Producing electrofuels needs a reliable, cost effective energy source

Producing hydrogen, ammonia and methanol requires an enormous amount of energy – wind and solar just don’t cut it on their own, argues Giulio Gennaro, technical director at atomic propulsion firm, Core Power.

A lot of consideration is currently being given to hydrogen as a means to decarbonise shipping. While hydrogen can be used only onboard very low powered / short-ranged vessels due to storage and handling issues, ammonia and methanol derived from hydrogen could see a widespread use. Both these electro-fuels have shortcomings, as with most energy sources, but they could be a suitable compromise. It is true that both of these fuels require a lot of tank space, but when obtained from green hydrogen produced by cheap solar photovoltaic (PV) or wind they would be either carbon free (ammonia) or carbon neutral (methanol). The total lifecycle greenhouse gas emissions and health, safety environmental emissions would be fairly low and diesel engines have the unparalleled ability to run on possibly every kind of fuel.

So, have we finally found the silver bullet to solve shipping decarbonisation? Not really.

Green electrofuels such as hydrogen, ammonia and methanol, as opposed to fossil fuels, are energy carriers. This makes their production an extremely energy hungry process. In fact, more energy is required to produce one ton of hydrogen than the energy that can be carried in the hydrogen itself. If that production energy is not clean, or is not durable, we’re not really helping the cause.

The idea of producing cheap, green electrofuels at industrial scale with surplus energy only from variable renewable energy sources (VREs) is a mirage. The reasons are simple:

  • Cheap energy surplus will no longer be a surplus, and it will no longer be cheap as soon as there will be a need for it.
  • The total system levelised cost of energy (LCOE) of offshore wind, the most vouched for VRE, is not cheap at all.
  • VRE, by definition, are characterised by low capacity factors and they are not dispatchable (meaning they do not supply energy on demand, but only when they can). As a consequence of the low capacity factor VRE need to be massively over deployed in order to provide the necessary amount of energy, and that would still not be sufficient, as being non dispatchable, a massive energy storage infrastructure is also needed to ensure a steady and secure power supply. And this energy storage, whether that is batteries, thermal or pumped hydro, has consequential severe repercussions on LCOE, use of resources, emissions and other external factors.

In order to understand the implications of the above it should be recalled that:

  • About 10 MWh are required to produce one ton of green ammonia.
  • IMO GHG Study 2020 estimates for 2018 a total HFO equivalent consumption by global shipping and fishing fleets of about 339m tons. Considering fleet growth and increase in propulsion efficiency it can be estimated that by 2040 more than 425m tons of HFO equivalent will be needed, equating to about 1bn tons of green ammonia equivalent.

The above can then be combined with the figures provided by the US Energy Information Administration (EIA) in respect of a total system that combines the levelised cost of energy and the levelised cost of storage, for new resources entering service in 2040, and presented below:

The final total system LCOE, relevant to achieve 90% capacity factor, necessary to power a large-scale industrial production of green ammonia are as follows.

The actual additional power generation then needed to produce green ammonia to cover one third of the needs of shipping, i.e., about 335m tons per year by 2040, can be assumed as follows:

The conclusions are straightforward:

  • The challenge in decarbonising shipping by means of green electrofuels ultimately lies in the durable, low-cost production of such fuels in large quantities.
  • VRE alone is not an option for such a large-scale industrial production due to the much higher power capacity needed, and the much higher total system LCOE.
  • Production at such a large industrial scale requires dispatchable power generation, not intermittent power.
  • Whatever energy source is used, the power generation capacity which will need to be deployed will be massive.
  • Advanced atomic power, from molten salt reactors would be ideal in this respect thanks to the low total system LCOE and the high-capacity factor.
  • VRE may be used alongside advanced atomic to supplement and provide ancillary power supply.
  • VRE could be used alone only for extremely small scale / local production with negligible impact on the global bunkering industry.

These conclusions are cascading across our industry. In a recent poll of 150 senior executives from maritime transportation, 90% agreed that energy density matters the most. 84% of those polled said we would see floating assets using advanced atomic power to produce zero-carbon fuels in the future, and when finally asked if advanced atomic is a viable solution to decarbonise the shipping sector, 47% agreed and 42% strongly agreed. Just 6% disagreed and 5% strongly disagreed.


  1. At last! An outbreak of well argued facts and common sense concerning future fueling options. The recent 100m dash for those with no sense of direction with lots of grandstanding announcements that do not stand close forensic scrutiny is put under pressure by this article.

  2. The changes needed to decarbonise shipping (and other industries) are huge, and development and utilization of all carbon free energy will be required. Some comments
    – I recognize the LCOE numbers outlined are attractive, but costs for wind (both onshore and offshore) are coming down, so they will be competitive with nuclear within 5-10 years.
    – I am not clear how costs for nuclear can be supported – there isn’t a single working molten salt reactor anywhere world-wide, and development of reactors in the US and Europe are typically a decade behind schedule and billions over budget
    – dispatchable power is not needed when you are generating green fuels, as the fuel itself provides energy storage
    – it is important to understand the timeline required for decarbonisation. In order to meet GHG obligations, the industry needs to decarbonise by 2050 and make significant reductions in GHG emissions by 2030. Even if there is a design certified by regulators, nuclear reactors take at least 15-20 years from concept to operation, so there isn’t enough time for nuclear to make any significant contribution to clean fuels before 2040

    1. Hello Keith,
      thanks for your comments.
      Please note that the projected LCOE are relevant to 2040, so they already take into account cost reduction.
      LCOE for MSR has been estimated on the bases of the one of advanced reactors, it is a yardstick evaluation based on higher efficiency in electricity generation and lower projected CAPEX and OPEX, it aims a providing a qualitatively assessment, not at hitting bullseye, which now would not be possible in any case.
      Construction of large nuclear plants on site, and few of them, is pron to budget overrun, the idea of building smaller, modular, in factories, for marine deployment aims at removing these issues.
      I beg to differ on the fact that dispatachable power is unnecessary when producing green fuel. I feel this is a common misunderstanding and I keep hearing it over and over. This would be true only in case the green fuel production is marginal, but if the goal is to produce hundreds of millions of tons of such green fuels, as it is being advocated, than such production must be at industrial scale, and that means that it must run at capacity 24/7, not for few hours per day at variable load.
      Finally in respect to time scale nuclear has proved being able to quickly scale up, an not as installed power only, but as power actually provided. Therefore, if we are serious about decarbonizing, it can be done, again, and in a sustainable way, also considering that nuclear as a striking advantage on VRE as far as use of resources, land occupation and lifetime.
      It will be done if there will be the political will to do it.

    2. Small modular reactors (SMRs) are defined as nuclear reactors generally 300 MWe equivalent or less, designed with modular technology using module factory fabrication, pursuing economies of series production and short construction times. Also designed with safety in mind not as an afterthought.

    1. Hello Brian,
      please note that there is a difference between “LCOE” and “Total System LCOE” as the latter also includes O&M (operations and maintenance) and Transmission costs.
      EIA puts LCOE of offshore ind for 2040 at about 59 USD/MWh, about the same as you quote, but then considers 29.4 USD/MWh for O&M and 2.4 USD/MWh for transmission costs.
      Please note that O&M for onshore wind is rated at only 7.5 USD/MWh, the striking difference is a telltale of how expensive is to maintain wind turbines at sea.
      Please see page 21 of the following document: “Levelized Costs of New Generation Resources in the Annual Energy Outlook 2021”, U.S. Energy Information Administration, February 2021.

  3. I keep hearing the objection that it takes more energy to produce hydrogen than you get back. But the locating, extracting, processing and delivery of fossil fuels is by no means 100% efficient, not to mention the loss through spills and leaks of oil and methane.

    And I rarely hear the admission that fuel cell energy is far more efficient than thermal systems with those smoke-stack thingies discharging waste heat and waste.

    Comparing hydrogen to the current approach makes it far more atractive.

    1. Hello Thomas,
      Fuel Cells are definitely much more efficient and less emitting than Internal Combustion Engines, however the big obstacle, same as batteries, is the low energy density, and also rather short running life. In the end of the day Fuel Cells are very high OPEX, same as Internal Combustion Engines, and very high CAPEX also. Therefore for either energy hungry assets or high powered assets they are a misfit, while for non energy hungry assets batteries are in general a much better proposition.

  4. I do support the use of nuclear power for ship propulsion, but I have difficulty with this:

    “Production at such a large industrial scale requires dispatchable power generation, not intermittent power.”

    It doesn’t actually seem to follow. Intermittent power is fine for production of hydrogen by electrolysis, surely?

    1. Hello Andrew
      As commented above, intermittent power is ok for small scale production using surplus power generation, but shipping (and not only shipping) needs an industrial scale production of green synthetic fuels, running continuously at capacity.
      The alternative is over deployment of VRE and green hydrogen / ammonia / methanol generating plants, running well below nominal capacity, and being much more expensive being under utilized. IMHO this is a losing proposition under each and every point of view.

  5. For ships over a certain size, straight nuclear propulsion is going to be cheaper and more efficient than any energy carrier. An Emma Maersk-class container ship could easily carry not just one but TWO NuScale reactors and gain quite a bit of useful load due to not carrying fuel oil except for emergency generators. Best of all, the ship could cruise at maximum speed outside of ports rather than “slow steaming” to save fuel.

Back to top button