# Diffusion

By 0x7df, Sun 05 March 2017, in category Physics

## Diffusion Equation

The time-dependent diffusion equation is:

$$\frac{\partial u}{\partial t} = \nabla \cdot \left( D\left(\mathbf{r}, t\right) \nabla u\left(\mathbf{r}, t\right) \right) + Q\left(\mathbf{r}, t\right)$$

where:

• $$\mathbf{r}$$ is position
• $$t$$ is time
• $$u \left( \mathbf{r}, t \right)$$ is the unknown, i.e. diffusing quantity
• $$D \left( \mathbf{r}, t \right)$$ is the diffusion coefficient
• $$Q \left( \mathbf{r}, t \right)$$ is the source density

From hereon we will display the quantities $$D$$, $$Q$$ and $$u$$ without their arguments, for simplicity of notation.

To derive the diffusion equation we begin with Fick's law (Fick, 1855):

$$\mathbf{F} = -D \nabla u$$

The vector field $$\mathbf{F}$$ is the flux, which is the rate of transfer per unit area; the integral of the normal component of $$\mathbf{F}$$ over a given surface is equal to the rate of flow through the surface. The direction of the flux vector is normal to the surface of constant concentration. Fick's law says that for a given concentration $$u$$ at a point $$\mathbf{r}$$, the flux is proportional to the concentration gradient there, and has the opposite direction.

$$Q$$ is the source density: the concentration produced per unit time per unit volume.

The divergence of $$\mathbf{F}$$, i.e. $$\nabla \cdot \mathbf{F}$$, is the rate of loss of concentration per unit time from the volume element.

Hence, by conservation:

$$\frac{\partial u}{\partial t} = - \nabla \cdot \mathbf{F} + Q$$

Hence:

$$\frac{\partial u}{\partial t} = \nabla \cdot D \nabla u + Q$$

As we saw above, this is the diffusion equation (sometimes known as Fick's second law).

The equation as written is linear, although it would be non-linear if we allowed the diffusion coefficient $$D$$ to vary with the unknown $$u$$, as well as with $$x$$ and $$t$$.

## Special Cases

In the special case that $$D$$ is constant in space and time, then this equation simplifies to:

$$\frac{\partial u}{\partial t} = D \nabla^2 u + Q$$

and in steady-state, if $$Q$$ is also constant, $$\partial u/\partial t = 0$$, it reduces to Poisson's equation:

$$\nabla^2 u = \frac{Q}{D}$$

and further to Laplace's equation if the source term is zero:

$$\nabla^2 u = 0$$

Returning now to the full equation, we recall that:

$$\nabla = \hat i \frac{\partial}{\partial x} + \hat j \frac{\partial}{\partial y} + \hat k \frac{\partial}{\partial z}$$
$$= \hat e_{\rho} \frac{\partial}{\partial \rho} + \hat e_{\phi} \frac{\partial}{\partial \phi} + \hat e_z \frac{\partial}{\partial z}$$
$$= \hat e_{r} \frac{\partial}{\partial r} + \hat e_{\theta} \frac{\partial}{\partial \theta} + \hat e_{\phi} \frac{\partial}{\partial \phi}$$

in Cartesian, cylindrical and spherical geometries, respectively.

Consider first spherical geometry:

$$\nabla \cdot D \nabla u =$$
$$\frac{1}{r^2} \frac{\partial}{\partial r} \left( r^2 D \frac{\partial u}{\partial r} \right) + \frac{1}{r \sin \theta} \frac{\partial}{\partial \theta} \left( \frac{D \sin \theta }{r} \frac{\partial u}{\partial \theta} \right) + \frac{1}{r \sin \theta} \frac{\partial}{\partial \phi} \left( \frac{D}{r \sin \theta} \frac{\partial u}{\partial \phi} \right)$$

In cylindrical geometry:

$$\nabla \cdot D \nabla u = \frac{1}{\rho} \frac{\partial}{\partial \rho} \left( \rho D \frac{\partial u}{\partial \rho} \right) + \frac{1}{\rho} \frac{\partial}{\partial \phi} \left( \frac{D}{\rho} \frac{\partial u}{\partial \phi} \right) + \frac{\partial}{\partial z} \left( D \frac{\partial u}{\partial z} \right)$$

and in Cartesian geometry:

$$\nabla \cdot D \nabla u = \frac{\partial}{\partial x}\left( D \frac{\partial u}{\partial x} \right) + \frac{\partial}{\partial y}\left( D \frac{\partial u}{\partial y} \right) + \frac{\partial}{\partial z}\left( D \frac{\partial u}{\partial z} \right)$$

In one dimension:

$$\frac{\partial u}{\partial t} = \frac{1}{r^2} \frac{\partial}{\partial r} \left( r^2 D \frac{\partial u}{\partial r} \right) + Q$$
$$\frac{\partial u}{\partial t} = \frac{1}{\rho} \frac{\partial}{\partial \rho} \left( \rho D \frac{\partial u}{\partial \rho} \right) + Q$$
$$\frac{\partial u}{\partial t} = \frac{\partial}{\partial x} \left( D \frac{\partial u}{\partial x} \right) + Q$$

These equations can be generalised:

$$\frac{\partial u}{\partial t} = \frac{1}{x^p} \frac{\partial}{\partial x} \left(x^p D \frac{\partial u}{\partial x} \right) + Q$$

where:

• $$p = 0$$ for plane geometry
• $$p = 1$$ for 1D cylindrical geometry, and
• $$p = 2$$ for 1D spherical geometry.

## Heat Flow Equation

The phenomena of heat conduction and diffusion are basically the same, and Fick, 1855 first put diffusion on a quantitative basis by adopting the mathematical equation of heat conduction derived by Fourier (Fourier, 1822).

The heat flow equation of Fourier is:

$$a \frac{\partial \theta}{\partial t} = \frac{\partial}{\partial x} K \frac{\partial \theta}{\partial x}$$

where $$a$$ is the heat capacity of the material per unit volume and $$K$$ is the thermal conductivity. For constant conductivity this becomes:

$$\frac{\partial \theta}{\partial t} = \frac{K}{a} \frac{\partial^2 \theta}{\partial x^2}$$

where the corresponding equation for diffusion is:

$$\frac{\partial u}{\partial t} = D \frac{\partial^2 u}{\partial x^2}$$

For the two equations to correspond, we equate temperature $$\theta$$ with concentration $$u$$, which clearly implies that $$D = K/a$$. However, it is also the case that:

$$\mathbf{F} = -K \nabla \theta$$

which, when compared with Fick's law for diffusion, implies that $$D = K$$, and therefore that in diffusion, unlike in heat conduction, $$a = 1$$. This is because we have identified $$C$$ with $$\theta$$, whereas in heat conduction, the diffusing 'substance' is actually heat, not temperature. The factor $$a$$ is needed to convert temperature to the amount of heat per unit volume, whereas in diffusion, the concentration is already by definition the amount of substance per unit volume, so $$a = 1$$.