Millions in funding for brain and quantum research

One of the two projects, named XrayConnectomics, will make intensive use of the Swiss Light Source SLS at PSI. «We want to understand how the nerve cells in the brain are interconnected with each other and how they process information,» says Adrian Wanner, a neuroscientist in PSI’s Laboratory for Nanoscale Biology. To that end, he will be setting up a new experiment station at SLS. In an initial step, he will measure the activity of hundreds of nerve cells and subsequently use the X-ray light of SLS to elucidate the neuronal structure of the corresponding brain tissue, down to its finest branchings. "That will make it possible for us to gain new insights into the brain’s circuit diagram," Wanner says. "The connections between nerve cells in the brain are formed on the basis of experiences and learning processes." It remains unclear, however, what these interconnection patterns look like – networks that form, for example, the basis of so-called working memory. Wanner wants to find this out with his newly approved research project worth around 2.5 million euros.

"The project will generate an enormous amount of experimental data," Wanner says. "This will present a major challenge that can only be overcome by means of artificial intelligence." Wanner hopes the project will yield new insights into how the brain works and suggest potential new approaches for medical therapies. "At present it is not possible to heal neurodegenerative diseases." One reason for this is that we still know too little about the organisational principles of the neuronal networks and about the individual cell types in the brain. "With our project, we want to make a contribution towards changing this situation," says Wanner.

Qubits as the key to new quantum computers

Another ERC grant has been awarded to PSI physicist Alexander Grimm, for his research on quantum bits and his new project COOLCCAT. Quantum bits are the information storage units and thus the foundation of quantum computers, which in certain applications should be far superior to classical computers. At this time, quantum computers exist only as single prototypes. Researchers are faced with the problem that stable quantum bits – also called qubits for short – are difficult to implement in practice. This is precisely the challenge that Grimm’s research addresses.

"Qubits are based on the rules of the quantum world. Here there are not only the two states 0 and 1 of classical bits, but also what we call superpositions of the two," Grimm explains. Thus there are many more possible states that a qubit can assume, an advantage that is also linked to a disadvantage: "Conventional qubits carry only one quantum of energy. That is the smallest possible amount of energy, and as a result they are extremely susceptible to disturbances and thus to errors." Disturbances are ever-present in a quantum computer, because in it the quantum world, with its own rules, encounters the familiar, everyday world of classical physics. So each qubit is connected to the world of classical physics, for example with cables and electronics, since ultimately qubits too need to be written and read out. These interfaces bring disturbances into a quantum computer.

Therefore Grimm wants to create a type of qubit whose behaviour will maintain the greatest possible stability with respect to disturbances. His candidates: oscillator qubits, also known as bosonic qubits. These consist, for example, of an extremely thin and narrow piece of superconducting metal several millimetres long. "In such a wire-shaped component, we can selectively stimulate oscillations of the electromagnetic field. These oscillations exhibit quantum mechanical behaviour and can serve as information carriers," the physicist says.

The oscillations are comparable to those of a pendulum. "You can first move a pendulum to the left and then let it go. Or you can first move it to the right and let go. These two possibilities lead to two oscillations that differ physically in terms of their so-called phase," Grimm explains. If you now wanted to get one such swinging pendulum into the other phase, you would first have to take hold of it briefly. That would be a considerable disturbance. An event like that, especially if it must occur spontaneously, is also extremely improbable in the quantum world. "The fact that we use the phase as an information carrier is what makes our qubits so stable," Grimm says.

With the 2.5 million euros that have now been awarded to him in the framework of the ERC grant, Grimm plans to seek experimental proof that such qubits actually are useful for quantum computers. "In this area, up to now, there have only been initial proof-of-concept experiments. That means we know qubits of this type work in principle, but we now need to show that this is also scalable, and that we can practically combine such qubits to make a quantum computer."

Text: Paul Scherrer Institute