ML23151A550

PRM-050-044 - 51FR31341 - Committee to Bridge the Gap: Petition for Rulemaking

ML23151A550
Person / Time
Issue date:	09/03/1986
From:	Chilk S NRC/SECY
To:
References
PRM-050-044, 51FR31341
Download: ML23151A550 (1)

Text

ADAMS Template: SECY-067 DOCUMENT DATE: 09/03/1986 TITLE: PRM-050-044 - 51FR31341 - COMMITTEE TO BRIDGE THE GAP: PETITION FOR RULMAKING CASE

REFERENCE:

PRM-050-044 51FR31341 KEYWORD: RULEMAKING COMMENTS Document Sensitivity: Non-sensitive - SUNSI Review Complete

STATUS OF RULEMAKING PROPOSED RULE: PRM-050-044 RULE NAME: COMMITTEE TO BRIDGE THE GAP: PETITION FOR RULEMAKING PROPOSED RULE FED REG CITE: 51FR31341 PROPOSED RULE PUBLICATION DATE: 09/03/86 NUMBER OF COMMENTS: 28 ORIGINAL DATE FOR COMMENTS: 11/03/86 EXTENSION DATE: 02/03/87 FINAL RULE FED. REG. CITE: 52FR37321 FINAL RULE PUBLICATION DATE: 10/16/87 NOTES ON PETITION WAS CONCERNED WITH GRAPHITE REACTORS. PETITION DENIED BY ATUS EDO ON 9/23/87 RULE FILE LOCATED ON P-1.

TO FIND THE STAFF CONTACT OR VIEW THE RULEMAKING HISTORY PRESS PAGE DOWN KEY HISTORY OF THE RULE PART AFFECTED: PRM-050-044 RULE TITLE: COMMITTEE TO BRIDGE THE GAP: PETITION FOR RULEMAKING PROPOSED RULE PROPOSED RULE DATE PROPOSED RULE SECY PAPER: SRM DATE: I I SIGNED BY SECRETARY: 08/27/86 FINAL RULE FINAL RULE DATE FINAL RULE SECY PAPER: SRM DATE: I I SIGNED BY SECRETARY: 10/06/87 STAFF CONTACTS ON THE RULE CONTACT!: JOHN PHILIPS MAIL STOP: 4000MNBB PHONE: 492-7086 CONTACT2: MAIL STOP:

• DOCKET NO. PRM-050-044 (51FR31341)

In the Matter of COMMITTEE TO BRIDGE THE GAP: PETITION FOR RULMAKIN G

DATE DATE.OF TITLE OR

• DOCKETED DOCUMENT DESCRIPTION OF DOCUMENT

- ! 08/28/86 08/27/86 FEDERAL REGISTER NOTICE - RECEIPT OF PETITION FOR RULEMAKING 09/22/86 09/19/86 COMMENT OF WASHINGTON STATE UNIVERSITY (W.E. WILSON) ( 1)

10/07 /86 I

10/01/86 LTR U.S. DEPT. OF COMMERCE (RABY) REQUESTING 90-DAY EXTENSION OF COMMENT PERIOD

. 10/15/86 10/10/86 LTR OREGON STATE UNIVERSITY (JOHNSON) REQUESTING EXTENSION OF COMMENT PERIOD 10/16/86 10/09/86 LTR WORCESTER PLOYTECHNIC INSTITUTE (NEWTON)

REQUESTING EXTENSION OF COMMENT PERIOD

• 10/20/86 10/14/86 LTR NORTH CAROLINA STATE UNIVERSITY - SCHOOL OF ENGINEERING (WEHRING) REQUESTING EXTENSION OF COMMENT PERIOD

10/21/86 10/13/86 COMMENT OF UNIVERSITY OF CALIFORNIA, LOS ANGELES (WALTER F. WEGST) ( 2)

. 10/22/86 10/10/86 LTR UNIVERSITY OF VA. DEPT. OF NUCLEAR ENGINEERING AND ENGINEERING PHYSICS (MULDER) REQUESTING A LONGER EXTENSION OF COMMENT PERIOD 10/23/86 10/23/86 FRN - PETITION FOR RULEMAKING; EXTENSION OF COMMENT PERIOD I

I I 10/27 /a6 10/22/86 LTR RHODE ISLAND ATOMIC ENERGY COMMISSION

' (DIMEGLIO) REQUESTING 90-DAY EXTENSION OF COMMENT PERIOD I

10/27 /86 10/20/86 LTR GA TECHNOLOGIES INC. (HOGAN) REQUESTING AN EXTENSION OF 90-DAY FOR COMMENT PERIOD

10/28/86 10/10/86 LTR DOW CHEMICAL (KOCHER) REQUESTING 90-DAY EXTEN-SION OF COMMENT PERIOD 10/29/86 10/20/86 LTR UNIVERSITY OF FLORIDA - NUCLEAR FACILITIES DIVISION (VERNETSON) REQUESTING 90-DAY EXTENSION OF TIME

DOCKET NO. PRM-050-044 (51FR31341)

. DATE DATE OF TITLE OR

, DOCKETED DOCUMENT DESCRIPTION OF DOCUMENT

• 10/30/86 10/27/86 COMMENT OF JAMES V. SPICKARD ( 3) 10/30/86 10/27/86 COMMENT OF UNIV. OF WASHINGTON - COLLEGE OF ENGRG (MAURICE A. ROBKIN) ( 4) 10/31/86 10/27/86 LTR PENNSYLVANIA STATE UNIVERSITY - COLLEGE OF ENGINEERING (VOTH) REQUESTING AT LEAST A 60-DAY EXTENSION OF COMMENT PERIOD 11/03/86 10/31/86 LTR GAS-COOLED REACTOR ASSOCIATES (MEARS) REQUEST-ING 90-DAY EXTENSION OF COMMENT PERIOD 11/03/86 10/31/86 COMMENT OF OHIO CITIZENS FOR RESPONSIBLE ENERGY (SUSAN L. HIATT) ( 5) 11/03/86 11/01/86 COMMENT OF MARYLAND NUCLEAR SAFETY COALITION (BffiY SCHROEDER) ( 6)

11/03/86 10/31/86 COMMENT OF GE STOCKHOLDERS' ALLIANCE AGNST NUC. PWR (PATRICIA BIRNIE) ( 7)

11/03/86 10/31/86 COMMENT OF NUCLEAR-FREE BERKELEY COMMITTEE (LEN CONLEY) ( 8)

, 11/03/86 10/29/86 COMMENT OF HARRY PEARLMAN, PH.D. ( 9) 11/03/86 10/29/86 COMMENT OF UNIVERSITY OF MISSOURI-ROLLA (ALBERT E. BOLON) ( 10) 11/03/86 10/29/86 COMMENT OF IOWA STATE UNIV./DEPT. OF NUCLEAR ENGRG (BERNARD I. SPINRAD) ( 11) 11/05/86 11/02/86 COMMENT OF NATURAL RESOURCES CMTE - LWV BALTIMORE (PAT LANE) ( 12) 11/07 /86 10/18/86 COr-t1ENT OF ECOLOGY/ALERT (EMERY NEMETHY) ( 13)

• 11/18/86 11/06/86 COMMENT OF WORCESTER POLYTECHNIC INSTITUTE (THOMAS H. NEWTON, JR.) ( 14)

. 01/28/87 01/23/87 COMMENT OF RHODE ISLAND ATOMIC ENERGY COMMISSION (A. FRANCIS DIMEGLIO) ( 15)

• 01/30/87 01/28/87 COMMENT OF OREGON STATE UNIV. - RADIATION CENTER (A.G. JOHNSON) ( 16)

• 02/02/87 01/28/87 COMMENT OF GA TECHNOLOGIES, INC. (KEITH E. ASMUSSEN) ( 17) 02/02/87 01/28/87 COMMENT OF UNIV. OF MO. - RESEARCH REACTOR FACILITY (WALT A. MEYER, JR.) ( 18)

DOCKET NO. PRM-O5O-O44 (51FR31341)

DATE DATE OF TITLE OR

• DOCKETED DOCUMENT DESCRIPTION OF DOCUMENT

• 02/02/87 01/30/87 COMMENT OF UNIVERSITY OF MISSOURI-- TRTR EXE. CMTE.

(DON M. ALGER) ( 19) 02/03/87 01/28/87 COMMENT OF GA TECHNOLOGIES (R.A. DEAN) ( 20) 02/04/87 01/29/87 COMMENT OF UNIVERSITY OF WISCONSIN (R.J. CASHWELL) ( 21)

! 02/04/87 01/30/87 COMMENT OF GAS-COOLED REACTOR ASSOCIATES (L.D. MEARS) ( 22)

- 02/05/87 02/06/87 02/02/87 02/02/87 LTR UNIVERSITY OF MO - RESEARCH REACTOR FACILITY (MEYER) RE CORRECTIONS TO COMMENT LETTER OF 1/28/87 (SEE COMMENT NO. 18)

COMMENT OF NUCLEAR REACTOR LABORATORY OF MIT (LINCOLN CLARK, JR.) ( 23) 02/06/87 02/02/87 COMMENT OF UNIV. OF FL. - NUCLEAR FACILITIES DIV.

(P.H. WHALEY) ( 24)

I I

. 02/06/87 01/23/87 COMMENT OF PUBLIC SERVICE COMPANY OF COLORADO (R.O. WILLIAMS, JR.) ( 25)

02/09/87 01/30/87 COMMENT OF GEORGE W. NELSON ( 26)

I

02/10/87 02/03/87 COMMENT OF UNIVERSITY OF TEXAS AT AUSTIN

- i 02/10/87 i 09/16/87 01/23/87 09/16/87 (THOMAS L. BAUER) ( 27)

COMMENT OF PENNSYLVANIA STATE UNIVERSITY (CHARLES L. HOSLER) ( 28)

ABSTRACTS OF COftt1ENTS SUBMITTED IN RESPONSE TO PETITION PRM 50-44

DOCf<ET NUMBER 5J PETITlON RULE PRM - 44 c j-/ rt_ .J 13 -f (}

• a7 SEP 16 p 3 :48 I.,.

I I ABSTRACTS OF COMMENTS SUBMITTED IN RESPONSE TO PETITION PRM 50-44

,,. NUCLEAr< ~£GU!.A10RY COMMISSlv.

6o(t<EflNG & Sf~VICE SECTION OFFICE '"';: :~!~ S~'"~ETA~Y 0:: *r* 1

:::* ._:u\i3Sl01'~

lt . . * **,.~

PRM II !I ii 50-44 UNIVERSmEs ii ABSTRACT COMMENTS

~or 11/3J88 Yea X Request mension b ltt>mlUal of comments of at INat 60 days. Dr. Solon's oornnenll are Yfll'/ c,ypelo, howwer, he makes aeveral n o ~

Missouri. 60d Also Indicates that 1he CBG propoted dala of Jaiuuy 1, 11l87 tor argtKn8nta.

Aolla,MO aubmltllll of fire respome plan9, evacuation plant. meuurement 1, Geometio and atrucl1nl conaiderationl In cerlaln research reactor& may of Wigner energy la fat IOo eady. SUggesta at least a yea,' be rule out IJal graphle lnteracllona. Thia lhoud be examined. 11 may be allowed lor rellpOl'U If NRC deckles lo accep( petltloneta .-queet poulble IO group !hue reaotJnl with those having power levels e<pll 10 or Clles design features of lhelr research roactot 0-*** graphite leu f1lln 100 Wu notreqtiring bther cotwlderation.

refleclDr Is aeparatsd tom core bJ 1fl. Inch of 11&81; 4 lndwa of 2. If It l a ~ !hat reeponses trom lndlvldual reactota must be ~

lead and 8 lno.hea of walet) makes the need b tire responee

• he commentot'a recommendallon on responM lim& ll'lowd be plans, ervacualion plans and me&llfflTl8llt of Wigner enetgy c:oneklenld.

unnec:eaary lilC8 a s,aphlte fire coud not mull In the release of 3. His uaeument of lhe aca.ncy poealble In delOOnlnlng &10red energy In fiulon products from the oore. graphite b/ eilhef analytical or experinenlal meat'lil u t.5()% must be Also 1U1J9H11 that orq a Jew reactor taci!Nel should be used IO addreaaed In !he NRC staff position on experimental measurement of

.maka *experimemaJ mellSUl'efflllfll of stored energy In graphlle

• nxod energy and on !he leaslbffity or calculallng alDAld energy using a if NRC concludea that c:unent experimental data are not aufflclent. aandard methodology.

Suggests lhat these experimental dall. should permit calct di.lion o the locallons of maxlroom slOred energy and 111 quantitative value lor otler reactors.

Wrilerdoes not beleve the slDfed energy ca\ be detem1lned wltNn :10%. A more reallllllc l"IUTiber would be :t.50%.

Nor1tl carollna 10J20/86 v.. Requosta extension or commert period.

State Univer9lty 90d

• . Unlvenity of 10/8/86 Yea Requests extension of comment period.

Miuouri, 90d Columbia, MO .

Page 1

PAM -

50-44 ml UNIV~mES UnlYeflltyof 11 II 10/30/86 No 1! ~8 ABSTRACT X ~ProfllAOI' Maurice A. Robkin, Dept. of Nuclear Energy)

COMMENTS lnfotmdon IUbmined Is largely a reb.ltlll of each point of lhe pethion. SIOf6d Washington "'* ,:_ ** allegallons we grgc.indleN and can be cbmNed bv con- energv'ln grapnll8 convnentl largely auack CBG'a 8ll8ftion of corwmallon of fieli of some almple physlca and knowledge of IOITl9 exper- calculated 39 caL\,n b UCLA reactor~ (1;63) l:,v lhe measured value of lments. Presents an mgunent that although C8G dlN the mea- 33.2 cal/gm b fle l¥gh fut tux region of h t ~ center Island. ArQll"ll8l'lla IIJ18fl18nt of( slored energy In the graphite of lhelr Argonaut reactot are qualtatlve, but merit conaldaralion In* relation to development of the BNl raa a muiTun of 33.2/caL'gm when restriclad ti a email volLme ~ oalcwllional melhod. Also provides argunenta on lempet'lllln

!does not.reprnent a hazard nor does It represent 1he average rise il lhe UCLA graphillt center leland of 110-C lor adiabe.tio heat u p ~

level of lt0fed energy In tie bulk of fle s,aphile. Allo rem.IN lltontd energy releale. The equation extraclad from Nightengale'$ book,~

!the resuta ot lhe CBG cab*tlona of IIDred energy wfth a claimed Graptjte. may be useful lo the deYelopment of a calcwalional method.

~.-.:.- *-' *- of 113 callgm corrupondlng tD 38 cai11J11 ii IQ.83, Then Statementa on graphite fires** a bit OYet'Saled, but aoudl on l8YMIII key points:

.,-.v_,.i1a:

• Hanford exptrimenta (oxyacetylene l0rch) lblraln the difflculty In ~sh-

• If the meaured maxlnun st0c'ed energy la 33 callgm the Ing cmdltione b Nlf-austalned graphi&e tuning of large bloeb oc logs of CBG admaled mk1inllm value of 3'I oa1.vn II a fJ'ON graphite 0Ye1Nlima11t * "There la, ii fact no eYldence !hat hire was a graphite atacf< -&-e,**!hat II l&lf.

• If lhe C8G esllmale representa hi mlnmm ~ Jot ht aumJned combl.don of maulw s,aphilt, either In the Chernobyl reactor maxkrun It la llkewlu lawed or ii the Wllldscale reaclOr."

• Also ~ ai extrapoldon of UCLA operallona back lo

• Refers ID tests ~ TSX graphite blocb containing a channel hough, the 1983 bv ratio~ fin a value of 64 caJ.lvrn- much larger cenlllr (8" X9" c:.I) prehealad ID 600"C

• showed . . lie chemical reaction Nn the measured value with air was ilconaequentlal compan,d to decay heal from reacm.

• UCLA me8lJIXelllGllll of center laland graphite are subetan-

• Graphlle mutt be finely c:lvided n:I expoNd lo a saeady suppl-/ of oxygen tor ially bek>w the maximum selr-.u.tainlng c:ombuslon

• Making certain COflNMllille auunptlons (18.5 cal/gm)

• No credible operational event In Argona.it reac'>l'a which can disrupt the

(.tlabatlo heating) calculalea a ~ rise of 110"C g,aphile auffldenlly to either p,ociJC:e large quanblies of hot fine partlclea or that II .way below fl.lei claddi-v meUlng poi,t or fuel 1111,oy IIJge ai' Iowa or both suflclentlr large to cai.u slgnMicm s,aphite oxklatlon.

, melting point The argunents and base& l11JSt be considered In developing an HRC stat position.

• Refe.- lo experlmenta al Hanford rela!lng lo graptila oxlda-don and comments on graphillt lint at awtrnobyl and

\Yondscale and concludN graphite COll'll:qllon wa de'8t-ri1ed lo be baslcaly not poaible with surface to vobne ratio ln*large blocks plua oxygen transport lo the surface-conlrola oxidation

• Sunvnarizet that ltored energy ll"ICler the most pesalnistlc

~dons la lnsuflident 10 heat the graphite to a tempera*

IUre !hat can damage the fuel, and even If healed, tie graphite wil not bum. Wigner energy and graphhe Ires lnAtgonaut reactors are non-lnues.

Page2

PRM d

a5 wl 50-44 Cl) 0..

UNIVERSITIES Wuhlng10n Stato ii ~; §§ 9122186 No X ABSTRACT COMMENTS Strongly opposes petition. CHu cause or Chernobyl accident to Pmvldes COPV of Nuclear New apeclal repol't on Chemobyl--4 and the aueaamena Unlve,dy be ** prompt critical reaclivlty exa.nlon and* 1'8am explollon" that lenonl 10 be learned from Chernobyl acddent may require changee In certain not by graphite fire (attaches article on Chernobyl from 9111/86 ia,eas regulated by NRC. Recommenda a review of al factl betote ldentllyi:tg "Nudear News;. Main comibudng factor& went human error and problems and appropriate remedy. Delcrlption or CBG pet111on as being premalln falure to folow procedu'9S and prud&nt safety preoarllona not ahould be considered In developing NRC staff position.

graphite In the reactor.

Chernobyl subetantialel NRC position 1hat

• graphlle lire caused by Wignef effect In a small research reaclo( Is a ~

event.

States lhat the lmpoailion of 1.11neceaa,y regulalionl on ~

search rectors will decrease safety by dlvetsion of staff efbt.

l.eaonl lo be learned from Chernobyl accldenl may require changes In NRC ~tionl in eettaln areu, however CBG politlon premalllre before all facts revealed, consequendy does not address real p,oblem.

Iowa Slate ~0/29/88 No X Or. Splnrad prOYidet, a detailed and systemallc argtment In lha ISU position can be used as a reference polnl br NRC staff conslderuon of University addteulng the aLJggGded requlrementa of the C8G petition. !tis power level po kW maximum allowable Md 6700 kW twa of kllltgraled powef)

The Departm~mt of Nuclear Engineering 181.J takes Ille lolowlng In olher n,actora. The 7 alep prooeu outlined by Dr. Splrvad ahould be considered poaitlonl: by BNL In lhelr developmeot of a ltardard calcuJallonal methodology. Dr.

• The posalbllity of a graphlle lire In ht UTR-10 reacto, la Splnrad'a atatament. in any case, bounding calculatlona, In Ole apirit ol those

' extremely remote, reported here, should deterrrine whether luch refleok>ra po!IG any putaliYe pro-

• S10red energy In tie graphite of Ille UTA-10 la too 1111181 lo blem requlrlng elaborala meaalrea lo resolve," la worthy of consideration as a way Initiate ot contribu1e 10 a serious accident, and 10 acteen the non-power reactora to delermlne those lhat roost provide more

• Existing emergency response procedurN are adequa&e lor delalledanalysia.

flefacity. -

The argumems supporting their politlon are aummarlzed as folowa:

1. Because the graphite In the UTA-10 la enc:losed In concrete, the poaaibihty of a large graphite int la very remolB.

2. There Is amply no credible reackM' accident lhat coud Ignite the graphite.

3. A sudden release of all the slored energy In the graphite adjacent lo the core would result In at mosi a 62° C temperature Increase.

Page3

~I PRM 50-44 iS UNIVERSmES ii ~i IB 11 ABSTRACT COMMENTS Jowa&tala 4. Only an act of sabotage, reautklg In ht btal demJcion of I Unlvenllty 1he faolltv, coud cause a graphilt h.

(Conlinutd) 6. Exlating emergency rnponae p,ocedurea ue adequate lor any crtd:>le reac10r acddent becaule lhe A>NJd energy Is lrwufflclent to cauae or conii>ule to an Ll'lll'11lclpaled accident scenario.

[The c:ommenlor concludea full:

• No crecible accident at the IJTR-10 oan reet1lt In a graphite Int.

• The NRC should deny the C8G pellllon and all or Its req.n-

• No RMll8d SAR la neceuary slra IU'ICI ht NRC SER ccn-lideres acddenta more extreme lhan the releue of s,aphite alDredene,gy.

I

• A ffl8f1'IOf8J'ld "8lofed Energy In ReleotDc' Graphltll of Research Reaclof'a,

• is pnMded showing a de1alled cab dallon of ht ll0red energy In 118 ISU reactor (le., 10.6 CaVgm In s,aptite adjacent to cote) and conctudlt there II no hazard uaoclated with the audden release or Na energy.

• ~ and measurement of alot8d energy ia not nec:easary.

It WOtJld expend umeceasarly scarce fuida, hundreds of per10n-ho1n and reeult In a few penson-rems addtlonal doses.

Univerllty of 10J2118f No X Proleaor Wegat comments as lollow-a: Proreuol' Wegst's rebuttal of the posiliona alaled In lh& CBG petition on graphite Caflo,nia.Lo. 1. lha Chernobyl graphite fire hu no relevance ti use of Ires and alOr8d energy are relavant 10 the development or lh& NRC atafl'a Angele& graphile In non-power resewch reaclOl'a. Since 118 high recommendatlona References 2 aid -4 have not previously been ldenllfied and graphite l&mperature (100"' C), lolls or Inert blank.et gas, should be reviewed. The ~ Impact conc:em can be deferred un11 the NRC mal<es numerous verticle channels for air flow and the lal'ge mass ltll decision on flt C8G petition for nN making.

(1700 tons) or V9fY hot waptite characteristics or Chernobyl

. . not Pf&sent In U.S. non-power reacm.

Page4

PRM 50-44 ~t;

~1 ~I sl Ii iS g~ ABSTRACT UNIVERSmes COMMENTS IJnMH8ity of 12, The ltl8ua of alofed Wigner enellJt II also of oo ~

California. Loa because of the geome1ry of research reactors. Very 8l'l'alJ Angelu voune of lJ8Phlte II lrradaled by last neutron tux (e.g., UCLA (Condnued) Of'q 6% of the graphlle was lmdaled by fut neulrons) leaving a trernendoia YOk.me of unaffeoled graphite ID abeorb II\)' al0fed energy reJeue. For research rHOfDIII M'IOU graphite le lmmffled In pools (lhoulandJ of gallora or waler) the WallK heat elrj( will preclude heating the graphile to Iha tempera.1urea (several IUldfed degrees Centigrade) required for lloted energy release,

3. Long cl1c11nkm of C8G'a mlsule of ltal8monts llken from lia 1es*monv cluing the UC1.A hearings. 1983, and provides bases for hll tesdmony. Also dies the lnacctney of lhe CBG ltalement, *p.it anoller way, auflk:ient Wigner energy could be 111:nd 10 In effect lower the 660" C lgnllon lempendl.n by l8V8f&I hundred degree$.* The atotage of Wigner energy, per ae, has no effect on It-. Ignition~-

4. CitN cost Impact of propoaed requirement In peMtion. Also dfflic:ulty of taking l8J11)lea from alurnbm clad graphlla *ling-era. Ono rec:owse of Uriverllty reseatch reactora wootd be lo replace old graphltit wffl Lrirradlaled c,aptillt. Cl1H auapi-

! don that lhe "hidden agenda* behind peltlon la to llllA down research reacllors. No lncreale In publo health and safety would be gailed by meut.rlng IIDred Wigner 11"11N\11, Pages

PRM

~ ~

~~

50-44

~1 ii i5 i§ w~ I UNIVERSITIES ABSTRACT COMMENTS Wo<<:eater 11/18/118 No X Thoma& H. Newt>n, .k., Di'ector Nuclear Reactor Facllty, llddres&ea Unfortunately, commentor does not present detail of hla calcuJalion of slDred Polytecmlo sewral Issues related to the CBG petllon: ef'lllf'VY k1 i1e WPI reactor graphite. Questions about location of stored energy, Institute (WPQ

• Calculatlons made tor the WPl reac10r (JBphlte, based on !hose per- average or maximum luence, and dlstrubllon of relea.aed al0f&d energy rel-formed lor fie UCLA reaclor, rjNe current stored energy values of adve IO lemperature rise remain. His suggestion of using lluence ralhet lhan 0.13 callg and an estimated value of 0.22 callg 8':lted energy by p0¥l8l level has merit It may be pouible lo establish a level of lntegra.Ed hi year 2000. ftuence below whJoh *wed energy Is alnw of no concem. His suggestion lhat a coat I benellt anatysls lhould be pertonned prior to Imposing the re-

• For lhe WPl graphlre, an lnstantaneout release of f1e calculated qulrementa proposed In PMR-50-44 are appropriate If a d&clsion ii made 10 stored energy would result In a lemperaklro rile of leu '1an 1"C. pursue any of the PMR-60-44 proposed requirements.

• Rather lhan power level (e.g.100W) he auggesta thatneub'On The commentota position on graphite ,:ii.. (lrnpoaslble) Is one of the Issues lluence be used 10 determine a re<pJ!rement llmil for WigrMw being conlested by CBG.

111:>led energy (e.g. 10 ~ ').

• Relatively low operallng temperatutea In mearch reactor com-blned with void and temperature coefllclenta mltigale agalnst a power excursion resuting In tempera!Llret neeeasary tor graphite Ignition (e.g. lhe WPI reactor la l.l'lable lo even cause lhe coolant water to boll).

• &lgge&ls many research reactors rather lhan U11)1ementing I

response plans and equipment and Wigner energy measure-mentl would opt to remove graphite where feasible and change reacto( core conllgi.ntlona. Cites polenlial for fuel loading errors and pensonnel exposure that la small, but mote lkely than a

£Jlll)hhe fire.

• If NRC considers approvilg PRM-50-<< It lho4..dd fund a program to slandardlze Wigner energy meaautements In research reactor pphlte. PAA shouJd be ir.tlgal8d In o,der lo verily the necessity of ltis ru6e making.

• Opposes petition - atal8S that from In pok1t of view, graphite fires are non--<:redlble (lmposslble). Meastnmenta of Wigner energy to verlt;, calculaliona Ml show stored energy to have a

. - contribution In an accfdent scenario.  ! ,

Page6

PAM

~

50-44 UNIVERSmES IJnlverllty ol sl II 11 2/8187 No ii X

ABSTRACT U of F proposes 10 show flit petidoner's 81118ttion that Wl$Jler energy In graphite moderated Argonaut l8aOlota repwenta a int COMMENTS U of F convnenaa are well staled and provide additional ~ to consider in evakJa-.

ing Ile Argonaut reactors. NI explarallon of the varlaliora In ht calculations of (Uof F) hazard Is groundleM n should be dlsndiaed. U of F addreues A>red energy In the UCLA Atgooaut reac1or la provided. Thia and U of Pa point Iha.I the BEPO Equivalent Flux and the appcoprlut factor of U 1 to the petl1ionen calculaled value of al0red energy (39 ca1lg) If totaly releasable lnclcate the proper faat neutron exposure* a foocllon of lhemwl adlabebly would not be a elgnllcant lempera.lU"e rise In the i:,aphile (-150" C) neulron luenc:o. U of Falto W'llllyzea fie NUREGJCR-2079 lhould be contldered In the Staff's evaluation.

calculadon of stored enerw for an ArljlOl'18IANNICbr. ~

appropiate correcllon faclora ID the 5 caUg ltored 8l'lefVI value In NUREG'CR 20Til(corrects for operalqj hlslory and 10 BEPO equlvalent lux) ta obtain a oorrected value ol 20.9 oaUg In good agrMment with UCLA data at 33.5 oa1.1q also e,q:ilalna rruinun, minimum and voll.me averaged stored energy valueL Cbea the desontpancy between pelilloniera minlrrun value b UCLA Argonaut of 38 oaLlgm wlil the meaued VIIII.- of 19.2 caLva

. U of F anaJpls of 11-. clltrlbtem of R>nNi erie,gy ii Atgonaut moderator graphhe le~ them lo c:oncludt that ,wn hx.91 11-.

original analysis was belled on auunpdont tat only allow orders of magnitude IIClCI.UCY, ht va.luea are ~ b llh.aatrallng that lhe Wigner Eftect la not a relevant llaue. 11w conected cal-culationll Im meaurod values show ht peak and.,,.,.

values of Wigner eneqw a1orage are mremet, low and u. In-IUl'llclent If '"'8Sed ID heal the grapljlt bf mon1 "8n an inlig-nlllc:anl amount.

U of F de1Jcribn lhe max1nun tempetat,.na achievable bt/

releae of alored enersw aid tlnda that a atr:wed-enetgy re1e...

cur.,,e la uaelul I n ~ an upper bomdaty of the hi! value for graphila IBmpemJrn eublect to amedng and report the lol-loYting:

i 150 callg .,400 MWDIAT AT30"C

• i;.. OF36CfgJ AMMIT-.

<1000 MWO/AT AT30"C. i;.. OF,w/J' 1111>-101r c Also report lhat continued heating don not reacit In apontaneooa teq,erature Increases..

U of F npecllon of cuve b 100 callg suggeeta a lemperab.n rise of km lhen 150

• C br the petitioner's catcula1ed 39 caVg for UCLA graphilB: commenls that allhou,Jtl CBG grosely Page 7

PAM 50-44 UNIVERSITIES ABSTRACT COMMENTS University ol

• oventttinalN WilJler GneflW atorage valuea, 1he values do not Florida present. hazard since the adlabaMc reli!tae of tt. pedtloner's UofF value wll not red In exceNlYe graphile ~ . Sid I (Continued) viewed

• an energy dla1rlbudon, the magnitude of 1IOnld 9l"illfVV hM no _.ty l9liflc:ance. Ches i,aphlta bl.rnk1g experlmenla In krnaoe aid fabe to bun \11111"110 CFM of air flow past ..nple.

Also ohN Thermlle In g,aphlla ~ wllh peak graphite lilff1>er*

aan ol 1426- C and no continued bl.ming toUovi,ing COffl)lelion of Thermbe burn.

States Wildlcale and Chemot,,,,I were not princfpdy graphlle Graphlle IIOA)ld buning when fuel oxidation c:eued.

U of F expWna ht following conctionaat UFTR rule ow a graptite In:

0 lack of 9lilion IOOl'Ce 0 large volume to IUl'face ratio for the graphite o low exposure area of graphile ~ to free alr apacea 0 low air low through reaolor void spaces 0 low operdng lemperallns.

Masaad'luletll lrlal;1ine of No X IT Reaotot (MITA-II) ha a waphifa reflector, however, lhe urr. comments on geometry (Le., barriers), temperaues In 1he grapl'ite, me*

T~*MIT pnM)fudea lhe possl,iity that ill gnipMe colJd be a hazard. auremerda made on releclor gtaphite, g,aphlll bl.mklg experimentl and observ*

IT also.concluded that Ile chemlcaJ llabllty of graphl1a II such tiona and relevance of exlsdng approY6d Emergency Response P1ml 10 CBG'a It cannot be

• hazard In non-powered rNOl0nl In general and request for new rn rnponM pia,. lhould be considered In h Staff'a evaJuation ht pelllon at>OUd .be denied. of PRM-60-44.

J)etlCt'll>H Iha ITl4l1Y baniera between lief and s,aphile:

0 lbnn.m fuel ciaddqJ 0 ah.l'rwun C0fe houllng o lght wa1a1 moderak>r o aluminl.m core tank o heavy waler reflector o akr'Mn.m refl8ctor tank.

Barriefs sepa.nm fuel and graphlae b)' a ntinun ol 17 lndlea. In

........,....., ., graphilD la blanke1ed b)' an Inert gas.

In MITR-11, maxlnun fluxn occur In a waphile region llat opet*

aees at 150" C ~ range lor a reactor power ol .U MW. MIT concludea lhat stbred energy released on healing 10 C 0-16 Page8

I PRM

• r 50-44

m 11 UNIVERSffiES Mauachuseu.

~1 1!

~i i5 ABSTRACT never be aua.iled). Also concludea flat even If a saluratlon value COMMENTS Institute of of 47 cal/gm was reached It woud not lead lo exceaslve lef11>et'-

Technology

• urr atuea. For MfTR-1 operaied al 300* C graphite temperalJr8S so (ContinJed) alored energy values were negligible.

MfT "°k a sample from a flennal counn slringe( at position of lower 1ux and lower lmdalfon 1empenwre and compared 1ta

'8rnperatUre cblng annealing with an unlrradlated sample. No difference noled ~ to 500" C (I.e., no ln88SIX8ble stored energy release).

MfT cltaa l'Nt.llta from aeveral IIIHlta n.ri m evaluala cornbwtibllity

[Of reactor grade graphite. Up lo a maximl.ln lemptrature of 1<400-1600" C there was no evidenc:e of burning.

MfT dH operadon of the graphite electrode In a carbon arc at ip temperakJres of 4000'" K. Volalizallon and oxidation occur but no b..lrrq. Also cites 1hermlta In carbon crucib&e that demonst.rat&

lx.mlng orly occurs under lf)8()iaJ comblnadon of llow channef geo metry, forced air flow, graphlle temperatures and concludes flat lhese conditions are not applloable to non-power reactora.

MIT ataJes that it Is not dear that a s:,aphlte tire occt.lmld In 1he Chernobyl plant Fuel tuning may have been mistakenly ldentl-lied as graphlle burning. If graphba bomed at al, It was probably hours aftsr 1he accident when decay heat caused lhe ~ I I J l e b rise. MfT also states that CBG's assertion lotes credblllty where It anempta to compare the Chernobyl RBMK plants 'Mth U.S. non-power reactors licensed l:7f NRC.

MIT concludes that au of 1he Information known about the llt.9Cept*

iblllty of graphite to Mlf.auslai'led cornbudon and all of 1he experl-ments that have been conducted Indicate that graptite la not a lire hazard In non-power reacton. Thet'elo<<I, there Is no need lor a

- I new regWlllon requiring fire response plan&. Evacuation plans already submitted and approved are adequate. Toe presence of graphite doea not oonslitu1a an urvevlewed hazard.

Page 9

PAM 50-44

- ~I UNIVERSmES Oregon State ii II is 2.4.78 No 1!

X ABSTRACT o.-----,.. lhe NRC I> reject h pelldon b h folowlng nmona:

COMMENTS 08U hM pANl&nted a knowledgeable atalemenl regarding lhe petlaionen 81M1-Uriversity

• lnadeqlallt leC:hnlcal juallflcation of pellloner'a conl9ntlon Ilona and r8qUl8l9d requnment9. Aesulls . . given In qualllallve 11kma lor OSTR

• CBG propoula are not reqt.ired 1"101' appropri.111, lo ensure lafe and mended lo llmilarly detigned l'H0101'S.

operation of lhe Oregon saa1e ~ TRIGA Reactor The cornmenta lhould be COl'llldered In the dewlopme,t of the NRC atalt lll"llllysil (OSTR) and oiler non-power FNCtors L111U1g graphlllt. and rNpOl'1N to 1ht petlllon.

• For TRIG.A reaclOts In plricuar a gniphlte ftre Is not a credible event IO the publlo la not at risk.

~ OSTR graphlla Is fou'ld In Ive locatlona:

• M part of the fuel elemert (top and botbm releotora lnllde the atalnleu""' fuel daddjng)

• As stalnleea alael dad graphite relleolot elements In IOll1!I of fle leaclDr s,ld positions

• Man am.dar rvllector sealed In an alunlnum hoUling

• As pert of a aealed, alwnirun-cowred thennalzl111 column

• Man abnll'llJfTl-COYeAt flennal colurM.

ITiwt l'nt lhrM locallona are lotally wi1Nn hi reac1or tank coventd 1bV 4,500 gals. otwa1er. G,aptite n 111t lot.di 1oca11on 1s boooded

~ botl tn:11 with 4,500 gals. of waler. The graphlle In lhe lfth lo--

~ protudes lnlo the reactor tank nkSe a apeclal auriotm hoLWlng and ii bounded on ht outer end by a 19 ton c:oncreta doot'.

• - .-* ack:lruaes each of Iha peGllonan aHrllonl or

~ts.

• Dillriues the argument flat ChemoboJI c:lemonsntea waphill

&res .,. creclble evenll. C1t1111 that ChernobtJI Is not relllvanl lo non-power reactofa, muted from pi'Oqlt critioal pov.,er ncu-sion caUli,g rapid lemperatur9 rile and complete cent dlarup-lion. Can not occur In OSTR becauee of large negallve temper-aw coeflclent of reaotMtv. Chernobyl was not due to Slorad energy. OSTR SAR ahowa tlere are no creclble acenarloa or -

~ tl8t red In fuel and graphite ~ high enough ID melt fle fuel or ID restAl In g,aphlle bl.mng.

• • Responds lo pe'1kmet'a statement that NRC hu failed lo require bulc safety l1l8UU'e& lo reduce the threat of a gra,:tille fire. Auerta flat bulo safety measLl'U .... lncoq>orated In OSTR lo reduce the llveat of 8JVJ fire iduding qrapnle ha.

Page 10

PRM 1! ii d1!

50-44 UNIVERSITIES ABSTRACT COMMENTS Non-ftlmmable conalrUotion (metal, c:oncrele building)

• graphlt8

-Oregon State UnMtrllty ii dad and lmmefSed In a large volume of water. Isolated from (Continued) lgnlllon 10WCM and atmolpheric oxygen. Cil8S lire fiw1ling

~ from tralnt<<t COf'l'Vl"Ullty Fie Dept.

Dilrriuu pelilloners clam !hat lcenleea have~ tlr6 mporll4t plan conlllined In tie OSTR NAO approYed emwgency rnpanae plan. CoMdls Ire lghlera and aupervisora niied and prepared 10 deal with lhe types of ires hy could el'lCOUllered lndudlng "8J)hl1a.

Reeporld& to pelillc:>nef'a aaaer1ion flat reeean:h reactors do not have adeqliate emergency pans to evacuata members of Ile pul:illc In lhe evenl of a grapfvle h. SlldN flat flffl'/ analyals of polanllal emergenciea.tot research ~ of leaa lhan 2 MW hu sl'IO'M'I f'latevacuabl of ht pubic ii not neceuary. Allo oilu N faHute ol lhe petition<< 10 ealablllh a lechrical buia which demonstrales why a graphb h should reqlh 1111 evacu-allronplan.

Addreaes lhe pellondonar's llm&ment that NRC Uldered-malBs Iha actual amount of A>red energy and hJa, ataM lhat 1he buildup and release of ltOred energy In graphite doea not translate lrlao an i1cf8ued Ire rillk, pa.rtieularly when graptiu la clad to exclude atrnospleric oxygen and immerled In aignlllcant amount ot waler. klenlitles most lmpoftant graphite components In relation lo al0red ene,gy. osu analyall conllrml lhat tempera-tum will remain wlthil fuel lafety lmlta. They also show In ilei' anaJyal& of h grapt-i!e reflector ring lhat 11:lmperature lnc(eaau resulli'1g from 11ored energy teleasea remain below Ole point where *any damage lo lhe graphite or al1.1t111Um clad "11luld oc:our. Cites ~ In a10red energy calculallo,-.

Including:

• Factor i.ed In converting from neutron fllJ)C to 1101od energy

• Calcualed maximun slored energy (i~. occurs only at lhe point of maximum neutron b and lowest graphite lempef'ature)

• Only a fraction of lhe stored energy la avaiable lot release, but OSU aaaumed 100% released Page 11

PAM

~

50-44 el ~I UNIVERSITIES ii II Is ~§ ABSTRACT

• cab dallon or maximum temperaue of graptillt aft8r atored COMMENTS

()regon Stall UnlV9fllly energy release OSU uaumed no heat ltanafer (adlaba11o); Con-(Continued) dudes lhat conelderallon of al ht abcM1 lilOWl lhat lhe OSTR and other almilat non-power reaclots' design la auc:h that a gra-phl1e ** hm release of stored energy .. not ctedlbkt

• SalH lat the petibler'a request lhlll actual emperica,I mea-aurementa of Wigner stored energy be reql,hd la lmpracllcal and agaiisl Al.ARA phlosophy at OSTR Mor9 lmpoi1anfy, how-ever, la the lack of technical basee ID establsh end enlltlW as a polBntial problem so itll meuurement II Ul"lllSOHUI')' aki Ul'M'ise.

Pennsylvania Stats 2/10187 No X PSU summarlzea the pelitiona contendona: PSlh C01M181 IIII aAt mostly qualitative. The 1n1orma11on onthe deaii;p1 features of

~

• 08G condendll that lhe aloled eno,w and the potendal coose- the PSSR aho!Jc:fbtt conaJde,ed In tie Slaff waluallore of the TRIGA type reacior.

quences of*a fire Involving graphite In INClor cores and relleclDr Refentnc:e ID F. C. Foulhee'a 1004 report lit cil8d u the buls llr fie alaled conckJ-have been mdereatlmated; and &Iona. Staff and connctor review of lhil report la nte0mmended.

• that specilo addll6onal regulallons should be promjga1ed.

PSU. concludes lhat lheae contenlionl have an lnlldequma tech-

' nk:al buis and hit the ntqUeat8d changes 1o regulallona are In-app,opriaie mlhe PeM Stale Bteazeale Reac10r (PSBR) and other slmBa- reactora.

Clea the lack of techrical buls for CBG cordullona based on Chernobyl event. States that no lachnlcal iteraue 11 ~ lo aug-geet lhal lfDred energy In graphite WU a contrb.lllng faclol' to ht aeverlty of the event Exmooes the pelillons concem wbh the dfference between mea-swed and calculated alored energy In the UCLA. A,gonauHype rnearch reac10r. PSU explains ht \11111hout Lnbmdon on fflh-awement procedurea o, on 118 calculation 1hey oamot c:omment.

They nollt, however, that n>red energy In and or belf does not condue an Impending problem. PSU explalnl 1hat equally Im-ponant concitions are:

• lhe tempemlln attained upon release of slored energy '

• fie presence of oxygen t o ~ a flame, and

• Iha exposed aurlace area of corri:luslille malerlaL PSU Gtatea that PSBR does not have the combination of conditions lreqwed to make dlsoerston offlsslon =:-......... rrom Page 12 t"

PRM

~

50-44 UNIVERSITIES Pemsylvanla St.ale Univefalty

~1 Ii§ d

m 11 ABSTRACT graphlte-f'elated lncidenta a ctedble evenL CBG has not pre-COMMENTS (Continued} senl&d lntormallon 10 conclude lhat tueh Is the case for any reac-tor type.

PSU prollldea cha.racteristb of ile PSBR:

• Swlnvnlng pool reaolor, core aha In a pool of walllt';

• Graphite II p,esent In fuel as top and bottom retecsor. i, each element;

• Grapli18 Is p,resen1 as spedal rellector element;

• All graptile Is clad in aialnless ateal widl a dnlgned gap dad and graphite 10 reduce heat lransfer and produce elevated lemperatutee lo promote 1elf anneal,g:

• All i;,aphilll Is below 18 fL of waler;

• The reac10r power Is such that ellher no algrillcant rut llux existl and therefore no significant stored energy aoa.m..ilatea, o, lhe flux and graphite lemperalllre are aufliciemly tidl tor auatalned selJ..annealing, and therefore, minimum 1tored energy accuroo-lallon.

PSU addreNed the CBG's three apedtlo regulatory reqtiremena:

• Graphlie fire respome plans

• Reaclor are evacuallon plans, and

• A graphite stored 81"f81lW measurement program.

PSU believes Iha pelltion lacks a Justiftab'e IOOhnioal basis for a credible graphite fire and therefore the propoaed regulallona are without basla and are moot The convnentar states his lack of belief that a credible scenario exiata for a Ire In PSBR lnvoMng graphite and the release of !is-lion producta. He cites the PSBR approved Emergency Plan In aooordance with 10CFR Part 60, Appendix E requirements and usoc:la'8d procedures es1abllahed 10 deal with credible fire acenariol. Stales In IUlllmary that the petition should be denied because It lacks an adequa19 technlcaJ basis and Is Irrelevant lo the PSBR and other similar reactors.

Page 13

PRM - I 50-44 UNIVERSITIE$ ~1 II 1! ii UniYenllty of Tex. 2/10/87 No atAudl'I ABSTRACT X Cl8s CON80lionl of COO and states !hat U of T haa dellrmlned that it. oonllllntionll do not haw an adoqllllle IIIChnlcal balls and COMMENTS Comnientl . . cplllative. Prinapdy endorses GA repo,1.1 and Pem Stale letter, and alalltS opposlllon ID ht petition.

(Uoll) lhal the roql.Mled changes lo regulationa Ml lrmpproprlaa ID ht U of T motor and other ainlar reli0IOl'L U of T states 1tmt hly have reviewed ht -..ra! docmlenta on radiation eflecla i, graphill

• It perllli'l8 ID TRIGA fuel and refloo.

1or graphlle and lhe Ruaajan report on Chemobyl iThov concludo on ll8 bull of htir review !hat

• There II no Indication of an Invalid analylla ot oonculon on the degree of tmard auocfatlld with research raactor graphlle In pool reactrn of ht type operated 17t Ile U of Tat Austin.

• Thele Is no clear bull or eatimale of risk for* lfgnlflcant hazard from Iha graphille 8IOnld energy,

• Thero wotAd be apedflo hawdl and greater risk from

• program Implemented to meaaln 1he stored energy.

U of T c:onet.n wl1h ht 1m11om8n1a IUJflVltQd 17t lhe PemsylYanla Stale lkiveniity (let18r daled Jan. 23, 1987).

LlniYer.ity of 2/4187 No X Stain lhelr bolef 1hat Iha contentlorw of 1he polftioner are wltlout Commonla are of general and qualta1lw nun. Polnta on graptita encapsuatlon Wi&ccnain merit bocaule ol no leotvica1 basil for applleabl lo* reactor and inmfflion In waler ant notoworlhy. NRC approved emergency plan provides (UolW) lifflllar 10 the 0n1t at U of W. The potldone(a couping of Ire In a fof l'lllpOnN lollrN. No halon produota In lhlNmal colmn waphie. Posib1 on USSR power Nactor wllh W9l8f eno,gy kl graphite Is conlnVy 10 m&IIIU'ltfflffl of ltored Ol'Mlfgy. c:onmlent with Al.ARA philosophy.

lnbmatlon avalable lo u of w. Operalion ~ I n g r a -

phht of USSR reaci>r wu high enough ID uan condnuoUI annealing of stored ene,gy.

U of W does not c:omment on h dffnnce between calclialed and rneu.nd uw.d energy levels. But conclude that Iha llcel-hood of. graptite ftre being lritlal&d l7t. ~ releate and

• 1'811.dng In lht release of fialon producls ii more ~

In u of w 1'811CtDf, ~ Is Uled In:

• Top and bonom end rellectors In l i e ! ~

• Aeflec:1or elemen!a

• Thermalcoll.mn.

The Int two components are oncapsuCat&d In alunmum or ,au,.

leas steel 10 llat oxygen II excluded. Bofl componera operate at high enought te!Jl)eratures that lnl6alng keeps Wlgne, ener-9'/ at low level. Bolh operate Wlder 20 ft. of water that povldes Page 14

PAM ~

60-44 UNIVERSITIES IJnlwfaq of Wlaconsul ii II 1! d .

ABSTRACT adequate coollng lhoud a Wigner energy NMeaae lake place and .

excludn oxygen makfng c:ombultlon lmpoaalble.

COMMENTS (ConUnued) The lhermal column graplile ls oon1)0Md of 4 In. square ~ 3 ft.

~ lfringera. The llringef'8 are not encapeulded to excbie alt, a YJlgner enwgy release resulti1g In a fire would not CRUM tialon product ...... because:

• The graphite pieces Int large enough to be clllota't to lgnll&

• The fut neutron I'm In ht ilemlal eolwm II IIO low 1hat a ldgrj ficanl amowl1 of lklted energy la not generaled

• There ll no fuel In lhe lhermal ~ ao no liulon pnxtucta woud be releaeecl If* . . did OCCll'.

Clms NRC approwd Emergency plan u provldng lire mponllff and concludea no furlhlW plans are needed.

Dedlll'N ht pelltloner'a reql.lNt lor a 11of'ed energy meuu,,...

mentl would lead to umec:eeaa,y raclallon expoan IO peraonnel since hQ la no hazard.

Univeraltv of 219187 No X The Director (14 yrs.) of fl& lJriver8lty of Arizona Nucla* Raaclof' Commenler prcwldes specUlc lnformallon for U of A Malk I TAIGA reactor showing Arizona (U of A) Laboralocy ltales that his operadng phlosophy and 1he phllo- low ~led power (c8 MWD), and geometry comlderationa that rule out graphile aophv IIIUght to the studenta II *M. no work lhould be done which burring In 11s normal c:onfigura1ian. &aft should look at LOCA condlliona relating 10 imfolves expoaure to radiation If lhat work Is of no benefit* conci1lona neceasary ro, graphite bumi1g.

Reminds ht NRC of lhe no greater lhan background radlallon In addlllon, Staff evaluation muat consider 1he Al.ARA prlnclple

• namely, reqwo-condNoM of h U of A Mark I TRIGA and the amual rapol1 of no l'T'llmla lhat lead to umecessa,y radlallon expcmna. Comments on annual repor1a moasurable rad1allon exposure dose to any alUdent or facuty IO NRC that conlalns lnfonnatlon on lntegraled powe,. This lnfonnatlon should be member above natural background. provided lo BNL Descrlhn ~ U of A TAIGA and the locaMom of graphb& In lhe reacb' (I.e., 3.4 inch-long fuel element end pieces. rellector, aJu.

ninum clad, 36 inch Inside dame1er). Stams flat fuel and graphhe are always covered wfft at least 14 ft. of water. Sleady-alate power 100 KW. Pubsing mode i. 600 MW conesponcflng to 16 mega-jouea. Operation for laad"ing p!MpOMa la Ina than 1 KW and the lnlegraled powe, ro, the reactor la leas hln 8 MWO.

Belleves

• WlMlfSity la trai'llng ground for atudenl for majo, respondliliha In fie mx:lear lield.

CUes that under walef graphite dad with alumh1m canno1 bum.

• Reactor la designed and conatructed IO that fuel and graphil8 lranlfer heat lo lhe water whether It~ . - In sleadv atate Page 15

PRM l

~1 ii 11 ii 50-44 UNIVERSITIES ABSTRACT COMMENTS Unlvefslty of opera11on,

• pulse or Wigner energy release (namely, syalem la Mzona not adabdo). Fuel and reflectDr are Wider water and wl not be (Contlnu!td) removed.

• Requlremenla ID remove graphlle tor Inspection are without benefit and would red In algnlftcant radiallon lo pe,..omel In vlolallon of the A1.AAA pmclple. Ns,aphltit relleotor WU remowd from pool and c:ladclng rell'IOY8d and were lo bum, lhe fuel W0\'1 be safely slol'ed U1der water and so woold not be ii-VOMKI. Therefore, It could not reaut In a Chemob>tl type release

• Oetenda dle right of lhe petitioner ID uk queatlona about 18Jety but ~ ooneceuary expendllln of university and gowmment resources.

• States his oppoaition lo lhe peffon and lhe p,apoaod Mt u being U"IWIN, unneceasa,y and a very nega1lve polk:y.

Univel'lltV of 2/87 No X Ar;lmenta using Information auembled to, Vie TRTR c:onutllrll- One of the more ~ responses lo PRM-60-44. MURR'a c:onments Miuouri Reaean:h ty are presented to estabbh flat fie CBG pel!llon ehould be re- and accompenylng appendlcea p'OVide delala not bn:f In ht rnajot1tv of Ile Reactor (MURR) jected. The argumenla deal with mistakes and lnaccU'acy In the oonvnenta receNeCI. TheA comments desetW careful nwlew b)' lhe Staff and pelillon and present evidence why CBG'* concerna art unfounded BNL a a part of .... utety aueumenta and In preparing fie NRC l9SpOflle regardng MURR'a specllio appllcatlon of graphlle. lo the pellllons.

MURA COITl1'lentll en Iha two lnues raised by tie pefllon: (1) The

~ of * ~ Ire at research reaclOrs, and (2) The potential contribudon of Iha Wi1J,er (stored) energy to the auto-lgni1ion of research reactor c,aphhe.

Comp.non of research reactors lo Chemobvf by fie CBG Ignores lhe extreme differences In power level, core llze, laaion product Inventory, operating lemperature, reaccr control systems, and Inherent deais,I characlerislk:a.

MURR also disaQr89S wllh CPG's inference that graphila fltea were lhe Initialing evenlll In boti Chernobyl and Wlndscale acd- -

dents.

Commema ant Pfovlded 10 lndlca1e lhat graphite ma at Wlndscale and Chernobyl were corollary evenla. MURR alao di:smlasea CBG'I conlenlion 1hat stored energy in lrradialed graphite played any role

~ Chernobyl because of the 700" C graphite temperatures. Cites lhe l&pOl"IS from the Commitllle of lnquery lo eslabllah that stoted Page16

PAM 50-44 UNIVERSffiES Univermty of Uluouri Reseatch ii ii 1! 1! ABSTRACT energv,dd not result In high local 1e""8rature In h t ~

COMMENTS levent. 81atn lhat for both accidents relerred to by CBG. lhl melt-Reactor (MLIRR) Ing of W lritated the b.mng of wapfile and not vice veru u (Cont/rl.Jed) CBGlnfet'a.

MURR then prooeeda wilh dlscunlona of &he folowlng ll0pb and relovanl ~ and conc:luslo,,..

II. Crd:lllilvofs,aplbeflrealn ruearc:happllcdona.

SUmmarizee that lht tact lhat graphlle will bl.In ca.mot be refuted.

However, the queadon cemera on whelher or not research reactct graphite firu . . credible or even If there la a credible lnltlallng event A transient lhat could lake grapNte In a ctedllle lnlValng large block geomelly 10 600° C above operetlng lempefatute II not COl1llk:lered crecible. Tranalenll ol tlia ~ a,e p-e-duded In research reacl:>I' lpf)licatlons by lie Inherent safely design IHtLns (paaslve) of reselllCh reactors as wel as eOQi-neered redlRlant Alely &ystelTB (active),

Ill. Wigner energy In research reaotor graphltit.

Murr stuaa lhal C8G doM not explore 1ho fact hit the release of Wgner *l"llflW IIBell requires an illtlaling event h c:lleS caJori.

n.trlo and combustion *xperlment& 1hal haWt conslatenlly llhown ht until Ile annealing IDi1lpenmff exceeds the lrradiaJlon

' lamperalure by some deflnlle amoont, stored enet;Y will not be

' ' released. The ttveshold lernpera1J.Jre lnorease required before any releae of 8lored energy oc:cwa varies mn 50" C 10100" C above lhl Irradiation lemperature. CBG'a conl8nlion lhal lhe a10rec1 energy In grapt-Mte la aufflcient 10 ralle the temperatunJ or s:,aphillt several hundred degn)es. MURR concludes 1hal even If CBG's conl8nlion were In.le, lhe lempetatUre excuralon poAt-lalltd II lar leu lhan the Ignition IBmperalUnt of graphite ot the melllng temperall.re of fuel or fuel oladc:lng.

IV. Reeponses ID other CBG commenl:s.

MURR cia the CBG'a atlack on tho AJdy done for arelyzlng cntdiblt acddenta in Argonaut reacl0fs and lta extrap:,latlon ol what flay perceille as a deficiency In lhla document '> lhe aa'8ty

..ttidlA,t MnA for ell -* ,__..__ t".Rr.l dont not P~17

PAM i

II d1!

50-44 UNIVERSITIES ii ABSTRACT COMMENTS IJnhN&ltyol aubatantln 1ta arguemem regarding ht lnldequacy of iw Mlaaouri Re8eaf'ch Hawley document rwch lea the safety Ill.des done by other re-Reaclor(MURR) aeardl reactors (e.g., MURR design analpis ooraldera both the (Continued) al0red enerav c:on1ent and the potemllll Wowfl In r e ~ graph-lie).

CBll's atatementa In their petition ab:MJt the Inadequacy of em<<*

gency plana Is rebutted. MURA dies recent upjJ"&da8 In emer-gency pans for researoh reaclora ID l1andan:lze and ID meet the requlremenls of ANS 16.6 Reg. Gulde 2.S. Propoeed Revtalon 1 10 Reg. Gulde 2.8. NRC haa reviewed end llflPRMld dl8le plans.

MURR concludee that CBG does not pnm that !he etTMlf98'1CV plam are Inadequate; they l1lOf'8ly alllle lhat they ant, u Htheir "0plnlon" la fact.

MURR cltea CBG'a auggeation tat NRC have lcenseea eu:,mft

~ plluw baled on a 25% release ID fie environment of

.,.. equllbium radloacive lodnt lnYenlory and lhell cllim tat Illa release fraction waa derived from ANS 15.1~ MURA alal89 lhat llelr nMeW oC AN815.16 Reg. Gulde 2.6 and 11s proposed nwislon 1 and find no auch inveM>ry re!eae hclion Sliggelted.

MURR b1ler stale8 that !his lnvenlofy la derived from Reg.

Gulde 1.4 and represents* ~ t e oont melt and lou of con-l8inmenl ldu'e. ReseatQh l1Nlel0r ... accldentlS would not approach 1hltl CBG lnventofy release; In fact, 0. oalcuated iodlnE releu.e fractions for a pe.r1lal core melt wl1h pa(dal lou of laolallori at a poMr reactor are leu flal 0.06% (8X10 *4).

MURR also take& luue wfth CBG'a poposed requirement tor the lc:enaees of research actora lo perform ernperical ----

ID determine the slol'8d energy content of their graphite reflectora to+/- 10% accuracy. MURR raises questions relative lo CBll'a underalanding of what wolAd be lnvofved In making lhla detet-mlnalion lncludng the expost.na resuling lrom hancl1ng kracl-ated graphill. MURR slates lhat Iha local 8b'ed energy ls not aa lmponant aa the average value <?f atored energy OY9f a graphite

' block alnc:e fie klnetk::s of energy release h a block o f ~ .

~ from possible short lived local transients, fie maxlrrun Page18

i I

PRM 50-44 UNIVERSmES lWveralty of Miuourl Research 11 II 1! 1! ABSTRACT

~ reached wlH oorreapond to an average value ol ltofedenergy.

COMMENTS Reactor (MURR) V. Applca11or1Sof Graptita al MURR.

(~ MURR clescribN lhe cl-.raclerletlc of the graphlle , . ~ and hlmw colunn The relector Is camed In alumlnun and during l'8IIQ)I' operadon has a YOlLme average IIN'll)OtatLn >160" C.

The themlal colunvl g,aphlle la within a wuar-jackeled abnlnum cuing. aeparated from the releclor by

• lead gamma lhlald and enclosed on the opposite end by a ll8el hnnal colurm doot.

Agu,ea lhowlng lheae rMCtor compontnll are provided.

VL Analylla of ltOred energy In the MURR graptite reflector.

A. Slonld energy calcwltlona.

MURR slaleS lhat the domlnadng faclDr c::oMOllng the dA'lbu-ton of alor9Cf energy IICOlN a waPR1e block is the local l1.lfV'llrll teqlperatunt. To estimate ll0red energy accurataty, hH dlmeneional dl:llrlbullorm In lhe graptile ol the fast lux and of llmperaue are needed. Since lhla Information wu not avaf.

able, the *lored energy was cabialed by using average lem-pendl.na and average fast flux valun derived from compu1e1_

code and delVI analyses data. Uulg a maflodology un1lar to 1"1111 used by Pearlman, and the alored energr equallcna and parameters presemed In Nlghtingal'a nuclear graphite (p. 329-331) w.o meflods wem 1.1118d. One using lhe Wlndscale cuvea and dala and the other using the Hanford dala (30" C t.lWW) -

MWOIAT COl'1Yetslon tac10r from fast nvt used wa tie one tug-gealed by Nadonal Carbon Co. E.amatea made using lhe Windscale data 5J&'-'.9 an average stored energy ot m cal/gm, IJlil,g Ile Haiford data a calculated value of 133 callgm. _/

B. AnujaJa of lhe poalble hazard moclued whh lhe calcl Mt8d m-edenergv.

MURR concludes lhat even l.U'IQ lhe \Nlndlcale dala, conaer*

valve heat translef' aaumpdona and reSeate of tie total alOnKI energy In i,aphile the lolaJ waptite energy release la leala lhan the core decay heat rale. The LOCA a,alysia fot MURR using the decay heat source ftnda no consequent lJel damage. A '

lower heat load would oresent even leala of a concem wilh Page19

t PAM

. 5D-44 UNIVERSITIES I.Jnlwrefty °'

~1 Ii 1! 1! ABSTRACT regard D tua damage. MURR alao contends ht ht >150" C COMMENTS MluourtRelNnlh operalng temperature of lhe bulk of flt g<<-aphhe rellector meana Reacklr (MURR) flat regardleu of h IDfal 8lor8d energy In t h e ~ . llt no (Contiooed) point during 1he releaM of energy can ile graphbl achieve a self suslalned llllmperaue eXCW'lion.

C. Credibily of graptite lglltlon In 1he MURR retlector appllcatlon.

MURR concludes that:

• 8el ll.8lained lemperalUte excuralonl ant not possl>le In ....

MURR ieflector

• Even wumlng a adaballo temperu,lnt lncreue of 700" C

• oxidation or combuation cannot occur In flt encloaed environ-ment with. designed ablence of oxygen (camed graphlbt, heltlffllilled)

• Retleotora are nonnaly under 23 ft. of waler wi1h an emer-gencv pool fill syalem adeqLut ID ITIUllaln 3 ft. of watat' above a complelety severed 6 Inch beam port which enaa.ns  :

flat relector Is CCMtf8d wkhwater even under 1l1Ne accktenl COlldllionla.  ;

VIL Analpls of ld0nKI energy In thermal ooh.um graphite.

MURR concludes lhat '

• The graphite In Ile 1hennaJ column hu

• very low stored en&rgy content.

• Even the maximum local atored ene,gy la Inadequate to create a peak adlabalio temperature rlae great enough 10 cause graphile Ignition.

• The graphite stack Is ccmpletaty enclosed In an aluminum water jacketed caalng with entry gmed lluough a greater than 12 ton lf8et and masonlte door. This graphlla Is not exposed ID 11\Jffiden air vokJrrw to support a fire even If one could be lnl1!ated.

• There are no hazards associated wifl Ile calcuated amoun1S of ak>fed energy In the thennal column

• There are no credible IAlllatlng evera for a Wigner release or butrwlg of gtaphl18 In lhe !henna.I coltMnn.

Page20

PAM Ii 11 II 50-44 UNIVERSmES sl ABSTRACT COMMENTS Univfflityof MURR lncludM the tollolNing Appendices~ delah of their MlsaouriResea,ch calc:uldona and analya1a:

Reactor (MURR) Apperllix A* Graptvte Relect>r Elenwl1a (ContlnJed) Appendix B

• Thermal Column Appendix C. Analysi& of Hlmwds ANoclaled \Wf1 S1anld Energy Release In MURR GrapNte Re1eotor Appendix D _, Analyail of Hazards Auodated Wllh Sl0f'ed Energy Releaae In Thermal Coluzm. "

Test Research rd 2/2187 No X ITtia IUbmit:bll from Ile TRTR group was Jnlendsd to be an amo- The Information la pow:led ID NRC as an aid 10 ht Slaff and their cor1raC1ora In Training Reactor tated reference 10 relevant Information regarding stored energy In perfonnlng !heir satety asseament and evaluation or !he CBG peUon. It 1s a uae-(TRTR) Group graphite and graptilB oxidation and combudon. The doament was lJI compldon of lnlormalion and referencea 1hat ahould be used by BNl and lhe prepared by Wall A. Meyer, ..-. of the Univensl1f of MtllOUtl Research Staff.

FIHokl, (MURR) tacatv: It was povlded 10 !he TRTR grot.>> 10 aal&t lhem In aaeaslng s,aphlte *klnld energy and Ql8Phl1e c:oni:luatlon In ,.lion 10 their apeclllc reactor lnatallallon. The following IDplca ant c:llscuNed, quantitatlYe data lf9 pesenl&d, and reteiencea and

• blblography .,.. provided.

• HstDrloal perapecllve.

• Comrlfllons between MWDIAT and llYl

• Theory of energy storage In lrradal&d garphlte*

• Ameaing of amred energy.

• Suntlon of sl0red energy accumulalon In graphl1e at Hitt lrradallon exposure.

• Melhodl of detemnng stored energy In Irradiated graphite.

• Exarq:lles of pat *lored energy measurement and maxlnun amperatures 'reached dumg anneal at variot.m rHllltln.

• Graphite OlCldallon and combustion.

-

• The aclabatlo model: delermlning !he maxinum te111)en11Ure aaaodal&d wit! tl0r8d energy In graphlle.

(

Page 21

PRM  !

50-44 wi INDUSTRY GaaCooled

~1 II 1! is 11/3186 Yes ABSTRACT Requests ex1enalon of comment period. None COMMENTS Aeador Aaaoo. 90d GCR DOW Chemlc&I 11/3186 Yn Requeala exlllnlion of comment period. None

.U.8A 90d GA Technologlea 11/3188 Yea Requem ex1enlJon of comment period. None 90d I

Pub6c Seivice 11/3186 Yes Reques1s extension of comment period. None Company of 80d Colorado

/

Page 1

PAM I z z Q

50-44

~ffi I INDUSTRY We.Vnghouse ii ii 11/3J88 1! ~i X

ABSTRACT Owm and opera'99 a 10KW pool-type trainl~ reactor. ReaolOr COMMENTS Westlnghoc.me'a aaesement of lhe problem and their recommended approach la l:lecwlc rarely operates above 100W. AWfll98 power alnce oont modified ~ reaonable. liowevef, It probably reflU1S from htlr conculon l1at the Corporation to Include grapNte reflector has been 19W. *...c:hltlo -.-..: propoeed by CBG ~ not be warranted for f1e Wednghouse Comment& l1at pelllion le Impulsive, emotionally based, and NTR. I BNL la able ti provide a recommended l1andan:I cak:uladonal method 1or founded on f1e Chernobyl accident which la efgnifloanly different u:i,ed energy, 1he ***.aynmado approach. **" they recommend shoud be i'dJded from research re8">1'S. Chernobyl ac:ddent was the result of a, a poatie oplion. Unlortunamly lhe detah of their stlnd energy cak:ulalion irnerrelated human factora and design weakneaea. are not provided.

Bellevea posslbl~ of graptite Ire In a US reactor Bhoud be aaseued, however, wants aaaesvnent ti be completed In a.

logloal, systemallc manner. Westlnghouu auggem lhe Jollow..

~ ayu,malio approach:

1. Each lcensee woud perlonn a detaled analyais 10 quan1lta-tlvely deterl'nh, lhe stored energy In lhe lloeflle8'a graphite.

2. From lhe9e .cak:uallons, lhe licensee would determine the probability or occurrence of a gra.phlle llre.

3. Al lleenseea would aubmit these anaJyaea to the NRC lor review.

,4. Hthe IOtal stlred energy In the graphile or lhe probability of a graphite fire OCOI.ITlng exceed& t,Of'M predelemllned value, then the lkjel'llleM should be reqlhd to verify tu exlating evacuation plana would be suitable fell' response 10 a graphite fire emergency or fonnulate and IUbnit new fire lghtlng and evacualion plans and FSAR revision.

Westinghouse provldea p!'811n'inary assessment of ileir reactor concluding that lhe reactor without waler la subcritlcal and a!ao becaule of low power opetalion contalra wry 1maR quanlilies of fission products. Analyail of graphite re~. aqur,ing flux at refledor same aa llux In !he core, calculataa 0.0806 ~ of stored energy In relleclor auemblle1. This stored energy, If released, woukf Increase graphlle tempenwre by 1°F. Maximum fuej plate lernperall.lre for rnaxlrmm ctedlble accident Is no grearec than 320°F. Conclude that even If rellector graptile were at temperature of fuel plate that lhe oxidation of graphlle is negligi)Je at 321 °F.

Page2

PAM z z

~0-44 ~ ...

ell (0 ~1 ~i Q

INDUSTRY Gas-COOied ii 213/87

~i !s is No ABSTRACT X GCAA rep,eseoll a nwnber of US utllltiea Interested in gas-cooled COMMENTS Similar in content to the comments received from GA Technologles compamg Reactor reaolor development and convnerclallzatlon. Comments apply to Cherno¥ Nactor lo 1he commercial HTGR Msoclalea tie pelitlon only as It relates lo commercial gu-cooled mactor lnfonnaion Is relevant_, F$V and should be faclDred Into tie NRC staff response (GCRA) deslgnL to petition 1'9qUBll1s regardng Fon St Vraln.

Baaea tor petltkln: The generic HTGR concerns lhoold be conaldeled In the ongok,g safety evakJa-lion of ile apecifto designs for these plants.

-

• Graphite llrea should be considered .. credible events by the NRC aa a result of tie reponed graphite Ure which ocourred at Chernobyl

• Stored energy effects have been undentstimaled aa shown bynewdala iThe crec:IUltt of s,aphile fires Is adchned ~ that the CBG beale faUa to recognize the requirements for a graphite fire and lhe algnlflcantly c:lfferent probabitlell of events, dependant on reactor dealgn. that coYld fd dealgn moaaures taken lo avoid such fiJeS.

Commercial gas-cooled reaclDr dealgns operate at lemperaturea high enough to permit rapid oxidation, but use helium cooJant

\6AJich la Inert and excludea air/oxygen from cortact with the i,apl'ila. The Inert coolarw Is contahled by

• high-Integrity pes-MH vessel. Chernobyl reaclof had hiljl 18mperal:Ln graphite

proleeled by a He and N I rrix1l.re cover gas contained only by a low-pvuure hermellcally seaed vaut Explains that the Chemobyt accident was cau&&d by one or more rapid power excwslona due to the Inherently poslllw faecllack from the void coefflcienl of the waler coolant. Rapid healing of the waler In the reactor coolant channels led IO an exploalw volumel*

rlo expansion of the waler lo steam resulllng In grou lab& of bt reactor valAt allowing air aocesa 10 the graphite moderator.

It Is argued that Chemobyl accldent has m impa.ct on the auus-ment of the credibility of a ainllar event occ:urmg In the convner-clal gas-cooled graphlle moderated reactors.

Coocludea that deslgne for commercial gas-cooled graptile mod-erated reactors operate at lemperalUreS weU above the lempera*

ture at which any eignlllcant H>ted energy occwa.

Recommenda pelidon be denied lo the extent It appllea to com-merclal gas-cooled graphite moderated reactors.

Page 3

PRM dd 50-44 INDUSTRY ii II ABSTRACT COMMENTS Public 8ervicl 1/87 No X Summarizes petition requests and lheir bales. Adclrelaes c::red- PSC's ldenlificatlon of design differences belW&en FSV and Chemol:l'il should be Compenyof blU1v of graphlle lints. ~ fahe or CBG t1 dta lie C8I..UlhM Included In responae to pedtlon. Si.mmary of NRC'11 ovaJuatlons and coocluliom Colorado mechaniam or lhe signffioant dffentncea In ANldor design of PSC's submittal on the hypolhetical oxygen lngreu event shoud allo be stated.

between 1M ~ plant and Fort St Yraln (FSV) plant. psc. aaaeumant of atored energy should be reported.

ldenliflea OIWMIM!I mechanlama at Chemobyl:

• Power eurge due lo large reaotlvlty lnsefllon

• Feedback 1rom poelllve void coefficient

• Operation of reactol' outside Its design basis limits PSC has addressed varioul reactor aocldenta reaulling In graph-Ile Olddallon In CP & OL proceedlnga. CWrenlly, '1ie lnformalion may be found In updated FSV-FSAR. None or the poslUlated ac-cldenll lead lo lignlfloant c,aphlte oxldam. PSC concludod that an accident Involving slgniflcm,t graplite oxktdon at FSV iii not c:redlble. AJso states NRC reached this lllm8 oonduslon. P8C recentty determined that lour un!Jlllne01.18 iidependent HUO-11Jral fallutea would have lo OCCU' 10 permit oxygen (u) low lnlo the FSV PCAV neoenary to lnillata and ti suatul av,llleant oxl-dalion of the FSV graphite. Even 1holl(11 no credible mech&nlam was b.m ti lnlroduce aufflclent air for a!Qniacant graphite oxkla-lion, PSC haa Identified a mechod that could lllnnlnale auch an l event Further c:lescri>es lnfonnalion provided NAO on 1lu metlOd and lie design (lfferences between Qiemobyl and FSV.

FSV hu an emergency plan~ povldea for acdon lo J)IOleCt the heallh and safety of the publlo baaed on oflslte c:loae8. Deems a apedal graphite ire evacuallon plan unnecesamy and not coat effective UlCO resoun:ea would have lo be expended umeces-aarflv lot development and maln1enance of unneceaaary pro-cedures and documents.

PSC clles !heir evaluation of 8D'8d energy:

• Not sigllflcant operating concern at FSV because It be-comes negligible at temperatures >500°F (FSV graphite exceeds this temperalwe at powers >7% or full power)

• Uoperated at 5% lor 30 days. adlaballo release results In only -'°C calculaled temperaue rile, which ls Insignificant reladve to design temperallKe of FSV core Page 4

PAM, 50-44 INDUSTRY Public Se~

Company or ii 1/87 ii 1! is No wl X

ABSTRACT Recommends lhat petition tor rulenaldng, to ht extent lhat It applN 10 FSV be denied and conaudes that requetts made by COMMENTS Qolorado the pedtlon are wilhaYt merit when applied to FSV. lmplemenla-(Corutnued) lion would place a luden on PSC'a resources wit! no muting benaffla k) publlo health and safely.

GA Technologlea, 2/3/87 No X Provides IUITIITl81Y of COO petition and GA._ Ul8SSfTlent of tie Conments 11111 relevant to helu'n-<:ooled HTGRs lrduding Fort St Vrain. These Inc. Chemobtl IICddent. Presents lhe dlflerencu between the HTGR ~ should be laclOred Into staff's response 10 pelllloner'B concems tor nlheRMBKreao10ra. Fort St. Yrah

• HTGR has He coolant and wffl not react with QH'alT1lc fuet pal1lcle coatings or grapti1&

• AMBK bollng watef coolant can reactwilh zlrcalloy rue!

cladding wry quickly at 1empera.11.ns reached under transient condltlonl

• HTGR has negallve reactivity feedback OWlf Its entire range or opeta1lon and Iola of ltl single phase coolant II negative

'

• RMBk'a two-phaae bolling waler coolant In comblnalfon with other factors can result In a positive reaotMI)' iwertlon and feecllaok when coolant ii lost or voided

• HTGR fuel is llermaly coupled with lhe graphite moderalot',

resuUilg In a core flat heats up slowly dutflll power or llow 1nlnslen1s

• AM SK'a fuel is not lhermaly coupled wilh lhe moderator, and the lhennal reaponae lime of i1e cote Is linilar to lhatof LWAa Tone ~ preclude an HTGR accident of 1he type that occurred at Chemobyl and make an HTGR accident that Involves signlkant graptile oxidation highly lnctedll:;,le. &ates that 1he ire lnitiaMy ob&en/ed at Chernobyl WU cmaed by bunTig fuel and daddlng, not bl.ming graphite.

GA's evaluation of stored energy at Fort St Yraln Is operaled at 6% power (below 250°C) for 30 days ~ lhal If releaaed adabalically would lnaease oore l8mperature by less flan 5-C.

Recorrmenda that NAC deny petition.

Pages

PAM  % z 2

Ii !I Is 5().44 wi INDUSTRY ii ABSTRACT COMMENTS GA Technologies, Inc.

212187 No X GA lmfts lheae corrrnenlil on lhe petition aa It,..._ to t,aphile In mlGA research reactors. Commenls on the contenllont .-id GA convnents provide a good summary c:lascripllon of ht mtGA graphll& com-ponena, their looalion, mt the remts and conclusions !hey have reached as a request f o r ~ in the regulaliona 10 requlr9 Operu>rl of result of hllr analyslL The lll1lllyaea In the auachment present aome numerical research reactors IO: mulls and asaumptlona, but the delala of the analyses and caloulaliona are not

• Prepare and submit t.> NRC for approval fire reaponse provided. BNL and the staff will have ID detemine the reasonablenesa of lheit plana and evacuation plans for a graphite fin, concluaions latgely on the baela of qualltallve ltalel'nents, the conservatism In

• UellSU'e the energy alofed i't the graphhe !heir asaumptlons, and 1he acceptabilHv of their argmients. Only one relerence la 80 maat luenoe (60 Kev* 10 Mev) :!. 10 nvt after nearly 30 years of intennittsnt oper

• Revise SAAa k> consider the rieks and consequences of a graphlle Ire In lhei' lacllity atlon. BNL may want to compue the GA table for releuable swred energy as a function of lrrada11on IBmp&raue (l'IYt > 5 x 10.,) wilh their own results.

Cites the basis tor the petition la CBGa COOIDl"dfon lhat lhe acddent at Chemobyl ahowa graphlle lra:a lo be ctedtie events and that licensees of rnearch reactora do not have adequate fire responee no, evacualon plane br a s,apf1la tire In lhelr facUities. CBG also conlenda thal based on hllr 1n1erpretation of recent reports by UCL.A lhlt Wigner alored energy has been severely undet'Ntimaled In earlier all.ldlea.

GA befleves thal theae propoaed requnmenla are unneceuary

/

and wookl be uncill)' burdensome. GA ht addreuea the Wls,l8t' stored energy and *ta.tea the C8G hu made Yef'/

aelootive uee of lhe data In the UCLA paper ID euppo,t Ill contentions explaining lhat the high value of Wigner et'l8fVY mrage ~ferred to by the petlloner was a alngle dala point and was not representadve of the behavior of graphil9 at a whole or of lhe tolal stored energy In the core. The COfflllele data In the UCLApeperaupportstheredam~lnbMC-fwded generic *Lldv of Wigner energy al0rage (NUREGJCR-2079)

In argonaut reactors.

GA presents argunents flat wl1 demonatrale:

• no graphite firq' can mull for any pou~aled reactO<

operation and conclude lhat no Juatification exists to revise existing safety analyses with regard to the potential consequencea outl1ned In lhe petition or k> ~ conlingency plans for graphlle lites or 10 obtain measurement of stored energy.

Pages

PRM 50-44 1Nousmv

~, d ii l1 i§

~

~~

z ABSTRACT I

COMMENTS GA Technok>gie11 212/87 No X GA also llal8s that experimental measuremet"ltl of 1Dred lno. energy 'NOU1d Involve conaiderable adcllJonal riaks arillng from (Conlrud) handling Irradiated fuel lhalwould be connry 10 Al.ARA. The pelilioner referet"IC:CM the attached letter aa providing l!IUf)POling cakl:!lla'°'19 pe11inent to these concluslona. GA ....-nrna.!izea the Information provided In the attachment u Jollowa:

• A table of lnllXffiJffl releasable atDred ene,gy u a blcliofl of lrraclatlon tempera.ll.ns togelher IMlh lhe auoclaled temperature lna'eues In Ile event lhe waphl1e reaches lemperalU'el > 1204'C to Initiate the releae of stared energy .

• Ast evaluation of the effecta of 1h11 potental energy release tor aeveral graphi1e c:omponenl9 (I.e.; lhe graphlle end refleclons In each TRIGA fuel element; lhe graphhe radial relleclora In 00i1Bln TRIGA facllltiea; and tie 1J11phll&

dunvny elements uaed In some TRIGA cores)

Presenrs, In sunmary, the result& of lheu analyaes:

• Graphlla end rell&ctora * &laol clad and alurilum clad fuel aurvlvea al nonna1. abnormal, and accldenta (Including LOCA) for operation at250 KW and 1 MW

• For the graphite radial relleclofa operating at 1 MW, the graphlle lempel'U.lle la >120°C so that releaaable atored enerw Is small (S 45 caL'gm). Hthla energy should be released, the Interface temperatt.Ke of lhe ah.mJnt.m.cfad graphl1e lnlerface la far below lhat required 10 endanger lleclad

• For operation at 250 tWI, reaulla show tat the graphite temperaues a.lwaya remain low (S 60"0), and Iha maxlm.un releasable energy (260 calfgm) may be available. However, because of fie relatively ama1I imgra.led tux, tie actual l maximum stored energy wil llao be low. YMI LOCA, the maximum graphlte-alumlnum clad Interface temperature Is only 70"C which Is below that requred to lnldate fie rele89111 of stored energy. Therefore, GA concludes lha1 at 250 MW, the graphite reflecior Is not at risk from the siored &l'lflll1Y In graphite Page7

PAM 50-44

~I INDUSTRY GA Tec:hnologles.

ii II Is 2f2/87 No 1!

X ABSTRACT

• For 1he aluminum-dad graphila dummy elementa, the effect COMMENTS Inc. or releasable atored energy la 1nconsequen11a1 In an normal (Conllud) and abnormal operations or al TAIGA rnctora. For lhe LOCA, bolh at 260 KW and 1 MW TAIGA reaotors, the alumh.m clad wil probably melt; howev<<, no graphlle h can reat.it I

GA concfudea 1hal In al lnllances aneot1ng v11a1 corr.,onen1a ol the TAIGA reactot (e.g., fuel elernenlt and graptillt radial,..

leolor), the release of aloted entt'l1/ from IJllPhfta componenlB causes ordy benign resulls. For 1ht I.OCA, lhe poalble lose of the abnlnJm daddlng on a few dummy elements does not con-alllule a hazard for 1he facity or ID the publlo from ehher radio-acdvily or graphite ire. Since 1here is no rilk to 1he public heallh and safety from 1he a!ored energy In lhe graphite In TAIGA l'NClors, and u demonstraled, the pelillon u applied lo TAIGA reactDrll Is not based on proper1'/ evaluatad facts and Is widlout IIUbltanc:e, GA recommends that the pelllion for ruSemaklng be denied.

Paga8

PRM 50-44 t:iOV~RNMENT AGENCIES Stale of Rhode ii 10/20J86 II 1! 1!

Vu QOd ABSTRACT Requests extelllion of corrvnent period. None COMMENTS Island Alomlc Energy Committee National Blnau 10ll/88 Yes Requests ex1enslon of comment period In behalf ot TRTR. None of&andardelor 90d TRTR (Teat, Research, ,.

Training Reacto11)

StaolR.hode 1128/87 No X Comments ate baled on lhe ope,don of a plate type reaclDr at 2 Commenla cootaln general 1ta1ernent11 but no details on the actJaJ l8mpefllh.ns IIW'tdAIOmlc MW wlti water moderation and cooling. The ruc10f' Ul8I graph- adveved In ltMt graphltll reflectors. Thefefofe, the oonclU&lon on stored energy EnofgyConmttN 118 c:amed In alurrinum a a relecb'. The OBA lor 1h11 l'MC8)( II mutt be conaldered qwlitallve. However, tie qualttatlw auesament of the a LOCA In which the bollot'!l 7 In. of tie bl elementt and tie polentlal lor h i'I tho graphite are based on phyalcal reali1les and 5houd be use-reloctDr pieces remain lmmenled In pool waler. lul In developing tho NRC response 10 the C8G peHon.

Dtling operation of tho reactor, temperun of lhe sraphlllt re-ftector and the hoallng from core neuTOnS and gamma rays II tuffldent to cause conllnuout annealng of some of the Wigner entJtW In tie graptila. Aa irradiation racb:ea f1e lhermal con-C11C1ivily, lhe graphlle temperatl.lrll ilcfeases and annealilg

~ more pronounced.

CIIN 1he need to have an lnl1latirlJ event 1hat Increases tho graph Ile temperalUre tome 50-100" C ovw the temperaJ:tn of 11w

~ ~ normal operation. Noles that 1he ll0fed energy released la proportional only to lhe above ~ lncreue and hi the lolal stored energy la not ntleued.

By comparison to data presented In referenced documents IO a 2 MW reactor they conclude that cbing a OBA that the tempera-ln In 1he graphhe retleclor la not lllfflclenl IO begin the release of alDr8d energy and that even If an Wldelined mechanism In-creased the temperature of lhe graphite IUflldent ti begin the release of stored energy, the tsmperatin reached In the graph-lie from the release would be too low ti create* problem Jor ellhe

' ht g,apN1e refleclor or the iJel They atate that the potanaJ for

&re In ht graphite relleclora la very amal duing alDrage UlCe ht grapjle la enclosed kl evacued akJri'wm cans. Ntit, dl.l'i'lg ot after ~ In the reactor the canned graphite rellectors are atored Page 1

PRM -

50-44 GOVERNMENl ~iI, 1! IsWI AGENCIES State of Rhode '

11 ABSTRACT I

COMMENTS in the reactor pool undet large quanlltiel of waler. Conclude that Island Alomlc the NRC haa correc1ly aueased ht ufetv ~Ilona of graph-Energy Committee lie* ll la Ul8d i"I research reacl)nl and flat the CBG peffon (c.ontlnutd) ahould be denied.

I Page2

I

.e PRM I I

50-44 I INDIVIDUAL cmzENS Harry Pearlman, PhD 11 11/3188, No*

I Ii 1! 1!

I X ABSTRACT Alegallonl and olalms are not supported by available facts.

COMMENTS References on measurements of stontd energy In UCLA Argonaut Reactor may I

Graphite SIDred Energy Measu-emenls

• Cites paper~ at prOYide an inportant means of establlshlng credibility of any BNL recommended ANS Nevada Meeting, June 1986 by Alhbaugh, Osb'ander, and . calc:ulatlcnl methodology.

Pearlman. The paper reports resulll of meaaurements of lhe Excepllona tD pelltlonet'a dalms muat be considered.

110fed eoorgy In aa""'8s of graphite from llNo locatlons (Lr., 19.2 Pearlman'a meltod of calcuating graphite skited energy needs ID be examined calfgm nea, Ile cenlllr of lhe graphile Island and 33.2 callgm at In detai In developing a l'9COl'Ml8llded cak:ualional model.

the edge of lhe cenlet fuel box of the UCLA Argonaut Reactor). All argi.wnents preeented are worthy of conslderalion In establllhfng NRC alalf The 332 waa changed IO 33.3 oaL'gm lO be consleter. wilh reconmendatlona. Flefer8nces 1, 3, 7, and 8- are worthy of further lnvedgallon.

rounding proceclurea.

Conmenls on the reeubl

• tollows:

• Tola.I reactor operation of 21 Mwd

• Thermal l\ience

• 3 lC 1011nut

• Adlabaqtlo cordllona largely

• lheoredcal COl'1C4lpl. dVllcut IO achieve In pmcllce

• 8lored energy decntues wltl lncrNllng dlatance flom 00NI region, e.g., 15.61 ~ O 18"; 1.34 cat1sJn O 22", and U"lll8Utnble O 26"

• Nlhin graphlla Island, and energy deoreuea from 33.3 IO 19.2 caL'gm In going a dlllance of only about 3'", from bl!

box edge IOWard center of graphlla Island

• A zone aboLi a* Wide )ult outlldl Iha luel bolCea Is a lnifsd volume where adiabatic re1aaso of stored energy cowl I red In laq)erature lncteaaea of too- to 139" C

• resulll from cooler graphite temperaturn and a peak of ht fut llux

• Cilee report, "BEPO Wlgnet Energy Release,* Geneva 1958, which gives data on enel'Q'I release In aneal of reactor grapNta In place and In tie labolaloty, e.g., 80"C lete temperatu'e rise over 6 IY. period compared ID 1 tv. In laboratory I

I

• Graphhe Stored Energy Calcuationl-NRC ISpOlWOClld calculations (NUREG/CR 2079) uses h i g h l y ~

'I' and empirical model applied to

• hypolhellcal Argonaut reactor Page 1 of 3

PAM 50-44 INDIVIJ;)UAL CITIZENS Harry Peatlman, PhD ii 11/3/88 II 1! 1!

No X ABSTRACT

• Al sublequent calcuationa (lncludng CBG's) have been baled on iu same model The procedure depends only on the COMMENTS lhennal flux at

• given position

• lnaenlon of com,ct valuea tor tie approxlmut values lmproYea ht results

• CBG'a clalm or substantialion of ltlff mlnlnum vu of 39 cai/

gm by 1he 33.3 cal/gm meaatnment la In error by a tactor of 2

• Peadman.. calculaeed maxirum value at centef of the graphite '

lliland la 17.8 oaVgm. These ruua. llhow 1hal usefu n'11aN of n:ired energy can be made wflhout me-..ement at leua --,

at lhe "8J)hite llland cen1er

• Petitioner lgnonts the dltl'lculv of making meuurements of g.raphllit ltDred energy

• ~ 8fTOr In CBG-. atatemertrega,d lloredMetlW elleellvely decreasing graphlle Ignition lernperab.n and lalJure lo demon-strate a llnkage between alot'ed energy awJ graphl1e ftres Graphite Ficn

• Cilea CBG'a ldure demonnmea how graphlle wl bwn under research reactor conc:llticn

• Faults CBG'a citation of ht Wigner 81181'0'/ effect being a I

significant contributor 1o lhe Wlndsoale reaccor fire I

• Citea lempelal!Jre of Cl>>mobyl NMICIOr graphite of 700-c ~

' ellrilallng acx:umulation of alor8d energy

• Cili8a fracturi,g of s.,aptita a1rUCtUte (logl) and ea1abllshlng convective air !lows al Chemobyt Thue COldtioill pita the

- lnillal ~ temperature (700"0) led 10 a selHustalnlng ~

Ire. CBG haa nol demonsnled how tda relates_, arral researchreactota.

Condudes 1hat CBG does not Sl.bllantlala Ila dalm that massive graphilllt flrea are credible acddenls i'l US graphlte-bearin reao-Iota, npedaly the amal research reaclot n that CBG fails lo

' demonltratu that graphl1a sSored eneru, can contrliu1e 1o such llres. Feels that samplng graptite lo meU1.1111 stored energy coold have radiological and aa,fety ~ n c e s that must be

' carefuly ewlualed.

Page 2 of 3

I PRM I

50-44 I

• 1NOIVJDUAL CITIZENS JamNV.

Spickard, PhD ii: Ii 1!

10J30/86 No I

X 11 p,§ ABSTRACT

~ looa1lon of graphlie reactots on univeralty oampl.U8 In urban area make mishaps afrnlar to Chernobyl and Yllndscale COMMENTS No new lnformadon provided. Realale& petitions dams and COOC&111S wilhout providing any additional bull or sul:>ldanliallo.

evenlB equaly dangeroua. Preae rapo,1a on Chernobyl tnigedy do not 11CM 1hat mal'1'/ us cl'9ts host reactors wllh almllat dell9lS without contalrvnent boildi,ga.

Local Ire lghUng unlta unprepared 10 light graphlle Ires.

NRC hu undereallmated tie danger o1 graphbe llres.

New data shows measurement ol 33.2 caUgm for UCLA a opposed to 5 e&'9n Ntlmmad by NRC.

Stored energy can help cause graphite Ire. 0peralora lhould be required so measure Wgner energy preNnt In g,aphJle and develop an effecwe plan tor pnMlnllng dlaasler.

EmeiyNemtlhy 11nJ86 No X CilM the aaaertlona of lhe CBt3 peO!lon No new arguments or i1lonnallon presenled. staff reaponN to pellllon wlB addreu

• Chemobvl poved graphlla llrea . . not non-crec:Nble. al of the mled conc:ems.

• Thete - . dozens of graphite moderated research reactora.

• Experiments have shown NRC'a preclctlon ol Wigner energy haw been undereatlmaled.

I j

I I

Page3 013

PRM . .

50-44 CITIZEN'S GROUPS Nude1t* Freo ii 1113188 ii 1! ~s No X 11 ABSTRACT Cltea Chernobyl dlsulet and grapffie lire al Wlndleale Reactor COMMENTS No new lldllllcal lnfonnallon presentad. Resralel CBG intonnalion and position.

Berkeley as Indication that It Is lq,rudent for NRC lo malmul the posillon Concen-. . .: (1) role of graphite In Ile Chernobyl and W'lndacale events, (2) the Committee lhat a g,aptile In la. "non-credble" ewnt Urgea "

NRC ID order crec:iblitr of graphite bwnlng, and (3) the need for an evacuation plan '°' lhe h pntparallon or an evacuation plan at Berkeley Aneach Befkeley reactl:>r and l.lnlYen!llty of caJlfomia campta at Berkeley. Response ID Reactor. C8G pelltion llhould addreaa tlele concema. .

GE-S!ockholdet'a 11/3/88 No X Cites Inadequate Mitty nlgt.ationa al U8 reaearch reactors No new tlldvllcal i1lonnallon pmerned. ReslalN C8G Information and position.

AjlanceAgalnlst uaklg graphila. Supports PAM 50-44. NfllJ a1atas lhal Chemobvl Concema are: (1) meuwemem of alored.energy In reacl)r graphlle, (2) plans, NucleatPOWIM' demonatrates Iha great danger of graptill Im and the dltlcuty equlpment, and nlni,g In extinguishing a graphlle fire, (3) preparation and pub-In ex'1gulthlng them. Restates CBG pellklra p,opoeed rut.. llcation of evacuation plane for on-site pensoonel and for nearbif residents.

Urges ru1H IO safeguard publlo heallh and reaeaunt pubic. Response 10 CG8 petition lhoud addntu these coocems.

Maryland Nuclear 11/3186 No X llfgea adoption of new ,ues lo lmp'Ove safety of l8SN.ICh No new lachnlcal lnformalk>n p-esented. AealalH CBG petition iifonnalion and Safety Coai1lon ntaCIOnl using graphite. Cites Chernobyl accidenl .. demonllra" poaltJon. Concerra ..-.: (1) dlW'lget from graphla fires, (2) dHicuty In elCtingulsh-Ing how dangel'0U& s,aphile fires are and how dlfllc:tjt they me ID big g,aphil& ha, (3) lrellghlers lack of nkllnQ or experience In exlingulahilg put 0iJl Urges that PRU 50-44 be adopted and lmplemerud. graphft9 Ira, and (4) Ille need for evacuallon of laborulry WOfkers and olhers nearby. Response to CBG pelilion llholJd addrea lhele concerns.

Ohio Cltilena Fot 11/3/86 No X 'Suppor1s PAM 50-44. Coosiders NRC's policy of regardng No new or technical lnformallon presented. Restates CBG lnlormalion and pas!-

Responaible li,"aphife fires as non-crecli)le evenia u a llOl'H:Rldlble am lrra- tlon. ~ are fer NRC lo make a proper l'esp0f1l8 to the Chernobyl lessons

' Enetgy, Inc. tional proposition, consldemg graphite !Ire at ChemobiJI. NRC learned. Reapo,lse to CBG petition shotAd address these c:oooama.

lhould consider the reality of grapNte fire and evoke appro-prial8 response.

DOCKET JUMBER iETlTlON RULE PRM j;?-44 {ij)

{ ~I l="f 813 -4 1) .1 TH E PENNSYLVANI A ST AT E UNI ~tt'~~ ITY VICE PRESIDENT FOR RESEARCH AND DEAN O F THE GRADUATE SCHOOL 114 KERN GRADUATE BUILDING UNIVERSITY PARK, PA 16802 *s1 rEB ' o P\2 *.26 C. L. Hosler Telephone:

Vice President and Dean 814-865-2516 January 23, 1987 814-865-6331 Comment on Petition for Rulemaking PRM-50-44 Submitting by the Penn State Breazeale Reactor License No. R-2, Docket No.50-005 Secretary U.S. Nuclear Regulatory Commission Attn: Correspondence and Records Branch Washington, DC 20555

Dear Sir:

This letter submits comments from the Penn State Breazeale Reactor (PSBR) concerning Petition for Rulemaking PRM-50-44. In the referenced petition the Committee to Bridge the Gap contends that the stored energy and the potential consequences of a fire involving graphite in reactor cores and reflectors have been underestimated and that specific additional regulations should be promulgated. We have determined that these contentions do not have an adequate technical basis and that the requested changes to regulations are inappropriate to the PSBR and other similar reactors.

The first basis for this petition is that a graphite fire is a credible event as demonstrated at the Chernobyl plant. While the Chernobyl plant included a graphite moderated reactor, the petition does not link the presence of stored energy in graphite with the severity of the event. To the contrary, operating data suggest that the operating temperature was sufficiently high that the graphite was continually in a self-annealed state, precluding the buildup of significant stored energy. Technical reports of post-accident analysis show that the cause of the accident was a series of operating procedure violations and errors resulting in a prompt, super-critical condition. No technical literature is cited to suggest that stored energy in graphite was a contributing factor in the severity of the event.

The second basis for the petition is the difference between measured and calculated stored energy in the UCLA Argonaut-type research reactor. We have neither the measurement procedure and data nor the calculations upon which to base a comment. However, it should be noted that stored energy in and of itself does not constitute an impending problem. One must consider the temperature attained upon release of the energy, the presence of oxygen to sustain a flame, and the exposed surface area of combustible material. The PSBR does not have the combination of conditions required to make dispersion of fission products from graphite-related incidents a credible event (nor has the petitioner presented information to conclude that such is the case for any reactor type).

ref. * ,

FEB 13 1987 r,.-,,, .-r. .--. ..-

AN EQUAL OPPORTUNITY UNIVERSITY

u s

?cs Cop Ad.: I Speci I D

Secretary, U.S. Nuclear Regulatory Commission Page 2 January 23, 1987 The PSBR is a swimming pool reactor structure; that is, the core sits in a pool of water. Graphite is present in two forms, the top and bottom reflector section of each fuel element and special reflector elements used for low power experiments. In all cases the graphite is clad in stainless steel. The graphite pieces are specifically designed with an air-gap between the graphite and stainless steel cladding. The air-gap reduces heat transfer, keeping the graphite at an elevated temperature where it self-anneals. Incidents with graphite ignition are not credible in the PSBR because:

1. The high density graphite is in the form of a solid cylindrical slug, a geometry which is not conducive to flame propagation even when exposed to an open flame;

2. The graphite is encapsulated in metal which isolates it from an ignition source as well as the oxygen required to sustain a fire;

3. The encapsulated graphite resides below 16 feet of water which also isolates it from oxygen while providing a continuous quench capability should any fire occur; and

4. The reactor power is such that either no significant fast flux exists and therefore no significant stored energy accumulates or the flux and graphite temperature are sufficiently high for sustained self-annealing, and therefore, minimum stored energy accumulation.

A detailed analysis leading to the conclusion stated in item 4 was performed by F.C. Foushee of General Atomics. The report, dated November 27, 1965, is entitled, "The Consequences of Radiation Damage to the Graphite Reflector of a 1 MW TRI GA Mark II."

The referenced petition continues to propose three specific regulatory requirements, graphite fire response plans, reactor fire evacuation plans, and a graphite stored energy measurement program. With the lack of a justifiable technical basis for a credible graphite fire, as discussed in the previous paragraph, we believe the proposed regulations are without basis and therefore moot. While we do not believe that a credible scenario exists for a fire in the PSBR involving graphite and the release of fission products, our safety evaluation report does include an analysis of other design basis accidents.

We maintain an NRC-approved emergency plan designed to assure adequate emergency preparedness for the protection of the health and safety of the general public in accordance with 10 CFR Part 50, Appendix E requirements.

Procedures are also established to deal with credible fire scenarios.

Secretary, U.S. Nuclear Regulatory Commisison Page 3 January 23, 1987 In summary, we request that the referenced petition be denied because it lacks an adequate technical basis and is irrelevant to the PSBR and other similar reactors. We appreciate the opportunity to comment on this matter.

Yours very truly,

~~~ -

Charles L. Hosler Vice President for Research and Dean of the Graduate School CLH/pka

DOCKET NUMBER COLLEGE OF ENGINEERING fi>ETl f,ON RULE PRM (51 P~ .81.5,1u fl-4,f 0"

{e[,?I THE UNIVERSITY OF TEXAS AT AUSTIN Department of Mechttnica/Engineering*Nuclear Engineering Program*Austin, Texas 787 P\2. <:if }471-5136 February 3, 1987 *s1 f£B 10 Comment on Petition fQrr iv~~a,ki,~,.,rpi-50-44 Submitted by the Nucl~~Kittgd.he~.th:n'g 'Teaching Laboratory Bl\f\t-\'.

License No. R-92, Docket No. 50-192 Secretary U.S. Nuclear Regulatory Connnission Attn: Correspondence and Records Branch Washington, DC 20555

Dear Sir:

This letter submits comments from the Nuclear Engineering Teaching Laboratory concerning Petition for Rulemaking PRM-50-44. In the referenced petition the Committee to Bridge the Gap contends that the stored energy and the potential consequences of a fire involving graphite in reactor cores and reflectors have been underestimated and that specific additional regulations should be promul-gated. We have determined that the contentions do not have an adequate technical basis and that the requested changes to regulations are inappropriate to The University of Texas reactor and other similar reactors.

A review has been made of several of the documents on radiation effects in graphite as it pertains to TRIGA fuel (including General Atomics report Nov. 27, 1965, "The Consequences of Radiation Damage to the Graphite Reflector of a lMW TRIGA Mark II"; Review by W.A. Meyer, University of Missouri Dec. 10, 1986, "Stored Energy in Irradiated Graphite"; and the USSR State Committee on the Uti-lization of Atomic Energy, "The Accident at the Chernobyl Nuclear Power Plant and Its Consequences"). There are no indications from a review of the various materials that indicates an invalid analysis or conclusion on the degree of hazard associated with research reactor graphite in pool reactors of the type operated by The University of Texas at Austin.

It should be emphasized that, while there is no clear basis or estimate of risk for a significant hazard from the graphite energy, there would be specific hazards and greater risk from a program implemented to measure the stored energy.

The University of Texas concurs with the statements submitted by The Pennsylvania State University (letter dated Jan. 23, 1987).

Sincerely, Thomas L. Bauer, Ph.D.

Assistant Director NETL SOP //3664 TLB:dlw cc: D. Klein G. Fonken FEB 1 J 987 Acknowledged by card ********** *. *** ..-~..-.

- ?rrx'?? sr1i/

-z_, l ,tP JT/r NOIS

DOCKET NUMBER PETITION RULE PRM ,04 -44@

{ .j7 P£ .J/J4!)

201 East Hyde Street *a7 FEB -9 A11 :o~

Tucson Arizona 85704 January 30, 1987 Gff -

DOC t\[ i", . :- .,*,-.1 Secretary of the Commission U~/, ~JL' United States Nuclear Regulatory Commission 1717 H Street NW Washington DC Docket PRM-50-44 Re: Petition to require inspection of graphite used in research reactors

Dear sir:

I am responding to the notice published in 51 FR 31341 regarding a petition requesting the Commission to amend its regulations and require operators of reactors that use graphite as a moderator or reflector to:

1) prepare fire response plans for a graphite fire

2) measure the energy stored in the graphite

3) revise the safety analyses to consider the risks and consequences of a graphite fire.

I am the director of the University of Arizona Nuclear Reactor Laboratory, a position I have held for 14 years. I have 20 years experience teaching graduate and undergraduate reactor physics and reactor engineering, and I served as a staff member of the International Atomic Energy Agency for two years. I also serve as a member of the National Standards Committee, ANS-15, for Non-Power Reactor standards. The following is my personal evaluation of the proposed rule, and is based on my experience with research reactors, and TRIGA reactors in particular.

The Unive,.-sity of Arizona Mark I TRIGA contains standa,.-d TRIGA fuel elements with 3.4-inch-long g,.-aphite end pieces, and a standa,.-d Ma,.-k I

,.-eflecto,.- app,.-oximately 12 inches thick and 36 inches in inne,.- diamete,.-.

The reactor operates at a maximum steady-state power of 100 kilowatts, and in the pulse mode the peak power is 600 Mw corresponding to an energy release of 15 MJ. The reactor is in a 21-foot-deep pool, and the minimum water level over the core and reflector is required to be 14 feet under any conditions. This water provides shielding for the core, such that under any mode of operation, the level of radiation in the restricted area beside the reactor pool is practically indistinguishable from natural background, and is less than 10 percent of the permissible level of radiation in unrestricted areas by 10 CFR 20.105.

I consider that a University is a training ground for students who some day will have major responsibilities in the nuclear field, and I set a standard which I hope they will maintain throughout their professional careers. In particular, in my teaching, I stress that no work should be done which involves exposure to radiation if that work is of no benefit.

The fruits of this can be seen in the reports which I have sent to you Acknowledged by card ...* ff t ]JJ9S7

f U S. NUCLE E h

D.:>

Pos~mark D C<, JCS R e I /\dd 1 Co.-.t~ r

~peci,. t., ~. 1 .i

annually since I became Reactor Director: no measureable radiation exposure dose of any student, faculty member, or staff member above natural background due to the operation of the reactor.

The TRIGA reactor at the University of Arizona is a valuable part of the whole education program, and it is used for both teaching and research. Much of the operation for teaching is at low power, less than 1 kilowatt, so the total energy release since the facility first went into operation is less than 8 megawatt days. This information is also provided to you in my annual report.

Although the TRIGA graphite reflector and fuel element end pieces will contain stored energy, I can see no fire hazard whatsoever. The temperature increase of the graphite if all of its stored energy were released would be less than that due to a typical pulse, and would not be damaging to the reactor. The TRIGA is designed to produce energy, and is constructed so there is heat transfer to the water, whether the energy is produced by steady-state operation, a pulse, or "Wigner energy". Since the graphite is under water, it will not burn. The fuel and reflector are not removed from the water, and will never be removed unless it is required for inspection under the rule you are proposing in PRM-50-44. Removal of the fuel and reflector for such an inspection, which clearly is without benefit, would involve significant exposure of persons to radiation and would be a violation of the ALARA principle and of the principles under which I have worked and taught.

Finally, if the graphite reflector were removed from the pool and its aluminum cladding removed for inspection under the proposed rule, the fuel would not be with it. In such an operation, the fuel would have to be underwater in the storage racks in the reactor pool. Thus even i f there were a fire in reflector graphite it would not involve fuel, and thus would not involve a Chernobyl-like release, which certainly the concern here.

I believe the proposed rule is unwise, unnecessary, and would be a very negative policy. It makes me sad to see how much time and money has been wasted both by the Commisssion and by the research reactor community in dealing with this proposed rule. I hold nothing against the persons who proposed the rule. It is reasonable for them to ask whether the use of graphite in reactors is safe, and I hope they continue to ask questions about other possible risks in this and other aspects of our society. At the same time, it is ridiculous for me to write to tell you that the graphite will not burn underwater. However if I do not respond, I face the possiblity of this unnecesary rule being imposed upon me, and I would then be responsible for the radiation exposures which it would entail.

~

• George W. Nelson, PhD

l t

1 0 Public ServtceN DOtK[T((1

~JSNRC c:...._,,o1c111rado P.O. BoxlMO co o.nwr. 80201.(JN) 2420 W. 26th Avenue, Suite 100D, Dt!5yert;[8<H~r-~ :5ff)211 R.Q WILLIAMS. JR.

VICE PAESIDENT NUClEAA OPERATIONS January 23, 1987 OFFIL Fort St. Vrain IOCK 4 Unit No. 1 i

• N P-87037 U. S. Nuclear Regulatory Corrrnission ATTN: Document Control Desk Washington, D.C. 20555 Attn:* Correspondence and Records Branch Docket No . 50-267

REFERENCE:

Response to the Committee to Bridge the Gap Petition for Rulemaking of July 7, 1986

1) Federal Register, Vol. 51 No. 170, 9/3/86, Docket No. PRM-50-44, Page 31341

2) NRC Letter, Heitner to Williams, dated 9/16/86 (G-86502)

3) PSC Letter, Brey to Berkow, dated 12/4/86 (P-86641)

4) NRC Letter, Denton to Lamm, dated 5/29/86 (G-86287)

5) Federal Register, Vol . 51 No. 208, 10/28/86, Docket No . PRM-50-44, Page 39390 Gentlemen:

Publ ic Service Company of Colorado (PSC) considers that the additional graphite reactor regulatory requirements requested by the Comn ittee to Bridge the Gap in their July 7, 1986 petition for rulemaking (Ref. 1) are unnecessary and technically inappropriate for application to the Fort St. Vrain Nuclear Generating Station.

~knowt EB 6 1 7 by

f ICLEt- r , A rrv,v r"lMM SSIO DO'""'~ ~ION

P-87037 Page 2 January 23, 1987 Rulemaking Petition Summary The petition requests that the NRC amend its regulations to require operators of reactors that use graph ite as a moderator or reflector to:

1) prepare and submit for NRC approval fire response plans and evacuation plans for a graphite fire, and

2) measure the energy stored in their graphite and revise their safety analyses to consider the risks and consequences of a graphite fire in their facilities.

The basis for the first requested amendment is the petitioner's assertion "that the occurrence of a graphite fire at the Chernobyl plant in the Soviet Union demonstrates that graphite fires are credible events." The "NRC and reactor licensees have held that graphite fires are 'non-credible' events and as a result have failed

• to take measures to help mitigate or extinguish such fires, should they occur."

The basis for the second requested amendment is the petitioner's assertion "that the NRC's generic analysis for energy stored in research reactor graphite signi ficantly underestimates the actual amount of stored energy, and thus, underestimates the associated risk of graphite fire."

Credibility of Graphite Fires The petitioner cites the Chernobyl plant accident as evidence of the credibility of a graphite core fire, yet does not take into account the causative mechanism or the significant differences in reactor design between the Chernobyl plant and the Fort St. Vrain (FSV) plant.

The Chernobyl c~usative mechanism, described in the Soviet report presented at the IAEA Conference in August, 1986 (Ref. 2), was a power surge due to a large positive reactivity insertion, which was exacerbated by the reactor's positive void coefficient of reactivity.

The Chernobyl reactor was being intentionally operated at that time in a regime outside its design basi s l imits.

PSC has addressed the possibili ty of various reactor accidents resulting in graphite oxidation in both the original construction pennit and the operating license proceedings for FSV. These are currently described in Section 14 of the Updated FSV FSAR, where it is found that none of these postulated accidents lead to significant graphite oxidation. Both the NRC and PSC have concluded, in these

P-87037 Page 3 January 23, 1987 licensing proceedings, that an accident involving significant graphite oxidation at FSV is not credib le.

PSC recently detennined that fo ur simultaneous independent structural failures would have to occur to pennit the oxygen flow into the FSV Prestressed Concrete Reactor Vessel (PCRV) necessary to initiate and sustain significant oxidation of the FSV graphite. No credible mechanism was found whic h could realistically create the air ingress necessary to sustain such oxidation.

Nevertheless, PSC has identified a method that could tenninate such non-credible graphite oxidation and described it in Reference 3. In this paper, PSC also pointed out some of the important design differences between FSV and Chernobyl, which preclude the type of accident which occurred at Chernobyl. These include FSV's negative reactivity coefficients, inert helium coolant, and high PCRV design pressure capability. The Chernobyl event was caused. by specific design features of that plant for which there is no analogy in FSV .

• In their initial evaluation, the NRC staff has concurred with these conclusions (Reference 4).

FSV already has an emergency plan which provides for action to protect the health and safety of the public based on projected offsite doses. A special "graphite fire" evacuation plan is unnecessary. Additional special plans for this specific hypothetical accident would require devotion of resources for the development and maintenance of unnecessary procedures and documents.

The requirements requested in this petition would be duplicative of regulatory actions that the NRC has already appropriately considered and handled with respect to the FSV graphite reactor.

Stored (Wigner) Energy Storage of Wigner energy is not a significant operating concern at

• FSV simply because it becomes negligible at temperatures higher than 500 degrees F. The average FSV graphite temperature exceeds this value at power levels above approximately 7% of full power. Even if FSV were to operate at 5% power for a month the accumulated stored energy would be only 1.1 cal/g. The maximum possible ad i abatic temperature rise due to the release of this Wigner energy would be only 4 degrees C, which is insignificant considering the design temperatures of the core and PCRV.

P-8703 7 Page 4 Ja nuary 23, 1987 RECOMM ENDATION PSC reconmends that the NRC deny this Petition for Rulemaking. to the extent it may apply to the Fort St. Vrain plant. In particular, it has been shown that FSV has been designed such that neither a graph ite fire nor significant graph ite oxidation is credible, and that this plant operates at temperatures well above those at which Wigner energy can be stored in graphite.

PSC concludes that the requests made in this petition are without merit when applied to FSV, as the petitioner's allegations cannot validly be concluded from the evidence presented. The stated concerns have been detennined to represent no undue risk to the public health and safety based on FSV Safety Evaluations and actual plant operating experience. Implementation of these requests would place a burden on PSC's resources wi th no resulting benefit to public health and safety .

• Very truly yours,

/4La~

R. 0. Williams, Jr.

Vice President, Nuc lea r Operations ROW/RS:jmt CC: Mr. Samuel J. Chilk Secretary of the Conmission Regional Administrator, Region IV Attention: Mr. J.E. Gagliardo, Ch ief Reactor Projects Branch Mr. R.E. Farrell Seni or Resident Inspector Fo rt St. Vrain

DOCKt I NUMl:St.t( ,.

P-ETITIOl I f< ULE PRM fl.- 44 ~

J. DIAZ, DIRECTOR

(~Ir/! 3/34/\ ~

NILS W.G. VERN ETSON, REACTOR MANAGER NUCLEAR FACILITIES DIVISION  !/

NUCLEAR REACTOR BUILD ING UNIVERSl1Y OF FLORIDA oesf1fRrr*,

GAINESVILLE, FLORIDA 32611 PHONE (904 ) 392-1429 TELEX 56330

• a7 FEB -6 p 3 :Z'4hruary 2 J 1 98 7 Secretary U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Attention: Correspondence and Records Branch Re: Committee to Bridge the Gap; Petition for Rulemaking (Docket No. PRM-50-44)

Gentlemen:

With reference to the subject petition for rulemaking alleging that the stored Wigner Energy in graphite moderated Argonaut reactors represents a fire hazard, the comments in this letter will show that the petition is groundless and should be dismissed.

The University of Florida Training Reactor response to the Petition by the Committee to Bridge the Gap will address three specific points:

1. Calculation of stored graphite energy, with reference to the alleged error in NUREG/CR-2079 (Analysis of Credible Accidents for Argonaut Reactors);

2. Maximum temperatures achievable by release of Wigner Energy; and

3. Graphite combustability.

Calculation of Stored Graphite Energy

• There are three basic units of graphite exposure to fast neutrons< 1 >:

Hanford dose, Calder dose, and BEPO Equivalent Flux. BEPO Equivalent Flux is the British Experimental Pile Oscillator thermal flux that corresponds to a (BEPO) specific localized value for fast flux. A BEPO Equivalent Flux of 1.8 x 10 12 thermal N/cm 2-sec represents a fast flux (>1 mev) of 6.2 x 10 10 N/cm 2-sec. The ratio of fast neutron flux to thermal neutron flux for a BEPO Equiva-lent Flux is 1:29. The University of Florida Training Reactor (UFTR/ fast flux is about 1.3 x 10 11 n/cm 2-sec while the thermal flux is 1.1 x 10 12 , 2 > indica-ting a flux ratio of 1:8.5. The BEPO Equivalent Flux conversion factor for the UFTR is then fast to thermal modified by 29/8.5, or a factor of 3.41 to indi-cate the proper fast neutron exposure as a function of thermal neutron fluence.

NUREG/CR-2079, Analysis of Credible Accidents for Argonaut Reactors, as-sumes an operating history for the UCLA reactor of 12 MWD and a stored energy of 5 cal/g. The UCLA reactor records a 21 MWD operating history< 5 >_ The esti-mated 5 cal/g stored energy should 1en be scaled to 21/12 x 5 = 8.75 cal/g to correct for actual operating history 3 >. The estimated 5 cal/g also used an uncorrected conversion factor (BEPO Equivalent Flux), strongly cautioned against in Nuclear Graphite <1 ). Corrected to a BEPO Equi.valent Flux based on the UFTR fast and thermal fluxes, the 8. 75 cal/ g is scaled by 3.41 to 29.9 cal/g, in agreement with UCLA data at 33.5 cal/g.

EQUAL OPt'ORTUNITY/AFFIRMATIVE ACTION EMPLOYE knowte ged by carcf. -  ! ~

• TJ""JT""e * * * *-* . ~* .-..,......

13SlON.

DO" CH Orr ,c., ur Do s Pos!ri..irk Date -,..

- ---'--...L...::..._:_------I C()J:1 J r ed /

i Add I Co.

!':>pcci I

~R D1W1liu" , ;c,_!_iA:5;

"/J ' Z.....

-~- 5

- a+::

U.S. NRC February 2, 1987 Page Two Note that the petitioner suggests a minimum value of stored energy of 39 cal/g( 6 ) while measured values indicate a minimum stored energy of 19.2 in the center island graphite and a maximum value of 33.5 cal/g< 4 >_ It should also be noted that (theoretically and as measured) stored energy is distribu~d and follows a near exponential decrea~e with distance from the fuel boxes 4 >, fol-lowing the fast flux distribution< 2 >. This distribution is important because the maximum value of s tored energy has reduced physical significance for the temperature increase of a volume of graphite due to a hypothetical release of Wigner Energy when compared to the volume average stored energy. The maximum Wigner Energy is measured at 33.5 cal/g at a point: assuming a 1 cm cubed block of graphite in the maximum position of stored energy with a homogenous stored energy distribution over a plane parallel to the fuel boxes, the aver-age stored energy content of that cube calculated by a conservative linear average is 7% less (32.4 cal/g) than the maximum stored energy value (33.5 cal/g) based on relaxation length for fast neutrons of 14.5 cm. If the block is extended to 10 cm from the core, where the value should be 16.8 cal/g, a linear average (conservative approach) indicates the average energy release from that 1 cm x 1 cm x 10 cm volume should be 24.9 cal/g, 25% less than the maximum stored energy. The average value of stored energy for the center is-land graphite (again using a conservative linear average) is 35.5% less than the maximum value: on averaqe, less than 21.6 cal/g will be released in center island graphite in the 1 cm section of graphite that represents the volume of greatest energy s tored between the fuel box and the core centerline.

Although the original analysis was predicated on assumptions that only allow orders of magnitude accuracy, the values are appropriate for illustrat-ing that the Wigner Effect is not a relavent issue. The corrected calculations and measured values show that peak and average values of Wigner Energy storage are extremely low, and "this level of energy is insuffic:i,ept, if released, to heat the graphite by more than an insignificant amount."< 6 )

/

Maximum Tenperatures Achievable by Release of Wigner Energy An increase in graphite temperature from an external heat source is most nearly represented by annealing processes that remove stored Wigner Energy.

This section of the response to the petition uses data from annealing experi-ments involving average values of stored energy. The release of the maximum amount of stored energy occurs in graP,htte (most limiting conditions) at 560 MWD/At. A s tored-energy release curve< 1 is useful in determining an upper boundary of the final value for graphite temperatures subject to annealing.

The use of these curves indicates that for 150 cal/g of stored energy from operating to 400 MWD/At at 30°C, the maximum temperature of the graphite from a Wigner Energy release is about 360°C, a change of 160°C above the annealing temperature. The curve indicates that the maximum energy release that could occur at less than 1000 MWD/At could only bring the final graphite temperature to 450°C, a change of 260°C above the annealing temperature. Continued heating does not result in spontaneous temperature increases.

U.S. NRC February 2, 1987 Page Three The petitioner suggests 39 cal/g presently stored in UCLA graRhi te at 21 MWD of operation, with 113 cal/gin another 17 years of operation< 6 >. The stored-energy release curves do not extend to values less than 100 cal/g( 1 )

but the 100 cal/g curve suggests a temperature rise of less than 150°C.

It should be noted that even if the petitioner's values applied to all the graphite in the core uniformly, temperature increase from Wigner Effect would be less than 150°C above annealing temperature. In fact, the maximum stored ener?y value is measured to be less than the petitioners alleged mini-mum value< 4 , and any increase in temperature will occur as an average kinetic energy increase over a volume rather than a specific isolated point. Average energy storage values are more appropriate to determining the graphite tem-perature rise in a hypothetical energy release, again reducing any potential for effects of a hypothetical Wigner Energy release

• Although the petitioner grossly overestimates Wigner Energy storage values, the petitioner's values do not present a hazard because an adiabatic Wigner Energy release does not cause exessive graphite temperatures; and when viewed as an energy distribution, the magnitude of stored energy under con-s ideration loses all significance to safety.

Graphite Combustabili ty Graphite will only " burn," actually oxidize, under a very controlled set of conditions< 1 , 7 >_ With a strong external heat source and a good oxygen sup-ply, the surface of graphite will burn starting at about 650°c< 1 >. At the UFTR, there is no strong graphite heat source(Z); air flow across the graphite is restricted and essentially eliminated fo r the high stored-energy gra-phite(2) and the surfa9e to volume ratio of the graphite and the pile strongly discourages combustion(?)_ Recent experiments note that graphite heated in a furnace with forced air at 10 CFM will not flame, has internal temperature increase of 24°C, and only produces about 24 cal/sec(B)_ Thermite was fired inside a graphite crucible (3.7 kg) releasing 3000 Kcal of energy in 11 seconds. Peak temperature rose to 1426°C. When the burn was completed, the graphite did not continue to burn< 9 >. Strong evidence has demonstrated re-peatedly that graphite will not burn without a heat source and will not under-go selfsustained burning except under specifically controlled conditions.

The classic graphite fires, Windscale( 1 0), and Chernobyl, were not prin-cipally graphite fires. Oxidation of fuel caused a heat producing system that burned the graphite. If the heat sources were removed, the graphite would/did discontinue burning.

Any attempt to establish a scenario for a graphite fire at the UFTR is blocked by a lack of ignition source< 3 > and physical characteristics of the UFTR system such as the large volume to surface ratio for the graphite(?), low exposure area of gra~hite surfaces to free air spaces< 2 >, low air flow through reactor void spaces 2 > and low operating temperatures< 2 >.

U.S. NRC February 2, 1987 Page Four In summary, Wigner Energy storage effect will not store a significant amount of energy in low power reactor graphite such as in the University of Florida Training Reactor; the maximum temperatures achievable by such a re-lease are insignificant; and graphite fires are not achievable due to the de-sign of reactors such as the UFTR. Clearly Wigner Energy and graphite fires in Argonaut reactors are not issues of concern. Therefore, we request that the pe t ition be dismissed.

Thank you for your consideration.

!!;;?_ly,

• P.M.W~

Acti ngh:!:~t o ianager

~S-~~

w . ;3 .\J ~

William G. Vernetson Director of Nuclear Facilities PMW/WGV/ps Enclosures cc: Reactor Safety Review Subconunittee

REFERENCES

1. Nuclear Graphite, Nightingale, pages 231/225/337/326/416, (1962).

2. "Safety Analysis Report," University of Florida Training Reactor, N.J.

Diaz, W.G. Vernetson, page 4-23, ( 1981).

3. "Analysis of Credible Accidents for Argonaut Reactors," (NUREG/CR-2079, PNL-3691).

4. "A Selected Annotated View Graph Set: Graphite Stored Energy in the UCLA Research Reactor," C.E. Ashbaugh, N.C. Ostrander, H. Pearlman, (1986) .

• 5.

6.

Private Communication, C. Ashbaugh, H. Pearlman, N. Ostrander to W.G.

Vernetson, (1986).

10 CFR Part 50 (Docket No. PRM-50-44) Committee to Bridge the Gap: Peti-tion for Rulemaking.

7. "Parametric Study of Reaction of Air With Reactor Moderator Graphite,"

(Contract Report SA-06252, UNC Industries), A. Barsell, M. Richards, C.

Dahms, ( 1986).

8. "Test Report: ASTM E-136 Combustibility Testing of TSC Graphite," S.

Bogert, UNC Nuclear Indus tries, Inc., ( 1986) .

• 9. "Preliminary Test Report on TSX Graphite - Thermite Burn Demonstration,"

Revision 1 , S. Bogert, ( 1986).

10. "Graphite Task Force Technical Report on Initial Feasibility of Burning Graphite With Oxyacetylene Torches," S. Bogert, E. woodruf, ( 1986).

DOCKET NUMBER iETITION RULE PRM fl -44@

(51 F£ ..Pl} -4 ))

NUCLEAR REACTOR LABORATORY oocK[ TED

~*SHRC AN INTERDEPARTMENTAL CENTER OF MASSACHUSETTS INSTITUTE OF TECHNOLOGY

• s1 FEB - 6 O.K. HARLING 138 Albany Street Cambridge, Mass. 02139 L. CLARK, JR .

Director (617) 253- 4 202 ~t Director of Reactor Operations oc February 2, 1987 Mr. Samuel J. Chilk, Secretary U.S. Nuclear Regulatory Commission Attn: Correspondence and Records Branch Washington, D.C. 20555

Subject:

Comments on July 7, 1986 Petition of Committee to Bridge the Gap

Dear Mr. Chilk:

The Nuclear Reactor Laboratory of the Massachusetts Institute of Technology (MIT-NRL) submits herein its comments on the July 7, 1986 petition of the Committee to Bridge the Gap (CBG). The petition requests that the Nuclear Regulatory Commission amend its regulations to require certain actions by operators of reactors that use graphite as a moderator or reflector. Such amended regulations would apply to the MIT Research Reactor (MITR-II), because it has a graphite reflec-tor. The design of the MIT Reactor, however, precludes the possibil-ity that its graphite could be a hazard. In addition, the chemical stability of graphite is such that it cannot be a hazard in non-power reactors in general. Consequently the CBG petition should be denied .

• 1. Design Considerations Specific to the MITR-II The design of the MITR-II precludes any possibility of a graphite fire in the reflector, the principal reasons being the properties of graphite and the fact that there is no source of ignition in the graphite. The des ign is described in the Safety Analysis Report 1 , and it provides many barriers between the fuel and the graphite, i.e., aluminum fuel cladding, the aluminum core housing, the light water moderator, the aluminum core tank, the heavy water reflector and the aluminum reflector tank. The light and heavy water systems operate at temperatures below 55 °C and are not pressurized. These barriers separate the fuel and the graphite by a minimum of 17 inches. In addition, the graphite is blanketed by an inert gas in order to limit the production of argon-41.

j U. S. N lJCI.E,\R REG Ult', Qrn Do,:-; I w; 0 ~1, - GF HiL ~l, Or I HE. COi/

& S:::.h,IL Do u, 1ent Statis;1c. s Pn° e" , / -

(;d

,., r: l

• _) '- , , *LllJn

2. Continuous Annealing and Stored Energy Nightingale 2 provides a comprehensive discussion of stored energy in irradiated graphite, including the effects of irradiation temperature on the amount of stored energy that can be released upon heating the graphite to higher temperatures.

The data contained in Chapter 12 can be used to determine an upper bound on the energy stored in MITR-II graphite.

The maximum neutron fluxes occur in a graphite region that has been measured to be in the 150-158°C range for operation at a power level of 4.9 MW. Nightingale's Fig. 12.9 (attached) shows that the stored energy released on heating to 600°C is only 10-16 calories/gram after irradiation at close to MITR-II temperatures for an exposure to 193 megawatt days/adjacent ton (Mwd/At), much higher than will ever be attained in the MITR-II. Fig. 12.10 (attached) shows further that the effect, which is reported to tend toward saturation at high exposures, is only 47 calories/

gram for irradiation at 150°C. Even if present and all released, this would not lead to excessive temperatures.

For the MITR-I, which operated from 1958 to 1974, the maxi-mum graphite temperature was close to 300°C, and so stored energy values were negligible (see Fig. 12.10).

3. Qualitative Measurement of Stored Energy In order to confirm the validity of the 10-16 calories/gram upper bound estimated above, a qualitative measurement of the energy stored in MITR graphite has been made. A 1/2 inch dia-meter sample of the reflector graphite was core-drilled from the inner end of a stringer located beneath an irradiation thimble (3GV2) at a point about 30 inches below the core centerline. Al-

• though the flux is down by a factor of about three, the graphite temperature -is estimated to be 30-40QC cooler, resulting in less annealing.

The measurement was made by heating the irradiated sample (1/2 11D x 1/2"1) in an evacuated electric furnace. An unirradi-ated graphite sample was heated simultaneously as a control.

Both were equipped with thermocouples. The sample temperatures were raised from room ambient to an equilibrium value of 190QC over a two-hour period. The two temperatures tracked each other precisely, with no indication whatsoever for the temperature of the irradiated sample to rise more quickly due to the release of stored energy. After reaching equilibrium, the sample - tempera-tures were further increased at a rate of about 1 QC/minute to 500QC over the next five hours in order to determine if any energy might be released in that temperature range. Again there were no observable effects. Given that the specific heat of graphite is 0.2-0.4 calories/gram-QC in this tempe:f'ature range, a temperature rise of several degrees would be expected for each calorie per gram of stored energy released.

4. Combustion Tests Two tests were conducted to evaluate the combustibility of graphite. In the first, an oxy-acetylene torch was used to heat an unirradiated block of graphite approximately l" T x 2" L x 2" W. A spot on the block was heated to 800-900°C (bright red), the maximum achievable with the torch over a period of several minutes. No flame other than the torch resulted, and there was no flame at the flowing spot after the torch was removed. The same torch cut through a 1/4" thick piece of cold-rolled steel in seconds.

The purpose of the second test was to determine whether irradiation in a reactor would affect the combustibility of graphite. The two samples used for the experiment described in the previous section were mounted on a graphite block with their thermocouples and heated over several minutes to 1200°C, close to the upper limit for the thermocouples, using an oxy-acetylene torch. The samples were then held as close as possible (+/-S0°C) to 1200°C for five minutes. Although the samples were bright yellow, no evidence of flaming was observed. When the torch was removed, both samples immediately cooled. Just at the end of the five-minute period, the thermocouple for the irradiated sample burned through, ending the experiment with that sample.

The unirradiated sample was allowed to cool momentarily and was then reheated to 1200°C, at which time the acetylene supply to the torch was valved off. The purpose was to determine if hot graphite would burn in an oxygen-enriched atmosphere. The sample immediately cooled without burning.

As was expected, oxidation of the graphite did occur, sub-stantially reducing the samples in size, but at no time was there evidence of burning, and there were no observable differences between the two samples.

In order to see if burning could be initiated at even higher temperatures, the thermocouple was removed from the unirradiated sample and was replaced by a 1/8 inch diameter carbon steel rod as a support for the sample. The oxy-acetylene torch was again used to beat the graphite, this time to probably 1400-1500°C, at which temperature the steel softened and could no longer support the sample. Again there was no evidence of burning.

General Considerations

1. Graphite Combustibility The resistance of graphite to burning, even at high tempera-tures, is a well known phenomenon. For example, a one-half inch

1. Graphite Combustibility (cont.)

diameter all-carbon arc lamp electrode operates in the open air with the tip at about 4000°K, and a consumption rate of no more than three inches per hour - half of which is the result of direct volatilization from the arcing tip, and half due to oxida-tion. When the arc current is interrupted, the electrode prompt-ly cools, showing that ignition, or self-sustained combustion, does not occur, even after heating to this very high tempera-4 ture . Many other examples of graphite uses in high temperature applications exist, e.g. a graphite crucible successfully con-tained a 2100 °C thermit r~action, whereas a steel container melted through in 12 seconds .

In addition to the experiments described above that have been conducted at MIT to demonstrate the resistance o~ iraphite to burning, many other experiments have been performed - which support the MIT results. Only under special circumstances in-volving high temperatures (700°C and above) and high air flow has it been possible to demonstrate burning, and such burning should not be considered self-sustaining due to the need to maintain high, forced air, flow rates (reference 2, p. 416, and reference 9). Such conditions are not applicable to non-power reactors.

2. Chernobyl Comparison The Federal Register notice (September 3, 1986, p. 31341) regarding the petition states that CBG asserts that the occur-rence of a graphite fire at the Chernobyl plant in the Soviet Union demonstrates that graphite fires are credible events. The CBG assertion itself loses credibility where it attempts to com-pare the Chernobyl RBMK plants with U.S. non-power reactors (NPR) licensed by NRC. There is no resemblance in design, size~ Qpera-ting temperatures or pressures, or nuc l ear b e h av1or

. 1 u-11 and nothing to support the contention that graphite fires in NPR' s are credible events. In fact, the many attempts to induce com-bustion indicate just the opposite.

It is also not at all clear that a graphite fire occurred in the Chernobyl plant. If it did, it was probably hours after the accident when decay heat caused the temperature to rise. How-ever, what was reported to be graphite burning may well have been fuel. Also, high graphite temperatures may have given the appearance of its burning, whereas it may have been only glowing due to its temperature without self-sustaining combustion. In any event, increases in the graphite moderator temperature were a result of other factors and only secondarily, if at all, due to graphite combustion. In a parametric study of the reaction of air with N Reactor moderator graphite, it was concluded that, for realistic air flow rates, the rate of graphite temperature rise due to chemical reactions <\ioes not increase substantially over that due to decay heat alone .

Conclusions All of the information known about the succeptibility of graphite to self-sustained combustion and all of the experiments that have been conducted lead to the conclusion that graphite is not a fire hazard in non-power reactors. Consequently there is no need for a new regulation requiring reactors using graphite as a neutron moderator or reflector to formulate graphite fire re-sponse plans, as requested by CBG. Likewise, there is no need for a new regulation requiring a special evacuation plan for a reactor fire; evacuation plans already submitted and approved in accordance with 10 CFR 50. 54 (q) are adequate. The presence of graphite does not constitute an unreviewed hazard.

A new regulation requiring measurement of the Wigner effect is also not needed. Due to low exposures and/or continuous annealing, it probably does not constitute a problem in most, if not all, non-power reactors. As shown above, it is clearly not a hazard in the MITR-II.

It is, therefore, our strong belief the CBG petition should be denied.

Sincerely, L ~ &~-

Lincoln Clark, Jr. r LC/crh cc: MITRSC

References

1. MIT Reactor Staff, "Safety Analysis Report for the MIT Research Reactor (MITR-II)", MITNE-115 (October 1970), as amended.

2. Nuclear Graphite, Ed. R.E. Nightingale, Academic Press (1962).

3. Almasoumi, A.M., "Characterization of MITR-II Facility for Neu-tron Activation Analysis", S .M. Thesis, Department of Nuclear Engineering, Massachusetts Institute of Technology (September 1978).

4. The Industrial Graphite Engineering Handbook, Union Carbide Corp., p. SD.03.01 (1962).

5. Bogert, S. R. , "Pre 1 iminary Test Report on the TSX Graphite-Therm.it Burn Demonstration, Revision 1", UNC Nuclear Industries (July 1986).

6. Bogert, S.R., "Test Report on the TSX Graphite-Charcoal Burner Demonstration", UNC Nuclear Industries (July 1986).

7. Bogert, S.R. and Woodruff, E.M., "Technical Report on Initial Feasibility of Burning Graphite with Oxyacetylene Torches", UNC Nuclear Industries (April 28, 1986).

8. Bogert, S.R., "Test Report: ASTM E-136 Combustibility Testing of TSX Graphite", UNC Nuclear Industries (June 1986).

9. Barsell, A.W., Richards, M.B., and Dahms, C.F., "Parametric Study of the Reaction of Air with N Reactor Moderator Graphite", Report No. GA-Cl8609 (September 1986) .

• 10. Quapp, W.J., Bogert, S., Woodruff, G., Barsell, A. and Richards, M., "A Post-Chernobyl Review of Nuclear Graphite Oxidation and Combustion", presented at the Eighth Annual International HTGR Conference, San Diego (September 1986).

11. Research Trainin Test and Production Reactor Director, Second Edition, Editor R.R. Burn, American Nuclear Society 1983).

0.08 Ill "C 0.06 I 9"C 0

"' 0.04 i-:

~ 0.02, 0.00

-0.02 ' - - - - -

100 200 300 400 500 600 Fw. 12.9 Stored-energy releaae curve. of CSF graphlte'" expoeed to 193 Mwd/ At.

The higher inadiation temperaturee greatly reduced the 200°C peak Numbem under the curves give the stored energy releaaed to 600°C.

Q.215 u

0.20

• w

~0.1!1 1-

'D 0.10 43'48 llwd/At, 250 0.05 100 zoo 300 400 F10. 12.10 The effect of irradiation temperature on stored-energy relea.ee curves for high e.xJ)OllW"ea..,

• Numben1 under the curves give the etored eneray releued to 400°C.

2 From Nuclear Graphite, Ed. R.E. Nightingale

DO CKf!\iijearch Reactor Facility USNH1.,

UNIVERSITY OF MISSOURI

,r. ".:I. Research Park

'87 FEB -5 P\7cc)ll;mbia, Missouri 65211 Telephone (314) 882-4211 February 2, 1987 uff wOC Secretary

• U.S. Nuclear Regulatory Commission Washington, D. C. 20555 ATTENTION:

SUBJECT:

Correspondence and Records Branch Corrections to Comment Letter from University of Missouri with regard to Docket No. PRM-50-44 This comment submittal contains corrections to the letter submitted January 28, 1987. The need for corrections arose specifi-cally because of errors in the Figure numbering sequence. This docu-ment replaces the previous comment submittal in its entirety.

~cerely,

• Attachment xc: R. E. Carter COLUMBIA KANSAS CITY ROLLA ST. LOUIS an equal opportunity institution

U. S- NUCL EAR REGULATORY COM DOCK T NG & SEf<VlCE CR\NC,,

,ON Or, ,l,t vf H U 11 t: L, I ii

. ~

LVJl I ~I

TABLE OF CONTENTS SECTION NO. DESCRIPTION PAGE NO.

I. INTRODUCTION *** ........ .. ....... 1 CREDIBILITY OF GRAPHIRE FIRE IN RESEARCH REACTOR APPLICATIONS * * * * * * * * * * . . . . . .

II.

3 III. WIGNER ENERGY IN RESEARCH REACTOR GRAPHITE ** . . . 8 IV. RESPONSES TO OTHER CBG PETITION COMMENTS *** * *

• 13

v. APPLICATIONS OF GRAPHITE AT MURR *** * * * * ... 16 A. The MURR Graphite Reflector ....... .. 16 B. The MURR Thermal Column * * * * * * * * * *

• 17 VI. ANALYSIS OF STORED ENERGY IN THE MURR GRAPHITE REFLECTOR * * * * * * * * * *

• 23 A. Stored Energy Calculations **** . ... . . 23 B. Analysis of The Possible Hazard Associated With The Calculated Stored Energy * * * * *

• 25 C. Credibility of Graphite Ignition in The MURR Reflector Application. * * * * * * * *

• 26 VII. ANALYSIS OF STORED ENERGY IN THERMAL COLUMN GRAPHITE * * * * * * * * * * * * * * * * * * * * *

• 27 REFERENCES * * * * * * * * * ............ 28

• APPENDIX A APPENDIX B GRAPHITE REFLECTOR ELEMENTS REFERENCES **

THERMAL COLUMN A-1 A-20 B-1 APPENDIX C ANALYSIS OF THE HAZARDS ASSOCIATED WITH STORED ENERGY RELEASE IN THE MURR GRAPHITE REFLECTOR . . . C-1 REFERENCES * * * * * * * * * * * * * * * * * * * *

• C-6 APPENDIX D ANALYSIS OF HAZARD ASSOCIATED WITH STORED ENERGY RELEASE IN THERMAL COLUMN * * * * * .. . . . D-1 1

I. INTROOUCTION The petition by the Committee to Rridge The Gap (CGR) addresses two separate issues: (1) the credibility of a graphite fire at research reactors and (2) the potential contribution of the Wigner (stored) energy to the autoignition of research reactor graphite. The CBG cites the occurrence of burning graphite at Chernobyl as their basis to allege that research reactor applications of graphite are unreviewed safety questions.

To imply that the consequences of an accident involving research reactor graphite are analogous to the consequences of the Chernobyl accident ignore the many extreme differences in the design and operation between the

• Chernobyl RBMK reactor and even the highest power research reactor. These extreme differences include power level, core size, fission product inven-ory, operating temperature, reactor control systems, and inherent design characteristics.

The RBMK reactor is a 3200 MWt plant operated with graphite temperatures in excess of 700°C during normal operation. It can be characterized as having a large core with a positive core void coefficient, subject to local power instabilities. By comparison, the largest licensed research reactors

• are 20 MWt and have coolant operating temperatures ranging from 30°C to 75°C.

These research reactors all are designed to have inherent safety features including negative core temperature and void coefficients.

The CBG also infers that graphite fires were the initiating events in both the Chernobyl and Windscale accidents, when in fact, they were corollary events in both cases. The graphite fires may have exacerbated the accident consequences but were not the cause of either event.

The cause of the Chernobyl accident was determined to be a prompt criti-cal reactivity excursion with rapid fuel failure and a steam explosion which resulted in core disassembly and subsequent destruction of the reactor enclosure {building). 1 , 2 The fragmented graphite (1500 tons) was subjected to temperatures in excess of 2000°C (fuel temperature was estimated to be 3000°C) and burned. This fire was sustained due to the continued heat source of the 1TJelted fuel and an unrestricted supply of oxygen. With a normal operating graphite temperature of 700° - 750°C there was expected to be neg-ligible Wigner energy to have contributed to the graphite fire at Chernobyl.

The immediate cause of the accident at Windscale, on October 10, 1977 awas the application too soon and at too rapid a rate of a second nuclear heating to release the Wigner energy from the graphite, thus causing the failure of one or more uranium fuel cartridges, whose contents then oxidised

• slowly *** The exposed uranium smouldered throughout the course of the day and gradually led to the failure of other fuel cartridges and their combus-tion and to the combustion of graphite.a3

• Attention was given to the possibility that, irrespective of the second nuclear heating, a large local release of Wigner energy occurred in a pocket of graphite which had not been annealed for some time, and that as a result high local temperatures were caused, sufficient to result in the failure of a fuel cartridge, or a lithium-magnesium cartridge or even combusion of graph-

• ite. The Corrvnittee of Inquiry studied the thermocouple records, laboratory data on the failure of cartridges, and general information.

rejected explanations of this type.u 4 The C~111JT1ittee One 111.1st keep in mind when trying to evaluate the hazards associated with graphite oxidation and combustion, that the burning of graphite per se is not a radiological hazard, unless the burning graphite can elevate tem-peratures to the point that fuel cladding melts with subsequent release of fission products.

In both of the accidents cited by CBG, the melting of fuel initiated the burning of graphite and not vice versa as CBG infers.

Il. CREDIBILITY OF GRAPHITE FIRES JN RESEARCH REACTOR APPLICATIONS 8

Safety analyses for research and test reactors are based on the concept of a postulated Design Basis Event (DBE), an event for which the risk to the public health and safety 1s greater than from any event that can be mecha-nistically postulated.n 5 If a facility can meet the 10CFR20 and 100 require-ents for public health and safety for a DBE condition," then the capability of the facility to withstand normal and abnormal operational transients and a broad spectrum of postulated credible accidents without undue risk to the public would be defined within the DBE." 6

• There is no doubt that the burning of graphite is possible, if the temperature of the graphite is elevated past the ignition temperature and the theoretical volume of oxygen required for combustion is continuously applied.

The question to be answered, however, is whether graphite in research reactor applications can be made to burn.

In most research reactor graphite configurations, the conditions for substantial combustion are difficult to postulate even under accident condi-tions. In most applications even if ~he ignition temperature were exceeded,

• oxygen supplies would be insufficient to sustain combustion.

The credibility of graphite fire in large block geometry, oxygen restricted (or vitiated atfllOspheres in canned graphite) and water submerged applications at research reactors does not lend itself to a simple generic engineering answer. The applications and operating conditions in various research reactors are diverse, and the postulated initiating events are endless. However, one can argue that a graphite fire 1s a tertiary or secondary event at best, requiring some initiating event to be possible (a massive temper-ature excursion of at least 700 - 800°C magnitude). If an initiating event of this magnitude is required to initiate a graphite fire

• I and is credible, one would expect this initiating event to have been pre-viously analyzed and within the DBE envelope for each particular research reactor.

In other words, perhaps a fllOre important determination is whether or not there is a credible initiating event for graphite fire and whether this initiating event is previously analyzed and within the accident consequences of the Design Basis Event.

If there are no such credible initiating events, then graphite fire in a research reactor's particular configuration is of no practical significance.

There have been several studies undertaken to determine the oxidation

• and combustion characteristics of artificial graphite. The results are varied due to difference in experiment setup and conditions. The*results of three prominent studies will be presented here.

1) "If one defines a "threshold oxidation temperature* as that at which a sample loses 1% of its weight in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, pure graphite has a threshold oxidation temperature of 520°C to 560°C depending ori coke source and processing variables ....

In-the temperature range 200°C to 250°C above the threshold

• oxidation, the oxidation is penetrative in character, such that the weight loss is greater than the volume loss. At still higher tem-peratures, the oxidation is diffusion-controlled. with virtually all I

oxidation taking place at the outside surface ** *

• Similar oxidation effects on graphite are produced by steam and CO2 atmospheres. However, the threshold oxidation temperature is higher for these gases. Corrnnercial graphite will have a threshold oxidation temperature of about 700°C in steam and 900°C in C02.n 7

2) *rests were conducted (at Brookhaven National laboratory) to determine the exact conditions for self-sustained combustion. Sustained combustion was defined as that temperature which remained constant or increased without external heat input ***

The minimum temperature for self-sustained combustion of unirrad1-ated graphite was found between 830°C and 930°C. It was determined that a fire could be put out either by suffocation or by chilling the graphite The irradiated pieces of graphite (exposure of 5.6 x 10 20 nvt) were heated to different temperature levels between 400°C - 900°C. At each

• temperature. the flow rate was changed and evidence for self-sustained combustion carefully sought.

evidence.

Up to and including 700°C there was no However. at 790°C combustion took place and the graphite temperature 1ncreased. 118

3) S. C. Hawley in his aAnalysis of Credible Accidents For Argonaut Reactors* cites the results of a graphite burning study in Nightingale's Nuclear Graphite. Hawley derives his 650°C ignition temperature from a graph in Nuclear Graphite (see Figure 1). His is a very conservative

• interpretation of ignition temperature (and rightfully so for the generic study he was providing), however. further investigation of the graph in Figure 1 is warranted.

This information was taken from Dahl's MExperimental Evaluation of the Combustion Hazard to the Experimental Gas-Cooled Reactor {EGCR).u Review of this docu~ent can lead to interpretations of ignition tem-perature significantly different from Hawley.

The experiment was performed in a channel that simulates a fuel channel in the EGCR. ulnitial conditions {for the experiment depicted by Figure 1) were 607°C, 4 lb/hr air flow, inlet air temperature 150°C and

• t sufficient heat input to offset radial heat losses. The test proceeded for 70 minutes with constant air flow **** graphite temperature rose nearly linearly at approximately .6°C/minute during this period. The air flow was then.increased fifteen fold and an expontial temperature rise occurred. (The 650°C point coincides with this step air flow increase.)

The rate reached 20°C/m1nute. After a short time the air flow was stopped and the entire asseMbly began to cool ****

Apparently oxygen starvation was retarding oxidation at the flow rate of 4 lb/hr and when air flow was increased, heat generation was accelerated to a much higher degree than heat removal * * * *

• Combustion in many cases depleted the oxygen supply to such an extent that oxidation could not be supported in downstream sections. 09 It is not surprising that the graphite oxidation accelerated with the step change in air flow during Dahl's experiment. The forced air supply is the principal behind achieving high temperatures in forced draft furnaces that are used to burn coke for the melting or annealing of metals. The forced draft provides a fresh supply of oxygen to the fuel as well as purging combustion products that would inhibit continued combustion.* 10 , 11

• It 1s evident from Figure 1 that at both 850°C and at 900°C when the forced a1r flow at rates of 50 to 60 lbm/hr was reduced, the graphite assembly began to cool. Therefore, if graphite at 850°C - 900°C 1s not subjected to forced air flow. the self sustaining temperature rise at these temperatures can not be maintained. The applications of graphite at research reactors preclude

/

these magnitudes of air flow rate (as opposed to the large graphite piles for which these experiments were designed that used forced gas flow for cooling).

I

' l

-*-Ai, Flew 1000 --* c,.,._,. Tu11,.,.,... Ufl from llll&t 10 *0

- l'ewar ID tlealanl

• Flo. JU& An uample of e e l f ~ burning of Speer-Nuclea.r 2 graphite ~

air... Inlet. temperature ill J&o*C.

Figure 1 12 In summary, the fact that graphite will burn can not be refuted. How-ever, the question here centers on whether or not research reactor graph-

• ite fires are credible or even if there 1s a credible initiating event. A transient that could take graphite in large block geometry to 600°C above operating temperature does not appear to be credible. Transients of this magnitude are precluded in research reactor applications by the inherent safety design features (passive) of research reactors as well as engineered redundant safety systems (active). The Committee to Bridge The Gap contends that the release of Wigner (stored) energy might be such an initiating event or could contribute to such an initiating event. This contention will be addressed in the following section.

f III. WIGNER ENERGY IN RESEARCH REACTOR GRAPHITE The Corm1ittee to Bridge The Gap contends that the release of Wigner energy in research reactor graphite could lead to a graphite fire or contrib-ute to the ignition of such a fire. They provide no proof to substantiate this claim. Instead they focus their attention on a perceived error in the methodology used by S. C. Hawley in creating a generic estimate of stored energy 1n Argonaut Reactors. They use half of their petition to argue that their method of estimating stored energy is better than Hawley's fllethod. But regardless of whose estimate is more correct, they never show proof of the linkage between the release of stored energy and the autoignition of graphite

• The Committee to Bridge The Gap does not appear to understand the mecha-nism behind the release of stored energy or the difference between total stored energy and stored energy releasable as a function of annealing tempera-ture. aAs soon as stored energy in graphite was observed, it was recognized that measurement of the total stored energy alone would be inadequate for the prediction of possible temperature excursions in the graphite moderator. It is necessary to know the amount of energy released as a function of the annealing temperature.Al3

• The shape and intensity of the graphite energy release curves are the main concern from a reactor safety standpoint. The most important feature of the energy release curves is the 200°C peak. The obvious reason for concern in the 200°C peak region is that the release rates exceed the specific heat of the graphite, a condition where self-heating of the graphite can occur (see Figure 2). This intense peak is only valid at low temperatures (30°C-70°C). 14 At higher irradiation temperatures (even irradiation temperatures below 200°C),

the 200°C peak is substantially reduced 15 (see Figures 3 and 4). In fact, the energy release peaks for irradiation temperatures exceeding 120°C are less than the specific heat for the annealing temperatures and hence can not result

. (

in self-sustained heating of the of the graphite. Other investigations have made the same observation (see figures 5 through 7).

Several other considerations 100st be understood about the release of stored energy in graphite: (1) the energy releasable to 800°C is only a fraction of total stored energy (i.e. for exposures > 5000 Mwd/At. energy releasable is 40% of total stored energy; for exposures to 600 Mwd/At energy releasable u soi of total stored energy) 16; (2) the release of stored energy occurs over time and not instantaneously (20 to 60 minutes for adiabatic laboratory releases 17 and 3 to 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> in actual reactor anneals) 18 and; (3) the peak graphite temperature reached during graphite. annealing under

• reactor conditions (not adiabatic} is significantly less than the laboratory achieved peak temperatures under adiabatic conditions for a particular stored energy release (428°C peak local temperature predicted for the second BEPO release. the actual local peak temperature during anneal was 306°C). 19 The Committee to Bridge The Gap does not explore the fact that the release of Wigner energy itself requires an initiating event. It is observed consistently in both calorimetric and combustion experiments that until the annealing temperature exceeds the irradiation temperature by some definite

• amount. stored energy will not be released. 20 The threshold temperature increase required before any release of stored energy is encountered varies frOfTl 50°C 21 , 23 to 100°C 24 above the irradiation temperature.

CBG contends that the storage of energy in graphite is sufficient to raise the temperature of graphite several hundred degrees. 24 Even if this were true, the temperature excursion they postulate is far less than the ignition temperature of graphite or the melting temperature of fuel or fuel cladding.

• t u, .. osu,iu 1ti 1t.c11a, O.T - * * - 11700

- - 11000

- - '>>&40 0.6 too 1076

?60 f.... cu eso

....* 100

i. 0.4

• .... 0.1 O.t 0.1

• 0 IOO too Flo. 12.6 Stored-energy relea11e BOO Annealing Tem.,.ratura * *c curves""

Figure 22s of CSF graphite irradiated 11e&r IOO aocc.

• ....p

~Q04 J.,:

'Cl

~

o.os a.oz tll"'C 0.00

-a.oz 100 200 aoo 400 l500 fiOO Ar1Mallng Tamperature, '"C Flo. 12.D Stored-energy releue curves of CSF grapbitr' expoaed to 193 J.Iwd/At.

The higher irradiation temperatures greatly l't!duced the 200*C peak. Numben under the curves sive the ston:d energy releued to 600°C.

Figure 3 2 6

.i.

~-3*-----,-----,-----,----.::r----ir----r----r-----ii.

)187 Mwd/AI*

• ~o"'t

/

I M*OIAt = ~.a/Ate oto

)

u

• ....... . C:;h.C.' 1-,c.,

~T

\0.1$

QJO

<11348 ll*d/Ate 2~"C 7.3 ccl/9

• 100 zoo  :&00 4100 F1G. 12.10 The effect of irradiatiOI1 temperature on Btored-energy release cun*es for high exposures.*** Numbeni under the curves give the stored energy released to too*c.

Figure 4 27 Specific Heat Curve Superimposed on Drawing

• 'i

~

-b0*8t-------1r---;-------~----=---=--~---:...-------a

~ 0oH5 temperature 1*74 60 l

1*70 85 1"9

. $.)g a lD'°n 135 148 &a CC

~l I

" T I

-T I I 0 100 1.00 500 600 700 800 temperature 6 -c F,z11rt! J. EncrJY rdcase from irradiated graphite (no intermediate anneals)

Figure 52a t

Ol 0

f

-!:.. 01 0

O*

i-0

.:5.

i

.* Olt 0

0

• '11 .. ~LaATVll&l"C)

FJ:G. 6. Effector &empt'rat.urr on aha.pt' of rate of energy releue curve for eoru,t.e,nt dose of 4.4 x JOH nvt.

Figure 529

• 0..$-

to

• , 2 lll>Sl&O,A'tt0Hf, n.1090..,IIO"I;;

s.,clO.,, ad\

§"" 02;-

'\&,!.'

~r; 0 ,:.

I (I,'

0.:0 200 ~

nMP[UTUSIEt"c)

• --- ,oo---

i' IV. RESPONSES TO OTHER CBG PETITION COMMENTS The Committee to Bridge The Gap uses their petition as a forum to indict the reactor safety studies of all research reactors. The CBG attacks the study done for analyzing credible accidents in Argonaut reactors and extrap-olates what they perceive as a deficiency in this document to the safety studies done for all research reactors. They do not substantiate their argu-ment regarding the inadequacy of the Hawley document, ITlJch less the safety studies done by other research reactors (i.e., MURR design analysis considers both the stored energy content and the potential growth in reflector graph-ite). The CBG also makes blanket statements in their petition about the

• inadequacy of emergency plans for research reactors. Research reactor emergency plans were recently upgraded and standardized and meet the require-ments of the Nuclear Regulatory Corrmission (ANS 15.16, Regulatory Guide 2.6, Proposed Revision 1 to Regulatory Guide 2.6). The evaluation of the adequacy of these emergency plans lies with the Nuclear Regulatory Corrmission using as their basis facts and engineering analyses about each specific licensee. The CBG does not prove that the emergency plans for research reac-tors are inadequate, they merely state that they are, as if their "opinion" is fact.

The CBG suggests that the Nuclear Regulatory Convnission have licensees submit evacuation plans based on a 25% release to the environment of the equilibrium radioactive iodine inventory. CBG states in footnote 9, page 4 of their petition, that the 25% release fraction is derived from ANS 15.16 "Standard for Emergency Planning for Research Reactors" (November 29, 1981).

Review of this document as well as Regulatory Guide 2.6 (January 1979) and Proposed Revision 1 to Regulatory Guide 2.6 (March 1982) show no such inventory release fraction suggested.

This inventory fraction is utilized for power reactors (Regulatory Guide 1.4) and is based on complete core melts and loss of containment for power reactors (NUREG/CR-3011). The consequences of a research reactor fuel accident would not approach that of a complete core melt and loss of con-tainment at a power reactors. The calculated iodine release fractions for a partial core melt with partial loss of isolation at a power reactor are less than .08% (fraction of core inventory release= 8 x 10- 4 ). 32 CBG further suggests that licensees perform empirical measurements to determine the stored energy content of their graphite reflectors. They do not suggest what importance they place on locating the point of maximum

• stored energy or why they want to determine it to.::!:. 10% accuracy. Do they suggest dismantling an entire reflector element to determine this point of maximum energy storage? Do they recognize the radiation dose that workers

~ould have to absorb to take samples for making these measurements? (MURR recently removed a graphite reflector element for replacement. The dose rate for this element exceeded 350 R/hr@ 1 foot.)

The maximum local stored en~rgy is not as important as the average value of stored energy over a graphite block. The absorption of the stored

• energy released will occur first in the graphite and result in a temperature rise in the entire graphite assembly as the heat is conducted to other reactor materials (i.e., reactor structure and fuel).

"The kinetics of energy release in a block of graphite with nonuniform stored energy is difficult to determine. Presumably the real behavior lies between that of a block containing the average stored energy and the other extreme of supposing each element of the block to rise locally to the maximum temperature characteristic of the local stored energy **** The q

relaxation time for this nonuniform temperature to equalize itself over the cell is only about ten minutes * *

• therefore, apart fron possible short-lived local transients, the maximu~ temperature reached will correspond to an average value of stored energy.* 33 V. APPLICATIONS OF GRAPHITE AT r-t.lRR A. The MURR Graphite Reflector The graphite reflector comprises 2/3 of an annular right circular cylinder that surrounds the beryllium reflector and the reactor pressure vessel. The remaining 1/3 of the annular cylinder is comprised of sample irradiation positions with little or no graphite. The graphite reflector is composed of eight 30° wedges {Numbers 1, 2, 3, 4, 6, 7, 8, 9 on Figures 8 and 9). A vertical perspective of the graphite reflector loca-tion relative to the reactor core is presented in Figure 10 *.

The inside radius of the graphite in each wedge is 9.684 inches

• (24.60 cm) and the outside radius of each wedge is 18.262 (46.39 cm).

The graphite height is 34.89 inches {88.62 cm). The volume of each 30° wedge is 33,736 cm 3 (with the non-graphite volume of the beamport access ports subtracted).

The total volume of the graphite in the reflector is 8 x 33,736 cm 3 =

269,888 cm 3 (9.5 ft 3 ). Using an average density of artificial graphite as 1.6 g/cm 3 , the mass of each 30° wedge is 1.6 g/cm 3 x 33,736 cm 3 = ~3,978 g

(~ 119 lbs}. The total mass of the graphite in the reflector is 8 x 53.978 Kg= 431.8 Kg (~ 953 lbs)

• Seven of the 30° wedges have accumulated exposure history equivalent to total reactor operating history (original wedges). One wedge was re-placed in November 1986 by a new graphite wedge with sample irradiation space.

Each graphite wedge is canned in 1/8 inch aluminum, with a nominal Helium gas gap of .040 inches between the graphite and the canning. The canning-of the graphite raises the internal volume average temperature to greater than 150°C during reactor operation. The principle upon which the canning is based is presented in Nightingale's Nuclear Graphite. uA 100re uniform temperature can be achieved if the graphite is insulated from the coolant channels in regions of lower power by a semi-stagnant gas layer between the moderator and the coolant." 34 B. The MURR Thermal Column The thermal column consists of a 60 inch thick graphite pack contained within a water-jacketed aluminum casing (see Figure 11). The column has a lead gamma shield positioned between the reflector ring around the core and the inner end of the case. The protrusion of the thermal column into the pool is a portion of the thermal column casing. This section 1s 37 1/2 inches square and 12 1/4 inches deep. This section is then stepped into

• a box 50 inches square and 68 inches deep. The graphite in this section forms a 48 inch cube (48 inches long x 48 inches wide).

Total volume of graphite in the thermal column is 2.095 x 10 6 cm 3

(~ 74 ft 3 ). Total mass of graphite is 3,350 Kg (~ 7400 lbs).

Figure 12 is included to show perspective of thermal column location with respect to the reactor core

• Figure 8

~.

!I I

II I

I °'s..

a, I en

• e-.. .................. -

c::n LL.

1

---

*-**--****--**-*It---

\ / POOL WATEn IN

---

~ ...-CENTER TEST HOLE CTORINLET WJ BERYLLIUM REFLECTOR

- GRAPHITE REFLECTOR

.,.._ 312A2mm DIA.

"'0-° e

-ti)

N

....anE ,,.,.,,. REFLECTOR TANK

- POOL OUTLET 38.10mm Figure 10 t

I I N I I.

' THERMAL COWMN DOOR RECESS 1

1 I

II I

I

- II I

I Figure 11

SCHEMATIC DIAGRAM Of THE REACTOR AND BIOLOGICAL SHIELD A - 4 x 4 Inch RenDvable Graphite Stringers

B .. Steel Thermal Colu.nn Door C .. Graphite Pack D - Aluminum \iater Jaclcet E - Pressure Vessel and Fuel Asseubly F - Core Centerline G - Spent Fuel, Gamna Irradiation 0 Im l):==::::sJ Facility*

l! - Pneumatic Tubes F

- - - - - - - l"l""t""'t1'-m-----~-

a E

Figure 12

\ '

I t YI. ANALYSIS OF STORED ENERGY IN THE MOR~ GRAPHITE REFLECTOR A. Stored Energy Calculations The existence of damage gradients in observations of irradiated graphite is well documented. These gradients occur due to the existence of pre-existing defects in graphite before irradiation; the energy spectrum of the neutron flux in the irradiated graphite and the tempera-ture differentials within the graphite. 35 These damage gradients are difficult to predict since the flux and temperature during irradiation affect both the creation of the defects as well as the removal (annealing)

• of the effects.

The dominating factor controlling the distribution of stored energy across a graphite block 1s the local running temperature. 36 *tn the face of uncertainties due to variations in temperature because of fluctuations in power distribution (changes caused by control rod manipulation. fuel burn-up or loading fresh fuel). it is difficult to assign an exact temp-erature to a particular region of a graphite structure. Moreover. the effect of the temperature of the structure depends on such design features

• as the manner in which graphite pieces are supported. aligned and re-strained with respect to each other; on clearances; and on mechanical loads. All factors considered. the behavior of a moderator (reflector) structure 1s predicitable only within a broad range of values.* 37 To estimate stored energy accurately, three dimensional distributions 1n the graphite of the fast flux and of the temperature are needed. 38 Since this detailed information was not available. the stored energy was

'calculated by using average temperatures and average fast flux values derived from computer code and design analyses data (see Appendix A).

A methodology similar to that used by Pearlman to estimate the stored energy at the UCLA reactor was ut111zed. The Mwd/At conversion factor fr0f1'1 fast nvt was one suggested by the National Carbon Company 1n HURR's design analysis and is comparable to the Mwd/At-to-fast nvt conversions cited by Nightinga1e3 9 and Oavidson~ 0 in separate studies.

The methodology incorporates the stored energy equations and param-eters outlined in Nightingale's Nuclear Graphite (p. 329-331). Two methods were used to calculate the total stored energy, one using the Windscale curves and data (155°C, 205°C, 255°C curves) and the other using the Hanford data (30°C curve and the information from reference 8 and 10 from Figure 12.2 of Nuclear Graphite)

• The first method (using Windscale data) resulted in the highest cal-culated average stored energy (275 cal/gram). This value is more than twice the calculated value by the second method (133 cal/gram); which it-self is high compared to the empirical data for irradiation exposure at

/

greater than 150°C (the MURR graphite volume averaged temperature a 166°C).

The Windscale curves *were obtained from irradiations carried out inside hollow fuel elements in high-flux reactors (DIDO. PLUTO and DMTR}

• for which the intensity of the damaging flux was considerably greater than normal for a graphite-moderated reactor. It is possible that under such conditions the radiation effects produced at a given total dose are significantly greater.* 42 The overestimation of this technique is verified by the 150°C and 200°C points on Figure 12.2 (Nuclear Graphite) which were obtained from samples irradiated in the Hanford controlled-temperature facility. These points correspond to approximately a factor of 1/2 with respect to the Windscale curves. This factor would bring the two methods into closer agreement. Another reason why the Windscale curves and data seem suspect is that information derived from these curves

.\ .

extrapolates to a saturation level of stored energy at 30°C almost twice that of the expected 600 to 700 cal/gram. 4 3,44 Woods noted that *irradiation at 150°C effects an order of magnitude reduction in the changes of most physical properties corrtpared with those incurred by irradiations at 30°t.* 45 This order of magnitude change is not reflected in the W1ndscale data.

B. Analysis of The Pos~1ble Hazard Associated with The Calculated Stored Energy In spite of the probable overestimation of stored energy using Method 1 of Appendix A, the higher value was analyzed with respect to

• possible hazard consequences in Appendix C.

Since the graphite reflector elements are not physically located adjacent to the fuel elements. conservative heat transfer assumptions had to be made to determine the possible effects on the fuel from a total release of the graphite stored energy.

The results of this analysis show that even at unrealistically short release times, the heat rate associated with a total graphite energy re-lease 1s less than the core decay heat rate. The LOCA analysis for MURR

• considers the decay heat rate load as its heat source with no consequent damage to the fuel. A lower heat load would present even less of a concern with regard to the core.

It must be stressed also that the consequences of the LOCA are less than that of the MURR Design Basis Accident. This would place a graphite energy release (if it were credible) well within the Design Basis Event envelope discussed earlier.

Evidence presented 1n Section III supports the contention that no energy release peak in excess of the specific heat of graphite is possible with exposure temperatures in excess of 120°C.

Since the bulk of the MURR reflector operates at temperatures greater than 150°C, this means that regardless of the total stored energy in the graphite reflector, at no point during the release Of energy can the graphite achieve a self-sustained temperature excursion.

c. Credibility of Graphite Ifnition in The MURR Reflector App ication If the .evidence presented in Section III is accurate, self-sustained temperature excursions are not possible in the MURR reflector application.

Even if the maximum* stored energy could be converted to an adiabatic temperature increase of 700°C, it is unclear how oxidation, much less

• combustion, could occur in an enclosed environment with a designed absence of oxygen (canned graphite, helium filled). uA self-sustained ignition can occur only in those situations which are capable of support-ing self-sustained combustion. For example, if the ambient pressure or ambient oxident concentration is insufficient for sustained combustion, it will also be insufficient for ignition.H 45 The analysis above also neglects the fact that these reflectors are normally under 23 feet of water. The MURR has an emergency pool fill system with a verified flow rate great enough to maintain 3 feet of water above a completely severed 6 inch beamport. This source of water ensures that the reflector is covered with water even in such an accident con-dition.

VII. ANALYSIS OF STORED ENERGY lN THERMAL COLUMN GRAPHITE The graphite 1n the thermal column has a very low stored energy content (see Appendix D). Even the maximu~ local stored energy is inadequate to create a peak adiabatic temperature rise great enough to cause graphite ignition.

The graphite stack 1s completely enclosed in an aluminum water jacketed casing with entry gained through a greater than 12 ton steel and masonite thermal column door and radiograph facility. This graphite is not exposed to sufficient air volume to support a fire even if one could be initiated *

• There are no hazards associated with the calculated amounts of stored energy in the thermal column. Furthermore. there are no credible initiating events for a Wigner release or the burning of graphite in the ther~al column *

' I RE_FERENCES

1. s. Rippon. *thernobyl:The Soviet Report.* Nuclear News (Oct. 1986/volume 29/no. 13) p. 64.

2. Preliminary Statement of the Select Panel for Post-Chernobyl Safety Review.

August 1986, Scientists and Engineers for Secure Energy

3. *Accident at Windscale #1 Pile on October 10, 1957,* Presented to the

• Parliament by the Prime Minister by Command of Her Majesty, Nov. 1975, Cmnd 302 (NP-6539) pp. 5-7.

4. ibid, pp. 9, 10.

5. NUREG-0849, Standard Review Plan for the Review and Evaluation of Emergency Plans for Research and Test Reactors, p. 1.

ibid, p. 1*

6.

7. L. M. Currfe, V. C. Homister and H. G. MacPherson, *The Production and Properties of Graphite for Reactors," Proceedin s of the United Nations 1

International Conference on the Peaceful Uses o Atomic Enersx (1955),

Vol. 8, p. 470.

8. R. W. Powell, R. A. Meyer, R. G. Bourdeau, 11 Control of Radiation Effects in A Graphite Reactor Structure." Proceedings of the Second United Nations International Conference on the Peaceful Uses of Atomic Energy (1958).

Vol. 71 p. 293.

,. 9. R. E. Dahl, *Experimental Evaluation of the Combustion Hazard to The Experimental Gas-Cooled Reactor - Preliminary Burning Rig Experiments,*

(Nov. 1961), HW-67792, pp. 15-17.

10. F. D. Jones, P. B. Schubert, Engineering Encyclopedia, (New York:Industrial

• 11.

12.

Press Inc., 1963), pp. 507-527

• T. Baumeister, ed. Standard Handbook for Mechanical Engineers, (New York:

McGraw-Hill Book Co., 1978), p.9-25, 9-26.

R. E. Nightingale, ed. Nuclear Gr~phite, (New York:Academ1c Press, 1962),

p. 416. ,

13. Nightingale, Nuclear Graphite, p. 331.

14. T. s. Neubert and R. B. Lees, *stored Energy 1n Neutron-Bombarded Graphite, 11 Nuclear Science and Engineering (1957), Vol. 2, p. 761 (Figure 7).

15. _ Njghtfnga.le,~ Nuclear Graphite, p. 341.

16. ibid, p. 338 (Figure 12.7).

17. ibid, p. 336 (Figure 12.5).

REFERENCES (cont'd) 18.

19. ibid, pp. 261, 264.

20. J. F. Kircher and R. E. BoW1'1an, ed. Effects of Radiation on Materials and Components (New York:Reinhold Publishing Corp., 1964), p. 352.

21. A.H. Cottrell, J.C. Bell, G. B. Greenough, W. M. Lorrier and J. H. W.

Simmons, NTheory of Annealing Kinetics Applied to the Release of Stored Energy from Irradiated Graphite in Air-cooled Reactors,N Proceedings of the Second United Nations International Conference on the Peaceful Uses of Atomic Energy (1958), Vol. 7, p. 320.

22. E. Fast, F. O. Smith and J. o. Cord, "A Proposal for the Controlled Release of Stored Energy in the MTR Reflector Graphite," Report #100-16656, 1959.

23. W. K. Woods. L. P. Bupp and J. F. Fletcher, "Irradiation Damage to Artificial Graphite," Proceedin~s of the First International Conference on the Peaceful Uses of Atomic nergy (1955), Vol. 7, p. 466.

24. Petition for Rulemaking Submitted by The Convnittee to Bridge The Gap.

July 8, 1986, p. 6.

25. Nightingale, Nuclear Graphite, p. 337.

26. ibid, p. 340.

27. ibid, p. 341.

D. J. Littler, ed. Properties of Reactor Materials and the Effects of 28

• Radiation Damage (London:Butterworths, 1962), p. 2i2.

29. H. Bridge, B. T. Kelly and B. S. Gray, ustored Energy and Dimensional Changes in Reactor Graphite,* Proceedings of the Fifth Conference on Carbon (1962), Vol. 1, p. 295.

30. Cottrell, et al, p. 319.

31. C. Dalmasso and G. F. Nardelli, *The Wigner Release in Graphite-Moderated Reactors,* Ene~ia Nucleore, (English translation in USAEC Report AEC-tr-4545), y 1961, p. 26.

32. "Dose Projection Considerations for E~ergency Conditions at Nuclear Power Plants," (NUREG/CR-3011), pp. 2.5-2.9

33. Dickson, p. 264.

34. Nightingale, Nuclear Graphite, p. 487.

REFERENCES (cont'd)

35. Nightingale, Nuclear Graphite, p. 231.

36. Cottrell, et al, p. 231.

37. Nightingale, Nuclear Graphite, p. 486.

38. C. E. Ashbaugh, N. C. Ostrander and H. Pearl~an, *Graphite Stored Energy in the UCLA Research Reactor,* Transactions of the Anerican Nuclear Society (1986), Vol. 52, p. 372.

39. Nightingale, Nuclear Graphite, p. 286 (Figure 9.20).

40. Davidson, Woodruff, Yoshikawa, *High Temperature Radiation Incuded Con-traction 1n Graphite," Proceedings of the Fourth Conference on Carbon, New York, Perganon Press, 1960, p. 600.

41

• Nightingale, Nuclear Graphite, p. 341 (Figure 12.10).

• 42.

43.

44.

ibid, p. 330.

ibid, p. 329.

J. J. Newgard, "Simple Semi-empirical Model for Neutron Incuded Stored Energy in Graphite, Journal of Applied Physics, Vol. 30, pp. 1449-1451, 0

1959.

45. Woods, et al, p. 471.

46. Gordon P. McKinnon, Fire Protection Handbook, 15th ed., (Quincy:National Fire Protection Association), pp. 3-4

• APPENDIX A GRAPHITE REFLECTOR ELEMENTS I. TOTAL HOURS AT 10 MW (Equ1valency)

December 1986 - 119,039 total hours at full power 43,726 MWD total 119,039 hours4.513889e-4 days <br />0.0108 hours <br />6.448413e-5 weeks <br />1.48395e-5 months <br /> ~ ]

• 4,959.96 u 4,960 days full power operation 5(4960 - x) + lOx = 43,726 24,800 - 5x + lOx = 43,726

• 5x = 18,926 x

• 3,785.2 days at 10 MW 4,960 - x

• 1,174.8 days at 5 MW heck: (3,785.2 days)(lO MW)~ 37,852 MWD (1,174.8 days)(5 MW) = 4,874 MWD 43,726 MWD 1,174.8 days at 5 MW~ 587.4 days at 10 MW (equivalent MWD) 3,785.2 days at 10 MWD

+ 587.4 equivalent days at 10 MWD 4,372.6 days. at 10 MW x 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />s/day 104,942 II! 1.05 x 10 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> at 10 MW 1.05 x 10 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> x 3600 sec/hour~ 3.78 x 10 8 seconds at 10 ~

• A-1

,\  :.

• 11. GRAPHITE VOLUME AND HASS r c 46.39 cm (18.262 )n) 2

/ r -= 24.60 cm (9.684 1n) 1 h = 88.62 cm (34.89 1n) 2 2 2 A= w (r - r ) c 4,859.6 cm h 2 1 V =Ah= 430,661.5 cm3 (360° reflector)

V = 1/12 V 30° wedge I

I 3

\ -= 35,888 = 35,890 cm for solid wedge (i.e. no beamport holes Beamport Access Holes 1 = 21.8 cm (8.587")

Three holes+ 4* diameter r = 5.08 cm (2 in)

Q_) Three holes+ 6* diameter r 1

= 7.62 cm (3 in) 2 2 3 3 V4 11 .. "lfr L = 1,767.4 cm x 3

• 5,302.2 cm 1

2 3 3 V6 11 -= 1rr L = 3,976.6 cm x 3 = 11,929.9 cm 2

3 V z 17,232.1 cm (all six beamport holes) tot This beamport access volume distributed among 8 graphite wedges (30° each) 3 V = 17,232.1 .. 2 154 cm b.p./wedge 8 '

3 V = 35,890 - 2,154 = 33,736 cm 30°wedge A-2

3 Density art1f1cfa1 graphite R 1.6 g/cm 3 . 3 M = (33,736 cm )(1.6 g/cm) = 53,977.6 grams= 119.2 lbs.

wedge (1 lb.* 453 grams)

M c 53,977.6 grams x 8 = 431,820.8 grams

• 953.2 lbs. total total A-3

111. TEMPERATURE PROFILES IN LARGE G~APHITE ELEMENTS Temperature distribution information was taken from Design Data I, *oes1gn Analysis of the Graphite Reflector Elements*. The result for graphite thermal conductivity of 1/8 the unirradiated value is used. These temperature distri-butions were generated by computer code HEATING. which performs a numerical solution of the general steady state and/or transient three-dimensional heat conduct~on equations. These calculations were made neglecting axial con-duction. *rhe axial variation of the heat generation was assu~ed to follow the nvt curves of Figure 4.3. The actual axial heat generation distribution

• 1n the graphite would be expected to be somewhat flatter than this, hence this assumption is conservative.*l Assuming the ends of the graphite to be at pool water temperature (50°C)

\

1s also conservative, because the graphite canning isolates the graphite from direct pool contact and a helium gap insulates the canning from the graphite to reduce axial conductance

• A-4

' l TEMPERATURE ZONES USED IN ANALYSIS Volume I Ave. Temperature

[15.6] Zone I {top) a3.s*c

[15.6] Zone 11 (bottom) 83.s*c

[37.9) Zone III (central front) 2so c 0

[30.9] Zone IV (central back) 1so c0 graphite log height= 34.89* (88.62 cm) 2 A -= 405 cm wedge cross-section

• I I

,/

/_

.... ..... \

A 1:,f:J A

III s:

s

.5506 A

.4494 A cross-section

/

/ IV cross-section

/

Volume S Zone I+ II= 2x5.445 s 31.2%

34.89 tlOT 'TO Sc.ALE.

Volume I Zone Ill+ IV= 68.81

• Volume I Zone III= 68.8 (.5506) = 37.91 Volume I Zone IV ~ 68.8 (.4494) = 30.91 A-5

'-r, '5 I Cc. t,r,r\ ~u-J,)

- I .,-is v,,.

• ----1 Cent.er line ff=E~ftti=i~JJ-: 4 5

- 6

• 0.8S8'

'( = 2** ~(.I, r-

--h ........ ~e.J.1f of 'tlt,,.fl.,...:,;:.

--10 9

=-----

~311 Large Reflector ElemE:nc Point:.* of Contact Between Between Gr*phite and Can EB 1/16 inch spacer 1 /8 inch spacer ~P<.C'-,....-rv-'-'E p~t::,1,.1; C>Y '1ltf..

cP.oS..S-<,:Ec..'Tio.lA'- " ' ' ~ ti,{' 1'"i,(<:T'l(.1'\~

Y/~ ~ Y5 c~-ri.~1.-,..lt... or &e..(\E'~,-n.

0 L. &

(.1:>t-St*--K.-D f ~ f\G,,>,.~"" 3,'!> ,wJ.:> ?> *.C of. Pt-'-'"rl ~ ,....ro~-t... 'l:. , Q_~S*"'-N Plt491\.'(,.,t ;;~ -mt. c;f.41,fl-\'ln. _t..'i-_!!:;."f -:.n::!:;

E\.~<cA!!-) ( po+{- t (\-1,

• r.-1)

A-6

-: !':'1 320 300 *:-,

.*. I 280

• u 260

~~~

?

240

• ==

200 A

• • l.

• • .1:,

180 ** l-

-:::-y.::l 1601:

140 I ~-r ** .: _: ~,=i.;.:.;...i~=

. -t

- ._. ! -:T~:

---

~ ---1 *

~

~

• 1 2. 3 4 .5 6 8-circumferen~ial Inde~. 3 9 11 13 f'igure 3.3 CJrcumferential Graphite Tempera~ure Profile~ for Large Elements II -,

340 300 u 260

"'w II.I 220 (t

11,1 f-4 ID

~"'

.c

0. 180

...ufl!

140 100 7 9 . 1 11 1 13 17 19 2S '

• Radial Index

• I figure 3.5

~*Afal ~-**~'** * - * * - . . _ ___ *--***--

~u NS 14

.....u '

• C 0

"'::I II C

12 . ' ~

I ,J; 0

N - 'J

• "i\1~

I c:_

....M 0

/

/

10 A

II1~

~ llod1 Bal~ tn II

c .. A~,_

3'2

.... ~

N GO

'AO/

V *~ ,.,. '2'fO

.\. 8 V 'I'

\

"'~

~

J

- ~

JII.

V

'2J.o/

V r\.. '24,o

t3 0

Cl I

'i\ ,* 0\

C l.39 I ~ 'Z :JO

!/

0 6

z

.,J

I II 'Z4'D V illefe ence Rods Full Cast Out '\ '2-t,-. \ <= \,)

..J I / ~

'O II 1,0 I

...* 4

/,v ~

CIO II

...."'C

• 2 llo l'°IG ~ ~~

I/ , /

- ~

/

' '\'

P\l~ol\.

~

p

!o / ~~

~~ V I" ~~.....

0

~20

-- 0 ...i+l6 +l. lZ *16 v

-=-

-20 -2 4

{/)

• D

{A Inchu Above Core Centerline i* Axial Yari&C _ _ _ _ _ _ _ .......

Plux for 7200 Hwd

1* s::

14 0

2

.J<j G,I C:

I 12 * . ~j 0

N I d J)

....,c 0

V

"' \

a:

u 10

'~\,)*V

/

,ti .J1 .,,

Bod* ll&l ta

'5 "5

N G0

{.

l/' /* . .~

-~t' 3

.\

8 11~:,

V

, ... 110*

j

" \ \

~

I r t...

- ~

s..

\Z./

/ V I\. l'lo

~

0 i::

...... 6 I1.) I 11,,IIQ

• ! ~

V /

~ '

a

~

1"1>

Refe euce CaH Rods Full Out \ l1re> \ ""-'

~

.*... I ,/

\q p\

4 Oio I-!' \ g CICI tj "i:

£.:> j.

r 0,0

/  !\~O \

~

V

• 2 /0 I/ .,./ \

l!>O / / ~~

• 0

~ J~ V I I I I"- ~~

r --' I

+: *Jtl. 4

-~ ~

i

~\A r-o'::r' Incbu Above Core Centerline \fl

~ _j 1~ rt .:~-:,;

l~F--:r- Figure 4.3 * ~ . ~ C.

.... '¥ IJ j ~

Axu.l Vari&

f Iute&rated Neutron Flux 00 HIid " ~ *'-,.J

IV. FAST FLUX DISTRIBUTION (>.821 Mev)

The flux distribution within the graphite was derived from EXTERMINATOR II code (pages A-12 through A-14).

Average flux values used for analysis:

11 2 Zone I = 8.16 x 10 neutrons/cm -sec 12 2 Zone II E 1.31 x 10 neutrons/cm -sec 12 2 Zone II I = 5.06 x 10 neutrons/cm -sec 12 2

• Zone IV = 1.09 x 10 neutrons/cm -sec Integrated Fast Flux:

11 2 8 Zone I = (8.16 x 10 n/cm -sec)(3.78 x 10 sec) 20

= 3.08 X 10 nvt 12 2 8 Zone II = (1.31 x 10 n/cm -sec)(3.78 x 10 sec) 20

= 4.95 X 10 nvt 12 2 8 Zone III = (5.06 x 10 n/cm -sec)(3.78 x 10 sec) 21

= 1.91 X 10 nvt 12 2 8 Zone IV = (1.09 x 10 n/cm -sec)(3.78 x 10 sec) 20

= 4.12 X 10 nvt A-11

~ ~ ~ ... .c J-;- ~ ~ ~ :r: .I:

--*-----*- --*- -*~. -~r~~---

-**--

• I n

In l',t or:

• .u 21 t

or 09t*9* 12 u

• 0C111*n or z-1 on*o* ,1 r an*t , --

• n1a 1 SNnu*., t ,f ,1dt t,fOlflN1Nla z* Zlt*o H~A**l---

oz9*t ,

Yl '1JG I

.... .... . ......... o*o oogz*o oosL*o nnru, .1:n1nas.::.N1Jtnrr-

• ,,v, 3n"1V~NJ~li NY st CHY

• l~.L.1NOi, Z*H SYM JS¥, IIMl SdNOl ti S,O) OQ s*o~ Lt

'5Nr.U Y) I .:I a,Jd4'

-~--------r,- *o* ..5 **is*-T:-o o o ,--,,--11 -,-ll" 0 D o , , , ,*-r-an, Nf>U.drJ

....... -:It .;,9-..;~

.>' .,,,.,.* . NUMA f*l&Gfl' C*R* a* .......... :; ""~*-----; *r.;- . *-* *: - ..... $ **

---

fltOUP o.o a

IU fLU* { > 'n ,,...,,, )

u ii;---,, I

• 1 H

u--*-,o I o.o o.o

• o.o o.o

• a,o o.o o.a o,o o,o a hOVIEtll 'il.967E. IO 9a7HEtlO 1,034.l!tlO' 4aUll IO .1,lttO~tlP lit7J~f~'q (,f*~E** 0 3,UlfE+ot 0,01**1*,ot.~'/~-hw-tJ.:

~

4

... ,sal!+u l*l,Ul! l2 4e.:1921!tll I .0321! l2

• • ia1&!tll .:s.2 Ht~ a z.auuu, * **4HE*u a.19,£ -ia-, ;*i141 ia. *1;1:i1E'+'i1 .. ~-.-:. 7i9El

.***~,**o: *,.0,11,10 a.11af!+10. o,t i.:.-;-;,-,.,i.li \. r.ar.r.rz;U'9l lO-o.*----------

$ 2e097E 12 2e3*0E+U 2 .2dOE l2 le5J7E l2 ~e929f l1 6*1071!+&1 Je*ZOE II le 44E*ll 3e48VE IO OeO

• , 3e7841 &2

,.o.iauu

,h209E U 3el10E+IZ l*009E+&2 lel61£ 12 7*652E+ll 49a39Etll 1,771&,II **IIU+IO OtO

• • IIIE U 4*0726 11 ~*1431 12 1,1671+11 913471+1& 0,1291 11 J,113J ll ltOllltlO 0*0 la I Ui!t 12 **-r;11'1iitil!'+Ta* -1*;-d *,u~*Ti:_.6"';"0-,sl! I 0

• 6o*67E U &**usu,2 , . I 291i!t 12 c-i;-s,,im-1r;tue*.-.-cr<<o-:-o-- -*----*--*-

9 ,.927£02 6e44illt! l2 ~.24Jct I 2 l*75~E li 2*2*6ctl2 lallZE+IZ 70095!+11 21¥20£*11 ***~*ftlO o.o u 9eUlk!tl2 7eO'.lll: U 7,l~*Et12 *** oo~tl2 216121+12 1,511!+11 11,1101+11 l,3tll II , *** 1£ 10 o.o SI ae09tal!. lJ det.L!Ol! U &,6tottl2 beOCt~ II z.Si9itl2 le724E 12 9,2l3~+ll 3e77~1 11 lel57E+IO OeO

&2 la.itSIEtl:.I leOUll! U 'w, '7115£t ll a. 115 iE+a ll!. ;;;*3*72i";az-- i~;wuli-1a*-*,-;c,:,*3*E*n*a i.21!tlfi'ii-9";a7ou10. o.-o-ll hU21HU hl47E 13 1 .102E+&J

• 0*4Jde+&2 3a7Sol+12 2el*dE+l2 1*14JE 12 4e651E II le08H+II OeO

"'11 l .t.161i. 13 hZ171?t-U 1,Z2IH IJ ,.a&IIEtlZ .,,3,*** , a.Jstc*** ,~21a***a ,,0901 11 *- l*lN£+11 o.o t,'l.14~ U

---,. I aUdl!tlJ I olllllSl!tU 1.uu 1:a lalllllt+U leUll!,tU

&*3471:\tU a.oza+u lel6ltt1J 7,77et£t12 "d,'J~&!t rz de~4l~+la 4,,4"Ql+II 11*H71+U

,: *~~*oe***a***a-;,:'11*1**~*-

&tlllEtll a,~14ttll h,U71£til l*ll&! II ***3611 11

&.2961+&1

• ~**s9'Eif21r;91ift+i1-c*..e,l! u* *

• o ; o - - - - - - - * -

l150ijftll OeO 0*0 11 JoOJb~tlJ

):,o I

t.,J,._,

II /hUOUIJ leit:l?E+ll I* PIOhlJ le1hEt13 l eOlJt. I .I hUll! II I ,722! 1.J 9* *UhUi 9,7171+11 9,9Vl! &°2 l'4UIC+La , :l,'~l*U*~I i ,U.llt II

,,.ao1 12 ,.11:su_, ,,1011 11 1,1.*e+1.,

lt1Hlt~ l l Ii'~ ;.*tirft"il'- -,1;*,~e.,z-T:1;01!-U*- *7;6':11lr+n--,.,,.-,'lfl-**ij;o----

I pt.(!OOU a.**~~..

0* 0 o.o H z;JO*UU I el&OEtlJ 21 ZelUt!tll hdl7E ll , le74d£+l3 la007t*IJ &1M07Ul2 3eZ981!:tU I a110!t12 7 .. 755£ 11 '. ,e.ve:. 11 o.o 12 2.u.wu.1 ltA*JUll 1.,,i~**.1 a.oort+&J t,7idE &2 3,2ue:u2 a,,e.71!+12 7*75H Ji le771f!tll o.o u 2a3JletU a.UD ll t.7J4Etl3 9 9421:+12 5e7lcU!tl2 :s.uxu~ lt7U£ U 1.,,.,~**' l1fHE ll o,o o~o--*-----*- - - - -

z* 2a275e+IJ a.779£+13 l,o~litll 9e6GJt l2 H 2eUtaEtlJ I a 709!t13

--~*-- **-*-*-.J.-..... '.. . *-*--**--..-----**-** -- -*-*** _,.. - .--*-*

NUffA f*I &Ofl' GAQUP ,Lu,t

--- ai"" - !12 59---* - 6 0 -

H 2*0Hlt13 le61.llt13 I ,!UJ£ 13 lt760l!t&.I 1,02o!tU 2t8"11Ul2 1,Ullt12 6e1lfftll 1 e5:H! II Oe.O a, I ,9Ul!tt.1 lt4Vll!UJ ****n-i~

• e,*,u~ua,. . **650Et'U 2*6.J61!tU 1, lHt f.t' -,,*az*E l 1

. ' . h~2Ctf: U 0,0-a,aaoeu;a o.~

" 7,U3EtU le726E+U '*lH!t&J .,209! 11 2ellll!+l2 1 e2961!+11

-29""'. &elS08E*l3

~-

30 1e263E. 13 'llo920E U 9*441Etl2 &e469l! U! l*lf>81! -l2 le806t! l2 V1736E+ll 4,29BE II 111,9131! 10 0,0 9e9911Ul2 7,*UaE U 7e5Pll+tJ ,* &eU2 .2*11YIC l2 &e*B~ijf12' 8,0071!:+H :11,G H+lf. 8,2444 la .a.a u

-;i:s*

7*409!U2 a,,nuu 1,7131!.tll .:s,.Oll! U *1.H~l! JI ** ,,.,~ * ., ~.atHtli ** ,,,.,11 .,,us+ao o.o 1-.io9e 1.2 4el111! U . *; 002e: t I 2'". 2

• 3o"IW!+ l2 **.-.-.:ufll! 12-*t;v.lZ!"fl)-4.:, f6E t'l-1~9*&*CH'l 4-; 898!'.+ Io- o . o - - - - - - -

,'H 3e241fH2 2e620f! U Z, 4v9f t 12 1, 09E+&2 8el42E+ll 4e605E+11 *,,Sl531! 11 1,1321! 11 3,44ot: IO 0,0

.n **72lftll '* 1601: ll Oed4*~tll J,d20~+11 2,171E+ll le2l1~ 11 6,a,~i+IO 3t~l~t10 lelOOE+IQ OeO 30 1.ddot:tll l 16J0Etl I le&72!tll ~,23$1! 10 S!~,,srio ~*07~etl0 J,710Gtl! ~~,~~, ~, *3,o,4E 09 O,O l1 o.o o.o o.o * *-*-*-o~o--**** -1;-o"----0-;o----b-;o * * 'o.o * -;-o -* o;o-----------

~ ci.sis t..-11. 'S,i;l..3EWL S.1:l'4~4:11._ .. l,IIJIH+J~

. :*( , ... 1*...

1.1 s_i._i.... ,~ * ,

• OC> 7*.IU...l.:i!l~-ih.WLSr.:~..:..~::::;l!*~e..::.,;:.t,11.:.:!;:.'...:..J:.*1..t.1~b:.:;;:..:*.;.:1~1.=-*..;*:,.::'1:.:,,.:.~.:::SS:.i::=,*~.:.,,!.:.1-1*

V. EQUIVALENT MWD/AT (USING DESIGN DATA CONVERSION) 3.08 X 1020 nvt Zone I a:

17 = 3,080 MWD/AT

, 1 X 10 nvt/(MWD/AT)

Zone II 4.95 x 10 20 nvt

= = 4,950 MWD/AT 1 x 1017 nvt/(MWD/AT)

Zone II I 1.91 X 1021 nvt a: = 19lJ100 MWD/AT 1 x 1017 nvt/(MWD/AT)

• Zone III Saturation

=

13 x 1020 nvt 1 x 1017 nvt/(MWD/AT)

C 13,000 MWD/AT 2

Zone IV = 4.12 x 10 20 nvt = 4,120 MWD/AT 1 x 1017 nvt/(MWD/AT)

A-15

VI. CALCULATING THE AVERAGE STORED ENERGY PER ZONE

1) USING WINDSCALE DATA AND CURVES (Fig. 12.2, 155°, 205°, 255° Curves)3

2) USING HANFORD DATA AND CURVES (Fig. 12.2. 30° Curve)~

-kE and applying the equations S = S (1 - e )

and S(T) "" a e-T/B from Nightingale's Nuclear Graphite. pp 329-331 *

• A. Method 1 (using W1ndscale Data and Curves for 150° and 250°)

Zone I a) S 30

= 685 [l - e-(.526)(3.08)]

S = 685 [.802]

• 549.5 30

-30/71.2 b) S(30) =a e 549.5 =a ( .656)

• c) as 837.6 S(83.5} "" a e

-83.5/71.2 S(83.5) = 837.6 (.3095) = 259.2 cal/gram A-16

[ -(.526)(4.95)]

Zone II a) S = 685 1 - e 30 s  : 685 (.926) C 634.3 30

-30/71.2 b) S{30) c a e 634.3 = a ( .656) a ., 966.9

-83.5/71.2 c) S(83.5) = a e

• Zone III a)

S(83.5) = 966.9 (.3095)

S 250

=

[

200 1 - e

= 299.3 cal/gram

-( .151) (19.1)

]

S = 200 [.944]., 188.8 cal/gram 250

-(.242)(4.12)

Zone IV a) S = 600 [ 1 - e ]

150 S s 600 [.631] = 378.6 cal/gram 150

• Volume Weighted Average Stored Energy Per Wedge (Method 1)

Zone I c 259.2 cal/gram x .156 = 40.4 Zone II = 299.3 cal/gram x .156., 46.7 Zone III = 188.8 cal/gram x .379 = '71.6 Zone IV = 378.6 cal/gram x .309 ~ 117.0 275.7 cal/gram A-17

B. Method 2 (using Hanford Data and Curves)

Zone I a) Same as Method 1 S(83.5) = 259.2 cal/gram Zone II a) Same as Method 1 S(83.5) s 299.3.cal/gram

• Zone III a) S S

30 30

• 685 [l - e

= 685 [l J

• -(.526)(19.1)]

• 685

-30/71.2 b) ( ) aae S30 685 = a ( .656) a s 1044.2

• c) S( 250) 1: a e

-250/71.2 5(250) = 1044.2 (.0298) a 31.2 cal/gram A-18

.\*

-(.526)(4.12)]

Zone IV a) S c 685 [ 1 - e 30 S = 685 [.885] = 606.5 30

-30/71.2 b) S(30 ) = a e 606.5

• a ( .656) a* 924.5

-150/71.2 c) S{150) = a e

• S(150) = 924.5 (.1216) E 112.4 cal/gram Volu~e Weighted Average Stored Energy Per Wedge (Method 2)

Zone I

• 259.2 cal/gram x .156 = 40.4 Zone II

• 299.3 cal/gram x .156 = 46.7 Zone III = 31.2 cal/gram x .379 c 11.8 Zone IV = 112.4 cal/gram x .309 c 34.7 133.6 cal/gram A-19

APPENDIX A REFERENCES

1. E. Robert Schmidt, *oesign Analysis of the Graphite Reflector Elements,N Missouri University Research Reactor Design Data, Vol. I, p. 18.

2. W. K. Woods, L. P. Bupp and J. F. Fletcher, "Irradiation Damage to Artifical Graphite,N Proceedin s of the International Conference on the Peaceful Uses of Atomic Energy , p. 6.

3. R. E. Nightingale, ed. Nuclear Graphite (New York:A~ademic Press, 1962}, P. 329.

4. ibid, p. 329

• A-20

APPENDIX B THERMAL COLUMN I. AVERAGE FAST FLUX DISTRIBUTION (>.821 Mev)

Derived from Missouri University Research Reactor Design 0 Data, Vol. I, *Radiation Heating 1n the University of Mimssouri Research Reactor.

A. FRONT FACE:

10 2 2 x 10 n/cro -sec (65 cm from reactor core centerline)

B. FIVE CENTIMETERS FROM FRONT FACE:

10 2 1 x 10 n/cm -sec (70 cm from reactor core centerline)

• C. TEN CENTIMETERS FROM FRONT FACE:

9 6 x 10 n/cm -sec 2

(75 cm from reactor core centerli.ne)

D. THIRTY CENTIMETERS FROM FRONT FACE:

8 2 7 x 10 n/cm -sec (95 cm from reactor core centerline)

E. MID PLANE:

6 2 6 X 10 n/crri -sec {141 cm from reactor core centerline)

• F. BACK FACE:

1 X 4

10 n/cm -sec 2

{217 cm from reactor core centerline)

B-1

II. EXPOSURE CALCULATIONS A. FRONT FACE:

10 2 8 18 (2 x 10 n/cm -sec)(3.78 x 10 sec}* 7.56 x 10 nvt

8. FIVE CENTIMETERS FROM FRONT FACE:

10 2 8 18 (1 x 10 n/cm -sec)(3.78 x 10 sec) z 3.78 x 10 nvt C. TEN CENTIMETERS FROM FRONT FACE:

9 2 8 18 (6 x 10 n/cm -sec)(3.78 x 10 sec)~ 2.27 x 10 nvt D. THIRTY CENTIMETERS FROM FRONT FACE:

• E. MID PLANE:

(7 x 10 8

6 2 8 17 n/cm -sec)(3.78 x 10 sec)= 2.65 x 10 nvt 2 8 15 (6 x 10 n/crn -sec){3.78 x 10 sec)= 2.27 x 10 nvt F. BACK FACE:

4 2 8 12 (1 x 10 n/cm -sec)(3.78 x 10 sec)= 3.78 x 10 nvt B-2

III. MWO/AT EQUIVALENCY A. FRONT FACE:

7.56 x 1018 nvt c 75.6 Mwd/At 1 x 1017 nvt/(Mwd/At)

B. FIVE CENTIMETERS FROM FRONT FACE:

3.78 x 1018 nvt = 37.8 Mwd/At 1 x 1017 nvt/(Mwd/At)

C. TEN CENTIMETERS FROM FRONT FACE:

• 2.77 X 1018 nvt 1 x 1017 nvt/(Mwd/At)

= 22.7 Mwd/At D. THIRTY CENTIMETERS FROM FRONT FACE:

2.65 x 1017 nvt = 2.65 Mwd/At 1 x 1017 nvt/(Mwd/At)

E. MID PLANE:

• 2.77 X 1015 nvt 17 l x 10 nvt/(Mwd/At)

• .023 Mwd/At F. BACK FACE:

3.78 x 1012 nvt c 3.7 x 10-5 Mwd/At 1 x 1017 nvt/(Mwd/At)

)

B-3

I " ) .

IV. TEMPERATURE Temperatures within the thermal column have been measured by thermocouple and all points exceed 100°F (37.8°C). For conservatism, the graphite temperature will be assumed to be 30°C.

V. CALCULATION OF STORED ENERGY A. FRONT FACE:

-kE S=S [1-e]

m

-(.526)(.0756)]

S = 685 [1 - e

• S =

-2 685 [3.90 x 10 ] = 26.7 cal/gram B. FIVE CENTIMETERS FROM FRONT FACE:

-(.526)(.0378)]

S = 685 [ 1 - e

-2 S = 685 [1.97 x 10 ] = 13.5 cal/gram C. TEN CENTIMETERS FROM FRONT FACE:

• S = 685 [ 1 - e S =

-(.526)(.0227)]

2 685 [1.19 x 10-] = 8.1 cal/gram D. THIRTY CENTIMETERS FROM FRONT FACE:

-(.526)(2.65 X 10- 3 )

S = 685 [ 1 - e ]

-3 S = 685 [1.39 x 10 ] a .954 cal/gram 8-4

E. MID PLANE:

-(.526)(2.3 X 10- 5 )]

s C 685 [1 - e S = 685 [1.21 x 10-S] z .008 cal/gram F. BACK FACE (negligible)

VI. THERMAL COLUMN GRAPHITE VOLUME ANO MASS

• Back Section: 48 inches x 48 inches x 48 inches = 110,592.00 in3 Front Section: 37.5 inches x 37.5 inches x 12.25 inches = 17 1 226.56 1n 3 Total 127,818~56 in 3 127 ' 818 *56 1n 3 x {2.54 cm) 3 *

(1 in)3 cm 3 6 Total Volume + 127,818.56 1n 3 x 16.387 in 3 = 2.095 x 10 crn 3

Total Mass + (2.095 x 106 cm3)(1.6 g/cm 3 ) = 3.35 x 10 6 grams ( = 7400 lbs)

B-5

1,-t,,:Pf~PJ:+=ii:JPJ:+l~PJ:+:+:J~PJ:+:f~*=J:t::J~~+::+:J=Pf+:J:J[::f:f~=i=ll'.::i=:r:fJi:J:l:J;:crlJi:li!*Eri 12 l O -r 1--+T-t,-1-rTT I

, . . . . . . 1..

11* -:  :' :. l * ** 1-, *I* - " Ii

. I .... t.....:,-_ - .. . - '.:.

- ~  :- t::; i...... :- : .....

._ =~~I: *1 r-

• - u 4,1 C

"--I*

1

=. . .

~-=,': *"

i:: .~

,= t - ':- .;. * ,_

,. i:::. re : * : ~ >:- I-:

• -- -1 :c:

~ ":'

-= ~ ., ~.: ; 8 .:.. ~ I:- : ~_: r= c:

r::

=-

. :i ':.'_-:I*: ~ 'c: *:; '::.  :. . : ':

N e

u 41:)

t: ...... _ - ,_ -

• 0 I'\. - lo-i.-

.. I -

~

~ *.., t-u 1010

-1,~tt~1~~~1~~1-~tttt~r~j~~i1,~i~~~tl~~1=t~~1,~~=;~*~*-~:*;':*~\~tt~1-~~*-~~-;:*i*:~~t1:t~~~~;:t~-~:t~~t:t~~~:tj~~:f:lj~i:t:tjti:t~~ij

- GJ i::

... . , 1-, I, C

11 -IC I.< *

,;i  : 11:: i ,_.. ""

-- - <,: - *_- ~:~I~.~,

':- t. .*. - ~- =- ::

':::  : I<: ~a 1:.:. ,_ * ..:.. :..

-- -  : . *= = - ~ ~ : ;.: -- f.,

--= ~ '::

- - - -  : '\

~--*-i.-

Ji.... - - -

~

.' \

• o \

C

,~ I,; r.: =-:-  ;

'=

Dit;;t.&nce from Reac*r.or C.~rt:. C.t:nterline (cm)

B-6

. .-,.., r ',

l'igure o ic*.; ... d~

101 11 I* I II \. I* I ii I I II I J,.,,I J U ti J *i II* J.t 11 I\ I I I I I I I 11 l I I I Ill I I I I t I I I I I I I I i\ 1, .

l \ ~ .. I'\ 1~=,.. 1: *1=-: 1-

\

,\ \

.. \ -

... !t=f--1 1: ~

-1::.. i-

. F*.:. ,:t-=.t-:'t:

,3:: ,:. i.~ .. ~:;J:-.1::1-  :::.~-: -*-t-:i:li-§l 1:~t-f-£t ~:.- t--::-._ :t*-*-LJ.., ~t::f.t*t:

1:.1-:1:.f:1\I+ 1-*11- 1

..._.._ **:.~+

... ,-- *f-1*1-t- *t::tt:: :i: *---- ij..ffll~utt.-- ~ ....... +-+-f-1

.m,: ~~

r::t:t:1-

• +-1-t-+-

1-1g:1-+1-  !'- ....

l-+-+-11*1++-

f1° 1111-1-111111 IJum 1111111 ttU' u1,1 111 Hffl Ill IIIll Il11111111 IU I\. -:.

!IF: =-

u

• 1::

.r...:a...t= ';.

\

I\

"f:-:f-=-l-=-1 .I\!

\

\"

I

• 8-:1"1 j~L- ,j:fi += ."r-;+;-

~al\ a.---*- :I§ e....* I:-

-II-+-

a-:1~1 .. 1,-..1-*1 ;--t-t,,-1 -f--1 I l '---1:.m:i.:-- ~:.t:: -lj=_ttt f::.N:,1-1-+  :::r-.-r:- .r.......

1-f ::-t:-.*-

r-i:::::i;::

---

*- I-

i:.i-i:-~~

.~

f:f 1-+-1-1$U :.t::

1: I

...gu

'II 10' IIIIIIBAIIffm1-ltiffl-1111Wfffltf+ftNII UIOOl1/211 I\ \.

-*--=---*- .., \*=*= I::

F.f.l=l"*t-;~ :1:t.=  ;

1;;14: f - -1***. *I-:-

~

. J-:: -;'fif::t:, I"~ -t~t:..

t:""

I,\

\

lit

~£ 1-'I\

.., ~

..-c.

.1-1*¥.-M~h=t=-t:ff-t-l=c!-:t~

4-:1=-

~:.r::l. -* .... -

....... .. 1---i* r,

=

~

.... _. f-

tJ*. -rfJtftf,=... 1 ~-i--1--*-

J-t ............... :t::t:t *-....-

10 5

J:J+tttmm:ttmttt~ttt1EEt1-1111fflt1ffl11ffl-t-t-t-1 C

nmLllLmm---m~

I-

"" F ... F-  ;: I\, ** ..

I \I' I"

~f..-1=-'

~ F-f!:tlt.flt 1--

n - .. -ttf~ -t.~ f:: I- ..:

I-4

\0 100 1"50 200 Dlatance from Reactor Core Cente~llna (cm)

B-7

r

• APPENDIX C ANALYSIS OF THE HAZARDS ASSOCIATED WITH STOREO ENERGY RELEASE IN THE MURR GRAPHITE REFLECTOR I* CALCULATING TOTAL STORED ENERGY Calculated (Theoretical) Mass Total Stored Average Stored Energy (7 Wedges) Energy

!HIGH 275.7 cal/gram X 377,843 grams . 1.04 x 10 8 cal

,MID I

-205 cal/gram X 377,843 grams . 7.75 x 10 7 cal iLOW 133.6 cal/gram X 377,843 grams ,::

5.05 x 10 7 cal

~

II* ENERGY RELEASE FRACTION The amount of total stored energy releasable fn an anneal to 1000°C is 50%

to 80% of total stored energy. 1 For conservatism a 100% release of energy will be assumed.

III. CONSIDERATIONS FOR TRANSFERRING GRAPHITE ENER~Y TO TH£ CO~E The graphite reflector forms 2/3 of a right circular annular cylinder sur-rounding the core and fs separated from the core by the materials listed in Figure C-1.

Considering the total stored energy calculated 1n Step I, the amount of heat that can be transferred to the core depends on the energy conducted in the core direction.

Area of inner surface= 2w(9.684)(34.89) = 2123 in 2 Area of outer surface= 2w(l8.262)(34.89) = 4003 in 2 Total heat transfer area 6126 in 2 2123 Ratio of inner surface area to total heat transfer areas bl2'6 s .346 C-1

ALu.M uJ I.I ,.._

f.f fl~ cr.,L.

1rw~

FIU_L ,,.J (1.0.,)

Figure C-1 For conservatism, a fraction of .40 will be used to determine the total amount

• of energy in Step I) directed toward the core.

following reasons:

1)

This is conservative for the Heat dissipation in the intervening materials is not considered.

2) Heat conduction in the vertical direct'ion is not considered.

3) The heat transfer area of the outer reflector surfate is twice the area of the inner surface.

4) The distribution of highest stored energy is towards the outside of the graphite element (away from the core).

5) The AT, the driving force for heat conduction, is much higher towards the bulk pool area than towards the core.

C-2

ENERGY DISSIPATEO TOWARDS CORE HIGH 1.04 x 108 cal x .40 s 4.16 x 10 7 cal MIO 7.75 x 10 7 cal x .40 = 3.10 x 10 7 cal LOW 5.05 x 10 7 cal x .40 = 2.02 x 10 7 cal IV. CONVERSION OF ENERGY TO BTU HIGH 4.16 X 10 7 cal x 4.1819 E 1.74 x 10 8 Joules MID 3.10 X 10 7 cal x 4.1819 E 1.30 x 10 8 Joules LOW 2.02 X 10 7 cal x 4.1819 = 8.45 x 10 7 Joules

• = 1.65 HIGH 1.74 X 10 8 Joules x (9.47086 X 10- 4 ) X 10 5 Btu MID 1.30 X 10 8 Joules x (9.47086 X 10- 4 ) = 1.23 X 10 5 Btu LOW 8.45 X 10 7 Joules x (9.47086 X 10- 4 ) = 8.00 X 10 4 Btu V. CONVERTING TOTAL ENERGY TO ENERGY RELEASE RATE Stored energy is released over a period of time varying from 20 min to 60 min 2 in totally adiabatic laboratory conditions to 3 to 6 hours 3 in actual reactor annealing releases of energy *

• ENERGY RELEASE TIME n n HIGH [1.65 x 10 5 Btu] 45.8 Btu/sec 91.7 Btu/sec 1R3.3 Btu/sec 367.7 Btu/sec

• 1n MID [1.23 x 10 5 Btu] 34.2 Btu/sec 68.3 Btu/sec 136.7 Btu/sec 273.3 Btu/sec LOW [8.00 ~ 10 4 Btu] 22.2_Btu/sec 44.4 Btu/sec 88.9 Btu/sec 177.8 Btu/sec C-3

VI. COMPARING GRAPHITE ENERGY RELEASE RATES TO CORE DECAY HEAT Fission product energy release rate frOfTI a reactor which has heen operated to near equilibrium fission product concentrations 1s equal to approximately 6 percent of operating power.~ Thus the MURR core will produce decay heat at a rate of 0.6 MW which is equal to 568.7 Rtu/sec. Even at the most conserva-tive estimate of energy release time (7.5 minutes) the energy release rate from the graphite would be less than that of core decay heat.

VII. COMPARING THE CONSEQUENCES OF A GRAPHITE ENERGY RELEASE TO THE [OSS OF COOLANT ACCIDENT ANALYSIS FOR MURR

• The loss of Coolant Accident (LOCA) is the most severe credible accident for MURR (a double-ended rupture of the reactor inlet pipe between 5078 and the pool liner). Analysis of thfs accident is contained in MURR Hazards Summary Report, Addendum 4, Appendix E. Considering the core decay heat as the energy source to be dissipated, the results of this analysis show that DNB and fuel damage will not result from this accident. In fact, results of the analysis show that fuel cladding surface temperature would not exceed 281°F {with a starting temperature of 165°F). This accident results 1n a 116°F {46.7°C) increase in the cladding temperature of the fuel. This temperature is well below 1184°F (640°C), the temperature at which fission product release from UAlx fuel is appreciable.

Since the most conservative estimate of heat load from the graphite is less than the heat load (decay heat) considered for the LOCA analysis, the con-sequences of the release of graphite energy to the core would be less severe.

C-4

MAXIMUM ADIABATIC TEMPERATURE EXCURSION IN GRAPHITE REFLECTOR 1000 1000 C ~ 1.86 T + (8.34 x 10- 3 T2 ) _ (5.27 x 10-6 T3) 200 / p 2 1200 pooo c -= 3.748.25 cal/g-mole _ oc X 1 9-111ole 200 p 12g

= 312.35 cal/

g -

oc Ave C over ~his =

200I 10 e~

SOO = 312.35 cal/ 9 _ oc 1 x mrc; = .390ca1/

9

_ oc temperature range

• HIGH MAXIMUM ADIABATIC TEMPERATURE RISE 275 cal/gram +

275 cal/ gram

_390 -= 705°C 205 cal/

HIGH 205 cal /gram,., +

.390 gram = 525°C 133 cal/ gram LOW 133 cal/gram + _jgQ = 341°C C-5

APPENDIX C REFERENCES

1. R. E. Nightingale. J.M. Davidson and W. A. Snyder. *oamage to-Graphite Irradiated up to 1000°c.u Proceedin s of the Second Nations International Conference on The Peaceful Uses o 0111 c 9 *

2. same as 17 (main report)

3. same as 18 (main report)

4. s. Glasstone and A. Sesonske. Nuclear Reactor Engineering 3rd ed. (New York:Van Nostrand Reinhold Co ** 1967). p. 124

• C-6

I __ ,,

' .~'

APPENDIX D ANALYSIS OF HAZARD ASSOCIATED WITH STORED ENERGY RELEASE IN THERMAL COLIJMN The maximum stored energy at the front face of the thermal column is 26.7 cal/gram. The stored energy falls off to 13.5 cal/gram five centimeters into the front graphite section and less than 1 cal/gram at the back of the front graph-ite section {30 centimeters from front face). The maximum stored energy could con-tribute to a maximum adiabatic temperature increase of 128°C. This would correspond to a maximum temperature at the front face of 158°C under adiabatic conditions. Even this exaggerated temperature at the front face is incapable of autoignition of

• graphite.

Five centimeters into the graphite stack the maximuM adiabatic temperature increase would be 65°C. The maximun adiabatic temperature increase falls off even more drastically 30 centimeters fron the front face (4.8°C rise). The back section of graphite (48 inch cube) has negligible stored energy and a maximum adiabatic temperature rise of less than 4.8°C.

CALCULATIONS:

• A JB cp C

p a

C 1.86 + (8.34 x 10- 3 T) - {5.27 x 10-6 T2 cal/g-mole -

1.86T + (8.34 X 10-3 2

T2) 0 c

130 30 J cP = (241.B + 70.473 - 3.859) - (55.8 + 3.753 - o.47)

= 308.414 - 59.506 = 248.908 cal/g-nole - °C x 1 12 g-mole g

= 20.74 cal/g - °C D-1

p ,  ; ' **

• - ' l 130c .

Ave cp 30 1 rfur over this

• 100 P

• 20.74 cal/g - °C x c .2074 cal/g - °C temperature range FRONT FACE:

Max adiabatic temp increase c 26

• 7 cal/gram = 12s.1°c

.2074 cal/g - 0 c

, . 5 CENTIMETERS FROM FRONT FACE:

Max ad;abatic temp increase= 13 *5 cal/gram 0 = 65.1°C

.2074 cal/g - c 30 CENTIMETERS FROM FRONT FACE:

Max adiabatic temp increase 1 c

.2074 cal/ ram cal 7g - 0 c*

= 4.8°C 0-2

DOCKET NUMBER E.TITION RULE P~M

( a'/;;-~ _i;,g4 ))

.fib-44 @

S-COOLED REACTOR ASSOCIATES JOCK iff U:3NP' 10240 Sorrento Valley Rd. ste. 300

-iiiiill. .W-'111(" Son Diego, CA 92121-1605 (619) 455-9500

'87 FEB - 4 P12 :20 I,

30(.,  !:.. *it' ,

L.D. MEARS ... R:...,~,.., *~

General Manager January 30, 1987 Mr. Samuel J. Chilk Secretary of the Commission U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Attention: Correspondence and Records Branch

Dear Mr. Chilk:

Comments on Petition for Rulemaking on Graphite; 51 FR 31341 This letter provides Gas-Cooled Reactor Associates'

( GCRA) comments on the petition for rulemaking dated July 7, 1986 that was filed by the Committee to Bridge the Gap regarding graphite-moderated reactors. GCRA represents a number of U.S. utilities and potential users who are interested in the development and commercialization of the gas-cooled reactor in this country. As such, our comments apply to the petition and its application only to commercial gas-cooled reactor designs.

Two bases for the petition are identified. First, that graphite fires should be considered as credible events by the NRC as a result of the reported graphite fire which occurred at the Chernobyl plant in the Soviet Union.

Second, that there are new data which show that the Wigner stora energy effect ha s b~en u*nd~rasti:ma ted. Th e applicability of each of these bases to the design of commercial gas-cooled graphite-moderated reactors in the U.S. is addressed in the following.

Credibility of Graphite Fires The first basis fails to recognize both the requirements for a graphite fire and the significantly different probabilities of events, which are very much dependent on the reactor design, that could fail design measures taken to avoid such fires. The line between "credible" and "non-credible" must be drawn on the basis of a particular design and not on a generic basis.

f£8 S 1987 ref .** **o******..

f U S. NUCt.EAR REGULATORY COMMISS ION, DOCKETING & SERVICE BRANC H OFF ICE OF TH£ SECRETARY OF THE co I M,~!:>10 Docl .s Postmark D 1te _ _,_.::::..::'-l--=--- -- - --*r C

Add1 Cc,

~pecjal O~,*J0uuon

Mr. Samuel J. Chilk Page Two January Jo,* 1987 The conditions for a graphite fire include the following:

I. that the graphite be hot enough and remain hot enough to initiate the rapid oxidation process (the fire), and

2. that sufficient air (oeyg.;m) be av.:::!.lable to austain the process .

• Commercial gas-reactor designs in the U.S. operate at high enough temperatures that the first condition is mat.

However, the design prevents the second conditions from being met by inerting the graphite in its helium coolant which is contained by a high integrity pressure vessel.

Likewise, the graphite moderator in the Chernobyl-type designs also normally operates at high temperatures.

However, the Chernobyl-type reactors inert the graphite in a nitrogen/helium.mixture contained only by a low pressure, hermetically sealed vault to prevent meeting the second condition.

In the Chernobyl accident, the initiating event was one or more rapid power excursions due to an inherently positive feedback from the void coefficient of the water coolant. Apparently, these power transients caused rapid heating of water in the coolant channels of the reactor leading to an explosive volumetric expansion of the water to steam which grossly failed the integrity of the reactor vault allowing air access to the graphite.

By comparison, the gas-cooled reactor systems in the U. s. have an inherently negative temperature reactivity feedback such that the autocatalytic initiating excursions which took place in the Chernobyl design are not possible.

Further, even if rapid heating in the core did take place, the single phase helium coolant used in U.S. gas-cooled reactors is not capable of the explosive volumetric expansion undergone in the Chernobyl reactor.

In sum, the failure which occurred in the Chernobyl plant exposing the hot graphite to air was caused by an explosive event triggered by the specif.le characteristics

Mr. Samuel J. Chilk Page Three January 30, 1987 of that system for which there is no analogy in a gas-cooled reactor in the U.S. Thus it is argued that it has no impact on the assessment of the credibility of a similar event occurring in the commercial gas.-cooled graphite-moderated designs in the U.S.

• stored Wigner Energy The designs for commercial gas-cooled gr*aphi te moderated reactors in the U.S. operate at temperatures well above that at which any significant Wigner energy storage occurs. Thus, this basis :for *the petition is not applicable to commercial gas-cooled reactor designs.

Recommendation We therefore conclude that neither of these bases is relevant to the design for commercial gas-cooled graphite-moderated reactors in the u. s. Thus, to the extent that the proposed rule would apply to commercial gas-cooled graphite-moderated reactors, we recommend that the NRC deny the peti ti,on for the proposed rul.emaking.

We would be pleased to provide additional details on our comments if required .

• Sincerely, L. D. Mears General Manager LDM:bl cc: GCRA Management Committee Andy Millunzi, DOE/HQ Gene Northrup, GA R. o. Williams, PSC

uvvru:.1 1-.vn11Ji..,n l?ETITION f3ULE PRM ~?J-44 /47)

( ..51 F£ o/3-4}) (!!!:ti University of Wisconsin

, QC.KE:!:.

NUCLEAR REACTOR LABORATORY ADDRESS: I '.iNRC NUCLEAR ENGINEERING DEPARTMENT 130 MECHANICAL ENGINEERI NG BUILD ING PHONE 282-33112, AREA CODE eos MADISON, WISCONSIN !53708

'87 FEB -4 P12 :18 ufl iv~ '

Comment on Pet it ion IOtK Ri.11lemak i ng*

PRM-50-44 Re: License R-74, Docket 50-156 January 29, 1987 Secretary U. S. !l<<Jclear Regulatory Commission Attention: Correspondence and Records Branch Washington, D. C. 20555

Dear Sir:

This letter comments on Petition for Rulemaking PRM-50-44. The comments are those of the operating organization of the University of Wisconsin !l<<Jclear Reactor Laboratory (License R-74).

We believe the contentions of the Petitioner are without merit because of no technical basis for application to a reactor similar to ours.

First, the Petitioner couples the fire in a USSR power reactor with stored Wigner energy in graphite. rt> information available to us indicates Wigner energy to have been a factor in the USSR accident. In fact, operating temper-

• atures assure in the graphite of the USSR reactor appear to have been high enough to continuous annealing of the stored energy.

The Petitioner indicates a basis of their contentions as the difference between calculated and measured energy levels in the UCLA Argonaut-type reactor graphite moderator. We do not have the capability to comment on either the measurements or the calculations since the data are not available to us. More important is the likelihood of a graphite fire being initiated by a Wigner release and resulting in release of fission products. We comment on this below.

Graphite is used in three components of the University of Wisconsin reactor:

A. Top and bottom end reflectors in fuel elements; B. Reflector elements; C. Thermal Column.

U. S. NUCLE R REGULATORY COi\iMISSI ON.

DO"K .T1NG & sr1NICE BR'.NC H Ot r ICE OF T t. SCC~ET,,flY Or THE COM -~ ,I D~ ~ .... t :.:, ...... ,.... s Postmark D e_ _,__ --l'- -

Co..,1es 1 Ar d1 C

~p~c1 u,!., 1-.;n

Secretary, U.S.N.R.C. January 29, 1987 Components A and Bare encapsulated in aluminum or stainless steel so that oxygen is excluded. Both operate at high enough temperatures that annealing keeps Wigner energy at low levels. Both operate under 20 feet of water so that adequate cooling is available shoul d a Wigner release take place and combustion is not possible due to the cool ing and oxygen excluding action of the water.

The thermal column graphite is in 4 inch square by three foot long stringers. Although these stringers are not encapsulated to exclude air, a Wigner energy release resulting in a fire wuld not cause fission product release because:

a. The graphite pieces are large enough to be difficult to ignite;

b. The fast neutron flux in the thermal column is so low that a significant amount of stored heat is not generated;

c. There is no fuel in the thermal column and, thus, no fission products would be released if a fire did occur

• The Petitioner asks for rulemaking requiring graphite fire response and evacuation plans. Fire responses are inc l uded in the existing NRC-approved Emergency Plan and further response plans would not be needed even if the graphite presented any special hazards.

The Petitioner also requested that a gra phi te stored energy measurement program be required. Such measurements would lead to unnecessary radiation exposure to personnel and are not needed because of the lack of hazard.

We urge that the petition be denied due t o lack of technical basis.

Very truly yours,

~

• Reactor Director RJC:mld

DOCKET NUMBER N RULE PRM 51J-44 /ij)

(ff!F,,, JIJ47) ~

_..__________________ GATechnologies-------------**---~

• DOCK ET EC*

~

GA Technologies Inc.

P.O. BOX 85608

'87 FEB - 3 P3 :51 SAN DIEGO, CALIFORNIA 92138 (619) 455-3000

&ff DOC

• January 28, 1987 Secretary U.S. Nuclear Regulatory Commission Washington, DC 20555 Attention: Correspondence and Records Branch

Subject:

Gentlemen:

DOCKET NO. PRM-50-44 This letter provides GA Technologies' comments on the petition for rulemaklng flied by the Committee to Bridge the Gap (CBG), Docket No.

PRM-50-44. Comments In th Is Ietter are conf I ned to the requested rulemaklng as It relates to the high temperature gas-cooled reactor

( HTGR). Comments re I at Ive to the TR IGA research reactor are be Ing provided In separate correspondence.

The petition requests that the NRC amend Its regulations to require operators of reactors that use graph Ite as a moderator or ref Iactor

( 1) to prepare and subm It for NRC approva I f Ire response p Ians and evacuation plans for a graphite fire, and (2) to measure the energy stored In their graphite and revise their safety analyses to consider

• the risks and consequences of a graphite fire In their facll !ties.

The basis for the petition Is CBG's contention that the accident at Chernobyl shows graphite fires to be credible events for which I teen-sees do not have adequate fire response and emergency plans. CBG also contends, based on their Interpretation of a recent report by re-searchers at UCLA, that WI gner energy stored In graph Ite has been severely underestimated In earl !er studies.

GA Technologies bel Ieves that these proposed requirements are unneces-sary and, If adopted, would have an unduly burdensome Impact on I teen-sees with reactors containing major graphite components, with no cor-responding Increase In publ le health and safety.

10955 JOHN JAY HOPKINS DR. , SAN DIEGO, CALIFORNIA 9 2121 Acknnw1t~.i:,d bv carer. .. ~.. ~

U, S. NUCLEAR REGULATO'<" C .*.t..,310 N o o-::KET 1NG & SER' ,Cr: 8 ,, , , ..: rl OFF CE OF THE S  : 1<{

Of- I HE CO M .. 1,..,_1. ,,

Do~~r cnt St.- : .~s Pos;r.1rk :) e _ \/_?-Cf ~1

Secretary U.S. Nuclear Regulatory Commlsslon January 28, 1987 Page Two The basrc cause of the Chernobyl accldent, as documented rn the Sovret report presented at the IAEA Conference rn August 1986, was a reactor power surge lnltlated by a large posl tlve reactrvrty Insertion and accelerated by the RBMK reactor's posttrve vold coefflctent of reac-tivity. At the time of the accident the reactor was be(ng Intention-ally operated outside Its design bas rs In violation of numerous safety procedures. The fun damenta I d t f ferences between the HTGR r n the Unrted States and the RBMK reactor are wel I known (e.g., Refs. 1, 2, and 3)1. Primarily, these dlfferences are as fol lows:

• 0 In the HTGR, the he I rum coo Iant rs r nert and cannot react wIth the graphite or the ceramic fuel particle coatings. In the RBMK reactor, an exothermic reaction between the boll Ing water coolant and the ztrcal toy metal fuel cladd ing can occur at temperatures that can be reached very quickly under transient conditions.

o The HTGR has negative reacttvrty feedback over rts entire range of operation. The reactrvrty effect assocrated wlth loss of the slngle phase coolant rs negative. The two phase bol I Ing water coolant of the RBMK reactor, In comblnatton wlth other factors, can result In a posttrve reactrvrty rnsertlon and feedback when primary coolant ts lost or voided.

0 In the HTGR, the f ue I rs therma I Iy coup Ied wrth the graph t te moderator; resultlng rn a core that heats up slowly during power or flow transients. In the RBMK r eactor, the fuel rs not ther-mally coupled wlth the moderator, and the thermal response time of the core rs stmllar to that of I lght water reactors.

These differences preclude an HTGR accident of the type that occurred at Chernoby I and make an HTGR acer dent that Involves s Ign If Icant graphite oxidation highly rncredlb le.

With regard to the credlbll rty of graphlte "flres", the fire rnrtrat ly observed at Chernobyl was caused by burn lng fuel and claddlng ejected from the core rn the accident, not burn ing graphite (Ref. 4). The lsee attached I (st of references.

Secretary U.S. Nuclear Regulatory Commission January 28. 1987 Page Three major fission product releases were not associated with graphite oxi-dation. but with the Initial steam explosion and later with the build-up of decay heat energy trapped In the core by the tons of sand.

boron. and dolomite used to smother the core (Ref. 5). The Soviets estimate that about 10% of the core graphite was either ejected In the accident or "burnt" over the next nine days (Ref. 6). Conservatively ass um Ing that the tu 11 10% "burnt" over the n I ne day per Iod. th Is corresponds to an average loss rate of 0.05% per hour. definitely not a flaming. smoking fire In any normal sense of the word; "slow oxlda-t Ion" wou Id be a better descr I pt Ion of the event. Hence. graph Ite

• "burning". to the extent that It occurred at Chernobyl. was a secon-dary et feet of the ace Ident that had re I at Ive Iy I Itt Ie Impact on otfslte doses.

Extensive studies of graphite at Argonne National Laboratory (Ref. 4).

GA Technologies. Oak Ridge. Brookhaven. and others (Ref. 7) have shown that burning of large pieces of pure nuclear grade graphite I Ike those used In U.S. graphite-moderated reactors Is very difficult to Initiate and sustain. Extremely high temperatures and an uni lmlted supply of oxygen are required. unless the graphite Is broken Into very smal I pieces. For the HTGR In the u.s ** a scenario resulting In core dis-ruption and such a "graphite fire" has been properly considered by the NRC staff to be Incredible based on the HTGR plant design features.

An examp Ie of the d If f Icu Ity of burn Ing graph I te Is I 11 ustrated by electric-arc lamps and electric-arc furnaces. Both are Important appl !cations of graphite and II lustrate the chemical stabll tty of that mater I al at very h Igh temperatures under atmospherl c condltl ons. In the e Iectr I c-arc I amp e Iectr Ic Ity Is passed through the gap between graphite electrodes. creating a source of I lght. In this appl !cation the temperature of the electrode tip Is typically about 3700oC. I.e **

near the vaporization temperature of graphite. At this temperature the consumption of a 0.5 Inch diameter graphite electrode Is about 3 Inches per hour. half due to volatll lzatlon of the tip and half due to oxidation In the open air. When th e electric current Is Interrupted.

the electrodes Immediately cool down and no self-sustained combustion of the graphite occurs. even after long operation at these very high temperatures (Ref. 8).

In the specific case of the Fort St. Vraln HTGR. referred to In the CBG petition. an Improbable accident scenario (frequency less than 10-9 per year) resulting In maximum graphite oxidation was hypothe-sized In May 1986. fol lowing the Chernobyl accident. Means to stop

Secretary U.S. Nuclear Regulatory Commission January 28, 1987 Page Four the oxidation were also Identified (Refs. 1 and 2). The scenario and methods for dea I Ing wIth It were rev Iewed and accepted by the NRC staff, which determined that continued operation of Fort St. Vraln Is Justified (Ref. 9).

With regard to evacuation plans that address graphite fires, the emer-gency planning regulations (10CFR50.47) already Incorporate protective action requirements that are governed by projected offsfte doses. The question of whether the radlonucl Ide release results from a "graphite fire" or some other accident scenario Is Irrelevant. When the crlter-

• 1on for protect Ive act Ion Is met, predeterm I ned protect Ive act Ion (sheltering or evacuation) ls taken. In the case of the Fort St.

Vraln Radiological Emergency Response Plan, these protective actions already consider the f I sslon product releases cited In the petition.

Although the supporting design data for Fort St. Vraln Indicate It was not necessary, It was conservat Ive Iy assumed, to be cons I stent wIth the emergency planning basis for I lght water reactors, that 25% of the equll lbrlum Iodine Inventory Is available for release In establ lshlng the Emergency Planning Zone radii (Ref. 10). Additional special plans for any particular type of accident are unnecessary and would require I Icensees to devote extra resources for the Ir deve Iopment and ma In-tenance with no added benefit to the publ le health and safety.

With regard to Wigner energy storage, discussions between the staffs at GA Technologies and UCLA Indicate that CBG has made very selective use of the data In the UCLA paper to support Its content Ion that WI gner ener gy storage has been severe Iy underest Imated. The h I gh va I ue of WI gner energy storage ref erred to by CBG was a s Ing Ie data point occurring In a very cold piece of graphite In a very high fast flux region of the UCLA reactor core. It was not representative of the behavior of the graphite as a whole, or of the total stored energy In the core. The Important cone I us Ion from th Is study Is that th Is stored energy Is concentrated In a th In Iayer no more than 3 Inches wide around the outside of the fuel boxes In the UCLA Argonaut reac-tor. The comp Iete data presented In the paper ref erred to by the petftftoners support the results of t he NRC-funded generic study on Wigner energy storage CNUREG/CR-2079). One of the authors of the UCLA paper has addressed these points In detail In his comments on the CBG petition (Ref. 11).

Fort St. Vra In and other graph I ta-moderated power reactors norma I Iy operate at temperatures above 260°c. At these temperatures, graphite Iatt Ice dIs Iocat Ions are annea Ied out as they are produced, and the stored Wigner energy Issue ls Irrelevant for these reactors (Ref. 12).

Secretary U.S. Nuclear Regulatory Commission January 28, 1987 Page Five Evaluations at GA Technologies have shown that even If FSV were opera-ted at 5% power (I.e., below 260°C) for an entire month, the accumu-1ated stored energy wou Id, If re Ieased ad Iabat Ica I Iy, Increase the core temperature by Iess than 5oC. Th Is temperature Increase Is Insignificant.

In summary, the petitioner's allegations are unsupported by the evi-dence, and the NRC has already adequately addressed potential problems

• due to graph Ite ox Idat Ion and WI gner energy re Iease In Its gener Ic studies and I lcenstng activities. No additional rulemaklng Is neces-sary, and GA Tech no I og Ies recommends that the NRC deny the pet Iti on for rulemakfng.

If you wish to discuss any of these comments In further detall, please cal I me at (619) 455-2120.

PI I R. A. Dean Senior Vice President Reactor Programs

REFERENCES

1. H. L. Brey CPublJc Service Company of Colorado) letter to H. N.

Berkow CNRC), "Chernobyl Nuclear Reactor Accident and Its lmpl !-

cations Upon Fort St. Vraln", P-86358, May 9, 1986.

2. H. L. Brey CPSC) letter to H. N. Berkow CNRC), "Response to Re-port on Chernobyl Accident", P-86641, December 4, 1986.

3. R. A. Dean (GA Technologies), Statement Submitted for the Record, Senate Energy and Natural Resources Committee, United States Senate, "The Chernobyl Accident and lmpl Jcatlons for the Domestic Nuclear Industry", June 19, 1986.

4. "Report of the U.S. Department of Energy's Team Analysis of the Chernobyl-4 Atomic Energy Station Accident Sequence", DOE/NE-0076, November, 1986

5. "The Accident at the Chernobyl AES and Its Consequences", State Committee for Use of Atom Ic Energy In the USSR, August 1986.

(Translated from the Russian, Department of Energy, NE-40, August 17, 1986.)

6. "Chernoby I : The Sov Iet Report", Nye I ear News, Vol. 29, No. 13, October, 1986, p.65.

7. R. P. Wlchner,"Alr Oxidation of Graphite-Imp! !cations for HTGR Ace I dents, 11 ORNL/GCR-80/7, Oak RI dge Nat Iona I Laboratory, Apr I I 1980.

8. C. L. Mantel I, Carbon and Graph i te Handbook, lntersclence Pub-I lshers, 1968, p.349 *

• 9.

10.

H. R. Denton CNRC) letter to R. D. Lamm (Governor of Colorado),

May 29, 1986.

D. W. Warembourg CPSC) letter to B. K. Grimes CNRC), "Fort St.

Vraln Unit No. 1 Emergency Planning", P-80066, April 1, 1980.

11. H. Pearlman letter to the Commissioners, "Committee to Bridge the Gap; Petition for Rulemaklng", Docket No. PRM-50-44, October 29, 1986.

12. R. E. Nightingale, Nuclear Graphite, Academic Press, 1962.

DOCKET NUMBER ~

~ ... 44 &

-:r..

UNIVERSITY OF MISSOURI

~ET) TION RULE PRM

(~IF~ ~l-31))

January 30, 1gg7 OOC.:KETEC U~ 1 mc Research Reactor Facility Research Park

• s7 FEB -2 P2 :36 Columbia, Missouri 65211 Telephone (314) 882-4211 Secretary U. S. Nuclear Regulatory Commission Washington, D. C. 20555 ATTENTION: CORRESPONDENCE AND RECORDS BRANCH

SUBJECT:

DOCKET NO. PRM-50-44 The attached literature review is submitted for consideration by the Nuclear Regulatory Commission in their determination of the need for proposed rule-making, PRM-50-44, that appeared in the Federal Register/Vol. 51, No.

170/September 3, 1986.

This literature review contains general information regarding graphite stored energy and graphite oxidation and was compiled to assist the reactors in the Test Research and Training Reactor (TRTR) community with their specific responses. This literature, though general in nature, shows how difficult it is to start a graphite fire in large block geometry, how difficult it is to maintain it without abundant supplies of oxidant, and how relatively easy it can be put out by removing the oxidant. The literature provides examples where the re l ea se of stored energy ha*s created limited temperature excursions,

• during actua l reactor anneals.but does not support or suggest the possibility of autoignition of graphite from the release of stored energy.

Sincerely,

~---~ .Jol\ e-....

Don M. Alger \

TRTR Executive Committee Attachment COLUMBIA KANSAS CITY ROLLA ST. LOUIS an equal opportunity institution 2 1987

~

j bycar'1. *m-r -- r,r.rr.W

I I er "' 0

~ ~ .....

0

~

-0 0

U>

C I 0 <fl i:i

- n U> 3 0 z

i 9  ;:o "'x 0

~

0

('

~ "1 r 0,

('

0 i, II l C -j

,, ~ (l) .... 2

-t-:::. ~"'

Cl, C

..,. w I'

I .... ""  !:'

I I I

- 1-,..) ;'.

('

C'

. r C I

I I -,~°"

~ t '

(.

.~

r

~

c

~

c;

"' (.' .

I

STORED ENERGY IN IRRADIATED GRAPHITE By

• Walt A. Meyer, Jr.

University of Missouri Research Reactor Facility December 10, 1986

PREFACE As is true of many engineering problems, there is no indication of what "technical steps should be started for solution. In fact one is not quite sure whether there is any technical work to be done.

The questions of the hazard of stored energy in graphite and of the com-bustibility of graphite have been answered before as is evidenced by the litera-ture presented here. The only problem is that the previous results and answers were not generalized and extended to specifically cover the graphite applications in research reactors.

There is certainly considerable research that can be extended to show that stored energy is not a problem in the current applications at research reactors.

Research reactors appear to cluster at the low end of the stored energy accumu-lation continuum, even though there is great disparity between reactors with

• respect to accumulated neutron dose. One set of research reactors (i.e., Trigas, Argonauts) do not appear to have a problem because of low integrated flux (UCLA nvt corresponds to 40 Mwd/At, a point that virtually corresponds to the origin of the majority of stored energy vs exposure data). Another set of reactors are those with high integrated flux, such as MURR (great enough for saturation ef-fects to be expected) which appear to not have a problem due to the elevated temperature of the canned graphite (>150°C).

This review of the literature was intended to be an annotated reference to relevant infonnation regarding stored energy in graphite. Each reactor will no doubt generate its own approach to answering the stored energy question (i.e.,

Iowa State University has devised a stored energy calculation that allows the calculation of an upper bound on stored energy).

One question regarding low power reactors is whether a core transient could create the 50° to 100°c increase in graphite temperature needed to start a stored energy release, regardless of the stored energy content?

• On a generic scale, air cooled reactors with much higher stored energy than one could expect to accumulate in a research reactor and with a much greater vol-ume of graphite have been shown to have maximum temperature transients during annealing in the neighborhood of 300°C (certainly not high enough to damage fuel). One would not expect research reactors, if annealing temperatures could be reached, to experience anywhere near these transients, even disregarding the fact that most research reactor graphite is submerged in water (a much better heat transfer medium than graphite).

Walt A. Meyer, Jr.

Acting Reactor Manager University of Missouri Research Reactor Facility

TABLE OF CONTENTS Preface .. . . . . . ... . . . .. . ii Li st of Tables iv Li st of Figures . . . . . . . . . . . . . V SECTION I. Historical Perspective *** 1 II. Conversion between Radiation Dose Units . . . . ... ..... 3 II I. Theory of Energy Storage in Irradiated Graphite . . . . . 7

• IV.

v.

Factors Affecting The Distribution of Stored Energy in Irradiated Graphite **************

Anneal 1ng of Stored Energy ****

9 14 VI. Saturation of Stored Energy Accumulation in Graphite at High Irradiation Exposure *** 19 VI I. Methods of Determining Stored Energy in Irradiated Graphite 21 VII I. Examples of Past Stored Energy Measurement and Maximum Temperatures Reached During Anneal at Various Reactors .. . 29 IX. Graphite Oxidation and Combustion 31 x* The Adiabatic Model: Determining the Maximum Temperature

• Associated with Stored Energy in Graphite References (Footnotes)

Bibliography

• • • • • • • • 38 R-1 through R-5 B-1 through R-4 iii

LIST OF TABLES Table No.

1 Conversions of Mwd/At to nvt . . . . . . . . . . . .. . .

2 Definitions of Units of Irradiation Dose . . . . . . . . . 6 3 Property Changes in CSF Graphite at 3O°C . . . . . . . . . 17 4 Table 12.1 Parameters for the Equation S = s.., (1 - e-kE) * * * * * * * * * * * .......... 26 5 Table 1 - Derived Expressions for Ignition Temperatures of Oxidizing Systems * * * * * * .... . 36

• 6 Relative Combustibility of Powdered Metals in CO2 .... 37 iv

LIST OF FIGURES Figure No.

1 Figure 38 Variation of stored energy across a lattice eel l. * * . . . . . . . . . . . . . . . . . . . . . ..*

• 13 2 Figure 3.4 Sample positions in E4/l ............ 13 3 Fig. 1 Simplified plan view of the UCLA research reactor *** 13 4 Fig. 12.7 Stored energy released from TSGBF graphite irradiated at 30°C. The numbers inside the areas give the stored energy released in the corresponding temperature range. The specific heat integrated over each temperature range is given at the top of the figure figure. All stored energy was removed by the 1800°C anneal * * * * * * * * * * * * * * * * * * * * * * * * * ... 18 5 Fig. 12.6 Stored-energy release curves of CSF graphite irradiated near 30°C * * * * * * * * * * * * * * * * * * . .. 18 6 Figure 12.2 The accumulation of total stored energy (S) at several irradiation temperatures. Equivalent Calder irradiation temperatures are shown in parentheses for each curve (see Sec. 12-2.3) * * * * * * * * * * * * * * * .. 26 7 Figure 12.13 The relation of radiation-induced c-spacing changes and total stored energy (S). Polarization and geometrical corrections have been applied to the c-spacing measurements * * * * * * * * * * * * * * * * * * :- **** .. 27

• 8 9

Fig. 9.6 Expansion of c spacing at different irradiation temperatures. * * * * * * * * * * * * * * * * * * * * * *.

Fig. 21 Irradiation induced changes in crystal lattice parameter * * * * * * ........ .......... ...

27 28 10 Figure 15 Percent growth and stored energy vs.

C-axis changes * * * * * * * * * * * * * * * *

• 28 11 Fig. 1 - Rate of reaction as a function of temperature. . .. 35 V

f' STORED ENERGY IN IRRADIATED GRAPHITE

.I. Historical Perspective E. P. Wigner called attention to the effects on graphite of the displacement of atoms from lattice sites by momentum transfer (the Wigner effect) as early as 1942 and 1943, and published his reference in 1946. 1 The necessity that the internal energy of artificial graphite be increased by reactor-bombardment was first emphasized by Szilard. 2 Its significance was not appreciated on either side of the Atlantic until involuntary releases were observed during operation of the early

• graphite piles. This stored energy has been called the compl ication 11 as well as 11 Wi gner energy 11

• 3 11 Szilard It was during the planned release of Wigner energy that the accident at Windscale occurred on October 10, 1957. Even though it was concluded that there was no direct contribution to the fire from the stored energy release itself, the Windscale fire concentrated attention on this phenomena and produced a large investigation at a number of international laboratories. The bulk of these studies still exist in the form of

• government reports, many of which are not readily available.

A nLm1ber of references were reviewed to compile this information regarding stored energy. However, the book Nuclear Graphite, edited by R. E. Nightingale, provided the most condensed information source and reference. Nightingale described his book as 11 a reference for those concerned with the development and use of nuclear graphite. The objective of this book has been to survey the literature and provide a comp re hens i ve, selected bib l iog raphy of useful references. 11 4 1

• Nuclear Graphite was published in 1962 and comprises the bible of the research in nuclear graphite through the mid to late 1950 1 s.

In 1953, the Ad Hoc Committee on Graphite, which had been formed to oversee Hanford problems, went out of existence by declaring there were no problems. That verdict was a tremendous compliment to Hanford, but in the end sealed the door of on-going broadly based research and development programs for nuclear graphite. 5 Very little research on nuclear graphite has been forthcoming since that time. We have been fortunate at MURR to have contacts with two of

• the early researchers in graphite, Dr. R. L. Carter, University of Missouri, and Walter P. Eatherly, Carbon and Graphite Programs at Oak Ridge National Laboratories. Their assistance on this graphite project as well as earlier ones has been invaluable

• 2

II. Conversion between Radiation Dose Units The most obvious obstacle one encounters in trying to utilize previously obtained data concerning the characteristics of irradiated graphite is converting the dosage units of the studies to some comparable quantity known for a specific reactor's graphite application. The majority of graphite irradiation data that has been obtained is expressed in terms of Mwd/At, Mwd/Ct and Hwd/t (definitions of these terms are in Table 2).

The success with which data based on various dosage units can be

• utilized to make comparisons in the behavior of graphite irradiated in different reactor facilities depends on the convertibility of these dosage units to more generally useful units, such as integrated fast flux. 6 There is some uncertainty in converting from Mwd/At and Mwcl/Ct to integrated fast flux. The nvt conversion is probably no better than+ 20%. 7 A second obstacle to utilizing graphite irradiation data is the uncertainty of correlating dosage data accumulated almost exclusively

• in large graphite moderated, natural uranium, gas-cooled, reactors to the dosage accumulated in small water moderated, graphite reflected, enriched-uranium reactors. The correlation of the damaging radiation dose to graphite "in similar positions in graphite moderated reactors is closely proportional to the flux above 1 Mev *** this is not true when irradiations are conducted in vastly different types of reactors.wB The inexact analogy between the reactors from which the data was compiled and the test and research reactors to which it is applied, has been documented by Ashbaugh, Ostrander and Pearlman in the UCLA stored energy determination 9 and by Gray and Thorne. IO 3

The methodology for making the conversions from reactor specific nvt to Mwd/At (the unit of the abscissa in most irradiation data graphs) in order to utilize graphite irradiation data is further complicated by the disparity between conversion factors from the different studies and whether the conversion is between thennal, fast or total nvt and Mwd/At (see Table 1).

The most conservative approach for this conversion, regardless of whether the conversion is from thermal, fast or total nvt, is to use the factor that will result in the highest Mwd/At after the conversion

• (overestimating the Mwd/At equivalency of a given nvt). For example, if a reactor facility has operated to an integrated flux of 1 x 10 19 nvt (thennal), using conversion factor 15 (Table 1) would be JOOre conservative than using conversion factor 16 (Table 1). Using factor

1. 5 wou l d yield: .

lMwd/At 1 x 10 19 nvt (thennal) x 3.6 x 10 17 nvt (thennal) = 27

• 8 Mwd/At

• Using factor #6 would yi'eld:

• 1 x 10 19 nvt (thennal) x . lMwd/At 3.9 x 10 17 nvt (thennal) = 25 *6 Mwd/At

• 4

Table 1 Conversions of Mwd/At to nvt

1. MURR Design Data 1 Mwd/At = 1 x 10 17 nvt (E > 1 Mev)

"Design Analysis of the Graphite Elements,* p. 3 (suggested by National Carbon Co.)

2. Davidson, Woodruff, Yoshikawa 1 Mwd/At: .87 x 10 17 nvt (E > .18 Mev)

"High Temperature Radiation Induced Contraction in Graphite, 11 Proceedings of Fourth Conference on Carbon, p. 600 Nightingale, Yoshikawa, Woodruff 1 Mwd/At = 1.3 x 10 17 nvt (E > .18 Mev)

Nuclear Graphite, p. 286

4. Bridge, Kelly, Gray 180 Mwd/At = 1 x 10 20 nvt 11 Stored Energy and Dimensional Changes in Reactor Graphite,u Proceedings of Fifth Conference on Carbon, p. 292

5. Powell, Meyer, Bourdeau 1 Mwd/At = 3.6 x 10 17 nvt (thermal)

"Control of Radiation Effects in A Graphite Reactor Structure,~

Proceedings of Second United Nations Conference on Peaceful Uses of Atomic Energy, p. 283 Nightingale, Davidson, Snyder 1 Mwd/At = 6.5 x 10 17 nvt (tota~)

"Damage to Graphite Irradiated Up to 1000°C," 1 Mwd/At: 3.0 x 10 17 nvt (thermal)

Proceedings of Second United Nations Conference on Peaceful Uses of Atomic Energy, p. 295

7. Woods, Bupp, Fletcher 1 Mwd/At = 6.46 x 10 17 nvt (total) uirradiation Damage to Artificial Graphite, 11 Proceedings Of the International Conference on the Peaceful Uses of Atomic Energy, pp. 455 5

Table 2 Units of Irradiation Dose

1. Megawatt-day per adjacent ton (Mwd/At)

_ the neutron irradiation received by the sample during the period required for the uranium fuel adjacent to the sample to generate one megawatt-day of fission energy* per 2000 pound/ton.

H. Bridge, et al, "Stored Energy and Dimensional Changes in Reactor Graphite," Proceedings of the Fifth Conference on Carbon, Vol. I, p. 292, 1962 Megawatt-day per centrol ton (Mwd/Ct)

_ considered equivalent to Mwd/At, however it is determined from the average power generated in the central region of the reactor rather than a direct measurement of the power in the fuel invnediately surrounding the irradiation experiment; R. E. Nightingale, Nuclear Graphite, p. 229, 1962

3. Megawatt-day per ton (Mwd/T)

_ irradiation exposure received by the sample during the period re-quired for the (2000 pound) ton of uranium metal in the i11111ediate vicinity of the sample to generate one megawatt-day of fission heat.

W. K. Woods, et al, 11 Irradiation Damage to Artificial Graphite,"

Proceedings of the International Conference on Peaceful-Uses of Atomic Energy, vol. 7, p. 456, 1955

4. Integrated flux [represented by nvt]

_ product of the average flux (neutrons/cm2-sec) and the exposure time (sec) a) nvt (fast) = product of (average) fast flux and the exposure time b) nvt (thermal) = product of (average) thermal flux and the exposure time Glasstone, and Sesonske, Nuclear Reactor Engineering, p. 424 6

III. Theory of Energy Storage in Irradiated Graphite The theory of radiation induced effects in graphite is thoroughly developed in Nightingale's Nuclear Graphite, Chapter 7. Graphites do not experience damage from exposure to gamma radiation but do exhibit property changes as a result of neutron bombardment. Neutron bombardment of the graphite results in the displacement of carbon atoms from the crystal lattice into interstitial positions. 11 The permanent effect of the dis-placements is generally manifested by a system of Frenkel-type defects (a system consisting of vacancies and interstitial atoms) generated in the

• crystalline lattice of the graphite.

The estimated energy of formation of vacancies 1£.fV within the plane of the graphite and that of the interstitial atoms 1£.fi between planes may be assumed as follows:

1£.fV "'6 ev; fff i "'5 ev These remarkably high values of the free energy of formation (i.e. in the case of metals the 1£.f V_"' 1 ev) point to a high heat content assoc1 ated with the point defects in graphite. Increasing the neutron radiation dose beyond a certain level (~ 10 18 nvt for 30°C irradiation temperature) causes the concentration of the defects to such an extent that they will exhibit the effects of mutual interaction, with the formation of complexes

("clusters" of interstitial atoms, 11 clusters" of vacancies, C2 complexes).

The phenom*ena of assoc1 at ion of simple defects tends to decrease the heat content of the system and hence the total stored energy. 12 Not all the damage is in the form of atom displacements. Some damage takes the form of distorted or misshapen cell configurations (edge dis-locations, screw dislocations).

7

The results of the neutron induced damage in graphite are changes in many mechanical, physical and chemical properties of the graphite. The extent of the radiation damage, and hence the extent of the property changes, depends on the neutron exposure (neutron spectrum as well as length of irradiation), the temperature of irradiation and is also greatly affected by the raw materials and fabrication history. 13 These three factors, the grade of graphite, the temperature distri-bution in the graphite and the flux distribution in the graphite greatly affect the distribution of stored energy within a graphite moderator or

• reflector. These factors make it difficult to predict the degree of damage resulting from the irradiation of graphite. The effects do not lend themselves to generalizations that are consistent across the various types of graphite and irradiation conditions.

Stored energy is one of the property changes induced by irradiation damage to graphite. Stored energy at a given temperature is defined as the difference in the heat content (ll-l)T between an irradiated crystal and the same unirradiated cryst~l in thermodynamic equilibrium at the same temperature. 14 wThis stored energy is basically an increase in the

• enthalpy of the material and can be measured through conventional heat of combustion measurements from which total stored energy is obtained or as an apparent decrease in the specific heat from which the stored energy release rate may be obtained as a function of annealing temperature." 15 8

IV. factors Affecting The Distribution of Stored Energy in Irradiated Graphite The existence of damage gradients in observations of irradiated graphite is well documented. These gradients occur due to the existence of pre-existing defects in graphite before irradiation; the energy spectra of the flux in the irradiated graphite, and the temperature differentials within the graphite. 16 These damage gradients are difficult to predict since the flux and temperature during irradiation affect both the creation of the defects as well as the removal {annealing) of the effects. Studies have shown that the most severe rate of stored energy accumulation occurs at points of high neutron flux and low graphite temperature within a moderator or reflector. These points of high energy accumulation are, however, not at all indicative of the bulk stored energy in a graphite structure.

The results of the second British Experimental Pile Zero (BEPO)

Wigner Energy release showed that the mean stored energy over a iattice cell was considerably less than the point of maximum stored energy. The samples taken were from ~raphite in an experimental hole which runs at

• right angles to two fuel channels adjacent to the BEPO core. The varia-tion of stored energy over the sampled section {an 8 inch x 8 inch section at a plane 9 feet from the unload face, [see Figures 1 and 2] showed that the stored energy near the fuel channel {the points of highest flux and lowest temperature) proved to be greater than the mean .by about a third.

Thus considerable errors could be made if samples from a nontypical part of the cells were used to determine the total stored energy if the results were not corrected.

9

It was presumed that the real behavior of the release of nonuniformly distributed stored energy nlie between that of a block containing uni-formly the average stored energy and the other extreme of supposing each element of the block to rise locally to the maximum temperature character-istic of the local stored energy. The relaxation time for the nonuniform temperature to equalize itself over the cell was about ten minutes ***

therefore apart from possible short-lived local transients, the maximum temperature reached would correspond to an average value of stored energy." 17

• The nonuniform distribution of stored energy in reflector graphite was documented by measurements at the University of California-Los Angeles reactor. "The highest value found was 33.2 cal/g, next to the fuel boxes.

At the island center, it was 19.2 cal/g. The stored energy was small, and further, was confined to the graphite volume adjacent to the fuel boxes, which is a small fraction of the total volume of graphite in the reactor* (see Figure 3). is Variation of The Grades of Graphite

• The grade of nuclear graphites is determined by the constituent materials and the method of manufacture. The production methods and the effect of variations in these methods on the properties of the end product are discussed in several references. 22 , 23 Artificial graphite is not a single substance, but comes in many varieties with various degrees of anisotropy, various densities, various crystallite sizes, various impu-rities, and various amounts of disorder in their initial structure. 24 Graphite with large crystallites have been found to store energy more rapidly than small micro-crystals. 25 10

Variation of The Neutron Spectrum in Graphite "The problem of calculating the rate of accumulation of damage in a given volume of graphite in a nuclear reactor involves an additional con-sideration; the energy distribution of the neutrons in that volume of graphite **** Within reactors the neutron spectrum varies from point to point depending on such factors as distance to the fuel, control rod positions, fuel element. configuration and type and amount of coolant." 26 The Effect of Thermal Conductivity Change on Stored Energy in Graphite

• The distribution of temperature of a graphite moderator as a func-tion of distance below the cooled surface depends upon the density of energy dissipated by penetrating radiation and upon the thermal conduc-tivity of the graphite. The , latter is, in turn, a function of the graphite temperature and of the extent to which radiation-induced changes have affected the graphite structure.u 2 7 An application of the solid state theory of thermal conductivity of graphite permits deduction of the irradiation-damage gradients in modera-tor structures. The surface layers of heavily damaged graphite insure

• radiation annealing of deeper lying graphite. 2 8 Thermal conductivity is an important factor in determining maximum temperature experienced by graphite when it is used as a moderator or reflector. The central temperature of the graphite will be increased due to a large thermal gradient created by the greater decrease in the surface thermal conductivity relative to the center during irradiation. 29 Empirical data have indicated that the effects of irradiation on thermal conductivity are very marked with thermal conductivity being re-duced by factors of 40 30 , 31 to 50 32 , 33 after irradiation to an integrated neutron flux of 10 19 nvt.

11

One thermal conductivity study observed *The appearance of a surface layer of very low thermal conductivity on the cooled surface of a graphite moderator exposed to a nuclear reactor. This surface layer was shown to provide a thermal barrier which insures that the internal temperature of the bulk of the graphite moderator is high enough to permit irradiation annealing * * *

• For a high flux reactor the existence of this high bulk temperature prevents extensive accumulation of stored energy.* 34 This effect of surface insulation of the interior of the graphite block was documented during the disassembly of a Soviet isotope reactor

• that used graphite as a moderator and reflector and water as coolant.

11 The alteration of thermal conductivity diminished in the 8 cm graphite layer from the cell axis to its periphery **** As a result, the ac-cumulation of radiation damage in the central part of the lattice, where the temperature was comparatively high, was nuch less than in the periphery." 35 12

I .,

1

• r / ~ / / / I/ I / / / ,/ I /

"I>'*\ / I / I I I / .. I * / I ' "

--- --~

I/ /

• • , r-i

... I ,

• I

/

' /

/ fi1

---

1

"' ~ IEaperu,._,11,1 hol* £4/1 8D t

e""

00 011 I

I. .

To-

!I

--~---.__

,;71\

, ,'r::: i E

i;,

l 'II I

A:::

.J!.., E! _,,l!.76 E-I~

1l!]

1

'\Q)

)

~

iI I

0

-~ I I

I

. / /

I

1

/

I C: I > I C '

~

C1t1'9"1C* ....,_

F,;urr 38 V0ri0t10n of 1t0rrC enrr&'f ocrou a Iott.ct' ice1l

..., . C &

0,11onc:r Inches F,gur* J.I Somplo po>11i0n1 ,n E-' '1 Figure 1 19 Figure 2 2 o GRAPH,1i.

-~--,.cub I

H r,t ___ GRAPH.if r 111 1 , tt l_J c,

I GflAl-'llll l E

I . --*7 l

'\ - - C.

I - :I I

• I

\.\

L_i _ __J r-:--

I I

I l

~-- - - ~ ~ / '---v,,-

EXHRI\.A~ fl:0R1H CE:,~ EP IS, A* JD  :...uu:H E XlE RNAL REH ECTOR CGi-:[ I *lf.t R[i'LlClOR 21 f,,~

?. - 0 6 35 cm f1i: I S1mrl1ll,*J rl:11, ,-.--- *'I *h \*Cl\ ,,_ ..* ,, ,.,, .. ,,

Figure 321 13

V. Annealing of Stored Energy

• As soon as stored energy in graphite was observed, it was recognized that measurement of the total stored energy alone would be inadequate for the prediction of possible temperature excursions in the graphite moder-ator. It is necessary to know the amount of stored energy released as a function of the annealing temperature." 36

• stored energy in neutron bombarded graphite is released over a wide range of temperatures *** it is therefore reasonable to conclude that there are no unique stored-energy centers, but rather there nust be a

• continuum of disturbance centers varying in quality and, in particular, having different activation energies for thermal annealing.w 37 It is interesting to note that the average stored energy releasable in the annealing temperature range of 25° to 800°C is about 280 cal/gram for samples irradiated to greater than 2000 Mwd/At. Figure 4 shows the the stored energy released in corresponding annealing temperature ranges for various graphite irradiation histories. Even though the sample with 5000 Mwd/At exposure at 30°C has a total stored energy of 620 cal/gram

• only 275 cal/gram is releasable in annealing temperatures up to 800°C. 38 Similar findings were reported by Nightingale, Davidson and Snyder in their article "Damage to Graphite Irradiated up to 1000°CH. Table 3 shows that annealing temperatures up to 1000°C cause the release of about half of the total stored energy after irradiated exposures of 3000 Mwd/At.

This again points out the importance of understanding the stored energy releasable up to some specified temperature rather than the total stored energy.

14

11 Models proposed to explain the phenomenon of thermal annealing are complex and not completely successful. Many models include mechanisms involving high-order reaction rates, continuous spectra of activation energies, or diffusion-limiting or chain-inhibiting effects. For graphite irradiated at room temperature, the initial annealing takes place below 400°C and is accomplished by diffusion of interstitial atoms to vacancies and results in vacancy-interstitial annihilation. Between 400 and 1000°C, recovery is accomplished by the formation and breaking up of groups of interstitial atoms and above 1000°C the damage recovers by migration of vacancies. Additional investigations revealed that only a small fraction of the displaced atoms migrated to surfaces during annealing at tempera-tures up to 400°C and that 80 percent of the displaced atoms combined with vacancies. 1139 The most important feature of the stored energy release curves versus annealing temperature is the peak at about 200°C (see Figure 5). This peak increases in intensity with exposures up to about 4.2 x 10 20 nvt. At high-er exposures this release peak is diminished and increasing proportions of the stored energy are released only at higher temperatures. 40

• Superimposed on the release curves of Figure 5 is a line representing the specific heat of unirradiated graphite. When the rate of stored energy release exceeds the specific heat of the graphite, the graphite then exhib-its an apparent negative specific heat and it can spontaneously .increase in temperature. 41 It is observed consistently in both calorimetric and combustion experiments that until the annealing temperature exceeds the irradiation temperature by some definite amount, stored energy will not be released. 42 15

The threshold temperature increase required before any release of stored energy is encountered varies in the literature reviewed from 50°C 43 , 44 to 100°C 45 above irradiation temperature.

Since the minimum temperature for a rapid substantial release is!

some 50°C above the temperature of irradiation, this effect is important in practice only for large transients. 46 An annealing phenomenon which is of great importance, irradiation annealing (variously termed *nuclear annealingn, 11 pile annealing 11 ,

11 neutronic annealing*) has been observed in graphite. This phenomenon

• manifests itself as an alleviation of radiation damage beyond that attrib-utable to thermal annealing alone. The probable mechanism is radiation-induced breakup of the interlamellar complexes at a temperature high enough to promote recombination of splinter atoms with lattice vacancies. 47 There are several excellent references on the theory and modeling of graphite annealing, including Nightingale's Nuclear Graphite, Chapter 13, Report IAEC-tr-4545, 11 The Wigner Release in Graphite-Moderated Reactors, 11 by C. Dalmasso and G. F. Nardelli, and the "Theory of Annealing Kinetics Applied to The Release of Stored Energy from Irradiated Graphite in Air-Cooled Reactors,* by A.H. Cottrell, et al.

16

Table 1. Property Changes in CSF Graphit~ at 30°(

/-_;i f'OllHt [ jf D ~T.

"\ 1'11,... rr,.,,.,. lh'rr11:011rn-II J1>(>,i 2001, Jnf}(, .J(HJ(J ,u01., Ol}(JIJ c,, Cn stal laH*r spacmg I A I 6.71  ; )CJ  ; 48 i 6:- 'i 'i3 'i. 78 7 78 Lr \ppareni ~ ~1all11e size 50(1 2-;:; 12, i5 50 25 25 in Cn d1ren1on (Al

},: Thermal rondurtivn, at Perpend1rnlar 0 26 0 0082 0 006CJ 0 0063 0 0063 0 0062 25°C teal, cm f>eC °C, Parallel o ~o Cl 0104 0 00i8 0 00i3 0 0071 0 OOiO SE Total storfil energ, 0 28, .iJ, 550 610 630 (cal*t:*

Total storfil energ, re-mammg aft1:r a 3-hr anr!f'al at J000°C 0

L. ~

60 160 z.._-,~

250 300

~l'D 330 3,r-o

~-..,,1J.~ "4-11.Lu U..t.VSl>flll.

u.f' -;-t, tc~*~

Table 347 ~ ~ vPNJ (,

1'v-rf .

17

04 t-- 2BO jcp di 03 0.2 5000 Mwd/A!

0.1 0.0 04

~

......,. 0.3

'c 0.2 u

2000

...- Mwd/At

(/)

0.1 c:,

00 04 0.3 02 600 01 Mwd/At 00 0 400 800 1200 1600 2000 Anneal1n'i1 T11mp1roture, *c Frn 12 7 Stored energy released from TSGBF graphite irradlllted at 30°C. The numbers mSJde the ar= give the stored energy released rn the correspondrng tempera-ture ran11:e The specific heat rntegrated over each temperature range LS given a.t the top of the figure All stored energy was removed by the 1soo°C anneal 10 F1gure 448 EXPOSURES IN 1,1,.d/AI 0.7 5700 5000 3540 0.6 1"190 1075 L) 0.5 760 560

....."' 100

.; 0"1 u

0.3 Cl) 0.2 01 0

100 200 500 F1G 12 6 f.tored--f'nergy releasp c-un*ec= of CSF fraphite irradiated near 30°C.

. 49 Figure 5 18

VI. Saturation of Stored Ener~ Accumulation in Graphite at High lrradation Exposure The most interesting point about the energy release data is the change in the shape of the rate of energy release curve with irradiation at doses above 1.5 x 10 21 nvt. At irradiations greater than 1.5 x 10 21 nvt, the peak of the release curve is smeared out, eventually becoming practically invisible. nThis change can be described as a coagulation during bombardment of single interstitials into large complexes containing at least two interstitial atoms * ... As would be expected, it is ob-served after heavier bombardments *** that a progressively smaller

• fraction of the damage anneals near 200°C and a larger fraction requires much higher temperatures for annealing." 50 The evidence of saturation of stored energy accumulation is presented .

in many graphite studies. The following are some examples:

"It is clear that as irradiations are prolonged an increasing fraction of the total stored energy resides in disturbances having higher activation energies for thermal annealing. At the same time the total stored energy is increasing although at a constantly decreasing rate." 51 "After an initial increase of stored energy almost proportional to exposure, it is generated at a decreasing rate as the exposure increases because the single interstitial atoms begin to form complex groups." 52 "Essentially none of the incremental amounts of stored energy accumulated after exposures greater than 13 x 10 20 neutrons/cm 2 can be released by annealing at a temperature of 1000°c." 53 nThe same irradiation dose (greater than saturation as compared to before) displaces the same number of carbon atoms in each case but the distor-tion and defects already present in the crystal modify the way in which these atoms are distributed and hence the process by which they are annealed.u 54 19

At a certain dose level, which 1s different for different irradiation temperatures, the disorder produced could be of such a nature that it will not be susceptable to be increased any further; it 1s then said that a "saturation level" has been reached.

The saturation level in graphite at 30°C is equal to 6500 Mwd/T for the adjacent uranium.nSG 20

VII. Methods of Determinin~ Stored Energy in Irradiated raphite Experimental Methods A. Measurement of Total Stored Energy nThe customary method of measuring total stored energy is to determine the room-temperature heat of combustion. If &11 and &lu are, respectively the heats of combustion of an irradiated and similar unirradiated sample, then S = &11 - &lu*

Conventional methods of bomb calorimetry are employed in the measurement of the heat of combustion of graphite." 57 B. Measurement of Stored-Energy Release Curves An experiment on the release of stored energy requires the mea-surement of the amount of stored energy released and the temperature of the sample as functions of time. The relation between these quan-tities depends on the state of the graphite and on the conditions of the experiment. It is convenient to choose conditions which approxi-mate to a simple temperature-time relation or to a special condition of heat transfer. 5 8

• The three most common methods used to generate energy release curves are the isothermal, linear rise and adiabatic.

for each of these experiments are as follows:

The conditions a) Isothermal with T = constant b) Linear rise with T = T0 + at, where T0 and a are constants; and dT dS 59 c) Adiabatic rise with S dt = - cit

• 21

These experimental methods are documented in section 12-3 of Nightingale's Nuclear Graphite. Other methods are located in the following references: 11 Energy Stored by Irradiated Graphite:Variation of the Energy as a Function of Successive Annealing" by M. Quetier and J. Rappeneau; 11 BEPO Wigner Energy Release 11 , by Dickson, et al; and 11 Stored Energy in Neutron-Bombarded Graphite", by T. J. Neubert and R.R. Lees.

C. Retrieving Samples of Irradiated Graphite for Experimental Analysis The Dickso.n, et al paper "BEPO Wigner Energy Release" analyzed

• the effect on irradiated graphite of coring samples. It was suggested 1 that the heat produced in coring might release some of the stored energy in the samples and invalidate the sample. Tests showed that the maximum rise observed in cutting the irradiated graphite was 20°C.

The temperature changes observed were too small and of too short a duration, to have any measurable effect on stored energy. 50 Semi-Empirical Calculations John J. Newgard devised a semi-empirical model to correlate

• empirical data over a wide range of neutron doses. This model is described in his article "Single Semi-empirical Model for Neutron Induced Stored Energy in Graphite* and in Chapter 12 of Nuclear Graphite. Newgard pointed out that exposure data could be fitted to the semi-empirical equations of the form:

S = S (1 - e-kE) (VII-1) or S = S tanh k'E (VII-2) 22

where k and k1 are constants and E is the neutron exposure. 61 See Table 4 for parameters for equation VII-1. Figure 6 shows the curves to which the equation can be fitted.

Another useful equation found in Chapter 12 of Nuclear Graphite, s( T) = a e- TI a ,

11 relates the total energy stored [S(T)] at a given exposure to the irradiation temperature (T). The stored energy at 0°C (a) is a constant for any given exposure. All the data of Figure 6 are satis-factorily reproduced with a= 71.2°C. 1164 A different a must be used for the 30°, 50°C curves and the 155°, 205°, and 255°C curves. An a of approximately 415 correlates the low temperature range exposures and an a of 892 correlates the higher temperature range exposures

(@1000 Mwd/At). The value of this equation is its ability to allow approximations of stored energy relative to the curves in Figure n but at temperatures other than those of Figure 6.

The Newgard equation was utilized by Pearlman (1983) to calcu-late the stored energy in the UCLA graphite reflector. The calculated values correlated well with actual measured samples. 65

• The methodology used for calculating the stored energy in the Generic Safety Analysis of Argonaut Reactors is in PNL-3691 (April 1981) and is outlined in the Ashbaugh, Ostrander and Pearlman refer-ence.

B. I. Spinrad of Iowa State University has developed a simplified bounding calculation for stored energy in graphite (see Appendix listing).

23

Correlating Stored Ener~ to Radiation Induced Changes in C-Spacng in Graphite I

There has been considerable interest in estimating stored energy from the measurements of other property changes that are both quicker and nondestructive. One of the most successful attempts has employed correlations of stored energy with changes in _£-spacing. 66 The correlation of stored energy and c-spacing is shown in Figure 7. ult should be noted that the _£-spacings in Figure 7 have been taken from Figure 8 to which polarization and geometrical cor-rections have been applied. The slopes of the line in Figure 7 are

• therefore steeper than would be obtained from c measurements to which these corrections have not been applied. The stored energy versus

.£_-spacing curves are linear, and the slopes appear to increase with irradiation temperature above approximately 150°C. The close cor~

rrelation between stored energy and c-spacing indicates that the radiation effects responsible for stored energy and lattice expansion are similar. Since c-spacing can be measured from small powdered samples, this correlation is useful in estimating stored-energy

• gradients throughout a moderator structure. 1167 At Brookhaven National Laboratory, diffraction power samples were made from scrapings taken from growth samples, dust scraped from reactor channels and scrapings from samples cored directly from the reactor graphite. The c-axis values from each of these samples was similar. 68 X-Ray Diffraction Measurements The presence of interstitials causes an increase of all spacings in isotropic media; in graphite only the _£-spacing is appreciably 24

altered. The _£-spacing increases linearly up to very heavy bombard-ments (greater than 1300 Mwd).69 Reintegration caused by thermal annealing decreases the c-spacing because it removes interstitial atoms. This process is practically complete near 600°C, since the _£-spacing anneals almost completely at this temperature. 70 , 71 There are several noteworthy references regarding the correlation of stored energy and _£-spacing: *stored Energy and Dimensional Changes in Reactor Graphite" by Bridge, Kelly and Gray; "Interpreta-

• tion of Damage to Graphite,* by Hennig and Hove; and Chapters 5, 9 and 12 of Nuclear Graphite. There is also an ASTM C558-69, Standard Method for Measurement of Lattice Spacing of Nuclear Graphite

• 25

Table 12.1- PARAMETERS FOR THE EQUATION S* s_ (1 - e*lrE)

Irradiation s_, cal/g It, per 1000 Mwd/ At temperature, *c 30 685 0.526 1 150 600 0.242 2.2 200 375 0.169 3.1 250 200 0.151 3.5 300 50 0.393 1.3 Table 262

• 700

- - - Referenco e

- - Referencn 9 on<:~ con .. ,to~ _ _ .3o*c

- - Referenct

• R1ferenc1 10, 50 *c ID, 150°C 600 from M*d /Ale ouum*~--- 0 Fteferenct 10, 2oo*c 1M"d/A1 *06 M~A1 1

/

/

/

500 /

,, /

/

/

.000 /

"' I

~

I "u I 30C* I

" / 50"C I'I 200 I/

/.

100 I*

2!>5°C 1223 l ii 305°C (26!?)

C 0 2 3 6 7 6 9 10 Fi<, 12.2 Tl1e arcumulation of total st.ored energy (S I at i;everal 1rraruat10n l<'mprrnrnrn Equn*alent Calder irradrntion temperatures are shown in p&rentheBPS for rarl, curn* (F.f-f Set IZ-2.3).

F1gure 663 26

IRRADIATION TEMPERATURE. "C e 30 CSF 600 ~ 1~0}

0 200 PGA e 2~0 500 400 0

300

• 7.1

c. A 7.2 7.3 7.4 FIG. 12.13 The relation of rad1auon-induced c-spacing changes and total stored energy (S) Polaruation and geometrical correct1on.s have been apphed to the c-spac1011 measurements

. 772 F1gure 0

7.1 70 Cl 6.9 -

Ge 400-500°C 6.7 0 ~00 1000 1~00 2000 21 F1G 9.6 Expansion of c spacing at difieren: irra.cliatJOIJ t.emper11tures * "

F1gure 873 27

• ,o* r I

I i

I ,oL I

r

~ .tf. *** :

Iw "*

a

~

,L E

IQt.r1-.i..1*t . . . ADl,AT1C)llrl

,0

~ I ~ *

• I I 1'0~

• ,0 d'"'" ** I

.!so,..~

g.

10 1

FIG :! I [rrad,atwn ,m.luce<l che.n1,1t>~ m c~ ~t,.l le.tt1c-e parameter.

Figure 974

• !>O

• ol I

I I

.!£

.32

,.28 c 30 Cl u GRCM'TH

~

...~.., 20 I

It

..,C.

a:

!?

.r,

,o I

7X

.... ,.~7: 68:

C

• A* 15 *'I.;_

f,gure 15 Per.en! g*owth one! stored en**gy 'WI C-axi* changes Figure l 0 75 28

VII I. Exam~es of Past Stored Energy.Measurements ana x1mum Tern eratures Reacnea ur,n nnea a ar ors Max. Temperature Reactor TrnR Irradiation History Peak Stored During Anneal or TArrneal Erierg,:t . S.C-Releasea*

a) UCLA 50-80°C 2.7xl0 19 nvt (th) 33.2 cal/g N/A b) Hanford 30°c 575 Mwd/ At 210 cal/g 160 cal/g (800°C)

Test 30°c 2023 Mwd/ At 425 cal/g 280 cal/g (800°C)

Holes 30°c 4965 Mwd/ At 630 cal/g 275 cal/g (800°C) c) X-10 145°c 16 Years 35 cal/g 236°C ( 100° )

( Oak (76°C rise)

Ridge Graphite Reactor) d) BEP0 110°c 325°c (120°c)

(1st Anneal) e) BEPO 110°c 95 cal/g 316°C O20°c)

(2nd Anneal) f) BGRR 50°C 1. 7xlo 20 nvt 50 cal/g 300°C ( 135°c)

Brook-

• haven Graphite Research Reactor g) Windscale 253°c 4.8 X 1020 49 cal/g 230°c 5.7 X 1020 55 cal/g 192°c 5.3 X 1020 61 cal/g 29

REFERENCES FOR SECTION VIII

a. C E. Ashbaugh, N. C. Ostrander and H. Pearlman, "Graphite Stored Energy in the UCLA Research Reactor,a Transactions of The American Nuclear Society, Vol. 52, p. 372, 1986.

b. R. E. Nightingale, J.M. Davidson and W. A. Snyder, "Damage to Graphite Irradiated to 1000°C," Proceedings of Second United Nations International Conference on the Peaceful Uses of Atomic Energy, Vol. 7, pp. 299 c.) R. E. Nightingale, Nuclear Graphite, Academic Press, New York and London, d.)- 1962, Chapter 17.

e.)

f. R. W. Powell, R. A. Meyer and R. G. Bourdeau, "Control of Radiation Effects

• g.

in a Grap;hite Reactor Structure," Proceedings of the Second United Nations International Conference on the Peaceful Uses of Atomic Energy," Vol. 7, pp. 282-294, 1958.

Bell, et al, Windscale (Stored Energy in Pile Irradiated Graphite) p. 110 30

IX. Graphite Oxidation and Combustion "Graphite is one of the most inert of materials with respect to chemical reactions with other elements or compounds. It is subject to only three types of attacks: oxidation, formation of lamellar compounds and reaction with and solution in carbide-forming metals in certain high temperature ranges * * *

• Graphite nrust be protected from oxygen at high temperatures other-wise it will burn to CD2 or CO. It is interesting to note that graphite is actually less reactive to oxygen at low temperatures than many

• metals * ...

If one defines a wthreshold oxidation temperaturew as that at which a sample loses 1% of its weight in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, pure graphite has a threshold oxidation temperature of 520°C to 560°C depending on coke source and processing variables ****

In the temperature range 200°C to 250°C above the threshold oxidation temperature, th~ oxidation is still penetrative in character, such that the weight loss is greater than the volume loss. At still higher temperatures, the oxidation is diffusion-controlled, with virtually all oxidation taking place at the outside surface ****

Similar oxidation effects on graphite are produced by steam and CO2 atmospheres. However, the threshold oxidation temperature is higher for these gases. Commercial graphite will have a threshold oxidation tempera-ture of about 700°C in steam and 900°C in co 2

• 1176 In order to discuss the research data regarding combustion of graphite it is worthwhile reviewing the definitions and relationships of the terms oxidation, combustion and ignition. Many of the studies do not 31

distinguish oxidation from combustion; the tenns are not necessarily interchangeable, without qualifying assumptions.

Oxidation is defined as the combining of oxygen with various elements and compounds, when oxidation becomes rapid it is known as combustion. 77 Combustion is defined as the chemical union of oxygen with other elements and compounds at a rapid rate. There are three stages of combustion:

1) The absorption of heat to raise a materials temperature to the point of ignition.

2) The distillation and burning of the volatile gases *

• 3) The combustion of the fixed carbon.

A theoretical amount of air is required for the complete combustion of a material. In practice, there is difficulty in getting air into con-tact with all particles of a material. It is impossible to obtain perfect combustion with the theoretical amount of air, and an excess sometimes equal to double the theoretical amount is required. 78 The ignition temperature of a material is defined as the degree of temperature at which a substance will combine with oxygen at a rate suf-ficiently rapid to produce a flame. 79

• 11 When two materi a1s react chemi ca 11 y (in this case we are interested in graphite and oxygen) there are several curves that can describe the course of events. We have the rate of reaction as a function of tempera-ture of the material undergoing reaction. An average curve may be similar to Figure 11 * *

• ordinary chemical reaction_s normally obey the Arrhenius equation, which means that the reaction rate varies exponential-ly with the temperature. This dependence of reaction rate on temperature is shown in zone 1 of Figure 11. As the temperature increases, somewhere 32

in the neighborhood of Tig (the ignition temperature), the reaction rate increases so rapidly that within a narrow range one cannot measure the rate at a specific temperature; this region is zone 2. 11 80 It seems clear from these definitions that combustion and oxidation are synonymous only after the reaction temperature reaches ignition temp-perature and since the temperature of ignition corresponds to the onset of combustion, the terms combustion temperature and ignition temperature might be considered equivalent.

The most exhaustive study of graphite combustion was found in the Powell, Meyer, Bourdeau reference performed at Brookhaven National 11 Laboratory. Tests were conducted to determine the exact conditions for self-sustained combustion. Sustained combustion was defined as that temperature which remained constant or increased without external heat input ...

The minimum temperature for self-sustained combustion of unirradiated graphite was found between 830 and 930°C. It was determined that a fire could be put out either by suffocation or by chilling the graphite * * *

• The irradiated pieces of graphite (exposure of 5.6 x 10 20 nvt) were

• heated to different temperature levels between 400 - 900°C. At each temperature, the flow rate was changed and evidence for self-sustained combustion carefully sought. Up to and including 700°C there was no such evidence. However, at 790°C combustion took place and the graphite temperature increased. With both irradiated and unirradiated graphite, once self-sustained combustion commenced the temperature increased at least as high as 1240°C without any evidence of leveling off. 1181 33

Other references differed in the combustion temperatures cited. In graphite burning studies perfonned by the AEC in 1961 "to determine con-ditions leading to runaway in an air-cooled graphite moderated reactor

• • • results have shown that even under unfavorable conditions the temperature range in which graphite becomes troublesome is greater than 600°C." 82 Another study revealed that ugraphite ignited generally in the temperature range 700 - 720°C. This is what we think is a CO-oxygen ignition.* 83

• One study looked at the reaction of steam and moderator graphite.

"The graphite crumbled after exposure of 177 hours0.00205 days <br />0.0492 hours <br />2.926587e-4 weeks <br />6.73485e-5 months <br /> in the range 840° to 860°C. This is lll.lCh higher than the range of temperature that might be anticipated in an accident transient." 84 Another study agreed with the Brookhaven National Laboratory refer-ence that "sustained oxidation can be detected with CO2 monitors and controlled by smothering."85 Dr. M. Robkin, Director of Nuclear Engineering Laboratories,

• University of Washington, has provided additional information about graphite combustion in his *Response to the Petition from CBGu (attached in appendix).

34

/

f ZONE 3 t 'I' I

z I

~

I I

C I ZONE 2

...a::

0 I

I I

I C

a::

r,,

TE'I.IPEl:;ATURE OF SURFACE -

Fig. 1 - Rate of reaction as a function or temperature.

Figure 11 86 35

T ABJ.E 1-DERI\'ED EXPRESSIO~S FOR 1G!'!!TION TE~1J>ER.-\Tt:HES OF OXlDIZING SYSTE)1S System Equation Linear T,ll = 6AQmrP"'

4.575 log JUT)pd Parabolic T,g = E 3QA'*rmP"-' 2 9.14 log J(AT)dpt'*

• Cubu: T 1g T

=-----,----

=.

13.72 log E

2.A'>rQmp"l:/3 J(.lT)t;,pd E

Logarithmic lg

• 6bmQArP'-

. -4.575 log J(l)T)pdCl*tll>

Nomenclature:

T,g = apl)ar~nt ignition temperature E = Arrhenius actl\*ation energy A = preexponenlial factor in Arrhenius rate expression b = a constant, about 0.4 or 0.5 usually m = mass of !!>ample

• Q = heat of reaction per p-am of oxide formed J "'heat-tr;msfer coefficient T =- temperature differential p = density ?f particles t = time' r = len~h of path of heal flow d "'mean particle diameter N = a constant 87 Table 5 36

Table 588 Relative Combustibility of Powdered Metals in CO2 T ignition Magnesium 749°c Boron 920°C Aluminum 420°C Titanium 670°C Chromium 870°C Manganese 696°C 37

x. The Adiabatic Model: Determiniat the Maximum Temperature Assoc1atedth Stored Energy in Graphite Regardless of whether the stored energy in the reflector graphite is experimentally determined or calculated from empirical data correlation, the next step is a determination of the peak temperature transient associ-ated with the stored energy. It is important to show that this tempera-ture is considerably lower than any temperature that may cause reactor fuel problems.

The extreme conservatism of the adiabatic model will be reviewed, as

• well as whether or not the use of the peak stored energy or average stored energy in the peak temperature determination is most appropriate.

The model for the determination of the peak temperature associated with stored energy can be described as both a closed and isolated system.

(In a closed system matter does not cross the boundaries, in an isolated system no external work is done and there is no flow of heat into the system.)

Although properties which depend on the mass such as volume, surface

• area and all kinds of energy are known as extensive properties, it is apparent that extensive properties evaluated on a unit mass basis (specific values such as energy per unit mass) become intensive proper-ties (i.e. independent of mass). 89 This model will evaluate stored energy in graphite as an intensive property (based on unit mass).

Stored energy has been defined as the difference in heat content (ll-1) between an irradiated crystal and an unirradiated crystal (or the increase in enthalpy due to neutron exposure). It is this (&I) that is released upon annealing and can be associated with a unit mass temperature 38

rise. 11 In any exothermic reaction carried out directly at constant pres-sure, the amount of heat evolved will simply equal the loss of enthalpy

(&1) that occurs in the reaction. 90 In this model, the enthalpy lost by the unit mass graphite equals the heat available to be gained by the surroundings. However since this process is adiabatic (no heat enters or leaves the system), the heat available is applied to self-heating of the unit mass of graphite. The graphite is considered to be completely insu-lated, such that the temperature in the system will have to rise according to the heat capacity of the material within the system. 91 This model determines the temperature rise associated with the stored energy in a single point mass of graphite neglecting any heat flow into or out of the unit mass system. Obviously it would require the inte-gration of an infinite number of unit mass temperature increases to cor-relate to an actual graphite reflector. This in turn would require detailed knowledge of the three-dimensional stored energy distribution throughout the reflector. Since this information is generally not available, one must decide whether it is best to use peak stored energy or average stored energy to calculate the unit mass rise in temperature

• knowing that the result using the peak stored energy will not be repre-sentative of the maximum temperature of the bulk graphite.

It may be justified to use the temperature derived from the peak stored energy to set an upper bound to temperature, but it must be remembered that this temperature will be a local maximum. It would be a gross exaggeration to assume that this temperature could ever represent the bulk graphite temperature.

39

The effect of distributing the energy released over an entire reflector using an averaging technique is illustrated in a proposal for the release of stored energy at the Materials Testing Reactor. nThe maximum temperature rise upon release of stored energy is plotted ***

for three cases:

1} All energy released remained in the graphite pebble region now storing energy (Tpeak = 366°C}.

2} The energy released is uniformly distributed throughout entire graphite pebble zone (Tpeak = 259°C} *

• 3} The energy released is distributed throughout all the reflector graphite (Tpeak = 73°C).* 92 The BEPO calculations for determining peak temperature of graphite in preparation for their second anneal were based afirst, on the deter-mination of the distribution of stored energy in a lattice cell and in the reactor as a whole; and second, on* the substitution of this experi-mental information, suitably averaged, into the theory to estimate energy releases in the reactor itself under various conditions ***

their model treated each point in the graphite as though it were inde-pendent of the rest of the reactor.* 93 The calculated "upper limit to any possible local transient maximum gave the value of 428°C. The normal lattice averaging allowance would reduce this to 388°C, but for safety this allowance was reduced to half value, so that it could be stated with confidence that, apart from local transients, the over-all highest temperature would not exceed 408°C. 1194 The actual maximum temperature measured for the second anneal was slightly in excess of 300°C (316°C).

40

REFERENCES (Footnotes)

1. D. J. Littler, ed. Pro erties of Reactor Materials and the Effects of Radiation Damage (Lon on: ut erwort s, *

2. T. s. Neubert and R. B. lees, nstored Energy in Neutron-Bombarded Graphite,*

Nuclear Science and Engineering (1957) Vol. 2, p. 748.

3. littler, p. 22

4. R. E. Nightingale, ed. Nuclear Graphite (New York;Academic Press, 1962),

p. vii-viii.

5. w. P. Eatherly, *Nuclear Graphite - The First Years," Journal of Nuclear Materials (1981), 100, p. 55.

6* Nightingale, Nuclear Graphite, p. 227

• 7.

8.

R. E. Nightingale, "Record of Proceedings of Session E-21," Proceedin~s of Second United Nations International Conference on The Peaceful Uses o Atomic Energy (1958) Vol. 7, p. 550.

Nightingale, Nuclear Graphite, p. 226

9. C. E. Ashbaugh, N. C. Ostrander and H. Pearlman, HGraphite Stored Energy in the UCLA Research Reactor,* Transactions of the American Nuclear Society (1986), Vol. 52, p. 372.

10. B. S. Gray and R. P. Thorne, *The Correlation of Graphite Irradiations,"

Journal of British Nuclear Energy Society (1968), Vol. 7., p. 91.

11. J. F. Kircher and R. E. Bowman, eds. Effects of Radiation on Materials and Components (New York:Reinhold Publishing Corporation, 1964), p. 331.

C. Oalmasso and 6. F. Nardelli, "The Wigner Release in Graphite-Moderated Reactors,H Energia Nucleore, (English translation in USAEC Report AEC-tr-4545) May 1961, p. 4.

13. Kircher, p. 331

14. Dalmasso, p. 1

15. R. E. Nightingale and J. F. Fletcher, *Radiation Damage to Graphite from 30°C - 185°c,* Reactor and Fuels Research and Development Operation, Report #HW-47776, p. 5. ,

16. Nightingale, Nuclear Graphite, p. 231

17. J. L. Dickson, G. H. Kinchin, R. F. Jackson, W. M. lomar, J. H. W. Sirmnons, "BEPO Wigner Energy Release,* Proceedin s of the Second United International Conference on the eace u ,

Vol. 7, p. 264.

R-1

18. Ashbaugh, et al, p. 373

19. Dickson, et al, p. 276

20. ibid., p. 271

21. Ashbaugh, et al, p. 373

22. Nightingale, Nuclear Graehite, Chapter 2

23. L. M. Currie, V. C. Homister and H. G. MacPherson, "The Production and Properties of Graphite for Reactors," Proceedinfs of the First United Nations International Conference on the Peacefu Uses of Atomic Ener 9 *

24. Record of Proceedings Session 11B, Proceedings of the First United Nations International Conference on the Peaceful Uses of Atomic Energy (1955) Vol 7,

p. 505 *

• 25.

26.

27.

Kircher, p. 349 Nightingale, Nuclear Graehite, p. 218 R. L. Carter, R. L. Eichelberger and S. Siegel, "Recent Developments in the Technology of Sodium - Graphite Reactor Materials," Proceedings of the Second United Nations International Conference on the Peaceful Uses of Atomic Energy (1958) Vol. 7, p. 76.

28. ibid, p. 81

29. R. A. Meyer and D. G. Schweitzer, "Post-Irradiation Measurements of Thermal Conductivity Between 200°C and 600°C for Low Permeability Graphites,"

Proceedings of the Fifth Conference on Carbon (1962) Vol. 1, p. 328.

30

• S. Glasstone and A. Sesonske, Nuclear Reactor Engineering, (New York:Van Nostrand Reinhold Company), p. 440. .

31. B. B. Brohovich, F. I. Ovchimnikov, V. I. Klimekov, P. V. Glazkov and B. M. Dolishnyuk, "Dissasembly of an Experimental Uranium - Graphite Isotope Reactor after Four Years of Operation," Proceedinss of the Second United Nations International Conference on the Peaceful ses of Atomic Energy (1958) Vol. 7, p. 246.

32. Record of Proceedings Session 11B, p. 506.

33. W. K. Woods, L. P. Bupp and J. F. Fletcher, "Irradiation Damage to Artificial Graphite," Proceedin s of the First International Conference on the Peaceful Uses of Atom c Energy o. , p. *

34. R. L. Carter, "Preliminary Analysis of Radiation Damage Gradients in Moderator Graphite," Atomics International Report ITID-7565, p. 33.

35. Brohovich, et al, p. 246 R-2

. )

I

36. Nightingale, Nuclear Graphite, p. 331

37. Neubert, p. 758

38. Nightingale, Nuclear Graehite, p. 339

39. Kircher, p. 353

40. Woods, et al, p. 465

41. Record of Proceedings Session llR, p. 507

42. Kircher, p. 352

43. A.H. Cottrell, J.C. Bell, G. B. Greenough, W. M. Lomer and J. H. W.

Sinnnons, aTheory of Annealing Kinetics Applied to the Release of Stored Energy from Irradiated Graphite in Air-cooled Reactors, 11 Proceedings of the Second United Nations International Conference on the Peaceful Uses

• 44.

45.

of Atomic Energy (1958) Vol. 7, p. 320.

E. Fast, F. 0. Smith and J. D. Cord, "A Proposal for the Controlled Release of Stored Energy in the MTR Reflector Graphite, 11 Report #ID0-16656, 1959.

Woods, et al, p. 466

46. Cottrell, et al, p. 325

47. Carter, et al, "Recent Developments .... ", p. 76.

48. Nightingale, Nuclear Graehite, p. 338

49. ibid, 337

50. G. R. Hennig and J. E. Hove, "Interpretation of Radiation Damage to Graphite," Proc

• of the International Conference on the Peaceful Uses of Atom, c o * , p. *

51. Neubert, p. 763

52. Kircher, p. 349

53. Woods, et al, p. 464

54. H. Bridge, R. T. Kelly and B. S. Gray, ustored Energy and Dimensional Changes in Reactor Graphite," Proceedings of the Fifth Conference on Carbon (1962), Vol. 1, p. 307.

55. R. E. Nightingale, J.M. Davidson and W. A. Snyder, Damage to Graphite 11 Irradiated up to 1000°C, Proceedin s of the Second United Nations Inter-11 national Conference on the Peacefu Uses o Atom c Energy 8 ol. 7, p.

295.

56. Dalmasso, p. 4 R-3

57. Nightingale, Nuclear Graphite, p. 326

58. Cottrell, et al, p. 318

59. ibid, p. 318

60. Dickson, et al, p. 271

61. Nightingale, Nuclear Graphite, p. 329

62. ibid, p. 330

63. ibid, p. 329

64. ibid, p. 331

65. Ashbaugh, et al, p. 373

• 66. Nightingale, Nuclear Graphite, p. 346

67. ibid, p. 347

68. R. W. Powell, R. A. Meyer, R. G. Bourdeau, "Control of Radiation Effects in a Graphite Reactor Structure,* Proceedin s of the Second United Nations International Conference on the Peace u Uses of Atom c Energy 195 Vol. 7, p. 291.

69. Hennig, p. 670

70. ibid, p. 674

71. Woods, et al, p. 465 72

• Nightingale, Nuclear Graphite, p. 346

• 73.

74.

75.

ibid, p. 267 Bridge, et al, p. 303 Powell, et al, p. 291

76. Currie, et al, p. 469, 470

77. F. D. Jones and P. B. Schubert, En ineering Encyclopedia (New York:

1 Industrial Press Inc., 1963) p. 89 *

78. ibid, p. 280

79. ibid, p. 679

80. W.R. DeHollander, uspontaneous Ignition," Proceedings of the 1958 AEC and Contractor Safety and Fire Protection Conference (1958), p. 33.

R-4

.I j ., J I

81. Powell, et al, p. 293

82. TID-11295, Nuclear Fuels and Materials Development, Oivision of Reactor Development, AEC, 1961.

83. D. G. Schweitzer, Proceedings of US/UK Meeting on the Compatibility Problems of Gas-Cooled Reactors held at ORNL, 1961, Report ITID-7597, p. 502.

84. S. Peterson, "Ignition and Combustion of Reactor Materials under Accident conditions,* Journal of Nuclear Safety (Winter 1965-1966) Vol. 7, p. 169.

85. Fast, et al, Report #ID0-16656

86. DeHollander, p. 37

87. ibid, p. 37 s.

88

• Peterson, "Ignition and Combustibility of Fuels and Structural Materials,"

Journal of Nuclear Safety (Fall 1964) Vol. 6, p. 41.

89. R.H. Perry, Perry 1 s Chemical Engineer 1 s Handbook (New York:McGraw-Hill Book Co., 1963) p. 4-26.

90. W. L. Masterton and E. J. Slowinski, Chemical Principles, (Philadelphia:

W. B. Saunders and Co., 19!3) p. 77.

91. DeHollander, p. 36 92 *. Fast, et al, Report #ID0-16656

93. Cottrell, et al, p. 321

94. Dickson, et al, p. 264 R-5

BIBLIOGRAPHY A. R. Anderson, N. K. Taylor, R. J. Waite and J. Wright, "Measurement of Oxidation Rates of B.E.P.O. Graphite,u US/UK Meeting on Compatibility of Gas-Cooled Reactors, Oak Ridge National Laboratory, Report #TID-7597, 1960.

C. E. Ashbaugh, N. C. Ostrander and H. Pearlman, *sraphite Stored Energy in the UCLA Research Reactor,u Transactions of The American Nuclear Society, Vol. 52, p. 372, 1986.

D. E. Baker, uGraphite Studies For Gas-Cooled Reactor Program, 11 Information Meeting on Sas-Cooled Power Reactors, Oak Ridge National laboratory, Report

1. TID-7564, P. 189, 1958.

J. c. Bell, and G. B. Greenough, uStored Energy in Pile Irradiated Graphite,u Research and Development Branch, Windscale Works, Sellefield, England, Report

1. TID-7565, p. 110.

Bezjak and S. Maricic, "The Correction For Instrumental Broadening of Two-Dimensional hk-X-Ray-Reflection in Carbon Black Crystallite-Size Measurements,"

Proceedings of the First International Conference on The Peaceful Uses of Atomic Energy, United Nations, New York, Vol. 8, pp. 491-499, 1956.

H. Brid.ge, B. T. Kelly and 8. S. Gray, "Stored Energy and Dimensional Changes in Reactor Graphite," Proceedings of The Fifth Conference on Carbon, Vol. 1, The McMillen Co., New York, pp. 289-313, 1962.

B. B. Brohovich, et al, "Disassembly of An Experimental Uranium-Graphite Isotope Reactor After Four Years of Operation," Proceedings of the Second United Nations International Conference on The Peaceful Uses of Atomic Ener ,

Un te at ons, ew Yor, o. , pp. , 1

• R. L. Carter, *preliminary Analysis of Radiation Damage Gradients in Moderator Graphite," Atomics International, Report fTID-7565, pp. 33-45.

R. L. Carter, R. L. Eichelberger and S. Siegel, "Recent Developments in The Technology of Sodium-Graphite Reactor Material$," Proceedings of The Second United Nations International Conference on The Peaceful Uses of Atomic Energy, Vol. 7, pp. 72-81, 1958.

"Chernobyl: The Soviet Report," Nuclear News, Vol. 29, No. 13, October 1986 A.H. Cottrell, et al, "Theory of Annealing Kinetics Applied to The Release of Stored Energy from Irradiated Graphite in Air-Cooled Reactors," Proceedings of The Second United Nations International Conference on The Peaceful Uses of Atomic Energy, Vol. 7, pp. 315-327, 1958.

A.H. Cottrell, *Annealing A Nuclear Reactor: An Adventure in Solid-State Engineering,u Journal of Nuclear Materials, 100 pp. 64-66, 1981.

L. M. Currie, et al, "The Production and Properties of Graphite for Reactors,u Proceedings of the First United Nations International Conference on The Peaceful Uses of Atomic Energy, Vol. 8, pp. 451-473, 1955.

B-1

BIBLIOGRAPHY {cont'd)

R. L. Cushing and J. Pieroni, *Experimental Determination of Stored Energy in Graphite Combustion in Air," US/llK Meeting on Compatibility Problems of Gas-Cooled Reactors, Oak Ridge National Laboratory, Report ITID-7597, 1960.

c. Dalmasso and G. F. Nardelli, *The Wigner Release in Graphite-Moderated Reactors," Energia Nucleore, (English translation in llSAEC Report AEC-tr-4545, May 1961.

J.M. Davidson, NStored Energy in Irradiated Graphite in US/UK," Graphite Conference, St. Giles Court, London, December *16-18, 1957, USAEC Report TID-7565, (Pt. 1), pp. 11-20, 1959.

Davidson, Woodruff, Yoshikawa, AHigh Temperature Radiation Induced Contraction in Graphite," Proceedings of The Fourth Conference on Carbon, New York, Perganon Press, 1960.

w. R. DeHollander, *spontaneous Ignition," Proceedings of the 1958 AEC and

• Contractor Safety and Fire Protection Conference, pp. 33-39; 1958.

J. L. Dickson, et al, "BEPO Wigner Energy Release", Proceedings of the Second United Nations International Conference on the Peaceful Uses of Atomic Energy, Vol. 7, pp. 250-281, 1958.

R. E. Dahl, "Experimental Evaulation of The Combustion Hazard to The Experimental Gas-Cooled Reactor {ESCR), 11 Materials Development, Report

1. HW-67792, 1961.

W. P. Eatherly, *Nuclear Graphite - the First Years,* Journal of Nuclear Materials, 100, p. 55, 1981.

8 E. Fast, E. O. Smith and J. D. Cord, A Proposal for the Controlled Release of Stored Energy in the MTR Reflector Graphite,a Report NO. ID0-16656, 1959.

R. Foster and R. L. Wright, Basic Nuclear Engineering, 3rd ed. {Allyn and Bacon, Inc., Boston), pp. 346-350.

J.P. Genton, Properties of Reactor Materials & the Effect of Radiation Damage, ed. D. J. Littler, London, Butterworths, (TK 9006.L77), p. 22, 1962.

s. Glasstone and A. Sesonske, Nuclear Reactor Engineering, (Van Nostrand Reinhold Company; N. Y.), pp. 424, 438-442.

B. s. Gray and R. P. Thorne, *The Correlation of Graphite Irradiations,*

British Nuclear Energy Society, Vol. 7, p. 91, June 1968.

Handbook of Chemistry G. R. Hennig and J.E. Hove, "Interpretation of Radiation Damage in Graphite,*

Proceedings of the International Conference on the Peaceful Uses of Atomic Energy, pp. 666-675, vol. 7, 1955 B-2

BIBLIOGRAPHY (cont'd)

The Industrial Graphite Engineering Handbook, National Carbon Company, 1962.

F. D. Jones, P. B. Schubert Engineering Encyclopedia (New York:lndustrial Press Inc ** 1963 p. 897.

G. H. Kinchin, "The Effects of Irradiation on Graphite,* Proceedings of the International Conference on the Peaceful Uses of Atomic Ener , United at ons, ew ork, o *

• pp.

• J. F. Kincher and R. E. Bowman, ed., Effects of Radiation on Materials and Components, Reinhold Publishing Corporation, New York, March 1964.

D. J. Littler, ed. Properties of Reactor Materials and the Effects of Radiation Damage (London:Butterworths, 1962).

w. L. Masterson and E. J. Slowinski, Chemical Principles (W. B. Saunders and Co., Philadelphia), p. 77, 1973 *

• R. A. Meyer and D. G. Schweitzer, "Post-Irradiation Measurements of Thermal J.

Conductivity Between 200°C and 600°C for Low Permeability Graphites. 11 Proceedings of the Fifth Conference on Carbon, New York.

s. Nairn and V. J. Wilkenson, 11 Prediction of Conditions for Self-Sustaining Graphite Combustion in Air," US/UK Meeting on Compatibility Problems of Gas-Cooled Reactors, Oak Ridge National Laboratory, Report ITID-7597, 1960.

T. s. Neubert and R. B. Lees, "Stored Energy in Neutron-Bombarded Graphite,"

Nuclear Science and Engineering, Vol. 2,pp. 748-767, 1957.

J. J. Newgard, "Simple Semi-empirical Model for Neutron Induced Stored Energy in Graphite," Journal of Applied Physics, Vol. 30, pp. 1449-1451, 1959.

R. E. Nightingale, Nuclear Graphite, Academi c Press, New York and London, 1962.

R. E. Nightingale, J.M. Davidson and W. A. Snyder, "Damage to Graphite Irradiated to 1000°C," Proceedings of Second United Nations International Conference on the Peaceful Uses of Atomic Energy, Vol. 7, pp. 295-300, 1958.

R. E. Nightingale and J. F. Fletcher, "Radiation Damage to Graphite from 30°C - 185°C," Reactor and Fuels Research and Development Operation, Report IHW-47776.

Sigfred Peterson, "Ignition and Combustion of Fuels and Structural Materials,"

Journal of Nuclear Safety, Vol. 6, pp. 40-43, Fall 1964.

Sigfred Peterson, "Ignition and Combustion of Reactor Materials Under Accident Conditions," Journal of Nuclear Safety, Vol. 7, p. 169, Winter 1965-1966.

Sigfred Peterson, "Ignition and Combustion of Reactor Fuels, Coolants and Structural Materials," Jou rnal of Nuclear Safety, Vol. 8, pp. 25-30, Fall 1966.

8-3

... t BIBLIOGRAPHY (cont'd)

R.H. Perry, Perri's Chemical Engineer Handbook, McGraw-Hill Book Company, New York, 19b.

P. I. Perry, "Reactor Hazards," Information Meeting on Gas-Cooled Power Reactors, Oak Ridge National Laboratory Report #TID-7564, p. 72, 1958.

R. w. Powell, R. A. Meyer and R. G. Bourdeau, "Control of Radiation Effects in a Graphite Reactor Structure," Proceedings of the Second United Nations International Conference on the Peaceful Uses of Atomic Energy," Vol. 7, pp. 282-294, 1958.

Preliminary Statement of the Select Panel for Post-Chernobyl Safety Review, August 1986, Scientists and Engineers for Secure Energy (SE2).

M. Quetier and J . Rappeneau, 11 Energy Stored by Irradiated Graphite Variation of The Energy as A Function of Successive Annealings," Properties of Reactor Materials and The Effects of Radiation Damage, D. s. littler, ed. (London,

• N.

Butterworths), 1962.

s. Rasor and J. D. McClelland, "Properties of Graphite, Molybdenum and Tantalum to their Destructive Temperatures," Report WADC-TR-56-400 (Pt. 1),

1956.

Record of Proceedings of Session E-21, Proceedings of the Second United Nations International Conference on the Peaceful Uses of Atomic Energy, Vol. 7, pp. 548-551, 1958.

Report ITID-7569, Proceedings of the 1958 Atomic Energy Commission and Contractor Safety and Fire Protection Conference, held at Atomic Energy Commission Headquarters Building, Germantown, Maryland, June 24-25, 1958.

Report #TID-11295, Nuclear Fuels and Materials Development, Division of Reactor Development, Atomic Energy Co11111ission, 1961 *

• M. Robkin, "Response to Petition from CRG," University of Washington, 1986.

E. R. Schmidt, "Design Analysis of the Graphite Reflector Elements," Technical Memorandum, TM-ERS-62-2, Missouri University Research Reactor, Des1gn Data, Vol. I, 1962.

D. G. Schweitzer, Proceedings of US/UK Meet i ng on the Compatibility Problems of Gas-Cooled Reactors held at Oak Ridge National Laboratory, 1961, Report ITID-7597, p. 502.

R. E. Sonntag and G. J. Van Wylen, Introduction to Thermodynamics: Classical and Statistical, John Wiley and Sons, New York, 1971.

B. I. Spi nrad, "Stored Energy in Reflector Graphite of Research Reactors,"

Iowa State University, 1986.

w. K. Woods, L. P. Bupp and J. F. Fletcher, "Irradiation Damage to Artificial Graphite," Proceedin¥s of the International Conference on the Peaceful Uses of Atomic Energy, Vo. 7, pp. 455-471, 1956.

8-4

DOCKET NUMBER ri>ETITION RULE PRM 53-44 'liJ (5'1 r£ $/g4!) <!ft/

Research Reactor Facility OOGKETL US~RL UNIVERSITY OF MISSOURI Research Park Columbia, Missouri 65211 "87 FEB -2 P2 :36 Telephone (314) 882-4211 January 28, 198ZF

[) C, Secretary

u. s. Nuclear Regulatory Commission Washington, D. C. 20555 ATTENTI ON: Correspondence and Records Branch

SUBJECT:

DOCKET NO. PRM-50-44

• The University of Missouri Research Reactor (MURR) at Columbia, Missouri is submitting these comments in response to the July 7, 1q8fi petition filed by the Committee to Bridge The Gap (CBG) regarding the possible hazards associated with graphite stored energy and the cred-ibility of a graphite fire in research reactor applications. These comments were solicited by the Nuclear Regulatory Commission for their consideration of the need to institute proposed rule-making, PRM-50-44 that appeared in the Federal Register/Vol. 51, No. 170/September 3, 1986.

The arguments included in these comments will establish that this petition for rule-making should be rejected. The petition makes many claims and analogies that are based on speculation and opinion and are not supported by fact. The arguments presented here will focus on the mi stakes and inaccuracies in the petition and will also present evi-dence why these concerns are unfounded regarding MURR's specific

• application of graphite *

,s~~cn~~~,

~Meyer Acting Reacto Reviewed and Approved:

~ r-l\ Don M. Alger Associate Director

'>>ti\~ \

XC: R. E. Carter COLUMBIA KANSAS CITY ROLLA ST. LOUIS an equal opportunity institution FEB 2 1987 r- lrnn,ulana"rl hv NIM .......

I u S. NU CLEtK l<EGULATORY CO\\\jJ:,~lON DOCKET, ~ , & :c.EJNICE r-'" .1~CH OFfl Ct:. lJF THE SLCk'.:TAtlY OF H' v.,1: ,~i:::JN Do- ,l 'lt ..; L.:>~k:: s Pcs' ro,, ~

.. r~ lJJ+e r "-

I I 1 ~7 z,. 'i{

A *l J

,J,_ ,, ~

TABLE OF CONTENTS SECTION NO. DESCRIPTION PAGE NO.

I. INTRODUCTION **** . . . .. . . .. . .. . . .. 1 I I. CREDIBILITY OF GRAPHIRE FIRE IN RESEARCH REACTOR APPLICATIONS * * * * * * * * * * * * * ... 3 I II. WIGNER ENERGY IN RESEARCH REACTOR GRAPHITE. . . . . 8 IV. RESPONSES TO OTHER CBG PETITION COMMENTS *** . . . 13

v. APPLICATIONS OF GRAPHITE AT MURR. . . . . . . . . . 16 A. The MURR Graphite Reflector . . . . . . . . . 16 B. The MURR Thennal Column *** . . . . . . . . 17 VI

• ANALYSIS OF STORED ENERGY IN THE MURR GRAPHITE REFLECTOR * * * * * * * * * ... . . . . . 23 A. Stored Energy calculations **** ...... 23 B. Analysis of The Possible Hazard Associated With The calculated Stored Energy * * * * *

• 25 C. Credibility of Graphite Ignition in The MURR Reflector Application * * * * * * . . . . 26 VII. ANALYSIS OF STORED ENERGY IN THERMAL COLUMN GRAPHITE * * * * * * * * * * * * * * * * * * . . .. 27 REFERENCES . ..... .... . . . .. . . . . . . 28

• APPENDIX A GRAPHITE REFLECTOR ELEMENTS * * * * * * * * * * *

• A-1 APPENDIX A REFERENCES * * . . . . . . . . . . . .. . . . . . . A-20 APPENDIX B THERMAL COLUMN . . . . . . .. . . ... . . . . .. B-1 APPENDIX C ANALYSIS OF THE HAZARDS ASSOCIATED WITH STORED ENERGY RELEASE IN THE MURR GRAPHITE REFLECTOR ... C-1 APPENDIX C REFERENCES * * * * * * * * * * * * * * * * * * . . . C-6 APPENDIX D ANALYSIS OF HAZARD ASSOCIATED WITH STORED ENERGY RELEASE IN THERMAL COLUMN ***** ..... 0-1 i

~~ I. INTRODUCTION The petition by the Committee to Rridge The Gap (CGR) addresses two separate issues: (1) the credibility of a graphite fire at research reactors and (2) the potential contribution of the Wigner (stored) energy to the autoignition of research reactor graphite. The CBG cites the occurrence of burning graphite at Chernobyl as their basis to allege that research reactor applications of graphite are unreviewed safety questions.

To imply that the consequences of an accident involving research reactor graphite are analogous to the consequences of the Chernobyl accident ignore the many extreme differences in the .design and operation between the

• Chernobyl RBMK reactor and even the highest power research reactor. These extreme differences include power level, core size, fission product inven-ory, operating temperature, reactor control systems, and inherent design characteristics.

The RBMK reactor is a 3200 MWt plant operated with graphite temperatures in excess of 700°C during normal operation. It caA be characterized as having a large core with a positive core void coefficient, subject to local power instabilities. By comparison, the largest licensed research reactors are 20 MWt and have coolant operating temperatures ranging from 30°C to 75°C

• These research reactors all are designed to have inherent safety features including negative core temperature and void coefficients.

The CBG also infers that graphite fires were the initiating events in both the Chernobyl and Windscale accidents, when in fact, they were corollary events in both cases. The graphite fires may have exacerbated the accident consequences but were not the cause of either event.

The cause of the Chernobyl accident was determined to be a prompt criti-cal reactivity excursion with rapid fuel failure and a steam explosion which resulted in core disassembly and subsequent destruction of the reactor enclosure (building). 1 , 2 The fragmented graphite (1500 tons) was subjected to temperatures in excess of 2000°C (fuel temperature was estimated to be 3000°C) and burned. This fire was sustained due to the continued heat source of the melted fuel and an unrestricted supply of oxygen. With a normal operating graphite temperature of 700° - 750°C there was expected to be neg-ligible Wigner energy to have contributed to the graphite fire at Chernobyl.

The immediate cause of the accident at Windscale, on October 10, 1977 nwas the application too soon and at too rapid a rate of a second nuclear heating to release the Wigner energy from the graphite, thus causing the failure of one or more uranium fuel cartridges, whose contents then oxidised

• slowly *** The exposed uranium smouldered throughout the course of the day and gradually led to the failure of other fuel cartridges and their combus-tion and to the combustion of graphite." 3 "Attention was given to the possibility that, irrespective of the second nuclear heating, a large local release of Wigner energy occurred in a pocket of graphite which had not been annealed for some time, and that as a result high local temperatures were caused, sufficient to result in the failure of a fuel cartridge, or a lithium-magnesium cartridge or even combusion of graph-

• ite. The CoJ1111ittee of Inquiry studied the thermocouple records, laboratory data on the failure of cartridges, and general information.

rejected explanations of this type. 114 The Committee One lll.lst keep in mind when trying to evaluate the hazards associated with graphite oxidation and combustion, that the burning of graphite per se is not a radiological hazard, unless the burning graphite can elevate *tem-peratures to the point that fuel claddin~ melts with subsequent release of fission products.

In both of the accidents cited by CBG, the melting of fuel initiated the burning of graphite and not vice versa as CBG infers.

II. CREDIBILITY OF GRAPHITE FIRES IN RESEARCH REACTOR APPLICATIONS

• safety analyses for research and test reactors are based on the concept of a postulated Design Basis Event (DBE), an event for which the risk to the public health and safety is greater than from any event that can be mecha-nistically postulated." 5 If a facility can meet the 10CFR20 and 100 require-ents for public health and safety for a DBE condition," then the capability of the facility to withstand normal and abnormal operational transients and a broad spectrum of postulated credible accidents wi~hout undue risk to the public would be defined within the DBE." 6

• There is no doubt that the burning of graphite is possible, if the temperature of the graphite is elevated past the ignition temperature and the theoretical volume of oxygen required for combustion is continuously applied.

The question to be answered, however, is whether graphite in research reactor applications can be made to burn.

In most research reactor graphite configurations, the conditions for substantial combustion are difficult to postulate even under accident condi-tions. In most applications even if the ignition temperature were exceeded, oxygen supplies would be insufficient to sustain combustion

• The credibility of graphite fire in large block geometry, oxygen restricted (or vitiated atmospheres in canned graphite) and water submerged applications at research reactors does not lend itself to a simple generic engineering answer. The applications and operating conditions in various research reactors are diverse, and the postulated initiating events are endless. However, one can argue that a graphite fire is a tertiary or secondary event at best, requiring some initiating event to be possible (a massive temperature excursion of at least 700 - 800°C magnitude). If an initiating event of this magnitude is required to initiate a graphite fire and is credible, one would expect this initiating event to have been pre-viously analyzed and within the DBE envelope for each particular research reactor.

In other words, perhaps a more important determination is whether or not there is a credible initiating event for graphite fire and whether this initiating event is previously analyzed and within the accident consequences of the Design Basis Event.

If there are no such credible initiating events, then graphite fire in a research reactor's particular configuration is of no practical significance.

There have been several studies undertaken to determine the oxidation

• and combustion characteristics of artificial graphite. The results are varied due to difference in experiment setup and conditions.

of three prominent studies will be presented here.

The results

1) "If one defines a wthreshold oxidation temperature" as that at which a sample loses 1% of its weight in 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, pure graphite has a threshold oxidation temperature of 520°C to 560°C depending on coke source and processing variables In the temperature range 200°C to 250°C above the threshold

• oxidation; the oxidation is penetrative in character, such that the weight loss is greater than the volume loss. At still higher tem-peratures, the oxidation is diffusion-controlled, with virtually all I

oxidation taking place at the outside surface **

Similar oxidation effects on graphite are produced by steam and CO2 atmospheres. However, the threshold oxidation temperature is higher for these gases. Connnercial graphite will have a threshold oxidation temperature of about 700°C in steam and 900°C in C02." 7

2) uTests were conducted (at Brookhaven National Laboratory) to determine the exact conditions for self-sustained combustion. Sustained combustion was defined as that temperature which remained constant or increased without external heat input ***

The minimum temperature for self-sustained combustion of unirradi-ated graphite was found between 830°C and 930°C. It was determined that a fire could be put out either by suffocation or by chilling the graphite The irradiated pieces of graphite (exposure of 5.6 x 10 20 nvt) were heated to different temperature levels between 400°C - 900°C. At each

• temperature, the flow rate was changed and evidence for self-sustained combustion carefully sought.

evidence.

Up to and including 700°C there was no However, at 790°C combustion took place and the graphite temperature increased. 11 a

3) S. C. Hawley in his uAnalysis of Credible Accidents For Argonaut Reactors 11 cites the results of a graphite burning study in Nightingale's Nuclear Graphite. Hawley derives his 650°C ignition_ temperature from a graph in Nuclear Graphite (see Figure 1). His is a very conservative

• interpretation of ignition temperature (and rightfully so for the generic study he was providing), however, further investigation of the graph in Figure 1 is warranted.

This information was taken from Dahl's uExperimental Evaluation of the Combustion Hazard to the Experimental Gas-Cooled Reactor (EGCR). 11 Review of this docu~ent can lead to interpretations of ignition tem-perature significantly different from Hawley.

The experiment was performed in a channel that simulates a fuel channel in the EGCR. 11 Inittal conditions (for the experiment depicted by Figure 1) were 607°C, 4 lb/hr air flow, inlet air temperature 150°C and sufficient heat input to offset radial heat losses. The test proceeded for 70 minutes with constant air flow **** graphite temperature rose nearly linearly at approximately .6°C/minute during this period. The air flow was then increased fifteen fold and an expontial temperature rise occurred. (The 650°C point coincides with this step air flow increase.)

The rate reached 20°C/minute. After a short time the air flow was stopped and the entire asseMbly began to cool ** *

• Apparently oxygen starvation was retarding oxidation at the flow rate of 4 lb/hr and when air flow was increased, heat generation was

\

accelerated to a much higher degree than heat removal ....

• Combustion in many cases depleted the oxygen supply to such an extent that oxidation could not be supported in downstream sections." 9 It is not surprising that the graphite oxidation accelerated with the step change in air flow during Dahl's experiment. The forced air supply is the principal behind achieving high temperatures in forced draft furnaces that are used to burn coke for the melting or annealing of metals. The forced draft provides a fresh supply of oxygen to the fuel as well as purging combustion products that would inhibit continued combustion." 10 , 11

• It is evident from Figure 1 that at both 850°C and at 900°C when the forced air flow at rates of 50 to 60 lbm/hr was reduced, the graphite assembly began to cool. Therefore, if graphite at 850°C - 900°C is not subjected to forced air flow, the self sustaining temperature rise at these temperatures can not be maintained. The applications of graphite at research reactors preclude these magnitudes of air flow rate (as opposed to the large graphite piles for which these experiments were designed that used forced gas flow for cooling).

- - - Aor FI-IOOO - - - llrophite Te111perotwe L5 ft fro., llllet 10 110

- Power to Heaten

_! 900 BD*M

--.; JE i

E aoj** ~

l!-

100 I!

=

~

~

ii:

a 40i 52 ~

a IL 700 Z.0 16

• 110 100 l50 Tiffie, 111iD l00 HO 0

FIG. 14.16 A.D.. example of eelf-flUBtained burning of Speer-Nuclear 2 graphite ~

air." Inlet temperature ii 150°C.

0" Figure 1 12 In surrnnary, the fact that graphite will burn can not be refuted. How-ever, the question here centers on whether or not research reactor graph-ite fires are .credible or even if there is a credible initiating event. A transient that could take graphite in large block geometry to 600°C above operating temperature does not appear to be credible. Transients of this magnitude are precluded in research reactor applications by the inherent safety design features (passive) of research reactors as well as engineered redundant safety systems (active). The Corrvnittee to Bridge The Gap contends that the release of Wigner (stored) energy might be such an initiating event or could contribute to such an initiating event. This contention will be addressed fn the following section.

III. WIGNER ENERGY IN RESEARCH REACTOR GRAPHITE The Committee to Bridge The Gap contends that the release of Wigner energy in research reactor graphite could lead to a graphite fire or contrib-ute to the ignition of such a fire. They provide no proof to substantiate this claim. Instead they focus their attention on a perceived error in the methodology used by S. C. Hawley in creating a generic estimate of stored energy in Argonaut Reactors. They use half of their petition to argue that their method of estimating stored energy is better than Hawley's method. But regardless of whose estimate is more correct, they never show proof of the linkage between the release of stored energy and the autoignition of graphite *

• \

The Committee to Bridge The Gap does not appear to understand the mecha-nism behind the release of stored energy or the difference between total stored energy and stored energy releasable as a function of annealing tempera-ture. *As soon as stored energy in graphite was observed, it was recognized that measurement of the total stored energy alone would be inadequate for the prediction of possible temperature excursions in the graphite moderator. It is necessary to know the amount of energy released as a function of the annealing temperature.* 1 3

• The shape and intensity of the graphite energy release c~rves are the main concern from a reactor safety standpoint.

the energy release curves is the 200°C peak.

The most important feature of The obvious reason for concern in the 200°C peak region is that the release rates exceed the specific heat of the graphite, a condition where self-heating of the graphite can occur (see Figure 2). This intense peak is only valid at low temperatures (30°C-70°C). 14 At higher irradiation temperatures (even irradiation temperatures below 200°C),

the 200°C peak is substantially reduced 15 (see Figures 3 and 4). In fact, the energy release peaks for irradiation temperatures exceeding 120°C are less than the specific heat for the annealing temperatures and hence can not result in self-sustained heating of the of the graphite. Other investigations have made the same observation (see Figures 5 through 7).

Several other considerations must be understood about the release of stored energy in graphite: (1) the energy releasable to 800°C is only a fraction of total stored energy (i.e. for exposures ) 5000 Mwd/At, energy releasable is 40% of total stored energy; for exposures to 600 Mwd/At energy releasable ~ 80% of total stored energy) 16 ; (2) the release of stored energy occurs over time and not instantaneously (20 to 60 minutes for adiabatic laboratory releases 17 and 3 to 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> in actual reactor anneals) 18 and; (3) the peak graphite temperature reached during graphite annealing under

• reactor conditions (not adiabatic) is significantly less than the laboratory achieved peak temperatures under adiabatic conditions for a particular stored energy release (428°C peak local temperature predicted for the second BEPO release, the actual local peak temperature during anneal was 306°C). 19 The Committee to Bridge The Gap does not explore the fact that the release of Wigner energy itself requires an initiating event. It is observed consistently in both calorimetric and combustion experiments that until the annealing temperature exceeds the irradiation temperature by some definite

• amount, stored energy will not be released. 20 The threshold temperature increase required before any release of stored energy is encountered varies from 50°C 21 , 23 to 100°C 24 above the irradiation temperature.

CBG contends that the storage of energy in graphite is sufficient to raise the temperature of graphite several hundred ~egrees. 24 Even if this were true, the temperature excursion they postulate is far less than the ignition temperature of graphite or the melting temperature of fuel or fuel cladding.

EXPOSURE& IN 1,1,rd/AI 0.7 - - * !1700

- - - eooo

- - 11~0 0.6 i4110 107!1 760 O.&

f.... !160

.....,. 100 i 0.4

....en "D

(U "D

0.2 0.1

• 0 100 zoo :goo Anneoli"'il Temperature, *c Fro. 12.6 Stored-energy release curvd' of CSF graphite irradiated near 30°C.

Figure 2 2 s 0.08 111-C QM 39-C

....~

~ 0.04 1i 0

"D

~ 0.02 000

-0.02 100 200 300 400 &00 600 Ann ollna Temperature, -C Fro. 12.9 S~ergy releue curves of CSF graphite'" exposed to 193 Mwd/At.

The higher irradiation temperature. greatly reduced the 200°C peak. Numbers under the curves give the stored energy released to 000°C.

Figure 326

025 u

0.20

<1633 Mwd/Ale

.... 200-C ii

~ 0.15 0.10

<1348 Mwd/Ale 250"C 0.05 100 200 300 <100 AnM0lln11 Te,.peratur ;"I:

FIG. 12.10 The effect of irradiation temperature on stored-energy relelll!e curve11 for high exposures.*** Numbers under the curves give the stored energy re1eued to 400°C.

Ff gure 4 27 Specific Heat Curve Superimposed on Drawing "i

~

"i Cl()-8

~ Dos*5 1-~

Temperature 60 1-70 85 1-49 £ x1D2°n 135 14B Sp oC Oii

'-Al I

\

CD

~0-4

~

"IJ I

02 ,;;;;---

0 500 600 700 800 lemperature 6 oC F,gurr 5. Ena-sY rck:asc from irradiated graphite (no intcnnediatc anneals)

Figure 5 28 07 0

i;'

• OS 0

z D

0-*

f OJ

~

. 02 0

C 01

• TlMPEIUTUCl[("C)

Flo. 6. Effect of temperature on mape of rate of energy release curve for colllltant dose of 4.4 x 10'° nvt.

Figure 629 Cl>'

200 ~ *-- -*oo---

n MP£ RA"TUAE(_"c) figure 9. Typicol lineor..f'iu am***

Figure 730,31 IV. RESPONSES TO OTHER CBG PETITION COMMENTS The Committee to Bridge The Gap uses their petition as a forum to indict the reactor safety studies of all research reactors. The CBG attacks the study done for analyzing credible accidents in Argonaut reactors and extrap-olates what they perceive as a deficiency in this docunent to the safety studies done for all research reactors. They do not substantiate their argu-ment regarding the inadequacy of the Hawley document, ITK.ICh less the safety studies done by other research reactors (i.e., MURR design analysis considers both the stored energy content and the potential growth in reflector graph-ite). The CBG also makes blanket statements in their petition about the

• inadequacy of emergency plans for research reactors. Research reactor emergency plans were recently upgraded and standardized and meet the require-ments of the Nuclear Regulatory Commission (ANS 15.16, Regulatory Guide 2.6, Proposed Revision 1 to Regulatory Guide 2.6). The evaluation of the adequacy of these emergency plans lies with the Nuclear Regulatory Commission using as their basis facts and engineering analyses about each specific licensee. The CBG does not prove that the emergency plans for research reac-tors are inadequate, they merely state that they are, as if their wopinion"

• is fact.

The CBG suggests that the Nuclear Regulatory Commission have licensees submit evacuation plans based on a 25% release to the environment of the equilibrium radioactive iodine inventory. CBG states in footnote 9, page 4 of their petition, that the 25% release fraction is derived from ANS 15.16 11 Standard for Emergency Planning for Research Reactors" (November 29, 1981).

Review of this document as well as Regulatory Guide 2.6 (January 1979) and Proposed Revision 1 to Regulatory Guide 2.6 (March 1982) show no such inventory fraction suggested.

This inventory fraction is utilized for power reactors (Regulatory Guide 1.4) and is based on complete core melts and loss of containment for power reactors (NUREG/CR-3011). The consequences of a research reactor fuel accident would not approach that of a complete core melt and loss of con-tainment at a power reactors. The iodine release fractions for a partial core melt with partial loss of isolation at a power reactor are less than

.08% (fraction of core inventory release= 8 x 10- 4 ). 32 CBG further suggests that licensees perform empirical measurements to detennine the stored energy content of their graphite reflectors. They do not suggest what importance they place ~n locating the point of maximum

• stored energy or why they want to determine it to+ 10% accuracy. Do they suggest dismantling an entire reflector element to determine this point of maximum energy storage? Do they recognize the radiation dose that workers would have to absorb to take samples for making these measurements? (MURR 1

recently removed a graphite reflector element for .replacement. The dose rate for this element exceeded 350 R/hr@ 1 foot.)

The maximum local stored energy is not as important as the average value of stored energy over a graphite block. The absorption of the stored

• energy released will occur first in the graphite and result in a temperature rise in the entire graphite assembly as the heat is conducted to other reactor materials (i.e., reactor structure and fuel).

uThe kinetics of energy release in a block of graphite with nonuniform stored energy is difficult to determine. Presumably the real behavior lies between that of a block containing the average stored energy and the other extreme of supposing each element of the block to rise locally to the maximum temperature characteristic of the local stored energy **** The relaxation time for this nonuniform temperature to equalize itself over the cell is only about ten minutes * *

• therefore, apart froM possible short-lived local transients, the maximum temperature reached will correspond to an average value of stored energy." 33 V. APPLICATIONS OF GRAPHITE AT MURR A. The MURR Graphite Reflector The graphite reflector comprises 2/3 of an annular right circular cylinder that surrounds the beryllium reflector and the reactor pressure vessel. The remaining 1/3 of the annular cylinder is comprised of sample irradiation positions with little or no graphite. The graphite reflector is composed of eight 30° wedges {Numbers 1, 2, 3, 4, 6, 7, 8, 9 on Figures 2 and 3). A vertical perspective of the graphite reflector loca-tion relative to the reactor core is presented in Figure 4.

The inside radius of the graphite in each wedge is 9.684 inches

• {24.60 cm) and the outside radius of each wedge is 18.262 {46.39 cm).

The graphite height is 34.89 inches {88.62 cm). The volume of each 30° wedge is 33,736 cm 3 {with the non-graphite volume of the beamport access ports subtracted).

The total volume of th~ graphite in the reflector is 8 x 33,736 cm 3 =

269,888 cm 3 {9.5 ft 3 ). Using an average density of artificial graphite as 1.6 g/cm 3 , the mass of each 30° wedge is 1.6 g/cm 3 x 33,736 cm 3 = 53,978 g

{~ 119 lbs). The total mass of the graphite in the reflector is 8 x 53.978 Kg= 431.8 Kg {~ 953 lbs)

• Seven of the 30° wedges have accumulated exposure history equivalent to total reactor operating history {original wedges). One wedge was re-placed in November 1986 by a new graphite wedge with sample irradiation space.

Each graphite wedge is canned in 1/8 inch aluminum, with a nominal Helium gas gap of .040 inches between the graphite and the canning. The canning of the graphite raises the internal volume average temperature to greater than 150°C during reactor operation. The principle upon which the canning is based is presented in Nightingale's Nuclear Graphite. 11 A more uniform temperature can be achieved if the graphite is insulated from the coolant channels in regions of lower power by a semi-stagnant gas layer between th~ moderator and the coolant." 34 B. The MURR Thermal Column The thermal column consists of a 60 inch thick graphite pack contained within a water-jacketed aluminum casing (see Figure 5). The column has a lead gamma shield positioned between the reflector ring around the core and the inner end of the case. The protrusion of the thermal column into the pool is a portion of the thermal column casing. This section is 37 1/2 inches square and 12 1/4 inches deep. This section is then stepped into a box 50 inches square and 68 inches deep. The graphite in this section forms a 48 inch cube (48 inches long x 48 inches wide).

Total volume of graphite in the thermal column is 2.095 x 10 6 ~m 3

(~ 74 ft 3 ). Total mass of graphite is 3,350 Kg (~ 7400 lbs).

Figure 6 is included to show perspective of thermal column location with respect to the reactor core *

'Ill I

I I

I I

I N

QJ I co

~

Cl I LI-

I I I I I I I I

I I

I I

I I I I I I I I I I I I I I I I I I I I I I I I I I I I

=,- -rrlf-~~/~:F=~~~---,~~!.tR1__ L

~

~

.l 1* o !l!i* *

• I  :~:I

• ,=:- i * -*

I J

-~--------

- I --1

= ....

I I

:-------1-*

t_______ __._ ____

L~------~---_-_-_-_-_-+------+

I g

..":,~,:;=' ~"'!.":.,,,

-- /-:--

\ /

_ _ _ _ _ _ _ _ _ _ _ _ _..:.J.::_ POOL WATEn IN

.-CENTER TEST HOLE TOR INLET

~ BERYLLIUM REFLECTOR

- GRAPHITE REFLECTOR

--312A2mm DIA.

• ,._---+-~+-4---f---~""'- 952.50 mm DIA.

,0 0-REFLECTOR TANK POOL OUTLET 38.10mm Figure 4

_l]

I II I II I I 4'-i'so I

N THERMAL COWMN DOOR RECESS ,,I "I

II I

"II :11*** II I II Figure 5

SCHEMATIC DIAGRAM OF THE REACTOR AND BIOLOGICAL SHIELD A - 4 x '1 Inch RenDVable Graphite Stringers B - Steel Thennal Colunn Door C - Graphite Pack D - Alunitllml Water Jacket E - Pressure Vessel and Fuel Assemly

• F - Core Centerline G - Spent Fuel, Ganma Irradiation 0 lm O=====ll Facility*

H - Pneunatic Tubes

• H

• F la--l:1AIH:tttil!'tt-- - - - ---

B E

Figure 6 VI. ANALYSIS OF STORED ENERGY IN THE MURR GRAPHITE REFLECTOR A. Stored Energy Calculations The existence of damage gradients in observations of irradiated graphite is well documented. These gradients occur due to the existence of pre-existing defects in graphite before irradiation; the energy spectrum of the neutron flux in the irradiated graphite and the tempera-ture differentials within the graphite. 35 These damage gradients are difficult to predict since the flux and temperature during irradiation affect both the creation of the defects as well as the removal (annealing)

• of the effects.

The dominating factor controlling the distribution of stored energy across a graphite block is the local running temperature. 36 "In the face of uncertainties due to variations in temperature because of fluctuations in power distribution (changes caused by control rod manipulation, fuel burn-up or loading fresh fuel), it is difficult to assign an exact temp-erature to a particular region of a graphite structure. Moreover, the effect of the temperature of the structure depends on such design features as the manner in which graphite pieces are supported, aligned and re-

• strained with respect to each other; on clearances; and on mechanical loads. All factors considered, the behavior of a moderator (reflector) structure is predicitable only within a broad range of values." 37 To estimate stored energy accurately, three dimensional distributions in the graphite of the fast flux and of the temperature are needed. 38 Since this detailed information was not available, the stored energy was calculated by using average temperatures and average fast flux values derived from computer code and design analyses data (see Appendix A).

A methodology similar to that used by Pearlman to estimate the stored energy at the UCLA reactor was utilized. The Mwd/At conversion factor from fast nvt was one suggested by the National Carbon Company in MURR's design analysis and is comparable to the Mwd/At-to-fast nvt conversions cited by Nightingale 39 and Davidson 40 in separate studies.

The methodology incorporates the stored energy equations and param-eters outl.ined in Nightingale's Nuclear Graphite (p. 329-331). Two methods were used to calculate the total stored energy, one using the Windscale curves and data (155°C, 205°C, 255°C curves) and the other using the Hanford data (30°C curve and the information from reference 8 and 10 from Figure 12.2 of Nuclear Graphite) *

• The first method (using Windscale data) resulted in the highest cal-culated average stored energy (275 cal/gram). This value is more than twice the calculated value by the second method {133 cal/gram); which it-self is high compared to the empirical data for irradiation exposure at

/

greater than 150°C (the MURR graphite volume averaged temperature a 166°C).

The Windscale curves "were obtained from irradiations carried out inside hollow fuel elements in high-flux reactors (DIDO, PLUTO and DMTR)

• for which the intensity of the damaging flux was considerably greater than normal for a graphite-moderated reactor. It is possible that under such conditions the radiation effects produced at a given total dose are significantly greater." 42 The overestimation of this technique is verified by the 150°C and 200°C points on Figure 12.2 (Nuclear Graphite) which were obtained from samples irradiated in the Hanford controlled-temperature facility. These points correspond to approximately a factor of 1/2 with respect to the Windscale curves. This factor would bring the two methods into closer agreement. Another reason why the Windscale curves and data seem suspect is that information derived from these curves extrapolates to a saturation level of stored energy at 30°C allTlOst twice that of the expected 600 to 700 cal/gram.43,44 Woods noted that "irradiation at 150°C effects an order of magnitude reduction in the changes of most physical properties compared with those incurred by irradiations at 30°C." 45 This order of magnitude change is not reflected in the Windscale data.

B. Analysis of The Possible Hazard Associated with The Calculated Stored Energy In spite of the probable overestimation of stored energy using Method 1 of Appendix A, the higher value was analyzed with respect to

• possible hazard consequences in Appendix C.

Since the graphite reflector elements are not physically located adjacent to the fuel elements, conservative heat transfer assumptions had to be made to determine the possible effects on the fuel from a total release of the graphite stored., energy.

The results of this analysis show that even at unrealistically short release times, the heat rate associated with a total graphite energy re-lease is less than the core decay heat rate. The LOCA analysis for MURR considers the decay heat rate load as its heat source with ~o consequent damage to the fuel. A lower heat load would present even less of a concern with regard to the core.

It must be stressed also that the consequences of the LOCA are less than that of the MURR Design Basis Accident. This would place a graphite energy release (if it were credible) well within the Design Basis Event envelope discussed earlier.

Evidence presented in Section III supports the contention that no energy release peak in excess of the specific heat of graphite is possible with exposure temperatures in excess of 120°C.

Since the bulk of the MURR reflector operates at temperatures greater than 150°C, this means that regardless of the total stored energy in the graphite reflector, at no point during the release of energy can the graphite achieve a self-sustained temperature excursion.

c. Credibilit~ of Graphite Ifnition in The MOR Reflector App fca~ion If the evidence presented in Section III is accurate, self-sustained temperature excrusions are not possible in the MURR reflector application.

Even if the maximum stored energy could be converted to an adiabatic temperature increase of 700°C, it is unclear how oxidation, nruch less combustion, could occur in an enclosed environment with a designed absence of oxygen (canned graphite, helium filled). 11 A self-sustained ignition can occur only in those situations which are capable of support-ing self-sustained combustion. For example, if the ambient pressure or ambient oxident concentration is insufficient for sustained combustion, it will also be insufficient for ignition. 114 5 The analysis above also neglects the fact that these reflectors are normally under 23 feet of water. The MURR has an emergency pool fill system with a verified flow rate great enough to maintain 3 feet of water

• above a completely severed 6 inch beamport. This source of water ensures that the reflector is covered with water even in such an accident con-dition.

VII. ANALYSIS OF STORED ENERGY IN THERMAL COLUMN GRAPHITE The graphite in the thermal column has a very' low stored energy content (see Appendix D). Even the maximum local stored energy is inadequate to create a peak adiabatic temperature rise great enough to cause graphite ignition.

The graphite stack 1s completely enclosed in an aluminum water jacketed casing with entry gained through a greater than 12 ton steel and masonite thermal column door and radiograph facility. This graphite is not exposed to sufficient air volume to support a fire even if one could be initiated *

• There are no hazards associated with the calculated amounts of stored energy in the thermal column. Furthermore, there are no credible initiating events for a Wigner release or the burning of graphite in the thermal column

• REFERENCES

1. s. Rippon, 11 Chernobyl:The Soviet Report," Nuclear News {Oct. 1986/volume 29/no. 13) p. 64.

2. Preliminary Statement of the Select Panel for Post-Chernobyl Safety Review, August 1986, Scientists and Engineers for Secure Energy

3. "Accident at Windscale #1 Pile on October 10, 1957, Presented to the 11 Parliament by the Prime Minister by Co!Mland of Her Majesty, Nov. 1975, Cmnd 302 {NP-6539) pp. 5-7.

4. ibid, pp. 9, 10.

5. NUREG-0849, Standard Review Plan for the Review and Evaluation of Emergency Plans for Research and Test Reactors, p. 1.

6. ibid, p. 1*

• 7. L. M~ Currie, V. C. Homister and H. G. MacPherson, 11 The Production and Properties of Graphite for Reactors, 11 Proce~din s of the* United Nations International Conference on the Peace u Uses o tom c Energy 955, Vol. 8, p. 470.

8. R. W. Powell, R. A. Meyer, R. G. Bourdeau, "Control of Radiation Effects in A Graphite Reactor Structure, 11 Proceedings of the Secona United Nations International Conference on the Peaceful Uses of Atomic Energy (1958),

Vol

• 7, p. 29 3.

9. R. E. Dahl, "Experimental Evaluation of the Combustion Hazard to The Experimental Gas-Cooled Reactor - Preliminary Burning Rig Experiments, 11

{Nov. 1961), HW-67792, pp. 15-17.

10. F. D. Jones, P. B. Schubert, Engineering Encyclopedia, {New York:Industrial Press Inc., 1963), pp. 507-527 *

• 11.

12.

T. Baumeister, ed. Standard Handbook for Mechanical Engineers, {New York:

McGraw-Hill Book Co., 1978)~ p.9-25, 9-26.

R. E. Nighting~le, ed. Nuclear Graphite, {New York:Academic Press, 1962),

p. 416.

13. Nightingale, Nuclear Graphite, p. 331.

14. T. S. Neubert and R. B. Lees, 11 Stored Energy in Neutron-Bombarded Graphite, 11 Nuclear Science and Engineering (1957), Vol. 2, p. 761 {Figure 7).

15. Nightingale, Nuclear Graphite, p. 341.

16. ibid, p. 338 {Figure 12.7).

17. ibid, p. 336 (Figure 12.5).

REFERENCES (cont'd)

18. J. L. Dickson, G. H. Kinchin, R. F. Jackson, W. M. Lomar, J. H. W. Simmons, "BEPO Wigner Energy Release," Proceedin s-of the Second United Nations International Conference on the eace u tom c nergy ,

Vol. 7, p. 265.

19. ibid, pp. 261, 264.

20. J. F. Kircher and R. E. Bowman, ed. Effects of Radiation on Materials and Components (New York:Reinhold Publishing Corp., 1964), p. 352.

21. A.H. Cottrell, J.C. Bell, G. B. Greenough, W. M. Lomer and J. H. W.

Simmons, "Theory of Annealing Kinetics Applied to the Release of Stored Energy from Irradiated Graphite in Air-cooled Reactors," Proceedings of the Second United Nations International Conference*on the Peaceful Uses of Atomic Energy (1958), Vol. 7, p. 320.

22. E. Fast, F. O. Smith and J. D. Cord, "A Proposal for the Controlled Release

• 23.

of Stored Energy in the MTR Reflector Graphite," Report IID0-16656, 1959.

W. K. Woods, L. P. Bupp and J. F. Fletcher, "Irradiation Damage to Artificial Graphite," Proceedin~s of the First International Conference on the Peaceful Uses of Atomic nergy (1955), Vol. 7, p. 466.

24. Petition for Rulemaking Submitted by The Committee to Bridge The Gap, July 8, 1986, p. 6.

25. Nightingale, Nuclear Graphite, p. 337.

26. ibid, p. 340.

27. ibid, p. 341.

28. D. J. Littler, ed. Properties of Reactor Materials and the Effects of Radiation Damage (London:Butterworths, 1962), p. 212 *

• 29.

30.

H. Bridge, B. T. Kelly and B. S. Gray, "Stored Energy and Dimensional .

Changes in Reactor Graphite,* Proceedings of the Fifth Conference on Carbon (1962), Vol. 1, p. 295.

Cottrell, et al, p. 319.

31. C. Dalmasso and G. F. Nardelli, "The Wigner Release in Graphite-Moderated Reactors," EnefGia Nucleore, (English translation in USAEC Report AEC-tr-4545), y 1961, p. 26.

32. uoose Projection Considerations for EMergency Conditions at Nuclear Power Plants, 11 (NUREG/CR-3011), pp. 2.5-2.9

33. Dickson, p. 264.

34. Nightingale, Nuclear Graphite, p. 487.

REFERENCES (cont'd)

35. Nightingale, Nuclear Graphite, p. 231.

36. Cottrell, et al, p. 231.

37. Nightingale, Nuclear Graphite, p. 486.

38. C. E. Ashbaugh, N. C. Ostrander and H. Pearlman, "Graphite Stored Energy in the UCLA Research Reactor," Transactions of the Arlerican Nuclear Society (1986), Vol. 52, p. 372.

39. Nightingale, Nuclear Graphite, p. 286 (Figure 9.20).

40. Davidson, Woodruff, Yoshikawa, nHigh Temperature Radiation Incuded Con-traction in Graphite," Proceedings of the Fourth Conference on Carbon, New York, Perganon Press, 1960, p. 600.

41. Nightingale, Nuclear Graphite, p. 341 (Figure 12.10).

• 42. ibid, p. 330.

43. ibid, p. 329.

44. J. J. Newgard, 11 Simple Semi-empirical Model for Neutron Incuded Stored Energy in Graphite," Journal of Applied Physics, Vol. 30, pp. 1449-1451, 1959.

45. Woods, et al, p. 471.

46. Gordon P. McKinnon, Fire Protection Handbook, 15th ed., (Quincy:National Fire Protection Association), pp. 3-4

• APPENDIX A GRAPHITE REFLECTOR ELEMENTS I. TOTAL HOURS AT 10 MW (Equivalency)

December 1986 - 119,039 total hours at full power 43,726 MWD total 119,039 hours4.513889e-4 days <br />0.0108 hours <br />6.448413e-5 weeks <br />1.48395e-5 months <br /> rl24l day hours

] = 4,959.96 = 4,960 days full power operation 5(4960 - x) + lOx = 43,726 24,800 - 5x + lOx = 43,726

• • 5x = 18,926 x = 3,785.2 days at 10 MW 4,960 - x = 1,174.8 days at 5 MW heck: (3,785.2 days)(lO MW) = 37,852 MWD (1,174.8 days)(5 MW) = 4,874 MWD 43,726 MWD 1,174.8 days at 5 MW ~ 587.4 days at 10 MW (equivalent MWD)

• +

3,785.2 days at 10 MWD

\

587.4 equivalent days at 10 MWD 4,372.6 days at 10 MW x 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />s/day 104,942 "' 1.05 x 10 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> at 10 MW 1.05 x 10 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> x 3600 sec/hour= 3.78 x 10 8 seconds at 10 MW A-1

I}. GRAPHITE VOLUME AND MASS r = 46.39 cm (18.262 _in) 2 r = 24.60 cm (9.684 in) 1 T

h = 88.62 cm (34.89 in) 2 2 2 A=n(r2 - r 1

) = 4,859.6 cm rt V C Ah = 430,661.5 cm3 ( 360° reflector)

V = 1/12 V 30° wedge

• Beamport Access Holes

= 35,888 ~ 35,890 cm 3

for solid wedge (i.e. no beamport holes) 1 = 21.8 cm (8.587")

Three holes+ 4u diameter r = 5.08 cm (2 in)

Q_) Three holes + 6" diameter r 1

= 7.62 cm {3 in) 2 2 3 3 V411 = nr 1 = 1,767.4 cm x 3 = 5,302.2 cm 1

2 3 3 v6" = nr 1 = 3,976.6 cm x 3 = 11,929.9 cm 2

3 V = 17,232.1 cm (all six beamport holes) tot This beamport access volume distributed among 8 graphite wedges {30° each) 3 V = 17,232.1 = 2,154 cm b.p./wedge 8 3

V = 35,890 - 2,154 = 33,736 cm 30°wedge A-2

3 Density artificial graphite a 1.6 g/cm 3 . 3 M = (33,736 cm )(1.6 *g/cm) = 53,977.6 grams= 119.2 lbs.

wedge (1 lb.= 453 grams)

M = 53,977.6 grams x 8 = 431,820.8 grams ~ 953.2 lbs. total total A-3

III. TEMPERATURE PROFILES IN LARGE GRAPHITE ELEMENTS Temperature distribution information was taken from Design Data I, "Design Analysis of the Graphite Reflector Elements". The result for graphite thermal conductivity of 1/8 the unirradiated value is used. These temperature distri-butions were generated by computer code HEATING, which performs a numerical solution of the general steady state and/or transient three-dimensional heat conduction equations. These calculations were made neglecting axial con-duction. 11 The axial variation of the heat generation was assumed to follow the nvt curves of Figure 4.3. The actual axial heat generation distribution in the graphite would be expected to be somewhat flatter than this, hence this assumption is conservative. 111 Assuming the ends of the graphite to be at pool water temperature {50°C) is also conservative, because the graphite canning isolates the graphite from direct pool contact and a helium gap insulates the canning from the graphite to reduce axial conductance

• A-4

TEMPERATURE ZONES USED IN ANALYSIS Volume % Ave. Temeerature

[15.6] Zone I (top) 83.5°C

[15.6] Zone II (bottom) 83.5°C

/1-----,

/ I ', [37.9] Zone II I ( central front) 250°c A

.(_

....._(

[30.9] Zone IV (central back) 150°c I ' "

I '

graphite log height= 34.89" {88.62 cm) 2 I A

• 405 cm wedge cross-section I

• /

I I

I

,/

/_

' '\

' '\ A

= .5506 A

"".4494 A cross-section

/ .....

' \

IV cross-section tlOT TO

':f Volume I Zone I+ II~ 2x5.445

• 31.21 34.89 SCALE.

Volume 1 Zone III+ IV= 68.8%

• Volume I Zone III* 68.8 (.5506) ""37.9i Volume I Zone IV ""68.8 (.4494)

• 30.91 A-5

Lt, '5/(,.. ~ ~~}

= ,.,,8 ~

\

• \

---

1 Centerline

~ct;S_:~,_-Jj_1r-+-+-i1_LJJU--:

- 4

---6

~7 J

0. 858'

, -:: 24, S"'-1. c.v,..., ----8

-h> w-.~eJ.~

of- t<~"'-~'\c:

-io9

~11 Large Refleccor Elem~nt ~3 Poincs of Contact Between Between Graphite and Can t":tf<o.~1?.c......

-h, .t,..} S-i..L~ e_d"J'-

EB 1/16 inch spacer o ( ~r~f\.-,r~

1/8 inch spacer

~p,;, (f_~ft'c. pjl.,,o f., L. 1..- bl=-  :}tt..

1/Ko -:: 1/5 CRo~.S -<;.~c.T1o-lAl- '-' ii,..u,) fJr\' -./ t;.~1 lf'I..,

C~'T~RL1,.)t_. Of- G~f'~\~ L,o(:,,

(. i)f..~'~ D F /2.<NY\ f I C:,,..1..~..t-5 3, 3 t\-1'JD 3 -.(

oF Pt.'-1<.r-l Pfff"A ,-.so~~ I , Q,;c;,eat-l f'tr,IP,L'(S*t oi=- -nt1:. b~f\-\1TI...__~ .. fL~ r...rcr-EL'Utc.,*lTS.) (p~* A-l, , A-1)

A-6

• .u ,.,

--~..1..,1 QI

~

,I.I Ill

~

~

f QI

,I.I

.c; Q.

Ill

~

Cl Circumferential Index, J

• Figure 3.3 Circumferential Graphite Temperature Profiles for Large Elements A-7

• ~r*r*,1, *-* .. 1.** !l**l 340 t~.: ~:: .:: ~-:*

r,, '*:

=~~

... 1.:~ *-*l--

I

- :*--=r:..:.: I**,

o

l. ~., ""'*'......

• .. +-~:::~::

I I ,

--~11. ..

.~ I"'. ,_i ..

.l *

~~

::)~tr~.:*;

H* 1 1*,:

. . . ;.J

• J: .......

1Jffe ;hit~~~;-

"'"l""l:7:;,

I .... *11. , ,,.,1 I.to ln:ti!nttal .. n, ~- 1,U, fr-,, ... " ,

• 1, ...... ", * * , .. , , *I.,.,.... ,... .,

I I'*

• 1 ,

ru** -.;~~

• Jf*'!..'1Tt*::..**:., .... , *! Lfr.!r'::,

• • • ~ .. I,-:

I*

. ,1 j',

o, I

.:~.

• .,: i,, .. ,,::;:-'.""";*,

.j.

.i

• t

~!

I"'!""~

1Wffi,E"**

1 ::,.

L! *.. ~ :*~: * -,k:.;: ;~~.~:.,-: :*::*~J= :-:~(-f=, . *.*~:-:-.r:~;:::--'-:-::;:: ;**:-=:* !r.-.:, -g. ~:: -:-<*-**:

.'ii~

t * ** * *

• • i

• I

• itt .. -ir'1 ,!1 v...*t:::-r--

*. *:-:;-":t_

I ; *

,}ilif,;.i ,:~

I* * **

,~ *- ....

1 . _. * * - *

~~- ~f-~:.::;Sfr:::::.=~~;:*-:::

I

-.,:.* :::.:-- ~:L::. : --- . . *=-~*- ..... ,..... _:-..:~:-: .. '

.+/-* --~~---:1**- ... -~-f:..;:?T- . -- -.

• =

-~-r~~=r~:;~*~:

- . . - *,*i -** . * * * *

.. *r

• • .*:.:*:1 ol *

. :..* .. :: ,.*. ,** lt'~(r:-:-- !1" 1* . h. ,*, 11 ::- ~ *.ff.L::*1 :,*i!:iµ t**:L~r: **:;J"t"r*i.::i:rntt 1 11 :iJ1 -~ .,l, +<* ......:-:1'.~.!;F.-:1 7,-t-'..:,;.., *:z:~:d::.~

~

,*; 1 *** !1: jl

• 1",y1"'7 l'f ! .!l' :r ,..~~*t

' * - ,, .. *~tl ' *. .t'

• • .,., ** 1*-;* 1 I .,.,'* *:.Li; J 1, l, 1

1 j' 1

! !1n i!:i ;;;~': .,. :r.;:. ~~ i-:tnJti 11 11.!'.. -~H ll cl'1/2-'

• +J rn i1; E1 '-J I ...

. *.J'~ '!J. ~...,

..:}i t"~

-* 1$1;~105i=i:: ~**.= =~:-:r:7-

.. - ** - ~ . ,... _ '..1,t:::::..;._ ..........

.. . ~ . : fflj rn 1 .l ' *m Hi 1 J~llp,~ :1: ,r:!

1 i;r~Jf

• iffi fl t!i W ..;rr fJ. 1.~

. .r-* ~.....:1:::':.!:J.:::tdtf . *. fut=:*

".. : 1:-11

.

• TI ,...... . . - '

J" 1 I~:, ~ t .. " . , \ ju t:::  ;;,,,,'.!":T, :J::. ~--**- .,'j .1 **.

H

,.t .u ... , . . . . . ***

"*H*- . ,

• jtP14 n-... ~ !...tt, I 1JJ..1.. 1t: ffi1 ~l =

,,,.;,.,;.,1 ,... * ....;.

300 .I" :t *::: :f ~:::1:~ :r.i: Tr:-: j . ~n. :::* r;-;-:.,I

_,,_..,.. ... - I' .. -- -~1*--*I

f ili

........ !. ~

... j?; jf'"' ' I+ l .....

. . lli!:* .....:-*

l:"

.1fll. ~; r. ' ~ .i,i l- l*! . 1-1

'tn r.:: r;:;

~ff*

.cii fik .,, * .i. J.

It*

LH t!.!' . -* .-a+/- ::;-, iffi Hf- fil :§ ui-:

tJ ~ .~ gt!.

'.if.

260 T,ri tf 1~+ I I 1 . ~

• - .1!-

tl:1!

Wti ..... imn:! !;L% ~µ u

~n ifa 0

. l.i:~ j.ttj .ti !f:l il 'il ...

?!-

. ]:~

~

.:1~~

Ill * :tT .. -

+~ rim 5: lt 1 ~1* f: h.:_;

i::i~ s*; --

tt1 t!'

~ ~

r.i::Jifi Hi:.

m1

l.l t ~

. ~ ~

II

}= ,m~:~ l*.r1 *i?

lt; t:~

"*lf.m

!I

'. :-:~ -~:. ti:. ,i-J: l-,.~ ;~! µ,s

~ :-J!

I

• • !* L:;.-  !~* ~

11. *~:;

~

Ill ..,, . ,~

7~

-... p L!

J t

. 11. *-

g*~ill : filli *iif_t.:i'. I\W. iH tJ; '::i ~

I,... tfil tl"j t:Ji ,,:r:.... :.t;-:::T*:.

iii 1l *: h: l

..11,

~ ill t .? l * .... 1....

""Ill 220 p** *** ftii ~-* .;;, II i,* . 1ITi @. ,.*

.... [,

t1r mt i I "i'i:. I':

-~*  ! . * . * ;l!: *:. ::::

I *fl~ :ril :.! , ;f;I jl!t lt:.i r::-.

mi w -:j llif .l  :

i '1i QI

,!. I

i:,,

f-,1

• - 1 =

..... - ....1~-;r *fftrj ilit Jl 1~-I lr- u Ulti~  : ~ *j.!...

11 I a, -r ~f :q * *It- * **1- *= *tf*. .****

co ~

~

.c  !! ;tlj ~ :!tl 1 jt :t. . ' + .. .

lli q:. LJ

.!,~ ,J J:

.:tv,..:

I ,...... **.  :!:! . :ili r~1:*.tI

,... ,:-- ,l. ' ::.rl-~ I , -**:t!. t , **.

180 '. [ :q C.

a, ~ l. 1

~

h :r w ...rl' f .,, ~i~~.:r*1:

- -!.: ~t*r *4t.t:

ff!: fi;;lli:gt c.,

"" =

~ '.u~ i I "I

h: H .: t::

J:

II l! 1H1tfo lr.t ~~

-*r****::r:

I:{; .,~ ....

,* t:1:-. :;:'

. I :iJ iP *- !J ~'1

~

• i l

~ *t:j  !'  ; 'I . .

j "i:i*~

140 i :H .

11t : I II- ~t I

. u. l:IJ 11'

!ii id r I

f t I ~;  !

~

t

~

t:'I,

!ill ,,.,i:

t I l

• trn*

1 ii l .~* :u I *.1: I l *i I j I:

qn:t :ii :

I*

. . l I JP I 100 I

1,,,

l1l. .

Ii"1: ..1m lil I i 1! ..

in= ,ff ,: 'I. ,!J! 'I I I

' I 1;::: !t I

,.. , 1ir! r; 60 *:!i jjj I :I I 1 3 .5 6 7 I 9 l 11 L 13 _j .15 I 17 19 21 ' .. 25 Radial Index - I figure 3.5 Radial Gra~hlt~ Temperature Profile*

11'nP T.at-11ta 1'1--nt-,a

LU NS 14 u

.....a, s::

0 4J

~

s::

12 .

I 0

N V

I 0

.-4

~

10 '

~ *~

In 3?-1 K--... I\ A./i.

llodt Bal V 7* ~

.-4

• ~ ... 2'11l N

ao .

' icfO/

.'\. 8 V

"' ~ \

I/'

r2J.o/ / "-. '24.o O'I I

c:(

ti 0

6 7.-'J.o I V

V ltefe ence Ca1u*

.. '2:Jo \

z

~

Q) 'ZAro I/ I/ Rods Full Out ~ "Zo-o \

"t:I Q)

,&.I Ill 4 1'70 I , / i\ ,- ~\

/.u (II)

Q)

,&.I s::

\Yo ;

I

~{;,,

~

/ \

'~ ~

r\.J '-/o \

2 }\0

,., /

io / / ~~

- J~[ / 1"- ~~

I"

,. '--="

0

+20 oU +16 +l +8 .f-4 0 -4 -8 -Hz -16 v -20 -2 4

{Ji t. '

0 V'

Incbu Above Core Centerline Flux for tit Hwd

l6 Ne 14 ID r::

0

~

~

GI 12 J '.1 0

N c::I "

J3 j J)

I a::.

....M V ~

0 u.

/ 0 10

>IJ

a:

.-1 I ~o l

I---.. \ A~~

lloch Bal: In ..,,

~

"5 N

Cl0 j,

8

(_

13~/

~

I

/* . ~ .... ) '.l() Q

-~

Jj ~

V

\

"A r t... I

....~ \Z,0/ J I\.. 1-Z. ci:

i .,,

fl.<

~

\1v I V V ._.,, 0

\ ii

z 0

Qj 6

\ 11T) V I/ IRefe ~ence Ca&E-Rods Pull Out

\ I= \ '-../

~

"'O Qj I ,/

\q D\

g

.,.i

'7 Oio

...00cO 4 1~.., /

~

u r::

£0)

\ ~~o

' \

~

'.I' H

2 /0 V ,/ \

l90 / / ~~

--~

~

V 0

+

- - - ~ .. - ,

- 1 ' ~~ -,- ~

- 4 r--

Inches Above Core Centerline Vl

~ ~\Ji ~..9i-=~

,:I O  ::r a: ~ F- -s Figure 4.3

• r" _j ~

l * .. *'-..)c 7

~

~

.s r-=-

.::::, Axial Variat f Integrated Neutron Flux fo Hwd \9 ~

IV. FAST FLUX DISTRIBUTION (>.821 Mev)

The flux distribution within the graphite was derived from EXTERMINATOR II code (pages A-12 through A-14).

Average flux values used for analysis:

11 2 Zone I = 8.16 X 10 neutrons/cm -sec 12 2 Zone II = 1.31 X 10 neutrons/cm -sec 12 2 Zone II I = 5.06 X 10 neutrons/cm -sec 12 2 Zone IV = 1.09 X 10 neutrons/cm -sec Inte9rated Fast Flux:

11 2 8 Zone I = (8.16 X 10 n/cm -sec)(3.78 X 10 sec) 20

= 3.08 X 10 nvt 12 2 8 Zone II = (1.31 X 10 n/cm -sec)(3.78 X 10 sec) 20

= 4.95 X 10 nvt 12 2 8 Zone III = (5.06 X 10 n/cm -sec)(3.78 x 10 sec) 21

= 1.91 X 10 nvt 12 2 8 Zone IV = (1.09 X 10 n/cm -sec)(3.78 x 10 sec) 20

= 4.12 X 10 nvt A-11

lt..

NUklC r* I I Qt- CeRo foTo CUT ~60NwD . O A D l'IIIIAL QPTl tjN CARD

---.J--! I 1 1 g Q Q l2. . _..IL .Q. - -" _g_ Q ..il. _JL .Q_Q_g__g_ -P--A~-P ..Q.._ .Q.. . Q. _g _ .i _ ~ _ _ _ _ _ _ _ __

la PE Cl Fl CA Tl.INS 37 RO*S ~O CCILS

• GkPS U C014PS 0 NUC:i LoT,Rod BNO I 0 0 0 EPI ,.ooooe-o* NORN FAC a.ooooooE 17 IHIS CASE HAS H-Z GEONlaTRY, . - .. ,-- ~..,... -- -** ..

~NQ JS AN EIGfNVALUE CALCo

--**Ji:ilQN~5CURCE CtillKl o.1soo 0.2500 o.o o.*o-* ----- -* -- -- -*------------*---------- -----* -- --------------

MESH SPcClrlCATl~~s I DELTA I 7o6ZII

• 2o5*Q J* 7oQ2Q 37 I

- - - . . . l .. DllU I l

• 270 3 o.::.oa 0.112 42 o. o,u **

lt*.. * . I D1sr.

~ a II

.JO laOZQ

• 0*6"0 6lo!5i:IO 12 21

.a "

.ll 22 Zi 860360 JO JI J our.

2 Oe6l5 l

~7,264 12 j! 1*

j*ns I 020 JO 38 16 ** 02 J'il

• 7 llle&ll *a

~SI JQ, I H1 ~7

_t:11!~.H-~J~_. S.~------------- - - -

.::C:=- ( H ~ ~ 1/2

r-=- ~ J; ~

I :: *:

...

.__ ___ HU~R

!i'-014P 51 J5IIOF l fLUll C,Ro 52 10

( > , i!.1 ('(\,s.,,J) f'

• T

• SJ aur ~qa101P 1

5

. 04g 55 flN~&.

56

--*~ *-: 'T.--*-;*r** .:.. *: - * *

• 57 I ',

I 58 59-- --60 I 0.0 ilo 0 0.0

• 0,0 OoO Oo0 0o0 OoO 0o0 00 0 2 l,1Hi11Etl 1 'ilo 9o7Etl 0 7,0J Et10 3o2d<<sE 051 0

9,72ijEtlQ **Q21Et-10 3,2"0Et11) lo'il 711;~-~*q (,*~*~E .. JO Oo0}--l,ot.~'1/~-l--v'J..'

~ ,76i!! ll **J'il2Etll **~alEtll J,2 :U:tll ~**"*e:*n l . , IIU! JO I

o8JUa 12 lool2Et12 1, 59,9Ei--l 2 - l ;-i;HE+ i 2 2~2~61;t_l_1 .

7o537Etl , .. *- ~7f9E~l

,d*~~~*-~-.

f-i;i11*e:+11

,-~,mitlO 0,9 lo &4TE~c-2*.*1a9e*+10--o;-*~-----------

s 2oo97Etl2 2o3it0Ef'l2 2o2d0Etl2 l o537t:t-12 \lo 51291: l l 6ol07E ll 3,4201: ll lo 4 Et-ll 3o .. 8\1Ef' 10 0.0 0 lo784Et12 Jo2L19Et&2 J,110Etl2 2o0091;tl 2 lo265C:tl2 7.65;:E-t l1 *,iJ9e .. , 1 &,775Et1' 4,25ijE IO 0,0 I 5o00laftl2 4to21 BE+l 2 ** 0121:. 12 ~-5431: 12 I ol;l67E l2 ~,3-.li;f-l 1 !j,li9Etll ti!* UJEtl I 15, 01;121; 40 o,q

---

d 6o467Etl2 6o.J25Etl2 5,129Etl2 3.12ai: 12 --l,d'il6Etl2 .. l,llliE l2 6o085ftll -2;!)"&6ft-ll 5e96!>>E IO *0~0 ------*- ---

9 7o927Etl2 Oo4 11lf+l 2 1

~,24Jft12 lo75LIEtl.:! 2o2 6ftl2 l o3l2Et12 7 o0515E l l 2,920E l l Ood~ C:t-10 o.o Ill 9o4.?l.ll:t12 1. o~Jf.t 12 7,J..,~Et12 **'-00ftl2 2o612iatll 1,lilbEtli! tl,150E ll l,J IE ll /.8~1Et10 o. 0 u lo096E+1J do920ftl2 d,6toftl2 ~.ilo~Etl2 2.SUi'ilf l2 1, 72 E l2 'ilo2.JJ~ II 3e77;1E I l 8.8S1Etl0 0.0 12 lo.:!61EtlJ lo0l'iolC:+1J 0

1o1,7tHietl.? So751-E l.! j . 3721; 12

.-; ~ 61;- -l 2 *-* lo033Et12 4.21.JE ll 9ot170E+l0 0. ---- ------ --

13 lo*l2i:tlJ lol*7Ef'lJ l,102E ll Oo43dE l2 3,750Etl2 2o UtlEt-12 lol JE+l2 o655Et-ll l oOEl'ilE ll o.o l4 l ofl1St.t13 lo277f+ll lo225Et13 7olldEt12 4;1~tt;tU 2ol5~!it12 le25~1itl.; ~* 0'i151;t-U 1.1921:+11 o.o lS lo7Jdftll lo .. ll5Et1J lol47Etll 7,770Etl2 o49Q~+l2 ~.557E+12 1.~57Et12 ij,5J ~tll lo295E II 0o0

---

lo 1 od',lt>ftll l

• 620f t 1 J loito2Et1J do'J91Etl'2 4o.,8*9E U 2,7:'71: 12 lo459E l2 5o975Etll 1.3971:: ll OoO 11 .!oOJbl!tlJ 1,bJlf.tll 1,~hlftll d,~41~tlJ 6 1 162f~li 2,~24dt&a lo6aijft12 oo*J6ftll l,~0Uftll 0o0

t::-

I

___c...i.._

• ~

Iii 20 21 22 2ol!IOl!tlJ lo237Etll 2~JO*Etl3 2.J,at:t 13 2035.lctl.J lo 7101! t I J lo761tftll l.8101! .. 13 Io dl7Etl J 1, 8*31:tl 3 loOJJftlJ hOdlEtU l .722Ei'l3 le74df l3 lo76ZE+ll

"*'tl2ft12

~.7271Hl2 9o'ilit\1Etl 2 lo001E ll lo007iaf'l.j 0

t'l,*Jl~tL2

~ofl2Dlal2 5-. 7*6E--i2 5.EI07E l2 5o7~dE 12 J1'~11**en~

Jol'i13et~i!

3.2'il8E l2 J,2~2Et-l2 l,UQE &a h100Et&a lo170E+l2 lolb7E+l2 ti,V5~~ .. Jl 1o31t41:tl,I 7o755E ll 7o75~E~Jr 1;4qOEtl&

1~6~:J,f; ll 0e0 lo7b11E ll l.771E+ll 0o0

. -3;-2ose.-,2- i ;1,oe -12*- *1;637E+Jl--r;'r40*e-.=i1- **o;o _ _ _ _ _ _ _ _ _ _ _ __

OoO OeO 2J 2.J.Jll;tll I ,8231; & J 1,7J*EtlJ 9o94t21:tl2 5e71~tl2 le2 ~t1i 1,7.1Et12 7o6i7~ JI lo7 8E+ll 0,0 2* 2o275ff'1J lo779Etll loo<alEtlJ 9eot1.JC:tll 5~5o2E+l2 *--J,f5JE -&.692E l2***1;*-lt>>E--iC-1;101e:t=11 o.o 25 2oldoEtl.J lo70'ioEtlJ 1.0.?~c ll ~.2tldE l2 5 0 .Jjlftl2 3o021E+l2 lo021Etl2 7el25E ll loblOE+II OeO

--*-' --- ,_ ' -- ----~---*-

NUkR 1*1101' Coke lb t'

I

. GROUP fL.Ul(

r.i"" - 52 -60 26 a.oe.51:! .. 13 lo61-SEUJ lo533E .. IJ 8e760E .. l~ 5o02ciE+l2 2o84dE l2 le528f l2 6*719Etll l o538Etll 0-.0 n lo911E03 I ~.'il2EUJ I,* 18E -13, 13e.lQ21: l2 0

• • 650EU2 2e6Jol!Ua 1'! 161:tJ~-~* &~.eta l 1.*aoEtll o.o

~* h72bl:t-U I eli7EU3 1.aaJ~t12 ~.6501:+11 lo296E+l& Oe~

~ - - * -"T -

29 lo508E .. 13 lol7~ .. ll f. 1 soe + 11 - -o*;-o-- - - -

30 lo26lf .. 13

.Ji 9*i9dft12 7,93&etl2 7o573E+J2 p4 6iatli z.5~1;+12 &~4~Jij**~ s.O01e+r1 -~,~,6E+1r* s.a**euo a.o

,12 1 ** Qift-12 5o977Et12 5.723Et12 3* 0JE+la loii~f .. Ja 1,1,J~*Ja ~*IWJC l~ **7t~~ll 6,644lil l0 OeO 33* !i, i 09E l2 4* l 7 &Et l2 . *~ 002E+ 12**. 2. 30*~ 12 ..f.-J SJE .. .-z--,.~*21:..-.-.-.-;-3 i6E+i ,-,-;-9541:.:n .-;-s9se 1O -o-;-o

H lo2-Uf l2 20620£+12 2o4~9Etl2 lo409Et12 8ol 2E+ll 4o6C15E .. ll 0

)!e!a53EU l lo232E+ll le ot!tlO OoO

=-~ tie UJE .. 1 I 7ol60c II Cled ~~tll lod20F+II 2el7~E II le2ldl; II Ci,~~~l;UO J~~&~tlO lelO@EtlQ OeO 5!~,,etlQ

-- :IQ J7 I oddciEtl I 0.0 lo6l0f II 0.0 1,572Etll o.a

~e23~ 1Q o.o o.o

~* 01,1e~10 o.o J,HP,ijt~~

o-;-o

~~,~~~*ff le0~4E- O~

o-;-o OeQ

- o;-o

' ' *

• I~ * :* *' '

• 0 1 IJ~~- _j_, 00 J_l=!:,.!,~!.::l:!o!..---....!ll.!.l:!..U:e..::.-1-.!.l'Z.:..:.-_.::'3~-"U\=,!!l:..:e..::..:.....i:.:'2.:...*-1!.'1.!..1:,:C>:.:;,;:.'*_t-;,.:'2.:,._ *. ; . *~~.:.;.*h~*~*;i.:~:::.**~,;.,l:.;1_,! - - - - - - - - - - - - - - - - -

- -- - . -*- *-- - -- - - - -- . --*----- ~

V. EQUIVALENT MWD/AT (USING DESIGN DATA CONVERSION)

Zone I = 3.08 x 10 20 nvt

= 3,080 MWD/AT 1 X 10 nvt/(MWD/AT)

Zone II = 4.95 x 10 20 nvt = 4,950 MWD/AT 1 x 1017 nvt/(MWD/AT)

Zone III 1. 91 x 1021 nvt

= = 19,100 MWD/AT 1 x 10 17 nvt/(MWD/AT)

• Zone III Zone IV Saturation

=

=

13 x 10 20 nvt 1 x 1017 nvt/(MWD/AT) 4.12 x 1020 nvt

= 13,000 MWD/AT

= 4,120 MWD/AT 2

1 x 10 17 nvt/(MWD/AT)

I A-15

VI. CALCULATING THE AVERAGE STORED ENERGY PER ZONE

1) USING WINDSCALE DATA AND CURVES (Fig. 12.2, 155°, 205°, 255° Curves) 3

2) USING HANFORD DATA AND CURVES (Fig. 12.2, 30° Curve) 4

-kE and applying the equations S = S (1 - e )

00

-TI f3 and S(T) = a e from Nightingale's Nuclear Graphite, pp 329-331 *

• A. Method 1 (using Windscale Data and Curves for 150° and 250°)

Zone I a) S 30

= 685 [1 - e-(* 525 )( 3.0 8)J S = 685 [.802] = 549.5 30 *

-30/71.2 b) S ( 30 ) = a e 549.5 = a (.656) a = 837 .6

-83.5/71.2 c) S(83.5) = a e S(83!5) = 837.6 (.3095) = 259.2 cal/gram A-16

[ -(.526)(4.95)

Zone II a) S = 685 1 - e J 30 S = 685 (.926) = 634.3 30

-30/71.2 b) S{30) =a e 634.3 = a (.656)

\

a = 966.9

-83.5/71.2 c) S(83.5) =a e

• Zone III a)

S(83.5)

S 250

=

= 966.9 (.3095) 200 [ 1 - e

= 299.3

-(.151)(19.1)

]

cal/gram S = 200 [.944] = 188.8 cal/gram 250

-(.242)(4.12)

Zone IV a) S = 600 [ 1 - e J 150 S = 600 [.631] = 378.6 cal/gram 150

• Volume Weighted Average Stored Energy Per Wedge (Method 1)

Zone I Zone II

=

=

259.2 cal/gram x .156 ::: 40.4 299.3 cal/gram x .156 = 46.7 Zone III = 188.8 cal/gram x .379 = 71.6 Zone IV  ::: 378.6 cal/gram x .309 = 117.0 275.7 cal/gram A-17

B. Method 2 (using Hanford Data and Curves)

Zone I a) Same as Method 1 S(83.5) = 259.2 cal/gram Zone II a) Same as Method 1 S(83.5) = 299.3 cal/gram

• Zone III a) S = 685 [1 - e S

30 30

685 [1]

-(.526)(19.1) 685

]

-30/71.2 b) S(30) = a e 685 = a ( .656) a = 1044.2

• c) S(250) = a e

-250/71.2 S(250) = 1044.2 (.0298) = 31.2 cal/gram A-18

-(.526)(4.12)]

Zone IV a) S = 685 [ 1 - e 30 S = 685 [.885] = 606.5 30

-30/71.2 b) S(30) = a e 606.5 = a ( .656) a = 924.5

-150/71.2 c) S(150) = a e

• S(150) = 924.5 (.1216) = 112.4 cal/gram Volume Weighted Average Stored Energy Per Wedge (Method 2)

Zone I = 259.2 cal/gram x .156 = 40.4 Zone II = 299.3 cal/gram x .156 = 46.7 Zone III = 31.2 cal/gram x .379 = 11.8 Zone IV = 112.4 cal /gram x .309 = 34.7 133.6 cal/gram A-19

APPENDIX A REFERENCES

1. E. Robert Schmidt, "Design Analysis of the Graphite Reflector Elements," Missouri University Research Reactor Design Data, Vol. I, p. 18.

2. W. K. Woods, L. P. Bupp and J. F. Fletcher, "Irradiation Damage to Artifical Graphite," Proceedin s of the International Conference on the Peaceful Uses of Atomic Energy , p. 46.

3. R. E. Nightingale, ed. Nuclear Graphite (New York:Academic Press, 1962), P. 329.

4. ibid, p. 329

• A-20

• APPENDIX B THERMAL COLUMN I. AVERAGE FAST FLUX DISTRIBUTION (>.821 Mev)

Derived from Missouri University Research Reactor Design Data, Vol. I, "Radiation Heating in the University of Mimssouri Research Reactor."

A. FRONT FACE:

10 2 2 x 10 n/cm -sec {65 cm from reactor core centerline)

B. FIVE CENTIMETERS FROM FRONT FACE:

10 2 1 x 10 n/cm -sec {70 cm from reactor core centerline)

• C. TEN CENTIMETERS FROM FRONT FACE:

9 2 6 x 10 n/cm -sec {75 cm from reactor core centerline)

D. THIRTY CENTIMETERS FROM FRONT FACE:

8 2 7 x 10 n/cm -sec {95 cm from reactor core centerline)

E. MID PLANE:

6 2 6 X 10 n/cm -sec {141 cm from reactor core centerline)

• F. BACK FACE:

1 X 10 4 2 n/cm -sec {217 cm from reactor core centerline)

B-1

II. EXPOSURE CALCULATIONS A. FRONT FACE:

10 2 8 18

{2 x 10 n/cm -sec){3.78 x 10 sec) = 7.56 x 10 nvt B. FIVE CENTIMETERS FROM FRONT FACE:

10 2 8 18

{l x 10 n/cm -sec){3.78 x 10 sec) = 3.78 x 10 nvt

c. TEN CENTIMETERS FROM FRONT FACE:

9 2 8 18 (6 X 10 n/cm -sec)(3.78 x 10 sec) = 2.27 X 10 nvt D* THIRTY CENTIMETERS FROM FRONT FACE:

• E. MID PLANE:

(7 X 10 8

6 2

2 8

n/cm -sec)(3.78 x 10 sec) 8

= 2.65 X 10 17 15 nvt (6 X 10 n/cm -sec)(3.78 x 10 sec) = 2.27 X 10 nvt F. BACK FACE:

4 2 8 12 (1 X 10 n/cm -sec}(3.78 x 10 sec) = 3.78 X 10 nvt B-2

III. MWD/AT EQUIVALENCY A. FRONT FACE:

7.56 X 1018 nvt = 75.6 Mwd/At 17 1 x 10 nvt/(Mwd/At)

B. FIVE CENTIMETERS FROM FRONT FACE:

3. 78 X 1018 nvt

= 37.8 Mwd/At 17 1 x 10 nvt/(Mwd/At)

C. TEN CENTIMETERS FROM FRONT FACE:

• 2. 77 x 1018 nvt 17 1 x 10 nvt/(Mwd/At)

= 22.7 Mwd/At D. THIRTY CENTIMETERS FROM FRONT FACE:

2.65 X 1017 nvt

= 2.65 Mwd/At 17 1 x 10 nvt/(Mwd/At)

E. MID PLANE:

2. 77 x 1015 nvt

= .023 Mwd/At 1 x 1017 nvt/(Mwd/At)

F. BACK FACE:

3. 78 X 1012 nvt -5

= 3.7 x 10 Mwd/At 1 x 1017 nvt/(Mwd/At) 8-3

IV'. TEMPERATURE Temperatures within the .thermal column have been measured by thermocouple and all points exceed 100°F (37.8°C). For conservatism, the graphite temperature will be assumed to be 30°C.

V. CALCULATION OF STORED ENERGY A. FRONT FACE:

-kE S = S a,

[1 - e J

-(.526)(.0756)

S = 685 [1 - e J

• B.

S = 685 [3.90 x 10 FIVE CENTIMETERS FROM FRONT FACE:

-2 J = 26.7 cal/gram

-(.526)(.0378)

S = 685 [1 - e J

-2 S = 685 [1.97 x 10 J = 13.5 cal/gram C. TEN CENTIMETERS FROM FRONT FACE:

• S S

=

=

685 [1 - e

-( .526) ( .0227}

685 [1.19 x 10- J 2

=

J 8.1 cal/gram D. THIRTY CENTIMETERS FROM FRONT FACE:

-(.526)(2.65 X 10- 3 )

S = 685 [1 - e J

-3 S = 685 [1.39 x 10 J = .954 cal/gram B-4

E. MID PLANE:

-(.526)(2.3 X 10- 5 )

S : ,: 685 [1 - e J S = 685 [1.21 X l0- 5 ] = .008 cal/gram F. BACK FACE (negligible)

VI. THERMAL COLUMN GRAPHITE VOLUME ANO MASS

• Back Section:

~ront Section:

48 inches x 48 inches x 48 inches Total

=

37.5 inches x 37.5 inches x 12.25 inches ~ 17,226.56 in 3 110,592.00 in 3 127,818.56 in 3 cm) 3 =

127 ' 818

• 56 i n 3 x (2.54 (l i n)3 Total Vo*lume + 127,818.56 in 3 x 16.387 1'n3 cm 3 = 2.095 x 10 6 cm 3 Total Mass + {2.095 x 10~ cm 3 )(1.6 g/cm 3 ) = 3.35 x 10 6 grams {"" 7400 lbs)

B-5

l I

I

. *. .,. c::::

!"'1

-:-~~-- ~-

1

- l'c- * -: ~

• re - F : ;: - ':"'~ - ~ ':..*-

c:. '"'  :; '= ~ ;~-' : .~ ,:c ~: =*: ,~ r.=.,:: '"

,,~~~~-~~=~~~f ~- ~ '-' *-:. .:

-u QI a:

l

\*

II.

~.:L:-;-:- -  ; I=: ~ * =-~; ,,. . c:= ;;: ::::

~

~==*..-.:--~~~-

-*-1-- *- - . r--,-

N eu -r- ~ - - ~ .:.,~--

I ll

~

0

~

~

111 s=

...., '- - , - II II J, ~ ti - J I, ~ ..r -,

.- --  :

• 3 j ;..,_ - -

- r~. ..: ..

~

• -*=--'7

- =:

1-, ,_ ,_

1 f-1-~-

. L- - '\" - f - t:~,:~~:~

-~ ,_ ,_ -

. 1 1 - ,-: -

- -=- : -~ - - :

..... _. - c.: ~ 1 - I-:- :: "-: -= ~ i.:: ~ ,- .: C -

~

,=-~- 11

,_,.. - ~--~1--~=~=-:=~--~*~*--->-

-~=~ =-, -*-~-~- -- .::

- it-,,-- --\

~-

-- >-...- - - - \ *-~ ~ ___ .._Ht*,~++++++++~~+~1--

10 8 I - ..... ~ - -*- --

)0 1tO -io 'ID 1 )0 1 ,o 20)

Distance from R.Ea.c-r.or C,c,rE. C.E.nterline (cm)

B-6

108 ._....,....,,........._......,..,_._..,...,_.__._.__.......,__.._.__......,,...,...,,__,,_.,,__,......_...,_.,._.......,.........,..........._.....,...,..,'1

'l ' '

I\

• k

' -**--------. - ~- - -

--_.... ______ -- *....::t- - -

r-- - i-- -- - ..* l _

-- ...., _....,,..,,_,,.... ..... r - - * -

1-r--- - -~

.._..,._..,._t-t-l-l--t-t-t-t-*t-l-l-HiHHl-1, * - 1-- i--- +-f-,1-i~ I -+-*+--t-+-t-+-*+-.... 1-t-Ht-l~ll-f

\ '

u 0

\ '

at .-_..,._ -

~ -

e

--** -r-* .---

• - - - - - - 1-- - - - - -

'Ill

-r-

- f--*-

f-- ~ - - - - - ~ - - --- t- \" - *- - .__ - - *~ 1,....- ~ - -

C 1--~--~ *-* 1- ~ --

0 u 10 6 t-1-1-t-t-HHH~- -~=,:~-,-*~-~-*------~---~-~=~~*+*-+-+*-~~~.~~~:~I~~::!:!:~-+--~=-==

, 1 \

~

- - -- ..: -=. ...:

~ " '1 C:

0 u

1-----=~7.:..:

a, r - .: -- _- ~...:

z 5

10 C ' 1

\

I F ' .:

Po- - --

1--1---,-

4 -*-

lO loo l 0 200 'Z.td 2-'Zo 140 Di tance from Reactor Core Cente~line (cm)

B-7

APPENDIX C ANALYSIS OF THE HAZARDS ASSOCIATED WITH STORED ENERGY RELEASE IN THE MURR GRAPHITE REFLECTOR I

• CALCULATING TOTAL STORED ENERGY Calculated (Theoretical) Mass Total Stored Avera9e Stored Energy p* Wed9es) Energy

\

HIGH 275.7 cal/gram X 377,843 grams = 1.04 x 10 8 cal MID ~ 205 cal/gram X 377,843 grams = 7.75 x 10 7 cal

/LOW

._ 133.6 cal/gram X 377,843 grams C 5.05 x 10 7 cal

• II. ENERGY RELEASE.FRACTION The amount of total stored energy releasable in an anneal to 1000°C is 50%

to 80% of total stored energy. 1 For conservatism a 100% release of energy will be assumed.

III. CONSIDERATIONS FOR TRANSFERRING GRAPHITE ENERGY TO THE CORE The graphite reflector forms 2/3 of a right circular annular cylinder sur-

• rounding the core and is separated from the core by the materials listed in Figure C-1.

Considering the total stored energy calculated in Step I, the amount of heat that can be transferred to the core depends on the energy conducted in the core direction.

Area of inner surface= 2n(9.684)(34.89) = 2123 in 2 Area of outer surface= 2n(18.262)(34.89) = 4003 in 2 Total heat transfer area 6126 in 2 Ratio of inner surface area to total heat transfer area = 6126 2123 = .346 C-1

A\..u...M ,,.J v. M RHLc;o:vL Fv.E.L cS~f\P}-llfE-Kf:fL..C.mC

~.I:,,,)

r 1MIL c,.o")

'\. '\_

"'\./

/ /

/ .

/

-, /

'\

- ' >-)'\,

/

• ). .r

• ,.. /*

/

.. \ . 'y /.___ /

,, '*"' ~ > / .

'""*'\,\

• ',,/ "' .

/

.:\. '\. _, / >-

. .,_ "' ,, " . , X

• ).,, /\...._ .

• <.,,\*,.,/ >/ Y""- , ,

/ '.

) ( *,

t Wl'.\~'i.L GPI

(_, 11. - )

W1JT~

(, 15"

,1, 1

')

Figure C-1 For conservatism, a fraction of .40 will be used to determine the total amount of energy in Step I) directed toward the core. This is conservative for the following reasons:

1) Heat dissipation in the intervening materials is not considered.

2) Heat conduction in the vertical direction is not considered.

3) The heat transfer area of the outer reflector surface is twice the area of the inner surface.

4) The distribution of highest stored energy is towards the outside of the graphite element (away from the core).

5) The ~T, the driving force for heat conduction, is much higher towards the bulk pool area than towards the core.

C-2

.ENERGY DISSIPATEO TOWARDS CORE

. HIGH

• 1.04 x 10 8 cal x .40 c 4. Hi x 10 7 cal MIO 7.75 x 10 7 cal x .40 = 3.10 x 10 7 cal LOW 5.05 x 10 7 cal x .40 = 2.02 x 10 7 cal IV. CONVERSION OF ENERGY TO.BTU HIGH 4.16 X 10 7 cal X 4.1819 = 1.74 X 10 8 Joules MID 3.10 X 10 7 cal X 4.1819 = 1.30 X 10 8 Joules LOW 2.02 X 10 7 cal X 4.1819 = 8.45 x 10 7 Joules

-** HIGH MID LOW 1.74 X 10 8 Joules x (9.47086 X 10-'+)

1.30 X 10 8 Joules x (9.47086 X 10-'+)

8.45 X 10 7 Joules x (9.47086 X

= 1.65

= 1.23 X

X 10 5 Btu 10 5 Btu 10-'+) = 8.00 X 10 4 Btu V. CONVERTING TOTAL ENERGY TO ENERGY RELEASE RATE Stored energy is released over a period of time varying from 20 min to 60 min 2 in totally adiabatic laboratory conditions to 3 to 6 hours 3 in actual reactor annealing releases of energy.

ENERGY RELEASE TIME n n

• ln HIGH [1.65 x 10 5 Btu] 45.8 Btu/sec 91.7 Btu/sec 183.3 Btu/sec 367.7 Btu/sec MID [1.23 x 10 5 Btu] 34.2 Btu/sec 68.3 Btu/sec 136.7 Btu/sec 273.3 Btu/sec LOW [8.00 x 10 4 Btu] 22.2 Btu/sec 44.4 Btu/sec 88.9- Btu/sec 177.8 Btu/sec C-3

VI. COMPARING GRAPHITE ENERGY RELEASE RATES TO.CORE DECAY HEAT Fission product energy release rate from. a reactor which has heen operated to near equilibrium fission product concentrations is equal to approximately 6 percent of operating power. 4 Thus the MURR core will produce decay heat at a rate of 0.6 MW which is equal to 568.7 Btu/sec. Even at the most conserva-tive estimate of energy release time (7.5 minutes), the energy release rate from the graphite would be less than that of core decay heat.

VII. COMPARING THE CONSEQUENCES OF A GRAPHITE ENERGY RELEASE TO THE Loss OF COOLANT ACCIDENT ANALYSIS FOR MURR The Loss of Coolant Accident (LOCA) is the most severe credible accident for

• MURR (a double-ended rupture of the reactor inlet pipe between 507B and the pool liner). Analys'is of this accident is contained in MURR Hazards Summary Report, Addendum 4, Appendix E. Considering the core decay heat as the energy source to be dissipated, the results of this analysis show that DNB and fuel damage will not result from this accident. In fact, results of the analysis show that fuel cladding surface temperature would not exceed 281°F (with a starting temperature of 165°F). This accident results in a 116°F (46.7°C) increase in the cladding temperature of the fuel. This temperature is well

• below 1184°F {640°C), the temperature at which fission product release from UAlx fuel is appreciable.

Since the most conservative estimate of heat load from the graphite is less than the heat load (decay heat) considered for the LOCA analysis, the con-sequences of the release of graphite energy to the core would be less severe.

C-4

MAXIMUM ADIABATIC TEMPERATURE EXCURSION IN GRAPHITE REFLECTOR

, . 3 2 6 3 1000 Jl000 C = 1.86 T + ,s.34 x 10- T ) _ ,5.27 x 10- T )

200 p 1200 Jl000 C 1 g-rnole 200 p = 3,748.25 cal/g-mole _ oc x 12g

= 312.35 cal/ _ oc 9

Ave C

= 200 110eop _ 1 over ~his SOO - 312.35 cal/g _ oc X °8()0 = .390 cal/

9

_ oc temperature range

• HIGH MAXIMUM ADIABATIC TEMPERATURE RISE 275 cal/ gram +

275 cal/~ram

.390 = 705°c 205 cal/

HIGH ~ram = 52s c 0 205 cal/gram + .390 133 cal/gram LOW 133 cal/gram + .390 = 341°c C-5

APPENDIX C REFERENCES

1. R. E. Nightingale, J. M. Davidson and W. A. Snyder, "Damage to *Graphite Irradiated up to 1000°G," Proceedin Nations International Conference on The Peaceful Uses o tom c , o. 7, p. a e 1 *

2. same as 17 (main report)

3. same as 18 (main report)

4. S. Glasstone and A. Sesonske, Nuclear Reactor Engineering 3rd ed. (New York:Van Nostrand Reinhold Co., 1967), p. 124 .

C-6

APPENDIX D ANALYSIS OF HAZARD ASSOCIATED WITH STORtD ENERGY RELEASE IN THERMAL COLIJMN ,

The maximum stored energy at the front face of the thermal column is 26.7 cal/gram. The stored energy falls off to 13.5 cal/gram five centimeters into the front graphite section and less than 1 cal/gram at the back of the front graph-ite section (30 centimeters from front face). The maximum stored energy could con-tribute to a maximum adiabatic temperature increase of 128°C. This would correspond to a maximum temperature at the front face of 158°C under adiabatic conditions. Even this exaggerated temperature at the front face is incapable of autoignition of

• graphite.

Five centimeters into the graphite stack the maximuM adiabatic temperature increase would be 65°C. The maximun adiabatic temperature increase falls off even more drastically 30 centimeters from the front face (4.8°C rise). The back section of graphite (48 inch cube) has negligible stored energy and a maximum adiabatic temperature rise of less than 4.8°C.

CALCULATIONS:

c CP ~ 1.86 + (8.34 x 10- 3 T) - (5.27 x 10- 6 T2 cal/g-mole - 0 f BC = 1.86T + (8.34 x 10-3 T2)

A p 130 cP = (241.8 + 70.473 - 3.859) - (55.8 + 3.753 - o.47) 30 J

~ 308.414 - 59.506 = 248.908 cal/g-Mole - °C X 1 r2m~le

= 20.74 cal/g - °C D-1

Ave Cp f13oc

= 30 100 p "'

• 20 *.74 . 1 / g - °C x Too

• ca l = .2074 cal/g - °C over this temperature range FRONT FACE:

Max adiabatic temp increase = 26.7 cal/gram =

128.7°C

.2074 cal/g - °C

• 5 CENTIMETERS FROM FRONT FACE:

Max adiabatic temp increase = 13.5 cal/gram = 65.1°C

.2074 cal/g - °C 30 CENTIMETERS FROM FRONT FACE:

Max adiabatic temp increase = 1 cal /~ram = 4.8°C

.2074 calg - °C D-2

U v l,l'\C:.. I l'IUIVIDC."'

i?E. ULE PRM Jd:-44 {iJ)

{..flM .P.J4J)

DOCKETED

!I C I'

.al------------------GATechnologies - - - - - - - - - - * * * - - - - - - .

GA Technologies Inc. *s7 FEB - 2 P2 :30 P.O. BOX 85608 SAN DIEGO, CALIFORNIA 92138 January 28, 1987 (619) 455-3000 GEN-1010 Secretary U.S. Nuclear Regu latory Commission Washington, D.C. 20555 Attention: Correspondence and Records Branch

Subject:

Docket No. PRM-50-44; Comments on Petition for Rulemaklng -

Graphite In Non-Power Reactors

Dear Sir:

Contained herein are GA Technologies lnc.'s (GA) comments on the petition for rulemakfng flied by the Committee to Bridge the Gap CCBG) ,

Docket No. PRM-50-44. Comments f n th fs Ietter are cont l ned to the requested rulemaklng as It relates to graphite fn TRIGA research reactors. Comments relative to the high temperature gas-cooled reactor CHTGR) are being provided under separate cover.

The petition requests that the NRC amend Its regulations to require operators of reactors that use graphite as a moderator or reflector

( 1) to prepare and submit for NRC approval f I re response p Ians and evacuation plans for a graphite fire, and (2) to measure the energy stored fn their graphite and revise t hef r safety analyses to consider the risks and consequences of a graphite fire In their facf lltles.

The basis for the petition ls CBG's contention that the accident at Chernobyl shows graphite ff res to be credible events for which I fcensees do not have adequate ff re response and emergency p Ians. CBG a I so contends, based on their Interpretation of a recent report by researchers at UQ.A, that WI gner energy stored f n graph lte has been severely underestimated In earl ler studies.

GA bel Jeves that these proposed requl rements are unnecessary and, Jf adopted, would have an unduly burdensome Impact on licensees with TRIGA reactors, with no corresponding Increase in pub I le health and safety.

With regard to Wigner energy storage, discussions between the staffs at GA and UQ.A Indicate that CBG has made very selective use of the data In the UQ.A paper to support its content ion that Wigner energy storage has been severely underestimated. The high value of Wigner energy storage referred to by CBG was a single data point which was not representative of the behav tor of the graph lte as a who Ie, or of the tote I stored energy In the core. The complete data presented In the paper referred FEB 2 1987 nowledged by card .******.* .-.,..,.-. ,-"T.

10955 JOHN JAY HOPKINS DR. , SAN DIEGO, CALIFORNIA 92121

PI i i

["r- ,.

p

~11.2,.

/ 1 vs,k~_

to by the petitioners support the results of the NRC-funded generic study on Wigner energy storage CNUREG/CR-2079). One of the authors of the UCLA paper has addressed these po I nts In deta I I In h Is comments on the CBG petition (Ref. 1).*

In the specific analyses for TRIGA reactors reviewed below, we wit I demonstrate that no graph lte f I res can resu It for any postu I ated reactor operation. We can, therefore, conclude that no Justification exists to revise existing safety analyses with regard to the potential consequences outlined In the petition for rulemaklng or to require contingency plans for graphite fires. Of course, It also fol lows that no need exists to obtain experimental measurements on stored energy In representative samples from operating reactors. For TRIGA reactors, such a measurement would Involve considerable additional risks related to Al.ARA considerations since handling of radioactive fuel Involves risks of personnel radiation exposure, and Hot Ce I I man I put atl*on of such fuel would Involve cuttl ng Into fuel elements with the consequent release of fission products.

Details of supporting calculatlons pertinent to this discussion are given In the attachment. A tab Ie of the max Imum re Ieasab Ie stored en.argy Is presented as a function of the graphite Irradiation temperature together with the associated temperature Increases should sufficiently elevated temperatures be reached (>120oC) to cause the energy to be released. The effects of this potential energy release from Irradiated graphite ls evaluated for the severl!II graphite components; namely, the graphite end reflectors which are In each TRIGA fuel element, the graphite radial reflector which Is In certain TRIGA facl I ltles, and finally the graphite dummy elements which are used In some TRIGA cores. The first two of these are vital reactor components since fuel elements contain fission products and the graphite reflector assembly Is a major structural component of the core. The graphite dummies are of lesser concern since they contain no fission products and only the minutest quantity of radioactivity.

The analyses described In the attachment can be summarized as fol lows:

(1) Graphite end reflectors - Steel clad and aluminum clad fuel survives al I normal, abnormal, and l!ICcldent (Including LOCA) operations at 250 KW and 1 MW.

(2) Graphite radial reflector - The temperature of the graphite In the radial reflector Is sufficiently high c~120°c> during operation at or above 1 MW that the releasable stored energy Is very smal I (~45 ca I/ gm)

• Even If th Is energy were ever to be re Ieased, the resulting temperature of the grap~lte-alumlnum clad Interface ls far below any value that endangers the aluminum clad. A separate

• See attached 11st of references.

analysts for 250 KW operation shows that the operating temperatures of the graphite always remain low (~OoC) with the result that up to the maximum releas21ble energy (260 cal/gm) m21y be aval Iable.

However, the actual maximum sotred energy wlll also be low In this case becaue of the relatlvely smal I Integrated flux from 250 KW operation. For oper21tlon at this power level, the accident conditions al so carry minima! consequene::es; even with loss of cool ant CLOCA), the maximum temperature of the graph lte-al uml num clad Interface Is only 70oC. The l21tter temperature ls f21r below that req u I red to In It I ate re Iease of WI gner energy. As a conse-quence, the graphite reflector assembly for 250 KW operation ls not at risk from the stored energy In graphite.

(3) Graphite dummy elements - The effect of releasable stored energy on the aluminum cl adding on graphite dummies Is Jneonsequentf21l In al I norm21l and abnormal operations of al I TRIGA reactors. For the loss of coolant CLOCA) with 250 KW and 1 MW TRIGA reactors, the aluminum clad on the graphite dummy elements wll I probably melt. However, as noted above, no graphite fire can result.

We conclude th21t In al I Instances affecting vital components of the TRIGA reactor (e.g., fuel elements and the gr21phfte radial reflector), the release of stored energy from grap~fte components causes only benign results. The possible loss of the 21lumfnum cladding on a few graphite dummy elements In the rare Instance of loss of coolant does not constitute a hazard to the fact I tty or to the pub I le from either radioactivity or graphite fire.

In view of the fact that no risk to the pub I Jc health and safety ext sts from the stored energy In the graphite In TRIGA reactors and since, as demonstrated herein, the petition as applied to TRIGA reactors ls not based on properly evaluated facts and ls without substance, we strongly recommend that the petition for rulemaklng be denied.

If you have questions regarding our comments, please contact me at (619) 455-2823 or Dr. Wil IIBm Whittemore at (619) 455-3277.

Very truly yours, Keith E. Asmussen, Manager Licensing, Safety, and Nuclear Comp!lance KEA/mk Attachment

January 28, 1987 ATTACHMENT TO GEN-1010

1. GENERAL At least as early as 1965, GA Technologies Inc. made analyses of the consequences of Wigner energy storage in the graphite contained in TRIGA reactors. The conclusion at that time was that the graphite reflector and its cladding were not at risk from the stored energy. Subsequent analyses have extended these evaluations and have confirmed the satisfactory findings for all graphite components in the TRIGA reactors.

A review of the available literature confirms that no purely graphite fire has occurred at any nuclear reactor. The fire initially observed at Chernobyl was caused by burning fuel and cladding ejected from the core in the accident, not burning graphite (Ref. 2). The major fission product releases were not associated with graphite oxidationj but with the initial steam explosion and with the buildup of decay heat energy trapped in the core by the ton~ of sand, boron, and dolomite used to smother the core (Ref. 3). Graphite "burning," to the extent that it occurred a*t Chernobyl, was a sec9ndary effec*t of the accident that had relatively little impact on offsite doses. Similar-ly, while graphite did burn at Windscale, again, hot uranium was responsible for burning the graphite. In TRIGA reactors, no such circumstances could exist. Even in the extreme case with unclad, hot uranium-zirconium hydride, this fuel will quench itself in water and, in the ab~ence of water, it will not burn in air. Hence, this fuel is not itself a source of heat sufficient to support burning graphite.

Recent tests have evaluated the burning of bulk graphite under the stimulus of acetylene torches. Although the graphite was heated to 1,0000 C and did in fact burn directly under the torches, when the acetylene was shut off leaving the oxygen now uninterruped, the graphite ceased burning and cooled down. A further example of the difficulty of burning graphite is illustrated by electric-arc lamps and electric-arc furnaces. Both are

important applications of graphite and illustrate the chemical stability of that material at very high temperatures under atmospheric conditions. In the electric-arc lamp, electricity is passed through the gap between graphite electrodes creating a source of light. In this application, the temperature of the electrode tip is typically about 3700°c, i.e., near the vaporization temperature of graphite. At this temperature, the consumption of a 0.5-inch diameter graphite electrode is about 3 inches per hour, half due to volatili-zation of the tip and half due to oxidation in the open air. When the electric current is interrupted, the electrodes immediately cool down and no

• self-sustained combustion of the graphite occurs, even after long operation at these very high temperatures (Ref. 4).

Extensive tests at Brookhaven National Laboratory (Ref. 5) have demon-strated that it is extremely difficult to burn graphite at 650°c and then only under carefully controlled conditions of ducted gas flow, graphite prepara-tion, and the correct mixtures of air, CO and co 2* Obviously, the special requirements to support the burning of graphite in a TRIGA reactor at or near 0

65D c are not credible. Further, as is shown below, no mechan.ism in TRIGA reactors related to stored energy in graphite can produce graphit~ tempera-

, tures significantly above 650°c. Thus, graphite fires associated with stored

• energy in TRIGA reactors are*not credible.

2. ENERGY RELEASE FRCM IRRADIATED GRAPHITE For ease in discussing the several cases of Wigner energy storage in graphite in TRIGA reactors, it will be convenient to summarize the pertinent parameters. The exact details of the processes involved are complicated and have been well summarized by Meyer (Ref. 6). A large compendium of useful results is contained in the work edited by Nightingale (Ref. 7). We will concern ourselves only with the part of the Wigner energy that can be released

- 0 0 since only this is active in rai~ing the graphite temperature to 650 -775 C.

The different amounts of releasable energy result in different adiabatic temperature rises that can be found in published curves. It will be

sufficient for our purposes to assume a constant specific heat C c-o.4 ca11°c) and calculate the temperature rise from the releasable energy.

Graphite irradiated to high neutron fluences (approximately 5 x 20 10 nvt) will yield a released energy Erel that depends on the temperature Ti of the graphite during irradiation (Ref. 8). E increases with nvt up r 20 re1 to nuences of 5 x 10 , at which point it remains constant (Ref. 7). In the 20 following table we present a useful summary of peak (i.e., for nvt = 5 x 10 )

releasable energy as a factor of graphite irradiation temperature. Also shown 0

is the peak adiabatic temperature rise AT calculated from Ere ;

1 c.

Table. Releasable Energy E and Peak Adiabatic 1

Temperature Rises ATeas a Function of Graphi~o Irradiation Temperature Tir (for nvt >5 x 10 )

0 Tir ( C) Erel (cal/gm) /jT (°C) 35 260 650 55 - 70 150 375 70-100 125 313 1 00-135 45 113

>135 26 65

3. GRAPHITE IRRADIATED IN TRIGA REACTORS Several graphite components are found in TRIGA reactors. Not all are found in all TRIGA reactors. Such components are:

(a) Top and bottom end reflectors in each TRIGA fuel element.

(b) Graphite reflector around core. A graphite thermal column acts as a partial core reflector in some TRIGA reactors.

(c) Graphite dummy elements used to fill core positions in some TRIGA cores.

The graphite in these locations has been analyzed for the concerns of releasable Wigner energy. Account is taken for the actual construction of the components, the types of cladding and appropriate flow paths for heat into and out of the affected component. In certain instances, the actual core location of the component has been considered.

3. 1 GRAPHITE IN TRIGA FUEL ELEMENTS F. Foushee (Ref. 9) evaluated the consequences of irradiating the top and bottom graphite reflector plugs in the fuel in a 1-MW TRIGA reactor. In all 1-MW TRIGA reactors, the fuel cladding is stainless steel. Although Tir for these graphite end reflectors in contact with hot fuel assemblies clearly exceeds 100°c, Foushee conservatively assumed the maximum releasable energy (i.e., E - 260 cal/gm). Even under these conditions, the maximum re 1 0 temperature excursion (/J,T -550 C) would be less than that produced in large TRIGA pulses for which the clad is not at risk.

After about 1971, all TRIGA fuel was manufactured containing graphite end reflectors with a diameter 2.2 mm smaller than the inside diamete~ of the clad. For 1 MW operation, these end plugs will have Tir in the range 100°-130°c. For the earlier fuel with the graphite essentially in contact with the clad, Tir is computed to be in the range 90°-100°c for which the maximum subsequent temperature excursion will not exceed 313°c. The above results are for steel clad fuel in TRIGA reactors with power levels of 1 to 1.5 MW.

The case for 250 KW operation has also been evaluated. Steel clad fuel cannot be at risk because, as shown above, even with release of the maximum amount of stored energy (260 cal/gm), the maximum temperatures would be less than routinely experienced in large TRIGA pulses. For aluminum clad fuel having no extra gap between the graphite end plugs and the clad, careful ev.aluation shows that the bottom graphite reflectors are most affected by the Wigner energy. In all these fuel elements, the bottom reflector contains

regions with operating temperatures well below 55°c. For this low power operation {<250 KW), the total fluence of fast neutrons (50 KeV-10 MEV) is

- 20 small in the graphite end reflector(~ 10 nvt after nearly 30 years of intermittent operation) and is as much as seven times less in the cooler region farthest from the fuel meat. The experimental work reported in Simmons (Ref. 10) demonstrates that the maximum stored energy will be less than 75 cal/gm everywhere within these graphite end pieces. Additional computer calculations for the loss of coolant accident (LOCA) show that some portion of each graphite end piece will exceed the release temperature of -120°c as a result of the hot, air-cooled fuel. However, in the resulting release of Wigner energy,. the maximum graphite temperature will not be high enough to melt aluminum. Since neither the air-cooled fuel nor the graphite end plugs reach the melting temperature for aluminum, the aluminum clad is not at risk.

Conclusion:

Both the older aluminum clad fuel operated at 250 KW and steel clad fuel in operation at either 250 KW or 1 MW have no safety problems from Wigner energy under normal, abnormal, or accident (LOCA) conditions.

3,2 Stored Energy in Graphite Reflector Foushee has evaluated the effects of Wigner energy on the radial graphite reflector (Ref. 11). In that work, one extreme example was considered in which the operation was specially adjusted to accumulate the largest possible releasable energy. An elevated graphite temperature was then postulated whereupon all the stored energy was assumed to be released (far greater than the now accepted 260 cal/gm). Even in this extreme case, the integrity of the aluminum clad of the reflector was maintained. For the more realistic case, the initial graphite reflector temperature (for 1 MW operation) is 90 + 30 =

120°c (see Fig. 9 Of Ref. 11). After about 1 MW-Year of operation, the graphite heat conductivity suffers sufficient deterioration (from radiatiorr damage) that T.

lr -

> 140 + 30 = 170°c. Well within the first MW-Year of operation, the graphite temperature will exceed 120°c at which the reflector

continuously releases its available energy. If the water cooling was lost (LOCA), the aluminum under-surface of the reflector cladding (half-inch thickness) will be air cooled to about 100°c. The possible contribution to the temperature rise from the energy in the graphite, should it be released, would be small, in the range 65°-113°c at most.

Operation of the TRIGA reactor at 250 KW produces lower temperatures in the graphite reflector. The initial Tir is estimated at about 22.5 +JO=

52.5°c (Fig. 9, Ref. 11). Only after 2.5 MW-Years (10 years of continuous operation at 250 KW) will Ti > 82.5°c. We assume conservatively that 260 r -

cal/gm will be released whenever the graphite temperature reaches= 120°c. No normal or abnormal operation with water in the reactor tank can result in this graphite temperature; consequently, none of the stored energy can be released.

Even during loss of coolant (LOCA) for the low power reactor, the maximum temperature of the aluminum clad for the reflector will not exceed 70°c. This temperature is far below the 120°c considered to be the minimum required to initiate release of stored energy. We conclude that the cladding integrity for the radial graphite reflector is not endangered even with loss of coolant.

Conclusion:

The graphite reflector and its clad for 1 MW as well as 250 KW operation have no safety problems due to energy stored in graphite under normal, abnormal, and accident (LOCA) conditions.

3.3 Graphite Dummy Elements Graphite dummy elements are used to fill some of the core positions near or at the edge of the reactor core. These have aluminum cladding and, in more recent times, have an increased gap between the graphite and its cladding.

Calculations show that even for the gapped units, operation at 1 MW is not sufficient to raise the graphite temperature during irradiation to 80°C.

Consequently, we must assume that the maximum releasable energy (e.g., 260 cal/gm) is available whenever the graphite temperature is elevated to at least 120°c. The total graphite temperature will then be 120 + 650 = 770°c. No

normal or abnormal operation will produce a temperature as high as the initiation temperature of 120°c. Even if it were produced, water cooling of the aluminum clad would preserve the integrity of the clad. Loss of water cooling presents a different situation. In LOCA conditions, we expect that the aluminum clad-graphite interface will reach 77rPc. Cooling with hot air at -200°c would not be sufficient to prevent melting of the 0.030-inch aluminum clad. The hot graphite at 770°c or less cannot burn because the specific requirements set forth in Ref. 5 (Part I, p. 43) for graphite temperatures in the range 6S0°c to 750°c cannot be met. Further, the graphite radiates its energy rapidly and quickly cools to the ambient air temperature c-200°c). It should be noted carefully that no release of radioactivity or fission products is involved.

Conclusion:

Melting of the aluminum clad on the graphite dummies can occur only in the extreme event of loss of coolant. Even in this unlikely event, no risk to the personnel at the facility or to the public results because there can be no graphite fire, and there is no measurable release of radioactivity.

REFERENCES

1. H. Pearlman letter to the Commissioners, "Committee to Bridge the Gap; Petition for Rulemaking," Docket No. PRM-50-44, October 29, 1986.

2. "Report of the u. S. Department of Energy's Team Analysis of the Chernobyl-4 Atomic Energy Station Accident Sequence," DOE/NE-0076, November, 1986.

3. "The Accident at the Chernobyl AES and Its Consequences," State Committee for Use of Atomic Energy in the USSR, August 1986. (Translated from the Russian, Department of Energy, NE-40, August 17, 1986.)

4. C. L. Mantell, Carbon and Graphite Handbook, Interscience Publishers, 1986, p. 349.

5. Donald G. Schweitzer et. al., "Oxidation and Heat Transfer Studies in graphite Channels," Nucl. Sci. and Eng. _g, 39-45 (1962) Part I; E, 45-50 (1967) Part II; _g, 51-58 (1962) Part III; _lg, 59-62 (1962) Part IV.

Donald G. Schweitzer, "Thermal Properties of Air-Cooled Graphite Channels, 11 Nucl. Sci. and Eng. J_, 275-282 (1962).

6. Walt A. Meyer, Jr., "Stored Energy in Irradiated Graphite," a private report from the University of Missouri Research Reactor Facility, December 1 O, 1986.

7. R. E. Nightingale, ed. "Nuclear Graphite," Academic Press, New York, 1962.

8. T. J. Neubert and R. B. Lees, "Stored Energy in Neutron-Bombarded Graphite," Nucl. Sci. and Eng.~' 748-767 (1957).

9. F. c. 'Foushee, "Release of Stored Energy in Graphite End Pieces," GA Technologies internal memorandum, dated August 27, 1986.

10. J. H. W. Simmons, Radiation Damage in Graphite, Pergamon Press, Ltd.,

London, 1965.

• 11. F. C. Foushee, "The Consequence of Radiation Damage to the Graphite Reflector of a 1-MW TRI GA Mark II," GA Technologies document SAF-14, November 27, 1965.

I

\

DOCKET NUMBER PETITION l<ULE PRM ,.M-,44 @

r - -- , (5/FI!, .3JJ4!)

M(;KETEO USNRC Oregon

~tate.

Radiation Center U n1vers1ty Corvallis, Oregon 97331 (503) 75-4-2341

'87 JAN 30 P3 :16

\. _)

January 28, 1987 Off!

IOC Secretary U.S. Nuclear Regulatory Cormnission Washington, D.C. 20555 Attn: Docketing and Service Branch

Subject:

Comments on the Committee to Bridge the Gap Petition for Rulemaking, Docket No. PRM-50-44; Submitted by the Oregon State University TRIGA Reactor, License No. R-106, Docket No. 50-243 Gentlemen:

The Oregon State University TRIGA Reactor (OSTR) staff would like to respectfully submit the following comments on the above-referenced petition for rulemaking, PRM-50-44, and in so doing we would concurrently like to request that the Commission reject the subject petition for the reasons which follow. After evaluating the issues specified iD the petition, we believe that there is inadequate technical justification for the peti-tioner's contentions and that their proposals are not required nor appropri-ate to ensure the safe operation of the OSTR and other non-power reactors using graphite in a similar manner. In addition, we submit that for TRIGA reactors in particular a graphite fire is not a credible event and therefore the public is not at risk from such an occurrence. We intend to support this position by addressing each major point listed in the Supplementary Information section of the petition for rulemaking, as publis hed in the Federal Register on September 3, 1986. However, we would first like to clarify the location of graph ite in the OSU TRIGA reactor.

Infonnation on the Location of Graphite in the OSTR Graphite is found in the OSTR in five locations:

a. As part of the fuel elements, acting as top and bottom reflectors inside the stainless steel fuel cladding.

b. As stainless steel clad graphite reflector elements l ocated in some of the reactor grid positions around the outer edge of the core.

c. As an annular reflector for the core, sealed within an aluminum housing.

Oregon State Umversity Is an Affirmative Action /Equal Opportunity Employer 1.ckn db ref.:,;;£~.~'-.,.{!l.

U. S. NUCLEAR REGULATORY COMMISSION DOCK ET ING & SER'/ICl:: BR'INCH Or ICE OF TH_ sr { u Or THE cor* *1.1.., ,; ,'I Do~u ~ ~s l Fv l,

• k ') te F£ ( ii-e;J___

I -

~/~"°/ eaa,,10 I' I C

d. As part of a sealed aluminum-covered thermalizing column which has one end adjacent to the al um inum-covered core reflector in the water-filled reactor tank and the other end facing a water-filled bulk shield tank.

e. As an aluminum-covered thermal column with one end adjacent to the aluminum-covered core reflector in the water-filled reactor tank and the other end covered by a 19-ton concrete door.

It should be noted that graphite locations described in a,b and c above are totally within the reactor tank and are thus covered by approximately 4,500 gallons of water. Also, the graphite described ind is totally sealed and bounded on both ends by about 4,500 gallons of water, while the graphite described in e protrudes into the reactor tank inside a specially shaped aluminum housing which is part of the reactor tank wall configuration.

Infonnation in Support of Our Request for Rejection of the Petition Credibility of Graphite Fires

a. The petitioner asserts that "the occurrence of a graphite fire at the Chernobyl plant ... demo ns trates that graphite fires are credible events."

The Chernobyl incident clearly does not relate to non-power reactors and does not establ i sh that graphite fires at such reactors are credible events. Based on a number of reports in the open litera-ture, it is evident that the fire at Chernobyl was caused by a very large prompt critical power excursion (109 MW) resulting in a rapid temperature rise and complete core disruption, and was not due to stored energy in the graphite. The kinetic response of the Chernobyl reactor, wh ich led to the large power rise, and the kinetic response of small non-power reactors is completely different. In particular, the OSTR has a very large negative temperature coefficient of reac tivity, which means that any reac-tivity excursion results in the reactor automatically shutting itself down. Indeed, pulsing is a normal, approved, licensed method of operation for the OSTR. The Safety Analysis Report (SAR) for the OSTR shows that there is no credible scenario or mechanism which results in the fuel and graphite becoming hot enough to melt or bu rn.

b. The petitioner states that "the NRC has failed to require basic safety measures to reduce the threat of a graphite fire."

We assert that this is incorrect and that quite adequate "basic safety measures" have been incorporated into our facility to reduce the threat of any fire, including a graphite fire. The facility was designed with fire prevention in mind. For example, the building is of concrete and metal construction with virtually all of the materia l s used being non-flammable. In addition, considering the fact that essentially all of the graphite is

clad in either stainless steel or aluminum and is immersed in a large volume of water, it becomes apparent that it is also well isolated from ignition sources and appreciable amounts of atmospheric oxygen, and is in direct proximity to a large volume of substance (i.e., water) capable of extinguishing fires. A drop in the water level of only six inches (in a 20-foot tank) provides an alarm which results in an immediate response 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> per day. The city of Corvallis fire department has stated that they could easily provide water make-up of 2,000 gallons per minute, or more, to our reactor tank if needed. Clearly, this is more than sufficient to cope with any conceivable water loss.

To enhance fire safety beyond the above precautions, at our request, a state of Oregon fire marshal and a city of Corvallis fire marshal inspected our reactor facility and our fire response plan. The i r conclusion was that there were no sources of ignition in and around the reactor room other than normal electrical wiring, and, more importantly, no sources of combustible fuel to allow a fire to continue or spread. This led to their further conclusion that it was extremely unlikely that a fire would start within the reactor room and that any propagation would be very minimal.

It should also be noted that there are no electrical cables or wires in, or close to, any of the OSTR graphite.

c. The petitioner alleges that "licensees ... have no fire response plans for graphite fires. 11 While we assert that no credib le scenario exists for a graphite fire at our facility, it is entirely incorrect to say that we have no fire response plan for a graphite fire. The fire response plan for a graphite fire is exactly the same as that for any other fire in the facility, and this plan is an integral part of our current NRC approved eme rgency response plan. Corvallis fire fighters and their supervisors support a large research-oriented

• university and are therefore trained and equipped to deal with fires from a wide variety of sources. In particular, these indi-viduals are regularly trained in fire fighting procedures for our facility and they are prepa red to deal with the types of fires they could encounter, inc luding in their opinion, a fire involving graphite.

d. The petitioner asserts that "research reactors do not have adequate emergency plans to evacuate members of the public in the event of a graphite fire. 11 Every analys is of potential emergencies for research reactors of less than 2 MW has shown that evacuation of the public is not necessary. This position is reflected in the NRC s own regu-1 latory guide on emergency planning for research reactors. Further-more, the petitioner does not establish a sound technical basis which demonstrates why a graphite fire, in particular, should even require an evacuation plan.

Stored (Wigner) Energy in Graphite

a. The petitioner states that the NRC "underestimates the actual amount of stored energy and thus, underestimates the associated risk of graphite fire."

The buildup and release of stored energy in graphite does not automatically translate into an increased risk of fire. This is particularly true when the graphite is contained so as to exclude appreciable amounts of atmospheric oxygen and is immersed in a significant amount of water.

For the purposes of considering Wigner energy, graphite inside the OSTR fuel elements and the graphite in the reflector (or in the reflector elements) are the most important since the fast neutron fluxes in the two thermal columns are much lower. In a further refinement, the graphite in the reflector ring is of slightly greater interest than the graphite in the reflector elements because the graphite elements are clad in stainless steel while the reflector is clad in aluminum, and both are equi-distant from fuel in the core.

A conservative analysis performed at Oregon State University dealing with temperature increases in graphite due to the release of stored energy confirmed that the fuel element temperatures will remain within the fuel safety limit specified in the OSTR Technical Specifications. This result also agrees with the data reported by GA Technologies, Inc. in their submission to the NRC dated September 3, 1986. In addition, a similar independent OSU analysis for our graphite reflector ring produced data showing that energy releases and subsequent temperature increases in this structure would remain below the point where any damage to the graphite or aluminum would occur; and therefore these temperature increases would not be a problem if they were to take place. This conclusion agrees with the work reported by General Atomic in their document entitled "The Consequences of Radiation Damage to the Graphite Reflector of a 1-MW TRIGA Mark I I, 11 F. C. Foushee, November 27, 1965.

Moreover, a number of factors need to be kept in mind when discus-sing these stored energy calculations:

i) The factors which were used to convert from neutron flux to stored energy were conservative.

ii) The calculated stored energy values are the maxima, which occur only at the points of highest fast neutron flux and lowest graphite temperature. Therefore, the calculated values are conservative, since the mean values for stored energy will be considerably lower.

iii) Only a fraction of the stored energy is available for release.

However, in the OSU calculations it was conservatively assumed that 100% of the stored energy was released.

iv) Calculations of maximum temperature after stored energy release typically assume an adiabatic process (no heat transfer).

OSU calculations included this assumption, which is conserva-tive because it takes no credit for the fact that there will be heat transfer and the stored energy release will occur over an extended period of time.

Bearing all of these considerations in mind, we have concluded that the design of the OSTR and other similar non-power reactors is such that a graphite fire from the release of stored energy is not credible.

b. The petitioner stresses that "actual empirical measurements of Wigner energy will be required to assess the magnitude of the energy stored in research reac t or graphite."

A review of the location of graphite in the OSTR will quickly show that it is totally impract ical and entirely against the ALARA philosophy to attempt to measure the Wigner energy in graphite samples from reactors such as ours. More importantly, however, without a valid techn ical basis to establish the stored energy as a potential problem, its measurement becomes unnecessary and unwise.

We would like to thank the Commission for the opportunity to submit comments on this petition for ru lemaking and for extending the comment period beyond the original deadline. The additional time made it possible for us and for others in the non-power reactor community to more thoroughly study the issues, to research the literature, and to perform analyses which, from our viewpoint, resulted in a response based on facts. Should there be questions or a need for additional information, please let me know.

A. Johnson Acting Director AGJ/ef cc: D. Alger, UMC T. V. Anderson, OSU R. Carter, USNRC, Wash., D.C. S. E. Binney D. Feltz, TxA&M Univ. B. Dodd T. Raby, NBS J.C. Ringle Marc Voth, Penn State Univ. A. H. Robinson B. Wilson, Wash. State Univ. R. A. Schmitt Director, Oregon Dept. of Energy S. A. Stone Regional Administrator, D. L. Willis Region V, USNRC

DOCKET NU MBER PETITION RULE PRM fl-44@

(al ne ~1.J4 I STATE OF RHODE ISLAND AND PROVIDENCE PLANTATIONS nae KETL I

NR

"'oP'-

RHODE ISLAND ATOMIC ENERGY COMMISSION Nuclear Science Center

• a1 JAN 28 P2 :38 South Ferry Road Gfr 1..

Narragansett, R .I. 02882-1197 IOCK January 23, 1987 .~

Secretary U. S. Nuclear Regulatory Commission Attn: Correspondence and Records Branch Washington, DC 20555

Dear Sir:

This letter is a response to your request for comment concerning the petition for rulemaking of the Committee to Bridge the Gap which appeared in the Federal Register of September 3, 1986 on pages 31341 and 31342. This petition concerns the use of graphite as a moderator or reflector in research reactors.

These comments are based on the operation of a plate type reactor at 2 MW with water moderation and cooling using graphite reflectors canned in aluminum. The design basis accident for this reactor is a loss of coolant accident in which the bottom 7 " of the fuel elements and the reflector pieces remains in pool water.

During operation of the reactor, neutrons and gamma rays from the core cause heating in the graphite reflector. The temperature in the graphite from this heating is sufficient to cause continuous annea l ing of some of the Wigner energy deposited in the graphite, thus reducing the accumulation of stored energy. This effect becomes more pronounced as more energy is stored because the changes in the thermal conductivity of the graphite produce higher graphite temperatures.

In addition, in order to start a release of stored energy, the graphite temperature must increase some 50 to 100 degrees centigrade over the temperature of graphite during normal operation. Even with an increase in temperature it is important to recognize that the stored energy released is proportional only to this temperature increase and that the total stored energy is not released.

FEB 2 1987 knowtedged by card . ******** ; **** * ,n -

1 co,~* , iN U S. N JCLE DC,, r,.,, JrtANCr OFF I ur , E.TAtU Or 1 iJ Doc 0

s nark Dc:te _ _

s ,~e1ved _

'/-z.

/

3 / ~7

' r ~ Rep* * *£;

1. u.~. 1t. .ion _ _ / /M, ,le.5t:lAl!!,J

Secretary/U. S. Nuclear Regulatory Commission January 23, 1987 Page 2 We have examined the data presented in references 1 and 2 and compared the data with the characteristics of a 2 MW reactor during normal operation and during a design basis accident. We have concluded that during a design basis accident , the temperature obtained in the graphite reflectors is not sufficient to cause a release of the stored energy . We have further concluded that if the temperature of the graphite reflectors increased (by some mechanism unknown to us) sufficient to begin the release of the stored energy, the temperature achieved in the graphite from the release would be too low to create a problem for either the graphite reflector or the fuel .

The potential for fire in graphite reflectors is very small during storage and use. This is because the elements are enclosed in evacuated aluminum cans and during or after use in the reactor are stored in the reactor pool under large quantities of water.

Based on our review of the information presented in references 1 and 2 , we believe that the Nuclear Regulatory Commission has correctly assessed the safety implications of graphite as it is used in research reactors . We therefore believe that the petition of the Committee to Bridge the Gap should be denied.

References :

( 1) Stored Energy in Irradiated Graphi'te, Walt A.

Meyer, Jr. , University of Missouri, December 10, 1986 .

(2) Development of Uncanned Graphite Reflector Elements for Pool Reactors, Franz S. Holzer, Research Reactor Journal, Volume 3, No. 4, Summer 1963, pp . 10- 13 .

Very truly yours ,

a -~ 0 ,

A. Francis DiMeglio Di re c tor AFD:cd cc: RIAEC RUC

DOCKET NUMBER PETITION LILE PRM '-'°d-44

{~! r£ .J/.142) {BJ

[][))[JJ WORCESTER Worcester POLYTECHNIC Massachusetts 01609 INSTITUTE November 6, 19~f!h7r93-SOOO USNRC'

.86 NOV 18 P6 :oa Secretary U.S. Nuclear Regulatory Commission Washington, DC 20555 Attn: Correspondence and Records Branch Re: Docket No. PRM-50-44

Dear Si rs:

In reviewing the petition for rulemaking named above, several facts should be noted concerning both the WPI 10 kw nuclear reactor and the research

- reactor community at large.

Performing calculations for the WPI reactor graphite based on those done for the UCLA reactor (the results of which differed from those of the Committee to Bridge the Gap by only 20%), it was found that the WPI graph-ite currently contains 0.13 cal/g of stored energy. It was further esti-mated that by the year 2000, the graphite will contain 0.22 cal/g. This stored energy, if released instantaneously, would result in a temperature rise of less than 1°c. It should also be noted that several unrealis-tically conservative assumptions were made in these calculations.

Since the amount of Wigner energy stored in graphite is dependent on the neutron exposure of the graphite, the basis of these requirements should not be power level greater than 100 W, as proposed, but neutron fluence in the graphite (in either n/cm 2 or MWd, perhaps 10 20 n/cm 2). With require-ments based on fluence, many facilities could opt to replace the graphite present when the fluence approaches the requirement limit, instead of having to rely on safety equipment and evacuation plans in the case of an emergency.

Most nonpower research reactors operate at relatively low temperatures.

Even under accident conditions, the void and temperature coefficients of the reactors are such that any power excursion could not cause the reactor temperature to approach that of graphite ignition. The WPI Reactor, for

.l.cknowf deed by carct. ~-~C

• 9 1986

LI. S. NUCLEAR R~GL~ ATORY COMMrSSION DOCKFT" 1r:: ,, ... r ... ,,, r,: f""r-rTt()

er Poslm "r 1# -

Copic Add 'I  :,~ -

Special

/1t.r: l'll'h/s

example, under its present configuration, is unable to even cause the coolant water to boil.

In fulfilling the proposed requirements, many reactors, instead of going to the expense of implementation of response plans and equipment as well as Wigner energy measurements, would simply opt to remove the graphite where feasible and change the reactor core configuration. In making, these changes, potential exists of fuel loading errors and personnel exposure.

Even though this potential is small, it is probably more likely than a graphite fire.

If the NRC considers approving this petition, it should also consider the funding and implementation of tests to standardize Wigner energy measure-ments is research reactor graphite. Probablistic risk assessment studies should also be instigated in order to verify the necessity of this rule-making petition.

In short, I do not feel that I can support these proposed requirements because in speaking from a small research reactor point of view, a graphite fire due to a reactor accident is not only non-credible but impossible and

- measurements to verify the accuracy of Wigner energy calculations will stil l show the Wigner energy of our graphite to have a negligible contri-bution in an accident scenario, even if the measurements are several orders of magnitude higher.

Sincerely, Thomas H. Newton, Jr.

Director Nuclear Reactor Facility THN/ns

@)

DOCKET NUMBER

~ETITION RULE PRM

( 5'! F/!., ~ 1141 Ja-.{4 /.6 DO CKET *:

'. ., , NPl

• Box 355 Westinghouse Power Systems Pittsburgh Pennsylvania 15230-0355 Electric Corporation .86 NOV 10 A11 :59 IA 86-85 October 31, 1986 Secretary of the canmission U. s. NUclear Regulatory camd.ssion Washington, DC 20555 Attention: Correspondence and Records Branch Gentlemen:

This letter is in response to the proposed amendment to 10CFRS0 in the Federal Register, Volume 51, No. 170, september 3, 1986 regarding graphite fires. our carments are given below.

Westinghouse owns and operates a l0KW pool type training reactor, Docket No.

50-87, License No. R-119. Although the reactor is licensed to an operating power level of l0KW, the reactor is rarely operated at power levels in excess of 100 watts. The average operating power since the time our core was modified to include graphite reflector assemblies has been 19 watts.

The regulation proposed by the petitioner would require all licensees whose reactors enploy graphite as a neutron moderator or reflector to inmediately fonnulate and suJ::mit fire response plans, evacuation plans and a canplete revision of their Final safety Analysis Report (FSAR). This petition seems to be impulsive, E1110tionally-based, and founded on the Chernobyl accident which was the result of inter-related human factors and design weaknesses involving a reactor that uses a unique design significantly unlike nuclear plants in the U.S.A. While Westinghouse does not disagree that the possibility of a graphite fire in a U.S. reactor should be assessed, proper engineering practice dictates that this assessment be canpleted in a logical, systematic manner. Westinghouse suggests the following systematic approach:

1. Each licensee would perfom. a detailed analysis to quantitatively detem.ine the stored energy in the licensee's graphite.

2. Fran these calculations, the license would detem.ine the probabili ty of occurrence of a graphite fire.

3. All licensees would suJ::mit these analyses to the NRC for review.

4. I f the total stored energy in the graphite or the probability of a graphite fire occurring exceeds sane predetem.ined value, then the AOV 17 1986 Ackl10Medget1 PJy card.:.;;;;;;., ..... ,, ..,..

• i NUJ::LEAR P.Fr" ,, ATORY cqM~ISSIOII DOCKETINC om --*

0, O

'VI CE SECT ION N!ETARY

,; JON ostmark D, opies c 1/lI dd'I c ai!)

- I S°.

/l:trs

u. s. Nuclear Regulatory 0almission october 31, 1986 Page TWo licensee should be required to verify that his existing evacuation plans would be suitable for response to a graphite fire anergency or formulate and subnit new fire fighting and evacuation plans and FSAR revision.

The Westinghouse Nuclear Training Reactor consists of Uraniun fuel elements and graphite reflector elements sul:lnerged in a pool of water which operates a.t low temperature and atmospheric pressure. Without the presence of the water, the reactor is not capable of lmdergoing a chain reaction. Also because of the low power of operation the graphite and fuel contain only small quantities of fission pmducts.

we have perfoJ:med a. preliminary analysis on the graphite used in the Westinghouse NUclear Training Reactor. In this analysis, sane extremely conservative assumptions were made. The graphite reflector assemblies sun:ounding the core were assumed to be exposed to a flux equal to the average flux level in the core at power. Utilizing the integrated power of the reactor over the time the graphite has been installed, there would be no ni::>re than .0806 ca.1/gm of stored energy in the graphite assemblies. This would result in an approximate temperature increase in the graphite of 1°F, if all of the stored energy were released simultaneously. In addition, the maximLml fuel plate surface tarpe,:ature reached ~ the Maximm Credible Accident analyzed by our FSAR is no D)re than 320°F. Making the very conservative assumption, that the graphite were to attain a tempe"='8.ture equal to that of the fuel surface 4urin;J the transient, the additional one degree increase fran the stored energy is certainly insignificant. The oxidation of the graphite at temperature on the order of 320°F is negligible. It follows fran this conservative analysis that the drastic steps proposed by the petitioner would not be warranted for the Westinghouse Nl'R.

In conclusion, Westinghouse would like to stress that a systematic approach be taken. If an analysis of the licensee's facility can show that the amount of stored energy in their graphite is insignificant and/or that the effects of the release of this energy is negligible, the licensee should be exempt fran any requirement to subnit graphite fire fighting and evacuation plans and a revision to the safety Analysis Report. Westinghouse also recxmnends that the NRC identify acceptable data or references which oould be used in all licensee's analyses. In this way, all of the facilities oould be using the same basis for their calculations.

1. .;zt1_[-

A. J. Nardi, ~ er ES License Administration ATN/dh

I

• Westinghouse Power Systems Box 355 Pittsburgh Pennsylvania 15230-0355 Electric Corporation IA 86-85 October 31, 1986 secretary of the ccmnission

u. s. Nuclear Regulatory Ocmnission Washington, DC 20555 Attention: Correspondence and Records Branch Gentlemen:

This letter is in response to the proposed amendment to 10CFRS0 in the Federal Register, Volume 51, No. 170, September 3, 1986 regarding graphite fires. our oc:mnents are given below.

Westinghouse owns and operates a 10KW pool type ~ reactor, Docket No.

50-87, License No. R-119. Althou';Jh the reactor is licensed to an operating power level of 10KW, the reactor is rarely operated at power levels in excess of 100 watts. The average operating power since the time our core was modified to include graphite reflector assanblies has been 19 watts.

The regulation proposed by the petitioner would require all licensees whose reactors anploy graphite as a neutron mod.era.tor or reflector to jDIJlelU.ately foJ:mUlate and sul:mit fire response plans, evacuation plans and a oanplete revision of their Final Safety Analysis Report (FSAR). This petition seems to be impulsive, emotionally-based, and fOlmded on the Chernobyl accident which was the result of inter-related human factors and design weaknesses involving a reactor that uses a unique design significantly lmlike nuclear plants in the U. s .A. While Westinghouse does not disagree that the possibility of a graphite fire in a u.s. reactor should be assessed, proper engineering practice dictates that this assessment be oanpleted in a logical, systematic manner. Westinghouse suggests the followilq systematic approach:

1. Each licensee would perfom a detailed analysis to quantitatively dete:rndne the stored energy in the licensee*s graphite.

2. Fran these calculations, the license would dete:rn,i ne the probability of occurrence of a graphite fire.

3. All licensees would sul:mit these analyses to the NRC for review.

4. I:f the total stored energy in the graphite or the probability of a graphite fire occurriD;J exceeds sane predete:rn,ined value, then the

u. s. NUClear Regulatoxy ccmnission October 31, 1986 Page TWO licensee should be required to verify that bis existuq evacuation plans would be suitable for response to a graphite fire anergency or foi:mulate and sul:mit new fire fighting and evacuation plans and FSAR revision.

The Westinghouse Nuclear TrainiJq Reaetor consists of Uranium fuel elements and graphite xeflector elements subDerged in a pool of water which operates at low tanperatm:e and atmospheric pressure. Without the presence of the water, the reactor is not capable of undel:going a chain reaction. Also because of the low power of operation the graphite and fuel oontain only small quantities of fission pJ:Oducts.

We haVe perfoi:med a preliminary analysis on the graphite used in the Westinghouse NUclear Trai.n:iiq Reactor. In this analysis, sane extranely conservative assumptions were made. The graphite reflector assemblies surroundiJ¥:1 the oore were assumed to be exposed to a flux equal to the average flux level in the oore at power. Utilizing the integrated power of the reactor over the time the graphite bas been installed, there would be no nt>re than .0806 cal/gm of stored energy in the graphite assanhlies. This would result in an appi:ax:i.mate teq,emture increase in the graphite of 1°F, if all of the stored energy were released simultaneously. In addition, the maxi.mm fuel plate surface tfflpn:-ature reached 4u:!"irq the Maximum credible Accident analyzed by our FSAR is no more than 320°F. MaJdJq the vexy conservative ass\l!l)tion, that the graphite were to attain a tanperature equal to that of the fuel surface during the transient, the additional one degree increase fi:an the stored energy is certainly insignificant. The oxidation of the graphite at ta,penture on the order of 320°F is negligible. It follows fi:an this oonservative analysis that the drastic steps proposed by the petitioner would not be warranted for the Westinghouse Nl'R.

In oonclusion, Westinqhouse would like to stress that a systematic approach be taken. If an analysis of the licensee*s facility can show that the amount of stored energy in their graphite is insignificant and/or that the effects of the release of this energy is negligible, the licensee should be exempt fi:an any requirement to sul:mit graphite fire fighting and evacuation plans and a revision to the Safety Analysis Report. Westinghouse also recxm:nends that the NRC identify acceptable data or references which oould be used in all licensee*s analyses. In this way, all of the facilities oould be using the same basis for their calculations.

~ Lt A. J. Nardi, ~ e r ES License Administration Al'N/dh

• Box 355 Westinghouse Power Systems Pinsburgh Pennsylvania 15230-0355 Electric Corporation IA 86-85 0ctober 31, 1986 Secretary of the o::mnission

o. s. Nuclear Regulatory o::mnission Washington, DC 20555 Attention: O>rrespondence and Reoords Branch Gentlemen:

This letter is in response to the proposed amendment to 10CFRS0 in the Federal Register, Volume 51, No. 170, Septanber 3, 1986 regarding graphite fires. our cc:mnents are given below.

Wes~house owns and operates a 10D pool type trainin;J reactor, Docket No.

50-87, License No. R-119. Altholqh the reactor is licensed to an operating power level of 10D, the reactor is rarely operated at power levels in excess of 100 watts. The average opera~ power since the time our core was modified to include graphite reflector assemblies has been 19 watts.

The regulation proposed by the petitioner would require all lice-sees whose reactors anploy graphite as a neutron moderator or reflector to bmediately fomul.ate and suJ::mit fire response plans, evacuation plans and a oanplete revision of their Final safety Analysis Report (FSAR). This petition seems to be impulsive, anotio:nally-based, and founded on the Chernobyl accident which was the result of inter-related human factors and design weaknesses involving a reactor that uses a unique design significantly unlike nuclear plants in the U.S.A. While W e s ~ does not disagree that the possibility of a graphite fire in a u.s. reactor should be assessed, proper

~eering practice dictates that this assessment be oanpleted in a logical, systanatic manner. W ~ so;Jgests the followi.Jq systematic approach:

1. Each licensee would perfo:m a detailed analysis to quantitatively detennine the stored energy in the lioensee*s graphite.

2. Fran these calculations, the license would detennine the probability of occurrence of a graphite fire.

3. All licensees would suJ::mit these analyses to the NRC for review.

4. l:f the total stored energy in the graphite or the probability of a graphite fire occurrirq exceeds sane predetennin~ value, then the

o. s. Nuclear Regulatory Oc:mnission October 31, 1986 Page Two licensee should be required to verify that his existing evacuation plans would be suitable for response to a graphite fire anergency or foDlllllate and sul:mit new fire fighting and evacuation plans and FSAR revision.

The Westinghouse Nuclear Training Reactor consists of Uranium fuel elements and graphite reflector elements sul:merged in a pool of water which operates at low 1:er{>e,:ature and atmospheric pressure. Without the presence of the water, the reactor is not capable of undergoing a chain reaction. Also because of the low power of operation the graphite and fuel contain only small quantities of fission products.

we have perfozmed a preliminary analysis on the graphite used in the Westinghouse Nuclear Training Reactor. In this analysis, sane extremely conservative assumptions were ma.de. The graphite reflector assemblies surrounding the core were assumed to be exposed to a flux equal to the average flux level in the core at power. Utilizing the integrated power of the reactor over the time the graphite has been installed, there would be no nt>re than .0806 cal/gm of stored energy in the graphite assanblies. This would result in an BR;>rox:i.mate temperature increase in the graphite of 1°F, if all of the stored energy were released simultaneously. In addition, the maximum fuel plate surface taiperature reached ~ the Maximum credible Accident analyzed by our FSAR is no mre than 320°F. Makixq the very conservative assumption, that the graphite were to attain a temperature equal to that of the fuel surface 4uri.iq the transient, the additional one degree increase fran the stored energy is certainly insignificant. The oxidation of the graphite at temperature on the order of 320°F is negligible. It follows fran this oonservative analysis that the drastic steps proposed by the petitioner would not be warranted for the Westinghouse Nl'R.

In conclusion, Westinghouse would like to stress that a systematic BR;>roach be taken. If an analysis of the licensee* s facility can show that the moount of stored energy in their graphite is insignificant and/or that the effects of the release of this energy is negligible, the licensee should be exempt fran any requirement to sul:mit graphite fire fighting and evacuation plans and a revision to the Safety Analysis Report. Westinghouse also recxmnends that the NRC identify acceptable data or references which could be used in all licensee*s analyses. In this way, all of the facilities could be using the same basis for their calculations.

1. .;qr-L A. J. Nardi, Manager F.S License Administration AJ'N/dh

O0CKET NUMBER .~

t,.,~TION RULE PRM fl-4,4l.f!,./

{_£ii M $194))

Oct 18 - 86 from the pen of EMERY NEMETHY , Se&)tk?*

ECOL GY/ALERT BOX 6 2 1 '86 NOV -7 Al0:38 BLO0~S URG 1781 9 Sec ' y - NRC V ash , DC 20555 OFF IL::.

00CK£-l ll ~*

u,, ,i Flv *(f ATT: COrtllliS..t-01DEI~C~ & h.ECORDS BrtA~f ©Nt.1-1 .

Re : Docket PRIV1 44 Petition by Committee to Bridge the Gap Gentlemen ::..

In view of the fo l lowing:

Th at the disaster at Ch ernobyl proved tl1at graphite fires are not "non- credible",

Tha t appare n tly there are e.ozens of non-power research reactors moderated by graphite ,

That exneriments have shown NrtC' s nredic-tions of Wigner Energ y in graphite have been un oe re stim a te d, We feel the peti tiori by the Commi tee to Bridge t h e Gap is fair , reasonable Pnd in the nublic interest .

NOV 1 ..

Acknowfed,ed by card. * ********~ ******'***

D.S. NUCL r T""~Y COMMISSIOIJ DOCK ~f SECTION 0

• TARY N

J>oslmorl- r*

Copies *

~dd'I c lpecial D f.

~

DOCKET NUMBER i£.TIIION RULE PRM 5J-# BOOTS WINTERSTEIN President ~

(~! j:::,e,'.9;g4J) Irene Willams . . . . . . 1st Vice Presk:lent Patty Folard . . . . . . 2nd Vice President lru Baltimore Cit;y Jane Harrison . . . . . . 3rd Vice President Doris Johnson . . . . . . . . . . . Secretary Susan Greenwood . . . . . . . . . Treasurer LEAGUE OF WOMEN VOTERS OF BALTIMORE CITY

• 2318 NORTH CHARLES STREET, BALTIMORE. MARYLAND 21218

• 301 889-5353/889-5354 November 2 , 1986 co Secretary of the Commission o .,,

,;:r, US Nuclear Regulatory Commission M r**

-t Washington , D. c . 20555 a,-

~ ::z

,,. C1

z

,,it" '

Dear Sir  :

i:

~r"" -0 Only yesterday did the Natural Resources Cornrn~iitee ~

the Baltimore City League of Women Voters become a-iiai:-e cf:!i the Petition for Rulemaking PRM 50-44 which speaks to the issue of the use of graphite in research reactors . We hope that our letter is not too late for consideration .

We understand that the petition asks the NRC to make specific plans for a response to a graphite fire both in terms of fighting such a fire ( requiring special knowledge of what to use and how) and in evacuating persons in the area .

Anyone who has read the INSAG Summary Report on the Post-Accident Review Meeting on the Chernobyl Accident held in Vienna , 30 August - 5 September recognizes that new and more stringent safety plans , on and off- site , power or research reactor , are in order . Since human error played a large role in both the TMI - 2 and Chernobyl accidents , redundancies in safety systems cannot insure an accident-free future .

Graphite as a fire problem at Chernobyl warns us to seek specific response plans wherever graphite is used with reactors in our own country .

Stored energy in research reactor graphite should be measured . Certainly there could be no harm and much reassur-ance to be gained by requiring measurement of "Wigner energy" in graphite wherever reactors use this material .

In summary, we support the measures addressed in the Petition for Rulemaking PRM 50-44 and urge the Nuclear Regu-latory Commission to address the question posed by the use of graphite in research reactors in regard to public safety and preparedness in case of emergency .

Sincerely

~ /_.a-x-L---

Pat Lane Natural Resources Committee LWV Baltimore City

l>osf ar Copi

~d ,

!Spec,

DOCKET NUMBfiR 5>>ETlTlON RULE_ PRM .jl -44, (ip (SI~ .!/-34!}

Department oM&._ij~lfilligineering

- * ~15\'s'weeney Hall IOWA STATE Ames, Iowa 50011 UNIVERSITY GFFI' r*

October 29, 1986 OCCh Tl Secretary, U.S. Nuclear Regulatory Commission Correspondence & Records Branch Washington, D.C. 2055~

Dear Sir:

- The Department or Nuc 1ear Eng i nee,* i ng at Iowa State University i c, submitting this comment in response to the July 17, 1986 peti-tion by the Committee to Bridge the Gap (CBG) to the U.S. Nuclear Regulatory Commission regarding the possible hazards associated with graphite fires. CBG proposed an amendment to 10 CFR Part 50 Lhat would require implementation of additional emergency plans to combat raphite fires in reactors, special evacuation plans in the event of such fires, and measurements or stored energy In reactor* graphite to ascertain its potential to initiate or exacerbate a reactor accident. It is the position of the licensee (ISU) that:

the possibility of a graphite fire in the UTR-10 reactor is extremely remote, stored energy in the graphite of the UTR-10 is too small to initiate or contribute to a serious accident, and existing emergency response procedures ar*e adequate for the fac i l i ty.

REGULATORY COMMISSION NG & f~n ICE SECTION E OF T' t ~CRETARY THE C"C'VMI SSION ocum

• 1ti,tic*

/d/2f/J<_~-

d 2_

• hon

In our opinion, therefore, th0 proposed amendment is unnecessary For our reactor (and any other reactor with a similar power history); moreover, compliance would fmpose an undue burden upon our facil lty personnel (including unnecessary radiation exposures while collecting graphite samples ror analysis).

There is approximately 183 ft 3 or graphite in the UTR-1O reactor (8300 kg) serving as reflector, thermal column, and a neutron duct to the shield tank. The facflity fs described in the Safety Analysis Report (ISU-ERI-Ames 82418, August, 1981). The graphite 4t is completely enclosed in the concrete shielding surrounding the reactor, and as a result very little graphite is exposed to*a sufficiently large volume of air necessary to sustain a fire.

There fs a foot-high air space above the reflector, but flow of afr to this exposed graphite is restricted by the very small gaps in the top closures.

Not only is there insufficient air to support a fire in the

- reflector, there is no credfble accident that could ignite such a fire.

1. There cannot be a prompt critical reactor excursion because the license technical specifications limit the excess reactivity to .005 delta-k/k, or less. Thus a Chernobyl~l ike power surge fs physically* impossible.

2. The reactor power is 1 imited to 10 kW, far too low to initiate combustion fn graphite. The same is true for the maximum possible power transient described in the Analysis Report (insertion of +.005 delta-k/k 2

while at 10 kW). The energy production in this transient is 11 kW-sec, with a resultant fuel temperature increase of less than .01 °c.

3. A buildfng fire cannot ignite the graphite because ft is i~olated inside the concrete shield. Also, a very hot ffre of long duration would be required to heat the graphite enough to start a fire. Such a fire could not go undetected, and the Ames Fire Department Is able to respond to a fire In the lab within minutes.

4. An attachment to this letter contains a conservative estimate of the Wigner energy stored in graphite In the UTR-10 reactor as a result of its cumulative 6700 kWhr energy production. The stored energy is about 4.5 kWhr in the 370 kg of graphite adjacent to the core tanks, or 10.5 cal/g. If the energy was suddenly released adiabatically, during the power transfent of item 2 above, this graphite would experfence a 62 °c temperature increase. The energy release fs greater than the transient1 but the resulting graphite temperature, about 80 °c, is far below that required for combustion.

The only conceivable way a graphite rire could be ignited might be vfa an attack by determined, professional saboteurs. A powerful incendiary explosive, capable of destroying the reactor structure and pulverfzfng the graphite would be needed. However, such an explosion itself would pose a much greater danger to 3

building occupants and passers-by than any subsequent release of radioactivity. The consequences of a graphite fire are negligible compared to the fnitfatfng event.

Finally, ft should be emphasized that the UTR-10fac111ty has an NRC approved emergency plan for responding to fires. Evacuation procedures exist for the Nuclear Engfneerfng Laboratory and they are exercised annually in drills. NRC did not require the establ fshment of an emergency planning zone beyond the building itself because the inventory of radioactfvfty is simply too smal 1.

To summarize the above analysis, we have reached the following conclusions:

1. Because the graphfte in the UTR-10 fs enclosed in concrete, the possibility of a large graphite fire is very remote.

2. There is simply no credible reactor accident that could fgnite the graphite.

3. A sudden release of all the stored energy in the graphite adjacent to the core would result in at most a 62 °c temperature increase.

4. Only an act of sabotage~ resulting in the total destruction of the facility, could cause a graphite fire.

5. Existing emergency response procedures are adequate for any credible reactor accident because the stored energy is insufficient to cause or contribute to an unanticipated accident scenario.

4

We conclude that no credible accident at the UTR-10 can result in a graphite Fire, and therefore, recommend the NRC deny the CBG petition requirements of additional emergency and evacuat1on plans. Also, since an NRC Safety Evaluation Report of the faci 1 fty (NUREG-1016) has already considered accident scenarios far more extreme than the release of graphite stored energy, we contend a revised Safety Analysis Report is unnecessary.

Finally, our calculations (see attached memorandum) show the stored energy in the graphite is at most 4.5 kWhr, or 10.5 cal/per gram of graphite adjacent to the core. There fs no hazard associated with the sudden release of this energy.

Therefore, we see no reason to sample the graphite for stored energy measurements. Such a requirement would cost our facility hundreds of person-hours (al 1 the fuel would have to be removed to storage), result in a few person-rems additional doses, and force us to spend very scarce funds for unnecessary laboratory tests (unless NRC was willing to contract for the tests).

We urge the Commission to reject thfs attempt to amend 10 CFR 50.

Sincerely, Bernard I. Spinrad Professor and Chair BIS:rpa Encl: Stored Energy in Reflector Graphite of Research Reactors 5

STORED ENERGY IN REFLECTOR GRAPHITE OF RESEARCH REACTORS by Bernard I. Splnrad A sfmpl ffied method of computation is adequate to determlne whether there is any possibility that enough stored energy might be present fn the reflector graphite of research reactors to present any problem of severe overheating on its sudden release.

The steps of th1s computation are sufficiently transparent so that elaborate codes or careful measurements need only be employed when the results of this simplified approach indicate a high degree of damage. In other words, this fs a simple bounding calculation. The steps are:

1. From the reactor eQergy production history, estimate the total energy that has been released in the form of kinetic energy of fission neutrons.

2. From the element8ry reactor neutronics models, estimate the fraction of these fission neutrons that are moderated in the

- reflector graph f te.

3. From published results determine the fraction of the energy transferred to carbon atoms by neutron moderation that goes into atomic displacement energy.

4. From the typical kinetic energy of a dislocated carbon atom a number that is also useful in step 3 -- and the energy stored in such a dislocation, derive the conversion factor that measures the efficiency of energy storage.

5. From experiments that compare stored energy from irradiation at cryogenic temperatures with that from irradiation

at actual temperature, determine the fraction of stored energy that ts not "self annealed" during irradiation.

6. From known neutron slowing down characteristics of graphfte, determine the effective volume of graphite that has been subject to signfficant energy storage.

7. From the previous estimates of energy storage and graphite volume (converted to mass), the total temperature rise of the graphite in case of sudden release of stored energy can be estimated.

Steps 1, 2 and 6 are reactor-dependent. Therefore, we first calculate the results of steps 3, 4 and 5, as wel 1 as one of the more important pieces of data used in step 1.

PRELIMINARY ESTIMATES Step_!_

Whf le the total energy generation of a reactor is hfstory-dependent, the fraction of flssfon energy that goes fnto kinetic energy of neutrons fs not. The ffssion of 235 u produces approximately 200 MeV of energy at steady state, including 5 MeV of neutron kfnetfc energy, from approximately 2.5 neutrons of 2 MeV each. Thus, 1/40, or 2.5% of reactor energy is 1 iberated as neutron kinetic energy.

Step l The "displacement energy" of graphite -- the energy required to displace a single carbon atom from its lattice -- fs approximately 25 eV (1). Different models converge on a estimate 2

of approxfmately 22000 displacements in graphite for a 2 MeV neutron completely thermalized in graphite (2). Combfning these two numbers gives 550,000 eV of displacement energy arising from moderation of a 2 MeV neutron. This is a fractfon 0.275 (27.5%)

of the neutron energy dissipated in the graphfte.

This fraction is a reasonable one, consfderfng that; (a) For the most energetic carbon atoms that exist from early col 1 isfons of neutrons during moderation, electronic excitation, which does not produce atomic displacements, is a major source of energy dissipation.

These coll is ions with high energy neutrons transfer most of the neutron energy.

(b) For carbon atoms in the 25-100 eV range, a thermal spike, which does not produce further displacement, fs a major source of energy dissipation.

(c) Most atoms that are displaced are actually more energetic than 25 eV; therefore, the consideration just mentioned in item (b) applies to most displacement events.

Step .1.

The energy stored by formation of a vacancy and a matching displaced atom has been estimated in a variety of ways. A value corresponding to 3 eV per vacancy (70 kcal/gram-atom) is near the upper end of estimates, and has been generally adopted. To this must be added the energy of the displaced atom. If the heat of sublimation of graphite is used for this -- an effective upper 3

1 imit -- the energy of an interstitial would be about 7.1 eV (170 kcal/gram-atom). The actual energy stored thus lies between approximately 3 and 10 eV. Thus, of the 25 eV needed to create the displacement-vacancy pair, something between 12 and 40% is actually stored, the rest being dissipated as heat. We adopt the upper value 40% -- as more conservative; ft is used by Nightingale In a major text (3).

Step 2.

All estimates in previous steps assumed that stored energy remains -fixed in place. This is the case -for radfatfon damage to a sample maintained at cryogenic temperature. I-f irradiation is at room temperature appreciable annealing occurs. It is estimated that about 1/4 of the storable energy will anneal out under these circumstances (4).

ESTIMATE FOR THE ISU REACTOR To Illustrate the remainder of the process, we shal 1 use data

-from the UTR-10 reactor at Iowa State University. This is a two-slab Argonaut-type reactor. We shall re-fer to it fn what -fol lows as "the ISU reactor".

~ J_

Since it was -first put into operation, the ISU reactor has logged 6700 kWh of energy generation. 0-f this, 2.5% or 167.5 kWh was neutron kinetic energy.

4

Step .Z.

We shal 1 denote inf"inite multiplication Factor as "K". K f"or this reactor has been estimated in several dff"f"erent ways. The highest estimate is 1.48. Since the Fraction of" neutrons leaking f"rom the core is 1 - 1/K, this leads to a conservative estimate of the leakage fraction, 32.5%. We assume that all leakcige is of Fission-energy neutrons into graphite, which is 1 ikewise conservative because: (a) some of the leakage is upward or downward into water ref"lectors; and (b) some of" the leakage is of neutrons that have first made moderating collisions with the core water. Applying the Factor of" .325, nevertheless, leads to a calculated 54.4 kWh of" neutron energy deposited fn graphite.

Steps l.t_ 1 and~

The numbers For these steps have already been presented. They conclude as factors of" .275, .4, and .75. The product of thege three factors is .0825. No more than thfs fraction of the neutron kfnetic energy deposited in graphite remains as stored energy fol lowing neutron irradiation. Applying this factor to the !SU reactor leads to an estimate of 4.5 kWh of" energy stored in graphite.

Step§.

A given amount of energy may be signif"icant if it is deposited in a smal 1 sample, trivial in a very large sample. It is necessary 5

to estfmate the affected volume of graphite, in the case under discussion. In this case, conservatism favors using the smallest reasonable amount of graphite. For the ISU reactor, the ~~ti~t~

f5 as follows:

(a) The root-neutron-age in graphite fs approximately 20 cm.

Fast neutrons entering the graphite are attenuated --

f .e., lose their energy -- accordfng to a curve for which this is a characteristic decay length. Although many neutrons leak through, we here assume that all the energy fs stored fn this thickness of graphite.

(b) The core consists of two water-cooled slabs, each approximately 60 cm high and 48 cm wfde. The two slabs are separated by 45 cm of graphite. We assume that the cross-sectional area of potentfally affected graphite is Just four times (for two faces each, of two slabs) this 60x48 core area. In other words, the considerable axial and lateral leakage of fast neutrons is ignored. The volume of graphite under consideration fs therefore 4x20x60x48 cubfc centimeters, or 0.23 cubic meters. The graphite has a specffic gravity of 1.6, so this amounts, finally, to 370 kg.

Step]_

In thfs, the final step of the calculation, we compute how much temperature rise would occur in the affected graphite from release of all the stored energy in it: i.e., the release of 4.5 kWh of energy in 370 kg of graphite. Using the handbook value of 6

0.17 cal/g-°C and standard conversion factors, this amounts to 62 °c.

DISCUSSION The computation presented above has enough conservatism in ft to be Justfffably labeled as a bounding calculation. For the IS.Li reactor, the maxfmum temperature rise so calculated would be 62 °c, which means that even a sudden release of stored energy would not brfng the graphite above the temperature of bofl ing water. This would not be a damaging event, and consequently there fs no safety-related reason to consider determining the real value -- whfch is likely to be less -- either by more precise theoretical methods or by experiment. Indeed, the specific stored energy fs so ./

small that ft would be dffffcult to measure at al l ..

There are, of course, other reactors with graphite reflectors.

- Many of these reflectors are far from the reactor cores, which would mean both that the fraction of neutron energy deposited in them would be small, and the mass of graphite fn which damage might be concentrated would be large. Both these factors might compensate for greater operatfng energy production than that of the ISU reactor. In any case, bounding calculations, fn the spirit of those reported here, should d~termlne whether such reflectors pose any putative problem requiring elaborate measures to resolve.

7

Only after simplified estimates of specfffc energy storage have been obtained, and have indicated the possibility of large temperature rises, should more detailed estimates of safety implications, and such confirming data as may be Indicated, be called for. This clearly should be done on a case-by-case basis.

REFERENCES

1. R. E. Night i nga 1 e, Ed. , "Nuc 1 ear Graphite". D. R. De Ha 1as, Ch. 7, "Theory o-f Radiation E-f-fects in Graphite", p. 197.

Academic Press, New York (1962).

2. fbid, p. 217.

3. Ibid, R. E. Night f nga 1 e, Ch. 12, "Stored Energy", pp. 3 43-3-4 5.

4. ibid, p. 340.

8

DOCKET NUMBSR

~ETITION RULE PRM ~

5J

--,44 Z?ff}

r:.

UNIVERSITY OF MISSOURI-ROLLA (5IF£ .JIJ4i) cJ6 OOCK£TE NM~ r Reactor Facility Nuclear Reactor Mlw,II.ii,+1iti0j-lt~540 1-0249

• ~elepfione~34'4) 34 1-4236 October 29, 1986 Offl Cc.-

0OCKETING & !::>f vAft.,

I er.

BRANCf-l Secretary U.S. Nuclear Regulatory Corrvnission Washington, D.C. 20555 Attention: Correspondence and Records Branch Docket No. PRM-50-44

Dear Sir:

We have the following objections to the proposed petition for rule-making regarding "credibility of graphite fires" and "stored (Wigner) energy":

1. The deadline for submittal of comments of November 3, 1986 is too early and should be extended by at least sixty (60) days.

2. The deadline of January 1, 1987 for submittal of fire response plans, evacuation plans, measurement of "Wigner energy", etc. is far too early and, if implemented, should be extended by at least one (1) year.

3. Some reactor faci l ities, such as ours, where the graphite reflector is physically separated by approximately1/2 inch of steel, 4 inches of lead, and 6 inches of water (at closest approach), should not be required by the proposed amendments to prepare such fire response plans, evacuation plans, measurement of "Wigner energy", etc., because a fire in the graphite would not lead to the release of fission products in the reactor core.

4. It should not be necessary for each reactor facility to experimentally measure the "Wigner energy". If the U.S.

NRC staff concludes that the current experimental data base on the matter is insufficient, then a limited number of facilities cou ld perform such a study.

5. The determination of the location of the maximum stored energy and its quant itative value would be calculated based upon neutron flux distributions for other reactors.

an equal opportunity institution

IJ. S. LEAR REGU C>O<:KETI &

OFFICE Of OF THE Docurr

,6tfmark Oa I Copie, ec iv d Add' I Copies R lpecial Oistribuh

U.S. NUCLEAR REGULATORY COMMI SSION October 29, 1986 Page 2

6. The determination of the max imum quantity of stored energy to within 10% is not practical from either an experimental or an analytical app roach. An uncertainty of+/- 50% for such a physical property wou ld be more realistic.

Sincerely,

@104.l-Q,~

Albert E. Bolon, Ph.D., S.O.

Reactor Facility Director AEB/lp cc: T. Raby (TRTR Group )

\

t)t>CKET NUMBfiR PRM n AA

,ju--v,y rl HAR RY PEARLMAN, PH.D. ~E"'9TION RUL E----..- CONSULTANT IN ENERGY 11818 Porter Valley Drive [51 F;e, _gl.341}

Northridge, California 91326 (213) 360-6132 DOCKETED USNRC October 29. 1986

• 86 NOV -3 P2 :23 Chairman Lando Zech, Jr.

Commissioner James Asselstine Commissioner Frederick Bernthol Commissioner Thomas Roberts U.S. Nuclear Regulatory Commission (NRC)

Washington, DC 20555 Docket No. PRM-50-44

Subject:

Committee to Bridge the Gap; Petition for Rulemaking Gentlemen:

I'm responding to NRC 1 s request (Federal Register/Vol. 51, No. 170/September 3, 1986) for comments on subject petition, which seeks to have NRC impose re-strictive conditions on all graphite-bearing reactors operating in the United States. The petition alleges that a massive graphite fire is now a credible accident because of recent events at Chernobyl. Further it claims that high levels of stored energy in the graphite contribute to its combustibility.

While a few power reactors are mentioned in the petition, the great majority of reactors affected by the proposed rule would be non-power research reactors.

They would be required, under the proposed rule, to take samples from the re-actor graphite and measure the stored energy.

The petition should be rejected. Its allegations and claims are not supported by available facts, as I shall demonstrate below, beginning with the stored energy issue .

• Graphite Stored Energy Measurements In a paper delivered at the American Nuclear Society (ANS) meeting in Reno, NV this past June, I reported measurements of the stored energy in samples o) reactor graphite taken from several locations in the UCLA research reactor.O Reference (1) is a Summary, which contains results at two locations, only:

19.2 cal/gm near the center of the graphite island (see Figure 1); and 33.2 cal/gm at the edge of the center fuel box. The June 18, 1986 presentation included additional results, at successively greater distances from the core.

All the data are summarized in the attached Table 1. The sample locations are indicated on the attached Figure 1, where the earlier 33.2 value was corrected to 33.3, to be consistent with the numerical rounding procedure used for the later samples. Both the Table and Figure were presented on June 18, among other vugraphs.

Several observations may be made from these results. First, the stored ener-gies, which were accumulated in a total reactor operation of 21 Mwd (thermal fluence about 3 x 1019 nvt). One measure of the importance of the stored energy is the increase in temperature in a given graphite sample, when the energy is Materials Nuclear Energy

RICI.EAR REGULATORY C(?MMl$6t@tl DOCICETfllG & SERVICE SECTION OFFICE OF THE SECRET ARY Of THE COMMISSION Document tatisliC$

,e u ':J F~

_ .ceived /

'4 C i s Re producf "2...

istribution

released under conditions that prohibit any transfer across the sample boundary.

This is the adiabatic temperature increase, shown in the last column of the tab le. Adiabatic conditions are largely a theoretical concept, difficult to achieve in practice without the special arrangements that can only be made in a laboratory. Second, the stored energy decreases sharply with increasing 11 11 distance from the core region, becoming 5.61 cal/gm at 18 1.34 cal/gm at 22 11 and unmeasurably small at 26

• Within the graphite island, the stored energy decreases from 33.3 to 19.2 cal/gm, in going a distance of only about 3 inches, from the fuel box edge toward the center of the island. Clearly it would be still lower at the center, which is 2 1/4" farther in the same direction (but where no meas urement would be made}.

Referring again to Figure 1, the graphite volume where adiabatic temperature increase of 100° to 139°C could be expected, seems to be limited to the center island and a zone about 3 inches wide just outside the fuel boxes. Further, the mass of graphite affected, which is only a few percent of the total, is in good thermal contact with the unaffected remainder.

The above resu l ts are consistent with what is kno~n of the graphite stored en-ergy phenomenon (Wigner effect, or Wigner energy).<2) The primary cause is col-lisions between energetic neutrons and lattice points in the graphite, creating displaced a toms that produce the extra graphite energy. The high energy neutron flux peaks near the fuel boxes in this reactor. Also, the graphite is cooler there which favors stored energy accumulation. Once the stored energy accumu-lates, it may be reduced as well as increased, by continued reactor operations depending on the temperature. If the temperature is higher than the previous level, the extra energy will be released as heat; that is, the graphite will be annealed. Or, the annealing could be induced by an external source of heat.

The released e nergy could be conducted away either to cooler graphite or to cooling water. The irradiation temperature is an important parameter, as well as the total exposure.

In real reactor situations, the energy release is neither instanteous nor adiabatic. For example, in a gas-cooled, natural uranium-fueled reactor, the peak graphite temperature reached on annealing was 80°C lower than in a lab-oratory test under adiabatic conditions with a sample of the same graphite.

Further, the time when peak temperature occurred was six hours( c1fter heating began, compared with about l hour for the laboratory experiment. 3J Graphite Stored Energy Calculations Most of the data base on stored energy as well as its analysis, relates to ex-perience in large natural uranium-graphite production or power reactors. <2)

Exposures are in the range 1021 to 1022 total nvt. (The early data are quoted in units of Mwd/AT). Total fluence in the small, highly enriched core research reactors can be as low as 1019 nvt. Also, their geometry results in sharply different neutron spectra and fluxes, and temperature distributions, from the large reactors.

The first attempt to calculate stored energy in an Argonaut research reactor seems to have been in the NRC-$ponsored generic study of a 11 a spec ts of the safety of this reactor design.(4J The calculation made use of a highly approxi-mate and simplified empirical model, which was applied to an essentially hypo-thetical Argona ut reactor. Although the treatment in Ref. 4 mentions the UCLA reactor, it does not employ the exact parameter values for the reactor (see below}. All subsequent calculations of stored energy in the UCLA reactor have been baseaon this same model. The. procedure. depends only on the thermal flux at a given position. For con serva ti sm, the 1oca tion chosen is the point of maximum thermal flux, which is at the center of the graphite island (Figure 1}.

At several places in the petition (top of p. 5, middle of p. 6 and top of p. 7}

CBE argues the NRC sponsored study gives a "wrong" and even "grossly under-estimated" value for the UCLA stored energy. My review of the stored energy ca 1 cul a tion in Reference 4 revea 1s that it uses approximate va 1 ues for the reactor flux and operating time. Insertion of the correct values improves the results.

CBG erroneously claims that their calculated minimum value (39 cal/gm} is sub-stantiated by our 33.3 cal/gm measurement. They overlook the fact that their calculation is based on the same model, (4} and therefore gives the stored energy at the ra hi te island center. Their error ignores the important fact that our 33.3 ca gm measurement ,s at the edge of the fuel box. As is shown above in this letter, our measurements indicate a graphite center value less than 19.2 cal/gm. Therefore, CBG's minimum value is more than 2 times too liTgFi

--hardly the close approximation asserted in the petition (p. 6).

In this assertion, as well as elsewhere in the petition CBG seems not to under-stand that the stored energy in the UCLA reactor type depends strongly on location in the graphite, and varies sharply with distance from the core.

(Figure 1 and Table 1}. Their failure to recognize this essential feature,as well as the demonstrated low stored energies actually measured, undermines their arguments about the importance of stored energy.

Finally, my calculated maximum value at the graphite center is 17~8 cal/gm(5}.

Although I used the same model as in Ref. 4 my calculational procedure was dif-ferent, as explained during my June 18 presentation. This result is lower than the geometrically closest measurement, 19.2 cal/gm only 3 11 away, at the fuel box edge; but the difference is in the sense to be expected, because of the higher temperature at the center. The small difference is probabl.Y. fortuitous, given my published reservations about the accuracy of the model Cl>. Nevertheless, the result shows that useful estimates of the stored energy can be made without measurement, at least at the graphite island center.

The petition's claim that estimates/ predictions of stored energy are not "reli-able" (with the exception, apparently of CBG's own calculations} is thus refuted by available facts.

The petition calls for measurements of the stored energy, in samples removed (periodically?} from the suspected regions of highest exposure. This would re-quire (periodic} shutdown of the reactor. Accessing the highest exposure graph-ite entails risks, radiological and otherwise. It was one thing to sample the UCLA reactor graphite, already partly dismantled. It is quite another to ex-tract samples from the core region of a functioning reactor, even if it is shut down. The petition ignores this problem.

The petition I s scenario for a cause-effect relationship between stored energy and combustibility (bottom of p. 6, top of p. 7} is literally impossible. The scenario postulates that stored energy may raise the temperature by "several hundred degrees", and lead to "graphite ignition". As is shown above in this letter, the affected graphite is only a few percent of the total. In the event of a stored energy release, means are readily available to transfer heat to cooler graphite, or to water.

A limited volume of affected reactor graphite, whose adiabatic release tempera-ture may be calculated to be several hundred degrees, simply does not equate to a graphite conflagration.

The petition has not demonstrated a 1inkage between stored energy and graphite fires *

• Graphite Fires The issue in research reactors is not whether graphite will burn, but whether it will burn under research reactor conditions. The petition does not demon-strate how this could occur.

The claim that *** "the Wigner effect" was a significant contributor to the Wind-sea le reactor fire" (petition, p. 5) is not supported by any reference. The reason may be that none could be found. In a review of the Windscale accident prepared by a committee of UK scientists, it was concluded that there was no direct contribution to the fire from the stored energy release itself. Mis-management of a deliberately planned stored energy release resulted in a rapid oxidatiof) Qf the uranium fuel, and the heat released led to some graphite com-bustion. l6J This has no relevance to sma ll research reactors, where the accum-ulation of stored energy overall is too low to warrant a deliberate annealing.

During normal operation of the RBMK reactor, unit 4 at Chernobyl, graphite con-ditions were vastly different from research reactors such as the Argonaut de-sign. Perhaps the most important difference was the large inventory of very hot graphite (up to 700°C). 7) This temperature effectively eliminates the accumulation of stored energy. However, i t creates a situation where a thermal excursion in any part of the graphite mass can only result in driving the tem-perature still higher. The reactor struc t ure was shattered, following a large Core Disassembly Accident that led to subsequent steam (or chemical) explosions, which fractured th~ otherwise massive graphite logs and exposed them to con-vective airflow.CS} Under these conditions, a self-sustaining graphite fire can and did occur. Yet, apart from the dramatic and thought-provoking aspects of this accident, it is not evident how it re l ates to small research reactors. The petition offers nothing beyond the statement that graphite burned at Chernobyl.

• Conclusion The petition does not substantiate its claim that massive graphite fires are credible accidents in U.S. graphite-bearing reactors, expecially the small re-search reactors. The petition further fails to demonstrate that graphite stored energy can contribute to such fires.

The requirement in the proposed rule to sample graphite from functional reactors (even if shut down for the purpose) could have radiological and other safety consequences that must be carefully evalua t ed. In the absence of a demonstrated need to conduct such sampling, it is unjustifiable to enforce it under an NRC rule.

I do hope you will reject the petition in its entirety. Please let me know if you have any questions.

Sincerely,

• y/OMf /J~

HARRY PEARLMAN HP:eh FIGURE l

• I October 28, 1986 l

'OORl'H GRAPHITE ISLAND

/ CENrER SAMPLES_

I I

I

~ ' l ' I V E GRAPHITE LAYER, I N D I ~ ll)RIZCNOO, CXXR>INATES CF SAMPUS,

~ RESEAlCH REACroR TABLE l

• I October 28, 1986 STORED ENERGY MEASUREMENTS, UCLA RESEARCH REACTOR GRAPHITE LOCATION COORDINATES STORED /! Tad SAMPLE H(a) y(b) ENERGY ESTIMATED (d)

DESIGNATION {CM}(IN) {CM}(IN) {J/GM) {CAL/GM) ,oc) 1N 13.5(5-1/4)S 5.4 .6(21-1/2) 104 (24.9) 103° 2N 13.5(5-1/4)S 38.1(15) 139 (33.3) 139° 3N 13.5(5-1/4)5 *68.6(27) 131 (31.3} 130° F4 5.8(2-1/4)S 61 (24) 80.3(19.2) 80° 1E 45.7(18)W 61 (24) 5.61(1.34) 5.6° 2M 55.9(22)W 61 (24) 1-¥<:\0.33) 1.4° 3W 66(26)W 61 (24) n.d.

NOTES (a) H-horizontal distance from graphite island center, south(S) or Wesi (W)

(b) V = vertical distance above core bottom plane (c) Not detectable on DuPont 1090 DSC (d) Based on assumed average Cp = 0 0.24 cal/gm - c REFERENCES (1) Ashbaugh, C.E., N.C. Ostrander and H. Pearlman 11 Graphite Stored Energy in the UCLA Research Reactor" Transactions of the ANS g, 372 (1986).

(2) Nuclear Gra~hite, Edited by R.E. Nightingale, Academic Press, New York/

London, (19 2).

(3) Dickson, J.L. and other 11 BEPO Wigner Energy Release", Proceedings of the Second U.N. Conference on Peaceful Uses of Atomic Energy, Geneva 1958, 7, 250 (1958). -

(4) Hawley, S.C., R.L. Kathren and M.A. Robkin "Analysis of Credible Acci-dents of Argonaut Reactors" PNL-3691 (NUREG/CR-2079), April 1981.

11 (5) Rebuttal to CBG's Wigner Energy Testimony", by H. Pearlman, November, 1983.

This is part of UCLA's testimony, submitted November 7, 1983, more than two years prior to the measurements.

Yet, in discussing UCLA's calculations of stored energy, the petition negligently ignores this written testimony. Instead, it refers only to a single, qualitative statement that I made, dissected out of the entire 130 page context of my testimony. Because my name is used in the body of the petition, I feel it is not inappropriate to make a direct reply:

(a) I did not use the words "real" or 11 realistic 11 in the statement. The petitiooerroneously attributes them to me (p. 6). The Transcript verifies that they are not my words.

(b) At the hearing, the intervenor immediately moved to strike the state-ment. A later formal motion, signed by the CBG president (August 26, 1983) again requested striking it from the record because, among other deficiencies such as not being supported by written testimony,it had been--and the fo 11 owing is a direct quote--* 1 thrown in I on re-re-di rect after two days of thorough examination of completely different conclusions. 11 (My underline).

But the petition now ignores everything but the "thrown in" statement.

(6) "Final Report of the Committee Appointed by the Prime Minister to Make a Technical Evaluation," etc. Cmnd 471, H.M. Printing Office, London (July, 1958).

11 The full title is very long. The report is usually called simply Cmnd 11 471 *

-a-

(7) "The Accident at the Chernobyl Atomic Energy Station and Its Consequences" USSR State Cammi ttee for Utilization of Atomic Energy, prepared for IAEA Conference, August 1986, Vienna.

(8) "Summary Report on the Post-Accident Review Meeting on the Chernobyl Accident". Safety Series No. 75-INSAG-l. International Nuclear Safety Advisory Group. Vienna, Austria, September 1986.

• OtlyKET NUMB&R /ii',

~ETITION RULE PRM 1 5 -</g/F,e ~134t)\;J/J NUCLEAR-FREE B ERKELEY COMgITTEE 48 Shattuck Square #177 Berkeley, CA 94704-1140 Len Conly 526-2746 Paul Miller 835-2635 "86 NOV -3 P2 :52 OFF ICE Ot .,_ v~1kcr".* October 31, 1986 OOCKETIHli l. S[

IR~NCH The Secretary of the NRC Commission Nuclear Regulatory Commission Washington, D.C . 20555 Attn: Docketing and Service Branch To: The Secretary of the NRC Commission We urge you to issue an amendment to 10 Code of Foo.eral Regulations Part 50 as requestoo. in PRM-50-44. In view of the Chernobyl disaster and the occurrence of a graphite fire at the Windscale reactor in England in 1957, it does not seem prudent that the NRC continue to hold to the position that a graphite fire is a non-credible event.

In particular , we urge you to order the preparation of an evacuation plan at the Berkeley Research Reactor in Etcheverry Hall on the University of California campus at Berkeley , California which would minimize the hazards of radiation exposure to people on the Berkeley campus as well as to residents of the surrounding communi ties , +~,c eve,r.t :.f S14cl,. o p*t ,

Thank you for your attention .

Sincerely yours,

~~

Len~ nly Secretary, Nuclear~Free Berkeley Committee

'It'

(

I Tl")RY COMMf§flJII 1

CE SECTION

'"RETARY JS~

1stics-k D 1~/Jt 'j'f,t, 8 (

.,du z

GE Stockholders' Alliance Against Nuclear Power P.O . Box 966

• Columbia, MD 21044 * (301) 730-0178 l)ea(ET NUMBiR '"

fETITION RUL PRM j-"J-44_

(5/ F£ J/$4 !) (I .B6 NOV -3 PS :09 Chairman Patricia T. Birnie October 31 , 8f~.,.1~N...; - *~

t. 1

, . r .,,

<;f ;,, ,;,r 1 fl ANC'1 ,*

Secretary of the Commission Board of Advisors U. S. Nuclear Regulatory Commission (In Formation) Washington, D. C. 20555 Larry Bogart Citizens Energy Council Attention: Docketing and Service Branch Leo Goodman (1910-1982)

Subject:

Petition for Rulemaking PRM 50 - 44 Split Atom Study Group

Dear Sirs:

Judith Johnsrud, Ph.D.

A Vice President We have learned that safety regulations are very inadequate

- Solar Lobby at the many research reactors in the US which use graphite.

Charles Komanoff Komanoff Energy We support a petition for rulemaking (PrlM 50 - 44) which Associates addresses many of these deficiencies . We urge you to adopt Claude Lenehan, OFM these suggested provisions without delay .

Corporate Responsibility Advisor The experience at Chernobyl gave a tragic demonstration of Paul L. Leventhal President, Nuclear the great danger of graphite fires and how diffidult it is Control Institute to extinguish them. The proposed new rules will :

Grigsby Morgan-Hubbard 1 . help reduce the possibility of a fire by requiring Writer and Energy Consultant measurements of stored energy in the reactor graphite; John R. Newell 2. require preparation of appropriate plans , equipment Bath Iron Works and training of personnel in extinguishing a graphite President (Ret.)

fire; and Miles H. Robinson, M.D.

A Citizens for J. require preparation and publication of an evacuation

- Health Inform ation plan for research lab personnel and all others working Nathan H. Sauberman or living near the reactor.

Professional Engineer (Ret.)

Whil e we understand that the authorized power l evel of most of these reactors is very small in comparison to nucl ear power John Somerville, Ph.D.

President, Union of American reactors, t his does not mean that the danger is proporti on-and Japanese Professionals ately less.

Against Nuclear Omnicide Irving Stillman, M.D. We urge that you establish rules that will safeguard public Physicians for Social health and reassure the public that no detail is being Responsibility overlooked in learning from the tragic experience of Chernobyl .

Faith Young Energy People, Inc.

Sincerely, Affiliations for Identification Purposes, only.

Patricia Birnie

U.S. NUaEAR REGUlATOi<Y COMMf5SION POCKETING & SfRVlCE Se°CTIO OFFIG er: T' i: ('r-r .... r * .,v o:=

It II IfC, I

2-,

~ *'

MARYIAND NUCLEAR SAFETY COALITION DOCKET NUMB&R fi>ETITION RULE PRM :JtJ-44 ~ DO(K[Tf'"

U5NHC (SI F£ ~l!J4!)

.86 NDV -3 Ps :oa Secretary of the Commission US Nuclear Regulatory Commission Washington, D. C. 20555 RE: PETITION FOR RULEMAKING PRM 50 - 44

Dear Secretary:

We urge the adoption of new rules to improve safety of U. S. research reactors using graphite .

Chernobyl ' s accident proved how dangerous graphite fires are, and how difficult they are to pu"l' out . Most community fire fighting systems are not trained or experienced in how to put out graphite fires . Using conventional techniques could compound and expand the damage .

The proposed petition for rulemaking (PRM 50-44) recognized the great need for better emergency planning, both for how to fight graphite fires effectively, and planning for evacuating laboratory workers and others who are nearby.

We feel it is important to adopt and implement these needed new rules .

Thank you for your prompt action on this important matter .

Sincerely,

~::~

P.O. BOX 902

• COLUMBIA, MARYLAND 21044 (301) 730-0178 * (301) 454-5601

U. 5. Fro"CltAR lEGULATORY COMMfSSICR DOCKETING & S~P 11rc c;{rr10M OFH, F --

Ad Special C1 1,

DOCKET NUMB&R VETITION RULE PRM .:Jd-44 4/:)

C~I F£ .3/~4!) ~

DOC K[]u, October 31, 1986 US NHC COMMENTS OF OHIO CITIZENS FOR RESPONSIBLE ENERGY, INC. (

• oc rtg* >Nov -3 P2 :45 ON PRM-50-44 on September 3, 1986 the NRC published in the Federal 8ff/(,£:.~ u* -t

• Register a notice o receipt Of a petition for rulemaking, CW#f.tLJNG&s:-Pv1Kr 0-44, filed by th Commit tee to Bridge the Gap. 51 FR 31341. BRANCH CF.

his tition seeks amendments to the NRC's regulations requiring operators of reactors using graphite as a moderator or reflector to consider in their safety _analyses the risks and consequences or graphite fires, including the preparation of response plans for graphite fires and the measuring of Wigner energy, OCRE supports PRH-50-44. The petition advances a sound and rational concept. The NRC's policy of regarding graphite fires e as non-credible events, however, in a non - credible and irrational position, particularly in light of the occurrence of a graphite fire at Chernobyl, Just as the THI accident woke people up to the reality of hydrogen generation and combustion during degraded core and severe accidents, which evoked the appropriate regulatory response (i.e., enhanced hydrogen control requirements for small and intermediate size pressure suppression containments), the Chernobyl accident should make the NRC consider the reality of graphite fires and again evoke the appropriate response.

The appropriate regulatory response is contained in PRH 4, Which OUld be adopted.

Respectfully submitted,

~~

suscm L. Hiatt OCRE Representative 6275 Hunson Rd.

Mentor, OH 44060 (216)255-3158

JJ,la NIJQ.EAR REGULATOR G & SE

()fflCE Of :rHE OF THE C Docu

~ o~ _ 1t/4!/__/,_ _

C..,l.uived Add' f Copi s Repr ~l ed s,ecW D'stribution

DOCKET NUMBER fETlTION RULE PRM 5lf-44

(~IF~ .;'l~4j)

GAS-COOLED REACTOR ASSOCIATES DOLK ETE '

10240 Sorrento Valey Rd. Ste. 3CX) JS C San Diego, CA 92121-1605 (619) 455-9500

.86 NOV -3 P1 :QB OFF IL::.* **rt. ~*

OCK EW ,11 Altv!Cf ANCI--!

October 31, 1986 Mr. Samuel J. Chilk Secretary of the Commission U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Attention: Correspondence and Records Branch

Subject:

Comments on Petition for Rulemaking on Graphite, 51 FR 31341, Request for Extension

Dear Mr. Chilk:

GCRA represents a number of U. s. utilities and potential users who are interested in the development and commercialization of gas-cooled graphite-moderated power reactors in this country. Because of the potential importance of the subject petition to such reactors and because of its relatively detailed technical content, we request a 90 day extension to the comment period to allow time for ourselves and other interested parties to prepare an adequate technical response.

Sincerely, L. D. Mears General Manager LDM:APK/dll cc: GCRA Management Committee A. Millunzi, DOE/HQ G. Northrup, GA R.O. Williams, PSC

DOCKET NUMBER PETITION RULE PRM ffd-44 l5! F~ .IIJ4!)

THE PENNSYLVANIA ST A TE UNIVE R SI To!c:,ffr(("

SNR UNIVERSITY PARK, PENNSYLVANIA 16802 v, C College of Engineering 1J6 OCT 31 n-, :Q4e 814

'Aria Breazeale Nuclear Reactor 865-6351 October 27, 1986 Fll'E OF SE 0 t<fT/NG ~ " ~VIC::f'.

!rt.A NCH Secretary U.S. Nuclear Regulatory Commission Washington DC 20555 Attn: Correspondence and Records Branch

Dear Sir:

Penn State Breazeale Reactor, License No. R2, Dockett No. 50-5, Request for Extension of Comment Period on the Petition for Rule Making Submitted by the Committee to Bridge the Gap The Pennsylvania State University, licensee of the Penn State Breazeale Reactor, respectfully requests that the comment period for the petition for rulemaking submitted by the Committee to Bridge the Gap be extended for at least 60 days beyond the published November 3, 1986, deadline.

With the timing of the comment period coinciding with the most active time of the academic calendar, we find that we have had insufficient time to adequately address the issues involved. The additional 60 days will allow us to prepare a more meaningful response to t he petition.

We thank you for your consideration of this request *

• m~~t)Jl Yours very truly, Marcus H. Voth, Director Assoc. Professor of Nuclear Engineering MHV/pka cc: D. Alger, Univ. of Missouri Tawfik Raby, U.S. National Bureau of Standards W. F. Witzig AN EQUAL OPPORTUNITY UNIVERSITY

I OJ._*t""

DOCKET NUMBER ~

PjTITION RULE PRM jd-44 {!!)

(_.SIF~ .J/~4(}

UNIVERSITY OF WASHINGTON ooc;KETED USNRC SEATTLE, WASHINGTON 98195 "86 OCT 30 P4 :1 3 College of Engineering Department of Nuclear Engineering FIC£ o~ -,C.vl ,..,f(Y OOCKE. Tlllu J S* RVICf.

B!'t~KCH Secretary Nuclear Regulatory Commission Washington, DC 20555 October 27, 1986

Dear Sir:

This Is a comment In response to the notice appearing In the

• Federal Register vol. 51, No. 170, of Wednesday, September 3, 1986, page 31341 regarding a petition by the Committee to Bridge the Gap CCBG) al leglng that the stored Wigner energy In graphite moderated ARGONAUT reactors represents a fire danger.

This al legation Is groundless and can be dismissed by consideration of some simple physics and knowledge of some experiments.

The CBG point out that measurements made at UCLA determined that the level of stored energy In the graphite of their reactor was "as high as 33.2 cal/gm". A smal I volume of even this level of stored energy does not represent a hazard nor does It represent the average level of stored energy In the bulk of the graphite.

In addition they claim that their calculations Indicate that by the year 2000, the stored energy would have reached a minimum of 113 cal/gm corresponding to 39 cal/gm In 1983 *

• If the measured maximum stored energy Is 33 cal/gm, their esti-mate of a min I mum va I ue of 39 ca I/gm Is a gross overestimate. If the Ir est I mate represents a mI n I mum v a I u e for the max I mum, It 1 lkew lse Is f I awed. There ts no reason to expect that the duty cycle of the UCLA reactor, had It continued to operate, would have been any different In the next 17 years of Its I lfe than It had been In the first 22 years. Thus, the assertion that the stored energy would have Increased by a factor of nearly 3 fold (39 cal/gm In 1983 to 113 cal/gm In 2000) during a less than doubl Ing of operational time Is unjustified. Taking an effective start of 100 KW operation to be In 1962 and extrapolating their year 2000 figu r e back to 1983 on the basis of operation time Imp Iles a va I ue of stored energy of about 22/39 times 113 ca I/gm or about 64 ca I /gm, near I y doub I e what was observed as a max I mum.

Their calculations are thus clea r ly In error.

lnaddltlon to the error In the CBG estimate of the maximum level of the stored energy, the measurements of the center Is I and graphite In the UCLA reactor show that the average stored energy Is substantially below the maximum.

Benson Hall, BF-10 / Telephone: (206) 543..2fJ.f Q'Ci S 1 ,9 Acknowledged by card .. ......................,

U. $. NUCLE, < ISSION

,\ I OF .r, b-

• Pos~mark D c C 1es k A C'o

The effectiveness of a neutron f or causing a graphite lattice displacement Is dependent on the neutron energy, wi th higher energy neutrons being more effective. Depending on how the weighting Is done, the average effective fast neutron flux In the graphite center Island may range between 10% and 50% of the peak value near the fuel boxes with the smaller value corresponding to a higher effective energy. Even If one takes, for the sake of a very conservative estimate, an average value of 50% of the observed maximum, or 16.5 cal/g m and assumes adiabatic heating which adds additional conservatism, then the fol lowing estimate of the temperature rise In the center Island graphite can be made.

The specific heat of nuclear grap h ite Is given In tabular form In the book "Nuclear Graphite", (R.E. Nightingale, ~~.e..a.r:. Graphite, Academic Press, New York, p.122). These data may be f It to the analytic function Cp=1.86+8.34E- 3*T-5.27E-6*T 0 2 cal/mole over the

• temperature range between 25 a nd 650 degrees centigrade (S.C.

Hawley, et al, Analysis of Credible Accidents for Argonaut Reac-tors, NUREG/CR-2079, Apr I I, 1981). This expression may be Inte-grated to give the specific energy required to raise the tempera-ture adlabattcal ly anywhere within the fitted range. The adiaba-tic release of 16.5 cal/gm throughout the graphite would result In a temperature rise of under 110 degrees centegrade. Even without cool Ing by the circulating water, this temperature Is far below the melting point of both the aluminum cladding and the fue I a I I oy. SI nee the actua I average stored energy Is much smaller and since heat conduction In the graphite Is high, the actual temperature rise would be much lower.

In addition to the above arguments which Indicate that there Is I ntr Ins lea I I y no hazard from s to red energy, there Is now c I ear exper I manta I ev I dance that graph lte b I ocks wI I I not susta In com-bu st I on on their own. Experiments done at Hanford In which a block of graphite was heated In air until white hot by an oxyacetylene torch showed that as soon as the torch was removed, the graphite cooled rapidly. The only effect after sustained heating was a sl lght erosio n under the flame. The red glow of the Cher nob I y graph lte Is more I Ik e I y to have been the resu It of lncadescence due to both the nor ma l operating temperature of the reactor plus any additional heat generated by fission product decay, metal-water reactions, metal-air reactions, hydrogen burn-Ing or oxidation of hot fine gr ap hite particles produced In the exp I os Ion. There Is, In fact, no ev I de nee that there was a graphite stack "fire", that Is, self-sustained combustion of massive graphite, either In the Chernobyl reactor or In the Wlndscale reactor.

An analytical model developed for determining the oxidation/combustion characteristics of TSX graphite blocks containing a channe l through the center of an 8"x9" unit eel I of graphite preheated to over 500 degrees centigrade showed that the chemical reaction of air with graphite was Inconsequential (W.J.

Quapp, private communication to M.A. Robkln). Decay heat was the dominant energy source for subsequent cor e heatup. The combustion effect on the temperature of the blocks was only a few degrees.

Graphite combustion was determined to be bas teal ly not possible for any credible temperature and air flows. Graphite wt I I ox Id I ze at h tgh temperatures. However, In I arge b I ock geometry, the surface to volume ratio Is so low that the kinetics of the surface reactions combined with mass transfer I Imitations of oxygen transport to the surface of the graphite does not provide enough chemical reaction to result In a fire *

• For graphite to burn In self-sustaining combustion requires that It be finely divided and be exposed to a steady supply of oxygen.

These conditions are not met In ARGONAUT reactors which have large graphite blocks, are contained under a massive concrete shield, and have limited airflow around the graphite. There Is no credible operational event In Argonauts which can disrupt the graphite sufficiently to produce either large quantities of hot fine particles or large air flows or both, sufficiently large to cause significant graphite oxidation.

In summary, the stored Wigner energy under the most pessimistic assumptions Is Insufficient to heat the graphite to a temperature wh !ch can damage the fue I and even If heated, the graph lte wI I I not burn. Wlgner energy and graph lte f I res In ARGONAUT reactors are non-Issues.

In addition to the above comments, other comments by the TRTR community wl 11 be forthcoming. In order to al low suff le lent time to prepare these additional comments, a 90 day extension of the comment period untl I February 3, 1987 Is needed.

~ : y, a /?L Maurice A. Robkln Professor c.c. W.G. Vernetson, University of Florida

DOCKET NUMBER PETITION RULE PRM ..j'?J-44 A)

(51 F/2 -=31$4!} (?,I 414 Rose Avenue--Bo~~fSTE Aromas, California 95~~C 27 October, 1986 U.S. Nuclear Regulatory Com ission "86 OCT 30 P3 :31 Attn: Docketting and Service Branch Office of the Secretary Off ICt. .., it\t< v Washington, DC 20555 IICKfTIN(j

• I: PIVICf IRANCli RE: PRM 50-44

Dear Sirs:

understand that a Petition for Rule-Making has been filed with the NRC to require operators of graphite-cored research reactors to more realis-tically prevent graphite fires, and to prepare adequate response plans in the case of such fire outbreak

• I support this petition and urge you to adopt strict guidelines to prevent fires at graphite reactors, which have to this date been relatively unregulated.

After Chernobyl I do not think I have to stress the urgency of having a plan for responding to fires at graphite reactors. Though the rela-tive size of U.S. graphite reactors is much smaller, their locations on col-lege campuses and in urban areas makes mishaps similar to the Chernobyl and Windscale <1957) events equally dangerous. In all the press attention to the Chernobyl tragedy, I nowhere saw it noted that any U.S. cities host reactors with similar designs, likewise without con t ainment buildings.

Were a fire to break out today, ost local fire-fighting units would be as unprepared as were the Russians for the type of response necessary.

Requiring reactor operators to develop a plan for fire-fighting (and evacua-tion of the local area) seems essential to me *

• What must be pointed out from a scientific perspective, however, is the special danger of graphite fires--a danger that the NRC has heretofore underestimated. New data presented at the Annual Meeting of the A erican Nuclear Society indicate that the Wigner energy in the graphite of older reac-tors is much higher than expected. UCLA researchers easured as much as 33.2 cal/ gram, as opposed to the 5 cal/gra the NRC estimated.

This energy can help cause graph ite fires, and would certainly in-crease the size of the fire were one to begin from whatever cause, it seems minimally necessary that reactor operators learn how uch Wigner energy is present. Only by periodically measuring the energy can the operator develop an effective plan for preventing disaster.

Please approve the petition for rule- aking submitted to you.

PhD.

98611 6 l.00

~ledged by card ****** , , ** , ,;. *,, * ,w

lJ.

S. NU CLE,\*~ RECU:../\TCRY C  ::.:iSSION OOCK t.:TING & Sf Hi'",: c;.. l::rl OFFI CE GF TH:.. -- !Y Doc:..rr.cn .. "S Pos!ma rk Da e /0 2-. ?

Copies Hcc:e .cJ

---

'"--i'--- -=------~

/__ _ _ __ _

Add') Co-  :; r *, **

I ~pecial D,~, l~ul,\., 1

- ---- - - - -----1

DOCKET NUMBER PETITION RULE PRM .!J-4,4 NILS J. DIAZ, DIRECTOR W.G. VER NETSON, REACTOR MANAGER NUCLEAR FACILITIES DIVISION (S-/ FJ! ~/J4!)

NUCLE AR REACTOR BUI LDI NG UNIVERSllY OF FLORIDA DOCKET GAINESVILLE, FLORIDA 32611 PHONE (904) 392-1429 TELEX 56330 USNHr October 20-",1986 oo OCT 29 A9 :43 0 Fl ._ _ A .,

Secretary DOCKETING & il kVtr*i U.S. Nuclear Regulatory Commission BRANC~ '

Washington, D.C. 20555 Attention: Correspondence and Records Branch Re: Committee to Bridge the Gap; Petition for Rulemaking

{Docket No. PRM-50-44)

Gentlemen :

• With reference to the petition for rulemaking, the Univer-sity of Florida Training Reactor {R-56 License) would like to re-quest an extension on the November 3, 1986 deadline date for re-ceiving comments. We are in the process of developing a detailed set of comments. However, the extent of the petition and the breadth of areas potentially involved require that we have more time to develop a well-supported set of comments. Specifically, the petition for rulemaking requires that a broad spectrum of conditions be considered concerning the credibility of graphite fires and the extent of buildup of Wigner Energy in the graphite of research reactors. The wording of the proposed amendment im-plies potential far-reaching administrative consequences for re-search reactor facilities. To respond properly necessarily in-volves considerable resource allocation to assemble data sources, evaluate data and determine applicability especially in light of specific design features *

• Because of the time and resources that must be directed to develop the response to this petition for rulemaking, we would respectfully request a 90-day extension on the November 3rd dead-line date for making comments on the petition for rulemaking and the proposed amendments to 10 CFR part 50. We feel this extension i s well justified by the need and desire for well-supported com-ments on the proposed amendments. * >1Y.MI'" , u,~.,lJ. 1o~~i; Thank you for your consideration of this request.

Sincerely ,

w~AJ u~

William G. Vernetson Director of Nuclear Facilities WGV/ps cc: P.M. Whaley J.S. Tulenko Reactor Safety Review Subcommittee EQUAL OPPORTUN ITY/AFFIRMATIVE ACTION EMPLOYER

V U.S. NUCLEAR REGUl.ATOIY COMMISSICti DOCKETING & SERVICE SE'cTION OFFICE OF THE SECRET ARY OF THE COMMISSION Document Statistics Postm!lf'lc Date /t'ft-4 I

Copies Recei ed _ _/_

Add' I Copies Re produced Special Distributien

DOCKET NUMBER PETITI N RULE PRM ~?J-4,4

($/ p,e ..J/~4?)

DOW CHEMICAL U.S.A. DOlKETE.r USNRC.

October 10, 1986 Secretary U.S. Nuclear Regulatory Commission Washington, D. C. 10555 ATTN: Correspondence and Records Branch RE: GRAPHITE IN RESEARCH REACTORS: DOCKET NO. PRM-5O-44

Dear Sirs:

The Dow Chemical Company owns and operates a TRIGA nuclear research reactor which utilizes small amounts of graphite as a moderator. The peti~on for rulemaking announced in the Federal Register of September 3, 19'>, impacts on the company, which wishes to present the following comment:

The Dow Chemical Company requests that the Commission extend the comment period for a minimum of 90 days beyond the November 3, 1986, deadline announced in the original notifi-cation. The Dow reactor staff is working with the manufac-turer of the fuel (GA) and with other members of the Test, Research, and Training reactor (TRTR) group to prepare a more detailed and technically competent response to the request for comments.

Very truly yours, C. W. Kocher Analytical Laboratories Inorganic Materials Science and Characterization 1602 Building AN OPERATING UNIT OF THE DOW CHEMICAL COMPANY

U. S. NUCLEAR REG I AT RY COMMISSI DOCKET! G & e r SECTION omr. T RY OF vN r

,cstmark D* 1~/;1 Copies Add'I c

. I Special Di r I ~ 1~2,1/2,7,, s

Certification #P 645 196 984 DOCKET NUMBER PETITION RULE PRM t)°d-44

( ~/ Fte .J/~41 STATE OF RHODE ISLAND AND PROVIDENCE PLANTATIONS 00(.K£TED USNRC

~RHODE ISLAND ATO MIC ENERGY COMMISSION Nuclear Science Center .86 OCT 27 P2 :45 South Ferry Road Narragansett , R .I. 02882-1197 OFF IGE r: ,..,

OOC ET IN1..1 .,f Vlf.f October 22, 1 986 BRA c1.;

Secretary

u. S. Nuclear Regulatory Cornmisison Attn: Correspondence and Records Branch Washington, DC 20555 Gentlemen:

Please refer to the Federal Register notice Vol. 51, No.

170, September 3, 1986, page 31341 , concerning a request for comments on a proposal for rulemaking from the Committee to Bridge the Gap, Docket Number PRM-50-44. The proposal for rulemaking concerns the use of graphite in reactors. The comment period currently extends to November 3, 1986.

The research reactor operated by the Rhode Island Atomic Energy Commission does utilize graphite. We are, therefore, anxious to comment. It is, however, impossible to meet the current deadline because of the short time between the pubic notice and the November 3 deadline.

The analyses required to show that graphite can be safely used in this research reactor are complex. In addition, since these analyses have not been performed in recent years, considerable time is required simply to review the literature and perform the calculations on a plane appropriate to the problem.

In addition, we are currently approaching a time of year which is both busy and uncertain with end-of - year inspections, end-of-year academic requirements and traditional time off for staff holidays.

For these reasons, it is requested that the comment period be extended for an additional 90 days.

Thank you for your consideration.

Z1-~tJi&r ~

A ; Fran cis DiMe glio Di r e c tor

• AFD:cd

J. S. NUCLEAR REGULATORY COMMISSl0Jil DOC KET ING & SERVI CE SE.CT ION OFFICE OF H'E s~r 0 ET 'RY OF TIE -o ':"ION Po*tmork Cc i s 'I Ad,..'I lpecial Distr.' 1 ,

DOCKET NUMBER ION RULE PRM S~-4,1 (51 Fl2~!E4 /)

• DOC KU EL Jlll------------------ GATechnologies--------***--------*

GA Technologies Inc.

P.O. BOX 85608 .86 OCT 27 AlO :25 SAN DIEGO, CALIFORNIA 92138 (619) 455-3000 OFF ICE ~ t At r Law Department QOCKETIN '" E~v,cr BRANCH October 20, 1986 Secretary of the Commission

u. s. Nuclear Regulatory Commiss i on Washington, D.C. 20555 Re: Docket No. PRM-50-44

Dear Sir:

This is to request an extension of 90 days in the time within which comments are invited in response to NRC's notice that it has a petition for rulemaking concerning fires and Wigner energy measurements in reactors which use graphite as moderator or reflector. The petitioner's claims are rather sweeping and its accusations that NRC and l i censees lack credibility demand a thorough response.

Very truly yours, JPH:jr 10955 JOHN JAY HOPKINS DR. , SAN DIEGO, CALIFORNIA 92121

1J NUO.EAR REGULATORY COMMISSI 00CKETING & SEDVICE cECTION o~ - ~ -,ARY ON

DOCKET NUMBER PETITION RULE PRM SIJ-4t/ 7590-01 (57 !=£. .31.84t)

Nuclear Regulatory Corrmission DOCKETE f 10 CFR Part 50 USNRC Corrmittee to Bridge the Gap; Petition fo~u11rfa23'19P5 :Ql Extension of Conment Period OFFICE Of :>L 1..Kt AK Y

[Docket No. PRM-50-44] DO KETINli & SE v1r:r:

BRANCH AGENCY: Nuclear Regulatory Commission.

ACTION: Petition for rulemaking: Extension of conment period.

SUMMARY

On September 3, 1986 (51 FR 31341), the NRC published a notice of receipt of a petition for rulemaking filed by the Committee to Bridge the Gap. The petition requested that the Conmission amend its regulations to requ ire operators of reactors that use graphite as a moderator or reflector to (1) prepare and submit for NRC approval fire response plans and evacuation plans for a graphite fire and, (2) measure the energy stored in their graphite, and revise their safety analyses to consider the risks and consequences of a graph i te fire in their facilities. The notice of receipt requested public

- corrment on the petition and established a corrment closing date of November 3, 1986.

In response to requests from the U.S. Department of Conmerce, University of Missouri, Oregon State University, Worcester Polytechnic Institute, and North Carolina State University, the NRC has agreed to extend the comment period on PRM-50-44 for 90 days from the original comment closing date.

DATE: The comment period for PRM-50-44 has been extended from November 3, 1986 to February 3, 1987.

ADDRESSES: A copy of the petition for ru l emaking is available for public inspection in the Commission's Public Document Room, 1717 H Street, NW.,

Washington, DC. A copy of the petition may be obtained by writing to the Division of Rules and Records, Office of Administration, U.S. Nuclear Regulatory Commission, Washington, DC 20555.

Al l persons who desire to submit written comments concerning the petition for ru l emaking should send their comments to the Secretary of the Commission,

- U.S. Nuclear Regulatory Commission, Washington, DC 20555, Attention:

Docketing and Service Branch.

FOR FURTHER INFORMATION CONTACT: Michael T. Lesar, Acting Branch Chief, Rules and Procedures Branch, Division of Rules and Records, Office of Administration, U.S. Nuclear Regulatory Commiss i on, Washington, DC 20555, Telephone: 301-492-7758 or Toll Free: 800-368-5642.

Dated at Washington, DC this ~~ day of (!J~P'U,/ 1986.

For latory Commission.

Samuel J. Chilk, Secretary of the Commission.

DOCKET NUMBER .

PETITION RULE PRM ~1k.11.B4!)

UNIVERSITY OF VIRGINIA DOCK[i *r, DEPARTMENT OF NUCLEAR ENGINEERING AND ENGINEERING PitMsllis NUCLEAR REACTOR FACILITY SCHOOL OF ENGINEERING AND APPLIED SCIE1l!E Der CHARLOTTESVILLE, VA 22901 22 p4 :53 October 10, 1986 8~f1-fft?mtr;r-:,::

/JJ: .uf:hi\iw-t(t~4-7136 fR .

IFUNC~ VIC{

Secretary Attn: Correspondence and Records Branch U.S. Nuclear Regulatory Commission Washington, DC 20555 Re: Committee to Bridge the Gap; Petition for Rulemaking 10 CFR Part 50

[Docket No. PRM-50-44]

Dear Sir:

Our university research reactor facility first learned of the GAP petition for rulemaking at the September 25-26, 1986 annual Test ,

Research and Training Reactor (TRTR) Meeting, when this information was presented by NRC personnel. Given the nature of the subject of the petition, which requires a time consuming review and analysis of the available scientific literature on the subject of Wigner energy deposition in graphite by radiation, the allotted comment period with the November 3 , 1986 deadline is far too short. Accordingly, we request that the NRC grant a longer extension of the comment period .

At this time we would like to convey our initial impression that the concern expressed by GAP is unjustified , and that the petition should therefore be rejected on grounds of insufficient merit. We expect to make the case for our facility, and for similar reactor facilities, in the near future .

R UVA Reactor Facility Sworn to and subscribed b~fore me this N/2 day

. of -f.aLL /) * -I-______ '19--Ll._0 CJ/

~ -,,,..,,,_ '

- -~ rfat"y Public

U.S. NUCLEAR REGUI ATORY COMM(~S DOCKETING & sr 0 VICE SECTION OFF trE QF T'--

• SECRET ARY OF T! E CvW iSS ION

COCKE:. T N'UMBER J..

P.£.Tl110N- RULE PRM 5J-,14 ~

UNIVERSITY OF CALIFORNIA, LOS ANGELES

. ( §'I r-,e .31341) UCLA OO~TEO BERKELEY

• DAVIS

• IRVINE

• LOS ANGELES

• RIVERSIDE

• SAN DIEGO

• SAN FRANCISCO u "* SANTA CRUZ

.86 OCT 21 P3 :07 COMMUNITY SAFETY DEPARTMENT OFFICE OF RESEARCH 6c OCCUPATIONAL SAFETY Qff'~~c~~A~ 90024 DOC KCTING .,, : : F" ICF BR N .

October 13, 1986 Chairman Lando Zech, Jr.

Commissioner James Asselstine Commissioner Frederick Bernthal Commissioner Thomas Roberts U.S. Nuclear Regulatory Commission Washington, D.C., 20555

Dear Sirs:

Committee to Bridge the Gap; Docket No. PRM-50-44 Petition for Rulemaking I submit the following comments concerning both the credibility and relevance of CBG's Bases for this Petition.

1. The Chernobyl graphite fire has no relevance to the use of graphite in nonpower research reactors. The normal operating temperature of the graphite in the Chernobyl reactor was in the range of 7oo 0 c. 1 This temperature is above the minimum temperature of 65o 0 c2 at which a self sustaining oxygen-graphite reaction might occur.

Hence, any event that released a substantial amount of energy and at the same time breached the inert gas blanket isolating the graphite from air could perhaps cause a graphite fire to start (this did happen at Chernobyl). Further, the extremely large mass (1700 tonnes) of very hot graphite could not easily dissipate the energy generated by the beginnings of an oxidation reaction (fire).

Hence, the temperature of the graphite stack would rise further and increase the rate of oxidation. Finally, the numerous vertical channe l s in the Chernobyl reactor allowed air to reach the hot graphite and hence sustain the oxidation reaction.

None of t hese conditions occur in U.S. research reactors. All operate well below the boiling point of water (100°c): all use relatively small quantities of graphite (even an Argonaut contains only 4 tons) and hence have large area to volume ratios which greatly increase the dissipation of internal heat; all, except for Argonauts, operate in a large pool of water with the graphite in sealed cans; and none of the graphite stacks are breached by hundreds of channels to provi de pathways for air to reach the graphite. There is simply nothing about the Chernobyl reactor graphite fire that has relevance to non-power research reactors.

2. The issue of stored Wigner energy is also of no relevance because of the geometry of research reactors. That i s, all research reactors have concentrated fuel cores, surrounded by graphite OCT 2 2 1986 Acknowledged by card *** , , , , , , , , , , ** -'* ,w

If.I. NUCLEAR ~,..

/IJ/tf, j>le r I Add'I C ial IJ z

reflectors. The significance of this point is that a very small volume of graphite is irradiated by fast neutron flux. Hence, only a small fraction of the graphite is subject to lattice displacement with subsequent Wigner energy storage. For example, in the UCLA Argonaut, less than 6.0~ of the graphite was irradiated by fast neutrons. No matter what the magnitude of the Wigner energy stored, there is a tremendous volume of unaffected graphite to absorb the single energy pulse that might be released from the affected graphite. Thus, the overall temperature of the stack would increase very little.

Note that in those research reactors immersed in pools containing thousands of gallons of water, the release of stored Wigner energy is probably impossible. The immense heat sink provided by the water precludes heating the graphite to the temperatures required (several hundred degrees centigrade) for Wigner energy release *

• 3. Since Mr. Aftergood quoted my testimony in his petition to the NRC, I feel I have the right to comment on the credibility of his arguments. He has used the standard tactic of CBG of carefully picking bits and pieces of facts from the literature, quoting them out of context, and totally ignoring the vast body of scientific data that totally refutes his claims.

I did in fact make the statement quoted by Mr. Aftergood on pages two and three of his petition. However, if you read all of my testimony 3 (and I urge you to do so before making a decision on this petition) you will find some interesting facts about graphite "burning".

a. Graphite combustion is a surface diffusion controlled phenomenon. ~ That is, the carbon atoms do not vaporize and leave the surface to react with oxygen, but rather the oxygen must reach the solid surface before combustion occurs. This is completely unlike the burning of wood or other combustible materials, and in fact is more analogous to the oxidnt~on (rusting) of iron. (Note even iron will burn if it is heated to a sufficiently high temperature and if oxygen is forcefully directed at the molten surface.) - - -- --

b. Mr. Aftergood fails to mention that the cited ignition 0

temperature for graphite of 650 C was measured under certain very special conditions: 1) Forced airflow through graphite channels; 2) An external heat source to maintain the necessary high temperature since the oxidation of the graphite alone did not maintain the necessary temperature; and 3) Very small graphite samples with large surface to volume ratios.

The significance of the latter two conditions is that the loss of energy by radiation from relatively small quantities of graphite is so high that it is impossible to maintain the high temperature necessary for continued oxidation of the graphite.

It is not possible to take the conditions that exist in massive quantities of graphite heated uniformly throughout to very high temperatures (Windscale and Chernobyl reactors) and scale them down to apply to small quantities of graphite, heated locally to a few tens of degrees (<100) centigrade. The implication that the conditions that resulted in the graphite

--2--

"fires" at Windscale and Chernobyl have any relevance to small graphite reflected research reactors, is either based on scientific ignorance or scientific dishonesty.

c. Finally, I want to point out a graphic example of how Mr.

Aftergood twists words to give a decidedly wrong impression. In his cover letter to the NRC Commissioners he states: "Put another way, sufficient Wigner energy could be stored to in effect lower the 650 ° ignition temperature by several hundred degrees." Aside from the fact that the 650 0 temperature applies only to experimental conditions that do not exist in research reactors, the storage of Wigner energy, per se, has no affect on the ignition temperature.

This proposed rule making would make serious cost impacts on all research reactors and would in fact be impossible to meet for those reactors in which each graphite stringer is sealed in an aluminum can

• The integrity of such cans would be destroyed by the sampling process and hence sampling would be useless. If this proposed rule is implemented, I suspect that one would simpl y throw the old graphite away and replace it with unirradiated graphite. Obviously, such a course of action is a costly process that would probably shutdown many University research reactors (probably the real hidden agenda behind this petition). I have already made it clear, that in my opinion there is no benefit to be gained by measuring stored Wigner energy in research reactors, (i.e., no increase in public health and safety), and I strongly urge that this petition be denied in toto.

Very Truly Yours,

• WFW:si Walter F. Wegst, Ph.D.

Director, Research &

Occupational Safety

--3--

Notes

1. "The Accident at the Chernobyl AES and Its Consequences", State State Committee for Using the Atomic Energy of the USSR, prepared for the International Atomic Energy Agency Expert Conference, 25-29 August, 1986, Vienna, translated from the Russian, Department of Energy.

2. "Experimental Evaluation of the Combustion Hazard to the Experimental Gas-Cooled Reactor -- Preliminary Burning Rig Experiments", R.E. Dahl, HW-67792, Nov. 1961.

3. "Testimony of Dr. Walter F. Wegst Concerning the Safety of the UCLA Research Reactor," in NRC Docket No. 50-142, transcript of the July 25, 1983 hearing *

• 4. "The Prediction of Conditions for Self-Sustaining Graphite Combustion in Air", J.S. Nairn & V.J. Wilkinson, Proceedings of Conference at 0RNL, 1960, TID 7597

--4--

.Ovv ,

• I I * ...,,.,[JER PETlTION RULE PRM fl -44

. (~Ir-~ .Jl~4T) ._v. "'~

North Carolina State University DOLK U5 School of Engineering

'°l'IOGll"\1' Department of Nuclear Engineering

• 86 OCT 20 p2 :o9 Nuclear Reactor Program Box 7909 Zip 276g5-7909 OFFt c  ;; .

(919) 737-2321 DOCK Tt G ' t A1d1 BRANC ,.v,ct October 14, 1986 Secretary U. s. Nuclear Regulatory Commission Washington, DC 20555 9

Dear Sir:

I am writing in regard to the "Committee to Bridge the Gap; Petition For Rulemaking" (Docket No. PRM-50-44) concerning the proposed rule for nonpower reactors that contain graphite.

In order to develop a proper response, it is requested that the Commission delay the deadline 90 days for receiving comments.

The "Basis for Petition" provided by the Committee to Bridge the Gap require detailed review that cannot be adequately accomplished in the existing comment period. Rushing this proposed rule with -

out a thorough review would be extremely detrimental to the long term viability of research reactor facilities across the United States.

In summary, we feel that an extension in the comment period is in order.

Sincerely, Bernard W. Wehring Director, Nuc l ear Reactor Program BWW:edt North Carolina State University is a Land-Grant University and a constituent institution of The University of North Carolina.

U. S. UCLEAR P~r-*1

• CKETI y

Po n r Cc A

Spec

DOCKET .NUM~ER P.ET\TlON RULE PRM ~-44

/,5; Fl! ..J/j//-z}

[))O[P WO RCESTER *- Worcester POL YTECHN IC (_ c Massachu~~~~g INSTITUTE (617) 793-SdotNHC "86 OCT 16 p 3 :56 October 9, 1986 Secretary U.S. Nuclear Regulatory Commission Washington, DC 20555 Attn: Correspondence and Records Branch

Dear Sirs:

In response to the petition for rule making docket no. PRM-50-44 concerning stored energy in graphite at nonpower research reactors, I respectfully request that the conrnent period be extended so that a more extensive evaluation of the technical aspects of this proposed rule can be made.

This proposed rule could seriously affect operatons at some nonpower research reactors, therefore, every effort should be made to obtain valid technical information. I believe that the extension of the conrnent period would make this technical information more reliable and complete.

Sincerely, Thomas H. Newton, Jr.

Director, Nuclear Reactor Facility THN:ama

s V

J NO 3 1 :10 J.'H'v' 1 :DI.HO 13)000 MlJOlll ~ 'D

DOCKET NUMBER P.Er1r10N RULE PRM .:r0-44 ...

r (ff'! F£ JI.J4!)

UO(KCTEr.

USNRC Oregon state.

Radiation Center Umvers1ty Corvallis, Oregon 97331 csoa>

"86 OCT 15 p 3 :so 154.2341

~

FFict o~ si t. '"t ..

OCKETING 31. Vl(.f BRANG~

October 10, 1986 Secretary U.S. Nuclear Regulatory Commission Washington, D.C. 20555 ATTN: Correspondence and Records Branch

Subject:

Oregon State University, License No. R-106, Docket No. 50-243; Request for Extension of Comment Period on the Petition for Rulemaking submitted by the Committee to Bridge the Gap Gentlemen:

Oregon State University would like to respectfully request that the Nuclear Regulatory Commission extend the comment period on the above-mentioned petition for rulemaking for a period of 60 days beyond the published Novem-ber 3, 1986 deadline.

In support of this request, we submit two primary reasons why we feel it would be beneficial to extend the comment period. First, we did not receive the Federal Register listing of the petition for rulemaking until the last week in September and did not have an opportunity to discuss the details of the petition with other members of the research reactor community or with representatives of the Commission's staff until the annual meeting of the National Organization of Test, Research and Training Reactors, which occurred September 25 and 26, 1986. Upon returning from this meeting, it became apparent that we would have only approximately 30 days to prepare our comments in order to meet the November 3rd date.

We feel that this is a very tight time constraint which will not promote in depth responses by licensees, and is a limitation which exists for essentially all of the affected non-power reactors.

A second consideration, which is perhaps applicable only to Oregon State University, involves the fact that our fall academic quarter began about October 1, 1986. As you can imagine, this is the busiest time of the entire year for our staff, and it is therefore very difficult for us to assemble the necessary personnel to prepare an appropriate response Oregon State University is an Affirmative Action/ Equal Opportunity Employer

tm'Cl:tAlf lfECl.JtAi01h' CO'MMt~

DOCK T *r; c :ov1cE SECTION

• 'T RY Ju-(/!/

/

J:2

/td.r; /%:t/_;--

Page 2

• U.S. Nuclear Regulatory Commission October 10, 1986 at this particular time. Providing an add itional 60 day comment period will significantly help Oregon State Un ive rsity, and from our viewpoint the rest of the community as well, to perform a more thorough evaluation of the petition and to submit a more meani ngful response.

We thank you for you r consideration of this request.

hnson Actin Director, Radiation Center Reac tor Administrator AGJ:jrl cc: Don Alger, University of Missouri Tawfik Raby, U.S. National Bureau of Standards Marcus Voth, Pennsylvania State Un iversity T. V. Anderson, OSU S. E. Binney, OSU B. Dodd, OSU

DOCKET NUMBER

~?J-,44

r..

PETITION RULE PRM

(_511=eJ1J41) Research Reactor Facility OOlKETE USNRC UNIVERSITY OF MISSOURI Research Park October 1, 1986 Columbia, Missouri 65211

• 86 OCT -8 PS:\ 7 Telephone (314) 882-4211 Secretary U. s. Nuclear Regulatory Commission Washington, D. C. 20555

• Attention:

REFERENCE:

Correspondence and Records Branch Docket 50-186 University of Missouri Research Reactor License R-103

SUBJECT:

Request extension of comment period for the petition for rulemaking (Docket No. PRM-50-44)

The University of Missouri Research Reactor , in the interest of gathering more information regarding the petition for rulemaking, Docket No .

PRM-50-44 , requests that the comment period for this petition be extended 90 days .

This additional time period is requested to allow a thorough review of the original petition, requested by separate letter from the Division of Rules and Records, and hence a more comprehensive comment regarding the petition .

wmh n Walt A. M~~-V Acting Reactor Manager Endorsement:

Reviewed and Approved

~c,._ \'I\ 'l!G.V\

Don M. Alger \

Associate Director COLUMBIA KANSAS CITY ROLLA ST. LOUIS an equal opportunity institution

• t I

• u.s, NUCtEAR RCG:Jr HOR V (Q /1,,lr,.AIS )1:

D0CKF 7 fN(. & c: c ' C ,.

• Q <F r.

()F T ~

,fJs nark C I

/,.

it,

UNITED STATES DEPARTMENT OF COMMERCE National Bureau of Standards Gaithersburg, Maryland 20899 00( KEf Er*

U'."" NRC October 1 , 1986

.86 OCT -7 All :39 Secretary ,

u. s. Nuclear Regulatory Commission Washington, DC 20555 OFFICE Of SL1.,1,t. iAhY DOC ETIHu l SEPVICF:

BRANCH Attention: Correspondence and Records Branch

Subject:

Petition for Rulemaking, Docket No. PRM-50-44 Gentlemen:

At their meeting, September 25-26, 1986 , the directors and managers of Test, Research and Training Reactors (TRTR) a sked me to request, in their behalf ,

a 90-day extension in the comment period of the above subject petition for

~ulemaking . Many TRTR members were not aware of the petition until recently.

Additional time is needed in order to carefully evaluate the issues raised in the petition and to make appropriate conunents. The extension will afford everyone concerned this opportunity.

Sincerely ,

Tawfik M. Raby Chairman , TRTR

\

s. NUCLEAR REGI JI~ T(' MI S IOS DOCKET! G s:i ON OFFICE C OF° ri.i Pos tM rL C .

Add I

, SpecH

DOCKET NUMBER PETITION . RULE PRM S/-44 (f) "/to (51 F£ .J/14!)

Washington 00tKETEI:

USNRC State University Nuclear Radiation Center, Pullman, Washington 99164-1300 / 509-335-8641 *86 SEP 22 P3 :45 September 19, 1986 Off lU. St 1.,ht. At' Y OOCKET INli & Sf RV lcr:

BRANCH Secretary U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Re: Docket PRM-50-44 Gentlemen:

I strongly oppose the proposed amendments to 10 CFR Pact 50 made by the Committee to bridge the gap in their petition of July 7, 1986. The recent conference held in Vienna, Austria on August 25 to 29 on the Chernobyl accident clearly indicates that the cause of the accident was due to "a prompt critical reactivity excursion and a steam explosion" not by a "graphite fire " (see attached article from the September 11, 1986 issue of Nuclear News). The main contributing factors were human error and the failure to follow prudent safety precaut ion s and written operation procedures, not the presence of graphite i n the reactor .

The proposed amendment would serve no useful purpose in decreasing the likelihood or mitigating the effects of a Chernobyl type accident. The Chernobyl accident in effect substantiates the NRC's position that a graphite fire caused by the Wigner effect in a small research reactor is a "non-credible" event. The imposition of unnecessary regulations and requirements upon research reactors will really decrease overall safety rather than increase safety. A diversion of effort on the part of the staff of a research reactor from managing the day to day operations of the facility and bona fide safety considerations to "non-credible" events lessens the attention given to "credible" events and increases the likelihood of human error precipitated events .

Notwithstanding, there are lessons to be learned from the Chernobyl acci-dent, and changes i n the NRC regulations in certain areas may well be advisable. The proposal by the Committee to bridge the gap, however, was obviously made prematurely before all the facts were revealed and conse-quently did not address the real problem but only a perceived problem.

Sincerely, C)V. r:91/~

W. E. Wi 1son Associate Director WEW:mb Encs.

ACkftO.Vtedged by eard .. SEP 261981

• , , * ,,-,-.,

• niiil

LJ. S. NUCLFA, r 11

• '"0, '.Ml SSION DOCKET- ' - . .. SECTION C l'r. j , DY N

Postm rk Copies Add 'I C iSpecial C

nuclear news Special Report-Sept. 11 , 1986 Helicopter inspecting the damaged plant (photo: Sovtet Life)

Chernobyl: The Soviet- report From August 25 to 29, the lntunational has been used in a number of countries to ertial energy of the turbme rotor. The Atomic Energy Agency held a unique provide power to feedwater pumps and latest test was intended to see if this conferen~ in iJs headquarters city of Vien- emergency core-cooling systems could be overcome with a new voltage na, Austria. The muting was devoted en- (ECCSs), and to relieve some of the regulation system.

tirely to the April accident at the Cher- wear and tear that a very rapid startup The human errors started here. Al-nobyl-4 power reactor in the Soviet imposes on diesels. though there had been plenty of discus-Union, and featured the presentation by It came as a surprise to some Western sion of the justification of the experimen-Soviet officials of a large volume of infor- experts that the Soviets seemed also to tal program in genernl, the specific test mation-surprisingly large, in the view of be trying to supply the main circulation program is said by the official Soviet re-some attendees. NN European Editor pumps from the rundown of the tur- port-released a few days before the A Simon Rippon was present for the confer- bogenerator. The flywheels on these start of the conference (NN, Sept 1986, W ence, and he covered both the Soviet reve- pumps are already designed to provide a p. 23)-to have been improperly pre-lations and the analyses of them by out- longer coastdown until natural circula- pared: ,

side observers. tion can be established. The initial mis- "The quality of the program was poor understanding may have come from a and the section on the safety measures Why the experiment? mention of the connection of four of the was drafted in a purely formal way. (The The Cbernobyl-4 accident took place main circulation pumps to the test safety section said merely that all switch-because of a variety of poorly conceived generator and the other four to the grid. ing operations carried out during the ex-actions and procedures related to an ex- This connection, it seems, was used to periments were to have the permission of periment. This was one of many candid maintain the reactor at power with cool- the plant shift foreman, that in the event assertions made in a remarkable five- ing from four pumps during the tur- of an emergency the staff were to act in hour presentation on August 26 by Val- bogenerator rundown, thus allowing the accordance with plant instructions and ery Aleksevich Legasov, head of the test to be repeated quickly, if necessary, that before the experiments were started Soviet delegation to the conference. The . through the opening up and shutting off the officer in charge-an electrical en-Soviet disclosures may not have included of the steam valves once again. The four gineer, incidentally, who was not a spe-enough technical detail or procedural pumps connected to the test tur- cialist in reactor plants--would advise justification to satisfy every conference bogenerator would have been discon- the security officer on duty accord-attendee, but by Soviet standards the* nected from that supply in the normal ingly.)"

Legasov address was unusually informa- way when the steam valves were first The program made essentially no pro-tive and self-critical. shut off, leaving only the feedwater vision for additional safety measures, The experiment was intended to dem- pumps drawing energy from the running- though it called for the deactivation of onstrate that, in the event of tur- down turbine. the ECCS, so that it would not trip in as bogenerator disconnection along with 'the meeting was also told that there the circulation pumps ran down The loss of offsite power, the inertia of the had, m fact, been earlier tests of this kind procedures also placed extra demands on turbine rotor could contribute to aux- at Chernobyl-4, in 1982 and 1984. In the auxiliary power supplies.

iliary electricity supplies during those these tests the regulation of the field coils In reply to press questionmg in Vien-vital seconds before the startup of of the generator had allowed the voltage na, a representative of the Soviet dele-standby diesel generators. This technique to fall off much more rapidly than the in- gation said that it was personnel of a Copyright ©1986 by Amencan Nuclear Society

Chernobyl special report systems that had been connected to this Thus, the operators at Chemobyl-4 de-turbogenerator, including four of the cjded to press on, and at 1:03 and 1:07 main circulation pumps and two feedwa- a.m., they started the sixth and seventh ter pumps, were switched to the grid bus- main Circulation pumps in immediate bars of the turbogenerator that was still preparation for the tests. Since the reac-on line. tor power, and consequently the hy-At 2 p.m., the ECCS was isolated to draulic resistance of the core and the recir-I J prevent it from kicking in automatically. culation circuit, were substantially lower The start of the test, however, was then than planned, the full eight pumps pro-postponed at the request of the local duced a massive coolant flow through the electricity dispatcher. As a result, the reactor, 56 000 to 58 000 m3/hr. At some plant was maintaine(i in the unauthorized individual pumps, the flow was up to state with no ECCS for the next nine 8000 m '/hr, compared with a normal

'hours, although this particular violation operating level of 7000 m3/hr. This was did not in actuality play any important another violation, because of the danger part in what followed. Still, the delay that pump breakdown and vibration may have aggravated operator impa- could be caused by cavitation at the tience over the test, and contributed to pumps. But the most serious conse-the "mindset" that led plant personnel to quence of the increased flow was the cre-ignore procedures and block safety sys- ation of coolant conditions very close to tems in their effort to get the plant to the saturation, with the possibility that a proper power level for the test. small temperature increase could cause At 11: lO p.m., the load demand was extensive flashing to steam. The steam lifted, and preparation for the test re- pressure and the water level in the steam sumed with power reduced to the re- separation drums had also dropped quired level, 700-1000 MWt. The au- below emergency levels--but, as part of tomatic control system that operates on the continuing attempt to keep the reac-

>---< 10 km groups of control rods in 12 zones of the tor running long enough for the test to be

,___ __, JO km core, to stabihze power density distribu- started, the operators also blocked the tion, was switched off, in keepmg with a resulting signals of the low levels to the low-power operation requirement. At emergency protection system.

higher power levels, these zonal rods also At 1:19 a.m., the feedwater supply was Map from Soviet rep_ort to IAEA regulate the average power automati- increased-to as much as four times its cally. When the local controllers are initial value-in an attempt to restore the switched off, automatic controllers work- water level in the steam separation "commercial electro-technical" organiza- ing on a signal of the average power of drums. This reduced . both the reactor tion, Domtechenergo, that had asked for the whole core come into play, but it ap- coolant inlet temperature and fuel chan-the tests on Chemobyl-4. Dom- pears that the operators did not syn- nel steam production, with consequent techenergo had, presumably, been re- chronize this automatic system quickly negative reactivity effects. Within 30 sec-sponsible for the development of the new enough to the required power setpoint. onds the automatic control rods had fully voltage regulation system that was being There was an overshoot in the power re- withdrawn in response to the negative tested in the expenment. duction, and the level fell below 30 MWt. reactivity, and the operators attempted By 1 a.m. on April 26, the operators to withdraw the manual rods as well. But The sequence of events were able to stabilize the power back at the operators again overcompensated, The detailed sequence of events lead- 200 MWt, but this was as high as they and the automatic rods began to move ing up to the accident at Chemobyl-4 was could get it due to the xenon poison build- back in.

presented by Legasov on the first after- up that had started during the excursion noon of the aCCident review meeting in to lower power and was still continuing.

Vienna. He followed the written descrip- To drag the reactor up to 200 MWt, the tion fairly closely, but added one or two operators had pulled far too many of the significant asides and comments. He manual control rods out of the reactor, noted, for example, that there would and the neutron flux distribution in the have been pressure on the operators to core was such that the reactivity worth of complete the tests as they shut down on those rods that would be effective m the this occasion, because the next planned first few centimetres of travel back into maintenance period would be more than the core was limited to the equivalent of a year away. He also ~aid that, in six to eight fully mserted rods.

hindsight, 1t con be seen that technical According to the rules, thb operating mean~ could easily have been used to margin of reactivity should not be al-prevent the operators from overriding lowed to go below 30 rod equivalents safety protection system~ and otherwise without special authorization from the violating procedure. Failure to provide chief engineer of the power station.

adequate protection for ~uch human Lega~ov said that if the margm ever falls error represented "a tremendous below 15 rod equivalents, "nobody in the psychological m1~take'" on the part of the whole world-not even the Prime Mims-designers of the RBMK reactor. ter--can authorize continued operation The run up to the accident started at I of the reactor." But the operators were a.m. on Apnl 25, with the reduction of so intent on getting the reactor up to an reJctor power over the next live minutes acceptable power level* for the test-trom 100 percent (3200 MWt) to halt that another attitude attributed to the much. Then the unwJnted tur- mindset-that they ignored the touchy Soviet delegation leader Legasov bogenerator wa~ shut down. The plant ~late of the reactor. (photo: AP1W1de World Photos) 2 NUCLEAR NEWS I SEPTEMBER 11, 1986

At 1:22 am .. the reactor parameters BOILING WATER were approximately stable, and the deci-sion was made to start the actual turbine PRESSURE TUBE test. But in case they wanted to repe11t GRAPHITE MODERATED the test agam quickly, the operators blocked the emergency protection signals REACTOR from the turbine stop valve, which they were about to close, so that it would not trip the reactor. Also, just before they shut off 'the steam to the turbine, they sharply reduced the feedwater flow back to the inillal level required for the test  :,.

conditions. This boosted the coolant inlet temperature, creating a transient situa-tion that could not be addressed because safety systems were cut off.

At 1:22:30 a.m., the operators obtained a printout from the fast reactivity evalua-tion program, giving them the position of all the rods and showing tliat the operat-FEED PUMPS ing reactivity margin had fallen to a level that required immediate shutdown of the REClRCUI.ATK)N PUMPS reactor. But they delayed long enough to start the test. There was clearly a failure GRAPHITE to appreciate the basic reactor physics of Schematic diagram of the RBMK-1000, a heterogeneous water-graphite channel-type aie system, which had rendered the con- reactor (s~rce: Soviet report to IAEA)

W-ol rods relatively worthless. The neu-tron flux distribution in the core had with an increase in reactivity and power, had not reached their lower stops. He been pulled into such a distorted shape and further increases in temperature and therefore deactivated the rods to let them that the maJority of the rods would have steam production-producing a runaway fall by gravity.

to go well into the core before they condition. At about 1:24 a.m., observers outside would encounter sufficient *neutron flux At 1:23:40 a.m., the scram button- the plant reported two explosions, one for their absorption to be effective. which would drive all control rods into after the other; burning lumps of mate-At 1:23:04 a.m , the turbine stop valve the core-was pushed. Legasov told the rial and sparks shot into the air above the was closed. With the isolation of the tur- Vienna meeting that there seemed to be reactor and some fell onto the roof of the bine, four of the primary circulation some ambiguity about the motivation for turbine hall and started a fire.

pumps started to run down-another this action, as unearthed during sub- In his presentation of Table I, which transient situation for which the automat- sequent questioning by investigators of delineates the operator violations, at the ic responses had been cut off. the fatally ill shift foreman, who had Vienna meeting. Legasov said that if any Shortly after the beginning of the test, given the order-he may have been be- one of the first five violations had not the reactor power began to rise sharply. latedly responding to the printout of been committed, the accident would not The bulk of the coolant was very close to reactivity margin; he could have been re- have happened.

the saturation point at which it would sponding to the sharp rise in reactor flash to steam, because the operators had power; or he may simply have beheved earlier run an excessive level of coolant that the test had now run long enough to Inside the reactor Aow with all eight pumps on during low- allow him to shut down the reactor. The mechanism of the accident, par-

~wer reactor operation. The RBMK After a few seconds a number of ticularly in the last few seconds before reactor, with its positive void coefficient, shocks were felt in the control room, and the explosion that literally blew the top responds to any such formation of steam the operator saw that the control rods off the reactor. was the subject of intense interest for one of the working groups at the meeting. By the end of the week, the consensus of international experts was TABLE I that the accident mechanism as described THE MOST DANGEROUS VIOLATIONS OF OPERATING PROCEDURES m the Soviet report-a prompt critical AT CHERNOBYL-4" reactivity excursion and a steam explo-VJO!auoo Motivation Consequence sion-was a wholly plausible explanation Reducing operational Attempt to overcome Emergency protection for what happened. There 1s still a need reactiVJty margin below xenon poisoning system was ineffective for more detailed understanding of the penmssible limit mechanism, and some doubts linger on 2 Power level below that Error m switching Reactor difficult to control the cause of a second explosion that was specified in test program off local auto-control reported to have taken place three or 3 All circulating pumps on with Meeting test requirements Coolant temperature close four seconds after the first.

some exceeding authorized to saturation The* Soviet analysis is based mainly on discharge computer modeling of the reactor condi-4 Blocking shutdown signal To be able to repeat tests Loss of automatic tions starting from 1:19 a.m., some four from both turbogenerators 'if necessary shutdown possibility mmutes before the accident (see chart, 5 Blocking water level and To perform test despite Protection system based on next page). This was the point at which steam pressure trips from unstable reactor heat parameters lost the operators started to introduce a sig-drum-separator nificant perturbation on the reactor sys-6 Switclung off emergency core To av01d spurious Loss of possibility to tem by increasing the feedwater flow to cooling system triggenng of ECCS reduce scale of accident restore the water level in the steam

• From lhe Soviet Union summary of its report to lhe IAEA. separator drums. The data-logging sys-NUCLEAR NEWS / SEPTEMBER 11, 1986 3

Modeling of the Chernobyl accident

<il © (11) (~

. \.!. c,:: (,i Q! Ci\ JI I I

Our ha.tl f 1,un:ul

,~,.,

- *- I""' -, ~

' I ,, * © (i I

I \

• m (i tu \ \,Y

(!) Ci I \t C

,r

\ V

- ,~

IH

('1 J

H I

\

\@ ':;/

1 Y\

~" ([) I" ~ ~

le\

r I

I (.II I c L

(L) c) 1~ ,

• 17',,.

,_ I ,. I I I \,/'AG IV I r c ~ - I '\. I I ~

I I I In I I 1"ti I

"" I I

I IIY,

....... t--L '

I  ; H

' I

~

I

• p "~ I I ~ I I ** I I I I (H) I I I ____.I I 119 00 1.19.30 1 2000 1 2030 1-21.00 1 21 30 1 2200 1 2230 Notes: Vertical lines on chart represent 1D-second intervals from one-second intervals. At 1 :23:43, the neutron power curve switches 1 *19:00 to 123:30. At that point, the vertical Imes begin representing from A to D, wrth a change In the vertical scale (see legend).

Parameter Scale (Mm.) Scale (Max ) Parameter Scale (Mm ) Scale (Max )

A Neutron power. low range(%) 0 120 K Main circulation flow (m'/h) 2 8 8 React1V1ty, ~urri (%) -1 +5 L Feedwater flow (kg/s) 0 600 C Stearn drum pressure (bar) 54 90 M Stearn flow (kg/s) 0 600 D Neutron power. high range(%) 0 480 N Fuel temperature (0 C) 200 2000 E Auto-rod. group I (fraction iii) 0 1.2 0 Mass steam quality (%) 0 6 G Auto-rod, group 2 (fraction m) 0 1.2 P Volumetric steam quahty (void fraction) 0 1.2 H Auto-rod, group 3 (fraction in) 0 1.2 S Steam drum water level (mm) -1200 0 tern had also recorded the position of all tual turbine rundown test, the steam of feed water flow, would have allowed three sets of automatic control rods at quality in the fuel channels increased increased steam production in the fuel this tJme, providing a good reference again, and the automatic rods started to channels (curve P) despite competition point for the modeled curves. reinsert (curves E and H) and managed from increasing steam pressure in the Actual measurements from the data- to compensate for the resulting reactivity drums (curve K). The condition of the logging system, and information gleaned transient. reactor was.such that a small increase in from the questioning of operators, are in- A detailed printout of power density power increased the volumetric steam dicated on the chart (with corresponding distribution and rod positions at 1:22:30 quality much more than,it would at nor-letters in circles) with the curves obtained has provided a picture of the neutronic mal power, and resulted* in a large posi-from the computer model, and they all state of the reactor core at this point in tive reactivity insertion.

seem to tie in fairly well. Unfortunately, time. It indicates that in the radial- After 1:23:31, the volumetric steam there were relatively few reactor mea- azimuthal direction, the neutron flux for quality (curve P), reactivity (curve B),

surements from the data-logger because all practical purposes showed a smooth and neutron power (curve A) all began much of its capacity had been switched to convex shape, but that in the vertical di- to increase. At 1:23:40, the scram button record information relevant to the *tur- rection, the curves showed a -double was pressed, but the automatic-rods were bine rundown test. hump, with a greater release of energy in already inserted, and the reactor power As the feedwater flow was increased the upper part of the core. This neutron was on the brink of taking off. (On the (curve L, 1:19 to 1:22), the water level in distribution is consistent with a burned- chart, the neutron power curve switches the steam separator drums was restored out core, practically all rods withdrawn, from A to D at 1:23;43, with a change in (curve S), and the steam pressure de- volumetric steam quality in the upper the vertical scale.)

creased. ( curve C). As the colder water part of the core much greater than lower The prompt critical excursion took the from the drums reached the reactor core, down, and greater xenon poisoning in the power first to around 530 MWt at the steam generation in the fuel channels central region than in the periphery. The 1:23:40, and only the Doppler effect of probably decreased, and the steam qual- reactor would have been in an unusual the fuel heating up to an estimated 3000 ity went down (curves O and P). Re- and impermissible state, with the excess °C pulled it back down briefly. The con-sponding to the negative reactivity that reactivity worth equivalent to only six to tinuing reduction of water flow through this would have introduced, the automa- eight rods. the fuel channels during the power excur-tic control rod~ withdrew (curves E, G, But at 1:23, the reactor parameters sion led to intensive steam production, and H move down, indicating less ab- would have appeared to be closer to the destruction of the fuel, a rapid surge sorber in the core). lt is believed that the stable than they had been for some time. of coolant boiling (with the particles of operators, in their attempt to maintain At 1:23:04, the turbme stop valve was destroyed fuel entering the boiling the power at 200 MWt, attempted to closed for the start of the test. A reduc- water), a rapid and destructive increase "help the automatic rods with manual tion in total coolant flow occurred (curve of pressure in the fuel channels, and fi-rods (dotted curve. 1:19:30) and further K, 1:23:12) as the four main circulation nally the explosion that destroyed the reduced the react1v1ty margm. pumps that had been connected to the reactor.

As the feedwater flow was cut back at test turbogenerator started to run down. A second power excursion at I 23:45, l *22, a minute before the start of the ac- This, together with the earlier reduction to more than 1000 MWt. is represented 4 NUCLEAR NEWS I SEPTEMBER 11, 1986

Chernobyl special report Modeling of the Chernobyl accident

, 0 l'i'I-'"""'

~ \! ~

., .. ~

., l I -I I I I

I I I I I I 11 I <P

~

I I I t*J I

fl I C I I

--_;V--~

I

_,- I  ! I I I ,'I I *_;..t::v- ' / (1)

'I

---

I .

I I

= -- ~ - ~

F

- I  !

i H I I

I I

l T

I I t I I

~

I* I I D v~

I

\

<D I

\ I - I I  ! I I I ~ I ,__. 't ,., \

\.... / I I i  ;  ! )--,-" I \ \

H I I  :

I I~ I I C I I ,) \1 ...../' ' .....

I ~'I-. --- \

C

, .,,-r I -

...I I I I

'Jrn rr.

I I I

\£)

.C

~

I I I I I

I I

I I "I->-::::;..-

II ~ \ I

\

• I ~' I I e I/

- I '

I I  :. --1 ~ 0 I I I I I I I I

• A II ~ \ H

©' \

I ~"I  : I I I I I I  ! I N M I J..,Lo I VI j '-"

I 1 2300 1 2330 12335 1 2340 Scram in the computer modeling by redistribu- cause nobody could imagine that a pilot because more rods are available to be tion of the disintegrating fuel in the boil- would be stupid enough to try. He. dropped in-one must consider the role

. ~ g water and graphite moderator. suggested that the Soviet Union had rec- of lost neutrons in the chain reaction bal-W At precisely the moment of fuel dis- ognized somewhat later than other coun- ance equation. The description of the ruption, which was simulated in the tries the need to protect against this kind RBMK reactor as one with a high neu-model when the power density in the fuel of human fallibility. tron efficiency means that the losses of exceeded 1260 J/g, there was an abrupt On the specific question of the most neutrons by leakage from the very large fall of the coolant flow (curve K) as serious violation-the operation of the core, and by absorption in the various check valves on the main circulation reactor far below the authorized limit.for materials within the core, are relatively pumps closed in response to the in- reactivity margin-Legasov said that an small. Under these circumstances, the creased pressure in the core This loss of automatic system to prevent this had light-water coolant becomes one of the flow was also recorded by the data-log- been considered at the early design stage. more significant absorbers in the core, ging system. The flow from the pumps But at that time such a system, which and any reduction caused by boiling will would have been partially restored after would rely on a fairly complex calcula- have a significant positive effect on the the rupture of the fuel channels, but the tion of the power distribution in the core neutron multiplication. If, on the other water was now directed into a mass of and the reactivity worth of all the reactor hand, a great many control rods are still

• damaged zirconium and hot graphite. control rods, was not considered to be partially inserted, the removal of some The ensuing reaction would have pro- sufficiently reliable to incorporate as an water becomes less significant.

duced large amounts of hydrogen and automatic shutdown system. Among the immediate measures being carbon monoxide, which-upon contact Since the triggering event of the Cher- taken on other RBMK reactors is a lock with air above the reactor--could have nobyl accident was a prompt critical reac- on the rod drive mechanism th&t ensures caused the second explosion . tivity excursion, a great deal of the criti- at least 1.2 m of insertion into the core.

cism of the design centers on the subject Also, the authorized minimum operating

• BMK modifications of reactor physics-although if one lis- margm of reactiVJty has been increased One of the most sensitive issues for tens to many of the experts trying to ex- from 30 rod equivalents to 80. This both Soviet and W estem experts was the plain the situation, one can sympathize means that in their first second of inser-extent to which design features had con- somewhat with the operators accused of tion, the available rods must have a reac-tributed to the accident. The Soviets, having an inadequate understanding of tivity effect-sometimes referred to as while stressing the overabundance of their reactor. The positive void coeffi- reactivity bite-equivalent to the full in-human errors, were relatively frank cient of the light-water coolant in the fuel sertion of 80 rods.

about the few identifiable design weak- channels is clearly the most significant Modifications proposed for the slightly nesses-if only to assert that forthcoming characteristic of this reactor, though in it- longer term include the installation of modifications are sufficient to allow con- self this does not make the reactor im- more control rods and the provision of a tinued operation of other RBMKs. Ex- possible to control as long as there is an diverse rapid shutdown system that perts from other countries sought to sub- adequate number of effective control would use some form of fluid injection.

stantiate the claims that many of them rods. The large size of the reactor, with The absence of a diverse shutdown sys-had made prior to the meeting, to the ef- low-enrichment fuel and highly efficient tem is the RBMK aspect that was fect that such an accident could not hap- graphite moderator, also tends to lead to perhaps the most criticized by specialists pen in their reactors because of funda- a system that is subject to local power in- , from other countries, particularly those mental design differences from the stabilities. These can be controlled, sub- from Canada, who pointed out that they RBMK. ' ject to a system of automatic regulation learned the lessons about the need for One of the design flaws specifically re- coupled to good instrumentation. But at such a system after the criticality accident ferred to by Legasov in his opening re- low power, the instabilities are more ap- at the NRX research reactor at Chalk marks was the lack of automatic systems parent, and the instrumentation is less ef- River back in the 1950s.

to prevent the operators' violations. He fective. Another change designed to help over-compared the situation to that of an air- To understand why the withdrawal of come the positive void coefficient is the craft designer considering it unnecessary too many rods can be a dangerous situa- introduction of fuel with an enrichment to provide automatic locks to stop a pilot tion in the RBMK--even though the situ- of 2.5 percent instead of 2 percent. It was from testing the doors during flight, be- ation appears to be a relatively safe one stated that this change-over will begin NUCLEAR NEWS / SEPTEMBER 11, 1986 5

Chernobyl special report next year, but will take some time to be Western specialists was that there was no bris above the reactor. There has obvi-fully effective, smce fuel is changed on- possible mechanism in their reactors for a ously been some editing of these pic-load over a period of years. Higher en- prompt critical reaction that could pro- tures, since the IAEA visitors to Cher-richment fuel has been developed for the duce fuel-coolant interaction similar to nobyl in May were apparently shown larger design of RBMK, which gets an that which appears to have taken place at much longer shots of the glowing core.

output of 1500 MWe fr.om. a reactor of Chernobyl-4. The only scenarios for pos- But the remaining video, and Soviet de-the same size as the 1000-MWe units at sible - steam explosions involve core scnptions of the damage, were enough Chernobyl. The apparent contradiction meltdown and melt-through, with a sig- for the working group on accident dam-of improving the situation by putting nificant time delay and thus much less se- age to determine that the whole of the more fissile material into the core is, ac- vere release consequences if a steam ex- top plate of the reactor had been lifted cording to one knowledgeable reactor plosion were to occur. But on the question off by the explosion and deposited at an physicist, also related to a greater pro- of whether an explosion of comparable angle to one side of the reactor. In the portion of non-water atoms capturing neu- energy to that at Chernobyl-4 would process, *au of the fuel and control rod trons in the critical balance equation. breach their containments, the general channels-roughly 2000 in total-had Another RBMK feature that has come view was that it might cause cracking and been ripped open.

m for criticism from abroad is the high some openings, but would not com- The working group agreed that the temperature of the graphite moderator pletely destroy the structure, which power excursion and steam explosion during normal operation-but it is not would still have some effectiveness in re- could nave produced the necessary yet certain that this contributed sigmfi- ducing radioactive releases. energy. The energy release calculated by cantly to the severity of the Chernobyl ac- Inevitably, there was much talk in Vien- the French delegation was on the order cident. At a temperature of 700-750 °C, na of the need to "improve the man- of 200 MJ, generating a pressure of some the graphite represents a significant machine interface." The Soviet special- tens of atmospheres under the top plate.

source of heat in an RBMK reactor, ists seemed to acknowledge that they Rough calculations also indicated that it compared to other graphite-moderated have lagged behind the West in thIS area. would only have taken about two atmo-and heavy-water reactors, where the They also accepted the need for im- spheres to lift the plate.

moderator acts as a large heat sink. Dur- proved training and retraining of For some observers, the severing of ing the low-power operation of the Cher- operators, with greater use of simulators. the fuel channels pomted up a weakness nobyl-4 reactor just prior to the accident, But, scoring a rather perverse point, they in the RBMK design. With the graphite it is also likely that the nitrogen-helium noted that the excellent routine perfor- hotter than 700 °C during normal opera-gas mixture, which is used for partial mance of thetr plants to date was one of tion, and dependent on the coolant chan-cooling of the graphite, would have been the reasons why the operators at Cher- nels for heat removal, the Zircaloy pres-changed to pure nitrogen, which has nobyl were Ill-equipped to deal with an sure tubes could easily be subjected to a poorer heat remov 9 1 properties. This abnormal condition. The fact that the temperature at which they would rupture would have placed additional reliance on Chemobyl-4 unit had been the top- readily, especially in the region of the the coolant m the fuel channels to re- ranked reactor in the performance fig- transition joint to stainless steel just move heat from the graphite, and may ures of Soviet plants was also cited as a above the reactor.

have weakened the transition joints be- possible cause of complacency on the The fire on the roof of the turbine hall tween the Zirconium alloy and stainless part of the operators. was the most immediate cause for con-steel at the tops of the fuel channels, in cern for firefighters. The hot lubricating the area where they were ruptured by the Accident consequences oil in the turbines and the hydrogen cool-initial steam explosion. The damage to Chernobyl-4 was ant for the generators were vulnerable, On the vexing question of contain- shown in a screening of the video pic- givmg nse to fear that the fire could ment, the Soviets asserted that much of tures taken mainly from helicopters on spread to the adjacent Chemobyl-3 unit the plant, in the strong-box compart- the second day after the explosions. and even to Units 1 and 2, which share ments that formed the containment for These included two brief glimpses of the the same long turbine hall. The fires the design basis loss-of-coolant accident, red glow of the core seen through the de- above the reactor were dealt with mainly appeared (in the available video pictures) to be still intact. With a design pressure of 4.5 atmospheres (0.45 MPa), these compartments are capable of providing a high degree of protection of the primary circuit components and pipe work. To prevent radionuclide escape into the area above the reactor, the design relies on the huge volume of the building that houses the fueling machine and spent-fuel storage pool to provide pressure re-duction and containment. It goes without saying that the steam explosion m the Chemobyl-4 reactor was beyond the de-sign basis accident. The Soviet specialists maintain. however, that there was, and still is, no practical possibility of provid-ing an, all-embracing pressure contarn-ment building of the light-water reactor type over the top of this very large reac-tor. Instead, the Soviets try to ensure that Jn accident beyond the design basis cannot occur.

Asked at press briefings 1f an LWR containment could withstand a similar steam cxplo~ion. the first response of Spraying water on streets 1n towns and villages near Chernobyl (photo* Sov,et Life) 6 NUCLEAR NEWS I SEPTEMBER 11, 1986

with fire c\tinguishcn- ,ind in!.tallcd fire planned. Thi~ wa!, the rca!.On given by hydrant!.. All of thc!,C fire~ were extin- Lcgasov for the apparent delay in evacu-guished by 5 a.m. on the morning of ation, which had drawn some criticism April 26. Only then wa!. the adjoining from outside obo;crvers. Once safe evacu-Chcmobyl-3 reactor shut down. This rcv- al!on routes were established, said clat1on came a!, a surpri!.c to some dele- Lega!,()V, the evacuation was carried out gates; so did the Soviets' statement that with what was termed remarkable effi-the other two units were not shut down ciency, in 2.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> with a fleet of 1000 until the following day. The Soviets said buses that thi!, wa!. un indication of how the The pattern of radioactive release (see damage had been confined to the one Table II), calculated on May 6 with al-unit. lowance for decay of short-lived mate-The video of the damaged reactor in- rial, obviously started very high on the cluded !,()me shots of components in the first day (amounting to 12 MCi), then fell equipment vaults around the reactor, and to 2 MCi on days four and five, and then indicated that, at the lower levels, much rose rather alarmingly to 7 and 8 MCi on of this equipment had survJVed almost in- days eight and nine before falling off tact. On one side of the reactor, the cells sharply. The increase on days eight and containing four of the main circulat10n nine was attributed to a rise of tempera-pumps were intact; on the other side, ture in the core as various materials were away from the added partial support of dropped on top of the damaged reiictor the turbine building wall, the other four to seal it off. The release fell off again as circulation pumps were visible standing some nitrogen gas cooling of the core was out of the rubble. established and as the sealing became ef-It is estimated that about 3 5 percent fective.

A of the fuel material was ejected from the W core, and that some 10 percent of the Cover-up operation graphite was ejected or ignited. Much of Reports of helicopters dropping a vari-the fuel (0:3-0.5 percent) was deposited ety of materials onto the burning reactor as heavy particulate matter, some tens of sounded like a fairly desperate effort to microns in size, around the site. A cover it with anything at hand. The further 1.5-2 percent was distributed Soviets maintained in Vienna, however,"

over a 20-km zone, while 1-1.5 percent that it was a rather more carefully was d1stnbuted

• as small particulates, thought-out operation. Spraying buildings with decontammant down to micron size, over the rest of the Immediately after the aCCJdent, at- (photo: Soviet Life) 30-km evacuation zone. tempts were made to get some cooling The initial large release from the reac- water into the damaged core via the With a loading of something like 5000 tor fortunately missed the nearby town of emergency auxiliary feedwater pumps, tons from the covering on top of the reac-Pripyat, but caused considerable con- but this proved unsuccessful. Considered tor, and with the possibility of continuing tamination of the forested areas through next were covering of the open reactor high temperatures, there was real con-which evacuation routes had to be vault or allowing the fire to burn itself cern for the supporting structures of the out. The former was adopted for the reactor. The Soviets stated that it was fatrly obvious reason of trying to limit this fear, rather than the speculation TABLE II radioactive releases, but it raised the about the danger of core melting, that ESTIMATED RELEASE OF problem of fuel heatup and the remote led to an urgent decision to construct a RADIONUCLIDES FROM THE possibihty of some fuel melting into large slab of concrete below the reactor.

CHERNOBYL ACCIDENT- masses that might go re-critical. This has a heat exchanger on top of it and The first thing dropped on the core is described in the Soviet report as an Released actJY1ty (MC1) Rele.ned (percentage) was some 40 tons of boron carbide, to re- "artificial heat-removal horizon." Be-Nuclide by May6 byMay6 duce the possibility of re-criticality. This cause of the high radiation doses still pre-was followed by 800 tons of dolomite vailing around the reactor, the concrete Xc-133 45 up to 100 Kr-85m up to 100 (limestone), to absorb heat as it decom- was pumped in through tunnels dug to Kr-85 0.9 up to 100 posed and also to release carbon dioxide the basement of the reactor.

I-131 7.3 20 to help extinguish the graphite fire. Next The next stage will be to build some Te-132 1.3 15 there came 2400 tons of lead, also to ab- sort of entombment around the whole of Cs-134 0.5 10 sorb heat as it melted but additionally to the damaged reactor. Work is already Cs-137 1 13 run down through the core, if possible. It under way on the construction of walls, Mo-99 3 2.3 was also hoped that the lead would build particularly between Unit 3 and the dam-Zr-95 38 3.2 up some shielding against gamma radia- aged reactor, and the long-range plan Ru-103 3.2 2.9 2.9 tion, not the least for the benefit of the calls for all of the debris to be roofed Ru-106 1.6 Bli-140 4.3 , 5.6 hehcopter pilots. The covering was com- over. The Soviets have not finished the Ce-141 2.8 2.3 pleted with large quantities of sand and detailed design of the entombment, and Ce-144 2.4 2.8 clay, both to smother the graphite fire seemed eager in Vienna to get views Pu-238. 0.8E-3 3.0 and to filter escaping fission products. from other countries, especially on the Pu-239 0.7E-3 3.0 The covering did indeed heat up the relative mcnts of a natural circulation Pu-240 lE-3 3.0 fuel and increase releases of radioactivity open cycle cooling sc_heme versus some Pu-241 0.14 3.0 until it was possible to establish cooling form of closed cycle cooling.

Pu-242 2E-6 30 by pumping nitrogen gas-from a com-Cm-242 2.lE-2 3.0 pressor station on the site-into the Radiation effects Sr-89 22 4.0 Sr-90 0.22 4.0 space below the reactor vault. By May 6, As of late August, 31 people had died Np-239 1.2 3.2 the temperature was stabilized, and the as a result of the Chemobyl-4 acetdent.

• Estrmated error +/-50% release of rad10activity fell to a low level. All of these were operating personnel, NUCLEAR NEWS / SEPTEMBER 11, 1986 7

Chernobyl special report lm:fightc,~. Jnd c.:mcrgcncy workers who had dcalt with the: immediate conse-quence~ of the accident. Apart from two people killed at once, one from steam hums and one from falling dehris, all the tleaths have occurred among the 203 people ho~p1tahzcd with acute radiation sickness ( others were hospitalized with less severe symptom~) Medical special-1~ts from other countries praised the speed with which expert medical teams reached the site and the efficiency with which the severe cases were selected.

Biodosimetry on the 203 severe cases has revealed that they all received doses m excess of 1 Gy (100 rads), with 35 re-ceiving more than 4 Gy and a few ex-posed to extreme doses of 12-16 Gy. All the deaths thus far have been among those who received more than 4 Gy.

A mass of invaluable information has been provided by the Soviet doctors on the treatment of the victims, which, again, was judged by other experts to have been excellent. There was praise for the good conventional medicine applied to the majority of the victims, as opposed A radiation checkprnnt at the edge of the 30-km evacuation zone (photo* Sovfoto) to the much-pubhcized bone marrow transplants, which were applicable only limit of additional cases as a result of the laminated food, is much more compli-for cases within a small band of radiation accident. cated. The Soviet report has attempted dose and which were largely unsuccess- The Soviets said that a highly efficient to produce highly conservative figures ful. operation employing youth volunteers using maximized assumptions at all Nobody beyond the bounds of the ensured the widespread d1stribut10n and stages to obtain a quick assessment of Chernobyl site is reported to have suf- use of potassium iodide tablets in the whether any special medical provisions fered any symptoms of direct radiation town of Pripyat and some surrounding need to be made for the regions of the sickness. The majority of the 135 000 areas. This has provided the first large- Ukraine and Byelorussia, where some 10 people in the 30-km evacuation zone, in- scale test of this technique for blocking percent of the activity released from cluding the 45 000 from the town of iodine doses to the thyroid. The first re- Chernobyl is estimated to have fallen Pripyat, received external radiation ported indications are that the technique out. For the external radiation from this doses of less than 25 rem from the has proved effective and that there have fallout, the upper limits of the collective radioactive cloud. A few people living in been no undesirable side effects. Mea- doses are put at 8.6 million person-rems villages situated in the most contami- surements indicate that the majority of for 1986 and 29 million person-rems for a nated areas may have received between the people in this area would have re- period of 50 years.

30 and 40 rem. These external doses are ceived a thyroid dose of less than 30 rads. On the still more difficult question of estimated to account for a collective dose In the period after the accident, a large estimating the internal doses from con-of 1.6 million person-rems. Taking ac- number of the population from the sumption of food contaminated with count of the projected spontaneous evacuated zone and beyond, including al- cesium, a figure of 210 million person-cancer deaths for this population over the most 100 000 children, were checked for rems for the next 70 years has been pro-next 70 years-14 000 cases-the Soviet radioiodme in their thyroids. The mea- duced-but discussion in the working report suggests 2 percent as an upper surements were reported to have shown party at Vienna concluded that, in their levels significantly below those that could attempt to produce the most pessimistic cause any health effects. estimate, the Soviets may have overesti-Outside the 30-km evacuation zone, mated by a factor of 10. Some support direct radiation measurements of several for this view came from whole-body mea-times natural background of 0.008-0.012 surements that have already been carried mR/hr were recorded, and in Kiev, levels out on about 1000 people from the re-peaked at 1 mR/hr before falling off gion. Of these, 97 percent showed levels slowly. The averaging of the radiation that were 10 times lower than the expec-measurements for the whole of the popu- tation based on the pessimistic assump-lation of the European part of the SoVIet tion of cesium ingestion. The Soviet re-Union outside the 30-km evacuation. port. stated that on the basis of its zone gives' values of individual doses of maximized figures, the cancer mortality external radiation that do not exceed 1.5 rates in the Ukraine and Byelorussia may rem for 1986, nor 50 rems for the next 50 be increased by no more than than 0.05 years The .Soviet report therefore con- percent as a result of the external radia-cluded that there is no health danger to tion, and less than 0.4 percent as a result this population as a result of the external of the internal radiation.

rad1at10n from the Chernobyl cloud. This report was prepared principally by The question of doses from the fallout European Edilor Simon Rippon, wilh of radioactive material, both external contributions from E. Michael Blake, Jon gamma radiation from the ground and in- Payne, and others on the NUCLEAR ternal doses from consumption of con- NEWS staff 8 NUCLEAR NEWS / SEPTEMBER 11, 1986

00(,;_KETEO Uf?MB90-0l]

NUCLEAR REGULATORY COMMISSION "86 AUG 28 P3 :31 10 CFR PART 50 O-FF1cr (H' ~~ c:-*:,, *-. 'r"

[DOCKET NO. PRM-50-44] OOCKc.T,r,.; :, ~,::*.1;1 i

[,--r J ~. ,:, , *

• COMMITTEE TO BRIDGE THE GAP; PETITION FOR RULEMAKING AGENCY: Nuclear Regulatory CoIT111ission.

ACTION: Receipt of Petition for Rulemakin~.

SUMMARY

The Nuclear Regulatory CoIT111ission requests public coITDTients on this notice of receipt of a petition for rulemaking dated July 7, 1986, that was filed by the Committee to Bridge the Gap (CBG). The petition was docketed by the Commission on July 7, 1986, and assigned Docket No. PRM-50-44. The petition requests that the Commission amend its regulations to require operators of reactors that use graphite as a moderator or reflector to (1) prepare and sub-mit for NRC approval fire response plans and evacuation plans for a graphite fire and (2) measure the energy stored in their graphite, and revise their

- safety analyses to consider the risks and consequences of a graphite fire in their facilities. The petitioner believes this action is necessary to ade-quately protect the public in the event of a fire

• DATE: Submit coIT111ents by . Corranents received after this date will be considered if it is practical to do so, but assurance of consideration cannot be given except as to co11i11ents received on or before this date.

ADDRESSES: Submit corrments to: Secretary, U.S. Nuclear Regulatory Corrmission, Washington, DC 20555, Attention: Correspondence and Records Branch.

For a copy of the petition write: Division of Rules and Records, Office of Administration, 4000 MNBB, U.S. Nuclear Regulatory Commission, Washington, DC 20555.

Inspect and copy the petition or comments received on the petition at:

The- NRC Public Document Room, 1717 H Street, NW., Washington, DC.

FOR FURTHER INFORMATION CONTACT: John Philips, Chief, Rules and Procedures Branch, Division of Rules and Records, Office of Administration, U.S. Nuclear Regulatory Commission, Washington, DC 20555, Telephone: (301) 492-7086 or Toll Free (800) 368-5642.

SUPPLEMENTARY INFORMATION:

I. Basis for Petition The petitioner states that two recent developments indicate that the potential

- for a graphite fire at U.S. reactors has been inadequately addressed.

Credibility of Graphite Fires

\

The peti~ioner states that NRC and reactor licensees have held that graphite fires are "non-credible" events and as a result have failed to take measures to help mitigate or extinguish such fires, should they occur. The petitioner asserts that the occurrence of a graphite fire at the Chernobyl plant in the Soviet Union demonstrates that graphite fires are credible events.

The petitioner asserts that because the NRC has deemed graphite fires non-credible, it has failed to require basic safety measures that could help reduce the threat of such a fire. Petitioner alleges that licensees whose reactors use graphite, including dozens of nonpower research reactors, the Fort St. Vrain plant in Colorado, and the Department of Energy's N reactor, l/ have no fire response plans for combatting graphite fires. The petitioner further alleges that research reactor licensees do not have adequate emergency plans to evacuate members of the public in the event of a graphite fire.

r Stored (Wigner) Energy The petitioner states that new experimental data show that the NRC's generic analysis of energy stored in research reactor graphite significantly under-estimates the actual amount of stored energy and thus, underestimates the associated risk of graphite fire. The petitioner states that in a generic study (NUREG/CR-2079) which discusses the amount of stored energy present in research reactor graphite, that NRC contractors predicted that 5 calories per gram might be stored in the graphite of an Argonaut-type research reactor. The petitioner states that the UCLA research reactor is an Argonaut-type and in contrast to predictive calculations by both NRC and UCLA, UCLA researchers recently reported measurements of stored energy in the reactor graphite as high as 33.2 cal/gram; while CBG, the intervenor in the UCLA reactor relicensing proceeding calculated a minimum stored energy of 113 calories/gram in the year 2000, which corresponds to 39 calories/gram in 1983.

l/ This reactor is not within the licensing and regulatory authority of the NRC.

The petitioner alleges that NRC's generic estimates of Wigner energy storage are inaccurate and stresses that as a remedy to the problem, actual empirical measurements of Wigner energy will be required to assess the magnitude of the energy stored in research reactor graphite and the magnitude of the fire hazard that it presents.

II. Proposed Amendments to 10 CFR Part 50 The petitioner requests that the NRC adopt regulations that would require all licensees* whose reactors employ graphite as a neutron moderator or reflector by January 1, 1987, to:

(a) Formulate and submit for NRC approval fire response plans for combat-ting a reactor fire involving graphite and other constituent reactor parts (e.g.

fuel) which might be involved, in such a fire, taking into consideration the potential for explosive reactions. Response plans shall identify precisely which materials will be used to suppress a fire without increasing the risk of explosion, and shall indicate where and in what quantities these materials will be stored.

(b) Fonnulate and submit for NRC approval evacuation plans for a reactor fire. Plans should include evacuation out to a sufficient distance from the reactor such that no member of the public receives a dose to the thyroid greater than 5 rem, assuming a release to the environment of 25% of the equilibrium radioactive iodine inventory.

(c) Measure the "Wigner energy" stored in the graphite of their reactor.

Revise the reactor safety analysis report to consider how a release of stored energy would affect the outcome of other accident scenarios (e.g., fire, reactivity accidents).

A sufficient number of graphite samples shall be measured to identify the location of maximum stored energy, and to determine the maximum quantity of stored energy to within 10%.

• Zero power or critical facilities 11 11 11 11

, defined here as reactors which operate at 100 watts or less, are exempted from these requirements.

III. Conclusion In conclusion, the petitioner contends that the Chernobyl accident proves that it is a mistake to assume that graphite fires are non-credible, and yet the NRC has based its regulatory approach to nonpower reactors on this assumption.

Petitioner further states that just as the Soviets began after Three Mile Island to recognize the necessity of reactor containment, the NRC should learn from Chernobyl that graphite fires are credible accidents and regulate graphite reactors accordingly. Above all, the petitioner believes that the NRC must require preparation of fire response plans that include the prevention and mitigation of graphite fires and evacuation plans adequate to protect the public in the event of a fire.

New measurements, the petitioner further concludes, indicate that the NRC has underestimated the amount of stored energy in the graphite of nonpower reactors and, consequently, that the potential for a graphite fire in such reactors has also been underestimated. The petitioner argues that since licensee calculations (such as those made by UCLA) are similarly unreliable, the NRC should order

actual empirical measurements of stored en~rgy in all nonpower reactors that use graphite. Finally, the petitioner states that safety analysis reports and hazards analyses should be revised to consider generally the consequences of a release of stored energy and the risks and consequences of reactor fires.

Dated at Washington, DC this -;}::7 ~ day of 1986.

For the Nuclear Regulatory Corrmi~sion.

Secretary of the Co111Tiission.

DOCKET - -R 1'1u IYh...11-I COMMII IEE TO BRIDGE THE GA~E, *. lvl; 11

• u1:.£ PR"'62}-'1LJ 1637 BUTLER AVENUE #203 _rT LOS ANGELES, CALIFORNIA 90025 (213) 478-0829 C

July 7, 1986 Chairman Lando Zech, Jr. DOCKETED cOIIITlissioner James Asselstine USNRC Commissioner Frederick :semthal coomissioner Thomas Roberts u.s. NUclear Regulatory corrmission washington, o.c. 20555 near carmi.ssioners:

The graphite fire at the Chernobyl nuclear power plant raises the question of the potential for a similar fire at u.s. reactors which use graphite, including dozens of nonpower research reactors, the Fort st. vrain plant in Colorado, and the Department of ,Energy"s N reactor. We believe that the possibility of such a fire at a U.S. reactor has been inadequately addressed by the NRC. While graphite fires are notoriously difficult to extinguish, no forethought has been given by NRC or reactor licensees as to how to fight such a fire; none of the facilities which use graphite have fire response plans which even address graphite fires, and the NRC does not require any.

FUrther, new data indicate that the NRC appears to have incorrectly assessed the amotmt of "Wigner energyn stored in research reactor graphite.

In a generic study which discusses this problem (NURffi/CR-2079, p.37), NRC contractors predicted that 5 calories per gram might be stored in-the graphite of an Argonaut-type research reactor. In contrast, UCLA researchers at the recent Annual Meeting of the Amer'ican Nuclear society reported stored energy in reactor graphite as high as 33.2 cal/gram. over its operating lifetime, such a reactor could store sufficient energy to raise the temperature of graphite several hundred degrees. PUt another wa ,

sufficient Wigner energy could be stored to in effect lower the 650 6C' ignition temperature by several hundred degrees. It seems clear that_ NRC has significantly underestimated the potential for a graphite fire.

We have therefore prepared the enclosed petition for rulemaking, which formally requests that NRC issue a rule requiring operators of reactors that use graphite as a moderator or reflector (1) to prepare and submit for NRC approval fire response plans and evacuation plans for a graphite fire,

- and (2) to measure the energy stored in their graphite, and revise their safety analyses to consider the risks and consequences of a graphite fire in their facilities.

Very truly yours,

~(frrvl Steven Aftergood EXecutive Director enclosure cc w/enclosure:

rxx::keting and service Branch Office of the secretary

Committee to Bridge the Gap JUly 8, 19~6 1637 :sutler Avenue, suite 203 Los Angeles, CA 90025 (213)478-0829 DOCKETED USNRC 16 JI. -7 Pl :37 OFFICE OF 5::Rt:TARY DOCKETING t. SERVIC[

r BEFORE THE BRANCli NUCLF.AR REGULA'IORY COMMISSION UNITED STATES OF AMERICA PETITION FOR ROLEMAKIN3

'IO REDUCE FIRE HAZARD FROM NUCLF.AR REAC'IDR GRAPHITE I. Introduction TWO recent developments indicate that the potential for a graphite fire in U.S. nuclear reactors has been inadequately addressed. First, the occurrence of a graphite fire at the Chernobyl plant in the soviet Union demonstrates that such fires are indeed credible events, though they had been dismissed as non-credible by regulatory officials and reactor licensees. second, new experimental data show that the NRc's generic analysis of stored energy in research reactor graphite significantly underestimates the actual amount of stored energy, and thus underestimates the associated risk of graphite fire.

Because it had deemed graphite fires non-credible, the NRC failed to require basic safety measures which could help to reduce the threat of such a fire. Licensees whose reactors use graphite, inclyding dozens of noopower research reactors and one commercial power reactor , have no fire response plans for combatting graphite fires in their reactors. Research reactor licensees do not have adequate emergency plans to evacuate members of the 1 In addition, one unlicensed Dept of Energy reactor, the "N" reactor in Washington state, which produces commercial electric power as well as plutonium, also uses graphite.

public in the event of a graphite fire or other severe accident.

we therefore petition the NRC to issue an amendment to 10 Code of -

Federal Regulations Part 50, as follows:

All licensees whose reactors employ graphite as a neutron rroderator or reflector shall, by January 1, 1987:

(a) Formulate and submit for NRC approval fire response plans for combatting a reactor fire involving graphite and other constituent reactor parts (e.g. fuel) which might be involved in

_such a fire, taking into consideration the potential for explosive reactions. Response plans shall identify precisely which materials will be used to su:r;:press a fire without increas:i,ng the risk of explosioo, and shall indicate where and in what quantities these materials will be stored.

(b) Formulate and submit for NRC approval evacuation plans for a reactor fire. Plans should include evacuation out to a sufficient distance from the reactor such that no member of the public receives a dose to the thyroid greater than 5 rem, assuming a release to the environment of 25% of the equilibrium radioactive iodine inventory.

(c) Perform measurements of the "Wigner energy" stored in the graphite of their reactor, and submit these measurements to NRC-for review together with a revised safety analysis, which shall address the risks and consequences of a reactor fire.

A sufficient number of graphite samples shall be measured to identify the location of maximum stored energy, and to determine the maximum quantity of stored energy to within 10%.

• zero power" or "critical facilities*, defined here as reactors which operate at 100 watts or less, ~re exempted from these requirements.

I I. The NRC and Licensees Mistakenl Dismissed Gra ite Fires AS Non-credib e

'Ihe NRC and reactor licensees have held that graphite fires are *non-credible* events, and as a result they have failed to take measures to help mitigate or extinguish such fires, should they occur. The NRC considers the dropping of a single fuel element to be the *maximum credible accident" that can be 'suffered by most research reactors. 2 One reactor licensee has gone so far as to testify under oath that *graphite is considered a non-combustible material* and that "the so-called 'burning' of graphite is 2 see, e.g., NURE'G/CR-2079, "Analysis of credible Accidents for Argonaut Reactors", and NUREG/CR-2387, "Credible Accident Analyses for TRIGA and TRIGA-FUeled Reactors."

2

actually a surface reaction more analogous to the rusting 3f iron than to burning in the sense of a self-sustaining, propagating fire.*

These erroneous not;ons were first refuted thirty years ago by the Windscale reactor fire in England in 1957. In a discussion of that incident, the worst reactor accident prior to Chernobyl, Dr. c. Rogers Mccullough of the U.S. Atomic Energy commission stated:

By the way, this is an amusing point. 'llle belief had grown up on the part of many people in this country that graphite will not burn. This is nonsense. Graphite is carbon, and anyone knows that carbon will burn if you get it hot enough. But this gl~b remark, that graphite will not catch fire, had becone prevalent.

But the fire hazard associated with graphite arises not only from its combustibility. A graphite fire, once burning, is also exceptionally difficult to extinguish. There could be great danger in using either water or carbon dioxide to put the fire out, since combustible gases may be produced as a result. This is why advance planning to fight a graphite fire is so important.

Dr. Mccullough's report on the Windscale incident describes how those fighting the fire tried various methods over a couple of days to put the fire out, all to no avail, and how they had to try, as a last resort, water:

Now they were faced with the decision either to use water or to let the fire burn up. They decided there was nothing left for them to do but put water in. There was some trepidation about this, as you can_ imagine, because they well knew that water on glowing uranium makes hydrogen. Water on glowing carbon makes hydrogen and CO; you have then a nice mixture of hydrogen, CO and air, ~ you might have an explosion. But they had no other choice. -

As things turned out, an explosion fortunately did not occur. McCullough concludes:

I think it took a great deal of courage on the part of these people to put water on this reactor. They did it with fear and trepidation, and in talking with them they will not guarantee that 3 *Testimony of Dr. Walter F. Wegst concerning the safety of the UCLA Research Reactor, pp. 7-8, in NRC oocket No. 50-142, transcript of the July 0

25, 1983 hearing, after transcript page 2028.

4 For a description of the accident, se T.J. Thompson and J .G. Becker ley, editors, The Technology of Nuclear Reactor safety, MIT Press, vol. 1, pp.

633-636.

5 TID-7569, "Proceedings of the 1958 Atomic Energy Commission and contractor safety and Fire protection conference, 0 June 24-25, 1958, p. 83 6 ibid., p. 76 3

they could do it a second time without an explosion. 7 The situation at Chernobyl seems to have been at least as precarious.

It should be noted that under normal circumstances the Chernobyl reactor core and graphite were maintained in an inert, helium-nitrogen atmosphere, in which graphite combustion is truly nnon-credible,n in fact impossible.

Many people have difficulty understanding that what may ordinarily be impossible can become quite feasible under accident conditions. Thus reactor licensees argue graphite is "non-combustible" and the NRC Staff has argued that there may be insufficient air flow to sustain a graphite fire.

sut at Chernobyl there was no airflow-- under normal conditions- and yet a graphite fire did break out, and the soviet Union was, like the British, was confronted with the formidable difficulties of trying to extinguish it.

According to news reports, the soviets were even forced to app~oach SWedish and West German experts for information on fighting such fires.

Whereas British and soviet scientists have faced graphite fires with "fear and trepidation," the NRC and reactor licensees have been satisfied to declare such fires "non-credible" and therefore unworthy of attention.

Yet graphite is used as a moderator in the Fort st. vrain nuclear power plant in Colorado, and in the dual purpose N reactor operated by the Department of Energy. Graphite is also used as moderator or reflector in the majority of the more than sixty nonpower research reactors throughout the U.S. Because graphite fires have been deemed non-credible, none of these facilities have considered how they would go about trying to extinguish a graphite fire, and none of them have prepared fire response plans for such fires. Graphite fires are also ignored in safety Analyses, Hazards Analyses, Safety Evaluation Reports, Environmental Impact statements, etc. Most of the nonpower reactors have evacuation plans which extend no further than the reactor room itself. Since they usually have no containment and no low-population buffer zone (most of these facilities are based on university camp~es), prompt evacuation after an ac~ident at one of these reactors is crucial-- and currently unlikely to occur.

7 ibid., p. 78 8 "soviet says Fire at Atomic Plant Followed Blast," New York Times, May 6, 1986, p. 1, at p. 4.

9 our proposed requirement for an evacuation plan (see p.2 above) cites a maximum permitted dose to the public of 5 rem. 'Ibis is derived from Table 1 of ANS 15.16, "Standard for Emergency Planning for Research Reactors,"

November 29, 1981, as reaffirmed in Table 1 of NURffi-0849, "Standard Review Plan for the Review and EValuation of Emergency Plans for Research and Test Reactors." 'Ihe estimated 25% release to the environment of radioiodines is derived from the same ANS document, and may be conservative, considering that a reactor fire could propel a significant amount of radioactive materials including particulates into the environment.

4

III. The NRC Bas Miscalculated Wigner Energy in Research Reactor Graphite; Empirical Measurements are Necessary "Wigner energy" is energy stored in nuclear reactor graphite during reactor operation as a result of neutron bombardment. It can be released suddenly as ~aat if the reactor temperature is raised above normal operating tenperature. such a release of energy can potentially heat the graphite beyond its ignition temperature, cited in the literature as approximately 650 degrees Centrigrade. This phenomenon (*the Wigner effect*) was a significant contributor to the Windscale reactor fire.

The question of how much stored energy might be present in research reactor graphite was the subject of vigorous technical dispute in the UCLA reactor relicensing proceeding (which terminated in 1985 after the reactor was permanently shutdown). That dispute and the measurements described below reveal the unreliability of calculated estimates of Wigner energy by the NRC Staff and licensees, and indicate the need for empirical measurements.

The UCLA reactor was a 100 kilowatt Argonaut-type research reactor which was reflected and partially moderated by graphite. By 1983 the, reactor had operated for the equivalent of 194 full power days. Had its request for relicensing been approved, it could have operated for a total of 560 full power days by the year 2000. Th ~mal neutron fl~ at f~ll power had been measured variously as 1.5 x 10 1 and 3.0 x 10 n/cm -second.

Graphite tel]lPeratures during irradiation were generally quite low, on the order of 5o 0 c. (Wigner energy storage is much higher at low temperatures than at high, where the graphite self-anneals.)

The committee to Bridge the Gap (CBG), the Intervenor in the UCLA reactor relicensing proceeding, had contended that under these circumstances (substantial integrated flux at low temperatures), a significant amount of Wigner energy could be stored in the reactor graphite and that therefore measurements should be made of the actual energy storage. UCLA and the NRC Staff argued that no measurements were necessary because, they asserted, little Wigner energy could possibly be stored in the graphite.

- Each of the three parties in the UCLA reactor relicensing proceeding (the NRC Staff, UCLA, and ca:;) prepared different estimates of stored energy in that reactor.

The NRC aff predi~ted that 1 calories/gram might be stored in the UCLA graphite. 11 This prediction was part of a generic safety analysis used by the NRC to evaluate the safety of a whole class of research reactors (nArgonaut-type" reactors).

lO See nstored Energy* in the definitive NUclear Graphite, edited by R.E.

Nightingale, Academic Press, 1962, H?* 325-353.

11 s.c. Hawley, et al, *Analysis

• of cred1b

• 1 e Acc1'd ents for Argonaut Reactors*, NUREG/CR-2079, 1981, p. 37.

5

UCLA's witness Dr. Harry Pearlman testified under oath that the "real n amount of Wigner energy that could be stored in the UCLA graphite by the year 2000 was sufficient merely to produce a 15°c temperature rise~ 2 if released; this corresponds to about 3 cal/gram by the year 2000 or about 1 cal/gram in 1983. -

The committee to Bri~e the Gap calculated a minimum stored energy of 113 calories/gram in 2000, which corresponds to 39 calories/gram in 1983.

UCLA withdrew its renewal application before a final verdict could be rendered on the Wigner energy storage. BUt subsequent to reactor shutdown, researchers at U:,If4 have made some measurements of Wigner energy in a few graphite samples. ,

• The peak measured value reported by UCIA for Wigner energy in graphite, 33.2 calories~ rn_, is several times higher than the NRC s figure of 5 caI,?gram. It exc~ !?l ! factor of thirty UCLA's own "realistic* estimate of 1 cal/gram.

In contrast, UCLA's measured value of 33.2 ffl/gram is closely approximated by coo's minimum estimate of 39 cal/gram.

It is clear that the NRC's generic- analysis grossly underestimated the amount of Wigner energy stored in research reactor graphite, and therefore the potential for graphite fire at such facilities has likewise been underestimated. The storage of Wigner energy in research reactor graphite, like the hazards of a graphite fire, is an unanalyzed safety problem.

A research reactor which continued to operate could store sufficient energy to raise the graphite temperature several hundred degrees. Thus a 12 Transcript of July 23, 1983 hearing, p. 1870 13 "Testimony of COO Panel II - Chemical Reactions,n p. 13, in transcript of October 13, 1983, hearing, following page 2889. CBG estimated an upper bound on stored energy of 189 cal/gm in 2000 or about 65 cal/gm in 1983.

14 C.E. Ashbaugh, N.C. Ostrander, H. Pearlman, nGraphi te Stored Energy in the UCLA Research Reactor ,n Transactions of the American Nuclear society, June 15-19, 1986, Volume 52, pp. 372-373. The University has refused to permit independent confirmation* of its measurements, and th~e measurements may still understate the stored energy, due to the small number of samples taken and the apparent failure to take samples from locations that might produce higher values ( i.e. the optimum combination of coolest irradiation teq,eratures and highest fast flux).

15 The measurements also confirm coo's assertions (and contradict those of OCLA) as to the location of the peak fast flux, and hence the peak Wigner energy storage.

6

small powe~ surge, 16 or other event which by itself might cause little damage, could lead to graphite ignition and fuel melting.

The NRC's generic estimates of Wigner energy storage are wrong; licensee calculations have proved no better. Actual empirical measurements of Wigner energy will be required to assess the magnitude of the energy stored in research reactor graphite, and the magnitude of the fire hazard that it presents.

N. conclusion The Chernobyl accident proves once again that it is a mistake to assume that graphite fires are non-credible. Yet the NRC has based its regulatory approach to nonpower reactors on this mistaken assumption. Just as the Soviets began after Three Mile Island to recognize the necessity of reactor containment, we should learn from Chernobyl that graphite fires are credible accidents, and regulate graphite reactors accordingly. Above all, the NRC must require preparation of fire response plans which include the prevention and mitigation of graphite fires, and evacuation plans.adequate to protect the public in the event of a fire.

New measurements indicate that the NRC has grossly underestimated the amount of stored energy in the graphite of nonpower reactors, and consequently that the potential for a graphite fire in such reactors has also been underestimated. Since licensee calculations (such as those made by UCLA) are similarly unreliable, the NRC should order actual empirical measurements of stored energy in all non:p:,wer reactors which use graphite, as described above. safety analysis reports and hazards analyses should be revised to consider the consequences of a release of the stored energy, and the risks and consequences of reactor fires generally.

Respectfully submitted, Steven Aftergood Executive Director 16 POwer surges can result when large amounts of "reactivitT' are inserted into a reactor. This occurred recently at the campus reactor at Texas A & M University, when more than one "dollar" of positive reactivity was accidentally inserted into the reactor. see NRC preliminary notification Pro-N-86-16.

7

COMMITTEE TO BRIDGE THE GAP 1637 BUTLER AVENUE #203 OOC:KETEI.:

LOS ANGELES, CALIFORNIA 90025 USNRC (213) 478-0829

~ Jl 25 All :07 July 22, 1986 Chief oocketing & service Branch OFF ICE OF SEC~t IAt<Y OOCK ETlttG & SERVICF:

Office of the secretary BRANCH U.S. Nuclear Regulatory corranission oocKr.r r *1 ~ Fr' Washington, o.c. 20555 PEl I 11'-'i~ r,'-°U: PRM

• SO* '/--f-

Dear Sir or Madam:

Enclosed is a slightly revised version of the petition for rulemaking that we filed on July 8.

In particular, we have made two changes. On page 2, we have altered the wording of item (c) of the proposed rule. On page 4, footnote 9, we have changed "conservative" to "non-conservative."

We would appreciate it if you would acknowledge receipt of this letter.

Thank you.

Sincerely,

~~J steven Aftergood Executive Director

~sl ON P0

committee to Bridge the Gap July 8, 1986 OOC.KETED USNRC 1637 sutler Avenue, suite 203 (rev. 7/22/86)

LOS Angeles, CA 90025 (213)478-0829 "86 Jl 25 All :08 D""""'""T r . OFFICE Of Sti~Rt TARY DOCKETING & SEf?VICf.

PE 11 f1J1 RULE r,

--5~-t/cj BRANCH BEFORE THE NUCLEAR REGULA'IORY COMMISSION UNITED STATES OF AMERICA PETITION FOR RULEMAKIOO

'IO REDUCE FIRE HAZARD FROM NUCLEAR REAC"IDR GRAPHITE I. Introduction TWO recent developments indicate that the potential for a graphite fire in u.s. nuclear reactors has been inadequately addressed. First, the occurrence of a graphite fire at the Chernobyl plant in the soviet Union demonst.rates that such fires are indeed credible events, though they had been dismissed as non-credible by regulatory officials and reactor licensees. second, new experimental data show that the NRC's generic analysis of stored energy in research reactor graphite significantly underestimates the actual amount of stored energy, and thus underestimates the associated risk of graphite fire.

Because it had deemed graphite fires non-credible, the NRC failed to require basic safety measures which could help to reduce the threat of such a fire. Licensees whose reactors use graphite, including dozens of nonpower research reactors and one commercial power reactor 1, have no fire response plans for combatting graphite fires in their reactors. Research reactor licensees do not have adequate emergency plans to evacuate members of the 1 In addition, one unlicensed Dept of Energy reactor, the "N" reactor in Washington State, which produces commercial electric power as well as plutonium, also uses graphite.

public in the event of a graphite fire or other severe accident.

We therefore petition the NRC to issue an amendment to 10 Code of Federal Regulations Part 50, as follCMS:

All licensees whose reactors employ graphite as a neutron nooerator or reflector shall, by January 1, 1987:*

(a) Formulate and submit for NRC-approval fire response plans for combatting a reactor fire involving graphite and other constituent reactor parts (e.g. fuel) which might be involved in such a fire, taking into consideration the potential for explosive reactions. Response plans shall identify precisely which materials will be used to suwress a fire without increasing the risk of explosion, and shall indicate where and in what quantities these materials will be stored.

(b) Formulate and submit for NRC approval evacuation plans for a reactor fire. Plans should include evacuation out to a sufficient distance from the reactor such that no member of the public receives a oose to the thyroid greater than 5 rem, assuming a release to the environment of 25% of the equilibrium radioactive

• iodine inventory.

(c) Measure the "Wigner energy* stored in the graphite of their reactor. Revise the reactor safety analysis *report to consider how a release of stored energy would affect the outcome of other accident scaiarios (e.g., 'fire, reactivity accidents).

A sufficient number of graphite samples shall be measured to identify the location of maximum stored energy, and to determine the maximum quantity of stored energy to within 10%.

"Zero power" or "critical facilities", defined here as reactors which operate at 100 watts or less, are exempted from these requirements.

II. The NRC and Licensees Mistakenly Dismissed.Graphite Fires As Non-credible

'Ihe NRC and reactor licensees have held that graphite fires are "non-credible" events, and as a result they have failed to take measures to help mitigate or extinguish such fires, should they.occur. The NRC considers the dropping of a single fuel element to be the !!maximum credible accident" that can be suffered by most research reactors. 2 One reactor licensee has gone so far as to testify under ~ath that "graphite is considered -a non-

-* combustible material" and that "the so-called burning' of graJ;>hite is 2 see, e.g., NUREG/CR-2079, "Analysis of Credible Accidents for Argonaut Reactors", and NUREG/CR-2387, "Credible Accident Analyses for TRIGA and TRIGA-Fueled Reactors."

2

actually a surface reaction more analogous to the rusting 3f iron than to burning in the sense _of a self-sustaining, propagating fire.*

These erroneous notjons were first refuted thirty years ago by the Windscale reactor fire in England in 1957. In a discussion of that incident, the worst reactor accident prior to Chernobyl, Dr. c. Rogers Mccullough of the u.s. Atomic Energy commission stated:

By the way, this is an amusing point. 'Ihe belief had grown up on the part of many people in this country that graphite will not*

burn. This is nonsense. Graphite is carbon, and anyone knows that carbon wi1*1 burn .if you get it hot enough. But this gl~b remark*, that graphite will not catch fire, had become prevalent.

But the fire hazar*d associated with _graphite arises not only from its combustibility. A graphite fire, once burning, is also exceptionally

• difficult to extinguish. 'Ihere could be great danger in using either water or carbon dioxide to put the fire out, since combustible gases may be produced as a result. i:rbis is why advance planning to fight*a graphite fire is so important.

. . . , or. MCCUllough~s report on the Windscale incident describes how those fighting the fire tried var'ious methods *over a couple. of days. to put the fire out, all to no avail, and how they had to try, as a last resort, water:

Now they were faced with the decision either to use water or* to let the fire burn up. -They decided there was nothing left for them to do but put water in. There was some trepidation about this, as you can imagine, because they well knew that water on glowing urani.um makes hydrogen. Water on glowing carbon makes hydrogen and CX>; you have then a nice mixture of hydrogen,. en and air, ~ you might have an explosion. But they had no other choice.

As things turned out, an explosion fortunately did not occur. McCullough concludes:

I think it took a great deal of courage on the part of these people to put water on this reactor. They did it with fear* and trepidation, and in talking with them they *will not guarantee that 3 *Testimony of Dr. Walter F. Wegst Concerning the safety of the UCLA Research Reactor,* pp. 7-8, in NRC Docket No. 50-142, transcript of the July 25, 1983 hearing, after transcript page 2028.

4 *por a description of the accident, se T.J. Thompson and J.G. Beckerley, editors, The Technology of Nuclear Reactor Safety, MIT Pre*ss, vol. 1, pp.

633-636.

5 TID-7569, "Proceedings of the 1958 Atomic Energy commission and contractor safety and Fire Protection conference," June 24-25, 1958, p. 83 6 ibid., p. 76

they could do it a second ti.me without an explosion. 7 The situation at Chernobyl seems to have been at least as precarious.

It should be noted that under normal circumstances the Chernobyl reactor core and graphite were maintained in an inert, helium-nitrogen atmosphere, in which graphite combustion is truly "non-credible,* in fact impossible.

Many people have difficulty understanding that what may ordinarily be impossible can become quite feasible under accident conditions. Thus reactor licensees argue graphite is "non-combustible" and the NRC staff has argued that there may be insufficient air flow to sustain a graphite fire.

Bllt at Chernobyl there was no airflow-- lll'lder normal conditions-- and yet a graphite fire did break out, and the soviet Union was, like the British, was confronted with the formidable difficulti~s of trying to extinguish it.

According to news reports, the soviets were even forced to aw~oach- Swedish and West German experts for information on fighting such fires.

Whereas British and soviet scientists have. faced graphite fires with "fear and trepidation," the NRC. and reactor licensees have been satisfied to declare such fires "non-credible" and therefore unworthy of attention.

Yet graphite is used as a moderator in the Fort st. vrain nuclear power plant i'n Colorado, and in the dual purpose N reactor operated by the Department of Energyo Graphite is also used as moderator or reflector*in

- the majority of the -more -than- sixty-* nonpower research

• reactors throughout the U.S. Because graphite fires have been deemed non-credible, none-of these-facilities have considered how they would go about trying to extinguish a. graphite fire, and none of them have prepared fire response plans for -such fires. Graphite fires are also ignored in Safety Analyses, Hazards Analyses, Safety Evaluation Reports, Environmental Impact statements, etc. Most of the nonpower reactors have evacuation plans which extend no further than. the reactor. room itself. Since they usually have no containment and*no low-population buffer zone *{most of these facilities are based on university campuses), prompt evacuation after an ac§ident at one of these reactors is crucial- and currently unlikely to occur.

7 ibid., p. 78 8 "soviet Says Fire at Atomic Plant Followed Blast," New York Times, May 6, 1986, p. 1, at p. 4.

9 our proposed requirement for an evacuation plan (see p.2 above) cites a maximum permitted dose to the public of 5 rem. 'Ihis is derived from Table 1 of ANS 15.16, "Standard for Emergency Planning for Research Reactors,"

November 29, 1981, as reaffirmed in Table 1 of NUREG-084~, *standard Review Plan for the Review and EValuation of Emergency Plans for Research and Test Reactors." 'Ihe estimated 25% release to the environment of radioiodines is derived from the same ANS document, and may be non-conservative, considering that a reactor fire could propel a significant amount of radioactive materials including particulates into the environment.

4

III. The NRC Has.Miscalculated Wigner &iergy in Research Reactor Graphite; Empirical Measurements are Necessary "Wigner energy" is energy stored in nuclear reactor graphite during reactor operation as a result of neutron bombardment. It can be released suddenly as

• terrperature.

~r-t if the reactor temperature is raised above normal operating such a release of energy can potentially heat the graphite beyond its ignition temperature, cited in the literature as approximately 650 degrees. centrigrade. This phenomenon ("the Wigner effect") was a significant contributor to the Windscale reactor fire.

The question of how much stored energy might be present in research reactor graphite was the subject of vigorous technical dispute in the UCLA reactor relicensing proceeding (which terminated in 1985 after the reactor was permanently shutdown). That dispute and the measurements described below reveal the unreliability of calculated estimates of Wigner energy by the NRC staff and licensees, and indicate the need for empirical zreasurements.

The UCLA reactor was a 100 kilowatt Argonaut-type research reactor which was reflected and partially moderated by graphite. By 1983 the reactor had operated for the equivalent of 194 full power days. Had its request for relicensing been approved, it could have operated for a total of 560 full power days* by *the year**2000.

• Th ~mal neutron fl~ at f~ll pow.er had been.measured variously as 1.5 x 1~ 1 and 3.0 x 10 n/cm -second.

Graphite teIB?=ratures during irradiation .'were *generally quite low, on. the order of so 0 c. (Wigner energy storage is much higher at low temperatures than at high, where the graphite self-anneals.)

The committee to Bridge the Gap (CBG), the Intervenor in the UCLA reactor relicensing proceeding, had contended that under these circumstances (substantial integrated flux at ,low temperatures), a significant amotmt of Wigner energy could be stored in the reactor graphite and that therefore measurements should be made of the actual energy storage. UCLA and the NRC Staff argued that no measurements were necessary because, they asserted, little Wigner energy could possibly be stored in the graphite.

Each of the three parties in the UCLA reactor relicensing proceeding (the NRC staff, UCLA, and CEG} prepared different estimates of stored energy in that reactor.

The NRC 9.\aff predicted that ~ calories/gram migh:t, be. stored in the UCLA graphite. This prediction was part of a generic safety analysis used by the NRC to evaluate the safety of a whole class of research reactors

( "Argonaut-type" reactors).

lO see "stored Energy" in the definitive Nuclear Graphite, edited by R.E.

Nightingale, Academic Press, 1962, H?* 325-353.

11 s.c. Hawley, et al, "Analysis of credible Accidents for Argonaut Reactors", NUROO/CR-2079, 1981, p. 37.

5

UCLA's witness Dr. Harry Pearlman testified under oath that the "real" amount of Wigner energy that could be stored in the UCLA graphite by the year 2000 was sufficient merely to produce a 15°c temperature rise1. 2 if released; this corresponds to about 3 cal/gram by the year 2000 or about 1 cal/gram in 1983. - -

The committee to Bri~e the Gap calculated a minimum stored energy of 113 calories/gram in 2000, which corresponds to 39 calories/gram in 1983.

UCLA withdrew its renewal application before a final verdict could be rendered on the Wigner energy storage. BUt subsequent to reactor shutdown, researchers at u::f have made some measurements of Wigner energy in a few graphite samples. 4

'!he peak measured value reported by UCLA for Wigner energy in graphite, 33.2 calories per ~m, is several times higher than the NRC s figure of 5 cal/gram. It exc -El a factor of thirty UCLA's own "rea-listic" .estimate of 1 cal/gram.

In contrast, UCLA's measured value of 33.2 ffl/gram is closely approximated by coo's minimum estimate of 39 cal/gram.

- It is clear that the MC's generic analysis grossly underestimated the amoont of Wigner energy- stored in-research -reactor -graphite, and therefore -

the potential for graphite flre at'such facilities has likewise been underestimated. The storage of Wigner energy in research reactor graphite, like-the hazards of a graphite fire, is an unanalyzed safety prd::>lem.

A research reactor which continued to operate could store sufficient energy to raise the graphite temperature several hundred degrees. Thus a 12 Transcript of July 23, 1983 hearing, p. 1870 13 "Testimony of COO Panel II - Chemical Reactions," p. 13, in transcript of October 13, 1983, hearing, following page 2889. CBG estimated an upper bound *on stored energy of 189 cal/gm in 2000 or about 65 cal/gm in 1983.

14 c.E. Ashbaugh, N.c~ Ostrander, H. Peariman, "Graphite stored Energy in the UCLA Research Reactor," Transactions of the American Nuclear Society, June 15-19, 1986, Volume 52; pp. 372-373. The University has refused to permit independent confirmation of its measurements, and those measurements may still understate the stored energy, due to the small number of samples taken and the apparent failure to take samples from locations that might produce higher values (i.e. the optimum combination of coolest irradiation tenperatures and highest fast flux).

15 'Ihe measurements also confirm ca;'s assertions (and contradict those of UCLA) as to the location of the peak fast flux, and hence the peak Wigner energy storage.

6

small power surge, 16 or other event which by itself might cause little damage, could lead to graphite ignition and fuel melting.

The NRC's generic estimates of Wigner energy storage are wrong; licensee calculations have proved no better. Actual empirical measurements of Wigner energy will be required to assess the magnitude of the energy stored in research reactor graphite, and the magnitude of the fire hazard that it presents.

rv. conclusion The Chernobyl accident proves once again that it is a mistake to assume that graphite fires are non-credible. Yet the NRC has based its regulatory approach to nonpower reactors on this mistaken assumption. Just as the Soviets began after Three Mile Island to recognize the necessity of reactor containment, we should learn from Chernobyl that graphite fires are credible accidents, and regulate graphite reactors accordingly. Above all, the NRC must require preparation of fire response plans which include the prevention and mitigation of graphite fires, and evacuation plans adequate to protect the public in the event of a fire.

New measurements indicate that the NRC has grossly underestimated the amount of stored energy in the graphite of nonpower reactors, and consequently that the potential for a graphite fire in such reactors has also been underestimated. Since licensee calculations (such as those made by UCLA) are similarly unreliable, the NRC should order actual empirical measurements of stored energy in all nonpower reactors which use graphite, as described above. safety analysis reports and hazards analyses should be revised to consider the consequences of a release of the stored energy, and the risks and consequences of reactor fires generally.

Respectfully submitted, Steven Aftergood Executive Director 16 Power surges can result when large amounts of "reactivity" are inserted into a reactor. This occurred recently at the campus reac:tor at Texas A & M University, when more than one "dollar" of positive reactivity was accidentally inserted into the reactor. see NRC preliminary notification PID-IV-86-16.

7

DOCKET NUMS!

COMMITTEE TO BRIDGE THE G~ TJTION RULE PRM fa-1/f 1637 BUTLER AVENUE #203 LOS ANGELES, CALIFORNIA 90025 (213 ) 478-0829 July 7, 1986.

Chairman Lando Zech, Jr. DOCKETED Corranissioner James Asselstine USNRC Commissioner Frederick aernthal Corranissioner Thomas Roberts u.s. Nuclear Regulatory corrmission ~ Jl -7 ~-1:35 Washington, o.c. 20555 OFFICE OF Sf

Dear commissioners:

DOCKETING & iivARy BRANCH .VICt.

The graphite fire at the Chernobyl nuclear power plant raises the question of the potential for a similar fire at U.S. reactors which use graphite, including dozens of nonpower research reactors, the Fort St. vrain plant in Colorado, and the Department of Energy's N reactor. we believe that the possibility of such a fire at a U.S. reactor has been inadequately addressed by the NRC. While graphite fires are notoriously difficult to extinguish, no forethought has been given by NRC or reactor licensees as to how to fight such a fire; none of the facilities which use graphite have fire response plans which even address graphite fires, and the NRC does not require any.

Further, new data indicate that the NRC appears to have incorrectly assessed the amount of "Wigner energy" stored in research reactor graphite.

In a generic study which discusses this problem (NUREG/CR-2079, p.37), NRC contractors predicted that 5 calories per gram might be stored in the graphite of an A~gonaut-type research reactor. In contrast, UCLA researchers at the recent Annual Meeting of the American Nuclear society reported stored energy in reactor graphite as high as 33.2 cal/gram. over its operating lifetime, such a reactor could store sufficient energy to raise the temperature of graphite several hundred degrees. PUt another wal, sufficient Wigner energy could be stored to in effect lower the 650 c ignition temperature by several hundred degrees. It seems clear that NRC has significantly underestimated the potential for a graphite fire.

we have therefore prepared the enclosed petition for rulemaking, which formally requests that NRC issue a rule requiring operators of reactors that use graphite as a moderator or reflector (1) to prepare and submit for NRC approval fire response pl ans and evacuation plans for a graphite fire, and (2) to measure the energy stored in their graphite, and revise their safety analyses to consider the risks and consequences of a graphite fire in t hei r facilit ies.

Very truly yours,

~ ~,.11/

Steven Aftergood Executive Director enclosure cc w/enclosure:

Docketing and service Branch Office of the Secretary

\l S- N 0

DOCKET NUMBER *~ -

corrmittee to Bridge the Gap July ~ T:r RUL..E_P_R_ _"'!"'

1637 Butler Avenue, suite 203 LOS Angeles, CA 90025 (213)478-0829 DOCKETED USNRC

~ JI. -7 Pl 35 BEFORE THE NUCLEAR REGULA'IDRY COMMISSION UNITED SI'ATES OF AMERICA PETITION FOR RULEMAKING

'ID REDUCE FIRE HAZARD FROM NUCLEAR REACIDR GRAPHITE I. Introduction TWO recent developments indicate that the potential for a graphite fire in u.s. nuclear reactors has been inadequately addressed. First, the occurrence of a graphite fire at the Chernobyl plant in the ,.soviet Union demonstrates that such fires are indeed credible events, though they had been dismissed as non-credible by regulatory officials and reactor licensees. second, new experimental data show that the NRC's generic analysis of stored energy in research reactor graphite significantly underestimates the actual amount of stored energy, and thus underestimates the associated risk of graphite fire.

Because it had deemed graphite fires non-credible, the NRC failed to require basic safety measures which could help to reduce the threat of such a fire. Licensees whose reactors use graphite, inclyding dozens of nonpower research reactors and one commercial power reactor, have no fire response plans for combatting graphite fires in their reactors. Research reactor licensees do not have adequate emergency plans to evacuate members of the 1 In addition , one unlicensed Dept of Energy reactor, the "N" reactor in Washington State, which produces commercial electric power as well as plutonium, also uses graphite.

public in the event of a graphite fire or other severe accident.

We therefore petition the NRC to issue an amendment to 10 Code of Federal Regulations Part 50, as follows:

All licensees whose reactors employ graphite as a neutron rroderator or reflector sha~l, by January 1, 1987:

(a) Formulate and submit for NRC approval fire response plans for combatting a reactor fire involving graphite and other constituent reactor parts (e.g. fuel) which might be involved in such a fire, taking into consideration the potential for explosive reactions. Response plans shall identify precisely which materials will be used to suppress a fire without increasing the risk of ~xplosion, and shall indicate where and in what quantities these materials will be stor~.

• (b) Formulate and subm;it for. NRC approval evacuation plans for a reactor fire. Plans should include evacuation out to;a sufficient distance fr:om the reactor such that no member of .the public receives a dose to the thyroid greater than 5 rem, assuming a release to the environment of 25% of the equilibrium radioactive iodirie inventory. - *

(c) Perform measurements of the "Wigner energy" stored in the graphite of their reactor, and submit these m~asurements to NRC for review together with a revised_ safety analysis*, which shall address the risks and consequences*of a reactor fire.

A sufficient number of graphite samples shall be measured to identify the location of maximum stored energy, and to determine the maximum quantity of stored energy to within 10%.

"Zero power" or "critical facilities", defined here as reactors which operate at 100 watts or less, are exempted from these requirements.

II. The NRC and Licerisees Mistaken! Dismissed Gr ite Fires As NOn-cred le

'llle NRC and reactor licensees ha_ve held that graphite fires are nnon-credible" events, and as a result they have failed to take measures to help mitigate or extinguish such fires, should they occur. The NRC considers the drcwing of a single fuel element to be the "maximum credible accident" that can be suffered by most research rea.ctors. 2 One reactor licensee has gone so far as to testify under oath that "graphite is considered a non-combustible material" and that "the so-called 'burning' of graphite is 2 see, e.g., NUREG/CR-2079, nAnalysis of Credible Accidents for Argonaut Reactors", and NUREG/CR-2387, ncredible Accident Analyses for TRIGA and TRIGA-Fueled Reactors."

2

actually a surface reaction more analogous to the rusting 3f iron than to burning in the sense of a self-sustaining, propagating fire."

These erroneous notJons were first refuted thirty years ago by the Windscale reactor fire in England in 1957. In a discussion of that incident, the worst reactor accident prior to Chernobyl, Dr. c. Rogers Mccullough of the U.S. Atomic Energy commission stated:

By the way, this is an amusing point. The belief had grown up on the part of many people in this country that graphite will not burn. This is nonsense. Graphite is carbon, and anyone knows that carbon will burn if you get it hot enough. But this gl~b rerrark, that graphite will not catch fire, had become prevalent.

But the fire hazard associated with graphite arises not only from its combustibility. A graphite fire, once burning, is also exceptionally difficult to extinguish. There could be great danger in using either water or carbon dioxide to put the fire out, since combustible gases may be produced as a result. This is why advance planning to fight a graphite fire is so important.

Dr. McCUllough's report-on the Windscale incident describes how those fighting the fire tried various methods over a couple of days to put the fire out, all to no avail, and how they had to try, as a last resort, water:

Now they were faced with the decision either to use water or to let the fire burn up. They decided there was nothing left for them to do but put water in. There was some trepidation about this, as you can imagine, because they well knew that water on glowing uranium makes hydrogen. Water on glowing carbon makes hydrogen and CO; you have then a nice mixture of hydrogen, CO and air, ~ you might have an explosion. But they had no other choice.

As things turned out, an explosion fortunately did not occur. McCullough concludes:

I think it took a great deal of courage on the part of these people to put water on this reactor. They did it with fear and trepidation, and in talking with them they will not guarantee that 3 "Testimony of Dr. Walter F. Wegst concerning the safety of the UCLA Research Reactor," r:p. 7-8, in NRC Docket No. 50-142, transcript of the July 25, 1983 hearing, after transcript page 2028.

4 For a description of the accident, se T.J. Thompson and J.G. Beckerley, editors, The Technology of Nuclear Reactor safety, MIT Press, vol. 1, pp.

633-636.

5 TID-7569, "Proceedings of the 1958 Atomic Energy Commission and contractor safety and Fire Protection conference," June 24-25, 1958, p. 83 6 ibid., p. 76

they could do it a second time without an explosion. 7 The situation at Chernobyl seems to have been at least as precarious.

It should be noted that under normal circumstances the Chernobyl reactor core and graphite were maintained in an inert, helium-nitrogen atmosphere, in which graphite combustion is truly "non-credible," in fact impossible.

Many people have difficulty understanding that what may ordinarily be impossible can become quite feasible under accident conditions. Thus reactor licensees argue graphite is "non-combustible" and the NRC Staff has argued that there may be insufficient air flow to sustain a graphite fire.

BUt at Chernobyl there was no airflow-- under normal conditions-- and yet a graphite fire did break out, and the soviet Union was, like the British, was confronted with the formidable difficulties of trying to extinguish it.

According to news reports,, the soviets were even forced to app~oach Swedish and west German experts fbr information.on fighting such fires. 8 Whereas British and soviet scientists have faced graphite fires with "fear and trepidation," the NRC and reactor licensees have been satisfied to declare such fires "non-credible" and therefore unworthy of attention.

Yet graphite is used as a moderator in the Fort St. vrain nuclear power plant in Colorado, and in the dual purpose N reactor operated by the Department of Energy. Graphite is also used as moderator or reflector in the majority of the more than sixty nonpower research reactors throughout the U.S. Because graphite fires have been deemed non-credible, none of these facilities have considered how they would go about trying to extinguish a graphite fire, and none of them have prepared fire response plans for such fires. Graphite fires are also ignored in safety Analyses, Hazards Analyses, safety Evaluation Reports, Environmental Impact Statements, etc. Most of the nonpower reactors have evacuation plans which extend no further than the reactor room itself. Since they usually have no containment and no low-population buffer zone (most of these facilities are based on university campuses), prompt evacuation after an ac~ident at one of these reactors is crucial- and currently unlikely to occur.

7 ibid., p. 78 8 "soviet says Fire at Atomic Plant Followed Blast,n New York Times, May 6, 1986, p. 1, at p. 4.

9 our proposed requirement for an evacuation plan (see p.2 above) cites a maximum permitted dose to the public of 5 rem. This is derived from Table 1 of ANS 15.16, "Standard for Emergency Planning for Research Reactors,"

November 29, 1981, as reaffirmed in Table 1 of NUREB-0849, "Standard Review Plan for the Review and EValuation of Emergency Plans for Research and Test Reactors." The estimated 25% release to the environment of radioiodines is derived from the same ANS document, and may be conservative, considering that a reactor fire could propel a significant amount of radioactive materials including particulates into the environment.

4

III. The NRC Has Miscalculated Wigner Energy in Research Reactor Graphite; Empirical Measurements are Necessary nwigner energyn is energy stored in nuclear reactor graphite during reactor operation as a result of neutron bombardment. It can be released suddenly as ~r-t if the reactor temperature is raised above normal operating terrperature. such a release of energy can potentially heat the graphite beyond its ignition temperature, cited in the literature as approximately 650 degrees Centrigrade. This phenomenon ("the Wigner effect") was a significant contributor to the Windscale reactor fire.

The question of how much stored energy might be present in research reactor graphite was the subject of vigorous technical dispute in the UCLA reactor relicensing proceeding (which terminated in 1985 after the reactor was permanently shutdown). That dispute and the measurements described below reveal the unreliability of calculated estimates of Wigner energy by the NRC staff and licensees, and indicate the need for empirical IIEasurernents.

The UCLA reactor was a 100 kilowatt Argonaut-type research reactor which was reflected and partially moderated by graphite. By 1983 the reactor had operated for the equivalent of 194 full power days. Had its request for relicensing been approved, it could have operated for a total of 560 full power days by the year 2000. Th mal neutron f ~x at f~ll power had been measured variously as 1.5 x 1012 and 3.0 x 10 1 n/cm -second.

Graphite temperatures during irradiation were generally quite low, on the order of so 0 c. (Wigner energy storage is much higher at low temperatures than at high, where the graphite self-anneals.)

The Committee to Bridge the Gap (CBG), the Intervenor in the UCLA reactor relicensing proceeding, had contended that under these circumstances (substantial integrated flux at low temperatures), a significant amount of Wigner energy could be stored in the reactor graphite and that therefore measurements should be made of the actual energy storage. UCLA and the NRC Staff argued that no measurements were necessary because, they asserted, little Wigner energy could possibly be stored in the graphite.

Each of the three parties in the UCLA reactor relicensing proceeding (the NRC staff, UCLA, and CEG) prepared different estimates of stored energy in that reactor.

The NRC 9.\aff predicted that ~ calories/gram might be stored in the UCLA graphite. This prediction was part of a generic safety analysis used by the NRC to evaluate the safety of a whole class of research reactors (RA.rgonaut-type* reactors).

lO See nStored Energyn in the definitive NUclear Graphite, edited by R.E.

Nightingale, Academic Press, 1962, pp. 325-353.

11 s.c. Hawley, et al, nAnalysis of Credible Accidents for Argonaut Reactorsn, NUREG/CR-2079, 1981, p. 37.

5

UCLA's witness Dr. Harry Pearlman testified under oath that the "real" amount of Wigner energy that could be stored in the UCLA graphite bv the year 2000 was sufficient merely to produce a 15°c temperature rise1 2 if released;* this corresponds to about 3 cal/gram by the year 2000 or about 1 cal/gram in 1983. -

The Committee to Bri1~e. the Gap calculated a minimum stored energy of 113 calories/gram in 2000, which corresponds to 39 calories/gram in 1983.

UCLA withdrew its renewal application before a final verdict could be rendered on the Wigner energy storage. BUt subsequent to reactor* shutdown, researchers at U':_Lt have made some measurements of Wigner energy in a few graphite samples. 4 The peak'measured value reported by UCLA for Wigner energy in graph~te, 33.2 calories pet gram, is several times higher than the NRC s figure of 5 cal7grarn. I~ exceeds _eY a* factor of thirty UCLA"'s own "realistic" estimate of 1 cal/gram.

In contrast, UCLA's measured value of 33.2 ffl/gram is closely approximated by ca:;'s minimum estimate of 39 cal/gram.

It is clear that the NRC's generic analysis grossly underestimated *the amount of Wigner energy stored in research reactor graphite, and therefore the potential for graphite fire at such facilities has likewise been underestimated. The storage of Wigner energy in research reactor graphite, like the hazards of a graphite fire, is an unanalyzed safety problem.

A research reactor which continued to operate could store sufficient energy to raise the graphite temperature several hundred degrees. Thus a 12 Transcript of July 23, 1983 hearing, p. 1870 13 "Testimony of COO Panel II - Chemical Reactions, 0

• p. 13, in transcript of October 13, 1983, hearing, following page 2889. CBG estimated an upper bound on stored energy of 189 cal/gm in 2000 or about 65 cal/gm in 1983.

14 c.E. Ashbaugh, N.c. Ostrander, H. Pearlman, "Graphite Stored Energy in the UCLA Research Reactor , 0 Transactions of the American Nuclear Society, June 15-19, 1986, Volume 52, pp. 372-373. The University has refused to permit independent confirmation of its measurements, and those measurements may still understate the stored ~ergy, due to the small number of samples taken and the apparent failure to take samples from locations that might produce higher values ( i.e. the optimum combination of coolest irradiation temperatures and highest fast flux).

15 The measurements also confinn ca:;'s assertions (and contradict those of UCLA) as to the location of the peak fast flux, and hence the peak Wigner energy storage.

6

small power surge, 16 or other event which by itself might cause little damage, could lead .to graphite ignition and fuel melting.

The NRc's generic estimates of Wigner energy storage are wrong; licensee calculations have proved no better. Actual empirical measurements

• of Wigner energy will be required to assess the magnitude of the 'energy s,tored in research reactor graphite, and the magnitude of the fire hazard

, that it presents.

rv. conclusion

'lbe Chernobyl accident proves once again that it is a mistake to assume that.graphite fires are non-credible. Yet the NRC has based its regulatory approach'to nonpower reactors on this mistaken assumption. Just as the soviets began after Three Mile Island to recognize the necessity of reactor containment, we should learn from Chernobyl that graphite fires are credible accidents, and regulate graphite reactors accordingly. Above all, the NRC must require preparation of fire response plans which include the,prevention and mitigation of graphite fires, and.evacuation plans adequate to protect the public in the event of a fire.

New measurements indicate that the NRC has grossly underestimated the amount of stored energy in the graphite of nonpower reactors, and consequently that the potential for a graphite fire in such reactors has also been underestimated. Since licensee calculations (such as those made by UCLA) are similarly unreliable, the NRC should order actual empirical measurements of stored energy in all nonpower reactors which use graphite, as described above. safety analysis reports and hazards analyses should be revised to consider the consequences of a release of the stored energy, and the risks and consequences of reactor fires generally.

Respectfully submitted,

~

Steven Aftergood Executive Director 16 POwer surges can result when large amounts of "reactivity* are inserted into a reactor. '!'his occurred recently at the ~ s reactor at Texas A & M University, when more than one *dollar* of positive reactivity was accidentally inserted into the reactor. see NRC preliminary notification POO-IV-86-16.

7