Big improvement in colour camera technology

http://phys.org/news/2014-04-pixels-nanowires-paradigm-digital-cameras.html

I had previously learned about how conventional colour cameras work and noted that they waste at least half the photons of light through colour filters before they can reach the light sensors and that halves the cameras sensitivity to dim light in dim light conditions. But here they found a way to do away with those wasteful colour filters and this should at least double the sensitivity of colour cameras to dim light. They achieve this by making layers of vertical silicon nanowires each with all having a different radius that causes them to absorb a different range of wavelengths of light -one wavelength range for each primary colour. The link also states this has other advantages of “higher pixel densities and higher resolution”.

But one quote that confuses me though is where it says:

“...the pixels with different color responses can be defined at the same time through a single lithography step.”

does that imply that all the silicon nanowires for the different primary colours are put in exactly the same layer rather than stacked one in front of the other with one layer for each primary colour? Because, if so, surely that would mean wasted photons and that would defeat the whole point!? Or am I probably just mentally visualizing this “ single lithography step” incorrectly (somehow ) ?

Originally posted by humy
http://phys.org/news/2014-04-pixels-nanowires-paradigm-digital-cameras.html

I had previously learned about how conventional colour cameras work and noted that they waste at least half the photons of light through colour filters before they can reach the light sensors and that halves the cameras sensitivity to dim light in dim light conditions. But here they ...[text shortened]... am I probably just mentally visualizing this “ single lithography step” incorrectly (somehow ) ?

I was wondering why the image shown of the child's toys was such low res, it was based on a 100X100 array.

They would not be vertically stacked, they would be side by side. If you note, the sizes they mentioned like 80 nm, presumably for blue which is around 320 nm, which makes these like a 1/4 wavelength antenna if we were talking 100 mhz or so. That means you can stack a lot of them together so you could have a section that was 80 nm, say 4 of them side by side, then a 100 nm section and so forth, because they are way less than a wavelength you can put a number of them together and it would all work out.

They do mention difficulties in real world manufacturing however so new process engineering will have to go hand in hand with this device for it to reach the real world marketplace. When they do, the resolution will more than double over today's camera's and still maintain low light level sensitivity.

One thing they noted, making them bigger or smaller can extend the sensitivity into the UV or IR range so you could see a camera that went from IR through visible through UV in one device with great sensitivity which implies to me the military would glom onto them and not allow them into the marketplace. Just speculation of course. That same technology could extend to astronomy and maybe get WAY deep into IR for gas analysis around stars and such.

It might even be able to resolve Terahertz radiation, millimeter wavelengths, where each post would now be 250 microns wide, still 1/4 wavelength wide.

Of course I am just expanding and speculating on the future of such techniques.

Since each post is 1/4 wavelength I wonder if this could lead to a holographic display if they can be made to emit light since now you would have an array of lines each less than a wavelength so if you could phase array them, you could have a projector capable of sending light out at programmable directions and any wavelength. It would be like a radar phased array but for light frequencies.