Computational studies of reacting flows with applications to zinc selenide nanoparticle synthesis and methane/hydrogen separation

This work is a numerical study of the design and operation of two reacting flow systems, each with great potential in their fields. The design of reacting flow systems by computer simulations are successfully used in science and engineering to evaluate design geometries and operation, without resorting to experimental trial and error that is expensive, time consuming and, in some cases, dangerous. The models of the two systems described in this work are based on fundamental conservation equations for momentum and mass transfer coupled with chemical reaction kinetics and particle dynamics.; The first part of this work is a study aiming to elucidate the transport phenomena and chemical reactions that control the size of ZnSe nanoparticles formed by a new vapor-phase synthesis route. The nanoparticles are synthesized by reacting vapors of (CH₃)₂Zn:N(C₂H ₅)₃ adduct with H₂Se gas (diluted in hydrogen) fed continuously from opposite sides into a counterflow jet reactor. The nuclei of the nanocrystals are formed by a direct condensation reaction near the stagnation point. The nuclei grow into nanoparticles by coalescence/coagulation and by surface growth reactions. A 2D model of an axially symmetric reactor was developed that includes descriptions of flow, mass transfer by convection and diffusion, chemical kinetics, particle nucleation, coagulation and surface growth. The coupled nonlinear partial differential equations of the model were solved using the Galerkin Finite Element Method. The model was used to study the relative importance of the underlying physical and chemical phenomena in controlling particle size and particle size distribution. Model predictions compared well with the limited experimental data available for this system. The model was also used for model-assisted design of the experimental counterflow jet reactor, where vapor-phase synthesis of ZnSe nanoparticles was demonstrated for the first time.; The second part of this work involves the development of a predictive model describing pressure and concentration dynamics during Pressure Swing Adsorption (PSA) of binary (or pseudo-binary) gas mixtures. The separation of metane-hydrogen mixtures over 5A-zeolite was used as an example. The PSA cycle considered in this study includes the following 5 steps: (1) pressurization with product, (2) high-pressure adsorption, (3) cocurrent depressurization, (4) countercurrent blowdown and (5) countercurrent purge with product at low pressure. The PSA mathematical model describes the following processes gas flow in the bed (as axially dispersed plug flow) and the mass balance of the components of the mixture coupled to adsorption/desorption kinetics. The model results in a system of coupled partial differential equations in the axial bed dimension and time. The Galerkin Finite Element Method was used to discretize the equations in the axial direction of the bed. The resulting system of ordinary differential equations (ODE's) in time is solved by using an Euler full-implicit scheme. The model is being used by Chemical Design, Inc., for the initial design of PSA units.