The problem is that the transition produces enough energy to boil the LH2. As explained on the old sci.space.history group:
Skipping the gory quantum-mechanical details... there are two energy
states of the hydrogen molecule, ortho and para. At room temperature,
hydrogen is about 3/4 ortho. At liquid-hydrogen temperatures, the
stable state is almost all para. But the ortho-para conversion is
slow, so if you just liquefy hydrogen, what you get is still 3/4
ortho, and it slowly converts itself to para. This is trouble because
the ortho-para transition releases a modest amount of energy, and it
doesn't take much energy input to boil off liquid hydrogen. In fact,
the transition will boil off all the liquid.
The fix is to find a catalyst which will speed up the transition, and
put some of that in your hydrogen liquefier, so the transition will
happen while the liquefier is sucking heat out of the hydrogen anyway.
This is now quite routine.
The difference in properties between the two are quite small; some
thermal properties are slightly different, as I recall.
I suppose ortho-hydrogen ought to have a very slightly higher Isp than
para-hydrogen, but the difference is slight and the difficulties of
handling and storage loom large.
This particular problem is pretty much unique to hydrogen,
fortunately. Hydrogen and helium, and to some extent neon, show
oddities in their behavior which are visible effects of quantum
mechanics -- they are sometimes spoken of as the "quantum gases".
(This contributes to some of the peculiarities of liquid hydrogen,
like its very low boiling point and its very low density.) Hydrogen
is the only one of the three which has molecules containing more than
one atom, and hence the only one which shows an ortho-para
In practical terms this means
As a result of this slow change, a thermally isolated tank
of liquid hydrogen prepared without conversion to the
para form will lose about 1 percent of its volume during
the first day of storage.
(this is from the reference linked below)
A number of catalysts have been used: chromic oxide, paramagnetic ferric oxide, hydrous manganese oxide. More info here.