All Glossary

Co-efficient of Thermal Expansion

The co-efficient of thermal expansion of ceramic bodies and glazes determines how well they fit each other and their ability to survive sudden heating and cooling without cracking.

Key phrases linking here: co-efficient of thermal expansion, thermal expansions, thermal expansion, high expansion, high-expansion, low expansion, low-expansion, cte, coe - Learn more

Details

Co-efficient of thermal expansion (CTE) is a measure of the reversible volume or length change of a ceramic material with temperature. The more the expansion during heating the more contraction must occur while cooling it back down. Expansion values are very small and recorded in scientific notation (e.g. 6.5 x 10-7 which is 0.00000065). Typically the power-of-ten is dropped, so a number might simply be: 6.5. Higher numbers indicate higher expansion.

It might not seem like a CTE of 0.00000065 in/in/degree C is much but in a brittle material like ceramic it is. The phenomena of thermal expansion usually comes to the attention of a potter, for example, as a crazed glaze (having a crack pattern that is a product of the glaze being stretched onto the ware). This is a serious problem - crazing impacts food safety and ware strength. The opposite problem, shivering, is where a glaze is under compression; this is even more serious since sharp flakes of glaze can pop off ware into food and drink. Some people promote the idea of having a lab measure the CTE or your clay body and glaze to get a better fit. That is not practical (see more on that a couple of paragraphs down).

Thermal expansion is also the basic phenomenon behind why ceramic ware often cracks when suddenly heated or cooled. True flameware bodies have almost zero thermal expansion and some can withstand a flame directly on the surface. Bodies high in free quartz particles (by virtue of the percentage of silica in the recipe and not being too vitreous) have the highest expansion. Vitreous porcelains (where the feldspar glass has dissolved much of the silica) and bodies made from low-expansion minerals (like phyrophyillite, mullite) have the lowest expansions.

When heated from room temperature to 2000F fused silica (non-crystalline) exhibits a CTE of almost zero whereas quartz mineral, having the same chemistry, has an amazingly high expansion of 1.5%! By comparison fused alumina (at 1400C) is 0.9% and stabilized zircon 0.8% (also very high). It is possible to fit a glaze to an alumina body by formulating it to have a very high thermal expansion. Likewise, it is possible, although much more difficult, to fit a glaze to an ovenware body by formulating it to have an extremely low expansion.

With glazes, thermal expansion is mainly a product of their chemistry. But there is more. The degree to which all the material particles are dissolved in the melt, the degree to which the melt has become homogeneous and the degree to which phase separation and crystallization have happened on cooling in the kiln also affect the CTE. Understanding a specific glaze is thus a study of all the variables against their CTE, something that can only be done over time.

Thermal expansion of clay bodies is more complicated than for glazes. It is a product of the complex microstructure of fired ceramic (grains of unchanged minerals, others that have melted and flowed, others that have altered their crystal form, others that have reacted together to create new mineral species). That micro-structure is a product of many factors (e.g. the degree to which a body is vitrified, the firing schedule, the particle size and shape of the materials, the distribution of particle sizes) determine its complex microstructure and thus its thermal expansion. Does refiring ware affect CTE? Yes, it doubles the heat treatment on the piece. Like understanding glaze CTE, a history of data that relates all the variables to measured CTEs is needed.

So how do you get a glaze to fit a clay body? Glaze fit is dealt with on many pages on this web, we recommend physical stressing to reveal misfit (crazing or shivering), calculation to move the glaze in the direction needed to improve fit and then mixing and testing the new recipe. The cycle is repeated until fit is demonstrated.

Related Information

Substituting alumina in a clay body dramatically lowered thermal expansion

These are glazed test bars of two fritted white clay bodies fired at cone 03. The difference: The one on the right contains 13% 200 mesh quartz, the one on the left substitutes that for 13% 200 mesh calcined alumina. Quartz has the highest thermal expansion of any traditional ceramic material. As a result the alumina body does not "squeeze" the glaze (put it under some compression). The result is crazing. There is one other big difference: The silica body has 3% porosity at cone 03, the alumina one has 10%!

No crazing out of the kiln. But an ice-water test did this.

The side of this white porcelain test mug is glazed with varying thicknesses of V.C. 71 (a popular silky matte used by potters), then fired to cone 6. Out of the kiln, there was no crazing, and it felt silky and wonderful. But after a 300F/icewater IWCT this happened (it was felt-pen marked and cleaned with acetone). The glaze was apparently elastic enough to handle the gradual cooling in the kiln. However, the recipe has 40% feldspar and low Al₂O₃ and SiO₂, in a cone 6 glaze these are red flags for crazing.

No matter what anyone tells you, glaze fit can rarely be fixed by firing differently (that just delays it). If someone needs to cool their kiln slowly to prevent crazing it simply means the glaze does not fit - its needs to be adjusted to reduce its co-efficient of thermal expansion.

Simple dilatometric curve produced by a dilatometer

Dialometric chart produced by a dilatometer. The curve represents the increase in thermal expansion that occurs as a glass is heated. Changes in the direction of the curve are interpreted as the transformation (or transition) temperature, set point and softening point (often quoted on frit data sheets). When the thermal expansion of a material is quoted as one number (on a data sheet), it is derived from this chart. Since the chart is almost never a straight line one can appreciate that the number is only an approximation of the thermal expansion profile of the material.

Same body, glaze, thickness, firing. Same thermal shock. Only the tile crazed?

Why did the glaze on the tile craze? It is double the thickness of the walls of the mug. Thus, when quenched in ice water (BWIW test), a greater gradient occurs between the hot interior of the clay and the rapidly cooling surface.

Can you make things from zircopax? Yes.

Only 3% Veegum will plasticize Zircopax (zirconium silicate) enough that you can form anything you want. It is even more responsive to plasticizers than calcined alumina is and it dries very dense and shrinkage is quite low. Zircon is very refractory (has a very high melting temperature) and has low thermal expansion, so it is useful for making many things (the low thermal expansion however does not necessarily mean it can withstand thermal shock well). Of course you will have to have a kiln capable of much higher temperatures than are typical for pottery or porcelain to sinter it well.

How much silica can some glazes accept?

G2922B is a cone 6 clear glaze that started as a well-known recipe "Perkins Studio Clear". We substituted Gerstley Borate with a frit (while maintaining the chemistry) and then noted that the glaze was highly fluid. Since I wanted to keep its thermal expansion as low as possible, I added 10% silica. 2926B shows that it is very well tolerated. Then I added 5% more (2926D) and 10% more (2926E which is still very glossy). That means that E represents a full 20% silica addition! SiO₂ has no real downsides in any well melted glossy glaze, it hardens, stabilizes and lowers expansion.

Shivering on terra cotta is a red light not to ignore

Low fire terra cotta mugs have cracked. Why? The white glaze is under compression, its thermal expansion is too low (that is why it is also shivering off the rim). As the piece is cooling the kiln the thick layer of white glaze first solidifies. As cooling proceeds the body shrinks (thermally) at a faster rate than the glaze. The puts the glaze under compression and stretches the body. As some point (e.g. last stages of kiln cooling, a thermal stress during use) the body cracks to relieve the stress (notice how the white glaze is pushing the cracks apart). Neither the body or glaze are at fault, in this case they are simply made by different manufacturers and are thermal expansion incompatible. One solution would be to mix it with a white glaze that is crazing (the opposite problem). Or you could add some nepheline syenite to the glaze to increase its thermal expansion (maybe 10% by dry weight).

Dilatometer curve of vitreous porcelain (red) vs. stoneware body

The 500-600C zone is the alpha-beta inversion of quartz. Notice the vitreous body experiences a bigger expansion change there. But in the 100-270C cristobalite inversion region the stoneware undergoes a much more rapid change (especially in the 100-200C zone). This information affects how ware would be refired in production to avoid cracking (slowing down in these two zones). In addition, that stoneware would not be a good choice for an ovenware body. Photo courtesy of AF

High thermal expansion talc body cannot be COE-calculated

Talc is employed in low-fire bodies to raise their thermal expansion (to put the squeeze on glazes to prevent crazing). These dilatometer curves make it very clear just how effective that strategy is! The talc body was fired at cone 04 and the stoneware at cone 6. The former is porous and completely non-vitreous and the latter is semi-vitreous. This demonstrates something else interesting: The impracticality of calculating the thermal expansion of clay bodies based on their oxide chemistry. Talc sources MgO and low fire bodies containing it would calculate to a low thermal expansion. But the opposite happens. Why? Because these bodies are composed of mineral particles loosely sintered together. A few melt somewhat, some change their mineral form, many remain unchanged. The body's COE is the additive sum of the proportionate populations of all the particles. Good luck calculating that!

Potters can learn from how glazes are fit on ceramic tile

These are thermal expansion curves for body, engobe and glaze (from a dilatometer, a device that measures it against increasing temperature). The upper line is the body. The center line is the engobe. The lower line is the glaze. The ceramic tile industry is very conscious, not only of glaze-fit but also engobe-fit. Engobes (slips) are employed to cover brown or red burning bodies so they glaze like a porcelain. Typically technicians tune the formulation of the engobe to have an expansion between the body and glaze. The body is highest so that during cooling, as it contracts, it puts a squeeze on the engobe (the engobe thus never finds itself under tension). The glaze has the lowest expansion, it is under a state of compression by the engobe (and slightly more by the body). This equilibrium enables the tile to wear for many years without crazing or shivering. Chart courtesy of Mohamed Abdelmagid.

A high expansion glaze is bowing the foot of the bottom bowl

The glaze has a calculated thermal expansion of 8.8 (because of high KNaO and low SiO₂). Very high. It is basically stretched on. These plates are not glazed on the bottom. The glaze on the inside of the upper plate fits, the base is flat. But the glaze on the inside of the lower plate is pulling the base upward. The built-in stresses will eventually cause the piece to fail (likely fracturing into many pieces) if bumped. It is also almost certainly crazing. And the low SiO₂ implicates it for leaching. The solution? Reduce the KNaO in favour of MgO and increase the SiO₂ as much as possible without compromising the fired character.

Two matte mechanisms: One crazes the other does not

These two glazes look the same, they are both cone 6 satin mattes. On the same porcelain. But the matteness "mechanism" of the one on the left, VC71, is a low Si:Al ratio melted by zinc and sodium. The mechanism of the one on the right, G2934, is high MgO melted by enough boron to also have plenty of SiO₂ and Al₂O₃. The "baggage" of the mechanism on the left is high thermal expansion and crazing (drastically reducing strength and providing a space for a germ zoo). If your ware develops this your customers will bring it back for replacement. No change in firing will fix this, the body and glaze are not expansion compatible. Period.

Example of COE curves considered a good fit for body and glaze

These are from a sanitaryware plant in India. Long-term glaze fit is essential for their products. The glaze thus needs to be under some compression. That means the body must have a slightly higher coefficient of thermal expansion (COE) than the glaze. These two charts were created on the same dilatometer by the same person using well-defined procedures (the glaze and clay each have their own procedures). A history of measurements and associated knowledge of how the data relates to the quality of the fired products provides a context to interpret these reports. In other words, technicians have learned that the difference shown here is what is required to achieve optimal glaze fit for this specific body/glaze combination. Of course, some sort of database system (e.g. lab notebook, an account at insight-live.com) is needed to record the history of testing to be able to effectively compare the past with the present.

Inbound Photo Links

A dunting crack

Cone 10 mug is crazing after a year. That's OK because it's high-fire, right?

Links

By Tony Hansen Follow me on

Got a Question?

Buy me a coffee and we can talk