NukeInnovation

The Nuclear Innovation Workshop’s white paper was largely technical. But “Future nuclear vision (regional, national and international outlook)” is also part of its scope.

The workshop attendees were mostly engineers and scientists, so they focused on the technical aspects. But if nuclear power is going to increase its contribution to electrical generation by 2030, the issues go beyond technical. I have a lot more questions than answers on how that is going to happen. The workshop mostly didn’t discuss those issues.

First, some technical issues we didn’t discuss.

A host of technical innovations are beginning to be exploited by new companies: Molten salt reactors, liquid sodium reactors, thorium as a fuel and more. Perhaps technical innovation is not the primary innovation that is needed. We know how to build light-water reactors, and a great many safety improvements resulting from the experience of various types of failure are incorporated into their new designs. The leading small modular reactor is a light-water reactor from NuScale. DOE Undersecretary Franklin Orr said that new nuclear must start coming on line by 2030, which is fifteen years away. Energy generation by innovative designs is likely to be 20 or more years in the future. Do we even need those new designs? Where will they be preferable to light water reactors?

I’m not dissing the alternate designs, just trying to fit them into the timeline. It’s always a good idea to have backup plans. It’s likely that the American nuclear program was harmed by a too-early narrowing of options. But resources are limited, and the question is how they are to be prioritized.

The funding issues, particularly for the national laboratories, are horrendous. Researching nuclear reactor materials and processes requires time. Only two examples: materials must be irradiated in a research reactor; test loops must be built to check out flow issues and validate modeling. Materials irradiations may take years; test loops require extended runs of weeks and months. Congress’s lack of ability to pass budgets on time means that experiments are shut down prematurely and personnel may be furloughed. Although the DOE has done some things to improve continuity of funding, the national laboratories are forced to focus on budgets with accompanying anxiety through most of the year.

Another trick is for Congress to add funding for “innovative” projects to the budget one year and leave it off the next. A year is just about what is needed to start a project up. Whatever was done in those projects doesn’t even get written up. The next time, they start from scratch.

Part of the budget difficulties lies in lobbying, of both Congress and DOE. The national laboratories are not supposed to lobby Congress, but they have staff whose job it is to talk to senators, representatives, and their staffs. The labs’ staff put forth their laboratories and programs in the best light. Congressional delegations tend to favor their own states.

Private companies, of course, have no restriction on lobbying Congress. Someone obviously versed in such things was in the Albuquerque workshop session. He singlemindedly pursued the objective of a particular facility, which he presumably expected to be located at his company. He refused to discuss anything else and at times tried to physically take over the discussion from the moderator.

An idea that surfaced several times but didn’t get a lot of support was that there should be a national energy plan, and that the roles of the national laboratories, universities, and private industry should be more explicitly defined. Scientists and engineers are rational people, who plan out their research that way. It’s an attractive idea. And it will never happen, for the reasons in the last two paragraphs. Part of the rationale of having several national laboratories working on nuclear reactor issues is that they will compete and produce a better project. Another rationale is that having a lead laboratory that focuses effort will produce a better project. The latter rationale is often applied to university research centers as well. Both of those rationales can’t be right.

The competition is at least as destructive as creative. Besides the time wasted on dealing with the lack of a budget, a fair amount of time goes to figuring out how to get the money away from someone else. That would include having a center or program lead moved to your organization. A few of the suggestions illustrated this: “Fund organizations besides [organization].” Private comments reinforced this.

Kirsty Gogan talked about societal acceptance of nuclear power, which is necessary for it to expand rapidly. Societal acceptance will encourage Congress to increase funding for initiatives like the testing facility and expansion of Nuclear Regulatory Commission regulations beyond large light-water reactors that are likely to be workshop recommendations.

But public support is not that firm for nuclear. What do we do about that?

I was disappointed to see comments in this vein as uninformed as what I see on internet discussion groups. One of the criteria in the white paper had to do with decreasing weapons proliferation danger. Quite a few comments indicated ignorance of the Nuclear Nonproliferation Treaty and its relevance to civilian nuclear power, or seemed to advocate that it be ignored.

I inserted a few suggestions outside the technical. Using community approval to site plants met with extremes of approval or disapproval in our live session, and more disapproval and even disbelief in the large-group comments. Again, many seemed not to understand the process or how it has been used in Sweden, Finland, and Great Britain.

The dreaded linear no-threshold model showed up only once in the group deliberations. I had a good discussion with a participant with whom I totally disagreed. We were on the way to what could be a reasonable synthesis when break time ran out. In my opinion, this is the most useless and counterproductive argument pro-nuclear people make.

Another counterproductive wish that came up once or twice was that the NRC would promote nuclear power as well as regulate it. Eliminating that conflict of interest was, of course, the reason for splitting the AEC into NRC and ERDA. It’s hard to see how that could be reversed, or why it should.

The gender differential was huge. The Albuquerque session of about 25 people included one other woman besides me. Other sessions were in the same range.

I suspect that little will be said in the final report about any of these topics. The funding difficulties can’t be expressed openly within the DOE system because they will irritate Congress, and there’s not much point in talking about the competition among the laboratories, universities, and industry. In both these cases, everyone knows about them, and there’s not much that can be done about them.

The issues of societal acceptance barely surfaced during the workshops, and they certainly didn’t get the discussion they deserve. Maybe another day.

 

facebooktwittergoogle_plusredditpinterestlinkedinmailby feather

Leave a Reply