Font Size: a A A

Study On Crossflow Microfiltration Process Enhanced By Turbulence Promoter

Posted on:2014-01-16Degree:DoctorType:Dissertation
Country:ChinaCandidate:Y F LiuFull Text:PDF
GTID:1221330395998678Subject:Chemical Engineering
Abstract/Summary:PDF Full Text Request
The permeate flux of microfiltration sharply declines with filtration time owing to the phenomenon of membrane fouling, which seriously hinders the wide applications of MF due to the increased energy consumption. The use of turbulence promoters can significantly increase the crossflow velocity or wall shear stress under the same operation condition, which can effectively prevent particle deposition on the membrane surface, thereby improving the permeate flux of MF. The flux improvement by turbulence promoter depends on the improved hydrodynamics of fluid flow. However, the hydrodynamics effects and the mechanism of flux enhancement by turbulence promoter have still not been clearly understood. In this thesis, central baffle and wall baffle were used as turbulence promoter to enhance the permeate flux during the crossflow MF of particulate suspension. A novel helical screw insert with a rectangular section was designed. Computational fluid dynamics (CFD) simulations of fluid flow in the tube filled with turbulence promoters were conducted to investigate the hydrodynamics effects responsible for the flux improvement. In order to explore the intrinsic mechanism of flux enhancement, the effects of turbulence promoter on the cake properties were investigated. The turbulence promoter-assisted MF process was successfully modeled by the artificial neural network (ANN) to predict the efficiency of flux enhancement. The operation variables in MF process were optimized by ANN to achieve high flux enhancement efficiency, which provides a useful guide for the applications of turbulence promoter.Firstly, both central baffle and the wall baffle were used to enhance the permeate flux during crossflow MF of calcium carbonate suspension. The effects of baffle configuration, baffle geometric parameters including baffle constriction ratio (β), baffle spacing (L/D) and baffle arrangement on the flux enhancement efficiency were experimentally investigated. It reveals that both β value and (L/D) value of baffle play an important role in the flux enhancement efficiency. The baffle with a larger β value can obtain higher flux improvement efficiency. The optimum (L/D) value of central baffle strongly depends on the β value. While the optimum (L/D) value of central baffle is independent of the β value. The central baffle can achieve higher flux improvement efficiency than wall baffle in terms of same geometric parameters. The combination use of central baffle and wall baffle achieves higher flux enhancement efficiency than the use of central baffle or wall baffle alone. In order to explore the hydrodynamics effects responsible for flux improvement by the baffles, CFD simulations of fluid flow in the baffle-filled tube were conducted. It reveals that the vortex formation due to the presence of baffles induces the intense velocity fluctuation, thereby increasing the turbulence intensity of fluid flow, which greatly disrupts the development of the boundary layer and effectively prevents the particle deposition on the membrane surfaces. Therefore, the permeate flux of MF membrane is significantly improved by the baffles.Then, a novel helical screw insert with a rectangular section was designed. CFD simulations of fluid flow in the tube with the newly designed helical screw insert were conducted to investigate the hydrodynamics effects responsible for flux improvement. The effects of geometric parameters of helical screw insert on the flow hydrodynamics were theoretically studied. Due to the presence of helical screw insert, the fluid flow in the tube is mainly divided into two parts, that is, helical flow within the helical groove and axial flow through the radial clearance gap. The turbulence intensity of flow and wall shear stress is significantly increased owing to the intense mixing of helical flow and axial flow, which is responsible for the flux enhancement. There is no stagnant region or dead zone at the neighborhood of tube wall and no secondary flow or vortex formation within the helical groove. Compared to the helical screw insert with semi-circular section reported in the literature, the pressure drop along the tube can be reduced by25%and wall shear stress is increased by6.7%when using the newly designed turbulence promoter. Simulative results indicate the geometric parameters of helical screw insert have a significant influence on the flow hydrodynamics. Both wall shear stress and pressure drop along the tube increase with the increase in either the helical diameter (Dh) or the central rod diameter (Dr), and decrease with the increase in the width of helical groove (λ). In terms of the critical value of wall shear stress, the geometric parameters of helical screw insert was optimized as following:Dh=11mm, λ=12mm, Dr=4or5mm. The helical screw insert without a central rod produces a smaller pressure drop and a lower wall shear stress.In order to explore the intrinsic mechanism of flux enhancement, the effects of newly designed turbulence promoter on the cake properties were experimentally investigated during the crossflow MF of particulate suspension. And the effects of operation conditions on the cake properties were studied. It reveals that the cake thickness diminishes from0.59to0.1mm, the cake porosity increases from0.55to0.65and the average particle size decreases from5.15to1.99μm due to the presence of helical screw insert under the same operation condition. The specific resistance of cake increases2.45times due to the dimished average particle size, indicating the negative effect of turbulence promoter on the permeate flux. The positive effect of turbulence promoter on the permeate flux due to the remarkable reduction in cake thickness overwhelms its negative effect due to the increased specific resistance. Therefore, the cake resistance is significantly reduced by turbulence promoter. The operation conditions of MF have significant influences on the cake parameters. The cake thickness increases, the cake porosity decreases and the average particle size increases with an increase in transmembrane pressure (TMP). The cake thickness decreases, the cake porosity increases and the average particle size decreases with an increase in the inlet velocity. The cake thickness increases and the average particle size decreases with an increases in the feed concentration. The cake porosity increases with an increase in the feed concentration when without helical screw insert, and almost keeps stable when using a helical screw insert. The effects of both TMP and feed concentration on the cake properties can be weakened to some extent, and the effects of inlet velocity on the cake properties can be strengthened by helical screw insert.At last, the turbulence promoter-assisted MF process was successfully modeled by ANN, which can predict the flux improvement efficiency under various operation conditions. The optimal ANN model architecture is that neuron numbers in the hidden layer is12and transfer functions in the hidden layer and output layer are logsig and tansig, respectively. The effects of operation conditions on the flux enhancement efficiency were analysized using ANN model. It reveals that the flux enhancement efficiency first increases and then decreases with an increase in TMP, and increases with an increase in both the inlet velocity and feed concentration. TMP has more important influence on the flux enhancement efficiency than the feed concentration or the inlet velocity. The optimal operation conditions were optimized to achieve the highest flux improvement efficiency, which provides a useful guide for the applications of turbulence promoter. It suggests that the highest flux enhancement efficiency can be obtained by applying both a high inlet velocity and a low TMP at low feed concentration, and by applying both a high inlet velocity and a high TMP at high feed concentration.
Keywords/Search Tags:Microfiltration, Turbulence Promoter, Flux enhancement, CFDSimulation, Neural Network
PDF Full Text Request
Related items