Hydrogen Economy Humbug


Aerospace engineer (and Mars Society president), Robert Zubrin thoroughly debunks hydrogen hooey in his excellent New Atlantis article, "The Hydrogen Hoax." To wit:

The spokesmen for the hydrogen hoax claim that hydrogen will be manufactured from water via electrolysis. It is certainly possible to make hydrogen this way, but it is very expensive—so much so, that only four percent of all hydrogen currently produced in the United States is produced in this manner. The rest is made by breaking down hydrocarbons, through processes like pyrolysis of natural gas or steam reforming of coal.

Neither type of hydrogen is even remotely economical as fuel. The wholesale cost of commercial grade liquid hydrogen (made the cheap way, from hydrocarbons) shipped to large customers in the United States is about $6 per kilogram. High purity hydrogen made from electrolysis for scientific applications costs considerably more. Dispensed in compressed gas cylinders to retail customers, the current price of commercial grade hydrogen is about $100 per kilogram. For comparison, a kilogram of hydrogen contains about the same amount of energy as a gallon of gasoline. This means that even if hydrogen cars were available and hydrogen stations existed to fuel them, no one with the power to choose otherwise would ever buy such vehicles. This fact alone makes the hydrogen economy a non-starter in a free society.

And even if you are among those willing to sacrifice freedom and economic rationality for the sake of the environment, and therefore prefer hydrogen for its advertised benefit of reduced carbon dioxide emissions, think again. Because hydrogen is actually made by reforming hydrocarbons, its use as fuel would not reduce greenhouse gas emissions at all. In fact, it would greatly increase them.

To see this, let us consider an example. Let's say you wanted to produce hydrogen. You choose to do it via steam reformation of natural gas, the most common technique used commercially today. The reaction is:

CH4 + 2H2O => CO2 + 4H2 ?H = +59 kcal/mole (1)

As the positive enthalpy change indicates, the reaction is endothermic (that is, heat-absorbing) and will need an outside source of energy to drive it forward. This can be obtained by burning some methane, which releases 205 kcal/mole, via the following reaction:

CH4 + 2O2 => CO2 + 2H2O ?H = 205 kcal/mole


Assuming an optimistic 72 percent efficiency in using the combustion energy to drive the steam reformation, this would allow us to reform 2.5 moles of methane for every one that we burn (or 5 for every 2). So if we take five units of reaction (1) and add it to two units of reaction (2), the net reaction becomes:

7CH4 + 4O2 + 10H2O => 7CO2 + 4H2O + 20H2


As far as usable fuel is concerned, what we have managed to do is trade seven moles of methane for twenty moles of hydrogen. Seven moles of carbon dioxide have also been produced, exactly as many as would have been produced had we simply used the methane itself as fuel. The seven moles of methane that we used up, however, would have been worth 1435 kilocalories of energy if used directly, while the twenty moles of hydrogen we have produced in exchange for all our trouble are only worth 1320 kilocalories. So for the same amount of carbon dioxide released, less useful energy has been produced.

The situation is much worse than this, however, because before the hydrogen can be transported anywhere, it needs to be either compressed or liquefied. To liquefy it, it must be refrigerated down to a temperature of 20 K (20 degrees above absolute zero, or -253 degrees Celsius). At these temperatures, the fundamental laws of thermodynamics make refrigerators extremely inefficient. As a result, about 40 percent of the energy in the hydrogen must be spent to liquefy it. This reduces the actual net energy content of our product fuel to 792 kilocalories. In addition, because it is a cryogenic liquid, still more energy could be expected to be lost as the hydrogen boils away during transport and storage.

As an alternative, one could use high pressure pumps to compress the hydrogen as gas instead of liquefying it for transport. This would only require wasting about 20 percent of the energy in the hydrogen. The problem is that safety-approved, steel compressed-gas tanks capable of storing hydrogen at 5,000 psi weigh approximately 65 times as much as the hydrogen they can contain. So to transport 200 kilograms of compressed hydrogen, roughly equal in energy content to just 200 gallons of gasoline, would require a truck capable of hauling a 13-ton load. Think about that: an entire large truckload delivery would be needed simply to transport enough hydrogen to allow ten people to fill up their cars with the energy equivalent of 20 gallons of gasoline each.

Instead of steel tanks, one could propose using (very expensive) lightweight carbon fiber overwrapped tanks, which only weigh about ten times as much as the hydrogen they contain. This would improve the transport weight ratio by a factor of six. Thus, instead of a 13-ton truck, a mere two-ton truckload would be required to supply enough hydrogen to allow a service station to provide fuel for ten customers. This is still hopeless economically, and could probably not be allowed in any case, since carbon fiber tanks have low crash resistance, making such compressed hydrogen transport trucks deadly bombs on the highway.

And it gets worse. Read on.

In February 2004, in my "Hydrogen Bombs" column, I asked,

I'm as big a techno-optimist as you're apt to find, but hydrogen doesn't seem to have much of an immediate future as a replacement for our current energy system. Besides, why use electricity to make hydrogen to make more electricity? Why not just use electricity for what you need and instead do a lot of research on improving battery technologies?

Last week saw the launching of the Automotive X Prize (modeled on the Ansari X Prize for reusable space craft) that will award a prize to whomever develops a commercially viable car that gets 100 miles to the gallon. I think a better prize (say $100 million) would be for developing a commercially viable battery that can charge and recharge in less than a hour, last at least 10 years, and power a car for 300 miles. (Of course, I am more than happy to listen to more technically savvy people suggest more "reasonable" battery prize goals, but you get the idea.)

I am less sanguine about Zubrin's conclusion that we should mandate flex-fuel vehicles because I think that turning food into fuel is an economic and ecological deadend. (Though I do have hopes that cellulosic ethanol may turn out to be economically and ecologically viable.)

I realize the urge to comment on this topic will be nearly overwhelming, but I don't think you'll regret reading Zubrin's entire article.

Disclosure: I own a few shares of stock in two hydrogen fuel cell companies which have declined more than 90 percent in value since I purchased them 6 years ago. In fact, a shareholder suit against one of the companies just netted me something like $30 as compensation though the class action lawyers got a bit more than that. If you run out and buy such stocks based on what you read here, you've clearly not understood the arguments. On the other hand, if you nevertheless feel like mandating or subsidizing hydrogen production, one side effect may be that value of my almost worthless stock will go up.