(87e) Computational Fluid Dynamics Modeling of Mixing and Reaction of High-Viscosity Liquids in a Continuous Flow Reactor

Process intensification is a strong trend in chemical manufacturing which aims to conduct chemical processes with drastically increased energy and resource efficiency, lower capital and/or operating cost, and reduced physical and environmental footprint. In this context, the transition from traditional batch to continuous processes, which has been conducted in the pharmaceutical industry over the past decade, is now reaching the much larger scale of specialties and fine chemicals production. Among the many challenges faced during this transition is the requirement for rapid mixing of reactants in order to achieve optimum conversion and selectivity and consistent product quality. This problem is more pronounced in cases where the reactants are immiscible and/or have high viscosities at operating temperatures, as is often the case for specialty chemicals such as lubricants and dispersants.

It is well known that mixing is affected by material properties such as density, viscosity and surface tension; by operating conditions such as pressure and temperature; and by flow conditions and geometries. Chemical reactions between the mixing liquids further affects the mixing patterns in the reactor by varying the properties and local temperatures at different points in the fluid. However, while mixing has been studied for many decades, the detailed mixing behavior of such complex systems is still not fully understood, yet critical for reactor design and operation.

In the present study, we present computational fluid dynamics (CFD) simulations aimed towards understanding the mixing behavior of highly viscous liquids (polymeric anhydrides and amines) that are typically used in manufacturing dispersants via amination of the anhydride. Beyond a systematic analysis of the mixing behavior inside a continuous flow tubular reactor as a function of some of the variables listed above, we will discuss the use of appropriate metrics to study mixing, such as coefficient of variation, exposure and reaction conversion. Furthermore, the effect of different inlet geometries—ranging from co-axial flows to spray-type feed configurations—as well as the use of in-line static mixers will be discussed. Key results from CFD simulations are validated against laboratory experiments with non-reactive flow visualization as well as reactive tubular reactor experiments. The results of this study are being used to assist optimal design of experiments and scaling up to pilot-plant scale.