next up previous
Next: Fermi-Dirac Statistics Up: Quantum Statistics Previous: Illustrative Example

Formulation of Statistical Problem

Consider a gas consisting of $ N$ identical non-interacting particles occupying volume $ V$ , and in thermal equilibrium at absolute temperature $ T$ . Let us label the possible quantum states of a single particle by $ r$ (or $ s$ ). Let the energy of a particle in state $ r$ be denoted $ \epsilon_r$ . Let the number of particles in state $ r$ be written $ n_r$ . Finally, let us label the possible quantum states of the whole gas by $ R$ .

The particles are assumed to be non-interacting, so the total energy of the gas in state $ R$ , where there are $ n_r$ particles in quantum state $ r$ , et cetera, is simply

$\displaystyle E_R = \sum_r n_r \epsilon_r,$ (8.16)

where the sum extends over all possible quantum states, $ r$ . Furthermore, because the total number of particles in the gas is known to be $ N$ , we must have

$\displaystyle N = \sum_r n_r.$ (8.17)

In order to calculate the thermodynamic properties of the gas (i.e., its internal energy or its entropy), it is necessary to calculate its partition function,

$\displaystyle Z =\sum_R {\rm e}^{-\beta E_R} = \sum_R {\rm e}^{-\beta (n_1 \epsilon_1+n_2  \epsilon_2+\cdots)}.$ (8.18)

Here, the sum is over all possible states, $ R$ , of the whole gas. That is, over all the various possible values of the numbers $ n_1, n_2,\cdots$ .

Now, $ \exp[-\beta (n_1 \epsilon_1+n_2 \epsilon_2+\cdots)]$ is the relative probability of finding the gas in a particular state in which there are $ n_1$ particles in state 1, $ n_2$ particles in state 2, et cetera. Thus, the mean number of particles in quantum state $ s$ can be written

$\displaystyle \bar{n}_s = \frac{\sum_R n_s \exp[-\beta (n_1 \epsilon_1+n_2 ...
...on_2+\cdots)]} {\sum_R \exp[-\beta (n_1 \epsilon_1+n_2 \epsilon_2+\cdots)]}.$ (8.19)

A comparison of Equations (8.18) and (8.19) yields the result

$\displaystyle \bar{n}_s = -\frac{1}{\beta}\frac{\partial \ln Z}{\partial\epsilon_s}.$ (8.20)

Here, $ \beta\equiv 1/(k T)$ .


next up previous
Next: Fermi-Dirac Statistics Up: Quantum Statistics Previous: Illustrative Example
Richard Fitzpatrick 2016-01-25