Silicon-based general-purpose quantum computer sets new world record for number of quantum bits

Researchers at the QuTech Quantum Computing Research Center in the Netherlands have engineered a record number of six silicon-based spin quantum bits in a fully operational array, according to a paper published today in Nature, "Universal control of six-qubit quantum processors in silicon" [1]. Importantly, with a new chip design, an automatic calibration procedure, and new methods for initialization and readout of quantum bits, these quantum bits can operate with low error rates: an advance that will contribute to the realization of a scalable silicon-based quantum computer.

It should be noted that the world's first atomic-scale quantum integrated circuit, which contains 10 quantum dot quantum bits, was developed a few months ago by Australian quantum computing manufacturer SQC in collaboration with Professor Michelle Simmons' team at the University of New South Wales, but it is not a gate-based general-purpose quantum computer. The previous record for a general-purpose quantum computer based on silicon spins (or semiconductor quantum dots) was four quantum bits.

01An important step for silicon-based fault-tolerant quantum computers

Different materials can be used to produce quantum bits, but no one knows which material will turn out to be the best choice for building large-scale quantum computers. So far, only smaller-scale silicon quantum chips have demonstrated high-quality quantum bit operation. Now, researchers from QuTech, led by Professor Lieven Vandersypen, have produced a six-quantum-bit chip in silicon that runs at a low error rate. This is an important step toward a fault-tolerant quantum computer using silicon.

To create quantum bits, individual electrons are placed in a linear array of six "quantum dots" spaced 90 nanometers apart. Quantum dot arrays are made in silicon chips with a structure very similar to that of a transistor: a common component in every computer chip. A quantum mechanical property called "spin" is used to define a quantum bit whose orientation defines the logical state of 0 or 1. The team uses finely tuned microwave radiation, magnetic fields and electrical potentials to control and measure the spin of individual electrons and make them interact with each other.

The quantum processor with six quantum bits described in this experiment. These quantum bits were created by adjusting the voltages of the red, blue and green lines on the chip. The structures known as SD1 and SD2 are extremely sensitive electric field sensors, which can detect the charge of even a single electron. These sensors coupled with an advanced control scheme allowed the researchers to place individual electrons at positions labeled (1)-(6) and then operate them as quantum bits.

Single quantum bit gate characterization. a) Rabi oscillations for each quantum bit, spin fraction is the spin fraction of quantum bits Q2-Q5 and the spin fractions of quantum bits Q1 and Q6. b) Quantum bit frequencies for each of the six quantum bits. c) Rabi frequency of each quantum bit as a function of the applied microwave power. d) Random benchmark test results for each quantum bit, the reported fidelity is the average single quantum bit gate fidelity, and the uncertainty (2σ) is calculated using the fitted covariance matrix. e) Table showing the phase shift time T2*, the echo decay time (T2h) and the visibility (vis.) for each quantum bit.

Two-qubit gate characterization. a) Quantum circuit for measuring controlled phase (CPhase) oscillations between a pair of quantum bits. b)-f), measured spin probabilities as a function of the total evolution time of adjacent quantum bit pairs Q1-Q2(b), Q2-Q3(c), Q3 -Q4(d), Q4-Q5(e) and Q5-Q6 as a function of the total evolution time. f) Used for different virtual barrier gate voltages 0 and 1 corresponding to exchange off and at maximum. g) Measured maximum exchange coupling for each quantum bit pair and the corresponding remaining exchange for the other pairs. coupling. h) Exchange coupling versus virtual barrier gate voltage for all quantum bit pairs. j) Pulse shape of the exchange amplitude across a gate voltage pulse used for the CZ gate and the corresponding pulse shape converted to gate voltage.

Experimental results show that all average single-quantum bit gate fidelities are between 99.77 ± 0.04% and 99.96 ± 0.01%: this indicates that even in this extended quantum bit array, the researchers achieved high-fidelity single-quantum bit control. The two-quantum bit gate is achieved by pulsing a (virtual) barrier gate between neighboring points while remaining at the point of symmetry. By creating single- and double-quantum bit control on six-quantum bit arrays, scientists continue to create and quantize pairwise entanglement across quantum dot arrays as a measure of the quality of quantum bit control.

Bell State Chromatography Scanning

Three-quantum-bit GHZ state laminar scan

The results show that the density matrix obtained from measurements on a six-quantum dot array has a fidelity of states ranging from 88% to 96%: this is significantly higher than the Bell state fidelity of 78% to 89%. In scaling the quantum dot system to a record number of quantum bits, the team achieved Rabi oscillations per quantum bit with 93.5-98.0% visibility, implying high readout and initialization fidelity.

Stephan Philips, first author of the paper, explains, "Today's quantum computing challenge consists of two parts, developing quantum bits of sufficient quality, and developing an architecture that allows one to build large systems of quantum bits. Our work is a combination of both. And because the overall goal of building quantum computers is a huge endeavor, we are contributing in the right direction."

The spin of an electron is a fragile property. Small changes in the electromagnetic environment can cause fluctuations in the spin direction, which increases the error rate. the QuTech team built on their previous experience with quantum dot engineering with a new way to prepare, control and read the spin state of an electron: using this new arrangement of quantum bits, they can create logic gates and entangle systems of two or three electrons as needed.

02Silicon quantum devices hold promise for industrial production

Quantum arrays of more than 50 quantum bits have been produced using superconducting quantum bits. However, it is the global availability of silicon engineering infrastructure that holds the promise of easier transfer of silicon quantum devices from research to industrial production. Silicon presents certain engineering challenges, and prior to this work by the QuTech team, only arrays of up to 3 quantum bits could be engineered in silicon without sacrificing quality.

The requirements of having a large number of quantum bits and operating with high fidelity are often conflicting. "This paper shows that with careful engineering it is possible to increase the number of spin quantum bits in silicon while maintaining the same precision as a single quantum bit. The key building blocks developed in this study can be used to add more quantum bits in the next iterations of the study." Co-author Dr. Mateusz Madzik said [2].

Prof. Vandersypen said, "In this study, we have driven the number of quantum bits in silicon and achieved high initialization fidelity, high readout fidelity, high single-quantum-bit gate fidelity, and high two-quantum-bit state fidelity, but what really stands out is that we have collectively demonstrated, in a single experiment, on a record number of quantum bits all of these properties."

Reference links：

[1]https://www.nature.com/articles/s41586-022-05117-x

[2]https://phys.org/news/2022-09-full-six-qubit-quantum-processor-silicon.html