Salt rejection in nanofiltration for single and binary salt mixtures in view of sulphate removal

Three commercial membranes (NF70, NF90 and TFC-SR) were firstly characterized in terms of pure water flux and the rejection of uncharged (alcohols and sugars) compounds. Subsequently, the rejection of monovalent (sodium and chloride) and divalent (calcium and sulphate) ions in single (NaCl, CaCl, and Na₂SO₄) and binary (NaCI/Na₂,SO₄ CaCl₂/CaSO₄, NaCI/CaCl₂, and Na₂SO₄/CaSO₄) salt mixtures was studied. According to the pure water permeability the TFC-SR membrane is a loosely packed NF membrane (12.3 L.m ⁻².h⁻¹.bar⁻¹), while both NF70 and NF90 are tightly packed (2.6 and 3.6 Lm⁻².h⁻¹.bar-). According to the uncharged solute rejection, the MWCO_NF70 = 60, MWCO_NF90= 200 and MWCO_TFC-SR > 500. NF70 and NF90 were equally efficient in rejecting 1-2, 1-1 and 2-1 salts (>90%), while TFC-SR showed typical negatively charged surface behaviour, i.e., R (1-2) salt > R (11) salt > R (2-1). Sulphate rejection decreased in the presence of sodium chloride more significantly than in the presence of calcium chloride due to the more efficient retention of the bivalent calcium.     

Relation between salt rejection and electrokinetic properties on Shirasu porous glass (SPG) membranes with nano-order uniform pores

The salt rejection by Shirasu porous glass (SPG) membranes having nano-order uniform pores was investigated for understanding the electrokinetic mechanism resulting from the surface charge developed on the membrane when in contact with salt solutions. Due to the dissociation of the hydroxyl groups such as silanol groups on the membrane surface, the membrane was negatively charged over a pH range of 3–10 from electrophoretic measurements. Cross-flow filtration experiments showed that up to 63% of NaCl was rejected by an SPG membrane having a mean pore diameters of 33 nm in a 1 mol m⁻³ NaCl solution at pH 7 under a transmembrane pressure of 74 kPa, even though the pore diameter is much larger than the ion diameter. This is a consequence of the electrostatic repulsive interaction between the co-ions (Cl⁻ ions) and the membrane surface. At the same pH, the rejection factor of NaCl decreased with increasing salt concentration due to an increase in the ionic strength. More negative charge on the membrane surface at higher pH resulted in higher rejection factors of NaCl for a fixed salt concentration. Higher rejection factors of NaCl by SPG membranes with smaller pore sizes for a fixed concentration are due to the higher ratio of the thickness of the electric double layer (Debye length) to the pore radius. The SPG membrane showed a salt rejection sequence: Na₂SO₄, NaCl and CaCl₂ at the same pH. This is because divalent anions (SO₄²⁻) are more strongly repelled by the negatively charged membrane, while divalent cations (Ca²⁺) adsorb specifically onto the membrane surface than monovalent cations (Na⁺). The salt rejection factor increased with increasing permeate volume flux. Due to the stronger acidity of the membrane materials, SPG membranes had a higher rejection factor and a lower isoelectric point (IEP < 3) than ceramic membranes.     

Streaming effect of single electrolyte mass transfer in nanofiltration: potential application for the selective defluorination of brackish drinking waters

This work deals with electrical properties of nanosurfaces in contact with electrolyte solutions. Single halide ion solutions were studied by streaming potential (SP) measurements and observed retention (R_obs) of the F⁻, Cl⁻, and Br⁻ ions across nanofiltration (NF) membranes. The detailed understanding of an electrolyte solution mass transfer requires an intimate knowledge of the physicochemical interactions occurring between nanoporous materials and electrolyte solutions across the first-generation composite membranes called NF55, NF70 and NF90. These membranes are composed of a polysulfone mesoporous sublayer and a microporous skin layer in polyamide. In order to get a better understanding of these effects, it seems attractive to compare the mass transfer permeation of the monovalent ions F⁻, Cl⁻, and Br⁻ with the electrokinetic characterizations deduced from a properly developed SP apparatus. SP measurements is a very simple method to show the intrinsinc charges on membrane pore walls. The membrane's electrical properties are studied with SP design modeling pH, ionic strength and kind of electrolyte solutions. We have observed that the isoelectric point (IEP) of the membrane materials is both dependent on the ionic strength and on the kind of electrolyte solution. The IEP in the presence of KCl is 4.4 at 0.0001 mol/L and 5.8 at 0.001 mol/L, showing an increasing adsorption of the cation K⁺ by increasing its solution concentration. For a fixed concentration, the effect of the electrolyte solution has shown that a higher adsorption of Ca⁺⁺ occurs in comparison to K⁺ and Na⁺. But the adsorption of these electrolyte solutions is essentially reversible as observed under dilution conditions. Furthermore SP measurements were used for the first time to characterize the transmembrane pressure ranges where a convective and/or a diffusional mass transfer occurs. Such an approach was developed to correlate the R_obs of the halide ions F⁻, Cl⁻ and Br⁻ with the kind of mass transfer (diffusional and/or convective) occurring predominantly under transmembrane pressure variations. Thus the NF70 membrane shows at low pressure (under 3 bar) the order of R_obs following the hydrated ionic radius: R_obs.(F⁻)>R_obs.(Cl⁻)>R_obs.(Br⁻). For a higher pressure (> 3 bar) an inversion occurs between Cl⁻ and Br⁻, but F⁻ was not affected. These results open a new prospective area for selective defluorination of brackish drinking waters using NF membranes under low transmembrane pressure.     

Formation and Characterization of Carboxymethyl Cellulose Sodium (CMC‐Na)/Poly (vinylidene fluoroide) (PVDF) Composite Nanofiltration Membranes

Abstract

A series of carboxymethyl cellulose sodium composite nanofiltration membranes were prepared through the method of coating and cross‐linking. Effects of the preparation techniques and the operating conditions on the rejection performance of the resulting membranes were investigated, respectively. It suggested that the resulting composite NF membrane with excellent rejection performance should be prepared through a certain preparing technology. Attenuated total reflection infrared spectroscopy and atomic force microscopy were employed to characterize the resulting membrane. The rejection of this kind of negatively charged membrane to the electrolyte solutions decreased in the order of Na₂SO₄, NaCl, MgSO₄, and MgCl₂.     

Influence of Chloride,Nitrate, and Sulphate on the Removal of Fluoride Ions by Using Nanofiltration Membranes

Abstract

This study investigates the removal of fluoride from various solutions(NaF, NaCl, NaNO₃, and Na₂SO₄) by using three commercial nanofiltration membranes denoted NF70(Filmtec), DESAL5 DL(Osmonics), and MT08(PCI) under 8 bars, 293 K in batch recirculation mode. The membranes were tested for their rejection potential of fluorides in the presence of chlorides, nitrates and sulphates. To identify the mechanisms of selectivity, we successively analyzed the rejections in turn of different combinations of NaF, NaCl, NaNO₃, and Na₂SO₄ in order to simulate the behavior of a real brackish water from Senegal. Fluorides rejection efficiency ranges from 78% to 95%. The efficiency of nanofiltration membranes improved is closely linked to the nature of the solution. The high rejection level(above 98%) of the divalent sulphate ions(50 or 200 mg/L) induces a Donnan effect establishment. In brackish water conditions, for all the membranes, there was a noticeable influence of anion size and hydration energy on fluoride rejection.     

Characterization and control of foulants occurring from RO disc-tube-type, membrane treating, fluorine manufacturing, process wastewater

This study was carried out to find a way to limit fouling of disc-tube-type reverse osmosis membranes in the treatment of acid rinse wastewater from the fluorine manufacturing process as well as to pretreat the wastewater before it entered the membrane process. Experiments showed that the scale consisted of Ca²⁺, SO₄⁻² and F⁻. Complex scales were removed in a subsequent procedure where the membrane was rinsed with NaOH followed by citric acid which could consequently recover its flux up to 86%. Cleaning chemicals had to be used regularly for efficient recovery of permeate flux. Ultrasonic cleaning could also improve the recovery of permeate flux up to 83%. Calcium salts were used to remove fluoride ions. CaCl₂ removed fluoride ions up to 11% more than Ca(OH)₂ at 0.5 [Ca²⁺]/[F⁻]. At acidic pH 4-7 or alkaline pH 7 and above, residual fluoride ions increased as Ca²⁺ reacted more efficiently with Cl⁻, OH⁻ and SO₄⁻² rather than F⁻. On the other hand, fluoride ions were best removed at pH 7. Adding Ca²⁺ salt above pH 7 caused an increase of residual Ca²⁺ salt in the effluent, even if fluoride ions can be ideally removed in the form of CaF₂ at a pH over 11.     

Evaluation of a modified spray-applied interfacial polymerization method for preparation of nanofiltration membranes

Nanofiltration (NF) membranes were fabricated by using piperazine (PIP) and trimesoyl chloride (TMC) by conventional and spray-applied interfacial polymerization methods, studying the effect of the application method for both phases, the number of applied layers, and the displacement speed for the spray application. A polysulfone ultrafiltration membrane was used as support. NF membranes were characterized by different spectroscopic, microscopic, and physicochemical techniques. Rejection capacity was evaluated for sodium chloride (NaCl), sodium sulfate (Na₂SO₄), and magnesium sulfate (MgSO₄) salts; the decreasing rejection order was Na₂SO₄ > MgSO₄ > NaCl for each NF membrane. NF membrane prepared with one layer of the sprayed out TMC solution and conventional application of PIP solution exhibited the highest salt rejection (99% for 1000 ppm Na₂SO₄) and a permeated flux of 10.28 L m⁻² h⁻¹ at 0.55 MPa. The modified method is a facile-reproducible preparation methodology that reduces the consumption of time, effort, and reagents leading to a scalable manufacturing process for separation technology. © 2019 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2019 , 136, 48129.     

Novel charged chitosan composite nanofiltration membranes containing chiral mesogenic group

To improve the performance of nanofiltration (NF) membranes, a chiral mesogenic compound, a positively charged compound, and a negatively charged compound were grafted to chitosan, respectively. Series of novel composite NF membranes were prepared by over‐coating the polysulfone ultrafiltration membrane with the mixture of chitosan and modified chitosan. The chiral mesogenic compound, the positively charged compound, the negatively compound and their chitosan derivatives were characterized by infrared spectrophotometer, differential scanning calorimetry, polarized optical microscope; the structure of the membrane was characterized by scanning electron microscopy. The performance of composite NF membranes was strictly related to the novel compounds grafted to chitosan and its composition. The rejection reached the maximum of 95.7% for CaCl₂ with P_2‐7 composite NF membrane, corresponding flux was 3155 Lm^?2h^?1. The rejection reached the maximum of 93% for Na₂SO₄ with P_3‐5 composite NF membrane, corresponding flux was 3879 Lm^?2h^?1. Comparing with conventional NF membranes, the membranes were used in low pressure with high flux, especially for the separation of high‐valence ions from solution. The membranes were typical charged NF membranes. POLYM. ENG. SCI., 57:22–30, 2017. © 2016 Society of Plastics Engineers     

Studies on integration of ion exchange and nanofiltration for water desalination

This work addresses an integrated treatment to produce drinking water involving the ion exchange (IEX) and nanofiltration processes. In the first stage, the experimental procedure consisted of evaluating the IEX equilibrium and the dynamics of fixed-bed saturation/regenerations, for the system Cl^?/SO₄², using an anionic resin, Purolite A850, in the form sulphate. A mathematical model which considers axial dispersion for the liquid phase and linear driving force for intraparticle mass transfer predicted well the breakthrough curves of Cl^? and SO₄^2?. In the second stage, the selective separation of Cl^?, Na⁺ and SO₄^2? ions through a flat membrane nanofiltration TS80 was assessed.     

Dynamic polyacrylamide membranes in the reverse osmosis

Dynamic formation of Polyacrylamide membranes on a Millipore filter in the reverse osmosis was optimised concerning rejection of a standard NaCl water solution.Optimised membranes were tested under 70 bars with selected salts in single-solute water solutions.Exclusion of salts with alcali cations depend on the kind of anion and is in the order SO^2-₄ > F^- > HCO^-₃ > Cl^- > NO^-₃ > J^-.Rejection of salts with alkaline earth cations is generally lower than corresponding salts with alcali cations.Especially concentration dependence of rejection of SO^2-₄ and NO^-₃ through these membranes was tested using different cations.Remarkable is the relatively high rejection of Na₂SO₄, which is 93,6% for a 5 g/l water solution under a flux of 820 l/m²· d.Dynamic PAA membranes are not useful for sea water desalination but they could be used for treatment of water with higher SO^2-₄ content.     

Preparation and characterization of a composite nanofiltration membrane interfacially polymerized from cis,cis‐1,3,5‐triaminocyclohexane and trimesoyl chloride

cis,cis‐1,3,5‐Triaminocyclohexane (TAC) was synthesized and used to prepare composite nanofiltration (NF) membranes by interfacial polymerization with trimesoyl chloride (TMC). The surface elemental composition, morphology, and hydrophilicity of the prepared NF membranes were characterized. The separation performances were examined with various salts and polyethylene glycol (PEG400, PEG600) solutions. The effects of preparation conditions were also systematically studied. The NF membrane was negatively charged and exhibited a salt rejection in the order Na₂SO₄ (98.2%) > MgSO₄ (90.8%) > MgCl₂ (84.5%) > NaCl (54.6%). The water permeability was 1.56 L m^?2 h^?1 bar^?1, and the molecular weight cutoff was 600 Da. The TAC/TMC membrane exhibited some characteristics that were different from the ones made from common diamines such as m‐phenylenediamine: (1) the surface was smoother, without a ridge‐and‐valley structure; (2) there were two kinds of crosslinking points in the polyamide chains; (3) the active layer was formed faster (only 5 seconds was required to reach a Na₂SO₄ rejection of 98%). © 2016 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2016 , 133, 43511.     

Preparation of graphene oxide/polyamide composite nanofiltration membranes for enhancing stability and separation efficiency

Polyamide (PA) NF membranes are synthesized on a hollow fiber support by the interfacial polymerization (IP) of piperazine (PIP) and trimesoyl chloride (TMC). Then, GO is coated on the PA layer to decorate the NF membrane surface (denoted GO/PA-NF). This strategy aims to improve the hydrophilicity, chlorine resistance and separation stability of the membrane. The optimization, chemical composition, morphology, and hydrophilicity of the synthesized GO/PA-NF membrane are characterized. Results indicate that the optimized GO/PA-NF in terms of rejection rate and flux are with 0.05 wt% GO. The rejection of GO/PA-NF for Na2SO₄ and MgSO₄ is 99.4% and 96.9%, respectively. Even if the GO/PA-NF is immersed in 1000 ppm NaClO solution for 48 h, the NF membrane still maintains stable salt rejection. The developed NF membranes exhibit excellent treatment performance on dying wastewater. The permeate flux and rejection of GO/PA-NF toward Congo red solution are determined to be 44.2 L/m²h and 100%, respectively. Compared with the PA membrane, GO/PA-NF presents a higher rejection for Na₂SO₄ (99.4%) and a lower rejection for NaCl (less than 20%), which shows that the NF membranes have a better divalent/monovalent salt separation performance. This study highlights the superior performance of GO/PA-NF and shows its high potential for application in wastewater treatment.     

Novel thin-film composite nanofiltration membranes fabricated via the incorporation of ssDNA for highly efficient desalination

In this work, the biomacromolecule, single-stranded deoxyribonucleic acid (ssDNA) was innovatively incorporated into the polyamide layer to tailor the permeate flux and antifouling performance of the nanofiltration (NF) membranes. With active amines groups, the ssDNA was as the aqueous phase monomers along with piperazine (PIP), and reacted with trimesoyl chloride on polyethersulfone substrate to fabricate thin-film composite (TFC) NF membranes. The NF membrane prepared under optimal ratio of ssDNA/PIP had a pure water permeability of 75.8 L m⁻² h⁻¹ (improved 58% compared to PIP NF membrane) and Na₂SO₄ rejection of 98.0% at 6.0 bar. The rejections for different inorganic salts were the order: Na₂SO₄ (98.0%) > MgSO₄ (89.2%) > MgCl₂ (72.8%) > NaCl (23.0%). Furthermore, the TFC NF membranes showed good antifouling performance in long-term running with 300 ppm bovine serum albumin and humic acid solution. © 2018 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2018 , 135, 47102.     

Separation performance of a nanofiltration membrane influenced by species and concentration of ions

Nanofiltration (NF), which has been largely developed over the past decade, is a promising technology for the treatment of organic and inorganic pollutants in surface and ground waters. The ESNA 1 membrane from the Nitto Denko Corporation of Japan is made of aromatic polyamide, which provides salt rejection from 50% to 90%. In this paper permeation experiments of aqueous solutions of five chlorides (NH₄Cl, NaCl, KCl, MgCl₂ and CaCl₂), three nitrates (NaNO₃, Mg(NO₃)₂ and Ca(NO₃)₂), and three sulfates (NH₄)₂SO₄, Na₂SO₄ and MgSO₄) were carried out. The effects of species and concentration of salts on the separation performance of the ESNA 1 membrane were investigated. The experimental results showed that the rejection to most salts by the ESNA 1 membrane decreased with the growth of the concentration. Then, the reflection coefficient and solute permeability of ESNA 1 membrane were calculated by the Spiegler-Kedem equation from experimental data. The reflection coefficients of the ESNA 1 membrane to salts are all above 0.95. The salt permeabilities, except for magnesium and calcium salts, increased with the growth of concentration. The sequence of rejection to anions by the ESNA 1 membrane is R(SO²⁻₄) > R(Cl⁻) > R(NO⁻₃) at the same concentration which ranges from 10 mol/m³ to 100 mol/m³. The sequence of rejection to anions by the ESNA 1 membrane can be written as follows: R(Na⁺) > R(K⁺) > R(Mg²⁺) > R(Ca²⁺) at 10 mol/m³ concentration and R(Mg²⁺) > R(Ca²⁺) > R(Na⁺) > R(K⁺) at 100 mol/m³ concentration.     

Scavenging Ratios in an Urban Area in the Spanish Basque Country

For a period of six years (1995–2000) the scavenging ratio, which is the ratio of a pollutant's concentration in water to its concentration in air, collected at an urban site in the Spanish Basque Country was studied. The aerosol is characterized by SO₄ ^2? and NO₃? with 1.79 and 1.61 μg m^?3, respectively. Greater fractions of SO₄ ^2?, NO₃?, and NH₄+ ions were present in the fine particle range, while greater fractions of other ions appeared in the coarse range. The most important species found in the precipitation is SO₄ ^2? with 3.0 mg l^?1. NO₃?, Ca²⁺, and Cl^? are the second most important ions. The volume-weighted mean concentration of H⁺ is 4.6 μg l^?1 (pH = 5.3). The concentration of all analyzed ions (except H⁺) decreases throughout the rain event, showing the washout phenomenon of the rainwater. The scavenging ratio for the anthropogenic ions NO₃?, SO₄ ^2?, NH₄+, and K⁺ is lower than the scavenging ratio for the marine-terrigenous ions, Cl^?, Na⁺, and Ca²⁺.     

High‐performance composite nanofiltration membranes fabricated via ternary mixture: Complementary preponderance of the fluorine‐containing monomer 2,2′‐bis(1‐hydroxyl‐1‐trifluoromethyl‐2,2,2‐triflutoethyl)‐4,4′‐methylene dianiline and the rigid monomer bisphenol F

In a previous study, we proved that tailoring the polyamide backbone stiffness is an effective way to fabricate high‐performance polyamide nanofiltration (NF) membranes. However, in the previous study, we mainly focused on the flat membrane and did not consider its chlorine tolerance. In this study, by regulating the aqueous‐phase compositions in the interfacial polymerization process, chlorine tolerance on NF hollow‐fiber membranes was endowed while the membrane performance stayed high. The experimental results show that when the ratio of Piperazine (PIP)–bisphenol F (BPF)/2,2′‐bis(1‐hydroxyl‐1‐trifluoromethyl‐2,2,2‐triflutoethyl)‐4,4′‐methylene dianiline (BHTTM) was 5:1:4, the NF membrane possessed a permeate flux of 21.0 L m^?2 h^?1 bar^?1 and an Na₂SO₄ rejection up to 90.0%. X‐ray photoelectron spectroscopy analysis also confirmed that the polymerization degree of the PIP–BPF–BHTTM NF membrane was the highest. Moreover, the NF membrane could tolerate active chlorine to over 10,000 ppm h Cl. After the active chlorine treatment, the permeate flux increased over 30.0 L m^?2 h^?1 bar^?1, and the Na₂SO₄ rejection was about 90.0%. Although the PIP–BHTTM NF membrane also possessed good chlorine tolerance, its permeate flux (after active chlorine treatment) was only 60% of that of the PIP–BPF–BHTTM NF membrane. Therefore, the PIP–BPF–BHTTM NF membrane possessed a combination of high flux and high chlorine tolerance and showed good potential in water treatment in rigorous environments. © 2018 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2018 , 135, 46482.     

Preparation and characterization of a composite nanofiltration membrane from cyclen and trimesoyl chloride prepared by interfacial polymerization

A novel nanofiltration (NF) membrane was prepared with cyclen and trimesoyl chloride by interfacial polymerization on a poly(ether sulfone) ultrafiltration membrane with a molecular weight cutoff of 50,000 Da. The effects of the reaction time, monomer concentration, and heat‐treatment temperature are discussed. The physicochemical properties and morphology of the prepared NF membrane were characterized by Fourier transform infrared spectroscopy–attenuated total reflectance, scanning electron microscopy, energy‐dispersive spectrometry, and atomic force microscopy. The NF performances were evaluated with solutions of Na₂SO₄, MgSO₄, Mg(NO₃)₂, and NaCl. The salt‐rejection order of the prepared NF membrane was as follows: Na₂SO₄ > MgSO₄ > Mg(NO₃)₂ > NaCl. The resulting rejection of Na₂SO₄ and PEG600 (polyethylene glycol with the average molecular weight of 600) were more than 90%, whereas that of NaCl was approximately 10%. After the addition of silica sol in the aqueous phase (silica sol concentration = 0.1% w/v), the salt rejection of the membrane changed slightly. However, the water flux was from 24.2 L·m^?2·h^?1 (25°C, 0.6 MPa) up to 38.9 L·m^?2·h^?1 (25°C, 0.6 MPa), and the resulting membrane exhibited excellent hydrophilicity. © 2015 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2015 , 132, 42345.     

Catalytic template assisted interfacial polymerization for high-performance acid-resistant membrane preparation

Highly permeable acid-resistant nanofiltration (NF) membranes are of critical significance for the efficient treatment of acidic streams. Enhancing permeability while maintaining the high solute rejection of acid-resistant NF membranes remains a great challenge due to the low reactivity of monomers. In this work, a novel catalytic template assisted interfacial polymerization (IP) strategy of 3-aminobenzenesulfonamide (ABSA) and trimesoyl chloride (TMC) was provided to prepare a poly(amide-sulfonamide) membrane. Aminopyridine doped graphene quantum dots rich in acylation catalytic sites and ZIF-8 nanoparticles are co-loaded on a substrate as template. Benefiting from the enhanced phase integrity and self-inhibition effect of the template assisted IP process, the resulting ultra-thin acid-resistant membrane exhibits an excellent water permeance (20.4 Lm⁻²h⁻¹bar⁻¹) with a high Na₂SO₄ rejection of 90.5%, which outperforms almost all the reported acid-resistant NF membranes. Our work paves a versatile way for synthesis of special separation membranes.     

Electrocatalytic reduction of bromate ion using a polyaniline-modified electrode: An efficient and green technology for the removal of BrO3 in aqueous solutions

The electrocatalytic reduction of bromate ion (BrO₃⁻) was investigated in a three-electrode system using polyaniline (PANI) as the electrode material. Bromate ion reduction and Br⁻ removal were observed during electrochemical treatment because of the catalytic and doping capabilities of the PANI film. BrO₃⁻ removal efficiency in the 0.10 mol L⁻¹ Na₂SO₄ supporting electrolyte achieved 99% at pH 7 in 25 min, with no bromide ion detected in the solution. Optimal removal was found in pH range 6-7, and the pH of the solution had a significant impact on bromate reduction. A reduction mechanism was also discussed by analyzing the cyclic voltammograms of the reduction process and X-ray photoelectron spectra of the main elements (N 1s and Br 3d) on the PANI surface. We propose that during the electrocatalytic reduction process, bromate is reduced to bromide because of the loss of electrons from the nitrogen atoms on the PANI chains. The doping of the resultant Br⁻ ions in the PANI film has an important role in avoiding further oxidation of Br⁻ to BrO₃⁻. The used PANI film can be regenerated by de-doping the Br⁻ ions with a 0.5 mol L⁻¹ H₂SO₄ solution. Thus the process can be considered efficient and green.     

Impact of pH on the removal of fluoride, nitrate and boron by nanofiltration/reverse osmosis

The objective of this study was to evaluate the impact of pH on boron, fluoride, and nitrate retention by comparing modelled speciation predictions with retention using six different nanofiltration (NF) and reverse osmosis (RO) membranes (BW30, ESPA4, NF90, TFC-S, UTC-60, and UTC-80A). Retention was explained with regard to speciation, membrane properties, and ion properties such as charge, hydrated size, and Gibbs energy of hydration. Flux was independent of pH, indicating that pH did not alter pore size and hence permeability for all membranes except UTC-60. Membrane charge (zeta potential) was strongly dependent on pH, as expected. Boron and fluoride retention depended on membrane type, pH, which correlated closely to contaminant speciation, and was due both to size and charge exclusion. While retention at low and neutral pH was a challenge for boron, high boron retention was achieved (> 70% above pH 11). Fluoride retention was generally > 70% above pH 7. Nitrate retention depended on membrane, and was mostly pH independent (as was the speciation). The presence of a background electrolyte matrix (20 mM NaCl and 1 mM NaHCO₃) reduced nitrate and boron retention (at high pH) due to charge shielding, and enhanced the retention of fluoride in single feed solutions, suggesting preferential transport of Cl⁻ compared to F⁻ with Na⁺.