Pioneering High-speed – Interview with Marco Eppenberger
Marco Eppenberger is an Integrated Photonics Engineer specializing in quantum technologies, holding a Doctorate in Science from ETH Zürich. With a current role as an R&D Engineer at PsiQuantum in Palo Alto, California, Marco brings a wealth of expertise to the field. His academic journey at ETH Zürich, where he completed a Master of Science and a Bachelor of Science in Electrical Engineering and Information Technology.
Polariton: Your paper introduces plasmonic micro-racetrack modulators with impressive electro-optical bandwidth and high-speed operation. Can you explain why is this important?
Marco: In today’s information age, we rely every day on services that need to communicate data. Just imagine sending your friends the latest vacation picture or buying your train ticket on your smartphone. Even more important: the need for data transmission continues to grow exponentially with emerging applications, for example self-driving cars, next generation supercomputers, or assistants using generative artificial intelligence. All such services rely on performant interconnects that move data from to and from datacenters and within datacenters themselves with as little energy as possible. The introduced modulators enable the next step of parallel optical data transmission at highest speeds and can be co-integrated, unlocking these energy savings in the long run.
POL: This technology could have a big impact on things like faster internet and better data centers. Can you give us examples of how people might use this technology in real life?
I see the prime application for this new technology within datacenters. Currently, the power consumption of datacenters is a big challenge. By using these new modulators adjacent to data-hungry processors or memories and interconnecting them, more compact, high-speed, and parallel optical interfaces could be engineered, omitting the power-inefficient electrical cables. In the end, improving datacenter efficiencies means that more powerful supercomputers can be realized. An example for an immediate benefit of these would be better climate forecasts, faster medical drug discoveries or more efficient batteries.
What do you think could be the next steps for this technology?
An immediate, to me very exciting near-term development could involve co-integration with the associated electronics. We have shown that the parallelizable optics can be designed for highest-speed operation. However, that alone is not enough to enable connectivity. Co-designing the appropriate electronics and a suitable electro-optic receiver and demonstrating end-to-end connectivity could be a very worthwhile goal for realizing such new, lowest-power interconnects.
Thank you Marco
Interview by Helena Echeverri
- Eppenberger, M., Messner, A., Bitachon, B.I. et al. Resonant plasmonic micro-racetrack modulators with high bandwidth and high temperature tolerance. Nat. Photon. 17, 360–367 (2023). https://doi.org/10.1038/s41566-023-01161-9