Using my standard crystalline Cone 9-10 base, I tested
5 common colorants, and their combinations. All the tests included
rutile, since some titanium is desirable for seeding the crystals, but
by itself it yielded a white background with gold crystals.
These are the recipes:
Note: When combining these glaze samples, an equal volume
of glaze was added of each. The resulting glaze thus is diluted to
one half of the strength of colorant.
That is, combining B and C above, the recipe would have 1.5 grams CuC and 1.0 grams of CoC (all the ingredients above these two are the same).
When viewing the tiles, consider that glaze application of small tests inevitably varies. In general, whether crystals were plentiful or not mostly depended on the thickness of the glaze. These tests were to show color response more than crystalline growth. They were all fired on the same shelf in the kiln, at the Cone 9-10 range.
In each group of tiles below, the first is the pure colorant
represented by the code letter, then the other combinations are shown.
I liked a number of the tests. Manganese made a lovely warm beige, but I tend to shy away from using it much, partially from the gasses it gives off in firing, which have become a health concern in recent years.
H65A H65AB H65AC H65ADH65AE
A=Rutile B=Copper Carb. C=Cobalt Carb. D=Manganese Diox. E=Red Iron
H65DE H65AD H65BD
H65AD A=Rutile B=Copper Carb. C=Cobalt Carb. D=Manganese Diox. E=Red Iron Oxide
Because Copper is a strong flux, some of the tests with Copper ran off the tiles (always a possibility with crystalline glazes). The Copper tile was the only glaze with some cratering, probably caused by glaze boiling. If you wanted a green glaze, I'd consider dropping the whiting if you're using Copper.
Brad's crystalline primer page