October 31, 2008

"black" Silicon
— Purple Avenger

Interesting stuff. They have manufacturing scalability issues though.

...captures nearly all of the sun's light. "It is basically a sponge for light, both visible and infrared," says CEO Stephen Saylor...
The cool thing is that its picking up in the IR spectrum as well as visible, which traditional silicon is pretty useless for. Its usefulness as a power generating technology remains to be seen, but it looks pretty good for improving light sensor performance. Sounds like they've gotten a ~200X reduction in the amount of power needed to operate a light sensor. More info HERE and HERE.

That they talk about doing a melt and recrystallization suggests pretty high power pulses. There's no mention of what the beam width is when they do this, so it could be a pretty narrow beam -- which is where their scalability problems may be coming from. You can't do nano-scale operations and produce shit by the square mile in any practical manner.

We were initially doing this sort of pinpoint stuff, but managed to get good results with a beam the optics fan out very wide. Using wafer based technology is also a very limiting factor for scalability and manufacturing cost -- one of the basic problems with traditional silicon. Nano-solar got around this with the magic goop they spray on a flexible substrate and dispensed with wafers entirely. We'll be eliminating wafers as well in what I'm working on. They cost too much and are limiting on your final material form factor. Wafers are fine for proving concept in the lab, but they really kinda suck when you want to roll high volume manufacture and have nonrestrictive form factors in the final product.

I've occasional mentioned that there will be a slew of new material science developments over the next 10 years that will be transformational for the US economy and industry. What these folks have managed to do with silicon is a small taste of what's to come. There'll be a lot of novel processing of common materials that get them to behave in ways that are very uncommon. Science is just starting to scratch the surface of how materials really behave at the nano-level, and traditional processing methods, like ion implantation, have intrinsic limitations to the kind of elements you can process with them. That is going to change. Actually it has already changed ;->

Posted by: Purple Avenger at 07:38 PM | Comments (32)
Post contains 397 words, total size 3 kb.

1 Reminds me of discussions I have with people who want to renew their three-year-old computer's warranty. Eventually, getting someone to use the old-fashioned way of doing xyz to fit the standards of the computer when it was built would be like getting someone to manufacture vacuum tubes. Technology moves so quickly.

Posted by: Sekhmet at October 31, 2008 07:43 PM (QiWeI)

2 Not to sound like an idiot, but can we use this for sledding or some other kind of, um, ah, er, lubricant?

Posted by: mrcaniac at October 31, 2008 07:47 PM (Rbulg)

3 I remember a couple of SF stories with "black plants" that would use all the light -- apparently normal plants only use about 10% or so.

Posted by: Dead Career Sketch at October 31, 2008 07:50 PM (JTN0y)

4 Why are you so racist?

Posted by: Christoph at October 31, 2008 07:51 PM (hawOV)

5 I follow this stuff as well, it looks like they're awfully close to some of the white papers I've read about nanoscale rectifying antennas replacing conventional photovoltaics, at least in the lab.  I'm hopeful it's 3-5 to productize, 8-10 to get a few large installations.

Posted by: leoncaruthers at October 31, 2008 08:06 PM (JSO4h)

6 seriously with the racism. why does it seem like everything is about color for you Purple Avenger?

Posted by: phil at October 31, 2008 08:08 PM (fDO1N)


Won't this make solar energy too affordable?

I mean, with all the alternative energy lobbyists out there protecting thier little pet technology, would this get squashed?

And because it might let the little guy have efficient solar on his roof, he can cut the umbilical cord to big electric?


I do agree there is alot of Nano scale tech coming, but its too expensive for now. When it hits mass productions and the cost goes down the curve .....

Posted by: Cromagnum at October 31, 2008 08:08 PM (j5MnB)


energy efficiency from solar at 90% wouldn't be a research project.   Fuckers would be funding that fucker en mass, without a second thought.

Can't forget the energy input to create (I know you don't PA, remember I love your blog) this material to begin with.  Might be an intro, but it's not the answer.

Posted by: Wickedpinto at October 31, 2008 08:10 PM (ul7te)

9 Is that the shit that was in Scully's eyes?

Posted by: Dave in Texas at October 31, 2008 08:13 PM (eiOZw)


Then we still need transportation energy. Unless we can store alot of 'juice' QuantumSphere might have that, as well as a better fuel cell

Gas2Go might have found a solution in Cellulose to Sugar to Gasoline

(via Instapundit)

Posted by: Cromagnum at October 31, 2008 08:16 PM (j5MnB)

11 Cromagnum, You can also flat-out make methane with reasonably priced electricity. Compressed natural gas vehicles aren't too outlandish. So there's plenty of options... if electrical production actually starts exceeding usage growth dramatically.

Posted by: Al at October 31, 2008 08:34 PM (Lk931)


read that too cromagnum, have gas2 in my "science" folder, didn't need insty.

Problem is that while roome temperature cellulose to sugar creation is less energy intensive, doesn't change the fact that you still need energy, and ethenol, even from sugar is still 120% plus energy in-efficient.

It can only work if we have nukes, and even then, we would need a LOT of nukes.

Posted by: Wickedpinto at October 31, 2008 08:42 PM (ul7te)


I would actualy prefer the manipulation of biologics, than nano's.

Partly cuz most of the "nanotech" that has been "proven" hasn't actually ben proven.

why can't we propogate termite ass bacteria, and just produce sugar in a big ass vat of termite ass bacteria?

Posted by: Wickedpinto at October 31, 2008 08:51 PM (ul7te)

14 > Won't this make solar energy too affordable?

Oh, Geez.


The Solar Constant at this distance from the sun -- the actual, perfect-condition amount of energy which reaches the earth's surface -- is roughly


No ifs, no ands, no buts. This isn't a fudgeable factor. No magic tech will improve upon it. It's a simple, uncomplicated physical constant. That is the MAXIMUM possible amount of energy which can be derived from ANY solar collector -- a single kW for each square meter (square yard in Americanspeak).

Not one microwatt more.

To derive one kw of power, you need to cover a square meter with cells.

However, you have to cover more than that, because, the fact is, you don't have sunlight 24 hours. So assume perfect sunlight, and double the coverage, since on average you only have sunlight half the day.

Now we have *2* m2 per kW.

And you have to keep the power for the evening times, which means you need some kind of battery/storage technique. Batteries are pretty inefficient. Double it once again -- *4* m2 per kW.

Current generating capacity for the USA is in the terawatt range -- 1 billion kW

So, in order to produce the power in the USA alone, we need to cover not less than 4 BILLION m2   in order to get that power. That's 4000 km2 -- 75% of the State of Delaware -- all with little "black" cells.

And that's our theoretical maximum. Start tacking on other losses (say the cells are only 75% efficient) and the fact that we really don't have perfect sunlight, but variable sunlight and cloud cover -- and you're pretty much looking at covering MORE than the land area of Delaware with these cells.

And the current cost of a solar panel? About $615/m2 -- say we manage to halve that, and round it off -- call it $300/m2 -- That's 1.2 TRILLION dollars.

And power consumption is unlikely to go down if we switch to all these electric cars, you know? So we're going to be going UP from that.

And that's not even getting to the fun part.
1) Cleaning. Solar cells need to be cleaned on a regular basis, or they lose about 30% of their efficiency. What, praytell, is the major cause of accidental deaths after vehicular accidents? Try Falls. Yes, accidental falls. And is that going to go up when you have people going out in the snow up onto a slick roof with the intention of clearing off the snow and leaves and what not on the panels?

2) Toxic waste. Semiconductor production is not a "clean" process. This is clearly a "new" process, so it's possible it will be cleaner than classic cells -- but how do we know this? We don't. One more big IF added to the expense category.

3) Lifespan. A typical solar cell does not have a long lifespan -- typically somewhere between 5 and 10 years and it drops off steadily from there. So that 1.5 TRILLION dollars isn't amortized across 30-40 years the way it is with oil, gas, or nuke plants -- it's every 5-10 years.

4) Disposal. Every 5-10 years, we're looking at disposal of 4,000 square KILOMETERS of solar panels. Recycle? Landfill? How much is THAT going to cost? My money is, it's not going to be CHEAP.

Q.E.D. -- solar power SUCKS.
Big Time.
And solar cells -- are always going to suck.

There's only ONE form of solar that actually has ANY promise -- and that's Ocean Thermal. OTEC basically works by using the temperature differential between the surface of the ocean and water about 90 feet down. Right -- it uses the entire ocean as a solar collector. But that's a low-temperature gradient, and we don't have an effective technique for extracting that energy yet (one possibility looks to change that -- the JTEC), but it's still not cost-effective, despite the free energy to be derived from using the surface of the ocean as your solar collector. So it's not ready for prime time.

Posted by: OBloodyhell at October 31, 2008 09:28 PM (00mT8)

15 > That's 4000 km2
Oop. 2000. Sorry. Adjust as needed. was merging this with an older form of it regarding standard cells, which are under 50% efficient and thus had another doubling in them that I was removing on the fly.

Posted by: OBloodyhell at October 31, 2008 11:54 PM (00mT8)

16 But will it make better breast implants?

Posted by: chad at November 01, 2008 12:13 AM (YICPL)

17 I discovered black silicon years ago as a substitute for gaskets when installing exhaust headers.

Posted by: Baron Von Ottomatic at November 01, 2008 02:03 AM (Ulsfn)

18 Black silicone?  So now flat-chested Afraican-American girls can get some mondo implants, too?  Truly a great day for America and race relations.

Posted by: Stinky Esposito at November 01, 2008 02:13 AM (MMC8r)

19 Black silicon?  Kinda reminds me of Ice Nine.

Posted by: Jabba the Tutt at November 01, 2008 05:09 AM (/fxRW)


There's only ONE form of solar that actually has ANY promise -- and that's Ocean Thermal. OTEC basically works by using the temperature differential between the surface of the ocean and water about 90 feet down. Right -- it uses the entire ocean as a solar collector. But that's a low-temperature gradient, and we don't have an effective technique for extracting that energy yet (one possibility looks to change that -- the JTEC), but it's still not cost-effective, despite the free energy to be derived from using the surface of the ocean as your solar collector. So it's not ready for prime time.


I had the misfortune of talking to some idiot who thought that the way we needed to go was 1) stop ALL exploration for oil period. Basically, we were raping Mother Earth. Stop me if you've heard this one before... and 2) we needed solar. Lots and lots of solar. I told stupid shit (I won't grace him by calling him a moron) that solar may be a workable solution on an individual basis, but on a regional level, no bleeding way. I mean, look at SoCal. We have approx 15M people here. Enough solar energy to power that would require covering the Mojave Desert with panels. It just isn't a practical solution. Consider your argument, and solar simply fails. Like I said- ok as an individual solution, not so much as a regional one.

Posted by: Bill H at November 01, 2008 10:24 AM (q8CmE)

21 way to crap on everyone's hopey-changey O-phoria, Obloodyhell!  Scalability and technological innovations will take care of alot of your concerns.  I suspect someone will find a way to recycle or reuse old solar cells and manufacturing waste.  Think of cost-performance for computers; 10 years from now, your pricing assumptions may be way off.  I saw some other reseachers developed this ripple shape for panels that would improve absorbsion over the day as the sun moves.  Hell they found worms that eat the heavy metals in British brownfields...

Don't think there is a need to store it if you can sell extra on the grid.  Daytime is peak usage anyway.  Batteries are most important for electric vehicles.

Also: 8 hrs/day for 10 yrs is about 29,000 hrs.  cost of $500/m2 and theoretical energy collection of 1kw/m2 is 1.7 cents/kwh.  That's cheap.  Even 50% efficiency gives us 3.4 cents/kwh.  still cheap.  Recycling?  Unless there are some chemical reactions going on, its a piece of silicon being recycled into... a piece of silicon.  Besides, we don't need to be getting 100% of our energy from this one source; it frankly wouldn't be as good in certain areas anyway - high latitudes, Seattle...

You have some special affinity for Delaware?  Drove through there a year or two ago and they charged me an assload in tolls for what was like a 5 minute drive.  My mind's made up... bulldoze it and set up solar cells!  I don't want to force a totally immature technology to market like Archbishop Algore, but Delaware deserves it after that crap.  I would have been fine with people just putting them on their roofs or back yards; thanks for the suggestion!

It's african-american silicon.  racists.  Or was that racist?  That whole thing confuses me.

Posted by: A.G. at November 01, 2008 11:14 AM (JoIvi)

22 Toxic waste. Semiconductor production is not a "clean" process.

Ummm, that is going to has changed ;-> 

In the future, you won't even need a traditional fab.  The ovens will be gone, clean rooms gone, tedious gazillion step processes will collapse down to small handful.  

The time from design to device will collapse to a matter of minutes.  Small scale manufacture of custom designed semiconductors will fit within a standard 40' cargo shipping container space.

What this means is the virtual fabs can become real real fabs if they want to, and the entry cost to do so will be under a million bucks for low volume manufacture capability.

Posted by: Purple Avenger at November 01, 2008 02:38 PM (OqXyp)

23 I know this because we're designing/building the gear right now ;->

Posted by: Purple Avenger at November 01, 2008 02:46 PM (OqXyp)

24 Purple: Are you allowed to say who you work for ?

Posted by: RebelWithoutAClue at November 01, 2008 06:12 PM (+RA9S)

25 Thanks for sharing, I found this story while searching for rock lyrics Advertising signs, useful comments and great points made.thanks for great informations It’s a wonderful Writing Board! Great site. A lot of useful information here neon signs California. I’m sending it to some friends!I’ve recently started a blog Los Angeles neon signs, the information you provide on this site has helped me tremendously. Thank you for all of your time & work.

Posted by: Advertising signs at January 07, 2011 10:10 PM (JsxNx)

26 Vibram shoesl is designed to cradle the foot, and not just protect it from the pounding, but Vibram Five Fingers shoes to optimize a Five Finger shoes runner's gait in such a way that a person can run longer and faster.Buy without worrying about injury.

Posted by: saly sa at April 11, 2011 05:08 AM (jQ3Gm)

Posted by: oakley uk at May 26, 2011 10:42 PM (aF7XF)

Posted by: oakley uk at May 26, 2011 10:43 PM (aF7XF)

Posted by: oakley uk at May 26, 2011 10:44 PM (aF7XF)

Posted by: oakley uk at May 26, 2011 10:44 PM (aF7XF)

31 Good post and straight to the point. I am not sure if this is actually the best place to ask but do you people have any ideea where to employ some professional writers? Thanks in advance drm removal software

Posted by: free drm removal at June 30, 2011 01:36 AM (o74nb)



Posted by: donghanjin at October 13, 2011 08:25 PM (Ia16k)

Hide Comments | Add Comment

Comments are disabled. Post is locked.
92kb generated in CPU 1.25, elapsed 4.2354 seconds.
62 queries taking 4.0263 seconds, 268 records returned.
Powered by Minx 1.1.6c-pink.