Toward a Large-Scale Superconducting Quantum Computer: Error Management and Scalability

点击翻译按钮获取中文摘要

Practical quantum computers are closer to reality than ever before. Among many implementations, trapped ions, semiconducting devices, and superconducting quantum circuits are among the most promising candidates to implement medium- and large-scale quantum computers with ~100-100000 qubits. Superconducting qubits have already been integrated in arrays of ~50 qubits and can in principle be further integrated. However, dissipation phenomena due to imperfect materials or inadequate engineering can cause a qubit to fail and, to date, continue to be a major challenge. Additionally, moving from medium- to large-scale systems will require truly scalable classical qubit wiring and control techniques that are not yet available. The community working on practical quantum computers has evolved dramatically in the last five years. Major ventures such as IBM and Google are investing hundreds of millions of dollars to improve and scale up qubit arrays. In order to be competitive, it is imperative to choose a research plan clinically. The classical integrated circuit industry should serve as an example. Major ventures build the hardware and software required for a complete computing platform; research labs investigate the open scientific and technological challenges and find solutions that are then adopted by the large companies. The long-term vision of my research program is to reduce qubit errors and develop scalable technologies for a large-scale quantum computer. The two main objectives are: O1)Error management: a.Banish dissipation due to imperfect materials by integrating a novel superconducting circuit: The pocketmon resonator. b.Realize a new qubit: The pocketmon qubit. O2)Scalability: a.Develop a new qubit wiring technique for a large-scale quantum computer: Pin-chip bonding for fully vertical interconnects. b.Benchmark pin-chip bonding with qubits. Material imperfections in superconducting on-chip devices are due to the presence of oxidized layers in proximity of the qubits. The pocketmon qubit is a new type of superconducting device where, instead of removing the oxidized layers, we store the qubit energy in a region of space that is only marginally affected by the unwanted layers. Pin-chip bonding is a novel qubit wiring technique that I invented. This technique makes it possible to operate two-dimensional arrays with more than 100000 qubits, permitting to reach any qubit in the array from above instead of laterally, as presently done with traditional techniques. If a square chip contains NxN qubits, accessing the qubits from above allows us to reach N2 qubits instead of just 4N qubits from the sides. Canada in general and, in particular, Waterloo have already created a strong "quantum ecosystem." My research program aims at strengthening this effort in the direction of practical quantum computing technologies. This will generate a strong class of highly skilled Canadian students with expertise in a rapidly blooming field of research.