Cooling Dave Finkelnburg

Instructions

An often overlooked part of firing ceramic work is cooling the kiln. Surface, color and integrity of the work are all affected by the rate at which the kiln is cooled.

Defining the Terms

Cristobalite: (inversion @ 428°F) The mineral cristobalite is a high-temperature polymorph of quartz—it has the same chemical composition as quartz but a different crystal structure.

Dunting Point: A crack or fault that can occur during the firing as a result of a thermally induced stress. Dunting is caused when a ware is cooled too quickly past the temperatures at which silica undergoes a shift in crystalline structure from beta back to alpha.

Quartz Inversion: Crystals of alpha-quartz turn to beta-quartz between 1022°–1063°F and the reverse occurs during cooling over the same temperature range. During heating, there’s a 2% volume expansion of crystalline quartz between 1070°–1110°F, and an identical 2% contraction on cooling.

Thermal Expansion and Contraction: Thermal expansion is the amount of change in volume in response to a change in temperature. When a substance is heated, atoms within it move more and thus maintain a greater average separation. The degree of expansion divided by the change in temperature is called a material’s coefficient of thermal expansion (CTE).

Thermal Shock: Stress in an object caused by a rapid change in temperature. When a ceramic object, which is brittle by nature and also a good insulator, is exposed to a very rapid change in temperature, the thinnest, most exposed parts of the object changes temperature first. These heated parts expand while the cooler parts do not expand as much. The different amounts of expansion in different parts of the piece can create stresses large enough to cause the piece to crack.

What Goes Up Must Come Down

A glaze is a glass formed by melting its ingredients on the surface of an object during firing. All glazes are the same while they are molten—they are liquid glass. Some glazes remain glassy when they cool provided they are fully melted during the firing.

Crystals precipitate from a glaze when the glaze is saturated with one or more chemical elements in concentrations too high to remain in the glass as it cools. The effect is crystals in the glaze. Below the saturation temperature, atoms of the elements attempt to move through the molten glass toward points where they can link up with other elements to form crystals. Just as sugar crystals precipitate from a strong syrup over time while the syrup is liquid, glaze crystals can only form while the glaze is molten. Once the glaze sets, atoms within it can no longer rearrange themselves into crystals. These crystals may form anywhere from the glaze surface to the contact layer between the glaze and clay body. The starting point for the crystals is always some seed material (zinc, silica, frit) that is in the glaze or on the body or tiny crystals that nucleate spontaneously from the melt.

Matte glazes composed of crystals, which precipitate from the glaze melt, require time to develop. Rapid cooling may actually prevent the formation of matte glazes. Cooling a glaze quickly can prevent the elements that would make up crystals from rearranging themselves into the crystals. Cooling a glaze quickly does not guarantee a glossy surface because some glaze compositions just want to crystallize. On the other hand, slow cooling will not guarantee development of a matte glaze. The time necessary for crystal development and the temperature at which the crystals grow is frequently different for each glaze and may even be affected by the clay body.

We have said this before, but it bears repeating; the rate of cooling does not cause crazing. Crazing is a fundamental mismatch between the coefficient of thermal expansion (CTE) of the body and applied glaze. It occurs because the glaze shrinks more than the body during cooling but the rate of cooling has no effect on the CTE of either body or glaze. Thus slow cooling cannot prevent crazing. However, fast cooling may reveal crazing that otherwise might not have been apparent for days, weeks or even months.

So how fast can you cool? As fast as you want, as long as the rate of temperature fall does not dunt the work and the surface results are what you want. If damage is observed due to thermal shock, most often from rapid cooling through quartz inversion (1100°–1000°F), then it is necessary to slow down the rate of cooling. The rate of cooling depends on the amount of quartz in the clay body, the thickness of the ware, and its shape. Cooling fracture is due to a differential in thermal contraction. The first part of the ware to cool to the quartz inversion temperature converts from beta to alpha quartz with a consequent volume shrinkage while the hotter center of the ware has yet to reach that point and shrink. In extreme cases this cracks the ware. Wide, flat rims are prone to radial cracks while tall, small diameter forms are much less sensitive to it.

Cool Work

All glazes, in general, may benefit from a hold or soak at peak firing temperature to help heal glaze flaws such as pinholes and blisters. Glossy glazes, however, are usually not improved by slow cooling. Once the surface of a glossy glaze has fully smoothed out it can be cooled. The first 600º of cooling of high-fire glossy glazes usually is best if done as rapidly as the kiln and the ware permit.

Mid-range and high-fire matte glazes formed from precipitating crystals seem to develop interesting matte surfaces when held or cooled slowly from somewhat below 2000ºF down to about 1500 ºF.

Most electric kilns cool too fast for maximum crystal development in the glaze, thus requiring some firing down (manual control of the rate of cooling) to achieve the greatest possible surface variation due to crystal growth. Since every kiln load, glaze, and ware combination may be a bit different, experimentation, testing, and record keeping are essential to determining the cooling cycle that produces the best matte glazes under particular circumstances. Holding a high-fire or mid-range matte glaze for two hours at 400°–600ºF below peak firing temperature is a good start for such testing.

If you fire a gas fueled kiln, it is important to note that the cooling cycle is actually a very long period of oxidation. Once the gas is turned off, the kiln atmosphere immediately is oxidizing. When you close the exhaust damper at the theoretical end of the firing, you are actually prolonging a period of oxidation at an extreme temperature by minimizing the kiln’s ability to cool. In effect, you are forcing the kiln to cool more slowly. Unless you intentionally create a reduction atmosphere, all gas-kiln cooling is in oxidation. Just closing the damper with the fuel off will not maintain a reducing atmosphere due to secondary air leaks. To circumvent this, fuel firers sometimes cause reduction during the cooling cycle by introducing fuel at a low rate and starving it of air.