A recent paper by Lindsay Krall, Allison Macfarlane, and Rodney Ewing has garnered a great deal of attention:
The paper claims that small modular reactors (SMRs) will generate waste streams that are disproportionately large relative to larger reactors, and that overall, the SMR technical approach is inferior to the approach we’ve taken today. I’m sure that no one will be surprised to learn that various media outlets have seized upon this paper to paint the entire small modular nuclear reactor effort as somehow fundamentally flawed.
The typical media outlet relishes any opportunity to paint nuclear fission energy in a negative light and so has responded with glee to the claims made in the paper without any effort to challenge them and to discover if they are, in fact, true across the board. I can only contrast this with the assertive effort they undertake to challenge any negative claim made against wind or solar energy.
The published paper mentions five different molten-salt reactor designs: Seaborg, Terrestrial Energy, FUJI, ThorCon, and Moltex all make it into their paper. I found it particularly interesting that this study omits the work that Flibe Energy has done on its lithium-fluoride thorium reactor (LFTR) design. Flibe Energy is the oldest MSR company, incorporated in 2011, and the one that has a design most derivative of the work done on the Oak Ridge Molten-Salt Reactor Program from 1957 to 1976. Flibe’s LFTR uses the same basic triad of materials at the MSRP anticipated: Li2BeF4 salt, graphite moderator, and Hastelloy-N structures. The LFTR also uses an improved version of the thorium fuel cycle. And Flibe’s design is by far the most “American” of all the MSR concepts reviewed in this report.
It’s not as if the authors have not had any exposure to the LFTR concept, since I briefed a panel of the National Academy of Sciences on our LFTR design on January 13, 2021, and it included two of the authors on this PNAS study, Alison Macfarlane and Rod Ewing.
Krall, Macfarlane, and Ewing express concern that small modular reactors would have excessive neutronic leakage from their cores that would adversely affect their metallic structural materials. This is a valid concern, but it is a design-specific concern. It is difficult to make the case that this is a concern generic across all small modular reactor types. At Flibe Energy, our lithium-fluoride reactor designs employ a blanket fluid that contains a neutronically-absorptive material. In the LFTR design, this blanket fluid has a large amount of thorium in it, and this thorium productively absorbs neutrons and acts as an “internal radiation shield” for the reactor vessel. This addresses the concern about the neutronic activation of the reactor vessel that was raised in the study. In my presentation to the National Academy of Sciences, I begin addressing the concern about neutronic leakage and neutron activation of structural materials at time 3:50:47.
Here’s a neutronic image of a reactor very similar to our LFTR design in the relevant aspects. You can’t see the “edge” of the reactor (the vessel) itself because the neutrons don’t make it that far. They don’t “light” it up. So, it’s dark to them and that means the neutrons aren’t striking the metallic alloy (Hastelloy-N) that makes up the vessel and weakening it or leading to damage. Now I realize that their study focuses on Terrestrial Energy’s IMSR, which doesn’t have this attractive feature that the LFTR does, but Krall et. al. should have considered LFTR in their analysis. Indeed, co-authors Alison MacFarlane and Rodney Ewing were both part of the panel listening to my presentation. Dr. MacFarlane starts asking questions at time 4:10:35 and Dr. Ewing asks questions at time 4:14:00. Dr. Ewing says that I had the “rosiest” picture of the reactor types. I don’t know if that’s a gentle way of saying that he doesn’t believe what I’m saying, but it’s clear that Dr. MacFarlane doesn’t believe what I have to say. Whether or not they believe me, it seems rather suspect that they would author a report that says that all small modular nuclear reactors have particular problems that I had explained in my presentation are not a concern in our two-fluid molten-salt reactor design.
At one point Dr. Ewing complains that companies are keeping the details of the fuel-cycle approaches proprietary and not sharing them with a community that could review them:
We shouldn’t be the ones doing this kind of study. The vendors, those who are proposing and receiving federal support to develop advanced reactors, should be concerned about the waste and conducting research that can be reviewed in the open literature.
I can’t fathom how he could make a statement like this about Flibe Energy and our LFTR design. We have consistently released an enormous amount of detailed information about our fuel cycle approach and the detailed aspects of its design. The EPRI report on LFTR was published on October 22, 2015. That’s a long time ago, Dr. Ewing. That’s a long time ago, Dr. Macfarlane. Did you read it? Here’s a link to it in case you didn’t.
I think you’ll find that it addresses nearly all of your concerns. If not we’re wide open to additional discussion. I can attest that at no time since January 2021 did either of these authors reach out to Flibe Energy to ask questions about the LFTR design, and unfortunately, I’ve lost a lot of respect for both of them because of this. It sounds like Dr. Ewing is right, and indeed, he should not be doing this kind of study, because he is excluding important data.
This study appears to be one of those studies, all too common, where the answer was determined beforehand and the inputs were carefully chosen to support the predefined conclusion. The conspicuous absence of Flibe Energy’s LFTR design despite the presentation to the National Academy of Sciences supports this proposition.
You can say whatever you want about our competitors’ designs, and indeed, I rarely have many good things to say about them, but so long as an instance of a small modular reactor exists in the form of the LFTR that handles issues of waste and safety, you cannot accurately conclude that there is anything wrong with small modular reactors in a general sense.