Nucleating our carbon-managed future 1

Leave a Reply

Your email address will not be published. Required fields are marked *

Subscribe to Comments:


  1. Lack of agreement on effective solutions. The Renewable Portfolio Standard (RPS) that became widespread has not worked. Instead, the RPS let us take our eye off the goal of emissions reductions to focus on increasing the penetration of renewables.

    I have been saying this for years.  Measuring “renewable energy” is bogus.  The ONLY metric that matters is GHG emissions.

    Solar and wind, as intermittent energy sources, require backup generation for the majority of their nameplate capacity.

    Rod noted years ago that no less a luminary than RFK Jr. admitted that wind and solar rely on natural gas.

    Somehow, use of natural gas was back-doored, allowing gas generation to expand like a weed beneath the thin veneer of renewables, despite its huge emissions and ecologic footprint.

    Proving once again that “renewables” lock in gas consumption and all the associated emissions, including fugitive methane.

    Fortunately, legislatures in a few states are beginning to replace the RPS with the Clean Energy Standard (CES). These policies call for requiring set amounts of emissions reductions by certain dates—not prioritizing a particular technology solution.

    This should have been done in 1988, when the IPCC was founded.  Now it’s too late to avoid major consequences of that failure.

    One must look beyond the propaganda coming from both fossil and renewable competitors and seek out the data.

    In many cases, those competitors are one and the same.  Gas interests finance “renewables” to lock themselves in as the backup.

    And Valerie… you need a copy editor!

    1. Thank you . . . and so true! My apologies for such a long post! I suspect it was a breaking the (rapidly nucleating) ice thing with too many stored up thoughts that I have been meaning to write about.

      1. It’s not the length, it’s typos like this:

        within the remaining carbon budge,

        Spell-check misses those, but a good copy editor doesn’t.

  2. I think in every respect nuclear power is the best option. Whether political, economic, ethics, etc…

    There are so many advantages to an energy source that we have barely scratched the surface of yet, with little to none of the drawbacks of other sources of energy generation.

    The waste issue becomes manageable when you include waste burning fast reactors like some of the new designs.

    It’s not rationality that is keeping us from a systemic shift to a totally fossil free energy model it is the opposite. Too many people are still stuck in the mode of thinking of nuclear power as the most dangerous form of energy generation when it in fact has been one of the safest all along.

    And with new designs will probably beat that record by an order of magnitude or more.

    The way I look at it is that massive fossil fuels use is rapidly closing the doors on all our options. And nuclear power begins to open them and present us with a future that isn’t focused on “peak oil”, catastrophic climate change, truly nasty global petro-politics and such.

    Instead nuclear power allows and in facts encourages long term economic, ecological and social planning because there is no uncertainty in how we will power our societies. With waste burning fast reactors for instance the available supply of fuel when you include depleted uranium is so vast that by the time we had to start looking at alternative energy sources our technology would be so advanced the transition would be seemless.

    A real future with nuclear power or a world tumbling down into total failure with fossil fuels, I know which one I fully support.

    1. Doug, I completely agree with you and would, if I could, support that option. But lovely as that is, our current reality, is not giving us that option. The majority of voters and politicians have prioritized the deployment of renewables, so those of us who support nuclear have two options.

      One: we insist that renewables are a net negative, throw harsh facts and figures around and try to get renewables fans to admit their errors and change their minds to recognize that nuclear is the “best option.” Since that approach has failed spectacularly, has not garnered those working that approach any meaningful traction (at best they pick up a few supporters here and there and lose a few just as quickly), the pronuclear community will continue to be seen as extreme, argumentative, out of touch with reality and will remain at logger heads with those who are in the real positions of power at key environmental groups and in political office.

      Two: we work to educate folks about the true virtues of nuclear but we accept that our partners in climate action are enamored of renewables, so we choose to not criticize renewables and we pick our fights and focus on showing why nuclear is a far better complementary energy source than natural gas, thereby allowing the disappointing realities of renewables to speak for themselves.

      We desperately need a world without fossil fuel use asap and we have to build trust and connections to get nuclear to be embraced within the solution set of the accepted technologies, so I know which one I fully support.

      1. I agree with all of that, this isn’t a question of picking either low density renewables like wind and solar power over the very energy dense nuclear power, it’s about using all the options on the table right now.

        As things stand with the limits on implementing low density renewables on the scale needed and the vastly understated need for grid scale energy storage with a renewable energy model there is simply no way to carry out the essential fossil fuels phase out without nuclear power. In a very real way supporting low density renewables while refusing to even discuss the nuclear power option is taking a pro-fossil fuels stance. Because some baseload generation is needed to make up the shortfall and intermittency issues with renewables and without nuclear power that is done with gas and even coal in countries like Germany.

        I think even a 1/3 nuclear and 2/3 renewables mix would work but a 50/50 is probably the best mix. This is what needs to be effectively communicated to those who understand how vital change is in our energy model but are still stuck in thinking that nuclear power can never be the best option.

        Some reactors now being designed like molten salt fast reactors determine their energy output by how much fuel salt is pumped through the core. The higher the rate the more heat and electricity generated. This also good load following and would be an excellent fit with renewables on days with overcast or no wind for instance. These reactors are also designed to run on spent nuclear fuel and plutonium from decommissioned nuclear weapons. Another coherent selling point for those who are interested in the best options.

        The Russians have a saying, “perfect is the enemy of good enough”. We simply do not have the time or the resources to create a “perfect” alternative to fossil fuels using low density energy generation. Without the crucial contribution of nuclear energy base load generation our future becomes even bleaker than it already is. I live in the west and now fear summer for the mammoth wildfires we now get due to climate change. And that is just the start.

        1. When you say 1/3 nuclear and 2/3 renewales you have to include all the energy/electricity we need to electrify transportion or public heating/conditioning (through heat pumps, for example), including bio-fuels like methanol, DME or Fischer/Tropsch fuels, for uses not easy to direcltly go electric. And it’s a LOT energy more !

  3. Advertising. Without ads all the new cool snazy tech will stay on the drawing board and not make any money. You want to invest? Make sure that all companies you invest in commit to spend a large amount on major advertising.

    1. Especially when you consider that generations have been taught the opposite of what needs to be communicated. That includes movies like “The China Syndrome” and “Silkwood” which are essentially horror movies with nuclear power as the villain.

      There’s no question that the entire nuclear sector has a serious image problem that can only be addressed by effectively communicating the almost impossible to overstate benefits of nuclear power when compared to our current fossil fuels reliant energy model.

  4. The word ‘science’, has been ruined for me since 2020. I’m a STEM professional; I’ve dedicated my life to what science was, and now just reading this word, in its now typical context, gives me the sensation of hearing a grating/scraping sound.

    Examples of ‘science’ and context below:

    “the loss of 90% of ALL coral reefs”
    “He [the science-respecting president] has shown that he is guided by science and seeks real solutions.”

    I must accept the governance of louder, more emotional and generally lesser people because it is their world and I simply live in it. Will my job as a nuclear engineer be phased-out? Maybe. Not any time soon; there are old plants and new plants and I have an esoteric skill set. Will nuclear power expand to enable broad decarburization in the first world? Nope. It’s literally NOT happening, at this moment. Even if USA and Europe put a lot of effort into “clean” or “green”, 7/8 of the world’s population is going to take the path of least resistance to comfort and quality of life. Is it first world man’s burden to create/identify problems and then “lead by example?

  5. “If you’ve studied chemistry, you’ll know that the nucleation point describes the start of a change in physical state, such as from a solid to a liquid, or liquid to gas. Water starting to crystallize into ice nucleates where the first H2O molecules reorganize as a solid.”

    I think of it more as tritration, were drips of sanity around nuclear power have been falling into the information beaker with no real sign of change for years. Until the last drip and the indicator suddenly changes color.

    There will be a sudden paradigm shift in regards to broad acceptance of nuclear power that may seem impossible now, but will be seen as inevitable in retrospect in my opinion.

    1. I really like Doug Coombe’s observation about the effect of a
      critical drip on a supersaturated chemical solution.

      If I may add another metaphor for the moment of nucleation,
      it is the moment at which a molten-salt reactor’s supersaturated
      fuel mixture, while sitting at rest in a passively cooled dump
      tank following an earthquake-caused loss-of-power emergency,
      suddenly crystallizes into a solid mass of subcritical fuel
      salt, thus avoiding the dreaded moment of meltdown, followed
      by hydrogen gas explosions and intractable environmental
      contamination that would have occurred with a standard,
      solid-fueled, water-cooled design.

      1. Chris – interesting metaphor.

        As intriguing as molten salt reactors are, I’m still not clear about what happens after a “dump.” What steps must be taken to put the plant back into full power operation?

      2. In an MSR drain, the fuelsalt does not suddenly crystallize
        to a solid. Decay heat wil keep the salt liquid for the better
        part of a month, even in drain tank(s) that are arranged
        for efficient cooling. This is good because the transfer back
        into teh primary loop depends on the salt being liquid.
        And if it freezes, the non-eutectic mixtures does so
        over a temperature range that can be 50C or more.
        Some designs put heaters in the drain tanks, so they can
        thaw the salt if this happens.

        1. I the case of the Elysium reactor everything is placed inside a larger vessel also containing molten salt in this case pure chloride salt.

          This ties the entire system together and can be cooled by a heat pipe exhanger. In this case you’d never be solidifying your fuel salt if it drains into storage. Until the end of the primary reactor vessel after about 40 years. Even then it probably makes more sense to keep the fuel load molten until it’s moved to a new reactor vessel.

          The fuel salt load in this design is essentially permanent with a fission product cleanup at between 40-80 years. That’s another advantage of the fast spectrum reactor, the isotope capture cross-sections are much smaller meaning you can run a fast molten salt reactor with something like a 40% load of fission products.

      3. [Please note, I’m replying to Rod’s comment using a
        ‘reply’ button attached to my own comment, because I
        don’t see one attached to Rod’s comment.]

        Rod Adams says:
        May 1, 2021 at 6:34 PM
        Chris – interesting metaphor.

        As intriguing as molten salt reactors are, I’m still
        not clear about what happens after a “dump.” What steps
        must be taken to put the plant back into full power

        Good question but I’m afraid I’m not qualified to fully
        answer it. I’m a retired former software engineer from the
        programming languages arena, and everything I -think- I
        know about nuclear reactors is based on what I’ve read
        from nuclear engineering sources such as Nick Touran’s
        web site and Rod’s. My most specific trusted source on
        thorium molten-salt reactors is a 2018 spec sheet from
        Seaborg Technology describing their Compact Molten Salt
        Reactor (CMSR), on p. 208 of IAEA’s biannually updated
        report on small modular nuclear reactors.


        Regarding dump tanks, the 2018 CMSR status update says:

        “The reactor core will passively shut down if overheated
        as a result strongly negative fuel temperature coefficients.
        In case of a continued inability to cool the core, the fuel
        will eventually drain itself through an actively cooled
        freeze plug to a passively cooled dump tank.”

        Regarding the process of restoring a CMSR to full power operation
        following a core-dumping emergency, I assume you would need to
        direct the question to Seaborg Technology’s engineering staff.

        Best Regards,
        Chris Aoki

        1. Chris and all,

          I was able to change the settings to allow for six levels of embedded comments, so hopefully solving the problem Chris had with this comment.

      4. Ideally the molten salt reactor would be unaffected by the earthquake – or other natural disasters – and supply power for emergency relief.

      5. Rod and Valerie, one final comment about that metaphor.

        I’m grateful to both of you for thinking to use it, but it’s also
        from Robert Persig’s famous book, “Zen And The Art Of
        Motorcycle Maintenance”, in which Persig’s main character
        recalls a moment of nucleation or crystallization from
        school chemistry lab. It’s key to the story which is
        one of my all time favorite books, and goodness, the
        metaphor shakes me to my core.

        Little did I know at the time I read the book in 1976,
        that one of the world’s leading nuclear reactor engineers,
        Alvin M. Weinberg, must have had a Zen-like moment years
        before, in which he realized that draining the liquid fuel
        away from the thermal reactor’s neutron moderator was key
        to the safety of the reactor design.

        I sincerely hope that people all over the world soon
        experience their moment of nucleation!

        1. Thank you for sharing this recollection. Time for me to re-read “Zen and the Art of Motorcycle Maintenance,” as it has been a long time since I did.

  6. I appreciate the passion and optimism for nuclear power. I do not appreciate linking strongly fact based nuclear power with the theory of anthropogenic climate change.
    We apply for licenses based on telling the truth, facing the facts, and answering all questions in a transparent matter.
    Climate change activists seem to BELIEVE – more or less in a religious fashion a doctrine that has been beset with fraud (Hide the Decline / IPCC), and opposing views are shot down and discredited rather than debated.

    Geologists (scientists by any definition) show us that Earth alternates from Ice Age to Warming Periods at regular intervals and has done so for millions of years – without industrial man.

    It is for this reason, I implore others in the nuclear industry to not hitch the future of such a valuable energy source to *science* that avoids debate. It’s not how we roll.

    1. “Geologists (scientists by any definition) show us that Earth alternates from Ice Age to Warming Periods at regular intervals and has done so for millions of years – without industrial man.”

      Because of well understood radiative forcings as part of the Milankovitch cycles. Which are on the order of a few tenths of a watt per meter squared acting over thousands of years to first cause an ice age then warm the Earth again melting most of the ice.

      The radiative forcing from all the carbon dioxide we’ve emitted in the last 150 years has a positive radiative forcing of over 2.5 watts per meter squared. We have overwhelmed the natural cycles and are the dominant factor in the Earth’s surface heat budget.

      Which is rapidly rising threaten a number of crucial systems on which we depend on for life.

      If we can mitigate that with large scale nuclear power then it is in everyone’s interests we do so. And we can and should do so.

    2. I am convinced Global Warming & Ocean Acidification due to dumping of fossil fuel CO2 into the air are real things, but since there are people who don’t believe that, it is well worth while to point out all the advantages of nuclear besides the lack of CO2 emissions.
      Eg: the lack of soot & fly ash to damage the lungs of people who breath those in. The small amount of land occupied by nuclear plant leaving more land for other things….

      1. There are so many advantages to nuclear power over fossil fuels especially coal that climate change really doesn’t need to be brought into the picture.

        That includes the emission of radioactive materials. Carbon is a molecular sponge and coal contains relatively large amounts of radioactive isotopes. These are released when the coal is burned any many go out the smoke stacks. Coal power puts out two orders of magnitude more radioactive material than nuclear power as things stand now.

        Also products that come from coal like cinder blocks are radioactive, you wonder how many people who otherwise think nuclear power is dangerous would feel about that in a structure using these materials. Which are still well withing safe exposure limits.

    3. I too find the hype that emanates from passionate pro-nuclear proponents to be nearly as bad, if not AS shrill as the anti-nuclear disinformation, such as we saw last week among those celebrating the closure of the ‘dangerous’ IP3 (concluding ~50 years of safe operations on that parcel). Seems there is religion on both sides – and both claim to worship Gaia and ‘science’, such as AGW.

      To the MSR fans that hijacked this comment thread, the reason most of the Gen4 concepts haven’t been built (let alone detail designed) in the literally 20-years since the term ‘Gen4’ came into parlance, is that they’ve all been built/tested already through the ’80s and they weren’t easier or cheaper to operate, more reliable or less of a source of dose to workers/public, etc. IOW, they were not better than H/LWR, and more PITA.

      1. Based on facts not the kind of disinformation that underlies the anti-nuclear movement and the fossil fuels sector that gave it its seed money with FOE.

        MSRs aren’t a sport, and I’m not a fan. I understand the benefits and drawbacks and am in fact excited by something like the Elysium MCSFR that takes the elegant molten salt reactor design and improves it.

        That includes a high negative co-efficiency of reactivity, high efficiency, low waste stream, high thermal efficiency and virtually limitless fuel.

        At the fast spectrum and neutrons at over 1 MeV U-238 become weakly fissile. Which allows not just spent nuclear fuel but depleted uranium to be burned in a MCSFR.

        There are currently close to 100,000 tons of spent nuclear fuel in the US alone and about 7 times that of depleted uranium all being stored as high or low level waste.

        If you replaced ALL us power generation with MCSFR at current energy demands that would be over 2,000 years of fuel. That means no mining ever again for fuel.

        That’s before we even look at all the other advanced reactor designs.

        As you are so deprecating of nuclear power, what other options do you present that are even close to the potential of nuclear power?

      2. As for MSRs as compared to PWR/BWRs for ease of operation, safety and fuel efficiency.

        1. The primary containment is chemical not physical and they operate at near atmospheric pressure and without water. Meaning that massive primary vessel and equally massive secondary containment isn’t needed.

        2. The fuel salt is pumped through a graphite core in a slow spectrum MSR and through an empty can in a fast spectrum MSR. While the slow still uses control rods, the fast spectrum doesn’t even need those. It’s output is dependent on how fast you pump the fuel salt through the can.

        3. As far as fuel efficiency they aren’t even comparable. Most of the potential fuel with solid fueled reactors is discarded as “waste” as the core array is swapped out every few years due to the increase in neutron robbing fission products and the physical decay of the rods under intense heat and neutron bombardment.

        The core in a molten salt reactor is…. molten. It’s never removed and some fission products like Xenon 135 can be removed as the reactor is running. Try that with a PWR.

        A fast spectrum reactor with much smaller capture cross sections of the isotopes can run with a fission product load as high as 40%. The Elysium reactor will not need a fuel cleanup for between 40 and 80 YEARS for instance.

        And the actinides are never removed from the fuel salt. They remain in the fuel until transmuted into something fissile and burned up. Producing fission products with far short half lives removing the long term HLW issue.

        And on and on, it’s hard to even categorize all the advantages of MSRs over the current reactor designs. Which are all still far preferable to fossil fuels.

        It’s not a question of picking one reactor design or type over another we need them all now. And I really have to wonder why anyone would take such a divisive position.

        1. Doug, thank you for the summary of the MSR. The MSR bypasses one limitation of the solid fuel fast reactor, where (according to Till and Chang) the cladding ages from bombardment by fast neutrons. In a fast spectrum MSR, the fast neutrons transfer excess momentum to the liquid itself without accumulating damage.

          As I read it, some (not necessarily massive) containment would still be needed, at least for emergency shutdown when an MSR dumped its liquid fuel into pans of larger area. Still hot, the liquid would be releasing radioactive volatiles, with associated sputtering of radioactive liquid into the (contained) atmosphere.

          Emergency cooling would be needed too. The dumped liquid would need cooling, as the shortest-lived fission products would continue to add 6% of the previous heat power for the first hour or so, declining to 1% across the first day.

          1. You get a variety of options with the molten salt reactor design.

            In the case of the Elysium reactor the primary reactor vessel will be in a larger vessel also filled with the same molten salt minus the fissile material, fertile and fission products. The dump tanks will also be in this larger vessel thermally linking all the different components together with a heat pipe heat exchanger using sodium or cesium to cool the combine system once the reactor is shut down and the fuel load drained to sub critical tanks.

            In the case of the MSRE they built their own radiator system to cool the core load once it was dumped into the drain tanks. Something repeated multiple times for maintenance and when they switched from the uranium fuel cycles using U-235 to the thorium cycle using U-233 as the fissile fuel.

      3. Sorry for hijacking the subject in order to
        exploit molten salt/MSR as another metaphor
        for crystallization/nucleation of public awareness.
        I am a retired software engineer, and I do not
        have the audacity to insist that MSR be adopted
        at the expense of all other designs currently
        being worked on by the world’s nuclear engineers.

        My main complaint about the anti-nuclear
        movement is that it assumes, with apparent
        success, that the public does not understand
        the engineering distinction between laws of
        physics and the design/implementation of
        machines that use them and are constrained
        by them.

        The defects in safety, fuel efficiency, and
        weapons proliferation are in various nuclear
        reactor designs, and it is reasonable to want
        them to be fixed, tested, and regulated, all of
        which takes time, money, and legislation.

        A big reason it has taken so long is defunding
        and subsequent public unawareness, which
        began in the early 1970s during the Nixon
        Administration. For details, see this earlier
        Atomic Insights entry which links a Google
        Tech Talk by Kirk Sorensen:

        FWIW, I am an avid reader, and my favorite book
        about climate skepticism and other forms of
        science denialism is “Merchants of Doubt” by
        Naomi Oreskes and Erik M. Conway.

        1. The subject is nuclear power and coalescing a initiative to implement it on the scale needed to deal with some very critical issues right now.

          Discussing something as potentially groundbreaking as molten salt reactors isn’t hijacking anything in my opinion.

          While some may attempt to claim that molten salt reactors are some fringe movement by crackpots the reality is far different.

          The scientist who came up with the concept of the MSR is the very same Eugene Wigner who did the groundbreaking work on semi-conductors that allowed the development of transistors and our modern information based world.

          The scientist who took the idea and turned into reality was his protege Alvin Weinberg, the man who designed the first nuclear power reactors.

          The reason both he and Wigner conceived and built MSRs was because of the distinct advantages they saw in them compared to other designs that Weinberg himself had first created.

          When you consider where people like Ed Pheil have taken the design and a fast reactor in development that will use waste and weapons grade plutonium as fuel with almost none of the drawbacks of other reactor designs, what I have a hard time understanding is the resistance to this kind of design.

          If it’s organizational inertia, does that have any real relevance in our current situation.

          1. Thanks Doug for the historical clarifications on the roles
            of Wigner & Weinberg. I studied the writeup of the Elysium
            MCSFR in the IAEA SMR catalog:


            which further explains some details of the Elysium MCSFR,
            but I’m still unclear about how it handles long-lived fission
            by-products. Is the idea that an MCSFR breeds and burns
            all fertile isotopes (recursively?) leaving only stable by-products
            as waste? If not, then what is it?

            1. They won’t be taking ANY actinides out of the fuel salt except in the long term cleanup which will be between 40-80 of reactor operation. In that case it will be something like 1% of the actinides coming out in the cleanup a very small amount to deal with.

              All the fertile material will eventually be bred into something fertile and burned up.

              As far as the fission products, they will leave most in the reactor for decades the fast spectrum has much smaller capture cross sections allowing this. The cesium and strontium FPs have similar chemistry to the actinides and will be a chemical proliferation protection for the fissile uranium and plutonium in the fuel load.

              Some of the fission products that will come of of solution as the reactor runs like the noble gases Xenon-135 and Krypton-85 will be sold as they have commercial value.

              Of the rest from what I understand most of the fission products will be at ground state in a decade, it’s something like over 85%. Of the remainder the longest lived stuff needs to be stored securely for about 300 years. Still a challenge but the volume is still much smaller than the large amounts of SNF now being stored all over the world over a much shorter time scale than the thousands of years needed to store the spent fuel with large amounts of radioactive actinides..

    4. All the articles on agree with Rob Brixey’s advice that nuclear power professionals and advocates promote nuclear power on its own merits and not on claims that nuclear power can influence global warming and climate change. Atmospheric carbon dioxide plays only a minor role in determining climate change.

      To abandon fossil fuels and their by-products is a existential threat to the modern world. What kind of people want that?

      1. John:

        One of the merits of nuclear power – among many others – is that the power source is clean enough to run inside submarines.

        It not only eliminates CO2 production wherever it replaces fossil fuels, it also eliminates air and water pollution caused by all of the other gaseous and particulate emissions caused by burning fossil fuels.

        As a strong advocate for abundant energy, I do not advocate abandoning fossil fuels. I advocate replacing them wherever possible with cleaner, more abundant fission fuels. In some applications, hydrocarbons are the best available option because fission does not work on the scale of personal transportation, small boats or lawnmowers.

        1. Rod:

          We agree that nuclear power should be expanded as fast as possible to support a peaceful, prosperous world.

          We have witnessed the lack of progress in the USA and Europe since the 1980s. Yes there is a lot of talk about good new designs. That is great.

          But the problems that have held nuclear power back for fifty years have to be dealt with head on and quickly. Reform in political leadership to support nuclear power. Reform in regulatory licensing. Reform in radiation safety guidelines.

          Carbon dioxide is not a pollutant. It does not cause catastrophic man-made global warming. That is fantasy of alarmists who are determined to deny fossil fuels to the world. The true story of carbon dioxide is explained in over 900 articles on in the section, ENVIRONMENT. It is explained at,,,,, and many other sites.

          Carbon dioxide is the key molecule of all life on Earth. We need CO2 in the atmosphere.

          I want nuclear power to expand as fast as possible just like you. But, I want it so that fossil fuels are saved for future use that nuclear power can not do yet.

          What is that? Operating billions of automobiles for personal freedom and mobility, powering ships for global commerce, operating planes for long distance travel, agriculture, food processing, cooking, manufacturing, mining, space heating, etc..

          For the long term future, nuclear power will provide all the energy of the world including production of synthetic liquid fuels to replace depleted fossil fuels.

          You and I have witnessed near zero progress for nuclear power for forty years in the United States and Europe. China and Russia are making great progress. India will also.

          It is up to us (who else?) to clear the problems in politics, government bureaucracies, and alarmists blockades that led to this disaster.

          Spending time claiming nuclear power is good because it minimizes carbon dioxide production is not good science. It distracts from correcting the real blockades to nuclear power.

          Best wishes.

          1. John:

            Have you ever considered the possibility that fossil fuel interests have used their still impressive political clout to slow nuclear energy development? There are many bits of evidence pointing in that direction. It’s also important to understand that people whose wealth depends on continued fossil fuel dominance – with accompanying periods of high, profitable pricing – are highly motivated to slow the deployment of an abundant, less constrained source of energy & focused power.

            You and others ignore the fact that SOME of the evidence claiming that CO2 emissions increases and an ever rising concentration of CO2 are not problems to worry about comes from people that want to keep merrily selling natural gas, oil and coal.

            Many tell me that CO2 is plant food and necessary for existence. I agree to a point, but can’t agree that it’s a good idea to keep increasing the concentration of atmospheric CO2. There are many chemicals that are useful in moderation, but become ever less useful as concentration increases. Fertilizer, for example, feeds plants. But farmers who seek ever increasing yields can quickly find that too much fertilizer can damage their crops (as well as cost a lot of extra money.)

            I’ve spent close to 30 years engaged in energy policy discussions with people who claim that natural gas is cheap and clean enough – even though “cheap” natural gas at $2.00/MMBTU costs 3x as much as commercial nuclear fuel and even though the best available natural gas technology produces about 20x as much as routine nuclear power.

            For the record, I recognize the need to use hydrocarbons – mined now and synthetic later – to power many of the important sectors you mention. But I have some disagreements.
            Ocean ship propulsion is well suited for direct fission power plants. It’s an application that has close to 70 years worth of operational experience.
            A significant fraction of space heating needs can also be met using nuclear plant “waste” heat, especially when the nuclear reactors are small enough to be placed in industrial areas of cities, college campuses and industrial parks that have district heating systems.
            Manufacturing electricity and process heat are also applications well suited for nuclear energy.

  7. My personal dream is to see a PBS NOVA sequel to the
    2017 episode “The Nuclear Option”, which explains to
    the viewing public what a “thermal breeder” is, and
    what it can do to provide walk-away safety, eliminate
    the need to store highly radioactive nuclear waste for
    100Ks of years in 100% impermeable underground
    caverns, and provide enough safe, reliable, load-following
    nuclear energy to ditch fossil fuels before 2050.

    “Coming up next: Closing The Nuclear Fuel Cycle.
    Right now, on NOVA!”

    1. That’s another advantage of fast molten salt reactors, they are suited to load following.

      Reactor vessel geometry creates the barely critical zone in the reactor interior when the reactor is filled with a specific volume of molten salt with a specific concentration of fissile fuel. When pumping is active the reactor is always barely critical. Power output is determined by how much fuel salt is pumped through the core per unit time. To follow the grid load with a fast MSR reactor you vary the reactor pumping flow rate from the core to the heat exchangers. Which is then converted to electrical power in your generator loops.

      If you’re building a fast spectrum reactor to burn SNF it also makes more sense to make them as large as possible. A 1,200 Mwe MCSFR would burn about 4 tons of SNF fuel a year. You’d want to build lots of these to deal with the almost 100,000 of SNF in the US alone.

      Plus you can also keep building other solid fueled reactors in the meantime as their SNF when it has cooled can also be considered stored nuclear fuel for the arriving fleet of MCSFR.

      To clarify my comment about actinides in the Elysium reactor, I meant they are left in the fuel salt and subjected to neutron flux that eventually breeds them into fissile material that then undergoes fission.

      All forms of nuclear power that in fact do work together like solid fueled reactors providing fuel for fast burner reactors are an elegant solution to our energy needs that extends centuries at least into the future.

      When you also factor in how developments in harvesting uranium from sea water are highly likely in the coming centuries the fuel supply for nuclear power becomes almost unlimited. That’s a possible 4.5 billion tons reserve of fuel to be tapped.

      1. That’s another advantage of fast molten salt reactors, they are suited to load following.

        Isn’t that a property of molten-fuel reactors in general?  A strong negative temperature coefficient makes them essentially self-regulating?

        IMO, load-following is a mistake.  The reactor costs about the same if it’s running flat-out or just idling.  It would be far better to find alternative loads to make productive use of power that would otherwise not be generated.  Plasma gasification of garbage to make syngas and clean fill is one such alternative load.  Gasification of biomass such as crop wastes is another.  I’ll have to dig through my spreadsheet again looking for errors, but my first cut on the numbers suggested that the USA could gasify 1 billion tons of biomass with a little under 200 GW of average thermal power.  You’d need a temperature of about 1000°C to run the reaction to completion, so you may need to use electricity rather than direct nuclear heat.  Besides, you want to distribute the gasification to minimize shipping of bulk biomass.  Once converted to methanol, CO2 and perhaps slurried char, you can move things by your choice of tankers and pipelines.

        1. Yes, molten salt reactors have a high negative coefficient of reactivity, when the fuel salt heats up it expands forcing some out of the core and increasing the distance between the fissile material in the fuel. This slows the fission reaction.

          Thermal depolymerization is probably a more efficient way to convert organic waste into usable products. It’s almost self powering and the end product is a light oil similar to Texas sweet crude.

          You could utilize nuclear power to produce synthetic gas and diesel directly from air using it’s electricity, water and atmospheric carbon dioxide.

          The Elysium MCSFR is planned to reach about 1,300 C with later models using high temperature alloy reactor vessels, you can almost process concrete at that temperature.

          Also many areas are at critical levels for water, with their high operating temperatures molten salt reactors are well suited to desalinate large amounts of sea water.

          I think the real challenge in the future is going to be finding things that nuclear power and it’s products can’t do. For instance every MSR would also be a medical radionuclide producer as well. There would be no shortage of Technetium 99m, Iodine-131 or Bismuth -213 again.

          Or how about advanced propulsion systems for the space programs, they’re already looking at nuclear power to get to Mars in a meaningful time scale and Kirk Sorensen was researching power options for NASA when he stumbled on MSRs that had largely been forgotten by the sector. A molten salt reactor doesn’t need to be cooled by water, you could run one on the Moon to power any facilities there for instance.

          High efficiency ion rockets used in deep space missions use Xenon as fuel. Molten salt reactors remove the Xenon-135 fission product as they run which could provide this fuel. You can also configure an MSR to produce all the Pu-238 you’d ever need for thermal nuclear power generator systems which are often used to power deep space missions.

          The list goes on and on, if we have the intelligence to ever go nuclear power as our energy model base, I think the term nuclear revolution will be a massive understatement.

          1. Yes, molten salt reactors have a high negative coefficient of reactivity, when the fuel salt heats up it expands forcing some out of the core and increasing the distance between the fissile material in the fuel. This slows the fission reaction.

            And even without a core as such; you say Elysium doesn’t have one, just a container.  I’ve read that critical mass scales as the inverse cube of density.

            I’ve done a bit of investigation into the prospects of liquid actinide fuel in SiC containers (rods) and lead coolant.  It would eliminate issues of fuel fabrication and allow gaseous fission products to escape the fuel as such and collect in the headspace of the containers.  Both uranium and plutonium form carbides and silicides, but they have low heats of formation so the container material would tend to remain as SiC.  Alumina (sapphire) may be another candidate.

            Thermal depolymerization is probably a more efficient way to convert organic waste into usable products. It’s almost self powering and the end product is a light oil similar to Texas sweet crude.

            This might work with animal products (especially fats), but lignocellulose first torrefies by dehydration and then chars.  The simplest route to liquid fuel that I’ve found starts with gasification followed by conversion to methanol.  Two steps, turnkey chemistry.

            You could utilize nuclear power to produce synthetic gas and diesel directly from air using it’s electricity, water and atmospheric carbon dioxide.

            If you start with lignocellulose, the carbon capture and a large fraction of the energy generation is already done for you.  It’s the cheapest route.

            Dry lignocellulose is about 45% carbon by mass (wood higher, grasses lower IIUC).  If we go by the DOE’s estimates and assume 1 billion dry tons per year, that’s enough to make 1.2 billion tons of methanol.  That’s 25.8 quads worth, within a stone’s throw of net US petroleum consumption.  Electrify a few things that currently use petroleum and that’ll do it.

        1. They ran into some challenges with the slow spectrum thorium cycle MSRs such as fuel processing being a lot more complex than expected.

          Plus flibe is an exotic, expensive and toxic salt and the highly enriched Li-6 is a proliferation issue. And the competition with battery makers who also use large amounts of lithium in their products.

          Elysium went with the fast spectrum and a chloride salt to avoid this. Fuel processing is basically chopping up SNF fuel rods and dropping them into the molten salt. Plus no need for a graphite core that needs to be replaced every few years. Plus because it’s a fast reactor it can use depleted uranium as part of its fuel input of which there are huge stockpiles now being stored as low level waste.

          I think if it ever gets traction the MCSFR is going to take off.

          And as I commented above, because it does treat SNF from the current fleet of sold fueled reactors and their follow ons as fuel, a MCSFR is a solution to the issue of long term radioactive actinide waste like Pu-239. A MCSFR burns any Pu-239 you put into it plus all the other TRUs in SNF and almost all its waste stream is relatively low half life fission products. It’s a natural complement to next generation PWRs and BWRs. No need for a Yucca Mountain style SNF long term storage site when your growing fleet of MCSFR is burning up the SNF at 4 tons a year per one 1,200 Mwe plant.

          1. A MCSFR burns any Pu-239 you put into it plus all the other TRUs in SNF and almost all its waste stream is relatively low half life fission products. It’s a natural complement to next generation PWRs and BWRs.

            Sounds good so far, but….

            No need for a Yucca Mountain style SNF long term storage site when your growing fleet of MCSFR is burning up the SNF at 4 tons a year per one 1,200 Mwe plant.

            Trying to crunch this…

            10% burnup is roughly 100 GWd/t, so 100% burnup is about 1 TWd/t.  4 t/yr is ~4000 GWd/yr, ~11 GW(t).  That’s WAY more than 1200 MW(e) using any reasonable heat engine.  For a molten-salt reactor you should be getting at least 45% thermal efficiency (ultrasupercritical steam) so make that closer to 5 GW(e).  Scale that down to 1 GW(e) and you’re talking 0.8 t/yr consumption.

            Current LWRs are achieving something like 40 GWd/tHM, so we’re producing way more SNF than we have the fissiles to burn in fast-spectrum reactors.  It’s the fissile loading that’s critical with fast-spectrum reactors; they require a lot more per unit heat production than thermal reactors.  Net breeding comes at a price.

            The limiting factor is the availability of fissiles.  Ironically, thorium MSRs can breed up faster than anything starting from natural uranium, but we have way more uranium refined and on hand.  It poses a dilemma, it does.

            I’m just an educated layman, not a pro.  I can deliver reality checks, but the resolution belongs to the specialists.

            1. My source on the 4 tons of SNF burned a year in a 1,200 Mwe MCSFR is Ed Pheil who has decades experience in the field, I’ll see if I can source his data.

              As I’ve already commented, at the fast spectrum the U-238 which makes up most of SNF and almost all of depleted uranium is a weak fissile at the fast neutron spectrum. That along with the much higher amount of neutrons emitted per fission at the fast spectrum creates the neutron economy to get you over the threshold with the kick from Pu-239 or other fissiles.

              As for breeding fissiles with thermal spectrum MSRs, you still have to deal with the issues with that reactor design which types like the Elysium are being created to overcome.

              That includes complexity, cost, toxicity and proliferation issues with the salt itself. the lithium in flibe needs a high purity of Li-6 which makes it weapons grade for thermonuclear weapons. Not to mention the tritium release from lithium in the salt as the reactor runs.

              The MCSFR uses table salt blended with some potassium chloride to modify the melting point in an empty reactor can. No graphite moderator, or other internal components. The differences between the breeder and burner MCSFR are straight forward. The burner uses all the walls as reflectors and the breeder the top and bottom as reflectors and the breeder blanket around the vessel circumference.

              The challenge now is designing the correct vessel geometry.

          2. Elysium went with the fast spectrum and a chloride salt to avoid this. Fuel processing is basically chopping up SNF fuel rods and dropping them into the molten salt.

            Color me highly skeptical of this.  SNF is generally less than 2% fissiles.  Even a fairly large fast-breeder core like PRISM requires over 16% fissiles to function.

            Maybe you can continue to operate a MCSFR by adding SNF to it, but you won’t start one that way.  It’s going to take an enrichment or reprocessing operation to make the starting fuel charge.

            1. It will be a blended fuel mix, the Elysium MCSFR is designed to run best on Pu-239 and SNF, but you can run any fuel cycle on it including thorium.

              As for fissiles for startup in the long term, have a small percentage of your MCSFR fleet as breeders and produce all the fissiles you need that way.

              Also at neutron energy over 1 MeV U-238 become weekly fissile with 10% of that undergoing fission. So SNF and even depleted uranium becomes a lower grade fuel source for an MCSFR. You just need to blend up with whatever amount of Pu-239 or U-233/U-235 depending on your cycle to get above the threshold.

            2. The Elysium reactor is being designed to work best with SNF AND Pu-239. You do need enough fissiles to bring the reactor to barely critical when the reactor vessel is filled with fuel salt.

              At the fast spectrum SNF is above 2% fissiles as 10% of the 96% U-238 in SNF will undergo direct fission and the rest will be breed into fissiles as the reactor runs. This is in addition to all the fissile TRUs in the SNF and whatever Pu-239, U-233 or U-235 you decide to blend into the fuel to bring it to barely critical status in the reactor to start it.

              Any reactor needs enough fissiles to reach criticality in their specific design, MCSFR are no different than this. When reactor chemistry is mature in a MCSFR you’re breeding all the new fissiles needed plus the additional fissiles added when you add more SNF or depleted uranium to keep the reactor running.

              U-238 is a partially fissile fuel at the fast neutron spectrum of over 1 MeV.

            3. At the fast spectrum SNF is above 2% fissiles as 10% of the 96% U-238 in SNF will undergo direct fission and the rest will be breed into fissiles as the reactor runs. This is in addition to all the fissile TRUs in the SNF and whatever Pu-239, U-233 or U-235 you decide to blend into the fuel to bring it to barely critical status in the reactor to start it.

              Yes, but you’re talking a LOT of extra fissile Pu to reach the fissile percentage specified for e.g. PRISM with the balance supplied by SNF.

              This leaves major questions:

              1.  What IS the fraction of non-SNF fissile required?

              2.  Where do you get it?

              3.  What are the limits of this approach, WRT e.g. existing inventory and annual rate of increase?

              For thorium thermal breeders, several of these questions DO have answers:

              1.  Non-SNF fissile required is zero; it’s suppllied by e.g. HALEU.

              2.  Enrichment of NU.

              3.  Not much in the medium-term.  At 1 t/yr/GW(e) consumption, 100 kg fissile/GW(e) inventory and 1.03 breeding ratio, each reactor in operation could fuel a new reactor in about 3.3 years.  Assuming that reactors are started as quickly as the fissiles to fuel them are available, the doubling time could be as little as 2.3 years.  This means octupling in 7 years and a 16-fold increase in just over 9 years.

              Compared to a radical up-scaling of uranium mining and enrichment, this looks like the way to go.  On the other hand, the USA has no experience with such reactors at the GW scale.  It’s always six of one, half-dozen of the other.

            4. Obviously, you have to take out the uranium (mostly, U-238 and some 235) from the Pu+MA+FP waste stream

          3. In a steady state, it’s more like (a bit less of) one ton per GWyear, whether it’s depleted uranium, LEU or natural uranium or LWR spent fuel. I don’t know where the 4 tons/year figure comes from, maybe you’ re confusing with the start-up fissile (but that is no less than 8-10 tonn per GWe) ?

      2. “would burn about 4 tons of SNF fuel a year”

        One tonne per year per gigawatt would be more accurate. Taking 200 MeV per 235 amu gives 82 PJ/t, or 2.6 GW-a/t. That’s 2.6 GW-annum thermal per tonne of fission products. When converted at 38.5 % efficiency, the generation of 1 GW would produce a waste stream of just 1 tonne of fission products per year.

  8. @Rod Adams

    “As intriguing as molten salt reactors are, I’m still not clear about what happens after a “dump.” What steps must be taken to put the plant back into full power operation?”

    The core load flows into the sub critical drain tanks, cools and solidifies. When the reactor is ready for operation the tank is heated to above the salt melting point and the fuel salt in pumped back into the reactor vessel and the circulation pumps started up restarting the reactor.

    In the case of the MSRE at ORNL in the 1960s this process was repeated multiple times as maintenance was done.

    That reactor used freeze plugs of the same salt in the drain pipes cooled by fans. Turn off the cooling fans and plugs melt and the core drains.

    Something like the Elysium MCSFR won’t even use freeze plugs, it requires constant active pumping from the drain tanks to keep the reactor vessel at the fill line. If the pumps lose power or there is a high or low temperature trip the core salt automatically drains into sub critical drain tanks.

    That are also thermally connected to the ground around them carrying off the decay heat as the salt cools.

  9. Good Article:

    I look at the following link.

    It shows renewables generated 20 percent of the electricity as nuclear in terms of MegaWatt hours. At first I think, they are catching up with nuclear. Then I think some more. Nuclear is available 24 hours a day and 365 days a year. Renewables are available when there is enough water in the hydro pond with the fish protected, when the sun shines and the wind blows. Renewables are subject to the whims of nature to make them available.

    I see the news. General Motors is investing a billion dollars in a plant in Mexico to make electric cars. They are also investing in electric car plants in the US. The Chevy plant in Lordstown, Ohio closed in 2019. It was followed by the investment in a large battery plant for electric vehicles. General Motors is typical of all the car companies. They are all investing in electric cars.

    I envision that as people will be transitioning to electric cars, they will be using them daily for work transit. They will need to be charged nightly. The sun doesn’t shine at night and the wind doesn’t always blow. If we want greenhouse gas free energy, these renewables won’t be enough.

    Necessity is the mother of invention.

    Nuclear plants will quietly be built again. Perhaps they will be part of new renewable energy complexes. There will be acres of windmills and the generation IV nuclear plants will be hidden underground generating the bulk of the necessary energy.

    It looks like this article was written by one of the money people. My old man said, “The Beancounters Run The World.” Now that I see the Beancounters jumping in, I see new nuclear will be a reality.

      1. Looks like it’s based on one of the designs for the molten salt reactor powered bomber from the 1950s. I don’t know much about that design, but it sounds interesting.

        There will be two versions of this reactor, one is a fast spectrum waste burner and the other a slow spectrum using a graphite moderator.

  10. @Engineer-Poet

    “Yes, but you’re talking a LOT of extra fissile Pu to reach the fissile percentage specified for e.g. PRISM with the balance supplied by SNF.”

    The Elysium MCSFR is far different than the PRISM style fast reactors, it’s a liquid fuel circulated through an empty can. The wall geometry creates the necessary zone in the reactor for it to reach barely critical.

    I don’t know the specific details because they haven’t been nailed down yet.

    Ed Pheil and his team at Elysium have about a 300 years combined experience designing, running and maintaining multiple reactor types, this is one to keep your eye on is all I’m saying.

    1. Basically, the start-up fissile is going to be almost the same for any fast reactor, about 8-12 ton of WG or RG plutonium equvalent per GWe.
      As far I understand I don’t think that depleted uranium or U-238 is anyhow useful for this scope. But at last it’s not even that important, you can separate (with all the fission products) uranium from transuranics in the LWR (or Candus) waste stream for the fast reactor start-up and use that slightly enriced uranium (or depleted uranium from enrichment tails) for conversion in a steady state.

      Elysium team explains it well in their slides,
      as “Mode I and II” (UNF Method I – UNF w/ Plutonium addition and UNF Method II – Uranium Depletion (No plutonium added))

    2. Basically, the start-up fissile is going to be almost the same for any fast reactor (liquid or solid fuel based) at about 8-12 tons per GWe of WG or RG plutonium equivalent and for that scope I think uranium depleted or 238 is basically useless.

      But at last this is not that important, you have a lot of transuranics from LWR (or Candus) waste stream, simply you have to separate it with all the fission products from slightly enriched uranium (about 1%) in a proliferation resistant way (like Elysium’ s)

      The Elysium team well explains it in their slides
      when they say ” Mode I and Mode II “:

      UNF Method I – UNF w/ Plutonium addition
      UNF Method II – Uranium Depletion (No plutonium added)

    3. Doug,
      basically, the fissile start-up is almost the same for any fast reactor (either solid or liquid fuel), at about 8-12 ton per GWe as WG or RG plutonium equivalent; for that scope I think U-238 (or depleted uranium) is useless, even if it’s slightly fissile in a fast spectrum.

      But at last it’s not so important, we have plenty of transuranics+plutonium from LWR (or Candus) waste stream, once you separate it together with fission products, in a proliferation resistant way, from slightly enriched uranium (about 1%).

      The Elysium team well explains that in their slides with “Mode I and II”

      UNF Method I – UNF w/ Plutonium addition
      UNF Method II – Uranium Depletion (No plutonium added)

      1. @Alex P

        Thank you for the link to the Elysium document by way of explanation of how they concentrate used LWR fuel to the fissile density required for start-up. However, it doesn’t actually explain how they achieve the technically difficult step of separating the fissiles from the excess of U238 and fission products. Instead it is just a box on a flowchart. Linking us to a document describing such a major technical achievement might reassure us that these guys are serious.

  11. @Engineer-Poet

    “Yes, but you’re talking a LOT of extra fissile Pu to reach the fissile percentage specified for e.g. PRISM with the balance supplied by SNF.”

    This leaves major questions:

    1. What IS the fraction of non-SNF fissile required?

    2. Where do you get it?

    3. What are the limits of this approach, WRT e.g. existing inventory and annual rate of increase?”

    This isn’t a PRISM fast reactor that uses solid metallic fuel, it’s a molten salt reactor.

    I’m not sure what what the fissile load will be, the design isn’t finished yet or they haven’t published. But as I’ve explained the fissile load in the SNF is not the 2% you have stated, it is over 10% because 10% of SNF, U-238 IS fissile at that spectrum. Making SNF low enriched fuel at that spectrum.

    The last figure I have from the mid 2000s was there was enough Pu-239 stockpiled globally to produce 300,000 nuclear weapons, that would be an excellent place to start for fissiles needed for MCSFR start ups.

    What’s the limits of this approach?

    Virtually limitless in terms of fuel supply. If you build a fraction of your MCSFR as breeders then there is a constant supply of all the fissiles you need.

    The available fuel supply is massive. In the US alone close to 100,000 tons of SNF are being stored, and about seven times that of depleted uranium. And as I keep explaining, at the fast spectrum U-238 isn’t just fertile eating up neutrons in the reactor, it is also fissile meaning every 10% of U-238 atoms undergoing fission is adding 7-8 neutrons to your neutron budget.

    So we’d be looking at centuries of fuel for MCSFR and if by that time we’d need another source the technology to harvest uranium from sea water would almost certainly be mature. That’s a potential uranium reserve of 4.5 billion tons

    Not to mention a bunch of other energy technologies that will almost certainly be developed in that time frame.

    1. (part 1/2)

      This isn’t a PRISM fast reactor that uses solid metallic fuel, it’s a molten salt reactor.

      That much was obvious from the outset.

      I’m not sure what what the fissile load will be, the design isn’t finished yet

      So it’s not even a paper reactor yet; how can you make firm claims for it?  Where’s the knowledge base for such projections?

      the fissile load in the SNF is not the 2% you have stated, it is over 10% because 10% of SNF, U-238 IS fissile at that spectrum.

      What makes Elysium so special, when LMFBRs require so much more fissile?  PRISM requires 16.39% fissile PU in the core to be loaded. and it has the same fast spectrum that Elysium has (maybe even harder).  Fermi I used 25.6% enriched uranium!  The same fissile nature of U-238 should be operative in all, no?

      PRISM only achieves 69.91 kg annual fissile gain out of 2458.8 kg fissile Pu loaded.  That’s a gain of only 2.84% per year, a glacial rate of expansion compared to how fast our deadlines are looming.  If Elysium is going to somehow do a lot better, the question stands:  HOW?

      (BTW, I note that the Dubberly paper appears to assume 1 GW(t) output for a PRISM, whereas the power claimed for PRISM now is about 840 MW(t).  This will affect both fissile consumption and production negatively.)

      the last figure I have from the mid 2000s was there was enough Pu-239 stockpiled globally to produce 300,000 nuclear weapons, that would be an excellent place to start for fissiles needed for MCSFR start ups.

      All well and good, but this still begs the question:  how MUCH is that, and how far can you stretch it?

      1. It’s not a LMFBR reactor either, it’s a fast molten salt reactor.

        And it doesn’t need high fissile loading to work, it just needs enough to be barely critical when the reactor vessel is filled with a molten fuel salt load.

        What makes the Elysium reactor so special.

        1. Simplicity. It’s an empty stainless steel can with top mounted pumps for ease of access and replacement. One reactor vessel can be used for power output as low as 50 Mw or as high as 1,200 Mw, depending on the number of heat exchangers and fuel pumping capacity. The power output is a factor of how quickly the fuel salt is pumped through the reactor vessel.

        2. Passive safety on multiple levels including using a non toxic salt for the medium. It’s table salt, not the toxic lithium, fluoride, beryllium mix in LFTRs.

        3. Efficiency. Almost all the fuel loaded into the reactor is burned up leaving virtually no long half life actinide wastes. Almost all the waste coming out of the reactor is fission products and most of that will be at ground state in a decade.

        4. Cost. It’s designed to be on par with current coal and gas generation plants.

        It’s not just on paper, Elysium is bringing it to market as quickly as they can. The fuel is already undergoing certification as it’s not really fuel processing. It’s simply chopping up SNF and melting it in the salt.

        Once they get the precise reactor geometry nailed down, then it will be prototyping then production. It’s going to be here in a matter of a few years if they get the funding, that’s the whole point of the project.

        If you want to go into the details here’s some data.

        1. Doug,

          Your newer document does not contradict this earlier slideshow which states:

          • High Assay LEU 10-20% enriched
          High fuel load due to fast reactor, open core, etc.
          • Feed-In starts as mostly 20% enriched

          Page 10 states that it would require more than 10 tons of reactor-grade Pu to start.

          Fast-spectrum reactors ALL require high fissile loadings because fission cross-sections are smaller at high neutron energies.  You need a higher fissile concentration to compensate for this, and Elysium is no exception.  Yes, it appears that you CAN start quite a few Elysium units on reclaimed Pu from SNF, but you have to concentrate it rather than using straight SNF.  There’s nothing in the document about having to use an enriched feed when starting on reclaimed Pu, so it does appear that they expect to be able to use nothing more than reclaimed U or SNF when starting on a Pu (rather than enriched U) charge.  I’d want clarification on this, however, and the slideshows do not go anywhere near the required level of detail.

          1. Do you know why theycould increase the reactor T_out (from the first doc to the latter) from 600 °C to 750 °C ?

            1. To be frank, the first document only states that T(out) would be greater than 600°C.

              And of course I don’t know why, because the details of materials &c that would allow that are not given in anything I’ve read so far.

        2. They really should change the name “Elysium” to something that sounds less sinister.

          “Elysium Industries” is the kind of fancy, elitist name that Hollywood always gives to evil mega-corporations that claim to be visionary and utopian but are secretly plotting to turn as all into zombie cyborgs or something.

          Hollywood actually made a sci-fi movie titled “Elysium” in which Jodie Foster and her rich, elitist friends live on a luxury spaceship called Elysium and spend their time oppressing all the poor saps down below on a dystopian Earth. Is that the kind of psychological association you want for your reactor?

          The company is headquartered near Schenectady, so how about calling itself the Schenectady Power Company, or something else that’s nice and ordinary and down-to-earth. Nobody is afraid of Schenectady.

          And, duh, since there’s table salt in the reactor, call it the Table-Salt Reactor.

    2. (part 2/2)

      What’s the limits of this approach?

      Virtually limitless in terms of fuel supply. If you build a fraction of your MCSFR as breeders then there is a constant supply of all the fissiles you need.

      You must be new here, because that hasn’t been news to me for over a decade.  The problem is how fast you can get this thing rolling at the required scale.  1 TW(e) of PRISMs using the reference core makeup and assuming 311 MW(e) output requires 3215 units fueled with roughly 8.2 THOUSAND metric tons of fissile Pu, plus whatever’s not in reactors at the moment due to cooling off or being reprocessed.

      The information I have available puts the US plutonium inventory at 95.4 metric tons.  That’s a bit over 1/9 as much as we’d need.  Can MCSFR somehow stretch that by an order of magnitude?

      This is where the Th-U fuel cycle shines.  It can achieve a substantial positive breeding ratio in a thermal spectrum, which minimizes fissile inventories.  At 100 kg fissile per GW(t), our 95,400 kg of Pu could fuel 954 GW(t) of reactors producing perhaps 430 GW(e).  If a 1.03 breeding ratio can be achieved, we can double the reactor fleet in less than three years.  THAT will get us where we need to go before our deadline for action hits.  It’s not that I don’t like fast-spectrum reactors (we definitely need enough to burn our TRU instead of disposing of it as waste, and leaving all that uranium just lying around offends my sense of thrift), it’s that we probably have less than 15 years to deal with our climate problem, and 2.84%/yr expansion rate just won’t cut it.

      1. Engineer-Poet and Doug Coombes,

        do you know whether Elysium team wants to enrich chlorine and, if so, at which level of enrichment ?

      2. As far I understand it, those 95.4 tons are WG plutonium only, not all the plutonium we have from current LWR fleet. Assuming ~ 350 kg of Pu+MA per GWyear and a production of US nuclear electricity of 800 TWh/year, we have already 30 tons/year of RG Pu, or about 2000 tons in the last 60 years. It’s enough to start 200 GWe of fast reactors (at about 10 tons per GWe of fissile start-up)

      3. The Elysium reactor is being designed to be as simple and easy to certify as possible. The first fleet will use materials already certified and the fuel processing system is already well on the way to certification. As it is basically chopping up SNF rods and dropping them in molten salt.

        The salt itself is mostly sodium chloride – table salt – which is available in massive amounts and is inexpensive. An entire fuel load would be less than $500 from a wholesaler.

        The fuel cost is in the black, the reactor burns waste that current nuclear companies now pay millions of dollar a year to store as waste.

        There’s no issue with a supply of fissiles as I keep pointing out, designing a breeder MCSFR is a straightforward process of having the top and bottom of your reactor as neutron reflectors and the outside circumference of the reactor as a breeder blanket.

        The reactor itself isn’t pressurized meaning it doesn’t need a massive pressure vessel that takes half a year to fabricate in plants that are now rare globally. And is very expensive. The entire all up cost of a MCSFR is on par with current coal and gas fired power plants. It is being designed that way. With the follow on Elysium reactors using hastelloy or other high temperature alloys and running at 1,300 C you could directly replace the coal and gas burner assemblies in current plants with a MCSFR.

        The reactor doesn’t run predominantly on Pu-239, 95 tons would be able to start multiple reactors. As long as some are breeders then there is a constant supply of fissiles as long as U-238 or Th-232 are being bred into U-235 or U-233. This would start new reactors, the U-238 in the SNF will be bred into fissiles keeping your already running reactors running till the reactor vessel is retired. Then the fuel load is moved to a new reactor with an already mature fuel load.

        As for the thermal spectrum molten salt reactors I’ve already addressed the issues with those that this reactor was intentionally designed to overcome. It is far simpler and far more efficient as it requires no new fuel source other than what is already being stored as waste….. ever.

    3. Sorry, I screwed up.  We have more like 1/90 of the fissiles we’d need, so MCSFR would have to stretch our available fissiles by 2 orders of magnitude.

      (this is what I get for not using a calculator despite having several of them ready to hand)

      1. Engineer-Poet and Doug Coombes,

        do you know whether Elysium team wants to enrich chlorine and, if so, at which level of enrichment ?

        1. <shrug>

          Elysium expertise is Doug’s thing, not mine.

          Though I do wish Rod would get on the ball and approve part 1 of my response to Doug.  I waited 2 days before dealing with the auto-censor because I wanted the complete reply to appear at once, but had to give up.

          1. You should really go to the source, there’s plenty of information on this.

            Ed Pheil has given multiple presentations on their reactor design.


            Plus regulatory data which will become available as the reactor passes certification testing.


            Also he and his CFO have discussed the economic and political aspects the Elysium reactor is designed to address.


            I’m at my limits and beyond here, I’ll let the experts speak for themselves.

        2. I think it was a marketing decision by co-founder Carl Perez who comes from an entreper

          I haven’t heard anything about chloride enrichment, I’m foggy about the capture cross sections on different chloride isotopes so I don’t know if that will be an issue.

          Ed Pheil has talked about sourcing the sodium chloride from commercial wholesalers and that a full load for one MCSFR would be about $300. I think it’s safe to assume that would not include enriching the salt.

          1. sorry the start of that comment included one for another post about the naming of the company. Please just ignore it.

  12. Doug Coombes writes:

    “And as I keep explaining, at the fast spectrum U-238 isn’t
    just fertile eating up neutrons in the reactor, it is also
    fissile meaning every 10% of U-238 atoms undergoing fission
    is adding 7-8 neutrons to your neutron budget.”

    This is quite interesting and it is exciting to anticipate public
    release of detailed design specifications and testing schedule.
    I must admit that I have a bit of concern about a possible safety
    issue. My question is this: is the margin between “barely critical”
    and what’s called “prompt critical” a thin one? How can one be sure
    that the margin will not be crossed?

    I must explain that what I’ve read recently about “prompt criticality”
    and nuclear reactor designs says that the ability to control reactivity
    depends crucially on some portion of the reactor’s “neutron budget”
    coming from “delayed neutrons” which are emitted, after some delay,
    from fission product nuclei, as opposed to “prompt neutrons” which are
    emitted directly from fission reactions. This is my source from

    My understanding is that the timing of “delayed neutrons” is
    crucial to control of reactivity because if a chain reaction
    is achieved entirely through prompt neutrons, its growth will
    be orders of magnitude faster than control mechanisms based on
    feedback of temperature, thermal expansion, etc. can be expected
    to respond to it.

    So if my understanding of all this is correct, the occurrence
    of a prompt-critical chain reaction could lead to a major
    accident before anything could be done to prevent it.

    If that is not the case, I would greatly appreciate an
    explanation of how the proposed design could guarantee that
    a prompt-critical accident cannot occur, and how such a guarantee
    could be demonstrated by testing prior to release of the product.

    1. Your question has an answer, and Hans Bethe answered it some decades ago.  He calculated the maximum explosive yield of a fast-spectrum reactor if the fuel rods melted in the middle and the top half fell into the bottom half at the acceleration of gravity.

      He came up with a number in the tens of kilograms of TNT, not enough to breach a decent containment let alone a shield building.  It’s just too slow to get things happening before the reactor core dis-assembles itself from thermal effects.

      Modern designs hold fuel rods from the top, so melting would create a void in the middle and shut the reactor down almost instantly due to neutron leakage.

      1. That wouldn’t apply to reactors with a homogeneous molten core, rather than fuel rods. The Moltex fast salt reactor, unlike Elysium’s design, does keep the molten salt fuel in tubes similar to standard fuel rods. From what I can understand, a heterogeneous core has a fast fission bonus of about 1.03 percent, versus a homogeneous reactor, since the neutrons’ first collisions are more likely to be with another fissile nucleus.
        I recall the French Rhapsodie fast reactor had some unexplained power excursions, and the Russians have delayed development of the BN1200 serial production sodium-cooled reactor while they get a handle on the BN800.
        Robert Heinlein had a science fiction story decades back, where the reactor operators were trying to control what sounded like a molten metal reactor with a generation time much faster than their reactions could be. They had a psychiatrist watching them all the time to see if they were cracking under the pressure. Don’t think anything proposed is quite that twitchy!

    2. Fast spectrum nuclear reactors run on prompt neutrons, they aren’t moderated as in a thermal reactor.

      The design advantage of a MCSFR is that the fissile material is in solution in a salt, it’s not in a concentrated location as with a solid fueled reactor where the neutron flux has to be carefully controlled to avoid a critically incident.

      In the case of the MCSFR, the walls of the reactor are designed in such a way that they reflect neutrons back into the central region of the vessel and given a certain volume of fuel salt, with a certain concentration of fissile material the reactor barely reaches critically. The point where one fission will create an infinite chain of following fissions if everything remain in a steady state.

      In a MCSFR the power output is controlled by how much fuel salt with its fissile material is pumped through the reactor in a specific time. The faster the salt is pumped through the reactor the more fissions there are. But always at barely critical. Your reactor design is your primary control.

      I think it’s a truly elegant design, in its simplicity, efficiency, fuel sourcing and more.

      The guy primarily responsible for the design has a PhD in plasma physics but went into nuclear fission research not fusion. He worked as civilian naval contractor at Bectel I think for 30 years and his entire team came with him to Elysium energy. They have a huge amount of experience to draw on, if anyone can make this work they can.

      1. Doug

        Both prompt and delayed neutrons are “born” as fast neutrons. The thermalization process in moderated reactors doesn’t affect the fixed delays that produce some neutrons later than the instant that fission occurs.

        IOW, delayed neutrons are still important in controlling fission and the rate of power level changes in a critical reactor.

  13. Engineer-Poet writes:

    Your question has an answer, and Hans Bethe answered it some
    decades ago. He calculated the maximum explosive yield of a
    fast-spectrum reactor if the fuel rods melted in the middle
    and the top half fell into the bottom half at the acceleration
    of gravity.

    He came up with a number in the tens of kilograms of TNT,
    not enough to breach a decent containment let alone a shield
    building. It’s just too slow to get things happening before
    the reactor core dis-assembles itself from thermal effects.

    That is an interesting conjecture, but please remember that
    we’re discussing a fast-neutron, molten-salt breeder reactor,
    in which the concentration and distribution of fissile nuclei
    may change within a liquid medium to enable a prompt-critical
    chain reaction, whose growth is exponential in a fast nuclear
    time frame, and whose magnitude could render the strength of
    the containment structure irrelevant.

    This scenario suggests the potential relevance of a particular
    prompt-criticality accident which occurred in a plutonium
    reprocessing plant. The accident occurred when a worker
    switched on a mixer containing an aqueous solution of plutonium,
    in which a vortex increased the density of the dissolved plutonium
    enough to cross a prompt-critical threshold. Details:

    If such an event is indeed possible in a fast molten-salt breeder
    reactor, then I think the reactor’s design specifications, public
    review, and testing schedule would need to be planned to address it.

  14. @Engineer-Poet

    “I read papers for information. I don’t watch YouTube for anything but entertainment.”

    I watch it mostly for its excellent science and technology content. It’s whatever the content creators make of it including the award winning Fermilab channel.

    In the case of the Ed Pheil Elysium videos he goes into depth on all aspects of the MCSFR he’s building at nuclear sector industry conferences. That often include good Q&As with other professionals at the end.

    It would probably answer many questions you may have about the Elysium MCSFR, but your choice.

    1. My problem is that video is just too SLOW and very hard to scroll back to cover details that need review.  I can read 3-5x as fast as I can listen.

      1. Ed Pheil always includes graphics to go along with what he is talking about so I find it fairly easy to jump back and forth if I need to review anything using the progress bar.

        There should be more stuff coming out about this, I’ll keep an eye out. There may be more detailed PDFs out there as well.

      2. @Engineer-Poet

        I agree. Video and audio are great formats for entertainment and for stimulating interest in a topic. But they are nearly impossible to search or to reference as a source of facts and numbers.

  15. There is an other interesting and quite long (one hour) Ed Pheil’ s talk on “Titans of Nuclear”, available on Spotify, too (if you prefer audio rather than video format)

  16. I posted a comment a while back asking about the
    possibility of reaching a prompt-critical state in a
    molten salt fast breeder reactor. Does anyone have
    any information about how such an event might be
    detected before it actually occurs?

    Alternatively, does anyone have info about how the
    design of the reactor might prevent a prompt-critical
    state from occurring?

    I’m a bit worried about small variations in density
    or distribution of breeder-generated Pu239, in a plain
    stainless-steel tank containing tons of it. I assume
    that all it would take is a few kilograms in the wrong
    place to sustain a couple dozen generations of
    prompt-critical chain reaction. The result might not
    explode like a carefully designed implosion device,
    but it might do a very good impersonation of a big
    dirty bomb, which I think would be the end of the

    If someone could answer the question, or refute my
    aforementioned assumptions, in a manner that could
    be easily understood by the general public, I think
    that would also help.

    I notice that Ed Pheil is listed as one of Nucleation Capital’s
    advisors, so maybe Rod or Valerie could start by asking him.

    1. Chris Aoki — “Plentiful Energy” describes the EBR-II which ran for 30 years without incident despite some serious tests for stability.

  17. I found a collection of NRC training docs that answer
    some basic questions about fast reactor physics, last
    updated in 2019:

    A good example from this collection is one about delayed
    neutron fractions for isotopes of interest, and solid-fuel pin
    configurations including breeders and burners:

    “Fast Reactor Physics 2 – Reactivity Feedbacks and Fuel Cycle”

    This is good material, but I think it would need to be updated
    for molten salt fast breeder reactors like Elysium MCSFR.

  18. You know, Rod, it would be nice if you wouldn’t let comments languish in the mod queue for days on end while others get through.  I’m still waiting for you to allow comment-165543 out into the light of day.

  19. Hi Engineer-Poet,
    For what it’s worth, I just started fresh with a new comment
    under the “Leave a Reply” box at the end of the page. The website
    responds to me promptly with an e-mail message asking me to
    visit a URL to validate my e-mail address. I think the “Reply” button
    may deposit your comment in somebody’s inbox to languish.

    1. I get those mails on some replies; when that happens, I see the comment with a URL that includes a moderation hash and a note on the comment that it’s being held for moderation.  On others, the comment simply vanishes instead of any of that happening.

      I keep an archive of comments whether they post correctly or not.  The file has almost 400,000 words in it.

      1. He is a busy man and as he has had this Atomic Insights thing going for a long time may be getting tired of it.

        1. And the comment count here has jumped from 96 to 98 (without MY comment being approved), and they must be so old that I can’t find them when I search for the date.  I went back several days in June, got nothing.

          How am I supposed to read a “new” comment if I have no way to find it but re-reading everything?

  20. Off-topic, but UtilityDive reports Georgia Power has apparently invited a special NRC inspection of construction at Vogtle-3. It isn’t clear how significant this will turn out to be — it’s apparently triggered by GP/Bechtel’s rework of some safety-related cable raceways. I don’t know, I’m not involved, I wasn’t there, and it’s not my problem. It may be a “totally routine NRC special inspection” but once there, NRC is welcome to expand their inspection as their little hearts desire.

    Of course, they are always welcome to do so anyway. We’ll see.

Recent Comments from our Readers

  1. Avatar
  2. Avatar
  3. Avatar
  4. Avatar
  5. Avatar

Similar Posts