Robust Multi-Qubit Operations for Donors in Silicon
Abstract
This Proposal, in response to the U.S. Army Research Office Broad Agency Announcement for Qubits in Silicon (W91 INF -16-R-0018), describes the work plan and milestones for the development of a silicon quantum computer where information is encoded in the electron/nuclear spin states of individual 31P atoms, introduced in the silicon crystal by means of counted single-ion ion implantation. This work continues and expands part of the research supported in the period 2012 - 2016 by the Quantum Computer Concept Maturation project "Solid-state quantum computer in silicon" (W-91 INF-13-1 -00 24), which resulted in the first single-atom electron spin qubit in silicon [Nature 489,541 (2012)], the first 3IP nuclear spin qubit [Nature 496,334 (2013)], the record coherence times for single qubits in the solid state(9.56 seconds for the electron spin, 35seconds for the nucleus) [Nature Nano. 9, 986(2014)], the demonstration of Kane s "A-gate" [Science Advances 1, e1500022 (2015)] and the violation of Bell s inequality [Nature Nano. 11,242 (2016)]. Implanted 31P donors in isotopically enriched 28Si have already demonstrated outstanding 1-qubitgate fidelities (99.94%for the electron, 99.99% for the nucleus).This proposal aims to address and resolve what we consider to be the two key obstacles to the demonstration of scalable, high-fidelity multi-qubit operations for donors in silicon: 1) the need for exchange interaction to mediate 2-qubit logic gates; 2) the randomness in number and location of the ion-implanted 31P donors. Exchange-based 2-qubit gates will be pursued in the initial stages of the project, implementing a scheme we developed to make them robust against donor placement uncertainties. For the longer term, to avoid the use of exchange interactions altogether, we will demonstrate a new donor-based quantum bit, called the "flip-flop" qubit, where quantum information is encoded in the combined electron-nuclear states of the donor. This qubit can be driven electrically, and has an associated electric dipole, which introduces a long-range dipole-dipole coupling between two flip-flop qubits. This removes the need for the donors to be placed at ~10 nm distance for the exchange coupling to mediate 2-qubit interactions. For the scalable and reproducible fabrication of implanted donor-based quantum computers, we will develop and apply a novel setup for deterministic, counted single-ion implantation. This system will include on-chip P-I-N detectors integrated with low-noise electronics at room temperature, and a dedicated AFM nanostencil to achieve a lateral placement accuracy better than 10 nm. The final deliverable of this proposal is a device where two deterministically-implanted 31P donors are individually operated through electrical microwave excitation, and coupled through long-range (~200 nm) electric dipole interaction. We target 1-qubit and 2-qubit gate fidelities in excess of 99%, as predicted by error models based upon existing charge noise data. This demonstration will establish the suitability of ion-implanted 31P donors for large-scale, fault-tolerant quantum computers in silicon. The project will be undertaken by PI Morello and Co-PI Dzurakat UNSW Australia, focusing on the coherent multi-qubit operations and the development of nanofabricated Si-MOS structures for qubit control and on-chip ion detectors. Co-PI Jamieson at the University of Melbourne will lead the effort on the counted single-ion implantation.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Sep 11, 2018
- Source ID
- W911NF1710200
Entities
People
- Andrea Morello
Organizations
- Army Contracting Command
- National Security Agency
- University of New South Wales