Which Thermodynamic Books Cover Non-Equilibrium Systems?

I tend to pick books based on how I plan to use them. If I'm working on transport coefficients or hydrodynamic limits, I almost always have 'Non-Equilibrium Thermodynamics' by S. R. de Groot and P. Mazur open — it’s dense but precise. For intuitive, example-driven learning (especially useful when teaching or preparing a talk), 'Modern Thermodynamics' by K. Kondepudi and I. Prigogine frames irreversibility in approachable language and gives lots of physical context.

When my focus is microscopic — like memory kernels, projection operators, or Green–Kubo relations — I gravitate toward R. Zwanzig’s 'Nonequilibrium Statistical Mechanics' and the Evans–Morriss book 'Statistical Mechanics of Nonequilibrium Liquids' because they bridge theory and simulation tricks. On the modern front, stochastic thermodynamics (fluctuation theorems, Jarzynski/Crooks relations) is best learned from papers by C. Jarzynski and G. E. Crooks, Seifert’s reviews, and Sekimoto’s 'Stochastic Energetics'. I also recommend supplementing books with lecture notes (e.g., Markus Esposito or Udo Seifert) — those notes are often fresher on applications like molecular motors or nanoscale heat engines.

Whenever I open the bookshelf to hunt down non-equilibrium thermodynamics, I get this excited, slightly nerdy rush — there’s so much variety depending on whether you want rigorous statistical foundations, continuum-level irreversible thermodynamics, or the modern stochastic-fluctuation perspective.

If you want a classic, go for 'Non-Equilibrium Thermodynamics' by S. R. de Groot and P. Mazur; it's a solid continuum treatment of irreversible processes and transport with clear derivations. For a broader, more conceptual introduction that blends classical and modern views, I really like 'Modern Thermodynamics' by K. Kondepudi and I. Prigogine — it’s readable and connects ideas to chemical and biological examples. On the statistical side, 'Nonequilibrium Statistical Mechanics' by R. Zwanzig and 'Statistical Mechanics of Nonequilibrium Liquids' by D. J. Evans and G. P. Morriss dig into projection-operator methods and computer-simulation friendly techniques.

If you’re fascinated by fluctuations, small systems, or molecular machines, explore U. Seifert’s review pieces and books/notes on stochastic thermodynamics, and K. Sekimoto’s 'Stochastic Energetics' for Langevin-level energetics. For a mathematically rigorous route, D. N. Zubarev’s 'Nonequilibrium Statistical Thermodynamics' and N. G. van Kampen’s 'Stochastic Processes in Physics and Chemistry' are invaluable. My study path usually mixes one continuum book, one stat-mech classic, and a couple of modern papers to see how theory meets simulations and experiments.

I usually tell friends who want a quick roadmap: read 'Modern Thermodynamics' by Kondepudi and Prigogine for motivation, then switch to de Groot and Mazur for the formalism. If you love math and microscopic derivations, go to Zwanzig and Evans–Morriss. For contemporary small-system topics, track down Seifert’s reviews and Sekimoto’s 'Stochastic Energetics'.

Beyond books, try some hands-on practice: implement a Langevin integrator, measure work distributions, and test Jarzynski’s equality numerically — the theory jumps off the page when you see fluctuation relations in simulated data. Also, keep an eye out for lecture notes by Esposito or Seifert and review papers about fluctuation theorems and stochastic thermodynamics; they often summarize the best modern perspectives and bridge the gap between the classics and current experiments. If you want, I can sketch a study plan that mixes chapters and papers with short coding exercises.

I like to think about these books as tools for different problems, so I compare them depending on the question I'm asking. For macroscopic transport and linear irreversible processes, 'Non-Equilibrium Thermodynamics' by de Groot and Mazur gives you constitutive relations, entropy production formulas, and practical examples; it’s the workhorse. If you want a pedagogical, less formula-heavy entry that still touches on irreversibility and complex systems, 'Modern Thermodynamics' by Kondepudi and Prigogine is friendlier and full of conceptual insights.

For anyone dealing with microscopic derivations, correlation functions, and memory effects, 'Nonequilibrium Statistical Mechanics' by Zwanzig is the deep-dive. Its projection-operator formalism explains how macroscopic irreversibility emerges from reversible dynamics. Evans and Morriss’s 'Statistical Mechanics of Nonequilibrium Liquids' is superb if you plan to run or interpret molecular dynamics simulations. For stochastic, fluctuation-driven problems — single-molecule experiments, small engines — Sekimoto’s 'Stochastic Energetics' and Seifert’s body of work (lecture notes and reviews) are where you’ll find modern theorems like Jarzynski and Crooks discussed in context.

Practical tip: pick one continuum book and one statistical book, and concurrently read review articles on fluctuation theorems. Doing exercises, coding simple Langevin simulations, and replaying classic derivations by hand makes the material stick much better than passive reading.

If I had to recommend a compact set for someone just diving in: start with 'Modern Thermodynamics' by K. Kondepudi and I. Prigogine for big-picture intuition, then read 'Non-Equilibrium Thermodynamics' by de Groot and Mazur for continuum theory. For microscopic approaches, Zwanzig’s 'Nonequilibrium Statistical Mechanics' is indispensable.

To round things out, pick up Sekimoto’s 'Stochastic Energetics' and Seifert’s papers on stochastic thermodynamics for fluctuation theorems and small-system energetics. Throw in van Kampen’s 'Stochastic Processes in Physics and Chemistry' if you need firm grounding in stochastic methods. That mix covers continuum irreversible thermodynamics, statistical treatments, and the modern stochastic perspective, which is where a lot of exciting research is happening.