• Membrane Journal
• /
• v.18 no.1
• /
• pp.7-25
• /
• 2008

Sodium dodecyl sulfate, which was anionic surfactant, at a concentration higher than its critical micellar concentration was added to calcium solution for forming micelles. Then aggregates were formed by adsorption or binding of calcium ions on the surface of micelles, and gathering between them, and then rejected by two kinds of ceramic membranes to remove calcium ions. As result, rejection rates of calcium were higher than 99.98%. And in our experimental range the higher TMP trended to increase the resistance of membrane fouling ($R_f$), total permeate volume ($V_T$), dimensionless permeate flux ($J/J_o$) and permeate flux (J) because TMP was driving force. And we investigated effects of $N_2$-back-flushing time and filtration time, that was back-flushing period, during periodic $N_2$-back-flushing on ceramic membranes. As result, optimal BTs for NCMT-623l ($0.07{\mu}m$ pore size) and NCMT-7231 membrane ($0.10{\mu}m$) were 10 sec and 15 sec, respectively. Also, optimal FT was 5 min for both membranes, and the frequent $N_2$-back-flushing could decrease membrane fouling effectively. Then, the optimal conditions resulting from our experiments for synthetic calcium solution were applied to groundwater using as washing process of soymilk package. As result, rejection rates of calcium were higher than 99.98%.

• Membrane Journal
• /
• v.24 no.3
• /
• pp.201-212
• /
• 2014

This study was performed to discover the optimum operation conditions for the advanced water treatment using the ceramic membrane, introduced the first in the nation at the Y water treatment plant (WTP). The result of investigation to find the optimum operation conditions which can continue preserving the filtration performance as well as satisfying both the economics and the water quality is as follows. In the ordinary water quality condition of the Y WTP, the optimum filtration time(the backwash period), which can minimize the production of backwash waste and preserve the membrane performance was examined to be 4.0 hours on basis of institution capacity ($16,000m^3/day$). Examining the recovery rate of TMP from the chemical cleaning (CIP) discovered that the inorganic contaminants, which cause membrane fouling, such as iron, manganese, aluminum, were removed through the acidic cleaning using citric acid, whereas the membrane recovery rate was found to be low. But, on the other hand, the TMP was recovered to the initial value from the alkali cleaning using the NaOCl. Therefore, the main contaminant causing the fouling was determined to be hydrophilic organic compound( biopolymer). The membrane recovery rate is highly influenced by the temperature of the cleaning chemical. That is, the rate increased with increasing temperature.

• Membrane Journal
• /
• v.22 no.6
• /
• pp.461-469
• /
• 2012

This study was carried out to find the optimum chemical cleaning (CIP) conditions for ceramic microfiltration membrane process of Y drinking water treatment plant. B train consists of coagulation as pretreatment process with membrane was chemically cleaned 9 times more than that of A train with ozonation and coagulation. The frequent CIP of B train was due to improper CIP method suggested by membrane manufacture as well as different membrane fouling between A and B train resulting from the different pretreatment processes. That is, recovery rate of CIP was overestimated because the rate was calculated based on normalized trans membrane pressure (TMP) rather than normalized permeability. And also, iron oxide fouling was ineffectively removed by citric acid. By using a mixture of 1% citric acid and 0.1 N sulfuric acid as reagent for acid CIP step, the recovery rate of CIP was the highest while CIP efficiency by 0.1 N sulfuric acid was the lowest. When sulfuric acid concentration increased from 0.1 N to 0.3 N in mixture, total recovery rate of CIP was not increased due to the decreased CIP efficiency in alkali CIP step by 0.3% NaOCl although its rate in acid CIP step was increased. It was proved through the experiment result of CIP sequence changes that an acid followed by alkali CIP was more effective than that of the reverse method.

• Membrane Journal
• /
• v.16 no.1
• /
• pp.31-38
• /
• 2006

The ceramic microfiltration system with periodic $N_2$-back-flushing was operated for treating paper wastewater discharged from a company making toilet papers by recycling milk or juice cartons. Two kinds of alumina membranes with 7 channels used here for recycling paper wastewater. The optimal filtration time interval for HC04 membrane with $0.4{\mu}m$ pore size was lower value of 4 min than 16 min for HC10 with $1.0{\mu}m$ pore size at fixed back-flushing time 40 sec, trans-membrane pressure $1.0kg_f/cm^2$ and back-flushing pressure $5.0kg_f/cm^2$. From the results of TMP effect at fixed filtration time interval and back-flushing time, the lower TMP was better on membrane fouling because high TMP could make easily membrane cake and fouling inside membrane structure. However, we could acquire the highest volume of total permeate at the highest TMP for the reason that TMP was driving force in our filtration system to treat paper wastewater. Then the permeate water of low turbidity was acquired in our microfiltration system using multi channels ceramic membranes, and the treated water could be reused in paper process.

• Membrane Journal
• /
• v.11 no.4
• /
• pp.190-203
• /
• 2001

In this study the discharged wastewater from a paper plant was filtrated by 4 kinds of tubular carbon ceramic ultrafiltration membranes with periodic water-back-flushing. We could investigate effects of watch-back-flushing period, transmembrane pressure (TMP) and flow rate, and find optimal operating conditions. The back-f1ushing time (BT) was fixed at 3 sec, and fi1tration times (FT) werc changed in 15~60 scc, TMP in 1.00~2.50$kg_{f}$/$cm^2$, and the flow rates in 0.27~1.75 L/min. The optimal conditions were discussed in 7he viewpoints of dimensionless permeate flux (J/J$_{0}$), total permeate volume ($V^T$) and resistance of membrane fouling ($R^f$). Optima1 back-flushing period was BT/FT=0.20, suggesting that the frequent back-flushing should decrease membrane fouling. Optimal TMP in the viewpoint of $V^T$ was 1.00~1.55$kg_{f}$/$cm^2$, suggesting that rising TMP should increase membrane fouling and decrease permeate flux. But, rising f1ow rate should decrease membrane fouling and increase permeate flux. Then, average rejection rates of pollutants filtratedby carbon ceramic membranes were 88~98 % for turbidity, 48~72% fort $COD_{cr}$ and 37~76% for TDS.

• Membrane Journal
• /
• v.25 no.5
• /
• pp.415-421
• /
• 2015

$SrCo_{0.8}Fe_{0.1}Nb_{0.1}O_{3-{\delta}}$ oxide was synthesized by solid state reaction method. Dense ceramic membrane was prepared using as-prepared powder by pressing and sintering at $1250^{\circ}C$. XRD result of membrane showed single perovskite structure. The oxygen permeability were measured under 0.21 atm of oxygen partial pressure ($P_{O_2}$) and between 800 and $950^{\circ}C$. The oxygen permeation flux of $SrCo_{0.8}Fe_{0.1}Nb_{0.1}O_{3-{\delta}}$ membrane was increased with the increasing temperature. The maximum oxygen permeation flux was $1.839mL/min{\cdot}cm^2$ at $950^{\circ}C$. Long period permeability experiment was carried out to confirm the phase stability and $CO_2$-tolerance of membrane containing Nb in the condition of air with $CO_2$ (500 ppm) as feed stream at $900^{\circ}C$. The phase stability and $CO_2$-tolerance of $SrCo_{0.8}Fe_{0.1}Nb_{0.1}O_{3-{\delta}}$ were investigated by XRD and TG analysis. The result of $SrCo_{0.8}Fe_{0.1}Nb_{0.1}O_{3-{\delta}}$ which exposed carbon dioxide for 100 hours indicated 8wt% of $SrCO_3$. But it was known that the level of $SrCO_3$ production dose not have a significant effect on oxygen permeability.

• Membrane Journal
• /
• v.33 no.6
• /
• pp.439-446
• /
• 2023

Ceramic membranes are generally used for various industrial processes operating under extreme conditions because of its high thermal and chemical stability. However, due to the trade-off phenomenon of permeability and mechanical strength, preparation of high permeability-high strength membrane is necessary. In this study, the change in characteristics and performances of ceramic membranes was analyzed depending on the type of polymer binder and its mixing ratio. Because the solubility between solvent and polymer binder was higher in PSf (polysulfone) than in PES (polyethersulfone), the viscosity and discharge pressure of the PSf-based dope solution were higher than those of PES-based dope solution. When PSf was used as a polymer binder, ceramic membrane showed high mechanical strength and low water permeability due to the dense structure. On the other hand, in case of PES, the mechanical strength was slightly reduced and the water permeability was increased. It was confirmed that the optimum mixing ratio of the PSf and PES with high water permeability and mechanical strength was 9:1.

• Membrane Journal
• /
• v.22 no.2
• /
• pp.135-141
• /
• 2012

This study was carried out to find the optimum coagulation conditions for ceramic microfiltration process of Y water treatment plant. When pH of raw water from Y Dam was adjusted to 7, the efficiency of coagulation was the best and the optimun dosage of coagulant was 3 mg/L(as $Al_2O_3$) for turbidity of raw water less then 10 NTU in Jar test. In mini module test, the decay rate of specific flux was the lowest when PAC (poly Aluminum Chloride) was used among coagulants and pH was adjusted to 7. The decay rate of specific flux for raw water turbidity of 10~30 NTU was greatly decreased with increase of dosage of coagulant (PAC) while the rate was not significantly decreased for turbidity more than 50 NTU. In conclusion, the optimum dosage of PAC (11% as $Al_2O_3$) was 30 and 50 mg/L for raw water turbidity of less than 10 NTU and more than 50 NTU, respectively. The dosage of PAC should be increased linearly 30 to 50 mg/L depending on raw water turbidity of 10 to 50 NTU.

• Membrane Journal
• /
• v.23 no.4
• /
• pp.267-277
• /
• 2013

The effect of water back-flushing period (FT) and water back-flushing time (BT) was compared with the previous study of nitrogen back-flushing in viewpoints of resistance of membrane fouling ($R_f$), permeate flux (J), and total permeate volume ($V_T$) in hybrid process of tubular ceramic microfiltration and PES (polyethersulfone) beads loaded $TiO_2$ photocatalyst for advanced drinking water treatment. As FT decreasing, Rf decreased, but J and $V_T$ increased. Turdity treatment efficiency was the maximum at NBF (no back-flushing) and increased a little as FT decreasing in both water and nitrogen back-flushing. Organic matter treatment efficiency was the maximum at FT 4 min in water back-flushing, but increased as FT decreasing in nitrogen back-flushing. As BT increasing, Rf and resistance of reversible membrane fouling ($R_{rf}$) decreased, but J and $V_T$ increased. The turdity treatment efficiency was almost constant beyond 98% in water back-flushing, but increased as BT increasing except NBF in nitrogen. The organic matter treatment efficiency was the maximum at BT 6 sec in water back-flushing, but increased as BT increasing except NBF in nitrogen. The $V_T$ was the maximum at BT 30 and FT 2 min, and optimal condition was BT 30 sec per FT 2 min in this experimental range.

• Membrane Journal
• /
• v.27 no.3
• /
• pp.263-272
• /
• 2017

There are several different methods to characterize membrane pore size distribution, however, it is yet difficult to accurately measure pore size range of 10-50 nm. In this work, we employed gas-liquid displacement porometer (GLDP) and liquid-liquid displacement porometer (LLDP) to characterize in-house alumina hollow fiber membrane (K-100) and commercial membranes (A-100, A-20) that exhibit pore sizes between 10-100 nm. GLDP method was more suitable for measuring the maximum pore size, and the measured mean pore size of the membranes by LLDP were better correlated with water permeability and solute rejection. It was determined that LLDP is effective for measuring pore sizes between 10-50 nm; however, the method holds intrinsic disadvantages such as low precision and high sensitivity compared to that of GLDP. Nevertheless, it is expected that the recently commercialized LLDP technique can provide useful data that other methods cannot.