‘Up-scaling’ potential for polyelectrolyte multilayer membranes

Author's Accepted Manuscript

‘Up-scaling’ Potential for Polyelectrolyte Multilayer Membranes Nithya Joseph, Pejman Ahmadiannamini, Pulluru Sai Jishna, Alexander Volodin, Ivo. F. J Vankelecom

www.elsevier.com/locate/memsci

PII: DOI: Reference:

S0376-7388(15)00472-X http://dx.doi.org/10.1016/j.memsci.2015.05.042 MEMSCI13724

To appear in:

Journal of Membrane Science

Received date: 11 February 2015 Revised date: 21 May 2015 Accepted date: 23 May 2015 Cite this article as: Nithya Joseph, Pejman Ahmadiannamini, Pulluru Sai Jishna, Alexander Volodin, Ivo. F. J Vankelecom, ‘Up-scaling’ Potential for Polyelectrolyte Multilayer Membranes, Journal of Membrane Science, http://dx.doi.org/ 10.1016/j.memsci.2015.05.042 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

‘Up-scaling’ Potential for Polyelectrolyte Multilayer Membranes Nithya Josepha, Pejman Ahmadiannaminib, Pulluru Sai Jishnaa, Alexander Volodinc, Ivo. F. J Vankelecoma* a

Centre for surface chemistry and catalysis, Faculty of Bioscience Engineering, K.U.Leuven, Kasteelpark

Arenberg-23, P.O.Box 2461, 3001 Leuven, Belgium. Fax: +3216321998

Tel: +3216321594 E-mail:

[email protected] (I.F.J. Vankelecom). b

Department of Civil and Environmental Engineering, Michigan State University, East Lansing, MI 48824, USA

c

Laboratory of Solid-State Physics and Magnetism, Department of Physics and Astronomy, K.U. Leuven,

Celestijnenlaan 200D, B-3001 Leuven, Belgium

Abstract The layer-by-layer method is an attractive technique for preparing ultrathin nanostructured polyelectrolyte multilayer membranes (PEMMs) with tailored composition and tunable properties. This paper investigates the optimization of the membrane preparation to render it more feasible from a practical view point. PEMMs were prepared with minimal number of bilayers and rinsing steps for two polyion combinations: poly(diallyldimethylammonium chloride) with poly (vinylsulfonic acid sodium salt) and the same cation with poly(sodium 4styrenesulfonate). Three bilayers proved optimal for SRNF applications. The commonly applied rinsing process turned out to be non-essential. In addition, two minutes dipping time per deposition step already proved sufficient for defect-free SRNF membrane preparation. Thus, the build-up protocol for PEMMs was overall reduced to a 6-step coating procedure taking 12 minutes without compromising the SRNF properties, in contrast to the previously reported 20 to 80 steps which required upto 7 hours. Therefore, the ‘up-scaling’ potential of this type of membrane is significantly increased and potential production costs reduced. Keywords: Polyelectrolyte. Multilayer. Membrane. Layer-by-layer technique. Solvent resistant nanofiltration 1. Introduction Preparation of thin multilayer films using layer-by-layer technique (LBL) is reported as a versatile method to produce thin nanostructured membranes [1]. The LBL method involves alternate dipping of a charged substrate into oppositely charged polyelectrolytes (PEs) followed by washing after each deposition as schematically shown in Fig1[2]. Each adsorption step then leads to a charge inversion of the substrate, thus forming a layered complex consisting of several bilayers. These layered membranes are formed and stabilized with the help of electrostatic interactions between the charged layers [3]. These self-

assembled PE membranes are unique, thanks to their strong internal coloumb interactions and the tuning facility of their electrostatic nature by adapting the composition and characteristics of individual PE constituents. In addition to electrostatic interactions, hydrophobic interactions [4], hydrogen bonding [5, 6], and co-ordination chemistry [7] play an important role in the LBL method. The popularity of this method lies in its simplicity, precise thickness control and its tunability based on composition. Moreover, this method has a broad processing window allowing the film properties to be customized by controlling different parameters, such as the type of PEs [8], the charge density of polymers [9], possible optimization of various deposition conditions including ionic strength [10], pH [11] and temperature [12]. Additionally, a wide variety of charged substrates with different geometries, and possibility of post-assembly modifications, like grafting [13], crosslinking [14], hybridization [15] and annealing [16] further improve the separation properties. The use of these PE multilayer membranes (PEMMs) has been explored in different membrane processes, like pervaporation (PV) [17], nanofiltration (NF) [18], reverse osmosis (RO) [19], gas separation (GS) [20] and solvent resistant nanofiltration (SRNF) [21]. The latter is a recently developed technology for the separation of organic mixtures down to the molecular level by applying a pressure gradient [22]. SRNF membranes should preserve their separation properties in aprotic or strongly swelling solvents, extreme temperatures and other aggressive conditions [23, 24]. To date, SRNF membranes based on PEMs have been prepared successfully from PE pairs, such as sulfonated poly(etheretherketone) (SPEEK) /poly(diallyldimethylammonium chloride)(PDDA) [21], (PDDA/Polyacrylicacid) (PAA) [25], (PDDA/Polyvinyl sulfonic acid) (PVS) [26], (PDDA/Polystyrene sulfonate) (PSS) [26], on hydrolyzed polyacrylonitrile supports (HPAN) and PDDA/SPEEK on charged silicon composite/H-PAN supports [27]. These membranes are generally utilized for the separation of organic dyes from solvent. In all cases, Donnan exclusion was a key factor, while size exclusion played a less important role in the separation performance. All membranes mentioned above showed promising solvent stability and separation characteristics in organic solvents, including tetrahydrofuran (THF) and N,N-dimethylformamide (DMF). These are generally quite troublesome in SRNF, since they have a strong dissolving/swelling effect effect on most polymers that form membranes. Even though PEMMs prepared by the LBL method have immense potential in the different areas of membrane separations, these membranes are not yet commercially available. This is mainly due to the cumbersome and time-consuming preparation procedure for these membranes. Previous studies have shown that a large amount of PE deposition steps are required to obtain selective and defect-free membrane, sometimes even up to 100 coating and rinsing steps [28]. Any increase in the number of bilayers further limits the permeability and the advantages of the LBL method. There are few membranes reported with low numbers of bilayers [29, 30, 31]. However, these are prepared under special deposition conditions, such as by using vibration [32], in the presence of electrical fields [33], using the dynamic LBL technique [34], or by applying special post-deposition conditions [35]. Other approaches for the reduction in processing time of PEMMs that were reported used automatic spray system

[36, 37], dip-coating in agitated solutions [38] spin self-assembly [39] or applied the shift time approach [40]. Considering these limitations, it would be interesting to optimize the laborious synthesis procedure of PEMMs by reducing the number of preparation steps. The preparation of PEMMs in as few deposition and rinsing steps as possible would make the industrial application of the latter much more feasible. Therefore, this study investigates the SRNF performance of PEMMs with PDDA as polycation and PVS or PSS as polyanions, with a minimal number of required bilayers, hence with the least possible synthetic effort, but still with highest possible permeances and excellent selectivity. Aspects that will further be explored are the influence of dipping time, coating temperature, coating solvent and rinsing time. 2. Experimental 2.1 Materials The polyethylene-propylene (PE/PP) non-woven fabric (Novatex 2471) was kindly provided by Freudenberg (Germany). PAN (Mw = 150,000 Da) was purchased from Scientific Polymer Products. PDDA(Mw = 200,000-350,000 Da) was obtained from Sigma-Aldrich as a 20 wt% aqueous solution. PSS (Mw = 70,000 Da) in sodium form as a 30% aqueous solution and PVS in sodium form as a 25 wt% aqueous solution were bought from Sigma-Aldrich. Isopropanol (IPA), absolute ethanol (99.5%), tetrahydrofuran (THF), dimethylformamide (DMF) and dimethyl sulfoxide (DMSO) were obtained from VWR. Bengal rose sodium salt (RB) (dye content 95%,), acid fuchsin (AF) (dye content 70%) and bromothymol blue (BTB) (dye content 95%) for filtration were purchased from Sigma-Aldrich. The characteristics of the solutes used in this paper are listed in table 1. 2.2. Membrane Preparation 2.2.1 Support H-PAN support was prepared following the procedure reported earlier [26]. Briefly, a 15 wt % solution of PAN in DMSO was cast on a nonwoven PE/PP support via the phase inversion method, followed by a hydrolysis procedure in-order to convert PAN to H-PAN. The hydrolysis was done by the immersion of the PAN support in 10 wt% sodium hydroxide (NaOH) at 50°C for 40 min. The remaining NaOH was removed by washing with water. 2.2.2 LBL coating of PEs The LBL coating was done using an automated dip coating machine, High Throughput Membrane Leuven (HTML, Belgium) [41, 26]. The automated dip coating machine consists of four separate vessels, as depicted in fig 2: two for the PE solutions and two for rinsing purposes. Dipping time, drip time, number of cycles, and the oscillation frequency while immersing in the PE solution or while rinsing can be controlled. PE solutions were prepared with a composition of 0.2 wt% in aqueous media. PDDA was used as the polycation and PSS or PVS as polyanion. The chemical structure of these PEs are shown in Fig 3. For the

preparation of each bilayer, the negatively charged H-PAN was immersed in the polycation solution and the polyanion solution respectively, with a rinsing step in de-ionized (DI) water after each PE deposition. The PE deposition time and rinsing time was 5 min. Drip time after each coating and rinsing time were both 30 s. During all immersions, the membranes were moved up and down at a frequency of 0.5 Hz over an amplitude of 0.5 cm. PEMMs with 1, 2, 3, 4, and 5 bilayers were thus prepared. Three bilayers of (PSS/PDDA) or (PDDA/PVS) were prepared to study the effect of rinsing, dipping time, temperature and coating solvents on the PEMMs for SRNF performance. Membranes were prepared at different dipping times (2, 5, 10, 20, 30 or 60 min) and at different temperatures (10, 20, 40 or 50°C). In order to investigate the effect of solvent composition, PE solutions were prepared from water/ethanol solutions with different volume ratios. Neither precipitation nor clouding was observed in the PE solutions with increasing the temperatures up to 50°C. A thermostated water bath was used to maintain constant temperature for the dipping solutions. Rinsing times used were five, four, three, two, one or zero min (without rinsing). All solutions were used at their native pH values and the membranes were kept in DI water after preparation in order to keep the membrane structure intact.

Fig 1: Schematic representation of the preparation of PEM-based membranes by the layer by layer assembly method via dip coating (adapted from ref 2)

Fig 2: Demonstration of automated dip-coating machine (HTML, Belgium).

Fig 3: Structural formulas of polyions used in this work.

2.3 Characterization 2.3.1 Atomic Force Microscopy (AFM) AFM was used to measure the surface roughness of membranes. AFM measurements were performed using an Agilent 5500 AFM. The topography and phase images were recorded simultaneously while operating the instrument in tapping mode under ambient conditions. Commercial silicon ‘‘soft tapping mode’’ cantilevers (PPP-NCSTR probes from NanoAndMore GMBH) were used with an integrated pyramidal tip having a radius of curvature lower than 7 nm, nominal spring constant ~ 7 N/m and nominal fundamental resonance frequency ~ 170 kHz. The average roughness (Ra) and the root-mean-square roughness (Rrms), which is given by the standard deviation of the height measurement data (Z values) were calculated using Eq. 1. 𝑅𝑎 =

1 𝑁

𝑁 𝑖=1

𝑍𝑖 − 𝑍 and 𝑅𝑟𝑚𝑠 =

1 𝑁

𝑛 𝑖=1

𝑍𝑖 − 𝑍

2

---------- (Eq.1)

where Zi is the current Z value , 𝑍 is the average of the Z value in the given area. 2.3.3 Quartz crystal microbalance with dissipation Real-time monitoring of PEMs build-up was carried on an E4 quartz crystal microbalance with dissipation (QCM-D) (Q-Sense, Stockholm, Sweden) for the characterization of PEMs. Silica-coated quartz crystals (4.95 ± 0.05 MHz) (Q-Sense E4, Stockholm, Sweden) were treated by a UV-ozone cleaning protocol and were used as substrates for PEM build-up. PEMs were prepared via the LBL technique by passing alternating positively and negatively charged PE solutions throughout the QCM-D chamber using a peristaltic pump (Ismatec IPCN 4, Glattbrugg, Switzerland). Three bilayers were coated for each PE pair through alternate pumping of PE solutions (2 g/l PE solutions in Milli-Q water (18.2 MΩ.cm)) into the QCM-D chambers at 0.3 ml/min. 2.3.4 Water permeance measurement DI water permeance, expressed in terms of l/m2.h.bar, of the substrate and the PEMMs with one or three bilayers of PDDA/PSS or PDDA/PVS were determined using a high-throughput filtration cell (HTML)[41]. The measurement done will be described in section 2.4.

2.3.5 Atomic absorption spectrometry (AAS) In order to check the contamination of the rinsing bath, Na analysis is performed using Spectr AA20 plus atomic absorption spectrometry. Estimation of Na content in the rinsing bath of PSS after the deposition of 2, 6, 10, 15, 20 and 25 bilayers is done and compared with Na content of PSS solution. 2.4 Filtrations SRNF measurements were performed using a high-throughput filtration cell containing 8 filtration cells, each with 0.000177 m2 effective membrane area. The membrane disks were supported with porous stainless steel disks and sealed with Viton O-rings. The system was pressurized with nitrogen to 25 bar. The feed solution was stirred at 500 rpm during filtration to reduce the concentration polarization. For the filtration tests the following solutes and solvents were used: two negatively charged dyes, RB and AcF, and the neutral dye, BTB with IPA. In-order to check the long term stability of PEMMs in aprotic solvents at lower bilayers, filtration experiments were done for 30 h using RB and THF as solute and solvent respectively. All feed solutions were prepared at 35µM and measurements were performed at room temperature. The SRNF performance was assessed using two parameters, namely, permeance and retention. Permeance was measured gravimetrically by weighing the collected permeate expressed as l.m-2h-1bar1. Retention values were calculated from the permeate and feed concentration according to Eq.2, using a Perkin-Elmer Lambda UV-Visible spectrophotometer at a wavelength of 556 nm for AcF, 555 nm for RB and 419 nm for BTB. 𝐶𝑝

𝑅 % = 1 − 𝐶𝑓 ∗ 100 ---------- ( Eq.2) where cp and cf are dye concentration in the permeate and feed respectively. Average values were reported for at least three membrane coupons taken from one sheet.

Table1: Main solute properties

Component

Structure

RB

MW Charge (g/mol) 1017

-

Molar Volume (cm3/mol) 273

585

-

246

AF

BTB

624

zwitterion 242

3. Results and Discussion 3.1.1 Surface roughness Table 2 gives the calculated surface roughness of the PDDA/PSS and PDDA/PVS membranes respectively, measured on 1*1µm2 scans. There is no significant difference in the average surface roughness (Ra) and root mean square roughness (Rrms) of the membranes with the number of bilayers with either pair of PEs. Note that Rrms is the most accurate characteristic of the membrane surface roughness [42], since the Ra takes less account of the low spatial frequencies of the surface height variations. The surface roughness of the substrate before and after hydrolysis is shown in table S1 and fig S1 of supporting information (SI).

Table 2: The surface roughness of the PDDA/PSS and PDDA/PVS membranes measured by AFM. Membranes were prepared from a concentration 2g/l, dip time-5min, rinse time - 5min, solvent MQ water and at room temperature. Membranes PDDA/PSS PDDA/PVS

R rms Ra

1bilayer 6.3 ± 0.2 5.0 ± 0.2

2 bilayer 7.5 ± 1.0 6.0 ± 0.9

3 bilayer 7.3 ± 1.4 5.7 ± 1.1

Rrms

8.3 ± 1.8

8.2 ± 3.1

8.1 ± 3.5

Ra

6.7 ± 1.5

6.6 ± 2.5

6.4 ± 3.0

3.1.2 QCM-D The frequency and dissipation shifts during the PEM build-up were acquired on QCM-D equipment using Milli-Q water to obtain the baseline. The results for (PDDA/PSS)3 and (PDDA/PVS)3 multilayers are shown in Fig 4 and 5, respectively. Deposition of materials onto the crystal sensors results in alteration of the resonance oscillation and frequency amplitude. In the QCM-D results of the (PDDA/PSS)3 multilayer, each layer deposition can be distinguished by a stepwise decrease in measured frequency, indicating successful deposition of PEs onto the quartz crystal. The constant and regular decrease of the relative frequency of the (PDDA/PSS)3 multilayers reflects a regular increase of the total amount of material deposited on the quartz crystal. Introduction of each PE solution is accompanied by an increase in dissipation, indicating that the viscoelasticity of the PEM film is increased with each layer. The observed sharp frequency and dissipation jumps in (PDDA/PSS)3 multilayers imply quick adsorption for each PE deposition. For (PDDA/PVS)3 multilayers, the frequency shifts in a stepwise fashion after introduction of PE solutions followed by a slight frequency increase. The frequency increase is attributed to materials loss from the surface which might be either due to the loss of PEs during the PE rearrangement or due to the captured water. The dissipation of (PDDA/PVS)3 films was increased after each PDDA deposition step and was decreased after each PVS deposition step, indicating a less rigid structure after PDDA adsorption and a more rigid one after PVS adsorption.

Fig 4: Frequency and dissipation graphs for the 3rd overtone of PDDA/PSS PEMM measured with QCM-D. PE deposition was done by using a con 2g/l, dip time - 5min, no rinsing in between PE deposition, solvent - MQ water and at room temperature.

Fig 5: Frequency and dissipation graphs for the 3rd overtone of PDDA/PVS PEMM measured with QCM-D. PE deposition was done by using a con 2g/l, dip time - 5min, no rinsing in between PE deposition, solvent - MQ water and at room temperature.

3.1.3 Water Permeance The water permeance of the supporting layer and PEMMs with 1 and 3 bilayers is shown in table 3. As expected, the water permeance decreases with the increase in the number of bilayers. The results confirm the deposition of PEs on the charged surface. The coating takes place in principle only over the surface, but might possibly also partly occur within the pores of the membrane which would block the pores partially [43]. Table 3: Water permeance of the supporting layer and the PEMMs. Membranes were prepared by a con of 2g/l dip time - 5min, rinse time - 5min, solvent -MQ water and at room temperature.

Membrane

Water permeance (L/m2.h.bar)

H-PAN

406 ± 7

(PDDA/PVS)1

232 ± 3

(PDDA/PVS)3

63 ± 8

(PDDA/PSS) 1

105 ± 3

(PDDA/PSS) 3

33 ± 5

3.2.1 SRNF performance Fig 6 displays the SRNF performance of membranes with different numbers of bilayers prepared from PSS and PVS. Both sets of membranes show comparable retention values and the PVS based membranes provide slightly higher permeance as compared to the PSS based

membranes. Observed retentions for two negatively charged dyes, RB and AF, proves that even 2 bilayers can already create a defect-free membrane. In the case of these negatively charged dyes, the retention is almost constant (up to 99%) from three bilayers onwards. All membranes showed lower retentions for the zwitterion BTB, as indicated in table 4. The lower retentions for BTB, compared to the negatively charged dyes is due to the more prominent role of the Donnan exclusion, compared to the size exclusion [26]. In addition, Fig 7 indicates the excellent long-term stabilily of (PDDA/PSS)3 membranes in a strong aprotic sovent like THF for a period of 30 h. The results thus show a stable retention and permeance as a function of time, confirming the potential of PEMMs in SRNF. PDDA/PSS

PDDA/PVS

Fig 6: SRNF performance of PSS/PDDA and PVS/PDDA membranes for different charged solutes from IPA solutions: (left) AF; (right) RB . Membranes were prepared by a con 2g/l, dip time - 5min, rinse time - 5min, solvent - MQ water and at room temperature.

Fig 8 compares the performance of different types of previously reported SRNF membranes with the current PEMMs. It is observed that PEMMs with three bilayers showed comparable SRNF performances with literatures; but the overall permeance is low for PEMMs. Compared to the water permeances, the IPA permeances are much lower. During permeation, water acts

as a good solvent and IPA as a bad solvent for the PEs. In the presence of a good solvent, polymer chains will try to maximize the polymer-solvent interactions and thus make the membrane swell. In a poor solvent, PE chains will adopt a more compact conformation in order to reduce the polymer-solvent interactions [44,45]. The lower permeability of PEMMs in organic solvents is thus probably due to the lower swelling in organics. Moreover, PEMMs with a high charge density have a high degree of ionic cross-linking, which leads to decreased permeability along with the high selectivity.

Table(4): SRNF performance for IPA solutions for the zwitterion (BTB) of membranes containing two and three bilayers of (PDD/PSS) and (PDDA/PVS).

Membrane (PDDA/PSS)2 (PDDA/PSS)3 (PDDA/PVS)2 (PDDA/PVS)3

Retention (%) 71.4 ± 1.3 83.9 ± 0.2 55.5 ± 0.6 72.3 ± 0.3

Permeance (l/m^2 h bar) 0.92 ± 0.13 0.18 ± 0.06 1.74 ± 0.08 0.18 ± 0.03

Fig 7: Longer term SRNF performance of (PDDA/PSS)3 membranes using RB/THF feed. Membranes were prepared by a con 2g/l, dip time 5min, rinse time - 5min, solvent - MQ water and at room temperature.

: Fig 8: Comparison of SRNF performance (RB/IPA) between current PEMMs and other SRNF membrane systems reported in literature

3.2.2 Dipping time Adsorption kinetics of PEMs show a bimodal nature wherein the first step is a fast process involving the transport of chains to the surface and their adsorption is the slower step relating to the internal reorganization of the polymer chains in the multilayer [46,47]. The PE adsorption is mainly determined by charge density and concentration of the polyelectrolytes [48,49]. A higher concentration leads to faster adsorption kinetics, whereas the dipping time has a significant role in establishing a plateau for the adsorption equilibrium. Fig 9 shows that all membranes prepared with different dipping times possess similar SRNF properties. The results confirm that two minutes dipping is enough for the adsorption of 2 g/lPDDA, PSS or PVS. All membranes showed rather smooth areas with relatively low roughness, irrespective dipping time, as shown in fig 10 (on a 1*1µm scale). On a larger scale (5*5 µm), roughness is slightly reduced with increasing dipping time (fig S4).

Fig 9: SRNF (IPA/Af) performance of the (PDDA/PSS)3 membranes prepared at different dipping times. Membranes were prepared by a con of 2g/l, rinse time - 5min, solvent - MQ water and at room temperature.

Fig 10: AFM images of (PDDA/PVS)3 membranes in a 1*1 µm scale prepared at different dipping time:(a) 2 min and (b) 30 min.

3.2.3 Coating temperature The coating temperature is a key parameter in the preparation of PEMMs. An increased temperature is reported to generate thicker films [50]. Adsorption is faster and larger numbers of inter-PE bonds are formed at higher temperature [51, 52]. Fig 11 shows how temperature increase during coating adversely affects the SRNF properties. Moreover, the best SRNF performance is shown at ambient temperature. The potential formation of loops and tails at higher T should thus result in thicker films, which may prove insufficient to close surface membrane defects, thus leading to reduced SRNF properties.

Fig 11: SRNF performance (IPA/Af) of (PDDA/PSS)3 membranes prepared at different temperature. Membranes were prepared by a con of 2g/l, dip time - 5min, rinse time - 5min and solvent - MQ water.

3.2.4 Dip coating solvent Adapting the solvent composition of the coating solutions also permits control over the formation of PEMMs. Alcohol addition to the aqueous PE solution, results in a decreased dielectric constant of the medium, which is inversely related to the electrostatic interactions [53, 54, 55]. The multilayer build-up in such a solvent system exhibits an exponential growth pattern, whereby a sharp increase in thickness and mass loading is observed [53]. In order to study the influence of the dip coating solvent composition on PEM preparation, PE solutions were prepared by varying the ethanol concentration in the dip coating solution. The PVS solution precipitated at 100% ethanol, due to excessive changes in the dielectric constant of the solvent. As seen in Fig 12, retention is reduced with increased ethanol concentration. PE chains in poor solvents collapse into more compact structures [55]. Thus the increase in electrostatic interactions, combined with this collapse, results in the formation of holes and aggregates in the PEMMs. Moreover, Fig 13 and fig 14 shows an increase in roughness with increase in ethanol concentration. This indeed confirms the formation of holes and aggregates in PEMMs at higher ethanol concentration. These drastic changes, related to the increased alcohol content during multilayer build-up, probably cause the poor performance of the resulting membranes.

Fig 12: SRNF performance (AF/IPA) of (PDDA/PSS)3 membranes prepared from aqueous solutions containing different ethanol concentration. Membranes were prepared by a con 2g/l, dip time - 5min, rinse time - 5min and at room temperature.

Fig 13: Surface roughness (10*10 µm) of the (PDDA/PSS)3 membranes prepared from aqueous solutions containing different ethanol concentration

Fig (14):AFM images of (PDDA/PVS)3 membranes on a 10*10 µm scale prepared at different ethanol concentration: (a) 0% (b) 5% (c) 10% and (d) 25%.

3.2.5 Rinsing time The rinsing process has been proposed to be crucial for the removal of loosely attached polyions and for the rearrangement of the pre-adsorbed polymer layer to render the multilayers more stable [9,56]. To study the effect of rinsing on SRNF properties, (PDDA/PSS)3 membranes were prepared at different rinsing times by keeping all other conditions constant. Fig 15 shows that all thus prepared membranes have similar retentions and permeances. Membranes prepared without rinsing show a smooth surface at a smaller scale, but a slight increase in roughness on a larger scale (fig 16). The observed filtration performances prove that rinsing time and even total absence of rinsing doesn’t affect the SRNF performance. Kovacevic et al. [57] and Saarinen et al. [58] also reported that rinsing doesn’t have a significant role in the multilayer layer build-up. The different rinsing methods previously studied, either as in pure water, or with a PE solution similar to which is used for multilayer formation or in a salt solution, all showed only a minor effect on the overall multilayer build-up [59]. Thus, the PEMM preparation without rinsing step in-between the addition of oppositely charged PEs helps to reduce the overall multilayer build-up time without compromising membrane performance. It should be mentioned that the membranes are kept in water after their preparation in order to preserve the membrane structure. This treatment, could possibly help to remove the loosely attached polymer chains that were left behind and could serve as a rinsing for the outermost layer. Absence of rinsing steps could of course bring along contamination of one coating bath by another. To verify this, the amount of PE going into the rinsing bath was determined by estimating the Na content of the PSS rinsing bath. This Na present in the rinsing bath is persumed to be from the PSS bath which would be similar to the contamination of the PDDA coating bath. During PEMM preparation, the Na content in the rinsing bath was determined as shown in table 5. The Na content is increasing from 0.022 ppm (2 bilayers) to 1.24 ppm (25 bilayers). 1.24 ppm Na is negligible compared to the 112.51 ppm Na content in the PSS solution. It does indicate that some crosscontamination occurs, but that these amounts are too low to merit inclusion of so many rinsing steps.

Fig 15: SRNF performance (IPA/RB) of the (PDDA/PSS)3 membrane prepared using different rinsing times. Membranes were prepared by a con 2g/l, dip time - 5min, solvent - MQ water and at room temperature.

Fig (16):AFM images of (PDDA/PSS)3 membranes prepared without rinsing in between: a (1*1µm scale) and b(5*5µm scale).

Table 5 : The determination of the Na content in the rinsing bath (Na content in the PSS solution was 112.51 ppm)

Number of deposited bilayers

Na content in PSS rinsing bath (ppm)

2 6

0.02 0.16

10 15 20 25

0.37 0.87 1.11 1.24

4. Conclusions SRNF membranes, based on PDDA/PSS and PDDA/PVS PE systems, were successfully prepared with only 3 bilayers. The 20 to 80 processing steps and 100 to 400 min needed for the previously reported membranes were thus drastically reduced to 12 steps in the case of 3 bilayers, realized all together in less than 12 min. Importantly, also the rinsing process during the multilayer build-up didn’t show any influence on the final membrane performance. The most practical ambient temperature seemed to be the preferred T to prepare membranes with the best SRNF properties. A decrease in the overall preparation time could thus drastically reduce the membrane manufacturing time, thus significantly increasing the ‘up-scaling’ potential of these membranes and reduce their production costs.

5. Acknowledgements We thank the financial support from the OT (11/061) from K U Leuven, I.A.P – P.A.I. grant (IAP 6/27; Belgian Federal Government), and of the long term Methusalem funding (‘CASAS’) by the Flemish Government. We also thank Prof. V. V. Tarabara (Michigan State University) for the QCM measurements. References [1] G. Decher, Fuzzy Nanoassemblies: Toward Layered Polymeric Multicomposites, Science. 277 (1997) 1232 -1237. [2] N. Joseph, P. Ahmadiannamini, R. Hoogenboom and I. F. J. Vankelecom, Layer- by-layer preparation of polyelectrolyte multilayer membranes for separation, Polym Chem, 5 (2014) 1817-1831. [3] A. V. Dobrynin and M. Rubinstein, Theory of polyelectrolytes in solutions and at surfaces. Prog. Polym. Sci, 30 (2005) 1049-1118. [4] N. A. Kotov, Layer-by-layer self-assembly: The contribution of hydrophobic interactions, Nanostruct. Mater, 12 (1999) 789-796. [5] W. B. Stockton and M. F. Rubner, Molecular-level processing of conjuguated polymers. 4. Layer-by-layer

manipulation

of

polyaniline

via

hydrogen-bonding

interactions,

Macromolecules, 30 (1997) 2717-2725. [6] A. B. F. Antunes, M. Dierendonck, G. Vancoillie, J. P. Remon, R. Hoogenboom and B. G. De Geest, Hydrogen bonded polymeric multilayer films assembled below and above the cloud point temperature, Chem commun, 49 (2013), 9663. [7] T. K. Sievers, A. Vergin, H. Mohwald and D. G. Kurth, Thin films of crosslinked metallosupramolecular coordination polyelectrolytes, Langmuir, 23 (2007) 12179-12184. [8] S. T. Dubas and J. B. Schlenoff, Swelling and smoothing of polyelectrolyte multilayers by salt, Langmuir, 17 (2001) 7725-7727. [9] N. G. Hoogeven, M. A. C. Staurt and G. J. Fleer, Formation and stability of multilayers of polyelectrolytes, Langmuir, 12(1996) 3675-3681. [10] M. Schönhoff, Layered polyelectrolyte complexes: physics of formation and molecular properties, J. Phys.: Condens Matter,15 (49), 2003, 1781-1808. [11] S. S. Shiratori and M. F. Rubner, pH dependent thickness behavior of sequentially adsorbed layers of weak polyelectrolytes, Macromolecules, 33 (2000) 4213-4219.

[12] H. L. Tan, M.J. McMurdo, G. Pan and P. G. V. Patten, Temperature dependence of polyelectrolyte multilayer assembly, Langmuir, 19 (2003) 9311-9314. [13] J. H. Dai, A. M. Balachandra, J. I. Lee and M. L. Bruening, Controlling ion transport through multilayer polyelectrolyte membranes by derivatization with photolabile functional groups, Macromolecules, 35(8) (2002) 3164-3170. [14] J. A. Jaber and J. B. Schlenoff, Mechanical properties of reversibly cross-linked ultrathin polyelectrolyte complexes, J. Am. Chem. Soc, 2006, 128(9), 2940-2947. [15] J. L. Stairs, J. J. Harris and M. L. Bruening, Enhancement of the ion-transport selectivity of layered polyelectrolyte membranes through crosslinking and hybridization, Chem. Mater, 13(8) ( 2001) 2641-2648. [16] H. Mjahed, G. Cado, F. Boulmedias, B. Senger,P. Schaff, V. Ball, J-C. Voegel, Restructuring of exponentially growing polyelectrolyte multilayer films induced by salt concentration variations after film deposition, J. Mater. Chem, 21(23) (2011) 8416-8421. [17] L. Kraseman and B. Tieke, Ultrathin self-assembled polyelectrolyte membranes for pervaporation, J. Membr. Sci. 150 (1998) 23-30. [18] S. U. Hong and M. L. Bruening, Separation of amino acid mixtures using multilayer polyelectrolyte nanofiltration membranes, J. Membr. Sci. 280 (2006) 1-5. [19] A. Toutianoush, W. Jin, H. Deligoz and B. Tieke, Polyelectrolyte multilayer membranes for desalination of aqueous salt solutions and sea water under reverse osmosis conditions, Appl. Surf. Sci. 246 (2005) 437-443. [20] J. Levasalmi and T.J. McCarthy, Poly(4-methyl-1-pentene)- supported polyelectrolyte multilayer films: preparation and gas permeability, Macromolecules, 30 (1997) 1752-1757. [21] Xianfeng, Li, S. De Feyter, D. Chen, S. Aldea, P. Vandezande, F. Du Prez, and I.F.J. Vankelecom,

Solvent-resistant

nanofiltration

membranes

based

on

multilayered

polyelectrolyte complexes, Chem. Mater, 20 (2008) 3876-3883. [22] P. Vandezande, L. E. M. Gevers and I. F. J. Vankelecom, Solvent resistant nanofiltration : separation on a molecular level, Chem. Soc. Rev, 37 (2008) 365-405. [23] Y. H. See-Toh, F. W. Lim and A. Livingston, Polymeric membranes for nanofiltration in polar aprotic solvents, J. Membr. Sci. 324 (2008) 220-232. [24 ] H.J. Zwijnenberg, S. M. Dutczak, M. E. Boerrigter, M. A. Hempenius, M. W. J. LuitenOlieman, N. E. Benes, M. Wessling and D. Stamatialis, Important factors influencing molecular weight cut-off determination of membranes in organic solvents, J. Membr. Sci, 390-391 (2012) 211-217.

[25] P. Ahmadiannamini, Xianfeng. Li, W. Goyens, N. Joseph, B. Meesschaert and I.F.J. Vankelecom, Multilayered polyelectrolyte complex based solvent resistant nanofiltration membranes prepared from weak polyacids, J. Membr. Sci. 394-395 (2012), 98-106. [26] P.Ahmadiannamini, X. Li, W. Goyens, B. Meesschaert, W. Vanderlinden, S. De Feyter and I. F. J. Vankelecom, Influence of polyanion type and cationic counter ion on the SRNF performance of polyelectrolyte membranes, J. Membr. Sci. 403-404 (2012) 216-226. [27] D. Chen, Solvent resistant nanofiltration membranes based on multilayered polyelectrolytes deposited on silicon composite, J. Appl. Polym. Sci, 129 ( 2013) 3156-3161. [28] P. Stroeve, V. Vasquez, M. A. N. Coelho, J. F. Rabolt, Gas transfer in supported films made by molecular self-assembly of ionic polymers, Thin Solid Films, 284-285 (1996) 708712. [29] S.U.Hong, L. Ouyang and M.L. Bruening, Recovery of phosphate using multilayer polyelectrolyte nanofiltration membranes, J. Membr. Sci. 327 (2009) 2-5. [30] S.U.Hong, M.D.Miller and M.L. Bruening, Removal of dyes, sugars, amino acids from NaCl solutions using multilayer polyelectrolyte nanofiltration membranes, Ind. Eng. Chem. Res, 45 (2006) 6284-6288. [31] R. Malaisamy and M. L. Bruening, High-flux nanofiltration membranes prepared by adsorption of multilayer polyelectrolyte membranes on polymeric supports, Langmuir, 21(2005)10587- 10592. [32] M. Yin, J. Qian, Q. An, Q. Zhao, Z. Gui and J. Li, Polyelectrolyte layer-by-layer selfassembly at vibration condition and the pervaporation performance of assembly multilayer films in dehydration of isopropanol, J. Membr. Sci. 358 (2010) 43-50. [33] P. Zhang, J. Qian, Y. Yang, Q. An, X. Liu and Z. Gui, Polyelectrolyte layer-by-layer self-assembly enhanced by electric field and their multilayer membranes for separating isopropanol-water mixtures, J. Membr. Sci. 320 (2008) 73-77. [34] G. Zhang, W. Gu, S. Ji, Z. Liu, Y.Peng and Z. Wang, Preparation of polyelectrolyte multilayer membranes by dynamic layer-by-layer process for pervaporation separation of alcohol/water mixtures, J. Membr. Sci. 280 (2006) 727-733. [35] Z. Zhu, X. Feng and A. Penlidis, Layer-by-layer self-assembled polyelectrolyte membranes for solvent dehydration by pervaporation, Mater. Sci. Eng. C, 27(2) (2007), 612619.

[36] H. Tang, G. Zhang and S. Ji, Rapid assembly of polyelectrolyte multilayer membranes using an automatic spray system, AIChE Journal, 59 (2013), 250 -257. [37] A. Izquierdo, S. S. Ono, J. C. Voegel, P. Schaaf and G. Decher, Dipping versus spraying: Exploring the deposition conditions for speeding up layer-by-layer assembly, Langmuir, 21 (2005), 7558-7567. [38] Y. Fu, S. Li, J. Xu, M. Yang, J. Zhang, Y. Jiao, J. Zhang, K. Zhang and Y. Jia, Facile and efficient approach to speed up layer-by-layer assembly: dipping in agitated solutions, Langmuir, 27(2) (2011), 672-677. [39] J. Cho, K. Char, J. D. Hong and K. B. Lee, Adv. Mater, 13 (2001), 1076-1078. [40] D. A. Hagen, B. Foster, B. stevens, J. C. Grunlan, Shift-time polyelectrolyte multilayer assembly: fast filmgrowth and high gas barrier with fewer layers by adjusting deposition time, ACS Macro Lett. 3 (2014) 663-666 [41] http://www.html-membrane.be [42] K. Boussu, B. Van der Bruggen, A. Volodin, J. Snauwaert, C. Van Haesendonck, and C. Vandecasteele, Roughness and hydrophobicity studies of nanofiltration membranes using different modes of AFM, J. Colloid Interface Sci. 286 (2005) 632–638. [43] J. Kochan, T. Wintgens, J. E. Wong and T. Melin, Properties of polyethersulfone ultrafiltration membranes modified by polyelectrolytes, Desalination, 250(3) (2010) 10081010. [44] N. G. Hoogeveen, M.A.C. Stuart and G. J. Fleer, Formation and stability of multilayers of polyelectrolytes, Langmuir,12 (1996), 3675-3681. [45]E. Poptoshev, B. Schoeler and F.Caruso, Influence of solvent quality on the growth of polyelectrolyte multilayers, Langmuir, 20 (2004) 829-834. [46] M. Schonhoff, Layered polyelectrolyte complexes: physics of formation and molecular properties, J. Phys. Condens. Matter. 15 (2003) R1781 – R1808. [47] E. Guzman, H. A. Ritacco, F. Ortega, R. G. Rubio, Growth of polyelectrolyte layers formed by poly(4-styrenesulfonate sodium salt) and two different polycations: New insights from study of adsorption kinetics, J. Phys. Chem. C, 116 (2012) 15474-15483. [48] C. Magnenet, S. Lakard, C. C. Buron, B. Lakard, Functionilization

of organic

membranes by polyelectrolyte multilayer assemblies : Application to the removal of copper ions from aqueous solutions, J. Colloid Interface Sci. 376 (2012) 202-208. [49] O.Mermut, C. J. Barrett, Effect of charge density and counterions on the assembly of polyelectrolyte multilayers, J. Phys. Chem. B, 107 (2003) 2525-2530.

[50] M. Salomaki, I. A. Vinokurov, J. Kankare, Effect of temperature on the builup of polyelectrolyte multilayers, Langmuir, 21 (2005) 11232-11241. [51] K. Buscher, K. Graf, H. Ahrens, C. A. Helm, Influence of adsorption conditions on the structure of polyelectrolyte multilayers, Langmuir, 18 (2002) 3585-3591. [52] W. Lenk, J. Meier-Haack, Polyelectrolyte multilayer membranes for pervaporation separation of aqueous organic mixtures, Desalination, 148 (2002) 11-16. [53] S. T. Dubas, J. B. Schlenoff, Factors controlling the growth of polyelectrolyte multilayers, Macromolecules, 32, (1999) 8153-8160. [54] W. Dong, S. Liu, L. Wan, G. Mao, D. G. Kurth, H. Mohwald, Controlled permeability in polyelectrolyte films via solvent treatment, Chem. Mater, 17 (2005) 4992-4999. [55] E. Poptoshev, B. Schoeler, F. Caruso, Influence of solvent quality on the growth of polyelectrolyte multilayers, Langmuir, 20 (2004) 829 –834. [56] Y. Lvov, K. Ariga, M. Onda, I. Ichinose, T. Kunitake, A careful examination of the adsorption step in the alternate layer-by-layer assembly of linear polyanion and polycation, Colloids and surfaces A: Physicochem. Eng. Aspects, 146 (1999) 337-346. [57] D. Kovacevic, S. van der Burgh, A. de Keizer and M. A. Cohen Stuart, Specific ionic effects on weak polyelectrolyte multilayer formation, J. Phys. Chem. B, 107 (2003) 79988002 [58] T. Saarinen, M. Osterberg and J. Laine, Adsorption of polyelectrolyte multilayers and complexes on silica and cellulose surfaces studied by QCM-D, Colloids and Surfaces: Physicochem. Eng. Aspects, 330 (2008) 134-142. [59] Z. Feldoto, I. Varga and E. Blomberg, Influence of salt and rinsing protocol on the structure of PAH/PSS polyelectrolyte multilayers, Langmuir, 26(22) (2010) 17048-17057.