SSSAJ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Web of Science (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Aydin, M.
Right arrow Articles by Kilic, S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Aydin, M.
Right arrow Articles by Kilic, S.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Aydin, M.
Right arrow Articles by Kilic, S.
Related Collections
Right arrow Hydraulic Conductivity
Right arrow Soil Surface Chemistry
Right arrow Experiment Design
Published in Soil Sci. Soc. Am. J. 68:450-459 (2004).
© 2004 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA

DIVISION S-2—SOIL CHEMISTRY

Dependence of Zeta Potential and Soil Hydraulic Conductivity on Adsorbed Cation and Aqueous Phase Properties

Mehmet Aydin*,a, Tomohisa Yanob and Seref Kilica

a Faculty of Agriculture, Dep. of Soil Science, Mustafa Kemal Univ., Antakya, 31040 Turkey
b Arid Land Research Center, Tottori Univ, 1390 Hamasaka, Tottori, 680-0001 Japan

* Corresponding author (maydin{at}mku.edu.tr).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
This paper deals with the effects of pH, electrolyte concentration, and exchangeable Na percentage (ESP) on electrophoretic mobility (EM)/zeta potential ({zeta}p) of clay particles and hydraulic conductivity (HC) of the clay/sand mixtures. The soils taken from Japan and Kazakhstan were used for obtaining clay fractions. For EM determinations, clay suspensions were prepared at a concentration of 4 g of clay per 100 mL of distilled water (DW) or electrolyte solution. The electrophoretic mobilities were measured using Burton apparatus with water-cooling system and converted into {zeta}p. For HC measurements, clay/sand mixtures were designated 16:84. Columns of these mixtures were prepared by packing 120 g of mixtures into 50-mm diameter plastic cylinders to a bulk density of about 1.4 g cm–3. Results showed clearly that the mobility was very sensitive to the ion valence adsorbed on the clay. The negative electrophoretic mobilities of homoionic Na– Ariake soil (AS) clay and Na–Kzyl-Orda soil (KS) clay were 2.13 x 10–8 and 2.14 x 10–8 m2 s–1 V–1, respectively, whereas Ca clays flocculated. The {zeta}p values of AS clay and KS clay, as a function of the ESP, varied between –12.83 and –26.84 mV, and –5.68 and –27.00 mV, respectively, at ESP > 30. Although the smectitic AS clay was less sensitive than the micaceous KS clay to pH changes during electrophoresis experiment, its HC also was affected by pH changes. Decreased pH from 7 to 5 could easily result in two to three times high HC values for both clay/sand mixtures. The EM of both soil clays was similar at pH 10 to 12, and exchangeable Na percentages 90 to 100. Sharp increases in EM and decreases in HC of AS clay were observed at exchangeable Na percentages 50 and 60, respectively. Similar trends related to EM were also obtained for the KS clay. However, salt concentration of the suspension solution did not have consistent effect on the EM values. This behavior of the clays was consistent with HC observation. The results indicated that HC of the clay/sand mixtures could be correlated to {zeta}p. The saturated HC of the mixtures was found to change as an exponential function of the {zeta}p of clay particles.

Abbreviations: AS, Ariake soil • DW, distilled water • EC, electrical conductivity • EM, electrophoretic mobility • ESP, exchangeable sodium percentage • HC, hydraulic conductivity • KS, Kzyl-Orda soil • XRD, X-ray diffraction • {zeta}p, zeta potential


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE INTERACTION OF clay surface with ions has great influence on physicochemical behavior of soils. The types and concentration of ions in soil solution govern the dominance of attracting and repelling forces and the resulting flocculation or deflocculation of clays. Divalent exchangeable cations result in flocculated clay systems while monovalent exchangeable cations produce dispersed systems. In other words, by replacing the Na adsorbed on the clay with Ca, the thickness of the electrical double layer is reduced and thus soil colloids flocculate. On the other hand, numerous studies have been conducted to examine the effect of pH on clay dispersion–flocculation (Chorover and Sposito, 1995; Goldberg and Forster, 1990; Kretzschmar et al., 1997; Lebron and Suarez, 1992; Miller et al., 1990).

In many practical situations the value of the {zeta}p obtained from electrophoresis experiments can be used as an estimate of the diffuse layer potential. It is also valuable in discussing the tendency for the soil colloids to disperse. However, the determination of {zeta}p is indirect and is done via the measurements of EM. Apart from charge density, EM of a particle depends on a number of other factors, such as electrolyte concentration (ionic strength), ion species, pH, dielectric permittivity of the medium, viscosity, temperature, particle size and density, and the shape of the suspended particles. The effects of these factors on the EM and {zeta}p of pure mineral particles and/or soil clays were investigated by many researchers (Anderson and Bertsch, 1993; Cerpa et al., 1999; Dixit, 1982; Kretzschmar et al., 1997; Lebron and Suarez, 1992; Osei and Singh, 1999). For example, van Olphen (1977) reported that the EM of clay particles in water was typically in the range of 1 x 10–8 to 3 x 10–8 (m2 s–1 V–1).

To understand the electrokinetic properties of clay particles, most studies were performed for relatively pure minerals (Bar-on et al., 1970; Vane and Zang, 1997; Yeung et al., 1997). However, soil clay particles often differ appreciably in properties from those of the homogeneous particles. On the other hand, the existing information in electrophoretic migration of clay particles is mostly related to very dilute suspension. In dilute clay colloid suspension, the flat electrical double layer on the charged surface can be extended fully without interaction between two double layers (Jiang et al., 2001). However, in concentrated suspensions the flat clay particles begin to approach each other, and interaction of the double layer on the clay surfaces occurs. Ideally the suspension densities should be as dilute as possible for description of the EM of individual particles. However, it appears that concentrated suspensions may reflect the complex behavior of soil–clay fraction better and the possible interaction between particles. Although concentrated clay suspensions have disadvantage of rapid flocculation, for specific cases the suspension density is dictated by the objectives of investigations.

The {zeta}p determined from EM measurements can also be considered a fundamental property that influences soil permeability. In this regard, Quirk and Schofield (1955) showed that HC of a given soil decreases with increasing ESP provided that the electrolyte concentration is below a critical level. Dispersion and swelling of clays within the soil matrix are interrelated phenomena, and either can reduce soil HC. Swelling reduces soil pore sizes and dispersion clogs soil pores. Different opinions are evident in the literature as to whether swelling or dispersion is the major cause of reduced permeability of sodic soil (Chaudhari, 2001). The dominant process restricting permeability of arid-zone soils is clay dispersion followed by clay migration and plugging of soil pores (Frenkel et al., 1978). Numerous factors have been related to clay dispersion (Goldberg and Forster, 1990). Sodium saturation percentage, the electrolyte composition, and concentration of the soil solution are two most important factors influencing the clay dispersion–related problems of reduced HC and infiltration rate of soils (Shainberg and Letey, 1984). Some studies have indicated that EC-SAR relation may not be sufficient. For any given soil there may be a good relationship, and there is a strong bias to developing simple two-dimensional plots (such as EC–SAR) but this provides a rough guideline at best (Pratt and Suarez, 1990). For example, Suarez et al. (1984) examined the effect of pH on saturated HC of three mineralogically different arid land soil types using laboratory soil columns, and found an adverse impact of pH on HC and flocculation. They also concluded that the assignment of the threshold values for soil permeability must consider pH along with previous concepts of Na adsorption ratio (SAR) and electrolyte concentration. As stated above, the effects of mineralogical composition, organic matter, oxide content, ESP, pH, and electrolyte concentration on soil HC and EM (or {zeta}p) of soil clays are well known. However, there is insufficient quantitative knowledge to link HC with {zeta}p.

The objectives of this study were (i) to demonstrate the effects of solution composition on EM/{zeta}p of two mineralogically different soil-clay fractions and on hydraulic conductivity of the clay/sand mixtures, and (ii) to examine relationship between {zeta}p and HC.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil Samples and Analyses
Two soil samples originating under widely different environmental conditions were used in this study due to mainly their different mineralogical composition. One of the soil samples was taken from a paddy field derived from marine alluvium in Japan (AS). The other sample was obtained from Kzyl-Orda Region in Kazakhstan (KS). The samples were taken from surface layers (0–20 cm). Some physical and chemical analyses were conducted on the samples to link soil properties with measured EM and HC. The analyses conducted were: texture (Bouyoucos, 1962), CaCO3 equivalent (Nelson, 1982), organic matter content (Nelson and Sommers, 1982), cation-exchange capacity (CEC) (Rhoades, 1982), pH in the soil paste, and EC of the saturation extract (U.S. Salinity Laboratory Staff, 1954). The Fe oxide extracted with dithionite-citrate-bicarbonate (Mehra and Jackson, 1960) was determined with an atomic absorption spectrophotometer.

Qualitative and quantitative analyses of clay minerals were undertaken according to Ohtsubo et al. (2002). After removal of carbonates with acid sodium acetate, soil samples suspensions were treated with 7% (wt/wt) H2O2 to remove organic matter, followed by deflocculation by adjusting pH at 10 with 1 M NaOH after sonication. The <2-µm clay fractions were collected from the soil suspensions by siphoning. For X-ray diffraction (XRD), duplicate clay suspensions containing 50 mg of clay were prepared. One was saturated with Mg2+ by washing with 0.5 M MgCl2 and other with K+ by 1 M KCl. Excess salt was removed by washing with water. One cubic centimeter of water was added and an aliquot of the suspension containing 30 mg of clay was dropped onto a glass slide (28 by 48 mm), air-dried, and X-rayed. The K-saturated specimen was heated at 300 and 550°C, and the Mg-saturated specimen was solvated with glycerol, followed by X-raying. Filtered CoK{alpha} or CuK{alpha} radiation from a Rigaku diffractometer (Rigaku Denki Co. Ltd., Tokyo) was used for the XRD. The relative percentage of the clay minerals in clay fractions was estimated based on the peak intensities calculated for the respective diffraction spacing by multiplying the peak height with the peak width at a half of the peak height. The respective equations were formulated according to Ohtsubo et al. (2000)(2002). X-ray diffraction pattern of the <2-µm clay fraction of the AS sample revealed that smectite was the principal mineral, accompanied by mica and kaolinite. The major clay minerals of the KS were mica, kaolinite, and chlorite (Table 1).


View this table:
[in this window]
[in a new window]
  		Table 1. Some physical and chemical properties of the soil samples.{dagger}

 
Separation and Preparation of Stock Clays
A clay-size fraction of soil samples was obtained by allowing large-size fractions to settle out of suspensions and then siphoning off the suspensions. For this purpose, Na-hexametaphosphate (NaHMP) was used as dispersant. Sodium clay and Ca clay were prepared by saturating the colloidal fraction with 0.5 M NaCl and 0.25 M CaCl2 solution, respectively. Thereafter, the clays were washed with ethyl alcohol (successive centrifugal treatments) until the equilibrium solutions were free of Cl as indicated with AgNO3. The salt-free clays were dried at 45°C. These clay fractions saturated with Na and Ca ions were kept as stock clays.

Electrophoresis Experiments
Preparation of Sodium Clay and Calcium Clay Suspensions
For EM determinations, 4 g of air-dried homoionic Na- and Ca-clay fraction was dispersed in 100 mL of DW (pH {cong} 5.7, EC = 1.7 µS cm–1). In addition, a sodium acetate solution and a calcium acetate solution with the same electrical conductivity (EC) as those of Na-clay and Ca-clay suspensions, respectively, were prepared. The electrical conductivities of Na-AS and Na-KS clay suspensions were 0.75 and 0.67 mS cm–1, and those of Ca-AS and Ca-KS clay suspensions were 0.32 and 0.49 mS cm–1, respectively. The bi-ionic (Na-Ca) clay suspensions were prepared by mixing the two homoionic suspensions with the fraction of 0, 10, 20, 30, 40, 50, 60, 70, 80, 90, and 100% by volume of Na-clay suspensions. To ensure equilibrium, the electrophoresis experiments were performed at least 24 h after mixing the appropriate homoionic suspensions (Bar-on et al., 1970).

Preparation of Clay Suspensions at Different pHs
Some amount of stock Na clay was washed by 0.01 M NaCl until the conductivity of supernatant was about equal to that of the washing solution. Finally, 4 g of clay fraction was suspended in 100 mL of 0.01 M NaCl solution. The pH of suspensions was varied, in the range 2 to 12, by drop-wise addition of HCl or NaOH as needed (Cerpa et al., 1999).

Preparation of Clay Suspension at Different Ionic Strengths
As mentioned previously, pretreated Na clays were dispersed in 0.001, 0.01, and 0.1 M NaCl solutions. The samples were shaken then centrifuged at 1190 x g (3000 rpm), and supernatants were decanted. These procedures were repeated until ECs of supernatants were close to those of treating solutions. All suspensions were prepared at a concentration of 4 g of clay per 100 mL of treating solution for electrophoresis processes.

Experimental Apparatus and Electrophoretic Mobility Measurements
Electrophoresis experiments were performed using an apparatus similar to Burton tool (Janse and Bolt, 1958; Yesilsoy et al., 1983).

Commercially available power supplies for electrophoresis can be operated under constant-voltage or constant-current conditions. In general, the constant-current mode is preferred if thermal effects are of concern. Under constant-current conditions, however, the system voltage is not an experimentally available parameter; thus, a constant-voltage is usually applied to obtain constant field strength. While the electrical potential across the electrodes was maintained to be constant, the electrical current passing through the system increased with time, and resulted in increases in temperature. These increases cause turbulence in clay suspensions during the electrophoresis processes. Therefore, the Burton apparatus with a water-cooling system was fabricated at the glass factory by us. The glass U-tube was surrounded with an outside tube for cold water circulation to minimize the temperature increases during the process. Direct current (Kenwood PA500-0.1 A regulated DC, Kenwood, Japan) power supply was connected via platinum wire electrodes to the U-tube. A schematic diagram of the apparatus described herein is presented in Fig. 1 .



View larger version (30K):
[in this window]
[in a new window]
  		Fig. 1. Schematic diagram of electrophoresis apparatus (U-tube).

 
A long plastic tube was filled with acetate solution then connected to the stopcock at the bottom of U-tube (sodium acetate and calcium acetate solutions were used for Na-clay and Ca-clay experiments, respectively). With the stopcock open, about 15 mL of the solution was introduced into the U-tube. The stopcock was closed when the solution reached a calibration mark previously scratched on the one leg of the U-tube. After that, the plastic tube filled with relevant clay suspension was connected to the same stopcock. This tube was raised, and then, the stopcock was slowly opened to inlet clay suspension (about 10 mL) into the U-tube, and to force acetate solution upward through the legs of the U-tube. This operation is very important to obtain a sharp interface between solution and suspension. After closing the stopcock the initial position of the solution–suspension interface in the U-tube was recorded, and the apparatus was ready for electrophoresis operation.

A constant electrical potential of 200 V was applied via platinum wire electrodes immersed in the acetate solution at the upper end of the solution. Displacement of suspension boundary was measured every 5 min. During the processes four readings (in millimeters) were taken, and the system was closed. All the experiments were performed at constant field strength of 9 V cm–1, and at room temperature. At the beginning and end of electrophoresis processes, the temperature of suspension and/or solution was measured. During the operation, the increases in temperature were minimized by means of cold water circulation through outside glass tube. Each treatment was replicated three times.

Electrophoretic mobility is defined as the ratio of the terminal velocity of the particle, {nu}, to the applied electric field, F: EM = {nu}/F. The mobility, EM, was converted to {zeta}p by using the Smoluchowski equation: {zeta}p = EM{eta}/, where {eta} and are the viscosity and permittivity of the medium, respectively (Ohshima, 1999; Vane and Zang, 1997).

Hydraulic Conductivity Studies
Determination of Saturated Hydraulic Conductivity of Clay-Sand Mixtures with Different Exchangeable Sodium Percentages
The bi-ionic clay mixtures were prepared by mixing the appropriate amounts of the two stock clays (Na clay and Ca clay) to obtain the following ESP: 0, 20, 40, 60, 80, and 100. Some amount of quartz sand (0.1- to 0.6-mm diam.) was first divided into two subsamples. Thereafter, one of the subsamples was leached with 0.5 M NaCl. The other was leached with 0.25 M CaCl2. Subsequently, the both samples were washed three times with distilled water. The proper proportions of Na- and Ca-treated quartz samples were also mixed to obtain ESP of 0, 20, 40, 60, 80, and 100. After then, clay/sand mixtures with the same ESP containing 160 g of clay mixed with 840 g quartz sand were prepared. Columns of these mixtures were prepared by packing 120 g of mixtures into 50-mm diameters plastic cylinders to a bulk density of about 1.4 g cm–3. A rubber stopper with a hole to accommodate an outflow tube provided button support for the mixture columns. Cheesecloth covered with a 5-mm layer of sand served as a filter (Regea et al., 1997). Three replications of the columns for each ESP were prepared. All columns were slowly wetted (to saturation) from bottom with DW, and kept saturated long enough for Na and Ca clay to be in equilibrium with each other. Then the DW was applied from above using a constant head device (height 20 cm for ESP ≤ 40, and 35 cm for other ESP experiments). The constant head was maintained by using Marriotte Bottle systems. The effluent was collected using a fraction collector, and the drainage rate was calculated. The amount of suspended clay was determined by gravimetric procedure. The saturated HC determinations were made after steady-state flow was attained.

Determination of Saturated Hydraulic Conductivity Using 0.01 M NaCl Solution with Different pHs
Stock Na-clay/sand mixtures were prepared. This mixture was designated 16:84. The pH of solution was adjusted, in the range 3 to 9, by drop-wise addition of HCl and NaOH. The columns were slowly wetted from bellow by capillary rise and then the same solution was applied from above (with a constant head of 35 cm). The HC measurements were taken when the flow rate and pH of the effluent had stabilized (Suarez et al., 1984).

Determination of Saturated Hydraulic Conductivity under Different Electrolyte Concentrations
The pretreated Na-clay/sand mixture columns were initially prewetted from bottom with a 0.1 M solution of NaCl. Once saturated, the columns were leached from the top with the same solution (the applied hydraulic head was 35 cm). The columns were then leached with solutions of successively decreasing salt concentration (0.01 and 0.001 M NaCl) and finally with DW until new steady-state HC and effluent EC were achieved (Frenkel et al., 1978; Regea et al., 1997).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Electrophoretic Mobility and Zeta Potential
In this study, clay particles of both soil samples exhibited negative EMs regardless of changes in pHs and ionic strengths, that is, they migrated in the direction of increasing electrical potential. The absolute values of EMs are being used in the text to facilitate the presentation. Decreases and increases in mobility are given here in terms of the absolute value. The standard deviation of triplicate EM measurements was <10–9 m2 s–1 V–1. It should also be noted that the effect of gravitational forces on the contramigration of clay particles is not clear in this present study, because EM is usually measured horizontally to eliminate the influence of the gravitational field. However, as pointed out by Yeung et al. (1997), the electric field effects appear to be much greater than those of gravity.

Effect of Sodium and Calcium Proportion
The EM and corresponding {zeta}p as a function of the ESP are presented in Fig. 2 . The EM and {zeta}p of clays saturated with Na and Ca ions in different proportions show clearly that these properties are very sensitive to the valence of the ion adsorbed on the clay. The negative EMs of pure Na-AS and Na-KS clay were 2.13 x 10–8 and 2.14 x 10–8 m2 s–1 V–1, respectively, whereas Ca clays flocculated. In a system containing Ca as the only cation, the thickness of diffuse double layer is much smaller than for mixed Ca–Na systems (Lebron and Suarez, 1992). Yesilsoy et al. (1983) also reported the flocculation of Ca-clay suspensions during the electrophoresis processes. On the other hand, the suspensions densities used in our study might contribute to the rapid flocculation.



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 2. Variation of electrophoretic mobility (EM) and corresponding zeta potential ({zeta}p ) of Ariake soil (AS) and Kzyl-Orda soil (KS) clays as a function of Na percentage (the complementary ion was Ca).

 
Because of the preparation method, presumably Al might become an appreciable and exchangeable cation on hydrolysis of the exchangeable cations. This is probably mostly compensated by the subsequent re-equilibration with the solution used to measure EM but not all the Al may readily exchange. Suarez and Frenkel (1981) also found substantial hydrolysis of Na-saturated kaolinite during the conventional preparation procedure of successive washings to remove excess salt. In addition, a partial conversion of Na clay into Ca clay is possible, because a substantial quantity of calcium carbonate was present in both soils. Under dilute conditions, Na in solution is low thus Ca solubilized from calcite might cause appreciable exchange and partial formation of Ca clay from the initial pure Na clay. It is possible but unlikely that all the calcium carbonate dissolved during preparation of the homoionic clays. The electrophoretic mobilities of particles at ESP ≤ 30 could not be determined because of the problems with flocculation. When the ESP approached 40%, clay particles moved toward the anode under the influence of the applied electric field. With further addition of Na into the exchange complex, the electrophoretic mobilities of both soil clays increased very rapidly. When the Na proportion reached a value of 70%, the EM of AS clay was already identical to that of pure Na clay; whereas this threshold for KS clay was 90%. However, Bar-on et al. (1970) reported that a slight addition of exchangeable Na to Ca-saturated montmorillonite had a considerable effect on the EM of the particles. They also found that the maximum EM equal to that of the homoionic Na clay was reached at ESP of 35. These discrepancies may be due to differences in mineral composition (clay types) and clay particle concentration in the suspension, as well as possible differences in total electrolyte concentrations. Additionally, the presence of a mixed Na-Ca clay rather than Na clay would cause the Ca fraction of the subsequent Na-Ca mixtures to be greater than assumed. This may also provide an explanation as to why Bar-on et al. (1970) found that the maximum EM occurred around ESP 35 rather than ESP 70 and 90 in this present study. Lebron and Suarez (1992) looked at micaceous clay and also found a rapid change in EM around ESP 15 to 30. The fact that different soils may exhibit different Na saturation for clay dispersion is usually attributed to the effects of clay mineral type and content (Frenkel et al., 1992). For a more reasonable direct comparison, many studies were performed for reference clay minerals. However, using reference clay minerals as models for electrophoretic behavior of field soils may not be appropriate. Arora and Coleman (1979) concluded that the montmorillonitic soil acted like reference montmorillonite and micaceous soil clays, which behaved like reference illite and vermiculite and therefore, flocculation–dispersion behavior of the soil clay fractions was influenced by their clay mineralogy and could be related to the behavior of reference clays. Unfortunately, their data do not allow such a clear-cut conclusion (Goldberg and Forster, 1990). Thus, a reasonable direct comparison of the EM of heterogeneous soil clays is difficult. Our micaceous KS clay was sensitive to changes in sodium percentage for ESP between 30 and 90, while smectitic AS clay was sensitive to Na change at ESP 30 to 70. The results shown in Fig. 2 indicate that the EMs of AS clay were higher than those of KS clay at the same ESP except homoionic Na clay. Similarly, the {zeta}p values of AS clay and KS clay, as a function of the ESP, varied between –12.83 and –26.84 mV and –5.68 and –27.00 mV, respectively, at ESP > 30. The results indicated no differences in EM of homoionic Na-AS clay and Na-KS clay. Thus, further experiments were conducted only with Na-saturated clay.

Effect of pH
The influence of pH on EM and {zeta}p of Na-saturated AS clay and KS clay in 0.01 M NaCl-treated suspensions is shown in Fig. 3 . While clay particles still moved toward the anode, their EM decreased with a decrease in pH value.



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 3. Effect of pH on electrophoretic mobility (EM) and corresponding zeta potential ({zeta}p) of Ariake soil (AS) and Kzyl-Orda soil (KS) clay particles in 0.01 M NaCl-treated suspensions.

 
Negative EM of KS clay was found to be a strong function of pH between 4 and 11. There was another important fact that the colloidal suspensions of KS clay flocculated at low pHs. At pH 2 and 3, the EM of particles could not be determined due to rapid flocculation. It is also apparent from Fig. 3 that the behavior of AS clay in response to pH changes is different from that of KS clay. In other words, the smectitic AS-clay was considerably less sensitive than the micaceous KS-clay between pH 5 and 10, as indicated by the lower slope of the curve. The changes in the mobility were only observed at a pH range of 5 to 8. Moreover, the EM of this clay was also typical of particles that carried mainly negative charges, and there was only a slight decrease in the mobility in response to pH changes. At pHs between 2 and 4, the suspensions flocculated thus the EM could not be measured (data not plotted). Over most of pH range, EM of AS clay was found to be more negative than that of KS clay. We attribute the difference to the mineral composition: KS containing large amount of mica, kaolinite, and chlorite has more pH-dependent charge than that of AS containing smectite as dominant mineral (Table 1). Because of the nonswelling nature of micaceous KS clay and its smaller CEC, it would also have a greater proportion of pH-dependent charge than smectitic AS clay.

The possible mutual flocculation of positive edge faces and negative surfaces was gradual in KS clay. This behavior of KS clay was probably due to the presence of high organic matter content and the large amount of mica with medium pH-dependent charge. A substantial quantity of organic matter was present in the KS soil. Thus the colloidal organic matter might control the flocculation of clay particles (Chorover and Sposito, 1995; Kaplan et al., 1993; Heil and Sposito, 1993). At pH 3, however, the negative permanent charge on the faces of KS clay was probably balanced by the positive charge on the edges, resulting in flocculation consequently zero EM (not plotted). With increasing pH of KS clay suspension, EM of these clay particles was continuously shifted to more negative values. The pH increases, especially in the case of KS clay might have decreased edge and face attractions due to increased negative charges, thereby causing the double layer expansion and high suspension stability. As reported by Arora and Coleman (1979) and Kretzschmar et al. (1997), it was probably because of a reversal of edge charge from positive to negative preventing edge-to-face flocculation. Dixit (1982) also reported that above pH 8, the EM of a Luvisol suspension increased very rapidly.

The large amount of smectite, probably deposited on positively charged edges of other minerals (kaolinite, chlorite, and partially mica), may mainly be responsible for the sharp flocculation of AS clay around pH 4 and consequently zero EM (not plotted). At pH ≥ 5, AS clay had a strong negative charge. From pH 5, the presence of smectite might inhibit edge-to-face interactions and promote dispersion. Similarly, Miller et al. (1990) concluded that a mixed kaolinitic–smectitic soil was less dispersive than the kaolinitic soils. Frenkel et al. (1978) also reported that the addition of small amount of montmorillonite to kaolin soils promoted the dispersion of kaolin flocs. Arora and Coleman (1979) emphasized similar phenomena for smectite and kaolinite mixtures.

Although many experimental studies on EM and {zeta}p of clay particles have been performed (Cerpa et al., 1999; Dixit, 1982; Lebron and Suarez, 1992; Vane and Zang, 1997; Yeung et al., 1997) in the past, the types and source of clay, method of preparation, particle concentration in suspension, and the composition of aqueous solution were rarely the same among references. In spite of these facts, the pH range regarding EM changes is generally consistent with the published data for clay minerals including smectite and mica. The general shapes of EM and {zeta}p versus pH curves (Fig. 3) are similar to those in the literature (Anderson and Bertsch, 1993; Chorover and Sposito, 1995; Dixit, 1982; Kretzschmar et al., 1997; Lebron and Suarez, 1992; Osei and Singh, 1999), although differences were observed at low pH. Vane and Zang (1997) also emphasized this disagreement between different studies, especially, at low pH. Some studies on EM of different mineral types in the literature concluded that a positive {zeta}p developed at sufficiently low pH, indicating a net positive charge. For example, Kretzschmar et al. (1997) found that EM measurements showed that pure kaolinite had positive net total particle surface charge at low pH and negative surface charge at high pH, with an isoelectric point at pH 4.8. Chorover and Sposito (1995) found that EMs of their soil clays were predominantly negative between pH 2 and 6. However, the mineral composition of soils used in the present study complicates the properties and behaviors of clay fraction. Similarly, Dixit (1982) concluded that the nature of clay was found to be a prime factor in determining the effect on EM.

It should also be noted that Fe-oxides could affect surface charge act as a cementing agent. Changes in pH affect the edge charge on clays and the surface charge of variable charge minerals such as Fe and Al oxides (Suarez et al., 1984). At low pHs we expect edge-to-face bonding to occur, as well as bonding of positive Fe and Al oxides to negative clay surfaces (van Olphen, 1977).

Effect of Ionic Strength
The EM and {zeta}p of clay particles versus electrolyte concentration are presented in Fig. 4 . The KS clay was more sensitive than AS clay at low electrolyte concentration. The negative EM of KS clay reduced from 1.85 x 10–8 m2 s–1 V–1 at 0.001 M NaCl-treated suspension to 1.11 x 10–8 m2 s–1 V–1 at 0.1 M ionic strength. The {zeta}p ranged from –23.35 to –14.01 mV over the same concentration range. Nevertheless, no tangible change in EM of the AS clay suspension was observed when electrolyte concentration increased from 0.001 to 0.01 M. The negative EM of particles in 0.01 M NaCl-treated suspension (EM = 2.03 x 10–8 m2 s–1 V–1) was close to that observed in 0.001 M (EM = 2.00 x 10–8 m2 s–1 V–1). On the other hand, the EM of the same clay decreased to a value of 1.11 x 10–8 m2 s–1 V–1 when the salt concentration in the suspension increased to 0.1 M ionic strength. The effect of salt concentration on EM (or {zeta}p) of different soil clays is, however, not clear in the literature. For example, Anderson and Bertsch (1993) observed a decrease in EM of kaolinite with increasing salt concentration and no changes in EM of bentonite with variation in ionic strength for pH values between 4 and 7. Several researchers (Hunter, 1981; Vane and Zang, 1997) reported that the EM of some clays did not fluctuate with changes in concentration between 10–5 and 10–2 M of different electrolyte. In addition, it has also been a common observation that the colloidal stability of soil clays especially for kaolinitic and montmorillonitic soils is many times greater than that of comparable reference clays (Arora and Coleman, 1979; Frenkel et al., 1992; Goldberg and Forster, 1990; Heil and Sposito, 1993; Miller et al., 1990).



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 4. Electrophoretic mobility (EM) and corresponding zeta potential ({zeta}p) of Ariake soil (AS) and Kzyl-Orda soil (KS) clays versus electrolyte concentration of NaCl.

 
Hydraulic Conductivity
The pore volume of the columns was about 40 mL. For successive different electrolyte concentration experiment, eight pore volumes (320 mL) of displacing solution were enough to displace the original solution in relation to steady-state flow and effluent EC. It was also observed that the flow rate reached the steady-state values within the time that six pore volumes passed through the columns for independent pH experiments and five to seven pore volumes for independent ESP treatments. The HC readings taken after the necessary pore volumes for steady-state flow were presented.

Effect of Exchangeable Sodium Percentage on Hydraulic Conductivity
Saturated HC of KS and AS clay/sand mixtures ranged from 0.3 to 13.5 and 0.2 to 8.0 mm h–1 over the range of ESP between 100 and 0, respectively (remember that this experiment was conducted with DW). There was a sharp drop in HC values of AS-clay/sand mixtures when the ESP reached 60 (Fig. 5) . This threshold for KS-clay/sand mixtures occurred at ESP between 40 and 60.



View larger version (16K):
[in this window]
[in a new window]
  		Fig. 5. Hydraulic conductivity versus exchangeable sodium percentage for Ariake (AS) and Kzyl-Orda (KS) clay/sand mixtures (distilled water was applied to columns).

 
At ESP > 40 a small amount of clay was found in effluents. The concentration of clay in effluents changed between 0.7 and 1.2 g L–1. The changes in HC of KS-clay/sand mixtures probably were governed mainly by dispersion of particles. It seems reasonable that the onset of dispersion should precede clay movement in a packet mixture column. Also, decreases in HC of AS-clay/sand mixture were mostly because of clay dispersion and plugging of pores, because decreases in HC were not reversible (see below Effect of Salt Concentration on Hydraulic Conductivity). Although AS clay contained low swelling smectite (Ohtsubo et al., 2000) swelling might play a minor role in HC. However, Regea et al. (1997) reported that in low swelling smectite, the contribution of the swelling to HC deterioration was negligible. Opinions also differ as to the effect of clay mineralogy on HC. Velasco-Molina et al. (1971) reported that, in the virtual absence of electrolyte, the order of soil dispersion at a given ESP was: montmorillonitic > halloysitic > micas. At low ESP values, the micaceous soil sometimes dispersed more than the halloysitic-kaolinitic soil. Many laboratory studies have also demonstrated the effect of high Na level on decreasing HC, resulting from swelling and of expansible clay and from dispersion of nonexpansible clays (Frenkel et al., 1978; Miller et al., 1990; Suarez et al., 1984).

Effect of pH on Hydraulic Conductivity
Plots of saturated HC vs. pH are presented in Fig. 6 . Hydraulic Conductivity values showed a great sensitivity to pH. Slight increases in HC occurred initially with increasing pH from 3 to 5. But at pH 7 a sharp drop in HC started. Suarez et al. (1984) showed that high pH has an adverse effect on HC of soils. Variable-charge components are considered important parameters in soil stability studies (Goldberg et al., 1988). The potential importance of micaceous soil or heterogeneous clay edges is, in part, the result of having variable charge in contrast to permanent charge of the faces. It is, therefore, reasonable to consider the relative importance of the variable edge charge with respect to face charge; presence of positive and negative charge is also important for linkage of soil clay particles. We consider huge differences in HC between pH 5 and 7 as due primarily to changes in dispersive forces. The HCs of KS-clay/sand mixture were 1.6 and 1.7 mm h–1 at pH 3 and 5, respectively. Thus, at pH 3 and 5, no important change in HC occurred. However, at pH 7 and 9, a sharp drop in HC of this mixture occurred. While HC at pH 7 was 0.7 mm h–1, it was reduced to 0.5 mm h–1 at pH 9. Similarly, no tangible changes in HC of AS-clay/sand mixture occurred at a low pH. Hydraulic conductivity of AS-clay/sand mixture was 0.8 and 1.0 mm h–1 at pH 3 and 5, respectively. But a greater decrease in HC was evident at pH 7 (0.4 mm h–1) and 9 (0.1 mm h–1). It was probably due to a full dispersion of negatively charged particles. At a low pH we can expect edge-to-face bonding to occur, and this type of bonding should result in high HC. It is expected that the sensitivity of soil HC to pH changes depends on the quality of variable charge minerals and organic matter present in the soil. Soils with large amounts of variable charge should be most susceptible to pH effects (Suarez et al., 1984). On the other hand, Goldberg and Glaubig (1987) observed that the effect of pH was much greater for Na-kaolinite, which was flocculated in DW at pH 5.8 than for Na-montmorillonite. Increases in pH probably cause the development of greater negative charge on the clays, enhancing repulsive forces and therefore dispersion. Although the KS clay showed greater absolute decreases in HC, the AS clay actually showed a greater relative decrease. The HC decreased by a factor of about 3 for the KS clay and 8 for the AS clay, as pH was increased from 3 to 9. Gupta et al. (1984) also found increasing clay dispersion with increasing pH from 6.5 to 10.5 for a Na-saturated Indian Soil shaken in 0.1 M NaCl for 1 d. On the other hand, the greater decrease in HC of our clay/sand mixtures from 5 to 7, compared with that from 7 to 9, was not reflected in proportional increases in HC between those pH levels. Although large amount of low swelling smectite (Ohtsubo et al., 2002) was present in AS clay with less pH dependence, in low swelling smectite, the contribution of swelling to HC deterioration can be neglected (Regea et al., 1997).



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 6. Hydraulic conductivity as a function of pH for Ariake (AS) and Kzyl-Orda (KS) Na-clay/sand mixtures under 0.01 M NaCl solution.

 
To interpret the observation of HC in the entire pH range, it is necessary to consider the possible effects of clay mineralogy, the content of organic matter, Fe and Al oxides, and CaCO3. For example, the relationship between flocculation and pH can be altered in the presence of organic matter (Chorover and Sposito, 1995; Heil and Sposito, 1993). On the other hand, in our clays, a mixture of minerals exists each probably with a different point of zero charge. At low pH we expect edge-to-face bonding to occur, as well as bonding of positive Fe oxides to negative clay surfaces. This type of bonding should hinder dispersion and thus should result in optimum HCs (Suarez et al., 1984). But it is difficult to resolve the relative contribution of this mechanism and the effects of other factors for our clay/sand mixtures.

Effect of Salt Concentration on Hydraulic Conductivity
Plots of HC vs. salt concentration for both clay-sand mixtures are given in Fig. 7 . The average HC of AS-clay/sand mixtures was about 1.0 mm h–1 under the leaching solution of 0.1 M, and markedly reduced to 0.3 mm h–1 when leached with 0.01 M solution. However, it was not affected by changes in salt concentration between 0.01 and 0.001 M. When leached with DW, HC decreased to a value of 0.2 mm h–1. The EC of the leachate also decreased drastically and suspended clay particles started to appear in the effluent with the breakthrough of diluted solution (in this treatment the maximum concentration of clay in the effluent was 0.8 g L–1). Saturated HC of KS-clay/sand mixtures decreased with decrease in salt concentration. On average HC of KS mixtures was 1.2 mm h–1 at 0.1 M solution. Reduction in HC started immediately after the solution of 0.01 M was applied to the column (0.6 mm h–1). Upon leaching with more diluted solution and then DW, the HC of the mixture was reduced to 0.4 and then 0.3 mm h–1, respectively. Frenkel et al. (1978) concluded that equivalent reductions in HC occurred at higher salt concentration with montmorillonitic soils than with kaolinitic soils. They also reported that the HC of 31% clay, kaolinitic soil was reduced to essentially zero at all ESP leached with DW. However, in our coarse-textured mixtures, full plugging did not occur because the pores were probably large enough and the water still moved throughout columns. In addition, at the end of this experiment to test reversibility of HC decreases, we applied our most concentrated solution (0.1 M) to the columns. The decrease in HC was not reversible on the application of the solution for both clay/sand mixtures.



View larger version (15K):
[in this window]
[in a new window]
  		Fig. 7. Hydraulic conductivity versus electrolyte concentration of NaCl for Ariake (AS) and Kzyl-Orda (KS) clay/sand mixtures at exchangeable sodium percentage 100.

 
The Relationship between Zeta Potential and Hydraulic Conductivity
In general, there is a good agreement between the effect of ESP, pH, and salt concentration on {zeta}p and saturated HC for AS and KS soil clays but, less for AS. In other words, high level of ESP, low ionic strength, and increased pH promoted an increase in {zeta}p of both soil clay-related decreases in HC of both clay/sand mixtures. Decreased HC can be attributed to the probable increase in thickness of diffuse double layer and, as found in our experiment, {zeta}p. Shainberg and Letey (1984) reported similar dispersion behavior for their soils. Chaudhari (2001) also concluded the same mechanism for three texturally different soils. Dilute sodic solutions enhanced dispersion of our clays in both EM and HC experiments. The {zeta}p, defined as the potential at the shear plane, can be considered a fundamental property that is influenced by salt concentration, exchangeable Na, and pH (Lebron and Suarez, 1992). A high {zeta}p suggests that the double layer is more diffuse (Oster et al., 1980.). In addition, determination of {zeta}p has also been considered a good index of the magnitude of the repulsive interaction between colloidal particles (Hunter, 1981). In other words, {zeta}p is valuable in discussing tendency for the soil colloids to disperse. All these phenomena may also influence HC. In this regard, to evaluate the relationship between {zeta}p (assuming as an independent variable) and HC, all data of {zeta}p and HC vs. ESP, pH ,and salt concentrations were pooled; although clay-particle percentages were different. Then simple regression analysis of the data was performed using a standard procedure (Fig. 8) . Hydraulic conductivity looks like a sensitive parameter to changes in {zeta}p. This indicates that the saturated HC of porous medium such as clay/sand mixtures can be correlated to {zeta}p, which is a function of both solid and aqueous phase properties.



View larger version (22K):
[in this window]
[in a new window]
  		Fig. 8. Relationship between zeta potential ({zeta}p ) of clay particles and hydraulic conductivity of clay/sand mixtures for Ariake (AS) and Kzyl-Orda (KS) samples.

 
However, the EM measurements are not easily conducted thus simple prediction of HC from {zeta}p values is not possible at present. Also, the relationships between {zeta}p and HC do not provide information at the point most needed—when the HC is first adversely affected. On the other hand, in evaluating HC, the most common approach is to determine the flocculation values, because maintaining adequate soil permeability and favorable soil structure is largely dependent on the flocculation–dispersion characteristics of the soil clay fraction. Previous investigations have evaluated the interactive effects of ESP, SAR, electrolyte concentration, and pH on the flocculation–dispersion behavior of the soil clay fraction (Frenkel et al., 1992; Goldberg and Forster, 1990; Heil and Sposito, 1993; Kretzschmar et al., 1997; Lebron and Suarez, 1992; Miller et al., 1990; Oster et al., 1980) and on hydraulic conductivity (Frenkel et al., 1978; Suarez et al., 1984; Chaudhari, 2001). In further research, it will be useful to quantify the relationships among three variables (EM, flocculation values and HC) rather than two to consider.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
The existing information in electrophoretic migration of clay particles is mostly related to very dilute suspensions of homogeneous clay samples, which rarely occur in the soils. However, in this research, clay fractions separated from two different soils and in relatively concentrated suspensions were used to determine the EM of soil clays. Results show that concentrated suspensions and heterogeneous samples can be used in electrophoresis experiments with variable degrees of success. It should also be noted that the simple apparatus with water-cooling system used in this study was capable of providing reliable information on EM.

It was clearly observed that EM of clay particles increased with increasing Na proportion. The EMs of particles at high Ca-clay proportions could not be determined due to rapid flocculation of concentrated suspensions. Mineral composition of our smectitic and micaceous soil clays played a dominant role in their behaviors. The importance of pH effect was also demonstrated by the observation of differences in EM and saturated HC. The assignment of threshold values for HC of some specific soils calls for the consideration of pH along with other concepts such as SAR, ESP, and salt concentration. The micaceous clay was more sensitive than smectitic clay at low electrolyte concentration in the studies of EM and HC.

Finally, it can be concluded that HC is a sensitive parameter to changes in {zeta}p. However, the EM measurements are not easily conducted thus simple prediction of HC from {zeta}p values is not possible at present.

Received for publication October 14, 2002.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Web of Science (5)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Aydin, M.
Right arrow Articles by Kilic, S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Aydin, M.
Right arrow Articles by Kilic, S.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Aydin, M.
Right arrow Articles by Kilic, S.
Related Collections
Right arrow Hydraulic Conductivity
Right arrow Soil Surface Chemistry
Right arrow Experiment Design

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome