d-Wave superconductors and quantum computers A.M. Zagoskin D-Wave Systems Inc. and Physics and Astronomy Department, The University of British Columbia, Vancouver, Canada Abstract Quantum algorithms promise enormous speed up in dealing with certain problem classes, but only in large enough quantum computers (about thousand qubits). Therefore the scalability of solid state devices make solid-state-based qubit prototypes an attractive choice. On the other side, it is difficult to preserve quantum coherence in such devices, in the presence of macroscopic number of degrees of freedom. The coherent ground state and suppression of low-energy excitations in superconductors could help solve the problems. Recent experiments on several mesoscopic qubit pro- totypes, using conventional superconductors, demonstrated coherent quantum behavior. Here I review the current results on qubit prototypes based on high-T c superconductors, and discuss the open questions and the further directions of research. Ó 2001 Elsevier Science B.V. All rights reserved. Keywords: Qubit; Scalability; Time-reversal symmetry breaking The interest attracted by quantum computing within the research community, as well as among a wider audience, grows remarkably fast and strong for a field, which currently has only a few useful applications (quantum Fourier transformation, Shor’s factorization algorithm, Grover’s search algorithm [14]), and only embrional stage tech- nology. The reason it is touted as ‘‘the’’ technology for this millennium [3] is that, if successful, quan- tum computing will be a qualitative breakthrough, not just a better way of solving certain important problems, but the only way to do so. The conditions, which must be met by any quantum computer prototype (DiVincenzo’s ‘‘five criteria’’ [6]), require both (A) an ability to ini- tialize, manipulate, and read qubits (or pairs of qubits), which implies a sufficient coupling of in- dividual qubits to each other and the outside world, and (B) to preserve coherent unitary evo- lution long enough to allow the algorithm to run, thus requiring a minimal coupling of qubits to anything. (With quantum error correction, the latter means decoherence time in excess of 10 4 gate application times [6].) The scale necessary to begin realizing the po- tential of quantum computers as such is about a thousand quantum bits. This indicates that, as with conventional computers, we should be look- ing at solid state (or at least condensed matter) devices. The essential scalability of such systems is bought for a price of having a huge number of low-energy excitations in condensed matter, which would couple to qubits and destroy quantum co- herence. Superconductors are a natural choice due to coherence of their ground state and absence or suppression of dangerous low-energy excitations. Physica C 368 (2002) 305–309 www.elsevier.com/locate/physc E-mail address: zagoskin@dwavesys.com (A.M. Zagoskin). 0921-4534/01/$ - see front matter Ó 2001 Elsevier Science B.V. All rights reserved. PII:S0921-4534(01)01186-8