Correction – 2 Megawatt Walk Away Safe Truck Movable Nuclear Reactors

NANO Nuclear is a startup that has raised over $8 million to develop micro nuclear fission reactors with up to 2 megawatts of power. They will be transported by Semi Trucks.

I, Brian Wang of Nextbigfuture, was contacted by the Nano Nuclear team for a correction that their initial nuclear fission reactors will be 1-2 megawatts of power. Previously, I reported 20 megawatts because the DOE report was talking about 20 megawatt microreactors but this was just a definition.

They plan to sell 1,000 microreactors on a yearly basis for a trillion-dollar industry. Nuclear currently supplies some 18% of US power needs off 95GW of installed capacity, according to the Energy Information Agency (EIA). The Department of Energy (DoE) forecasts the country will need around 200GW of new nuclear capacity to reach net-zero by 2050.

The products are ZEUS which is a solid core battery reactor and ODIN, a low-pressure coolant reactor, each representing advanced developments in portable, on-demand capable.

The Zeus nuclear microreactor is engineered for safe operation in remote locations. Its ‘walk-away safe’ feature ensures stability and safety, minimizing risks in varied environments. This aspect is critical for ensuring consistent operation without the need for active intervention.

Portability and Adaptability: The modular components fit within standard shipping containers, facilitating transportation to remote sites. This feature enhances its utility in areas where traditional energy infrastructure is not feasible.

The second Advanced Nuclear Reactor (ANR) design in development at NANO Nuclear, ODIN aims to diversify its technology portfolio. The ODIN design will utilize conventional fuel form with up to 20% enrichment, helping to minimize the required development and testing program schedule and costs. The proprietary reactor design ODIN will usee low pressure coolant to minimize the stress on structural components, improve their reliability and service life. It will also use a unique reactivity control system design, aiming to have high reliability and robustness through minimizing the number of moving parts.

The reactor will operate at higher than conventional water-cooled reactor temperatures, which will allow resilient operation and high-power conversion efficiency in generating electricity. The ODIN design will aim to take maximum advantage of natural convection of coolant for heat transfer to the power conversion cycle at full power and for decay heat removal during reactor shutdown, operational transients, and off-normal conditions.

19 thoughts on “Correction – 2 Megawatt Walk Away Safe Truck Movable Nuclear Reactors”

  1. In essence, we should probably bet on Molten salt Reactor. Easy to maintain and relatively smaller than the conventional one

    • Is that what you learned on Reddit? That the reactor we don’t operate at this time is “easy to maintain”? What makes it smaller? It still requires heat transfer area right?

  2. But what happens if you shoot it like individuals are doing with Tesla’s cyber trucks? What will a 50 caliber machine gun do to this beast?

    • TRISO fuel. You bury it with sand and lead shot if you can’t clean it up right away. Then you come in (hopefully with robots by then), and clean it up with long handled tools and shielded flasks to carry it all away.
      Time distance and shielding. The TRISO controls almost all the possible site egress of activated materials from possible water solution (not all), TRISO htgr’s can air cool and remain safe.
      So even if you blew it up. It would be harder to clean up (not impossible), it would not melt down, or cause the steam and metal fires which are the main threat in legacy reactors.

      • The TRISO fuel was developed specifically for helium cooled gas reactors. Helium is inert. Unclear how the material will react in other types of coolants.
        Historically, production costs are reduced by making machines larger and more efficient. Reactors are not microchips

        • Isn’t SiC sort of like a super material? It doesn’t melt but decomposes near the melting point of UO2 (5100F). It is used for abrasives, crucibles, disk brake rotors, etc. I would venture to guess that the durability of SiC and TRISO is second to none… so much in fact that it is likely un-reprocess-able. It is expensive though, because the TRISO is built using vapor deposition. The larger pebbles can be graphite compacts, but the TRISO kernels are made using PVD IIRC.

  3. I went to the Nano Nuclear Energy web site:

    https://nanonuclearenergy.com/

    I can’t find any technical information on what the reactor coolant for ZEUS and ODIN might be (helium gas, lead-bismuth, molten salt, liquid sodium, etc.), or what the power conversion device might be (e.g., sterling engine, helium gas turbine, etc.). There is apparently no information on how the core is designed, what the control rods are, or even whether either design is a thermal or fast neutron reactor. This is all pie in the sky wishful thinking nonsense intended to seduced starry-eyed idealists and gullible investors. Cave Emptor!

  4. I took a look at their technology, as outlined in Nano website. Pretty light on information, particularly the end use. That is troubling from a technology standpoint, and really worrisome from an investment standpoint.
    Turning heat into something commercially useful like electricity is not a trivial exercise. Simply heating up a room with heat given off by fission and radioactive decay is not particularly useful.
    High efficiency power production invariably means high temperatures, which can be a problem for the reactor, depending on the coolant. Unclear on how the machine will do something useful. I have a number of patents on advanced energy systems, so I have a pretty deep technology understanding, particularly nuclear systems.

  5. It really doesn’t matter 2 or 20MW since zero will be built and the net will be the same, zero.

    Using micro reactors to supply baseload is not going to work. The darn little things will need millions of dollars in fuel only to hopefully make 2MWe and sell it for what? $300/MWe? And where? Manned by whom? Paid what? $300/hr? I don’t even need to argue against micro reactors – they just will never matter.

    I put some thoughtful comments together in the KP-FHR article, describing how N2 is a rather strong poison, which makes the loss of coolant accident for the ‘Adams Atomic Engine’ lead to a power excursion. I performed a set of calculations for Rod that showed near prompt criticality on depressurization from 10 bar, and more than double prompt critical on depressurization from 50 bar. The AAE is unsafe for its choice of coolant, which was chosen by Rod to use COTS aviation compressor/turbine.

    • HALEU fuel is a no go in terms of cost, also this is HALEU in TRISO for even higher cost:

      HALEU is frightfully expensive (calcs)
      https://energyfromthorium.com/2024/01/09/25k-haleu/

      @scaryjello – Rod Adams was planning to use N15 as a coolant. The Russians are planning on using N15 in a uranium nitride fuel, and this is the fuel being used in the Transformational Challenge Reactor.

      Can you give a link to your comments? The comments system doesn’t allow viewing of a users comment history…

      • In a fast reactor the nitrogen would be transparent. It is a thermal poison with an xsect of about 2b compared to 0.0035b for carbon. A graphite moderated nitrogen cooled reactor would have a positive depressurization coefficient.

        You can search Atomic Insights for the last article about the AAE. The assumptions used in the worth calc were based on a PBMR geometry.

        • The UN and UC fuels are basically metallic and react with water. If slightly hypo-stoichiometric, metallic U appears at the grain boundaries, and it burns in air. I have in my hand a book from 1964, Ceramic Fuel Elements, Robert B. Holden, for the American Society For Metals. All these compounds and quasi ceramics, such as uranium-silicon/nitride/carbide/phosphide were proposed and tested back then, back before the decision was made that UO2 is the least-worst trade-off wise for power reactors. The Na-cooled reactors in Asia use UO2; although it isn’t as dense as alloy, it is very stable. The Na-cooled reactors that have operated in North America used metallic fuel. With metallic fuel, you either must reprocess it or convert it to oxide for safe storage. In fact, every fuel material other than oxide or TRISO will need to be converted to oxide to facilitate short and long term storage (5 to 5E5 years), which is a point that MacFarlane makes over and over again much to the chagrin of gen4 fans.

    • They won’t be used in baseload markets. They will be used in remote locations, hard to reach, hard to supply, no roads, dangerous access.
      There are many places in Canada’s far north where the real cost of a MWhr without subsidies can possibly approach $2000MWhr. More than likely however, all you need is a minesite with diesel gensets that have to airlift, sealift, or ice road all their fuel in for a whole year to reasonably reach $400MWhr to $700MWhr. This is comfortably at the low end.
      Also portable/sealed core reactors are more likely going to be more acceptable to the local populations as the reactor and fuel leaves the site at the end of its service life.
      They will not be contributing to baseload.

      • “They will be used in remote…”

        Yes. That is the only pitch anyone dares to make.

        This could have been done at any time in the last 50 years with thermo-electrics bottomed by a compact steam engine (triple expansion or turbine). A 20 MWt 500 psi nuclear boiler would be rather easy to develop (1000 liter volume, 20 kW/liter), and an air-cooled radiator.

        Thing is, once the neutrons start zooming around, the NRC wants (actually has enforcement authority) to make sure a qualified team is running the apparatus. The job couldn’t be left up to whatever space marines were willing to do a 6 month jump for $100k.

  6. Small reactors just make more sense to me then very big ones that take decades and uber-billions to build. Ad to that technology does change rather quickly these days. A reactor, designed 20yrs ago is already obsolete. You need to incorporate new technologies into reactor design as quickly as possible. It just seems obvious; it’s easier to do that w/something smaller then bigger. I’m a fan of nuclear power, if it’s done rationally.

  7. A thousand reactors per year for a $1 Trillion industry? $1 Billion per 2MW reactor? $500 thousand per kW? That’s 500x the cost per kW of a diesel electric generator. There may be markets where it’s worth that much of a premium for avoiding the need to import diesel fuel. Maybe the South Pole. But that’s not 1000 units per year.

    I suspect a typo.

  8. They bothered to contact Brian but failed to mention any actual details of their design beyond repeating they are low pressure primary coolant loop with HALEU. If they can’t communicate basic design details that explain their value proposition then they are no better than any other fluff SMR startup.

    Their renders suggest compact air cooled balance of plant, that’s a pretty important detail.

    • Your comment is well taken. Perhaps they consider providing certain detail, “proprietary” Perhaps. But I always ask questions when someone says “it’s that” or is classified. That can be another way of saying “We don’t know”.

Comments are closed.