< Back to previous page

Project

FWO-SB-Beurs: Picosecond Optical CMOS Samplers for Sensing High-Speed Time-Resolved Phenomena (FWOSB10)

I. State-of-art and objectives Image sensors have become mainstream consumer devices. Furthermore, high-end image sensors with high-pixel count and high frame-rates have become essential building blocks in industrial systems. When making the step to the world of much shorter time-scales, i.e. down to the nano- and picosecond range, sensors become available that can indirectly sense other things than just light. By measuring high-speed time-resolved optical signals in response to stimulation by short light-pulses, many applications that deal with photo-luminescence and fluorescence-lifetime imaging become feasible. Fluorescence Lifetime Imaging Microscopy (FLIM) is an exemplary domain for the state-of-the-art of high-speed time-resolved sensors. The application space linked to this photo-luminescence domain is booming, but a big step forward on high-speed sensors is needed. Expectations are that with the advent of low-cost and simple systems based on CMOS sensors, other domains will benefit and new applications can arise. Therefore research is needed to improve the capturing performance down to picosecond time-scales, certainly if this is doable in a low-cost CMOS technology. Conventional fluorescence lifetime microscopes use a gated image intensifier in combination with a Charge-Coupled Device (CCD). They have a low quantum efficiency and excitation rate and are additionally bulky and expensive. Demodulating CMOS imagers are being developed for FLIM applications. They don’t require an external gated image intensifier, avoiding its cathode quantum loss and drastically reducing size and cost. The current CMOS imagers don’t yet reach the performance and specifications of the conventional microscopes. A new revolution is necessary to be able to build imagers that beat the conventional microscopes in performance, specifications and are small and cheap. It is our belief that by combining the state-of-art with new concepts and possibilities, added with the knowledge, skills and talents available in the department, we can bring the optical CMOS samplers to a next level. II. Methodology and technology that will be used to achieve objectives The way of achieving this is by exploring the possibilities that come with backside illumination (BSI). BSI is an upcoming technology where the imagers are illuminated from the backside. The transistor circuitry and the metal connections now appear at the other side of the detector, which gives a lot of freedom and new possibilities for designing imagers. What we hope to achieve is by combining the Current-Assisted Photonic Demodulation (CAPD) technology, developed at our department, with a BSI set-up, following improvements could be made simultaneously: 100% fill-factor, speed-up by being able to apply a much larger electric drift field, speed-up by eliminating the last diffusive part of the travel, smaller pixel areas and multi-tap pixels. Several runs of concepts-simulations-fabrication-testing-documenting-improving should be conducted. Industrial partners will guide us for testing, decision making, valorization… III. Researches and research groups that will cooperate The research will take place at research group LAMI, part of the ETRO-department of the VUB. Information and knowledge will be spread between researchers to achieve maximal efficiency. The research will be under the supervision of Prof. Maarten Kuijk, who founded two spin-offs: Eqcologic NV, founded in 2005 and successfully acquired by Microchip Technologies in 2013 and Optrima NV, now called Softkinetic Sensors NV, founded in 2009 employing now 100 FTE. Other PhD students are doing research in related subjects and industrial partners will ensure possible valorization. The combination and collaboration of all these people will create an ideal ecosystem for this research.
Date:1 Jan 2016 →  1 Aug 2019
Keywords:CMOS
Disciplines:Image processing