Transistors are the nerve cells of the information age.
These tiny switches control the flow of the electrical signals in everything from smartphones to supercomputers. If the signal is allowed to pass through a gate on the chip, it is represented as a 1; if not, it is a 0. Those bits of 1s and 0s form the basis of all of today’s computing.
These transistors are so small that billions of them can fit on a computer chip the size of your fingernail. They are measured these days on the scale of a few DNA strands.
Unfortunately, they are hitting their limit. If they are made even smaller, quantum effects will play havoc with the ability of transistors to control the flow of the current. That is why some people think that Moore’s law, the postulate that the number of quantum transistors on a chip can double every two years, is at an end.
But if computing is to take a quantum leap into the future, these already mindbogglingly tiny transistors must become tinier still. A quantum computing age will need quantum transistors.
This is the challenge that Eva Dupont-Ferrier’s lab at Université de Sherbrooke in Quebec is tackling.
A few years ago, Sherbrooke established an agreement to develop quantum transistor technology with Interuniversity Microelectronics Centre (IMEC), an international research and development organization that has been pioneering some of the smallest and most sophisticated nanoelectronics and digital technology in the world today.
Sherbrooke’s Institut Quantique is part of the recently established Quantum Colab (Quantum Colaboratory) research and development environment that also includes the Stewart Blusson Quantum Matter Institute at the University of British Columbia and the Transformative Quantum Technologies (TQT) technology vector of the Institute for Quantum Computing at the University of Waterloo.
Those institutions and their industry partners are working together to leverage Canada’s facilities and experts in order to bring quantum technologies from the lab bench to commercialization.
Dupont-Ferrier is using the Quantum Colab facilities at Sherbrooke in order to employ what are known as dopants on silicon chips to help generate property known as spin that can be used as quantum bits or qubits in quantum computing.
Just as a current that is on or off can be translated into bits of 0s and 1s for today’s computing, the spin state — up and down — would be used as a quantum bit, or qubit, to do quantum computing.
“If we want to make this transistor quantum, we can use the spin as the quantum bit,” Dupont-Ferrier says.
Dopants are commonly used in making computer chips. A piece of pure silicon would normally not conduct electricity very well, but the semiconductor industry solves this problem by inserting dopants which might be atoms of arsenic, boron or some other substance that helps to conduct the current by creating a different charge within the silicon matrix to help transport the electrons along the transistor paths.
The problem for the microelectronics industry is that as transistors continue to shrink, not only can quantum effects disrupt the flow of the signal, but also, the dopants can become misaligned, and this too ruins the ability of the transistors to control the on/off flow of the current.
By reimagining the transistor, Dupont-Ferrier’s lab is turning that disadvantage into an advantage. Her lab is making use of the dopants themselves for quantum computing.
“We can do two things. If we want to make this transistor quantum, we can use the spin of the dopant as a qubit. Or, more simply, what we can do is confine the charge in the transistor. So instead of having a current that is flowing, you confine the electrons in the channel of the transistor and then you use the spin of these electrons as the qubits,” she says.
The lab is making structures known as quantum dots, which can be imagined as being like puddles that trap the electrons so that their spin within a confined region can be utilized for computing. “In our case, we managed to have two traps for electrons which would be equivalent of the channels of two transistors connected in series,” Dupont-Ferrier adds.
In doing this, her team is creating a system that puts quantum computing capability on a silicon chip. “Basically, we are using the same devices that are used for state-of-the-art technologies, but now, we are using the spin of the electrons instead of just the charge,” Dupont-Ferrier says. “It’s actually a way of taking the size limitation in classical electronics and turning it into an advantage,” she adds.
The research is progressing well. “We have managed to have electrons confined in two quantum dots and then control the coupling between these two and we have measured this. The next step to put them in the magnetic field and look at the spin in these two dots, or two qubits.”
The other aspect of the research involves doing experiments on state-of-the-art transistors from industry partners by putting them into Sherbrooke’s ultra-cold cryogenic equipment and to see if, at low enough temperatures, those devices can be controlled and ultimately used for quantum computing. The experiments are yielding good results. “They are working with much better performance at low temperature.”
Dupont-Ferrier says it is exciting that engineers and physicists in academia and industry can now talk about an entirely new way of thinking about transistors. “They have their own path in the microelectronics industry, but now, we can come in as physicists and say: ‘You can look at your device differently. You can do quantum information.’ It’s really a change in the way of thinking.”
Collaboration with other academic institutions and with industry partners is the key to Canada’s quantum success, she adds. “The more you exchange ideas, the more ideas you can have,” she says of the power of collaboration. “Canada is a small country in terms of population but when it comes to quantum research, it is a strong country. We have a lot of collaboration, and that is really helping the quantum science here,” she says.