Copyright © 2010 jsd
DRAFT Review of Schroeder, An Introduction to Thermal Physics
This book has some good points and some not-so-good points. We all know that writing a book is a tremendous amount of work ... but in this case it seems that with a relatively minor amount of additional work this could have been a much, much better book.
Here is a list of crucial concepts, with a discussion of how the book handles the each concept.
- On page 111 it defines «thermodynamic identity» which is virtually synonymous with the «first law» equation on page 18. The importance of this identity is grossly overstated. The range of validity is mistated: You can’t say “no other relevant variables are changing”. There are dozens of variables (energy, entropy, enthalpy, free energy, free enthalpy, volume, pressure, temperature, altitude, velocity, magnetic field, chemical potentials, et cetera). You simply cannot hold constant every variable that is not explicitly being differentiated; you to be specific about which ones are being held constant.
This is a problem. I know of physics professors and chemistry professors who are absolutely convinced that S ≡ k log Ω by definition. Then these professors get called upon to teach the thermo course (or the “physical chemistry” course) and the problem spreads.
Conservation is such a fundamental idea that inadequate coverage seems like a serious omission, but I rate this a (0) not a (−) because the book doesn’t say outright wrong things about this issue, and supplementing a book is very much more pleasant than contradicting a book.
Under the heading of spiral development, here’s another remark: When the number of particles gets down to 2 or 3, additional issues crop up. These issues require a more nuanced understanding of temperature and its relationship to the Boltzmann factor. The way equilibrium and temperature are handled here is entirely reasonable for a first introduction, but there should be at least a hint of the limits of validity, and a hint as to where the next layer of detail might be found.
Alas, the rest book does not uphold the high standard we see on page 1. There are lots of naïve ideas set forth as if they were the last word on the subject.
It would be a lame excuse to argue that it is not possible to explain the limitations of the approximations being used. This is lame because we see on page 1 that it actually is possible to handle this in a reasonable way.
More importantly, there is no excuse – lame or otherwise – when the limitations are actually stated, but stated wrongly, as on page 111.
It is unreasonable to expect that everyone who reads the book will work all the exercises. This is a problem, because some important results are hidden in the exercises:
|S = −∑ p log p (1)|
can be found in problem 6.43 on page 249. The equation is evidently considered so unimportant that it is not even given a number.
The following items don’t closely correspond to any of the “checklist” items.
(−) On page 19 : exercise 1.26 : As always, we should emphasize ideas, not terminology. Alas, this exercise is at best a question about terminology, and does little or nothing to clarify the ideas. How can any answer to this exercise be graded right or wrong? Experts wage holy wars about the terminology in this area.
(−) Starting on page 149, in the long section on «Free Energy as Available Work», the whole idea is fundamentally flawed. In the expression F = E − TS, the free energy is a property of the system itself, just as E, T, and S are properties of the system itself. However, the «available work» (to the extent that the idea makes sense at all) depends on the temperature of the environment.
Copyright © 2010 jsd