Among a number of other strong points, this WWF report (PDF) on “The end of the oil age” highlights some of the problems with hydrogen as a fuel, particularly for vehicles. One major issue raised is the difficulty of transporting the stuff:
Despite having a high specific energy (i.e. energy content per unit mass) of 142 MJ/kg, the physical density of hydrogen is just 84 g/m^3, which means that one kilogramme of the gas occupies around 12 m^3 at normal temperature and pressure (NTP). By comparison, one kilogramme of natural gas displaces 1.4 m^3 and packs a specific energy of 54 MJ/kg. This means the volumetric energy density of hydrogen is only one-third that of natural gas, making the cost of a hydrogen pipeline around six times higher than a natural gas pipeline of equivalent energy capacity. The IEA projects that worldwide investment required to develop a hydrogen pipeline network might be in the order of US$ 2.5 trillion, while noting that the energy required transporting hydrogen via pipeline is on average 4.6 times higher per unit of energy than for natural gas. This equates to an efficiency loss of ten percent over a distance of 1,200 km; the same energy would move natural gas 5,000 km.
As an alternative to pipeline distribution, like natural gas, hydrogen may be either compressed to around 200 atm or chilled close to absolute zero for transportation via truck or ship. Both processes are energy intensive, resulting in additional efficiency losses in the hydrogen supply chain, and super-cooling requires venting that can further deplete the stored fuel. According to one study, it takes 22 tube trailers at 200 atm or 4.5 liquid hydrogen tankers to carry the energy contained in a single gasoline tanker of the same gross weight.
Comparing the hydrogen distribution efficiencies with our electron pathway, we know that electricity grid transmission and distribution (T&D) losses of around 6-8% are typical in OECD countries. Whether carried by pipeline, tanker or ship, it is therefore inconceivable that centrally-produced hydrogen will ever match the efficiency of the electricity grid. Only if it is synthesised at or close to the point of use would hydrogen avoid significant energy losses associated with distribution. Even then, mindful that our guiding principle is the exclusive use of energy from sustainable renewable resources, hydrogen produced in localised facilities would still need to be compressed for storage and/ or delivery directly to the vehicle, which would of course incur further energy losses.
As I have said several times before, hydrogen is a low quality fuel, difficult to handle with low energy per volume. Even with electrolysis at 80% efficiency, the report concludes that hydrogen vehicles would have an overall efficiency of 28%, when production, transport, and the operation of fuel cells are taken into account. Given that vehicles using hydrogen fuel cells would actually be using electricity to drive their motors anyhow, it seems more sensible to focus our efforts on battery technology, which the report concludes to be 23% more efficient, with a lot less new infrastructure to build.