A false-color scanning electron micrograph of the qubit structure used in this study.  The area imaged is approximately 1,500 nanometers in diameter.  For comparison, a human hair is between 50,000 and 100,000 nanometers wide.

A false-color scanning electron micrograph of the qubit’s structure showing the “moving wells” developed by UW-Madison researchers to improve the accuracy of quantum computers. The area imaged is approximately 1,500 nanometers in diameter. For comparison, a human hair is between 50,000 and 100,000 nanometers wide. The University of Wisconsin-Madison

Conventional computers rarely make mistakes, largely thanks to the digital behavior of semiconductor transistors. They are on or off, corresponding to the ones and zeros of the classic bits.

On the other hand, quantum bits, or qubits, can be zero, one, or an arbitrary mixture of the two, allowing quantum computers to solve certain calculations that are beyond the capability of any classical computer. A complication with qubits, however, is that they can occupy energy levels outside of the one and zero of computation. If these additional levels are too close to one or zero, errors are more likely to occur.

A false-color scanning electron micrograph of the qubit structure used in this study.  The area imaged is approximately 1,500 nanometers in diameter.  For comparison, a human hair is between 50,000 and 100,000 nanometers wide.

Marc Friesen

“In a typical computer, all aspects of a transistor are super-uniform,” says UW-Madison Scientist Emeritus Mark Friesen, author of both papers. “Silicon qubits are in many ways like transistors, and we’ve gotten to the point where we can control the properties of the qubit very well, except for one.”

This property, known as valley splitting, is the buffer between computational energy levels at one zero and additional energy levels, helping to reduce quantum computational errors.

In two papers published in Nature Communications in December, researchers from the University of Wisconsin-Madison, the University of New South Wales, and TU-Delft showed that altering the physical structure of a qubit , known as a silicon quantum dot, creates enough valley division to reduce miscalculations. The results overturn conventional wisdom by showing that a less perfect silicon quantum dot can be beneficial.

In the past, the most common way to fabricate qubits was to embed the quantum dot in a layer of pure silicon, called a silicon well, and then sandwich this layer between two layers of silicon-germanium with sharp boundaries between layers. In these new studies, the semiconductor qubits are fabricated in the same way, with a layer of silicon located between two layers of silicon-germanium. New is the addition of germanium to the silicon layer itself.

Portrait of Bob Joynt

Bob Joynt

“Everyone always thought that the only thing you shouldn’t mess with in qubit design was pure silicon,” says Bob Joynt, professor of physics at UW-Madison. “And we decided, well, let’s play around with that a bit.”

The researchers modified the design by intentionally adding germanium to the silicon layer in a slightly different way. Like so many successes in physics, theirs started with theoretical calculations. Joynt asked what would happen if the concentration of germanium “wriggled” throughout the well, in evenly spaced waves with peaks and valleys. Friesen’s group, who also contributed to the accompanying TU-Delft article, noted that some germanium is still pouring into the well, even when researchers try to keep it out. They asked what would happen if low levels of germanium were randomly sprinkled into the well.

“Theoretical calculations have shown unambiguously that it’s better to include germanium, and that it’s better to stir than not to stir,” says Mark Eriksson, professor of physics at UW-Madison, whose group tested the new qubits. with agitated wells.

Portrait of Mark Erikson

Mark Erikson

After confirming that the stirred wells did not significantly alter the electronic properties of the quantum dot, Eriksson’s group measured the size of the energy buffer, or valley split, in these new structures. For a quantum dot embedded in a swinging well, the theory predicted an increase of 20 microelectronvolts. But in the lab, the largest valley split measured was nearly 250 microelectronvolts. Additionally, the division of the valley changed when the quantum dot was moved to new locations in the well, where the atoms making up the ripples were in different locations. The hustle and bustle of the concentration was clearly not the only factor affecting the valley split – the individual atoms mattered too.

This fact turned out to be the key: the atoms in the swinging well were distributed somewhat randomly. The introduction of this random distribution into numerical simulations made it possible to explain with precision the variations observed in the division of the valleys. In the second paper, Friesen and co-workers at TU-Delft confirmed that even uniformly distributed random germanium in the well leads to greater valley splitting on average than when no germanium is present.

“Contrary to many people’s fears, germanium in wells is a very good idea,” says Eriksson. “If you sprinkle it completely randomly with no focus swings, you’ll do pretty well. If you sprinkle it with focus swings, you’ll do better.

Both qubit structures have been filed for patent protection by WARF, and the UW-Madison team is actively working on designs that further improve the valley split.

“It’s possible that when quantum computers are made from silicon/silicon-germanium qubits, they will want to have germanium in the wells,” says Joynt. “But we don’t know yet. We are just getting started.

At UW-Madison, Thomas McJunkin, Benjamin Harpt, Yi Feng, Merritt Losert, JP Dodson, Michael Wolfe, Don Savage, Max Lagally, and Sue Coppersmith also contributed to this work. Both studies were funded by the army research Office, a branch of the Army Research Laboratory of the US Army Combat Capabilities Development Command (W911NF-17-1-0274).

Source link

Leave A Reply