ML20213A678

Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants.Lwr Edition.Proposed Revision 0 to New SRP Section 6.5.5, Pressure Suppression Pools as Fission Product Clean-Up Systems. for Comment

ML20213A678
Person / Time
Issue date:	04/30/1987
From:	Office of Nuclear Reactor Regulation
To:
References
NUREG-0800, NUREG-0800-06.5.5-R0, NUREG-800, NUREG-800-6.5.5-R, SRP-06.05.05, SRP-6.05.05, NUDOCS 8704280166
Download: ML20213A678 (51)

Text

_ - -. . . . ._ _ __ __-__-

e. . . .

1 U.S. NUCLEAR REGULATORY COMMISSION NUREG-0800

" STANDARD REVIEW PLAN FOR THE REVIEW OF SAFETY ANALYSIS REPORTS FOR NUCLEAR POWER PLANTS" l

NOTICE OF ISSllANCE AND AVAILABILITY 5 "FOR COMMENT" PROPOSEDNEWSECTION6.5.5ANDPEGULATORY ANALYSIS AND SUPPORTING TECHNICAL REPORTS i

l The U.S. Nuclear Regulatory Comission (NRC) has published proposed new Section 6.5.5., "Suppressien Pools as Fission Product Cleanup Systems," of NUREG-800,

" Standard Review Plan for the Review of Safety Analysis Reports for Nuclear Power Plants " LWR Edition (SRP).

l The proposed new SRP Section 6.5.5 consists of revision 0, its supporting Regulatory Analysis, and the supporting Technical Report " Effectiveness of :l H

l BWR Pressure Suppression Pools in Retaining Fission Products." This new l i

! section would permit pressure suppressicn pools to be considered as miti-

! gative engineered safety features capable of retaining some fraction of fission products that might be directed into them following potential reactor E!*** ate '7o* o 0800 R PDR

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

. .. ...s 2

accidents. Its review procedures would provide means of obtaining estimates O

of the fractions of fission products that should be assumed to be retained by the pools. These procedures would become necessary following a proposed future revision of Regulatory Guide 1.3, which currently denies retention capability to suppression pools. .)

,l This revision would be reouired for future plants, but would be optional for present licensees and are part of a proposed series of SRP and Regulatory Guide changes interded to implement the Commission's Severe Accident Policy and to introduce the result of recent severe accidont regulatory research into staff practices. Comments are being solicited from interested organizations, groups and individuals. The staff will evaluate the comments received, and address them, as appropriate, in the final documents.

Copies of the "For Comment" documents will be available after April 6,1987.

Copies will be sent directly to utilities, utility industry groups and associations and environmental and public interest groups. Other copies will be available for review at the NRC Public Document Room, 1717 H Street, NW, ,

l Washington, D.C., and the Commission's Local Public Document Rooms located in the vicinity of nuclear power plants. Addresses of these Local Public Document l l

Rooms can be obtained from the Chief, Public Document Branch, U.S. Nuclear Regulatory Commission, Washington, D.C. 20555, telephone (202) 634-3273.

l l O

i

}

i.

3 4

i Coments should be sent to the Secretary of the Comission, U.S. Nuclear

Regulatory Comission, Washington, D.C. 20555, Attention

Docketing and Service Branch, by June 5. 1987.

f Dated at Bethesda, Maryland, this 6th day of April 1987.

- FOR THE NUCLEAR REGULATORY COMMISSION l

t J

j Parold R. Denton, Director Office of Nuclear Peactor Regulation 5

I i

1 i

l i

i

NURE2-0800 i (Formerly NURE2-75/087) I

/'paArGO 's U.S. NUCLEAR REGULATORY CONINI8SSION C) iM, g n ofi STANDARD REVIEW PLAN OFFICE OF NUCLEAR REACTOR REGULATION e..e*

Proposed Revision Standard Review Plan PSRP-6.5.5, Rev 0 This proposed revision of the Standard Review Plan and its supporting value/ impact statement and associated technical documentation have not received a complete staff review and approval and do not represent an official NRC staff position. Public coments are being solicited on the proposed SRP section and the associated regula-tory analysis and technical support document A3788, " Effectiveness of BWR Pressure Suppression Pools in Retaining Fission Products" prior to a final review and deci-sion by the Office of Nuclear Reactor Regulation as to whether this proposed re-vision should be approved. Coments should be sent to the Secretary of the Com-mission, U.S. Nuclear Regulatory Commission, Washington, D.C. 20555, Attention:

j Docketing and Service Branch. All coments received by June 5,1987 will be consi-dered, and all of the associated documents and comments considered will be made publicly available prior to a decision by the Director, Office of Nuclear Reactor Regulation, on whether to implement this revision. Single copies of each of these documents are available upon written request to the Division of Information Sup-port Services (Attention Distribution Section), U.S. Nuclear Regulatory Commis-sion, Washington, D.C. 20555. )

l i

Rev. O USNRC STANDARD REVIEW PLAN Star.dard review plans are prepared for the guidance of the office of Nuclear Reactor Regulation staff responsible for the review of applications to construct and operate nuclear power plants. These documents are made available to the public as part of the Commission's policy to inform the nuclear industry and the general public of regulatory procedures and policies. Standard review plans are not substatutes for regulatory guides of the Commission's regulations and compliance with them is not required. The i' standard review plan sections are keyed to the Standard Format and Content of Safety Analysis Reports for Nuclear Power Plants.

Not all sections of the Standard Format have a corresponding review plan.

Published standard review plans will be revised per.odically, as appropriate, to accommodate comments and to reflect new informa-tion and emperience.

Comments and suggestions for improvement will be considered and should be sent to the U.S. Nuclear Regulatory Commission, 1 Of fice of Nuclear Reactor Regulation Washington, D.C. 20666.

NUREG-0800 (Fcrm:rly NURE"-75/087) u

[] /'ps nea ,IS U.S. NUCLEAR REGULATORY COMMISSION U STANDARD REVIEW PLAN QW k% /! OFRCE OF NUCLEAR REACTOR REGULATION e.ees Proposed New Standard Review Plan, Rev 0 6.5.5 PRESSURE SUPPRESSION POOLS AS FISSION PRODUCT CLEAN-UP SYSTEMS

. REVIEW RESPONSIBILITIES Primary - Plant Systems Branch Secondary -Reactor Systems I. AREAS OF REVIEW Pressure suppression pools are reviewed under this plan only when the applicant claims credit for fission product scrubbing and retention by the suppression pool. The pressure suppression pool and the drywell, when considered as a barrier to the release of fission products, are reviewed to assess the degree to which fis-sion products released during postulated reactor accidents will be retained in the suppression pool. Leakage paths which allow fission products to bypass the pool v are identified and reviewed, and the maximum fractional bypass leakage is obtained, for use in the evaluation of radiological dose consequences.

1. Fission Product Control Requirement Sections of the SAR related to accident analysis, dose calculations, and fis-sion product control are reviewed to establish whether or not fission product scrubbing of the drywell or reactor compartment atmosphere is claimed or re-quired for mitigation of offsite consequences following a postulated accident.

2. Desian Bases A comparison is made to establish that the design bases for the suppression pool and the drywell or reactor compartment are consistent with the assump-tions made in the accident evaluations of SAR Chapter 15.

Rev. 0 USNRC STANDARD REVIEW PLAN Star.dard review plans are prepared for the guidance of the Office of Nuclear Reactor Regulation staff responsible for the review of applications to construct and operate nuclear power plants. These documents are made available to the public as part of the Commission's policy to inform the nuclear industry and the general public of regulatory procedures and policies. Standard review plans are not substitutes for regulatory guides or the Commission's regulations and compliance with them is not required. The standard review plan sections are keyed to the Standard Format and Content of Safety Analysis Reports for Nuclear Power Plants.

Not al! Sections of the Standard Format have a corresponding review plan.

l / Published standard review plans will be revised periodically, as appropriate, to accommodate comments and to reflect new informa-tion and experience.

Comments and suggestions for improvement will be considered and should be sent to the u.S. Nuclear Regulatory Commission.

Office of Nuclear Reactor Regulation, Washington, D.C. 20565.

1 1

\

3. System Design l The information concerning the suppression pool is reviewed to familiarize the reviewer with the expected temperature histories, depth of fission pro-duct entry expected during postulated accidents and potential leakage paths through drywell penetrations.

4. Testing and Technical Specifications The details of the applicant's proposed preoperational tests, and, at the operating license stage, the surveillance requirements, are reviewed under section 6.2.1.1.C. The results of that review are examined to assure that pool depth and amount of leakage bypassing the pool are maintained consistent with the assumptions used in assessing the pool's effectiveness in fission product cleanup.

II. ACCEPTANCE CRITERIA The acceptance criteria for the fission product cleanup function of the sup-pression pool are based on the following requirements from Appendix A of 10 CFR 50:

A. General Design Criterion 41 (Ref. 1) as related to the control of fission products following potential accidents.

B. General Design Criterion 42 (Ref. 1) as related to the periodic inspection of engineered safety features.

C. General Design Criterion 43 (Ref. 1) as related to the periodic functional testing of engineered safety features.

Where they can be shown to be in compliance with these criteria, suppression pools may be given appropriate credit for fission product scrubbing and reten-tion (except for noble gases, for which no pool retention is allowed) in the staff's evaluation of the radiological consequences of design basis accidents.

Specific criteria which must be met to receive credit are as follows:  !

l

1. The drywell and its penetrations must be designed to assure that, even with a single active failure, all releases from the core must pass into the suppression pool, except for small bypass leakage.

2. The bypass leakage assumed for purposes of evaluating fission product retention must be no less than that accepted in the review under sec-tion 6.2.1.1.C, and must be demonstrated in periodic tests by the license technical specifications also reviewed under that section.

3. For plants which have already received a construction permit, the iodine i retention calculated using this section must not be used to justify re- i moval of the standby gas treatment or other filtered exhaust system from '

status as engineered safety features. For such reviews, criterion II.5 of SRP 6.5.1 shall not be applied, and the charcoal absorbers must be at least maintained to the minimum level of Table 2 in Regulatory Guide 1.52, Revision 2.

6.5.5-2 Rev. 0

l Acceptable methods for computing fission product retention by the suppression pool are given in Subsection III, " Review Procedures."

\

(V III. Review Procedures The first step in the review is to determine whether or not the suppression pool is to be used for accident dose mitigation purposes. If no fission pro-duct removal credit is claimed in the accident analyses appearing in chapter 15 of the SAR, no further review is required.

If the suppression pool is intended as an engineered safety feature for the mitigation of offsite doses, then the reviewer estimates its effectiveness in removing fission products from fluids expelled from the drywell or directly

~

from the pressure vessel through the depressurization system.

1. Pool decontamination factor The decontamination factor (DF) of the pool is defined as the ratio of the amount of a contaminant entering the pool to the amount leaving.

Decontamination factors for each fission product form as functions of time can be calculated by the SPARC code (Ref.2), and this calculation should be performed whenever the pool design is judged by the reviewer to differ significantly from those found acceptable as fission product clean-up systems in past reviews. If, however, the time-integrated DF values claimed by the applicant are 10 or less for particulates and 100 or less for iodine v por, the applicant's values may be accepted without any need to perform calculations (Ref. 3). A DF value of 1 (no retention) should be used for noble gases and, unless the applicant demonstrates otherwise,

[m for organic iodides as well.

G) If cciculation of fission product decontamination is done using the SPARC code, the review should be coordinated with the Reactor Systems Branch, which is responsible for establishing the accident assumptions needed to i assemble the input for the calculations.

2. Pool bypass fraction The fraction of the drywell atmosphere bypassing the suppression pool by leaking through drywell penetrations is obtained as a product of the re-view under section 6.2.1.1.C. If B is the bypass fraction and DF is the time-integrated pool decontamination factor, then the overall decontamina-tion, D, to be reported to the Reactor Systems Branch for use in chapter 15 dose calculations may be taken as:

DF 0=

1 + B(DF-1) or 1 1-8 g=B+y v

6.5.5-3 Rev. O

The reviewer should clearly distinguish that fraction of B which may be further treated by the standby gas treatment system from that fraction of B which also bypasses secondary containment.

3. Other containment atmosphere cleanup systems Plants having drywell or containment spray systems for which fission prod-uct cleanup credit is claimed are reviewed separately under section 6.5.2, and credit for both suppression pool and spray cleanup can be given as a result of the separate reviews.

4. Technical Specifications The technical specifications are reviewed to assure that they require periodic inspection to confirm suppression pool depth and surveillance tests to confirm drywell leak tightness consistent with the bypass fraction used in computing the overall decontamination. Technical speci- ~

fication review is coordinated with the Facility Operations Branch as provided in NRR Office Letter No. 51.

IV. EVALUATION FINDINGS The reviewer verifies that sufficient information has been provided by the ap-plicant and that the review and any calculations support conclusions of the following type, to be included in the staff's Safety Evaluation Report:

We have reviewed the fission product scrubbing function of the pres-sure suppression pool and find that the pool will reduce the fission product content of the steam gas mixture flowing through the pool following accidents which blow down through the suppression pool. We estimate the pool will decontaminate the flow by a factor of for molecular iodine vapor and by a factor of for particulate fission products. No significant pool decontamination from noble gases or organic iodides will occur. The system is largely passive in nature, and the active components are suitably redundant such that its fission product attenuation function can be accomplished assuming a single failure. The applicant's proposed program for preoperational and surveillance tests will assure a continued state of readiness, and that bypass of the pool is unlikely to exceed the assumptions used in the dose assessments of Chapter 15.

The staff concludes that the suppression pool is acceptable as a fission product cleanup system, and meets the requirements of General Design Criteria 41, 42 and 43.

V. IMPLEMENTATION Except in tho m teses in which the applicant proposes an acceptable alterna-tive method Nr complying with the specified portions of the Commission's regulations, the methods described herein are to be used by the staff in its evaluation of conformance with Commission regulations.

l 6.'5.5-4 Rev. O

Implementation of the acceptance criteria of subsection II of this plan is as follows:

(a) Operating plants and OL applicants need not comply with the provision of this review plan section.

(b) CP applicants will be required to comply with the provisions of this revision.

VI. REFERENCES f

l 1. 10 CFR Part 50, Appendix A, General Design Criteria 41, " Containment Atmosphere Clean-up," 42, " Inspection of Containment Atmosphere Cleanup Systems," and 43, " Testing of Containment Atmosphere Cleanup System."

2. P.C. Owczarski, R.I. Shreck and A.K. Postma, " Technical Bases and Users

~

Manual for the Prototype of a Suppression Pool Aerosol Removal Code i

(SPARC)," NUREG/CR-3317, 1985.

3. P.C. Owczarski and W.K. Winegardner, " Capture of Iodine in Suppression Pools," 19th DOE /NRC Nuclear Air Cleaning Conference, Seattle, 1986.

4 4

1, i

i l

O Rev. O 6.5.5-5 i

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

sg 4

REGULATORY ANALYSIS OF THE REVIEW OF

! SUPPRESSION POOLS AS FISSION PRODUCT CLEANUP SYSTEMS

1. Statement of the Problem

~

Regulatory Guide 1.3, "Assurptions Used for Evaluating the Potential j Radiological Consequences of a Loss of Coolant Accident for Boiling Water Reactors," states, as Regulatory Position C.1.f, that "No credit is given i for retention of iodine in the suppression pool." Before the time this

! guide was first published, November 2, 1970, experiments had demonstrated the efficacy of suppression pcols in removing iodine in several chemical forms from air-steam mixtures. The adoption of Regulatory Position C.1.f.

I therefore, was deliberately conservative. Factors which may have .

f influenced its adoption are:

4 I (1) drywells are generally leaky, permitting significant bypass of the suppression pool.

l

! (2) suppression pool retention of fission products varies markedly l with conditions pertaining during the accident, and would have required more complicated inodels than any being used in 1970. l

{

(3) because of heavy reliance on standby gas treatment systems, suppression pool credit was not needed by boiling water reactors to j meet the dose guidelines of 10 CFR 100.

1 I

2 Standard Review Plan 6.5.3, " Fission Product Control Systems and Structures," contradicts Regulatory Guide 1.3 by stating that suppression pools may be considered as fission product control systems, although no guidance or reference is supplied as to methods to be used in their review. In NUREG-0979, supplement 4, " Safety Evaluation Report related to the final design approval of the GESSAR II BWR/6 Nuclear Island Design,"

~

the staff agreed to consider suppression pool retention in any application referencing the approved design. Revisions prompted by new source term information and the replacement of TID-14844 by more realistic accident assumptions will result also in the revision of Regulatory Guide 1.3.

Regardless of whether an accidental release is assumed using the current Regulatory Guide 1.3 or using the most modern methods, it is an undue conservatism to ignore the capability of t,he suppression pool to mitigate off-site dose consequences, provided that recognition of such capability does not degrade safety.

The effectiveness of suppression pools in retaining gaseous todine and particulate matter varies markedly with the conditions under which these materials are swept into the pool. While the overall effects of such variation can be calculated for any given postulated accident, this calculation would be uncertain in its predictions of the relevant conditions and would be very expensive to perform. It would be inappropriate to solve the problem of ignoring suppression pool effectiveness by replacing it with a required set of calculations that are O

3 x

impractical for use in assessing effectiveness. To avoid this further problem, the present proposal takes a narrow interpretation by replacing the undue conservatism of omitting credit in favor of moderately conser-vative simplifications.

t l

l 2. Objectives The objective of the proposed action is to establish the degree to which suppression pools can be considered as fission product cleanup systems and by revising the Standard Review Plan (SRP) to include procedures and criteria for suppression pool design evaluation.

3. Alternatives b

\g The existing SRP 6.5.3, " Fission Product Control Systems and Structures,"

in II.5, states that " Fission product retention credit assumed by the l

applicant for other systems, e.g., pressure suppression pools, may be acceptable provided that justification is supplied by the applicant."

This provision has been applied, so far, only in the review of t,he GESSAR-II application. The existing SRP, however, contains no precedures for reviewing pressure suppression pools. One alternative to the proposed new section, therefore, would be to continue to review pools on a-case-by-case basis. This course would not consistently apply computer code and model validation experiments which have been devised for pur-poses of developing a means of calculating pool retention of fission products.

4 SRP 6.5.3 could be revised to remove the statement allowing pool credit, and the GESSAR II SER could be amerded to retract the earlier position.

This course would remove the inconsistency between the SRP and Regulatory Guide but, in addition to ignoring the large volume of research data supporting pool credit, would provide an undue degree of conservation to the staff's review and be contrary to commission policy. (Goal 2.4, NUREG-0885, Policy and Planning Guidance, 1986)

The alternative selected is to propose an additional review plan section which would provide a consistent use of the available data without de-gradation of safety.

In proposing review procedures, two decisions were made concerning the means by which the review could be simplified.

(a) time-averaged decontamination factors (DF) were introduced, (b) minimum DF values were stated, such that only applications claiming larger DF's would require plant specific computer runs.

These decisions were prompted by practical considerations in conducting reviews; not taking the proposed course would have required great computer expense in any review. If a novel suppression pool feature were proposed, such as, for example, a chemical additive or increased submergence depth of the downcomers or quenchers, the Source Term Code Package computer codes could be run to quantify the effectiveness of the pool. Use of the Source Term Code Package costs about $25,000 per accident sequence.

I I

i 5

lO D The minimum DF values chosen are designed to be sufficiently small that no accident sequence is likely to be found to have a smaller time-averaged  !

value, even allowing a margin of safety for uncertainties.

4. Consequences i

l By resolving the contradiction between Regulatory Guide 1.3 and SRP 6.5.3 l in favor of the former, the staff would be denying a large body of j i

evidence proving the efficacy of suppression pools in reta1ning fission products. This might be defensible on the grounds of being conservative, l

but would not permit the realistic consideration of core melt accidents as they are currently being moc',eled to be used in licensing decisions.

By continuing the present situation, i.e., taking no action, the con-

! tradiction would remain. Licensees could request suppression pool credit, based on the staff's GESSAR 11 statement, but the staff would have no censistent guidance for performing the review, and would be reduced to j

either accepting or rejecting the licensees' submittal, or running the i source term code package repeatedly.

i The consequences of the proposed new section would be the effect that increased pool credit would have upon the efficiency required of other fission product control systems in order to meet the dose guidelines of 10 CFR Part 100. A licensee could request pool credit to justify a relaxation i of the maintenance and surveillance requirements placed on other systems.

1 O

i

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

6 l

Apart from the containment buildings themselves, the most important acci-dent off-site dose mitigation features of boiling water reactor plants under the SRP are the standby gas treatment systems (SGTS). These filtered exhaust systems are designed to have maximum effectiveness against the forms of fission product iodine assumed to be released by TID-14844. When reviewed against the fission product releases predicted by the new source term code package, however, suppression pools are capable of a high degree of retention of fission products. The proposed change will focus attention on sup-pression pools as dose mitigation features, and as a means of providing defense-in-depth in fission product mitigation capability.

The development of regulatory requirements for suppression pools might lead existing licensees to upgrade the quality of drywell penetrations, as part of measures to minimize pathways bypassing the pools. Drywell pene-trations are already subject to leak testing at each refueling under SRP 6.2.1.1.C.

At present, Regulatory Guide 1.3 assumes that 22.75% of the core iodine inventory as molecular iodine,1.25% as particulate and 1% as organic iodide are available for release from containment. Typical standby gas treatment systems (SGTS) serving BWR secondary containments as filtered exhaust systems are maintained at 99.% efficiency against all of these forms, i.e., after one-hundred-fold decontamination 0.25% of the core iodine is exhausted into the environment as the sum of the primary containment and main steam line isolation valve leak rates following a DBA-LOCA.

O

7 f

Against the same release to containment, the minimum decontamination factors in the proposed SRP 6.5.5 would reduce the 25% of the core iodine inventory available for primary containment leakage to 3.5%, assuming 10%

suppression pool bypass leakage. For 1% pool bypass leakage, 1.6% of the core iodine inventory would be computed as available.

Using the assumed release in Regulatory Guide 1.3, and obtaining sup-pression pool scrubbing credit with 10% bypass, a typical BWR could meet 10 CFR Part 100 thyroid dose guidelines and still reduce the plant SGTS efficiency from 99% to 95%. It should not, however, be assumed that by reducing bypass leakage and claiming suppression pool credit a licensee t

could greatly reduce the surveillance testing requirements of their SGTS.

Other design basis accidents, for example the fuel handling and instrument line break accidents, also require the use of the SGTS to meet the acceptance criteria of SRP 15.6.2, 15.6.5 Appendix B, and 15.7.4. 1 131 For a typical plant, the release of 3000 Ci of 1 would lead to dose I

consequences in excess of the guidelines of 10 CFR Part 100. This amount is equtvalent to only a few parts per million of the core inventory of iodine fission products. For a typical BWR, a million-fold reduction in iodine is mostly achieved by a low leakage containment (0.5% per day in 2 l

hours leaks 4X10~4) and to a lesser extent by the SGTS (10

• penetration by iodine). Since suppression pools are virtually useless against organic iodide, and since it is not feasible to eliminate bypass completely, i overall decontamination factors of more than about 15 cannot be practically achieved using the current iodine chemistry assumptions in Regulatory

)

l l

8 Guide 1.3, even if molecular and particulate iodine forms are totally absorbed by the pool.

A 15-fold reduction in SGTS effectiveness, i.e., from a penetration test of 1% or less to one of 15% or less, would reduce its organic iodide absorption efficiency from 99% to 85%. Unfortunately, the SRP section dealing with SGTS review, 6.5.1, states that systems requiring iodine absorption efficiencies of less than 90% may be reviewed under SRP 11.3.

Charcoal absorbers reviewed under SRP 11.3 may follow Regulatory Guide 1.140 rather than 1.52, and are not built or maintained to engineered safety teature standards. To prevent use of suppression pool credit to justify not maintaining and testing SGTS absorbers to Regulatory Guide 1.52 criteria, prior to revision of SRP 6.5.1, explicit mention of SGTS surveillance tests has been added as a criterion.

A typical SGTS contains about $20,000 of impregnated charcoal per train, with a comparable additional labor cost for renewing and testing if filter replacement is needed to pass a surveillance test. If the surveillance test criteria of a SGTS were relaxed, charcoal change-out would be required less often, perhaps reducing maintenance by of the order of 10 4 dollars per year.

It is also possible that a licensee might wish both suppression pool and maximum SGTS credits, while requesting an increase in allowable containment leakage. Again, a very large saving in the costs of containment integrated leak rate tests would not be expected, since the large degree of iodine fission product retention would not be associated

9 f

.f'Ni with any change in the postulated noble gas releases during a LOCA.

Licensees electing this course would be limited by the 10 CFR Part 100 guideline for whole body doses at the low population zone boundary over the course of the accident. For most BWRs, a doubling of the containment leak rate would bring the noble gas release consequences to the guideline, although for some plants having favorable meteorological parameters and large low population zones several-fold increases would still meet the guidelines.

While granting credit for suppression pool scrubbing, as proposed, would allow the deterministic licensing calculations of accident dose to be more easily met, the primary thrust of the change will be to allow greater BWR containment leak rates and more noncondensible accident l

fission products past SGTS filters. That is, existing BWR containment leak rates of about 0.5 volume percent per day maybe increased to as much as 5 volume percent per day, and 99 percent elemental iodine filter efficiencies maybe reduced to 90 percent. The change, therefore, may l result in increases in the quantities of fission products postulated to be released during design basis accidents. However, regulatory guidelines would still be met, and the change in risk is expected to be very small since the bulk of public risk is attributed to accidents in which the containment fails or is bypassed (i.e., severe accidents not designbasisaccidents).

O

10

5. Decision Rationale Strategic goal 2.4 of the NRC Policy and Planning Guidance, 1986, lists as objectives the completion of the reassessment of source terms and the implementation of appropriate revisions in staff practices. The source term revisions will involve many related changes to the SRP and regulatory guidance, and may also include rulemaking and revision of existing regulations. The proposed action is perceived as an early step in this process, since it will put in place the review procedures and criteria necessary for considering the mitigation of new source terms by suppression pools.

The proposed section is equally applicable to the source term assumptions contained in Regulatory Guide 1.3 and to the fission product releases cal-culated by the Source Term Code Package. For both applications the pro-posed acticn offers the following advantages:

1.) Suppression pool fission product retention can be assumed to be described by conservatively chosen decontamination tactors. The use of these factors avoids the large expense of computer analysis needed to quantify suppression pool response using the available computer codes. As discussed earlier, very large decontamination factors can be calculated, but the net effective decontamination achievable is limited by the possibility of pool bypass leakage.

O

11 (O

') 2.) Existing plants have the possibility of reducing maintenance costs for their charcoal absorbers by being able to retain the absorbent for longer periods of time between changes.

The cost savings of these advantages would vary with the degree to which licensees and aplicants elected to claim suppression pool fission product cleanup credit, and the number and diversity of j accident sequences necessary to represent the effectiveness of the pool.

While releases of fission products as assumed in Regulatory Guide 1.3 i are effectively reduced by filtered exhaust systems, the releases calculated for many accident sequences by the Source Term Ccde Package are more effectively reduced by s'uppression pool scrubbing.

By adding guidance for the review of suppression pools as fission product cleanup systems in the form proposed, conservative but appropriately realistic credit would be assessed without significant loss of the safety afforded by existing filtered exhaust systems. l

6. Implementation The proposed action requires no acti.on of existing licensees, except as they might voluntarily elect to reanalyze the accident consequences and submit an.FSAR amendment to reduce reported iodine doses. This action would take effect upon publication of the proposed revision.

O

APPENDIX A TECHNICAL REPORT A-3788 8-1-86 EFFECTIVENESS OF BWR PRES $URE SUPPRESSION POOLS IN RETAINING FI$510N PR000 CTS H. P. Nourbakhsh, R. Davis, and M. Khatib-Rahbar Accident Analysis Group Department of Nuclear Energy Brookhaven National Laboratory Upton, New York 11973 August, 1986 Prepared for U.S. Nuclear Regulatory Commission Washington, DC 20555 Contract No. DE- ACO2-76CH00016 FIN No. FIN A-3788

O ABSTRACT The effectiveness of BWR suppression pools in retaining fission products released during severe accidents is assessed. Scrubbing models are reviewed and sensitivities to input parameters of SPARC Computer Code used in' Source Term Code Package (STCP) are also discussed. An assessment of the effective decontamination f actors for the suppression pools based on the results of recent STCP calculations performed by BNL and BCL is also presented.

1 ACKNOWLEDGEMENTS The authors are grateful to W. T. Pratt (BNL), J. Read, L.* fof fer, and

2. Rosztocsy (USNRC) for their review and many helpfd remarks on th13 manu-script. The work reported herein was conducted under the auspices of the United States Nuclear Regulatory Comission (USNRC), Of fice of Nuclear Reactor Regulation.

A-2 0

CONTENf5 ss Page t

s 1

1. INTRODUCTION........................................................

3

2. DESCRIPTION OF PRESSURE SUPPRESSION P00LS...........................

7

3. PREDICTIVE METH005..................................................

3.1 SPARC (Suppression Pool Aerosol Removal Code) .... . ..... .. .. .. . . 7 3.2 Experime ntal Val idation of the SPARC Code. . . . . . . . . . . . . . . . . . . . . . 8 3.3 SPA RC Se n s i t i v i t y An al y s e s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 3.4 Sol ubl e . Ga s Sc rub b i n g . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 17 4 EFFECTIVE DECONTAMINATION FACTORS FOR THE SUPPRESSION P00L..........

17 4.1 Peach Bottom (Mark 1)..........................................

j 4.2 G r a n d Gu l f ( Ma r k ! ! ! ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

5.

SUMMARY

AND CONCLUSIONS ............................................ 23

6. REFERENCES.......................................................... 25 LIST OF FIGURES Page figure Title BWR Ma r k I cont ai nme nt sy s t em. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1

5 2 BWR Mark !! containment system.................................

3 BWR Ma rk !! ! cont a i nme nt sys t em. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 I

4 Comparison of SPARC calculated decontamination factors with BCL experimental values for 1/2 in diameter horizontal injector....................................................... 9 5 Variation in bubble aspect ratio with mean bubble diameter..... 11 1 6 Ef fect of bubble diameter and aspect ratio on 0F............... 12 7 E f f ect of bubbl e swa rm ri se vel oci ty on 0F . . . . . . . . . . . . . . . . . . . . . 15 l I

1 l

4

)

I A-3 J

J 4

t

. . _ _ _ , , . _ . , . ,-r ---m. .,_y _ . - -

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

O LIST OF TABLES Tabl e Title Page 1 The Input Parameter Values to SPARC (Calculated by Precedin Codes in the STCP..........................................g.... 13 2 Effective (Time Averaged) In-Vessel Release Decontamination Factors for the Suppression Pool (Peach Bottom Mark !)......... 18 3 Ef fective (Time Averaged) In-Vessel Release Decontamination Factors for the Suppression Pool (Grand Gulf Mark !!!)......... 21 4 Effective (Time Averaged) Ex-Vessel Release Occontamination Factors for the Suppression Pool (Grand Gulf Mark !!!)......... 21 A-4 O

1

1 i.

1. INTRODUCTION The radiological source terms resulting from postulated severe reactor To accidents have important implications regarding health and publje . risk.

i assess the radiological consequences of reactor accidents, an evaluation must I be made of the quantities and characteristics of releases of radionuclides from the fuel pins to the environment.

The fission product release and transport is strongly influenced among other things by reactor type, containment design and the engineered safety features.

In a boiling water reactor (BWR), the pressure suppression pool is de.

f In most accident sequences involving I

signed to serve as a passive heat sink.

j* severe core damage, soluble gases and aerosol-laden gases vent through the suppression pool prior to escape to the outer containment building. The pas-sage of these materials (gases, vapors, and particulate materials) through the J water in the pool results in the removal of certain fission products.

l This report presents information for the mitigative potential of pressure l

suppression pools in order to develop a technical basis for changes in reguia-tory requirements for such engineered safety features.

l I

I l

i 4

i i

i c

t i

i

{

' A-5 i

}

i

3-

2. DESCRIPTION OF PRESSURE SUPPRESSION POOLS Tne pressure suppression pool is primarily designed to reduce the primary containment pressure following a design basis accident. The thr.ee. basic BWR containment designs (Mark 1, 11, and 111) are illustrated in Figures 1 through

3. These three types of designs are similar in concept.

The Mark i design has a separate toroidal pool (wetwell) that is con-nected to the main part of containment (drywell) by large vent pipes. Typi-cally, the suppression pools contain approximately 120,000 f3 t of water. The torus containing the water has a major diameter of about 110 feet and a minor diameter of 30 ft. Ducts several feet in diameter connect the drywell to the ,

wetwell torus. The large ducts branch through a vent header into nultiple (typically 2-ft-diameter) downcomers that have their open lower ends submerged in the water. Steam can also be directed into the pool by separate lines from

• the safety / relief valves on the reactor's primary system.

The Mark !! design is called the "over-under" design because the drywell is located directly' above the wetwell . Steam released during an accident to the drywell is conveyed into the suppression pool by multiple vertical steel downcomer pipes. The downcomers penetrate the diaphragm floor separating the drywell and the wetwell .

Vent valves in the floor allow free flow from the top of the wetwell back into'tfie"drywell .

In the Mark !!! design the wetwell is an annular region at the periphery of the containment. Water is retained by the weir wall (height approximately 20 f t) and steam discharges into the pool from the drywell through submerged, horizontal vents in the lower drywell wall in the event of a steam system rup-ture in the drywell . The safety relief valves on the primary reactor system discharge directly into large pipe headers that terminate at spargers sub-merged in the suppression pool. The suppression pool volume is typically about 160,000 f3 t , similar to a Mark !! plant.

A-6 O

t j -4

\

.I I

1 i I

i .  !

l , SHIELD PLUGS TOP OF SHIELD i WALL I i

. . F r:

' ". .:U...:

I. W ,

qQ..%.. ,

E ,' f

"[ ^\ ~<.* , REACTOR VESSEL e

i 4

c.

- 'g i ,

4 1

DRYWELL ,

l  ?

i

• DRYWELL SUPPORT I - ,'

• j 1,.

.I

'ORYWELL I, SKIRT L ., E E ..

. 1 l CgrVENT VENT -

] k Q

l q),. ..

h .. .*a PRESSURE PRESS'URE SUPPRES$10N gyppgggsion

CHAMBER poot 1

(TORUS) t I

i i

l i

l Figure 1 BWR Mark I containment System.

\

A-7 i

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

5 O

L:. e ,' f, . , ~*

r ,* .'.

I '," ' \ : '- REACTOR -

<. . VESSEL DRYWELL

\

s%

\ /

I

' )l t

l ~ '

m-- ',,.*.

• ' ' . . - -, .i VENT PRESSURE f .';' SUPPRESSION f CHAMBER

= ~

_~x-q -

~

= s.~h. .'

= -: + :-_-: , W -3: e ~

e

= ~2hMg

,~

Eg SUPPRESSION POOL ll*'.* **l l

Figure 2 BWR Mark 11 containment system.

l A8 l

i I

i l

l i

i 5 TEEL CONTAINMENT

\ / SHIELD BUILDING DRYWELL HEAD f  :

5 t-1 U PERCONTAINMENT p

n g- '. -

1'F - REACTOR

', g c .

.- -- 7. l, ,

,i l

I  !

Is a:: BIOLOGICAL

~

SHIELO WALL DRYWELL  :

-- DRYWELL WALL i .

A $ '

i

-- WEIR WALL i

- ~. =_ _ _ ll

21. Kg "

j -SUPPRESSION POOL g ,

,, - HORIZONTAL VENTS i

l l

i I i i  ;

I l I

figure 3 BWR Mark 111 containment system.

l

}

A9 i

l i

i T

7-

3. PREDICTIVE METH005 Several models have been developed for predicting aerosol scrubbing ef f t-ciencies in BWR suppression pools. The Fuchs' model of particle removal from single spherical bubbles 2 is the basis of all particle scrubbing models and codes curr{ntly in use for nuclear reactor analysis. These models include; (1) SPARC. developed under NRC sponsorship by Battelle Pacific Northwdst Lab-oratory, (2) SUPRA,' developed developed by General Electric.,under EPRI sponsorship by SAIC and (3) a model Several models have also been developed for scrubbing efficiency of solu-ble gases. Diffey et al.6 proposed a model for scrubbing efficiency of ele.

mental iodine based on the assumption that iodine in the gases leaving the pool is in equilibrium with iodine in the water pool. The experimental mes-surement reported by giffey et al. seems to support the plausibility of their model. Devell et al. carried out experiments with 1 2 in water at 100'C and concluded that todine in gas bubbles did not necessarily reach equilibrium with todine in the liquid and thus, extended the Dif fey et al. model to account for the degree of saturation. SUPRA also includes models for scrub-bing gaseous fission products. More recently, models for elemental and organic todine scrubbing have also been added to the SPARC code.a In the following sections, the SPARC (as used in the STCP) aerosol scrub.

bing model will be discussed and the code results will be used to illustrate the variation in scrubbing decontamination factors over a range of input parameters selected to reflect the current uncertainty in their values.

3.1 SPARC (Suppression Pool Aerosol Removal Code)

The SPARC computer code 2 has been developed to calculate the behavior of aerosol particles in the pressure suppression pool under conditions that may be predicted to result from a postulated accident. The code calculates the scrubbing of the aerosol particles from the gas mixture bubbling through the pressure suppression pool. This calculation is handled in terms of a decon-tamination f actor (DF) per particle size.

Fuchs' model of particle removal from single, spherical bubbles is the  !

basis of particle scrubbing models used in the SPARC code. In the Fuchs' model, the dominant scrubbing processes take place inside rising bubbles.

This model identifies three mechanisms of particle removal. They aret ,

1

1. Brownian diffusion of particles to the bubble wall (dominant for smaller particles).

1

2. Gravitational settling of particlgt to the lower bubble wall '

(dominant for larger particles).

3. Inertial deposition of particles on the bubble wall driven by the centrifugal acceleration produced in the internal circulation of the gas in the bubble (doininant for larger particles).

A-10 O

I l

8-l O In addition to the three removal mechanisms modeled by Fuchs, the SPARC code considers the following additional mechanism.

. Steam condensations (no particle size dependence assumed).

. Convection caused by vapor flux to or f rom the bubble walls. The con.

vection velocity is added algebraically to the deposition velocities calculated for other deposition mechanisms.

. Inlet impaction during gas injection into the pool.

. Particle growth in the bubble from water acquisition by deliquescing material in the particles. This is not specifically a removal mechan.

ism, but it will enhance removal of larger particles by larger parti.

- cle dominant mechanisms and degrade removal of smaller particles by small particle dominant mechanisms.

3.2 Experimental Validation of the SPARC Code The phenomenological models included in the $PARC code are well supported by separate ef f acts testing as found in the extensive literature on bubble dynamics and mass transfer between rising bubbles and liquid media.

Experimental studies of pool scrubbing have been conducted at Battelle Columbus Laboratories. The available data base consists of particle scrubbin measurements taken in a pool using a 0.5 in diameter horizontal injector.g The following conditions were varied during 56 dif ferent experiments: inert gas composition (air or helium), steam composition, gas flow rate, injector

( depth, pool temperature (ambient or near boiling), and aerosol (Cst. Te02 . Or

$n), sire, solubtitty, density, and aerosol concentration.

Decontamination factor (DF) measurements for each experiment consist of the time. integrated particle mass flow rate into the pool divided by time.

integrated particle mass flow rate out of the pool. Figure 4 presents a com-parison of experimental values and calculations by the $PARC code as used in the $fCP. These comparisons correspond to an underprediction by $PARC by an average factor of 6.2.'

3.3 $PARC Sensitivity Analyses l

The sensitivity study of the $PARC code involves variations in the fol.  !

lowing irmortant input parameters: l

. Particle size of aerosols borne through the pool by gases

• The site of gas bubbles passing through the pool (OlAM)

+ 1he aspect ratio of the gas bubbles (RAfl0)

. The swarm rise velocity of the gas bubbles (V5 WARM)

• The volume fraction of steam in inlet gas O A-11 U

9 O

A 10' .

1

o o o

. o di 10 --

S,

u. l Lo O  : o t .

? 2 o

g10 _, o o

m  :

e o 2  : <

't 10' d t 4"

t o h

t, ;f!f i

10 ' ' '' l ' ' "I ' "I 10 10' 10 8 10 10' Calculated DF Figure 4 Comparison of $PMC calculated decontamination factors with BCt experimental values for 1/2 in diameter horizontal injector, A-12

i

-10 There are other less important input parameters to SPARC such as pool (O) v temperature, pool depth and percent of soluble material in particles.

Aerosol particle size is a parameter obtained from the result of calcula.

tions with the VANESA and TRAPMERGE models of Source Term Code Package (STCP). ThesensitivityoftheSPARCanalysestothisparamete{o is reduced as The volume the breadth of the particle size distribution is increased.

fraction of steam in inlet gas is a sequence-dependent quantity calculated by the MARCH code, in this section decontamination curves calculated by SPARC will be presented to illustrate the importance of the user input parameters, namely, the bubble size, the bubble shape, and the bubble rise velocity. T r.e parameter ranges for these variables that are chosen reflect a reasonable range of uncertainty.

BCL.1{xperimental studies of gas liquid hydrodynamics have been condu steam) and noncondensable (air, helium, or hydrogen) gases injected into water pools through single hole and multihole configurations typical of those found in BWR quencher pipes. In an actual accident situation, swarms of bubbles i

.' rather than single ' bubbles will be encountered. The bubble size in these swarms is a distribution. The bubble size distribution has been found to be inoependent of the injection flow rate and injection angle. There is, how.

ever, a dependence on condensable steam fraction. The distribution is well 3

described by a lognormal distribution with mean diameters of 0.55 cm and 0.35 cm for low and high steam volume fraction respectively, and with a constant i standard deviation of 1.5. The bubble diameters selected for the sensitivity study are 0.3 to 0.9 cm to reflect the range of uncertahty associated with c

this parameter.

Aspect ratio and bubble diameter are related, the larger bubble being more elliptical . Tnts relationship also depends on water purity. Figure 5 shows the aspect ratto of bubbles as a functio of their equivalent spherical diameter. Two correlations from Clif t et al. 2 are given.

One is for pure water and the other for contaminated water. Impurity levels of parts per mil.

lion range are suf ficient to produce more nearly spherical bubbles. Figure 5 also shows a correlation developed by BCL based on their experimental resul ts . The aspect ratio selected is based on the Clift correlation for con.

taminated systems.

The resulting SPARC sensitivity to bubble diameter and aspect ratio is presented in Fig. 6. The important input parameters for these cases are pre.

sented in Table 1. The input parameters to SPARC which are calculated by preceding codes in the STCP are taken from a recent BNL calculation for a typical time frame in the Peach Bottom TC2 sequence (TIME = 90 min).

The bubble swarm rise velocity d'etermines the residence time for scrubbing. For a single bubble rising in an infinite pool, terminal veloci measurementsforalargenumberofgas-liquidsystemshavebeenperformed.g For an air bubble rising in a stagnant water pool, the experimental data reported by Haberman and Morton,' shows that the terminal rise velocity is nearly constant at about 0.24 m/s for bubbles with equivalent diameters between 2 and 20 mm. For a swarm of bubbles the drag force between the bub.

i bles and the surrounding itquid will create significant circulation current in

< the liquid. Inside the bubble column the rising gas bubbles pump liquid from A-13 i

l I - - _ . . _ _ - _ _ . . . _ _ . . _ _ _ . __ _

O d # * * * , ' ' .

n 4.0 -

M w -

M PURE e 3.5 -

LIMIT _

3 -

\

o _

a.

o 3.0 -

n M

CONTAMin4rgo 32.s -

100' C =

y o

g 2.0 -. \ '

8 -

h 1.s -

5 - _- /

= , '

1.0 - BCL Correlation f s ,

' 0 05 os o,7 ,, o,9 ,o '

De (cm)

M gure S Varf atfon in bubble a5pect ratto w(th mean bubble diameter, A-14 O

- 12-i l

i i

t i

i i

i i

'o . .

I DIRN RATIO 0.5 '5. _

1 o.3 . 2 b... .

.. .....M....

.... A d .... ....,

i ,

b  ! ,

J -

I d-I l 5

l l H t

! E

• i I

1 \ l 1

1

~o .

~~

\

\'*'

,/ /

/ l' l

,,/ ,

/

i. .,'.. ' l

.. ............ =',*.=,

1 l

t

}

i ,

i 1 .

1 o . ,

..,1 10., 10

] "" 10'8 10

) PARTICLC Ol0 METER INICRONI l I

i i 4 r J

l L

l Figure 6 (ffect of bubble diameter and aspect ratto on DF. ,

1 I F A-15

'l 4

)

4 I r

(

{

l I

-w-- m we g-TTT'N'hWNC""'F M8M*WY N S WN * ' 7'W'*-~~D CCF *N"W'*'W-"NY""'"W W T7"FF"WT"VYY'W"~"'""*- -*---T""C'*-*7 wm- " - -

Table 1 The Input Parameter Values to SPARC (Calculated O

by Preceding Codes in the STCP)

TIME = 90 min TIME = 116 min Pool Temperature (*C) 113. 115 Pressure Above Pool (ATM) 4.75 5.58 Inlet Gas Flow Rate (G/SEC)

HO2 1.27E+4 4.03E+4 H 5.12E+2 g 1.06E+2

. O.

C0 0, Air 0. O Inlet Gas Temperature ('C) 517 451 Inlet Gas Pressure (agut MI6h 49.5 61.

Particle Material Density (G/CC) 3. 3.

Percent Soluble Material 0.35 0.37 A-16 O

A k' the bottom of the pool to the surf ace. The local liquid velocity inside the bubble column increases the bubble rise velocity relative to a stationary observer outside the pool. In the BCL experiments.II the bubble rise veloci-ties have been measured relative to a stationary observer outside the tank for different gas injection rates. The typical spatial distributton- of bubble rise velocities in the bubble column varies between 20 cm/sec at the outer edge of the bubble column to 100 cm/sec at the centerline of th6 bubble col umn. The values ranging from 20 to 116 cm have been chosen for the purpose of the present analyses. The higher value of 116 cm/sec was chosen bec it corresponds to the value used in the BNL and the BCL STCP calculations.qg8 .

The result of sensitivity of SPARC with respect to bubble swarn rise velocity is presented in Fig. 7. The important input parameters for these cases are also presented in Table 1 and are based on BNL STCP calculations of Peach Bottom TC2 Sequence (TIME = 116 min).

The various uncertainties identified in the sensitivity study of the SPARC model are estimated to lead to an order of magnitude uncertainty in the pool decontaminatio.n. ThisconclusionisconsistentwiththeOVESTstgyfor the Grand Gulf TC sequence performed by Sandia National Laboratory. . It

! should be noted that the SPARC underpredicts the DF values due to both unmodeled phenomena such as fragmentation and coalescence of bubbles as well as uncertainties associated with the code input parameters. As indicated earlier, the DF values in the BNL and BCL recent STCP calculations are similar to the lower bound estimates of the present sensitivity study.

3.4 Soluble Gas Scrubbing V Mechanistic models for elemental and organic iodine scrubbing have been A good comparison between avail-added recently to the SPARC computer code.

able experimental data and the SPARC prediction has been observed.a An inte-gral decontamination factor of the order of 7000 for glgental Dueiodine (1 2sol to high ) has u-been calculated for the Peach Bottom TC1 sequence.

• bility of H! in water (relative to 1 ),2 a higher integral decontamination factor for hydrogen todide is expected. An exact quantification of pool scrubbing efficiency for various soluble gases requires detailed calculations using any of the available models discussed previously.

A-17 v

O

'a . .

VSWARM

50. CN/SEC  :

116. CM/5LG I o .

'. '.2.Q;.'C.nl.1C9.:. .

b .

e '

R '

. l 8

w '.'.  :

s '

I Wo:

e- '.

E . '

E '

e 8

o u

Q "o . .

~

o , ,

i 10 10' 10 10'  !

PARTICLE DIANOTER INICRON) i 1

Figure 7 Effect of bubble swarm rise velocity on DF, A-18

4. EFFECTIVE DECONTAMINATION FACTORS FOR THE SUPPRESSION POOL

} One of the major considerations in predicting the scrubbing effectiveness

' of suppression pools is the definition of the environment and conditions that could conceivably challenge the pool. This section of the report presents a sequence-based assessment of the effective decontamination factors for the BWR suppression pools. The information is based on the results pf Sour'ce Term 16 and BCL1 ,18 for Peach Code Package (STCP) calculations performed by BNL Bottom (Mark 1) and Grand Gulf (Mark Ill) plant subject to a postulated severe

, accident condition.

4.1 Peach Bottom (Mark I)

Peach Bottom Unit 2 Power Plant was included in the 8CL and BNL STCP radionuclide release calculations. Peach Bottom, which is a General Electric BWR 4/ Mark I design, has been in operation since early 1970. The accident sequences selected for BCL detailed source term analysis includes of (1) TC, l an anticipated transient without scram, (2) TB, a station blackout scenario, l

and (3) V, an interfacing system LOCA sequence. These sequences were selected on the basis of preliminary ASEP results on accident sequence probabilities as well as preliminary SARRP containment event tree quantification. In this sec-tion, the results of Source Term Code Package Calculations r one variation of TC (TC2) sequence and one variation of TB (TB1) sequence ' used to assess the effective decontamination factors for the BWR/ Mark I suppression pools. ,

l '

l In the TC2 sequence the failure to scram is accompanied by the failure to

achieve early power reduction as well as the failure to achieve emergency depressurization. The primary coolant inventory is maintained by the combina-tion of the HPCI, RCIC, and the CRD systems. As the suppression pool heats up I

due to the continuing large steam input through the safety / relief valves, failure of the safety systems could take place due to loss of lubrication oil cooling, seal overheating, etc. In the present analysis the HPCI was assumed l

to fail at a suppression pool temperature of 200*F, and the RCIC was assumed

to fail at a containment pressure of 25 psia, due to high turbine exhaust back i pressure. The CRD system, which takes its suction from the condensate storage

' tank, would continue to operate as long as the water in the latter was available. The CR0 flow is insufficient to keep the core covered and cooled, and eventual core melting would take place. The containment would be intact i during the initial core melting in this sequence, but would fail shortly after l the reactor vessel f ailure.

I i

For the TB1 scenario loss of all off-site and on-site AC power leads to l

the loss of all active engineered safety features except the steam powered i emergency core cooling systems. The latter, however, require DC power for operation and would fail when the station' batteries are depleted; the latter has been estimated at six hours after the start of the accident. In such an l

event, core uncovery and melting takes place with the containment initially

intact; containment 4eMuse is assumed to f ail late in the accident sequence.

In both sequences considered, the in-vessel fission product release due to core degradation and melting which consists primarily of Cst, Cs08, and Te are free to pass down the safety relief lines and into the suppression pool through the quenchers, and these are subject to pool scrubbing. Table 2 pre-sents the implied decontamination f actor (DF) for the in-vessel phase, for l

Ib

! A-19 t

O Table 2 Effective (Time Averaged) In-Vessel Release Decontapiqation Factors for the Suppression Pool (Peach Bottom Park 1)

In-Vessel DF Fission Product Group TC2 Sequence TB1 Sequence CSI 200 Large=

CSOH 300 Large Te 250 Large

• Due to inconsistencies in the reported values, an exact quantification is not possible.

O b

A-20

1 both BNL TC2 and BCL TB1 calculated sequences. The variation of DF with fis-sion product species is due to the f act that the various species are released at dif ferent times and thus experience different conditions in the pool.

In the TC2 sequence, the containment was assumed to fail at tTie time of pressure vessel f ailure. This ensures that the fission products released dur-ing core / concrete interactions can bypass the suppression pool . However, in the TB1 sequence, it is assumed that the containment failure occurs late and therefore most of the Te and refractory fission products released ex-vessel are passed through the suppression pool. These fission products consist pri-marily of Ba, Sr, La, and Ce with lesser quantities of Te (a proportionately larger fraction of the puff release at the time of pressure vessel failure is Te). In this case, the DFs corresponding to the ex-vessel release phase were found to be negligible as comp; red with DFs for the in-vessel release phase.

The BCL STCP results for another variation of TC, TC3, which is identical to TC2 except for inclusion of containment venting, was also studied. With venting all the releases pass through the suppression pool but due to incon-sistencies in the reported values, no quantification of ex-vessel release decontamination factors was possible at this time.

4.2 Grand Gulf (Mark III)

Selected severe accident scenarios for the Grand Gulf Unit 1 Power Plant were included in the BCL STCP radionuclide release calculations. Grand Gulf Unit 1, which is a General Electric BWR 6 with Mark III containment, began operations in June 1982. The accident sequences selected for BCL detailed O

source term analysis consists of (1) TC, an anticipated transient with scram, l and (2) TB, a station blackout scenario. These sequences were selected on the  !

basis of preliminary ASEP results on accident sequence probabilities as well as preliminary SARRP containment event tree quantification.

For the TC sequence, the containment was assumed to fail by overpressuri-zation prior to core melting due to the elevated power input to the suppres-sion pool associated with the failure to scram; containment failure was (

assumed to lead to failure of the emergency core cooling system pumps. It was l also assumed that the Automatic Depressurization System (ADS) would be acti- .

l vated prior to containment failure and subsequent core uncovery. In the analysis of the containment response, nominal leakage between the drywell and the outer containment bypassing the suppression pool was assumed.

Two variations of the station blackout (TB) scenario were considered. In the first, late containment failure was considered and in the second, the con-tainment was assumed to fail at the time of reactor vessel failure. With the complete loss of electric power in this sequence, all the active engineered safety systems, with the ext.eption of the steam turbine driven emergency core cooling systems, would be unavailable. The turbine driven pumps would operate as long as the station batteries were available. The latter were assumed' to be depleted at six hours after the start of the accident. Also, in the absence of electric power, the ADS, upper pool dump, and the hydrogen igniters would not be able to perform their functions. Thus, core overheating and melting would take place with the primary system at elevated pressure. For the late containment failure variation of the station blackout sequence, nomi-nal leakage between the drywell and the outer containment was assumed.

A-21 l'

- - - - _ _ - - - _- - -- , ,- e - ~ - - -

Containment failure in this case would be expected due to the buildup of non-condensables during the attack of the concrete foundation by the core debris.

For the early containment failure variation of the station blackout scenario, containment failure was assumed to occur immediately after reactor vessel failure due to a large hydrogen burn. The expulsion of the hot, core debris from the primary system is the obvious ignition source. In the analysis of this scenario, a large leakage between the drywell and the outer containment was assumed af ter vessel and containment failure; this implies some degrada-tion of the boundary between the drywell and containment due to the events associated with primary system failure or the hydrogen burn.  ;

1 In all three sequences considered, the in-vessel fission product release '

due to core degradation and melting, primarily of Cs!, Cs0H, and Te, are sub-ject to pool scrubbing. Table 3 presents the implied decontamination factor (DF) for the in-vessel release phase for the three calculated sequences.

In the TC and TB2 sequence, the fission product released during core / con-crete interactions can bypass the suppression pool. However, in TB1 sequence it is assumed that most of the Te and refractory fission pr.oducts during the ex-vessel release phase pass through the suppression pool . These fission pro-ducts consist primarily of Ba, Sr, La, and Ce with lesser quantities of Te (a proportionately larger fraction of Puf f release at the time of pressure vessel failure is Te due to its later release time, during the melt release phase).

Table 4 presents the effective ex-vessel release decontamination factors for TB1 sequence.

The DFs corresponding to the ex-vessel release phase are smaller than the DFs for the in-vessel release phase because:

1) The gases evolved ex-vessel ccntains less condensable gas (steam)

2) The pool temperature is highar later in the accident sequence

3) The ex-vessel aerosol particle sizes are smaller

4) The depth of the suppression pool during th release under water is smaller for the ex-vessel release.

The values for in-vessel release decontamination factors for the TC sequence shown in Table 3 are of.l0the same order of magnitude as the lower bound estimates in the QUEST study for the Grand Gulf TC sequence performed by Sandia National Laboratories. (In-vessel release DF values for Cesium, Iodine and Tellurium reported in the QUEST study is 111). The lower bound ex-vessel release DF value for Tellurium reported in the QUEST study is 5 com-pared to 10 obtained in the present study.

A-22 9

4 Table 3 Effective (Time Averaged) In-Vessel Release Decontamination Factors for the Suppression Pool (Grand Gulf Mark III) ,

Fission Product Group TC Sequence TB1 Sequence TB2 Sequence Csl 85 50 60 Cs0H 80 55 65

~

Te 40 40 50 i

Table 4 Effective (Time Averaged) Ex-Vessel Release Decontamination Factors for the Suppression Pool (Grand Gulf Mark !!!)

Fission Product Group 6-Vessel 0F Sr 25 1

Ba 20 La 15 i Ce 30 Te 10 f

i A-23 O

P

___,_.. __.___,_. -__.._ _.. ___.--- ~. ..___. -_ ~ .__ _ ______-- _--

5.

SUMMARY

AND CONCLUSIONS Tne scrubbing models and sensitivity to input parameters of SPARC com-puter code used in Source Term Code Package (STCP) have been discussed. The various uncertainties identified in the sensitivity study of SPARC ~model were estimated to lead to an order of magnitude uncertainty in the decontan)ination factor by the suppression pool .

An assessment of the effective decontamination factors for the suppres-sion pools based on the results of Source Term Code Package (STCP) calcula-tions performed by BNL and BCL has also been presented. The DF values in these calculations correspond to the lower bound estimates of the sensitivity ~

study. It is seen that variation in pool decontamination factors are a func-tions of sequence and system being considered and the DFs corresponding to the ex-vessel release phase are smaller than the DFs for the in-vessel release phase.

O O

A-24 l

l

/3

6. REFERENCES

1. " Technical Basis for Estimating Product Behavior During LWR Accidents,"

NUREG-0772, U. S. Nuclear Regulatory Commission, Office of Nuclear Reactor Regulation,1981.

2. N. A. Fuchs, "The Mechanics of Aerosols," Pergammon Press, New York, 1964, p. 240.

3. P. C. Owczarski, R. I. Schreck, and A. K. Postma, " Technical Bases and User's Manual for the Prototype of a Suppression Pool Aerosol Removal Code (SPARC), NUREG/CR-3317, PNL-4742, 1985.

4. A. T. Wassel et al ., " Analysis of Radionuclide Retention in Water Pools,"

J. of Nuclear Engineering and Design, Vol . 90, pp.87-104,1985.

5. F. J. Moody, " Derivation of an Elliptical Supprassion Pool Scrubbing Model," General Electric Company,1983.

6. H. R. Dif fey et al ., " Iodine Clean-Up in a Steam Suppression System,"

AERE-R-4882, UKAEA, Harwell, United Kingdom,1965.

7. L. Devell et al ., " Trapping of Iodine in Water Pools at 100"C," Pro--

ceedings of IAEA Symposium on Containment and Siting of Nuclear Power Plants, CONF-67042, 1967.

i 8. P. C. Owczarski, Private Communication, August 6, 1986.

9. P. C. Owczarski and W. K. Winegardner, " Validation of SPARC, A Suppres-sion Pool Aerosol Capture Model ." Proceedings of an International Sym-posium on Source Term Evaluation for Accident Conditions, Columbus, Ohio, October 28 - November 1,1985.

10., P. K. Mast et al. " Uncertainty in Radionuclide Release Under Specific LWR Accident Conditions," Volume IV TC Analysis, SAND 84-0410, December 1985.

11. D. D. Paul , et al . , "Radionuclide Scrubbing in Water Pools Volume 1:

Gas-Liquid Hydrodynamics," EPRI NP-4154, August 1985. i

12. R. Clift, J. R. Grace, and M. E. Weber, Bubbles, Drops, and Particles, Academic Press, New York, 1978.

13. G. B. Wallis, "Tne Terminal Speed of Single Drops or Bubbles on an Infi-l nite Medium," Int. J. Multiphase Flow, Vol .1,1974, pp. 491-511.

14. W. L. Haberman and R. K. Morton, "The David Taylor Model Basin," Report No. 802, 55715-102, 1953. .

15. P. C. Owczarski and W. K. Winegardner, " Capture of Iodine in Suppression Pools " To be presented at the 19th DOE /NRC Nuclear Air Cleaning Confer-ence, Seattle, Washington, August 19, 1986.

f O

A-25 O

16. M. Khatib-Rahbar et al . , " Independent Verification of Radionuclide Release Calculations for Selected Accident Scenarios," NUREG/CR-4629, BNL/NUREG-51998, July 1986.

17. R. S. Denning et al . , "Radionuclide Release Calculations

• fo~r Selected Severe Accident Scenarios, Vol.1 BWR, Mark I Design," Battelle Columbus Laboratories, Draft, November 1985.

18. R. S. Denning et al ., "Radionuclide Release Calculations for Selected Severe Accident Scenarios, Vol. IV, BWR, Mark !!! Design," Battelle Columbus Laboratories, Draf t, February 1986. ,

O t

t f 1

• l 4

e l

A-26

\ /

NRC corm 33s U. NUCLE AR REGULATORY COMMISSION 87 778 NUREG-0800 j B LIOGRAPHIC DATA SHEET Section 6.5.5. Rev. 0

( "/

4. TITLE AND SURTITLE s4da u,ne No., et aparwris tel Standard Review Plan for the Review of Safety Analysis

2. (Leave blmkl /

/

Reports for Nuclear wer Plants LWR Edition, Proposed 3. RECIPIE ACCESSION NO.

Revision 0 to New SRP Sect,1on 6. 5 "Presgur(e Suopression Pools as Fission Produ t Clean-U Systent For Comment) /

7. AUTHOR (SI 5. DAT( REPORT COMPLETED MONTH l YEAR Mtrch 1987

9. PERFORMING ORGANIZAlsON N AME A O M AILING ADORESS (sactode Zip codel DATE REPORT ISSUED Office of Nuclear Reactor egulation ppy l '*1$7 U.S. Nuclear Regulatory Co ission 8 "' "** * #

Washington, DC 20555

8. (Leave Nankl

- 12. SPONSORING ORGANIZATION NAME AND MAIL G ADORESS (include lip Code / /

10. PROJE CT/T ASK/ WORK UNIT NO.

Office of Nuclear Reactor Regulagion U.S. Nuclear Regulatory Commissio 11. cONTR ACT NO.

Washington, DC 20555

/

13. TYPE OF REPORT \ PE RIOD COVE RED (lactusive dates 1 Proposed SRP Section (Guide) /

p

15. SUPPLEMENTARY NOTES ''"'#"*#

P 6 0

16. ABSTR ACT (?OO avords or sess/ /

,m, j

[ Proposed new SRP Section 6.5.5 would $rovide acceptance criteria and review

(/ procedures to be used in assessing tbe role of gpressure suppression pools as fission product cleanup systems foTlowing potential reactor accidents. A cal-culational model to account for drywell bypass lp given, and minimum fission product decontamination factors detailed calculations of pool,s/are crubbing listed for use in instances have beenhperformed. The proposed in which no section is accompanied by a regulatory analysis and a supporting technical report. \

\

17. KEY WORDS ANO DOC ENT ANALYSIS 17s. DESCRIPTORS

\

Pressure Sup ession Pool Fission Pro uct Cleanup System 17b. IDENTIFIER /OPEN-ENDEO TERMS j G 18. AVAILA LITY STATEMENT 19. SECURITY CLASS (This reportl 21.'NO. OF P AGES Unelassified TY S Un1imited 20g'ssiNeg traispei 22. ,P RICE NGC *ORM 33S (7 771

• U.S.COVERhmthf petNTINC otrICE 1987 191-692:63049 l

120555078877 1 1A011X US NRC-0 ARM-ADM DIV 0F PUB SVCS l PGLICY & PUB MGT HR-POR NUREG W-501 DC 20555 WASHINGTON 1

e O

l l

1 l

l 9

. . _ _ _ _ . _ . . . _ _ _ . _ _ . _ . _ _ . _ _ _ _ _ _ _ _ _ . , , _ . . _ _ . _ _ . . . . _ _ _ _ . _ _ _ . _ _ _ _ _ . . . . _ _ _ _ . . _ . _ _ . . . _ . .