ML20008D667

Responds to 790110 Request for Addl Info on H-451 Graphite, Info Concerns Stress Limits,Strain Behavior,Clarification of Previously Submitted Data & Cesium Transport

ML20008D667
Person / Time
Site:	Fort Saint Vrain
Issue date:	02/06/1979
From:	Wessman G GENERAL ATOMICS (FORMERLY GA TECHNOLOGIES, INC./GENER
To:	Gammill W Office of Nuclear Reactor Regulation
References
FOIA-81-127 NUDOCS 7902140093
Download: ML20008D667 (12)

Text

1 e -, . , , 1 av y .,

. . = = . = = ===-

i n.u..

{!/

.:w  : - = == =

I c m ,. c ::. c: x:n ,

Po rct un1 s:sc.m w can aa *

(714:4 4 /,00 t

i February 6,1979 i

i I'

l:r. Ull11a:a Ga:: nill ,

Assistant Director for Advanced Roactors  !

Division of P.aject !bnag;;;at U. S. Iluclear llaculatory C=aissicn Uachingten. D.C. 20555 Caar 14r. Gmn111, '

Gar.aral Atc:aic Cc:pany is enclosing fifty (CO) ccpics of its response l to your Raquast fer Additional Information (CAI) ca 11-451 Cra:;hito, dated t Jan':ary 10, 1979. f;o respanse to itc:a 231.18 is being provided since it l was prc0ca':0d in tha RAI as a point of clarificatica caly. '

If you have any questions regarding those responses, plesso do not hasitate to let us knew.

Sincerely.

.A-c&>s m-G. L. L'essman, Director Plant Licensing Division GLU:mk 4

1

.]

i

)

i 1

MAC1CO l

+ ppe y y .;

[ .O l .

STATI CF CALIFGr< ilia )-

, ) ss.

! COUlTI CF SA 1 DIEGO )

After bain9 duly st orn, tha parcen tnce to ma to be G. L. Wessman of Ganoral At0mic Cc pany, signed the within do::: ant this 6 day of February 1979.

l WITi4E55 ny hand and official scal.

1 I

":_^:::: :::_ _- - - _- - - - - - -_- - :_-_ g j j' )j y,

,rjjy Q g

! cotorytvuoftc ct 4 % 7' c .L.a.g l s

Ud' R1'4 J9 d [ ,.+1j *g d.IJJ1 m u.... w...vw:::::::::::::.

f l

l e

l 1

l O

s

.]

. e R;;;rmi TO ;Dr. GOG7",^Ji U0.2 ~ yO P 7WW t

CUSTICI 130.2 In your responso to Q130.la, you indicated that unirradiated H-451 graphite dcos not experience fatigua damaga under reversed stross cycling unless stressed above a h:malogous stress limit of 0.G3 in the axial direction or 0.74 jn tha radial o iontation.

Thaco hcmalogous st.'oss limits are ut.carstcod to be established on the basis of 50 porce:t survival. Indicate ycur ratienale for not using limits for 99 percent survival. If the limits for 99 porcent survival are used, will your conclusien be still the sama?

RISPCNEI Tha h::cigous fatiguo stress limits c;uotad in the respcaso to l Quastion 130.la were inadvertently taken from the wrong secticn of Refer-ence 1. The cmanded values of the H-451 graphite h :alogous fatigue stress limits for revorced stress cycling for 50% speciman survival at j 100 cycles are 0.60 in the axial direction and 0.75 in the radial direc-tion (Ref. 1).

The homologoue Nico stress limits for reversed stress cycling for 99% survival w% TS sonfidence at 105 cycles are 0.44 for the axial direction and 0.d ':a r:Jial direction (Ref.1). If these limits for 99% survival ate used to qvaluate the fatigue lives of the H-451 elemants, the conclusions in the prior response to Question 130.la remain the sama. Both secondary and primary stresses have been calculated to ba well below the stress limits for 99% st rvival, and during normal oper-ation no fatigue inducing mechanism exists.

)

f I

REFERENCE

, i

1. Price, R. J., " Cyclic Fatigue of Near Isotropic Graphite: i Influence j of Stress Cycle and Neutron Irradiation", GA-A145S8. Noverrter 1977. 1

._ 1

=- __ J

1 i

.ul. :. .:.: TC :;.~,C O. :37'C:1 CFM .17 '" :. 4:' T *?"!T: l*

i i

D RSTIC1 231.17 The respanse to ist round question 231.3b indicated that because ,

of the high degree of cceplexity of strain behavior in the radial orientation (as ccupared with axial strain behavior), no general l ccmparison of stresses caused by irradiation strain in H-327 and  !

H-451 graphites has been provided. Vet, the effects of these j radial shrinkages are said to have boca included in the analyses.

Notwithstanding the asserted ccmplexity in radial strain cohavior, scma linkage must be provided, at least in the fonn of an excmplar  ;

calculation, between the radial strains and the stress provided I in Table 2-1 of GLP-5583. l l

RESP 0NSE The ccmparison of stresses caused by irradiction strains in H-327 and i H-451 gr; hite is included in Table 2-1 of GLP-5588 for strains in both e the axial and radial directions. The behaviors of the two graphites in the axial direction were compared in the text of Section 2.4.1.4 because the irradiation strains are so well behaved in the axial direction that they I can be described and ccmpared in simple terr- (The axial strain behaviors of H-327 and H-451 graphites are shown in Figures 4-1 and 4-5 (a) of GLP-5588).

• In contrast, the radial irradiation strain behaviors of both H-327 and H-451 graphite are complex (Figures 4-2 and 4-5 (b) of GLP-5588),

and it is difficult to provide a general description of this behavior in I words. Although the radial strain behavior is not described in the text of Section 2.4.1.4 (GLP-5588), it is included in the calculation of stresses and is reflected in the analytical results. All material properties, in-ciuding the radial strain behavior, are included as input to the stress analysis computer codes, FESIC and SAFE'GRAFITE. A detailed understanding of the impact of the radial strain behavior for the specific temperature and fluence history that produced the stresses listed in Table 2-1 can be obtained only by examining the details of the analysis.

c.

+- e Y

j . .

piSpr WE TO HRC CUESTION 231.19 03 H-491 GPAPHITE l

CUISTTON 231.19 (a) Althougn *here are no data for the diffusion of strontium in H-451 graphite, it is stated in the report that, based on the data shcwn in Ta'olo 7 and Figure 3. "there appears to be no dependence on the type of graphite." please indicate which i. ' the referenced data are for near-isotropic graphites. If little or no strontium diffusion data exist for near-isotropic needle-coke graphites, discuss the applicability of l tha data to H-451 graphite. .

(b) The strontium diffusion data in Figura 3 of the report have a r fairly utde scatter (1 to 2 orders of magnitude) in the lower ranga af j tcmparatures whereas at high tcmperaturas the scatter is smaller. % ease

, indicate tha ranga of expected operating tcaperatures in the Fort St.

j Vrain H-451 blocks.

~

I l

l RESPC;i3E TO 231.19(a) l

( None of tha data shown in the response to Question 231.12 are for the dif-fusion of strontium in near-isotropic graphite. The conclusion that strontium diffusion in H-451 graphite does not differ significantly frcm that in H-327 graphite is Inferred based on the fact that the available data, taken on many graphites, indicate no dependence of strontium diffusivity upon graphite type.

RESPONSE TO 231.19(b)

As noted in the response to Question 231.16c, the expected range of operat-ing temperatures in the Fort St. Vrain H-451 fuel blocks is 420-1210*C. It should be noted that most of the scatter in strontium diffusivity data at low temperatures is below the fit to the data shown in Figure 3 of the response to Question 231.12.

l J

d 3

D79.?r:MT TO NDP DMeSTfDN 211.20 ON H 451 GPap4!TE ,

t GUISTION i i

231.20 (a) The report indicates that, whereas strcntium diffusion in graphite can be described by Fick's Law, cesium transsort is a more ccmplex I phencmenoa. Yet cesium transport is described in the report in terms of a portaatien coefficient, defined by the equation J = n (Cl-C2 )/L.

(report equation 9), which is a form cf Fick's Law. Please discuss this apparent contradiction; that is, discuss how a process that is acknowledged to be more complex, can be described by Fick's Law, when Fick's Law is known to be generally applicable only to very simple diffusion cases.

I (b) There appears to be only one dat;. (see report Figure 4) indi-cating the effect oi ir-adiation aa ccaium permeation coefficients for H-451 graphite and ncne fcr :t-32'/. It is difficult, therefore, to accept the conclusicn tha+, 'f a po*aastion coefficient for Cs in .

H-451 graphite is significan;fy e2duced in-pile." Please discuss tha .

effect of using the permeat.cg coefficients for unirradiated graphite {

as interim values until care data are obtained on cesium diffusion -

i in H-451 in-pile. Please indicate whether such data are to be ob- i tained on the 8 test elemants currently scheduled for the first li reload in Fort St. Vraia. ,

RESPONSE TO 231.20(a)

The definition of the permeation coefficient, J. "(ci-co) n (( = , (3) is analogous but not equivalent to the definition of the diffusion coefficient where Fick's laws apply. In Eq. (1), J is the flux, n the permeation coeffi-cient, C1 the concentration of the diffusant at the source boundary of the madium in which diffusion occurs, C2 the concentration at the sink boundary, and L the thickness of the medium, assuming slab geometry.

Fick's first law of diffusion is (Ref.1)

J= -

0(3C/3X). (2) l I

• _ .wme* w wee goy . 4 . .m__ _,

C Ca.nida.', far sim311aity, the steady-state ccaditi n for a slab cf gra;hite ci thickness L, also, let C2 = 0. Then, Fick's first law beccmes J = CC1/L. (3) 1 Fick's laws (first and second) also imply that the leading of the graphite par unit area is H = C1L/2, (4) '

i i

where H is tha leading. Thus, the flux and leadir.g are related according to i J = 2CM/L2 . (5) i l i

! I It is this combination of flux and loading which is characteristic of diffu- j sion in media to which Fick's laws apoly under steady-state conditicns. Anal- {

ogous relaticns can be obtcined for non-steady-stato conditions. When Fick's ows apply, steady-state leading anJ nass flux are simultanacusly attained. f The perceation ccefficient (Eq. 1) can be used to describe the rate of [

Cs migration thrcuch and release from graphite. Ho.ever, nothing is implied

{

by Eq. I about the leading of the graphite. The fact that Cs loadings observed I when studying Cs diffusion in graphite (Ref. 2) did not attain steady-state f values at the sama tiraa as the Cs mass flux at the sink boundary reached its

{

steady-state value leads to the conclusion that Cs diffusion coes not obey l Fick's 1cws.  !

I RESPONSE TO 231.20(b)

Transport data for cesium in H-327 granhite in-pile are available (Refs.

2-4) and also support the conclusion that cesium transport is significantly reduced in-pile. In the t.ase of H-451 graphite, the one comparison between in-Dile and out-of-pile transport was based on two experiments in which the sane apparatus and source loading were used in-pile and cut-of-pile.

Nevertheless, the effect of using pomcation coeffic!ents for unirradiated H-451 graphite as interim values has been examined. As shown in Figare 4 of tne response to Question 231.12, the pemeation ccefficient of cesium in unir-radiated H-451 graphite (solid line) is greater than values given in the FSAR 1

' broken line) by about a factor of seven. The cesium source tenr.s in the FSAR I are based on the per eatien coefficients given in the FSAR. If it is assumed,

_- ..-______m_ _ _ _ _ _ .

o es a first approximaticn, that casium release is directly prcportional to the  ;

p2emaatica coef ?icienc in graphice, then the calculated ces'um-137 "Cesign" plateout inventory of 5460 C1 given in Table 3.7-2 of the FSAR beccmes approxi-mately 33,000 Ci.

3 Tha effect of this increased cesium source term on accident consequences l discussed in Chapter 14 of the FSAR was investigated. The two-hour exclusion

{

area boundary (EA3) bone dose for Design Basis Accident fic. 2, Rapid Depres-l surization/Blcudown, is the only dose significantly affected by the hypothetical increase in casium inventory.

i Using the same calculational methods used in the FSAR analysis, the two-hour EAB bone dose for 03A-2 is determined to be increase:j by about 0.5 rem if the Cs-137 in/entory is increased to about 38,000 Ci. The increase is relatively small because the boae dose effectiveness of Cs-137 ccmpared to strontium-90  !

is relatively small. The resultant bone dose rcmains will below eccepted limitt. '

With regard to test elements FTE-1 through FTE-8, the currently olanned DOE-funded post-irradiation examination program (PIE), described in Amendment 2 (Appendix A) of General Atomic Repo t GLP-5494, includes gamma spectroscopy

[

of graphite c. nents and destructive fission product examinations. These tests will allow the general characterization of fission product migration, including cesium transport through H-451 graphite.

Although analysis of data obtained from the test elements can be used to check our ability to calculate fission product migration in HTGR fuel within

'the uncertaintfes of PIE measurements, material properties, and operating con-ditions, this anal sis

/ will frobably noc enable accurate determination of I permeation parameters. Precise determination of permeation coefficients requires accurate control of tesc conditions, including temperature histories and fission product source concentrations. Such control is not attainable during test element irradiation in the FSV core.

Ll i

^

f.

n

.az; = ricIs i

1. J. Crank, "The Ibthematics of Diffusion", 0:: ford University Press,1956.  :

2. D. Chandra and J. H. Norman, " Diffusion of Cesium Through Graphite",

J. Nucl . ibter. 62, 293-310 (1976). .

i 3. C. F. Wallroth, et.al., " Post Irradiation Exaraination of Peach Bottom t l Fuel Test Elemant FTE-3", USAEC Report GA-A13004, General Atomic Company, August 15, 1974.

4. C. F. Wallroth, et.al., " Post Irradiaticn Examination of Peach Bottom Fuel Test Element FTE-6", USERDA Report GA-A13943, General Atomic Company, September 1977.

t t

1 Y

i

. _ . _ . _ _ _ _ _ _ _ _ _ _ _ t

a t

• i DRSpG15R 70 tmC GIRSTM:1 231.21 0:1 H-F,1 O'? HIT.

CUISTI0tl i

231.21 The response to 1st-round Quastien 231.16 indicates (Table 1) that the creep tests on H-4519 aphite will not be completed until 1983, whereas H-451 gra::hito reloed fuel elements could be placed in Fort St. Vrain as soon as late 1980 or early 1981.

Moreover, much of the curre.at creep data base on near-isotropic graphite appears to have L;an generated on Gilsocarbon-based, near-isotropic graphite retiar than petroleum coke-based, near-isotropic nraphites. Invicwof(a)thecurrentlimitations on the data base on H-451 grar'11te and (b) the uncertainty regarding the applicability oY Gilsocarbon graphite creep data, please indicate how the test data yet to be developed would lead the expected irradiation exposures of reload elements.

Also, show hcw the planned post-irradiation examination program .l on the 8 test elcments and future relcad elcments will provide }

dimensional change data that would be used to verify the pre- '

dicted changes that are based, in part, on the expected creep behavior.

-RESPON3E Based on current projections, earliest possible use of H-451 graphite in Fort St. Vrain reload segment 9 would begin in early 1981. Figure 1 shows the schedule for completion of H-451 graphite irradiation creep tests at JRC Petten and at ORNL. Creep tests to 7.4 x 1021n/cm2 (E >

0.18 MeV, HTGR) at 840*C in the H1gh Flux Reactor, Petten, will be com-  ;

pleted at about the end of 1979 and confirmatory tests at 900*C at ORNL will reach full fluence in mid-1981. (The1 Ptter test was inadvertently ted from the response to Question 231.10.) 600*C creep test data fro.a ORNL will be available to 3 x 1021 n/cm2 by mid-1980 and full exposure data at 6.00*C will be available in mid-1982. ORNL creep tests at 1250*C, planned for a later date, are not directly relevant to FSV operation since most of the graphite in FSV operates at significantly lower temperatures. Thus, full exposure creep data at 840-900*C will be available in time for the earliest use of H-451 in Fort St. Vrain.

Full exposure creep data at 600*C will be available when the first H-451 reload segmtit has ac:umulated about 1 x 1021 n/cm2 l 1

l m . , . - - _ - - . - _. .

j

i

. d.

. . c iia corruutty pianaaa Lud-punued pusc-irrauiac.un enuniiation vic/

r

7. am 6- tact ch::nt: FTE-1 thr ugh FTI-010 d:::ribd in . tar.d :n: 2 '

('ppand1x .O of Ganaral At;mic P,epart GLP-Edg4, June IJ, '977. The pro- ,

gram includes detailed graphite block metrology for all eight test elcmants.  ?

Cata will Oe recorded en magnetic tape and will be ccmpared with pre-irradiaticn maasure:r. ants to establish irradiation induced graphita strain and bow.

Maasured strain and bcw can be ccapared with strain and bow calcu-lated for each test element to provide an indication of the validity of such calculations. Tha maasured strain and bcw should agree with calcula- l tions within the uncertainties of the PIE measuremants and uncertainties in material preparties and cperating conditions. Input to such calculations, however, includes numarcus physical properties of graphite; it is, accord-

.ingly, not possible to assess the accuracy of any single input property, .

such as creep behavior, by examiati.g behav* r which is simultaneously dependent upon several properties.

The irradiation creep tests described aNve and in the response to question 231.16 will provide the primary data base for H-451 graphite creep behavior.

mm-

c l

l ICl9 I l'MO l 1991 l 1 % 't l 19S3 l i 4'2E l 1925 i i ci2 6 i M CEO.9 (;;.ygy lato ufg 2xt;[*'  ?,280 Guto 62i0 (G Ru c-;7 0 O U 9 9 gx:,g e;y [77/ / / / // - // , / // /// , ,

2 7.imo caca, 65=t0' y r,4 d'gleg n

0;T53: cfc)'- .' , e i oern. casp Sn= to"N/cs* 9 ud'ufen*

q r g ,3 (,qg c;i / -

/ / /1

~

3 # to*' ajc?  % no*' ujen -

cast. cga p 9 q sm,(w*c} / / / / / / i ! i i i i / , ,;

FIG 10E 1 Schedule for completi;r: of H-451 graphite irradiation creep tests and earliest possible use of H-451 in Fort St. Vrain (reload segmant 9),

l l

1 I

1 l

1 1

1 l

-- - -- - ~ ~ _ _.

j

-l

!