Although hydrogen contains a lot of useful energy per unit mass, the gas density is very low; consequently, the energy per unit volume is extremely small, too small in fact to make the gas at normal pressure useful as a transport fuel. The energy density can be increased to close to that of conventional liquid fuels by compressing to as much as 700 bar, but the penalty is that a tank able to withstand this pressure is required. In addition, refuelling is much more difficult.
There is an energy cost as well. Though compressing the gas does not affect the intrinsic chemical energy, additional mechanical energy is required for compression which, if not recovered, will indirectly reduce the energy value of the stored hydrogen. The energy can be fully recovered if the gas is compressed adiabatically (meaning no heat is allowed to escape the vessel). If the gas is allowed to expand back to normal pressure prior to consumption in a fuel cell, the additional energy that was put in can be converted back to mechanical energy which can potentially be used to generate electricity.
However, this is not feasible in practice and the compression energy is almost always lost. That being the case, one can ask what is the most effective strategy for compression that minimises the mechanical energy required? It turns out it is better to try to maintain the temperature at ambient as the gas is compressed. This is possible if the heat of compression is extracted, or allowed to escape the vessel by the use of appropriate heat exchangers. Ideally the compression should be isothermic, though this too is difficult to achieve in practice without making the process too slow.
The model above makes the comparison (click anywhere to restart the simulation). Each vessel contains 1 mole of gas with energy content 237 kJ. The vessel on the left is thermally insulated and the one on the right allows unconstrained heat transfer. The model is constructed by making fractional changes to the initial volume (22.4 litres) on each step until the volume drops to 0.224 litres. The calculation of course scales to any initial volume at STP. The changes in pressure (p)and temperature (T), and the work done over each step are:
Adiabatic Compression
dp = - γ p/V dV
dT = -(1 - γ) T/V dV
dE = - p dV
Isothermal Compression
dp = - p/V dV
dT = 0
dE = - p dV
Remember, the volume change dV is negative. γ is the ratio of specific heats, 1.41 for hydrogen. The energy calculation refers to the work done on the system.
Note how the energy required increases dramatically even for isothermal compression as 100 bar is approached, equivalent to 4% of the stored chemical energy.
Summary
Although much of the energy used in isothermal compression is not recoverable, this makes little difference as the energy is never really recovered in practice. That being the case, compression should take place under (as far as is possible) isothermal conditions to reduce the energy cost of compression.
Investigations