For Science Watch, Zhi-xun Shen of Stanford University, who is a coauthor of #3 and 7, offered the following introduction to the physics of cuprate superconductors. "As you cool down a conventional superconductor it acts as a metal until it reaches the critical temperature T_c, below which it rapidly goes superconducting. However, the underdoped cuprates investigated in these papers seem to be preparing for the transition well before they reach T_c." Doping refers to the dramatic changes that occur in these substances when a small fraction of electrons is chemically removed, so that the holes can move freely and the insulator converts into a metal. In short these superconductors have complex heterostructures in which itinerant charge carriers and polarons coexist.

   In a classical superconductor an energy gap opens up at T_c, reflecting the energy needed to break the Cooper pairs of electrons with opposite spins participating in superconductivity. The "gap" means that no conducting particles have the particular energy associated with the gap. Classically the energy gap opens up only at T_c, but what researchers see for copper oxide superconductors is the energy gap opening up way above the superconducting transition, at 200K for samples with a T_c of 80K. This is serious new physics, which is under intense investigation because any kind of crossover has much to say. The natural interpretation is that Cooper pairs form far above T_c in the cuprates. If this is correct it means that the superconducting transition in copper oxide superconductors is due to something other than the presence of Cooper pairs, such as a phase coherence transitin. This would signal that the mechanism is different from the BCS theory that applies to classical superconductors.

   High-T_c superconductors are not like normal metals, where the Fermi-liquid model holds. Bi2212 deviates far from a Fermi liquid. Of papers #3 and 7 which report that the energy gap occurs well above T_c, Zhi-xun Shen says, "While the full implication of this observation is still under intense observation, a natural interpretation is that the Cooper pairs are formed well above T_c. If this holds then the critical step for the superconducting transition in the cuprates is the formation of the Cooper pair, rather than something else such as a phase coherence transition. This idea is further supported by our finding that the size of the superconducting gap in the underdoped regime does not scale with T_c"

   The Stanford group has recently confirmed that the gap collapses very rapidly in higher doping regimes. Furthermore they have seen the same behavior in another cuprate system, Bi2201. The latest unpublished results report that the gap is seen in La-Sr-Cu-O, the system investigated in newcomer #6 which looks at the lattice distortions at optimum doping. Antonio Bianconi, Rome University, Italy, and colleagues report on structural measurements at the mesoscopic scale. According to Bianconi, "It is getting clear that the high T_c superconductors belong to a family showing a dynamic phase segregation of charge carriers." This idea of phase segregation connects the cuprate perovskites showing superconductivity with their cousins the manganese perovskites which exhibit giant magnetoresistance (see Science Watch, 8[5]:7, September/October 1997. Evidence is now mounting for a connection in the cuprates between antiferromagnetism and superconductivity, a convergence that will be explored in conferences slated for June 1998 to be held in Rome and Erice, Italy.