# From Fermi-Dirac to Maxwell-Boltzmann statistics

I have a little question I can't seem to find the answer to. It is as follows:

When does Fermi-Dirac statistics reduce to Maxwell-Boltzmann statistics?

-

$$f = \frac{1}{1+e^{(E-\mu)/kT}}$$
where, $E$ is the energy, $\mu$ is the chemical potential and $kT$ is Boltzmann's constant times temperature reduces to the Boltzmann distribution
if $E \gg \mu$ then $e^{(E-\mu)/kT} \gg 1$ and so you can write $$f \approx e^{-(E-\mu)/kT}.$$
The statement $E\gg\mu$ does not make much sense. Physically it is the value of $T$ that matters. –  Peter Kravchuk May 23 '13 at 14:38
From Atkins, Physical Chemistry, p. 800: "For energies well above $\mu$, the 1 in the denominator can be neglected...". –  TMOTTM May 23 '13 at 15:00
Well, by $E\gg \mu$, TMOTTM clearly meant $E-\mu\gg kT$. –  Luboš Motl May 23 '13 at 15:36