REMOVAL OF FINE PARTICULATE MATTER FROM EXHAUST GASES BY METALLIC MICROSIEVES
Esther Stahl, Josef Robert and Görge Deerberg (pages 36-40)
Metallic microsieves are promising filter media for removing fine particulate matter from gases. Therefore, a study has been carried out to show their filtration performance (pressure drop and filtration efficiency) with respect to different boundary conditions (volume flow, temperature). Experiments showed high particle retention rates after a short time, depending on the hole diameter. By applying microsieves with an average pore diameter of 8 µm, a retention rate of more than 99% could be obtained after 90-160 s. The practical application of microsieves as a filter in the exhaust gas of wood fired furnaces is shown to give good results.
DESIGN OF A NEW HIGH PERFORMANCE DRUM FILTER FOR THE CHEMICAL INDUSTRY
Thomas Langeloh and Reinhard Bott (pages 40-45)
Drum filters represent the most important and most applied rotary filter technology that is used in a wide range of industrial applications with very different products. For filtration tasks in the chemical industry BOKELA has developed a new drum filter generation which satisfies the needs and requirements of these applications, e.g. the filtration of crystal suspensions. These modern high performance drum filters are characterised by high specific solids throughput, intensive cake wash, sharp filtrate separation, low moisture contents, complete cake discharge, reliable operation and ease of maintenance. The fundamental aspect is a well calculated and well-designed hydraulic system.
DETERMINATION OF THE PRESSURE DEPENDENCE OF PERMEABILITY CHARACTERISTICS FROM A SINGLE CONSTANT PRESSURE FILTRATION TEST
Eiji Iritani and Nobuyuki Katagiri (pages 46-50)
A single constant pressure filtration test procedure has been developed for easily determining the pressure dependence of permeability (or average specific cake resistance). The method makes use of the variation of the effective pressure drop across the filter cake with time generated by using a filter medium with high resistance. The validity of the method was confirmed by constant pressure filtration of bentonite slurry using an ultrafiltration membrane with a nominal molecular weight cut-off of 1000 Da. It was also revealed that the threshold value of pressure below which the average specific cake resistance remains constant can be readily obtained from the method.
MITIGATING AND MODELLING MEMBRANE FOULING BY BIOACTIVE MIXTURES
Zaid S. Saleh, Judie M. Farr and Steffen A. Friedrich (pages 50-57)
In this study, the fouling that occurred during batch nanofiltration processing to separate phenolic phytochemical compounds from clear apple juice concentrate was characterized and modelled using various operating parameters. Polysulphone SelRO® spiral wound membranes with a molecular weight cut-off (MWCO) of 1 kDa and 0.25 kDa were used on a pilot laboratory scale rig. The retention of various compounds including total and individual phenolics, total soluble solids, proteins and certain types of sugars were determined analytically. The effects of temperature, pH, trans-membrane pressure and feed concentration on permeate flux, fouling and membrane performance were determined.
Different types of fouling mechanisms, including reversible and irreversible concentration polarization, cake formation, pore blocking or a combination of the aforementioned, were investigated. The trends in fouling rates were compared with predictions from available fouling models, and the fouling mechanism was considered. It was found that cake formation was the dominant reversible fouling mechanism in the 1 kDa membrane and that pore blocking was the dominant type of fouling occurring in the 0.25 kDa membrane. Reversible cake formation was greater at higher feed concentrations and temperatures and lower at low pH values and pressures. Furthermore, performing diafiltration increased the permeate flux remarkably by reducing the concentration polarization effect.
POLYMER EXTRUSION FILTER DESIGN WITH A HYBRID PSO-GA OPTIMIZATION APPROACH
Kathleen R. Fowler, Eleanor W. Jenkins and Brian M. McClune (pages 58-64)
We discuss an optimization study of two layer extrusion filter designs using a three-dimensional computational simulator and derivative free hybrid optimization methods. The simulator models flow of a non-Newtonian fluid through a multi-layered filter with debris deposition. Previous studies used a derivative free sampling algorithm to maximize different performance measures of one- and two- layer filters, relative to changes in porosity and pore diameter in each layer. A challenge in those studies, and the motivation for this work, is that the single search optimization algorithm used would converge to a sub-optimal filter design for certain starting points.
In this work, we apply a new hybrid optimization algorithm that combines two heuristic search methods: a genetic algorithm (GA) and a particle swarm optimization (PSO) algorithm. Both are known to exhaustively search the design space and are less likely to stagnate at a local minimum. They do, however, require a significant number of calls to the simulator. This is computationally expensive, as each call may require over an hour of computation time. The efficiency is improved by incorporating surrogate functions (i.e. a cheaper approximation to the real objective function) into the search phase. This paper presents numerical results for a two layer extrusion filter design and discusses extensions to more complicated filter designs.