If you step back from the details, a hydrogen economy is a way to get non-oil energy into the transportation segment, in particular, with minimal emissions of either tailpipe pollutants or greenhouse gases. Hydrogen represents only one of many possible pathways for doing this. A recent article in Technology Review focused on another possibility, methanol. It suggests that improvements in methanol manufacturing and advances in fuel cells that convert methanol directly into electricity, without first having to produce hydrogen, could bypass the hydrogen economy model and give us an alternative fuel that can be made cheaply from many different sources. I'm not sure things are quite so rosy, however. Methanol could also deliver a classic case of unintended consequences.
Methanol is made industrially today from natural gas in a process that consumes a significant portion of its energy content, between 25-35%. That's why cutting out the methanol-to-hydrogen step required for current fuel cells is important. Even though methanol is a liquid fuel with an acceptable energy density--about half that of gasoline--and avoids the complications of storing hydrogen on a vehicle at high pressures or low temperatures, turning it into hydrogen incurs further energy costs. If methanol can be made from non-fossil energy sources, such as wind or nuclear power, as the article suggests, then at least its greenhouse gas balance, if not its energy balance, could be significantly positive.
Unfortunately, this doesn't reckon with the challenges of distributing methanol on a large scale. The basic problem is that, unlike gasoline or ethanol, methanol is a neuro-toxin. Ingesting even a small quantity can lead to blindness or death. This is a concern for both bulk handling and at the point of sale. Now, gasoline isn't exactly water, but at least if you spill it on your hand, you don't need to be taken to a hospital. In a perfect world, no one would ever be exposed to either liquid gasoline or methanol, but we know that storage tanks, pipelines and tankers occasionally leak, and most of us have spilled a bit of gasoline while refueling our cars.
Now, methanol accidents wouldn't automatically be disastrous. In the presence of enough water, methanol dilutes quickly. Naturally occurring bacteria would break it down before it would have an opportunity to do much harm. It's much less clear, though, what happens in the case of a methanol spill concentrated enough to kill off the bacteria and remain in the environment for an extended period. There's not much in the literature on this.
A methanol economy is an intriguing notion, providing many of the benefits of a hydrogen-based energy system, but with somewhat fewer complications. The toxicity issue is a potential show-stopper, though, and deserves careful attention and rigorous field tests--as opposed to laboratory-scale experiments--to see what would happen should large quantities spill under real-world conditions.
Note: due to business travel, there won't be any posting tomorrow.