Heat Transfer by Conduction

The Physics of Heat Conduction

At the atomic level, thermal conduction arises from two mechanisms. In non-metallic solids, heat is transported primarily by lattice vibrations—quantised packets of vibrational energy called phonons. When one region of a crystal lattice is hotter, its atoms vibrate more vigorously, and these vibrations propagate to neighbouring atoms, carrying energy along. In metals, a second—and usually dominant—mechanism comes into play: free electrons. The sea of delocalised conduction electrons that gives metals their electrical conductivity also serves as an extremely efficient carrier of thermal energy. This is why metals are both good electrical and good thermal conductors, a connection formalised by the Wiedemann-Franz law.

The thermal conductivity $k$ of a material quantifies how easily it conducts heat. Copper, at $k \approx 400 \; \mathrm{W/(m \cdot K)}$, is among the best conductors. Diamond, thanks to its extremely rigid lattice, reaches $k \approx 2000 \; \mathrm{W/(m \cdot K)}$. At the other extreme, aerogel—a silica-based material consisting mostly of trapped air pockets—achieves $k \approx 0.015 \; \mathrm{W/(m \cdot K)}$, making it one of the best thermal insulators known.

Fourier's Law of Heat Conduction

The cornerstone of conductive heat transfer is Fourier's law, proposed by Jean-Baptiste Joseph Fourier in 1822. In one dimension, it states that the heat flux—the rate of heat transfer per unit area—is proportional to the negative temperature gradient:

$${\displaystyle q = -k \, \frac{dT}{dx}}$$

Where:

• $q$ is the heat flux $\left[\mathrm{W/m^2}\right]$.
• $k$ is the thermal conductivity of the material $\left[\mathrm{W/(m \cdot K)}\right]$.
• $dT/dx$ is the temperature gradient in the direction of heat flow $\left[\mathrm{K/m}\right]$.

The negative sign reflects the second law of thermodynamics: heat flows spontaneously from hot to cold. The total heat transfer rate through a surface of area $A$ is simply:

$${\displaystyle \dot{Q} = q \, A = -k \, A \, \frac{dT}{dx}}$$

In three dimensions, Fourier's law generalises to a vector equation involving the temperature gradient $\nabla T$:

$${\displaystyle \vec{q} = -k \, \nabla T}$$

This compact expression is the starting point for virtually all analytical and numerical treatments of heat conduction.

The Heat Equation

Combining Fourier's law with conservation of energy yields the heat equation (also called the diffusion equation), the fundamental partial differential equation governing transient heat conduction in a solid:

$${\displaystyle \rho \, c_p \, \frac{\partial T}{\partial t} = \nabla \cdot \left( k \, \nabla T \right) + \dot{q}_{\text{gen}}}$$

Where:

• $\rho$ is the material density $\left[\mathrm{kg/m^3}\right]$.
• $c_p$ is the specific heat capacity $\left[\mathrm{J/(kg \cdot K)}\right]$.
• $\dot{q}_{\text{gen}}$ is the volumetric internal heat generation rate $\left[\mathrm{W/m^3}\right]$, for example from electrical resistance heating or chemical reactions.

For a homogeneous material with constant thermal conductivity, this simplifies to:

$${\displaystyle \frac{\partial T}{\partial t} = \alpha \, \nabla^2 T + \frac{\dot{q}_{\text{gen}}}{\rho \, c_p}}$$

where $\alpha = k / (\rho \, c_p)$ is the thermal diffusivity $\left[\mathrm{m^2/s}\right]$—a single property that tells you how fast temperature changes propagate through a material. Copper, with $\alpha \approx 1.17 \times 10^{-4} \; \mathrm{m^2/s}$, responds thermally far faster than brick at $\alpha \approx 5.2 \times 10^{-7} \; \mathrm{m^2/s}$.

Steady-State Conduction: Thermal Resistance

When conditions do not change with time ($\partial T / \partial t = 0$) and there is no internal heat generation, the heat equation reduces to Laplace's equation: $\nabla^2 T = 0$. For the simple but extremely useful case of one-dimensional, steady-state conduction through a plane wall of thickness $L$, the temperature profile is linear and the heat transfer rate is:

$${\displaystyle \dot{Q} = \frac{k \, A}{L} \left(T_1 - T_2\right) = \frac{T_1 - T_2}{R_{\text{cond}}}}$$

where the thermal resistance for conduction through the wall is defined as:

$${\displaystyle R_{\text{cond}} = \frac{L}{k \, A}}$$

This analogy with Ohm's law ($V = IR$) is extraordinarily powerful. Just as electrical resistances can be combined in series and parallel, thermal resistances can be chained together to model composite walls, insulation layers, contact resistances, and convective boundary conditions in a single unified framework.

Composite Walls

Consider a wall composed of three layers (e.g., plaster, brick, insulation) with convective boundary conditions on both sides. The total thermal resistance is:

$${\displaystyle R_{\text{total}} = \frac{1}{h_1 A} + \frac{L_1}{k_1 A} + \frac{L_2}{k_2 A} + \frac{L_3}{k_3 A} + \frac{1}{h_2 A}}$$

and the overall heat transfer rate follows directly as $\dot{Q} = \Delta T_{\text{overall}} / R_{\text{total}}$. This approach is the backbone of building energy calculations and is used daily by architects and mechanical engineers to evaluate insulation performance and U-values.

Radial Conduction: Cylinders and Spheres

Many engineering components—pipes, cables, reactor vessels—have cylindrical or spherical geometry. For steady-state conduction through a hollow cylinder (inner radius $r_1$, outer radius $r_2$, length $L_{\text{cyl}}$), the temperature profile is logarithmic and the heat transfer rate is:

$${\displaystyle \dot{Q} = \frac{2\pi \, k \, L_{\text{cyl}} \left(T_1 - T_2\right)}{\ln\left(r_2 / r_1\right)}}$$

The corresponding thermal resistance for radial conduction through a cylindrical shell is:

$${\displaystyle R_{\text{cyl}} = \frac{\ln\left(r_2 / r_1\right)}{2\pi \, k \, L_{\text{cyl}}}}$$

An interesting consequence arises when insulation is added to a thin pipe or wire: there exists a critical radius of insulation $r_{\text{cr}} = k_{\text{ins}} / h$ below which adding insulation actually increases heat loss, because the larger outer surface area enhances convection more than the insulation reduces conduction. This counter-intuitive result has practical implications for the insulation of electrical cables and small-diameter tubing.

Transient Conduction and the Biot Number

When temperatures change with time—a part being quenched, an oven preheating, a brake disc absorbing frictional heat—we enter the domain of transient conduction. The first question is whether the body's interior temperature can be treated as spatially uniform or whether significant internal gradients develop. The Biot number provides the answer:

$${\displaystyle \mathrm{Bi} = \frac{h \, L_c}{k_s}}$$

where $h$ is the surface convective coefficient, $L_c$ is a characteristic length (typically volume/surface area), and $k_s$ is the solid's thermal conductivity. When $\mathrm{Bi} < 0.1$, internal temperature gradients are negligible and the lumped capacitance method applies—the entire body is treated as being at a single, time-dependent temperature described by Newton's law of cooling:

$${\displaystyle T(t) = T_{\infty} + \left(T_0 - T_{\infty}\right) e^{-t/\tau}}$$

with time constant $\tau = \rho \, V \, c_p / (h \, A_s)$. When $\mathrm{Bi} > 0.1$, spatial temperature variations within the body become important and the full heat equation must be solved—analytically for simple geometries (via separation of variables and Heisler charts) or numerically using the finite element method for complex shapes.

Engineering Applications

1. Building Insulation and Energy Efficiency

Conduction through walls, roofs, and floors is the primary pathway for heat loss in buildings. Engineers use the thermal resistance network approach to evaluate the overall U-value of construction assemblies and to determine where additional insulation delivers the best return. Materials like expanded polystyrene ($k \approx 0.035 \; \mathrm{W/(m \cdot K)}$), mineral wool, and vacuum insulation panels are selected specifically for their low thermal conductivity.

2. Electronics Thermal Management

In microelectronics, heat generated by transistors must be conducted through the silicon die, through thermal interface materials (TIMs), and into the heat spreader and heat sink before it can be dissipated by convection. Each layer presents a thermal resistance, and the total junction-to-ambient resistance determines the chip's operating temperature. Reducing any link in this chain—for example by using high-conductivity copper or diamond heat spreaders—directly lowers the junction temperature and improves reliability.

3. Metallurgical Heat Treatment

Processes like annealing, quenching, and tempering depend critically on controlling the rate and uniformity of temperature change within a metal part. Internal conduction determines how quickly the core of a thick steel forging reaches the target temperature compared with its surface. Getting this wrong results in residual stresses, distortion, or undesirable microstructures. Predictive thermal analysis using FEA is widely employed to design heat treatment cycles that achieve the desired properties while minimising defects.

4. Thermal Protection Systems

Spacecraft re-entering the atmosphere encounter extreme heating—surface temperatures can exceed 1500 °C. Ablative heat shields and ceramic tile systems exploit materials with low thermal conductivity to keep the vehicle's structure cool, buying time for the heat pulse to pass before it penetrates to critical components. The design of these systems involves solving transient conduction problems with temperature-dependent material properties and surface ablation.

Case Study: Heat Loss Through an Insulated Pipe

A steel pipe ($k_{\text{steel}} = 50 \; \mathrm{W/(m \cdot K)}$) carries steam at 200 °C. The pipe has inner radius $r_1 = 25 \; \mathrm{mm}$, outer radius $r_2 = 30 \; \mathrm{mm}$, and is covered with 40 mm of mineral wool insulation ($k_{\text{ins}} = 0.04 \; \mathrm{W/(m \cdot K)}$, outer radius $r_3 = 70 \; \mathrm{mm}$). The ambient temperature is 20 °C with an outer convective coefficient $h_o = 10 \; \mathrm{W/(m^2 \cdot K)}$. Per metre of pipe length ($L = 1 \; \mathrm{m}$):

$${\displaystyle R_{\text{steel}} = \frac{\ln(r_2/r_1)}{2\pi \, k_{\text{steel}} \, L} = \frac{\ln(30/25)}{2\pi \times 50 \times 1} = \frac{0.1823}{314.16} \approx 5.8 \times 10^{-4} \; \mathrm{K/W}}$$

$${\displaystyle R_{\text{ins}} = \frac{\ln(r_3/r_2)}{2\pi \, k_{\text{ins}} \, L} = \frac{\ln(70/30)}{2\pi \times 0.04 \times 1} = \frac{0.8473}{0.2513} \approx 3.371 \; \mathrm{K/W}}$$

$${\displaystyle R_{\text{conv}} = \frac{1}{h_o \, 2\pi \, r_3 \, L} = \frac{1}{10 \times 2\pi \times 0.07 \times 1} = \frac{1}{4.398} \approx 0.227 \; \mathrm{K/W}}$$

The total resistance is $R_{\text{total}} = 5.8 \times 10^{-4} + 3.371 + 0.227 \approx 3.60 \; \mathrm{K/W}$, and the heat loss per metre is:

$${\displaystyle \dot{Q} = \frac{200 - 20}{3.60} \approx 50 \; \mathrm{W/m}}$$

Notice how the steel wall contributes almost nothing to the total resistance ($0.016\%$), while the insulation layer provides $93.6\%$ of the thermal resistance. Without insulation, the heat loss would jump to roughly 800 W/m—a sixteen-fold increase. This type of thermal resistance analysis is routine in industrial piping design, process engineering, and energy auditing.

Numerical Methods for Complex Conduction Problems

Analytical solutions exist only for relatively simple geometries and boundary conditions. In practice, engineering components have irregular shapes, temperature-dependent material properties, internal heat sources, contact resistances at interfaces, and complex multi-material assemblies. For these problems, the finite element method (FEM) is the standard numerical tool. FEM discretises the geometry into a mesh of elements, approximates the temperature field with piecewise polynomial functions, and solves the resulting system of algebraic equations to produce a detailed temperature map throughout the component.

Modern FEA software can handle steady-state and transient conduction, coupled with convective and radiative boundary conditions, nonlinear material properties, phase changes, and even thermo-mechanical coupling where thermal stresses and deformations feed back into the thermal problem. When the fluid flow itself is part of the problem—forced or natural convection, conjugate heat transfer through a heat exchanger—Computational Fluid Dynamics (CFD) takes over, solving the flow and energy equations simultaneously. Together, FEA and CFD form the backbone of modern heat transfer analysis, enabling engineers to predict temperatures, heat flows and thermal stresses in components of any complexity.

Conclusion

Heat conduction is the most intuitive and mathematically tractable of the three heat transfer modes, yet its engineering importance is immense. Fourier's law and the thermal resistance concept provide elegant analytical tools for quick estimates and design guidance, while the heat equation—solved numerically by the finite element method—handles the full complexity of real-world problems. Whether the goal is to minimise heat loss through a building envelope, manage temperatures in a microchip, control the cooling rate during a quenching process, or design a thermal protection system for re-entry, conduction is invariably part of the story. For engineering challenges where accuracy matters and the stakes are high, expert thermal analysis services provide the rigour and insight needed to turn thermal understanding into reliable, optimised designs.