Turning CO2 into gasoline — A new way to waste energy

Last week, NYT climate Andy Revkin blogged about a federal laboratory that says it can take atmospheric carbon dioxide and turn it into gasoline:

One selling point with Los Alamos’s “Green Freedom” concept, and similar ones, is that reusing the carbon atoms in the captured CO2 molecules as a fuel ingredient avoids the need to find huge repositories for the greenhouse gas.

The only problem with that exciting statement is that it is almost certainly not true, a point I will come back to.

Now the NYT has published an article on the subject that also overhypes the technology:

There is, however, a major caveat that explains why no one has built a carbon-dioxide-to-gasoline factory: it requires a great deal of energy.

To deal with that problem, the Los Alamos scientists say they have developed a number of innovations….

Even with those improvements, providing the energy to produce gasoline on a commercial scale — say, 750,000 gallons a day — would require a dedicated power plant, preferably a nuclear one, the scientists say.

Hmm. Let’s see. Problem one: Motor gasoline consumption in this country is almost 400 million gallons a day. So we would need more than 500 nuclear power plants … just in this country … and just for gasoline (you’d have to more than double that to displace all the other petroleum products we consume, like diesel fuel). And that would probably require another 5 Yucca mountains just to store the waste, although I’m not sure the word “another” is right ’cause this country can’t even agree on one friggin’ storage site in the middle of nowhere.

Problem two: According to the Los Alamos “Overview of Green Freedom,” each 750,000 gallon a day plant (with accompanying nuclear reactor) costs $5 billion. So cutting under half of all petroleum use in this country would cost over $2.5 trillion (not counting this cost of uranium or disposal)!

This supposedly yields a gasoline price of $4.60 a gallon, though the authors say with a couple more technological breakthroughs, that could drop to $3.40. How about if instead of assuming more breakthroughs, which hardly ever happen in the energy sector, we apply Romm’s Rule of Costs for Future Energy Sources.” Romm’s Rule says that for any new energy technology that is not yet commercial (and in this case we have a “concept” for which the patent was still pending in November), take the inventor’s highest projected cost and double it. Also flip a coin and if it comes up heads, the technology will never be commercialized — think fusion. And that’s generous — in reality, if the coin comes up head or tails (i.e. doesn’t land and balance on its edge) it will probably never be commercialized — remember the fuel cell was invented in 1839 and commercial fuel cells are just a tad more common than time machines. [Please note this rule does NOT apply to technologies that are already commercial.]

Problem three: Romm’s Rule of Energy Transformation. This rule, developed for analyzing hydrogen cars, says: You can probably make a sow’s ear from a silk purse if you try hard enough, but why would you do that? Zero-carbon electricity is arguably the most premium energy carrier you can make in a carbon-constrained world in part because electric motors are so efficient. Electricity can directly run a motor to move your electric car or plug in hybrid for under $1.00 a gallon, even using expensive nuclear power. You lose maybe one-fifth of the original electricity in the process. The entire Green Freedom process is so inefficient it probably throws away more than three-fourths of the original nuclear power (if not much more). Basically, after spending all that money and wasting all that premier power you are stuck with a low-grade (but conventional) fuel that has to be run through an inefficient gasoline motor. Why would you do that?

[Yes, we don’t quite yet have commercial plug ins, but they are straightforward extension of already commercial hybrids, we don’t need any technology breakthroughs, and multiple manufactures will almost certainly be selling them within three to five years. EVs will be common in other countries within the same time frame, as I’ve written. All of this will happen decades before “Green Freedom,” assuming it even proves feasible.]

Before coming to the last problem, let me complain about the NYT article, which, while skipping happily over the myriad problems with Green Freedom, bizarrely says of other alt fuels:

Hydrogen-powered cars emit no carbon dioxide, but producing hydrogen, by splitting water or some other chemical reaction, requires copious energy, and if that energy comes from coal-fired power plants, then the problem has not been solved. Hydrogen is also harder to store and move than gasoline and would require an overhaul of the world’s energy infrastructure.

Electric cars also push the carbon dioxide problem to the power plant. And electric cars have typically been limited to a range of tens of miles as opposed to the hundreds of miles that can be driven on a tank of gas.

Yes, if the energy comes from coal, neither hydrogen or electric cars make sense. But the same exact thing can be said of Green Freedom: It makes no sense if you use coal plants, but the NYT never mentions that fact. That’s why the Los Alamos inventors go the nuclear route. But if you can assume, say, 500 nuclear plants for Green Freedom, surely you can live with maybe 100 nukes for electric cars, which brings us to ….

Problem four: We are going to need a vast quantity of zero-carbon electricity in this country just to reduce emissions 80% in the electricity sector while supporting population growth and increased living standards. In the very unlikely event we would build 500 nukes and 5 Yucca-sized storage sites (and find the necessary uranium, given that, presumably, ever other country is going to be doing the same thing) to make carbon-neutral gasoline, it is safe to say that’s all the nuclear power plants we will be building this century. So the electricity will have to come from renewable power and … yes, coal power with carbon capture and storage (CCS). And if we keep dawdling, we are going to overshoot the safe level of carbon dioxide concentrations, and need to pull carbon out of the air and then NOT burn it. So Green Freedom does not save us from “the need to find huge repositories for the greenhouse gas.” If coal with CCS is practical and affordable, which strikes me as much more likely than Green Freedom achieving both of those goals, we are going to be doing as much CCS as we possibly can.

The scale of the climate solution is enormous — so great you would never contemplate it if the result of doing nothing were not so irreversibly catastrophic. That’s why, in spite of all these problems, in spite of the fact that it is exceedingly unlikely we would choose to use much Green Freedom by 2050 even in the equally unlikely event it is actually feasible on a large-scale in that time, it probably deserves some exploratory funding. Although that assumes we have a President who triples federal clean tech funding — no I wouldn’t waste much money on it at current federal R&D levels. It certainly shouldn’t be hyped by the media (or anyone else — this means you, Roger Pielke, Jr.), it is certainly no silver bullet, it probably isn’t even one of the ten silver bullets we need — but we can’t afford to ignore any solution that has even some microscopic chance of working.

[Tip o’ that hat to Earl K.]

28 Responses to Turning CO2 into gasoline — A new way to waste energy

  1. David B. Benson says:

    No Joe, it doesn’t deserve exploratory funding. Not even a dribble or a drab. Use the $$ on something which would be immediately useful.

    Anthracite biocoal comes to mind. Current South Africa spot price for metalurgical anthracitic coal is an astounding $(US) 275 per tonne. At that price it certainly pays to run a hydrothermal carbonization plant to produce very high grade (only kind hydrothermal carbonization makes) bio-anthracite.

  2. Steven Kimball says:

    Hello Dr. Romm,

    I had just finished posting a similar, albeit less detailed and less erudite, comment at Daily Kos in response to a diarist who was puffing the New York Times piece, when I noticed that your post. I’m beginning to think that all my reading is paying off.

    Throwing modesty to the wind, I’ve copied my comment below.

    I have a lot of doubts about this proposal.

    According to the Times’ article:

    “Even with those improvements, providing the energy to produce gasoline on a commercial scale — say, 750,000 gallons a day — would require a dedicated power plant, preferably a nuclear one, the scientists say.

    According to their analysis, their concept, which would cost about $5 billion to build…”

    Let’s do some quick back-of-the-envelope calculations. The US uses 388.6 million gallons of gasoline per day. 388,600,000 divided by 750,000 gallons of gasoline gives 518 as the required number of processing facilities and corresponding nuclear power plants. At 5 billion a plant, the figure given in the article, the cost would be 2.59 trillion dollars.

    A couple of questions:

    First, assuming that we want to spend enormous sums building more nuclear power plants, which is, I believe, a questionable assumption, why should we bother to make gasoline instead of using the electrical power directly to replace coal-fired power plants? After all burning coal produces far more greenhouse gases than burning petroleum does.

    Second, if this process really can remove carbon from the air why would we put it back in to the air? Shouldn’t we sequester it instead and use either renewable energy sources or even more nuclear to generate electricity and power our vehicles?

  3. Information says:

    The idea of storing renewable or nuclear energy by recycling CO2 with water is a good one which may be better than biofuels in terms of land use and environmental impacts, and better than hydrogen in terms of distribution and end use. We should use as much electricity directly as possible (e.g. in plug-in hybrid vehicles), but when we need dense chemical fuels, this is a valid option.

    The Los Alamos group appear to be “proposing” a process that was studied in the 1970s by Steinberg and several times thereafter by Bandi 1995, Stucki 1995, etc. Capture of CO2 from the atmosphere using the same chemistry has been studied ( , ). Electrolyzing the absorbent, whether it is carbonate or bicarbonate after CO2 absorption, has patents dating back to the 1960s and studied by the previously mentioned workers. They admit that it has all been studied but they are looking at how to piece together the parts in more detail than before. Unfortunately then, this is not a breakthrough and does not warrant the level of hype it is receiving.

    There are a few research groups in the world working on the same goals with new, different processes that haven’t been demonstrated in the 1970s. The major energy consuming step is the decomposition of H2O and/or CO2 (which Steinberg and the Los Alamos guys partially integrate this with the CO2 air capture absorbent regeneration step).

    Solid oxide electrolysis is one very promising, high efficiency and likely economical way:

    And thermochemical decomposition is another:

    And thermolytic decomposition is another:

    Then people are also working on methods that split water and/or CO2 using sunlight more directly (photoelectrochemical and photolytic methods).

  4. David B. Benson says:

    Lots of other solutions which appear to be less costly. Anyway, climate problems made the NAE cut for ‘Grand Challenges’:

  5. Ronald says:

    The article talks about using nuclear power plants and turning CO2 into gasoline. I’m curious how this process is different and the energy input/output compared to the
    1) production of hydrogen (for use in fuel cells) or
    2) the production of NH3 (ammonia) for the use as a fertilizer.

    I’ve been to a wind power conference in North Dakota and one word is the big 800 pound elephant in the room, Stranded. The wind velocities and power potential is great in North Dakota, but it’s stranded energy, not enough power transmission lines to bring it to the places to sell the electricity.

    What many in wind areas are looking for is an industrial process that needs electricity and can be done on the schedules that wind allows, more at night and in winter and intermittent. The authors of this article talks about nuclear power because of it’s non carbon, but apparently wind power producing electricity would also work. North Dakota does have the coal plants to use as the CO2 feed.

    Some people are looking at making NH3 (ammonia) from the wind energy produced, but I’m sure that process is something like this process to make gasoline and uses quite a lot of electricity to get back much less product, NH3. Making the fertilizer NH3 would be better (cheaper) done the old fashioned way, with natural (methane) gas.

    So I suppose, making H2, gasoline or NH3 with electricity is very energy inefficient.

  6. John Mashey says:

    I can understand why we want to store wind/solar energy, but I’m not sure why we’d want to “store” nuclear (i.e., good for baseload)-generated electricity … although shifting more and more Los Alamos researchers to thinking about sustainably problems is a good idea, unless it automatically leads to “more reactors”.

    Algae & plants do a good job of using CO2 at low concentrations, which is what we have. IF I had a wish, it would be to develop more uses of plants like switchgrass for non-fuel uses so the carbon stays sequestered, i.e., like insulation. was interesting.

  7. Paul K says:

    Gasoline from CO2 is a waste because gasoline is on the way out for automobiles. We don’t yet have the plug-ins or switchgrass ethanol that are soon expected. Major car companies and many micro shops are actively pursing EV. What is the private and public involvement in switchgrass? Is there any venture capital in it. Are there labs studying it. Joe, how about a post on switchgrass?

    I too have marveled at the magnificent wind turbines in the Dakotas and across the Plains. Those who have opposed offshore wind farms for aesthetic reasons have a different sense of beauty. I wondered why some of the turbines didn’t spin. Sometimes it is maintenance, but more often it is unneeded capacity. There is nowhere for the power to go. Stranded energy also hinders large scale desert solar. We cannot move industries and people closer to the power source. The changeover to alternatives requires sufficiently efficient 1,000 mile power lines. I believe on the West Coast huge amounts of hydro-electric power are sent 1,200 to 1,500 miles using direct current. Is there an electrical engineer in the house?

  8. Hans says:

    re: “Electricity can directly run a motor to move your electric car or plug in hybrid for under $1.00 a gallon”

    Can you clarify what that even means? If you have an electric car then you have no “gallons”, just batteries. Do you mean that with an electric car you can get the distance equivalent of 1 gallon’s worth of travel on a conventional car, (eg 25 miles), on one dollar’s worth of electricity?

  9. Beefeater says:

    Being a non-scientist myself, and having lived through the oil crunch of the early 70’s, I think that I’ll stick to my cow magnets and water injector until something else comes along. They don’t work either, but at least they’re cheap.

  10. Ronald says:

    One thing about that North Dakota wind conference that I went to, both North Dakota U.S. Senators spoke at the thing. I suppose when you are a state with 600 000? people, you get to see your U.S. Senators a lot more than states like New York or California.

  11. Joe says:

    Hans — yes, the per mile cost running on electricity is about one quarter that of running on gasoline.

  12. Lou Grinzo says:

    I thought I’d seen the most ridiculous potential application possible for nuclear power when I read about the proposals to build nukes to power Canadian tar sands operations. But I was wrong. This item wins by a neutron.

  13. Benjamin says:

    While there are many things that can and should be done to reduce our dependence on gasoline, it will be many, many decades until we need no gasoline at all. First of all, trains and cars can be electrified, but airplanes and ships will probably always need another fuel source. Second, if this project is as expensive as it seems, then why not develop it for niche applications, like classic cars or emergency generators? Sure, gas might be $6 a gallon, but it’ll be worth it to someone. Meanwhile, the rest of us will be driving electric cars (or not driving at all!), and everyone will be carbon neutral.

  14. Earl Killian says:

    Ronald, I’ll try to answer your question. The “future efficiency goal” for making hydrogen from water using electricity is 50kWh per kg of H2 at 6000 psi. They aren’t there yet as far as I know; there is one commercial device that can produce 6000 psi at 60kWh/kg. The goal of the FreedomCar program is a fuel cell that can turn H2 back into electricity again at 20kWh/kg. They aren’t there yet either. My guess is the new Honda FCX gets less than 14kWh/kg. Anyway, just using the goals, one can see that the hydrogen world is hoping to get back 40% (20 divided by 50) of the electricity they put into making H2 from renewable electricity. Today they are at less than 25%. Delivering electricity over the grid and charging batteries is much more efficient, at about 80%.

    The Los Alamos paper doesn’t give all the details, but it looks like 55% or so of the electricity energy ends up in the gasoline (there is also heat energy that is required as an input that I am not counting here). Since your typical ICE is 21% efficient, you’re getting about 12% of the energy out of the engine than the electricity that went in. The Prius is better, and using it instead of a typical ICE would give you about 19% of the starting energy back.

    One way to think about this, if we need 40,000 sq.mi. of wind turbines to power our 2050 passenger vehicles directly with electricity, then you will need 80,000 sq.mi. of wind turbines if you want to do it with hydrogen (assuming they achieve their goals), 160,000 sq. mi. if you want to do it with gasoline Priuses, and 320,000 sq.mi. if you want to do it with non-hybrid gasoline cars. I prefer the 40,000 sq.mi. solution.

    The cost of fueling vehicles is higher than the land areas above would indicate since the H2 production plant or gasoline production plant have capital costs that must be depreciated and included. My estimate is that the per-mile cost of driving directly on renewable electricity is 2.4 cents a mile, hydrogen from renewable electricity about 7.5 cents a mile, gasoline from renewable electricity about 19 cents a mile (using 35 MPG from the new CAFE standards).

  15. John says:

    Using this process to produce gasoline for automobiles is stupid, I agree. But what about fuel for airplanes? Ships? Even large trucks seem unlikely candidates for battery-based propulsion. There’s a lot more fuel being used out there than just for the evil SUVs.

  16. David B. Benson says:

    Transimitting large amounts of electric power long distances is quite efficient: the losses are about 7% per 1000 km, plus or minus depending on many factors.

    However, this requires obtaining right-of-way and some moderately expensive construction of transmission line towers.

    There is indeed an HVDC line from the Pacific Northwest to Southern California and at least one more back east somewhere.

    Fuel for airplanes, ships and large trucks can be produced from biomass, fairly inexpensively. For reasons I don’t understand, the initial tests of biofuel for airplanes use a 50% biofuel, 50% fossil fuel mixture.

  17. DOE is apparently happily conflating two problems here: carbon free energy and jet fuel.

    Nuclear is a serious competitor in the first category.

    In the second category, carbon from somewhere and energy from somewhere will be combined to create the jet fuel. Air travel is not going away and there simply isn’t any other energy source light enough to fly around with.

    The idea of getting automobile fuel this way instead of making a big push to convert to electric cars (not to mention shared transport, local services and telecommuting) seems ridiculous to me.

    If there is some reason these two issues are tightly linked so far I haven’t seen it. It seems to make sense as a usage for episodic wind and solar as John Mashey points out.

    However, this does have potential for solving the aviation problem and should not be sneered at. I’m not sure that it’s needed for land freight; reinventing rail might be more effective. Since rail pretty never took note of the 20th century, a pretty substantial leap seems plausible directly from the 19th to the 21st.

    If that’s right, I guess the trucking industry will have to go away.

  18. Peter Foley says:

    David B. Benson, They use the legacy jet fuel(kerosene) to allow the use of legacy gaskets and seals that rely on the aromatic components to “swell” the gaskets and protect various metal combustion chamber parts with the soot from the impurities in the ‘natural’ kerosene. It is really embarrassing when the rubber bands(jet engines) break at 35000 feet Similar types of issues exist with the use of sulfur free diesel–various addtives are used to keep legacy engines running with out damage from very low lubricity low sulfur diesel.
    If you think we’ve actually reached natural peak oil, how can any rational actor not support replacement fuels that don’t add to alleged carbon AGW?
    Joe, don’t forget that large percentage of present Nuke waste is from the Cold War and current bomb arsenal. Thus 518 plants wouldn’t require any where near five ‘Yucca’s.

  19. Marcel F. Williams says:

    This is an idea that I’ve been promoting for more than a decade. Using nuclear energy to produce synthetic industrial chemicals and transportation fuels such as gasoline, diesel fuel, aviation fuel, methane, methanol, and ammonia is the only solution to the world’s problem with greenhouse gases and their energy needs.

    And don’t worry about building more Yucca Mt. sites, building thousands of nuclear power plants would require practically all of the plutonium from spent fuel to be reprocessed and converted into more energy. If anything, using nuclear power to replace transportation fuel would probably produce a severe shortage of plutonium.

    Plug-in-hybrids cars would probably be the perfect vehicles for a nuclear economy since they would make even the most expensive nuclear gasoline affordable while also utilizing the much cheaper nuclear electicity. However, the mass production of thousands of nuclear power plants, hopefully, in remote centralized locations, should dramatically reduce the capital and labor cost of nuclear power plants which should also significantly reduce the cost of gasoline produced from nuclear power facilities.

  20. David B. Benson says:

    Peter Foley — Thanks for the explanation.

  21. Glen says:

    This whole thing seems to be predicated upon the idea that it is carbon neutral only because it uses as it’s source carbon that is already in the atmosphere.

    Isn’t this exactly what is supposed to be the idea behind cellulosic ethanol? When a gallon of ethanol is burned, it puts 30% less greenhouse gas into the air, but it also gets you 30% less distance down the road. The only reason that it is thought to be carbon neutral is that it uses an equal amount of carbon from the environment in order to produce it.

    This scheme doesn’t seem to be doing anything that the environment isn’t already doing for you when it produces the feedstock for ethanol.

    What gives?


  22. M.R. says:

    Hello all,

    Don’t forget the potential for hydropower.
    See these links:



  23. Matus1976 says:

    Do your criticisms adequately take into account that batteries are terrible for energy storage density (~300 Wh/kg) and completely impractical for long distance travel of vehicles? And additionally, no civilian aircraft will be battery powered any time soon, let alone military vehicles. Hyrdogen is great for gravimetric energy storage density, but it’s volumetric storage density is terrible.

    To get adequate ranges and storage densities both Hydrogen and Batteries are impractical, so everyone is racing to find the best “hydrogen binding” medium, from metalic matrixes to buckyballs. But we all ready have a great hydrogen binding medium, it’s called carbon. The fuel economy of the future will need to include a liquid hydrocarbon at room temperate, whether that is Methonal or Gasoline, etc. Nothing compares for energy storage densities, Methonal fuel would enable much more efficient than ICE’s – Direct Methonal Fuel Cells and would allow us to keep our existing infrastructure. That methonal or similiar liquid hydrocarbon would necessarily require the manufacture of synthetic fuel via the application of energy in some form to Water for the Hydrogen and CO2 for the Carbon.

    The “Green Freedom” process doesnt sound very practical, but some form of synthetic liquid hydrocarbon or similiar fuel will ultimately be the only thing which makes any sense, where carbon functions only as a temporary carrier.

  24. Joe says:

    Batteries aren’t terrible for the kind of energy storage needed by plug in hybrids. Hydrogen is going nowhere. Biofuels will run planes and long-distance travel, when we stop using petrol.

  25. Dan says:

    The thing that needs to be considered, especially with any hydrocarbons, is that it must be clean energy, in the sense that pollutants are accounted for in the process. If you ignore the by-products, then you’re ignoring the cost of dealing with those by-products. Sending carbon into the air is an example of this. If you don’t build in the cost of then you’re just postponing the cleanup costs untill later and in many cases pushing costs off into other industries such as water sourcing and healthcare.

    When you consider various fuels, it isn’t an accurate comparison to compare gas and hydrogen without dealing with the handling of the byproducts. Hydrogent has some greater efficiencies in that regard that I don’t see being discussed in these comparisons.

  26. craig t says:

    why not convert electric dams uhmmm lets say somthing like the hover.

  27. RC says:

    Please correct your the post in the post by:

    Information Says:
    February 19th, 2008 at 9:17 pm

    The link is not correctly hyper linked the link is only being created up to 7480. The complete link should be

  28. i guess that from the first response that we should just give up and burn at the stake all those who propose such a idea, remember you first have to start a with general idea and then move on down to get it right!