The most intriguing aspect of the new “high Tc” superconductors is that we still have no consensus about the nature of the basic mechanism for superconductivity in these materials, nor about why Tc is so much larger than in conventional superconductors. As is now well known, these materials include compounds of the type La2-xSrxCu2O4-y (where Sr can also be replaced by Ba) in which the maximum Tc attained is between 35 K and 40 K, and of the type YBa2Cu3O7-δ (where Y can also be replaced by most rare-earth atoms) in which the maximum Tc attained is about 98 K. Very recently, new compounds such as Bi2Sr2CaCu2O8+y have been discovered where Tc's of up to 120 K have been observed, but the physical properties of these at the time of writing are much less well known. The interesting thing about all these compounds is that they all result from doping of cupric oxide insulators, and they all contain planes of Cu-O atoms.
Perhaps not entirely by coincidence, high Tc superconductivity thus occurs in a class of materials, namely transition metal oxides, whose electronic ground states are currently least well understood in terms of conventional band theory. It is now generally accepted that the d-shells of transition metals ions (including Cu) have associated with them a fairly large Coulomb repulsion that makes it energetically unfavorable for two holes to be on the same d-shell simultaneously. For Cu, the energy U is estimated to be between 4 and 8 eV. This implies that Cu++ (d9 configuration) is favored, while Cu+++ (d8 configuration) is not. This is in accordance with spectroscopic measurements of these compounds.