Thermal Properties

thermal shock

Take a glass dish from a hot oven and plunge it into cold water and it cracks with a snap. That sudden failure from a fast temperature change is thermal shock. The surface cools and tries to shrink, but the still-hot interior holds it stretched, and if the tension beats the material's strength a crack runs. Ceramics and glasses — strong but brittle — are the usual victims.

A rapid quench of delta_T sets up a surface stress of order sigma = E x alpha x delta_T / (1 - nu), where E is Young's modulus, alpha the expansion coefficient, and nu Poisson's ratio. Fracture happens when this reaches the material's strength. Turning that around gives a thermal-shock figure of merit R = sigma_f x (1 - nu) / (E x alpha), the maximum sudden temperature drop a material can survive. To resist shock you want high fracture strength sigma_f and low E and low alpha. Add high thermal conductivity k, which flattens the internal gradient before the stress can build, and you get a second merit index R' = R x k.

This is why fused silica and borosilicate (Pyrex, both low alpha) shrug off temperature swings that shatter ordinary soda-lime glass, why cookware and kiln furniture use low-expansion ceramics, and why silicon carbide (high k, moderate alpha) survives thermal cycling. Metals rarely suffer thermal-shock fracture because they yield and conduct heat well — they relieve the stress by bending a little, not by breaking.

Borosilicate lab glass (alpha about 3 x 10^-6) can survive a sudden drop of well over 100 degrees C, while soda-lime glass (alpha about 9 x 10^-6) typically cracks around 40 to 50 degrees C — the merit index R scales inversely with alpha.

Low expansion, not high strength alone, is what buys thermal-shock resistance.

Thermal shock is a brittle-fracture problem, so it depends on the worst flaw, not just the average strength — the same reason ceramic strength scatters and needs Weibull statistics. A single strength number can overpromise.

Also called
thermal shock resistance熱衝擊抗熱震性