Always flue gas, never atmospheric concentrations.
We have loads of good ideas for scrubbing flue. We have no good ideas for atmospheric CO2.
420/1000000 isn't a lot of particles per million. Moving and concentrating those particles and grabbing them is going to require an _incredible amount of energy_ no matter how you swing it. People have not comprehended the scale of this challenge yet.
If the source of the flue gas is ~renewable (e.g. wood) and the energy generated is used usefully, this isn't a bad idea for carbon capture. It's much easier to capture the carbon technologically like this, and rely on broader capture (e.g. trees) to pull it out of the atmosphere. What really counts is what happens to the concentrated carbon being extracted - it needs to not go back into the atmosphere.
direct air capture has been demonstrated by a variety of methods, including the literally oldest human industrial chemical process, lime burning. i've done the calculations, it's not such a bfd, even with lime
I would be very curious to see that math that you solved the problem with, given that all of the scientists in the world working on this missed your solution.
as https://en.wikipedia.org/wiki/Direct_air_capture#Environment... explains, it is only 250 kWh/tonne CO₂, or 900 kJ/kg in SI units. To remove the ≈60 Gt/year of anthropogenic CO₂ currently being emitted and get us to carbon-neutral with direct air capture would consequently require a theoretical minimum of 1.7 terawatts, which is only about 10% of current world marketed energy consumption, and presumably about 5% of world marketed energy consumption 10 years from now. Kicking climate change into reverse would require a bit more than that, maybe double. Depending on the sorbent system, this energy can be solar thermal; it does not have to be electrical.
Existing direct air capture systems like Climeworks's do not closely approach the theoretical minimum. ...
Point source capture is of course much cheaper but it cannot get us to net negative CO₂ emissions.
https://news.ycombinator.com/item?id=29327736 describes the lime approach. it uses a huge amount of energy, 1600 kWh/tonne CO2 (5700 kJ/kg CO2), much larger than the theoretical minimum above and much more than Climeworks, so the purpose of explaining it is only to demonstrate that there's no practical scalability limit to direct air capture. you don't have to ramp up your production of triethanolamine or aluminum formate or whatever the fuck. those are just a matter of optimizing the energy efficiency
None of this explains how using 10% of total human energy production (at a bare minimum, using technology that doesn't exist) isn't a "BFD", or how in the world lime burning - a process that emits CO2 - is in any way a solution.
I have absolutely "read the research", spent months in training and worked in the industry, which is why I found your claim that you've solved it and that it's not a big deal to be so ridiculous.
direct air capture normally requires cycling a sorbent, as you would know if you had worked in the industry as anything other than a janitor. you heat it up, it emits CO2 (hopefully into your underground caverns or wherever the fuck you're sequestering the stuff), you cool it back down, and it starts absorbing CO2 (for example, from the atmosphere). that's the same way scrubber systems like diethanolamine, triethanolamine, and zeolites work. calcium carbonate/calcium oxide/calcium hydroxide is one such cycle, just a particularly energy-hungry and slow one. conceivably there might be other approaches that don't rely on thermal cycling (pressure cycling, some kind of electrochemical thing) but afaik nobody has demonstrated one. they won't repeal the fundamental entropic limit anyway so they'll only be marginally better than existing thermal-cycling systems, we can't expect an order of magnitude improvement
using 70 percent as much, or even several times more, energy than the humans currently produce is not a bfd because the amount of solar energy reaching earth is five orders of magnitude greater than current world marketed energy consumption. historically that was irrelevant because burning coal was cheaper than solar panels. now solar panels are cheaper and it's just a matter of scaling up production, which will happen unless people somehow run out of economically profitable ways to use energy. see my notes from 2008 at https://dercuano.github.io/notes/solar-economics.html for details
i note a total absence of any quantitative arguments in your comments thus far; instead they rely on handwaving and gee-whiz pop science claims of incredibleness and authority
You edited your comment after I posted, but it doesn't matter, you haven't talked about the topic at hand: atmospheric CO2, a pollutant that makes up 0.04% of the air.
All of your calculations assume an infinite stream of pure CO2 - which is why sorbent based DAC work great for flue gas. Making that work with atmospheric CO2 requires tremendous pretreatment, which is hard and expensive and not something that can be handwaved away. It is the fundamental problem.
no, that is not correct; all of the calculations i have made and cited assume 400 ppm atmospheric air, not pure co2. it wouldn't even be coherent to talk about a theoretical thermodynamic minimum energy requirement to capture co2 from pure co2, which is the '10% of total human energy production' you're talking about upthread
rather than handwaving it away i linked to the calculation of what that theoretical minimum is, provided my own calculation of a trivially feasible and scalable approach that takes less than seven times as much energy as that, and mentioned climeworks, which is one of several companies doing direct air capture today for less energy than that (but obviously more than the theoretical minimum)
the top of the article has a flow diagram of a lye-catalyzed version of the lime process i was describing, a chemistry that has been used in submarines, anesthesia, and scuba diving for about a century
Bro, you have not answered the question. How do you move and prepare the air? Think about the volume and weight of the gas that needs to be moved. See what adding fans and pretreatment does to your napkin science project and tell me how many times the total GDP of the world it costs.
There is a reason why all of science is searching for new methods for DAC right now - because what you're suggesting doesn't work at the cost and scale we need.
fans use an obviously insignificant amount of energy compared to burning lime (as you'd know if you'd ever done a process engineering calculation involving fans), air pretreatment is unnecessary for the methods i'm talking about (as i already explained), and 'all of science is searching for new methods for DAC' is the sort of nonsense that suggests you've never met a scientist in your life
the issue has only ever been that clean energy has historically been too expensive for terraforming. new methods for DAC may be economically important in a hypothetical competitive DAC market, and knocking off 50 percent or even 5 percent of the energy cost would mean a significant reduction in the absolute resources required, but they aren't going to improve on the minimum energy thermodynamically required, which is inherent to the 400ppm concentration
what's changed is that now we have a cheap source of carbon-free energy that scales to five orders of magnitude more energy than we need for this
ten years ago we didn't; without that the problem was basically unsolvable
Alright, man, good luck. I didn't see a single answer about how you're going to physically move and prepare 5 quadrillion tons of air or why that's not necessary, but I'll sleep tight knowing that you've cracked it.
Everybody else who's reading this, maybe look up what all of the proposed DAC plants in the world look like, notice they're pretty much just a gigantic wall of fans, and think about why that might be. It's because DAC requires moving a shit-ton of air, because 400ppm isn't a lot of parts per million, and that matters.
Or 14.9 trillion kilowatt hours, which at plausible retail prices of £0.10/kWh is approximately one and a half trillion pounds. Now, who's going to pay for that?
It’s not much on a global scale, given the nature of the problem it would be solving. And if it works we have a nice little renewable power infrastructure as a byproduct
The real problem is people and our near inability to coordinate at that scale
agreed, or even, as demonstrated in parts of this thread, people's inability to come to an agreement on widely-known facts that conflict with their ego defense
i'm not sure what to recommend. https://pubs.rsc.org/en/content/articlehtml/2022/ee/d1ee0352... is recent, open access, and seems to be well written, but i haven't finished reading it. it includes, among other things, a process flow diagram with theoretical numbers for the calcium loop which gives slightly lower numbers than my own calculations i linked upthread
i did read https://doi.org/10.1016/j.isci.2022.103990 which is also recent and open access. it is very poorly written and contains a lot of embarrassing factual errors; however, it has a number of specific and hopefully correct numbers about energy costs lifted from other publications which are hopefully more trustworthy. on the gripping hand, none of them that i've tracked down so far report operational energy consumption numbers from an actually built large-scale dac plant
none of these papers incorporate the awareness that energy costs have just gone through the first big drop in a century and a half, perhaps because that future is not yet widely distributed
and, as i said, you can probably use solar thermal for sorbent regeneration
note, though, that the number you're using is the theoretical minimum energy used. existing practical technologies, as i said, consume several times more energy than that. it is unlikely that we will ever closely approach this theoretical efficiency
there is indeed a real difficulty in the current human incentive structure where everybody hopes someone else will fix the problem so they don't have to bear the cost themselves. but that's not a question of engineering infeasibility due to resource limits; that's a question of politics
Yes we do, they don't work in all climates, but we do have a quite well-working approach:
Step 1: base-resistand swamp cooler structured packing.
Step 2: NaOH or KOH solution in water, at equilibrium with air (temperature& humidity).
Step 3: blow air sideways while running liquid down via gravity.
Step 4: regenerate liquid via salt metathesis reaction on column packed with Ca(OH)2.
Step 5: dry-distill the CO2 out of the CaCO3 in a lime kiln (carbon capture for those is easier than for a gas turbine, as they exhaust like around iirc 3~5x as much C02 for each oxygen you feed in, assuming you fuel both with the usual methane).
Step 6: inject the captured CO2 into old gas/oil formations.
Of course this would be better if you combine it with already-needed air handling, for example in central forced air HVAC installations where adjusting the concentration of the lye in step 2 through adding water or vacuum-distilling allows it to humidify/dehumidify the air and where the liquid/air contact can also easily do the heat exchanger task (that strongly basic water shouldn't support growth of any harmful micrograms, I think (for KOH it would typically be half the molarity of a saturated solution, for NaOH more like 80~90%)).
They use similar scrubbing but in dry columns (Ca(OH)2 beads soaked with NaOH solution, then left to drip and packed into canisters) for anesthesia rebreathers to scrub a patient's exhaled CO2, and similar lime/lye cycling in paper Mills (Kraft process; green liquor (sodium carbonate + sodium sulfite) to white liquor (sodium hydroxide+ sodium sulfite) recausticizing).
Assuming negligible cost of the initial wet air scrubbing (via fusing the air handling with HVAC and in the process even providing extra-low-CO2 air), assuming a methane-fired lime kiln, this overall prices should be able to compensates for around 2~4x as much fossil fuel burn in, say, vehicles, than the burn/consumption of the DAC plant.
And much of the process is quite amenable to using solar electricity (with about a day's worth of storage, but part might be able to use thermal storage in rock piles for pre-heating the kiln gas feed), especially as nothing really requires the lime kiln to work in the winter.
One adult tree captures 25kg per year. One transatlantic flight releases 500kg per passenger. Twenty trees, per person, per flight. Get serious.
IPCC projections for a "good" scenario still depend on magical BECCS processes that don't yet exist., not flue scrubbing, which I think if anything will be away to keep coal plants online for _longer_.
It's still valuable science, but don't get excited until there are major, major advancements in non-flue DAC/CCS.
You're failing to put things in perspective. There are 500 trees on earth for each human being. That would mean they can absorb 12500 kg per year.
Also, tree are NOT the largest natural CO2 sink, oceans absorb more.
Overall, the biosphere absorbs about 50% of total human emissions (25% by land plants and 25% by oceans). I think that once we drop the CO2 emissions enough, the nature will quickly scrub it out of the air, we don't need to get negative ourselves, just drop close to zero and nature should balance itself pretty quick.
It's a carbon _cycle_. We've spent a hundreds of years adding new carbon to that cycle as fast as our technology will allow, which is now very fast. It will remain in the cycle unless we sequester it back into the ground. Plants (for the most part) aren't CO2 limited, they're nitrogen limited.
The climate also isn't something that will just snap happily back into place. We're in a local minima. If we climb up this hill, we won't like what we find in the next valley.
Oceans absorbing CO2 results in acidification which is already harming the oceanic life cycle. We really want to tilt the balance away from the oceans, because the consequences of a dead ocean are dire.
At the power plant it is a matter of putting together technology that already exists. All of the elements such as amine strippers, pyrolysis, oxy-fuel combustion, etc. have been around since the 1970s. The literature has a "stopped clock" appearance in that people have talked about these things in the context of "clean coal" and biofuels for a long time without anything being built.
There are two problems with it.
One of them is that any kind of carbon capture is expensive and competes directly with a power plant (maybe the same power plant) without the CCS equipment. To win that competition, the operator of the power plant either needs to be fined for emitting CO₂ or has to be paid for capturing it, to the tune of $50-$100 per tonne.
Another is that the ecological accounting for biofuels is frequently not favorable. When you consider water consumption, effects of land use, and the effects of inputs, many biofuel schemes (such as Ethanol from corn in the US) seem to be a net negative.
The cheapest BECCS scheme is to capture CO₂ from the alcohol fermentation process, this has been implemented with great success
Ethanol from sugarcane in Brazil is much better from an ecological and economic perspective and it would cost something like $30 a tonne to capture CO₂ and inject it into a saline aquifer but politically it seems impossible given that most international bodies think that the real problem in Brazil is deforestation and that somehow promoting agriculture in Brazil is going to cause deforestation despite the fact that the Ethanol industry is mostly around São Paulo and nowhere near the rainforest.
In principle there might be some agreement that Brazil gets credits for BECCS and also gets credits for protecting the rainforest but the latter is somewhat nonsensical in that there is a mismatch in time between processes that emit CO₂ (say you spend 5 hours on a plane) and the need to protect a forest forever to sequester a finite amount of CO₂.
> but politically it seems impossible given that most international bodies think that the real problem in Brazil is deforestation and that somehow promoting agriculture in Brazil is going to cause deforestation despite the fact that the Ethanol industry is mostly around São Paulo and nowhere near the rainforest.
It's not like Bolsonaro's term hasn't proven that deforestation is a massive problem. No one is trusting Brazil on a geopolitical stage at all, particularly as it looks like Lula is heading off in a lame-duck term against parliament and regional governors being dominated by Bolsonaro allies, who are bought off by Big Ag.
agriculture has been the most contentious issue in world trade and it is the reason why the World Trade Organization has been deadlocked since 1999. A while back Brazil asked the question: "What industries can we lead the world in?" and one of the answers was meatpacking so they got behind
That is, JBS is to Brazil what Boeing or Apple Computer are to the US and it is by no means anomalous that it has a great deal of political influence.
When we tell them what to do with their land they react the same way we'd react if they said "It's unfair that we have to pay royalties for software like Microsoft Windows and Hollywood Movies, not to mention GMO seeds".
Most of the people in the US I talk to who rightly want to see a stop to deforestation in the Amazon are ignorant about agriculture in Brazil (e.g. they think that JBS is backwards, not the fiercely competitive corporation it is) and ignorant about agriculture in general and not particularly understanding of the process by which deforestation happens in Brazil.
It doesn't help that many questions are not well understood such as the relationship between land use in the area south of the Amazon and local climate changes that could cause the rainforest to recess northward.
There could be some compromise but so long as developed country NGOs are speaking to Brazilians in a patronizing way it isn't going to happen.
Why is sugarcane in Brazil any better than corn in the US? If they use petrochemical fertilizer, the problem is the same.
You cannot sequester a ton of CO2 for $30/tn anywhere in the world right now. You can't even do it for five times that. If what you say were true, I'd be able to clean up after my own existence right now, which I can't do. I can only buy scammy "credits" and "offsets".
The advantage of corn over other crops is that it flourishes in the presence of nitrogen fertilizer. Sugarcane is not so hungry.
The cost of capturing fermentation CO₂ is low because fermentation CO₂ is almost pure with very little nitrogen in it. Thus almost all of the cost is the cost of compressing the CO₂, pumping it, and injecting it underground which is about $30 a tonne. Even a few percent of nitrogen will cause the CO₂ to misbehave while pumping, so capturing CO₂ from a combustion stream requires some kind of separation which historically has been an aniline stripper or something like the rectisol process, but maybe it will be aluminum formate or something else in the future. The aniline stripper costs about $50 a tonne.
The trouble with capturing fermentation CO₂ is that it is not scalable. There is a certain amount of it produced and it isn't enough to "save the Earth" but it is a low hanging fruit and it would help in the process in validating sequestration, as there are all kinds of questions about the permanence and safety of saline aquifer injection. (Don't get me started about the CarbFix water sequestration project...)
As for why you can't buy any good carbon credits it is the proliferation of junk carbon credits at low prices that keeps good ones off the market. Another problem is that many schemes are using the CO₂ to produce more oil
which on one hand is a real market for the CO₂ (in Texas you can drill and get CO₂ in some places and since the 1980s they will pump it sideways and use it for enhanced oil recovery) but doing so seems to be more problem than solution and it ties the project economically to the up-and-down cycles of the oil industry.
This is quite interesting, thanks for sharing. I had written off ethanol because of how gunked up it got with the American corn industry and the stupid fact that Iowa votes first in the US primary elections. Maybe there's something to this.
We have loads of good ideas for scrubbing flue. We have no good ideas for atmospheric CO2.
420/1000000 isn't a lot of particles per million. Moving and concentrating those particles and grabbing them is going to require an _incredible amount of energy_ no matter how you swing it. People have not comprehended the scale of this challenge yet.