iiiiiiii --_ -- ' - ' iiiiiiiiaiiiii M M 0 0 0 0 0 l I l I 0 N --N iiiiiiii JEBRET ··- C · · llIISJ OC MEN r CONSISTS OF - t · To te of Issue March 4 1964 Report Number KOA 1237 • S A Levin L R Powers E Von Halle· Operations Analysis Division D M Iang Superintendent - UNION CARBIDE CORPORATION NUCLEAR DIVISION Oak Ridge Gaseous Diffusion Plant Oak Ridge Tennessee SECRET- 01 COPI E S SERi EQ K-0A-1237 FtE t i G 1 ---··-·--· ·-- -----·---·--- -· Nth POWER EVAWATION 1A · 1 D l l l f llifl l l l ijlfl Il lf ll f l ·---- - 0 A 1 Z 3 7 OF ' PAGE J SECRET 3 Report Number K-OA-1237 Title Nth POWER EVAWATION ABS CT In this report an attempt is made to correlate the probability of some country an Nth power successfully producing a nuclear weapon or weapons by means of an overt or clandestine program involving the production and operation of a gas centrifuge plant with the industrial capability of that country For this purpose the c ountries of interest have been divided into·three groups designated by X Y and z Group x countries are those possessing a relatively high degree of technical competence and which have a high degree of industrial activity Group Z countries are those which possess relatively little technical skills and which have relatively little industrial activity Group Y countries are· those which lie between and have limited internal industrial activity The over-all time investment and work force required for the construction of a centrifuge weapons facility and the cdst and manpower required for its operation for various centrifuge models is presented for X Y and Z nations It is felt that it is feasible for the countries described in this report which do not have a nuclear weapons program to produce enriched uranium by means of a small gas centrifuge plant The current status of f J3 S centrifuge development programs in foreign nonComrriunist countries is reviewed on the basis of information obtained from both the open and c assified scientific literature the press and ntelligence reports This report supersedes reports KOA-662 and KOA-916 prepared at Oak Ridge in 1960 and 1962 · CkE -· - SECR r - 5 TABLE OF CONTENTS ' 9 summary ••• • 11 Plant Description ••••••••••• •• • 24 Introduction • Centrifuge Facility •••••••• ••• •••••••• 24 Zippe Centrifuge 24 Present 6-Inch Subcritical Centrifuge ••••••••••• •••••• •••••••• Proposed Subcritical Centrifuge ••••••••••••••••••••••••••• •••••• Proposed Supercritical Centrifuge •••••••••••••• ••••••••••• ••••• Future Subcritical Centrifuge •• • • • • Future Supercritical Centrifuge •••••• •••••••••• ·•••••••••••••••••• 27 27 30 30 • ••••••••••••••••••• # 26 Centrifuge Plant Characteristics ••••• ••••••••••••••• •••••••• •• • 3l PJ ant Iayout 31 Electrical Eq_uipment 38 Process Piping 40 Process Controls 40 Process Auxiliaries 42 Operating Costs _ 45 Feed Pl ant 46 Metals Plant · 47 Discussion 72 72 Feasibility of Clandestine Operation Minimum Time Required to Produce the First Nuclear Weapon Opera ting and Maintenance •••• •••••••••••••••• •••• • • • • • • • • • • Capital and Operating Cost Estimates Key Areas Indicating Potential for Weapons Production 74 74 74 75 Correlation of Industrial Capability of Nations • ••••• •• •• • 75 Clandestine Production of Enriched Uranium by the Sweep Diffusion Process 78 Review and Evaluation of Foreign Cent•rifuge Programs ••• • •• 80 Bo Introduction ••••••••••••• • Brazil · England •• •• ••••• • ••• • Japan The Netherlands •••••••••••• West 81 82 86 91 Ge ny • • 93 Other Foreign Countries ••••• 99 Acknow ledgem ent s • • • • • • • • • 100 References 101 -· SE6RET y • '1 •- · · ----·- -- - 7 · LIST OF TABLES 1 2 3 4 5 6 7 8 rr· -- Ilescri tion of Centrifuge Models •• • • • ••• • 16 Gas Ce trifuge Plant Summary for Class X Nations • • • • 17 Gas Centrifuge Plant Summary for Class Y Nations •••••••• • ••• 18 Gas Centrifuge Plant Summary for'Class Z Nations • •• • ••• •• • 19 Gas Centrifuge Plaut Summary for Class X Nations •••••• ••• • • 20 Gas Centrifuge Plant Summary for Class Y Nations • • ••••••• 21 Gas Centrifuge Plant Sumrrary for Class Z Nations •• ••••••• ••••• 22 Minimum Time Re uired to Produce First Nuclear Weapon •••••••• •• 23 Estilnated Manpower and Capital for Centrifuge Plant Construction and Operation in s - Model A Original Zippe ••••• •••• • •••• 49 t • j D I 21 22 23 24 25 O_peration in U S 25 TtJ Per Year 60 Estimated Manpower and Capital for Feed Plant Construction and Operation in U S 50 TU Per Year 61 Estin ated Manpower and Capital for Feed Plant Construction and Open3 tion in U S 100 Tu Per Year ••••• ••••••••••••••• •••• 62 Estimated Manpower and Capital for Feed Plant Construction and Operation in u s 500 TU Per Year 63 Estimated Manpower and Capital for Metal Component Facility Construction and Operation in U S 50 Kg U Per Year 64 Estimated Manpower and Capital for Metal Component Facility Construction and Operation in U S 500 Kg U Per Year 65 ---- - - - - - - - - - - - - - - - - - i ff Or r 17 Sf Crt E P - · •' ' •• 8 - iRD JP ___ _ • LIST OF FIGURES 1 2 Schema tic Representation of Typical Zippe Centrifuge •••• • 25 Scher iatic Representation of the Proposed Subcritical Centrifuge 28 3 Schematic Representation of a Supercritical Centrifuge •••••• • 4 One-Half of a 10-Centrifuge Parallel Cell • •• • ••• • ••••••• 5 10-Centrifuge Parallel Cell •• • •••••• • • ••• - • • • • • • • • 6 10-Centrifuge Series Cell One Centrifuge Per Stage •• ••••• 7 Typ cal 10-Cel Sub Cas de •••••••••••• • ••• 8 Typical Electrical Distribution System • ••••• •• •••••• 9 layout of Ideal Plants Containing Subcritical Centrifuges ------ l- 3 in • z _25 Q m sec •••••• - - _ _ ••• ·-- - -· ••• ·--•-• _0 _ _ _• _ •__ - _ _ _ '-- fl5 16 33 34 · 36 37 39 f t e L 'll @ i Ar 7 9_6 - J r r 29 Cori'OlatTonOf Relative Industrial Capabilities Schematic Representation of a Sweep Diffusion Column •••• •••••• 79 EER l f' _ - SECRET 1 --SEGRH 9 c· Nth Power Evaluation · INTRODUCTION The nations of the Western Alliance have been engaged in a series of sporadic negotiations with the Soviet Bloc on the subject of nuclear disar mament sinc_e the mid 1950' s Thi_s _st dy w as__ mcierlaken in order to assess the potential of the gas centrifuge as-a means of attaining a nuclear weapons potential in various size foreign countries and to provide valuable up-to-date ba kground material for future disarmament conferences · - ' ·' In a previous e - i ···- prelilllinary evaluation was made of the gas centrifuge as a means of producing a small number of nuclear weapons either overtly or covertly in a colllltry currently not known to have a nuclear weapons program This study is an extension of the earlier work covering a wider range of production rates and incorporating the present gas centrifuge technology developed over the past three years in the USAEC gas centrifuge development program The effects of advanced models anticipated as a result of future development effort are also presented Two of the most promising routes a country might follow in order to pro duce a small number of nuclear weapons are the gas centrifuge for the production of U-235 and the natural uranium reactor for the production of At the t i ine ·is- r··tbe j ire1ffous···cen'fa·Ifuge eva1t iati6u··mentioriea 1 plutonium I ·a bove · ari evaluation of the natural uranium-graphite reactor route for the production of plutonium was made by the Richland Operations Office J The present report is limited to the evaluation of the potential of the gas centrifuge route I V In the report an attempt is made to correlate the probability of some '_ colllltry an Nth power successfully producing a nuclear weapon or weapons by means of an overt or clandestine program irrvolving the production and operation of a gas centrifuge plant with the industrial capability of that country For this purpose the countries of interest have been f ' divided into three groups designated as X Y and Z Group X countries are those which possess a relatively high degree of' echnical compete -- and which have a high leyel of industrial activit DELETED -- i • V 1 j ' i' l ' - ' - _ _ _ _ _ _ _ _ _ _ _ _ ' · --SECRET - ----- 10 J DELETED i --· ------- f c wt2 cn 1Group Z countries are those possess-rel 3 ti- ely·--ltt c1e- fe hnical - -1 - n _yhich have relative y l_' i ttle industrial_ actiYi -a roup y countries are those whic 19 fe41I ·ana -·wEfrli J iEtve ___ · 1 i nited in1ernal industrial activityf l ·- --------' - DELETED _ - For this evaluation two sizes of production facilities are considered l Capable of producing 50 kg per year of highly enriched U-235 which should be a sufficient amount of fissionable material for the fabrication of at least one nuclear weapon per year 2 Capable of producing 500 kg per year of highly enriched U-235Th is J roduction rate should be about the upper limit for which the centrifuge process would be used The production facilities may be considered as consisting of three separate processes These are l The feed plant in which the ore concentrate is converted to process gas 2 The isotope separation plant ftself in which the concentration of U-235 is raised from that of the feed 0 71 weight per cent to tbat re uired for a nuclear weapon greater than 90% 3 The metal reduction plant in which the material is converted to uranium metal and then machined to finished metal parts for the nuclear weapon Six centrifuge models are considered for the two production-rate plants studied The centrifuge models considered can be broken into three categories based on the progress made in the AEC gas centrifuge development program l Those centrifuges which could be operated without the necessity of other development work 2 Those centrifuges which are presently being wo·rked on and which may be operable within a year or two 3 Those centrifuges which require considerable advances in materials and operational technology These centrifuges might be available in about 5 to 10 years Table l presents a sUIIIIm ry of the more important characteristics of the various centrifuge models In fuble 1 each of the centrifuge models is assigned a code letter which will be used to refer to the models in all of the following cost and s ry tables Model A the original Zippe machine is included only for comparison purposes and is not considered a feasible machine even _ for the smaller production rate · ---- f 7fy' £ JOit J fLr j · ll -fflD SUMMARY It is felt that it is feasible for the countries described in this report which do not now have a nuclear weapons program to produce enriched · · · uranium by means of a small gas centrifuge plant The centrifuge process · lends itself to clandestine operation however in a country not having • · its own uranium ore supply safeguards requiring adequate inspection •- would probably prohibit clandestine operation A class X country would 1 need no outside assistance while a class Y country would probably have to import some of the hard ware and auxiliary equipment nec·essary to fabricate the centrifuge Plant A class Z country would probably have to purchase prefabricated centrifuges and almost all of the auxiliary equipment for the centrifuge plant from foreign vendors In addition a class Z country would need technica L advisors t'rom the outside to aid _ in the construction and operation of the centrifuge plant r A summary of the over-all time investment and work force required for construction of a centrifuge weapons facility and the cost and manpower required for its operation for each of the centrifuge models described in Table 1 is presented in Tables 2 throu 4 for the X Y and Z nations for the 50 kg·U-235 per year production rate and in Tables 5 through 7 for the 500 kg U-235 per year production rate A detailed cost time and manpower breakdown of the centrifuge weapons facility for its three separate processes -- isotope separation plant feed plant and metals plant based on estimated United States requirements is also presented in the report A correlation which is used to obtain factors for converting United States requirements into requirements of other nations is presented in Figure 15 DELETED -1ii i flllleS presenl ea · ·which assume knowledge of thedetails o_f_a_p_a_rt_icu lar centrifuge model reflect the time required by a class X Y or Z country to develop the centrifuge fabricating operating and cascading techniques to the point where detailed plant designs could be initiated plus the estimated ti r · to build the plant and produce the first weapon If no knowledge o the details of a centrifuge model is available from the United States or from some _other highly advanced country additional development time would be required which is reflected in the higher ti mes presented in Table 8 for these cases The total capital investment which amounts to about $43 5 million for the model C ceatrifuge facility for the low production rate coupled with operating costs of about $4 5 million per year will in the case of a class Z country be quite a burden on the economy A class Z countr-f would have to be highly _motivated to undertake such an expensive project However it is est in ated that these costs may be reduced by as much as 75% with the more advanced centrifuge models these much lower costs would certainly make the project more feasible for a class Z country au 12 The physical concealment of the small centrifuge plant should present no problems because of the relatively small size of the plant The ground area would range from a small fraction of an acre maybe up to 3 4 acre depending upon which centrifuge model was used While the centrifuge plant might be about three stories high for conventional construction this would not be noticeable in industrial areas of X and Y nations the building height could be lowered without sacrificing much in additional cost This would however increase the ground area required The lower building height might be desirable in the case of class Z countries in which three-story buildings could not be easily camouflaged without using costly underground construction The feed and metals processing facilities are relatively small operations which could be performed within the centrifuge separation plant The power reg_uirementp for the centrifuge plant will be relatively small ranging from 0 5 to 8 4 megawatts for the small_plant and 2 9 to 21 megawatts for the large plant depend ing upon which centrifuge model is assumed ' he effluents from the centrifuge plant could be handled easily The waste streams from the plant over a period of -a year which is essentially the same amount as the feed could be contained in a relatively few lO ton UF cylinders which could be stored conveniently anywhere 6 within the plant The off-gases from the feed and metals plant could probably be neutralized with caustic and the product deposited in seepage pits Factors Influencing the Choice of the Centrifuge Route in Countries X1 Y and Z In a class X country the choice of whether to produce a limited amount of nuclear weapons by the centrifuge or the reactor route is not clear cut at present The centrifuge has the potential for low capital and operating costs but the techno_logy is unavailable in the unclassified literature and the centrifuge at the present time will present a higher risk of failure It is felt however that a class X country will have the experienced scientists and engineers necessary to bring a centrifuge facility into successful operation Therefore those class X countries having an advanced centrifuge program of their own or access to the results of another country's program may hoose the centrifuge route for attainment of their first nuclear weapons if they plan on modest expansion of their nuclear ca ability in the future and r foresee a use for either slightly or highly enriched uranium in their future nuclear energy programs Those countries having no centrifuge program o_f their own and no access to the results of programs in other countries would not be li ely to choose the centrifuge route As more advanced centrifuge models are developed and the reliability of the centrifuge is established the choice may well be overwhelmingly in favor of the centrifuge process in those countries having access to the results of the development effort In a class Y country if the goal is pr µnarily the achievement of a clandestine very limited nuclear capability the reactor-pluto nium route would undoubtedly be mo re attractive from the standpoint of certainty at the present time because of the availability of infonnation in the open •r -· L -l -literature Again however if the country has access to the results of advanced centrifuge development work and they plan a modest expansion of their initial nuclear capability in the future and or foresee the need for enriched uranium in their future nuclear energy programs they may risk following the centrifuge route for their initial nuclear capability As the more advanced centrifuges are developed and the class Y country obtains detailed information about these advanced machines their choice between the reactor-plutonium route or the centrifuge route could favor the centrifuge route A class Z country will find the construction and operation of either a centrifuge or plutonium facility a difficult task These countries would need much help from a class X or Y country I ue to the nonindustrial nature of a class Z country c1 c omplete J cland estine facility could probably not be built The choice of whether to build a centrifuge or plutonium facility in a class Z country would probably depend upon which appeared most attractive to the_ class X or y· country with which they were collaborating _ It probably would be easier- to justify a power reactor than an unspecified centrifuge facility a nd if the primary objective is an attempt at a clandestine operation the plutonium route would-probably be followed Again as the more advanced centrifuges are developed the centrifuge site will probably become smaller and easier to hide and the choice may swing to the use of the centrifuge method Security Considerations - The continued classification of centrifuge technology and the exercise of export control over advanced centrifuges or critical components by the United States and her allies is highly desirable The declassification of the centrifuge development efforts by the United States or her allies especially the work done in the past three years will increase the number of countries which might obtain a nuclear capability through the centrifuge route and reduce the time required for the attainment of a nuclear capability by those countries presently considering the use of the centrifuge • The release of the results of future development work will have an even more pronounced effect since it is expected that future work will make the centrifuge route more attractive from the standpoint 0f both tim -and· money · Key Items There ar·e certain key items which indicate the possibility that an Nth power may be constructing a centrifuge facility for enriched uranium production The more important of these items are listed in the report § ' ep Diffusion Process Although only the centrifuge process has been considered in this report for the clandestine production of enriched uranium there are otb er processes which might be better matched with the lower technological capabilities e f Y and Z foreign powers Such a process for example would be sweep diffusion In the sweep diffusion process a gaseous mixtll1' 'e - - - ----------·· ·--··· · - · - - - - - - - - - - - - - - - - 14 o f isotopes is conf'ined in a column across which flows a current of a third component called the sweep gas or sweep vapor As the sweep vapor flow·s through the process gas mixture it tends to sweep along the heavy or less diffusible component A swe p dif'fusion plant req_uires process equi pn1ent such as nickel plated pipe and screen rotary lobe blowers to move the process gas between columns centrifugal pumps to pump the condensed sweep ·vapor to the evaporators and the evaporators themselves A preliminary investigation -of the sweep diffusion process for the separa- tion of other isotopes is being considered this investigation would also determine its potential for uranium isotope separation One of the more serious problems which would have to be overcome is the solubility of process € J3 S in the condensed sweep vapor Evaluation of Foreign Centrifuge Programs The current status of gas centrifuge development programs in foreign nonCommunist countries is reviewed and an evaluation of the program in each country is made with respect to its scope and direction ultimate goals and chances for success The information used in the preparation of this report was obtained from the open and classified scientific literature the press and from intelligen e documents ·which were rnade available for this purpose This work supersedes a previous ·report 2 From the inf'orI l8 tion available it appears that only four foreign countries outside the Communist Bloc are still making· any serious effort to devel on __ a gas centrifuge process for the enrichnent of t he uranium isotopes • se are Weat Gernany Engl and the Netherlands and Japan In additYon Brazi l is maintaining its small interest in the centrifuge process and France has recently been reported engaged in ·centrifuge dev lopr ient ·· · In general due in large part to the fact tbat ·the uropean nations agreed at the end of 1960 to impose security restrictions on all future inf'o i m9 tion arising from their development programs little new inforroa tio i hs s beco' le a• -ai 1E ble si ice the preparation of t 1 re _ 1 t tvro yea rs ago sx e pticn s to t 1ia a r e England with which the United 81 a tes has an ' brrorr J 9 tion exchange program and ra o a be --e p 'obably for 7 J 1 tical rea_ ons the work s - r lsr 'cl c -o f -d ·--h···e··-- - - 1 •- ·y·· J ·1 ·· b1e· •- n-f o- t- on· ·J _ _l l -- _• -•- J-• J • · Q s·n·' e· J ' L LI V J •uo _ i G is most _probable· t t-··as a result of the furor which ceni rifuge developments evoked in the press in 196o almost every class X country has under n a centrifuge develo-nt effort of ome sort everi if 3-uite 'l L 1 1 1 - _ ™ DELETED l --- --- ' f the five countries mentioned above West Germany is probably carrying on·the most extensive program It is comprised both of the development work headed by Dr w Groth at Bonn and the development program being Ji 'sued q y DEGUSB P at Frankfurt under the dir ac tior i of Dr fl- Zippe The progress made by the Germans prior to ·che ua oi '19 i' when the West of - - ---- 16HET 15 06'· German Govermr ent classified all further developments ca i be assessed reaso -iably · ell fro o the availab --J iteratu ' '· DELETED ---- - - 1'-1 -f __ u 4 j It-Ts planned to consolidate all of the ·German ce t ri ' 1 e--· - - r· _ - _n_a__ _ new facility at Juli ch near Aachen where the work will be conducted directly for the West German Government 0 ----- pot c I I I ' I 1 The Japanese in 1958 initiated a modest centrifuge development program apparently aimed at the development of an economical Groth-type machine It has recently been reported that the Japanese are also investigating a Zippe-type machine Although the progress in their first two years of work on the centrifuge process was most impressive the accomplishments made during the second two years of their program appear to be Quite limited Their work is not as yet classified Although it is known that centrifuge work has been going on in the Netherlands for some time very little about their work has been made public The Dutch along with the Germans have encouraged the use of centrifuges for· a proposed Euratom isotope separating facility In May of this year the Netherlands an otmced that they are embarking on a three-year centrifuge development program which if successful will lead to a pilot plant for uranium isotope separation DELETED All five of these countries have the stated or implied goal of developing an-econoffiical rocess for the production of U-235 for peacef'ul uses However as has already been pointed out in this report a workable centrifuge process regardless of the purpose for which it was developed provides a capability for the production of nuclear weapons While many of the scientists working on centrifuges are probably interested in separating uranium for power reactors as they claim the national leaders supporting these programs may well be interested in eventually using the centrifuge process to obtain nuclear weapons ---· ·--· -·-·---··- CJ' £ - Y- TABLE l DESCRIPTION OF CENTRIFUGE MODELS Model DescrlJ _tion rq - A Original Zippe Length in 12 -· -- ·-- - _ __ Diameter in 3 Speed m sec 350 Separative Capacity 6U max kg u yr Fabrication Techni9 ue Material Al 7075 Extruded 1 12 - - - Separative Over-all Capacity Centrifuge oU expected Efficiency kg uzYr 'f _ o 537 - ' ·• 30 u • I q I I i 'J DELETED li 1 ---------_ - ·-------- _________ __ I-' O' - ____ I ' - _ _- ___ _r_ - --------- _ _ · These figures represent actwH measured separative capacities for these centrifuge models All other efficiencies and s1cparati• apacitiea are calcnlated values Qt l · ' ii ' 1' ' - ' ' -- 'rABLE 2 GAS CENTRIFUGE PLANT SUMMARY FOR CLASS X NATION 50 Kg U PER YEAR AT 90% U-235 Machine Capital Investment Centriruge Plant Feed Plant Mntals PJ ant A ¢ 1 l 1· n1 1 C Construction Manpower Man-Months Over-all Constructlon Time yr Metals Plant Totai Power Required centrifuge plant only kw Operating Work Force Centrifuge P1-ant Technical 'fatal Feed Plant Technical Total Metals f'lant Technical Total Plant Work Force _ ----------__ 1 I I l_ ····--···-··-- -· L 11 568 000 1 918 000 364 ooo 13 850 000 14 9 16 000 1 420 100 30 790 4 8 860 60 8 250 2 6 440 380 3 940 2 4 3 540 2 2 11 066 000 621 000 l53 000 11 840 000 3 386 ooo ·497 000 153 000 4 036 000 1 495 000 -429 000 153 000 2 077 000 1 4 8 ooo 4 9 000 l i3 000 2 o -0 000 84lO 540 000 412 000 153 000 1 105 000 319 000 412 000 153 000 U84 _ooc 2110 888 2520 470 l1- -v 32 821 13 270 10 118 10 100 h5 33 6 33 6 29 6 29 5 27 27 2 9 40 2 'l'otal K 28 271 000 2 150 000 364 ooo 30 785 000 Operating Cost ¢ yr Centrifuge Plant Feecl Plant I 104 078 000 2 626 000 36li ooo 107 060 000 cali Co11str11ctio11 -Tork Force 'l'utal Number of Men Technical G 1 2 664 ooo 1 9-L8 ooo 36h ooo 30 2 4 870 000 1 860 000 364 ooo 7 09lf OOO 4 3 5· 000 1 860 000 3611 000 · 6 5 19 oou 280 20 2 450 261 20 30 2 2 2 7 2 170 2 2 7 5 6 6 6 2 6 6 867 309 153 135 78 66 CentrifUge Model Identification - - DELETl D -------- -···- ___ ----- N - ----- --·-· _ -- 6 f ' TABLE 3 G S CENTRIFUGE PLANT SUMMARY FOR CLASS Y NATION 50 Kg U PER YEAR AT 90% U-235 Machine T nie A Capital Investment ¢ Centrifuge Plant 1 eed Plant Metals Plant 'rotal Peak Construction Work Force 'l'otal Number of Men 'l'echnical Construction Manpower Man-Months Over-all Construction Time yr Operating Cost ¢ yr Centrifuge Plant Feed Plant Metals Plant Total Pnwer Required centrifuge plant Operating Work Force Centrifuge Plant Technical Total Feed Plant ' Pechnical Total only kw Metals Plant Technical Total C 125 566 000 3 168 000 439 000 129 173 000 37 141 000 1 710 120 34 108 oop 2 594 000 't I 1 040 80 540 l o 450 40 51 080 13 680 6 7 3 6 6 530 3 3 5 870 1 330 30 4 070 3 590 3 1 3 1 8410 3 654 000 521 000 161 000 4 336 000 2110 1 567 000 450 000 161 000 2 178 000 888 1 52E ooo 45c ooo 161 000 2 135 000 2520 39 990 16 3f 6 12· 142 12 121 54 8 3lf 8 34 6 6 33 33 2 2 8 8 184 8 2 2 8 163 95 8 81 11 598 000 651 000 161 000 12_ 410 000 10 49 8 4o • 374 ____ DELETED - L 5 875 000 2 244 000 439 000 8 558 000 h59 ooo -- -- _ ' ent-rJ fuge J ii el Identification 'i4 I K 15 27€ 000 2 31 1 000 h3 i _OOO 18 031 000 2 --- J 13 956 000 2 314 000 439 000 16 709 000 8 l Olf7 Tot_al Plant Work Force G _ _ 439 rx• 4 3 0 1 J 161 000 1 032 ocn 470 9 4o ·- - I t - 470 9 310 50 566 ooo h32 ooo 161 000 ° 1 159 000 -·--1-_ I - 5 2 lH u·JO 2 2411 n ' l i13' cou ·r 901 ooo L - J · I • ' JI- Pf t IJ TABLE - 4 GAS CENTRIFUGE PLl lf 1' SUMMARY FOR CLASS Z NATION 50 Kg U PER YEAR AT 90' U-235 Machine A Capital Investment ¢ Centrifuge Plant l 'eed Plant Metals Plant Total C r 6 889 000 2 631 000 514 ooo 10 034 000 2 631 000 514 000 9 263 000 2 000 140 78 550 8 8 1 210 90 21 040 4 8 630 530 4o 9 030 4 1 390 35 360 30 11 970 000 672 000 166 000 12 808 coo 8lno 3 771 000 538 000 166 000 li 475 000 2ll0 1 617 000 461 ooo 166 000 2 247 000 1 5 1 161 Feed Plant · Technical Total 12 5' Metals Plant Technical rl'otal Total Plant Work Force __ -- -- I I L 17 914 000 2 713 000 514 000 21 1li1 ooo Over-all Conscruction 'I'ime yr 1 _ C c - K 16 364 000 2 713 000 514 ooo 19 591 000 ConsLruction Manpower Man-Honths Power Required centrifuge plant only kw Operating Work Force Centrifuge Plant Technical Total __L_ 39 993 000 3 ot 2 000 514 ooo li3 549 ooo ll'otal NUJJ1ber of Men ' 'echnical '1 'ype 147 231 000 3 715 000 514 ooo 151 460 000 Peale Construction Work Force Operating Cost ¢ yr Centrifuge Plaut Feed Plant Metals Plant 'I'otal G ------ l 5 J 0 040 lt 3 6 c'Go 5 530 4 1 4 1 888 l ' 77 000 464 ooo 166 000 2 207 000 2520 585 000 446 ooo 166 000 1 197 000 1 70 19 382 14 167 14 1110 10 64 9 9 4o 4o li7 6 118 000 9 lr' l1 ooo i 4G ooo J 66 n O 1 oGf r · r 1170 10 117 7 7 39 5 1 2 2 9 2 9 2 9 2 2 1 227 9 h38 9 9 216 189 ll2 95 C e t e Model Identification DELETED _ _ J f ___V- _ j - J -- l '· o ta 'rABLE 5 GAS CENTRIFUGE PLANT SUMMARY FOR CLASS X NATION 500 Kg U PER YEAR AT 90'fo U-235 Machine T _c_ _ Capital Investment ¢ Centrifuge Plant Feed Pl ant Metals Pl ant 'J 'otal Peak Construction Work Force Total Number or Men Technical Construction Manpower Man-Months Over-al l Construction ' l'ime yr Operating Cost ¢ yr Centrifuge Pl ant I•'eed Plant Metal a Pl ant Total l K 1 05 827 000 3 941 000 51 8 000 110 286 000 36 973 000 3 448 ooo 518 000 lw 939 000 33 356 000 3 387 000 518 000 37 261 000 1 1 60 1 30 64 120 8 6 1 lf90 1 00 1 490 90 23 860 3 1 l 240 65 1 0 770 2 5 990 60 8 6 0 2 4 3 399 000' 919 000 566 000 4 8811 000 25 200 5 321 000 810 000 566 000 4 691 000 4870 2 522 000 791 000 566 000 3 885 000 291 0 24 070 5 1 8 626 000 903 000 566 000 10 095 000 8900 58 l 54l 26 638 29 718 13 249 182 10 59 1 0 53 10 53 10 50 10 49 3 16 1 616 3 16 3 16 3 16 701 787 315 3 1 6 2111 -- ' L L 96 782 000 3 869 000 51 8 000 101 169 000 21 250 Total Plant Work Force I 255 665 000 4 466 ooo 518 000 260 649 000 21 262 000 1 035 000 566 000 22 863 000 l'ower Required centrifuge pl ant only kw Operating Work Force Centrifuge Plant Technical Total Feed Plant Technical Total Metals Plant Technical Total ' r -·--- G 10 J DELETED · '--·----- - ------ ••• - - - · - - - · · --- --✓ _ ' __ I I - I 0 TABLE 6 GAS CEN'rRIFUGE PLANT SUMMA RY FOR CLASSY NATION 500 kg U PER XEAR A 90% U-235 Cnpltal Investment 1 entrif Plant Peed Plaut Metals Plant Total C $ l'csak Construction Work Force Total Number of Men Technical i Col struct ion Manpower Man-Months Over-all Construction Time yr Operating Cost ¢ yr Centrifuge l'lant l'eeu Plant -letals Plant Total f' wer Req_uired centrifuge plant only kw Work Poree Centrifuge Plant Technical Total Feed Plant Technical Total Metals Plant Technical Total I 127 675 000 11 754 000 625 000 133 054 000 44 606 ooo 4 160 000 625 000 49 391 000 l10 2lf 3 000 4 086 0 625 100 411 95 1 oo 1 880 160 106 380 12 0 1 790 120 39 930 11 3 1 800 110 1 190 37 930 4 2 1 500 80 17 870 3 4 1 1 3 22 284 000 1 084 000 593 000 23 961 000 21 250 9 o4o ooo 946 ooo 593 000 10 579 000 8900 3 562 000 963 000 593 000 5 118 000 25 200 3 480 000 849 000 593 000 4 922 000 4870 2 113 000 835 000 593 COO 4 071 000 291 0 70 1859 32 770 34867 16 300 12 219 12 71 12 64 12 64 60 12 12 59 3 19 1 949 3 19 853 3 19 950 3 19 3 19 379 297 Centrifuge Model Identification ------- _ - --- ·-·- DEL reo· ---------- - --··· ·rn 3 3 --- - ' -·-·- -------- - _______ '• _ L 1 16 763 000 4 667 000 625 000 122 055 000 - ---r-· K 308 41 9 000 5 389 000 625 000 314 463 000 tJp rA ilng Total Plant Work Force Machine T J' Ee G • -------------··- --·· -- _ ·-• - -- - - - _ ___ · f _ I-' ' TABLE 7 GAS CENTRIFUGE PLAN'r SUMMARY nlR CLASS z NATION' 500 Kg U PER YEAR AT 90i U-235 Machine 'l' lCEe 1p1tal Investme nt ¢ Cenl riflll3e Plant Feed Plaat • eta la Plant 'Potal __ L C G I K 361 667 000 6 318 000 7 53 000 3W 718 ooo 136 909 000 5 tf73 ooo 73 5 000 11 3 115 ooo 149 7o4 ooo 5 474 000 733 000 155 911 000 52 302 000 4 878 000 733 000 57 913 000 li 7 186 000 1 791 000 7 53 000 52 710 000 2 200 180 163 5' 0 15 8 2 100 140 61 400 5 7 2 110 130 58 330 5 6 1 760 90 27 480 4 5 1 400 80 22 020 23 000 000 1 119 000 612 000 24 731 000 21 250 9 331 000 977 000 612 000 10 920 000 8900 3 676 000 994 ooo 612 000 5 282 000 25 200 3 592 000 876 000 612 000 5 080 000 4870 2 728 000 862 000 612 000 4 202 000 2910 82 2 180 37 903 4o 1 01G 19 352 14 259 14 84 14 75 14 75 14 70 14 69 4 11 22 2 285 22 1 000 4 22 1 113 4 22 444 h 2' · ' 348 -·• 1k Construction Work Force 'l'otal Number of Men Technical Construction Manpower Man-Months Over-all Construction Time yr Operating Cost ¢ yr Centrifuge Plant Feed Plant Metals Plant Total Power Required centrifuge plant only kw Operating Work Force Centrifuge Plant Technical Total ' s j 1 - Z Jil i' Feed Plant Technical Total 11 3 i Metals Plant Technical •rotal Total Plant Work Force r - ··Centrifuge Model Identification I --·---c J- DELETED - ' ·- -- -- - •----------- ---- v • _ ___ - -- - ' ' ABLE 8 MINIMUM TIME REQUIRED TO PRODUCE FIRST NUCLEAR WEAPON Time Required years Class Country Assum E_tions Y X - 1 Knowledge of' model C details 1 · 41 t • i 2 No knowledge of' model C details 3 Knowledge of models Kand L details after they are developed ·X· I z - --• DELETED us - -P•- -- - -- C • ' ·-- · - Without details or considerable assistance from another count cy having details of the centrifue e model this class country would probably not develop the model on its own ---1r---- ' --···-·- ---- -· --· --- -- ---· --•--- - ---- -·· ---- t1'i D e i 1 '·J C _ LO ·- ' - ---------------- ---•• --•- • -- « •- P ·- -- --- -- - -· - ' £' - - _' ·--- -- J '0 '_ 24 PLANT DESCRIPTION CENTRIFUGE FACILITY Estimates of the capital investment operating costs construction schedule construction manpower and operating manpower req_uirements for each of the three processes of the centrifuge production facility are presented in Tables 9 through 25 based on estimated US costs Plants having production rates of 50 and 500 kg uranium per year of highly enxiched U-235 were investigated The 50 kg uranium per year product rate was chosen to reflect the production of at least one nuclear weapon per year The 500 kg of uranium per year product rate is approximately the largest production rate for which the cent rifuge plant might present attractive advantages over a gaseous diffusion ulant Six ba85C centrifuge models were studied for ea h of the above pr duction rates Centrifuge Models The details of the six centrifuge'models studied are presented in Table l the centrifuges can be grouped into the following three categories Present Machines These are centrifuges which have already been built and tested Although in all cases there does not exist as much experimental and operational data as'would be desirable for the design and construction of a centrifuge plant using these machines the separation performance of the machines is essentially known Fabrication of a large number of machines of the types in this group could be started almost immediately by the United States Two machines are included in this group These are ZIPPE CENTRIFUGE Zippe 3-in 350 m sec subcritical centrifuge with a separative work of kg u yr This centrifuge model is included in the study since considerable infonnation and reference is made to this centrifuge in the unclassified literature It is not felt that anyone would actually attempt to build a facility to produce highly enriched uranium using this centrifuge because o ' he high ca pi J'l Il 'Lo p_erating costs and _ -1 £€i' _ number of centrii'ug 9 _reg_uired-l -·-___ _ o 34 ° -j ' DELETED _ __ - - _ _- ---- The original Zipp •· - ntrifuge 'as con tZ-U ted at the University of Virginia consisted of a 3-in -dia extruded rotor which was rota ted at a peripheral ' speed of 350 m sec A schematic of the Zippe centrifuge is shown in Figure 1 The rotor was prepared from extruded 7075-'Ib 0 040-in -thick aluminum 25 __ __ __ PRODUCT FEED WASTE MAGNETIC BEARING DAMPING ASSEMBLY TOP SCOOP TOP BAFFLE MOLECULAR PUMP CASING I ROTATING BAFFLE _ f ARMJ ' 'JRE BOTTOM SCOOP MOTOR WINDING ' V _ ' NEEDLE BOTTOM DAMPING ASSEMBLY Figure 1 SCHEMATIC REPRESENTATION OF TYPICAL ZIPPE CENTRIFUGE 26 alloy was threaded at the ends and eq uipped with threaded aluminum end caps The bottom bearing and needle assembly and top magnetic bearing were similar in principal to those used in the current subcritical machine The top and bottom damping syste GJ s were not as refined as in current models but operated with similar objectives The rotor was driven with an axial air gap motor where the armature was a simple steel disk attached to the bottom of the centrifuge rotor A stationary scoop syst em was inserted in the center of the rotor for gas feed and gas removal pur poses While models of this Il9 chine were successfully produced at ORGDP and a number of units were operated in the experimental centrifuge cascade over a considerable period of time the characteristics of this centrifuge never were considered to be wholly attractive for cascading purposes owing to such factors as high inventory loss and nonreproducible separative performance For purposes of comparison however the Zippe centrifuge will be considered here as a base point in the evaluation of the centrifuge process The Zippe centrifuges produced for experimental testing at ORGDP and those tested by Zippe and personnel at the UniFity of Virgir i i s had separat i ve cal'§ cities in the rang of 0 34 kg u · DELETED RSSENT 6-INCII SUBCRITICAL CENTRIFUG i --·------- ----------·----- -·-- po6 b - q ·- 21 DELETED l ---· I Pro iosed Machines· These are centrifuges for which the separation performance has not yet been extensively measured experimentally In both cases some nechanical tests on the model have already been made These centrifuges are the logical descendants of the group 1 machines It is felt that there are no major obstacles in the path of the development of these machines and that sufficient information has been obtained on their forerunners to permit the prediction of their performance w1th reasonable accuracy The development and testing of these rrachines required to bring them to a production capability in the United St at A would be anticipated to be in the neighborhood of one to three years PROPOSED SUBCRITICAL CENTRIFUGE - - - 'k •--• • - Qo G- DELETED _ I • _f5ooling 'be achieved-eitherwith a circulating coolant system or an evaporative cooling system using the atmosphere as a heat sink - -- -- -------- __ -·--- __ __ -·--·-----·-----------·--- DELETED t C _ ·-· ------- --· PROPOSED SUPERCRITICAL CENTRIFUGE ·-- ·---- z I t € I ft7 ti DELETED ' -- ---- _ --1 A sche c Ja t C of tfrfs-- proii'oie l - supercritical centrifuge is shown in Figure 3 7ne unit wilJ_ be driven from the closed end of' the bowl by a di LE_c_ u f_ling with a conventional·------- sc uirrel-cage i _ µ9-µction motor •' tJli6 --·-·· DELETED J ·· 28 PROCESS GAS LINES MAGNETIC DAMPING UPPER ZONAL CRASH SHIELD TOP BAFFLE --- ___ CENTRIFUGE BOWL VACUUM LINE BOTTOM BAFFLE BOTTOM DAMPING SYSTEM Figure 2 SCHEMATIC REPRESENTATION OF THE PROPOSED SUBCRITICAL CENTRIFUGE f ' · ' -- a ' ' ' - - -·- '1'E ·c · ·-- f·' c ' -tJ c · l p I fr I ' I '' I ' ' 'i oo 3 2 t7- DELETED I I I I • I I I i• I 'I I i' l 30 I · DELETED magiietfo bearing is employed at the open end of the rotor A servo -sys em which will automatically maintain the appropriate gap between the rotating and nonrotating magnetic portions of the bearing will be used to regulate· the magnet gap spacing A conventional Zippe-type gas extraction system will be incorporated in this machine being supported from an assembly located at th _ open end of the rotor U·--- ooc_ 6 I l DELETED Owing to the greater length of th rotor and h igher···pe1 -ipheral sl eeds gr a er se parati ve C E§l Ci ties will be -- available ETED --------- through -·-- the use of this unit Hff • DEL J -- ' - --- - - - ---------•-•J ---•'------··------------- - - - - - l 'uture Centrifuges The development of these machines is dependent upon appreciable advances in technology and it would be ex pected to require 5 to 10 years before these machines would be ready for plant installation Two machines are included in this group They a e FUWRE SUJ3CRITICAL CENrRIFUGE At this time there appears to be no basic limitation on the diameter of subcritical mac ires the optimum size will be established by economic considerations relating to the cost of building such machines their operating costs and their se parative capacities Research currently being conducted by the various glass filament manufacturP rs seems likely of developing higher strength glasses Moreover winding techniques undoubtedly will be developed to utilize higher fractions of the av · able stren h of' th gl ss filaments in the centrifuge rotor DELETED - ----- _j 1 FUTURE SUPERCRITIC AL CErrrRIFUGE _____ -·-·- ---· -·· '' · -· - ' - --· · 1 DELETED I i ·---·- ·-- _ _ _ __i --------------------------- It would z 31 ' not appear unreasonable that such a chi e to Y § r ELo_f__ de JJ elapme_nt - ffo rtf o _ be · vai_ fu with fiv DELETED -•-••------- '-------•------ N-- •·••----CENTRIFUGE PIANT CHARAG'IERISTICS The centrifuge is characterized as a high-enrichment low-throughput device In a plant design the centrifuges are connected in parallel to obtain the·desired throughput and these parallel connected ba o ks of centrifuges are then connected in series to obtain the desired over-all enrichment The tails concentration that i1 the degr_ee of stripping would depend upon many factors such as the cost of feed_ equipment costs operating costs etc An estiil19 te was made of the amount of stripping which would probably be attractive with the various centrifuge model and the tails concentration associated with this_ stripping is indicated in the Tables of Centrifuge Costs 9 through 19 --- The amount of stripping varied from very little tails concentration 0 0065 for the Zippe model to a considerable amount tails concentration 0 0041 for the future supercritical centrifµges operating in the larger plants _ Ideal plant tapers for those plants studied here are shown in Figures 9 through 14 These figures show the number of centrifuges in parallei for each stage of the plant along with the number of stages length required In describing a centrifuge cascade for the production of U-235 it is probably best to discu ss at length the individual factors which are involved in the layout and design of the plant The following paragraphs will describe the main items of interest PIANT IAY0UT In order to reduce capital operating and maintenance charges the centrifuges in a production plant will be pipea · ·i wired so that vacuum electrical instrumentation and control equipment can be shared by a number of machines The smallest grouping of Il19 chines will be referred to as a cell A description of a ty pical centrifuge cell follows Centrifuge Cell There will probably be two ty pes of cells in the centrifuge plant one will be the parallel cell the other the series cell The two cell ty pes are described below ----------------------------------------·-•-'---- 32 Parallel Cell A group of centrifuges operating in parallel valved wired and instrumented so that the entire cell of machines must be started operated a d stopped as a unit The cell would be the smallest grouping of machines which could be isolated from the cascade A schematic of one-half of a ten-machine parallel cell is shown in Figure 4 A complete ten-machine parallel cell layout is shown in Figure 5 As shown in Figures 4 and 5 the ten machines share common vacuum and process piping The cell vacuum piping can be isol §3 ted frpm t rner vacuum fJCJF s stem by the automatic block valves shown 6_ Gi '1 DELETED volume of the parallel celis-Ts estimai3ea fo be sufficiently J ·ge what failure of a single machine will not release sufficient gaseous inventory to cause damage to adjacent o perat ing r P nt r1fue es It is asst' led th t the su pcrcritical machine will also be capable of tolerating modest' pressure excursions After a cell has been initially evacuated and is ready to be placed on stream the vacuum valve is closed thus isolating the cell from the n ain vacuum system The process valves are opened in appropriate seq uence and the ma chines placed on stream The pumping action of the rotating centrifuge bowl will tend to preserve the initial vacuum conditions Should one of the machines in this cell develop trouble the appropriate alarm would be given in the control room and the process block valves would close isolating the cell from the cascade and minimizing the loss of process inventory If this particular cell was being used as a purge station the vacuum block valve would remain open during normal operation and would be automatically closed in the event of trouble Series Cell A group of centrifuges opera ting in series valved wired and instrumented so that the entire cell must be started operated and stopped as a unit The difference between -the parallel and series cells is that the parallel cells contain centrifuges all operating in the same cascade stage at the same gas concentration while the series cells contain stagesff of centrifuges each handling a higher concentration than the centrifuge stage below A schematic of a ten-machine series Centrifuges operating in the same cascade stage each being fed gas of the same concentration and each producing upflow and downflow streams of the same concentration A stage in a centrifuge plant would consist operating at the same concentration level large centrifuge plant the feed stage could centrifuges operating in parallel while at small plant the product stage might consist of centrifuges all At the feed point of a consist of hundreds of the product point of a of a single centrifuge -VACUUM BLOCK VALVE CELL VACUUM HEADER CENTRIFUGE CASING 1 Figure 4 ONE-HALF OF A 10-CEHTRIFUGE PARALLEL CELL ' I 4' I ' J · 7 I 7 I 2' I 2' 7' • I -c ' I ' ' I 2-1 2 FEED 4 FEED HEADER FOR SUB GROUP I I 14 VACUUM HEADER FOR SUB GROUP w -i -- I - ' 3- l 2 WASTE HEADER FOR SUB GROUP 3-1 2 PRODUCT HEADER FOR SUB GROUP PROCESS PIPES - STACKED Figure 5 10-CEHTRIFUGE PARALLEL CELL _ t 3'5 cell coJtaining one centrifuge per stage is shown in Figi re 6 The ten mach nes share a commcn vacuum piping but each machine has i dividual p ocess pip ng in that it receives feed from both the machines above and below it in the cell and sends a product stream to the centrifuges above and a waste stream to the centrifuges below It may be necessary to have small automatic block valves in the process lines connecting stages in the series cells to avoid a buildup of light gases in the tc P stages -when the series cell is iso ated from the cascade The buildup of light gases could cause excessive pressures which might lead to the destruction of the centrifuge For the economic evaluation presented in this report these block valves were not included but should they be needed the effect on the unit cost of separative work would be very small Each stage of machines has a flow metering device on one of the exit process streams The machines share a common process control valve at he top na nottom of the cell which establishes t e desired flows into and out of the cell The vacuum system can be isolated from the pumping system by use of the autonE tic block valve shown The process system can be isolated from the cascade by the automatic block valves at the top and bottom of the cell When a series cell is isolated from the cascade the cascade is either split at that point or the cell is by-passed hence the valves and piping necessary for both of these operations are required In order to obt in additional economies groups of parallel ocells will probably be piped and instrtrmented to share a common vacuum instrumentation and process control system wherever possible Such oa grouping of parallel cells will be called a sub-group Series cells because of the concentration range spanned in the cell and the necessity to minimize roiY ing of assay when trouble arises probably will not be grouped into subgroups A description of a typical sub-group follows Sub-groups Groups of parallel cells operating in parallel which share a common vacuum and process control system Figure 7 shows a schematic of a sub-group which contains ten parallel cells The sub-group has a common vacuUlil system consisting of an appropriate size diffusion pump cold trap and a rough pump vacuum valves and a properly sized vacuum header The sub-group also has a common flow control system which regulates the feed pro·Juct and waste flow out of the sub-group process headers Eac•' cell as st 3 earlier can be isolated from the vacuum and process pipi nj of the rest of the sub-group Each sub-group in turn can be isolated forom the process system of the rest of the cascade stage and the flow by-passed by the three process block valves shown at the end of the sub-group in Figure 0 7 The number of centrifuges connected in a cell and the number of cells connected in a sub-group will depend upon IT any factors such as machine reliability cost of vacuum pumps cost of vacuum and process hardware separation factor of the individual centrifuge etc The larger the cell and sub-group the greater the capital savings but the larger the losses in separative wor•k will be when a centrifuge or an auxilia S l € te l l _g ---- trouble req Jcrtng the shut do m of a cell or sub- e ED J '-t r'io£ WASTE FROM STAGE ABOVE PRODUCT FLOW TO STAGE ABOVE - PRODUCT FLOW FROM STAGE BELOW VACUUM BLOCK VALVE l CELL VACUUM EADER i CENTRIFUGE CASING - NOTE ARROWS INDICATE FL W DIRECTION ll3 lCltll l PRODUCT FLOW FROM CENTRIFUGE WASTE FLOY FROM CENTRIFUGE FEED FLOW TO qNTRIFUGE igj' '· Figure 6 10-CEHTRIFUGE SERIES CELL OHE CEHTRIFUGE PER STAGE I ··i a 39• Z f 1 I L______ 1 1 L ____ J II TO STACK VACUUM HEADER - 0 ------- · - --------7 I I I l_ _______ - FORE PUMP __J I - --1 I _ DIFFUSION PUMP --- _TRAP J -WL _____ J TYPICAL VALVll G AT VACUUM Pl MP DIFFUSION ------- I ' PUMP I II 1- I_ _ _ _ _ _ _ _ _jI 2 FORE PUMP __0 I- v 6 0 O _ TRAP I -------7 I I -- i I 1--- --1 I I _______ 0 - UJ• - i ' It ' -- · ---- I I I -I ________ I E V 1a I I 2 PRODUCT 2-1 2 FEED 2 WASTE 14 VACUUM IE -- SEE DETAIL A -- - TYPICAL 10-CELL SUB CASCADE DETAlL A Figure 7 TYPICAL 10-CELL SUB CASCADE 4 PRODUCT - - t 6 FEED 4 ' ' 'ASTE TYPICAL PIPl1'1G VALVING AT POINT WHE E THE SUB GROUP PIPING TIES INTO STAGE PIPING T r •• · f rv- f a$ 38 L DELETED 1 • ' i P a ··7 These reliability predictions were - eT LminaryccSst--e-st±ma for various cell and sub-group layouts to arrive at a reasonable economic cell and sub-group size A reasonable first estioate of cell size would be about ten achines for the parallel cells and anywhere from a few up to ten stages for the series cells depending upon the number of centrifuges per stage and the separation factor of the individual centri 'uges At the feed point of very large plants where r any hundreds of centrifuges must be operated in parallel to supply the large interstage flow requirements of the feed stage sub-groups in the range of ten parallel cells i e 100 centriiuges seem reasonable For smaller plants and for stages far removed from the feed poi t where the interstage flow rates are lower smaller sub-group sizes will be desirable The use of series cells will be desirable as the product or waste point is pproached since in thece locationc i --r u J '---' L U J J C L a J c J C½ _U L Lt U to supply tbe necessary interstage flow rates Po-ry -i - i - o -c LUt 111 L 1- i 1 L t J -- - --- - - - - - • - __ ELECTRICAL EQUifl 1 ENT Process Electrical Equipment Motor Drive System The centrifuges will be driven from a high frequency power system supplied by alternators The machines must be started and brought to operating speed from a lower frequency start-up system to minimize overheating during start-up In the plant the electrical drive systems will be divided into appropriate size units where each u it will contain the necessary drive alternators including spares and the necessary start-up equipment for a block of centrifuges The number of sub-groups included in the block -will depend upon the plant size power required nranber of machines operating in parallel etc The size and number of alternators requir d for an electrical unit will vary with plant size and stage size A typical high frequency electrical distribution system is illustrated in Figure 8 Figure B also shows the electrical distribution system for the auxiliary equipment lighting etc Centrifuge Motors Each centrifuge will be driven by its own individual motor which in the case of the subcritical machin s will probably ea hysteresis-synchronous axial air gap motor with th· rotor consisting of a steel disk attached to the bottom of th$ centrifuge bowl In the case of the supercritical machi es the motors presently evisioned will be a conventional squirrel cage induction motor coupled to the centrifuge bowl by means of a drive shaft Distribution System The alternators will be connected to busses to equalize loading with SJ E re alternators being available to take up the load should one of the alternators fail The power busses will be connected to the individual sub-group control panels which will serve as feeder stations for the individual centrifuges Each centrifuge will be supplied with an ammeter and fusing protection for the motor Switchyard Above certain size electrical loads it will become necessary to provide switchyards to step down the voltage from the power distribution network In order to have a reliable power supply the switchyards should be double ended t o-I§ DELETED ' i'' I J 4o Protective System The high freq_uency power supply will probably have the following protective equipment A light to indicate that a motor fuse has burned out an overload protection for the alternator which -will cut out the alternator if the current gets too high a power failure alarm Auxiliary Power Supply It will be necessary to supply power and distribution networks for the vacuum pumps instrumentation isolation valves lighting crane operations welding machines power tools etc The costs of the over-all electrical system will depend upon the size of the plant the power req_uirements of the indiyidual centrifuges and the arrangement of centrifuges in the cascade For supercritical machines a positive source of power must be available to actuate the drive bearing constraint and to ve ry the m 8 enet ge p cont-roller s iI1ce means are needed to pe it cc fe deceieration in the event of a general power failure A source of DC provided from storage batteries appears satisfactory for this purpose PROCESS PIPING The process piping to each centrifuge will consist of a feed line a product line and a waste line The line sizes to the individual machines will be quite small and will probably not exc ed about 1-in -dia The lines from· individual ma chines in the parallel cells will tie into header lines for the cell which will in turn tie into headers for the sub-groups which will tie into headers for the stage The pro ess lines in general will be fairly small probably not exceeding 12-in -dia for the largest piping inter-connecting sub-groups or stages By-pass piping and valving isolation valving and control valving will be required for each cell and sub-group While the pipe sizes will in general be quite small the vacuum requirements will be stringent and careful assembly and extensive leak testing will be required A flexible coupling to tie the actual centrifuge scoop system to th process piping will probably be required to allow easy installation and removal of the· centrifuges The process piping will probably be Niplated ·steel to assure low consumption and freedom from corrosion products which could plug scoop lines and cause other troubles with the centrifuge operation The piping must be sized so as to keep the pressure drops very low since_ the available pressure from the scoop systems is 10w and it is necessary to keep the feed pressure quite uniform on all the entrifuges in a stage to assure uniform feeding to the individual centri£uges PROCESS CONTROLS It will be necessary to control the process flow rates in order• to assure optimum and stable plant operation It also will be necessary to monitor the speed and temperature of the centrifuge bowl and the vacutnn jacket pressure in order to assure safe constant performance from the centrifuge In a large centrifuge pl ant due to the very large number of centrifuges required the design criteria which must be adopted for the control and instrw nentation ystems are those of minimizing the actual operating information obtained and- recorded for individual machines which are performing satisfactorily • Since the temperature and speed of the centrifuge bowl and - - 41 the pressure of the vacuum jacket are not controlled in the usual sense it is desirable to adopt a supervisory type of control for these variables This supervisory control would initiate protective or corrective action only if certain safe limits are exceeded Thus the variables of speed and temperature need only be scanned routinely and if the values are within control limits there is no need to record the actual values When deviations from acceptable limits do occur it is only necessary that suitable alarms be activated An increase in vacuum jacket pressure above a given level will be interpreted as a machine failure and the necessary instrumentation alarms and recording equipment must be available to innnediate y detect trouble sound alarms and isolate equipment Equipment must also be available to measure and record the centrifuge bowl speed and temperature for trouble shooting individual machines when they exhibit troubles or when a cell of machines is bein·g started or stopped and it j s necessary to observe carefully the machine operating conditions This use of supervisory control will permit important savings in instrument costs and also in operating costs since it results in a central control room equipped with simple alarm instruments which require little maintenance are reliable and easy to interpret Process Gas Controls The centrifuges will be fed by a single feed tube at the approximate center of the bowl A scoop system will remove the product and waste stream from the bowl through individual lines The individual ceni rifuge feed product and waste tubes must be Aesigned and manufactured so that they present uniform flow restrictions to the process gas and thereby act as flow regulating devices assuring uniform and predictable flow into and out of e ch machine in a given cell A control valve will be needed between the product header of a sub-group and the product header for the stage to control the fore pressure of a critical metering orifice The control valve will be operated and controlled pneumatically to regulate the product flow from the sub-group 'I' his will control the cut of the sub-group and provide a means of damping pressure disturbances in the cascade A ma riual valve will be needed in the waste header of a sub-group to maintain inventory levels in the sub-group at desired levels Once this valve is set for a particular steady state operation l tttle additional adjustment is anticipated Automatic block valves will 012 available in the feed waste and product headers for each cell In the event a machine fails in the cell these valves will immediately close and isolate the cell from the cascade No difficulty is anticipated for the remaining cells as the feed rate will simply be increased typically by 10 - 25% per machine This of couxse will result in non-optimum machine performance for the period that the cell is off stream but assures continuous plant operation For the parallel cells and subgroups control valves will be used between the waste header of one stage and the feed header of the stage below and also between the product header of one stage and the feed header of the stage above For those portions of the plant i e near the product and waste where the series type cells are used control valves and metering orifices will be used at both ends of the cell to maintain the desired upflow through the cell Limited tests p rformed in the ORGDP 35-machine centrifuge facility have indicated that this type of control system should perform satisfactorily ----·-·-·-··- -----------------•• ______________________ T 42 Centrifuge Controls Speed The speed of the centrifuge will be controlled in principle by the alternator frequency The alternators are driven by induction motors which are slip controlled with load and motor size so that the alternator frequency is reasonably constant Each machine will have a speed pick-up probe and the speed will be automatically monitored on each of the machines of a sub-group every few minutes The speed will not however normally be recorded or indicated but will be automatically cpecked against the design speed If the speed is more than a set amount from design an alarm will sound in the central control room and indicator lights will be turned on at the cell to indicate the offending cell The ·process operator will then check the equipment locate the offending machine and take the necessary corrective action The necessary switching and recording equipment will he available to allow switching to any machine in a sub-group and monitoring and recording its speed to locate troubles and to follow the speed during start-up operations Bowl Temperature A thermocouple will be installed on the scoop syste to provide an estimate of the bowl temperature As w-ith the speed the temperature will be monitored automatically on each machine of a sub-group or one instrument might handle several sub-groups every few minutes The temperature will not be indicated or recorded but should the temperature of a centrifuge indicate above a pre-set upper limit an alarm will sound in the control room The necessary switching and recording equipment will be available to check the temperature of the machines in the offending cell and to permit trouble shooting in order to decide what action to take Vacuum Pressure A sensitive Pirani-type gauge will monitor the pressure in the vacuum header of each cell Should this pressure exceed a pre-set limit indicating inlea kage or machine failure an alarm will sound in the control room and the process and vacuum isolation valves will close iso1 ating the cell from the sub-group process arid vacuum system The cell can then be checked to determine the cause of troubles and appropriate action taken Other Controls In addition to the above instrumentation controls and alarms will be needed to handle the failure of power coolan pumps vacuum pumps centrifuge motors and for isolation and by-pass of sub-groups during periods of troubles PROCESS AUXILIARIES There axe many auxiliary systems needed in the centrifuge plant important are listed and briefly described below The more Vacuum System In order for the centrifuge bowls to rotate at the high speed desirable for good separation performance it is necessary that a goo·d vacuum be maintained in the volume surrounding the centrifuge bowl i e 0 1 to 1 0 µ depending upon speed size etc Properly sized diffusion and rough pumps must be supplied to evacuate the centrifuge jackets s_o that start-up operations can commence Once he machines are at operating speed the centrifugal action of the centrifuge itself tends to maintain the necessary vacuum environment The vacuum system must be large enough to allow for fairly rapid pump-down of the system to reduce the start-up time and the system must be tight enough to eliminate problems due to in-leakage of atmospheric gas to the system The diffusion pumps must be protected by cold traps to remove any process gas which may be in the vacuum system All vacuum piping will be nickel plated to minimize corrosion and improve the pumping characteristics of the system Cooling System The normal heating due to the inefficiency of the centrifuge motor and the heating produced by the drag of the scoops and of the molecular pumps necessitate the removal of heat from the centrifuge S the centrif g _ l o erSCes in I DELETED this heat is rather har l to _remove 's ·1 A circulating oiT_ system which wilr Coor--tRer cefrtri fu ge motor · 'Sow · an ad amping system and also supply oil for damping in the supercritical machine is presently being used The oil is pumped through a water cooler and then returned to the centrifuge The use of a self contained evaporative coolant system for the subcritical units is presently in the early design stage In the evaporative system the heat at the motor and bowl would be removed by evaporating a low boiling liquid such as ethyl ether The vapors would be cooled by radiation to the outer centrifuge casing be condensed and return to be evaporated a gain Such a system would eliminate the external piping pumping and water cooling system and as such should be a simpler more economical system Lubricating Oil System Subcritical Centrifuge In the Fu bcritical centrifuge the bowl rotates on a very small diameter needle shsft which rotates in a hemispherical cup in a bard metal support plate This simple bearing system' is filled with oil at startup and indicatiolrn are that it can operate for extended periods years with no oil additJon or maintenance Therefore t 1ere is no lubrication system as such for the su1Jcritical machines Supercritical Centrifuges 1he more elaborate bearing and damping system of the presently proposed supercritical centrifuge requires the use of a high pressure oil system for damping and the removal of a sizeable a mount of heat from the damping and drive systems It is expected that in a plant the lubricating oil system and the cooling system would be combined in a single system dNJt ' Process Gas Purge System Experience with operation of the present ORGDP 35-machine cascade has shown that lights will be passed through the cascade by the scoop system Provisions must be ma de to remove any in-leakage at various stages in an operating plant to prevent undesirable pressure buildup The vacuum system which is installed for start-up of the centrifuges will also act as the purge system The volume surrounding the centrifuge bowl below the molecular pump which will keep most of the heavy UF out of this volume can be 6 pumped when needed with the existing vacuum equipment during operation to remove the lights from the system The volume of the vacuum system of each sub-group is sufficiently large that the rise in pressure due· to the failure of a single machine will not cause damage to the other machines in purge cells connected to a common vacuum system nor would the resultant pressure surge overtax the vacuum equipment Process Gas Feed Waste and Product System These could be similar to the present facilities installed in the UFh gaseous diffusion plant Controls metering equipment heating facilities and refrigeration equipment would be required Site Auxiliaries These items would include roads parking lots security provisions fire protection maintenance and assembly shops administration buildings laboratories cafeterias dispensaries etc The size type and cost of these facilities will depend upon the size and location of the particular plant being designed Process Building The process buildings would probably be constructed of concrete block The load levels will be quite small so that a minimum of reinforcing and heavy structural steel will be required The centrifuges and their utilities would probably be installed in one level with a gallery above for machine installation and removal operations The wachine casings couid be supporteo by steel fra11 1--works as shown in the partial cell schematic Figure A-8 Sufficient overhead clearance will be available to allow easy removal of the centrifuges from and installation into their casings The process buildings will include central control rooms offices rest rooms storage areas some maintenance facilities in addition to the centrifuge cascade areas For the large plants a total floor area per centrifuge is estimated at approximately 20 square feet The process buildings would be heated and ventilated as required for the health and comfort of the operators and maintenance force - T 45 OPERATING COSTS Power Costs The annual power cbarges for a centrifuge will consist of centrifuge drive power 1 lighting and maintenance power and centrifuge auxiliary power The centrifuge drive power will consist of the power necessary to drive and control the centrifuge bowl at design speed power needed to overcome·the gas drag friction bearing load damping load etc This power will be determined mostly by the centrifuge model i e size speed damping characteristics vacuum system etc The lighting and maintenance power will represent power necessary to assure adequate lighting for the process building maintenance areas assembly areas and a ninistrative areas This power will also include the power necessary to carry out the plant maintenance requirements i e welding power power for cranes and lifts power tools etc This lighting and maintenance power will depend upon the plant size and the number of centrifuges in the plant cells and_sub-groups The auxiliary power includes the pow·er required to operate the necessary auxiliary equipment such as vacuum pumps coolant pUillJ s refrigeration ·etc The auxiliary power will depend upon the plant size and the number of centrifuges in a cell and sub-group Operating and Maintenance labor Costs At this particular stage of the centrifuge development estimating the required maintenance and number of operators needed for the safe efficient operation of a gas centrifuge plant is extremely difficult Estinates obtained now must be taken as very preliminary and subject to change It is felt that the costs presented here reflect optimistic thinking and future costs are not eXJ ected to be lower Material Cost The -cost of worked materials was based on • he assumed failure rate overhaul requirements and routine maintenance requirements The estimated cost is not considered very reliable due to the lac of actual operating eXJ erience with the centrifuges This cost for any practical centrifuge plant will probably be low· compared to oth7 costs and therefore should not have much effect upon the unit easts cost of' $40 per machine per year was assumed for tl le subcritical cent r ifuges and $100 per year for the supercritical centrifuges as a minimum cost for a large installation i e 10 000 centrifuges For smller installations the costs were scaled based on ORGDP experimental centrifuge cascade eXJ erience 't 46 FEED PIA NT The amount of stripping performed in the plant will affect·the amount·of normal assay feed required for a given pr6duct rate and concentration The various centrifuge models and plant sizes considered here require an estimated range in desired feed rate from 17 5 tons per year to 700 tons per year Because of this wide range in desired feed rate 'the costs of con truction and operation of feed plants with capacities of 25 50 100 and 500 tons per year were estimated and are presented in Tables 20 through 23 The costs for feed plants within this range were obtained by interpol a ting between the estimated costs and those slightly outside this range were obtained by extrapolating these estimated costs Briefly the process which would probably be used can be described as Colluws The ore concentrate assaying appr oximately 60r o u raniUl l is treated with nitric acid and the uranium is dissolved 1 2 The· uranium is purified by solvent extraction using tributyl phospbate in kerosene as the solvent Extraction scrubbing and product stripping are accomplished in glass and stainless steel columns Stainless steel mixer-settler equi Illent is employed to clean used solvent and to remove traces of solvent impurities from the product uranyl nitrate hexahydrate solution Product concentration to about 10 pounds of uranium per ·gallon is carried out in vertical tube evaporators and boildow tanks heated by steam coils Solvent extraction raffinates are neutralized and disposed of in seepage pits Nitric oxide fumes from the denitration step are piped to the waste bancll ing area and are neutralized with caustic in a scrub tower however for the 500 tons U yr case approximately 50% of the nitric acid may be recovered by contacting with water in a vertical tower no For the 25 50 and tons yr cases the uranium trioxide is fluorinated dir ctly to uranium hey a fluoride with elemental fluorine using a flame reactor 3 U o reacted rnaterrar r la tively_ DELETED __ bk w Rf · 6JL 'Rohigh in beta radioactivity is caught in an ash receiver bolted to the bottom of the reactor When an ash receiver is filled it is replaced and held for about three months to permit decay of the uranium daughter products of short half life The ash is subsequently ground and refed to the fluorinator A small amount of ash is also collected in the fluorinator off-gas filter systemj this material contains the largest part of the nonvolatile impurities and is accordingly reprocessed through the ore concentrate solvent ext raction system The uranium hexafluoride is separated from the product gas stream by· batch cold traps and the uranium hexafluoride that is collected is drain ed as a liquid to cylinders All items of equipment and piping in contact with uranium hexafluoriae are constructed of Monel The oxide hopper and the rotary dispersers are made of steel For the 500 tons U yr case it is probably expedient to reduce the uranium trioxide to uranium dioxide with hydrogen and to convert the uranium dioxiue to uranium tetrafluoride with hydrogen fluoride prior to treatment with fluorine To accomplish the reduction and hydrofluorination two-stage fluid bed reactors are used for each step with hydrofluorination reactors being of the stirred fluid bed variety uo 3 li2 uo2 4 HF DF4 F2 uo2 • • • UF 4 rr2o 4 2 H 0 2 5 6 UF6 Fluorination is carried out in a flame reactor and ash handling is as described above for the other cases Except for the mild steel feed hoppers reduction equipment -- piping valving etc -- is constructed of stainless steel Bydrofluorination reaction equipment is made of Inconelj all hopper'• _ screws filter tubes piping valves etc are Monel Fluorine may be produced from 6 000-ampere Monel cells and hydrogen fluoride is removed from the fluorine stream by cold traps followed by sodium fluoride Hydrogen is generated from cracked ammonia Hydrogen fluoride could be bought in 6 000-gallon tank car lots for the 500 tons U-yr case 200-pou d cylinders should suffice for the other three cases METALS PIA NT Two different size metals plants were estimated here one to handle 50 kg of metal per year the other 500 kg per year The plantJ capital operating and maintenance costs are presented in Tables 24 and 25 Briefly the process which probably would be used can be described as follows 48 ··f Casting skulls will be burned to oxide leached extracted and returned to wet chemistry for preci itation Machining chips will be recycled to reduction All massive metal will be returned to casting The wet chemistry and reduction salvage will probably be discarded TABLE 9 EST IMA'rED MANPOWER AND CAPITAL FOR CENTRIFOOE PLANT CONSTRUCTION AND OPERATICN IN U S Machine Identification A Original Zippe 350 m sec Product Rate 50 Kg U pe r year Capital Cost Centrif'- ge Electrical Casing 6 831 000 20 434 000 7 861 000 9 464 000 8 395 000 Instrumentation Mechanical Building 52 985 000 21 1 86 000 1 762 000 88 427 ood Test and Start up Total Capital Cost Operating Manpower and Costs Direct Operating Labor Direct Maintenance Labor Total Dil ·ect Labor Overhead at 1ooi Works Laboratory Technic ans Technical Supervision R j Tecimlcal and Scientific Staff Total Labor Materials Power Peak Manpower Men Engineering 'fon Manual funual 72 J 40 860 Tot il Manpower il ngineering Total Direct Construction Cost Indirect Construction Cost v1·g· i'J Construction Manpower and Scheduling 12 494 000 ifon Manual l anual Man-Months 1 780 2 310 14 450 Min tmum Time Reg 'd I months _j Year 335 335 2 97 000 2 97 000 4 i94 000 4 ' i9l• 000 6 110 000 l1 _32 000 120 000 1 0 697 ¢ Year fl 9 h80 ooo 'i92 000 68 000 Total Mate rial s 1 160 000 1 0 640 000 Total Operating Cost No of Centrifuges Feed Rate tons rr 19 800 70 Tails Concentration U-235 0 0065 115 t C a E I I I i I I t ------- ' · ' TABLE ll ·'GT1Mfl1'Eli • • · WEH AND Ci l'l'rAL FOR CENTRIFUGE PLANT CONSTRUCTION AND OPERA'l'ION Ill U S • ____ ·- - ---- - ------- ----- I i I i I i -I 1 iIv I-' I l ' '•' _ ·1 i ' _ __ - · - _ _ ___···- ·--------------- - - ------- 1 ------ TABLE l2 ESTIMATED MANPOWliR AND -·- - - - -· CErITRIFUGE PLANT CONSTRUC'rION A fD OPERATIO0f JN TIIE J l - -•-- I I I '1 i _ f ' - t Jl J c· -- ' '· l - ---·· ·--·------ -- - •-W-' - - - - l i J TABLE l3 _ ES'l'lMATED MJ N AND CAPITAL FOR CENTRIFUGE PLA N' r CONSTRUCTION ND OPERJ TlC'N JN -1J S - ------------------- -·- · _ _ ___ lWER _ ____ _ ---- --'----'--- __ --- - ---__ ------------ Ii I I ' I I Jl w iJ I ' l l '· ---- __ --··- - - --·-· ·--- - - ·--- _ _ _____ _ - --· _ _ ' --· s to -------------- ' TAJJLE l4 -------- - ------ 0 - _ 1 __ CAPITAT FOR r E J P _ON 3'l'Rl' C C N wO ' RAT -S - -----· I 1 I I 1 J-' - - t J •cJ· f ' - aJ 1· ' •·-----·------------·---- _ t ' · ------- l --- - __ _____ ·- 'rABLE 15 ESTIMATED MANl'ln L R Al'ID CAl'ITAL F'OR CENTRIFUGE PLAN'r CONSTRUC' l'ION AND OPERAT JN HI 11 f -- cc -· · • - - • _ --- -'- ---- -- --- f I' I I i r t I 1 I I ' i i l o ' ' 'a '_ · - ' t - l 1i fof' l I ' i I I I _____ -- ' ·------ -- i ·---- TABLE l6 ESTJMATED MANPO'•r R AND CAPI'l'AL FOR CENTRIFUGE PLANT CONSTRUCTION AND OPERATirn TN U S --• ' ------ - -- - • -r • c - - ' ' _ ---- ' •-M•• ' II i I i l · r ' ·l ·- i IJ ' ·· · j •' ° I 1 ·--- ---- __ i 5 ' ·' TABLE 17 - · ---- ____ _ ESTIMATED '' AND DAPITAL_ro '' ' PLM T DOlffiTROCTION m DPJlRATIO Ill •' I l l ----· 'LJ1 T-- I I t i i I I • I i ' I I I i I - -·- ---·------- · _ i'ABLE l9 ------------ _ Gl'lMA'PED M NFDWF f1 AND CAPrrAL FOR CEN'rRIFUGE ·------ _ rwrr - - '• CONSTRUC' f'ION AND OPERATirm l1I u s '• -- -- - - 'I i f 'tiI 4 f J1 JJ ·11 'l I _ ' ' G Ji $ b _QJ i ' i t II ' __ ·····•• -N ·---- - ___ - ---··- -- -- - --- _ ' --- _ ----- --- '' 60 TABLE 20 ESTIMATED MANPOWER AND CAPITAL FOR FEED PLM'T CONSTRUCTION P_l'ID OPERATION IN U S Feed Rate 25 TU per year Capital Costs Dollars Construction Manpower and Scheduling Installed Eg_uipment Pea Manpower Men Solvent Extraction Fluorinator and Cold Trap Fluorine System Auxiliary Chemistry Systems SP3-re Equipment 133 000 73 000 106 000 18 000 65 000 Piping Instrumentation Utilities Maintenance Facility Administrative laboratory Building 210 000 58 000 33 000 32 000· 36 000 209 000 Total Manpower Direct Construction Costs 973 000 Time Required Start-Up Engineering Design and Inspection Indirect Construction Costs Contingency 99 000 126 000 189 000 194 ooo Total 6 Engineering Non Manual Manual 9 72 Man-Months 6o Engineering Non Manual Manual 90 690 ll Months 1 581 000 Operating Manpower and Costs Dollars Per Year 5 days week 24 hours day Ls bar 9 Chemical Operators 9 Maintena ice l Supervis ·· l Clerk Overhead at 100% 2 Laboratory Technicians including overhead l Engineer · · including overhead Total Labor Materials 59 000 59 000 9 000 6 000 133 000 32 000 18 000 316 000 Direct Materials Maintenance ¥ aterials other Materials Total Materials 15 000 79 000 Total Operating Cost 395 000 24 000 4o ooo 61 TABLE 21 ESTJY t ATED MANPOWER AND CAPITAL FOR FEED PLANT CONSTRUCTION AND OPERATION IN U S Feed Rate 50 TU per year Capital Costs Dollars Construction M npower and Scheduling Installed Equipment Solvent Extraction Fluorinator and Cold Trap Fluorine System Auxiliary Chemistry Systems Spare Equipment Piping Instrumentation Utilities Maintenance Facility Administrative Laboratory Building Direct Construction Costs Peak Manpower 143 000 97 000 153 000 26 000 82 000 261 000 74 000 44 ooo 42 000 46 ooo 254 ooo 1 222 000 Men Engineering Non Manual Manual 8 12 90 Total Manpower Man-Months Engineering Non Manual Manual Time Required 8o 120 86o 11 Months Start-Up 135 000 Engineering Design and Inspection 154 ooo Indirect Construction Costs 240 ooo Contingency 238 000 Total 1 989 000 OPerating Manpower and Costs Dollars Per Year 5 days week 24 hours day Labor l2 Chemical Operators 11 Maintenance l Supervisor 1 5 Clerks · overhead at 10 3 Laboratory Technicians including overhead 1 5 Engineers including overhead - Total Labor Materials 79 000 75 000 9 000 B ooo 171 000 48 ooo 27 000 417 000 Direct Materials Maintenance Materials Other Materials Total Materials 44 ooo 51 000 20 000 115 000 Total Operating Cost 532 000 62 ___ _ - RD TABLE 22 ESTIMATED ¥ ANPOWER °AND CAPITAL FOR FEED PLANT CONSTRUCTION AND OPERATION IN U S Feed Rate 100 ' LU per year Capital Costs Dollars Construction Manpower and Scheduling Installed Equipment Peak Manpower Men Solvent Extraction Fluorinator and Cold Trap Fluorine System Auxiliary Chemistry Systems Spare Equipment 188 000 129 000 202 000 33 000 108 000 344 ooo Piping Instrumentation Utilities Maintenance Facility Administrative Laboratory Building 98 000 58 000 52 000 57 000 308 000 l 5TI 0OO Direct Construction Costs Start-Up Engineering Design and Inspection Indirect Construction Costs Contingency ' f-otal Engineering Non Manual Manual 10 15 116 Total Manpower Man-Months Engineering Non Manual Manual T lllle Required 100 14o 1 110 11 Months 173 000 210 000 316 000 315 000 2 591 000 Operating Manpower and Costs Dollars Per Year 5 days week 24 hours day Labor 14 15 L 2 Materials Chemical Operators Maintenance Supervisors Clerks Overhead at 100% 3 5 Laboratory Technicians including overhead 2 Engin ers including overhead Total Labor 92 000 98 000 14 000 11 000 Direct Materials Maintenance Materials other Materials Total Materials 216 000 54 000 Total Operating Cost 36 000 521 000 83 000 6'7 000 2 000 173 000 TABLE 23 ESTD ATED MA l 'POWER AND CA IT lL FOR FEED PLANT CONSTRUCTION AND OPERATION IN U S Feed Rate 500 TU per year Capital Costs Dpllars Construction Manpower and Scheduling Peak Manpower Installed Equipment Solvent Extraction Fluorination and Cold Trap Fluorine System Auxiliary Chemistry Systems Spare Equipment Piping Instrumentation Utilities M lintenance Facility Administrative Laboratory Building Direct Construction Costs Men 295 000 213 000 284 ooo Engineering Non Manual Manual 106 000 201 000 19 29 215 Total Manpower 628 000 199 000 115 000 Man-Months 68 ooo Engineering Non Manual Manual 70 000 426 ooo 18o 28o 2 06o 2 605 000 Time Required 11 Months Start-Up 290 000 Engineering Design and Inspection 4o7 ooo Indirect Construction Costs 592 000 Contingency 578 000 Total Operating Manpower and Costs Dollars Per Year Overhead at 100% 4 Laboratory Technicians including overhead 2 5 Engineers including overhead Total Labor dE ys week 24 hours day Materials Labor 17 Chemical Operators 28 Maintenance 2 Supervisors 2 Clerks 5 n ooo Direct Materials Maintenance Materials Other Materials Total Materials 324 ooo 64 ooo Total Operating•Cost 112 000 183 0 18 000 45 000 757 000 254 ooo 124 ooo 28 000 li66 ooo 1 163 000 64 TABLE 24 ESTIMATED IviANPOWER AND CAPITAL FOR METAL CCMPO11ENT FACILITY CONSTRUCTION AND OPERATION IN U S Product Rate 50 Kg U per year Construction Manpower and Scheduling Capital Costs Dollars Peak Manpower Installed Equipment Wet Chemistry Process 24 000 Reduction Bombs 8 000 Casting Equipment 44 ooo Machining and Testing Facility 50 000 Recycle Equipment 20 000 Piping Instrumentation Utilities Building Direct Construction Costs Engineering Design and Inspection Indirect Construction 14 000 Men Engineering Non Manual Manual Man-Months 32 000 24 000 221 000 25 000 Engineering Non Manual Manual Time Requir d 64 ooo Operating Manpower and Costs Dollars Per Year 4o-Hour Week OPeration Labor 2 Operators 1 Maintew nce 30 000 l Engineer 1 Supervisor 30 000 Overhead Materials Total 5 Total Manpower 5 000 310 000 Total 2 2 6o ooo 28 000 148 ooo 20 20 50 9 Months TABLE 25 ESTIMATED MANPOWER AND CAPITAL FOR METAL COMPONENT FACILITY CONSTRUCTION AND OPERATION IN U S Product Rate 500 Kg U per year Capital Costs Dollars Construction Manpower and Scheduling Installed Equipment Peak Manpower Wet Chemistry Process 49 000 Reduction Bombs 16 000 Casting Equipment 44 ooo Machining aud Testing Facility 70 0000 'Rec le 'Eg_u i e t Piping Instrumeutation Utilities Building Direct Construction Costs Engineering Design and Inspection Indirect Construction % f I Men Engineering Non Manual 1v vvv 19 000 6 000 Total Manpower 42 ooo Man-Months 50 000 326 000 30 000 83 000 Engineering Non Manual Manual Time Required Total Op7rating Manpower and Costs Dollars Per Year 4o-Hour Week Operation Labor 6 Operators 2 3 7 l f' 'l 3 Maintenance 90 000 2 Engineers 1 Supervisor 1 Chemist 57 000 Overhead Materials 147 000 250 000 Total 544 ooo _r- 20 30 70 ll Months 66 250 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - V' LU er Product Rate 50 Kg yr Feed Concentration 0 0071 Product Concentration 0 90 Waste Concentration 0 0065 Total Centrifuges 19 800 Total Centrifuge Banks in Series 219 Centrifuges in Parallel at Feed 618 200 LU V' z vi' LU 0 I 150 V' z a I LU I I 100 u 0 er LU a z 50 100 200 300 · 400 500 NUMBER OF CENTRIFUGES IN PARALLEL Figure 9 LAYOUT OF IDEAL PLANTS CONTAINING SUBCRITICAL CENTRIFUGES MODEL A 3 in 350 m sec 600 700 I ' I l l t ' I ' l lI --' i ' i f l I' SECRET t 06 6 Z P I 63 -· l ' I I t ' ' i J r J I I I I i l I l i I I i -- - ------ - · t6 1l i r· I' 1l· t I i l I ' I l I _ I I I i ' l ' I ·-- -- - • iEtRH • I SEeRET ----- ---- ----- 70 I _fiis l • • ll Q ---- l I I ' ' ----- n - -----yO b - -------- - -· ________ -· --- 'I • il O' 7 ' I • l I l I I ' P l I I I I I i i I I I I _ ---· l I Il I I --c SECRET 72 DISCUSSION FEASIBILITY OF CLANDESTINE OPERATION Any nation in groups X Y and Z could if sufficiently motivated build a gas centrifuge plant which would produce sufficient fissionable material for the construction of a nuclear weapon The gas centrifuge plant lends itself to clandestine operation The power requirements for the centrifuge plants under consideration will be relatively small and since it is possible to obtain a relatively large separation in a single centrifuge the number of centrifuges required for the plant particularly if they are of an advanced design is considerably less than the number of gaseous diffusion stages which would be required for this same production goal Therefore a small centrifuge plant can be contained in a building o f modest dimensionsj such a plant would be difficult to detect especially in an industrial country A class X country could build a clandestine centrifuge cascade for production of even the 500-kg U per year quantity with very little outside assistance and with little drain on its economy A class X country will have several large universities and in general all will have conducted research of varying description pertaining to problems in isotope separation Any type X country will have the· experienced scientists and engineers necessary to bring a centrifuge plant into successful operation Similarly a class X country will have no problem in obtaining the services of skilled machinists for constructing the centrifuges nor in recruiting trained operators and maintenance men for running the plant The special materials required for construction of the isotope separation plant and the relate·d facilities would most likely be readily available in a class X country or if not could be purchased without arousing any suspicion due to the high level of domestic industrial activity Many o f the special items such as spectrometers special instrumentation etc could be purchased by a class X country through a university for its research department without inviting attention Furthennore the countries in this category already have or may be expected in the very near future to have nuclear power programs Thus these countries will have a valid requirement for research re ctors and uranium ore It would not be possible without safeguarcle fuel or adequate controls to detect the relatively small diversions of uranium necessary to provide feed to a small isotope separation cascade under these circumstances A class Y country may be characterized as a country which possesses technological competence but which has limited industrial activity A class Y country could not build a centrifuge plant without some outside assistance however it could probably adequately disguise the nature of its activities from the outside world at least for the low production rate facility A class Y country may be assumed to have a sufficient number of scientists and engineers to bring·a centrifuge facility to successful completion Since these men may lack specific experience of this nature it may be assumed that it would take appreciably longer for a class Y country to achieve successful operation than it would a class X country A class Y country would have some difficulty in recruiting the skilled 73· machinists operators and maintenance men necessary to construct and o perate the isotope separation plant A class Y country would in all liklihood have to import much of the hardware necessary·to fabricate the centrifuges and also some of the auxiliary eg_uipment required for the plant The material of construction for the centrifuge bowl would probably have to be imported however since further machining and assembly could be done after delivery the use to which this material is to be put may not be evident A class Y country would also have to import other materials such as motor generator sets for high frequency currents process control equipment mass spectrometers and other items of a specialized nature These orders could however be distributed among a large number of vendors in order to prevent detection of the construction effort for the small-production rate plants it would however be much more difficult to pravant d ataction with the larger· plants Class Y cowrbries may well have access to uranium ore either by virtue of their domestic nuclear power program or their own natural resources 'rf not they would have to procure it elsewhere and this may provide a method of detection There are however many countries able to export uranium ore and one· should remember that the 25 tons per year or so required for the small production rate plant is a relatively small quantity of ore The high production rates reg_uiring hundreds of tons per year of ore·would probably present a'problem of detection in a class Y country not having its own ore st 1-p ply class Z country would find the construction and operation of e ·er ' -rifi ge facility a difficult tas J h e toiai -capitalinve·stmeat vi li cn a TUounts---co bout· $1 3 ·5· illicin for the model c centrifuge facility for the low production rate coupled·with the operating costs of about $4 5 million per year wou la be in the case of a class Z co f 1' ' qu i te a bu 't' E n on the economy A class Z country - rould have -co lie·· highly motl-rated t - undertC te such an expensive project However the lower construction and operating costs estimated with the more a vanced centrifuge models would certainly make the project more feasible for a cla s Z cotm try A class Z country would need·a great deal of outside assistance both in manpower and in material in order to bring a centrifuge plant into successful completion Countries in this category would probably need technical advisers from abroad competent scientists and engineers to aid in dev ' c pment of 'o £1 operable facility Operators and ma i i J tenance men would have- t o be trained for their particular jobs A class z country would probably not have su fficient skilled· machinists to fabricate the centrifuges It would be expected therefore· thiit a type z·· country wou d purchase fabricated centrifuges Their alternat ive would be to ·train the necessary machinists and to purchase the lathes drill presses and other shop eg_uipment which would be required-for the manufacture of·centrifuges In addition to the centrifuges almost all of the auxiliary equipment reg_uirj J pr a ntrifuge plant woui L have to b l' chased froin foreign _ A r y ua rs ec - -· to· hid' e -DELETED _ __ I 'f G t 2 ' tfrtbennoice the plant Hseir wouia pt'1 5a5ly be more ditfiCui - n a nonindustrial c·ountry · In short it appears that a class Z ___________________________________ ncJ ____ _ country could not build a completely clandestine nuclear weapons facility It could however with the collaboration of a class X or Y country build such a facility but even then it would have much more difficulty hiding it than would a class X or Y country MINThfilM TlliE REQUIRED TO PRODUCE THE FIRST NUCIEAR WEAPON The minimum times estimated for the production of the first nuclear weapon with the centrifuge process in countries X Y and Z are presented in Taole 8 for both the present centri t'uge model C and the yen advanced C 9- l LK J J19 _ If - I c76 I 6' zf DELETED l 7' c _- fTheseTirnes can -be c o m j _ j a r r s f'or ' neuniteO states • · 1'fi e minimum time estimated for the develoment of _tne highly advancgd centrifuges models K and L is fi _ years DELETED '--ff t h e centrifuge development effortswere not available·from the United States or from some other highly advanced country and the countries have to develop the centrifuge on their own it is estimated that a class X country without an already well-developed centrifuge program would require an additional 2 to 3 years over the times presented for the present centrifuge model C A class Y country would require an additional 4 to 5 years for model C and would probably not attempt to develop a more advanced centrifuge and a class Z country could probably never develop the centrifuge on its own OPERATING AND lAINTENANCE At this particular time in the_ centrifuge development estimating the required maintenance and-the number of operators needed for the safe and efficient operation of a gas centrifuge plant is extremely difficult It is felt that the costs presented here refleet _ptimistic thinking and the future costs and estimates are not expected to be lower The actual staffs required for the operation of the centrifuge plants in the early stages until the problems and uncertainties are resolved could be larger than those presented here CAPITAL AND OPERATING COST ESTIMATES 1 DELETED l I ------ Off · y c t zei -- ·- •_ _ J Y J 75 7 li e roo e aOwnced cen _my $985 000 $1 032 000 and $1 065 000 for the class X comyared with $947 000 for the United States YJ be r J and Z countries · KEY ABE 4 S ThlJICATING POTENTIAL FOR WEAPONS PRODUCTION A class X country due to its large industr al caps city could probably construct and place in operation a centrifuge facility without any outside assistance and without arousing undue suspicion Both a class Y and Z country would require outside assistance and the purchase of many items outside their o m countrv Tne kev items whose m '3 l' IUfa t rrA o n r ch_e_cc in quantity would serve as an alarm that a centrifuge plant may be in operation or under construction are listed in Table 26 Any one of these items may mean nothing but a combination of them should indicate the possioility of a centrifuge facility - • --- - - J - - - ----i' ·----- - ----- tit I 9 b l tf DELETED I I ' CORRELATION OF Th DUSTRIAL CAPABILITY OF NATIONS To estimate the order in which the various nations can be expected to achieve an isotope separation process and thetime required for this accomplishment it is necessary to choose some feature or features of a nation's economy which are indicative of that nation's industrial competence The two such features -whic2 reflect all areas of modern industrial i' I C JRD activity are the consumption of electrical enerror and the consumption of steel Furthermore the per capita value of these quantities provides a measure of the level or relative effectiveness of application of a nation 1 s industry since per capita consu ption is itself an empirical statement of demonstrated effectiveness Accordingly the following quantities were defined Relative Industrial Size s Steel Consum 12tion US Steel Consumption Steel I PerConsCapita l2 Llori Relative Industrial Level L V Relative Industrial Capability ✓ Urn US Per Capita Steel Consumption s Electrical Energy Consumption US Electric Energy Consumption Per Capita Electric Ene r·gy Cor i surnptiou US Per Capita Electric Energy Consumption L The Relative Industrial Capability RIC was then relat·ea to actual time cost and manpower requirements by correlating the industrial experience of nations for which such data were available with the computed RIC 1 s of those nations The results of the correlations are shown in Figure 15 in which factors for converting United States requirements into the requirements of other nations are shown for man-months required for building a plant the time which will elapse between inception and end of construction the number of men engaged in construction or operation of the plant applicable to peak work force in the case of construction the construction or ca pi tal cost and the annual operating cost In each case an estimate made for the United States may be multiplied by the appropriate factor of Figure 15 to obtain the corresponding estimate for a nation for which the RIC has been computed The RIC was computed for a variety of nations and these are located on Figure 15 Also shown are the zones which define the X Y and Z categories of nations referred to in this report Statistical data used in computing the RIC were obt ained from Statistical Abstract of the United States which contains international statistical data drawn chiefly from the United Nations Statistical Office This Agency I s annual publication Statistical Yearbook rr provides a wide variety of detailed statistical data -·---- • ---· ' -1 '· 0B' ii '' l t·· ' ' - c· ' __ -- - - ----------- ·· -·- Figu- e 15 _ ·· CORRELATION OF RELATIVE INDUSTRIAL CAPABILITIES · -- -· --R· CL lJESTDIB PRODUCTION OF ENRICHED URA_Jl IlJM BY THE SWEEP DIFFUSICH PROCESS - I' u· Although the previous sections of this report have dealt exclusively with the possible production of enriched uranium by the centrifuge process there are other processes which might be better matched with the lower technological capabilities of Y andr-- Z foreign powers Such a process for exam ple would be sweep diffusion · In the sweep diffusion process J a gaseou - mixture of isotopes is confined in a separation unit through which flows a current of a third component called the sweep gas or sweep vapor As the sweep vapor flows through the process gas mixture it tends to sweep along the heavy or less diffusible component In COilll ' lOn applications of this process the sweep vapor is chosen so that it can be easily condensed on a cold vertical wall The falling film of condensed sweep vapor then drags the process gas downW1 rd with it setting up an axial circulation or countercurrent flow as ° own a n Ri g11re J 6 which multiplies the simple process separation factor It is envisioned that a i sweep diffusion column could be built in the form of two concentric tubes the inner one being porous and serving to distribute the sweep vapor which is introduced into the central region The separation region would then be 'in the annulus and the entire column would be immersed in a coolant bath so that the outer tube would act as a condenser for the sweep vapor Such columns would be rcla ti vely easy to construct and a number of the coiumns could be put in a common coolant jacket to reduce costs and make the plant more compact I ' · r --· - - - - ·- - a··--f - - - - ·- costs estunate or 1 u e sweep u1i 1 usion process musu ue regaru ecf_§he as of' a most preliminary nature since the process has not been proved ex perimentally The costs are based on an efficiency factor of 50% for the swee11 diffusion plant based on the Ill9 Ximum separative capacity of the columns The power requirement for sweep diffusion will be the order of tr 1 t for gaseous diffusion and will thus be much greater than for a entrifuge plant of similar size The 5olubility of the process gas in the condensed sweep vapor is one of the problems which must be overcome if the sweep diffusion process is to ceach its theoretical potential If the process gas is appreciably scluble and this mixture is then vaporized and reintroduced into the columns a serious loss in separative work would be incurred iecause the ·process gas in the sweep vapor stream is at the average plant concentration in the desired isotope and it is being remixed with process gas at - he point concentration in the concentration gradient A study of this r 1D ixing effect shows that separative work losses greater than· 50% are probable with sweep vapors of the Freon family If the solubility problem could be solved by proper choice of the sweep vapor or if the effect could be mitigated by careful process design the sweep diffusion process might be a very attractive means of enriching s n ' 79' LIGHT PRODUCT i I l i SWEEP VAPOR FALLING LIQUID FILM COUNTERCURRENT FLOW OF PROCESS GAS FEED I POROUS HOT WALL CONDENSER COLD WALL --- _ l - - - - • HEAVY PRODUCT ' EVAPORATOR - - - Figure 16 SCHEMATIC REPRESENTATION OF A SWEEP DIFFUSION COLUMN -prfi Bo i ___ 7 ' - r' uranium in a country that lacks the technological sophistication to build gas centrifuges A sweep diffusion plant requires process equipment such as nickel plated pipe and screen rotary lobe blowers to move the process gas from cascade to cascade centrifugal pumps to pump the condensed sweep vapor to the evaporators and the evaporators themselves 1 fjl --_ · A preliminary investigation of the sweep diffusion process for the separa3' tion of other isotopes is being considered which would also detennine· '-'· _2 _ts potential for uranium isotope s e p a r a t f the more serious pr615Temswn1 cnwou td-ha ve----to--be--ove·rcc ime1s the solubility of process gas in the condensed sweep vapor r - 1 RF VTF W AND lWAT UA'J'TON OF FORF IGN CENTRTF'TJGE P ROGRA M S INTRODUCTION The current status of gas centrifuge development programs in foreign nonCommunist countries is reviewed in this section of the report and an evaluation of th program in each country is made with respect to its scope direction ultimate goals and chances for success The experience acquired in carrying out the centrifuge development program in the United States has been utilized in the preparation of these evaluations The information on the current status of the centrifuge·pro ject in each country has been compiled from the following sources The open scientific literature The press Interview and trip reports prepared by AEC personnel which are available in the classified scientific literature Classified intelligence documents This work supersedes a previous report 2 From the information available it appears that only four foreign countries outside the Communist Eloc are making any serious effort to develop a gas centrifuge process for the en c bment of the uranium isotopes These are England Japan The Netherlands West Germany In addition Brazil has maintained an interest in its rather modest centrifuge program and France has recently been reported engaged in centrifuge development In general little new information has become available since the preparation of the earlier report This is due in large part to the fact that European nations agreed at the end of 1960 to accede ·to the request of the United States that security restrictions be imposed regarding developments in the centrifuge process Exceptions to this are England with which the United States has an information exchange program and Japan where probably for political reasons the work is entirely unclassified r · SbGRET S1 Nevertheless it is most probable that because of t e furor which centrifuge developments evoked in the i ress in 1960 that almost every cl ass·x country has u dertaken a centrifuge program f sone sort even if quite small The Co mnunist Bloc countries have been excluded from this evaluation for the simple reason that practically no infor nation is available It seems reasonable to presume that the centrifuge program which Russia pursued in the years following world War II from 1949 to 1954 was abandoned when the German scientists working on the project were reFatriated r Zippe one of the scie tists who worked on the Russian project has recently reported that the Russians have resumed work on centrifuge development It is reasonable to suspect that this report is correctj however no specific information has been made available to the West Similarly it is reasonable to believe that the East German scientists· some of whom also worked on the Russian project are rrreRent1y p1 1 rs1 1i r1e their investigations for the Fast German Government But again no details are available As complete a description as is possible and an evaluation of the program in each of the various countries is given in the following sections BRAZIL Description of the Centrifuge Program It is well known that Brazil has three centrifuges built in West Gennany which are located at the Institute of Technological Research at the University of Sao Paulo in Sao Paulo Brazil According to infonnation received Dr Massei is the director of the Institute and Professor I-vo Jordan is in charge of the centrifuge _ project A statement by Jordan verified that the three centrifuges owned 91 Jb by the Institute er _im Eorte frQ m Genna P -_19 5 8 s i ncl IL l§ d J J i Q - tion in March 1959 J _ 9 J f · § ' - ft5- J DELETED I l The only published a rticle containing experimental separation data obtained with tpe Brazilian machines deals with the separation of the isotopes of argon 3 Accord ing to this paper written by Groth et al in l960 the EGRET --------------------------------------- _________ __ - ii 82 Brazilian centrifuges are of the t r' e which Groth has designated as tte ZG3 The ZG3 has a rotor length of 66 5 centimeters and a dia eter of 18 5 centimeters and is designed for a peripheral speed of 302 m sec Further information on the ZG3 is'given in the section on the West Geman centrifuge program Admiral Cunha president of Brazil's nuclear energy commission tas stated that Brazil has no intention of producing a nuclear weapon is interested only in peaceful applications of nuclear energy e d that Brazil's centrifuges are intended only for research and training purposes Evaluation of Program The current Brazilian centrifuge program is relativezy small involving as far as is known only the op_eration of the t h_ - ' '_ns - h-l - r c v'ho' - h h ' ' T p11 rh c fl f' - m W ' ' · Ii ' -- i -- - ------ ' r --- p ' _' 2 l ' ' ' I I' I I I -------ENGLATul Description of rog a m The current status of the British cent ifuge program is presumably well known to the US AEC by virtue of an agreement between the two governments effective in 1960 to exchange ini'orrr ation on gas centrifuge technology Members of the United States and the United Kingdom centrifuge teams met in accordance with this agreement in 1960 1961 1962 and 1963 It is anticipated that the next meeting between the two groups wHl take place in tv a rch 1964 The information obtained during these meetings regarding the nature of the British program is contained in the L S classified reports of the rr eetings 4 5 6 7 and 8 and is sllJI 8 ' 'ized below 83 · DELETED ------------The United ---- ------- -- - ------ -- - ---_ Kingdom development program has been in progress for about three years having been initiated in October 1960 The first two-year program for which 300 000 pounds sterling or about $840 000 was budgeted was aimed at developing a respectable machine with a good efficiency and an extraction system which can be U sed by engineers to make a cascad In October 1962 a second two-year program was authorized at a budget level of 350 000 pounds sterling or about $980 000 The stated goal of the present program is the successful operation of a cascade of centrifuges I DELETED i I _DELETED I i' I i I l l The future plans of the c 1 ·Ie it UK centrifuge program include Operation of an imprmred relatively low speed subcritical machine The operation of a high speed subcritical machine The operation of a relatively low speed supercritical machine and --·--- Operation of a cascade SECRET --- E v-alua L -• L LOU Cl OI P2 ogra'Ut I I I I I I I I I DELE1Etl I II II I I I I I 'SECRET _j 06 86 I ------------ ----------· i J ·- JAPAN DELETED 11J I '° _ ' C Description of the Program · Japan has been conducting research on the enrichment of the isotopes of uraniwn by means of the gas centrifuge process on a relatively smali scale at the Institute of Physical and Chemical Research in Tokyo The program was initiated in FY 1958 All of the basic gas centrifuge work at the Institute on-uranium enrichment has been directed by Dr ' Yoshitoshi Oya ma of the Tokyo Institute of Technology The current staff working on the project at the Institute includes about 10 scientists and engineers plus three technicians The Japanese program is financed by the Science and Technics Aeency of the Japan Atomic Energy Bureau A number of research contracts wholly financed by the Government have been entered into with various research and manufacturing groups including the Institute of Physical and Chemical Research the Osaka Metal Industry Company Ltd Nippon Atomic Industry Group and the Chemical Engineering Association The first Japanese experimental centrifuge was cornple_t_e_d ur ng FY 1959 at a reported cost of I fi§ ooq l J- JThe un t ufactuxed by the Ishikawajima· --Turban Manui·acturing Gcimpa nf now a subsidiary of the Toshiba Company The machine was patterned after the centrifuges designed·by Groth in West Germany Countercurrent flow is presumably obtained by means of a temper-ature difference b tween the two end caps Most of the test efforts on this machine were directed toward working out the bugs in the system and dealt with balancing motor performance effectiveness and durability of· the seals and determination of operating temperatures of the various components Separation tests were attempted with neon argonJ and UFh However the neon separation tests failed because lubricating -oil lea R ed into the roto r and the UF separation experiment failed due 4o the 6 reaction of' c with t t 11 onl t ·n rrgon se earation t1 _s_ts ere at all successful ·· · ·----- 1 DELETED r bM 87 • DELETED I i I I l l I Although most of the uranium enrichment research work has been under the direction of the Institute of Physical and Chemical Research beginning in FY 1963 the Japan Atomic Fuel Corporation was to take over this function including the experimental work at its Tokai-Mura laboratories Laboratory facilities are now being constructed to accor unodate the work and the two gas --------------------------------------------- centrifuges will be tranoferred from the Institute to the Takai I abcratcries Financing ill continue through the Science and Technics Agency o further tests are planned at the Institute on the unit which is now oeing dismantled for installation in the Japan Atonic Fuel Corpo - ation laboratory at Tokai-Mura where resumption of testing ¥as scheduled for September 1963 The ultimate objective of the AFC work is to d e- relo p a practical ultracentrif ige for use in a pilot-pla 'lt operation ' All of the support ng research work on aterials bearings seals etc is aimed at achieving this objective The technical problems and costs encoi mtered to date are of _greater _ma itude than wlmt was at firE i ant icipat ed f ·· ------·-·--- DELETED 1use o- ' li -u on anexTieriinentiiJ h is scheduled to s art- n I 19-Y fi'C is now making plans to purchase UF production equipment including a mass spectrometer 6 _ - - -----• - -••- • ' · - • - - - _ • • • ¼ S - rf - The work being done by the Hippen Atomic Industry Group HAIG of which the Toshiba Manufacturing Company is a predooinant member includes basic research work on the development and testing of key items such as the rotors bearings seals etc aimed at the eventual desiesn and construction of a p 'actical gas centrifuge unit 3 capable of operating on UF 6 In this work NAIG is experimenting with different high-strength light weight UF and HF corrosion resistant materials such as various grades 6 of duraluminum and fiberglass The group hopes to be able to develcp a practical gas centrifuge by 1966 capable of operation in a small pilot plant The new unit would incorporate any improvements resulting from the testing of the two existing machines at AFC with the best engineering design developed at NAIG NAIG is confident that a machine can be developed within this period which can be used to separate the isotopes of uranium economically NAIG is currently engaged in dismantling and transferring ur1its 1 and 2 to Tokai-Mura They will supply three teclmic ans to help operate the units The work at NAIG is under the lirection of Mr Kunio Yoshimura who has helped with the design of the two units at the Institute He has traveled extensively in Europe in connec ton with the study of the gas centrifuge and appears to be well acgus ed with the program under way both in Europe and in the United States Tne Osaka Metal Industries Company the 2 ead ing fluorine· producer in Japan he s performed research work since FY 1958 aimed at the product on of UF Development work on T JF -handlin€ apparatus is also in progress 6 6 An over-all study of plant design for u raniu n isotope separation was nade by the Society of Chemical Engineers under direct contract with the Japan Atomic Energy Bureau These studies include a feasibility analysis of the gaseous diffusion gas centrifuge and separation nozzle method s to determine the most practical route for Japan to follow and to outline the technical proble s ir volved with each of these methods DELETED - DELETED I I j In addition to the experimental work described above two papers of a theoretical nature by Ka ngawa and Oyama appeared in the Journal of the Atomic Energy Society of Japan in late 1961 The first of these dealt with the effect of the flow pattern on t'he separative capacity of a centrifuge and the second with the expected variation of the over-all separation factor and separative capacity of the centrifuge with the feed rate to the centrifuge These papers are comparable with those written by the US investigators at the start of the USAEC centrifuge program and indicate a familiarity with the isotope separation theory developed by Karl Cohen and a basic understanding of the isotope separation processes in the gas centrifuge However in none of the published Japanese work do hey make any comparison between the theorJ and their experimental results Evaluation of the Program Despite the auspicious start made by the Japanese centrifuge program as measured by the progress achieved du ring the first year the program seems to have found ered Their second machine wr h was to be essential y- the equivalent of Groth's ZG5 fell appreciably short of this goal from a mechanical point of view and furthemore on the basis of its separative capacity is much inferior to their original machine This lack of success can be attributed in part to the fact that there appears to be no great urgency imposed upon tr e current program to d evelop a gas centrifuge of high separative efficiency This attitude is probably influenced· not oniy by the high cost and tmexpected technical difficulties involved in d eveloping the mactine but also because it is general y- felt in Japan that a supply of enriched fuel for its second nuclear power plant 250-300 mw electrical and probably for other future power reactors will be available from the United States or the United Kingdom or from other sou rces Nevertheless t ecause it may become desirable s'ooner or later for Japan to produce her own enriched uranium supply Japan has undertaken a centrifuge _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ___ 0 _ _ -S T r 9D develc pn ent program ai 1 ed at constructing a practical centrifuge by FY 1966 which is to be followed by a s all centrifuge pilot pl nt fi i 'iJ cr 91 DELETED - THE NETHERI ANDS Historical Background A review of the information available · Dutch centrifuge development program makes quite apparent the the Dutch have divulged very little information regarding the their work nor have they given any indication of the progress may ba ve made I - on the fact that scope of which they According to Kolstad the Dutch centrifuge proj ct is about six years old and in 1958 employed about 30 people In spite of this there have been only two general papers plus a few patent applications from the Netherlands dealing w j th 9entrifuges and none of these is at all impressive The two papers lO lll originated at the laboratory for Mass Spectrography F O M Amsterdam where Dr J Kistemaker heads the group working on the centrifuge project In i958 it was reported that this group had no centrif-qges in operation but were actively studying the component parts It was reported by Kolstad however that the Germans believe that the best Dutch centrifuge -work is being done at the Werkspoo'r N V C9 ny Little is lmown about the work there Qne of the putch patents L2 J d a te i February 1958 describes a fiberglass centrifuge bowl According to th s patent fiberglass is preferable to metal because of the ease of fabricating centrifuges from this lll'3 terial Since no mention is made of ways to combat the problems of cor osion and of low tensile strength of the rotors in the axial direction it is quite probable that no fiberglass materials had even been tested at that time · There had been some co-operation between the Dutch and West Gerill9 ns up through l96o r garding centrifuge development In a joint report to a meeting of the Executive Committee of the Research Syndicate for the Construction of a European Uranium Isotope Separation Plant the Dutch and West Gen nan groups proposed t construction of a plant containing 40 000 to 60 000 centrifuges which would produce annually 200 tons of uranium containing 1 4% U-235 and 7 7 tons annually·of uranium containing 20% U-235 from 1000 tons of natural uranium ore·and would require from 30 to 42 megawatts of power No mention was made of the type of centrifuges to be used in the proposed plant The Dutch claim that the centrifuge which they are developing is different from the West German ma chine and requires less power f'9r 9peration In the paper which Kistemaker presented at Geneva in l95$llO he mentioned four centrifuge models which may be indicative of the immediate goals of the Dutch program The models are liited below 92 cm Peripheral Speed m sec Separative Capacity kg U yr Maxi mum Allowable Centrif1 ge Cost $ cent dfuge 6o 10 300 1 3 85 B 6o 10 350 2 1 200 C 210 7 350 8 3 750 D 120 5 4 4Do 8 3 750 Dia Model Length cm A Eacl1 of the oe_pa r8 Llve capc cii ies in i he above -cable is ' 59b of the maximum theoretical separative capacity of the machine T'ne maximum allowable centrifuge cost is the price that one could afford to pay for a centrifuge and still produce enriched uranium at a price competitive vith that given by US price list in circulation at that time Kistemaker arrived at these figures by assuming that the power requirement would be about 100 watts per machine and that the total capital cost of a centrifuge plant would be twice the cost of the centrifuges alone Kistemaker concludes by expressing the opinion that it appears doubtful that model A could be produced for $85 per machine it is possible that model B could be produced for $200 per machine and that there is a very good chance of producing models C and D for less than $750 per machine It should be pointed out that C and Dare supercritical machines Concerning this Kistemaker said in the 1958 paper 1 Je in Amsterdam have high hopes of surpassing difficulties with the gas extraction and those of passing the criticals 11 DELETED e - - vi 1 __ futch centrifuge program was cle ssi lea ia te Ln-1960 by the Government of the Netherlancls at the reg_uest of the United States 1 'l·t••'' ' -- ---- ------------------------- a a __ - - - - - - - - - - - - - - - - • ·-fflf' T 93 ' i 'hat the I etherlands was quite actively interested in the centrifuge process ear _y in 1963 was evidenced by their request that - he United States help them get ccpies of certain East German centrifuge patents hie were in the possession of the West Germn Gove 'tll lent In May of 1963 the Netherlands announced the fact that they were embarking on a tr ree year centrifuge development program which if successful would lead to a pilot plant for U-235-U-238 separation A special research facility for gas centrifuge work is being constructed at Duivendrecht a suburb of Amsterdam for this purpose Evaluation of Program It should be evident from the foregoing description of the centrifuge development work in the Netherlands that so little information is available that it is practically impossible to make any c- o luo tion o ' the D-u tch program OD E is fo1· eu then merely to speculate concerning tbe nature of their program and the direction it is taking r--- - -- I DELETED I _1 It-woi iiff°'be presmned f m Kistemaker' s prese htion-that ·E fie -Dut cli progr was concerned primarily wi- 11 the development of a supercritical machine The lack of any subsequent inf'ormation on the Dutch program is indicative only of the fact that they have maintained an effective security program in this area w EST GERMANY Historical Background The German centrj_•-'•1ge program has up to the present time 7 consisted of two separate G d distinct development efforts One part of the program has been the centrifuge project located at the University of Bonn7 directed by Dr W Groth This work is a continuation of the German wart inJ e work and bas been quite active since 1954 Dr Martin's work at the University of Kiel is considered to be a part 'Of this program The other part of the Ge -man program was carried on at the laboratories of DEGUSSA in Frankfurt where Dr G Zippe supervised the development efforts This work was a continuation of the Russian post-war centrifuge work in that it was based on the centrifuge which was developed by the Russian centrifuge group of which Dr Zippe was a member Thi project as been under way for about four years Dr Groth 1 s work at Bonn was unclassified until the end of 196o when the West Gerrea n Government followed the recommendation of the USAEC and j -sEeREf 94 classified all f'LU'ther developments in centrifuge technology i1uch of the significant open literature on the gas centrifuge has been corrtr Lbut ed lcy this grou p c f investigators ' L'his J rogram has aiways S ad a st cong enginee ing em hasis The work has been directed toward the development 3 _ llechanically reliable centrifug_ s · DELETED DELETEI» Tlleir program can be surniii arized most i conveniently by considering the various centrifuges they have either con- structed or plan to construct The characteristics of these centrifuges are listed below Length-toDiameter Ratio Peripheral Speed m sec Separative Capacity kg U _yr cm faulius cm UZ-I 4D o 6 o 3 33 302 0 582 UZ-III B 63 5 6 7 4 74 302 0 935 ZG-3 66 5 9 25 3 6o 302 0 97 __113 0 9 25 7 03 302 1 6 4 - - - i ength Centrifuge ZG-5 t X7c I Y b - DELETED - -- _ ··----·-- Model UZ-I was constructed during World War II and was successfully separating the isotopes of uranium early in 1942 In 1943 a double centrifuge unit the ·uz-rrr A was built · This machine which consisted of two rotors spi1ming in opposite directions with a cyclical fluctuation in speed in order to provide the gas flow between them was to be the basic processing uni t of the Gennan uranium isotope separation plant Although completed 1e UZ- III A was damaged by the war before any UF 6 tests could be made After the war an improved machine the UZ-III B was built which had the sariie dimensions s the UZ-III A but it as well as all subsequent models employed heated end caps to provide the internal countercurrent i'low The structural details of and the experimental results from the UZ-III B have been published in several places A good review of this work was presented by Groth and Beyerle 13 in 1958 Th ZG-3 centrifuge was described by Groth i4 at Geneva in 1958 and some data were presented on the separation of the isotopes of argon and xenon Although Groth stated that the ZG-5 was already in operation i f 1958 no_data on the ZG-5 experiments were reported until a 196o paperl3 which dealt with the separation of the isotopes of argon ---SEeftEf ' 95 The separative efficiencies have been computed for several of Groth's machines from his experimental data and are presented below Centrifuge UZ-III B ZG-3 ZG-5 Peripheral Velocity ml_sec Separative Efficiency UF6 252 70 UF6 28o 74 argon 268 30 argon 302 23 xenon 268 36 xenon 302 30 argon 298 39 Process Gas Groth was apparently disappointed by the low separative efficiencies obta ineq w ith the ZG-3 and ZG-5 The last paper pµ_blished by the group at Bonntl5 deals with the possibility of increasing the separative efficiency of his centrifuges by impressing a radial temperature gradi nt across the end caps The meager data presented in this article indicate -that an e p reciaole gai l in eff cie i cy may be attained in this ma 1_11 e It is doubtful tt- i t there will be any further publicatio- S ' q this group in the open literature until such time as the German Government changes its classification policy regarding the centrifuge process Several nuclear research institutes have been established by the German Federal Republic Groth is now supported in his research work by the Society for the Support Nuclear Science and Rese 2 ch of' the state of Nord Rhein-Westphalia Part of' this research organiz8 ion is the Institute for Scientific Apparatus Measuring and Control Engineering Aachen of' which Dr Konrad Beyerle is director Several members of the staff were transferred from the Institute of Instrument Design of the Max Planck Society Gottingen which under Beyerle's direction designed the earlier centri·tuges used by Groth i I DELETED ' -'a _ -- - ·_ i £ECRET DELETED I f 'f 6 I Pllt DELETED ____________ ·--------··-·- _ _ ___ _____ Another Eeyerle-designed centrifuge was put in operation in the summer of 1958 at Kiel Institute for Physical Chemistry Kiel which is headed by Dr Hans Martin The unit is si ilar to those used by Groth in Bonn Martin I s centrifuge is mounted verticaJly and is abo' lt 4-l 2 feet lo ig Martin is continuing his theore fr A 7 stn niPR 0n the t e0Ty cf c 'lov inside the centrifuge bowl DELETED ____ I The centrifuge developed by Dr Groth at Bon _ is a relatively complex machine compared with the centrifuge model which was developed in Russia after the last war 1946-1954 by a group headed by Dr Max Steenbeck Dr G Zippe an Austrian who was interned by the Russians and who was a member of Steenbeck's group came to the United States afte his release by the Russians and in the period from July 1958 to June 196o duplicated at the University of Virginia the machine which had been developed in Russia This centrifuge which differs from the Groth-type centrifuge primarily in that it eliminates the need for complicated sea ls and bearings has been termed the Zippe-type machine tr roughout this report At approximately the same time that Zippe was duplicating the Russian machine at the Unive sity of Virginia an associate of his on the Russian project Rudolf Scheffel was doing the same work in Frankfurt for T EGUSSA These two men had an agreer r it with one another regarding their w ork and when Zippers contract at V 1 - _ inia was terminated he joined Scheffel at DEGUSSA • DELETED 1 I I l I l I ECREI DELETED In connection with the German decision in 196o to classify the ce trifuge program it was their stated intention to consolidate all of the centrifuge development work in a new facility at Julich near Aachen Neither of the groups concerned was at all happy with this arrangement nevertheless Zippe said they are planning to move all of the DEGUSSA centrifuge work to Julich where it will be conducted directly for he West German Government S im ilarly Groth although he has on severe · occasions said that he was going to terminate his research on centrifuges because of government interference and because he has been severely hampered by his Government's security regulations apparently agreed in mid-1963 to move his laboratory to Julich and continue his work there It has just recently been reported that Dr Boettcher director of the Atomic Research Center at Julich has applied to the Nord Pli ein-Westpr alia authorities and to he Ministry for Atomic Energy for ap roval of a proposal to establish a centrifugal separation instituie at Julich Sufficie t funds were requested to employ a staff of about 100 scientists at the centrifugal institute Fifty percent cf the total expenditure would be met by the state of Nord Rhein-West pb alia and 50% by German s'ederal Goverrnrrent The DEGUSSA group under Dr Zippe and Dr Scheffel would form the nucleus of the centrifugal research staff ' i'b e total number of Ge rma n scientists currently engaged on the project is less than 25 However Dr 3oettcher 1 s plans are opposed by some influer ial n e r ters of the super isory boa -d at Julich e nd r o decision tas yet been reacted t f G 'IL' Evaluation of Program i I I I I DELETED I I I I I I I' The interest which Groth a professor at the University of Bo has in the centrifuge process is probably primarily scientific and pro essional His name has been associated with centrifuges for a long t e and the development of a successful centrifuge process would be a singular personal achievement The intere3t shown by the Gerrr an Government evidenced by its su port of the centrifuge project is much more difficult to assess It is granted that the Gerin 9 n Govermnent is interested in supporti g sc _entific activities at its institutes a l d universities___ _--·--------·-·· ··-- --·sp ' M 1----- DELETED o a 99 I I I li DELETED ' C1' t-$ t FOREIGN COUNIBIES - ' • ------- - I I l DELETED I I France DELETED DELETED I I ' ' ----------- - l fc SEBRET 100 ACKNOWIEDGEMENTS The authors wish to acknowledge the assistance given them in the preparation of this report by the Plant Engineering and Technical Divisions of ORGDPJ the Technical Division of Y-12 and the centrifuge grou_ps at AiResearch and the University of Virginia SEeREf ' - SECRET 101 REFERENCES l Levin s A 1 Hatch D E J and Von Halle E 1 Production of Enriched Uranium for Nuclear Weapons by Nations X1 Y and Z by Means of the Gas Centrifuge Process 1 Union Carbide Corporation Nuclear DivisionJ Oak Ridge Gaseous Diffusion Plant 1 February 26 196o KOA-662 2 Garrett G A Levin S A J and Von Halle E Review· of the Gas Centrifuge ·Process and Evaluation of Foreign Centrifuge Programs 1 Union Carbide Corporation Nuclear Division 1 Oak Ridge Gaseous Diffusion Plant 1 October 30 1961 KOA-916 3 Bulang w Groth w Jordan I Kolbe 1 W J Nann E and Welge 7 K- H 7 non the Separative Capscity of The n u lly D1·iven Countercurrent Gas Centrifuges ' z Phys Chem Frankfurt 24 249-64 196o 4 Garrett G A Vanstrum P R and Kuhltba u A R Report on Meetings with the UKAEA in England December 12-15 1960 Union Carbide Corporation Nuclear Division Oak Ridge Gaseous Diffusion Plant February 20 1961 KB-852 5 Vanstruro P R Evans E C and LevinJ S A J Report on Meetings with UKAEA Representatives at the University of Virginia July 1920 1961 Union Carbide Corporation Nuclear Division Oak Ridge Gaseous Diffusion Plant August 23 1961 KB-878 6 Friedericy J A 7 Meeting with UKAEA Representatives at the University of Virginia Charlottesville Virginia July 19-20 1961 University of Virginia August 1961 EP-4422-181-61s 7 Levin S A and Von Halle E • Report on Visit to Centrifuge Facilities at Capenhurst England June 14 1962 Union Carbide Corporation Nuclear Division 7 Qak Ridge Gaseous Diffusion Plant February 14 1963 KOA-1127 8 I ang D M Leed R E wwry R A Parker H M VanstrumJ P and Woc -1 w-orth L R Report on Visit by US Gas Centrifuge Team to UY • May 27-30 1963 Union Carbide Corporation Nuclear Division Oak Ridge Gaseous Diffusion Plant September 2 7 1963 7 KL-1670 9 Kolstad G A Discussions with Germans and Dutch on Centrifuge Collaboration FUR-50 July 1958 10 Kistemaker J 7 Los J J and Velclhyzen E J J 11 The Enrichment of Uranium Isotopes with Ultra centrifuges rr Proceedings of the Second United Nations Conference on Peaceful Uses of Atomic Energy Geneva Vol 4 435-8 1958 ·SECREt l02 ll J 2 Los J and Kistemaker J 11 0n the Influence of Temperature Distribution Inside a Gas Centrifuge If Proceedings of the International Symposium on Isotope Separation North Rolland Publishing Comran _v Amsterdam 695-700 1958 Netherlands Pa tent No 87 740 To Reactor Centrum Nederland Feo Uary 16 1958 13 Beyerle K and Groth w 11Enricbment of the Uranium Isotopes by the Gas Centrifuge Process 1r Proceedings of the International Symposium on Isotope Separation North Rolland Publishing Company Amsterdam 667-94 1958 14 Groth W Beyerle K Nann E and Welge K- H• Eu dcbment of the Uranium Isotopes by Lhe Gas Centrifuge Process Proceedings of the Second United Nations Conference on Peaceful Uses of Atomic Energy Geneva Vol 4 439-46 1958 15 Bulang W Groth 25 283-5 196o W and Nann E • z Physik- Chem Frankfurt I 1 1 1 I 1 1 1 1 1 1 1 1 -_ SECRET • I 1