influence of monovalent electrolytes on the electrochemical studies of newly synthesized thermally...
Embed Size (px)
Desalination 329 (2013) 103–114
Contents lists available at ScienceDirect
j ourna l homepage: www.e lsev ie r .com/ locate /desa l
Influence of monovalent electrolytes on the electrochemical studiesof newly synthesized thermally stable inorganic–organicnanocomposite membrane
Md. Ramir Khan 1, Rafiuddin ⁎,1
Membrane Research Laboratory, Physical Chemistry Division, Department of Chemistry, Aligarh Muslim University, Aligarh 202002, India
H I G H L I G H T S G R A P H I C A L A B S T R A C T
• Sol–gel method for the synthesis ofcadmium tungstate nanocomposite
• Counter-ion transport number andperm-selectivity from potential
• Utilization of different methods toestimate fixed charge density
• Selectivity order: Li+ N NH4+ N K+ N Na+
• Physicochemical and spectroscopic char-acterization of the prepared membranes
⁎ Corresponding author.E-mail address: [email protected] (Rafiuddin).
1 Tel.: +91 571 2703515.
0011-9164/$ – see front matter © 2013 Elsevier B.V. All rihttp://dx.doi.org/10.1016/j.desal.2013.09.008
a b s t r a c ta r t i c l e i n f o
Article history:Received 27 December 2012Received in revised form 12 July 2013Accepted 12 September 2013Available online 2 October 2013
Keywords:Cadmium tungstate nanocompositeCounter-ion transport numberPermselectivityPhysicochemical propertiesTasaka, Aizawa and Kobatake methodsFixed charge density
Ananostructured inorganic–organic compositemembrane has been synthesized by cost effective sol–gelmethodunder acidic conditions. Membrane potential measurements have been carried out in various monovalentelectrolytes (KCl (aq), NaCl (aq), etc.) at different concentrations to scrutinize the relationship between fixedcharge concentration and electrochemical properties of the membrane. Effective fixed charge densities arefound to follow the order: Li+ N NH4
+ N K+ N Na+. The calculated values of permselectivity, ionic mobilityratio and counter-ion transport number reveal that the inorganic–organic nanocomposite membrane is morecation-selective towards Li+ ions; however the selectivity decreases with the increase in concentration for allmonovalent electrolytes used. Thesemembraneswere comprehensively characterized for their physicochemicalproperties, morphology, molecular interactions, crystallinity and thermal stability by various analyses. The ion-exchange capacity, volume void porosity and water uptake of the membrane are found to be highly dependenton polystyrene content in the membrane phase and these properties decrease with the increase in the amountof polystyrene. Furthermore, membrane with 25% blend ratio with polystyrene is found to exhibit good selectiv-ity along with moderate ion-exchange capacity, which may be used for their application in electro-drivenseparation at high temperatures or for other electrochemical processes.
© 2013 Elsevier B.V. All rights reserved.
104 M.R. Khan, Rafiuddin / Desalination 329 (2013) 103–114
Ongoing research has gradually acknowledged and understood theimportance of synthetic membranes and their employment on a largeindustrial scale due to a numerous number of practical applications[1,2]. The magnificent properties of inorganic membranes provide aset of tools for solving many of the problems that the society is facing,from environmental to energy problems and from water quality tomore competitive industries [3–7]. Such a wide variety of issuesrequires a fundamental strategy, together with the specific descriptionof applications provided by those researchers that have been closeto the industrial applications. In many applications such as water desa-lination and purification, the membrane processes compete directlywith the more conventional water treatment techniques. However,compared to these conventional procedures, membrane processes areoften more energy efficient, simpler to operate and yield a higherquality product.
Inorganic–organic nanocomposites, a rapidly emerging andpromising research field, demonstrate the possibility to controlthe hydrophobic or hydrophilic micro-domain, surface chargesand porosity for designing cation selective ion-exchange mem-branes of new generation. Such hybrid materials show the attrac-tive properties of a mechanically and thermally stable inorganicbackbone and the specific chemical reactivity as well as flexibilityof the organic component [8–11]. The membrane systems preparedby such hybrids provide numerous remarkable impacts such as sig-nificantly lower eco-environmental impact factors, higher finalproduct quality, good selectivity for heavy metals, etc. as comparedto the pure inorganic as well as organic materials in a variety ofindustries.
In recent times, it was proposed by many researchers to use in-organic–organic nanocomposite membranes as a substitute toNafion membrane for fuel cell applications because of their highwater retention capacity at elevated temperature as well as theircomparable physicochemical and electrochemical properties tothe Nafion which possesses superior stability and conductivity[12–14]. The present article represents the results of our overallexploration on membrane performance of polystyrene blendedcadmium tungstate nanocomposite, synthesized through a sol–gel route, as reflected by its physicochemical properties, transportnumber, permselectivity and electrochemical properties. Both thepure cadmium tungstate membrane and polystyrene incorporatedcadmium tungstate membrane have been studied to figure out thechanges in their properties such as water content, porosity, andcrystalline nature as a function of the amount of the binder. For afurther investigation of the cadmium tungstate nanocompositemembrane blended with polystyrene, its effective fixed chargedensity, which is considered as the most valuable parameter con-trolling the membrane phenomena, has been evaluated from theexperimental membrane potential values by means of differentmethods put forwarded by many researcher groups [15–19].
2. Experimental section
2.1. Reagents and solutions
Deionized water (water purification systems, integrate; withreverse osmosis (RO) conductivity 0–200 μs/cm) was used to pre-pare all solutions of the reagents, which are of analytical grade.Pure sodium chloride, potassium chloride and ammonium chloridewere obtained from E. Merck (India) Limited while lithium chlo-ride is from Loba-Chemie Indoaustranal Co. (India). Cadmium (II)chloride (from Qualigens Fine Chemicals, Mumbai, 90.978% purity)and sodium tungstate (99.90% purity, E. Merck) had been used. Theworking solutions of the electrolytes (KCl, NaCl, LiCl and NH4Cl) of
the required concentrations were prepared by appropriately dilut-ing their stock solutions.
2.2. Preparation of cadmium tungstate membranes
Inorganic–organic nanocomposites have been prepared by sol–gelmethod. Cadmium (II) chloride solution (0.25 M) was reacted with0.25 M solution of sodium tungstate to prepare the white cadmiumtungstate precipitates by the method that had been described in ourearlier reported work [20,21]. The resulting mixture is stirred wellwith a magnetic bar keeping the temperature constant at 80 °C for24 h until one-phase solution is formed that goes through a solution-to-gel transition. The mixture was adjusted to pH 1.0 by adding dilutedHCl solution. The precipitates of cadmium tungstate were keptovernight in the mother liquor for digestion. After decanting off thesupernatant liquid, the remaining precipitate was washed withdemineralized water to remove any excess reagent of electrolytes andthen dried at 100 ± 1 °C for another 24 h. These were powdered andsieved through 200 meshes (Granule size b 0.07 mm). Pure crystallinepolystyrene was also ground to fine powder and sieved through 200meshes. The inorganic precipitates were then mixed with granulatedpolystyrene with the help of a pestle and mortar to get ion-exchangemembranes having a varying percentage (by mass) of polystyrene(15–35%). Such membranes were used to understand the changes intheir physicochemical behavior as a function of the quantity ofpolystyrene that is blended with the inorganic component whilemembranes having 25% polystyrene were only selected for transportstudies.
2.3. Instruments and membrane characterization
A digital potentiometer model 118 (Electronics India) was used formeasuring the membrane potential. A Shimadzu TG/DTA simultaneousmeasuring instrument, DTG-60H (Kyoto, Japan) was used forthermogravimetric analysis (TGA) and differential thermal analysis(DTA). Infrared (IR) spectra were recorded on a Fourier transform infra-red (FTIR) spectrometer from Perkin Elmer (1730, USA), using KBr diskmethod, operating under a nitrogen atmosphere with a heating rate of10 °C/min from 25 to 800 °C. A Rigaku Miniflex X-ray diffractometerwas utilized for crystallographic investigation by employing a mono-chromatic X-ray beamwith calcium-filtered CuKα radiation and settingthe diffraction angle at between 20° and 80°. A scanning electronmicro-scope (LEO, 435 VP) instrument, with gold sputter coatings, operating at10−2–10−3 Pa with EHT 15.00 kV with 300 V collector bias was usedfor the electron micrographs for figuring out the morphological struc-ture of the inorganic–organic nanocompositemembrane. Optical absor-bance spectra of the samples were obtained using Perkin-Elmer UV–visible spectrophotometer.
2.4. Water uptake, volume void porosity and swelling
Different samples of the inorganic–organic nanocomposite mem-branes prepared with different quantity of polystyrene were immergedin distilled water as well as in 1 M NaCl solution for 24 h to figure outwater content in terms of water concentration, void porosity and swell-ing [22,23]. Their surfaces were wiped with filter paper and then thewet membranes were weighed. The thickness of the samples wasmeasured using a micrometer screw gauge and membrane density forwet membrane was determined by dividing the wet membrane weightby its volume. Subsequently, they were dried at 100 °C in an oven toconstant weight. The weight, thickness and density of dry membraneswere also estimated in the same way. Water content and porosity ofthe membrane were determined in terms of amount of water absorbedby the membrane.
105M.R. Khan, Rafiuddin / Desalination 329 (2013) 103–114
Volume fraction of water (ϕw) of the membrane is evaluated by therelation:
ϕw ¼ Δw=ρw
Δw=ρw þwd=ρm: ð1Þ
Here, Δw(=ww − wd) is the weight difference between wet anddry membranes, ρw and ρm are the densities of water and mem-brane, respectively. Water content of the pure cadmium tungstateas well as polystyrene–cadmium tungstate samples with differentquantities of polystyrene are determined in terms of water concen-tration (in mol/dm3) in the membrane phase  by the followingequation:
Wcm ¼ ww−wd
� �� ρw ð2Þ
whereWcm is the concentration of water in the membrane andMW is
the molar mass of water (18 g/mol).The volume of free water within the membrane per unit volume of
wet membrane of the composite, defined by Volume void porosity(τm), is evaluated by the relation :
τm ¼ ΔV1þ ΔV
Here, ΔV symbolizes the volume increase of the membrane uponabsorption of the water per unit of dry membrane volume which isevaluated by the relation: ΔV ¼ ww−wdð Þρd
ρWwd, where ρd is the density of
the dry membrane and ρW the density of water which enters into themembrane.
2.5. Ion exchange capacity
The ion-exchange capacity of the various samples of cadmiumtungstate with different amounts of polystyrene was determined bycolumn (0.5 cm, internal diameter) operation. The ion-exchanger inthe H+ form was placed in the column with glass wool support; and a0.1 mol L−1 solution of sodium nitrate solution was used as the eluent.The flow rate was maintained at 1.0 mL min−1. The H+ ion contentof the effluent was then determined by titrating against a standardsolution of 0.1 mol L−1 sodium hydroxide.
Fig. 1. Schematic diagram representing m
2.6. Chemical stability
ASTM D543-95 method  was used to test the chemical stabilityof the inorganic–organic nanocomposite membranes. The membraneswere exposed to a number of media commonly utilized such as H2SO4,NaOH, K2Cr2O7, and HNO3. The membranes were scrutinized after 24,48 and 168 h, analyzing variation in color, texture, splits, holes, bubbles,brightness, decomposition, curving and stickiness.
The synthesized inorganic–organic nanocomposite membrane wasalso tested for chemical resistance in acidic, alkaline and strong oxidantmedia.
2.7. Measurement of potential
The freshly prepared inorganic–organic nanocomposite membranewas cemented in a Pyrex glass tube cell having two compartments inwhich a saturated calomel electrode was placed for measuring themembrane potential; the schematic diagram of the constructed electro-chemical cell of this type is shown in Fig. 1. The monovalent chlorideelectrolytes of concentrations c1 and c2, in both the compartments ofthe cell were vigorously stirred by a magnetic stirrer to minimize theeffects of boundary layers on the membrane potential (mV ± 0.5).The experiment was conducted at room temperature and atmosphericpressure.
3. Results and discussion
3.1. The physicochemical properties
The cadmium tungstate nanocomposite membrane has been syn-thesized by the cost-effective sol–gel method in acidic medium; thetemperature of water coagulation bath was thermally controlled at80 °C to reduce synthesis condition variation. Our attempt was to setup a membrane system of adequate chemical, thermal and mechanicalstability. So, in order to acquire suchmembranes, the selection of binderis also of immense importance. Easily available polystyrene with a lowcost is found to be an appropriate binder, as its cross-linked rigid frame-work provides adequate adhesion to the cadmium tungstate whichaccounts for the mechanical stability to the membrane over otherbinders like polyvinyl chloride (PVC) and cellulose acetate. So, polysty-rene blended cadmium tungstate results in optimal thermal and
easurement of membrane potential.
Fig. 2. Plots showing dependence of amount polystyrene on physicochemical properties:(A) water concentration, (B) volume void porosity and (C) ion-exchange capacity of cad-mium tungstate nanocomposite membrane.
106 M.R. Khan, Rafiuddin / Desalination 329 (2013) 103–114
mechanical stabilities provided by the inorganic part and furnish goodflexibility due to the organic component i.e. polystyrene supportedcadmium tungstate nanocomposite membrane can be a potential andbetter candidate than many conventional membranes which degradeunder harsh conditions frequently encountered in various industrialsettings. Furthermore, it appears to be an efficient and cost-effectivematerial having solvent resistance and thermal resistance characteris-tics .
When the composite membrane was tested for chemical resistancein acidic, alkaline and strong oxidant media, it was found that in acidic(1 MH2SO4) and alkaline (1 MNaOH)media, few considerable changeswere observed after 24, 48, and 168 h, indicating that the membrane iseffective in such media. In strong oxidant media like K2Cr2O7, however,the synthesizedmembrane became fragile after 48 h, losing itsmechan-ical resistance.
Different amounts of polystyrene (15–35%) have been blendedwiththe inorganic compound resulting in various blend ratios. The mem-branes with 25% organic binder are found to be quite stable and furnishreproducible results; higher or lower than this quantity of polystyreneresults in the decrease in stability. Membranes prepared in this waydid not show any dispersion in water or in other electrolyte solutions.They were subjected to microscopic and electrochemical studies forfiguring out cracks and homogeneity of the surface; only thosemembranes that had smooth surface and generated reproduciblepotentials were assured by carefully controlling the conditions offabrication for further studies.
Bothwater content in terms ofwater concentration and volumevoidporosity of the polystyrene blended cadmium tungstate membrane,calculated by means of Eqs. (2) and (3), were found to decrease withthe increase in the quantity of polystyrene owing to the decrease ininterstitial volume. This shows that these parameters depend uponthe amount of polystyrene that has been used for blending. The cross-linked clusters of polystyrene may be the reason for this decrease. Thevalues are plotted as a function of polystyrene in Fig. 2A and B. All theinvestigated membranes of cadmium showed a negligibly small swell-ing when immersed in NaCl solution for 24 h. The narrow pore sizedistribution of the membranes increases the diffusive resistance andwill enable a precise control over molecular transport. The low ordersof water concentration and volume void porosity with negligible swell-ing of the membranes also suggest that interstices are negligible anddiffusion across the membrane would occur mainly through exchangesites. Ion-exchange capacity (IEC) indicates the density of ionizablehydrophilic groups in the membrane matrix, which are responsiblefor the ionic conductivity in the ion-exchange membrane. The IEC(mequiv./g dry memb) values can be evaluated from the quantitativeanalysis of Na+ ions by the following equation:
IEC ¼ CNaþ Vsol.
where, CNaþ is the concentration of Na+ ions (mmol/cm3 =mequiv./cm3) of the NaNO3 solution, Vsol is the volume of the solu-tion andWdrymemb is the weight of dry membrane (g). The results ofthe ion-exchange capacity of different samples of cadmium tung-state are represented in Fig. 2C, which reveals that the ion-exchange capacity decreases with the increase in the weightfraction of non-charged polystyrene that is incorporated with cad-mium tungstate. This decrease may be attributed again to the crosslinked clusters of polystyrene. The values of ion-exchange capacityshow that cadmium tungstate behaves as a weak cation-exchanger.The values of water concentration (Wc
m), volume void porosity(τm), and IEC for cadmium tungstate nanocomposite membraneare somewhat higher than the values of our previous reportedwork of calcium tungstate membrane .
Fig. 3. Surface image (A), cross sectional image (B), percentage content (C) and EDAX spectrum (D) of polystyrene blended cadmium tungstate nanocomposite membrane.
107M.R. Khan, Rafiuddin / Desalination 329 (2013) 103–114
3.2. Membrane characterization
3.2.1. SEM analysisSEM studies had been carried out to figure out the surface texture,
composite pore structure, thickness, homogeneity and cracks of the syn-thesized nanocomposite membrane. The electron micrograph image(Fig. 3) reveals a clear demonstration of the surface morphology andthe homogeneity in mixing the organic and inorganic substrates to getthe hybrid i.e. no phase separation and no visible cracks of the mem-brane surface can be observed, suggesting that the synthesized compos-ite films are homogeneous in nature and hence form a densemembrane. The surface thickness is approximated to 2 μm as shownin the image. The membrane is macroscopically uniform in thicknessand the porous nature of the membrane is clearly exposed in theimage. The pores that can be modeled as uniform capillaries are evenlydistributed throughout the surface of the membrane. Entrance and exiteffects can be ignored since the pore radius is small as compared to thethickness of the membrane and it is assumed that the membrane andthe adjacent solution are in equilibrium. The cross sectional image ofthe hybrid material is also given in Fig. 3B. The elemental content ofthe hybrid membrane is clearly seen in the EDAX spectrum (Fig. 3A)and the weight percentage of different elements that is present in thehybrid material (Fig. 3D).
3.2.2. Thermal stabilityThe thermal stability of the pure cadmium tungstate as well as the
polystyrene blended cadmium tungstate nanocomposites was illustrat-ed by their TG-DTA studies (Fig. 4). The TGA curve measured underflowing nitrogen is presented in Fig. 4A for representative pure inorgan-ic membrane, which shows no any appreciable weight loss except asmall continuous weight decay (about 12–15%) up to 400 °C which is
attributed to the removal of adsorbed water molecules on the surfaceas well as those present in the interstitial sites of the membrane matri-ces; the corresponding endothermic peak is observed in theDTA curves.The TGA curve of the inorganic–organic nanocomposite (Fig. 4B), onthe other hand, shows a well defined weight loss of more than 34% ina temperature range between 380 °C to 450 °C, which attributes tothe decomposition of the polymer in the hybrid membrane, wherebythe midpoint of decay is observed at 418 °C. The DTA curve reveals anexothermic peak (heat energy releases) at 495.66 °C which may beattributed to the transition of the substance to oxide . From theTG-DTA analyses, it suggests that cadmium tungstate blended withpolystyrene results in better thermal stability furnished by the inorganicpart as well as good mechanical stability and flexibility provided by theorganic component.
3.2.3. FTIR studiesFig. 5 shows the FTIR spectra for pure cadmium tungstate and its
composite membranes. Negative shift of the composite relative to freepolystyrene or metal salts has been ascertained from of the bands inthe spectra. Bonding characteristics of the composites have beenascertained from the FTIR spectral analyses. For the pure cadmiumtungstate (Fig. 5A), the absorption band at around 1056 cm−1 corre-sponds to the ν(O\W\O) stretching mode and the broad absorptionband between 1568 and 1628 cm−1 is assignable to the stretchingmode of terminal W_O . The wide and broad absorption peaknear 3471 cm−1 may be attributed to the ν(O\H) stretching mode ofthe non-bonding water molecules which are adsorbed on the surface.For the polystyrene blended cadmium tungstate composite membrane,the FTIR spectrum is depicted in Fig. 5B which reveals the shifting of theν(O\W\O) stretchingmode to 1066 cm−1 and a slight increase in thestretching mode of terminal W_O to 1637 cm−1. The resonance and
Fig. 4. TGA/DTA curves for (A) pure cadmium tungstate and (B) polystyrene incorporated cadmium tungstate nanocomposite.
108 M.R. Khan, Rafiuddin / Desalination 329 (2013) 103–114
electronic effects possible in the tungstate ion and the interaction withthe polymermatrixmay be the reason for this increase. The polystyrenebackbone of the composite exhibits characteristic frequencies ofaromatic ring at 693, 749, 1022, 1359, 1454, 1494 and 3100 cm−1 inthe inorganic–organic hybrid; the stretching bands between 2366 and2921 cm−1 are attributed to the aliphatic C\H groups and the peak at3028 cm−1 corresponds to the aromatic C\H groupswhich are presentin the organic binder.
3.2.4. X-ray studiesThe X-ray diffractograms, which can be better explored to provide
information on crystal morphology, for the pure cadmium tungstateand its nanocomposite membranes are shown in Fig. 6; the presenceof sharp peaks reveals their crystalline nature. The XRD patterns showa broad peak at 27.6° for pure cadmium tungstate crystal (Fig. 6A)while for the inorganic–organic nanocomposite crystal, the broad peakis observed at 30.9° (Fig. 6B). It is possible to correlate peak width tothe size of crystallographic perpendicular planes using the Debye–
Scherrer formula: D = 0.9λ/β cos θ, which has been used to calculatethe crystallite size (D) from the corresponding X-ray spectral peak.Here, λ is the X-ray wavelength (1.54060 × 10−10 m) of the incidentlight, β is the full width at half-maximum (FWHM) which is the peakwidth at half of the total peak height of the compound and θ is thepeak diffraction angle. The average crystal granular size of the purecadmium tungstate nano-particles, calculated from the most intensepeak using the above equation, was found to be 9.14 nm while thecrystallite size of the nanocomposite was 26.46 nm. It is also graspablefrom the diffractograms that the crystalline nature of polystyreneblended cadmium tungstate is less than that of the pure cadmiumtungstate; the result is obvious as polystyrene is amorphous in nature which decreases the crystalline nature of cadmium tungstate.
3.2.5. Spectral analysesAbsorption spectroscopy is an influential non-destructive technique
to investigate the optical properties of the nano-particles. The opticalabsorption spectra of pure and polystyrene blended cadmium tungstate
Fig. 5. Fourier-transform infrared spectra of (A) pure cadmium tungstate and (B) polystyrene supported cadmium tungstate nanocomposite.
109M.R. Khan, Rafiuddin / Desalination 329 (2013) 103–114
nano-particles are shown in Fig. 7. The absorbance is expected todepend on several factors, such as band gap, surface roughness andimpurity centers. Absorbance spectra show an ultraviolet cut-off ataround 450–600 nm, which can be attributed to the photo-excitationof electrons from valence band to conduction band. In order to calculatethe direct band gap we used the Tauc relation: αhυ = A(hυ − Eg)n
where α is the absorption coefficient, A is a constant, and n = 1/2 fordirect band gap. An extrapolation of the linear region of a plot of(αhυ)2 vs hυ gives the value of the optical band gap Eg. The measuredband gap was found to be 3.1 eV for pure cadmium tungstate nano-particles, which is less than the value of polystyrene incorporated
Fig. 6. X-ray diffractograms of (A) pure cadmium tungstate and
cadmium tungstate (3.4 eV) and this can be attributed to the quantumconfinement effect of the nano-particles.
3.3. Membrane potential and electrochemical studies
Themembranepotential valuesψm, measured across thepolystyreneblended cadmium tungstate nanocomposite membrane for differentmonovalent electrolytes such as LiCl and KCl of unequal concentrationsranging between 1 and 0.007 mol dm−3 at 146 MPa, are plotted as afunction of − log C, where C = (c1 + c2) / 2 is the average concentra-tionwith the ratioγ(=c2/c1)fixed at 10 (Fig. 8). The potential difference
(B) polystyrene blended cadmium tungstate membranes.
Fig. 7. Absorbance spectra of (A) pure cadmium tungstate and (B) cadmium tungstatenanocomposite containing 25% polystyrene.
Table 1Calculated values of transport number (t+) and mobility ratio (ω ) of the cadmiumtungstate nanocomposite membrane from the measured membrane potential values
110 M.R. Khan, Rafiuddin / Desalination 329 (2013) 103–114
develops across the membrane due to the tendency of oppositelycharged ions to move with different mobilities. The quantity of chargenecessary to generate the potential is small, particularly when dilutesolutions are used. The charge imparts some significant electrochemicalproperties to themembrane such as the differences in the permeabilitiesof co-ions, counter ions and neutral molecules. The magnitude of themembrane potential depends on many factors like applied pressure atthe membrane preparation stage, counter-ion to co-ion mobility ratio,exchange characteristics of the membrane material for various cationsin addition to the nature and concentration of the equilibrating electro-lyte solutions [30,31]. It was observed that the cadmium tungstate nano-composite membrane prepared at higher applied pressure exhibitedhigher membrane potential. The values of membrane potential atpressure 166 MPa are also represented in Fig. 8. The results reveal thatthe potential values follow the order Li+ N NH4
+ N K+ N Na+. The highermembrane potential observed in the case of LiCl may be due to the factthat Li+ ions are not thermodynamically favored in the tungstatemembrane phase as the Stoke radii of their hydrated cations(2.4 × 10−10 m for Li+) are higher than those of other cations such
Fig. 8.Membrane potentials,ψm against−log C across cadmium tungstate nanocompositemembrane equilibrated with various monovalent electrolytes at different applied pres-sures 146 MPa (–) and 166 MPa (- -).
as K+ (1.3 × 10−10 m) andNa+ (1.8 × 10−10 m) and partially hydratedcations transfer in the case of cadmium tungstate membrane andpermselectivity persists to higher immersion concentrations becausethe lithium salts do not cause Nernst breakdown . The membranepotential was observed to increase with time at first; attained a maxi-mum value after a certain interval and then fell off slowly. For a concen-trated solution, the time taken for the attainment of maximum potentialwas observed to bemore than that of a dilute solution but it was found todiffer with different electrolytes.
3.4. Electrochemical studies and fixed charge density
Themeasurement of ion activity by means of a membrane electrodeismost successful in the concentration range over which themembranebehaves as ideally permselective and obeys the Nernst equation. Anideally permselective membrane is the one which allows a negligiblepermeability for co-ions as compared to that for counter-ions. The co-ion transference and the dependence of the exchange of cationsbetween the solution and the membrane phase and on the electrolyteconcentration  is possibly the reason for the deviation. The data inFig. 8 illustrate that the potential values are found to be positive andincreasewith the decrease in concentration of all the tested electrolytes,which shows that themembrane is negatively charged i.e. cation selectiveas cations easily pass through the negatively charged cation exchangemembrane. The overall outcome of the process is that one cell of thepair becomes depleted of ions while the adjacent cell becomes enrichedin ions, however, the selectivity increases with dilution which may bedue to the structural alteration produced in the electrical double layer atthemembrane-solution interface. The increase in selectivitywith dilutionis also supported by the increasing values of the counter-ion transportnumbers, t+ (Table 1). But in the case of some 2:1 and 3:1 electrolytes,the selective character of ion-exchange membrane is inverted i.e. anionselective . This change in the selectivity of the membrane mighthave been caused by the adsorption of multivalent ions leading to astate where net positive charge is left on the membrane surface makingit anion selective.
The influence of penetration of mobile species into the membranewhen an ionic gradient is maintained by two solutions of an electrolyte
with various monovalent electrolytes at different concentrations with c2/c1 = 10 at anapplied pressure 146 MPa at 25 ± 1 °C.
Electrolyte c2 (mol/dm3) t+ ω
NH4Cl 1 0.639 1.770.7 0.648 1.840.5 0.669 2.030.25 0.713 2.480.1 0.832 4.960.07 0.882 7.50
KCl 1 0.626 1.670.7 0.635 1.740.5 0.652 1.870.25 0.695 2.280.1 0.804 4.100.07 0.858 6.06
NaCl 1 0.615 1.600.7 0.624 1.660.5 0.650 1.860.25 0.682 2.150.1 0.782 3.590.07 0.834 5.04
LiCl 1 0.641 1.790.7 0.654 1.890.5 0.678 2.110.25 0.735 2.770.1 0.843 5.380.07 0.897 8.76
Table 2Comparison of the values of the effective fixed charge densities, calculated from differentmethods, of the cadmium tungstate nanocomposite membrane in contact with differentelectrolytes with their counter-ion transport number, t+m in the membrane phase.
Electrolyte NH4Cl KCl NaCl LiCl
Xfa 0.0208 0.0192 0.0177 0.0213
φXb 0.0158 0.0148 0.0137 0.0160θXc
c 0.1174 0.0690 0.0508 0.1246θXd 0.1277 0.0813 0.0631 0.1497t+me 0.6246 0.6151 0.6083 0.6356
a From the slope of Fig. 9A with Eq. (5).b From the slope of Fig. 9A with Eq. (6).c From the slope of Fig. 9B with Eq. (7).d From the value of log C of Fig. 10.e With the intercept of Fig. 9A and Eq. (5).
111M.R. Khan, Rafiuddin / Desalination 329 (2013) 103–114
of different concentrations on both sides of the membrane is greater inthe case of counter-ions than in the co-ion. The values of the ratio of themolar mobilities of the cation and anion ‘u+/u−’ (themobility ratio (ω)of the membrane) calculated for cadmium tungstate nanocompositemembrane are also represented in Table 1. The values ofω in themem-brane phase were found to be increasingwith dilution for all themono-valent electrolytes used. The high mobility is attributed to highertransport number of comparatively free cations as compared to theanion (Cl− ion) of the electrolytes.
When a slightly charged membrane is imposed between two solu-tions of an electrolyte of unequal concentrations c1 and c2 (c2 N c1)with the effective concentration of ion-exchange site of the membraneis much lower than electrolyte concentration (C), the following equa-tion was derived for the potential developed across the membrane[15,31]:
ψm ¼ RT=Fð Þ 2tmþ−1� �
lnγ þ 2 γ−1ð Þγ
where Xf is the membrane fixed charge concentration and t+m is the
counter-ion transport number in the membrane phase. Eq. (5) repre-sents a linear relationship between ψm and 1/c1 (Fig. 9A) which allowsthe evaluation of Xf and t+
m values of the nanocomposite membranefrom the slope and intercept for different electrolytes. The values areshown in Table 2. The t+
m values are found to be proportional to thevalues of Xf i.e. both are found to increase with the decrease in electro-lyte concentration. The results show that the counter-ion transportnumbers follow the order Li+ N NH4
+ N K+ N Na+ in the tungstatemembrane. The calculated values also reveal that the nanocompositemembrane exhibits moderate selectivity towards all cations and thatfor the same concentration, the selectivity of the investigated tungstatemembrane is comparatively more towards Li+ ions as selectivity isproportional to counter-ion transport number, as suggested by thevalues given in the table.
The fixed charge density of the membrane can also be calculated byanother widely accepted approach derived by Tasaka et al.  whosuggested the following relation:
−ψm ¼ RTF
Fig. 9. Plots for: (A) potential values, ψm and (B) 1/t− app against 1/c1 polystyrene blen
Here, φX is the effective fixed charge density of the negativelycharged membrane. Eq. (6) indicates that the plot of ψm against 1/c1will be linear. The plot shown in Fig. 9A has been used but with a differ-
ent slope equal toRTF
� fromwhich the values ofφX for different
electrolytes have been evaluated. The calculated values are also repre-sented in Table 2.
In the same above experimental condition, it was also shown thatthe inverse of the apparent transport number of the anion, t−app in ahigh salt concentration range could be expressed by the followingequation [17–19,35]:
þ α γ−1ð Þ1−αð Þγ lnγ
Here, α is the ratio of molar mobility of cation to the sum of molarmobilities of cation and anion, θXc is the effective fixed charge densityof themembranes under investigation and c1 (in mol/dm3) the concen-tration of the monovalent electrolyte in the lower concentration side ofthe cell. The apparent transport number of the anion t−app is defined bythe Nernst equation: ψm ¼ RT=Fð Þ 1−2t−app
� �ln c2
c1. Eq. (7) shows that
the values of α and θXc can be evaluated from the values of intercept
( 11−α) and slope ( α γ−1ð Þ
1−αð Þγ lnγ θXc) from the linear plot of 1/t− app against
1/c1 (Fig. 9B). The calculated values of the fixed charge densities θXc of
ded cadmium tungstate nanocomposite membrane for different salt electrolytes.
112 M.R. Khan, Rafiuddin / Desalination 329 (2013) 103–114
the polystyrene blended cadmium tungstate composite membrane arealso given in Table 2 which unveils the order LiCl N NH4Cl N KCl N NaCl.
3.4.1. PermselectivityThe term, permselectivity Ps is defined as a measure of preferential
permeation of counter-ions inside the membrane as compared to solu-tion (outside the membrane); and ion selectivity of an ion-exchangemembrane can also be expressed as a function of it. Applying approachproposed by Helfferich , permselectivity has been evaluated by thefollowing equation: Ps = (t+m − t+)/(1 − t+) where t+
m and t+ arethe true transport numbers of the counter-ion in the membrane andsolution respectively. The permselectivity arises because of the natureof the membrane for inequity between counter-ions and co-ions. Suchtype of discrimination arises as a result of the nature and magnitudeof the charge, the so-called concentration of fixed charge on membranesurface (Xf) that is associated with the membrane matrix. Xf can beexpressed in terms of permselectivity, Ps by the relation: X f ¼ 2CPsffiffiffiffiffiffiffiffiffi
. The calculated values of Ps are plotted against log C of the polysty-rene blended cadmium tungstate membrane in contact with variousmonovalent electrolytes (Fig. 10). The decline in permselectivity valueswith the increase in electrolyte concentration may be due to thereduction of the Donnan exclusion.
It was also suggested that the fixed charge density of a membranecould be calculated from the data of permselectivity . The massfixed transference number of anion τ− in the membrane is given bythe equation:
τ− ¼ 1−α
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi4ξ2 þ 1
4ξ2 þ 1q
þ 2α−1ð Þð8Þ
where ξ ¼ CθX
and α ¼ uþuþþu−
Here, the product θX is termed as the thermodynamically effectivefixed charge density of the membrane.
It was observed that the difference between τ− and t− app was lessthan 2% in a wide concentration range when the average concentration
Fig. 10. Permselectivity (Ps) against log C for polystyrene blended cadmium tungstatemembrane with different monovalent electrolyte salts.
(c1 + c2) / 2was replaced byC as suggested in literature . So Eq. (8)can be rearranged after replacing τ− by t− app in the following form:
1ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi4ξ2 þ 1
q ¼ 1−t−app−α
α− 2α−1ð Þ 1−t−app
� ≡ Ps: ð9Þ
Here Ps is the permselectivity of the membrane-electrolyte systemthe value of which can also be calculated from the membrane potentialdata by inserting the values of α and t− app. Eq. (9) indicates that if Ps =1, then the transport number of the co-ions (t− app) is zero i.e. themembrane is perfectly selective; while if Ps = 0, then t− app = 1 − αi.e., anions behave as in the bulk solution or as in a membrane havingno fixed charge. It is evident from the left hand side of equation thatwhen C becomes equal to θX , Ps attains the value of 1=
The value of C corresponding to which Ps = 0.447 will give the valueof the fixed charge density (θX). The values of the fixed charge densitiescalculated from the plots of Ps against log C (Fig. 10) for variouscadmium tungstate membrane-electrolyte systems are also includedin Table 2. The results furnish information that the fixed charge densityis highest for LiCl and lowest for NaCl for the same electrolytic concen-tration, indicating that the polystyrene blended cadmium tungstatenanocomposite membrane shows the highest selectivity towards Li+
ions. The same result has also been explained in terms of counter-iontransport numbers.
3.5. Electrical properties
The membrane resistance (Rm), capacitance (Cm) and impedance(Z) of the membrane/electrolyte system have been evaluated from themodel equivalent electrical circuit using the following equations.
Rm ¼ Rx 1þ Xx
� �2� �ð10Þ
Xx ¼ 1=νCx ð11Þ
Cm ¼ Xx
Z ¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiR2x þ X2
Here ν = 2πf and f is the frequency used to measure Rx and Cx. Thevalues of Rm, Cm and Z, thus calculated as a function of both bathingelectrolyte concentrations and the applied frequency for TMM aregiven in Tables 3 and 4.
The double layer theory was also used to measure the changesproduced in magnitude of Cx with change in electrolyte concentration.The magnitude of Cx is determined by the dielectric constant andthe membrane thickness, i.e. called geometric capacitance (Cg). Thepolarization charges on the geometric capacitor in the form of double
Table 3Electrical resistance (Rx) and capacitance (Cx) observed across themembrane equilibratedwith various concentrations of electrolyte solutions at 1 kHz (at 25 °C).
Cx (×10−2 μF) Rx (kohm)
KCl NaCl LiCl KCl NaCl LiCl
1 1.165 1.133 1.045 0.344 0.470 0.5200.7 0.795 0.814 0.840 0.723 0.856 0.8440.5 0.549 0.458 0.322 0.988 1.144 1.0980.25 0.332 0.321 0.184 1.325 1.423 1.4110.1 0.185 0.095 0.091 1.735 1.789 1.8020.07 0.094 0.082 0.044 2.180 2.202 2.234
Table 4Calculated values of impedance (Z) and double layer capacitance (Cd) from Eqs. (13) and(14) respectively, across the membrane equilibrated with electrolytes solution at 1 kHzfrequency (at 25 °C).
(kΩ) (×10−2 μF)
KCl NaCl LiCl KCl NaCl LiCl
1 0.447 0.376 0.448 33.72 2.34 2.090.7 0.842 0.667 0.822 16.00 1.43 1.080.5 1.110 1.126 1.423 10.42 1.22 0.840.25 1.524 1.605 1.772 9.11 0.62 0.290.1 1.841 2.850 2.982 8.74 0.19 0.200.07 2.834 4.494 5.625 7.98 0.95 0.08
113M.R. Khan, Rafiuddin / Desalination 329 (2013) 103–114
layer plays an important role and affects the overall electrical capaci-tance of the membrane. Cg in series with the two double layers andgiven by the expression
where Cm is the capacitance of the interfacial electrical double layer.For high electrolyte concentrations and significant surface
charge, 1/Cg ≫ 2/Cd, so that Cm ≈ Cg.Now, taking the value of Cm as Cg, the different values of Cd at other
electrolyte concentrations are calculated by using Eq. (14). It wasfound that the value of Cd increases with the increase in electrolyte con-centration. Cm should differ considerably from Cg when l/Cg = 2/Cd.This situation prevails in the absence of surface charge at lower electro-lyte concentrations.
The relatively high chemical as well as mechanical stability of poly-styrene blended cadmium tungstate nanocomposite membrane inwater renders this material potentially useful for electrochemical pur-poses. The nanocomposite membrane prepared with 25% polystyrenegives the best reproducible results and hence can be a potential and bet-ter candidate than many conventional membranes which degradeunder harsh conditions often encountered in various industrial applica-tions. The physicochemical properties (water content and volume voidporosity) of cadmium tungstate nanocomposite membrane are foundto be comparatively higher than other previously reported works[20,21] and decreasewith the increase in the percentage of polystyrene.The inorganic–organic composite membrane behaves as a weak cation-exchanger and its exchange capacity decreases abruptly with increasingamount of polystyrene. The counter-ion transport numbers, perm-selectivity and effective fixed charge densities are found to be in theorder LiCl N NH4Cl N KCl N NaCl for the same electrolytic concentrationindicating that cadmium tungstate membrane is more cation-selectivetowards Li+ ions. As theporosity is higher than that of calcium tungstatemembrane, cadmium tungstatemembrane shows higher potential .Moreover, the values of fixed charge densities calculated by differentmethods (Table 2) are found to be in close agreement. The electricalproperties of the membrane have been calculated (Tables 3 and 4).Characterization by TG-DTA, SEM, FTIR, and X-ray analyses revealedan adequate thermal andmechanical stability aswell as themorpholog-ical structure of the nanocomposite membrane, which is essential forthermally stable cation-exchange membranes.
List of symbols and abbreviations
c1, c2 concentrations (mol/L) of the monovalent electrolyte in thelower and higher concentration sides of the membranerespectively
C average electrolyte concentration (mol/L)SEM scanning electron microscopeTGA thermo gravimetric analysisDTA differential thermal analysisFTIR Fourier transform infra red spectroscopeXRD X-ray diffractionD crystallite sizeβ full width at half-maximum (FWHM)θ diffraction angle (degree)Wc
m water content in terms of water concentration (mol/L)ww wet membrane weight (g)wd dry membrane weight (g)τm volume void porosity of the membraneρw density of water (g/cm3)ρd density of the dry membraneψm membrane potential (mV)t−app transparent transport number of aniont+m counter-ion transport number in the membrane phaset+ transport number of the cation in the solution phaseτ− mass transference number of the anion
We are grateful to the University Grant Commission (UGC), NewDelhi for financial assistance for our research work. We also expressour great appreciation to the All India Institute of Medical Sciences(AIIMS), New Delhi for providing SEM images and the Departmentof Applied Physics, Aligarh Muslim University (AMU) for X-raydiffractograms.
 T. Xu, J. Membr. Sci. 263 (2005) 1–29. B.P. Tripathi, V.K. Shahi, Prog. Polym. Sci. 36 (2011) 945–979. J.N. Armor, J. Membr. Sci. 147 (1998) 217–233. S.K. Sikdar, D. Grosse, I. Rogut, J. Membr. Sci. 151 (1998) 75–85. V.B. Bart, V. Carlo, V.G. Tim, D. Wim, L. Roger, Environ. Prog. 22 (2003) 46–56. In: W.S.W. Ho, K. K. S. (Eds.), Membrane handbook, Chapman & Hall, 1992. K.P. Lee, T.C. Arnot, D. Mattia, J. Membr. Sci. 370 (2011) 1–22. M.A. Hickner, H. Ghassami, Y.S. Kim, B.R. Einsla, J.E. McGrath, Chem. Rev. 104 (2004)
4587. R.K. Nagarale, G.S. Gohil, V.K. Shahi, R. Rangarajan, Macromolecules 37 (2004)
10023. V. Ganesan, A. Walcarius, Langmuir 20 (2004) 3632. T. Chung, L.Y. Jiang, Y. Li, S. Kulprathipanja, Prog. Polym. Sci. 32 (2007) 483–507. M.S. Kang, J.H. Kim, J. Won, S.H. Moon, Y.S. Kang, J. Membr. Sci. 247 (2005) 127. P. Xing, G.P. Robertson, M.D. Guiver, S.D. Mikhailenko, S. Kaliaguine, Macromole-
cules 37 (2004) 7960. R. Souzy, B. Ameduri, B. Boutevin, G. Gebel, P. Capron, Solid State Ionics 176 (2005)
2839. M. Aizawa, S. Tomono, S. Suzuki, J. Membr. Sci. 6 (1980) 235. M. Tasaka, N. Aoki, Y. Kondo, M. Nagasawa, J. Phys. Chem. 79 (1975) 1307. Y. Toyoshima, Y. Kobatake, H. Fujita, Trans. Faraday Soc. 63 (1967) 2814. N. Kamo, M. Oikawa, Y. Kobatake, J. Phys. Chem. 77 (1973) 92. Y. Kobatake, H. Fujita, J. Phys. Chem. 72 (1968) 2871. M.R. Khan, Rafiuddin, J. Mol. Struct. 1033 (2013) 145–153. T. Arfin, Rafiuddin, Electrochim. Acta 54 (2009) 6928–6934. Y. Mizutani, M. Nishimura, J. Appl. Polym. Sci. 14 (1970) 1847. R.K. Nagarale, G.S. Gohil, V.K. Shahi, R. Rangarajan, J. Colloid Interface Sci. 287 (2005)
198–206. R.N. Reddy, R.G. Reddy, J. Power Sources 132 (2004) 315–320. S.W. Capeci, P.N. Pintauro, D.N. Bennion, J. Electrochem. Soc. 136 (1989) 10. ASTMD543-95, Standard Particles for Evaluating theResistance of Plastics to Chemical
Reagents, 1998. 27–33. U. Razdan, S.V. Joshi, V.J. Shah, Curr. Sci. 85 (2003) 761. A. Lehmani, P. Turq, M. Perie, J. Perie, J.-P. Simonin, J. Electroanal. Chem. 428 (1997) 81.
114 M.R. Khan, Rafiuddin / Desalination 329 (2013) 103–114
 E. Boccaleri, A. Arrais, A. Frache, W. Gianelli, P. Fino, G. Camino, Mater. Sci. Eng. BSolid 131 (2006) 72–82.
 R.K. Nagarale, V.K. Shahi, R. Schubert, R. Rangarajan, R. Mehnert, J. Colloid InterfaceSci. 270 (2004) 446.
 G.S. Gohil, V.K. Shahi, R. Rangarajan, J. Membr. Sci. 240 (2004) 211–219. W.U. Malik, S.K. Srivastava, V.M. Bhandari, Satish Kumar, J. Electroanal. Chem. 110
 F.G. Helfferich, Ion Exchange, Dover Publication, New York, 1962. F.A. Siddiqi, I.R. Khan, S.K. Saksena, M.A. Ahshan, J. Membr. Sci. 2 (1977) 245. H.R. Khan, S. Ozeki, J. Colloid Interface Sci. 177 (1996) 628–632. V.K. Shahi, G.S. Trivedi, S.K. Thampy, R. Rangarajan, J. Colloid Interface Sci. 262
(2003) 566–573. Y. Kobatake, N. Takeguchi, Y. Toyoshima, H. Fujita, J. Phys. Chem. 69 (1965)