Gary K. Patterson

APPLICATION OF TURBULENCE FUNDAMENTALS TO REACTOR MODELLING AND SCALEUP

Abstract Applications of the fundamentals of turbulent mixing become clear once those fundamentals are understood. The first article in this series presented those fundamentals, in order to show how to apply turbulent mixing fundamentals modelling and scaleup, this article covers the following topics: 1. reaction types and their interaction with mixing; 2. closure of the Reynolds equations for mixing and reactions; 3. application to complex geometries; 4. random coalescence-dispersion modelling; 5. application to complex chemistry. The most difficult problem in applying our knowledge of turbulence to mixer modelling and scaleup is the choice of model complexity. The levels of model complexity available and how to apply them to various problems are presented following the introduction.

Effect of Degradation by Pumping on Normal Stresses in Polyisobutylene Solutions

Normal stress measurements were made on dilute solutions of polyisobutylene in three low‐viscosity solvents at high shear rates with a jet thrust apparatus in which the thrust tube was mounted on the frame of an analytical balance. Thrust measurements could be easily read to 0.01 g and replicate measurements agreed closely with the original ones. Mechanical degradation of polyisobutylene in cyclohexane and toluene caused significant reductions in (P11−P22)w values at high shear rates and in the amount of drag reduction in turbulent flow. Molecular weight distribution curves and estimates of viscosity‐average molecular weights of the fresh and degraded solutions in toluene indicate that these effects noted in this solvent are caused by the breakdown of small amounts of high molecular weight polymer, with only a small change in the viscosity‐average molecular weight.

Dynsyl: a general-purpose dynamic simulator for chemical processes

Abstract DYNSYL was developed at Lawrence Livermore Laboratory (LLL) in order to provide a means of simulating chemical process dynamics in nuclear fuel processing systems. The program is a modification of DYNSYS with several new subroutines for process unit simulation. The program can be used to generate process data or to provide estimates of process performance; it simulates both steady-state and dynamic behavior. This report describes the overall structure of DYNSYL and includes some example problems.

Turbulence structure in drag reducing polymer solutions

Longitudinal turbulence intensities measured by split‐film anemometry in degraded polymer solutions in pipe flow peaked much farther from the wall than longitudinal turbulence intensities in a Newtonian oil at the same Reynolds numbers. The peak values were at about the same level for the degraded solutions as for the oil, but were higher and nearer the wall for fresh polymer solutions. Radial turbulence intensities were lower for the polymer solutions at all locations under all conditions. Drag reduction for the polymer solutions ranged from 1 to 21% based on Newtonian fluid of equal viscosity. For the fresh solutions the Reynolds stresses dropped to zero or low values much farther from the wall than normal, indicating a region of almost complete turbulence shear strain recovery near the wall (positive uv excursions balanced negative uv excursions). Numerical computations with a turbulence model involving balance equations for turbulence energy and turbulence energy dissipation rate (the k‐e model) show ...

Simulation with Validation of Mixing Effects in Continuous and Fed-Batch Reactors

Publisher Summary This chapter describes the effects of scale-up of stirred reactors on product yield for competitive reactions with various rate constant ratios, for two impeller types, and for continuous and fed-batch modes of operation, by numerical simulation. Many of the chemical reactions that are simulated, have been studied experimentally, so comparisons under identical reaction conditions are used to validate the numerical method. The simulations are done using FLUENT as the transport computation engine with added subroutines to simulate mixing rates and chemical reaction rates as a function of the degree of mixing. The closure used for the chemical component mass balances is the Paired-Interaction (P-I) Model. The P-I model treats component segregation (lack of mixing) as a conserved quantity requiring a balance equation for its production, dissipation and transport. The chapter also explores typical strategies for scaling-up the mixing power for stirred chemical reactors: constant power per unit volume and constant impeller tip speed. Most of the simulations are for continuous flow reactions since they require about one-tenth of the time to compute than simulations for fed-batch reactions. Some of the cases simulated are for fed-batch operation, which generally give higher yields than continuous flow operation for the same mixing rates, and concentrations, but directional effects for changes in size, feed location, and impeller rotation rate are usually the same for the two modes.