New sensor material 'Black silicon', 1000x more sensitive to light?

What I like is this part:

"The SiOnyx patented laser process produces silicon photodetectors with extraordinary performance at visible and infrared wavelengths from 400 to

1550 nm.? The measured optical responsivity of 100 A/W at 950 nm demonstrates a 100 fold increase in sensitivity over traditional detection methodologies.? This equates to an external quantum efficiency of 10,000%.? Remarkably, this performance is achieved at a mere 3 V of operational bias, enabling direct integration with hybrid and digital circuitry.? The extension of silicon"s spectral sensitivity out to 1550 nm has profound implications for laser sensing applications at the 1064, 1330, and 1550 nm nodes."

Hmm, a quantum efficieny of 10,000%. Connect that to a LED with 10% efficieny, shine 50% of the LED output on the detector and I finally have free light at home. Gone is the global warming problem!!!

Or would I really need 3V of bias to get the amplifying effect?

Arie de Muijnck

Crappy web site. I wonder how thay manage the "photonic gain."

John

What's the current state of silicon solar cell technology? About 15% IIRC.

So they can get 150x the incident power at the output? WoHoo! Perpetual motion id here!

--
Paul Hovnanian     mailto:Paul@Hovnanian.com
------------------------------------------------------------------
A vacuum is a hell of a lot better than some of the stuff that nature
replaces it with.       -- Tennessee Williams

On a sunny day (Thu, 5 Feb 2009 19:58:36 +0100) it happened "Arie" wrote in :

I think what they are doing is increasing surface area. So if you say light captured per cm^2, then it will be more then for a flat surface, that should match the numbers perhaps.

It's a photoconductor, not a generator. Still, there must be one hell of a multiplier mechanism in there somewhere.

It is pretty slow.

John

It is probably similar to an APD.

The quantum efficiency of normal InGaAs photodetectors is almost 100% with APDs having gains beyond that of hundreds or thousands but requiring very precise voltage control since you run them in avalanch mode.

Increasing the area doesn't help as it just spreads the photons out more thinly.

Now being able to this at low voltage and with silicon is a significant advance.

kevin

The significance is stretching the response out to almost two microns. Normal silicon quits at 1.1 -1.2 microns.

Steve

On a sunny day (Thu, 5 Feb 2009 14:30:59 -0800 (PST)) it happened kevin93 wrote in :

Maybe, as they call it 'black silicon', it sort of looks black, means it does not reflect anything, so more light actually lands on that bigger surface, and for a broader spectrum.

But it is late at night there, maybe I think too simple.

Photoconductors improve the responsivity but not the sensitivity. You get 1000 times the photocurrent with 1000 times the noise, and (incidentally) 1000x lower bandwidth due to the carrier lifetime.

The real ticket for improved performance is the electron-multiplying CCD.

Cheers,

Phil Hobbs

I agree; no way to move the image to see everything. I question the comment "hundreds of volts".

Photoconductors exhibit gain equal to the carrier lifetime divided by the transit time--essentially you get to re-use the carriers many times. That's also the origin of the speed/gain tradeoff.

Unfortunately both the photogeneration and recombination are stochastic, so they have twice the shot noise of a photodiode. This is sometimes called 'generation-recombination noise', but that causes confusion with normal thermionic G-R noise in IR photoconductors.

Cad sulphide cells (as used for the last century) are similar, have even higher gain, but have all kinds of hysteresis and things.

Cheers,

Phil Hobbs

Psychrotronic turboencabulation?

;-) Rich

Is there anything in the physics that would prevent it from acting like an ordinary vacuum tube electron multiplier?

Thanks, Rich

No, but every photon would have to, on average, create a couple of hundred electron-hole pairs, when it usually is lucky to knock loose

0.5 on average. Great if it works, I guess. But gain is cheap at these speeds, so there's no huge advantage over pin diodes.

John

To be fair Sionyx also has a finger in that pie. And although their announcement is a bit garbled about black silicon with its improved band gap properties and higher photon conversion efficiency was one of the things they announced. eg

It has been a bit quiet of late. I guess they reckon the sensor market will be higher added value and so more lucrative on small yields.

It is on the face of it a clever technology. Sort of a bit like a photon detector coupled with an integral solid state electron multipler (needing only a small bias voltage). It is clever and should be useful.

Regards, Martin Brown

This is the real science. Phil is on the ball.

We had single element bolometers with FETs integrated right at the transducer that were pretty sensitive. has to be very carefully installed, and had to be careful which way the optics got pointed, even if it was off.