ML20058K969

Safety Evaluation Supporting Amend 20 to License R-103

ML20058K969
Person / Time
Site:	University of Missouri-Columbia
Issue date:	08/01/1990
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20058K963	List: ML20058K959 ML20058K965 ML20058K969
References
NUDOCS 9008030195
Download: ML20058K969 (20)

Text

. - . _ . - . _ _ _ . _- - .- - _ -

I

'ge.asg j k UNITED STATES

[f 3sJ NUCLEAR REGULATORY COMMISSION l 5 W ASHING T ON, D. C. 20665

%,, l

.... 1 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION SUPPORTING AMENDMENT NO. 20 TO j FACILITY LICENSE NO. R-103 UNIVERSITY OF MISSOURI AT COLUMBIA DOCKET NO. 50 186 I 1

1.0 INTRODUCTION

l By letter dated September 12, 1986, as supplemented on September 11,1987, March 11,1988, July 22,1988, February 1,1989, April 25,1990, and May 17, 1990, the University of Missouri at Columbia (UMC) requested changes to j Appendix A of Facility License No. R-103. Technical Specifications for the UniversityofMissouriResearchReactor(MURR). The changes would allow the )

4 use of extended life aluminide fuel (ELAF) in the reactor.

R. E. Carter, Project Manager, Office of Nuclear Reactor Regulation, NRC, i initiated this review. A. Adams, Jr., a Project Manager in the same office continued and completed the task. R. G. Ambrosek of the Idaho National Engineering Laboratory (INEL) conducted the principal technical review under contract to the NRC.

2.0 EVALUATION 2.1 INTRODUC110N The requested license change will modify the technical specifications to allow '

using a new fuel element design (ELAF or 1.270 kg fuel element) which would significantly reduce the fuel cycle cost and reduce the amount of U-235 needed per MWD of energy produced.

I

• 008030195 90oeot l

(, DR ADOCK 03000166 eoc

l 2

Because the ELAF has the same dimensions as the current fuel element (0.775 kg fue' element), no geometry modifications to the core structure are required.

The proposed fuel element changes are graded fuel loading per plate instead of a constant fuel loading and boron carbide included as a burnable poison in

~

selected plates. These changes are proposed to decrease nuclear peaking and allowanincreaseinburnup(fissions /cc)whichresultsinlongercycletimes before fuel replacement.

I The safe operation of nuclear reactors is dependent on maintaining the l integrity of the clad on the fuei mterial. There are two major ways in which clad failure can occur - failure due to corrosion and failure of the heat transfer process resulting in high clad temperatures.

I 5

Studies on aluminum clad fue1 have shown that corrosion of the cladding is l dependent on the plate-coolant interface temperature,.the heat flux magnitude, the element operating time, and the reactor coolant chemistry. ELAF will operate for a longer interval, and at different power densities. This will result in changes in corrosion and the parameters influencing corrosion must be addressed.

The flow channels, mass flow rate, element geometry, etc., are not being changed. The heat transfer process will only be changed due to changes in power density, and oxide thickness. The power density is influenced by peaking factors, fuel density, neutron spectra, rod worths and position, experiment loadings, and placement of neutron absorbing materials. The resistance to heat transfer is influenced by the thickness of the oxide layer.

The rhte of heat transfer can change reactor behavior during transient opera-

. tion. The review was conducted to verify that these parameters which influence the clad integrity were examined and that the evaluations demonstrate the new fuel elements can be operated within acceptable limits.

p.' .

h .-

2.2 EVALUATION OF EXTENDED FUEL LIFETIME The proposed operating conditions for the new fuel element would require a

-burnup limit of 2.3 x 10 21 fissions per cubic centimeter. The license limit for the current fuel at MURR is 150 MWD which is between 1.7 x 10 21 and 1.8 x 10 21 fissions per cubic centimeter. To achieve this increased burnup, the in-reactor time will essentially double. The parameters which influence fuelfailureasaresultoftheincreasedburnupandirradiationtimeare(1) swellingandblistering,(2)pittingcorrosion,and(3)oxidefilmformation.

2.2.1 Swelling and Blistering Fuel plate swelling occurs as a result of irradiation and fuel burnup. Blis-tering ecturs when fission gases agglemerate in interstitial voids and the gas pressure becomes high enough to cause clad deformation. UMC addresses these ,

par 6ineters and concludes that these failure mechanisms are not expected to occur for the proposed fuel elements within burnup limits as requested. The fuel plate swelling and blistering acceptability is based on test data in Reference 4.

Examination of this data supports the conclusion by UMC that the increased burnup levels will not result in excessive swelling or blistering. Thus fuel plate f ailure by these mechanisms is not significantly more likely than in the

. current fuel at the same reactor oower level and coolant conditions.

2.2.2 Corro g n Corrosion can result in aluminum clad failure in two ways: pitting and oxide film formation. Oxide film forms as a result of essentially uniform corrosion.

Pitting occurs when the corrosion rate is accelerated in a local area. If the oxide film exceeds a thickness which results in spallation, the spalled zones have been shown to have localized attack in the form of subsurface voids pene-trating into the 6061 aluminum alloy5 ,

~ - _ . . . . . . . .. .

[' 4 The corrosion of aluminum in water is dependent on the operating time, the chemistry control of the water, the temperature of the fuel plate surface, and the heat fluxb . Since operating time is a factor in the amount of corrosion, the consequences needed to be evaluated.

The licensee's discussion on fuel plate failure due to pitting notes that pitting failures are not catastrophic and any failure of the clad can be detected by the primary coolant radiation monitoring system. _

During development of the ELAF, irradiation of fuel plates was conducted "

intheAdvancedTestReactor(ATR). As reported in Reference 4 an equation was derived which correlated the maximum pit corrosion rate. Howevar, it is noted that most pits do not propagate at the maximum rate. The average of thetendeepestpitsintheELAF1rradiationprogramwas4.3 mil (0.11mm).

One of the pits would have penetrated the clad if it had not been in the side plate area. _

MURR has experienced no fuel element failures due to pitting corrosion with the current fuel. Based on the MURR, ATR, and ELAF irradiation experience it ~

is concluded that pitting is not a major concern. Technical Specification 3.8b does not allow the reactor to operate with fuel in which anomalies have been detected. Therefore, failed fuel will be removed upon discovery. The licensee's evaluations project the bounding primary coolant activity level as 5 0.131 uti/ml if a clad failure due to pitting was to occur. After removal of the failed fuel element, this activity level could be reduced to the maximum _.

allowable level for reactor operation as given in Technical Specification 3.9c by the coolant demineralizer system in approximately three hours. It is noted that the consequences of pitting erosion are bounded by the Maximum Hypo-tnetical AccM:nt (MHA).

Based on the review and MURR and ATR operating expe'ience, the staff agrees that clad failures due to pitting are not a major cancern and that consequences

,a

. to the pubite are bounded by the MHA. The increased irradiation time will

~

result in deeper pits, and the probability for clad failure due to pitting I will be increased as a result of longer operating times.

5

The study on aluminum film corrosion reports that the oxide film rate of formation is dependent on the coolant pH, the operating time, the surface temperature, and the heat flux magnitude. The report concludes that corrosion will occur at a rate 2.7 times greater at a pH of 5.7 to 7.0 than at a pH of 5.0. Results also indicate that the rate of oxide formation for heat fluxes in I the range of 0.5 x 106 Btu /h-ft2is about half that observed at heat fluxes of 1 to 2 x 106Btu /h-ft with 2

(

other conditions being the same. Projected oxide thickness can be determined by using the Griess correlation.

Oxide thickness measurements were performed and data was submitted on 17 elements at MURR. The data correlation predicts a multiplier on the Griess correlation of 0.7. The maximum nominal predicted oxide thickness for the ELAF with its operating conditions is 0.854 mils (.0217 mm), with a three standard deviation value of 1.75 mils (.0445 m).

Observations at the Idaho National Engineering Laboratory (INEL) have also shown that the Griess correlation overpredicts the oxide thickness for the ATR operating conditions. Correlation of the ATR data indicates a multiplier of 0.5 for the maximum nominal prediction. The oxide deposition has been shown to be dependent on heat flux and primary coolant pH. The MURR operates with a higher pH than the ATR but with a lower heat flux.. Thus a direct compariscn is not possible. The MURR derived multiplier does appear to be consistent with ATR experience.

The spontaneous spallation oxide thickness is not predicted for the maximum

~

nominal or three standard deviation values based on the measured data base.

Therefore, the staff concludes the concerns with spallation were not relevant.

h i

E

- 2.2.3 Conclusion Based on the UMC submitted data and ELAF irradiations, it is concluded that fuel plate swelling, blistering arid corrosion will not preclude operation of the ELAT to burnup levels of'2.3 x 10 21 fissions per cubic centimeter. pitting failures would not result in excessive primary coolant cleanup times or exposures to the public in excess of the MHA.

2.3 EVALUATION OF NUCLEAR CHARACTERISTICS The MURR safety limits derived in 1973 are based on a nuclear peaking factor of 3.678 which corresponds to a peak power density of 1114 watts per cubic centimeter of core region. This peaking factor was determined using a two-dimensional neutron diffusion code that does not have fuel' depletion capability. To compensate for fuel depletion effects an azimuthal peaking factor between elements of 1.112 was used.

The power densities for the various combinations of ELAF and current fuel elements were calculated by UMC with the BOLD YENTURE-IV and AMPX-II diffusion theory code systems. These code systems can perform three dimensional neu-tronics analysis, including feel depletion.

The code systems and MURR computer models were benchmarked against the results ofadestructiveanalysiscompletedona775gramfuelelement(775-F3)with 3

82.5 MWD power history ,

2.3.1 BOLD VENTURE-IV Code System Benchmark This section evaluated the code systems benchmarking and the physics results for the new element irradiation in various configurations.

The BOLD VENTURE-IV code system was benchmarked by comparing calculated burnup with burnup neasurements obtained by post-irradiation examination of the 775-F3 fuel element. The measurements were performed longitudinally along the fuel element centerline.

l 1

1

\ .

=

7 The BOLD VENTURE-1V calculation showed good agreement with the measured data.

In general, the calculated burnups are higher than the measured data. This indicates that the flux predictions are higher and predicted power densities should be higher, thus conservative. The good agreement lends confidence to the accuracy and applicability of the calculations performed using the BOLD VENTURE-1V code system.

2.3.2 Analytical Evaluations for Upgrade Fuel The UMC submittals ar.d Reference 3 present calculations performed for the ELAF. These analyses have been reviewed and are discussed in the following subsections.

2.3.2.1 Total Control Blade Worth Reference 3 discusses the methodology used to derive effective diffusion theory constants for the reflector co . trol blades. The discussion notes that adjustments were made to the control blade " black" absorber constants for the group three and group four neutron energy levels to force agreement with a measured reactivity worth with C.775 kg fuel elements. This methodology worked for cores configured with the 0.775 kg fuel elements. The ELAF is going to have a harder neutron spectrum :aused by the higher U-235 loading and the burnable poison loaded in select inner and outer plates. As a result of these changes, the leakage neutron energy spectrum and the amount of neutron leakage from the core should change. The small distance (1.68 cm) from the core edge to the poison material should make the control blade worth sensitive to spectral and leakage changes. If these effects upset the

" balance" between the fraction of neutrons absorbed in groups three and four in the poison region, then the cross section set derived by forcing agreement with a measured worth for the present core might not be adequate for the ELAF core ' analysis . UMC was asked to demonstrate that the cross sections for the current core are sufficiently valid for the proposed cores and that control blade worths and reactivity conditions will be consistent with the Technical Specifications.

e ,

,4-(

8-Analyses reported by UMC relate that the total rod worth for a core of 1.270 kg elements could be 10% less than the worth in the present core. This value is based on comparative calculations with microscopit cross sections for boron-10 using XSRDNPM and for control rod worths with four group neutron flux energy levels using BOLD VENTURE.

Technical Specification 3.1 has three requirements that apply directly or indirectly to reactivity of the control blades; maximum reactivity insertion rate, shutdown margin, and excess reactivity. The decreased rod worth will

-result in a smaller reactivity insertion rate wH ch is conservative. Excess reactivity will continue to be demonstrated to meet Technical Specification limits for each core. Compliance with the Technical Specifications for shutdown margin and excess reactivity will be demonstrated by measuring control rod worths and cold clean critical rod height.

2.3.2.2 Power Peaking The BOLD VENTURE-1V code system was used to calculate nuclear power peaking factors for.various configurations of fresh and depleted 0.775 kg and 1.270 kg fuel elements. UMC notes that for a complete core of 1.270 kg fuel elements the peak power density occurs in plate 4 with the control blade in and plate 23 with the control blade out. These calculations assumed only water was in the flux trap.

The licensee states that the highest nuclear power density occurs in a core of four fresh 0.775 kg fuel elements and four depleted 1.270 kg fuel elements. This peak occurred in plate 1 of the fresh 0.775 kg elements and had a nuclear power peaking factor of 3.225. The evaluations show that the hot spot occurs on Plate 7 for the l'.270 kg elements. The azimuthal peaking factors for the 1.270 kg element are larger than the 1.04 originally assumed by the licensee on the hot plates. However, the highest calculated power density for the 1.270 kg element is lower than that on Plate 1 of the 0.775 kg element by a f actor of 1.45.

4 -

I $-

9 Analyses for dependency on flux trap loading by the licensee indicate that there are small changes in fuel elenent power densities. The power increase is approximately 3% with location of maximum power dependent on control rod position and reactivity sign. The hot spot power densities are all below those for plate 1 in the current element.

- The cross section sensitivities to temperature and burnup are also analyzed by the licensee. The results show that the cross sections are not sensitive to changes in fuel material temperature and boron and fuel depletion for the currently proposed MURR operating ranges, it is noted that the benchmarking and power peaking analyses were performed with cross sections representative of a cold, clean core.

Based on the data presented, the staff concurs that the reported results are applicable and cross section corrections for temperature and fuel boron-depletion are not required for the proposed operating conditions.

2.3.2.3 Core Temperature and Voiding Reactivity Coefficients A reactor core loaded with ELAF is expected to have a different neutron spectra than the present core. Because negative reactivity coefficients due to reactor core temperature change and core voiding are dependent on the neutron spectra, an analysis of the expected changes and consequences in regard to technical specification limits for these parameters was completed.

The core void reactivity and temperature reactivity coefficients were analyzed for the present core and the proposed core. The void reactivity coefficient was calculated using an R-Z model with BOLD VENTURE and the AMPX four-group uross section set. The results show that the void reactivity coefficient is approximately 3% lower for the proposed core, but is within the Technical Specification limits.

, s.

The core temperature coefficient of reactivity was calculated using the seven group EPRI-CELL cross section library and BOLD VENTURE. The results show that the temperature coefficient is predicted to be -5.64 x 10-5 6 k/'F for a complete ELAF core which is less than the Technical Specification value of -6.00 x 10 -5 a k/'F.

2.3.3. Conclusions Based on this review, the staff concurs that the control rod worths will be less in a core composed of only ELAF by approximately 10%.

Calculations have demonstrated that cross sections have a small dependency on burnup and temperature and thus " cold, clean core" cross sections are satisf actory for power peaking calculations.

The void reactivity coefficient for the ELAF core is approximately 3% less than the present core, but is withir Technical Specification limits. The calculated core temperature reactivity coefficient would be reduced by approxi-mately 19% which is less than the current Technical Specification limit.

Therefore, UMC could not load any complete ELAF cores withou?. further analysis and possible amendment of the Technical Specifications.

However, step wise introduction of ELAF fuel into the core to create mixed ELAF/ Standard (.775 kg) cores is acceptable as long as the licensee demonstrates compliance with the existing Technical Specification limits for each core configuration.

2.4 ENGINEERING FACTORS EVALUATION

-The safety limits are derived using hot channel factors. These factors are composed of nuclear peaking factors, engineering hot channel factors and flow-related factors. The nuclear peaking factors were evaluated in Section 2.3. The engineering and flow-related factors will be evaluated in this section.

l

L .,

. 2.4.1 Enthelpy Rise Evaluation The nuclear peaking factors, engineering hot channel fr. tors for enthalpy rise and flow-related factors are combined for an en',nalpy hot channel factor. The engineering hot channel factors considered are fuel content variation and fuel thickness / width variation.

The initial analyses for the upgrade core assumed the factors used for the-existing fuel are applicable. There is no approved fuel specification for the upgrade fuel at the present time. The three charges VMC deems necessary in the present fuel specification for the new fuel tiement are higher UA1 x loading in selected plates, burnable poison in selected plates, and the use of UA12 phase UA1 x powder. Adherence to the cittrent specification tolerances will not assure that the fuel content variation f actor and fuel thickness / width variation factor can continue to be used for the new fuel. The licensee will develop ELAF full specification tolerances as required to meet the constraints of the Technical Specifications and the safety analyses.

The current specification applied to the upgrade fuel would result in no change to the flow-related factors. However, the additional oxide thickness projected for the upgrade core can influence these factors. These factors were ev61uated and documented for limiting operating conditions by the licensee.

Based on these evaluations, the staff concludes that conservative engineering and flow-related hot channel factors demonstrate that there is high confidence the safety limits can be met.

2.5 UPGRADE FUEL PERFORMANCE ASSESSMENT The ELAF analyses and characteristics have been assessed in previous sections.

These assessments form the basis for an evaluation of upgrade fuel performance under normal, transient, and accident conditions as compared to the current fuel.

l

o 2.5.1 Normal Operation The increased oxide thickness predicted as a result of the increased operating time required to reach the elevated burnup limit w'll result in different ,

fuel plate temperatures and a small change in coolant channel flow area. The predicted maximum oxide thickness is less than the demonstrated spallation thickness. Due to the increased operating time, it is concluded that the probability of clad failure dee to corrosion pitting is increased, but analyses show consequences are minor and the MHA bounds the potential radiological consequences.

The engineering hot channel factors and flow-related factors used to establish safety limits have been conservatively estimated and analyses show safety limits can be maintained.

The use of the ELAF will result in larger neutron spectral shifts during core lifetime. These shifts will result in changes to control blade wort.h (~10%

less) and in reactor response to experiment loading in the center flux trap, The change in neutron spectra will also result in changes to the void and temperature reactivity coefficients. As addressed in Section 2.3.2.3, both have been shown to decrease in the ELAF core.

Power shifts due to experiment loading in the flux trap are small (~3%) and do not impact the safety margin as currently approved.

2.5.2 Transient Operation The time-temperature response of the reactor fuel will be different for transient operation after the oxide layer exceeds that for the present core.

The increased oxide will result in a slower energy transfer to the coolant in response to a power increase. This will result in higher fuel temperatures and slower coolant heatup. The staff concludes that these conditions will not result in fuel failure during normal operational transients. The changes in control blade worth, negative reactivity coefficients, and hot

4 1

l spot location with respect to experiment loading and control blade position will result in different transient behavior. The staff concludes that none of these changes would result in fuel failure for operational transients.

'2.5.3 Operation Under Accident Conditions l As with operational transients, those accidents which result in reactor power <

increases will have slower energy transfer to the primary coolant as a result of increased oxide thickness. Thus, higher fuel plate temperatures will result. The decreased energy transfer rate to the coolant will influence the reactivity feedback characteristics for the reactor. This has more signifi-l l cance for accidents than for operational transients because postulated I perturbations are faster and more severe and inherent reactor feedback characteristics limit the consequences. The increased oxide thickness has a l

small influence on safety margins during loss of flow or pressure events due to the slower energy transfer f rom the fuel. -

The ELAF has a flatter power profile than the current core. The current core L power peaking is in the exterior plates, which are close to the flux trap and L

reflector. The ELAF. power peaking will occur on the inner plates and is of a lower magnitude. These characteristics will alter reactor response to reactivity l

y accidents. The reactor response to a reactivity transient has been analyzed by the licensee. The results show that fuel melt will not occur.

l l The additional fuel and boron in the ELAF core can be expected to result in a different neutron spectra. The reactor core reactivity coefficients are dependent on neutron spectra and are different for the ELAF core. A change in reactivity coefficients will result in a different response to accident )

conditions. Sensitivity analyses were completed by the licensee. The results l demonstrate that fuel melt is not predicted. I The review documentation provides one conservative evaluation for a reactivity perturbation with an increased oxide thickness. The assumed oxide thickness  ;

l 1'

~

14 is larger than projected, but the results indicate large changes. The fuel centerline temperature goes from 437.5'F to 765.6'F and the clad temperature goes from 379'F to 396'F. The peak power is predicted to increase from 51.48 MW to 57.9 MW. The transient profile is changed and the integrated energy is larger. The model for this event was not provided by the licensee and was not reviewed.

The sensitivities to critical heat flux correlation, channel modeling, inlet temperature, coolant flow velocity, and moderator direct heating were evaluated

-by the licensee. The results show small sensitivities and an analysis with worst case input is reported to result in fuel plate temperatures below the expected fuel plate blistering temperature.

The staff concludes that the behavior of the ELAF under accident conditions is acceptable.

2.6 ELAF - RADIOLOGICAL CONSEQUENCES The licensee states that the ELAF will not cause any changes to the MHA and environmental considerations. The most current evaluation for the MHA is in l

Addendum 4 of the Hazards Sunnary Report . This evaluation considers the four fuel plates at the peak flux position which contain 78.58 g of U-235.

This develops the basis for the percent of core that is assumed to melt.' The iodine activities are then calculated based on operation at 10 MW for 80 days.

If four fuel plates at the peak flux position are taken for the ELAF core, they contain 203.3 g of U-235. With a total core inventory of 10,163 grams the percent of meltdown is 2% as compared to 1.3%. Thus the current MHA evaluation as contained in the Hazards Summary Report, Addendum 4, is not applicable for the new core, and the licensee reevaluated the MHA for the upgrade core.

l y

l l

l The revised analysis bounds the consequences for the ELAF elements to within the current elements since the power density is higher on plate 1 of the present core.

The proposed percent of melt is equivalent to the previous basis and the staff l concludes it is acceptable.

1 2.7 CHANGES IN TECHNICAL SPECIFICATIONS l I

Technical Specification 3.8.a. is amended to limit peak burnup for UA1 x intermetallic fuel to a calculated value of 2.3 x 1021 fissions per cubic i centimeter. This burnup limit has been evaluated by the staff in this review and found to be acceptable. Tha reference to uranium-aluminum alloy type fuel is removed from the Technical Specifications because this fuel is no longer used at the MURR. The basis for thh Technical Specification is also changed to reflect the new reference for the peak burnup limit.

Technical Specification 3.8.d. is amended such that fuel elements or fueled devices outside the core-shall be stored in a geometry such that the calculated K,ff is less-than 0.9 under all conditions of moderation. The limit was K,ff of 0.8. It was not certain from calculations perforned if ELAF fuel with its higher U-235 loading would be able to meet the K,ff storage limit of 0.8 in the I

existing storage arrangements. A limit of K,ff of 0.9 for fuel storage outside of the reactor core is specified in American National Standard ANSI /ANS-15.1 i "The Development of Technical Specifications for Research Reactors" which is followed by the research reactor community and supported by the NRC. Many I res. orch reactors have approved Technical Specifications with a fuel storage limit of K,ff of 0.9. The staff has concluded that fuel elements or fueled aevices can be stored safely with a K,ff of 0.9.

Technical Specification 4.1.b. is amended to remove the reference to uranium-

]

aluminum alloy type fuel as was done in Technical Specification 3.8 a.

e g

.1 The basis for Technical Specifications 4.1.a. and b. are expanded to include.

ELAF fuel.

Technical Specification 4.1.c. and its basis are amended to add a description of ELAF as reviewed by the staff. This includes maximum U-235 loading in grams per element and the maximum amount of boron carbide per element as a burnable poison. The Technical Specification allows any combination of ELAF and current 775 gram loading fuel to be used in the core. Mixed cores will be subject to the existing Technical Specifications concerning core performance. Mixed cores were reviewed by the staff in this review and found to be acceptable when they meet the other requirements of the Technical Specifications.

Technical Specification 5.5 is amended to require a fuel plate inspection of one element out of every eight elements that have reached their end-of-life fcr anomalies. Reference to aluminum alloy fuel has also been removed from this Technical Specification. This Technical Specification restates what ks-w .

4een acceptable to the staff.

3.0 ENVIRONMENTAL CONSIDERATION

This amendment involves changes in the installation or use of facility components located within the restricted area as defined in 10 CFR part 20 and changes in inspection and surveillance requirements. The staff has determined that the amendment involves no significant increase in the amounts, and no significant change in the types, of any effluents that may be released offsite, and there is no significant increase in individual or cumulative occupational radiation exposure. Accordingly, this amendment meets the eligibility criteria for categoricalexclusionsetforthin10CFR51.22(c)(9). Pursuant to 10 CFR 51.22(b),noEnvironmentalImpactStatementorEnvironmentalAssessmentneed be prepared in connection with the issuance of this amendment.

~ .

.g.

- s 17

4.0 CONCLUSION

S The increased operation time to reach the proposed burnup limit for the ELAF will result in higher oxide thicknesses on the fuel plates as predicted by the Griess correlation and the experimentally derived multiplier for MURR 10 MW operation. The projected oxide thickness at the hot spots is below the spalla-tion thickness.

The increased oxide thickness will also result in higher fuel plate temperature ano different heat transfer characteristics during transient operation. These changes have been evaluated for the new fuel, and the results show higher fuel temperatures but fuel plate blistering or melt is not predicted.

The nuclear peaking for the new fuel- has been calculated using the BOLD VENTURE-IV code system. This system was benchmarked by comparing calculated burnup with burnup measurements of an element that operated at a reactor power of 5 MW for 82.5 MWD power history. The benchmark showed good agreement with the measured results. The submitted data has sufficient detail to verify that the following differences Detween the benchmark and proposed operation have been-considered and consequences assessed: 1) the coolant, clad, and fuel meat temperatures are different as a result of 10 MW reactor power and the increased oxide thickness, and 2) the heavier U-235 loading and burnable poison are expected to result in different neutron spectra than that in the present core.

The spectra will also shift as burnup occurs. The change in cross sections and the sensitivity of control blade worths, fuel plate peaking, and core tempera-ture and voiding reactivity coefficients have been evaluated.

Tre evaluations show that cross sections are not very sensitive to temperature end burnup. The control blade worths are influenced by the fuel element type.

Analyses show that the shutdown margin will be less with the ELAF elements.

UMC states that Technical Specification compliance will be demonstrated by test.

- d 9 -

The core temperature and void reactivity coefficients are decreased in ELAF and the core temperature coefficient for a complete ELAF core is calculated to be less than current Technical Specification limits. UMC has to operate within the bounds of the Technical Specifications. Cores that contain ELAF fuel will have the core temperature coefficient measured to ensure Technical Specification compliance.

The power peaking in a core composed of ELAF has been evaluated with limiting experiments in the flux trap to verify that maximum fuel plate peaking occurs with water in the flux trap. The analysis shows that changing the flux trap composition within Technical Specification limits does result in significant power shifts. They are small and power densities remain less than the peak power produced in Plate 1 of the current element.

Based on this review, the staff concludes that operation with ELAF elements in the MURR at power levels no greater than 10 MW is acceptable.

The proposed changes to the fuel specification do not ensure that the engineering factors are appropriate. The engineering factors used for deriving safety limits need to be verified for the new fuel based on the approved specification for fuel fabrication. Scoping analyses have demonstrated that the fuel fabrication should not result in unacceptable limits.

However, the licensee will perform an analysis using actual fabrication specifications that verifies the scoping analysis before ELAF fuel is placed in the reactor.

The staff has concluded, based on the considerations discussed above, that:

(1) because the amendment does not involve a significant increase in the probability or consequences of accidents previously evaluated, or create the possibility of a new or different kind of accident from any accident previously I

- l

. ig . I evaluated, and does not involve a significant reduction in a margin of safety, the amendment does not involve a significant hazards consideration, (2) there 1 is reasonable assurance that the health and safety of the public will not be endangered by the proposed activities, and (3) such activities will be conducted "

in compliance with the Commission's regulations and the issuance of this 1 amendment will not be inimical to the common defense and security or the  ;

health and safety of the public. I Dated: August 1, 1990 l l

l l

l l

l l

l l

1 l

\

i

9.0 REFERENCES

1

'1. MURR Hazards Summary Report, July 1,1965,- and Addendum 1-5.

2. Technical Specificaticns for University of Missouri Research Reactor Facility, July 9,1974, as amended. ,

1

3. Soon Sam Kim and Charlie McKibben, MURR Upgrade Neutronics Analysis Using l AMPX-II/ BOLD VENTURE Computation System Benchmarked to the Destructive l Analysis of Fuel Element 775F3, MURR Internal Report, Septenter 1,1986.

4. L. G. Miller and J. M. Beeston, Extended Life Aluminide Fuel Final Report, EGG-2441, June 1986.

5. J. C. Griess, et al., Effect of Heat Flux on the Corrosion of Aluminum by ,

Water. Part IV, ORNL-3541, February 1964 l I

I 1

I I