1 AIT Asian Institute of Technology

Mathematical modelling of transient behaviour of deep-bed filtration

AuthorChang, Jing Song
Call NumberAIT Diss. no. EV-89-02
Subject(s)Filters and filtration--Mathematical models

NoteA doctoral dissertation submitted in partial fulfilment of the requirements for the degree of Doctor of Engineering
PublisherAsian Institute of Technology
AbstractMost of the mathematical models in deep bed filtration have been developed for predicting clean bed filter efficiency and efficiency during ripening period of filtration. Most of them do not take into account the size distribution of particles in suspension and pores of granular medium. The effect of chemistry of liquid on transient behaviour of filtration has not been studied in detail. In this research therefore, the above aspects have been included in the mathematical models and the models were tested with experimental results. The O'MELIA-ALI model for ripening stage of filtration has been modified to describe the entire cycle of filtration. Two mathematical models developed based on two different concepts to describe the entire cycle of filtration were tested with experimental results. The first model was based on the detachment of deposited particles. The degree of detachment was assumed proportional to the hydraulic gradient and the number of particles already retained on the filter grain. The second model was based on the concept that there exists a maximum limit of particle deposition on the filter grain. It was found that the first model can simulate better than second model can. This may be an experimental proof for the concept of detachment which is now agreed by most of the researchers. The ultimate specific deposit was estimated experimentally for a wide range of filtration velocities to give a qualitative justification on the applicability range of the proposed models. A mathematical model was proposed incorporating one of the important parameters, the size distribution of particles in suspension, which is the case in real situation. The improvement of removal efficiency of finer particles in the presence of coarser particles in suspension is explained based on the assumption that some of the retained coarser particles in addition to retained finer particles act as "particle collectors" in the removal of finer particles. The concept of detachment of retained particles due to hydraulic gradient is also incorporated to describe the decrease in removal efficiency during the filter breakthrough period. The four coefficients appearing in this mathematical model were determined for different operating conditions using experimental results obtained from filtration of suspension of uniform pollen grains. Although these model coefficients were different for different operating conditions, a uniform trend was observed in the variation with operating conditions. These model coefficients calculated for suspensions with single size particles were then used to predict the removal efficiencies when particles of different sizes are in mixture in suspension. Some data collected from Antwerp Water Works, Belgium, were used to verify the assumption made in the mode l formulation. Another important parameter of filtration, the pore size distribution has not been taken into account in the existing mathematical models. In this study, the effect of pore size distribution has been studied theoretically. For this, the O'MELIA - ALI model which falls in the category of external flow model was modified to the internal flow model and this modified model was then incorporated with the model developed using Effective Medium Approximation (EMA) approach to investigate the effect of pore size distribution in deep bed filtration. Six different pore size distributions were assumed and the filter performances were simulated for these cases to get the preliminary idea on the effect of pore size distribution in deep bed filtration. The removal efficiency of particles in deep bed filtration depends both on transport and attachment mechanisms of particles. The particle attachment mechanisms depend mainly on chemical characteristics of aqueous phase, suspended particles and filter media. In this study, the effect of ionic strength on the particle removal efficiency during ripening period was studied experimentally. The removal efficiencies were related to ionic strength and zeta potential. The clean filter bed efficiency and removal efficiency during the ripening period on different ionic strength were quantitatively related to the two model coefficients appearing in O'MELIA - ALI model. These model coefficients represent the clean bed filter efficiency and ripening stage filter efficiency respectively. An observed uniform trend in the variation of these coefficients with ionic strength of suspension showed the importance of quantitative incorporation of chemical effects in filtration mathematical models.
Year1989
TypeDissertation
SchoolSchool of Environment, Resources, and Development (SERD)
DepartmentDepartment of Energy and Climate Change (Former title: Department of Energy, Environment, and Climate Change (DEECC))
Academic Program/FoSEnvironmental Engineering and Management (EV)
Chairperson(s)Vigneswaran, S.;
Examination Committee(s)I, Fude;Chongrak Polprasert;Gupta, Ashim Das;Huynh, Ngoc Phien;Ives, K.J.;
Scholarship Donor(s)Japanese Government;
DegreeThesis (Ph.D.) - Asian Institute of Technology, 1989


Usage Metrics
View Detail0
Read PDF0
Download PDF0