DEADLY CROP IN THE TANK FARM

An Assessment of the Management of High-Level Radioactive Wastes in the Savannah River Plant Tank Farm, Based on Official Documents

Arjun Makhijani, Ph.D.* Robert AIvarez** Brent Blackwelder, Ph.D.***

* Associate Professor, Capitol Institute of Technology ** Director, Nuclear Weapons and Power Project, Environmental Policy Institute *** Vice-President, Environmental Policy Institute

Corrections were made to several pages, in January 2012. Changes to pages 27 and 29 are described at http://www.ieer.org/errata.html.

TABLE OF CONTENTS

Preface, ...... i 1 . Summary and Recommendations ...... 1 Part I 2. Site Characteristics ...... 8 3 . Introduction to Tank Farm Operation ...... 16 4 . Technical Problems ...... 33 5 . Catastrophic Events ...... 55 6. Environmental Contamination...... 71 7 . Worker Exposures...... 100

Part I1 Tables Compiled From The Savannah River Plant 200 Area Fault Tree Data Bank PREFACE

Although much attention over Union where U.S. intelligcncc sources the past decade has been focused on and exiled Sovict scientists suggest an accidental large-scale releases of explosion in 1957 scverely contaminated radioactivity from commercial nuclear several hundred square miles and reactors, a similar potential exists at resulted in a major loss of life. A federal nuclear facilities. In recent potential for similar expIosive years, the risks and consequences of accidents or loss of containment from major radiation accidents as well as the earthquakes also exists at the Savannah health and environmental effects from River Plant site. the routine operations of the facilities The tragic accidental release of owned by the U.S. Department of Energy methyl isocyanate at a Union Carbide (DOE) have also become the subject of plant in Bhopal, India which killed more increasing concern, One principal reason than 2,500 people and injured tens of is that the various DOE installations thousands with long term effects have now been operating for two to four underlines the need to pay much more decades, and the evidence of adverse attention to events which may have a health effects from exposure to small probability of occurring but which radioactivity is now accumulating at an are disastrous if they do occur. alarming rate. This applies to various Initiated in 1981, this study is an exposed populations such as workers at independent evaluation of the management DOE facilities, the veterans of atomic of the high-level radioactive wastes in tests and the people who lived in the the Tank Farm at the Savannah River path of the fallout from these tests. Plant (SRP). These wastes come primarily There also exists the potential for from the manufacture of plutonium for serious damage from large releases of nuclear warheads. Plutonium is produced radioactivity resulting from operating in nuclear material production reactors accidents or from events outside the and extracted at SRP's two reprocessing control of DOE, such as earthquakes. plants. Very large quantities of radionuclides Our study is based upon that part of in liquid form are present at some DOE the official record which has been made facilities. In particular, the public (some of it through Freedom of facilities at the Savannah River Plant Information Act requests filed by the in South Carolina and at Hanford in Environmental Policy Institute in 1981). Washington state, contain very large In all over 14,000 summaries of reported quantities of radionuclides in liquid accidents, worker exposures, spills, form, which are particularly dangerous equipment failures and non-routine because of their mobility. Large maintenance at the SRP Tank Farm were quantities of toxic non-radioactive made available to us for analysis. wastes also exist at these sites. The However, a great deal more data, terrible actuality of a massive accident particularly technical monthly reports, involving nuclear high-level radioactive and more detailed measurements of worker wastes has already been experienced in exposures and environmental the Ural mountain region of the Soviet contamination is still held secret by the U.S. Department of Energy and its calculations of failure probabilities contractor, E.I. du Pont dc Ncmours and and thc statistical model used by DOE Company. This secrecy, which we believe and DuPont for risk estimates. Robcrt for the most part is not justified, Alexander, Chief of the Occupational means that limitations have been put on Standards Branch of the Nuclear . our research. Nonetheless, many Regulatory commission*^ Office of conclusions emerge rather clearly, and Research and his staff, reviewed Chapter we have tried to delineate these and 7 which deals with occupational other issues as best the available exposures. Dr. Alice M Stewart, MD, information permits. Senior Research Fellow at the University The study comes not only in the of Birmingham School of Social Medicine context of a heightened concern about in England reviewed various DOE nuclear facilities, but also at a epidemiological studies pertaining to time when the public and Congress are the Savannah River Plant and- offered considering renewal of a federal law advice on their interpretation. Dr. Karl known as the PricelAnderson Act. 2. Morgan, former Director of Health Enacted by the U.S. Congress in 1957, Physics at Oak Ridge National and renewed every ten years, the Laboratory, reviewed Chapter 6 and 7. Price/.Anderson Act is designed to deal William Lawless, professor of with major nuclear accidents occurring Mathematics at Payne College, Augusta, at federal and civilian nuclear Georgia, and former senior engineer for facilities. The principal mechanism in the Department of Energy at SRP, the law is a ceiling on the total limit reviewed the entire report and made of liability that private utilities and several helpful comments and criticisms. the federal government can bear. Under Yaaron Sternberg, Professor of Civil the law, some 280 DOE nuclear facilities Engineering, University of Maryland, are covered with the liability limit set reviewed sections of Chapter 6 at $500 million, DOE contractors are pertaining to hydrology at SRP. Bernd completely shielded from liabilities Franke, Senior Research Fellow at the arising from a major nuclear accident. Institute for Energy and Environmental We hope that the data and analysis Research in Heidelberg, West Germany, presented in this study will be a reviewed Chapter 6. Dr. Uta Boiket of contribution to the public debate on the University of Bremen in West Germany this question which is not only also reviewed Chapter 6 and made important for our generation, but for comments on soil contamination. David future generations at well. Afbright, research fellow at the We would like to thank the people Federation of American Scientists, who reviewed our report and provided reviewed portions of the Fault Tree Data usef ul criticisms and comments. Dr. Bank on occupational exposures and Peter Bickel, Professor of Statistics helped derive dose estimates for and Divisional Dean of the College of plutonium depositions. Dr. Roland Arts and Sciences, University of Cali- Finston, Health Physics officer at fornia, Berkeley, reviewed statistical Stanford University (California) portions of-the end of Chapter 3 and the provided advice on estimating cancer portion of Chapter 5 dealing with risks. The conclusions and recommendations of this report are ours alone and not necessarily those of the reviewers. Their comments were vital in improving the report and we are deeply indebted to them. We would also like to thank Jackie Williams and Diana Kohn, who helped type, edit and prepare the the report for publication. Finally, we would like to express our thanks to the Mary Reynolds Babcock Foundation whose financial support made this report possible, and to Frances Close Hart who helped encourage its initiation and final completion. Washington, D.C. July 1986 Arjun Makhijani Robert Alvarez Brent Blackwelder CHAPTER ONE SUhlhlARY AND RECOMMENDATIONS

Eight hundred million curies of the major equipment--such as the deadly high-level radioactive wastes are reactors and reprocessing plants, as stored in the Savannah River Plant well as many of the tanks, date back to (SRP). Although 27 million gallons of the 1950s. This is a field in which these wastes constitute about one third technological change and safety of the total volume of military standards have changed rapidly. Yet in high-level radioactive wastes in the recent years the basic approach to waste U.S., they contain about 78 percent of management at the Savannah River Plant the total radioactivity in all U.S. Tank Farm has changed but little. military high-level wastes. SRP's In fact, the operating record of the high-level wastes pose a serious threat obsolete facility shows that its very to the plant's workers, to the people design basis was faulty and dangerous. who live in substantial portions of There is also substantial South Carolina and Georgia, to future evidence that these problems have been generations and to the environment. The compounded by unsatisfactory management rates of radiation=-related cancers among in many areas crucial to safety. Both workers are already significantly higher DuPont and DOE appear to be more anxious than expected. The plant site borders to minimize any adverse consequences and the Savannah River and sits atop the thus allay public fears than to address Tuscaloosa aquifer, one of the most operating problems and risks from prolific and used sources of fresh water accidents in a scientific and in the eastern United States. The 300 technically responsible manner. square mile site and the shallow We will summarize the issues aquifers above the Tuscaloosa are so discussed in detail in the main body of severely contaminated that it is the report (where the references art reasonable to conclude that it has been provided) under the following four treated by the federal government as a headings: national sacrifice area for the US. nuclear weapons prbgram. o Routine Environmental Contamination The high-level radioactive o Accidents and Risks wastes which continue to build up at the o Worker Exposures and Cancer Risks Savannah River Plant result from the o Long-Term Waste Management production of radionuclides for the US. nuclear weapons program, In particular most of the wastes come from the production of plutonium in nuclear "-Severe contamination in the upper reactors and the subsequent reprocessing aquifer (at SRP) poses an imminent of the reactor fuel rods in chemical threat to a deeper aquifer that separations plants. The SRP is owned by supplies drinking water to plant the U.S. Department of Energy (DOE) and employees and off-site communities." operated under contract by E.I. du Pont Environmental Protection Agency de Nemours and Company (DuPont). Most of October 1983 "...hnrardorrs srcr fnce thc toxic wastes in cardboard boxes and cortcerr~rariorts (of nrercrcry) nray drums (until 1984 a common practice at exist rvell beyortd the period over SRP) remains to be seen. which larrd corrtrol cart be The problem of clean-up is arrticipated and indeed nray exist bound to be scvcrely complicated by a cerrturies or meltenia into rhe lack of data and unreliability of such future ...' data as there is. In regard to I~iternal Savannah River non-radioactive toxic materials other Laboratory Report. than mercury, such as PCB's and organic solvents, hardly any data exist. Even The design of the Savannah the data on high-level waste contain River Plant assumed that radioactive serious uncertainties, notably in wastes could be routinely discharged relation to plutonium content of the into the soil because the soil would waste. trap them and prevent them from We have two widely differing contaminating wa ter supplies, estimates for plutonium in the particularly in the case of some of the high-level waste tanks. Data in DuPont more deadly materials like plutonium and Safety Analysis Report (issued in 1978) cesium-1 37. Little thought appears to yields, an estimate of about 170 have been given to pollutants from kilograms of plutonium containing non-radioactive toxic materials and less 300,000 curies of radioactivity, for to interactions between the two kinds of 1980. However, in 1980 DuPont supplies pollution. an estimate of 1 million curies of Time has shown both the design plutonium - about 400 kilograms - to the premise and the omissions to be serious National Academy of Sciences. This errors. Radioactive materials and enormous discrepancy, serious both for non-radioactive toxics have contaminated plutonium accounting from the security the shallow aquifers beneath SRP. point of view, and for its potential Interaction between solvents and environmental consequences, is plutonium has caused it to migrate into unexplained at least in the public the groundwater in twenty years - record. compared to a predicted time of hundreds Finally, despite the evidence, of thousands of years. Despite repeated and the judgment of the Environmcntal internal and external efforts to stop Protection Agency, DOE and DuPont these dangerous and technically obsolete continue to operate the plant as if and erroneous practices. DuPont and DOE there is no danger to the deeper continue routine discharges of toxic aquifers which are used by the public. materials into the soil. Indeed, the official plans to start up If the SRP site is not to become the L-reactor to increase plutonium a permanent national sacrifice area, a production call for considerably greater massive clean-up of the site will be water use, which could further increase required. DOE has estimated that the the likelihood of contamination of the first seepage basin cleanup would require vital Tuscaloosa aquifer. a billion dollars. Whether this will consist of something more than putting Accidents and Risks the U.S. Geological Survey. DOE and DuPont also assume that water use Some of the most technically patterns and many others factors will not difficult aspects of waste management at change significantly for a hundred years SRP have to do with the problems or more. This is not merely arbitrary; associated with accidents and risks to it is contrary to evidence. Water use the public and to future generations. patterns have changed immensely in the Both in relation to accidents that have past few decades with DOE and irrigation occurred and the risks from potential being major contributors to that accidents, DOE and DuPont discount as change. Indeed, in other reports, DOE "insignificant" problems related to plans on continuing to contribute to groundwater contamination, using significant increases in water use. scientifically flawed assumptions and The data on hydrogen methods. The data base has been concentrations are also contradictory. A arbitrarily kept and is of' little 1978 public Safetv Analvsis Rebort does statistical validity when it comes to not cite any explosion in the tanks some of the most crucial accidents. resulting from a hydrogen build-up. The Consider, for example, the computerized mta Bank. cites one question of hydrogen build-up in the explosion in Tank 6 due to a build-up of high-level waste tanks. Hydrogen gas is hydrogen to only 15% of the lower generated in the tanks due to the action explosive limit-that is to an amount far of radiation on hydrogen-containing less than the minimum required for an chemical compounds in the waste. A explosion, according to official build-up of hydrogen to high enough estimate. As another example of levels, due to partial or total failure defective data, the Data Rwrecords of tank ventilation systems for example, a maximum hydrogen conceration of 150% could cause an explosion severe enough of the lower explosive limit in one to destroy the tank and send millions of instance. But the offical Wetv curies of radioactive waste spewing into Analvsis Rebort implies a maximum of the air and onto the land, Such an only 100%. accident could cause up to 20,000 cancer In other areas also data casts in addition to genetic damage and management has been poor. The evidence other ill-health effects. In addition, a is that, tens of thousands of very large area of land would have to be non-routine maintenance problems and written off essentially forever. It equipment failures have been nbitrarilv would also have unpredictable omitted from the record of data. In repercussions, possibly very severe, for turn, this data has been used to groundwater contamination. estimate failure probabilities of The DOE and DuPont approach to components. This is only one example such accidents irresponsibly assumes among many of serious deficiencies and that groundwater contamination can be omissions, all of which downplay ignored as "insignificant" because the systematically the risk from accidents soil will retain the radioactive wastes. to which the public is being exposed. As This assumption has been shown to be noted above, even the estimates of the invalid by SRP's own operating inventory of plutonium in the waste experience and has been criticized by appears to vary widely. Besides the dangers from con tamination. These doses averaged about operating accidents and design and 260 mrcms pcr year - about 150% more construction problems (for example, tank thandhe doses received off-site from and pipe leaks), there is also the background sources. In addition, various danger of earthquakes. SRP was not typcs of work involve additional designed to withstand severe exposures. earthquakes. In the last few years, Plant data show that the work however, both the U.S. Geological Survey in the reprocessing and waste management and the Nuclear Reguta tory Commission is especially hazardous in this regard. have concluded that severe earthquakes, The greater dangers arise both from the comparable to the one in Mexico City in nature of the work involving highly 1985, cannot be ruled out. The Nuclear radioactive materials and from the Regulatory Commission has criticized the design of the plant. The total recorded SRP assumptions of moderate earthquakes exposure to external radiation to at most, to "contain a strong clement of worlgers from 1954-78 was 50,000 Person-' speculation." rem. The waste and reprocessing area A severe earthquake could workers constitute one-third of the cause hundreds of millions of curies of workforce but received more than half radioactive wastes to contaminate the this dose. air, soil, and water of the area. Even The external radiation doses to using the non-conservative assumptions SRP workers alone can be expected to of DuPont and DOE, it would cause from cause between 16 and 330 excess cancers 11,000 to 230,000 excess cancers and up among SRP workers, with more than half to 2,500 genetic defects in future of these expected among waste and generations. The direct cost, moderately reprocessing area workers. Already, estimated, would be from $800 million to there are definite indications at SRP $14 billion - excluding the cost arising and at other DOE owned nuclear from writing off of large areas of land, facilities around the country that from contamination of water supplies, workers are contracting and dying from from property, agricultural, and radiation related cancers. Some business losses. examples: Current law specifies maximum DOE liability as $500 million. DuPont is o At SRP, the incidence of myloid exempt from liability to the public in leukemia has been more than the event of accidents, earthquakes, and double the expected number (6 other catastrophic events. It exempts occurred versus less than 3 contractors from liabilities arising expected); even from their own negligence. o At the DOE owned Oak Ridge Worker ExDoSUT~Sand Cancer Gaseous Diffusion Plant, a study found "excess deaths due to lung Workers at SRP receive and brain cancers and respiratory considerable doses of radiation just by disease..." being on the site because of routine emissions and radiation from site

* When radiation doses are measured for large populations, the unit 'person-rem' is used. This is calculated by multiplying the total number of people exposed times their average dose in rems. Or, it can be the actual sum of all doses they receive. o A study of 2,509 DOE workers In the early years, it was simply exposed to more than 5 rems assumed, without significant geologic or between 1947 and 1978 showed a other systematic scientif ic rate of cancer of the rectum at investigation, that the wastes could be three times the national average safely pumped into the bedrock among them. underneath the plant site, and below the much used Tuscaloosa aquifer. Pending The emergence of an alarming such long-term disposal, it was decided pattern of excess cancers has elicited a to store the wastes in carbon-steel curious response from DuPont. An tanks which were much cheaper than internal 1976 study by DuPont found stainless steel tanks. However, this "evidence ... that lung cancer and required the neutralization of the leukemia were significantly highly acidic wastes discharged from the increased. .." among workers. Instead of SRP reprocessing plants, so that the publishing the study, DuPont attempted acid would not corrode the carbon-steel. to erase the significance of its This created a much larger volume of findings through statistical waste, including sludge which is manipulations. Even an advisory difficult to handle. committee to DOE found these Eight of the first sixteen tanks manipulations "inappropriate," and developed leaks in the primary?- recommended that the data be reanalyzed containment in about a decade. This has by a non-DOE/DuPont group. required much more handling and moving The data themselves are not in of the wastes than planned - which in good shape. One of the principal sources turn causes - more equipment and process of data - the computerized Data Bank for problems, worker exposures and accidents and non-routine maintenance - environmental contamination. is missing thousands to tens of The plan to dispose of the thousands of entries. Moreover, there is wastes into the bedrock under the plant very little data on internal radiation has been abandoned in favor of exposure - through inhalation, ingestion solidifying the wastes by encapsulating and wounds. Most of the reported data is them in glass. Solidification of the gathered by obsolete and discredited liquid wastes is urgently needed. methods. This is a crucial area for However, the current evaluation of safety practices, and glassification plans, which are being liability, since internal exposures are implemented, face some serious problems. emerging as a principal cause of The operating record does not radiation related cancers. The records bode well for the proposed waste of DOE and the Defense Nuclear Agency vitrification facility at SRP called the (Department of Defense) are particularly Defense Waste Processing Facility. This poor in this regard. will require much waste movement and remote operation. If heavy maintenance and repair are required, worker exposures may be increased. Further, Little attention was paid to there is no operating experience even at the problem of long-term waste the pilot plant level for vitrification management when the plant was designed. of radioactive sludge, which has been radioactive wastes be stopped pending the source of considerable handling pilot plant construction and resolution problems. Unanticipated breakdowns or of other issues related to long-term failure of the plant to operate as management. predicted could result in costly dclays in the irnplemcntation of long-tcrm waste General Policy Recommendations rnanagemcnt, while at the same time leaving the wastes in the current o The Department of Energy should not dangerous liquid form. be allowed to continue to regulate DOE also plans to dump very itself or its contractors. large quantities of "low level" wastes, o Current limits on DOE liabilities solidified in concrete, as part of its under the Price/Anderson Act should program. This will increase the be lifted. Further, DOE contractors radioactivity in the low-level burial should be held financially grounds many fold. It almost certainly accountable for major accidents will contaminate the groundwater with stemming from their negligence. Both very much larger quantities of would be great incentives for radionuclides than are already present. safety. In particular, it will increase o Independent studies on various plutonium-238 contamination by about aspects of the plant such as health w. and that by iodine-129 and and safety decommissioning and technicium-99 by sveral rn~lllon. . ti- long-term disposition of the site over the amounts that have been should be initiated. All documents discharged already as "low-level" wastes relating to these matters should be into the seepage basins. made public. Moreover, two of the three sites picked by DOE, for the first Technical Recommendations repository for long-term d3sposai of high-level radioactive wastes have o The Savannah River Plant should be characteristics which may rapidly barred from producing any more destroy glass. high-level liquid radioactive waste At Hanford the relatively high until the long-term questions are water velocity in the repository could satisfactorily resolved. erode the glass much faster than o Interim solidification, such as required by current regulations. At the calcining of the existing high-level Nevada Test Site, low pressures could wastes (including sludge) should be cause steam to form around the glass, done while setting up any pilot with the same result. In either case, plant efforts for long-term failure of the glass could cause large management. quantities of radioactive materials to o Research and development of remote be released into the environment. equipment and working methods to For these reasons we recommend protect workers from radiation and that solidification of the wastes by other hazards should be broadened calcining to be done immediately and and intensified. further generation of liquid high-level o The practice of using soil as a disposal medium and surface and groundwater as transportation mediums of toxic materials (radioactive and non-radioactive) should be stopped. All .the seepage basins should be shut down on an expedited basis. (Some are now due to be shut down in 1988.) o An urgent program to clean up contaminated aquifers should be initiated. CHAPTER 2 SITE CHARACTERISTICS The Savannah River Plant near Aiken, Chemical Separations or 200 Area South Carolina, occupies a nearly Two spent reactor fuel reprocessing octagonal area of 200,646 acres or about plants are in the 200 Area (F and H). 300 square miles on the coastal plain of Once spent fuel elements are chemically South Carolina bordering the Savannah separated at the F and H facilities, River. The tract extends over portions low-concentration radioactive and of Aiken, Barnwell and Allendale non-radioactive waste products are counties. (See Figure 2.1) discharged into a series of seepage Owned by the Department of Energy basins. There are three in operation in and operated under contract by E.I. du the F Area and four in the H Area.(3) Pont de Nemours and Company, the Other low-level and transuranic wastes Savannah River Plant is the principal are stored in shallow burial pits producer of radioisotopes (plutonium and making-up about 195 acres known as the tritium) for the nuclear weapons Burial Ground.(4) High-level wastes as program. The plant site was picked in well as some 'Low Activity" wastes (See 1950 by the Atomic Energy Commission in Chapter 3) from the reprocessing plants the wake of the decision by President are stored in a series of tanks, known Truman to manufacture thermonuc1ear as the Tank Farm. There are 51 waste weapons.(]) tanks in the 200 Area. The 200-Area also In December 1953, the first contains a tritium processing facility. production reactor began operation and the following year the first warhead Fuel and Target Fabrication materials were chemically separated and Facilftles or 300 M Area sent of f-site. Uranium fuel elements for SRP's Currently, the Savannah River Plant production reactors are fabricated at operation employs about 8,000 workers this site. This 300 M area contains one and comprises five areas. They are: seepage basin which has received industrial solvents and uranium wastes. Reactor or 100 Area Since 1953 SRP has had five heavy A Heavy Water Extraction Plant water pressurized ' reactors operating or the 400 Area over the following periods: This facility was started up in 1952 R-Reactor 12/1953 - 6/1964 and has provided deuterium or heavy (placed on standby) water which serves as the neutron P-Reactor 2/1954 - present moderator for the production reactors. L-Reactor 7/1954 - 2/1968 (may be restarted) The Savannah River Laboratory K-Reactor 11/1954 - present or 700 Area C-Reactor 3/1955 - present The Savannah River Laboratory has The 100 Area reactors discharge three test reactors and provides radioactive and non-radioactive liquids- supportive research for the plant's into a series of seepage basins,(2) of activities. The laboratory also has which two are currently in operation. four seepage basins which receive site. Discharges from these formations radioactive and non-radioactive occur into plant streams or into the wastes.(S) Savannah River. Beneath these shallow formations are CAIX-TNX Area the Ellenton and Tuscaloosa formations. This facility performs experimental The two formations have a combined research in support of the plant's thickness of about 250 meters, and operations and discharges radioactive consist mainly of highly impermeable and non-radioactive wastes into two sands which yield large amounts of water seepage basins. of high quality. Beneath the Tuscaloosa and Ellenton formations is bedrock, Proximity to Population Centers which has been considered for many years Approximately 700,000 people live as a potential storage area for SRP's within a sixty-mile radius, from the high-level radioactive wastes.(7) center of the Savannah River Plant. The Tuscaloosa and Ellenton Major population centers near the plant formations are thought to be separate include: Augusta, Georgia (20 miles and distinct from each other. The northwest), Atlanta, Georgia (155 miles recharge of these formations is about 18 to the west and north), Columbia, South miles north and east of the SRP site. Carolina (65 miles northeast), and ' They discharge into the Savannah River Savannah, Georgia (85 miles southeast). near Augusta, Georgia. Although nearby (See Figure 2.2) communities draw their water from these formations where they are closer to the Geology surface, SRP withdraws the largest About 80 percent of the SRP area amounts for its purposes. In recent lies in the Aiken Plateau, a level plain years, however. irrigation in the extensively eroded by surface streams counties near SRP have been responnsible with the remainder as alluvial terraces for significant withdrawals. which are adjacent to the Savannah The deeper portions of the River. The plant complex itself ranges Tuscaloosa aquifer extend into Georgia, between 300 and 385 feet in northern Florida and eastern Alabama. It elevation.(6) is used extensively throughout the The geologic profile of the SRP site coastal plain for drinking, agricultural consists of an overburden with an and industrial purposes. (see fig. 2.4) average thickness of 300 meters and The bedrock beneath the Tuscaloosa consists of six layered unconsolidated and Ellenton formations is composed of sedimentary formations interbedded with two substances, crystalline and triassic thin layers of clay. (See Figure 2.3) rock. The hydrogeology of the bedrock The uppermost sediments comprise about relative to the aquifers above has been 60 meters and contain the Hawthorne, studied extensively, but according to Barnwell, McBean and Congaree the National Academy of Sciences in 1981 formations. All except the Hawthorne "their relationships to the overlying yield water. Water in these shallow aquifer are still not certainT(8) aquifers is recharged by the percolation Adding to the uncertainty are of rainwater. It outcrops on the plant assumptions that the formations along the coastal plain may be vulnerable to average annual humidity is 70%. Rainfall major earthquakes of the magnitude of averages 47 inches a year. the Charleston earthquake of 1886 Occasionally the area is subjected (Modified Mercali Scale X). SRP is to severe storms in the forms of considered to be in a seismic risk zone hurricanes and tornados. capable of experiencing ma jar damage. (See Figure 2.5 An earthquake of lcsscr magnitude is capable of References dramatically changing the hydrogeology of the area, as witnessed recently in 1. E.I. du Pont de Nemours and Idaho. Company. S-t I n construction Historv, Vol. 1, DPES 1403; Surface Hydrology January 1957. Six small streams on the plant site 2. Marter, W.L., "New Criteria for flow into the Savannah ~iverto form the Seepage Use." DPST-77-44. September surface drainage system. Five streams 1977. flow diagonally across the area towards 3, Ibid. the southwest and are used to dispose of 4. Fenimore, J.W., "The Burial radioactive and non-radioactive Ground as a Containment System: 25 Years discharges from SRP facilities, Since of Subsurface Monitoring at the Savannah the surface of the SRP site gently River Facility," DPST-82-725, June 1982. slopes toward the sea, the five streams 5. Op. Cit. Reference 2. descend 100 to 200 feet before 6. Final Environmental Impact discharging into the Savannah River. Statement. Corm-term Manament of Additionally there are over 50 tive Wastes; artificial impoundments covering about ;enAikcn. S.C, 3000 acres, the largest of which is Par November 1979. Pond (2700 acres). Water flowing to the 7. Radiaactive Waste mment Savannah River is held intermittently in $he Savannah River Plant: A Technical marshes and over 200 naturally occurring Review. NAS-NRC. December 1981. basins known as "Carolina Bays." A large 8. Ibid. swamp borders the Savannah River as it 9. Op. Cit. Reference 6, runs past the plant site. The Savannah 'River is used for fishing, boating and drinking. Approximately 70,000 people downstream from SRP rely on the river for drinking wa ter.(9) Climate and Meteorology The climate in the vicinity of the Savannah River Plant is characterized as having mild winters and long summers with temperatures averaging 48 F in the winter and 85 F in the summer. The c*(LLC ~*rur(C.C.&re~~,m~ImAawlil . m*nn M tw)mTmlwm 0 ~mnrRawtan * Irrmhb-mmd -Ivw FIGURE 2.1 Ct c8arushap rn'ylr

Source: DOEIEIS-0108b voll p.3.3 11 L 4

NORTH CAROLINA

a

0

FIGURE 23

SRP location in relation to surrounding population centers.

Source: DOE/EIS-0108d voll p.3.2 12 Sea Level

0 5 10 15 kilometers

0 Hawthorn McBean Ellenton Tuscalwsa

Tenlery rucks Dunbarton undlfferentlated T~I~SSIC rock Basln

FIGURE 2.3

Generalized northwest to southeast geologic profile act088 the Savanndh Rivet Plant.

Source: DOE/EIS-0108b voll p.3-14 13 Cretaceous Aquifer Tertiary Sand Aquifer Floridian Aquifer

FIGURE 2.4

Sources: United States Geologic Survey Water Supply Paper 2275, (National Water Summary 1984) Draft Environmental 1mpact Statement L-Reactor Operation Savannah River Plant, DOE/EIS-0108D SUSMIC RlSK lUAP FOR THE UNmD STATES. The contlplou, United States may be dgned to risk categories u indicated in the flgute. Tbme a~oriercornrpond to krtcadde, on the Modified MedIntensity Sale, u indicated (Figure o*: "United States EYtbqurke, 1968," US, Dapurmtnt of Commerce, U.S. Corrt rad Gdetfc Survsy, 1970.)

ZONE 1 - Minor damage: dWmt urMqwko may cruse amage to sVuCtm with fund.nnrrtrl mnagnrter than 1.0 mcond8: emponds to intrnsitirr V and VI of ma M.M? Sale. ZONE 2 - Mouent, bunrgw correspond8 to intmdg VII of ltw M.M.. Sale.

FIGURE 2.5

+ Sourc8t Naro, A.V.,-A Guidabook to Nuclear Rea~tots,~University of California -Press (Berkeley and Los Angelas), 1979, Pig. 4-2, p. 57. 15 CHAPTER 3 INTRODUCTION TO THE TANK FARM OPERATION

The extraction of plutonium, the wastes is similar except in two and uranium from irradiated fuel rods respects. results in large volumes of liquid First, the high level waste must be wastes which contain the waste products allowed to decay for a year or more from the fission of uranium-235 and ("aging" of the waste) to decrease the plutonium. The wastes also inevitably radioactivity and heat content before contain some plutonium, uranium and further treatment. The "low activity" neptunium since they cannot' be waste is not "aged." Second, "low compl~tely extracted. These liquid activity" waste is stored in uncooled, wastes are highly radioactive and highly single-wall tanks, while most high-level acidic. They are classified into waste is stored in cooled, double-wall "high-level waste" and "low-levelw or tanks. Figure 3.1 shows the treatment of "low activity wastesw*according to the wastes in the reprocessing "canyons" amount of heat they produce per unit before discharge to the pipes that carry volume. The high-level wastes produce them to the tanks. decay heat of 0.5 to 5 BTU/hr./gal.** Large quantities of other The "low-activity" waste contains 3 to 5 radioactive wastes are generated at the orders of magnitude less fission Savannah River Plant. Principal among products per unit volume than high-level these is tritium which is released to waste. The variety of radioactive the atmosphere, to the seepage basins, materials contained in each type of and to the Burial Ground. waste is the same and the handling of

*The official papers and documents contain conflicting uses of the term "low-levelw waste. In presentations to the National Academy of Sciences in 1969, "low-level" waste was defined as above. However, the report on waste management DPSTSA-200-3, of the E.I. du Pont de Nemours and Company, refers to "low-level liquid wasteswas those which are directly discarded to the environment. It refers to the dilute wastes from the secondary recovery cycles which are processed in the Tank Farm as "Low Activity Wastesw.We will use the designation "low activity wasten in quotes in this report, for these latter wastes, and "low-levelw wastes in quotes for liquid wastes discharged to the seepage basins or to other areas of the environment directly. **One Btu, or British thermal unit, is the amount of heat required to raise the temperature of 1 pound of water by 1 degree Farenheit. Figure 3.2 shows a schcmatic diagram with it. Most of this entrained cesium of the handling of the high-level waste is rcmoved in a zeolite ion-exchange in the Tank Farm. The waste pipe column. Then the vapor (callcd ("header") carries the waste by gravity "overheads") is condensed and sent to flow from the reprocessing area to a the seepage basin. These discharges to "diversion box." This contains the the seepage basin contain some piping enabling the selection of the long-lived radionuclides, including tank to which the waste is first routed. cesium-137 and cesium-135, as is The routing of the waste has turned out discussed in Chapter 6. The evaporation to be a complicated and dangerous process substantially reduces the volume process, as we shall see. The selected of radioactive waste stored in the tank is filled with the waste, which is tanks. known as "high heat wasten at this The concentrated solution is sent to stage. It is allowed to "age* for at a tank for further settling, followed by least one year in this tank. During this further evaporation. The salts in the period the relatively short half-life concentrate gradually precipitate out of radionuclides like zirconium-95 decay it in the tanks where they form complex consiaerably. Also during this period irregular structures. (5) The most of the radionuclides, with the concentrate-containing tanks thus major exception of the cesium isotopes, gradually become filled with salt. The settle to the bottom of the tank and radioactivity in these tanks consists form a fairly distinct layer of mainly of cesium-137. Finally, when the "sludge." This sludge contains from 80% zeolite in the cesium removal column is of the zirconium95 and ruthenium-106 to depleted, it is "dumped into the waste 98% of the strontium and plutonium tank directly underneath it and replaced isotopes. However, 95% of the cesium with a fresh charge." (6) remains in the liquid above the sludge, In summary, there are five kinds of known as "supernate" (supernatant high level radioactivity waste in the liquid), or in the interstices in the Tank Farm at the Savannah River Plant: sludge. (4) The sludge is only 10% of the volume o "Tresh" high heat waste, of the high-heat waste, but it generates discharged from the reprocessing 50% of the heat since it contains most "canyons," being held for "aging"; of the radioactive materials. The radioactivity in the supernate consists o the sludge, which is about 10% primarily of radio-cesium. But there of the total volume of wastes but are also significant quantities of other contains most of accumulated very long-lived radionuclides, such as radionuclides and generates 50% of the technicium-99 and iodine-129. heat; After this "aging" and settling process, the supernate is pumped out and o the supernatant liquid in sent to another tank. From there it is various stages of concentration pumped to an evaporator where it is containing principally cesium isotopes heated. The water in it evaporates and but also some quantities of other the vapor entrains some cesium isotopes radionuclides; o the salt "cakes" which form in The othcr radionuclides shown the concentrate holding as liquid is in Tablcs 3.3 and 3.4 are also of great cvaporatcd and the remainder bccomes concern, even though the amounts of more concentrated with salts; radioactivity (compared to ccsium-137 and strontium-90) are relatively small, o spent zeolite which has absorbed because of their very long half-lives radiocesium from the evaporator and, in many cases, their very high overhcads which is dumped into tanks toxicity. Further, it takes a relatively directly beneath the cesium removal large quantity of a radionuclide with a columns. 10p.g half-life to yield 1 curie of radioactivity. Thus, the amount of As of 1981 there were in all about stronium-90 in the tanks is about 1,100 27.7 million gallons of high-level kilograms which yields an activity of radioactive waste in the Tank Farm at about 150 million curies. But only 3 the Savannah River Plant. (7) About 10% curies of activity yields about the same of this is in the form of sludge, about weight of uranium-235. 25% is in the form of salt mounds, and There is some question as to the the rest is in (low viscosity) liquid accuracy of these estimates since form. (See Table 3.1) official data appear to yield Tables 3.2 through 3.4 show estimates substantially different values. of the quantities of principal fission The estimates for the actinides like products, long-lived fission products plutonium and americium in Table 3.4 and actinides like uranium and above are based on figures drawn from an plutonium. There were according to one Energy Research and Development estimate a total of about 700 million Administration rebort. This data yields curies in 1981. The estimate for the end a figure of about 300,000 curies for the of 1984 would be about 800 million plutonium-238 inventory for 198 1. This curies extrapolated from an Oak Ridge is also the figure we get using the National Laboratory estimate of 776 data from the 1978 DuPont Safety million curies through 1983.(8) Roughly Analysis Report.(-10) However, the data half of this is in relatively long-lived supplied by DuPont in 1980 to the radionuclides, principally cesium-137 National Academy of Sciences cites a and strontium-90. total inventory of 1 million curies for 1981 or more than three times the above figure.(l 1). These enormous

'Average waste composition .changes only very slowly so that the estimates based on 1978 compositions and 1981 compositions should be in close ,agreement, unless there was a major unreported discharge of plutonium between 1978 and 1981. discrepancies can have serious Savannah River Plant, including the implications for long-term waste Fault Tree Data Bank(l3) into which management as well as for assessrncnts of non-routine noccurrences" are logged in risks posed by the 200 Area Tank Farm chronological order. The official operation in case of catastrophic events record, at least that part of it which like hydrogren explosions or severe is public, is far from complete. It also earthquakes. Table 3.5 summarizes its suffers from a number of other discrepancies for plutonium. deficiencies which are discussed below Most of the high-level waste is and in subsequent chapters. This report contained in cooled tanks which have a is, in many ways, also an assessment of secondary containment partly or fully that public official record. lined with steel.** Due to the The Fault Tree Data Bank, which is requirements of waste processing and, as stored in a computer, is a principal discussed below, due to problcms with source of official information in the tanks, the liquid waste and the assessments of waste operations. It is sludge are moved from one tank to used, for instance, by the Savannah another. There is an extensive network River Plant management to estimate the of underground pipes in each area. The probabilities of breakdowns or routing of waste between tanks is done accidents and the .risk tq . the public through jumper changes in "diversion from such accideirts, Yet it is sorely boxes." Some of the flow of the waste incomplete and uneven both with regard into the tanks is by gravity. The rest to the number of entries and the quality is accomplished with the use of pumps of the data. Sometimes it is also and steam jets. inconsistent with other official Extensive instrumentation is documents, even regarding major necessary to keep track of the wastes, incidents. to detect leaks, contamination, etc. From a statistical point of view, lnstrumcntation is also required to perhaps the most serious deficiency measure radiation levels and to estimate concerns the number of entries. Table worker exposures and releases of 3.6 shows the number of entries in the radionuclides to the environment. Fault Tree Data Bank during various The following chapters provide a periods since the start-up of the preliminary assessment of worker Savannah River Plant in 1953. During the exposures, environmental contamination, 1950s. there were on an average about 4 technical problems and risks due to entries per year. During the 1960s. the catastrophic accidents in the figure jumped to about 55 per year. In radioactive waste. Tank Farm at the the first half of the 1970s (1970-76) it Savannah River Plant, Our assessment is was about 290 per year. Since 1977 there based on the official documents of the

"Type I and -11 tanks have a concrete outer shell lined with steel to a height of 5 feet of th; tanks total height of about 25 feet. Type I11 tanks of later design have a secondary containment which is fully lined with steel, (12) have been about 1800 entries per year -- The arbitrariness of the recording about four hundred and fifty times the procedure until 1965 is revealed by the frequency during the 1950s. following entry dated August 24, 1965, Neither the Department of Energy in the Pata Bonk (Table 11, Part 11, (DOE)nor the management of the Savannah entry 08-24-65H): River Plant (E.I.du Pont de Nemours and Company) have claimed that the rapidly Prior to 1965. inforntntio~t ojr increasing frequency of entries i~urr~m~etrtfailz~re, pimp failirre, corresponds to an increase in problems leaks irt rhe waste rank sysrent on the radioactive waste Tank Farm. Nor are not recorded totless the does the evidence available to us irtdiridtral occtirrelrce is of indicate that this is so. Thus we must parlictrlar inreresf. infer that the increasing frequency of entries was the result not of technical The official studies which we have factors but of periodic management examined do not set forth any criteria decisions as to what was to be for which leaks and failures might be of recorded.* This means that by the "particular interest.' From a technical standards of 1977-82 (so far as the and statistical standpoint, no failure number of entries into the data base is can be so dismissed as being concerned) tens of thousands of events uninteresting. Indeed, sound estimation involving equipment breakdowns, worker procedures of failure rates and hence of exposures, non-routine releases of accident probabilities require the radioactivity to the environment, etc. recording of every failure. Further, were not recorded.** when data recording procedures are Moreover, the frequency and magnitude changed suddenly, corrections must be of the management changes in introduced in the prior data. We have entry-making procedures does not lend not seen any discussion of such confidence that all the events which corrections in the official report. need to be recorded are now being Nor was the matter taken care of in entered into the Data Bank. Even 1965. The frequency of entries has recently events of great importance, continued to increase. The average for such as corrosion pitting in new tanks, 1960-65 was about 32 entries per year, have been omitted from the Qata B;ink for 1966-69, 75 entries per year. As (Chapter 5). We defer discussion on the noted above this increased to 290 per quality of entries and documents to year during 1970-76 and 1800 per year following chapters which deal with the during 1977-82. specifics of the management of high level radioactive wastes on the Tank Farm. *The probability that there was a purely fortuituous increase in the number of entries, related neither to technical factors nor to management decisions, is close to zero. **If one assumes that there should have about 1800 entries per year in the pre-1977 period as tliere were during 1977-82, then we get an estimate of almost 40,000 abnormal events which were not recorded during 1953-76. References 6. R.M. Girdler, mfLI&. . Pndioactive Wastes at the Savannah rive^ 1. Management of Radioactive Wastcs at Plant, E.I. du Pont de Nemours and the Savannah River Plant, Presentations Company, Aiken, South Carolina, July made to the Committee on Radioactive 1969; pp. 6-7, Waste Management of the National Academy 7. National Research Council, of Sciences at the Savannah River Plant, -active Waste Manaaement at thg January 20 & 21, 1969. TID 26524, gv Reprinted 4/1/71; p. 32. Bcview. Panel on Savannah River Wastes, 2. Ibid, p. 32, charts 2 and 3. National Academy Press, Washington, 3. Energy Research and Development D.C., 1981; Table 1. Administration, Waste 8. Oak Ridge National Laboratory, D~emhnSSa vanM R iver Pbt-- Aikea, en t Fuel ctjve Invcntorlm uth Carw. Final Environmental . . . . Impact Statement, ERDA-1537, National ~~~~6~~ cb"~~b"5;'~~d066 Technical Information Service, Washington, D.C., Sept. 1984; Fig. 2.20. Springfield, Virginia, September 1977; 9. Op. cit. ref. 2, p, 11-65. 10. Op. cit. ref. 4; Table 6-1 1. 4. E.1. du Pont de Nemours and 11. Op. cit. ref. 7; Table 2. Company, Wetv mvsis ... 12. Op. cit. ref. 6; pp. 3-6. mdijStor= 13. 00 Area Fault Tree Data B 0 200 Area) -- SavwRiver, F&H 'At- Farms, ennah DPSTSA-200-3. Aiken, South Carolina, River Plant, computer printout of Data August 1978; p. 6-41. Bank provided by the Savannah River 5. Op. Cit. Ref. 1, pp. 35-7. Plant management (DOE) to the Environmental Policy Institute in 1983 with entries dating from December 1953 to November 1982. Table 3.1

I Volumes of High Level Wastes Stored at SRP (Gallons), (I),(2), 1981

Yaste TVD~ uluLI. TVD~111 Total Supernatant liquid 7,760,000 10,000,000 17,900,000 Crystalline salt 4,040,000 3,000,000 7,070,000 $1udge 2,630,000 180,000 2,810,000 TOTAL 14,400,000 13,300,000 27,700,000

Notes: 1. Source: National Research Council, jladioactive Waste flanaaement at the Savannah River Plant - A Technical Review, National Academy Press, Yashington, D.C., 1982. 2. All numbers rounded to 3 significant figures Table 34

pRnrCIPAt FISSION PROW- (CURIE COW) fW THE HIOH-teVEt WASTES AT SRP

mAI, QUANTZTY AND RADIOACTIVXTY BSTIHhTES8 1981

Radionucl~de Half-Lif e (2) Avarage Spcifio Total Total Rabio- (yeam) ~ctivity ~ctivityof Quantity activity of in aZl nuclide (a) of mdio Radionuclide High-me1 culcien/grams nuclide ~ies wastes kilograms wieo/liter

7, ~adionuclidee[a <1 with lese thtm one-year half-life Notes to Table 3.2

1. The total quantity of high-level wates as of 1981 was assumed to be 27.7 million gallons or 105 million liters. See Table 3.1 above. 2. Source: Ronnie 0. Lipshutz, adioactive Waste: Politics, Technoloav and Risk, Ballingir, Cambridge, Mass., 1980; Appendix A. Numbers preceeded by "'"are estimates from the graph in Appendix A (Figure A). Note that the half-lives cited In various sources differ somewhat. 3. Data on activity per liter of waste are from Table 6.11 of Dupont's Safetv Analvsls Re~ort,DPSTSA 200-3, 1978.

4. Source: 6. Friedlander et 81. Nuclear and Radiochemistrv, Wiley, N.Y., 1981; Appendix D. 5. The estimates for re1atlvely short-1 ived radionucl ides is a rough one, based on the charts of fresh waste In DuPont documents, addusted for decay. 6. THe estimate for relatfvely short-lived radionuclides is a rough one, based on the charts of fresh waste In DuPont documents, adjusted for decay. Table 3.3

W(G-LIVED PISSIOH PROWCTS nt SRP HIGH-mL WASTE ESTIHATE OF QUANTITY AND RADIOACTIVITY, 1981 Radionuclide ~alf-life(1) Average Spbcific Total Total Yeam activity activity (1) quantity radio- in sludge curies/gram of radio- nuclides (3) 121, (3) kilogtatEa curies cories/liter

sludge J \ Notes to Table 3.3

1. Source: Ronnle D. Llpschutz, adloactive Waste: Polltlc&, echnoloav. and Rlsk, Balllngfr, Cambridge, Mass., 1980; Appendix A. 2. Sludge conslstutes about 10% of the wastes or about 10 million llters In 1981. The radlonuclldes ln this table are located princl ally In the sludge except Cs435 and posslbly Tc- 99 (see notes g and 6 below), so that an approximate estimate of the total can be obtalned by multlplylng the speciflc actlvlty In the sludge wlth the total volume of sludge. Source: ERDA 77-42/l.,blternatfves for Lana-Term Manaaement of Defense Iah-level Waste. Table 111-4. 3. Rounded to one slgnlflcrnt figure. 4. Estlmated from Llpshutz, note 1 above, Flgure A-1. 5. A considerable amount of techniclum-99 Is also contrlned In the salt-cake. Thjs Is lndfcated by levels of technjcfum-99 whlch are expected to be found after rocesslng Into nsaltcretem Source: DOE - EIS-0082, fm~act~tatLment, Defense Yaste Proce!:;:: 1982. F%m., 6. Most of the teslum Is In the supernate and salt-cake. Salt-cake content of Cs-135 estlmrted as follows: In sludge Cs-135 = (6x10'~ curles/lltrr) x (10~llters) - 60 curies In tuprrnatc Cs-135 - (1x10" curlrs/lltrr) x (~xl~~llter) SO0 curles Total CS-135 is about 600 curles, excluding that present In salt-cake. Table 3.4

ACTINIDES IN SRP HIGH-LEVEL WASTES: AN ESTXHATB OF QUANTITY AND RADfOAC!l!IlKTY

Radionucl ide Half-Life (1) Average Spacif ic Quantity Total Years activity activity (1) of radio- radio- in sludge (2) Cuties/gram nuclide (3) activity (3) curies/liter kilograms curies

llm Notes to Table 3.4

1. Source: Ronnie Dm Lipshutz, Radioactive Waste: Politics, ~echnoloav,and RislE, Ballinger, Cambridge, 1980; Appendix A.

2. Sludge is about 10% of the volume of high-level waste. The volume of sludge in 1981 was about 10 million liters. The radionuclides in this table are located principally in the sludge. An approximate estimate of the total quantity in curies of the radionuclide can be obtained by multiplying its activity per liter of sludge with the volume of sludge. The figures for average activity in SRP sludge are taken from ERDA 77-42/lX. te at ves Ifiah -Level Waste , *able 111-4. 3. Rounded to one significant figure.

4. Source: G. Friedlander et al., Baaochemistw, Wiley, N.Y., 1981; Po 646.

Table 3.6

FREQUENCY OF ENTRIES INTO THE FAULT TREE DATA BANK

Period Average number of Entries per year

Source: 200 Area Fault Tree Data Bank - F h H Area Waste Tank Farm, Savannah River Plant, with entries from December 1953 to November 1982. LOW ACTIVITY WASTE PROCESSING

ACID

-.?;a 7'- ... ii.i* .%.2f ;4.;;.:.Fr ;,.i. r,'...... tq6y:, . ., G.t,..,.;,; .Y: 3, WAN . . . :.;.;i:;(; :a" *;el: ...+,,).

i + 'I UWWADER 10 TMK FARM

HIGH ACTlVnY WASTE PROCESSING

FIGURE 3.1 31

CHAPTER FOUR TECHNICAL PROBLEMS There has been a wide variety of printnry tn~tks.Alntry sttch technical problems at the Radioactive srcrveys...are ntade by Waste Tank Farm since it began direct obserrario~tsthrough operation. In an industry not dealing operred access risers a~td/or with highly toxic, explosive and irtspectiorr holes in the persistently radioactive materials, many roof, usirrg either lowered of these problems, such as those with itrcandescenr lighu or a some measuring instruments or small pipe mirror-directed sutrbeam for leaks, might be considered minor. illtrminatio~t.*(l) However, in the context of management of a set of tanks containing about 800 As many entries in the million curies of radioactive waste indicate, when a tank riser is which generate explosive gases, there opened, fission products and plutonium are few, if any, problems that can be rise up from the vapor space and considered as minor. Moreover, the level contaminate the air.(2) In addition, of technology appears, in some some worker exposure is probably instances, to be surprisingly primitive unavoidable as a result of such an in situations where the price for this inspection, though the p- does is often paid by increased risks of not cite any. worker sxposures and environmental A wide sampling of technical contamination. problems, which are usually some For example, one method of inspecting combination of equipment, operating the high-level waste tanks is to open and/or process problems, is given in the "riser" covers (akin to manhole Tables 3 through I1 of Part 11. All covers) and look in: these entries are based entirely on the Eault* Inspection oj equipment The premise on which the Radioactive used /or handling and Waste Tank Farm was built was starkly storing radioactive waste simple: is diljicult because 01 radiation and contamination The high-level waste could problems ... be safely stored in tatiks Since 1959, the most gunril natiorral policy and important and recurrent criteria can be agreed upon waste tank inspections /or the long-term storage of (other than routine these wastes."(3) surveillance) have been visual surveys in the No time period after which the tanks annular space and, to a might become unsafe seems to have been lesser extent, inside the specified. Hence a timetable for the

*Other methods such as periscopic examination are also used for inspections emptying of the tanks and the initiation Wc will discuss technical problems of a long-term management method seems under two headlines: not to have been part of the original design. The implicit conclusion must be o Process and Operating Problems; that the Department of Energy (then the o Equipment Problems. Atomic Energy Commission) and its contractor DuPont assumed that the This categorization, as any other, is to wastes could be stored in tanks for an some extent an artificial aid to indef initc period. discussion since there is a considerable This assumption was made explicit as overlap and interrelationship between recently as 1977 by DOE (then the Energy the two categories. For instance, Research and Development equipment problems such as those with Administration). This report explicitly level measuring instruments or with pipe considered continuing operation of the corrosion can lead to operating problems Tank Farm into the indefinite future, and accidents such as overfills and with the tanks being replaced every leaks. Similarly, process problems such fifty years, as an alternative for as high sludge temperature can long-term waste management.(4) (See accelerate equipment problems such as Chapter 5 for further discussion on the tank stress-cracking. assumed longevity of the tanks.) This confidence in tank storage was Process and Operating Problems accompanied by the premise that long-term management could be relatively The nature of the materials and of easily accomplished by pumping the the reprocessing techniques at the wastes directly into the bedrock beneath Savannah River Plant meant that a large SRP. For about two decades this was an volume of high-level radioactive waste implicit premise in the design and would bc generated. It was decided to operation of the Tank Farm. neutralize this acid waste with sodium But this original assumption of hydroxide since the SRP management a quick, cheap and safe solution to calculated that "the storage o/ the long-term management has proved elusive. wastes in alkaline solution [was the] (see below.) Similarly, the operating most economical."(S) This conclusion history of the Tank Farm does not stems largely from the fact that acid support the original premise that the waste must be handled with specially wastes can be stored safely in tanks for resistant equipment and stored in an indefinite period -- even though this stainless steel tanks. Acid dissolves continues to be the official view of the carbon steel. Neutralized waste does operation. We cannot, of course, present not do this, and it could thus be put in a detailed assessment here. Our much cheaper carbon-steel tanks. resources only permit a sketch , a Unfortunately, the decision to preliminary overview of the technical neutralize the waste has had a number of issues as they relate to the operation severe impacts on the question of of and to the long-rerm questions posed long-term waste management. As it by the Waste Tank Farm. turned out, even the near and medium term problems became so severe as to elements."(8) Much of this equipment, operation that is frequently required to such as the pumps and jets, have been route waste: the source of considerable operating problems, accidents and breakdowns. The IZ'aste martagement attd T & T evaporators have also been the source of usirtg the shielded crane, considerable operating problems. artentpted to chattge ttitmerow These problems have in turn caused jrcntpers in CTS fconcenrrare worker exposures and environmental transfer system] cell bur contamination. For example, radiation were 'etnsuccessftil dtte to exposure rates during replacement of an crane failtcre or ittability evaporator feed pump in November 1968 to see while nroritrg ranged up to 30,000 millirads/hour and equipnrertr. The exposure the total exposure was estimated as 800 dose rate in the cab millirads. (See Summary Table 1) In remained less than I mr/hr another example, the total exposure and the radiation level over during the excavation of the F area open cell was 2R/hr. T & T concentrate transfer system pump pit in later made jumper changes October 1975 was 21.6 rem. (See Table 1, encountering dose rates of Part 11, entry 07-75F) The following up to 1.5R. The total entry is an example of environmental estimated exposure involving contamination due to equipment problems: 36 personnel entries was I115 mr. 02-29fsicJ-61. Soil Contamination. [Pl'ettonium) The successful development of adequately solutions have leaked from shielded equipment such as cabs could this patre1 4 times in recent lower exposures substantially for ntottths by low level wasre operations which now require entry of concentrate leaking through personnel into extremely radioactive cell walls of rhe waste from environments. evaporator. Soil removed: Many of the operating problems - asphalt mostly replaced. such as the plugging of jets and pipes, (See Table 2, Part II) and pump failures as well as the failure to deveIop adequately shielded remote It is also important to note maintenance and operating procedures -- that with three decades of operating have a direct bearing on the prospects experience, it has proven impossible to for long-term waste management in develop remote maintenance methods for general and for vitrification in many jobs essential to Waste Tank Farm particular. (See below.) operation, given the present level of There have been many other process resources devoted to it. Thus in and opcrating problems. A sampling of November 1982, the plant management these is to be found in Summary Table 3, unsuccessfully attempted to use a while a more extensive selection from shielded crane in order to change the Data Bank is given in Tables 3 jumpers (pipe connections) in the through 11 of Part 11. For instance, concentrate transfer system, a dangerous summer lightning frequently causes loss explosive limit" and that on all other of power. Usually, the emergency power occasions it has been below this limit. takes over, but this has failed on (10) However, the Data Buentry dated occasion. Loss of power on July 31, 09-00-56 records a level well above this 1972, caused a loss of make-up cooling limit (Table 5, Part 11): *h)drogen water "for a short period of time" in content of the vapor space of Tank 9 some areas. (See Table 7, Part 11) 150% oj tower explosive limit." Prolonged loss of cooling water to Besides this, the BuPont Safety high level waste tanks could cause bnalvsis Rebort does not mention any serious problems. The hotter wastes explosion in tanks resulting from would come to a boil in about a week. hydrogen build-up. The Data Bank An accident in December 1956 caused a cites one such explosion in tank '6 due loss of cooling water to Tank 12 for six to a hydrogen concentration of only 15% days. of the lower explosive limit. The confused and contradictory data in the [This] resulted in waste Safetv AnamReba and the Dm temperatlrres that approached does not permit us to arrive at any the bdiling point and a conclusions about these events, visible discharge oj except that it shows the unreliability condensing steam emerging of the data and the public reports for jrom the fiberglass-packed even the most serious accidents that verrt jitter, and release oj could happen in the Tank Farm operation. less than I Ci oj airborne activity. There was rto Equipment Problems appreciable spread oj contamination beyond the Many operating problems have immediate vicinity oJ the been rendered more severe by equipment tanks.(9) problems. We have already mentioned the failure to develop adequately shielded remote work stations for many jobs. ted in the Fault Tree Dm Bank. Furthet, process and operating problems Rather an earlier less serious loss of such as those with failed pumps or cooling water for 2 days to Tank 3 in plugged lines can be interpreted in some August 1956 is listed. (See Table 4, cases as those of equipment which is not Part 11) fully suited to the materials, As a final example, we consider the temperatures and radioactive conditions data of hydrogen formed by the action of which are encountered on the Waste Tank radiation ("radiolysis") in the tanks. Farm. Again, it is important to remember A substantial accumulation of hydrogen that it is the highly toxic and could result in a tank explosion. (See persistent nature of the radioactive Table 5, Part XI) The DuPont Safety materials which is a primary cause of Analysis Report on Waste Management many of tbese difficulties with states that on "one occasion... the instruments, pumps, jets, etc. A hydrogen concentration in the vapor selection from the entries in the Data space of a waste tank reached the lower &mk is given in Tables 3 through I I in Part 11. In this section we will 1969. (Table 4, Part XI, H-area): "Tank concentrate on the problems related to 11 had 16 cooling coil leaks following the tanks only, since the questions of sludge removal. None had been detectcd safety of worker and public exposures bcfore sludge removal." and long-term risk depend centrally on Cooling coil designs have been the integrity of the tanks. repeatedly modified. The initial design There are four types of tanks on this of Type JII tanks called for removable Tank Farm. (See Figures 4.1 and 4.2) cooling coils, but the later ones are Type IV tanks are the single-walled, being built with fixed coils as was the uncooled tanks which are used to feed case with the design of the very first the evaporator and for low-activity tanks. waste storage.(ll) Types I, I1 and I11 Various incidents, some of frequen t are cooled tanks. Types I and 11, the occurrence, have cast doubt on the earlier designs for high-level wastes, integrity of both the primary and have a secondary concrete containment secondary containment of the tanks. In with a partial steel lining for 5 feet the leak from Tank 16 in 1960, discussed of their approximately 25 foot height. in Chapter 6, the liquid waste seemed to They are sometimes known as have escaped into the soil very soon "double-walled" tanks.(l2) Type 111 after its level reached above the outer tanks have a full-height secondary steel steel pan and came into contact with the liner. There are 51 tanks in all: 16 of concrete container. Other events, such Types I and 11.27 of Type 11 and the as the escape of high heat waste from rest of Type IV. the concrete encasement of the pipe at There have been several kinds of Tank 8, also discussed in Chapter 6, problems with the tanks. For instance, also indicate that'concrete is not an one major problem relates to cooling effective barrier to high-level wastes. coil leaks. A large number of cooling The situation has been somewhat coils on the first batch of 12 Type I alleviated in this regard by the tanks and the second batch of 4 Type 11 construction of full height steel liners tanks leaked. The leaking coils had to in the secondary containment structure, be taken out of service, reducing the but almost all the older tanks continue available cooling in the tanks. In some to be used and they contain large cases a large proportion of the cooling quantities of radioactivity. coils had to be blanked off because of Further evidence of the leaks. Partly due to such leaks, some ineffectiveness of the secondary tanks are more hazardous than others for containment is provided by the frequency storing wastes. The cooling coil leaks with which groundwater and rainwater have thus been one of the causes flow into the annular space, sometimes in necessitating an increased number of large quantities. For instance, several transfers from one tank to another, hundred gallons of water leaked into one Ironically, such transfers have of the newer tanks (Tank 40) in August sometimes created cooling coil leaks 1982 due to inadequate sealing. The where none were previously evident, as water carried some clay with it into the was apparently the case in the transfer annulus. (See entry dated 08- -82 of sludge from Tank 11 during November H-area in Table 3, Part 11) Upon occasion, large quantjties of rainwater corrosion pitting in an H-area entry have also entered the primary dated 02-04-81. (See Table 4, Part 11). containment. For instance, 4,800 gallons of rainwater flowed into Tank 11 due to Co~~strttctionhas completed "excessive rainfall over a short period over eight percetit oj of time and failure to expedite repair nreasrcrittg attd mapping of of cracked risers immcdiately upon pits i~rTattk 38. To date discovery." (H-area entry dated 08-1 8-82 the deepest pit is 0.061. in Table 3, Part 11) Inspection of the cleaned The primary steel containment has jloor is revealing hutrdreds been subject to cracking and to leaks. of shatlo~vpits between 30 Four of the first sixteen tanks had ro 60 mil deep." developed leaks by the late 1950s (13) and a total of nine by the mid-1970s. The deepest pit found in the tanks was Some tanks have had extensive corrosion 170 mils or about one-third the wall cracking. While the leaks from these thickness. (16) cracks tend to get sealed with salt According to the testimony of William deposfis in most cases, these deposits Lawless, a former Senior Engineer in may not similarly keep the wastes Waste Management in the Department of confined in the event of mechanical Energy and stationed then at the stress, such as that from hydrogen Savannah River Plant. DuPont was aware explosions or earthquakes. The presence that 'corrosion pitting could penetrate of many cracks and leaks in some tanks the tanks in very short order." has possibly compromised their However, this "was not considered a structural integrity. (See Chapter 5.) problem at SRP for many reasons The substantial redesign of the tanks including the new design of these new after the experience with cracks and Type 111 tan ks...[ and] also because of leaks in Type I and 11 tanks was the construction quality assurance supposed to have addressed these procedures that DuPont was using."(l7) problems. The failures were This DuPont assessment was made in investigated and as a *result of this March 1980, six mom prior to the investigation new cooled tanks were discovery of the corrosion pitting. designed to eliminate the cracking After the discovery of that pitting, problem and to obtain other advantages according to Lawless, "DuPont felt that indicated desirable by experience to there should be no date."(lS) restrictions...[p laced on] these tanks." Unfortunately a large number of the A consulting firm, Arthur D. Little, new Type 111 tanks, at least fourteen of Inc., hired to do an assessment of the twenty-seven, were already compromised problem, recommended that restrictions during construction by severe corrosion be placed on three tanks regarding the pitting. Such pitting was found on kind of wastes to be put in them.(l8) (Tank numbers 38 The Department of Energy overruled these through 52) under construction during recommendations and certified them for 1980 and 1981.(15) Yet the Data Bank unrestricted service. contains only pne entrl regarding Tank 38 was considered by all parties to be sound. We undcrstand that a few decades, the long-term approach which months after it was put into scrvice, SRP management assumed would be viablc, the primary containment Icaked, possibly and the only one it considered in some due to a welding problcm. detail, was direct disposal in bedrock. Disposal of liquid high-level wastes in a deep rock formation beneath the SRP site was first proposed in 1951 The initial decision to by the Atomic Energy commission. Because neutralize the wastes and store them in this formation was located beneath two relatively inexpensive carbon-steel major aquifer systems, major concerns tanks was madc without much planning for over this mode of disposal were soon the long-term consequences. Three raised. But it was not until 1955 that a factors appear to have been largely Committee on Geological Aspects of responsible for this: Radioactive Waste Disposal of the National Academy of Science -- National o initial attention was focused Research Council (NAS-NRC) was convened mainly on production of plutonium, to advise the Atomic Energy Commission etc. without serious anticipation on current and proposed geological of possible domestic negative disposal methods of radioactive wastes. consequences; In March 1960 the Committee reviewed the safety and feasibility of storing o confidence that the tanks could radioactive wastes beneath the SRP plant hold the wastes for as long as site and recommended that SRP proceed necessary; and with test borings. In December of 1962 the testing o an assumption that a quick, cheap program for bedrock disposal at SRP was way would be found for long-term complete. The NAS-NRC Committee disposal. concluded that "storage of liquid radioactive wastes in excavated chambers Attention continues to be was technically feasible."(l9) Six months focused on the production aspects. later, the NAS-NRC Committee concluded However, the confidence regarding the further that underground disposal at SRP long-term viability of the tanks and the was safer than storage in surface tanks. quick availability of some other method But they added a note of caution by has not been borne out by time. pointing out that the test drilling at Storage in tanks has faced a SRP may have "invalidated" some of their number of problems and cannot be a data because it disturbed groundwater long-term option because of the severe flow. From 1964 and 1966 further tests dangers that this kind of storage poses were made. in case of accidents or natural By 1965 the NAS Committee membership catastrophes. The numerous leaks which changed, with only one member from the developed in the tanks in the first few previous group remaining. In 1966 the years after they were built also pointed majority of the newly reconstituted to the necessity of some other long-term Commit tee opposed direct disposal in management solution. For about two bedrock at SRP. They recommended that the entire program be discontinued on studied any alternative.(22) EPA also the grounds that P.. the placement of cited the existence of faults in the high-level wastes 500 to 1000 feet below rock formations beneath SRP which have a very prolific and much used aqzrifer is the potential for movement -- in its essence dangerotts and certainly threatening the integrity of a will lead to public controversy." The repository located in these formations. Committee also stated "dhere is doubt Throughout the 1970s. the state of that it will be possible to prove the Georgia voiced strong opposition to the sa fery oj the proposed bedrock system bedrock disposal option at SRP, lor high-level liquid or soluble particularly when Jimmy Carter was wastes," Underscoring their opinion was governor. In 1979, under President the 'unpredictable nature of groundwater Carter's administration, the DOE in a flow through fractured rock." The decision on the final Environmental minority view held that 'work on bedrock Impact Statement on the "Long-term disposal at SRP would be continued."(20) Management of Defense High-Level The Atomic Energy Commission Radioactive Wastesa at SRP, suppressed this study until 1970. administratively foreclosed the Moreover, they ignored the majority option.(23) This, however, has not opinion and proceeded along the same satisfied the current Georgia state course set out in 1951. The NAS-NRC government because they believe that Committee went out of existence after DOE'S decision is legally non-binding. submitting their report, and a new Adding to Georgia's suspicions, the NAS Committee was convened in 1972 with released yet another report on disposal different members. This time, the of high-level wastes at SRP in 1981, Committee concluded that ',there is a which endorsed (a') the drilling of an reasonable prospect of achieving exploratory shaft for bedrock disposal; [adequate] protection by storing the (b) the grouting of radioactive wastes waste in vaults in rocks underlying the into rock formations underneath SRP; (c) Tuscaloosa formation beneath the the direct disposal beneath the SRP site Savannah River Plant site." However, as a safe and cheaper option than the Committee felt that *,no putting the wastes in solid form and reasonable amount of exploration from storing them at r tepository the land surface can conclusively of f-sit e.(24) demonstrate the saftty of waste storage in deep vaults." The Committee then recommended that *an exploratory shaft be sunk and exploratory tunnels be While DOE and DuPont have not driven into the rock selected."(il) yet officially acknowledged the dangers Plans for executing the NAS of the direct disposal approach, they Committee's recommendations were blocked appear to have abandoned it in practice in 1972 by the Environmental Protection for the present in favor of Agency (EPA) on the grounds that deep encapsulating the wastes in glass (which geological disposal of high-level waste is similar to the Pyrex kitchenware). was 'environmentally unsound' and that This process, known as vitrification, is for more than 20 yean DOE had not to bt carried out in the Defense Waste

The discussion, the data and quotes in it are based on D.O.Eas final EIS for the project, unless otherwise stated (23). Processing Facility now undcr The first, and perhaps most construction at SRP. Most of the significant fact about routine radionuclides will be cncapsulatcd in operation of the vitrification plant, is glass. The rest of the waste will be in to note that. .there is no ex~etiencewith thc form of a chemical salt 3uch_a~jlot ~Lad_t solution. This is to be processed by evaporation and mixed with cement. The full-scale industru badIn its large radioactive cement blocks, called Final Environment Impact Statement for "saltcrete", are to be disposed of as the project, DOE summed up the "low-level" wastes in shallow unlined vitrification experience at SRP burial pits at the SRP site. as follows: The production of glass is scheduled to begin in 1989. No schedule At SRL the borosilicare glass exists as yet for placing the glass in a process is being successfully repository, but it is reasonably certain dentottstrated on an engineering scale that no repository will be in operation with simulated (non-radioactive) waste before- 1998 -- the earliest date and tested on a laboratory scale with announced for repository opening by DOE-- acaral SRP waste. Physical property data since there have already been haie been obtained on jull-size substantial delays in the repository non-radioiactive Jorms and on program. Thus, the glass is to be stored small-scale forms made with actual at the Savannah River Plant for about a waste. decade, and probably longer. DOE plans to dispose of the saltcrete in an In other words, individual "engineered low-level" waste burial site full-size glass cylinders (0.61 meters near the glassification plant. diameter and 3 meters high) have been Successful solidification of the produced, but without radioactive wastes wastes would be a step forward in incorporated into them. The protecting the public from accidents production of single glass cylinders such as hydrogen explosions. It would without radioactive wastes is also pose less immediate danger in case qualitatively different in major of earthquakes. However, we have serious respects from a large-scale plant reservations about the way in which this producing 500 canisters every year, plan is being carried out, since there tach containing up to 200,000 curies of are a number of unresolved issues and radioactivity. In fact, mvof t& problems relating to both the routine operation of the facility as well as to potential accidents. In addition, there are basic questions relating to the mver been OD-v tested w& glass and saltcrete and the proposed ive materials for any disposal of saltcrete in the "low3evel" t leagtb of tjme. Further, the waste burial ground. entire process has not been tested on an industria) scale, with the actual equipment even wjth simulated (non-radioactive) wastes. Nor has a

high-level waste have been common in discussed. The analysis is based on a Tank Farm operations. Most repair and number of assumptions, many of which do replacement and much maintenance has not derive from direct operating required "hands-on" work. Breakdowns in experience. It comes to the conclusion such essential equipment have therefore that radiation doses to the public would led to worker exposures and sometimes to be small. lengthy shut-downs. It is particularly Since there is no significant important to note that, after three operating experience, and since the decades of Tank Farm operation, DuPont public literature is meager -- and DOE have not been to able develop A~~enmon Accidents in the FmEIS remote working methods for such routine does not contain even one reference -. work as jumper changes (required for the it is not possible for us to make an proper routing of waste). assessment of DOE'S conclusions. The formation of the waste-glass However, we note that the approach to mixture, the production of the canisters the analysis is similar to that which filled with radioactive glass and many DOE and DuPont have used for analyzing of the other processes in the Defense the safety of the Tank Farm operation. Waste Processing Facility will be As we will see in the next chapter, that remotely controlled and operated. approach is statistically and technically Under such conditions, maintenance, unsound. repair, and replacement of equipment DOE'S analysis of accidents in and clean-up af ter accidents may pose the Final EIS omits any substantial even more serious problems than those discussion of what might become a which have been encountered so far in critical problem for the continued Tank Farm operation. For those reasons operation of the plant, and hence for we have concluded that yen~f~cam. I. the future vitrification of the wastes and their placement in a repository. This is the problem of the effect of accidents, ranging from small spills to fires or explosions, on the operability of the plant itself. Failure of the This conclusion is reinforced by the plant to operate substantially as possibility of accidential contamination predicted would have substantial of the facility. environmental and economic impacts, and The Final Environmental Impact possibly major health impacts as well. Statement of the Department of Energy DOE dismisses "minor incidents" for the project contains a list of the out of hand: accidents "in which significant amounts of radioactive materials could be Occasionally minor incidents released into the environment. "The EIS will occur during plant operation briefly discusses events such as "spill because of operator error or failure oj from a slurry receipt tank" and "steam a plan! component or system. Such events explosion in glass melter: However, will resttl! in the release of little or only the consequence for immediate no radioactivity to the environment and releases to the environment are are, therejore, no! discussed in this report. In a similar vein, the noted above such a process has been consequences for operability of a major suspected to cause failures of several spill are discussed. In the event of a tank cooling coils on at least one spill of 16,000 liters (more than 4,000 occasion. gallons) of highly radioactive sludge Failure of a large number of cooling slurry onto the cell floor, the spilled coils in the new tanks would seriouly material would be flushed into the sump impair their ability to hold high-level with about 1000 L [liters] of water and wastes. The older tanks already have then transferred through the dump system many leaking cooling coils, rendering to a vessel for reprocessing . . , them unfit for holding hot sludge. During the accident and subsequent Most of the older tanks also transfer, aerosols from the evaporation have leaks in the primary containment of the spilled material would be carried vessels which are currently plugged with through the canyon ventilation system." salt which has crystallized on the There is no discussion of the cracks. Thus, the transfer of wastes effects of such a spill on the from older tanks to newer ones and the subsequent operation and maintenance of preparation of feedstock for the the plant. vitrification plant could lead to One effect of this and other serious problems, including leaks of such spills would be to make the radioactivity and impairment of Tank processing areas very radioactive. This Farm operations. would make any entry into the process These problems must be area much more dangerous. Consequently, confronted in any effort to solidify the all maintenance, repair, and replacement wastes. However, the likely intensity of requiring such entry by personnel would the problems will depend on the demands involve higher exposures and work under of the specific solidification process morc hazardous conditions. It is and equipment. We do not know if these possible that some of the accidents problems have been taken into account in listed would seriously impair the the design details which are not public. maintainability, and hence the Public documents, such as the Final EIS operability, of the plant. dating to the Defense Waste Processing There is another class of Facility, do not deal with them events that could seriously affect the substantially. operation of the vitrification plant which has apparently not been Ssftcrctc considered, This relates to the ability The salt in the tanks is to produce consistently the kind of feed proposed to be dissolved and most of the solutions and slurries that will be cesium and strontium isotopoes are required for vitrification, without removed from it. The rest of the seriously impairing the integrity of the solution is to be mixed with cement and Tank Farm operation. poured into trenches to set as concrete, For instance, water and hot called saltcrete. The saltcrete chemicals will be added to the sludge wmonolithsaare to be formed in trenches and the mixture agitated to produce a about 10 meters below the soil surface. slurry-feed for the glass melter. As Each trench will be 20 feet deep by 21 feet wide by 52 feet long. This arrangement and the spicification of the saltcrete are to conform to the Nuclear 60,000 curies of technicium-99 Regulatory Commission's criteria for (half -1if e 2 14,000 years), and 200 "low- level" waste. The saltcrete burial curies of iodine-129 (half-life of ground will be treated as such. 16 million years). In addition, substantial quantities of alpha emitters like The designation "low-level" plutonium-238 and americium-241 will be waste docs not apply to the overall discharged. Further, these alpha radionuclide content of the wastes, but emitters do not decay into stable rather has been'created as a catch-all elements but generate other radioactive category for wastes not falling into elements like uranium-234 and other categories. Ronnie Lipshutz has neptunium-237, some of which are even noted that these wastes were renamed more long-lived. "nonhigh level wastes" and were defined -In 1984 the DOE adopted a new by the federal government to include internal nuclear waste management any waste that is not high level and standard (DOE order 5820) which raised contains less than ten nanocuries of the control limit for alpha emitters in alpha activity per cubic foot. (28) soil from 10 nanocuries per gram to 100 ,Thus, a sufficiently large nanocuries per gram. This increase in dilution of radioactive waste would alpha contaminated soil appears to be allow its redefinition as 'low-level designed to accomodate the saltcrete waste and hence permit its discharge process, Support for this relaxation in into dumps from which the long-lived DOE'S standard assumes that substances nuclides could pollute the soil and the like plutoniirm can be compared to groundwater. This appears to be the case natural analogs(29) - an approach not with the proposed production of widely accepted in the industry.(30) In 'saltcrete" in the Defcnse Waste Chapter 6 we discuss how DuPont's soil Processing Facility. transport models which support this Table 4-1 shows the quantities increase in alpha emitters in soil bas of some of the long-lived radionuclides been invalidated by data from its own which are to be disposed of in the new operating experience. mlowfevel"waste burial ground to be Saltcrete falls into the built as a part of the Defense Waste category of 'low4 evelw waste only Processing Facility. The cumulative because of dilution. Very large quantities of some long-lived quantities of it are to be produced - 530 radionuclides at the end of the proposed cubic meters per week, which is about 1 28-year operation of the plant will be million cubic feet per year. The planned very large, even after adjusting for overall production in 28 years of almost radioactive decay. Roughly three million 30 million cubic feet of low-level" tons of "saltcrete" will be disposed off waste in the form of saltcrete alone is in the "low-level" dumps at SRP over 28 about d-the which. had years, The cumulative amount of various BC~mLlla radionuclides in this will be pperatcd low-level radioactive waw approximately: 30,000 curies of s ~n the U.S. comb ned. (311 It will cesium-137, 50,000 curies of e the cum\Ltative.. "low-level" . samarium-1 5 1 (half-life' 90 years), Y'aste neneratcd at SBP In - The total quantity of most maximum permissible concentration (which long-lived radionuclides which have been will do some harm, particularly to discharged into the seepage basins in children, even at those levels). the first 30 years will pale by Thus, 200 curies of iodine-129 comparison with the plannicd discharges will require a dilution by 100 billion in the form of saltcrete. For example, gallons to achieve the maximum the planned accumulated discharge of permissible concen tration for 30,000 curies (decay-corrected) of iodine-129. The final Environmental cesium-137 will be 100 times the Assessment made by DOE for saltcrete accumulated decay corrected discharges considers only the effect of discharges of cesium 137 to the seepage basin until of radionuclides with the Savannah 1976. Similarly, the planned cumulative River, where a large dilution can be discharge of about 100 curies of assumed. As in the other assessments plutonium-238 will be about 30 times the discussed in this report, the estimated accumulated dischanges to the contamination of shallow aquifers is not seepage basins until 1976 of 3.4 curies. given due importance and the possibility (See Chapter 6 for discussion of of the contamination of the much-used routine discharges.) Tuscaloosa aquifer is not even The situation with regard to seriously considered. technicium-99 and iodine-129 will be The present plans for disposal especially serious. The cumulative of large quantities of radionuclides in amounts discharged to the seepage basin saltcrete are the combined result of until 1976 of these radionuclides was inadequate technological development and about one hundredth of a curie and 20 inadequate efforts at using known microcuries, respectively. The planned technology. For example, in the case of discharges in saltcrete .will . increase technicium-99, the technology to recover these figures. . by . . it from the waste exists, as discussed for technlclum-99 and t wentv in an official report on Hanford waste 1Pp1ne-129,. . There is no prospect that operations.(32) But DOE and DuPont do saltcrete can contain these long-lived not plan to use it at SRP. elements for anything approaching their Mercury in saltcrete could also half-lives. Thus, much of the over 3 present problems. There are about 90 to tons of technicium-99 and 12 tons of 100 tons of mercury in high-level waste iodine-129 is likely to find its way tanks at SRP. Present plans call for into ground and surface waters from the extraction and cleaning of this mercury "lowdevela saltcrete burial site. They for reuse, as part of the Defense Waste will then present a substantial hazard Processing Facility. On the order of one to people. percent of this, or 1 ton, may be The "maximum permissible present in the saltcrete, due to concentrationa of iodine-I29 in water is imperfect recovery techniques. Yet the SO nanocuries per cubic meter (about final EIS admits that studies of the 0.2 billionths of a curie per gallon). leachability from saltcrete are still This means that it would require only in the "preliminary" stage. about 5 billion gallons of water to In sum, the construction of a dilute 1 curie of iodine-129 to the full-scale vitrification plant appears to be technically premature. It may even result in cosiderable delays in the velocity of the water in the aquifers vitrification of wastes if an accident which would soon saturate the repository irretrievably contaminates a vital could cause a relatively rapid erosion section of the plant. of the glass. (34) Our recommendations for At the Nevada Test Site, the solidification by calcining and relatively low pressure in the repository construction of a pilot plant are aimed could cause water coming into contact at achieving the relative safety that with the hot glass to flash into steam. solidification of the wastes in the In experiments at the Argonne National tanks will bring (particularly the Laboratory, researchers discovered that sludge and supernates), and at the same steam attacks glass and can cause its time minimizing the risks of failure in rapid disintegration.(36) the long run. This mismatching of waste form More time is also needed to to the repositories could result in select the best waste form for the violation of Nuclear Regulatory repository site to be selected. The Commission regulations for long-term choice of borosilicate glass may disposal at both these sites. These turn out to be an unfortunate one. The require that: controversy about choice of materials known as "waste forms" for waste o the waste form remain almost solidification has had a long and completely intact for at least 300 complex history, which we will not go in years and up to 1,000 years; to here. We only note here that it is by now well recognized that the choice of o the rate of disintegration of the encapsulating material and the choice of waste form should not exceed 1 part the location of the geologic repository per 100,000 per year thereafter. must be such that the two complement tach other.. It is quite possible that in the In December 1983, the Department case of glass one or both these of Energy announced three locations requirements would not be met at Hanford from which it will choose its first and the Nevada Test Site. (We have not geologic repository.. They are: made an investigation of geologic questions relating to the Deaf Smith o Hanford, Washington o Nevada Test Site o Deaf Smith County, Texas Two of these three sites - Hanford and the Nevada Test Site - have characteristics which could result in the relatively rapid destruction of the glass and, hence, release of the wastes to the surrounding environment. At Hanford, the relatively rapid

*While the selection is formally only for civilian high-level wastes at this stage, military high-level wastes are likely to be put there also because of the cost of additional repositories and the political difficulty of finding a site. DOE has already filed one evaluation in which it claims a common civilian-military geologic repository would be cheaper. STET(33) County site. We should note here, unanticipated problems and even however, that the county is breakdowns, which have not been factored agriculturally very productive for which into cost and exposure. There is at it depends on clean underground water least some experience with calcining, su pplies.) which is also required, anyway, as part This possible mismatching arose of the vitrification process. We believe in part because of the haste with which that if risks and exposures to be both decisions were made, and the minimized, it is more prudent to proceed apparent lack of serious coordination carefully, and step by step, rather than between them. The initial repository rushing headlong into full-scale selection documents do not examine in production without pilopplant depth the question of waste form experience, as DOE and DuPont are now interaction with the specific sites doing. These risks and exposures can be selected for study as possible further minimized by stopping the repositories. production of high-level liquid The result of this haste is radioactive wastes, pending the likely. to be increased cost, higher resolution of the long-term questions. worker exposures, and less long-term Ref erences security from radioactivity for future generations, It would, in our opinion, 1. Energy Research and Development be more prudent and conservative to Administration, JVaste ' M- proceed in the following manner: ation. Sav- River P- Final Environmental Impact o solidify the existing high-level Statement, ERDA-1537, National Technical wastes,-including the sludge by Information Service, Springfield, VA, calcining them; September 1977; p. 11-102. 2, Savann ah River Pfant - 200 Ara o conduct appropriate pilot plant Was@ studies, including development of Farm Computer Printout of Data better worker protection measures, Bank provided to the Environmental for waste encapsulation in glass Policy Institute in July 1983, with and, possibly, other waste forms. entries dating from December 1953 to November 1982. o stop the further production of 3. Barah, A.A, The Nature, liquid high-level radioactive Handling and Storage of Separations wastes until the long-term Process Wastes: in -nt af questions are resolved. Padioactive Wwat themahRiver Plant. presentations made to the The Department of Energy has Committee on Radioactive Waste opposed calcining on the grounds that Management of the National Academy of this might be more cgstly and cause more Sciences at the Savannah River Plant, worker exposures? We believe this is a January 21, 1969, p. 42. short-sighted view. Rushing in to 4. Alternatives for --Term full-scale vitrification without pilot ent of Defma-Leva- - - plant experience is likely to cause many tive WwSavannah River Plant,

'calcining is a proven method that has used to solidify high-level radioactive wastes at DOE'S Idaho National Engineering Laboratory for over 20 years.

Capacity for the Disposal of Defense High-Level Waste, DOE/DP-0020 (Draft), Washington, D.C., July 1984. 34. J.E. Mendel (compiler), Fim ort of the Defense H ah-Level Waste Leacm Mechanisms Pronran, prepared for the U.S. Department of Energy, Pacific Northwest Laboratory, Richland, Washington, August 1984; pp. 1.28-1.33. 35. Arjun Makhijani, and Kathleen Tucker,. . mt. High Water and Rock LDstabllltv at Haford, Health and Energy Institute, Washington, D.C.; February 1985; pp. 46-56. 36. J.K. Bates et al., "Hydration Aging of Nuclear Wastes," Science., Vol. 218, 1 October 1982; pp. 51-53. TABLE 4-1

Planned Annual Discharges amd Cumulative Quantities of some Radionuclides in Saltcrete, Corrected for Radioactive Decay

Specific Cumulative Content in Content Half -life Saltcrete Saltcrete after Radionuclide in years nanocuries/gm 28 Years

1. Tritium 12.3 21 6,000

Source: DOE/EIS-0082 UNCOOLED WASTE STORAGE TANK Prestressed Concrete Walls 1,300,000 Gallons

I

I

PNEUMATIC CONCRETE STEEL LINER WITH TENSlONlNG BANDS

COOLED WASTE STORAGE TANK Type I, Original 750.000 Gallons

CONDENSER RISER

EARTH COVER

5,7s8-0~4 FIGURE 4.1 53 COOLED WASTE STORAGE TANK Type 1,030,000 Gallons

-INTAKE FAN EXHAUSTl

COOLED WASTE STORAGE TANK Stress Relieved Primary Liner 1,300,000 Gallons TYP~m ,EXHAUST FAN AIR INLET PIPES

SECONDARY LINER PRl MARY LINER

2'-6* ANNULUS SPACE

54 FIGURE 4.2 CHAPTER FIVE CATASTROPHIC EVENTS In a situation where 800 million critique of the method and suggest more curies of radionuclides, in mostly reasonable ways of approaching the liquid and sludge form, are being held problem. in tanks, there is the ever-present The basic approach of DuPont and the possibility that some event may Department of Energy is to calculate precipitate a catastrophic release of probabilities for events that might radioactivity to the environment. There initiate or contribute to releases of are three broad areas that must be radionuclides and then estimate the addressed to assess the probabilities magnitude of the releases. The method and consequences of such releases: is called "fault tree" analysis. The human consequences of these releases are o analysis of possible equipment then calculated by estimating how the and. process failures which could radioactivity might reach people result in large releases - for ("critical path") and how many cancers example, tank leaks or hydrogen and genetic effects it might therefore explosions in tanks; cause. Both the calculations of the probabilities and the ways in which they o an assessment of the effects of are used are seriously defective. Some events not caused by the Savannah of the calculations of the consequences River Plant but which could have of releases fall into the realm of the a severe impact on its operation absurd. -- for instance, an earthquake: Process and Equipment Fatlures o an assessment of the consequences of such accidents and releases The first step in the official risk once they do happen. assessment is to catalog events which might result in releases and to Both the Department of Energy and calculate their probabilities. For DuPont have produced reports on this events related to the plant itself: subject, In 1977, the Energy Research and Development Administration presented Probabilities /or primary calculations of risk to the public posed events oj Tank Farm Fault Tree by the possibility of catastrophic were obtained primarily from events.(l) DuPont completed a Safctv three sources: in 1978 which is practically the same in regard to the o 'The SRL [Savannah River risk calculations.(2) We will not Laboratory] computer-stored analyses the results of their data bank called 200 Area calculations but rather present a Fault Tree Data Storage and Retrieval System."

*this report remained secret until 1982. o "U'ASH- 1400, Appendix 111, made. If there is any significant Failure Data." change in recording procedures, this o "Judgnrerrts by experienced should be clcarly motivated and relatcd Techrtical persorrrtel of SR P to field conditions. Finally, when artd SRLfSavatt~~ahRiver different recording procedures are used, Lnboralory ]."(3) a systematic attempt to bring all periods to a comparable statistical The second of these three items has basis must be made. None of these nothing directly to do with the Savannah criteria have been followed in the River Plant Waste Tank Farm. WASH-1400 recording of the mtTree Data Bank. is a study entitled Reactor Safetv We noted in Chapter 3 that the Studv: An Assessm- frequency of Data Baentries have U.Wercia1 Nuclear Power Planu varied from 4 per year in the 1950s to published by the Energy Research and about 1,800 per year in the late 1970s. Development Administration in 1975. Moreover, these changes in recordkeeping Apart from the considerable controversy procedures seem to have been largely surrounding this study, the fact of the arbitrary and dictated by non-technical matter is that commercial nuclear power considerations. Many events of plants do not have tanks full of considerable significance such as tank reprocessed, neutralized, liquid leaks were explicitly not recorded since high-level radioactive waste. Some of they were deemed not to be of the same equipment used in commercial "particular interest." The quality and power plants might be usea on the Tank consistency of the data also leave much Farm, but it is used under significantly to be desired (as we have discussed in different conditions. Thus, while it Chapter 4). Statistical estimates based would be prudent to check on the on such data are likely to have performance and breakdown frequencies of considerable margins of error. such equipment under power plant Deficiencies of data when they relate conditions,* it is inappropriate to use to hundreds of pieces of equipment and such data as primary source for accident process details, and which are of such or breakdown probabilities for SRP's enormous magnitude, cannot possibly be Waste Farm operations. made up by technical judgment of a few This leaves only one document as the experienced people. Further, primary source of data for the probability reliance on Savannah River Plant calculations -- The Fault Tree Dam employees' opinions to estimate the Baok Statistical calculations based on probabilities of failure contains this document will, general, be of inherent conflicts of interest. A dubious validity, to say the bast. finding of high failure probabilities Good statistical estimation requires would reflect unfavorably upon the a clear philosophy of data recording and a uniformity of recording data over periods comparable to or longer than the periods for which estimates are to be

*It would be interesting to know how the breakdown probabilities for Tank Farm equipment have been changed after the 1979 accident at the Three Mile Island nuclear power plant. quality of the technical work and the Thus, for example, if the recorded inadequacy of corrective measures. It data yield any average failure rate for might also jeopardize the jobs of the a piece of equipment of once a month, very personnel making the estimates if the confidence in that estimate could there was a finding that the vary considerably, With excellent data probabilities of serious damage to the -- that is, data covering many identical public were large because such a finding pieces of equipment, under the same might imply a lack of due care or operating conditions, uniformly recorded responsibility. All of this would be over periods of time much longer than true even if the management were the average time between failures -- we committed to a scientific and thorough would be able to use a maximum failure evaluation of the dangers of plant rate of a few percent more than the opera tion. average rate yielded by the data with a Judgment of technical personnel can, high confidence level. This might be in general, only be a supplement to an the case with some estimates for evaluation based on reasonably equipment at the Tank Farm which have comprehensive data. It cannot make up failure rates of once a year or greater, for lack of essential data. on the assumption that all such This essential inadequacy of the data failures have been recorded for the six is compounded by the way in which the year period 1977-82. (However, as noted statistical estimates are used and in Chapter 3, the frequency and nature presented. Given the defective data of the recording changes does not lend base, it would be more reasonable to itself to having a high confidence that calculate a range of failure rates. all equipment failures and operating Instead, only one number, an average, is problems were recorded during this, or presented. It is then used as if it any, period.) were utterly reliable. That is, the In the case of equipment such consequence is calculated using these as tanks or evaporators for which the average rates as if these rates are the recorded failure rates are comparable to actual failure rates -- with 100 percent or smaller than once a year, the certainty. In practice, there is some Bank provides an inadequate data base, uncertainty associated with any estimate particularly due to changes of recording of failure rates, This will vary with procedures. Under such circumstances an the adequacy and amount of data, with estimate of failure rate in which one the specific piece of equipment and with could have a high confidence might well the probability model used. Sound be substantially larger than the statistical practice, particularly in estimate based on the recorded data, the presence of inadequate data, is to especially since many tank leaks have estimate ranges for parameters using 95 explicitly been excluded from the data percent, 99 percent or higher confidence base as discussed in Chapter 3, limits. The higher limits would be used Consequently, improving the calculations if consequences of misestima tion are by requiring high confidence levels in serious. The upper failure frequency, the estimates of failure rates of such with the specified confidence limit, equipment could well result in much should form the basis of further estima tion of consequences, higher estimates or risks compared to applied to cases where some real data those given in the official has been compiled. publications. In addition, it is good The DOE and DuPont approach to statistical practice to estimate the evaluating probabilities of system variance, or spread of the quantity failure is also defective. For estimated around the average. For instance, the probability of a waste example, if the average interval between tank explosion resulting in a collapse breakdowns of a pump is 60 days, then it of the tank roof is estimated as is,important to know whether most follows: failures occur at or near the average of 60 days, or whether the average is Waste Tank Explosion. A hydrogen composed of widely bspread values, for explosion in a waste tank requires example from 10 to 100 days. This the successive failure of several practice also appears not to have been equipment or procedural safeguards: followed by DuPont and DOE. For some events, such as those which o Failure of tank ventilation are dependent on social factors, or for system. which- requisite technical data are not available, there appears to be no o Failure of pressure alarm to statistical basis for the probability detect ventilation failure or estimates. For example, DOE has failure of operating persortnel to considered the consequences of possible heed the warning. abandonment of the Tank Farm without th e removal of the wastes in it. The basis o Spark initiation in tank after of its assumption of a probability of explosive gases have been ope chance in one hundred thousand per generated in the tank. yc'ar is the general assertion that the "waste tank farm is less likely to be o Failure of procedural sa feguards abandoned than other engineered (in routine check of blower facilities that have been maintained for operation, routine measurement of centuries such as dikes and hydrogen composition in gas space aqueducts."(4) While some dikes and of waste tank, etc.) to detect aqueducts have been maintained for and correct ventilation failure. centuries, many, perhaps the majority, have not. DOE does not cite any Based on estimates of the individual analysis of the matter, nor does it probabilities oj these conditions, a justify using a probability of hydrogen explosion is estimated to hav5 abandonment that might be orders of a probability of approximately I x 10 magnitude lower than that for aqueducts [one in I,OOO J per year, and dikes. When probability The waste tank explosion assumptions are of the 'hand-wavinga postulated ... involves failure and variety, as appears to be the case with collapse of the tank roof. It is this example, it is meaningless to estimated that one tank explosion in I0 discuss improving them by the use of would result in such an extensive statistical techniques which can be accident. The probability of the waste expression for the during construction and that the primary probability of a failure in containment of one tank has already the next time period t, P(t) leaked. = I-A t) reduces approximately to P(1) Earthquake Potential (events/hr) (I hr) =At . This expressiott was used in In 1886 Charleston, South Carolina, calcula!io~ts throughout the experienced a severe earthquake, with an analysis. estimated intensity of X on the Modified Mercalli scale.* This earthquake had an This statement does not explicitly intensity of VIl to VIII in the SRP rule out the possibility that failure area, which is about 100 miles from rates may vary with time. That is, as Charleston. The earthquake was felt equipment grows older, the possibility over a 2 million square mile area and of failure might increase, in some cases caused damage as far away as Chicago, dramatically. However, in the Illinois. This intensity of VII to VIII application of the model, constant on the Modified Mercalli scale, with an probabilities of occurrence appear to be acceleration of 0.26, is the design used throughout to cover long periods of basis earthquake for SRP. time. This might be adequate for Until 1982, the U.S. Geological equipment, such as reel tapes, which are Survey (USGS) rated "the likelihood of a frequently repaired and replaced, but it Charleston type event in other parts of is not a reliable way to estimate risks the coastal plain and the Piedmont [as] involving long periods, particularly very low." But the USGS altered its over periods which are comparable to or views dramatically in 1982. The greater. than the entire period of Charleston region and elsewhere along observation. This is especially crucial the coastal plain indicated that "...the in estimating the probabilities of general geological structure of the events relating to the integrity of the Charleston region can be found at other tanks. locales within the eastern seaboard DuPont estimates the frequency of (Appalachian, Piedmont, Atlantic Coastal tank failures to be 1 per 200 tank-years Plain, and Atlantic Continental for Type I and I1 tanks and 1 per 17,000 Shelf)."(l 1) Furthermore, the USGS tank-years for Type 111 tanks.(lO) The stated that "...the historical record is latter figure means that one would not, of itself, sufficient grounds for expect one leak in 500 to 600 years in a ruling out the occurrence in these other Tank Farm consisting of 30 Type 111 regions of strong seismic ground motions tanks. The estimated frequency of similar to those experienced near failure of 1 in 17,000 tank-years is Charleston in 1886."(12) They advised made on very inadequate data. Little the Nuclear Regulatory Commission (NRC) confidence can be placed on it, that the design basis for nuclear particularly when we note that corrosion facilities should be determined on their pitting has already occurred in 14 tanks revised assumption of a major earthquake

'Modified Mercalli Earthquake Intensity Scale. Source DuPont Safety Analysis Report; Liquid Radioactive Waste Handling and Storage Facilities (200 Area), DPSTA 200-3 August 1978; p. 6-59. occurring anywhere in the region and no surface offset due to earthquakes including the SRP site.(] 3) are known!' This means that the causes The Nuclear Regulatory Commission has of the Charleston earthquake are not yet taken the USGS seriously and has well understood.There are "literally launched short and long-term thousands of crustal structures" known probabilistic and deterministic studies. in the eastern seaboard, which if they In explaining their research plan to were active, "could produce strong the Commission, the NRC staff noted that earthquakes. But none have been shown to "the November 18, 1982 letter from the have been active for the last 2 million USGS represents not so much a new years."(l6) However, the difficulty is understanding but rather a more explicit that evidence of past inactivity cannot recognition of existing uncertainties be used for ruling out future large with respect to the causative structure earthquakes -- hence the "strong and mechanism of the 1886 Charleston elements of speculation" in any earthquake." Put more simply, the probability numbers. confident pronouncement by various experts that "a major earthquake near Calculating the Consequences SRP is improbable,"(l4) is simply speculation, according to the NRC staff Some of the most serious lapses in in 1982: the analysis of risks lie in the calculations of the consequences of Many hypotheses have been equipment or process failures or of proposed as to the locale in events like earthquakes or hydrogen the eastern seaboard of explosions. It is here, in their future Charleston-size eagerness to prove that nothing could earthquakes. Some of these possibly happen to gravely endanger the could be very restrictive in public, that the Department of Energy ,' location while others would and its contractor DuPont descend allow the earthquake to frequently into the realm of the recur over very large areas. incredible and the absurd. Presently, none of these The official reports present hypotheses are definite and calculations relating to a couple of all contain strong elements dozen "primary events -- that is, of specutation.(l5) initiating causes of releases of radionuclides. Our resources do not The NRC staff found "the permit an evaluation of each one, but primary problem with seismic hazard many of the more serious ones are characterization of the eastern seaboard similar since they involve ruptures of is that no causative mechanism for equipment containing large quantities of seismicity has been identified to date high-level waste. We will therefore concentrate on evaluating the official analysis of the consequences of earthquakes and of hydrogen explosions. Table 5.1 summarizes the major effects of earthquakes of various intensities on Tank Farm operations as wrong as we will discuss in Chapter 6. 'calculated by DuPont. For intensities The waste is assumed to be immobilized VII or less on the in the soil, and if it does reach the Modified Mercalli scale, no damage or water it is assumed to decay before it releases are assumed. An earthquake of reaches the surface wa.ter, before anyone intensity VI1 would cause difficulty in uses it. The possibilities of standing, plaster and tiles to fall, and substantial changes in groundwater use, furniture to break.(l7) At least some of or in hydrology (known to be caused by the calculations appear to be based on severe earthquakes) are explicitly rather optimistic assumptions. dismissed. For example, the calculations assume No synergistic interaction with that the cracks and corrosion of the events the earthquakes might cause tanks will not compromise the ability of elsewhere in the Savannah River Plant the tanks to withstand a "design basis appears to have been considered. For earthquake."(l8) These calculations instance, it is well known that the appear to rule out any brittle cracking presence of organic compounds like of the tank walls even though this has tributyl phosphate in the soil can cause been acknowledged as a possibility by a tremendous acceleration of the both DuPont(l9) and the Department of migration of some radionuclides - Energy.(20) plutonium in particular. Tributyl In earthquakes of intensity IX or phosphate ,being the principal solvent greater, rupture of some waste tanks and used to separate plutonium and uranium piping, etc., is explicitly considered. from fission products, is present in The only two modes of release considered large quantities at the plant. It is "significantw are those which result in stored in tanks which would likely direct release to the surface water or rupture in case of an earthquake. That to the air. Thus, in the worst possible would mean that vast areas of the soil case of an intensity XII, where on the site would have tributyl destruction would admittedly be total, phosphate which would spread because of the maximum possible release considered the heavy rainfall in the area. The is 46 million curies to the surface effects of even so obvious a contingency water(21) which accounts for about less on the contamination of groundwater seem than one tenth of the total radionuclide not to have been considered. content of the tanks (1977 inventory). The clear implication of the The consequences of releases into earthquake calculations is that the the soil and groundwater - which would Department of Energy and DuPont do not amount to about 500 million curies in consider soil and groundwater the worst case (1977 inventory) -- are contamination to be serious matters in explicitly dismissed because "their and of themselves (that is, except as consequences and risks are they affect surface water). In that insignificant compared to those of case, one wonders why they have not release by the surface water and carried their argument to the logical atmospheric pathway.' conclusion: if.soil and groundwater The major premises for such an contamination pose 'insignificantn assumption have already been proven risks, why have tanks at all? Why not just dig trenches, being careful to be further illustrated by the official cover them with tarps (to prevent conclusion that suspension of particles suspension of radionuclides), and into the air would be one of the two discharge the wastes directly into them most important sources of risk to the with hoses? It would be much cheaper -- public. Thus, in an intensity IX and get rid of the problem of long-term earthquake a relative damage of just 25% management all in one stroke! to tanks containing about a million Equally egregious are some of the gallons each of waste is postulated. assumptions about restoration of plant This would result in the suspension of operation. For instance, the about 1.5 gallons per tank in the air calculations assume near total according to DuPont. (See Table 5.1). destruction (90%) in an earthquake of This is assumed to be the second most intensity XI. The result would important source of radiation to people. undoubtedly be a plant site that would A closely related assumption is that be extensively and intensely only the present generation would be contaminated. We should note here that affected and that cancers would be soil excavation work on a spill of 50 practically the only effect. Doses to curies, which would be tiny compared to future generations appear to have been the consequences of an earthquake ignored. We do not know if cancer or releases of hundreds of millions of other effects on fetuses have been curies (plus large quantities of other considered. toxic materials), had to be stopped The risk calculations also appear to because radiation rates of 500 rads per omit the effects of non-radioactive hour were encountered. (F-Area entry toxic materials present in the tanks. dated 03-12-76. Table 1, Part 11s) For example, there were about 190,000 Ignoring its own records and experience, pounds of mercury in the tanks in the management assumes that cooling 1975.(23) The risks and problems posed water to tanks would be restored after by the much smaller quantities of 90 days. mercury in the seepage basin have been How workers would be mobilized to acknowledged both by the EPA and by work in an area which would be a DuPont researchers (see Chapter 6). Yet veritable radioactive hell and which the risks posed by almost 100 tons of would have to be abandoned essentially mercury in the tanks, particularly to forever, or how they would work to ground water, seem to have been ignored. restore piping and power in such an The effects of the errors and environment, is not .described in the defects in both the approach to and the official reports. Neither does the details of the risk calculation is to question of what use it would be to yield probabilities in catastrophic restore cooling water supply to tanks events which are very small. When these that no longer exist seem to have been are combined with the selective addressed. estimation of the consequences of The absurdity of the radioactivity releases, the overall calculations and the determination to effect is to yield small risk or "dosea show that there could be no serious harm estimates. This is because wdosea even as a result of big earthquakes, can estimates are calculated by multiplying the probability of events with the Reducing damage from a catastrophe releases which are considered to an average risk is contrary to common "significant." sense. If one adopted this view, one Besides the serious deficiencies and might advise people to cross the street errors in the analysis, this method of against "DON'T WALK" signals, if calculating risks of catastrophic events calculations showed it would save time is not a well-considered one. It assumes on the average. Similarly, no healthy that an event of a large probability young parent would be advised to with relatively small consequences is purchase life insurance because the equivalent to an event of small return on investment is higher in a bank probability with serious consequences.* account on the average. We reject risk The soundness, or lack thereof, of the estimates based on statistical av.erages procedure in evaluating risks can be in such cases because average gains illustrated by looking at it from the cannot compensate for the losses in the point of view of an insurance company or specific instances. a benk. An insurance company with a For this reason, an average measure thousand dollars in capital could write of risk is not a very meaningful measure many times a thousand dollars of to be used for catastrophic events. For insurance provided that the maximum these, the consequences, once a amount of any policy was small compared precipitating cause of catastrophe has to the capital. A prudent insurer would occurred, are more significant.* This not sell one policy for several million procedure should not be applied to dollars if the ability to pay is a few events that threatens substantial or thousand, even if the probability of complete destruction of large areas or having to pay on the policy was very numbers of people. For these kinds of small. This is because the consequence cases, fie calculations of risk must bc of the event, if it occurred, would be based not onlv on exoected orobabilities bankruptcy. The total amount of all of occurrence but also on the conseaucnces outstanding policies can safely be if ca tastrouhe strike$ higher than the capital only if each one Thus, the seriously defective is kept below it. Similarly banks calculations of DuPont admit to a generally do not make single loans possible dose of 34 million person-rems which art larger than their invested as the result of an intensity XI1 capital, even though the cumulative earthquakes.(24) On the basis of amount of all loans is usually much dose-risk relationships established by larger. In fact, banks are required by various scientists and committees, a law to keep the largest single loan to range of subsequent health effects can well below the capital invested. be estimated, though this range is substantial. (See Chapter 7 for

*This method is not peculiar to DuPont. It is used all too often by DOE and is widespread throughout the nuclear industry. *In theoretical terms: one must not only consider the probability of the intersection of two events El and E2 given by the product of the probability of El with the conditional probability of E2 given El, (pr(ElnE2) a pr(E1) x pr(EZ/El)). One must also consider the conditional probability, pr (E2/E1), by itself. detailed discussion.) The 34 million years in the first generation and about person-rems estimated by DuPont to occur 460,000 years for all generations.(28) from the rupture of the high-level waste Radiation geneticists assume that the tanks at SRP could give rise to anywhere bulk of adverse hereditary effects from from 11,000 to 230,000 excess cancer radiation will occur during the first cascs.(25) These estimates are based on three generations or 90 years after the 1978 data presented by DuPont. Current parents art first exposed. inventories (1984) are about 30% higher. In the case of a magnitude X Moreover, we have discussed above that earthquake, comparable to the 1886 the quantities of plutonium and other Charleston earthquake, the consequences actinides in the wastes could be based on DuPont release estimates would several-fold higher than the 1978 DuPont be about 60% of the above figures. data. The casualty estimates, even for These estimates are in no sense the limited model used by DuPont, based definitive. The large range of on current inventory data would likely uncertainties are ample evidence of this be still higher. fact. But for purposes of public health The calculations of risk of genetic planning, such estimates help us to harm also have a wide range. With the understand the general parameters of exception of cytogenic data from monkeys catastrophic risks and their possible and humans, genetic risk estimates have effects on the society, as best we know been heavily dependent on extrapolations them today. to humans from experimental rodent It is even more difficult to assign studies. These risks are expressed in monetary values to the health effects of terms of (a) first generation serious radiation exposure from man-made disorders, (b) cumulative genetic sources. Nonetheless, we provide some ef f ects over time (equilibrium), and (c) rough minimal figures as supplementary effective years of life lost. Because of indicators of the material consequences the major uncertainties in predicting of dangerous activities. We have taken generational effects from the initial values from the scientific radiation exposure of the parents, we literature(29) on the societal costs of cite low and high range estimates. radiation exposure and have applied them Thus, if a 34 million person-rem dose to a catastrophic accident involving the occurred in the vicinity of SRP from a high-level waste tanks at SRP. Assuming major accident at the Tank Farm: (a) an inflation adjusted discount rate of first generation genetic diseases could 6% per year, the estimated excess cancer range from 200 to 2,500;(26) (b) cases due to a 34 million person-rem cumulative effects over time could be dose could cost from $8 00 million to 2000 to 37,000;(27) (c) effective years $ 14 billion. Adverse genetic effects, of life lost are estimated at 80,000 which include long term institutional care for first generation offspring damage could cost from $90 million to over S1 billion. These costs do not reflect property damage or loss. Some accidents could also .cause very large releases and damage. Consider, for example, an explosion in a liability for catastrophic radiation waste tank due to a build-up of accidents. Private contractors were hydrogen. Such a build-up due to 'the also held accountable for accidents failure of ventilation systems in a stemming from their negligence. In 1957, tank. A severe explosion could cause however, a system was created by law the rupture of the tank and a roof which set a limit on liability for collapse. We have cited above the accidents at 500 million dollars. Known official probability estimates for such as the Price/Anderson Act, this system an event, along with a discussion of was added to the 1954 Atomic Energy Act their weakness. In the worst case mainly as an incentive for private officially postulated, an explosion utilities to purchase nuclear power could cause the release of about 7 plants, who were fearful of liabilities million curies of radioactivity into which could bankrupt them. The act also the soil, water, and air. covered federal nuclear facilities under The official calculation of the section 170d. emcts of such a release hae the same The central feature of the act is a general defects as those discussed limit on the total aggregrate liability above. for earthquakes. Even with the which arises from a major nuclear underestimation implicit in these accident. Although the act has been estimates, a severe hydrogen explosion changed over the years to increase could result in sufficient radiation coverage for nuclear power plant doses to cause from 1,000 to over accidents, the limit on liability for 20,000 excess cancer cases, as well as Department of Energy facilities is set hundreds or thousands of cases of at the 1957 level of $500 million. diseases from genetic defects in future As it applies to DOE nuclear generations. The costs, txcludina facilities, the Price/Anderson Act pronertv losses tnv- provides total indemnification for all damane. could range up to about 51.4 government contractors of billion. or nest-. Currently The inclusion of property there are over 100 DOE contracts with losses, environmental damage, losses of Price/Anderson protection, covering agricultural land and crops would about 50 prime contractors and 70,000 drive the total damage estimates subcontractors.(31) A total of 280 DOE considerably higher. facilities are covered under the Act.(32) Accident Llablllty and Accountablllty According to the Department of Energy: During the 1940s. it was recognized that certain federal nuclear facilities essentially, the contracf had the potential of causing indemnity provides that, in catastrophic radiological accidents the event of a nuclear involving massive loss of life, serious incident, the contractor or injuries, and extensive property any other person who may be damage. Up until 1957, the federal liable would be indemnified nuclear weapons program had no limit on by the DOE, up to the statutory, specijied ceiling prudence of the judgment to self-insure oj $500 million, for any from the first dollar of the indemnity legal liability resulting coverage."(35) from a covered nuclear The enormity of the possible damage, iricident arising out of or as well as the lack of scientific in con~iection with soundness of the analysess of DOE and contractual activity.(33) DuPont in their assessments of possible catastrophic events, leads us to If an accident occurs and the limit conclude that making contractors on liability is exceeded, as would accountable for major accidents stemming likely be the case if major ruptures from their negligence would add a much occurred at SRP's high-level radioactive needed incentive for safety. Further, waste tanks, an act of Congress would be the DOE liability should be made required to provide adequate commensurate with the damage that would compensation. Unfortunately, the speed occur. This would be another built-in by which the Congress bas moved to incentive for the federal government to provide even partial relief has been meet 'state-of-the-art" health and less than timely. For example, in 1947 a safety standards. chemical explosion caused by the collision of two ships under government Ref erences contracts killed 570 people and injured another 3,500 near Texas City, Texas. I. Energy Research and Developmcn t The entire Texas City dock was destroyed Administration. Waste along with residences and industrial Q~ewSa vw R iver Plant. Aiken, facilities causing property damages Final Environmental Impact estimated at S80Q300 million. Eight Statement. ERDA-1537, National Technical years later Congress passed disaster Information Service, Springfield, Va. relief awarding victims a mere $21 September 1977. million in inflated 1955 dolIars.(34) 2. E.I. DuPont de Nemours and Co., Additionally, the act does not cover Safetv Analvsis rebort: . . Radioactive Waste and wvang the cost borne by states and citizens in -- - -- the case of an incident at a federal res (200 Area). Savw nuclear facility which requires a River PmDPSTSA-200-3, Aiken, S.C. precautionary evacuation. Nor does it August 1978. cover acts of sabotage or terrorism. 3. Ibid, p. 6-4. High level radioactive waste 4. Alteration for Long-Term repositories are also not covered under ent af D_ef-level the Act. wveW- SavwRiver Pu The indemnification system for DOE en. S.6 Energy Research and nuclear contractors is such that Development Administration publication contractors are not required to purchase no. ERDA 77-42/1. National Technical private insurance since the government Information Service, May 1977, p. V-34. would have to pay the premiums. 5, Op. Cit. reference 1; p. 111-19. According to DOE, "our experience to 6. Op. Cit. reference 2; p. 4-35. date, of course, completely supports the 7. U.S. Atomic Energy Commission. 8. Op. Cit. reference 1; p. 111-129. Effects of Ionizing Radiations (BEIR 9. Op. Cit. reference 2; p. 6-13. Ill). The Effects on Po~ulations of 10. Ibid., p. 4-19. Ex~osure to Low-Levels of Ionjzins I I. J.F. Devine, Assistant Director Padia tion, 1980, NAS-NRC, p.7. for Engineering Geology. U.S. Geological 27. Ibid. Survey, letter to Robert Jackson, Chief, 28. Abramson. S., "The Genetic Geosciences Branch, Division of Impact of Low-Level Ionizing Radiation: Engineering, U.S. Nuclear Regulatory Risk Estimates for First and Subsequent Commission. November 16, 1982. Generations." Presented at the Annual 12. Ibid. Meeting of the National Council on 13. Ibid. Radiation Protection and Measurements, 14. Briefing paper, U.S. Nuclear April 1984, Regulatory Commission, November 20, 29. Voilleque, P.G. and Pavlick, R.A., 1982. "Societal Cost of Radiation 15. Ibid. Exposure," mthPhvsicg, Vol. 43, 16. Ibid. No. 3. September 1982, pp. 405-409. 17. Op. Cit. reference 2; p. 4-36. (See table 1,) 18:Ibid., p. 4-38. 30. Hacker, Barton, "Elements of 19. Ibid., p. 6-38. Controversy: A History of the Nuclear 20. Op. Cit. reference 1; p. 111-43. Weapons Testing Program," Praf t, 21. Op. Cit. reference 2; Table 6-27. Reynolds Electric Co., Las Vegas, 22. Ibid., p. 6-62. Nevada. 23. Op. Cit. reference 1; p. 11-38. 31. US. Nuclear Regulatory 24. Op. Cit. reference 2; Table 6-27. Commission Policy Issue Paper: BB, 25.The lower estimate has been derived fie Conaress as 'Reauired bv Section from the National Academy of Sciences BEIR Fnerev Act of 1954 I11 report, The Effects on Po~ulatipg am- US. Department of Energy, ure to Low Lev& of 10- . . August 1, 1983, &&tion, 1980; p. 195. The relative 32. Ibid. risk, LQ-L model risk cited is 182 33. Ibid. cancer deaths per million rads, giving 34. Kehoe, K., I Jnavaibble at Any an estimate of 6,200 deaths for 34 Pric~Environmental Policy Institute, million rads. The cancer incidence Washington, D.C., 1981. is obtained by multiplying the figure 35. Op. Cit, Reference 23. for deaths by 1.75. The high estimate. . is obtained from John Gofman Radlatlon Sierra Club Bo- -, -, 1981. Gofman estimates cancer deaths at about 3,800 per million rem yielding about 130,000 deaths for 34 million rem, The estimate for cancer incidence is 1.75 times this figure, or about 230,000 cases. 26. Committee on the Biological Table 5.1

~stimatedEarthquake Damage to Major containment Systems. Earthquake Intensity at SRP VII None leading to release. VIII Covers of evaporator cells fall into cells and shielding blocks tip over. Pipes and evaporator pots are ruptured. Waste sprays out top of evaporator cells at 20 gpm for 10 minutes. 0.01% converted to 10m spray. The concrete shells of some waste tanks crack, but the steel liners remain intact. Soil liquefies in the H-Area tank farm and slumps away from 11 waste tanks built partially above the grade. Some transfer lines may also be uncovered. The transfer line from the H-Area CTS pump tank to the waste tanks fails and spills 50 qpm to the ground surface for five minutes. IX Normal and emergency electric power lost. OA25 relative damage done to evaporator cells waste tanks and underground piping. Evaporator cells crack and pots rupture, spilling 0.25 of their contents to the ground surface (1500 L/evap). Roofs collapse on OA25 of waste tanks splashing 57,000 L of supernate per tank to the ground surface, and atomizing 52 L of supernate per tank (

Source: DuPont Safety Analysis Report, DPSTSA-200-3, Aiken, SC, August, 1978. CHAPTER 6 EN\'IRONhfENTAL CONTAhllNATION There are four kinds of The permissible levels of discharges en\lironmental contamination from are calculated from the 500 millircm radioactive and non-radioactive limit using a "dose-to-man" model. substances which occur at the Waste Tank These limits are very large. They are Farm opera tion: shown in Table 6.1. Overall they allow releases of over a million curies per o routine discharges to the year, including 800,000 curies of environment, notably of large tritium. Large releases of alpha quantities of contaminated emitters, like plutonium and uranium-235, liquids to the retention and are permitted. For instance, they allow seepage basins; the release of 600 curies of U-235 per year, which amounts to over 25 tons. o releases due to improperly These legally allowed limits, under working equipment, such as air which the Department of Energy and its f iliers; contractor for the Savannah River Plant operate, allow public radiation o accidental contamination, such as exposures. which are higher than those that from tank and pipe leaks; for commercial nuclear power plants by a factor of twenty. o disposal or abandonment of In practice, the routine releases contaminated materials and that have been documented so far are equipment used on the Tank Farm. considerably below the limits shown in Table 6.1 for most radionuclides. The Seepage Baslns However,thtse limits are applied to "the point at which e jfluerrts pass the The Savannah River Plant management site boundary" and not the more allocates to each part of the operation conservative EPA standard that places "release guides" for routine limits at the point of discharge onto radioactivity discharges to the the ground. In a previously secret environment. These are based on the report on the Savannah River Laboratory nature of the specific part of the plant (SRL) seepage basins, it was noted that and on overall total release limits "most discharges would be above ERDAM calculated to keep doses to the -524 [DOE'S] concentration guides," if surrounding population below certain limits were applied to the point of levels. discharge,(2) The non-conservative Legally the management is constrained practice of calculating release limits to keep discharges of radioactivity so at the plant boundary has enabled the that the dose to an individual member of DOE and DuPont to develop operating the public from its activities will not release guides which art well below the exceed 500 millirems per ycar.(l) "legal" limit in most casts. It has also made it impossible to assess the performance of these standards for individual facilities. Each area is allocated operating monthly release half-life radionuclides to decay beforc guides which may be exceeded if entering the soil and water, but it is "required" by operations. They not likely to have any significant frequently are exceeded. effect on the longer lived isotopes like Liquid releases to the seepage basins cesium-I 37 and strontium-90. make up the largest volumes of routine Routine releases to the seepage releases from the Waste Tank Farm. The basins are not the only discharges of operating release guides for some radionuclides from the Tank Farm. radionuclides from the F & H areas, Improperly working equipment, such as which incIude the reprocessing plants air filters, contribute to plant and the tank farm, are shown in Table radionuclide releases to the 6.2. atmosphere.* The Data Bank contains a While these are considerably below number of entries showing air filters the DOE'S legal limits shown in Table working well below specification 6.1, the releases from the other parts efficiency. This includes the HEPA (High of the operation, notably the Burial Efficiency Particle Activity) filters Ground have been much higher than which are installed on the waste tanks anticipated (see below). Further, the to filter out any radionuclides from the "dose-fo-man" model used to translate air in the primary tank and the releases of radionuclides to the annulus.** On one occasion two tanks were environment into doses of radioactivity operated for five days without air to people is a linear model whose filters. (See H area entry dated inadequacies have bten criticized by the 09-05-80, Table 2, Part 11.) We do not Environmental Protection Agency.(3) have estimates of the total Many entries into the Data Bank radioactivity which has been released show that the operating release guides due to poor filter operation. to the seepage basins are frequently Similarly, there have been problems exceeded on a monthly basis, and with radioactive cesium removal. On one sometimes on an annual basis as well. recorded occasion cesium-137 was (See for example, the H area entry dated released because the cesium removal November 81, Table 2, Part 11.) column was inadvertently bypassed for 11 Besides such routine releases, there hours (F area entry dated 1-15-65, Table have been releases due to operating 2, Part 11): 'Cs release, Cs columns problems in the plant. To prevent this inadvertently bypassed for 11 hours." radioactivity from going to the earthen No estimate of the quantity released is seepage basins, lined retention basins provided in the Data entry. have bten built to which large volumes There are some entries in the of liquids contaminated fb a greater &,& whose environmental implications degree than usual are diverted. The may be serious but which are not retention basins bold some of the explained sufficiently. From the data raaionucfides instead of the seepage given it is not possible to determine basins. This allows some of the shorter whether environmental contamination

C *Some of the radionuclides released to the seepage basins enter the atmosphere later by volatilization. This applies in particular to radioiodine and tritium. Some of these airborne releases come down with the rain in areas outside the plant. **Air circulation in the annulus was provided after many tank leaks were discovered. It was found that dry air helped to deposit salt on the cracks which prevented most cracks from leaking further. occurred. Perhaps most remarkable among On an average, about 30 million these is an entry dated March 1981, (F gallons of radioactive effluent is area entry dated 03- -81, Table 2, Part discharged annually onto the F & H area 11) which reads in full: seepage basins. Thus, in all an estimated 2.5 billion gallons of A valving error in JB-title contaminated liquids have been resrtlted it1 a trartsfer of 494 discharged during 1953-78.(4) The grants of Pu [pluronittnt] ittto chemical composition of the effluent is catrjlon tattk 9.7. This soliirion highly variable. It has varied from was subseqrierttly fed to the LAU' highly acidic (pH = 0.6) to highly basic [Low Acfi~*ifyWasre] evaporator (pH .P 13.2). Its content of salts, artd discarded. organic compounds and other materials is also highly variable since the water There is no indication as to where which is used to "clean-up" operating this large quantity of plutonium was areas is, in part, collected and sent to "discarded." the seepage basins.(5) Tables 6.3 and Finally, radionuclides, including 6.4 show the discharges of radioactive plutonium, are released when tank and non-radioactive wastes to the risers (pipes) and other similar pieces seepage basins. By 1976 over half a of equipment are opened. The Rata. million curies of radionuclides had been Bank has dozens of entries indicating discharged to the seepage basins such contamination during the two-year (decay-corrected figure). (See Table period 1976-77. .Yet no environmental 6.3.) In terms of radioactivity, the contamination data or worker exposure principal component has been tritium, data are given. There is also no but large quantities by weight of their explanation of why these entries occur radionuclides have been discharged. predominantly in this period in the )I Thus, while the total quantity of area. (See Tables 1 and 2, Part 11, and natural uranium (99.3% U-238 and 0.7% Chapter 7.) U-235) discharged up to 1976 was The importance of the seepage basins estimated at 13 curies (a small to environmental contamination lies in percentage of the total activity), it the fact that here the radionuclides amounted to almost 40 tons of natural come into direct contact with the soil uranium by weight. This material will and water. This n6t only a means of stay radioactive essentially forever, routine contamination, but yields some since the half life of U-235 is 710 important scientific data regarding the million years and that of U-238 is 4.5 assumptions on which risk calculations billion years. are based. Records pertaining to non-radioactive discharge are less complete, but DuPont has estimated that between 1954 and 1973 about 4,500 pounds of mercury were discharged into the F & H basins. The cumulative volume of nitrates from 1961 to 1970 was estimated at 5,260 tons. (See Table 6.4) Large quantities of It is the position of the Savannah other chemicals are also routinely River Plant management that SRP discharged. discharges will not result in serious Mercury poses significant problems. consequences. Apart from the DOE has apparently only recently begun controversy regarding the dose to consider it a serious environmental calculations mentioned above, this hazard at SRP. Generated as an impurity position rests on several premises: of sodium hydoxide and as a catalyst to (a) ion-exchange will immobilize a dissolve spent fuel elements in SRP's number of radionuclides -- in particular reprocessing plants, the amounts of plutonium in the soil so that they will mercury discarded annually have varied never enter the groundwater; widely, according to official documents. (b) only strontium-90 and tiitium By the late 1970s a few studies had been will contaminate surface and groundwater done at SRP on the mercury problem. One in any significant quantities; report stated: (c) groundwater contamination is by itself of no consequence and since there Long term behavior oj will be no contamination of deeper nun-rdioacrive mercury in freshwater aquifers, the only soils of separalions areas is "dose-to-mana path would be via surface lagely unknown and jurure water. environmenral impacts are (d) decommissioning of waste difficulr to predict. [It also facilities is a low priority because of co~rcluded rhar mercury] ... in favorable environmental conditions. soil and groundwater may impose In the following discussion we show the most severe restrictions on that there is sufficient data to throw furure at the sire...( 6) doubt on the scientific validity of each of these assumptions. Indeed, some data Additionally, very long-lived fission indicates that they are invalid as products have been discharged into the general assumptions. F Bt H' seepage basins. They include Researchers for the Savannah River zirconium93 (950,000 years), cesium-135 Laboratory, in studies on waste (2.3 million years) and iodine-129 (16 discharges in the 200 Area, claim that million years). (See Table 6.5.) These only tritium and strontium-90 enter very long-lived fission products were groundwater in appreciable also not taken seriously until recently. quantities.(8) This is not reconciled A 1977 Savannah River Laboratory report with historical data regarding states, "the amounts [of very long-lived ground water contamination. fission products] expected to be present For some 30 years, radioactive and do not constitute a hazard." The study non-radioactive discharges at SRP have prefaces this claim by acknowledging severely contaminated shallow aquifers that "only limited studies have been beneath the plant. Moreover, in 1983 made of very long-lived fission products some of these contaminants were found in and available data is inadequate to a deeper aquifer used for drinking water estimate amounts in soil columns and on-site.* Hundreds of wells have been groundwater." drilled to measure and trace contaminants, particularly in the 200

*Lawless, W.F., The Dupont Management of Savannah River Plant Radioactive Wastes: A Report to the U.S. House of Representatives. Committee on Energy and Commerce." Nov. 27, 1983. Area where the largest discharges have The seepage basins in the occurred. However, only a limited separations areas have beconre amount of data is publicly available on the predominatit source oj the extent of groundwater contamination strontium-90 release to the from the 200 Area operations. [Savannahj river.(] I) There is general agreement regarding the contamination of groundwater by The flushing of strontium-90 and tritium and strontium-90. Substantial tritium into the groundwater system in amounts have migrated to surface the 200 Area has caused a steady outcrops. As regards tritium, DOE (then build-up in the shallow aquifer in the AEC) stated: form of radioactive underwater contaminant plumes. Groundwa ter 0f the 420,000 Ci f curies] of inventories of tritium and strontium-90 tritium sent to the [seepage] in these plumes as observed in well basins through 1975, 100,000 Ci readings are consistently in excess of has migrated to Four Mile concentration guides: Creek. The remainder is in transis in the groundwater or Concentration [of strontium-90j has decayed (ha1f-li fe 12.3 in groundwater near the years) or evaporated from the separations are basins range surlace of the basins.(9) from 47 to 1100 rimes the concentration guide in the Similarly, a considerable portion of F-Area...concentrarions will the seventy-six curies of strontium-90 nor approach the concentration discharged to the F and H area seepage guide until several hundred basins until 1975 had reached the years after basin groundwater.. Nine curies of this had decommissioning.(J2) travelled to the surface outcrop at Four Mile Creek.(lO) According to a previously secret 1972 Based on 1973 estimates, the average DuPont study of seepage basins, 44% and annual amounts which outcrop at surface 59% of the tritium discharged, streams are about 1.6 curies for respectively, in the F & H basins will strontium-90 and 11,000 curies for reach the Four mile Creek.(l3) Only tritium. According to a DuPont analysis about 10% to 21% of the strontium-90 of seepage basins' operation: will be removed by decay in each area. These estimates of groundwater flush

'Note that the figure of 76 curies of strontium-90 discharged to the seepage basin 1975 cite by DOE does not match up with the DuPont figure of 65 curies to 1976 in Table 6.3 above. rates arc subject to wide uncertainties. concentrations for non-volatile For example, in 1974: beta-emitters and alpha-emmiters by factors ranging from 100 to 1000. (See predicrirtg furtcre emergence oj figures 6.2 and 6.3.) Alpha srrortrircm-90 (jronr the H contamination of groundwater near the F basitrsj it; the creek is jar & H basins was often in excess of more uttcertaitt sitice the soil drinking water standards, sometimes by a cohrntt between the basins attd factor of 1000 at depths of 23 feet and the creek is not near complete as far away as 46 feet from the breakthrotcgh and grouttdwater basins.(l7) During this same period, corrcentrations provide no non-volatile beta-emitters (primarily bdication where the strontium radiocesium and radiostrontium) were is.(l4) measured in shallow aquifers at levels hundreds of times above pre-operation "background readingsn. Until 1956, Similarly, tritium and strontium-90 certain alpha readings in water were have -migrated with the groundwater referred to as "uranium or underneath the 195-acre Burial Ground, plutonium."(l8) This description was where used contaminated equipment, dropped in later reports and statements contaminated mercury (until 1968), were made that no plutonium was being contaminated soil, etc. are put into detected in surface or groundwater. In 20-foot-wide by 20-foot deep trenches. spite of these facts DuPont researchers The tritium plume in the groundwater confidently state that 'most of the underneath the Bur'ial Ground is so plutonium is bound in the top few inches concentrated that according to William to one foot of seepage basin sediment." Lawless* a former DOE engineer, it The researchers also stated that uranium 'would give a dosage of about 18 rem/yr 'is barely detectable in groundwater to anyone drinking well water from this separation areas."(l9) This would imply area and exceeds the EPA drinking water that uranium migrates through the soil standard by 3500 times and would not be very slowly. However, uranium migration safe for 100 years."(l5) (See at SRP maybe comparable to the more Figure 6.1.)* rapid rate of strontium90.(20) We have There b also evidence of not independently researched this contamination of the groundwater with controversy, but it is potentially very other radioactive and non-radioactive important in view of the large toxic materials. Previously secret quantities of uranium both in the reports covering pre-plant operations seepage basins and in the tanks. (1951-53) to the year 1963 (DPSP Cesium-I37 has also reached the series)(l6) show that groundwater groundwater under the seepage basins contamination due to the F & H area seepage basins was quite severe, often exceeding maximum permissible

*There are no drinking water wells from the shallow aquifers on the SRP site discussed here. We cite the drinking water standard in this and other cases as a reference standard only. though ERDA claimed that the another source of plutonium dumped into "concentration of 137-Cs in the it. In the early 1960s tributyl groundwater in equilibrium with this phosphate contaminated with plutonium soil is below the concentration guide was burned in open pans in the Burial given in ERDAM-0524."(21) Some of this Ground. The ashes were left on the soil cesium has migrated all the way to surface and they were eventually washed surface outcrops and is "detectable in into the trenches by rainwater.(26) Savannah River water by the routine This, together with the plutonium monitoring program."(22) disposed of in the Burial Ground, have The leakage of high-level wastes in provided strong evidence of the 1960 containing plutonium directly into migration of plutonium in the soil, the groundwater has been documented by particularly in the presence of organic Savannah River Laboratory reports.(23) solvents like tributyl phosphate. This William Lawless, in suppressed appraisal is not only important of itself -- it is reports written in 1982 and 1983 noted of basic significance to one of the groundwater samples in the 200 Area major design premises of SRP. indicating the presence of It is one of the principal plutonium.(24) assumptions of the risk calculations The appraisal report related that ion-exchange, that is, selective primarily to the Burial Ground which absorption of radionuclides in the soil, contained 400,000 curies of transuranic will prevent plutonium and other waste stored on pads, an estimated 4,000 alpha-emitters from migrating through curies of transuranics in cardboard the soil site into the groundwater. boxes, between 1 million and 2 million This assumption, which often takes on curies of radiocobalt and between 14,000 dogmatic proportions in some public and 20,000 curies of strontium-90 and documents, is crucial to the cesium-137. In this report DuPont, in rationalization of the prolonged response to a DOE question, admitted the operation of obsolete waste disposal presence of plutonium in the practices at SRP - long after they groundwater: should have been discontinued. The Savannah River Plant waste Current plutonium levels in disposal system is based on the this groundwater are quite low experience of the "semi-works," a and generally do not exceed prototype nuclear explosives production drinking water standards. Only facility constructed in 1943 at Oak one well (G-21) in the area o/ Ridge, Tennessee, and the Hanford the solvent spill exceeds [the] Plutonium Works constructed near 5 pc/L [pico curies per liter] Richland, Washington, in 1943. SRP was standard.(25) the third and last plutonium production complex constructed and operated by However, according to testimony by DuPont for the federal government. The William Lawless, four wells out of forty common denominator of these operations had exceeded the drinking water standard which justified their design basis was for alpha-emitters (see below). the bresence of a large land base with The Burial Ground bas also bad abundant water supplies. These combined factors made a "buffer zone" whereby time of less than twenty years for radioactive and non-radioactive plutonium. Yet, the official risk chemicals discharged onto the analyses deny the possibility of such environment would be diluted to "safea migration in the case of accidents or levels at the plant boundary. At the other catastrophic events such as same time, heavy industrial water earthquakes. (See Chapter 5.) demands could be satisifed. The accelerated migration of The "buffer zone" concept assumed plutonium in the presence of organic that, more often than not, the most solvents has been observed at other dangerous waste products discharged into locations and reported in the scientific the environment could be absorbed literature.* For instance, Jess M. indefintely in the soils. This Clevelan and Terry F. Rees of the U.S. assumption is not supported, even by DOE Geological Survey reported on a similar and DuPont data. First of all, the phenomenon at the nuclear waste dump at build-up of waste chemicals in soil has Maxey Flats, Kentucky.(29) Indeed, the a finite absorption capacity. When that U.S. Geological Survey has also warned limit is reached, the molecular barriers against the generalization that the soil are overcome and a "breakthrough" is always an effective barrier to occurs. This has been observed with radionuclide migration: strontium-90 in the soil near the F & H area seepage basins.(27) This "breakthrough" effect, as well In some of the qualitative as migration of radionuclides prior to literature on waste disposal a 'breakthrough," is enhanced by the pancealike aura surrourtds the presence of organic solvents such as term "ion-exchange." In such tributyl phosphate in the soil. This literature it is implied that has been demonstrated by events both at when a11 else fails SRP and at other locations. ion-exchange processes will DOE has officially acknowledged that prevent movement of an accident involving a 40-gallon spill contaminants to points of water of contaminated solvent in 1962 and a we.(3I) leak of solvent from two storage tanks in 1968 (tributyl phosphate) contributed Changes in the rate of groundwater to the migration of plutonium as well as flow are another factor which could fission products into the groundwater cause the acceleration of radionuclide over a 12-acre portion of the Burial movement. As documented below, DOE and Ground.(28) In fact, the solvent DuPont explicitly assume in their appears to have accelerated the movement evaluations of accidents and risks that of radionuclides to such an extent that groundwater flow pattern$ will not DuPont has acknowledged that by 1981 change. We have already mentioned the 'most of the released activity from the possibility of change in these patterns spilled solvent has already reached the due to events such as earthquakes. water table330) This is a migration (Chapter 5.) They may also change due to changes in patterns of usage of Soil Intrusion by "cleavage" from rainfall also can enhance vertical plutonium migration: groundwater. Current groundwater flow groutrdwater level beneath adjacetrr rates are roughly 0.5 ft/day in the F seepage basins, thereby increasing the area and 1 ft/day in the H area.(32) But terrdertcy of coritan~inarrrs lo enter the DuPont has acknowledged that "water ground rvater attd nrigrate."(39) movement in the area is complex because sediments are not uniformly DOE and DuPont claim that the decpcr permeable."(33) Groundwater withdrawal, aquifers are protected from contaminants in the SRP area principally from the in the upper systems because of the Tuscaloosa formation, is estimated to be "extensive upward vertical gradient 70 cubic meters per minute.(34) SRP is between the Tuscaloosa and the estimated to be pumping about 24.3 cubic Congaree."(40) This statement is not meters/min. or roughly about one third reconciled with their own data which of the total withdrawaL(35) shows that upward vertical gradients, on Major withdrawals of this magnitude site, are being significantly reduced can have the effect of changing the due to present withdrawals.(41) Pixie relationships between the aquifer Newman, a University of Wisconsin systems, which in turn encourage hydrogeologist, has concluded that contariiinants under the 200 Area waste "pumpage in the H Area has reduced the operations to migrate and possibly enter vertical head difference between the the deeper systems. Over the next few Tuscaloosa and the Congaree to less than years, DOE plans to expand plutonium or equal to 0.6m (2.0 ft)."(42) production at SRP which will increase Further, she found it conceivable that withdrawals.(36) Moreover, in recent the "effects of additional L-Reactor years off-site pumping of the pumpage may induce flow and spread Tuscaloosa for irrigation has increased contamination away from inactive as well rapidly near SRP in Allendale and as active waste site, There is little Barnwell Counties.(37) doubt that (expanded plutonium "If this trend, coupled with production) will accelerate anticipated increases in groundwater use contamination problems in the F & H at the SRP facility continues," writes areas (nitrates and mercury) and the M Dr. Yaron Sternberg, a University of area (degreaser solvents)."(43) Maryland hydrogeolosit, 'the present The major source of evidence used by head difference of about 12 feet at the DOE/DuPont to suggest that contaminants L-Reactor may decrease and likely be in shallow formations will not enter the reversed."(38) deeper systems is a map "constructed by In more general terms, the EPA has subtracting two peizometric maps for agreed with Sternberg that additional which data are somewhat sparse."(44) The pumpage of groundwater by SRPUcould accuracy of this map is questionable cause additional drawdown of the since it is not based on data collected from several "nested" observation wells which measure pressure relationships at 2 or more depths within each hydrostratigraphic unit. Moreover, this interpolated map does not factor in anticipated cbanges in water flow from increased groundwater pumping. In essence, DOE and DuPont EPA standards. A 1977 Savannah River assume that the hydrogeology of the Lab report notes: aquifer systems underneath SRP will remain essentially unchanged Nitrates have been detected in indefinitely. growtd~vaterat seep lines at In addition to radionuclide Four Miles Creek in contamination, there is also the problem Co~tcentratio~tsrarrgirtg front 7 of groundwater contamination by to 30 rimes the &PA drinking non-radioactive toxic materials such as water sta~tdard. The average mercury and PCB's, some of which are co~tcerttratio~tof nitrates in also contaminated with radioactive Forrr Mile Creek at Road A-7 was materials. On October 29, 1982, G.C. 3.1 Mg/l corresporrdirig to a Halstad, Director of the Process and transport of 460 torrs per Weapons Division at SRP, indicated to G. year.(46) Smithwick, Director of Environment, that SRP could not meet EPA's Interim Final Thus about 460 tons of nitrates are Hazardous Waste Regulations because: flushed by groundwater into surface streams annually. Lesser amounts of The seepage basin exceeds the toxic mercury are present in groundwater allowable hazardous beneath the SRP 200 Area. In 1974, constitutent concentrations for DuPont researchers obtained soil samples many nun-radioactive nuclides. from an H-area seepage basin which had temporarily dried up. They estimated Liquid hazardous waste are that half of the mercury (assuming stored below grade in the uniform distribution) was bound to the burial ground. Over ten tons oj upper layers of the soil and the other mercury are intimately mixed half was leaching into the groundwater. with the environme~rt in the If the remaining half continued to leach Burial Ground. uniformly and if groundwater flow rates remained constant "in less than 200 An inventory oj the Hazardous years, concentrations of mercury would nun-radioactive nuclides in the exceed the EPA drinking water Burial Ground is not possible; standardT(47) In fact, a groundwater there, are no records excepr in sample taken in 1982 in the the case oj mercury and PCB's vicinity of the Burial Ground was (in recent years).(45) measured to have a level of mercury of 60% to 80% of EPA's drinking water Non-radioactive contaminants are standard. This mercury, as of 1977, "is present in significant amounts in detectable in soil from a swampy outcrop groundwater beneath SRP's nuclear waste along Four Mile Creek and button1 operations. Nitrates which constitute sediments and suspended solids from the the largest single non-radioactive creek show that some mercury from the substance discharged into the F & H H-area basins is migrating..."(48) basins are found in groundwater and at surface stream outcrops far in excess of DuPont has, in effect, acknowledged that mercury may pose one of the more serious long-term dangers. A Savannah radioactive waste, in River Lab has noted that: con~bbatiort wirh chentical waste. into trenches and in the hazardous surface ground does not represent cortcentratio~ts/of ntercury] may state-of-the-art technology and exist well beyortd the period in ntay violate RCRA [the which land cotttrol can be Resource Conservation and anticipated and indeed ntay Recovery Act] requirements. To exist cennrries or ntellettia increase the volunte of waste into the future whetr soil which must be handled by this erosion may expose ntercury on facility before the the surjace.(49) decommissioning plan has been developed, is oiii of logical Recent data analyzed by the EPA phasing. Practically speaking, also underscores the groundwater SRP needs to develop a proper contamination non-radioactive disposal facility to handle contaminants at SRP. present volunres of waste materials before any additional F & H area studies have shown wastes are generated.(53)* that chemicals, e.g. mercury, I- 1-1 trichlorethyelene and The AEC itself acknowledged some of chromium from the seepage these problems as long ago as 1971. basins have entered the shallow Although it prefaced its recommendation soil to outcrop zones ...this with the claim that waste disposal in gpbears to anstrate a ntethod soil was a 'well-established safe P f di-n~ ~olfUlonrsto a procedure," it issued a policy on the witholti a ~ernrit bt handling of low-level and intermediate wina iwor'ndwarer as a meditrm liquid wastes which called for an end of gf tr~."(emphasis "the use of natural-soil columns (by added)(50) means of cribs, seepage ponds and further facilities..."( 54) The EPA has concluded that The AEC further noted that: bevere contamination in the upper aquifer poses an imminent threat to a there is an explicit assumption deeper aquifer that supplics drinking that the favorable water to plant employees and off-site environmental conditions will communities."(5 1) Despite DuPont's claim exist until the radioactivity that continued use is desirablen (52). by 1983, the EPA had also reached the following conclusions about the Burial Grounds:

The present practice of disposing of low-level

.We understand that DOE and EPA have recently reached an agreement that DOE jacilities will conform to the Resource Conservation and Recovery Act. How and when this will be done remains to be seen. irr the soil decays to ir~t~ocito~cs with plans to significantly increase levels. Becaitse of the waste volumes over the next decade. The lo~tg-termburdert of corttrol artd current schedule for the closure of some stireeillatrce inheret~t itr the of SRP's seepage basins has been set for trse of the techtiiqrce that 1988.(60) No decommissioning plans exist resttlts in large locat for the Burial Ground. accunltlatiotis oj radioacti~~ity in soil, AEC sites are contamination Due to Accidents elinlittating the routine use of surface and near surface The facts of groundwater techniques that depend on soil contamination and the unexpectedly rapid to remove radioactivity from migration of some pollutants are not liquid wastes...( 55) only important in themselves. They are of central significance to the The AEC set a deadline to discontinue estimation of risk posed by the the use of seepage basins and other accidents and other catastrophes -- both similar disposal methods by 1976.(56) those that have already occurred and By September 1973, however, the AEC those that might occur in the future. reversed itself and promulgated Accidents in the course of operation radioactive waste management have ranged from relatively small spills standards(57) which in effect ignored during maintenance work to large ones their official conclusions of the year due to pipe and tank leaks. Tables 2 before and supported continued use of and 3 in Part I1 show the accidental soil as a disposal medium for releases that have been recorded in the radioactive wastes. The standards also pata Ba. For example, one suggested that there be an "acceptable accidental release was recorded on levela of radioactivity in soils. The 02-05-69 in the H area as follows: level was not specified. A year later Savannah River Laboratory proposed using Release to environs, about I their current levels of radioactivity in Ci, mostly Cs-137. to ground 30 seepage basin soils as the feet sourhwest of Tank 9. &st. 'acceptablew limit.(SB) In the 0.50 is contaminated liquid declassified 1977 report on SRP's flowed into storm drain and seepage basins, it was noted that 'most there to Four Mile Creek. dischargeswwould be out of compliance Cause: failed copper pipe used with the standards if measurements of back flush gravity pollutants were made at the point of drain... About 6 cubic yards of discharge.(59) It was perhaps with this earth and asphalt surjacing, in mind that the 1973 standards were containing an estimated 0.5 modified to incorporate a "buffer zone" curies was excavated and allowing measurements to be made at the transported to the Burial site boundary instead. Ground. Total personnel By 1976, the year proposed by AEC to exposure was 5.8R. discontinue use of natural-soil columns, SRP had IS seepage basins in active use A number of entries indicated serious o A leak resulted in the accidental environmental contamination, contamination of soil around Tank but no details are given in the pata 3. (F area entries dated Bank. nor have we found reference to 08-21-75 to 08-25-75. Table 1, such events in other publications which Part 11.) we have examined. The entry dated "02/29/6In [sic], for instance, records A summary of these incidents based on that plutonium leaked into the soil and official descriptions is given in Table onto the asphalt around the evaporator 6.6. In addition, a number of serious four times in one month. No estimates spills have occurred due to pipe of quantities are given, The soil was failures: removed and, presumably, taken to the o A leak in a flange in August 1962 Burial Ground. Worker exposures are not contaminated 200 square feet of cited. In another example the F area ground. (F area entry dated entry date 04/00/63 (see Table 2, Part 08-13-62. Table 2, Part 11.) 11) records the contamination. of 10,000 square feet of area downwind of the o A flexible hose ruptured in diversion box due to jumper changes. No February 1967 causing a spill of estimates of quantity of contamination 40 to 50 gallons of slurried are given. Nor is there an explanation radioactive sludge containing as to why these jumper changes, which about 200 curies of radioac~ivity are fairly frequent occurrences, (H area entry dated 02-01-67. resulted in such widespread Table 2, Part XI.) contamination. There have been a number of large, o Chloride cracking of concentrate accidental releases of radioactivity. transfer system pump tank piping The following four have been at or near in June 1975 caused a spill the waste tanks: containing roughly 100 curies of cesium-137;(64) o Leaks from Tank 16 into surrounding soil and water during o A pipe failure due to corrosion the 1960s (several H area resulted in the release of 500 entries, 1959 and 1960. Table 3, gallons of radioactive salt Part II.)(61) solution into soil (H area entry dated 08-01-77. Table 2, Part o Overflow of Tank 8 due to 11.) instrument error resulting in soil and water contamination (F Apart from the last incident, these area entry dated 04-00-61. Table spills have been discussed in various 3, Part II.)(62) official documents which are public. We will therefore not discuss them in o A spill of "100 to 200" gallons detail here. We will only comment on of high level radioactive waste the adequacy or lack thereof of some of due to a plugged riser on Tank 9 the official analyses. This enables us (H area entry dated 05-01-67. to assess the implications of the Table 2, Part IL)(63) official position and of possible future The estimated 700 gallorts of spills in terms of their cumulative waste that exceeded the level impact on long term contamination in the 01 the steel anntilris pan area of and around the Savannah River represents an upper lintit of Plant.* the amount that could have The first major leak to occur was in escaped from the tank into the Tank 16 in September 1960**. Since it was ground. This amount oj waste preceded by a number of cracks in the would contairt 5,200 curies oj primary tank, which were first observed radioisotopes. The actual in November 1959, it was amount oj waste that reached perhaps the accidental release which was the soil is not known, but lor the most closely monitored of all of the waste to leak into the soil them. Yet a considerable uncertainty, of requires passage through a an order of magnitude or more, remains tortuous route and probably as to the quantity of radioactivity that only a small fraction of the was released to the soil and water. 700 gallons lejt the Since the bottom of Tank 16 is below the artnu!us.(66) groundwater table it appears that some of the leaked radionuclides immediately Even the figure of 700 gallons is entered the groundwater. The leak was subject to some uncertainty, since the first discovered by the detection of height of the level of waste in the tank high-level waste in the water from a is the primary indicator of the monitoring well near the tank. (See quantity. The actual height at the Table 3, Part 11, H area entry dated maximum point used to derive the figure 10-00-60.) of 700 gallons is an inferred value, High-level waste began accumulating obtained by linear extrapolation of in the 5-foot high steel pan which lines level measurements. Moreover, these the concrete secondary containment. The level measurements are themselves tank height is 27 feet. In September subject to some uncertainty because of 1960 the level of waste rose above that chronic problems with the main level of the liner and some waste leaked from measuring instrument (the 'reel tape"). the concrete containment, possibly from Yet the official report does not cite a various joints. The total quantity of range of possible values of height (and the leak from the primary tank into the hence, quantity) but gives only one annulus was 185,000 gallons. figure. There is at least one earlier account which indicates a loss of more than 700 gallons. J. Christl, in a 1969 presentation to the National Academy of Sciences, stated that "inventory measurements indicated rhat as ntuch as 1,000 gallons may have escaped the encasernent."(67) By 1980, the DuPont estimate given to a new National Academy

*Technical issues relating to tank integrity have been covered in Chapter 4. **This account is based on an evaluation of the Tank 16 leak by W.L. Poe of DuPon t.(6S) of Sciences had decreased further to section of the radionuclides "about I00 liters" -- or 26 gaIlons.(68) would be delyed and sjould This leaves almost one thousands gallons require 2x10 to 12x10 of waste unaccounted for. years to reach the populated The possibility of as much as 5000 area. In any o/ these periods curies of radionuclides, mainly the radiocesiitnl wort Id hase radiocesium, getting into the decayed, leaving harnlless groundwater should be an event of very solrttio~rs of stable bariunt. serious concern whose long term The above caIcuIatiorrs assume consequences should be carefully that climatic corrditions, evaluated. Yet both the content of the hydrotogy, and grorotd water official evaluation of this incident, as usage remain unchanged during of the others we have read, convey the the periods covered by the impression that the management is calcutation. In times as long certain that this matter will have no as hundreds of rhousands of adverse effect on the public whatsoever. years, some oj these conditions We have already discussed the might change. I j these assumptions regarding radionuclide conditions were assumed to transport and groundwater contamination remain unchanged for a hundred on which such conclusions are based, and years or so and then the presented evidence to show that they are readioactivity from this leak essentially erroneous. The following was released to the creek by quote shows that such models, which have some mechanism or i/ the proved to be unreliable over the last radioactivity moved at the same three decades. are used to extrapolate rate as groundwater (travel over hundreds of years and even hundreds rime 70-350 years), the of thousands of years: concentration of radionuclides in the creek jrom this release Calculations based on the would be below the measured groundwater flow concentration guide limit for velocities and estimated flow cesium in drinking water.(69) paths and the measured ion-exchange capabilities of The second set of calculations the soil determined in involving hundreds of years is based laboratory samples indicate entirely on arbitrary assumptions. The that the leading edge of the first set involving parameters radionuclides should no! reach determined from measurement involve the nearest creek and thus projections of up to 1.2 million years ultimateIy$he popu(ped area and are on the face of it absurd. A for 1x10 to 8x10 years. mere glance at the history of the last [At this point a footnote adds: one hundred years would alert anyone to "The range of times denotes the the possibility that patterns of "ground range of uncertainty in travel water usage" might be subject to drastic path to tire creek.-] Movement change. In all cases attention , is of the tore' or central focused on travel time to the creek. The focus on the pathway of surface "first indication of subsurface water enables the further assumption of contamination" was discovered as a immense dilution by the creek and result of unrelated drilling of Savannah River waters. Thus, the monitoring wells in the area. By that possible doses appear small. The time it was impossible even to be sure possibility of the contamination that the radioactivity detected was a entering the Tuscaloosa aquifer was result of the 1961 overfill.(72) consistently dismissed. Thus, the The official report estimates that implications of the known migration of the contamination was probably the radionuclides, particularly in the result of the April 1961 overfill and presence of solvents, and of groundwater that the amount spilled was about 1540 use changes that are .already occurring gallons of high heat waste. This was have not been examined. estimated to contain anywhere from DuPont has also issued a special 25,600 curies to 260,200 curies, report on the spill from Tank 8 around depending on assumptions about the April 1961.(70) This spill apparently composition of the waste. While much of occurred when a faulty level reading this consisted of relatively short (from' a "reel tape") caused a tank half-life cerium isotopes such as overfill. The leak was not detected at -. cerium-144, approximately 7000 to 20,000 the time. The _Bank entry dated curies consisted of longer lived April 1961 does not mention a spill at isotopes like cesium-137 and all but explicitly states that the waste strontium-90 and associated "showed up in the catch tank via [the] radionuclides that remain in the encasement drain system." (See Table 3, soi1.(73) Part 11, F area entry dated 04-00-61.) From soil sameles, the contamination The report of the incident states was assessed to be localized. Radiation that about 15,500 gallons of high heat levels up to 300 rads per hour were waste overflowed into the catch tank as measured within 17 feet of the soil a result of the overfill. This was surface (the top of the tank is 13 feet detected on April 3, 1961, and below the soil surface).(74) This spot corrective action taken by April 5. is only around 30 feet above the maximum However, the "leakage from the fill-line height of the water table around Tank 8. encasement was not recognized at this As with the Tank 16 leak, the time."(7l) Apparently no steps to check long-term calculations are exclusively for possible soil contamination were oriented toward surface water taken at the time, even though the high outcrops. Even these concerns are beat waste followed to the catch tank effectively dismissed since travel time outside the primary steel pipe and for strontium90 and cesium-137 to the through the concrete encasement (or groundwater outcrops are claimed to be jacket) covering it, This was in spite "on the order of several hundred of the prior experience of leakage from thousand years" by which time they would the concrete secondary containment of have been rendered harmless by Tank 16 (see above). radioactive decay.(75) The investigations for a possible Only plutonium-238 and -239 are leak were begun only in 1973-74 when the recognized as possible long-term sources

of radionuclides and other toxic 11. Op. Cit. Reference 4; p. 7. materials. 12. Op. Cit. Reference 2; p. 13. Po sclentlf. .. rc assessment of the 13. Op. Cit. Reference 4; p. 47 of accidents or 14. Ibid; p. 48. hes such as earthauakes can be 15. Lawless, W., Fifth Complaint to mw DOE, 27 Aug 83. tvidence from routine onerations into 16. DPSU 54-1 1-12; Dp-27; Dp-92; DPSP WunL Far from attempting to do this 55-25-34; DPSP 56-25-13; DP-57-25-4; the DOE and DuPont risk assessments DPSP 57-25-1 5; DPSP 57-25-43; DPSP discussed in Chapter 5 as well as the 58-25-17; DPSP 58-25-38; DPSP 59-1 1-23; assessment of the accidents discussed DPSP 59- 11 -30;DPSP 60- 11-9; DPSP above largely ignore this evidence. 60-25-9; DPSP 61-25-4; DPSP 62-25-2; Thus, from a scientific point of view, DPSP 62-25-9; DPSP 63-25-3; DPSP these assessments are fundamentally 62-23-1 0. f lawed. 17. Ibid. (i.e. DPSP series). 18. Ibid. 19. Op. Cit. Reference 2; p. 13. References 20, Lawless, W.F., "The DuPont Management of Savannah River Plant 1. Energy Research and Development Radioactive Wastes: A Report to the U.S. Administration. Yaste Mana~ernent House of Representatives," Report to the Sav Aiu Energy and Commerce Subcommittee on th Car-. Final Environmental Oversight and Investigations, U.S. House Impact Statement ERDA-1537, National of Representatives, Nov. 27, 1973. Technical Information Service. 21. Op. Cit. Reference 1; p. 111-17. Springfield, VA, September 1977; p. D-1. 22. Ibid. 2. Marter, W.L., "New Criteria for 23. SRL reports citing Pu coat.; p. Seepage Basin Use," DPST-77-444, 118. September 1977; p. 5. 24. Lawless, W.F., Appraisal report, 3. Lawless, W.F., Savannah River Reference #3 Plant Burial Ground Management appraisal 25. Ibid. Report, June 2-13, 1980, unpublished. 26. Op. Cit. Reference 20. US. DOE-SR Draft Report (final draft, 27. Op. Cit. Reference 4; p. 13. Nov. 1, 1982). 28. Op. Cit. Reference 1; p. 111-20. 4. Fenimore, J.W. and Horton, J.H., 29. Op. Cit. Reference 24; p. 10. "Operating History and Environmental 30. 'Migration of Pu at Maxey Flats," Effects of Seepage Basins in Chemical w.26 Jan 81; p. 1506. Separations Areas of the Savannah River 31. USGS, quoted by Ronnie Plant," DPST-72-548, 1974. Lipshutz, -active Waste: Pw. . 5. Op. Cit. Reference 2; p. 28. Tcchn61onv,Ballinger, 6. Ibid. Cambridge, Mass., 1980; p. 131. 7. Op. Cit. Reference 2; p. 13. 32. Op. Cit. Reference 2; p. 39. 8. Op. Cit. Reference 4; p. 5. 33. Op. Cit. Reference 4, p. 9. 9. Op. Cit, Reference 1, p. 111-13. 34. US. Department of Energy, Draft 10. Ibid. p. 111-17. ETS: OD- Savapp~hRiver Plant. DOE-EIS - 01080. Appendix F. November 1976. 35. Ibid. 63. PuPont Safttv Analvsis Re~ort- 36. Ibid. ~d Radioactive... Waste Hand1 inn and 37. Sternberg, Y., comments on the aae Facrm 200 Area, Savannah Draft EIS. L -Reactor O~eration. SRP, River Plant, DPSTSA 200-3, Aiken, S.C., October 1983. 1977, pp. 3.210-21. 38. Ibid. 64. Ibid. 39. U.S. EPA Region IV, "Comments on 65. Op. Cit. Reference 61. L-Reactor Draft EIS," DOE-EIS-01080, 66. Ibid., p. 24. Oct. 1983. 67. Wement of Ramctive Waste 40. Op. Cit. Reference 33. pt the Savannah River PU. 41. Ibid. Presentations made to the Committee on 42. Pixie Newman, "Comments on Draft Radioactive Waste Management of the NAS EIS-01080," Oct. 1983. at SRP, Jan. 20 & 21, 1969; TID 26524. 43. Ibid. p. 48. 44. U.S. DOE Draft EIS-01080. 68. mioactive Waste h&uuwment at 45. G. C. Halstad memo to G. aeSav- River Plant - A Tecwcal smithwick, 29 Oct 1982. Beview. Panet on Savannah River Wastes, 46. Op. Cit. Reference 2, p. 14. National Academy Press, Washington D.C., 47. Ibid., p. 26. 1981, p. 12. 48. Ibid., p. 41. 69. Op. Cit. Reference 61. 49. Ibid., p. 26. 70. Op. Cit. Reference 62. 50. Op. Cit. Reference 39. 71. Ibid., Appendix A, 51. Ibid. 72. Ibid,, p. 70. 52. Op. Cit, Reference 2, p. 14. 73. Ibid., Table 3, p. 15. 53. Op. Cit. Reference 39. 74. I bid., p. 11. 54. "Plan for the Management of 75. Ibid., p. 16. AEC-Generated Radioactive Wastes," USAEC 76. Ibid., p. 17. Report WASH-1202, January 1972. 77. Op. Cit. Reference 1. 55. Ibid. 78. Op. Cit. Reference 63. 56. Ibid. 79. Testimony of William F. Lawless 57. *Radioactive Waste Management," in the Aiken County Magistrate Court, USAEC Manual, Chapter 0511, Sept. 13, January 9, 1984; unofficial transcript 1973. provided by the Natural Resources 58. Jacobsen, W.R., "Criteria for Defense Council. seepage basin USAEC," DPST-74-447, August 13, 1974. 59. Op. Cit. Reference 2, p. 5. 60. U.S. Dept. of Energy, FY 1984 Appropriations. 61. Poe, W.L. et a]., 'Leakage from Waste Tank 16: Amount, Fate and Impact," DP-1358, Nov. 1974. 62. Odum, J.V, "Soil Contamination Adjacent to Waste Tank 8," DPSPU-11-4, Table 6.1

ANNUAL RELZASE LIMITS FOR SRP OPERATION

Element Release limit Element Release limit curies/yr curies/yr tritium 800,000 iodine-131 300

zirconium-9 5 natural uranium + niobium neptunium-237

Source: Management of Radioactive Wastes at the Savannah River Plant. presentations made to the Committee on Radioactive Waste Management of the National Academy of Sciences at the Savannah River Plant, Jan 20 C 21, 1969; p.11. Table 6.2

Release Guides for Seperations Areas, 1978 (Curies/Year)

Element Release Guide Release Guide Total to Seepage basins to atmosphere tritium krypton-85 strontium-89 strontium-90 iodine-131 cesium-134 TABLE 6.3 IUVBNTORY OF RADIOACTIVIFX TO SEEPAGE BASINS TEROUGH 1976

tritium strontium-90 cesium-137 natural uranium thor ium-232 plutonium-23 8 plutonium-23 9 cur ium-2 44 Alpha - unidentified(2) Beta - unidentified (2) Short-lived (3)

Notes: 1, Decay corrected. Does not include 100-R basins, 2, Not corrected for decay. 3. Half-lives more than 1 year but less than 6 years. Source: Marter Wet,, 'New Criteria for seepage basin usew DPST-77-444,Sep 1977, pegw TABLE 6.4 BO1Q-RADIOACTIVE RELEASES OF MBRCURX & NITRATES TO SEPARATIONS AREA BASINS - CUHULATIVE TO 1975

Basins Mercury Nitrates pounds tons CMX-TMX 64 -

TOTAL 4,564 5,620 Source: From: Marter, W,L, , "New Criteria for seepage basin use*. DPST-77-444. 12 Table 6.5

VERY LONG-LIVED FISSION PRODUCTS IN SEPERATIONS SEEPAGE BASINS (ESTIMATED)

Radionuclide Half-Life Fission Yield % Total in F Years and H Ci

Total 1.5 x loo2

Source: Marter, W.L., @@New Criteria for seepage basin use.@@ DPST-77-444. 11, Table 6.6

ACCIDENTS AT OR NEAR HIGH-LEVEL WASTE TANKS INVOLVING RADIONUCLIDE SPILLS

Location Entry Date Volume Radioactivity Official Cause gallons curies

Tank 16 Nov 1959 Max 700 Max 5200 primary leak to annulus Tank 8 April 1961 1500* 7000 to 20,000 reel tape error, overflow of tank; discovered in 1973-74

Tank 9 May 1967 100-200 1500-2000 2' riser plugged (90% Cs-137) with salt causing tank overflow Tank 3 Aug 1975 3 SO*** suckback and (mainly Cs-137) leak-in steam supply H-Area Aug 1977 400 500 pipe corrosion (mainly Cs-137)

* High heat waste ** Estimate of what was in the soil around tank as of 1976 *** Soil radiation level of 5OOR/hrt soil removal had to be stopped because of high levels. TABLE 6.7

ACCIDElQTS AT OR lDEAR HIGH-LEVEL WASTE TARAS IRPOLVIIOG RADIONUCLIDE SPILLS

: Location Entry Date Volume Radioactivity Official Cause gallons curies

' Tank 16 Nov, 1959 Max, 700 Max, 5200 primary leak to annulus

' Tank 8 April 1961 1500* 7000 to 20,000 reel tape error, overflow of tank discovered in 1973-74) Tank 9 May 1967 100-200 1500-2000 2' riser plugged with salt (90% Cs-137) causing tank

' Tank 3 Aug. 1975 3 50 ++* auckback and leak in (mainly Cs-137) steam supply H-Area Aug, 1977 400 500 pipe corrosion (mainly Cs-137)

*High heat waste **Estimate of what was in the soil atound tank as of 1976 *++Soil radiation level of SOOR/hr; soil removal had to be stopped because of high levels,

FIGURE 6.2

%~me~- DPSV 5)-11-12; OP-27, DP-32; DPSP 55-25-34; DPSP 56-25-13; DPSP 56-25-54; OPSP 57-25-4; DPSP 57-25-15;DPSP 57.25-43; OPSP 50-25-1 7; DPSP 56-25-38; DPSP 53-1 1-23; DPSW 59-11-30; DPSW 69-11-9; DPSPU 60-253: OPSP 61-25-4; OPSP 62-25-2; DPSP 62-25-9; DPSP 63-25-3; DPSP 63-25-10.

g27501( 2mx

-1750X -15ooX -1250X -1mx /-7 -?SOX -500X 2H)X - 11 11 11 1 111111 L~kgro~nd(ll.3pc/l)( I I I I I I I I hPlmtoWr8tlw 1952 1% 199 1935 1956 1957 1958 1959 1960 l%l 1%2 1%3 Harsurraants of F6M pm&ttr 1951-53 -10 CO)sTwlmf WOW-VOLATILE BETA MlnERS IN THE 200 AREA F6W SEEPAGE BASIN WELLS BY ORDERS OF MG?ilfUDf ABOVE PREBPERATIONS BACKGROUND READINGS FIGURE 6.3 CHAPTER 7 WORKER EXPOSURES

There are .over 8,000 workers at the with "non-routinew matters such as Savannah River Plant. All workers, repair, or checking, or replacement of including the ones involved in waste malfunctioning equipment. In addition, management, are subject to some routine these workers are also involved in the exposure to radioactivity as a result of work resulting from accidents, for plant operations, of routine example, excavation of contaminated radionuclide discharges to the soil. During the course of such work, environment, and of any increase in they are exposed to the dangers of background radioactivity levels accidents which can result in external resulting from a build-up of exposures and internal exposures, as contamination on the site. For through inhalation of radionuclides. instance, "background" radiation levels Table 1 in Part I1 lists most of the in SRP's 200 area between 1954 and 1969 entries relating to worker exposures gave an average cumulative dose of 3,896 found in the 200 Area Fault Tree Data mrems, or about 260 mrems a year. This Bank. The rest are to be found in Table is about 2-1/2 times the normal 2, Part 11. (See Introduction to Part background levels for an off-site 11.) area.(l) In addition, they are also We have mentioned the general problem subject to the danger of exposures of of the uneven number of entries in the two other kinds. Bank. This applies in particular to worker exposures. There o Exposures which are an are about 300 entries where worker inevitable, and sometimes a exposures are specifically mentioned. routine consequence, of the Three-fourths of these entries are for specific nature of the work; the six-year period 1977-82, and only about 70 are for the period 1953-76. o Exposures as a result of Assuming that the recording frequency accidents. during 1977-82 expresses a management decision rather than increasing numbers It is not uncommon for large numbers of worker exposures (see Chapter 3). we of construction workers at SRP to be can estimate that there are roughly 1000 recruited for manual clean-up of highly occurrences involving exposures of the contaminated areas caused by type entered into the Data Bank accidents.* For the workers involved during 1977-82 which have been omitted in the operation of the Tank Farm, it is from it for the earlier period as the difficult to draw a distinction between result of varying management decisions "routine," "non-routinen and about what 'was to be entered regarding "accidentaln exposures since much of worker exposures. their "routine" work consists of dealing A large fraction of the 14,000

*Between Nov. 1970 & Feb. 1971, 850 people were involved in a manual clean-up of the K-reactor process room where a reactor source rod burst open. entries into the Data Bank. either pumps but which do not mention any describe people working in radioactive worker exposures. They are quoted below environments and with radioactive system in their entirety (Table 8, Part 11): components, or imply such work by describing equipment problems. Yet 1. 02-27-67, F area: worker exposures are mentioned only in "242-F Evaporator Feed pump about 300 instances -- or about 2% of failure" the total entries. For example, problems with the reel tape, an 2. 11-30-67, F area: instrument which measures the liquid "C.T.S. pump failed" level of radioactive waste in a tank, have been frequent. Reel tapes have 3. 01-00-68, F area: required frequent repair and adjustment. "Pump failure, after 40 days of They are quite radioactive since the service in tank loop of CTS. tapes and probes come into direct This is the 4th failure in 9 contact with the liquid waste and with months." the radioactive vapors above the liquid. Yet an entry such as "Repaired Tk.29 4. 06-00-68, H area: reel iape. 8-4 [shiftJm(2)gives no "Pump Jailure. CTS tank loop indication of any worker exposures pump failed and was replaced E M whatsoever. unit - bearing an EM unit packed It is possible that worker exposures with special grease." in some such instances might be small or even less than measurable. It is not Omissions of exposure data are only likely, however, that most of the work a part of the problem with the involving intensely radioactive Bank. Many of the approximately 300 environments or components would not Pats Bank entries which do mention result in measurable worker exposures. worker exposures are unsatisfactory in a For instance, problems with number of respects. These qualitative evaporator feed pumps and concentrate deficiencies can be broadly categorized transfer system (C.T.S.) pumps have been as those relating to external and to fairly frequent. The H area entry of internal exposure data. March 1975 records radiation exposure As regards external exposures, most rates of up to 1'0 R/hr during an of the entries do not give the actual evaporator pump replacement, and a measured exposure or even an estimated total exposure of 0.8 rem to 9 people. exposure. They simply cite the fact of Similarly an F area entry dated 09-26-74 contamination or give expousre rates or records an exposure of 1.4 rem to 14 radiation rates. Thus, data regarding people during feed pump replacement and the number of workers involved in the notes that the entire "evaporator feed work or the duration of the work, which pump job" during 1-1-74 to 09-26-74 would enable an estimation of the total caused a total exposure of 11.4 rem. dose, are often left out. Here are some (See Table 1, Part 11.) Yet there are examples: several entries which imply or mention changes of or repairs to radioactive 1. 04-00-67, H area: 18 to install foundations for salt removal equipment. Body "Released waste. During repairs dose rates to 100 mr/hr. Total to an evaporator can valve a leak exposure 515 mr. for 28 people." caused extensive contamination." (from Table 2, Part 11) 4. 01-00-81, H area: "Exposure rates to 30 2. "People that were contaminated rads/hr/2r/hr were encountered to remove B.G. [Burial Ground] went home at a wire that had become wrapped around a 7:00 p.m. 4-12 [shift]." sludge probe in Tank 15. Total exposure to remove the wire was 435 mrems." 3. 05-29-81, H area: "Installed draw-off valve in C-2 5. 04-27-81, H area: riser on Tk-3.6. Radiation level "Transfer jets screen plugged over open hole was 100 exposure rates during removal to rads/40R/hr. at 30 cms." 10R/hr. Total exposure to 27 workers was 3.2 R." Thi largest radiation rate in the Data Bank was 500 R/hr encountered by The largest single exposure of more workers excavating contaminated soil than 14 rems is mentioned in an H area around Tank 3. (See Table 1, H area entry of November 1968: entry dated 03-12-76.) For the entries which do record "Film badge of sep[arations) worker exposures to external radiation, dept. supv. indicated skin the amounts range from doses of a few exposure of 14,590 mrads during tens of millirems to ten or more rems. Oct. exceeding AEC manual These art collective doses to all quarterly standard of 10 rem. workers participating in a particular The exposure pro[bably) job. Some examples from Table 1, Part occurred when the employee I, follow: removed two acme nuts from a 1. 01-00-75, F area: 241.H waste transfer pump in '14 R received by personnel railroad tunnel airlock. See installing new feed systems in special hazards investigation 242-F." 266.0(Table I, Part If). 2. 07-00-71, F area: Most of the entries in the Total of 3.4R were obtained by Bank involving worker exposures record personnel while tightening such cases of exposure to external, packing glands on 3 valves in the primarily gamma, radiation. There are feedpump. Exposure resulted from also many entries which record high radiation levels in feedpump contamination of clothing and skin enclosure." through accidental contact with radioactive materials. For example, one 3. 06-00-82, F area: worker's left hand was contaminated to "Excavation around tanks 17 and 100 mrad/hr during the removal of a specific gravity dip tube from Tank 13. extent of the contamination, even within A second worker's wrist was contaminated the nose. Only large particles are to 40,000 c/m beta-gamma in the same deposited in the nose, while small ones accident (see Table 1, Part 11, H area pass through to the Iungs.(S) The nasal entry dated 03-28-80.) smears only demonstrate that the person Finally, there is the question of may have inhaled plutonium. The internal exposures -- that is, exposures measurements cannot be used to which result from the incorporation of reconstruct the amount of plutonium that radionuclides into the body by reached the lungs without some inhalation, ingestion or through cuts. information about the particle size The para Bank records very few distribution. cases of internal exposures. None of We have performed an illustrative these contain estimates of lifetime calculation for one case in which body exposures resulting from radionuclide counter data are provided in the Data incorporation. In fact, most of the data Ban. (Body counters can be a more is in a form which does not allow such reliable means of determinig internal dose reconstruction. For example, much radionuclide burden, if the time and of the monitoring for inhalation is from circumstances of contamination are nasal smears and urine samples. However, accurately known, and if contamination this data is not reliable enough to is sufficiently large to be reliably estimate lifetime dose commitment, which detected by the available counters.) The is a crucial indicator of the H area entry dated 08-23-72 (see Table probability of radiation-induced 1, Part XI) gives a chest activity diseases and problems. measurement of 33 nanocuries. We assume According to research performed at that it is a plutonium count, since it DOE'S Hanford operations, internal follows a measurement of plutonium in plutonium contamination estimates the urine. The accident report does not derived from urine samples in the best state when the chest measurement was of circumstances were questionable.(3) performed. Therefore, it is difficult to An evaluation of Hanford workers with determine the amount of plutonium that internal plutonium exposures showed that remains in the lung. We will assume that curves describing urinary excretion data the measurement was performed soon after for plutonium vary by individuals,(4) the accidental exposure. Therefore, any Therefore, any dose reconstruction drawn plutonium deposited into the pulmonary from excretion data may be unreliable, region of the lungs was still there.(6) particularly if only one measurement was Since the plutonium is contained in performed (as appears to be the case). waste from reprocessing, the most likely In addition, the Qata Budoes not chemical form is plutonium nitrate. state how long after the exposure the The B- does not give the urine sample was taken, which will particle sizes that were inhaled. heighten the unreliability of the dose Consequently, the plutonium particles reconstruction. are assumed to have an activity median Estimating inhalation doses from aerodynamic diameter (AMAD) of one nasal contamination is also unreliable. micron. This is a default value A nasal smear does not measure the recommended by the International Commission on Radiation Protection if no level wastes are opened, we can expect particle size distribution is given. some contamination of the surrounding Since plutonium nitrate is more air. There are many entries in the Data soluable than some other forms, various Bank which show the presence of authorities(7)(8) have estimated that plutonium as well as fission products in the amount of plutonium inhaled would be the air around the working areas when two to four times the amount of tank risers are opened. For reasons we plutonium measured. For this case it do not know, such Data Bank entries were would mean an estimate of 66 to 132 made mainly in the H area and primarily nanocuries inhaled by the worker. during 1976 and 1977. There is no Using dose conversion factors apparent reason why such contamination prepared at Oak Ridge National would not have occurred in the years Laboratory, Table 7.1 presents selected before or since, or why the entries were organ doses (actually organ committed concentrated in the H area. It is dose equivalents). The marrow dose possible that H-area management may have below is sufficient (according to a made a decision in 1976 to record such survey by Mancuso, Stewart and Kneale of measurements, but that this was Hanfard workers) to more than double the rescinded within a year or two. risk of this worker contracting a The presence of radionuclides in the blood-related cancer.(9) air in the working areas during certain Another accident report (H area, date kinds of work increases the possibility 02-28-74) records inhalation of of inhalation of radioactive particles. cesium-I37 and ruthenium-I06 by two We understand that protective equipment maintenance workers during an accidental is normally worn during such jobs. spraying of one pint of contaminated However, without detailed information on liquid. The chest counts on one worker the characteristics of the equipment, or were: Cs-137: 262 nanocuries, and its maintenance and condition, and on Ru-106: 43 nanocuries, For the second the size distribution of the suspended worker the count was Cs-137: 64 radioactive particles, we are not in a nanocuries. position to estimate the internal dose There are only a handful of entries that might result from working in such in the Qata Book with body counter conditions. measurements which enable internal dose The evidence in the Buis reconstruction. We do not know if so fragmentary and qualitatively uneven documents not yet made public contain that, taken by itself, it does not allow more extensive data. Of course, without us to reach general conclusions about more data we cannot know if the above worker exposures. It does document many calculations are representative of the specific cases of worker exposures, both dose commitments to workers from internal and external, and the radionuclide inhalations. But they do conditions under which those exposures illustrate the dangers workers face in occur. Moreover, we are in a position the course of their duties from to complement these specifics with some suspended radioactive particles. broad evidence regarding radiation In general, when equipment such as exposures and hazards of contracting tanks or evaporators containing high radiation-related diseases at SRP and other DOE facilities. An official estimate of whole-body 330 excess cancer cases, Over half of radiation accumulated between 1954 and these cancer cases are expected to occur 1978 by monitored SRP workers ia approx- among workers who handle radioactive imately 50,000 person-rems.( 10) About wastes.(l2) one-third of these workers performed This is only a partial picture of the duties in the 200 Area of SRP involving risk since it doesn't factor in processing and storage of radioactive exposures to internal organs from and other toxic wastes. These 200 Area inhalation and ingestion. The above workers received over half the total mentioned risk only pertains to external accumulated dose.(l 1) penetrating radiation. Savannah River Plant workers handling The paucity of data regarding nuclear wastes bear higher risks of internal exposures is not unique to SRP. contracting radiation-induced cancer due Indeed, it may be said to be typical. to their higher exposures. Yet these For instance, there is very little risks have not yet been ascertained by public data on internal exposures independent epidemiological studies. In relating to the nuclear weapons testing the absence of such studies, we can program. This was recently documented as estimate expected health effects among regards to the Bikini tests in 1946 by SRP -workers on the basis of their the revelations in the papers of the accumulated collective dose and risk late Col. Stafford L. Warren, a estimates derived from studies of other principal medical officer of the exposed human populations. However, Manhattan Project and the Chief of the there are considerable uncertainties Radiological Safety section at Operation associated with such estimates. Crossroads -- as the 1946 Bikini tests One of the main, uncertainties in were called.(I3) This relative neglect radiation-cancer risk estimates relates of internal exposures is a very serious to the questions now being raised about lacuna in the data because a large body the Japanese A-bomb survivor study - of medical evidence is beginning to the principal scientific reference used emerge that such exposures play a by standard-setting agencies. Current principal role in causing revisions in radiation doses, cancer radiation-related diseases. incidence and non-cancer effects data on So far as SRP workers are concerned the A-bomb survivors have made all one important piece of evidence is an present official cancer risk estimates unpublished DuPont study written in tenuous at best. 1976. (14) The researchers found Given these and other uncertainties 'evidence-that lung cancer and (discussed below), it is more leukemia were significantly increased" appropriate to evaluate when SRP workers were compared with radiationdnduced cancers in the context other DuPont employees and the general of a range of risks. Based on the public. A total of eleven cases of authorities citied above, we estimate leukemia were found among about 5,000 that the 1954-78 collective occupational SRP employees who were on the job up to radiation dose of 50,000 person-rems can 1974. Nine cases were found among wage be expected to result in between 16 and employees when 2 cases were expected. Of these, six were myeloid leukemias, when 2.86 was the expected number of when compared to the general cases. Myeioid leukemia is generally public. The summary states, recognized as being caused by ionizing "leukemia mortality did radiation. In October 1983, this data demonstrate a gradient with was reanalyzed by DuPont in a way which increasing radiation dose, but erased the statistical significance of was associated with long-term the previously observed employment in maintenance and positive findings. DuPont9s revised engineering jobs."(l7) analysis was deemed "inappropriate" by a DOE advisory committee reviewing the o In a "nested study. The committee also recommended case-controlw study of cancer this data be reanalyzed by a deaths among workers exposed to non-DOE/DuPont group.(l5) ionizing radiation at ORNL it Additionally, the health risks of SRP was found that "significant workers cannot be evaluated in a vacuum. excess risk was found for For example, the cancer findings among workers in workers at the Hanford facility in maintenance/construction and Washington state have great relevance to janitor/laborer job title SRP. Hanford was constructed and groups."(l8) initially operated by DuPont for the U.S. government in 1943, ten years o Data on employees 'who before SRP began operations. Like SRP, worked at the Oak Ridge Y-12 Hanford produced plutonium and processed Tennessee Eastman facility enormous quantities of radioactive between 1943 and 1947 show wastes. In 1974, excess cancer deaths "significant excess deaths from were discovered by the Washington Public lung cancer when compared to Health Department among Hanford workers, U.S. white male rates." A In 1977, DOE researchers published follow-up report states, findings that bone-marrow, pancreatic 'relative risk was found to and lung cancers were linked to Increase with increasing of radiation exposure levels I0 to 20 times lung dose even after below those considered uacceptabfen controlling for age, smoking under current f ederal standards.(l6) status and other work place Since that time, researchers at the exposures."(19) Oak Ridge Associated Universities (ORAU) and Los Alamos National Laboratory under o A mortality study of contract to DOE have discovered the workers at the Y-12 weapons following additional evidence of excess plant operated by Union Carbide cancer deaths among employees at DOE Corp. (until 1983) has found facilities: *excess deaths for cancer of lung, brain, and CNS (central o Workers at the Oak Ridge nervous system], Hodgkin's National Laboratory, according disease, other lymphatic to an ORAU Project summary, tissue." Brain tumors among show a 49% excess of leukemia Y-12 workers, was reported by Dr. C.C. Lushbaugh at an o A study of 2,529 DOE "in-housen meeting of DOE workers who were reported to researchers to be 100% to over receive over 5 rems between 400% in excess of what is 1947 and 1978 shows a expected in the general "significantly eleva ted" excess public.(20) for cancer of the rectum -- 3 times the national average.(25) o An epidemiologic study of deaths among workers There is a definite and alarming employed at the Oak Ridge industry-wide pattern here. The workers Gaseous Diffusion Plant, in the 200 Area of the Savannah River demonstrated "excess deaths due Plant form a part of it. For many of to lung and brain cancers and them internal and external exposures respiratory disease..." may be a cause of the radiation-related Enrichment workers exposed to diseases which they have or might nickel powder show "excess contract. The uneven quality of the data deaths from cancers of the on external radiation, the serious lack buccal cavity and of data on internal exposures, the use pharynx..."( 21) of obsolete and misleading methods for many of the measurements of internal o A 117% excess of deaths exposures as have been made does not from brain tumors has been speak well of the attitude of DOE and found among 7.1 12 workers at DuPont toward the health and safety of DOE'S Rocky Flats facility near the workers. We recommend a complete Denver, Co.(22) disclosure of data relating to worker exposures, and of the dose and cancer o At DOE'S uranium risk estimation methods. processing plant in Fernald, Measures also need to be taken to Ohio, ORAU researchers report reduce worker exposures. As has been that "after taking age at discussed in chapter 4, remote shielded diagnosis and age at hire into cabs have not yet been successfully been account there is an association developed for many operations in very between exposure to uranium and radioactive environments. There have the development of probably been many improvements in non-malignant respiratory operating procedures over the years. disease..? Furthermore, But the frequency with which non-routine "digestive cancers show a 36% maintenance is required and the excess..."( 23) persistence of problems such as plugging of pipes, faulty reel tapes, and pump o Deaths from malignant failures leads one to suspect that any skin cancers among workers at such improvement in operating procedures DOE'S Lawrence-Livermore may have been offset by the larger Laboratory was found to bc quantity of wastes (in terms of three times the national radionuclide content) and the average.(24) increasingly radioactive environment in Table 7.1

Range of Dose equivalents ORGAN from Plutonium Nitrate (rem)

Red Marrow Bone Surface Lungs which the workers must do their jobs. 9. Mancuso, T.F. et al., "Radiation In other words, until a more definitive Exposures of Hanford Workers Dying from analysis can be made based on more cancer and other Causes," Health quantitative data than we now have, it Phvsics, Vol. 33, No. 5, Nov. 1977. will remain an open question whether the 10. Sauer et a]., "The Risk of Death "learning curve" has resulted in any in Counties near the Savannah River decreases in exposures, or whether these Plant, 1949-1978," Draft, 19 October have actually stayed the same or even 1983. increased over the years. 11. Final Environmental Impact Statement, Lonn Term Mamement of oactive Wastes References -ah River Plant. Aiken, =November 1979 (DOE-EIS-003). 1 Franke, B. and Alvarez, R., 12. The low estimate is based on the "Analysis of external gamma radiation National Academy of Sciences BEIR I11 data around the Savannah River Plant, report, and the high estimate is based AMBIO Stockholm, May 1985. on John Gofman's book. See note 25, 2. -Savannah River Plant - 200 Area Chapter 5 for details. Fault Tree Data Bank - F&H Area. Waste 3 Arjun Makhijani. . and David Tank Computer Printout of Data Albright, The -n of Personnel Bank provided to the Environemtnal ation Crossra International Policy Institute in July 1983, with Radiation Research and Training entries dating from December 1953 to Institute, Washington, D.C., 1983. November 1982. 14. E.I. Du Pont de Ntmours and 3. Snyder, W.S., "Major Sources of Company, Savannah River Plant, Error in Interpreting Urinalysis Data Unpublished Report, 1976. Shy, C.E., to Estimate the Body Burden of Pu-239: A Epidemiologic Issue on Occupational Preliminary Study," Health Physics, Vol. Mortality and Morbidity Studies of 8, pp. 767-72. Department of Contract Facilities in: 4. Swanberg, W.S., "Major sources POE Radlatlon. . Er>idemioloev conttactors of Error in Interpreting Urinalysis Data Pro--, to Estimate the Body Burden of Pu-239: A April 13-14, 1982, US. DOE, Office of Preliminary Study," Health Phusics. Vol. Health and Environmental Research. 8, pp. 767-72. 15. Prajccts 5. Task Group 1966: Deposition and ered P~~siblcto Urntake ~n Retention Models for Internal Dosimetry Po~ulationsaround the Sav- River of the Human Respiratory Tract," Task m.Provided by the Epidemiology Group on Lung Dynamics, Hm Consultants to the U.S. Department of phvsics. Vol. 12, pp. 173-207, 1966. Energy, October 25-26, 1983; Final 6. Ibid. Draft, May 1984. 7. Ibid. 16, Mancuso, T.F, Stewart, AM and 8. Eckerman, K.F. tt a]., "Internal Kneale, G.W., "Radiation Exposures of Dosimetry Data and Methods, of ICRP-Part Hanford Workers Dying from Cancer and 2, Vol. 1: Committed Dose Equivalent and other Causes," Wth Ph- 33/5, Secondary Limits," NUREG/CR-1962, Vol. November 1977. I., 1981. 17. Goldsmith, R., Epidemiologic Project Summaries; Oak Ridge Associated Univtrsities/University of North Carolina, May 1984, U.S. Department of Energy, Office of Health and Environmental Research, p. 31. 18. Ibid. 19. Cookfair, D.L. et al., "Lung Cancers Among Workers at a Uranium Processing Plant", in: Proceedinas of oe 16th Mid-vw To~icalMeetinp. of the Health Phvsics Societv -- E~idemiolo~y d to Health Phvsi~~,Albuquerque, N.M., Jan, 9-13, 1983. 20. Op. Cit. Ref. 17 and Lushbaugh, C.C. et al., DOE Epidemiology Contractors Workshop, Program and Working Papers, April 13-14, 1983. U.S. Department of Energy, Office of Health and Environmental Research, p. 17. 21. Op. Cit. Ref. 17, p. 23, 2-7. 22. Op. Cit. Ref. 20 (see statement of G.L. Voelz ct al., p. 5). 23. Op. Cit.Ref. 17, p. 2-12. 24. Op. Cit. Ref. 20 (see statement of G.L. Voclz et al., p. 4). 25. Op. Cit. Ref. 17, p. 5-1, end INTRODUCTION PART 11

The tables on worker exposures, environmental contamination, technical problcms and accidents have been compiled from the "200 Area Fault Free Data Bank - F and ?-IArea Waste Tank- a computer listing of such occurrences obtained by the Environmental Policy Institute through the Frccdom of Information Act. This listing comprises about 700 pages of computer print-out in which there are about 14,000 entries. Each entry is a summary of some problem in the Radioactive Waste Tank Farm - whether this bc a "non-routine maintenance" of an instrument or a major radioactive waste spill or worker exposure. The record comprising the "Fault Tree Data Base" is very uneven both in quantity and quality of entries, as is discussed in Part I. The frequency of entries increased from about 4 per year during 1953-59 to about 1800 per year during 1977-82. The variation in the frequency of entries bears no apparent relation to the frequency of problems. Rather the range of problems and method of reporting seem to have changed. While the range of problems reported on has become much wider, the quality of entries has not improved. In some cases, it bas tended to deteriorate. For example, there are many entries during 1980 which simply state 'non-routine maintenance* on such-an-such piece of equipment, without even a brief statement of the problem or the nature of the corrective action. Further, as noted in Part I, there are instances when important problems have been left out of the Data Bank even after 1977. The Data Bank entries are divided into two chronological sets, one each for the F and H areas. The earliest entry is dated December 20, 1953, and the latest was on November 30, 1982. We bave chosen to group the entries into eleven tables based on the following problem areas: o Worker Exposures o Environmental Contamination o Tank Leaks and Overflows o Tank System Failures and Problems o Explosions: Potential and Actual o Equipment Plugging o Power Supply Failures o Pump Failures o Instrument Problenis o Miscellaneous Leaks o Miscellaneous Each table is organized chronologically with one set of entries each for the F and H areas. The first two tables, "Worker Exposures" and "Environmental Contamination" contain every explicitly recorded entry on such exposure or contamination in the Data Bank. The entries are not repeated, however, SO that there are many entries in the "Worker Exposuresn table which also deal with environmental contamination and vice-versa. This non-repetition of entries that apply to more than one category applies to all the tables. Our resources did not permit a fully cross-referenced compilation, which might provide further useful insights, The remaining tables, ie., Tables 3 through 1 1, contain selected entries whose purpose is to illustrate typical and frequent problems as well as to record the range of problems encountered. Frequent problems, such as plugging of certain equipment, or failure of cooling coils are noted in parenthetical comments, All our comments are given in parenthesis. The text of the description, when in quotation marks is directly from the Data Bank entry; when not in quotation marks it is a summary of the Data Bank entries. Our clarification inside quotation marks are inside square brackets. TABLE 1

WORKER EXPOSURES

DATE AND AREA DESCRIPTION AND COMMENTS

02-00-60F Skin and clothing contamlnatlon - 2 people - safety violation.

"A transfer jet was moved from Tank No. 19 and Installed in Tank No. 20 Body exposure rates to 1 R/hr. were experienced. Contamination to lR/hr. was released Inside the windbreak protected Job site when liquid leaked from the jet dip trlbe durlng rem~val.~

Leak in diversion box. Radiation level increased to 25 R/hr. Body exposure rate 5R/hr. Adjacent ground areas contaminated to 320 M/Rad/hr at 2 ubcges,

Felled tank 18 evaporator feed pump. "Body exposures ranglng to 30 Whr. at 18 inchesn durlng replacement. Asphalt also contamlnated to 5 rads/hr at 5 Inches. "Total estimated personnel exposure was 0.8 R."

Cooling water contaminated. Faulty valve and leaking tube bundle 1.5 curies of Cs-137 released to seepage basin. 'Estimated 2.6 R exposuret1 to workers during "clean up."

6 workers - nasal and skin contamlnatlon during removal of defective reel tape. (Reel tape problems are frequent.)

Skin contamfnation to wrist. Gloves not taped to coveralls.

Total of 3.4 R worker exposure "while lightenlng packing glandsew "Exposure resulted from high Irradlatlon levels In feed pump enclosure.^

Exposure during removal of a valve. "Nasal contamlnatlon 400 to 70,OO Dls/m. beta gamma," 4 people. "Hands, face and personal Items contaminated to 2000 c/m beta-gamma. Bloassay - 13 NCI, Cs-137/1.5 L. Body count = 84 NCI Ru-106, 368 NCI Cs-137.''

"242-F. Personnel received 13 R total exposure." (No explanation glven.)

Hlgh radiation levels near vent fllter up to 30 R/hr B 1 Inch. Replacement cased estimated 3.5 R exposure.

nExhaust filter; Radiation drain build-up of radloactivlty on the original exhaust fllter for tanks 18 8.20 resulted In exposure dose rates to 3 m/hr In the control roan and lunch room.8a

(The entry just above appears to have been repeated by mistake.)

Reel tape replacement, exposure was 30/30 mrads/mr/hr. at 12 inches."

Jumper problems on Concentrate Transfer System Tank. Exposed at top of open riser 2000 mrads/2000 mr/hr. WORKER EXPOSURES. Paae 2

DATE AND AREA DESCRIPTION AND COMMENTS

09-26-74F 1.4 R total estimated exposure to 14 people during feed pump replacement. Total 1-1-74 to 09-26-74 exposure due to "evaporator feed pump Joba1 above 1s 11.5 R

"14 R received by personnel installing new feed system In 242F.O

1 worker 4,000 c/m beta-gamma on shorts, 30,000 c/m on coveralls. "His flash suit pants had a hole.1a

Steam condensate leak repair. "Exposure rate 500/500 mrads/mr/hr. at 12 inches from spool piece."

2R exposure during 2 replacements of evaporator feed pump.

3.8 R due to concentrate tank lines be1ng uncovered.

1 R durlng concentrate transfer system inspection and excavatlon. 2-5 R/hr at 18 inches.

3.7 R during removal and shipment of 18 jumpers. "6.8 R during removal of CTS tank from pit. 10 R/hr. O 15 ft." ,

"Contaminated soil encounted tslcl durlng excavation around Riser 6 Tank 3. 350 R/hr ti? 1 Inch from steam supply line to the jet In Tank 3. Probably the result of suckback and leak. Soil contained about 50 Ci 137-Cs."

"CTS Plt - Waste management began steam cleaning inside the CTS pit. Work was stopped when steam vapor was observed seeping thru cracks in the hut. Sample taken outside the hut read less than 1 x 10'~ micro c1 FP/cc and 5.0 x 10'l2 micro cl PW Isicl/cc on Initial count.n

Total exposure at CTS pump pit excavatlon to date 21.6 R

Low actlvlty waste transfer to 241-F. "Got gassed out several tlmes."

10 R/hr radiation level O 2" durlng removal of piping. (Worker exposure not given.)

"Radlatlon rate 10R/hr close to Riser 6, Tank 3 durlng contarnlnated so11 removal. Sol I to 500 rads/200 R/hr at 2 Inches. 100 cu. ft. contained 8-1/2 C1 of 1374s. Excavation stopped."

1-5 R during evaporator feed pump replacement. Pump had leaked.

Change tral ler floor and bench contarnlnated. Up to 150 c/m beSa-gamma. successfully dec0nned.n

Tank 7 feed pump leak repair. Safety vlolatlon. 2 workers. One had about 1000 d/m beta/gamma in nostrils.

0.7 R durlng valve repair tank 31.

3.3 R exposure durlng repalr of pump to tank 7 evaporator. WORKER EXPOSURES, Page 3

DATE AND AREA DESCRIPTION AND COMMENTS

11 -30-77F 30 c/m on control room floor and 50 c/m on lunch rm floor found. Cleared. (No estimates of worker exposures. No data on posslble sources of contamlnatIon.)

Filter replacement. "ETE [Estimated Total Exposure1 = 3.2 R."

Dlp leg of transfer jet replacement Itstopped when exposure rates Increased to 30 R/hr. at 2 ft. of the dlp leg." (Total worker exposure not given.)

Workers rubber shoe came off. Personal shoe contamlnated through cloth cover to 40,000 c/m beta-gamma. Working In hut on Tank 34.

2 contamlnated hard hats found In regulated locker area of control room.

"Breathing alr compressor shut down by waste management. Puff workers uslng Iltl had to evacuate malntenance area. "No communlcat1on between groups. No audible alarm.I1

3 workers exposed to CTS loop. No badges. Exposure estlmated 65 mr. Cause accidental removal of a fence.

"Employee scratched the back of hls left hand I durlngl ...CTS l lve work. No beta-gamma or alpha ~ontamlnatlon.~~

~Construct1onworker got 8,000 c/m beta gamma on gloves. Worked In a 241-F regulated area wlthout health physics coverage."

"People that were contamlnated B.G. went home at 7 p.m. 4-12 [shlftll' (No further explanatton.)

Construction worker hands contamlnated: t92,000 c/m beta-gamma on palm and back of hls left hand, 200 c/m beta-gamma on back of hls rtght hand."

"Unauthorized llne break on an evaporator process Ifne." 100 ml Ilquld contamlnated 1.5 sq. ft. area In gang valve house to 500 mr/hr. at 3 feet. One mechanic hands contamlnated to 35 mr/hr., gloves and coveralls to 6,000 c/m. Two mechanfcs had nasal smears of 300 d/m and 275 d/m. Suckback fn the llne had occurred."

241f. Worker had 1000 c/m gamma on rfght palm and 600 c/m beta on left palm.

Dralnlng and drylng of probes In LDB's whlch had inIlquld In them. "No smearable contamlnatlon.~l

"A dip tube had to be mined loose from approximately 12 feet of salt In Tank No. 19. Total estlmated exposure for 36 people was 1.2 Ren

"Two constructlon carpenters entered a barrlcaded excavatlon adJacent to F area dlverslon box No. 2 whltle waste was belng transferred through Ilnes exposed by excavatlon. The carpenters recelved 230 and 160 mr respectively." WORKER EXPOSURES, Page 4

DATE AND AREA DESCRIPTION AND COMMENTS

04-2 1-8OF Tank 33: Ilquld radlatlong 500 mrad/hr. at 5 cms spilled onto ground. 3,000 to 80,000 c/m In a 6 sq. ft. area. 3 workers had skin contamlnation to 60,000 c/m beta-gamma.

04-2 1-8OF "Hard hats In change and trlaler floor smearlng 8,000 c/rn."

04-23-80F Tank 33: Repalrs to hut. 2 men got 30 mr. "The top of the tank, out approx. 2 m from the hut, was found kontamlnated to a max of 40,000 c/m beta-gamma at 2.5 em. transferable."

04-28-80F 'A Juno wlth 200 mrads/hr. of smearable beta-gamma was left in seat of [pick-up truck]. Seat and floor board contamlnated to 30,000 c/m. Health physics Inspector contamlnated the seat of hls personal trousers to 2,000 c/m.I1

07-1 1-8OF Worker had 1,000 c/m to 2,000 on both shoes and 1,000 c/rn on bottom of left sock. Had been worklng on tank 20.

08- -80F ~F1berglassf Ilter for Tank No. 1 removed.11 It contained estimated 0.8 CI actlvlty. "Total estimated worker exposure for 65 employees was 2155 mr."

12-1 2-80F "Employee reported to medlcal thls a.m. with a slight wound on the back of her left hand that she said she recelved at DS-1 on 12-4-80."

12-20-80F "Worker removed hls p l astlc suit top at bottom of DB-1 due to the hoses belng tangled. Nasal smears taken were less ms."

01-06-81F Tank 18 transfer jet repalr. Dose rate: 19250/250 mrads/mr/hr. at 30 ems."

02-24-8 1F Tank 25: Repairs. wExposure dose rate to 350 mrad/350 mr/hr.I1

03-16-81F HEPA filters changed. "T8T operator had 2000 c/m beta on his throat area." "North HEPA DOP tested 99.94%. ."

04-1 0-81 F Ph-12 alkaline solutlon ("lnhiblted wateru1) sprayed Into hut and on workers durlng probe of Tank 2.

05-1 9-81 F "As sampler was pulled from tank 1 It brought wlth It discarded steel tapes from the tank Intertor. Liquid on the tapes contaminated the wrlsts and nasal passages of the foremanf1 because he "grabbed the steel tape, broke It lnto Islcl and pushed It back In hole before h.p. could stop hlm. Skln: 5 mrads/hr. beta-gamma at 5 cm. on rlght wrlst; 60,000 c/m beta-gamma on left wrlst. Nasal contamfnatlon 590 d/m beta-gamma.lg

Mechanic changlng feed pump. Tank 7, had 2 pairs coveralls on mask. 20,000 c/m beta-gamma on right sleeve of inner coveralls and 2,000 c/m on skln on underslde of rlght forearm.

06-01-81F Removal and shortening of transfer Jet. 06-01-81; 3R to 19 men. Again 6-14-81. 1.7R for 16 men. One worker had 1,000 c/m beta-gamma on left forearm. Causes: repalr to jet; p l uggage; leak in diversion box; Inspectlon. Dose rate: 50/mr/hr.

07- -81F Lead Llner was Inserted In riserr#3. Radlatlon level over rlser reduced from 15 R/hr to 2.5 R/hr. Estimated total exposure to 12 workers: 565 mr- WORKER EXPOSURES. Paae 5

DATE AND AREA DESCRIPTION AND COMMENTS

07-20-8 IF Worker contamlnated hls left shoe to 1,000 c/m by stepplng out of shoe cover at hut of Tank 26.

07-21-81F Clean-out port to Dlversfon Box 5. '#Water and salt backed out Ifne and contamlnated employee." Rubber gloves 150 mr/hr. at 8 cm., left cover was 10 mr/hr. at 8 cm. Left shoe (personal ) 7,000 c/m beta-gamma. Cause: apparently a plugged Ilne.

07-25-8 1F Same clean-out. Worker contaminated to 2,000 c/m above port 815 after Ilquld sprayed from line.

12-21-81F Tank purge HEPA fllters fall. Estimated release to atmosphere: 10 mlcro- curies of flsslon products. Total exposure of 2 reds to 20 persons.

01-28-82F Further repalrs tank 28 Jet. "Body dose rates ranged to 8 r/hr. at 45 cms of open rlser...Employee knelt on damp cardboard." He had rlght knee: 3,000 c/m beta-gamma; left knee: 1,000 c/m beta-gamma; another had 2,000 c/m beta- gamma on rlght knee. Backflush valve tnstalled In Tank 25. "Total exposure to 5 men was 100 mr." Leak repalr concentrate transfer sysiem plt. ItTotal exposure to 15 men was 720 mr."

Work on failed sump Jet steam line. OITotal exposure to 4 people was 85 mr."

Extensive work on transfer Jets to replace gaskets, flush and allgn Jet. "Body dose rates ranged to 8 R/hr. at 45 cm. above open riser. Total estimated exposure lnvolvlng 30 people was 3,000 mr."

Blow back of water from "catheter Itnew: fused to unplug plpesl at Tank 25. ~Contamlnatlonto 500 mr/hr. at 8 cm....lnsIde the enclosure on the outer SWP clothlng of the operator.

Interarea pump plt repalrs. "Total exposure for 17 people was 580 mr."

Tank #27 transfer Jet leak. Jet ralsed 3 tlmes to regasket It. "Body dose rate was 2 R/hr at 0.5 meter above open rlser...Total exposure was 1,025 mr. for 6 people."

Power outage to Tank 33 and 44 ventllatlon system. Durlng survey of vapor space skln contemlnatlon to foreman: left ear, 600 c/m (beta-gamma); halr, 2,000 c/m; lnslde bath forearms, 20,000 c/m; and left side of abdomen, 600 c/m. Personal clothlng was contamlnated up to 10 mrad at contact.. .. Foreman's car and a personal wash cloth at the foreman's home were also contamlnated. The foreman proceeded wlthout monltorlng as required. Blame partly lald to ninsuffIclent tralnlng.n (In splte of extenslve body contamlnatlon, the report of the Incident found "no nasal contamlnatlon or body asslml Iatton.")

Hydrogen survey tanks 33 and 34. Foreman had "skin contmlnat1on to 20,000 c/m beta-gamma - deconned to no detectable. Her clothing and personal Items - several were contamlnated to 25,000 c/m beta-gamma probe." A Health Physics department truck "contamlnated to 100 c/m beta-gamma smearable and probed et 8,000 dm. Hot smear problng 280 mr/hr at 5 cms was found In the truck.O1 WORKER EXPOSURES, Page 6

New CTS ventllatlon system construction. 81Body dose rates ranged to 20 mr/hr. The total estlrnated dose was 560 mr for 43 people.

Excavation around tanks 17 and 18 "to install foundatlons for salt removal equlpnent. Body dose rates ranged to 100 mr/hr. The total estlrnated dose was 515 mr for 28 people." installation of pump tank transfer Jet. "Total exposure was 80 mr for four people." nConstructlon concrete worker probing from 2.000 c/m to 6,000 c/m all over h Is person.. .May have had radioactive Injection. Determined he had received an InJectlon of thalllum gtven by doctor."

Installation of new CTS ventllatlon system. "Body dose rates ranged to 50 mr/hr In the trenches. The total estlmated expousre Involving 201 personnel was 3565 mr."

Concentrate Transfer System vent line construction. CTS Just entered for work. "Total exposure for seven people was 550 mr."

CTS vent system construction. Body dose rates to 40 mr/hr. Total estimated dose to 158 people: 2030mr.

Tank #26 feed pump change. Dose rates to 2.5 r/hr. at 0.5 meters above open r lser. "Tota I estlmated exposure to four people was 165 mron

Leaking evaporator condenser replaced. Dose rates to 1 r/hr. Total estlmated dose to 30 workers was 635 mr.

Replacement of falled pump In the concentrate transfer system. Total exposure for 9 workers: 340 mr.

CTS pft reentered to check recirculation pump. 435 mr to 11 workers.

Check out of CTS vent system. 900 mr to 24 workers.

Tank 83 repairs. Dose rates to 1-5 r/hr at 50 cm above open rlser. Estimated exposure to 40 people: 895 mr.

2F Evaporator work. Exposure to 20 people: 400 mrem.

Tank 833 transfer jet repalrs. Total exposure of 865 mr to 7 people.

CTS vent system modifications. 50 mr total exposure to 4 people.

Instal latlon of asphalt on tanks 1,2,3,4. Body dose rates ranged to 100 mr/hr at 60 cm above tank tops. Total dose to 135 people estimated 1,390

"Installed a pump insert plug In the donre riserw at Tank 817. Dose rates to 800 mr/hr at 45 cm of r lser top. Total est imated exposure to 8 people: 140 mr.

Dlverslon box FOB-4 entered twice for lnspectlon and repair of leak. Total estlmated exposure to 15 people: 560 mr. Dose rates ranged to 500 mr/hr a+ 1 meter above open box. WORKER EXPOSURES, Page 7

DATE AND AREA DESCR lPT 1 ON AND COMMENTS

10-1 5-82F FDS-4 entered agaln to verify valve operation. Total exposure to 7 people was 470 mr.

11- -82F Replacement of leaklng nozzle gaskets on Tank 26. Body dose rates to 8 R/hr. at 0.5 meters above open riser. Total estimated exposure to 12 people: 2465 mr. 1 worker knelt on wet lead shleldlng contamlnatlng hls rlght knee to 4,500 c/m. Decontaminated.

1 1-1 9-82F CTS vent system check ouf preparation. Total exposure for 5 people: 135 mr.

F AREA L l ST1 NG ENDS

- ~ -- H AREA LISTING BEGINS

02-1 561H Re1 sing of thermocouple were from annul us plug on Tank 16 caused contam- Ination of 2 workers "up to 6,000 c/mgl and equipment contarnlnated up "to 15 r/hr. on a rlser plug.n llApprovedprocedures" not to1 lowed.

09-2 1 -65H Suck back, ascrlbed to leaktng valve. Contaminated control room to 5 mr/hr.

08-3 1 -66H "241-H tank 21. Hlgh exposureu to seven workers. Exposure levels to 7 r/hr. "wlthout H.P. [health physlcsl monitoring the max expose was 565 mr." Exposure was durlng excavatton of a trench over waste tank 24.

nRadlatlon exposure. See SHI 243."

"Film badge of sep. dept. supv. indicated skln exposure of 14,590 mrads during Oct. exceedlng AEC manual quarterly standard of 10 rem. The exposure probIably1 occured when the employee removed two acme nuts from a 241-H waste transfer pump in railroad tunnel airlcck. See speclal hazards lnvestlgatlon 266."

Worker contaminated hands to 150 mrlhr. at 3 Inches whlle aWemptlng repairs [to evaporator] wfthout proper protection equipment.

"Exposure rates to 20 rads/hrn during removal of transfer jet from tank 14.

Nasal contamination of worker during wor*n Tank 15 removal of reel tape to 30,000 c/m beta-gamma and face and hair contamination of 10,000 c/m. "Apparently occured durlng removal of protective equipment at end of Job."

Removal of slurry pumps and equlpmsnt from Tank 14. Exposure rates to 10 rads/hr. nEstlmated exposure to personnel was 9 radon

"8 constructton mn and 1 HP inspector received nasal contamlantlon durlng change of four slurry pumps frm tank 11. Contamlnatlon levels were 530 to 43,500 d/m beta-gamma.n

"Undetected leak resulted In airborne contamlnatlon to personnel and equlpment. 3 employees nasal contamlantlon to 7,200 c/m beta-gamma. A vehicle and pavement around DB-2 contamlnated." WORKER EXPOSURES, Page 8

DATE AND AREA DESCRIPTION AND COMMENTS

08-23-72H Worker exposure durlng grtndlng of contamlnated plplng. 4 workers had contamlnatlon to 1,500 Urn beta-gamma. "One constructlon personnel received an asslmllatlon. Bioassay = 0.65 d/m 1.X, chest - 33 NCI."

02-28-74H 1 plnt of contamlnated llquld sprayed from a leak. "Grating of the catwalk around the evaporator cell was contamlnated to 8 Rads/hr...2 maintenance mechanics were contamlnated by falllng droplets. Nasal contamtnat1on up to 1.345 d/m beta-gamma. Body contamlnatlon 300 mr/hr at 2" from arm, 1st mech. bioassay = 12 NCI; Cs-137/ 1.5L Chest count = 262 NCI; Cs-137, 43 NCI Ru-106. Body burden (1% WBBI. 2nd mech. bioassay = 64 NCl , CS-13711 .5L Rec'd (2% tPBB).I1

Student contamlnated rlght hand to max. of 1,500 c/m beta-gammb.

Jumper repairs DB-4. Liquid flooded bottom of box.

Worker s "dos Irneter was off scale at conc Ius lon of Job. H 1s f II m badge was pulled and thls revealed he had accumulated 590 mrads/590 mr.I1

Radlatlon rates to 40 R/hr. above open riser, Tank 29. Personnel exposure rates to 3 R/hr. "Transferable contaminatlon to 2.E06 d/m foot square alpha and 40r000 c/m foot square beta-gamma wlthln plastlc hut."

I1Exposure rates to 400 mr/hr to construct Ion personnel during Ilve connections between Tank 13 and evaporator."

Failed evaporator feed pump. Exposure rates to 10 R/hr. Total exposure to 9 people 0.8R.

CTS leak detectlon system Instal latlon, exposure rates to 100 mr/hr. Total exposure to date 4R.

CTS leak repalr. Radiation exposure rate 3 R/hr. at 3 ft. above all. 4.8 R personnel exposure durlng lnspectlon and leak repalr.

"Tank 15 - outslde coveral Is contamlnated 6,00 c/m beta-gamma. No alpha on upper rlght leg." Student glven used coveralls.

Val ve rep laced. Ma 1ntenance mechan 1c's covera l ls cont am1 nated to 20,000 c/m beta-gamma at 1 with 3,000 d/m alpha. Alr sample at hut door outslde, durfng the Job llrevealed 1-75 x 10-9 mlcroci FP/cc and 153 x 10'12 mlcroct Pu/cc 1n a1r.l1

CTS tank drawoff pump replaced. Exposure rate to 500 rnr/hr. Pump radiated 25 rad/hr. at 2 Inches.

Techn lca l eng lneer contamlnated personel shoes to "1,500 c/-beta-gamma" by stepplng on ledge of tank 31 to remove thermocouple.

Tank 31 HEPA filter replacement. Dose rates to 20 Rads/lOR/hr at 2 Inches and 1-5 R/hr at 2 feet. WORKER EXPOSURES, Page 8&

DATE AND AREA DESCRIPTION AND COMMENTS

04-27-76H Evaporator cel I: AIr sample taken during Jumper change: 0.44 x 10'9 micro Ci FP/cc and 40.6 x 10-12 mIcro C1 Pu/cc.

05-30-76H Steam leak into Tank 22 sump. Air sample taken 10 ft. east of hut downwind measured 0.18 x 10'9 mlcrocuries FP/cc of air.

07-1 1-76H Hosed salt deposits In feed pump cell wIll water. Body exposure rate of 200 mrads/100 rm/hr at 2 ft. from hole. "No spread of contamInatIonu from hosing.

08- -76H CesIum removal column replaced. Exposure rates 700 mr/hr. 60 cm. over open rfser and 100 mr/hr at 0.5 from the column.

08-1 6-76H Pump pit No. 3. 3 workers contaminated clothes up to 40,000 c/m. 1 had 1,000 c/m on hair over right ear. Nasal smears for foreman 242 c/m to 300 c/m. AIr samples: 10.66 x 10" mlcroCi FP/cc and 47.7 x 10'12 mlcroC1 Pu/cc. Slmllar levels of air contamination on 08-17-76.

Hard hats at 241-H contamlnated "up to 18 c/m beta-gamma, no alpha.11 Survey on 09-10-76 revealed hard hats contaminated up to 670 c, beta-gamma.

Tank 31. Air samples during equipment removal: 86 x mIcroC1 FP/cc and 400 x 10" mIcroC1 Pu/cc. Exposure rates to 150 rads/80 R/hr f lush wlth open tank rtsers. Somewhat lower levels In aIr on 09-23-76 and agaIn on 10-14-76. 1.9 x 10- microC1 FP/cc and 40.5 x m1croCI Pu/cc at Tank 13 on 11-10-76.

Tank 13 drill motor repalrs. A1r activfty 3 meters east of hut at HEPA exhaust -16 x 104 mIcroC1, W/cc and 56 x 10-l2 microC1 Pu/cc. SIml lar levels on 11-19-76 when all covers removed In CTS. Again at Tank 13 on 12-03-76. Entry into Diversion Box 1 on 03-28-77. 90.1 x 10-l2 mIcroCi Pu/cc. Siml lar level durlng survey of Tank 16 for maintenance on 04-26-77. SimI lar level during removal of Jumper 08-2. DB-1 work. installation of "U1l Jumper. 49 x lo-' microC1 FP/cc and 18.3 x 10-12 microCl Pu/cc.

Tank 13. Workers hoses contaminated to 7,000 c/m beh-gamma.

lgattempted decontamlnatIonw of ba II vat ue radlatIng 1,000 rad/hr at 58 cm.

Tank 29. Liquid sp1 Il durfng repairs. Exposure rates 150 rads/100 R/hr. at 5 cms. ltPersonal shoes" contaminated.

Tank 29 repairs. Total exposure 1,800 mr.

"Contaminated water 15,000 c/m/ml, beta-gamma1 bubbled from ground at Tank 13 darr?ng hydrostatic testlng of a Ilne, faulty weld. Exposure rates to 1 R/hr at 30 an." Total exposure 2.65(.

"High personnel exposures to TBT workers on hot Job." WORKER EXPOSURES. Page 9

DATE AND AREA DESCRIPTION AND COMMENTS

05-03-77H "New pumphouse - high actlvlty was recorded In control room at 7:00 pm... 1 /I mrad/mr/hr. gen. area.I1

05-1 7-77H 08-5 repairs. "Exposure rate (hand rate also) was 20 rads/.2R/hr. at 5 cm. nozzle to regasket."

06-01 -77H Tank 16 malntenence. "Jet radiated 50 rads/4 R/hr at 5 cm."

09-1 5-77H Removal of valve In Tank 31. nRadlatlon level flush wlth open hole 50 radslhr. 15R/hr." 08-4. Gasket replacement. Body exposures rates to 8 rads/3R/hr. at 45 cm.

02-09-7M Evaporator cell : Plugged vent and transfer I Ine. Exposure rate 25R/hr. at 3 cm. "Repair personnel got about 2R.N

02-1 5-78H LIqu Id leak fcorn evaporator cell. Exposure rates to 50 R/hr at 1 inch.

04-02-78H "Repalrs to valve on evaporator feed pump In Tank 13." Total exposure 0.8 R.

05-0 1 -78H Replaced gaskets In CTS pit. Total exposure to 8 people: 950 mr-

05-1 9-78H RepaIrs to CTS draw-off pump. 4,000/200 mrad/mr at 18 Inches. 3 days to repalr.

07- -78H Removal of CTS agitator. Total exposure 1 R.

07-1 6-78H Up to 15.000 c/m on personal shoes of worker, probably whlle taklng a readlng at P-Trap rlser #8.

08- -78H Unplugglng of vent I Ine from 242-H evaporator. Total exposure to 15 workers: 800 mr.

08-03-78H ~8ConstructIonplpe f ltter, to 8,000 c/m beta-gamma at 3 cms. on chest (small area)." Clothes also contamlnated. "An open bag contalnlng used regulated clothlng was.found radlatlng 20,000 c/m beta-gamma at 3 ems."

09-1 8-78H Worker had 7,000 c/m "on hls neck In one small spot."

02-05-79H nEmp loyees burned eyes wh 1 I e watch I ng we l dl ng operat Ion. .I1

02-1 5-79H Worker contamlnatead rlght Index fInger and left hand to max of 30,000 c/m whlle handling bottles of contamlnated mercury.

03-02-79H 2 workers contaminated personal clothlng and shoes up to 30,000 c/m whIle working at Tank 16.

04-1 7-79H 281-8H. Miscellaneous bastn malntenance. "One employee's results was 620 mrems/620 mrems. The other employee's results was 825 mrems/825 mrems-

05-03-79H Survey prior to repalr of bad valve. RadlatIon rates over open hole of 1000 mr/hr beta-gamma and 1000 mr/hr gamma. WORKER EXPOSURESD Page 10

DATE AND AREA DESCRIPTION AND COMMENTS

06-04-79H Leak In evaporator bottom I lne. "100 workers accumulated 4750 rems.I1 IThls mlght be an error. The flgure Intended Is probably 4750 mrems.1

12-05-79H "CIS area - employee contaminated the front of his shirt and pants to 2,000 c/m beta-gamma."

0 1 -02-8OH Sol1 near 502 line measured 1 rad/hr at 5 an. Two workers contaminated their shoes to 1,000 c/m beta-gamma due to contamlnated soll.

01-25-8OH Tank 16. Tape taplng worker's gloves to coveralls came loose. Rlght forearm contamlnated 20,000 c/m, left forearm to 40,000 c/m.

01 -27-80H Regulated hard-hat found contamfnated to a max. of 200 c/m beta-gamma during weekly routlne check.

02-07-80H Worker's palm and foreflnger contaminated to 10,000 c/m beta-gamma while sampllng Tank 31.

03-28-80H Removal of speclf lc gravlty dlp tube from Tank 13. One worker's left hand contamlnated to 100 mrad/hr. at 5 an; right hand to 40,000 c/m beta-gamma. Second worker's wrlst to 40,000 c/m beta-gamma.

04- -80H 6 sq. ft. area contaminated whlle removlng dlp tubes. 3 construction workers received up to 60,000 c/m contamInatIon on skin durlng cleanup.

04- -8OH Installation of new llne at evaporator. Total exposure of 1,100 mr to 30 people.

04-1 0-8OH l nsfde of vehicle cab found contamlonted to 20,000 c/m beta-gamma during a survey.

05-1WOH "Extensive decontarninatlon of salt and gravel on top of Tank No. 13 was performed." Source of contaminatton apparently a crack In reel Jet pl l l box. Total exposure to 11 people to remove tar and gravel: 140 mr.

05-30-80H Work on alr compressor. Exposure rate 10 mrad/lO mrlhr.

06-1 9-80H CTS area - workers' shoes contamfnated from 800 c/m up to 8,000 c/m beta-gamma. One worker "stepped In water that came up over hls workshoes."

09-08-80H Dlsrnantled purge blower. Exposure rate ltlO/10mrad/mr/hr gen. area."

10-02-80H Repal red leek I ng Jumper connector at 242-H evaporator. "Body dose rate flush wlth open cel l was 2,000/1500 mrad/mr/hr. 3 shlfts.

10-1 0-8OH Removal of faulty feed pump (Tank 13) and Its transport. Exposure rates to 3 R/hr. Relnstal latlon exposure rate 8,000/mrads/3,000 mr/hr.

1 2-03-80H Check on feed probe on Tank 13. Worker's personal Jacket contamlnated to w15,000 c/m beta-gamma (no alpha)."

01- -81H nExposure rates to 30 rads/hr.2R/hr were encountered to remove a wlre that had become wrapped around a sludge proben In Tank 15. Total estimated exposure to remove wlre: 435 mrems. WORKER EXPOSURES, Page 11

DATE AND AREA DESCRIPTION AND COMMENTS

01-1 1-81H Accidental pressurization of rfser on Tank 31 during work. 10 sq. ft. area contaminated. Worker's cloth1ng and shoes contaminated from 2,000 c/m to 35,000 c/m beta-gamma.

Entry Into Diversion Box 4 for repairs: worker contaminated shoe covers-: and clothing to 20,000 c/m beta-gamma.

Loop I ine to CTS and Tank 29 p l uggage and ma lf unct Ion. Repair causes exposure of 2,220 mrem. Total to 17 persons. Exposure rates to 8R/hr.

Worker's clothes, including lnsulated underwear contaminated to 3,000 c/m beta-gamma while welding.

By-pass valve cracked due to freezing and contaminated tank top and ground. Worker without protective clothing contaminated shoes and trousers to 1000 c/m and hand to 200 c/m.

CTS line radiating 80 R/hr uncovered late Friday night. Measurement taken on Monday.

Safety showers at evaporator replaced. General area dose rate 10 mrads/ 10 mr/hr.

"TST rigger's exposure of 1,145 mr exceeded exposure gulde of 600 mr for month. He was changing jumpers in Diversion Box 4. Failure to communicate was principal cause."

Excavations for concrete pad at rlser 3. Total exposure to 11 persons: 310 mrem.

Workers from "Puritan Janitorial Serviceff exposed to ammonia vapor.

Construction worker at Tank 15: gloves contaminated to 40,000 c/m and tarp over rlser and plastic on floor to 60,000 c/m beta-gamma.

Transfer Jet screen plugged. Removal exposure rates to 10 R/hr. at 30 cm. Total exposure 3.2 R to 27 workers.

Removal of obsolete equipment from insider rlser 6, tank 9. 40 workers accumulated 3.7 R exposure.

Failed draw-off valve removed from Tank 36. Total exposure over 2-day job to 7 workers was 190 mr.

lnstailatlon of Thermocouple on valve rlser C-2, Tank 36.

Removal and regasketting and relnstallatlon of jumpers in DB-6. 37 workers total exposure: 4.6 R.

Riser plug removals. Exposure rates to 1.5 R/hr. Total estimated exposure to 37 workers: 2265 mr.

Change of bad gasket on trasnfer Jet, Tank 36. "WM and T8T changed gasket on transfer jet at C-1 rlser using a radiation dose rate of 5 radsN R/ hr at 0.5 m from connector. Radlatlon level at top of riser was 20 rads/ 15 rads/hr." WORKER EXPOSURES, Page 12

DATE AND AREA DESCRIPTION AND COMMENTS

07-1 3-81H "Employee 1nadvertent Iy contaml nated h 1s personal clothlng t to 6,000 c/m beta-gamma1 after removing a rope [safety belt rope1 from an open bag ln the 242 supply room."

Rlser jet put on truck for dlsposal leaked. Both shoes of worker on truck contamlnated to 80,000 clm beta-gamma.

19Constructfonbollermaker and laborer Incurred skln contamlnatlon while handling a portable hand saw whlch was contaminated. The laborer also sustained nasal contamination."

Contlnuatlon of work on transfer plplng lnstallatlon. 43 workers exposed to a total of 1405 mr.

Installation of new nozzles in 08-6. Total exposure to 31 workers: 1885 mr.

"Employee contamlnated himself and clothlng when he removed draln plug rod...at the CTS pit. Procedures not followed.o1

T8T rigger contaminated arms and clothes up to 20,000 c/m after handling used tag I ine.

Construction plpefltter handled contamlnated weldlng lead. Right hand contamlnated. Sme area at 20 mrads/hr.

Failed reclrculation pump. "1035 mr for 11 men."

Radlatlon survey lnslde transitlon box showed max. radlatlon level of 300 R/ hr at 3 Inches.

Installation of new steam line. Dose rate 70 mrads/70 mr/hr. In the "general area.

4 bat l valves In CTS cel l rebul It. Exposure rates to lR/hr. Total exposure to 13 people: 750 mr.

Repalres to valve jumper, CTS pit. Exposure rates 500 mr/hr. at 30 ems. from jumper and 1R/hr over open cell. Total estimated exposure: 315 mr. Reentry Into cell on 09-29-81. Total exposure to 6 people: 180 mr.

"CTS plt - E81, CfMO, and T8T attempted to repair the autmatlc valve using a dose rate of 500 nu/hr."

Worker contamlnated hands and clothes up to 4,000 c/m.

New thermocouple installed "using a dose rate of 100 mrlhr."

Jumper placed in burial box.' Total exposure to 14 people: 2340 mr.

lnstallation of jumper ln 06-2. Total exposure of 8 people: 775 mr.

Replacement of felled transfer pump and other repairs In CTS pit. Total exposure to 62 people: 2400 mr. WORKER EXPOSURES, Page 13

DATE AND AREA DESCRIPTION AND COMMENTS

10- -81H "Repair a bent nozzle lnslde the cell. The total exposure to (Involving 70 construction, waste management, and T8T personnel) was 3900 mr."

10-04-81 H "Employee contaminated his right shoe to 2500 c/m beta-gamma while worklng In 281-8F retention

10-09-81H "Flush water jockey pump down..." Dose rates: up to 2,000 mrads/200 mr/hr at 30 cms.

10-12-81H Worker knelt In contamlnated water on cell covers. Rlght knee: 10,000 c/m beta-gamma.

10-16-81H Replacement of contaminated fllter on experimental gas sampler. Worker's hand and clothes contamlanted to 70,000 c/m beta-gamma. "No nasal contaw lnatlon or body asslmllatlon.l~

Tank 13 feed pump work. Total exposure to 38 workers: 1495 mr. CTS work: Exposure to 18 workers: 1990 mr.

lnstal lation of new Jumper for DB-4, DB-6 transfer system. Total estlmated 11 workers: 640 mr. , Work for startup of evaporator. Total exposure to about 20 workers: 950 mr.

Installation to replace failed feed pump. Exposure to 21 workers: 1245 I mr. Decontamination and repalr of another feed pump. "Total exposure to 40 men...was 33,000 mrad and 2,000 mr."

Installatlon to replace falled feed pump. Exposure rate to 2000 mrads/ I 100 mr/hr at 0.5 m from pump. Total estlmated exposure to 32 people: 1245 mr.

Tank 31 - Worker contamlnated right shoe to 3000 c/m beta-gamma, less than 500 d/m alpha.

Changes In DB-7. Total exposure to 9 workers, 70 mr.

Tank 38: lengthenlng of transfer jet. Total exposure to 5 workers: 570 mr.

Dlverslon Box 1. Jumper changes. Total exposure to 8 men was 1,030 mr.

Work in "HPT-3" mainly on transfer Jet. Total exposure to 13 men: 980 mr.

Installation of new Jet C-1 riser. Radiation level at top of open riser up to 20R/hr. Total estlmated dose to 5 workers: 605 mrem.

Tank 13. Removal and 1nsta l l atIon of pump. Total of 4 entr les needed to complete the Job. Radiation exposure rates to 9R/hr. Total estlmated dose to 29 people: 1560 mrem. [Same Item appears to have been entered on 01-02-82H wlth exposure of 1420 mrem to 15 people??]

Dlverslon Box 2 Jumper changes. "Total exposure to 11 men was 1,160 mr."

Replaced stator on reclrculatIon pump motor. Total exposure to 15 workers: 1,240 mr. WCRKER EXPOSURES, Page 14

DATE AND AREA DESCRIPTION AND COMMENTS

0 1-06-82H CTS pit - transport of contaminated metal angle rlser. 3 workers' coveralls, gloves contamlnated from 10,000 c/m beta-gamma to 100 mr/hr at 5 cms. Pick- up truck contamlnated to 10,000 c/m beta-gamma.

01-25-82H "Waste header tle-In. HP [Health Physicsl started to set a rate for construction and found lfquid drlpplng from canyon encasement raln Ilne. The radlatlon rate was 60 rads/20 R at 18 Inches from I lquld...Approxfmately 5 gallons had dralned Into sump.81

Cfft jumper vent llne repaired "using a dose rate of 1000 mrads/200 mr/hr. at 0.5 meters from jumper."

Hlgh radiation rate In Dlverslon Box #5 flug box - 200 rad/l5 R/hr due to a leak. Total exposure to 3 workers: 350 mr.

Work on jumpers - total exposure to 12 workers: 895 mr.

uConstructlon completed tie-In of the new jacketed waste headers. Problems were encountered when llquid radiatlng up to 150 rads/hr drlpped from several of the headers."

Tank 13 f a1 led feed pump. Exposure rates to 3R/hr above open r 1ser. Total estlmated exposure to 10 workers8 1035 mr.

Tank 31 work. Total exposure to 3 workers: 155 mr.

nConstruction chipped through four feet of concrete on top of Tank 38" to weld coollng weater line to tank prlmary. Radlatlon levels up to 600 mr/hr. at 5 cms. from tank top. Total estlmated exposure to 24 workers: 920 mr.

Dlverslon box 7 repairs. Total exposure to 27 workers: 4,370 mr.

Preparation for lnstallatlon of Jet for Tank 9. Total estlmated exposure to 17 people: 1030 mr.

Work In 08-5 flRadlatlon levels were reduced from 40 rads/2HIslcl/hr to 10 rads/lR/hr at 5 cms. Radlatlon dose rate of 2,000 mrads/2OO mr/hr at 0.5 m was establ lshed..."Total estlmated exposure to 7 workers: 215 mr.

Dfverslon box 85 hose Instal latlon. "Job was stopped when outer palr of coveralls were contamlnated to 15 mrads/hr at 2 Inches."

"DB-5 - WMO installed a flush valve on nozzle no. 9 flush valve uslng a dose rate of 3000 mrads/300 mr/hr. Estimated total exposure: 110 mr."

Entry unto 08-2: Jumper change. Total exposure to 7 men was 610 mr.

CTS - removal of radloadlve materials for burlal. Broken jumper radiated 4000 mrads/300 mr/hr at 5 cms.

Entered CTS plt for inspection of pump. [Exposure rate or total exposure not glven.1 WORKER EXPOSURES, Page 15

DATE AND AREA DESCRIPTION AND COMMENTS

05- -82H Failed feed pump decontaminated from 20 rads/lR/hr. at 30 cm to 1000 mrad/ 100 mr/hr at 30 cm. Total exposure to 17 workers: 555 nu.

05- -82H Miscellaneous maintenance In CTS cell durlng April and May. Total exposure to 33 workers: 1,605 mr.

05- -82H "Dry alrborne waste contaminated 9 personnel and a 2500 sq. ft. area south of Tank 16 durlng removal of a recirculation jet." Improper equipment and procedure. Skln contamination up to 40,000 c/m beta-gamma. Nasal up to 700 c/m beta-gamma.

wConstruction began repair of leak...usIng dose rates to 100 mr/hr at 0.5 m.

CTS work on drop valve in Tank 36. Total exposure to 3 workers: 220 mr.

2 entrles work on drop valve tank 37 "using a dose rate of 300 mr/hr at 0.5 from edge of rlser (at tank level). Radiation level Incrased to 50 R/ hr. over open rlser when valve was removeden Total exposore to 4 men was 165 mr.

Severe right forearm laceratlon durlng work In repalr cell (299-HI. Survey revealed, "Metal plate causing Injury: 2,000 c/m less than 500 d/m alpha; cut area of plastic sult 45,000 c/m Is 50 d/m alpha...employee's arm less than 10 c/m beta-gamma, less than 10 d/m alpha; and a blood smear less than 20 c/m beta-gamma, less than 10 d/m. [The last f lgure Is apparent1y an alpha count but thls 1s not stated.]

Puncture wound on finger to maintenance mechanic. Cut area of finger contaminated to 1000 c/m beta-gamma.

Leaking transfer jet, Tank 24 mechanlc had puncture wound: 1000 c/m on exclsed. Estlmated total exposure: 240 mr.

Tank 41 grav 1ty I Ine plugged. Cather Itat Ion requlred. Tote I exposure to 4 workers: 50 mr.

Unsuccessful at-tempt to install sump jet. DB-6. Total estimated exposure for 11 people: 580 mr. "The radlatlon level increased to 2500 mr/hr over the open cell. A dose rate of 500 mr/hr was established for the job.t1

Regasketing of Tank 37 CTS spool plece. Total exposure to 14 workers: 2165 mr

Regasketing of Tank 37 CTS spool piece. Max. radlatlon level 30 R/hr. Used an exposure date rate 500 mr/hr. at 1 m. from outsfde edge of riser- Estimated total exposure to 6 people: 910 mr.

Work procedure violation. Maintenance mechanlc received facial and hair contamination of 700 c/m beta-gamma.

Entry to evaporator cell. Total exposure to 6 workers: 60 mr.

Tank 38 transfer jet adjustment. Leak check. Total exposure to 7 workers: 145 mr. Again on 07-26-82. Exposure: 95 mr to 5 workers. WORKER EXPOSURES, Page 16

DATE AND AREA DESCRIPTION AND COMMENTS

07-22-82H Removal of sump Jet from 08-6. Total exposure to 5 workers: 75 mr.

08- -82H Worker contaminated soles of personal shoes to 200 mrads/hr by stepping on a plece of contaminated plastlc on Tank 35. Monitor house floor contaminated up to 4,000 dm; sample truck cab floor to 40,000 c/m and tank top up to 200 mrad/hr.

Tank 35. Worker contamlnated personal shoes to 200 mrad/hr by stepplng on contamlnated plastic. Sample truck cab contaminated up to 40,000 c/m beta-gamma, and top of Tank 35 up to 200 mrad/hr. IThls entry appears to have been repeated In sanewhat different from on 08-20-82H.l

08-6. Sump Jet lnstallatfon. Total exposure to 4 workers: 90 mr.

Pump replacement. Total exposure to 4 workers: 140 mr.

Removal of clean out port t3. 24 persons recelved a total of 980 mr.

Decontamlnatlon of Tank 13 In prepratlon for off plant vendor work to repalr cracks. Total exposure to 19 people: 1560 mr. Contlnuatlon In Octobeer: Total exposure -to 28 people: 1435 mr.

Unsuccessful attempts to change Jumpers remotely in CTS cell uslng shielded crane. Exposure rate In cab of shielded crane 1 mr/hr compared to 2R/hr over open cell. Later hands-on change gave 1115 mr over 30 personnel entries, and further work gave estlmated 900 mr over 50 personnel entrles. TABLE 2

ENVIRONMENTAL CONTAMINATION

DATE AMD AREA DESCRIPTION AND COMMENTS

03-00-60F 1800 square yard area around diversion box contaminated with partlculates.

02-29-61 F Plutonlum leaked from evaporator to ground 4 times in "recent months."

08-1 362F Flange leak between tanks 18 and 19. Radiation rates 4 to 10 R/hr at 4 inches., 186 cu-yds of soil and asphalt excavated. [Worker exposure not cited.]

10,000 square feet area downwlnd of dlverslon box contaminated "due to Jumper changes."

Column feed pump shut off for shows. Rad1oceslum released to seepage basln.

alCeslum release - Cs columns lnadvertently bypassed for 11 hours.Ia

Radioact Ivl ty by-passed exhaust f i Iter dur 1ng sludge transfer. "Nearby I# area: contaminated.

faCesium 137. Contlnuing problems wlth high Cs1I

"Suckback through steamlfne." 60 R/hr at gang valve area. llGang valve replaced and area decontaminated."

Pump gasket ruptured. Area contaminated.

4.03 curies Cs released to seepage basin. Monthly guide of 2.5 curies exceeded, during August 30 - Sept. 5 releases.

"Dif f iculty In meeting guldesn for alpha release to seepage basins due to h1gh alpha actlvlty In Savannah Rlver Laboratory waste.

Leaklng feed pump flush Ilne caused ground contamlnatlon. 100 R/hr. on line insulation and 15 R/hr. on adJacent ground. 80,000 c/m released to Four Mlle Creek after rain.

Dlscharge of near boilfng concentrate to Tank 34. Result: flashing of vapor and discharge of airborne activity from cracks In concrete and riser covers - 20 R/hr.

Increase of Cs to Four Mlle Creek - "Surface contamlnatlon In waste tank farm-storm sewer-rafnfall??w

Well FTF7, mlnor contaminatlon has been found in groundwater of this well, new waste tanks 3 & 5 in F area, since last October, shortly after the we1 Is were drilled." Cause of contamination unknown. ENVIRONMENTAL CONTAMINATION, Page 2

DATE AND AREA DESCRIPTION AND COMMENTS

04- -74F 10 cubic yards contemlnated asphalt removed. [Source of contamlnatlon not stated. I

04-05-74F nContaminatlon. Contaminated water was found between CTS pit liner 04-23-74F and concrete encasement at 242-F. 300 mrad/hr at 2 inch." The I lquid whlch was subsequently pumped out was found to have radloactlvlty levels of 7.07 x d/m/ml cesium-137 and 7.04 x d/m/ml Cs-134.

Ru-106 contarnlnation found in monitoring we1 1 FTF-6. Pumplng continued throughout June 1974.

Top of Tank 29 contaminated.

1.0 x 10'~ mlcrocuries/cc of air released during Jumper change ln diversion box 1. Worker has nasal contamination of 8200 dis/mn beta- gamma was not wearing "respirator prote~tion.~3 others and 1 pickup also contamlnated.

242F: Exhaust alr (wairoutu) 24 hr. count of plutonlurn was 4.1 x 10'~ mlcrocur ie/cc. Flsslon products: 6 x 10" microcur le/cc dur l ng jumper removal work.

"Steam vapors and minor contarnlnation escaped through HEPA fllter on exhaust from Tanks 18 and 20.I'

Contamination escapes Tank 18-20 filters. Filters replaced. Tank 20 samples: "62,000 cm beta-gamma 3.1 x 10-8 mlcrocuries FP/ccW and "96 cm alpha, 4.8 x 10-l2 microCi Pu/cc."

"242-f. The storm sewer monitors alarmed...point 4 which monitors tanks 17-20 went off scale. Special water samples pul led were c than m.s. [less than measurablel." [Cause of alarm appears not to have been found.]

08-25-75F 500 R/hr In dry wells adjacent to Tank 3. wContamlnation within 4 ft. of riser and 208 ft. deep."

09- -75F St-90 to seepage basing above gulde.

10- -75F Asphalt near Tank 8 contaminated. Operator error.

06-1 6-76F Concentrate Transfer System. Alr Actlvl ty-f lsson products. 166 x 10" mlcrcCl/cc.

05- -77F "A HEPA filter installed downstream of Tank 7 fiberglas [sic1 fllSer after release of 137 Cs to environment.

07- -77F High activity waste evaporator "continued deter loratIon." Release of Ru-102 to seepage basin "exceeded monthly guide In June." ENVIRONMENTAL CONTAMINATIONS, Page 3

DATE AND AREA DESCRIPTION AND COMMENTS

0 1-1 6-78F Tank 7 filter leak contamlnates road and platform. Operating error - portable filter left running over the weekend.

05-1 1-78F "Filter tested 99.50% Tank 8 tea pot. HP [Health Physics1 suspected leak."

05-24-78F Wrong valve opened on evaporator. Alarm Ignored. Road contarnlnated. 250 mr/hr at 5 cms; 20,000 c/m beta-gamma at 3 cm and "less than 500 d/m alpha.1t

02-26-79F "The canyon diverted segregated on 281-8F...due to alpha contamination 11 d/m/mI."

03-06-79F 2219 canyon diverted segregated coollng water to 281-8F basln - alpha contarnlnation 24.8 d/m/ml at water monitor.

03-07-79F Contamlnat1on leaked fran plastic construction hut to ground outside. n40 square feet contaminated to 60,000 dm."

03-1 3-79F Leak In a "loop lineu during modlficatfons. Construction hut floor contamlnated to 3500 c/m beta-gamma.

03-1 6-79F olradlatlon reading(s1 in trench from D8-%..are 3R/hr. at 3 cm. south end, 1R (at 1 foot) at east end of the Ilne, and 15-30 mr. at south end of trench."

03-27-79F Tank.820 west riser radiating at 1 R/hr at 5 m.

04-02-79F HEPA f 1 lters on DB-5 leaklng. Test ef f lclency 99.25%

04-04-79F Water diverted #'due to 13.4 d/m alpha, 180 d/m beta-gamma."

04-1 8-79F "281-5 segregated coollng water diverted at 10:10 p.m. lnltlal contamination 200,000 d/m.ll Water diverted agaln at 3:30 pm. To 8F due to contamination of 20 d/m/ml beta-gamma.

05- 13-79F "Got alarm on DRB-5 high alr activity. H.P. checked and O.K. now." [Not clear whether this was an Instrument problem or leakage of radlo- activlty or both.]

Leak from concrete encased line. Sol1 samples "showed 100 c/m beta- gamma and less than 10 c/m alpha transferable ~ontamlnatlon.~

Segregated cooling water diverted to seepage basin due to beta-gamma and alpha contamlnation.

"Caved In tdepresslonl area of berm on cast side of 242-F evaporatorM had radlatlon readlng of 4 R/hr. In depresslon and 40 mr/hr at 60 cm above depression.

tt281-6 hlgh actlvlty alarm went off."...Diverted to 281-8. Counts were about 400 d/m/ml. ENVIRONMENTAL CONTAMINATION, Page 4

DATE AND AREA DESCRIPTION AND COMMENTS

0 1-29-8OF Constructfon breached tank 2 posltlve pressure and tlcaused tank air to flow from the tank 2 riser 2 openlng. No consequence occured, but the incident resulted because the job was not reviewed by HP and operations as requlred by procedure."

02-1 1-8OF "Whl le digging holes for stream1 lne poles8t the sol l was found to have actlvlty of 2,000 c/m near seepage basin line near road C.

03-03-80F DIverslon box 81 opened for a leak check. Dose rate 20 R/hr at 60 cm. After remote flushing, dose rate 5 R/hr. Job stopped "due to high wlnds that spread contamlnatlon outslde the windbreak for a dlstance of 30 f t." Cause: leaking jumper.

"Trebler line - at 15 feet below p-30 manhole. Water seeplng out berm was detected that probed 1 to 4,000 c/hr. beta-gamma."

30 cc of contamlnated Ilquld released due to operatlng errors during work on Tank 25 to remove pluggage. Redfatlion was 500 mr at Ii. cm."

Tank 18. Condensate leaked from HEPA filter at temporary exhaust system. 5 square feet area contaminated to 15,000 c/m beta-gamma.

"Water leaking around segregated water; some at P-3 manhole- Each probed 6,000 c/m beta-gamma.

tlDurlng excavation for new tank" dlrt under road found radioactive to 4,000 c/pm.

"Contaminated zeollte "inadvertently released."

Contaminated Zeollte agaln released out to top of Tank 27 in attempt to clear a plugged Ilne.

Filter change on tanks 18 and 20. "Max rate 30 mr/hreng 20 gallons water dralned Into plastlc pages "with small amounts escaped to asphalt.ti

"Durlng startup, about 115 grams of uranlum fromk the uranium cycle were lost to OW waste stream when mlxer-seliler D was operated 20 minutes with

f a1 led impel lers. [Also see, DPSPU-80 272-238; Sl-80-12-153. )'I

"A valvlng error in JB-line resulted In a transfer of 494 grams Pu Into canyon tank 9.7. This solution was subsequently fed to the Law [low activity waste1 evaporator and dlscardedOn

3 gallons of flush water and zeollte contamlnated platform and tank top. l~Decontminatlonwll 1 require about 10 man-days."

"Segregated C.W. from 22-F is still diverted to 281-8F.81 Readings inlet: 209 d/m/ml; outlet: 227 d/m/ml.

Accidental f Ire In a "waste box stored In the wlndbreakev Box radiated 5 mrads/hr and ashes smeared 10,000 beta-gamrna.gt ENVIRONMENTAL CONTAMINATIONS, Page 5

DATE AND AREA DESCRIPTION AND COMMENTS

06- -81F Release of Sr-90 and Ru-106 "exceeded the prorated monthly guidew In April 200-F. Release of other beta-gamma also exceeded monthly release guide in April. Several causes.

06-08-81 F "Pumping from F-area retention basin to seepage basin was discontlnued due to high seepage basin level. AcId has been added to seepage basln to aid In seepage process.

High activity in circulated coollng water at 281-4F (561 d/m/ml beta-gamma). Evaporator 1 shut down.

1.3 milllon gallons cooling water collected In 8F basln. Samples "analyzed less than 50 d/m/ml. Pumping from 8F to seepage basins In progress. Seepage basins are ninety-two percent fil led."

Segregated cooling water with up to 700 d/m/ml diverted to 281-8F from canyon.

"DO? Test results - Tanks 18 exhaust greater than 99.97% effIclent. Tanks 17 and 19 exhaust 99.90% efficient." Cause of low efficiency at 17 and 19: inleakage of air at blower (06-25-81 entry). Increased efficiency to 99.60% by 06-29-81. Increased to 99.8% by 07-07-81.

"Buldllng 221-F clrculatlng cooling water was dlverted once and the segregated cooling water dlverted four tlmas to the 281-8F retention basln due to contamination from conyon equlpment during July.*' Levels of radioactivity in samples from 30 d/m/ml to 150 d/m/ml.

"2816F was dlverted to 281-8F. ..281-6F samples 182 to 362 d/m/ml."

"Segregated cooling water was again diverted to 281-F on August 7, 12, 17, 19, 21, and 27. IConiamination to 104 d/m/ml.lw

Canyon diverted water to 281-8F. Sample readlng 2150 d/m/ beta-gamma.

"241-F-T8T cmpleted removal of contaminated asphalt in front of Tank No. 26 and by west side of Tank No. 27.l'

Release of several beta-gamma emitters including Sr-90 and Cs-137 and alpha emltters exceeded monthly gulde. "Source of release was 211-F bui lding segregated cool ing water contamlnated by leeks."

Low level waste trailer leaked liquid "probing 8000 c/m to 10,000 c/m beta-gamma. 30 gallons of low level waste backed...out the roadway and Into an adjacent drainage ditch."

"Releases of Ru-103 and Ru-106 to F area seepage during July exceeded monthly guide." Cause: tube leak in building 221-F evaporator re boiler. Releases for other beta-gamma and alpha emitters also exceeded guide. Same cause.

"The combined 200-F and H releases to Four MI le Creak of Sr-89, Sr-90, Cs-134 CS-137 and other beta-gamma em1 tters exceeded the monthly and annual gul des 1n September.'* ENVIRONMENTAL CONTAMINATIONSI Page 6

DATE AND AREA DESCRIPTION AND CWNTS

11-21-81F RadIoactIvIty release to air saturated HEPA fllters. AIr fllter counter 20,000 c/m.

Monthly release guides again exceeded for several beta-gamma ~~~~~s In October. Annual guldes agafn exceeded for several beta-gamma emltters In October. Annual guldes for several radIonuclldes "had already been exceeded.

"5000 gals chrmated water was lost from system and diverted to the retentIon bas1n."

ttCesium removal column hopper overflowed. "Less than 500 d/m alpha smearable and less than 1000 c/m beta-gamma smearable."

RadIatIon from riser opening 100 mr/hr. Rainwater leaking Into annulus.

"Four transfers totalling 36,000 gallons from catch tank to Tank 7" In January 82. ContamInatIon levels from 6.05 x d/m/ml (last transfer) 2.48 x 10'~d/m/ml ( f lrst transfer). Heavy raln caused water leakage Into transfer line encasement.

Monthly release guide for several rad1onuclIdes exceeded In November 1981.

"7500 gal lons of water contaminated to 1.5 x d/m/ml beta-gamma was transferred from F-area catch tank to Tank 7." Cause: "unusually heavy rains."

Monthly releases fm a number of beta-gamma emitters exceeded release guides to seepage basins and to Four-Mi le Creak in December 81. They "had already exceeded the annual guide."

"Leak In No. 3 valve house. Smears showed 3500 c/m beta-gamma In draln.lt

Radioactive l lqutd ftcams out of gang va lvew accidental ly wh1 le f l ushing. wRadIatlon 5 mrad/5 mr and floor under gang valve where It leaked was 15,000 beta-gamma."

87,000 gallons of segregated coollng water diverted In February. [Radio- activity levels not c1ted.l

Tank #7 HEPA f Il ter tested 99.5% eff IcIent.

Fallure of coolIng coll on neutralization tank 12.1. Tank vessel replaced. 1.63 mill Ion gallons dlverted. ".08 total beta-gamma CI."

"Canyon is dralning 281-5F basin to 281-8F basin. Total gal lons to be drained - 380,000. Contamlnatlon was 164 d/m/ml."

281-8F retention basln, 530 gallons of sl Ightly contaminated water. 5 d/m/ml beta-gamma, was pumped to Four Mile Creek."

"Canyon diverted segregated cool Ing water to 281-8F.. .2,000 d/m/ml alpha." ENVIRONMENTAL CONTAMINATIONS, Page 7

DATE AND AREA DESCRIPTION AND COMMENTS

04-2842% Tank tY17 HEPA f Il ter tested on ly 99.6% ef f icient.

05-1 7-82F Changed hose from interarea pump plt leaked water radlatlng 1000 mr/hr. on top of cell cover.

05-1 7-82F Regulated tools taken to truck. 9 square feet of truck bed contaminated to 0815,000 c/m...beta-gamma fixed and 125 c/m beta-gamma transferable."

06- -82F Zr-95 and Ru-103 releases to seepage basin exceeded monthly guide in April "due to decreased recycle of acid recovery unit." Zr-95 also exceeded in May for the same reason [entry date 07- -82F1.

06- -82F 1.36 mi l l ion gal Ions of cool lng water containlng about 24 mi l Iicurles transferred from the retention baslns to seepage basln "677,000 gallons of sl ightl y contaminated water transferred to creek releasing approxlmatel y 9 mil l ic~rles.~~

07- -82F 720,000 gallons rainwater transferred from retention basln to Four Mile Creek on July 13 and 15. Esttmated activity released. 4.21 mlllicuries beta-gamma and 0.82 millicurles alpha.

07- -82F Sr-89, 90 and other beta-gamma releases ln May exceeded monthly gulde to the Creek. May release of Sr-89,90 was 35 curies compared to monthly guide of 2.916 curies.

07-26-82F HEPA f ilter test results: Tank No. 1 - 99.92%; #2 - 99.35%; 13 - 99.80%; 184 greater than 99.97%; 18 - 99.93%.

08- -82F June Sr-89.90 release to creek 3.12 mCi [Monthly guide stated as 2.916 -curies In previous (July) entry and 2.916 millicuries in thls entry.1 09- -82F 241-F, Tank #33. "Breakthrough of the tank purge exhaust system HEPA f 1lter caused an atmospheric release of an estimated 325 microcur les of Cs-137." Filters replaced.

11- -82F P-30 A manhole plugged and overflowed whl le pumplng salt and sand from 281-8F retentlon basin. 2000 sq. ft. contamlnated from 4,000 c/m to 15,000 c/m. "Contaminated soil was removed and the area returned to norma I on 11 /4/82.It

11- -82F 200 sq. ft. aspha It and sol l to a depth of 4 to 6 1nches removed - contaminated by leak. I8Radiation level reduced ferom 1.5 R/hr to 200 mr/hr at 5 cm.'l Body exposure dose rates to 1 R/hr. Total worker exposure to 27 workers estimated at 940 mr.

11 -03-82F 2000 sq. ft. ground area contarnlnated from 4,000 c/m to 5,000 c/m. Process sewer plugged from silt and mud being removed from 281-81: retentlon basin. "The potential existed for a larger quantlty of radioisotopes to be released to area if the overflow had not been detected."

- - -- F AREA LISTING COMPLETE -ENVIRONMENTAL CONTAMINATIONS, Page 8

DATE AND AREA DESCRIPTION AND COMMENTS

H AREA LISTING BEGINS

04-00-58H 1t241-H, Tank 9. Concern for several airborne contamination If dehumldifica- tion system Is not equipped with filtering device and a more annulus flushing.

01-00-59H "Ground water leak contamlnates diversio~.oox. Groundwater leak Into catch tank at 300 gallons a day since 11/57."

05- -6OH "H-canyon jumpers shipped to burial ground. 5 R/hr at 1 ft. from box. Maximum contamination 60,000 c/m beta-gamma."

10-00-60H "Hlgh actlvlty waste was detected In water dlscharged from a we1 l around Tank 16."

03-00-66H "About 4 curies of Cs-137 was released to segregated water whlle unplugging waste evaporator bottoms dlscharge line." "Sent flow to 281-3 retention basin for 7 hours. There was no release to 4 Mile Creek."

"Leak - envlronmental contamlnat1on. Llquld backed up the water flush line and leaked out ground and pavement near the backflush riser at Tank 24."

Leak of high level was-te due to rupture of flexible pipe. "50 gallons of slurry contamlnated approxlmately 1000 square feet of the ground surface and eq~lpment.~~Ground contamlnated to 200 mrad/hr. at 6 Inches was temporarily covered with earth and sprayed with asphalt to lmmobolize the actlvlty. 2 workers had "slight nasal contaminatlonf~but bloassay was flnegati~e.~

*Released waste. During repairs to an evaporator gang valve a leak caused extensive c~ntaminatlon.~No further lnformatlon.

NContamlnatIon of ground arean during resln removal from ceslum removal column. 1 gallon contamlnated water fell on ground.

llAn estimated 100-200 gal Ions of hlgh l y radloactlve l iquid waste contai ning 1500-2000 C1 (90% Cs-1371, overf lowed from rlser 6 of waste tank 9 in 241-H. Crystal lzed salts plugged She 2 foot diameter rlser causing waste to overf low the rfser. The waste f lowed aCross the ground fol lowing the grade to the lip of an open storm sewer." 1200 sq. ft. of earth and asphalt had "radiation Intenslties to 100 R/hr at 1 ft. The storm sewer effluent was Impounded wlthln a few hours by constructing a dam near the sewer outfall. The Impounded water was pumped to 281-3H retention basin (3 cl) and the seepage basln (4 Cl). The storm sewer was subsequently flushed wlth clean water. Flow of water was discontinued through the most highly contamlnated por-klon of natural stream bed at the sewer outfall and downstream of the temporary impoundment dam. Some of the waste escaped Into Four Mile Creek and as of May 29, 16.62, 9.32 and 0.47 curies had passed sample polnts at Road 4, C and A respe~tlvely.~No release detected In Savannah River. Ground covered wlth earth to ~ImmobolIzeradiation." About 150 cubic yards containing an estimated 1200 curles taken and burled In burlal ground. "Rainwater caused overflow of small dam at sewer." NOTE: Uncertalnty about quantity of leakage aid escaped activity.

"Cs releases f tom the 5/67 tank 9 lncldent In the Four Mile Creek are tabulated as of 7/24/67 - see DPSP 67-1-7, on page 405." ENVIRONMENTAL CONTAMINATIONSI Page 9

DATE AND AREA DESCRIPTION AND COMMENTS

07-24-67H wCeslum released to Four Mile Creek as of July 24 was measured as follows. 26.7, 16.1 and 0.9 total curies passing sample point at Roads 4, C and A respecti~ely.~

02-05-69H ttAbout 1 C1" Cs-137 released to ground near tank 9. Estimated 0.5 Ci flowed into Four Mile Creek. 6 cu. yds. of earth and asphalt containing 0.5 Cl removed to burial ground. Total worker exposure was 5.8 R.

12-22-69H "Poor performancew of zeolite column is causing Cs-137 to be released to seepage basin. [Quantity and perlod not cited.]

01-00-71H Radlation from tank 32 vent continues to increase. Magnitude not cited.

12- -71H 0.2 m CI to storm sewer. 200 sq. ft. asphalt l@contaminated1000 c/m beta-gamma at 1 inch. 100 gallons process liquid overflowed from E.P. 4 and 5 overheads." Further overflow recorded 12-09-71. Samples from sewer outfall were 20 d/m/ml.

01- -72H Tank 29 HEPA fllter weffIcfency less than the requlred 99.9%.

08-00-73H nContaminated water from water addition system under tank 16 pumped to seepage basin."

09-28-73H "An ertimated 2 ga l Ions of conam1nated Ilqu I d spi 1led.. .dur Ing attempts to remove salt blockage...between Tank 29 and the CTS system. A 5 foot by 20 foot area was contaminated to 500 mr/hr. at 3 feet above the asphalt." Rain storm fed activity Into sewer "which was diverted to the seepage basins." Estimated release to the plant streams of Cs-137 Is 20 mIcrocuries and 200 m1crocurIes to the seepage basin. "Personnel received an estimated 1.8 R exposure during cleanup operation^.^

Leak and subsequent rain contaminated 600 sq. ft. to 1 rad/hr at 2 inches of ground deposited from the top of Tank 13.

Evaporator concrete pad "hlghly contaminated and an additlonal 300 feet square ground was contaminated with low level radiati~n.~

"Trebler sample Iroutlnel off at 10 am, counted 38,348 d/m/ml beta-gamma and 96 d/m/ml alpha 1119,000 gallonsl. Total release calculated 8 curies gross beta-gamma and 0.019 curies gross alpha."

-9 "Air sample at piping from catch tank...Sample calclulationl 48.5 x 10 mlcrocur Ies Pu/cc and 11.044 x 10-I microcur les Pu/cc.

Tank 29 water valve contarnlnated - 1'3000 mrad/hr at 2 Inch beta-gamma. Valve connection radiating 40 rads/lOR/hr. at 3 Inches...AIr sample taken downwind at approxlmatel y 10 feet; 2-6 x 10'1° mlcroC1 FP/cc and 64 x lo-'* mIcrocurles Pu/cc of air..."

Tank 23 dip tube left In a trailer overnight before shipment to burlal ground. Contaminated shoes and trousers of 6 workers and 35 sq. ft. of asphalt. ENVIRONMENTAL CONTAMINATIONS, Page 10

DATE AND AREA DESCRIPTION AND MNTS

08-28-74H Llquld spllled on hut floor during reel tape change. Tanks 9 and 12. Radlatlon to 25 rads/hr at 2 Inches.

10- -74H Monthly release of Cs-134 to streams exceeded guide of 8.33 mcl for Sept. due to "runoff water from waste tank farms.Ig Ilndlcated a hlgh level of contamlnatlon of the sol1 In the waste tank areas.]

"Transfer between Tank 29 to 21 gassed out today. Steam vlslble... ground area at fllter radlatlng 120 mrad/hr. C. at 1 Inch..."

About 130 gallons of llquld waste "generated In RBOFN containing 1.9 micro cur Ies/gal Ion Cs-137 sent to seepage bas1 n. Total radloactlvlty discharged about 0.6 C1.

907-4H and Water monitor alarmed. "Water samples lndlcated up to 99 d/m/ml beta-gamma."

wSightglass on cesium removal column froze and burst.t1 Four gallons evapIorator1 overheads spllled on top of Tank 9. 'ILess than 2,500 d/m gamma." No further detalls are glven.

"Tank 31 and 32 - the following fllters were leaklng: Tank No. 31 lSA" fllter on condensate exhaust (99.92%). Tank No. 32 pumf 1 lter (inlet) (95.00%) ." t95% eff lclency have meant large releases of radloactlvlty Into the alr. No f lgures cited. 1 Efflclency stated at 95% durlng 08-06-75 report in splte of f Ilter change on 8/1/75. Tank 16 annulus f 1lter ef f lclency 97.60%.

'907-3H monitor - water diverted." Calculated 1028 d/m/ml beta-gamma.

Leaklng process line. nContamlnatlon 1000 mrad/600mr hr. 25 R/hr at 2 1nches from recycle l in8 Jacket." Ground contaminated.

"904-486 - Sample ca lcul atlon 1067 d/m/m 1 beta-gamma F-1 1 d/m/m l alpha on routlne sample...Source of actlvlty appears to be Tank 23 materlal.n

Cs-137 contaminated water 34 gallons to 56 d/m/ml pumped from leak detection sumps of Tanks 21, 22 and 24.

Tank 11. 2 leaklng HEPA fllters replaced.

Tank 16 annulus exhaust filters leaking and found Installed backwards. Fllters changed.

Exhaust f Ilters for Tanks 29 and 31 changed? Old f 1 lters radlatlng 1-5 RH 10R at 3 Inches.

"Tanks 21 and 22 - purge Inlet fllters...were 98.506 efflclent and... for Tanks 23 and 24, 99.80% efflclent."

'CTS - llquid detector alarmed..." samples was "less 1000 c/m beta- gamma and less 500 d/m alpha." ENVIRONMENTAL CONTAMINATIONS, Page 11

DATE AND AREA DESCRIPTION AND COMMENTS

11- -75H Ladder contamlnated to 20,000 c/m flxed beta-gamma put In clean scrap and sent to salvage yard. [Apparent safety vlolatlon.1

wOverflow of 242-H concentrate pump tank to sump. 50 gallon of very dilute waste. 13 Cl of 137cs. Fallure to follow approved procedures and false llquld level read1ngs.l1

1975 annual release guide of 4 CI for Cs-137 exceeded. Total 1975 release 6.22 curles. 2.25 CI released In December 75. Cause believed to be loosenlng of sedlmant In Ilne between H area and seepage basin.

"907-2H...water dl~erted.~~Samples up to 125 d/m/ml.

907-5H water dlverted to retention basln. 666 d/m beta-gamma and 3 d/m/ml alpha In water sample.

Low efflclency for HEPA fllters on tanks 9, 12 and 16.

Tank 29. Air sample 15 feet downwind 3.05 x 10" mlcrcCl FP/cc and 75 x 10-l2 mlcroC1 Pu/cc dulrng change of demister. Slmllar levels (1.6 x lo4 and 12+ x 10'12 respectlvel y) downwind of annulus plug on tank 16. 11 x 10-l2 mlcrocurles Pulcc near Tank 16 on 06-28-76.

14 x 10-l2 mlcroC1 Pu/cc In alr during work In dlverslon box 2.

"19,500 gallons (0.01 Cll waste tank coollng water leaked under road... Clrcumferentlal break [In the plpel due to heavy loads on road."

"FI lter tested 90.00%. New constructton. Locatlon not speclf led." 96.00% on 01-04-77H.

Llne or Jet pluggage. Ground contaminated up to 1 R/hr. Ten palrs of shoes contamlneted 15,000 c/m, 3 vehlcles had contamlnatlon - Wall dw~ntamlnated.~

Llquld splll In dlverslon box 2. Dose rate Increased from 1-5 R/hr. to 30 R/hr. Alr sample 3 meters downwlnd showed 8 x lo4 mlcrcC1 FP/cc. Total worker exposure 0.R.

Purge f 1 lter on Tank 31 tested on1 y 90.00% ef f lclent.

"The 221-H clrculatlng water has been Increasing In alpha actlvlty. If It continues the C.W. wl 1 1 have to be dlverted to seepage. The 5:30 sample had 8 c/m alpha.1U

Tank 29 f llter tested 90.80%. Tank 24: 99.29%.

Tank 29 f I lter tested 99.80%. "Rep1 aced because of high radl at ion." ENVIRONMENTAL CONTAMINATIONS, Page 12

DATE AND AREA DESCRIPTION AND COMMENTS

Slmllar changes on same day on other Tank 29 and Tank 31 fllters. But poor efflclencles seen agaln on 06-15-77. (99.50 to 99.88%) and 06-20-77.

nLlquld high level waste leaked from transfer llne as waste was pumped from Tank 16 annulus to tank 14. Llne thought to be acketed was In fact unjacketed 2-Inch carbon steel with no waterproofing. Presence of Jacket never conf lrmed. No prlnts aval table. Tech. std. vlolatlon. About 400 gallons of salt solutlon and about 500 C1 Cs-137 leaked Into earthen berm over Ilne. Plpe fa1 led from corr~sion.~

Radlatlon rates to 15 rads/lOR/hr. at 5 an. durlng start of clean-up and up to 40 rads/30R/hr. after plpe and some dlrt removed. [entry of 08-10-771 Pipe and some sol 1 sl lpped to 643-6. Alr samples: 2.5 x lo-' mlcro Cl FP/cc and 7.8 x mlcroC1 Pu/cc.

221-H coollng water dlverted to 281-8H "when alpha contamlnatlon was detected (max. of 190 d/m/mll.N After 4 hours alpha was 3 to 11 d/m.

Tank 35 Intake fllter 99.60%. Tank 37 Intake f Ilter 97.50%.

Valve stuck open on waste evaporator overheads tank. 0.34 Ct of Cs-137 sent to seepage basln.

Warners pond area bushes showed contamlnatlon to 6000 c/m beta-gamma.

Water monltor3H. nBeta-gamrna 28 c/m/ml = 235 d/m/ml - 1.0 x mIcroCl/ml."

Water leaklng out of seepage basln Ilne. Extent of contemlnatlon, If any, not glven.

Tank dl3 top contamlnated up to 1R/hr durlng HEPA filter lnstallatlon. "Sol1 moved to burlal ground."

nMlsvalvlng released 2-3 Cl of Cs-137 to H seepage basln f about one-thlrd of annual guldel In one week."

"904-8G...trebler sample calculated 65,000 td/m/ml. 1.42 curie based on 128,400 gallons released." Composltlon of release glven.

Pluggage of cool lng water 1 lne to high level waste neutral lzatlon tank and cool lng col l leak. "134,000 d/m/ml beta-gamma and 443 d/m/ml alpha measured at 281-4Hmonltor house." Dlverted to 281-8H retenslon basln: "This lncldent put 1.7 ml 1 lion gal Ion of water contalnlng 30 curles In retention basln 281-8H, capaclty Is 5.2 ml1 1 Ion gal Ion. Twlce wetel' overflowed a manhole In route to the seepage. The overflow and other water dlscherged per procedure to Four MIle Creek released 11-5 MCI."

Dlverslon box overflowed while dlvertlng water to 281-1. Sol1 leadlng to dltch at Warner's pond contamlnated to 40,000 c/m beta-gamma. [NOTE: repalr work glven to contractor (data book entry dated 10-10-78). 1

281-8H: contamlnated sol1 from around this basln being sent to burlal ground. 17 truckloads to date. ENVIRONMENTAL CONTAMINATIONS, Page 13

DATE AND AREA DESCRIPTION AND COMMENTS

07-29-79H uEnvlronmental we1 I pump Is leaking and f lowlng between tanks 15 and 16. Analyzed 15 d/m/ml beta-gamma and 0 d/m alpha."

Waste leaked into plt durlng repalrs. Exposure rate event from 600 mr/hr. to 3 R/hr. Total exposure not glven.

Leak due to corroded plpes - they were Improperly fabricated. Earth contam- inated to 3000 mrad/hr at 2 Inches.

Ceslum removal column. Valve leaked and an "area approx. 10 feet long t contam1natedl to 15,000 c/m beta-gamma."

Spill over top of pan being transported to burial ground contarnlnated 2 areas, lncludlng one near bul ldlng 643-0 recelvlng area to 2000 c/mo

Waste truck mlshandl lng resulted in spl l 1 on top of tank 21 contamlnatlng ground up to 20,000 c/m beta-gamma.

Old CTS vent1 latlon system - f Ilter accidental ly separated from housing durlng removal. Plastlc under f 1lter contamlnated to 400 rnr/hr transferable.

"The new purge ventflatfon system for tanks 23 and 24 was placed In service wlthout a HEPA fllter and operated from September 5 - September 10."

"Changed alr sample for tanker [?I 23 and 24 exhaust. 4.1 x lo-'* mlcroC1 Pu/cc air, 10 ft. south of tanks 23 and 24 exhaust filter."

"Exposure rates to 2 R/hr. were encountered durlng removal of a concrete pill box from Tank 9. The sol1 underneath the pill box Is contaminated and wf 1 l be rkved at a later time." Sol l removed In 01-81. 7i5 cubIc yards contained 115 curies. Taken to burlal ground.

llCharging hopper of cesium removal column backed up overflowing onto ground and asphalt whlch were contamlnated to 40,000 c/m beta-gamma at 9 Inches."

"Water coming from tank no. 14 and runnlng across road smeared less than 1000 c/m beta game and less than 500 d/m alpha.n

Sand and asphalt around riser 4, Tank 15 contarnlnated to 5 rads/hr at 8 cm. Removed.

Leak in line durlng welding work. [Tanke 14.1 "Swfpe Indicated 30 mads/ hr. Area was covered with plastlc.1~

Tank 29 drop valve. nlOO mrad smearable, 500 mr probe at 3 Inches on the ground...600 c/m to 100 mrad. Ground is Iced and also the llqufd at tank. Employee had 200 c/m on his hand, It cleaned up. He had 1000 c/m on rlght pants knee...contamInated asphalt removed 2/22."

Tank 24 HEPA f 1lter 99.92% eff lclent.

Segregated coollng water [sample 289 d/m/ml beta-gamma1 diverted to 281-8H. [Quentfty not stated. I ENVIRONMENTAL CONTAMINATIONS, Page 14

DATE AND AREA DESCRIPTION AND COMMENTS

04-20-81 H flLlquldwas found comtng out of the alr vent to the drarroff valve Tank 31, lfquld was smearlng to 1000 nu-.

05-05-8 1 H Segregated cooling water sent to seepage basin "for 3 days...as a result of hlgh actlvlty." Cause, quantltles, sample measurements not given.

06- -81H 0.294 C1 of CF144 released to seepage basin durlng 4/81, compared to "monthly gulde" of 0.125 Cl.

06-25-8 1 H Llquld, up to 400 mrad smearable, found on top of rlser 3, tank 13.

08-06-81 H "Tank 13 - HEPA radlatlon level remains the same the sllght 12 R/hr. at 8 cms. and 10 m/hr. at rope.1

09-01-81H Llquid leak onto rlser and tank area, Tank #13. "Maximum smearable 60,000 clm beta-gamma, less than 500 d/m alpha detected I fquld."

09-26-81 H Ground under leaklng CRC column fllter "radiates to 10,000

10-14-81H nOverflow from cesium removal column due to wear on valve no. 32, teflon seats worn due to zeollte particles. Potentla1 for serlous t.elease. 32 sq. ft. of ground and hopper contamlnated to 200 mrad/hr...l/2 gallon."

1 0-28-8 1 H 08-4 hut area: 1 sq. ft, sol1 contamlnated to 15 mradslhr.

11- -81H August releases of Ci-144 and trltlum to seepage basln exceeded month1 y 01- -82H gulde. Trltlum exceeded gulde in July as we1 I. "The year-to-date release of Cl-144 11.701 curlesl has exceeded the annual guldes of 1.5 cur1eson "Sr-90 release has exceeded.. .the annual gulde of 0.600 cur les." C1-144 increased to 2.005 C1 by November.

15,100 gal lons of contamlnated wafer to seepage basln ln 3 transfers. "One transfer of 6,500 gal lons measured 3,070 d/m/ml, exceeding the I lmlt."

Flre burned 3000 square feet bank near 241-H.

5700 gallons contaminated water, 3250 d/m/ml beta-gamma and 1 d/m/ml alpha sent to seepage basln In the month. Actlvlty exceeded release l lmlt of 1500 d/m/ml, but was "only 25 of the monthly discharge."

3.5 mllllon gallons of retentlon basin water released to creek from December 29 to January 5. "A sample measured 100 d/m/ml beta-gamma. 5.6 mlll1cur1es Is estimated to have been released durlng the transfer of +he last 50,000 gal Ions."

Top of tank 13 and 1000 square feet asphalt contamloated to 10,000 c/m beta-gamma smearable. Leak due to possl ble crack between pl l 1 box and encasement.

Splll durlng removal of mlxlng Jet from tank 816. 2500 square feet contam- Inated up to 20 mrads/hr. Workers had skln and personal clothlng contamlnatlon up to 40,000 c/m beta-garma. 3 workers nasal contamlnatlon - max. was 722 d/m beta-gamma. All persons skin decontaminated uslng soap and water. ENVIRONMENTAL CONTAMINATIONS, Page 15

DATE AND AREA DESCRIPTION AND COMMENTS

05-1 2-84H olDecontamfnation of ground area continues around tanks 15 and 16H. Roadway 06- -82H of tanks decontaminated from 30,000 c/m to 2,400 dm." Work completed In June.

06- -82H Total of 10.1 milllon gallons of mud and water transferred from retentlon basln to seepage basin over 2 months. "No annual release guides were exceeded as a result of the cleaning." One transfer, 662,000 gallons also made to creek. 81Activity...was less than mlnlmum sensltlvity of the monitor^.'^

06- -82H Transfer of about 12,000 gallons from catch tank to seepage basin. Activity average about 1100 c/m beta-gamma and less than 1 d/m/ml alpha. Total activity released 0.02 Cl.

06- -82H March releases to H area seepage: Sr-90: 0.150 Ci. Pu-247: 0.142 Ci.

06-1 4-82H 600-800 gallons evaporator overheads spllled to ground from seepage basin manhole. #'The release to ground was 0.5 percent of the annual gulde to plant streams from F and H areas. The contamination was actually carried to the retention basin via the storm sewer system."

11,700 gallons water contaminated to about 450 d/m/ml t0.009 CI1 released to seepage basln.

50 squre feet ground area contamlnated to 20,000 c/m beta-gamma during transfer to seepage basln.

'lBeg1nnfng In August, radioactivity up to 27,000 c/m has been detected lntermlttently In air exhausted from the annul1 of Tanks 29 and 31." Normal actlvlty Is 100-700 c/m.

Release to seepage basln in Agusut exceeded monthly guide: Sr-90: 0.190 C1 released tgulde 0.083 CII.

H area outfall-52 - soil contaminated to 12,000 c/m beta-gamma excavated to Improve drainage. TANK LEAKS AND OVERFLOWS

DATE AND AREA DESCRIPTION AND COMMENTS

04-00-61 F Tank 8 overflow. Reel tape error. Contaminatfon of 2-2 R/hr at 2" found In well on 10-9-74 possibly from this splll.

02-00-69F Tank 1 leak. [No details glven.1

08-1 2-73F Leakage of groundwater to 241-F catch tank.

03-1 5-74F Tank 18 overflow. Increased pressure caused overflow from riser. [Envtronrnental contamination not dlscussed.l

06-1 9-74F Catch tank overfill. Instrument problems.

12- -74F Miscellaneous serious problems with tanks 4, 8, 15, 29, 31 ranglnq from bulld up of salt to reel tape malfunction. "Tank 8 drove uncontrollably on 2 o~caslons.~~

04-1 7-74F 4,000 c/m In Tank 7 dehumldlflcatlon exhaust dust. Source unknown. Mat n ly Cs-137.

06-30-75F Waste solutlon sprayed as mist onto top pf rlser cover of Tank 19.

09-06-75F Tank 19 - cracks In riser contarnlnate sol1 - 500 mrads/hr.

09-06-75F Rain leaks Into annulus of Tanks I, 4, 5, 33, 54. [Frequent occurrence. 1

11- -76F 2230 gallon groundwater In leakage to catch Tank.

12-1 1-76F 67,000 gallon rain to waste tanks. Operating error durlng construction. Tank 7 above mx. operating level and 4.2 below max. fill Ilmlt.

03-08-77F Crack In tank sealer. "SSeam and condensate seeping out." Tank No. 6. 10,000 c/m beta-gamma; 4,000 d/mln. alpha.

05-29-77F Tank 7 feed pump packing leak. 1,500 mrad/200 mr/hr at 5 cm.

02-09-78F wSolutlon backup flush water tank overflow and It reads from 7 to 25 rads/hr.I1

07-25-79F Llghtnlng struck surge tank level instrument. Tank overflowed to chromate water tank. Chromate water tank full because the pump Is Inoperable."

12-24-79F "Underground water leak between Tank 2 and 4."

0 1-04-80F Condensate leaking from bottom hole at Tank 27 fllter encasement. Conden- sate drain valve cracked. Llquld reading 20 mrads/hr at 5 cms.

05-22-805: 5600 ib. of 51% nltrlc acld overflowed tank 35 and went to seepage basln. Cause: valve leak. TANK LEAKS AND OVERFLOWSD Page 2

DATE AND AREA DESCR IPT ION AND COMMENTS

01-07-81F Tank 14: Small crack found. Sealed Itself after "a very small leak occurred." wTank 14 has a hlstory of Inactive leak sltes, and this addltlonal crack does not slgn1flcantly change the tank lntegrlty."

06-20-8 1F Nozzle 6 leaking. Stopped transfer from Tank 26 to Tank 47. Extent of contamInatIon, If any, not clted.

07-2 1-82F 241-F - Tank #26. LIne leaking - "probed 45 mrad/5 mr/hr. at 5 cms and smeared 30,000 c/m beta-gamma and less than 500 d/m alpha."

08- -82F Rainwater In leakage Into annulus of tanks 40H and 47F. *'The leaklng penetratlon on tank 40 was found when a coaling water Ilne near the penetratlon ruptured."

08- -82F "The catch tank col lected.. 06,044 gal Ions of ground water that leaked 1nto the encasements. 408 gal lons were col lected In the prev Ious report per Iod.

10-02-82F High level waste requlred more sodium hydroxide for neutralIzatIon than tank 12.1 could hold. 300 lbs. of solutlon overflowed onto cell floor.

F AREA LISTING COWLETE

H AREA LISTING BEGINS

olTank 14 leaking." [No further 1nformatlon.l

"Tank 10 leaking - thls Is the thlrd apparent leak In H area waste tank farm. Tank 9 and 14 are also leaking."

"241-H leak. Waste detected In annular space of tank 16." No measurements of actlvlty In or near annulus glven.

"24131 leak. Tank 14."

"Leaking tank. Tanks 14 and 16."

"Tank 16. Waste leaking Into annulus at rate of 0.2 Inch Increa~e/hour.~

"241- leak. Hlgh actlvlty waste was detected In water discharged from a well around Tank 16." No measurements glven. Test north of tank also showed actlvlty. Also actlvlty beneath contructlons pad. [Entries 11-07-60 and 12-01-60. I

"Tank 16. Many leaks observed...ln annulus space."

"Tank 9. About 850 gal lons waste leaked lnto annulus In a week. Total about 5000 gallon^.^

"Tank 10. Leak Into annulus - sol1ds have been vlslble for 2 years." TANK LEAKS AND OVERFLOWS. Paae 3

DATE AND AREA DESCRIPTION AND COMMENTS

01-31-67H Waste tank 10 - hlgh pressure water flange leak. Leak developed "whtle llne was under 2000 Ibs. of pressure. Potentfal danger to pers~nnel.~'

01- -71H "Increase of 84 Inches noted In tank 21 leak detention sump." Max. actIv1ty u1200c/m/ml gross beta-gamma and 29 counts Cs-137. Source of actIvIty unknown."

03-00-71H Tank 21 sump leak 36 gallons/day. Cause unknown. ActIvIty about 30 c/m/ml. 04-00-71 H Pump out continues for months.

04-00-73H Tank 21 contamfnated water accumulates in b&m leak detention sump. Many tanks have this problem in the H-area.

07-00-73H ~lUnexplaInedIncrease In Tank 10 liquid level." In August sama for Tank 9.

09-00-73H llEvIdenceof leakage of groundwater In...Tank 24."

04-01 -8OH Undergroudd water leak Into annulus of Tank 11.

01- -82H New leak found In Tank 14 - winadIve.18 "About 40 other leak sites have been previously deteded In Tank 14."

06- -82H Several hundred gallons water In leakage tank 40 annulus. Inadequate seal Ing. Water carrfed sane clay Into annulus.

09-1 8-82H 4800 gallons rainwater entered tank 11H through Inadequate seals to tank rIsers...A means existed for uncontrolled water addltion to tank. Causes: heavy rain plus failure to repair cracks soon enough.

4600 gallons ralnwater Into Tank 11 through rIser #lo TABLE 4

TANK SYSTEM FAILURES AND PROELEMS

DATE AND AREA DESCRIPTION AND COMMENTS

08-00-56F Coollng water to tank 3 shut off for 2 days due to "operating error."

12-01 -64F Error shut off coollng water for llseveral hours."

01-2869F Coollng water contaminated to 80,000 cls./m/ml due to operatlng error.

03-2 1-69F Tank 7. "Half of coollng colt orlflces plugged."

12- -74F 64 leaks total In coollng colts of Tanks 1, 2, 3, 8, 9, 10, 11, and 14 - presumably for 1974.

10- -75F Waste tank coollng water contaminated due to leaks. 4500 d/m/ml of Cs-137.

09-1 9-79F Failure of refractory band on new tank (#26) before radIoactlve service. Corroslon caused by heating caused hlgh carbon steel to crack.

10-1 6-79F "Reheat box on Tank 7 drain llne Increased In radlatlon to 25 R/hr at draln Ilne." After flushlng radiation reading down to 1 R/hr.

01-1WOF Several unexplained entrles called "non-routine maintenance" on varlous tanks. 01-1 1-80F [Many slm1lar entrles thereafter a1so.l

01-21-80F Sample taken SIXmonths prior to entry date "was outside technical standard IlrnIts for the prevention of nitrate-Induced stress corrosion cracking."

04-06-80F Vent lfne on ceslum removal column radlated at 2500 mr/hr at 5 cms. "Vamp on Tank 19 alarmed."

05-1 7-80F "Flush tank L.L. flow levell 1s acting crazy, It has filled to overflow at anything frm 20 to 60%."

08- -8OF llAlpha actlvlty In process vessel vent discharge to sand fllters increased spotad1cally up to 20 tlmes normal. Alr sparge on Tank 16.1-2 prior to sampling of frame waste recovery product resulted In high alpha adl~lty.~

10-14-80F Radlatlon flled at monltor on tank 26 was 4 R/hr. Cause: llprobably leaking automatic valve."

05-26-81F Tank 2. Two shaft sectfons of a probe disengaged. One 10 ft. sectlon left In tank.

05-27-81 F 14' long 160 Ib. probe and shaft wuncoupled and fell Into waste Safety champ fa1 led.

07-01-81F 43' long dip tube (600 Ibs.) fell 6 feet to bottom of tank 19. TANI: SYSTEM FAILURES AND PROBLEMS, Page 2

DATE AND AREA DESCRIPTION AND COMMENTS

02- -82F Fission product decay heat in Tank 32 increased to 20% Iimit due to receipt of fresh waste - 1.e. to 37 x 10~Btn/hr.

05-1 3-82F Tank 38 containing 19,000 gal Ions of waste "was outside technical standards for inhibitor^.^^ tlllnhibitorsH inhibit tank corrosion from unneutralized high level waste.)

08- -82F Tank 827 cesium removal column (CRC) pluggage. Dead algae apparently present at inlet seem to be the cause. ICRC piuggage fairly frequent.]

09-1 5-82F "Unplanned transfer of Pu solution to canyon tank 9.7, due to defective valve handle installatton. 46 grams of Pu accidentally transferred. Tank 9.7 contents recycled through canyon second Pu cycle."

09-24-82F 100 grams Pu again accidnetally transferred to canyon tank 9.7 due to improper valving. Tank contents again recycled through second Pu cycle.

F AREA LISTING COMPLETE

H AREA LISTING BEGINS

02-00-58H wOutside air purged hot Jet line and got 200 mpc. Tank 9 annulus flushed and transfer Ilne was 4.5 R/hr with 1/4 Inch lead shield. Concluded they should have flushed with water sane how.@^

05-00-58H Tanks 10,11, 12 garnma radlatlon: max. measurements 306 R/hr at 1 foot, 1.5 R/ hr at 2 feet and 1240 I/hr at 1 foot respectively.

05-01-6lH wCorrosion pitting to 5 mlls observedn on tanks 21 and 22.

12-22-66H wRadiation. 5 R/hr. Fran unshielded feed Ilne from tank 23."

0 1-22-69H Tank 10 cooling coil leaking.

0 1-23-69H Tank 14 cooling coil leaking.

06-00-69H 3 cool ing coi I leaks in one month. Tank 10 on 5-30-69; Tank 14 cot I #10 on 6-1-69 and Tank 14, colt W4 on 6-15-69. Coils blanked off. lCooling coil leaks frequent.]

"241-H, Tank 9. Exterior wall of the tank 9 reel jet riser was contaminated to 30 radslhr. at 3 Inches. Jet discharge I ine inside riser had fa! led.

"Cracked refractory line of new tank. Affected tanks 29 through 32."

16 col l leaks in Tank 11 coi Is between 10-28-69 and 11-23-69 fol lowing sludge removal. "No coi 1 leaks had previous l y been observed in this tank."

11th leak In Tank 10 cooling coils. [Total no. of coils = 35.1 Twomore failed by 04-00-70. Two more failed during 06-00-70. Total number of failures by 05-00-72 was 19. TANK SYSTEM FAlLlJRES AND PROBLEMS, Page 3

DATE AND AREA DESCRIPTION AND COMMENTS

10- -72H Tank 32 temperature rise. Unexplained. Increased cooling reduced temperature but "sludge temperature remained at approximately 99O C."

Almost same entry as for 12-27-73 of 14,000 gallons waste slphoning into annulus of Tank 14. Slphon broke. ICould be a colncIdence or dupllcate entry with wrong date. Annulus alarm was Ignored for several hours.]

5000 gallons of hlgh level waste supernate containing up to 5000 Ci, Cs-137 inadvertently transferred to low level waste receiving tank.

"Tank 15 smear samples taken with wet cotton swipes on the pan floor disclosed contaminated much higher than found...In 1973." Actual values not g Iven.

Tanke 36 annulus: Air sample showed 746 c/m beta-gamma and 1202 c/m alpha. Alarm did not Indicate high activity.

Tank 16. 100 gallons of sludge at the bottom of tank.

Temperature of cooling water for Tank 16 found higher than standard due to faulty air lock and heat exchanger.

Valvlng error. Hlgh heat waste sent to wrong tank.

"Construction has completed over eighty percent of measuring and mapping of pits in Tank 38. To date the deepest pit is 0.061. Inspection of the clearing floor is revealing hundreds of shallow pits between 30 to 60 mills deep. "

Unsealed cooling coil penetration. Tank 38. This mallowed contaminated air to be drawn from the primary tank vapor space into the annulus."

Tank 32 annulus fan shut down during core drilling. "The WMO self Imposed Ilmit of 115 C on bottom temperature was exceeded due to loss of cooling air flow."

nA back-up system has been provided for annulus air exhaust from Tank 32H ...to prevent possible hardening of sludge onto the tank bottom upon interruption of cool lng ef fed of airflow through the air slots under the primary tank."

Erroneous transfer of 18,000 gallons of concentrate supernate between tanks.

Radloactlve waste accidentally sucked back into unshlelded above ground piping. Radlatlon rates were 4 R/hr at 5 cm. between tank top and manual sparger valves...

74 inch crack observed In Tank 16 primary wal I. Longest previous observed crack was 6 inches.

wUnlnhibited seal water" supplied to Tank 41 for 1 month in place of standard due to failure of automatic chemlcal addition system. EXPLOSIONS: POTENTIAL AND ACTUAL (due to bulld up of hydrogen, ammonia, organlc compounds)

DATE AND AREA DESCRIPTION AND COMfENTS

06-00-60F Hydrogen In Tank 18

04-00-61 F Hydrogen bulld-up to 95% of lower explosive llmlt - fan fallure on tank 5 and 8. [NOTE:- All percentages below refer to lower exploslve 1lrnlt.l

05-07-62F Hydrogen to 20%. Fan not started.

07-3 1-62F Hydrogen to 30%.

08-29-62F Hydrogen to 80% In Tank 4; 45% ln Tank 8; and 20% In Tank 3. To 100% fn Tank 2 durlng scheduled power outage. Operatlng error.

05-08-64F Tank #l H2 would read';00% In 2-4 dayn wlthout vent1 latlon. tComm8nt only.]

12-1 2-65F Organlc solvent ndegratfon" - fumes and smoke.

10-OM6F HHydrogen explosion." H2 to 15% In 11: hours. Vent blower off for 19 hrs. IPosslbly an erroneous entry. Intent might have been to wrlte posslble exploslon. Hydrogen explosions not mentioned In any other report.)

H2 to 5-158. "Temporary blowers belng used."

H2 to 12%. Failure to turn on hydrogen purge blower.

H2 to 10% In rlser 97 of Tank 8.

H2 to 101 In Tank 18 vapor space.

hmonla from added smnonfa nttrate evolves In Tanks 4 and 17.

140 pprn amnfa In Tank 7 alr exhaust.

Tank 8 exploslmeter readtngs, presumably hydrogen: "4 pm = 7s; 6 pm = 6%;

8 pm .I 4%; 10 pm = 3%. NO smoke. St111 smells same."

Amnonla at Tank 4 purge exhaust 1000 ppm durlng transfer of flush solution from 221-F.

Tank 8 purge had 2000 ppm ammonla 12-8 shlft.

Tank 8 purge exhaust had 1000 ppm ammonfa 8-4 shlft.

Tenk 8 purge exhaust had 5000 p@n mnla 12-8 shlft. . .. "241-F PP No. 2 and 3 purge exhaustw ...ammonle 1500 ppm. EXPLOSIONS: POTENTIAL AND ACTUAL* Page 2

DATE AND AREA DESCRIPTION AND COMMENTS

F AREA LlSTING COMPLETE

H AREA LISTING BEGINS

Hydrogen in vapor space of Tank 9 to 150%. [No forced ventilation.1

H2 bui Id-up to 40% In the alr space above Tanks 14 and 16."

Tank 14 H2 to 30%. Blower found off.

Tank 15. HZ to 10%.

"Tank 14. Leak of 25 gallon per day essentially self sealed."

"Tank 16. 26 previously undetected leaks seen through hole 42."

Tank 21, Hydrogen to 5%

"On 3-26-70 firecracker-like detonations occured durlng the removal, regasketing and reinstailatlon of the 242H evaporator...The potential problem from exploslve compounds was emphasized when Tank 21 jet pill box was entered ...several cap-pistol pops 1111th puffs of smoke occured on the sklnless steel floor under the operators rubber overshoes." Further Information in DPSP 70-1-6, pages 68-69.

Silver compounds may be responsible for detonations.

H2 to 5% in Tank 11.

"Tank 32 hydrogen bui Id-up." In leakage of air also Indicated.

H2 to 37%. Tank 11 . "Tank 15 - 4% hydrogen detected today on weekly routine."

Tank 8 - 8% HZ.

"Tank 35 - 350 ppm amm0n1a.~

"First transfer frm canyon to Tank 22 containing ammonia was started at 6:15 a.m." NH3 rose from 25 ppm to 450 ppm. "Next check due at 11 a.m." 650 ppm at 10:30 p.m. On january 2, 1978, HN3 reached more than 1000 ppm. Not thls Is an Intentional transfer. Purpose not stated. 1000 ppm on 0 1-04-78.

"...hydrogen was coming out of tank 1321 at 10% on exploslrneter scale.81 TABLE 6

EQUIPMENT PLUGGING

DATE AND AREA DESCRIPTION AND COMMENTS

04-00-61 F Evaporator llne plugged: contamlnatlon

04-1 1-6lF Evaporator line plugged - waste accumulated ln cell. I9Radlation was 5R outside cell wall."

07-06-64F Leak due to jet pluggage. Beta-gamma contaminatlon to 80,000 C/M

02-1 6-67F Evaporator bottoms lfne plugged. [Frequent occurrence.1

04-25-67F Concentrate transfer system (CTS) loop pump plugged. [Plugging equlpment such as pumps and pipes common ln thls system.1

10-1 4-68F Steam vent IIf t II ne p lugged. [Frequent problem. I

1 l-00-69F "4 major llne plugsn In one month.

07-20-78F Llft dropvalve plugged. Evaporator down 12 hours. ILlfe llne problems particualrly plugging common.1

08-1 7-78F Extensive pluggage in rec1rculatlon pump, 301 line to Tank 34 D/0 valve and Instruments.

03-1 6-79F nsomev pluggage In hlgh level waste header #l. Removed wlth heated water.

02-04-8OF Vent line plugged. [Frequent problern.1 Evaporator down due to thls and other problems.

03-04-8OF Both F-area evaporators down due to l ine pluggage and valve problems.

08-05-8 IF "Evaporator down 299.5 hours. Prlmarlly caused by plugging of both loop and vent Ilne~-.-~and some other factors.

1 l-12-8lF "CRC Is plugged." Even extensive efforts to unplug lt were not successful as per entry of 11-14-81.

F AREA LISTING ENDS

H AREA LISTING BEGINS

03-1 5-68H "242-H evaporator. Persistent

05-23-68H "Cs column. Plugglng problems."

06-1 Od8H "242-H evaporator. Bottoms Ilne plugged due to electrical power failure." EQUIPMENT PLUGGING, Page 2

DATE AND AREA DESCR IPT ION AND COMMENTS

09- -73H I1242-H Evaporator. 67.5 hours downtime due to pluggage of Tank 29 draw-off valve and tank loop Iine.I1

0 1 -07-74H During unplugging of CTS tank level instrument, the tank overflowered 2000 gallons to the sump. "Sump contents returned to the system and the cel l floor f lushed.'I

08-00-74H Unsuccessful attempt to unplug high level waste header by jetting hot 25% sodIum hydroxide.

08-23-77H Transfer Jet to Tank 31 plugged. "Could not be removed, stuck In salt. Mlning tool made of A1 drssoived when lowered into tank, liquid backed up contaminated Al tubing and contaminated hutn...Worker exposure not given.

10-1 7-8OH "Pluggage of lift line of evaporator. Attempts to catheter were unsuccessful because the lift pumper connector block was not properly designed. Pluggage was removed using causlfc and acid." TABLE 7

POWER SUPPLY FAILURES

DATE AND AREA DESCRIPTION AND COMMENTS

12-20-53F, 04-04-54F Both ccmplete and partlal fallures noted. Data on standby power 07-10-56F, 07-13-56F not glven as a rule. Most power fallures occur during summer 07-07-57F, 07-27-57F storms. Durations vary frcina few mlnutes to several hours. 07-31-59F, 07-25-60F Lfghtnlng 1s the most common problem. Some examples include: 08-08d0FD 08-14dOF 08-06-62FD 08-07-62F 06-ll-63FD 07-29-646 05-28-65F, 07-0865F 07-1 7-68F8 08-1 868F 05-25-69FD 04-05-71F 07-02-71F8 07-27-71F 08-04-71F, 08-31-81F 02-20-73F, 06- -73F 06-18-73F, 08- -73F 11- -73F, 06-17-75F 07-18-75F, 01-29-77F 09-16-77F, 07-09-78F 06-30-79F, 08-11-79F 08-12-79F, 08-27-79F

02-20-73F Outage due to operator error.

06-1 7-75F Emergency generator dld not start; then started and then failed after "several hours."

02-29-77F Emergency generator failed during routine 30 mlnute weekly test.

09-1 6-77F Diesel generator falled to start during test.

06-30-79F Electrical power from substatton durlng storm. Emergency power Iapparentlyl shorted out. No power for 10 hours in portlons of area.

08-1 1-79F Power completely out, including emergency power (briefly). Emergency supply stuck on. nEmergency feed breaker was smoking again."

09-08-8 1F Emergency generator failed to start due to fallure of breaker and compressor motor. Normal power falled due to electrical storm. Emergency power failed. Two apparently Independent reasons. Evaporator Instrumen- tatlon also lost. [Total down tlme or consequences not cited. 1

F AREA LISTING ENDS

H AREA LISTING BEGINS

07-31 -72H Total loss of power to wells 44, 45 and 48 and waste management substation 254-14: Loss of process coollng water make up to 241, 242, and 285-H "for a short perlod of time." POWER SUPPLY FAILURES, Page 2

DATE AND AREA DESCRIPTION AND COMNTS

06-1 5-75H Lfghtning. Maln power down. Emergency on. Scme emergency switch gear Instruments and controls damaged by Iightnlng.

09-03-78H Main power down due to Ilghtnlng. One emergency generator failed to start. 3 segregation valves fat led to close. Also much equipment, 1 nc 1udl ng emergency power related equ lpment, damaged. TABLE 8

PUW FAILURES

DATE AND AREA DESCRIPTION AM) COMMENTS

03-0264F Feed pump motor burn out [fairly cmonl.

02-2767F Evaporator feed pump failure. [Frequent pump faIlures.1

31- -68F Concentrate transfer system pump failure - 4th tlme In 9 months.

02-00-68F Flash tank pump "Inadvertently left off." 8 to 12 feet water accumulated in evaporator cell.

07-00-68F Tank loop pump bearing failure - 5th in 4 months. No deflnlte solutlon to problem.

10-0068F 13 day evaporator outage due to repeated pump faffures.

01- -71F Jacuzz 1: feed pump f a1 led, [frequent occurencel , motor bad leak1ng pack ing.

04- -72F Evaporator down 39 hours due for replacement of feed pump control valve. Extensive valve problems In evaporator pumplng area.

05-1 0-79F Concentrate Transfer System pump failed. Evaporator shut down.

07-1 7-79F Llghtnlng struck near west pump house I1knockIng out power to all c.w. pumpsw 2 blown fuses. Emergency power clrcuIt was open [apparently due to Ilgntnlngl. Posslbfe common made faflure.

0 1-25-80F t9Evaporator feed pump failed."

07-06-80F Tank 26 feed pump falled, 242-16F. Evaporator down 334 hours, caused by motor wlndIng short circult.

03-05-82F Tank 826 feed pump fa1 led March 5 after 1 year of operailon. New pump falled In 6 days. Both due to short circults In motor. I1dr/hr at 0.5 meter above open r iser hole. "

- -- - F AREA LISTING ENDS

H AREA LISTING BEGINS

10-00-68H Failed transfer pump belng recondltloned for use. [These pumps are radloact1ve.l

01-26-81H "Evaporatorreplaced chems pump - 40 mrad/40/hr general area.n Nature of problem and total exposure not given.

08-05-81 H Evaporator 131: down 357.5 hours nprlmarlly due to a felled CTS reclr- culatlng pump." Down again for 1 month (09-20-81 entry) due to pump failure. TABLE 9

INSTRUMENT PROBLEMS

DATE AND AREA DESCRIPTION AND COMMENTS

01-1 7d3F llExtensive troubles: due to water and 011 In lnstrument alrlInes.

03-25-65F Tank 7. "Necklace Alarm from heavy ralns. Alarm dIsc~nnected.~~

07-1 665F Instrument air lost [Frequent alr compressor problems.1

08-1 065F Entire alarm system lnoperatlve.

09-03-68F "Mass spec. and exploslmeter readings on H2 In tanks differ up to one decade.

Tank 3 reel tape assembly felled. [Very frequent problems wlth reel tapes.]

"Reel tape reads erroneously when It comes Into contact wlth salt.11

"Serious problemsn on new reel tapes for several tanks - "May hInder relIab1lIty In the future."

Erroneous annulus alarm during heavy rain occur "very frequently."

High Activity elan goes off. Apparently a false alarm. [Frequent occurence.1

"241-F South vamp Is out of order." [Frequent problems wlth Vamp. 1

On Tank 5, Vamp gave false alarm. Unplugged.

New solid state temperature recorders' performance "very poor.11

Storm sewer tadlation alarm. ActIvIty less than "MS." Frequent.

South vamp monitor out. Fa11 safe light is out 5 days.

F area evaporator start-up. Several instrument failures due to freezing.

LlghtnIng damaged clrcults 1n evaporator alarm system. Repan - 48 hours.

"Left palm on hand and shoe monitor will not respond to source." [Very frequent problem wlth these gauges.]

"Vlctoreen defective on 907-3F water monitor." [Frequent problems wlth vlctoreen.1

Reel for Tank 7 dropped. 4 shifts to replace tape and problem. No exposures glven.

"Hand and foot counter In 242-F control room repaired.@l INSTRUMENT PROBLEMS. Paae 2

DATE AND AREA DESCRIPTION AND COMMENTS

05-1 7-79F llStorm water monitor not operating."

05-3 1-79F llAmpllfers for radiation monItor...was returned to normal service."

06-20-79F "Tank 37 annulus alarm. No apparent cause, probably Instr~rnents.~

06-25-79F A number of switches on a monitor shorted to ground. Probably Ilghtnlng.

11-21-79F "Tank 28 hydrogen analyzes Is out of service."

05-29-80F "Hydrogen analyser for Tanks 44-47 need more work [repaIrlngl.n [Frequent problems with and repa1rs.needed for hydrogen analysers. Many entries beginning 1978, 1979.

08-08-80F "Instrument power Is off at 641-F [Iightnlngl."

10- -8OF Faulty measurement causes high level waste emitting 14R/hr to be put In high level waste dumpster.

12-27-80F Water In instrument airline freezing and stopping air flow. Annulus fans 1 to 3 and 33 and 34 went off 3 tlmes during the day as a result.

06-07-81F "Vamp alarmed at Tank 18 at 10:30 a.m. No unusual radiation.

F AREA LISTING ENDS

H AREA LISTING BEGINS

09-00-59H Accldental transfer of acidic waste to Tank 16 because Tank 12.1 (acid waste tank) sampler was out of order.

08-04-62H "Rain caused tank 15 annulus alarm. Cause not determined before alarm ceased."

06-27-69H Faulty reel tape. [Frequent problems with reel tapes In H area as well.1

05-25-74H Alarms Inoperable during power outage. "600 gallons of desalt-descale flush water overflowed the CTS pump tank...."

01-1 3-7M "907-4H and 907-3H monitors glvlng much trouble. Getting hi-activity alarms due to spiking victoreen."

07-1 4-77H Tank 35 and DB6. Air monitors dld not respond to beta-gamma source during test.

08-1 0-78H Vamp at Tank 37 would not alarm at 100 mr/hr. [Thls Instrument needs frequent attentlon.1 TABLE 10

MISCELLANEOUS LEAKS

DATE AND AREA DESCRIPTION AND COMMENTS 06-26-64F Tank 7-18 transfer llne leak: 40 R/hr. on unshlelded portlon.

"Contamlnatfon line leak during test...spread contamlnatlon.~

Valve leak. 60 R/hr at 1 inch on top of rlser. Tank #19.

llHeat exchanger. Leaklng badly.@'

"Custom designed valves at end of the Jumpers leaklng profusely."

Leak at valve "due to severely corroded valve plugs."

Contaminated water In leak detectton sump. tFalrly frequent.]

Chlorlde causes chipplng of concrete of 2 waste Ilners. [Chlorlde cracking caused such severe problems that pipes wlth such coatfngs had to be taken out of service.1

Steam Ilne leak. "Very badly" corroded. IStaam llne leaks frequently.]

Leaklng relief valve on Tank 34 Chromate cooling water plplng.

Leak In plplng occured In Jul y between dlverslon boxes 5 and 6 and pump plts 2 and 3.

Evaporator 242-f down 129 hr. due to leaklng gaskets and replacement of pump motor starter.

"Catch tank collected 4800 gallons of groundwater that leaked Into the concrete encasement." A jumper leak was found though It tested leak-f ree on 1/24/80.

Heavy rafnfal I flooded excavated trenches and cracked stalnless steel Ilner of concentrate transfer system pump pft. Thls llconstltuted a loss of secondary containment for the CTS pump pIt.*l

"Water dripping from steam line to evaporator in cell sump." Water reading 43,000 dm.

"Waste 1 ine carbon steel jackets have suffered corrosion and one had f Inch dlameter hole in It. Thls 1s a breach of secondary containment protection requlred for SRP operatlons. The defedlve lines were replaced. Radlatlon was reduced fran 1 R/hr at 5 cm to 100 mr/hr at 5 UII.~~

20,000 gallons of chromate cooling water leaked fran plpe serving tanks 9-16. Leak apparently began July 9 and contlnued at 30-40 gallons per hour untll repalred on July 29. [Chromate water losses are frequent.] 12,000 gal Ions lost agaln on 08-09-82 due to 360 break In plpe servlng tanks 29-32. -MISCELLANEOUS LEAKS, Page 2

DATE AND AREA DESCRIPTION AND COMMENTS

F AREA LISTING ENDS

H AREA LISTING BEGINS

05-00-70H i'Substantlal leakage...ln the 241-H heat exchanger...The five leaking tubes were plugged."

02-25-76H Leaking valve inside pill box. Radiation levels up to 20 R/hr. at 2 Inches of water flush line.

01-21-78H qiRepaIred many leaks in 3H water monitor. The cabinet handle broke off."

02-08-78H 40 yard section of process sewer line to seepage basin caved in, caustic soda passed through pipe. "No release of radfoactive material."

10-1 7-79H Leak In or near preheater. Concrete pad under preheater radiated to 50 R/hr. TABLE 11

MISCELLANEOUS

TYPE OF OCCURRENCE DATE AND AREA DESCRIPTION AND COMMENTS

GRASS FlRE 06-29-59F lvWeldlng sparks caused dry grass to burn. Bucket brigade."

GRASS FIRE 09-08-60F Flre due to weldlng sparks. Dry chemical used to put it out.

VALVE FA lLURES 11-03-61F Solenoid valve failure admitted compressed alr into evaporator, emptylng it.

VALVE PROBLEM 12-0064F Valvlng error not dlscovered for 12 hours.

WASTE UNACCOUNTED 08-1 8-65F 8,000 gallons of waste were unaccounted for in 5 days. FOR

ELECTRICAL SHORT 11 -28-66F Cell spray of radloactlve IIquId shorted out electrical CIRCUIT equ Ipment.

SALT LOAD 11 -00-67F Concern about salt load on coollng colls. "Thought to be safe."

ZEOLITE LOSS 08-25-69F Zeolite from Cs removal column lost in "grossw quantity.

PLUGGED SMR 07-00-71 F Could cause sewer water to flush into some tanks.

SNOW 02-1 2-73F "All operations shut down. Snow."

STEAM OUTAGE 01- -74F Area steam outage - 14.5 hrs. for evaporator.

GENERAL EVAPORATOR 03-0 1-74F 242-F evaporator down 70% of the month. Pump and Instrument fallures improper operatlon durlng operating perlod wlth hlgh adivlty in the overhead stream.

VACUUM PUMP lNLET 07-25-751: Irregu Iar ma Intenance procedure. CONTAMINATED

CONCENTRATE TRANS- 08-1 4-75F "Vessel radlated a maxlmurn of 320 R/hr at ~lde.~ FER SYSTEM VESSEL FAILURE

TANK ROOF LOAD1 NG 02-1 0-Z7F 18,000 Ib. concrete cask placed on top of tank 3 w 1thout check 1ng 1f Ioadl ng acceptable. l1lnadequate canmuni~atlons.~~No collapse.

FIRE ON CON- 03-1 2-77F Flre near tank constructlon. Probably cause: STRUCTION SITE cigarette.

WASTE UNACCOUNTED 04-1 2-77F 900 gallons waste from conyon unaccounted for due FOR to vartous problems. Transfer of waste from "canyon" to Tank Farm. M l SCELLANEOUS, Page 2

TYPE OF OCCURRENCE DATE AND AREA DESCRIPTION AND COMMENTS

GANG VALVE PROBLEMS 10-03-77F [Frequent.]

HIGH WIND 01-25-78F Earth collapse durlng excavatlon exposed 25' sectlon of concentrate transfer plpe. CTS llne sagged 6" and broke a 2" dmestlc water Ilne.

EXCAVATION ACCIDENT 10-17-78F Breathlng alr l lne broken by heavy equipment durlng excavatlon.

EXCAVATION ACCIDENT 10-24-78F 1" water llne broken by excavating equipment.

CONSTRUCTION 11 -22-78F "Constructlon truck dr Iver backed over probe plpe ACCIDENT to the Jacket South of tank 33."

EXCAVATION ACCIDENT 11-27-78F Excavatlon accidentally unearthed a 20' sectlon of concentrate transfer Ilne. "Soil under CTS collapsed al lowlng the CTS to sag about 10 Inches. Sl-78-12-138."

BURNED OUT MOTOR 12-28-78F Burned out motor on gate valve to creek removed but not replaced for 10 days.

INADEQUATE WASTE 03-1 5-79F Hlgh actlvlty waste sent to tanks wlthout adequate NEUTRALIZATION neutralfzatlon, In vlolatlon of the technical standards.

ASPHALT CAVING IN 08-06-79F llAsphalt Is cavlng In at LDB 122 Tank 25."

PLASTIC COVERING 1 1-02-79F "Construction plastlc covering over the end of core CAVES plpe gave wayn al lowlng ralnwater to flood several leak detectlon boxes.

SHOWER FLOODS 12-0 1-79F "Somebody left the shower running In the ladles change BUILDING room In the new bullding and flooded the bulldlng."

EVAPORATOR FAILURE 08-13-80F 242-F. evaporator #I down. Evaporator put Into service In 1969. Cause: Bundle tube failure. "There have now been 4 evaporator failures. 2 were In 242-F and 2 In 242-H. Servlce life has been 7-11 years between failures."

SUCK BACK TO Improper alr blow leads to suckback of contamfnated GANG VALVE solutlon lnto gang valve discharge plplng. *lRadlatlon was 10 rads/lO R/hr. at 3 cms. on lower lance llne In gang valve house."

CWUTER PROBLEMS 02- -81F s9WastePHA out of servlce for several weeks wlth an 03- -81F Inoperable floppy disk unit." nApproxlmately 6 hrs. downllne experienced on the sample PHA."

CONTAMINATED 07-08-82F Rear wheels of trailer loaded with hlgh level waste WHEELS fell off. Radiation dose rate 10 rmems/hr at 3 feet from end of tank. MISCELLANEOUS, Page 3

TYPE OF OCCURRENCE DATE AND AREA DESCRIPTION AND COMMENTS

ACCIDENTAL CHEMICAL 07-26-82F "Engineer Inadvertently discharged the halon fire suppressor system In the F area No. 2 evaporator control room." 4 people had to evacuate the building.

F AREA LISTING ENDS

H AREA LISTING BEGINS

INLEAKAGE OF 05-00-65H Pump pit #I - 10 feet of water; pump tank had 4f feet RA INWATER of water. Leakage of rainwater into operating areas was frequent.

PROBLEMS NOT 08-24-65H 'lPrior to 8-24-65 Information on instrument failure, RECORDED pump failure, leaks In the waste tank system are not recorded unless the indlvidual occurrence fslcl of partfcular interest.

NEW PIT 02-1 3-80H "New CTS pIt - durlng a1r blowlng of Ilne, nozzle 624 CONTAMINATION leaked water contamlnatIng wall to 1500 c/m and the floor to 3000 c/m beta-gamma."

FAILED VALVE 07-09-80H Failed dlaphram on CIS valve. Radiation rate: 12 rad/ DI APHRAM 2R/hr. "flush with open riser."

CONTAMINATED MOTOR 08-05-80H "Evaporator burped at about 9:00 pm, setting off Vamps on Tank 21 and CRC pump. Tank 5,500 mr/hr at 8 cms. and CRC tank 4,500 mr/hr at 8 cms."

CERIUM WINDOW 0 1-04-82H "15 psfg argon gas was applied to the annulus of the SHATTERED repalr cell shIelded window t299-Hl with the annulus vent valve closed. The pressure shattered a cerium glass section of the window."

VITRIFIED CLAY 03-1 2-82H "The vltrifled clay pipe seepage basfn line In the LINE H-area is being replaced with a new polyetheylene line."

MOTOR ROTATING 04-29-82H Reclrculatlng pump motor rotating backwards. Swltched BACKWARDS electrical connections.

WASTE BOX FIRE 06-24-82H "A waste box and Its contents were Ignited after cunlng In contact wlth hot metal slag in the 22-H warm sample isle. An explosion could have occurred along with the personnel injury due to the fire."