INTERVIEW
LiNbO 3 remains the best suited material for frequency conversion. It is not deposited as single crystal by pulse laser deposition( PLD) or epitaxy, so we purchase TFLN wafers and then do top-down processing. BaTiO 3 is attractive for electrooptics and can be deposited or grown more easily, though for frequency conversion LiNbO 3 is clearly ahead.
You have strong expertise in LiNbO 3 and BaTiO 3, how do they compare for your projects? LiNbO 3 remains the best suited material for frequency conversion. It is not deposited as single crystal by pulse laser deposition( PLD) or epitaxy, so we purchase TFLN wafers and then do top-down processing. BaTiO 3 is attractive for electro-optics and can be deposited or grown more easily, though for frequency conversion LiNbO 3 is clearly ahead. Rather than trying to beat performance records in this very competitive environment, we often differentiate our research by original design and features. For instance, we pushed Bragg modulators rather than only standard Mach-Zehnder layouts. Another recent development of those materials in my team, is their synthesis as solution processed films that can be easily nanoimprinted without the needs of complicated etching process. We just demonstrated metasurfaces and metalenses with frequency conversion and electro-optic effects.
Do you impulse collaborations with industry? When we were temporarily unable to apply to certain European calls from Switzerland, I deliberately diversified funding with industry-funded PhDs, always under conditions that protect the student’ s ability to publish, speak at conferences and finish a thesis. A good illustration of this partnership starts when a start-up wanted a compact, no-movingparts spectrometer for space payloads, since bulk optical spectrometers are heavy. We developed an electro-optic on-chip broadband spectrometer. The original idea was theirs, but we did the full device and characterization.
After the ETH News website offered us a space to highlight our research, on nonlinear metasurfaces, several companies reached out. One possible collaboration is related to solar energy where metasurfaces could concentrate light onto small, very efficient cells to power small devices. How do you position your group in the quantum optics landscape? We build enabling technology for quantum sciences. Miniaturization is the challenge: integrated sources from nonlinear media for entangled-photon generation, combined with superconducting detectors, with demanding high-speed and cryogenics performances. I do not believe there will be a single winning platform. Silicon, LiNbO 3, BaTiO 3, III-V each have a their advantages. Some companies are raising hundreds of millions round of financing. We will not compete on scale and record performance. Nevertheless, we may contribute original designs, clean proofs of concept, and robust building blocks.
How do you handle tech transfer and spin-offs from the lab? I try to support young founders from my team by providing the infrastructure they need to incubate deeptech projects for one to two years, helping to remove pressure from them. If their first attempt doesn’ t succeed, they can pivot to something new and still benefit from the knowledge and experience they have gained. We currently have three startups in motion, including Versics, founded in 2022, which develops telecom modulators in the 60 to 110 GHz range. I am a co-founder, but I focus on scientific support and infrastructure. I leave business execution to those who want to live it every day.
Photoniques 134 I www. photoniques. com 15