1 Amem NUCLEAR*-PR@PULSI&N "PRGJEC‘I‘ QUARTERLY PROGRESS REPORT for Perlod Endlng Mmh 01; 1951 i Phofosi‘at Pnce $ TR G e T Dll‘eCtorz ANP D1V151°n e .-._.Mlcrofalm Pnce S ll 10:' deted I:uy° _- e -_:Avqslcir!e frcm fhe ol o W B COttrell - | Office of Techmcal Servrces ok ” P K e L E Depar’rmenf of Commeme . ST e __:Washmgfon 25 D C -'Dgt‘e-'f_éfs{ued: A . Declassified with deletions Jawwary 7, 1960. GAK BIBGE NATIGNAL LABGRATORY S '_w-_. Operated by : CARBIDE AND CARBON CHEMICALS COMPANY G A Div1sion43ffln10n Carbide and Carbon Corporat10n1='” B Post fiffice Box P ' : Oak Ridgeg_ Tennessee : l E G A i N O T { C E : 'I'b:: report wan gmpamc! a8 an. mcmt of Govammem -ponsored work, Neithzr the United . States, nor the Commission, , BOT &Y, perso:: tcting on behali of the Cammissions. " - RN !zkcs mymmty 9T repri d:6r {mplied, with respect to: the aecu-: |° " raey, yer 5a.0f t.tw mfcrmatioz\ contained in this, repart. ot het the use . ] Tof any. iflormflofl :Wm matbnd o process djselosed in t.h,ls reparL mse.y mt lnfringe . : Pty el - | “privately owned #ightsior L : ] o S i AL Assimes any Habilitles with reapect ta the use: of orfor. dzmages resuiung from f.he' : ) el Ca U E us of aoy fnformattion, sppara.m method, oF prodess: discloed in this report, . RS v CE AR aaed in (the above,: persnn ;cung on ' behalf: o! the: Commisamu ihcludes: any em~ : = 2.1 Y ployee or. contracis ‘of the C fons o adpl 2 of such dontracter, fo.the’ extont. that - L§such émpt 5t or of fiw"‘ : Y lovee o #uck contractor ‘prepares;. - disséminates, of profldes acceas 20, my fn[ormtzon pursuaat o b.!s emptoyment or com‘.ract L : with we‘f‘bmmiasim o’ a[e empiujmfint wlth .mch ccnzracf.ar Vel . DISCLAIMER This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. DISCLAIMER Portions of this document may be illegible in electronic image products. Images are produced from the best available original document. " ;Beports prevzously 1ssued 1n thls series are as follows 8 ' 5ioBNL-s237 'ORNL 529 _ ' OBNL 768 "ORNL-ssa omwesis esodee e . REE S L SN 22N : esanee o ; Per1od Endlng November 30 1949; :'Perxod Endlng February 28 1950: :Perlod Endlng May 31 19503 Perlod Endlng August 31, 1950 Per1od Endlng'December 105,1950 ® seesee oY sooeen o i Yeedew e T e seases P e e L . YT T I : . P H——— TABLE OF CONTENTS S B . . Part I. REACTOR DESIGN | | A “f INTRODUCTION To PARTT _-j_”.' s 1. DESIGN OF THE 200-MEGAWATT ATRCRAFT BEACTOB S 28 S Core Design R R SR R U EPULIN TR IR 28 ' . Core materials ' _f L : Ut S 32 | Cantrol of the Aircraft Reactor’_ B S . SIS ;33wu.zfi Inherent stability ' : o o ' = 34 Liquid-fuel-level control system - . . 35 Primary Coolant Circuit ‘ | : | '1 '38 7 ~ Pumps. | R 39 . Intermedlate heat exchanger _ 42 2. BESIGN OF THE AIRCRAFT REACTOR EXPERIMENT " R 45 . Core Design ’ S o L 45 Control of the ARE Reactor . o o o 47 Fluid-~Circuit Design . S e 4T Test Facility Building for the ARE o B B 48 : Experlmental Engineering for the ARE | : L o : 51 Fuel circuits ‘ . o - _ BT ~ Coolant circuits - | | S o -7 S s Ifistrumentaticn ~ I T L 583 3. BREACTOR PHYSICS' L | R S TN -1 Introductlon B : R - e f.v'f:_ __"p_f ; _'56' Reflected-Reactor Cr1t1ca11ty Calculations. : 59— Mathematics of calculations : | A1 R - IBM calculations ' Lo e S0 863 Reflected-Reactor Solid-Fuel Calculations o 64 ~Effect of core composition and core size | - .68 Effect of reflector » ‘ ST 75 | . _ Effect of xenon 4 ‘ S o 80 o R Effect of uranium lumping : ' SRR -84 L "~ Effect of iteration of the source term . - 84 T ~ Statics of ANP Equivalent Bare Reactor | m o | 89 ' Statics of ANP Reflected Liquid-Fuel Reactor ; . © 93 D Kinetics of ANP Reflected Liquid-Fuel Reactor | ) 97 N Statics of ARE Bare Reactor _ 97 S Kinetics of ARE Reactors : ' o ' 106 — Perturbation theory studies o B S - 106 Maximum pressure in fuel tubes | . | 115 Critical Assembly Calculations ' _ | | - 116 — Glossary of Nuclear Energy Terms R o o o118 soan0 . . XXX XY ‘avaen e : _mITI(:AL EXPERIMENTS G | Apparatus for the Cr1t1cai Experlments e | ~Fabrication of uranium dlsks_; 9 ””ff Flrst crxtzcal assembly | s NUCLEAR MEASI}BEMENTS | __,_ng._ff _: Malyhdenum Cress-Sec*lon Meaaurements _ o i The 5-Mev Van de- Graaff Accelerator' = o D . 3 o S ‘Research program . e s 123 '-.3 Heehan:aal Velacxty Selector' :_fi B T e ol 6 cfi‘smsm&' OF LIQUID FUELS EOHRER R N e L ) e S Phase Stuéles of Fluorlde Systems” o f '-.]__'- ]31 7r_f: ffff7?':iIZ?f;flfl.f< “ L Sodium fluorxdew wranium fluoride o oo 0198 o Potassxum fluoride—~uranium fluorlde o 128 Sodium . fluor1de—»pot35a1um fluorzdewmuranlum fluorldea o130 SRR Sodium fluoride:- mberYlllum fluorxdeumuranlum fluor;de'-V"';‘ o 31397” o Fxltratlon Exper1menr5;_' B . R N A i ' N Snsp&nsxfins of Uranium Compounds in Sod1um deroxzde e L Stability of uranium suspensions = et 138 e g Ident:flcatzon of the u*anlgm_pnmpound:Q e o139 _ T ) Part III MATERIALSgEESEARCH; | S INTBODUCTION TO PART III "”'*1 ']f' SRR T e e s B g e e _' 11 c@fimszm Expzmxmm'rm\f o gl e Statxc Corroslon by quuld Metals R R _ -f___f {jVfi __'}: flj182f§v-J.~ Testing technxque e e 189 ~Static corrosion . B . Blffu31on Ba"rler as a Cerosxon Inh1bltor L e g fiynamzfi Corrosion by Liquid Mezals . .. o208 " Thermal convection 10093-««;harps I B ik e S Forced circulation locps - fxgure-elght laeps e e g e R Statla Corrosion by Fluoride Melts - R A o219 __‘“j"~ffi"- Pre*reatment of fluoride eutect1*--u=_':- o e o ey Hasults'f-- L s ey ~ Conclusions S L R e R LIQUID erL A‘\D HEAT TmNerR as.smcu SRR EEEEEEE 2:-2'6_””‘ _“Sodxum Hydroxlde Hean Transfer Studxes L 298 . Boiling Liquid Metals o e 280 © Heat-Transfer Coefficient of quuid Sodium 230 L Study of Free Convection in Liquid- Fuel Pxns 232 nHeat Transfer to Noncxrcular Ducts '13 o o '*;-f'_-77233'¢* oodoc IR S ° » - . £ ] Theoretlcal Thermal Entrance and Turbulent Flow Analy31s" 933 Physical Propertles - o A N R g3 Heat capacity LT -= S S R o237 Thermal conductlilty of llqulds | o 938 Thermal conducthlfy of solids .~~~ L 243 - VlSCOSlty of liquids . ::.;”__ _' ;”;.j . L+C; : ”“ S _: {.: '246_ L e Den31ty of llqulda e L T e T e e e D AR Moften Fluorlde Electrlcaleonduct1v1ty Measurements .i_' B '71 251 “'.‘ - 13. COMPONENTS OF LIQUID METAL SYSTEMS R o 283 } Pumps S T . SRR LR R R ,-f« }f_ '5' © 283 Eiectromagnetlc pump for flgure elght loops' s S 253 : . Centrifugal pump for figure- elght loops e o 254 - - Centrifugal pump for ARE | SRR o 254 Pump for 200- megawatt alrcraft reaczor-- R - 28T Canned rotor pump T S esT L,Turblne pump EER T _ _ ;u_.";,5 , '_'4fh. o 238 | Bearlng Tester Design S T T _260” Operation of Seal-Testing Dev1ce R o - 260 Electromagnetic Flowmeter o ' o 262 Calibration Loop ' T S T o L 263 Flange-Testing Device o _ SO L - S 263 - Stress-Rupture Tests ' B ' - o 263 - Insulation Testing | | "j_ ' ' SRS A 263 " Heat-Exchanger Tests : 1 RN 264 - - Cleaning and Disposal of Liquid Metals and Equlpment ' _ - 266 Sampl1ng Procedures = o : L _ 269 Purification of Liquid Metals and G&S&a ' o SRR S 270 _ Liquid Metals Safe:y Committee . S - R 271 Instrumentation ' e : S L R R 272 _Bulldlng Facllltleb far the Experlmental Englneerlng Group - _ 272 14, METALLUBGICAL PR{)C‘ESSES .' e L B Y ‘Welding Laboratory - e e o sfigfi’fi“’w Creep-Rupture Laboratory B - ... .. 293 Equipment installation ' _ ‘ o o 294 Tensile tests with cyelic stress f o 1295 ~ Physical Chemistry of Liquid Metals : o E 1295 ' Liquid-metal corrosion of single metal crystals. ' IR - 296 Processing of single spherical crystals o « 296 Hydrogenous Flulds | . _ o 3 SRR - 298 sboa e '3,”15.3 BABIATTON DA%AGE SO ' ' Creep Under Irradxat;on S - RIS RO SN '_ Cantllever creep apparatus e “:_ ,5'3 Tensile creep apparatus_'f_ 3 o L o Thermal CORdUCthltY of Structural Alloys at ngh Temperaturef Y-12 Cyclotron Creep Experxments :ffl; R Purdue Cyclotron o - R " Properties of. un1rrad1ated molybdenum i S Resxstlvzty of molybdenum after 1rrad1at1an'* Hug" - Creep lustrumentation __j R Effects of Radlatlon on a Fluoride Fuel Part IY ALTERNATIVE SYSTEMS n INTBGB{ICTION 'ro PART I‘V "'Tf*17.; VfiPQB.CYCLE ANE HELIUM«CYCLE SYSTEMS “'f | Yapcr-Cycle System | High-temperature materlals_ Haizum Cycie Analy31s ' 18 SUPEECRITICAL w.m:a CYCLE i EIE Heat Transfer and Fluld Flow 5 - SRR | Comparatlve heat-transfer stud1es of flaw types | :4;;55T;i - Study of flow znstabxllty - Sl " Review. ef heau transfer data Begqg;;v}_ty R RS o DTS . . SRR Equlvalenb water reactcrs fcr 1ron-water mlxtures NN Results of reactivity studies o S Bynamlcs and Cont*ols _' Properhles of Mater*als - Ll " Embrittlement of sneel by atomlc hydregen Cuxrasxan of s&eel in wateri" - = ;;19-._ SUPEBSO’\IIC TUG TGW SYS’I‘E\’( -1-6: HOMGGENESUS CIRCULATING FUEL AND CIECULATING %iODEBATUB REAC’IUBS o e i ot NPT ST St S . - XX SR N semeRn : L ) 3 ; - el : LT RS T2 ANALYTICAL CHEMISTRY Determlnatxon of Oxygéhlzfi Sodlum . . n-Butyl bromide method S ". Sampl1ng'techn1ques " Conclusions fBetermlnatlon of Oxygen in Argon and Hélzum — Brady method ..”}'JThérmal Stability of Dow Coraxng Sll1cone 011 550 - N __LCcntral Pregram for Spectrcgraphlc De:ermznatlon of Trace Metals : in Sodium '-_'Sarv1ce Analyses Analysis Analysis of uranium- compounds and mlxtures_ of berylllum fluoride .'ffiAnalysls | ngna1ys1sf <. Analysis- - Analysis of sodlum | of 1ead and blsmuth of metals and alloys of miscellaneous samples "f: fSummary of servxce analyses 22, LIST oF REPOHTS ISSUED Eo __ _23;- CHABT QF THE TECHNICAL ORGANIZATION OF THE ANP PROJECT | . '_.IQ » . - * ;'9000:9 - » sesemar vevsar o . @ se ‘:_-336‘ L 336 G wfia;-;fasa;j »gu-. a Q;33g:f“_,'p; o339 o339 : 349-'1 ST .f5341;'_';3~ & S 342 T 343 S LIST OF TABLES. | Table 1.1 -. E§ac£of Core Daga' l;g =l ; f3 7- i ':fi ; if '5 ;;.;; '; F i 29 i; ; - | Téblé 1;2 .,'Temperature and Préséure Summary.(Core, Primary Coolant ..: : ; ” : Cm R s Chrcu1ty'and,Part of SecondarY Coolant Circuit) o400 _ ,Tébi§ yAx1al Flow Pump SPeCLflcatlon Data f?f- f;f "-f'; '1} .]f f f4?* f-f n © Table 3 .4'T'_Heat Exchanger Spec1f1catlon Data l¥ ::? : : I._[f_irfif”:fJ{ ;w 44 1. 1 Effect of Core Composxtlfln én Crltlcal Mass of Standard Beactor -72 i Tabié 2 e . S ] o - e ; N © Density (p) Coefficients of Beact1v1ty ‘and Uranium Mass (n) NN . in the Standard Reactor SR T _-_2___ f'-f T. 'T3'; .”T: Tab1é. Sh1m Control Requlrements' ' .' _‘ '_¥ ;:f _' .J: _ : 1; 491T'. : ;Tabie - A | Table Table ~ Table ABE thm C@ntrol Bequlrementa'-- _75 R 1 } }; 'f'f _a17 'i:f1Oé_ :" ' :;00nstants Employed in the Klnetlc Calculatlon" _A£7; . .f fiii9:fi -~ WA b o Pawer and Average Fuel Temperature Equatlons for the ARE _ o ' Resuitzng from Perturbatlon Theory Analys1s - 11 ‘Table 3;8: _ Calculated Beact1v1ty by Su§d1v1d1ng Energy Groups'f o .. ,:117:>“ | Table 6.1 Determination of Ternary Eutectlc in NaF-KF-UF, from - '- i:.” ' b Flltratlon Experlmenta 13y SR ' Table 6;2., Effect of Tempera*ure and Leng*h of Agzng on the StabllltY _ _ _of Sodium Monouranate Suspensions in Sodium Hydroxide . 137 TS S | S AT " Table 11.1 Static Corrosion Data Obtained in 40-hr Tests Made in | R | ; Iron Capsules at 1000°C in Lead | 185 " Table 11.2 Static Corrosion Data Obtained in 400-hr Tests Made.in . '.':::H . o Iron Capsules at 1000 C in Lead ‘ 198 SRR Table 11.3 Static Cflrr051on Data for Tests Made in Nzckel Capsules : o o | - in Sodium at 1000°C | S “5;_fiw 198 nllTéhle-11.4 .Statzc Corr031on Data Obtained from Tests Made in Beryllla_' “ _ | Cru01bles at 1000 C with 2U- 9881 Alloy | - 199 i RILLE LN FEE N EN ] . EX) tTéble-ll,S ' Cérrbsionland Operation §§te$'of Thermal.Convéctiofi Loops.._fi_ 1 2Q6: " Table 11.6 Materials Under Test (Stacic Corrosion by Fluoride Melts) 213 - ~Table 11.7 Results of Corrosion Tests on Fluoride Eutectic o ' '}._224 Table 18.1 Calculated Equiv.len- Density of Iron as a Function of the | : Iron-to-Water daiis h - - . 315 Table 18.2 Fuel Requiremen® for Iron«Wanef.Reaetars | o 31T Table 21.1 Determination of Oxygen in Bulk Sodium by n-Butyl Bromide R Method ' | . 338 ‘Table 21.2 | Results o£ fietermihations of'Oxygen_in Helium ahd Argon - 340 - | Tab1e 21.3: .SufimarY.o£ Sérvice Analyées' 2 :;- __' , i _f f .L:-7;:; :;Tiflf3§4g ; :” e L P soadee. .. FYIitL ® : peswee o ,'3¢335L1qu1d Level Control System_fif-' .' Pr1mary C1rcu1t Pump {Axxal Flcw) _ ;: ,Intermed1ate Heat Exchanger :' ; f,Fig;;z;i;; _ pre11m1nary ABE Cere Arransefient | , ' fli Fié;€2:2: L Fl°°r Plan of ABE Test Faclllty Bulldlng | R _ Fxg 23 ._.._.._~’End Ele"atlons of AHE Test Facfluy Bulldmg : ' i . 50 j._ F{é; 3;i :;51ow1ng Dawfl Denslty vs. fiad;us (807) : ,_ A_.-_ '::?6d}}f} o fig,_3;2 : fS1ow1ng Down DenSIty vs. Radlus (839) _' o :. 8 H: '_5 T _; 1§2iJ Fig. 3.3 Spatial Power Distribution (Reactor 804) 66 ” Fi§,fi3;4 "fNOrmalxzed Fxss:onlng Specbrum (Raactor 804) e ' i _ . 3 §? ];_ :fi | _“;if{é{fl3;$ _;':Ncrmalzzed F1531on1ng Spectrum (Integrated over Core) ; ; ” ffig: ' '.: {F£é;;3;6;_: Spat1a1 Power Dlstrlbutxon (Heactor 806) _'.. | _-”' :-: ? ”ijfQfl}f _ liijié; 3;?:f_?ENormalxzed Fzss;onxng Spectrum (Reactor 806) | : .:~. _ 7i:' o i'Ffé,_3§8 ” 'Spat1a1 Power Dzstrlbutzon (Reactor 813) '13_ ._ _ ;_ f'“ ' f4f : 7” Fig::3;§:? _Spat1aI Power Dzstrlbut1on (Beacter 808) Ny L f.' ” ?6 : .. | 'Fig;.g;iflf_’ Normal;zed Fxsszonlng Spectrum (Beactor 808) .'._ : _ : _ : -7? ; if . Fié;'§;i1 . Spatzal Power sttrlbutlon (Heactor 812) .; A :”faf8_ fi o __Fig;:é.izttfi Effect of Reflector Thlckness on Reactlvxty | ."_ f'__Tf7§;' ”.. _ '-;Fi85.3{;3:. Spatxal Power Dlstrlbutxon (Reactor 855) | .H 81i.' R Fig.'3;14' Flux Dlstrlbut1on (Reactor 855) / R o ' :§2 f 5 Fig..3;15 | Normallzed F1551on1ng Spectrum (Reactor 839) B | - : : 831 : . :Fié. 3.16 Spatzal Power D13trxbut1on (Reactor 822) | ._ | .;._* i“ 35 _ Fig. 3,17 t Narmallzed Flsslonlng Spectrum (Reactor 822) _ | . _:. '86 TFig, 3.18 .Self Sh1e1d1ng Factor as a Functzon of Lethargy (10 m;I Fozl) .3?{ L R f§§¢f-'~.v. l._‘d' aseebe cashe . O, i . » P sseome seaeee. T T - Figsl.' ' :Flux Response to a Step Change in Beact1v1ty of 10"3 (No_-.- 7ff'Bfi1k Mean Fuel Temperature Response to a Step Change in ".:Flfix’Response to a. 75 C Step Change in Coolant Entrancef _Bulk Mean Fuel Temperature Hesponse to a Step Change in 'Nermalzzed Fzsszenlng Spectrum (ABE Bare Reactor - No Xenon) _Normallzed Flssrenlng Spectrum {Cr1t1cal Experlment _Total Cross Seetron Curve of Meiybdenum .Phase_Dlagram of the NaF-UF4 System fe :Self Shleldlng Fector as a Funct1on of Lethargy (20 mrl F011) '_rIteratlen of Source Term: | "_' 'u Normal1zed Flssxonlng Spectrum (ANP Equlvalent Bare. Reactor) 'Normallzed F1$Slen1ng Spectrum (ANP Reflected quuld Fuel Reactor) -iSpatlal Power Dlstrrbutlon (ANP Reflected quuld Fuel Reacter)i _jDelays -:Ne Xenon) '_rReaet1v1ty ef 10‘3 (No Delays - No Xenon) -Temperature (No Delays - No Xenen) 44'Bu1k Mean Fuel Temperature Bespense to a 75°C Step Change in | Coolant Entrance Temperature (No Delays ~ No Xenen) a Fleg Response to a Step Change in React1v1ty of 10'3 (No | .Xenon) AR React1v1ty cf 10' (Ne Xenon) .Normallzed Frssronrng Spectrum (ABE Bare Reacter'- Equlllb- :f-A o _ _ S o - 108 rlum Xenon) .Change in Power for 100 F Step Chenge in Inlet Coolant Change in Mean Fuel Temperature (°F) for 100°F Step Chenge Fn in the Inlet Coolant Temperature o Aseembly No. I) ' Cr1t1cal Experxment Apparatus '.Total Cross Sectlon Curve of Molybdenum (Low Energy Taxl) Phase Diagram of the KF-UF, System __The Ternary System NaF-KF-UF, - - seoees . LT sawsee se000 e e easese cous YYXT T 90 .u ‘1:9‘L ' .. | e _."'1"0-1- 102 13 S107 1i2 ' 113 119a 122 124 129 _;r131 i ‘fStatic Corros1on Testzng (Capsulat1ng Technlque) ”~ 3’ia.f ; i§3 . Stat1c Corrosxon Test:ng (Tabuiatlng Technlque) N € ':::; -<”i87: o Armco Iron and Type 405 SS Expesed to Lead ‘.i ' ] §83 | Type 430 SS and Type 446 SS Exposed toLead 189 _»'Nlckei and Beryli1fim Exposed to Lead 1 _; : ~i=. -:; a 1j_'”” 191E :; : ’_ T1tan1um and Type 304 S$ Exposed to Lead j : - :: 19é,:35 ~ Type 316 SS and Type 347 S Exposed to Lead LT g .:. '.Type 310 SS Exposed to Lead . ' 1 *;” "j__ _;*V ;_"' _; “{'194? '7;«11;1Gf‘ Molyhdenum,-Inconel and Inconel X Exposed to Sodlum -:1.}:;: 4':§Q§f ._ “:r'i1.11 ,‘Type 316 SS Tublng and Tnconel Tublng Exposed to 51 ] Alloy_ :: ‘édi.:_fi_ [ ' "11312 TYpe 310 SS Exposed to Lead ' DR SRR R :-.x ;;1 ‘265 '% 1 2 3 | | 4 = .”.,:11;5: ; Molybdenum and Tantalum Expcsed to Lead ;: 3 .'.'ffi;:'lE. '1§65 .:_f | . SN | 1 8 9 ig. 11.13 , Flaws in Welded Joints Which Tend to Promote Premature Con~ T ’ vectlon Leop Failure | R RRT L' ".-il.ié_; Steps in Capsule Preparatlon for Fluorade Corros1on Testlng o 216. ' ;'11.15’ quuzd Fuel Carr051on (Monel and Nickel) S .'2i7. | ig. 11.16 quUld Fuel Corrosxon (Inconel X and Inconél) o . '3._7 2ié: _ ig. 11.17 Liquid Fuel Corrosion (304 S3 and 304 ELC 5) I I TR ig. 11,18 Liquid Fuel Corrosxon (347 ss and 32188) i-'~ ,jzzq;" 11.19 Liquid Fuel Corrosion (446 SS and 430°SS) R ‘:']-:251f: 'f son - ee - otk s T L e T R e e T L e L R e T S R D T e e e . Fig. 12. :;- iF£ggT: 'f'fFig{“ . Fig. 12, - Fls Fig. _': . Fig 1 ._ '3_ f,:Fig;. ’ quuld Fuel Ccrr031on {410 SS and Nichrome V) quu1d Fuel Corr051on (Globelron) ,Test Sectlon for the Determlnatlon of the Heat Transfer Coefficlent ef NaGH : '\., IZS;TTemperaCure Dlstrlhutlen Along Walls cf Rectanguiar Condult f for Constant Wall Flux Heatzng " Temperature B;strlbutlcn Alang Walls of Equxlateral Trlangular fluct for Constanc Wall Flux Heatlng . .:5:Temperature fllstributlon Alang Walls of Blght ?rlangular :f Duct for Constant Wall Flux Heatzng | flHeat Exchange Sectlon af Ice Calorxmeter. %.:A53embled Calorlmetrxc Apparatus’i HQ ' :Therma1 COnductlvzty Apparatns - Stsady State Apparatus for3 quuxds '{Praposed Apparatus fer Measurlng Thermal B1ffu51v1ty of L1qu1ds:f Langxtudlnal Flow Thermal Conduct1v1ty Apparatus fcr Sollds-' B : Cell of Electromagnetzc Pump ._Seailess Pumplng System : Mock Up of Heat Exchanger (End Vlew) '$1ngls Crystal Spher of V'Assembly Detalls of Beacta 2 NucIear Power Plant Instal;atxon_ _' C&Rt’flffigai ?um? A$8&mbly .I S Bear;ng Tes er Assembly _Insulatxon Taster ”Mock flp of Heat Exchanger o LDensxty Apparatns : . E Tg&?f " ”_Conductxvxty of the NaF U? Fluorlde Eutectlc :i L 252 ectralytxcally Etched Copper o 'and Heat Exchanger sasess . i : ; L] | SUMMARY In th1s edltlon of the Quarterly Pragress‘Beporf of the Alrcraft Nuclear' .:Propu131on PrOJect the research is divided into four phases in order to _”-fa0111tate the presentatlon and the comprehen310n of the work Each of these | phases — Reactor Design, Shleldlng Research. Materlals Research and Altef- .natlve Systems - 18 separately 1ntroduced so that the reader may know the- ‘ spec1f1c areas covered in the subsequent detalled élscu331ons . The d351gn of the Alrcraft Reactor Experlment (Sec 2) has been developed 3from the deSIgn of the qulescent llquld fuelwnllquld metal cooled aircraft react0r (Secg 1), _whzch was establlshed early in the quarter. The'reactor tcons1sts of a 3. ft square cyllnder w1th 6111p801da1 endsx'empioylng berylllum :ox1de as. a moderator 'sodlum as a coolant a mlxture of UF, BeF NaF as fuel, 'and 1nconel as the structural materlal The llquld f1 uorlde fuel is contalned' in tubes suapended in the coolant streams whlch traverse the moderator Thls '”_1nt1macy of the fuel elementa and the heat transfer medlum ensures complete “continuity of heat flow from fuel to coolant = an essential feature of high- ;power ‘aircraft reactors. 'The maximum design temperature of both these reactors - is 1500 F (a materials llmltatlon) which is adequate for a subsonic aircraft power plant The ternary fluorlde eutectlc: UF'lBéF:'NaF a?péars"fio be a satisfaétéiyfi llquld fuel ~both in uranium content (about 80 lb/ft ) and meitlng polnt .(about 900° F) An alternative fluorlde fuel UF -KF.NaF, possessing a higher "uranlum concentratlon (about 120 lb/ft ) is available although its meltlng f p01nt (about 1920 F} is h1gh (Sec 6). The radlatlon stab111ty of these fuels ' appears te be good from the ev1dence obtalned S0 far (Sec 15). t The phenomenon of mass tranSfer in a blmetaillc'systém restricts the ..,Tfabrlcatlon of the fuel and coolant c1xflu1ts and the internal structure of the core tc the same monometallic material: From the atandpoznt of corrosion 'molybdenum is ideally suited for this fask ‘but fabrication difficulties have so far prevented its consideration for the aireraft reactor. Molybdenum is, nevertheless, hlghly regarded as the future. COHotTuCtlon material for high- temperature reactors Inconel has been specified as the construction material of both the alrcraft reactor and its prototype, as it is the only metal which, “at this tlme 'is known to be reasonably compatible with both the sodium ~ 18 casoae s Y Yy L 3 cosbee seee - tficcq!t - snaee ] B evoae - ae 3 [ EXFERS o 'are listed below. ‘:'cooiant an& the fluorlde fuel and at the ‘same tlme to possess the other B .metallurglcal requlrements ef hlgh temperature strength weldab111ty,_fabr1—e_' o catablllty, etc, (Secs 11 and 14) Berylllum is a better moderator than berylllum ox1de from nuclear ‘con- slderat1ons however, the latter is spec1f1ed for thls reactor as it 1s-I relatlvely inert and may be readlly contained. (Hereg_agaln, the potentlallty of moiyhdenum is evident as it is the only known contalner for berylllum ) " The coolant, sodium, is compatlble with the oxxde although thelr m1x1ng in - the reactor is not lntended,_ Sodlumy 1nc1dentally, is a satlsfactery coolant;'"“'"' f:'a welélng and stress rupture laboratory for the Metallurglcal Group (Sec 14). " The bu11d1ng medlflcatlons for the Experlmental Eng1neer1ng Greup (Sec 13) N are not. completed The ANP Cr1t1ca1 Fac111ty (Sec. 4) and the theld Testlng" 'Reactor (Sec. 8}, wh1ch were essent1ally completed durlng the last quarter;'. ~ have both achleved cr1t1ca11ty and are belng used. The 5-Mev Van de Graaff | .“(Sec 5) is now helng 1nstalled and may be applled to nuclear research by the ‘;eend of the next quarter The Test Fac111ty Bulldlng for the ARE (Azrcrafte:' ‘Reacter Experlment) 13 now belng de31gned (Sec 2). In addztlon,ethe persennel empleyed by the Azrcraft Nuclear Propulsxonf Progect has 1ncreased in number so that there are now 256 technlcal people *.’engaged in eil phases of the research work, not xncludlng tredes and sub- =techn1ca1 peeple,' The eapaclty of this organ;zatxen is supplemented by the' -Awerk and sklll of 21 consultants and eight subcontracted sc1ent1f1c labora- ~tories. A chert of the technlcal organlzatlen of the Aircraft Nuclear Pre— : ‘1.pluezon Pro;ect at the Oak Bzdge Natlonal Laboratory 13 glven in Sec. 23 ' azseaace RESULTS From the bady of research now underwey on all problems assocxated wzth. this progect, some of the more tangible results obtained during the quarter . :::f19 e _. R O O T L D B L R b R . R Y [ L5 ORE D e A% . be. 0 Ss8e ¢ s 06 e e TN TR TR T S RN I I e TN SN 3 SR R & & [ JO SR AR L U Y Y RN S ] B R ) » . ese @ . T R TR N PERL AL AT RSN N DA RSN N ] L] Hle L ele 8. IMRL U eE 0 PO U 5 B RO D8 o s03 oo The facxlxtles of the project have been augmented by the completion ofe«. ——— - . o — . . "Pfiys§¢s -.. | 1. The analyses (Secg 3) of a. berylllum 0x1de——moderated sodlum cooled stalnless steel«wfabrlcated reactor whose statlc characterlstlcs are very 51mllar to the aforementloned alrcraft (ANP) reactor show the f0110w1ng fi(a) A change of moderator from berylllum ox1de to berylllum would : gt 'decrease the crltlcal mass requirement by 20% R (b)) The increase 1in cr1t1cal mass caused by removing a volume of a : - given constltuent equal to 1% of the core volume is- approxi- "mately +3.4% for beryIIIum ox1de, -1.0% for stainless steel, -and +G 2% for sodlum U, -_--'-1' R AT o (e) A change in reéflector thlckness frem 5. 51 to 7. 09 in. would .~ . 'deecrease the critical mass 36% and the median energy of fission from 45 to 14 ev.. The savings in critical mass from subsequent increases in the reflectorn however, would.not be 51gn1f1cant0f | 2. The maximum react1v1ty change:xxthe ANP reactor that must be overcome by Shlm control totals -5.0%, of wh1ch 20.5% is for &epletlon, -2.7% for equlllbrlum xenon, wlfil% for fuel expan81on$ and -0.7% for additional maximum xenon (Secq 3). 3. The Calculated'kinetic responses of the ANP reactor to a step change in reactivity of 10°3 and to an entrance coolant temperature step change of ictivity of P p chang 167°F.indica£e-that the reactor is well damped and bbmpletély safe (Séc;'B); 4. The calculated critical mass expected in the crltlcal experiment with the berylllum -uranium-aluminum assembly underestimated the experimental critical mass by 50%. D15crepanc1es between the exper1mental setup and the calculations are being exam1ned_(S¢csfi 3_an&_4)@ 5. The slow-neutron transmission curve of molybdenum indicates a strong ' resonance at 46 ev with other dips in transmission at 145 and 180 ev. The ‘line which best describes the transmission 1n the low- energy reglon is given ’kby oy = 5.7+ 0.30E°% (Sec. 5). £ [XFIEL] ‘seoese gpno0e S seeGee "Heat‘Transfer N 10. An exper1ment to determlne the extent of free convection in the " reactor fuel tube 1nd1cated a temperature drop of from 1 to 2°F in a vertical tube fllled with mercury, in which the temperature drop without convection would have been about 13°F. This impliés that convection in the ARE fuel - tubes will lead to very good heat- transfer conditions (Sec 8). | 114' A 10ng1tud1nal flow apparatus for the determlnatlon of the thermal ‘-'conduct1v1ty of sollds up to 600°F has been developeé whlch is capable of . 'accurac1&s of from 5 to 10% (Sec 8) ‘Liquid Fuels Chemistry 12, The equilibrium diagram for the ternary fluoride fuel NaF-KF-UF, has been reasonably well established, The lowest eutectic found to exist consisted of 27 mole % UF,, 29.5 mole % KF, and 43.5 mole % NaF and had a melting point of 1020 + 20°F (Sec. 6). : | o 13.. The equilibrium diagram for the ternary system NaF-BeF,-UF, has not been completely established, but a eutectic consisting of 12 mole % UF4} 17 " mole % BeF,, and 71 mole % NaF with a melting p01nt of 900 % 20°F has been .located This uranium content is hlgh enough to be of 1nterest as fuel’ “material (Sec 6). T ' '4‘2-1, | L S s - : . E s ‘a8 o800 @ ° .8 N6 4 GBG & p83 0e o i el « 0 ® ¢ o & . ee 8. e -0 @ . 3.0 ee o ‘e e e . . ®. & &9¢ LR X o ® s 8. 8. & © Ly oen [ TR R A * 8 .9 . e e e e e B 8 e L3 L EE 2 1 [ IOK DR R *0. . 8P 8. oo0 0 ¢ 48 89 s 6. . 8 ‘eses o0 - ' 14. The 1n1t1al 1rrad1at10n of the f1u0r1de eutectlc NaF UF4 in a. neutron flux of about 1012 '1rrad1at10n 1nduced pressure rlse, as mlght have resulted 1f free fluorlne gas _-fi .'8_ had been formed (Sec 15) 15 One molyb&enum sample 1rrad1ated for 4 hr by a 3-#a beam of N Mev .“4 fdeuterons ev1denced no- detectable varlatlon in re51st1v1ty after cne day (Sec 15) v Héta!lurgyf ' S ié Mbnel and nlckel A have been found to have aatlsfactory corr0310n: ”"rates (about 0. 03 mll/hr) in the fluorzde eutectlc (NaF UF, ), ;although ‘the - 'compatlblllty of these metals w1th such’ other requlrements_as hlgh temperature ' fjstrength is not favorable (Sec ll) 'Fl?V-~An extensive'summary of the corrosion behavior of molten lead under ”Estatlc condltlcns has been prepared (Sec. 11). Results of”thése tests with “varlous metals 1nclude the follow1ng . (a)' Irfln showed no- metallagraphlc ev1dence of any attack althcugh' “ g ‘a small welght loss corresponding to a surface. er951on 0f ;_j O 0005 1n was detected after 40 hr at . 1000 C. '_(5) Steeis conta1n1ng 12 to 16? chromlum also showed very llttle . evidence of corrosion, although there was some surface pene- . - tratlcn to a depth of 0. 005 in. after 400 hr at 1000 C. o (e) Austenltlc stalnless steels at 1000 C show more extrefie surface_ instabilities than iron or the low-chromium stesl, including lead solution, 1ntergranular pefletratlon precipitation, and ':decarburlzatlona- : : - Do EETS 18, The statlc corrosion of metals with sodlum is 1in general less severe 't}mzw1§11na8 fiaclaleVLdences no corrosion, and molybdenum, tantalum, tltanlum, 4 1ncone1 347 stainless steel, and 446 staxhless steel appear:tb be glmost'as SatiSfactory (i.e., 0.003 in. or less thickness change in 40-hr tests at '1000 C) (Sec 11); 19. A dxffu51on barrler of chromlc ox1de has been trled as a corr031on. " inhibiter for 310 stainless steel. A static corrosion test w1th lead at 1000°C for 100 ‘hr 1nd1cated that the depth of the attack was not reduced but the attack was, nevertheless less severe (Seco 11) - - LS b . GmE. o @ 8. . e ®d B - sss & ave 8 Ceoe e e e e e e e le el e LK) ¢ 8 o o we ° . o e 8 - ¢ we . @ w0 . e eel Al ® s e e e e "8 .8 LR 28 Tele e e e e P L3 o o e S S ee see . 9 ses € e L s .80 &9 per cublc centlmeter per second ev1denced noV V:{AItétfifi#i#é:Systefis ' ”29 The supercrztlcal water reactor:u;belng analyzed by Nuclear Developfi ment Assoc1ates, Inca- Prellmlnary exploratory work on heat transfer,_fluld '_flow, and react1v1ty have been encouraglng (Seca 18) , 23: e a8 ‘adaB@e S s K Seseanw seases . se e . ] i ...- L] Z, W I ' o W) ; , ted = : N B [~ ; o @ : : et , -, S fad -4 # o+ * P ~*f}]zwtaaancifsfiereipaflr41 - The alrcraft reactorg bezng a hlgh power (290 megawatt) reactor,_ls ' character1zed by ‘a high heat flux, the requlrement of a large heat transfer': | area in a small ?olume, and consequently,,a rather 1ntrlcate core de31gn._. ;e;The last quarter has w1tnessed the estabilshment of *he prellmxnary d351gn of the qulescent l1qu1d fuelmwllqu1d metal ceoled 290 megawatt azrcraft reactor. This de31gn 1ncorporates a core arrangement in which three fuei tubes are_-' ' 1mmersed in each coolant tube In a berylllum ox1de matrlx.." The complex1ty of the overall alrcraft power plant system nece551tates' = the separate development of the major components of the system, Consequently,' the deta1led &es1gn ‘of the Alrcraf* Reactor Experlment (ARE)gza low- -power. test :-_reactor,_has been undertaken wzth the aircraft reaetor de31gn as a 901nt of ."departure.. Thls ARE core d331gn 1s 1ntenfled to dupllcate so far as ;s' 'practlcal the materlals and klnetxcs of the larger reactore Although'exacfi -dupllcatxon of the ANP reactor is net p0s51ble, the resultant comprcmlses do not mltlgate agaznst the usefulness of the ABE as a devzce for proving the Tpresent core concept in all detaxls con51stent with rapld progress"_The prellmlnary de51gn of this prototype reactor ‘has beefi ccmpleted afid procure- ment of the larger pieces of " the reactor has been initiated. The ée31gn of a "_bulléxng faclllty for'thls test. reactor has been completed and detalled spec1f1cat10ns are being prepareda’___' Throughout the desxgn of hoth the Ak? and ARE reactors the ealculatlons: of the ANP Phy51cs Grcup have dlrected the layout of the care,: Numerous multlgronp calculatzons have been requlred by the englneerzng variations and 3sh1fts of interests that were 1nvest1gated ‘The nuclear characterlstlcs of reactors which have been 1nvest1gated 1nclude statics (e.g., crltlcal massg flux dlstrlbutxonsg power dlstrlbutlons, reflector effects) an& &ynamlcs (e.g,, the tlme response of the liguid fuelu~l1qu1d metal cooled reactcr te ‘changes in react1V1ty coolant temperature failures) The program for corre- lating theoretically predlctedvalues of crlfilcal mass with those obtalned from -actual crltzcal experlments was 1n1t1ated "A'satisfactory liquid:fuel fqr.these reactors appears to have been found in the ternary fluoride system beryllium fluoride~-sodium fluoride— . uranium fluoride., This fuel provides a suitable uranium density in a fuel 25':e seeeee T LI smvae . ) se@ee e [y L d [ Y X X2 8 [XTIr ] : o es B0 e susbae sclutlon w1th a satlsfactory meltlng polnta- Other solutlens of uranlumf Ifluorlde 1n alkallne earth fluor1des ‘are belng 1nvest1gated._ A schemat1c drawzng of the reacter and core, representlng that of eltherfi “ the ANP or ABE reacter,'ls presented in Fig. 0. 2. The essentlal components 'cf these l1qu1d fuel-—llquld metal cooled reactors are noted together with | 'speclflc references to. the text for dxscus31on cf each feature.- R N 26-27 | L ] ecevae e ; . : 3909046 . ® endeo ‘ssosee . (XXX XE S sesens 2 ; BY SR snenss seee & * e 1 DESIGN OF THE 209 MEGAWATT AIRCRAFT REACTOB '_ B W Schroeder, ANP D1v181on | Of the f1ve QUlescent IIQuld fuel reactors Whlch were dlscussed in the:se -~Iast quarterly report (ORNL-919) the halrpln core arrangement in which three tubular haxrplns (fuel ‘tubes) are 1mmersed in each coolant tube was selected,: for the 200- megawatt (ANP) reactor,' A study of the other alternatlves, howfi ever 1nd1cated that any one mlght be capahie of successful developments since it is not poss1ble at this time to deslgnate any szngle deszgn that warrante'; .i_deveiopment to the exclu31on of all othersa_ Nevertheless, as the early_ 1n1t1at10n of a development program leadlng to the ANP reaeter is deslred and” B as much of the 1nformat10n to be acqulred will be general in nahureg 1t was "_'necessary maselect a part1cular reactor fromtfim:more attractlve p0551b111t1es; fThe overall arrangement of thls reaetor (Flga 1.1) wxll 1nvolve a minimum of ad61t1onal research and it 1mposes minimum constralnts on the size and shape-e _-cf aecessory components, It 1s emphas1zed that even thls deszgn as 1llustratede is prellmlnary in form and that many months of analy31s and component test1ng e_-w11l %e requlred prlor to the establlshment of all features._? : The major compenents of ‘the reacter arrangement 1nc1ude the reacter core and pressure shell assembly, four 1ndependent control assemblles,-51x Interw_ Imedlate heat exchangers, 51x ax1al flow prlmary coolant pumps, , i L '_“gTheSé in&ividual components are'descfibed separately. o CORE ::BIES'I_GIN_ The core. (Flgn'l 2) is nomlnally a-3- ft square cyllnder w1th e111p501dal ends_ 'employlng berylllum ox1de as moderator and reflecton@ a mlxture of UF, and NaF as fuel'-and inconel as the structural and cannlng material. The_ velume percentages of these compenents in the core are given in Table 1.1, The core 1ncludes 2268 parallel coolant tubes arranged logltudlnally and 'spaced by cransverse perforated and dimpled disks occurring every 2 in. . along the-longltudlnal core axis. Three "U-tube™ fuel elements are located ewithin.each coolant tube and are supported by the fuel manifold system. Te minimize the number of header welds, "U tubes™ rather than individual straight . e23 ; ew eae e e & se i oe 9 Wes & . sde oe L e ‘4 e & . ®.0 8.8 D ERS LR O CeLees @ el 8 8.8 e e 90 T 60 0l 4, SR8 et e s 8 e8e L LANL R L EERE el e e e e e @ ® e 9 o 0 e @ igel oae ° LY S : e & pas gob- '?'"Reactor'ebre néta NN . TABLE 1.1 MATERIAL COMPOSITION . MATERIAL (£t3)' * VOLUME f‘Voluée‘%.SOF‘ - TOTAL CORE © WEIGHT (b . Fuel ._Fhei tubes. . Mbderator o Mcderator cans _{'Structure ] Coclant - . Total inconel. - Total core Ffiééd_fluéfi&etéaits: - Inconel - Inconel - Inconel ‘Sodium Sje10, o ¢ 0.42 139 418 4 2. 01 L : 17. 68 ‘ 783 ._.?.5?.14. - 23.65 - 100.00 e 210 1600 - 3378 “uxschLAREQ§S'QATA' ' flbétQtféhsfer area * Fuel tube aggregate length . Free flow area for coolant (% af core cross-sectzoa) . Number of tubes ' Number of ‘tube sectzans Fhel per section Fuel available for control : Fhel tube diameter Fhal expansxon allowance (not in core) 'F13310n_gas accumzlation allowance (not in core) '_”1049 ft2 39,600 fr 23. 65% 15,700 (7350 “u cubes") 148 _fi.6’{_5%'of wtal fuel .16%~.' ©0.100 da. .'6%,. : - 10% - ceeesre p 2g”30 eee000 R 1 RO N aesees J o CONTROG, BECTOMN | et} o : . : U DG NQLY-FT-86 ) % i FQ3% O Cter, & Al O ConLART Thses N i . S T T . ’ DEPARATELY CABAED FROM AN ANE OF CORE b \I - . By, Rpk Bor YoM HEAZERS T Co I - ' . PR S0 SOOI = M. N . & %;&%%u‘:"’ HOLA Liva L » . Awmers M Ak : e PRessaRE SuL . . e PLEFENTS N R AR N . : - - - : i '. 2L HEASERS Maflxdb.m e e 820 foweL H . . : e 60 59 SN i - e SO T AL IFSULATION - . Lo . . e : Ll Ty OO e SO DR INULATI | . et op s b r _,” - L frESears A00AC ML\ ' "- / 500 ik GNP SU—— Covrams, Saerme Common it OF0 - i ‘ —— \\ DN & e oy G Bk, -0 R - L e TS SRy X fm TUBE BEND GETIS 7 8rr DETan D, D YFT 62 i hny 3 REpTar ot T . ) T P00~ T WALES ek et rmw Borvam L . : WeApERs . T Coor gy Fhsx 50000 DRy TAER T Sy SeProaT s K T WALz, T g - . R - ~ FIGURE 12 AIRCRAFT REAGCTOR CORE ARRANGEMENT tubes (as descrlbed in sectlon on. control of the alrcraft reactor) are ema; ployed for alI fuel elements except control elementse The fuel headers to'o whlch the fuel elements are welded are arranged S0 thattflualnd1v1dual ny- tubeme _ legs are connected to separate headers, de51gnated as “1nlet"«and "outlet™ for ‘load1ng?-unload1ngg flushlng, etc, The fuel headers are partlally empty durlng operation, serV1ng as accumulators to accommodate changes 1n.core fuel volume. assoc1ated w1th fuel temperature changes. One hundredifldiiftyxune fuel element sectors are employed each of whlch:_ ”contains a proportlonate share of fnel elements, one inlet header and one . ”outlet header‘_ Thls degree of compartmentallzatlon favors fabrlcat1on and '1nspect10n and minimizes the conseguences of any one or more fallures, since" ¥any ‘one fallure will involve the loss of less than 1% of the 'tot ai fuelf volume,- Several ‘such’ fa1lures can be compensated for by control’ afijustment,r Inlet and outlet f1111ng and flushlng tubes connected to the individual headers, are grouped together into a spider=shaped harness and hrought to a multitube conduit passing through the pressure shell. The conduit is routed 'through the shleld to an external filling and flushlng station, The arrangement 1llustrated emhodles features that were selected after rev1ew of many alternatives. It was found that thin fuel elements are required to minimize peak fuel temperatures which tend to become critical, because the thermal conductlvxty of the fuel is estlmated to be of the order of 1.0 to 1.5 .-Btu/(hr »ft-°F), It was decided to use cyllndrlcal fuel elements rather than plate type elements for maximum r831stance to differential pressures. Tubes_- of small dlameters and lengths not in excess of several feet minimize hoop etresses assoc1ated with fuel egress acceleration forees- These facters, plus conSIderatlons of eurface area and 1nc1uded volume requxrements, dzctated ‘the -dzmen51onal constants listed in Table 1. 1 _Gore Materials. The material selections which have been mentioned are " not necessarily final, but these materlals have been tentatlvely adopted on the basis of current data. Fuel. The fused fluorides have been specified to permit the negative temperature coefficient and because of the relative ease of fuel drainage and replenishment associated with liquid fuels., These factors are discussed in more detail in Seecs. 3 and 6. 32 L] - avesde L Do smenen ’ ) s o . 4 XXIT LN *RQO02D a - 'Cbolahf. Sodlum has been 6631gnated as the prlmary and seconéary coolant because exlstlngwiatalndlcate that sodium canlnecontalned at desired operatlng temperatures by several materlals9 1nclud1ng certaln stainless steels nlckel, __and 1ncone1a_ Strucffire; The ‘core structure 1ncludes an lnconel pressure shell cooled externally ‘and 1nsu1ated 1nternally. The top dome is free of prlmary coolant -p1p1ng9 fac111tat1ng its removal. Within the pressure shell a structural , 'cy11nder is’ supported by a spllne joint in a manner perm;ttzng freedom for : axlal and radial expansion but restralnlng center-of- graV1ty movement. Moder- 'ator welght 'COolant tube wéight and coolant- tube drag forces areflt:anéfii ted ' by compre551on to the bottom core dome, wh1ch 1n turn transmits the }oads to the maln flange, B Inconel has been spec1f1ed tentatlvelyg as prel1m1nary data indicate it to be compatlble with the sodium coolant and with the fused fluorides. If further data indicate type 316 stainless steel to be compatlble with the fuel this material may be preferred "ModérdtorQ The use of’metallic beryllium as moderator offers some decrease in fuel investment compared to beryllium oxide: Beryllium possesses favorable thermal conductivity and ductility, but there is evidence that it . may”allby with any of the readily available canning'matériaisp and further 'research is required on the eilmlnatlon or mxnlmlzablcn of this ailoylng, Pend:ng resolution of this problem, BeQ is regarded as the spec1flc moderator ”'ma;erzal, and the anaiyses included here are based on its use, 'CONTROL OF THE AIRCRAFT REACTOR E. S, .Be_t‘tis‘, NEPA Control Gréup’. " Consideration of several features of the aircraft reactor indicates that the control difficulties are mitigated by the reactor design which is inherently stable for fast power or temperature transients, This stabilivy is achieved by using fuel elements with high negative temperature reactivity : L o B .f AR . : e » :;“;, u é ' D s esw ‘& w - “aa s 6. ses ¥ ave 6% : . &% e a9 '8 el 8 . el o 8.8 @ s 8 o & L d - & L L3 * 8O o 06 L ] LS IR TR L Gcos .- [ I R 2 . @ e o et e 8 e e [ BEK ] AR R A . e [ ] s sae ) sed. 8 g ko 08 -0 ¢ s see o3 'cdefffcients.r Slnce the. temperature tlme response of the moderator 1s abou%“' .'two orders of magnltude slower than that of the fuel ‘stability of the reactor to fast power or temperature tran51ents can be effected by the ch01ce of fuel ‘elements only and is 1ndependent of moderator choice.. The response of the coolant is slower than that of the fuel and the sodium coolant is a fieor.: moderator, a The expans1on of the fuel prov1des the necessary fast control for the reactors; Thls automatic regulation must be supplemented by a slow act1ng shim: control to compensate for large ‘and relatlvely slow changes in. reactor power9 1901son1ng,_and depletlons, The shlm control is effected by a variable liquid- _fuel volume in - a separate fuel assembly w1th1n ‘the active Iattlceg This 'system i3 actuated by a temperature. 51gna1 from the cooiant outlet and auto- .matlcally reguiates the fuel volume to malntazn a constan*-cooiant outlet_ 'temperatare. ' Inherent Stabilltye A fuel consisting ofzacomblnaclon of moleen fluorldes ' has the highest volume coefficient of thermml expansion of any system studied. | .lThlS type of fuel has been selected ‘to provide the maximum change in reactivity w1th temperature rarlatlons. A reactor whose changes in reactivity result 'from changes 1n fuel volume provides a mechanism for holding the fuel témperaQ ture to relatlvely small excursions about a bulk mean value, This type of control will allow wide varlatlons in fission rate or power. Since the reactor _temperature must be kept under control for safe operation, 1t is felt that the temperature control prov1des adequate safeguard for the reactor. Whlle 1nherent Stablllty to fast power and *emperavure tran51ents appears to be a hrghly desirable feature of the fused-salt fuel elements, it cannot be cancluded that the system will not have obgectzonable oscillations in power '_or temperature ‘transients unell carefui quantl ative anaiysee of the response of the system to such transients are madea Such analyses will give fur;her. ‘information on ranges of temperature and power for whlch srablilty is satlsu 'factory. _ The anal?ses of the response of thevsystem to fast perturbations show that, even with the large damping factor contributed by the deiayed neutrons omitted from the calculations, oscillations are small in amplitude and are ~highly ‘damped in power ranges abeut the design point, The resultscxfthe studies . show that the pressures developed within the fuel tubes by fuel expansion are 34 ssscee : . soenes . ‘sodsow ' veo8s 0., . 'not excessxve even for assumed magnltudes of reactl?xty step changes much,' "more severe than could reasonably be expected to occur in- practlce, Liquld Fuel Level Caatrol Systemc The liquidmfuel reaétor permits_the' "use of a portion of the fuel fOr dynamic COHtrOla. The dynamic control s?stem of the a1rcraft reactor requ1res the use of approximately 10% of the fuel: volume.- This volume is located in the center of the core where 1gs removal- will have the Least detr1mental effect on the power dlstrlbut1ong_ To mlnlmlze Iumplng of the fuel in any one con%alner, the system has been divided into fou:.elements,- Each element is. 1ndependent1y aetlvated and 1nterlocked so ~that no'mere'than one element is actlve at one time., These contrel slements are llmlted in maximum rate of ?rawel so that ‘the Ak/E will not exceed approx- "zmately 2 x 10°% per second; The coolant ouble* temperature provides the 7331gnal £or operaelon of the contral system'_ The system has no clo:e=e01erance moving: parts in the hlgh temperature region, Flgure 1.3 111ustrates the .'prelxmlnary 11qu1d fueiuflllquld metal cooled reactor contrcl systema A descriptioneafthe control 3ystem'core element, fuel reservoir, actuator 'unit;,and operating cycle is given. ‘Core Element. The core control elements extend through the core in a vertical direction and are so spaced as to introduce a minimum local power - variation. The upper header normally contains gas maintained under a constant 'pressureg The lower header is full of fuel and is fed from the fuel reservoir. Excluding the header ?olume each of the four core control elements contaln' ”approx1mately 2.5% of the total fuel volume of the cere. The amount of fuel "1n thls volume is controlled from the fuel reservoir. Both headers are u{separateé from the core fiux by a B C shleid Fuei Reservoza, The fuel reservoir is attached to the upper pressure shell and so arranged that the reserve fuel will be heated or cooled by the ‘sodlum-cqalanta On the initial start-up it will be necessary to keep this reservoir above the fuel melting point, but after some interval of power . operation the problem is reversed to one of heat removal. ~ Actuator Unit, The actuator unit is similar in construction to the fuel reservoir ‘but contains the transmitter eutectic in one compartment and hydraulic. oiliin_the other. The diaphragm is spring loaded to provide a positive force for removing the fuelrffom the core. This force, augmented by the gas pressure above the fuel in the core element, will serve to remove the fuel in the event 35 ® T . [ X X N2 N3 sensae LR R X sasese . L > o - _.‘_.... Cewne . sescaes ssesae . B F&ééfifi&w«b&; U Do peeig: o . i & ee s sase0e . ebotad LI sesdes eecsns e w @ * ¢ . U SECsyN Pogiriant : T InoesroR A rue x’g&‘ 3 LupPwe S5/, Hrags uwre O Leroect Soans e ‘\\fiywfiwuflig o Fureerr Ling - EonTROL Feeas Bepeore; .. TR G F2rpaf Eirreeree L rusroe U S P, Lower o . Disconnteer &u‘ Coprace 7use Prve™ " 1 h .' ,y%k’} Eorearic 7 ) FIGURE L3. LIQUID LEVEL CONTROL SYSTEM ef'c6mp6néfit7fai1ufe&_ The volume of 011 in the system at any tlme is regua'_ :lated by the hydraul1c control valve, which 1n turn is fed by a 9051t1ve dlS“ : .'placement pumpe Thls pump is so adJusted that the 011 flow w111 not exceed ~ the value determlned as . safe for the 1n*roduct10n of fuel into the corea~-The '” control valve:&sPOSItloned by a Ilnkage wh1ch14;controlied by the Proport10n31 *““' ";soleno1d and the shaft connected to the actuator d1aphragme Th1s llnkage 1s so deSIgned that ‘the proportlonal solen01d controls the p031t10n of the dia= 'phragm and correspondlngly the 1evel of fuel 1n the core element, The solen01d Cis. 1nstalled in such a. way that it will remove the fuel. from the core control : B element in the event of electr1cal fallnrea There 1s a selsyn transmltter_ ccupled to the élaphragm shaft to 1nd10ate fuel level to the operator, A "heat1ng coil, which will have'nhIEid water carculat1ng through it is PEQVldéd:' “. n-'to protect the eutectlc from freezlngs.' Operatnon, The:coolant outlet temperaturefi whlch 13 mon;tored by a numher of thermocouples immersed 1n the coolant stream prO?ldes *he control ".¢szgnals, The 31gnals are flrst compared to the operatlon level set point and. -then ampllfled and sent to the proportlonal solenoid. The level set point may 'be Varzed by the operator to meet the demand, but, once adjusted, it should 'nct requ:re “any change durlng normal operatlon@" This set point is éute? 'matxcaily overrzdden to set hack the reacter temperature in the event in 'fallur& of engznea pumps; auxxlxary power supplys etcg .. ’ . . PYYYY LI XX E X . +aRIE . [T *e e sesne - I E X R 2 3 J - L ] - AL LN LE-EE ] YT L PRIMARY COOLANT CIRCUIT N Whe primary circuit employs generously sized heat exchangers and ducts to minimize pressure losses and to . maximize secondary circuit temperatures, _The_arrangement (Fig., 1.1) provides zintermediate heat exchangers each with its own primary circuit pump (as 111ustrated) ‘and its own secondary circuit (not shown). The separation of the heat exchangers and secondary circuits isolates the consequences of any secondary system fa11ure, any primary pump failure, or any intermediate heat exchanger 1nternal fallure, Primary coolant returnzng 38 L-E RN 2 ed . sace couses aseess 'to the eore from the : pumps is mlxed in’ the core 1niet scroll and header, ~before the coolant is d1rected through the ccre coolant annul1s therebys protectlng agalnst local core starvatlon in the eventtyfprlmary ‘pump stoppagee Back flow through any dead prxmary pump is mlnlmlzed by the resistance of the: 'heat exchanger and ducts in serles with the dead pump as well as the res1ste' 'ance of the dead rotor 1tself thereby obv1at1ng the need for check valvesg'*' _ In Table 1. 2 are glven the calculated pressures and temperatures through« _out the core, the prlmary coolant c1rcu1t and a portzon of the secqndary_i _circuit. Pumps;_ Various types of pumpss‘xncludlng electromagnetlcs canned rotor :‘an& mechanlcally driven pumps, have been studled. At thezr current respective ,stages of development the mechanlcally driven pumps appear to be guperior for "alrcraft usage from the standpolnts of we1ght 51ze; and eff1c1ency._ Wlahlne .thls category a broad band of rotor types may be des1gned ranging from pure centrifugal at the hzgh head—-iow flow end of the range, through mixed row 1mpellersg'to axial flow at the moderate head—high flow end of the‘range,. The flow rates and heads required by the preposed system fall within the range permlttlng the use of ax1al flow pumps, which are preferred for their com- pactnessa- Such an axial-flow pump (Flgg 1.4) has been de31gnedg Pertinent -pump data are ngen in Tabie 1 3 Shaft seallng is provlded by a comblnatlon of 1nert gas pressure and a -géentrlfugal slinger. The sllnger 111ustrated ‘should be adequate under the most adverse operating conditions. Seallng under static conditions is provided by locatlng the gas seal above the hlgheSu 901nt in the sodlum system, Inert ‘gas pressure is ‘applied to the surface of the sod1um in the primary fluid make - -up tank and also admitted, via 1nterconnect1ng lines; to the shaft' housxngs below the gas seals, This arrengemefiL should ensure, under static conditions, that the sodium level in the shaft housings is maintained hydro- statically by the make-up tank level, Rotor thrust is taken by anloil-lubrin’ cated thrust bearing installed above the gas seal, 'The'pfimps may be driven by hydraulic9 pneumatic, or electric motsrs, and the choice probably will be dictated by the overall power transm1381on system de31gn of the airplane. 39 ssaces . LR B R 1 U ® a0 ’ ® e ® ° - Y ] cseans - - 'TTIE Y [ X% X2 * covwen essnne @ IR T . it S PR H A" - a L awes ?ABLE f;2 : Temperature and Pressure Summary - T o GENERAL SPECIFICATIONS | Total power o o 200,000 kw (139 600 Btu/sec) A _ Flow rate, primary circuit . B I 1800 lh/sec : S : Flow rate, secondary circuit - ‘; 1800 1b/sec ' - Coolant average properties: 3 K ' | _'4 RO R S Specific heat o 0.3 Bw/(1b°R) Density . - SR 3 o N e 50 ib/ft3 . Thermal condnctIV1ty o B T o 3% th/(hr-ft- F) Visc031ty . SERTI R R o -],;_f":f* ST & 5 1b/(hreft)- R PRIMARY CIRCUIT PRESSURE (psi) TEMPERATURE* (°F) o Pupidnles. s o 113 : Pump discharge R e f :3 - L 1130 Péessure shell inlet x : ;‘88 : o 1130 . . Core matrix inlet - ' RN 80 c '~‘ 1130 ' ' Core matrix outlet - ' ' : 6T - 1480 _P“essure shell outlet ' , - 62 B SR . 1480 Intermediate heat~exchapger 1nlet ' . ; ' 57 S . 11480 ~ Intermediate heatmfixbhanger outlet IR 53 1130 Pump 1nlet SR ) : T ST | AR : o 1130 - SECONDARY CIRGUIT .Ifiterme&iaté fieat-ex;hgngef inlet _ :£S' = 10 : _ 1070 Intermediate heat-exchanger outlet . - SR _? B o 1420 " REACTOR : Division of power production | L T . §4% in fuel, 6% in moderator . Avg. power density in fuel ' o S T4 Btu/ (secsin.?) v - , Avg¢ power density in moderator : o - 55 Btu/(sa;¢in.s) ’ - Max. fuel temperature ' ‘ : B 2650?F’ .. Mex. temperature on inner surface of fuel tube : o . 1540°F c» ~ Max. temperature on outer surface of fuel tube ‘ 1500°F ‘ Max. primary coolant temperature . E 1480°F " Max. moderator temperature ‘ , - 0 1130°F . Max. temperature of moderator surface®* A _ 1690°F Max. temperature of inner surface of moderator can “ 1488°F Max. temperature of outer surface of moderator can 1483°F . Max. primary coolant temperature (ref.) _ 1480°F : *Neglecting minor heatrinputs and losses. _ y — ‘#*Based on assumption of contact thermal resistance coefficient of 400 Btu/(hr-ft2~°F). Tew T ees w e 2 8. 0@ B EAE F 49D 06 o w8 ». % ® e s o . o e. a 8@ ¢ 8 o0 e . o . @ “ slew @ o8 .0 e e . P o an . e 2 & e 8. @@ e e s @ 9 * . 'R RIS ° e o &% 808 ® 960 ® v ss &8 2 8 * sec e U pwawo. Y-FT-48 . HETE fi.:_du.«_g: Ie3 BATRN BLET AL . WA Weasgdrion B Fr.0d 1 ¢ (AXIAL FLOW) * L & E ] L] . L LE ] L] FIGURE 14 PRIMARY CIRGUIT PUMP | TABLE 1.3 _Axial-Flow Pump+ Specification Data . Flowperpup . gofs | . Pressure rise so'psi;*f”zsoffu=ef sodinm ‘Specific speed.*nj 7 R '6000 BRI R T o Act§§l speed o ' . o . - 6860 rpm- Est1mated efflclency .Z" . _30%, | - Pump power requxrement.: ;: 157'hp§a¢'.— | f_;Blade tlp diameter .,g _ o 6;9.inr - 'Hotor huh dlameter.' ; _ '3.6_ifiég - *The pump béafing is of the selfécenaering hydrestatic type. "T%e estlmated system pressure rise requirement is of the order of 43 psi. Pending experimental verification of this requirement, the pump has’ been designed for a more adverse requirement. : '**2? hp per pump for the calculated system pressure reqnirement of psxg : _ _ ' Intermediate Heat Exchanger° The problem of heat transfer in the inter- 'medxate heat exchanger is quite comparable to that of the reactor core. However, the relatlve 1nsen51t1v1ty of the airplane weight as a function of heatuexchanger size permlts the use of somewhat larger and thlcker tubular elements. Cyllndrical elements are preferred for max1mum re51stance to 'dlfferentlal pressures,'and counterflow 1s desired for max1mum secondary' coolant.outlet temperature, Two conventional methods of prov1d1ng for thermal ' VSxpansion'of the tubes are the "floating header™ with a convoluted bellows fseal and bent- tube construction, The former may involve bellows stresses that3 are unacceptable at the intended operatlng temperature unless the pressures in the prlmary and secondary 01rcu1ts are’ substantlally balanced at all tlmes, which may be undesirable as well as difficult to ensure. Accordingly, the bentfltuhe or "hockey stick" design (Fig. 1.5) has been selected. Heat-exchanger specification data are given in Table 1.4, in which the volumes and areas are based on one heat.eXchanger, 42 - ) S ‘ . s - Sy s R X F . B O, o T TR ¥ S . Mol Fa & G0 . .aEa ERLER e 89 o @ 8 @ .8 . e & 8 ¢ o g8 @ @ o e & e &S 8 M8 &8 . ° . ) son . e e o & e e 8 e g B O ¢ @ a9 '3 ¥ ®.e e [ X X L 3 e 9 .0 O 8. @8 P Y X T 3 [ LS .4 Las @ sebeey FYYEISH e e ’ . ‘seeeme. I XXX 5 . » -0 o seadse L ] eame . ey ) AR n& RN % Y 3 B N X 3 % 3 4 X A ) o S o Loy B = sScele: Sl 7 IGURE 15 INTERMEDIATE HEAT EXCHANGER TABLE 1.4 Heat-Exchénger spéc‘ific‘atio:fl Data : .: Thbelfifitér.diametér‘ :" ' B f3-._ 'ii. ': _. & §5 in' T wall thlckness R 0.018 in, | ks R Thbe spacxng, center to center 3 ._.__ .. ' '.... 0. 3134 in. (staggered p1tch) | .Act1Ve volume f :.1_.'_. . ; 2 4 : i; 4 ft | e ’ i Hydraullfi dlameuer, pr1mary f1“1d f. ' '_33_0 02383 ftr(1n51de tubes) .-'Hydraullc &1ameter, secondary f1u1d.' : | o ";_fl 01307 ft 'Free flow area, prxmary or seccndary ' fi . 0.4229 ft3 .'4 Heat~transfer area, mean = S ”'_ . 400 ft2 . ..144 .. e . ® ) T o [X X 2] e ] LA XX ) XY LY X [ sihanae FYY X T 3 AN coseeR. . oo o Lo : coeene . anes L4 * . ° sssone . TR R S le2a DESIGN GF THE AIBCRAFT REACTGR EXPERIMENT B W Schroeder,_ANP D1v1310n_-2 The general features of the 200 megawatt alrcraft reactor have beeng postulated "and a prellmlnary de51gn ‘of the core, shleld and fluxd CIrcu1t has been achleved (Sec. 1). With thxs deslgn as a point of departure, the desxgn of a low»powered test reactor,-the ARE 1ntended as a prototype of_ :',certaln features of the 200 megawatt reactor, was attempted The complexity: ” -_of the overall alrcraft power plant system appears to be such as to warrant '1fseparate development of the magor components prlor to any ettempt to operate " the entlre systemg Accordlngly the Alrcraft Reactor Experlment is regarded-e' 'as a core and control system development program with shielding and fluldjo"' _clrcult components de31gned for 31mpllc1ty and adequacy rather than for fazthful ‘aircraft slmulatlon The current desxgn status o{ ARE components 13_' - outllned below,_j CORE DESIGN - The core design:is intended to'dupiicate, insofar as is practicable, the ‘materials, the temperature patterns-and'the kinetics of the 200-megawatt reactora_ Studies have indicated that the temperature pattern and certain of ;the control klnetlcs constants would be almost dupllcated if the alrplane-31ze ~ fuel elements Were employed w1th the number of fuel elements reduced pro- | portlonately to the power reductlonn Such an arrangement woeld have in- . sufflczent fuel volume, however, and compromzses in the direction of more and '__'larger fuel elements appear to be necessary - The more conservatism employed in estimation of the critical mass, the greater will be the fuel volume pro- vzded and the larger the departure from the 1ntended temperature ‘and klnetlc' SImulatlon It therefore seems best to defer flnal determznatlon of requ1red fuel volume pend1ng criticality tests, although it is desired to initiate procure- - ment of the pressure shell, the moderator and reflector ceramic blocks, and the sheet metal and metal tubing as early as possible. A core has therefore been designed (Fig. 2.1) that will permit freezing the components requiring early procurement while retaining fuel volume flexibility. This flexibility el ees. e @ - . e B8 8 woR O Aes B i s 8 4 e @’@ PR T ) ] e Loe &0 : g P A AR S SIS U FETRNE S, S ST A RS ree . s stw 5 e ke Lo B Lese e w0, [ L - . e [ !. . se eee B . 086 8.3 88 . N 8 9 4 cas. 2 Y-12 DWGNG Y-F7-7Q s P T HERDERG SR SEIERATR ALY ErEL kb T CONNECTEDR o M QUTLET SR 77 LA (v BLEELR —~ WERIER SRPEDAT A3EK YR HEADER ~ Big Beatws, " ThCk - Eo -V B8, c.r TN . ! Toni Comaant Taar 2220y DZOWALL aaam@mfimr CBEFTNick " NV \W\ufl\\fl\ m@% ? \_\R WL\\ \\ \W\\ o \W\A\ @\ »_ \ \ 0N NS \ \\ \ \ A h B g o a T i T -\..4 P (S e ) T T et J ,flw - o 7 T -1 \\\“ \.\\w\\ P s i \fi\ Clffome ARGUPEIALT P /f““wflfifiwfrffibe NN == ] : . R 5 \ I/.. \\\ \\% \“ g 2 \“Mfikfi ¥ g T e \..VN‘ \W _m_ T.fi., : / ; fi,. x : 3 X m 3 2 A ; Yo R X m L * ; ) X 3 ¥ x ¥ WM X X ,n * ,lr E UH X % 5 X 1 m * * : i 4 B 3 i 3 e : 3 ¥ * % . f % ] Y i R . % T 4 H 'PRELIMINARY ARE CORE ARRANGEMENT A E * e : 200080 oo®e & & [ X212 Y] 'FIGURE 2 .Tj 1s achleved hy speclfying two alternate tube 51zes and by de51gn1ng the: :coelant tubes such that four,_flveg.sn:,_seven9 or ezght fuel elements af : ::Zielther size may be accammoda*ed by each caolant tubec;Thxs arrangement permlté_f ' a crltlcal mass range from approxlmateiy 4 to. 25 1b, based on a uranlum_ ‘i den31ty in the fused salts of 80 lb/ft ; CGNTROL OF THE ARE REACT&B ff E S Bettlss NEPA Control ffian "' The control is 1ntend99 to dupllcate as. ClOSdl;IJSIPOSéiblé the contrdl lof the 2OG~megawatt alrcrazt reactor | The ilquld fuel dynamxc control system 0£ the alrcraft reactcr zequlres the use of approxzmafely 10% of the fuelf “volume.. Thls volume 15 lelded 1nt0 four eiements each of whlch has 2%% of the fuel and the rate Of insertion is. 11m1ted to a maximum react1v1ty change ‘__“of 2 x 10“ per second. The lower powered ARE reactor dynamlc control elements : _'wxll be scaled 1n such a manner as to glve these same condltlons w In addltlon to the llquld fuel dynam1c control system; it is also planned_' £0 1ncorparate a solld absorber rod in the ARE reactor. This B C rod will ‘operate at a maxlmum rate cf w1thdrawal of 2 X 10”4 &k/k per second w1th prov131ens for fast 1nsert19n for safety. Fast 1nsert10n of the rod will be .;xnltlated by the flux level, whlch will be adgustable fiver the operatlng range faf the reactor The 1nsert10n w111 be by means of graylty plus a sprzng or _ other 1 means of acceleratzng the rod when it is released from its magnet. It ' w11I be necessary to. ccel th& rod to allow materlals for slldlng or bearlng' 5[surfaces to cperate at temperatures fer which they are d931gned This cooling :can be done by clrculatlng water or some low- meltlng alloy, such asNa K, through the rod. The soll& E c rod can also be used to glve the reactor a; small perturbatlon in reacb1v1ty in order to check the caIcuIatlons on the react0r dynamlcs : | S LTRSS T FLUID-CIRCUIT DESIGN The function of the ARE fluid circuit is to'diSpose of the heat'génerated _' - as séfely and:as'économically as practicableg It was decided to employ sed1um: as the prlmary coalant because ef metallurglcal compatlbxllty The ultlmate @ . el [ Y O EXXT S K DRI sesese - ioiqoo;'_ R ) ‘assvae ‘goaaes’’ PR : RRL PN deenee s shes j heat receptacle must be expendable and both air and water were. studled Water:} was selected as the preferred agent because of the very nom1nal flow and'f_ 'pumplng requlrements asDoc1ated w1th 1ts use Safety con51derat10ns 1nd1cate the de51rab111ty of employlng an 1ntern . '_;medlate fluid that is nonreactlve w1th either sodlum or water A llquld wouldg? “be preferred but a revxew of avallable llqulds d1d not reveal any that are ' known to meet all requlrements g Aecordlngly,_lt is planned to employ elther -nltrogen or hellum as the 1ntermed1ate heat tranefer agentq The prlmary clrcult pumps make up tanks and.lntermedlate heat exchangers E are to be iocated in a shlelded pit adJacent to the reaetor pit. The 1nter- ..Lmedlate gas is conveyed from the heat exchanger pit to a pump roomj whlchv _houses thegas to water heat exchangers flnagas blowersy and other accesserles“. *-rEST'FACILITYfSUILfiING FOR THE ARE x _ The Teet FaC111tY Bulidlng for the operatlon of the Alrcraft Beactor jeExperlment has been éeS1gned (F1gea 2.2 and 2.3). This bulldlng is to cen31st uof a concrete steel structure approx1mate1y 80 by 105 f¢ with a one»storv ;serv1ce wlng andai40 ft hlgh crane bay The Austin Company has been authorlzed to prepare detailed. plans and spec1flcatlons for the Test Fac111ty The profee ”l-posed locatlon is approxzmately 1500 ft east of the HRE bulldlng In- the service w1ng (low w1ng) of the Test Fa0111ty Bulldlng will be the rcontrol Toom; change room, countlng room, :shops and offices. The crane bay -wzll accommoda*e a 10 ton erane and w1ll house the test beé 1nclud1ng the‘ eactor plt dlsassembly area, heat exchanger'plt ank-storage room, and pump -room,_ In the Auatln Company des1gn and subsequent constructlen of ‘the build- :1ng no 1nstallatlcns exflept the brxdge crane will be made 1n the crane bay area. Upon completlen of this construction phase by the bulldlng contractcr, '-.j the Laboratory w111 begln 1nstallat10n of test components and shxeldlng It 13 expected that by June 1 the AEC w111 have awarded a contract for the bu11d1ng and that six to eight months will be allowed fcr its completlcn° hThe estlmated cost of the bu1ld1ng is $400 000, ‘ R A8 Ceie ® * ¢ ‘eeteve 7 coedww . sene ' ‘.O. o8 [ ] Ceeeese o Lo H . . Y'Y YL LR eeeved '.-oono- PO essase e e, @ e e Teowe e . - ®. FYXTIEY e o ceowe’ . . B ) s AeneGe o seeeem, XX XN S ‘e @ cedbed T LS ° . e .® e XY XN ® ® £ R SR T £ * ol ' o oo LEA Na L ARBAWAY THHTIRANLY PP LRUNTING (g - arPLL . = | aneutse)” Chrp yEEMERL AN [£3 % 5%-3-1:9 Ee /"] ;jg“l B 3 T [reTRTeS HERERERNREERNRNS WhaLR ShulAA GHDP HRTAUMENT SHOP [l LonTapL C BAMP r_, (=] WBedh RNOLE UR BaoR LRI M TG e e AR Laoe s (3a8 e P WU —L.. L vy £ b%__mi - < owe fea meh awath | e CHANG K Ae® 10CUER . ] WECHMTAMNMAT I Pj RLTRING WAL e’} G RAv AT LRGN R E TRUCH. LoADIMG } Al FIGURE 2_.2‘. FLOOR PLAN OF ARE TEST FAG!L!TY BU!LDING Lo GR,QUND e 'FLQQR*PLAN ‘ Tl Hian BEMMTY coneeniE- U8 L Ia:: : t:':vfj STRHBARR RUMAIEY R@udo B8 10584 g_&:gtss' Reaty | _ ELEVATION 'w: NQQyHH . et == S0UTH ELEVAT_& QN S —, FIGURE 2.3. END ELEVATIONS OF ARE TEST FACILITY BUILDING ...... ........... ....... ........... Ll eesese .0 ooooooo ......... ......... .IOIDO.. ....... ........ P [ X R R LRI | EXPERI@ENTAL ENQINEEBINQ FQR ‘THE ARE H W Savage, ANP D1v131onfe” The de31gn for the ABE has suggested a number of experlments and tests, 'the scope of wh1ch has been studled ‘and- outllned in some- detall by the Exper1?~ __mental Englneerlng Sectlon : In general it w111 be’ necessary to examine the core structure and fluid CIrcults segmentally to determlne the fahr1cab11;tyge .'-ease of aseemblyy.structural 1ntegr1ty under condltlons of thermal cycllngye and the hydrodynamlc and thermal prepertles of thelr components In addition .'eto testlng components as recelved development englneerlng ef such systems _ 7p01nt1ng toward ANP type appllcatxons w111 be. pressed . Fuel CLTCUItSo The technlques of handlxng a 11qu1d fuel whlch is a solld "at room temperature and whlch nust. be preheated to about 1000 'F before- any etransfer can. take place must be determlned experlmentally The foilowxng';' :systems must be develeped for such a fuel 01rcult (1) evacuatlonp (2) flush- ;jxngg (3) lcadlngs (4) preheatlng, (5) transferrlngg (6) cleenlngy (7) pressur« "fzzlngs (8) monltorlng, (9) purzfylng, (10) stor1ng, and (11) 1nstrument1ng The flrst handllng of fuel for the ARE system, as such is expeeted to .rlnvolve problems of cieanlng stoxage purlflcetlon (f11ter1ng) and trans- fferral ef the fuel to the eystem without contamination. The prohlem of trans- eferrlng the llquld fuel from one location to another w1il involve studies of “both parallel and ser1es Clrcultsfl probably in combzna%1on "and under con- ditions where the circuits are pressurlzed w1th an inert gas. Itis anticipate& “that the entire sye&em will requlre remote 1nstrumentat1on for temperature pressures.flew volume measurements and leak detectxonv' Experlments wzll 1nvolve ‘mock- ups ef proposed de31gns in which hydrody«-‘ _nam1c measurements a* low temperatures ‘can be madeg ané mock ~ups and actuai_7 components which can be tested with high- temperature fluids for determining hy&r0dynam1c propertles at these temperatures. A means of accurate volumetric 'control will be requlred whlch wzll be adequate under elther radloactlve or nenrad1oact1ve condltlons and alse unéer dynamlc condltlons Prebiems relatlng to fabrlcatlen will depend in part on - determlnatxen of ‘setlsfactory Je1n1ng techniques which will resist both thermal cycling and. corrosive and erosive actlon of the flulds involved; in part on the ablllty to ’assemble the various componenfs after fabrlcatlon, and in part on whether- 51 avcaed; ° . FEYYT X s80800 ess e sansaw POy e . seseee seos anee o 'certaln of the 1nternal components of the oore can be subgected to dxrectr' contact w1th the f1u1ds The fabrlcablllty of the structure would be con- s1derably eased 1f certaln condltlons were demonstrated for example if it could be shown that is 1s not necessary to seal the proposed moderator beryl - "11um ox1de3 from the proposed coolant, sodium or sodlum potas31um alloy. If seallng is not necessary it should be possxble to provide support for the. -members Wh1ch w1ll change dlmen51ons under thermal cycllng w1thout prov1d1nggu rlgld connectlons between them : Coolant Circuitsqu The components of the coolant circuits will 1r”lude' all the p1p1ng9 pumps; heat exchengersg waste eas systems, etc. necessury 0 dispose of the heat generated within the core. It is. proposed to gave ‘”p' ;:the components of the heat- transfer system and to determlne the adequacy and _propertles of 1ts components 1nd1v1dua11y and 1n combination in order to '-determlne the arrangement necessary for these systems in the ARE fac111ty The general requlrement ‘is for the experlmental determlnatlon of techniques of handllng a llquld coolant which, in the case of sodium, would be solid at roomn | temperature and Would require preheatlng to at least 300°F before any transfer | would take place If the coolant was sodium-potassium eutectic alloy, which S is llquld at room temperatures it would not require preheating but would requlre pressur1z1ng at a temperature of 1500 F, , The follow1ng systems w111 probablykm requlred (l)evecuationy(2) flush- ”zng, (3) loading, (4) purification, (5) c1rculat10n, (6) draining, (7) clean- | 1ng, (8) coolant dlsposal,s(9) pressurizing, (10) monitoring, (11) instrument- 1ngg (12) preheatlng,ln.the case of sod;um, and (13) waste-heat fluid systems. Experlments will 1nvolve mock-ups of core sections including tubes and ’headers to determine hydrodynamzc propertles and thermal charecter1stzcs of these sections. Also included will be development of valves and flowmeters and other devices for controlling the locstion and transfer of coolant. A pump will be required which will handle about 150 to 200 gpm; for the ARE application it is anticipated that commercially available pumps can be modi fied for this purpose. Studies of electromagnetic pumps will continue. The techniques of cleaning the cooling system of the reactor and transferring 'coolant»to it in'a_pure state and of providing equipment for maintaining ‘purity of the coolant require considerable development in order to assure satisfactory handling at a maximum coolant temperature of 1500°F, 52.”e 3 . . A o 3 i N . . et i e se 0 . = L LX se. » see © gos @O . e e & s & 0 e & o [ IOESRAE EE DESSERN SN N LR s o se s ® LI 8 e g8 ‘e 3o s 0 % e . - 00 " e e e L L I ] L L] o ® 9 4 @ ¢ LI I I 4 [ (X3 ose - sce ¢ & . 0@ o 2 @ 6 ems go Another development wlll center around sultable plplng for transfer of. ':the coolant from the core to the waste heat dlsposal system,-and this w1115 '1nclude development of monltorxng dev1ces for leak detectlon and development of sultable 1nsulat1onu The waste heat dlsposal system will be located some dlstance from ‘the: actual reactor core and may znvolve transfer of heat from the coolant elther to water or to a gas such as air or nltrogenfi_ The szmplest ”system would probably be one 1nvolv1ng water Whlch could bo dxsposed of dlrectly, but such a system would requlre careful fabrlcatlon and 1nstrumen= - tat1on for leak detectzon A system u51ng rec1rculat1ng hellum can' be bullt’ oof conventlonal equlpment but is somewhat more - complexo- A waste heat dxsposal 'system ventlng dlrectly to air would 1nvolve few fabrzcatlon problems but f'wouXd enta11 rather more complex conductlng systems and safety dev1ces to pre-' vent general a1r contamlnatlon in case of breakdown 5 Other problems assoc1ated w1th th1s c1rcu1t wzll be encountered 1n the' 1ntegrat10n of the control dev1ces monltorzng the coolant temperature with the fcontrol devxces for the reactor ‘proper. These wlll 1nc1ude devzces for vernier type control and also emergency type control equlpment whxch in generai will o:act to shut down the systemas It is ant1c1pated that in the course of the experlmentatlon a heat source comparabla to ‘the output of the ARE (1 to 3 megawatts) will be a nec9351ty for' 4pretest1ng components of the system Fabrlcatlon toohnlques are’ not conszdered.' fflto ‘be an 1nsurmountable obstacle 1nasmuch as for. these segments of the ARE : c1rcu1ts the structures may be of reasonable szze,ithh the basic requzrement that Jolnts and removable parts of this system be fabrlcated in: such a fashion that: tlghtness is guaranteed over a perlod somewhat in excess of the expected ‘11f8 of the reactor as a Whole | ' e o IfiStrumentationafl InSfrumentétion for the réaCtor“systems'desoribed in .the foreg01ng seotlons will be divided between a more routine type of instru-- mentation for general control purposos 1nclu&1ng the 1ndlcat1on of basic conditions existing in the circuits, and the more elaborate type of control equipment designed to maintain safe opération'througheut ~ The Expérimental | 'Englneerlng Section expects that the initial instrumentation will be of the '_flrst mentloned typev' The latter type of instrumentation tequlred is the responslblllty of the Reactor Control Group and will be dzscussed here only as it is requxred to be 1ntegratod w1th the first type. . 53 D S ) 5. . ; i G ' : o . L . FOSEETY R T - e . 88 o Fws 8 ere ¢6 : son e el el e [AN S0 DT ST T S S SR S ] ie-e se @ @ PR IR PEREEE T TN Y FUSl S X TINAY R 2 % & 8 e o8 [ A s 8 a @ L I S 6 & @ L ® EAN DU R TR I | ee L] R ) s o * 0 ‘8 ses o® The follow1ng types of 1nstrumentat10n are expected to be requlred (1) temperature respon51ve dev1ces, (2) pressure responsxve devzees (3) vo}ué:1 ‘metric control, (4) mcnltorzngg (5) flow measurement ané control (6) streseg lndzeators (7) power measurement (8) control rod actuat1en (9) radxoo” : act1v1ty§ (10) reacthlty cantrol (11) mlscellaneous e}ectronle eqn1pment? "and (12} dev1ces fer personnel safety Thermocoeple.'re51stance b1 metal and.llquld type temperature respen31ve _ devxces for 1nd1cat1ngglrecord1ng; ‘and controlllng purposes w111 have wide Veappllcatlonc Most of thls 1nsfrumentat10n will be of the remote type, “and gome of the probiems will. 1nclude locatlng the temperature sens1ng element in ~restr1cted spaces prov1d1ng leads and 1nsulat1on between the temperature«.- sensxng element and the 1nd1cat1ng or controlllng equ1pment “and the- posszblee use of such equlpment for 1nd1cat1ng clogged 11nes 1n the fluxd c1reu1cs,_ The characterlstlcs ef temperature sene1ng dev1ces in strong radlatlon flelds must: be determlned Pressure responszve dev1ces wzll 1nclude hydraulzc and gas actuatede'“ *3in$truments null type bellows and diaphragms, manometers, and electric ‘_31mu1ators such as the mleroformerc- Problems w111 1nclude measurement of hlgh pressnres (up to several atmospheres) at hlgh temperateres in the presence of 'tllquld metalsQ 11qu1d fuels and gases and the measurement of 10Wpressures_ :eand low pressure dlfferentlals9 which may occur at any temperatfireg' It 1s: ?antlcipeted that pressure and volumetrle control dev1ces can be 1nterre1atedfi' to some extent | Wlth regard to velumetrlc centro} dev1ces 3I:w111 be necessary to develop 1 _methodsexfdeteetlng 11qu1d levels with fair aceuraey (in some cases preczsely) ~for beth the liquid fuel and liquid coolant and also the means of malntalnlng and eontrolllng these. Determination of the instantaneous quantlty of fuel Lpresent in a reactor core involves the consideration of weighing methods, ‘volumetric changes resultlng'from temperature variations, the feasibility of ‘using mean measurements from gangs of small tubes as an indication of the ‘quantity of f1uid'present in any one of the tubes, the effects of vapor pressures of materials being pressurized, the application of integeating flow ~measurement devices, the detection of clogged tubes and the relatlonshlp of ;thls effect to quantity determinations, the amount of excess fuel over and Tabove that in the core required for satlsfactofy control, and the possible' _appllcatlon of overflow1ng fuel as a dev1ce for quantlty 11m1tat1on and pro- 'v1d1ng some contlnuous flow of fuel B4 ae L1 L REE S ] . ¢E BB 0 WBE & 086 o * 89 e e e o8 @ . X T FRCRA IR IR 9 IR0 T DI o e el e s e es 8 L P - . [ % X J l . 0.8 '_.- E 3 N L N TE 2 e e L L I e .¢ g _..'_ L 6. &L @ 380 9 a 40 - 68 L e so0 st In connectlon w1th reactor control there lS a p0581b1e appllcatlen ofe 'dynamlc -pressure- respon31ve equ1pment to “the detectlon of hlgh and low~“ frequency pressure var1at1onss'and there w1ll agaln be the problem of locatlng; "the pressure sensxng element and brlnglng out leads to remote 1nd1cat1ng and 'controlllng equlpment : : SRy e IR S i Mon1tor1ng dev1ces will be requlred for detectlng the entrance of sodlum | into the fuel, or vice versa, and partlcularly for detectlng leaks in both the fuel lines and the coolant lines. It is ant1c1pated that spectrographlc' methods w1ll be of value in determlnlng where leaks are occurrlng - Study of the general problem of power monltorlng radlatlen detedtlen is not con31dered. f::as a £unct10n of the Experlmental Englneerlng Group _ Flow measurement can be effected by ‘the measurement of pressure drops g p0551bly by the appllcatlon of electromagnetlc devices or rotameters. The '-:subsequent control of the fluid flow can be accompllshed by orifices, venturis, threbtllng dev1ces, and/or valves of more or less conventlonal types, ‘55 ar ane e @ 6 ed #%. 6 08e. 8 gus e : S R U T [ | 6.8 e . . el w ee e @ 8 e g e e & 8a & @ ele e e - ste % e e W [ A & L N e e & LA} L] e s & L R ] R ea sae -8 @ a 9. 49 . OS¢ -0 # H08. am 7_j3@: BEACTOR PHYSICS N M Sm:tth Chalrman: - ANP Physzcs Group, Phy31cs D1v131on-- | 5; :Introduct1ofi ) | | .'“ .8;}TBef1ected Beactor Crltlcallty Calculatlons . Sl 6¢ 'Ref1ected Beactor Solld Fuel Calcuiatlons "q'Q;f Stat1cs of ANP Equlvalent Bare Beactor ~ E. Statics of ANP Reflected Liquid- Fuel Reactor f f.j>K1net1cs of ANP Reflected quuld Fuel Beactor G. Statics of ARE Bare Beactor R ” ” i'H; 'K1net1cs cf ARE Reactor N .'f}:}Crltlcal Assembly Calculatlons | J,” G1ossary of Nuclear Energy Terms A mmfinncr'xor{ In the flrst part of thls quarter the IBM Ferml dlffuslon multlgroup 7ca1cu1atlonal procedures have come into frultlon through the cooperatzve .”':effort of the ANP Phy51cs Group and the Uranlnm Control and Computing Section of Y- 12 The productlve capacity of this operation ~- now some 13 reactors ‘a'week — has materlally advanced the capabliltles ‘of the group not only in the survey of reactor characteristics bug also in the keeping up with the englneerlng variations an& sh1fts of 1nterest that haVe develaped The appl1catlon of the IBM caloulatxons has not followed a smooth course, _howeverJ since. about three weeks of Calculatlonal time was lost in loc atgng and rectifying a mathematlcai source of divergence in the flux solutions. This dlfflculty is descrlbed in Sec. 3B, which also incluides a description of ' the cperatlonal problem of these calculations and of the act1v1ty of the IBM group’ : - e : . . : . Slnce the 0r1g1na1 1nterest of the ANP Project was 1n11qu1&nwtalwwcoeled. Sfilld fuel reactor desxgns? the flrst calculatlons produced concerned this N ANe e e 2. 80 e & 866 0. sde 8 g e e W e B e LI0F XN Sy o @ & ® [ ) Cevel ve . e e 0 el el e es 0 e &8 e aie g S m Teee . & e ee e e ®.@ e 89 . .8 B e 84 : e e S e. o a sae s see 3. e ae € 8 e [ 3 8 elass of'reaotOrsx' The "statlcs "‘1@35, the crltlcal mass, flux dlstrlbutlon '-etcgpof+hese models apfflly almost equally wellto llqu1d fuel de51gns prov1ded' :'lthey have: the same medlan energy of fission (mef) ~ The solid-fuel studles are lflreported below in some detall followed by a descrlptlon of the ANP liquid- .'lfuel 200 megawatt d351gn of Jan. 9, 1951. It so happens that the mef of the_. | latter is somewhat less than that of the SOlld fuel de51gns because of dlfferw ence 1n moderator percentagesa ' The solld fuel reflacted reactor calculat1ons are of 1nterest since thev “show the effect on cr1tlcallty,'cr1t1cal mass, etc@ of the follow1ng 1. Change of moderator from berleIufi ozlée to ber3 iium whlah has. . been shown to glve a substantlal increase in reactz 1Ly, o2;o_Change of densxty of coolant &mé moderatora-- “3; lChange in the reflector thzckness._ 74;2'Heplacement of beryll1um ox1de reflector w1th a nonmoderaulng -+ . . stainless steel reflector (Whlchluu;been shown to decrease criti- Lo 'callty) :«5.'1Xenon poisoning. '6;;:Heterogenelty of uranlum lump1ng@ T@_ Iteratlon of the source term. . Follow1ng a routlne 1nvest1gat10n of the ANP 11qu1d fuel desxgn of January 9. by means of study of the equmvalent bare reactor, IBM calculations of reflected reactors were undertaken, A serxes of caloulatlons was made 'whxch permlts the estimation of the klnetlc temperature coeff1c1enu of re- act1v1tya; _ The kznetlos of the llquzd fuel reactor has been the subject of a large amount of 1nvest1gat10n which has revealed the desirability ofsashort thermal ‘relaxat1on time of the fuel tubes and of a long neutron lifetime. The worst types of accidents to which the nuclear-powered aircraft may be subjected are ~thought to be (1) the simultaneous failure of two turbojet motors and (2)'the failure of one-third of the primary coolant pumping power. Neither one of these failures is sufflcxent to make the reactor go into prompt-critical condltxon. Considerations of the time required for failure and the presence of delayed neutrons lead to the conclusion that self- regulatlon of the reactor is such that it is well. dampoc and safo durang such emergencaesg Calculations 87 RYYIIYE e eaee Lyt cees ° esswen - - s0p0ae0 ;on the behavxor of the reactor in the absence of damplng effect of ‘the’ delayed. ;'neutrons lndlcate that the reactor would safely regulate changes whlch mlgnt 'even throw it sllghtly beyand the prompt crltlcal condltlone WIth the change of emphaSIStjfthe eng1neer1ng studles toward the Alrcrafn Reactor Experlment bare reactor studies of the proposed ARE d651gns were - ' begunsf It was de51red that the ARE possess as many of the kinetic charactero _1stlcs of the hlgh powered 200- megawatt ANP reactor as pOSSIhleg'Whlch Te- 7snlted in a problem which could be solved only through arbitrary dec151on,_ A design of the ARE- posse581ng exac»ly the same characterIS&lcs of the AhP ”would have 1nsuff101ent fuel volume for crlticallty, It has been necessary to 1ncrease the size of the fuel tubesy_thereby 1ncreas1ng the thermal re- laXhtIGfl tlme hut an increase 1n the number of fuel tubes makes the ARE more, near1y thermal and the 1ncrease of neutron llfetlme tends te 6ffset the departure from 1dea1 equlvalents,- Beflected ARE calculat1ons have been begun but reporfable resul s'a:e not ava11abfie,' The klnetlc responses of the propcsed ARE designs have been studied and "compared with the kinetic response of the ANP design. A great number of acceptable de51gns have been found to exist, Calculatxons of the critical mass expected from the current beryillum~ uranium-aluminunm crltlcal assemblies have been undertaken. It is found that the calculatlons have underestlmated the critical mass by about 50%. In view of the experience of other groups in the flrs .estlmatlcn af critical masses from fundamental data, this result is not too surprlslng. Investigation of the sources of error are proceedzng9 and a number of small discrepancies between the exPerlmental setup aad the calcu;ations have been discovered, 'althbugh”nofie sfifficient'tc account for the large divergence. Obviously some time'an&'ffirther-experimentation will'béfnéceSSarybefcretheoretical and experimental values will come into agreement, It should be the goal to calculate critical masses by the moderated Fermi-diffusion method to within 25%. It may not be practical at this time to attempu to improve calculations beyond this point. & It is interesting that our ‘present experience has made but little use of the perturbation theory. The reason for this is that it is generally easier to . 58 . e seseaes T sneone csone [ ) R XN J . seas. YY) -'oo_-ool_. Yy "1_recalculate the characterlstlcs of a new reactor than 1t is to makeziperturba=7 | ;tlon calculatlon. The perturbatlon theory may be requzred whenever Iarge_" o spatlal variat1ons are enceuntered s B,*~REFLECTEB+REACraa_BRXTchLIfY cALcULAT1fiNs- | mathematlcs of calculations (D K Holmes, Physxcs D1v1slon and() Ay "'Schulze, Reactor School) In the calculatlon of the sfiablc cha‘acterxsvxcs of ' ,:two zeactors thh s&axnless steel reflectors and‘mmnw1th very thxfl BeO reflec- 'tors,'the siow1ng down den31ty had alternately greater&nuismaiiex vaiues (at a -gx?en space p01nt) from lethargy group to lethargy group (Fxga 3, l)andfxnaxiy took on negatlve values.' Thls dlfflculty has been dlscussed at length in a f' separate report (1) where it 1s shown that the partlcular assumptlon made as - to the var1atzon of the flux across a lethargy group, uamely,"g. ..fifig:é" | average fiux in the Nth lethary grcup :% n;f slowxng~down den31ty out of the N*h group E ' 1 : : _average of [-m—] in the N*h group o : (o : S e . e ; fi?lb1. e SRR AR N EREER R T = “should Eefreplaced by the assamption' 1N : '.::.fi;i_ :Ma 1 qx P g1 o' _ B fo-? N . _ .(2). : . : R I JN A . V ..::7- :1 :7 :_ . . . S fl.. S 3.j ' '.n.. . .where | __.| = the value of at the higher lethargy limit of the N%® ‘(1} Hoimes, B. K., 4 Correctwn to the gultzgroup Hethod of Y-Fi@*i?i Y«-FlO -38 {Feh 19, 1951% (OHYL, Y-12 Slte) T ' - . . B4 . ssasee asosse . ] ° ‘sesmee RS i e asd aseewe ‘eweese d st bt Ry s 8 | B0 10 1 LEEA T it R e Eg ] ol i R . B 4 Pii W o o TR 5 O ==ty ar -t i W W L) o M i e P WAt ILY o W W b B I Wb ot LA-Su it R ) T SN Fi dn i e e ’ i ) : e i %.'-HM -9 08 UTH 15 O 8 BER N N - ’ém%h. Al 4+ st -t 3 i e b 0.9} 0.5 st 0 o o ~<—— ALISN3A NMOQ ONIMOTS 0.5 B 28 12 I6 20 24 2 345 ) _ ) LATTICE POINT NUMBER,n (WHERE v nAr, WITH Ar =2.05cm.) SLOWING DOWN DENSITY VS. RADIUS (807) FIGURE 3. ! S seae . . L) T eesecee ® e "o . L ectense soeoee LI e sessoe saseos B [ X} L Nt ] . :gréup._ The use of numerlcal equatlons based on the second assumptlon e11m1é ' *J' nated the dsz1cu1t1es (see Flg._S 2) and thxs method has now been adopteé _' .:fcr reactor calculat1ons._ The details of the method referrgd_to as.the ;jj7'£ ”fiPLAm method are glven 1n a separate report (2) ' In order to determlne whethertfluareact1v1ty calculatlonstxfthese reactors"”": on the ba31s of assumptlon (1), whlch showed no osc111at10ns from group to - "greup, were actually acceptable w1thout recalculatlon zicomparlsan calculatlafi: was. ma&e. The partzcular reactor chosen for test. was a solld fuel type w1thu _ 'the f0110w1ng propertles b Chlfe f‘“fi;us i -‘ ,fii.." T 41 cfil. Beflector thlckness RS 14 cm Core eenta1n1ng='7_ }"'_; f :120)15 of Uz;; u._huerB:ALi:_- _.1. -_: 1‘Knfl=.% IN’CxE§3 . | ...i fiWBL; % ]}q REFLECfitfi‘ ; U0§_ f_ & ;_ 7'_'[f lses in”T ___i : .A:_ 6i:i B0 s200 7500 Sfiaanlasa steel _ _' '} | - 7?90 _i._'__w __‘.' I ~ '5;°51 ' Sodium ey LT o 3610 LIRS 20,00 __.Vold E e | 2._.;.037: - .:. | .0- S ~ The comparison between the twoé methods follows: " JETHOD OF ASSUMPTION METHOD OF ASSUMPTION - CHAPACTERISTIC - . - (D . - @y ' flff .'_i ST ; _ ' . ;-05118. . ',.. R | 2;96293. Median ensrgy for figsion (ev) 70.3 T £ 13 " Percent thermal fissions _' 5.8 AR - - '5.5 ' Estimated critical méss (1b) 98 '_7 f; ;' _ ' v 92 .Thé results are sfiffidightly similar that ifi'iS'pefmiSSiblé to écéept re- sults given b& the method (1) (which were calculated before the introduction of the EPLA method and whlch dzd not show calculational treuble), however, () Holmes, D. K , ?%e Hultzgroup Methods as Used by the ANP Physzcs Group, ANP-58 (Feb 15, 1951) £ _61' * vase L LERS : 0 bt cese [ % ‘eBeER® Py B . pqooli]'-.‘ eseew i b - o & . ¢ S ecenee s : Teeeewe T Ceesewe . L) Cweseedi L] e ® o L e X-10 DWG. NO. 10671 * i N = = =5 bt o e ine Y e AN i 4 T L P SEans w_@ : P b T 4 Al s, ) 09 1 B é 20 . I H _ b 1 4 i et + 1. ma b 1 ! e t..f* : o ; der i1 i i ot W. 1 A b H i i ot T PR H 1 . i ! 1 : Hh ; i ; ! ; ] - H 4 H . - ok e i + 4 + : ¢ Lo 1 % § b I ‘W+ 3 1 I 4 d e H " m M i i & nw, g4 . gt I : e 4 4 i 1 bt d b ik il by i N i 1 mae = s g > %hiam% . LR 58 G A N T g - 0.8 0.5 05 LI o o e W oo e e o e ,..__f._.._l?_..m_,zum NMOG ONIMOTS -°.‘5._" 0.5 28 24 16 S 2 3 4 Ay, WITH Ar =2.05 cm.) [N Led o & T Z ond o« m. o z Ti = o a ] Q TI Tl Mg % B =AW T . lo4BE O 4 8 2 16 20 24 26 32 36 40 44 48 52 56 REFLECTOR THICKNESS IN cm. e FIGURE 3!2 EF’FECT OF REFLEGTOR TH!CKNESS ON REAGT!V!TY » o Cewemes The effect of the substltutlon of the nonmoderatlng reflector on thef': core power dlstrlbutlon wasg strlklnge Flgure 3. 13 shows thls new spat1al7 ' 'power dlstnbutlone The shape of the curve 1s now very close to that of the f;fundamental mode of the equ1Valent bare reactor,-',e,, a zero order Bessel”I 'functlon of. the flrst klnd. The flux dlstrlbutxon curves change very llttle 'from ‘this shape (Flge 3 14) ' The rise in’ power denslty near the 1nterface_ that is characterlstlc of the mcderatlng reflector assembly is completeiy3 “absent,_The peak to- average power ratzo 1s 1 7? 1nd1cat1ng;avery undes1rabie9 :-nonuniform dlstrlbutlon of heat sourcee Flgure 3 15 shows the energy dlstrlbutlon of f1331cn praducxng neutrans at Various 1att1ce po1ntsa “This dzstrlbutlon is nearly: unlform over the. core, f :the fractlon thermal fxssion productlan at. the 1nterface belng no dlfferent frem that at the core center&- The effecb on the averall energy distribution ef nentrons is thus qulte marked the median energy of f15510n produczng '_neutrons now: bexng 270 ev 1nstead of 45 ev as 1n the BeO reflected assembly=_ From the standp01nt of bath low crltlcal mass and flattest p0351b1e core power dzstr1but10ng it seems very d831rable to use a mederatlng reflector' o rather than a stalnless steel reflector, e.ga, the pressure shell, directly exterzor to the CQI&! Effacttfiernon j The effect on crltlcal mass and reactlvxty of the- ' ;fzssxan product polson Xxenon was 1nvest1gatede Flrst the equlllbrlum xenan ‘denslty was determzned at each space lattlce pclnt over the ~core, based on theil f1531on den51ty and flux energy spectrum at the point. Thls equilibrium ‘den31ty varied from 6 x 10® atoms of xenon per cubic centimeter a% the center to 1 x 1916:at the 1nterfaceg ance ‘the crltlcailuy mebhod 1s set up only 'to handle core comp051t10ns which are finlform over the core, it was necessary to. flnd an average xenon den31ty whlch would give ‘the proper change in k jf This average wWas fcund by welghtlng the densities at each lattice p01nt with the fract10n|:fneutron captures in xenon that would take place:&athe spherlcal shell at this radxus.;_‘ It was foun& that the equlllbrzum xenon caused a Ak/k of -4.2% or, R expressed in terms of cr1t1ca1 mass, 1it 1ncreased the crltlcal mass from the 98 1b‘ of the clean case to- 117 1b. This is a greater effect than might be . expected since the Spectrum is def1n1tely nonthermal. It is planned to 1nvest1gate thls effect cf f1331on product p01sons furthera ‘At the date of "8() - : L g F , E ik Ceeli eew el e . ee 80 0 0R9 8.8 D% " . & @ . 58 LI 3 L] o 6 [ A Cee ¢ o e . . s -8 & @ e . w 98 s e . 9 e o * e [ 2 ® 0 e 9. ® LI a8 & e w0 @8 LU 1 B R N BT N xS oae . Bed G g 8w e 9 ® e 8es oo . S » e e e Taeesed ; sty Y XLILES . o e e esoemae epewee ] o 80 L » [ TXE YY) Caeew sl i CORE . REFLECTOR fil REACTOR 855 FE i HINEE i R | CoRE plAM. =269t [ ; R R R T e REFLECTOR . THICK. =55“ i [ TN ' ' FiTHET coMPOSITION BY VOLUME[ || 2617 %U0, S L Wk g g e s ) il F N T T T T 5.2.000 % BeO wmt"i‘ ! 7.90 0/0 SS 8070 5»5 36.10. % Na 20% 1.383 % VOID. | [RERME SN i i sl g 2oy i R v- 1. TR gt ....... PR N ‘_l‘g | ] rpenfmig o R "SI IR I . ? 0 -.,,';_ ;Ek@ff 1. 02 Fans Seieseed “;""j"i“-«éPEAK-POWER asnsn'v IS AV R ok i} e + + e fu “LHEHAVER. POWER DENSITY L] b H } 3 H 'x TERLTG " : w1 i : e [ + ! H ‘[ i 1 i R i 1 T Y 0 LRI U HE G500 00 AR AR R U0 I B ; 1 Er o 1 e SR 0 RESS ST BN SRS 1 \ 3 ' LR L S s e TS I i ; 1 1 i i 3 i ':l b - -+t Vi ! : ey i e by Fa i - -4 : ] NEaS it R O N SV S 1 S IS N T ! : ! - 1 T LA 1 v DI i 1 LSRN ] B3l H SE58EEINE Mans Eaaka A Dann R o SN aSuE A - o Y SIS I R RS Y L M i) I, ! o 1 N B 1T o 1 T H * ' T 13 J 3 H i . 1] I 1§ P 4 4t RS NS bbb - g ; RERS Ruaa A maN it 4 1 }_," A : T 1 i £ A I LIes ; - bodgr ? ’ : P bl LN 5.0 G PERE S 1T [ N 5 NUMBER™ OF NEUTRONS PRODUCED PER CC. OF CORE J S T BT il el Ry A SRR - - i pays i e i 2 4 6 8 10O 2 14 6 18 20 22 24 26 28 O AR u.__w - __.__4 g l , ' LATTICE PT NO. (I LATTICE iNTERVAL 205 cm) o 1 ! ! ‘ . g 2 4 6 _a< .m | ta 14 l6..' 18 zc .22 "RADIAL DISTANCE Unches) | FIGURE 3:3 SPATIAL POWER DISTR!BUTION (REACTOR 855) e o o : . s gERaTTE _ E1stE: 13 T e . s HTTTT (SRS SEL aEORS LURES 1 SERTRER] T 3 it 11 el it 133 i m.: Arpipaiidiaiadiaiedy e : M .M g & T m _WJ "M m W ” m i wm g m ! .n Mm s m .M» M .»t : m% 3891 M,., H i} + P T W W 3 13 + ES 6841 W”w. . .. J2 B id s Bl B84 b i 3 o Jpe ) % + %3 ] 11t S u : i L gt + Lrites gL S s o L E R (B Hipe L _ : gt w .. : T n_._ nrw i 3 N H “ o i 3 wmm T Wi ¥ e N ,Wm ey . O |l it 14 agtlbsse i iet: » S : ; caRE e i o] BRI = it AASEERSEE 11 B i _M. H 13 TRERE 3 m& nvv L m.wm i La Il . L, : . : i HH R SR, s SSERREH] - T T 15 w = fi i i A Lt : i o ] flwmw“, E il {hl m&mm 4 ..M.f. »% ] Pt ] i wf & i 1 HEHS ; i ! e i B T e N 8 " @ ‘T E pred P ] t Fadl deded {4 h {4 tebstel .. %m * t P i3 mi bt i o ok - Eian p i H R By I e+ b o ,.Ma y o % LB S : : TS i “ S i o “U pobf 4 o s i H ¥ w{m o fl T W gy ; f ! “., 1ilyeest o 4 o ! 5 o b A v w it ! i 1“ e o ;, Rl 3 817 R T L o 1 | it . Ww . * S %m A v.% T S gEas h | ”WH., 1 .m nu > i ; 1 t 1 Ll . i G . o ! - e " T T T : o < . ] et REAgAE & B . i ISP RaRNE Sy K nm — it LT, T T k! + L w m» 113 -4 R ‘ “ gl | ; i ; ~— T f i i | I : : | T , T ger: RO S e 1134 i ! : t h i SyEeraia) ] A ¢ 1igl SN 1 . oy L i 8 o & N vove " A1 i : M 3 ! i gfi I g ?a 5y ! il o ‘ o e h + i 1 H ..&u m M TR oy w S iz O snssss shiet 5 oy T S e i ; ; : £ % “:#x y ! Hh : : o = ocn e " ol ' [REES T4 a3 $rd 1 T b ,1.. x i it % 8 . -l [ e P e BN el BHid HELHER gt Hiditih = - eeeess - N TH : i w ; st 11T ; ww,m ; m mw.“ 31t wm. ik L -5 m = U o | sesdee ALl 33 2 i H: i B : . = Hidgls HEiE diise . g i 4 T%& w 11 1 : : L e 2 4 w 111 gaue IR, { - it i 3104 Wfi ¥ : o 2 cseeds duds SRSEtises e T ] ; : il ieg TELL : : g A ) o i T ey el iy Hiths U o o X 8 paF S H BE = - bk 3 i b . B - h il = 2 ] et : it : O ot T ¢ H t 1y h . 0 . ‘. " e 13 ¥ : b 1'% | S LA . | T leiiae s e | i RE- ". H 4434 e w + L 1 1 - m i i : O . ] | ! t - : - + + e o sestdre T 1 ; By SRR . 13 ._» 1 1 4 w " 1 H { i w 1 e h o VA . . H i ° ® i ; M U meae | ! | it * Skt —d ded 1} { ! Vit - i ] m B YYIEY] ! m .m i L “ ) M : Fie : 14 i esew .‘ § 3 i e 3 . [TV iee e & AP £ + ! wm I muw fii 3 MRIR § poy eeesee i sl it : TR i A s N b i Afim : l \ 1 ! 3 ¢ i T SRR b el 141 f BE H : %s ' T s} fl“ G Sy ¢ A * H | oK 5 b, . . > Hfiw Mfll fluw Jv M; k& = ; Y 4 ; ; 11} W X # ! ”... J e”) Am . : ] Hi) TE 3 . I - i L5 i iy i ol A e R AL - - MO nrv b - 53 : LA A d., flwv x\_ m i : Wm - LA = i 2 o e LAl g b 1] ookt 141 s ] ] 1] ! h 41 fbe 2 i et * o ! .iw. x i Xs / Yoi ¢ o % 7 : : Q. E ) . v«.. g : m y, " m ge w E R L = i M ES 113 ) w . ; H . e i E o A Rt bt ! i L F ) SEL 18T T . 1 i : ~ O i : o \\.\< 50 X-10 "OWG NO. 106 - 7 4 ; = £ TS T T e . e aakat -3 2 2 z; 38 -4 3 : -~ fl‘w MM..W M SMG @ .~n T.%iwfl . 3 . m e W . oy ity el ® Sz 8 o ¢35 ) a % @ S &l = i “. 7 aeg T : < oS53 B ZWN : .m o ..8......1.. AE3e] gt 8 H9u0d . _ 35t z 2k SadZF . wg B o MT%% %..%. %.%” oo S s S8 .hyfiv‘wma S W2 B fln.,. oo ; 1 W ® - s : : - O ety e e Q. 0w ® O e Az - : W Qw. O . ~aNda G ug . Eac HeE o 0. m x Ed e : &' ! -,” 16 JObE ‘esse0e ss0® TITLLE esvees’ ssores o 10 i 10° ENERGY {e.v.} FIGURE 3.5 NORMALIZED FISSIONING SPEGTRUM LETHARGY (REACTOR 839) 1.0 o '7th1s wrltlng only the one calculat1on hasbeen made w1thxenon in- reflected ffreactors. Equlvalent bare reactor: studles of assemblies w1th the same. spectrum: “lnd1cate a much smaller change in react1v1tyu Slnce the xenon effect 1s ; . ’.estlmated by employlng the flux distribution of the unp01soned reactor, the f .-fxrst xteratlon must result in an overestxmatlon of the xenon effecte Thef o faverage of two 1terat10ns 1s claser to a true value, and 1n thls case glvesfa :Ja value of about 3% for the xenon effect. A Flgure 3 15 shows the spat1al power d1str1but10n. Compar1son wmth' rFag 3. 3 on the clean reactor 1nd1cates a marked decrease in power denszty in thfilflfiffl near the 1nterface due to the xenon.- However, this is deflnltely' ' evorumybasxzed in th1s calculatlon cw1ng to the approx1mat10n of un1form xenon' dens;ty. L F1gure 3 17 shsws the 1ntegrated energy dlstrlbutlon of the f1331on-' ..:produclng neutrons.. The percentage of thermal neutrons has been decreased 'from 6 8 to 1 9% hy the xenon. | . L . | ] The denszty of xenon at 1ts max1mum transzent after shutdewn is a some- what elu81ve quantlty to determ1ne because the time ‘after shutdown at which the maximum occurs depends on. the neutron spectrum and hence varies over the '_core. BY a process of xntegratlon It was found that the greatest Ak/k change from the case of ne xenon would be 4. 9% and the crltlcal mass w1th thls xenon" would be . 120 lba The maximum xenon effect 1s thus not much worse than the‘ equ111br1um xenon effect. - Effect af flranium Lumping,_ Uszng the method deveioped at KAPL(3) to determ1ne new effect1ve cross- -sections for uranlum ‘when it exzsts in lumped rather than homogeneous form, it was found that the critical mass increased by only 5 lb. The fuel was assumed to exxst in 0.,08-in. 1solated p1ns hav1ng, by valumes 39% BeO, 36% UG and 25% VOld The mutual self- shzeldang effect i" of bundieé pzns was not determlned but this additional effect should be less _than the Isolated pin effect, The self shielding effect (ratio of effective ~to actual abscrptlon cross- sectlons) for lG« and 20-mil uranium fells of 93% enrlchment is 111ustrated 1n Flgs. 3, 18 and 3. 19.’ Effect Of Iteration of the Sonrce Term. In the calculatlon of a reflected : reactor by the multigroup method as developed by kapL(s) and as used by the 'ANP Phy51cs Group a spatlal distribution of fission neutrons is assumed. The .'(3) Bartels, w. J. C., S:lfsAbsorption of Ebnoene?geticifieusrons, KAPL-336 (May 1, 1950). 84 Y YT Y LI o roene e S [T LR - ve [ (X R E X3} covean - YYYEY ) . sheene e 2 M MO It Pt S0 S K3 D 2% It W K P B e A L e L e I Tl i B o . & Py ) 1 | ‘ R e S CoRE DiAM. =265' A e R e L e L U REFLECTOR. THICK,=55" A At e e f e cowosmou BY VOLUME | | SR H r gsvel ~ CORE REFLEc'mR - 3R ' e ZEREIAEE uaRLRsns dRanen e 1963 % UOz | - : 11 - f ‘ : 1 +H : pRS: sntegase 5_2.0_00__"/0_ Beo o ?59/‘,830 [ NG T R 790 %8s, sws.s L L R f xs %;;F 1 NI - BhduRsunaan: H JHE 36'10 % Na .20'% NQ ) S e G e JLil 2037 wvom [ RSB LIPEONAA H PN gaete: | - apsit EQUILIBR!UM Xe AT 200 MW : HHHH ' TH- jRasS e . BRRSRERS keff = .01 . e T NN BERsEh o ;f]if;PEAK POWER DENSITY il | a.o ; ; .. _ psqss BR NN - : ] _ 2525 hmn g‘.é.s,;;; = R t TN SeRSERRRSREY EECEal .A\!ER POWER _DENS!TY_)_ oo s & 1 TN M 3 ; PR B i i ' ' ! bl A i i BEN AN RN T’}, T3 T il % N ENEEY AN > N NEN NS “{ilié!tl."{"["‘ Y IBRRRER| I o .h“l. _n. -ouo » e YT T Y ‘aoeoae el e e ;o e - + . = s T T L ] 2 E oy o] g 134 %“'g”“‘“flua " 4 ] -t i Al - % - eoeeey T A Cemwd NUMBER™OF NEUTRONS PRODUCED PER CC OF CORE | Gl L bbbl et el L Ll L AL L L ARREARERARATERA TTARA RudnaRAREuRunnnsans LRRdRRE iRa85 BRESIRRLSNESRASAS C 2 4 6 8 o 12 14 I8 18 20 22 24 26 28 2 32 LATTICE PT. NO. { | LATTICE INTERVAL=20Sem) Sy | e RADIAL DISTANCE (lnches) ] ” FIGURE 3.16 SF’AT!AL POWER msmmunom | (REACTOR 822) RIS B 0SSN REACTOR NO.822 CORE DIAM.: AT L] 1 COMPOSITION BY VOLUME 2.69 . - Bt iein ook LR IR RERER XN . THICK. INSEREERNENS STRAN] .00% Be0 _?tysa£) 3§[£5gE;? Na SERGRNAE SNARNERBER BE AN - ~ CORE 963 % U0, 1 ASREARERNREANEES NENNNAARBEESNES DERGEANNEE INEAN SRRARE 00 RE NN £ 3N BN 62.00% I SNEANEEREEE S ESEY YN le i LT LA L Kege=to1 S v. FISSION= B3 e. EQUILIBRIUM Xe AT 200 M.W. §IREaeEauTs cuupt pupus unacdeny pysTudagne Aty ey spvnnyunna sunnguunesy MEDIAN ENERGY FOR . " i b 1 : i Y BN MR L4 4 bry- A i - t ' s . nouwnwu3d | .159 ., LETH ARGY u=in, | 10° o 121 B B 02 e3seEn deeeen sasdae L Gpases 'FIGURE 3.17 NORMALIZED FISSIONING SPECTRUM (REACTOR 822) L8 i T q U 1.Op |"2345678910|a'521314i5 LETHARGY '9@ ; 6 IT 18 192 21 SELF SHIELDING FACTOR AS A FUNCT!ON OF LETHARGY (lO MIL. FOIL) FIGURE 3.18 LssoL 88 + ——— Y G {-- -—" T .0'". | 0<123456?8936II (20 MIL. FOIL) F!GURE 319 I2. 1‘3“14 'ls LETHARGY s u ln '(2/ SELF SHIELDING FACTOR AS A FUNCTION OF LETHARGY I6 l?’ la'.?l;e_i-_ 20 21 £ 89901 ON TG ‘rate of convergence of zhe”diszribution=to-a'1imiting'vaiue'is'estabiishéd“by a process of iteréfiioo‘ Thls process 1s 111ugtrated graph1cally 1n Flge 3 20.. The steps were as follows (a) A flau source was. assumed for the first calcuiatlona The elgena _vaiue obtalnod corresponded to a k, f of 1. 043353 ' *f”(b)o:For the second calculao ong'the dlst?lbfi»l@fl dezermlned from- '”*:*calcula*lon (a) was used as the source distribution and re- .Jsulted in a k eff of 1 05118g‘an increase over (a) of 0.00783. o(o}!fiFor the'*hlrd caicuiat 10:13 *he dlstrlbutlon de@ermlned from . calculation (b) *a';‘sfid as the source discribution and the Q;szf-was 105483, an iucrease over (b) of 0. 00365 and an increase 'ZOVer (a} of 0. 011“?_ e A L e ' For the abGVe desc lbrd reactor tha flax1mam change in k f‘ doe to the' cho1ce of a flat source dlsfirlbutlon iz greafer than 1. 15? but pfobahly less than 1 5%., This corresponfi; to a maxlmum change in’ cratxcai mass of more. than : 5 4 lb but less than ? 1 1b, b STATICS GF 4;\‘13 EQU’V%LENT BARE REAGTGR J, W: Websters NEPA The prel1m1nary oalcuiatlons made thls quarter on the ANP reactor deslgn 'used barewreacuor methods and a "reflector savings™ to account for the re- fleotore The 3-ft elliptical core was approx1maoed by a homogeneous sphere of the same volume (17.68 cu f:}, having a dlamecercny 232 ft. The reflector, U?wh:ch is abtualiy rather complicated in oomp051tlon from a computational -standpoznt was taken to be equivalent to 5 extra inches of core mixture -_around_the 3.232-f: spherical core. The ;olume fractions of the materials comprising the core that were used in the calculation were as follows: 7.83% UF;-NaFVliquidfifuel solution, 57.14% BeO, 11.38% stainless steel, and 23.65% Na coolant, This composition was modified slightly in the final ANP reactor design. The critical mass of :his assembly oas.calculamed to be 95 1b of uranium of 93% enrichment, About 15 1lb of this total is necessary to offset the Xenon poisoning., About 3 1b was shown to be necessary because of the lumping of the fuel. This calculated increase due fio lumping is “hought tc be some- what optimistic since not all the effects of heterogeneity ha¥%e been adequately . . A e e e e R SR R S S e e e g s : nEs S as 498 @ L e en. 00 8 a0 O 90808 Sleel e ‘e 88 0 2.6 .® o 0. 8 88 e o eeiie &’ PR TR BTN SR N T S R TR ] e e e e e e ee e e e el e . e e e s o8 .0 . [ N T e o i teR 8886 6 4 08 88 8 & 4. s0e 80 26 . W . -3 o bt b A g e o o R T : _ Beaceiffeectis - SOURGE IR BRREENRA Y REDEN N INNEENRRERRENEERERENE AK=.,00783 AK=.00365 24 o Keft = LOSIIZ} so0783 o S G 05483777 T 3 B L ! I 22 20 5 LATTICE PT. NO (1 LATTICE INTERVAL =205¢m) “REACTOR A B G { I8 [El!if!llllil]Illll]-'!!ll]i!lll" B0 020 0 MO0 0 L OIS ot et 6 4 W B Y asas FIGURE 3.20 ITERATION OF SOURCE TERM W e - : 2 pasRSssaak bEES Jiuussass sessanusss o6 sty H N I L ¢ ; . o - (o] [+ @ ) 60 5 ¥3d 03I0NAOHd SNO¥LININ J0,HIEWAN S B _ < %) - ~ 3¥02 40 Y YLl e ARG T Y YT TY LR _cqn6=6:5:1.5 'accounted for in the present self shleldlng method ' Judgzng from. the com- ' parlson of theory and experlment on the f1rst crltlcal assemblyF the crltlcal :-masses as predlcted by the theory are too large and hence at this tlme_; "the best that can be sald 1t that the cr1t1ca1 mass of the ANP de31gn w1ll' probably 11e in the range 65 to 100 1b of uranium, Slnce the desIgn fuel volume is 1. 39 cu ft, the 95- lb mass leads to a '1fue1 solutlon of 68 1b of uranium per cub1c foot hav1ng a den51ty of about 3 g/cc and 11 mole % UF '1n the UF, - NaF solutlon._ These bare reactor caiculatlons 1nd1cate that the medlan energy of the f1551on produ01ng neutrons will be about 15 ev and. that about 7% of the : f1351on pro&uc1ng neutrons are thermal Flgure 3.21 is a graph of the energy' dlstribut1on of fission- produc1ng neutrons. The mean lifetime of a neutron "was calculated to be 2 x ID 5 sec. 3= Table 3 3 below presents the results of a study of the react1v1ty effectse in the reactor that must be offset by Shlm control' TABLE 3.3 Shim Control Requirements EFFECT o bk () e _”‘.fDQpletiéfi.(Qoo.megeweuts for 100 hr or 1. 855 ib) - ' - o -0.5 S 'Equ111b$1um xenon (total core flux, 200 megawatts, 5 x 1015 o - o neutronq/cmz -sec) B 3 o S =27 Moderator (expanelon and change of thermal base with rise in _ : , S MOderate“ temperature}® : _ o o R 0.0 : Fuel expanslon (14@0 to 1815°F) | S o . -141 'Reasonable allewanee for temporary extra xenon due to power o _ L -reductions . _ . S 8.7 . *The moderator will be preheeted at starteup. '_It'ie.intended that shim control in the ANP. reactor will be obtained by fuel removal. It was found that if fuel was moved uniformly over the core, _ 20% of the fuel would have to be removed to accomplish the above 5% reactivity change (the conversion factor from Ak eff to uranium mass for this reactor was A e RS i . et dos e e ° ea ee s ses P ose 3 s e 0 e @ e s s @ e el el eLe e e . 0. e 0. - & . o0 . *-se s 08 88 e e e e . 8 eae e L ele el e el e e ¢ 9o * o @ [ 2 L 0. 9 L o e » : [} .8 ‘e @ L ] a9 i s oz o S R o O ueraig gt h e L 02 & EE R o o EEE B 29 o i, 9 Pme g ER RN Y N 12 F e g Y O o - : roatr St o B S ® : g N R = I v , L qwo 3 e« = == O wik@ & S I R o . WY 2 2 HHE M & g g W e P on . = X e 3t oo : : 5 Th Z o W o _ SETEeE e < > LB & oot et i i 14k voooee X 0 lINN ¥3d NOILD na oHd NOILD! ED FISSIONING SPEGTRUM FIGURE 3.2I NORMALIZ &n/:- /k l_1n the cyllndrzcal portlon of the reactor around the longltudznal ax1s,only ”';determlned to be . _1@% would have to be remQVeda- The . welghtlng factor of 2 for" axial remova11 -- ‘as compared to unxform removai was caiculated from Lhe bare reactor perturu-t_-f ':;batlon theorem thau.the effecb on react1v1ty cf removal of a fuel lamp is ._proportlonal to the square of the Flux at the p01nt of removal 'ThlS-Will be_ : "~checked later by more accurate methods," :" A - {'_g@f sraxzcs aF ANP REFLECTED LIQUID FUEL REAETDB D K Holmem3 Phys1cs lelSlon 'and 0. A‘Schulze§ Reactor Schonl The ba51c ANP reflected reactor has a core cf 49 2 cm radlus w1th a. 16 4 -cm reflector The com9031t1on is: o _gAIEBiALf;r B VOL. % IN- CGRE '1 - YOL. % IN REFLECTOR ' S&rfitfifirg: f.-?' __‘? (1ne©nel) | Lo ':'5465 (Stginlésa éteel) .The volume percent of fuei shown corresponds to 120 lb of U235 the static 'characterlst1cs of ‘reactors of the same COmp081t10n except for the fuel were ‘also CalculatedQ_ The fuel welghts in these reactors were 100 and 80 1lb of U335, The keff for the 120 1b fuel reactor is 1.0660 and from all three calbulatlcns the critical mass is es&mmated to be about 90 1b, Flgure 3.22 shows the normalized flss10n1ng spectrnm the median energy for fission for the 120-1b fuel reactor is 18.5 ev, with 8.4% of the fissions occurring at thermal ; _energy;: Flgure 3.23 shows the pewer dlsurxbuglan,_, - From tfie-resui s far these reactors the followzng kznetlc coefflclents have béefi7determined . 1. The change in kafj per degree Fahrenhfilt change 1n the thermal base | | Ak . v._ SN L-%%la47 * 10f6/°F_ ) e Ces eake e e . 8. 80 B sre ' aes 88 e e ® e 08 L R T AR I e W e e BT TEE Y SORAE S -8 : @ L IO B S IR X SOt R e e e el e eee [ s ee Cee e e * ee e s e e o e @ ig e e Nel ee ke @ w20 0 o o S a‘el .. 0wB e - ';4)};.H6&e%éf 1£ the fuel was removed fromthetubes 7 soweww T 'ob.ieo : B ) CCRT SO ."f. . b --oo BER 'FRAGTION FISSIONING PER UNIT LETHARGY 08 .06 04 .02 —fedd AN R 3 £1 AL k 2 Ll easanncarenanas Hibi f,'_- HEE R fl B REACTOfi “NO. 844 o CORE DIA. 3.24 REFLECTOR THICKNESS 0.54' - COMPOSITION. BY vo:_ums, PER CENT - CORE - REF‘LECTOR' UF,-NaF 078 0 i BeO 8309 7500 | INCONEL 763 0 STAINLESS STEEL 0 - 500 ‘Na coouxm 2983 2000 ERRLE nt S LR = Y PSR aL TR R hy JhanEhug sy g ¥ H’“’”‘“ 3 R g was & TH: : - [ H i bk, PSS T ) i ek bk ey e 8 7 6 5 4 3 2 | o0 LETHARGY (u) ~ FIGURE 3.22 NORMALIZED FISSIONING SPEGTRUM (ANP REFLEGTED LIQUID FUEL REAGTOR) | 26;3 N ANGNND: AMEERAREY 1E 0 i § ‘-Mflg I o w‘wh- i o .___umou__n__o 90 ¥3d oud SN esesee scsass « OMLAIN J0,MIAWAN o DS o 3. :28f?i” ~ SPACE NUMBER n (| space=205cm) 22 FIGURE 323 SPATIAL POWER DISTRIBUTION (ANP REFLECTED LIQUID FUEL REACTOR) | - 14 16 18 20 2 - 2. The change in keff per &egree Fahrenhelt due to a dens:ty change s % _1n : _ R _ , R o W R R .f_ca)f_Moderatbr;only: Jpeo _a,(b) Sodlum coolant only:f.u -1.15 x 107%/°F gm ; [%Skl’ ;:1 : _.:._ Z>71 ;?; :. fi2 ;(é)f:Ffi§15§§ly;; :J it L ey fi_% = «5.2T7 % 10°5/°F H_3a_ The:f£agtiofia1CBAhgénih k?ff'pét'fraéfiqnai:éhangg.inhdensity of @ Mederater only: '_Qp/fizn;é'f,-“' . ”- (5)}-Sodiufi.c6oIant 0317;"_“ “ "ak/é S ——| = 0.00572 Ap/p Ko | | .f (g) Fuel only : {Ak/fs] L giaee ' L Z}Ffl/fij F‘Q; 1- . o ';Comparé' thh Table 3 2 1in Sec; 3. C wh1ch g1ves comparable constants for the solzd fuel de51gn cf hlgher med1an energy of flSSlOH. T A wd Yef posmee LT : fepemer T ‘owmews D 'F. KINETICS OF.ANP“REFLECTEB LIQfiiB»FUEL REACTOR - T, Rubin;'NEPA.'; N. M..szth Physxcs D1V1s10n_ R R, R Coveyou Mathematlcs Panel , 1951 to step change in react1v1ty of 1073 and to an entrance coolant temperature The klnetxc responses of the ANP reactor as des1gned by Jan. 9 step change of 75° C ‘has been calculateé(4) for the case in which delayed _neutrons ‘and xenon have been neglected The variations of flux and fuel ':temperature about the orlglnal operatzng posztlon on appllcat1on of the above " fwo stlmuli are plotted 1nF1gs. 3.24 through 3 27.- The follow1ng‘constants . were used in the calculations 1hér§aI feléxatién'tifie (sec) DR ‘5 ' s o 0.40 " ‘3_ ;" 13.6 ’1muunmwuwufl/m' B 66 x10° 3.7 x 105 -:_Tbmperature risé rate in absence of ccolxng ( C/sec) o .678 : - o 7.8 | Beact1v1ty temperatnse coefficient (per °C) o -3.3 x 1073 .~3-6 x 1075 Integrated flux (neutrons/cm -sec) E TR _ 3 fi 1015 | ':Neutron lifetime (sec) 3 ” l . . 2.5 x 1075 '”_Furthér.caiculations'show that for small‘reactivity changes from critical, the delayed neutrons completely damp out the high-frequenéy‘oscillations ebtained in fihesefcélculations, This is'évident since the time scale of the 'procéés'isxndw determinéd'by the delayed-neutron periods. Figures 3.28 and f3-29 illustraté the effect cf the delayed”nentrons; ' | Work is now in progress on obtaxnlng better estimates for the react1v1ty' coefflc1ents from the IBM multigroup calculatlons. : 6. STATICS OF ARE BARE REACTOR ~J. W. Webster, NEPA The reactor phyéics calculations in regard to the statics of the ARE havé:thus far been largely of the bare-reactor preliminary type. Nine pro- posals were made by the Design Group and a critical mass study has been com- pleted of these nine assemblies. It has now become evident that critical mass (4) For derlvatlon of equat;ons see report .ANP- 62,, Perturbation Equations for the Kinetic Re%fonse ({ iquid Fuel Reactor, by N. M. Smith, T. Rubin, M. J. Nielsen, and R. R, Coveyou (Apr. 17, 1951)., (DPNL Y-12 Site). - 97 . | X-10 0WG No. 10857 e . ey Hepuin Cu o e AT Ao E DA, NS SR Scviimen SIS S Haae WA SRS SR i - H Av.tH!oJl.. : - . B R S S oaterd . T B Sibove g e - oo i b S SEERS Siuiie) Sevt SEEEY STt el SALS Sttt whaits 14 I I gy S S 5 f & . t { l«:_lm\l i 0 ®(NO DELAYS IN REACTIVITY OF 10° SN S FIGURE 3.24 FLUX RESPONSE TO A STEP CHANGE NO XENON) » t=s8C X-10 DWG. NO. 10658 '?\’ S 3 IAN. {0 DELIAYIS 4 t=sec EP CHANGE IN REACTIVITY OF 107" (NO DELAYS ~-NO XENON) FIGURE 3.25 BULK MEAN FUEL TEMPERATURE RESPONSE TO A ST 13 15 T Fomi— g : 1 s = s ——— [OOIS SHME Nivanil ke > [anawoy N - [ vt PR QI i hi Pt Pt SR S RO ok Sy - MM Sty somumiy s - RO Sy SN S TR Srndry fai s ; iom - . i ) pengsl e . . » Py apt DT Qupuiiiinl Ryl ety s suwas TN 64 SR AN VA vy T, & Capd A 0 i oy 5 e B Ty SRy Sy sy 3L D IR DU NP i T F¥ SRR S Sy Iy IS Suh AP S8 e ' | NTERMEDI GOOLIANT | [TEMRERAT AN o - Ak OF| P o' " 1 ENTR HGE h 5 DUCED] IN COOLANT ENTRANCE TEMPERATURE -NO XENON) Ry o Py INTR .vnfi. s 5o wr il - ECte Shesi SEpSa et Gt IR walas g B o e ST el (o u T : s el unia Sy TS I S yEa £ B SRS Ralonn SRS Ry i $7 ; H H [ f=aac FIGURE 3.26 FLUX RESPONSE TO A 75°C' STEP CHANGE ity L i (NO DELAYS i iy - oSSy ahas e sy ——— e _ & ~ ~ : e b bt poe i T i 4 ..... th R i ] 5 BSR4 2 pois - - + o e v : piad o prre— oy S s ; s i s B + * b R H G o S E Saaee S aws SR G Ses tett SR DOAS DS Siions oo wa i T ¥ T P b 00 KO0 B I DIV DU DO Sl DD SN SRS S S Pt 04 LY : Kes pla ITIIIIT FRUa Satad boey orewriret Sl H s T e DRSS " , : B e bl ~ AN g iy bt AR o o B e el i P B R PP SUrEng (AP Cuildngs S - rait sour I 77 FRRIEE i w5t o e e s s % P ERTOY Ay e i : 8 = o g e i ,w A 7 o S el SAes tit bsts Stace el edd Sl T s pe s oo RETS | e = =4 0 : e PO ) e = = reasire B R - e A ag gLt e o e 1 ,.12 rr{ S A : = R et B S ! v o e D Ry R . 3 (S e foua fas: TS Ty SR et EE B 4 & ay L: D P . . L RE - -y H; wt - < =i A w . o 3 it fekioo + . — 34. Tn . . = = ._ o 3 7 K E : : S =i MT - - ENTRANCE TEMPERATURE (NO DELAYS~-NO XENON) - S 0 e g - T I gt 00 o b 4 ; ; } A ik By oo A A 44 pater otk U it g L R e T B . 4 B it S SIS SR S e NN IO Qe S SIS 5 B + vy - — P S S v TLI S by SR G Y i . T . i : s . RSP it Rl ot T - bt inapeydl T iy e *Y s H P S r T 1 B S e aans THt Sy, A i o ek ggsi souTnusUassEn e * i i T e i i : : 5 FIGURE 3.27 BULK MEAN FUEL TEMPERATURE RESPONSE TO A 75°C STEP CHANGE L2 X R ) A S X X Y YYY L ems o6 - LB XXX ..”:: . H X-10 DW6. NO. 10 28 e | H] il R atifro f ¥R B HI0 ssdobeg | AVERAGE 18 20 22 24 2’ (~0:XENON) IN REACTIVITY OF 1073 18 t= sec 14 iz 48 8 [ e v 6 4 XY [ 2] s 4 & . - & 2 9 FIGURE 3.28 FLUX RESPONSE TO A STEP CHANGE - vl e L iy -l f L. IN REACTIVITY OF R} {|AVERAGE REAG i p : ¥ H LA {-ses i4 (NO XENON) TT L¥d ‘seaer beaded FIGURE 3.29 BULK MEAN FUEL TEMPERATURE RESPONSE TO A STEP CHANGE 1s an lmportant crltetzen in selectlng an. ARE de31gn._ If 1n order to 31muu7 late the control kinetics of the ANP de51gn (1) the number of fuel ‘tubes in the ANP reactor is scaled down by a factor equal to the ratio of the des1gn“: power of the ANP (200 megawatts) to that of the ARE (3 megawatts?; (2) the .7fuel tube dlameter 1is kept the same, (3) coolant volume in- the core 1is reducede' accordlngly, and (4) the moderator yolume is increased accordlngly, then the 'resultant de81gn contains insufficient fuel volumeforcnfitlcalltyevefithoughf _the moderatcr volume is over 90% of the core volume. The result .is that the:' - number of fuel tubes must be 1ncreased over this exact scale-&own, or the sive ' of the fuel tubes must be 1ncreased or both untll the reactor can be mad; _cr1t1cal w1th a satlsfactorv fuel solutzon even though the control character= "Ve'lstlcs of the ANP deszgn ma} not be 51muiated exactly,, Table 3.4 llsts the composxtlon and crltlcal mass - results on the nlnee .'?roposals._ A 5-in, reflector savzng was assumed for each reactor., The cr1t1ca1 masses .are. glven for a BeO den31ty of both 2. 8 and 2.6 g/cc._ Iv appears that expedzency in procurement may d1ctate acceptance of the lower values- _ Judglng from comparlson of the theory w1th experlment on the farst} _cr1t10a1 assembly, these results are pe551m15t1c,and futgre correlated me thods may predlct values as much as 33% less than those quoted. In any event the proposed reactors 1, 2, lA and 1B could not ‘be made critical with any reason- able fuel solutlon.\\Proposals 3 and 2A are marglnalu' Proposal 3B would not s1mulate the control characterlstlcs of ANP at all well, so 3A and 2B, or : somethlng 1ntermed1ete between these two,seem to be the 1og1ca1 des1gn. Table 3. 5 presents the results of calculatlons on react1v1ty effects in reactor SA that will hdve to be offset by shlm contral. / It is planned ‘to build into the ABE two methods of contrel. ‘One method | wxll ‘be the removal of fuel solution from a cyllndrlcal portion of the reactor arennd the longltudznal axis, and the second method will be the insertion of an'ebeorber control rod alohg the axis. To accomplish the 2.8% change in k (Table 3.5) would require removing approx1mately 7%% of the fuel if it is removed from the axial portion of the core, i.e., fuel tubes out to mbout 0.27 of the core radius will have to be devoted to shim control., These results are rough and will be checked in the future by more elaborate methods. . L *6 ans ) ) [N 8 00 *eud * 4N ‘0N s ‘o' 8 . o @ e ® e LR SRR Rl ¢ e . 0 00 - e - LB L - .8 o8 8 &8 L XAE e a e e L [ LR T LT L ) . 0.8 F N T ® ° [ e e K Y [ N - Hee 9 CY T TN S T X B *0 [ ] 3. Ao 9@ sonehe snodne Cel el PXT T S Fa ] . TABLE 3.4 ARE Critical Masses " REACTOR 3 1A 3A .1B No. of fuel tubés ' Fuel tobe L.D. (in.) Fuel volume {cu ft) VYolume fraction Fuel Moderator getai Coolant ~ Thermal fissinns (%} N Uranivm mass {1b) Ppeo = 2.8 : pBeG = 2.6 - Uranium per cubic foot of fuel solution (1b} _‘ pBeO = 2.6 0.00102 0.97794 1 0.00904 0 0120 Es11 CE*13 E*722 0.00357 0.95753 - 0.0104 - 0.0285 E*13 E*15 © E*238 225 - 0.200 0.111 0.0063 0.9365 0.0117 - 0.0455 77 14 16 144 563 - 0.080 0.045 0.00255 0.95475 0.0127 0.0300 E*12 E*14 “E*311 563 | 0.150 L 0.1575 0.00894 0.90297 0-01669 0.0714 1 20 127 o © 0.200 0.2775 0.01575 -_0;35039 - 0.02006 0.1138 | 95 1125 10080 0.090 - 0.0051 | 0.9157 - 0.0192 0.0600 7 .13u 200 : 15 95 1125 020 00314 0.7065 | 0.0343 02218 e 41 48 86 * E means estimated from the other calculated results. 3 'Th¢'absbeér'cbntfdl.rbd will'pfovi&é'afiXiliarg”sfiimg saféty,'éfia'thamid5': 'controla. The effectlveness of thls type of control in the ARE as a functlon- ' of rod size has not yet been calculated e ) TABLE ) 3 ‘_'.5 . AEE'Shim cdntrol Requirements CEFFECT . Ay () : Depletmn B R S P l- e "Equzlxbrxum xenon (thermal flux 1 x 1013 neutrons/cm -Bec for' e B ' '3 megawatts) ot _ _ » L <16 Moderator (expansxon an& change of thermal base w1th rise in moderatoz I temperature)’ . IR . o T ' "Fuel expansxon (1400** to 1815 F) B SRR B f_fi o . -1v1 E;Beasonable allowance for temporary ‘éxtra xenon due to power R reductxons | _ o R o : =0.1 ‘The moderator w111 be preheated at startmup . *‘1400 °F was selected somewhat arbitrarily as the temperature at which the fuel would be loaded in the core at start- up The normallzed flsszonlng spectrum of proposed reactor 3A thhout xenon and’ w1th equ;llbrzum Xenon is shown in Flgs. 3. 30 and 3. 31, respectively, The przmary dlfference between the two curves appears at higher lethargies, since tha area under the two curves is normalized to unity, - 'E. KINETIC& QF ABE &E&CTG%& M, C. Edlund Physlcs vaxszon | Pérturbatidn Theory Stufiiés. A study of the kinetic behavior of some of the proposed ARE core configurations(s) has been partially completed. The ‘responses of these reactors for short time intervals after a step change of (8 schroéder'.g, W., Proposed ARE Cove Cbnfigurations, Y-Fg-13 (Feb. §, 1951) .(OENL, Y-12 Site). 30, @ L4 L - e a8 e ade. 0 esE S8 ° e e - % B8 % & ¥ &, &g @@ o & do AT SRRy Sk TRy SR @08 H e 88 TE e e e el eee e LR SR SRR TN SR e ‘e e e g @ & 0 ea L DS o el e el s e see & Cope @ W el e e 4. e e oe s -3 LERESRRYR A2 : @ EBANA 4 a8 g RE S 2 H-J ¥ - (} 4 H L B BB S HE L ARE 3' ELLIPTICAL REACTOR e S i i e et i PROPOSED 3 A | T vm.uw: FRACTIONS-- L0158 FUEL. 020} INCONEL: 8504 BeO II38 ‘Na THE T i T MEDIAN ENERGY FOR FISSION i ::f‘§ oo HHE T : _ g i : -3 THERMAL $1i S ' jeii sos % THERMAL FISSIONING - NO XENON P i bonk - v 8 5 . ¥ B W e FRAY SRNNS v el £ B ey N Tesasee o g vave sb00de L] L desc ° [ ] e e YET XY ® T PR R Leeenel o B bt gt T A ey il A e B i Lty - - e e o FRACTION FISSIONING PERUNIT w u : M : ;o B i 15 14 13 12 010 9 8 7 6 5 4 3 2 I o LETHARGY u=inlC B S | ey P | ¥ | SR I 1 10 o 10 102 w® 0t 0% 108 | | ENERGY (ev.) | | | Th o- % @ DR R o FIGURE 330 NORMALIZED FBSSIONING SPEGTRUM 9 PR RS - ume BARE REAGTOR NO XENQN) seeaced . v ® Y TY I FRACTION FISSIONING PERUNIT u nnnnnn PROPOSED 3A . VOLUME FRACTIONS = i i i 8504 BeO .1138 No B i R s s 11 MEAN ENERGY FOR FISSION = THERMAL EQUILIBRIUM Xe i ARE 3' ELLIPTICAL REACTOR D iR s 0158 FUEL .020! INCONEL ¥ 690 % THERMAL FISSIONING' il WD 3 &yt e 7!0 e 7 : 5 5 e 3 2_,: LETHARGY wus=inliQ T | (R Y 10 0 0w w1 0 S ENERGY {e.v) A S - ’ ) | FIGURE 33| NORMALIZED FISSIONING SPECTRUM (ARE QARE REAGTOR EQU!LIBR!UM XENON) o e o 3 oy g _100 F in the 1nlet coelant temperature have been calcnlated uszng the perturq 'batlon technlque (5) el : The varlatlon of neutron flux in the ABE type of cores durlng short. thme 1nteIVals after a change in react1v1ty is essentlally governed by four ._-factors ' e | : 1. Fuel temperature coeff1c1ent of react1v1ty9 a negat1ve COefflfi -clenta _:2{"Fue1 tube expan51ons (a positive ca@ff1c1ent¢3freactIV1ty 11nke& . with temperatures of coolant) :f;iThermai relaxation tlme of f fli 4!T0tal heat Capac]_ay Qf fuel and tubingo The Doppler effect in fuel has not been con31dered ‘and the moderator co- efficient resultzng from dlrect heatzng in- the- mo&erator has bken neglected, : : 'sancethe thermal relaxatlon time of the moderatorsls'of the order of thousands of secondsa- The xenon -coefficient was not 1ncluded since it depends on the. ‘ temperature of the moderator. It 1s reasonable to suppose that a temperature- sensing servo-control SYSuem could react in tlmes much shorter than 10° sec. Hence all moderator coeff1c1ents are neglected in a f1rst survey of ARE klnetlcs,f' ' ' ' ' o Three of the proposed ‘ARE reactor des;gns were selected as representatlve' - of the range bexng con31deredg They varled from 71% thermal fissions to 40% ~therma1 flsslonsg the neutron 11fet1mes varying from an estxmated 1074 to 3 X 10“5 Sec. The thermal characteristics of these reactors also covered the _'range cén31dered For camparlsen the same calculatlon was made for a yroposed- _full power alrcraft reactor (ANP reactor) The calcula ions 1nclude an effectlve smngle grou? of delafed neutrons obtalned by lumping all delayednneutron precursors into a 81ngle group with ‘an average decay constan? of 0.081 sec and a fractional yield per fission neutron of 0,0075, ' , | A'summary of the constants employed in the calculations is given in Tab,l& 3@61 ‘ ' o The temperature coefficient of reactivity that is due to fuel expansion is estimated to be -6 x 10°%/°F. The temperature coefficient of reactivity due to the tube wall expansion is 2.34 x 10"%/°F. The temperature of the fuel {6) Edlufid, M. C,;.Kinetias of Proposed ARE Resciors, Y-F10-42 (%o be issued) (ORNL, Y-12 Site). B . - . , . SR T e DT T DT B e ; v eee -6 & ¢ 88 THe B ee8 0 380 08 ‘.. » e e W@ PR N e eie e el e e 8 ¢ 60 e @ e & & LR RO T R T 7 S T o8 e e eee e elel e L e e NIRRT RN A R . e [ SURRURL ST I e 8 & €8 809 6. SR 8.8 86 88 3 .6 0 882 98 ' TABLE 3.6 gonstants® Employéd'in'théfxinetic'Calcuiétion'_ BeO IN CORE . {"mi_. %) INCONEL IN CORE (vol. %) % THEBMAL FISSIONS _ APPROXIMATE LIFETIME o {sec) - 94 99 11 70 71 40 . 1 o107 w0ttt s x107 2 x187% | THERMAL RELAXATION ~ TIME OF FUEL (sec) TOTAL HEAT CAPACITY . oF FUEL TOTAL HEAT CAPACITY ~ OF TUBES (Bsu/°F) TOTAL POWER PRODUCTION = IN FUEL ' (Btu/F), T.49 - 10.62 37.44 80.3 3.30 . 6.40 16.50 45.1 (Btu/ses) 2,674 2,674 2,674 178,228 ;jEMGKm - MEAN FUEL TEMPERATURE, ' STEADY STAIE - (°F) MEAN COOLANT TEMPERATURE, ~ STEADY STATE R INLET COOLANT TEMPERATURE (°F) 1900 1570 1450 1900 1270 . 1319 1320 . 1310 - 1100 1140 1150 1140 . “*With the exception - 8, Y. Manson, 110 . L 2 aeeee R sesses : ._.....-" SR ) o e Siietl Ta ‘senswa i b S - B coew of the % Thermal Fissions and Neutron Lifetimes these values were obtained from m 'fltubes fellows the temperature of coclant more closely than that of the fuel'“ _and ‘can be consldered as a coolant temperature coeffx01ent. Each calculatlon' _was made wzth two neutron 11fet1mes, 10 4 and 2 5 X . 10 5 sec. In FlgSu 3 32 and 3. 33 are shown the results when flP equals the Lhange_ 2 1n total power in the fuel (1n Brltlsh thermal unlts per second) from steady.. fstate, AB equals the change in mean fuel temperature from steady statej ‘and ¢ “15 the tzme in seconds after the change in coolant temperature,;_Ihe perzodég 'may be seen from the equatxons in Tabie 3. 7 I-TABLE'sof_f” Power and Average Fuel ?elperature Equations for the ARE ',_. Resultlng from Pertnrbation Theary Anaiysas REAcnnzs . Opeo.238 - 7ese'73 Tt 902e‘2 75+ a3 9e'° L0875 C M-304s L 1aef?3 7t . 41, ce'3 ST e “575*4 REACfi%§2A ‘A?"%f;5§5; 535:‘7‘ e§ + 8733-1 79. 4 166e'° essat | ';'_56 ¥_26;4.§ 0. 5493‘74 85- s1. sé‘? SET - 29. sé’° JULE e _ | | S REACfiX&BB e | & Ap ;'7".837 . 366é”3°“* ‘ 830,«-9 473t 373@’9 05108 ' ’éélé 8. lsw-a ozsa’?fl4@ . 63 9e'9 4?3f - 55 5.=a 05165 _."’-.;'QP s ~16.?00 - 49, ebaé'flsgfi + 62, oooe'?f 8* + 3, saae'°°°543* e 200+ 1, 90e'389* - 5. ?e'17 8, g3, ae'? 06431 The tlme at which the maxamum mean fuel tempera?ure excurs:on occurs is '1nverssly propsrtlonal ta ‘the thermal relaxatxon time of the fuel. In all - cases the maximum- change in mean fuel temperature was about 45°F. The power excurs:ons; however9 1nd1cate a consxderably dlfferent power response for the S proposed ARE cores compared to the aircraft reactor. In general the frectlonal_ - ' ¢ L ee ey e e [y s e8¢ see 6. 460 86 L e e, @ 8.8 @ e el @l el e o ® ¢ & 08 @8 LI R 3 S s e ® &0 ».. RN TR SR N LR L) el el ele ER R B e 8 . 0 @ @ * .8 . e e e v 8 8.8 ‘oe ees. B * s e oo ¥ & ® . 83G a@ 0.07 SIS U WA A Farems 3 NO. 10665 0.06 A gy 2. o R i : (o 3 MRl i i ¥ X-10 DY 0.05 SN NN ) ] 0.04 S Tme e FIGURE 332 CHANGE IN POWER (BTU/SEC) FOR I00°F IN INLET COOLANT: 5 H ! e e + R + : i a3 BEE 0.03 e - 0.02 STEP CHANGE 4 0.0l g_ EmuRp F ) 2 4. 5 3 E 3 E o o i iR A N o 252 ©3 weEs/nieg 400 s00} 200 e 100E O 800 700 600 < 50 40 [ i b : j; . e _ 3 D ) . S = E 5 -4 o ) eev0a0 ........... 3 : ) 0000000 ..... Pomee . .D“...O_ ; B e H bk Fh R R X R R I PP T SHE ess0oe.; sseswe T ......... 0 DA e 2 4 6 8 10 12 I T S - TIME (SEC) FIGURE 333CHANGE IN MEAN FUEL TEMPERATURE (“F) FOR 500°F STEP CHANGE IN THE : ENLET COOLANT TEMPERATURE FRACTION FISSIONING PER UNIT u '-n~ s DlMENS!ONS 2! s 2! 323 22 e ’_ i E .. RREADSEARE NN Lo LA b by gm _-.soswe comz ® . R Aenn faestbare !0 Mil.. SELF SHEELDING iN&LUDED ,ME;D;IAN ENERGY FOR FlSSION-! le.v._'_: i 3.4 iy 254 14 S + T4t ‘, t g T RN N RS : I T ki - . ey 3 o PRI SEES i an t f-r'jr BRI 1T RSN N - ; s Te L Rny ! LETHARGY In '..,!%Z o FIGURE 3 34 NORMALIZED FISSIONING SPECTRUM (CR!T!CAL EXPER!MENT ASSEMBLY *fi) 69901 'ON OMQ OI-X | 5 power change]&;much greater than the fractlonal temperature change and dependsfi3" “strongly on the thermal coupllng of the fuel w1th the coolant system and on . the neutron llfetlme.' Whereas the change 1n temperature obta1ned in each case for the twoc.c dlfferent assumed neutron llfetlmes (10 4 and 2.5 x 10°°3 sec) was the same§ ' the power excurszon was found to be somewhat larger and to occur soomer for the shorter neutron 11fet1me.}= Haximum Préssure in'FueI'Tui}es° An upper 11m1t to the maximum prcssnres". Axn the Ilqu1d fuel system resultlng from the expan31on of the fuel has beenc_ 'cxbnazned by neglectlng the -expansion of the tube walls.(7) The fuel is treated ':as an 1ncompre551b1e fluzda The ve1001ty of the fluid along the tube can: ‘ then bc obtained by 1ntegrat1ng the equation of continuity, the time rate of -change of den51ty being determlned by t he fuel temperature behavior. The‘. ’-Vpressure 1s then obta1ned frcm Euler’s equatxons._ The max1mum pressure in the fuel. tube ‘occurs at the closed end of the tube and is given by where . pressure in fuel reserv01r. Comy [~ 2 i - l;%. ube length Py = steady-state fuel dens1ty @ = volume expansion coeff1c1ent of the fuei T = temperature of the fuelr The time derivatives of the temperature are to be evaluated at a time at which the above expression for P,,. is a maximum, (1) Edluwnd, M. C., Meximun Pressures in Fuel Tubes, Y-F10-43 (to be isaued).(ORVL, Y-12 Site). 115 sassas i ceesse Ce LR e e™D O It is xnterestlng to note that the pressure varles as, the square of the tube length rather than as’ the flrst power, ‘as would be the ‘case of a 11qu1d ‘of constant den31ty flow1ng in a t,ubes B The maximum pressures for the cases glven:u1Table 3 7 have been calculated “to be ahout 10 psi for the ANP and less than 1 p51 for the ARE cores. S S CEITICAL EXPERI&EKT CALCULATIGNS B. T. Macauleyi Reactor School Bare reactor calculatlons were mad& by the ANP PhySICs Group using multle' group methods -and material constants forwar&ed by the Critical Experlmenu 'Group, The program was ‘planned for the purpose of fiorreiaglng theoretical pre~ 'dlctlons of critical mass with those obtained. from experlment,u81ng multigroup 'calculatlons in use by the ANP Physlcs Group as well as furnlsh1ngtheCr1t1cal- Expar1ment Group with theoretlcal values of spatxalfluxdlsurxbutlon fission- 1ng spectrum, ‘and effects of heterogenelty on ‘eritical mass and reactivity. A prellmlnary calculation on. a crltlcal assembly hav1ng the follow1ng propert1es was performed Base area 24 by 24 in. '_Loadlng leagth o Varied from 22.22 to 24.24 in. "*.'Volume f:aczlona fhel (53.2% Uzss) ' 3 o 0.0071045 : o e Beryllium moderator _ - - §.9150503 ~ Stainless steel o 0.000697 O Aluminum e 0.064039 © Veid 3 -{~3_,~:' | 0;0131@8& | When self sh;eldlng effects for fozls of 10 mils thxckness were taken into account it was found that an assembly having the dimensions 24 by 24 by 24.24 in. would have a ksff of 1.00142, W1th a corresponding critical mass, m,, of 29.35 kg of uranium metal, and an mef of 1.2 ev. The normalized fis~ sioning spectrum resultlng from thls calculatlon is illustrated in Fig. 3.34. H@wever'the Critical Assembly Greup repdrted that an assembly 21 by 21 by 23.22 in, had gone critical with a corresponding critical mass of approximately 19 kg and an mef of the order of 0.5 to 1 ev, making a discrepancy of about 116 - . . % er . see v e 6. eeies s sas Wisss . eo e & @ * 8 e e - ¥ e e 4 . e 8- 0. ¢ oe 8 ° 4.8 @ e e 08 . 0. & & & [ R, e 4 . e8I €. 9 9 & ¢ & 0 0 e e 9 e e ® e @ LA [ e« e W Te® LA X 2 e Ses 5 .0 B8 *e e & £ d .00_. 0 :50% in crztlcal mass._ More exact materlal constants were obtalned and aH 'rerun on an assembly of 21 bY 21 by 23.22 1n,fwas performed to see what' _}theoretlcal k ff-would be predlcted The calculated k ff was 0 90346 and the." 1_mef was 1 1 ev.' Because of the large dlfference between theoretlcal predlctlons and . exper1mentaI values for critical mass,'tests Were made on the multlgroup ' ’ jmethod to determlne a p0531ble source of ‘error, One test was made to determlne_ the effect of subd1v1d1ng the first two energy groups. Results of the test are glven 1n Table 3 8 from whlch it appears that the division of rhe hxgher gtABLE 3;3 43 R Calculéte&'REa¢ti#ity by Snbdividing Enexg$ Groups CALCULATION . | GROUP S ENERGY V kLpy ‘Original calculation 1 10745 6.07 x 10° ev 1.01253 S 2 | 607210 to 3.68 x 10° ev | Test No. 1 | 1 Broken into § gmups 1.01653 | s 2 Bwoken into 5 groups ' - Test Nd; 2 1 . %ade into one gronp 0.9979 ' | 9 - Made into cone group | enernggfdups'was quite satisfactory in the present multigroup method. -The next theofétidal test was made aSs&ming a'bafe reactor. 1In the correspendlng experlment the rectangular parallelopiped was transformed into a spherlcal core,and the volume of aluminum in the honeycomb surrounding the critical assembly was converted into an aluminum reflector of reduced dens:lty9 hav1ng the same total volume as the honeycomb outside the core. _ It was assumed that the neutrons leaving the core would be allowed only " one collision (according to calculated mean free path in the reduced-density aluminum) in the reflector prior to rétufning to the core. The reflector neutron contribution to the core then depended upon (1} the solid angle and (2) the probability that a reflector neutron made no second collision in 117 r o - 4 s oes 8 & °: e8 G¢ s . eed .0 gda. o8 e e e el e & .8 8.8 e e e o .. e s CRE R T s @ LB L] LR S L T e [ o e 89 . @ o ee e ee e elel TR L Sl A e 8 B L [ e lae 8e Sesl mee 8 ose 8@ ; e e a9 ee returnlng to the core. Integrat1ng over the reflector gave an albedo for the section chosen) : It was found from this test that the effect of the alum1num ehoneycomb was to 1ncrease keff by 1 to 2%, which corresponded to a masse decrease of the order of 5 to 10%. Of all the approx1mat1ons mades the ':smearlng out of the alumxnum Wthh neglects the streaming in the channelsgl is probably ‘the worst, so that the above mass decrease is probably an over- est1mate. Prelzmlnary experlmental results reported verbally to us 1ndlcate" 'that the alumlnum grld decreases the crltlcal mass from 2 to 3%; "experxment agree more closely. P0351b1e sources of error may lie in nct' _ account1ng for (1) the Be(n En) reactlon, (2) the beryllxun:tetai.cross sectlon ::at 0.1 Mev and hlgher, whlch is not known aceurately, and (3) the faet that = :contlnuous slow1nghdown theory in the uranlum resonance region may not hoid Work is in progress to obtaln a better est1mate of resonance Capture as a funct1on of’energy. 'J. GLOSSARY OF NUCLEAR ENERGY TERHS C. B. Mills, ANP Physics Group | "jThe;reector theefy sectionof the ghmsaryofnuelear energy terms 1s nearing cempletion. The many suggestions by others on the panel for reactor theory ‘terms have been 1ncorporated:n1the glossary and the uermsarebenng generalized to 1nclude xntermedlate reactor theory. 118 - - o ; S BT IR YOS S S 2. @6 ee: & eee & ose e e : e e ® oG e e e e a P B s o . 08 @ e e e el el 60 U6 el e e BR300 SR SR ) C R e . e a9 LIS . Cale e e & e e e & 6. & 0 PR e ade e ‘266 . 9.0 08 be e @ 3 a0 f_reduced denslty alum1num of 0.02 to 0.03 (dependlng on the alum1num crOSSH_f. Further 1nvest1gat10ns are contemplated to attempt to make theory and_'u *{ '4 CRITICAL EXPERIMENTS A D. Callzhan,_Phy31cs DlVISIOD, and J F Coneyhear, NEPA P The purpose of the ANP Crltlcal Experlments Program 1s to 1nvest1gate _ "the properties of cr1t1ca1 assemblles composed of uranium and various neutron;' . ';'teflectors and moderators.. Other materlalsg_s1mulat1ng structural memberss;f' e coclants, etc.,of the alrcraft reactor, can be 1ncluded 1n the assembly. The “masse:furanlum requlred for crztlcallty,'the spectral and spatlal dlstributlon 0£ neutrons thhln the array, and the effectlveness of controln are some of :ithe Varlables to- be examzned as assembly components are changed : ‘"rlng ‘the . last three months the 1nstallatlon of the equlpment in. the labcratory was :compieted 'and sufflczent uranlum and beryll1um have been procureé tO begln_:;' fexperlmentatlon late 1n the quarter.*v- - APPARATUS FOR THE CRITICAL EXPERIMENTS The crxtical experxment apparatus cons1sts essentlally of two matrices ’af 3 Ins-square alumznum tublng 3 ft long,'mounted on two tables, one belng: o fistationary and the other so arranged that it can be motor drlven toward the' '7_f1rst. EanChed uranlum metal and the cther materlals under stndy as moder- 'ators, etc, ~are assembled in unlts approxzmately 3. 1n, square and of sultablei“' 'flength and are placed in the alumlnum tnutl'):x‘ng‘s The two parts of the fznal{_ farray are then breught together by remote contrci : The usual safety and - control rods,,the latter for fine adjustment are bullt lnto the assembly9 and the apPrcach to cr1t1cal1ty is monltored w1th neutren and gamma detectors, Figure 4. 1 is an everail plcture cf the apparatus, the movable table ' be1ng on the left. The end of one-half of an assembiy is shown in the topu _of the movable bundie of aluminum tubes and of cgurseg the other half is sxmzlarly placed in the fixed matrlx. Four safety rods, the longer cy11ndersa' and two control rods may be seen at’ the réér-of_each of the afrays;' Attached td'the'fixed half, ét'the interface, is the'motor drive for positioning the neutron éoufceQ Various neutron and gamma detectors are shown areund the apparatus- The sets of aluminum tubes, comprising the interface of half an assemblx are visible. One of the elements, which is partly removed, _cons1sts of blocks of berylllum 1 in, th1ck with uranium disks between them, all _strung on a horizontal rod. The source drive motor is at the top. | 119 2¢. eve. . € ° PP Y O S Y TR BT TR Y ) 2 M e B G e > ‘e # RS S R e e & e R Y T Tlelee L 3 X LN ] CIRE T RO FRIRER X T SRS ) e e e e 4 o el e e ¢ & & 8 e L AR R AN L 2 @ e ®6 008 YRS I LR Y ¥ T ‘.o [ 2 [ ] L) Fabricatxon of Uranium Dlsks,_ The uranlum metal 1s belng fabrlcated:. into disks approx1mately 3 in, in diameter and 0. 010 111a thzck by’fnrstroll1ng'f bzllets to the requlred thlckness and then punchlng the dlSkSa. Since it hasw :not been p0551b1e to: roil to unlform zhlcknes uhe fLfixshed disks were brought to target welght bY Punchlng small holes 1n rhem, " The fxvst batch.-.” eof 1200 dlsks dellvered were found to be heavzly coated with biack ox1dey probably formed durlng hot rolling, whzch because - of the nenadherence of the' "eax1de, presenteu a serious’ contamlnatlon and accountablllty problem;_ The;-e 'ox1de ‘was read:‘y “emoved by leaching in concenzrateé nztrxc aczd ‘after wh1ch”' . the metal rereinedzut room temperature in an atmosphere of about 15% relatlve - *f,humldxty‘ The oxade cca*1ng whzch is now belng lald down ‘in qulte adherent,"’ It has been neces sarvs however, to lower_the targe% wexght of all dlsks by'f “a about 5% beC¢use of the welght 1ess in the_flrst batch durlng leachlngg__; = First Critical Assembly, The flrst crifilcal assembly to be bullt was - ' s1mple 1n structure in order. that 1t would lend ltaEIf readlly to calculatlan, T was to have bery111um metai as a moderators'te be cublcal ‘and ‘to have no "ereflector.. In the proce&ure foliowed an assembly of herylizum was first made . w1th the stepw1se addltlon of uranxum 1n the center,' The. Be/U235 euomxc ratio - of the loaded elements was 386 Cr1t1ca11ty was first achleved with about 6 kg _-of'uranlum in the core and a 6- 1n, layer of berylilum as a reflector. As moree fnranzum was added the reflector was removed resultlng flnally in' an assemhly 1;_21 bY 21 by 23 1ne; with no reflector and centaznzng about 18 kg of U335, e_whlch was crltlcal with one control somewhai removed 1eav1ng a small v01df-: : near the center,- Extrapolatloa to the condlulon ef all yeds 1n, 1.e., no ‘voids, gives 17. 5 kg as the mass. The calculawed Va{ues are 40 to 68% grea*er‘_ than the value observed ' Consequentiyd attempt% "have been made te ascer@aln _ if the dlscrepancy can be attrlbuted to mzsxn?erp etation of the experlmenyg' “the mest lakely cause belng qurzous reflectlon of 1eakage neutrons back into the core by the supportzng structure or by the concrete fioor.i No effect has ethus far been found which will account for the difference. JSbme'verY'fireliminary:meésfiremefits of:the acti&ity ifiduéed in bare and "cadmz&m covered uranium foils at the center of the core shew that the median "energy for fission of the neutrons is of the order of 0.5 ev. " Estimates have "also been made of the power level, a110w1ng rough callbrablon of some of the~ _1nstruments and a survey of persennei shleldlnga _2.1203. BN S e - YT L] @ XX XN YY) ¢ e P eaeeee snse ; EIYLIIX] 5. NUCLEAR MEASUREMENTS '1nterest because molybdenum 1s frequently proposed as a cnnstructlan mater1al‘ in the alrcraft reactor The 5- Mev Van de Graaff aecelerator 1s now belngf' lnstalled in the Y 12 Area and 1s expe”ted to begln dellverlng a proton beam ”:sometlme in Aprll The tentatlve program for thls fac111ty emph331zee neutrofi_ efwerk pertxnent to shleldlng and reactor calculataons | Also of lnterest in The cross sectlon of molyb&enum, as determlned at Columbla Dn1ver51ty LA _exhlblts a strong resonance at 46 ev. Thls measurement is of partlcularg o “.thls f1eid is the Mechanlcal Neutron VeloCIty Selector whlch is atlll bexng'ee-r7”"“ ,.censtruche& mum}gwm causs sgc*rmm masmmsms ' Coiumbxa Unzve231ty The measuremefit of the crcss aectlon ef molybdenum 1ntro&ueed in theE-V jlaSt quar%erLy repcrt {GRNL 919 p 92 ), has been contlnued - The slowaneutron.e i "-[transmxsszon curves ei 25 82 g/cm 'of molybdenum were taken w1th the Columblaefr" _iefinzver31ty Neutron Velficlty Selector by Prof W W. Havena Fxgure 5. 1 shawsf fthe transm1331en frcm D ta 240 ysec/meterg_ The llne whzeh besf flts the data oh "fln thls energy reg1on 15 glven by the equatlen a = 5.7+0. 30E” / The trans~ie ”m1331en of these data has been deiayed beeauee of the resaltb whlch do not -efagree w1th thcse publlshed hy Egelstaff and. Taylsr of Harwell (1) They nge: " their resalts as al? 6. 4 + 0. 4359é ‘ The discrepaney between these two rene. ' fsu1ts is conszderably iarger than can be explalned by exyerlmental uneerwj 'Vtazntleso_ The. Ceiumbla data have been rechecked from the pexnt of view of . both samyie thxckne&s and transm1331ens and it is belxeved thafi the Columbla_jr” fresu§ts are correct The Harwell people state that thelr eample was 99. ?%e E pure“f Since 11ke1y 1mpur1t1es 1n molybdenum are menganeseg tantalum, and 'tfingsten,.all of whzsh have hlgh neutron’ crosQ sectxonag it is eonce1wab1e’l _fthat the 1mpur1€1es mlght change the slope Howeva , it is eXuremely dlfflcult:._,, to explazn the dxfference between 6. 4 and 5.1 for the extrapela*ed value of ' th1s Cross- seet;cn Slnce any ‘of the probable 1mpur1tles weuld tend to in- .eecrease bsth the 1ntercept and theflslcpe ef the llneg 1t is: though& thaz the 3_{_‘ ‘ee'Calumbxa resulzs are probahly the better ones aThe CQ?fillatlfln of data on the tetal cross-section of mo%?bdenum at Columbla Unzversxty is& contlnulag o ~The information presented here was excexpted and edited from a letter by W. W. Havens te N. M. Smith. “H_.(l) Egeisgaff P. A., and Teyler, B T "‘Slaw“Neutron Crees Sectlens ef %eiybdenum and Bremxné*’ wa_ 166, 825 (1950) S TRANSMISSION 04 02 b 100 (MEASURED AT GOLUME!A UNWERSITY BY PROFESSOR WW HAVENS) FIGURE s, I !O 50 ' I | ao u'oso4 93 | oz NEUTRON ENERGY (LOW ENERGY TAIL) 60 80 100 nzo 140 IGO'-' NEUTRON T!ME OF F’LIGHT MICROSECONDS PER METER iSO 200 5 0!5 IN ELECTRON VOLTS % Ovl . o . .. . 220 240 TOTAL CROSS SECT!ON CURVE QF MOLYBDENUM CROSS SECTION BARNES 031418 SYIOND T e D T , Flgure 5 2 shows the transm1531on from 0 to 20 psec/meterw As can be; j9 seen 'there is a strong resonance at 46 ev wlth other dlps in transm1551on at 145 and 530 ev These hlgher energy dlps may be due to multlple levels,;aslt _the resolutlon w1dth of the veloc1ty selector is: not too good in thls energy[fif ” : 'reglon Note also that the hlghest transm1351on values obtalned in this :hlgher energy reglon correspond to a crass -section of about 5 5 barns whlch_'JLA' is in agreement with. the present data 1n the lower energy reglang The results f' '?'ftherefore are self ccnsxstent 'f TEE 5 &EV YAN BE GBAAFF ACCELERATOR Canway Snyder, NEPA The 5 Mev Van de Graaff accelerator whlch has been purchased by the _f"f i Unlted States Alr Force for use 1n the 301nt NEPA ORNL accelerator laboratory}; .-_successfully passed 1ts acceptance tests et the ngh Voltage Englneerlngi o ”ftasts the acceieratcr produced a magnetxcally analyzed prcton beam of from 6 te T ya in 1ts optlmum operatzflg range of from 2, 5 ta 4 Mev and 0§erated -_fsnccessfully at somewhat lower beam currents at energy extremes cf 9 5 and 5. 4_f 'ffievgz The energy stabxllty of the maahlne was excellent througheut the entlref '-31fi¢e it has not been cansmdered desxrable tu'"push" it at thls eariy stagea':fq,'” ' :1ends sahstantlatlon to the clalm that thls is the flnest electrsstatlc S 7Qacceleratcr zn tfie Werid Durlng January and February the acaelerator was. dlsassembled palnted 'u crateé ané shlpped to Oak Rldge, where its reassembly is now in. prmgress 1n3“ ' ."Bnlldlng 92&1 2 in the Y- 12 Area By ‘the end of the quarter all components ':the pxplng was. ccmp1e639 an& the electr1c1ans were’ begxnn1ng to 1nstall ‘the con&u;ts for the rather large number of w1res reqnared to control the machine. '-‘}had been recelved the ma}or 1tems were in thelr placesa about one t%lxd cf" -; It is &nt1c1pated that the accelerator wzll begln del1var1ng a proton beam_' e 1:somet1me 1n Apr11 Rasearch Prfigram.. The research pregram for the acceleratcr is now belng' "7 formuiatéd The advantage of more than a million voits wh1ch this accelerator, :has over. athers w1ll make accesszble for the flrst tzme a wzde fxeld mf greatf ..fl COIPOr&t10Rg Cambrldge9 Mass. ;-on Dacember 28 1950 Durang the course of thef"*- ' energY range explared anfi tha actual upyer l;mlt te the energy is not known,_f? 7' .'ifThe fact that ne auch acce}erator has yet ylelded nuclear data above 4.25 Mev . i i E J 5 : NI s s L i es # se. . eew e i el se e e ess 8 0ss 80 RN TRt ¥ . & ® BRE SN TR ISR NN RN BRI AR SRECIL AN ¢ MY ST AR SRR S S AR R I RO RS S5 & AREL, AR LR AL TN ST R L1 DR 4 e e R e e PR NE B T BT TRSEES RO SRR SRR R SR PRI S N, S HR e B L8888 s 888 [ IEL TN B L UNCLASSIFIED X-10 DWG. NO. 10605 1 SLOW NEUTRON TRANSMISSION =, - o - ~ OF 25.82gm/cm? OF Mo % S R IR S CROSS SECTION BARNES . TRANSMISSION o b— S i g | | [ "o 2 4 6 8 I0 12 14 i6 18 20 NEUTRON TIME OF FLIGHT MICROSECONDS PER METER S T ‘ 3 P i - i - 10,000 500 100 50 40 30 20 15 - NEUTRON ENERGY IN ELECTRON VOLTS FIGURE 5.2 TOTAL CROSS- SEGTION CURVE OF MOLYBDENUM (MEASURED AT GOLUMBIA UNIVERSITY BY PROFESSOR WM. HAVENS) Cd24 00 S ees e @ e ee esie ces 6 sde we S e .8 . ® e e @p .8 Be .. @ @ TR e g @ . 6. 8 ® . : interest to nudlédr'pBYSiés;' After the original calibration and breaking-in | period, the primary emphasis will be on neutron work of interest for shielding. ~and reactor calemlations. | R Lk 125 sfvess. - LI Caeee Y F'S L by : ® s & XX Y B L2 XYL cestar -‘.IQ_O.I_'- e e snoeée cane L] L ] @ © a 2 L BECHANICAL VELOCITY SELECTGR : G s Pawllckl, ORINS and E. C. Smith, Physics Division - The chopper time- of filght veloc1ty selector for operatlon in the neutron:" energy range up to several thousand electron volts is in the process of con—_;: struction and 1nstallatlona The electrlcal equ1pment fer thls neutron spec?_;. trometer cons1sts of the followlng parts 1 Eight?Qfour'counting'channels ‘and cdntrol"circuits° 2. Power supply for 1tem 1. -f3e Ionlzatlon chamber and assoclated ampllflers 4 . Shutter drlve motor and speed control regulator The mechaniCal'equipment consists_of; 'Igfi_The roaatlng shutter,_ 2Q"Neutren beam colizmator and necessary support elements for the . rotor. o - ' .fS;fTInstallatlon fac111t1es for the shutter assembly The countzng channels are compiete and are undergelng rellabalxty checks' in the 1nstrument ‘shop. The necessary regulators for fllament voltage are on han&, but the B supply generator delivery has been-repeatedly del&yed-and_xs again:cfirrefitly due for shifimento The ionizatiefi chamber has been checked with commercial purity BF, and the performance has.been promising. Before final filling with enriched BF,, a study-ef-BFa'purifieation is being conducted in the Instrument Department,' The 5-hp shutter drive motor and the speed control are on hend 'The-rotating.shutter is retaf&ing the entire pfogram, To date the rotor is less than 50% complete. At the present time the supervision of the Research Shops'ie doing'é%efything-possible to expedite the rotor fabrication, and the current eompletienudate of May 1 may be realized. The collimator and associ- ated elements for the complete chopper are very near completion at present, - ~ The installation facilities in the pile building have been designed, and the work order has been submitted., The installation will be'begun as soon as field workers are available and can do the work without 1nterfer1ng with other research activities in the v1c1n1ty L ae . ® &0 29 d. Gee @ .88 s o o ® e 2 » 6. .o ¢ @ & o 0 - * @ [ e a o® e L 2 ) e 9 0 00 L 6. &0 [ ) L 4 o 6 9 - ¢ LI L3 O] e 8. % 50 . 8se@ 9. o o ow 'Y ¢ 8. o oo o8 6 CHEMISTRY GF'LIQUID FUELS : W B‘ Grlmes, Materlals Chemlstry DlVlSlon f The iast*twa'ANp ‘quarterly progress repasts'(ORNL 858 and 919) have ' dlscussed some of the characterlstzcs of llqu1d fuel systems and have de- “ scrlbed the prellm1nary phases of the research deszgned to produce satis---_ factory liquids. Two' types of fuels, suspen31ons of uranium compcunds in ‘sodium hydroxlde and solutlons of UF in alkallne earth fluorldes, are under study.. At this time it appears that ‘the ternary system sodlum fluoride— beryil1um fluorlde*fluranlum fluorzde w1ll find usa1n.the ANP and ARE- reactors, since thls fuel camblnes suitable uranium densxty (about 89 lb/cu ft} 1n a- fuel solutlon w1th a satlsfactory (below SGO°C) meltzng po1nt Other ternary' fluor:de systems Whlch permlt hlgher uranium concentration. are belng studled,;“ A stable suspensxon of a uranium compound in sodlum hydroxide has not been“ reallzed although a suspen31on of adequate uranlnm concentratlon may be - maxntaxned with but slzght stzrrzng. PRASE STUE&IEfi 0? FLUEEIBE SYSTEWS J.;P._Blakely - G;jJ.jNess1e “R. E. Moore = €. J. Barton Materials Chemistry Division . Studles af molten m1xtures of uranium tetrafiucrlde with alkaii fluor1des_ and berylllum fluoride have continued during the past quarter using the 'methods of thermal analysis and of phase separation. The equilibrium diagrams - for the binary systems KF- UR5 and NaF-UF, have been established.; A large. 'number of data on the ternary system NaF KF-UF, have been cbtalned and the » equllibrlum diagram for this very compl;cated system is reasonably well established." Additional data on the systemfNaF«BereUF4 haye been obtained, but considerably more will be required to complete the equilibrium diagram for 4this'sy5tem;; No work on the NaF»LiF"UF4 Sysfem was carried out during the past quarter, since it is not likely that the separated lithium isotopes will soon be available in quantity. The general characteristics of these systems, 197 n By Ay . i SR Lk .opcdofl . L @ ® aseeee o evntee svevw e il L f as 1ndxcated by thermal anaiy31s, ‘are d1scussed brlefiy under 1nd1v1dualffi headlngs belew,fl The letratlon technlque and resuits are discussed under af e 'separate headlng Sflfil&m ¥1uor1flawaran1um Fluorldeo, Thzs system was preV1ous1y studied uAT. "by Kraus.f‘P Cooi1ng curves on the 1ower eutectlc po1nt mxxture, 26 mole %f"'. _UF‘.and 74 moie % VaF and on a. mlxture of: one other composxt1sn 18% UFQ and. " 82% NaF, Were determxned at the beglnfilng 0£ %hls wark and the results obtalnedg"”'“" 'seemed to conflrm the publlshed data However, 1n the ccurse of the 1nvest1; :__e-- 'gatlen of the ternary sysnem NaF KF UF . it began to appear that Kraus 7 equ1l1br1um dzagram mlght be 1n error _partlcularly in the regzons abovefiif ; 35§ mole % flF In addltlon Zacharlasen s stndy of thls systam by ‘the X~ray . H_, dxffractlan methodfz} showad the ex1atence of several compounds 1n addltlon_ Cto tfie NaUFs compound that Kraus data 1nd1cated Therefore add1t10nal study? .w*cf thls system was éeemed Wcrth whllea;¢ The resnits ebtalned are: shown graphlcaiiy in Fzg. 6 1 In the range: ”:8 to 58 maie % UF, che equzixhrzum curve is in essentlal agreement thfi Kraus’ ‘ data., The meitlag p01nt of the eutectlc at 26 mole % UF, is 615 i 10°C ' f_There appears to be good. ev1dence for the ex1stence ‘of the Na UF compound 'falthofigh the thermal data in the 30 to 50 mole % UF range hardly permxt un- 'fequlvecal 1nterpretatlonrM The iow breaks 1a thls reglon whlch Kraus attr1~ _buted to supercooilnu. may be due to a solld transztlon,; Zacharlasen showed- *that Na UFE exxsts in two medxfzcatxonsn and it appears E1ke1y ‘that the Tow breaks in the O to 25 male % UF4 range are alao due to sai:d transztzon,;,Thg . "Tjhzgher eutectzc poznt is at abcut 56 mole % HF 1nstea& of the prevxsusly .accepted 6? msie %,' ‘The meltzng pexnt of bhzs euteetlc 'is 680 10°C Theur ‘Jthermai data do not: show the ex1stence af the Na UF7 compaund 1&&&5151&& by 'Zacharzasen in the 25 mcie % U? _regzcn _ Potassium Fiuarzée¢w€ran1um Fiuorlde Study of this system was. begun eariy 1n the present research program but was dlscontlnued after the eXIStence‘ " of the hzgh meitxng compound K UFn_was found and the absencet:fany low- meit1ng' 'eutectlc in. the 0 to 50 mole % UF ‘range was estabilshed Slnce more completek data en thzs system were needed to complete the NaF- KF- UF ternary diagram, the earlier work was repeated using the improved technlque now in use,_ The :éata obtalned essentlally Canlrmed the earlzer results e .{1}. 'Kfiausg C. A P,‘msa B ag*am ofSome Ccmplez Salfi‘s of Uranizm with Eaécdfl of the A!kai, amd Alkalme Em‘ih Me tals, Mem:a £ Resea‘rflh Labor a‘tary, Brown University, Report M-251 (July 1, 1943}, L (2.)* Zaf:,harlasefls W. l{ C"ys ta’ Structure Stadz,es of the Sysiems NaF-"HF ~and Naf" Lan, CC=34%1 (Jana 16, 00 o.o P R 1 e Tes e eee ::nn :.. Iy ‘@ ; e, e @ el e e 1 -.':; :- : * : Taliel N e e 80 8 e :: Sialelile e sas e ele a8 e e e B @ e ere S owiiel el L e L 8RS e eeE g SEG .8 Gl R 8.8l 18 889 “;_ Tfl j19Q°}“: ‘ sesw®s 0 8 . 8 @ & @ Jeees PR o e LETS LR & L P semaa’ 2098 [ ] asede - - L R 'IEENE SN ) P ne dodeee ssaneel RIRUEEI T 2 _.g,.l':- : ® L R 1 e YT L L AR o K] s . Tesee ol oooioq_'. ’ . L Raee 900 M- - Tefiperatére'¢C’ ._fl'}f,_ - 600 800 M i'?oo-m';i 500 B NaF + Na,UF; oo . NaUF, + lig. -_ 2 se0aed ..IHD.". -8 e @ R o tel esesonoe e L] . 'YX 1) 1000 L] o & [ 2 » 700 Temperature °C | 500 H 900! 600 ¥ 400 UR s Mo 0. K3 UF, + oK, UF,— N & amenioaas = KUFy + aK,UF, F- N 0% UF, . FIG URE 6.2 PHASE m A M "fOF THE ".70 PEREE KF - UFa ESXKEST{Effifi };f;; ;;9 .f | 8501 ON'OMG OI=X 90 100% - Jreport (ORNL 919) xt was 1nd1cated that llttle was knowa concernlng the purlty. '_of the berylllum fluorlde avazlabie in thls laboratory._ SPeCtTOgraphle an&}:;,; chemical analyses were performed on the two batches of Brush Beryillum Co.’ __.product that have been used in ahe present xnvestlgatxon.; The spectrographlcif:_j__, '_anaIYses shcwed no 1mpur1tles except ‘a ‘trace of magnesxum.r The flrst chemlcalfV fanalysxs,.however,'1nd1cated that the beryll1um fluorxde mlght be cc&tamxnate&flf 'thh beryilxum 0xyfluor1ée, whxch has been. reported te haVe the compaszt10n3" -:;2880 SBer.} Later analyses by an merOVed method (see Analy51s of Berylllumfl" E Fluorxde in Sec. 21) 1ndlcated that the berylllum fluor1de recexved from Brush__5_ ' Bery1I1um Cc,,has the theoretxcal compasztxon w1th1n the llmlts of experlnh*'” ":mentai ‘error.:. Hydraflucrxnatlon of the BeF under varxous condltlons preduced : _na sxgnlflcant change in analytlcal results._.' - Ternary mlxtures have a. eutectlc at 480 i IG°C wzth an apprax1mate ' fcempesxt1on of 12 mele % UF l? moie % EeF 'and 71 mele %—NaFG The uranlum .H. . | -_content af thas entectzc 15 prcbably hlgh enough ta make it ef ;nterest as aqf-*“"“ “';Efuei materxal . As mentxoned in the prevxous progress report (OH&L 919} e:f _Furanium content of the other known eutectzc 1n thls system 1s apparentiy too o -law to be of 1nterest Caollng curves have been run on a few binary mixtures of Eer and UF "These shcweé no. 1ndxcatlon Gf a low meislng eutectlc,r.." ' fin the hasxs ef the data S0 far obtaxned -the passzblllty of fandlng in 'thxs system mlxtures meitxng belew 590°C»and centalnxng more thafi 29 mele %] f' . UF appears rather remote.“ The study is belng contln&e& at present to aliow :.c&flstrnctlen ef the equzllbrlum dlagram, . FILTRATION gmmms | A ccnszderably 31mp11fled technxque has been shown to be useful fcr fxlterlng lcw melting fuseé salt mlxtures._ A “flltratlon stick™ COn31st1ng of & szatered glass filter dlsk on the end of a p?rex glass tube may be inserted élrectly 1nto a fused sait mlxture on which a coollng curve is bexng deter-_ 'mzned,_ét the d651reé temperature a vacuum is applied to the filtration stlck "and a sample of llquld is thereby drawn up the tube through the sintered glass medxum.; The filtration stick is then removed from the melt and brfiken to. vobtaxn the flltrate for analyszs__ Such giass fllters may be used w1th cautlon up to about 659°C ‘ I en eae s @ e esl ee 8 sesie ges 28 : T Rt - . e .8 & . k-] @ ] A . - v tite $:° e I LRI S Cee e e el Taee -,:" ee & :: S e I S A . . . . » Ces® tesi e sdeie e we . En. e ERT LN The proaedure employed tc Iacate ternary eutectlcs is as fQIIGWS,,.A Lo 'lternary mlxture thh a coollng curve showxng a deflnxte eutectlc halt 1s-'f-ff chosen as a startlng polnt Thls mlxture is melted and bruught to equ1llbr1um ;' ' at a temperatnre well abQVe the meitzng point.- The mlxture xs then cooled 'slowly to as Iow a temperature as 1is practlcabie and flltered Anether mlxture-: :15 prepared on the b331s of the analysxs of the flltrate ‘and thls sample 1s_ 'm81Ced ‘and filtered at a stlll lower temperature.; By repetltzon of thls _ 'procedure the ternary eutectlc can be approached ciasely 1n a few steps. The regxoa 1n the NaF KF UF4 phase dlagram ef p0s31ble 1nterest as a fuelj 2. 11es between 26 and 35 mele % UF and between 5 and 35 mcle % KF as shown 1n f' 7_Fxg, 6. 3.L Ccolxng curves zn thzs regzon 1ndlcate a ternary eatect:c meitxngxi - at agproxlmately 520°C intratlon experlments conducted by the technlque*" '-?descrlhed abave whlch started thh two dlfferent CQmp931tzons of meltxng poxnt:f :_f "635°C gave, nn successl?e fxltratlcns, the cempes1tzons 1n Table 6 1 TABLE G 1 - fie%erminat:sn of Ternars Eutectic in Na? KF UF R frnm Fiitr&tzon Exfleriments__; gEE -czg-m;xgsmm-w‘e 9 - TEMPERATURE L e '£” 635 _ : _f §2;$. -': f_.:_45:. _ L s s s - 6i$: R ;. §1;57 _1' - :' 5i.1 '% ?: '17~5:,_ om0 s | ':»53::' '-f 18.5 s oL s | 1 . ss6 | a25 o s5° 25 | These results would seem te zndxcate that a ternary eutectlc meltxng at -53@ + IG°C 11&5 xn the region near 27 mole % UF4 and 18 mole % KF.: The cooling c&rves seem to 1ndlcate, howeVer, another eutectlc of 31m11ar meltxng poznt_. - .-_. ;.1533:. . . :_, . H Boabes Taol o & & L 8 Tampal i : o o A $68968 ; : 4. : esesee Ll (XX R XN ® R LET L Yy . sapsen Cesee T o e - [ IO [ Lo DWE KO _-;** *:e - e. ww £ '!" ;.vmm».-_ vmyv A Aw.vfimfi RO mm ummmmmm ‘Ne¥ _Avmwmvmvv vv.vvvv.vmvv. .; 10599~ E R T F’:"GURE__"S.z ' THE TERNARY SYSTEM NaF-KF-UF4 8w e seeon e soaeed L) @ ‘wessas ‘near 27 mole % UF, and 26 mole % KF.: Filtration of this mixture at the .eutectlc temperature resulted 1n ‘a f1lfirate containing 26. mole % UF and 26 male % KF It is p0551b1e that two ternary eutectlcs exist 1in this region: of_V : the d1agram add1t10nal experlments w1ll however, be necessary before 3 ~ def1n1te conclus1on can be drawn._;r. 1@:“ By use of th1s technlque the eutectxc 1nd1cated 1n the lower r1ght-hand7. corner of Fig. 6.3 has been shown to melt at 620°C and to be very close to tne | 'comp031t10n 12 mole % UF,. 67 mole. % KF, and 21 mole % NaF | . SUSPENSIOHS OF vafimxfig*caeefifififis IN SODIUN HYDROXIDE :'efeJ;;D ”Bedmén i Q E.: V;cholson oo LG 0verholser- ' Materlals Chemlstry D1v151on Previous fepbrte"(OHNL4858 and 919) have described ettempts'te prepare “stable. suspens1ons of uranium compbunds in molten sodlum hydroxide and have dlseussed some of the " propertles of the materials so prepared Addltbonal efforts have been made to ascertain the effects of a number of variables on 'the stabxllty of the suspenszons and to 1dent1fy the uranium cempouné present' in the sodlum hydro%ige at - elevated temperatures.' Only a limited number of data are presented since in-most lnstances a qualltatlve statement sufflces to show the effect of dlfferent varlables on the system, Stabxllty of Urafllum Sespens;onsyr Vlrtualiy all the experiments have been performed in a 51lver reactor in the prevallxng atmosphere. A few - experlments which were run in a nickel reactor at 700°C in the prevailing "atmosphere showed that'nickei is_corroded\?apidiy7under these conditions. The pfesenee of nickel oxide appears to cause a decrease in the stability of suspensions of the monocuranate, and, since this behavior of the oxide may mask the effect of the variables under consideration, the uSe.of niekeliveesels in an oxidizing atmosphere was abandoned, Effect of Pfétredtment of U0, Uranium trioxide was used in the majority of the experiments since this source of uranlum appeared to give the most stable suspensions.: Samples of trloxlde were spec1ally prepared from the 135 ® seae FY T L F L XY I sose YT L XX N . sgaevee ’ EY L L ogo..'. PaBEeE L e L ] emawee » peroxxde and from ammonium. dluranate,-but because these products yleided no "better sus ensxon than dld the commer01al uranium tr10x1des the latter was.' p . used most extensively. The commer01al product has a water content approx1~' mating”fihe'mofiohydrate? dry1ng at 300°C gives the anhydrous tr1ox1def. EX?':l . ‘periments ifi-which‘théfhy&ratéd'triOXidé Wés used gave resultS'cofipaiable1t6: _those obtained using as sharge materlal the commerczal trloxlde dehydrated by " heating at either 300 or 450°C Ignztlon of the tr10x1de to U 0 g at 890°C' _ ylelded a materlal which produced less ‘stable suspen51ons. | Effect of Temperature of Farmat;on, Most of the early experlments were ' performed by heatxng ‘the sodlum hydroxade to 700°C and adding sufficient uranlum tr10x1de to glve a suspen31on containing ‘about 5% by welght Sub- . sequent studzes showed that somewhat more stable suspenszons resuited when the tr1ox1de was added to the sodium hydroxlde at 450 to 500° c 1nstead of 700°C - - and most of the ensuzng work utilized this lower temperature° The temperature”' ~ was ralsed to 700°C followzng the add1t1on of the trloxlde and heid there, in most 1nstances, durlng the aglng perlod | Effect of Agtng of SuspenSLGn Reference was made in ORNL 919 to the fact that the suspen51ons became less stable after aglng at 700°C Additlonai studzes have shown that thls behav1or, presumably due to a growth of the partxciesm may be a very serious problem. The data given in Table 6.2 show “the effect of aglng at various temperatures on the stabllzty° The values reperted show the percentage of the orlg1nal uranium remaining in the upper ~one-fourth of the suspension after the deslgnated settllng time (100 corre- sponds to no settllng} In all cases the uranium trioxide (heated-at 300°C, ”passed 150-mesh sieve) was added to the sodlum hydrox1de at 450 to 500°C and the suspensxon was heated at the indicated temperature for the perlod given. 1The uranium trloXide content was approx1mately 4.5% by welght in all aases., The results in thls table aiso show that drastic aging of the suspensions occurs at high temperatures.. The rate of growth durlng aglng increases with increasing temperature and apparently becomes very pronounced at 800°C. The suspensions appear to be more stable at 600°C than at the higher temperatures, but it is difficult to say what fraction of the increaéed.stability is due to particle size and what portion -arises from the increased density and viscosity of the sodium hydroxide at the lower temperature. At the higher temperature the decrease in density and viscosity must be responsible for a large portion 136 . LT L [ L : T E T XTI ¢ "8 @ PAES o ° o FYYYI YL H L ] * “pes80e . sneswe X L] [ o e PBABED » Etfect of-Témperatfiré_andiLength:of Aging on-thefStabilifyfiof-_- . TABLE 6.2 Sodiufi abnqaranate Suspensions in Sodiufi-fly&fofiidé AGING CONDITIONS TEMPERATURE (°C) . TIME (hr) SETTLING TIME (min) PERCENT OF URANIUM REMAINING SUSPENDED 800 oe00 | 700 20 _713' . : _42. 95 20 I R R CATE SRR ,. | 98. - 86 83 58 61 44 34 187 ® sosome LA R LB AS conoen ensvoe @ 9 b4 sevens . - T eae® of the instability absérvéd:{'fie@ever, the resu1t5~obtaified-at'300°c"show' -quzte markediy the decrease 'in stah111ty that results from the aging at’ thls'. “high temperature and suggest that the rate of partlcle growth is. accelerateda . by a rlse in temperature._ This behav10r may: prove to be a real 1mped1ment since not only must 1t be poss1ble to prepare a reasonably stable suspension. f;but the system also must remaln stable for reiat1vely long perlods of t1me atj_ "hlgh temperatures 1f such a fuel system is to have any practzcal value. Effect af Add;tzve In view of the drast1c changes noted above whxch aoccur on. aglng, stud:es weTe made 1n an attempt to find some additive that wouId Inhlblt or, prevent thls growth of partlcies.fi It appeared doubtful thata 'anv baneflclal eifect would result from the introduction of another solid A"phase because of the deieterlous effect of nlckel oxade on the system.fl This limited the study to those materlals which are soluble in sodlum hydrOXIda atq' .1?90°C The effects of the addltlon of some materlals meetlng thls requxrement. ".are descrlbed brlefly below,t' The addltlve was. placed in the s1lver reactor,‘sodlum hydrox1de was aadded the temperature was ralsed to 450 to 50G°C sufficient uranlum tr10x1de ?was added ta make the suspenslon approx1matelv 5% by weight, and the ‘tempera- ture was then raised to 700°C. Sampllng for the settllng rate determlnatzon' ”was usually done after the suspen51on had aged for about 24 hr&r The addltlon of 1 toe 5% sodlum silicate had very llttle effect amounts. of the order of 5 to 20% resulted in decreased stabzllty.j Addition of 1 to 20% of eather thngsfiic'a01d or sodium chromate produced very unstabie suspensxons.; The presénbe'of molybdate appeared to be without any marked effect. Sodium !avanadata, when presant in amounts apprcx1mat1ng 20 to 30% by welght resuitéd in suspensions which were more stable after aging for 24 hr than were those to which no material was added. Smaller amounts of vanadate showed no beneficial effect. It is impossible to_stata whether the increased stability is due to changés'in density and viscosity or is to be attributed to an effect on the paréicie size. The results obtained using these additives are not encouraging and one cannot be optimistic about the possibilities of finding a stabilizer ~ for such a system.; Suspensibns of U0,. Experiments in which wranium dioxide, UO,, was used _ 2 P . 2 as the charge material under a normal atmosphere showed that the uranium com- pound was oxidized and that ‘a slow transition to sodium monouranate occurred. 138 IS % 4 e T E ] FTILLE] *GHe pme e . ® sn0e fpee sege [EE TR A ® L 3 ‘Sonded geedAw L RUEN -] i LXK LK ] _It should be noted that the suspenslon farmed from the d10x1de was very VUunstable and that the stablllty 1ncreased as sodxum monouranate was formed. It s appears that if UOQ Were used in a redu01ng atmospheré 1t mlght be p0331ble;' "*to &Vfild reactlonj'lt 1s not certa1n..however; that Improved suspen810ns w1llj=' ';:'be p0531ble by thls means.__' ' :Identif1cat1on ef the Uranxum Comgeand : It has been 1nd1cated xh'a' previous- reyort (OBNL 919) that the addltzon of uranium compounds to soéxum]f- hydr0x1de 1n cxzdlzlng atmospheres ylelded a. materlal whose characterxstlc:f” X ray. dlffractlan spectrum was not 1dent1ca1 w1th that of any of the startlng ¢ , f;materlais., POSlthe 1dent1flcat10n of th1s materlal as sod1um monouranate,ff" ' Na¢UQ§, has not beefi possxble frcm X ray dlffract1on alone;,chemzcal studies 'haVe recentiy suggested Very strongly however, that Va UO -1s actuaiiy.; ”«.abnalned,; ;ff~7 Attempted Separatzon af the Cnmpoand’from NQOH ; Severai attempts haver w"_been maéetx)reCQVer the uranxum cempeund by vacuum dlstzllatlon of the caustxc ' at SGG to 859°C fram 51lver cruczbies ccntaxned 1n stalniess steei reterts.;f o Contamznatzen of the r831due, whlch was noted 1n the flrst &1st111at10n ‘was ';;prevented in the snbsequent dlstliiatzons hy COVerlng the 511ver cruc1b1e wzth; _ a ioase flttlng SIIVer platen“ With this modlflcatzen. reiatlvely pure arange?' iceiared resxdues wera obtalned afiter &1stlllat10fl of the sodlum hydr0x1de was- ';fcomp}ete,P Chemlcal and X«ray dlffvactxon anaiyses shewad that the resldues.' _cen31sted prln01paily of sadlum dluranate,‘ Thls was net anticzpated but may' ;3§¢ expla:ned_fiy assumlng.fihat ;he reactlcn_"fi . 2Na,U0, ~—> Na,U,0, *+ Na,0 ”~_§ccfirs.darifig théliééé§75téées ijthé distiiiéticn,; Syntfiesas of Sodvum M@neuranate Mlxtures of sodlum carbcnate and U 0 shawed ilttie ev1dence of react:on when heated at temperatures as hxgh as 21250°€ Reactxon was obtalned by heatlng mixtures cantalnxng U, O Na, CO~ "and VaCl to 950°C.- | Attempts to leach the unreacted Na, CO and NaCl from the product however. left an- orange r631due which was shown by X-ray dlffractlan- to be sodium dluranate5 Va U O When stbichiometrlc mzxtgres of sodium carbonate and uranium trioxide were fifiely grpund,‘intimétély mixed, and heated at approximately 1600°C fof, ‘we. . wmve & 8 E. 3 s Se g SEe 8§08 98 LR A e o & L A B LI TR TR IR S ] R I T e :R L] 6 ee e es 8 & 6. 8 L3 ® L4 s & @ LA L LA PO TR O » i e 88 FYE'Y X ] @ eed % &9 &€ 5@ 4 FER - @ 1 hr a pale orange SQIld whxch showed eVJdence of 31nter1ng was obtaxned 5 ' z;fChemlcal analels gave a sodlum/uranlum ratio of 1 97/1 00 and a carbonatef” 51 centent of apprexxmately 1%., The X- ray dszractlon pattern for thls materlaLE is 1dent1ca1 wzth that obtalned when uranlum tr10x1de is heated w1th a large ' 1excess of sodlum hydrOXIde at 700°C and xs assumed to be characterxstzc of 1isod1am monouranate.k Less than- 10% of scdlum dzaranate was faund by the X- ray :f '?dxffractlcn analys1s.; When the exper1ment was repeated at 1250°C a mlxture A3c . of sodzum monouranate and sodlum dluranate was obtalned 1n whlch the diuranate " '30atent was af the order of 40 to 50%,~ ' Since the precedlng results 1nd1cated that soé;um moneuranate mav be' "prepared frem sodlum carbonate and uranium trlfixzde if a temperature ef 1669°C;3' e hlgher ig used experiments were rnn u31ng an 1nduct10n furnace to atta1n " ?gthe de51red temperatures._ A run at 1750°C '1n whlch a platlnum dlsh was :” f partly meited gave a layer of reiatlveiy pure sodlum monauranate._ There= . gg_ "was ev1dence that the portlon af the materxal 1n contact w1th the meited piatznum had been fused suggestzng thac the sodlum monouranate may melt 1&1“ the VlClnltY of I?SOGC The bulk of the mater1al had remalned at a temperak_ JEture censzderabiy belew 1?50°C the failure of the p}atinum was due to local '“ ;averheat1ng.‘ To avozd thls dlfflcnlty wben us1ng platlnam.'magnesza C&ps_;;” 'preheatad at 1860°C were trled These cups proved to be too poreus$ the__ 'fmafierlai dlffused threugh ‘the magne51a cup and in contact with thejgraphlte « hoider burst 1nte fiame at a temperature ef appxoxlmateiy 14083C 1n an. argcn _,_atmosghere«_f The fazlare to fznd a’ saltabie contalner £or the hlgh temperature sefi1um 'Fcarbanate~«uran1um tr10x1de reactlon resnlted in further attempts to: pregare' '-soélnm moncuranauu by other methads | A mlxtflre contalnlna 2 meles of sodium nltrate per mole of uranium ”rtrloxzde was heateé fcr 1 hr at lQSGOCQF A 51m11ar run was ma&e at ISSGQC Both *uns ylelded apale crange residue which appearedtc be sodium monouranates _but the pray dlffractlon anaiyses of these materlals have not been compieted,f Stabzizty af Sod?um Monouranate A mixture of uranium tr10x1de (5% by “_wezght} ané sodium hydroxxde was heated in a silver crucxble for 2 hr at 600°C under an atmesphere of hydrogen;; X- ray diffraction analysis showed that only 'sndzum’monfiuranate-was present.f Thls compound apparently is not reduced by' 140 L] L] *EREBD saed wEEEX] YT seseed o pasese BB GBEE . A e e FYSETE | hydrogen findér-theSe éonditions.j A sample was taken from the uppér fortion" -for analysxs after the sollds had settled and was. found to contaxn 0 008% uranlum.g Apparently uranlum is not soluble to any slgnlflcant extent under o the condltxons used FY PR T 3 ®: . - a8 Pe & Cee-% €93 S0 T e w8 LI ST e e L L TR Y BRTE SE T I o LT e N * e B @ 88 L] .. e @ o L) X T RSN 28 8 8 & . @9 o ¢ & . .3 o 8 & @ . e *. 9 @& L2 ] " e L X L X ¥ 3 e o008 . 2 2@ a0 8 & L rY 3 MATERTALS RESEARCH s & edese -nn‘rt _ INTRODUCTION TO PART TII ~In the final analysis the ultimate'sucéess'of any~reactor'program'dépéndé;; upon the satlsfactory solutlen of many serious material problems.. The many_ ~" spec1f1c requirements of materlals for the aircraft reactor comprlse a . de51gnf'_" problem unlque in’ englneerlng experlence.;The most pertznent cf these requlre»-_' 'ments are that the materlal do the foilow1ng l.f;R§31st 11qu1d metal corr051on,_3 .'2.; Possess hlgh mechanlcal strength :3;-.W1thstand attack by fuel and moderatora_ - . 4:j‘Be formable 1nto thln walled structures.k ' 5fj Be amen&ble to productlon of soand welds._ GQQ:Be dlmensxonally stable at operat1ng temperatureé." fj}g.Not be subject to thermal grad1ent transfera 8:;:Have faVorable nuclear characterlstlcsn, | 9§: Have reasonable stabllxty under radlatlon bombardment : 10;f'Be avallable 1n unlform qualzty,f In add1t10n to the above requlrements, nost ef wh1ch pertaxn speczfxcally-' 'to the structural and contalnlng mater1al in the reactor, the liquid fuel and the ilquldumetal coolant must possess certain thermal and physical properties,. ‘notably gpod-thermal conéuct1v1tyg high heat'transfer'character1st1c53 and low "'_fiscosity- Furthermore; these liguid- metal systems will requzre the develop- ‘ment of handi1ng equlpment (plumhlng) such as pumps, valves, flowmeters, bearings, insulation, and slmliar fa0111t1est The bulk of §he research*on theSe problems is now being done in various laboratories of the Oak Ridge National Laboratory which are associated with therAN? Division, and is discussed under one or more of the following subject headings: Corrosion Experimentationfi Liquid-Metal and Heat-Transfer Research, Components of quuld -Metal Systems, Metallurgical Processes, and Radlat1on Damage —~- Secs.vll through 15, respectlvely._ At this time the corrosiveness of wsodium, ‘IS belng investigated with the family of metals whose known o } - act A RV R 3 N f \ ) 178 - ; \ : G L4 ] . . k “a® sas. e 9 ¢ " ep o0 & . eed b 286 04 e 8 o e 8. 0 @« e o & . @ ' ® L) s a 9 @9 ® @ o & [ L'y * ¢ o Go__‘o L3 o & . ® o 2D 6. e a o e e @ ¢ ® s &, @ ®. .0 ® TR o 8 e & ph . S0s. 6 @ » ne¢. 86 e e . 0 688 g8 characterlstlcs fulflll the requlrements listed abO?e.g These eontalner .materlals 1nclude e var1ety of stalnless steelsy:nlekel alloysg‘and re- fractery metals.} The corroslveness of ‘the fluorlde fuel must be examined w1th' these same metals sxnce the metalllc cong ainer for the fluorlde melt wxii- the reactor,_be 1mmersed in the coolant (11qu1d metal) streamfikThese corr031on~' problems (Sec.,ll) are now bexng thOTOthLY examlned under statlc condltxons_ although the subsequent tests 1n c1rculat1ng loaps are in progress for some-' i : systems., | Many fundamental data are requxred on the thermal aed phys1cal propertles - ef both 11qn1ds (coolants and fuels) and metals. The heat transfer coeffl-e '“c1ents ef [ : NaOH (whxch is stxll of 1nterest ‘as a reactor eAfuel — Sec. 16) are belng determlned These values and other 1nformat10n 'concernlng prznc1pally the thermal convectlon in the fuel pzns, flow char- \'acterlstlcs of the coelant in varlous passages and the electrlcal eonduc21V1ty of the fluerlde melt are dlseussed in Sec.‘12,j In. add1t1on the Heat Transfer. o Research Group has 1n1t1ated ‘the edztleg of a rev1sed quuzd Metal Handbook.: The deVelopmentof llquld metal systems and assoczated handlzng technzques beleng te the new technology ‘which is rapidly becoming well known as "reactor 'plumblng.h¥ A number of pumps,'lnciudlng axial flow, centrxfugal electro- - &agnetic.-end "canned roter"éare belng tested on these systems. Teete were .1n1t1ated this quarter on mock- up of heat exchangers. As a cohsequence of this research with llquzd metals -adequate safety measures, disposal proce» dures, product control’ testsfl 1nsu1at10n and other teehnlques and procedures have had to be estahllshed ' Outstandlng in this field is the work cf the _quuld ‘Metal Safety Commlttee which wzll soon issue a safety manual ‘for the 'handllng of llquxd metals,_ Tfiese deveiepmenes aredlscussed in Sec. 13. Labarateries for stréés%rfifiture testingrzfiofider mefeiiurgy; and welding have been established for the deveiopment of fabricating techniques for the ~reactor itself.. Of particular interest are the welding of melvbdenum | and" the stress- rupture testing of metals in a 11qu1d ‘metal environ- ment aithough both these programs were delayed by the installation of labora- tory equipment., The fabrlcation of solid-fuel elements, however, has been: ' eXtensivély'pflrSued and three techniques have been developed.: In addition, the examination of electrolytzeaiiy etched single- crystal copper spheres may lead to a better underotand1ng of the fundamental physzcal chemlstry of liquid-metal corrosion. These awtzfatxes are discussed in Sec. 14. 179 I * .8 » » sossee snaue : I3 . dnaes ° 'y osbeso semane . ) T eooeee ° sgento. sapn evpses 2008 0__. ® e ss0s ® L d £ o L ] The radlatlon stabllzty of core mater:als is under 1nvest1gat10n._Effects of radlat1on damage on the creep and thermal conduct1v1ty of 316 stalnless steel are. currently under 1nvest1gat10n A program has’ heen estabi1shed at Purdue Un1vers1ty for sxmlllar work on molybdenum,-and preparatlons to date include the'establlshment of the th31cal properties’ of molybdenum w1thout Irradlatlon.“ The ra&latlon stabllxty of the. fluorlde eutectlc NaF UF, has-:: been tentatlvely establlshedh These rad1at10n damage measurements aredzscussed 1n‘Sec. 15.- R | | 180 su s e e e ee . o6 4 Bee A ase o e e e e @ s 8 o 8 FRE :-:__._.: e e o8 B3 ° 8 e e @ e oo * s I FEEEY TR o @ ® 8.8 e ¢ 8 - & o o . # &8 e e : N e $8 [ esd % & 80 e LI : | _. ! 11 CBH&“MSI(HV Efi&flflBIMEEWTATI{WQ . E . C Mlller and W B Manly, Metallurgy DIVISIOH : B H’ W Savage,:ANP DlVISlon Sl '__' Hageiston, Isotope Research and Productlon ' ; P The past quarter has seen the substantlal completlen cf the 1n1t1al phase-~r~*'53 "_of the stat1c corr051on testxng.; The fxrst portlon of the prcgram was to Tdafiermzre 1n a rough qualztablve way whether or not a metal or. alicy had anyf_i' cLa%cn o thhstandxng attack by 11qu1d metals at elevated temperatures.;5sya* - '.téia SCleutlve weedlng, many metals and’ alloys have been e11m1nated from_j'° ]_flnfnrther ccnsmdflratxon._The second portlon of the program has ‘now been started.j"' The metals and aliays whxch shcwed some resxstance to attack are now helng ” fsub;ected to vlgorous, mere quantltatlve corr031on study thie certaln re» 'fi-fzfiements in bcth apparatns and technlque are. yet to - be made ‘and evaluated it s felt that ‘the recently adopted methcds for the studyof attack reslstance“ ; ’wxii produce worthwhlle 1nformat10n._ The evaiuatlcn of reactcr metals for use Wlth a- certazn coolant un&erf 'fl-ngen condltlons of. cycilng or varyzng temperature and of a lzmlted degree 9£ 3-"' 'Lp&rlty 1nvolves mare than the clase study of 8. cerroslcn.rate determ1ned under3 | *{test éandztlfins where very pure mater&als afld 1dea11y prepared surfaces are:. involved, no thermal or composztlen gradlent is present ‘and " temperature is fl ma1nta1ned at a constant level Dynamxc carras;ofi tests in bcth cenvectzon '7;and forced czrculafilon loops have been initiated and signzficant 1n£ormat19n ran the iaw veioclty czrculatxen of sod1um has already been ebtalfied The fundamentai metallurg1ca1 ccn31derat19ns 1nvclved in corrosion 1n~__zm L dzcate that a barrler fxlm mxght be fcrmed on the surface of the metai whxch}f 'woaid lnhzbit the corrosive propertles of the ceolant A search is . 1n progress"” to find s&ch a dszuszan barrzer for systems of 1nterest to 0therw1se feaslbie 'ixquxd me&al systems.i_ The static corroszon by flucrxde melts will determlne the p0531bie con- tainer for the izqulé (fluoride salt mlxture) fuel.: It is notable in this program-that the guiescent nature of the liquid fuel in the ANP reaétérf-' mitxgates aga:nst the necessity of dynamic corrosion testing of the fuel, Ialthough th1s would eventually be deszrable., Some materials, e.g., monel and ® L ‘semewe L] FEE X TS . L2 XX ) cRed Ceewesn. . T : son. b ees N BSL AL ¥ e @ - L BN B [N TT-F ® A : b ; . ; ERE SN L3 B : b ._. & e & 5 B ) . = d9 | eee. & : sae [ e ® ety nlckel A have been found to have satlsfactory corr051on rates (less than 0 03 mll/hr) but the compatlblllty of these metals w1th other requlrements (namely, high- temperature strength) is’ not favorable., The tests are by no means compiete.f_?fi%' STATIG CGRROSIQN BY LI&UID fiETALS ' A D Brasunas, Metallurgy DlVlSlon o U '*Additional’data:are'béing gathered concerning'the resistance of fariouéf"' metals and allfiys to llquld metal corroszon under statzcand_dynamlc condltlons._;_,“ The coolanfl _: sodium has besen investlgatad with refiaré to 1+s 1nter¢ct10ns Tifihrceftain % fi';solld mecals under statlc tegt condltzons at 1000°C. N . Yo A crltlcal study of the capsulatxng testlng technxque has 1ed to the. csmponent may be eilmznated by the tubulatlng technique now helng developed 3 Both these technlques are descr1bed below._ 'resting Techniqne' The manner of en01031fig specimens in capsules for statlc corrosion. testlng in molten- metal media is illustrated in Flg.{ll 1. The metal spec1mens are machlned and polxshed to size (1 by % by ¥ in.) and:. carefully welghed after vapor degre351ng.JBelat1vely inert metal test capsules :(n1ckel in the case of sodium : BRI ' e - may. be prepared by maahlnlng from bar stcck,, The specimen is first_inéeited,ln the capsule and liquid metal is poured into it, in a dry box chamber cohtaining' argon.: A cap, inserted in the capsule, is then heliarc welded into placeg 'While under vacuum, the extended tube is crimped, spot welded and bead- welded, thereby enabllng ‘the corr031on test to be conducted at temperature in vacuum. The major undesirable feature here is concerned with the relati#gly‘ inert me;a1 capsu1e.; Although the'solubiiity'of nickel in sodium is very low, 182 ; e i r 2 ;,,fl o - Cee e e el el e e 8 ase. 8. 680 08 -0 > @3 ® 9 ® @8 o ® iR 8w s 0 e W @ o-® ‘e "8 @ 8¢ - 0 68 .86 6 2?0 @ e e ¢ o8 e s o . e 8 ae Seeel e &8 @ BIiia .6 800 ¢ o 0. ‘80 408 @ esd .2 ae [ 2] 8 8 . % $80 @6 "éeveiopment of a more satzsfactory manner of testing. The undesirable thlrd 3“" Z-a substant1a1 layer of nlckel may be dep081ted on metal spec1mens ‘The_ --_presence ‘of such fllms serlously challenges the valldlty of the test datag The same phenomenon has been observedxn lead tests conducted in iron capsules.js Conszderable effort is currently belng placed on the development of the'Ae "'jtubulatlng technlque 1llustrated in F1g._11 2 in whlch the: tube and spec1men.z_-' are of 1dent1cal compos1tlon., ‘Mass transfer caused by compos:tlon variations ”_'should be nonexlstent under these condltlons Thls technlque is. essentlally.: '_Slmllar to the above capsulatlng tochnlque in other respects._ The maJor developméntal problem is concerned with the malntenance of a hlgh degree of :_molten metal purlty dur1ng the flllxng and seallng operatxons._ Aglng and-: fllterlng'the molten metal slzghtly above 1ts meltlng po1nt together with thelr L necessary getterlng, constltute the present plans for purlflcatlon., Statlc CGrrosion Lead Many corr031on tests in molten lead at 1000°C' :have been conducted for 40 ané 460 hr.; These have been made in evacuated 1ront_ ":capsules as shown in F1g 11. 1 - The extent and nature of corr051ve attack N have now been detenmlned for most of the materlals tested Forty hoer testf ?s data are reported in Table 11 1 and the 400 hr data are glven in Table 11 2. 'fPhotomlcrographs 111ustrat1ng the nature of and depth of attack on numerous metals are given in Flgs._ll 3 through 11. 9 ~ The tests in whxch mass transfer o :of oapsule materlal to sgec1men was eV1dent “as in Flg 11. 5 ‘are. 1ndlcated in . Table 11. l by the symbol "Mt.? These tests w1ll be repeated under condltlons ':where thls phenomenon oannot occur, Iron showed no metallographlc ev1dence of any attack (Fig. 11 3a) in 40? ehr at . 1000°C A welght loss of approximately 0.1 g was detected which cor- tresponded to the change 1n spec1men thlckness of 0. 001 in. {0.0005 in. 'removed from all exposed surfaces). Low- alloy steels containing 12 and 16? Cr like- - wise showed very 11ttle evzdence of corrosion. However, tiny gray spheroxds ‘shown in Figs.11. Saand,éq were noted along the ‘surface to a depth of approxi- 'mateiy 0. 001 in. in 40 hr and. about 0.005 in. in 400 hr. Type 446 stalnless steel (26% Cr)fi shown in Flg ll4b shows apprecxably more suscept1b111ty to such attack The presence of iron layers on the surfaces of molybdenum and tantalum is shown in Figs.11.5a and b, respectively; similap films were detected on columbium and zirconium.: These tests will be repeated.- 184 L * soesee” Cemee i "ewoeses - * ° (X LT LY BHrEE ‘eooopee . YT Y Y X i @ csesne T eees ° ® R TAB&E 11 1 static COrrosian nata {)htained in- 40~ hr ’xests Made in Iron capsules at mofl (} in Lead | * SUBSURFACE _ S S EeE s 4 : "Pmfii%gfinm fiufig?gg DECARBUR '.FIEUJ". i S s patn | werear | o ARBITRARY [THICKN s INTERGRANULAR | S - |DE = FIIM - - b o BA 1 Sares | cowpmyzgficxmc;g% PENETRATION | =~ GRAIN . | MATION - | IZATION | FOBMATION | -~ . R ANALYSIS | CHANGE - CMETAL | RATING(®2 1 (in.) (in.) Gn.) o} Gag o Ge) | (e e REMARKS o ((ppm)} ()} (mg) - Elements:| ~ Fe -~ 10 **0»0605 None None " None Ll None - . .- No ev1dence of abtack eray ;-Fq;'37 L _ -1203 ;ff.: i Sl L . : o . B R N T I dxffnaetlon detected no changg S b “ffw S 10 None None None - 'None e I Ndne _ ' _’ No- ev1dence oi attack X~tay FE. 155 : +35_f 'f:F' - ; o o - A . diffraction 1nd1cated the W, ND b : . : : o B _ S presence af We oo o : R S s e o Mo 9, Mt(d)' None | Nome - | None- | None - ] Quter 0.0010° ;Fe deposit detecbed on Mo " {Fe, N.D. | -~ .+250 = .:OQD.: - . . R . | . J.IT- . .-‘:. : Inngr.oiOOQI ) sur ace _.... | MOJ N De i o ?. e gy Ta 9, M Nonme | None - None None - | Outer 0.0002 | No evidence of é?tack;.:x-ray {Fe, 153 | 4150 | - o . _ | o _ R Inner 0.0001 | diffraction indicated TaC - |{Ta, N.D. S f’. ‘; cb 9, Mt +0.003 None - . None - None 0;013 . -'0,004 | No evidence éf‘attack;.Xéfay: Fe, 933 7-_.+1401'j,'f_31T : - ' ' S o _ oo b diffraction indicated Pb; Cb, N.D. ; - ' specimen 1ncreased in thxck- ness .- . el Zx L 9, Mt None Nooe = None | ‘Nome = | 0.025° | 0.0001 = | No av1dence of actack; X-vay |Fe, 3 : s _ ' _ o . o o : L dlffractlon 1ndgcated ZrO Zr, N 5 e Tes®ol B . 6 -0.004 | 0.001 ~ Nome | Nome - | Nome | Surface appears severely ~ |Fe, § | -300 | sneddle S ' y : S ] RO ] R ferw1te tranSak formation at surface and along grain boundaries teo deptg of penetration noted; - X-ray evidence also. showed presence of Cr203 ' Austenlte =3 ferrite trana- | formation at surface and ~ adjacent grain boundaries; formation of large white ‘metallic phase in.grain RON .'_(e);} |- (e) :(é)_ . (o) | :(é’;i :-.;; boundaries near: surface, caviq ties. w1th1n Austenlte ferrn*e ?ranSa formation at ‘surface and ad- Jacent grain boundaries; at- tack more shallow on flne« graxn steel Numeruua rain bnundury Voldu' _ (¢>1 in decarburized zones; X-ray | - diffraction indicates face- ': (§)*_ ;‘ : lims ) :€f9¢; "fi - (@) {b) o :(c) {d) - {e) A g 10 = excellent 5 = poor, 1 = N.D. = not detected. Mt = mass transfer probable - Not yet ‘available. LG, <0.04% C ELC, <0.006% C See. remarks. dissolved. -Values reported are for dxmensxonal changes from one surface only. / -+5 rfL if “€; . . . 33 A .. N NI o o o PREFERENTIAL ATTACK, HOWEVER = C TABLE 11.2 ‘Static Corresion Bata.Obtained.in §fl0»hr~Test$'mfid¢ in 1foh cgpsn1¢afat,10Qofic*1n-Lead-(a} _'~..STA;{NL ESS steELtt) ABBITRARY COMP ARATI VE RATING' ¢ THICKNESS cHANGE ) INTERGRANULAR| /1" PENETRATION (in.) | “{in.) ~ SUBSURFACE PRECIPITATION WITHIN © GRAIN {in.} SURFACE | TRANSFORMATION| DECARBUHIZATION {in. CFIM WHV’ATION ' REMABK$ ::;. vt | ';CRANGE B (mg) '.1fi 405. 410 Lo, 430 430 ELC . - 446 446 FIC 446 ELC . 304 apeaed send 304 LC 304 ELC a6 4T I -0.001 1 +0.003 -0.001 -0.001 ‘None -0.802 None None -0.001 None .0.002 +0.001 -0.004 0 <3;> CM: oo 001 010 {avg.) .026 (max } .007 {avg.) 015 {max.) 035 (avg. ) 045 (max ) . OfGOS(e}” 0,006 - 0.006 0.007 V 9)005(3) None - Noné: _.fione' - None “None . ~ Nome . - . None 'N0ne_" fan. ) None - None '; . None ~ . Nomne None NOne fi . 0.008 10.003 0.003 0.008 0,005 _'0,_093 : ) 1' 0 0.006 ~ 0.005 0.028 008 300125 Bt - 9925 {. . Nome - -~ None - None - - Nene < = .- None ; Nflne . “I'within -iap?h ¢ 3 *! Sy Ewsdace Lo depth of 0 006 1n SR ; ,S1mllar Lo ahova | decarburized zone reported forthis - | low-carbon alloy may actuaily be : : slgma free zone. - . © § Gamma - .Samé _ | gamma b 11ny gray globules pre01p1tated 0 005 in, o Flehuies pr301p1tated below Same except that the 3 Pearlite-like preclpltatlon along surface to depth of 0.005 xn : Fray phase observed in grain bound - aries and w1th1n granns at surface alpha transformatxon alon surfacc and adJacent graln houn arles _ Samejj Spncxmen com€1ately demarbuw1zed h pha trnnufuwmamzon_ ; alon o boun araaa "None w\.Gamma -—§> aipha mransfommatxon alon L bcun aries o . .Nonej - Gamma surface and adgacent grann alpha transforma&nmn a& surface; “grain boundary vund ob. aervad t@ depth of 0 0 o .+i35 'f:5_'”}f ains below 5ur£ace to;j_: L e +300 -}Hmfiff? . +90 ; "fi lff %256 : [”.“ £300 %75.:;i.g' :; :@135 e e +190 : EZo(‘f) = a0 | aurface. aad adgncent gra1n e _‘*45@2' ; fi;;: La) Bath analysis not yet avaxlabie (br LC ifl 04% C ELC <0. 006% - 0= exwekienh 5 # poor, -1 = dissolved. IS Values mapormed are for dlmenslonni changes from oae surface only See remarks ' (8) Y Weipht change of two. dampleo@afifj _ ' Tfie:titénium' beryiilflm and nlefii oP@“lMQns “showed appreczable welght' -lloss and reductlon in- 5126,, A close Dtud'y’ of thelr surfaceb‘revealed ba51c-. -,dlfferences 1n the nature of attaek Nickel was severely attacked inter- .granularly, as shown in F1g 11.6a. leaving grains completely isolated from the1r nelghbors;_ Berylilum. Fig. 11.6b, indicated’ preferenulal at s aCk'for'fi '"différeht'cr?séalxogxaphzc dxrectisns the graln boundarles were not sub}ecf ';to greater attack than the grain pvoper. ‘Titanium showed ne1 her graln' fiboundary nor preferled ‘directional ttack by 1ead as 111ustrated1n.F1g 11 7@. Thorlum and uranlum spe01men¢ were “ery severely aztacked under szmllar test “ _condltlons.; :' R SR I e Austenltlc stalnleos steeiav shown in Flgs. 11 7b through 11.9. have been' :fouud to be susceptlbie to the fcl;ow:ng types of surface 1nstab111ty at 10@0°C when in contacfi with lead. solution, 1ntergxanular penetratlon,ssz« surface pre01p1tab10n of unknown phases within grazns and in grain houndarles, ' o fapparent decarburlzaulon and transformatlon to a body centered cublc phase ‘in the lead affected reg1ong.; Furnhermerea 1ntermetali1c layers may also be present on the su;faces of steels or other metals. Their meiting p01nta, -however, may be close to the me;tlng point of lead and uhey may escape detec- tioncmdngto flw:present method of separatlng the 3p801men'frqm the bathu | The corrosion data for atalnless ateels in lead at 1000 °C reported in .Tables 11. 1 and 11 2 1nd1eaue thau alio"*'contamnlflg smalier amounts of _chromlum and-nlckex eapeflaily the la*ter. appear to w1thstand molten. lead'” .attack better than the hlgher alloyed metals. - The weight change data reported in these tables are not par 1culariy algnlflcant since they do not - conszder occasional mass- tran¢fex;effe £ 1_.,.}?:of-s,szlblcw:' lead retenL1on,'or the mode of. weight loss dls"“zbu ion (1. ea; 1ntérgranalar-or general éttack);j A certain - amount of general attack can be tolerated. whereas ihtergrafiuiar attack must be kept to a mihimum;i Likewise, thickness changes of the Spécimen'aré'nofi entirely free of extraneous influence. Mass transfer and unsatisfactory cleaning may be'cérrecméd, bat dimensional instability due to gammamtosaipha transformation or solid solution can cause. and has caused, an actual increase in speé¢imen thickness in spite of possible solution (alioy 310 stainless steel in Table 11 2) Lfithbum Addltlonai tests are belng made in lithium. but the test data are not‘yet avallabiefi; The tests are of 40 hr duration and include binary 196 ‘sepews w sese EX R TR A . P eal}oystxffe Nl, Fe Cr, szCr, and Fe- C whlch were prepareda& the Massachusetts_ Instltute of Technology Metallurgy Laboratory | - Several addxtlonal tests are '.bexng made to determ1ne the relat1ve corroslon reslstance of columblum,f _molybdenum, and tantalum in molten l1th1um.‘ | o ' S | Sodzum The corrOSion data of'Pure metals and al loys in'eodifie hafe eoto' nbeen S0 fully evaiuated as those of pure metals and alloys in: lead Neverthe- eelessq-a small number of data are available and are presented in Table 11 3. "eSodlum attack dlffers from that of lead and Ilthlum in that attack 1s re=_ 'olatlvely shallow and gamma to- alpha transformatlons at the surfaees of stain- | *iess steels are not observed,. Some of the corr031on phenomena observed in sodlum are:- caused by im- pur1t1es 1n the sodlum rather than the sodium 1tself Carbldes have been: detected on the surfaces of the streng carblde forming metals molybdeeum ‘tan- talum, tltanxum,_and tungsten._A n1cke1 rlch layer on molybdenum (Fig. 11, 10a)e f has been retested in. sodium at IOOG°C u31ng the speolal testing teohnlque in "whlch such ‘mass transfer is not possxble.fi The datas however, are not yetl avaliable, but it is qulte obv1ous that molybdenum has excellent cCOrrosion: re31stance to sod1um.; Gther metals lzsteé in Tab}e ll 3 also show good corrosion resxstance to ' sodlum at’ 1600°C;_ However, there 1s evidence of slight susceptlblllty to. Intergranular attack in the case of low-carbon 304 steznless steel (18 Cr, 8 Ni) and of a sllght solubxllty of tungsten in sod1um._ Inconel X is inferior ‘to xnconel in corr051on as‘xs ev1dent in F;g.»ll.lfib.; Bwsmuth Uranzum Several metels haVe been exposed to bzsmuth contalnxng 2 atom % uranium at 1000 C.. These are listed in Table 11. 4 togeuher with testxng txme and other pertlnent 1nformatxonol Figures 11.1la and & 111fletraue' the nature of corrosive attack on type 316 stainless steel and 1nconel " The severlty of 1ntergrenular attack is gquite pronounced for these alioys Zirco4 nium and titanium were very severely attacked, whereas tungsten,-tantaifim, -molybdenum,.and columbium were-relatively'unattacked.f Additional materials are being tested, and will be reported in the future, - DIFFUSION BARRIER AS A CORROSION INHIBITOR - Corrosion protection may be accompliehed by either of two methods: (1) an environment could be selected such that the free energy change fozr the 197 - C# » ce0eRs ® TIZILE Y ™ 3 aenvbe . seaso® EYY S L Y (] L ') sosone L) o L X} eowond ° e @ - 'S foup L .- e e @ PPrEY) @ F] satew o * eoeds - L4 » LI s @ ABAQD S B se4080% e ® LI R R R oo Be @ e 'S * *99000 " E X T T I TABLE 11.3 - Static Corrosion DPata for Tests Made in Nickel_tapsu}eg-;n_Sédium}ati1666°c 86T ARBITRARY | THICKNESS At o o COMPARATIVE | CHANGE _ o CHANGE - = METAL RATING * (in.) REMARKS - (mg) 400-hr TESTS s Ni. 10 Noné No-evidenca-ok attack ;*jD?;_ 7J'g; Co 9 0:008 Slnghn attack; part1cles or voids. at depth oi 0 0068 in. | _fi4175 T : Mo ? Mt" Spectregraphxc evzdence of Ni and C on netal surface (see Fig ll 10a) _ ;?5: : fff; Ta 9 None X~:szf§z££racbxon study 1ndicates TaC,unknown film 0 002 in. th1ck;yxrr§gular Ti 9 None X-ray and spectrographzc evidence of TxC on’ surface to depth of 0.004 in. §24¢{;;ffiEH ” W 8 -0 00,1 ) X-ray evxdence ef wC on nurface’ . : }19$0? ? ?;a:¥ I#¢°“°l : 9 None ,D@c“”b“rlz“tlfln to depth of 0,001 in.; no 1nter3ranular attack (Fig. 11. Ifib) . 4160f ' ;-i fi I:Inéonelvx . 8 None 0! 6006 in. decarburization; 0:002 1n3 1ntergranular.a§tack_ F .'§idd'_fi”;?; } : ;34155 8 +0:010° Decarburization and unidéntified phase_fienéafih!éhiffiée;:fiouintetgranular”aptatk_ ” -%50Q;eii;';;; 40-hr TESTS 304 SS (0.006% C) 7 None .Precxpxmate in grain boundnry and along cxyatallographnc planes, corrosxon nofied co ”1;2§0,:ff fi”:5 : a2 depth of 0.003 1 S | i347 S8 9 -0.003 E 050003 in. decarburization; 0.0005 ifi» zntargranular attack _ _fGOTTT ':h : . 310 S8 8 None Voids end decarburization to depth of 0:002 in. e 430 SS 8 0.002 in. \intergranular attack; 0, 001 in. decarhuwiznfiion; h;nu§ifomfi'sigm;wlike ' 3~1&5 ;'{fi:... pr661p1tate to a depth of 0.025 in. : o S I : R §~§.446 S8 9 ~0.001 No ewxdehce.of attack; sigma-like particles nopéd;;n sé;ins and in grain héufldari§3 _ ;$6{ ;£f3 _ *10 = S = poor, 1 = &1ssolved Mt = mass transfer probable. | B 13¢IsF11i‘fiu-' : __. _H:: 9 - -; . 4 ; © Titamiwm - L1 )4 . Armco ivem .} . 91 o4 o o Caiumbium‘.':-" S8 "' Molybdenum - Tantalum -9 fThngsten - | N g 4 W .Inconel £ o -3 leewems SEXEE L 8 @ 8 R T T Y A Paanst .:*1@,; / . 5, : Statie Corrosion Data mmes obtginga;féogaTeétgflaaaéai@;ggifilxgéflgpfiaxbiggjatfigefl°c withfafiésaaifoiayf PR | TIME . MATERIAL (br) :xfggx;Iggflf;Ficgr;QN S | REMARKS C s e Bfizflg., Zircaaiufi.:' _f.f- 2 . 4 -Tai UOZ’ B1203, E ¥ UOzs Bx‘ Gst Lol ':NQ data-_ . No dé;a . 316;§té£nléss steel] ?1. : Nb.datfi _? daca 1 Uneven sarface thh dark brxttle noncontlnuous flim less than 0 0005 1n o Very brittle and peroua e ';-'Intergranular penetratlon to a depth of 0 009 1n "Uneven surface 1nd1cat1ng some corrosxon attack_ no diffnsiéhgap?dréflt_f_f;flx,:* : S '__Sampie dxssolved 1n Bath : _Sample nearly dxssalved no metallagraph1c data Very eh1n surface deposxt (nonadherent), no othar corrosxon Fa1nt suggestlon ef aurface fxlm and no change in- structuwe, some graln growth:fijfff - no: other V1$1hle corrosxon ,Very-thxn fxlm an‘aurface Very unlform surfac@, na v1slble £ilm or. corzosxon observed f; complete leaka & hy penetratlon af the Bx»U alloy alan Tube aamilc tested oundarxes throughout samp e, very porous and brittle (Fig. 11 llb? grain BeO): arucxbie lesked dur1ng teat penetratlon thraugh the % 1n uthlck sampley_f . (Fag 11 11a) "Bafi cruclble le&ked durxng test, sample stlli had axtremely heavy 1ntexgrannlnri } _ penetratlon to a depth of 0 025 .’excelient 5 = poer,- 1‘ dxssolved ' **Xaray llnes cotrespond to: thuse of arxglnal 2 atomxc % Hw&QB -a@émi¢ %.Eiiallny{.f fihi@k?i’ff; 'corrodlng reactlon is posltlve, or (2) a barrler lem between the reactants 'cculd be developed whlch woulfi reduce d1ffus1on ta the extent that the reu __actxon rate is satlsfactorlly m1n1m1zed.: The latter technlque 1s the one most w1dely practlced since the former usually demands condltlons and materzais"' =ffithat are generally dlffxcult and 1mpractlcal to obtaxn._, _"H' L A | | The possxble react1ons* between a pure solld metal muia WHE ilquld metal' are un1qué1y descrlbed in an equlllbrzum diagram of the bxnary system.: How“e' '*QVer,'such 1nformat1cfi is either nouev1utent for: many systems of 1nterest or “is 1naccurate to the po1nt of belng ml.ffladlng. Much work.remalns to be done”' ”-&n uhls f1eifi The ideal s1tuat10n woa]d be a system in which an 1ntermeta111c 7phase coex1ets thh relat1vely ?ure companefiuv at the deszred temperature and__ fthrough which dlffuslon Tates are flnglglblY smdli The formatxon of such ane: ',1ntermetalllc lem on ‘the spec1men surface would constltute an excellent H“eeprotectlve Iayer._ The solld and I1qu1d metals wauld then be accompanled by - 7'very little corr031on (1£ the defznltlon of thls term could be hraadened to; _e ' 'Inciude such a reactxon) Slace it appears unllkely that such a fortultous relatlonshlp ex1sts 1n; “the 11qu1d salzd (coolant-container) metal comblnat1on belng con51dered ‘the "_addltlon of a suxtable th1rd (or more) component cauld concelvably have a :_szmllar 1nh1b1t1ng effect. The barrier film mlght be an’ ox1ée,'n1tr1de, carblde,,or posszbly some other nonmetaillc, metalllc, or 1ntermeta111c ) .layer.t HoweVer, the adVerse effects on thermai conduct1v1ty must not be.' ?IOVeriooked Thermodynamlc data pertainlng to metal axldes suggest that Cr 205 'should '_be stable in the presence of lead at all temperatures and in the presence of sadlum abeve approximately 600°C; assuming that no compllcatlng mlxed oxides form, e.g., chromates;j Since many alloys contain chromium as an allay1ng“' element, the selective oxidation of this constituent may contribute to cor- rosion resistance and thus form a diffusion barrier.: Fortunately, this can be accdmplishe& by exposing the metal to an atmosphere of controlled oxidizing' potential for a suitable time at a selected temperature.: The adherence of the oxide to the parent metal through the temperature cycles involved 1is 0bv10usly v1tal for protectlon._ A.25 Cr«m20 Ni alloy (type 310 staxnless steel) was pretreated in this ‘manner tordevelop a Cr203 layer and subsequently was subjected to a static *This does not 1nclude reactions between grain and graln boundary or the 1qteractlons among the various crystallographlc planes. L _ . . e, N oeemew. . soanse o " (.1 ] L1 X ] LY @ . 3 .. [ L e e a8 & * [ W O SRR N 3 4 08 ©& '} [ BN L] & oy B T * & 8 8. & * o @ 30 4 : g @ P @8 W g e e ¢ @ L : FIoneny ) e 43 RO . 8. ped B ¢ 0 U8 . A 889 A *corros1on test 1n lead at 1000°C for 100 hrw, A photomlcrograph of the expoaed; surface is shown in- F1g 11 12 together with a s1mllar photomxcrograph of an “_unpretreated sp301men.; Although the depth Gf eorroslve attask was not apw___ _ptec1ably reduced 1n thls prellmlnary test it was. nevertheleas less severeww e Additicnal explo:atory';ests'are_planned_using other;allby$ and cqoiantsfij._ BYNAMIC CORROSIGK BY LIQUID &ETALS _ H W Savage,_ANP va131on ) | W D Manly, Metallurgy va131on | e Whereas the statxc corrosxon testlng of varlous materlals by llquld “_metals may serve to screen out those which are obvxously unsatisfactory, the *.f;nal acceptance of any: system is dependent upon dynamzc corrosion tests.j "These tests are more dxff:cuit to perform, howevar pr1mar11y because they' flalsa znvolVe the development of a ccmplete llquxd metal system. - The dynamlc tests are currently belng performed both in convectzon loops (whlch requ1re a8 _'mxnxmum of development but have. low flow rates) and forced c*rculatloa loops '(wh1ch requlre a lxquld metal pump and have substantxai flow ates) Ob?1ously th1s latter technlque w1ll uitxmatel? yield the de51red results the convection 10ops have been used on an 1nter1m ba51s pendlng the development of forced 7'c1rculat1ng systems., Many data ha?& been obtalned §r1nclpaliy w1th sedlum from the convertlon _ iaops.; The forced c1rculat1ng xoops (flgure elght loops) are stlll belng' : "deve1oped _ Thermal COflvectIOfl Loops — Barps (E M. Lees, ANP BLVISIOn} Thirty nlne thermal convectlon loops centaznlng sodium and a few lead- coflbaznlng loops have been pperated by the ANP Experimental Engineering Gloupa_ Data on these are giVen.in'Téble 11.5.. The decrease in operational troubles may'bé attrib- uted to 1mproved operational technique and to the 1nstailat10n of adequate p control equipment.- Operatmonal szfmcuifzes Several premature loop fallures were attrib- “uted to unsat1sfactory welds and should notbe interpreted as belng 1nd1cat1ve of poor corrosion resistance. of the alloy material. Figure 11.13 illustrates - - 29 @ @ e S aé B 008 6 488 €9 2% ¥ W% s ete e e B0 ee ae [ Dt LR B e e B8 L LR L L B O] 5 8’2 6 o .. wes IR W R N ST B SRR R @ e LR e e 6@ & e 08 620 o 009 0 0 8§ S0 ..0.8 " 0.080.69 TABLE 11 5 corrnsian ané operatiun Nates of Therua! couveatien Loepa | MATERIAL . Loop BaTH | METAL ZONE ' TEMP °F) - "1 TERMINATION B f'* - hQ¢AIIbN {TIME OF | OF TEST - SAMPLING i SAPLE GXYGEN (%) mm*sxs OF LIQUID BATH METAL[__.' FfiR OOBBOSION PRODUCTS (ppm) Spactrograph;c Anaiysxa'f. | UPPER BULB LOWER | 1BULBy-f¥?aQ-Cx_ -Ni M9-3Mn;; C6 ' ;Wfl METALLOGBAPHIC EXAMINATION 316 88 | 316 S8 | 316 sS 316 S8 Na 978 1000 625 1350 | 1350 1500 1500 1110 1345 1165 s middle of hot 1eg 12-10:50 as scheduled 'level _ 12«10 50 88 scheduled- mlnatzon scheduled because of gas- line leak above sodlum : '_».-. 11- 24- SG faalure - .flnltxal "_1-Final RN o ”-”Final. :12 -6~ 50 eatly ter» o Initial | Final - ' fin?1: Final | Final | Inivial Final | Finat Initial | 159 ;&§_; | Top cup Bottom cup: -Ihp'égp’” Thp;cup g | Bottom cup Top cup prfCup_w” Top cup’ pr'cup.f; Batton cup.- -9}092” 0.028 0.056 0.067 0.021 Bottom cup B : ;9;086_ .' S 0.020 | 0.031 0.082 | 0014 | 0i013 | 0.087 | 0,012 <30_..- bl measure the physxcal-, propert1es of lzquld metals llquld salts, llquld caustics, and structural and '"'other materlals Whlch may be utilized in the design of the azrcraft reactor. The fac111t1es of the 1aboratory at present 1nclude four Bunsen ice calofi-- rimeters, a steady state apparatus for the measurement of the thermal con- ductxv1ty of llqulds a iongltudlnal flow apparatus to determlne the thermal{" '__conduct1v1ty of solids, a fall1ng ball v1scometer9 and an apparatus for the_ measurement of den31t1es In addition to these fac111tles all of whlch are' or soon will be in nperatlcn, several other pleces of apparatus which would provzée other information, such as thermal diffusivities, or alternate methods_ for measurlng ‘the above propertles are being considered. These existing ané ‘ propcsed f30111t1es and the 1n£ormat10n obtainable therewlth are discussed. ‘Heat Cayacity (R F. Redmond and J. Lones, Reactor Technology Division). The problem of de*erminlng heat capacities has progressed to the stage that the method has been selected ‘the apparatus has been designed, constructed, ~and given a preliminary check for accuracya and the 1nvest13at10n of some materlals has been znltlated The calorimeter.chosen for measuring the necessary enthalpies was the Bunsen ice calorimeter. The enthalpy measurements made at various tempera- tures make it possible to calculate the heat capacity of the material béing' {1) Rothfus, R. R., Monrad, C. C., and Senecal, V. E., "Velacity Distribution and Friction Factur in - SBmooth Cannentrxm Annulx * Ind. Eng. Chem. 42, 2511 (1950) i o 23T .'.E‘?'s-s..'-'“:.-;:. E..E.Eu: oo: § u:a_ : : :_‘i"_:’.. T " "771nvest1gated as. a funvtlon Gf temperature In deszgnlng afi& constructlng thef 'fapparatus attent1en was glven to maklng the calorlmeter s1mpia but capable o£;3f ” i the de31red accuracy (S?) The calorlmeter 1s shown in Fxg 12 5 and the jassemb}ed apparatus in Flg 12 6 At present four snch 1nstraments are 1n use; "V By checklng the enthalpy data avallable for berylllum wxih data obtalnedfi'f " -_exper1menta11y u31ng the calorlmeterg the accuracy of the agparatus couid be: f gauged It was found t:at the enthalpy values checked w1th1n 4% __ moreg[i "=:£harangh cgmgarlson us 1?9 SVnthetic sapyhlre wxll permlt ccrrectlons to. he gf_ ” IaPPlleé:to the results_gzq~'reduc1ng the probable errcrln the enthaipy Valuesw-pV* Qurrentlyg enthh 1y_%et¢;m1natlons are belng run on =:*:'j_ sa&;um'l ' ”hydrcxxdey 316 stalnleso_eteel _and nlcxel Determ1nat10ns for zirconlum aref H;to be started socng_It has been estlmateé that the tzme necessary to deteIMIneu “fl'” “-{fthe temperatnre dependence ‘of the heat CapaCItY for a gzven materlal w111 be i¥'” ",nearly one month With the fcur calorlmeters avazlable ‘the heat capacltles; . 'f f-_: af four materlals can be determ1ned for each month of operatzonq' ?hermal CQnductlvity uf Liquids (L Basel M Tbb1as -and S K Cla1~ fil fborne,-Beact0r Techneiogy DlVISiefi} Stvady S*ate Mefhad The Deem type of”,. fapparatus for measurxng thermal conduct1v1t1es:afllqfizds was descrlbed brzefly'i ['1n the iast qaarterly reyort (OH&L 919 P 196) Slnce thaa tzme the design and fabrxcatzon of the varxcas cempenents of such an apyaratus have baen ¢am~"_"' _ ;pEeted ané the assemhiy af the apparatus and auxxllary systems is: ayprexx« :,mately 90% compiete ‘An assambly draW1ng of the main. apparatus w1thout :t_-;” ”'5[flaux111ary eiectrlcal afld p1pzng systems is presented in Fzg 12.7. A mare' | fdetaxled descrlptlon of the apparatus and methcd w111 be presented in a forthu: '”comang repcrt _ Some d;fflcuity was encountered 1n arc- weldlng thermocoupiesv " to the apparatus but this trouble has been eilmlnated _ At presen the»er" :thermflecuples are belng calibrated. On cempletlan 0f the apyaraxns as&embly" | “and after cailbrazlon af the thermocouples 'the apparatus will be shecked out o with sadlum as the spec;men llquxd sxnce the exastlng data for thermal con- fductxvxty ef sodlnm are xn samewhat gand agreementg | Trans;ent Me*hod Scme thcught has been glven ba 2 traasxent meth@d for. .'determxnzng tharmal diffusivities of lzqu;dsfi It has been &@c1ded to attack- "thg convectlfln pvablem by using aultable convection shields rather than by; I attemptxng te maznsazn perfegtiy uniform longxtudznal heat flow. A schematic dzagram @f tha prcpesed apparatua is prefientad in Flg 12 8, It cona;gts_' .._,'&. _Q‘q ; g ; ,.:1 esilee ® 08d 2 :,c._:,.,.é._‘- I B R B PR SR DR . eoe e g e el @ e 9T e, 8. 2 e ie Tie i & 8RB eiene e e e " PO R ) . W R, SR FEER. S, SO : : R DR TN SR D TR e ol -."-_“}‘_'_' % 1 < g o ) = o o - < & i o L o = 2 = o L w it O “ < e O x w b < i T # o L @ - © b sge . s ed 289000 o g s . - opa0 ] ° uucuusssma g S X--iO DWG NO IOSSZ T T T v TN . - LTI I 777 77 77 N 777 77 777777 7T 777777 VA7 77777777777 I ITIT N RuO AR aAAaAL - [ il - =3 ::' \ \\ ,0 _/ SO O BAAANSIASISANIIRN iy mm NN\, Tmhm o ] VAELSY A a»{}B T TT LT ARG L A A SR LV A N A & T ® \ \ a N NS \ \ B N W R L LU SR SR B AU R R LR S W [EEEIRPR SRR R \\\\\\\\\\\‘5 2222 vtk H%l ’//1 L S 5 & _1' M c L e Loy (& [mn 3| ‘ Dlu'un T o) Hjfli it NOTE: aiuussas REFER TO DETAIL SKETCHES b F’iGURE _12,7 THERMAL CQNDUCTNITY AF’PARATUS : STEADY STATE APPARATUS FOFE LlQiHDS o8 sl alwl iw i eelllee 24l- e awa Fa SRR L T U R R e T T L I R bom e U e e e e B A R i R A . L e e el w as .o Le Lt Ll T ------- ....... INSULATION— | convecTion SHIE% . —CENTER THERMOCOUPLE | P mmnaidaiiraing L L LA FBGURE 12. 8 | PROPOSED AWARATUS FOR MEASUR&NG THERMAL DEFFUSIVBTY OF LIQU!DS oo INSULATION |~—FuRNACE 'essentlally of a long eyl1ndr1cal Eube 'w1th th1n dlsks spaced about % in. "apart along the length of the tuhe One thermocouple w111 be located at the e l axxs of the cyixnders_the other near the wall It has been de01ded to make noe'“ 'attempt to malntaln e1ther constant wall temperature or constant flux at the boundaryg but rather to measure the wail temperature as a functlon of time, The thermal dlffusxv1ty can be obtalned us1ng the solutlon for the 1nf1n1tee ~'cyl1nder9_surface at ¢(T) -1n1t1al temperature constant <2) De31gn of apparatus of the type descrlbed above w1ll soon. be started : Thermal Conductivity ef Solids (M Toblas, Reacter Technology D1v15103) -'Longafudznal Flow Method. Tt is des1red to have avallable apparatus capable of measurlng the thermal conduct1v1t1es of sollds to accuraC1es greater ‘than eilfi% A partlal answer to this need seems tohavebeenibundlntheapparatus of :_Kztzes and Hull:ngs ‘which has been modlfled for operatlon at higher tempera- :turesc. Prel1m1nary tests are belng made to evaluate the maxzmum error to be '_expected from the equ1pment3 using substances of kncwn thermal conduct1v1tye ;esuch as Armco 1ronj coppery ~and alumlnum Br1efly,.the apparatus 1s constructed as follows (see F1g 12 9) The fspecxmen of the materlal under 1nvestlgat1on is in the form of a cyllnder 5 cm ~in length and 2 cm in &1ameter It is placed between a heat saurce and a heat %sxnk whzch are. 1mpressed upon 1ts two flat surfaces The heat source is a ‘copper block weund with nichrome wire embedded in Sauereisen cement. The - copper bleck is a cylinder sllghtly 1arger in ‘diameter then the specimen and has a small shoulder in the contact face enabllng close flttlng to the -speci- men: The heat sink 13 ‘a water- cooleé copper hlock likewise having a shoulder fer seating the specimen flrmly The essemhly of the heaters'spe01men, and cooler is mounted inside a containing cyllnder about 5 in. in diameter. The ecooler is mounted on a lava block, and the three cemponents are presseé to- gether by a spring whe load is communicated to the top of the heater through a:la?a dlak ' e R : ST S o S After assembly in the contalnzng cyllnder, the heater, specimen, and | cooler are buried in Sil-0-Cel powder. Longxtudxnal heat flow is further ensured by means of an electric guard heater placed on the outside of the ~containing cylinder. Two thermocouples are placed in the 511-0-Cel at the (2) Ca.z'slawfl H 8., and Jaeger, J. C Conductwn of Heat in Sol:ds, P 176, Oxford University Press, New York, 1947. 243 YT NN EX TR Y] R S FEEIES L 2 % XXX N *soa e * ‘asevee woue UNCLASSIFIED '15 ' 1-Guide Screw for Spring : 11-Copper Water Cooler ~ 2-Spring Retainer - 12-Lava Block " 3-Inlet Thermometer 13-Outlet Thermometer Well S 4-Retainer Spring 14-Return Line to Const. Temp. Bath 5-Lava Block . 15-Outlet Thermometer ‘ 6-Copper Block - 16-Brass Container (Heater not shown) 7-Insulated Inlet Thermometer Well 17-Nichrome Heater Winding - 8-Sample | ' 18-811-0-Cel Insulation 9-Centrifugal Pump . 19-Specimen Thermocouples 10-Inlet from Const. Temp. Bath 20 -Guard Heater Couples FIGURE 12.9 LONGITUDINAL FLOW THERMAL 'CONDUCGTIVITY APPARATUS FOR SOLIDS C i U e ;’%&' | _K o .'o_o 2;;4 4;'.'_'. * L] ] XX X2 ) S RN @ so0osge [ X 1 XJ (2 X2 3 R . J L L] - . Py ® a 2 # L) seR090 [T XXX e LE X R 3 ] LR ] _same’ level as. correspondlng thermccouples in the sampleg.and the power to *he guard heater is adgusted 'so as. te make the temperatures in the Sil- 0 Cel equal ‘to those in the sample -_'_] A ' '-_fl_The_thermél‘conducti?ity_bf'the'spécimen is given by:the_equation' Sz - where k 1s the thermal conduct1v1t 7-&? ‘s the temperature dlfference measured by the two thermncouples in the spaavrnr 4 is the area of the c1rcular CIOSSQ .';mectlfifl of tne spec1men L is the gpsc:ren lkngth *and Q is the heat flow from H‘the heaner through the sampie 1ntc fhe wate" coelera Q@ is determined by fmeasurlng the. flow rate and rise 1in temperature of the coollng wateru The -coolant 1a'supplled from a cofistant temperature bath by means of a small:-'. :”u"Eastern" centrlfugal pump through 1nsulated % ln, copper tublng The rate ofjw . flow is measured by welghlng a sample of water'over a known tlme 1nterval as 1£ returns to the bath. The rise in- temperature is measured by calibrated " _mercury 1n glaSa thermfimeters placed in large wells as close as p0331ble to' '-the 1nlet and fiutlet of the csgpe* coalera ?he power supyly for the heatlng elements and the pump is a-c¢ and is manually adjusted by means of varlacse_ The progress of the test is recorded "on a Leeds an& Northrup Mlcromax mult1p01n* instrument, while a Bublcon portable pfe01slon pot entlometer is used to measure sample thermoccup}e emfs ~iwhen a steady state has ‘been reached Chromel alumel thermocouples are used_ ;throughou*: Sample thermocouples are installed generally by boring smali "._prec1sely spaced holes in the sample and arcxng ‘them into place by means of a 'condenser-dlscharge devzce; or, 1in the case of nonmetaillc samples, by simply :,burying'them.ih'the.fiolés with cement. So far, no 81ngle'method of thermofi .'Vfioaple installation has shown .marked superiority, but satisfactory results .- have been achleved without special effort.. ' | : _flThe_apparatus measured the ¢onductivitiés'of.28 aluminum and Armco iron wi£h an'é?eragé'err6r of only 5% over the temperature range 100 to 400°C. .;However, théfi;emperature drops4in the sample were large, particularly at higher temperatures. Thus, when the average.pemperatuie in a sample of Armco iron was about'590°F the temperature difference between the two thermocouples was 268°F. For substances whose temperature-conductivity variation is ‘markedly nonlinear, serious errors may be introduced. 245 N Eiss 08 e3m e 8 ° e ks & PR G 0% I S R R SR S o 0. & A [N 3 BN L IaE 3 R B S S (4 L e & % ® s 00 o se L S LI s g () L4 e &8 @@ e & . a @8 5 ® @ e .® ° ¢ e & ® 0 & 20 ey ) e8P ¢ & 908 se o & a 908 o® RddidZ'Fiow:Metfiod An apparatus has been de51gned in whlch the flow of_' : heat is to be from the center of a hollow cyllnder outward " The appavatus is a modlflcatxon ‘of that used elsewhere (3 Constructlon is belng 1n1t1ated | It is hoped that the dlfflcultles caused by the large temperature gradlents mentloned above may be ellmlnated Beta;ls of.the design will be_glven in a 3 future reportg,-_:-” Viscosity of quuids (S I Kaplangifleactor Technology D1v151on) Thea falllng hall apparatus for measurement of v1sc031ty of heat-transfer lzqu1ds-' 3 at hlgh temperatures which was &escrlbed in the last quarterly repert (OBNLwTI ;919 p' 198) is nearlng completlon° The furnace and power supply have been :mounted and constructlon of the vxsc031ty tube eomponentsgllncludlng the '.magnetle valvey-ls f1nlshed Flnal assemblys 1nclud1ngthe:detector mechan1sm '15 expected to. start durlng the next month The desxgn of a melt tank whlch w111 permlt f11ter1ng of llqulds and'i chargxng 1nto ‘the v1seometer and other physxcal propertles anlts as requzred-_ has been comyleted and is now be1ng fabrlcateé | _ For the testlng'of liqnlds havxng spec1f1c grav1ty greater than 8, % in. ; 'bearlng balls p}ugged with tantalum will be used. The dropping mechanzsm con31st1ng of a- solld stainless: steel come attached to a movable rod, ‘will be *c_remove& to aecommodate these larger ‘balls. A supply of these balls is belng_' - made up for 1n1t1al callbratlon of the unit u31ng molten blsmuth | Brookfleld Englneerlng Laboratorzesg of Stoughton, Mass. , has been con- "sulted regardlng the constructzon of a rotatlonal vxscometer for llqulds and f~pre11m1nary 1nvest1gat10ns indicate that such a unit may prove feasible. " Density of Liquids (S I, Kaplan Beactor Technology Division). An apyaratus for the measurement of densxtles has ‘been designed and is now being fabricated. The des;gn of this apparatus 1s based on application of the eArehime&es-prineiple; igeag weighing a suspended bob in the liquid (see Fig. 12.16). An analytical balance has been adapted for this purpose. This unit w111 be chargeé from the same melt tank used for the vxscemetera :;_ (3) Powell ‘R. W “Survey of Existing Data on Thermal and Electrlcal Condfict1v1t1es of Irons and Steels” Iron and Steel Institute Special Report Series No. 24, Second Repoxt of Alloy Steels Comm:lttee, p. 253, London, 1939 - R 'Y ) e e 2 o ] oa 0% ‘88e 0. 400 08 R B o & @ [0 N ) L R TR FEREE TR ST Say ) I LA 7 SN [ LR . 8. Lo 08 e AR 0.8 R N ] ® . o530 el ek e e @ i RE e e e @ & @ g @ A SRR I ) o o i WY BN ¥ %) ® [EY RN S 3§ 2 se & o oe e o v e THERGCOU?LE wd - §a) wumwv N ‘i"!‘l‘.l!..i:‘.d I{!“.“"} ] L R \ - ZiNnserT |} sAszNLET; g . ._\Qccc__ _ | "AHWMHHHQ h'/. TN - B ¢ ] 1gbuna HOLE DRILLED THROUGH PAN AND BASE ~.mmmwv\ Aam% _ TUBE FURNACE 12.10 DENSITY APPARATUS - . ....QO.'. ° oo.a R :09. QO | MGLTEN FLfiOBIBE ELECTRICAL”C@&DUC?IVITY MEASURE’EEK?S | F. J Sheehang ANP D1V181on The electrlcal conduct1v1ty of the eutectlc b1nary salt 26 mole % UF o 74 mole % NaF. was determined from 1112 to 1700°F. The conductivity flgures obtained with direct current were checked with 60- cycle alternating current. No szgnlflcant dlfference was noted. The ‘values obtained are presented in a plot of res;st1v1ty vs. temperature (Fzg 12.13). These values are estimated _tolmaaccurata to only about 20% and apply only m:the particular mixture which was not vacuum'melted There was con31derable corrosion of the quartz tube in whlch the fluorzde was contalned N B B _'_ G ae sce e % @« . se 68 o dee & ohe 8w - LRl e & a. 6 @ * e e Ex o & . 88 e 8 . BB B Y L] L] 8 0. @& G o L0 & S0 9.9 - LA B ) ® - R 8 1] L IR AR T A e & B30 ) . 8@ o 8e e e 8. 9 8. LI (0 e L3 ] s00 ® a0 . 8. 8 Be od B0 008 Be CmessTviTy (Q/zend 3 3 fi e §:: % -4 4 ~ 2 -uoo’ o0 300 . 1400 ssoo .00 moo L | TEMPERATURE ("F) S e F!GURE 12 13 CONBUCTW!TY OF THE NQF‘*"‘UF«Q FLUORIDE EUT ECTIC 13, COMPONENTS OF LIQUID-VETAL SYSTEMS - The development of pumps, heat exchangerss and other accesscry plumblngg_" Jg'1s belng undertaken to assure the ex1stence of & satlsfactory l;quld metal-u' system for eperatlon with the ABE In addltloni ua ¥ of these research data1” . i whlch are not avallabie are necessary to. the satxexa torv desxgntxfany liguid- metal reactor.e Accordlngly, the Experlmental Z:ghbeer*ng Group of the ANP _ jDiv131on is eugdged in a program of testlng and Efmlcr1ng the varlous com« 'ee_,penents of quuid metal systems requzred by a pcwer pr Jduclng reactor,_ Thls_: program 1nc1udes researflh on pumpsi,bearlngs,;seais, flowmeters,ilnsulatlon S flanges9 heat exchangersi and related mat“ers;-whlch are reported heres_; ' PU&PS ¥. G Cobb ANP D1v1s:.on | The verlety of aVaxlable pumps permlts con51derable freedOm 1n the ch01ce- -_ of a pump for any partxcular application. For aircraft use an axial- flow pump appears to be preferred because of the 1arge pumping volume and small_ *“pump size and welgh* requlrement but; a centrifugal pump is belng considered efor the ARE. Both of these types of pumpsg elec*romagnetlc pumpsgzand others,e _ e'wxll be used on var1ous experlmental fiuld 01rcu1tsq B Electremagnetic Pump fnr Figurefizight Laefis. The eleetromagnetlc pump 'de51gned and built for use in the flgure eight loops (see Secs 11 for a - discussion ef ‘the operat1en of these 1oops) has thus far been used exclus1vely for . operatzon of the loop. Neither a calibrated flowmeter nor pressure taps have been developed for accurate determination of operating characteristics., Initial Qperation‘showed that . the length of the electrode between the wall of the cell and the joint with the copper secondary connector must be held to an absolute minimum to prevent, excessive heating with resultant loosening of the 'uclamped Joxnt, The copper~-to~—stainless steel weld joint previously reported was not entirely satisfactory owing to thermal cracking of the cell wall. Substztutlon of the nickel electrode eliminated the weld problem but 1ntro- : duced a hlgher resistance., 40 oes e ®° ° v we @ ses 8 el s % o 8 - ¢ 8 @ PO PO BT 4 oy ¢ @ w0 . 8 @ R % e ee. € &8 s o8 8 . W L@ *De L ] 07_0_: ®, !_. 0.‘ LI A » B W E S L) & &9 & @ a. 2h . Tas e @06 B B Gr &S . 8 & . ¢ @06 80 Another dlfflculty encountered 1n the operatlon of the electromagnetlc' '-_pump is the bulld up of sodlum, whlch dep051ts in the 1nlet sec ion of thé- pump cell (Flgs 13 1), Gne effect of this depOSIt appears to. be nonunlform_;z | 'fluld flowa Thls effect is not ser1ous and po:s1bly this bu‘id up of sodzum _ in the pump - cell must be tolerated 31nce other groups have reported 31m11ar.57* f‘unexplalned dszzcultless_ ' ' - “ o PRSI S : An orlflce plate w1th sodlum manometer attached has been fabrlcate& andf . callbrated w1th water, 'and a pressu“ , ap plate for 1nsert10n hetween fianges f' ' 'has been prepared,; As soon as operatlng experlence 1n a small forced-circu- ‘iation loop has been acqulred wzth these devzces, they w1ll be 1nserted in the = pump connectlons to study 1ts operat1ng characterlstlcs. _: Centrlfugal Pump for Fignre Elght Loops,i The vertzcal shaft sump Lype rcentr1fuga1 pump (Flg. 13 2) is: under fabrzcaylone Thls assembly utilizes a packed gas seal w;th llquld 1evel malntenance by automatlc control of gas ' fiffeed,_ Salenoid valves 0perated by a system of probes regulate the gas feed. " The seal is cooled 1nternally by a coolant such as kerosene c1rcu1ated through _ - the squzrt tube and hollow- shaft arrangement. The seal housing is also. cocled' ~rby cxrculatlon of the coolant through the passages around it. Make- -up gas to o provzde for the seal 1eakage is fed into the labyrlnth below the seal in arder | -_that the gas leak;ng thraugh the seal will Carry a m1n1mum of sedlum vapor,4 Thls pump has adeslgn capa01ty sultable for operatlon in the flgure elght “loops in place of the electrsmagnetlc pump now being utilized. It will be -Ldrlven by an adgustable -speed motor and put through a series of water tests _ - before be1ng used on llquld metals. The use of a cooled mechanzcal seal has also been contemplatedgpand a means for its 1nse;t1on with comparative ease . Jhaslfieen'préfiided for;._ . e S e R BR S 7-Gentri£ugai Pump for ARE, A Centrlfugal pump of slmllar arrangement is also belng ‘studied for use in the ARE; two pumps operatlng in parallel are “contemplated. By mountlng the pumps at the top of the llquld metal cxrcu- létionféystem with flow vertically downward, the shaft seal will be subjected ‘to the miaimfim-system pressure, and the pump liquid-level control can be utxlzzed as &he systemwlevel contrel as well. The ratxng of such pumps for a typlcal de31gn of 350° F across the reactor would be 120 gpm each at 15 psi developed head of sodium at 1100°F. If driven at a sp&ed of 2300 rpm, which gives a speclfzc speed of 1506, the impeller ’._254. ._ . .- ¢ ,.&4 X -1 8 L5 X ] [ 3 & o, o9 P8 & ... & ¢80 ._. @ S0 e @8 e e e @ @ el e ¢ e ele e e W P AN JETES o 06 @ 86 8 3 o8 @ e e [ L SO e e .oe 8 8.0 @ P R e Sl ER P O S SN 0. Rae 6 .60 6 .8 068 1] [ LA L LA0 1L . UNGLASSIFIED " Y-12 DWG. NO. DSK~-15059 \ P W IIITITIILILIII LI LIELIITI LT mw Wm. o _=fiw¢ixmmmwfi Frb vivle & CIELITTTIIIFAIL L TE TN A 13.2 GENTRIFUGAL PUMP ASSEMBLY 256 FIGURE @- P L "..‘O. Lo B SEE R R : caeo6e T e B P . . - odeeed e Cessave LN 1 L . Ce 5 wouid be approx1mate1y 5. 3 in, in dlametere The powef requlred would be'“ 1.5 hp per pump thh an assumed efflclene} of ?0% A more conserVatlve des1gn ' . for operatlon at approx1mately 1750 rpm and uszng an 1mpe11er 6- 13/16 in, '1n: _dzameter gives a SPECIflc speed of 1289 and approxxmately as good an - effze clency. A '_A»c and d ¢ elefitromagnetxc pumps are bezng studledznsp0531b e alterna*es Vto mechanxcal pumps for the ARE opevaflenaz Howeverg 1t appears at presen»l that mechanlcal pumps for ‘this application will requxre IEoS development 'gwork besldes belng'mnre compact, flex1b1eg‘and efflClent and p0351b1y moxs _ _dependab1e.,_: "' ; anp for 23@ &egawatt Aircraft Reactnr. Caiculat1ons on iarge pumps 'fqr a 200»megawatt reacmor indicate that centr1fugal pumps would be too bulky f ‘“7_for this appllcatlon and that elther ‘axiale or mlxed flow types (w:th specific '7fi speed parameter of 4000 to 6000) would be more. satlsfactory. One such pumpn' 'd&slgn study 1nvolved six. ax1al flow pumpsg-each handling 2677 gpm, each developlng 80 p51a and each requ1r1ng 139 hp (tctal of 834 hp at 90% effi- 81ency) Each Totor was to be 6 in, in diameter and run at 6800 rpm, with S a specxflc speed of 6000a " This desxgn was very compact but it requlred a hlgh speed of revoiut1on and a 1arge pump xntake pressure of about 56 psia to _suppress cav1*at19n,i A more conservathe deslgns for 2677 gpm at 40 psza -wzth a: mlxed ficw rotor of 4000 specific speed, requlred a 9.4-in. rotor operatlng at 2706 rpmg with intake pressure of 16 psia. A‘fuller'set'of-data'_._ o owill appear in a fcrbhcomlng report., _0anned BotOr Fflm§ (A, R Frlthsen andh& Rlchardsong Reactor Technelogy Di#ision)' A sealless pumplng system ‘which utilizes hydrodynamlc journal bearzngs was de51gned fabr:cated and tested. This pumping system consists of two "canned™ rotors with a pump sandwzched between them. The "canned" rotors are each part of a ¥-hp motor modified so that machined "cans” have ing a wali thickness of about 10 mils fit snugly inside the motor stators The motor armatures were machined so as to fit inside the "cans™ with a clearance of 25 mils, thus having the same clearance between "can”'walland armature as ex:sted before modification in the alr gap. The pump was made by modifying two Aurora turbine type pumps so that they conld be flanged together in such a way that their suctions and dischargesr *Earlie? spacificatibn data for the ANP pump are given in Sec. 1. 257 e eeed - ‘seevnes 2490008 (XYY Y] T N EYX LN L e Y [T NE K] WO“ld OWGS& each other-_ ‘This was mecessary in order to keep radial loads to “a mlnlmum,_' s o ' o - R T T 5 Flgure 13 3 is a schematlc draw1ng showlng the relatzon of the varlous."' \components in the pumplng system to each other and the manner by whlch fluld is re81rculated through the "canned"rmotorss Thxs reczrculatmen of fluxd 1sf _necessary sxnce the operatlon of" the pump is dependent on the shaft. "floatlng“‘” 1n the fluld When in operatlonj the armatures functlon a's Jaurnal bearlfigsif _and the; cans“ act as | bearlng sleevese= At the ‘same tlme, the hydrodynamlcv' forces actlng on the 51des of the lmpellers make the 1mpe11ers funct1on as ' :thrust bearlngs@: Forces whlch tend to move the armature ot of the center of " -1ts magnetxc field are at least partlally counteracted by the magnetlc forces,'v .Thls pumplng system thus has a combination of hydroéynamlc and magnetlc thrust o bearlngs. AR Althcugh no test runs were of a duratzon Ionger than 3 hr the total operatlng tlme was about 40 hr with no notlceahle wear on any part except on 'j the lower face. of the bottom 1mpel£er,f Thzs ‘can “be explalned by the fact; " __that ‘when not in operationm, the shaft rests on the bottom impeller. The new ' hydrau11c bear1ngs descrlbed in the prevxous quarterly Teport (OBNL 919, 'pa 1?9) were. tested however, results were not entlrely satlsfactory because 'clearances between the shaft and the hearlng couid not be maintained, “which resulted in selengg- The bear1ngs have been reé651gne& to eliminate this ".troableg and new bearlngs are now bezng fabrlcateds Tflrbiue Pump° A study is’ be1ng made and experzments are ta be 1n1t1ated to determxfie desxgn parameters for turbine pumps sa that a high- head low-flow pressure source may be avallable w1thout the use of excessxve speeds or sxzesag Fump Sealsal A theeretlcal study of mechanlcal seals is in progress to ald in choosxng seal materlals and deszgnsg It appears that fluid-lubricated 'seals &epend mainly on caplllary action of the 11qu1d in the seal to maintain a lnbrlcatlng fllm‘when‘nexther material in the seal possesses lubricating qualitites, and they thus require a very fine finish and very flat sealing surfaces, within a few microinches of true flatness. Gas seals, on the other hand, rely on "hcundary” 1ubrication of the seal surfaces by oxide films or 'adsorbedlgaS‘films,,reqniring the use of graphite or other material having a low boundary friction coefficient and high melting point as one of the sealing surfaces. P . oo YR XX} (XX 23] e N segbe Ly e (R ELX 2] cesned B ] I T D e UNCLASS!F!EQ Motor & Can Shaft, Rotor & Impeller mount . Pump housing T . ,.L ‘- e /) FEGURE 133 SEALLESS PUMPING SYSTEM | (UTiL!ZiNG HYDRODYNAMEG BEARI NGS) A sodlum or Na K lubrlcated mechanlcal seal is also another posslblllty'v be1ng glven con51derat1ong-:”‘ An alternate seal de51gn under con81derat1en uses an 0il seal around the' ~shafy, Wlth the 1atter enterlng the pump. case from. belowb. A small centrlfugai_' separator rctor 1s mounted between the seal and’ the pump rotor.- A separat1on'__. ‘surface is maznta1ned between the sod1um and a small charge of ozl above the seal by the centrlfuglng actlon of *hxs rotoru An 011 coollng Jacket ma1nta1ns- 'a temperature of approxzmately 250 F in: the v1c1n1ty of the 011 seal and the - centrxfuge rotor to. keep sodlum melted but at a reasonable temperaturea_ Pre- ':l1m1nary tests 1ndlcate that DC 550 sxlxcone oil is qulte stab}e in contact. 'w1th sodlum at 400° F fnr extended perlods.: Other poss1b1e seai llqa1ds are - be1ng 1nvest1gated ' BEARING TESTEB BESIGN W B McDonald ANP D1v1510n A bearlng tester for testlng hydrostatlc bearlngs operatlng in’ llquld metals over a wide range of temperatures and unit bearlng pressures has been d931gned (Fig;l% 4) It conslsts of .”IQj A pr301310n splndle to accommodate an 1nterchangeable sleeve3 N fL/IH« 1n dlameter which 31mulates a pumy shaft, ;_2,f”A sealed pot for contalnlng lzquld metals whlch'also provideS'f = rigid mounting for the hydrostatic bearlng.' Entry is provided ~ for pressurized liguid metal over a wide range of pressures. - The pot is mounted on two universal jeints which permit the _5 hydrdstatichearing td-aiign-itvelf’with the shafe, .,_ ‘3;'.Externa1 means of applylng variable load to the bearlng 4, Pump for pressurlzang the hydrostatlc bearlng, : § 'S.f'Varzableflspeed motor for driving the splndlesa 6. Sump and expansion tanks. Q?ERATIQN GF SEAL-TESTING DEVICE W. B. McDonald, ANP Division Two seal-testing units have been constructed and are set up., One has been operating in water with good results., Seal materials which have been R R R - g . b . - R . . : e. .'.'.Q. *8 R . @ Py 28 0 ves. o oo e8P - E. e f k P 6. 0 e L R ® L3 A Do . e o @ : o. :. : P e e ee _q.? :: e e e e i eve ° bR -AED- SN AT G- e @ . o 0 @i W @@ R L . end e€ et PRI S X L L I B . b 8% = \\\\\\\\\ \.\_\_\E\\\\\ N S UNGLASSIFIED N Y12 DWG NO. DSK-15039 N n@%f \vieooooxx&@f__ P IO 7 .\\\\\\.\.fi g&\\\ 777 \\\\ \\\ \\ \\\\ Y OLY T N ~ N0 TN T W W T W NI e 1107717 \\\\\\..\\\\\ * ¥ : ww_&g}«‘kfiw ¥ A P - ANNUANNN L SECTION AAL .TESTER ASSEMBLY tested to'dété.afef . ;1i_ Other seal materlals to be tested in varlcus comblnatlons are carhalloyg nltralloyg Scottsonxzed stainless steel, and Ihriglzed steel. Future seal Stalnless steel’ rotatlng member agalnst brass rlng. ThiéISeai 'operated satisfactorily against water, hut the brass seallng_fl face was scored excessively after 20 min‘of cperatlon, Some copper was prec1p1tated out of the brass onto the face of the ~stainless ‘steel’ rotating member. . The scoring was probably due: ‘to poor initial operat1ng technzque resultlng in excessive ‘hydraulic pressure on the back side of the rotating members_ Two- sets of these seals were. run w1th 1dent1ca1 results. | 'Staxnless steel rotatlng member agaxnst whlte cast 1ron rlng., 'jThls set sealed agalnst water dur1ng 1 hr of operat1on. However, ~a deep groove was worn in the. cast iron member ow1ng to the abeve-mentloned faulty operatlon technlque," f;“; SR EStelIlte agalnst stelllte& For this test the operat1onal tech* -~ nique was changed. Means were prov1ded for pressurlzlng both "_'chambers of the seal-testing unzt ‘i,e., water in outer: chamber ~ and air pressure on the inner chamber. Gauges were provided to check the pressures and’ throttllng valves prov1ded to adjust the. pressures. . This seal operated successfully over a 6-hr period | ~with unit contact pressnres of 48.5 psi and water pressure of.f -.50‘951-._ 1; S e e R - _ SteI11te aga1nst stell:te. Thesé'éeéis'perfofméd Satisféctdrily" against 40 psi water pressure with unit contact pressure reduced to 10.9 psx._ The ‘rubbing veioc1ty of members was 1350 ft/mln.fil Sta1nless steel rctatxng member agalnst stelllte rlng. Tth{ v seal operated successfully for 50 hr against 30 psiofwater = with unit contact pressures reduced to 3 psi. The rubblng-" veloc1ty was/1350 ft/mlnfz_. A;testxng_wzll be done in 11qu;d_metais. An electromagnetlc flowmeter was 1nsta11ed in the f1gure elght loop to thaln an 1ndlcat10n and rough estimate of the flow veloczty attained in ~ operation. ELECTROMAGNETIC FLOWMETER strength of 2600 gauss. It will be calibrated as socon as practical. 262 . e REITRER e b e SRR ® kiR - ‘ele e'e RS IR e 8 ee Lo bR T S e e e we e de % 9 LSRR S S T L IR I T S e el e ele _ . L8 i %aT see s e es mel s s 00 8589 This flowmeter utilized the large permanent magnet with a field ~ CALIBRATION LOOP The fabr1cat1on of the callbratlon loop 1s approxxmateiy 70% compiese .in.our sh0p and operat1on is p}anned by Aprli 1 | o~ FLAN&E ’EESTING DEVICE W C. Tunnell ANP D1v1510n Since 1t is desxrable ‘to use’ flanged Jo1nts in experlmental llqnld meta17 "~sequ1pment so that test sections can be quxckly and easxly installed and re- ”mGVed ‘a flange testzng device is be1ng fabrlcated to test flanges at varlous e system temperatures and pressures. It 1s planned first to test, w1th vacuumg- Frecn and water pressures ags1nst bolt pressures obtalned with torqus wrenches ’for varzous types cf API gasketss e,ga; soft 1rong nlckel and stainless 'steel before test1ng w1th 11qu1d metals atvarlous temperatures and pressurese _ STRESS-RUPTURE TESTS ' 1fW;;CesTunsell, ANP.Bivisiés; | The equlpment dsscrlbsd in the last quartsrly report (ORNL 919 P 166)' for self weldlng and stress corroszsn is xnztlaiiy to be used for streSSu_s ‘,rnpture tests. Two pieces of apparatus are bezng set up in convsctlon loeps | for llqu1d sodium tests, and another piece will be set up in a furnace with . inert-gas atmosphere on the specimen being tested. It is planned that tests of 10 and 100 hr duratlan will be runin the furnace at 1400, 1500, and 1600°F, testlng flrst 316 and 347 stainless steel and nickel. Tests in sodium will be_ for 10, 100, ‘and 1000 hr duration using the specimen shown in the above reference wzth dead weight loadlng up to 50, 000 psi stress. | Imsunarxnm Tssrzss | R. T. Schomer, ANP Division Preliminary tests were made by placing samples of various insuiabiens in a pan of burnizng cadmiim. The samples tested were Failip Cerey and Johns— Mansviilesfiigthempsraturs insulations as well as FEagle-Picher and Baldwin»flil#_ . ese PO » ee ®e @ :oi :_.oao _‘;s. ] L - . X . . : . 08 . m : e & ® :‘: : - o e 88 .._.:_e._._:_:. Sedt e e se0 B see s RISt ; 3 e A SRR . : ; Ca e "o e v o B8 se 6 & e g8 2% lead-mill slag wools., Results with all these materials apgfiaruéqcpgfagifiga;fi _ | _ 1n all cases the rate of reaction appears Tto be slow enough Nto be safe.i At th1s time a sample of lime 3111ca was aiso tested and 1t - _appeared to be worthy of further conmderaflone ?It was felt at_é o the canclusxon of the sodlum tests thaL a better method of meterznw the amoun t | of llquxd ‘metal Jetted 1nto a’ sample could be dev1snd With thls in mind a t:measurzng tube has been added between th& reserv01r gud the ampie holdlng :plate. A probe in thls measur1ng tube determlnes Lhe ameunt<3f; quld metal ’used in ‘each test.d By changlng the length of the probe, varlous amounts of,i llquld metal can be expelled. In addltlon to permxttxng better standardlzlng'*" of the amount of metal used in’ each test .thls change allows the bulk of thez: 7{11qu1d metal to be kept at a relat1ve1y low temperature while only the small-' ':amount of metal 1n the measur1ng tube is heated to the d851red temperature: ':-for the test. Thls mod1flcat1on has béen made in con;unct1on with the Heat "Transfer Sectlon of the Reactor Technology DlVlslcn. Prlor to modlfylng the 1nsu1at10n testing apparatus the remaining sodium :1n the reserva1r and the assoczated piping was heated to- 1730°F (which is above the bozllng 901nt of ‘sodium at atmospherlc pressure) This- molten metal was’ squzrted into the air and the- results obtazned were merely a stream of molten metal burnlng as it struck the air. No indications of ?aporlzatzon - were seen. HEAT EXCHANGER TESTS A P Fraass ANP DlVlslon ;In the ANP reactor des1gn bundles of falrly closely packed tubes have -been 1ncluded in a 1arge proportlon of the proposals both for fuel elements in the reacter core and for the matrix of the intermediate heat exchanger, One of the major design problems for any of these proposals has been the problem of thé'préSshre dfop'in the stream flowing outside the tubes. Three important elements cantrlbute to this pressure drop, namely, skin friction, turbulence induced by the spacers, and losses in the cross-flow region at the 1n1et and outlet., ' (N u; ame e e W esiien e ews 2 see 30 .. : : e el e . el e st es a e ils el e e e ae e 80 dip. e e “ o FEY : 0 8. ¢ 0leiie g s N A & & 3 N & o ees "B i eee s .68 38 s e O 200 08 R . © o o w 59 - O 8 = " ; o 2 2 o - I o - 5o 1 3o It was declded that the fluld ffiow and pressure characterzstlcs of a. m 3 flow test of the model shown 1n Flge 13. 5 Wthh Was made uP °f 1 3 in. GQPPerf'” _qi the tnbe bundle was avoxde& by puttlng a small effset in eVery other tuba;;1; ]f7 '-'where It bent to entfir the header,. Flgure 13 7 Can. end v1ew cf the model "*ffi:shows how thzs Was done,_ Note that offsettlng half the tubes in thzs Way alsd fhelped tn spread the ‘tubes out 1n uhe hfiader so chat weld1ng of a full scale : ‘;f: header would be facllxtated Prellmznary results frem the tests 1nd1cate that the pressure érop in _the cross flow reglon at the end ‘is qult& small Slmziarly9 the pressure IossT b:f?zn the. tube bundle, appears to be about thce the caicuiated vaiues based "~~set~up 13 bezng madlfzed to permlt operatlcn at a hzgher Beynolds number, ER Reactar Technelagy EIVISIOH to determlne the ixmltatlons Qf hydraullc radzus:]' :fi7fas the fluld pressare Icss parameter.:-'“' CKEAKING &?JB E}ES?BS&fi fiF LIQUXB fiETALS ANfi E@}IIP%EN? F L._Hzll and R,.Devenlsh ANP Dzvls1on :”ffi ' be used by ali unzts in %he X 12 plant area whlch may have to clean equ1pmentf . ccntam1nated wzth llquxd metal,sznce 5611d1fled; REPA and the A%P Dl?151on '”rest1ng upoa e graVel be& w;gh adequate drainage and supplxed w1th steam ._heat and water¢;_ "¥ Large amounts of sodium have been remcved from small items of equlpment_ hxgh flash-yoint saturated hydreearbon type oil. The residual sodxum a&herxng: to the_equlgmentzfi then removed by steaming, degreasing, and thorough rinsing, Thg,moist'steam used has produced a relatively small amount of‘fuming_and 266 ® ‘@ R I L S : LS R A g eveeve ] O T LY caevee. L] Lemenea ae ressee Teeam e . s . . 'bnndle of closely spaceé tubes cofild be read:ly 1nvest1gated through dn a1r~:i¥f v weldzng rods.‘ Restr1ct1on of the flaw 1n the cross- fiow reg1on at: the end of 'across the spacers 1s also very reasonables The ma}or 1oss, from skxn fr1€t10n3 : om the use of the hydraullc radlas of the ch&nnel betwesn the tubes, The test;f1 :'”ané the Ilterature is be1ng re examlneé bY the thearet1cal graup of the7= 3 The faciilty'mnbe used far removxng lzquxd metals from pleces<3fapparatus}_”3-"”fi -has been de51gned and 13 1n the process cf censtructzon,. Thls faczlxty wzll .:')' g;a;e Joxfltly equ1pp1ng ‘this facxllty Whlfih conszsts of 8 sheet steel barrier ._: _. . 'by dra1n1ng when 1mmersed in a heated oil bath containing low- s§ecxfxc gravzty‘z,g e enEe » o mlnur exploslons,_ Dry steam is expected to yroduce llttle ‘or mno exple510ns.-" Large items of equxpment have been cleaned by meitzng the soélum in such: a '” manner that it falls through the atmosphere for: a short éistance ané *hen 1ntof '. ’V;Eanner,_untll suff1c1ent quant1ty 1s on hand to warrant recovery or d1sposa1 Up to now the waste sodlum accumulated has been used in’ tests of cleanlng-F methods and to create cantrolled flres fortestsof flre extlnguxshzng equxpfi wzll normaily be dxscardeda-,--- Waste sadzum Wl}l be fiumped 1nt0 an abandoned quarry to be reacted w1th'h:” the large mass Of water present,_ ?ésts are underway to determzne the pract1«_ callty Of dlSPGSIHg‘Of solid sodium by hflldlng unéer the water in a perforated steel cantaxner,: It has been noted that seélam stcred under hea?? 011 will not react 1nstantaaeously when placad in water, ané the éelay should affard; :' wasta metal; This metha& wlll also ‘be used to d;spase ef other was?e materlals'"" ontaxnzng a large percentage fif metallzc sodlum@_ o SA&PLING PR@&EDHE&ES B P Lc Hzll and J Fs Sheehan, ANP ElVlSlon__i Samples of 11qu1& cooiant beang tested are taken for chemxcal analysls.'” Sadxum is sampled when in the molten state by draw1ng the metal into eVacuateé f1'” | glass tub@s. The ‘tubes consast cf two bulbs of ahout 5 ce capaCIty connected;. " by a 2 - mm tublng, One tube has a long piece of tubzng endxng in a drawn: break off nxp whlch 1s broken by contact with the walls or battom of the sodzum centalning apparatusg Thas methed is bel1eved to give nenrepresentatzve‘ sampies becaase there is the 90351b111ty’9fp1ck1ng up precxpatazed sollds' "fl_269_~' el GEe e e & e 86 s sEe 6. a0 38 v el e PR EIRE X SR Jese el e e e PO T S S L . N LR LB DA b e s 88 » e Be R R R $ oW 22 o @ @ el e e e @ ERREEEE ST n_.-.t-. LX S [ X2 & - eed & 8 6@ $@ e 8 o.-og L &, o o L e o et s i e o an 011 bath Thls method appears to have promxse but needs further reflnemente * Waste metal Wlll be accumulated under 011 -or xn an otherwlse safe 3$_menta" Easposal requlrements vary w1th the metals concerned = Waste sod1umwl”” i '_tame for the operators ta take necessary safety precautxcns after dumpang the7 7*h "restzng on the contact surfacea Modxflcatzons of the techn1que are bezng- n ;fde#iSéau some. of them patterned after &he methad ased bY Knolls Atcm1s Power?i: ' '3 g:Laboratory.__-g_iz;_;fl After zhe sample is &rawn 1nto the tube *he sodxum SOIldlfl&S 1n the v. jbu1hsg5and the tlp is sealed to prevent exce331ve contamlnatlcn Wluh oxygenf from ‘the atmospheres_ The sodlum«fllled buibs are then glven to the chemicai' v ” ; anaIys1s laboratory for further handllng as descr1bed in Sec,‘21 Sampies Of thfi SOlldlfxed iaquxé metai are removed fromtfiuatest apparatus:J” -_for determznatxon of metalllc contamlnatzcns Such samples are not analyzeé '7_p for oxygen as are thase taken in. the glass bulbs but - are analyzed spectroJ" : -graph1calIy soleiy for plck up from the contalner materials (see Secx 21) PEBIFICA?IQN GF LIQflIfl fiE?&LS Afifi EASES P Lu fllll ANP BIVISIGH ;?. 1$ quuld sodxum used in the tests 1s purlfzed by fllterlng at about 259 F "flf thrpugh a 51ntered staxnless steel fllter w;th a pore size of approx1mately' 5 g Preczse determznat1on of the oxygen consent of thxs flitered soézum has ” ' ';ffinct heen made pendzng develfipment cf adequata sampllng and analyzxng technlw.' ; ques (see Append1x, Sect;on 21)*;, The argan and hallum used es an inert atmesphere will have to be purzf;ed:-,_: 'because three “eylinders of argon were analyzed and found to contazn 20, 30, ;._and 45 ppm of oxygen._ Two helium cyllnders checked were found to cantaan'_“,_ _ | _3500 and 56 ppm ef oxygen (see Seca_zi) Purzflcatzen w;ll be accampl:shed :]f'r“ o R S B o RN - i N a3 . eee el @l e e S ew R awe 0 sea a6 : RN R N S PR TR RRtear ¥ e e e e 8.6 I - N hT SRR, Sty TSR SR TR SR . o e 8 e . L@ oo e e lelell e e e E e R e e e e Lo g e e e ..'.a a0 ses @ - eed @ 6.8 sa e l'e s ese. 89, '_'by bubbllng the gas through a 2- ft depth of?%aK in addltlon to pa551ng it ove;{" ';hot copper turnlngs and hot titanium turnlngs- Testo will be made to determlne - 'the purlty of the gases after the scrubb&ng treatment (see Appendlx Sec, 21} Lmum mEmLs SAFETY ceaamrmze: P L. Hxll ANP DlVlslon S The quflld Metals Safety Commlttee is nearlng the completlon of a safety : manual for use by units handl1ng llquxé metals._ The completed manual should _be dlstrxbuted to 1nterested persons by Aprxl 15, 1951 As a result of numerous tests in. various areas,«the commzttee has recs lommended the use. of dry graphxte powder screened to 25 mesh as a fire extin- -galshlng agent for sodlum NauK calclum, uranlum RS _ and possxbly lmagneSIum fIIESe' It is readlly applled by scoop,-or with dry type pressurlzed ‘fzre extxnguzshers such as the Ansul MetaIwX or ‘the Pyrene G-1. The graphlte' is removed after use by vacuuming. Cleanlng is not difficult if the graphxte "has not been walked on or otherw&se forced 1nto the surface of flooringa The search for adequate protectlve clcth1ng contlnues. The commlttee_ v”recommends the 1nter1m use of chrome Ieather as partlal protect;en aga1nst ';fJettlng molten soélum DR R | | | SR : | § Flame retardant ciothlng is recommended for general wear 1n quuxdwmetals handling areas. It is not 1ntenéed to protect agalnst B dxrect metal burns but oniy agalnst general body burns resuitlng from 1gn1ted clathlng._;f Thermal 1nsu1at1ng materlals have been tested for re81stance to attack by 'fmciten metalsg with steel or lead-mill slag wools and °13Y bGnded dlatomaceous “silica offerxng considerable resistance to molten sodlum - up to . ghout 1700°F. Above this temperature the slag wools begin to deterlorate _from_t¢mperature effects alone and eventually fuse completely. Clay- bonded diatomaceous-silica high-temperature insulating materials, such as Johns-~" Ménfille'Sfiéerex or Philip Carey Hi-Temp No. 19, havefiroféd successful against 'éttack from molten sodium | \In many cases an 1n1t1&1 reaction occurs wh:ch appears to form a slag that retards further attack, 271 ° . N . e : . PR Y e @ .- s ° passae soes *a LYY X ] q...."n': '] ¢ esesssy YR LYY sseeee - .n.oo.oo'.' ) . seseee XLl INSTRB&EHTATIGN C D. Hatfleld ANP D1V1s1on Instrumentatlon for the llqu1d metals test systems described in the pre- cedlng sectlons will requlre the routine instrumentation for genmeral control ‘purposes as well as the indication of basic physical conditions existing in the systems. The following equ1pmen» has been developed for these liquid- metal systems: | | S T 4 ‘ . 1. Differential pressure measurements. A mpék«up of a liquid-metals - ecirculation loop has been designed and built for use in sodium . manometer operation training. The mock-up is complete in every detail, thus facilitating the manometer operatzon when used in the forced czrculatlon sodlum loopsg- 2. Control panels. The flrst of a group of exght control panels - has ‘been installed and is ready for operation of convection loops.. Twe auxiliary portable power supplies have been designed and built to supplement the control panels when these are used to operate figure-eight circulation rigs. These power supplies will be adaptable to other experimental testing equipment as -_démands for such a power supply develop. 3. Automatlc 11qu1d level controls, The two-probe level-control "~ system has operated with little trouble in both the figure-eight and convection loops. A baffle design change is being considered ~ in the surge tank of the figure-eight to eliminate the-actuation .- of the electrical relays in the level-control system., The—. ~agitation of sodium in. the surge tank is caused by electro- . . magnetic pump operation at high output levels. .BUILDI&G FACILITIES FOR TEE EXPERIME&TAL ENGINEESING GRBUP : P. L. Hlll and R. E erght ANP D1v1s1on N . Laboratory'area facilities are improving slowly with much still to be completed. One open-faced canopy type hood 48 ft long, 9 ft high, and 6 ft ‘deep has been installed and is in partial use. A similar hood 120 f¢ long, 15 ft high, and 12 ft deep is being designed with construction promised by June, 1951, With no fixed partitions along their lengths, these hoods can accommodate a great variety of test equipmeht so as te provide protection against fumes and/or fire hazards, 272 rosesad aeeves YL seeede ° e YY) : . A physlcal and chemxcal testlng laboratcry w1th benches, s1nksg furnaces, 'dry boxes, and hoods IS belng de31gned These fa0111t1es will be used for- ::prelzmlnary checks on propertles of materlals and for speczal small scale 'experxments that may be requlred | A - Areas for flow tests u31ng water and air are belng prepared Thé wétéréflw ,iflow test area. w1ll ‘be provzded w1th c1rculat1ng water up to about 2800 gpm capacity. . 1hfl air-flow test area has - one centrlfugal blewer rated at 1750 cfm - :and-a-heam'af,IZ.Trxnr-of;watsr,. Other blowers w111 be added as requlreé g Spesi*‘f'échine;'wélding;_and electrlcal shops have been establlshed ':which ére-p epared to do small or spec1a1 Jobs not otherwise handled in the main shops._ The weliding shop is prepared to do'fieliaré”éelding wherever -grequlred in the experlmental area. Exper1ments are’ underway to determine the best method for hellarc weldlng of thxn pxeces of materlals such as stalnless: ]steels.- ._;' - ' '2?3 epaw . eosoos [TA3 18] YL resves sassee e | _5_1_4;;;_'_ METALLERGICAL PR@CESSES '7E C Mlller and W D, Manley, Metallurgy DIJISIOH - The demands of the hlgh temperature reactor 1nc1ude the deVelopment ofhe | xmproved weldlng technxques for high-stress eorr0310n-re51stant welds.E_Weldf°' ing. and creep rupture laboratorxes for this work w1ll soon be in £ull use.: 3_Equlpment and materlals for these laboratorles are be1ng received daliy, ‘and -'the next quarter should see greatly expanded operatlons.m The Powder Metal-: flurgy Group has spent the past quarter 1n 1nvest1gat1ng and deVeloplng tech=fi_ _ nlques for seizd fuel-element fabr1cat10n for hlgh temperature serv1ce._ The1r :1nterest has centered chlefly around flat plates which are sandw1ches of ;claddlng mater1a1 wlth fuel between, and which may be used flat or rolled;- 'eznto tuhzng The method of plec1ng the proper'amount of fuel in the fuel 'iayer and bondlng thxs layer to the cover piate is the subgect of 1nte331ve; 1nvest1gat10n._ A program desxgned ta stndy the fundamental phy31cal chemlstry of I1qu1d~ metal systems has been 1n1t1ated The first phase of the program will be con- ':cerned with the effect of crystal orientation on the rate and type of attack _‘by lzquzd metais._ In order to study 31multaneously all possxble crystal ier1entat1ons, 31ng1e erystals mach1ned into spheres w1ll be used 27292 S . FTY 1 R a%0 k) resee ¢ () resew . ! . I E XXX ) .:”i?nghblfis Lsgégfi?bgy.i_.. - Peter Patriarca, Metallurgy Division The cperatlons of the Weld1ng Laboratory wh1ch heretofcre were performedc'f in borrowed facxlltles were transferred thls quarter to the new Reldlngj-'f" 'Laboratory.; As much. of the time durlng the last quarter was devoted to. thlssf-i' 'transfer and the 1nstallat10n of new equ1pment, there is lzttle progress ta: _report on the weldlng of columb1um or molybdenum although both these problemsf ”are stlll under actlve cons1derat10n._ The ma;orlty of hel1arc weld1ng equ1pment and asscc1ated apparatus has- "'been dellVered and is in use.: Stlli 1ack1ng are a d-c¢ motor- generator welder -and a varlable speed weld1ng turntable._ These should be delivered and in o cperatlon xn March | A.serles of thermal convect1on lcops of a new desxgn, for the Dynam1c _ ‘Corrosion Grcup, are being constructed using the facilities of the X~1@ Plpe- ‘ ' Shop.; These'loops, which will be of types 410, 430, and 316 stainless steel, i fIzstt iron, and inconel, are being fabricated as specified and under super- -:fisiofi;f S . - o S . Prellmlnary experlments have been conducted to determxne technlques of_ 'cspct weld1ng strap supports for fuel tube clusters.A This method will be used '::gto fabrxcate a test ‘cluster for use in the flgure e1ght thermal conVEctlon B 'ccloop at Y-12 | ' ' L Commun1cat1on thh the Iaboratorzes of the Internatzonal Nickel Ccmpany ijxndzcates that Jclnzng of thln walled stalnless steel or 1nconel tublng to flat or tube header by means of electrodeposztlon of nickel is feasible.: Specimens have been sent for experiment by the International Nickel Company '-'laboratory and will be returnsd to the Metallurgy Division for evaluation when '.'complete._ If these samples are enccuraglng it is ant1c1pated that further work along this 11ne w1ll be undertaken.J B | | ~ CREEP-RUPTURE LABORATORY f’.fl.:B.:OliVerg Metallurgy Division . The installation of equipment in the Creep-Rupture Laboratory should Bc ) compiéted by the middle of March., Other facilities for stress-rupture testing ; ) 5% : : Bk F . . ‘56 eed ‘e Ca s ee 66 @ see o coo o : e e e s & B ° K3 > . M :i . . 8 o o e 89 0 oo :_: o 8 @ s 9. 606 @ : : : :_:_ e ‘e e L B ] o : R : b e 266 9 '8 so .80 & & e ess :gzn ilquxd-metal env1ronment is in the de31gn stage.t Only a llmlted nuvber of tests were performed durlng the past quarter in the new Creep Bupture Laboraw' '_tory because of the construction and 1nstallat1on that was 1n progress, Ten81le tests were conducted to observe the deformatzon and recovery 'characterlstlcs of several possible metals for the reactor fuel tubes under _cycllc stressxng. Under the test conditions, 1500°F elastic recovery was not detected and the max1mum stress decreased w1th each cyclefi Equipment Installation.; If dellvery schedules have been met by the:QI | :varlous vendors, installation of the laboratory equlpment should be- comw' "pleted by the. mlddle of March Completlon of the 14 spec1al furnaces and chamber . far testing in ‘vacuum or in imert atmosphere is expected in’ 30 to 60 .-days._ These. furnaces are belng built in part by Oak Eldge Natlonal Laboratory and in part by L.'H._Marshall C0. The first unit has been delxvered and the "'yroductlon schedule calls for two or three furnaces per week. - Studzes are underway on the first furnace unit to determine the varlous operatxng charac- 'terlstlcs,_lncluélng the natural temperature gradlent and the values and dxstrabutlon of shunts to glve an adequately long constant temperature zone. A 25 kw gasollne driven motor generator w1th automatic transfer switch _’has been purchased ta furnish emergency electrlc power for the laboratory.: VAIternate sources for other serv1ces are provided but require manual sw1tch—. 'over,j Man- power requlrements have been surveyed -and.lt 1s.planned to have an 'operator 1n constant attendance.' | A procurement program has been 1nst1tuted to obtaxn sheet rod, _and 'tnb1ng of the various materlals that may be of Interest on the ARP Progect The productxon 0f test spec1mens of several types and from the various ma- *terzals has been dlscussed with the Research Shops. * Toollng is belng planned to prov1de speclmens 1n lats of about 25 at a mlnlmum preductlon cost“ _ 0w1ng to the lack of space ‘in Bulldlng 2000 ‘and- the f;re hazard 1nvolved '1n handllng liquid metals, space has been obta1ned in the basement of the Pilot Plant Bulldlng to install a laboratory for stress-rupture testing in 'llquldwmetal-env1ronments.j This laboratory is in the preliminary design stage. It is planned to install four tube-burst units and six lever-arm type - creep racks along with the necessary instrumentation and other-afiXiliaries.; ‘The planfiing provides for possible future expansion if the phqgram'at_any 294 L3 [ esanav -] L 4 ‘awvame eossee. @ - PTEYY R = ofuture tlme 1ndlcates such an. expan51on to be de51rable.; Tho ffifnaoés and test: chambers to contaln the specxmens and lquId metai are in the flnai deslgn stage.“ Cons1derab1e attent1on has been glven to the desxgn problems 'so that the amount of handlzng of the lzquld metal 'is minimized and the hazards are kept at a mlnlmum.‘ Thls laboratory wxll also be under the 24 hr. s&perv1s1on of an operator., | o R _ . . § U N No tests haVe been conducted durxng the past quarter in thxs laboratory.- -rarea.: Durlng the constructlon perlod the room has been crowded with the: serv1ce crews and the power has been frequently 1nterrupted maklng testzng i -_1mpract1cal.,- Tensile Tests with eyclic Stress ' Eighteén téhsile.fiests were conducted at 1500°F to observe ‘the deformation and recovery characteristics of typeso 310, 316, and 347 staxnless steel and inconel under cond1t1ons of cycllc g;res31ng The tests were conducted at two constant ‘strain rates of 0.05 and: : 0’107ih 7m1n.; The straxn rate was maintalned constant durlng loadlng until 'the ioad ceased to 1ncrease., The load was then released siowly The specxmen was held at zero stress for 10 min prlor to reloadlng., This was repeated for : four or flve cycles, g1v1ng a total test tlme of apprcxlmately 1 hrh _ Thxs series has Just been completed and ‘the data,'whlch have not been .3aéequately ‘studied, will be reported at ‘a iater date.; Followxng are a few_l -goneral:zed ohsorvatzons i IncreaSIng ‘the strain rate markedly 1ncreased the maxlmum stress o obtalned - TR - . ~2;;«Generally, the maximum. stress obtalned in each cycle decreased o owith each succe351ve stress cycle.; | 3.;_Under the test condltlons elast1c racovery was not detected "4,j'Tests performed at room temperature showed that these types of ~~ stainless steel exhibited considerable elastic recovery.: " PHYSICAL CHEMISTRY OF LIQUID METALS GQ;PA‘Smith, Motollurgy Divisiono 'A program to study the fnnéamental phy31ca1 chemlstry of izquzd metal corrosion has been started As various studies of the phy51cal chemxstry of ¥ 395 T T U St .. e 0o = :“ : :'I. :‘s % . s & P @« .4 : S i s iea e o e e e e s 0s. s a8 o2 2 * € . . & see b s s P IR SEE . s . & o - . ® L . =1 PR P "' ® @ . B ® L3 it o PYSRE T Y ] » a8 S0 0 8. &6 ¢ s .0 sse. ™ et many surface phenonena have shown a. marked dependence of rate on crystal face this program w1ll be concerned 1n1t1ally w1th large 31ngle crystals. Even a perfectly prepared polycrystall1nesur£acepresents a d1ff1cu1t problem because of the numbers of crystal faces and boundarles.. The dlfferences in behavzore3°' .of crystals of varlous or1entat10ns has been markedly demonstrated in the caseflfj of the ox1dat10n of copper at 1ntermed1ate temperatures and in. the case of ‘the catalytlc decomp051t10n of carbon monox1de on nlckel where entlrely &1fferent-; fprocesses take place on different crystal faces.: In otder to 11m1£ the number _of var1ables, 1n1t1al studis 'w‘el be devoted to galnzng an. understandlng of 3 the propert1es of the dzf ex3°t crystal faces by the use of metal 51ngle : \:crystais exposed to a varletv uf 11qu1d metals.f A qualltatlve survey will 'd=term1ne the extent of expeeteq dlfferenees and the most act1Ve crystal faces‘f : fer a glven process.;fi_ Liquid Metal Corrosian of Single Metal Crystais,; As sxngle crystalsz 3 machlned 1nto spheres present every crystal face at least thce on the sur-. ':a_face, all crystal faces may be 81multaneously compared in a 31ngle experlment N In F13.514 10 is shown a s1ngle crystal sphere of copper, ‘the surface of wh1ch was initially electrolytlcally polished in an orthophosphoric a01d; 'solutlon.; ‘The pattern result1ng from the different rates of etching attack on" | ”"dxfferent crystal faces are ev1dent In Flg._14 10a a 2100] axis is normal to | .f the plane of the page.“ Copper has a cubic lattice and the expected fourfold'. V Lsymmetry of the (100) pole may be seen.t ‘The (100) pole itself is the geOw '>}_ metric center of the fourfold symmetry. In F1g..14 105 a [111] axis is normal " to the plane of the page._ The expected threefold symmetry may. be seen. All. _;'the other crystallographlc poles may be located in like manner using ‘the. " methods of crystaliography For example, e (110) pole iles mldway between any | "e-palr of ad;acent (100) poles or any palr ef adgacent (lll) poies." ??eceasing of Single Spherica] Crystals., The feliowxng procedure will ?_be used with spher1cal single crystals The cold-worked surface layers will " be removed electrolytxcally until a smooth surface is obtalned the complete- Thess of cold- working removal being 1ndzcatedby the sharpness of the etched pat- e tern, Annealzng will then be carried out in an atmosphere which will reduce “‘the oxide film, where possible, and then the crystal will be immersed in the 'llquld metal without being subjected to atmospheric contamination. Ani ap~ paratus for lxqu1d metal--single crystal systems which do not attack quartz is essentlally complete.: The first system to be tried will be mercury-copper, as | vo ese e e s es ee w :'_" : '.:..!_ =S ‘. e . @ @ e '@ e & _ b ve TR $ o ® LR IL S : :‘N- T B N R LU : . ) 3 -.:o.'-'o'o'a' e e 8w ek - e¥ #0089 s 'the 51ngle crystals of copper are 1mmed1ately obtalnab e and the handilng of"' mercnry presents no magor problem._.-e To demonstrate the feas;bzllty of the method as d means of study of such‘ ”'systems, a 31ngle rather crude experzment was performed A 51ngle crystal" cf;. :copper was 1mmersed in’ a bath of mercury at room. temperature w1th no pre- cautxons taken agalnst atmospher1c contam1nat1an., The result, deflnltely' ‘prellmlnary, was that certaln rather sharpiy deflned crystaliographlc reglons" of the spherleal copper crystal weTre wetted_ g1V1ng a uniform’ fllm, whlle the other reglons were ccvered thh flne droplets._ The avaliabxllty of 31ngie crystals sultable for corrosion stud1es is a 'matter of some concern. Although a varzety of metals may be produced in th1se :_form by the Brldgman method few are commerc1aily obtainable, and those_ : crystals whxch must be grown by other methods are very scarce commerelally.t-i Anether problem yet to be selved is that of mach1n1ng the spheres from-e -'the 31ngle crystals in such a manner as to make possxble the removal of the ecold worked layer by eleetroiytlc pallshlng W1th some metals reasoneblefw u:success can be obtalned bat it may be necessary to grow others in spherlcal f-shape.‘ The major dszlculty is in the machlnlng of those metals which may he__ ':fraetured by cleavage.; In these metals deep cracks,. although not perceivable ',before corr051ofi sometxmes fcrm owlng to abnormal corr031on along the cleavage. planes._? _ Very Ilttle experxmentel attentlen ‘has been glven te surface preparatlen Teven though 1t has been generally recognxzed that physxcechemzcal processes “eccurrlng at the surface are frequently sen51t1ve to surface preparation .fmethods._ It is hoped that research 1nto surface preparatzon w1ll not be' " necessary exeept where the meanlng of corroslon results is in doubt _amomous- ‘F:;ufms-" A study 1s to be undertaken to determlne the approxzmate llqu1dus curves .for systems cantalnlng a high percentage of hydrogenous compounds. The pur- pose of this study is to determine which systems ‘might be vseful as moderator- - coolants in a hlgh temperature reactor on the basis of meltlng point.: Sub- ‘sequent studies are planned to determine the stability of selected mlxtures- and to establlsh methods of decreasing undesirable dlssoczetxon.AThe necessary _apparatus for conduct1ng thelzquuhm curve determinations is being designed.- 208 [(AEXE L) ° aesave . sesse g e : il seew - XTI sepese 7fifl15 RABIATION DAMAGE D S BxlllngtonF Phy31cs of Sollds Instltutq,Metaliurgy D1V1310n; _ The effect of radlatlon on the creep of metals w111 be determlned by.. cthe cantllever creep apparatus (ORNL 919) o Recent effcrts have been concerned 'cW1th 1myrov1ng methods cf constructlon and checklng the components of thc 'uapparatus to. assure the requlred accuracy A ten511e creep apparatus is also - belng bullt, The creep spe01men w1ll be tubular so. that 1t may be fllled wzth a lxquld metal for stress corr051on studxes under 1rrad1atlcnr- The apparatus has been de51gned.for1ncasur1ng thermal ccnductxvzty changes _durlng lrradlatlong.and bench test1ng 1is now im progress,_ The equlpment for 'jthe before after measurements of thermal ccn&uct1v1ty 13 also belng dcsxgned | The Y«I? Berkeley,_and Purdue Unlver31ty cyciotrcns have been appllcd to ‘L:radxatzcn damage problemsai A rotatxng cyllndrlcal target wh1ch w1ll be" }1qu1d metai cooled is belng developcd fcr the Y 12 cyclo!:ron° Thls target .iWIII prov1de creey data for. the target materlal whlle belng 1rrad1ated TwcckJ;f -'autempi;c: were cac.e i:o 1rr cfi a‘*c 13.6*{1”& mc**‘:a,ls in 1r0n ca*csui% usmc the €C-in, f_Berkeley cyclotrong but in each ‘case’ heater faliures prematurcly termxnated c;fthe experlment The Purdue Cyclatrcn Group has hegun a. study cf the prcpcrtlcs '_'ef unlrradlatcé moiyb&enum, preparatorytc raélatlcn éamagc work on thls metai '*H“Instrumentatlcn fcr crecp mcacurements under deutercn bombardments is be1ng 'fl:developed and a prcllmlnary measurement cf re31st1v1ty afcer 1rrad1atlcn has.cc=-": 'been:mach e The 1n1t1al 1rrad1at1cn of the fluorlde ectectlc Na? UF¢ cv1deficcd no. '”1rrad1at1cn 1ndaced prcssure rlseg as mlght have resulted 1f free fluorlne gas: . '3;had been fcrme& CREE? UE\EDER I RRABIATI GN J ol Wllson*f;_c 3 C Zukas M 5 Feldman ~ W. W. Dav1s Metallurgy Bzv131on Ne furthcr data from experlments on creep under 1rrad1at10n have been/ obtamned The maln work durlng the quarter consisted in 1mprov1ng methods ofr constructlon of the apparatus and checklng of its components o ._;-'ufi . [ _::299?. -« swaene U e & . e en e L eeee ..'.'_....‘. R » e eee e e S sonws T e X IS TR XN [ ] _ol;.t(: S o Operatlon of a new. creep cest in the 1ntense gamma fleld supplled hy a]_;.u' '”fC 60 source located 1n the canal at the Oak Rldge Natxonai Laboratory Beacter is cantemplated The geometry of the sourcels such that the present cantllever ' apparatus must be modlfled to place the long axis of the speclmen vertxcal ¢-fThe apparatus 1s now belng modxfled in the shop cantllever'Creep Apparatuso A preilmxnary test of effect of radlatlon onj S 'thermocouple characterlstlcs under 1rrad1at10n shows a p3331ble change of emf of chromel alumel and 1ron constantan Peupies at 1220 F after a perxod of tWo _:weeks° These thermocoupies were used to run coollng {meltlng p01nt) curves on -fl=-pflre alumlnum 1n an alumlna cruczble 1n the reactfir | Effects ef masses @f metal 1n the fleld of mlcroformers on the llnearlty fof the respanse were studled Furthermoreg temperature ceefflcxents of micro- 'fformers ané varlatlons 1n cores were checked to eilm1nate poss1b1e errors j th 1nterpretat10n of creep results°= thtle _1f any, effect on. the mlcroformer f_was observed when 1t was maunted on a slotted alumlnum baseozu; ' Tens:le Creep Apparatfls, A pzlot model of a tens;le creep apparatus.: 'icontalnlng 1ts ewn loadlng'welght on a lever arm is in the shep The spec1men : w1l1 be tubular so that materlals Such as sodlum may be 1ntreduced into the "itube ta s&udy stress corrasxon durzng creep testsa_ At least two complete 'tests at 1509 F en 347 or 316 staznles» steel w1ll have been run in the. 10ak dege Natlenal Laboratory Reactor in . March thh stresses sf the oréer af ~ EGGO p5133; 13_- - R _ ; THER?&A% flfi?\‘fiUCTIVITY fiF STRQCTUR&L ALLQYS AT fiIGH TEEP@R&T%RE A F Cchen Metallurgy BlVlSIOn‘ Des1gns have been prepared for the measurement of the change of thermal ,-conductlvzty is a functlon ef temperature under plie 1rrad1atlono Ccnstructlonfl- of fihe experxmental parts for the meamurements on 316 staxnless steel arg -'nearxng completlen and prelimlnary bench experlments wxll have been 1n1t1atedt' 1-by March. The other materlals to be con81dered for thermai cenduct1v1ty77 'fmeasurements ‘are 347 stalnless steel 1ncqnelg 1ncane1 X, and molybdenumfi 1 3§§_ ;~ ® e Cemewmew L o Cewwean ot =8 Y . : ® o H ° e L) : ® J e o v ieee @ ] o s @ : te et e 2 ® L - T ew o8 . LR IR g X .8 LR 2 [} Leaeeesl o o Out of pile experlmefits are belng planned to determlfie absolute values of thermal conduct1v1ty of alloys at high temperatures ‘The a uracy in hese out- of plle experlments is expected to be w1rh1n a few perflen* i Y 12 CYCLOTRON CQEEP EXPERIMENTS T ‘H. Blew1tt andhfi J Feidman Metallurgy Dlylszon': ‘R Sq_L1V}ng$;on Elecfiflamagne?lc Besearch Dlvx31on The experlmental arrangement to “‘be used in the 86 in, cyclotron for -determlnzng the effect of radiation on the creep of metala at elevated temper= atures has ‘been worked out in pr1nc1ple and the components are in probess of constructlon The 1nternal beam of the 86 in. cyclotron will be used: zo_- .:“1rrad1ate a thln walled cycllndrlcal oample wh1 h will be cooled by IquId: meta1 pumped through the cyllnder at the rate of apprOXImately 10 gpm by an ' electromagnetlc pump. The cyllndrlcal-bample will be rotated at 500 rpm to o fac111tate llquld metal pumplng andtx)obtaln unlform radxat1@n:u1the dlrectiont 'of retatlon _ In maklng creep measurementa it 1is nece oafyfthat the temperatufe:of’tfie ‘target not fluctua*e more 11"han a few degrees. The 11qu1d metal flow1ng through ~ the thin- walled cyllndrlcal sample results in efflclenf hea? transfer with a _-temperature gradlent of only a few degrees . depend1ng on ghe material under 3 test° The fluctuatlcns ‘of the sam@le temperature from cyclo*rfln "beam off“ to "beam en” Wlll be of fhe-aame arder [ ...'. ® * . 200839 : [ [ E X R X K PP e o ses0e hd " EETY X *s » *0000 ee0e8® R L EREEN X J e 'PURBUE CYCLOTRON The Purdue program empha31s has beefi placed on molybdenum and stalnlessf _esteels whlch are of 1nterest in high- temperature reactor deSIgns The ma;oru ~work has been w1th molybdenumg_ After technlques have been developed w1th. molybdenum9 the same apparatus w1ll be applled to the 1nvest1gat1on of sta1n~ less steel Prerequlslte to tbe meesuremene ef the prcpeftxee of molybdenum after - '-1rrad1at10ng a study of these propertles in unir. efi**fed molybdenum is being = "undertaken. opertlea eurrently under study are grain size, deBS1ty mlcroe hardnessg and resxst1v1ty Prellmlnary measurements of the resistivity af' molybdenum after 1rradla*1en have been made andzu:change was detected although. ‘one day elaysed after bombardment before a measurement ceuld be safely made. ' Instrumentat1cn for creep experlments under deuteron bemhardment is underway. fleLControl cxrcults are completec Propertles of Bn1rrafiiated Molybdenum, In order to obtain reproducible resultsP a study of grain size, degree of preferred-oriéntatiflng strain, and hardness of commercially avallable rolled molybdenum was made. Three standard :'thlcknesses e 5 10, and 20 mil — were selected With the 5~ and 20-mil "__samples, ewder and back reflectlon X ray: patterns show sharp lines, 1nd1eat=7 ing that ~grain 51ze is' in the range 10 2 to 2 % 19¢4"m5_ X-ray patterns from :ithe 10- mxl sample exhlblt spettlness in the rlngs 1nd1cat1ng that the graln_' - sxze is somewhat larger than 10 2 mm, Preferred orientation is marked in each E thlckness} but partlcularly so in the case of the 5 mll samgle Den51ty wasfi' 'calculated from the lattlee parameter determined from back- reflectlon patte?ns and the value obtalned was 10, 22 g/cca Agreement in denszty between S~ and_ 10-mil samples was within the experlmental error. These results were supplew' ,,‘mentediw'photomlcrographs of the pol1shed and etched surfaces of the dlfferent: :-sampies,. U31ng equlpment w1th a magnzflcatlon of only lefldlameters,'lt was. poss1ble to resolve graln boundaries in only the IG m11 sample. This is ccnsxstent with the X- ray determination of grain size. Wlth the 10- mll semple' the grain size was estimated at 7 X 10°2 mm. This value is greater than that ‘indicated by the X-ray patterns, but, -accordlng to the Oak Ridge Natlenal : Laberatcry grefipy thzs dlscrepaney is typlcal of the two methods9 has the. *Since we have no explanatlon for the appearance of two very weak extra lines in the back. refiectlon patterns, a measurement of the bulk density was made and the value obtained was 10.21 g/ce. ew. . eee a9 °: a8 €6 o S 68 & acs a0 .0 8 o 58 o B . RS s A el ‘.0 o8 o e e e # s e .98 6 @6 " e 8 L N R SR N 5. 1) . L % . L ] N IE T SRR AN Y L3 RS Beg B LR A L (X ] ¢as . 900 ®. 5 &% an 2 86 RO 8F ’Q;usual order of magnztudes “and is in the same'directien The same X ray and '-,photomlcrograph technlques are now belng employed to determlne the strueture of the relled sheets after dlfferent anneallng treatmentsw' Mlerohardness tests taken with a Tukon tester yield an average Knoop number of 245 for the_;%-” '5 mll sample, 267 for the 10 mil samples ‘and 233 for the 20-mil sampler Reszst1v1ty measurements using ‘the equlpment 1mmed1ately avallable have_ 5been troublesomefi_ Three different sample hclders have been desxgned and '-tested but further modifications are required before the de51gn will be sultable, The Just1f1cat10n for seekrng an’ extremely accurate and sen51t1ve 1.means ef measur1ng*resrst1v1ty is that bombardment lnduced chenges 1n re31s= - ,tlvxty are small in the case of metalsaf Re51st1v1ty of the 5~m11 sampleln,as recelved condltlon has been measured IW1th1n an error of 8. 4 x'10°9% ohm- cmg'the error belng averaged by applylng the 'method of least squares ' The re513t1v1ty p as a functlen of temperature T is - lxnear w1th1n experlmental error9 and can be expressed as p = 5, 022 - 0. GZZGST;.- .”over the temperature range 77 to 300°K. It should be added that reflnements of 'the technlques are not yet 'such that thle accuracy can always be malntalned | w:th successive measurements | Wlth less eccurate and earlier measurements. of ‘the 5--and 10-mil sheets, the re51st1v1ty agreed with the equation just given fthhln experzmental error. It is surprlslng that the dlfference:nxeeld werklng -between the Sf and 10 mil sheets, whxch is’ quite ev1dent in X- ray measurea ments and . hardness tests does not preduce an eas1ly dlscernlble dlfferenee in :-'re31st1v1ty Resistivity ef Meiybdenum After Irradlatmn° One ef the 5- mll sameles of molybdenum was bombarded in the Purdue cycletron fer 4 hr w1th a 3-pa beam of - 10- Mev deuteronsa; The sample was. cooled hy heat conduction through a brase bar 1mmersed in dry ice, and its temperature was measured during the bombardw 'ment thh a thermocoupleoy Whenever the temperature rose to 40°C, the bombard- / ment was temperarlly 1nterrupted ‘while the dry ice supply around the brass bar 'j'wes replenlshed Cross-sections for deuteron-induced transmutations in some ; of the naturally abundant isotopes of molybdenum are hlgh and after the’ bembardment the sample read over the tolerance on a health monltor at a _ distance of approximately a meter. After three days the tolerance value was obtained at'a'distafice of about 10 cm. A radiocautograph of the sample;‘made with a 5- ~min exposure, indicated a relatirely uniform bombardment aeross the semp1ea A more accurate measurement will be made after more deeay w1th a ]303' - sscoee . N R ®, ety segose seend o 8 seece a0 ¢ sessee ssoens Te el o Ry YTV YT N pestes e e e ‘sepans R Y'Y S — "' Ge1ger Mueller tubes'masklng dlfferent parta of the surface Wlth lead It'fiaéfifff 0 ' §fe1t chat as the recrystalllzatlon temperature of melybdenum is around 1280°C o -the effects of radlatlon mlght ‘be ”frozen 1n" by keeplng the sample below 40° °C H”durlng bombardmento. Not hav1ng hot- laboratory fa6111t1es for measurlngg{* 'reszst1v1ty, ‘one day ela?sed after bombardment before a measurement coald bej : : 'safely made. By that ‘time any varlatlons 1n re31st1v1ty wh1ch may have. been_ ' fproduced weré not detectable by the apgaratusu Hardness and X ray measurements ':awazt further decay cf the’ sample Be31st1v1ty measurements in place would be 'desxrable and ‘will be attempted i Smaller samples mlght also be used as theyrh ”f c0uId be bombarded Ionger and glve the same total actlvatlon &s a large sample '- 1rrad1ated for less t1meo_fv' Creep Instrumentation@ Eeman&s on 1nstrumentat10n for creep under bom~ o ;bardment are critlcai because of the short range of partlcie penetrat1on;' 'f?and fluctuatlons in Speclmen heat1ng Prc’dficed by the beamfl- A large assemblY _ _ of 1nstruments is requlred 'and thls part of the prcgram is. In the deveiepment '1”' ”:staga,' Control cxrcults have been completed These 1nclude tunxfig coil :,assemblles, tuned ampllflers, and recordlng 1nd1catcrs of creep and creep' 'rate. Tests of the czrcults have been satisfactary as. to SenSItIVItY and ”stablllty, but suppiementary shleldlng and fllterlng are requlred to. reduce ' stray plck up A mlcrofsrmer type of mutual 1aduetanca brldge has been -jdeveloped for. usa as the sens1ng dev1ce in the extensemetero The mechanlcal_.' :]~_desmgn of the creep machlne is complete except for the furnaceg and about 80%fi i of the parts for the machlne have been tfirned out by the machlne 3h0p Bew13 8 qulrements for centrollzng the temperature grolee acrass the wlre arg strzn—ff -gent,;and the development of a sultable furnace may take several months° 17' f Future plans are primarxly cancerned thh a satlsfactory cempietlon of *he” . work now nnderway However9 1f p0551b1e9 1t wculd be aévant&gecus to 1ncorwf ffiporate measurements of other phys1cal prapertles and enlarge tha scope of -T '_metals under 1nvest1gat10nn.fA re§et1tzon of a creep experzment uslngf 'pclonxum alpha partlcles 1n01dent on & szngle crystal cadmlum~w1re creep-' 'spe01men9 w111 be carrled out. 'Ewge?s‘ or ‘mmmw ON A "FLUORIDE FUEL W R Grtmes, Materlals Chemlstry Dlvxsxon. Fiuorlde fuels whlch appear satlsfactory for the ANP and ARE reactors_ have been developed The one reservatzen that is cemmonly expressed in connectlon wzth :he use: of these fuels, 31nce no 1nf0rmat10n to ‘the cantraryf“f”“'": Tientin : SUTERIRR IR bt BE . IRE 8 L #8885 0 688 P B8G B § e 8y BT TR AT Ayesey Sus T T MU Sl SRR . e . ] E N T Y TR TR SR R & el e e e e B e LR SRR TREERTE DERER . S SEREN [ 20 TR PR Tk e LR O AR e ORI St TS IR SR S SERERL o SHEV SN 4 CUIAL T RN ALY LDl e T IR X RSO Mt BN L R ex1sts,'15 that decomp031t10n and corr031on effects mlght appear under 1rrad1«"” ~ation. Consequently,'the controlled irradiation of a fluorlde fuel was. underm e "taken under the dlrectlon of Er V. P Calklns of NEPA The fluorlde eutectlc 26 mole % UF and 74 mole % NaF was 1rrad1ated in the X 10 plle._ The eutectlc was. placed in a heated ccntalner Whlch was' equlpped w1th pressure and temperature 1nstrumentat10n Prellmlnary checkse 13were performed outsxde the. plle, fellow1ng which the test chamher was inserted . in the X-10 plle in-a. neutron flfix of approx1mately 1012 per square centlmeter' - per second. Since the fluorlde sample would be too radloactlve for 1aboratory- 'analysxs, ceollng curves were taken while the sample wasxe.the plie to give an llndzcatlon of any change in its chemlcal comp031tlon However, these cooling. “curves exh1b1ted dlscontlnultles of unknown orlgln around the melting point of | ‘the eutectic, although the characteristic curves of the orlglnal eutectic were 'iobtalned w1th no neutron flux, Furthermore, no irradiation-induced pressure. 'rlse was observed From thls 1nformat10n it may be tentat1vely concluded pendzng further experlmentatlon, that the rad1at1on damage of the eutectic mzxture in a flux cf 1013 is negllglble° 305 (I XN X N L 2 L] & % (RN L X R 2l L3 EX R R YN L d [ Lo . e X F 3 _' . ® B ‘aeeeee e ... S veee | Ve SLe 0B L EX XY ] ST e P P T Y R X ] [ eseves aseses Part IV ALTERNATIVE SYSTENS . I_NTRGQUCTIQN **m-.mmi Ix;' e At the present txme the Alrcraft Nuclear Propu1s1on Program is p01nted'” ffltoward development of ‘a B- 36 type of aircraft powered by J-53 turbogets w1th a sodium- cooled reactor u51ng quiescent llquld fuel However,.severai alter» :natlve types of reactor systems are stlll rece1v1ng attent1on although none of these systems is yet known to be as feasible as the one now chosen for the ' maln-llne effcrt, They are, however, 1nterest1ng, ‘and several have patuntlaIQ. ltxes 1f thelr present dlfflcultles, chlefly 1n materlals, can he *’arcome,; The work of the Oak B:dge Nat1onai Lahoratory and its’ subconcra torg'dnj ’;the snpercrxtlcal water system, the sodium compressor Jet system, the hemé-' "”geneous reactor, the clrculatlng fuel reactor, the c1rculat1ng modera*or re« actors and the supersonlc tug- tow system is dlscussed in Secs. 16 through 19 ‘.7 307 e ‘a9 0 . e [ ] Tee a8 4. 638 ¢ ads sa e e @ - & e & e &. @ * ®. 0. . @ . L R N E Y e - . o * L] . S0 4 o8 [ 38 3 S TS S SRR . a0 @ s ¥ e, s 8 T8 el ete e e e e.e A L8 e e e e e +0. S0 . 4869 4.8 we ee 8 o . 000 o8 '_ieq Homooemeoos CIRCULATING- FUEL, AND ~ CIRCULATING-MODERATOR REACTORS "oeAtomie Energy'Diyision The H K Ferguson Co , Inc. The fxrst of three scheduled analyses of the fea51b111ty of nuclearu Q}a~ 'powered a1rcraft u51ng reactors of the homogeneous, c1rcn1at1ng fuel or cir- ..culatlng moderator type has been complete& under subcontract to the Oak Rldge ;'Natlonal Laboratory during the past quarter by The H. X. Ferguson Co. Thelrf' report, HKF- 109, entltled Homogeneous Beactor for Subsontc Azrcraft has - aIready recelve& wide c1rculat10n It concludes that flying an XB-52 plane at f35 000 ft and Mach O 8 w1th modlfled J-53 engines powered byzaNaOH homogeneous o reactor is entlrely feaszble, and’ that a sultable reactor can be built pro~ f .v1ded that a corroszon re31stant contalner for NaOH can be foun& and that efe ofsolutlon or suspen31on of ‘some uranium compound in %aOH whxch is radiatzon-\ - ofeteble can be obtalne& Research toward solv1ng these mater1a1s problems is now underway at ‘Oak : dege National Laboratory | The general prlnc1ples of the Ferguson desxgn are. b 'shewn in Flgs 16.1 and 16.2. | The H K Ferguson Co is how engaged in an anaiytlcal study of two other types of alrcraft reactor systems which have been mentlone& in prev1ous ANP_' 'dzscus31ons but whlch have never been thoroughly studled by Oak ‘Ridge Nat1onal'“ ”'Laboratory These systems utlllze the follow;og "1 The ozrculetlng fuel reactor basedcnxuranlum d1ssolved in liquid o blsmuth ‘as the prlmary worklng medlums w1th a sol1d moderator ~ 2. The c1rculat1ng moderator reactor usxng NaOH as the primary BRI worklng medium. Two types of fuel elements will be investigated . for this reactor, a hlghvconduct1v1ty solzd fuel element and a E 11qu1d fuel element of the NaF UF, type These analyses are Just gettlng underway at present It.is'expected that the : Work w111 be completed by the end of June, 1951 | | S EEE sl _-'308_ 'm » cEevse ceavee - esean@i IR *040D0 e L4 i LR sseees wemone ool el e E e Jemeag L _odi‘flh" AT ....... LAMINATED 42 N REFLECTOR cootine | DA © PASSAGES ~¢iTAlk 1 PREBBURE SHELL . X-10 DWG.NO.l0387 . . | ~TUBE SHEET ~ FROM JET RADIATORS . y A A R R b R LB B SR AR “‘__::“ N e _\.__.\: .. g I e .-f S . EXPANSION JOWT -~ CORRUGATED ~CLADDING . - = _ INSULATION —% SEE DEVAIL B — FOR OUTLET @ VANES. QUTLET SHIELDING VANES e R . _FIGURE 16,1 - | e ASSEMBLY. DETAILS OF REACTOR AND HEAT EXCHANGER Y THINBLE 23° ~QUTLEY BHIELDING VANES . - DETAIL "B" / Yo JET RADIATORS CONTROL MOB . V) viEw LooXme { - DOWN AT OUTLEY - " X-10 DWG. NO.10586 rB R /,-Pmma'f FLUID Fau. uns v.\a.w: uummflm | SMIELD FILLLME = | ~SECONDARY FLUID FILL LINE VALVES (AUTOMATIC) | o A—I VALVE fl““’“‘% SHOT FILL LINE VALVE (AUTOMATIC) ' e e s e a2 A . ‘4 Sl T LT / ) & e main T %= oy R4 ‘LANDING GEAR - . RADIATOR— LY \\‘ . A Y N .- a | . , . b 13 P - A : L DOWN LOCK . REACTOR SHIELD / _ " AND CONTROL RCD s . i COOLING AIR] INLET - ! 5= _ v _ N | : er eV e &fl;@é e N ENGINE AIR ' Lel %Ecoamanv FLUID DRAINN, ‘THREE way - VLN e TR s \ | ._va;_ygs_uurog,gm.;_.__; _BY-Pass vALVE/ g ~ * COCLANT CIRCULATING PUN?S _ \\ © 0 CHEGK VALVE m=e=r __! : =T ~'_GROUND LINE -y - : 3 KRR T - N R S SN T LoRiMARY FLUID DRAIN VALVE (AUTORATIC) Lni-:ga- MAIN LANDING GEAR SHOT DUMPING GATE AND LOCK ad COMTROL ROD HOUSING ' : - SHIELD DRAIN VALVE (AUTCHATIC) . VARIABLE. AREA COOLING AR mm.z\'A ‘ : T " ACCESSORY DEIVE - -1 ANDARD XB-B2 : ’—?. . P . .- 2 i smm | 7..’ " FUSELA i ER R «TURBO=- JET ™ W j y ) L ‘_;_ v audll ‘. ’ \-4 . Voo 9 i . . - ; ‘\ w-\.‘_ .z. fl\- 3 }.{ A, AT '\\k s } Al . \M | 4 ENGINE INLET | ‘REACTOR SUPPORT .‘ . ar puers - b STRUCTURE 'SECTION A-A - SECTION B-B " SEGTION G-¢ FiGURE 16. 2 &UCLEAR POWER PLANT INSTALLATION »e B8 9 8 L B0 9 % ses & .oaw 0 3 -3 e e e @ e L) L4 e e e e LR T D [ RTIEERT A ] PRE T IR X R .8 0. & '8 ete [ B LI Y ] e LI ® 8 @ o o % .. L. s & 8 " e..@ e . ‘e L3 28¢ 68 8" 80. 38 L A ) LI .68 1ifé;fYAPOB%CfCLE:ANfiffifiiivm4éycsg*SYSTEMS:7f‘ Atomlc Energy Research Department North Amerlcan AVIatlon fIncg_i North Amerlcan Av1at10n Incg, has completed prellmlnary studles of bcth:“q "the mercury vapor cycle reactor and the hellum CY613 system,'and is preparlngfifil ' " an extended report an each NAA conciudea that nelther of these spe91f1cfw' -L systems shows promlseg_but from thelr analy81s of the vapor cycle system 1t _: ('_appears that the use of Ilquld vapor sodlum may reaalt in a fea51ble raactorq“-f m ngurthermore3 NAA suspects that a- hlgh (approx1mately 22@0 F) temperature w1lif o be requxred of a Supersonlc vapor cycle reactormp S - VAPOR- c_sfc'Lf'é‘. '_sfs’rm e North Amerlcan Av1a£10n Incbg has been explerlng the probable conse~ f'fi tquences if iow L/D ratlas for a Mach 1 5 plane aheuld force the ANP 9rogramg:. ' 51nt0 hlgher reactor temperatures than are belng con31dered at the presenttlmeu:”:' Any 1ncrease 1n requlred reactor temperature above abaut 1890 F w1ll brlflg the '1fturb03et radlatar temperature up into the range where pretectzon agalnst alr-,: 1?021dat10n beccmes extremely dlfflcflita NAA therefcre suggests that, as anl' 'gsaiternatlve tc the research on hlgher tem§eratfire a1r reszstant metals now; g01ng on eIsewhere, the ANP program con51der the use of £ llquld vapor cemw; . presscr 3et in the secandary 1cop for these hlgher temperauure systems :§he" f_advantage Gf the compressor get is that the turblne is cempletely enclese& 1n[' fja nonoxzdzzxng atmosphere, whlie the azr radxator is at the lowar temperature; ‘end of the e1rcu1t and 'so may pfin$lbl¥ be kept belflw the dlfflcult 0x1&at10n ' 'jrangeg- For the secendary worklng fluxé of such a compressor Jet cycley_NAA- 'has expierad mercury ilquld vapor and is preparlng an extended regnrt thereon,:§ ‘ However, they flné that the temperature range avaxiable to the secondary 1009'” .'us1ng mercury is sa much cunsfrlcte& by the Low. cr1t1ca1 temperaturecd?mercury]u" 1that the cycie becomes 1nef£101ent for alrplane use The use of 11qu1d vapor - . 80&1um as the best secandary workxng fluld for. such a compresscr jet is '=theref0re proposed A detalled report on a Mach 1. 5 a1rcraft usxng thls'. -_'sodlum lxquld vaper cempressor get cycle is now in preparatlona 3@9:0 S e se T dee i ele R L A L T T o Sy Ty | o e e alel el iR e el e [ YEK SRR THY REUREE S N ERL SN § [T I LA L PR ST St & o8 e 8 LA POSER SR TSI RERER SRE T SRR ML JRIERLL FEE B e e elw e o e e e e el e L SR ¥ N S N TEN L SRR X TN JE LR M L L S L ORRS | ngh Temperature Materlalso_ At reactor wall temperatures above 2200 F it e'w1il probably be necessary to use prlmary coolants with low vapor pressures- aee g those temperatures ‘Such 11qu1d metals as tin, blsmuth magnes;um,fend '.alumxnum_are-belng cons;dered as’ prlmery_eoelantsq Varlous conta1ner mater1als sudaascarbon tungstens molybdenumgtantalum, 'and TlC will be tested for eorr051on res1stance to these suggested primary S 711qu1&s,_ In add1t1on; a con31derab1e amount of corrosion research is now ; 'underway on: both graphlte and the above contalner mater1als when exposed to _11qu1d sodlum and to sodlum vapor.wm 1n1tlally at 180fi°w-efid eventual‘v up to e '__3000 te 4000 F : | g Complementary to the cerr051on test program in llquld metals NAA is carryxng out ten31le strength creep, and fatlgue testscnlmolybdenum} tantalumg' ' and tungsten at temperatures up to 2000 C. "Data on propertles of graphlte at th1s temperature are already avallable at NAA from werkunderanother centract | BELIUMfCYCLE-SNALYSIS A fa1rly detalied ana1y51s of a nuclear powered alrcraft” based on a. .?closed cycle helium- cooled system, "has been completed and a. flnai report is "5be1ng prepared Thls analys1s has shewn that the helium cycle is inefficient '_jheceuse§ 64 clalmed 1n WASH-24). Cr1t1cal -mass con51deratlons may requlre the 1ntroductlon of regions of slow- _ mevxng dense ‘water as reported in WASH 24; some 1mprevement can probably be | _made 1n reduczng the ameunt of heat transfer surface wasted 1n centalning the, dense reglonsa The use of central entrance arrangements to beost average ) water den31ty was conszdered brlefiy but dld not appeer te be Very premlSlng Study ef Fiow Instehxlitya When several heated stream passagee are cen;e nected in parallel between common header53 it is pos sible for a type of flow 1nstab111ty to arise if the heated fluld ex?ands markedly w1th heat addition (cf. Hanford "belllng dxeease") Under certaln condltlens ifa smali dlsturbm__' ’ance tendstx)block the ‘low in a passage and If the heat 1nput to that passage 13 not reduced it is poss1b1e for the decreased flow rate to require en' anreesed pressure drcpg w1th the flow then belng further choked. Thus if :héi_ - mass. flow rate is decreased with. heat 1nput malntalned the coelant den51tyvf‘ will be decreased and the ve1001ty increased; it is p0551b1e, in the case of an expans1ble fiuld for the increased velocity to more than balance the dew - crease in mass flow and glve a net 1nerease in pressure drop necessary to malntaln the flewq'”' An analysas of the stralght through fiew case shewed that a 1% decrease in mass velocity would give a 0.4% decrease in pressure drop, .compared to thee .8%.decrease.fqr_the incompressible fluid. Thus the straight-through flow 7_313' YL L4 [ R I R e . RIEC T : . LT LY PR e CLme : S S R D . cenoes o . YY) oso0es . : Cesesee senaea 'system is mllély stable, but w1th a not very strong restorlng force :715 iSf-“j' 'uyerhaps somewhat surprlslng that the stralght through casels 1nherentiy stableer-f Cin v1ew of the large expanSIOn undergone by the watero The sltuatlon 13 helpedf" f“ by the fact that most of the expans1on takes place in the 1nterlor of the flow{_"flu "path wzth densxty changlng comparatlvely slowly thh respect to temperature;}'fie" r.near ‘the ex1t end The flow stablllty is less favorable for multlpass systems Cin whlch a. flow passage mlght haVe 1ts ex1t at a temperature 1n the reglon of Tf fe'hzgh expanszblllty o In any event 1f thls type of flcw 1nstab111ty shauld appearg it can beieb . f cured by 1ntroduc1ng erlflces at the entrances of the flow channels ‘the re»-e .3qu1red pressure drop across the orlflces would not seem to be large enough to_e wbe obgectzenable in any. cases of 1nterest Fcr a sharp edged orzfzce,-a 1% fdecrease in flaw rate would glve ‘a 2% decrease 1n pressure drop, and it 15;” "~e_thus merely necessary to. provzde an’ orlflce w1th equlllhrlum pressure drop?-5°' '~Liarge enough to make the overail (orlfice plus flow channei) changeaxlpressurel" '_f;drop negat1ve fsr a decrease 1n flow rate° Review af Eeat Transfer Data, Heat transfer to snpercrltlcal water can ';be expected tn foiiew a ccnventlonal Colburn type correlatlong although theree:' _Vare no. known data fer the pressures and flow rates under can51&eratzcng . Un- efiertaxntles:nipredlctxng heat- transfer coefflczents arlse from the wlde var:‘.afl_'.'_:i :?tlens of fiuld properties wzth temperature ané frem doubt -as’ to gust which- 'e temperature should 5e used in evaluatlng fllm prcpertlee_for large fllm dreps“lfe' The thermai conduct1v1ty and v1sc051ty depend markediy on temgerature,:ff-“ "hoth decrease w1th temperature when the water hae lxquzd characterlatxcs,_end'- :"both go thraugh mlnlma ané 1ncrease w1th temperature when the water has vapor“ '-,characterlstlcs Slnce there are few data near. and beyend the minima, extra- "pelatlon beyoné the mlnlma 1s Bncertalnu ~ The. specxflc heat has a sharp peak - at temperatures clese to the quasx phase cbange Of the water,_ but rellable- -Nspe81flc heat data are avallabiea R A aurve of the praperty dependent portlon of the heat tranafer correla» ---' _tlon as a functlon of" fllm temperature Was obtalned by extrapolatlng data onm' : beth tfiermal eonductlv1ty and v18c031ty The iatest heafi transfer correlatlonsfl' ':wlth hlgh pressurecs} are. glven for transpcrt propertles evaluated at an . *It is understoo& that the NACA Lewis Laboratory w111 extend its program on heat transfer to steanzta '_ - 1nclude pressures in the nelghborhocd of 5000 ps1a _(3)1 McAdams,'W- Hk, Kennel W: E and A&domS, J. N "‘Héet Transfer to Supefheafied Steamfat High.: P:essgres;*-ASHE_i}ans ?2 421 (1950). Y ade 4 & CR €8 . es s eae 9 goRre AW T ele, A JOR J 5K N J B % @ & @ e e e e as el & & L] LR ) . E . [ I T ) @ U # ass RAL ERTERTE U SO S SR LI T e PR SRR T ARV TN 1 Ve I TRY Ny B DR R el SEORL T X SR e X 48 B e pRe 69 AN faverage temperature and are satlsfactory for fllm drops up to 260 F The fllmf;fl: drops in the supercrltlcai water reactor d351gn9 hgwever are af the orde: of“V " .'500°F and uncertalntles of the order of 100 F maY therefcre be expacted 1n;f *i7 " 1est1mates of‘maximum wall temperature ”35 3E§cTi¥rTY¥ Estxmat1en cf the react1v1ty cf reactors contalnlng 1ron and water is: atf_;”" present a dlfflcult preblem because there are 1nsuff1¢1ent experlmental datao f.T ';jAn agpraach to the prob}em is to use. a procedure whlch would be SH€CQSDffil 1n:f jfthe case of plazn water reactors (for whzch there are'»ritlcai mags measuredJ’ - ments such as those glven in, report K- 343“}) mxitoutlllze only mlcrascopch_ ferass sectlans of fihe reactcr 1ngredlants When deveioped such & procedfireff° “5 _cauId be extended ta metai water mzxtures by the use of. mlcroscagzc data. for '-'fthe metai companefit Therefere the analy51s has proceeded u31ng the apprfiach 'L ~ fbasad en mzarosceplc data enly Eqaivalent Water Reactors for Irenmfiater &1xtures A procedure has been'-;3”7 *_developed whereby ‘the densxty of the 1ron 1n ‘an 1ron water mlxture (thh re- ';spect to zhe seatfierlng power of tha mlxture for neutrans of hlgh eaergy) may w':"_;"be expressed in terms of an equlvalent de331ty of plaln waterq_ This is. feau "331bie sxnce tha scatterlng Pewer sf metals (ether than hydrogen) COmPareS L Tl . H:favarabiy thh that of an equal valume ef water at: thfiuhlgh,energles where :_ '"]much ef the neutrou leakage tends to accur Tabie 18. 1. summarzzes the cal»_f1 T =cu1ated eqazvakent &efi51ty af 1ran as a functlon of the 1r0n towwater ratlo of': ' i ffthe mmxture TABLE 18 1 IBONQTB»?ATER VOLHME EATIO ’;f{i EQUIVALEVT TRON DEKSITY = R D i "'_f' i _ ‘g/°°} ; ‘ 00, 45_' o 0.60 - o D42 B £ RS R 103 8 ..(4),‘B§fl§; C. K ) Calllhan D Marf:tt J.‘W;, éhé Muff&?, R;'L;; Critical Hass Sindia#; Part IIf;}xl - Cal, K25 Repc t R 343 (A@: 1 1949) i In the calcalatlons thus far the 1ron slow1ng dewn (by 1nelast1c c0111§fr sxon) has been neglected Perturbatzon methods are now being developed to 1nwfi flclude the benef1c1al effect of the 1nelast1c slow1ng down by iron. The only{ result avallable as yet adds Q 11 to the equlvalent iron denslty of 0 41 (last i hne of Table 13 1) | . P Suppose that 1ron 1s, 1n fact equlvalent to water of some den31ty x 1nf' preventlng escape of neutrons whlle slowxng down. Then9 so far as neutron ~ 'slow1ng down is concerned a mixture of 1 volume of iron and z volumes of .water af dens:ty P is equlvalent to water of densxty b = (x + pz)/(l + z) (an the other hand 1ron (1n the form of stainless steel) absorbs (slow)' ~neutrflns ‘about 11 times as strongly as an equal volume of watero' Thus, as '_regards neutron capture, the same mlxture 1s equlvaient to water of dens1ty- (11 + pz)/(l + z): By means of these two "equlvalent den31t1es " a and b, -known results on ali water reactors can be transformed 1nto results for water~ flron reactorsy. _ gesults of Reactlvxty Studies By use of the equ1valent dens1tles de- f1neé above, ‘the volume and. fuel requlrements for any iron- -water reactor may be calculated ~ Let ¥V and My denote, respectlvely, the volume and fuel re- qulrements for a glven water reactoraf BReducing the density of water in this 'reacter from l ta b requlres 1ncrea51ng ‘the valume by a factcr 1/b3-and the mass cf fuel by a factor 1/52 S1nce ‘the slow- neutron leakage in water re- actors is relat1vely small " the effect of multlplylng the absorpt1on of the. :water in the cere by e/b can be taken 1nto account approxxmateiy1nfmult1ply1ng - the amount ‘of fuel by the same factor,_ Thus the final result is a reactor '”'hav1ng a- volume V = V /63 ané fuel mass M =M a/b3 Acurve of My vs. V, for_' flspherlcal water cores surrounded by infinite water “eflectors has beén ob- “tained.** Compar1a0n<5} w1th crltlcal mass expfirlments(4) lndlcates that 'the curve 1s censervatlve in that 1t overestlmates the fuel requlrement by - .‘ up t(} 25%, | : _ _ : . . : ' Consxderlng an example whlch roughly approx1mates the Wllllams reactor, mthe fuel requlrement M, for a water reactor hav1ng aV, of 132 liters is found***tokm 2.46 kg : Assumlng the equlvalent d6331ty of the iron to be 0.5 and that of the water to be also 0.5, 27 kg of fuel is requlred ' ._The resplt'just found'is for a spherical core of,total volumeKV': 1056 litérs'and cofitaining 88 liters of irdn' the balance of the core volume and ‘Actuaily, the 1ncreased water and fuel absorption will act to slightly reduce the leakage. The ~above procedure is thus conservative in"that it overestimates the fuel requirement somewhat. o e*The comparison between Greu%hfig s computatlon for spheres and the’ experlments on right cylinders was effected by comparing the U density in the experimental assemblies with that in a sphere havzng the same core buckling. (5) Nuclear Develapment Assocxates, Inc , Ptle Stzes and Fuel Requzrexsnts, C1rc1e II (Jan 7 1949) #4+% There is, of course, ‘no pretenszon that the crlsxcal masses are known to anyihifig lxke t%c uamber_ of significant figures given in Table 18.2; we would, in fact, be happy if we could believe them ‘to within, say, 20%. Also, it should be neted that no allowance for depletzon, xenon, ‘and other; leSOflS etc. has been 4 @ncludad here oo oo :-vv.:--o" o : - ) . ' o. l .__ :‘.__n . S e S ea ee e ..335 REE e s o e PN 0o . © . .0e8 8 9 co e . & ® PP . v oppc (X X 1] the surroundlng 1nf1n1te reflector are composed of half dens1ty water Upon :vary1ng the core voluum F’ wes1m1lar1y obtaln the resulta glven in Table 18 2 TABLE 18 2 Fuei Bequirement for Iron %ater Beactors-'. o (with 38 lite:s cof i:ang ; = 9=5'and o =0.5) - V '{voi; of hal_fndensit..y' 'watzer? ST - {liters} {2 ¥ (kg) S _'L vol. of iren . ] : o 176 EEEEE R 1 _ s 44;6'. T 24 b 2z s 440 4 - 29.3 (TSN o 2.8 1056 [ e oo b ;e e | 19 4 28,5 ',_These computatlons suggest that, w1th a prescrzbed amount of iron and hence a prescrlbed total heat- transfer surface in the reflector, there is cons:derable- leeway afforded 1n the d651gn as regards reactor size. S1nce the smaller re- “actors would have Iower sh1eld walghts and hence lower power requxrementsi'the _ " amount of 1ron and hence the fuel requlrement would be expected to be rela— - tlvely less in the smaller reactor sizes than is 1nd1cated in the table,_ On the ether hand the uncertaxnt1es1n the above computatzons, and the sen51t1ve~_ o 'ness of the fuel requlrement to changes in P and x 'wlll increase with the fi;fractmonal 1ron content of the corea- The above con51derat10ns have all env1saged uniform composxtlon through-m ';fout the reactor, Actually, the water densxty will change along a coellng lstream, as d1scussed above in the section on heat transfer and fluid flow‘ Only one prellmlnary calculatxon has yet been made on ‘the effect of space *varlatlon w:thxn the reactor. 'This was a rough one-group treatment of a case__ .1n which the water density varied as it would 1n flowing stralght through the 'reaéidr._ Apparently the react1v1ty of such a reactor can be higher than if 'the water were eVerywhere at its average den31ty, on the other hand, the power o generation will be more peaked than in the ‘uniform- densxty case, whzch is a d1sa6vantage from the standpo1nt of heat removal *Agaln,'anly prellmlnary calculatlons havg been made on-the yariatibgs.of' reactor reactivity with water density, neutron temperature, and other such e LT XS e @ L3 FY R L R Rt Y T R Y'Y ¥ i X TR e e 0 e 8 LI e L. eE e e e e TR R TR S SEEEE TIE TR 6 e e e Ge a8 el e e & e ees e e s e e 8. ¥ e e & 8 W ete e [ Y LR SR S L Sea. -8 B8O 8.4 B0 08 L E 9 v e0e. o quantltzes that enter lnto control con51derat10ns.. Flrst ruundmestlmates 1nd1-;_' 'cate ‘that the rate of change of react1v1ty with re8pect to (average) waterfg :den31ty,_dR/dp, 1s about 1/3 for a reactor hav1ng 10 parts of half éen51ty_' water to 1 part of iron. by volume,'“' BYNAMICS AND CGNTBOLS T%e study ofreactordynamlcs and controlshascmly recently been 1n1tlatedf and few conclu31ons have been reached although the nature of the analysis has-' ;:.been determlned -~ In order to obtaln results thhout exce351ve work and ane _;undue amount of. tlme, the time behav1or of the reactor is now?* belng analyzed | by perturbatlon studies’ ef small dlsplacements from equlllbrlum, This study_' ':_.w111 adequately determlne the fundamental stablllty of the system although 1te._v "';'wlll not cen31der condltxons 1nvolv1ng large changes from equllxbrlum values. : 1 the reactor turns out to be stable and well behaved fcr short tzmes; | the behav1or w111 then be con81dered for longer time 1ntervals (over many secends) in which the coupllng to the external circult and to any controls becemes 1mportantt The questloBS;of both acc1dents and start- -up have been consxdered to some extent, but further c0n31derat10n is de31rabie befere detazled conclus;sns can “be reached ‘A method has been dlscussed whlch may lead to a feasible, even 1f' tlme consumlng and complzcated start up procedure.' PROPERTIES GF MA?ERIALS | The presence of radlatlon 1ntroduces some doubt into what would otherw1se | be a rather satzsfactory materlals 31tuat10n, In the absence of radlatloni effects, 347 stalnless steel can be used in contact with. steaniatISOO F for _ several hundred ‘hours w1thout showing apprec1able corr051on ‘and 25-20 stain-- 'less steel can be used at temperatures up to about 1750°F. Cases of success-i ful cperataon at condltlons similar to those contemplate& here are common?* and there is, of course, a vast wealth of general background in alloy steel technology - With rad1at1on present, nevertheless the materials situation is | not unllke that for most other reactor materlals, i.e., radiation-damage ex- perlments and heat transfer tests will be needed (cf. section onheattranefer)-f and (because of the many incentives for pushing temperatures higher) attempts must be made to determlne practlcal operatlng limits. *At s later stage, when actual bu11d1ng'0f the reactor is contemplated, and detalled desxfinlsbelng carried out, a more thorough analysis should be performed with the aid of one of the high-speed digital computerse ~ **For example, sorfe experlmental supercrltlcal pressure bevlers have been erated in: tnxs country.” (Purdue} and in’Europe. Central station beilers are built {Babcock and Wag Zears with exit Gteam at 11000F (wall temperature about 1150°F). The Bureae of Mlnes Flscher T}epsch ydrogenatlon plant operates at pressnres up to 1, 300 psa. : _ P sy oo o Q.. . Qoo .0 e L ’. €3 %}. D B T 2 : v ¥ e a2 A w0 ¥ . - g. 05 * : o o Lle : s A o L ‘oeese ' > o e ee ssoe ; ..4. dasee ..... cox) to operate for 25 S Embrlttlement and corr031an, whlch are the prlmary materlals prohlems,. .fif depend markedly on the purlty of the water, whxchshould be . degassed and should£j f 'have less than 1 ppm of dlssolved solldss and the desxgn of the system=- Th¢“ i347 stalnless steel should be as low as p0531b1e in. ‘carbon and sulfur,-an& fl - ' ;str1ngent technlques are prescrlbed for plcklxng, anneallng,_weldlngi etc. zaNo castxngs should be used m-only forglngsa Embrlttlement of Steel by Atomis Hydrogen._ The embrittlement and flssur~ 5”‘” f}lng of steei by d1f£u31on of atomic hydragen 1nto the steel appears to be the 'fba51c troubi%«cau81ng mechanxsm.; Radiatlcn bombardment of the water in- the; 'supercrxtlcal water reactor may 1ncrease the atcmxc hydregen concentratlon.: " The agency of embrlttlement appears to be mechanlcal free hydregen atomsfx' ':dxffuse readlly along the grazn bcundarxes and elther form bonds w1th 1m-_' '5f pur1txes in. the steel (espeC1ally carbon, whlch is canverted to methane) or - -recemb:ne 1ntc H molecules (The amount of atomlc hydregen whlch can das—f 3selve xn steel is small ) ngh pressures are generated in rxfts ‘and valds.f' The: steei is Weakened by farmatlon of actual flssures an& is: embrlttle& by the ]bulld—up of hlgh 1nternai trzax1al stresses and deformatzon nf its fxne struc-._ ' [itare.; Thls concEpt of embrzttlement has been éeveloped thoroughly 1n recent: ‘”?_years by Zapffe (man? papers 1n Metals and Allays, Eetals Pragress, etc ) andff. “-ethers, and the ol& ‘concept of embrlttlement by hydrl&e formatzen seems . tolff Jhave faiien into dlsrepute.-ff“ Fartunately, 34? staznless steel 1s nct susceptlbie to the hydregen 7F *effects (in the absence of 1rradlatlon) sznce a pretectlve surface film of -.fi’chromzc zede recomblnes the free hy&rogen.. There are,_however,_few &ata on ': ;the temp&rature dependence of thls surface effect. It has been suggested that '_j hydrogen mlght be bled out of the system by the use of & yalladlum plug, smnce_f.f '-{pailadzum is aimost transparent to hydrogen. flarrasion uf Steel in Water.f Corraszon rates of kess than 0 GGGE 1n, u' :penetratlan are reported for 347 staanless steel at temperatures up’ to 1500°F 'and for 25- 2@ staxnless steel at temyeratures up to 1759 Fw1thout1rradxatlon.f“ Even wlth these small corroslcn rates of the nonradlatlon case, hcwever, it 4 appears that enough radloactlve iron may get into’ the ceelant stream to: pres:s' ~clude access to the water handlzng machlnery for malntenance, 1nspect10n,f etc.(8) Furthermere, dlssolved materxal wzll tend to be deposxted out on thez- reactor fuel plates where solub;llty 1nver310n occurs ' Thls takes place atfg (6} Lexm ton Pfo sct Nuciear Powered Flzght A Repori to the Atomc Energy Commsszors, LexP 1 %30 (Sfipt; 3o, 19a8). e e T o FY XY T T Sam "h%e - .'.oo.'oto.;-. - L ] wasees 0. oWy L SR 1 e wF @ + : Ceeele * ‘% - : .. : ATIRE A T o & ® & . L : : 30 ga L eew e Bge oo 4% oe a8 'the expan51on of the water from llquld characterlstlcs to vapor characterls~” 'ntlcs, The experlence of Rohsenow and Clark at M.I. T with'a hlghstemperatur&~‘ rig w1th corrosion sources (e g., cast ‘pump rotor) 1nd1cate Just what can: ___happen, an. ox1de layer that was depos1ted in the phase-change reg1cn in- 'f ;fcreased the f1lm &rop 50% in 2 hr,m h e e e e | 320 L te B ’ L] L) T YYY Y R Yy R SOSRue [ ) EYTLLES 206089 o o XX Y se L R XY K essaen ABe BN » - * PR ® '“f”fi9 SUPEESBNIC TUG TGW SYSTEM : C‘ B Ellls, ANP DlVlSlon _ Of the varlous pess1bllxt1es for a supersenlc alrcraft whlch can béf 'Esforeseen at the present tlme “the ‘tug-tow system stands out as demandxngfffi”' ]sre1at1vely smali alrcraft 'and therefore, a con51derably 51mp1er reactor and;ffi : Jsenglne than any etker system The pr1nc1pai burden of deubt regardlng thls”? " system1s.thrown on tne questlon of dynamlc stablllty of tug tow at- trans sonlc" .f'and supersonlc Siem“ AR Burlng the ras Gpcrter seme prellmlnary axpioratlon on the welghts cff .snxtable shleldsgjseabtors,_and engines have been carrxed out by Oak Rldgesf ' Nat1onsl Laboraturys VEFA :and Nuciear Development Assoczatess_lnc;l Some cf,f . the shleld welght calculatlens have been descrlbed in’ Sec 10. Varlous &eslgns; | f:for the tug plane have been sketcheé by NEPA and 11ft to- drag and gross welghtm .ssvalues estimated The present estlmste at the Oak Hldga Natlonal Laberatoryfs_ 'sizs that gross wexghts of the tug. and the tew planes w111 be approx1matel¥ii7 ” lii;lfi@ 600 and 20, 600 1b, _reSpectlvely Furthsrmore,.the shleld wezghts are :__ixkely to be smail enough so that the determlnlng factar in the welght of ‘the a{tug wxll probably be the welght of the eng1nes and plumhzng, rather than thsf ff_shxeld _ It is belleved that ealculatlons on the dynamlc stab111ty of the asupiede fTsystem cauld he reasonably trustworthy A determlnatzan of the numb&r ‘and’ *-Lplacement of englnes will have to be made sa as to fix the moments af 1nsrt137' fifs.af the two cr&ft befcre such calculatxans would be possxéle,;;,'5 PER L X SN T P & 2. A0 A BN B 828808 O T S R s R A e R R N e TR RSN R Y IR ISP EIE S JREAREE IR N ¥ S RELY SN .. e e e e e el ¢ E T 4 LN R I RRY TR SRS YRR TR AOpar by Y i) B IR T SROTi el TN T ) KRR T B SR T LR IR SR X AT Y SN N B & AEO B APPENDIXES . 1 s : R B S P X T THeBOO . e 80 a88R0e L govene o & 8 e ”'21 ANALYTICAL CHEMISTRY C— fl Susano Analyt1ca1 Chemlstry D1v1510n : .“'Rééulté-df'defiélopmefitIW6rk:on'éhalytical fiethddsref intefést to thé-ANPf. \\\\ ‘pation of ongen in metalllc sodlum ‘The. new method is based on. the fact that-'"' _ alkyl halldes react readzly w1th sodlum to produce metalllc halldes,_but aref practzcally inert to sodlum oxlde." kD Also under con51derat10n are methods for detectlng and estlmatlng small, _amonnts of oxygen ‘and nztrogen in inert gases (hellum and argon} A colorl- metr1c metho& for oxygen appears prom1slng, for nltrogen, an 1sotope dilution 'fmethod of analys1s, us1ng le as tracers is tc be 1nvest1gated Other studles 1nclude tests ef the stablllty of Dow Cornlng Slllcone 011 550 at hlgher temperatures (up to 1200°F) the chemlcal nature of a sublxmate' from a eutectlc mixture of sodium and uranlum fluorldes,'and statistical ‘analyses of spectrographlc test results for minor components in sodium,.the 'obgectxve belng to determlne the standard error at varlous ccncentratxon_- 'levels and to 3351st ‘in 1mprov1ng methods A total cf 427 sampies of interest to the Beactor Development Program - (1nc}ud1ng NEPA work) were analyzed durlng this perlod - 0f these, samples of meta111c sodzum derived from corr051on test work comprxsed the majority of_ ':samples analyzed for the ANP Progectn FUSIOH mixtures of alkali metal and- :uranzum fluorldes and of alkali metal, uranlum, and berylllum flucrldes, were anaiyzed in cen51derable number,3as were metals, alioys,_and bcron carbide. | liThe remalnlng samples were quite varlefizjxnature, often: requlrlng con31derab}e development work to estabilsh satlsfactory analtylcal procedures: DETEBMIN&TIGN OF OXYGEN IN SODIUM J C Whlte and W, J. Ross, Analytxcal Chemlstry D1v151on ‘ As prev1ously reported efforts are being made to develop a method forg- the determlnatlon of oxygen in sodium which w1ll permit a spectrographzcr ,analys1s on the same sample. It was hoped, also, that any such method would have certain other advafitages over the currently used Pepkow1tz and Judd Procedure (1’-. e e (1} ?; §0?1t§,)L P , and Judd W C.,"Determlnatlon of Sodium Monoxide in Sodium,” Anal. Chem. 22, 1950 ' - - 336 svaese. S ® ® sensse seveew XN X.T o e’ g » davses : [ oeew ® s k2 Durzng the znltlal phases of thls study, a large number of organlc com- pounds were 1nvest1gated in a search for a reagent ‘which. would take so&lumff:: into solut1on Wlthout dlssolv1ng or reacting with sodlum ox1de CAlL thesegf; vmater1a1s proved to be unsu1table for one or more reasons. Ethylacetoacetate “and- acetonylacetone at fzrst appeared promising but were later dlscarded T : partly because of the excessive react1v1ty they showed toward sodlum ox1de n- Butyl Bromlde Method Recentlyzamore frultful approach to thls problem: -has been dlscovered whlch depends on the well known Wurtz reactxon The, general reactlon is RX + 'smg +'-a"x s R;R'_'} + 2NaX .”It was felt that the reactlon mlght be useful as the baSIS for a method if an'. "excess_of-hallde were employgd? whlch is the reverse of the usual procedure, 'It_%és aisdlfdufid thgg.£he react;cn_rate could be controlled by the adéltlon 'df“an.ihért‘ailuént ib'thévreagent.. Al tol mlxture of n- butyl brom1de and hexane was selected as’ optlmum for pre11m1nary tests The effect of'n butyl bromxde on sod1um oxide was thea 1nvest1gated A 'fiwelghed amount of Na 0 was 1ntroduceé into the reagent (1 to 1 n-butyl bromide "and hexane) and the mixture was refluxed at 65 to 70 C for various periods ‘ef tlme._ Water was then added to the mlxture to dzssolve ‘the monoxide, and the total ba51c1ty of the resultlng two- phase system was determined by tltraw tion. The results of these experlments 1ndlcate that after 24 hr there is no " loss- of basxclty,‘after 36 hr there is a barely detectab}e loss, and after 72 hr there is a sllght loss whlch 1s, however, of questlenable sxgnlfzcance 'In __actual practlce,_however,_a contact time of more than 24 hr is nelther neces- _sary nor de31rab1e, In the actual determlnatzan the sodlum sample is placed in the reagent; mlxture, ‘which rapzdly attacks it as evidenced by formation of a blue salt._ As fresh sodium is exposed the react:on proceeds with greater vigor, especially when the temperature is raised to about 60°C. The reactlon is generally 'complete 1n 2 to 4 hr, although it is usually more convenient to allow the mixture to stand overnlght. After completion of the reaction, the 1nsoluble sodxum_bromlde and‘monox;dé are dissolved in watér; and the solution 1s titrated with dilute acid to a potentiometric end point. A bromide titration e moe @ 0 - L] LX) ek B0e ¢ G60 GO X ¢ e . s ¢ 86 & @& @ e e e e . : e © S0 & . e s e . & s 08 @ 86 L3 IR ) S I IO - . s se e 8 & © o @ L ; e el . e 8. @& L e o0 9 LK EREIIE T | 1 L} o ® sa® ¢ & 068 868 06 @ e aed 69 : ’”fon an allquot of the water solutlon allows calculat1on of the sample welght-w'” i (tatal sodlum) and the rest of the solutlon is processed for spectrographld g-ana1y51s Sampllng Technxques Several samp11ng technlques have been employed in an effort to abtaln standard samples Accordlng to one procedure sodium was-' ';melted under an atmosphere of argon and the molten metal was drawn 1nto_ ..évacuated glass bulbs The oxygen content of sodium sampleé in such a manner :?has been found to range between 0. 02 and 0. 05% as determlned by the Pepkow1tz i fand Judd method ' The unlformlty of such samples is usually questlonable ' becauSe of p0581ble segregatlon of the Na 0 in the sodium: Another technique of sampllng whlch prov1des more unlform samples is the double dlst1llat10n of .;sodlum 1nto glass hulbs under vacuum These results aiso range between 0. 02 and 0 05% A br1ck of bulk sodlum; furn1shed by the ANP Experlmental Englneerlng‘ g'Group, was sampled and analyzed for oxygen by the n- butyl bromlde method_ ;One fourfih inch slices were cut from the brlck and samples were taken from each sllcea Results obtalned to date are given in Table 21. 1 CTABLE 21.1 'Determination of nygen 1n Bulk Sadlum by ne Butyi Bramide Methoé . o i I OX?GEN} (%) G | SODILM SAMPLE SAMPLE 1 | SAMPLE 2 | SAMPLE 3 | SAMPLE 4 |First layer 0,033 -l o0.e3¢. | 0.033 f 0.033 |Second layer . 0.046 1 o0.047 . 0.048 0.048 | Third layer - .~ 0.055 .| o0.048 .| 0.053 S | Fourth layer ~ | 0.073 | 0.070° . | 0.081 ‘ Judglng from the prec151on of the test results on samples from any one -flayer on the brlck of bulk sodium, it appears that the n- butyl bromide methed'- is’ capable of satlsfactory pre0131on Varlatlens in test results on standard, samples prepared by other methods are 9robably due to segregatlon and’ non- uniform dlstr1but10n of sodlum oxide from sample to sample, even though such .samples recelve the same treatment.andare suppcéeély 1dent1cal:n1all respects 338 o LB g ‘esesce’ s8088s saepese | sesep e - . . ° 2 YYY S 'éoncifisidfi§, The results S0 far 1nd1cate that the n- butyl bromlde method s sultable fcr the determlnatlon of oxygen in sod1um and that 1t possesses‘ slmpertant advantages over the Pepkow1tz and Judd method among whlch are ;(1) an oxygen determlnatlon and spectrographlc analy51s can. both be made on " the same sample, (2) less time per determlnatlon 1s requlred (3) less Sklll:. “and Judgment are requlred on. the part of the operator, and (4) the apparatuss Cis much 31mp1er and less fraglle Developmental work on thls methea i3 contlnulng DETERMIN&TION GF QXYGEN IN AQGON AND HELIUM J C Whlte and J B Lun& Analytlcal Chemlstry D1v1s1on: _ The ‘amount of oxygen in argon or hellum is generally qulte small but some questxon has been ralsed.asto the unxformlty of these gases from cyllnder to cyllnder with regard. to purlty It is contemplated that these gases w111 'zto used as inert blankets in contact w1th llquld alkali metsls, and since 1t ~'Is known that as few as 10 ppm of oxygen will prove deleterious, it appears l_deslrable to. resort to some purlflcatlon of the gases before use. Thls pur1f1cat10n, as planned by the ANP Experlmental Englneer1ng Group, "7w”_con31sts in passing the gas through heated copper ‘oxide, titanium turnxngs —heated to 800 C sodlum patass;um alloy (Na- K), end, fznally,a.llquld -air trap. 'Sueh drastlc treatment will, presumably,'reduce the oxygen ccntent to far less-_' 'ethan the mlnlmam tolerable amount. The eff1c1ency and service llfe of thls _fltraln w111 be determ1ne& by testlng the gas fer oxygen after varlous 1ntervals 'eef tlme up to 1900 hr of operatlon“ Brady Methfld A method &escrlbed by Brady‘z) for the determlnatlon of 'exygen in’ gases, which is based on the ox1dat10n of sodium anthraqulnone B- s*sulfcnate from the red reduced form to the colorless oxldlzed form, has been _:applled to the analyses of hellum and argon. The reduction in colo: intensity A_of the sodlum anthraquinone B- sulfonate solutlon which is a funetiefi of the . oxygen content of the gas, is measured spectrophotometrlcally Iflvestigations' ‘are anderway toestab11sh the 1ower llmlt of detection affordedinrthls method The first determlnatlons were made dlrectly on cylzn&ers of argon an& helium. The results are glven 1n Table 21 2 (2) Brady, L J., “Determinetiefi of Small Ameunts of Oxygen in Gases,” 4nal. Chee, 20,_ 1033 (1948). 339" - scovom . e L seoeue . ‘seseed [ TEXEE ) . S L seanse TABLE 21 2 Besults of Betermlnatlons af fixygen in Heliue;énd'fifgen:-' o GAs - © CYLINDER - | OXYGEN (ppm} | | Helium 2 5T *Argon 9 R T SR The results demonstrate the w1de varlance from tank to tank w1th regard to oxygen content Plans are also underway to determxne n1trogen in hellum '°_and argon by means of lsotoplc dllutlon ana1y31s; using N 5 as a tracer. This _ -?ork w1ll be carr;ed eut_xneceoperatlon w1th_the_¥ 12,Assay Laboratoryu ?fiERfiAL ST&BILITY GF BOW*CQRNING SILIC@NE OIL 559 J C Whlteg Analytlcal Chemlstry D1v1510n | ' A study of the stablllty of Dow Cernlng Slllcone Gil 550 at elevated'. '.temperatures,'alene and when in contact with sodlumg has been prev1ously reported (ORNL 919 p. 260) The 011 polymerlzed at 909 to 1050°F in the,e 'presence of sodlum? but llttle change was noted at a temgerature of 1000° F"' _when sedlum was not present Further tests were recently made to determzne the thermal stab111ty of- efthe oil. In' one case the 011 was heated under helium and maintained at 1200 F for 15 min. Upon remeval from the furnace it was found that about 20% Qf the - Ilquld had been lost by evaporatlon and approx1mately 49% of the remalnlng fluid had polymerlzed into a dark- brown, hard, brlttle mass . In a second trial, the oil, sealed in a quartz ‘tube under argon and heeted to 1200° F, was 1fi'ebserved to un&ergo polymerlzatlon after 15 to 20 m1n at thxs temperature These results 1ndlcate that DC 550 is thermally unstahle at temperatures ~above 1200°F because of polymerization, even_when not in centact with sodium. 340 EXZY XY EEETEXN) [TEEEX ] s 8.8 i » _.qcnvo'_ | CONTROL PROGRAM FOR SPECTROGRAPHIC DETERMINATION OF TRACE METALS IN SODIUM B L. McCutchen, Analytical Chemistry Division The qua11ty control program assac1ated W1th the spectrographlc anaLy51s of metalllc sodlum has had two obJectlves (1) to determlne the preczslon of_i 'the test results9 at varlous concentratlon levels for 1ron9.nlckel chromlums;' and- manganese, and (2) to ass1st in a study aimed at 1mprovement of the pre« o c1s1on of the poreus cup method of spectrographlc analyszs The standard error at a 99 5% confldence level for the average of faur ' B determlnatlons was found to be essentlally the same (approx1mately 20%) for the four constltuents at the three concentratlon levels which were investi- = '-;,gated ThlS means that 99 3% of. all averages of. four determlnatlons for one 'constltuent in the sample will be w1th1n +20% of the amount of the constltuent_ actually present It has been establlshed as standard pracedure that 311 Qfdetermlnatlons be made in quadrupllcate It is belleved that the prec131on of the test results may be 1mproved by - '"zchanglng the 1nternal standard from platlnum to vanadlum and by carefully: 'Qcontrollxng the process1ng condltlons for the spectrograph1c platesa An 5,1nvest1gat10n of the use of vanadium lnstead of platlnflm as the 1nterna1 'a_standard is’ now hezng carrled out. -aSERViCE Awanysgs-- - During thls perlod the analytlcal work carrled out in support of the ANPf” "program censlsted chlefly in the determlnatlon of minor impurities in 11qu1d" _metals (coolants) 1n uranlum compounds and mlxturesj as well as in structuralf' ' materzais and in unrela*ed mlscelianecus samples Analysis of Sadium (J M Peele and L H Jenkzns} ‘Minor metallic 1mpurlt1es in sodlum metal were determlned Spectrographlcally and the oxygen* content was determlned by the chem1cal method of Pepkcwztz and Judd. (1} The_ more. recent ‘sodium samples have required a wider variety of tests than in the fpastg The determlnatlons of lead, copper, and cobalt are often requested requiring the preparatlon of spectrographlc standard so}utlons contalnlng_ these elements 341 esoe e : : Abee re vabWeS LX) amese el YT Tpeweee 0] L ° Although the porous cup technlque for the determlnatzon of the mlnor _‘“ metalllc components of. sodlum was applled w1thout 51gn1f1cant change durlng_ 'thls perlod ‘the results of a few prellmlnary tests 1ndlcate ‘an’ 1ncrease in preCISxon when vanadlum 18 substltuted for platlnum as an 1nternal standard It further tests confirm these flndlngss'and the presence of vanadlum is found; __ to cause no 1nterference in the chemlcal preparatzon of the sample, the use of '”platlnum as an 1nternal standard may be dxscontlnued and vanadlum substltuted[” ‘for thxs purpose Aualysis of Lead and B1smuth (J M. Peele) 'Sémpies 6f‘lead'éfid of ’“ 51smuth metal after use in corroszon tests9 were analyzed for m1nor com~ Sponentsn Practlcaliy the same. spectrographlc technlque was used in these. ;determlnatzens as is used for ‘the - analyszs of sodluma_The chemzcal preparatlon 9£ the blsmuth and leaé samples dlffers from- that empioyed in the case of : sadlum, howeversislnce the mlnor 1mpur1t1es aredlssolved along with the matrix materlal whlle in the preparatlen of sodlum samples the sodium is separated" 'frcm the mlnor 1mpur1t1es by dlssolution and prec1p1tatlon methedsg Most of the lead and bismuth samples were fcun& to contain relatlvely-' Iarge amounts of carr951on productsj Spectrographlc standards for use in .*carrylng out these analyses were prepared in such a manner ‘that the concen- tratlons of lead and blsmuth were approx1mateiy equlvaient to those in the _'samples to be tested | Analysis of Uranium Comgonnds ané Mixtures (E C Lynn and L J Brady) A varlety of uranium- bear1ng materlals conslstlng, for the most part- of ~ fusion products of alkall bases and salts with uranium oxxde, uranium and fberylllum fiuorxde, or mlxtures of these componentsg were analyzed for the' Reactor Chemlstry Group and for the NEPA plant._ | Sodifimlwas dEtEEmined flame»photémetrlcally | This method was also uééd" -_4£or pota551um except when high accuracy was required, in which case the per- 'chlerate grav1matr1c method was used for pota531um¢-' '_Potentiometricg grav;meurxb, and colorlmetrlc methcds were used fer the_f [P furanium'determinationsa " The chclce of method was governed by the range of uranium concentratlcns and by the accuracy requ1red Berylllum was determlned as follows° The total cx1&es (U Os plus BeO)- "were determlned as well as the uranxumu The dlfference between total cx1des ' and uranium (calculated as U 0 ) was assumed to be BeO, from Whlch the berylmi' lium content was calculatedg 0bv1ously, when the U/Be ratlo is high, the: 342 ® » pesese . el » e e Y » ' e # o » ) . .o o 8 e ne PR ; : &0 L wien 8 ® e 'y X} L A . PR T ) - 2 e . e & s s PR s 46 68 . 2. 489 O e o8 se & e ede ae '-accuracy of the beryll1um determlnatlon w1ll be of a low orderu To:improve" ‘this accuracy it 1s planned to chem1ca11y separate the uranium and berylllum: by one ef the follow1ng methodse (1) berylllum carbonate pree1p1tat10nse (2) dlethyl ether extractlon of the uranlums or (3) perox1de pree1p1tat10n ofe_s N _-the uranlumfi_V' Fluorldes Were separated from 1nterfer1ng elements by d15t111at1on and-" 'determlned volumetrlcally as lead chlorofluorldeq_ ‘H;The'lithiumwcontent«xftwo fusicnswas‘determined withthe flame'phetometera' For the analyses of the remalnlng uranlum bearlng fu31ons, conventlonal | well establlshed methods were used Analysxs of Bery111um Fluoride (E C. Lynn and J H. Hackney) The- ':fusual Wlllard and Wlnters dlstlllatlon procedure for the removal of fluorldec f;was found when applled to. berylllum fluorlde, to yleld low results unless anla' .exceedlngly long distillation period was used, and even then the results were "rather erratlc° A pyrohydrolytic procedure whereby berylllum fluoride 1sa hydrolyzed at an elevated temperature by superheated steam and the resulting .Zhydrogen fluorlde is colleeted for analy51s has been substituted, and is 'vyrov1ng te yleld qulte satlsfactory results in mueh less time than the W111ard' ; 'and Wlnters method | Analys:s of fietals and Alloys (S R, Buxton) " A number of samples. of _ferroes and nonferrous alloys were analyzed by chemlcal and by spectrographlc” -methodsq' The results obtalned from. spectrographlc analy51s were sufficient 1n some cases to permzt class1f1eat10n of the samples as to alioy types;wh11e wzth other samples the information gained from a semlquantltatlve spectro- graphlc analys1s ‘served as a vaiuable guide to chemical analy31se. Chemxcal*' " methods were employed where greater accuracy in est1matlng the coneentratlen : of the ma;or comgonents of the alloy was requlred Afialysis of é%sceilaneous Sampies (L H. Jenkins) A Qide variety of finfelated'mLSCellan eous samples were analyzed durlng this quarter. The follow« ing list is: typlcal of the diversity of these samples: determination of - sodium peroxxde in sodium oxide; determlnatlon of hydrogen, nieregen, and titanium in tltanlum hydride samples: the determination of hydrogen.ifi a fused melt of lxthlum hydrlde and lithium and berylllum fluorides; the determination of water ia barium hvdrflx1de determlnatlon of free 1od1ne in 11th1um 1od1de, determination of traces of copper in magnesium perchlorate; and the determ1~ nation of carbonate and sulfate in thallium carbonate. sbnees [ soussw X ] . TTIY e L 2 “eguee® The quantlty of materxal ava11able for analy51s was llmlted in’ the case 1e*:**'” '-.o.Of many of these samples and micro or semimlcro methods were necessarv' Smallel'°' _.scale equlpment ‘was constructed and used where standard mlcroequlpment was notf‘ - readlly avallable,~:“_ | | S e 5 : The complex1ty of the analyses in thls category varled from those requlr-efi" ing only SImple tests Whlch could be completed 1n a matter of a few mlnut889 fto samples for whlch standard metbeds of analy51s were not evallabie and Whlch "j‘requlred many days to complete Summnry of Service Analyses._ A brlef summary of tnsea;fliyr1ca¢ work fon” 'the ANP program during the pasf quurterly perlod 15 glven in le»le 21. 3 TABLE 21. 3 Summary of Service Analyses - No. OF SAMPLES ~ Backlog of samples as of Dec 1; 1950 ’ .:: )_ -f' ?9 J:.—. e 24 No.' of aamples recelved u o _ o _ B 234 . _ : .: .1.85' | Total no. of samples : f'v_l - o ':_'f ei_g 313: B 'f éGS No‘“of aamples reported - o k 7_""__'E- 234 193 Balance as of Feb. 2ay 1951 B S £ IR '.thél‘fio{fof'determinatiOns made: NEPA : ':i__z: B - . ANP Spectrographlc__ Chem;cal Sfiectrographice ‘ Chemicel‘ 1250 445 213 580 ‘344j, ] e T PO L R 208 eade . 'y roze T _ sevene E aedame. S o . . . soeeaw . S sessee ‘assess ‘sedpave” ) ® » » L) _ BEPORT NO. CvRn | 3;?15-$ Y-FI5-7 Y-F28'°3. . . AP-58 CY-F10-23 - Suppl. A~ Suppl. B - Y-Fi0-24 © Y-F19-28 Y-F10-29 Y-F10-31 ARE Core Design Status . TIME DESIGN OF THE ARE . i Pressure Bro;: CompansonfCrossflow vs. Count:er- = flow ' ‘ o ' o o KOfi Clrculaung Reactors _ Temperature in a Reaccor Havmg a Prescrlbed -'_Power SRR R SEERN S PR ' ' S g Curves Relatmg ABE Confxguratwns | . The Eff'lé.cit'of Exterhai Preséure' én the Tubes o uf the Fuel Annulus Design Ccnf:.guracmn o Raactor De51gn Handbook AM) Reactor Primary Coolant Flow Changes thh o Pump Fallures o BEACTOR:PHYSICS .The Mu1t1group Method as Used by the ANP ' e Ph’yn cs Group : \Elamentary Reactor Phys:.cs ' Calculat‘.zons of Age of Neutrons in Berylllum Fluoride by ANP Phys1cs Group : Temperature Vauatlons wx.th Space and Time in en Infinitely Long Cylinder Concentric with an Interior Cylinder in Which There Is a Steady SOurce of Heat P IEV! Multlgroup Procedure.s, Bev:tsed Introducuon ' The AdJ 01nt. Equatlons and Perturbatmn Theory . for a Reflectad Beactor 345 ° senbee sesese - PYTY T e L . 8. - seve00 o ssae L eeeweal U o 22, LIST OF REPORTS ISSUED © AUTHOR - . schroeder, R. . Manson, 8. V.. Scfirocdé;_, R Daié‘ééitér', ¢ hls Est.ahrfnbk,. J. 'Manéq.n,' S V Wesson, H. Menson, S. ’V. Duffy, J. G. 'Fraras-, A. P Mann,E R. ) H.Q}.mes,. b. K. Hoimes, D. K Smith, N. M. Rubin, T. Holmes, D. K. ‘Schulze, 0. A. Nielsen, M. DATE ".34;5-51’:“ mmw ".124£§-56f 242651 1-8#51 12-29-50 1-29f51* 2-15-51 12-7-50 2-29-51 1-31-51 12-5-50 12429050 12-20-50 1-8-51" 'REPORT NO. Y-F]_O-SS S '.ggzlculatlons of 'I'hermal Paramet.ers fo: 10- Megawatt Smlth N H. = o vEos - : -_'.'Y'-.F]..Q.'.% . Y-F10-37 . NEPA-1710 : -'1-_' :_First ANP Crltlcal Assembiy Y-691 o Cnmcal Mass Calculatmns for Survey ARE ,Proposals 'Heatmg in BeO Reflectors o "Shm Cont.rol Requlred for the ANP Beactor - St.ablhty of Coiumbxum as a Reactor Core Materlal W _CR_ITiCA_L EXPEfimms : Prehminary Cr1t1ca11tnj Calculatlons for the | b Edland M c. . Webgter, J. | :__.:?gbsier, 3; E&iund, M. C. Hooneyh'an':', Al 0. _ ' ---NfiCLzAR MEASUREMENTS = Temperature Dependence of Xenon Cross -Section 3 13 : | Progress Report #3 L Goe.rtzel . G. - LIQUID METAL AND HEAT TRANSFER RESEARCH ' Fdrced Convection Heat Transfer in Thermal _ Entrance Regmns - Part I ' : 'Investlgatmn of Matermls for Use in Heat o Transfer Systems Conbe.mmg Lead Alloys - | : Lead-Blsmunh Heat Transfer R 346 eaee N sevee @ seses 20 o, 8" estoen » seses e P Poppendiek, H Stanford Uni'?ersity_ Univ. _o_f Cal\s'.fo.rnia. Oppenheim, A. 1-24-51. 1-30-51 - 2-1-51 st 12-20-50 '1—23f51 11-29-50 3-20-51 1-31-51 3-13-51 _ REPORT NO. Y-R1-227 G127 . ANP-ST _‘ Y.m;g Y-FIS-6 S YRse2 ©YeR29-l TITE i _Me_t'al}\fi.x_'gicél'__.Ifi?esti'gg;iegs-' §f 'Materials_ e . - Subjected to Liquid Lead-Bismuth Alloy Environ- ' §¢5&'.'}' S I R T B R R P . "RzaaTbR'CHEQISIBY f;' | '..Stabzhty of Dow Cornlng lelcone Fitud 550 in - - the Presence of Sodlnm : ' :"{'UG'-TOW_S?ST}"&ZM._ N Desxgn Posmhhues wfl:h nght fiezght ’mg«'l‘ow | "Sh:.e}ds ' '”ExpfiaifiENTAL ENGINEE&ING._ E:ffects of Magor Pasrame%.es-s on t.he Perfoxmance‘ of Turbojet Engines . . E NEPA Program for Large L1qu1d Metai Rz.g e .Progress Repe ts on Stamless Steel Acid Pumps Reworked to Test Speczal Features for Operatlon thh quuz.d Metals _ : Fabmcawmn Develo;ament Prcgram of Velomty of Sound in Fused Sal&s RADIATION BAMAGE Hadlaf.mn Damage and t.he ABE ”'“sUPERCHITICAL'WATERISYSTEfl ' Quarterly Report on ANP Act.nxta.es Det:. y 1950 - - Feb., 1951 e » a* T TR LY A . e N TYET Y] CAUTHOR Uni‘ira"..af California :1 Rbéan, R{: Gz-uber-,.A'. R. | Fr'aas; A. P. Haines, J. F. .' Fraas, A .P.'I | Haines, .IF _ __Savage, H. Wfi RS Prehmnary Report on Expenmental Determmamon Mann, E R Smith, L. P, Nuclear Development Associates, Inc, '.fxngrgifl b o (31351 -.1-.17;.51 S '12e20;so_€ ':7 2-28-51 CHART OF THE TECHNICAL ORGANIZATION 23 OF THE ANP PROJECT s S TY ko . [ 3 LR 2 3 L X B L3 N R 00110 ) esecoo. S e o« e LA A s LTI * 2 e e e L X X2 ® - ® .l Chart of the Technical COrganization of THE AIRGRAFT NU AT THE QAK RIDG CLEAR PROPULSION PROJECT £ NAT!ONAL LABORATORY | USAF ENGINEERING - LIAISON OFF!CER ] R Bodemuller,” _ ANP-DIVISION DIRECTOR S TR ¢ Briat {8L0e, S704. 1) o, Hicvee, See. ANP.COORDINATOR C.B.Eliis. L. Boun, Sfc. USAF : : COORDINATING STAFF il [ADMiNiSTRATION CHIEF [ WoR Eerqqren . ADMINSTRAT! VE' BDE PROECT EDITOR . Y. B. Cottreil (840G, SP041) - W, CARDWELL E. Carven L. . Gool A LIBRARY - . Trpavew | | T Higes. - Y-i2 Fiwrt are included on the chart without specific indication of divisicnat . l'hn-tm o BEP Project at the Ubom!nry.]_ STAFF ASS!STANT I : : . - : R PHYSICS ¢ B o ARE PROJECT CHIEF . R STAFF ASSISTANT (STAFF ASSISTANT. - Mo Nneisen, Lt Col; USAF .- Elis W, M. Breazegle . R. C. Briant FOR METALLURGY - FOR CHEMISTRY. ! W moma, Ses. o EC : T CiMiller Lo R’ Grlmcs . o . 5 o BULK SHIELDING . ANP PRYSICS THEORETICAL RADIATION DAMAGE DESIGN CONTRACTOR ) N e i g " ARE DESIGN ;- - RSP SR | MEASUREMENTS MM Smite ANALYSIS g s- Sillington’ EEXNPGES;EE;L@L_ _ HEg; ;g:fiwcsraa METALLURGY LIQUID FUELS - PHYSICAL CHEMISTRY © M. [ .. Billington K ) : : H L E.P Blizacd 800, 970411 T. & Welton® >, Billington. ot Bt et T R w Schroeder e AR, £ ¢ Mitier - CHEMISTRY - SEPARATIONS: %, Prceamn, Sec. R : P T - WL vage : i : u. n--m-s. A G. B, AarHENS R ot s g tBape. 92012 (80w, ngx.fl.fl {BLog. 20007 \'{;RR- G;;’;:)s : gcl Ciewef : W, K. Eaces, S L Stars ) 6. . 2 4L Meem R, R, Covevow A ufim_:':e NG’IA cR, fi, Busen W MiLLS z PN gm,_m e : e e 18006; 5723 £ ;4 <. uunq’ 5. SOR. NEPA i ALEL o W c. P:::;sk H ;. g::/:::wk K” o 2 ‘nmm" . H o o Y. kaniy . (Jl ; g:flm [ Fommrq:xm LD TN b ecmaEr. BS 8. Siuows W, ONitS Y re bomd C W B Pappertéik: T . sawse S AEY . €. i . 7. MACAMRLET & ZimwinD," RS a5 P S THOMP SN, - ESTaBROON 1.E, BEMME N . - Ay Ri £, MCORE - G, MARROH ) . 8. Mg T o AND DTHERS L, F. HENPHGLE ¥ toee U F, aaseL 03 ? M. ADRazON G, d. MEISLE B M, MclEon® €. £. Criffors iy M L Nelson® S . 8w Ko : R CovrnisH A I & g B, E, Miows SO L P TRieL (Beoo. 0%} 0. A ' B, W PARINSON 3. LITIENBLRCER. NEPA AL G BRINDELL €. P. COUGREN -] a b d L. &, DveaoLieR £, 3 KA a9 0. @, WeUTEOMSL, NEPA W fousin . £.p. MR WS Fammr | & Do Ghnsomas . B R . : . ¥. mosser 4. 7. foasa Mueloar Bevalapmest E L iR e e B Pt mengn. WAL USAE b :n :!;ufi.s. T. Catavsic,® SLc. ¥ G tAw, ¥ Sed o m’m CompsYERS 1. €. ML kssociates, inc. E. M. Lees W B ARk FLW. Swosten - A P. Bacen. WEMA J. Iukas ) RN MASGR MW, HOSFMAN . A Lews NEPA 7. camer W streant, Sec. 3. Monid NEPA il vy P 55?' ;. ;2:;4) i* Tirmiincn L 1. Gomsen. Sec. b res RITICAL EXPERIMENTS & Yourg CONTROL GROUP e PR 8w, s CONSLTANT T V. Gosvo# B §. boenTzEL o€ A o TuouAF, SE&. B . N x " A.D Callihan” ; & R. GmistR R - . Shuwon - €. DA ; ©. & HrLL) bumg eyl LX) e TECHBAT 0 dm5 M. Tgacaars, NEPA {Bups] 9218Y S . mo peisheid E 3 BefTiS' NEPA :!‘ ; Mw’:_ o Y. Surv?u. src.. e CTANS R, = ro ] s . i Fannenyra. KAPS F. Coneyozar,” KEPA 86 M. CYCLOTRIN 806, 9704031 - B R s - remmic g w #ovracanes 3 LR i Sy T. Hoeasat - Sy, NEPA RAD EAT 10N, DAMACE - . 5. A enancen, S T, FuLTon " T T * W Hunatnss 1 MACHINE HAKE, NEFA - H. K..Ferquson Company. - F b 6. Burock 4. Gomiriez NON-METALLIC CHEMICAL ANALYSIS|: . HATER, NERR 8. S. Livingston® i W, JENSER, NEPA s 5. Ciasesoawg . R, Jownson T MATERIALS. -~ kit 32 COMPUTATIONS . pRESTET. MEPA . K. Gonen ‘ A, N, K TOHEN. NEPA O 1, Lonss 3, Powe . . s SR . (Bp6. $204-3) i~ B ML G, KiTCHEN. MEPA ! ", frcuanossu. . ®e Proars RESEARCH : usano SHELD TESTING REACTIR. F. utteimia” 5. Sinors T nreps s P “3 nasis 5. vy <. Sesier [Ty 4t Hess (0. B708 A1 oA S Pap e Sl e R Hugeisfon g s Baps. 210 ¥ 5. omsssens - NUCLEAR R4 moer X s‘c"f,: 'f";‘u = T Ceaeaers - . Fiey fuoc, o7es) W ae P 1. W Fiom, 2. 5, tuce? s Cumpir. - £. Soss : o s W. 9, Sasfe, NEPA Ve MEASUREMENTS. F. . RAIN® 3. Careacarm BILE LOOP PROECT i Mere L. H. BALLWES, Mas., USAF o R Kt - R, Seine . - &, sagiw T . S Boses £o T, o0t LN Ym:“?;ic R Cotnman H. ListreTss A H. Snelt. Bl %, SToCKBRIGGE? L dmizene, 6. Siennt o 4. BuTTRES ST TE cos 2 P ramninnt 56 OESIGN CONSULTANTS ' ¥ Lonen : Sivean HE I et - ' : wl‘u. o n ProgsaeR - - . ! . o CONSULTANTS 9. Biesin’ :_ ;4 :nt'n_s@u AL CE A ENOK . HEnay | MEPA HEUTRON VELOCITY B T . B B Baueass . METAL (l}um_ - P Raypagl,” SEC. &, HOLLAND Y - . - ot i - S 3 o B 4 € smccatomo SELEETOR - CONSLE TANTS w, Deutsor. MIT R RODEMALLEN. * USAE : - . R Gray® TECHYTIAY SPECTROGRAPHIC £ 5 ) o v, % povonen, - v - SPECIAL PROECTS : %, fascatan 3 AR F. 0. M-Hsnscuzm REPR ®. CLEL . ol R. HY KW, Newhoe. . Cre. U1V . R, 5. crosser ANMLESES DoLRs 3. GRANT. i . M . B . A . sg . taam. See. & Prudicki, RIS H G mcaiasaN, MAT. CATRON o M‘;;*mm e Oy, o M ’2‘7.‘:‘;‘“" EXCHAMIER} CONSULTANTS ik iiat D, K. STEVAKS, V. B A, Seeb1R0, MIT . o 3 : TECHRICOANS N SO TenEICIANS E 6 B L. P Suind, CORNELL DV, 1. W, @0, m:u;g:pogsm«; R,oW BOARTS. Univ, TEw. . CONSIE TANT Lo B Burwsart® |—-J R'c?:;;;g ‘:; e M, 3 GARSER, LMV, TEMG. T. M. MCYAY, Uery. Ra ARD QTHERS = 57 . . . . M . Usrv A . 2;::: . £, WIZNER, PRINZITON " o Lrear CONTRACTORS G, WisLicaus) B - 5 MEV VAN [E. GRAAFF ’ ’ b OTHERS , e Waacg i k Morth Amacicas Aviation, lnc. SPECIAL ARE : ms‘k?flfl' y CONSULTARTS €, SYARE. 240 CTHERS, PROJECTS I MM WaRD, uuv. Mo . A, BETME, Comtri Lt o . : - . b Pir oA . Purdue tniversity TBLoG. 97041} LCONTRACTORS CQRSULIMTS 10, KA : :«m fi” K. LARK-HORGMITZ NG OTHERS : B s G. CRiM - . Sryger, : A .} R %"';;m niversity 5‘ gaez:z c&msmz ST, 1:04. EBDING MATERIAS ) ) W, B, ot - G SMEFPARD. by OTMERS ¥ 1. et r;uma,L e . : : %, K, Enccy. MEPA . 3 a, 8. Kitres® 5. Bain, AEPR NEW_SYST_EMS e Eean Gniv. of Cal ifornid. £l ED STAusRuRc, Uiy, TEN : {Bros, 920415 H., Bruiaeg . 5, V. MANSON, NACA R, €. GRASST . temcs,® S, '%fi!:;.i;:‘g]s 5. 8. man, sers M. A domison - - - et . - - E - = r. Caeta, mera - : CONTRACTORS. - Comg‘rl M;worqutiul ratories HOTE: This chart shows ofiy the Fines of technical coordination or the ANP LW BT D SRS projact. The various individuala and groups of pespis listed are enga . : .engaged | aither wiolly or part-time on research and design which is cooraimied for the . ’ . .“"“”' throrial "'““"“ vanafit of the ANP project in.the msnnar indicated on the chart. Eseh groug; : H. SM.&.EV- Momevez, is Also responsible 10 its Division Disector for tha detailed progress o 5 m&& S of its research and for administrative matters. Parsosnei from 13 diffarent . . Pl Blvigions of the Onk Ridge Maticeal Latorstory ent Engineariog sections of the H I Revises ~ Fab. % | : i