Next Article in Journal
Impact of Mining Area Steep Slope Conditions on the Soil and Water Conservation Benefits of Ecological Restoration
Previous Article in Journal
Enhancing Coastal Aquifer Characterization and Contamination Inversion with Deep Learning
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Water
Volume 17
Issue 2
10.3390/w17020254
water-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Ionic Strength and pH-Responsive Ultrafiltration Membrane to Overcome the Typical Permeability-Selectivity Tradeoff

by
Yian Chen
and
Yoram Cohen
*
Water Technology Research Center, Department of Chemical and Biomolecular Engineering, Henry Samueli School of Engineering and Applied Science, University of California, Los Angeles, CA 90095, USA
*
Author to whom correspondence should be addressed.
Water 2025, 17(2), 254; https://doi.org/10.3390/w17020254
Submission received: 19 December 2024 / Revised: 11 January 2025 / Accepted: 14 January 2025 / Published: 17 January 2025
(This article belongs to the Section Wastewater Treatment and Reuse)
Browse Figures
Versions Notes

Abstract

:
Stimuli-responsive polysulfone (PSf) ultrafiltration (UF) membrane was developed via surface modification with tethered hydrophilic polyacrylic acid (PAA) chains of length greater than the native membrane pore size. The surface nano-structured (SNS) membrane was synthesized via atmospheric pressure plasma-induced graft polymerization (APPIGP) to form a surface tethered PAA brush layer. The SNS-PAA-PSf UF membrane demonstrated hydraulic permeability and selectivity in the ranges of 0.74–2.29 × 1013 m−1 and 1.8–15.0 kDa, respectively, in response to changes in pH (3–11) and ionic strength (~0.02–547 mM). Membrane performance characterization showed that, for the above ranges of pH and salinity, the SNS-PAA-PSf UF membrane can overcome the typical membrane perm-selectivity tradeoff. The above performance is attributed to the swelling of the tethered PAA chains, upon ionic strength decrease or pH increase, which provides a less hindered transmembrane solute transport path, but increased hydraulic resistance. Conversely, at high ionic stress or low pH tethered chain collapse leads to lower molecular weight cutoff (MWCO) but with hydraulic resistance below that of the swollen state. The study results suggest that there is merit for further tailoring and improving the performance of stimuli-responsive UF membranes, developed via APPIGP, for applications over selected ranges of pH and ionic strength.

1. Introduction

Ultrafiltration (UF) membranes are widely used for water treatment and liquid separations and the global UF application market is growing at an annual growth rate of ~6 and it is expected USD 7.08 billion by 2030 [1,2,3]. Over the past four decades, there has been a growing interest in responsive-UF membranes whose physicochemical properties (i.e., surface charge, chemical functionality, surface hydrophilicity, electrostatic characteristics) and perm-selectivity can be adjusted in response to external stimuli (e.g., pH, ionic strength) [4,5,6,7,8,9,10,11]. Stimuli-responsive UF membranes have been proposed for use in a wide range of applications including wastewater [12] and produced water [13] treatment, as well as for desalination feed water pretreatment [14]. Such membranes have been prepared by various approaches such as blending, solution casting, phase inversion, self-assembly, surface coating, radiation, and graft polymerization [7]. Of particular interest in the present work are UF membranes that are formed by tethering polyelectrolyte chains onto the surface and or within the pores of the target base membrane. Triggered by the environmental stimuli (i.e., pH and ionic strength), the tethered polyelectrolyte chains undergo conformational changes (e.g., extension/collapse) so as to alter membrane performance [7,8].
The swelling and collapse of the surface tethered polyelectrolyte chains in response to solution pH and ionic strength have been well established in the literature. For example, a change in solution pH can trigger proton association-dissociation equilibrium (i.e., protonation and deprotonation) of the polyelectrolytes [15]. As solution pH increases above the acidic polyelectrolytes pKa, the surface tethered polymer chains deprotonate, leading to chain swelling owing to electrostatic repulsion between charged segments [16,17,18,19,20,21,22,23,24]. On the other hand, at a low pH, protonation of the ionizable functional groups of the surface tethered polymer brush layer leads to chain collapse [25]. Similarly, tethered polyelectrolyte chains also respond to changes in solution ionic strength due to charge screening by solution ions that disrupt electrostatic interactions among the charged polyelectrolyte chain segments [26,27], leading to chain collapse with increasing solution ionic strength.
Most studies on responsive UF membrane have reported that polyelectrolyte chains that are tethered inside membrane pores can act as an “on-off valve” (via “through-pore” mechanism; Figure 1), where chain swelling and collapse lead to narrowing (“close”) or enlarging (“open”) of the membrane pores [16,17,18,19,20,21,22,23,24,28,29,30,31,32], thereby leading to selective passage of molecules through the membrane. For size-based membrane separation, tunable membrane pore size can be achieved via external environmental stimuli [8] to regulate porous membrane permeability and selectivity. For example, graft polymerization of methacrylic acid (MAA) onto polyether sulfone UF membrane surface and inside the pores [17] demonstrated 8% increased bovine serum albumin rejection (from 91% to 99%) and up to 77% declined permeate flux (from 87 L·m−2·h−1 to 20 L·m−2·h−1) as the feed solution pH increased from 2 to 10, attributed to the swelling of the tethered MAA chains narrowing base membrane pores [17]. In another study, commercial polysulfone (PSf) UF membranes pores (10–35 nm size range) were modified with tethered poly(acrylic acid) (PAA) chains (5–10 nm chain length) that were synthesized via reversible addition-fragmentation chain-transfer polymerization [24]. Upon pH increase from 3 to 11, the above modified UF membranes demonstrated ~32–115% decreased permeate water flux and up to 79% decreased molecular weight cutoff (MWCO)—a behavior attributed to swelling of the tethered PAA chains [24]. In a later study, a UF membrane was surface modified by depositing polystyrene nanospheres (size range of 45–121 nm) with surface tethered PAA chains (25–72 nm chain length) [16] onto a polycarbonate base membrane. Compared to the base membrane, the rejection of nanoparticles and dyes of the synthesized UF membranes increased by ~1–50% and ~27–31%, respectively, upon solution pH increase from 3.5 to 10.1. However, the rejection increase was accompanied by up to ~68% water permeate flux decline [16], and significant membrane performance degradation was observed with progressive operational cycles. Overall, the separations performance of the above-mentioned responsive UF membranes has been attributed, in part, to the “through-pore” mechanism whereby the tethered polymer chains size (i.e., length) was reported to be 75–28% lower compared to the membrane pore diameter and thus can be synthesized inside membrane pores. It is stressed that the modified membranes reported in the above studies were all challenged by the typical tradeoff of increased selectivity at the cost of decreased membrane hydraulic permeability [16,17,18,19,20,21,22,23,24,28,29,30,31].
UF membranes can also be synthesized by tethering polyelectrolyte chains of length significantly greater than membrane pore size to favor tethering of the polymer chains onto the outer membrane surface [32]. Solute transport through such membranes was reasoned to follow the “through-polymer” mechanism (Figure 1) which was first introduced in 1997 [28]. According to the above mechanism, when the tethered polymer chains extend away from the membrane surface, the underlying membrane pores are exposed hence enhancing solute transport through the membrane and thus lower solute rejection [25,32,33]. In contrast, collapsed tethered polymer chains form a tightened layer, thereby reducing the available free volume which impedes transmembrane solute transport and hence increased solute rejection [26,34]. Performance of the above class of membranes is expected to be responsive to ionic strength and pH which will impact membrane swelling and membrane transport will then be governed by the “through-polymer” mechanism. To the knowledge of the authors, the existing literature has not reported on the performance of this type of a responsive membrane as attributed to the “through-polymer” mechanism nor on overcoming its perm-selectivity tradeoff. Accordingly, the present work reports on an investigation of the above UF membrane type that is ionic strength and pH-responsive and its potential for overcoming the typical perm-selectivity tradeoff.
Given the above stated potential for UF membrane performance enhancement, the present study reports on a systematic investigation of the separations performance of a stimuli-responsive PSf UF membrane having a surface nano-structured (SNS) with tethered hydrophilic PAA chains. UF membrane surface nano-structuring was accomplished via atmospheric pressure plasma-induced graft polymerization (APPIGP) [35,36] to achieve tethered chains of length that were significantly longer than the membrane pore size. The APPIGP approach was selected given its proven potential for scalability [37], and the hydrophilic PAA was selected for the tethered polymer layer given its swelling/collapse response to pH and ionic strength, and demonstrated fouling resistance in membrane filtration and desalination [36]. In the current study, the impact of ionic strength and pH on the performance of the synthesized SNS-PAA-PSf membrane was evaluated, in terms of the membrane’s hydraulic resistance and MWCO. In addition, we rationalize the observed membrane transport performance as per the “through polymer” mechanism and report on the conditions at which the permeability-selectivity tradeoff is overcome for the SNS-PAA-PSf UF membrane.

2. Materials and Methods

2.1. Overview

An ultrafiltration (UF) membrane whose performance is responsive to changes in ionic strength and pH was synthesized and characterized with respect to hydraulic resistance and molecular weight cutoff (MWCO) using a model solute. This stimuli-responsive membrane comprised of a Base PSf membrane whose surface was nano-structured (SNS) with a layer of tethered PAA chains synthesized via atmospheric pressure plasma-induced graft polymerization of acrylic acid. The MWCO, hydraulic resistance, and surface hydrophilicity of the resulting SNS-PAA-PSf membrane were evaluated over pH and ionic strength ranges of 3–11 and ~0.02–547 mM, respectively, and also compared with a native PSf membrane of equivalent MWCO. In addition, the membrane permeability-selectivity trade-off was evaluated, relative to the performance of commercial and literature-reported membranes, to demonstrate the potential for overcoming the perm-selective tradeoff with the SNS-PAA-PSf membrane.

2.2. Materials

Polysulfone membrane sheets (MUF-20K and MUF-10K, Toray Membrane USA Inc., Poway, CA, USA) with manufacturer reported 100 kDa and 10 kDa MWCO, respectively, were selected as the base membranes for surface modification (Base-PSf) and as a reference membrane for comparison (Native-PSf), respectively. Prime grade 4″ silicon 〈100〉 wafers were utilized as a base for preparing model surfaces with PAA chains tethered onto a pre-coated layer of PSf. Sulfuric acid (96%) and aqueous hydrogen peroxide (30%) (KMG Electronic Chemicals, Inc., Houston, TX, USA) were used for silicon wafer cleaning to remove organic residues. The cleaned silicon wafer was then spin coated with poly(ethyleneimine) (PEI, Mw~750,000) solution (50 wt% in H2O, Sigma-Aldrich, St. Louis, MO, USA) to form an adhesion layer. A PSf layer was subsequently formed on top of the PEI layer by spin coating a PSf solution prepared from 0.25 wt% PSf pellets (Mw~35,000, Sigma-Aldrich, St. Louis, MO) dissolved in chloroform (≥99.9%, Sigma-Aldrich, St. Louis, MO, USA). Acrylic acid (AA) monomer (99%, Sigma-Aldrich, St. Louis, MO, USA) was used for graft polymerization to modify the Base-PSf surface post atmospheric pressure plasma activation. Helium (99.999%) and oxygen (99.999%) gases (Airgas, Los Angeles, CA, USA) were the source gases for plasma surface treatment. Nitrogen (99%) gas (Airgas, Los Angeles, CA, USA) was used for membrane surface drying, reaction mixture degassing, and pressurizing the UF filtration feed tank. The MWCO of the UF membranes was characterized using polyethylene glycol (PEG, Mw = 1000–20,000 Da; Sigma-Aldrich, St. Louis, MO). Sodium chloride (NaCl, ≥99.0%; Fisher Scientific, Chino, CA, USA) was used to adjust solution ionic strength, and aqueous solutions of 0.1 N hydrochloric acid and sodium hydroxide (Fisher Scientific, Chino, CA, USA) served to adjust the solution pH.

2.3. Preparation of PSf Surrogate Membrane Surfaces

PSf surrogate membrane surfaces were synthesized onto a silicon wafers to provide a relatively smooth membrane surface for synthesis and characterization of the tethered PAA layer thickness, surface roughness, average feature height, and surface contact angle. The silicon wafers were first cut into 1.5 cm × 1.5 cm square samples using Small Sample Cleaving Pilers (GC-SS-100, LatticeGear, Beaverton, OR, USA) and cleaned in a piranha solution (an aqueous mixture of 70 vol% sulfuric acid and 30 vol% hydrogen peroxide) at 90 °C for 10 min, followed by five cycles of thorough rinsing with D.I. water. Prior to spin coating, each wafer piece was washed with isopropanol and D.I. water, followed by nitrogen drying. A PEI adhesion layer was then formed on the cleaned silicon wafer substrates by spin coating (using 790 Spinner with PWM32 controller; Headway Research Inc., Garland, TX, USA) 0.1 mL of a 0.3 wt% aqueous PEI solution (centrally placed using a pipet onto the substrate surface) at 2500 rpm for 30 s. Immediately afterward, the PEI-Si surfaces were spin coated with 0.1 mL of 0.25 wt% PSf solution in chloroform at 2500 rpm for 30 s forming a ~26 nm PSf layer (determined via AFM characterization). The surrogate PSf membrane surfaces (PSf-PEI-Si) were then vacuum oven dried (40 °C for 24 h) prior to surface nano-structuring and characterization.

2.4. Atmospheric Pressure-Induced Graft Polymerization (APPIGP)

The tethered PAA layer was synthesized onto both surrogate membrane surface and the Base-PSf membrane surface via surface activation with atmospheric pressure plasma (APP) induced free-radical graft polymerization (APPIGP) of AA (Figure 2), following a well-established protocol [35,36,38]. In the above APPIGP approach the upper membrane surface is activated, to form surface free-radicals for acrylic acid graft polymerization [35,36]. Therefore, the PAA chains were tethered primarily onto the external membrane surface. Details of the above previously developed APPIGP membrane surface modification approach and its application to membrane surface nano-structuring, along with detailed physicochemical surface characterization are provided elsewhere [35,36,37,38,39]. Briefly, the Base-PSf membrane coupons (with active areas of 13.4 cm2) were extracted from the base membrane sheets (Toray MUF-20K) and then immersed in D.I. water for at least 24 h. The Base-PSf membrane coupons were subsequently blown dried with nitrogen gas using a polytetrafluoroethylene nitrogen/drying gun (International Polymer Solutions, Inc., Irvine, CA, USA). Surface activation of both the Base-PSF and surrogate PSf membrane surfaces was accomplished via helium/oxygen (He/O2) APP with helium and oxygen flow rates of 45 L·min−1 and 0.5 L·min−1, respectively. The plasma was generated with an Atomflo™ 500 APP system equipped with a source head mounted on an XYZ scanning robot (Surfx Technologies LLC, Redondo Beach, CA, USA). The plasma stream, generated at 150 W RF power, was delivered onto the PSf surface via two sequential scans at a speed of 100 mm·s−1 at a source-surface separation distance of 15 mm.
Graft polymerization of AA to form a surface tethered PAA layer was achieved by immersing the plasma activated PSf samples in 250 mL glass reaction vessels containing 20 vol% aqueous AA monomer solutions at pH~1.9 adjusted using aqueous NaOH solution (50% w/w). The capped glass vessels were immersed in a constant temperature water bath at 70 °C for 60 min. Nitrogen was bubbled into the monomer solution (via a perforated tube) during graft polymerization to promote mixing and scavenge dissolved oxygen that could inhibit the polymerization reaction. The above surface activation conditions (i.e., plasma type, plasma treatment and graft polymerization conditions) and AA graft polymerization onto the PSf surface were adopted based on previously established conditions for UF membrane performance tuning [35,36]. After graft polymerization, the SNS-PAA-PSf surfaces were rinsed with D.I. water, followed by immersion in D.I. water for ~24 h period prior to characterizations. It is stressed that the above protocol results in covalently anchored (i.e., tethered) PAA surface chains which form a stable modification layer [37,40,41]. It is noted that, under the current synthesis conditions (Section 2.4), the surface tethered PAA were previously determined [39] to have length of 96–145 nm and distance of 1.5–2.4 nm between neighboring tethered chains.

2.5. Membrane Surface Hydrophilicity

The impacts of solution pH and ionic strength on surface wettability of SNS-PAA-PSf and Native-PSf UF membranes were evaluated by captive bubble (CB) water contact angle measurements using an automated drop shape analyzer (DSA20; KRÜSS GmbH, Hamburg, Germany). The membranes were immersed in the target aqueous solutions (pH and ionic strength ranges of 3–11 and ~0.02–547 mM, respectively) at room temperature (~20 °C) for at least 30 min prior to the contact angle measurement. Membrane surface liquid contact angle measurements were carried out within 2 s of dispensing a 6 µL air bubble onto the immersed membrane surfaces using a ‘J’-shaped needle. Measurements were taken at 5 locations for each sample, and the reported values represent the average of each set of measurements. Membrane surface hydrophilicity was then quantified by the surface free energy of hydration (ΔGiw) calculated using Young-Dupré equation,
∆ G i = − γ ( 1 + c o s θ )
where γ is the liquid surface tension and θ is the contact angle. Surface tensions ( γ ) of the aqueous solutions at different pH levels and ionic strengths are provided in Table S1 (Supplementary Material).

2.6. Surface Topography

PSf surrogate membrane surfaces (Section 2.3), before and after surface nano-structuring, were characterized via AFM, using a Bruker Dimension Icon Scanning Probe Microscope (Bruker, Santa Barbara, CA, USA). Briefly, the PSf-PEI-Si surfaces were immersed in an aqueous solution (or D.I. water) for 30 min to allow equilibration prior to AFM characterizations using AFM probes (ScanAsyst-Fluid+, Bruker, Santa Barbara, CA, USA) in the Mechanical Force Tapping mode [42]. Surface topography was evaluated via AFM, in aqueous solutions, for 1 μm × 1 μm regions at 512 × 512-pixel resolutions, at a probe loading force ~1 nN and 0.8 Hz scan rate. The root-mean-square surface roughness ( R r m s ) was determined from the AFM measurements as,
R r m s = ∑ i N ( Z i − Z a v g ) 2 / N
where Z i and Z a v g are the feature height (FH) of the i t h data point and the average FY, respectively, and N is the total number of data points within the scanned area (1 μm × 1 μm). The tethered PAA layer thickness was determined from the AFM cross-sectional height profile in the aqueous solutions, as per the protocol in [36].

2.7. Membrane Performance Characterization

Hydraulic permeability and MWCO for the membranes were determined over a pH and ionic strength (I.S) ranges of 3–11 and ~0.02–547 mM, respectively. All feed solutions were prepared using D.I. water (I.S.~0.02 mM) with the pH (adjusted with 0.1 N HCl and 0.1 N NaOH solutions) monitored using a pH meter (Oakton pH 110; Cole-Parmer, Vernon Hills, IL, USA). The solution I.S (adjusted using NaCl solution) was monitored with a conductivity probe (WD-35604-00, OAKTON, Chicago, IL, USA).
The membrane hydraulic permeability coefficient ( L p ) and PEG MWCO were determined based on tests using a dead-end stirred UF cell (Amicon 8050, Millipore Corporation, Burlington, MA, USA), which accommodated an active membrane area of 13.4 cm2. The membrane volumetric permeate flow rate was monitored with an in-line liquid flow meter (SLS-1500, Sensirion AG, Stäfa, Switzerland). Post membrane compaction, at 3.5 bar (~50 psi) and ~20 °C for 3 h, the hydraulic permeability coefficient ( L p ) was determined, for the unmodified and modified membranes based on permeate flux measurements with feed solutions at the target pH and ionic strength conditions. L p was then determined from the slope of a linear plot of
J v = Q p / A = L p · ∆ P
in which Q p and A are the permeate flow rate and membrane surface area, respectively, and ΔP is the applied transmembrane pressure. The hydraulic membrane resistance ( R m ) was determined based on the flux expression of
J v = ∆ P / ( μ · R m )
whereby
R m = 1 / ( μ · L p )
in which µ is the aqueous solution viscosity.
Determination of the separation performance of the SNS-PAA-PSf and Native-PSf UF membranes was determined using a series of PEG Mw fractions in the range of 1000–35,000 Da. PEG was selected for UF MWCO characterization given that it is anionic with negligible interactions with membrane surfaces [43,44,45]. MWCO determinations were carried out using an aqueous 1 g·L−1 PEG solution at an initial permeate flux of 9 L∙m−2∙h−1 (~5.3 gallon⋅ft−2⋅day−1). PEG feed ( C f ) and permeate ( C p ) concentrations were determined via a Total Organic Carbon Analyzer (Aurora 1030W, OI Analytical, College Station, TX, USA). The observed (nominal) solute rejection for each PEG fraction was determined from
R o = ( 1 − C p / C f ) × 100 %
Based on correlation of PEG rejection with its Mw, whereby the Mw fraction corresponding to 90% rejection was then designated as the membrane PEG MWCO. Also, the membrane pore size was estimated given the membrane PEG MWCO using the previously reported empirical correlation [35]
dp = 1.3095 × MWCO0.3409
which is applicable for the range of MWCO in the current work.

3. Results and Discussion

3.1. Overview

The surface of a polysulfone (PSf) UF membrane was structured poly(acrylic acid) (PAA) chains that were terminally and covalently anchored (i.e., tethered) to the upper surface of the Base PSf membrane. The base membrane nominal pore size (~6.3 nm [35]) was significantly smaller than the average PAA chain length formed under the current synthesis conditions (Section 2.4) whereby the ratio of tethered PAA chain length to the base UF membrane pore size being ≥15 and thus a solute transport behavior that is governed by the “through-polymer” mechanism (Figure 1). The experimental characterization of the SNS-PAA-PSf membrane revealed that these membranes were hydrophilic (Section 3.2) and exhibited stimuli-responsive performance whereby its hydraulic resistance and molecular weight cutoff (MWCO) varied with the ionic strength and pH (Section 3.3 and Section 3.4). The membrane hydraulic resistance and MWCO (characterized based on PEG) both decreased at lower pH or higher ionic strength which enhanced both membrane permeability and enabled the SNS-PAA-PSf UF membrane to overcome the typical permeability-selectivity tradeoff (Section 3.5).

3.2. Surface Wettability

The captive bubble (CB) water contact angle for the SNS-PAA-PSf membrane decreased from 56.1° to 43.4° as the ionic strength decreased from 547 to ~0.02 mM (Figure 3, Table 1). Correspondingly, the surface energy decreased from −114.5 mJ·m−2 to −125.7 mJ·m−2 indicating increased membrane surface hydrophilicity. In contrast, for the above same ionic strength range, the Native-PSf membrane displayed minor CB water contact angle variability (50.4 ± 2.3°) (Figure 3, Table 1; also, Figure S1, Supplementary Material). Surface hydrophilicity of the SNS-PAA-PSf membrane decreased (as indicated by the increased membrane surface energy) with increasing solution ionic strength, corresponding to the charge screening effect of Na+ and Cl− ions that disrupt electrostatic interactions among the charged polyelectrolyte chain segments [26,27] and leading to chain collapse. Based on AFM surface characterization of the surrogate membrane surface (Section 2.3), it was assessed that the tethered PAA layer RMS surface roughness and average feature height decreased by 84% and 78%, respectively, and the PAA brush layer thickness decreased by ~83% (Figure S1 and Table S3) upon elevation of the ionic strength from 0.02 mM to 547 mM. The collapse of the surface tethered hydrophilic polymer chains corresponded with the decrease in surface hydrophilicity (as indicated by higher membrane surface energy), both of which are driven by the tendency of the polymer brushes to minimize their contact with the high salinity considered as a “bad” solvent for the PAA chains [46].
As the solution pH increased toward the alkaline range, the SNS-PAA-PSf membrane surface CB water contact angle decreased by ~44% over the pH range of 7–11, corresponding to surface energy decrease of ~11% (Table 1 and Figure 4). At neutral and alkaline pH, which is above the pKa of PAA (~4.8, [35]) the carboxylic acid groups of PAA are ionized to -COO- and protons [47], and electrostatic repulsion between chain segments (i.e., -COO- groups) would then leads to swelling (extension) of the tethered PAA chains. However, at acidic pH (below the pKa of PAA), protonation of the ionizable carboxylic acid side groups of PAA lead to chain collapse; hence there is little change in the contact angle and surface energy over pH range of 3–7.
Over the tested ranges of pH (3–11) and ionic strength (~0.02–547 mM), the SNS-PAA-PSf membrane surface energy varied over the range of −114.5 to −139.2 mJ·m−2 relative to a nearly constant value of ~120.8 ± 2.9 mJ·m−2 for the Native-PSf membrane (Figure 5, Table 1). The unaltered surface hydrophilicity of the Native-PSf membrane was consistent with other reported studies [14,48]. For the SNS-PAA-PSf membrane, on the other hand, swelling of PAA chains (at high pH and low ionic strength) was accompanied by increased surface hydrophilicity (i.e., lower surface energy), while the collapse of PAA chains (at low pH and high ionic strength) corresponds to increased surface hydrophobicity (i.e., higher surface energy) [7,8,26,49]. Although the subject of membrane fouling is outside the scope of the present study, we note that the surface energy of the SNS-PAA-PSf membrane was below the threshold of −113 mJ·m−2 (proposed by Van Oss [50]) for which a surface is considered hydrophilic. Hence, the hydrophilic surface layer of tethered PAA chains should reduce membrane fouling propensity. Moreover, since chain swelling/collapse can be modulated by external stimuli (i.e., ionic strength and pH) this behavior could be harnessed to improve membrane cleaning efficiency [5,36,39,51,52,53,54].

3.3. Ionic Strength-Responsive SNS-PAA-PSf UF Membrane Performance

In response to solution ionic strength, the resulting conformational changes of the surface tethered PAA chains are expected to affect the PSf UF membrane hydraulic resistance (Rm) and PEG MWCO. Indeed, the SNS-PAA-PSf membrane exhibited ~29% membrane hydraulic resistance decline (from 1.74 to 1.24 × 1013 m−1) and ~78% PEG MWCO decrease (from 8.0 to 1.8 kDa) in response to solution ionic strength rise from ~0.02 to 547 mM (Table 2). The above behavior is consistent with the “through-polymer” mechanism (Figure 1) whereby decreased brush layer thickness, upon collapse of the PAA chains, should result in a lower hydraulic membrane resistance relative to the status of a surface with swollen chains. Over the tested ranges of solution ionic strength and pH, the SNS-PAA-PSf membrane was hydrophilic (i.e., surface energy below −113 mJ·m−2, Section 3.2). Consequently, membrane hydraulic resistance is not expected to be impacted over the above variability surface energy. On the other hand, PAA chains collapse over the membrane pore reduced the SNS-PAA-PSf membrane effective pore size (from 2.7 to 1.6 nm) and correspondingly the membrane hydraulic resistance also increased. It is also plausible that the collapse of the PAA brush layer reduces the void volume within the PAA layer, hence reducing the effective solute diffusivity through the polymer surface layer. In contrast, at a low solution ionic strength, the surface tethered PAA chains are in a more extended configuration, and thus the membrane hydraulic resistance should increase with increased PAA surface layer thickness. However, it is reasonable to expect greater solute access to the underlying membrane surface pores upon chain swelling given the greater access provided to the underlying base membrane pores; the above is consistent with the “through-polymer” mechanism (Figure 1, Introduction) and previous observations regarding the filtration of bovine serum albumin and lysozyme using zirconia-based membranes with surface tethered poly(vinyl pyrrolidone) [25].
The Native-PSf membrane demonstrated negligible membrane hydraulic resistance variation (~3.9%) with increased ionic strength (over the range of ~0.02 to 547 mM) relative to a greater membrane hydraulic resistance decrease (by 28.5%) for the SNS-PAA-PSf membrane (Table 2 and Figure 6). However, a significant PEG MWCO increase of 53% (from 10 to 15.3 kDa) was observed for the Native-PSf membrane as the solution ionic strength was elevated from ~0.02 to 547 mM (Table 2 and Figure 7). The above trend could be due to reduction in the hydrated size of the PEG molecules. For example, it was previously reported that partial dehydration of PEG molecules led to effective PEG (Mw ~ 400–1000 g/mol) molecular size reduction by 9–22% (from Stokes radius of ~0.45–0.6 nm) induced by surrounding ions in aqueous salt solutions (0.1–1 M, KCl, LiCl, and MgCl2) [55]. Under the above conditions, it was reported that PEG (Mw = 600 g·mol−1) rejection by a ceramic nanofiltration membrane decreased by up to 40%. In the present study, it is estimated that dehydration of the PEG molecules was a likely contributor to the observed increased Native-PSf membrane PEG MWCO (by 53%) upon solution ionic strength rise from ~0.02 to 547 mM. In contrast, even with dehydration of PEG molecules, the SNS-PAA-PSf membrane exhibited decreased PEG MWCO with increased solution ionic strength. The above results suggest that the possible impact of the effective PEG molecular size decrease, due to dehydration, is insignificant relative to any potential effective membrane pore size decrease that may have resulted by the collapsed PAA chains. Here we note that as the solution ionic strength increases, the SNS-PAA-PSf membrane surface tethered PAA chains transition from a swollen layer to a more collapsed one which amplifies covering of the external membrane surface (Figure 8). Hence upon increased ionic strength, both solute rejection and membrane water permeation increased (due to the thinner PAA layer), as expected according to the “through polymer” transport mechanism (Figure 1).

3.4. pH-Responsive SNS-PAA-PSf UF Membrane Performance

Conformational changes of the surface tethered PAA chains due to pH change were also observed to affect the hydraulic resistance and MWCO of the SNS-PAA-PSf membrane. As the solution pH increased from 3 to 11, the membrane intrinsic hydraulic resistance (Rm) and PEG MWCO increased by factors of 3.1 and 2.2, respectively (Figure 9 and Figure 10, and Table 2), consistent with the “through-polymer” mechanism (Figure 1 and Figure 8). The above behavior is postulated to be due to swelling of the tethered PAA chains, away from the membrane surface, leading to increased PAA layer thickness and thus greater membrane hydraulic resistance. Yet, the swollen PAA brush layer provides a clearer path for solute permeation relative to the case of a collapsed surface PAA layer (Figure 8).
The surface hydrophilicity of the Native-PSf membrane remained unchanged over the pH range of 3–11 (Table 1 and Figure 4), where the surface CB contact angle (46.8°) variability was at most ±5% (Figure S1, Supplementary Material). Membrane hydraulic resistance for the Native-PSf membrane also demonstrated a nearly constant value (with variability of ±6%) over the tested pH range (3–11), confirming that the Native-PSf membrane is not pH-responsive. On the other hand, the Native-PSf membrane PEG MWCO varied in the range of 10–16.8 kDa over the pH range of 3–11 with a minimum MWCO at pH of 7; the latter is possibly due to PEG hydrolysis at low or high pH [56]. In contrast, the hydraulic resistance and PEG MWCO of the SNS-PAA-PSf membrane increased from 7.4 × 1012 to 2.29 × 1013 m−1 and PEG MWCO 6.9–15 kDa, respectively, as the solution pH increased from 3 to 11 (Table 2). The above indicates that pH-responsiveness of the SNS-PAA-PSf membrane is due to the surface tethered PAA chains.

3.5. Tuning of the Responsive SNS-PAA-PSf UF Membrane Performance and Surface Wettability

Over the present range of experimental solution conditions (i.e., pH 3–11 and ionic strength ~0.02–547 mM), wide performance ranges in terms of membrane hydraulic resistance (Rm = 7.4–2.29 × 1012 m−1) and PEG MWCO (=1.8–15.0 kDa) were observed for the SNS-PAA-PSf membrane (Figure 11). Stimuli-responsive SNS-PAA-PSf membranes with tunable perm-selectivity have the potential to increase UF membrane flexibility by expanding its applications to a wide range of operating pH and ionic strength conditions (i.e., food and beverage, pharmaceutical, petroleum, and chemical industries) without the necessity to utilize different membranes.
Over the ranges of tested pH and ionic strength, the Native-PSf membrane hydraulic resistance and PEG MWCO varied over narrow ranges of 1.37 × 1013–1.61 × 1013 m−1 and 10.0–16.8 kDa, respectively. This apparent pH and ionic strength responsive performance is likely linked to the variability of PEG effective molecular size with pH and ionic strength [55]. In contrast, the decrease in SNS-PAA-PSf membrane hydraulic resistance was accompanied by a measurable PEG MWCO reduction in response to increased solution ionic strength or reduced solution pH (Table 2). Increased SNS-PAA-PSf membrane hydraulic permeability (i.e., up to 68% lower Rm) and selectivity (up to 77% lower PEG MWCO) were achieved with decreased pH and increased ionic strength owing to transitioning of the tethered PAA chains to a more collapsed conformation. The above behavior is as expected for the SNS-PAA-PSf membrane in which water and solute transport are likely governed by the “through-polymer” mechanism (Figure 1). Consequently, the responsive SNS-PAA-PSf UF membrane can overcome the typical tradeoff between membrane hydraulic permeability and selectivity over the current range of tested ionic strength and pH conditions as illustrated in Figure 12. In this figure, the form of the conventional plot of membrane hydraulic permeability coefficient (Lp) versus selectivity, here expressed as the Lp/MWCO ratio, also shows the UF membrane upper bound tradeoff as assessed based on the literature data. The data for the SNS-PAA-PSf membrane indicates that at pH < 5 or ionic strength > 171 mM, this UF membrane overcomes the typical membrane perm-selectivity tradeoff.

4. Conclusions

A pH and ionic strength-responsive polysulfone UF membrane was synthesized with hydrophilic PAA chains of size (average length of ~96–145 nm) significantly greater than the base membrane pore size (~6.3 nm) tethered onto the external membrane surface. The resulting SNS-PAA-PSf membrane demonstrated stimuli-responsive transport properties (i.e., hydraulic permeability and molecular weight cutoff) over pH and ionic strength ranges of 3–11 and ~0.02–547 mM (as NaCl), respectively. Over the above pH and ionic strength ranges, the SNS-PAA-PSf membrane hydraulic resistance and PEG MWCO were in the ranges of 7.4 × 1012–2.29 × 1013 m−1 and 1.8–15.0 kDa, respectively, compared to much narrower performance ranges of 1.37 × 1013–1.61 × 1013 m−1 and 10.0–16.8 kDa for the Native-PSf membrane. The wider range of transport performance for the SNS-PAA-PSf membrane is attributed to conformational changes (i.e., swelling and collapse) of the surface tethered PAA chains upon changes in the aqueous environment ionic strength and pH. Decreased membrane hydraulic resistance accompanied by decreased PEG MWCO for the SNS-PAA-PSf membrane, upon decreased solution pH or increased solution ionic strength, resulted in simultaneous increased membrane hydraulic permeability and selectivity. Hence, the typical hydraulic permeability-selectivity tradeoff was overcome by the responsive SNS-PAA-PSf UF membrane. Given the results of the present study, it is suggested that there is merit in further exploration of the utility of the current approach for developing such membranes for targeted UF separations, particularly at conditions of low pH and high ionic strength. Also, further investigations are warranted to explore the stimuli-responsive performance of the SNS-PAA-PSf membranes under conditions of simultaneous variations of both pH and ionic strength.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/w17020254/s1, Table S1: Surface tension of different liquids used in the present study [60,61,62,63,64,65,66], Table S2: Membranes performance data used to establish UF membrane tradeoff upper bound [35,36,54,57,58,59], Figure S1: AFM 3-D topography of PAA-PSf-PEI-Si surfaces in NaCl solutions with ionic strength of (a) 2, (b) 17, (c) 171, and (d) 547 mM; Table S3: Surface roughness, average feature height (FH), and PAA brush layer thickness for the native PSf-PEI-Si and PAA-PSf-PEI-Si surfaces in D.I. water and saline solutions with an ionic strength range of 0.02–547 mM [36].

Author Contributions

Conceptualization, methodology, validation, formal analysis, writing—original draft preparation, and visualization, Y.C. (Yian Chen); investigation, resources, data curation, writing—review and editing, supervision, project administration, funding acquisition, Y.C. (Yoram Cohen). All authors have read and agreed to the published version of the manuscript.

Funding

The present work was supported, in part, by the U.S. Army (Grant W56HZV-16-0117) and the California Department of Water Resources (Agreement Number 4600011630).

Data Availability Statement

The data presented in this study are available in this article.

Acknowledgments

The authors acknowledge the assistance of Toray Membrane USA Inc.

Conflicts of Interest

The authors declare no conflicts of interest.

References

  1. Yang, J.; Monnot, M.; Eljaddi, T.; Ercolei, L.; Simonian, L.; Moulin, P. Ultrafiltration as tertiary treatment for municipal wastewater reuse. Sep. Purif. Technol. 2021, 272, 118921. [Google Scholar] [CrossRef]
  2. Nikhil, M.Y.P. Ultrafiltration Market Size, Share, Competitive Landscape and Trend Analysis Report by Type (Polymeric, Ceramic), by Module (Hollow Fiber, Plate and Frame, Tubular), by Application (Municipal Treatment, Industrial Treatment): Global Opportunity Analysis and Industry Forecast, 2021–2031; Allied Market Research: Portland, OR, USA, 2022. [Google Scholar]
  3. Global Ultrafiltration Membrane Filtration Market by Type (Organic Membrane, Inorganic Membrane), by Application (Industrial & Municipal, Food & Beverage), by Geographic Scope and Forecast; Report ID: 78328; Verfied Market Reports: Washington, DC, USA, 2024.
  4. Abbas, M.A.; Mushtaq, S.; Cheema, W.A.; Qiblawey, H.; Zhu, S.; Li, Y.; Zhang, R.; Wu, H.; Jiang, Z.; Sadiq, R. Surface Modification of TFC-PA RO Membrane by Grafting Hydrophilic pH Switchable Poly (Acrylic Acid) Brushes. Adv. Polym. Technol. 2020, 2020, 8281058. [Google Scholar] [CrossRef]
  5. Liu, H.; Yang, S.; Liu, Y.; Miao, M.; Zhao, Y.; Sotto, A.; Gao, C.; Shen, J. Fabricating a pH-responsive membrane through interfacial in-situ assembly of microgels for water gating and self-cleaning. J. Membr. Sci. 2019, 579, 230–239. [Google Scholar] [CrossRef]
  6. Huang, T.; Su, Z.; Hou, K.; Zeng, J.; Zhou, H.; Zhang, L.; Nunes, S.P. Advanced stimuli-responsive membranes for smart separation. Chem. Soc. Rev. 2023, 52, 4173–4207. [Google Scholar] [CrossRef] [PubMed]
  7. Wandera, D.; Wickramasinghe, S.R.; Husson, S.M. Stimuli-responsive membranes. J. Membr. Sci. 2010, 357, 6–35. [Google Scholar] [CrossRef]
  8. Liu, Z.; Wang, W.; Xie, R.; Ju, X.-J.; Chu, L.-Y. Stimuli-responsive smart gating membranes. Chem. Soc. Rev. 2016, 45, 460–475. [Google Scholar] [CrossRef] [PubMed]
  9. Darvishmanesh, S.; Qian, X.; Wickramasinghe, S.R. Responsive membranes for advanced separations. Curr. Opin. Chem. Eng. 2015, 8, 98–104. [Google Scholar] [CrossRef]
  10. Uredat, S.; Gujare, A.; Runge, J.; Truzzolillo, D.; Oberdisse, J.; Hellweg, T. A review of stimuli-responsive polymer-based gating membranes. Phys. Chem. Chem. Phys. 2024, 26, 2732–2744. [Google Scholar] [CrossRef] [PubMed]
  11. Pan, Y.; Liu, Y.; Yang, S.; Zhang, C.; Ullah, Z. Recent research progress on the stimuli-responsive smart membrane: A review. Nanotechnol. Rev. 2023, 12, 20220538. [Google Scholar] [CrossRef]
  12. Birmod, R.P.; Lade, V.G.; Shambharkar, R.D. Responsive membranes for wastewater treatment. In Handbook of Nanomaterials for Wastewater Treatment; Elsevier: Amsterdam, The Netherlands, 2021; pp. 673–697. [Google Scholar]
  13. Wandera, D.; Wickramasinghe, S.R.; Husson, S.M. Modification and characterization of ultrafiltration membranes for treatment of produced water. J. Membr. Sci. 2011, 373, 178–188. [Google Scholar] [CrossRef]
  14. Gancarz, I.; Poźniak, G.; Bryjak, M. Modification of polysulfone membranes 1. CO2 plasma treatment. Eur. Polym. J. 1999, 35, 1419–1428. [Google Scholar] [CrossRef]
  15. Sinha, M.; Purkait, M. Preparation and characterization of novel pegylated hydrophilic pH responsive polysulfone ultrafiltration membrane. J. Membr. Sci. 2014, 464, 20–32. [Google Scholar] [CrossRef]
  16. Zhang, R.; Zhou, T.; Peng, H.; Li, M.; Zhu, X.; Yao, Y. Nanostructured switchable pH-responsive membranes prepared via spherical polyelectrolyte brushes. J. Membr. Sci. 2019, 580, 117–124. [Google Scholar] [CrossRef]
  17. Shi, Q.; Su, Y.; Ning, X.; Chen, W.; Peng, J.; Jiang, Z. Graft polymerization of methacrylic acid onto polyethersulfone for potential pH-responsive membrane materials. J. Membr. Sci. 2010, 347, 62–68. [Google Scholar] [CrossRef]
  18. Iwata, H.; Matsuda, T. Preparation and properties of novel environment-sensitive membranes prepared by graft polymerization onto a porous membrane. J. Membr. Sci. 1988, 38, 185–199. [Google Scholar] [CrossRef]
  19. Wang, M.; An, Q.-F.; Wu, L.-G.; Mo, J.-X.; Gao, C.-J. Preparation of pH-responsive phenolphthalein poly (ether sulfone) membrane by redox-graft pore-filling polymerization technique. J. Membr. Sci. 2007, 287, 257–263. [Google Scholar] [CrossRef]
  20. Ye, Q.; Wang, R.; Chen, C.; Chen, B.; Zhu, X. High-Flux pH-Responsive Ultrafiltration Membrane for Efficient Nanoparticle Fractionation. ACS Appl. Mater. Interfaces 2021, 13, 56575–56583. [Google Scholar] [CrossRef] [PubMed]
  21. Yu, H.-Y.; Li, W.; Zhou, J.; Gu, J.-S.; Huang, L.; Tang, Z.-Q.; Wei, X.-W. Thermo-and pH-responsive polypropylene microporous membrane prepared by the photoinduced RAFT-mediated graft copolymerization. J. Membr. Sci. 2009, 343, 82–89. [Google Scholar] [CrossRef]
  22. Shim, J.K.; Lee, Y.B.; Lee, Y.M. pH-dependent permeation through polysulfone ultrafiltration membranes prepared by ultraviolet polymerization technique. J. Appl. Polym. Sci. 1999, 74, 75–82. [Google Scholar] [CrossRef]
  23. Fan, K.; Huang, J.; Yang, H.; Lu, R.; Sun, X.; Hu, J.; Hou, Z. pH and thermal-dependent ultrafiltration membranes prepared from poly (methacrylic acid) grafted onto polyethersulfone synthesized by simultaneous irradiation in homogenous phase. J. Membr. Sci. 2017, 543, 335–341. [Google Scholar] [CrossRef]
  24. Gao, K.; Kearney, L.T.; Wang, R.; Howarter, J.A. Enhanced wettability and transport control of ultrafiltration and reverse osmosis membranes with grafted polyelectrolytes. ACS Appl. Mater. Interfaces 2015, 7, 24839–24847. [Google Scholar] [CrossRef] [PubMed]
  25. Chen, Y.; Rovira-Bru, M.; Giralt, F.; Cohen, Y. Hydraulic Resistance and Protein Fouling Resistance of a Zirconia Membrane with a Tethered PVP Layer. Water 2021, 13, 951. [Google Scholar] [CrossRef]
  26. Welch, M.E.; Ober, C.K. Responsive and patterned polymer brushes. J. Polym. Sci. Part B Polym. Phys. 2013, 51, 1457–1472. [Google Scholar] [CrossRef]
  27. Zhao, Y.-F.; Zhang, P.-B.; Sun, J.; Liu, C.-J.; Zhu, L.-P.; Xu, Y.-Y. Electrolyte-responsive polyethersulfone membranes with zwitterionic polyethersulfone-based copolymers as additive. J. Membr. Sci. 2016, 510, 306–313. [Google Scholar] [CrossRef]
  28. Ito, Y.; Ochiai, Y.; Park, Y.S.; Imanishi, Y. pH-sensitive gating by conformational change of a polypeptide brush grafted onto a porous polymer membrane. J. Am. Chem. Soc. 1997, 119, 1619–1623. [Google Scholar] [CrossRef]
  29. Birkner, M.; Ulbricht, M. Ultrafiltration membranes with markedly different pH-and ion-responsivity by photografted zwitterionic polysulfobetain or polycarbobetain. J. Membr. Sci. 2015, 494, 57–67. [Google Scholar] [CrossRef]
  30. Gajda, M.; Ulbricht, M. Capillary pore membranes with grafted diblock copolymers showing reversibly changing ultrafiltration properties with independent response to ions and temperature. J. Membr. Sci. 2016, 514, 510–517. [Google Scholar] [CrossRef]
  31. Wang, X.; Zhang, D.; Wu, J.; Protsak, I.; Mao, S.; Ma, C.; Ma, M.; Zhong, M.; Tan, J.; Yang, J. Novel Salt-Responsive SiO2@ Cellulose Membranes Promote Continuous Gradient and Adjustable Transport Efficiency. ACS Appl. Mater. Interfaces 2020, 12, 42169–42178. [Google Scholar] [CrossRef]
  32. Yang, Q.; Adrus, N.; Tomicki, F.; Ulbricht, M. Composites of functional polymeric hydrogels and porous membranes. J. Mater. Chem. 2011, 21, 2783–2811. [Google Scholar] [CrossRef]
  33. Ito, Y.; Park, Y.S.; Imanishi, Y. Nanometer-sized channel gating by a self-assembled polypeptide brush. Langmuir 2000, 16, 5376–5381. [Google Scholar] [CrossRef]
  34. Kostaras, C.; Kati, D.; Christoulaki, A.; Spiliopoulos, N.; Anastassopoulos, D.; Vradis, A.; Toprakcioglu, C.; Priftis, G. Stimuli-responsive polymer brushes in nanoconfined geometry. In Proceedings of the 10th Panhellenic Chemical Engineering Scientific Conference, Patras, Greece, 4–6 June 2015. [Google Scholar]
  35. Chen, Y.; Kim, S.; Cohen, Y. Tuning the hydraulic permeability and molecular weight cutoff (MWCO) of surface nano-structured ultrafiltration membranes. J. Membr. Sci. 2021, 629, 119180. [Google Scholar] [CrossRef]
  36. Chen, Y.; Zhang, J.; Cohen, Y. Fouling resistant and performance tunable ultrafiltration membranes via surface graft polymerization induced by atmospheric pressure air plasma. Sep. Purif. Technol. 2022, 286, 120490. [Google Scholar] [CrossRef]
  37. Chen, Y.; Kim, S.; Kim, Y.; Walker, J.S.; Wolfe, T.; Coleman, K.; Cohen, Y. Scale up of polyamide reverse osmosis membranes surface modification with tethered poly (acrylic acid) for fabrication of low fouling spiral-wound elements. Desalination 2022, 536, 115762. [Google Scholar] [CrossRef]
  38. Chen, Y. Performance Tuning of Ultrafiltration and Reverse Osmosis Membranes Surface Nano-structured with Tethered Poly (Acrylic Acid) Chains; UCLA: Los Angeles, CA, USA, 2022. [Google Scholar]
  39. Kim, S.; Cohen, Y.; Moses, K.J.; Sharma, S.; Bilal, M. Polysulfone surface nano-structured with tethered polyacrylic acid. Appl. Surf. Sci. 2019, 470, 411–422. [Google Scholar] [CrossRef]
  40. Chen, Y.; Cohen, Y. RO membrane with a surface tethered polymer brush layer for enhanced rejection of nitrate, boron, and arsenic. J. Membr. Sci. Lett. 2023, 3, 100062. [Google Scholar] [CrossRef]
  41. Cohen, Y.; Lin, N.; Varin, K.J.; Chien, D.; Hicks, R.F. Membrane surface nanostructuring with terminally anchored polymer chains. In Functional Nanostructured Materials and Membranes for Water Treatment; Wiley: Hoboken, NJ, USA, 2013; pp. 85–124. [Google Scholar]
  42. Levy, R.; Maaloum, M. Measuring the spring constant of atomic force microscope cantilevers: Thermal fluctuations and other methods. Nanotechnology 2001, 13, 33. [Google Scholar] [CrossRef]
  43. Yushkin, A.; Borisov, R.; Volkov, V.; Volkov, A. Improvement of MWCO determination by using branched PEGs and MALDI method. Sep. Purif. Technol. 2019, 211, 108–116. [Google Scholar] [CrossRef]
  44. Toh, Y.S.; Lim, F.; Livingston, A. Polymeric membranes for nanofiltration in polar aprotic solvents. J. Membr. Sci. 2007, 301, 3–10. [Google Scholar]
  45. Hilal, N.; Al-Abri, M.; Al-Hinai, H. Characterization and retention of UF membranes using PEG, HS and polyelectrolytes. Desalination 2007, 206, 568–578. [Google Scholar] [CrossRef]
  46. Milner, S.T. Polymer brushes. Science 1991, 251, 905–914. [Google Scholar] [CrossRef]
  47. Jin, X.; Hsieh, Y.-L. pH-responsive swelling behavior of poly (vinyl alcohol)/poly (acrylic acid) bi-component fibrous hydrogel membranes. Polymer 2005, 46, 5149–5160. [Google Scholar] [CrossRef]
  48. Bryjak, M.; Gancarz, I.; Poźniak, G. Surface evaluation of plasma-modified polysulfone (Udel P-1700) films. Langmuir 1999, 15, 6400–6404. [Google Scholar] [CrossRef]
  49. Chen, T.; Ferris, R.; Zhang, J.; Ducker, R.; Zauscher, S. Stimulus-responsive polymer brushes on surfaces: Transduction mechanisms and applications. Prog. Polym. Sci. 2010, 35, 94–112. [Google Scholar] [CrossRef]
  50. Van Oss, C.J. Interfacial Forces in Aqueous Media; CRC Press: Boca Raton, FL, USA, 2006. [Google Scholar]
  51. Hosseinzadeh, A.; Ranjbar, P.R.; Bozorg, A. pH-responsive P (AA-b-SBMA)-grafted magnetic GO high-performance ultrafiltration membrane with improved antifouling tendency and cleaning efficiency. Desalination 2024, 574, 117299. [Google Scholar] [CrossRef]
  52. Zhao, X.; Su, Y.; Chen, W.; Peng, J.; Jiang, Z. pH-responsive and fouling-release properties of PES ultrafiltration membranes modified by multi-functional block-like copolymers. J. Membr. Sci. 2011, 382, 222–230. [Google Scholar] [CrossRef]
  53. Ma, R.; Lu, X.; Wu, C.; Zhang, S.; Zheng, S.; Ren, K.; Gu, J.; Wang, H.; Shen, H. Performance design of a highly anti-fouling porous membrane with dual pH-responsiveness. J. Membr. Sci. 2022, 660, 120886. [Google Scholar] [CrossRef]
  54. Zhao, X.; Qin, A.; Liu, D.; He, C. Tuning the antifouling property of PVDF ultrafiltration membrane with surface anchored polyelectrolyte complexes for sewage treatment. RSC Adv. 2015, 5, 63580–63587. [Google Scholar] [CrossRef]
  55. Bouranene, S.; Szymczyk, A.; Fievet, P.; Vidonne, A. Influence of inorganic electrolytes on the retention of polyethyleneglycol by a nanofiltration ceramic membrane. J. Membr. Sci. 2007, 290, 216–221. [Google Scholar] [CrossRef]
  56. Zhou, Z.; Murdoch, W.J.; Shen, Y. Synthesis of an esterase-sensitive degradable polyester as facile drug carrier for cancer therapy. J. Polym. Sci. Part A Polym. Chem. 2016, 54, 507–515. [Google Scholar] [CrossRef]
  57. Hoek, E.M.; Ghosh, A.K.; Huang, X.; Liong, M.; Zink, J.I. Physical–chemical properties, separation performance, and fouling resistance of mixed-matrix ultrafiltration membranes. Desalination 2011, 283, 89–99. [Google Scholar] [CrossRef]
  58. Dang, H.T.; Narbaitz, R.M.; Matsuura, T. Double-pass casting: A novel technique for developing high performance ultrafiltration membranes. J. Membr. Sci. 2008, 323, 45–52. [Google Scholar] [CrossRef]
  59. Lai, G.; Lau, W.; Goh, P.; Ismail, A.; Yusof, N.; Tan, Y. Graphene oxide incorporated thin film nanocomposite nanofiltration membrane for enhanced salt removal performance. Desalination 2016, 387, 14–24. [Google Scholar] [CrossRef]
  60. Qasm, N.A.; Generous, M.M.; Qureshi, B.A.; Zubair, S.M. A Comprehensive Review of Saline Water Correlations and Data: Part II—Thermophysical Properties. Arab. J. Sci. Eng. 2021, 46, 1941–1979. [Google Scholar] [CrossRef]
  61. Nayar, K.; Panchanathan, D.; McKinley, G.; Lienhard, J. Surface tension of seawater. J. Phys. Chem. Ref. Data 2014, 43, 043103. [Google Scholar] [CrossRef]
  62. Beattie, J.K.; Djerdjev, A.M.; Gray-Weale, A.; Kallay, N.; Lützenkirchen, J.; Preočanin, T.; Selmani, A. pH and the surface tension of water. J. Colloid Interface Sci. 2014, 422, 54–57. [Google Scholar] [CrossRef] [PubMed]
  63. Lü, Y.; Wei, B. Second inflection point of water surface tension. Appl. Phys. Lett. 2006, 89, 164106. [Google Scholar] [CrossRef]
  64. Fu, D.; Lu, J.-F.; Bao, T.-Z.; Li, Y.-G. Investigation of surface tension and interfacial tension in surfactant solutions by SAFT. Ind. Eng. Chem. Res. 2000, 39, 320–327. [Google Scholar] [CrossRef]
  65. Goncalves, F.; Kestin, J.; Sengers, J. Surface-tension effects in suspended-level capillary viscometers. Int. J. Thermophys. 1991, 12, 1013–1028. [Google Scholar] [CrossRef]
  66. Nikitas, P.; Pappa-Louisi, A. Thermodynamic and modelistic study of surface solutions: Aqueous solutions containing 2-butanol. J. Phys. Chem. 1990, 94, 361–370. [Google Scholar] [CrossRef]
Water 17 00254 g001
Figure 1. Schematics of (a) “through-pore” mechanism and (b) “through-polymer” mechanism of responsive membrane tethered with polyelectrolyte chains.
Figure 1. Schematics of (a) “through-pore” mechanism and (b) “through-polymer” mechanism of responsive membrane tethered with polyelectrolyte chains.
Water 17 00254 g001
Water 17 00254 g002
Figure 2. Schematics of APPIGP surface nano-structuring.
Figure 2. Schematics of APPIGP surface nano-structuring.
Water 17 00254 g002
Water 17 00254 g003
Figure 3. Dependence of the surface captive bubble (CB) contact angle for a Native-PSf and SNS-PAA-PSf membranes on solution ionic strength (0.02–547 mM NaCl) at pH = 7.
Figure 3. Dependence of the surface captive bubble (CB) contact angle for a Native-PSf and SNS-PAA-PSf membranes on solution ionic strength (0.02–547 mM NaCl) at pH = 7.
Water 17 00254 g003
Water 17 00254 g004
Figure 4. Dependence of Native-PSf and SNS-PAA-PSf membranes surface captive bubble (CB) contact angle on solution pH (=3–11).
Figure 4. Dependence of Native-PSf and SNS-PAA-PSf membranes surface captive bubble (CB) contact angle on solution pH (=3–11).
Water 17 00254 g004
Water 17 00254 g005
Figure 5. Membrane surface energy variability with ionic strength and pH for the Native-PSf (MUF-10K) and SNS-PAA-PSf UF membranes (Raw data: Table 1; Note: The Native-PSf membrane is Toray MUF-10K used for comparison, while the SNS-PAA-PSf membrane was synthesized using Base-PSf membrane (Toray MUF-20K); LIS designates low ionic strength < ˜ 0.02 mM).
Figure 5. Membrane surface energy variability with ionic strength and pH for the Native-PSf (MUF-10K) and SNS-PAA-PSf UF membranes (Raw data: Table 1; Note: The Native-PSf membrane is Toray MUF-10K used for comparison, while the SNS-PAA-PSf membrane was synthesized using Base-PSf membrane (Toray MUF-20K); LIS designates low ionic strength < ˜ 0.02 mM).
Water 17 00254 g005
Water 17 00254 g006
Figure 6. Normalized membrane hydraulic resistance (Rmo = Rm/Rm,IS=0, Section 2.7) of Native-PSf and SNS-PAA-PSf membranes over solution ionic strength (~0.02–547 mM NaCl) at pH = 7.
Figure 6. Normalized membrane hydraulic resistance (Rmo = Rm/Rm,IS=0, Section 2.7) of Native-PSf and SNS-PAA-PSf membranes over solution ionic strength (~0.02–547 mM NaCl) at pH = 7.
Water 17 00254 g006
Water 17 00254 g007
Figure 7. Normalized PEG MWCO (MWCOo = MWCO/MWCOIS=0) of Native-PSf and SNS-PAA-PSf membranes over solution ionic strength (~0.02–547 mM NaCl) at pH = 7.
Figure 7. Normalized PEG MWCO (MWCOo = MWCO/MWCOIS=0) of Native-PSf and SNS-PAA-PSf membranes over solution ionic strength (~0.02–547 mM NaCl) at pH = 7.
Water 17 00254 g007
Water 17 00254 g008
Figure 8. Illustration of tethered chains in the proximity of a membrane pore under collapsed (left), swollen (right) and intermediate (middle) configurations, and the corresponding impact on effective membrane pore size and tethered polymer layer thickness. (Note: length of the tethered chains is greater than the base membrane pore diameter).
Figure 8. Illustration of tethered chains in the proximity of a membrane pore under collapsed (left), swollen (right) and intermediate (middle) configurations, and the corresponding impact on effective membrane pore size and tethered polymer layer thickness. (Note: length of the tethered chains is greater than the base membrane pore diameter).
Water 17 00254 g008
Water 17 00254 g009
Figure 9. Effect of pH on membrane hydraulic resistance (expressed as Rmo = Rm/Rm,pH=3) for both the Native-PSf and SNS-PAA-PSf membranes at low ionic strength (0.02–2 mM).
Figure 9. Effect of pH on membrane hydraulic resistance (expressed as Rmo = Rm/Rm,pH=3) for both the Native-PSf and SNS-PAA-PSf membranes at low ionic strength (0.02–2 mM).
Water 17 00254 g009
Water 17 00254 g010
Figure 10. Effect of pH on MWCO (expressed as MWCOo = MWCO/MWCOpH=3) for the Native-PSf and SNS-PAA-PSf membranes at low ionic strength (0.02–2 mM).
Figure 10. Effect of pH on MWCO (expressed as MWCOo = MWCO/MWCOpH=3) for the Native-PSf and SNS-PAA-PSf membranes at low ionic strength (0.02–2 mM).
Water 17 00254 g010
Water 17 00254 g011
Figure 11. Mapping of PEG MWCO and hydraulic permeability for the SNS-PAA-PSf and Native-PSf membranes variability with solution ionic strength (~0.02–547 mM at pH = 7) and pH (3–11 at ionic strength of <2 mM) (Raw data: Table 2).
Figure 11. Mapping of PEG MWCO and hydraulic permeability for the SNS-PAA-PSf and Native-PSf membranes variability with solution ionic strength (~0.02–547 mM at pH = 7) and pH (3–11 at ionic strength of <2 mM) (Raw data: Table 2).
Water 17 00254 g011
Water 17 00254 g012
Figure 12. Selectivity (Lp/MWCO ratio) versus hydraulic permeability coefficient for the SNS-PAA-PSf membrane demonstrating performance above the tradeoff upper bound (set based on literature data for UF membranes [35,36,54,57,58,59]; Table S2). (Grey circles represent SNS-PAA-PSf UF membrane data for different ionic strength and pH solutions (Table 2). (Note: Lp/MWCO values are in the range of ~0.07–0.3 for data points that appear to be on the x-axis).
Figure 12. Selectivity (Lp/MWCO ratio) versus hydraulic permeability coefficient for the SNS-PAA-PSf membrane demonstrating performance above the tradeoff upper bound (set based on literature data for UF membranes [35,36,54,57,58,59]; Table S2). (Grey circles represent SNS-PAA-PSf UF membrane data for different ionic strength and pH solutions (Table 2). (Note: Lp/MWCO values are in the range of ~0.07–0.3 for data points that appear to be on the x-axis).
Water 17 00254 g012
Table 1. Membrane surface captive bubble (CB) water contact angle and surface energy for Native-PSf and SNS-PAA-PSf membranes.
Table 1. Membrane surface captive bubble (CB) water contact angle and surface energy for Native-PSf and SNS-PAA-PSf membranes.
Aqueous EnvironmentSurface
Tension a (mN·m−1)		Native-PSf Membrane bSNS-PAA-PSf Membrane
CB Contact Angle (°)Surface Energy c (mJ·m−2)CB Contact Angle (°)Surface Energy c (mJ·m−2)
Ionic Strength (mM)
~0.0272.850.5 ± 2.5−119.1 ± 2.443.4 ± 2.0−125.7 ± 1.7
272.855.0 ± 1.9−114.6 ± 2.045.6 ± 2.1−123.8 ± 1.9
1772.850.5 ± 1.9−119.1 ± 1.948.1 ± 1.9−121.4 ± 1.8
17173.048.0 ± 1.7−121.9 ± 1.651.2 ± 1.8−118.8 ± 1.8
54773.550.4 ± 1.5−120.4 ± 1.556.1 ± 1.9−114.5 ± 2.0
pH
372.846.4 ± 1.9−123.0 ± 1.741.6 ± 2.1−127.3 ± 1.8
572.844.0 ± 1.8−125.2 ± 1.643.6 ± 1.9−125.5 ± 1.7
772.850.5 ± 2.5−119.1 ± 2.543.4 ± 2.0−125.7 ± 1.8
972.847.7 ± 1.7−121.8 ± 1.636.5 ± 1.4−131.3 ± 1.0
1172.845.4 ± 2.1−123.9 ± 1.924.3 ± 1.8−139.2 ± 1.0
a The surface tensions for the listed solutions are provided in Table S1 (Supplementary Material). b The Native-PSf membrane is a commercial PSf UF membrane (Toray MUF-10K) with similar water permeability and MWCO as the SNS-PAA-PSf membrane, and thus this membrane was used for comparison with the SNS-PAA-PSf membrane. It is noted that the SNS-PAA-PSf membrane was synthesized on another Base-PSf membrane (Toray MUF-20K) with significantly higher water permeability and MWCO. c Membrane surface energy was calculated as per the protocol described in Section 2.5.
Table 2. Membrane hydraulic permeability coefficient and PEG MWCO for Native-PSf and SNS-PAA-PSf membranes.
Table 2. Membrane hydraulic permeability coefficient and PEG MWCO for Native-PSf and SNS-PAA-PSf membranes.
Native-PSf Membrane (a)SNS-PAA-PSf Membrane (b)
Lp
(L·m−2·h−1·bar−1) (c)		Rm (m−1) (d)Rmo (e)MWCO
(kDa) (f)		MWCOo (e)Pore Size
(nm)		Lp
(L·m−2·h−1·bar−1)		Rm (m−1)Rmo (e)MWCO
(kDa) (f)		MWCOo (e)Pore Size
(nm)
Ionic strength (mM) (g)
~0.0223.21.55 × 10131.0010.01.002.920.71.74 × 10131.008.01.002.7
223.21.55 × 10131.0010.11.012.922.71.59 × 10130.917.60.952.6
1723.11.55 × 10131.0010.41.042.923.81.51 × 10130.877.20.902.6
17122.31.59 × 10131.0211.41.143.024.51.44 × 10130.834.40.552.2
54721.11.61 × 10131.0415.31.533.327.21.24 × 10130.721.80.231.6
pH (h)
326.21.37 × 10131.0014.61.003.348.77.38 × 10121.006.91.002.5
525.11.44 × 10131.0513.80.953.237.49.62 × 10121.307.91.152.7
723.21.55 × 10131.1310.00.692.920.71.74 × 10132.368.01.162.7
923.81.51 × 10131.1010.50.722.917.72.03 × 10132.759.61.392.8
1124.91.45 × 10131.0516.81.153.415.72.29 × 10133.1115.02.173.3
(a) The Native-PSf membrane is a commercial PSf UF membrane (Toray MUF-10K) with similar water permeability and MWCO as the SNS-PAA-PSf membrane and thus was used in the present study for comparison. It is noted that the SNS-PAA-PSf membrane was synthesized on another Base-PSf membrane (Toray MUF-20K) with significantly higher water permeability and MWCO. (b) The SNS-PAA-PSf membrane is synthesized based on a commercial PSf UF membrane (Toray MUF-20K) surface modified by atmospheric pressure He/O2 plasma treatment (PSS = 15 mm, N = 2), followed by AA graft polymerization ([M]o = 20 vol%, 70 °C, 1 h). (c) UF membrane hydraulic permeability coefficient measured following the protocol described in Section 2.7. (d) UF membrane hydraulic resistance (Section 2.7). (e) Normalized membrane hydraulic resistance and MWCO. (i) Rmo = Rm/Rm, pH=3 and MWCOo = MWCO/MWCOpH=3 for tests in which pH was varied at low ionic strength (IS < 2 mM (ii) Rmo = Rm/Rm, IS=0.02mM and MWCOo = MWCO/MWCOIS=0.02mM for tests at pH = 7. (f) UF membrane PEG MWCO determined using a series of PEG Mw fractions, following the protocol in Section 2.7. (g) Aqueous solution ionic strength was adjusted with NaCl, and tests were carried out at pH = 7. (h) Aqueous solution pH was adjusted with 0.1N HCl and 0.1 N NaOH solution.
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Chen, Y.; Cohen, Y. Ionic Strength and pH-Responsive Ultrafiltration Membrane to Overcome the Typical Permeability-Selectivity Tradeoff. Water 2025, 17, 254. https://doi.org/10.3390/w17020254

AMA Style

Chen Y, Cohen Y. Ionic Strength and pH-Responsive Ultrafiltration Membrane to Overcome the Typical Permeability-Selectivity Tradeoff. Water. 2025; 17(2):254. https://doi.org/10.3390/w17020254

Chicago/Turabian Style

Chen, Yian, and Yoram Cohen. 2025. "Ionic Strength and pH-Responsive Ultrafiltration Membrane to Overcome the Typical Permeability-Selectivity Tradeoff" Water 17, no. 2: 254. https://doi.org/10.3390/w17020254

APA Style

Chen, Y., & Cohen, Y. (2025). Ionic Strength and pH-Responsive Ultrafiltration Membrane to Overcome the Typical Permeability-Selectivity Tradeoff. Water, 17(2), 254. https://doi.org/10.3390/w17020254

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop