I was thinking about this, I made Gibbs free energy calculations and I can't figure out if

$$\ce{SiO2(s) -> Si(s) + O2(g)}$$

is actually possible, even at a ridiculous high temperature.

  • 4
    $\begingroup$ I suggest having a look at Si-O phase diagram, e.g. doi.org/10.1007/s11669-007-9062-5 (or search images). By the way, you shared neither your Gibbs free energy calculations nor the temperature you arrived at; however, the $\ce{SiO2}$ decomposition temperature definitely lies above 3000 °C at 1 bar. Not sure it this counts as "ridiculous high" though. $\endgroup$
    – andselisk
    May 15, 2020 at 19:58
  • 1
    $\begingroup$ Per @andselisk's comment the melting point of $\ce{Si}$ is 1414 °C, and $\ce{SiO2}$ is 1,713 °C so the decomposition reaction would have to be: $$\ce{SiO2(l) -> Si(l) + O2(g)}$$ $\endgroup$
    – MaxW
    May 15, 2020 at 21:41
  • 1
    $\begingroup$ @MaxW I think it would be rather gas phase, as andselisk mentioned temp. needs to be higher then 3000 °C and SiO2 has boiling point (i.e. it doesn't decompose before boiling) $\endgroup$
    – Mithoron
    May 15, 2020 at 22:18
  • 2
    $\begingroup$ Well, gaseous SiO2 has to decompose at high enough temperature, but reaction may be rather be: SiO2 (g)⟶SiO (g) + O (g) $\endgroup$
    – Mithoron
    May 15, 2020 at 22:28
  • 2
    $\begingroup$ Anything decomposes at high enough temperatures, though at that point you may end up with a gas of free atoms or a plasma. There is no silicon dioxide in the sun (though silicon monoxide has been detected in stars ) $\endgroup$
    – JanKanis
    May 16, 2020 at 10:25

1 Answer 1


Silicon dioxide is difficult to decompose directly, but it can be broken down in the presence of other reagents at more moderate temperatures. In particular, the reaction of silica with elemental silicon has been known since 1905:

In 1890, the German chemist Clemens Winkler (the discoverer of germanium) was the first to attempt to synthesize $\ce{SiO}$ by heating silicon dioxide with silicon in a combustion furnace.[1]

$$\ce{SiO2 + Si ⇌ 2 SiO}$$

However, Winkler was not able to produce the monoxide since the temperature of the mixture was only around 1000° C. The experiment was repeated in 1905 by Henry Noel Potter (1869–1942), a Westinghouse engineer. Using an electric furnace, Potter was able to attain a temperature of 1700° C and observe the generation of $\ce{SiO}$.[2] Potter also investigated the properties and applications of the solid form of $\ce{SiO}$.[3,4]

The reaction may be used to remove silica from ores.

Cited references:

  1. C. Winkler Ber. 23, (1890) p. 2652.

  2. J. W. Mellor "A Comprehensive Treatise on Inorganic and Theoretical Chemistry" Vol VI, Longmans, Green and Co. (1947) p. 235.

  3. U.S. Patent 182,082, July 26, 1905.

  4. E. F. Roeber H. C. Parmelee (Eds.) Electrochemical and Metallurgical Industry, Vol. 5 (1907) p. 442. Link

  • $\begingroup$ Cool, but that doesn't answer the question. Reduction is sure more reasonable, but very different thing. $\endgroup$
    – Mithoron
    May 15, 2020 at 22:16
  • $\begingroup$ Did I not make that qualification at the beginning? $\endgroup$ May 15, 2020 at 22:28
  • $\begingroup$ You sure did, but why talk about entirely different reaction? $\endgroup$
    – Mithoron
    May 15, 2020 at 22:29
  • $\begingroup$ I rashly assumed that this approach could best be explored by giving an example of a possible reaction. $\endgroup$ May 15, 2020 at 22:38
  • 3
    $\begingroup$ Well, this does suggest that SiO2 may decompose rather into SiO, so it's still can be valid point... $\endgroup$
    – Mithoron
    May 15, 2020 at 22:42

Your Answer

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

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