Osaka team adds an extra quantum dot to make spin qubits faster and stabler

[Crossing]

Researchers in Japan have found a way around a common trade off in spin qubits. Asia Research News, republishing a University of Osaka release, reports on work by Takafumi Fujita's team that adds a third, auxiliary quantum dot to a flopping mode spin qubit in gallium arsenide

Spin qubits usually face a choice between fast control and stability. Strong coupling allows quick operations but makes the qubit more sensitive to voltage noise. The team's technique, which they call orbital band engineering, uses the extra dot to reshape the electron's energy levels, giving a wider operating region that is less sensitive to noise while keeping strong coupling

Machine learning feedback was also used to correct slow drifts in qubit frequency. With that feedback, the estimated dephasing time rose from about 52 nanoseconds to 580 nanoseconds

Qubit frequencies reached more than 100 MHz, and up to 271 MHz with different tuning. Gates lasting 4 nanoseconds achieved 99.56 percent fidelity with feedback, rising to 99.72 percent with extra pulse optimisation. Fujita said adding the auxiliary dot introduced an extra degree of freedom for tuning the qubit. The paper is published in npj Quantum Information

Breaking a fundamental trade off with a clever design tweak is always satisfying to see. The work so far is in gallium arsenide, while silicon and germanium are the main targets for scalable spin qubits. If it transfers to those materials, it could be very useful

Will tricks like this help spin qubits catch up with other platforms? And how important is machine learning becoming in qubit control?

23571www.asiaresearchnews.com