The contents of a well mixed or stirred reactor are assumed to be nearly spatially uniform due to high diffusion rates or forced turbulent mixing. In other words, the rate of conversion of reactants to products is controlled by chemical reaction rates and not by mixing processes. Thus we consider that the reactor is "limited" by reaction kinetics. An essential element of the stirred reactor model is the assumption that the reactor is sufficiently mixed to be described well by spatially averaged or bulk properties. For low-pressure processes (1 Torr and below), the dominance of species diffusion renders this assumption valid in many practical applications. For example, this is a very good assumption for the low-pressure, highly diffuse operating conditions of most plasma-etch reactors and some thermal CVD systems. The major advantage of the well stirred approximation lies in the relatively small computational demands of the mathematical model. Such a model allows investigators to easily consider and analyze large, detailed chemical reaction mechanisms or complex reactor networks.

In addition to fast mixing, the modeling of homogeneous reactors requires several assumptions. First, mass transport to the reactor walls is assumed to be infinitely fast. Therefore, the relative importance of surface reactions to gas-phase reactions is determined only by the surface-to-volume ratios of each material and the relative reaction rates (rather than by transport constraints). Second, the flow through the reactor must be characterized by a nominal residence time, which can be deduced from the flow rate and the reactor volume.

Further assumptions and limitations are described for specific model formulations in the following sections.