In ceramics color is often a matter of chemistry, that is, the host glaze must be compatible and have a sympathetic chemistry for the stain being added. Chrome-tin stains are a classic example.
Coming up with a pink glaze may sound like an easy matter. All you do is find a good transparent glaze for the needed temperature, mix in 5%-10% pink stain, and you have it; right?
In the real world that most of us live in, things don't work out quite that well. While it is possible to obtain some sort of blue, green or brown glaze using almost any transparent recipe and the appropriate stain, chrome-tin pink stains do not work this way. The color will only develop in a melt that has a chemistry that is sympathetic to the specific mechanisms under which the color develops.
In a quest for a transparent recipe that will perform well with chrome-tin colors like lilac, coral, maroon, and pink, potters have traditionally used the "text-book junkie" approach. With this method, one rounds up tons of transparent recipes and tests each until both the correct surface quality and color are achieved (you also wind up with a room full of materials you will never use again). This method may succeed on the first test but it may not work until the 101st!
What we need is a more thoughtful, fast-track approach, leaving the trial and error stuff for fine-tuning the final product. We also need to understand why the final recipe works, and even better, why others are unsuccessful.
The key to solving these problems is to adopt the Kiln God's formula viewpoint. He builds ceramic structures from oxides like calcia, soda, alumina, and silica. Again, he obtains the oxides from the materials we supply in a glaze and ,for most purposes, does not care what material sources a particular oxide. It is reasonable then, to appraise fired results on the basis of the oxide formula, not the recipe of powders used to mix the raw glaze.
My first step in making the pink work was to mix the stain into my favorite cone 6 transparent glaze and fire a test. Of course, the result was a disaster; it was gray instead of pink! The question was: Why? The secret to the answer lies in examining the formula, not the recipe. First, I calculated the formula using INSIGHT.
Following are the results of the calculation in a detailed format.
DETAIL PRINT - 3134 Clear Glaze MATERIAL PARTS WEIGHT CaO* Na2O* B2O3 Al2O3 SiO2 WEIGHT OF EACH OXIDE 56.1 62.0 69.6 102.0 60.1 ----------------------------------------------------------------------- MATERIAL Expan OF EACH OXIDE 0.15 0.39 0.03 0.06 0.04 Cost/kg ----------------------------------------------------------------------- ------- FERRO FRIT 3134.... 50.00 190.64 0.18 0.08 0.17 0.39 2.49 KAOLIN............. 30.00 258.14 0.12 0.23 0.24 SILICA............. 20.00 60.00 0.33 0.19 ----------------------------------------------------------------------- ------- TOTAL 100.00 0.18 0.08 0.17 0.12 0.95 1.36 UNITY FORMULA 0.68 0.32 0.63 0.44 3.64 PER CENT BY WEIGHT 10.43 5.43 11.99 12.36 59.79 Cost/kg 1.36 Si:Al 8.21 SiB:Al 9.63 Expan 6.89
Next, I phoned the stain manufacturer, Mason in this case, and asked what to do. Fortunately, all answers were given from the formula viewpoint. They told me that the CaO content had to be at least 10% but that 15% would be better to assure color development. B2O3 should not be too high and a little extra CaO would help counteract the solvent action of B2O3 on the color development. I also checked a few textbook indexes under the topic "Chrome Tin Pink" and found a reference in Parmelee that confirmed this. It got more specific, stating that "Calcium oxide tends to counteract the ill effects of B2O3 , particularly when the ratio of CaO:B2O3 can be at least 3:1". Little bits of light were appearing. These facts seem obvious now but they are a commentary on how helpless we sometimes are.
I increased the calcium oxide, then reduced the B2O3 formula value to 1 / 3 of the CaO amount. Now that a major flux had been reduced in the glaze, I also determined that it would be necessary to cut the SiO2 and Al2O3 to harmonize with more traditional cone 5 limits (although I was conscious of needing the SiO2 to keep the expansion down and retain the gloss). Also I introduced a little MgO to help reduce higher expansion caused by the loss of Al 2O3and SiO2 .
The calculations produced a new recipe as follows.
WHITING 15.00 CaO 0.75* FERRO FRIT 3134 25.00 MgO 0.12* KAOLIN 20.00 Na2O 0.13* SILICA 20.00 B2O3 0.26 TALC 5.00 Al2O3 0.24 ======== SiO2 2.29 85.00 Cost/kg 0.87 Si:Al 9.49 SiB:Al 10.56 Expan 6.97
With bated breath and unwarranted confidence, I tested this glaze. The results: Complete failure! No color at all! Ah, but I was learning! The 3:1 CaO:B2O3 and 15% CaO alone were not enough to achieve the color. What is more, the glaze was crazing badly. The measures I had taken to make up for the loss of SiO2 and Al2O3 were not enough. My 'Parmelee' textbook also mentioned that the SiO2 has a stabilizing influence and can also reduce the solvent bleaching action of B2O3 on the development of color.
I ran another test, increasing the SiO2 and MgO to lower the expansion.
WHITING 14.90 CaO 0.69* FERRO FRIT 3134 28.60 MgO 0.17* KAOLIN 22.30 Na2O 0.13* SILICA 26.30 B2O3 0.26 TALC 8.00 Al2O3 0.24 ======== SiO2 2.53 100.10 Cost/kg 0.85 Si:Al 10.62 SiB:Al 11.72 Expan 6.65
What were the results? Again, disaster! This time, I even lost the gray color, ending up with nothing but a clear glaze even with 10% pink stain! There is no question that I was learning. As Thomas Edison said when asked what he had learned in life: "I have discovered thousands of things that don't work!"
Well, back at the textbooks, I kept reading about the detrimental effects of zinc on this type of glaze, however, there was no zinc to remove.
Then I found another clue. One short sentence in the middle of a 6-page marathon of chemistry on chrome-tin stain topics said: "This stain also is unstable in the presence of MgO." I had introduced this very oxide early on to help keep the expansion down and had increased it in a subsequent test.
Few things disagree with magnesia but it was worth a try. So I removed the MgO and replaced it with even more CaO.
Here is how the recipe looked
Pink Glaze Base #4 ================== WHITING............. 37.70 18.85% FERRO FRIT 3134..... 55.80 27.90% KAOLIN.............. 35.50 17.75% SILICA.............. 50.70 25.35% CORNWALL STONE...... 20.30 10.15% ======== 200.00
CaO 0.84* 18.29% MgO 0.00* 0.02% K2O 0.01* 0.49% Na2O 0.15* 3.63% Fe2O3 0.00* 0.03% TiO2 0.00 0.01% B2O3 0.26 7.18% Al2O3 0.24 9.70% SiO2 2.59 60.65%
Cost/kg 0.81 Si:Al 10.61 SiB:Al 11.69 Expan 7.23
Guess what? I hit the jackpot! This time 10% pink stain gave color that would knock your eyes out. However, there was some crazing on porcelain, but this was not a problem since the glaze melts well and thus there is room for an increase in SiO2 or the B2O3 :Al2O3 :SiO2 complement, neither of which should affect color.
This project really demonstrated that you can learn much when something is unsuccessful. Every time a fired product comes out of the kiln, whether success or failure, you can relate its appearance to the formula and learn something valuable.
This shows clearly how well the M version works with a chrome-tin stain compared to the others. However the 6100 brown stain works best in the N recipe (which have MgO). Notice also that the M has a higher thermal expansion than the others.
Left: a cone 6 matte glaze (G2934 with no colorant). Middle: 5% Mason 6006 chrome-tin red stain added. Right: 5% Mason 6021 encapsulated red stain added. Why is there absolutely no color in the center glaze? This host recipe does not have the needed chemistry to develop the #6006 chrome-tin color (it lacks sufficient CaO and has alot of MgO). Yet this same matte glaze intensifies the #6021 encapsulated stain at only 5% (using 20% or more encapsulated stain is to develop the color is not unusual).
Dealing With Chrome Flashing
The development of chrome tin pinks is a combination of the stain and the right chemistry in the host glaze. Unwanted pink flashing occurs where there is a hostile chemistry in the glaze.
|Oxides||SnO2 - Tin Oxide, Stannic Oxide|
|Oxides||Cr2O3 - Chrome Oxide|