Search

A sophisticated method was developed for evaluating simultaneously and accurately both the average specific resistance and average porosity of the filter cake formed in unstirred dead-end ultrafiltration of nanocolloids such as protein solution and nanosilica sol. In the method, a step-up pressure filtration test was conducted by using a filter with a single-stage reduction in the effective filtration area. The influence of the pressure drop across the cake on not only the average specific cake resistance but also on the average cake porosity of highly compressible filter cake was evaluated using only flux decline data in one dead-end filtration test, taking advantage of the decrease in the cake thickness caused by the pressure increase. As a result, the cake properties were easily determined for a variety of nanocolloids. Constant pressure dead-end ultrafiltration data obtained under various pressures and concentrations were well evaluated based on the method proposed. © 2015 American Institute of Chemical Engineers AIChE J, 61: 4426–4436, 2015

A potential method has been developed for evaluating simultaneously both the average specific resistance and average porosity of the filter cake formed in unstirred dead-end ultrafiltration of nanocolloids such as bovine serum albumin solution and silica sol. The method consists of variable pressure filtration followed by constant pressure filtration. The relation between the average specific cake resistance and the pressure drop across the cake was determined from the evolution of the filtration rate with time in the course of the variable pressure filtration period, based on the compressible cake filtration model. The average porosity was evaluated from the significant flux decline caused by a sudden reduction in the cake surface area in the middle of the constant pressure filtration period. The pressure dependences of both the average specific cake resistance and average cake porosity were obtained from only two runs which differed from each other in the pressure profiles. © 2014 American Institute of Chemical Engineers AIChE J, 60: 3869–3877, 2014

The single step-up pressure filtration test was developed to determine the pressure dependence of average specific resistance of the cake formed in ultrafiltration of a variety of nano-colloids over a wide range of pressure drops across the cake. The values of the average specific resistance at extremely low pressures were obtained from only the flux decline data through the use of the distinct time variation of the pressure drop across the cake generated by using the ultrafiltration membrane with a high hydraulic resistance under the low filtration pressure in the first step of filtration. The values at higher pressures were obtained from the time variation of the filtration rate induced by a stepwise increase in the pressure. The correlations between the average specific cake resistance and the pressure drop across the cake were evaluated using only the flux decline data for a variety of different proteins and nanoparticles. © 2013 American Institute of Chemical Engineers AIChE J, 60: 289–299, 2014

Particle fouling in constant pressure submerged membrane filtration is analyzed. A particulate sample, with a wide size distribution ranging from submicron to micron, is filtered using a flat-sheet membrane module. The effects of filtration pressure and aeration conditions on the particle deposition probability, cake properties and filtration flux are discussed. Particle deposition is analyzed theoretically using a force balance model. The particle size distribution in a filter cake is smaller than that in the original suspension and gradually decreases during a filtration. An increase in filtration pressure leads to a higher filtration flux in the early period of filtration; however, after 3000 s the flux gradually approaches a pseudo-steady value. The filtration flux is significantly enhanced by sparging air bubbles. An increase in air volumetric flow rate or a decrease in bubble size leads to a smaller particle size distribution in the cake. The particle deposition probability, cake mass and average specific filtration resistance calculated by using the proposed models agree fairly well with available experimental data.

Microfiltrations of Saccharomyces cerevisiae yeast particles of various deformabilities were conducted to examine the blocking mechanism due to particle deposition during the initial stage of the microfiltration process. The S. cerevisiae particles were treated with glutaraldehyde to form particles of various deformabilities. To compare differences in the blocking mechanisms among the different particles during microfiltration, a blocking model was adopted to estimate the filtrate flux, which accounts for initial blocking due to pore obstruction and subsequent blocking due to the growth of a cake over these initially blocked regions. The surface morphology of the cake layer on the membrane surface and the blockage by particles in membrane pores was examined by scanning electron microscopy (SEM). Analysis of the blocking mechanism of particles of varying deformability during microfiltration based on the blocking model indicated that softer particles resulted in more severe pore blocking problems and greater flux decline. A comparison of the theoretical results with experimental observations by SEM showed good qualitative agreement.

The mechanism of cake formation is studied in 2–D cross-flow filtrations. By analyzing the forces exerted on a depositing particle, the value of the critical friction angle, βc, which leads the particle to be stuck on the contact position, is determined. Not only the probability of particle deposition but also the size distribution of deposited particles can be calculated from the theoretical value of βc. Numerical programs are constructed to simulate the packing structure and size distribution of particles in a formed cake and the path of cake compression. The local cake properties in the cake, such as local porosity and local specific filtration resistance, are also estimated by applying the design program and the data of filtration volume vs. time. The predicted values of average cake porosity and average specific filtration resistance agree very well with the experimental data.