ML19270G920

Forwards Response to 790529 Questions GA-LTR-23 to Facilitate Review of UC-2 Fissile Fuel

ML19270G920
Person / Time
Site:	Fort Saint Vrain
Issue date:	06/13/1979
From:	Fisher C GENERAL ATOMICS (FORMERLY GA TECHNOLOGIES, INC./GENER
To:	Gammill W Office of Nuclear Reactor Regulation
References
NUDOCS 7906210141
Download: ML19270G920 (14)

Text

. .

. . . .. ME mhe 4%4 . . - "wW-wwmutwee (a

. T hJ f p P,1 c. ' r tu g' ws -',-r 7 9 r-www'7 7 N7,="=rm, r,=W'm ,Tr="r Wr'8 WIW'."'d PT 7k MTIE%M*E NtM GENERAL AToMC COMPANY PO DOX 81608 SAN DIEGO. CALIFORNIA 92138 (714) 45KW June 13, 1979 Mr. William Gammill Assistant Director for Advanced Reactors Division of Project Management U. S. Nuclear Regulatory Cocmission Washington, D.C. 20555

Dear Mr. Gammill:

Enclosed are fif ty (50) copies of General Atomic Company's response to NRC questions transmitted via the May 29, 1979 letter, " Questions and Comments on GA-LTR-23". We trust that the enclosed information will enable you to complete your review of UC2 fissile fuel for Fort St. Vrain by your previously committed date of June 29, 1979 or, at the latest, within thirty days after receipt cf these responses.

If you have any questions regarding these responses, please do not hesitate to contact us.

Sin rely

(

A W

C. R. Fisher, Director Plant Licensing Division CRF:jr Encl.

2238 270 7906210191

STATE OF CALIFORNIA )

) ss.

COUNTY OF SAN DIEGO )

After being duly sworn, the person known to me to be C. R. Fisher of General Atomic Company, signed the within document this /fE day of June 1979.

WITNESS my hand and official seal.

g rr; + t gp.. $

21 ., _. ;A C ':1 -

1, . l.~ .1 v . J '. !'." C..

~

FW4 G r,b .rY .+k )

t r. t IN N.U' Did !);id) CQuieTf 9~ [l3 Wy Commission Exps'es May 29.1382 u u ,w ~... _

. . . . .w .c. n . ., v w. i

( Notary Public 2238 271

I RESPONSE TO NRC QUESTION 231.1 ON UC2 SAR QUESTION 231.1 (a) The second page of the response to Lead Item VII contains the statement that "The TRISO coated UC2 particle design is developed on the basis of calculated sic stress distri-butions which assure that the expected particle failure from internal fission gas pressure is less than or equal to that calculated for FSV TRISO coated (Th/U)C2 fuel." This state-ment appears to be inconsistent with the design basis pre-sented in earlier submittals and publications on TRISO UC2 particles (see, for example, Gulden ett al, "The Mechanical Design of TRISO-Coated Particle Fuels for the HTGR," Nucl.

Techol. 16,100 (1972) and General Atomic report GA-A12971).

Perhaps the quoted statement is intended to provide the design basis for only the Fort St. Vrain UC2 fissile particles and not the design basis for the large HTGRs in general. Please explain.

(b) The above-quoted statement appears to ignore potential fuel failure mechanisms such as kernel migration or sic-fission product interaction. In the case of the large HTGR appli-cations (viz. the Su=mit, Fulton, and GASSAR plant), kernel O migration was explicitly tied to the thermal design, and the buffer carbon coating thickness was related both to the thermal design and the kernel migration rate (i.e., are all interrelated).

Please explain, therefore, why the above-quoted design basis statement ignores kernel migration and other potential failure mechanisms and addresses only pressure vessel-type failure.

RESPONSE TO 231.la The TRISO coated UC2 particle design developed for Fort St. Vrain is based on calculated sic stress distributions which assure that the expected particle failure from internal fission gas pressure is less than or equal to that calculated for FSV TRISO coated (Th/U)C 2 fuel. This design basis is conservative but different from that proposed for TRISO coated UC 3 in large HTGRs. Reference 1 states that the LHTGR mechanical design basis for TRISO coated UC2 is <0.5% pressure vessel failure for fuel subjected to 78% FIMA, 8x 1025 n/m2 (E>29fJ), and 12500C. This design basis evolved from Reference 2, which supported the design assumption in Reference 1 that 50%

of the fuel having calculated Sic tensile stresses exceeding 30,000 psi would fail. In contrast, the FSV TRISO coated UC2 particle design is based on 2238 272

normalizing its performance to TRISO coated (Th/U)C2 fuel tested in capsule F-30 (Reference 3), the irradiation proof test of FSV fuel. Based on the concept of a sic failure criterion described in the response to NRC Lead Item VIII, the expected (Th/U)C2 pressure vessel failure fraction was cal-culated at 0.5% for fuel subjected to peak exposure conditions. This ex-pected failure level is consistent with the LHTCR design basis for IRISO coated UC 2 . Consequently, 0.5% was used as an upper prescu*e vessel failure limit to establish the TRISO coated UC2 particle design for FSV. As originally stated in the response to Lead Item VIII, the expected pressure vessel failure level for the nominal FSV UC2 design is 0.2%, which represents a considerable performance margin compared to (Th/U)C2 . (Refer to additional comments in response to Question 231.3.)

RESPONSE TO 231.lb The design of HTGR fuel is established to assure acceptable behavior af ter considering the impact of all perfor=ance limiting phenomena. Kernel and coating designs are first developed based on pressure vessel considerations.

After establishing fuel particle designs, the potential impact of kernel migration and sic-rare earth fission product reactions is also evaluated. The kinetics of kernel migration and sic-rare earth fission product reactions are given in sections 4.1.2 and 4.1.3 respectively of GA-LTR-23. As discussed in sections 4.1.2 and 4.1.3, these phenomena will not limit TRISO UC2 Performance in FSV.

REFERENCES

1. Smith, C. L. , " Fuel Particle Behavior Under Normal and Transient Con-ditions", GA-A12971, October 1, 1974

2. Gulden, T. D., et. al., "The Mechanical Design of TRISO Coated Particle Fuels for the Large HTGR", Nucl. Tech., Vol. 16, October 1972, p. 100.

3. Scott, C. B. and Harmon, D. P., " Post-Irradiation Examination of Capsule F-30", GA-A13208, April 1, 1975.

2238 273

3 RESPONSE TO NRC QUESTION 231.2 ON UC2 SAR QUESTION 231.2 Please provide the origins'or basis (and appropriate references) for the 231 MPa (33,500 psi) failure criterion listed on the third page of the response to Lead Item VIII.

RESPONSE

The failure criterion of 231 MPa (33,500 psi) for TRISO coated UC2 fuel was empirically determined from reference type UC2 fuel tested in irradiation capsules P13R and P13S. The post irradiation examination results for capsules P13R and P13S are documented in Ref. 1; however, the use of this test data to derive an empirical failure criterion for UC2 has not been pre-viously documented. The data base which supports the failure criterion of 231 MPa is summarized in Table 1. The data include ten separate batches of TRISO coated UC2 fuel tested over a range of irradiation conditions. Fourteen separate sic stress calculations were performed on these batches, and the average calculated failure criterion was 231 MPa.

REFERENCE

1. Scott, C. B. and Harmon, D. P., " Post-Irradiation Examination of Capsules P13R and P13S", GA-A13827, October 8, 1976 2L238 274

TABLE 1 EMPIRICALLY DETERMINED PRESSURE VESSEL FAILURE CRITERIA FOR TRISO COATED UC2 FUEL TESTED IN I'tRADIATION CAPSULES P13R & P13S 8 * "*

Calculated (*

Irradiation Conditions Failure Criterion Based on Fission Particle Batch Temper- Fluence r-85m as elease Data Retrieval Irradiation ature Burnup (10 25 n/m2 ) Fission Gas Number Capsule (C') (FIMA%) E>29fJ IITGR Release MPa psi 6151-00-010 P13R 1035 74 12.1 4.0 120.6 17500(b) 6151-00-035 P13R 1075 74 12.0 0.4 347.9 50500 6151-04-015 P13R 993 74 11.7 0.6 113.7 16500 6151-01-015 P13R 1070 73 11.5 1.4 330.7 48000 6151-09-015 .P13R 1070 73 11.1 1.8 120.6 17500 6151-09-025 P13R 1005 73 11.0 0.12 503.0 73000 6151-00-035 P13S 1015 71 10.7 0.3 351.4 51000 6151-03-015 P13S 987 73 11.8 0.2 223.9 32500(b)

,4161-01-021 P13S 1020 73 11.7 26.9 92.7 13500(D) y 6151-08-015 P13S 1020 73 11.7 0 >458.2- >66500 N

u 6151-02-025 P13S 960 73 11.6 4.3 134.4 19500 CD (a) Calculated based on a Monte-Carlo type stress calculation and observed fuel failure (methodology described in N response to Lead Item VIII.

( 71 (b) Average of two separate calculations, s.

5 RESPONSE TO NRC QUESTION 231.3 ON UC2 SAR QUESTION 231.3 The " expected" failure fractions of 0.002 and 0.005 that are given in the response to Lead Item VIII for TRISO coated UC2 and (Th/U)C2, respectively, are said to be based on nominal fuel properties, ex-pected property distributions, peak irradiation exposure conditions and empirically determined failure criteria for each fuel type.

This statement requires substantial elaboration with regard to the details of the assumptions and numerical values used for the cal-culation. For example, it would bs instructive to learn how the TRISO UC2 particles with approximately 75% FIMA at end-of-life are expected to have a lower pressure vessel failure rate than the TRISO coated (Th/U)C2 particles which have a maximum burnup of about 20% FIMA. That is, can you show what design features in the TRISO UC2 particle compensate for the effects of higher burnup (and higher internal fission gas pressure)?

RESPONSE

The expected failure fractions of 0.2% for UC2 and 0.5% for (Th/U)C2 were calculated using the input values listed in Table 1. These input values were used in a Monte-Carlo calculational routine for determining the sic stress distributions for UC2 and (Th/U)C2 fuel. Refer to the original response to Lead Item VIII for a detailed description of the methodology used. A com-parative evaluation of the input properties listed in this table shows that the UC2 particle design has a smaller kernel diameter in combination with a larger buffer thickness, sic thickness, and sic apparent failure stress as compared to the (Th/U)C2 design. These design features taken collectively compensate for the higher burnup in UC2 and result in the improved UC2 pres-sure vessel performance compared to (Th/U)C2' 2238 276

TABLE 1 INPUT VALUES USED TO DETER!!INE EXPECTED PRESSURE VESSEL FAILURE LEVELS FOR TRISO COATED (Th/U)C2 AND UC2 TRISO Coated Purticle Properties Empirically Kernel Buffer IPyC sic OPyC

" ate Diameter (pm) Thickness (pm) Thickness (pm) Thickness (pm) Thickness (pm) p 8

Particle Standard Standard Standard Standard Standard Stress Type Mean Deviation Mean Deviation Mean Deviation Mean Deviation Mean Deviation (psi)

(Th/U)C2 257 16.0 52 7.7 24 3.9 25 2.9 41 5.7 -2800 UC2 195 24.0 110 19.5 35 5.7 35 4.0 40 5.60 33500

(* sic stress distributions for different particle types are based on the following peak exposure conditions:

8 x 1025 n/m2 (E>29fJ) HTGR, 1250*C, 20% FIMA in (Th/U)C2, and 75% FIMA in UC2 N

u CD N

N N

e

7 RESPONSE TO NRC QUESTION 231.4 ON UC2 SAR QUESTION 231.4 Figure 3 in the response to Lead Item VII shows a comparison of Kr-85m R/B (rate of release / rate of birth) data obtained from irradiation test results with predicted Kr-85m R/Bs. The text then says that "....the observed Kr-85m R/B values are substantially less than predicted, which i= plies that... performance is equal to or better than the current (Th/U)C2 FSV fuel" (emphasis added). Yet no comparison is actually made with either predicted or observed Kr-85m R/B values on (Th/U)C2 particles. Please either show the comparison or delete the quoted statement from the text.

RESPONSE

The complete quoted text says that "....the observed Kr-85m R/B values are substantially less than predicted, which implies that the in-pile failure of UC, is less than that predicted by the TRISO coated particle stress model and that performance is equal to or better than the current (Th/U)C2 FSV fuel" (emphasis added) . The comparative statement regarding performance emphasized in Question 231.4 is based, in part, upon the observation emphasized above. As noted in the paragraph of the response to Lead Item VIII previous to that quoted above, TRISO coated UC2 failure fractions calculated in design analyses are less than those of (Th/U)C2 particles. Since the results for SSL-2 indicate that in-pile UC2 failure is less than that predicted by the stress model, the conclusion emphasized in Question 231.4 is made.

The statement that performance of TRISO coated UC2 fuel is equal to or better than the current (Th/U)C2 fuel is also supported by a comparative evaluation of in-pile end of life (E0L) fission gas release measurements in capsules F-30, GF-4, and SSL-2. Table 1 lists the EOL Kr-85m rate of release /

rate of birth measurements for these capsules along with the irradiation ex-posure conditions and supporting references. The table shows that TRISO coated UC2 fuel subjected to EOL peak exposure conditions in GF-4 and SSL-2 exhibits a lower Kr-85m R/B compared to the (Th/U)C2 fuel tested in capsule F-30. This comparison implies that the UC2 fuel performance is equal to or better than that of the current FSV (Th/U)C2 fuel.

2238 278

TABLE 1 COMPARATIVE EVALUATION OF IN-PILE EOL Kr-85m R/B MEASUREMENTS FOR CAPSULES F-30, GF-4, AND SSL-2 Irradiation Conditions EOL Kr-85m Irradiation Number of Fissile R/B

"""C" Capsule Fuel Type llTCR Fuel Temperature Particle Burnup Measurement Test Fissile / Fertile Rods Tested (OC) 10 n/m (E>29fJ)llTCR (% FIMA) (10-5)

F-30(a) (Tn/U)C2/ThC2 13 825 - 1090 3.7 - 9.4 11.9 - 20.1 1.5 - 6.0 (TRISO/ TRIS 0)

GF-4 cell 2 UC2 /Th0 2 3 940 % 1090 10 - 10.8 75.5 1.3(b)

(TRISO/ TRIS 0)

SSL-2 UC2 /Th02 27 1100 % 1250 4.3 - 9.1 72 0.68(C)

(TRISO/ BIS 0)

(*} Proof test for FSV fuel: Scott, C. B. and Harmon, D. P. ," Post Irradiation Examination of Capsule F-30,"

CA-A13208, April 1, 1975.

Pointud, M. L., " Irradiation de Combustible - Capsule GF-4, Degagement des Caz de Fission Pendant L'Irradiatica,"

Dossier No. 4, CEA Report DMG No. DR20/77, May 25, 1977.

( De B eaucou r t , Ph., and Maubac, Q. F., " Irradiation SSL-2 Resultats des Mesures de Produits de Fission Gazeux Radioactifs," CEA Report, EMT/76-182, October 13, 1976.

N U

CD N

N

• O co

9 RESPONSE TO NRC QUESTION 231.5 ON UC2 SAR QUESTION 231.5 The response to Lead Item VIII refers to tests conducted in cell 2 of capJule GF-4 without explanation of what capsule GF-4 was or where the tests were conducted. Please provide this information and list a reference where further information can be obtained.

RESPONSE

Irradiation capsule GF-4 was the fourth capsule.in a series of irradiation tests conducted under a cooperative agreement between GA and Commissariat al'Energie Atomique (CEA, French Atomic Energy Commission).

GF-4 was irradiated in the Silee' reactor in Grenoble, France. The capsule contained three separate cells designated for testing HTGR fuel rod and loose particle performance, with each cell designed for independent fission gas release (Kr-85m R/B) measurements. The irradiation test began in April, 1975, with the primary objective of evaluating the performance of reference type TRISO coated UC2 fuel in combination with TRISO coated Th0 2 (located in cell 2 of GF-4). The fuel was tested in fuel rods and as loose particles over the following conditions: 1050*C to 1100*C, 6 to 10.4 x 1025 n/m2 (E>29fJ) HTGR, and peak burnups of 4.0% FIMA for Th02 and 75.3% FEMA for UC 2 . A detailed description of the test objectives along with a preirradiation characterization of fuel properties is presented in Ref. 1. The post irradiation examination of this capsule was conducted in the Hot Cell facilities at Grenoble, France, and has been partially documented in Refs. 2 and 3.

REFERENCES

1. Kovacs, W. J. and Harmon, D. P., "Preirradiation Report: CA Fuel Materials for GF-4", GA-A13475, September 1, 1975.

2. Pointud, M. L. , " Irradiation de Combustible - Capsule GF.4, Degagement des Gaz defission Pendant L' irradiation", Dossier No. 4, CEA Report Dmg No. DR 20/77, >by 25, 1977.

3. Blanchard, R. , et al, " Experience GF.4 Examens Metallographique et Micro Radiographique du Combustible", Dossier No. Sc Partie), CEA Report DMG No. DR 23/77, June 9,1977.

2238 280

10 RESPONSE TO NRC QUESTION 231.6 ON UC2 SAR QUESTION 231.6 Please provide more background (i.e. previous irradiation history) for the CHST samples described in the response to Lead Item VII.

For example, list the reactor and time in reactor, nominal and peak operating temperatures, etc.

RESPONSE

Additional requested background information on the irradiation con-ditions of fissile CHST samples is provided in Table 1.

2238 281 0

. . 11 TABLE 1 BACKGROUND INFORMATION - FISSILE CHST SAMPLES limein Data Kernel Irradiation Temperature (OC) Reactor Retrieval No. Type Experiment Reactor Average reak (days) 6151-18-015 UC HB-2 GETR 105b") ND I) 112 2

6151-17-016 UC HB-5 GETR ND(b) 700(*) 112 2

4161-01-030 UC FTE-14 Peach Bottom 1095 1250 317 2

CU6A-6328 (c) (Th/U)C 2 F-30 GETR 1243 1623 269

(*} Design values

( }Not determined (c)FSV proof test fuel, 18.2%FIMA, 9.1 x 1025g 2 (E>29fJ HTGR 2238 282

. . 12 RESPONSE TO NRC QUESTION 231.7 on UC2 SAR Question 231.7 Manufacturing process variables for both the kernal and the coatings of the fuel particles are known to affect materials properties and performance in retention of fission products. While TRISO coated UC2 fissile particles have been demonstrated to be effective in re-taining both gaseous and metallic fission products under reactor conditions, how can this performance be guaranteed if the process variables used the feel particle manufacture are not to be included in their licensing basis?

Response

Manufacturing process variables are controlled by fuel manufacturing speci-fications. These specifications ensure a final product with kernel and coating characteristics which have been s1.cwn by irradiation testing to re-sult in satisfactory coated fuel particle performance. Quality control tests are specified at several points during fuel ranufacture to confirm that critical kernel and coating property requirements are met. Where necessary, process specifications are imposed to ensure a uniform product.

Changes to fuel specifications can be made only under Quality Assurance procedures which comply with the requirements of 10CFR50, Appendix B. These procedures are subject to periodic audit by both the customer and the NRC and are not changed without documented justification.

22'38 283