Statistical‐Mechanical Foundations for a Generalized Thermodynamics of Dissipative Processes

A statistical-mechanical formalism for nonequilibrium systems, namely the nonequilibrium statistical operator method, provides microscopic foundations for a generalized thermodynamics of dissipative processes. This formalism is based on a unifying variational approach that is considered to be encompassed in Jaynes' Predictive Statistical Mechanics and principle of maximization of the statistical-informational entropy. Within the framework of the statistical thermodynamics that follows from the method, we demonstrate the existence of generalized forms of the theorem of minimum (informational) entropy production, the criterion for evolution, and the thermodynamic (in)stability criterion. The formalism is not restricted to local equilibrium but, in principle, to general conditions (its complete domain of validity is not yet fully determined). A H-theorem associated to the formalism is presented in the form of an increase of the informational entropy along the evolution of the system. Some of the results are illustrated in an application to the study of a model for a photoexcited direct-gap semiconductor.