Professional Documents
Culture Documents
Research paper
a r t i c l e i n f o a b s t r a c t
Article history: Electro-osmosis can be a potential improvement technique for expansive soils. One-dimensional column
Received 29 January 2015 experiments were conducted to comprehend the impact of electro-osmosis on the consolidation behavior and the
Received in revised form 4 April 2015 geotechnical properties of a sodium bentonite with different electrode materials, including two reactive electrodes
Accepted 6 April 2015
(copper and iron) and two inert electrodes (graphite and stainless steel). The change in drainage rate, electric
Available online xxxx
current and voltage loss close to the anode were similar for the four electrodes, with a rapid change stage in the
Keywords:
first 2 h followed by a slow change stage in the next 6 to 8 h and finally a stable stage. The rapid voltage loss in
Electro-osmosis the first stage was caused by soil cracking near the anode due to severe volume shrinkage and gas generation.
Sodium bentonite Soil properties, including plasticity, swelling potential, zeta potential, and cation exchange capacity decreased
Electrode material emarkably in the vicinity of the anode. During electro-osmosis, the reactive electrodes caused more significant
Consolidation change in soil properties than the inert ones. Electro-osmosis technique can reduce the swelling potential of the
Geotechnical property bentonite, however, the zone of influence is restrained.
© 2015 Elsevier B.V. All rights reserved.
1. Introduction 2001; Asavadorndeja and Glawe, 2005; Ou et al., 2009; Abdullah and
Al-Abadi, 2010). However, their conclusions were different or even
Electro-osmosis is an electrically induced process in which pore contradictory when different kinds of soils were tested. The results on
water is driven from the anode to the cathode with the dissolved illitic clay and soft Bangkok clay indicated that both the liquid limit
electrolytes, leading to dewatering and consolidation of soil or and the plastic limit increased, while marine clay exhibited no change
other porous media. Casagrande (1948) made use of this phenomenon in neither the liquid limit nor the plastic limit during electro-osmosis
to consolidate soft soil and enhance the geotechnical properties for (Bjerrum et al., 1967; Esrig and Gemeinhardt, 1967; Lo and Ho, 1991;
engineering purposes. Since then, electro-osmosis has been successfully Bergado et al., 2000; Micic et al., 2001). Both the clay mineral category
employed for soft ground improvement, slope stabilization, dewatering and the content have considerable influences on soil behavior during
of tailings and sludge, pile foundation and soil remediation (Casagrande, electro-osmosis. Most of the previous electro-osmosis experiments
1983; Lo and Ho, 1991; Reddy and Saichek, 2002; Glendinning et al., were conducted on kaolinite or local soft clay, while studies on expansive
2007; Jones et al., 2008; Chien et al., 2009; Lamont-Black and Weltman, soils that usually contain large amounts of expansive clay minerals
2010; Cameselle and Reddy, 2012; Hu et al., 2012; Wu and Hu, 2013; (especially smectite) were rarely reported. Expansive soils are
Hu and Wu, 2014). characterized by severe swelling and shrinkage due to change in
The strength of soil is enhanced as a result of two processes: the water content, which may lead to significant volume change and result
discharge of pore water and the complex chemical reactions including in deformation of overlying buildings or pavements. Electro-osmosis
water electrolysis, electrode corrosion, cation exchange and cementation may be a potential technique to improve expansive soils since the soil
under the influence of an electric field. These chemical reactions are plasticity is changed during this process.
expected to alter the geotechnical properties of the soils. Previous studies The main objective of this paper was to study the impact of electro-
reported the change in Atterberg limits for different soils after osmosis on the consolidation behavior and the geotechnical properties
electro-osmosis (Bjerrum et al., 1967; Esrig and Gemeinhardt, 1967; of a sodium bentonite, which has similar characteristics (high swelling
Casagrande, 1983; Lo and Ho, 1991; Bergado et al., 2000; Micic et al., and low permeability) and components (smectite) as expansive soil.
One-dimensional column experiments were conducted on the sodium
bentonite with reactive (copper and iron) and inert (graphite and
⁎ Corresponding author. Tel.: +86 10 62797416; fax: +86 10 62773576. stainless steel) electrodes in order to analyze the effect of different
E-mail addresses: hui-wu10@mails.tsinghua.edu.cn (H. Wu), gehu@tsinghua.edu.cn
(L. Hu), wenqb@mail.tsinghua.edu.cn (Q. Wen).
electrode materials. During the experiments, water discharge, electric
1
Tel.: +86 10 62794154. current and voltage distribution along the soil column were monitored.
2
Tel.: +86 10 62788559. A camera was used to observe the behavior of the soil–anode interface.
http://dx.doi.org/10.1016/j.clay.2015.04.006
0169-1317/© 2015 Elsevier B.V. All rights reserved.
H. Wu et al. / Applied Clay Science 111 (2015) 76–82 77
After the experiments, soil samples taken from different positions of the V6 were placed in the vicinity of electrodes so that the voltage loss
soil column were prepared for geotechnical property tests, including close to the electrodes could be measured. In addition, a camera was
plasticity index, free swelling ratio, zeta potential, and cation exchange applied during the experiments in order to investigate the behavior of
capacity. soil–anode interface.
Previous studies indicated that the drainage and consolidation
2. Materials and experiments behavior was different when different electrode materials were
used during electro-osmosis (Lockhart, 1983; Bergado et al., 2000;
2.1. Materials Mohamedelhassan and Shang, 2001). In order to study the effect of
electrode materials on the electro-osmotic enhancement of bentonite,
A sodium bentonite from the city of Zhangjiakou in Hebei Province four types of electrodes were used including two reactive electrodes
of China was used for electro-osmosis experiments. The geotechnical (copper and iron) and two inert electrodes (graphite and stainless
properties and the chemical composition of the material are summarized steel).
in Table 1. The high plasticity index (124%) and free swelling ratio (540%)
indicate that the sodium bentonite possesses remarkable swelling and
shrinkage characteristics. The dominant Na+ cation, as reflected by the 2.3. Sample preparation
2.9 wt.% Na2O, is the main reason for the high swelling potential. The
specific surface area is 33.89 m2/g, and the cation exchange capacity is A porous stone was placed on the bottom of the testing cell to prevent
40.03 meq/100 g. the bentonite from being pressed out. The anode platen was then placed
upon the porous stone. The bentonite powder was first mixed with
2.2. Experiment apparatus deionized water at a pre-determined water content of 10 wt.% before
compacted in the column by five layers. After that, the compacted
Fig. 1 displayed the details of the 1-D column apparatus used in this bentonite was saturated by vacuum method to achieve an initial
study. The testing cell was made of 5 mm thick plexiglass, and had an water content of 150 wt.%. The cathode platen, along with a filter
inside diameter of 90 mm and a height of 250 mm. The wall was paper, was placed upon the bentonite sample.
equipped with two pipes for inflow and outflow of water. The upper
pipe was used for outflow of discharged water due to electro-osmosis. 2.4. Experiment program
The lower pipe was used for inflow of deionized water during the
sample saturation period and was sealed with glass cement to achieve A constant voltage gradient of 100 V/m was applied to the bentonite
an impervious bottom boundary before the test began. The anode platen samples for 24 h. After the experiments, soil samples from different
was placed on the bottom of the soil column, while the porous cathode locations of the bentonite column were prepared for subsequent physical
platen was placed on the top. The thickness of the electrode was 3 mm and chemical analysis, including the determination of Atterberg limits,
and more details about the cathode were illustrated in Fig. 1. Small holes free swelling ratio, and cation exchange capacity (Chinese Specifications
with a radius of 0.5 mm were drilled on the cathode platen so that the of Soil Testing, SL237-1999). The zeta potential was measured with a
discharged water could pass through. Particularly 8 circular grooves, Beckman DelsaNano C zeta potential instrument at a constant pH of 6.5.
0.5 mm in depth and 2 mm in width, and a vertical groove, 2 mm in
width, were slotted on the cathode to lead the discharged water to the
drainage pipe. The depth of the vertical groove was designed to be 3. Results and discussion
gradational and increased from 0.5 mm on the left side to 1.5 mm on
the right side. The discharged water can flow because of gravity to the 3.1. Water discharge and electric current
right side of the vertical groove, which was aimed at the drainage pipe.
Six voltage probes were inserted into the bentonite column through The water discharge and drainage rate for the four electrodes were
the small holes on the testing cell. The probes were marked as V1, V2, displayed in Figs. 2 and 3. The total volume of the discharged water
V3, V4, V5, and V6, which had a distance of 10 mm, 46 mm, 82 mm, was 69 ml, 47.5 ml, 30 ml, and 26 ml for the copper, iron, graphite,
118 mm, 154 mm, and 190 mm to the cathode, respectively. V1 and and stainless steel electrodes respectively. The copper electrodes
accounted for the best drainage performance while the stainless steel
electrodes accounted for the worst.
Table 1
Geotechnical properties and chemical composition of the sodium bentonite.
The initial drainage rate for the four electrodes was approximately
34 ml/h, and three stages afterwards for water drainage were observed
Properties Values (Fig. 3). In the first 2 h (Stage 1), the drainage rate decreased rapidly to
Geotechnical properties less than 5 ml/h. This was followed by a slow decrease stage (Stage 2) in
Specific gravity, Gs 2.625 the next 6 to 8 h and afterwards a stable stage (Stage 3). Fig. 4 showed
Initial water content, w (%) 1.33
the real time electric current through the bentonite columns during the
Liquid limit, wL (%) 155
Plastic limit, wP (%) 31 experiments. In the first half hour, the electric current increased slightly
Plasticity index, Ip (%) 124 due to the presence of salt precipitates that went into the solution, as
Free swelling ratio, FSR (%) 540 well as the desorption and mobilization of ions in the soil matrix
Cation exchange capacity, CEC (meq/100 g) 40.03 (Mitchell and Soga, 2005; Yukselen-Aksoy and Reddy, 2012; Cameselle
Zeta potential, ζ (mv) −36.05
Chemical composition (weight proportion, %)
and Reddy, 2013). Afterwards, the change in electric current was similar
SiO2 68.2 to that in drainage rate, with a rapid decrease stage (Stage 1) where the
Al2O3 15.1 initial values of around 0.3 to 0.35 A were decreased to 0.12, 0.08, 0.03,
CaO 4.2 and 0.03 A for the copper, iron, graphite, and stainless steel electrodes
MgO 3.8
respectively. There after a slow decrease stage (Stage 2) followed in
Fe2O3 3.1
Na2O 2.9 which the electric current further decreased to 0.05, 0.02, 0.01, and
K2O 1.6 0.01 A and finally a stable stage (Stage 3) was reached. Figs. 3 and 4
SO3 0.4 indicated that the copper electrodes had the largest drainage rate and
TiO2 0.4 electric current. The reactive electrodes exhibited a larger electric
Cl 0.08
current and therefore a better drainage effect than the inert electrodes.
78 H. Wu et al. / Applied Clay Science 111 (2015) 76–82
Fig. 1. One dimensional column apparatus for electro-osmosis experiment (length unit: mm).
3.2. Voltage loss losses. A Norwegian clay presented a voltage loss of 52% close to the
anode after electro-osmosis, while a clay from Canada showed a voltage
During electro-osmosis, the effective voltage applied on the soil loss of 25% (Bjerrum et al., 1967; Lo and Ho, 1991). Mohamedelhassan
samples encountered significant reduction close to the anode, and the and Shang (2001) performed experiments on a marine sediment in
voltage loss for this study was illustrated in Fig. 5. Similar to the drainage Korea using different electrode materials, and the voltage loss at the
rate and electric current, the change in voltage loss close to the anode anode-soil interface was about 9% to 24%. The results from experiments
also experienced 3 stages. About 58% (copper), 72% (iron), 79% conducted by Tao et al. (2013) on a Zhejiang clay from China showed a
(graphite) and 77% (stainless steel) of the applied voltage was lost 23% voltage loss for iron anode and an 83% for voltage loss for copper
in the first 2 h (Stage 1). In the next 6 to 8 h (Stage 2), a relatively anode. As for the sodium bentonite in this study, all four electrodes
small voltage loss occurred, which was about 22%, 13%, 10% and exhibited a voltage loss of more than 80% close to the anode.
10% respectively. The 3 stages represented in Figs. 3 to 5 had an identical During electro-osmosis, the soil close to the anode lost water and
duration since the decrease of the electric current and the drainage rate therefore a volume shrinkage appeared at the bottom of the soil column.
was mainly a result from the increase of voltage loss close to the anode. This was accompanied by an increasing gas pressure at the soil–anode
After the experiments, a total of 84%, 89%, 89% and 88% of the applied interface due to water electrolysis and the impervious bottom boundary
voltage were lost close to the anode for the copper, iron, graphite, and (Acar et al., 1990). Because the sodium bentonite used in this study
stainless steel electrodes respectively. possessed a more significant shrinkage property than other soils, soil
Voltage losses close to the anode have also been analyzed in previous cracking close to the anode was more severe (Fig. 6). For the reactive
laboratory and field studies, where the results showed that both the soil electrodes vertical cracks were formed after about 1.8 h, while a
type and electrode material had remarkable impact on the overall voltage horizontal crack appeared at the soil–anode interface for the inert
electrodes. These cracks, especially the horizontal crack, caused the
significant voltage loss in Stage 1 (Fig. 5), which further led to the
Fig. 2. Water discharge of the electro-osmosis experiments. Fig. 3. Drainage rate of the electro-osmosis experiments.
H. Wu et al. / Applied Clay Science 111 (2015) 76–82 79
than a horizontal crack was formed for the reactive electrodes. 3.3.3. Zeta potential and cation exchange capacity
Moreover, Figs. 7 and 8 further indicated that the copper electrodes Under an applied electric field, ion exchange reactions occur
had the best restraining effect on the swelling shrinkage potential, between the released metal ions and the pre-existing ions in the double
therefore the copper electrodes presented the least voltage loss layer. This leads to the change of ion species and concentration in the
and the best drainage performance. double layer, and further induces the change of double layer properties
such as zeta potential and cation exchange capacity (CEC) (Van Olphen,
1977; Shang et al., 1993; Wu and Hu, 2014; Du et al., 2015).
Fig. 7. Change of soil plasticity after electro-osmosis. Fig. 8. Change of the free swelling ratio after electro-osmosis.
H. Wu et al. / Applied Clay Science 111 (2015) 76–82 81
4. Conclusions
Jones, C.J.F.P., Lamont-Black, J., Glendinning, S., Bergado, D., Eng, T., Fourie, A., Hu, L., Pugh, C., Reddy, K.R., Saichek, R.E., 2002. Effect of soil type on electrokinetic removal of phenanthrene
Romantshuk, M., Simpanen, S., Zhuang, Y., 2008. Recent research and applications in the using surfactants and cosolvents. J. Environ. Eng. 129 (4), 336–346.
use of electro-kinetic geo-synthetics. Keynote paper, Proc. 4th European Geo-synthetics Shang, J.Q., Lo, K.Y., Quigley, R.M., 1993. Quantitative determination of potential distribution
Conf., Edinburgh, UK, Paper No. 329, CDROM. in Stern–Gouy double layer model. Can. Geotech. J. 31, 624–636.
Lamont-Black, J., Weltman, A., 2010. Electrokinetic Strengthening and Repair of Slopes. Stoltz, G., Cuisinier, O., Masrouri, F., 2012. Multi-scale analysis of the swelling and shrinkage
Ground Engineering, pp. 28–31 (April). of a lime-treated expansive clayey soil. Appl. Clay Sci. 61, 33–51.
Lo, K.Y., Ho, K.S., 1991. The effects of electroosmotic field treatment on the soil properties Tao, Y.L., Zhou, J., Gong, X.N., Chen, Z., Li, Y.W., 2013. Comparative experiment on influence of
of a soft sensitive clay. Can. Geotech. J. 28, 763–770. ferrum and cuprum electrodes on electroosmotic effects. Chin. J. Geotech. Eng. 35 (2),
Lockhart, N.C., 1983. Electroosmotic dewatering of clays. III. Influence of clay types, 388–394 (in Chinese).
exchangeable cations, and electrode materials. Colloids Surf. 6, 253–269. Van Olphen, H., 1977. An introduction to Clay Colloid Chemistry. Interscience, New York.
Micic, S., Shang, J.Q., Lo, K.Y., Lee, Y.N., Lee, S.W., 2001. Electrokinetic strengthening of a Wu, H., Hu, L.M., 2013. Analytical solution for axisymmetric electro-osmotic consolidation.
marine sediment using intermittent current. Can. Geotech. J. 38, 287–302. Geotechnique 63 (12), 1074–1079.
Mitchell, J.K., Soga, K., 2005. Fundamentals of Soil Behavior. 3rd Edition. John Wiley & Wu, H., Hu, L.M., 2014. Microfabric change of electro-osmotic stabilized bentonite. Appl.
Sons, Inc., USA. Clay Sci. 101, 503–509.
Mohamedelhassan, E., Shang, J.Q., 2001. Effects of electrode materials and current Yukselen-Aksoy, Y., Reddy, K.R., 2012. Effect of soil composition on electrokinetically
intermittence in electro-osmosis. Ground Improv. 5, 3–11. enhanced persulfate oxidation of polychlorobiphenyls. Electrochim. Acta 86, 164–169.
Ou, C.Y., Chien, S.C., Wang, Y.G., 2009. On the enhancement of electroosmotic soil
improvement by the injection of saline solutions. Appl. Clay Sci. 44, 130–136.