Repositorio Institucional de Documentos

The EM3E Master is an Education Programme supported by the European Commission, the European Membrane Society (EMS), the European Membrane House (EMH), and a large international network of industrial companies, research centres and universities; www.em3e.eu Page i NOVEL ION SELECTIVE MEMBRANES Mahbub MORSHED Student, Erasmus Mundus Master in Membrane Engineering. (EM3E) Academic Session: 2012-2014 Supervisors Dr. Ir. D.C. Nijmeijer Professor & Head of Membrane Science & Technology (MST), University of Twente, The Netherlands. Dr. Reyes Mallada Associate Professor, Department of Chemical and Environmental Engineering, University of Zaragoza, Spain. The EM3E Master is an Education Programme supported by the European Commission, the European Membrane Society (EMS), the European Membrane House (EMH), and a large international network of industrial companies, research centres and universities; www.em3e.eu Page a Acknowledgement I would like to express my gratitude and thanks to the head of MST group, Prof. Kitty Nijmeijer for giving me the opportunity to conduct my master thesis in MST group. I would like to thank Dr. Rayes Mallada for approving me to conduct my master thesis in MST group of University Twente. My special thanks to Dr. Wiebe Matthijs de Vos. Your suggestion and co-operation helped me a lot to learn and get positive energy to continue my master thesis in MST. Your opinion helped me to find a way forward and thank you very much for your correction and suggestion regarding to my report. I am grateful to Dr. Ir. W.M. de Vos. I also want to thank specially Joris de Grooth, Timon Rijnaarts for their support and discussion. Every discussion with Joris and Timon was thoughtful and full of learning for me. I wish to express my gratitude to all academic staff, technical staff, secretary and all the group members. Thanks to my daily supervisor Sinem Tas. I am grateful to the EM3E committee for giving me the opportunity to be a part of the EM3E program. The EM3E Master is an Education Programme supported by the European Commission, the European Membrane Society (EMS), the European Membrane House (EMH), and a large international network of industrial companies, research centres and universities; www.em3e.eu Page b DISCLAIMER EN-This project has been funded with support from the European Commission. This publication reflects the views only of the author, and the Commission cannot be held responsible for any use which may be made of the information contained therein. BG-Този проект е финансиран с подкрепата на Европейската комисия.Тази публикация отразява само личните виждания на нейния автор и от Комисията не може да бъде търсена отговорност за използването на съдържащата се в нея информация. CS-Tento projekt byl realizován za finanční podpory Evropské unie. Za obsah publikací odpovídá výlučně autor. Publikace nereprezentují názory Evropské komise a Evropská komise neodpovídá za použití informací, jež jsou jejich obsahem. DA-Dette projekt er finansieret med støtte fra Europa-Kommissionen. Denne publikation forpligter kun forfatteren, og Kommissionen kan ikke drages til ansvar for brug af oplysningerne heri. DEDieses Projekt wurde mit Unterstützung der Europäischen Kommission finanziert. Die Verantwortung für den Inhalt dieser Veröffentlichung trägt allein der Verfasser; die Kommission haftet nicht für die weitere Verwendung der darin enthaltenen Angaben. ΕΛ-Το ζσέδιο αςηό σπημαηοδοηήθηκε με ηην ςποζηήπιξη ηηρ Εςπωπαϊκήρ Επιηποπήρ. Η παπούζα δημοζίεςζη δεζμεύει μόνο ηον ζςνηάκη ηηρ και η Επιηποπή δεν εςθύνεηαι για ηςσόν σπήζη ηων πληποθοπιών πος πεπιέσονηαι ζε αςηήν. ESEl presente proyecto ha sido financiado con el apoyo de la Comisión Europea. Esta publicación es responsabilidad exclusiva de su autor. La Comisión no es responsable del uso que pueda hacerse de la información aquí difundida. ET-Projekti on rahaliselt toetanud Euroopa Komisjon. Publikatsiooni sisu peegeldab autori seisukohti ja Euroopa Komisjon ei ole vastutav selles sisalduva informatsiooni kasutamise eest. FIHanke on rahoitettu Euroopan komission tuella. Tästä julkaisusta (tiedotteesta) vastaa ainoastaan sen laatija, eikä komissio ole vastuussa siihen sisältyvien tietojen mahdollisesta käytöstä. FRCe projet a été financé avec le soutien de la Commission européenne. Cette publication (communication) n’engage que son auteur et la Commission n’est pas responsable de l’usage qui pourrait être fait des informations qui y sont contenues. GAMaoiníodh an tionscadal seo le tacaíocht ón gCoimisiún Eorpach. Tuairimí an údair amháin atá san fhoilseachán [scéala] seo, agus ní bheidh an Coimisiún freagrach as aon úsáid a d’fhéadfaí a bhaint as an eolas atá ann. The EM3E Master is an Education Programme supported by the European Commission, the European Membrane Society (EMS), the European Membrane House (EMH), and a large international network of industrial companies, research centres and universities; www.em3e.eu Page c HUAz Európai Bizottság támogatást nyújtott ennek a projektnek a költségeihez. Ez a kiadvány (közlemény) a szerzõ nézeteit tükrözi, és az Európai Bizottság nem tehetõ felelõssé az abban foglaltak bárminemû felhasználásért. ITIl presente progetto è finanziato con il sostegno della Commissione europea. L’autore è il solo responsabile di questa pubblicazione (comunicazione) e la Commissione declina ogni responsabilità sull’uso che potrà essere fatto delle informazioni in essa contenute. NLDit project werd gefinancierd met de steun van de Europese Commissie. De verantwoordelijkheid voor deze publicatie (mededeling) ligt uitsluitend bij de auteur; de Commissie kan niet aansprakelijk worden gesteld voor het gebruik van de informatie die erin is vervat. LTŠis projektas finansuojamas remiant Europos Komisijai. Šis leidinys [pranešimas] atspindi tik autoriaus požiūrį, todėl Komisija negali būti laikoma atsakinga už bet kokį jame pateikiamos informacijos naudojimą. LVŠis projekts tika finansēts ar Eiropas Komisijas atbalstu. Šī publikācija [paziņojums] atspoguļo vienīgi autora uzskatus, un Komisijai nevar uzlikt atbildību par tajā ietvertās informācijas jebkuru iespējamo izlietojumu. MTDan il-proġett ġie finanzjat bl-għajnuna tal-Kummissjoni Ewropea. Din il-publikazzjoni tirrifletti (Dan il-komunikat jirrifletti) l-opinjonijiet ta’ l-awtur biss, u l-Kummissjoni ma tistax tinżamm responsabbli għal kull tip ta’ uzu li jista’ jsir mill-informazzjoni li tinsab fiha ( fih). PLTen projekt został zrealizowany przy wsparciu finansowym Komisji Europejskiej. Projekt lub publikacja odzwierciedlają jedynie stanowisko ich autora i Komisja Europejska nie ponosi odpowiedzialności za umieszczoną w niej zawartość merytoryczną. PTProjecto financiado com o apoio da Comissão Europeia. A informação contida nesta publicação (comunicação) vincula exclusivamente o autor, não sendo a Comissão responsável pela utilização que dela possa ser feita. ROAcest proiect a fost finanţat cu sprijinul Comisiei Europene.<0}Această publicaţie (comunicare) reflectă numai punctul de vedere al autorului şi Comisia nu este responsabilă pentru eventuala utilizare a informaţiilor pe care le conţine. SK-Tento projekt bol financovaný s podporou Európskej Komisie. Táto publikácia (dokument) reprezentuje výlučne názor autora a Komisia nezodpovedá za akékoľvek použitie informácií obsiahnutých v tejto publikácii (dokumente). SLIzvedba tega projekta je financirana s strani Evropske komisije. Vsebina publikacije (komunikacije) je izključno odgovornost avtorja in v nobenem primeru ne predstavlja stališč Evropske komisije. SVProjektet genomförs med ekonomiskt stöd från Europeiska kommissionen. För uppgifterna i denna publikation (som är ett meddelande) ansvarar endast upphovsmannen. Europeiska kommissionen tar inget ansvar för hur dessa uppgifter kan komma att användas The EM3E Master is an Education Programme supported by the European Commission, the European Membrane Society (EMS), the European Membrane House (EMH), and a large international network of industrial companies, research centres and universities; www.em3e.eu Page I Table of contents List of Figures .......................................................................................................................... III List of Tables ............................................................................................................................ V 1. Introduction ........................................................................................................................ 1 1.1 Aim of the project ....................................................................................................... 3 2. Theory ................................................................................................................................. 4 2.1 Ion exchange membrane.............................................................................................. 4 2.2 Modification of ion exchange membrane.................................................................... 5 2.3 Layer by layer (LbL) self-assembly ............................................................................ 7 2.3.1 General features of polyelectrolyte multilayers (PEM) by LbL dip coating ....... 8 2.3.2 PEM phenomena by LbL ..................................................................................... 9 2.4 Variables for PEM by LbL ........................................................................................ 13 2.4.1 Effect of temperature ......................................................................................... 14 2.4.2 Effect of salt concentration ................................................................................ 15 2.4.3 Effect of ion type ............................................................................................... 18 2.4.4 Effect of pH........................................................................................................ 19 2.5 LbL polyelectrolyte membrane ................................................................................. 20 2.6 Diffusion dialysis and LbL ........................................................................................ 22 2.6.1 PEM and diffusion dialysis ................................................................................ 24 3. Experimental ..................................................................................................................... 25 3.1 Materials .................................................................................................................... 25 3.2 Methods ..................................................................................................................... 25 3.2.1 Preparation of polyelectrolyte multilayer by PDADMAC and PSS .................. 25 3.2.2 Preparation of the polyelectrolyte multilayer by PDADMACPSBMA-PSS ... 27 3.2.3 UV-Vis measurement......................................................................................... 28 3.2.4 Contact angle measurement ............................................................................... 29 3.2.5 Measurement of electrical resistance and current voltage curve ....................... 31 3.2.6 Limiting current density ..................................................................................... 32 The EM3E Master is an Education Programme supported by the European Commission, the European Membrane Society (EMS), the European Membrane House (EMH), and a large international network of industrial companies, research centres and universities; www.em3e.eu Page II 3.2.7 Ion flux measurement by diffusion experiment ................................................. 34 3.3 Data analysis. and calculation ................................................................................... 35 4. Result and discussion........................................................................................................ 36 4.1 Membrane Characterization ...................................................................................... 36 4.1.1 Characterization by UV-Vis spectroscopy......................................................... 36 4.1.1.1 UV-Vis spectra for PDADMAC-PSS modified FKB .................................... 36 4.1.1.2 UV-Vis spectra for PDADMAC-PSBMA-PSS modified FKB ..................... 38 4.1.2 Characterization by contact angle ...................................................................... 39 4.1.2.1 Contact angle measurement for PDADMAC-PSS modified FKB ................... 39 4.1.2.2 Contact angle measurement for PDADMAC-PSBMA-PSS modified FKB .... 42 4.2 Membrane performance ............................................................................................ 43 4.2.1 Membrane electrical resistance .......................................................................... 43 4.2.1.1 Electrical resistance for PDADMAC-PSS modified FKB ............................. 43 4.2.1.2 Electrical resistance for PDADMAC-PSBMA-PSS modified FKB .............. 46 4.2.2 Limiting current density ..................................................................................... 47 4.2.3 Membrane performance for PDADMAC-PSS modified FKB .......................... 50 4.2.3.1 Diffusion experiments for K+ ......................................................................... 51 4.2.3.2 Diffusion experiments for Li+ ........................................................................ 52 4.2.3.3 Diffusion experiments for mixture of K+ and Li+ .......................................... 53 4.2.4 Flux and selectivity ............................................................................................ 54 5. Conclusion ........................................................................................................................ 58 6. Recommendation .............................................................................................................. 61 7. References ........................................................................................................................ 66 Appendix .................................................................................................................................. 76 The EM3E Master is an Education Programme supported by the European Commission, the European Membrane Society (EMS), the European Membrane House (EMH), and a large international network of industrial companies, research centres and universities; www.em3e.eu Page III List of Figures Figure 1: A Scheme of a cation exchange membrane……………………………………. 5 Figure 2: LbL deposition scheme by dip coating ……………………………………....... 7 Figure 3: The LbL deposition by spray and spin coating ……………………………........ 8 Figure 4: A Scheme of PEM multilayer ………………………………….......................... 10 Figure 5: Schematic intrinsic and extrinsic charge balance in the PEM ………………… 11 Figure 6: Asymmetric growth phenomena of PSS/PDADMAC layer buildup. Layer number represents the PSS/PDADMAC alternatively …………………………………… 12 Figure 7: Effect of deposition temperature on the thickness of a (PDADMAC/PSS) 10 multilayer on Si wafer and deposited at 1M NaCl ……………………………………….. 15 Figure 8: (a) Fluorescence intensity varying with the layer number of the multilayer film made of ADPy-100 (The pyrene labeled polyanion) and PDADMAC deposited in NaCl solutions. 8(b) Thickness vs. layer number for PSS/PDADMAC deposited from 1.0 M NaCl………………………………………………………………………………... 16 Figure 9: Interior of a multilayer, scheme with two oppositely charged polymer strands…………………………………………………………………………….............. 17 Figure 10: For a PDADMAC/PSS system on Si substrate (a) Salt concentration vs. void water and swelling water and type of salt. (b) Salt concentration, type of salt and total water content……………………………………………………………………............... 18 Figure 11: Effect of pH vs. layer thickness for PDADMAC/PSS multilayers on Si/SiO2 surface…………………………………………………………………………………….. 20 Figure 12: Rejection model of multi-bipolar membrane by polyelectrolyte…………........ 21 Figure 13: Schematic drawing illustrating the principle of diffusion dialysis utilizing a cation exchange membrane to recover a base……………………………………….......... 23 Figure 14: Scheme of polyelectrolyte layer by LbL via dip coating…………………........ 26 Figure 15: Scheme of polyelectrolyte multilayer (PDADMAC-PSBMA-PSS) by LbL via dip coating……………………………………………………………………………. 27 Figure 16: UV-Vis absorption spectra of PDADMAC/PSS blockVN multilayer ……… 29 Figure 17: Dependence of PDADMAC/PSS films formed at various pH of solution on Si …………………………………………………………………………………………. 30 Figure 18: Schematic drawing of the six cell setup.…........................................................ 31 The EM3E Master is an Education Programme supported by the European Commission, the European Membrane Society (EMS), the European Membrane House (EMH), and a large international network of industrial companies, research centres and universities; www.em3e.eu Page IV Figure 19: Experimentally determined current vs. voltage curve measured in a laboratory electrodialysis stack with a 0.05 M NaCl-solution …………………………… 33 Figure 20: Schematic standard glass diffusion cell……………………………………….. 34 Figure 21: UV-Vis spectra for PDADMAC/PSS bilayers on (a) Quartz glass (b) FKB cation exchange membranes; n= number of bilayers from zero to seven………………… 37 Figure 22: UV-Vis absorbance spectra for PDADMAC-PSBMA-PSS multilayer and absorbance vs. number of trilayer at 280nm (inset). ……………………………............... 38 Figure 23: PDADMAC/PSS Contact angle vs. number of bilayers.………………............ 40 Figure 24: Contact angle vs. number of trilayers………………………………………… 42 Figure 25: Electrical Resistance vs. number of bilayers…………………………………. 44 Figure 26: Electrical resistance vs. trilayers. .…………………........................................ 46 Figure 27: Current-voltage curve for the relation between current through a membrane and corresponding voltage drop over the membrane and its boundary layer. (Measured at 50mM KCl and LiCl) ………………………….................................................................. 47 Figure 28: Limiting current density behaviour with different bilayer. ………………....... 48 Figure 29: Diffusion of K+ in Base, (top), bilayer 6 (middle) and bilayer 6.5 (bottom) … 51 Figure 30: Diffusion of Li+ in Base, (top), bilayer 6 (middle) and bilayer 6.5 (bottom)..... 52 Figure 31: Diffusion of K+ and Li+ in Base, (top), bilayer 6 (middle) and bilayer 6.5 (bottom). ………………………………………………………………………………….. 53 Figure 32: Iion flux vs. number of bilayers …………………………………………... 54 Figure 33: selectivity (K+/Li+) vs. bilayers.……………………………………………… 54 The EM3E Master is an Education Programme supported by the European Commission, the European Membrane Society (EMS), the European Membrane House (EMH), and a large international network of industrial companies, research centres and universities; www.em3e.eu Page V List of Tables Table 1: Ion exchange membranes (IEM) and its basic nature………………………….. 4 Table 2: Crystal radii (rc), Stokes radii (rs) and Gibbs hydration energy (-ΔGh) in water of various ions…………………………………………………………………………… 6 Table 3: General features of the polyelectrolyte by LbL dip coating…………………… 9 Table 4: List of different variables for PEM by LbL dip coating……………………….. 14 Table 5: K+ and Li+ flux in FKB and PEM by LbL……………………………………... 55 Table 6: Selectivity of K+/Li+ for single ion and mixed ion experiments………………. 55 7 2.3 Layer by layer (LbL) self-assembly There are many ion exchange membranes modification techniques are reported however, layer by layer (LbL) can be a promising alternative for the ion exchange membrane modification. In principle it can be defined as the alternating exposure of a charged substrate to solutions contains positive or negative polyelectrolytes, respectively [25,26]. The basic idea is that the sequential exposure leads to the adsorption on the surface due to the electrostatic interaction between the polyelectrolyte and the charged surface, therefore the surface either need to have charge or the neutral surface need to be modified for the adsorption. A charge inversion occurs for such self-assembly method. A charge inversion is the reversal of polarity due to an excess adsorption of oppositely chargedpolymer. The stability of such film depends on the electrostatic interactions [27,28]. Figure 2 shows a layer by layer (LbL) by dip coating where alternating exposure of the polyelectrolyte results in layer formation. Figure 2: LbL deposition scheme via dip coating [26]. The deposition cycle showed in the figure 2 repeats to achieve the desired multilayer on the substrate. A positively charged substrate is dipped in the solution that contains negatively charged polyelectrolyte. After an interval of time of adsorption, a washing step is followed to remove the excess polyelectrolyte from the surface. Then deposition of positively charged polyelectrolyte and a washing step completes one cycle of deposition and results one bilayer of polyelectrolyte multilayer (PEM) on the substrate surface [26]. 8 Apart from the LbL by dip coating two other layer deposition techniques are the dip coating and the spray coating and the spin coating. Figure 3 shows a schematic diagram for the spray and spin coating. Comparing among these techniques spray coating is faster but wastage of polyelectrolyte occurs due to the excess drainage during the coating. A spin coating is unable to coat a bigger surface area. In contrast the dip coating is a slower process then the other two but advantageous in coating a larger surface area with a small amount of electrolyte [29]. Therefore the layer by layer (LbL) by a dip coating technique is preferred. Figure 3: The LbL deposition by spray and spin coating [29]. 2.3.1 General features of polyelectrolyte multilayers (PEM) by LbL dip coating The remarkable feature of polyelectrolyte multilayer by LbL dip coating isa small amount of polyelectrolyte molecule can alter the surface as well as its properties. However, the multilayer by LbL dip coating propagates because of the reversal of charge. Each deposition step (Figure 2, step 1 & 3) leaves the surface primed for the next deposition step thus the PEM formation propagates with each immersion step in polyelectrolyte solution. Charge over compensation is the partial complexation of the second layer with the loops of the first layer and 1:1 polyelectrolyte stoichiometry is a general assumption for such association [30,31]. However, the polyelectrolytes multilayer by LbL dip coating gives unique features (Table 3) which also can be termed as the competitive advantage over bulk chemistry i.e. chemistry in general. For example a PEM by LbL dip coating allows achieving a vast surface coverage 9 with a tiny amount of PE material. In each deposition step it produces a thin, stable layer on the substrate but influence the surface properties significantly. A charge overcompensation and a charge reversal gives even more options for the surface properties tuning for example incorporating a targeted amount of charge by varying substrate, electrolyte types, deposition condition etc. As a consequence specific property like a monovalent ion selectivity and enhanced ion transport in the membrane are assumed to be achievable. Table 3: General features of the polyelectrolyte by LbL dip coating [31]. Variable Brief description Surface functional groups Accessible only from the solution side. Monolayer thickness 0.5 nm to 5 nm Typical surface area 0.20 nm2 per molecule, 5x1014 molecules per cm2. Mass density At a mass of 400 g/mol, 1 cm2 of a densely packed monolayer corresponds to 0.33 μg of material. Area coverage 5g (semi-preparative scale) covers 1500 m2. Monomolecular layers of polymer May be thinner and less dense and typically consist of 0.1 to 1.5 mg/m2. 2.3.2 PEM phenomena by LbL The formation of the layer is the combination of two consecutive steps; adsorption of polyelectrolyte on the substrate surface and the stability of this layer due to the electrostatic interaction. When a substrate is exposed in the polyelectrolyte solution, redistribution of the electrolyte particles occurs between the solution and the substrate interface which creates an interfacial potential difference [32]. Therefore the interface differs from the substrate and the bulk solution phase and ultimately leads to the adsorption. The electrostatic interaction between the polyelectrolyte chain and substrate surface results a stable layer. The PEM formation is highly dependent on the surface charge density (substrate) and polyelectrolyte chain in the solution. When surface charge density increases, the dilute chain of polyelectrolyte on the surface transforms into the semi dilute. As a result a thin layer forms on the substrate. These polyelectrolyte layers differ from the bulk polymer by leaving behind 10 an overcharged surface i.e. overcompensation of the surface charge by polyelectrolyte chain which results a molecularly layered multicomposite film with high degree of complexity on the substrate [33-38]. A fast adsorption is followed by slow rearrangement and transport of the chains to the surface occurs by diffusion so that electrostatic force can take place. A slow rearrangement enables the diffusion again into the inner region of previously deposited layer. By mixing positive and negative segments, irreversible complexation of the charges occurs. In each monolayer deposition we obtain the oppositely charged surface which is the charge reversal of the surface [37, 38]. The basic structure of a polyelectrolyte multilayered film can be divided into three regions as shown in the figure 4. Figure 4: A Scheme of PEM multilayer [40]. The PEM can be divided into three zones. The first zone is near to the surface, zone II is the bulk zone and zone III is the transition zone. Zone I is composed of few multilayers and the number of layer depends on the type of substrate. When a new layer is absorbed after having the arrangement shown in Figure 4, thickness of zone I and zone III stays the same and the zone II grows one layer more. However the thickness of each individual layer of zone I is smaller than the thickness of each layer of zone II. On the other hand all layers of zone II are composed of equal thickness of polyanion and polycation. Polyelectrolyte of the zone II mainly follows a 1:1 stoichiometry. It worth to mention that the transition between the layers are not as sharp as it is shown in the Figure 3. When PEM layer is fabricated, zone I completes first, then zone II and zone III. After forming zone III, each new layer addition increases the zone II by transiting one layer from I III II Substrate Interior multilayer film Surface 11 zone III, but the overall thickness of zone I and zone III stays the same [40]. While conventional multilayer growth suggests a symmetric trend, an asymmetric growth model is proposed for PDADMAC/PSS polyelectrolyte on the silicon surface [41]. One possibility of asymmetric growth could be the absence of the surface ions. In this case the charge reversal cannot occur symmetrically with each deposition step and thus an asymmetric growth originates. Whenever the charge is not balanced by ion pairing between polyelectrolyte repeat units counterions are found whether in the bulk or at the surface of the polyelectrolyte multilayer. In polyelectrolyte multilayer polymer and counterion charge balance is often termed as extrinsic and intrinsic charge balance. Extrinsic charge is defined as the charge by the polymer and counterion interaction and intrinsic charge occurs when polymer ions interact. In Figure 5, extrinsic and intrinsic charge balance is schematically shown. Figure 5: Scheme of intrinsic and extrinsic charge balance in the PEM [41] However the overall charge is balanced by the combination of extrinsic and intrinsic charges which also gives a stable multilayer on the substrate. After forming a number of layers it is observed that the alteration of the charge does not follow symmetric order rather complete alternation of the charge occurs only near the surface. Figure 6 shows that after forming a dozen layers, excess positive sites begin to accrue in the multilayer. Treating the surface as a reaction-diffusion region for pairing of polymer charges, a model profile was shown in figure 6. Different reaction-diffusion ranges of positive and negative polyelectrolyte charge lead to a coverage of glassy, stoichiometric complex growing on the top of a layer of rubbery, PDADMAC-rich complex [41] - ClClClClClClClClClClNa+ Na+ Na+ Na+ Na+ PDADMAC PSS Substrate Extrinsic sites Intrinsic sites 12 Figure 6: Asymmetric growth phenomena of PSS/PDADMAC layer buildup. Layer number represents the PSS/PDADMAC alternatively [41]. It is reported that the complete alternation of the charge occurs only near the surface and an intrinsic negative thin layer forms with almost all PDADMAC deposition steps. Also after 10-12 layers, PSS no longer compensates all the PDADMAC charges therefore PEM contains anions all the time which is the crucial point for asymmetric growth. The phenomenon can be explained as when PDADMAC is on the top layer, many positive sites are found and in addition PSS all positive sites are consumed. After deposited the PSS, a stoichiometric film produces but with PDADMAC multilayer forms with excess PDADMAC. Thus, the growth becomes asymmetric; however the salt concentration and the substrate are important in such system. In general, it worth to mention that that for PDADMAC/PSS system after a sufficient number of layers, a persistent layer of PDADMAC remains within the bulk of the film, and the alternation of charge occurs only near the surface. However polyelectrolyte repeat unit is a part of long chain, for this reason a thin layer has a strong dependence on polyelectrolyte molecular weight. The adsorption of PSS is rate limiting therefore PSS molecular weight has a stronger dependence on layer [41]. Nevertheless the layer formation is a complex method with a diverse dependency on the multiple variables which need to be considered for the desired surface structure. A number of parameters and their effect on layer growth are explained in the following section. Extrinsic negative Extrinsic positive Intrinsic Thickness 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 13 2.4 Variables for PEM by LbL A successful LbL growth is primarily guided by longer adsorption time, rinsing volume of water and surface coverage. Longer adsorption time favours to the reproducibility because the plateau of the adsorption depends on polyelectrolyte concentration and adsorption time. The polyelectrolyte (PE) concentration is easy to reproduce but at the initial phase of adsorption small difference of time causes large difference in the adsorbed mass. And close to the adsorption plateau even large difference of time lead only a small difference of adsorbed mass. [42]. The volume of rinsing water is important because of the possibility of cross contamination. Whenever the substrate is removed from one polyelectrolyte solution, it adhere some excess of polyelectrolytes and can create cross contamination with the next PE deposition step. Therefore a dilution factor needs to be considered to avoid cross contamination. Dilution factor is calculated by dividing the volume of the first rinsing bath by the estimated volume of the adhering liquid. The number and the rinsing bath are normally chosen so that the overall dilution factor is maintained at least 1:106. The surface coverage of the functional groups is one of the most important key parameters for reproducibility. Most of the LbL shows linear growth which is associated with the functional group densities on the surface. Increasing or decreasing the surface coverage of the functional groups or the molecule diffusing in the whole surface able to influence the polyelectrolyte multilayer (PEM) linear growth. However for PDADMAC/PSS system, the growth can be different then a linear which is explained by figure 5 and figure 6. The substrates with few functional groups permit a molecular orientation towards the surface when absorbed. The polyelectrolyte with a high degree of polymerization gives a large number of functional groups, thus the orientation become less. Therefore, reproducible result for successful polyelectrolyte multilayer growth on any substrate is highly dependent on surface coverage, deposition condition and molecular weight of the polyelectrolyte [43-45]. Nevertheless, polyelectrolyte multilayer formation depends on a wide range of parameters. Table 4 gives a list of different variable which shows that polyelectrolyte multilayer dependency is a function of not only different parameters but also different possible interactions. Therefore a careful consideration of parameter (Table 4) control needs to consider for a successful layer growth with a reproducible standard. 14 Table 4: A List of different variables for PEM by LbL dip coating [45]. Different interactions for LbL Important parameters (Primary) Other important parameters 1. Electrostatic interaction. 2. Donor/acceptor interactions 3. Hydrogen bridging 4. Adsorption/drying cycles 5. Covalent bonds 6. Stereocomplex formation or specific recognition 1. Individual layer & its thickness 2. Type of surface, salt & PE concentration, surface charge. 3. Surface properties such as nature and density of the charged groups, local mobility (in case of a polymeric surface) 1. Solvent 2. Concentration of adsorbing species 3. Adsorption time 4. Temperature 5. Nature and concentration of added salt 6. Rinsing time 7. Humidity 8. Drying 9. Agitation or rinsing 10. dipping speed etc. The deposition condition is a vast aspect for PEM as the target thickness with functionalities and desired properties are often expected; however the type of polymer, salt concentration and the deposition time are often considered as most important. It also worth to mention thatthe structure and the properties of each layer is governed by the respective polyanion/polycation pair and the deposition conditions. Therefore the choices of parameters are also different for different polyelectrolytes. As an example, the layer deposition for PSS/PAH is strongly pH dependent whereas PDADMAC/PSS is mostly not [45]. After several layer formation the properties of the substrate are hindered by the polyelectrolyte which means the properties of thin polyelectrolyte multilayer becomes dominant over the properties of the substrate. Therefore it is often said that polyelectrolyte multilayer films are independent of the underlying substrate. 2.4.1 Effect of temperature The layer thickness increases with increasing temperature and relative humidity. Even small differences in temperature can easily account for changes in film thickness of the order of 5– 15 10% depending on the swellability of the film, however can be overlooked with the standard laboratory conditions. A film thickness vs. temperature for a (PDADMAC/PSS) 10 is shown in figure 7. Figure 7: Effect of deposition temperature on the thickness of a (PDADMAC/PSS) 10 multilayer on Si wafer and deposited at 1M NaCl. [46] The thickness of layers increases with the increasing temperature. The possible explanation is the change of interaction with temperature. At elevated temperature the solubility of PSS in water decreases which can be related with the solvent quality i.e. water. Thus solvent quality is reduced when temperature is increases and results a thicker film [30,47]. In addition to that another argument is about the possibility of the hydrophobic effect. The solubility of hydrophobic materials decreases at high temperature. The reduced solubility is assumed to drive the polymer to the surface and increase the thickness of adsorbed layers. It seems that at high temperature both localized dissociation and conformational dynamics of polyelectrolyte molecules tend to increase which results an increased tendency to send the loops on the surface and tail to the solution [30, 46] 2.4.2 Effect of salt concentration The polyelectrolyte multilayer formed at different salt concentration shows different fluorescence emission intensity shown in figure 8. The measurement was performed by the fluorescence emission spectrometer on a solid sample holder. The incident angle was 45° to the film surface. The emission intensity was taken at 396 nm as the characteristics of pyrene 16 and evaluated after subtracting the intensity of a cleaned quartz slide (as a blank) to eliminate the substrate emission and scattering. It was shown that the salt concentration increases as the intensity increases. A second example is shown in figure 8(b), as the layer number increases the thickness increases. Thicknesses were determined using an ellipsometer with 632.8 nm radiation at 70° incident angle. [48,49]. Figure 8: 8(a) Fluorescence intensity varying with the layer number of the multilayer film made of ADPy-100 (The pyrene labeled polyanion) and PDADMAC deposited in NaCl solutions [42]. 8(b) Thickness vs. layer number for PSS/PDADMAC deposited from 1.0 M NaCl [49]. While increasing the ionic strength of the medium, the layer becomes thicker. At a higher salt concentration, the polyelectrolyte trends to be coiled more and increases the thickness which can be explained by the intrinsic charge compensation, schematically shown in figure 9 [48]. Compensation via polycation–polyanion complexation is called intrinsic charge compensation. However, increasing the salt concentration will increase the extrinsic charge compensation, which reduces the number of complexation points between two polyelectrolytes, allowing the more possibilities to coil or to move through the film which gives rougher and thicker structures. It worth to mention that, at a very high salt concentration the layer can be de-structured due to the high number of counterion interactions [50-53]. Therefore salt concentration effect is obvious on polyelectrolyte multilayer which indicates a careful choice of salt concentration; necessary to obtain the desired thickness. 7(a) 7(b) 23 has the limitation to the low processing capability and low efficiency [89]. The applications of diffusion dialysis are recover acids and valuable metals from industrial waste solution [90] The principle of the diffusion dialysis process is shown in the figure 13 (recovery of base) considering a cation exchange membrane. Figure 13: Schematic drawing illustrating the principle of diffusion dialysis utilizing a cation exchange membrane to recover a base [4]. The mass transport in diffusion dialysis is determined by the transport of ions through the membrane. The feed and the product solutions are separated by the membrane. In order to promote the process efficiency a practical consideration is to decrease membrane thickness and increase the membrane area. [7]. The electrodialysis is an electromembrane separation technique uses ion exchange membrane arrangement and also a difference in electrical potential to separate ions from solution and from each other [91]. Ion transport by the diffusion dialysis and electrodialysis, in general, is described by the Nernst-Planck’s equation (Equation 1). The equation illustrates three forces diffusion, migration and convection. The simplified expression is as- (1) Feed Salt solution Product (Base) Feed (Salt or base mixture) Na+ Na+ SO42OHOH- 24 Here J is ionic flux [molm-2s-1], v is convection velocity in [ms-1] c is concentration of component [molm-3] D is diffusion co-efficient of the component [m2s-1], x is direction coordinator [m] z is valence number of component I [-], F is Faraday constant [Cmol-1], R is gas constant [Jmol-1K-1] T is temperature [K] and ɸ is electric potential [V] In the equation first term represents the diffusion which is first Fick law; second term is the migration force and third is the convection. Since for the dense membrane no convection occurs the equation can be rewrite as equation (2) (2) In the following paragraph, definition of important parameters such as membrane degree of swelling, resistance and flux are given which often considered to describe diffusion dialysis and/or electrodialysis. 2.6.1 PEM and diffusion dialysis The modification of the cation exchange membrane and applying it in diffusion dialysis could give specific cation selectivity. The nanoscopic PEM allows significant changes in morphology and/or surface properties like surface charge density. These changes can effectively contribute to improve on cation separation, recovery of ions, water purification, acid and base separation or recovery. [92,93]. The LbL deposition by the electrostatic interaction affects the ion exchange sites [94], size of the ion and their mobility. Under the influence of diffusion dialysis condition these parameters could give monovalent ion selectivity by diffusion. A diffusion dialysis with the high cross linked dense membrane (by PEM) may be able to hinder the larger size ions and gives size exclusion. Positive layer on the cation exchange membrane can make higher repulsion to the multivalent ions which also affect the monovalent ion selectivity [2, 33, 94-99]. 25 3. Experimental 3.1 Materials The polyelectrolyte multilayer (PEM) samples were prepared on FKB Fumatech cation exchange membranes which were purchased from FuMA-Tech GmbH, Germany. The layer by layer technique was followed by using polycation PDADMAC*, polyanion PSS** and poly zwitterion PSBMA*** Polyelectrolyte multilayer deposition was conducted in a salt medium prepared by sodium chloride, NaCl (≥99.5 %,). Different inorganic salts including sodium chloride, NaCl (≥99.5 %,), potassium chloride, KCl (≥ 99%), and sodium sulphate, Na2SO4 (≥98%) were purchased from Sigma Aldrich. Lithium chloride, LiCl (≥98%) was obtained from Fluka analytical and hydrochloric acid, HCl (37%) from Mereck, Germany. MiliQ milipore demiwater was used for other auxiliary purpose. All chemicals were laboratory grade standard. *Poly (diallyldimethylammonium chloride), Mw = 150 kDa, 20 wt % in water from Kemira, Finland). ** Polystyrene sulfonic acid, Mw = 100 kDa, 20 wt % in water from Tosoh Organic Chemical Co., LTD (Japan). *** PolyN-(3-Sulfopropyl)-N-(methacryloxyethyl)-N,Ndimethylammonium betaine (SBMA) from Sigma Aldrich (The Netherlands) . 3.2 Methods 3.2.1 Preparation of polyelectrolyte multilayer by PDADMAC and PSS The polyelectrolyte multilayer sample was prepared on the fumasep® FKB cation exchange membrane. The membrane was preconditioned in demineralized water for six hours to obtain the optimal performance, minimal wrinkling and lowest electrical resistance. Since the membrane was already in H+ form therefore further acid treatment was avoided. In the following step, a PDADMAC and a PSS bath was prepared by mixing 1g/l of each polyelectrolyte in 0.2 M NaCl solution. Both the solution was vigorously stirred for approximately 30 minute in room temperature for proper mixing of the polyelectrolyte in the 26 salt solution. Another two baths were prepared by the demiwater for rinsing purpose. The volume water in each bath was kept as same as the polyelectrolyte baths. Preconditioned FKB membrane was then cut into approximate 3.5x 3.5 cm and taped over a square shaped glass support. Each four side of the square shaped membrane was taped on the glass support keeping approximately 3x3 cm membrane area to be exposed in the solution. The sample was vertically dipped in the PDADMAC solution bath and then rinsed in water bath. At this step a monolayer i.e.0.5 bilayer was formed. The next successive step sample was dipped in the PSS bath and then rinsed in the water. At this stage 1 bilayer was formed. The duration of the layer deposition and rinsing was followed as 30 min/bath. A scheme of LbL formation on FKB membrane.by dip coating is shown in figure 14. Figure 14: A Scheme of polyelectrolyte multilayer by LbL via dip coating. The cycle of deposition steps were repeated to obtain desired bilayers on the FKB membrane. All samples were prepared at room temperature and the duration of each deposition step was considered as 30 minute. In each deposition step membrane sample kept as vertical and rinsing water was replaced by new after two times rinsing. Several sets of sample were prepared up to 7th bilayer. The prepared sample was stored in 0.5M NaCl solution. + + + + + + + + + + + + + + + + + + + + + + + + + - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Demi H2O Demi H2O - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + + + + + - - - - - - - 1. Polycation 4. Rinse 2. Polyanion 3. Rinse 27 3.2.2 Preparation of the polyelectrolyte multilayer by PDADMACPSBMA-PSS The preparation method was same as described in the section 3.2.1. The difference is an additional deposition step was employed in between the PDADMAC and PSS; deposition of PSBMA which is a zwitterionic polymer, exhibits both positive and negative charge in the molecule. To prepare the solution 1g/L polymer was dissolved in the 0.2M NaCl solution. Vigorous stirring was dene until the polymer completely dissolves. Dissolving PSBMA took longer time then PDADMAC and PSS. The other condition of the deposition such as salt type and its concentration, deposition time kept as constant. After PDADMAC-PSBMA-PSS deposition, the sample considered as one trilayer. In this PEM sample preparation, PSBMA layer is sandwiched in between the PDADMAC-PSS layers. Figure 15 shows a scheme of the polyelectrolyte multilayer with zwitterion polymer PSBMA where at first step PDADMAC is deposited and successively PSBMA and PSS deposition war performed. Sample considered as PDADMAC terminated and PSS terminated layers. Sample was prepared up to seven trilayers. Figure 15: Scheme of polyelectrolyte multilayer (PDADMAC-PSBMA-PSS) by LbL by dip coating. 1. Polycation 2. Rinse 3. Zwitterion 4. Rinse 5. Polyanion 6. Rinse + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + - - - - - + + + + + + + + + + + + + + + - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + Demi H2O Demi H2O Demi H2O 28 Deposition of PEM on the FKB membrane was conducted at the salt concentration 0.2M NaCl because at salt concentration 0.2M, PDADMAC/PSBMA the layer deposition is observed as most stable [104] 3.2.3 UV-Vis measurement The UV-Vis was measured by the spec: USB 2000+Miniature fiber optics from Ocean Optics, Inc, USA. The UV-Vis set up consist a UV source (Deuterium light), optical fiber connections, membrane holding cell, detector and recorder (Ocean plus software). Absorption occurred in the membrane holding cell where the membrane was placed. During measurement UV light was allowed to passes through the optical fiber line to the membrane holding cell. Another optical fiber was connected from the membrane holding cell to the detector. After absorption in the cell the detector responses are recorded by the Ocean plus software that gives a UV-Vis profile with a characteristic peak. UV-Vis measurement was conducted for several sets of membranes. First of all PDADMAC/PSS polyelectrolyte multilayers (PEM) were deposited on the negatively charged quartz glass surface. Before layer deposition, quartz glass was cleaned by the acid piranha solution. Acid piranha is a 3:1 mixture of concentrated sulphuric acid (H2SO4) with hydrogen peroxide (H2O2). The methodology adopted from http://www.lamp.umd.edu/Sop/Piranha_SOP.htm. Each bilayer formation on the quartz glass was conducted by following the same procedure as described in section 3.2.1. After each bilayer deposition, the sample was dried under a stream of N2 and then UV absorbance was measured. Up to seven bilayers UV-Vis profile were measured and recorded. Referring to the section 3.2.1 and 3.2.2, two other sets of sample (FKB modified by PDADMAC/PSS and PDADMAC/PSBMA/PSS) up to seven bilayers were prepared and dried under N2 stream before the experiment. Ultraviolet visible spectroscopy is a quantitative measurement technique of characterization based on the Beer-Lambert law (Equation 4). A = log10 (I0/I) = εcL (4) 29 Where A is absorbance in AU, I0 is the intensity of the incident light at the given intensity and I is the transmitted intensity, ε is the molar absorptivity i.e. often called extinction coefficient, c is the absorbed species concentration and L is the path length through the sample. An example of UV-Vis absorption spectra is shown in figure 16. Figure 16: UV-Vis absorption spectra of PDADMAC/PSS blockVN multilayer [78]. It was reported that PSS gives absorption band at 228 nm and PDADMAC absorption is negligible in the spectroscopic region. Nevertheless there is no distinct maximum for PSS and one can observe in the range of 280-320 nm [102]. 3.2.4 Contact angle measurement The contact angle (CA) was measured by the Data Physics OCA20, Germany. The equipment is a telescope-goniometer in which a horizontal stage is used to mount a solid sample; a micrometre syringe was used to form a water liquid drop (ultrapure water as a probe liquid), an illumination source, and a telescope with camera. Before CA measurement, the membrane sample was dried overnight in the vacuum oven at 30°C to get a dry membrane surface. Then the membrane was placed as flat as possible on the horizontal stage. The micrometre syringe (filled with ultrapure water) was in vertical position on the membrane. The illumination source and the camera focus were adjusted to 30 have the vision of the membrane surface. OCA 20 software allows making a precise water droplet of 1μl and dispenses it on the membrane surface. Then a snapshot was taken and CA was measured. For a single sample, three to five droplet were dispensed in different position of the surface and corresponding snapshot were recorded which were used to measure contact angle. taken for each sample and in different place of the membrane and corresponding contact angle was measured. After dispensing water droplet (1μl) on the membrane, five second waiting time was considered for the consistency. Contact angle measurement gives the information of the hydrophilicity/hydrophobicity and surface energy of PEM. Small contact angles (<<90°) correspond to high wettability, while large contact angles (>>90°) correspond to low wettability. An example of water contact angle vs. number of PDADMAC/PSS multilayer on Si substrate is shown in figure 17. [64] Figure 17: Dependence of PDADMAC/PSS films formed at various pH of solution on Si. [64] It was reported that PDADMAC/PSS multilayer on Si substrate does not show any pH dependency however PDADMAC terminated layers give higher contact angle then PSS terminated. Nevertheless CA value differs significantly from substrate to substrate which gives the information of surface dominated characteristic of PEM [64,103]. Films terminated by polycation layers are more hydrophobic. Also amplitude of layer-to-layer contact angle oscillations and average hydrophobicity depends on the polyelectrolyte adsorption conditions. 31 3.2.5 Measurement of electrical resistance and current voltage curve The electrical resistance was measured in a six cell electrodialysis unit (Figure 18).in which two outer compartments contain two working electrodes; the anode and the cathode. These electrodes were used to apply an electric field though the membrane. In the middle there were two central compartments with Haber-Luggin capillaries. A shielding compartment in between each electrode and main compartment was used. The membrane under investigation was placed in between Haber-Luggin capillaries. Other cell membranes (often called principle auxiliary membranes) of the figure 18 were used to prevent transport of water dissociation product from electrode to main compartment. The Haber-Luggin capillaries were filled with 1 M KCl solution which was connected to calomel reference electrode (Schott B2810) by silicon tubing. The reference electrode was connected with Autolab potentiostats/galvanostats, which allowed controlling the current level and measures corresponding voltage drop in the test cell. Figure 18: Schematic drawing of the six cell setup. [1] In electrode compartments 0.5 M Na2SO4 (2L) and in other compartments 0.5 M NaCl (2L for shielding and 2L for central compartment) were pumped at a rate of 450ml/min and temperature was maintained as 25°C. The Haber-Luggin capillary tip of each electrode was placed as close as possible to the membrane but not in contact with the test membrane to CEM CEM Test AEM CEM Anode Cathode 32 avoid the interference of the electric field. All salt solutions were allowed to circulate for 1015 minutes to obtain a stable condition for electrical resistance measurement. Current was applied in the range of zero to 135 mA with stepwise increase (15mA/step). For each step of current increase, corresponding voltage were measured and recorded. Currentvoltage data were further calculated and plotted as current density (mA/cm2) vs. voltage drop (mV). A linear relation was found and the slope was calculated which eventually gave the electrical resistance, R (Ω.cm2). Before the measurement, the membrane sample was preconditioned overnight in 0.5M NaCl solution to be stabilized. The measurement with the sample membrane gives a total resistance of membrane and solution. Therefore a blank run (without any membrane sample) was conducted to measure the resistance contribution of the salt solution. Rmembrane = Rmembrane+solutionRsolution The electrical resistance of an electrical conductor is the opposition to the passage of an electric current through that conductor. The electrical resistance of ion-exchange membranes is one of the factors which determine the energy requirements of electrodialysis processes. The resistance (R) of an object is defined as the ratio of voltage across it (V) to current through it (I), Thus R=V/I (3) The specific membrane resistance is usually reported as [Ω cm2] or [Ω m2] 3.2.6 Limiting current density The limiting current density (ilim) was measured with the same set up as described in section 3.2.5 however higher level of current was used to determine ilim so that ohmic, limiting and over limiting region can be obtained From the experiment current (mA) and corresponding voltage drop (V) data was obtained which further calculated and plotted to obtain a current density (mA/cm2) vs. voltage (V) curve. For each measurement, two fresh samples were considered to confirm the reproducibility of the results. 39 The absorbance spectra are quite similar to the figure 21(b). The absorbance peak was assigned at 280 nm and the range of the absorbance peak (figure 22) could be a result of absorption of PSS (peak range 280-320 nm) and PEEK.(around 335nm) as explained before (section 4.1.1.1) [65,109]. A steady increase of the absorbance peak was found with the increase of number of trilayers which means that the trilayer growth on FKB membrane by PDADMAC-PSBMA-PSS was successful. The absorbance vs. number of trilayer at 280 nm gives a linear trend. Thus we can say that the multilayer deposition was stoichiometric. Comparing with figure 21(b) and figure 22, the absorbance for base membrane (n=0) and linear trend were found same but the slope of inset figures are different. Hypothetically we can say that PSBMA layer influences the deposition trend and thickness. PSBMA is zwitterionic and growth of PEM may not solely rely on charge-charge interaction [104]. Nevertheless, for PDADMAC/PSS layer thickness linearity depends on molecular weight along with other variables [110] but for PDADMAC-PSBMA-PSS, more literature investigation and experiments are required to have the idea about the layer thickness and other characteristics. 4.1.2 Characterization by contact angle The contact angle measurement was used to characterize the polyelectrolyte modified membrane by its hydrophilic/hydrophobic behaviour. The water contact angle of PDADMAC-PSS multilayers on FKB membrane and PDADMAC-PSBMA-PSS modified FKB membranes are shown in this section. 4.1.2.1 Contact angle measurement for PDADMAC-PSS modified FKB The contact angle (CA) vs. number of bilayer is shown in figure 23 in which bilayer 0 is for the FKB membrane without any modification. Other than bilayer 0 each (+) marked data shows the CA of PDADMAC terminated layers and (-) marked ones are for PSS terminated layers. (Appendix B for contact angle values) An oscillating change of the contact angles upon alternating deposition of PDADMAC and PSS can be observed from figure 23. In general, all membrane samples show hydrophilic behaviour except the 0.5 bilayer which is slightly hydrophobic. An odd/even and hydrophilicity/hydrophobicity trend is observed from figure 23. 40 0 1 2 3 4 5 6 7 8 30 40 50 60 70 80 90 100 -, PSS terminated +, PDADMAC terminated + + + + + + - - - - - - - Contact angle () Number of bilayers - + FKB Figure 23: PDADMAC/PSS Contact angle vs. number of bilayers. This odd-even trend is explained by the polar surface properties of PEM multilayers [103]. The polarity reverts with each monolayer deposition. Also we can draw attention to the charge reversal phenomenon of polyelectrolyte multilayer (described in section 2.3.2). Thus in each PDADMAC deposition step the surface become positive and shows higher CA value and in each consecutive PSS deposition step the surface become negative which gives lower CA value. In other words each PDADMAC terminated layer shows more hydrophobic nature then PSS terminated layers and contact angle values are dependent on outermost layer [59]. We also can observe that layer to layer oscillation is in between 20 -40°. Köstler et al (2005) [103] reported about CA odd-even trend when PDADMAC-PSS multilayer was deposited on PTFE and PET substrate but CA oscillation differs from substrate to substrate, for example PDADMAC-PSS multilayer on Si gives a CA oscillation about 10° whereas on PTFE gives around 20° for same electrolyte pair. The reason of such behaviour can be explained by the low surface energy character. It has been reported that PTFE has a low surface energy character and gives high layer to layer oscillation; in contrast Si has comparably high surface energy and gives low CA oscillation [65,103]. From this study we can say that our FKB might have a low surface energy character and thus gives a high contact angle oscillation. We can draw a contradiction here; if the outermost layer and underlying substrate are solely responsible for contact angle oscillation then CA for all PDADMAC terminated layers and 41 CA for all PSS terminated layers are expected to be same but the experimental results shows that for each PDADMAC and PSS terminated layers, contact angle values decreases with increasing the layer number. We can say that there might be an effect of underlying polyelectrolyte layer. Hsieh et al (1997) [111] has reported that water contact angles for same polyelectrolyte pairs exhibits different values depending on the underlying polyelectrolyte layers which suggests a high degree of layer interpenetration. Another study of Faibish et al (2002) [112] suggests that the contact angle value is related with the coating density. For first few layers the coating can be assumed as incomplete. As the layer number increase, coating density also increases which may can contribute to lowering the contact angle values. Nevertheless a high degree of interpenetration of polyelectrolyte gives a very thin layer and PEM in contact of water swells immediately and polymer become hydrated which increases the thickness. It means the reduction of degree of interpenetration at swollen state [113]. Hence water contact angle can be assumed to be affected only by the outermost layer. Meanwhile hydrophbization effect is relevant since the contact angle exhibits different value with different drying protocol [59,65,103]. We can summarise that the odd-even and hydrophilic/hydrophobic effect occurs because of the polarity change with each layer deposition. The FKB substrate contributes to significant effects on the oscillation together with the outermost layer. As the number of bilayer increases higher degree of layer interpenetration becomes gradually effective with an increasing coating density therefore the contact angle value decreases with increasing the layer number (considering the CA trend separately for all PDADMAC terminated and PSS terminated layers) i.e. increases the hydrophilicity with increasing the layer numbers while exhibiting the same odd-even trend. In general we can say that-as the number of bilayer increase (PDADMAC-PSS modified FKB membrane), no matter whether the membrane is PDADMAC terminated or PSS terminated, the hydrophilicity of the membrane increases with the number of bilayers. Extrapolating the hydrophilic behaviour it also can be assumed that at higher bilayer number PDADMAC terminated layer can be sufficiently hydrophilic and may be able to influence ion transport behaviour. 42 4.1.2.2 Contact angle measurement for PDADMAC-PSBMA-PSS modified FKB The contact angle (CA) of the PDADMAC-PSBMA-PSS vs. number of trilayer is shown in the figure 24 0 1 2 3 4 5 6 7 8 20 40 60 80 100 -, PSS terminated +, PDADMAC terminated - - - - - - - FKB + + + + + + - Contact angle () Number of trilayers + Figure 24: Contact angle vs. number of trilayers Contact angle for FKB base membrane and first monolayer gives the higher values than other trilayers. A significant decrease of CA can be observed at first trilayer and then an alternating trend of CA is observed. It can be noted that first PDADMAC terminated layer refers to a single monolayer on the surface and trilayer 1 forms with a PSBMA layer in between PDADMAC and PSS layers. At this point the effect of PSBMA can be noticed by comparing with Figure 23 which suggests that PSBMA make the membrane more hydrophilic and also reduces the layer to layer oscillation to 10° to 15°, similar to the PDADMAC-PSS multilayer on a Si [65,103]. Also we can say that PSBMA can change the hydrophilicity/hydrophobicity of the PDADMAC/PSS multilayer system and the surface charge characteristics. PSBMA grafting on hydrophobic surface can reduces contact angle about 15° reported by Chang et al (2011) [118]. Taking into account the effect of PSBMA we can say that the first PDADMAC terminated layer was found slightly hydrophobic. Later contact angle decreases significantly because of PSBMA and PSS layer (CA of trilayer 1). Comparing with Figure 22, it seems like the contact angle reduces due to the contribution made by PSBMA because the substrate, other polyelectrolytes and deposition condition was same. From trilayer 1 to 7, 43 the similar oscillating behaviour can be observed. We can assume that; at PSS terminated trilayer 1, surface charge density is already influenced by the PSBMA which affects the next PDADMAC deposition step and so on. Nevertheless substrate effect on PEM is reported as significant for PDADMAC/PSS system [111] whereas PDADMAC/PSBMA is substrate independent [104] however PSBMA affects the charge reversal process [119]; also can be understood by the UV-Vis linear trend comparison between figure 21(b) and figure 22. Thus hydrophilicity/hydrophobicity controlling by PSBMA could be an interesting aspect to investigate in future. 4.2 Membrane performance 4.2.1 Membrane electrical resistance The electrical resistance (R) of the membrane was measured by a direct current method in 0.5M NaCl salt solution. A current-voltage relation was measured in ohmic region which gives the membrane electrical resistance. Polyelectrolyte multilayer on cation exchange membrane and determining their electrical resistance behaviour were the primary focus. The membrane electrical resistance is significant because it can give information about the membrane characterization and performance. The electrical resistance of PDADMAC-PSS multilayers on FKB membrane and PDADMAC-PSBMA-PSS modified FKB membranes are shown in this section. 4.2.1.1 Electrical resistance for PDADMAC-PSS modified FKB The membrane electrical resistance (R) vs. number of bilayer is shown in figure 25 Bilayer 0 represents the FKB membrane without any polyelectrolyte layer. Other than bilayer 0 each (+) marked points are the resistance of PDADMAC terminated layers and (-) marked points are the resistance for PSS terminated layers. As the number of bilayer increases an alternating trend of resistances are observed. Considering all PDADMAC and PSS terminated layers separately, the membrane resistance increases with the increase of bilayer which is opposite to the contact angle values. Comparing figure 23 with figure 25 we can see that membrane electrical resistance increases as the number of bilayer increases whereas contact angle decreases i.e. hydrophilicity of the membrane increases. 44 0 1 2 3 4 5 6 7 8 2 4 6 8 10 12 14 16 18 20 - - - - - - - + ++ + + R (.cm2) Number of bilayers + - + (+) PDADMAC (-) PSS Figure 25: Electrical Resistance vs. number of bilayers. FKB cation exchange membrane, by nature it has a negative surface (See appendix C1 for the properties of the FKB cation exchange membrane). According to charge reversal process (described in section 2.3.2) each PDADMAC deposition step convert the surface from negative to positive which gives a high R value; in contrast each PSS deposition step make the surface as negative which gives lower R value. [27,28] The alternating deposition of PDADMAC-PSS thus gives a zig-zag trend of electrical resistances. This behaviour exhibits more likely an odd-even trend (similar to the contact angle result shown in figure 22). The reason can be the surface potential of the outer layer. The odd-even effect is defined as the reversible variations with the sign of charges of the terminating layer [106]. As evidence, ζ-potential of PDADMAC/PSS has been reported as reversibly alternating [105]. Thus one hypothesis can be the surface potential changes with every layer deposition results the odd-even electrical resistance trend [107]. When surface become positive due to the PDADMAC terminated layer, it contributes to repeal the cation to pass though i.e. Donnan exclusion may become dominant which can also contribute to increase the resistance. For the base membrane the electrical resistance was found as 3.98±0.24 ohm.cm2 which is an agreement with the supplier information, <4 Ω.cm2 (at 0.5M NaCl). At this point we can 45 comment that the membrane was correctly preconditioned and was in H+ level for the deposition of the polyelectrolyte. With a PDADMAC monolayer the resistance was found as 13.01±0.34 ohm.cm2. In successive PSS deposition steps the R value come back as 5.45±0.74. (Appendix C.2 for resistance data). In general monolayer thickness is reported as 0.5 nm to 5 nm (Table 3) and the FKB membrane thickness was 80-100 μm (appendix C.1). The thickness of monolayer on FKB membrane is very small compare to the original membrane thickness. We can say that the thickness of PEM layer solely should not contribute to high resistance rather a combination of surface potential, cation Donnan exclusion and thickness due to swelling behaviour can be assumed. Taking into account the asymmetric model for PSS/PDADMAC system which suggests a PEM with PDADMAC terminated layer has many positive sites and thickness increment is linear (<10-12 bilayers) [40]. Taking into account all information above; we can mention about several possibilities of the resistance zig-zag behaviour. First of all alternating surface potential occurs due to each monolayer deposition which gives the zig-zag resistance trend. Though each monolayer thickness is considerably small but the complexity of the structure obtained from each monolayer cannot be avoided at higher number of layers. Hence it can contribute to increase the resistances. Higher resistance values for PDADMAC terminated layer originates because of the swollen thickness and surface potential along with the Donnan exclusion as PDADMAC terminated layer has many positive sites according to the asymmetric model. Again taking into account the contact angle results shown in figure 23 it was assumed that as the number of bilayer increases the layer interpenetration and coating density increases which reduces overall contact angle however; the coating density and layer interpenetration also can give surface complexity which may can increases the overall resistance (Figure 23). Nevertheless the resistance measurement in a direct current method gives not only the membrane resistance but also include resistance of membrane-solution interface i.e. diffusion boundary layer and electrical double layer [101]. Therefore it could be an assumption that resistance of polyelectrolyte multilayer can be influenced by the electrical double layer and diffusion boundary layer, when exposed in the solution. 46 4.2.1.2 Electrical resistance for PDADMAC-PSBMA-PSS modified FKB The electrical resistance (R) vs. number of trilayers of PDADMAC-PSBMA-PSS modified FKB membranes are shown in the Figure 26. 0 1 2 3 4 5 6 7 0 3 6 9 12 15 18 21 FKB -, PSS terminated +, PDADMAC terminated - - - - - - - + + + - ++ + R(.cm2) Number of trilayers + Figure 26: Electrical resistance vs. trilayers. The layers which are PDADMAC terminated show higher resistances then PSS terminated layers and resistance changes alternatively with each monolayer deposition. The reason can be explained by the surface charge reversal. Though in every trilayer there is a PSBMA zwitterion layer in between, even so resistances zig-zag trend is comparable with the figure 23 which primarily gives the assumption of charge inversion due to cationic and anionic terminated layer depositions. [27,28]. From this comparison it seems that PSBMA layer does not effect on the electrical resistance. It should be noted that for first few PSS terminated layer R was found smaller than the FKB base membrane which ideally should not happen. Considering the increase of layer thickness with the increase of the number of trilayers (as shown in figure 22); it was expected to see the resistance increasing trend (considering R separately for all PDADMAC or PSS terminated layers) with the increase of the number of trilayers (similar results as shown in figure 25). However, the reason could be the experimental error during the measurement and partial decomposition of the PSBMA layer in the 0.5M NaCl solution. PSBMA is reported as an 47 ionic strength responsive polymer and at 0.5M NaCl solution it suffers melting off from the surface [104] 4.2.2 Limiting current density The limiting current density (ilim) of different membrane samples has been measured. First of all a typical current-voltage curve is shown in figure 27 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 0 5 10 15 20 25 30 35 Current Density (mA/cm2) Voltage drop (V) ilim Ohmic region Plateau region Overlimiting region Figure 27: Current-voltage curve for the relation between current through a membrane and corresponding voltage drop over the membrane and its boundary layer. (Measured at 50mM KCl and LiCl) We can distinguish three regions from the current-voltage relationship, a sharp linear increase of current density with voltage drop which is namely ohmic region. After the sharp increase a plateau region is observed where slow increase of current density occurs due to the concentration polarization. In plateau region, concentration in the dilute boundary layer decreases and therefore the resistance increases; which results a deviation in the linear behaviour. Finally again a sharp increase can be observed (over limiting region). In this region electro-convection occurs [1,2,8,101,102] The current-voltage relation gives important information about resistances against ion transport and boundary layer. [101]. Also ilim value indicates the operating current level for electrodialysis (ED) because in ED, it is unexpected to have the concentration polarization 48 (CP).since concentration polarization is directly related to the ED operating efficiency with power consumption. Thus ilim value gives the idea about the maximum operating current level at certain salt concentration. We have considered FKB base membrane, bilayer 6 and 6.5.membranes for ilim measurement. These samples were selected based on the previous characterization and resistance results. Previously measured UV-Vis, contact angle and electrical resistance shows that lower number of bilayers may have lower surface coverage, lower thickness and lower polyelectrolyte density. Therefore higher number of the layers was preferentially selected and considered for ilim measurement. For each individual sample current density (mA/cm2) vs. voltage drop was plotted from which the ilim value was calculated. The ilim of base membrane, PSS terminated (-) bilayer 6 and PDADMAC terminated (+) bilayer 6.5 were measured at salt concentration of 50mM KCl, 50 mM LiCl and mixture of 50 mM KCl and LiCl. These salt concentrations were selected because ion transport experiments were conducted with the same salt and same concentration. The changes of limiting current density with number of bilayers are shown in figure 25 FKB base Bilater 6 (PSS terminated) Bilayer 6.5 (PDADMAC terminated) 2 4 6 8 10 12 14 ilim @ 50 mM KCl+LiCl ilim @50 mM KCl ilim @ 50 mM LiCl ilimi (mA/cm2) Figure 28: Limiting current density behaviour with different bilayer. In general, ilim values decrease with the increase of number of bilayers. Different monovalent salt gives different ilim value (at certain concentration) which signifies valuable information to 55 The flux of K+ decreases more than the flux of Li+ which affects the selectivity of K+/Li+ (Figure 33) The flux of K+ and Li+ (both single ion and mixed ion experiment) in base and bilayer 6 shows more/less same behaviour. However comparing fluxes of base and bilayer 6, the K+ flux decreases with the increase of bilayers but Li+ flux stays almost similar, though a small increase for mixed ion experiment can be observed. At bilayer 6.5, flux values for both single ion and mixed ion decreases. For K+, flux decreases significantly then Li+ (Figure 32). Thus the selectivity of the K+/Li+ also decreases with increasing the layer number (Figure 33). In Table 5 and Table 6 flux and selectivity results are given. Table 5: K+ and Li+ flux in FKB and PEM by LbL. Type of Membranes JK+(single ion experiment) [mol/cm2.s]x10-8 JLi+(single ion experiment) [mol/cm2.s]x10-8 JK+(mixed ion experiment) [mol/cm2.s]x10-8 JLi+(mixed ion experiment) [mol/cm2.s] x 10-8 FKB Base 3.51±0.41 2.56±0.02 3.41±0.65 1.20±0.14 Bilayer 6 3.30±0.07 2.40±0.26 3.34±0.39 1.58±0.05 Bilayer 6.5 1.64±0.09 2.11±0.13 2.13±0.05 1.43±0.12 Table 6: Selectivity of K+/Li+ for single ion and mixed ion experiments. Type of Membranes K+/Li+ (single ion experiment) K+/Li+ (mixed ion experiment) FKB Base 1.37±0.17 2.82±0.21 Bilayer 6 1.39±0.18 2.12±0.31 bilayer 6.5 0.78±0.01 1.49±0.08 The Li+ has lower ionic radius but higher hydration shell then the K+ [117] and the bilayer 6 membrane is more hydrophilic (water contact angle 35.05±3°) then base membrane (Section 4.1.2.1). Bilayer 6.5 membranes is PDADMAC terminated and has a positive surface, exhibits the water contact angle 74.4±3.7° and electrical resistance as 17.68±0.92 ohm.cm2 (Figure 23 and 25). From these characterization values we can say that bilayer 6.5 bears the opposite surface characteristics then bilayer 6 in terms of the surface charge and conductivity. Therefore both K+ and Li+ receive stronger Donnan exclusion in bilayer 6.5 and as a result lower flux values was obtained [115] 56 If Donnan exclusion is solely responsible for lower flux of K+ and Li+ at higher bilayer numbers; then the flux changes is expected to show the same trend at bilayer 6.5 for either ion but we obtained the flux of K+ decreases significantly then Li+. Therefore the Donnan exclusion is not solely responsible but partially influencing the flux behaviour. As a second consideration thickness at swollen state and hydrated cation radius become important. PEM theory describes that the thickness increment occurs with every monolayer deposition which is approximately 3-5 nm when linear increment is observed [110]. Also we have found a linear increment trend for our sample, showed in figure 21(b). The thickness increment contributes significantly in swelling of the membrane in the salt environment [55]. Apart from it the FKD fumasep membrane (similar to FKB) exhibits approximately 30% degree of swelling. [100]. Therefore we can consider that thickness of our membrane is much higher at swollen state than in the dry state. Due to the higher membrane thickness in swollen state and approximate barrier properties from the PEM complexity, flux values of the K+ could be smaller than base membrane. The charge complexation is a related phenomenon with PEM growth where intrinsic and extrinsic charges make the membrane to have a highly complex structure [41] (also explained in section 2.3.2). Taking into account this complex structure we can comment that K+ is partially hindered in the membrane phase so that the flux of K+ is significantly low at bilayer 6.5 but not the Li+. It also affects notably in the K+/Li+ selectivity. Therefore one hypothesis could be the change of diffusion coefficient in the membrane phase due to the complex PEM structure, surface charge and significant thickness increment in swollen state. show the low K+ flux then Li+ at the bilayer 6.5. Jingyi et al [3] has demonstrated competitive diffusion based on ion size (smaller size exhibits higher mobility). Since Li+ has smaller ionic radii then K+, thus the permeation behaviour of Li+ and K+ could be affected differently due to the PEM layer. Regarding to the transport behaviour by diffusion, Stachera et al [116] demonstrated about a three phase model that describes the transport through hydrophobic polymer phase, active phase including ion exchange fixed sites and membrane interstitial phase where hydrated cation moves preferentially following a hopping and dragging mechanism. Bilayer 6.5 has a positive surface and comparably hydrophobic then the base FKB and bilayer 6. Thus K+ flux decreases more at bilayer 6.5 then Li+ due to the different hydrated radius (Li+>K+) and Donnan repulsion. Also the hydrophobic polymer phase transport in the three phase model might become effective to facilitate Li+ more than then the K+. 57 Hence we can compare the result obtained in ilim behaviour (figure 28). At bilayer 6.5, ilim at KCl and K+ ion flux both decreases significantly and almost synergic result can be observed. Taking into account the effect of Donnan exclusion, the hydrated radius of K+ and the surface charge of bilayer 6.5; it seems that due to the Donnan exclusion overall K+ flux reduces at the bilayer 6.5 but because of the bigger cation size of K+ than Li+, K+ is partially hindered and therefore it retains more than Li+ at bilayer 6.5. For monovalent cation selectivity Balster et al [117] has reported about the effect of water uptake and low charge density behaviour which describes that at a low charge density favours to increase monovalent selectivity. In our contact angle results we observed an oscillation of 20-40 ° which partially gives the information of low surface energy character [65,103] therefore could be a possibility for different flux values at bilayer 6 and 6.5. It worth to mention that in the diffusion dialysis experiment salt electronutrality was maintained by the equal amount of HCl and H+ exhibits higher mobility (from receiving to feed side) then K+ and Li+ (from feed to receiving side). Thus the actual mobility of K+ and Li+ is partially influenced by H+ to maintain electronutrality. At lower concentration diffusion boundary layer is less pronounced [100], however need to be considered. Counterion transport number of K+(0.49)>Li+ (0.32) and salt diffusion coefficient for KCl is higher than LiCl in CMX membrane[1] which also could be comparable for FKB and signifies to conduct more related experiments for clearer explanation of ion transport in the PDADMAC-PSS modified FKB membrane . 58 5. Conclusion The layer by layer technique was applied to modify cation exchange membrane (FKB) to improve the monovalent ion selectivity of the membrane. Cationinc polyelectrolyte PDADMAC, anionic polyelectrolyte PSS and zwitterionic polymer PSBMA were used to modify FKB cation exchange membrane. Two approaches such as PDADMAC-PSS modified and PDADMAC-PSBMA-PSS modified FKB membranes were considered for the sample preparation. For two polyelectrolytes modified membrane both the characterization and performance were conducted while three polyelectrolytes modified membranes were considered for characterization. Several techniques were used for this purpose namely measurement of the electrical resistances, UV-Vis spectroscopy, water contact angle and limiting current density. Membrane performance was investigated by diffusion dialysis. UV-Vis measurement gives the information of the layer formation trend with the multilayer propagation on the FKB. We obtained a linear absorbance increment with the increase of bilayers for the both two polyelectrolyte modified and the three polyelectrolyte membrane which suggests a stoichiometric layer growth on substrate. Thus we deposited PEM multilayer successfully on FKB substrate. The membrane electrical resistances and the water contact angle (CA) values showed a zigzag alternating trend (PDADMAC-PSS modified FKB) because of the surface potential change of outermost layer with each monolayer deposition. Oscillating water contact angle was found for our samples which gives the hydrophobicity/hydrophilicity behaviour due to the surface charge reversal process in each monolayer deposition. Also one important observation was the hydrophilicity increases for higher bilayer number which ultimately tunes the PEM membrane surface properties. An alternating surface charge with the effect of hydrophilicity/hydrophobicity was observed. The resistance trend and the contact angle trend suggest a higher degree of layer interpenetration and higher coating density when bilayer number is high. At the same time when bilayer number is high it gives a high resistance and more hydrophilic behaviour then the base membrane. A good barrier property with a hydrophilic nature thus can be expected which can improve the ion selectivity when bilayer number is high. For three polyelectrolyte system several effects of PSBMA was observed comparing with the characterization results with two polyelectrolyte system. The three polyelectrolyte system 59 was considered keeping the PSBMA deposition step in between PDADMAC-PSS deposition. The membrane UV-Vis spectra and the linear trend suggest a stoichiometric increment of the layer thickness but slope of the linear fit was found different than that of two PE systems. This means PSBMA affect the layer thickness. The contact angle oscillation for three PE systems was observed low as 10° which was 20°-40° for two polyelectrolyte system. Thus PSBMA influence the nature of the outermost layer and increases the hydrophilicity. The first PSBMA deposition step (after a PDADMAC monolayer) on FKB made the membrane significantly hydrophilic and it suggest that we can use PSBMA to tune the hydrophilicity of the membrane. Also contact angle trend suggest a stable oscillation behaviour which was different then the two polyelectrolyte system. We can say that the layer interpenetration also decreases when PSBMA is used in between PDADMAC-PSS. The resistance for the three PE systems gives the alternating trend with the trilayer propagation but for some PE trilayer the R was smaller than the base membrane. It was suggested that at 0.M NaCl, PSBMA suffers internal melting thus more experiment with PSBMA need to consider. The limiting current densitiy (ilim) values are indicative for future research to explore in migration transport. We have found that ilim of bilayer 6.5 decreases significantly when measured at 50 mM KCl salt environment. The strong possibility is the Donnan exclusion because of the positive surface. Also bilayer 6.5 may be able to give the barrier properties due to the high layer interpenetration as shown in contact angle and resistance data trend. Ion transport experiment of single ion and mixed ion (K+ and Li+) by diffusion dialysis was conducted and found that at bilayer 6.5, flux of the K+ decreases significantly then Li+ which ultimately lowers the K+/Li+ selectivity. In other word we can say that the Li+ selectivity is improved at bilayer 6.5. Donnan exclusion and strong barrier property due to the highly interpenetrated layer formation seems logical for such behaviour. We can say from the R and CA values of bilayer 6.5 that membrane resistance and hydrophilic nature increases with the bilayer number thus gives a strong barrier to K+ but not to the Li+ Thus a cation and it’s hydrated radius becomes more effective to improve the monovalent selectivity for PEM membrane. The results obtained from diffusion experiments have been used to calculate flux and selectivity of the ions and their effect on polyelectrolyte multilayer. Finally flux of different ions and PEM have optimized with in the experimental boundary. The important finding from 60 diffusion experiment is cationic multilayer (Bilayer 6.5) influence the flux value as such K+ flux decreases more than the Li+, apparently selectivity of either compound can be changed by PEM multilayer on a cation exchange membrane. The important result we have found that at bilayer 6.5 K+ decrease significantly than Li+ which occurs due to the combination of Donnan exclusion, hydrated ionic radius of K+ and Li+ along with the membrane surface charge. From this work several insights have been obtained which are briefly given the following discussion section. 61 6. Recommendation 1. Rinsing volume for LbL by dip coating The layer by layer (LbL) by dip coating is a simple modification technique to obtain numerous surface properties and remarkably potential method to influence characteristics of the object; from nanoscale to the bulk [45]. However this simple technique becomes critical when specific properties are targeted. While preparing the FKB-PEM sample by LbL, polyelectrolyte and rinsing water volume was used as 1:1. Electrical resistance trend of those samples were found as alternating but shows remarkably big error margin (not shown in this report). Thus data reproducibility became a question to consider. A root cause analysis suggests that deposition time, PE and salt concentration was maintained carefully and for achieving stable layer, therefore rinsing water could be the possibility. With this instance, rinsing water volume was increased couple of times for new sample preparation and resistances were found as reproducible standard. It was also reported by Decher et al (2011) [45] that amount of rinsing volume is often ignored in LbL modification and one of the guiding principles of the LbL. The term dilution factor thus addressed which can be defined by dividing the volume of the first rinsing bath by the estimated volume of the adhering liquid. The number and volume of the rinsing baths should be chosen such that the overall dilution factor is at least 1:106. We can recommend that -To obtain reproducible results from LbL modified FKB (PDADMAC/PSS) sample; it has found quite important to choose right rinsing volume and it also leaves behind a scope of determining the dilution factor by a model with some approximation like PE concentration change with each adsorption step, surface charge density, adhering PE volume, membrane surface coverage to effective PE volume in deposition bath etc. 2. Tuning the surface properties of PEM We used 0.2 M NaCl to prepare our PEM sample. It was explained that PEM surface properties can be tuned by varying its vast deposition condition. (Table 4), however salt concentration, pH, type of ion, polyelectrolyte pairs, type of substrate can be considered as most important which influences the layer growth as well as the surface properties. By varying either of the parameter we can have different properties of PEM which suggest a very big research scope in with PDADMAC-PSS polyelectrolyte system with a particular focus on 62 improving Li+ selectivity. We can assume from our results that PDADMAC-PSS multilayer of higher order may be able to give highly hydrophilic behaviour with a high barrier properties, however also can be tuned with varying the above mentioned deposition condition and thus can be recommended for further research on PDADMAC-PSS multilayer by varying deposition condition and substrate specially for Li+ selectivity. 3. Membrane surface and charges Our LbL dip coating fabrication was designed in such a way that one side of the membrane was exposed to the PE solution and other side is hindered by the glass support. It can be assumed that after depositing the cationic layer on to it, a possibility is to obtain a bipolar like membrane which can be named as for instance a quasi-bipolar membrane as shown in the following figure Scheme for the FKB membrane after PDADMAC coating Here we can argue that after each PDADMAC deposition steps on FKB, the membrane become positively charged on the one side but the other side of the membrane which was hindered from the PE solution is still bears the negative charge. If this phenomenon is considered as logical then each PDADMAC terminated layer should make the membrane as quasi bi-polar like membrane and the properties investigated could resemble the property of bipolar membrane. Though the FKB membrane thickness is 80-100 μm and each monolayer gives a 3-5 nm thickness increment however for higher number of bilayers and the swollen state thickness of the membrane cannot be overlooked. - - - - - - - - - - - - - - - - - - - - - + + + + + + + + + + After PDADMAC deposition FKB FKB with PDADMAC 63 May be if we coat both side of the FKB membrane in a single deposition step by PDADMAC, probably it will make the whole surface positive. One comparative study could be done by preparing different samples by coating on single side and double side of the membrane and comparing the evaluated properties with bipolar membrane. 4. Combined PEM of PDADMAC/PSS and PDADMAC/PSBMA Grooth et al (2014) [104] demonstrated the ionic strength responsive PEM which was prepared by the PDADMAC/PSBMA polyelectrolyte. At 0.5 M NaCl such PEM suffers internal melting. One of our PEM systems were considered with PSBMA and electrical resistance (R) was measured at 0.5M NaCl. Thus it can be assumed that while measuring the R we partially lost some layer in the salt medium which might have affected our consecutive characterization as well as the resistances itself. Therefore measurement of R for PSBMA contained PEM should be done at <O.5M NaCl. We can measure R at three or four concentration <0.5M and extrapolating the R vs C (concentration) relation to obtain standard resistance value.at 0.5M NaCl. Also PSBMA was found to adsorb on PDADMAC layer but not onto PSS layer because of its selective interaction behaviour [120] which means preparing PDADMAC/PSBMA/PSS sample leaves behind a doubt to be a coherent approach because there is a possibility of not having PSS layer at all on the PSBMA. Rather PEM by combining PDADMAC/PSS and PDADMAC/PSBMA could be logical and interesting since it exhibits ionic strength responsive behaviour [104]. 5. Electrical Impedance Spectroscopy (EIS) Electrical resistance (R) was measured in direct current method which gives not only the resistance of membrane but also includes electrical double layer resistance and diffusion boundary layer resistance.[101] And the boundary layer thickness in a two compartment system was found as 350μm [121] therefore cannot be avoided. To obtain the actual membrane resistance Electrical Impedance Spectroscopy (EIS) can be considered. EIS is regarded as a powerful technique for evaluating functional and structural characteristics along with membrane resistance [100]. Therefore EIS spectroscopy measurement for PEM could be highly effective to know the PEM modified IEM membrane in detail. 64 6. UV-Vis, reflectrometry and elipsometry UV-Vis measurement gives the co-relation of layer thickness in terms of absorbance in UVVis region, however the intensity absorbed by the material at the exposure point can be regarded to understand and evaluate the thickness however, to determine the thickness we need extinction co-efficient which is not very straight forward for measuring the absorbed amount. Therefore to quantify the membrane thickness is rather difficult. Instead/along with of UV-VIS, reflectometry and elipsometry measurement could be recommended predominantly to quantify the adsorbed amount and % swelling of PEM. 7. Contact angle, ξ potential and streaming potential Contact angle measurement (CA) was considered to obtain the hydrophilicity/hydrophobicity of the membranes and nature of the surface. However CA values of same sample largely depends on the drying protocol. A small change in drying condition gives different CA values for PDADMAC/PSS [65,103]. Along with CA measurement ξ potential and streaming potential measurement could be recommended for comprehensive understanding of the surface. 8. Diffusion dialysis and electrodialysis: Diffusion experiment was conducted for K+ and Li+ in which ionic neutrality was maintained by the HCl.to obtain ion flux. It was found that the PEM bilayer decreases the K+ more than Li+, in other word we have achieved the higher selectivity of Li+ which put two future work possibility. From our experimental result we have obtained that K+ reduced significantly whereas the Li+ stays more/less same at bilayer 6.5. Extrapolating this behaviour we can say that there is a strong possibility that we may be able to achieve a good improvement in Li+ selectivity when the layer number will be high enough. Thus PEM of higher bilayer number can be recommended to prepare and investigate under diffusion dialysis. According to the hydrated ion radius value (K+>Na+>Li+) [1], thus Li+ and Na+ effective separation might be possible at higher layer PEM number. Though ξ potential of higher order membrane does not show alternating charge behaviour and also swelling/deswelling [14] effect makes the resistance alternation diminished. A contradiction might be possible about the properties that can be predicted from the lower bilayer results; even though new interesting behaviour might appear 71 [67] Nazaran, P., Bosio, V., Jaeger, W., Anghel, D.F., and von Klitzing, R. (2007) Lateral mobility of polyelectrolyte chains in multilayers. J. Phys. Chem. B, 111 (29), 8572–8581; 6th International Symposium on Polyelectrolytes, Dresden, Germany (2006). [68] Jomaa, H.W. and Schlenoff, J.B. Salt-induced polyelectrolyte inter diffusion in multilayered films:Macromolecules, 38 (20), 8473–8480 (2005). [69] Kolasinska, M., Krastev, R., Gutberlet, T., and Warszynski, P. Layer-by-layer deposition of polyelectrolytes. Dipping versus spraying. Langmuir, 25 (2), 1224–1232 (2009). [70] Kohler, R., Donch, I., Ott, P., Laschewsky, A., Fery, A., and Krastev, R. Neutron reflectometry study of swelling of polyelectrolyte multilayers in water vapors: Influence of charge density of the polycation. Langmuir, 25 (19), 11576–11585. PMID: 19788217(2009). [71] Tieke, B. and Toutianoush, A. Chapter 7, Electrostatic layer-by-layer fabrication of ultrathin separation membranes, in Bottom-up Nanofabrication: Supramolecules, SelfAssemblies, and Organized Films, vol. 5 (eds K. Ariga and H.S. Nalwa), American Scientific Publishers, Stevenson Ranch, California, USA, pp. 201–217 (2009). [72] Liu, X. and Bruening, M.L. Chem. Mater., 16, 351–357 (2004). [73] Krasemann, L., Toutianoush, A., and Tieke, B. J. Membr. Sci., 181, 221–228 (2001). [74] A. Ouradi Q.T. Nguyen,A. Benaboura, Polysulfone–AN69 blend membranes and its surface modification by polyelectrolyte-layer deposit-Preparation and characterization Journal of Membrane Science 454 20–35(2014) [75] Lutz Krasemann and Bernd Tieke, Selective Ion Transport across Self-Assembled Alternating Multilayers of Cationic and Anionic Polyelectrolytes Langmuir , 16, 287290(2000) [76] Jin, W., Toutianoush, A., and Tieke, B. Appl. Surf. Sci., 246, 444–450 (2005). [77] Szczepan Zapotoczny, Monika Golonka, Maria Nowakowska, Novel Photoactive Polymeric Multilayer Films Formed via Electrostatic Self-Assembly, Macromolecular Rapid Communication,; DOI: 10.1002/marc.200500171(2005) [78] Bohmer, V. Angew. Chem. Int. Ed., 34, 713–745 (1995). [79] Asfari, Z., Bohmer, V., Harrowfield, J., and Vicens, J. (eds) (2001) Calixarenes, Kluwer, Dordrecht, The Netherlands (2001). [80] Gokel, G.W. Crown Ethers and Cryptands: Monographs in Supramolecular Chemistry, Royal Society of Chemistry, Cambridge, UK (1991). [81] Kimura, E., Sakonaka, A., Yatsunami, T., and Kodama, M. J. Am. Chem. Soc. (1981). 72 [82] Katsuhiko Ariga, Jonathan P. Hill and Qingmin Ji, Layer-by-layer assembly as a versatile bottom-up nanofabrication technique for exploratory research and realistic application, Physical Chemistry Chemical Physics DOI: 10.1039/b700410a (2007) [83] S.U. Hong, R. Malaisamy, and M.L. Bruening, Optimization of Flux and Selectivity in Cl - SO 4 2Separations with Multilayer Polyelectrolyte Membranes J. Membr. Sci. 283 , 366-372 (2006). [84] S.U. Hong, R. Malaisamy, and M.L. Bruening, “ Optimization of Flux and Selectivity in Cl- /SO42Separations with Multilayer Polyelectrolyte Membranes” J. Membr. Sci. 283 , 366372 (2006) [85] Andreas Fery, Bjorn Scholer, Thierry Cassagneau, and Frank Caruso, Nanoporous Thin Films Formed by Salt-Induced Structural Changes in Multilayers of Poly(acrylic acid) and Poly(allylamine), Langmuir, 17(13), 3779-3783 (2001) [86] Jingyi Luo, Cuiming Wu, Yonghui Wu, Tongwen Xu, Diffusion dialysis of hydrochloride acid at different temperatures using PPO–SiO2 hybrid anion exchange membranes. Journal of Membrane Science 347 240–249(2010) [87] K.A. Stancheva, Application of dialysis, Oxide Commun. 31 (2008) 758–775. [88] A. Narebska, M. Staniszewski, Separation of carboxylic acids from carboxylates by diffusion dialysis, Sep. Sci. Technol. 43 (2008) 490–501. [89] Jingyi Luo, Cuiming Wu, Tongwen Xu, Yonghui Wu, Diffusion dialysis-concept, principle and applications, Journal of Membrane Science 366 1–16(2011) [90] T.W. Xu, W.H. Yang, Turning the diffusion dialysis performance by surface crosslinking of PPO anion exchange membranes-simultaneous recovery of sulfuric acid and nickel from electrolysis spent liquor of relatively low acid concentration, J. Hazard. Mater. 109 157–164. (2004) [91] Novalin, S. and T. Zweckmair, renewable resourcesgreen bio refinery: Separation of valuable substances from liquid fractions by means of membrane technology. Biofuels Bioproducts & BiorefiningBiofpr, 3(1): p-20-27 (2009) [92] Saracco, G. Transport properties of monovalent-ion-permselective membranes, , Chem. Eng. Sci., 52, 3019−3031 (1997) [93] Balster, J. H. Membrane module and process development for monopolar and bipolar membrane electrodialysis. Ph.D. Thesis, University of Twente, Enschede, (2006). [94] Chapotot, A.; Pourcelly, G.; Gavach, C. J. Membr. Sci., 96, 167−181 (1994) [95] Mourad Amara, Hackne Kerdjoudj, Modification of cation-exchange membrane properties by electro-adsorption of polyethyleneimine, Desalination 155, 79-87(2003) 73 [96] M Kumar, BP Tripathi, VK Shahi, Ionic transport phenomenon across sol–gel derived organic–inorganic composite mono-valent cation selective membranes, Journal of Membrane Science 340 (1), 52-61 [97] Gwenaël Chamoulaud, Daniel Bélange, Modification of ion-exchange membrane used for separation of protons and metallic cations and characterization of the membrane by current–voltage curves, Journal of Colloid and Interface Science 281 179–187 (2005). [98] B. Tieke, F. van Ackern, L. Krasemann and A. Toutianoush, Ultrathin self-assembled polyelectrolyte multilayer membranes, Eur. Phys. J. E 5, 29-39 [99] Said Abdu, Manuel-César Martí-Calatayud, John Erik Wong, Montserrat GarcíaGabald n, and Matthias Wessling, Layer-by-Layer Modification of Cation Exchange Membranes Controls Ion Selectivity and Water Splitting, ACS Appl. Mater. Interfaces, 6, 1843−1854 (2014) [100] Hong-Joo Lee, Heiner Strathmann, Seung-Hyeon Moon, Determination of the limiting current density in electrodialysis desalination as an empirical function of linear velocity, Desalination 190 43–50(2006) [101] Piotr Długołecki, Benoît Anetb, Sybrand J. Metzb, Kitty Nijmeijera, Matthias Wessling, Transport limitations in ion exchange membranes at low salt concentrations, Journal of Membrane Science 346, 163–171 (2010). [102] Piotr Długołecki, Piotr Ogonowskib, Sybrand J. Metzb, Michel Saakesb, Kitty Nijmeijera, Matthias Wessling, On the resistances of membrane, diffusion boundary layer and double layer in ion exchange membrane transport. Journal of Membrane Science 349 , 369–379(2010). [103] Stefan Köstler, Angel V. Delgado, Volker Ribitsch, Surface thermodynamic properties of polyelectrolyte multilayers, Journal of Colloid and Interface Science 286, 339–348(2005). [104] Joris de Grooth, Mo Dong, Wiebe M. de Vos, and Kitty Nijmeijer, Building Polyzwitterion-Based Multilayers for Responsive Membranes, Langmuir , dx.doi.org /10.1021 / la500857b. (2014) [105] G.B. Sukhorukov, E. Donath, H. Lichtenfeld, E. Knippel, M. Knippel, A. Budde, H. M¨ohwald, Layer-by-layer self-assembly of polyelectrolytes on colloidal particles, Colloid Surf. A 137 253–266. (1998) [106] J.E.Wong, F. Rehfeldt, P. Hanni, M. Tanaka, R.V. Klitzing, Swelling behaviour of polyelectrolyte multilayers in saturatedwater vapor, Macromolecules 37 7285–7289. (2004) [107] Monika Schonhoff, Vincent Ball, Andreas R. Bausch , Christophe Dejugnat ,Nicolas Delorme , Karine Glinel f, Regine . Klitzing g, Roland Steitz , Hydration and internal 74 properties of polyelectrolyte multilayers, Colloids and Surfaces A: Physicochem. Eng. Aspects 303 14–29(2007) [108] Bernd Schwarz and Monika Schonhoff, Surface Potential Driven Swelling of Polyelectrolyte Multilayers, Langmuir, 18, 2964-2966 (2002) [109] Liang Chen, Youhai Yu, Huaping Mao, Xiaofeng Lu, Lei Yao, Wanjin Zhanga,, Yen Wei, Synthesis of a new electroactive poly(aryl ether ketone), Polymer 46 2825–2829 (2005) [110] Peter Nestler, Malte Paßvogel, and Christiane A. Helm, Influence of Polymer Molecular Weight on the Parabolic and Linear Growth Regime of PDADMAC/PSS Multilayers, Macromolecules, 46, 5622−5629 (2013) [111] Meng Che Hsieh, Richard J. Farris, and Thomas J. McCarthy, Surface “Priming” for Layer-by-Layer Deposition:Polyelectrolyte Multilayer Formation on Allylamine PlasmaModified Poly(tetrafluoroethylene), Macromolecules, 30, 8453-8458(1997) [112] Ron S. Faibish, Wayne Yoshida, and Yoram Cohen, Contact Angle Study on PolymerGrafted Silicon Wafers, Journal of Colloid and Interface Science 256, 341–350 (2002) [113] J.B. Schlenoff, in: G. Decher, J.B. Schlenoff (Eds.), Multilayer Thin Films, Wiley– VCH, Weinheim, p. 106. (2003) [114] J.J. Krol, M. Wessling, H. Strathmann, Concentration polarization with monopolar ion exchange membranes:current-voltage curves and water dissociation, Journal of Membrane Science 162,145-154(1999) [115] G. Saracco, Transport properties of monovalent-ion-permselective membranes, Chem. Eng. Sci. 52 (17) 3019–3031. (1997) [116] D.M. Stachera, R.F. Childs, A.M.Mika, J.M. Dickson, Acid recovery using diffusion dialysis with poly(4-vinylpyridine)-filled microporous membranes, J. Membr. Sci. 148 119– 127 (1998) [117] Johan Mähler and Ingmar Persson, A Study of the Hydration of the Alkali Metal Ions in Aqueous Solution, Inorg Chem;51(1): 425–438. (2012) [118] Yung Chang, Wan-Ju Chang, Yu-Ju Shih, Ta-Chin Wei, and Ging-Ho Hsiue, Zwitterionic Sulfobetaine-Grafted Poly(vinylidene fluoride) Membrane with Highly Effective Blood Compatibility via Atmospheric Plasma-Induced Surface Copolymerization, ACS Appl. Mater. Interfaces, 3, 1228–1237 (2011) [119] Kharlampieva, E.; Izumrudov, V. A.; Sukhishvili, S. A. Electrostatic layer-by-layer self-assembly of poly(carboxybetaine)s: Role of zwitterions in film growth. Macromolecules, 40 (10), 3663−3668 (2007) 75 [120] Mary, P.; Bendejacq, D. D.; Labeau, M.-P.; Dupuis, P.Reconciling lowand high-salt solution behavior of sulfobetaine polyzwitterions. J. Phys. Chem. B, 111 (27), 7767−7777. (2007) [121] J.-H. Choi, J.-S. Park, S.-H. Moon, Direct measurement of concentration distribution within the boundary layer of an ion-exchange membrane, J. Colloid Interface Sci. 251 311– 317. (2002). 76 Appendix A. Flame photometer calibration and basic scheme of the BWB flame photometer. A.1. Flame photometer calibration For the calibration of the Flame photometer, five concentrations points were considered depending on the approximate concentration of the ion at the start of the diffusion experiment. For example let’s say Initial K+ ion concentration (at the start of the experiment) = 50mM Sample considered = 0.3 mL/ 15 min. Dilution factor = x10 Thus after dilution the concentration of initial sample becomes= 5 mM. Based on this above mentioned information 5 concentration points were considered for the calibration keeping the concentration ranges above and below then the initial sample concentration Thus the calibration for the BWB flame photometer was conducted with 0.5, 1, 2.5, 5, 10 mM KCl solution. The same method was used for all the diffusion measurement. A.2. Basic component of BWB Flame photometer: Basic component of BWB flame photometer (left) and the process involved in flame photometry (right). [http://www.bwbtech.com/] 77 B. Contact angle changes with different bilayers on the FKB membrane. Number of bilayer Water contact angle(ϴ)° 0 73.1±2.01 0.5 95.88±1.3 1.0 52.13±5.7 1.5 86.7±0.8 2.0 58.1±2.5 2.5 78.9±2.1 3.0 42.35±4.2 3.5 77.48±1.7 4.0 42.48±1.1 4.5 70.68±3.3 5.0 42.41±1.2 5.5 70.93±3.3 6.0 35.05±3.0 6.5 74.38±3.7 7.0 34.46±2.2 Number of bilayers and the corresponding contact angle value 0 0.5 1 1.5 2.0 2.5 3 3.5 4 4.5 5.0 5.5 6.0 6.5 7 78 C. FKB membrane properties and polyelectrolyte membrane properties. C.1 Physical and chemical data of FKB cation exchange membrane Physical and chemical data of FKB cation exchange membrane. [http://www.fumatech.com/NR/rdonlyres/58CD330B-882C-426F-8399C856BC481FC3/0/fumasepFKB.pdf] C.2 Membrane electrical resistance (Direct current method) Number of layers Average Resistance* (Ω.cm2) 0 3.99±0.35 0.5 13.01±0.35 1 5.46±0.75 1.5 14.68±0.23 2 5.23±0.67 2.5 16.40±1.07 3 4.53±0.12 3.5 17.31±0.12 4 7.16±0.27 4.5 18.10±1.17 5 6.26±1.83 5.5 18.66±0.23 6 9.64±0.38 6.5 17.69±0.78 7 11.20±1.23 *each data is the average of three sample measurement. 79 D.1 ilim values for different PDADMAC/PSS bilayers in different salt solution Sample* ilim 50mM 1:1 KCL+LiCl ilim, 50mM(KCl) (ilim, 50mM(LiCl) FKB base 12.33±0.237 7.907±0.116 4.57±0.12 Bilayer 6 11.32±0.236 7.485±0.233 4.239±0.114 Bilayer 6.5 9.651±0.225 3.81±0.707 4.065±0.12 *each sample data point refers to the 3-5 fresh sample measurement.