Thermal Shock Resistance of Metals and Ceramics: Mechanisms and Design

Thermal shock fracture occurs when a rapid temperature change generates constrained thermal strain faster than a component can relieve it elastically or plastically, producing stress that exceeds the material’s fracture resistance. This guide develops the Hasselman thermal shock parameters, explains why ceramics and refractory metals are especially vulnerable, and outlines the material selection and design levers used to prevent thermal shock failure in service.

The Physics of Thermal Shock Stress

When the surface of a body is suddenly cooled relative to its interior, the surface layer tries to contract by an amount alpha*deltaT but is restrained by the still-hot, still-expanded interior to which it remains bonded. This constraint converts the free thermal strain into an elastic tensile stress at the surface. For the idealized case of an infinite flat plate or the surface of a long cylinder under a step change in surface temperature, the peak stress is:

sigma_max = (E * alpha * deltaT) / (1 - nu)

where:
  E      = Young's modulus
  alpha  = coefficient of thermal expansion
  deltaT = temperature differential (surface - reference)
  nu     = Poisson's ratio

Fracture occurs when sigma_max reaches the material’s tensile fracture strength, sigma_f. Rearranging for the critical temperature differential gives the classical thermal shock resistance figure of merit, discussed further below and closely related to the residual stress concepts covered in our quenching and tempering guide.

Hasselman’s Thermal Shock Resistance Parameters

Hasselman formalized thermal shock resistance into two families of parameters: those predicting resistance to crack initiation, and those predicting resistance to crack propagation once fracture has started. The distinction matters because the two families depend on different, sometimes opposing, combinations of material properties.

Crack Initiation Resistance: R, R’, R”

R  =  sigma_f * (1 - nu) / (E * alpha)          [infinite Biot number, instantaneous quench]

R' =  R * k                                      [low Biot number, slow heat transfer]

R'' = R * (thermal diffusivity term)              [constant heating/cooling rate]

R defines the maximum survivable temperature differential under the most severe quenching condition (infinite heat-transfer coefficient). Because it is inversely proportional to both E and alpha, materials with low thermal expansion and comparatively low stiffness resist crack initiation best. R’ incorporates thermal conductivity, k, to account for the finite rate of heat transfer at moderate Biot number, favoring high-conductivity materials that reduce the actual temperature gradient achieved during a real quench.

Crack Propagation Resistance: R”’ and R””

R'''  =  E / (sigma_f^2 * (1 - nu))     [minimum elastic energy available for crack growth]

R'''' =  ( gamma_wof * E / (sigma_f^2 * (1 - nu)) )^(1/2)   [related to extent of crack growth]

where gamma_wof = effective fracture surface energy (work of fracture)

These parameters invert the strength dependence: once a crack has nucleated, a high elastic modulus and high fracture toughness limit how far the crack can run and how much residual strength is lost, while a very high strength (which stored more elastic energy at the moment of first cracking) can actually work against propagation resistance. This is why brittle-but-strong ceramics can show excellent R but poor R”’, catastrophically losing strength after a single severe thermal shock event even though they resisted first cracking well, a behaviour connected to the crack-driving-force concepts in our Paris Law fatigue crack growth article.

The Role of the Biot Number

The Biot number, Bi = h*a/k, compares the rate of heat transfer at the surface (governed by the heat transfer coefficient, h, and characteristic dimension, a) to the rate of heat conduction within the body (governed by thermal conductivity, k).

This is why thin-walled components and high-conductivity materials tolerate much faster thermal cycling than thick sections of low-conductivity material, and why furnace refractories are engineered specifically around thermal conductivity as much as strength.

Why Ceramics Are More Vulnerable Than Metals

Ceramics lack an effective plastic deformation mechanism at low to moderate temperature, so essentially all imposed thermal strain converts directly to elastic stress with no yielding to cap the peak value. Metals, by contrast, begin to deform plastically once local stress reaches the yield strength; this plastic flow both limits peak stress and dissipates strain energy, giving ductile metals a substantial practical margin over their brittle counterparts even when thermal expansion coefficients are similar or higher. The comparative behaviour parallels the ductile-versus-brittle fracture distinction developed in our fracture mechanics fundamentals guide.

Refractory Metals and the Ductile-to-Brittle Transition

Body-centred-cubic refractory metals, including tungsten, molybdenum, and to a lesser extent tantalum and niobium alloys, exhibit a ductile-to-brittle transition temperature (DBTT) that can sit above room temperature, particularly after recrystallization, neutron irradiation, or interstitial embrittlement. Below the DBTT, these metals lose the plastic stress-relief mechanism that normally protects ductile metals from thermal shock and behave in a manner closer to a brittle ceramic. This is a critical design consideration for plasma-facing components in fusion reactors, rocket nozzle throats, and high-temperature furnace elements, where rapid thermal transients combine with an operating window that may cross the DBTT during startup or shutdown.

Transformation Toughening in Zirconia Ceramics

Partially stabilized zirconia (PSZ) and tetragonal zirconia polycrystal (TZP) ceramics contain metastable tetragonal grains that transform to the monoclinic phase under the stress field ahead of an advancing crack tip. This stress-induced transformation absorbs fracture energy and generates local compressive stresses that resist further crack opening, substantially raising fracture toughness compared to conventional oxide ceramics such as alumina. Because R”’ and R”” scale favorably with fracture toughness, transformation-toughened zirconia shows markedly better thermal shock damage tolerance and residual strength retention after severe thermal cycling, even where the crack-initiation parameter R is not dramatically different from other oxide ceramics. This mechanism is developed in more detail in our dedicated guide to zirconia transformation toughening.

Material Design Strategies

• Minimize the thermal expansion coefficient where possible; low-expansion glass-ceramics and cordierite-based refractories owe most of their thermal shock resistance to this single variable.
• Increase thermal conductivity to reduce the actual gradient developed during a given quench rate, particularly important once Biot number is in the moderate range.
• For ceramics intended to survive repeated thermal cycling rather than a single severe event, prioritize fracture toughness (R”’ and R””) over raw strength, since strength alone can leave a component vulnerable to catastrophic crack growth after first cracking.
• Use compressive surface residual stress, from tempering, ion exchange, or surface treatment, to offset the tensile thermal stress that develops on rapid surface cooling.
• For metals operating near or below their DBTT, control grain size, minimize embrittling interstitials, and account for irradiation or thermal aging effects on transition temperature.
• Reduce component size or wall thickness and avoid sharp geometric discontinuities that concentrate thermal stress, in the same way stress concentrators are avoided in classical fatigue design.

Testing and Verification

The standard laboratory method is the water-quench test: specimens are heated to progressively higher temperatures and quenched into water held at a fixed lower temperature, with the critical deltaT at which residual strength drops sharply, or visible cracking appears, recorded as the practical thermal shock resistance. For high-heat-flux applications such as plasma-facing components or rocket nozzles, cyclic induction or flame heating tests with acoustic emission monitoring are used to detect crack initiation in real time. Residual strength after a defined number of thermal cycles is often reported alongside the single-shock critical deltaT, since these two metrics probe crack initiation and crack propagation resistance respectively, in keeping with the two-parameter Hasselman framework above.

Industrial Significance

Thermal shock resistance governs material selection for furnace refractories, foundry ladles, turbine blade coatings, brake rotors, glass and ceramic cookware, rocket nozzles, plasma-facing fusion reactor components, and rapid-cycle induction heating tooling. Selecting the correct material and geometry against the appropriate Hasselman parameter, matched to the actual Biot number of the service quench or heating event, prevents both single-event catastrophic fracture and slow cyclic degradation over the component’s operating life.

Frequently Asked Questions

Recommended Reference Reading

Disclosure: MetallurgyZone participates in the Amazon Associates programme. If you purchase through these links, we may earn a small commission at no extra cost to you. This helps support free technical content on this site.

Further Reading