Could you suggest any if it's possible to deposit a layer of $\rm SiO_2$ via PVD?

Why doesn't it widely used? Is that correct that I need $~1600\ C$ in vacuum?

  • $\begingroup$ Can you clarify what exactly you mean by PVD? There are many techniques for producing thin film coatings, and a number of them could be called "Physical Vapor Deposition." I'm wondering whether you are referring to a specific type of process, or just to vapor deposition in general. $\endgroup$
    – Colin K
    Commented Dec 27, 2010 at 17:12
  • $\begingroup$ Well, in general. Mainly, Evaporative deposition(for example, on tungsten heater), then sputter deposition. $\endgroup$ Commented Dec 29, 2010 at 1:23
  • $\begingroup$ The wikipedia page for CVD has a whole section on depositing SiO2. (en.wikipedia.org/wiki/Chemical_vapor_deposition#Silicon_dioxide) I don't see any reason to believe that it's not possible or that it isn't widely used. $\endgroup$
    – wsc
    Commented Dec 29, 2010 at 2:09
  • $\begingroup$ If the substrate is Silicon it is simply grown. $\endgroup$
    – my2cts
    Commented Dec 2, 2021 at 19:43

2 Answers 2


$\mathrm{SiO_2}$ is very commonly used in optical coatings. A typical example is the production of Dielectric Mirrors which works just like a silver or aluminum mirror, but can achieve very high reflectivity, upwards of $99.9\%$, and can be made to work only at very specific wavelengths.

In this application $SiO_2$ (called silica, generally) is often used in combination with $HaO_2$ (hafnium oxide, or hafnia).Alternating layers are deposited on a glass substrate in very specific thicknesses to produce the desired optical qualities.

I'm not really an expert on optical coatings, but I work closely with people who are, and I don't believe that the temperature needs to be quite as high as $1600C$ . For typical processes, temperatures are in the range of $900C$ for the silica source, but much lower for the substrate being coated. More recently, I've heard that we are using some sort of plasma assisted coating technique, which I gather can operate at much lower temperatures, $400C$ or below. This is because the energetic ions in the plasma knock silica off of the source without much heating, and apparently produces higher quality coatings as well.

EDIT: As Georg pointed out, the material being removed from the coating source is not molecular silica, but some other combination of silicon/oxygen compounds. I assume that an $SiO_2$ crystal structure re-forms on the coated optic, but I do not really know for sure.

The wikipedia page on Silicon Dioxide is quite thorough, and has good references which should give you a good starting point if you want to do more research. I also found this list of books, and this book as well, although I cannot personally tell you if they are good books or not.

  • $\begingroup$ Could you please ask for temperature and books they would recommend? :-) $\endgroup$ Commented Dec 29, 2010 at 12:30
  • 2
    $\begingroup$ @BarsMonster: I edited the answer, but keep in mind this is second hand information. If you're going to do any serious work you should talk to an expert. $\endgroup$
    – Colin K
    Commented Dec 29, 2010 at 14:45

Yes, you can deposit $SiO_2$ with PVD, ion plating.

The lower the pressure the lower the evaporation temperature needed.

The question is what do you want to deposit the $SiO_2$ onto and how thick?

You can use an RF plasma ($13.56 \ \mathrm{MHz}$ at about $5-15 \ W /m^3$, $Ar$ gas) with $DC$ bias on the part. Use powdered $SiO_2$ in a tunsten or Ta boat with negative melting current. Shoot at about $5 × 10-4 \ \mathrm{millliTorr}$.

Bias the part, on an isolated steel or copper tray, with $30-40V$ $DC$ $+$ at up to $7 \ amps$.

The other side of the $DC$ circuit is the grounded vacuum chamber wall. You must ground everything to the chamber and then ground the chamber.

Look at www.lesker.com or svc.org for details on setup. Or you could send it to me or any number of coaters and they can coat it for you. Look up vacuum coating PVD.

  • $\begingroup$ Thanks for the details - I was thinking about depositing SiO2 as gate dielectric in ASICs, but now I realize that I would not get required interface quality compared to thermal oxide... $\endgroup$ Commented Jun 16, 2011 at 8:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.