Wednesday, May 29, 2019

Automotive SWIR Startup Trieye Raises $17M

VentureBeat, Reuters, Globes, TechTime: Israel-based SWIR image sensor startup TriEye announces a $17M financing round led by Intel Capital. The new funds will be used to develop its sensors for low-visibility situations for connected and autonomous vehicles.

TriEye's SWIR technology is CMOS-based, enabling the scalable mass-production of SWIR sensors and reducing the cost by a factor of 1,000x compared to InGaAs-based technology. The initial SWIR camera samples are expected to enter the market in 2020.


  1. what is the detection principle they use?

  2. The CTO is Prof. Uriel Levy ( Skimming through his quite interesting publication list, one can see he worked on defect assisted absorption as well as Schottky diodes, whereas his recent papers seem to suggest that the latter is the preferred method - supposedly also CMOS compatible and hence, probably a good guess on the underlying technology

    1. to me it seems that the schottky diodes paper (i only found 1 with schottky and infrared - still refers to lower wavelenght:
      We demonstrate a nanoscale broadband silicon plas-monic Schottky detector with high responsivity and im-proved signal to noise ratio operating in the sub-bandgapregime.
      Bandgap of SI is 1.14 eV according to, which corresponds to a wavelength of 1087nm at 300K according to There is no photoelectric effect in SI by photons beyond 1087nm wavelength because the energy of this photons is simply too low. So a SI based detector that relys on the photoelectric effect cannot work. I have to admit - I did not read the paper, just the abstratct. But there they clearly speak of 'sub bandgap regime', to me this is a NIR related topic. They speak of SWIR in their product, so another paper seems to be more likely to be an explaination: While this talks about MWIR:
      It is shownthat this model can absorb light effectively with a maximumof 100% at the absorption peak wavelength. In addition,the absorption peak wavelength can be adjusted flexibly bychanging parameters, including the chemical potential appliedto graphene or other structural parameters of the structure.
      I guess the product described above is more likely to be based on such a graphene type of approach than schottky diodes.

    2. There are a bunch of papers to be found on their website: E.g., look at this one describing Schottky detectors for use up to 2.5 um as well as the operating principle: The idea is that you don't propagate carriers from valance to conduction band but over over a Schottky barrier.

    3. thanks Andreas, very interesting to learn there is yet another aproach to detect IR photons beyond the SI bandgap energy. Apart from the articles and links in the blog posts, I like the comments very much. Sometimes insights from "gurus" in their field, sometimes links to papers like this you sent. Since a few years, opening imagesensorworld is the first thing I do every day at work, I've learned a lot with this. Big thanks to Vladimir for running this site and to you all for posting comments ;-)

    4. What is old is new again. There was a lot of work on infrared Schottky-barrier photodiodes in the 80's, by Walter Kosonocky at RCA for one.
      For example, returns 144 results.

    5. For now it is only an hypothesis that they're using Schottky detectors. I haven't seen a statement from them yet. Their papers do proper referencing back to the 80's - so I'd say they do a good job there. Let's see if they can make it a commercial success.

  3. imaging in night conditions ?


All comments are moderated to avoid spam and personal attacks.