No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Relation between transport number and concentration of Donnan salt in membranes
Abstract
In this study the relation between the transport number of counterion and the concentration of Donnan salt was investigated to elucidate the decrease of the transport number in higher concentrations of electrolyte solution equilibrated with the membrane. In the experiment three different types of membranes were examined, and the membrane potential and membrane conductance were measured. In addition, the ionic concentration within the membrane was determined to estimate the transport number due to Donnan salt, tD. The tD was given as a function of Donnan salt concentration, and the mean mobilities of ions which exist as Donnan salts within the membrane were estimated. The values of ŪD also depended upon the valences of counterions and types of membranes. As a result, it was pointed out that the transport numbers of Donnan salts in cases of higher concentration ranges of outer electrolyte solutions were apparently close to those in the aqueous solution, and the decrease of transport numbers evaluated from membrane potential were attributed to the low values of tD.
... A decrease in SO 4 2− ions transfer with the increasing in concentration is essentially due to the decrease in the membrane selectivity (Rodrigues et al. 2008;Vallois et al. 2003;Raissouni et al. 2007;Amado et al. 2004;Boucher et al. 1997;Yamauchi et al. 1991;Hirata et al. 1992). ...
... The Faradic process at the anode (oxygen evolution) results in the regeneration of free protons in the feed compartment. This leads to a reduction in current efficiency for electrodialysis due to the competing transport of hydrogen ions across the membrane, especially as the free protons are more mobile than the metal ions (Caprarescu et al. 2012a, b;Amado et al. 2004;Boucher et al. 1997;Yamauchi et al. 1991;Hirata et al. 1992). ...
The main goal of this study was to explore the suitability and performance of bicomponent polymer membranes based on acrylonitrile copolymers-polyvinyl alcohol (PVA) mixtures and small quantities of anion (Purolite A100) or cation exchange resin (Purolite C150), prepared by phase inversion. Membranes were used for copper removal from synthetic wastewater solutions. A three detachable cylindrical compartment electrodialysis cell without recirculation of the electrolytes and synthetic solutions of various concentrations, similar to a copper electrowinning electrolyte, were used. The electrodialysis unit operates under galvanostatic control. The effect of pH on electrodialysis separation of Cu2+ and on the solution conductivity has been also investigated. The laboratory electrodialysis cell performance was evaluated in terms of percent of extraction (pe) and current efficiency (ce). Experimental results showed that the ionic transfer in electrodialysis cell was especially affected by concentration. The highest values for the pe (>81 %) and the ce (>25 %) of copper ions were obtained at maximum concentration in copper ions (3 g/L), indicating a better performance of the ion extraction. The transport of copper ions was also correlated with flux data. The ion exchange membranes were characterized using FT-IR spectroscopy, ESEM and Electrochemical Impedance Spectroscopy.
... Amphoteric charged (ion-exchange) membranes are expected to be the next-generation ion exchange membranes for the following features: controllability of the charge property by changing the pH of the outer solution and their potential as an anti-fouling material that prevents adsorption of organic molecules and biological macromolecules on the surface [181]. They are expected to be utilized in the biomedical and industrial fields, such as medical devices (e.g., hemodialysis membrane) [182][183][184], separation of ionic drugs and proteins [181,[185][186][187][188] and depletion of electrolytes by nanofiltration [189] or piezodialysis [190][191][192][193] as well as high performance gel actuator [194]. ...
During the last 50 years, ion exchange membranes have evolved from a laboratory tool to industrial products with significant technical and commercial impact. Today ion exchange membranes are receiving considerable attention and are successfully applied for desalination of sea and brackish water and for treating industrial effluents. They are efficient tools for the concentration or separation of food and pharmaceutical products containing ionic species as well as the manufacture of basic chemical products. The evolvement of an ion exchange membrane not only makes the process cleaner and more energy-efficient but also recovers useful effluents that are now going to wastes, and thus makes the development of society sustainable. Therefore, the intention of this review is to give a brief summary of the different preparation and characteristics of ion exchange membrane as well as their potential applications. The most relevant literatures in the field are surveyed and some elucidating case studies are discussed, also accounting for the results of some research programs carried out in the author's laboratory.
... The potential for the blended membranes was measured according to conventional methods [17, 18]. The membrane potentials in two membrane blend ratios were measured in two parts of KCl concentration — lower (10 !3 to 10 !1 mol dm !3 KCl) and higher (10 !2 to 2 mol dm !3 KCl) — concentration ranges, respectively. ...
Nafion (fluorocarbon polymer) and collodion blended with Nafion in the ratio of CN5:5 composite solutions were used to produce thin membrane films on Au electrodes. It was found that the transport properties of water and KCl, such as the filtration coefficient, Lp, reflection coefficient, σ and salt permeation, ω, across the two composite collodion (C) and Nafion (N) membranes depended on the quantities of the Nafion component. Moreover, the study indicated that a better diffusion coefficient of redox substances can be achieved by blending the Nafion solution with an ionic group to collodion solution compared to Nafion only. The uptake of neutral species of ferrocenedimethanol (FcDM), positively charged species of methyl viologen (MV2+), and negatively charged potassium ferrocyanide [Fe(CN)6]4− ions were studied by cyclic voltammetry and chronopotentiometry. The diffusion coefficient of MV2+ and FcDM in blended Nafion:collodion showed higher values compared to that in Nafion alone. The diffusion of positively charged species, MV2+, was decreased in the membrane matrix compared to the neutral species, FcDM. This may be attributed to the existence of an interaction between SO3− functional groups of the fluorocarbon polymer and MV2+ that led to mobility restriction of MV2+ compared to FcDM. The diffusion of the [Fe(CN)6]4− species was not clearly observed due to the electrostatic repulsion effect between SO3− functional groups of the flourocarbon polymer and the negative species, [Fe(CN)6]4−.
Elekrochemical modeling description of ion-exchange sistems.
Corrosion test and surface analysis were conducted in the H2S-Cl-environments to elucidate the role of alloyed molybdenum on the corrosion resistance of Ni-Cr-Mo-Fe alloys. The alloyed molybdenum improves the localized corrosion resistance. The surface film is double layers which comprise sulfide including molybdenum and chromium oxide. However, the Ni-Mo-Fe alloy also shows good corrosion resistance in the H2S-Cl-environment. This good corrosion resistance is considered to be due to the cation selectivity of molybdenum sulfide, which can form in such environment. The role of alloyed molybdenum on the corrosion resistance of Ni-Cr-Mo-Fe alloys in H2S-Cl-environments is proposed.
The effect of N,N-dimethylacetamide on the selectivity of heterogeneous (MK-40, MA-40, and MA-41) and homogeneous (MF-4SK)
ion-exchange membranes is studied for the first time. Concentration dependences of electrical conductivity and diffusion permeability
of the membranes were measured experimentally over wide lithium chloride concentrations; on their basis, electrodiffusion
coefficients of the co- and counterions were calculated. The interrelation between the electrodiffusion coefficients and the
specific moisture capacity of the heterogeneous and homogeneous membranes (which affect their selectivity) is revealed. The
calculated electrodiffusion coefficients were used in the calculations of the electromigration transport numbers of counterions
in the initial membranes and those processed in mixed solvent. It is shown that the heterogeneous membrane selectivity either
increased under the action of the aprotic solvent to polymer material (MA-40, MA-41) or remained practically unchanged (MK-40);
the selectivity of homogeneous perfluorinated membranes (e.g., MF-4SK) decreased, thus approaching that of the studied heterogeneous
anion-exchange membranes.
In partition equilibria of Ca2+, K+ and Cl– between a swollen charged membrane and an aqueous bath of mixed KCl and CaCl2 electrolytes, the equilibrium concentration of the bivalent ion in the membrane decreases to a minimum value as its concentration in the bath increases. The value of the concentration minimum decreases with increasing KCl : CaCl2 concentration ratio in the bath. Simulations and analysis have been performed on the basis of Donnan distribution theory.
Piezodialysis in weakly amphoteric polymer membranes is discussed from the viewpoint of non-equilibrium thermodynamics. Succinyl chitosans were used as the weak polyampholytes. With varying pH in the outer solution, the apparent isoelectric point of the membrane was defined as the point of zero membrane potential, where solute permeation is much enhanced. Piezodialysis was confirmed from the phenomenological hydraulic permeability coefficients, reflection coefficients and solute permeability coefficients of the membrane, which were based on non-equilibrium thermodynamics. Permeation experiments with mixed ion systems were also performed. The permeation behaviour was in accordance with that expected for the interactions of ions with polyampholytes in aqueous solutions.
Two different composite collodion membranes were prepared to examine the correlation of material transport to membrane charges or membrane structure from comparative study. One is a composite collodion membrane incorporating only perfluorobenzoic acid (PFBA) and the other is the same membrane but with pores formed by adding NaClO4. The membrane parameters defined by irreversible thermodynamic consideration such as filtration coefficient of water, salt permeability and reflection coefficient were estimated. The differences between two membranes or the dependence of membrane transports on salt species were discussed on the basis of the frictional coefficients between water-membrane skeleton or salt-water interactions.
Crosslinked sodium alginate (SA) membranes which have a protective polyelectrolyte complex layer at a membrane surface were prepared by a simple technique for the reverse osmosis separation of an anionic surfactant that has been used as an emulsifier in PTFE emulsion polymerization. With this technique, when a nascent SA film cast onto a glass plate was dipped in the reaction solution that contained both CaCl2, a crosslinking agent, and chitosan, a cationic polymer, the stable SA membrane with double layer structure was fabricated by crosslinking with the divalent ion Ca2+ inside the membrane and the simultaneous polyelectrolyte complexation of anionic SA with cationic chitosan at the membrane surface. The effects of the fabrication parameters on the characteristics of the resulting membrane were investigated. The complex membrane gave excellent membrane performance with rejections higher than 99% due to electrostatic repulsion between anionic solute molecules and the anionic membrane. Also, a comparison of membrane performance between crosslinked SA membranes with and without the protective complex layer was made at different operating conditions in the reverse osmosis separation of the anionic surfactant from waste fluids. The role of the protective polyelectrolyte complexes formed at the membrane surface is discussed in terms of membrane stability.
In an amphoteric membrane-mixed electrolyte system, the ionic concentration within the membrane and the membrane conductance were measured. The additivity concept proposed in our previous paper [J. Membrane Sci, 48 (1990) 25] was applied to the present mixed electrolyte system and each ion transport number in the mixed system was estimated. The transport number of Ca2+, tCa, was larger than tNa in the whole range concentration and decreased with increasing concentration of the outer electrolyte solution. Additionally, to ascertain whether the predicted values are reasonable, electrodialysis in the mixed electrolyte system was carried out under the same experimental conditions. The transport numbers estimated in the mixed electrolyte system were approximately in good agreement with the values obtained from the electrodialysis experiments. Furthermore, the transport numbers for two different ionic states in the membrane were evaluated and the concentration dependence of the total transport number, tM, was discussed on the basis of the results.
A new technique has been established to fabricate thin film composite membranes, by which a hydrophilic polymer could be coated in thin film on a hydrophobic support membrane. The new technique was composed of two steps: dispersion of a reactant to the hydrophilic polymer in the hydrophobic support membrane and interfacial reaction between the reactant and the hydrophilic polymer to produce thin film of the hydrophilic polymer on the support membrane. Composite membranes in which a thin film of sodium alginate is coated on a polysulfone support membrane were prepared by the new technique for the reverse osmosis separation of anionic surfactant–water mixture. Two methods were employed to fabricate a thin film of sodium alginate on the support membrane: (1) dispersion of the crosslinking agent, CaCl2 alone in the support membrane and (2) dispersion of CaCl2 in the support membrane with help of PVA which adheres fast to the support membrane. The formation mechanism of the thin layer was suggested schematically on each method. Both the methods could produce successively a thin layer of SA on the support membrane. Especially, method (2) gave a strong bonding of the thin layer on the support because of the large contact area with the support through the PVA layer which sticks fast to the SA layer. From the SEM pictures and permeation experiments, the method (2) was confirmed to be better to produce a defect-free thin film of SA on the support membrane.
Composite collodion membranes modified by the addition of different amounts (0–1.0 wt.%) of NaClO4 were prepared keeping a fixed concentration of perfluorobenzoic acid (PFBA) within membrane. The composite membranes indicated relatively high cation transport number and proved to be available as a cation exchange membrane. From volume flow and salt flow studies across membranes, the filtration coefficient, Lp, and the salt permeability, ω, were estimated as a function of added NaClO4. As a result, while Lp increased gradually in proportion to the amounts of NaClO4, ω indicated an abrupt increase of around 0.2 wt.% of NaClO4. The similar critical behavior was found in the measurement of membrane conductance and the result assured the critical change in ω.Furthermore, based on the above membrane parameters such as Lp, ω and Λm, the frictional consideration regarding the interactions between water and membrane matrix or between salt and membrane matrix was done and the transport properties were discussed in relation to the membrane structure.
The present study indicated that stable blended collodion/Nafion membranes can be obtained by adding Nafion solution to 50% in volume to collodion solution. The mixed solution beyond the critical volume of Nafion solution leads to the bringing about of phase separation. The blended membranes obtained as the cast membrane indicated almost the same permselectivity as ordinary cation exchange membrane. Regarding three different membranes including pure Nafion membrane, water flow and KCl flows were investigated systematically. The transport properties of water and KCl across the composite membrane and other characteristics such as water content in the membrane depended on the quantities of Nafion component and they increased with the increase in Nafion component. It was suggested that the transport properties changed according to the existence of water accompanied by Nafion polymer in the composite membrane. In addition, the effective charge densities of composite membranes were estimated from the membrane potential measurements and were compared with conventional cation exchange membrane. As a whole, the composite membranes indicated membrane performance comparable to the commercial ion exchange membranes.
In order to get a better understanding regarding material transport through charged membranes, two kinds of the composite collodion membrane were prepared and investigated by systematic and comparative studies. One of the membranes is a composite collodion membrane incorporating different quantities of perfluorobenzoic acid (PFBA), and the other is the same membrane but with pores formed by adding NaClO4. Several membrane parameters defined by thermodynamic consideration such as membrane conductance and electroosmotic coefficient were estimated from corresponding experiments and were examined on the basis of frictional consideration. The differences between the two membranes were also discussed in relation to the structure of the membrane skeleton or charge densities within the composite collodion membrane.
A charged mosaic membrane with parallel array of different negative and positive charges was prepared from microsphere gel. Several characteristics on the novel membrane were investigated through experiments concerning transport studies, membrane potentials and membrane resistance. From analysis of the volume flux and salt flux, preferential salt transport across the charged mosaic membrane was suggested. Membrane potential did not indicate a constant value and the absolute value decreased rapidly in short time. The large time dependence supported the interpretation on salt flow in transport studies. From potential measurement, cationic and anionic transport numbers in membrane also were determined to =0.41 and =0.59. Membrane resistance of this mosaic membrane indicated slightly higher values than that of ordinary charged membrane.
Several characteristics of a charged mosaic membrane with parallel array of negative and positive charges were investigated by using transport studies and the related analysis. From an analysis of the volume flux and salt flux based on irreversible thermodynamics, preferential salt transport across the charged mosaic membrane was clearly demonstrated. Additionally, transport properties of amino acids and sucrose through the charged mosaic membrane were estimated, relatively, on the basis of KCl transport. As a result, amino acid transport depends largely on the charged states and molecular weight; however, for sucrose transport, non-electrolyte was rejected under all experimental conditions.
Excess charges in negatively ionized membranes carried by the mass movement were determined experimentally from the data of electroosmosis and streaming potential for systems of collodion based polystyrene sulfonic acid membrane in KC1 solutions. The charge density effective to the mass movement in the membrane was found to be several percent of the stoichiometric density of charges fixed on the polyelectrolyte skeletons constituting the membrane. From this value together with the experimental data of the hydraulic permeability of the same membrane-electrolyte system, the effects of mass movement on the membrane potential and on the permeability of the electrolyte component were evaluated to be a few percent for the membrane potential and less than ten percent for salt permeability, respectively. Thus the mobilities of co- and counter-ions in charged membrane determined from the data of membrane potential and salt permeability of the membrane, as reported in Part I of this series, may be valid within ten percent. This was confirmed by comparing the experimental and theoretical values of the DC conductance for the membranes in various KC1 concentrations of the external solution.
It is well known that the asymmetry of a membrane structure is closely related to its functions. In the asymmetric membrane with respect to the membrane charge distribution, facilitated and reverse transport of electrolytes were theoretically predicted and the reverse transport of NaCl through the asymmetric membrane was observed experimentally. The object of this article is to make clear the mechanism of membrane functions resulted from the asymmetric membrane charge distribution. At first, the effect of the asymmetric membrane charge distribution on the membrane potential is described. In addition, the way to determine the values of parameters which characterize the membrane properties is shown. Next, the mechanism of the facilitated and the reverse transport of electrolytes is described correlating with the membrane potential. Finally, it is shown that the facilitated and the reverse transport can not take place in a composite membrane formed by piled up two symmetric membranes which differ one another in membrane charge density.
In an amphoteric ion exchange membrane-electrolyte system, three independent experiments were carried out, that is, the membrane potential, membrane conductance and concentration of ions within membrane were measured, respectively. The electrolytes adopted in the systems were NaCl, CaCl2, LaCl3. From the analysis of the results, the specific conductance attributed to Donnan salt within membrane was separated. As a result, it was pointed out that the dynamic state of Donnan salt is obviously different from that of salt in aqueous solution. Furthermore, the comparison between the ions constituting Donnan salt and the counterions which interact with the ion-exchange sites suggested us the existance of two different dynamic states on the ions within membrane.
In order to clarify the membrane phenomena in an amphoteric membrane, four different experiments were carried out in a simple electrolyte-amphoteric membrane system. The ion transport numbers obtained from the membrane potential and electrodialysis were approximately in agreement with each other. It was indicated that the transport numbers strongly depended on the ionic valence in electrolyte solution. The membrane permeability coefficient representing the dynamic behavior of ion within membrane was evaluated in terms of membrane potential, transport number, and membrane conductance. In addition, the ionic concentrations within membrane equilibrated with outer electrolyte solution were measured to obtain the mobilities within membrane. The membrane phenomena were discussed on basis of the present experimental results.
A method for the preparation of ion exchange membranes called the “paste method” has been developed into a semicontinuous preparation.The roll-coating system being employed, a base membrane 1 m. wide and 50 m. long is prepared continuously; the ion exchange group is then introduced to it.The ion exchange membranes thus prepared have been proved to have excellent electrochemical and mechanical properties as well as an excellent appearance.
The water transference due to electroosmosis was studied for several cationexchange resin membranes of sulfonated phenol resin type, mainly in sodium chloride solution. It was shown that (1) the electroosmosis depended on the transport number of the membrane and the amount of water contained in the resin, and (2) the experimental results could be explained by introducing the idea of fixed water and mobile water in the membrane. The fixed water is supposed to be composed mainly of the hydration water attached to the polar groups of the resin.
Measurements of bi-ionic potential across four porous cation-exchange membranes, a porous anion-exchange membrane, and a nitrobenzene liquid membrane for KCl/NaCl and KCl/LiCl systems were carried out and the experimental data were analyzed by a theory based on non-equilibrium thermodynamics. For porous charged membranes the value of bi-ionic potential depends on the diffusion potential in the membrane phase. That is, the change of bi-ionic potential is based on the change in the concentration of co-ions in the membrane phase. However, Donnan potentials at the two solution/membrane interfaces compensate each other and are negligible compared with the total membrane potential. For the nitrobenzene membrane, the value of bi-ionic potential varies largely with the electrolyte concentration even though the transport number of the counterions in the membrane is nearly unity. The behavior of the concentration dependence of the bi-ionic potential across the nitrobenzene membrane is markedly different from that across porous charged membranes.
Rates of permeation of water and electrolytes through an anion exchange membrane, prepared by the adsorption of a polycation on preformed collodion membrane, were measured. The experimental results are in agreement with a theory presented previously. An equation for the salt-concentration dependence of the reflection coefficient was derived and compared with both the theory of Kedem and Katchalsky and the experimental data. The agreement between the two theories and experiments is good. It is shown quantitatively that the total entropy production in negative osmosis is positive, because the absolute value of the entropy production due to solute is very much larger than that due to water.
Membrane potentials and apparent transport numbers of the cation are reported for cured cellulose acetate membranes bounded by HCl, NaCl, KCl and MgCl2 solutions, using Ag/AgCl electrodes and a flow-cell method. Membranes cured at 70°, 80° and 90° are used. Bounding solution concentrations vary from 0.005 to 0.05 M at the high concentration side (bounding the dense side of the membrane), and are kept constant at 0.002 M for the low concentration solution. In the KCl 90° membrane case the low concentration solution is varied as well, from 0.0001 to 0.002 M. Results show that cured cellulose acetate membranes are permselective towards univalent cations. This is interpreted as resulting from a low cation-exchange capacity of the dense layer of the cured membrane. The permselectivity increases with increased curing temperature. Addition of a non-electrolyte to the low concentration side reverses the osmotic flow and leads to higher apparent transport numbers of the cation. It is concluded that diffusion in small pores contributes significantly to the transport of ionic solutes through uncompacted membranes.
The membrane phenomena in an amphoteric membrane-single-electrolyte or mixed-electrolyte solution systems were examined. The mobilities of the ions within the membrane were evaluated from the relation between the transport number and membrane conductance after determination of the ionic concentration within the membrane. In the single-electrolyte system, the mobilities depending on the ionic charges were observed. The facts supported that ion-transport phenomena were mainly controlled by the electrostatic interaction between the ion-exchange sites and the ions within the membrane. On the other hand, the cation with the smaller valence was always excluded out of the membrane in the mixed-electrolyte system. When comparing the ratios of mobilities in the single-electrolyte system with those in the mixed-electrolyte system, large differences between them could be observed. It is conjectured that one of the reasons may be the difference in the Donnan adsorption salt concentration within the membrane.
Transport numbers of several aliphatic and aromatic carboxylate ions across an anion exchange membrane were determined on the basis of the transport number of the co-ions across the membrane in electrodialysis. Also, the specific conductivity and water content of each carboxylate ionic-form membrane were measured.The transport number of lower aliphatic carboxylate ions decreases slowly with increasing molecular weight, but it decreases markedly where the carbon number is over five. In particular, the transport number of caprylate or of caprate ions is below 0.5, and these ionic-form membranes apparently lose the characteristics of an anion exchange membrane. For aromatic carboxylate ions, the benzoate ion a higher transport number, but p- or O-hydroxybenzoate ions only about 0.6. The specific conductivity and water content of each carboxylate ion having a lower transport number are very small compared with other carboxylate ions of higher transport number. This suggests that the hydration number and the activity of ions with a lower transport number are small in anion exchange membranes.
To obtain a better understanding of the ion selectivity in an amphoteric ion-exchange membrane—mixed electrolyte solution system, three independent experiments were carried out. The transport numbers of ions across the amphoteric membrane were evaluated through electrodialysis experiments. The membrane conductances were obtained from the measurements of membrane resistances. The ionic concentrations within the membrane were given as the total amount of Donnan salt and ion-exchange capacity. The conventional selectivity coefficient and ionic mobility within the membrane were calculated by means of three kinds of data, and given as a function of calcium ion concentration in the solution. The order of Ca2+ > Na+ > Cl− in selectivity turned out to be determined by the product of ionic mobility and ionic concentration within the membrane. In particular, it was pointed out that the partition coefficients (M/aM) played an important role in ion selectivity in this mixed system. A new physical quantity for estimating selectivity, i.e. the permeability coefficient, P, was proposed.
Transference number of counterions in charged membranes with various concentrations of KCl solution were measured both by the Hittorf and by the emf methods. Two transference numbers obtained for a given membrane differed from each other in their magnitudes, notably in concentrated salt solution. Theoretical equation for the transference number of cations in a negatively charged membrane was derived on the basis of the thermodynamics of irreversible processes in conjunction with an approximate hydrodynamic equation for the movement of the local center of mass. The derived equation illustrates that the transference number obtained by the Hittorf method is composed of two terms; one associated with the movement of individual ion species relative to the local center of mass, and the other with the mass flow. The first term was evaluated with use of our previous results on mobilities and activities of mobile ions in a charged membrane for the same system studied here. Comparison between theoretical and experimental values revealed that the transference number relative to the mass movement was practically the same as that determined from the emf method. The second term of the Hittorf transference number of counterions could be evaluated from the data on the electroosmotic coefficient for the system. It was found that the contribution to the observed Hittorf transference number of the term associated with the mass movement predominates in the concentrated salt solution, and that the marked depression of charge density effective to hydrodynamic properties of the membrane was attributed to decrease of amounts of the counterions carried by the mass flow.
In order to clarify the mechanism of ion transport across an amphoteric membrane with sulfonate and N-methylpyridinium groups as exchange sites, membrane potential and membrane conductance measurements were carried out with varying valences of cations and anions of the electrolyte solution. It was shown that the membrane studied behaves as a cation exchange membrane in 1-1 valent and 2-2 valent electrolyte solutions. On the other hand, the membrane function was strongly influenced by the valence of the cation; the membrane in the LaCl3 system in particular was preferentially anion selective. The permeability coefficients were evaluated according to the theory of nonequilibrium thermodynamics, and ion transport was discussed.
The bi-ionic membrane potential has been studied on a system consisting of electrolyte solutions separated from another one by an ion exchange membrane.The slope of the linear portion of the curve of potential vs. log activity ratio, log , was found to be ) mv at 20°C, where aM and zM denote respectively the activity and the charge of permselective ion M present on one side of the membrane and aN and zN, those of the ion N on the other side. This was in good agreement with the theoretical prediction provided that the membrane is permselective to either cations or anions and the relative membrane permeability is a constant.