The ‘hydrogen economy’ is being promoted as the solution to the world’s energy problems. The debate is filled with hype and resembles that surrounding nuclear power in the 1950s and '60s, when it was claimed that electrical energy would be ‘too cheap to meter’. Notwithstanding that nuclear power reactors are the least polluting, safest, most reliable, and most economical means for producing electricity in the world today, the energy is definitely not too cheap to meter. Both nuclear power and the hydrogen economy have been promoted with great fervor by advocates, who are often motivated by causes other than the reasoned selection of our energy future. The decision as to whether a hydrogen economy should be developed will likely be made on the basis of anything but scientific and economic rigor. It may well contain a good portion of political expediency in response to pressure by special interest groups.

One crucial aspect of the debate is dihydrogen (H2) production. All of the estimated 9 billion kg of hydrogen used annually in the US is manufactured, predominantly by reforming carbonaceous fuels. Accordingly, we need to think about the amount of hydrogen required in a fully fledged hydrogen economy. In 2002, according to the Department of Energy, the US consumed 7.19 billion barrels of oil, 990 billion kg of coal, and 0.63 trillion m3 of natural gas. Accepting conversion factors of 1 barrel of oil = 1700 kWh, 1 kg of coal = 6 kWh, and 1 m3 of natural gas = 10.6 kWh, the electrical energy and hydrogen mass equivalents per year are 2.61 × 1013 kWh and 1600 billion kg, respectively, as estimated from the energy content of H2 (16.343 kWh/kg for the conversion of hydrogen into electricity using a fuel cell of 50% efficiency). To replace the carbonaceous fuels currently consumed, we need nearly 200 times the amount of hydrogen produced today: a truly daunting task. To extrapolate these amounts to a world scheme, multiply by a factor of four to eight.

If global warming is established unequivocally to be the result of human activity, with CO2 as the culprit, then the world must abandon or severely regulate the delayed carbon cycle, in which carbon that was fixed by nature tens to hundreds of millions of years ago is released into the atmosphere over a few centuries. This would severely curtail the most popular method for producing hydrogen – steam reforming of hydrocarbons and coal – unless methods for sequestering CO2 can be developed. Steam reforming of carbonaceous materials produced in a prompt carbon cycle (e.g. biomass) could supply only a small fraction of the required hydrogen for an acceptable ‘footprint’.

The only alternative technology for producing hydrogen in the quantities required is nuclear (fission) power. Nuclear systems can be configured to be ‘product agile’ (hydrogen/electricity) and can be readily mated with electrolysis and thermal splitting hydrogen generation technologies. However, even more fundamental questions need to be answered: why use electrical energy to split water into hydrogen and oxygen, transport the hydrogen to some other location, and then recombine oxygen and hydrogen in a fuel cell to regenerate electrical energy; why not use the electricity directly with advanced batteries as storage devices? After all, we have a well-established infrastructure for the transmission of electricity, but do not have a comparable system for the transport of hydrogen. Perhaps we would be better served if an ‘electron economy’ was examined along with the hydrogen economy in the debate that is now gathering steam. The critical question is whether we could stand to have 3600 nuclear power reactors operating within the US, and up to eight times that number elsewhere in the world (the total energy production from carbonaceous fuels in the US is about 35 times that from the 104 nuclear power plants currently operating). This would almost certainly require a transition to breeder reactor technology and would most likely result in a ‘plutonium economy’, with all the pendant risks of nuclear weapons proliferation. In principle, of course, all these problems would be circumvented by the successful development of fusion reactor technology.

In addition to technologies based on the prompt (biomass) carbon cycle, other ‘renewable’ technologies capable of producing hydrogen include the direct solar photoelectrolysis of water; water electrolysis using wind, solar, and tides as sources of the required electrical energy; and photobiological processes involving bacteria. While each of these processes will almost certainly make a contribution to the hydrogen economy, it is unlikely that they will be major players because they are too diffuse for our industrial societies. One aspect of biomass is particularly intriguing: what would happen if the efficiency of photosynthesis, which is only about 0.1-5%, could be increased by a factor of ten by genetic engineering? This could result in an enormous increase in biomass production and might make significant hydrogen production from the prompt carbon cycle quite feasible. Gone would be our worries about global warming, but we might have to mow our lawns ten times a week!

[1] Digby D. Macdonald is a distinguished professor of materials science and engineering at The Pennsylvania State University.

Read full text on ScienceDirect

DOI: 10.1016/S1369-7021(04)00319-0