bhauth
First, I’d just like to check: was that response written by AI?
The cheapest sources of CO2 are from ammonia production and fermentation tanks. But if you mean removing CO2 from the air, biomass is definitely the cheapest option.
The simplest thing you can do is bury byproducts like sugarcane bagasse, and do something (there are a few options) to prevent decomposition.
The most economically attractive option on a large scale, in my opinion, is conversion to levulinic acid + furfural for chemical products and fuel, and burying the hydrochar. But...
Most countries simply don’t have enough extra land to grow enough grass to replace / compensate for their CO2 emissions.
That requires a better process for conversion than is currently in use and some new uses for those products. Which I have some thoughts on, but that’s a big project.
As for good ways to reduce CO2 emissions in the first place, I think those include:
more working from home where practical
continue improving insulation where it’s bad
shut down old coal plants and build more HVDC lines
Consider this, we’re proposing a moonshot here, not just an incremental product improvement.
If it’s a moonshot, you should either: (1) be working on better chemistries in a university lab or (2) have some experience with manufacturing chemical products relevant to bringing manufacturing costs down or (3) be able to impress people with your understanding of industrial chemistry costs.
In 2022, Hemmatifar showed a stackable bipolar cell capturing at 400 ppm with electrical work of ~0.7 MWh/t while maintaining >90% efficiency[1]. They even ran it continuously for 100+ hours without fouling issues.
-
That citation also only shows release of CO2 at similarly dilute concentrations. A bigger difference between absorption and release concentrations obviously tends to require more energy.
-
poly(vinylanthraquinone) + carbon nanotube electrodes aren’t particularly cheap.
-
When such devices have shown a good cycle life, that’s in a lab with pure materials, not in open air with its dust and various organic compounds.
-
Recent TEA in ACS Energy & Fuels modeled a 200 kt/yr electro-swing system with wind power and projected $56-97/t.
That citation says:
The CapEx is estimated by assuming analogies: the regeneration cell of the AEC process is assumed to have the same relative CapEx as redox flow batteries, and the electrochemical cell of the ESA process is assumed to have the same relative CapEx as lithium-ion batteries.
That’s...a non-analysis. Here’s something easy to understand and more accurate than that: the CapEx of an electrochemical MOF thing is much higher than the CapEx of alkaline CO2 direct air capture. This is always going to be true. The stuff required is just more expensive than “sheets of something or other with liquid running over it”. Even if the energy costs are zero, I can’t see total costs being lower. I know about how much it costs to make such stuff, and it’s just too expensive.
Basalt mineralization (specifically the Carbfix method) injects CO₂-water directly into porous basalt formations.
Ah, you’re pressurizing the CO2 and drilling; I didn’t bother reading that far before. That’s certainly possible, though the basalt isn’t specifically necessary for mineralization. Also, while you’re focusing on fast mineralization, that’s kind of irrelevant for underground injection.
Our breakthrough combines metal-organic frameworks with electrochemical triggering to slash costs from $500+ to under $100 per ton.
No, it doesn’t. The cited paper used over 10x atmospheric concentration at the lowest end, the cost estimates you might be thinking of used higher concentrations than that, and those cost estimates were also incompetent and overly optimistic. The lower the concentration, the more fans and MOF you need for the same amount of CO2, and the more MOF degradation happens per CO2 captured. And the fan requirements are worse with a cyclic system than a continuous flow of alkaline liquid. For CO2 capture from air, the equipment required with this approach is much too expensive, and there’s no way you’re getting even close to $100/ton of CO2.
In general, any chemical process involving electricity is more expensive than just running a mixture over a catalyst, because...well, think about it, it’s a lot more complex. You need 2 electric sides and something between them, with high surface area.
(And even if you could get it, $100/ton is apparently still too high. There are lots of things with a CO2 mitigation cost from $70 to $100/ton that aren’t done because they’re too expensive. Well, there are companies paying high rates for direct air capture right now, but that’s in the hope that it’ll become much cheaper in the future, which it won’t. But this is a moot point here.)
As for storage, the cost of digging up rocks and especially grinding them into small-enough pieces is already too expensive for the amount of CO2 they absorb. If you’re going to do that, you don’t need the MOF system for capturing CO2 anyway, you could just spread the rocks around. Also, it’s kind of weird that you say “basalt mineralization” specifically when there are a bunch of rocks that can absorb CO2 if you grind them up; kind of makes it seem like you don’t fully understand the chemistry involved.
Anyway, the economics don’t work out and it’s not even close. Doing stuff with biomass is a lot cheaper, if you have land available for that.
The main problem with this post is that it assumes “AI” is a monolithic thing that can’t include different systems in the future. LLMs can translate from English to German. AlphaZero can beat the best human players at Go. Different systems can do different things. ChatGPT can’t evaluate the result of Javascript programs accurately, but if you set it up to run Node.js, it suddenly can.
I haven’t looked it up; I’m going to live with the risk that I’m wrong, and so are you.
No, I don’t think I will. I started learning about chemistry largely to be able to read ingredient labels and know what things in them are safe. And now, when it comes to molecular toxicology, I’m one of the best in the world.
birds and mammals independently evolved intelligence
I spoke with one of the inventors of bridge recombinases at a dinner a few months ago and (at least according to him), they work in human cells.
Hmm. I don’t believe that, not without a bit more evidence.
There are already several perfectly good languages for schemas, such as CUE, Dhall, and XSD
This won’t find deception in mesaoptimizers, right?
OK, in that case we can talk by DMs as well. Some LLMs tend to make formal and polite writing with somewhat awkward wording and can do a cursory web search to add more citations than you should expect someone to read, but maybe you’re a student in a country that also speaks something besides english.
You shouldn’t just be looking at biochar; there are other approaches, like drying, adding CaCl2, etc. I’ve seen some lower estimates for biomass burial, eg $50/ton CO2 here. Burial where gas from decomposition won’t escape is another option, eg this paper.
My general advice to you would be to trust cost estimates in papers less. Professors will effectively lie to make their research seem more useful, and there are bad techno-economic analysis papers too. Judging the quality of such papers and learning what parts are trustworthy is just a skill you have to practice.