Silver vaporized in a propane oxygen flame condensed onto borosilicate glass is highly reactive to the flame atmosphere. The silver coating seems to disappear in an oxidizing flame yet turns yellows and browns in a reducing flame. The level of reduction and even the pressure in the gas line will dramatically change the appearance of the silver in the finished glass work.
What is causing this to happen? Is the vaporization process in the flame forming something other than elemental silver? Why does it change color/disappear in different flame environments? How is this effecting the outcome in final product? Any help is greatly appreciated! I use this glass coloring technique extensively, I teach other glassworkers how to use it but have very little understanding as to how or what is happening. An array of colors can be made by controlling the gas pressure and flame atmosphere. There is no data I could find that explains this behavior of silver. I want to add that if the glass and silver is not heated beyond the softening point of the borosilicate glass (about 1500f) the silver can be made to turn yellow/brown to clear over and over again by simply adjusting the amount of oxygen in the flame. It is a repeatable process. The amount of reduction in the flame when the silver is vaporized also has an effect on the outcome. Less reduction will give more transparent colors ranging from blues to greens reds and yellows while more reduction is associated with more yellows, white and more opaque or translucent colors.

  • 1
    $\begingroup$ Thin films of pretty much anything can have surprisingly different colors, for reasons having nothing to do with chemistry. $\endgroup$ Aug 21, 2019 at 18:50
  • $\begingroup$ Yes, this is true. I’d like to know those reasons! $\endgroup$ Aug 21, 2019 at 18:56
  • $\begingroup$ In one word: diffraction. $\endgroup$ Aug 21, 2019 at 18:57
  • $\begingroup$ How would this be so greatly effected by flame atmosphere ? $\endgroup$ Aug 21, 2019 at 19:00
  • 2
    $\begingroup$ Thickness of the layer is one of many obvious ways. Then again, besides all that thin film diffraction stuff, there may be some genuine chemistry going on, after all. $\endgroup$ Aug 21, 2019 at 19:04

1 Answer 1


Gold nanoparticles have been used to make ruby glass for hundreds of years. The color is due primarily to surface plasmons. In a similar way, the color of metallic silver varies with particle size. If the particles are large, there is a large conduction band and the silver is highly reflective across the spectrum. Nanoparticle silver, though, exhibits various colors due to surface plasmons. According to that reference, 10 nm particles would absorb best at ~400 nm wavelength... i.e., absorb blue, leaving yellow reflected.

That said, there is certainly oxidation and reduction of silver, which is much more reactive than gold, in different parts of the flame, as you surmise. Perhaps various ionic compounds of silver form , such as $\ce{Ag2SiO3}$, silver silicate. It is even suspected that gold might react in glass!

It's hard to know which effect predominate, colored compounds or surface plasmons. Perhaps could add $\ce{Ag2SiO3}$ to a glass frit and heat it by radiation or conduction in vacuo to avoid oxidizing or reducing the compound, and check its color? That would be interesting to report here!

  • $\begingroup$ Very nice answer! And the Lycurgus cup goes back to the 4-th century. +1! $\endgroup$
    – Ed V
    Aug 23, 2019 at 2:23

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.