The future for sensors is becoming commercially available.

[dabateman]

It looks like the startup company Gigajot, a spin out from Eric Fossum labs. Whom is the inventor of the CMOS sensor, will have the newest QIS sensor technology available to purchase soon.

https://www.gigajot.tech/qis-cdk

 

The quantum efficiency in UV isn't stellar. But you will be able to count those photons. UVC looks more promising.

[JMC]

David, I've seen a few of these sensor charts now where the sensitivity seems to climb below 300nm. What's your thoughts on this, is this just what normal sensors do, or are they doing something to them to improve sensitivity?

 

Unfortunately my camera sensor sensitivity setup wont allow me to look below 280nm to get quantitative data, but when I did some imaging at 254nm, I was surprised by how sensitive my cameras seemed to be (tried the Sony A7III with the Bayer filters still intact as well as one of my monochrome cameras).

[dabateman]

I believe this data as its coming from people whom have developed this technology and have really advanced sensors.

What interesting is it starts climbing before 300nm and keeps climbing after 250nm. So this is not just due to increase signal from ionization, as that would just be the typical bump you see around 250nm for back side illuminated sensors.

I don't know why it dips 350nm and then increases. I will see if I can ask Eric about that. He is a regular on the dpreview science forum.

[JMC]

Thanks David, yes please check if you get chance as it'd be interesting to know.

[Stefano]

According to Wikipedia silicon sensors are theoretically capable of seeing from 190 nm to 1100 nm. I don't know if every (monochromatic) sensor can do that, but I do think that common sensors, if modified to be monochrome, can see much deeper than we think, easily going below 300 nm.

[dabateman]

Yes I am also surprised that I can image at 254nm with my Olympus Em5mk2 camera. Sony pulled something magical out, as my Panasonic sensor inside my Em1mk1 can't see that low, both being converted by Kolari. The Sony A6000 might be similar as Bernard had success at 300nm.

 

Its possible that Sony now blocks this achievement with coatings on the actual coverglass of the sensor.

 

I think its been reported Stefano that BSI CMOS sensors can see much deeper into X-rays. GSENSE has optimized their sensors for high energy detection. But these are custom and I don't know if they would still perform well into UV and visible wavelengths. See here:

https://www.gpixel.com/news/gpixel-announces-new-pulsar-technology-enabling-gsense-products-for-vuv-euv-soft-x-ray-and-electron-direct-detection-imaging/

 

This is a link to the paper that discusses the read noise characteristics of these new sensors:

https://ieeexplore.ieee.org/document/9402860

[Stefano]

That GSENSE sensor looks really cool. The problem with UV below 100 nm is the lack of transparent materials, literally nothing is transparent there. But at least we have a sensor capable of recording that radiation.

[dabateman]

Jonathan,

You can see Eric's response here:

https://www.dpreview.com/forums/thread/4573692?page=2#forum-post-65156530

 

Decent hand waving responses are here:

https://www.dpreview.com/forums/thread/4576263

 

Possibly due to quantum yield, where higher energies, wavelengths bellow 400nm results in two photons not just one being detected. Not sure that would account for it all.

 

Ken's response might be more correct, but still hard to know.

 

This no doubt now will spin out into a research paper.

[dabateman]

Stefano,

This plot answers my question in my last response to you and I didn't even realize it.

https://www.gpixel.com/wp-content/uploads/2019/11/GS2020BSI-QE-curve.png

 

The blue curve is the regular BSI uv optimized sensor with an AR coating.

The lower red curve is the newer x-ray optimized sensor, shaved down and without AR coatings. So the cost for deeper detection isn't too bad. And the quantum efficiency follows that of an uncoated sensor.

[JMC]

Thanks David, I'll take a look at that.

[Stefano]

Who knows, one day someone will take reflected X-ray photos and videos in his garden holdind this sensor mounted on a lead pinhole with a berillium window and an X-ray tube on top, all battery-powered. Most things would look like they are made of glass. I hope to see this being done one day by someone, it would be so cool.

[OlDoinyo]

Instead of a lead pinhole use a pass-through Fresnel mirror--much brighter image that way.

[Andy Perrin]

I think you mean a zone plate, olDoinyo?