Quantum computation (QC) features known examples of hardware acceleration for specific problems but is challenging to realize due to its susceptibility to small errors from noise or imperfect control. The fault tolerance principles may enable computational acceleration with imperfect hardware, but they place strict requirements on the character and correlation of errors.
For many qubit technologies, some challenges to achieving fault tolerance can be traced to correlated errors arising from the need to control qubits by injecting microwave energy matching qubit resonances. HRL Laboratories, LLC, has published the first demonstration of universal control of encoded spin qubits. The experiment demonstrated universal control of their encoded qubits, which means the qubits can be used successfully for any kind of quantum computational algorithm implementation.
This newly emerging approach to quantum computation uses a novel silicon-based qubit device architecture that traps single electrons in quantum dots. Three of these single electron spins have energy-degenerate qubit states governed by nearest-neighbor contact interactions that swap neighboring spin states in part.
Since the experiment showed that their encoded qubits could be controlled universally, any quantum computational technique could be implemented effectively using the qubits. The encoded silicon/silicon germanium quantum dot qubits use three electron spins and a control technique whereby voltages applied to metal gates partially swap the directions of those electron spins without ever aligning them in any one order.
During the demonstration, dozens of these carefully calibrated voltage pulses were applied one after the other in close proximity for a few millionths of a second.
The isotopically enriched silicon used, the all-electrical control of partial swap operations with low crosstalk, the configurable insensitivity of the encoding to specific error sources, and the quantum coherence they provide all work together to provide a vital pathway towards scalable fault tolerance and computational advantage, which are crucial steps towards a commercial quantum computer.
HRL scientist and first author Aaron Weinstein said, “Beyond the obvious challenges of design and fabrication, a lot of robust software had to be written, for example, to tune up and calibrate our control scheme. Significant effort was placed into developing efficient, automated routines for determining what applied voltage led to what degree of partial swap. Since thousands of such operations had to be implemented to determine error levels, each one had to be precise. We worked hard to get all that control working with high precision.”
HRL group leader and co-author Mitch Jones said, “This was very much a team effort. The enabling work of talented control software, theory, device growth, and fabrication teams was crucial. Additionally, many measurements of devices were needed to understand enough of the internal physics and to develop routines to control these quantum mechanical interactions reliably. This work and demonstration is the culmination of those measurements, made all the better by working alongside some of the brightest scientists I’ve met.”
Thaddeus Ladd, HRL group leader and co-author said, “It is hard to define what the best qubit technology is, but I think the silicon exchange-only qubit is at least the best balanced. Real challenges remain in improving error, scale, speed, uniformity, crosstalk, and other aspects, but none of these requires a miracle. For many other kinds of qubits, at least one aspect still looks hard.”
If scaled up, quantum computers would be different from conventional supercomputers in that they would utilize the fragile property of quantum physics known as quantum entanglement to execute some calculations that would typically take years or decades on traditional computers. The simulation of the behavior of big molecules is one such computation among many conceivable applications.
Journal Reference:
- Aaron J. Weinstein et al., Universal logic with encoded spin qubits in silicon, Nature (2023). DOI: 10.1038/s41586-023-05777-3