ML20137S832

Review of Fuel Rod Metallography on Fort St Vrain Fuel Element Sn 1-2415

ML20137S832
Person / Time
Site:	Fort Saint Vrain
Issue date:	07/31/1984
From:	Ranken W, Reiswig R SANDIA NATIONAL LABORATORIES
To:	John Miller Office of Nuclear Reactor Regulation
Shared Package
ML20136F700	List: ML20136F694 ML20137F771 ML20137S832 ML20137S873
References
CON-FIN-A-7258 NUDOCS 8509300516
Download: ML20137S832 (8)

Text

.. _.

s Review of Fuel Rod Metallography on Fort St. Vrain ~

Fuel Element SN 1-24} 5 .

NRC Fin No. A.725g July 31, 1984

- Los Alamos National Laboratory William A. Ranken, 0-13 Robert D. ReisWi9, MST-6

=__

Responsible NRC Individual and Division J. Mi11er/DRB3 Prepared for the U.S. Nuclear Regulatory Commission

,, Washington, D. C. 20555 6

i

/

8509300516 850820 PDR ADOCK 05000267 PM S

l

'. F; 1

. j DISCLAIMER This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, or any of their employees, makes any warranty, expressed or implied, or assumes any legal liability or responsibility for any third party's use, of any infortnation, apparatus, product

~

or process disclosed in this report or represents that its use by such third party would not infringe privately owned i

~

rights.

i

/

-2 l

l l

- . _ _ . _ _ _ . ~ . __ . _ . _ _ _ _ , . _ _ , , - . . , _ . _ _ _ _ . ~ , . .

. + .

ABSTRACT This report reviews the documentation submitted to the NRC by the Public Service Co. of Colorado, on the metallographic examination of a section of fuel stack 308, as found in cracked fuel element SN 1-2415 removed during the third refueling of the Fort St. Vrain HTGR. Los Alamos essentially concurs with the PSCo conclusions, in that the performance of the fuel during irradia-tion was acceptable, that no evidence of fuel and graphite element interaction has been found, and that no kernel migration was observed. The report provides comments to the NRC as to whether the licensee's technical information is cor-rect and consistent with the information gained by Los Alamos in previous fuel

. studies, and in their review of graphite slabs from Segment 2 cracked fuel element SN 1-2415.

FORWARD

,..'~ This technical evaluation report is part of the technical assistance pro-gram, " Review of Selected Fort St. Vrain Issues" FIN No. A-7258, and is sup-plied to the U.S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation, by Los Alamos National Laboratory.

a f

0 Review of Fuel Rod Meta 11ography on Fort St. Vrain Fuel Element SN 1-2415 Technical Evaluation Report Introduction During the Segment 2 reload of the Fort St. Vrain High Temperature Gas-Cooled Reactor, two colinear graphite fuel elements were found cracked axially down the B-face, and radially inward two and three webs deep. This report reviews the metallographic examination performed by GA Technologies on a fuel rod from a stack adjacent to the crack region of the more severely cracked fuel element.

Backcround The GA Technologies report entitled "Heta11ographic Examination of a Fuel Rod from Segment 2 FSV Fuel Element 1-2415" (Ref.1) describes the metallo-graphic examination of a fuel rod from the stack in fuel hole 308 of Fort St.

Vrain fuel element SN 1-2415, which developed a crack during irradiation.

Stack 308 was centered in the second row of fuel and coolant channels on the B-face side of the hexagonal graphite element, and was adjacent to the crack

..;(which actually passed through fuel hole 307). The metallographic examination was conducted to see if any evidence of unusual themal effects could be seen that would help in ascertaining the cause of the crack, and was perfomed in con' junction with a more general postirradiation examination on the fuel ele-ment.

l Each graphite fuel element is a hexagonal right prism, with 210 fuel holes and 108 coolant channels drilled parallel to one another in a triangular array with 0.74 inch pitch spacing, resulting in a basic ratio of two fuel holes for each coolant channel. Each fuel hole is drilled from the top surface of the element, to within about 0.3 inches of the bottom. Bonded rods containing coated fuel particles, each about 1/2 inch long, are stacked within the hole, and the hole is capped with a graphite plug.

The fuel is in the form of carbide particles, coated with a highly reten-tive coating, and bonded into fuel rods with a coal tar pitch binder2,

.  :. ~

i .

j - :

I The fuel rods contain a homogeneous mixture of fissile particles containing both uranium 235 (93.15% enriched) and. thorium, and fertile particles con-taining only thorium. As schematically shown in Figure 1, the' fuel particles are coated with a four-layer TRISO coating, sequentially consisting of an inner layer of porous pyrolytic carbon called the buffer layer; a high density iso-tropic pyrocarbon (IPyC layer); a thin layer of sic, which is highly impervious to metallic fission products; and an outemost layer of strong high density isotropic pyrocarbon (OPyC layer). Important particle parameters are repro-duced from the FSAR in Table 1.

4 Discussion l Our general evaluation of the metallographic examination of this fuel rod, l and the coated fission and fertile particles it contained, is that it appears to have been done competently. The main criticism relates to the selection of the examined fuel rod from a' stack (308) adjacent to the stack in fuel hole 307, through which the crack in the fuel element passed. Presumably this se-1ection was made because the thermal history of stack 307 might have been dis-torted by the crack and hence would be less representative than stack 308 of events that precipitated the cracking. Nevertheless, it would have been more infomative if the fuel rods from 307-had been measured for dimensional changes

,;and one of them chosen for metallographic examination in addition to the rod actually selected. Additional examination would have helped to rule out fuel nonuniformity effects as a cause for crack formation.

' The performance of the fuel was found acceptable in Ref 1. Specific ob-servations and conclusions from the document are summarized as follows:

i 1. Fuel rod 13 from stack 306 was in good condition although minor cracking in the matrix end caps and some debonding of particles from the rod sur-face were observed. -

2. There was no fuel rod-block interaction as evidenced by the visual exami-

nation of the rods, by the small pushout force of the stack, and by the small amount of debris collected from the emptied block.

3 .' The measured macroporosity for rod 13 was 17.5% which was within the (14-29%) range of macroporosities observed for fuel rods from capsule F-30.

i

4. A total of 231 fissile and 184 fertile particles from rod 13 were examined.

For the (Th,U)C2 and ThC2 particles, respectively, the OPyC coating

, failure was 0.4% and 7.6%, and the sic coating failure was 0.91 and 3.8%.

i However, these coating failure rates should be regarded as the upper limits since coating failures can be caused during manufacture and during the

grinding and polishing procedure, as well as during irradiation. ,

l 5. The chemical behavior of the particles was acceptable. There was no kernel j migration observed. However, there was evidence of fission product inter-action with the sic coating. 3.9% of the fissile particles and 3.2% of the fertile particles showed fission product--sic interaction with a pene-tration depth of % 5 un, which was higher than the expected value of

< 1 un. The increased penetration may have been caused by fuel dis-4 persion resulting from chlorine trapped in the buffer layer during the sic coating and/or by higher operating temperatures than the average calculated C~ for the element.

6. Fuel dispersion, which is attributed to chlorine diffusing through a low density IPyC into the buffer layer during the sic coating process, and j IPyC debonding were observed in some of the TRISO (Th,U)C and ThC 2 2 particles. . Fuel dispersion and IPyC debonding did not detrimentally affect the perfonnance of the particles.

l  ; In general the conclusions reached in the report are reasonable, but there

[ is a need for caution. Visual examination, the ease with which the rods could

be pushed out of the block, and the porosity measurements all confinn normal behavior for fuel stack 308. However, the fuel dispersion results, the amount of fission product interaction with the sic layers, the inner pyrolytic carbon coating debonding (shrinkage of the pyrocarbon), the observed coating failure rates, and the lack of data on total coating failure (because of itydrolysis of the metallographic specimen mount') all raise questions.

As stated in the report, one interpretation of the depth of fission product interaction (~ 5 un versus < 1 um expected) is that the operating tem-l perature was higher than the average calculated for this fuel element. Higher i l than expected operating temperature could presumably also explain the greater i l than usual amount of fuel dispersion observed in the sectioned fuel rod. Tem- )

i perature may also have been a factor in the increased irradiation - induced )

shrinkage of the inner pyrolytic coatings that led to debonding.

+

As to questions of whether broken coatings were caused by metallographic l

j sample preparation or by irradiation ,it would have been helpful if the GA

! investigators had cited past experience on sample preparation and used this to make a determination of the most probable cause of the observed cracking.

The lack of total-coating failure data is indeed unfortunate. It is not known whether the number of failures was a very small fraction of' total par-ticles or a significant added population to that of particles with just one of l'

the three containment layers cracked.

Conclusions Despite the uncertainties in interpretation of some of the fuel rod metal-lographic data, there is little evidence that the thermal history was abnormal. '

. The observed departures from the nom, particularly in fuel dispersion and I.. .. fission product interaction, may indeed have been caused by variations in

~

coated particle fabrication as suggested by GA. If the particles were uniform in-their properties, o+would expect a temperature effect to show up in all I of them instead of just'a fraction. On the other hand, the variation of par-l ticle properties could mask a temperature effect, because one would not be j sure, for instance, how much of the increased fuel dispersion was caused by j chlorine and how much by a higher than expected temperature. It appears that j ,, fuel rod meta 11ography is not, under the current conditions, a very sensitive

~ determinant of fuel and thermal history. In any case the GA promise to monitor the fuel dispersion and fission product - sic interaction in future FSV fuel

surveillance should be seized upon. If any future work is done on fuel element i 1-2415, it should include dimensional checks and metallography on the fuel j

rods from stack 307.

) References ,

1

1. " Meta 11ographic Examination of a Fuel Rod from Segment 2 FSV Fuel

] ,

Element 1-2415", GA Technologies Doc. No. 906968, 1983.

t 1

2. Fort St. Vrain Nuclear Generating Station, " Updated Final Safety i Analysis Report", Public Service Co. of Colorado.

i i i 4

i

e e

• e s MODEL TRISO COATED PARTICLE gvi*s'M DUTE R ISOTROPlc f

=.s%.N 8'

PYROLYTIC CARBON ,

,-, ' 'hg

[ d , #h f

E ,e ? . ' . \ ',

3

~

\

'4 slLICON C ARBIDE i / \.

',' i rVEL PARTICLE b f* rg $je SARRIER COATING jQ h 1. .

( '.l' . _121LL , g h',.* '2 i'l fh ,I l k

k. 6'd ,'t INNE R isOTROPlc PYROLYTIC CARSON 5 Ut L PARTICL F

% 4 f */

f

,'h ' gi) ,'a ',

\

N,,..

'g%

3

,s y' g

j ' surrER PYROLYTIC CARBON k.' .

. /../

Figure 1. Model of TRISO Coated Fuel Particle Table 1. Coated Fuel Particle Parameters Parameter Fissile Fertile Th:U 3.6:1.4.25:1 All Th Kernel composition (Th:U)C, ThC, Small large Small targe

. .. ' Average fuel kernel diameter, micron 140 225 375 525 Average coating thickness:

Buffer carbon layer, micron 50 50 50 50 Isotropic carbon layer, micron 20 20 20 20 sic layer, micron 20 20 20 20 Isotropic carbon .

layer, micron 30 40 40 ' 50 Average coated fuel diameter, micron 380 485 635 805 inches (045) (.019) (.025) (.032)

- - - =_ . . - - , , - , -- - - . - ,