UNSW scientists have shown that their pioneering single atom technology can be adapted to building 3D silicon quantum chips – with precise interlayer alignment and highly accurate measurement of spin states.
The 3D architecture is considered a major step in the development of a blueprint to build a large-scale quantum computer.
UNSW researchers at the Centre of Excellence for Quantum Computation and Communication Technology (CQC2T) have shown for the first time that they can build atomic precision qubits in a 3D device – another major step towards a universal quantum computer.
The researchers, led by 2018 Australian of the Year and director of CQC2T, Professor Michelle Simmons, have demonstrated that they can extend their atomic qubit fabrication technique to multiple layers of a silicon crystal – achieving a critical component of the 3D chip architecture that they introduced to the world in 2015.
The group is the first to demonstrate the feasibility of an architecture that uses atomic-scale qubits aligned to control lines – which are essentially very narrow wires – inside a 3D design.
What’s more, team members were able to align the different layers in their 3D device with nanometer precision.
Simmons said the 3D device architecture is a significant advancement for atomic qubits in silicon.
“To be able to constantly correct for errors in quantum calculations – an important milestone in our field – you have to be able to control many qubits in parallel.
“The only way to do this is to use a 3D architecture, so in 2015 we developed and patented a vertical crisscross architecture. However, there were still a series of challenges related to the fabrication of this multi-layered device. With this result we have now shown that engineering our approach in 3D is possible in the way we envisioned it a few years ago,” said Simmons.
In this paper, the team has demonstrated how to build a second control plane or layer on top of the first layer of qubits.
CQC2T researcher and co-author, Dr Joris Keizer, said it’s a highly complicated process, but in very simple terms, the team built the first plane, and then optimised a technique to grow the second layer without impacting the structures in first layer.
“In the past, critics would say that that’s not possible because the surface of the second layer gets very rough, and you wouldn’t be able to use our precision technique anymore – however, in this paper, we have shown that we can do it, contrary to expectations,” said Keizer.
The team members also demonstrated that they can then align these multiple layers with nanometer precision.
The researchers were also able to measure the qubit output of the 3D device single shot, with a single, accurate measurement, rather than having to rely on averaging out millions of experiments.