Black Silicon Used For Surveillance?

An anonymous reader writes "For the past decade, 'black silicon' has been touted as a way to make super-sensitive image sensors and ultra-efficient solar cells. That's because the material — silicon wafers treated with sulfur gases and femtosecond laser pulses — is much better at absorbing photons and releasing electrons than conventional silicon, at least over certain wavelengths. In 2008, Harvard spinoff SiOnyx went public with its plans to commercialize black silicon. But what happened to those plans? Today SiOnyx revealed in another exclusive that it has raised new venture financing from Microsoft co-founder Paul Allen and other big investors. It also has formed a key strategic partnership to scale up manufacturing of black silicon — and go after markets in security, surveillance, automotive, consumer devices, and medical imaging."

Comments

I don't like black silicon

[Anonymous Coward]

I think black silicon is less productive, despite all the effort we've put into giving it a chance based on historical misuse and underuse. Basically, it doesn't have the right bond valency to adequately function in our society, so we really ought to let the markets sort of segregate so that people who prefer white silicon can go there without having to worry about intermixing.

Re:I don't like black silicon

[0100010001010011]

They also prefer African American Silicon.

Re:Misleading summary

[couchslug]

"Black Silicon Kills Babies!"

An objective test is required. I suggest a sample of ten babies of the same weight.

Drop a 200KG slug of Black Silicon from a height of two meters on five of them, then drop a 200KG slug of White Silicon on the other five. High-speed video could monitor plastic deformation and splatter.

Nightvision?

[Fibe-Piper]

The article states the this allows people to see where they have previously been blind. Obviously the speaker means that people cannot see in the dark and this gives them this ability. I wonder how this compares to standard night vision technology which sounds like it does the same or similar thing.

Re:Nightvision?

[Anonymous Coward]

Standard night vision uses near-infrared light to 'see'. It requires an infrared emitter to actually 'see' things. Normal human eyes cannot see this light. Military/industrial grade night vision uses sensors that picks infrared light generated from heat. This is the stuff you usually see in movies. (See FLIR entry in wikipedia)

This dark silicone picks up visible light, although it will be far more sensitive than current sensors. As long as it's not pitch black, a tiny amount of light that normal eyes cannot see will be sensed by it.

Re:Nightvision?

[ChrisMaple]

Normal silicon (not silicone) is already quite sensitive, coming close to the ideal of responding to each photon. It is simply not possible to be "far more sensitive than current sensors".

Re:What's the significance?

[vlm]

Why is black silicon being used in security and surveillance significant? Title should read more like "Paul Allen and others invest in Black Silicon."

A photodiode is a really tiny solar cell. Or a CCD is vaguely like an array of really tiny solar cells with a bunch of glue logic (actually way different but at a simplistic enough level thats a useful mental model of a CCD even if its implementation is different ..)

Anyway the short version is high efficiency works, but apparently failed economically for bulk energy production. Ooops. Time for a new business plan. The purpose of yer low light camera sensor isn't to charge a battery, so its possibly useful regardless of manufacturing dollars per watt delivered.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Man, that's racist

[FatdogHaiku]

To be Politically Correct, you must call it African American Silicon.

You should never call prople like Desmond Tutu a Black. He must be called an African American.

So, the color of his skin gives him American citizenship?
If you really want to phrase it this way, he must be called an "African South African" which seems to be the epitome of redundancy...
Also, I had no idea he was a "prople".

Any questions?

[sevenofdiamonds]

I'm getting my Ph.D researching black silicon. If you have science or engineering questions about it, post them in reply to this comment. I'll check back at around 3 PM EST and will do my best to answer the questions I find then.

Re:Any questions?

[vbraga]

The summary defines black silicon as:

silicon wafers treated with sulfur gases and femtosecond laser pulses

Is this a thin film deposition (pulsed laser?)? Can you give a more accurate description? Maybe pointers to interesting papers?

Thank you!

Re:Any questions?

[DrWho520]

- Other posters have noted/claimed this is a result of high manufacturing costs making this material prohibitive for solar cell production. Could the manufacturing costs of this material be brought down to a point as to make it a good substance for solar cells? How close are we?
- What wavelengths does this material respond too/detect? Could it be modified/designed to image UV/Vis/IR?
- How linear is the response function, or perhaps would it require an exotic calibration procedure to translate photons into radiance?

Thank you for volunteering to answer questions and good luck in your academic endeavors! I wish I was in graduate school now and was positioned to work in this domain.

Re:Any questions?

[sevenofdiamonds]

The improvement realized by black silicon depends both on the kind of detector in which it is used and the wavelength it's trying to detect. An application for black silicon that the research community takes very seriously at the moment is detection of light at a wavelength of 1064 nm. This is the main emission line of Nd:YAG lasers, which are already used in a variety of applications (see: http://en.wikipedia.org/wiki/Nd-YAG_laser#Military_and_defense). At that wavelength, black silicon detectors are at least twice as good as traditional silicon devices and can be made 100-1000x thinner.

Re:Any questions?

[sevenofdiamonds]

Black Silicon is not made in a deposition process. Instead, ordinary silicon is shot with a femtosecond pulsed laser in the presence of a sulfur-containing gas. The laser causes sulfur to be incorporated while also structuring the surface of the silicon. Thus it changes both the chemical state and physical morphology of the material. I encourage you to check out the following freely available Ph.D thesis for more information: mazur-www.harvard.edu/publications.php?function=display&rowid=648.

Re:Any questions?

[sevenofdiamonds]

I work on silicon-chalcogen alloys like the material SiOnyx uses but also including Si:Se and Si:Te. I encourage you to attend next year's black silicon symposium (http://www.army.mil/-news/2009/08/26/26478-bent-laboratories-line-of-sight-goes-beyond-cannon/). The article is a year old, but the most recent symposium did include discussions of several types of black silicon beyond what is used by SiOnyx.

Re:Any questions?

[sevenofdiamonds]

Because thin layers of black silicon can absorb as much light as thick layers of ordinary silicon, it is possible that black silicon solar cells would be cheaper than ordinary silicon solar cells. The reason that the most progress has been made on detectors rather than solar cells is that it is easier to make a profit selling detectors at smaller scale. It is not so much the cost floor that has thus far prevented the appearance of black silicon solar cells, but rather just that it is currently not made at a scale that would lead to affordable solar cells. As technologies are developed for making larger quantities of black silicon, I would not be surprised if it or a related material started finding its way into solar cells. An example response spectrum can be found here: http://www.sionyx.com/advantage.html. Note that this plot shows internal gain, which some variants of the material possess at >2V reverse bias. The response function when running at small 1V reverse bias is comparable to that of ordinary silicon, but extended deeper into the IR (out to 1300 nm instead of silicon's 1100 nm).

Re:Any questions?

[sevenofdiamonds]

Here is an example response curve with internal gain: http://www.sionyx.com/advantage.html. The QE for a black silicon solar cell can be found in figure 4 of this patent: http://www.freepatentsonline.com/20100224229.pdf. Solar cell performance, measured at 0 bias, is an experimental way to look at the response without internal gain.

Re:Not having RTFA

[sevenofdiamonds]

The S is typically either SF6 or H2S gas. The wavelength of the femtosecond laser isn't especially important; the key is that the laser fluence (energy per area) be above the ablation threshold of the silicon (between 0.1 and 1 J/cm^2 for the relevant pulse durations). The laser spot size is typically a fraction of a millimeter on a side, but it can be rastered over a silicon wafer to make a large-area black silicon film. There is a recent Ph.D thesis available for free at: mazur-www.harvard.edu/publications.php?function=display&rowid=648 that gives a complete recipe for making black silicon.

Re:What's the significance?

[blind biker]

Anyway the short version is high efficiency works, but apparently failed economically for bulk energy production. Ooops. Time for a new business plan. The purpose of yer low light camera sensor isn't to charge a battery, so its possibly useful regardless of manufacturing dollars per watt delivered.

Using femtosecond lasers for treating silicon surfaces was never going to be price-competitive for solar panel production. DRIE black silicon on the other hand, could be made competitive, if/when production scale DRIE equipment appears, specifically modified (and simplified) for black silicon forming. The strong plasma that is required, however, limits scalability. Still, not entirely impossible.

Re:Man, that's racist

[mcgrew]

Woosh!

Re:Misleading summary

[Red Flayer]

An objective test is required. I suggest a sample of ten babies of the same weight.

Well, now you're just inducing selection bias. What if all babies who weigh 5 kilos have a splatter radius of only 0.2 meter, and all babies that weigh 7 kilos have a splatter radius of 2 meters? Then your objective test would be inherently flawed as to the predicted result of dropping a 200 kg slug of silicon on a baby.

What we need to do is to take a random sample of babies, of sufficient number compared to the total population of babies to achieve a low margin of error, split them into a control and a test group, and *then* drop the slugs of silicon on them.

Oh, also, for true statistical rigor, make sure to stab the babies with a pencil in both eyes first to ensure the study is double-blind.