Silicon-based qubits take a big leap forward

Over the last few years, the big question in quantum computing has shifted from "can we get this to work?" to "can we get this to scale?" It's no longer news when an algorithm is run on a small quantum computer—we've done that with a number of different technologies. The big question now: When can we run a useful problem on quantum hardware that clearly outperforms a traditional computer?

For that, we still need more qubits. And to consistently outperform classical computers on complicated problems, we'll need enough qubits to do error correction. That means thousands of qubits. So while there's currently a clear technology leader in qubit count (superconducting qubits called transmons), there's still a chance that some other technology will end up scaling better.

That possibility is what makes several results being published today interesting. While there are differences among the three results being announced, they all have one thing in common: high-quality qubits produced in silicon. After all, if there's anything we know how to scale, it's silicon-based technologies.

Quality issues

The idea of crafting qubits out of silicon has some history, and we've made progress with the technology in the past. That's because making qubits from silicon is relatively easy when using techniques developed for the semiconductor industry. For example, the intentional contamination called "doping" that is used to tweak the properties of silicon could also be used to embed atoms that can act as qubits. Similarly, our ability to place wiring on silicon can be used to make structures that create quantum dots where an individual electron can be controlled.

The best part is that these approaches require very little space to implement, meaning we could potentially squeeze a lot of qubits onto a single silicon chip. That's a big contrast to alternative technologies like transmons and trapped ions, both of which are bulky enough that the companies working with them are already talking about (or even implementing) spreading processors across multiple chips.

The problem so far has been that silicon-based qubits are rather error-prone. Ultimately, we want to use groups of these individual qubits as a single logical qubit that implements error correction. But if errors occur faster than they can be corrected, this won't be possible. And so far, silicon-based qubits are definitely on the wrong side of that error threshold.

High-quality dots

Two papers take a similar approach to improving the performance of qubits based on quantum dots. One is from a group of researchers based at the Delft University of Technology, and the other is primarily from Japan's RIKEN, with some collaborators at Delft. Both groups used silicon with wiring on it to create a quantum dot that trapped a single electron. The spin of the trapped electron was used as the basis for the qubit. And both groups took a similar approach, testing their gate under a wide range of conditions to identify the ones that tended to produce errors and then operating the qubit in a way that avoided those errors.

In the work at Delft, entangling the two qubits was done by manipulating the quantum dots so that the wave functions of the trapped electrons overlapped. After optimizing the use of the hardware, the researchers found that both the single-qubit and two-qubit gate operations have a fidelity rate of over 99.5 percent. That's above the threshold needed for getting the most commonly considered form of quantum error correction to work.

To show that the qubits are actually useful, the researchers use their two-qubit setup to calculate the ground state energy of molecular hydrogen. This calculation is relatively easy to do on classical hardware, so the results could be checked.

The RIKEN group did something similar and generally found that speeding up operations had a major effect on error rates. Again, managing this problem produced gates with a fidelity of 99.5 percent, well above the threshold needed for error correction. To show that the gates worked, the team implemented a couple of quantum computing algorithms and showed that they were completed with a success rate in the area of 97 percent.

Going nuclear

Electrons aren't the only things in silicon with a spin; given the right isotope, the nucleus of the atoms can have a spin as well. And that's what a group from Australia's University of New South Wales focused on in a separate paper.

Atomic nuclei are largely shielded from the environment by the shells of electrons that surround them. This makes them relatively stable repositories of quantum information, which tends to decay due to environmental interactions. Indeed, nuclear spins are often viewed as a great way of creating quantum memory. But their isolation also makes it harder to interact with the spins in the first place, which can make manipulating nuclear qubits a challenge.

To work around this issue, the team embedded phosphorous atoms in silicon and then entangled the nucleus of two of those atoms with an electron. For a single nucleus, this resulted in impressive fidelity, reaching up to 99.95 percent. This figure dropped once operations and readout were included, but it still stayed above 98.9 percent. The problems here seem mostly related to the shared electron, so that is something that can potentially be addressed by focusing on its behavior.

To show this idea worked, the team applied a series of operations to the qubit and then reversed the entire process. It ended up back in its initial state about 90 percent of the time.

Future prospects

So what does it mean to have two functional qubits at a time when a competing technology offers over 100? That all depends on what comes next. It should be relatively easy to create larger arrays of the sorts of qubits used in these experiments and see how having nearby devices influences their behavior. It should be equally easy to see whether connections can be established across the array quickly enough that larger algorithms can be implemented before fidelity issues reassert themselves.

Usefully, both of the papers point out how to characterize these devices to optimize performance. But there's a great deal of engineering needed before we even get to those measurements.

The group working on nuclear spins, in contrast, is already looking at different ways of scaling. The researchers note that heavier elements (like tin or iodine isotopes) provide additional accessible spin states, such that six qubits could be stored in just two atoms. These spin states can also be controlled electronically, which would considerably simplify the system.

In these systems, things like entanglement can potentially be mediated by shifting the electron to different pairs of nuclear qubits. And in some circumstances, the electron itself could contribute another qubit toward the total needed for error correction. Again, all this looks very promising, but we'll need to see how quickly the system's potential to scale translates into actual scaling.

Nature, 2022. Summary with links to papers: 10.1038/d41586-022-00047-0  (About DOIs).