Specifically, I'm looking for something more general than the usual "calculating the behavior of a (monatomic|diatomic|...) gas" that many books seem to focus on...some topics I'd be interested in would be:
- Liouville equation / measure (derivation, uses)
- Derivation of various ensembles (gibbs, etc)
- Brownian motion
- Classical path integrals
In addition to those basics, some modern flare would give bonus points:
- Geometric formulations (e.g. using differential forms / bundle connections)
- Lie group / representation theory
- Functional / operator calculus
I spent a lot of time looking at ToCs yesterday, and I found one book that looked like it could be a decent model of what I'm looking for: Kinetic Theory, by Liboff. The Landau / Lifgarbagez texts -- particularly the first part -- also seems to be a decent candidate (though it seems to be very old-school -- missing all of the 'flare' I mentioned above).
However, there are numerous examples of what I'm NOT looking for:
Hill - Introduction to Statistical Thermodynamics
Tolman - Principles of Statistical Mechanics
Mayer & Mayer - Statistical Mechanics
Am I just too picky? I'm envisioning something like what Boltzmann or Maxwell would have been able to write, but with modern language and concepts...not necessarily totally devoid of 20th century physics, but I'm looking for something that doesn't treat statistical mechanics as simply a step on the way to quantum mechanics.