Performance Evaluation of 1-Cyclohexylpiperidine as a Draw Solute for Forward Osmosis Water Separation and CO2 Recovery.

Membrane-based technologies, such as forward osmosis (FO), offer the advantage of treating water through a spontaneous process that requires minimal energy input while achieving favorable water permeability and selectivity. However, the FO process still has some challenges that need to be solved or improved to become entirely feasible. The main impediment for this technology is the recovery of the draw solute used to generate the osmotic potential in the process. In this paper, we discuss the use of a switchable polarity solvent, 1-cyclohexylpiperidine (CHP), as a draw solute that responds to external stimuli. Specifically, the miscibility of CHP can be switched by the presence of carbon dioxide (CO2) and is reversible by applying heat. Thus, in this study, the hydrophobic CHP is first converted to the hydrophilic ammonium salt (CHPH+), and its capability as a draw solution (DS) is thoroughly evaluated against the typical osmotic agent, sodium chloride (NaCl). Our results show that the water permeability across the thin film composite membrane increases by 69% when CHPH+ is used as the DS. Also, the water permeability when using different feed solutions: aqueous solutions of (a) urea and (b) NaCl were evaluated. In both cases, the CHPH+ generates water fluxes in the range of 65 ± 4 LMH and 69 ± 2 LMH, respectively. We then separate the diluted DS by applying 75 °C to the solution to recover the pure CHP and water. The results of this work provide a proof-of-concept of a CHP wastewater and desalination method via an FO process.

Introduction

Wastewater reclamation and seawater desalination are two important sources of obtaining fresh water. However, current technologies to reclaim such water require the use of energy-intensive methods. One example is reverse osmosis (RO), which operates at high pressures (>50 atm), recovers only 35–50% of fresh water, and produce large volumes of waste brine.[1,2] In the past years, forward osmosis (FO) has emerged as a viable technology for low-cost desalination and wastewater treatment.[3,4] FO allows the movement of clean water from a feed solution across a semipermeable membrane. The driving force for FO is the osmotic potential that is generated between the feed and the draw solution (typically a hypertonic solution, see Scheme ). FO is a low-energy process because it operates at low pressure and yields high water permeance across the membrane. The efficiency of the overall FO process largely relies on two aspects: (1) the membrane selectivity and water permeability, and (2) the osmotic potential of the DS employed. The challenges for the FO process involve the availability of membranes and the osmotic strength of the DS.

Scheme 1

Representation of the FO Process, Including Post Treatment of the Draw Solution to Recover Draw Solute and Clean Water

Thin film composite (TFC) membranes, based on polyamide as the selective layer, are considered at the forefront due to their high water permeability and good solute rejection. These membranes are composed of a polysulfone porous support and a highly cross-linked polyamide thin layer.[5−7] On the other hand, numerous draw solutes have been studied for the FO process, among them, inorganic salts (e.g., NaCl) are the most widely used due to their high osmotic potential, abundance, and low cost.[8,9] The use of inorganic salts requires desalination or the removal of the draw solutes post-purification. Therefore, finding new draw solutes that can be easily regenerated without the need of an energy-extensive separation process is important for the above-mentioned applications.

One of the uniqueness of the FO process is its ability to produce the required driving force from different draw solutes. Stimuli responsive materials are considered the next generation of draw solutes and these include: thermo-responsive polymers,[10] zwitterions,[11] magnetic nanoparticles,[12,13] polyelectrolytes,[14] and switchable polarity solvents (SPS).[15] An SPS is a solvent with the ability to exist in two different forms; one that is hydrophobic and has low miscibility in water, while the other is hydrophilic and highly miscible in the aqueous phase (Scheme ). The switch between these two forms is achieved by disrupting the equilibrium between the two forms. Amongst the previously studied SPS, tertiary amines have been identified to switch between the two forms by the addition or removal of CO2 from the system.[16] It is well known that CO2, due to its partial solubility in aqueous media, forms carbonic acid when dissolved in water. This in turn creates the scenario for the tertiary amines to promote an acid–base reaction (eq ), resulting in the form of an SPS-ammonium salt complex (i.e., the hydrophilic form).[17] As such, Le Chatelier’s principle dictates that the reversed equilibrium to such reaction causes the CO2 molecules to be released from the solution resulting in the recovery of the pure SPS and water. The cited literature suggests that the equilibrium disruption in these systems may be achieved by applying stimuli such as vacuum and/or heat.[18]

Scheme 2

Representation of the SPS Change in Miscibility Caused by the Presence or Absence of CO2

Among the tertiary amine-based SPS, 1-cyclohexylpiperidine (CHP) is an interesting case study due to its high stability, high osmotic potential in their ammonium salt form, and compatibility with the TFC membranes.[19] Also, the high carbon to nitrogen ratio (11:1) in CHP is beneficial as a draw solute since its hydrophilic nature promotes high osmotic potentials,[18] and it has a boiling point that may help in preventing significant losses during the switching process.[17,18]

In this work, the typical TFC membranes were fabricated, and their FO performance was evaluated using sodium chloride and the hydrophilic form of the CHP as draw solutes. As part of the FO evaluation, we studied different feed solutions to determine water permeability, membrane selectivity, and reverse solute flux. This work provides a comprehensive study on the tertiary amine CHP as the draw solute and its potential when combined with a TFC membrane for water treatment. The combination of TFC membranes with the tertiary amine SPS may improve the FO efficiency, primarily because it requires a lower-energy input (i.e., low temperature) to recover the draw solute and water. It also expands the FO scope in terms of its feasibility for water treatment as it can be used with complex feed solutions without affecting the driving force and still obtaining high water permeability. Specifically, it could be used directly as a desalination method, and unlike the RO, it would not require pressurization.

Results and Discussion

Evaluation of the Parameters Required for Switching the Equilibrium of CHP

As already established, tertiary amines can be protonated via an acid–base reaction with carbonic acid leading to a phase separation (i.e., from two phases to one). Moreover, in the presence of water and carbon dioxide, tertiary amines are converted into water-soluble ammonium salt.[20] Thus, the effect of different reaction parameters to the phase separation rate was first evaluated. In order to conduct these experiments, the setup in Figure was utilized with a 1:1 CHP to water ratio as previously published.[17] The results in Figure S1 include a control sample (A), where no CO2 was purged through the solution and the experimental sample (B), with an input of CO2 at a flow rate of 3 mL/min. As can be seen in the control sample, the volume of both phases remains intact over time. However, when CO2 is added to the experimental sample, the two phases merge together with at least 1 h of contact time (Figure S1B). This suggests that the phase change using pure CO2 requires a cutoff time of 1 h in order for the CHP to fully interact to form H2CO3. After the conversion starts, the average switch rate was 0.17 ± 0.03 mL/h, and after 30 h, approximately 0.8 mL of CHP remains unreacted (15% of the amine). This experiment verifies that CO2 is needed to obtain a single phase, though it also reveals that the switch time is relatively high.

Figure 5

Experimental setup for the switchability experiments using CHP.

Thereafter, the effect of the amine to water ratio to the switch rate was evaluated with the intent of optimizing the phase change process (Figure S2). In this study, the two limits were first evaluated: (1) excess of amine in the 2:1 ratio (Figure S2A) and (2) excess of water in the 1:2 ratio (Figure S2B). These two experiments were performed with a CO2 flow rate of 3 mL/min as previously executed. A closer look at these results reveals that when the amine is in excess, the switch rate is 2 times slower than when run with a 1:1 ratio. After 24 h, approximately 1.9 mL of amine remains unreacted, meaning that 25% of the amine is able to react. Meanwhile, when the water is in excess, 90% of the amine is in the hydrophilic form after 8 h. This could indicate that when the water is in excess the reaction is faster and more efficient.

Another parameter that may be optimized is the gas flow rate. In all previous experiments it remained constant at 3 mL/min. Thus, higher CO2 flow rates (10 and 30 mL/min) were studied for comparison. Results in Figure S3 show that the switch rate increases with a higher CO2 flow rate. This is likely due to the solution becoming saturated in less time at a higher flow rate, which translates into a faster phase change process. Interestingly, at a flow rate of 30 mL/min (sample B) 90% of the amine was switched after 8 h, and after 24 h the switch was completed. Similar to previous experiments, the reaction started after a 1 h breakthrough.

After completing these experiments, we were aiming to increase the reaction rate and reduce the overall time to convert CHP to CHPH+. In an effort to achieve this, the sample volume was increased and a glass frit was used for CO2 bubbling to improve mass transfer.[17] However, the trade-off is that the glass frit increases the gas pressure, necessitating a higher flow rate of CO2 (i.e., 800 mL/min). The results show that after 3 h of reaction, 50% of the amine is converted, and after 6 h, 10% of the CHP remain unreacted (Figure S4). Therefore, for further CHP conversion experiments, these optimized parameters were used: a 1:1 amine to water ratio, a larger sample container, gas frit, and a flow rate of 800 mL/min of CO2.

The chemical composition of the pure CHP and CHPH+ can be evaluated using FTIR, where distinct vibration patterns for each sample is observed, see Figure . The CHP spectrum shows characteristic vibration bands present within the CHP molecule, a C–H stretch in the range ca. 3100 cm–1, and at lower wavenumbers C–H, C–N, and C–C stretch peaks are observed. In the CHPH+ spectrum, a new strong and broad band is present in the range of 3690–3000 cm–1, where there is an overlap between the N–H stretch from the ammonium salt and the O–H stretch from −COOH. The new band at ca. 1620 cm–1 corresponds to the antisymmetric stretch of the COO– group in the bicarbonate compound. Additionally, the conductivity of pure CHP was 10.05 μS/cm, while the CHPH+ increased by a 1000 times factor to 15.24 mS/cm. This significant increase is attributed to the new ionic species in the solution (i.e., CHPH+, COO–, H3O+, and HO–). These results corroborate the chemical transition from CHP to CHPH+.

Figure 1

FTIR spectra comparison of pure CHP (black line) and after adding carbon dioxide, CHPH+ solution (red line).

We then proceeded to study the reverse switch (i.e., from the CHPH+ to CHP and water). In order to do so, we immersed the CHPH+ solution into a water bath adjusted to 75 °C and monitored the process. In Figure A, the increase in volume in the organic phase (i.e., CHP, top layer) is notable, while the aqueous phase (i.e., water, bottom layer) changes from dark to light yellow with time. After 3 h of conversion, the process was intentionally stopped and a characterization of the chemical composition of both layers was executed. We used FTIR to analyze both layers (Figure B). The organic phase spectrum exhibits the same vibration bands as that of pure CHP, suggesting the separation of pure CHP after 3 h. The aqueous phase shows two bands. The band in the 3700–2900 cm–1 range could represent an overlap of the following vibrations: (a) O–H stretch in water, (b) O–H stretch in bicarbonate, and (c) N–H stretch in the ammonium salt. The characteristic bands for C–H or N–H stretching are not present in the spectrum, suggesting that there is no CHPH+ in the solution. However, a sharp vibration band ca. 1630 cm–1 could indicate the presence of COO– from the bicarbonate, which can be verified by the pH values. The CHPH+ solution has a basic pH of 10.6 due to the presence of the bicarbonate, whereas the aqueous phase has a pH of 9.8, which can also suggest the presence of bicarbonate in the solution. These results suggest that the separation at these parameters (3 h at 75 °C) was completed successfully, and no CHPH+ remained present.

Figure 2

Conversion from CHPH+ to CHP and water. (A) Phase separation with time: upper layer is the organic phase, and bottom layer is the aqueous phase. (B) FTIR comparison; CHPH+ starting solution, organic phase, and aqueous phase after 3 h of separation.

Forward Osmosis Evaluation

The forward osmosis process has been widely studied for wastewater reclamation as a passive mode of separation that may be advantageous compared to other technologies.[3,4,9] However, salt ions have dominated as the preferred draw solute due to its accessibility and high osmotic potential. The main concern with the use of hypertonic solutions is the post-treatment of the diluted draw solution (i.e., desalination process). Thus, new and novel draw solutes that require less energy to be separated from the reclaimed water are needed.[18] We have demonstrated to be able to separate the CHP and water from the CHPH+ mixture by applying heat for short periods of time. However, the potential of CHPH+ to be employed as a draw solution with a custom-made TFC membrane has not been demonstrated. In order to demonstrate this, an FO flow cell was employed with different feed solutions and run for a period of 1 h. FO experiments using NaCl as a draw solute for comparison in efficiency were also conducted as part of this study. All the experimental combinations and results are summarized in Table .

Table 1

Studied Combination of Solutions where FS and DS Are the Feed and Draw Solutions, Respectivelya

combination (C)	FS/DS	JW ± SD (LMH)	JS ± SD (GMH)	JS/JW ± SD (g/L)
C1	Npw/NaCl	22.7 ± 0.8	1.4 ± 0.8	0.6 ± 0.03
C2	urea/NaCl	17 ± 4	1.1 ± 0.3	0.6 ± 0.03
C3	Npw/CHPH+	38 ± 4	2.1 ± 0.8	0.6 ± 0.02
C4	urea/CHPH+	65 ± 4	2.6 ± 0.5	0.3 ± 0.02
C5	NaCl/CHPH+	69 ± 2	0.7 ± 0.3	0.010 ± 0.005

The FS used were nanopure water (Npw), 1.3% w/v of urea, and 1.0% w/v NaCl; and the DS were 5.0% w/v NaCl and the mixture of CHPH+. Jw represents the water flux, and Js Is the reverse flux of the draw solute (i.e., salt or CHPH+).

First, the water permeability and reverse salt flux of the TFC membrane using a 5% w/v NaCl solution as draw solution were evaluated using different feed solutions. When nanopure water (Npw) was used as the feed solution (C1), the membrane reached 22.7 ± 0.8 LMH and 1.4 ± 0.8 GMH. Next, the feed solution was changed to a 1.3% w/v urea solution (C2) to study membrane selectivity. Urea is typically used because it is a small, uncharged molecule, and it also provides a source of carbon to measure any crossover. The water flux through the TFC membrane was 17 ± 4 LMH, which represents a 24% decrease in comparison with nanopure water. The reverse salt flux also decreased to 1.1 ± 0.3 GMH (23% decrease). These results suggest one of the following: (a) The efficiency of the TFC membrane decreases when a model molecule such as urea is added to the feed solution, or (b) the osmotic gradient is reduced by the presence of urea in the feed solution, which translates to a lower ability for the water to permeate through the membrane.

The specific reverse solute flux (Js/Jw) was determined as this parameter provides an insight into membrane selectivity because it quantifies the draw solute leakage per water permeated. Using both feed solutions, the average Js/Jw was 0.06 ± 0.03 g/L. The Js/Jw was kept constant while varying the FS, suggesting that the membrane performance was not affected by the different compounds in the FS studied. Therefore, the decrease in water permeability when including urea in the FS can be attributed to the draw solute, NaCl, which the performance can be easily affected by smaller differences in the osmotic pressure difference across the transmembrane. This could represent an additional problem to the overall FO process when employed with complex feed solutions.

Next, the CHPH+ as the draw solution and three different feed solutions were studied: (C3) nanopure water, (C4) 1.3% w/v urea solution, and (C5) 1.0% w/v NaCl solution. In these FO runs, the fabricated TFC membrane performance increased significantly with all the FS. First, the Jw obtained when water was the FS (C3) was 38 ± 4 LMH, which is 69% higher water permeability than using NaCl as DS. This first result indicates that the osmotic potential of CHPH+ is approximately 1.7 times better than the 5% NaCl solution. In C4 we studied the FO using urea as the FS, and the water flux increased to 65 ± 4 LMH (ca. 1.7 times higher than C3), which is the opposite trend than when using NaCl as DS. Finally, in C5 we used NaCl as the FS to evaluate the potential of the CHPH+. Interestingly the membrane reached a water flux of 69 ± 2 LMH, which is very similar to that obtained in C4. Also, the Js/Jw was calculated, and the obtained values are good indicators that the membrane operates efficiently with CHPH+ as DS. These are promising results since the osmotic potential of CHPH+ is not affected by the presence of an uncharged molecule (i.e., urea) or the salt in the FS.

For a better understanding on the TFC membrane performance difference when the DS is changed, we proceeded to evaluate the membrane morphology using SEM. As showed in Figure , the front side of the membrane exhibits the typical polyamide ridge and valley morphology. This morphology was not affected after the FO experiments (C3–C5). The back side of the membrane exhibited the polyamide morphology, which suggests that the amide polymerization was completed through all the membrane pores. Also, there is no noticeable difference between the membrane before or after CHPH+ exposure. Finally, we studied the membrane cross section where a finger-like morphology was obtained and remained after the FO experiments.

Figure 3

SEM micrograph comparison of the fabricated TFC membrane (A) before and (B) after FO experiments using CHPH+ as the DS (C3–C5).

The membrane morphology seems intact after being exposed to the CHPH+ solution in the FO experiments. However, we studied the wettability properties of the front side of the membrane to determine if there was a significance change. The TFC membrane before being used for the FO experiments had a contact angle of 40.0 ± 3.7 °, which decreased to 33.2 ± 2.0 ° after the FO runs. These results indicate that there was a slight increase in the membrane hydrophilicity. This suggests that some interactions between the molecules in the FS and the membrane surface occurred. To validate this, the surface free energy (SFE) of the membranes was determined. The SFE is the energy that arises from intermolecular interactions at the studied interface. The membrane before the FO run had an SFE of 58.9 ± 2.0 J m–2 and increased to 62.5 ± 1.0 J m–2 after the run. These results can be associated with the obtained trend in the FO experiments. The membrane performance increased significantly when CHPH+ was used as the DS and so did the SFE, which suggests that chemical interactions between the solutes (i.e., urea and NaCl in the feed side, and CHPH+ in the draw side) and the polyamide layer at the surface are occurring.

Finally, after the FO experiments (C3–C5) using CHPH+ as the DS, both solutions, FS and DS, were examined via FTIR to determine the chemical integrity after the process. In Figure A a comparison of the FTIR spectra of all the studied feed solutions is shown. The same trend as the reverse solute flux (Js) was obtained, where no presence of the amine is observed. Figure B shows the spectra of the diluted draw solution after each FO run. Even though the intensity was slightly reduced, these spectra present the same vibration bands as that of CHPH+ in Figure . Also, the conductivity of all three diluted DS after the FO runs was measured and compared to the original conductivity measurement of CHPH+, which was 15.54 mS/cm. The conductivity of the draw solution from C3, when water was used in the feed side, increased to 21.24 mS/cm. We suggest that the water that permeates to the draw side promotes the formation of additional CHPH+ and HCO3–, which translates to a higher conductivity. A similar increase was noted in C4 (17.61 mS/cm) and C5 (25.74 mS/cm).

Figure 4

FTIR spectra comparison of the (A) feed solutions (C3: nanopure water, C4: urea solution, and C5: NaCl), (B) draw solution after the FO experiments with CHPH+ as the draw solute, (C) aqueous phase, and (D) organic phase after the conversion of CHPH+ to CHP and water.

As a proof-of-concept, the recovery of water and pure CHP by heating was studied, and the diluted draw solutions were held at 75 °C for 4 h after each experiment. The recovered phases (i.e., organic: CHP and aqueous: water) were analyzed with FTIR and compared to the results in Figure and Figure B. Results in Figure C,D suggest that the separation of the layers was successful, which demonstrates the potential of employing CHP as a draw solute for different applications, including desalination.

Conclusion

We were able to study and optimize the conditions to convert the organic and hydrophobic CHP to an aqueous and hydrophilic solution of CHPH+. We compared the FO performance of the fabricated TFC membrane using different combinations of feed and draw solutions. The results showed that the membranes improved their performance when CHPH+ was used as a draw solute by 69%, and their efficiency was not affected by the different molecules (i.e., urea and NaCl) studied in the feed solution. After the FO performance was evaluated, we switched back the amine to its original organic and hydrophobic form by heating the CHPH+ solution. This resulted in a successful separation, where pure CHP and water can be recovered. Hence, we demonstrated, as a proof-of-concept, that CHP can be used as a feasible draw solute, and a desalination process can be completed via an FO process.

Materials and Methods

Materials

Polysulfone (PSF, average Mn ≈ 22,000), N-methyl-2-pyrrolidone (NMP, 99%), m-phenylenediamine (MPD, 99%), trimesoyl chloride (TMC, 98%), hexane (anhydrous 95%), sodium chloride (NaCl, ACS reagent 99.0%), and urea (ACS reagent 99.0–100.5%) were all purchased from Sigma–Aldrich. 1-Cyclohexylpiperidine (CHP, 97%) was purchased from Alfa Aesar. All chemicals and solvents were used as received and without further purification. The polyester mesh (PE), 105 micrometer–52% open area, was purchased from Elko Filtering Co. Nanopure water (18.2 MΩ·cm2) was used at all times.

Methods

Evaluation of the Switchable Polarity Solvent: 1-Cyclohexylpiperidine

We studied the 1-cyclohexylpiperidine (CHP) conversion to the hydrophilic form and evaluated the effect of different parameters to the switch rate. In order to do so, we used a custom-made system (Figure ) composed of (1) a graduated container to monitor the change in volume with time, (2) a condenser adjusted to 2–4 °C to avoid loss of the phases, and (3) a stainless steel or glass carbon dioxide input with a constant and controlled gas flow rate from a pressurized tank.

Several sets of experiments were performed to compare the effect of all the studied parameters: (1) presence of CO2, (2) amine to water ratio (v:v), (3) CO2 gas flow rate (mL/min), and (4) gas input method; see Table for the details of each studied parameter. In brief, CHP and water were added to the graduated cylinder, the condenser temperature was adjusted, and the pure CO2 was purged.

Table 2

Summary of Studied Conditions to Perform the Switchability Experiments of CHP

parameter	studied conditions
CO2 input	On/Off
CHP:Nanopure water ration (v/v)	2:1, 1:1, 1:2
CO2 flow rate (mL/min)	3, 10, 30, 800
CO2 input method	1 mm tube, inlet with porous frit

After obtaining one phase (i.e., CHPH+), we studied the capability to switch back to the two phases (i.e., CHP and water) by increasing the temperature to 75 °C. We used the same setup but immersed the solution container in a water bath.

Chemical Characterization of the CHP and CHPH+

All liquid samples were analyzed, and the infrared spectra were recorded on a Bruker Alpha Platinum-ATR spectrometer using a transmittance mode. A 500 μL drop was analyzed; the spectral width ranged from 4000 to 600 cm–1, with a 4 cm–1 resolution, and an accumulation of 64 scans. Conductivity and pH measures were obtained using a YSI multiparameter meter with a glass electrode.

Fabrication of Polysulfone Thin Film Composite (TFC) Membranes

The polysulfone support was fabricated via the non-solvent-induced phase separation (NIPS) process following our previous work.[21] The casting solution was prepared by dissolving PSF in NMP (12% w/v) at room temperature. After complete dissolution, the PSF was cast over the PE mesh, and the thickness of the film was adjusted to 150 μm using a doctor blade. Thereafter, the film was immersed in a nanopure precipitation bath. The support was then modified with polyamide via interfacial polymerization. First, the support was soaked in an aqueous solution of MPD (2% w/v) for 2 min, and the excess was removed with an air knife. Then, the support was soaked in a TMC solution in hexane (0.1% w/v) for 1 min, and the excess was removed using an air knife. Finally, the membrane was immersed in an aqueous sodium carbonate solution (0.2% w/v) for 5 min, rinsed, and stored in nanopure water.

Evaluation of the Forward Osmosis Performance

The water permeability of the membranes was tested with a custom-made flow cell using nanopure water, urea solution (1.3% w/v), and NaCl solution (aq 1.0% w/v) as the feed solutions (FS), while a mixture of NaCl solution (aq 5.0% w/v) and CHPH+ solution was used as the draw solution (DS). The exposed membrane area was 81.75 cm2; the active layer was facing the feed solution, and the results were collected at a constant flow rate at room temperature for 1 h. The water fluxes, Jw in LMH (L·m–2·h–1), were calculated using the following equation:where ΔV is the change in volume in the draw solution in L, AM is the active membrane area, and t is the time of the test in hours.

The reverse salt, or solute flux, Js, in GMH (g·m–2·h–1), from the draw solution to the feed solution was calculated using eq :where C and V are the salt or solute concentration and the feed volume at the end of the FO experiments, respectively. The concentration of salt in the feed was determined by ion chromatography, using a cation column (CS12) and anion column (AS4A) with a conductivity detector, and preparing a calibration curve with a standard solution of ions. The solute concentration (i.e., carbon-based) in the feed side was quantified using a total organic carbon (TOC) analyzer from Shimadzu model TOC-LCCH.

Membrane Characterization after FO Experiments

Scanning electron microscopy (SEM) images were taken with a Hitachi S-4800 field emission SEM with an accelerating voltage of 10.0 KV and a current of 5 μA. Air-dried samples were attached onto a sample holder and sputtered with a palladium film (ca. 5 nm-thick). Contact angle measurements were carried out using a Krüss drop shape analyzer DSA25S (Krüss Optronic, Hamburg, Germany) at room temperature. Air-dried membranes were cut to obtain a 1 cm2 coupon that was fixed to a glass plate using carbon tape. For the analysis, a 4.50 μL nanopure water droplet was released from a syringe with a 25 gauge flat needle (0.51 mm inner diameter and 0.26 mm outer diameter) onto the surface of the sample. The data was recorded every 0.5 s up to 120 s and analyzed in real time using Advance software (version 1.8).