Sciencemadness Discussion Board

Silicon deposition

hashashan - 11-3-2010 at 10:37

Hello all,
does anyone know if it is possible to make silicon deposition onto a metal at home? i mean no high vacuum devices.

watson.fawkes - 11-3-2010 at 12:45

Quote: Originally posted by hashashan  
Hello all,
does anyone know if it is possible to make silicon deposition onto a metal at home? i mean no high vacuum devices.
"At home" for many of us does indeed include high vacuum. In practice, that means a roughing pump (such as a refrigerant service type) and a vapor diffusion pump. Not real cheap, but not hugely expensive either, if you're willing to do some fabrication. Don't be too afraid of vacuum, though. Thomas Edison was sputtering gold over a hundred years ago with the vacuum equipment of his day.

As to non-vacuum techniques, I'm at a loss. All the chemical techniques that come to mind involve reducing ionized silicon, which means strong reagents of one sort or another.

arghh - 14-3-2010 at 18:14

Hello
I have been lurking for some time now and have been waiting until I had something to contribute before posting. I was hoping that would be something from my experiments but this thread seemed like a good place to start.

Depending on what your level of experience is, what equipment you have on hand and what you want to do with the silicon on metal, you can probably produce crude films of silicon on metal without vacuum equipment. Commercially, silicon is deposited on metal and other substrates by a process called Chemical Vapor Deposition (CVD). The basic description of it is that either silane (SiH4) or some chlorosilane (usually Trichlorosilane SiHCl3) is decomposed on the substrate. In its most simple form, the substrate is heated above Silane's decomposition temperature of 420 deg C (Higher for chlorosilnes). There is a thread on production of silane at https://www.sciencemadness.org/whisper/viewthread.php?tid=12... Woleen has a writeup of the procedure on his website as well : http://woelen.homescience.net/science/chem/exps/silane/index... . When Magnesium and silica (or also metalurgical grade silicon - available as "Silicon Metal" at several hobyist friendly suppliers) is heated to a sufficiently high temperature, Magnesium Silicide Mg2Siis formed. Silane is produced by reacting the MgSi2 with HCl or H2SO4. Silane, being pyrophoric, explodes in little pops on contact with the air. No if one repeated Woleen's experiment (OUTSIDE) while blowing an inert sheilding gas like CO2 or Argon over it (like in mig or tig welding) while holding your heated metal substrate over the top of the watch glass, you might be able to produce a silicon film on it. You could heat the metal by passing a current through it. There is mention in the literature of atmospheric pressure CVD of silicon. There is not a lot of mention though and I suspect that this results in poor film quality for most electronic purposes. The electrical and physical properties of Silicon thin films are highly dependent on the deposition conditions and low pressure plasma enhanced CVD is usually used for device quality films.

I would caution against building some spiffy kipp generator type apparatus and unclosed CVD reactor unless you really know what you are doing. By all accounts, silane doesn't just combust on contact with air it explodes and I'm fairly sure it is heavier than air. If air gets into your enclosed apparatus, it will turn it into a pipe bomb. Also, unreacted silane will collect in the device and go kaboom when you open it up if it is not properly purged with an inert gas. And of course the purging will release the silane to the air where it will make a (hopefulluy) smaller but more prolonged deflegration where the puge gas hits the atmosphere.

Now a possible alternative to the above is sputtering. The vacuum requirements for it are not as steep as for a lot of other vacuum processes and you can get a single stage refrigerator service pump (good down to about 0.1 tor under realistic use) for 89 bucks at Harbor Freight. http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnu... . You would still have to build the rest of the sputtering apparatus and make sputter targets out of MG silicon somehow but it is not totally beyond the amateur. See the following link: http://www.sas.org/E-Bulletin/2003-06-27/labNotesAS/body.htm...

hashashan - 22-3-2010 at 01:58

I know these processes I wondered maybe there was somehitng simpler that will not require vacuum. If I wanted to use vacuum I guess the simplest way to coat Si would be vacuum arc deposition

arghh - 23-3-2010 at 20:47

Electrodeposition from an aprotic polar solvent might be a possibility that is within the home chemist's grasp: http://dx.doi.org/10.1149/1.2127237

The reagents used in the above seemed somewhat hard to get (and for SiHCl3, somewhat dangerous) but some creative substitution might yield a workable process. The biggest concern is that you will either have to dissolve a pyrophoric gas (Silane) in a flamable liquid (the solvent) or work with chlorosilanes which are volatile and release HCl on contact with water (like in eyes and lungs). Safety precautions for the silane route would include:
1.A very low flammability solvent in the smallest quantity necessary. Keep the quantity of silane low as well.
2. Keep the solvent under an inert gas blanket.
3. Have a fire extinguisher near by.
4. Do it outside or in a place that could contain any fire (fireplace, metal fume hood with metal exhaust) until one got a handle on how the silane coming out of the solvent behaves.
5. The solvent should not release anything toxic if burned either.
6. Degas the solvent when done. Possibly crash the solution with water to drive the last silane out. Again, not using enough silane to be dangerous if released all at once and not confining the vapors in anything too close and brittle are good precautions.

OK, so propylene carbonate (used in the article) is starting to look fairy good. I recall some journal article on synthesis from propylene glycol and urea with a mixed zinc and iron oxide catalyst. Does anyone have any other ideas on a good, safe solvent that is easily obtainable?