Interface Behavior from Suction-Controlled Direct Shear Test

on Completely Decomposed Granitic Soil and Steel Surfaces

L. Borana1; J.-H. Yin2; D. N. Singh3; and S. K. Shukla4

Abstract: A soil-structure interface is defined as the contact surface between a soil and a structure through which stresses are transferred
from the soil to the structure or vice versa. The ultimate shear strength at the interface is one of the key parameters required for the design and
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.

safety assessment of a structure in the soil, such as pile foundations, retaining walls, buried pipelines, and soil nails. In this paper, the shearing
behavior of completely decomposed granite soil and steel interfaces is examined using a modified suction-controlled direct shear apparatus. A
series of direct shear tests was performed on two different types of soil-steel interfaces under different stress state variables. The experimental
results were used to investigate the influence of counterface roughness on the failure envelopes of soil-steel interfaces. Test results show that
matric suction has a significant influence on the shear behavior and shear strength of the interfaces. It has been observed that the critical inter-
face shear strength for a specific counterface roughness and net normal stress (NNS) depends on the applied matric suction. Also, both the
peak and the postpeak shear stresses are greatly influenced by variation in NNS. Furthermore, the experimental shear strength data are com-
pared with an analytical model that considers the influence of suction and dilation on an apparent interface friction angle. It is noted that, for
the applied NNS and matric suction, the analytical model works well for both rough interfaces. DOI: 10.1061/(ASCE)GM.1943-
5622.0000658. © 2016 American Society of Civil Engineers.
Author keywords: Unsaturated soil; Interface; Counterface; Completely decomposed granite; Matric suction; Soil dilatancy; Direct shear
test.

Introduction
The variation of soil moisture content significantly influences the soil
behavior, especially soil properties, such as permeability, volume
change, and shear strength. (Fredlund and Rahardjo 1993; Ng and
Menzies 2007; Rao et al. 2011). When the unsaturated soil comes in
contact with materials such as steel, concrete, geosynthetics, cement,
wood, and so forth, it forms an interface layer (Yoshimi and Kishida
1981; Uesugi et al. 1990; Zhan and Ng 2006), also referred to as the
unsaturated soil interface, and is common to many geotechnical engi-
neering structures, e.g., buried pipelines, soil nails, retaining walls,
shallow foundations and pile foundations, and so on. The importance
and necessity of understanding the behavior of interfacial interactions
in soils have been advocated by many researchers (Kim and O’Neill
1998; Khoury et al. 2010; Pei et al. 2013).
Estimating the interface shear strength of unsaturated soil is an
important aspect in many geotechnical engineering applications
(Escario and Sáez 1986; Fredlund et al. 1978; Borana et al. 2015).
The interface shear strength between compacted soil and structure
1 In Hong Kong, the application of open-ended steel pipe piles
Research Associate (formerly, Ph.D. Student), Dept. of Civil and
Environmental Engineering, Hong Kong Polytechnic Univ., Hung Hom,
Kowloon, Hong Kong. E-mail: lalitborana@gmail.com
2
Chair Professor, Dept. of Civil and Environmental Engineering, Hong
Kong Polytechnic Univ., Hung Hom, Kowloon, Hong Kong (correspond-
ing author). E-mail: cejhyin@polyu.edu.hk
3
Professor, Dept. of Civil Engineering, Indian Institute of Technology
Bombay, Powai, Mumbai 400 076, India. E-mail: dns@civil.iitb.ac.in
4
Associate Professor, Dept. of Civil Engineering, Edith Cowan Univ.,
Joondalup, Perth, WA 6027, Australia. E-mail: s.shukla@ecu.edu.au
Note. This manuscript was submitted on October 3, 2014; approved on
January 20, 2016; published online on March 2, 2016. Discussion period
open until August 2, 2016; separate discussions must be submitted for
individual papers. This paper is part of the International Journal of
Geomechanics, © ASCE, ISSN 1532-3641.
is a key factor in assessing the reliability and stability of the struc-
ture (Potyondy 1961; Boulon 1989; Mashhour et al. 1996). It is
believed that the interface shear strength (in a saturated soil condi-
tion) mainly depends on the surface roughness of the counterface.
For instance, Hamid and Miller (2009) have conducted extensive
experimental investigations to study the interfacial interaction
between the unsaturated silt-steel interfaces with rough and smooth
counterfaces and found that the interface with rough counterfaces
exhibits more shearing resistance when compared with the smooth
steel counterface. However, the shear strength of soil was found to
be greater than that of interfaces under similar stress state condi-
tions. Sharma et al. (2007) performed a series of experiments on the
soil-geomembrane interface and observed that soil suction contrib-
utes to the shear resistance only at a lower suction. Hossain and Yin
(2014) pointed out that the dilative characteristic of soil is signifi-
cantly influenced by the stress state variable, especially by the
matric suction. Thus, it is important to consider the influence of
interface dilatancy while determining the interface shear strength,
particularly while designing or assessing the overall safety and sta-
bility of the structural system.
is gaining popularity and is often used as the foundations of
bridges, seawalls, cellular walls (Okamoto and Katayama 2015).
The pile-soil interface behavior is of great interest, especially in
the granular soils. This is because the shaft friction plays an im-
portant role in resisting the applied load. Good understanding of
the mechanism of interface shear behavior during pile installa-
tions and static loadings assists in the rational design of the shaft
capacity (Yu and Yang 2011). Nonetheless, it is noted that there
have been quite feeble attempts by earlier researchers to study the
behavior of interfaces in unsaturated soils, especially the unsatu-
rated interface shear behavior corresponding to a specific counter-
face roughness under different stress state variables. Particularly, it
has been realized that there is a dearth of literature on the behavior

© ASCE D4016008-1 Int. J. Geomech.

Int. J. Geomech., 2016, 16(6): D4016008

 100
90
80
Percentage finer by weight (%)
70
60
50
40
30
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.
20
10
0
1E-3
Fig. 1. Particle size distribution characteristics of the CDG soil
of unsaturated interfaces in coarse-grained soil. With this in view,
the present study has been conducted to investigate (1) the ele-
mentary interface behavior of the rough soil-steel interface under
different stress state variables and counterface roughness and (2)
the suitability of an existing analytical model, which considers
the influence of matric suction and interface dilatancy, for pre-
dicting interface shear strength.
Shear Strength of Unsaturated Soils
Earlier researchers have formulated shear strength equations (Bishop
1959; Fredlund and Morgenstern 1977; Lamborn 1986; Gan et al.
1988; Fredlund et al. 1996) based on different stress state variables,
such as net normal stress (NNS) and matric suction.
Vanapalli et al. (1996) proposed a simple and practical model
based on the soil-water characteristic curve (SWCC)
t f ¼ c0 þ ðs n  ua Þ tan w 0 þ ðua  uw Þf ðHk Þ tan w 0
where t f = shear strength; c0 = effective cohesion; w 0 = effective
friction angle; s n = total stress; ua = pore-air pressure; uw = pore-
water pressure; ðua  uw Þf = matric suction at failure; ðHk Þ = nor-
malized water content obtained from SWCC; and k is the soil-
specific parameter (ranging from 1.0 to 3.0).
Hossain (2010) modified Eq. (1) by including in it the dilatancy
induced by matric suction and has proposed the following equation:
t f ¼ c0 þ ðs n  ua Þ tan ð w 0 þ c Þ
þ ðua  uw Þf ðHk Þ tan ð w 0 þ c Þ
where c = dilation angle.
By considering the effect of the interface angle between soil and
steel, Eq. (2) can be rewritten as follows:
t f ¼ ca þ ðs n  ua Þf tan ðd 0 þ c i Þ
where c i = interface dilation; d 0 = effective interface friction angle
with respect to NNS angle; and
© ASCE
Int. J. Geomech., 2016, 16(6): D4016008
0.01 0.1 1 10
Particle size (mm)
ca ¼ c0a þ ðua  uw Þf tan d b (4)
where c0a = effective adhesion intercept; and d b = interface friction
angle relative to matric suction.
Eq. (3) is a logical extension of the extension of Eq. (2), but thus
far its validity has not been explored for soil-steel interfaces. As
mentioned earlier, very limited efforts have been made to explore
the shear strength of unsaturated soil interfaces.
Test Materials
Soil
Locally available completely decomposed granite (CDG) soil was
used in this study, and it has been characterized by following
British Standards Institution guidelines (BSI 1990a, b; 1999). The
(1) laboratory results show that the CDG soil has a specific gravity of
2.59, liquid limit of 31%, and plastic limit of 21%. The maximum
dry density (MDD) is noted to be 1.84 Mg/m3 with an optimum
moisture content (OMC) value of 13.4%. Fig. 1 shows the particle
size distribution curve of the CDG soil. The SWCC curve for the
CDG soil was obtained using the suction-controlled modified direct
shear apparatus (MDSA), at zero NNS, by following the procedure
proposed by Hossain (2010). Fig. 2 depicts the water retention char-
acteristics of the CDG soil along with the best fit curve based on the
model proposed by Fredlund and Xing (1994). The values of the
best fitting parameters (including k , proposed by Vanapalli et al.
(2) 1996) for SWCC are presented in Table 1.
Counterface
A square stainless steel plate with the dimensions 100  100  20
mm was used as the counterface material (Fig. 3). The normalized
(3) surface roughness (Rn) of the counterface is based on the roughness
profile proposed by Kishida and Uesugi (1987) in Eq. (5)
Rn ¼ Rmax =D50 (5)
D4016008-2 Int. J. Geomech.
 1.0
0.8
Normalized water content
0.6
0.4
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.
a = 18
0.2 n = 1.8
m= 0.29
Ψr = 3000
0.0
1 10
Fig. 2. Variation of the normalized water content with the matric suction at zero NNS
Table 1. Fitting Parameters for the SWCC of CDG Soil as shown in
Fig. 2
Parameter
Preload stress (kPa)
(ua − uw)r (kPa)
A 16
N 2.5
M 0.26
K 2.2
where Rmax = maximum peak to valley height; and D50 = grain size
(diameter) corresponding to 50% fines. In this study, D50 was found
to be 75 m m from the grain size distribution curve.
Two different types of counterfaces were used in this study
(INT-R and INT-M) with an Rmax value of 0.75 and 0.375 mm,
respectively. Using Eq. (5), the Rn equals 10 for INT-R and 5 for
INT-M.
Test Apparatus
Fig. 3 shows the schematic diagram and photographic view of the
MDSA used in this study. The testing device used in the present
study is based on the concepts of unsaturated soil testing, as
reported by Gan and Fredlund (1988). Nonetheless, to incorporate
interfaces, the basic concepts of unsaturated interface testing by
direct shear test apparatus, as reported by Miller and Hamid (2007),
have been used. Briefly, The MDSA consists of an air pressure
chamber, diffused air volume indicator (DAVI), high air entry po-
rous disk (HAEPD), a pressure/volume controller (for draining out
and measuring the volume of air bubbles in the connecting tubes),
measuring/monitoring devices such as LVDTs for monitoring hori-
zontal and vertical displacements, a load cell for measuring the hori-
zontal shear load, a pore-water pressure transducer, and an auto vol-
ume meter (AVC). The load cells and LVDTs are provided with the
manufacturer’s calibration data; nevertheless, for the purpose of
corroboration, the calibration of the load cell, LVDTs, DAVI, and
© ASCE
Int. J. Geomech., 2016, 16(6): D4016008
Best fit curve (Fredlund and Xing,1994)
Experimental result
100 1000 10000 100000 1000000
Matric suction (kPa)
AVC were performed before conducting the tests. All the necessary
corrections were applied for calculating and controlling the NNS,
Value matric suction, and the shear load, corresponding to different air
pressures in the chamber. For instance, an increase of pressure in
0
the air pressure chamber exerts an upward force through the vertical
3,000
loading ram; thus, to counteract this uplift force an equivalent sup-
plementary vertical load is applied. A top steel platen (for interface
testing consisting of a water chamber, HAEPD, and drainage lines)
is placed on top of the sample specimen. A small opening of 2 mm
in the surrounding area of the top steel platen and test cell creates a
provision for the air to access the soil specimen. High-pressure pol-
yvinylidene fluoride tubes 3 mm in diameter are used as drainage
lines. The DAVI device is used to quantify the volume of diffused
air, and the AVC device is used to examine the movement of water
to and from the specimen. Also, a GDS pressure/volume controller
flushing device (supplied by Geotechnical Digital System, U.K.)
was used for flushing air bubbles from the water chamber to all the
drainage lines. Pressure regulators and a pressure gauge (of 1-kPa
resolution, supplied by Wykeham Farrance International Ltd, U.K.)
were used to control and monitor the air pressure and water pressure
in the MDSA.
Test Procedure
The testing procedure consists of preliminary checks, sample prepa-
ration, and direct shear testing.
1. Preliminary inspection: Before starting the shear test, the
HAEPD was completely saturated and there was no leakage in
the air pressure chamber, valves, and drainage pipes by follow-
ing the guidelines presented by Hossain (2010).
2. Sample preparation: In this study, two types of interface speci-
mens with different counterface roughnesses were prepared in
addition to CDG soil specimens. Oven-dried soil, passing
through a 2-mm BS sieve, was matured for 48 h at OMC. The
matured soil was compacted in a layer of 10 mm (e.g., two
layers for interface specimen and four layers for soil specimen)
to achieve a target dry density of 1.75 Mg/m3 (95% of MDD).
The gap of 0.8 mm between the upper and lower half of the
D4016008-3 Int. J. Geomech.
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.

Fig. 3. Schematic diagram of the MDSA for testing a soil-steel interface (not to scale)

shear boxes was filled with grease and tightened together using
the screws.
3. Direct shear testing: Axis translation technique is one of the
most popular techniques adopted by researchers to control soil
suction (Vanapalli and Nicotera 2008), and the same has been
used in this study to achieve the target matric suction (by apply-
ing required air pressure in the pressure chamber and back pres-
sure in the water chamber, which is located on the top of the
ceramic disk). The process of unsaturated direct shear testing
can be divided in three stages: saturation, equilibration of the
specimen at a constant target NNS and matric suction (applied
sequentially to the specimen), and drained shearing at constant
target matric suction and NNS.
The shear box comprised of the compacted specimen was kept
in the pressure chamber, and a porous disk was placed on top of it.
The specimen was saturated by following the guidelines proposed
by Gan and Fredlund (1994). The heights of the specimen before
and after saturation soaking were recorded, and no appreciable
change was noticed. After achieving saturation, the chamber cap
was mounted and a predetermined axial load, air pressure, and

© ASCE D4016008-4 Int. J. Geomech.

Int. J. Geomech., 2016, 16(6): D4016008

 100
Soil
INT-R
INT-M
80

Volume of water (ml) 60

40
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.

20
Suction = 200kPa
NNS= 50 kPa

0
0 1000 2000 3000 4000 5000 6000
Time (min)

Fig. 4. Migration of water from the pure soil specimen and soil-steel interface specimens during the equilibration phase under a suction of 200 kPa
and NNS of 50 kPa

I - Initial state
IO - Saturation phase
OAB - Equilibration phase
BC - Shearing at constant suction and net normal stress
Shear stress

on
cs u ct i
M atri

B
O
Net A
nor
ma
l st
re s s

Fig. 5. Sketch of the loading path used for the suction-controlled direct shear tests

water pressure were applied sequentially to attain equilibration. The equilibration, the specimen was sheared at a rate of 0.004 mm/min
equilibration was presumed to be complete while the flow of water under drained conditions, until the horizontal displacement of 15
(Dw) draining out was ≥0.2%, as shown in Fig. 4. After attaining mm was reached.

© ASCE D4016008-5 Int. J. Geomech.

Int. J. Geomech., 2016, 16(6): D4016008

 Loading Path of Direct Shear Test Path BC exhibits the progression of the shearing phase under tar-
get NNS and matric suction.
The loading path adopted for the test is shown in Fig. 5, where
Point I marks the initial condition of the soil specimen before
achieving the saturation and Path IO shows the phase of satura-
Experimental Results and Discussion
tion. Path OA presents the phase of equilibration under the influ-
ence of target NNS, whereas Path AB represents the equilibration
To study the influence of counterface roughness on the interface
phase under matric suction. Point B indicates the conclusion of
shear strength, a series of direct shear tests was conducted on soil-
the equilibration phase and the start of the shearing phase. Last,
steel interface specimens using two counterface plates (made up of
stainless steel) designated as INT-R (Rn = 10) and INT-M (Rn = 5)

400
INT-R 400
Matric suction = 0 (kPa) INT-R
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.

Matric suction = 200 (kPa)

300
300 -NNS 300
Shear stress (kPa)

Shear strength (kPa)

-NNS 300

200
200
-NNS 150 -NNS 150

100
100
-NNS 050
-NNS 050

(a) 0
(a) 0
-0.06
-0.06

-0.04
-0.04

-0.02
-NNS 050
v / H0

-0.02
v / H0

-NNS 050 -NNS 150

0.00 -NNS 150

-NNS 300
0.00
-NNS 300

0.02
0.02

(b) 0.04 NNS 50 NNS 50

(b) 0.04
0.4 NNS 150 NNS 150
NNS 300 0.4
NNS 300
0.2
0.2

0.0
Δw (%)

0.0
-NNS 300
Δw (%)

-0.2 -NNS 150

-NNS 050
-NNS 050
-0.2
-NNS 150
-NNS 300
-0.4
-0.4

-0.6
0 2 4 6 8 10 12 14 16 -0.6
(c) u (mm) 0 2 4 6 8 10 12 14 16
(c) u (mm)
Fig. 6. Behavior observed during direct shear tests on the soil-steel
interface (INT-R) under a matric suction of 0 kPa and NNS of 0, 150, Fig. 7. Behavior observed during direct shear tests on the soil-steel
and 300 kPa: (a) variation of shear stress with horizontal displace- interface (INT-R) under a matric suction of 200 kPa and NNS of 0, 150,
ment; (b) variation of normalized vertical displacement with horizon- and 300 kPa: (a) variation of shear stress with horizontal displacement;
tal displacement; (c) variation of water content with horizontal (b) variation of normalized vertical displacement with horizontal dis-
displacement placement; (c) variation of water content with horizontal displacement

© ASCE D4016008-6 Int. J. Geomech.

Int. J. Geomech., 2016, 16(6): D4016008

 400 600
INT-M INT-M
Matric suction = 0 (kPa) Matric suction = 200 (kPa)
300

Shear strength (kPa)

400
Shear stress (kPa)

-NNS 300

200 -NNS 300

-NNS 150
200
-NNS 150
100

-NNS 050 -NNS 050

(a)
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.

(a) 0 0
-0.06 -0.06

-0.04 -0.04

-0.02 -0.02 -NNS 050

v / H0

v / H0
-NNS 050
-NNS 150
-NNS 150 -NNS 300
0.00 0.00
-NNS 300

0.02 0.02

NNS 50 NNS 50
(b) 0.04 (b) 0.04
NNS 150 NNS 150
0.4 0.4
NNS 300 NNS 300

0.0 -NNS 300

-NNS 150
0.0
Δw (%)

Δw (%)

-NNS 050
-0.4
-NNS 050

-NNS 150
-NNS 300
-0.4
-0.8

0 2 4 6 8 10 12 14 16 0 2 4 6 8 10 12 14 16
(c) u (mm) (c) u (mm)
Fig. 8. Behavior observed during direct shear tests on the soil-steel Fig. 9. Behavior observed during direct shear tests on the soil-steel
interface (INT-M) under a matric suction of 0 kPa and NNS of 0, 150, interface (INT-M) under a matric suction of 200 kPa and NNS of 0,
and 300 kPa: (a) variation of shear stress with horizontal displacement; 150, and 300 kPa: (a) variation of shear stress with horizontal dis-
(b) variation of normalized vertical displacement with horizontal dis- placement; (b) variation of normalized vertical displacement with
placement; (c) variation of water content with horizontal displacement horizontal displacement; (c) variation of water content with horizon-
tal displacement

under different matric suctions (of 0, 50, and 200 kPa) and different
NNS (50, 150, and 300 kPa). The typical results from experimental
interface tests are presented and discussed subsequently.
Load-Displacement and Volume Change Behavior
during Shearing
The influence of NNS on the interface shear strength can be exam-
ined from the curves of shear stress-horizontal displacement at a
constant value of matric suction. Figs. 6–9 illustrate the relationship
between (1) shear stress and horizontal displacement, (2) volumet-
ric strain (v/H0, where v is the vertical displacement and H0 is the
initial specimen height) during shearing, and (3) change in water
content (Dw) with horizontal displacement for INT-R and INT-M
specimens at different NSS of 50, 150, and 300 kPa and constant
matric suctions of 0 and 200 kPa. Some important observations
from experimental results are discussed in the following:
1. For zero suction, the curves of shear stress versus horizontal
displacement for INT-R and INT-M show a gentle hardening

© ASCE D4016008-7 Int. J. Geomech.

Int. J. Geomech., 2016, 16(6): D4016008

 behavior for all of the applied NNS. The INT-R has a higher
peak shear strength value compared with INT-M. Also, it is
noted that shear compression increases with NNS for both
interfaces.
2. For the suction value of 200 kPa, curves of INT-R and INT-M
show that the NNS has a noteworthy influence on the postpeak
shear strength. Particularly at lower NNS, a notable reduction
in postpeak shear strength is observed for both interfaces.
Moreover, with an increase in the value of NNS, a gradual
hardening stick-slip behavior is evidenced. It shows that at
higher NNS both interfaces tend to behave like a smooth
interface. Shear dilation is observed for both interfaces at lower
NNS, whereas shear compression is observed under higher
NNS.
3. For NNS 50 kPa, the peak shear strength increases with the
matric suction. The postpeak shear strength of interfaces INT-
R and INT-M is not affected much at a lower matric suction.
However, at a higher matric suction of 200 kPa, a greater
degree of strain softening behavior is evidenced for INT-R
compared with INT-M. At lower NNS shear behavior of both
interfaces is nearly similar; this shows that the contribution of
the counterface roughness to the interface shear is highly

400
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.

INT-R
350 INT-M
Linear Fit
300
Shear strength (kPa)

250

200

150

100

50

0
0 100 200 300 400
Net normal stress (kPa)

Fig. 10. Relationships between shear strength and NNS from direct shear tests on interface tests at 0-kPa matric suction

400

INT-R
350 INT-M
Linear fit
300
Shear strength (kPa)

250

200

150

100

50

0
0 100 200 300 400
Net normal stress (kPa)

Fig. 11. Relationships between shear strength and NNS from direct shear tests on interface tests at 50-kPa matric suction

© ASCE D4016008-8 Int. J. Geomech.

Int. J. Geomech., 2016, 16(6): D4016008

 500

INT-R
INT-M
400 Linear fit

Shear strength (kPa)

300

200
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.

100

0
0 100 200 300 400
Net normal stress (kPa)

Fig. 12. Relationships between shear strength and NNS from direct shear tests on interface tests at 200-kPa matric suction

Table 2. Peak Shear Strength Parameters of the Failure Envelope with Matric Suction

INT-R INT-M Soil

0 0 0
Matric suction (kPa) c0a (kPa) d þc ( ) o
r2
c0a (kPa) d þc ( ) o
r2 0
c (kPa) f þ c (o) r2
0 38.8 36.3 0.9971 33.0 31.8 0.9997 0 35.9 0.9734
50 39.1 39.9 0.9991 41.0 32.6 0.9991 34.8 41.3 0.9998
200 51.5 40.2 0.9998 64.0 33.6 0.9989 97.1 44.0 0.9997

dependent on the level of NNS. This primarily happens when
the interface dilates under lower NNS, possibly creating a vac-
uum condition within the soil fabric and as a result water enters
into the specimen.
4. For NNS of 150 and 300 kPa, the curves of shear stress versus
horizontal displacement for interfaces indicate gradual harden-
ing behavior at lower suction and partial hardening/softening
behavior at higher suction. The effect of suction on rough inter-
faces is noted to be more profound at lower and intermediate
NNS.
5. The increment in the interface shear strength with respect to the
normalized counterface roughness varies nonlinearly with
change in stress state variables. In other words, for a specific
constant NNS and matric suction, the increment in the value of
interface shear stress with respect to the counterface roughness
is nonlinear.
6. The total volume change behavior for the rough interfaces is
noted to be significantly influenced by the counterface rough-
ness. For a specific constant NNS and matric suction, the incre-
ment in the value of shear dilation (or shear compression) with
respect to the counterface roughness is nonlinear. Nonetheless,
the overall degree of dilation (or compression) is directly influ-
enced by the stress state variables. At higher NNS, the overall
volume change is lower primarily because of the lower dila-
tancy and greater diffused air production, particularly at higher
suction values.
7. During shearing, disruption and probably ruptures are likely to
happen to the contractile skin, resulting in an increase of the
pore-water pressure, draining water out from the specimen.
Water drained out of the specimen was observed to be in the
range of 0.12–0.46%.
Failure Envelopes for the Interfaces and the Soil
Figs. 10–12 show the variation of the peak shear stress versus NNS,
for interface tests, under different matric suction values of 0, 50,
and 200 kPa. It must be noted that area correction is applied to cal-
culate the shear stress. Based on the raw test data, the failure crite-
rion was considered as the point at which the shear load starts
decreasing or remains reasonably constant compared with the peak
shear load. From the experimental results, it is noted that the appa-
rent interface friction angle d max and adhesion intercept ca
increases with matric suction. The variation of the apparent friction
angle is probably caused by the change in dilation angle with matric
suction. Interestingly, this observation is in agreement with previ-
ously published observations for soil-cement grout interface
(Hossain and Yin 2012). Furthermore, change in the value of adhe-
sion intercept is caused by a change of suction and d b angle. The
values of w max ¼ w 0 þ c , d max ¼ d 0 þ c , and adhesion (ca ) or
cohesion intercepts c0 , as defined by Eqs. (3) and (4), for different
suction values of INT-R, INT-M, and pure soil, are presented in
Table 2.
Figs. 13 and 14 show the variation of interface shear strength
t f with matric suction ðua  uw Þf (suction envelope) for INT-R
and INT-M. The d b parameter for the interface shear test, irre-
spective of the counterface roughness, exhibits a nonlinear

© ASCE D4016008-9 Int. J. Geomech.

Int. J. Geomech., 2016, 16(6): D4016008

 500
INT-R
NNS 50 kPa
NNS 150 kPa
400 NNS 300 kPa

Shear strength (kPa)

300

200
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.

100

0
0 50 100 150 200 250 300
Matric Suction (kPa)

Fig. 13. Relationship between shear strength and matric suction from direct shear tests on INT-R for different NNS

600
INT-M
NNS 50 kPa
500 NNS 150 kPa
NNS 300 kPa

400
Shear strength (kPa)

300

200

100

0
0 50 100 150 200 250 300
Matric Suction (kPa)

Fig. 14. Relationship between shear strength and matric suction from direct shear tests on INT-M for different NNS

Table 3. Value of d b with Matric Suction for INT-R Table 4. Value of d b with Matric Suction for INT-M

Matric suction (kPa) d b (°) Matric suction (kPa) d b (°)

0 36.3 0 31.8
50 0.3 50 9.1
200 3.6 200 8.1

relationship with suction (Vanapalli et al. 1996; Hamid and
Miller 2009). The values of d b for INT-R and INT-M obtained
from Eq. (5) under different matric suctions are presented in
Tables 3 and 4, respectively. The analytical values of interface di-
lation angles and apparent friction angle for different matric
suction obtained from volume change behavior curves are sum-
marized in Tables 5 and 6.
Fig. 15 shows the variation of interface shear strength with
respect to matric suction and normalized counterface roughness at
different NNS. As mentioned earlier, it is reaffirmed that the

© ASCE D4016008-10 Int. J. Geomech.

Int. J. Geomech., 2016, 16(6): D4016008

 Table 5. Values of Dilation Angle and Apparent Friction Angle for Different Matric Suctions Obtained from the Volume Change Behavior Curves for
INT-R

Matric suction NNS Interface dilation Average dilation Effective adhesion Effective interface Apparent interface friction
(kPa) (kPa) angle (°) angle c i (°) c0a (°) friction angle d 0 (°) angle d max ¼ ðd 0 þ c i Þ (°)
0 50 2.7 0.9 38.8 35.4 36.3
150 0
300 0
50 50 3.7 2.4 38.8 35.4 37.8
150 1.9
300 1.7
200 50 10.7 7.9 38.8 35.4 43.3
150 8.4
300 4.6
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.

Table 6. Values of Dilation Angle and Apparent Friction Angle for Different Matric Suctions Obtained from the Volume Change Behavior Curves for INT-M

Matric suction NNS Interface dilation Average dilation Effective adhesion Effective interface Apparent interface friction
(kPa) (kPa) angle (°) angle c i (°) c0a (kPa) friction angle d 0 (°) angle d max ¼ ðd 0 þ c i Þ (°)
0 50 4.6 2.2 33.0 29.6 31.8
150 2.1
300 0
50 50 5.0 2.6 33.0 29.6 32.3
150 2.5
300 0.3
200 50 9.2 6.5 33.0 29.6 36.2
150 6.7
300 3.6

400
INT-R
INT-M
NNS 300 (kPa)

300
NNS 300 (kPa)
)
Shear strength (kPa

200 NNS 150 (kPa)

NNS 150 (kPa)

NNS 50 (kPa)
100 NNS 50 (kPa)

0 12
0 10
50 8 ss
100 h ne
6
ro ug
S uc 150
ti on
4
z ed
(k P
200
2 ali
a) 250 rm
0 No
300

Fig. 15. Variation of interface shear strength with respect to matric suction and normalized counterface roughness at different NNS

average dilation angle increases with matric suction for both the
interfaces irrespective of the counterface roughness. Furthermore,
for a specific suction, the interface dilation angle increases nonli-
nearly for the normalized surface roughness values used in the pres-
ent study. Nonetheless, it would be worthwhile to study the varia-
tion of interface dilation angle and interface shear strength with
respect to suction for a Rn value above 10 and below 5, especially
because the optimum normalized counterface roughness, which
possesses the maximum interface shear strength (under specific
stress state condition), varies with the applied matric suction. The
value of optimum Rn is found to be nonuniform and is significantly
influenced by the stress state variables.

© ASCE D4016008-11 Int. J. Geomech.

Int. J. Geomech., 2016, 16(6): D4016008

 600
INT R Experimental
NNS 50 (kPa)
NNS 150 (kPa)
500 NNS 300 (kPa)

Hossain (2010)
NNS 50 (kPa)
400

Shear Stregth (kPa)

NNS 150 (kPa)
NNS 300 (kPa)

300

200
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.

100

0
0 50 100 150 200 250 300
Matric Suction (kPa)

Fig. 16. Comparison between experimental interface shear strength data for INT-R and analytical results obtained from the Hossain (2010) model

600
INT M Experimental
NNS 50 (kPa)
NNS 150 (kPa)
500
NNS 300 (kPa)

Hossain (2010)
NNS 50 (kPa)
Shear strength (kPa)

400 NNS 150 (kPa)

NNS 300 (kPa)

300

200

100

0
0 50 100 150 200 250 300
Matric Suction (kPa)

Fig. 17. Comparison between experimental interface shear strength data for INT-M and analytical results obtained from the Hossain (2010) model

Comparison of Experimental Results with Existing
Shear Strength Model
A modified model for predicting the interface shear strength
between CDG soils and cement grout, by considering the effect of
interface dilation angle on apparent interface friction angle, was
proposed by Hossain (2010). However, this model has not been tes-
tified or verified to consider the shear strength between soil-steel
interfaces, especially with different counterface roughness.
Figs. 16 and 17 show a comparison between experimental shear
strength data for INT-R and INT-M specimens and analytical
results obtained from Eq. (3) using effective interface shear strength
0
parameters (c0a and d ) of soil-steel interface at saturated condition,
and analytical values of interface dilation angle c i and d b angles
under different suctions (refer to Tables 5 and 6). It is noted that the
interface shear strength predicted from the modified model from
Hossain (2010) agrees well with the experimental shear strength
data for both interfaces (INT-R and INT-M) under different NNS
and matric suctions. In particular, the study shows that the analyt-
ical model successfully considers the nonlinear behavior of the
interfaces, especially at lower NNS and higher matric suctions.
Interestingly, the shear strength behavior appears to be roughly
coinciding with the air entry value of the soil (as seen in Fig 2). In
other words, the nonlinear shear strength behavior is governed by

© ASCE D4016008-12 Int. J. Geomech.

Int. J. Geomech., 2016, 16(6): D4016008

 600
Soil
500
Shear strength (kPa)
400
300
200
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.
100
0
0 50
Fig. 18. Comparison between experimental interface shear strength data for soil and analytical results obtained from the Hossain (2010) model
the rate (or increase in the value) of air entering the soil fabric. At
lower suction values (e.g., 20 kPa), it is noted (see Figs. 2, 16, and
17) that the interface shear strength is nearly linear probably from
lack of air entering the soil fabric. However, with an increase of air
entering the soil fabric, the nonlinear behavior is pronounced. A sim-
ilar nonlinear shear strength behavior with respect to matric suction
has been observed for the pure soil, and the experimental data
closely matched with the analytical model, as depicted in Fig. 18.
Nevertheless, it would be interesting to see the prediction of the ana-
lytical model for relatively smoother interfaces. The experimental
results reaffirm that interface dilation has significant influence on
apparent friction angle and on interface shear strength. Thus, it is im-
portant to consider the interface dilation while designing and assess-
ing safety of different soil-structure interactions.
Conclusions
Based on the experimental results of single staged direct shear tests
performed on soil-steel interface and pure soil in saturated and un-
saturated states, the following conclusions can be drawn:
• Matric suction and NNS significantly influence the hardening/
softening behavior of soil-steel interfaces and soil. However,
the degree of influence varies depending on the counterface
roughness. The influence of NNS on the hardening/softening
behavior becomes more profound with an increase in suction.
• Counterface roughness significantly influences the shear
behavior and interface shear strength for specific matric suc-
tion and NNS. The interface with greater counterface rough-
ness (INT-R) gains a higher peak shear strength value due to
larger soil dilation. The variation in interface shear strength for
specific stress state variables and counterface roughness is
nonlinear.
• The variation of the d b angle is inversely proportional to the
matric suction, and the interface dilatancy increases with
matric suction and counterface roughness. The soil dilatancy
for both interfaces is at its peak, particularly under higher suc-
tions and lower NNS.
© ASCE
Int. J. Geomech., 2016, 16(6): D4016008
Experimental
NNS 50 (kPa)
NNS 150 (kPa)
NNS 300 (kPa)
Hossain (2010)
NNS 50 (kPa)
NNS 150 (kPa)
NNS 300 (kPa)
100 150 200 250 300
Matric Suction (kPa)
• The interface dilation has a significant impact on the interface
shear strength. The experimental data for both interfaces (INT-
R and INT-M) and soil agree well with values predicted from
the analytical model from Hossain (2010). It is important to
consider the impact of dilation on the interface shear strength
while assessing or designing interfacial interaction related to
geotechnical applications.
• The optimum normalized counterface roughness that pos-
sesses maximum interface shear strength is not unique and is
closely related to the applied stress state variables.
Acknowledgments
The work in this paper is supported by a National State Key
Project “973” grant (No/2014CB047000) (subproject No.
2014CB047001) from the Ministry of Science and Technology
of the People’s Republic of China, PolyU Shenzhen Research
Institute, and Hong Kong Polytechnic University, China.
References
Bishop, A. W. (1959). “The principle of effective stress.” Lecture delivered
in Oslo, Norway, in 1955.” Tecknisk Ukeblad I Samarbeide Med
Teknikk, 106(39), 859–863.
Borana, L., Yin, J. H., Singh, D. N., and Shukla, S. K. (2015). “A modified
suction controlled direct shear device for testing unsaturated soil and
steel plate interface.” Mar. Georesour. Geotechnol., 33(4), 289–298.
Boulon, M. (1989). “Basic features of soil structure interface behaviour.”
Comput. Geotech., 7(1), 115–131.
BSI (British Standards Institution). (1990a). “Methods of test for soils for
civil engineering purposes.” BS 1377, London.
BSI (British Standards Institution). (1990b). “Methods of test for soils for
civil engineering purposes—Part 2: Classification tests.” BS 1377-2,
London.
BSI (British Standards Institution). (1999). “Code of practice for site inves-
tigations.” BS 5930, London.
Escario, V., and Sáez, J. (1986). “The shear strength of partly saturated
soils.” Geotechnique, 36(3), 453–456.
D4016008-13 Int. J. Geomech.
 Fredlund, D. G., and Morgenstern, N. R. (1977). “Stress state variables for
unsaturated soils.” J. Geotech. Eng. Div., 103(GT5), 447–466.
Fredlund, D. G., Morgenstern, N. R., and Widger, R. A. (1978). “The shear
strength of unsaturated soils.” Can. Geotech. J., 15(3), 313–321.
Fredlund, D. G., and Rahardjo, H. (1993). Soil mechanics for unsaturated
soils, Wiley, New York.
Fredlund, D. G., and Xing, A. (1994). “Equations for the soil-water charac-
teristic curve.” Can. Geotech. J., 31(4), 521–532.
Fredlund, D. G., Xing, A., Fredlund, M. D., and Barbour S. L. (1996). “The
relationship of the unsaturated soil shear strength to the soil-water char-
acteristic curve.” Can. Geotech. J., 33(3), 440–448.
Gan, J. K. M., and Fredlund, D. G. (1988). “Multistage direct shear testing
of unsaturated soils.” Geotech. Test. J., 11(2), 132–138.
Gan, J. K. M., and Fredlund, D. G. (1994). “Direct shear and triaxial
testing of a Hong Kong soil under saturated and unsaturated condi-
Downloaded from ascelibrary.org by Universidad De Lima on 04/18/23. Copyright ASCE. For personal use only; all rights reserved.
tion.” GEO Rep. No. 46, Geotechnical Engineering Office, Hong
Kong.
Gan, J. K. M., Fredlund, D. G., and Rahardjo, H. (1988). “Determination of
the shear strength parameters of an unsaturated soil using the direct
shear test.” Can. Geotech. J., 25(3), 500–510.
Hamid, T. B., and Miller, G. A. (2009). “Shear strength of unsaturated soil
interfaces.” Can. Geotech. J., 46(5), 595–606.
Hossain, M., and Yin, J. (2012). “Influence of grouting pressure on the
behavior of an unsaturated soil-cement interface. J. Geotech.
Geoenviron. Eng., 10.1061/(ASCE)GT.1943-5606.0000585, 193–202.
Hossain, M., and Yin, J. (2014). “Dilatancy and strength of an unsaturated
soil-cement interface in direct shear tests.” Int. J. Geomech., 10.1061
/(ASCE)GM.1943-5622.0000428, 04014081.
Hossain, M. A. (2010). “Experimental study on the interface behavior
between unsaturated completely decomposed granite soil and
cement grout.” Ph.D. thesis, Hong Kong Polytechnic Univ., Hong
Kong.
Khoury, C. N., Miller, G. A., and Hatami, K. (2010). “Unsaturated soil geo-
textile interface behavior.” Geotext. Geomembr., 29, 613–624.
Kim, M., and O’Neill, M. (1998). “Side shear induced in drilled shaft by
suction change.” J. Geotech. Geoenviron. Eng., 10.1061/(ASCE)1090
-0241(1998)124:8(771), 771–780.
Kishida, H., and Uesugi, M. (1987). “Tests of the interface between sand
and steel in the simple shear apparatus.” Geotechnique, 37(1), 45–52.
Lamborn, M. J. (1986). “A micromechanical approach to modeling partly
saturated soils.” M.Sc. thesis, Texas A & M Univ., College Station, TX.
© ASCE D4016008-14
Int. J. Geomech., 2016, 16(6): D4016008
Mashhour, M. M., Ibrahim, M. I., and El-Emam, M. M. (1996). “Variation
of unsaturated soil shear strength parameters with suction.” Proc., First
Int. Conf. on Unsaturated Soils, Balkema, Rotterdam, Netherlands,
1487–1493.
Miller, G. A., and Hamid, T. B. (2007). “Interface direct shear testing of un-
saturated soil.” Geotech. Test. J., 30(3), 182–191.
Ng, C. W. W., and Menzies, B. (2007). Advanced unsaturated soil mechan-
ics, Taylor &Francis, Abingdon, U.K.
Okamoto, M., and Katayama, T. (2015). “Capability and applicability of
steel sheet pile cellular structure for Hong Kong.” Technical Seminar on
Application of Steel Cellular Structure for Reclamation in Hong Kong.
Pei, H., Yin, J., Zhu, H., Hong, C. (2013). “Performance monitoring of a
glass fiber-reinforced polymer bar soil nail during laboratory pullout test
using FBG sensing technology.” Int. J. Geomech., 10.1061/(ASCE)GM
.1943-5622.0000226, 467–472.
Potyondy, J. G. (1961). “Skin friction between various soils and construc-
tion materials.” Geotechnique, 11(4), 339–353.
Rao, B. H., Venkataramana, K., and Singh, D. N. (2011). “Studies on the
determination of swelling properties of soils from suction measure-
ments.” Can. Geotech. J., 48(3), 375–387.
Sharma, J. S., Fleming, I. R., and Jogi, M. B. (2007). “Measurement of un-
saturated soil–geomembrane interface shear-strength parameters.” Can.
Geotech. J., 44(1),78–88.
Uesugi, M., Kishida, H., and Uchikawa, Y. (1990). “Friction between dry
sand and concrete under monotonic and repeated loading.” Soils Found.,
30(1), 115–128.
Vanapalli, S. K., Fredlund, D. G., Pufahl, D. E., and Clifton, A. W. (1996).
“Model for the prediction of shear strength with respect to soil suction.”
Can. Geotech. J., 33(3), 379–392.
Vanapalli, S. K., and Nicotera, M. V. (2008). “Axis translation and negative
water column techniques for suction control.” Geotech Geol. Eng., 26,
645–660.
Yoshimi, Y., and Kishida, T. (1981). “Friction between sand and metal sur-
face.” Proc., 10th Int. Conf. on Soil Mechanics and Foundation
Engineering, Vol.1, Balkema, Rotterdam, Netherlands, 831–834.
Yu, F., and Yang, J. (2011). “Mechanism and assessment of interface shear
between steel pipe pile and sand.” Proc., Advances in Pile Foundations,
Geosynthetics, Geoinvestigations, and Foundation Failure Analysis and
Repairs, 56–64.
Zhan, L. T., and Ng, C. W. W. (2006). “Shear strength characteristics of an
unsaturated expansive clay.” Can. Geotech. J., 43(7), 751–763.
Int. J. Geomech.