concept is also useful for estim ating the trend in an environmental situation before the equilibrium conditions are reached. In such a case the predicted annual dose in the future equilibrium situation is num erically equal [4] to the DC per year of practice leading to the expected equilibrium. In routine equilibrium conditions the annual dose incurred by groups of individuals in the population is a convenient and satisfactory m easure of the situation [ 5]. Thus future annual doses from environmental contamination by long-lived m aterials exceeding the recommended D Ls can be prevented by applying the annual D Ls not to the actual annual doses but to the annual dose com m itm e n t, i.e. to the DC per unit practice. In this context the DC can be defined as the infinite time integral of the average dose rate (internal plus external) to a given population group, i.e: (1) о The population groups for which the DC is calculated may be chosen to be composed of individuals with sim ilar irradiation characteristics, e.g. infants drinking cow milk from a contaminated pasture, people living down wind at some distance from the plant or farm ers working in contaminated f ie l d s . Obviously, if the DC is defined as in Eq. (1), the population group over which the dose rate is averaged does not consist of the sam e individuals: the defined DC gives the average dose accumulated over all the future years in the group as a group rather than as a selection of individuals. It seem s worthwhile to point out that in practice the discharge of radioactive w astes from the plant into the surrounding environment is limited and controlled on the basis not of the ERC, but of the so-called Stipulated Environmental Radiological Capacity (SERC), that is the fraction of the ERC authorized for utilization by the competent national authority. As is pointed out elsewhere [ 6], the SERC is based on the principle of eliminating not only the unacceptable risks, i.e. those deriving from exceeded DLs, but also of the undue risks, that is those not justified by actual needs. 3. COMPARTMENTAL MODELLING OF THE RECEIVING ENVIRONMENT The system "environment plus man" can be subdivided into many multi-compartmented subsystem s, the number of which is proportional to the degree to which the interaction m echanism s among the components are known. Generally the system components are not simply connected, i.e. the radioactive m aterial can diffuse or turn over from one component to m ore than two components. IAEA-SM-180/60 427 The compartments need not be spatial regions, but they must be distinguishable on some basis, e.g. different species of plants or animals, successive tracts of a watercourse, chemical phases, etc. Ideally the compartments are assum ed to be homogeneous and uniform with no internal gradients in their volumes (m^, m^, kg). If, however, a component cannot be considered to be homogeneous, it can be assum ed to have an em pirical retention function (e.g. a sum of power functions) and be included in the system alongside all the homogeneous compartments. The transfer of radioactivity between the various system components is caused by various processes, i.e. chemical processes (fixation of elements in soil or in river sediments), by m ass transfer (cow eating contaminated grass or the transfer of radioactivity from a watercourse to edible crops or forage by means of irrigation water), by biological turnover (the milk being extracted from the cow twice a day), by an external driving force (e.g. deposition on the ground of airborne m aterial due to wind, rain and gravity), by transport within an environmental medium (the movement of radioactivity downstream within a watercourse or the diffusion in marine coastal waters). Although non-linear and time-varying transfer functions are the rule in nature, it is in practice very useful to assum e that the dynamical behaviour of radionuclides in the environment can be modelled by means of linear, tim e-invariant transfer functions, that is the dynamics of the model can be expressed by a system of first-order linear equations with constant coefficients. The linear assumption is often a fairly good approxi mation to real system s because the radioactive substances released to the environment in the w astes are often m ass-less and it can be assum ed that the transfer mechanisms are based on first-order reactions, i.e. th e r a t e at which the radioactive m aterial is transferred from one component to the adjacent ones is proportional to the amount in the compartment itself or, if the distribution volume is constant in that time scale, to the concen tration in the component. Several different models of varying complexity may describe the physical system equally well for different purposes. Setting up a suitable model requires some knowledge of the system , analytical judgement and intuitive insight. The manner in which the various system components are "lumped" into a compartmental model is often partially dictated by the kinds of obser vations one can make and the quality of available data. Thus, if the information on a system is incomplete or of questionable accuracy, a highly detailed model is probably not warranted. Very elaborate models, though desirable and necessary for quantitative predictions of the behaviour with time of radioactive contamination along the different environmental pathways and of the resulting human exposure, can be justified only when accompanied by accurate transfer rates that have been carefully verified. Indeed the accuracy of the model predictions does not necessarily increase when a model is made more complex: the inclusion of additional param eters can sometimes result in less accurate predictions because each param eter has a measurement error associated with it. 4. MATHEMATICAL FORMULATION OF THE MODEL If the receiving 'local environment' can be modelled by means of a system of n homogeneous components, in a simple linear representation, 428 FRITTELU the Laplace transform s of the concentrations of a radionuclide in the ith component of the system owing its entry into an environmental sector can be expressed, if the distribution volumes are time-invariant, by means of the equations: c (p ) = V 'l ((Р + ДЦ - a ) * * (Vc(0) +_f(p)) (2) in the m atrix notation explained in Footnote 1. The time behaviour of the concentration C¡(t) in the component can be calculated by inverting the Eq. (2). The inversion of c¡(p) is straight forward but time-consuming, because we must solve the n degree equation: det ((p + X)I_- a) = 0 and find the partial fraction expansion of c^ (p). The knowledge of C¡ (t) is very useful in setting up and operating a program m e of environmental monitoring by choosing the sampling locations and the sampling and analysis frequencies of environmental sam ples. Indeed the radioactive contamination along the exposure pathways canbe satisfactorily controlled with reduced effort by monitoring in the compartment(s) where the radionuclide concentration varies more slowly. The exposed individual, whose exposure begins at age T, can be included in the model as a subsystem consisting of several compartments, not neces sarily homogeneous. The total whole-body or organ dose received over the time t equals the sum of the contributions the internal irradiation due to the uptake of the radionuclide from the various environmental compartments and its deposition in the organ at risk and of the external irradiation by the contaminated environment. It can easily be shown that the DC(T) in rem can numerically be c o m p u te d a s [ 1 ] : DC(T)=^(e^DCJT)+b.)cJO) (3) 1 = 1 where DC^ (T) is the DC to the whole body or to the organ by the intake in the body of unit activity (^Ci) at time zero at age T. It has been assumed ^ Symbols used in E q.(l): c(p) = n-row column vector of the Laplace transforms of theconcentrationsC(t) of radioactivity in the compartments of the system. f(p) = n-row column vector of the LapLace transforms of the alimentation functions in the \ = Radioactive decay constant. ((p + \)I_- a)"^ = inverse matrix of ((p + \)I - a). D(p + \) = determinant of ((p + \)I * a). Dij(P + M = ith row and jth cofactor of ((p + X)I_- a). c(0) = lim it of c(p ) as p tends to zero. IAEA-SM-180/60 429 1 river water E x te r n a ! 2 fish Irradiation 3 irrigated surface soil 4 horticultural crops 5 (orage 6 subsurface soil 7 cow gut 8 body fluid umbounded 9 lower large intestine 10 reservoir thyroid 11 body fluid bounded 12 kidney 13 udder milk that the functions e¡(t) and b¡(t) relating the internal exposure and the external exposure to the concentration of the radionuclide in the ith compartment are time-invariant. Thus, if we assum e that the radionuclide is released only into compartment 1 (e.g. 'surface w aters'), the ERC of the receiving sector 'surface w aters' can be calculated by putting DC = DL in Eq. (3), that is: (4) if the e¡, b¡ and a¡j are assum ed to be average values over the exposed population groups. 5. APPLICATION OF THE MODEL Figures 1 and 2 summarize the compartmental models for estimating the time behaviour along the different exposure pathways of 13li I37çg ^ d 9°Sr released in the low-level liquid wastes from a nuclear reasearch centre into a w atercourse. There is no edible vegetation on the river bed and it has been assum ed that, downstream of the input point of the w astes, the interactions between the watercourse and the aquifer can be ignored as far as the present application is concerned. The water from the river is used 430 FRITTELU 1 river water 2 fish 3 irrigated surface toi) 4 horticuitura) crops 5 forage 6 subsurface soi) 7 cow gut 8b)ood 9 tower large intestine 10 reservoir soft tissue 11 reservoir meat 12 kidney 13 udder mi!h 14 reservoir bone fSr) a red marrow b bone FIG.2. Compartmental mode! for the environmental behaviour of strontium and caesium released to surface waters. to irrigate horticultural crops and forage fields, such forage being used to feed cows in milk whose drinking water is drawn from the river itself. The exposure pathways in the model are: (a) Ingestion of contaminated fish. Since accurate values of the transfer coefficients in the aquatic system are not available at present, it has been assum ed that the concentration in the fish is proportional to the concentration in the water, the constant of proportionality being a Concentration Factor, F C [7]. (b) Ingestion of contaminated horticultural crops. The feedback loop through the soil to the crops or forage compartment models the experimental results. The shallow soil and subsurface soil compartments account for the fraction of radioactivity in the soil that is not available for uptake from the plants. The values of the transfer coefficients are derived from Ref. [ 7], with some modifications accounting for the characteristics of the local environment. (c) Ingestion of contaminated cow products. The model for the cow is the one derived from Ref. [ 8]. The model for iodine includes the body fluids compartment for accounting for the iodine that, because of its chemical form , is not transferred to the milk. The single reservoir is the thyroid gland. The cow model for caesium and strontium have more than one reservoir, i.e. meat, bone and soft tissue (brain, liver, spleen) for strontium and meat and soft tissue for caesium . The transfer rate constants in the model IAEA-SM-180/60 431 TABLE I. VALUES OF THE ERC (Ci/a) OF THE RECEIVING SECTOR "SURFACE WATERS" FOR THE "HYPOTHETICAL" CRITICAL GROUP Radionuclide "'1 '"C s '°Sr Organ thyroid whole body bone marrow bone Adults 1.4 x 10" 3 .3 x 10° 2 .1 x 10* 3.1 x 10* Children 1.1 x 10* 3 .3 x 10° 1 .3 x 10* 1.9 x 10* are derived from Ref. [9], with some modifications accounting for some typical conditions of the site, e.g. dry weight density of the forage in the field, daily milk production of the cow, feeding practices in the farm s. It has also been assum ed that the cows are not continuously grazing on the forage fields but that the forage is harvested and supplied to them some time after irrigation. (d) External irradiation from contaminated soil. The soil has been con taminated by irrigation water. Other pathways that were investigated and found to be unimportant are (i) the external irradiation from the contaminated water in the river bed, (ii) the inhalation of radioactive substances re-suspended from the soil, and (iii) the external irradiation (or inhalation of) radioactive substances from contaminated fishing or farming equipment. The exposed groups of the population consist of the people: (A) Eating the fish caught in the river; (B) Eating the horticultural crops produced in the farm s along the river; (C) Eating the cow products (Ci - milk, C 2 - m e a t , C 3 - soft tissue) from the farm s along the river; (D) Working or living on soil contaminated by irrigation water. Some inquiries about the habits of the people living in the area sur rounding the plant made it possible to conclude that only the population g r o u p s B , C i and D, A and D, C 2 an d C 3 can in practice overlap and be the same individual who is open to more than one exposure pathway. For example the group (A, D) consists of some hobby fishermen, their fam ilies and friends, while the group (B, C i, D) consists of,the farm ers living along the river. Indeed the cows are not slaughtered on the farm s, and their meat is not consumed by the individuals of the group (B, C^, D). However, for sake of safety and to account for future changes in the habits of the population, the values of the ERC of the receiving sector "surface waters" for ^1, -^C s and s°Sr reported in Table I have been calculated for a "hypothetical" critical group consisting of individuals open to all the exposure pathways discussed. The values of DC^ for the adults and the children are those proposed b y B r e u e r [10]. 432 FRITTELLI REFERENCES [1] FRITTELLI, L., 2nd European Congress on Radiation Protection, Budapest, 3-5 May 1972. [2 ] BRANCA, G ., BREUER, F ., CIGNA, A .A ., AMAVIS, R ., Euratom Rep. EUR-4636e(1971), [3 ] BOOTH, R .S ., KAYE, S .V ., ROHWER, P .S ., 3rd National Symp. Radioecology, 10 -1 2 May 1971, OakRidge, Tenn., USAEC Rep. CONF-71501 (1971). [4 ] UNITED NATIONS SCIENTIFIC COMMITTEE ON THE EFFECTS OF ATOMIC RADIATIONS (UNSCEAR), Ionizing Radiations: Levels and Effects, UN, New York (1971). [5] INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION (ICRP), Publication N o.9 (1965). [6] FRITTELLI, L., these Proceedings, paper IAEA-SM-180/59. [7 ] BRANCA, G ., BREUER, F ., CIGNA, A .A .. AMAVIS. R ., Euratom Rep. EUR-4897e(1973). [81 GARNER, R .J., Health Phys. 13 (1967) 205. [9 ] BOOTH, R .S ., BURKE, O .W ., KAYE, S .V ., Nuclear Methods in Environmental Research, Topical MtgAm. Nucl. Soc. Aug. 1971, Univ. Missouri, Columbia, USAEC Rep. CONF-710818 (1971). [10] BREUER, F., personal communication, 1973. IAEA-SM-180/10 RADIOACTIVE GAS AND AEROSOL PRODUCTION BY THE CERN HIGH ENERGY ACCELERATORS AND EVALUATION OF THEIR INFLUENCES ON ENVIRONMENTAL PROBLEMS A. PEETERMANS, J. BAARLI CERN, Geneva Abstract RADIOACTIVE GAS AND AEROSOL PRODUCTION BY THE CERN HIGH ENERGY ACCELERATORS AND EVALUATION OF THEIR INFLUENCES ON ENVIRONMENTAL PROBLEMS. in the atmospheric air is the main process to form*'A, ^CI, ^Cl, S's, ^ p, s'si, 24^ for the CERN improved synchro-cyclotron machine are presented. The implications of the release of radioactive gases and aerosols from this high energy accelerator installation into the environment are discussed, taking into account the available local meteorological data. 1. INTRODUCTION Since the early years of the CERN high energy accelerator operation the problem s of the radioactivity released to the CERN site and its environment has been investigated [1-5]. Outside the CERN boundaries, according to the Sw issi and the French^ radiation safety regulations, the radiation level due to CERN operation should not exceed 5 rem in 30 years. The stray radiation dose rates, produced mainly by the two high energy accelerators (a 600 MeV synchro-cyclotron (SC) and a 28 GeV proton synchrotron (PS)) as well as the Intersecting Storage Rings (ISR) are continuously m easured at thirteen places on the site. Radioactivity in air, precipitation and effluent water has been continuously monitored s in c e 1969. The location of the different monitoring stations can be seen f r o m Fig.l. Although the measured integrated doses beyond the CERN boundaries are well below the 5 rem /30 years limit, only a sm all frac tion is due to the release of radioactive gases and aerosols. In view of the planned ten-fold increase in beam currents of the CERN accelerators [6] and the construction of the 400 GeV SPS the problem of radioactive gases and effluents has been reconsidered more extensi v e ly [7,8]. The present paper mainly deals with the radioactive pollution problem s connected with the new 600 MeV SC facilities. While at present ' Ordonnance du Conseil Federal concernant la protection contre les radiations (19 avril 1963) rendue applicable au Canton de Genève par Règlement d'exécution ( 1^ octobre 1965). ^ Décret général du 11 décembre 1963 sur les installations nucléaires et Décrets N"66.450, 67.228 et Arrêtés du 18 au 24 avril 1968 sur la Protection du Travailleur contre les Rayonnements Ionisants. 433 4ь ETRAS n BAARU andPEETERMANS FIG. 1. General plan of the CERN facilities. IAEA-SM-180/10 435 Ep(MeV) FIG. 2. Excitation curves for the production of ^O , ^N , ^ C , ^Be, ^O . the PS is the main source of stray radiation, the SC will be expected to contribute mostly by the release of radioactive gases and effluents. Nevertheless the main results apply to high energy accelerators in g e n e r a l. 2. GENERALITIES Non-elastic interactions of high energy nucleons with complex nuclei are generally described by the two stage Serber model [9,10]: during the intranuclear cascade relatively high-energy nucleons and pions have a good probability of escaping from the nucleus, mainly in the forward direction. At the end of this cascade the nucleus left will de-excite by the em ission of particles, or fission may occur. As a consequence, when a high energy particle beam strikes a target element of atomic m ass At, there are well defined probabilities for a production of isotopes, radio active or not, with A less or equal to At- 436 PEETERMANS and BAARU FIG.3. Excitation curves for the production of "C, 'Be, "N , *°C. The specific activity of the radioisotope j (in pCi/m ^) produced in the atm ospheric air of a ventilated experimental area can be calculated by the following expression: ^ o - Aj 1 - e x p ( j b j + ^ ) t ) — where Xj is the decay constant of the radioactive isotope j produced, D the ventilation rate, V the volume of the room (or experimental area), an d <2 ij is given by: = У К E withKbeingaconstant, ф them eanhighenergyparticlefluxdensity(cm '^ *s*^) N¡ the number of parent atoms concerned, and?¡j the mean production cross-section of radioisotope j in the nuclear process with nucleus i. IAEA-SM-180/10 437 We suppose that the atmospheric air is composed of N 2 (75.7% by weight), Од (23%) and 4°A (1.3%). Besides the high energy particles, therm al neutrons have to be considered. This is of particular importance for the production of 41Д. The main radioactive nuclides formed by spallation and therm al neutron capture in the atmospheric air are listed in Table I. The excitation curves for the production of ^Be, 1°C, *^C, 13N, 14Q and 15Q are represented in Figs 2 and 3. More details con cerning the available experim ental data and calculated values are given in Ref.[11]. The 40A(p, )X nuclear reaction cross-sections have been calculated using the Silberberg-Tsao sem i-em pirical formula [12], which for low-Z targets seem s to give better results than the Rudstam formula [13]. For light nuclei such as nitrogen and oxygen, which have equal numbers of protons and neutrons, it is expected that, apart from Coulomb effects, the cross-sections for neutrons and protons will be the same because of charge symmetry. We will suppose that this also is valid for argon. Table I also contains values of the maximum perm issible concentra tions in air for the population at large as reported in the relevant "Ordonnance Suisse" or "Décret Français". If (МРС)д values are not given they have been calculated following the ICRP recommendations. Secondary particles (protons, neutrons, pions, a, y, etc.) are also produced in high-energy non-elastic nuclear interactions. The knowledge of their number, their energies and spatial distribution is of great importance in 'air activation' calculations. This information can be obtained from the Monte Carlo intranuclear cascade-evaporation calcula tions (see for example Refs [14, 15]). Some experimental data are available from heavy emulsion studies. The average number of cascade nucleons emitted per inelastic interaction increases with increasing particle energy. The ratio of neutrons to protons emitted during the cascade is independent of the bombarding energy but increases with the neutron excess of the target nuclei. For light nuclei about equal numbers of neutrons and protons are emitted, whereas for heavy targets twice as many neutrons as protons are observed [16]. Nuclei emitted in the evapora tion process are expected to have relatively low energies (a few MeV). Table II gives some experimental and theoretical values of the average number of neutrons emitted per inelastic event for incident protons on various elements [15]. The total inelastic cross-section for high energy protons and neutrons (E > 150 MeV) on a target m aterial of atomic m ass At (A¡ > 3) can be calculated by the expression [17]: °ln el. (mb)=43.1At*^ In cyclotrons and synchrotrons operating with an internal target the calculation of the yield of secondary particles requires the knowledge of target efficiency defined as the fraction of protons undergoing nuclear interaction in the target. In the particular case of the CERN improved SC a maximum target efficiency of 48% has been calculated for a 10 cm thick internal beryllium target [18]. 3. EXPERIMENTAL AND CALCULATED RESULTS A great number of the radioactive isotopes produced in atmospheric air by high energy particles have been identified by different authors 43 8 ETRAS n BAARL1 and PEETERMANS TABLE I. MEAN RADIONUCLIDES PRODUCED IN ATMOSPHERIC AIR Type and decay energy Radionuclide H a l f - l i f e (MeV) and (%) element for 600 MeV protons ( p C l / m ^ ) ( m b ) 12. 26 a B" (0.018 - 100%) N 3 5 0. 2 ?" О 4 0 A ¡ B e 5 3 . 6 d EC N 8 , 5 4 x 1 0 ' ^ y(0.48 - 10%) 0 8,2 A 0 , 7 2 . 5 x 1 0 * 2 a " г 19.4 s 6+(l. 9 - 99%) N 1 . 8 у (0.72 - 98%) 0 2 . 6 A < 0 . 1 6 x l 0 ' 2 a " c 2 0 . 3 m i n B+(0. 97 - 100%) N 1 4 0 1 2 A < 1 10.0 min 8+(l. 19 - 100%) N 5 . 5 5 X 1 0 ' 2 ^ О 2 . 5 A < 1 " o 7 1 . 0 s 6+(1.81 - 99?.) N - 1 . 5 X 1 0 * 3 3 у (2.31 - 99%) 0 1 . 4 A < 0 . 1 ^0 2. 03 min 8+(1.74 - 100%) N - 4 X 1 0 * 2 3 0 3 5 A <0.1 + (0 . 6 3 5 - 9 7 % ) 1 . 1 1 0 9 . 7 m i n A 9 X 1 0 * 2 E C ( 3 % ) IAEA-SM-180/10 439 ^ х м * " ^ 5Х 10*з 3 X 1 0 *' 4X10*2 ^ 4x10*2 1.3X 10*2 а 4X 10"* ^ 2 4х 10*2 а 3X10*2 . 8 . 8 . 8 1 0.2 1.8 0 6 6.8 4 .6 А А А 2 .5 А А А 3.2 А А .) .) .) 1007 1007 (1.1 - 87.) (0. 880 - 8?.) (0.54 - 90%) (1. 59 - 427.) (2. 754 - 1007.) (2.43 - 157.) (0.950 - 297.) (0.391 - 317.) (1. 013 - 307.) (1. 4 - 30%) * (1.98 - 92 1 1 1 AI ! 3 Í HMg 15^ ílN a HMg HA1 HSi p 30 values were calculated in accordance with (Publ. 1CRP No. 2) recommendations. The (MPC)A values of these elements are not reported in the Ordonnance Suisse or Décret Français concerning radiation protection values. The given ! 1 1 1 T A B L E I. (cont.) cross-sections Radionuclide H a l f - l i f e (MeV) and (%) e l e m e n t for 600 M e V protons ( n C i / m ^ ) ( m b ) 32 p 1 4 . 3 d 6'(1.71-100%) А 1 2 . 8 2 x l 0 ' 3 1 5 ^ 33 P 2 5 d ß* (0.249 - 100%) А 12. 0 6 х 10 *за 1 5 ^ 9 x 1 0 * 3 16 s 8 8 d 3" (0. 167 - 100%) А 16. 8 37 с 5. 06 min 6'(4.3-10%) А 1 . 5 4 x 1 0 * 2 a 16 ^ (1.6 - 90%) у (3. 09 - 90%) 38 с 2 . 8 7 h 6 " ( 3 . 0 - 5 % ) А 0 . 2 4 X 1 0 " ' " BAARLI andPEETERMANS 16 ^ (1.1 - 95%) у (1. 88 - 95%) 32 min 8"*"(4.5-47%) А 1 . 6 ( ? ) (1.3 - 26%) у (2.13 - 100%) (3. 3 - 32%) 3 7 . 3 m i n В "(4.8-53%) А 2 1 . 3 7 x 1 0 * 2 1 7 ^ (2. 77 - 16%) (1.11 - 31%) у (1.64 - 38%) ( 2 . 1 7 - 4 7 % ) 39 (-1 5 5 . 5 m i n В* ( 3 . 4 5 - 7 % ) А 2 0 . 9 3 x 1 0 * 2 a 1 7 ^ (2.18 - 8%) (1. 91 - 85%) у ( 1 . 2 7 - 1 0 0 % ) ( 1 . 5 2 - 8 5 % ) A 8*(1.198 - 99%) А 600 (thermal 4 x 1 0 * 2 18 ^ 1 . 8 3 h у (1.293 - 99?°) lAEA-SM-180/10 441 TABLE II. EXPERIMENTAL" AND THEORETICAL AVERAGE NUMBER OF NEUTRONS EMITTED PER INELASTIC EVENT FOR 660 MeV PROTONS ON VARIOUS ELEMENTS (H.W. B E R T I N I et al., O R N L - T M - 3 1 4 8 ) Theoretical T a r g e t Experimental C a s c a d e T o t a l С 1 . 0 5 0 . 5 7 1. 6 2 1. 5 ± 0 . 2 A l 1 . 4 5 0 . 9 6 2. 4 1 2 . 8 ± 0 . 3 C u 2 . 1 3 . 3 5 . 4 4 . 4 ± 0. 4 P b 3 . 1 1 2 . 1 1 5 . 2 1 1 . 9 i 1 . 0 и 3 . 3 1 4 . 5 17. 8 16. 8 ± 1.2 Vasilkov, R. G . , et al., Sov. J. Nucl. Phys. 7 (1968) 64. [1-3, 5, 8, 19, 20]. The qualitative and quantitative analysis of the radio activity in air is important since the (МРС)д of an u n k n o w n mixture of radioactive isotopes is as low as 10*2 pCi/m^ [21]. For the measurement of high specific activities (pCi/m3) w e use air flow g a m m a - c o m p e n s a t e d ionization chambers [22]. For identification and evaluation of very low activities the n o r m a l air sampling technique on filters (Microsorban 99/97) combined with low-level beta and g a m m a counting is applied. The g a m m a spectroscopy is carried out with a 55 cm3 high resolution Ge(Li) detector. Identification of pure beta emitters is based on the well known least- squares numerical analysis of the decay curve. The most important radioactive isotopes identified near high energy accelerators are ^C, 13N, 1SQ artet 41A which, with the exception of argon-41, are all relatively short-lived positron emitters. A list of radioactive isotopes detected in the ventilation air of the SC-Isolde experimental area are summarized in Table III. Isolde (Isotope Separator On-Line) is an extracted proton beam underground experimental facility [23]. M e a s u r e d and calculated values of specific activities in air for a particular experimental condition in Isolde are presented in Table III. The experimental values are the results of measurements over a period of 100 days. The filter collecting efficiency was assumed to be 100% although the physical size of the produced effluents when collected is not known. The agreement between experiment and theory is good. During the summer of 1974 the C E R N SC machine will be recon structed, with an internal b e a m intensity increased to 10 ^A. This tenfold increase involves among other things important shielding modifications. The ventilation system of the Isolde b e a m line is modified in order to minimize the release of radioactive gases and effluents to the environ ment [8]. The chimney will have an increased height, reaching about 25 m above the ground. A s in the past, absolute filters will provide added safety. The post SCIP (SC Improvement Programme) calculated maximum specific activities in air are s u m m a r i z e d in Table III. F o r the Isolde area the n e w molten metal target facilities have also been considered [24]. 442 TABLE III. MEASURED AND CALCULATED ACTIVITIES IN THE VENTILATION ATMOSPHERIC AIR OF THE SC-ISOLDE COMPLEX (before June 1973) Radionuclide I s o l d e ( H C i /гцЗ) ( C C i / m ' ) ( ^ C i /щ З ) ( ^ C i / m 3 ) ( C i / a ) not measured 1 . 6 x 1 0 * s 2 . 6 x 10 " 4 1.1.x 10"* 9.1 x 10* 3 ETRAS n BAARLI andPEETERMANS ? B e 2 . 7 x 1 0 " * 3 . 2 x 1 0 " * 5.1 x 10*3 2.2 x 10*3 1 . 8 x 10*2 ' ° C 6 . 8 1 . 6 2 . 4 1 . 3 x 1 0 2 total specific "c 1 . 9 2 1 . 5 7 . 5 6 . 8 x 102 a c t i v i t y "N including all 1 . 4 1 1 . 4 3.7 3.5 x 10" other radioactive N i s o t o p e s " 0 1 8 ± 3 0 . 5 0 . 6 0 . 3 2 3 ' S O . - 8 . 4 1 8 . 8 8 . 2 6 . 7 x 102 18 p ( a ) 1. 5 x 10"* 2.2 x 10*3 8. 9 x 10"* 7. 5 x 10*2 " N e (a) 6. 9 x 10"* 2. 6 x 10-3 9. 5 x 10*' 8. 5 x 10*2 ' ^ Na traces 1. 8 x 10*' 2. 8 x 10* ' 1.2 x 10*' 9. 9 x IO"' " N a 3.6 x 10"S 4. 8 x 10*s 7.6 x 10*" 8.3 x 10"* 2. 7 x 10*2 2 ' M g (b) 3 . 4 x 1 0 * 3 2.7 x 10*2 9.1 x 10*3 8. 5 x 10*' 2 ' M g 1.3 x 10*s 8. 6 x 10'" 1.4 x 10"* 5.9 x 10*5 4. 9 x 10"^ " A l ( b ) 3 . 2 x 1 0 * 2 8. 2 x I O * ' 3 . 4 x 1 0 * 2 2.8 " A l ( a ) 8 . 7 x 1 0 * 3 5 . 7 x 1 0 * 2 1. 8 x 10*2 1 . 7 " S i 1.8 x 10 " * 5 . 8 x 1 0 * " 8. 9 x IO*" 3.7 x 10*3 3.1x10'' Mp not identified 8.0 x 1 0 '" 2.2 x 10*2 9 ; 0 x 1 0 * 3 7 . 6 x 1 0 ' * Ир 6.0 x I O ' ' 8 . 4 x 10*6 1.3 x 10*4 5. 8 x 10*6 4 . 7 x 1 0 * 3 ЗЗр 2.0 x 10 *" 4. 5 x 10"' 7. 3 x 10*3 3 . 1 x 10*6 2 . 6 x l 0"3 Mg not identified 1.8 x IO*' 2. 9 x 10*6 1.2 x 10*6 1 . 0 x l 0 ' 3 0 1 / 0 8 1 - M S - A E A I " s ( b ) 3 . 6 x I O * " 1. 9 x I O * ' 6.6 x 10*3 6. 0 x 10 '* 3=S ( a ) 8. 0 x 10 *" 2. 5 x 10*з 9. 8 x 1 0 ' " 8 . 4 x 1 0 * 2 M ^ m ( b ) 6. 6 x 10 "' 8. 4 x 1 0 * 3 3 .1 x 10'3 2.7x10"' !Sci ( b ) 7.5 x IO*" 9. 8 x 10*2 3.7 x 10'2 3.2 ^C1 ( b ) 5 . 1 x I O ' " 7 . 1 x 1 0 * 2 2.7 x 10'2 2 . 4 ^ A not identified 1 . 5 x 1 0 *s 2.4 x 10*3 1.1 x 10'6 Э ^ х Ю " * "A ( b ) 2. 5 x 10"^ 0.7 0.2 2.0 total T < 30 min 1860 (98.6?.) total T > 30 m i n 2 6 443 444 PEETERMANS and BAARLI 1. BISE 4. VENT Greatest frequency 200°-250*. Bad weather sector (rain, snow, showers), occasionally poor visibility and low ceiling (800 - 1500 m /200 - 500 ft), especially when associated with snow. 5. Quick veering of winds by cold fronts passing by the station. 6. JORAN tAEA-SM-180/10 445 TABLE IV. SOME VALUES OF THE a, ß, y PARAMETERS (SEE FORMULA (3)) USED FOR THE CALCULATION OF THE DISPERSION PARAMETERS a AND Dispersion parameter Pasquill category a 6 У A - 1.907 1.15 - 0. 0157 У F - 3 . 0 3 4 0. 9 8 0 4 - 0. 0 0 4 6 3 A 1 0 . 1 7 9 - 3.644 0.4473 ° z F - 5. 0 1 4 3 1. 5 3 8 - 0. 0 6 2 5 The calculations (see results Table III, column 4) assume a 3 X 1013 p/s proton beam (50% extraction efficiency) on a 30 g thorium target. The b eam length in open air is not supposed to be greater than 2 m. The calculations for the S C hall of the air activity (column 5) suppose a 10 p A internal proton b e a m on a 10 c m thick beryllium target (target efficiency 48.4%) [18], the particles produced being spread out into a cone of half angle at half m a x i m u m of about 20° [25]. The production of 4iA was extrapolated from thermal neutron flux density measurements in the SC hall [26]. The m a x i m u m calculated release per year of radioactive isotopes into the atmospheric air due to the n e w C E R N 600 M e V S C are given in column 6 of Table III. These figures are based on an SC opera tion time of 80%, the Isolde facility being used 2 0 % of the b e a m time. T h e a m o u n t of activity in air due to radioisotopes of a lifetime less than 30 minutes is of special interest. 4. EVALUATION OF RADIOACTIVE AIR POLLUTION OUTSIDE THE CERN BOUNDARIES T h e C E R N site within the Canton of Geneva, wh i c h is a w e d g e in the southwest extremity of the Swiss Plateau, is located between the Jura mountains and the Alps. It follows from the orographical configuration of the surrounding country that winds are subjected to a marked canalization into two preferred directions: northwest (bise) and southwest (vent). A ve r y exhaustive statistical study of the climatology of G e n e v a - Cointrin Airport from 1949 to 1958 has been published by Aubert [27]. S o m e interesting surface wind statistics are s u m m a r i z e d in Fig.4. Typical of the entire Swiss Plateau is the continuous low stratus with marked inversion experienced in autumn and winter. This situation m a y last over a wide range of time, varying between 1 and 20 days [27]. T h e atmospheric diffusion effects on the specific activities in air X (^Ci/m3) have been computed by Pasquill's method as recommended b y the I A E A [28]: 446 PEETERMANS and BAARLI ÍAEA-SM-180/10 447 where Q is the source strength (pCi/s), u the average wind speed (m/s), h the height of the source above the ground (metres), a..- ^ tne lateral (crosswind) distance from the plume axis (metres). The dispersion parameters of ay and erg (in metres) as a function of downwind distance x, and for various meteorological categories can be calculated using empirical formulas and written as follows: ^(y)(z) ""(y)(z) +<3(y)(z) in* + Y(y)(z)l"^* (3) Typical values of <2, ß and y for meteorological categories A (extremely unstable conditions) and F (moderately stable) are given in Table IV. A computer programme has been written so that the values of x for each specific radioisotope can be calculated as a function of x and y for different meteorological categories. The values are corrected for de c a y by multiplying x (Eq.(2)) by: N o other corrections such as building effects and wash-out were considered. A m a x i m u m ground concentration of about 1.5 X 10*2 p C i / m 3 at a distance of about 100 m from the SC (post-SCIP), decreasing rapidly to 1.5 X 10*4 ^tCi/m^ at 1 k m was calculated for the meteorological category.A. F o r category F a m a x i m u m of 1.5 X 10*3 p C i / m 3 could be reached at a 1.5 k m s ource distance. 5. CONCLUSIONS Under some exceptional experimental conditions the radioactivity in the atmospheric air in the immediate vicinity of the targets or b e a m lines can be extremely high. However, the radioactive gases are generally to be considered as a second degree danger as compared to the risk of external ß-y irradiation from the machine parts. Outside the C E R N boundaries the stray radiation levels will very often remain predominant. F rom the improved C E R N SC point of view the m a x i m u m mean annual dose contribution beyond the C E R N boundaries, due to activated air released in the atmosphere, is about 5 m r e m / a . This c orresponds to a m a x i m u m average value of 2 X 10*4 ;nCi/m3 f,-,r ^ g isotopes concerned. As can be seen from Fig.5, no significant popula tion concentration is located within 1 k m radiüs around the C E R N SC. The C E R N site is however surrounded by agricultural land and vineyards so that deposition and wash-out should only be considered for ^Be and Зц, which are the long-lived isotopes of significance. M a x i m u m instantaneous ^Be concentrations in air due to the SC-Isolde (post-SCIP) operation have been calculated to be about 2 X 10*6 and 5 X 10'^ p C i /щЗ at distances of 100 m and 1 km, respectively. These values are to be compared with the 1.6 X 10*8 to 6.3 X 10*8 jnCi/m3 gué to production by cosmic radiation [29]. It is therefore evident that TBe surface pollution (pCi/m^) due to the C E R N high energy accelerators will be barely detectable. 448 PEETERMANS and BAARLI REFERENCES [1] RINDI, A., CHARALAMBUS, St., Nucl. Instrum. Methods 47 (1967) 227. [2] CHARALAMBUS. St., RINDI, A., Nucl. Instrum. Methods 56 (1967) 125. [3] HOFERT, M.. Rep. CERN Lab I - DI/HP/117 (1969). [4] PRANTL, F. A.. BAARLI, J., Rep. CERN Labi - DI/HP/122 (1969). [5] PRANTL, F. A., BAARLI, J.. CERN 72-15 (1972). [6] MICHAELIS, E.G., CERN Lab I - MSC/M-6 (1972). [7] RINDI, A., Rep. CERN Lab II - RA/72-5 (1972). [8] PEETERMANS. A., Rep. CERN Lab I - DI/HP/157 (1972). [9]SERBER, R., Phys. Rev. 72 (1947) 1114. [10] GOLBERGER, M.L., Phys. Rev. 74 (1948) 1269. [11] PEETERMANS, A., DELTENRE, R., Rep. CERN Labi - DI/HP/171 (1973). [12] SILBERBERG, R., TSAO, C. H.. Astrophys. J. 25 (1973) 315. [13] RUDSTAM, G., Nucí. Phys. A126 (1969) 401. [14] BERTINI. H.W., CUTHRIE, M.P., Nucl. Phys. A169 (1971) 670. [15] BERTINI, H.W., GUTHRIE, M.P., CULKOWSKI, A. H., USAEC Rep. ORNL-TM-3148 (1972). [16] HUDIS, J., in Nuclear Chemistry (YAFFE, L., Ed.), Academic Press, N e w York (1968). [17] SCHIMMERLING, W., etal., Phys. Rev. С 7 (1973) 248. [18] MICHAELIS. E.G., SKAREK, P.. Rep. CERNLabI-MSC/M-19/70 (1970). [19] VIALETTES, H., Proc. 2nd Int. Conf. Accelerator Dosimetry and Experience, SLAC, Stanford, U S A (1969) 121. [20] A W S CHALOM, M., LARSEN, F., SCHIMMERLING, W., Nucl. Instrum. Methods 75 (1969) 93. [21] Radiation Safety Manual, CERN, Geneva (1971). [22] PEETERMANS. A.. Rep. CERN 72-12 (1972). [23] KJELBERG, A., R U D S T A M , G., The Isolde Isotope Separator On-Line Facility at CERN, Rep. CERN 70-3 (1970). [24] RAUN, H.L., SUNDELL, S., WESTGAARD, L., 8thlnt. Conf. Low Energy Ion Accelerators and Mass Separators, Skovde, Sweden, 12-15 June 1973. [25] RENBERG, P. U., MEASDAY, D. F., PEPIN, M.. S C H W ALLER. P., FA VIER, B., RICHARD-SERRE, C., Nucl. Phys. A183 (1972) 81. [26] PERRET. Ch., Rep. CERN - MSC-21/613 (1965). [27] AUBERT, C., "Climatology of Geneva-Cointrin Airport", Swiss Meteorological Institute, Zurich (1963). [28] INTERNATIONAL A T O M I C ENERGY AGENCY, Application of Meteorology to Safety at Nuclear Plants, Safety Series N.29, IAEA. Vienna (1968). [29] JUNGE, ., Air Chemistry and Radioactivity, Academic Press, N e w York (1963). DISCUSSION E. N A G E L : I should like to point out that the wind rose of Ge n e v a - Cointrin airport concerns only advective winds. Light winds blow in a different direction towards Lake Geneva and are not canalized. Although C E R N m a y not be affected, other nuclear installations might be. M.B. BILES (Chairman): Those who operate high-energy accelerators are a w a r e that these facilities can produce radiation exposures higher than other nuclear facilities, unless care is taken in design and operation. It might be of interest to note that on the basis of 12 days' operation of the new accelerator of the National Accelerator Laboratory at Batavia, Illinois, at 400 G e V , the estimated annual exposure at the site boundary is 0.5 m r e m to the whole body. The dose from gases was too low to m e a s u r e . IAEA-SM-180/14 ANALYTICAL QUALITY CONTROL AND INFORMATION VALUE OF RESULTS IN ENVIRONMENTAL ANALYSIS D. BEYER, D. MERTEN Kernforschungsanlage Jülich, Federal Republic of Germany J. HEINONEN International Atomic Energy Agency, Vienna Abstract ANALYTICAL QUALITY CONTROL AND INFORMATION VALUE OF RESULTS IN ENVIRONMENTAL ANALYSIS. T h e results of analytical operations in environmental studies are the basis for far reaching decisions, for example as to protective and remedial measures to be taken, in diagnosis or therapy. T he information value of such data, therefore, has to be clearly indicated. Th e information value depends on the analytical involved. T h e analytical reliability of the results has to be characterized by stating the precision, accuracy and detection limits (where appropriate). Suitable methods for its determination are established in the paper. Analytical quality control measures have to be standardized with respect to the method applied and frequency restrictions due to the specific system. T he representativeness of samples is characterized by the "Operations Characteristic" (ОС) of the sampling scheme. The ОС specifies the probable error due to sampling. The sampling schemes have to be standardized with respect to frequency of sampling and number of samples to be taken in a certain area, taking into account the acceptable tolerances of the sampling error. The biological properties of the matrices involved, such as biological half-life, organ function, or composition of food, m a y 1. INTRODUCTION Reliable data required in many fields of human enterprise are obtained by analytical operations. Of particular importance are data relevant to the qualitative and quantitative composition of m a n and his environment. Both any necessary immediate action and future progress in environmental research, medicine and agriculture depend considerably on results of analytical investigations. Far reaching decisions in law, those regarding protective or remedial measures, and in diagnosis or therapy, to give but a few examples, rest on such investigations. There is an evergrowing need for reliable information in all these fields. How e v e r , the reliability of scientific routine w o r k which depends on the results of analytical investi gations is not always assured and m a y not suffice either in relation to what is needed or in what is technically feasible [1*3] . 449 450 t al. et. R E Y E B FIG. 1. System "Enviromental Analysis". IAEA-SM-180/14 451 This unsatisfactory situation m a y be due to either an unsatisfactory strategic concept of what would be suitable for the system being investi gated or an unsatisfactory system performance or the lack of a suitable experimental method, such as (a) an imperfect description and normal ization of analytical methods and results or(b) insufficient normalization and application of methods of quality control. 2. APPLICATION OF SYSTEM THEORY TO "ENVIRONMENTAL ANALYSIS" Analysis is a relatively well-defined scientific tool, which provides information on the character and the quantity of certain components of material matrices. It has to provide replies to the following questions [4j: " W h a t is present?" "How much is present?" "in what f o r m is it?" Such a task at first s e e m s quite unambiguous. H o w e v e r , an analytical operation, for example, in the case of environmental investigations, can only be a tactical step. The solution of more complex problems requires a strategic concept for the production of data of high information value. In this, analysis cannot be regarded as limited task. O n the contrary, all relevant factors and interrelations of the entire system, such as the nature of the investigation and its aim, and the influence of the individual elements on each other and on the system, have to be taken into account in the planning and performance of the analysis and in the interpretation of results. To carry out such evaluations,methods of system analysis and systems design have to be developed. 2.1. The system "EnvironmentalAnalysis" A n attempt is here m a d e to analyse the s y s t e m "EnvironmentalAnalysis", taking into account all elements and their relations which have an influence upon the information value of the results and which, therefore, have to be known to allow of a meaningful planning of the investigations. The System "Environmental Analysis" contains a large number of different s u b s y s t e m s and elements, viz: 2.1.1. Subsystem "Legislative and Administrative" Legislative and administrative bodies should define the scope of the investigations required with respect to the materials to be investigated, the type of nuclides to be determined, their permissible concentrations, frequency of analysis, required accuracy and reliability of data and their information value. There are informational relations between the elements in this s u b s y s t e m and be t w e e n all the other elements as well. 2.1.2. Element "Source oí Contamination" The source of contamination is characterized by contaminants of which the species, amount and time of release are unknown. This element is considered as a basic constituent of a series of subsystems and elements, between which material relations exist. 452 BEYER et al. 2.1.3. Subsystem "Environment" The Subsystem "Environment" consists of a large number of elements such as air, precipitation, water, soil, flora, fauna, ma n . There is a material relation between all these elements and the following Element "Sampling". E v e r y release leads first of all to a disequilibrium of the distribution of contaminants in the environment. This disequilibrium tends to change to dynamic equilibrium due to the material transport which immediately sets in. With an adequate knowledge of all the biological, chemical and physical factors, every step and transition of this equilibration can be understood and mathematically described by a model. T h e a i m of this kind of investigation is to describe the contamination of the environment, taking into account potential risks to man,while neither the Element "Source of Contamination" nor the details of material transport are sufficiently well-known, to calculate the exact b u r d e n to m a n . 2.1.4. Element "Sampling" A perfect description of the contamination of the environment requires a complete control of all elements and a knowledge of their variations with space and time. Due to the large number of elements, A to I (Fig.l.) in the Subsystem "Environment" and the inadequate know ledge of this subsystem, such knowledge and control is, however, practically impossible to attain. The investigations have to be restricted, therefore, to random samples. Because the content of a random sample can be entirely different from the characteristic content of the whole population, the information value of a random sample has to be clearly defined with respect to: (a) Its representativeness intime; (b) Its representativeness in space; (c) Its relevancy before one m a y extrapolate the investigations for one single element to the whole subsystem. 2.1.5. Element "Analysis" The Element "Analysis" involves all the operations and steps which are necessary for performing measurements and quality control. Analytical reliability has to be characterized with respect to: (a) Precision; (b) Accuracy; (c) Detection limit; (d) Counting statistics w h e r e appropriate. 2.1.6. Element "Result" T h e result of a m e a s u r e m e n t illustrates first of all numerically the properties investigated in a sample, e.g. the radionuclide content of milk. The interpretation of numerical data requires a statement of the information value, i.e.: (a) The relevancy of the sample; (b) The representativeness of the sample; (c) Analytical reliability. IAEA-SM-180/14 453 If the evaluation of a result depends on how quickly the data are available, then the time required has to be stated, too. T h e costs have to be indicated as well. It is true that the costs do not have any influence on the information value, but they m a y under practical circumstances, restrict the degree of information value, thus being significant for the selection of a procedure. 3. OPTIMIZATION AND STANDARDIZATION IN "ENVIRONMENTAL ANALYSIS" A system's analytical pattern covers the various influences and feed backs in the System "Environmental Analysis". The system can produce quantitiative statements on(i) the degree of information value, which it is possible to attain from the analysis of samples and (ii) the information value which is to be aimed at on the basis of legal and administrative regulations. O n the basis of the above consideration one m a y then determine: (a) The sampling scheme; (b) The analytical procedure. The relative importance of the individual components of the information value derives from these relationships. Thus it does not make sense to strive for extremely high sensitivity when the relevancy and representative ness of the sample is not adequate. It inevitably follows that all factors such as sampling, analytical procedure, methods for quality assurance,have to be optimized, i.e. they have to be adapted to the requirements and conditions of the entire system. After an optimization, these procedures can be standardized and optimized, at least for routine investigations. Such standardization has to cover: (a) Sampling; (b) Analytical procedure; (c) Quality assurance; (d) Statement of results. 3. 1. Unification of the sampling In the selection of sampling procedure, particularly in the establish ment of representativeness with respect to space and time, as well as to relevancy, the degree of information value has to be taken into account. Such evaluations have to be undertaken for every kind of contamination and for every element of the Subsystem "Environment" separately. Under these conditions the sampling intervals, which determine the representativeness of the sample, have to be fixed to be at regular intervals; 0.25 months, 0.5 months, 1 month, 2 months, 3 months, 6 months or 12 months. In order to determine the representativeness with regard to space, special investigations have to be undertaken in each case, because the variations with respect to space cannot be stated and calculated in advance. The relevancy of a sample can be estimated with an adequate knowledge of the Subsystem "Environment" and with a clearly defined problem. For instance, indicator-elements (plants and animals) have a very high degree of relevancy if it is possible to extrapolate from their degree of contami nation to that of the whole subsystem. 454 BEYER et al. 3.2. Analytical procedure T h e analytical procedure has to be described in detail. F o r its characterization one should indicate [5, 6]: (a) Matrix; (b) W o r k i n g level; (c) Amount of samples; (d) Detection limit; (e) Standard deviation of method; (f) Influence of substances which are foreign to the matrix; (g) Time requirements; (h) Costs. T h e latter data are necessary in order to m a k e the best choice b etween procedures which are otherwise equivalent. 3.3. Analytical quality control (quality assurance) In spite of the standardization or harmonization of the analytical pro cedure, the quality of results fluctuates because of random and systematic errors, the degree of which varies with time. It is, therefore, necessary to control the quality of the analytical data at regular intervals. 3.3.1. Analytical quality control in the case of single investigations The magnitude of random and systematic errors depends on the routine and skill of the analyst, on the state of the equipment, its calibration, etc. If a certain analytical procedure is s e l d o m used, it is necessary to test the results, taking into account all its properties, i.e. one m u s t undertake a total control. T h e effort involved is great, but necessary, in order to assure the reliability of the result. 3.3.2. Analytical quality control in routine investigations In the case of regularly repeated investigations, a total control of the reliability of an analyticalprocedure is neither practicable nor necessary. Random tests are quite adequate. The frequency of random tests depends on the degree of the reliability required to obtain the correct quality of results. The type and frequency of random tests can be specified according to the type of investigation. 3.4. Statement of results In order to mak e the results of analytical operations interpretable, data on their information value are necessary in addition to the normal data. JL-— J^ta^or^amp_le_ ^JData^orjs^nda^jprocedure — Matrix — Characterization of procedure — N u m b e r of sa m p l e s -* Standard deviation *" Component investigated — Detection limit IAEA-SM-180/14 455 3. Numerical result 4. Information value Numerical value Sampling Mean value Representativeness N u m b e r of individual Analysis determinations Standard deviation, accuracy, detection limit The production of appropriate, reliable analytical data in environmental analysis requires a systems analysis study of the entire system "Environmental Analysis". For this purpose, all the subsystems, elements and relations which might have some influence on the information value have to be deter mined, taking into account the object in view. F r o m this kind of study it is possible to extract qualitative and quantitative indications as to the mutual interrelationships and effects which are necessary for planning, performance of investigations and the evaluation and compilation of the results. F r o m this it follows that the procedure of sampling and analysis as well as assurance of the quality of the result have to be optimized and standardized, and the results have to be presented together with data on the information value. REFERENCES [1] MERTEN. D . , WORTLEY, G . . The Reliability of World Wide Monitoring in the Light of Accuracy of Low Level Radiochemical Analysis, IAEA Internal Report, 30 June 1967. [2] TUGSAVUL, A. , MERTEN, D. , SUSCHNY, O . , The Reliability of Low Level Radiochemical Analysis during the Period 1966-1969, IAEA, Internal Report, 30 March 1970. [3] Report on a Joint IUPAC, ILO, FAO, W H O and IAEA Meeting on Analytical Chemistry and its Influence on Economic and Social Development, IAEA Internal Report No. 68/8346 (1968). [4] G O T T S C H A L K , G . , Zukunftsaspekte der Analytik, Z. Analytische Chemie 258 (1972) 1-12. [5] G O T T S C H A L K , G . , Standardisierung in der angewandten Analytik, private communication. [6] BEYER, D . , M E R T E N , D. , "Vorschläge zur Standardisierung in der Ausscheidungsanalytik", 2. Sitzung des Arbeitskreises Inkorporationsüberwachung des Fachverbandes Strahlenschutz, Waldshut, 28-29 January 1973. IAEA-SM-180/12 DEVELOPMENTS IN THE UNITED KINGDOM IN THE DERIVATION OF EMERGENCY REFERENCE . LEVELS IN ENVIRONMENTAL MATERIALS Pamela M. BRYANT National Radiological Ptotection Board, Harwell, Didcot, Berks., United Kingdom Abstract DEVELOPMENTS IN THE UNITED KINGDOM IN THE DERIVATION OF EMERGENCY REFERENCE LEVELS IN ENVIRONMENTAL MATERIALS. Following an accidental escape of radioactive material from a nuclear installation to the environment, limitation in the amount of public radiation exposure might be required. Because of the great variability of the circumstances in which remedial action might be considered, the International Commission on Radiological Protection does not m a k e specific recommendations on 'actionlevels' but suggests that the setting of such levels is the responsibility of the national authorities. Guidance on m a x i m u m permissible contamination of diet and respirable air published by the United K i n g d o m Medical Research Council during the period 1959-1961 is being reviewed in order to take account of various developments in the intervening period. Th e concept of m a x i m u m permissible radiation exposure in connection with an accidental escape has been superseded by the m o r e flexible concept of an emergency reference level of dose. This is the basic standard; it is not a firm 'action level' but a level at which responsible authorities should judge whether countermeasures should be introduced. Secondary standards are described as derived emergency reference levels; these levels, which m a y refer to milk, air or other environmental materials, are intended to provide a basis for estimates of potential radiation exposure on which decisions regarding the application of countermeasures m a y be made. Other developments considered are changes in reactor types and in the radionuclides which might escape accidentally; also, advances in the metabolic, dosimetric and food chain data required to derive values for emergency reference levels in environmental materials. Present knowledge regarding radiostrontium metabolism indicates that a completely n e w approach is necessary, involving a change in the portion of bone INTRODUCTION Operation of a nuclear power industry entails some risk, however slight, of accidental escape of radionuclides to atmosphere. In the event of an "uncontrolled" escape of activity from a nuclear installation, radiation exposure of members of the public can be limited only by counter measures in the public domain that interfere with normal conditions. Such counter measures include evacuation of members of the public, control of food supplies and, if radioiodine has escaped, use of tablets containing stable iodine. Because of the great variability of the circumstances in which remedial action might be considered, the International Commission on Radiological Protection (ICRP) does not make specific recommendations on "action levels". The Commission considers that it will be appropriate to institute counter measures only when their social cost and risk will be less than those resulting from the radiation exposure, and that the setting of action levels for particular circumstances is the responsibility of national authorities./1/ 457 458 BRYANT In the United Kingdom the basic standards for such action levels are now commonly termed "emergency reference levels (ERLs)" of dose. So that situ ations can be rapidly assessed in practice, it is necessary to have secondary standards expressed in terms either of exposure or of activity in environ mental materials. These are termed "derived ERLs" and they correspond to appropriate ERLs of dose. During the period 1959-1961, the UK Medical Research Council made recommendations on maximum permissible radiation exposure and contamination levels in certain environmental materials related to maximum levels of dose./2-4/ The Council pointed out that, in recommending these levels, they used the fullest information available at the time, but where information was inadequate they used the more cautious, rather than the more likely, assumption. The Council stressed that knowledge of the metabolism of radio isotopes was increasing and that it was probable some of the recommended levels were lower than would subsequently be seen to be necessary. In the intervening period there have been substantial developments in knowledge of subjects inherent in the recommendations and a review of the recommendations is in progress under the aegis of the Council. The conclusions of this review are not yet available for quotation. This paper describes developments relevant to the derivation of ERLs in environmental materials. The developments involve revised and more compre hensive calculations using metabolic and dosimetric data not available when the Council's reports were prepared. They include a comprehensive re- evaluation of metabolism of radiostrontium in bone and also of the passage of radiostrontium through the food chain to man. Fresh metabolic, dosimetric and food chain data are also available for radioiodine and radiocaesium. These advances mean that calculations of values for derived ERLs in environ mental materials are now on a firmer basis than they were during 1959-1961. SUMMARY OF UK PUBLISHED GUIDANCE ON MAXIMUM PERMISSIBLE CONTAMINATION OF DIET AND RESPIRABLE AIR After the accidental escape of activity frcm No. 1 pile at Windscale in 1957, /5/the Medical Research Council (MRC) published recommendations on the maximum permissible daily intake in diet of each of a small group of radio nuclides which are produced in high yield by fission of reactor fuel^¡_ and which are in addition highly concentrated in certain organs of the body./2_/ Subsequently the Council published recommendations on the maximum intake of the same radionuclides by inhalation during their passage in a cloud./4_/ In another report during the same period, the Council published recommendations on emergency exposure to external radiation which might be derived from the passage of the cloud or from deposited material._/3_/ It was pointed out that persons working or residing in the immediate neighbourhood of the site of an accidental escape of radionuclides may be exposed simultaneously to external radiation from the cloud and internal radiation from inhaled activity, and subsequently to external radiation from radionuclides deposited from the cloud and to internal radiation from ingested activity in contaminated foodstuffs. In applying the guidance on maximum permissible contamination of diet and respirable air, account must be taken of all routes of exposure. The recommendations made during 1959-1961 /2,4/ for ^Sr, ^ S r , and Cs intakes and contamination of diet and respirable air are given in Table I for children aged 6 months. This age is selected from various ages, including adulthood, discussed in the recommendations because it is, with one exception, the most restrictive age. The exception refers to inhaled 137cs lAEA-SM-180/12 459 TABLE I. 1959-1961 RECOMMENDATIONS ON MAXIMUM PERMISSIBLE CONTAMINATION OF DIET AND RESPIRABLE AIR FOR CHILDREN AGED 6 MONTHS .34 Radionuclide Ю S r 9 " S r i 3 ? c s DIET 0 . 2 ^ 0 . 002 ^ 0 . 0 6 0 . 1 5 0 . 0 6 5 0 . 1 5 in mi l k {(iCi/1) RESPIRABLE AIR Total intake (pCi) 5. 5 0. 0 5 7 0. 88 ^ 1 9 b 0 . 0 8 0. 0 0 0 8 0 . 0 1 3 0 . 2 8 ( C i - s /щЗ) Maximum level of dose 15 rad 1 . 5 r a d / a 2 5 r a d 1 0 r a d a Associated with a normal calcium intake of 1 g daily. b T o allow for the contribution to dose of other short-lived isotopes of iodine and ^Te, these values should be reduced by a factor of 2 if the reactor has been operating for long enough for for which the time integral of concentration in air is marginally more restric tive for children aged 1 year than aged 6 months. The maximum levels of dose given in Table I refer to bone for radiostrontium, the thyroid gland for 131l and the whole body for 137cs. It was not clear in 1959 /2_/ whether the calculations for radiostrontium should be based on mean radiation dose in the skeleton or local dose at sites of maximum concentration. The values in Table I are based on the more cautious assumption that the radiation dose or dose rate is not to be exceeded at any site of radiostrontium concentration in bone. It was pointed out that if future work showed mean skeletal dose, and not local dose, to be the signi ficant criterion, it would be possible to increase the values of dietary intake given in Table I. The second difficulty arose because of the big difference in the half-lives of ?Ogr (28 yr) and 89gf (53 d). It was decided that the maximum permissible contamination by the long-lived 90sr should be calculated on the basis of the maximum yearly dose rate at a site of concen tration, while the criterion for the short-lived 89gp should be the maximum total radiation dose at any such site. The corresponding daily intakes in diet are applicable irrespective of the duration of intake. While it was recognised that these levels might be lower than necessary if intakes were limited to short periods, it was concluded that more data on certain aspects of the metabolism of calcium and strontium were required before any higher figures could be recommended./2/ Similar reservations regarding the lack of knowledge of skeletal growth and remodelling, particularly for young children^_were noted 2 years later in the recommendations for inhaled radiostrontium./4/ In contrast with the comparatively long duration envisaged for intake in diet, intake by inhalation following an accidental escape of activity could be of very short duration. 460 BRYANT In consequence, the variation with time of the radiation doses delivered to bone, and perhaps the distribution of the radiostrontium in the bone, will be different in the two cases. The rate of entry of radiostrontium into blood from lung may vary with the site and type of deposition in the lung. After reviewing the available evidence, it was decided that the radiation dose at any site of maximum concentration following an intake of short duration may be assumed to be about five times greater than that which would be obtained if all the radiostrontium were distributed uniformly throughout the skeleton./4/ In the calculation of the values given in Table I for radiostrontium in res pirable air, it was assumed that there was no biological elimination of radiostrontium from bone. 131 137 The values for I and Cs in diet given in Table I were calculated from data then available on the likely sequence of rise and fall of food con tamination and are expressed in terms of maximum contamination soon after the accident. These values and those in respirable air take into account the then available data on metabolic and dosimetric behaviour of ingested and inhaled and 137cs in humans. In all cases of uncertainty, cautious assumptions were made. There was no cause for uncertainty, however, in the choice of the thyroid gland as the critical organ for radioiodine and the whole body for radiocaesium. EMERGENCY REFERENCE LEVELS The ERL has been defined as: An emergency reference level is a value, usually of dose but sometimes of an environmental measurement, below which counter measures are unlikely to be justified. If the dose seems likely to exceed the emergency reference level, counter measures should be undertaken if they are likely to achieve a substantial reduction of dose and if they can be carried out without undue risk to the community. The counter measures appropriate to doses that would only moderately exceed the emergency reference level should be ones which do not involve appreciable risk to the community, measures involving greater hazard being called for only if radiation exposures would otherwise be considerable./6/ The levels are not put forward as firm action levels but as levels at which the responsible authorities should judge whether counter measures should be introduced. This judgement should take full account of the disadvantages and risks such counter measures may create. There is increased flexibility in the concept of ERLs compared with that of maximum permissible or "acceptable" doses. At the same time it should be mentioned that even when doses were described as acceptable, they did not represent levels below which there was no effect whatsoever and above which there would necessarily be serious con sequences. The flexibility inherent in the concept of ERLs of dose is also embodied in the derived ERLs in environmental materials used to assess situations rapidly so that executive decisions on counter measures can be made. The objective of this paper is to describe developments leading to revised numerical values for derived ERLs corresponding to a given ERL of dose. ERLs of dose on which derived ERLs in environmental materials in this paper are based are 25 rads to_the thyroid gland and 10 rads to the whole body as in the published guidance./2,4/ It is assumed that 10 rads also applies as the ERL of dose to red bone narrow, although the ICRP recognises that limits for red bone marrow, for workers or for members of the general public, [AEA-SM-180/12 461 might be raised to a moderate extent to become_consistent with the levels adopted for the whole body and other tissues ./7_/ In this paper the term ERL of dose is taken to mean the total dose commitment arising from the circum stance under consideration. NUCLIDES IMPORTANT IN VARIOUS TYPES OF ACCIDENTAL ESCAPE The radiological importance of nuclides which have escaped to atmosphere depends upon a combination of many factors, including quantity of activity, degree of chemical similarity of the nuclides to elements normally present in biological materials, and existence of means by which the nuclides are trans ferred to members of the public. Quantities of activity encountered in the nuclear power industry are generally several orders of magnitude greater than quantities used in the application of radionuclides in medicine, research and industry. Because the potential for a substantial, uncontrolled escape of activity to atmosphere is greater in nuclear power reactors than elsewhere, the nuclides chosen for detailed discussion in this paper are those which may be important in the context of these reactors. An assessment of environmental hazards resulting from the escape of fission products due to a postulated uranium fire in a gas-cooled thermal reactor of the type currently operated in the United Kingdom by the Central Electricity Generating Board showed that 131i would usually be the most impor tant fission product in an escape to atmosphere, but that the effects of certain other fission products could not be ignored./8/ Among these were other isotopes of iodine, and certain isotopes of tellurium, strontium, caesium, cerium, ruthenium and the rare gases, krypton and xenon. A similar assessment was subsequently carried out^ for reactor systems using uranium dioxide pellets as fuel in metal cans./9/ This fuel is used in many present-day thermal reactors, such as Advanced Gas-Cooled Reactors, Pressurised and Boiling Water Reactors, and the Steam Generating Heavy Water Reactor. Three basic types of release of fission products from fuel were assessed in the report, namely, melt-out due to overheating and release by can-puncture, both of which involve release of volatile and gaseous fission products, and a release of noble gas fission products only. In the context of possible dietary contamination and inhalation of contaminated air, release by melt-out followed by quantitative escape to atmosphere provides a reason able basis for detailed consideration. It is concluded that the group consisting of 89gr, 90gr, radioiodine, 132те and T^Cs, for which guidance was originally published in the period 1959-1961, is still of interest. Additional nuclides s4iich may also be of importance in the context of internal radiation doses include ^06Ru and 144ce. DERIVED ERLs FOR RADIOSTRONTIUM Metabolic model In a report prepared by the ICRP Task Group set up to review the radio sensitivity of the tissues in bone, it was concluded that the important tissues in bone with respect to bone cancer are the endosteal cells, particularly those associated with trabecular bone, and with respect to leukaemia the red bone marrow./10/ These specified tissues replace mineral bone, regarded as a single uniform medium, as the potential critical organ for ingested and inhaled radiostrontium. There have been a number of attempts to formulate a model representing radiostrontium metabolism in man; the most comprehensive of these is probably that developed by Marshall./11/ This model is employed in the report by the ICRP Task Group on Local Retention Functions of Bone Seekers,/12/ and is discussed by Papworth and Vennart in a paper which correlates the results 462 BRYANT o f United Kingdom measurements o f ^ S r in human bone and d i e t ./11/ The Marshall model considers retention of 90sr in various body tissues for adults only, whereas the Papworth and Vennart model describes uptake and reten tion o f ingested 9C*Sr as a function of age. A method of calculating Dg and D^, the doses received by endosteal tissues and red bone marrow respectively in trabecular bone, has been described by Spiers and his colleagues./13-15/ They consider the dose in a small, tissue-filled cavity surrounded by bone mineral uniformly contaminated with the radionuclide in question and extending in all directions to a distance greater than the range of the ß rays concerned. They give factors Dg/D^ and D^/D^ by which Do must be multiplied to obtain Dg and D^. In the present context (in rads) is the radiation dose to which bone is committed following an intake of radiostrontium. Values of the factors Dg/Do and D^/D^ for ^Sr and 90sr in^ trabecular bone show that for both nuclides in the ratio Dg/D^ is about 1.5 ./11/ Values of the ratio w ill approach 3 if Dg is averaged throughout a ll endosteal cells including those in compact bones./Tl/ Dose commitments to bone marrow per m icrocurie ^ S r and ^ S r ingested a t b ir th , a t 6 months and a t each age from 1 to 20 years were calcu late d by Papworth and Vennart /11/ using th e ir model and values fo r reten tio n functions and the Spiers method for estimating doses. For both nuclides values are higher in infancy than at any later age. Models for the deposition and retention of inhaled material in the human respiratory tract were proposed in a report by the ICRP Task Group on Lung Dynamics. /16/ The deposition model describes dust deposition in terms o f three anatomical compartments, broadly identified as the nasopharyngeal, tracheobronchial and pulmonary regions. The dust retention model considers retention only as an index of the rate or extent of dust clearance. Amounts o f m aterial deposited and amounts and ra te s o f clearance depend on many complex factors, including size, density and chemical state of particles. The Task Group proposed constants, consisting of values of biological half-time and regional fractions, for use with the clearance model. The constants are given for three classes of material: Class (D) exhibiting rapid clearance, Class (W) having intermediate clearance times and Class (Y) being avidly retained. Amended constants for use with the clearance model were subsequently given by the ICRP Task Group_on the metabolism o f plutonium and related elements and th e ir compounds./17/ Following clearance from the various compartments of lung, inhaled radiostrontium either goes direct to blood or via the gut to blood. There is no suggestion in the Task Group report /16/ th at the model depends markedly on age and i t may be assumed th at the calcu lated fra c tio n a l contributions to blood apply a t a l l ages. Dose commitments per m icrocurie ^ S r and 9°Sr inhaled may be calculated from appropriate values of dose commitment per microcurie ingested by multiplying by the sum of the fractional contributions to blood, the fraction reaching blood direct being weighted by a factor of 5 because the fractional absorption from gut to blood is about 0.2./11/ Environmental model An environmental model to estimate transfer of deposited radiostrontium through the food chain to man requires data derived both from experiments and from observations on fallout from nuclear weapons tests. Since fresh milk will often be the main source of radiostrontium in diet, especially for children, the concentration in milk will normally provide a valid basis for counter measures. IAEA-SM-180/12 463 The concentrations of Sr and Sr in milk that would occur in the first 30 days after a single deposition on pasture grazedj% cattle have been derived from the results of experiments with dairy cattle./18/ The calculation assumed that cows obtain their entire diet by grazing pasture of average pro ductivity and do not receive any supplementary uncontaminated food. This is the situation which leads to the highest estimate of milk contamination. The curve of concentration in milk against time rises to a peak and then falls away; the total intake of 89g¡- and 90gy activity by humans due to consumption of milk contaminated as a result of recent deposition can be calculated by integrating the total area under the respective curves, an exponential tail being assumed fo r the purpose. The peak a c t iv it y i s op eratio n ally most re a d ily available and provides the basis for countermeasures. 90 Because of the long half-life of Sr, account must also be taken of the contamination of milk that will continue at a lower level in later years. Information on longer-term transfer is available from extensive studies of world-wide fallout and a multi-term equation was derived by Bartlett, Russell and Jenkins from United Kingdom data./19/ In the case of fallout, contamination of milk arises from recently deposited ^-*Sr and from previously deposited 90gj- which enters grass from soil and from the plant base region and which is present in stored cattle food. If it is assumed that the same processes occur in the case of a single deposition of 90g^ such as might occur following an accidental escape of limited duration, appropriate terms in the fallout transfer equation can be used to predict contamination of milk in each successive calendar year following the year in which deposition takes place. It is recognised that the equation, when used in this context, w ill frequently overestimate the integrated concentration of 90g^ in milk because the relationships established from fallout data imply that all cattle feeding stuffs are contaminated. In the case of an accidental escape, feeding stuffs imported from outside the area will make no contribution. Summation of 90gr intakes calculated (a) from the curve representing transfer of recent deposition to milk and (b) from the terms of the equation relevant to subsequent transfer to milk leads to a value which is almost certainly an upper estimate of what would be likely to happen in practice. Evaluation of derived ERLs in environmental materials Derived ERLs of practical value in the event of an escape of radio strontium include the ERL of peak activity in milk and the ERL of time inte gral of concentration of activity in air. To calculate these, representative values of daily milk intake and volume of air breathed per unit time for appropriate ages are required in addition to the metabolic, dosimetric and food chain data already described. 90 In the case of ingested Sr, intake may continue for many years and the total dose commitment to an individual is the sum of the contributions arising from each year's intake as the individual grows older. In the case of inhaled radiostrontium, intake w ill take place almost entirely during the passage o f the cloud downwind. Present knowledge o f bone metabolism suggests that weighting for the short duration of intake is not necessary. The numer ical values for derived ERLs for radiostrontium given in Table II result from application of the above methods,/20/ on the assumption that the ERL of dose i s 10 rads to bone marrow. DERIVED ERLs FOR RADIOIODINE AND RADIOCAESIUM Metabolic model The thyroid gland remains the critical organ for intakes of radioiodine and the whole body for radiocaesium. Radionuclides of interest are 464 BRYANT TABLE II. DERIVED EMERGENCY TABLE III. DERIVED EMERGENCY REFERENCE LEVELS FOR REFERENCE LEVELS FOR CHILDREN FOR RADIOSTRONTIUM RADIOIODINE AND RADIOCAESIUM IN ENVIRONMENTAL MATERIALS" IN ENVIRONMENTAL MATERIALS [1] Radionuclide s ' S r 9 ° S r i3,. Radionuclide ' 3 ? C s DIET DIET c h i l d r e n c h i l d r e n ERL of peak 3. 5 0. 1 5 0. 2 5 6 . 7 i f ^ C i / 1 ) in mi l k (;iCi/l) ' RESPIRABLE AIR RESPIRABLE AIR c h i l d r e n a d u l t s E R L o f t i m e ERL of time integral of 0. 8 5 0 . 0 5 0. 0 3 " 0. 9 3 concentration in air ( C i * s / m p ) ( C i * s / m 3 ) ERL of dose 25 rad 10 rad " T o allow for the contribution to dose of other short-lived isotopes of iodine and ^ T e , these values 132те, which i s the parent o f 132^ and ^37(-g. Improved metabolic and dosi metric data for these radionuclides, especially in respect of children, have become available since the period 1959-1961. Some of these data were discussed by the present author in 1969 when revised values were presented for certain parameters./21/ These parameters include the fraction of the intake reaching the critical organ, and effective half-life and effective energy per disinte gration of the radionuclide in the critical organ. It is possible to make further revisions of values for the fraction reaching the critical organ, using the ICRP Task Group lung model /16/ and the amended constants /17/ described in the section on radiostrontium, and of values for effective energy per disintegration, using more recent data. These revisions would lead to results for radioiodine and *37çg differing, but only slightly, fron those given in reference /21/. Environmental model 131 An environmental model to estimate transfer via milk of deposited I and 137cg may be based on experimental data similar to that described for radiostrontium. The model and resultjji^ numerical values were discussed in detail by the present author in 1969./21/ It should be noted that only the 131i component of a mixture of iodine isotopes contributes to dose via milk, owing to radioactive decay of the shorter-lived isotopes during the time between deposition on pasture and human consumption of milk. Furthermore the rela tively short radioactive half-life of 131i of g days ensures that there is no contribution to dose in subsequent years, nor any contribution from stored IAEA-SM-180/12 465 cattle foodstuffs even if they were initially contaminated. The converse applies to 137cs with its long radioactive_half-life of 30 years. The anal ysis of data relating to fallout ^3^Cs /22/ provides a method for predicting contamination of milk in future years. Evaluation of derived ERLs in environmental materials The derived ERLs of peak activity in milk and of time integral of __ activity in respirable air given in Table III are quoted from reference /2JL/. The values for 131i were recommended for use in the United Kingdom by the MRC Committee on Protection against Ionising Radiations; the values for 137cg are used in the UK. Atomic Energy Authority but were not submitted to the MRC Committee .^/21^ It is important to note that the values for 131i in air should be reduced by a factor of 2 if the reactor has been operating long enough for isotopes of iodine and Te to be in equilibrium or by a factor of 10 for a reactor at start-up.^4,2l/ It is also important to note that the ERL for adults for ^37cg contaminated air is more restrictive than that for children. DISCUSSION The derived ERLs for radiostrontium in environmental materials given in Table II result from the use of new metabolic, dosimetric and food chain data, on the assumption th at the relevan t ERL o f dose i s 10 rads to bone marrow. This represents a change in critical organ from mineral bone considered as a single uniform medium. On this basis the derived ERL for the peak concentra tion of 89gr in milk is nearly 20 times and the ERL for the time integral of concentration in a ir about 10 times the resp e ctiv e values cau tio u sly recorrmen- ded during 1959-1961 (Table I). Although no values were given in the recom mendations for the peak concentration in milk, the recommended maximum perm issible d a ily intake o f 0.2 pCi 89gj-/d associated with 1 g Ca/d /2/ may be taken as equivalent to 0.2 pCi 89g^./i because there is about 1 g Ca/1 in milk. The same reasoning applies to ^Sr. on the basis adopted the derived ERL for 90gj- in milk is nearly 80 times and in air about 60 times the respective recommended values (Table I). The greater ratios applicable to 90sr compared with ^ S r r e s u lt from the greater caution inherent in the e a r lie r calculations for the radioactively long-lived 90g¡- than for the short-lived ^Sr. The derived ERLs for radioiodine and radiocaesium in environmental materials given in Table III correspond to ERLs of dose recommended in 1959 for the thyroid and the whole body respectively.^/2^ There is no change in the critical organs, and the values for derived ERLs are those published in 1969./21/ On this basis the derived ERL for the peak concentration of 131i in milk is about four times and the ERL for the time integral of concentration in air about twice the respective values cautiously recommended during 1959-1961 (Table I). The corresponding ratios for derived ERLs for 137cs are about 40 for milk and about 3 for air. The greater ratio applicable to ^37¡-g in milk compared with 137cs in a ir and ^31^- ^ environmental m aterials r e s u lts partly from the greater caution inherent in the earlier calculation for the radioactively long-lived 137cs in the milk pathway compared with 137cg in air and the radioactively short-lived ^31^ in the milk pathway and in air. The greater ratio is also connected with the biological half-life of 137cs in the human body. In the earlier calculations a long biological half-life was assumed to apply at all ages, whereas the h alf-life is now known to be much shorter in children than adults. This change contributes directly to the ratio of values for 137cg in milk because children constitute the critical group in both cases. But in the case of inhaled 13*?Cs it results in a 466 BRYANT change in the critical group from children to adults and a small ratio between the derived ERL in air and the previously recommended value. The respective changes in values demónstrate the numerical effect on the bases adopted in this paper of the developments described. REFERENCES /1 7 INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, P ublication 9, Pergamon P ress, Oxford (1966). /2/ Maximum permissible dietary contamination after the accidental release of radioactive material from a nuclear reactor: Report to the Medical Research Council by its Committee on Protection against Ionising Radiations, British Medical Journal 1 (1959) 967. /3/ Report on emergency exposure to external rad iatio n , Recommendations of the Medical Research Council Committee on Protection against Ionising Radiations. In: Hazards to Man of Nuclear and Allied Radiations, Appendix K, Qnnd 1225, HM Stationery Office, London (1960). /4/ Maximum permissible contamination of respirable air after an accidental release of radioiodine, radiostrontium and caesium-137: Report to the Medical Research Council by its Committee on Protection against Ionising Radiations, British Medical Journal 2 (1961) 576. /5/ ATOMIC ENERGY OFFICE, Accident at Windscale No.l Pile on October 10, 1957, Qnnd 302, HM Station ery O ffic e , London (1957). /6/ MARLEY, W.G., Evaluation of environmental monitoring data for nuclear facilities. In: Rapid Methods for Measuring Radioactivity in the Environment, IAEA, Vienna (1971) 799. /1 7 INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION,Statement issued following the meeting of the International Comnission on Radiological P rotection in November 1972, Health Physics 24 (1973) 360. /8/ BEATTIE, J.R., An assessment of environmental hazards from fission product releases, UKAEA Rep. AHSB(S)R-64 (1963). /9/ BEATTIE, J .R ., BRYANT, Pamela M., Assessment o f environmental hazards from reactor fission product releases, UKAEA Rep. AHSB(S)R-135 (1970). / Й 7 INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, A review o f the radio-sensitivity of the tissues in bone, ICRP Publication 11, Pergamon P ress, Oxford (1968). ----- 90 /11/ PAPWORTH, D ., VENNART, J . , Retention o f Sr in human bone a t different ages and the resulting radiation doses, Phys. Med. Biol. 18 (1973) 169. /12/ MARSHALL, J.H. et a l., Alkaline earth metabolism in adult man. Health Physics 24 (1973) 125, and ICRP P ublication 20, Pergamon Press, Oxford (1973). tAEA-SM-180/12 487 /13/ SPIERS, F.W., Proceedings of the First International Congress of Radiation P rotection . Pergamon P ress, Oxford, (1968) 165. /14/ SPIERS, F.W ., Delayed E ffe c ts o f Bone Seeking Radionuclides, Eds. C.W. Mays et a l., University of Utah Press (1969) 95. /15/ WHITWELL, J .R ., SPIERS, F.W ., Proc. 5th Cong, o f the French Society of Radioprotection, Grenoble, France (1971) 401. / K 7 INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, Deposition and retention models for internal dosimetry of the human respiratory tract, A repo rt prepared by the Task Group on Lung Dynamics, Health Physics 12 (1966) 173. / l V INTERNATIONAL COMMISSION ON RADIOLOGICAL PROTECTION, The metabolism of compounds of plutonium and other actinides, ICRP Publication 19, Pergamon P ress, Oxford (1972). /18/ SCOTT RUSSELL R ., (Ed.) R ad ioactivity and Human D iet, Pergamon P ress, Oxford (1966). /19/ BARTLETT, B .O ., RUSSELL, R .S ., JENKINS, W., Improved relatio n sh ip between the deposition of strontium-90 and the contamination of milk in the United Kingdom, Nature, London 238 (1972) 46. /20/ vmNART, J . , Personal communication. /21/ BRYANT, Pamela M., Data for assessments concerning controlled and accidental release of ^31^ and 137cg to the atmosphere,Health Physics 17 (1969) 51. /22/ BARTLETT, B.O., Reliability of predictions of the contamination of milk with fis s io n products in the United Kingdom, Nature, London, 216 (1967) 385. DISCUSSION J. PENSKO: What is your opinion on the need to estimate emergency reference levels in terms of collective dose, taking into account not only the dose absorbed but also the number of people involved? P a m e l a M . B R Y A N T : W e have not e x a m i n e d this question in an y detail. However, I do know that the collective dose concept is used in Sweden. B. LINDELL: I should explain that the collective dose is used in Sweden as yet only in the control of normal operations and not for accidental situations, for which w e still use the old United Kingdom action levels. This does not mean, however, that information on collective dose, e. g. per unit quantity of contaminated food, might not be useful as a basis for cost/benefit evaluation of some remedial action at a lower level of expected dose. N. G. GUSEV: I listened to your paper with the greatest of pleasure; these investigations are without doubt of great practical value. Your paper indicates that the m a x i m u m dose levels for accidental exposure have not changed in the United Kingdom. At the same time you supply time integrals of concentration which exceed those formerly applying by a factor of approximately three for and by a factor of ten for S9Sr and by a 468 BRYANT factor of 60 for ^°Sr. This implies that corresponding increases in accidental releases can also be permitted. Can you please tell m e whether these values have already be e n officially adopted? Pamela M. B R Y ANT: The values given in m y paper for the time integral of concentration are levels at which the need for remedial action should be considered. The revised values for ^^1 and ^ C s are those given in m y paper published in 1969. The revised values for^I were recommended by the Medical Research Council for use in the United Kingdom; those for -^Cs were not submitted to the Medical Research Council but they were recommended for use within the United Kingdom Atomic Energy Authority. The revised values for radiostrontium have no official status whatsoever; they illustrate the effect of applying the developments in knowledge based on the assumption of 10 rads as the E R L of dose to bone marrow. This E R L of dose has been assumed for the purpose of m y paper. The ERLs of dose for **^1 and ^ C s adopted in m y paper are those recommended by the Medical Research Council in 1959. The whole question of values for E R L s of dose is currently under review by the Council. N. G. GUSEV: Did you use a power function or an exponential model for determining the dose f r o m 9°Sr? Pamela M. BRYANT: The dose from 9°Sr was estimated on the ass u m p t i o n that retention in bone is exponential but that the rate constant varies with age. Allowance is also m a d e for different rate constants in cancellous and c o m p a c t bone in adults but not in children. T h e total dose commitment is the summation of the dose commitment from each annual intake for a lifetime. M.B. BILES (Chairman): In the United States of America we are m o v i n g towards establishing action levels at each facility for e m e r g e n c y response. Of course these action levels must be based on data such as those being developed by Miss Bryant and her colleagues. W e believe that specific action levels for each facility, taking into account population density, site characteristics, facility characteristics, m e t e o r o l o g y etc. , m u s t be developed if w e are to be able to respond to an e m e r g e n c y in a very short time. It is our goal to m a k e a decision whether or not to take action within several minutes. At present this is not possible, but we are working in that direction. M. DELPLA: As a doctor I should like to make a comment concerning your reference to the 1968 I C R P publication. This publication m a k e s the assumption that the effect is proportional to the dose. F o r strontium, a bone-seeking element, this effect is m a n i fested as cancer in the bone tissue, or as leukaemia, due to irradiation of the bone marrow. Various publications, particularly since 1968, indicate the existence of a practical carcinogenic threshold for bone tissue. This threshold gets higher according to the sequence: radium-224 and plutonium-239; radium-226; and strontium-90. However, radium is not produced by nuclear power plants. I should like to ask whether the rules applicable in the United K i n g d o m for nuclear accidents take into consideration the existence of a practical carcinogenic threshold. Pamela M. BRY A N T : The existing rules in the United Kingdom are based firmly on I CRP principles. J. V E N N A R T : The Medical Research Council will almost certainly adopt the relationships suggested by its Panel and described by Miss Bryant IAEA-SM-180/12 469 for the doses to critical tissues per unit concentration of the various radio nuclides in air and in milk. The Council has not yet decided on the limits for the E m e r g e n c y Reference Levels of Do s e and will do so only after consideration of the following and other related factors: (a) The risks from the doses for the individuals concerned and their future offspring; (b) The numbers of people who might be involved; and (c) The risks of any countermeasures such as the evacuation of people from the region concerned. M. B. BILES (Chairman): Does this imply that you might be placing an absolute m a n - r e m limit on dose exposure for emergencies? J. VENNART: Not necessarily. M . D E L P L A : I should like to point out once again the dangers of using m a n - rems in this context. The m a n - re m concept recently introduced in radiation protection is based on the hypothesis that the effect is proportional to the dose. T h u s the n u m b e r of m a n * re m s is equal to the product of the number of persons (in 'men') and the dose (in rem). A very low dose applied to a large number of persons gives a number of m a n - rems which can be very high — increasing with the number of persons. B u t w e m u s t beware; this can lead to extraordinary and d ramatic results which might cause regrettable decisions to be taken, such as the evacuation of a town! However, the m a n - rem has never produced a detectable effect either on an individual or on a population. Thus, I urge you all not to waltz about lightly with m a n - re m s . E v e n in a situation where there was no real danger, the whirl might culminate in a dance of death. P. P E LLERIN: I should like to congratulate our British friends once m o r e on their realism and the courage with which they take sensible decisions which serve to reassure the public and put an end to the inter minable disputes about the hypothetical effects of small doses. Miss Bryant's paper demonstrates this admirably. In Fr a n c e w e have already taken s o m e steps in the s a m e direction, in that we no w discount all doses below the recordingthreshold of personal dosimeters. A s for the ' m a n - rem', I should like to repeat that it is psychologically dangerous to use this unit in dealings with the public, because it is open to false interpretation. Finally, one should be very prudent in the use of the term 'lethal dose'. In this regard it would be far better to absorb three times the so-called lethal dose of plutonium (which might cause leukaemia or cancer in 15 or 20 years) than to take a single lethal dose of potassium cyanide, because the latter would cause death for certain. J. M. MAT U S Z E K : The remarks about the incongruities which m ay result from multiplying small dose values by large numbers of people to derive a 'man - r e m dose' in emergency situations can also be extended to routine situations. The United States Environmental Protection Agency, in preparing n e w drinking-water guidelines, at first suggested a 3000 ma n - r e m per year limit for public water supplies. Such a low limit would have resulted in doses from the N e w York City water supply exceeding the m a n - re m limit by a factor of mor e than 30 for ^&Ra alone. W h e n one con siders that the 2 %Ra concentrations are only 0. 1 pCi/litre, such a limit is obviously incorrect. The m a n - r e m limit is now being revised to allow water supplies involving very large population groups to have a higher limit. 470 BRYANT М. В. BILES (Chairman): This discussion reinforces m y belief that the establishment of individual facility action levels is the only possible approach to this problem. It is apparent that w e cannot establish general action levels covering every type of facility and population distribution. CHAIRMEN OF SESSIONS Session 1 B. LINDE LL S w e d e n Session 11 P. PELLERIN France Session III J. SCHWIBACH Federal Republic of Germany Session IV Rebeca M. de NULMAN M e x i c o Session V M.B. BILES United States of America Session VI J. C H Y S K Y Czechoslovak Socialist Republic Session VII P.R. KAMATH India Session VIII J. P E N S K O Poland Session IX J.R. BEATTIE United Kingdom SECRETARIAT Scientific G . E . S W I N D E L L Division of Nuclear Safety Secretary and Environmental Protection, IAEA Administrative K. DALLIBOR Division of External Secretary Relations, IAEA Editor D.F. NASTASIA-SCOTTI Division of Publications, IAEA Records Officer P.B. SMITH Division of Languages and Policity-Making Organs, IAEA Liaison Officer J. PENSKO Government of Poland 471 HOW TO ORDER tAEA PUBUCATtONS Exclusive sales agents for IAEA publications, to whom all orders and inquiries should be addressed, have been appointed in the following countries. UNITED KINGDOM Her Majesty's Stationery Office, P.O. Box 569, London SE 1 9NH UNITED STATES OF AMERICA UN!PUB, inc., P.O. Box 433, Murray HiH Station, New York, N.Y. 10016 In the following countries IAEA publications may be purchased from the sales agents or booksellers listed or through your major local booksellers. Payment can be made in local currency or with UNESCO coupons. AUSTRAHA Hunter Pubtications, 58 A GippsStreet, CoHingwood, Victoria 3066 BELG4JM Office )nternationa) de Librairie, 30, avenue Marnix, B-1050 Brussets CANADA ! nformation Canada, 171 Slater Street, Ottawa. Ont. К 1 A OS 9 C.S.S.R. S.N.T.L., Spáiená 51, CS-11000 Prague A!fa, PubHshers, Hurbanovo námestie 6, CS-80000 Bratistava FRANCE Office tnternationa) de Documentation et Librairie, 48, rue Gay-Lussac, F-75005 Paris P.O. Box 149, H 1011 Budapest 62 )ND!A Oxford Book and Stationery Comp., 17, Park Street, Caicutta 16 ¡SRAEL HeHiger and Co., 3, Nathan Strauss Str., Jerusatem )TALY Librería Scientifica, Dott. de Biasio Lucio "aeiou", Via Meravig)i 16, ¡-20123 Mi)an JAPAN Maruzen Company, Ltd., P.O.Box 5050, 100-31 Tokyo tnternationa! NETHERLANDS Marinus Nijhoff N.V., Lange Voorhout 9-11, P.O. Box 269, The Hague PAKtSTAN Mirza Book Agency, 65, The Mat), P.O.Box 729, Lahore-3 POLAND Ars Potona, Centrata Handtu Zagranicznego, Krakowskie Przedmiescie 7, Warsaw ROMANtA Cartimex, 3-5 13 Decembrie Street, P.O.Box 134-135, Bucarest SOUTH APRiCA Van Schaik's Bookstore, P.O.Box 724, Pretoria Universitas Books (Pty) Ltd., P.O.Box 1557, Pretoria SPA!N Nautrónica, S.A., Pérez Ayuso 16, Madrid-2 SWEDEN C.E. Fritzes Kung). Hovbokhande), Fredsgatan 2, S-10307 Stockhoim U.S.S.R. Mezhdunarodnaya Kniga, Smotenskaya-Sennaya 32-34, Moscow G-200 YUGOSLAVtA Jugos!ovenska Knjiga, Terazije 27, YU-11000 Betgrade Orders from countries where sates agents have not yet been appointed and requests for information should be addressed directly to: Publishing Section, Internationa) Atomic Energy Agency, Kärntner Ring 11, P.O.Box 590, A-1011 Vienna, Austria INTERNATIONAL ATOMIC ENERGY AGENCY VIENNA, 1974 PRICE: US $23.00 SUBJECT GROUPHI Nuclear Safety and Environmental Protection/ [i960; F.Fr.42; DM58,-) Radiological Safety