Ring current on the nanoscale as it arises in the hexagonal configuration of the Shell 105 catalyst under laser irradiation - seems to be another major piece of the Holmlid puzzle if not opening up a broader understanding of LENR in general.
It should be admitted up front that Holmlid does not buy into the SPP formation hypothesis, at least not yet . (according to an answer given by Ólafsson). First problem. Rust is a poor electrical conductor in a bulk or linear sense, as it is a semiconductor but it is photoelectric. At the nanoscale, rust never sleeps and will circulate huge equivalent current in the form of localized ring current. In terms of amperes per unit of area, large current is required for SPP formation. This ring current implies local superconductivity at the nanoscale or extremely low losses in iron oxides and several other hexagonally arranged semiconductors. http://newscenter.lbl.gov/2012/09/06/rust-never-sleeps/ To summarize the argument. SPP formation requires photons intersecting with large electrical current. The electrical current, or lack thereof, was the issue which seemingly was the hardest to imagine or explain in the context of ceramic reactors, like the glow tubes. When photons are absorbed on a photoactive semiconductor the photoelectric effect provides free electrons. When these electrons are localized as nearly lossless ring current, then the problem of SPP formation requiring huge current (megamp/cm^2 equivalent) has been essentially solved. The beauty of this is that it could never happen with a good electrical conductor. Computer simulation indicates that the nanostructure of the iron oxide catalyst forms the requisite hexagons at nanometer geometry. Regular hexagons are the key for ring current. There are several images here which show the hexagonal porosity and the 12 iron atoms which form the ring, with oxygen. http://www.sciencedirect.com/science/article/pii/S2352214315000106 The oxygen atoms which line the ring on its interior diameter are not fully reduced, and the result is a conductive pathway for ring current reminiscent of copper oxide HTSC. The pathway appears to be superconductive on the local scale but low loss is acceptable for this hypotesis. The Casimir geometry of the ring could explain the low losses or pseudo-superconductivity. The SPP magnetic field vortex is then responsible for the densification of deuterium. All the pieces of the puzzle are fitting together elegantly, thanks to Holmlid and his openness to completely share relevant information (unlike Mills or Rossi). Now we need experimental confirmation.

