Donor atoms in silicon are a versatile platform for experiments in quantum information processing, as well as quantum foundations. The electron [1] and nuclear [2] spin of a 31P donor were the first qubits demonstrated in silicon, and went on to become among of the most coherent qubits in the solid state, with coherence times exceeding 30 seconds [3], and quantum gate fidelities approaching 99.99% [4].
While magnetic resonance is the standard method to achieve coherent spin control, in recent years we have demonstrated or predicted that similar control can be achieved using oscillating electric fields, or even acoustic waves.
Using magnetic resonance, we have demonstrated an exchange-based 2-qubit gate for electron spins [5], in a device where we implanted a high dose of 31P donors. Future experiments will focus on using deterministic, counted single-ion implantation, for which we have recently demonstrated the capability to detect an individual ion with 99.85% confidence [6]. With nuclear spins, we have achieved the landmark result of universal 1- and 2-qubit logic operations with >99% fidelity, and prepared a 3-qubit GHZ entangled state with 92.5% fidelity [7].
The ability to control the qubits with electric fields would provide great advantage in terms of addressability and compatibility with control and readout electronics. We have thus invented a new type of qubit, called “flip-flop” qubit, composed of the anti-parallel spin states of the electron and nuclear spin. We have demonstrated its coherent electrical control of using microwave electric fields [8], and verified the microscopic mechanism that enables the electrically-driven spin resonance.
Heavier Group-V donors atoms possess a high nuclear spin quantum number and a nonzero electric quadrupole moment. In 1961, Bloembergen predicted that nuclear spin transitions could be induced by the electrical modulation of nuclear quadrupole couplings [9], but the experimental attempts were soon abandoned due to the very weak coupling strength in bulk materials. We have serendipitously re-discovered the phenomenon of nuclear electric resonance in our nanoscale device, where the local electric fields are extremely strong, and used it to coherently control a single 123Sb nuclear spin [10].
Finally, the application of local lattice strain is predicted to enable nuclear acoustic resonance [11] of quadrupolar nuclei such as 123Sb, and to provide a unique atomic-scale probe of local lattice strain in silicon quantum electronic devices.