Loading...
| Friend's email | |
| Your name | |
| Your email | |
| enter code | |
This page was sent successfuly
- Type of Document: M.Sc. Thesis
- Language: Farsi
- Document No: 44440 (06)
- University: Sharif University of Technology
- Department: Chemical and Petroleum Engineering
- Advisor(s): Molaei Dehkordi, Asghar
- Abstract:
- Fluidized bed spray granulation is an important powder production process with several key advantages compared to other powder production processes, such as spray drying. Moreover, in fluidized bed granulators, whole process takes place in just one apparatus and there is no need to side processing. Most important characteristics and essential part of fluidized bed granulation is the wetting of primary particles by atomized liquid binder to coating the primary particles and producing granules. To date, a lot of research has focused on modeling and understanding of the separate particle growth mechanisms which have been integrated into the population balance models. Although, these concepts can be useful in micro-level modeling of the process, there are a few works which has been done in this field. To the author knowledge, discrete element modeling (DEM) is the best method of modeling in capturing these micro-level sciences. Population balance models, supposing existence of sufficient knowledge about process mechanism which lead to use appropriate kernels, is so powerful in predicting the process behavior and dynamic variation of particulate phase quality. In using of population balance models in processes such as crystallization and polymerization, usually, a 1-dimensional population balance model (usually particle’s size or volume takes as the internal coordinate) is sufficient to simulate the process. But in the processes which deal with granulation, except in so simplified situations, 1-dimensional population balance models is not an appropriate choice for modeling. This is because of important internal coordinates such as granule’s porosity and liquid saturation which they should capture in the model. Therefore, to modeling granulation processes, we need to use multi-dimensional population balance. Until this date, several powerful and accurate models are presented for solving 1-dimensional populations balance equations, while solving multi-dimensional population balance equations is challenging yet. In discrete element modeling, we can accurately predict hydrodynamic behaviors of the fluidized beds; beside this, using this modeling method, lead us to use appropriately our knowledge about micro-level mechanisms in an optimized way. Also, due to large CPU TIME in numerical solution of discrete element models, it is not possible to use these models in simulating the process in industrial scales. But, DEM modeling can be considered as a valuable learning tool which can be applied to gain more insight into process mechanisms required for engineering scale modeling. Until today, accurate DEM models for simulation of fluidized bed hydrodynamics are released, but there is not much work about fluidized bed granulation process. Thus, presenting comprehensive model which considered important mechanisms of the process can be so useful and valuable. The present work can be separated in to general sections. In the first section, an accurate method for solving multi-dimensional population balance equations including coalescence and any differential term such as layering, attrition, wetting and consolidation is developed. In this section, at first, a new method for discretization of 2-dimensional population balance equation presents and then this method will be generalized to the n-dimensional situation. In the second part of the project, the fluidized bed spray granulation process is modeled and simulated using DEM method. In this model, it is tried to present a detailed model and various granulation phenomena such as wetting, nucleation, layering, coalescence, breakage, and gas encapsulation are accounted. Internal properties of the granules which are considered in this model are granules radius, mass, porosity, liquid saturation, linear velocity and angular velocity. Moreover, although this model is constructed on the 3-dimensional Cartesian, because of limitations of computational resources, simulations are done in the 2-dimensional geometry. In the following of the second section, effect of important process parameters including fluidization velocity, mass flowrate of the nozzle, deviation angle of the injection droplets from the centered line and the nozzle position on the product quality and time variations of total number of granules are examined carefully.
- Keywords:
- Modeling ; Simulation ; Granulation ; Fluidized Bed ; Discrete Element Modeling (DEM)
Digital Object List-
محتواي کتاب - view
Bookmark