Initial Period Of Filtration Research Articles

Preliminary Study of Hospital Wastewater Treatment with a Submerged Membrane Bioreactor (SMBR): Case Study of Songklanagarind Hospital – Thailand

The objective of this research was to study hospital wastewater treatment using a submerged membrane bioreactor (SMBR). The lab scale unit of SMBR with a working volume of 20 liters was operated at a hydraulic retention time of 0.416 day and 0.208 day at F/M ratio of 0.18 day–1 and 0.29 day–1. The operating conditions were set up to provide good biological treatment without sludge extraction, and two different permeate flux values were studied. The performance of membrane was studied by monitoring the variation of transmenbrane pressure (TMP) during filtration runs. The efficiency of MBR was investigated according to the daily measurements of pH, dissolved oxygen and temperature. The COD, BOD5, NH4+–N, TKN, color, turbidity, SS and Escherichia coli (E.coli) in influent and effluent were analyzed. Membrane fouling intensity occurred slowly when the system operated with flux at 10 L/h/m2 and 20 L/h/m2, which induced high TMP at the initial period of filtration. The fouling rate was at about 0.3022 mbar/day for the permeation of flux at 10 L /h/m2. The fouling rate still remained at 0.2774 mbar/day for 20 L/h/m2. The results showed the great effect of membrane use for total biomass retention and the removal rate of COD, BOD5 and E.coli were over 90%. The characteristics of sludge in SMBR showed healthy floc formations with good settling. Although the ratio of MLVSS/MLSS was lower than the normal range (about 0.2), it was found that the average values of COD and NH4+–N in permeate were lower than 2–80 mg/L and 0.05–6.755 mg/L while the value of turbidity was also less than 3 NTU.

Modeling of filtration characteristics during submerged hollow fiber membrane microfiltration of yeast suspension under aeration condition

A modeling of filtration characteristics during the submerged hollow fiber membrane microfiltration of yeast suspension under aeration condition was developed on the basis of hydrodynamic mathematical equations that describe the local flux distribution, particle deposition and gas–liquid two-phase flow. The effects of average operating flux, fiber length, inner diameter and aeration rate on the distributions of local flux and filtration resistance were investigated using the developed model. The model was validated using the literature data. The simulated results showed that the asymmetry of initial local flux distribution was intensified with the increase of aeration rate. The local fluxes along the fiber experienced the self-adjustment and reconstruction due to the changes of filtration resistance as the filtration processed. Furthermore, the local flux distribution along the fiber with a short length and large fiber inner diameter at a high operating flux would tend to become relatively uniform at the initial period of filtration.

Filtration and drying characteristics of casein

Casein precipitated by acidolysis of the skimmed milk was filtered under constant pressure conditions followed by the cake drying in a vacuum tray dryer. Filtration characteristics of the cake were expressed in terms of average specific cake resistance (αav) and cake solids fraction or porosity. An interesting trend was observed at lower feed concentrations of casein in the filtration slurry. In the initial period of filtration, αav, was found to be lower under the prevailing conditions as compared to the increased value of αav in the later part of filtration. This phenomenon accentuates the strong relationship between the differential pressure across the filtration unit and αav. To corroborate these findings, the pressure drop across the cake was found to influence its compressibility, n, that varied between 0.4 and 0.8. The differential pressure studied was in the range of 0.5–2.5bar. The influence of gas deliquoring on cake saturation (S) was studied to obtain the threshold pressure, an useful design parameter for pressure filtration.The drying of casein cake in a vacuum tray dryer showed the typical drying rate curve with constant rate period followed by the extended falling rate period. The drying curve was analyzed in terms of moisture diffusivity, which in turn was found to be a strong function of temperature. To overcome long drying time of few hours, the casein cake after partial dehydration was extruded to get the benefit of higher surface area to volume ratio. The extruded pellets were effectively and efficiently dried to the final desired moisture content. It was shown that the total drying time and drying technique influence to a large extent the protein denaturation.

Evaluation of cake filtration biological reactors (CFBR) vs. membrane biological reactors (MBR) in a pilot scale plant

In this study, cake filtration biological reactors (CFBR) and membrane biological reactors (MBR) were operated in the same environment. The initial flux of CFBR was 1,517.5L/h/m2 and the start-up cake resistance was around 6kPa. The flux dropped sharply during the first run, and at the end of first run flux was 110L/h/m2. During the second run, the flux declined further from 60L/h/m2 to 48L/h/m2 and during the last 4,000hours period flux was considered to be stable. However, the pressure difference (∆p) increased steadily for about 2,300hours until it reached a stable ∆p value at the termination of the run. On the other hand, the design operation flux of the membranes was 17.5L/h/m2. The system was operated around design flux values; however, more stable operation was achieved when the flux was around 9L/h/m2. Both the cloth filter and the membrane produced effluents with suspended solid concentrations of less than 10mg/L, but the effluent quality of the cloth filter was inferior to the membrane with respect to coliform removal. Activated sludge cake, which formed on cloth media during filtration, was evaluated according to conventional cake filtration theory using the plots of V vs. t and t/V vs. V. The plot of V vs. time t for filtration of different MLSS concentrations in laboratory cell was best fitted to second order polynomial regression (R2>0.98); a linear relationship between t/V against V was not observed. Separately, an evaluation of V vs. t showed that CFBR was best fitted to second order polynomial regression (R2=0.995 for first run and R2=0.989 for second run); however, a linear relationship between the reciprocal of filtration rate (t/V) and the cumulative volume was observed individually. A transition between two linear plots was considered to be a change in the filtration characteristics. Standard blocking model plots of CFBR were fitted better (R2>0.95) than complete blocking and intermediate blocking models with the measured data for the initial period of filtration; the latter period of filtration was best fitted to the cake filtration model (R2>0.99).

Measurements of Growth Rate of Filter Cake in Vertical Single-Pass Ultrafiltration Using Hollow Fiber Membrane Module

Vertical ultrafiltration experiments of silica colloid and bovine serum albumin solution were conducted in the single-pass mode by using a hollow fiber membrane module and beneficial in measuring the time evolution of the growth rate of the filter cake during filtration. The extremely small mass flux of the concentrate enabled us to highly concentrate the feed solution on the principle of vertical ultrafiltration in which the filter cake formed on the membrane surface is exfoliated continuously. Both growth and re-entrainment rates of the filter cake formed in vertical ultrafiltration were evaluated from the experimental data of the filtration rate and the mass fraction ratio of the concentrate on the basis of the mass balance within the hollow fiber membrane module. As a consequence, it was found that the re-entrainment rate of the filter cake increased almost linearly with the filtration time in the initial period of filtration and then tended to rapidly approach a constant value. The filter cake stopped growing under this dynamically balanced condition. The variations of the average specific resistance of the filter cake with time were also determined from the time evolutions of both the filtration rate and the growth rate of the filter cake.

Fouling of ultrafiltration membrane during secondary effluent filtration

Ultrafiltration (UF) membranes have been increasingly used for the pretreatment of secondary effluent in reverse osmosis wastewater reclamation plants. The fouling behavior of UF membranes during secondary effluent filtration was studied under constant pressure conditions in a cross flow filtration unit. The rate of flux decline was found to be strongly dependent on the permeate flux level, and drastic flux decline was observed at high flux during the initial period of filtration (within the first 30 min of each fouling test). Further flux decline was insignificant and a limiting flux behavior was observed. Membranes with high initial flux (>the limiting value) approached asymptotically to an identical stable flux, i.e., the limiting flux. The stable flux was strongly affected by solution chemistry (both pH and calcium concentration), with more severe membrane fouling occurred near the isoelectrical point of the foulant (pH 6) or at greater calcium concentration. Zeta potential measurements revealed that the stable flux may be correlated to the square of zeta potential of fouled membrane samples. The current study seems to suggest that foulant-foulant electrostatic repulsion plays a significant role in membrane fouling by effluent organic matter.

Effect of membrane pore size on the particle fouling in membrane filtration

The effects of membrane pore size and operating conditions on the particle fouling in “dead-end” microfiltration are studied by use of membrane blocking models. Two track-etched membranes (Isopore® membrane) with the mean pore diameters of 0.2 and 0.4 μm, respectively, are used as the filter media to filter 0.15 μm polymethyl methacrylate (PMMA) particles. A blocking chart is established for relating the operating conditions and the particle fouling for the testing materials. The major factors affecting the blocking index, particle accumulation and filtration flux, are discussed thoroughly. The blocking index varies gradually at the initial period of filtration, keeps at ca. 0.5 for a while and then suddenly drops to zero at a critical condition. The blocking index for 0.4 μm membrane is always larger than that for 0.2 μm membrane under the same filtration pressure and filtration flux due to more severe membrane blocking. The critical point at which the filtration transforms from membrane blocking to cake filtration occurs at a lower filtration flux or a lower particle accumulation for the membrane with larger pore size or under lower filtration pressure. The normalized resistance coefficient can be correlated to a unique function of blocking index under various operating conditions and for different membranes.

Network simulation for deep bed filtration of Brownian particles—a supplement

The track of individual particles moving through a filter bed is simulated by applying the Brownian dynamics simulation method and the modified square network model. The effect of the Raleigh-type size distribution of particles and pores on the permeability reduction of porous media is also investigated. We find that the pore size distribution of porous media has a more profound effect on the reducing of the permeability ratio than that of the particle size distribution especially at the initial period of filtration. Straining is the main mechanism to reduce the permeability at the initial period of injection for the case of Raleigh distribution of the pore size, and vice versa for the case of unique size value where the direct deposition mechanism becomes dominant.

An experimental and theoretical study on the initial period of cake filtration

Experimental measurements were made of the average specific cake resistance during the initial period of cake filtration, and the theoretical calculations about the period were also performed. The “filtration-permeation method” in the filtration cell of small area was used to measure the flow rate during the initial period of filtration, which is essentially characterized by the large flux due to fast flow rate and the rapid change of flow rate within a relatively short time interval. The measured average specific cake resistances of thin cakes which represent the cakes of initial period had very large values compared to the overall average specific cake resistance. This experimental result was contrary to the conventional theory about the initial period. Applying the “unified theory on solid-liquid separation” to the initial period, the average specific cake resistances at the initial period can have the large values--more than two times greater than that of the overall value.

Fouling of membranes during microfiltration of surimi wash water: Roles of pore blocking and surface cake formation

Studies were conducted to investigate the development of membrane fouling in the microfiltration of protein solutions such as surimi wash water, containing both soluble and suspended proteinaceous solids. A laboratory scale plate-and-frame crossflow membrane filtration unit was fabricated and installed. Continuous filtration was performed to recover the suspended myofibrillar proteins from the wash water at refrigerated temperatures (12–15°C) using polysulfone membranes. The transmembrane pressure was controlled at a constant pressure of 34.5±2.2 KPa, and the feed was pumped to the filtration unit at a crossflow velocity of 0.262±0.026 m/s. It was hypothesized that the development of membrane fouling was a dynamic process of two distinctive stages. The pore blocking resistance was the dominant resistance at the initial period of filtration, and the cake resistance began to dominate following the initial pore blocking. Experimental results showed that the initial membrane fouling process could be modeled by the standard pore blocking law, and the development of fouling continued with a continuous process of cake formation. A new concentration-dependent power law cake resistance model was found. The filtration resistance of the cake layer increased with the feed concentration and was found to be one order in magnitude higher than the initial dominant pore blocking resistance.

Dead-End Flow Filtration of Solid Suspension in Polymer Fluid through an Active Kaolin Dynamic Membrane

ABSTRACT In this study an active kaolin dynamic membrane formed on a SS-316 porous support was used to investigate the microfiltration of a solid suspension in polymer fluid. It was found that a homogeneous sodium acetate suspension in polymer can be completely removed by the active kaolin dynamic membrane to yield a very clear polymer product. The filtration rate as a function of operating temperature and pressure were experimentally determined. When the operating temperature was increased, the decreasing polymer viscosity resulted in an increase of filtration rate but was counteracted by the swelling effect of the membrane layer. Similarly, when the pressure was increased, the increase in driving force for filtration was counteracted by a simultaneous increase in particle packing. The filtration behavior of this dynamic membrane system was also simulated by the general blocking model d 2 t/dV 2 = k(dt/dV)q, where q was found to have negative values. Negative q values mean that our system reached a maximum blocking rate during the initial period of filtration and then decreased gradually. When the blocking rate became small enough, the system behaved like cake filtration. A satisfactory fit between experimental data and theoretical calculations was demonstrated.

Migration and Deposition of Submicron Particles in Crossflow Microfiltration

The migration and deposition of submicron particles in laminar crossflow microfiltration is simulated by integrating the Langevin equation. The effects of operating conditions on the particle trajectories are discussed. It is found that the Brownian motion of particles plays an important role in particle migration under a smaller crossflow velocity of suspension or a smaller filtration rate. Based on the simulated trajectories of particles, the transported flux of particles arriving at the membrane surface can be estimated. The particle flux increases with an increase of filtration rate and with a decrease of particle diameter; however, the effect of crossflow velocity on the particle flux is not obvious. The forces exerted on particles are analyzed to estimate the probability of particle deposition on the membrane surface. The probability of particle deposition increases with an increase of filtration rate, with a decrease of crossflow velocity, with a decrease of particle diameter, or with an increase of zeta potential on the particle surfaces. The simulated results of packing structures of particles on the membrane surface at the initial stage of filtration show that a looser packing can be found under a larger crossflow velocity, a smaller filtration rate, or a smaller diameter of filtered particles. Crossflow micro-filtration experiments are carried out to demonstrate the reliability of the proposed theory. The deviation between the predicted and experimental data of filtration rate at the initial period of filtration is less than 10% when the Reynolds number of the suspension flow ranges from 100 to 500.

Patchy cleaning of rigid gas filters—II. Experiments and model validation

Rigid ceramic filters have emerged as the most promising technology for cleaning of hot gases due to their resistance to attack by aggressive gases and high temperatures. Their potential may be further extended by using them in combination with injection of a dry sorbent to remove acid gases and other chemical contaminants. A crucial factor in their successful utilisation is the ability of the cleaning pulse to remove the deposited cake effectively from the filter surface. In this study, laboratory experiments have been carried out using a single ceramic candle filter. The non-steady-state behaviour of the filter in the initial period of filtration was followed. ‘Patchy’ cleaning of the filter was observed and is consequently identified as the major cause of the inefficiency of filter cleaning indicated by the measured residual pressure drops. A simple correlation has been proposed to determine the cleaned fraction from pressure differences as a measure of the efficiency of cleaning. The conditioning curves have been successfully simulated using a recently developed probabilistic model and it has been shown that the model can also be applied to simulate the experimental results obtained in a pilot plant operating at high temperatures. The modelling results are consistent with the experimental observation that patchy cleaning with a thin residual dust layer in the cleaned areas can explain the conditioning behaviour.

Long-Term Filtration Behaviour of Soil-Geotextile System

Abstract In addition to soil retention and permeability requirements for a geotextile filter, a geotextile should be able to function without having a significant number of its openings clogged. In order to evaluate the long-term filtration behaviour of a soil-geotextile system, a permeameter is developed in which both the long-term flow rates as well as the gradient ratio tests can be conducted simultaneously for different soils at different densities. Uniformly graded Ottawa sand mixed with various percentages of fine-grained silty soil is used to represent the unstable soil condition for internal soil migration. Two types of needle-punched nonwoven and one type of woven geotextiles are used whose opening sizes satisfy the soil retention criteria for the soil mixtures used. An unstable flow behaviour is observed during the initial period of filtration. Fine soil particles move toward the geotextile and are either trapped within the geotextile or retained upstream, decreasing the flow rate of the soil-geotextile system. Continuous reduction in flow rate is observed which finally reaches a constant stabilized value. It is also observed that the fine soils adjacent to the geotextile pass through it and the flow rate increases. The duration of such unstable behaviour and time taken to attain a stabilized flow condition increase with fine contents. The gradient ratio also attains a stable value in the long run after the formation of a soil filter cake in the soil-geotextile system. The initial unstable period does not indicate true clogging behaviour of the system. In view of this, it is recommended that a long-term stabilized gradient ratio be determined instead of the 24-h gradient ratio.

Effect of inline filtration on the potency of drugs administered intravenously

The binding of cephalothin sodium, phenobarbital sodium, dexamethasone sodium phosphate, isoproterenol hydrochloride, digoxin, dactinomycin and phenytoin sodium to three intravenous inline filters--a 5-micrometer stainless steel depth filter, a 0.2-micrometer cellulose ester membrane and a 0.2-micrometer polycarbonate membrane--was studied. The experiments were conducted simulating inline i.v. filtration of these drugs in three i.v. solutions--lactated Ringer's, 5% dextrose and normal saline. Cephalothin was assayed colorimetrically, phenobarbital and phenytoin spectrophotometrically, and the other drugs by radiotracer technique. Binding of the drugs to the filter was found to be insignificant from a therapeutic standpoint, except for dactinomycin which bound to the 0.2-uicrometer filters. The binding of dactinomycin was approximately 13% of the amount administered through cellulose ester and polycarbonate membranes, respectively. The binding occurred during the initial period of filtration of the drug solution. It was concluded that inline filtration of drugs administered in relatively high does should not present any problem concerning the reduction of the therapeutic potency because of filtration.

Particle deposition in fibrous media with dendrite-like pattern: A preliminary model

A preliminary model of the formation of chain-like particle agglomerates on fibers during filtration of aerosols in fibrous media is proposed. Such dendrite-like growth has been observed experimentally and occurred beyond the initial period of filtration. Unlike most of the previous studies which are confined to clear filters, the present work is intended for the description of filtration performance (both filtration efficiency and pressure drop) over the entire loading period.