Fundamentals of Statistical and Thermal Physics

Reif reads like thermodynamics is a brand new result and he has written the first giant paper on it. Not only does this book cover every area under the Sun, from the law of mass action to kinetic theory, but he develops each topic in a formal way, with logical consistency and that curious, gray-haired insight. I continually look to Reif when more "advanced" books fumble explanations. With the firm conceptual grounding I get there, I can then intuit what other authors are trying to convey. How many astute readers of other thermodynamics textbooks have any idea when Boltzmann's canonical distribution can be used? Or who know the difference between this and Boltzmann's equation? If you are an instructor, the illuminating end-of-chapter problems will be a boon. If you are a student, they will also be, yet less appreciated likely. No matter who you are, if you want to really know thermo and stat mech and are willing to _think_, then buy this book before some competing, flashy, colorized textbook drives it out of print.

One thing that distinguishes this book from the rest on stat mech is its emphasis on the physical content behind the equations. The author takes great pain to develop and elucidate, at every turn, a coherent physical picture for the edifice of statistical mechanics, in much the same style as the classic book by Tolman. However, if you are learning stat mech/thermo for the first time, you probably will be too busy familiarizing yourself with the equations to be able to appreciate the value of his explanations and motivations, which may just seem pointless rants. So, simply, if you are a newbie, try one of the more mediocre textbooks on stat mech, to learn the subject at a superficial level. Once you've mastered the more superficial stuff, you will come around to love this book, assuming that you are a reflective person.

My first exposure to stat mech was through Callen's book "Thermodynamics and Thermostatistics" in my junior year of college. I hated the book, and decided right then that I never ever wanted to become involved with condensed matter physics. However, I needed to know stat mech for my grad school qualifying exams, and not wanting to go back to Callen, I picked up Reif's book, following a professor's recommendation. What a difference. The clarity of presentation in this book is matched only by Landau's texts. You'll be amazed when doing the problems by just how much interesting physics you can do with just the fundamental relation of thermodynamics and the canonical formalism. Don't be turned off by the length of this book (>600 pages); the reason for this is not overly wordy descriptions (i.e, Griffiths), but simply huge amount of material presented (Griffiths would take over 1000 pages to chat about the same amount of material).<p>Perhaps the best praise I can give this book is that not only did it teach me stat mech, it really sparked my interest in the field of condensed matter physics, enough to contemplate my choosing this as a research topic in grad school.

Never mind that this book was published in the mid '60s (before I was even born); if you must choose one book to learn from, choose this one. It is so concise, so well thought out that I have yet to encounter a more instructive text on the subject. It contains a fabulous overview of statistical and thermal physics, and -even though the book contains quite a lot of material- I regret that the author (quote) "resisted the tempation to include applications of Onsager's reciprocity relations" and other irreversible phenomena. I used to teach physics and chemistry, and I must say that books of this quality just don't seem to be written these days. Given the inherent subtlety of its subject matter and the problems many students have grasping it I therefore strongly recommend using this book to study from.