Thermal inertia

Thermal inertia is a material parameter used to describe resistance to rapid temperature change during heating. In the SFPE treatment of solid conduction, it is the product of thermal conductivity, density and specific heat capacity. It is particularly relevant to the response of an exposed surface while heat is conducted into the material beneath it.

Two numerical conventions occur in the sources. Rockett and Milke call the product kρc thermal inertia, while Van Rossum uses its square root, written b, for the surface-heating comparison. These expressions contain the same underlying properties, but their values and units differ. A quoted thermal inertia is therefore meaningful only when its definition is identified.

Thermal inertia helps explain why a material can have a rapidly heating surface while its interior remains relatively cool. It also appears in ignition and compartment-heating calculations. It does not independently determine ignition, a component's temperature history or its fire resistance; those calculations require further exposure and material information.

Identifying the product and square-root conventions

Thermal conductivity describes the transport of heat through a temperature gradient. Density and specific heat capacity together describe the energy associated with changing the temperature of a material volume. Thermal inertia combines these transport and storage properties through their product. Its use therefore concerns more than the total mass of an object considered in isolation.

The product convention can be written as I = kρc. The symbol I is used here to keep that product distinct from the square-root convention b. Taking the square root of the product gives Van Rossum's parameter. Conversely, squaring b returns the product convention used in the SFPE chapter.

The product and square-root forms describe related quantities with different values and units. Based on Rockett and Milke (2002), p.1-37, and Van Rossum (2027), §5.2.

I = k ρ c

b = √(k ρ c) = √I

Symbol	Description	Unit
k	Thermal conductivity	W/(m K)
ρ	Density	kg/m³
c	Specific heat capacity	J/(kg K)
I	Thermal inertia in the product convention	W² s/(m⁴ K²)
b	Square root of the product	J/(m² K √s)

Specific heat capacity is measured per unit mass, whereas the density converts it into a property per unit volume. Neither of these quantities alone determines the surface response. Conductivity governs the transfer of energy away from the exposed surface into the solid. The product brings the three properties together in the transient surface-heating relationships.

The units provide a practical way to distinguish the conventions. A value carrying square-root time in its units belongs to a different numerical quantity from a value carrying time in the product units. Squaring or taking a square root changes both the number and the units. Merely copying a value into a formula labelled thermal inertia does not establish compatibility.

Material-property units also need to remain consistent. The equations here use conductivity in watts and specific heat capacity in joules. NUREG-1805 expresses some thermal-inertia quantities using kilowatt-based units. These are representations of the same property combination, but the numerical values cannot be exchanged without the associated unit conversion.

Why surface heating depends on the interior

An exposed solid does not generally warm everywhere at once. The surface receives heat first, while conduction carries part of that energy deeper into the material. Energy is also stored as the material temperature rises. Van Rossum therefore separates the speed of surface heating from the rate at which the temperature disturbance spreads into the interior.

A low square-root inertia parameter allows a relatively rapid surface-temperature increase under the conditions discussed in the introductory book. A material with a higher parameter can absorb and conduct more heat into its interior while its surface temperature rises more slowly. The comparison concerns the response to heating, rather than a guarantee about combustibility or a prescribed limiting temperature.

Rockett and Milke illustrate the role of the product using an idealised thick solid. Its initially uniform temperature is followed by a sudden change to a maintained surface temperature. Heat continues flowing into the interior even though the imposed surface temperature stays fixed. The required conductive heat flux depends on thermal inertia and decreases as the heating period progresses.

For a positive temperature rise ΔT between the maintained hot surface and the initial interior, the magnitude of that conductive flux can be written as follows. The expression contains the square root of the thermal-inertia product divided by time. It represents the stated temperature-step boundary condition, rather than a constant imposed heat-flux experiment.

q″(t) = ΔT √[I / (π t)]

Symbol	Description	Unit
q″	Magnitude of heat flux conducted into the solid	W/m²
ΔT	Maintained surface temperature minus initial solid temperature	K
t	Time after the idealised temperature change	s

At a fixed elapsed time and temperature rise, the required flux scales with √I, or b. At fixed material properties and temperature rise, it decreases as one divided by the square root of time. The ideal equation is singular at the instant of the temperature step. That feature belongs to the imposed instantaneous change and does not describe an infinite measured fire heat flux.

The same source discusses a different case in which an external heat flux heats a solid that also loses heat through convection and radiation. That case requires a different temperature solution. The distinction prevents a prescribed surface temperature from being confused with a predicted surface temperature. Thermal inertia participates in both problems, while the boundary condition determines the appropriate relationship.

Worked example using wood and concrete

Van Rossum compares a thick wooden wall and a thick concrete wall using rounded material properties. The wood has conductivity 0.13 W/(m K), density 500 kg/m³ and specific heat capacity 1,600 J/(kg K). The concrete has conductivity 1.7 W/(m K), density 2,300 kg/m³ and specific heat capacity 900 J/(kg K). These are the inputs for this particular comparison.

Multiplying the wooden-wall properties gives a product of 104,000 W² s/(m⁴ K²). Taking its square root gives about 322 J/(m² K √s). The concrete calculation gives a product of 3,519,000 W² s/(m⁴ K²) and a square-root parameter of about 1,880 J/(m² K √s). The arithmetic uses each material's complete property set.

Square-root inertia parameters calculated from the example material properties. Based on Van Rossum (2027), §5.2.

I_wood = 0.13 × 500 × 1,600 = 104,000 W² s/(m⁴ K²)

b_wood = √104,000 ≈ 322 J/(m² K √s)

I_concrete = 1.7 × 2,300 × 900 = 3,519,000 W² s/(m⁴ K²)

b_concrete = √3,519,000 ≈ 1,880 J/(m² K √s)

The concrete's square-root parameter is about 5.82 times the wood's value. This ratio follows from the unrounded calculated parameters. It is also the ratio of conductive fluxes in the idealised maintained-temperature problem when elapsed time and temperature rise are the same. That comparison concerns the mathematical model with these fixed properties, rather than two measured fire exposures.

The introductory example simultaneously compares heat penetration. After one hour, its characteristic penetration estimates are about 24 mm for the wood and 54 mm for the concrete. The concrete can have a more slowly heating surface while temperature change reaches further into it. The two observations concern different aspects of the response and are consistent with the same material data.

The relevant penetration property is thermal diffusivity, k/(ρc), rather than the product kρc. Its ratio places conductivity above volumetric heat capacity instead of multiplying them. Consequently, a statement about thermal inertia cannot automatically be translated into the same statement about penetration depth. The wood-and-concrete comparison illustrates why both combinations can be useful.

The rounded example is also distinct from a universal material specification. Temperature, moisture, composition and material changes affect the relevant properties. Van Rossum treats the fixed-property calculation as an explanation of thermal behaviour and then discusses its high-temperature limitations. The resulting parameters do not define every timber or concrete assembly exposed to fire.

A characteristic scale for surface response

The SFPE chapter gives a characteristic time scale for a surface responding to hot-gas exposure. It divides the thermal-inertia product by the square of the convective heat-transfer coefficient. That coefficient describes heat transfer between the gas and the material surface. The expression therefore combines an internal material property with an external heat-transfer condition.

t_s = I / h² = b² / h²

Here, t_s is the characteristic surface-response scale in seconds and h is the convective heat-transfer coefficient in W/(m² K). Van Rossum includes a coefficient of 25 W/(m² K) in its example treatment of fire-side convection. Using that coefficient as an explicit illustration gives approximately 166 s for the wooden-wall property set and 5,630 s for the concrete set.

t_s,wood = 104,000 / 25² ≈ 166 s

t_s,concrete = 3,519,000 / 25² ≈ 5.63 × 10³ s

These are characteristic scales, not predictions of ignition time or complete heating. A full temperature calculation still needs the initial state and appropriate boundary conditions. The coefficient used here is part of the illustration, rather than a mandatory value for every fire. Changing it changes the scale even though the material's thermal-inertia parameter remains unchanged.

Calculated surface-response scales using a selected coefficient of25 W/(m² K). Based on Rockett and Milke (2002), p.1-37, and Van Rossum (2027), §5.2.

The squared coefficient also clarifies the effect of surface heat transfer within this model. Doubling h while retaining the same material product divides the characteristic time by four. By contrast, doubling the product while holding h fixed doubles the time scale. These comparisons follow directly from the relationship and distinguish material response from the imposed transfer conditions.

Ignition and thermal response parameters

Time to ignition depends partly on the rate at which a combustible surface is heated. Thermal inertia is relevant because conduction into the interior removes energy from the exposed region while its temperature rises. NUREG-1805 relates this material-property combination to thermal-response parameters used in ignition calculations. The report was prepared for fire-hazard assessment in United States nuclear power plants.

Thermal inertia is not the only required ignition input. The report's treatment also includes the imposed heat flux and a critical heat flux for ignition. Its fitted critical flux is obtained by extrapolating a relationship between measured heat flux and ignition time. The material-response parameter and the flux threshold consequently describe different parts of that correlation.

Iqbal and Salley distinguish a fitted critical flux from the minimum flux for ignition under practical conditions. The fitted value depends on the model used to correlate experimental data. Thermal inertia alone cannot supply that empirical threshold. Nor does identifying a low-inertia material establish that any arbitrarily small thermal exposure will produce ignition.

The distinction between thermally thin and thick materials also affects ignition modelling. A thin specimen can heat through much of its thickness, while a thick specimen initially retains a cooler interior. NUREG-1805 identifies thickness or weight as significant for thin materials. A thermal-response relationship developed for a thermally thick specimen therefore has a limited domain of application.

Rockett and Milke further distinguish initial ignition from sustained burning. Their heated-solid discussion describes how heat conduction into a cooler interior can cool the surface when the external heating is removed. Reaching an ignition temperature does not automatically establish that burning will continue under the changed exposure. The stored temperature distribution remains relevant after the first ignition event.

Enclosure heating and structural calculations

Thermal inertia also affects heat transfer from compartment gases into the enclosing construction. NUREG-1805 distinguishes the early heating period from later conditions in which heat has penetrated through a lining. During the earlier period, the lining stores much of the transferred energy. The report uses the thermal-inertia product in its treatment of that transient boundary heat transfer.

The exposed construction therefore participates in the compartment heat balance. Heat absorbed by a wall or ceiling is energy transferred from the compartment gases into the solid. The response changes as the lining warms and as heat penetrates towards the unexposed side. A single steady-state resistance does not represent every stage of that transient process.

A parametric fire curve supplies a time-dependent exposure for a subsequent thermal calculation. The Australian Building Codes Board's 2020 data sheets distinguish the imposed temperature or heat-flux history from the prediction of temperatures within construction elements. They then relate those element temperatures to structural analysis using appropriate material properties. An exposure curve therefore does not by itself provide the material response.

The thermal-inertia parameter contributes to that response where the relevant conduction model uses it. The complete analysis can also require geometry, high-temperature properties and the changing boundary condition. The ABCB discussion ranges from simple thermal-mass calculations to more detailed spatial models. These are modelling choices within the stated Australian guidance, rather than interchangeable consequences of one inertia value.

Limits of constant-property descriptions

The simplified property product assumes that its constituent values adequately represent the heating period. Van Rossum notes that conductivity and specific heat capacity can change substantially at elevated temperature. Moisture evaporation also consumes energy that is not described by an unchanged sensible-heat capacity alone. Fixed-property parameters therefore have limits when applied across a large temperature range.

Material changes can alter more than temperature-dependent numerical properties. The introductory book identifies melting, charring, material loss and expansion of protective materials as examples. These processes can change the material distribution, the geometry or the composition of the remaining layer. The original product is not automatically a description of that altered construction.

The semi-infinite surface-heating treatment also requires a suitable thickness and duration. The distant boundary is assumed not to control the early response. When heating penetrates through the component, the rear boundary and the finite amount of material become consequential. Rockett and Milke treat thermally thin and thick solutions separately because the corresponding temperature distributions and assumptions differ.

Radiation and convection introduce further modelling conditions. The SFPE chapter notes that simplified analytical expressions can require linearising the heat-transfer terms. Over the large temperature ranges encountered in fires, such approximations may not provide satisfactory quantitative predictions. The chapter discusses numerical conduction methods when varying properties or more complicated energy effects need to be represented.

See also

• Thermal diffusivity

References

• Australian Building Codes Board. (2020). Fire safety verification method data sheets: Handbook annex (Version 1.1).
• Iqbal, N., & Salley, M. H. (2004). Fire dynamics tools (FDTs): Quantitative fire hazard analysis methods for the U.S. Nuclear Regulatory Commission fire protection inspection program (NUREG-1805). U.S. Nuclear Regulatory Commission.
• Rockett, J. A., & Milke, J. A. (2002). Conduction of heat in solids. In P. J. DiNenno (Ed.), SFPE handbook of fire protection engineering (3rd ed., chapter 1-2, pp. 1-27–1-43). National Fire Protection Association.
• Van Rossum, E. (2027). Introduction to Fire Safety Engineering. Fire Safety Academy.