Critique of a proposed stability criterion for chemical systems far from equilibrium.

Noyes [Noyes, R. M. (1981) Proc. Natl. Acad. Sci. USA 78, 7248-7249] has obtained a result for the Gibbs free energy per unit mass in a continuous flow stirred tank reactor, which states that the steady-state free energy increases monotonically with decreasing residence time for any stable stationary point. We show that his derivation is based on an erroneous stability criterion. We further investigate the dependence of the Gibbs free energy on residence time for the case of mass-action kinetics and ideal mixtures. We then apply this development to Schlögl's trimolecular reaction mechanism [Schlögl, F. (1971) Z. Phys. 248, 446-458] and show that Noyes' conjecture is violated for this simple example.