Skip navigation
Share via:


Synthetic fuels from CO2, powered by advanced nuclear reactors



The most effective way to "create step-change improvements in the GHG emissions associated with the production of fossil fuels" is to burn much less fossil fuel.

Neverthess, civilization needs liquid fuels. Preferably hydrocarbon fuels, to minimize disruption to existing systems.

The solution is synthetic fuels produced from ambient CO2 using nuclear energy. The fossil fuel companies are ideally positioned to make this a reality.

Category of the action

Fossil fuel sector efficiency

What actions do you propose?

To make this work, we need two things: an inexpensive high-termperature reactor, and a way to produce carbon-neutral liquid fuel

The Reactors

The following reactors have the potential for low cost, excellent safety, and proliferation resistance. They operate at high temperatures, providing process heat that can be used directly for chemical processes. All these reactors operate at atmospheric pressure.

Three of these are molten-salt reactors, using liquid fuel. In general, this type of design has a number of advantages. Safety is excellent; in the event of trouble, the fuel is passively dumped into cooling tanks. Decay heat is minimal since, with liquid fuel, fission products can be removed continually. There's no fuel fabrication cost, nor any offsite reprocessing. There's nothing to drive any sort of chemical explosion (such as the famous hydrogen explosions at Fukushima).

Denatured Molten-Salt Reactor

The DMSR is a molten-salt reactor designed for low cost and ease of development. It's a thermal-spectrum, non-breeder reactor. While it wouldn't scale as far as breeders, it may be an effective interim solution, and the quickest route to production molten-salt reactors. This design is also considered particularly proliferation-resistant.

Liquid Fluoride Thorium Reactor

The LFTR is a molten-salt thorium breeder in a thermal spectrum. It has the potential to be inexpensive, but does have more complex chemical processing than the DMSR, and probably a longer development time. However, once started it can be fueled by non-fissile thorium (which it breeds to fissile U233), and its startup fissile requirement is low, allowing rapid rollout.

Integral Fast Reactor

The IFR is a fast reactor, developed at Argonne National Laboratory over a period of thirty years. The design is near production-ready, but the program was cancelled at the final stage. Nevertheless, GE-Hitachi has a version which it's attempting to bring to market.

The IFR is the only reactor described here which uses solid fuel. However, unlike most solid-fueled reactors, the fuel is a metal rather than an oxide, which provdes good passive safety, easy fabrication, and easy on-site reprocessing. The coolant is liquid sodium at atmospheric pressure, allowing a pool with very large thermal capacity. The fuel includes a mix of plutonium isotopes, which is impractical for bombs and difficult to separate into weapons-grade pure isotopes (more difficult than just enriching uranium).

The IFR is estimated to have costs similar to conventional reactors. It requires a substantial fissile inventory. As such, rollout with IFRs alone may be somewhat slow, with a doubling time of about eight years.

We can speed the rollout dramatically by using IFRs to breed fissile, which is used as startup for LFTRs, which require a tenth as much startup fuel as IFRs for the same size reactor. Using this method, very rapid conversion of civilization to nuclear energy sources could be achieved.

Dual-Fluid Reactor

The DFR is a hybrid of the above concepts. It's a fast reactor, using molten salt fuel and molten lead coolant. However, at this point it's purely a concept, and hence would be the longest to get to production.

As a fast reactor, the DFR has very high "burnup," fissioning almost all its fuel, and hence producing very little long-lived nuclear waste. Since it produces a lot of neutrons in a very fast spectrum, it's also quite good at eliminating existing nuclear waste, or breeding new fuel from non-fissile feedstock. The reactor would have a doubling time of four years. Used as a breeder for LFTRs, it would simplify LFTR design by producing U233 as startup fuel, instead of plutonium.

The operating temperature would be 1000 degrees C, the highest of any of these designs.

The reactor is claimed to have no unsolved materials challenges. 

The reactor is claimed to have no unsolved materials challenges. The estimated cost is $1.2/watt capital cost, and less than one cent per kWh for electricity.

The Fuels


If CO2 can be economically concentrated from the atmosphere, gasoline can be produced by the methanol-to-gasoline process (MTG) already used in production by the oil industry. In theory, the energy needed to concentrate CO2 is only about a tenth the energy that would be contained in the produced fuel.

One well-known proposal to make gasoline from ambient CO2 is Green Freedom, proposed by Los Alamos National Laboratory. Using process heat from a GenIII nuclear reactor, along with MTG, Los Alamos estimated it could produce gasoline for $4.60/gallon conservatively, or $3.40/gallon assuming two modest technological improvements. About half the cost was energy cost, so a much cheaper reactor such as the DFR should lower significantly lower the price of fuel, allowing it to be quite competitive.

Nobel-winning chemist George Olah is working on another method of absorbing CO2 from the ambient air, using an inexpensive polymer adsorbent.

The first commercial plantproducing methanol from CO2 started operation in 2011.


Using a similar process, DME can be synthesized, and used in diesel engines with minor modifications.

Who will take these actions?

Multinational oil companies are the perfect candidate. They have the capital for a major development effort, experience producing synthetic fuels, and fuel distribution networks.

By transitioning to cheap synthetic fuels, oil companies can protect themselves from competition from other energy sources and electric cars, and secure their futures in a world becoming increasingly concerned about carbon emissions. They could even expand their businesses, taking energy generation from the coal industry.

The conventional nuclear industry could also make the effort. Arguably, the legacy industry hasn't been especially progressive regarding radically new reactor designs (though GE-Hitachi is attempting to promote a sodium-cooled fast reactor). Newcomers to the field may be more effective, and molten-salt reactors are a sufficiently different from molten-salt reactors that conventional industry experience might not be that helpful anyway.

On the other hand, a partnership may be best of all, particular with GE-Hitachi if using the IFR design.

Where will these actions be taken?

Developing new nuclear reactors in the U.S. is quite difficult due to overbearing regulation and political resistance. China is much more amenable, and is vigorously developing advanced reactor types. This includes a billion-dollar molten-salt effort.

An additional consideration is that China is rapidly developing and already the world's largest CO2 emitter.

Other nuclear powers could also play a part, if willing. The U.S. would certainly be a powerful competitor if political issues can be overcome.

How much will emissions be reduced or sequestered vs. business as usual levels?

The amount of reduction will depend on how quickly these technologies can be deployed, which in turn will depend in part on their cost.

Ultimately, these technologies could make civilization entirely carbon-neutral, even while using hydrocarbon fuels into the indefinite future. "Peak oil" will cease to be even a theoretical concern.

What are other key benefits?

If low costs can be achieved with innovative reactor designs, it could have a substantial impact on electric power production.

What are the proposal’s costs?

Initial capital requirements would be high, with billions required for development of new reactors and CO2 extraction processes. Deployment costs would also be high. But it wouldn't be pure cost; these would be for-profit ventures.

Time line

The first demonstration DMSR and IFR reactors could come on line within a decade.

The first IFRs would be fueled by stockpiled nuclear waste. The U.S. waste stockpile would be sufficient to start 70 gigawatt-size IFRs. Each would produce sufficient fissile each year to start 70 gigawatt-size LFTRs.

(If China were to implement this plan, perhaps it could offer the U.S. a good deal on waste disposal, as a way to secure that enormous fuel supply.)

Oil companies with ownership of these technologies could use profits to build synthetic fuel plants, selling the fuel through existing distribution networks. Deployment can be relatively rapid since we only need infrastructure at the source, instead of changing distribution or end-user equipment.

Of course, fuel supply isn't the only limiting factor. Others include available capital and construction time. Inexpensive modular reactors in mass production could help significantly with both.

Related proposals

Integral Fast Reactors Can Power the Planet

Different kind of nuclear power solves world energy needs without pollution

Molten Salt Reactors for Process Heat in the Oilsands


Tom Blees, Prescription for the Planet

David MacKay, Sustainable Energy: Without the Hot Air

Robert Hargraves, Thorium: Energy Cheaper than Coal

Dual-Fluid Reactor:

George Olah paper

Green Freedom