Superconducting-nanowire-based memories

A persistent current in a superconductor provides a nearly perfect physical token for data storage. It is nonvolatile (so long as the loop remains cold), and can be read out precisely, thus enabling digital (and perhaps even multi-level) electronic memory. However, the quantisation of flux in units of Φ_o, means that relatively large inductors (on the order of 10s of μm's long) are typically required for such systems, somewhat limiting their potential. The solution until recently was to use Josephson junctions as a compact source of inductance, but even this approach resulted in memory-cell dimensions at the many-μm² level. Recently, however, high-quality thin superconducting films with long penetration depths (and thus high kinetic inductance) have become available. These materials enable much smaller persistent-current loops without requiring the use of a Josephson junction. Novel nanowire-based electronic devices, also not using Josephson junctions, have also emerged that can be used to read out the nanowire state. With this, working single-cell memories with negligible bit-error rates have been realized. I will describe these developments. I will also present a new family of novel superconducting electronic elements that can be used with these memories to enhance their capabilities and function.