PROBABILISTIC RISK ANALYSES

Mr. Ahearne, what role do probability risk assessments have in safety analysis and licensing?

Mr. AHEARNE. At the present time the NRC does not have them as a formal part of the licensing process, but has encouraged all of the plants to go through these probabilistic risk analyses. The main thing that they do, they identify system weaknesses, vulnerabilities, where are problems that might be present that one had not expected. It has turned out to be very valuable, sufficiently valuable that most of the industry is now going forward to doing them.

Senator MCCLURE. According to Department of Energy documents, HWR safety research at the Savannah River plants will take 4 years. If HWR is the chosen NPR technology, do you believe that these safety analyses should be completed before construction begins?

Mr. AHEARNE. I think that there are several types of safety analyses you are speaking about. There are analyses being done as was already mentioned by Secretary Salgado on the existing Savannah River plants, and PRA's, probabilistic risk analyses, are being done on those.

Those are safety analyses focused upon existing designs, existing plants, taking them where they are, as he mentioned, they are 30-yearold plants. Those are the kind of analyses that are going to have to be completed to give some high confidence that those plants could go up in power. Separate analyses would probably be required for a new design.

Senator MCCLURE. Is that because the new design would be sufficiently different from the older?

Mr. AHEARNE. It would be different. Obviously, you can capture some of the work that has already been done, some of the work will be directly applicable, but I don't think that automatically you can assume that the work being done for the older design is going to be applicable to the new design.

Senator MCCLURE. In your opinion, are light water reactor codes easily adapted for use with heavy water technology?

Mr. AHEARNE. No.

Senator MCCLURE. You can't simply transfer the codes.

Mr. AHEARNE. The light water codes, it is not just the moderator which is a light water, they were really designed to handle the temperatures and the pressures, and the rest of the system fluid flow, and the type of fuel that exist in light water reactors.

The heavy water reactor is quite different, and I do not believe that it is a simple matter to take those codes and transfer them to a new reactor. One of the big issues that one has in any code work is to then validate the codes, and the more complex the code is, the harder time one has in validating; that is, to give confidence that the results of the codes are things that you want to believe.

DEFENSE DEPARTMENT CONCERNS

Senator MCCLURE. Dr. Duncan, you have been kind of left alone down there at the end of the table. I don't want you to leave here without any questions.

We have kind of an understandable, but, nevertheless, unique relationship here. You are the customer.

Dr. DUNCAN, Yes.

Senator MCCLURE. You are not the producer. So we have one government agency that is responsible for the production and another government department that sets the production requirements. You don't tell DOE how to do their business, and they don't tell you what it is you need. Is that fair to say?

Dr. DUNCAN. Yes; we are the sole customer for the particular subject. that we are talking about here.

Senator MCCLURE. There are historical reasons that it evolved that way, and I think they are imbedded in our public policy, and I support that public policy. I don't think that anybody is trying to change that. That relationship does work well, it has worked well, and I think it promises to continue to work well.

Dr. DUNCAN. We are very pleased with the relationship and it has worked well ever since World War II.

Senator MCCLURE. I do note that in your statement, you stress that your primary concern is that, whatever they do, they come up with a plan that will give you an assured supply of your needs.

Dr. DUNCAN. A safe and assured supply of our primary need. Then, of course, cost effectiveness is important to us; that is, the cost of the tritium should be minimized consistent with safety and reliability of supply. That is our position. We do not involve ourselves deeply in the specifics of how they do that because that is their responsibility as a supplier to us, and seem to have lots of help with this.

Senator MCCLURE. They have lots of advice.

In that process, obviously, as you look at those issues of safety and reliability, and assured supply, you do look at various aspects. Even though you trust them, you are looking over their shoulder, too, aren't you?

Dr. DUNCAN. Yes, we are.

Senator MCCLURE. So you trust them a little.

Dr. DUNCAN. If they would do something that would concern us, we would tell them about it. We have a mechanism for doing this: the Nuclear Weapons Council, of which Mr. Wade and I are members, along with General Herres, the Vice Chairman of the Joint Chiefs of Staff.

Senator MCCLURE. In that process, I know that you have received advice too, and that deals with questions of redundancy. That is not an unfamiliar term for people in the Department of Defense. We often times look at multiple suppliers, or two or more, to make certain that we have an industrial capacity to meet the needs of production. Is that also something you have looked at with respect to the materials supply?

Dr. DUNCAN. We have stated that we would like a safe and assured supply, and if they choose redundancy of sources and redundancy of technology to do that, we would give that serious consideration as the right way to do it.

Senator MCCLURE. I noted that statement on page 3 of your statement. Obviously, I welcome that because that is my major concern. I really am concerned that those who want us to build one reactor, with 100 percent of the supply at one reactor, are flirting with disaster for the security of this nation.

I don't know how to stress it more, and I note that the ERAB panel also recommended diversity. I think that almost every study that has even been done says that we ought to have a capacity that isn't tied up in the success or failure of one facility. I stress that because to me that is the No. 1 concern that I have as we go through this.

It isn't so much the choice of technology as much as it is making certain that we have something that works, and that we don't put all of our eggs in one basket and that we don't take the risk that one may prove to have problems. The embrittlement problem on the C Reactor was, we know now, a design flaw; it was not something that we anticipated. We didn't know it 5 minutes before we discovered it, and we couldn't anticipate it.

This is why I went into a little bit earlier that had we had all four reactors with precisely the same design that developed precisely the same problem, in that same period in their life history, we would be in trouble today. I again want to reiterate what I think is your concern as well, that we don't embark upon a program that exposes our future to the hazard of that kind of an unforeseen happening. We have enough trouble designing around those that we can foresee without getting into those that we cannot.

I don't believe, Dr. Duncan, that DOD has taken a position with respect to how to achieve the reliability.

Dr. DUNCAN. We have not taken a position with respect to how to achieve it. One gets reliability in rocket systems or in weapons systems by parallelism. Sometimes this parallelism is taking place at the component level or the subsystem level and not at the total system level, and sometimes it is taken at the total system level. The thought of two separate reactors for reliability is parallelism at the total system level.

We didn't want to take a specific position as long as we had a safe and assured supply, thinking that, perhaps, the Department of Energy, in their desire to do a cost-effective effort, might come up with some way of getting that reliability where the parallelism takes place in subsystems, or whatever.

We are quite concerned, and it is very necessary that we have a safe and assured supply.

Senator MCCLURE. Are you in a position to say whether or not you see any disadvantage in having half of your suply come from one reactor and half from another?

Dr. DUNCAN. I see no disadvantage in having half come from one and half come from another unless it runs up the cost of the tritium itself to a very high level.

Senator MCCLURE. To a very high level?

Dr. DUNCAN. Yes.

Senator MCCLURE. I would assume, and I think the preliminary cost figures, and they are simply that, are that it will cost modestly more to build two reactors at two sites, whether or not they are the same technologies, but that that increased cost is marginal. A marginal cost increase would not be something that DOD would object to?

Dr. DUNCAN. No, we don't. I would say that we would prefer a marginally higher cost for the product if we were assured always that we would get it than to have the lowest cost product and then all of a sudden not have it.

DETAILED REACTOR DESIGN

Senator MCCLURE. Mr. Ahearne, you made a comment a while ago, which I think requires at least some further comment. You suggested that you had not reviewed detailed designs in making any comment with respect to safety. I think that it is also safe to say that ERAB didn't either, because there isn't detailed design on any of these plants, with the possible exception of HTGR and the possible exception of a commercial light water reactor in which detailed design is available.

Isn't that correct; there is not a detailed design of a new HWR?
Mr. AHEARNE. That is correct.

Senator MCCLURE. There is not a detailed design of a liquid metal reactor.

Mr. AHEARNE. I am not certain, but I don't believe so.

Senator MCCLURE. You are not the only one who didn't have the opportunity to make a review of detailed design.

Dr. Graham, the National Academy of Sciences report on DOE's production reactors concluded that the existing level of understanding of severe accident behavior for the production reactors is inadequate to permit a realistic assessment of these designs to mitigate the consequences of severe accidents. Are you aware if this situation has changed?

Mr. GRAHAM. Since Secretary Herrington requested that study, Senator McClure, and since it was submitted to him, I know the Department of Energy has put into place a substantial program to address just those issues, and they have that program underway today. Department of Energy representatives could give you a more precise status than I could, but as I understand it, it will continue for at least the next 3 years.

CLASS IX ACCIDENT

Senator MCCLURE. In particular, do you know if the Savannah River safety experts have resolved the issue of recriticality with a class IX accident?

Mr. GRAHAM. I don't know the answer to that, Senator.

Senator MCCLURE. I think I asked the Secretary that question earlier in a slightly different form, and if I understood correctly, he was going to supply the answer for the record.

Mr. SALGADO. Yes, we will, Senator.

Senator MCCLURE. I am sorry that Senator Hollings had a 12 o'clock appointment and had to leave because I wanted to assure him that I raised the question of the class IX accident not to indicate that I think that the Savannah River reactors are operating unsafely today, but to indicate that a heavy water reactor to be constructed in the future, with a new design to new standards, is going to have to meet that kind of a test, and can't be built and operated until it does; isn't that correct? Wouldn't that be your view, Mr. Ahearne?

Mr. AHEARNE. Senator, I certainly believe that any of the new reactors have to go through a sound safety review. A class IX accident is a term of art that has been used for years in the NRC. In many places what it means by definition is beyond the basis on which the design was made. In a way there is a tautology there that says if you are considering class IX then you are already analyzing something for which the system wasn't designed. I don't want to step into that area. I would agree that one has to go through detailed safety analyses.

POWER GENERATION FROM NPR

Senator MCCLURE. Dr. Graham, the ERAB report seems to indicate that electricity sales are not of major consequence. I think I state that correctly.

Mr. Schoettler, if I state it wrong, I would invite your intervention, but that is the way I read the ERAB report.

It seems to me that that ignored projections of the need for at least 110 gigawatts of additional electrical power in the United States by the year 2000, and the growing concerns about global warming and acid rain, and the political volatility, which I hope we haven't forgotten, in the Middle East. As Science Advisor to the President, do you think the decisions on the NPR technology should, to the extent possible, factor in concerns about energy policy?

Mr. GRAHAM. Senator McClure, along with my colleagues testifying before you, I believe that the reliability of the tritium supply is the paramount issue before the country, and that should be given the principal focus. To the degree that that can be satisfactorily addressed, then I believe other concerns can also be considered, including the national energy policy and technical transfer issues.

Senator MCCLURE. Some concern has been expressed that if steam were produced and sold from an NPR, that those arrangements for sale might interfere with plant operation with respect to the production of tritium. Do you see any inherent unresolvable conflict in that regard?

Mr. GRAHAM. No, Senator, because I think the clear point would have to be made at the outset of any such sale of steam, that the production of tritium was the paramount priority, and it would take precedence over the production of electricity.