I have two questions for Robin (or anyone else who might know). First, some context.
In a metal lattice, normally the electrons are highly localized around the lattice sites. In the example of palladium, if you look at a density map of the electron orbitals, they are highly clustered around the palladium atoms, and there is a very low density in the interstitial sites. In relatively unloaded palladium, hydrogen occupies the octahedral sites. I recently read a calculation looking at the electron density, and it suggested that almost none of the electrons were located in the vicinity of the octahedral sites. The implication was that there was little to no shielding. They were arguing against a reduction in the Coulomb barrier for two deuterium nuclei occupying adjacent sites (they might have been considering the highly loaded case, where some tetrahedral sites are also occupied; I don't remember). They found the effect of electrons in the host lattice to be negligible on the tunneling rate, if I understand the calculation. That was a calculation for equilibrium conditions (e.g., not much going on). They thought they could draw conclusions on the basis of it for nonequilibrium conditions as well. In this context I note that the ionization cross section is a function of energy. At higher energies, you're likely to get ionization, and at lower energies you're likely to get an excited (Rydberg) state, but not ionization. When there is no ionization, the electron steps up into an orbital of increased reach -- a Rydberg state. It goes out further. Now imagine a host lattice with a population of Rydberg-excited electrons. The density map of the lattice would look somewhat different. There would still be the tightly bound inner shell electrons clustered around the lattice sites, but in the rest of the lattice, the density would be evened-out somewhat. Perhaps there would be an overall increase in the level of screening, even out in the octahedral sites (note that nickel is also fcc). Presumably, as the energy of the hydrogen population is increased there will be increasing Rydberg excitations of the lattice site electrons, and perhaps this would not necessarily get to the point of ionization because the energy is still quite low in relative terms. My two questions for Robin (or anyone else): - Do you have a sense of how tunneling would be affected at the locations that hydrodgen/deuterium pairs are likely to be if a significant population of nickel electrons were excited into Rydberg states? I think we can assume Ron's mechanism is also at play, but perhaps not. (If we get set aside Ron's mechanism, we have gammas to deal with.) - Is there a basic distinction between tunneling and Coulomb penetration, or are they both ultimately reducible to the amount of time that the two nuclei spend in proximity to one another? In the latter case this would presumably be because they approach so close. One interesting point to add -- I am not arguing for the importance of an fcc lattice. I think there are some very interesting things that could happen with this screening phenomenon and Ron's mechanism in cracks, for example (e.g., a dense plasma focus). Eric
