Single Nanofiltration Research Articles

Hybrid multi-chamber system for enhanced removal of boron during water treatment/desalination without chemical usage

The efficiency of boron removal during water treatment and desalination is insufficient to meet the water quality standard set by the World Health Organization. Boron can exist in its two natural forms in water, depending on the water pH. At low and high pH, boron tends to exist as boric acid and borate ion, which are neutral and negatively charged, respectively. In this study, we developed a multi-chamber boron control system. For the control group (nanofiltration (NF) and reverse osmosis (RO)), the pH was controlled to enhance the boron removal efficiency. As expected, the boron removal efficiency at pH 11 (RO: 95–98%, NF: 91%) was higher than that at pH 8 (RO: 85–87%, NF: 80%). Additionally, activated carbon was introduced in the developed system (forward osmosis-NF/RO hybrid processes within the draw-solution chamber). Powdered activated carbon was more effective than granulated activated carbon in adsorbing boron molecules, with the maximum adsorption capacities (qm) of the former and latter being 2.33 and 1.33 mg g−1, respectively. A simple economic analysis was performed to determine the applicability of the multi-chamber boron control system for the economical production of high-quality water with minimal boron concentration. The unit cost of the multi-chamber boron control system (0.26 USD ton−1) was significantly lower than that associated with single NF and RO processes (0.64–1.62 USD ton−1). The multi-chamber boron control system thus represents an economical alternative that can achieve a higher product water quality compared with existing processes.

A field study of desalination of high-salinity surface brackish water via an RO-NF hybrid system

In this work, a pilot-scale membrane desalination demonstration system for producing drinking water from surface brackish water with a total dissolved solids (TDS) of ≤ 10 g/L was designed and constructed. The production capacity and water recovery were at least 1 m3/h and 50%, respectively. Brackish water was pretreated with microflocculation – precision filtration – ultrafiltration, and the quality of the produced water met the feed water requirements of nanofiltration (NF)/reverse osmosis (RO). The performances of various design configurations, including single NF, single RO, and NF-RO hybrid systems, were evaluated and compared in terms of water recovery and product water TDS. The hybrid RO-NF systems were found to be more effective than either NF or RO alone, and the 4RO-4NF design configuration was most favorable. At a TDS of 10 g/L in the feed, the processing capacity and product water TDS of the 4RO-4NF hybrid system was able to meet the design specs at a feed flow rate of 2 m3/h and an operating pressure of 1.5 MPa. Its long-term stability was examined as well, and it was demonstrated that the desalination performance of the system was mainly affected by seasonal temperature changes and rainfall. The total process cost was estimated to be $1.56/m3, which is considered economically suitable for remote areas where safe drinking water cannot be provided centrally.

Strong improvement of nanofiltration performance on micropollutant removal and reduction of membrane fouling by hydrolyzed-aluminum nanoparticles

Increasing attention has been focused on the removal of micropollutants from contaminated drinking source water. However, low rejection efficiency and membrane fouling still inhibit further application of nanofiltration membrane in this field. Interesting results were found that the residual hydrolyzed-aluminum nanoparticles from supernatant after coagulation and sedimentation strongly improved the nanofiltration performance for micropollutant removal. A simulated raw water containing humic acids, micropollutants and kaolinite clay was employed to investigate the factors of water matrix affecting the nanoparticle-enhanced nanofiltration for micropollutant removal. Results of experiments showed that these hydrolyzed-aluminum nanoparticles easily induced the aggregation of bisphenol-A (BPA) and humic acids in the supernatant. The enhancement of BPA removal was mainly attributed to the repelling interaction between the Al-BPA-DOC complexity and in situ-modified membrane surface during nanofiltration.‘This in situ surface modification by the hydrolyzed-aluminum nanoparticles improved membrane hydrophilicity, roughness and positively-charging capacity. For the treatment of River Songhua water spiked with micropollutant, the percentage removal of BPA was improved to be 88.5%, much more than the case of single nanofiltration without coagulation (60.7%). Meanwhile, the membrane fouling was reduced by 2.13 times than the case of single nanofiltration without the dynamically deposited-layer of nanoparticles. This in situ modification of nanofiltration membrane by hydrolyzed-aluminum nanoparticles achieved excellent removal efficiency for micropollutants from River Songhua water background.

Desalination and removal of pesticides from surface water in Mekong Delta by coupling electrodialysis and nanofiltration.

The shortage of drinking water is a major problem in the rural areas of the Mekong Delta, especially, when surface water, a main local direct drinking water source is being threatened by pesticide pollution and salinity intrusion. A hybrid process coupling electrodialysis (ED) and nanofiltration (NF) is proposed as an effective process easy to setup in a small plant to treat complex matrix with high salinity and pesticide concentration as is the Mekong Delta surface water. Performance of the ED-NF integration was evaluated with synthetic solutions based on the comparison with a single NF step generally used for pesticide removal. Both energy consumption and water product quality were considered to assess process efficiency. The ED stage was designed to ensure a 50% removal of salinity before applying NF. As expected, the NF rejection is better in the hybrid process than in a case of a single NF step, especially for pesticide rejection. The integration of a NF stage operated with NF270 membrane consumes less energy than that with NF90 membrane but its efficiency was observed not high enough to respect the Vietnamese guidelines. Using NF90, the optimal recovery rate of the NF stage varies from 30 to 50% depending on the salt content in the feed.

Rapeseed hull oil as a source for phytosterols and their separation by organic solvent nanofiltration

Hulling of rapeseed leads to hulls as an additional valuable by‐product of rapeseed oil production, containing remarkable concentrations of phytosterols and tocopherols compared to the concentrations in the entire seed and double‐pressed kernel. It was shown that the content of these substances in hull oil is significantly higher than in the entire seed. In particular, the sterol concentration with 3% is remarkably high. Another characteristic of hull oil is the high concentration of alpha‐tocopherol. Therefore, the hull oil can be an interesting source for sterols and tocopherols, which are used as health promoting additives in food industry. The nanofiltration with organic solvent resistant membranes was tested for the enrichment of these substances. The experiments were performed in a dead‐end cell at pressures up to 20 bar at ambient temperature using n‐hexane and n‐heptane as solvents. Fluxes of pure solvent and solutions of triglycerides and sterols were determined. Rejections of sterols and triglycerides were measured in binary (solvent‐sterol or solvent‐oil) and ternary (solvent‐sterol‐oil) solute solvent systems. A five‐fold increase in sterol concentration in the solvent free permeate was possible with a single nanofiltration step. After removal of the solvent from the permeate, sterol crystallization was observed. Further investigations have to be done to optimize membrane separation and coupling with a crystallization of sterols.Practical applications: The hull fraction of rapeseed can be used as a source for phytosterols and tocopherols which are both food additives. Thus, in addition to the advantages with respect to the quality of the oil and protein, hulling leads to additional valuable products, which are oil fractions rich in sterols and tocopherols. The feasibility of an organic solvent nanofiltration process for the separation between unsaponifiables and oil is demonstrated.Rapeseed hull oil has a remarkably high content of alpha‐tocopherol and sterols. It was shown that organic solvent nanofiltration is a process, which can be used for a partial separation of sterols and triglycerides. An enrichment by a factor of five is possible, leading to oils with such a high sterol concentration that crystallization occurred.

Concentration of Milk and Whey by Membrane Technologies in Alternative Cascade Modes

The aim of the present study was the evaluation of a membrane cascade comprised of ultrafiltration (UF) in series with reverse osmosis (RO) or nanofiltration (NF) in comparison to a single-stage process. It was found that the upstream UF accelerated the NF and the RO, whereby the effect was more distinct for the NF. The maximum volume reduction ratio (VRR) during skim milk and sweet whey concentration could be increased by 78 and 96%, respectively, by substituting a single NF by an UF-NF cascade. The replacement of a single RO by a UF-RO cascade during concentration of skim milk slightly increased the VRR by 3%. However, the energy demand could be reduced by approximately 16%. For the concentration of sweet whey, it was found that it is more advantageous to conduct the RO at a higher transmembrane pressure (TMP) instead of applying an UF-RO cascade.

Optimal design of nanofiltration system for surface water treatment

Both reverse osmosis (RO) and nanofiltration (NF) membranes have been increasingly used for water purification and desalination. However, the salt rejection of NF membranes is quite different from that of RO membranes, which makes a significant distinction in their process designs. This work started from the performance investigation of a single NF membrane element and then focused on the process design of the NF system for surface water treatment. In experimental tests, it was found that the observed rejection of the NF element becomes nearly constant when the concentrate flow is large enough, while the membrane flux of the NF element is quite stable regardless of the water flow across the membrane surface. These findings can be used to instruct the process design of the NF system for surface water treatment. In process design, a two-stage arrangement is sufficient for the NF system to reach the highest water recovery, while the RO system requires a three-stage arrangement.

Removal of perfluorooctane sulfonates from water by a hybrid coagulation–nanofiltration process

Perfluorooctane sulfonates (PFOS) as a persistent, bioaccumulative and toxic pollutant has been detected in surface water around the world and attracted great attention in recent years. It’s important and urgent to develop effective technology to remove PFOS. The effect of coagulation pretreatment on the removal efficiency of PFOS by nanofiltration (NF) was investigated. Two types of NF membranes were used to test the PFOS rejections in three different solutions including DI water, humic acid (HA) solution and coagulation solution. The characterizations such as HA and PFOS accumulation on the membranes, scanning electron microscopy (SEM) and surface ζ potential analysis of virgin and fouled membranes were conducted to study the separation mechanism. The experimental results showed that PFOS can be partly removed by coagulation. The NF270 membrane exhibited a higher PFOS rejection (>95%) than the HYDRACORE membrane (40–60%) in single NF process. The surface negative charge density of both NF membranes reached a minimum value after filtering coagulation effluent, which might be a result of floc adsorption on the membrane surface. Coagulation pretreatment was effective to enhance the PFOS rejection of the NF membranes. In particular, the rejection efficiency of the HYDRACORE membrane increased from 55% to 86% at a trans-membrane pressure of 0.4MPa. The SEM images showed that the fouling layer formed by flocs was loose and highly permeable.

Two-stage nanofiltration process for high-value chemical production from hydrolysates of lignocellulosic biomass through hydrothermal liquefaction

Two-stage nanofiltration (TSNF) process was proposed for recovering high-value chemicals, including monophenols, cyclopentenones, glucose and acetic acid, from hydrolysates of lignocellulosic biomass (rice straw) through hydrothermal liquefaction (HTL). The separation performances of three single nanofiltration (NF) processes and three TSNF processes were studied. Results showed that at the first stage of (DL+DK) TSNF process, DL membrane had high glucose rejection of 97.12%, and lower acetic acid rejection as well as lower aromatics rejections than DK membrane. At the second stage, DK membrane had rather low acetic acid rejection of 5.04% to ensure acid separation from aromatics. Unlike glucose or acetic acid, aromatics were unable to be recovered into one fraction due to scattered rejections of different aromatic compounds on NF membrane. The (DL+DK) TSNF process was proved to be a feasible way to fractionate hydrolysates into three parts: glucose concentrate, monophenols and cyclopentenones concentrate, and acetic acid permeate.

Improvement of overall water recovery by increasing RNF with recirculation in a NF–RO integrated membrane process for seawater desalination

This study explored the effect of increasing nanofiltration (NF) permeate recovery (RNF) by recirculation and dosage of chemicals on NF separation performance, concentration polarization (CP) and accumulation of scaling-prone ion concentration on NF membrane surface. The related coupling effect on seawater reverse osmosis (SWRO) desalination was also evaluated. The results show that when RNF for a single NF membrane element increased from 40% to 65%, pH of NF retentate and recycling ratio (Rr) increased from 5.02 and 0.75 to 5.10 and 3.58, respectively, with the dosage of both acid and antiscalant. In the same RNF range, the rejection of Ca2+, CO32−, SO42−, TDS and total hardness decreased from 35.1%, 78%, 96.3%, 18.2% and 56.1% to 28%, 76.9%, 96.1%, 10.9% and 33.9%, respectively, for NF in without chemical dosage. The CO32− and SO42− ion concentrations both in SWRO brine and on SWRO membrane surface were only around 0.12, 98, 0.12 and 99mg·L−1, respectively, which were much lower than those in typical SWRO brine. Higher rejection for SO42− ion and higher selectivity of divalent ions over monovalent ions by the NF element were demonstrated in this study, which indicates that ESNA3 membrane yields a relatively higher efficiency of seawater pretreatment.

Discussion on calculation of maximum water recovery in nanofiltration system

In desalination process, the effect of concentration factor (CF) and concentration polarization degree (CP degree) on water recovery in nanofiltration (NF) system displays a significant difference from that in reverse osmosis (RO) system due to their different salt rejections. In this work, the relationship between CF, salt rejection and water recovery was firstly discussed in a single NF element. In order to keep a stable performance, a spiral NF2 membrane was taken as an example, and water recovery for the single NF2 element should be about 25% instead of 15%, which is generally considered suitable for the single RO element. Then, the CP degree was calculated according to the theoretical model and experimental test, and the ultimate value was pointed out to be below 1.2. Thus, when calculating the scaling tendency of the sparingly soluble salts, the salt concentration near the membrane surface was recommended to be 1.15 times as high as that in bulk solution. Finally, a calculation method for maximum recovery was proposed for NF system. Because of the varied rejection with different NF membranes for the same feed water, which is significantly different from that in RO system, the pre-test is very necessary for NF system design.

Multilayered PEC nanofiltration membranes based on SPEEK/PDDA for anion separation

Layer-by-layer (LBL) deposition of polyelectrolytes (PEs) on porous substrates has emerged as a versatile and inexpensive method to form high flux nanofiltration (NF) membranes. PE complexes (PECs) were prepared by alternative deposition of poly(diallyldimethylammonium) chloride (PDDA) and sulfonated poly(ether ether ketone) (SPEEK) on hydrolyzed polyacrylonitrile (PAN-H) supports. PDDA/SPEEK multilayers consisting of 5, 10, 15 and 20 bilayers were deposited from solutions with different NaCl concentrations. Film thickness was studied by AFM. Thicker films were formed with increasing the bilayer number and the salt concentration. For the first time, multilayered PDDA/SPEEK PEC membranes were applied as NF membrane in an anion separation application. The NF performance of PDDA/SPEEK membranes was studied for single salt solutions (chloride, sulfate and phosphate) as well as mixed-salt (binary and ternary). Membranes showed different retentions and fluxes for anions in single and mixed-salt solutions. When using mixed-salt solutions, the retentions were found to be higher compared to the single salt NF experiment results. HPO 4 2− ions showed a remarkable influence on the performance of the PEC membranes via formation of intermolecular bonds within the polymer chains. In the best case, membranes showed Cl −/SO 4 2− and Cl −/HPO 4 2− selectivities of 20 in the separation of mixed-salt solutions.

Batchwise and Continuous Organophilic Nanofiltration of Grubbs‐Type Olefin Metathesis Catalysts

A mass-tagged N-mesityl imidazolinium salt with four additional -CH(2)NCy(2) substituents was synthesized, leading to a molecular mass of nearly 1100 g mol(-1) in the corresponding carbene ligand. This mass-tagged ligand was used to generate the respective Grubbs II and Grubbs-Hoveyda type complexes. The catalytic activity of the latter complex was tested in several olefin metathesis reactions and found to be slightly superior to that of the related N-mesityl based complex. In batchwise solvent resistant nanofiltration experiments the ruthenium complex dissolved in toluene and following a metathesis reactions was efficiently retained (>99.8 %) by a single nanofiltration; the permeate contained less than 4 ppm of Ru. Equally efficient catalyst retention was observed in a membrane reactor utilized for the continuous synthesis of a RCM product.

Modeling the permeate transient response to perturbations from steady state in a nanofiltration process

Knowledge of the transient response to perturbations from steady state is of importance in treatment processes in order to predict the time required to achieve the maximum effect of any feed water concentration change. Experimentation was conducted to evaluate the membrane permeate transient response for single-stage and three-stage nanofiltration membrane systems when each system's feed water experienced a perturbation in water quality with a sudden increase in chloride. The transient response of a single nanofiltration membrane was described by a first order system with a time constant of 3.8 min. The transient response to a permeate concentration change in a nanofiltration membrane process was 99% realized in 7 min. Mathematical models were developed that accurately described permeate chloride concentrations as a function of time after a step-change occurred in feed water chloride concentration levels feeding the systems. Results indicate that first order modeling is satisfactory for the description of...

Reclamation of acid dye bath wastewater: Effect of pH on nanofiltration performance

This paper describes the effect of pH on the selection of the best treatment scheme for the reclamation of acid dye bath wastewaters (ADBW) of carpet manufacturing industry. ADBW, having acidic pH (5.0–5.9) originally, was subjected to pre-filtration through 1.0 μm microfiltration (MF) media prior to nanofiltration (NF). Three treatment alternatives were adopted: (1) single NF without pH neutralization; (2) sequential NF without pH neutralization; (3) single NF with pH neutralization. Color was removed completely and turbidity was rejected by 79–92% in all alternatives. However, COD, total solids, total hardness and conductivity were partially removed in single NF without pH neutralization. The inadequate rejection of pollutants lead to the implementation of sequential NF consisting of three stages, where COD rejection increased from 55%–77% to 94% at the end of the third stage. Finally, the wastewater pH was neutralized by adjusting to 7.0, which provided COD rejection efficiency as high as 97% in single NF. The flux decline in single NF at acidic pH was 14–23%. In sequential NF, the flux decline decreased from 21% to 9% in three stages. The flux declines increased up to 25% due to pH neutralization, however chemical cleaning was found to be effective in restoring the fluxes. Besides, the main cause of flux decline, which was fouling at acidic pH became concentration polarization at neutral pH. Therefore, the best treatment scheme for the recovery of ADBW came out to be single NF with pH neutralization preceded by 1.0 μm MF. These results revealed that pH is an important operational parameter affecting the selection of the best treatment scheme for ADBW.

Reclamation of printing effluents of a carpet manufacturing industry by membrane processes

The effluent from the print dying process of a carpet manufacturing industry was subjected to nanofiltration (NF) and ultrafiltration (UF) processes aiming at water reuse. Three alternatives were adopted; (i) NF, (ii) loose UF (MWCO 20,000 Da) followed by NF and (iii) tight UF (MWCO 1000 Da) followed by NF. The separation performances and flux decline levels were compared for the detemination of the best process. Printing effluent was chemically precipitated using alum and fed to a lab-scale plate-and-frame membrane module under a trans-membrane pressure of 1.75 × 10 5 and 5.90 × 10 5 Pa (1.75 and 5.90 bar) for UF and NF processes, respectively. All the alternatives were tested in concentration mode of filtration where a volume reduction factor of up to 28.5 was achieved. The NF process provided very high separation performance for COD, color, turbidity, total solids and total hardness. The flux decline started at 16% and reached up to 31% at a volume reduction factor of 11.8. The loose and tight UF processes applied before NF to reduce this flux decline were not effective with a very severe flux decline of about 60% in both processes. The comparison resulted in a recommendation with single stage NF with adequate effluent characteristics.

Nanofiltration and reverse osmosis of model process waters fromthe dairy industry to produce water for reuse

In dairy plants the process waters generated during the starting, equilibrating, interrupting and rinsing steps contribute to the production of effluents. They correspond to milk products (milk, whey, cream) diluted with water without chemicals. The treatment of these dairy process waters by nanofiltration (NF) or reverse osmosis (RO) operations was proposed to concentrate dairy matter and to produce purified water for reuse in the dairy plant. The study reports one-stage and two-stage (NF + RO and RO + RO) spiral-wound membrane treatments with five model process waters representative of the main composition variations observed in dairies. Performances (permeate flux, milk components rejection, purified water characteristics) of the different operations were compared. Discussion was focused on the comparison between quality of produced waters and vapour condensates (from product drying and evaporation processes) reused in dairy plants. Accordingly, both total organic carbon (TOC) and conductivity of water treated by a single RO or NF + RO operations were convenient for reuse as heating, cooling, cleaning and boiler feed water. With the two-stage RO + RO process, a more purified water complying with the TOC drinking water limit was achieved.

Removal of Dissolved Organic Matter by Nanofiltration

A nanofiltration module is studied to determine its ability to remove dissolved organic matter (DOM) from two water sources: An effluent-impacted ground water and a low-turbidity surface water. Both feed and product waters from a single nanofiltration module are analyzed for trihalomethane precursors, as measured by bulk water dissolved organic carbon (DOC), UV absorbance, and trihalomethane formation potential (THMFP), as well as molecular weight (MW) distributions of these parameters. Results indicated excellent rejection of DOM accompanied by good rejection of inorganics, as reflected by conductivity data. Nanofiltration is very effective in rejecting higher molecular weight DOM (>1,000 amu) while moderately effective in rejecting lower molecular weight material (<1,000 amu).