ML20206S599
| ML20206S599 | |
| Person / Time | |
|---|---|
| Issue date: | 06/23/1998 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-3104, NUDOCS 9905210167 | |
| Download: ML20206S599 (11) | |
Text
.-
na Powers 7/8/98 hbfb
)
f,PR ADVISORY COMMITTEE ON REACTOR SAFEGUARDS REACTOR FUELS, ONSITE FUEL STORAGE, and DECOMMISSIONING SUBCOMMITTEE MEETING MINUTES APR!!. 23-24,1998 ROCKVILLE, MARYLAND The Advisory Committee on Reactor Safeguards (ACRS) Subcommittee on Reactor Fuels, Onsite Fuel Storage, and Decommissioning held a meeting on April 23 and 24,1998 in Room T-283,11545 Rockville Pike, Rockville, Maryland, with representatives of the U.S. Nuclear Regulatory Commission (NRC) and Industry. The purpose of this meeting was to discuss the basis of the proposed NRC fuel failure criterion for high burnup conditions, and the adequacy of the NRC fuel codes to predic! fuel behavior under accident conditions. The Subcommittee also heard presentations by industry representatives regarding the reactor fuel high burnup issues and programs. Dr. Medhat El-Zeftawy was the cognizant ACRS staff engineer for this meeting.
The meeting was convened at 8:30 a.m. on April 23,1998, recessed at 4:30 p.m.; and then reconvened at 8:30 a.m. on April 24,1998 and adjourned at 5:20 p.m.
ATTENDEES ACRS D. Powers, Chairman R. Seale, Member j
T. Kress, Member W. Shack, Member A. Cronenberg, Fellow M. El-Zeftawy, Staff N!1C G. Holahan, NRR T. King, RES R. Meyer, RES H. Scott, RES F. El-Tawila, RES M. Chatterton, NRR E. Weiss, NRR H. Vandermolen, RES U. Shoop, NRR/RES H. Richings, NRR J. Mitchell, EDO F. Sturz, NMSS D. Carlson, NMSS S. Wu, NRR INDUSTRY r0 J. Butler, NEl B. Dunn, FTl i
T. Coleman, Framatome D. Risher, Westinghouse
'JD >b R. Yang, EPRI L. Neimark, ANL P. Kumar, DOE J. Rashid, Anatech a.'
d D. Diamond, BNL A. Mottz. Penn State c
R. Montgomery, Anatech G. Meyer, Framatome AM 9905210167 990623
3104 PDR i
!! ' t" i n ' cr
' SAL s
_fh EO
'h
~ na ;;
- l..
I i.
2 No written comments or requests for time to make oral statements were received from members of the public. A list of attendees is available in the ACRS office and will be made available upon request.
OPENING REMARKS BY THE SUBCOMMITTEE CHAIRMAN Dr. Dana Powers, Chairman of the Subcommittee, convened the meeting at 8:30 a.m. and stated that the purpose of this meeting is to discuss the proposed NRC fuel failure criterion for high burnup conditions and the agency-wide program plan for high bumup fuel. The Subcommittee will gather information, analyze relevant issues and facts, and formulate proposed positions and actions as appropriate for deliberations by the Full Committee.
Dr. Powers indicated that there are benefits from the operation of high burnup fuels ( e.g.,
secietal benefits ), however, the question that need to be answered is it safe to do so? NRR has now approved some licensees to operate fuel for up to 62 GWd/t. This is nearly twice the bumup of the experimental data base. The ACRS is aware of experiments performed in France and Japan that suggest burnup compromises an important barrier to defense-in-depth. Also EPRI has' objected to these tests as insufficiently prototypic. Furthermore, the Subcommittee is aware of arguments that realistic patterns of energy inputs predicted by computer codes are much slower than the tests. Now we have judgements from NRC contractors that suggest the codes are not conservative and sufficiently accurate to draw conclusions.
i NRC STAFF PRESENTATION Mr. G. Holahan, NRR, stated that the NRC staff is developing an agency-wide plan to address safety issues related to high burnup fuel. The plan addresses current and extended burnup levels. For bumup extension strategy, the staff is recommending the following:
Small increments up to 62 GWd/t will be considered on a case-by-case basis Industry should take the lead for further higher burnup levels, and additional testing and analysis is required.
For higher burnup levels, the staff recommends the use of risk-informed approach to focus the emphasis.
The research plan contains the following elements:
Cladding integrity and fuel design limits Control rod insertion problems Criteria and Analysis for reactivity accidents Criteria and Analysis for LOCA Criteria for BWR power oscillations ( ATWS)
Fuel rod and Computer Codes Source term and Core melt progression l
Transportation, dry storage j
High enrichments ( greater than 5% )
m l
I 3
i The staff anticipates that licensees will not seek fuel enrichments above 5% in the future. The use of risk-information is to allow certain relaxation, e.g., in the traditional clad failure criteria.
The staff emphasizes that the Material condition of the fuel and " Performance Monitoring" are better investment than testing with correlation for bumup. Based on the low probability of the reactivity insertion accidents, the staff is not recommending any backfit analysis.
For higher bumups ( greater than 62 GWd/t ), the staff is proposing an " Analysis Method",
which focus on obtaining Mechanical properties data at higher bumup and use Failure Modeling rather than Enthalpy Criteria to determine cladding failure.
The issue on source term and core melt progression is not being addressed actively. A brief consideration of bumup related factors leads the staff to conclude that it is unlikely that high l
bumup will have a significant effect on source terms or core melt progression. The current source term is considered to be adequate. A more thorough assessment of these possible effects, utilizing recent French data, had been originally been planned for FYg8, but reductions in the severe accident area have eliminated funding for that work.
Criteria and Analvsis for Reactivity Accidents Dr. R. Meyer(RES)
The specific accidents of concem are the rod drop accident in a BWR and the rod ejection l
accident in a PWR. For these accidents, the NRC uses criteria to ensure that fuel rods remain l-coolable and that fuel particles are not dispersed into the coolant ( 280 cal /g peak fuel enthalpy) and to indicate the occurrence of cladding failure (DNB, MCPR,170 cal /g peak fuel enthalpy) for the purpose of dose calculations. Tests in the French CABRI reactor in late 19g3 with some highly degraded commercial fuel resulted in cladding failure at very low fuel enthalpy ( 30 cal /g average for a fuel rod) and substantial fuel dispersal. Analysis of these and similar tests showed that failures were occurring in a partially brittle manner, as a result of the mechanical expansion of the pellets, rather than by dryout and overheating of the cladding as addressed by l
the current criteria. It thus appear that the current criteria may not achieve their purpose for l
high bumup fuel.
Dr. Meyer indicated that the frequency of occurrence of a BWR rod drop accident is below the l
range of interest for consideration as a generic issue, whereas the frequency for a PWR rod ejection accident is just within that range. Shortly after the CABRI results, RES suggested l
tentative interim criteria as follows:
Oxide spalling -
none allowed Cladding failure -
100 cal /g ( enthalpy increase)
Coolability -
200 cal /g (enthalpy limit) less than 30 GWd/t Nu Cladding failure for Greater than 30 GWd/t A fixed bumup limit was not given for the above criteria because bumup did not seem to be the most important varireble ( lt is oxidation). The' data base for these criteria includes bumups up to 62 GWd/t and oF.ide thickness to 130 microns, with most oxide thickness below 80 microns.
The main limitation appears to be that oxidation should not be severe that spallation occurs because that introduces known phenomena that can cause localized embrittlement.
I u
1 4
Dr. Meyer stated that the current data base has substantial limitations. To avoid some of these limitations, the staff will participate in new programs through intemational agreements. In France, a new water loop will be constructed to test more current PWR cladding types with prototypical pulse widths, water as the coolant, and appropriate coolant flow to investigate
- cladding failure and the effects of dispersal fuel particles. In Japan, a new high-temperature, high-pressure capsule will be constructed to test more current PWR and BWR cladding types and pulse widths effects.
Criteria and Analysis for LOCA Dr. Meyer stated that for the LOCA a'.:cidents, the NRC uses cladding embrittlement criteria (2200 degree F peak cladding temperature,17% cladding oxidation) to ensure that coolable geometry is not lost: and related models must be used in safety analyses for oxidation kinetics, ballooning, rupture, and flow blockage to demonstrate that long term cooling is maintained.
Additional analyses are performed to show that seismic and blowdown loads do not fragment the fuel or interfere with control rod insertion during and after LOCA. Some high burnup fuel rods accumulate heavy oxide coatings ( approaching 17% )during normal operation and experience some loss of ductility from hydrogen absorption.
Fuel behavior during LOCA accidents is assessed with embrittlement criteria and several types of analyses:
the initial stored energy in the fuel is calculated with the NRC's FRAPCON-3 code or similar vendor and licensee codes during the transient, the amount of oxidation and the peak cladding temperature are calculated for comparison with the embrittlement criteria, and the deformation of the rod is calculated to provide related flow blockage. NRC's.FRAPTRAN code can calculate these quantities, and the models of these phenomena are usually built into vendors' systems codes systems codes like the NRC's TRAC-P and TRAC-B codes calculate the entire plant a
transient, including the long term cooling phase finite-element structural mechanics codes are used to calculate the fuel assembly and core response to seismic and LOCA loads in 1997, RES initiated a major program to establish a data base for LOCA criteria and models utilizing typical high bumup fuel from U.S. power reactors. The program is being carried out in the hot cells at Argonne National Laboratory (ANL) and will also provide fundamental mechanical properties. Cooperation on obtaining and preparing fuel rods for the tests is being obtained from EPRI and DOE, and collaboration on technical matters is also being obtained i
from France, Japan, and Russia.
j i
1 l
i
5 1
A Probabilistic Profile for Reactivity Accidents Dr. David J. Diamond Dr. Diamond, Brookhaven National Laboratory (BNL), briefed the Subcommittee regarding a probabilistic profile for reactivity accidents. The objective of this study is to help put NRC program plan for high bumup fuel in context. The probabilistic profile equals the frequency of l
occurrence of unacceptable fuel damage. The study considered the current (280 cal /g) and the proposed (100 cal /g) criteria for unacceptable fuel damage. In addition, design-basis (BWR rod drop & PWR rod ejection accidents) and beyond design-basis events were also considered.
j Some of the observations on the NRC program plan that raise questions are:
Work-details on ATWS have not yet been defined Very little work on Zirc-4 with spalling oxide Very little work on Zirc-2 without liner Multiple tests for less than '65 GWd/t on Zirc-2 with liner and Zirc-4 for RIA and LOCA No work on Niobium alloys for less than 65 GWd/t for RIA or LOCA No work for greater than 75 GWd/t except stored energy methods There are also beyond design-basis events for the PWR that are important. Of particular interest are boron dilution events. Frequencies were calculated for events with a large boron dilution to achieve a fuel enthalpy of more than 280 cal /g. With a smaller fuel enthalpy criterion, a smaller dilution is required and some of these events may have frequencies of occurrence j
that are in the same range as design-basis accidents. Other beyond design-basis events could l
be refueling events.
l Dr. Diamond indicated that for BWR rod drop accident (RDA), deterministic calculations show l
fuel damage is only possible at lower criterion, and the RDA has such low probability and l
should not warrant significant attention in the NRC research effort. For PWR rod ejection l
accident (REA), deterministic calculations show fuel damage is only possible at lower criterion.
l The results show that high burnup fuel does not behave in ways anticipated by simple l
extrapolation of data for lower burnups. Dr. Diamond stated that in general reactivity accidents
~
L are not major contributors to risk. BWR ATWS events have a significant frequency.
Uncertainties are large and profile has gaps with further investigations are warranted.
Dr. Powers questioned the treatment of delayed neutrons and indicated that the current methods of neutron kinetics are flawed. Dr. Powers cited the study performed by Downer and Ott. Dr. Diamond disagreed and stated that the enemy deposited is about the same, however the power guide is different, and at higher bumup there is more uncertainty in the beta factor.
i NMSS High Burnup Activities Fritz C. Sturz Dr. Sturz stated that current approvals for fabrication and handling of power reactor fuel do not exceed 5% U-235 enrichment. NMSS, however, is studying the criticality safety issues for 5-10% enrichments. The new criticality safety implications for 5-10% enrichments include the following:
7
.a.
[
~6 3
i Unmoderated or weakly moderated criticality becomes possible i
Must evaluate configurations with intermediate neutron energies Critical masses become smaller, configurations more varied
==
Criticality codes and validation Criticality issues apply to full range of activities for production, transportation, and storage of materials '
Dr. Sturz indicated that commercial power industry has requested higher fuel bumups to allow longer operating cycles. The plan of action for NMSS is to continue research work on criticality methods and data, validation and range of applicability. The NRC and industry will be gathering criticality benchmark information from foreign sources. Recent user-need memo from NMSS to RES to update spent fuel technical licensing bases include cladding and fuel integrity issues, and nuclide inventory calculation issues.
High Burnup Fuel Performance Exnerience Margaret S. Chattedon, NRR Ms. Chatterton stated that in late 1995 and early 1996, several control rods failed to insert fully during scrams at two PWRs (South Texas and Wolf Creek). All of the affected control rods were positioned in high burnup fuel assemblies. Upon inspection of the rods and fuel assemblies, the control rods were found to be in good condition, but the fuel assemblies were
' deformed. Related evidence was fcond in North Anna and at a number of plants of similar design in Europe.
The root cause for the observed cases of control rod sticking was determined by Westinghouse to be the fuel assemblies response at high bumup to several aspects of fuel design including creep, oxide thickness, operating temperatures, holddown spring stiffness, thimble tube thickness, and dashpot dimensions.
Ms. Chattedon summarized high bumup fuel performance experience as follows:
Oxidation levels found to be significantly higher than predicted (using approved models) for at least one vendor fueltype Excessive intamal gas pressure in IFBA rods Incomplete rod insertion events Large axial offsets High crud buildup Accelerated growth of rods and assemblies Low energy fuel failures during RIA tests Much more demanding operating environment Ms. Chatterton indicated that the safety concem regarding oxidation levels is brittle fracture can occur post LOCA. The regulatory requirements is oxidation must be less than 17% of clad thickness, post LOCA (10 CFR 50.46). The fuel is operating in a more demanding environment with reduced margins. The safety questions are not enough to cause a backfit restrictions, but enough to require attention. Industry and NRC programs are being planned to confirm existing fuel operating margins. Utility awareness has been heightened, and it is believed that utilities have the means and ample motivation to avoid this problem in the future.
t
i' 4
7 Electric Power Research Institute (EPRI) Rosa Yang Ms. Yang stated that to address fuel performance and regulatory issues proactively, the utilities asked EPRI to formulate a comprehensive program leading to more robust fuel program (RFP).
The robust fuel is characterized as a fuel with high reliability, sufficient operating margins, and capable of providing the economic benefits needed in the electricity market. The objectives of the RFP are:
)
Resolve outstanding performance, reliability, operational and regulatory issues of f
current fuel designs and core conditions, Understand and quantify performance margins for current fuel designs and avoid regulatory mandates of overly restrictive operationallimits, With the direct involvement of NEl, serve as the focal point on industry-wide fuel related regulatory issues, Provide technology and gain regulatory acceptance for fuel designs and operation to higher burnups.
The RFP is to be run by the utilities and managed by EPRI, with close and direct participation -
by all major fuel vendors, INPO and NEl. A utility steering committee structure has been put in place for the direction and management of the RFP. An initial five-year plan identifying the research and development projects needed most urgently has been formulated under the direction of utility working groups with input provided by all fuel vendors.
The current plan focuses the RFP effort into the following four major areas:
Fuel / Water chemistry-to address fuel performance problems associated with exposure to primary coolant environment at operating temperatures Response to Transients-to address the transient behavior of high burnup fuel and the impact of this behavior on licensing High Burnup Properties-to address the need for data and performance-based technical j
requirements that will ensure the presence of adequate margins at high burnup and permit the licensing of burnup extensions Failure Mechanisms / Mitigation-to develop an understanding of the root causes of fuel failures and the mechanisms of postfailure secondary degradation.
Work to be performed in support of each of the above areas is being directed by a separate utility Working Group. However, issues such as cladding corrosion, crud formation / deposition I
and their effect on the axial offset anomaly (AOA) in PWRs, or the resolution of the burnup extension licensing issues require coordinated effort on the part of all four Working Groups.
Evaluation of RIA-Simulation Exoeriments R.O. Montgomery, Y.R. Rashid (ANATECH
{
Corp.)
The objective is to evaluate the response of RIA-simulation tests on pre-irradiated test rods and to assess the applicability of the results to LWR fuel. The approach taken included understanding the RIA-simulation experiments, defining the in-reactor fuel rod conditions, and
H 8
identifying method to translate experiments to in-reactor. To define the in-reactor fuel rod loading and response conditions, the following steps were taken:
Evaluate pulse characteristics using 3-D spatial kinetics methods for both realistic and l
conservative control rod worths Evaluate fuel rod behavior under typical LWR in-reactor coolant and power conditions To translate experiments to LWR conditions, a Thermo-mechanical basis for quantifying fuel rod response was developed. To assess the cladding materiallimit, Anatech used a strain l
energy density (SED) model as a measure of loadinMntensity. The SED is the mechanical l
energy stored by the cladding during PCMI-driver, transient. CSED is a measure of cladding failure potential. Cladding failure occurs when SED reaches a critical value CSED; CSED is the i
energy released at failure.
Summary of Anatech analysis results include:
Good agreement observed for residual cladding hoop strains in the REP Na tests l
Good agreement observed for the fuel and cladding axial elongation in the CABRI REP l
Na, NSRR, and SPERT-CDC test CSED relations based on irradiated cladding properties allows for differentiation between success and failures I
CSED empirical model can be used to estimate cladding failure by PCMI when coupled with best-estimate fuel behavior analysis For in-reactor conditions, wider pulse causes heat conduction from the pellet rim for high bumup fuel during pulse. Higher cladding temperature causes more ductility in conclusion, fuel rod response in LWR is different than what has been observed in tests. The SED approach is a better translation method than simply using cal /g.
l Hiah Burnun Effects and Ernected Marains in LOCA Plant Analvsis M.E. Nissley, Westinghouse Electric Corp.
Mr. Nissley stated that the current Westinghouse, operating strategy is to place a higher demand on fuel rod design, such as the capability of power upratings, longer fuel cycles, and reduced leakage and increased peaking factors. The current fuel managements typically result i
in two or three cycle residence.
10 CFR 50.46 acceptance criteria for peak cladding temperature and local oxidation are PCT less than 2200 deg. F, and local oxidation of less than 1700%. Limits embrittlement and potential for fuel fragmentation during quench. Westinghouse conservatively interprets 17%
limit as steady state and LOCA transient oxidation. Westinghouse used the best estimate LOCA evaluation model in which the parameters that affect the hot assembly transient response (e.g., break flow, power distribution) are analyzed using W COBRA / TRAC code.
I Uncertainties in the hot rod response for a given W COBRA / TRAC transient are analyzed using HOTSPOT code. Licensing basis PCT (PCT 95%) is determined using a combination of
i, 9
response surfaces and Monte Carlo simulations. Maximum local oxidation was determined using a plant-specific transient which exceeds PCT 95%.
The modeling approach used for LOCA bumup study is to select an existing transient which gives PCT of approximately 2200 deg.F ( greater than PCT 95%0 for beginning of life (BOL) conditions, then replace BOL hot assembly with an assembly with characteristics of bumed fuel.
Repeat the transient to quantify the e#ect on peak cladding temperature and oxidation.
Westinghouse concluded that there are significant PCT margins exist for high bumup fuel.
Transient oxidation decreases with bumup, but can approach 17% maximum oxidation limit with Zr-4 clad fuel if pre-transient and transient oxidation are considered. Improved cladding alloys provide additional margin to oxidation limit, by reducing pre-transient oxidation.
In4 ell Proaram on LOCA laauen and the Mechanical Preaar+Ia= of Hiah Burnup LWR Cladding Dr. Lawrence A. Neimark / ANL
]
Dr. Neimark stated that the NRC regulatory criteria for LOCA is based mainly on experiments with nonirradiated early Zircaloys. High bumup fuels use advanced variations of Zircaloy-4 (PWRs) and Zircaloy-2 (BWRs) and Zirconium-niobium alloys. Older design claddings are susceptible to oxidation and hydriding, and advanced claddings emphasize corrosion resistance.
In-reactor tests in France, Japan, and Russia indicated reduced failure threshold for simulated RIA with fuel at bumups greater than 45GWd/t. The degradation of cladding mechanical properties at high bumup would affect fuel response to RIA and LOCA events. The issue: are current NRC LOCA criteria applicable to fuel rods clad with current materials and operated to bumup of 60 GWd/t, significantly above previous extrapolations, or are modified criteria necessary?
ANL is conducting a hot-cell program on fuel rod segments to assess the applicability of current NRC criteria for cladding embrittlement in 10 CFR 50.46 to the advanced cladding materials now in use. In addition, the ANL program will determine the mechanical properties of high l
bumup cladding necessary to analyze transients important in licensing safety analysis. The purpose of the program is to provide input to the NRC-sponsored codes and models, specifically FRAPCON, FRAPTRAN, and BALON2 swelling and ballooning model in l
FRAPTRAN.
Issues with high bumup fuel rods are related to the following:
l Cladding oxidation resulting in reduced wall thickness and hydrogen uptake, leading to embrittlement and reduced load-bearing capability Significant cladding oxidation during normal operation may result in insufficient margin to the 17% limit during LOCA l
Radiation damage, contributing to cladding embrittlement Thermal shock behavior of embrittled, high bumup cladding may be di#erent than that represented in the current database
D I
i I
10 Peripheral fuel restructuring rim", probably resulting in a more friable fuel in proximity to cladding breaches Chemical bonding of fuel-to-cladding could affect access of released fission gases to the cladding and thereby affect cladding ballooning in consideration of the above issues, three types of tests under simulated LOCA conditions have been identified,i.e. (1) thermal-shock tests by direct quenching and by slow cooling and quenching at less than 700 deg.C following ballooning and rupture of the cladding and high-temperature oxidation, (2) quenching test under combined thermal shock and bending moment applied in simulation of seismic loading, and (3) instrumented impact testing of the intact cladding specimens that survived the thermal shock.
Dr. Neimark indicated that the industry fuel design and operating conditions are expected to undergo further changes as more advanced cladding materials are developed. Under these circumstances, mechanical properties of high burnup fuel cladding are being studied so that results LOCA, RIA, operational events, and power ramping can be extrapolated to advanced cladding materials without repeating major integral experiments in test reactors. To provide the most applicable data base on fracture toughness and tensile properties and rnechanistic understanding of these properties, three types of tests were selected, i.e., the modified ring-stretch test, electromagnetic ring-expansion test, and the pin-loading fracture-toughness test.
NRC Staff Commitments The staff will brief the ACRS, possibly, in October 1998 regarding the NRC participation in the CABRI program. In addition, the staff will keep the ACRS posted regarding the development of the high burnup fuel program.
i Subcommittee Discussion and Follow-un Actions I
The Subcommittee is not convinced that the NRC has a plan to assure that the knowledge and tools are available to respond quickly to adequately formulated proposals from licensees for higher burnups.
The plans to address issues of fuel performance during ATWS events need to be further developed.
The effect of high bumup on radionuclide source terms need to be studied further.
l Anticipatory research need to be added to the plan to respond to line organizations requests.
The Subcommittee did not agree with NMSS staff to further study enrichments in excess i
l of 5%, since there is no indication from the industry to use fuels with higher l
enrichments.
L
7 t
F 11 The NRC staff may have been too conservative in the lead test assembly (LTA) program.
Background material orovided to the Subcommittee No documents were submitted to the Subcommittee prior to the meeting.
i Presentation Slides and handouts Provided Durina the Subcommittee ":: tina The presentation slides and handouts used during the meeting are available in the ACRS Office files or as attachments to the meeting transcripts.
i NOTE:
Additional details of this meeting can be obtained from a transcript available in the NRC Public Document Room,2120 L Street, N.W., Washington, D.C. 20006, (202) 634-3274, or can be purchased from Ann Riley & Associates LTD.,1250 l Street, N.W., Suite 300, Washington, D.C. 20005, (202) 842-0034.
l 1
i I
L