29 Aug '13 22:33>1 edit
A thought just struck me about this. Suppose we have a fairly high energy "Standard Model" (specifically I'm assuming it's massless) photon trundling along. It can split into an electron positron pair, the electron emits a W- particle and turns into a neutrino. The W- emits a photon and is then absorbed by the positron which turns into an anti-neutrino. Since the electron type neutrino has a mass less than 0.1 eV (based on a cosmological bound on the sum of neutrino masses being 0.3eV see the Wikipedia page on neutrinos) an upper bound on the lowest energy photon that can do this has an energy of 0.2eV which is near-infrared. The Feynman diagram corresponding to this process is non-zero (the photon can't connect to the electron as switching the electron and positron around picks up a minus sign so the graph cancels), so unless it's suppressed by Ward identities or otherwise canceled it should contribute to photon decay. It won't happen at much of a rate as there are 4 powers of Planck's constant and it's mediated by the weak force (which reduces the rate a lot), but I don't see an immediate reason why this shouldn't be a process whereby a sufficiently high energy (>~ 0.2eV) massless photon could decay into a couple of neutrinos.
This process won't work if the photon is massive but lighter than the 2 neutrino masses we need at the threshold, as then one would be able to find a frame of reference where the photon is stationary - in that frame it is prevented from decaying by conservation of energy - the same must apply in all other frames, so if the photon has a mass this decay channel is ruled out.
This process won't work if the photon is massive but lighter than the 2 neutrino masses we need at the threshold, as then one would be able to find a frame of reference where the photon is stationary - in that frame it is prevented from decaying by conservation of energy - the same must apply in all other frames, so if the photon has a mass this decay channel is ruled out.