The Challenging Economics of Electrofuels
When we consider the pathways to a low-carbon energy system, there are certain areas of the economy that are labeled “hard to abate”, such as long-distance trucking, aviation, shipping, iron and steel, and many chemicals. This is not to imply that other sectors such as electricity and passenger vehicles are easy to abate, as experience has shown, but the hard sectors are those for which there do not exist obvious, cost-effective solutions.
For certain transportation needs, electrofuels, which are carbon-neutral fuels synthesized with electrolysis, are a proposed solution. Today, we will look at what electrofuels are, the rationale behind their use, and why it is so challenging to do so.
Why Electrofuels?
“Electric cars work OK, so why not electric airplanes?” seems like a reasonable question. This is why.
Electric cars have large and heavy batteries compared to the fuel tanks of combustion cars, and this is a drawback but a manageable one. For an long-haul aircraft, long-distance trucking, and transoceanic shipping, energy density is a show-stopper. Battery performance might improve with better lithium-ion design, novel battery chemistries, or entirely different solutions such as flywheels, but for now, it does not look like this picture will change substantially any time soon.
There are plenty of other solutions on the table for transportation needs that cannot reasonably be met by batteries. For trucking, there is dimethyl ether and compressed natural gas, though while energy density is sufficient for such trucks to operate, compared to diesel, the fuel tanks will be larger and cut into valuable cargo space. Hydrogen fuel cells are getting some traction, despite the Nikola scandal.
A consortium of shipping companies has identified alcohol, biomethane, and ammonia as the most promising low-carbon shipping options. Methanol is also considered. Nuclear propulsion is used for submarines, aircraft carriers, and ice breakers, and while there is no fundamental reason why it could not be used for cargo ships, this has not been done beyond a few demonstration ships such as the NS Savannah. An industry consortium NuProShip is working to change that.
Aircraft are particular difficult because the need for high fuel density limits options. Sustainable aviation fuel, produced from various biomass feedstocks, can be blended from 10-50% into jet fuel. Aside from synfuels such as SAF or electrofuels, I don’t see a realistic option for decarbonizing aviation, unless you count carbon offsets.
The next thing we would ideally want from a low-carbon alternative to fossil-based fuels for transportation is a drop-in alternative. This means that the fuel can be used in existing vehicles without modification. A fuel switch to something that is not a drop-in alternative adds additional cost in production chains for new vehicles and/or retrofits of existing vehicles, and also a transitional period where infrastructure for both fuels need to be available.
Economics of Electrofuels
With enough electricity, you can do just about anything, including producing any matter of other kind of fuel. Let’s start with hydrogen.
The many theoretical uses of hydrogen comprise the hydrogen economy, though the problem is that while hydrogen can theoretically be used for many things, such as for a reducing agent in steel production, there are not many things for which hydrogen is the best solution. That would be a good topic for another day. But hydrogen is used for many applications already, and those uses are not going away. As of 2020, 48% of hydrogen was produced from steam methane reforming, 30% from petroleum fraction, 18% from coal gasification, and 4% from electrolysis.
I haven’t found more recent figures, but the 4% electrolysis number has not changed much recently, perhaps because this route in the United States is about twice as expensive as SMR. Keep in mind that electrolysis is not a low-carbon route unless the electricity is produced from low-carbon sources. Hopes for a more favorable outlook for electrolyzed hydrogen rest heavily upon learning curve cost reductions for both the electrolyzers and for electricity feedstocks such as solar photovoltaics.
Ammonia is a widely used chemical, most of all for fertilizer, but for many other uses too. Today, ammonia, chemically NH3, is produced by the Haber-Bosch process, which converts an N2 and 3 H2 molecules into 2 ammonia molecules. Hydrogen production, as described above, is an inherent part of the Haber-Bosch process, but there is active research into alternative pathways that would bypass H2 production. As is the case of hydrogen, electrolyzed ammonia has a levelized cost more than twice as high as that of ammonia from conventional sources. Also as is the case of hydrogen, projections of cost reductions rely heavily on learning curves.
Methanol, chemically CH3OH, also has many industrial uses, including as a gasoline additive, and it is considered as an alternative fuel. Methanol can also be refined into dimethyl ether, which is less energy dense but carries other properties that are potentially attractive as a trucking fuel. You’ll never guess the story here. About 99% of methanol today is produced from coal and natural gas, and given that the cost of methanol production is at least 3-5 times higher for electrolysis than from natural gas, I don’t expect to see much more electrolyzed methanol any time soon.
It would be worth taking a digression into efficiency. The efficiency of methanol production by natural gas, coal, and biomass respectively is estimated at 64-72%, 65%, and 50-60%. For electrolysis, accounting for the primary energy basis of electricity, efficiency is about 15%. Low-carbon sources of electricity need to be just not cheaper, but more than four times cheaper than natural gas if the electrolysis route is to be competitive. A similar problem applies to other electrofuel routes.
Synthetic hydrocarbons, made from electrolysis rather than refined from fossil fuels, would likely be five or more times expensive than refined fossil fuels. Once again, we have the same problem that, on top of the inefficiency of primary energy conversion into electricity, the synthesis of electricity into synthetic hydrocarbons is about 40-50% efficient. This report also finds extreme ratios between the prices of electrolyzed hydrocarbons and fossil hydrocarbons, perhaps a factor of 10 if direct air capture is used for the CO2 source.
Finally, methane—chemically CH4—comes to us primarily in the form of natural gas, but it can be synthesized with the Sabatier process, which converts CO2 and H2 into methane. Aside from producing carbon-neutral methane on Earth, the Sabatier reaction is of interest for manufacturing methane fuel on Mars for Starship rockets, eliminating the need for the rockets to carry return fuel on the trip to Mars. At present, the Sabatier reaction from low-carbon electricity sources (or any electricity sources) is way too expensive to be financially viable.
How are electrofuels carbon-neutral? Carbon-based synfuels, such as methanol and methane, release carbon dioxide when combusted. However, the carbon neutrality derives from the fact that captured CO2 is used to manufacture the fuels. However, electrofuels are only as carbon-neutral as the source of electrical energy. One gotcha here is that the carbon source for carbon-based fuels might be from carbon capture and sequestration. In that case, the CO2 emissions are spread over two applications: whatever the CCS was applied to, and then to the fuel combustion, but neither of them could be properly classified as a low-carbon process.
Do Electrofuels Have a Future?
Maybe, if learning curve trends work out, we will see hydrogen and ammonia electrolyzed from low-carbon electricity. For more complex fuels, though, the cost barriers for electrolysis are so severe that a substantial rethinking of the problem is required.
Without other advances, we would need electricity less than 1¢ per kilowatt-hour, which is less than a fifth of prevailing prices in the United States. Learning curves won’t cut it; getting to this price point would require an energy breakthrough, such as with an inexpensive method of producing fusion power, the likes of which does not appear to be on the horizon. It may require substantial reform to nuclear fission, a technology which has been stymied by misguided regulation and a poor industry structure.
Terraform Industries is attempting to do just this with solar power. A few months ago, Terraform announced that they had synthesized hydrogen for $2.50 per kilogram, which is more expensive than steam methane reforming but in the ballpark. They apply direct air capture for $250 per ton of CO2 captured, an improvement over the state of the art. Then they synthesize methane via the Sabatier process. The claims need to be validated by the market, but they are promising.
New processes of producing synthetic fuels would help. I mentioned an alternate ammonia pathway above. Another possibility is widespread use of thermochemical water splitting to produce hydrogen, which could be powered by a high-temperature gas cooled reactor or concentrated solar. There is also a process termed artificial photosynthesis to produce hydrogen more efficiently than with electrolysis. Hydrogen is a building block of more complex fuels.
We have grown accustomed to a can’t-do attitude with energy. Nuclear power is inherently too expensive. Transmission has insurmountable regulatory barriers. NEPA is unreformable and we have to live with it as it is now. And the economics of electrofuels will never work. This may very well turn out to be the case, and the effort spent pursuing electrofuels would be better spent elsewhere, but defeatism is not conducive to solving major challenges.
Quick Hits
There was a major computer outage yesterday that resulted from a faulty software update from CrowdStrike, a cybersecurity company. The outage disrupted air travel, banks, medical procedures, knocked some television stations off the air, and much else. Troy Hunt, a cybersecurity expert, described the incident as the largest outage in IT history and wrote,
This is basically what we were all worried about with Y2K, except it's actually happened this time.
To be clear, there is no evidence that this incident was itself a cyberattack; it appears to have been entirely accidental on the part of CrowdStrike. But it is a sobering reminder of the vulnerability of the modern IT system and a failed test of preparation on the part of critical service providers.
I recently became aware of a blog post by Jason Crawford from last year entitled, “The spiritual benefits of material progress”. There is a notion in some traditionalist circles that prosperity is spiritually enervating and is therefore something to be resisted. This idea is, in my view, as flawed as the degrowth notion that the solution to environmental challenges is material deprivation. This is a topic that I have been meaning to write about for a long time, but I haven’t been sure how to approach it.