AE-221 UDC 62] .039.536 .2:539.56

CM CM

Swedish Work on Brittle-Fracture Problems in Nuclear Reactor Pressure Vessels

M. Grounes

AKTIEBOLAGET ATOMENERGI

STOCKHOLM, SWEDEN 1966

AE-221

SWEDISH WORK ON BRITTLE-FRACTURE PROBLEMS IN NUCLEAR REACTOR PRESSURE VESSELS

M Grounes

ABSTRACT

After a short review of the part of the Swedish nuclear energy program that is of interest in this context the Swedish reactor pressure vessels and the reasoning behind the choice of materials are surveyed. Problems and desirable aims for future reactors are discussed. Much work is now being done on new types of pressure vessel steels with high strength, low transition temperature and good corrosion resist­ ance. These steels are of the martensitic austenitic type Bofors 2RMO (13 % Cr, 6 % Ni, 1. 5 % Mo) and of the ferritic martensitic austenitic type Avesta 248 SV (16 % Cr, 5 % Ni, 1 % Mo). An applied philosophy for estimating the brittle -fracture tend­ ency of pressure vessels is described. As a criterion of this tendency we use the crack-propagation transition temperature, e. g. as measured by the Robertson isothermal crack-arrest test. An estimate of this tran ­ sition temperature at the end of the reactor ’s lifetime must take in­ creases due to fabrication, welding , geometry, ageing and irradiation into account. The transition temperature vs. stress curve moves to ­ wards higher temperatures during the reactor ’s lifetime. As long as this curve does not cross the reactor vessel stress vs. temperature curve the vessel is considered safe. The magnitude of the different factors influencing the final tran ­ sition temperature are discussed and data for the Marviken reactor ’s pressure vessel are presented. At the end of the reactor ’s lifetime the estimated transition temperature is 115 °C, which is below the maximum permissible value. A program for the study of strain ageing has been initiated owing to the uncertainty as to the extent of strain ageing at low strains. A study of a simple crack-arrest test, developed in Sweden, is in progress . An extensive irradiation-effects program on several steels is in progress . Results from tests on the Swedish carbon-manganese steels 2103/R3, SIS 142103 and SIS 142102, the low-alloy steels Deger- fors DE-631A, Bofors NO 345 and Fortiweld, the new steels mentioned above, and the U.S. steels ASTM A212B and ASTM A302B, are given. Results from a series of irradiations at temperatures between 60 °C and 300 °C to doses between 2 x 1 0* ® and 3x1 0*^ n/cm (> 1 MeV) show that a saturation effect is obtained at higher temperatures in some, but not all, steels. This means that, after some irradiation, further irra ­ diation causes no further change in the properties. The experiments show that similar steels, and even different heats of the same steel, may behave in an entirely different manner as a result of irradiation. Experiments showing the effects of differences in neutron flux and neutron spectrum are described. Calculations show that the effect of differences in neutron spectra between materials testing reactors and pressure vessels may be considerable.

- Printed and distributed in March 1 966 LIST OF CONTENTS

.ESS? 1. Introduction 3

2. Nuclear reactor pressure vessels in Sweden 3

3. Choice of materials in existing and future reactors 4

4. Estimates of the brittle -fracture tendency of pressure vessels 7 4. 1 General philosophy 7 4. 2 Analysis of the influence of different factors on the transition temperature 9 4.3 Design philosophy 17

5. Work on non-nuclear aspects of the brittle - fracture problems 18

6. Irradiation-effects studies 19 6.1 Materials 19 6. 2 Mechanical testing 21 6. 3 Irradiations of commercially available pressure vessel steels 22 6. 4 Irradiations of new types of pressure vessel steels under development 25 6. 5 Study of effects due to differences in heat treatments and compositions 25 6.6 Study of effects due to differences in neutron atmospheres 28 6.7 Irradiation-effects monitoring programs 29

7. Acknowledgements 29

References 31

Appendix I

Figures 1-23 1. INTRODUCTION

The brittle-fracture work described here was carried out by the Structural Materials Laboratory of Aktiebolaget Atomenergi (the Atomic Energy Company of Sweden), a company in which the majority of shares is held by the state, the remainder by private and municipal interests. About 1,500 persons are employed, of whom some 20 per cent are university graduates. The activities of the company are fi­ nanced by the state, the total grant between 1 947 and 1965 amounting to around $ 200 million. For the financial year 1965-66 funds total ­ ling $ 32 million were granted. The first Swedish industrial prototype reactor, a combined power and heat reactor station, was built at Agesta outside Stockholm and went critical in July 1963. It has been operating since 1964. Its initial thermal output was 65 MW; 10 MW electricity and 55 MW heat [l ]. The second power reactor is being built at Marviken. This will have a power of 1 40-200 MW and will be used for electricity produc ­ tion only. The station is planned to be in operation by 1969 [2 ]. These reactors use natural and low-enriched uranium dioxide fuel, respec­ tively. They are heavy water moderated and cooled and housed in steel pressure vessels. Agesta is a pressurized water reactor and Marviken will be a boiling water reactor (140 MW power) , which later on will be operated with internal superheat (200 MW power). Another future reac ­ tor project, which is a development of the Agesta type, is the PHWR (Pressurized Heavy Water Reactor) [3 ]. Recently a decision was made to build a large light-water cooled boiling water reactor at Oskarshamn , having a power of 400 MW. This reactor will be in operation in 19 70. Besides various reactors for physical research etc. the com ­ pany operates the materials testing reactor R2 at Studsvik. R2 is a 30 MW tank-in-pool reactor of U. S. origin, similar to the ORR.

2. NUCLEAR REACTOR PRESSURE VESSELS IN SWEDEN

The dimensions and design pressures of the three Swedish reac ­ tors are shown in the following table. - 4 -

Reactor Diameter Height Pressure Material Wall thickness m ' m bar mm

Agesta 4. 5 5. 9 40 2103/R3 65 Marviken 5. 2 21.5 58 DE-631 A 80 Os karshamn 5 1 7 85 *) *>

The pressure vessel temperatures are 220 °C for Agesta and 265 °C for Marviken. The neutron doses received during the reactors ’ lifetimes are estimated to be about 2x10^ n/cm (> 1 MeV) and 6x10^^ 2 n/cm (> 1 MeV) at Agesta and Marviken, respectively. In the PHWR reactor the neutron flux at the reactor vessel will be larger, about 3x10 ^ n/cm^ (> 1 MeV) during the reactor ’s lifetime.

The Agesta and Marviken pressure vessels are shown in Figures 1 and 2,

3. CHOICE OF MATERIALS IN EXISTING AND FUTURE REACTORS

At the time of the selection of the pressure vessel material for the Agesta reactor , in early 1958, not much was known regarding the possible differences in susceptibility to irradiation effects of different materials. Since the main cause for concern was the irradiation-in ­ duced increase in transition temperature it was decided that the choice of material should be regarded as a conventional brittle-fracture prob ­ lem: thus the material with the lowest transition temperature should be chosen. The best steel from this standpoint, as recognized in the Swedish pressure vessel regulations, was at that time a carbon-man ­ ganese steel designated SIS 142103 which is similar to ASTM A 21 2B. The analysis was somewhat changed and special melts of this steel, designed 2103/R3, were made according to the following specifica ­ tions . Analysis in per cent

C Si Mn P S Cr Ni Cu Co N <0.16*^0.15-0.5 <1.6*^ <0.02 <0.03 <0.1 <0.1 <0.1 <0. 020 < 0.015

Not yet decided 'If C £0.15 %, Mn £1.7 % is allowed If C <0.14 %, Mn <1.8 % is allowed - 5 -

The steel is aluminum-treated and normalized and has the fol ­ lowing mechanical properties at room temperature

Lower yield Tensile strength Elongation strength ^ kg/mm^ % kg/mm s 30 48 - 58 S 18 The impact energy should be not less, than 2.6 kgm/cm^ at -20 °C.

The special melts of this steel were made from quite pure ma ­ terial and the steel is very clean, containing very few inclusions etc. This has given a very low transition temperature, around - 90 °C. The steel is made by Degerfors Jarnverk. The wall of the Agesta vessel is made of 65 mm thick pressure vessel steel (and 5 mm stainless steel) in the cylindrical part. For the Marviken vessel the increased diameter and pressure would give a rather large wall thickness if the 2103/R3 steel were used. In the meantime development work had started in Sweden on low-alloy pres­ sure vessel steels with higher strength. The use of such steels for a nuclear reactor vessel was considered to be advantageous since the weld inspection and works hop-to-site transports would be facilitated. Irradiation and other experiments were therefore started on such steels *), viz. manganese-molybdenum steels Bofors NO 345 and Degerfors DE-631A, both similar to ASTM A302B, and molybdenum- boron steel Fortiweld. The final choice was the DE-631A steel. The specification for the DE-631A steel is as follows : Analysis: C Si Mn P S Cr Ni Cu 50.20 0.15-0.35 51.7 50.035 £0.040 50.20 <0.20 50.20 Co Mo ' N 50.050 0.40-0.55 £0.015

The steel is aluminum-treated, normalized at 875-925 °C and annealed at 550-625 °C. The mechanical properties of material that has been annealed for 4 hours at 600 °C are:

T emperature Lower yield strength Tensile strength Elongation kg/ mm^ kg/mim in 5 x D %

RT 2 42 a 56 £ 1 7 275 °C > 33

Analysis of the actual plates of these steels tested by us are shown in Appendix 1. - 6 -

The impact energy (mean value of 3 tests) at - 20 °C should be not less than 2. 8 kgm. The need for larger pressure vessels in future reactors requires steels with greater tensile strengths than the steels discussed so far. The lower the initial brittle-ductile transition temperature the greater the margin of allowable error in assessing a minimum safe working tem­ perature '. In this connection the low-alloy steels do not have very low transition temperatures , these being about - 50 °C, and it seems un­ likely that a yield stress greater than about 40 kg/mm can be achieved at 300 °C in these steels. On the other hand, significant economies could be made if the need for plating the inside surface of the pressure vessel with stainless steel could be eliminated. The search for higher strength g, lower transition temperatures and better corrosion resistance leads one to alloy steels. We are now very interested in assessing the properties of a 1 3 % Cr 6 % Ni 1.5% Mo steel (Bofors 2RMO) in the martensitic austenitic condition 1 and a 16 % Cr 5 % Ni 1 % Mo steel (Avesta 248 SV) in the ferritic martensitic austenitic condition. The tensile strength in the unirradiated condition at 300 °C is more than 70 kg/mm . Thus, al ­ though more expensive steels, their high strength and excellent tough ­ ness would make their general use for pressure vessels (as well as for thermal shields etc. ) acceptable economically and from the point of view of safety. But until sufficient knowledge is available concerning the use of these steels as pressure vessel materials, their stability and the influence of many variables must be assessed. Irradiation- effects studies , brittle-fracture work and technological investigations on these steels are now being carried out in co-operation between us and the steelworks concerned. The mechanical properties of the steels are shown in the following table. The data represent guaranteed values, actual data from commer ­ cial heats being shown below each figure. Typical impact test results are shown in Figure 3.

The concept of minimum safe working temperature is discussed in part 4.

Most of the work so far has been concerned with a similar steel, 2RM2, which does not contain mu./bdenum. - 7 -

Steel Yield strength Tensile strength Elongation type RT 2 300 °C 2 RT 2 300 °C 2 RT 2 300°C 2 kg/mm kg/mm kg/mm kg/mm kg/mm kg/mm IV in 2RMO :> 62 i n 2> 85 ;> 71 :> 15 ;> 11 (72) (66) (92) (78) (20) (17)

248 SV :> 55 ;> 55 :> 90 :> 75 a 16 a 12 (75) (65) (95) (81) (20) (15)

4. ESTIMATES OF THE BRITTLE-FRACTURE TENDENCY OF PRESSURE VESSELS

4. 1 General philosophy

The philosophy used so far in our estimates is based on a pro ­ posal by Cottrell [4] in I960 that the crack-propagation transition tem­ perature should be used in conservative estimates of a steel’s brittle- fracture tendency. This temperature is in each case obtained by the addition of increases due to irradiation, ageing etc. to the initial tran ­ sition temperature. This reasoning is based on the fact that three con ­ ditions are required to obtain brittle fracture: a sharp notch, a high stress and a low temperature. Although the attempt is made to avoid unsuitable stresses and notches during design and control, it is only the temperature that can be completely known, and estimates of the brittle -fracture tendency should be based on this parameter alone. Thus the general conclusion is that the only safeguard against brittle fracture is to maintain the vessel above the maximum temperature at which a crack can propagate. In this way a mistake in the design or control is taken into account. Fellini et al. [5,6] developed a method for analyzing the brit- tie -fracture tendency in terms of temperature and stress '. In a stress vs. temperature diagram (Figure 4) the yield strength (cr ) and tensile strength (c ) increase with decreasing temperature (when crack-free materials are tested). Since the yield strength is increased most, the

A diagram, similar to Fellini’s, was published several years earlier by Kihara et al. [7] - 8 -

curves cross at a very low temperature (e. g. - 200 C). The existence of small cracks in the test specimen will decrease the tensile strength with decreasing temperature along the curve AB in Figure 4. The tem­ perature where this curve crosses the a y vs. T curve is called the nil- ductility temperature NDT. An increase of the crack size gives a lower ­ ing of the minimum stress at which a fracture propagates. In order to prop ­ agate fracture from a crack of given size, a higher stress or lower tempera ­ ture is required. The curve ACD shows how the crack-arrest tempera ­ ture CAT varies with stress. The temperature where CAT = a is called FTE (fracture transition elastic) and denotes the highest crack-arrest temperature at a pure elastic load. The temperature where CAT = o is called FTP (fracture transition plastic) and denotes the temperature above which only shear fracture occurs. At temperatures below NDT the stress must be lower than the low value a . if safety against brittle fracture is to be maintained. The apperanee of this type of diagram for irradiated steel is not known in detail and has not been discussed in the literature. Some as ­ pects have been discussed by Mogard [8], It is well known that irradia ­ tion increases the yield and tensile strengths, the former being increased most. Results of tensile tests on irradiated pressure vessel steels per­ formed at varying temperatures (including very low temperatures) have not been published to any great extent. Chow et al. [9 ] published such yield strength data for ASTM A212B steel, and their data have been used in Figure 4 which shows an assumed diagram for a steel in the unirradiated and irradiated conditions. It is appreciated that conventional irradiation ex ­ periments .ageing studies etc.on tensile and impact specimens can give the approximate value for NDT and the yield and tensile strength, but the stress a . and the CAT curve are not known. mm The irradiation-induced increase in NDT has been measured simul­ taneously with impact tests and drop-weight tests at NRL [10] and is ap ­ proximately equal in both cases. A similar experiment to compare in­ creases measured by impact tests and Robertson tests has been suggested by Nichols et al. [11] and such tests with the Marviken reactor ’s pressure vessel material are being carried out in co-operation between the UKAEA and AB Atomenergi. - 9 -

Lately investigators in Great Britain have suggested a new approach to these problems , based on experiments with large pressure vessels having different artificial cracks .

4. 2 Analysis of the influence of different factors on the transition temperature

Cottrell’s original suggestion has been worked on by an AB Atomenergi group of metallurgists and designers, and an analysis of the factors of interest in this connection gave the equations :

t , = tr. + At + At. end fin age irr where t,. = t , + At, , + At , + At fin pi fabr weld geom

and t = t + At pi cv prop

These equations give

t , = t + At + At, , + At , , + At + At + At. end cv prop fabr weld geom age irr where t , = transition temperature of the pressure vessel at the en end of its lifetime

t transition temperature of the finished unused pressure fin vessel

t , = crack-propagation transition temperature of the flat ™ heat-treated plate before fabrication

t = transition temperature of the flat heat-treated plate as measured by Charpy V-notch impact tests

At = increase in transition temperature corresponding to Pr°P the difference between the crack-propagation transi ­ tion temperature (e. g. as measured with Robertson tests) and the Charpy V-notch transition temperature

Atfabr - increase in transition temperature due to fabrication (bending etc. and heat treatment in connection with fabrication but not welding) - 10 -

At . , = increase in transition temperature due to welding We (heat-affected zones)

At = increase in transition temperature due to differences geom -n geome try between specimens and pressure vessel

At = increase in transition temperature due to ageing during a^e the reactor ’s lifetime

^irr = increase in transition temperature due to neutron irra ­ diation during the reactor ’s lifetime.

Thus t :> t , sop end whe re t = lowest safe reactor vessel temperature at the end of the s op reactor ’s lifetime

It might also be possible to decide on a value

t < t pop sop where tpop = lowest permissible reactor vessel temperature.

Aspects of the choice of t will not be discussed here. pop The various quantities in the equation for t ^ will be discussed in the remaining part of this chapter and numerical examples will be given which apply to the Marviken reactor ’s pressure vessel. It should be noted that we have assumed that none of the quan ­ tities At influence one another. In most cases there is no experimental evidence for this assumption. The question of how irradiation influences At . that is if the difference between transition temperatures as meas - prop ured by different testing methods is the same before and after irradiation has been discussed above. In one of our experiments we showed that At e and At^rr were independent of each other , thus aged material was changed as much as unaged material due to irradiation [12],

Transition temperature of plane plate (t )

The demands of crack-arrest tests , e, g . the Robertson test, make these methods unsuitable for the comparison of the properties of many different steels , particularly in connection with irradiation-effects studies. - 11

When one wishes to place steels in order of increasing brittle-ductile transition temperature, arbitratily defined, one may use an impact test; this need not require the use of ordinary Charpy bars, but miniature specimens as used in part of the present study can be equally satisfactory. The value of t is easily obtainable from the manufacturer or through tests. It should be noted that the impact test is a techno ­ logical test method in which no physically well-defined property is measured. The transition curve depends on the test specimen, test­ ing machine and testing technique. The transition temperature is evalu ­ ated from the transition curve according to some arbitrary standar ­ dized method.

Difference between t_^_ and crack-propagation transition temperature

prop In impact tests a mixture of factors influencing crack-initia ­ tion and crack-propagation tests are measured. Thus the crack-propa ­ gation transition temperature should always be higher than t . The crack-propagation transition temperature is often measured by iso ­ thermal Roberts on tests [13,14], Such tests have been performed for us by the UKAEA Reactor Materials Laboratory, Culcheth, on two different heats of the DE-631A steel. The results are shown in Figure 5, which is discussed further below. At a stress of 20 kg/mm the CAT is 25 °C.

Increase in transition temperature due to fabrication, excluding welding (Atfabr )

This increase obviously depends on the fabrication process, which varies from case to case, so that it is difficult to draw gener ­ alized conclusions from past experience. As an example it can be mentioned that Holliday and No one [15] investigated an aluminum-treated carbon-manganese steel in which the increase of transition temperature due to hot-pressing, normalizing and annealing was about 20 °C; after further stress -re­ lieving in situ the total increase was about 30 °C. For the Agesta reac ­ tor the corresponding figure was about 15 °C [16], and for the Marviken reactor we have assumed that the increase will not be higher than 2O' °C. - 12 -

Increase in transition temperature due to welding (At j^)

This increase is due to changes in the he at-affected zone arid must be investigated in each case since it depends on electrode materials and welding procedures \ In the Marviken reactor ’s pressure vessel there will be no such increase.

Increase in transition temperature due to geometry (Atg eom )

This increase, if any, is the least known and the most difficult to in­ vestigate. It has not even been shown to exist, although theoretical considera ­ tions indicate that there should be such an effect. It has been assumed that the effect is negligible for the Marviken reactor ’s pressure vessel.

Increase in transition temperature due to ageing (At e)

The magnitude of this quantity is not very well known. The ageing effect may be both due to precipitation hardening and strain ageing. During long anneals of steels one can get some precipitations that have a hardening effect and also some phase changes etc. among existing precipi­ tates. Even in a slowly cooled steel there is some super-saturation of nitro ­ gen etc. which can produce such an effect. Strain ageing has attracted some interest in our discussions. In some cases the increase in transition temperature due to ageing has been investi­ gated by measurements of the properties of materials that have been strain- aged under conditions that were thought to give a maximum ageing effect, e.g. straining 10 per cent followed by ageing 1 hour at 250 °C. Such a treatment gives a change of about 40 °C in the transition temperature of 2103/R3 steel. In pressure vessels the following alternatives can occur:

... • / . *•)(•) 1. straining during hydrostatic tests (cold pressure vessel) ' 2. straining during use (hot pressure vessel) 3. no plastic straining at all

In the last case there is no strain ageing. In the first two cases, which may differ considerably, the plastic strain, if any, will be rather small.

* T The weld metal can also have a higher transition temperature than the base material; this is taken into account by making a separate analysis of t and all At for the weld metal. cv These tests are usually done with a cold pressure vessel. If necessary, however it is possible to test the vessel at higher temperatures . - 13 -

The problem of strain ageing after a very small strain has not been investigated to any great extent. One does not quite know what is happening in a pressure vessel when the strain is lower than the Laider's strain. The material will be inhomogeneously deformed, thus some parts will be deformed up to the Luder’ s strain and some parts not at all (Figure 6). The question is then how large volumes are deformed, for only in these volumes will there be an ageing effect. Since the phi­ losophy used here should preclude brittle-fracture propapagation (and not only initation), it might be argued that the existence of small em­ brittled areas of strain-aged material does not matter. For the Mar ­ viken reactor's pressure vessel it has been estimated that the increase due to ageing will be not more than 20 ° C.

Increase in transition temperature due to irradiation (At., )

Discussions regarding the irradiation effects in pressure vessels of power reactors are usually based on experiments performed in other reactors. The interpretation of such experiments involves several prob ­ lems. In general some property of a material is measured after ir­ radiation for a certain time in one reactor and one is then interested to know how long irradiation in another reactor will be needed to obtain a given change in this property. The following factors are then of interest:

1. Material 2. Test methods 3. Neutron atmospheres in the materials testing reactor 4. Neutron atmospheres in the reactor pressure vessel 5. Irradiation temperature in the material testing reactor 6. Irradiation temperature in the reactor pressure vessel 7. Damage theories

It is obvious that the material of interest should be investi­ gated. When this is not possible, great care must be exercised in extrapolations from tests on other materials, even if these are quite similar. For the moment insufficient knowledge exists of how differ­ ent factors in the composition and heat treatment etc. of steels in­ fluence their irradiation effects and considerable unexplained differ­ ences between similar materials have been observed. A conservative approach is to regard all materials not yet investigated as equal to the poorest material known. - 14 -

It is also obvious that the property of interest should be measured if this is at all possible. The problems involved in comparisons of Charpy impact tests and crack-arrest tests have been discussed above. In our irradiation-effects studies much use has been made of miniature im­ pact specimens. Their use is further discussed in part 6. The neutron atmosphere is a convenient name for the combina ­ tion of neutron flux, neutron dose and neutron spectrum. These quanti ­ ties are difficult to measure and calculate. In materials testing reactors their gradients in time and space may be considerable and measurements must always be made in connection with the actual irradiation of speci­ mens. In power reactor pressure vessels the neutron atmosphere is gen ­ erally not well known at the time when the selection of pressure vessel materials has to be made. The problems connected with threshold de­ tectors will not be discussed here; it should be noted, however, that much work remains to be done if a fairly good picture of the neutron spectrum is to be obtained from such measurements. It is also pos ­ sible to calculate the spectra to some extent. The irradiation temperature in materials testing reactors may be difficult to keep constant at a pre-determined value. The gamma heat can vary greatly in time and space and in our materials testing reactor R2 it can generate up to 0. 2 kW per Charpy impact specimen. Temperatures reported in the literature are often quite imprecise and are sometimes estimated or guessed. The influence of temperature on irradiation effects differs quite considerably between different materi ­ als; this is discussed more fully in part 6. The influence of neutron flux is not well known. Experiments made at low neutron doses and with one order of magnitude difference between fluxes indicate that there is no effect. A decrease of the dose rate could be advantageous since recovery processes are time -depend­ ent. Generally this possible influence is disregarded. The effect of neutron dose has sometimes been expressed in equations like At = K • or At = K(log 0+A), where At is the increase in transition temperature, f is the neutron dose, K, A and n are con ­ stants. These equations are discussed further in part 6. The effects of neutron spectra are not very well known and might lead to quite important differences between estimated and actual changes. - 15 -

In most cases neutron doses are reported as the number of neutrons with energy above 1 MeV; and since neutrons with lower energy can also have quite a large influence, comparisons between irradiations «•) in different spectra are difficult '. The damage models, that is models to show how the changes in properties can be calculated as a function of neutron atmospheres (in­ cluding neutron spectra), are not very well known. Although much in­ teresting work is in progress in different laboratories, there is still quite a wide gap between the more theoretical work being done and the actual application of data. Some calculations have been made in order to show how attemps to "translate" irradiation effects obtained in one spectrum into an equiv­ alent dose in another spectrum can give different results when simple damage models are compared [17]. In the first case we use a simple cascade model according to Ross in [18] which considers the overall number of vacancies formed. The second model, which corresponds to the "depleted-zone " model by Seeger [19], does not take the num­ ber of vacancies formed into account but assumes that these form clus­ ters and that only the number and not the size of the clusters is impor ­ tant. Neutron doses defined in different ways are included for compar - ison. The spectra we used were calculated and are shown in Figure 7 ' The following table shows the comparisons between these spec­ tra when the number of neutrons with energy above 1 MeV is constant and the figure for the R4-10 spectrum is 1.0 in every case.

In the Swedish D^O-moderated reactors the neutron spectrum at the power reactor pressure vessels differs considerably from that in the H^O-moderated RZ reactor.

R4-10 denotes an earlier version of the Marviken reactor with a 1 00 mm thick thermal shield. R4-5 denotes an earlier version of the Marviken reactor with a 50 mm thich thermal shield. G1 denotes the French graphite-moderated reactor G1 MTR denotes a MTR-type reactor like RZ • Fission denotes Watt’s fission spectrum 16 -

Spectrum ^

Basis for comparison R4-10 R4-5 Gl MTR Fission

Dose (E > 3 MeV) 1.0 1. 3 0. 9 1.6 1.9 Dose (E > 0. 5 MeV) 1 . 0 0. 9 0. 9 0. 7 0. 7 Dose (E > 0. 2 MeV) 1.0 0. 9 0. 8 0. 6 0. 5 Dose (E > 0. 002 MeV) 1.0 0. 8 1.1 0. 5 0. 3 Number of vacancies i. o 1. 0 0.9 0. 8 0. 7 Number of clusters 1. 0 0. 8 1. 1 0. 4 0. 2

Now, as an example, if equal neutron doses as measured with **) . ... a sulphur neutron dose monitor ' are compared in G1 and in a fission spectrum, the number of defect clusters will be about 12 times as large in Gl. If the total number of vacancies is compared, their number will be only 2.7 times as large in Gl as in the fission spectrum. Thus the "conversion factor" between these two spectra differs by a factor of about 4. 5 depending on which damage model is used. For a given flux this can mean a difference Of a factor of 4. 5 in the lifetime of the reac ­ tor vessel. Irradiation experiments with the Marviken reactor's pressure vessel material have been made at the operating temperature, 265 °C, in the DR3 reactor and an interpolation between these values , which are described in detail in part 6, gives an increase in transition temperature of about 50 °C for a dose of 6x10^ n/cm^ (> 1 MeV).

*) R4-10 denotes an earlier version of the Marviken reactor with a 100 mm thick thermal shield R4-5 denotes an earlier version of the Marviken reactor with a 50 mm thick thermal shield Gl denotes the French graphite-moderated reactor Gl MTR denotes a MTR-type reactor like R2 Fission denotes Watt's fission spectrum **) E > 3 MeV 17 -

Summary of changes in the Marviken reactor's pressure vessel

A summary of the points discussed above gives the following results:

t + At = 25 °C cv prop At. = 20 °C fabr At = 0 weld At = 0 geom At = 20 °C age At. = 50 °C irr Thus EAt = At + At i j + At + At + At. * 20 + 0 +0 + 20 + 50 — fabr weld geom age irr

= 90 °G and t , =t + At + EAt = 25 + 90 = 115 C end cv prop

4. 3 Design philosophy

In order to be able to find a practical use for the evaluation of the transition temperature in a reactor pressure vessel at the end of its lifetime, one needs a '‘design philosophy" which can indicate the highest value one should tolerate. A publication from the U.S. Navy [20] specifies that the tem­ perature should be above FTE when the pressure is more than 20 per cent of the design pressure. Porse [21 ] has described Westinghouse ’ s philosophy, loads up to 10 per cent of the yield strength being allowed at temperatures below (FTE - 110 °C). The permissible load then in­ creases to 20 per cent of the yield strength of FTE. At higher stresses the vessel should be maintained at temperatures above FTE. It should be noted that the limits mentioned are not prohibitive, they merely de­ fine a "zone of caution" within which a more detailed analysis of stresses, stress concentrations etc. is made. Mogard [8] suggested that the maximal stress in the reactor vessel (a^) should be entered as a function of temperature into the - 18 -

stress vs. temperature diagram of Pellini et al. Since the CAT curve is moved to the right in the diagram by irradiation etc. (see Figures 4 and 5), the maximal permissible change (or neutron dose or irradiation time) can be obtained from the first point of tangency of the two curves when the CAT curve is moved to the right. For the Marviken reactor the calculated membrane stress in the cylindrical part of the pressure vessel is shown in Figure 5. Stresses in other parts of the vessel are higher and it is thought that the corresponding curve giving the maxi ­ mum stress max i-n an Y part of the vessel passes through the point 13 kg/mm^ at 150 °C, as shown in Figure 5. Our estimate gave XIAt x = 90 °C for the pressure vessel steel in the Marviken reactor. A CAT curve displaced 90 °C towards higher temperatures is shown in Figure 5 and does not come very close to the vs. T curve. Thus the vessel is safe.

5. WORK ON NON-NUCLEAR ASPECTS OF THE BRITTLE FRACTURE PROBLEMS

All our earlier work was concentrated on the irradiation-effects problems . Lately some work on a few specific non-nuclear problems has been started and some works is also in progress in co-operation with the steel companies concerned. From the preceding discussion it is obvious that we have an in­ terest in crack-propagation tests. There is no Robertson tester in Sweden, and so far we have had tests on a few heats of pressure ves­ sel steels made in Great Britain by the UKAEA Reactor Materials Lab ­ oratory. Work has recently been started with another test, developed in Sweden by Noren in 1951 and called the NC-test [22,23], In this NC -test large flat tensile specimens are being used. On the edges of the specimens a brittle layer of cast iron is deposited by means of welding (Figure 8). Tensile tests are then made at varying temperatures. When the specimen is strained, sharp brittle cracks develop in the brittle layer, run through this layer and are either stop ­ ped or not when they reach the base material. At temperatures below CAT (for a given stress) the cracks should propagate and at tempera ­ tures above CAT they should be stopped. Thus for each temperature 19 -

the stress corresponding to the CAT curve is obtained as the stress when the specimen breaks (Figure 9). Tests are now in progress with #) . . several of the steels of interest to us ' in order to develop a minia ­ ture NC'-test specimen, suitable for irradiation experiments. As soon as such a specimen has been developed it will be used in our irradiation- effects studies. It was mentioned previously that strain-ageing studies are of interest to us since the increases in transition temperature due to strain ageing can form a substantial part of the total change (including irradiation effects) during a reactor ’s lifetime. Some strain-ageing studies are in progress in order to investigate the extent of changes in yield strength due to a large variety of combinations of strains , temperatures and ageing times. To some extent impact tests will also be made.

6. IRRADIATION-EFFECTS STUDIES

Most of our earlier efforts were concentrated on irradiations of tensile and impact specimens of several commercially available steels under conditions as close as possible to those of the actual pressure vessels. Later work has included irradiation of steels that are still in the development stage as pressure vessel materials. Some work has also been started on the influence of various parameters con ­ nected with the heat treatments and composition of the materials and with the irradiation conditions.

6. 1 Materials

The following steels have been irradiated:

21 03/R3, an aluminum-killed carbon-manganese steel which is a modified version of the Swedish standard steel SIS 14210 3. Specimens were made from a normalized 65 mm thick plate, clad with 5 mm austenitic stainless steel on one side. This plate was excess material from the fabrication of the pressure vessel for the Agesta reactor.

In co-operation with Professor T. M. Noren, Royal Institute of Technology, Stockholm. - 20 -

SIS 142103, an aluminum-killed carbon-manganese steel made according to the Swedish standards. Specimens were made from normalized bars with 19 mm diameter. This material has been irradiated because of its extremely low transition temperature.

SIS 142102, a silicon-killed carbon-manganese steel made ac ­ cording to the Swedish standards. Specimens were made from normalized 20 mm thick plates. This material is similar to the two steels previously mentioned but is not aluminum-killed. This material has been irradiated in order to study the effects of variations in grain size.

ASTM A212B, an aluminum-killed carbon-manganese steel made according to the U. S. standards. This material has been supplied by the U.S. Steel Corp. [24]. Specimens were made from a normalized 102 mm thick plate.

DE-631A, an aluminum-treated manganese-molybdenum low- alloy steel which is similar to ASTM A302B but has been mod ­ ified according to Sxyedish melting practice. This material is used in the Marviken reactor ’s pressure vessel. Specimens were made from a normalized and annealed 1 00 mm thick plate.

NO 345, a heat-treated manganese-molybdenum low-alloy steel. This material was under discussion in connection with the choice of materials for the Marviken reactor. Specimens were made of a 35 mm thick quenched-and-tempered plate from an early heat and also of 125 mm thick normalized and quenched-and-tempered plates .

ASTM A302B, an aluminum-killed manganese-molybdenum low- alloy steel made according to the U.S. standards . This material has been supplied by the U.S. Steel Corp. [24]. Specimens were made from a quenched and tempered 152 mm thick plate.

248 S V, a ferritic -martensitic-austenitic chromium-nickel-mo ­ lybdenum alloy steel which is under development as a pressure vessel steel for future reactors , Specimens were made of 70 mm thick plates . - 21

2RM2, a martensitic-austenitic chromium-nickel alloy steel which is also under development for future reactors . Spec­ imens were made of 22 mm rods and of cast material.

6. 2 Mechanical testing

Irradiated and unirradiated control specimens were tested on the same apparatus . Hardness, tensile and impact properties were measured. In addition to ordinary Charpy specimens we have made much use of a miniature impact specimen with the dimensions 3. 33x 3. 33x 27. 5 mm, 0. 8 mm deep notch with 45° angle and 0.11 mm bottom radius : eighteen such specimens occupy the same volume and have the same shape as a single Charpy specimen (Figure 10). These miniature specimens have been thoroughly tested and found suitable for comparing the impact properties of steels, and are extremely valuable in irradiation studies [25-28]. Brittle ductile transition tem­ peratures have been determined from the 4. 2 kgm (30 ft. lb. ) fracture energy for ordinary Charpy V-notch bars; for miniature specimens the approximately equivalent value 0.17 kgm was used [28]. Examples of transition curves as obtained with miniature im­ pact specimens are shown in Figures 11 and 12, which are discussed more fully below. These diagrams enable one to appreciate at a glance the quality of results obtained with the small specimens; scatter is no more evident than with ordinary Charpy bars and, when it occurs , is believed to illustrate more of the inherent properties of the material than any shortcomings in experimental method. An interesting aspect common to all steels studied is that irradiation produces a more ho ­ mogeneous product, the fracture energy values showing appreciably less scatter for irradiated than for unirradiated steels; especially is this true of steels having considerable scatter in the unirradiated con ­ dition. Data as shown in Figures 11 and 12 allow a clear comparison of the susceptibility of different steels to irradiation effects., Other

Since this work was started it has been decided (for other reasons than those connected with irradiation effects) that another steel, 2RMO, which has properties similar to those of 2RM2 but which contains 1. 5 % Mo in addition, should be developed as a pressure vessel steel. Irradiation-effects studies on 2RMO steel are in progress . - 22

general features common to many other steels are that, after, irradia ­ tion at temperatures above 100 °C, the fracture energy vs. tempera ­ ture curve assumes a lesser slope as irradiation dose increases. This, of course, means that the knee at the high fracture energy portion of the curve is subject to greater change in position than the foot of the curve, thus greater or lesser increments in transition temperature can be measured dependent upon the arbitrary definition of the transition temperature .

6, 3 Irradiations of commercially available pressure vessel steels

The first irradiations were made in the French graphite-moder ­ ated reactor G1 at Marcoule. The results from the first experiment have been published previously [12, 29]. The irradiation temperature was 160 - 195 C and the neutron dose 2. 5x10 n/cm (>1 MeV). A report concerning results from a second experiment (neutron dose 18 2 about 5x10 n/cm ) is in preparation. A series of thirty irradiations has been'made in the Danish heavy water moderated reactor DR3 at Riso; most of these were made at carefully controlled temperatures in the range 110 - 335 °C. Among these experiments were the first ones performed at the actual pressure vessel operating temperatures. Part of the results have been published [29]. Due to the operating conditions of the pressure vessels mentioned previously we have primarily been interested in irradiation tempera ­ tures between 200 and 300 °C and in neutron doses between 10* ^ and 10 ^ n/cm^ (>1 MeV). However, as the present results show, it is of interest to extend the work to lower irradiation temperatures and higher doses . In Figure 13 are presented Vickers hardness HV, ultimate tensile strength ct , yield strength og , total and uniform elongation e and e all measured at room temperature , and the value of the t u brittle-ductile transition temperature T^ as determined from im­ pact tests with miniature specimens, for the carbon-manganese steel 2103/R3. The different quantities are shown as a function of neutron dose for two irradiation temperatures 195 and 265 °C. Irradiation at both temperatures clearly causes significant hardening and reduction 23 -

of ductility in terms of total elongation, but it is apparent that irra ­ diation at a temperature of 265 °C produces lesser changes than ir- . radiation at 195 °C. At this latter temperature all properties change and, at the highest dose achieved, 12x10 n/cm (> 1 MeV), the curves show pronounced positive gradients. Even if the curvature decreases with dose there is no saturation although the indication is that this might arise for somewhat higher doses. After irradiation at 265 °C HV, cr^ and e^. show very nearly the same change with irradiation dose as found for the lower temperature, but significant differences are found for a g , e and T , the latter quantities clearly indicating sat ­ uration. Figure 14 shows the corresponding data for one of the two investigated heats of the low-alloy steel Fortiweld. Here we notice that all properties suffer less change due to neutron irradiation at the higher than at the lower irradiation temperature, and that again a trend to saturation of the irradiation-induced changes is observed. Miniature impact specimens of 210 3/R3 and ASTM A212B steels have been irradiated together at the same temperature 230 °C to simi­ lar values of total neutron dose, and the measured impact energy curves for these steels are shown in Figure 11. The corresponding data for DE-631A and ASTM A302B steels irradiated at 265 °C are shown in Figure 12. Charpy impact specimens of DE-631 A steel have been ir­ radiated at 265 °C (the Marviken reactor's pressure vessel tempera ­ ture) to different neutron doses. The results are shown in the follow ­ ing table and in Figure 15.

Neutron dose T ition temperature Increase in transi ­ n/cm2 (>l MeV) ?4. 2 tion temperature °C 0 - 45 5. Ox 101 8 0 45 1.4x 1019 + 15 60 3. 3 x 101 9 + 40 85

In connection with the choice of materials for the Marviken reac ­ tor's vessel, miniature impact specimens of the various low alloy steels considered originally were irradiated at 265 °C to a neutron dose of 18 2 4x10 n/cm (> 1 MeV). The resulting fracture energy vs. temperature - 24 -

curves for the irradiated and unirradiated steels are shown in Figure 16, a to f. It is clear that, with the exception of Fortiweld A, the Jow-alloy steels show good resistance to neutron irradiation and have acceptable impact properties under the said condition. Of the low-alloy steels dis­ cussed here, NO 345 has shown the least susceptibility to irradiation embrittlement, although DE-631A shows similar promise. The increase in transition temperature caused by neutron irra ­ diation at different temperatures is presented graphically for the dif­ ferent steels studied in Figure 17, a to f. The results show that certain steels irradiated at temperatures above 250 °C show saturation effects, no further change in mechanical properties arising after a certain dose is attained. As a consequence the "square root law" often mentioned in connection with the dependence of the irradiation effects on neutron dose, is not always applicable. The corresponding complete transition curves for steel 2103/R3 are shown in Figure 18. The changes in yield strength in steels 2103/R3 and Fortiweld are shown in Figures 19 and

20. Specimens from the welded joints in the Agesta reactor's pres­ sure vessel, made of 2103/R3 steel, were irradiated in the DR3 reactor at the vessel’s operating temperature, 230 °C. The following results were obtained.

Results of impact tests with Charpy specimens

Material Neutron dose Transition temp, n/cm.2 (>1 MeV) T4.Z °C

Weld metal 2. 8xl018 25 center of plate 6. 8 x 10* 8 62 1.4xl0 19 81

AT = K1 where AT is the change in transition temperature and ft is the neutron dose. K is a "constant" which depends on the material ’s properties, the irradiation temperature , the neutron dose rate and the neutron spectrum. ■ - 25

Results of impact tests with miniature specimens

Transition temperature, Tq ^ °C

Neutron dos e Weld metal Weld metal Heat-affected He at-affected n/cm^ (>1 MeV) Surface of Center of zone zone plate plate Surface of Center of plate plate

(2.8-3.1)xl018 10 44 9 - 10 (6.7-7.6)xl0 18 32 61 40 47 (1.4-1.5)xl 01 8 46 66 50 42

The results correspond to a crack-propagation transition tem­ perature not higher than about 110 °C at the end of the reactor's life­ time. This is less than the estimated permissible values .

6. 4 Irradiations of new types of pressure vessel steels under development

So far a few irradiations have been made of miniature impact specimens from plate of the Avesta 248 SV steel and from castings and rods of the Bofors 2RM2 steel. The results of irradiations at 265 °C are shown in Figure 21. The 0.17 kgm transition temperature is very 18 2 low after an irradiation of about 5x10 n/cm (> 1 MeV). A joint program for irradiation of four pressure vessel steels has been agreed upon with the Danish Atomic Energy Commission, Riso. This program includes the Swedish Degerfors DE-631 A, Bofors 2RMO and Avesta 248 SV steels and irradiations will be made at 230°C , 265 °C and 300 °C to the neutron doses 1 0* 8, 2x10*^, 5x10^ and 10 9 n/cm^ (> 1 MeV) in the DR3 reactor, Riso, and to 10^, 2x10^ , 5x10^ and 10^ n/crn.^ (> 1 MeV) in the R2 reactor , Studsvik.

6.5 Study of effects due to differences in heat treat ­ ments and compositions

Three special questions in connection with the studies of 210 3/R3 steel have been investigated, namely the effects of very low initial tran ­ sition temperature, of strain ageing and of variations in grain size. - 26 -

The very fine-grained steel SIS 142103 was irradiated in ord

that the corresponding figure for the Marviken vessel would be 30 per cent. The difference is due to significant differences in the calculated neutron spectra in the two cases. It is considered that boron fission in Fortiweld steel should not limit its use in this connection. The first test results agreed with this conclusion, since the increase in transition temperature as measured in the first experiment in G1 [29 ] was about the same in Fortiweld B as in most other steels irradiated simultaneously. Similar results were obtained from irradiations in the Danish reactor DR2 [31]. A detailed study was , however, made on a later heat, here called Fortiweld A, and irradiation was found to produce large increases in the brittle-ductile transition temperature. Results for material from this heat obtained with normal Charpy specimens are illustrated in Figure 17 f; corresponding data for miniature speci­ mens irradiated simultaneously are shown in Figure 17 e. On the other hand, Fortiweld B irradiated at 35 °C in DR2 was found to be much less susceptible to irradiation embrittlement; results obtained in DR2 with ordinary Charpy bars are also depicted in Figure 17 f. The ques­ tion arose as to whether these differing results were due to different inherent properties in the two heats or to different neutron atmospheres. To settle this point, miniature impact specimens of both qualities of Fortiweld were irradiated at 265 °C in adjacent positions in a constant- temperature rig in DR3. The results (Figure 16, f and g) prove that the two heats are inherently different. Metallographic examination showed the two steels to have significantly different microstructures , the steel which was very susceptible to embrittlement possessing prom ­ inent retained grain boundaries from the former austenite phase: these boundaries surrounded smaller grains of pearlite and bainite. If the retained former austenite boundaries are a source of weakness, then we may associate the high susceptibility of Fortiweld A to irradiation embrittlement with the large grain size, about 3. 5 in the ASTM scale. Fortiweld, of course , contains boron and, as mentioned, fission of B"*' ^ may aggravate fast neutron irradiation effects; it is sufficient to point out here that whatever effect boron may have, it should be the same in both Fortiweld A and B, unless of course boron is concentrated in the retained former austenite grain boundaries . - 28 -

If the boron is not concentrated at grain boundaries , and such a concentration has not yet demonstrated for this steel (although this is a difficult task), then the influence of boron fission is dependent pri­ marily on the ratio of thermal to fast neutron flux at the reactor pres­ sure vessel. This latter ratio may vary significantly with reactor de­ sign, and the problem must be assessed separately for each proposed application. It must be admitted, however, that the initial high tran ­ sition temperature , the influence of boron fission and the somewhat uncertain metallurgical role of boron make this a difficult steel to assess. In order to investigate this question further, material from three laboratory-melted heats containing natural boron, B ^ and B , respectively, have been irradiated simultaneously and tests are now in progress. NO 345 steel has been irradiated in the form of plates of two thicknesses, one of them after two different heat treatments. The dif­ ference between the different materials was insignificant.

6. 6 Study of effects due to differences in neutron atmospheres

The effects due to differences in neutron doses and irradiation temperatures have been discussed previously (cf. Figure 17). The influence of neutron flux (dose rate) has been investigated by two irradiations at 250 °C in the R2 reactor. One of these irradia ­ tions was made during 6 hours at full power (30 MW), while the other was made during 60 hours at 3 MW; thus the total dose was the same, but the dose rate varied by a factor of 10. Some of the results are now available and are shown in Figure 22. The effect of dose rate at this low neutron dose seems to be negligible. Tensile and impact specimens of five pressure vessel steels (ASTM A212B, 2103/R3, ASTM A302B, NO 345 and Fortiweld) have been irradiated in the R2 reactor at temperatures between 50 °C and 1 00 °C to neutron doses of 7x10*** - 1. 1x 10^ n/cm^ (> 1 MeV).

Irradiations were made in two reactor positions, one (12) out ­ side the reflector where the neutron spectrum is more moderated, the other (E5) near the center of the core. - 29 -

Impact test data for the steels ASTM A212B, 210 3/R3 and NO 345 did not show any difference due to variations in neutron spectra when materials irradiated to the same dose (E > 1 MeV") were compared. In Fortiweld steel there was a difference in transition temperature of about 20 °C at a dose of 1. 4x10"*'^ n/cm^ (> 1 MeV"). In ASTM A302B steel the change might be 1 5 °C at the same dose (Figure 23). Tensile test and hardness data show that, with an equal dose (E > 1 MeV), a somewhat greater (10-15%) hardening was obtained in the softer spectrum than in the harder. An attempt was made to correlate this difference to the two damage models discussed in sec­ tion 4. 2, using calculated neutron spectra for the two reactor posi ­ tions. The cascade model gave a 7. 6 % difference between the changes obtained in the spectra, while the depleted-zone model gave a 15.2 % difference. The last-mentioned model gave results most in accordance with the experiments. However, the difference between the two neutron spectra is too small to allow quantitative conclusions to be drawn.

6. 7 Irradiation-effects monitoring programs

Monitoring programs will be organized for the power reactors . For the Agesta reactor we obtained a large amount of excess material from the actual melts and also from plates used in the pressure vessel. Tensile and impact specimens of all materials used in the vessel, in­ cluding welds and heat-affected zones, are being irradiated inside the vessel in channels in the outer side of the thermal shield. One set of specimens of each material will be withdrawn and tested each year during the reactor ’s lifetime. The total program consists of more than 3, 500 specimens. A similar but less extensive program is planned for the Marviken reactor.

7. ACKNOWLEDGEMENTS

The author is very much indebted to several of his colleagues , whose contributions are gratefully acknowledged. Mr. H. Mogard made numerous important suggestions, Dr. H. P. Myers and Mr. N. E. Hannerz initiated and executed much of work described here, Mr. K. Luthman participated in the analysis mentioned in part 4. 2, Mr. J Cervacek assisted in the study of neutron spectrum effects, Mr. P. 30

Lindhagen evaluated many of the results, and Messrs. K. Eriksson, Y. Haag, E. Hellstrand, L. E. Johansson, G. Stalberg and C. Sundstrom assisted in different parts of the experimental work. The commissioning and operations staffs of the DR3 and R2 reactors assisted in many ways. - 31 -

REFERENCES

1. NY LANDER, P L, HJERTBERG, P and LARSON, B, Manufacturing and Installation Experience of the Heavy-Water Systems of the Agesta Nuclear Power Plant. Int. conf. peaceful uses of atomic energy. 3, Geneva 1964. Paper No. 602.

2. MAR GEN, P H, LEINE, L and NILS ON, R, The Design of the Marviken Boiling Heavy-Water Reactor with Nuclear Super-Heating. Int. conf. peaceful uses of atomic energy. 3, Geneva 1964. Paper No. 603.

3. JONSSON, A and McHUGH, B, A Large Swedish Nuclear Power Station with a Pressurized Heavy-Water Reactor with an Open Uniform Lattice Core. Int. conf. peaceful uses of atomic energy. 3, Geneva 1964. Paper No. 605.

4. COTTRELL, A H, Theoretical aspects of radiation damage and brittle fracture in steel pressure vessels. Steels for reactor pressure circuits. Iron & Steel Inst. Spec. rep. No. 69, London 1961 p. 281 - 296.

5. PELLINI, W S, STEELE, L E and HAWTHORNE, J R, Analysis of engineering and basic research aspects of neutron embrittlement of steels. IAEA Symp. on radiation damage in solids and reactor materials. Venice May 1962. 2, Vienna 1962, p. 113 - 151.

6. PELLINI, W S and PUZAK, P P, Fracture analysis diagram procedures for the fracture-safe engineering design of steel structures. 1963. (NRL Report 5920.)

7. KIHARA, 1-1 et al. , Effect of stress relieving on brittle fracture strength of welded steel plate. Int. inst. of welding. Doc X-218-59.

8. MOGARD, H, AB Atomenergi. Internal Memorandum 1963. (In Swedish.)

9. CHOW, J. G. Y. , McRICKARD, S B and GURINSKY, D H, Mechanical properties of irradiated iron and iron alloys.. 1963. (ASTM-STP-341, p. 46 - 64.) 10

10. HAWTHORNE, J R and STEELE, L E, The effect of neutron irradiation on the Charpy-V and drop- weight test transition temperatures of various steels and weld metals. 1960. (NRL Report 5479.) (ASTM-STP-286, p. 33 - 56.) - 32 -

11. NICHOLS, R W et al. , UKAEA Reactor Materials Laboratory, Culcheth. Personal communication.

12. GROUNES, M and MYERS, H P, Irradiation effects in a strain-aged pressure vessel steel. Nature 193 (1962) p. 468 - 470.

13. ROBERTSON, T S, ’ - Propagation of brittle fracture in steel. Iron & Steel Inst. J. 175 (1953) p. 361 - 374.

14. ROBERTSON, T S, Brittle fracture - test methods and theoretical considerations. Iron & Steel 28 (1955) p. 161 - 166.

15. HOLLIDAY, W C and NOONE, M J, Metallurgical problems associated with the construction of gas- cooled reactor pressure vessels. Steels for reactor pressure circuits. Iron & Steel Inst. Spec, rep No. 69. London 1961 p. 207 - 239.

16. ULFF, C, Degerfors J&rnverk. Internal Report.

17. GROUNES, M, BestrcLlningsfdrdndringar och neutronatmosfcir i reaktortryck- tankar - n&gra synpunkter. 1962. (AES-2) (in Swedish.)

18. ROSSIN, A D, Radiation damage in steel: considerations involving the effect of neutron spectra. Steels for reactor pressure circuits. Iron & Steel Inst. Spec, rep No. 69. London 1961, p. 393 - 409.

19. SEEGER, A K, On the theory of radiation damage and radiation hardening. Int. conf. peaceful uses of atomic energy. 2, Geneva 1958. Vol. 6 p. 250 - 273.

20. Tentative structural design basis for reactor pressure vessels and directly associated components. 1958. (PB 151987.)

21. PORSE, L, Reactor-vessel design considering radiation effects. ASME Winter Annual Meeting, Nov. 1963. Paper 63- WA-100. (CONF-102-7. ) 22 * * * *

22. NOR^N, T M, The nominal cleaviage strength of steel and its importance for welded structures. North East Coast Inst, of Eng. and Shipbuilders Trans. 73 (1957) p. 87 - 112. - 33

23. NOREN, T M, Columbium as a micro-alloying element in steels and its effect on welding technology. 1963. (SSC-154.)

24. ALGER, J V and PORTER, L F, Evaluation of reference pressure-vessel steel for neutron- irradiation studies. 1964. (U. S. Steel, Appl. Res. Lab. , Project No. 40. 002-006 (4).)

25. GROUNES, M, A small impact test specimen for irradiation experiments. EAES Symposium on Reactor Materials, Saltsjobaden Stock ­ holm 1959. Paper S:6.

26. GROUNES, M, Discussion in: Steels for reactor pressure circuits. Iron & Steel Inst. Spec. rep. No. 69. London 1961, p. 422 - 426.

27. MALMBERG, G, Jamforelse mellan slagseghetsvarden erhallna med standard- provs tavar och med miniatyrprovs tavar . Jernkont. Ann. 145 (1961) p. 183 - 204. (In Swedish.)

28. MYERS, H P, GROUNES, M and BANNERZ, N E, Irradiation behaviour of Swedish steels for reactor pressure vessels. Int. conf. peaceful uses of atomic energy. 3, Geneva 1964. Paper No. 420.

29. GROUNES, M and MYERS, H P, Irradiation effects in some Swedish pressure vessel steels. EAES Symposium on Steel in Nuclear Engineering, Studsvik 1962. Paper S-62:1.

30. MYERS, H P, Neutron damage in steels containing small amounts of boron. 1961. (AE -53.)31

31. GROUNES, M and HANNERZ, N E, Unpubl. reports. - 34 -

Appendix Analysis and properties of the steels that we

Analysis. % Steel Maker plate C Si Mn P S Cr Ni Mo Cu N A1

C-Mn 2103/k3 Degerfors 65 0.15 0.41 1.59 0.014 0.012 0.02 0.01 0.05 0.05 0.006 0.034 mild steels SIS 142102 Oegerfors 20 0.14 0.36 1.58 0.032 0.032 0.06 0.08 - 0.18 0.005 - A212B US Steel 102 0.26 0.22 0.80 0.012 0.036 0.12 0.28 0.034 0.26 0.007 -

Mn-Mo NO'345 Before 35 0.16 0.28 1.52 0.010 0,008 0.17 0.16 0.27 0.08 0.010 low alloy NO 345 Before 125 0.17 0.27 1.37 0.009 0.007 0.18 0.13 0.28 0.05 0.011 0.035 steels DE631A Degerfors 100 0.19 0.29 1.46 0.013 0.014 0.06 0.07 0.45 0.15 0.008 0.031 A302B US Steel 152 0.24 0.23 1.34 0.011 0.023 0.11 0.18 0.51 0.20 0.008 -

B-Mo low Fortiweld A Degerfors 75 0.13 0.34 0.77 0.012 0.007 0.06 0.05 0.56 0.12 0.006 0.051 alloy steel Fortiweld B Degerfors 65 0.12 0.38 0.68 0.023 0.011 0.07 0.04 0.45 0.12 0.007 0.100 0.003 % B

Cr-Ni alloy 2RM2 Before cast 0.07 0.44 0.66 0.011 0.009 12.9 5.3 0.1 steels 248 SV Aveeta 70 0,038 0.53 0.76 0.017 0.005 16.0 4.7 1.0 - 0.038 -

Properties Yield Tensile ElongationReduction Steel strength strength 65 in area kg/mtn x) kg/mnv % %

C-Mn 2103/tU 31 53 38 76 mild steels SIS 142102 ASTM grain size 6.5 30 51 - - ASTM grain size 9 35 55 - - A2I2B 32 53 - 68

Mn-Mo NO 345 (35 mm) Quenched and tempered 47 59 19 75 low alloy NO 345 (125 mm) Normalised and tempered 40 55 28 60 steels DE631A 47 64 - 64 A302B 48 64 - 67

B-Mo low Fortiweld A 50 61 21 63 alloy steel Fortiweld B 49 62 18 - 0.003 % B

Cr-Ni alloy 2RM2 63 88 18 61 steels 248 SV 64 92 18 50

x) 0.2 % offset yield strength 1. Vessel wall 7. Upper thermal shield 2. Filler ring 8. Vessel head 3. Filler body 9. Flange 4. Outer thermal shield 10. Silver seal ring 5. Inner thermal shield 11. Fuel assembly 6. Distribution header 12. Control rod

Fig. 1. The Agesta Reactor ’s Pressure Vessel Refuelling machine

Superheater inlet ports Control rod Steam separators Water level Boiling element Superheater element Moderator tank Moderator inlet nozzles Feed water inlet Steam outlet Impact energy kgm

2RM0

DE- S31A

Fig. 3. Impact energy transition curves for 248 SV and 2RMO steels. As a comparison the corresponding curve for a Mn-Mo-steel (DE-631 A) is included.

a Croc* arrest ‘emperctur*.. heat"» Z™2 ° —------:i Stress kfl/mml ------urwrrsdieted

temperature iCAT) I CAT of th< vessel at the , en/*enc AtOf,itS IH lifetime

fabrication, I Oflcng and ] irrnd.otion 1

* Point on the envelope i of Cr-T curves ter . drfferert ports of the

300 Te-r-oe-aiv-e ®C

Fig. 4. Assumed Fellini diagram Fig. 5. Fellini diagram for DE-631 A for unirradiated and ir- steel showing the vessel-stress radiated steel. vs. temperature diagram for the Marviken reactor ’s pres­ sure vessel. /Sniff /*( Off

•trail

•train

Em-ll W

Fig. 6. Stress-strain diagram Fig. 7. Comparison of neutron for mild steel. When the spectra used. $(E) is the total elongation is between flux of neutrons with en­ a and b, part of the mate ­ ergy E in neutrons/cm^, rial will have elongation a s, MeV. ■while part has the elonga ­ tion b.

Fig. 8. Noren’s NC-test specimen Fig. 9. Results from Noren's NC-test

------NC-test results ------Yield strength - Charpy transition curve

—r -t wIN ... ! H- ! p 1 ^ 1 °.02

Fig. 10. Miniature impact test specimen. ASTM A212B

Fig. 11. Impact energy transition curves as measured with miniature impact specimens. Steels type 2103/R3 and ASTM A212B irradiated to various neutron doses • at 230 °C.

DE 631 A

-200 50 100 ASTM A302 B

>+ +

Fig. 12. Impact energy transition curves as measured with miniature impact specimens. Steels type DE-631 A and ASTM A3202B irradiated to various neutron doses at 265 °C. ZI05/R3 irradiated at 05 *C _ ]QQ

0 50

Fig. 13. Changes in yield strength a , tensile strength a , total elongation et> uniform elongation e . Vickers pyramid hardness HV and transition temperature T as measured with miniature impact specimens in 210 3/R3 steel as a function of neutron dose at the irradiation temperatures of 195° and 265 °C. Fortiweld irradiated at 26S°C

10^-Or

Fig. 14. Changes in yield strength ct , tensile strength a , total elongation e , uniform 8elongation @ , Vickers pyramid hardness^HV and transition temperature T as measured with miniature impact specimens in Fortiweld steel as a function of neutron dose at the irradiation temperatures of 195° and 265 °C. impact

,5,0*10 n/cm(>1 MeV) Jfi* 1019 ------' 19 33 10------

Temperature *C

Fig. 15. Impact energy transition curves as measured with Charpy V-notch impact specimens for DE-6 31 A steel irradiated to different neutron doses at 265 °C. Data points for one one of the curves (1.4x 1 0* ^ nvt) are not shown. S/5 142102 Grain Size ASTM 9

+200°C

5/5 142102

Grain Site ASTM 6,f J______I______L

Fig. 16. Impact energy transition curves as measured with miniature impact specimens of different steels irradiated to about 4 x 1 01 8 n/cm2 (>1 MeV) at 265 °C . ASTM A212B

AT C 30x10 nvt

AT C 30x10 nvt

NO 345 as normalised

30x10nvt

NO 345 as quenched and tempered

Fortiweld miniature specimens

-30x10nvt

-200 Fortiweld Charpy specimens

IRRADIATION TEMPERATURES x 335 C

-100 n 19? C A 16? C - V 12? C " a 11? c • 3? C FORTIWELD B, IRRADIATED IN DR2 REACTOR

Fig. 17. Changes in transition temperatures of different steels as a function of neutron dose at different irradiation temperatures. Curves a - e are based on tests with miniature specimens. Fig. 13. Impact energy transition curves of 2103/R3 steel irra ­ diated to different neutron doses at different irradiation temperatures . NEUTRON DOSE (> I MeV) > 10 , n/ema

Fig. 19. Changes in 0. 2 per cent offset yield strength of 2103/R3 steel as a function of neutron dose at different irradiation temperatures .

-10 / / /

NEUTRON DOSE (> I MeV)»10"” n/mi

Fig. 20. Changes in 0. 2 per cent offset yield strength of Fortiweld steel (heat B) as a function of neutron dose at different irradiation temperatures . AVESTA24&SV _ 70 mm plate O unirradiated ' 1e -|- irradiated 265°C 5^x10 nvt

BOFORS 2RM2 caststeel O unirradiated -|- irradiated 265°C 5,3x10 nvt

22 mm rods. o'

SO/nCV?S 2/?/72 O unirradiated irradiated 265’C l.ixld nvt

Fig. 21. Impact energy transition curves of the new types of steels as measured with miniature impact spec- mens . o» unirradiated X= 30 MW power 0.6 — □ - 3 MW power

Fig. 22. Impact energy results as measured with miniature speci­ mens of 2103/R3 steel. Irradiation 2x10^ n/cm^ (>1 MeV") at 250 °C . Fig. Increase in transition temperature ®C

240 200 160 120 2

3.

- - - - noted irradiations specimens Increase 3

by 4

.a in 5

vertical

of

transition in

7 different

the

more line.

temperature pressure

moderated

4 vessel

A V o ® ® Neutron 5

as ASTM neutron A NO NO Fortiweld

STM 7 measured

steels. 345 345

10

A212B A302B

dose

(quenched (normalized) spectrum

Data

with n/crcr

points

miniature and are (^1

de tempered)

for MeV) ­

LIST OF PUBLISHED AE-REPORTS 185. A new method for predicting the penetration and slawing-dawn at neutrons in reactor shields. By L. Hjarne and M. Leimdarfer. 1965. 21 p. 1—145. (See the back cover earlier reports.) Sw. cr. 8:—. 146. Concentration of 24 trace elements in human heart tissue determined 186. An electron microscope study of the thermal neutron induced loss in by neutron activation analysis. By P. O. Wester. 1964. 33 p. $w. cr. 8:—. high temperature tensile ductility of Nb stabilized austenitic steels. By R. B. Roy. 1965. 15 p. Sw. cr. 8:—. 147. Report on the personnel Dosimetry at AB Atomenergi during 1963. By K.-A. Edvardsson and S. Hagsgard. 1964. 16 p. Sw. cr. 8:—. 187. The non-destructive determination of burn-up means of the Pr-144 2.18 MeV gamma activity. By R. $. Forsyth and W. H. Blackadder. 1965. 148. A calculation of the angular moments of the kernel for a monatomic gas 22 p. Sw. cr. 8:—. scatterer. By R. Hakansson. 1964. 16 p. Sw. cr. 8:—. 188. Trace elements in human myocardial infarction determined by neutron 149. An anion-exchange method for the separation of P-32 activity in neu- activation analysis. By P. O. Wester. 1965. 34 p. Sw. cr. 8:—. tron-irradited biological material. By K. Samsahl. 1964. 10 p. Sw. cr 189. An electromagnet for precession of the polarization of fast-neutrons. 8:“. By O. Aspelund, J. Bjorkman and G. Trumpy. 1965. 28 p. Sw. cr. 8:—. 150. Inelastic neutron scattering cross sections of Cu6: and Cu65 in the energy 190. On the use of importance sampling in particle transport problems. By region 0.7 to 1.4 MeV. By B. Holmqvist and T. Wiedling. 1964. 30 p. Sw. cr. 8:—. B. Eriksson. 1965. 27 p. Sw. cr. 8:—. 191. Trace elements in the conductive tissue of beef heart determined by 151. Determination of magnesium in needle biopsy samples of muscle tissue neutron activation analysis. By P. O. Wester. 1965. 19 p. Sw. cr. 8:—. by means of neutron activation analysis. By D. Brune and H. E. Sjoberg. 1964. 8 p. Sw. cr. 8:—. 192. Radiolysis of aqueous benzene solutions in the presence of inorganic oxides. By H. Christensen. 12 p. 1965. Sw. cr. 8:—. 152. Absolute El transition probabilities in the dofermed nuclei Yb177 and Hfm. By Sven G. Malmskog. 1964 . 21 p. Sw. cr. 8:—. 193. Radialysis of aqueous benzene solutions at higher temperatures. By H. Christensen. 1965. 14 p. Sw. cr. 8:—. 153. Measurements of burnout conditions for flow of boiling water in vertical 3-rod and 7-rod clusters. By K. M. Becker, G. Hernborg and J. E. Flinta. 194. Theoretical work for the fast zero-power reactor FR-0. By H. Haggblom. 1964. 54 p. Sw. ci. 8;— 1965. 46 p. Sw. cr. 8:—. 154. Integral parameters of the thermal neutron scattering law. By S. N. 195. Experimental studies on assemblies 1 and 2 of the fast reactor FRO. Purohit. 1964. 48 p. Sw. cr. 8:—. Part 1. By T. L. Andersson, E. Hellstrand, S-O. Londen and L. I. Tiren. 1965. 45 p. Sw. cr. 8:—. 155. Tests of neutron spectrum calculations with the help of foil measurments in a D2 G and in an H2 G-moderated reactor and in reactor shields of 196. Measured and predicted variations in fast neutron spectrum when pene­ concrete and iron. By R. Nilsson and E. Aalto. 1964. 23 p. Sw. cr. 8:—. trating laminated Fe-D2 C. By E. Aalto, R. Sandlin and R. Fraki. 1965. 20 p. Sw. cr. 8:—. 156. Hydrodynamic instability and dynamic burnout in natural circulation 197. Measured and predicted variations in fast neutron spectrum in massive two-phase flow. An experimental and theoretical study. By K. M. Beck­ shields of water and concrete. By E. Aalto, R. Fraki and R. Sandlin. 1965. er. S. Jahnberg, 1. Haga, P. T. Hansson and R. P. Mathisen. 1964. 41 p. 27 p. Sw. cr. 8:—. Sw. cr. 8:—. 198. Measured and predicted neutron fluxes in, and leakage through, a con­ 157. Measurements of neutron and gamma attenuation in massive laminated figuration of perforated Fe plates in D2 0. By E. Aalto. 1965. 23 p. Sw. shields of concrete and a study of the accuracy of some methods of cr. 8:—. calculation. By E. Aalto and R. Nilsson. 1964. 110 p. Sw. cr. 10:—. 199. Mixed convection heat transfer on the outside of a vertical cylinder. 158. A study of the angular distributions of neutrons from the Be9 (p,n) B9 By A. Bhattacharyya. 1965. 42 p. Sw. cr. 8:—. reaction at low proton energies. By. B. Anfolkovic', B. Holmqvist and T. Wiedling. 1964. 19 p. Sw. cr. 8:—. 200. An experimental study of natural circulation in a loop with parallel flow test sections. By R. P. Mathisen and O. Eklind. 1965. 47 p. Sw. 159. A simple apparatus for fast ion exchange separations. By K. Samsahl. cr. 8:—. 1964. 15 p. Sw. cr. 8:— 201. Heat transfer analogies. By A. Bhattacharyya. 1965. 55 p. Sw. cr. 8:—. 160. Measurements of the Fe54 (n, p} Mn54 reaction cross section in the neutron 202. A study of the "384" KeV complex gamma emission from plutonium-239. energy range 2.3 —3.8 MeV. By A. Lauber and S. Malmskog. 1964. 13 p. By R. S. Forsyth and N. Ronqvist. 1965. 14 p. Sw. cr. 8:—. Sw. cr. 8:—. 203. A scintillometer assembly for geological survey. By E. Disslng and O. 161. Comparisons of measured and calculated neutron fluxes in laminated Landstrdm. 1965. 16 p. Sw. cr. 8:—. iron and heavy water. By. E. Aalto. 1964. 15 p. Sw. cr. 8:—. 204. Neutron-activation analysis of natural water applied to hydrogeology. 162. A needle-type p-i-n junction semiconductor detector for in-vivo measure­ By O. Landstrdm and C. G. Wenner. 1965. 28 p. Sw. cr. 8:—. ment of beta tracer activity. By A. Lauber and B. Rosencrantz. 1964. 12 p. 205. Systematics of absolute gamma ray transition probabilities in deformed Sw. cr. 8:—. odd-A nuclei. By S. G. Malmskog. 1965. 60 p. Sw. cr. 8:—. 163. Flame spectro photometric determination of strontium in water and 206. Radiation induced removal of stacking faults in quenched aluminium. biological material. By G. Jonsson. 1964. 12 p. Sw. cr. 8:—. By U. Bergenlid. 1965. 11 p. Sw. cr. 8:—. 164. The solution of a velocity-dependent slowing-down problem using case's 207. Experimental studies on assemblies 1 and 2 of the fast reactor FRO. eigenfunction expansion. By A. Claesson. 1964. 16 p. Sw. cr. 8:—. Part 2. By E. Hellstrand, T. L. Andersson, B. Brunfelter, J. Kockum, S-O. 165. Measurements of the effects of spacers on the burnout conditions for Londen and L. I. Tiren. 1965. 50 p. Sw. cr. 8:—. flow of boiling water in a vertical annulus and a vertical 7-rod cluster. 208. Measurement of the neutron slowing-down time distribution at 1.46 eV By K. M. Becker and G. Hernberg. 1964. 15 p. Sw. cr. 8:—. and its space dependence in water. By E. Moller. 1965. 29. p.Sw.cr.8:—. 166. The transmission of thermal and fast neutrons in air filled annular ducts 209. Incompressible steady flow with tensor conductivity leaving a transverse through slabs of iron and heavy water. By J. Nilsson and R. Sandlin. 1964. 33 p. Sw. cr. 8:—. magnetic field. By E. A. Witalis. 1965. 17 p. Sw. cr. 8:—. 167. The radio-thermoluminescense of CaSOj: Sm and its use in dosimetry. 210. Methods for the determination of currents and fields in steady two­ By B. Bjarngard. 1964 . 31 p. Sw. cr. 8:—. dimensional MHD flow with tensor conductivity. By E. A. Witalis. 1965. 13 p. Sw. cr. 8:—. 168. A fast radiochemical method for the determination of some essential 211. Report on the personnel dosimetry at AB Atomenergi during 1964. By trace elements in biology and medicine. By K. Samsahl. 1964. 12 p. Sw. K. A. Edvardsson. 1966. 15 p. Sw. cr. 8:—. cr. 8:—. 212. Central reactivity measurements on assemblies 1 and 3 of the fast 169. Concentration of 17 elements in subcellular fractions of beef heart tissue reactor FRO. By S-O. Londen. 1966. 58 p. Sw. cr. 8:—. determined by neutron activation analysis. By P. O. Wester. 1964. 29 p. Sw. cr. 8:—. 213. Low temperature irradiation applied to neutron activation analysis of mercury in human whole blood. By D. Brune. 1966. 7 p. Sw. cr. 8:—. 170. Formation of nitrogen-13, fluorine-17, and fluorine-18 in reactor-irradiated H2 0 and D2 0 and applications to activation analysis and fast neutron 214. Characteristics of linear MHD generators with one or a few loads. By flux monitoring. By L. Hammar and S. Forsen. 1964. 25 p. Sw. cr. 8:—. E. A. Witalis. 1966. 16 p. Sw. cr. 8:—. 171. Measurements on background and fall-out radioactivity in samples from 215. An automated anion-exchange method for the selective sorption of five the Baltic bay of Tvaren, 1957—1963. By P. O. Agnedal. 1965. 48 p. Sw. groups of trace elements in neutron-irradiated biological material. cr. 8:— By K. Samsahl. 1966. 14 p. Sw. cr. 8:—. 172. Recoil reactions in neutron-activation analysis. By D. Brune. 1965. 24 p. 216. Measurement of the time dependence of neutron slowing-down and Sw. cr. 8:—. thermalization in heavy water. By E. Moller. 1966. 34 p. Sw. cr. 8:—. 173. A parametric study of a constant-Mach-number MHD generator with 217. Electrodeposition of actinide and lanthanide elements. By N-E. Barring. nuclear ionization. By J. Braun. 1965. 23 p. Sw. cr. 8:—. 1966. 21 p. Sw. cr. 8:—. 174. Improvements in applied gamma-ray spectrometry with germanium semi­ 218. Measurement of the electrical conductivity of He13 42plasma 5 induced by conductor dector. By D. Brune, J. Dubois and $. Hellstrom. 1965. 17 p. neutron irradiation. By J. Braun and K. Nygaard. 1966. 0 p. Sw. cr. 8:—. Sw. cr. 8:—. 219. Phytoplankton from Lake Magelungen, Central Sweden 1960—1963. By T. 175. Analysis of linear MHD power generators. By E. A. Witalis. 1965. 37 p. Willen. 1966 . 44 p. Sw. cr. 8:—. Sw. cr. 8:—. 220. Measured and predicted neutron flux distributions in a material sur­ 176. Effect of buoyancy on forced convection heat transfer in vertical chann­ rounding av cylindrical duct. By J. Nilsson and R. Sandlin. 1966. 37 p. els — a literature survey. By A. Bhattacharyya. 1965. 27 p. Sw. cr. 8:—. Sw. cr. 8:—. 177. Burnout data for flow of boiling water in vertical round ducts, annuli 221. Swedish work on brittle-fracture problems in nuclear reactor pressure and rod clusters. By K. M. Becker, G. Hernborg, M. Bode and O. Erik- vessels. By M. Grounes. 1966 . 34 p. Sw. cr. 8:—. son. 1965. 109 p. Sw. cr. 8:—. 178. An analytical and experimental study of burnout conditions in vertical round ducts. By K. M. Becker. 1965. 161 p. Sw. cr. 8:—. 179. Hindered El transitions in Eu155 and Tb181. By $. G. Malmskog. 1965. 19 p. Forteckning aver publicerade AES-rapporter Sw. cr. 8:—. v 180. Photomultiplier tubes for low level Cerenkov detectors. By O. Strinde- 1. Analys medelst gamma-spektrometri. Av D. Brune. 1961. 10 s. Kr 6:—. hag. 1965. 25 p. Sw. cr. 8:—. 2. Bestralningsforandringar och neutronatmosfar i reaktortrycktankar — 181. Studies of the fission integrals of U235 and Pu239 with cadmium and nagra synpunkter. Av M. Grounes. 1962. 33 s. Kr 6:—. boron filters. By E. Hellstrand. 1965. 32 p. Sw. cr. 8:—. 3. Studium av strackgransen i mjukt stal. Av G. Ostberg och R. Attermo. 182. The handling of liquid waste at the research station of Studsvik,Sweden. 1963. 17 s. Kr 6:—. By S. Lindhe and P. Linder. 1965. 18 p. Sw. cr. 8:—. 4. Teknisk upphandling Inom reaktoromrddet. Av Erik Jonson. 1963. 64 s. 183. Mechanical and instrumental experiences from the erection, commis­ Kr 8:—. sioning and operation of a small pilot plant for development work on 5. Agesta Kraftvarmeverk. Sammanstallning av tekniska data, beskrivningor aqueous reprocessing of nuclear fuels. By K. Jonsson. 1965. 21 p. Sw. m. m. for reaktordelen. Av B. Lilliehook. 1964 . 336 s. Kr 15:—. cr. 8:—. Additional copies available at the library of AB Atomenergi, Studsvik, 184. Energy dependent removal cross-sections in fast neutron shielding Nykoping, Sweden. Transparent microcards of the reports are obtainable theory. By H. Gronroos. 1965. 75 p. Sw. cr. 8:—. through the International Documentation Center, Tumba, Sweden.

EOS-tryckerierna, Stockholm 1966