Many of the best frits do have a published chemistry, or the chemistry is approximate. Can we still use them in glaze chemistry? Yes.
Is this possible? Yes. How? Relativity is the key. Consider thermal expansion. Rather than try to predict absolute thermal expansions for specific chemistry profiles, it is more practical to try move the chemistry of a glaze in a specific direction to adjust the fired expansion. For example, a crazing glaze needs to have the thermal expansion reduced to reduce the crazing. A cycle of calculations and then physical testing will achieve the desired goal.
An effective strategy to stop crazing is to reduce the amount of KNaO in a recipe and substitute fluxing oxides of lower thermal expansion (like MgO, Li2O, CaO, Sro). Suppose a recipe contains a frit of approximate chemistry (where the frit company releases a chemistry that shows a range for each of the oxides rather than exact amounts) also contains significant feldspar and silica. Without touching the frit, we are in a position to reduce the feldspar and introduce materials that source the desired fluxing oxide(s) (and then add or remove clay and silica to rebalance the SiO2 and Al2O3 amounts). Even if the mystery frit contains KNaO it does not matter, its contribution will be constant, we are adjusting with other materials that also supply it. But there are a guideline or two to follow to make success more likely:
-You must maintain the same total for the recipe and the mystery frit amount also must remain unchanged.
-Compensate when adjusting material amounts where a significant LOI is involved. For example, whiting has a 50% loss on firing. That would mean that for every 1% calcium carbonate lost the recipe total drops by 0.5%. Watch the LOI that Insight calculates, keep the total minus the LOI the same as the original. For example, if the original glaze has a 10% LOI that means you only actually have 90% that goes into making the fired glaze. If you make changes that produce a recipe having a 95% LOI, the recipe total is now greater and the frit is proportionately less.
-In your calculations, use an average of the oxide ranges given for the frit as if it were the actual chemistry, this enables you to make better decisions about the most appropriate kinds of changes you could make to the chemistry to achieve the desired goal. Often the amount of the frit may be less than 30% and thus the calculation of the final chemistry of the glaze is not altogether inaccurate.
By Tony Hansen