Thomas Signamarcheix, why did the start-up, C12, set up in January 2020 and from the physics laboratory at the École Normale Supérieure in Paris, seek support from the CEA in Grenoble?
The first short-term goal of C12 is to make quantum chips, which will later be used in quantum computers. It is a technology that they have developed on their own. It is based on a carbon nanotube, carbon 12, hence the name of the company, C12. They claim to make quantum bits (Qubits) – a Qubit is the building block for quantum – based on this carbon nanotube, which is placed on top of a silicon chip. The chip makes it work.
The C12 team made its first demonstrations under laboratory conditions. It then wanted to move on to more mature, more standard manufacturing technologies in the world of microelectronics. She came to visit us because we in France, as an R&D operator with such a production capacity, are a bit alone.
Do you have the capacity to produce on an industrial scale?
We have a large prototype capacity, including a large clean room of 11,000 m2 where silicon chips are made. It’s colossal in France and in Europe! When ideas are born in laboratories, we try to reach a level of maturity that is credible to the industry. All our manufacturing processes are the same as in the industry. We buy the same equipment. We take the risk, for example, by damaging certain machines. We standardize silicon wafers that contain a lot of chips, we reach levels of quality, yield … Our goal is to reach this level of maturity until the industrialist decides.
Is this work carried out only at CEA Grenoble, or is the Saclay plant also affected?
Semiconductor manufacturing is in Grenoble. At Saclay, we have several options in the form of algorithms and chip programming or in the form of fundamental research in other quantum technologies such as superconductors, for example.
How many CEA employees are involved in this partnership with C12?
For this partnership, the collaboration aims to remove locks, one by one. Depending on the locks, we set a more or less strong intensity. On average, this collaboration mobilizes the equivalent of four to five researchers. This can triple tomorrow depending on which lock we manage or will not lift.
How many locks do you need to lift?
We know we have about ten interdependent criteria that define the quantity. You have to show complications, have coherence times, etc. Once the first lock is lifted, the next three may fall at the same time, or on the contrary, the second may become more difficult, which will require returning to the first. This is quantum complexity. The problem is knowing which way to go.
When do we see the first quantum computers?
In the quantum realm, the chip is a major problem with many Qubits. It takes about a million to make high-performance calculations. The next steps (the algorithm, the software to run the quantum computer) depend on the behavior of the Qubits, their quality, their number, etc. The time scale depends on the C12. It is up to them to develop this hardware and software sector. They claim to have a first quantum chip prototype in 2024. On our side, we are developing a different Qubits strategy, and our time scale ambitions are 2030.
For which applications will quantum computers be used?
There are several worlds in quantum. The most stimulating is high-performance computing. Quantum technologies will be in cryostats and not in your cell phone. We can compare this with data centers, but on a smaller scale. On the other hand, the computer power that can be developed is without measure. This will be used for the most greedy applications in terms of bandwidth, such as research in materials, health, energy, combinations of molecules in biology, climate evolution … A calculation that today would take a million years would only take a week with quantum. These are the promises of the quantum, it still needs to be demonstrated!
Interview by TR