[comicjk]

That's not what "entropically preferred" means. If something is a lower-energy state, it's energetically preferred, but may or may not be entropically preferred. Water is NOT - entropy would prefer simpler molecules over more structured molecules.

I think the concept you're thinking of is free energy, which determines the final destination of a process. The equation relating these things is (change in free energy) = (change in energy) - temperature × (change in entropy). Entropy only becomes the dominant component when temperature is high. And, as expected, water molecules dissociate at high temperature.

[kurthr]

No, since this was chemistry, I was talking about enthalpy of an exothermic reaction (negative enthalpy), and interstellar space is pretty close to constant pressure, but not constant entropy for Gibbs free energy.

In any case: dH=TdS + Vdp note dp is small in space but V can be large and dS is the change in entropy.

...and enthalpy of ideal (interstellar H & O) gases does not depend on pressure, unlike entropy and Gibbs energy. If you really just mean free energy U, then they are basically the same thing in open space (but not in a plasma), a distinction without difference.

For further pedantics I recommend Wikipedia, since I doubt we are helping anyone else.