For a given firing temperature, whether a glaze comes out glossy, matte, or somewhere in between depends on whether the glaze is saturated with flux. Matte glazes are flux saturated. Glossy glazes are not.
Defining the Terms
Dunting: Cracking of a ceramic body caused by thermal stress induced when the body is cooled too rapidly after firing.
Flux: A mineral which, when added to a glaze or clay-body recipe, lowers the melting point of the recipe.
Flux Supersaturation: The condition where the quantity of flux dissolved in a molten glaze exceeds the saturation point of the flux as the glaze cools. If the glaze is still molten when the saturation temperature is reached, crystals of flux aluminum silicate precipitate from
Saturation Point: With reference to a glass, the maximum amount of a flux that can be dissolved into a given glass composition.
Shifting a glaze from glossy to satin to matte is pure glaze chemistry, thus very geeky! Gloss to matte exists on a chemical continuum of flux element abundance relative to silica and alumina.
Viewing the analysis of a glaze in the unity molecular formula (UMF)1 format of ratios of flux to alumina to silica helps to understand how a glaze can be changed from gloss to matte. Take a hypothetical glossy cone-10 glaze. Its UMF is 1:0.4:3.0. That is, for each mole of flux it contains 0.4 moles of alumina (Al2O3) and 3 moles of silica (SiO2). The mole of flux is 0.1 sodia (Na2O), 0.2 potassia (K2O), and 0.7 calcia (CaO). The calcia is right at the limit in this glaze. Adding some whiting (CaO) to the recipe will likely tip the glaze from gloss to satin. Making a line blend with this glaze on one end and the same plus 10% added whiting on the other end will show just where the tipping point lies.
1 The original version of the Stull glaze chart, drawn by R. T. Stull in 1912, plots silica and alumina levels of a glaze by unity molecular formula. The chart depicts the properties of a group of glazes fired to cone 11. Silica in the glazes increases from left to right and alumina increases from bottom to top. Fluxes are 0.3 KNaO, 0.7 CaO. Republished with permission of The American Ceramic Society. ²
A matte glaze is supersaturated with flux elements. They precipitate from the cooling glaze and form flux aluminum-silicate crystals. As R.T. Stull showed over 100 years ago (1), when the ratio of silica to alumina in a glaze fired to cone 11 is less than 5:1 and the fluxes are 0.3 KNaO and 0.7 CaO (in UMF format), matte surfaces form. A glossy glaze is not saturated with flux elements so it solidifies as a shiny glass. A satin glaze is between the two and is just a matte glaze that’s only supersaturated by a small amount. Early industrial ceramic ware was praised if its fired glaze formed what was then called “good glass.” That meant the glaze melted fully at the firing temperature used.
We know that when fully melted glazes cool and harden to a matte surface, that result is because crystals have formed. The crystals are made up of flux elements plus alumina and silica.
The crystals of a particular flux form from a glaze because there is more of that flux element than the glass can keep dissolved as the molten glaze cools. If there were not such an excess, which is termed flux supersaturation, the glaze would be glossy.
Different flux elements may have different saturation points. The proportions of glass formers and modifiers can also change the saturation point for given flux elements. For example, magnesium oxide, MgO, may have a different saturation point in a low-alumina glass than in a glass with more alumina.
While testing is necessary to learn what these limits are for a particular glaze system, there are approximate limits published that give helpful hints about how much of particular fluxes to use in glazes intended for specific firing temperatures. These have been published by a number of ceramic researchers and are found in books, articles, and most commercial glaze calculation software.
2 This diagram suggests R.T. Stull’s original high-fire work with porcelain may be viewed more globally as an island of fully-melted glazes within a plot of glass forming oxides. Besides silica, oxides of boron and phosphorous are glass formers and should be included on the horizontal axis. Titania, TiO2, is included with alumina on the vertical axis. In the lower left corner there are not enough glass formers for a glaze to melt. At the upper right, glazes won’t melt because the refractory properties of silica and alumina overwhelm the fluxes. This generalization applies for any family of glazes of a specific ratio of fluxes matured to a specific firing temperature. Stull’s work (Diagram 1) shows ratios of SiO2:Al2O3 between 5:1 and 15:1 for glossy glazes with the fluxes and firing temperature Stull used. Those ratios are a good starting point for testing for glossy glazes even if B2O3 or P2O5 are included with SiO2 as glass formers. Matte glazes should be found at or below 5:1.
Fast Versus Slow Cooling
Formation of a satin or matte glaze surface can be affected by the rate at which the ware is cooled. Allowing the kiln to quickly cool 200–300°F (93–149°C) below the peak firing temperature, then holding or cooling slowly from there until the glaze solidifies can improve matte surfaces. The extended cooling cycle allows time for crystals to grow at the glaze surface. It’s necessary to cool the glaze a bit to permit some crystal formation in some glazes (but not all). Precisely how much cooling is necessary is a matter of trial and error. Some crystals form at peak firing temperature and slow cooling isn’t necessary. It depends on the glaze.
It is also sometimes possible to obtain a glossy surface from a matte-glaze recipe by rapidly cooling the firing to below the solidification temperature of the glaze. This process simply cools the molten glaze so fast there is no time for crystals to form in it.
There are risks involved, however. Crash cooling a kiln from the top of its firing cycle may cause dunting of the ware and can shorten the life of the kiln refractory. Depending on the kiln, the cooling rate may not be fast enough to prevent crystal growth in the glaze and the risks may not produce the desired result.
A more certain path to a glossy glaze is to increase the ratio of glass formers to flux above the flux-saturation point for the recipe. Finally, remember that no amount of slow cooling can cause crystals to precipitate from a glaze that is not flux saturated.
1 See Gebhart, Tina. 2011. “Techno File: Glaze Unity Formula,” Ceramics Monthly, Vol. 59, no. 6 (June/July/August): pp. 14, 15.
2 Stull, R.T. 1912. “R.T. Stull Chart I” from “Influences of Variable Silica and Alumina on Porcelain Glazes of Constant RO,” Transactions of The American Ceramic Society, Vol. XIV: pp. 62–70.
the author Dave Finkelnburg is a studio potter and practicing engineer. He earned his master’s degree in ceramic engineering from Alfred University.