The Environmental Panacea of the Hydrogen Economy Recedes Further


Both presidential candidates offer lip service to the environmental panacea of the 21st century: the hydrogen economy. "My plan calls for a hydrogen-based energy economy by 2020," said John Kerry in a speech to the League of Conservation Voters. In his 2003 State of the Union speech, even failed Texas oilman, President George W. Bush promised, "Tonight I'm proposing $1.2 billion in research funding so that America can lead the world in developing clean, hydrogen-powered automobiles."

However, British economist Andrew Oswald shows just how much of a mirage the hydrogen economy is. Oswald calculates that it would take at least 1 million windmills covering an area half the size of California or 1000 new nuclear power plants to produce the hydrogen needed just to run our vehicles. Since one-third of the energy we use is for transport that means that it would take 3 million windmills covering about 240,000 square miles or 3000 nuclear power plants to produce all the hydrogen we need to fuel our economy. (For comparison, all of the solid structures,including highways, buildings, parking lots, in the lower 48 states today cover just 43,480 square miles.)

But why waste energy making hydrogen anyway? My bet is that by mid-century, humanity will be well on the way to the nuclear and solar power-electric battery economy thanks to nanotechnology.


NEXT: Define 'Republican Enough'

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. The Navy has a large stock of helium. We could use that.

  2. Seriously! Politicians talking about technology can only mean bad news for the country. From Reagan’s missile defense to everything Al Gore ever said. There is no greater example of speaking in total ignorance.

  3. The demand for electricity fluctuates dramatically over the course of a day. At night, much of the system’s capacity goes unused, while during the day, “peakers” are used. Peakers often produce a great deal more pollution than the rest of an energy company’s system – they’re the really dirty plants that are only used when they absolutely have to be, in order to keep the overall pollution emissions low. They tend to be older oil and coal plants.

    By using existing cleaner (like natural gas, nuke, hydro, or wind) plants to manufacture hydrogen during the night, when there is a lot of excess capacity, the peak demand can be met without needing to turn on the peakers. Maybe not as clean as windmills, but one hell of a lot better than firing up each energy company’s oldest, crappiest plants every day during hot, humid summer days!

    What’s more, the pollution created by a power plant 1) is located some distance away from heavily populated areas, as opposed to cars, which spew out most of their emissions in congested cities 2)can be more efficiently captured and sequestered than is the case with 100 million non-point sources (tailpipes) and 3) is likely to be less, because a handful of highly profitable power plants are more likely to be kept in good working order than 100 million private automobiles.

  4. By using existing cleaner (like natural gas, nuke, hydro, or wind) plants to manufacture hydrogen during the night, when there is a lot of excess capacity, the peak demand can be met without needing to turn on the peakers.

    But why make hydrogen? Converting to H2 will require the construction and maintenance of a vast and inherently unstable infrastructure. Batteries, capacitors and fuel cells are a much better solution.

  5. Warren,

    It is my understanding that sequestering energy in hydrogen is both more efficient than other practical methods, and cleaner. In terms of transportation, a lighter-than-air gas would take a loss less energy to move a given distance than a tractor-trailer loaded with batteries. And batteries are a significant source of chemical pollution.

    As for the unstable infrastructure, that’s a technological problem with a technological fix.

    There’s a power company in western Mass that has a hyrdo-peaker. At night, it uses its power plant to pump water to a reservoir at the top of a hill. During the day, it opens the spillway, and the water turns a turbine. Obviously, there’s power inefficiency built into the system, but when you consider the peak hour prices of electricity compared to the costs of the extra natural gas they use at night, you can see how such a system can be profitable. Also, the water way was left over from an old mill system, so their construction costs were minimal.

  6. Joe,
    There’s nothing wrong with your peak hours vs. continuos production thesis. But frankly, the disadvantages of hydrogen over competing technologies make the question a ‘no brainer’.

    But let’s suppose I’m wrong. Investment and development in hydrogen, as with everything else, should be left to the free market. If hydro-hippies(tm) can make it work, more power to them! (hee) But no way should the federal government be digging into my pocket and funneling (a portion) of it to delusional, semi educated, tree huggers in order to buy their patronage.

  7. Joe you miss one other thing… it’s not whether the demand is met by new power stations or poewr stations running “off peak” its about using fossil fuel. So if its irrelevant if the power comes from a new staion or a more efficiently used EXISTING station. Each will be burning coal, gas, or oil to generate power for Hydrogen production.

  8. Joe L,

    The efficiency improvements of new plants over old is considerable. Plus, not all fossil fuels are created equal. There are major differences in the amount of pollution they create, and the power they can generate from the same amount of fuel.

    Do you really believe that there would be no difference between living downwind of a 40 year old coal plant, and living downwind of a brand new natural gas plant?

  9. Warren,

    Do you have a good source for your contention that hydrogen is a less-attractive means of energy storage, vis a vis the solid state systems you prefer?

  10. Joe, the point IS that energy will still have to be expended to produce hydrogen. Now natural gas may be more efficient than coal or cleaner, but it still produces emissions and heat.
    As to hydrogen, the gas is extremely difficult to contain, so the storage of it is problematic, as a liquid it requires cryogenic storage and the only other option is a solid or a liquid that is an admixture of hydrogen and something, such as lithium hydride… Hydrogen is still at the experimental stage, yet.

  11. One problem with hydrogen, is that while it is 4 times as energy dense per pound than gasoline, it is much less energy dense per unit volume. The example I saw showed that a 15 gallon gasoline tank could be replaced by a 60 gallon liquid hydrogen tank. When you couple that with the fact that hydrogen tanks would have to be heavier than similar gasoline tanks, it takes more energy to transport hydrogen than gasoline.

    I’ve read that boron/boron oxide energy exchange could be done more cheaply than even gasoline, so there is definitely the possibility that furture battery technology could be better than hydrogen.

    10-15 years ago, people were talking about getting around the peak energy problem by building giant superconducting coils that could store huge amounts of power. Had the government put its thumb on the scale for that in a bigger way, maybe it would have come to pass, and wasted a lot of resources that would better have been spent on fuel cell or whatever now and in the future…

  12. Its like the holy grail of an “Electric Car” – stupid idea – oil is a much more efficient way of distributing energy. An electric car fired by a coal fired power plant is I would believe more polluting than a oil (gas) car. On the other hand – hybrids are a great idea.

    Another stupid idea – corn based fuels – from my understanding it takes something like a 1.5 barrels of oil to make the equivalent of 1 barrel of ethanol.

  13. all-

    I do fundamental research (courtesy of the DOD) on an integral component of a direct methanol fuel cell. When the technology has matured it will be possible to fill your car with methanol much like your would gasoline. The products are electricity (transfered to mechanical energy), water, and carbon dioxide. Methanol is easy to make through fermentation (let the sun and bacteria do all the work) and is an energy dense liquid. This is technically a hydrogen powered fuel cell. The product of a water methanol reaction is hydrogen which then passes through the fuel cell membrane accompanied by an electrical current (no batteries needed). The beauty of this technology is that the current infrastructure used to distribute gasoline can also be used to distribute methanol. It is also a market driven technology. Moterolla, Toshiba, GM, and a host of other companies are dumping money all over.

  14. Uh. Windmills “covering the land” is not apt.

    They will cover farmland. In that regard including access roads they are not a big land user. About 1/2% of a farm would be used for the wind towers and access roads combined.

    The big problem with hydrogen is that there is no viable fuel cell for vehicles at this time. The much touted PEM cells are about as efficient as a gasoline hybrid in converting fuel to transportation. Not much incentive to change there, when you consider the losses in the rest of the system.

    In addition there is no very good way of storing hydrogen in vehicles that is not a road hazard. And then there is the fueling problem.

    Notice I haven’t even mentioned electrical supplies.

    To have a lot of hydrogen powered vehicles on the road in 15 years is very unlikely. It will be done but 50 to 75 years is more like it.

    BTW most of the windmills will be sited in the Mid West plains. Because that is where the wind is you know.

    America is the Saudi Arabia of Wind. As wind turbines get larger the cost of electricity generated by them will decline below the cost of nukes. I fail to see how that makes nukes a source that will become much larger than they are today.

    It seems like many of the people here prefer the pie in the sky future based on technology that is barely in the R&D stage over the economics of technology that is already being deployed.

    I don’t get it. Then again I’m an engineer. There are two rules in engineering. It has to work. The customer must be able to afford it. I don’t see either aspect mentioned here.

    BTW platinum is currently the catalyst of choice for fuel cells. There is no enough of it currently available for an American hydrogen economy let alone a world one.

  15. B,

    You are correct about methanol being the most likely hydrogen carrier.

    And the money bit too.

    There are a few snags. It produces a heat rise which is unacceptable inside portable eqpt. i.e. it would be good as a battery charger not a battery – with current designs.

    There is still tthe platinum problem.

    Now powering a lap top is a high value operation. To power a vehicle the costs will have to drop by a factor of 20 to 50 from the first viable laptop versions – which are just around the corner. Another year or two at most. Which is the state they have been in for about three or four years. Evidently there are snags.

    Did I mention the current fuel of choice is 10% methanol? Now in a vehicle the water produced by the fuel cell could be used. But that adds more plumbing and motors and stuff. And it is another energy consuming system.

    If this stuff was easy it would already have been done. After all the fuel cell battery is something like 150+ years old.

  16. I’m confused by these numbers. According to, there are 104 operating nuclear reactors in the US providing about 20% of our total electrical energy.
    If we say it would take 1000 nuclear plants to replace cars, we come to the conclusion that automobiles use 200% as much energy as all our other electricity needs. This seems a bit high.

    It’s also noted above that replacing all direct electricity with hydrogen (presumably to be used in a mini-generator in your home? WTF?) would take 2000 nuclear plants (I’m assuming the cars still run on gas), or the equivalent energy production of FOUR TIMES what we use today. Are we saying that hydrogen is only 25% efficient compared direct electric delivery or is my math bad?

  17. Galius,

    I’ve seen much lower estimates for the number of nuclear plants it would take to replace gasoline in cars, etc. There may be some advocacy going on…

  18. “Joe, the point IS that energy will still have to be expended to produce hydrogen. Now natural gas may be more efficient than coal or cleaner, but it still produces emissions and heat.”

    Joe L, you are the one missing the point. Existing gasoline engines already produce emissions and heat. The question is whether replacing those gasaoline engines with hydrogen engines (with the hydrogen created at power plants) would reduce the pollution created in order to power the cars. Arguing that burning natural gas produces emissions and heat is irrelevant.

  19. JDM,

    Flywheels are probably as good as superconductors for energy storage. Especially if you consider that it is currenly possible to mass manufacture them.

    The volume produced, cheap superconductor is not here.

    To make volume possible you have to at least be able to extrapolate down the cost curve to where the technology is viable. Which means finding a big enough market niche that would be viable at near current prices. I have looked at this quite a bit. The short answer. No such market exists.

    Production wise the current gasoline engine produces power for $10 per hp in automobiles. The best current projection for fuel cells is $100 to $1000 per HP. You want a $20,000 car with a $10,000 engine? Your $20,000 car would have to retail for $50,000 to $100,000 or possibly as much as $1,000,000. Show me your salesmen.

    It is not just a technology problem. It is a dollars problem. No short term (less than 20 years) government program is even close to a solution. In fact at this point no one knows what the solution is.

    Nanotechnology will not solve the watts or joules per pound issue. Let alone the dollars per joule or dollars per watt issue.

    If the level of thinking I have seen here is applied by the government to the issues involved we will spend a lot of money and get the usual.

    Government is politics. It is not science. It is not engineering. And it is certainly not marketing. Which is why it is politics.

  20. I am technically illiterate, thats true Joe, but I’m not missing the point… Energy is being expended, either in my car or at a power plant. The question is, which is the maximimum point on several axises, of cost, efficiency, cleanliness, and convenience. The fact that IN MY car hydrogen is cleaner is not the same thing as saying, OVERALL hydrogen is cleaner. And IF the production of hydrogen involves the burning of more coal or even natural gas, then the overall porecess may or may not be economical or environmentally friendly.

    And non-peak use of power plants will involve the increased use of fuels such as coal, oil, or natural gas, so I’m asking, not disputing, is that REALLY any better than the current situation or any better than the construction of new plants SPECIFICALLY for the production of hydrogen?

  21. Galius,

    Currently the energy loss in producing hydrogen is about 50%. The PEM fuel cell (the automotive panacea) is at best 50% efficient (it is somewhat better partially loaded). So it winds up that in terms of energy the fuel cell SYSTEM as currently envisioned is on a par with a gas/electric hybrid. BTW I haven’t included any losses from converting coal or natural gas into electricity.

    It is the same old problem in terms of the publc. Every one with access to Scientific American and a light switch is an energy expert.

  22. HAH, M. Simon I don’t even have a subscription to Scientific American, its too wordy and dull! I just have access to a light switch! Though I have VISITED coal mines and power plants….

  23. M. Simon,
    I myself am an engineer with a BS in electrical engineering. Your critique of various technologies is valid and on point as far as I can tell.

    However, I don’t understand your defense of wind power. Your contentions that “covering the land is not apt” and “America is the Saudi Arabia of wind” do not persuade. To paraphrase you; If this stuff was easy it would already have been done. After all the wind turbine is something like 150+ years old.

    I personally would bet on nuclear power plants (already proven) and hybrid vehicles for the near term. In the long term, nuclear power plus? maybe fuel cells, maybe batteries, I myself think capacitors are the best bet. The last thing I think can be made to work is hydrogen (as hydrogen in gas or liquid form).

  24. We need nuke power now (actually, needed to start building more nuke plants 20 years ago) and need to use nuke until solar gets cheaper & more efficient. Then, areas with good sun exposure can generate their own power and generate enough to produce hydrogen for use in areas that don’t have enough sun. Hydrogen has problems, but if energy is really cheap, like from more efficient solar power, Hydrogen is really cheap to produce, much more so than mining coal or drilling for oil/natural gas.

    At least that’s what Scientific American told me last month.

  25. Warren,

    I think your problem is that your info is 20 years out of date. The fact is that currently wind is lower cost than natural gas and is complimentary. i.e. they dispatch in about the same time frame. Right now it is touted as helping to cap natural gas costs for electrical generation.

    When land based turbines reach the 3 – 5 MW peak range the electricity will cost the same as coal or nuke (current series production is in the 1.5 MW range). The cost curve is: double the size reduce costs per watt by 30%. About the same as the cost curve for electrical generation plants from 1900 to about 1950 or so. For similar reasons.

    Wind right now is within range of coal or nukes (less than 2:1 difference). Compare that to fuel cells. Wind turbines are being mass produced. Fuel cells are not.

    As to the land issue. I think that very few people will care whether farms are producing just wheat or electricity and wheat.

  26. I think another major thing to look at is how many power plants have been built in the US in the last however long, 20 years, 30 years, whatever. The reason we have to use those stinky dirty old peak plants is because of generation costs and regulations. You can’t build new plants because of whatever reason: NIMBYs, fishies, or birds, or pollution. Unless that changes, the idea of generating enough excess capacity to make enough hydrogen by electrolysis to replace fossil fuels in even just cars is ludicrous. I also don’t see the point to using oil, or natural gas power plants to produce electricity to make hydrogen for cars. Why not use the oil or natural gas itself, without going through 2 efficiency losses.

    I’d love to see more done with biological processes, such as the methane production by bacteria, or even hydrogen production by bacteria that has been known to work on small scales. Will it scale up to large scales? Hopefully. That would certainly be better than producing Hydrogen by electrolysis. I have to also admit that I’m not overcome by warm and fuzzy feelings about having a car with a big hydrogen tank in it. Not very palatable, but maybe that’s the Hindenburg talking.

    What we really need is a better way to store energy. It’s telling of the state of mass energy storage that one of the best ways to do it is pump water up a hill. While a novel idea, it’s not gonna work everywhere (or hardly anywhere, for that matter). So I think we’re waiting on a breakthrough in energy, preferably electricity, storage. So does anyone have the secrets of shipstones on their desk anywhere? I think we need to work toward that, instead of trying to make do with our litany of not-so-good alternatives now. That doesn’t mean stop progressing, because more efficient cars are better, and cleaner electricity generation is better, but our problem is both total energy production and storage.

  27. M. Simon,
    Hmmm, OK. I’d be happy to see wind generation take off. If you keep in on the farm I don’t see a problem.

    My feeling is that you are over playing the amount of wind power available and under playing the square footage needed to tap it. Time will tell either way.

    The point is, we don’t need politicians gazing into crystal balls and betting our future on their visions.

  28. Guy,

    Scientific American left out one important factor.


    The only current alternative energy producer that is even close to the costs of nuke electricity is wind.

    Solar is about 3X to 5X the cost of wind electricity. Good for high value low energy useage situations. Like weather stations or oil field monitoring (yep – solar is favored for oil field monitoring – AMOCAMS).

    There is no way at current prices to get back your investment in solar electricity if the grid is available.

  29. Warren,

    Fact check my ass. I’m an engineer. Every thing I know on the subject is available from open sources.

    I’d welcome verification. Or correction.

    On the politician bit: you are correct. Market forces are such that the politicians ought to butt out. At most I’d favor one more 3 year extension of the wind Production Tax Credit (which is a tax reduction). It would give us a bit of pump priming until turbine size reaches 3 to 5 MW peak.

  30. Solar power is great for remote uses but I seem to remember that it takes a huge amount of energy to make a solar cell.

  31. Just use perpetual motion machines to make the H2, like the guys profiled here claim to:

  32. Last week at the Nanobusiness conference in Chicago one of the presentations was about a company that makes solar cells out of plastic and titanium dioxide. (Using TiO2 for solar cell technology has been around for a while; up to now there’s been a problem with the lifetime.) They were quoting expected 10 years lifetime already–low efficiencies so far (single digit) but the improvement curve has been good. I expect them to reach 20-30%. At moment company in niche markets (military) but this is an area where economies of scale mean something. (As opposed to standard solar cell technology, where we’re always using the silicon wafers rejected by the chip industry.)

    Yes itwouldbenice if we continued to be able to have hydrocarbons around since they pack so much potential energy into so little space. Unfortunately…..Peak Oil.

    So who you gonna call? It’s not that the present alternatives may be able to get down to being cost-wise equivalent; it’s that we won’t have a choice.

  33. It should be pointed out that we already have a hydrogen economy. For example, a company called Air Products makes hydrogen for all sorts of industrial processes.

  34. tzs,

    We may in fact not be at the end of the hydrocarbon age or at the peak oil point.

    Take a look at Thomas Gold on the subject.

    In fact the recent spike in oil prices may be a market manipulation by folks who want Bush to lose.

    As to not having any choice. We will know that we are there when electricity prices start rising.

    So far the price of steam coal is declining. The price of wind is declining. I don’t see it.

Please to post comments

Comments are closed.