Just a side comment here about nuclear vs electronic spins. One of the things that sets the limit on how well you can cool is how strongly the spins are coupled to each other. Nuclear spins tend to be much less coupled to each other than the electron spins are, so in principle you can cool the nuclear spins much lower in temperature. On the other hand, it is very frequently the case that the nuclear spins do not couple very strongly to things like phonons either --- as a result you can build a cooling device that ends up cooling the nuclear spins to some extremely low temperature, but the vibrational (phonon) temperature of the material may remain high -- the two systems not being in thermal equilibrium.
« Back to index