Wu, H. & Hu, L. Geotechnique [http://dx.doi.org/10.1680/geot.12.P.

133]

TECHNICAL NOTE

Analytical solution for axisymmetric electro-osmotic consolidation

H. WU and L. HU

Electro-osmotic consolidation is a potential method for soil improvement. An axisymmetric electroosmotic consolidation model with coupled horizontal and vertical seepage is proposed and the
analytical solution is derived without the equal strain hypothesis, which was used in previous models.
The proposed analytical model can predict the distribution of pore-water pressure in the radial
direction, and comparison with previous analytical models indicates that the degree of consolidation is
overestimated if the model is simplified from an axisymmetric model to a planar model. A design
chart is developed to facilitate the evaluation of applying electro-osmotic consolidation for soil
improvement.
KEYWORDS: consolidation; dewatering; electrokinetics; ground improvement; seepage; theoretical analysis

INTRODUCTION
Electro-osmosis, in which an electrical field is applied to a
saturated soil to cause water flow from the anode to the
cathode, has been proved to be an effective method to
consolidate soil and enhance the bearing capacity of ground
(Bjerrum et al., 1967; Casagrande, 1983; Casagrande et al.,
1986; Chew et al., 2004; Hu et al., 2010). Esrig (1968) first
proposed a one-dimensional analytical model to calculate the
excess pore-water pressure. Wan & Mitchell (1976) considered the surcharge preloading, and coupled the two methods.
Thereafter several analytical models were proposed for electro-osmotic consolidation, including the two-dimensional
model in a vertical plane (Shang, 1998; Xu et al., 2011) and
the two-dimensional model in a horizontal plane (Su &
Wang, 2003). Prefabricated vertical drain (PVD) and electric
vertical drain (EVD) are often installed in an equilateral
triangular pattern in the ground so that an axisymmetric
model is more appropriate to describe the behaviour of soil
than the planar models (Glendinning et al., 2008).
The unit cell theory proposed by Barron (1948) was the
most popular method to analyse axisymmetric seepage problems, such as ground treatment with PVD (Carrillo, 1942;
Barron, 1948; Hansbo, 1981; Onoue, 1988; Xie & Zeng,
1989; Indraratna et al., 2005; Wu & Hu, 2012a), as well as
electro-osmotic consolidation (Li et al., 2010; Wu & Hu,
2012b). Li et al. (2010) developed an axisymmetric model
to analyse the soil behaviour of electro-osmotic consolidation with surcharge preloading. Wu & Hu (2012b) proposed
an axisymmetric analytical model for electro-osmotic consolidation coupled with vacuum preloading. However, both
of the above analytical models ignored the vertical flow of
pore water and used the equal strain hypothesis, which
assumes equal vertical soil displacement along the radial
direction at an identical depth. The equal strain hypothesis is

not applicable for all situations, and only average excess

pore-water pressure can be obtained.
In this paper, an axisymmetric model considering both the
radial and vertical flows for electro-osmotic consolidation is
proposed, and the analytical solution is derived to describe
the behaviour of the soil mass. Results from the analytical
model are compared with previous model predictions and
design charts are obtained to evaluate the suitability of
electro-osmotic consolidation to given site conditions.
THEORETICAL ANALYSIS
Two-dimensional axisymmetric model
A schematic diagram of the two-dimensional axisymmetric model for electro-osmotic consolidation with a central
EVD is shown in Fig. 1. The top of the model and the
central EVD are permeable boundaries, while the peripheral
and bottom boundaries are impermeable. The radii of the
model and the drain are re and rw , respectively. The voltage
is distributed along the radial direction and is independent
of time and the vertical position z. The electrical capacitance
per unit volume and the electrochemical reactions are
neglected. The electro-osmosis conductivity and the electrical conductivity in r and z directions are supposed to be the
same and remain constant.
Governing equations, boundary conditions and initial
condition
The combined pore-water flow under the hydraulic gradient and electrical gradient can be described as (Esrig, 1968;
Hu et al., 2012)

Manuscript received 9 September 2012; revised manuscript accepted

25 March 2013.
Discussion on this paper is welcomed by the editor.
 Department of Hydraulic Engineering, Tsinghua University, Beijing, Peoples Republic of China.
Key State Laboratory of Hydro-Science and Engineering, Tsinghua
University, Beijing, Peoples Republic of China (corresponding
author).

vr k r

@H
@V
 ke
@r
@r

vz k z

@H
@z

(1)

where H is the total water head and V is the electrical

potential; v r and v z are the pore-water flow velocity in the
radial and vertical directions; k r and k z are radial and
vertical hydraulic conductivities, and ke is the electro-osmosis conductivity.
Based on the principle of conservation of pore water in a
1

WU AND HU


2
Power source

kr

1 @H @ 2 H
2
r @r
@r


kz

@2H
@H
mv w
@z2
@t

(8)

The peripheral boundary is impermeable so that the

boundary condition can be expressed as
vr (re , z, t) k r

Permeable

@H
@V
 ke
0
@r
@r

(9)

Substituting equation (7) into equation (9), the following

boundary equation can be obtained as
@H(re , z, t) k e V a 1

@r
k r ln n re

Soil

Impermeable

Impermeable

EVD

(10)

The bottom boundary is impermeable and since the voltage

is distributed only in the radial direction, the mathematical
expression for the bottom boundary can be described as

@H(r, z, t) 
0
(11)

@z
zL
where L is the depth of the model.
The boundary of the EVD and the top boundary can be
expressed as

Cathode

Anode

Fig. 1. Diagram of the two-dimensional axisymmetric model of

electro-osmotic consolidation

saturated soil system, the following equation can be derived

under axisymmetric conditions
vr @vr @vz
@H

 mv w
@t
r
@r
@z

(2)

where mv is the coefficient of volume compressibility and

w is the unit weight of water.
The governing equation for electro-osmotic consolidation
under axisymmetric condition can be obtained by substituting equation (1) into equation (2)


1 @H @ 2 H
@ 2 H k e @V
@2V
kr
2 kz 2
k e 2
r @r
@r
@z
@r
r @r
(3)
@H
mv w
@t
The equation for the electric field can be obtained based on
the law of conservation of electrical charge
2

@V
1 @V
0

@r2 r @r
The boundary conditions for V can be expressed as

(4)

V (re , z, t) V a

(5)

V (rw , z, t) 0

(6)

Va
r
ln
ln n rw

(12)

H(r, 0, t) H 0

(13)

where H0 is the total hydraulic head. When the other

methods such as vacuum preloading are used together
with electro-osmosis, H0 can be a certain value according
to the actual conditions. When the electro-osmotic consolidation is used alone for the soil improvement and
there is no water at the top of the model that can enter
the model, the top boundary is actually a 0 flux
boundary, which can be described by a one-dimensional
axisymmetric model.
The initial condition can be described as
H(r, z, 0) 0

(14)

Analytical solutions
According to equations (8) to (14), the analytical solutions
for H can be obtained by the separation of variables (details
are given in the supplementary data available on-line)
H(r, z, t)
1 X
1
X
p
k e V a
F im D0 ( i r)
2
k ln n rw
i1 m1 r
"
#)

 (
2m  1
(2m  1)2 2 k z
z exp C v i
3 sin
t
2L
kr
4L2
1
X
p
k e V a
Ei D0 ( i r)
2
k
ln
n
r
w
i1 r
(
"r
#)
r !
kr
kr
3 exp
i z exp
i (2L  z)
kz
kz

where Va is the applied voltage.

Then the solution for equation (4) can be derived as
V (r)

H(rw , z, t) H 0

(7)

where n denotes the ratio of the diameter of the model to

the diameter of the EVD. Considering equation (7), equation
(3) can be simplified as

ke V a
r
ln H 0
k r ln n rw
(15)

ANALYTICAL SOLUTION FOR AXISYMMETRIC ELECTRO-OSMOTIC CONSOLIDATION

F im

p
2
4
D0 ( i r e )
i
(2m  1)
p 
2 p 
D1 i rw  n2 D20 i re

kr
4L2
i
k z (2m  1)2 2
kr
4L2
1 i
k z (2m  1)2 2

p
8
H0
p rw D1 ( i rw )
(2m  1)
i
p
p

D21 ( i rw )  n2 D20 ( i re )
(16)
p
2
D0 ( i re )
i
"
Ei
r !#
p
p
kr
2
2
2
i L
[D1 ( i rw )  n D0 ( i re )] 1 exp 2
kz
p
p
p
p
J 0 ( i r w )
p
Dj ( i r)
Y j ( i r)  J j ( i r)
Y 0 ( i rw )

(17)
(18)

where J j and Yj are the first and the second Bessel functions
respectively with an order of j, and i is the solution for
D0 (1=2 3 re ) 0 where i equals 1,2,3. . .; Cv k r /(mv w ) is
the coefficient of consolidation in the radial direction. Then
the degree of consolidation can be calculated as

Table 1. Basic parameters for electro-osmotic consolidation in the

numerical and analytical model
Parameter

Value

Radius of model, re : m
Depth of model, L: m
Unit weight of water, sat : kN/m3
Hydraulic conductivity in radial direction, k r : m/s
Hydraulic conductivity in vertical direction, k z : m/s
Coefficient of volume compressibility, mv : MPa1
Electro-osmosis conductivity, ke : m2 /s per V
Applied voltage, Va: V

0.5
1
10
5 3 109
2.5 3 109
0.3
2 3 109
10

Impact of vertical seepage on electro-osmotic consolidation

In order to focus on the effect of vertical seepage on the
electro-osmotic consolidation process when vacuum preloading is also used, H0 is set at 0 to remove the effect of
vacuum preloading in the following analysis. The values for
the ratio of vertical to radial hydraulic conductivity, k z /k r ,
were varied from 0.01 to 1 with a fixed k r ; the comparison
is shown by the real lines in Fig. 2. The horizontal ordinate
Tv denotes the time factor, which can be expressed as
Tv Cv t/L2 : The vertical ordinate denotes the ratio of the

(
"
#)
 p p  

(2m  1)2 2 k z
F im 1= i D1 i rw 2L=(2m  1) exp C v i
t
kr
4L2
i1 m1
(
" r #
)
U 1
1

p  p p




P
k
r
k z =k r 1=i exp 2
i L  1
L r3w =4 n2 (2 ln n  1) 1 H 0 r2w =2 (n2  1) Ei D1 ( i rw ) = i
kz
i1
1 P
1
P

(19)
Equation (19) indicates that the degree of consolidation is
independent of the electro-osmosis conductivity. The applied
voltage will contribute to the excess pore-water pressure and
settlement, but not the degree of consolidation if the electrochemical reactions are neglected.
When electro-osmotic consolidation is used alone, the
one-dimensional axisymmetric model can be employed, in
which the governing equation, the boundary conditions and
the initial condition are shown as follows
8 

>
1 @H @ 2 H
@H
>
> kr
2 mv w
>
>
r
@r
@r
@t
>
>
>
< @H(r , t) k V 1
e
e
a

(20)
@r
k
ln
n re
r
>
>
>
>
>
H(rw , t) 0
>
>
>
:
H(r, 0) 0

settlement at the anode to the depth of the model. The solid

curves represent the results of k z /k r 1, k z /k r 0.5,
k z /k r 0.1 and k z /k r 0.01 respectively.
Figure 2 illustrates that with the increase of k z , the
settlement at the anode decreases. When k z /k r 0.01, the
seepage in the vertical direction can be neglected and
the ratio of the settlement to the depth of the model
reaches 12%.

Tv
0001
0

p 
2=i D0 i re
p 
C i 2 p 
D1 i rw  n2 D20 i re

(21)

10

100

kz/kr 1
kz/kr 05

(22)

DISCUSSION
The basic parameters adopted in the analytical model are
listed in Table 1.

S/L: %

1
X
p
k e V a
ke V a
r
ln
C i D0 ( i r) eCv i t 
2
k ln n rw
k r ln n rw
i1 r

01

The analytical solution can be derived as (details are

given in the supplementary data available on-line)
H(r, t)

001

kz/kr 01
kz/kr 001

10

12

Fig. 2. Relationship between the ratio of settlement at the anode

to the depth of the model and time under different k z /k r

WU AND HU

Ground surface (permeable)

Impact of the equal strain hypothesis

Wu & Hu (2012b) neglected the vertical seepage and
derived an analytical solution for the one-dimensional axisymmetric model with the boundary conditions shown in
equations (10) and (11) under the equal strain hypothesis
 1  et=Br
u(t) MV a (1  et=Br ), U

(23)

Electrical
vertical
drain

 are the average excess pore-water pressure

where u and U
and the average degree of consolidation. Br w r2e mv F i ,
M ke w (F j /kr ), Fi n2 /(n2 1) (ln n  3/4) + (1  1/4/n2 )/
(n2  1), F j n2 /(n2  1)  1/(2 ln n).
Figure 3 illustrates the excess pore-water pressure of the
two analytical solutions shown in equations (21) and (23),
representing the analytical solution with and without the
equal strain hypothesis respectively. Because of the equal
strain hypothesis, the excess pore-water pressure obtained
from equation (23) is the same along the radial direction.
Nevertheless, the proposed analytical solution can reflect the
variation of the excess pore-water pressure from 0 at the
cathode to a relative high negative excess pore-water pressure at the anode.

Comparison with previous analytical solutions

Shang (1998) and Xu et al. (2011) simplified the problem
to a two-dimensional model in the vertical plane as shown
in Fig. 4. Shang (1998) assumed the voltage to be 0 at the
top boundary, which does not account for the actual situation. Xu et al. (2011) accounted for this boundary condition
and derived the analytical solution
u(x, y, t)

1
X
(2i  1)y
i1

(
3

2L
1
X

(mi  mi )eC mi t mi 

(24)

m1

)
(2m  1)x
4k e w V a x

3 sin
2B
k h B(2i  1)

r /rw

Fig. 4. Diagram of the two-dimensional model in vertical plane

for electro-osmotic consolidation

C mi C x

(2m  1)2 2
(2i  1)2 2
Cy
2
4B
4L2

mi

32k e w V a B(1)m1
k h (2m  1)2 (2i  1)3

mi

8k e w V a BC y (2i  1)(1)m1
C mi k h (2m  1)2 

(25)

where Cx and C y are the coefficients of consolidation in the

horizontal and vertical directions; kh is the hydraulic conductivity in the horizontal direction; B is the distance between the anode and the cathode.
Figure 5 shows the comparison of the proposed twodimensional axisymmetric model with models proposed by
Shang (1998) and Xu et al. (2011). The x-coordinate, Tx , is
the time factor in the horizontal direction. The model
proposed by Shang (1998) overestimates the degree of consolidation for the two-dimensional model in the vertical
plane by about 8% at Tx 0.01, as there is a voltage
gradient in the vertical direction attributable to the assumption of zero volts at the top boundary. The model proposed
by Xu et al. (2011) overestimates the degree of consolida10

10

125
Two-dimensional
axisymmetric model

01

Degree of consolidation

Excess pore water pressure: 100 kPa

Anode
Cathode
(impermeable) (permeable)
z

08

02

04

03

Free strain: t 02 d
Free strain: t 1 d
Free strain: t 2 d
Free strain: t 5 d
Equal strain: t 02 d
Equal strain: t 1 d
Equal strain: t 2 d
Equal strain: t 5 d

Fig. 3. Comparison of previous analytical result and proposed

analytical result regarding excess pore-water pressure

Two-dimensional
model in Xu et al. (2011)
Two-dimensional
model in Shang (1998)

06

04

02

0
00001

0001

001

01

10

Tx

Fig. 5. Comparison of proposed analytical model and previous

two-dimensional model in vertical plane about the degree of
consolidation

ANALYTICAL SOLUTION FOR AXISYMMETRIC ELECTRO-OSMOTIC CONSOLIDATION

5
tion by about 15% at Tx 0.01 and 50% at Tx 0.1 owing
solution is derived to show the distribution of the excess
to the simplification from axisymmetric model to planar
pore water along the radial and vertical directions. Based on
model.
the developed analytical solutions, the following conclusions
can be drawn.
Field application of the proposed analytical solution
Figure 2 gives the relationship between the degree of
consolidation and the time factor. However, these curves are
generated for a specific problem, given size of the EVD,
basic properties of the soil and the electrical gradient. In
order to develop useful information for practitioners, Fig. 6
was developed using three dimensionless variables, namely,
re /L, S/L and G. The re /L value describes size of the treated
zone, S/L is the ratio of final settlement at the anode to the
thickness of the clay layer and G describes the electroosmosis properties
G

ke V a
k r [(re rw )=2] ln n

(26)

With a designed EVD and treatment process, if the basic

soil properties are measured, then Fig. 6 can be used to
capture the relation between the final settlement and the
electrical gradient. In the design of electro-osmotic consolidation, specifically for a given site with soil properties, it is
possible either to select the treatment zone r or the treatment
voltage to obtain desired settlement based on Fig. 6.
An example is analysed and the properties of the ground
are supposed to be as follows: ke 2.5 3 108 m2 /(s V),
k r 5 3 1010 m/s, k z 2.5 3 1010 m/s, mv 0.312 MPa1 ,
L 10 m. The radius of the EVD is assumed to be 0.04 m
and the applied voltage is set to be 12 V. If a settlement of
0.9 m is expected, then according to Fig. 6, re should be
0.75 m. When re is 0.75 m, G equals 520 and re /L equals
0.075. Then from the curve with G equals 520, S/H is 0.09,
which means the settlement is 0.9 m. Therefore, during the
installation of EVDs, the space between the anode and the
cathode should be 0.75 m in order to obtain a settlement of
0.9 m, and this is also the optimal choice for re in this
situation.

SUMMARY AND CONCLUSIONS

A two-dimensional axisymmetric model for electro-osmotic consolidation is proposed to describe the coupling of
radial and vertical pore-water movement, and the analytical
re /L
0

01

02

03

04

05

005

S/L

010

015

020

025

G 400
G 600
G 900
G 1200

G 500
G 700
G 1000

G 520
G 800
G 1100

Fig. 6. Relation between S/L and re /L under different G

(a) The proposed analytical model couples the pore-water

movement in horizontal and vertical directions with
electro-osmotic consolidation; the result indicates that
with the decrease of vertical hydraulic conductivity, the
settlement at the anode caused by electro-osmotic
consolidation increases.
(b) The proposed analytical model is more general and can
exactly describe the distribution of excess pore-water
pressure along the radial direction, whereas previous
models can only give the uniform excess pore-water
pressure, owing to the equal strain hypothesis.
(c) Since EVDs are often installed in an equilateral triangular
pattern, the axisymmetric model is more appropriate to
describe the behaviour of soil than the planar model. The
degree of consolidation is overestimated if the model is
simplified from an axisymmetric model to a planar
model.
(d ) Design curves are developed for the electro-osmotic
consolidation, which describes the relation between the
final settlement, the electrode spacing, the soil properties
and the applied voltage. Hence, the final settlement can
be predicted using these charts for a given site
configuration.

ACKNOWLEDGEMENTS
Financial support from the National Natural Science Foundation of China (NSFC50978139), National Key Basic Research Program (2012CB719804) of the Ministry of Science
and Technology, and Tsinghua University (2010THZ021) is
gratefully acknowledged.
NOTATION
B length of the model
Br calculating factor for 1D axisymmetric model under equal
strain condition, Br w r 2e mv Fi
Ci calculating factor for 1D axisymmetric model without
equal strain condition
Cv coefficient of consolidation in the radial direction
Cx , C y coefficients of consolidation in horizontal and vertical
directions for the planar model
D0 function related to the first and second Bessel functions
with an order of 0
D1 function related to the first and second Bessel functions
with an order of 1
E0 Youngs module
Fim calculating factor for 2D axisymmetric model without
equal strain condition
G electro-osmosis properties of soil
H total water head
H0 total hydraulic head applied on the model
J j, Yj the first and second Bessel functions with an order of j
ke electro-osmosis conductivity
k r , kz hydraulic conductivities in the radial and vertical directions
L depth of the model
M calculating factor for 1D axisymmetric model under equal
strain condition, M ke w (F j /k r )
mv coefficient of volume compressibility
n ratio of diameters of the model to the EVD
rw, re radii of the central EVD and the model
S/L ratio of diameters of the final settlement at anode to
thickness of clay layer
Tx time factor in horizontal direction for the planar model
U degree of consolidation for the 2D axisymmetric model
 average degree of consolidation for the 1D axisymmetric
U
model

WU AND HU

u pore-water pressure for the 2D model in the vertical plane

u average excess pore-water pressure in the radial direction
for the 1D axisymmetric model
V voltage in the model
Va applied voltage on the whole model
v r , v z pore-water flow velocities in the radial and vertical
directions
w unit weights of water and saturated soil
v volumetric strain of soil mass
i the solution for D0 (1=2 3 re ) 0 where i equals 1, 2, 3. . .
 Poissons ratio

REFERENCES
Barron, R. A. (1948). Consolidation of fine-grained soils by drain
wells. ASCE Trans. 113, 718754.
Bjerrum, L., Moum, J. & Eide, O. (1967). Application of electroosmosis to a foundation problem in Norwegian quick clay.
Geotechnique 17, No. 3, 214235, http://dx.doi.org/10.1680/
geot.1967.17.3.214.
Carrillo, N. (1942). Simple two and three dimensional cases in the
theory of consolidation of soils. J. Mathl Physics 21, No. 1, 15.
Casagrande, L. (1983). Stabilization of soils by means of electroosmotic state-of-art. J. Boston Civ. Eng. ASCE 69, No. 3, 255
302.
Casagrande, D. R., Casagrande, L. & Castro, G. (1986). Improvement of sensitive silty clay by electro-osmosis: Discussion. Can.
Geotech. J. 23, No. 1, 9596.
Chew, S. H., Karunatatne, G. P., Kuma, V. M. (2004). A field trial
for soft clay consolidation using electric vertical drains. Geotextiles and Geomembranes 22, No. 1, 1735.
Esrig, M. I. (1968). Pore pressures, consolidation and electrokinetics. J. Soil Mech. Found. Engng Div., ASCE 94, SM4, No.
7, 899921.
Glendinning, S., Lamont-Black, J., Jones, C. J. F. P. & Hall, J. (2008).
Treatment of lagooned sewage sludge in situ using electrokinetic
geosynthetics. Geosynthetics Int. 15, No. 3, 192204.
Hansbo, S. (1981). Consolidation of fine-grained soils by prefabri-

cated drains. Proc. 10th Int. Conf. Soil Mech. Found. Engng,
Stockholm 3, 677682.
Hu, L. M., Wu, W. L. & Wu, H. (2010). Theoretical analysis and
numerical simulation of electro-osmosis consolidation for soft
clay. Rock Soil Mech. 31, No. 12, 39773983 (in Chinese).
Hu, L. M., Wu, W. L. & Wu, H. (2012). Numerical model of
electro-osmotic consolidation in clay. Geotechnique 62, No. 6,
537541, http://dx.doi.org/10.1680/geot.11.T.008.
Indraratna, B., Rujikiatkamjorn, C. & Sathananthan, L. (2005).
Analytical and numerical solutions for a single vertical drain
including the effects of vacuum preloading. Can. Geotech. J. 42,
No. 4, 9941014.
Li, Y., Gong, X. N., Lu, M. M. & Guo, B. (2010). Coupling
consolidation theory under combined action of load and electroosmosis. Chin. J. Geotech. Engng 32, No. 1, 7781 (in
Chinese).
Onoue, A. (1988). Consolidation by vertical drains taking well
resistance and smear into consideration. Soils Found. 28, No. 4,
165174.
Shang, J. Q. (1998). Electro-osmotic enhanced preloading consolidation via vertical drains. Can. Geotech. J. 35, No. 3, 491499.
Su, J. & Wang, Z. (2003). The two-dimensional consolidation
theory of electro-osmosis. Geotechnique 53, No. 8, 759763,
http://dx.doi.org/10.1680/geot.2003.53.8.759.
Wan, T. Y. & Mitchell, J. K. (1976). Electro-osmotic consolidation
of soils. J. Geotech. Engng Div., ASCE 102, No. 5, 473491.
Wu, H. & Hu, L. (2012a). Analytical and numerical solutions for
vacuum preloading considering a radius related strain distribution. Mech. Res. Comm. 44, No. 1, 914.
Wu, H. & Hu, L. M. (2012b). Analytical models of the coupling of
vacuum preloading and electro-osmosis consolidation for ground
stabilization. J. Tsinghua Univ. (Sci. & Tech.) 52, No. 2, 182
185 (in Chinese).
Xie, K. H. & Zeng, G. X. (1989). Consolidation theories for drain
wells under equal strain condition. Chin. J. Geotech. Engng 11,
No. 2, 317 (in Chinese).
Xu, W., Liu, S. H., Wang, L. J. & Wang, J. B. (2011). Analytical
theory of soft ground consolidation under vacuum preloading
combined with electro-osmosis. J. Hohai Univ. (Nat. Sci.) 30,
No. 2, 169175 (in Chinese).