Soil Science Society of America Journal 66:1518-1525 (2002)
© 2002 Soil Science Society of America
DIVISION S-2—SOIL CHEMISTRY
Electrochemical Water Splitting at Bipolar Interfaces of Ion Exchange Membranes and Soils
Brian M. Desharnais*,a and
Barbara Ann G. Lewisb
a Dep. of Civil and Environmental Engineering, Tri-State Univ., 1 University Ave., Angola, IN 46703
b Dep. of Civil Engineering, Northwestern Univ., 2145 Sheridan Road, Evanston, IL 60208
* Corresponding author (desharnaisb{at}tristate.edu)
 |
ABSTRACT
|
|---|
Under the influence of a direct current, the dissociation rate of water in contact with bipolar ion exchange membranes (BPMs) is 106 to 107 times faster than in free solution. This phenomenon, termed accelerated water splitting, is well known in industry where BPMs are designed for electrosynthesis of acids and bases. In this work, it is hypothesized that (i) accelerated water splitting can also take place at the bipolar interface between ion exchange membranes (IEMs) and the ion exchange surfaces of soils, and (ii) electroosmosis plays a key role. If the IEM has an electrostatic charge opposite in sign to the predominant charge on the soil colloidal particles, the interface is, in effect, bipolar. If an external electric field is then applied, conditions can give rise to accelerated water splitting. Laboratory experiments performed on various mixtures of Ottawa sand, bentonite, talc, and anion exchange resin indicate that accelerated water splitting occurs when the free pore solution in the low permeable soil moves away from the bipolar interfaces due to electroosmosis, thus causing an unsaturated zone at these interfaces. Accelerated water splitting then initiates at these interfaces since there are not enough counterions in contact with the IEMs to maintain an ionic current. Very little cation exchange capacity (CEC), anion exchange capacity (AEC), or clay is needed for water splitting to occur at a bipolar IEM and soil interface.
Abbreviations: AEC, anion exchange capacity • AEM, anion exchange membrane • BPM, bipolar ion exchange membrane • CEC, cation exchange capacity • CEM, cation exchange membrane • IEM, ion exchange membrane
|
INTRODUCTION
|
|---|
TRADITIONAL BOUNDARIES separating various fields of science, and boundaries between science and engineering, are rapidly blurring; interdisciplinary research is demonstrating that new scientific and engineering discoveries can arise from disciplinary crossovers. This is particularly true in soil science, which application has spread beyond the agricultural to contaminant fate in soil, and to soil remediation. We describe here a soil colloidal phenomenon not previously reported, that we encountered in the process of applying soil colloid chemistry to an engineered system.
Under the influence of a direct current, the dissociation rate of water in contact with BPMs is 106 to 107 times faster than in free solution (Simons, 1985; Strathmann et al., 1993). This phenomenon, termed accelerated water splitting, is well-known in industry where BPMs are designed for electrosynthesis of acids and bases. Separation of H+ and OH- is accomplished without chemical addition or gas evolution, a process whose energy requirement is theoretically 
40% of that of water electrolysis with gas evolution (Jorissen and Simmrock, 1991). This water splitting is believed to occur within a region of 1 to 10 nm above the surface of the membrane.
While developing a sampler for soil solutions utilizing electrokinetic techniques and IEMs, we observed phenomena that could only be explained by postulating that accelerated water splitting was taking place at the interface between the ion exchange surface of the soil and IEM. If the IEM has an electrostatic charge opposite in sign to the predominant charge on the soil colloidal particles, the interface is, in effect, bipolar. If an external electric field is then applied, conditions can give rise to accelerated water splitting.
Laboratory experiments to test our hypothesis are described here; in brief, accelerated water splitting was reproducibly demonstrated. As far as we have been able to ascertain, this is the first report of the phenomenon in soils, and these findings indicate that accelerated water splitting is an important consideration when utilizing electrochemical devices that employ IEMs in contact with low permeable soils. For example, one enhanced technique for the electroreclamation of low permeable soils is to wrap the electrodes with IEMs in order to mitigate the transport of undesirable electrolysis products from entering the soil being treated. Specifically, it may be advantageous to inhibit the acid front generated at the anode with anion exchange membranes (AEMs) or the alkaline front generated at the cathode with cation exchange membranes (CEMs). The encapsulation of anodes with AEMs, and cathodes with CEMs, may not be feasible with low permeable soils possessing significant net cation exchange capacities and anion exchange capacities, respectively. These two IEM and media interfaces are bipolar since the fixed charged functional groups on the IEM and media are oppositely charged, and accelerated water splitting may occur, thus negating the intended purpose of the IEMs.
|
Bipolar Ion Exchange Membrane Theory
|
|---|
The BPM generally utilized in industrial processes contains an anion selective layer on one side and a cation selective layer on the other side (Fig. 1)
. Membranes are manufactured with a variety of fixed charged functional groups including (i) sulfonate, carboxylate, and sulfonyl groups for cation selective layers and (ii) quaternary, tertiary, secondary, and primary ammonium groups for anion selective layers. Sulfonate groups and quaternary ammonium groups, the most common, are strong acids and bases, respectively. The rest are weak acids and weak bases. In the presence of a direct current, ions from the interface within the BPM migrate to their respective electrodes. After essentially all of the salt ions are depleted from the solution at the interface, maintenance of the direct current can only be maintained by the hydrogen ions and hydroxide ions available from the ionization of water (Strathmann et al., 1993). Due to the principle of electroneutrality, counter ions must be present in the outside solution to balance the charge of ions. For the system illustrated in Fig. 1, the counterbalance ions are K and chloride ions in the anolyte and catholyte, respectively. This process is highly dependent on the ability of the membrane to diffuse water to the interface and the strength of the direct current.
Several mechanisms have been postulated to explain this phenomenon of water splitting (Strathmann et al., 1993). One mechanism is the increase in the water dissociation rate constant (kd) due to electric field effects. If only this mechanism is considered, the calculated water dissociation rates are at least three orders of magnitude smaller than measured experimental values. A second mechanism is a catalytic proton transfer reaction between the fixed charged functional groups and the interfacial water. Experiments suggest that this mechanism originates primarily in membranes possessing quaternary and tertiary ammonium groups (Simons, 1979; Simons, 1985; Ramirez et al., 1992). Actually, the responsible functional group in the anion selective layer is the tertiary ammonium group, but the phenomenon readily proceeds at quaternary ammonium groups after a significant amount of the quaternary groups have degraded to the tertiary form. Water splitting also occurs within membranes containing weaker basic ammonium groups and weaker acidic carboxylic and phenolic groups, but these rates are at least 1 to 2 orders smaller than those for membranes containing quaternary and tertiary ammonium groups (Simons, 1985). The second mechanism appears to occur only at a threshold current density or limit current density, which is 0.005 mA cm-2 or 0.08 mA cm-2 for the membranes possessing ammonium groups or carboxylic groups, respectively (Simons, 1985). This phenomenon was not noticeable within membranes containing sulfonate fixed charged sites if the system is sufficiently clean and void of (i) dissolved charged impurities or (ii) adjacent solutions containing appropriate ionizable groups (Simons, 1979). Monolayer IEMs containing the above mentioned responsible functional groups behave similarly to BPMs when they possess significant depositions of charged impurities on their surfaces, such as long-chained charged polymers (Johnsson and Boesen, 1984). The catalytic reactions for acid and base groups are shown below (Simons, 1985). B denotes a neutral base group such as an alkyl amino group (i.e., R3N, where R is an alkyl), and AH denotes a neutral acid group such as a carboxylic group (i.e., COOH). Proton Catalytic Reactions Involving Base Groups
Proton Catalytic Reactions Involving Acid Groups
|
Electroreclamation of Low Permeable Soils
|
|---|
When a direct voltage is applied across a soil, electrokinetic transport processes (electrophoresis, electromigration, and electroosmosis) occur. Electrokinetic theory in soils has advanced markedly, especially with respect to soil chemistry effects on electrokinetic transport (Casagrande, 1949, 1952; Gray and Mitchell, 1967; Segall et al., 1980; Lageman, 1989; Cabrera-Guzman et al., 1990; Alshawabkeh and Acar, 1992, 1996; Acar and Alshawabkeh, 1993; West and Stuart, 1995; Denisov et al., 1996; Dzenitis, 1997). Electroosmosis (the transport of solutes in the pores of the soil due to the movement of the soil solution in an applied electric field) is particularly relevant here and bears some discussion.
Double layer theory in soils postulates (i) the free water layer and (ii) the boundary film of water within the diffuse electric double layer surrounding clay particles, organic colloids, and metal oxides. To maintain electroneutrality, counterions are associated with the charged particle surfaces. Under normal hydraulic flow, it is usually assumed that the boundary water film is not affected and only water within the free layer moves under the hydraulic gradient. (In actuality, the two water layers are a continuum, and the boundary between the layers is not clearly defined.) Under the influence of an electric field, the counterions and their associated water molecules on the particle surfaces will migrate towards their respective electrodes. This movement imparts a net strain on the pore fluid around the hydration shell of the counterions, causing a net shear force to develop through the viscosity of the pore fluid (Acar et al., 1995). The net shear force and momentum cause the boundary film of water and free water to move. The thicker the diffuse double layer and the smaller the pore size, the more uniform is the strain field and the farther it extends into the center of the capillary (Acar et al., 1995). Because of the pore size effect, electroosmosis is only significant in low permeable soils.
Electrokinetic techniques can effectively extract a variety of inorganic contaminants, organic contaminants, and radionuclides from soils, especially those with low permeability (Hamnet, 1980; Hamed et al., 1991; Acar et al., 1992, 1995; Bruell et al., 1992; Pamukcu and Wittle, 1992; Lageman, 1993; Shapiro and Probstein, 1993; Will, 1995; Cox et al., 1996; Coletta et al., 1997). In addition, bench-scale studies suggest that electrokinetics may be effective in transporting nutrients and terminal electron acceptors to microbial populations within low permeability biodegradation zones, thus enhancing in situ bioremediation (Segall and Bruell, 1992; Acar et al., 1997; Budhu et al., 1997; Thevanayagam and Rishindran, 1998). In the simplest cases for contaminant removal, the electrodes are inserted into the ground and a voltage subsequently applied. To optimize the efficiency of electroreclamation, enhanced removal techniques have been attempted, such as (i) enhancing the transport of the contaminant with the addition of a reagent or additional process or (ii) conditioning the electrolysis products. One enhanced removal technique is to wrap the electrodes with IEMs in order to prevent the transport of undesirable electrolysis products into the soil being treated (Rosand et al., 1995). Inhibition of the acid front generated at the anode with AEMs (Fig. 2)
, or inhibition of the alkaline front generated at the cathode with CEMs have also been attempted.
|
Bipolar Interface Between Ion Exchange Membranes and Low Permeable Soils: A Hypothesis
|
|---|
On the basis of the well-established existence of ion exchange surfaces in soils, we suggest that when an IEM is placed in contact with a low permeable soil possessing a significant ion exchange capacity of opposite surface charge, the resulting configuration is, in essence, that of a BPM and accelerating water splitting will occur at the IEM and soil interface. When a direct voltage is then applied across the bipolar interface, the ionic current across the IEM is maintained by the counterions in the free pore solution of the low permeable soil. With time, the free pore solution in the low permeable soil moves away from the bipolar interface due to electroosmosis, thus causing an unsaturated zone at the interface. Accelerated water splitting then begins at the interface since there are not enough counterions in contact with the IEM to maintain an ionic current (Fig. 3
and Fig. 4)
.
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 3. Accelerated water splitting at the bipolar itnerface of an anion exchange membrane (AEM) and low permeable soil possessing a significant net cation exchange capacity (CEC). The free pore solution is absent due to the electroosmotic flow towards the cathode, which is caused by the electromigration of cations on the surface of the soil.
|
|
View larger version (30K):
[in this window]
[in a new window]
|
Fig. 4. Accelerated water splitting at the bipolar interface of a cation exchange membrane (CEM) and low permeable soil possesing a significant net anion exchange capacity (AEC). The free pore solution is absent due to the electroosmotic flow towards the anode, which is caused by the electromigration of anions on the surface of the soil.
|
|
In this work, it is hypothesized that (i) accelerated water splitting can take place at the bipolar interface between IEMs and the ion exchange surfaces of low permeable soils and (ii) electroosmosis plays a vital role. To test these hypotheses, an experimental program was designed on the assumption that accelerated water splitting in soil systems could be identified by monitoring of H+ and OH- under appropriate experimental conditions.
|
MATERIALS AND METHODS
|
|---|
Cell Description and Configuration
The acrylic electrochemical testing cell (Fig. 5)
consisted of the anode compartment, anode sandwich, middle compartment, cathode sandwich, and cathode compartment. Each sandwich consisted of three Viton rubber seals, an AEM, and a CEM (Fig. 6)
fastened together and connected to the three main compartments with sixteen nylon screws and wing nuts. The center Viton seals of the sandwiches contained syringe holes for injection of deionized water into the space between the membranes. This space provided an accumulation zone for ions that migrate into the center of the sandwich as a result of electrodialysis. Two Ebonex conductive ceramic electrodes were placed 22.9 cm apart within the anode and cathode compartments, respectively. The electrodes were snugly fit 0.5 cm from the ends of the acrylic-testing cell. The cell configuration used to examine accelerated water splitting at interfaces of IEMs and soils is illustrated in Fig. 7
. Potassium chloride (0.5 M) was placed in the cathode and anode compartments, deionized water in the center of the sandwiches, and soil in the middle compartment [a 0.5-M concentration of KCl will not compromise the Donnan exclusion principle within the IEMs (Overbeek, 1956)]. Accelerated water splitting may occur at the AEM and soil interface or the CEM and soil interface. If accelerated water splitting occurs at the AEM and soil interface, hydroxide ions will migrate through the AEM nearest the anode and hydrogen ions will migrate across the soil towards the cathode. If accelerated water splitting occurs at the CEM and soil interface, hydrogen ions will migrate through the CEM nearest the cathode and hydroxide ions will migrate across the soil towards the anode. To effectively observe the phenomenon, the entrance of electrolysis products into the soil and ion exchange sandwiches must be inhibited, thus isolating the effects of the accelerated water splitting phenomenon at the interfaces. Isolation of the water splitting products was important in order (i) to inhibit the neutralization of the water splitting products by the electrolysis products and (ii) to verify that the acid and alkaline fronts in the soil were not the result of electrolysis products leaking through the IEMs.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 7. Cell configuration. AEM = anion exchange membrane; CEM = cation exchange membrane; DW = deioinized water.
|
|
Raipore Ion Exchange Membranes
Raipore R4010 CEMs and Raipore R4030 AEMs (Pall-Gelman Sciences, Inc., Ann Arbor, MI) were utilized in this study (Table 1). Raipore IEMs are synthesized by a radiation grafting technique that grafts ion exchange functional groups to a polytetrafluoroethylene film.
Media
Experimental media (Table 2) included Ottawa sand (Soiltest, Inc., Pelham, AL), bentonite clay (Avocado Research Chemicals, Ltd., Heysham, UK), talc clay (Fisher Scientific, Hanover Park, IL), and IONAC A-554 (Polyvinylbenzyldimethylethanol ammonium chloride) anion exchange resin containing dopant chloride ions (Mallinckrodt Baker, Inc., Phillipsburg, NJ), selected for their wide range of ion exchange capacities and particle size ranges. IONAC anion exchange resin was used to represent a soil with high AEC. The bentonite was primarily in the sodium form and exhibited high swelling characteristics.
Experimental Procedure
IEMs were soaked in deionized water for 24 h prior to an experimental run to ensure complete hydration and expansion. The porous medium under investigation was saturated with 1.0 M KCl, mixed for 30 min, and allowed to sit for at least 3 h. This incubation period prior to any experimental run was especially important for media containing swelling bentonite. Saturating with KCl provided an ample supply of counterions to maintain the ionic current across an IEM in the event that free pore solution was present at the bipolar IEM and soil interface. According to our hypothesis, accelerated water splitting would occur at the interface if there were not an ample supply of counterions, due to the electroosmotic flow away from the bipolar IEM and media interface. The experimental cell was assembled (Fig. 8)
with care that the membranes did not dry out. Deionized water (14 mL) was injected into the membrane sandwiches, and the anolyte and catholyte (300 mL of 0.5 M KCl) placed in their respective compartments. About 200 cm3 of the saturated porous medium was placed in the middle compartment of the test cell in layers 1 to 2 cm thick, each layer compressed with the blunt side of a spatula before adding the next layer. Bulk material in the compartment was placed to cover the top of the membrane by 0.5 cm, such that no part of the membrane was exposed. This was done to minimize short-circuiting of the ionic current over the free surface of the specimen.
A constant current of 300 mA was then applied to the test cell with a Hewlett Packard DC power supply (MPB-3 Series, Model 6299A, Hewlett Packard, Palo Alto, CA). The constant current density across the soil and IEMs was 10.5 mA cm-2 for all runs. At timed intervals, the voltage was measured with an Extech digital data acquisition multimeter with a PC interface (Fig. 8). The voltage histories were similar for all runs. After reaching a maximum applied voltage of 20 to 52 V at 1 min, the applied voltage approached a relatively constant value ranging from 10 to 25 V after 3 min. The average voltage gradients ranged from 1.6 V cm-1 to 3.8 V cm-1 (which corresponds to the normal range for electroreclamation of contaminated soils), based on the assumption that the voltage applied between the electrodes was primarily distributed across the soil (i.e., the resistance of the catholyte, anolyte, and membrane sandwiches were very low). This assumption was checked by measurements made during the experimental runs.
The constant current was terminated after 0.5 h. This period was selected for two reasons: (i) preliminary results suggested that if water splitting occurred, it would begin well before 0.5 h, and (ii) in order to avoid the neutralization of the water splitting products in the IEM sandwiches from the electrolysis products, it was necessary to have a low percentage ratio of the delivered charge to the initial number of moles of K+ or Cl- in the anolyte and catholyte. This ratio was (5.60 x 10-3 mol. of delivered electron)/(1.50 x 10-1 mol. of K+ or Cl-) or 3.7% for the selected test conditions at 0.5 h. Since this ratio was low, the primary ions entering the anode sandwich from the anolyte were potassium ions, and the primary ions entering the cathode sandwich from the catholyte were chloride ions. Towards the end of the run, the secondary ions entering the anode sandwich from the anolyte were hydrogen ions as the pH of the anolyte significantly decreased. Similarly, the secondary ions entering the cathode sandwich from the catholyte were hydroxide ions as the pH of the catholyte significantly increased. Overall, the effect of the electrolysis products on the final pH in the sandwich solutions was minimized by the 0.5 M KCl solutions.
Upon completion of each experiment, three samples (each approximately a 2.3-cm thick section) of the porous medium were quickly taken from the test cell. Immediately after, solutions from the various compartments were extracted. If necessary, the soil samples and solutions were cooled to room temperature, as the temperature of the media, especially those containing clay, increased by 5°C during the runs. The pH values were measured by inserting a combination pH electrode (Orion pH and ISE meter, Model 290A, Orion Research, Inc., Boston) directly into the solutions and saturated media samples. Additional details of the experimental procedures can be found in Desharnais (2000).
|
RESULTS AND DISCUSSION
|
|---|
The experimental program was designed to investigate accelerated water splitting at various IEM interfaces in contact with porous media possessing different cation exchange capacities, anion exchange capacities, and pore sizes; as indicated above, electroosmosis is expected to occur to a greater extent in porous media with smaller pore sizes. Mercury intrusion porosimetry of separate dried samples of the media confirmed what was expected, that is, that the pore sizes of the various mixtures significantly decreased with an increasing percentage of clay (either talc or bentonite) in the mixtures. The greatest reduction in pore size occurred in pores of the largest diameters (for example, Fig. 9)
. The mercury porosimetry data is used here as qualitative comparison only, since the saturated media in the test cell were compacted for the tests, while the samples taken for porosimetry were not.
Replicated measurements of pH values of the solutions in the IEM sandwiches (Fig. 10 through 13)
were in good agreement as indicated by the small size of the error bars (95% confidence limits). Test media with net CEC included the (i) bentonite and Ottawa sand mixtures, and (ii) talc and Ottawa sand mixtures. Tests with bentonite and Ottawa sand mixtures (Fig. 10) and talc and Ottawa sand mixtures (Fig. 11) exhibited high pH values within the solution of the anode sandwich for mixtures containing >1.5% bentonite (CEC 5 cmolc kg-1 and AEC 0 cmolc kg-1) and 1.5% talc (CEC 3 cmolc kg-1 and AEC 0 cmolc kg-1), respectively. These results can be explained by water splitting at the media and AEM interfaces, with subsequent transport of hydroxide ions across the AEM of the anode sandwich and migration of hydrogen ions across the soil towards the cathode. Water splitting was absent at the nonbipolar CEM and soil interfaces, a result consistent with the hypothesis that water splitting will occur only at bipolar interfaces. Water splitting was also absent with 100% Ottawa sand, a result consistent with the hypothesis that electroosmosis is a key factor to water splitting at bipolar interfaces (i.e., electroosmosis in sand-sized media is negligible). Electroosmosis was observed with both the bentonite and Ottawa sand mixtures and talc and Ottawa sand mixtures. Evidence of electroosmosis included (i) the movement of a drying front in the mixtures originating from the anode sandwich towards the cathode, which was visible through the transparent body of the acrylic testing cell, (ii) the simultaneous pooling of water on top of the mixtures in the middle compartment due to the electroosmotic flow having nowhere to go in the middle compartment, and (iii) the lower moisture contents and densification of the mixture sections closest to the anode sandwich upon completion of the runs. The pH values of the solution within the cathode sandwich for all mixtures were consistently high due to the minor transport of hydroxide ions through the AEM from the catholyte while the pH values of the solution within the anode sandwich of the 100% Ottawa sand runs were consistently low due to the minor transport of hydrogen ions through the CEM from the anolyte.
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 10. pH values in the ion exchange membrane sandwiches for mixtures of bentonite clay and Ottawa sand.
|
|
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 13. pH values in the ion exchange membrane sandwiches for mixtures of IONAC, talc, and Ottawa sand.
|
|
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 12. pH values in the ion exchange membrane sandwiches for mixtures of IONAC resin and Ottawa sand.
|
|
Test media with net AEC included the (i) IONAC resin and Ottawa sand mixtures and (ii) IONAC resin and Ottawa sand mixtures with 10% talc. Tests with IONAC resin and Ottawa sand (Fig. 12) exhibited high pH values within the solution of the cathode sandwich and low pH values within the solution of the anode sandwich for all mixtures. This can be explained by the absence of water splitting at all IEM and soil interfaces. Water splitting was absent at the nonbipolar AEM and soil interfaces, a result again consistent with the hypothesis that water splitting will only occur at bipolar interfaces. Water splitting was absent at the CEM and soil interfaces of the coarse-grained sand and IONAC mixtures, a result consistent with the hypothesis that electroosmosis is a key factor to water splitting at bipolar interfaces (i.e., electroosmosis in sand-sized media is negligible; therefore, electroosmosis should not occur in the sand and IONAC mixtures). Also, evidence of electroosmosis was not observable with the IONAC and Ottawa sand mixtures. The pH values of the solution within the cathode sandwich were consistently high due to the minor transport of hydroxide ions through the AEM from the catholyte while the pH values of the solution within the anode sandwich were consistently low due to minor transport of hydrogen ions through the CEM from the anolyte.
Tests with IONAC resin and Ottawa sand mixtures with 10% talc (Fig. 13) exhibited high pH values within the solution of the anode sandwich for mixtures containing <25% IONAC resin (CEC 2 cmolc kg-1 and AEC 83 cmolc kg-1). This can be explained by water splitting at the media and AEM interface, with subsequent transport of hydroxide ions across the AEM of the anode sandwich and migration of hydrogen ions across the soil towards the cathode. Water splitting was absent at the AEM and soil interface for mixtures containing >25% IONAC resin. Low pH values within the solution of the cathode sandwich were observed for mixtures containing >5% IONAC resin (CEC = 3 cmolc kg-1 and AEC = 17 cmolc kg-1). This can be explained by water splitting at the media and CEM interface, with subsequent transport of hydrogen ions across the CEM of the cathode sandwich and migration of hydroxide ions across the soil towards the anode. The presence of water splitting at both the AEM and soil interface and CEM and soil interface for mixtures containing <25% IONAC resin demonstrates the existence of a transition whereby localized areas in the soil contain both a net AEC and CEC. Although the mixtures containing 5 to 25% IONAC had greater net anion exchange capacities as a whole than cation exchange capacities, there may have been some localized areas in direct contact with the AEM that still had appreciable net CEC, which explains why accelerated water splitting was observable at both IEM and media interfaces. As more IONAC was added, the chances of there being localized areas in contact with the AEM with appreciable net CEC decreased, which explains the decreasing trend of the pH values in the anode sandwiches (Fig. 13). Similar to the aforementioned results with low permeable medias containing net cation exchange capacities, evidence of electroosmosis was mainly present at the CEM and media interface for IONAC resin and Ottawa sand mixtures with 10% talc. Water splitting was absent at the nonbipolar CEM and soil interface for the mixtures containing 0% IONAC resin (CEC 3 cmolc kg-1 and AEC 0 cmolc kg-1), a result consistent with the hypothesis that water splitting will only occur at bipolar interfaces, evidence for this being the relatively high pH values within the cathode sandwich due to the minor transport of hydroxide ions through the AEM from the catholyte.
Although the beginnings of migrating acid and base fronts were observed at interfaces where water splitting was present, acid or alkaline fronts were not sufficiently developed after 0.5 h. If they had been sufficiently developed, the acid or base front would be very noticeable. For example, Fig. 14
demonstrates the migration of the acid front after 3 and 6 h due to accelerated water splitting at the interface of an AEM and low permeable mixture comprised of 10% bentonite and 90% Ottawa sand (CEC 13 cmolc kg-1 and AEC 0 cmolc kg-1). The pH within the solution of the anode sandwich was high, and a migrating acid front developed across the soil specimen (i.e., pH values were much lower within soil sections closer to the anode side). This can be explained by water splitting at the AEM and media interface (i.e., at the interface of the anode sandwich and soil section 1), with subsequent transport of hydroxide ions across the AEM of the anode sandwich and migration of hydrogen ions across the soil towards the cathode. Accelerated water splitting was absent at the CEM and soil interface, otherwise, the pH of the solution within the cathode sandwich would have been low. The pH of the anolyte was low due to the oxidation of water into hydrogen ions and oxygen gas at the anode. The pH of the catholyte was high due to the reduction of water into hydroxide ions and hydrogen gas at the cathode.
View larger version (36K):
[in this window]
[in a new window]
|
Fig. 14. pH profiles for long-term experiments with a low permeable mixture comprised of 10% bentonite and 90% Ottawa sand.
|
|
The experimental results described above indicate that very little CEC, AEC, or clay is needed for water splitting to occur at a bipolar IEM and soil interface. The phenomenon can be exhibited at any IEM and media interface if a low permeable soil possesses both a significant cation and AEC.
Extension of these findings to the enhanced electroreclamation of contaminated soils points to possible modifications of some current practices in the field. For example, the encapsulation of anodes with AEMs and cathodes with CEMs may not be feasible in low permeable soils possessing significant net cation exchange capacities and anion exchange capacities, respectively. These two IEM and soil interfaces are bipolar and accelerated water splitting will occur, thus negating the intended purpose of the IEMs.
Received for publication June 26, 2001.
|
REFERENCES
|
|---|
- Acar, Y.B., and A.N. Alshawabkeh. 1993. Principles of electrokinetic remediation. Environ. Sci. Technol. 27:2638–2647.
- Acar, Y.B., R.J. Gale, A.N. Alshawabkeh, R.E. Marks, S. Puppala, M. Bricka, and R. Parker. 1995. Electrokinetic remediation: Basics and technology status. J. Hazard. Mater. 40:117–137.
- Acar, Y.B., H. Li, and R.J. Gale. 1992. Phenol removal from kaolinite by electrokinetics. J. Geotech. Eng. 118:1837–1852.
- Acar, Y.B., F. Rabbi, and E. Ozsu. 1997. Electrokinetic injection of ammonium and sulfate ions into sand and kaolinite beds. J. Geotech. Geoenviron. Eng. 123:239–249.
- Alshawabkeh, A.N., and Y.B. Acar. 1992. Removal of contaminants from soils by electrokinetics: A theoretical treatise. J. Environ. Sci. Health 27:1835–1861.
- Alshawabkeh, A.N., and Y.B. Acar. 1996. Electrokinetic remediation. II: Theoretical model. J. Geotech. Eng. 122:186–196.
- Black, C.A., et al. (ed.) 1965. Methods of soil analysis. Agron. Monogr. 9. ASA, Madison, WI.
- Bruell, C.J., B.A. Segall, and M.T. Walsh. 1992. Electroosmotic removal of gasoline hydrocarbons and TCE from clay. J. Environ. Eng. 118:68–83.
- Budhu, M., M. Rutherford, G. Sills, and W. Rasmussen. 1997. Transport of nitrates through clay using electrokinetics. J. Environ. Eng. 123:1251–1253.
- Cabrera-Guzman, D., J.T. Swartzbaugh, and A.W. Weisman. 1990. The use of electrokinetics for hazardous waste site remediation. J. Air Waste Manage. Assoc. 50:1670–1676.
- Casagrande, L. 1949. Electro-osmosis in soils. Geotechnique 1:159–174.
- Casagrande, L. 1952. Electroosmotic stabilization of soils. J. Boston Soc. Civil Eng. 39:51–83.
- Coletta, T.F., C.J. Bruell, D.K. Ryan, and H.I. Inyang. 1997. Cation-enhanced removal of lead from kaolinite by electrokinetics. J. Environ. Eng. 123:1227–1233.
- Cox, C.D., M.A. Shoesmith, and M.M. Ghosh. 1996. Electrokinetic remediation of mercury-contaminated soils using iodine/iodide lixiviant. Environ. Sci. Technol. 30:1933–1938.
- Denisov, G., R.E. Hicks, and R.F. Probstein. 1996. On the kinetics of charged contaminant removal from soils using electric fields. J. Colloid Interface Sci. 178:309–323.
- Desharnais, B.M. 2000. Electrokinetic sampling of ionic contaminants in soils. Ph.D. diss. Northwestern Univ., Evanston, IL.
- Dzenitis, J.M. 1997. Soil chemistry effects and flow prediction in electroremediation of soil. Environ. Sci. Technol. 31:1191–1197.
- Gray, D.H., and J.K. Mitchell. 1967. Fundamental aspects of electro-osmosis in soils. J. Soil Mech. Found. Div., Am. Soc. Civ. Eng. 93:209–236.
- Hamed, J., Y.B. Acar, and R.J. Gale. 1991. Pb(II) removal from kaolinite by electrokinetics. J. Geotech. Eng. 117:241–271.
- Hamnet, R. 1980. A study of the processes involved in the electro-reclamation of contaminated soils. Dissertation. Univ. of Manchester, Manchester, UK.
- Johnsson, G., and C.E. Boesen. 1984. Polarization phenomena in membrane processes. p. 125–148. In G. Belfort (ed.) Synthetic membrane processes: Fundamentals and water applications. Academic Press, Orlando, FL.
- Jorissen, J., and K.H. Simmrock. 1991. The behaviour of ion exchange membranes in electrolysis and electrodialysis of sodium sulphate. J. Appl. Electrochem. 21:869–876.
- Lageman, R. 1993. Electroreclamation: Applications in the Netherlands. Environ. Sci. Technol. 27:2648–2650.
- Lageman, R., W. Pool, and G. Seffinga. 1989. Electro-reclamation: Theory and practice. Chem. Ind. (London) 18 Sept.:585–590.
- Overbeek, J.T. 1956. The Donnan equilibrium. Prog. Biophys. 6:57–84.
- Pamukcu, S., and J.K. Wittle. 1992. Electrokinetic removal of selected heavy metals from soil. Environ. Prog. 11:241–250.
- Ramirez, P., H.J. Rapp, S. Reichle, H. Strathmann, and S. Mafe. 1992. Current-voltage curves of bipolar membranes. J. Appl. Physics 72:259–264.
- Rosand, T., Y.B. Acar, and G. Breedveld. 1995. Electrokinetic extraction of lead from spiked Norwegian marine clay. p. 1518–1534. Y.B. Acar and D.E. Daniel (ed.) Geoenvironment 2000: Characterization, containment, remediation, and performance in environmental geotechnics. American Society of Civil Engineers, New York.
- Segall, B.A., C.E. O'Bannon, and J.A. Matthias. 1980. Electro-osmosis chemistry and water quality. J. Geotech. Eng. Div. 106:1148–1152.
- Segall, B.A., and C.J. Bruell. 1992. Electroosmotic contaminant-removal processes. J. Environ. Eng. 118:84–100.
- Shapiro, A.P., and R.F. Probstein. 1993. Removal of contaminants from saturated clay by electroosmosis. Environ. Sci. Technol. 27:283–291.
- Simons, R. 1979. The origin and elimination of water splitting in ion exchange membranes during water demineralization by electrodialysis. Desalination 28:41–42.
- Simons, R. 1985. Water splitting in ion exchange membranes. Electrochim. Acta 30:275–282.
- Strathmann, H., B. Bauer, and H.J. Rapp. 1993. Better bipolar membranes. Chemtech 23:17–24.
- Thevanayagam, S., and T. Rishindran. 1998. Injection of nutrients and TEAs in clayey soils using electrokinetics. J. Geotech. Geoenviron. Eng. 124:330–338.
- West, L.J., and D.I. Stewart. 1995. Effect of zeta potential on soil electrokinesis. p. 1535–1549. In Y.B. Acar and D.E. Daniel (ed.) Geonvironment 2000: Characterization, containment, remediation, and performance in environmental geotechnics. American Society of Civil Engineers, New York.
- Will, F.G. 1995. Removal of toxic substances from the soil using electrochemistry. Chem. Ind. (London) 15 May:376–379.
TOP
ABSTRACT
INTRODUCTION
