POLLOC: Polariton logic

An all-optical photonic approach will boost computational energy efficiency

Context

In 1965, Gordon Moore predicted that the number of transistors on a chip would double every year to reach 65 000 by 1975. When that remarkable prediction proved true, he revised the doubling rate to every two years, and that became known as Moore’s Law. Almost 50 years after Moore’s seminal prediction, traditional chip architectures are reaching their technological, practical and economical limits. The EU-funded POLLOC project is exploiting an all-optical approach that takes us beyond current transistor technology. By replacing electrons with photons, optical transistors and all-optical logic gates are envisaged that could bypass the fundamental limitations of the current electronic transistors. Moreover these novel devices offer processing at the speed of light to achieve energy-efficient massive processing required for tomorrow’s high-efficiency and high-power computing platforms.

Objectives

For energy-efficient computation beyond the current CMOS paradigm, tweaking the current nanoelectronics roadmap will be neither enough nor sustainable, but requires to completely rethink transistor devices and circuits. Leveraging recent breakthroughs in perovskite nanomaterials and room-temperature exciton-polariton devices achieved by the consortium partners, we believe that now the time has come to take this beyond the scieFor energy-efficient computation beyond the current CMOS paradigm, tweaking the current nanoelectronics roadmap will be neither enough nor sustainable, but requires to completely rethink transistor devices and circuits. Leveraging recent breakthroughs in perovskite nanomaterials and room-temperature exciton-polariton devices achieved by the consortium partners, we believe that now the time has come to take this beyond the scientific publication level and build a novel technology that can leapfrog established architectures.
Within POLLOC we aim for the development of a complete technology platform for universal photonic information processing based on exciton polariton condensates in microcavities with inorganic perovskites. We will validate this new technology with respect to the key parameters power, energy-efficiency, size, frequency, and cost. In the digital processing domain, we aim for optically programmable, cascadable logic gates with less than 100 attojoule switching energy and sub-picosecond switching speed. To fulfil the requirements of this disruptive all-optical device and circuitry approach, POLLOC assembles the whole gamut of necessary expertise from chemistry, physics, theory and technology. The carefully chosen, well-balanced consortium consists of leading partners from academia, SME and large end-user with excellent track records that are uniquely positioned to tackle the ambitious goal to unleash the potential disruptive performance gains of this technology and to establish a new kind of digital and analog circuitry paradigm.

Partners

IBM RESEARCH – UNIVERSITY OF SOUTHAMPTON – CNRS – Foton OHM – ETH Zurich – AMO

Coordinator

IBM RESEARCH

Coordinator iFOTON: Laurent PEDESSEAU (Foton-OHM)

Fundings

EU (2.740 M€, FOTON: 80 k€)

See on line