I've studied the idea of creating synthetic fuels--mostly ethyl alcohol--from 'salvaged carbon' from biomass, municipal waste both solid and liquid, by utilizing bulk hydrogen to upconvert the carbon-oxygen chains using a Fischer-Tropp conversion process. While this doesn't eliminate CO2 emissions, it extends the usefullness of petroleum carbons by using it twice--once for packaging and what not, and then again as a fuel. Carbon from biomass could supplement this, but could not provide the necessary amount for complete displacement of petroleum based carbon. Coal could make up the difference, but again you'll get a net carbon emission.

The use of boron has been mentioned as a fuel a couple of times because it can be easily recycled--I wonder if something like borane or diborane--boron analogs of methane and ethane--could be used as a fuel. Diborane spontaneously ignites (hypergolic) with air and is extremely toxic, making fuel spills an immediately lifethreatening problem.

Still, a borane fueled engine, using perhaps a high pressure, helium filled multicylinder Stirling cycle engine could achieve automotive-like performance.
The oxide particles, if the combustion chamber were insulated well enough, would remain molten and pool one a collection plate, where further heat extraction would cause it to solidify. The collection plate could then dump the boron-oxide 'brick' into a storage bin for returning to a 'fueling station.' The purpose of the boro-hydride compound is ease of combustability--straight boron is difficult to burn. I would imagine that something like a pellet stove could do it, if the boron were in small BB sized pellets--which isn't a bad idea now that I think of it...

Hydrogen infrastructure for widescale transportation use is a technically and energy intensive business. Handling of cryogenic fluids is not trivial--especially liquid hydrogen. Leaks are common, difficult to detect, and very quickly become a major explosion and conflagration hazzard. Hydrogen fires in air are almost entirely invisible too (I've actually seen this--in daylight you can only detect it by the sound of the flame and the presence of 'heat waves' from the flame as seen from shadow cast on the ground!) Also compounding the problem is that gasseous hydrogen is so light I don't think it can carry an oderant like natural gas and propane, so leak detection by smell is out. And liquid hydrogen is too cold to mix an oderant in anyways, it would freeze and clog any supply lines. Liquid hydrogen is so cold, that air will liquify and freeze around uninsulated lines. The air that condenses and dripss off to pool someplace will undergo a spontaneous 'fractionation' as the more volatile nitrogen boils out, leaving almost pure liquid oxygen behind. Liquid oxygen is a major combustion hazard to anything flammable all by itself--and in close proximity to liquid hydrogen--well, it's 'rocket fuel!'

Approximately 1/3 of the energy contained within the liquid hydrogen as heat of combustion must be expended to chill and compress hydrogen gas to liquify it. Also, large storage tanks are required that must be rigourously insultated with vacuum jackets and thermal radiation shields (multiple layers of aluminum foil spaced with a very light layer of fiberglass batting.) And in large storage fascilities, a reliquification plant is necessary.

And because of liquid hydrogen's very low energy density (about 26,000 BTY/gallon, versus 127,000 BTU/gallon for a typical hydrocarbon fuel,) one must have a really big fuel tank.

This makes me think that the best use of hydrogen is to use it as a primary ingredient for a synthetic fuel which is then used for a transportation energy source. In large installations, where storage space isn't a problem, hydrogen might be effectively used as a peaking fuel to run gas-turbine peaking plants. The hydrgoen is generated during off peak hours--but this only makes any sense at all if you've already replaced fossile fuels as the primary energy source. Otherwise, hydrogen doesn't make sense at all...