Computational fluid dynamics for design and optimization of tubular low-density polyethylene reactors

Kolhapure, Nitin
Major Professor
Rodney O. Fox
Committee Member
Journal Title
Journal ISSN
Volume Title
Research Projects
Organizational Units
Journal Issue
Chemical and Biological Engineering

Polymer reactor models often assume that the reactants are mixed rapidly and thus the concentrations can be considered to be uniform at small scales. However, for fast reactions or for viscous systems, poor mixing of chemical species significantly affects the reactor performance while adversely affecting product quality. The purpose of this research is to formulate a generalized algorithm based on state-of-the-art computational fluid dynamics (CFD) techniques such as full or presumed probability density function (PDF) methods to better understand the role of micromixing in mixing-sensitive chemical processes. The impressive capabilities of the algorithm are investigated using an industrial test-case of tubular low-density polyethylene (LDPE) reactors. The precise control and optimization of these reactors are of primary industrial concern due to tight coupling between fluid dynamics and complex LDPE chemistry under extreme operating conditions. CFD simulations are carried out by combining the CFD model and detailed LDPE chemistry into a single FORTRAN code as well as into a commercial CFD software--Fluent(c). The technique of in situ adaptive tabulation enables the computationally efficient inclusion of the stiff and non-linear LDPE chemistry. Results include temperature profiles, various species profiles and prediction of polymer quality with and without chain transfer mechanisms under various inlet and operating conditions, along with comparisons against pilot-plant scale data and/or comparison of various CFD techniques for accurate and efficient predictions of micromixing effects. Interesting features such as a bimodal temperature distribution and local hot-spots as well as global decomposition after an induction time or due to pulsating initiator feed are also observed under certain conditions using the full PDF simulations near critical points where instabilities occur. Considering the advantages of the two CFD methods, efforts are also directed towards efficient combination of the two techniques in order to obtain reactor stability plots and catalyst efficiency profiles, which are extremely helpful in operational decisions as well as design of control strategies. Thus the study not only illustrates the importance of mixing effects on LDPE polymerization in tubular reactors, but also yields insight into choosing appropriate operating conditions for maximizing catalyst efficiency while controlling reactor dynamics and product quality in plant-scale tubular LDPE reactors.