07 January 2013

Air Capture Update 2013: Still Progressing

One of the most tantalizing possibilities for dealing with the accumulation of carbon dioxide in the atmosphere is to take it out using brute force methods of chemistry, biology or even geology. I discuss such "air capture" of carbon dioxide in Chapter 5 of The Climate Fix. As the climate debate continues to generate more heat than light, technologies of air capture are continuing to improve.

On Saturday, the New York Times had a smart article with an update on the technology:
[A] Canadian company has developed a cleansing technology that may one day capture and remove some of this heat-trapping gas directly from the sky. And it is even possible that the gas could then be sold for industrial use.

Carbon Engineering, formed in 2009 with $3.5 million from Bill Gates and others, created prototypes for parts of its cleanup system in 2011 and 2012 at its plant in Calgary, Alberta. The company, which recently closed a $3 million second round of financing, plans to build a complete pilot plant by the end of 2014 for capturing carbon dioxide from the atmosphere, said David Keith, its president and a Harvard professor who has long been interested in climate issues.

The carbon-capturing tools that Carbon Engineering and other companies are designing have made great strides in the last two years, said Timothy A. Fox, head of energy and environment at the Institution of Mechanical Engineers in London.

“The technology has moved from a position where people talked about the potential and possibilities to a point where people like David Keith are testing prototype components and producing quite detailed designs and engineering plans,” Dr. Fox said. “Carbon Engineering is the leading contender in this field at this moment for putting an industrial-scale machine together and getting it working.”
A crucial question of course is cost. In a 2009 paper on air capture (here in PDF), I compared the idealized costs of using air capture as the main mechanism to achieve stabilization of carbon dioxide concentrations, based on existing cost estimates of the technology, with the costs of stabilization under conventional mitigation policies, as estimated by the IPCC and Stern Review.

In that analysis I found that under the identical assumptions used by IPCC/Stern, air capture which cost between $100 and $500 per tonne of carbon would lead to overall economic costs of 0.5% to 3.0% of overall GDP to achieve stabilization at 450 ppm. This was a surprising result, because for conventional mitigation Stern estimated the costs at up to 4% of GDP and IPCC up to 5.5%. Critics of my paper complained that air capture was not yet possible. My reply was that neither was conventional mitigation possible, so why not pursue both?

A more recent literature review of air capture technology and economics by Goeppert et al (2012, here, $) found estimated costs of $50 to $3,700 per tonne of carbon. They conclude:
Direct CO2 capture from the air is still in its infancy. The cost of a commercial plant will depend on many factors including the process used as well as the cost of labor, materials and energy. While there is no question that the capture of CO2 from the air is possible, more research and development is clearly needed to optimize this technology and determine its economic viability. Only with the construction of demonstration and pilot plants will we have a clearer understanding of the total cost associated with DAC. A few start-up companies including Carbon Engineering, Kilimanjaro Energy, Global Thermostat and Climeworks have started such an effort. Some of the proposed devices and prototypes for the capture of CO2 from the air are shown in Fig. 14. It should also be pointed out that the costs associated with DAC units are not ‘‘stand alone’’. Once captured, the CO2 will be used for applications such as enhanced oil recovery (EOR) or recycling into chemicals and fuels including methanol, DME and hydrocarbons (CCR). This will give an economic value to the captured CO2, lowering the de facto cost of DAC and provide a more favorable overall picture of the process. Water (moisture) could also be separated from the air at the same time as CO2, which could provide clean water as an added value. . .

CO2 from the atmosphere provides a nearly inexhaustible carbon source for humankind. Combining carbon capture and storage with subsequent withdrawal for technological recycling based on an anthropogenic chemical carbon cycle offers a feasible new solution to our carbon conundrum. As fossil fuels are becoming scarcer and increasingly depleted, carbon capture and recycling offers a renewable and safe source for carbon containing fuels and their products. It also liberates humankind from the limitations associated with the biological natural carbon sources including crops and biomass. The chemical carbon cycle constitutes humankind’s practical technological analog of nature’s photosynthetic CO2 recycling. At the same time, the anthropogenic chemical carbon cycle (CCR) also helps to mitigate the environmental harmful effect of excessive CO2 in the atmosphere. Instead of just a greenhouse gas harmful to the Planet’s ecosystem, CO2 should therefore be considered as a valuable industrial C1-feedstock.
Ultimately, the test of air capture will not come from journal articles or policy debates, but actual engineering in the real world. Of the actual costs, Goeppert tells the NYT, "We won’t know for sure until someone builds a pilot plant."

The good news is that far from the glare of the climate debate, scientists and engineers are hard at work on advancing air capture technology. And guess what? They are making progress. Watch this space.

References cited

A. Goeppert et al. 2012. Air as the renewable carbon source of the future: an overview of CO2 capture from the atmosphere, Energy & Environmental Science, 5:7833-7853 DOI: 10.1039/C2EE21586A

R. A. Pielke, Jr. 2009. An idealized assessment of the economics of air capture of carbon dioxide in mitigation policy. Environmental Science & Policy 3:216-225

31 comments:

Eric said...

"CO2 from the atmosphere provides a nearly inexhaustible carbon source for humankind."

Yes, but we need energy (non-carbon energy!) to make such ideas practical and to benefit from CO2 as a chemical feedstock. And that's sort of how we got in this mess in the first place. The only obvious solution? Nuclear power, and lots of it!

Mark B. said...

"Direct CO2 capture from the air is still in its infancy."

No - direct CO2 capture has yet to prove it will ever grow beyond the fetal stage. Infancy implies the known progression to child and adult. Only a naive belief in progress says otherwise. As to only knowing the cost after the plant is built - I call BS on that one. Whatever the final cost of a sow's ear to sink purse factory, I know that it would cost a lot more than just raising silkworms, making silk thread and weaving silk cloth.

John M said...

"...estimated costs of $50 to $3,700 per tonne of carbon..."

Is that a typo or are they taking precision lessons from climate scientists?

Mark Bahner said...

It's very difficult to see how ambient capture will ever be less expensive than capture of CO2 from power plant exhausts, particularly natural gas power plants, which are essentially free of particulate and sulfur dioxide.

Therefore, unless or until most gas-fired power plants have carbon capture, I don't expect to see ambient air capture.

But what ambient air capture does is to render false all claims of multi-century substantial damages from CO2, such as those made by Nicholas Stern, Joseph Romm, Ken Caldeira, and others.

There would be no reason to endure centuries of substantial damages, when for a fraction of the cost of the claimed damages, future generations could reduce the ambient concentration of CO2 to the pre-industrial level of 280 ppm.

Les Johnson said...

We purchase liquid CO2, which is a byproduct of liquefying N2 and O2. Its been a few years since I was in that side of the business, but we used to pay 50 to 60 dollars per tonne of CO2, FOB the plant.

Air liquefaction is an energy intensive business, so I could see energy sources like wind and solar being used as the primary power source. Assuming it becomes competitive with fossil fuels, of course.

HowardW said...

I found this telling from your second reference: "many advocates of greenhouse mitigation support aggressive action not simply because of the direct climate benefits, but also because of the ancillary benefits, including ‘‘limiting the aggregate scale of human population and economic activity’’ (Parson, 2006)."

Roddy said...

As a layman, my presumption was the same as Mark #4 - CC from power or industrial source surely can only always be cheaper than air capture. With storage the same; transport I guess could differ.

CCS in the EU is keeping a VERY low profile given it is a plank of energy and emissions policy.

Mark Bahner said...

"We purchase liquid CO2, which is a byproduct of liquefying N2 and O2. Its been a few years since I was in that side of the business, but we used to pay 50 to 60 dollars per tonne of CO2, FOB the plant."

Interesting! The question in my mind is, "Does removing CO2 from ambient air in that manner produce more than one tonne of CO2 from fossil fuels per tonne of CO2 removed from the atmosphere?"

MIKE MCHENRY said...

The Times article indicates this company uses sodium hydroxide (lye) to a capture the CO2. Sodium hydroxide is made via the chloralkali process. This involves the electrolysis of sodium chloride (salt)solutions. This requires electricity which comes from....

n.n said...

Eric:

Yes, nuclear technology is the only technology suitable to be exploited as a primary energy producer. It can be reasonably isolated from the environment (i.e. independently reliable), and is capable of high density energy production (i.e. it does not commit large tracts of land to a single use).

In general we should assess the value of each technology as it is suitable for an application. Oil, as a high-density chemical energy, is ideally suited for mobile applications. Others, including solar, wind, etc., are well suited for applications constrained by geography, reliability, etc.

This is a systems problem. We need to characterize the system, identify its optimal function, and model it appropriately.

The characterization and subsequent model need to be objective within the constraints imposed and assigned to the system.

The Right Wing Professor... said...

I have to say I'm far less impressed than you with the progress. The statement that air capture requires only 2 - 4 times more energy than flue gas capture (I haven't run the math, but that's not unreasonable) is pretty damning, considering we don't have an operational flue-gas capture system yet. Aside from the intimidating thermodynamics, any capture system would have to run hundreds or more likely thousands of cycles just to recover the CO2 produced in manufacturing it. Things tend to wear out. And it would have to be huge to make any impact at all on the atmosphere, and require huge amounts of energy.

Reviews of the sort you cite are written by workers/enthusiasts in the field who have a strong vested interest in maintaining or better yet increasing funding for the field. I wouldn't credit the optimism they evince with any objectivity whatsoever. Example: I've read dozens of papers and grant proposals on solar energy conversion, and they're all unanimous that we're just one funding cycle away from developing an economic method. There have probably been literally thousands of such proposals written. yet it still isn't close to being economically competitive.

When a scientist lies about methods or results, it's rightly called fraud. Yet we routinely lie far more egregiously in research proposals about the prospects for our work, and it's simply the norm.

An honest proposal would say that carbon sequestration, particularly from the atmosphere, is an intimidatingly difficult problem, may never be solved in a practical way, and at best will take many years of research. However, given the non-zero probability that it will work, and the significant probability we will have no better alternative, we ought to check it out. Beside, the research itself is fairly cheap.

Not exactly an easy sell, but not dishonest.

The Right Wing Professor... said...

Interesting! The question in my mind is, "Does removing CO2 from ambient air in that manner produce more than one tonne of CO2 from fossil fuels per tonne of CO2 removed from the atmosphere?"

In principle, it doesn't have to. I ran the math a while back, and even if you make electricity by burning coal in a old fashioned power plant, you can in principle capture more CO2 from the atmosphere than you put into it. But you have to be pretty efficient to do much better than break even.

It's more efficient to capture and sequester flue gas. Diluting CO2 from 100,000 ppm (flue gas) to 400 ppm (atmosphere) means a significant increase in free energy (about 14 kJ/mol CO2). Given you might get a bit better than 100 kJ/mol CO2 from a coal power plant, you don't want to be giving that up.

Mark Bahner said...

I wrote: "Interesting! The question in my mind is, 'Does removing CO2 from ambient air in that manner produce more than one tonne of CO2 from fossil fuels per tonne of CO2 removed from the atmosphere?'"

The Right Wing Professor (must be lonely ;-)) responds: "In principle, it doesn't have to."

Well, it does have to remove more CO2 from the atmosphere than it puts in, in order for it to be comparable to what Carbon Engineering is doing. Les Johnson said he used to pay $50 to $60 per tonne of liquified CO2. If the CO2 emitted to the atmosphere based on that process was less than one tonne of CO2 emitted per tonne of liquified CO2 produced, then we *already* have something that can remove CO2 from the air at a cost of approximately $50 to $60 per tonne of liquified CO2.

"I ran the math a while back, and even if you make electricity by burning coal in a old fashioned power plant, you can in principle capture more CO2 from the atmosphere than you put into it. But you have to be pretty efficient to do much better than break even."

Yes, I'm wondering if the process that Les Johnson was describing is better than breakeven. I suspect it's not. If it were better than breakeven, I don't think Carbon Engineering wouldn't get much publicity.


Les Johnson said...

All: The process I described in an incidental one. Liquefying N2 and O2 will also liquefy CO2.

In terms of energy use, I would have doubted that it "breaks even" in terms of CO2, since such a large volume of air would have to captured and chilled. But, the following suggests that about 10% of generated power would be needed to liquefy and transport CO2.

https://docs.google.com/viewer?a=v&q=cache:Y073jDictrYJ:www.xdos.co.jp/CCSJp.pdf+&hl=en&gl=ca&pid=bl&srcid=ADGEESjRy5bjOpGJ9YSQTFPp-4WxATDg1Xxp5aB7jXtkPMWFKN9aqt_ucOh_j0Q3KjeBrffygC8pLUhI6Qmh8ZDktT3hgXDOVyvX3DGhZZ_4Xq32xmswegLjib_gnN4npx5fO1_WfT0n&sig=AHIEtbSyJS0hy5X8BXQxNRRcP54BJESs3w

We were not concerned with CO2 as a GHG, but for industrial purposes.

I note that this paper also uses a 50/tonne figure, from about the same time frame I was using CO2.

Les Johnson said...

Hmmm...doing the calculations myself, the energy to liquefy CO2 is quite low, compared to the energy source.

One bbl of oil has about 1,450,000 kcal energy when burned. If we assume 36 kcal to liquefy one kg of CO2, then burning one bbl of oil at 100% efficiency (yeah, yeah) would be able to liquefy over 40,000 kg of CO2.

One bbl of oil produces 433 kg of CO2, which would require 15,577 kcal of energy to liquefy. This is a little over 1% of the energy in a bbl of oil.

I am amazed that the cost is so little (relatively).

The Right Wing Professor... said...

If we assume 36 kcal to liquefy one kg of CO2

To liquify CO2 at ambient temperature, you have to do two things. The first is concentrate it to 100% at 1 bar. If you assume flue gas is 10% CO2, that requires 6 kJ/mol or 130 kJ/kilogram at perfect efficiency. The second thing you have to do is compress it to a pressure at which it will liquify, which would be 64 bar at 25°C. That requires another 10 kJ/mol or 234 kJ/kg (assuming ideal gas behavior). The total is 364 kJ/kg, or 87 kcal/kg.

I get about 50 MJ/kg for combustion of hydrocarbons, generating about 3 kg of CO2. So you generate about 16 MJ of heat per kg CO2. At an optimistic conversion efficiency of 35% (ultimately limited by Carnot), that 's about 5.5 MJ useful work/kg CO2. So at perfect thermodynamic efficiency, you'll waste about 7% of your energy recovering the CO2 from flue gas. No current sequestration method is even close to perfect thermodynamic efficiency.

Please note all the above is dictated by the first and second laws, and it therefore not subject to engineering improvements. I've also ignored dozens of other complications (such as, for example, dealing with the water vapor in flue gas)

Tamara said...

What are the downwind effects of that big array - changes in air flow, low CO2/H2O concentrations? Hope they don't build them next to the windmills in the middle of farmland.

Les Johnson said...

yeah, that paper, and my calculations, were pretty optimistic.

Its still much lower than I expected.

They might be able to reduce costs using a gas filtration system, like a pressure surge adsorbtion, but I agree. Engineering will only do so much. Maybe 20% cost?

Mark Bahner said...

17-Tamara

"What are the downwind effects of that big array - changes in air flow, low CO2/H2O concentrations? Hope they don't build them next to the windmills in the middle of farmland."

Actually, I don't think farmers will be very happy to have the CO2 capture devices anywhere near them, regardless of whether the farmers have windmills on their land.

I'm pretty sure the "photo" that headlines the story is Photoshopped. In part, because there should be a lot of brown grass on the downwind side! Seriously.

The desire, in order to cut both capital and operating costs, would be to remove *all* CO2 from the atmosphere (all the way from its current value of about 400 ppm down all the way to near zero ppm)...rather than to remove CO2 down to 280 ppm, which is the pre-industrial atmospheric concentration. That's because if one only goes from 400 ppm to 280 ppm, that's only 120 ppm difference, whereas if one goes from 400 ppm down to zero, it's a 400 ppm difference. So if one goes from 400 ppm to 280 ppm, one has to treat 3.33 times as much air to get the same removal as if one goes from 400 ppm to 0 ppm. But the trouble is that plants *need* that 280 ppm to survive. So one could expect that crops downwind of one of these arrays would have a lot of problems.

John M said...

I'm a bit puzzled by Les Johnson's comments above about obtaining CO2 from air separation. My understanding of current industrial CO2 production is consistent with these two passages from Wikepedia:

"Carbon dioxide is mainly produced as an unrecovered side product of four technologies: combustion of fossil fuels, production of hydrogen by steam reforming, ammonia synthesis, and fermentation. It can be obtained by or from air distillation, however, this method is inefficient."

"Industrial carbon dioxide can be produced by several methods, many of which are practiced at various scales.[16] In its dominant route, carbon dioxide is produced as a side product of the industrial production of ammonia and hydrogen. These processes begin with the reaction of water and natural gas (mainly methane).[17]

Although carbon dioxide is not often recovered, carbon dioxide results from combustion of fossil fuels and wood as well fermentation of sugar in the brewing of beer, whisky and other alcoholic beverages. It also results from thermal decomposition of limestone, CaCO3, in the manufacture of lime (calcium oxide, CaO). It may be obtained directly from natural carbon dioxide springs, where it is produced by the action of acidified water on limestone or dolomite."

http://en.wikipedia.org/wiki/Carbon_dioxide

I would suspect that in the vast majority of air separation plants (if not all), the CO2 is removed by an adsorption bed and ultimately vented back into the atmosphere, since its value and concentration is so little that it's not worth bothering with.

Ben Harding said...

I'm going to give Roger the benefit of the doubt here (I have not read The Climate Fix) and interpret his advocacy for this method to be for its application in removing the _inventory_ of CO2 in the atmosphere ("...dealing with the accumulation..."). Of course, the first rule of holes is to stop digging, which, once you've decided to reduce CO2 concentrations, is where much more thermodynamically efficient flue gas recovery would come in (assuming emission reduction is more costly or has inadequate capacity).

But, then Roger goes and refers to air capture as "...the main mechanism to achieve stabilization..." What's up Roger? Is there something about thermo or econ I don't understand?

Les Johnson said...

John: No, in most air plants, CO2 is the first extracted, and sold, because of the realtively high temperature it liquefies.

CO2 is valuable in many industrial processes, and is sold as such.

While its concentration in air is low, enough N2 and O2 are needed for other industrial processes, that the CO2 produced as an incidental process is still valuable.

This any account for the relatively high price of CO2 vs N2, in spite of more energy needed for making liquid N2. A large demand, with a low supply (400 ppm).

John M said...

Les,

Are you sure you're not thinking of cryogenic gas treatment facilities that recover gases from oil fields?

Everything I see indicates that air separation plants aimed at serving the N2/O2/Ar markets purge the CO2 back into the atmosphere.

See here for example.

http://www.uigi.com/cryodist.html

If you have a technical paper or trade journal article that says otherwise, I'd love to see it.

eduardo said...

One point that is not clear to me is the logistic to store all that carbon. If we are able to split the oxygen and carbon in carbon dioxide molecule, we still would need to store the same amount of carbon by weight that we dig out - i.e. we would need to roughly duplicate the carbon mining and oil mining infrastructure to put the carbon again back into the Earth crust . If we dont split the carbon dioxide molecule , then the weight to be stored would be three time the weight we dig out.

Les Johnson said...

John: Yes, you are correct, and I was mistaken. I had assumed (always dangerous), since we bought the CO2 from the same producers we got N2 from, that it was the same process.

Looking into those CO2 suppliers, they all have CO2 plants attached to refineries, gas plants or fertilizer plants. Some, such as Ferrus, specifically say they are taking the waste stream.

My mistake, my apologies.

The Right Wing Professor... said...

Eduardo:

Most of the sequestration proposals involve pumping liquid CO2 back underground, or even to the bottom of the deep ocean, where the pressure will keep it as a liquid, which, as it happens, is denser than and immiscible with water. It would sit there in pools, slowly dissolving.

John M said...

Les,

Thanks, no problem.

Not often that someone admits a mistake on a blog.

I'll have to try it sometime to see how it feels. :)

eduardo said...

@26

Right Wing Prof.

Ok, you would need to liquefy the CO2, transport it to adequate places and sink it into the ocean. The amount by weight to be transported would be three times the amount that is currently shipped in form of coal and oil and gas. I think the required infrastructure would be quite substantial.

The Right Wing Professor... said...

Eduardo:

Indeed. As I said, there are lots of additional costs not factored into the simple thermodynamic analysis.

Mark Bahner said...

25- Les Johnson:

"John: Yes, you are correct, and I was mistaken. I had assumed (always dangerous), since we bought the CO2 from the same producers we got N2 from, that it was the same process.

Looking into those CO2 suppliers, they all have CO2 plants attached to refineries, gas plants or fertilizer plants. Some, such as Ferrus, specifically say they are taking the waste stream.

My mistake, my apologies."

27- John M.

"Les,

Thanks, no problem.

Not often that someone admits a mistake on a blog.

I'll have to try it sometime to see how it feels. :)"

:-)

I'll try that. Tomorrow. ;-)

Regardless of where the CO2 comes from, what interests me is the price. If the price was $50 to $60 per tonne FOB at the plant, that probably means that the costs were less to make that liquid CO2.

Les Johnson's cost number seems generally compatible with this estimate by Stanford University:

http://news.stanford.edu/news/2011/december/extracting-carbon-air-120911.html

"After a detailed comparison, the research team concluded that the cost of removal from air is likely to be on the order of $1,000 per ton of carbon dioxide, compared with $50 to $100 per ton for current power-plant scrubbers."

EliRabett said...

Flue gas is not a pure CO2 stream, and depending on what else is there capture in the flue may be neither easy nor cheap.

Post a Comment