statistical mechanics

[anon134]

Consider an ideal gas of oxygen atoms in equilibrium with oxygen atoms absorbed on a planar surface. here are N_s sites per unit surface area at which the atoms can be absorbed, and the energy of an absorbed atom is -e compared to one in the free state. The system is under 1 atm and at 300K.

Should the atoms be described by classical statistics or quantum statistics? I need to show this qualitatively and here's what I have:

\frac{1}{e^{(\epsilon-\mu)/kT} \pm 1}
In the classical limit, [itex]\epsilon-\mu>>kT[/tex]

But we do not know what e is, or mu. Not sure where to go from here, any hints physics forums?

[badphysicist]

Your equation encompasses both Fermi-Dirac and Bose-Einstein statistics because you have a +/- sign in there. Look up both Fermi-Dirac and Bose-Einstein statistics to find out the differences.

[anon134]

Your equation encompasses both Fermi-Dirac and Bose-Einstein statistics because you have a +/- sign in there. Look up both Fermi-Dirac and Bose-Einstein statistics to find out the differences.

Exactly, since quantum statistics becomes Boltzmann maxwell statistics at the classical limit, so I have the +/- for generality. Thus, regardless of a BE or FD distribution, taking the classical limit will give back classical stat.

Now, the question is, do I use classical statmech or quantum statmech for this problem?