He has indeed written that this transition releases "a few hundred eV per atom", so believe he must be aware that it is strongly exothermic.
Ok, so there's good reason to think that Holmlid will agree with something in the ballpark of your 640 eV bond energy calculation.
Holmlid has written that it this H(0) material has a strong bond and that it is stable so I believe that he must be aware of this as well.
Yet a strong magnetic field can apparently cause it to easily transition to H(1). I do not think I am able to provide a precise explanation of why this occurs. However, since it is supposed to be a room-temperature superfluid and superconductor, it may also be a consequence of these properties.
How does one undo a 640 eV bond with a magnetic field, even a strong one? There's an energy balance problem that deserves close attention here which I think is being neglected. Energy must be conserved (if we're not to stray too far from the assumptions of physics as it is currently practiced). If the 640 eV bond can be broken with an external magnetic field, either the bond was not really 640 eV, or we have another store of potential energy lying around that was somehow tapped by the magnetic field. If you click on the button on an umbrella, it shoots out, which is analogous to the 640 eV release. Now you have to apply force to push the umbrella back to its initial position. But instead the suggestion seems to be that you can click on another button in the umbrella, and it shoots in back to its initial position.
The same could be said if any nuclear reaction occurred within the lattice as in countless other LENR claims. The reaction sites would eventually all get destroyed. Craters and local melting may visibly appear too. I think these are common problems in LENR systems that have been reported to be working.
The point I made earlier about lots of energy being needed pertained to the undoing of the 640 eV bond and was not an objection about what is seen in LENR. If you have that kind of energy being released which, I have assumed, is needed to break the 640 eV bond, you'll no longer have a solid within which the presumed H(0) can be turned into H(1) or H(2). Instead perhaps you'll have a plasma, which is a different situation than the one in which you suggested the H(0) would be re-inflated (i.e., in a metal). My point pertained not to a high energy release, but to the explanation about H(0) being turned into H(1) or H(2) in the context of a metal.