ChatGPT & DALL-E generated image of a truck with a very large bubble of CO2 coming out of its exhaust.

Chevron’s Fig Leaf Part 7: Carbon Engineering’s Fuel Is At Best 25x The Cost, 35x The CO2 Emissions Compared To An EV

April 27, 2019November 13, 20235 years ago Michael Barnard 0 Comments

Sign up for daily news updates from CleanTechnica on email. Or follow us on Google News!

Carbon Engineering recently garnered $68 million in investment in its air-carbon capture technology from three fossil fuel majors. This is part 7 of the 5 article series assessing the technology and the value of the investment. Yes, a 6th and 7th were tagged on because the story just gets worse.

Carbon Engineering will produce transportation fuels that cost 18-25 times more and have 22-35 the CO2e emissions as just using electricity in an EV.

The first piece summarized the technology and the challenges, and did a bottoms-up assessment to give context for what Carbon Engineering is actually doing. The second piece stepped through Carbon Engineering’s actual solution in detail. The third piece returned to the insurmountable problem of scale and dealt with the sheer volume of air that must be moved and the scale of machinery they have designed for the purpose. The fourth article looked at the market for air carbon capture CO2 and assessed why three fossil fuel majors might be interested. The fifth article looked at the key person behind this technology and the expert opinions of third parties.

In part 6, looking specifically at air-to-fuel, we worked through Carbon Engineering’s public statements and settled on methanol as a higher probability liquid hydrocarbon for it to be synthesizing. We looked at the lower energy consumption and the use of electricity for a portion of it to align the CO2 debt. We established that we’d do energy comparisons in MWh as the common terms.

In this piece, we’ll work through the meaty set of chemical processes to figure out how much more energy and CO2 they are going to emit even with BC’s very low CO2.

How much hydrogen do we need?

Let’s start with how many molecules of CO2 there are in a ton. According to this source, it’s 22,700 mol CO2. As a reminder for people like me who don’t do this every day, mol is short for mole which is 6.022×10²³ molecules of CO2. It’s easier to work in mols than in the result of moles times 6.022×10²³. But the weight of the C specifically is a lot lower. Carbon is only 12 grams of the 44 gram weight of a mole of CO2. So the carbon is only about 272 kilograms of the ton of the CO2.

Assuming a 100% efficient process (they never are), that means that we need 4 times as many hydrogen atoms as carbon atoms. That’s 90,800 mol H, but hydrogen only comes in molecules, mostly H2. We need 45,400 mol H2 to combine with the 22,700 mol CO2. Hydrogen is really light stuff with a molecular weight for H2 of 2. Each atom is one-sixth the weight of carbon and one-44th of the mass of a single CO2 molecule.

The easy way to figure out how much this weighs is to multiply 45,400 mol by 2 g/mol. Turns out you need 91 kilograms of hydrogen to add to the ton of CO2. There’s a process-efficiency catch in creating methanol, which is that you convert H2 at relatively low efficiencies so you have to take multiple passes in most processes, leading to about a 95% final efficiency. So you actually need a bit more hydrogen, about 96 kg.

There are couple of major paths to get hydrogen out of water, and the company is vague on the specific electrolysis process it is using, but all are energy intensive. High-efficiency, high-volume PEM electrolysis sees about 80% efficiency and it takes about 50 kWh of electricity per kilogram of hydrogen. We need 96 kilograms of hydrogen, so that’s 4.8 MWh of electricity. And at BC’s very low 15.1 grams of CO2e per kWh, that’s another 0.07 tons of CO2 emissions as debt for the hydrogen. In Alberta next door, as was pointed out in part 1, the 820 grams of CO2e would mean that hydrogen from electrolysis would have a 3.9 ton CO2e debt.

A 2014 PEM electrolytic hydrogen economic study clustered likely production costs — not retail price — around $5 USD / kg H2 for larger-scale facilities. That suggests that the hydrogen cost is about $475 for the 95 kg.

How much oxygen do we need?

How much oxygen do they need to bond with 272 kilograms of carbon to make methanol? CH3OH is one carbon and one oxygen, so basically half of the remainder of the ton is the amount of needed oxygen. That’s 364 kilograms.

Oxygen should be easy. There’s a lot of oxygen already bonded to CO2. The problem is that the carbon-to-oxygen ratio for CH3OH is the different than CO2. You have to break the bonds of an oxygen atom from every CO2 molecule to get one of each for a methanol. However, its process assertion is that it receives pure O2 from the hydrogen electrolysis process and will use that as its source. So over on the electrolysis side the company spends a bunch of energy to create hydrogen and oxygen, then in the CO2 process it spends a lot of energy to bond it to carbon, then spend more energy to break the bonds leave CO2, and then in the fuel synthesis cycle it spends a bunch of energy to break another chemical bond.

Guess what, breaking up is hard to do. It requires energy to peel atoms off of stable molecules. Remember that creating CO2 is normally exothermic, which means that you get it by burning carbon in the presence of oxygen, resulting in heat. Basic physical laws say it takes as much energy to break the bonds as you get when you form them, plus a little bit for entropy. Does this seem like it might be adding up to an odd energy balance?

It’s likely worth looking in more detail at its process cycle to see if the company could use some intermediate molecule to get carbon without a lot of oxygen attached.

Well, no. The CO2 is the result of stripping stuff off of CaCO3 to leave CO2 and CaO which is added back to the process. Going further back we see K2CO3. Its process is already spending a lot of energy binding extra oxygen and then breaking it back out as part of the process of capturing it. There’s no simpler feedstock from its model that I can see for methanol synthesis than the CO2 output, and most of the literature on processes is CO2 to methanol in any event.

The diagram above is only one of the chemical processes of course. The company also has a hydrogen electrolysis process, then you have to bind them together of course, and making liquid hydrocarbon fuels is creating exploitable chemical bonds by pumping energy into them.

How much energy to make methanol?

That’s also well known, as people work on CO2 to methanol processes all the time. There’s a good 2016 study in Applied Energy, titled Methanol synthesis using captured CO2 as raw material: Techno-economic and environmental assessment, looking at multiple plants producing thousands of tons a year, then an economic buildup. We’ll pluck this out.

Process energy needs for low-energy CO2 + H2 > methanol.

So we need to turn our H2 and CO2 into methanol. The numbers above add up to 1.47 MWh, but we are only creating 0.73 tons of methanol, so that’s another 1.1 MWh and has another CO2 debt assuming BC electricity only, of 0.02 tons. Given the preference for heating using natural gas, the company may be using that instead, in which case the CO2 debt would go up.

272 kilograms of carbon, 91 kilograms of hydrogen, and 364 kilograms of oxygen creates 727 kilograms of liquid fuel give or take a bit. In the end, its ton of CO2 turns into about 0.73 tons of methanol.

What does energy & CO2 look like so far?

Yeah, that’s close to $1,000 USD per for the 0.73 tons of methanol, about $1,250 per ton. What’s the market value of methanol? As a reminder, I mentioned that Methanex sets the global market for methanol. It’s the 800-lb gorilla globally in this 20 million ton a year market. I happen to know this because I consulted briefly with the company and so learned about them. It’s a boom and bust market with long periods of low costs and then periods of high demand, like the jump in 2017 by 180% to $500 USD per ton.

The chart is read as 2015 being the index price at 100%. So the best possible scenario for Carbon Engineering appears to be a price 2.5 times the historically high price of methanol. And the price in Feb 2016 was about $170 per ton, over 7 times cheaper. And no one was able to make methanol to gasoline work economically off of cheap methanol.

So right now, this is looking like it’s economically non-viable, and we haven’t even made it to the worst part yet.

How much energy is in 0.73 tons of methanol?

Methanol has an energy density of 20-22 megajoules (MJ) per kilogram. This means that the embodied energy in the 0.73 tons of methanol is 14.6 to 16.1 GJ or 4.1 to 4.5 MWh.

Wait. What’s that? So far we’ve spent 7.4 MWh (without any distribution or further process costs) and we’re only getting back 4.5 MWh equivalent energy? We’ve already lost 40% of the energy just for the chemical processes which create the useful fuel.

What about when it’s used?

Let’s go back to what the company’s claim is, which is “to generate fuels that are drop-in compatible with today’s infrastructure and engines.” So that means not special engines, but today’s engines. And its examples are all transportation. Let’s do a little thought experiment.

Most car internal combustion engines average about 20% efficiency, meaning that 80% of the energy is wasted as heat. Toyota has a prototype internal combustion engine that hits 38%. Diesel are a bit better than basic gas engines at 30% or so. Jet engines vary drastically as well, but average for a big Boeing is around 36%.

Taking the car example, the simplest model is to add perhaps 10% of the methanol to gasoline to create a mixture that the car engine can run on. With adaptation you can run engines on pure methanol, and in fact some racing series do that. Methanol burns a bit more efficiently than gasoline but has only about half the energy per unit of mass. Blending is a very low energy process, but still you have to distribute the liquids, blend them, do process quality work, and store and distribute the result. Let’s ignore that energy cost for now.

When you burn methanol to power an internal combustion vehicle, with the greater efficiency you still throw away about 75% of the energy in the liquid as heat.

How many MWh equivalent are we left with? That’s about 1.1 MWh. We’re now down to 15% of all of the energy inputs being turned into useful work.

The energy density compared to gasoline means you have to burn close to twice the methanol to go the same distance. That 0.73 tons of methanol is the equivalent of about 0.36 tons of gasoline. That’s about 130 gallons or 500 liters of gasoline capable of driving the average 28 mpg car about 3,700 miles.

What if we actually took the final step and made a fully engine-compatible fuel for cars from this process? The process efficiency for methanol to gasoline is 50% to 60%, so multiply the badness by 2. Gasoline energy is around 90% of the methanol feedstock, so you might be able to travel 3,400 miles on the resulting of synthesized gasoline and the carbon debt would go up quite a bit. The process is pretty inefficient, but let us, once again, be nice to Carbon Engineering and suggest we’re up to 0.6 tons CO2 debt.

What if we just used the 7.4 MWh in an electric car?

Electric cars take about 15-30 kWh to travel 100 miles. With 7.4 MWh, that’s 25,000-50,000 miles. Let’s work with the best case for air-to-fuel and the worst case for electricity.

Those are some interesting apples-to-apples numbers. The methanol path costs 18 times more for a unit of distance traveled and has 23 times the CO2 emissions for fuel as the same starting energy used directly in the worst-case EV. The methanol to gasoline path is worse at 25 times the cost and 35 times the emissions.

What if we just used the electricity that the process consumes, not the natural gas? That’s almost exactly 6 MWh and you could drive almost 20,000 miles on it without burning any natural gas at all.

Who thinks this is a good idea again?

Well, these companies just put $68 million into the company. Assuming that they are actually interested in this completely absurd approach, why would they be?

Could it be because it keeps a lifeline for keeping the engine technology which burns their fuel around for a lot longer? And with a total world market for methanol of only 20 million tons at 40% of the cost at peak commodity markets, this isn’t going to be opening up new demand. The major economic study cited regularly concluded that even with CO2 being free because governmental funds would pay for the cost of ton of it, methanol was far too expensive.

Carbon Engineering’s $100 per ton for CO2 claim is meaningless when the rest of the process makes the entire thing uneconomic. This conclusion has been reached over and over again by analysts. Personally, this is the first time I’ve bothered to do a bottom-up workup of the end to end efficiencies, costs, and emissions because I’ve read conclusions like this from credible sources about a dozen times.

That Carbon Engineering’s technology is just as abysmally bad at achieving anything of actual market value in the syngas space was so obvious to me that I went and looked at other use cases, assuming that it must be doing something else. That’s why part 4 pointed at enhanced oil recovery, the only place its central technology has any fiscal merit without having any environmental merit.

What could the money have been spent on instead?

Carbon Engineering has so far accrued about $6 million USD from the Canadian government and $68 million from the oil and gas majors. The company spent a bunch of money already to get to that point, so let’s assume that it is up at $80 million in funding to date, likely a conservative number.

What if that $80 million had been spent on a wind farm? The rule of thumb is $2 million per MW of capacity, so that’s a 40 MW capacity wind farm. In a year, it could generate about 150 GWh of electricity and electric cars could drive about 470,000 miles on the electricity, about 130 times as far.

Where does this leave Carbon Engineering?

Carbon Engineering has a process that could produce methanol under about the best possible circumstance at 2.5 times the highest market cost in the past decade. It could produce transportation fuel that costs 18-25 times electricity for an EV and would have emissions just for the fuel of 22-35 times the EV. If the same money was spent on renewables, you could drive cars and trucks over 100 times further with CO2e emissions per mile two orders of magnitude less than this air-to-fuel, Rube Goldberg device.

If the CO2 was priced so that its central process was free per ton from the air, they’d still be uneconomic. Carbon Engineering will never find a market for its air-to-fuel products.

The only useful thing that its technology can do is be plunked down on a tapped-out oil well, run off local natural gas not worth shipping to market, and drive the extraction of more oil. Oh, and provide a nice green fig leaf for Chevron, Occidental and BHP.

References

Have a tip for CleanTechnica? Want to advertise? Want to suggest a guest for our CleanTech Talk podcast? Contact us here.

Latest CleanTechnica.TV Video

CleanTechnica uses affiliate links. See our policy here.