Help with an approximation

[demonslayer79]

I'm doing a problem on Van-der Walls interaction and was told in the hint of the problem to use the approximation kT>>hw to simplify

{-hw/(2kT)}-Ln[Exp[-hw/(kT)]-1]

I have no idea how to apply this approximation to simpify the problem.
Thanks

[Hurkyl]

Do you remember differential approximation? (Or taylor series?) If kT >> hw, then (hw / kT)^2 is going to be negligibly small...