According to its preface, this new book is intended for a one-semester undergraduate course in thermal physics, aimed at third- and fourth-year physics majors. Expecting a stereotypical physicists' book--long on logical rigor and mathematical sophistication, short on clarity and physical intuition--I was pleasantly surprised to find it very accessible and enjoyable reading. Given the demands of covering both classical and statistical thermodynamics in a single semester, I was also somewhat surprised to find that the two subjects are treated essentially independently: the classical theory is presented first (Chapters 1-10), and after a transitional chapter on the kinetic theory of gases (Chapter 11), Chapters 12-20 give a more general introduction to statistical mechanics. Inasmuch as this approach emphasizes the independence and generality of the classical theory and also traces the historical order of development, I think it pedagogically superior to the more "unified" alternative in which the classical results are presented as special cases of the more general statistical theory. It does, however, have the notable disadvantage of taking more time, thereby limiting the number of applications that can be considered if adequate coverage is to be given to the fundamentals.

This book generally does a very good job of laying out the fundamentals, and it also has its share of interesting and novel material. In particular, there is an emphasis on real-world implementation of thermodynamic concepts. Examples include an unusually detailed description of the experimental determination of temperature scales, a calculation of hole sizes for which effusion experiments can be considered quasi static (for a given pressure) and therefore make use of the Maxwell-Boltzmann velocity distribution, and a discussion of the similarities and differences between Carnot's idealized refrigerator and the typical household model. Likewise, in discussing the third law, considerable attention is paid to actual laboratory methods, such as laser cooling and adiabatic demagnetization, for achieving very low temperatures. More esoteric topics, rarely seen in more chemistry-oriented books, include a derivation of the cosmic background spectrum by considering the universe to be a blackbody at T = 2.735 K and an application of Fermi-Dirac statistics to explain how an internal electron-gas pressure prevents the collapse of white dwarf stars. There is a section devoted to a detailed consideration of the conditions for mechanical and thermal stability, and an entire chapter introducing information theory and its connection to thermodynamics. The derivations are compact and easy to follow, and the mathematical level should be accessible to anyone with a good grasp of basic calculus.

Since this book is intended primarily for physicists, it is perhaps understandable that the subject of chemical reactions is allotted slightly less than two of its more than 400 pages and that there is no mention of the concepts of activity and fugacity. A couple of other omissions are not so easy to understand. For example, the only mention of fluctuations is in a brief discussion of the equivalence of ensembles in the thermodynamic (i.e., large-system) limit. There is no coverage of the intimate relationship between fluctuations and physical properties, which is of central importance in so many facets of thermal physics: phase transitions, transport processes, and computer simulations, to name a few. Also, the idea of ensembles in general is given only minimal coverage. All the detailed analysis is confined to isolated systems consisting of "weakly interacting" molecules (or phonons, in the case of crystals), so that the system is effectively a canonical ensemble whose subsystems are the molecules themselves. For beginning students, a collection of molecules is certainly easier to conceptualize than a collection of states of a many-body system, but exclusive use of this approach could leave them with the false impression that the applicability of statistical mechanics is limited to such idealized situations. Along the same lines, greater coverage of classical statistical mechanics is probably in order, but it should also be noted that what coverage there is (the chapter on the kinetic theory of gases) is excellent, especially in its extremely clear and compact derivation of Maxwell's velocity distribution on the basis of intermolecular collisions.

Most aspects of this book should prove interesting and helpful to chemistry students. One notable exception is the use of several nonstandard (from the perspective of the chemist) conventions that could be a source of confusion for beginning students. For example, in strict adherence to the "MKS" system of units, the unit of amount of substance is the kilomole (the amount needed to give a mass in kilograms numerically equal to the molecular weight), so that Avogadro's number is given as 6.02 x 10²⁶ and the gas constant as 8.314 x 10³. Also, work is defined on the basis of the system's internal pressure, which leads to the statement that irreversible processes cannot be represented on a P-V diagram. Of course this is not the case if work is defined as in most chemistry texts, in terms of the external pressure, which can be well defined even when pressure gradients exist within the system. The sign convention for work is also opposite that to which chemists are accustomed: dW = +PdV (i.e., work is defined in terms of the surroundings). This leads to the statement of the first law as dU = dQ - dW, which is much less intuitive than the more familiar alternative. The inverse temperature employed in statistical mechanics also differs by a sign from the usual convention. (b is given as identical to -1/k_BT.) Finally, a strange definition of "degree of freedom" is used (based on the number of terms in the Hamiltonian, rather than the number of independent coordinates needed to specify the configuration) that leads to the statement that a diatomic molecule has two vibrational degrees of freedom ("one kinetic and one potential"). Obviously, this will contradict what most students learn in their other coursework.

All things considered, there is a lot to like about this book. While I would not recommend it as stand-alone text for a chemistry-oriented course in thermodynamics--indeed that is not its intended use--its clear presentation, unusual topics, and especially the author's ability to develop topics in a minimum of space merit a place in the library of anyone who teaches such a course. Also, provided students are cautioned about the nonstandard aspects noted above, its user-friendly style could make it a nice supplementary reference for physical chemistry as well as more advanced courses, and therefore a useful addition to any chemistry library.