Principle of the Laser/Light Flash Thermal Conductivity Analyzer (LFA)

Overview Thermal transport testing methods are broadly divided into steady-state and transient techniques. Steady-state methods (such as heat flow meter, guarded heat flow, and hot plate) determine thermal conductivity directly from Fourier’s law, but they operate over relatively narrow temperature and conductivity ranges. They are best suited to medium temperatures and materials with low to moderate thermal conductivity. Transient methods offer wider applicability, particularly for high-conductivity materials and high-temperature measurements. Among them, the flash method (also called the laser flash method) is the most widely adopted and validated in the thermophysics community. The flash method requires small specimens and covers a broad measurement range, enabling characterization of most materials except highly insulating products. It is especially effective for materials with medium to high thermal conductivity. Beyond conventional solid discs or plates, appropriate fixtures, sample containers, and thermal models allow measurement of liquids, powders, fibers, thin films, molten metals, coatings on substrates, multilayer laminates, and anisotropic materials. Relevant standards include: ASTM E1461 (Standard Test Method for Thermal Diffusivity of Solids by the Flash Method), DIN EN 821, and DIN 30905. Principle of the Flash Method The flash method directly measures thermal diffusivity (α). At a set temperature T (isothermal conditions controlled by a furnace), a laser or xenon flash delivers a short energy pulse uniformly to the specimen’s rear surface. The surface absorbs the pulse, its temperature rises instantaneously, and heat propagates one-dimensionally through the thickness toward the front surface. An infrared detector continuously records the temperature rise at the center of the front surface, producing a transient temperature (detector signal) vs. time curve. Under ideal assumptions—a near-instantaneous pulse, purely one-dimensional through-thickness conduction without lateral heat flow, and perfectly adiabatic surroundings without heat loss—the front-surface temperature would rise to a maximum and then remain constant. In that ideal case, the thermal diffusivity is obtained from the half-rise time t50 (also noted as t1/2), defined as the time for the front-surface temperature to reach one half of its maximum increase after the pulse. The Parker relation is used: α = 0.1388 × d^2 / t50 where d is the specimen thickness. Deviations from ideality require model-based corrections. Examples include boundary heat losses, radiative losses from surfaces and along the radial direction, nonuniform irradiation that induces radial heat flow, sample translucency or insufficient surface coating leading to partial transmission or subsurface absorption of the pulse, and finite pulse width effects when t50 is very short. From Diffusivity to Conductivity Thermal conductivity λ(T) relates to diffusivity via: λ(T) = α(T) × Cp(T) × ρ(T) Given α, specific heat Cp, and density ρ at temperature T, conductivity can be calculated. Density is typically measured at room temperature; temperature dependence can be corrected using coefficients of thermal expansion, with a simultaneous correction for thickness d if needed. For modest temperature ranges and small dimensional changes, density is often approximated as constant. Specific Heat by Comparative Method in LFA Cp may be sourced from literature, measured by differential scanning calorimetry (DSC), or determined in the LFA using a comparative approach alongside α. The comparative method uses a reference standard (std) with known Cp, matched to the sample (sam) in cross-sectional shape, similar thickness and thermal properties, and comparable surface finish. Both are coated to ensure similar optical absorption and infrared emissivity, then measured sequentially under the same conditions. By definition: Cp = Q / (ΔT × m) Thus: Cpsam / Cpstd = (Qsam / (ΔTsam × msam)) / (Qstd / (ΔTstd × mstd)) With identical pulse energy and identical absorbed area and absorptivity at the rear surface, Qsam = Qstd. If the environment, detected area, and emissivity are consistent at the front surface, the conversion factor between temperature rise ΔT and detector signal change ΔU is constant, so ΔT can be replaced by ΔU: Cpsam / Cpstd = (ΔUstd × mstd) / (ΔUsam × msam) Given Cpstd, mstd, and msam, and reading ΔU from the plateau of the adiabatic response, one obtains: Cpsam = Cpstd × (ΔUstd × mstd) / (ΔUsam × msam) Heat-Loss Corrections and Practical Notes In real experiments, conditions deviate from adiabatic. Heat losses occur during the temperature rise, so the signal does not form a perfect plateau. Even the peak temperature rise ΔTmeas differs from the ideal adiabatic ΔTadiabatic. Therefore, a heat-loss correction must be applied to obtain a corrected temperature rise ΔTcorr for Cp calculations. Commercial LFA software (e.g., NETZSCH LFA Proteus) incorporates these corrections within the diffusivity and Cp evaluation workflow. If the pulse energy differs between the standard and the sample, an appropriate scaling factor must be introduced into Q. If detector gain differs, a scaling factor must be applied to ΔU. These instrument-specific adjustments are handled by the evaluation software.

This article was originally published on our Chinese website. You can view the original version below: