Thermal Expansion in Ceramic Bodies and Glazes

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: thermal expansion in ceramic bodies and glazes, thermal expansions, thermal expansion, high expansion, high-expansion, low expansion, low-expansion, cte, coe - Learn more

Details

Coefficient of thermal expansion (CTE) is a measure of the rate at which a material changes dimension as its temperature changes. In ceramics we are normally concerned with linear thermal expansion, expressed as change in length per unit length per degree of temperature. A typical ceramic might have an average CTE of 6.5 × 10⁻⁶/°C (the same as 65 × 10⁻⁷/°C). For convenience this is often referred to simply as “6.5”. Higher numbers mean greater expansion during heating and correspondingly greater contraction during cooling.

These numbers appear very small, but over hundreds of degrees the dimensional change accumulates. More importantly, ceramics are rigid and brittle, so even very small differences in expansion between two bonded materials can generate large stresses.

This phenomenon most commonly comes to the attention of potters as crazing. A glaze having a higher thermal expansion than the body wants to contract more during cooling. Once it has cooled sufficiently to become rigid, the body prevents it from doing so and the glaze is put under tension — it is, in effect, being stretched onto the ware. If that tension exceeds what the glaze can withstand, it cracks to relieve the stress. Crazing reduces ware strength and creates a network of fissures in an otherwise smooth, cleanable surface.

The opposite condition occurs when the glaze has a lower thermal expansion than the body. The glaze is then placed under compression. A small amount of compression is desirable and can strengthen ware. Too much, however, can cause shivering, where flakes of glaze release from edges and contours, or can even put enough stored stress into the piece to fracture it. This is especially serious on functional ware because the flakes can be razor sharp.

Thermal expansion is also an important factor in the ability of ceramic ware to survive sudden heating and cooling. Low-expansion bodies generally have better thermal-shock resistance, although CTE is not the only factor — thermal conductivity, strength, elastic modulus, wall thickness, shape and the severity of temperature gradients also matter. Flameware bodies therefore employ mineral systems having very low thermal expansion, commonly cordierite, spodumene or other lithium aluminosilicate phases.

The remarkable influence of mineralogy can be demonstrated using SiO₂ itself. Fused silica, an amorphous glass, has an extremely low thermal expansion, around 0.5 × 10⁻⁶/°C. Crystalline quartz has exactly the same chemistry but expands many times more. In addition, at 573°C quartz undergoes its reversible alpha-to-beta crystal inversion, producing an abrupt dimensional change. Thus chemistry alone cannot predict the thermal expansion of a crystalline ceramic material — the mineral form in which that chemistry exists matters enormously.

With glazes, thermal expansion is mainly a product of chemistry because most of the fired material is glass. This is why oxide chemistry calculations are so useful for adjusting glaze fit. But chemistry is not the whole story. Undissolved particles, crystallization, phase separation and incomplete homogenization of the melt can all alter the actual expansion. The temperature at which the glaze becomes rigid during cooling also matters because stresses between body and glaze develop primarily after the glaze can no longer relieve them by viscous movement.

With clay bodies, thermal expansion is considerably more complicated. A fired body is not normally a homogeneous glass. Its microstructure can contain residual grains of quartz and other minerals, newly formed crystals such as mullite, glass produced by feldspar melting, pores and microcracks. The proportions and characteristics of all of these depend on the recipe, particle sizes, firing temperature, firing schedule and degree of vitrification.

Residual quartz is especially important. A body that preserves plenty of quartz particles can have a significantly higher expansion than one in which much of that quartz has dissolved into the developing feldspathic glass. This helps explain why the thermal expansion of a porcelain can change as it is fired toward greater vitrification even though its overall chemistry has not changed.

Can refiring change the thermal expansion of a clay body? Yes, it can. Another firing can further dissolve residual quartz, develop more glass, alter crystalline phases or change the microstructure in other ways. But the magnitude of the change depends on the body, its existing degree of maturity, the refiring temperature and the schedule. A refiring therefore does not automatically produce a large CTE change, but in some highly vitrified bodies it can be significant.

What about having a laboratory measure the CTE of your clay body and glaze? A dilatometer can provide extremely useful information, especially in manufacturing where body and glaze expansion curves are monitored consistently over time. But simply obtaining two numbers and trying to make them equal is not a reliable shortcut to glaze fit. The expansion curves can have different shapes, glazes have different set points, a slight compression is normally desirable, and firing history influences the result. The real value of dilatometry comes from building a history relating measured curves to the actual performance of ware. This is how ceramic manufacturers use it.

So how does a potter get a glaze to fit a clay body? Test the actual glaze-body combination. Physically stress fired ware to reveal crazing or shivering. If there is a problem, use glaze chemistry to move the calculated expansion in the required direction, make the adjusted recipe and test again. Repeat this cycle until the fired ware demonstrates the fit required. Calculated expansion tells you which way to move; physical testing tells you when you have arrived.