New bottles for some new wine

A naive view might be that thermodynamics is an exhausted seam. Chemists - at whom this book is primarily directed - have incorporated the concepts of thermodynamics into their universe of discourse and, largely through the deeply thoughtful work of J. W. Gibbs in the late 19th century, now routinely, especially as undergraduates, calculate the equilibrium properties of chemical reactions from thermodynamic data.

So how can thermodynamics be rendered modern? Two routes present themselves. One (a cheat) is simply to pour old wine into new bottles. Another is to acknowledge the revealing word of the previous paragraph, equilibrium. The new wine of classical thermodynamics is the study of systems that are far from equilibrium and where structures - the dissipative structures of the subtitle - can emerge from either actual or notional flows of matter through systems.

To some extent, this text is a mixture of new and old in both wine and bottles. Broadly speaking, the text covers in its initial section what we once thought undergraduate physical chemistry courses could cover; in its later sections I judge that it becomes too difficult for most graduate courses.

Despite the seduction of its subtitle, I would guess that the content would not be particularly useful for engineers. On account of its range, I think the text is most suitable for instructors who want to achieve a synoptic vision of the range of modern (equilibrium and non-equilibrium) thermodynamics.

The presentation of the early material is straightforward, and made more enjoyable by the asides that illustrate thermodynamics in the context of modern science.

However, because the power of thermodynamics stems from the precision of its language, I was unhappy to see so much looseness in the presentation, careless use of language, and the disregard of modern conventions for symbols and units.

Although it is interesting to see photographs of so many giants of the field, I would have preferred - at this level of presentation - more attention paid to the selection and presentation of diagrams.

Potential readers should also be advised - warned is too strong a term - that much of the presentation of the material is in terms of de Donder's concept of chemical affinity, which is not a concept of common currency, except perhaps in Belgium. However, there are good reasons for hanging the presentation on this concept, as becomes clear as we approach the second half of the book and the modern wine - too heady for some - of systems far from equilibrium. This part of the text is essentially an accessible and welcome exposition of Prigogine's own contribution.

One deficiency of the exposition is the almost total lack of physical interpretation of the enormous number of equations, some of them of great complexity. A truly modern account of modern thermodynamics would impart more insight. Nevertheless, this text, together with its Mathematica code for some of its equations, is a useful instructor's resource. It may help to make non-equilibrium studies accessible to a wider professional public; but not, I suspect, to undergraduates.

Peter Atkins is professor of chemistry, University of Oxford.