advanced27 min studyLesson 357 in Physics

Transport momentum and energy between neighboring layers

Compute dilute-gas transport coefficients with their signs, units, and model limitations.

On this page

What you’ll understand

  • Relate molecular transport to viscosity and heat conductivity.
  • Check dynamic versus kinematic coefficient dimensions.
  • Explain the BGK Prandtl-number restriction.

Particles crossing between neighboring fluid layers carry the momentum and thermal energy characteristic of the layer they last visited. That transport produces viscous stress and heat conduction even in a gas with no lasting molecular bonds. The coefficient estimates depend on collision physics and the chosen closure, but their dimensions and smoothing signs are robust diagnostics. We will use the explicit BGK predictions while keeping them separate from exact transport coefficients of a specified intermolecular potential.

Momentum transport opposes shear differences

Suppose mean flow in the x direction increases along y. Molecules arriving from larger y tend to carry excess x momentum into a slower layer, while molecules arriving from smaller y carry a deficit. Random motion therefore transports x momentum down its gradient. In the kinetic pressure tensor convention, the off-diagonal momentum flux is minus viscosity times the velocity gradient. The viscous part of the Cauchy stress uses the opposite sign because stress is traction, while the pressure tensor appears as an outward momentum flux.

A dimensional estimate gives dynamic viscosity of order mass density times thermal speed times mean free path. At fixed temperature in a dilute gas, increasing number density increases the carriers but shortens their mean free path approximately inversely. These effects roughly cancel, explaining weak density dependence in the dilute regime. This cancellation fails near dense-fluid correlations, where the simple binary transport picture and collision frequency no longer suffice.

Pxy=−η ∂yux,η∼ρvthℓ,ν=ηρP_{xy}=-\eta\,\partial_yu_x,\qquad \eta\sim\rho v_{\rm th}\ell,\qquad \nu=\frac{\eta}{\rho}
η is dynamic viscosity in Pa·s=kg/(m·s), ρ is mass density in kg/m³, and ν is kinematic viscosity in m²/s. The estimate assumes dilute near-equilibrium transport and simple shear.↔ Scroll wide equations horizontally.

Thermal transport has an additional heat capacity

Molecules crossing a temperature gradient carry excess or deficient random energy, yielding heat flux down the gradient. Thermal conductivity therefore involves not only the flight scale but also energy per temperature carried by each molecule. Fourier’s law relates a heat flux in watts per square metre to a temperature gradient in kelvin per metre; its coefficient has units watts per metre kelvin. It must not be inserted directly as the diffusivity of a temperature equation.

To obtain a thermal diffusivity, divide conductivity by the relevant heat capacity per volume. For an ideal gas at approximately constant pressure, that capacity is mass density times specific heat at constant pressure. In a constrained fixed-density setting, a different caloric factor may appear. The dynamical and thermodynamic constraints matter just as much as the material coefficient. Calling conductivity and diffusivity by the same symbol can conceal this physically important distinction.

q=−κ∇T,α=κρcp,cp=5kB2m\mathbf q=-\kappa\nabla T,\qquad \alpha=\frac{\kappa}{\rho c_p},\qquad c_p=\frac{5k_B}{2m}
q is heat flux in W/m², κ conductivity in W/(m·K), α constant-pressure thermal diffusivity in m²/s, and cp specific heat in J/(kg·K) for a classical monatomic ideal gas.↔ Scroll wide equations horizontally.

One relaxation time ties two coefficients together

First-order BGK gives viscosity equal to pressure times relaxation time. Its conductivity is five halves the gas constant per mass times that viscosity. Their dimensionless Prandtl ratio is therefore one. A more faithful elastic collision calculation can produce the near-two-thirds value characteristic of dilute monatomic gases. This is a model limitation with a concrete consequence: fitting the momentum diffusion time does not independently fit the thermal diffusion time.

The heat lab represents a diffusivity after such conversion. Reset and set mode to 2 and diffusivity to 0.25, then Step twice; predict amplitude exp(−1). Reset and set mode to 2 and diffusivity to 0.5, then Step twice; predict exp(−2). Read the decay rate as diffusivity times squared mode number. The lab has no mass density, viscosity, or heat-capacity control, so it cannot derive gas coefficients or compare Prandtl numbers.

ηBGK=pτ,κBGK=52kBmpτ,Pr=cpηκ=1\eta_{\rm BGK}=p\tau,\qquad \kappa_{\rm BGK}=\frac52\frac{k_B}{m}p\tau,\qquad \mathrm{Pr}=\frac{c_p\eta}{\kappa}=1
p is pressure in Pa and τ in s. These are first-order transport predictions of the single-relaxation-time monatomic BGK model, not universal exact coefficients of the Boltzmann equation.↔ Scroll wide equations horizontally.

Convert a relaxation time into transport

For an illustrative BGK gas take p=8×10⁴ Pa, τ=2.5×10⁻¹⁰ s, kB/m=200 J/(kg·K), and mass density 1 kg/m³. Find η, κ, ν, α, and Pr.

  1. The viscosity is pτ=8×10⁴×2.5×10⁻¹⁰=2×10⁻⁵ Pa·s. Pressure multiplied by time has the required dynamic-viscosity units, rather than diffusivity units.

  2. The specific heat is cp=(5/2)×200=500 J/(kg·K). Conductivity is cpη=500×2×10⁻⁵=0.010 W/(m·K). This equality is the BGK transport relation for the supplied monatomic model.

  3. Divide viscosity by density to get ν=2×10⁻⁵ m²/s. Divide conductivity by ρcp to get α=0.010/500=2×10⁻⁵ m²/s. Their ratio, or cpη/κ, is one, showing that this model ties thermal and momentum diffusion together.

The result

The coefficients are η=2×10⁻⁵ Pa·s and κ=0.010 W/(m·K), with ν=α=2×10⁻⁵ m²/s and Pr=1.

Explore the model

Interactive lab

Interactive lab · heat

Use diffusivity rather than conductivity

Reset and set n=2, κ=0.25, then Step twice; predict amplitude exp(−1). Reset and set n=2, κ=0.5, then Step twice; predict exp(−2).

Use diffusivity rather than conductivityThe exact periodic heat-equation mode is u(x,t) = exp(−κn²t) sin(nx), on [0,2π]. Playback advances physical time. The initial sine is shown separately, and the unnormalized L² norm is √π exp(−κn²t). This analytic instrument does not run a finite-difference solver or test discretization stability. Current readings: Current amplitude: 1; Spatial L² norm: 1.772; Decay rate · κn²: 3.6 /s. All parameters are adjustable with the labeled controls below.Periodic heat equation mode 3 at time 0: the amplitude is 1.0π/2π3π/22π−1−0.500.51x · periodic intervalTemperature perturbation uExact mode · n = 3
Current temperature modeInitial sineCurrent amplitude
Current amplitude
1
Spatial L² norm
1.772
Decay rate · κn²
3.6 /s
t = 0 s

The exact periodic heat-equation mode is u(x,t) = exp(−κn²t) sin(nx), on [0,2π]. Playback advances physical time. The initial sine is shown separately, and the unnormalized L² norm is √π exp(−κn²t). This analytic instrument does not run a finite-difference solver or test discretization stability. Parameter changes pause playback and recompute the model at the current elapsed time. Reset restores both parameters and returns time to zero. Playback starts paused. Adjust either parameter, then choose Play or Step.

Look forThe control κ is the heat equation’s diffusivity. It is not gas thermal conductivity unless the required heat-capacity conversion has already been performed.

Check your understanding

Which units belong to thermal conductivity rather than thermal diffusivity?

Try it yourself

A near-equilibrium gas has η=3×10⁻⁵ Pa·s, κ=0.020 W/(m·K), shear gradient ∂yux=20 s⁻¹, and temperature gradient ∂yT=−50 K/m. Find Pxy and qy, including signs.

Take a moment to make a prediction or work it out on paper. Open a hint when you need one.

Give me a hint

Use the kinetic momentum-flux convention and Fourier’s minus sign; state the transport directions.

Compare with the solution

The off-diagonal kinetic momentum flux is Pxy=−η∂yux=−6×10⁻⁴ Pa. Heat flux is qy=−κ∂yT=+1.0 W/m², toward increasing y where temperature is lower. The associated viscous Cauchy shear stress is +6×10⁻⁴ Pa, opposite in sign to this pressure-tensor contribution.

Take it with you

  • Dynamic viscosity and kinematic viscosity have different dimensions.
  • Conductivity becomes a diffusivity only after a heat-capacity conversion.
  • Basic BGK predicts Pr=1 and cannot fit both transport channels independently.

Read further

Original sources and deeper explanations to follow at your own pace.

Optional progress, saved in this browser when available. Revisit any lesson at any time.

Lesson not marked complete.