Lithium carbonate is now ultra-expensive. Yet the glaze on the left needs it. Spodumene has a high enoughLi2O concentration to be a practical source here. It also has a complex chemistry, but the other oxides it contains are those common to glazes anyway. Using my account at insight-live.com, I did the calculations and got a pretty good match in the formulas (lower section in the green boxes). Then I made 10 gram balls and did a melt flow test at 2200F (notice the long crystals in the glass pools below the runways). Not surprisingly, this recipe is very runny, that's why the tiny yellow crystals grow during cooling, they produce the gold effect this recipe is known for. The spodumene version is very similar, perhaps better. The calculated cost shown is outdated, in 2022 for us it is $17.84 vs $10.40 per kg (based on purchasing 2.5kg amounts of the materials).

This was applied at leather hard stage on Plainsman M370, bisque fired on, dipped in clear G2926B glaze, then fired at cone 6. The transfer was purchasing online. Since the pigment contains cobalt it does feather somewhat at the edges, this would be less of an issue at low temperature.

Why? Glaze fit. These are available on Aliexpress (as Drip Pottery or Drippy Pottery) and they are made by a manufacturer that has close control of body maturity (and thus strength) and the capability to tune the thermal expansion fit of glaze-on-body. It has to fit better than normal because of the absence of an outside glaze. Too low an expansion and the compression (outward pressure) will fracture body (these are thin-walled pieces making them vulnerable). Too high and it will craze. And the glaze is thick, it will shiver or craze with far less forgiveness than a thin layer. And how did they get the glaze on this thick? They likely deflocculated it, up to 1.7 or more, glazed the inside, let it dry, then glazed the outside. These pieces are a visual and technical achievement. If you are a potter you had best think twice before attempting the same.

Potters are used to making dipping glazes that they weigh out and mix from recipes. Hobbyists commonly use bottled commercial brushing glazes. Did you know that a dipping glaze can be turned into a brushing glaze by the addition of Veegum (or Veegum CER) and water? Do you know what a base-coat dipping glaze is? Here is a quick overview: Dipping glazes need to go on to bisque ware evenly, be thixotropic enough to hold on at thickness and drain and dry quickly. But they don't need to dry hard. Brushing glazes need a cohesive slurry that dries slowly and hardens well on drying. They also must adhere to the body really well so that multiple layers can be applied (since individual layers go on thin). Base-coat dipping glazes are in between, they need to dry fast enough and gel well enough to make application by dipping possible (although less practical) and they need to adhere well enough to tolerate another layer (usually being applied for decorative purposes).

L4496, this clay, is the top two bars, fired at cone 4 and 04. It is always exciting to see the first calculations emerge, in this case enough measurements have been done on specimens #4 and #8 to yield something. The bottom bars are Plainsman L215 terra cotta. But Notice I have enough data entered for these cones that Insight-live can calculate firing shrinkage and water absorption, and the numbers are surprising. Specimen #8 has a firing shrinkage of -0.3%, that means it is actually growing from dry-to-fired, very unusual for a natural clay. Notice also the absorption is only 3.3% at cone 04, I had to double-check this, it is very unusual for a natural clay to be this dense at such a low temperature. Even though the surface of the L215 bar on the bottom is much smoother and denser appearing it is actually much less dense (at 12% porosity). The cone 4 numbers are also interesting. The clay is melting at cone 5 (not shown) but the cone 4 bar is not bloating or bubbling and its firing shrinkage is low.

These screen fragment overlays are from the recipe panel at insight-live.com. The table of data shown here is from the SHAB test only. The row numbers are the specimen numbers of each bar. The first five columns are the data we collected by measuring the bars before and after firing (dry length, fired length, dry weight, fired weight). The last three red columns are the results of calculations it does on that data to produce values for drying shrinkage, firing shrinkage and fired porosity. The graph above charts the firing shrinkage (ascending line) and absorption (descending line) against temperature. These two lines are like a "fired maturity fingerprint". Finding meaning in this data enables characterizing the firing behavior of the clay. In the next step, we will compare it to a terra cotta clay body.

These are the same glazes. The slurry of the one on the left had a specific gravity of 1.45, it was creamy and appeared to be good. However, when this bisque porcelain mug was pulled out after the dip it dried so fast that it would not even out around the lip (in spite of my efforts to roll it). To fix this I added water to increase the specific gravity to 1.43 (making it quite watery). Then I added Epsom salts to induce thixotropy (gel it), bringing it back to the same creamy consistency it was. This time it went on evenly and dried slowly enough that it evened out. Notice the darker color, is it still damp. Although the piece dries enough to handle in less than 30 seconds, it does take longer to dry completely because there is more water in the slurry.

These test bars are fired from cone 5 (top) down to cone 06 (bottom). We are processing hundreds of these bars are any given time, managing the simultaneous testing of dozens of body, glaze and engobe projects in our group account at insight-live.com. These took about a month to work their way through our system, all the measurement data has been entered (we will look at that in the next step). These bars show visually how this clay matures across a wide range of temperatures, from most-porous at the bottom to beginning-to-melt at the top. One thing is obvious: Most terra cottas shrink much more as they approach cone 2, commonly reaching 8%, then they begin to expand above that. This one is much more dimensionally stable, it is only shrinking about 3% at cone 2 (the #2 bar).

This is the G3939A recipe (a 90:10 mix of G2934 and G2926B), it normally produces a silky matte if not cooled too quickly. Shown on the left is our original addition of 8% Mason 6021 red stain and 4.5% titanium dioxide. This not only did not produce the desired marbled effect, it actually made it more glossy! A 1.5% titanium addition completely transforms it to what you see on the right. Rutile, as a source of TiO2, is often used for this, but it is high in iron and would completely muddy the red color. Pure titanium dioxide, by contrast, is iron free.

These porcelain mugs have the same glaze, the one on the left was fired at cone 10R (gas), the other at cone 10 oxidation (electric). This is our standard cone 10R magnesia matte, G2571B. We have added a 5% Mason 6021 encapsulated red stain and 4% titanium dioxide (producing recipe code number G2571D). While the reduction version looked good the oxidation one turned out much more vibrant. And it feels much better, being very pleasant to touch. The marbling is a bit excessive so in G2571D1 we reduced the titanium by 1% (and increased the stain by 1%). MgO matte base recipes are very receptive to this type of adjustment and they work across a wide range (from low to high temperatures). Titanium is much better for variegating bright colored glazes than rutile, because the latter contains lots of iron that muddies the color.