Is it okay if I don't like statistical mechanics?

[TomServo]

I'm taking stat mech right now (Kittel book) and...I might hate it. I love e&m, quantum, classical, relativity, but I like almost none of what we're doing in this class. It's not the type of physics I'm interested in and it's abstract in a way I don't like (I normally like abstraction). I know I'll face it again in grad school but everybody says stat mech is the most widespread, fundamental kind of physics, that it shows up everywhere, etc.

Will I learn to love it or is it okay if I hate it? I want to do research in gravity/general relativity. If I don't like it, is that a sign I shouldn't be a physicist or what?

 

[Simon Bridge]

It's OK to hate it - it may motivate you to find another approach.
It does get better as you delve into it - the link to quantum statistics and hamiltonian mechanics will probably intrigue you for example. But I don't think anyone actually "likes" everything in the physics they do - don't sweat it.

If it gets too much, there is always the Feynman suggestion...

 

[WannabeNewton]

Well there's no rule that says you have to like everything in physics in order to be a physicist but statistical mechanics is arguably the most pervasive subject, as you have already been told. Have you considered the possibility that just maybe you got off on the wrong foot with the subject perhaps due to a bad lecturer or bad textbook? There are a plethora of statistical mechanics texts out there so see if there's any that gets you back on the right foot. I personally like Kardar's text the best: https://www.amazon.com/dp/0521873428/?tag=pfamazon01-20

I love general relativity but in my opinion statistical mechanics is way more interesting because it is so rich in theory and needless to say rich in applications, both classically and quantum mechanically.

That being said, I personally know a few physics majors who hate statistical mechanics however I can tell you that they do so because of their lecturer.

 

[TomServo]

Which Feynman suggestion was that?

And WannabeNewton, why do you like that book?

 

[jesse73]

If you expand into statistical physics you have a huge amount of material. Statistical Physics is absurdly interesting. You have aggregation process, applications in biophysics and biosystems, entropy in information theory, complex networks etc.

A book on kinetics.

https://www.amazon.com/dp/0521851033/?tag=pfamazon01-20

networks are also interesting
https://www.amazon.com/dp/0199206651/?tag=pfamazon01-20

and phase transitions and critical phenomena
https://www.amazon.com/dp/0199577226/?tag=pfamazon01-20

I think all of those should be doable by any older undergrad.

 

[Simon Bridge]

Which Feynman suggestion was that?

"Move to a different Universe." ;)

... sorry, I thought it was well known.

That is, if your dislike of statistical mechanics comes from a dislike of the field itself, i.e. the statistical nature of the physics involved. If it is a dislike of the course, which is often taught in a very "mathy" way, that's a different story. I remember totally loathing the course before post-grad.

I still have the text someplace ... totally impenetrable to this day.

The breakthrough for me was in the S.M. Tan lectures:
http://home.comcast.net/~szemengtan/

 

[atyy]

In addition to things like phase transitions, Bose-Einstein condensation etc, an interesting aspect of stat mech is that it works, yet isn't fundamental.

Since it isn't fundamental, how can it be derived?

The overlap of QM, gravity, and thermo/stat mech is very interesting. Quantum black holes have a temperature. Do they obey stat mech? If so, what are the quantum states to be counted, especially since classically black holes have "no hair"?

Incidentally, the theory of critical phenomena in classical stat mech is how the meaning of quantum field theory was understood, and people (who were not rigourous mathematicians) stopped worrying that renormalization was this bizarre process of subtracting infinities, because the stat mech picture provided a good physical picture of renormalization.

Also, don't forget that the dawn of the quantum age started with Planck's derivation of the black body formula using stat mech!

https://www.amazon.com/dp/0521873428/?tag=pfamazon01-20

And WannabeNewton, why do you like that book?

The lecture notes from which the book was made are available free at http://ocw.mit.edu/courses/physics/...chanics-of-particles-fall-2007/lecture-notes/. The notes for the second part of the course which covers critical phenomena are at http://ocw.mit.edu/courses/physics/...-physics-of-fields-spring-2008/lecture-notes/.

I too like Kardar's books very much. Some books like Reif place stat mech above classical thermodynamics, but Kardar presents succintly all of classical equilibrium thermodynamics first. A wonderful thing is his presentation of Clausius's derivation of entropy as a state function, and the second law of thermodynamics from simple English statements (the Kelvin and Clausius statements). I think Clausius's discovery is as beautiful as Einstein's derivation of the Lorentz transformations from the Principle of Relativity and a speed of light that is the same in all inertial frames. Kardar doesn't hesitate to briefly discuss "philosophical" questions like the origin of irreversibility, one ingredient of which is a large number of particles. Kardar is also careful to note interesting details like the differences between the microcanonical and canonical ensembles (p87 of the first set of notes).

Another very good book, but less philosophical is Peliti's "Statistical Mechanics in a Nutshell". For introductory thermo, I like Adkins's "Equilibrium thermodynamics".