Formation Of Dynamic Membrane Research Articles

Formation of dynamic membranes for oily water separation by crossflow filtration

This paper presents the results of a study of dynamic membranes for oily water separation. The formation of the dynamic membranes and their performance in oily water separation were investigated. The results showed that a magnesium hydroxide suspension, made by a simple reaction process with salt and sodium hydroxide, was a suitable material for dynamic membrane formation on the ceramic support. A dynamic membrane with a uniform and compact structure was formed using the crossflow filtration technique. The formed dynamic membranes were reproducible and gave good performance in separating oil from water using a model oil emulsion in an alkaline environment. The oil retention was more than 98% and the permeate flux was more than 100 L m −2 h −1 at 100 kPa in a batch concentration test.

Membrane bioreactors for water reclamation

Singapore has been using dual membrane technology (MF/UF RO) to produce high-grade water (NEWater) from secondary treated sewage. Membrane bioreactor (MBR) has very high potential and will lead to the further improvement of the productivity and quality of high-grade water. This study was focused on the technical feasibility of MBR system for water reclamation in Singapore, making a comparison between various membrane systems available and to get operational experience in terms of membrane cleaning and other issues. Three MBR plants were built at Bedok Water Reclamation Plant with a design flow of 300 m3/day each. They were commissioned in March 2003. Three different types of submerged membranes were tested. They are Membrane A, plate sheet membrane with pore size of 0.4 microm; Membrane B, hollow fibre membrane with pore size of 0.4 microm; and Membrane C, hollow fibre membrane with pore size of 0.035 microm. The permeate quality of all the three MBR Systems were found equivalent to or better than that of the conventional tertiary treatment by ultrafiltration. MBR permeate TOC was about 2 mg/l lower than UF permeate TOC. GC-MS, GC-ECD and HPLC scan results show that trace organic contaminants in MBR permeate and UF permeate were in the same range. MBR power consumption can be less than 1 kwh/m3. Gel layer or dynamic membrane generated on the submerged membrane surface played an important role for the lower MBR permeate TOC than the supernatant TOC in the membrane tank. Intensive chemical cleaning can temporarily remove this layer. During normal operation conditions, the formation of dynamic membrane may need one day to obtain the steady low TOC levels in MBR permeate.

Parameters affecting the properties of dynamic membranes formed by Zr hydroxide colloids

Membranes exhibiting separation properties similar to those of conventional UF membranes can be dynamically prepared on a porous substrate by in-situ deposition of particles from a circulating colloidal suspension of certain hydrous oxides, notably Zr hydroxide. The objective of this work is to elucidate parameters affecting Zr hydroxide colloidal properties in order to determine optimal conditions for dynamic membrane (DM) formation on substrates of different porosity. The main parameters studied were the effects of pH, Zr concentration and ionic strength on colloidal properties. The main results were as follows. Over the pH range from 3 to 4, a stable mono-disperse colloidal dispersion was obtained, composed of sub-micron particles. The particle size increased systematically with the pH level of the suspension. At a pH value of near 4, a bi-modal distribution was obtained, composed of both sub-micron particles and very large agglomerates. At pH values higher than about 4.5, large agglomerates formed instantly. Particle agglomeration was also affected by the ionic strength of the solution and the procedure of the colloid preparation. The above results find full support in the flow and separation properties displayed by DM prepared with Zr suspensions of different chemistries (notably the pH level) on two types of substrate: a porous fabric having a pore size of tens of microns and a microfiltration membrane having a pore size of a few microns.

Dynamic formation of a self-rejecting membrane by nanofiltration of a high-formula-weight dye

Nanofiltration of a high-formula-weight dye, Cibacron F3G-A, FW=840, was investigated at low-to intermediate-crossflow velocity and pressure using a zirconium hydrous oxide-poly(acrylic acid) membrane. A sharp minimum in the rejection vs. pressure curves obtained at constant crossflow velocities provided a distinct determination of the pressure required to initiate dynamic membrane formation. These minima occurred at a pressure near 2MPa and only at a crossflow velocity less than 0.4 m/s, indicating that a critical pressure is required for membrane formation while also implying that the concentration at the membrane surface is involved in the modification.

Formation of dextran-Zr dynamic membrane and study on concentration of protein hemoglobin solution

In this work, a dynamic membrane composed of Zr colloids was formed on a SS-316 porous tube to investigate its feasibility of concentrating hemoglobin (Hb) solutions. The very fine and uniform Zr colloids was prepared with the [SO −2 4]/[ZrO +2] ratio below 0.5 and at 50°C for 24 h. If the sulfate to zirconium ion ratio was above 0.5, the agglomeration would increase the Zr particle size to make it less desirable for separation applications. Depending upon the size of Zr colloids and the layer thickness, the dynamic membranes can offer a dextran MWCO from 300 kDa to 3 MDa. The rejection to Hb from those Zr dynamic membranes was at best 70%. However, if the Zr dynamic membrane was modified by a small amount of 2 MDa dextran, the resulting composite membrane exhibited 100% rejection to Hb molecules. Our results indicated that the adsorption of Hb on Zr colloids helped with its rejection by the dynamic membrane, but only at the expense of flux reduction. The presence of NaCl salt in Hb solution showed detrimental effects toward Hb rejection, which was probably caused by desorption of Hb from Zr colloids. However, at the isoelectric point of Hb (pH=6.8), the detrimental effect due to NaCl was not observed. Brief comparison between this composite dynamic membrane and commercial polymeric membranes, YM-30 and PM-30 from Amicon, showed that our dynamic membrane had better rejection to Hb (100% versus 99.5% and 98.5%), but lower fluxes (7.7 versus 16.6 and 14.6 l/m 2 h).

Use of Chemical Species as Dynamic Membranes with Crossflow Microfiltration

ABSTRACT The feasibility of utilizing the phenomenon of dynamic membrane formation with crossflow microfiltration in treating domestic wastewater was investigated. The primary membrane, used throughout the investigation, was made of woven polyester. Different chemical species, such as CaCO3, FeCl3, and NaAIO2, were used in forming dynamic membranes on top of the primary membrane. Secondary effluent from a domestic activated sludge wastewater treatment plant was treated. A calcium carbonate dynamic membrane produced a stabilized permeate flux of 90 L/m2·h, with a permeate turbidity of 0.21 Nephelometric Turbidity Unit (NTU), at optimum conditions. Ferric chloride produced optimum results when it was mixed with tap water. A permeate flux and turbidity of 70 L/m2-h and 0.16 NTU, respectively, were obtained. Sodium aluminate produced a stabilized permeate flux of 77 L/m2·h when it was mixed with tap water during the formation of the dynamic membrane. The permeate turbidity was 0.16 NTU. The fouling mechanism of the three dynamic membranes was investigated, and empirical models were produced.

Crossflow microfiltration with dynamic membranes

The effect of dynamic membrane formation on the performance of crossflow microfiltration in treating domestic wastewater was investigated. The dynamic membrane was formed on top of a woven polyester primary membrane by circulating a precipitate of MnO 2. The dynamic membrane formation was investigated at different conditions of pH values and KMnO 4 concentrations. Secondary effluent from a domestic wastewater treatment plant was used thoughout the study. The MnO 2 dynamic membrane was found to improve the performance of crossflow microfiltration by producing higher flux values, extending the running time, and increasing rejection of solids. The optimum permeate flux values were obtained when the dynamic membrane was formed at a pH of 10 and using 100 mg/l of KMnO 4. At optimum conditions, the permeate turbidity was found to stabilize at values of less than 0.2 NTU. Cleaning the membrane was achieved easily and efficiently by brushing the outside surface of the primary membrane.

Self-forming dynamic membrane for ultrafiltration of pineapple juice

The formation of self-forming dynamic membrane on a porous ceramic support was studied. Pineapple juice of 12° Brix concentration was used in the experiments which were carried out at 25°C by circulating the pineapple juice at the applied pressure of 100, 200, and 300 kPa and at cross-flow velocities of 1.30–2.95 m s −1 through the ceramic membrane module for 1 h. The experimental data of flux and rejections showed that the dynamic membrane was well-formed after 30 min of circulation under the applied pressure of 300 kPa and at a cross-flow velocity of 2.0 m s −1 in which the steady values of flux and rejections of macromolecules and sugars obtained from the filtration mode were 6.0×10 −3 m 3/m 2 h, 84–87% and 6%, respectively. The corresponding values for ultrafiltration by alumina membrane of MW cut-off 50,000, using equivalent conditions, were 15.8×10 −3 m 3/m 2 h, 91% and 10.5%. Ultrafiltration was found to be more promising. The stability of the self-forming dynamic membrane was acceptable when subjected to change of filtration conditions. The permeation flux increased with cross-flow velocity and decreased when the applied pressure was reduced. The resistances for filtration by dynamic membrane and by ultrafiltration were calculated. For a porous support of large pore sizes, an in-pore blockage of solutes which were smaller than the membrane pores reduced the pore volume and induced fouling. Internal fouling resistance ( R f) was, therefore significant and responsible for the values of flux and rejection and was approximately 70% of total resistance. While in ultrafiltration, in which membrane with a smaller pore diameter was used, R f was only 20% but R p, the polarized layer resistance, was as high as 60% of total resistance.

Use of MnO 2 as a dynamic membrane with crossflow microfiltration: Slow membraning technique

During crossflow microfiltration, dynamic membranes are formed from particles contained in the feed to be filtered. The formation of dynamic membrane was designed from MnO 2 particles by circulating KMnO 4 for 24 h. The primary membrane used in the investigation was made of a polyester woven fabric. Secondary effluent was used throughout the research. The dynamic membrane was formed under different conditions of pH and KMnO 4 concentration. The investigation showed encouraging results in terms of higher flux values and improved permeate quality. The optimum results were obtained when the dynamic membrane was formed at pH 5 and KMnO 4 concentration of 50 mg/l. A permeate flux of more than 40 l/m 2.h was obtained after more than 50 h of running time, a value which was reached within 3–4 h without using the dynamic membrane. At optimum conditions, the permeate quality was seen to stabilize at about 0.2 NTU. Cleaning the membrane was found to be easy and efficient when using brushes on the outside surface of the membrane.

Ultrafiltration separation of aqueous solutions of poly(ethylene glycol)s on the dynamic membrane formed by gelatin

Ultrafiltration separation of aqueous solutions of poly(ethylene glycol)s with molecular weights ranging from 300 to 40 000 Da on the dynamic membrane formed by gelatin have been experimentally studied and theoretically discussed. The membrane was formed on the cellulose acetate ultrafiltration membrane UAM-500 (Vladipor). Dynamic membrane formation causes an essential growing of the retention coefficient of solutes studied. This effect depends on the working pressure and the molecular weight of the compounds being separated. It rises as values of both parameters increase permitting the change of the efficiency of the ultrafiltration separation over sufficiently wide ranges.

Formation of dynamic membranes with crossflow microfiltration

A dynamic membrane is one of the anti-fouling techniques in the literature, but it has not been thoroughly studied. The formation of a MnO 2 dynamic membrane on top of a polyester primary membrane was investigated. The MnO 2 used in the investigation was precipitated from potassium permanganate (KMnO 4) solution using sodium formate (HCOONa). The study showed that formation of dynamic membranes with particle sizes less than the pore size of the primary membrane was found to proceed according to the standard law of filtration, in the first stages of membrane formation. Later, as particles start to bridge the pores and precipitate on the membrane surface, dynamic membrane formation was found to obey the cake filtration model. General equations to describe each case of dynamic membrane formation were obtained. When wastewater was treated using the dynamic membrane, particles were found to precipitate in the same pattern as in the case of dynamic membrane formation.

Formation and characteristics of dynamic membrane for ultrafiltration of protein in binary protein system

A self-rejecting type dynamic membrane was formed by filtering a protein solution through a porous ceramic tube. In the dynamic membrane formed by a single protein (ovalbumin) solution, the effects of the linear velocity along the membrane, the bulk protein concentration, the pressure and pH in the feed solution on both membrane formation and ultrafiltration performance were studied in detail. The apparent protein rejection reached more than 0.8 in all cases. The volume fluxes at steady state increased with increasing linear velocity and with decreasing bulk concentration, but the pressure hardly influenced the flux. At the isoelectric pH, the flux in the early stages was the largest and that in the steady state of membrane formation was smallest. This behaviour is characteristic of a partially permeable dynamic membrane. The formation of a dynamic membrane was further attempted by using a binary mixture of proteins (ovalbumin and γ-globulin) in solution. It is suggested that the formation of deposits occurred independently for the two proteins. The flux at steady state in the binary protein system agreed with the smaller flux in the corresponding single protein system.

Formation of dynamic UF membrane with fine Zr particles

The performance of dynamic ultrafiltration (UF) membranes depends on the characteristics of the metal colloids exploited as membrane materials. Due to the lack of chemical and physical stability of metal colloids, membrane performance is not always constant. However, the authors have succeeded in preparing stable inorganic dynamic UF membranes on a porous ceramic support with fine Zr particles prepared by Matijevic's method. The deposited layer of the particles produced an excellent rejection ability as an UF membrane. The membranes were superior in stability to conventional metal colloid membranes. The performance of this membrane can be controlled by the Zr particle size and the thickness of the deposited layer.

An investigation of the mechanism of dynamic membrane formation on rotating microporous surfaces

The mechanism of formation of Zr(IV) hydrous oxide membranes deposited under dynamic conditions on a microporous support has been examined. The support was rotated at speeds of 1200 to 2000 rpm, corresponding to a linear velocity of 2.7 to 4.6 m/sec. A small pressure difference across the porous barrier of 0.5 to 0.7 atm was maintained. Observations of the decline of transmembrane flux are consistent with a two-step model of membrane formation comprising: (a) a period of pore clogging and bridging, followed by (b) the thickening of a colloidal cake. Increasing the rate of rotation significantly decreased the filtration constant for both the pore clogging and the cake formation periods, but the time required to begin cake growth remained essentially the same. The rotation velocity also influenced the zirconium rejection, which reached a plateau of 80 to 88% at 1200 rpm and 90 to 95% at 2000 rpm. The selectivity of the dynamically formed Zr(IV) hydrous oxide membrane was tested with a Rhodamine dye solution. Rejections of 85 to 95% were obtained for over 1 hr with essentially constant filtrate flux.

The formation mechanism of dynamic hydrous Zr(IV) oxide membranes on microporous supports

A study of Zr(IV) hydrous oxide dynamic membrane formation was carried out on 0.1 μm Millipore filters in unstirred batch cells. Observations of flux decline during the filtration process are consistent with a mathematical model involving two stages of membrane growth: (a) a “pore-clogging” step, in which the pores are clogged and eventually bridged by colloidal particles; and (b) the growth of a conventional filter cake. The model also correctly predicts the effects of feed concentration, and pressure on the Zr content and the hydrodynamic resistance of the dynamic membrane. These predictions are also shown to be applicable to dynamic membranes formed in cross flow filtration. Some evidence suggests that irreproducibility of the steady-state flux in crossflow filtration systems may be attributed to the presence of pump oil in trace amounts (<20 ppm).

Dynamic deposition of polyacids on porous membrane supports

AbstractPoly(acrylic acids), poly(styrenesulfonic acid), and their block and random copolymers were tested for their ability to form dynamic membranes on partially cured asymmetric cellulose acetate. Chemically modified porous polypropylene (Celgard) was also used as a support for poly(acrylic acid). Salt rejections, water fluxes, and streaming potentials of membranes were tested under hyperfiltration conditions. Sorption of the polyelectrolytes by the cellulose acetate supports was studied using spectrophotometric, 22Na tracer, and electron microscopy techniques. The dynamic membrane formation was noted only for poly(acrylic acid) and for its 1:4:1 block copolymer with poly(styrenesulfonic acid). The uneffectiveness of other polyelectrolytes was discussed in terms of a negative zeta potential of cellulose acetate. The increase in salt rejection (R) due to the polyelectrolyte is strongly dependent on the initial Ri of the support. Sharp maxima in the ΔR ‐versus‐Ri curves have been noted for Ri in the range of 40–55%. The most significant improvement in the hyperfiltration characteristics of cellulose acetate was attained with the 1:4:1 block copolymer. Flux of 17 gfd at 350 psi and R = 93% was obtained in short‐term tests for a 0.1N feed solution. Long‐term tests did not reveal any flux or salt rejection decline for membranes in which poly(acrylic acid) was complexed with phosphoramidic groups grafted onto Celgard.