ML20207J704

Decision ALAB-841 Affirming LBP-85-35.NRC Shall Ensure That Final Analysis of Hydrogen Control Sys Includes More Detailed Review of Containment Heat Removal Capability & Effects of Combustible Gas Release.Served on 860725

ML20207J704
Person / Time
Site:	Perry
Issue date:	07/25/1986
From:	Shoemaker C NRC ATOMIC SAFETY & LICENSING APPEAL PANEL (ASLAP)
To:
References
CON-#386-104 ALAB-841, LBP-85-35, LPB-85-35, OL, NUDOCS 8607290231
Download: ML20207J704 (71)

Text

_ __

{b t  ;.

[ff0 UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION gg ATOMICSAFETYANDLICENSINGAPPEALgpg"Sp U? " ~ - -

DOC Administrative Judges: ff(' $(')f Alan S. Rosenthal, Chairman July 25, 1986 Dr. W. Reed Johnson (ALAB-841)

Howard A. Wilber -

) SERVED JUL35T986 In the Matter of )

)

CLEVELAND ELECTRIC ILLUMINATING ) Docket Nos. 50-440 OL COMPANY, ET AL. ) 50-441 OL

)

(Perry Nuclear Power Plant, )

Units 1 and 2) )

)

Susan L. Hiatt, Mentor, Ohio, for the intervenor Ohio Citizens for Responsible Energy.

Terry J. Lodge, Toledo, Ohio, for the intervenor Sunflower Alliance, Inc.

Jay E. Silberg, Harry H. Glasspiegel, Michael A. Swiger and Rose Ann C. Sullivan, Washington, D.C., for the applicants The Cleveland Electric Illuminating Co.,

et al.

Colleen P. Woodhead for the Nuclear Regulatory ,

Commission staff.

DECISION Before us are the appeals of intervenors Ohio Citizens for Responsible Energy (OCRE) and the Sunflower Alliance, Inc. (Sunflower) from the Licensing Board's September 1985 concluding partial initial decision and various earlier orders and rulings in this operating license proceeding 8607290231 860725 PDR ADOCK 05000440 G PDR 1502-

~

, 3 5

2

_ _ . .e 71 -

involving Units 1 and 2 of the Perry Nuclear Power Plant.

That decision authorized the issuance of licenses for the operation of Units 1 and 2 of the Perry Nuclear Power Plant, subject to the applicants' fulfillment of certain specified conditions.

I.

The Sunflower Appeal The Sunflower appeal need not long detain us. It seeks to challenge both (1) the rejection at the threshold of twenty Sunflower contentions relating to the Perry emergency response plan; and (2) the disposition on the merits of other contentions of that intervenor on the same subject.

On neither score, however, is the challenge adequately briefed.

The twenty contentions were denied admission to the proceeding because, in the Licensing Board's view, their bases were not set forth with reasonable specificity as l

l l

l l

l 1 See LBP-85-35, 22 NRC 514 (1985). The Perry facility consists of two boiling water reactors, each rated at 1265 megawatts electric. The facility is located on the shores of Lske Erie, in Lake County, Ohio, approximately 35 miles no'.'theast of Cleveland.

-, -- -- -w - -e e,

t t

3 required by the Commission's Rules of Practice, 10 CFR 2.714(b).2 Sunflower has not favored us with an explanation as to why the Board was wrong in so concluding. Rather, its brief simply restates the contentions and their purported bases. Similarly, in disputing virtually all of the Board's findings on those of its contentions that were tried on the merits, Sunflower does no more than to repeat, essentially verbatina, proposed findings that had been submitted to and rejected by the Board.

In the circumstances, we have no hesitancy in summarily rejecting the Sunflower appeal in its entirety. In passing in March 1985 upon Sunflower's appeal from an earlier partial initial decision in this proceeding, we took note of the fact that, with respect to several of its appellate assertions, Sunflower had " failed to provide any explanation why its claim of error is correct." That being so, we announced, the assertions were being treated "as waived or abandoned."4 It is difficult to understand why Sunflower's counsel chose to attach no significance to that result in -

the subsequent preparation of his brief on the present Memorandum and Order (Admissibility of Contentions on Emergency Plans and Motion to Dismiss) (January 10, 1985)

(unpublished) at 8-10.

ALAB-802, 21 NRC 490, 496 n.30.

4 Ibid.

P-

,-w - - . -

3 9

4 appeal. Whatever may have been the reason, however, the same outcome is warranted here. The short of the matter is that, if Sunflower wished us to take seriously its insistence that the Licensing Board committed error, its counsel was duty-bound to illume the foundation for that insistence.

II.

The OCRE Appeal In its brief, OCRE expresses disagreement with the vast majority of the Licensing Board's substantive holdings on those issues litigated by that intervenor. Although each of its claims of error has been examined, we follow the course recently pursued in the Shoreham proceeding and specifically address in this opinion only those of sufficient possible merit or significance to require further discussion.5 The assertions meeting this test fall into six categories:

(1) the adequacy _of the plant's system for the control of hydrogen in severe accident situations; (2) the reliability of the existing emergency diesel generators; (3) the necessity for automatic initiation of the Standby Liquid Control System (SLCS) to aid in the rapid shutdown of the plant's reactors in the event of an anticipated transient without scram (ATWS);

! See Long Island Lighting Co. (Shoreham Nuclear Power l Station), ALAB-832, 23 NRC'135, 143 (1986).

s 5

(4) the risk of damage to safety equipment from

" turbine missiles";

(5) the effect of radiation on polymers used in the plant's safety equipment; and (6) the need for the pressure testing of the containment personnel airlocks.

A. Hydrogen Control. In the event of a loss of coolant accident (LOCA) at a nuclear power plant, the temperature of the fuel and the fuel cladding will rise.

Most water-cooled power reactors have cladding that consists primarily of zirconium, which, if it reaches a sufficiently high temperature, may react with water or steam to generate hydrogen. This hydrogen, in the improbable event that it were to accumulate to high concentrations,6 could ignite violently to threaten the integrity of the containment structure and the operability of components inside the containment.

During the 1979 accident at Three Mile Island, the deflagration of hydrogen resulted in significant pressure and temperature increases in the reactor containment 6

The generation of large quantities of hydrogen during the course of a reactor accident is not normally expected.

Commission regulations (10 CFR 50.46 and Part 50, Appendix K) governing the design, analysis and functioning of emergency core cooling systems (ECCS) in nuclear power plants were promulgated to limit, inter alia, the amount of hydrogen produced by the fuel cladding zirconium-water reaction to that resulting from oxidation of less than one percent of the cladding. 10 CFR 50.46(b)(3).

i

?

6 building. Following that event, in order to accommodate severe accidents in which hydrogen is generated in quantities greater than that allowed by the ECCS rule (see supra note 6), the Commission amended its regulations to require improved hydrogen controls.7 The nature of these controls is dependent on the design of the particular nuclear power facility.

For certain non-inerted reactor containment 3 such as that at Perry (i.e., those with atmospheres that contain oxygen, a necessary ingredient for hydrogen combustion), the new hydrogen rule envisions a system that will provide for the controlled burning of hydrogen as it is generated during the course of an accident. This will prevent hydrogen buildup to concentrations at which violent deflagration or l

an explosion might take place.8 The rule requires that the hydrogen control system be able to burn safely the amount of hydrogen that would be generated if up to 75% of the fuel cladding in the active fuel region reacted with water.'

l Further, the systems and components necessary to establish and to maintain safe shutdown, and to preserve containment l

10 CFR 50.44.

See generally 50 Fed. Reg. 3498 (1985) (Statement of Consideration for the hydrogen rule).

9 10 CFR 50.44(c)(3)(v)(B).

1 r

E.

t t

7 integrity, must be capable of performing their functions i

during and after exposure to the environmental conditions created by the burning of hydrogen.10 In order to obtain a full-power operating license, the hydrogen rule requires the utility to submit an analysis to the staff that, among other things, provides an evaluation of the consequences of hydrogen generation and combustion during staff-accepted severe accident scenarios.Il This i submittal need not, however, be a completed " final" i analysis. Instead, all that is required at the operating license stage is "a preliminary analysis which the staff has determined provides a satisfactory basis for a decision to support interim operation at full power until the final analysis has been completed. 12

1. The Perry hydrogen control system consists of 102 igniter assemblies distributed throughout the containment.

l When activated during an accident, each igniter will exhibit l

a surface temperature high enough to cause the ignition of hydrogen as it reaches a combustible concentration. The electric power necessary for igniter operation is to be 10 10 CFR 50.44(c)(3)(v)(A).

11 10 CFR 50.44(c)(3)(vi)(B).

12 10 CFR 50.44(c)(3)(vii)(B) (emphasis supplied).

J P

8 supplied by either normal offsite sources or the emergency diesel generators.13 As required by the current hydrogen rule, the applicants conducted a preliminary analysis of this system and submitted the results to the staff. Using accident scenarios that had been approved by the staff in the context of the hydrogen control analysis for the Grand Gulf facility (a boiling water reactor recently licensed by the NRC with safety-related equipment identical or similar to that found in Perry),14 the applicants modeled the temperature and pressure response of the containment environment by means of the CLASIX-3 computer code.15 They then determined the capability of the containment to maintain its integrity despite the predicted pressures.16 With respect to the 13 Buzzelli, et al., fol. Tr. 3241, at 32-34.

l 1 The hydrogen igniter systems installed at Grand Gulf i and Perry are similar. Applicants Exh. 8-1 at 22-23. In addition, the hydrogen control analysis conducted for Perry utilized the same accident scenarios and the same degree of >

cladding reaction with water (75%) as employed in the Grand Gulf analysis. Id. at 18-21, 28-30. However, as Grand Gulf has a larger reactor core, and thus more zirconium metal to l

react, a greater total amount of hydrogen would be released and available for combustion during a severe accident at that plant. Notafrancesco (Second Prepared Testimony (#2)),

fol. Tr. 3676, at 2-3; Garg, fol. Tr. 3676.

1 Buzzelli, et al., fol. Tr. 3241, at 41-43.

16 Applicants Exh. 8-1 at 13-17. See generally Applicants Exh. 8-4. l l

1 l

\

s 9

thermal survivability of necessary equipment, a comparison of the temperature response profiles of Perry and Grand Gulf revealed that lower equipment temperatures should result during hydrogen burning at Perry. As verification, a calculation was performed that showed that the maximum temperature of an igniter assembly would be lower at Perry than at Grand Gulf.1 For purposes of pressure survivability analysis, the qualification or design pressures of the equipment were compared to the peak pressure resulting from hydrogen burning during the hypothesized accident.19 Based on its review of the applicants' submittal, the staff found the preliminary analysis to be acceptable and in compliance with the regulations.20 In so finding, the staff identified certain issues that the final analysis is to address.21 1

Applicants Exh. 8-1 at 21A-21D.

Buzzelli, et al., fol. Tr. 3241, at 49-50.

l 19 Applicants Exh. 8-1 at 21D.

20 Notafrancesco (#2), fol. Tr. 3676, at 6.

l Staff Exh. 8, NUREG-0887, Safety Evaluation Report for Perry Nuclear Power Plant, Supplement No. 6 (April 1985) at 6-12 [hereafter cited as " Staff Exh. 8"]. Those issues include (1) mechanistic hydrogen-steam release histories i that are representative of degraded core accident sequences; l (2) containment diffusion flame environment; (3) containment (Footnote Continued) l

o o

10

2. OCRE's Issue 8 put into question the effectiveness of the applicants' hydrogen control system:

The Perry hydrogen control system is inadequate to assure that large amounts of hydrogen can be safely accommodated without a rupture of the containment and a release of substantial quantities of radioactivity to the environment.22 In this regard, OCRE complained that the applicants' preliminary analysis was defective and thus should not have been accepted by the staff. According to OCRE, the applicants had failed to select for analysis the most severe accident scenarios. In addition, OCRE asserted that the analysis was flawed because of uncertainties.in the CLASIX-3 computer code, inadequacies in the applicants' calculations of the structural capacity of the containment, and the applicants' reliance on the Grand Gulf analysis.

(Footnote Continued) l environmental aspects related to deflagrations, such as heat transfer modeling; (4) drywell environmental analysis; (5) l confirmation of equipment survivability; and (6) devel:pment of combustible gas control emergency procedures. ,

See LBP-85-35, 22 NRC at 530.

l l 23 OCRE further maintained that the final analysis of the system had to be completed before full-power operation could be allowed. Its thesis appeared to be that, l

otherwise, there would be an improper delegation to the l

staff of the ultimate resolution of the hydrogen control l issue. The Licensing Board correctly rejected that line of reasoning. Id. at 531. As we have seen, the hydrogen control rule now in effect plainly sanctions full-power operation on the strength of a satisfactory preliminary analysis. See supra p. 7.

l

4 11 On the basis of its review of the record developed on the issue, the Licensing Board found reasonable assurance that the applicants' hydrogen control system at Perry will function as designed, that the containment will retain its integrity during the predicted hydrogen burn, and that the necessary systems and components can survive in the environment resulting from the burn.24 Provided certain imposed conditions were met, the Board concluded, that system would be in compliance with the applicable regulations.25 Those conditions require that, before operation in excess of five percent of rated power, the applicants must demonstrate to the staff that (1) written procedures for operation of the system are available, and (2) further confirmatory analysis of the pressure survivability of equipment not so qualified or with narrow qualification margins has been made.26

3. On appeal, OCRE renews its claims raised below and asserts that the Licensing Board committed a wide variety of errors. Our discussion will be confined to those assertions l

l 24 LBP-85-35, 22 NRC at 549-50.

Id. at 551.

26 The conditions were incorporated into

-Id. at 588-89.

the operating license for Unit 1. See Perry Nuclear l Power Plant, Unit No. 1 Facility Operating License (License No. NPF-45), March 18, 1986, Section 2.C.(10).

12 of Licensing Board error that appear on the surface to have possible merit.27

a. The applicants selected two scenarios to be analyzed in order to determine the temperature and pressure response of the containment to a postulated degraded core event with hydrogen generation, release and combustion.28 One initiating event was the opening of a safety relief valve and its subsequent failure to close as expected. The other scenario started with a small steam-line break in the drywell. For each scenario, the emergency core cooling systems were assumed not to operate at the beginning of the accident. It was further assumed that only one of the two containment spray trains would be initiated after the first hydrogen burn.29 According to the scenarios, just prior to reaching a metal-water reaction equivalent to 75% of the cctive fuel cladding, recovery of the ccoling systems would occur and the hydrogen generation portion of the accident would be terminated.30 At the hearing belou, OCRE embarked upon cross-examination on subjects such as (1) the availability of 27 See supra p. 4.

28 Buzzelli, et al., fol. Tr. 3241, at 38.

29 Applicants Exh. 8-1 at 28.

30 Buzzelli, et al., fol. Tr. 3241, at 38.

s 4

13 containment sprays and the functioning of the Reactor Core Isolation Cooling (RCIC) system during a hydrogen release; (2) the use of containment venting in the event that the hydrogen control system could not be operated; and (3) the effect of station blackout on a hydrogen generation event.

Although these issues were said to relate to the functioning of the hydrogen control system, in actuality they challenged the accident scenarios used by the applicants. The Licensing Board granted OCRE's representative a fair amount of leeway in conducting such cross-examination but ultimately concluded that these questions went beyond the scope of the hearing.31 On appeal, OCRE asserts that the preliminary analysis should have addressed issues beyond the selected accident scenarios, such as the availability of containment sprays and the effect of station blackout. We think otherwise.

. Section 50.44(c)(3)(vi)(B)(3) provides in terms that the evaluation of the nydrogen rel' ease is to "[u]se accident scenarios that are accepted by the NRC staff." As earlier noted, the scenarios selected by the applicants were approved by the staff. We need not decide here whether there are any circumstances in which staff action of this nature might be subject to challenge by an intervenor. For, 1

LBP-85-35, 22 NRC at 548-49.

L

D 14 be that as it may, the approval was not open to successful attack in this instance.

Given the complexity of a nuclear power plant, there is virtually no end to the sequences of failures and errors that might conceivably result in hydrogen production. But the likelihood of the occurrence of most of the sequences is extraordinarily remote: in order for them to materialize, there would have to be such unlikely developments as the concurrent failure of redundant safety-related equipment or an equipment malfunction accompanied by improbable operator error. Manifestly, the Commission did not intend to require utilities to include in their analyses -- preliminary or final -- every one of these sequences, irrespective of how divorced from reality it might be. Moreover, it is plain from the terms of the rule itself that the Commission was fully prepared to leave it to the staff to decide which of i the vast number of possible scenarios should be analyzed.

l Assuming, again without deciding, that the exercise of the staff's broad discretion in that regard is reviewable at all, the intervenor seeking to challenge the choice of scenarios must do much more than simply allege that there ar other scenarios that the staff might appropriately have it isted be factored into the analysis: it must also allege and establish that, without the inclusion of the additional scenarios, the analysis could not fulfill its intended purpose. We are satisfied that no such demonstration was

i Q 15 l made here. Stated otherwise, this record does not ectablish that the staff acted capriciously in approving the use of the two chosen scenarios for preliminary assessment 32 purposes.

b. We now turn to the applicants' use of the CLASIX-3 computer code to analyze the Perry containment response to hydrogen combustion during the selected scenarios. Using this code, the calculated peak temperature and pressure in the containment would be 1760' Fahrenheit (F) and 21 pounds per square inch gage (psig). Before the Licensing Board, and again before us, OCRE c,hallenges the validity of this outcome. The primary basis for its attack is a report prepared for the NRC by Sandia National Laboratories in which CLASIX-3, as it was used in analyzing the hydrogen control system for the Grand Gulf nuclear i

facility, is evaluated by comparison with results obtained 32 The staff may require additional scenarios to be included in the final analysis yet to be performed.

33 Buzzelli, et al., fol. Tr. 3241, at 48. Pressure is ordinarily discussea In terms of pounds per square inch gage, which is the pressure measured above or below atmospheric pressure so that the measurement is independent of variations in atmospheric pressure. We note that the figures that depict the time dependence of temperature and

pressure in the Perry containment during the course of a hydrogen generation event indicate that these peaks are of very brief duration (less than a minute), apparently lasting i only during the periods of intermittent hydrogen burning.

See Applicants Exh. 8-1, Appendix A, Figures 3-8, 22-27.

,.r-9 - -- - _ _

_--__._..--.-y ,+,,..-,,,_y,w.-.mw.w_--__ -m- .y,-%.r,_,-- -.7-_,-.,_._.-m.. .

O 16 from a Sandia code, HECTR. That document takes issue with many of the assumptions used in CLASIX-3, such as hydrogen ignition limits, flame propagation speeds, and combustion completeness. Further, the use of HECTR generally yields higher temperatures and pressures in the containment than the results from CLASIX-3. OCRE asserts that the Licensing Board ignored this evidence contained in the Sandia report.

Contrary to OCRE's claim, the Licensing Board's decision contains a full discussion of the containment response models, and the uncertainties and assumptions associated with both CLASIX-3 and HECTR. At bottom, the Board accepted the CLASIX-3 analysis because the temperatures and pressures predicted by that code were OCRE Exh. 21 (NUREG/CR-2530). This document was admitted into evidence over the objection of the staff. See Tr. 3691. It was not the basis for any direct testimony and was used very sparingly in cross-examination (e.g., Tr.

3688-91, 3744-46). The document was, however, liberally cited in OCRE's proposed findings of fact, and in its appellate brief. The Board should not have admitted such a large (226 pages) unsponsored document without clearly limiting the evidentiary purpose to which it could be put.

We normally decline to address procedural matters not raised on the appeal. We have made an exception in this instance because the treatment given the document by the Licensing Board placed an undue burden on our review of the hydrogen control issue.

35 See LBP-85-35, 22 NRC at 538-42.

s 17 higher than those produced during large-scale experiments; 6

i.e., the code predictions were conservative. Our review of OCRE's Exhibit 21 and the rest of the record gives us no reason to quarrel with the Board's approach on this issue.37 The presented evidence indicates that the response model used by the applicants provides reasonable values of containment pressure and temperature as a result of a hydrogen burn.

c. The dual-train Residual Heat Removal (RHR) system at Perry is designed to remove decay heat from the suppression pool and to provide containment spray in the event of an accident.38 The combustion of hydrogen would 36 Id. at 539, 577. See also Tr. 3621, 3733-34.

In opposing OCRE's Motion to Compel the Appearance of Dr. Marshall Berman (March 18, 1985), the staff provided an affidavit by Dr. Berman, who directed the Sandia study which led to NUREG/CR-2530. See NRC Staff Response to OCRE's Motions for Continuance and to Compel the Appearance of Dr. Berman (March 27, 1985). Dr. Berman noted that much experimental work had been done since the document was published; in particular, experiments confirming hydrogen f ignition at mixtures of 6-8% by volume (CLASIX-3 assumed 8%,

l HECTR 10%). He also stated that he had no objection to the staff's position regarding the acceptance of CLASIX-3 results at Perry, if it were the same as that at Grand Gulf.

The Licensing Board quoted Dr. Berman's affidavit in denying OCRE's motion, concluding that it could not find genuine scientific disagreement between the NRC staff and Sandia.

Memorandum and Order of March 29, 1985 (unpublished) at 4.

See also Tr. 3724-26.

8 The water Tr. 3453; Applicants Exh. 8-1 at 25-26.

in the suppression pool within the containment serves i primarily to quench the steam release during a LOCA.

l l

{

18 produce additional heat that must be removed from the containment atmosphere. However, because hydrogen combustion would occur early in the accident and the maximum temperature would not be reached in the suppression pool until much later, applicants' witness John D. Richardson concluded that the heat from hydrogen combustion would not have a significant effect on the suppression pool temperature.39 Accepting this evidence, the Licensing Board determined that an adequate method existed to remove heat from the containment if a degraded core accident occurred at Perry.40 According to OCRE, the Licensing Board should not have accepted the adequacy of the heat removal system on the basis of this testimony but, instead, should have insisted upon an analysis of Perry's heat removal capability. We disagree. For purpoces of the preliminary analysis, there is no reason to reject the applicants' thesis: 1.e., the heat from hydrogen combustion would not cause the containment heat removal capability to be exceeded because the combustion would occur early in the accident, before the 39 Tr. 3611.

LBP-85-35, 22 NRC at 547.

19 I

i maximum suppression pool temperature would be reached.41 In 41 It is not entirely clear from the Board's opinion why it believed that the fact that heat from hydrogen burning comes early in the accident scenario had a bearing on the question of the ultimate capability to remove heat from the containment. But, while not considered explicitly ~

j in the testimony, the record does provide the answer.

The Mark III suppression pool contains a large amount of water, which has the capacity to store a great deal of heat (7.35 million pounds of water with a heat capacity of 1.0 Btu /lb 'F). See Applicants Exh. 8-1, Appendix A, Table

8. The worst case degraded core scenario results in the burning of 2290 pounds of hydrogen. Id. at 30. If all of the energy resulting from the hydrogen burning were added suddenly to the suppression pool, the pool's temperature would rise only 19*F. (This computation employs thg heat-of-combustion value for hydrogen of 6.096 x -10 8 Btu /lb.

Hence, the hydrogen burn yields a total of 1.40 x 10 Btu.

See Chemical Engineers' Handbook at 3-145 (R.H. Perry & C.H.

Chilton editors, 5th ed. 1973).) Heat is removed from the

suppression pool by the RHR system, whether by the i containment spray mode or the pool cooling mode, althcugh the cooling capacity is somewhat reduced (by perhaps 15%)

when in the spray mode. See Tr. 3453, 3455, 3476, 3481-82.

As one might expect, however, the heat remcval rate due to RER system operation increases as the suppression pool temperature increases, and, at the suppression pool temperature design limit of 185 F, one of the two loops of ,

the RHR system operating in the containment spray mode I

(assuming a 15% reduction in heat removal rate) will gemove heat from the suppression pool at a rate of 1.41 x 10 Btu / hour. Perry Final Safety Analysis Report [FSAR1, Amendment 15 (December 31, 1984), Table 6.2-3, Figures 6.2-8 and 6.2-9; Applicants Exh. 8-1, Table 5.4-3. Thus, the entire amount of heat added to the pool due to hydrogen combustion would be removed in about one hour. The peak suppression pool temperature (185') due to reactor decay

1. heat does not occur until about three hours after an I accident if only a single RHR loop is in operation. Perry FSAR, Figure 6.2-8.

In summary, the suppression pool has the capacity to accept the heat resulting from hydrogen burning with only a J

(Footnote Continued) f I

!' _ - - , . _ , . _ _ . _ _. ._ . _ _ _ _ _..__. _ . _ ,._ --- _ ~ - _ _ _ _ . _ _ _ _ _ . . - .

20 connection with the final analysis, however, the staff ,

should ensure that the applicants have performed a more detailed review of containment heat removal capability.

d. OCRE asserts that the Licensing Board failed to address the potential for release of combustible gases from electrical cable insulation that becomes heated during a hydrogen burn. The intervenor refers to a paper in evidence that suggests that such gases, if they also burned, could affect the pressure-temperature response during a 2

hydrogen combustion event. A staff witness, who was a co-author of the paper, testified, however, that the relevance of the paper was not clear.43 Because, from a review of the paper, it appears that for the conditions at Perry the combustion of gases released from heated insulation would not be a significant factor in the overall consequences of a hydrogen burn, its omission from the (Footnote Continued) modest increase in pool temperature. Assuming the operation of only one, of two, RHR cooling loops, this heat will be removed well before the onset of peak pool temperature. In fact, the evidence establishes that, even if the hydrogen combustion heat were added tc the pool at the time of peak pool temperature (185* + 19* = 204*), the heat removal system will still function. Tr. 3459-60, 3606-07. As shown above, the system would bring the pool temperature back to its design value in about an hour.

42 See OCRE Exh. 24 at 1205.

43 Tr. 3730-31. The subject paper dealt with a dry containment building of a pressurized water reactor, while Perry is a boiling water reactor with a Mark III containment.

21 applicants' preliminary analysis was acceptable. But the possible effects of this phenomenon are to be included in the final analysis.

e. The regulations allow the structural integrity of the containment to be demonstrated by an analysis showing that the Service Level C stress limits of the American Society of Mechanical Engineers Boiler and Pressure Vessel Code (ASME Code) have been met.44 Using these limits, the applicants performed an analysis to determine the location in the Perry containment with the lowest pressure-retaining capacity. This analysis showed that Containment Penetration 414 has the lowest capacity with a value of 50 psig.45 Thus, the applicants concluded, the containment would retain its integrity so long as that pressure was not exceeded.

44 10 CFR 50.44(c)(3)(iv)(B)(1), (vi)(B)(5)(i).

Under the ASME Code, the Service Level C stress limit for simple loadings, such as the membrane stress in a steel containment shell, is the yield stress of the material.

This ensures that, for such loadings, material behavior will remain in the elastic range; hence, though the object may deform under loading, it will return to its original shape upon removal of the load. For more complex loading, higher stresses are allowed under Service Level C. See Tr.

l 3584-85; Applicants Exh. 8-4, Table 8; ASME Code,Section III, NCA-2142.2 (1983 edition).

45 Buzzelli, et al., fol. Tr. 3241, at 28.

l l .

- - - ,,---y----

, , - - ,- .-- , - - - -,---------n-- , - - - - - - - - - - , - , ,- - - - - *~-

e 22 The capacity of Penetration 414 had been determined by utilizing specified minimum material strength values for steel of the type surrounding the penetration -- values that were derived from the ASME Code. Had the same values been employed in the case of Penetration 205, for example, pressure capacity lower than 50 psig would have been obtained. Instead, Penetration 205 was analyzed on the basis of the actual physical properties of the material surrounding it, as ascertained from material certification data furnished by the supplier. That procedure produced a pressure capacity in excess of 50 psig.47 The regulations specifically permit the use of actual material properties for analytical purposes.48 We therefore cannot subscribe to OCRE's insistence that the applicants were required to appraise the pressure-retaining capacity of Penetration 205 on the same basis as Penetration 414 was examined.

OCRE also asserts that the analysis erroneously failed to consider the effects of dead load and elevated temperature. Had they been included in the analysis, OCRE claims, a lower value would have been given to the pressure 46 Tr. 3285.

47 Applicants Exh. 8-4 at 16-17.

48 See 10 CFR 50.44(c)(3)(iv)(B).

g 23 capacity of Penetration 414 and thus to that of the containment. Although that may be so, it is of no consequence here. For, the record establishes that the dead load and elevated temperature factors would reduce the allowable stress by approximately five and ten percent, respectively.49 Thus, even if those factors were assumed to have a cumulative impact, Penetration 414 would still have a pressure capacity well in excess of the hydrogen burn pressure of 21 psig.50

f. As earlier noted, the Licensing Board imposed a condition that required, prior to operation at levels 49 Tr. 3286, 3586-87: Applicants Exh. 8-4 at 16. OCRE also claimed that containment vessel out-of-tolerance, had it been considered, would have resulted in a lower pressure capacity value. But containment vessel out-of-tolerance only affects the steel shell and not the penetrations. Tr.

3596-97. The pressure capacities of the cylindrical and dome regions of the Perry containment are 79 and 78 psig, respectively, which are well above the 50 psig limiting penetration capacity. Ibid.

50 For low probability loading events the ASME Code permits the use of higher, Service Level D stress limits.

Using these limits, the applicants calculated a containment capacity of 57 psig, which they characterized as "more realistic." Buzzelli, et al., fol. Tr. 3241, at 28.

Although the Licensing Board took note of this calculation (LBP-85-35, 22 NRC at 536), there is no merit to OCRE's claim that the Board placed crucial reliance upon it.

Appellate Brief of Ohio Citizens for Responsible Energy (October 21, 1985) at 24. In any event, the 50 psig containment capacity was demonstrated using the required Service Level C limits, notwithstanding the applicants' additional Level D analyses or the Licensing Board's nondispositive opinion of those analyses.

24 above five percent of rated power, " confirmatory analyses" of the pressure survivability of certain components that either had not been qualified or had " inadequate margins" of survivability.51 On appeal, OCRE complains that the additional analyses of pressure survivability required by the Board will not be simply confirmatory but, rather, will require an evaluation of sufficiency by the staff. OCRE takes this to represent the delegation of a contested issue to the staff for post-hearing resolution. In this regard, OCRE condemns as " illogical" the Board's finding that the Perry containment has the capability to cope with a vacuum (i.e., " negative pressure").52 OCRE notes that this finding was predicated on the operability of the vacuum breakers, yet the Board found the same vacuum breakers to be insufficiently qualified to withstand the hydrogen burn pressure.

We agree with OCRE that, having concluded that the staff should not have accepted the applicants' evaluation of the pressure survivability of those components with design 51 LBP-85-35, 22 NRC at 544. Of those components for which a preliminary evaluation of pressure survivability had been completed, only the containment vacuum breakers, the hydrogen mixing compressors, and the hydrogen mixing compressor check valves did not have qualification or design pressures that exceeded the calculated peak pressure from a hydrogen burn. Applicants Exh. 8-1 at 21D.

OCRE Brief at 21. See LBP-85-35, 22 NRC at 536.

s t

s .

25 pressures below the burn peak pressure,53 the Licensing Board was required to retain jurisdiction over the matter until a satisfactory evaluation was produced. The error on that score, however, has turned out to be harmless.

Contrary to the view of the Licensing Board, the applicants' preliminary evaluation of the pressure qualific'ation of this equipment is acceptable.54 Insofar as concerns the containment vacuum breakers and hydrogen mixing compressor discharge check valves, the record shows that only the external design pressure is exceeded by the hydrogen burn peak pressure.55 Because the active componepts of the vacuum breakers and check valves are not exposed to the hydrogen burn pressure, at this preliminary stage it is reasonable to proceed on the basis that this equipment would survive a hydrogen event. With respect to the hydrogen mixing compressor, identical equipment at Grand Gulf has been shown to survive pressures exceeding the hydrogen burn peak pressure.56 Again, at least in the context of a preliminary evaluation (which is all the Commission's l

I Id. at 578. See Staff Exh. 8 ct 6-11.

54 The staff will require equipment survivability to be confirmed in the final analysis. Id. at 6-12.

55 Applicants Exh. 8-1 at 21D.

56 Ibid.

26 m

hydrogen rule requires at this point), this allows the conclusion that the compressors will function through a hydrogen event at Perry.

g. Finally, OCRE urges that 10 CFR 50.44(c)(3)(vii)(B) precluded the applicants' reference to the equipment survivability analysis conducted for the Grand Gulf facility. While the provisions of that section may be open to different interpretations, the Statement of Consideration accompanying-its promulgation indicates a Commission intent to allow a hydrogen burn analysis for one facility to make use of a previous and staff-accepted analysis for another similar facility.58 The question remains, of course, whether, as a matter of fact, the Grand Galf analysis was appropriately employed in the Perry analysis. We conclude that it was. As previously indicated, not only are the two facilities including their hydrogen igniter systems similar, but also l

57 See supra note 14.

0 l

As the Commission stated:

Previously approved generic or reference analyses may be employed in lieu of plant specific analyses where the generic analyses can be shown to be

! applicable. It is believed that the adoption of the above approach will eliminate the need for repetitive calculation of accident scenarios.

t 50 Fed. Reg. 3502.

l l

l.

t

4 27 the hydrogen control analysis performed for each was essentially identical.59 In fact, because of the larger reactor core at Grand Gulf and the additional hydrogen that would be generated, more severe equipment temperatures are predicted to occur during a hydrogen burn event at that plant.60 B. Diesel Generators. General Design Criterion (GDC) 17 requires a nuclear plant to include, inter alia, a reliable onsite electric power system to permit the functioning of equipment needed to maintain the plant in a safe condition in the event of the loss of other sources of power. 61 To meet this requirement at Perry, the applicants installed four diesel generators manufactured by Transamerica Delaval, Inc. (TDI), two for each Perry unit.

The reliability of TDI diesel generators in general came into question as a result of the identification of l deficiencies in these units at other nuclear power l

l l

See supra note 14.

60 Applicants Exh. 8-1 at 21B-21C.

1 The General Design Criteria for Nuclear Power Plants are found in Appendix A to 10 CFR Part 50. As the Introduction to the Appendix states, these standards

" establish minimum requirements for the principal design criteria for water-cooled nuclear power plants similar in design and location to plants for which construction permits have been issued by the Commission."

O o

28 facilities.62 On this apparent basis, OCRE submitted a contention that the applicants had not demonstrated that Perry's diesel generators could be relied on to generate the necessary power in an emergency.

At the hearing, the applicants and the staff submitted a plan that had been prepared jointly by the twelve or so owners of nuclear plants with TDI diesel generators. The Owners Group plan provided for an in-depth assessment of each facility's TDI diesel generators through a combination of design reviews, quality revalidations, engine tests and component inspections.63 The Board found that the plan was "a well-thought-out program which, if implemented properly, provides reasonable assurance that TDI diesels will reliably carry out their intended function."64 On the strength of the applicants' commitment to follow the plan, _the Board concluded that emergency onsite power will be available when needed and that the applicants met the regulatory requirements.65 62 Staff Exh. 1, Safety Evaluation Report on Transamerica Delaval, Inc., at 1. See also, Duke Power Co.

(Catawba Nuclear Station, Units 1 and 2), ALAB-813, 22 NRC 59, 79 (1985).

63 Kammeyer, fol. Tr. 2179, at 8.

64 LBP-85-35, 22 NRC at 561.

Ibid.

i

O 29 OCRE challenges this conclusion on three grounds.

First, it claims that, in view of its source, the Board's reliance on the Owners Group plan was improper and in violation of the Commission's quality assurance requirements. Second, according to OCRE, many of the Board's findings respecting the sufficiency of the plan were contrary to the weight of the evidence and based upon the application of erroneous evidentiary standards. Third, the Board is said to have violated the Atomic Energy Act in delegating to the staff the responsibility for monitoring the applicants' implementation of the plan. As a separate matter, OCRE argues that the Board improperly denied its motion to reopen the record to consider the implications of defective check valves associated with the TDI diesel generators at the Grand Gulf facility. We consider these claims seriatim.

1. There is no merit to OCRE's insistence that, in l relying upon the Owners Group plan, the Licensing Board erroneously failed to attach significance to the fact that the plan was devised by the interested utilities themselves rat'..er than by an independent technical organization.

l Contrary to OCRE's apparent belief, there is nothing in l

Appendix B to 10 CFR Part 50 (the quality assurance criteria for nuclear power plants) that places either an express or an implied limitation upon who may prepare a quality assurance plan. Rather, such a plan can be formulated by

30 any entity or organization and then, irrespective of its source, is judged on its own merits. The plan in question was so assessed by the staff, which found upon an extensive review that (if properly carried out) the plan was adequate to provide reasonable assurance that the diesel generators at Perry would function when needad. Unless that ultimate finding.was shown to be crucially flawed, the Board could rely upon the plan in making its own reasonable assurance finding.

2. OCRE attacks the adequacy of the Owners Group plan on a variety of grounds. We address here only those of its arguments that pertain to a genuine safety concern.66

a. Each TDI diesel generator is supported by chock plates that rest on the concrete floor. The amount of surface contact between the diesel engine base and the chock plates is important in determining the stress that will be

\

exerted on each plate. OCRE claimed that the Owners Group plan did not provide assurance of sufficient surface contact.67 The staff agreed and required the applicants to i

66 OCRE asserts that the burden of proof was not placed on the applicants as required. We have reviewed those instances cited by the intervenor and conclude that the Board's treatment of each was proper.

OCRE Response to Applicants' Motion for Summary Disposition of Issue 16 (February 27, 1985) at 47-49.

l l

i l

. l I

31 I

establish that the chock plates would withstand the stresses placed upon them.68 ,

To satisfy the staff's requirement, the applicants presented an engineering evaluation outside of the Owners Group p3in. On the basis of the ascertained surface contact between the engine base and the chock plates, the evaluation concluded that the plates would withstand the O

stresses. The Licensing Board considered this conclusion and other evidence, and determined that the foundations of the diesel generators were acceptable. 1 On appeal, OCRE does not directly attack the evaluation but claims that it had been rejected by a staff witness. In actuality, however, that witness did not reject the evaluation but stated that, before taking a position, he would require additional information regarding the minimum amount of surface contact specified by the generator manufacturer.72 The applicants thereafter supplied a witness who testified that, as the evaluation that had been 68 Berlinger, et al., fol. Tr. 2281, at 54-55.

69 Tr. 2496-97.

O Ibid. See OCRE Response to Applicants' Motion for Summary Disposition of Issue 16 (February 27, 1985), Exhibit 56.

1 LBP-85-35, 22 NRC at 560.

2 Tr. 2417-19.

32 conducted by the manufacturer had determined, the surface contact was sufficient in this instance.

In the circumstances, we are satisfied that the concern for adequate stability of the Perry diesel generators has been properly resolved even though the Owners Group plan did not fully consider this matter.73

b. The Owners Group plan provided for a detailed design review of the cylinder block.?4 This was prompted by the development of stud-to-stud and ligament cracks in the blocks of TDI. diesel generators.75 On the basis of this review, the Owners Group recommended a certain allowable depth for stud-to-stud cracks that develop in the cylinder block. A staff witness disagreed with the recommendation.

On the assumption that ligament cracks would also be 3

Tr. 2496-97. Given OCRE's total reliance upon the staff's asserted misgivings, it is worthy of passing note i that, since the Licensing Board issued its decision, the

staff has accepted the applicants' evaluation of the diesel

! generator foundation. See NUREG-0887, Safety Evaluation Report for Perry Nuclear Power Plant, Supplement No. 8 l (January 1986) at 9-10.

! 74 Kammeyer, fol. Tr. 2179, at 11-12.

5 See Wood, fol. Tr. 2179, at 55-62. As discussed at the Shoreham hearing on TDI diesel generators, a stud-to-stud crack extends from one stud counterbore to a stud counterbore of an adjacent cylinder, while a ligament crack extends from the cylinder head stud counterbore to the cylinder liner counterbore. See Long Island Lighting Co.

(Shoreham Nuclear Power Station, Unit 1), LBP-85-18, 21 NRC 1637, 1646 (1985).

l

i 4

33 present, he opined that a lower limit should be imposed upon 6

the allowable depth for stud-to-stud cracking.

For its part, the Licensing Board noted that no ligament cracks had as yet been identified at Perry.77 That being so, the Board concluded, the disagreement was

" irrelevant" and, therefore, the owners Group recommendation l should be accepted. 8 Challenging this outcome, OCRE points to staff testimony that indicated that, with the small amount of operating time accumulated by the diesel generators, there had been little opportunity for ligament cracks to develop. 9 Given this limited operating experience, we agree that the Licensing Board erred in dismissing the concern regarding the acceptable depth of stud-to-stud cracks in the presence of ligament cracks.

Subsequent developments, however, have rendered the error harmless. Earlier this year, the staff imposed a condition upon the Perry Unit 1 license requiring that, in the event of the identification of stud-to-stud cracks, the affected diesel generator be considered inoperable and the i

76 Tr. 2372-74.

LBP-85-35, 22 NRC at 559.

8 Id. at 559, 561.

l 70

' Tr. 2413.

i 1

34 NRC staff be notified. Under the condition, the diesel generator may not be returned to an operable status until corrective actions have been approved by the staff. This stringent license condition obviously satisfies any concern pertaining to the minimum acceptable depth of a stud-to-stud crack in the cylinder block.81

c. On the basis of its evaluation, the Owners

, Group concluded that the Perry diesel generators are capable of supplying continuous power during an emergency, in satisfaction of the specifications set forth in the Final Safety Analysis Report. In the interest of ensuring adequate maintenance of the generators, however, the Group suggested that the cylinder block of each generator be inspected after 572 hours0.00662 days <br />0.159 hours <br />9.457672e-4 weeks <br />2.17646e-4 months <br /> of operation.83 Seizing upon this recommendation, OCRE argued below that it established that the TDI generators failed to meet the requirement for a 80 See Perry Nuclear Power Plant, Unit No. 1 Facility Operating License (License No. NPF-45), March 18, 1986,

' Attachment 4 at 1. As the license condition does not mention ligament cracks, we assume the condition holds regardless of the presence or absence of that type of cracking.

81 The diesel generator requirements imposed upon Unit 1 will also be made applicable to Unit 2. Tr. 2483.

82 Tr. 2194-95.

83 Wood, fol. Tr. 2179, at 62; Kammeyer, fol. Tr. 2179, Exhibit A at 3.

35 continuous emergency power supply.04 Citing testimony that was withdrawn later, the Licensing Board rejected the argument and found that the diesel generators could " fulfill their basic purpose even with the 572-hour inspection limit."85 On appeal, OCRE charges that, inasmuch as it rested on withdrawn testimony, that finding must be set aside. We think otherwise. To be sure, the Licensing Board erred in relying on withdrawn testimony. But there is sufficient evidence in the record to establish that the recommendation for cylinder block inspections at periodic intervals does not cast doubt on the ability of the Perry diesel generators to provide needed power during an emergency.

For example, a detailed review of the cylinder block to establish the adequacy of its design was conducted by the

! Owners Group and approved by the applicants' consultants and the NRC staff. Further, TDI diesel generators of the same model as those at Perry have been operated at the Catawba 84 See LBP-85-35, 22 NRC at 558.

Id. at 559. The testimony in question was to the 85 effect that emergency power is needed for co e cooling purposes for no more than one week (168 hours0.00194 days <br />0.0467 hours <br />2.777778e-4 weeks <br />6.3924e-5 months <br />). Tr.

2221-22. The witness later repudiated that testimony. Tr.

2274.

86 Christiansen, fol. Tr. 2179, at 34-35; Kammeyer, l

! fol. Tr. 2179, at 12; Berlinger, et al., fol. Tr. 2281, at 12-16.

I

- - . - - . , - - - - - - - - - - - . - _ . . . - . , . - - . - - - - . - . , , - - . . . - - - . . . . . . , _ _ --.,__.,-.-----,-n .

m,-.. . , . - . . . - - . - . - - , . . . - ,

36 nuclear facility for 1600 hours0.0185 days <br />0.444 hours <br />0.00265 weeks <br />6.088e-4 months <br />.87 In this connection, notwithstanding the inspection interval recommendation, the Owners Group explicitly noted its belief that, should it prove necessary because of an ongoing emergency, the generators could be counted upon to continue to operate properly well beyond the 572-hour period.88

d. Following a full appraisal, the staff concluded that the actions already undertaken or planned by the applicants are adequate to ensure that the Perry Unit 1 diesel generators can reliably generate emergency power.* '

Nevertheless, the staff indicated that it would reassess the Owners Group program, and its own review of that program, after the first refueling outage.90 OCRE charges on appeal that the Licensing Board erred in relying on staff approval O Christiansen, fol. Tr. 2179, at 27.

88 Tr. 2194-95, 2268-72. The probability of offsite power being unavailable, hence the need for diesel generator operation, for more than 163 hours0.00189 days <br />0.0453 hours <br />2.695106e-4 weeks <br />6.20215e-5 months <br /> (one week) is very low.

Tr. 2273. Thus, in all likelihood, no difficulty will be encountered in maintaining the recommended 572-hour inspection intervals for the diesel generators.

09 Berlinger, et al., fol. Tr. 2281, at 12-13. As earlier noted, the Unit -2 diesel generators also will have to satisfy the requirements imposed upon the Unit 1 generators.

90 Tr. 2305.

37 of the diesel generators when (according to OCRE) that approval was " preliminary" and for an " interim duration."91 It is clear, however, that the staff's approval of the use of the Perry diesel generators was not preliminary but involved an extensive review of their adequacy.92 With respect to the planned reassessment of the Owners Group program, it is part of the staff's ongoing responsibility -

during operation of a nuclear power plant to review the success of various programs that are under way at the plant.

There is thus-no merit to the argument that the staff's planned reassessment program undermines its approval of the adequacy of the Perry diesel generators.

3. In Supplement 6 (April 1985) to its Safety Evaluation Report (SER) at 9-6 to 9-7, the staff informed the applicants that they would be required to take certain actions with regard to the diesel generators before an operating license would be issued for Unit 1. For reasons that are not entirely clear, the Licensing Board imposed a license condition that obligated the applicants to complete those actions.93 OCRE Brief at 29.

92 Berlinger, et al., fol. Tr. 2281, at 12-13. See generally Staff Exh7 17 Safety Evaluation Report on Transamerica Delaval, Inc.

LBP-85-35, 22 NRC at 588.

o 38 On appeal, OCRE insists that the license condition violated Section 189a. of the Atomic Energy Act of 1954, as amended,94 in that it delegated a contested issue to the staff for post-hearing resolution.95 As is manifest, however, no such delegation took place. OCRE had a full opportunity at the hearing to litigate the merits of any or all of the actions that the staff thought should be taken (many of which had, in fact, already been completed, albeit not yet reviewed by the staff).96 The condition here in question mandated simply that the staff confirm that the SER requirements had been fulfilled -- a step that the staff undoubtedly would have taken even in the absence of the condition.97 94 42 U.S.C. 2239(a).

In this connection, OCRE cites Union of Concerned Scientists v. NRC, 735 F.2d 1437 (D.C. Cir. 1984), cert.

! denied, 105 S. Ct. 815 (1985). There the court overturned a Commission rule that foreclosed litigation of the results of

emergency preparedness exercises.

' See LBP-85-35, 22 NRC at 554, 556-58, 560. See also, e.g., Christiansen, fol. Tr. 2179, at 12-16, 19, 34-35; Wood, fol. Tr. 2179, at 78-81; Berlinger, et al.,

fol. Tr. 2281, at 8-10, 12-13; Tr. 2265, 2302-03, 2324-26, 2415-19, 2422-26, 2496-97, 2511; Staff Exh. 1 at 10-11, 18.

It is worthy of note that the staff has accepted the applicants' submittal regarding each of the matters in question. See Perry Safety Evaluation Report, Supplement No. 8 (January 1986) at 9-7 to 9-11.

l

s 39

4. On the basis of a staff report that disclosed defects in check valves in the air start system of a TDI diesel generator at the Grand Gulf nuclear facility, OCRE moved to reopen the record on the diesel generator issue.98 The motion contended that the defects demonstrated a failure of the Owners Group plan to ensure the adequacy of the TDI diesel generators. In denying the motion, the Licensing Board concluded that there was "no conceivable circumstance by which this Board's decision on (the diesel generator issue] -- whatever its decision may ultimately be -- could be affected by the proffered evidence." '

As explained at the hearing below, the Owners Group gave priority attention to those diesel generator components with known problems.100 Other components were subjected to a design review or a quality revalidation (or both), based on their importance, past operational experience, and the 98 Motion to Reopen the Record on Issue #16 (April 30, 1985). In addition, the staff indicated in its response to the motion that cracks had been identified in similar check valves in TDI diesel generators at the Shoreham nuclear power plant. NRC Staff Response to OCRE Motion to Reopen the Record (May 15, 1985), Affidavit of Drew Persinko at 3.

99 Memorandum and Order of May 28, 1985 (unpublished) at 2. See Kansas Gas and Electric Co. (Wolf Creek Generating Station, Unit No. 1), ALAB-462, 7 NRC 320, 338 (1978).

100 Kammeyer, fol. Tr. 2179, at 11.

40 engineering judgment of the Owners Group. At Grand Gulf, although an engineering application review was performed, the air start check valves were not subjected to a quality revalidation because they were not manufactured by TDI nor had there been evidence of past failure of this type of valve.102 In the circumstances, we agree with the Licensing Board that the new information brought forward by OCRE would not have affected its decision oa the adequacy of the Owners Group plan at the Perry facility.

C. Standby Liquid Control System. Nuclear power facilities utilize control rods containing neutron-absorbing material to help regulate the fission rate in the reactor core. Emergency shutdown of the reactor (referred to as a reactor " trip" or " scram") is achieved by the fast insertion of the control rods into the core. To provide the capability to terminate the fission process in the event the control rods fail to be inserted, many (if not all) reactors are equipped with a supplementary method for injection of neutron-absorbing material on an emergency basis.

101 Id. at 16.

102 Applicants' Answer to OCRE Motion to Reopen the Record on Issue #16 (May 9, 1985), Affidavit of Edward C.

Christiansen at 4-5. At Perry, the TDI diesel generators do not employ air start check valves such as those that were found to be defective. Id. at 3.

s 41 Early in this proceeding, the applicants indicated that they would install a standby liquid control system (SLCS) as that supplementary method.103 It was unclear, however, whether the system would be automatically or manually initiated.104 Thereafter, Sunflower submitted a contention to the effect that the applicants should install an 05 automatic SLCS.

Following the publication of a Commission rule on June 26, 1984, requiring an automatic SLCS for boiling water reactors " granted a construction permit prior to July 26, 1984, that have already been designed and built to include this feature,,106 OCRE moved for summary disposit' ion of the contention and an order directing the applicants to automate the SLCS prior to exceeding five percent of full power.107 According to OCRE, the Perry SLCS came within the scope of the rule even though, at the time of its issuance, the SLCS was being constructed for manual, rather than automatic, 103 The SLCS supplies a highly-concentrated boron solution to the reactor core.

104 LBP-81-24, 14 NRC 175, 220 (1981).

105 Id. at 219-20.

106 This rule is codified as 10 CPR 50.62 and became effective July 26, 1984. See 49 Fed. Reg. 26,036 (1984).

107 OCRE Motion for Summary Disposition of Issue No. 6 (July 6, 1984).

42 initiation.108 OCRE claimed that the SLCS was designed with the capability for automatic initiation and that, by the applicants' own admission, the SLCS could be converted to automatic at a cost of only about $100,000.109 In OCRE's view, as read in light of its legislative history, the rule applied to a facility that has "the capability to be automated at low cost (i.e., before commercial operation)."110 The Licensing Board rejected this interpretation of the rule. Over the dissent of its then chairman, who thought that the rule should not be read "so inflexibly" as to exclude consideration of the relatively small cost of conversion,111 the other two members of the Board concluded that a " literal interpretation" of the rule compelled denial of the motion.112 As the majority saw it, the rule exempted from its requirement a SLCS that was not fully completed for automatic initiation at the time the rule became effective.

108 Id. at 2.

109 As the construction status of Unit 2 of the Ibid.

Perry plant was substantially behind that of Unit 1, the discussion of the Board and parties focused on the first unit.

110 OCRE Brief on the History and Intent of the ATWS Rule (September 7, 1984) at 11-12.

11I LBP-84-40, 20 NRC 1181, 1193 (1984).

112 Id. at 1188.

43 To the majority, it was immaterial that the incremental cost l of completing the conversion of the SLCS to automatic was l relatively low.113 In its view, the significant factor was that the Commission had decided to exempt from backfitting those plants "in an advanced stage of construction for which an automated SLCS has not been designed and built."114 The majority found that to be the case at Perry and, accordingly, denied OCRE's motion and dismissed the contention.

On appeal, OCRE admits that the plant would require further construction work before its SLCS could become operational in an automatic initiation mode. Nonetheless, it maintains that the Licensing Board erred in finding the i

plant exempt from the rule's requirement for an automatic SLCS. According to OCRE, the Board should not have applied the rule literally but, instead, should have given it a 11 In this connection, the majority was unwilling to accept, without further analysis, $100,000 as the cost of conversion as maintained by the dissenting Board member and OCRE. According to the majority, this figure did not take into consideration other costs such as " sunk costs" and

" costs of delay." Id. at 1189-90.

114 Id. at 1188. The majority also saw no "important unconsidered or unresolved issue of reactor safety" in deciding that the Perry plant was not required to convert its SLCS to an automatic initiated system. It assumed that in deciding to exempt certain plants from the requirement for automatic initiation, the Commission had found either

! type of SLCS to be safe, i

44 flexible interpretation consistent with its legislative history, as advocated by the dissenting Board member.115 We disagree. Unlike OCRE, we find no clear legislative history indicating that, in issuing the rule, the Commission intended the words "already designed and built" to have any other than its ordinary meaning. Those words plainly refer to an SLCS that is already completed for automatic initiation -- i.e., an SLCS that is " wholly ready" for operation at the time of the rule became effective.116 That this is so is clearly evident not only from the commission's use of the past tense of the words " design" and " build," but also from its addition of the word "already."

At the time of promulgation of the rule, the SLCS for each Perry unit was not "already designed and built" for automatic operation. In fact, the installation of the SLCS with a manual initiation feature at Perry Unit 1 was nearing 115 OCRE also argues that a literal interpretation

! would allow plants such as Perry to escape the rule's requirement for an automatic SLCS. A short answer to the argument is that the Commission did not see that as a safety problem, for it could have required backfitting of all plants without an automatic SLCS. We need add in this connection only that OCRE does not attempt to explain why the Commission should have deemed the absence of an automatic SLCS to constitute a threat to safety.

116 See Webster, New Collegiate Dictionary 34 (1977).

l l

.,,._.w-___. ._y. .,_.._.,.-n_,_--,-....,-,.__,,_._-.--__.____-.y _.

l 45 ll completion at that time and, according to unchallenged evidence, conversion to automatic initiation for the Perry Unit 1 SLCS would have required at least "the additional installation, modification or deletion of approximately forty cables, ten relays and numerous wires, switches, indicating lights and annunciators. 118 In the circumstances, we are satisfied that the Board's decision is amply supported by the record and fully in accord with the Commission's rule.119 We, therefore, affirm the Licensing 0

Board's decision on this score.

11 LBP-84-40, 20 NRC at 1185. See Applicants' Response to ASLB Request for Information on ATWS Rule and the Perry SLCS (September 7, 1984), Affidavit of Gary R.

Leidich.

110 LBP-84-40, 20 NRC at 1187. See also Applicants' 1

Response (September 7, 1984), Affidavit of Gary R. Leidich.

119 To reach this conclusion, we need not and do not decide whether the estimated $100,000 cost of conversion relied on by OCRE and the dissenting Licensing Board member accurately reflects the entire cost or, if' accurate, is so small that it can be considered incidental to an "already designed and built" SLCS.

120 OCRE also complains that the Licensing Board improperly assigned to the staff the responsibility for determining whether the SLCS at Unit 2 should be automatic.

There was no evidence, however, indicating that the SLCS for that unit, which received its construction permit prior to July 1, 1984, was any more designed and built for automatic initiation than the SLCS for Unit 1. Given the Board's decision for Unit 1, it was clear that the SLCS for Unit 2 was also exempted from the rule's requirement for automatic initiation. Thus, in reality, there was no question to be deferred to the staff for resolution.

i 4

- , . -- , - - < , , - - - , ~ , , - - . - , - ,m ~---------,-,---------,-,------r--------m.r,-,---

46 D. Turbine Missiles. Another issue raised below concerned the potential danger to safe operation of the Perry facility due to the placement and orientation of the plant's General Electric turbine generators. According to OCRE, while in operation parts of the turbine might break off and form missiles that, because of the turbine's orientation, could strike and damage structures, systems and components of the plant essential to its safe operation.121 Pursuant to 10 CFR 2.749(a), the staff moved for summary disposition of the issue in its favor.122 While admitting that the turbines were unfavorably oriented, the staff asserted that the risk of possible turbine missile damage at Perry is acceptably low.123 According to the 121 The breakup of the turbine could generally be .

caused by either (1) the running of the turbine at excessive speeds due to a sudden loss of its electric load (i.e.,

"overspeed"); or (2) stress corrosion cracking (which can result from a combination of corrosive elements and the relatively high stresses occurring not only at startup but during normal operation as well.) See Virginia Electric and Power Co. (North Anna Nuclear Station, Units 1 and 2),

ALAB-676, 15 NRC 1117, 1118-20, 1130 (1982).

' NRC Staff's Motion for Summary Disposition of Issue No. 13 (May 31, 1983). The required statement of material facts incorporated pertinent provisions of the SER, Supplement 3 (April 1983). In a supporting affidavit, a staff expert assumed as his own the statements contained in the Supplement concerning turbine missiles. This SER Supplement was admitted into the proceeding as Staff Exhibit

5. See Tr. 1224-25.

Staff's Motion, Statement of Material Facts, at (Footnote Continued)

o 47 staff, no General Electric turbines have broken as a result of crack propagation within a period of three years of startup. Further, during the same time interval, no cracks of greater than one-half the critical size have been observed in such turbines.124 In light of that prior history, the staff maintained that a turbine testing, inspection, and maintenance program will prove effective in protecting against missile damage. In this connection, the applicants were said to have agreed to carry out, on at least an interim basis, a staff-developed program along such lines. Within three years of commencement of plant operation, and following the completion of a General Electric study on missile generation probabilities in relation to time intervals for conducting inspections, the applicants are to submit their own testing, inspection, and maintenance program.

The applicants' response in support of the motion was accompanied by two affidavits. One was executed by two General Electric employees, D.P. Timo and L.H. Johnson, and (Footnote Continued) 2-3; Perry Safety Evaluation Report, Supplement No. 3 (April 1983) at 3-7.

I4 Staff's Motion, Statement of Material Facts, at 3.

A crack reaches critical size at that point when it might cause the turbine to fail. See North Anna, 15 NRC at 1123.

- - - , - - - . _ _ . - - . - - - - - . - - _ , . , , , , . - --,-,---y--.,

a 48 the other by an applicant employee, Edward J. Turk.125 The Timo-Johnson affidavit described in detail the turbine-generator's overspeed protection system and the program for testing and inspecting it, as well as the evaluation procedure for developing inspection intervals for the detection of stress corrosion cracking. That affidavit also declared that the inspection interval selected will be such that no existing crack might reach the critical size before the next inspection.

The Turk affidavit confirmed that, in Supplement 3 to the SER, the staff had established requirements for a turbine maintenance program for Perry. In particular, the program was said to require turbine inspection for stress corrosion cracking at intervals not to exceed approximately three years, or two refueling cycles. According to the affidavit, the applicants had also committed themselves to periodic testing of the overspeed protection system and testing and inspection of the turbine steam valves. j In opposing the motion, OCRE relied on a paper presented at a seminar on turbine missiles by Patrick G.

Heasler of Battelle Pacific Northwest Laboratories, in which the author estimated a greater probability of turbine 1

Applicants' Answer in Support of NRC Staff Motion for Summary Disposition of Issue No. 13 (June 27, 1983).

1 1

o 49 failure during the first year of operation than had been estimated by the staff.126 OCRE did not, however, supply a supporting affidavit of Mr. Heasler. Rather, the sole affidavit attached to OCRE's response was that of its representative, Susan L. Hiatt. In it, she did not address the merits of the motion but, rather, simply set out a claimed need for additional time to conduct further discovery and to study materials OCRE had recently received from the NRC and the applicants.

Concluding that there was no good reason to withhold action on it, the Licensing Board granted the staff's motion. In doing so, the Board noted that the Heasier paper was not supported by an affidavit that would " establish its admissibility."127 Nonetheless, the Board decided to consider its content "because of the significance of granting a motion for summary disposition,.128 and the importance of the safety question involved. Focusing on 1

OCRE Response to NRC Staff's Motion for Summary Disposition of Issue #13 (June 23, 1983), at 3. We have I

considered the other documents referred to by OCRE in opposing the summary disposition motion and conclude that none has significance concerning this issue.

j 127 LBP-83-46, 18 NRC 218, 222 (1983).

128 Ibid.

- , - , - , - - - , , - - - - - - . - - - - - - - - - , - , - - - , - ---,-,w - ------ -- - - - -

~>-.w- . , a --- , ,

a 50 cracking due to stress corrosion,129 the Board compared the findings in the Heasler paper with those in an earlier article on the subject written by S.H. Bush, then also of O

the Battelle Pacific Northwest Laboratories. For the reasons explained, it thought the Bush article "more credible."131 In addition, the Board observed that the 129 According to the Licensing Board, OCRE was not challenging the overspeed aspect of the inspection program; it was challenging only that part of the program dealing with stress corrosion. Id. at 219-20. On appeal, OCRE does not dispute this Board conclusion, but explains that its limited challenge was due to the unavailability of information on the overspeed protection program until the information was presented by the applicants' June 27, 1983 response supporting the staff's motion. Because OCRE does not challenge the Board's view of the scope of the contention, we thus confine our review to the stress corrosion aspect of the contention.

130

-Id. at 222-25. The Bush article was first mentioned in an applicants' filing. It was later furnished to the Licensing Board and the parties by the applicants at the Board's request. Id. at 221, n.13.

131 These reasons included the fact that the Bush article was published in a " refereed journal" while the Heasler paper was "in the nature of a draft" and was not similarly published. Id. at 224. The Board also noted that:

Bush was a member of the Advisory Committee on Reactor Safeguards from 1966 to 1977 and was its chair in 1971. Bush presents a variety of assumptions in carefully presented statistical form. His article contains

" comments" that in two instances indicate that at least two events could not properly be considered by Heasier as reflecting adversely on operating experience with turbines. These events occurred during a (Footnote Continued)

I 1

4 51 Timo-Johnson affidavit filed by the applicants in support of the staff's motion corroborated Bush's article. According i

to the Board:

In [that] affidavit [], D.P. Timo and L.H. Johnson present the detailed, empirically based analysis of turbine missile failures that this Board relied on at the outset of this opinion. We note that this affidavit, which has not been controverted, postdates Bush and Heasler and derives support from research results that were not available to them. Given its later date, there may well be other data available to its authors that were not previously available. In this instance, we need not prefer the Timo-Johnson analysis to Bush's.

We merely accept their ggglysis as additional corroboration for Bush.

On this basis, the Licensing Board decided that the Heasier paper was insufficient to raise a genuine issue of material fact concerning any possible danger to the facility from turbine missiles.

Before us, OCRE claims that the Licensing Board's dismissal of the issue was in error for a number of reasons.

According to OCRE, the Board (1) decided the turbine missile 4

issue with respect to only the first three years of plant operation, while improperly referring the long-term aspects (Footnote Continued)

" factory test" or were "preoperational."

Heasler's article does not explain why it was appropriate to include these events in his

! analysis.

l Ibid. (footnotes omitted).

12 Id. at 225.

-~,.m -,.ne.,- r-- -

,----.-,e--.e,-- ,.,,,---.--n.r,-r- -,-, - -,, ,.-n.,,-.n.c, .

,,-...-n,, -..--.rn--, - - - - - - - . , , ~--n.-----v, ,n n ----n.-r- -,--. ,v

a b

52 l

1 of the issue to the staff for post-hearing resolution and l l

thus depriving it of a right to a hearing under Section 189a. of the Atomic Energy Act; (2) did not base its decision on reliable evidence in the record; (3) applied the wrong legal standards for adjudging summary disposition motions; and (4) improperly denied its request for a delay in deciding the motion to allow it time for further discovery and study of the issue.

We find each of these claims to be without merit.133

1. Contrary to OCRE's assertion, the Licensing Board did not decide the turbine missile issue with respect to only the initial three years of plant operation. Upon its consideration of the evidence presented by the parties, the Board concluded that there was no reason to question the adequacy, over the long term, of the inspection, testing, and maintenance program established by the staff for the Perry facility.134 That program, as the Board indicated, will not necessarily lapse at the end of the three-year period, but will remain in effect unless and until the 13 In opposing OCRE's appeal, the applicants maintain that the turbine missile issue was improvidently admitted by the Licensing Board, thereby implying that OCRE's appeal was moot. Applicants' Brief in Opposition to Intervenors' Appeals from the Concluding Partial Initial Decision (December 2, 1985) at 80-82. We need not address this claim in view of our disposition of OCRE's appeal.

134 LBP-83-46, 18 NRC at 220.

n----

.m w. , - - - - + --,.e ,, -- , , , - . - . --7-,-n,. .

.-._,e,-~,- . . , v-. . - , - - , - , , . . , - ,, - , , -

53 applicants put forth a revised program which is found satisfactory by the staff.135 (Although, as earlier noted, the applicants are obliged to submit a program at the end of three years, nothing precludes them from adopting the staff's program as their own if experience has shown that course to be warranted.136)

The fact that the staff may be called upon to approve a revised program does not mean, however, that there has been an improper delegation of adjudicatory functions to it. One of the staff's major continuing responsibilities is to monitor the operation of all nuclear power facilities and, thus, to review all proposed changes in existing inspection, testing, and maintenance programs at such facilities.137 7f OCRE becomes concerned with any modification to the turbine program that should be suggested by the applicants, it may raise those concerns by means of a petition for action under 10 CFR 2.206.138 In that petition, OCRE can advance any

! reasons it might have for believing that a hearing should be 135 See supra p. 47.

136 Ibid.

137 See, e.g., 10 CFR 50.59.

I 138 We assume that, should they be asked to do so, the applicants will agree to advise OCRE of any proposed modification to the current program.

e 54 held to consider the appropriateness of the applicants' revisions.

2. The affidavit-supported statement of material facts upon which the staff's summary disposition motion rested brought to light the staff's determination that its testing, inspection, and maintenance program would avoid the generation of turbine missiles. To oppose the motion, OCRE was obliged to set forth " specific facts," by affidavit or other appropriate means, to establish that "there is a genuine issue of fact. 139 While citing the Heasler paper, however, OCRE failed to support it with a suitable affidavit or otherwise.140 139 10 CFR 2.749(b). More completely, the relevant portion of this paragraph reads:

When a motion for summary decision is made and supported as provided in this section, a party opposing the motion may not rest upon the mere allegatiens or denials of [its) answer; [its]

answer by affidavit or as otherwise provided in this section must set forth specific facts showing that there is a genuine issue of fact. If no such answer is filed, the decision sought, if appropriate, shall be rendered.

Unless properly controverted, the regulations specify that

"[alll material facts set forth in the statement required to be served by the moving party will be deemed to be admitted j . . . .

10 CFR 2.749(a). See Houston Lighting & Power Co.

(Allens Creek Nuclear Generating Station, Unit No. 1),

ALAB-629, 13 NRC 75, 78 (1981).

140 We note in passing that there is no room for

! lingering doubt that a properly designed and implemented (Footnote Continued)

[

O G

55 OCRE points out that the Licensing Board cited the Bush article in granting summary disposition, even though it too was not supported by an affidavit. We agree that, in common with the Heasler paper, Dr. Bush's article could not serve as a basis for deciding whether a genuine issue of material fact existed. But we are totally persuaded that the content of the staff's and applicants' affidavits alone was sufficient to justify the conclusion that no issue of material fact was present. Accordingly, the Board's reference to the Bush article was unnecessary and, at most, constituted harmless error.141 (Footnote Continued) testing, inspection, and maintenance program is effective in assuring the safety of plants from turbine missiles. As observed in North Anna, "it is possible, utilizing empirical data, to determine with reasonable certainty the length of time that will elapse before an initiated stress corrosion-induced crack might reach critical size." 15 NRC at 1132. While the Perry turbines were manufactured by a company different from that responsible for those at North Anna, we know of no reason why this observation on the value of empirical data is inapplicable here. Moreover, the Heasler paper contains nothing that would prompt a reexamination of our North Anna conclusion.

141 In SER Supplement 3 (January 1986), the staff discusses its shift in emphasis from the calculation of the probability that generated missiles will strike important structures, systems, and components to the potential for turbine failure and its prevention. See Staff Exh. 5 at 3-1 to 3-7. OCRE claims that the Licensing Board erred in j relying on this assertedly " preliminary and unapproved" l staff position. OCRE Brief at 46. While correctly noting l

that the staff's position in the SER differs from that in a l

regulatory guide and a section of the Standard Review Plan, j OCRE does not suggest any deficiency in the new position and (Footnote Continued)

o 56

3. OCRE's next complaint is that the Licensing Board erred in placing the burden of proof for summary disposition on it rather than on the staff as the movant. It did so, according to OCRE, by requiring a supporting affidavit for the Heasler paper and by accepting the staff's and applicants' statements "without a moment's hesitation" while subjecting those of OCRE "to the most exacting scrutiny."142 As explained earlier, however, the staff's motion and the applicants' response were supported by a statement of material facts and appropriate affidavits as called for by the rules; OCRE's response in opposition consisted of arguments only, without proper evidentiary support. In the circumstances, the staff's and applicants' statements were entitled to be considered by the Licensing Board while those of OCRE were not.

4. Finally, OCRE complains about the Licensing

! Board's rejection of its request for additional time to respond to the staff's summary disposition motion. Citing

(

l l

(Footnote Continued) we see none. Further, regulatory guides and the Standard Review Plan do not have the status of Commission regulations and are subject to changes by the staff. In the circumstances, we conclude that the Licensing Board did not err in its reliance on the staff position in SER Supplement l

l 3.

142 OCRE Brief at 44-45.

O 57 10 CFR 2.749(c),143 OCRE insists that the Board failed to 1

apply the rules liberally in accordance with comparable federal judicial practice and thus erred in refusing to allow it to conduct further discovery and to provide it with time for further study of the contention.

In her affidavit in support of the postponement request, OCRE's representative, Ms. Hiatt, alluded to the need for time to analyze documents she had recently received from the applicants and the NRC. She also cited the need to obtain further data, including the ultimate results of the ongoing General Electric study then in progress. In rejecting the request, the Licensing. Board pointed out that

"[w]hatever that study may say, applicant [s] [are] bound to an inspection and maintenance program as to which there is no genuine issue of material fact."144 As to the asserted need for further time to analyze the documents, the Board concluded that OCRE's insistence upon an additional six 143 ihat section reads:

Should it appear from the affidavits of a party opposing the motion that [it] cannot, for reasons stated, present by affidavit facts essential to justify [its] opposition, the presiding officer may refuse the application for summary decision or may order a continuance to permit affidavits to be obtained or make such other order as is appropriate and a determination to that effect shall be made a matter of record.

144 LBP-83-46, 18 NRC at 226.

58 months was unjustified. In addition, the Board determined that OCRE had had adequate time for discovery.145 Inasmuch as cched'uling is a matter of Licensing Board discretion, we do not inject ourselves into scheduling controversies, absent a "truly exceptional situation. More particularly, we ' enter the scheduling thicket cautiously' and then only 'to entertain a claim that a [ licensing] board abused its discretion by setting a hearing schedule that deprives a party of its right to procedural due process.. 146 OCRE has not demonstrated that it was denied due process by the Board's actiSu, nor even claims such

~

injury.147 In the circumstances, we find no basis for disturbing the Licensing Board's rejection of OCRE's postponement request.

145 Ibid.

146 Virginia Electric and Power Co. (North Anna Nuclear Power Station, Units 1 and 2), ALAB-584, 11 NRC 451, 467 l (1980) (quoting Public Service Co. of Indiana (Marble Hill I Nuclear Generating Station, Units 1 and 2), ALAB-459, 7 NRC 179, 188 (1978)).

j 147 It should be noted that, according to our information, the General Electric report that OCRE desired had still not been submitted to the staff as of early this year. Therefore, a six-month delay beginning in June 1983 (as requested by OCRE) would not have yielded any additional information regarding that report.

l l

l i

b 59 E. Polymer Degradation. Nuclear power plants use polymers (material generally having the characteristics of plastics or rubber) in various applications, at least some of which have a bearing upon safe operation.148 When exposed to radiation over an extended period of time, the molecular structure of polymers can be affected in a manner that results in changes in specified properties, such as embrittlement or reduced electrical resistance. In order to ensure that electric cables and other equipment utilizing polymers will function properly in their radiation environment, utilities test the radiation-resistance cf these polymers to determine at what point their replacement would be necessary.

During the course of the proceeding below, at the behest of OCRE, the Licensing Board admitted a late-filed contention that provided:

Applicant (s] ha[ve] not demonstrated that the exposure of polymers to radiation during the prolonged operating history of Pe{gg would not cause unsafe conditions to occur.

14P' For example, polymers serve as insulation for safecy-related electrical wiring and are also found in the seals and gaskets of safety-related pumps and valves.

14S LBP-82-53, 16 NRC 196, 202 (1982). The contention was prompted by research performed at Sandia National Laboratories, which suggested that polymers in use at nuclear plants (such as Perry) might degrade more rapidly than previously thought. In this connection, the Sandia studies found that when different samples received equal (Footnote Continued)

O 4

60 Subsequently, the staff filed a motion for summary disposition of the contention.1 0 One of the assigned grounds was that the applicants had committed themselves to carry out a surveillance and maintenance program that would allow replacement of significantly degraded equipment before it could become a problem and that, therefore, exposure of polymers to radiation during operation of Perry will not bring about an unsafe condition.

For a variety of reasons, the Licensing Board granted the staff's motion. None of those reasons invoked, however, the applicants' commitment to maintain the surveillance and maintenance program.1 1 It may well be, as OCRE insists, that the Licensing Board's rationale will not withstand analysis.152 (Footnote Continued) doses of radiation, the polymers that received the dose at a l lower rate experienced greater degradation than those that j received the dose more rapidly. See id. at 200; LBP-83-18, l

17 NRC 501, 502-04 (1983).

150 NRC Staff Motion for Summary Disposition of Issue

1. 9 (January 14, 1983).

151 LBP-83-18, 17 NRC at 512, as orally modified at Tr.

810-28.

l For example, we encounter difficulty in accepting the Board's basis for disposing of the question concerning l

polymers used in mechanical equipment. The Board determined that OCRE had "not demonstrated the existence of a genuine issue of fact concerning the relationship between the more rapid degradation of polymers (that are not used for electric insulation) and the safety of the Perry plant."

(Footnote Continued)

b 61 Nonetheless, we are satisfied that the Board's result on the staff's motion was correct -- i.e., that there is no genuine issue of material fact respecting whether a significant safety problem might arise from polymer degradation.

The record establishes without contradiction that serious polymer degradation does not develop overnight. To the contrary, it affirmatively appears that the material would have to be subjected to a radiation environment for a number of years before becoming degraded to such an extent that a safety problem might arise.153 Thus, the existence (Footnote Continued)

Id. at 507. As we previously stated in the construction permit proceeding, the party seeking summary disposition on a particular matter has the burden of establishing that no genuine issue of material fact exists on that matter.

ALAB-443, 6 NRC 741, 753 (1977). It was thus the obligation of the NRC staff and the applicants to show that degradation of mechanical polymers would not cause a safety problem.

Without finding that the staff and the applicants had made such a showing, however, the Board disposed of this issue on the strength of the intervanor's failure to make a contrary showing.

Further, in reaching its conclusion on this issue, the Board stated that there is a " natural inference" that degradation of polymers used in electrical systems would cause safety problems (although, curiously, the Board was unwilling to assume, "without evidence," that degradation in

"[s]eals, gaskets and the like" would give rise to similar concerns). LBP-83-18, 17 NRC at 507 n.16. We do not endorse the Licensing Board's view respecting what inferences, if any, exist with regard to the relationship between safety problems and polymer degradation.

153 Using figures found in the Sandia reports upon which OCRE relied, together with the maxinum dose rate at Perry supplied by the applicants, the Licensing Board (Footnote Continued)

o 1.

62 of a reliable inspection and maintenance program should suffice to provide timely detection and correction of such a problem.-

With their answer in support of the staff's motion for summary disposition, the applicants submitted an affidavit that discussed a surveillance and maintenance program for Perry designed to " provide assurance that radiation degradation of polymers in safety related equipment will either be prevented, or discovered and corrected, before it can cause unsafe conditions to exist."154 The affidavit indicated that the program would be completed prior to fuel loading of Unit 1.155 According to the affidavit, "[olne (Footnote Continued) calculated that "50% degradation" in some of the properties of the polymer materials will not occur for approximately nine years. Id. at 509. The Board also estimated that it would require six years of continuous exposure at 357

, rads / hour before "significant" radiation effects would appear. Id. at 508. More recently, the applicants supplied an affidavit to the effect that the highest radiation dose rate to which polymers in Perry will be exposed is now considered to be 160 rads / hour rather than the 357 rads / hour earlier reported to the Licensing Board. Applicants' l Response to Licensing Board's May 9, 1983 Order concerning l Issue No. 9 (August 4, 1983), Affidavit of David R. Green at l 2 & n.1. Thus, the time intervals before degradation would occur are even longer than those estimated by the Licensing Board. 1 154 l Applicants' Answer in Support of NRC Staff Motion for Summary Disposition (February 8, 1983) at 4. Although the staff's motion had referred to the program as involving electrical equipment, the applicants' description indicated that it covers all safety-related equipment.

155 Id., Affidavit of David R. Green at 1.

t

4 4

63 function of the program will be to detect equipment degradation, including degradation to polymeric materials from radiation,"1 6 and that periodic perfornance tests will be " performed to monitor system and/or component . . .

operation, and determine unacceptable component degradation."1 Hence there is no dispute that, as the staff asserted in its motion for summary disposition, the applicants have committed themselves to a program of inspection and maintenance.158 OCRE has not endeavored to explain why the program might not accomplish its intended purpose. Further, no deficiencies appear to us. In the circumstances, the record provides a sufficient foundation for the conclusion that polymer degradation does not pose a threat to safe facility operation. r l

l 156

d. at 2.

157 Id. at 3.

158 The program has, in fact, been fully developed and is now in place. In a letter dated April 25, 1986, responding to our inquiry regarding the status of the program, counsel for the applicants stated that "[t]he development of the surveillance and maintenance program

[ described in an affidavit of David R. Green] was completed prior to fuel loading of Unit 1. Implementation of the program is underway and will continue throughout the life of the plant." Letter from Jay E. Silberg to Appeal Board (April 25, 1986) at 2.

l

a d

64 F. Air Lock Testing. To protect against the uncontrolled release of radioactivity, the reactor containment of a facility such as Perry must be essentially leakproof.159 To this end, access to the containment is achieved through the means of an " air lock" -- i.e., a compartment with two airtight doors, the outer one of which is closed and sealed before the inner one is opened to allow entry.160 If the air lock is opened during a period when containment integrity is required (e.g., during power operation or hot shutdown conditions), it must be tested within three days thereafter. If, however, the opening 4

occurs when containment integrity is not required (e.g., the reactor is in a cold shutdown condition with no fuel movement), the test may be deferred until such time as that integrity is once again necessary.161 The reactor containment is the structure that encloses all those pressure-containing components of boiling and pressurized water nuclear power reactors (such as the reactor pressure vessel, piping, pumps, and valves) that are part of the reactor coolant system or connected to that ,

system to some extent. 10 CFR Part 50, Appendix J, Section l II.A.

See Perry Final Safety Ar:alysis Report [FSAR],

Amendment 14 (August 22, 1984), at 6.2-67.

I 10 CFR Part 50, Appendix J, Section III.D.2(b).

b 65 There are two recognized means of testing air locks.162 The simpler, and thus less time-consuming, procedure involves the application of pressure only to the air lock seals themselves for the purpose of ascertaining whether those seals are tight (seal testing). The more elaborate procedure calls for the pressurization of the entire chamber between the two air lock doors, followed by the measurement of any leakage from the chamber.

In terms, the regulations allow seal testing in circumstances where the three-day test requirement applies (i.e., when containment integrity is then-mandated). For some unexplained reason, however, that explicit permission does not cover the case where the test need not be conducted within three days (i.e., when containment integrity was not required at the time of the air lock opening). As to that situation, the regulations are wholly silent respecting i

i whether seal testing would be sufficient.

Desirous of utilizing seal testing in connection with all air lock openings, the applicants requested, under 10 CFR 50.12(a), an exemption from so much of the regulations as might be construed as implicitly directing pressurization of the entire air lock chamber where the three-day testing I

Perry FSAR (Amendment 14, August 22, 1984) at 6.2-101 to 6.2-102.

o O

66 requirement does not come into play.163 In support of the request, the applicants in essence maintained that no safety consideration dictates full chamber pressurization simply because the test need not be performed within three days of the air lock opening but, rather, can be postponed to such date as containment integrity once again is necessary.

163 See letter from Murray R. Edelman, Vice President-Nuclear Group, The Cleveland Electric Illuminating Company, to B.J. Youngblood, Chief-Licensing Branch No. 1, Division of Licensing, U.S. Nuclear Regulatory Commission (April 8, 1985), attached to OCRE's Motion to Reopen the Record and to Submit a New Contention (July 5, 1985).

At the time of the applicants' request, Section 50.12(a), provided in pertinent part:

The Commission may, upon application by any interested person or~upon its own initiative, grant such exemptions from the requirements of the regulations in this part as it determines are authorized by law and will not endanger life or property or the common defense and security and are otherwise in the public interest.

Effective January 13, 1986, Section 50.12 was amended in several respects. The above-quoted portion of subsection (a) now reads:

(a) The Commission may, upon application by any interested person or upon its own initiative, grant exemptions from the requirements of the regulations of this part, which are --

(1) Authorized by law, will not present an undue risk to the public health and safety, and are consistent with the common defense and security.

50 Fed. Reg. 50,764, 50,777 (1985).

0 h

67 Through the vehicle of a motion to reopen the record to allow it to introduce a new contention, OCRE sought to challenge the exemption request as unauthorized by law.164 The Licensing Board denied the motion.165 We concur in that result.

Contrary to OCRE's insistence, we find nothing in the Atomic Energy Act that might preclude the grant of the sought exemption. More specifically, OCRE points to no provision in that Act that could possibly be construed as forbidding the commission to allow these applicants to use seal testing in all instances of air lock opening. Nor is there any merit to the intervenor's argument that the exemption request is tainted because it rests, at least in part, on a desire to reduce the cost of testing. In recently revising Section 50.12,166 the Commission recorded its belief that judicial precedent and long-standing Commission practice confirm that, within the confines of carrying out its paramount responsibility to protect economic public health factors in its and safety,making.y decision it may g9nsider 164 OCRE's Motion to Reopen at 2-4.

165 LBP-85-33, 22 NRC 442 (1985).

166 See supra note 163.

10 50 Fed. Reg. at 50,767.

i

, ,, , . - - - - . - -.,,-_..n. , . , - . - - . _ - . , _ . - . , ,--a , - , - , , , -

,w -

o 68 The result reached in LBP-85-35, 22 NRC 514 (1985), is affirmed.168 The staff shall ensure, however, that the final analysis of the hydrogen control system includes both (1) a more detailed review of containment heat removal capability; and (2) a further consideration of the potential for and effects of the release of combustible gases from heated cable insulation. See supra pp. 19-20, 21.

It is so ORDERED.

FOR THE APPEAL BOARD k,

C. Je Q$=: Nb Sh6emaker Secret ry to the Appeal Board The concurring opinion of Dr. Johnson follows.

168 The substantial majority of the issues raised before and decided by the Licensing Board were encompassed by the appeals. Our examination of the relatively few substantive determinations not appealed has disclosed no error requiring corrective action.

w 69 Concurring opinion of Dr. Johnson:

Although joining in substantially all of the foregoing opinion, as well as in the result, I part company with my colleagues on one point. They have reserved decision on whether an intervenor may challenge the staff's approval of the accident scenarios used in the analysis of an applicant's hydrogen control system (see supra p. 13). I would however decide the issue now, for it is clear to me that the Commission intended to preclude from exploration in licensing hearings the details of those scenarios that lead to the generation of large quantities of hydrogen. 10 CFR 50.44(c)(3)(vi)(B)(3) states that the evaluation of the hydrogen release must "[ulse accident scenarios that are accepted by the NRC staff." An interpretation of this statement that would foreclose intervenor challenges to hydrogen generation scenarios is not only reasonable, but compelled by the unusual circumstances that the rule itself addresses.1 In any consideration of the hydrogen rule, one must bear in mind that the rule was promulgated to provide a I

Under my interpretation of the hydrogen rule, OCRE's claims on appeal that relate to the hydrogen generation scenarios (e.g., functioning of the RCIC, station blackout, and containment spray operability) or to the consequences of a failure or delayed operation of the hydrogen control system would necessarily be denied. See OCRE Brief at 15, 17, 18, 20.

r, l e, ,

I l

70 mitigating capability, beyond the safety systems already required, for unlikely and unexpected nuclear power plant accidents that could lead to the generation of large quantities of hydrogen. In the rule's most significant i

provision, the Commission adopted as an upper bound that {

l amount of hydrogen generated by the oxidation of 75% of the i fuel cladding. In order to establish a framework for an analytical evaluation of the functioning of the hydrogen control system and the survivability of the safety-related equipment exposed to a hydrogen burning environment, the Commission further directed that staff accepted accident scenarios be considered. Following a mechanistic scenario enables the analyst to determine the time dependence of hydrogen generation in the reactor core and the time-and-spatial dependence of its subsequent release and buildup in various locations within the containment -- input information necessary for the analysis.

The emergency core cooling system (ECCS) of a nuclear power plant is designed pursuant to the requirements of 10 l

CFR 50.46 to cool the reactor core in the event of a LOCA and, in particular, to limit the hydrogen produced to that resulting from oxidation of less than 1% of the fuel cladding. But the hydrogen' rule specifies a hydrogen yield amounting to oxidation of 75% of the cladding. Hence any accident scenario considered under the rule envisions failures of emergency core cooling equipment well beyond l

i

c ,

e

(

71 those that would be expected for redundant systems and could ordinarily be litigated in licensing proceedings. A contention proposed by an intervenor that postulated such failures would be deemed to challenge the ECCS rule and thus would not be allowed under 10 CFR 2.758(a).2 It is therefore unlikely that the Commission intended to permit intervenors, on the one hand, to propose their own hydrogen generation scenarios under the hydrogen rule while, on the other, prohibiting the postulation of such scenarios as a challenge to the ECCS rule.

As a practical matter, there are many scenarios that could lead to hydrogen generation. But their sole purpose is to provide the basis for a severe analytical test of the hydrogen control system and the survivability of safety systems. The NRC staff is best qualified to determine which scenarios provide this severe test.3 2

10 CFR 2.758(b) and (d) provide the route by which an intervenor may challenge an NRC rule. With regard to the hydrogen rule, if an intervenor can make a prima facie case that some accident scenario other than one " accepted by the staff" would provide a more severe test for the hydrogen control system, hence the rule is not serving its intended purpose, Section 2.758(d) provides that the question be put before the Commission for a possible waiver (i.e., that the intervenor proposed scenario be considered rather than, or in addition to, the staff accepted scenarios).

3 And, as pointed out in the footnote above, there exists a mechanism for considering an intervenor's scenario if it can be shown to provide a more severe test of the e hydrogen control system.

n O.

72 Finally, as noted, the rule sets a conservative upper limit on hydrogen production -- that due to 75% cladding oxidation. In order to meet this provision, however, an accident scenario must generally include the arbitrary assumption that the ECCS recover in time to terminate the metal-water reaction at this point. It is understandable that the rule would require the postulation of scenarios that assume arbitrary (and unlikely) failures and recovery of equipment in order to set up the mechanistic framework for an analysis of the hydrogen control system. I believe the Commission recognized, however, that this is an exercise wholly dependent upon technical and scientific expertise, not amenable to an adjudicatory hearing, conducted under the rules of evidence.