How Much Hydrogen?
Although my postings have been pretty petro-centric for the last few weeks, I still intend for this to be an energy blog, not just an oil blog. Where better to seek a bit of balance than with hydrogen, which was in the news frequently last year? On another blog I ran across a recent article from the Financial Times discussing the likely connection between nuclear power and a hydrogen economy. There's also a feature article on this subject in the latest issue of Wired. On the surface, this is hardly welcome news for those who see a hydrogen future as being synonymous with green energy. A few quick calculations will indicate why people might be thinking along these lines.

Let's begin with the assumption that we'd like to replace 100% of gasoline consumption with hydrogen in twenty years. This is highly ambitious but nicely frames the scale of the challenge. The US currently consumes about 9 million barrels per day of gasoline from domestic and foreign refineries. The energy content of all that gasoline is roughly 16 quadrillion BTUs (quads for short) per year. If the hydrogen-powered cars of 2025 were to use energy three times more efficiently and drive about 60% more total miles per year than today's cars (based on long-term trends in vehicle miles traveled), then we'd need to produce 8.6 quads a year of hydrogen for them to run on.

Almost all of the hydrogen used today is produced from natural gas, at an efficiency of about 70%, i.e. 30% of the energy content of the gas is lost in the process. If that still held true in 20 years, then we'd need an incremental natural gas supply of 34 billion cubic feet per day for hydrogen production. This quantity is more than half of current US natural gas consumption. So even with the efficiency improvements inherent in the hydrogen fuel cell, a true hydrogen economy--even just the transportation component of it--will require an enormous new source of primary energy from fossil fuels, renewables, or nuclear power.

Staying with natural gas for the moment, the North American gas industry will have its hands full simply maintaining current supply levels for current uses--electricity generation, home heating and industry--over that timeframe, without adding anything for hydrogen. Since liquefied natural gas (LNG) is the industry's current answer to its supply problems, it's worth noting that the amount of gas cited above for future US hydrogen needs is equivalent to the output of 40 new LNG plants such as this one planned for Indonesia, or about 5,600 fully-loaded LNG tankers per year. Importing even a fraction of this much LNG will be a big challenge, given the resistance that most proposed LNG receiving facilities are meeting (see for example my postings of 11/15/04 and 5/17/04.)

So if natural gas isn't the long-term energy source for the hydrogen economy, what is? Frankly, it's daunting to contemplate getting the approvals necessary to install sufficient new capacity of any kind to fill this gap, whether we are talking about LNG, wind turbines, or solar arrays. All of these, at this scale, will encounter enormous opposition. Of all the options, nuclear power would require the fewest new facilities in the smallest number of locations. Perhaps this explains its attraction for some hydrogen advocates.