Terrestrial Energy – Molten Salt Reactor Designed to Be Commercial Success

There is a growing roster of innovative organizations populated by people who recognize that nuclear technology is still in its infancy. Terrestrial Energy is one of the most promising of those organization because of its combination of problem solving technology, visionary leadership, and strong focus on meeting commercial needs.

Nearly all of the commercial nuclear power plants operating and under construction today use the basic design components of solid fuel arranged into a critical mass that can produce massive quantities of heat in a compact volume. The heat produced in that solid fuel is almost always moved by water, with the heat energy that has been transported from the fuel transformed into steam that turns a turbine generator.

During a relatively brief period of rapid manufacturing, a capable component production infrastructure was established with tooling, processes, quality assurance procedures, and skilled personnel. Regulators learned how to review water cooled reactor designs.

Nuclear engineering departments focused on solid-fueled, water-cooled reactor technology because they recognized there was an established market for graduates who were primed with the applicable knowledge. Even though that period of construction came to a virtual halt, operators have continued to invest in developing their skills in extracting the greatest possible value from the plants that were completed.

All that infrastructure results in a substantial inertia that has encouraged most developers of new nuclear technology to stay with the basic model of prior generations — with some evolutionary improvements based on many decades worth of documented lessons learned.

However, one of the lessons learned from using the conventional technology is that there are certain unavoidable cost and schedule limitations associated with the technological choice. Water does not want to remain in liquid form at the temperatures that are desirable in a steam power plant; the solution is to use high pressures to control the physical state of the water. High pressures mean thick-walled containers and piping; thick walls add to welding challenges, make it difficult to control forming processes and lead to lengthy production cycles due to the need to control rates of heating and cooling to add process inspection stages to ensure all quality standards are met.

There are organizations that have developed procedures, built the necessary tooling, and demonstrated that they can perform the difficult tasks well, but the best in the business have reached a stage where there are only marginal improvements available. When done by the best at a sufficient scale, water-cooled, solid-fueled nuclear power plants can compete against most fossil fuels, especially when they get credit for producing clean power that replaces fossil fuel power that inherently must discharge at least some of its waste materials into the environment.

Unfortunately, the marginal advantage is often not sufficient to overcome the risk that the nuclear plant will not be built by the very best and will suffer schedule interruptions that drive prices into the uncompetitive range.

Innovators like Terrestrial Energy believe there are fundamental choices that can alter the competitive balance. TEI’s choice has been to design a reactor that is more akin to a chemical reactor, with fuel that is a dissolved reactant in a solution (in this case, a salt solution) where the solution provides the transport mechanism for the heat produced in a strongly exothermic reaction. Of course, the reaction in this case is not a chemical reaction; it is a fission chain reaction.

The hot reactor fluid is circulated through multiple redundant heat exchangers sealed into the same container as the reactor. Fluoride salt without any actinides circulates on the other side of the primary heat exchangers to transport the reactor heat to a second set of heat exchangers where water receives the heat and boils into high temperature, high pressure steam.

The salt circuits operate at high temperature but low pressure. Low pressure enables containers that are simpler, cheaper and quicker to produce compared to the containers performing similar functions in a water-cooled reactor.

The heat and pressure conditions in the steam generator are more similar to those in a fossil fuel boiler than those in a pressurized water steam generator.

Terrestrial Energy has chose to operate its molten reactor on low enriched uranium — which it describes as a dry tinder — vice thorium, which is the frequently targeted molten salt reactor fuel. According to TEI’s web page explaining that choice, thorium is analogous to “wet wood” and needs a “torch” like plutonium-239 or highly enriched uranium (either 235 or 233) in order to be lit and sustained.

TEI knows there is plenty of available natural uranium at an affordable cost, and that there is plenty of capability to produce the correct enrichment with the ability to expand capacity as needed. Uranium fuel has a well-established supply chain; using it will simplify licensing. TEI is aggressive about commercialization; it is aiming to simplify both designs and related processes in order to drive down schedule-related costs.

TEI understands that graphite is a well-proven and understood moderator and structural material for high temperature, liquid-fueled reactors, but TEI also understands graphite’s characteristics of storing energy and changing dimensions under a sustained neutron flux. Replacing graphite components would be complicated; designing them to last the lifetime of the reactor would require research and development with uncertain results.

TEI has a solution for that issue in the form of producing sealed reactor/primary heat exchanger units with installed redundancy that will last for roughly seven years before needing to be replaced. Each unit will have a shielded space for two reactor modules, one will be in use and one will be cooling off. The design philosophy is similar to that used in staged rockets; the difference is that TEI will not throw away used reactors; they will contain useful materials that can be recycled when conditions are right for that activity to begin.

TEI has developed conceptual designs for three different power outputs aimed at various niches in the power market, ranging from 29 MWe to 290 MWe. Any of the basic power modules can be combined at a power station to provide a large total output power level.

One of the more important commercial decisions that TEI has made is to put its headquarters in Canada and to plan to use the Canadian performance-based licensing process. That process should take several years less than the one that would be required for a US license. Once there are operating units in Canada, presumably it will be easier to show US regulators how their system works.

Terrestrial Energy has a lot of work to do to achieve their goal of producing power by the early part of the 2020s, but the principals have established a plan that has strong potential for success.

Additional Reading

April 12, 2013 A simple and “SMAHTR” way to build a molten salt reactor, from Canada (Note: TEI’s current design reflects several refinements since this 16-month-old article.)

Corrected copy (9/5/2014) – Based on feedback from TEI’s Chief Technology Officer, this version corrects the secondary fluid from “solar salt” to fluoride salt.

About Rod Adams

34 Responses to “Terrestrial Energy – Molten Salt Reactor Designed to Be Commercial Success”

Read below or add a comment...

  1. Kim L Johnson says:

    Hi Rod,
    You have probably heard that Terrestrial Energy Inc (TEI) is targeting Canada’s Oil Sands as its most likely First large-scale Commercial App of nuclear Molten-sal. or Ms. Science.

    By delivering to the field site simple moderate-pressure Steam at ~110 to ~220°C for melting & freeing Tar from the Sand *sans* CO2 emissions, TEI avoids the costs & time needed for Brayton-cycle or other ~complex turbomachinery R&D. TEI could also transport Process Heat to larger or more distant sites using NO-pressure Ms. Mixtures (include More than just Fluorides) that remain 100% fluid & stabile from ~120 to ~450° C.

    • Rod Adams says:

      @Kim L Johnson

      During my chats with principals at TEI, I learned that they are not focused on any particular first commercial application. They have identified numerous potential early adopters.

  2. matt says:

    A timely article on TEI. It’s worth noting that their recent round of financing was oversubscribed, and I’m not surprised.

    For a long time I felt that all the new reactor concepts – molten salt, metal-cooled, SMR, thorium etc. were sideshows. Distractions from the important business of a global nuclear build-out: AP1000s with supply chains ramped up to fill hundreds of orders… Which I had very little hope of actually happening.

    When TEI recently came to my attention, I felt a spark of optimism that grew into genuine excitement the more I learned. A reactor design with a real KISS philosophy, evidence of commercial and political acumen… This is such an obviously better way to do it! Low pressure, no stored mechanical energy, LOCA impossible by definition, worst-case loss of integrity leaves the fuel salt spread out and passively cooling in the plant bund with all fission products captive. It saves money, time, resources… Any power plant based on these will be able to use off-the shelf turbines rather than giant nuclear saturated turbines. I hope to see these, or something like these, being retrofitted to China’s coal fleet inside 20 years on favourable economics alone.

    • Rod Adams says:


      I, too, have become more interested in TEI’s design concept. It has a sound basis with a lot of documented testing underlying the design, but the team has also recognized that trying to produce an ideal machine with all of the capabilities sometimes claimed by enthusiasts complicates the process of getting a product to market in a timely fashion.

      Evolution is a great way to improve on ideas; sometimes revolution is required to jump over to a new development path with greater opportunities for evolutionary improvements.

      Clay Christensen’s work on disruptive innovation would describe this process as another ‘S’ curve development opportunity.

      • Eino says:

        A link to Wikepedia and Clay Christensen:


        As noted above maybe the KIS principle is what the nuclear industry needs. Another expression used by Henry J Kaiser that comes to mind is, “Find a Need and Fill It.” Kerosene replaced whale oil when it was becoming scarce, the incandescent light replaced dangerous burning lamps and candles, today’s need for more efficient lighting is replacing that with LEDs. The type of reactor given in this article satisfies the need for affordable clean energy and could replace coal.

  3. John Tjostem says:

    The USA regulators need to move past ALARA in response to unwarranted fears of radiation and find a way to facilitate revolutionary improvements in reactor technology.

    • Stephen Galperin says:

      Absolutely. The rest of world is going to develop this technology and if the US doesn’t get on board it will be yet another technology we are buying from other countries (China).

  4. David says:


    I agree that this is exactly the type of design that needed to be done. Swapping out the cores – on line! Talk about capacity factor – wow. Also, these should load follow instantly making the grid incredibly stable.

    I am planning to follow them very closely. MSR’s are the right way to go and I am interested to see how they accomplish criticality with low enhanced uranium.

    Finally, one point I did not understand was the avoidance of plutonium. I understand that plutonium is simply a fuel. Why avoid it’s production?

    • Rod Adams says:


      As I understand it, the technology does not avoid the production of plutonium; it reduces the accumulation of plutonium per unit of output power. It does this by efficiently burning more of the produced plutonium than is possible in solid-fuel, light water-cooled reactors, especially those systems that depend on distributed boron in the coolant for chemical shim.

      • David says:


        Thanks, I read their website more carefully after posting and noted that they are talking about reducing Plutonium waste – not reducing plutonium production during power production. In other words, burning up the stuff so it gives us power rather than being caught in a waste stream.

        I am excited about this product. They have a great website and that shows thought. Their focus on actual production rather than research at the edges is refreshing.

  5. Andrew Jaremko says:

    Rod – thanks for this post. I’ve been following Dr. LeBlanc via Gordon McDowell’s videos, and it’s great to see an announcement of the concepts he’d been hinting at for the last couple of years. I agree that the ‘nuclear heat battery’ is a product whose time has come, and I also applaud TEI’s choices in the design and concept. Using existing supply chains is simply the right thing to do.

    I did some mental arithmetic, trying to ballpark how much a reactor would cost by comparing it with other types of engines. I came up with a WAG around $50 per kilowatt, once the production line has reached about 100 units and is running smoothly. That gave me thermal energy at about $.01 per kilowatt-hour amortizing the overnight cost of the reactor over a seven year life span. Refurbishing and recycling will drop that again. I think we will see action from TEI!

  6. Kim L Johnson says:

    Rod, Matt (R?) and David (W?),*
    . . . . .*(have i guessed your last init. right ??)
    My over-all opinion of TEA is also basically the same as yours, especially Re. their minimizing of Finan$ial and Regulatory Risks by astutely restricting a “World-First”
    {MSR Commercialization} to *Canada*!
    So in M.H.Op, TEI certainly has “Got the Reg’s and the $ Rounds Right” for the West’s 1st-comersh MSR, which is why, I bet, you gents Now are *bullish* on Molten Salts.

    Only Care & Concern I’ve got regarding TEI is that the Board & all its management teams lack *Chemical* Scientists (Ch.Engs & chemists) who think longterm *and* understand the Complex Chemistry of Fuel & Solvent Fluorides, Fission-Prod. fluorides, and the various catalytic compounds (∑of which *aren’t* Fs), which comprehension is *essential* for half-way-efficient functioning of even the “simplest” DMSR. (latter “Denatured” MSR category definitely covers TEI’s ‘ĭmzer’ – a more efĭsh Pronunci. of the o/w *4*syllable I.M.S.R…).

    So to ensure an ĭmsr or any “Ms. Whatever” of *timely* Technical Success – i.e., enjoying *decent* Chem. & Cat.s for improved Fission-efĭsh, Vessel-life, etc, all within Real-project constraints – the decision-making Management has *got* to be *familiar* with the right answers (and the reasons for them) to following 2 Ms. Chem examples, just for starters(!):
    1. the Difference entre “Lewis” Acids & Bases and which should most Ms. Baths want to have much more of…
    2. Are Baths best if their “Rēdox” Potential is maintained mostly in a {somewhat “reducing” state} *versus* {deciding “oxidizing” state} ??

    -Kim J

    • David says:

      HI Kim,

      I understand your concern about the lack of Chemical Engineers and chemists. I would note that silence does not mean lack. Also, there is a great opportunity here to partner with another business – http://flibe-energy.com which has exactly the type of expertise you are asking about. Also Ohio State has several departments, especially in the Cleveland area who are very up on Molten Salts and reactors. If these guys are as smart as they seem they will have several consulting contracts.

      • Martin Kral says:

        David, Good reply. Only the best and brightest will do. Glad to hear that you guys are open to outside help when needed. Good luck, I too have been following your progress.

  7. Benjamin Haas says:

    How important is serial production of the reactor system compared to efficient construction of the rest of the plant (rebar, concrete foundations, buildings, etc.) in reducing capital costs? I see a lot of emphasis on the design and simplicity of the reactor. What about making sure excessive concrete and steel are avoided for the rest of the facility?

    • Ed Leaver says:

      I think Toshiba-Westinghouse and GE-HItachi have shown how to minimize structural concrete and steel while increasing safety with ESBWR and AP-1000. These designs really are impressive. I doubt their lessons are lost upon the competition.

      • Eino says:

        I went to their website to have a look.


        Note how much smaller they are in their comparison to other proposed small reactors. How do they get such a better power to size ratio? Is it more efficient use of fuel? Do they not need the passive cooling measures that the others have? This small size may mean less concrete and steel.

        • Eino says:

          I took another look. Much hotter. More thermodynamically efficient.

        • SteveFost says:

          My understanding is that the small size, i.e. large power density, is made possible by the large temperature range in molten state of the fuel (at atmospheric pressure too) coupled with excellent specific heat capacity and by the fact that the fuel and coolant are one and the same. With fuel integrated with coolant, there is no problem of heat transfer between solid fuel elements and liquid coolant, so heat transfer can be accomplished in a smaller volume without damaging fuel elements.

          • EdP says:

            SteveFost, you got it 3 aptly right on all counts. Higher delta T and higher coolant capability and NO fuel heat transfer will make the entire plant smaller, including energy conversion system vs a nuclear turbine system, but not quite as small as for a gas or coal turbine, as the temperature is not as high as those. Power density is much more than an order of magnitude higher. High temperature allows high radiation cooling decay heat removal efficiency designs, but reduced heat loss at low temperature to retain heat to keep it liquid for a faster restart, if desired. Recall that 40% of the decay heat is separate from the reactor do to fission gas removal, but that needs to be passively cooled too.

        • matt says:

          Passive cooling is supposedly inherent. The fuel and reactor operates at a far higher temperature than a PWR and can survive going even hotter without containment problems. With fission shut down so there is only decay heat to deal with, at that temperature it can easily transport heat by convection and shed it by radiation and conduction. The heat capacity of the salt will deal with the first few hours of intense decay heat; after that it may be able to shed the rest through conduction into the ground and radiation from the top of the reactor vessel (haven’t read any analysis on this.)

          Another big saving in concrete and steel is in the containment. A PWR containment has to be able to deal with a full-bore pipe rupture and water at 160 bar suddenly flashing to steam. It needs to be large enough to accommodate the steam expanding to a containable pressure, airtight, and strong enough to hold that pressure. PWR containments are monsters because they have to be.

          By contrast, if this thing ruptures, not a lot happens. No pressure, no stored mechanical energy, just red-hot salt spilling into a sump lined with furnace refractory, much the same as pizza oven cement. The real bad stuff – iodine, caesium and strontium – is all chemically bound in the salt and will not boil out or escape. Your containment doesn’t need to be able to hold a fully-fledged steam explosion or any internal pressure. You let the decay heat drop over days and weeks until the fuel freezes solid and that’s as bad as gets.

          Additionally you don’t need thick-walled vessels with ASME-N documentation from alloy-source onwards for these things. Not nearly as many valves or welds, no redundant coolant loops with redundant motors and redundant backup generators and RCIC… They can’t help but be a whole lot cheaper.

          • jmdesp says:

            One should stay very careful about how that spilling might get described.
            Keeping in memory the way sodium fires have carefully been misrepresented into something absolutely horrible which they never have been. The fire at Monju just stopped by itself never damaging much anything, and causing no radioactive release. It still has been used as the absolute proof sodium based fast breeder reactors are completely uncontrollable.

  8. Cyril r says:

    This is an extremely exciting development, and I am in the amazing and fortunate position to actively contribute to the technology design process. This is actually happening with this company, a very exciting time ahead for all to look out for….

  9. David L. says:


    Thanks so much for your article, well thought out and written as usual.

    A small issue I just wanted to correct. We are investigating an option that includes the use of “Solar Salt” (nitrite and/or nitrate based salts) as a third salt loop but we would not consider it for the secondary coolant that enters and exits the IMSR core-unit. Solar salt is an oxidizer and we would not want to introduce a chemical potential into the core. The secondary coolant is meant to be a fluoride salt that can play nice with the fuel salt if there is ever any leaks in the heat exchangers (many options under review). An added third loop of Solar Salt is simply a nice option (well studied by ORNL back in the 70s) that eases steam generation complexity and offers a good way to transmit heat energy for industrial users.

    • Rod Adams says:

      @David L

      Thank you for your compliments and your comment. I have corrected the article.

    • Engineer-Poet says:

      An idle question:  was the restriction of oxidizers (nitrates) to a tertiary coolant loop decided on BW (Before WIPP) or AW (After WIPP)?

      (I am still flabbergasted that anyone would put organics with nitrates.)

      • Cyril R. says:

        The idea of a nitrate third loop has been around for decades, the first work was actually done around the MSRE project time. It was already available at that time as a high temperature heating fluid in other industries, and it was known to be compatible with steam with a lower melting point to boot. It is also good in mopping up tritium (as still volatile but not mobile T2O). It works well with a steam cycle.

        Quite a few tests have been done with hydrocarbons and nitrates, its not as bad as you think, though the IMSR uses no organics in any quantity. Though nitrates are strong oxidizers, they are not very volatile, so are more of a risk because of high temperature as an ignition source than donating oxygen to a fire. Hydrocarbons don’t dissolve into molten nitrate, so reaction rates are small.

        Oxygen into the fuel salt however, in any quantity, is something to avoid. Oxygen increases corrosion and can cause uranium to come out of solution. In all cases the IMSR is designed with fluoride salt secondary loop so a leak is chemically inconsequential.

  10. Jeremy Owston says:

    Of all the proposed reactor designs out there to replace LWR technology this for me has had the most thought from a commercial standpoint. I like the thought process of dropping any intension of going down the breeder route which no customer is really interested in, also not getting carried away with thorium because it is the in thing at the moment.

    It would be interesting to know what route they are taking with salt selection (i.e. will it include lithium and beryllium salts with there cost and handling difficulties). Also will be interesting to know whether they are going to use drain tanks or not and whether from a safety standpoint are they necessary. For instance could you just let the reactor stay in a hot shutdown mode where the negative temp coefficients stop the reaction thus removing the complexity of either a control rod or rods, or in the case with drain tanks having freeze plugs with a heating method in the tanks to melt the fuel and a pump to move the fuel back into the reactor?

    I could imagine the reactor core having no moving parts and in its simplest form having heat exchanges above the core with convection currents driving the heat extraction. Thus it would be just a tank with graphite, some heat exchanges and a separator for volatile fission products above it. Such a system would be very cheap and easy to build. Shutdown achieved through the expedient of stopping heat extraction.

    Lets hope they make this happen

  11. Jeremy Owston says:

    Also I would be very interested to see a company like Rolls Royce on board with a project like this. They would bring a desire to be in the civil nuclear field, huge experience and knowledge in nickel super alloys from there aero engine business, and the manufacturing and engineering capabilities and facilities to build and commercialise the product. A model operating like there aero engines would suit very well where rather than sell the customer a reactor you sell the output nuclear heat via leasing the reactor as per with aero engines where RR lease the engine to the operators so they can manage all the assets and get stable returns as opposed to the lumpy nature of sale contracts.

    Such a setup would mean that the same reactor core could end up a number of sites as and when the customers needed it without the customer needing to have the sole risk of financing the reactor and predicting long term demand for its product (i.e. nuclear heat)

  12. Murray Chapman says:

    Interesting idea to get Rolls Royce on board. If the power generation switches from steam to Brayton Cycle with nitrogen it’s an even better fit with their gas turbine power generation business. They can also bring their nuclear engineering experience…