Engineering Structures 141 (2017) 596606

Contents lists available at ScienceDirect

Engineering Structures
journal homepage: www.elsevier.com/locate/engstruct

Cyclic tests on slip resistance of squat heavyweight concrete shear walls

with construction joints
Keun-Hyeok Yang a,, Ju-Hyun Mun b, Yong-Ha Hwang c, Jin-Kyu Song d
a
Department of Plant
Architectural Engineering, Kyonggi University, Suwon, Kyonggi-do, Republic of Korea
b
Technical Research Center, GL Construction Co., Ltd., Gwangju, Jeonnam, Republic of Korea
c
Department of Architectural Engineering, Kyonggi University Graduate School, Seoul, Kyonggi-do, Republic of Korea
d
Department of Architectural Engineering, Chonnam National University, Gwangju, Jeonnam, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: The objective of this study was to examine the effectiveness of different slip-resistance bars (X-, W-, and
Received 19 September 2016 O-shaped bars) on reducing sliding shear displacement at the base interface of squat heavyweight con-
Revised 26 March 2017 crete (HWC) shear walls with construction joints. In addition, the structural applicability of welded wire
Accepted 27 March 2017
fabric (WWF) reinforcement was investigated as an alternative to the individual deformed bars conven-
tionally used for the shear reinforcement of shear walls. All the wall specimens had the same concrete dry
density of 3380 kg/m3, geometrical dimensions with barbell-shaped cross-sections, and shear ratio of 1.0.
Keywords:
The ultimate failure of the walls subjected to under constant axial loads and reversed cyclic lateral loads
Squat shear wall
Heavy-weight concrete
was associated with a kinking in the longitudinal and vertical shear bars and splitting of the surrounding
Sliding shear concrete at the base interface due to the dowel resistance of the bars. The slip-resistance bars were effec-
Construction joint tive in reducing the shear slip displacement at the base interface after the peak load of the walls was
Slip resistance reached. The WWF reinforcement performed comparably to the conventional shear reinforcements in
enhancing the shear strength of the squat shear walls and preventing a rapid decrease in the applied
loading after the peak strength. The shear strength of the squat HWC shear walls can be conservatively
estimated using equations proposed by EC8 for earthquake resistance, even for avoiding sliding shear fail-
ure at high wall drift ratios.
2017 Elsevier Ltd. All rights reserved.

1. Introduction buildings and safety-related nuclear structures. In particular,

As high-density materials are known to be effective in attenuat-
ing and absorbing high-energy X-rays, gamma rays, and neutrons,
heavyweight concrete (HWC) is used for shielding against radioac-
tive rays in the walls and roofs of building structures requiring
impenetrability such as nuclear power plants, medical units, and
accelerator laboratories [1,2]. Experimental data [3,4] have
indicated that HWC is beneficial for enhancing the modulus of
elasticity and shear strength. However, it is not easy to find data
that verify the structural advantages of HWC elements, although
thorough evaluation of the structural performance including ser-
viceability, strength, and ductility is essential for the rational
design of concrete members.
Reinforced concrete squat shear walls with a shear ratio below
2.0 are frequently applied to provide seismic resistance to low-rise
Corresponding author.
E-mail addresses: yangkh@kyonggi.ac.kr (K.-H. Yang), mjh352002@nate.com
(J.-H. Mun), hyongha89@gmail.com (Y.-H. Hwang), jgsong@jnu.ac.kr (J.-K. Song).
nuclear structures are likely to be subjected to multiple cycles of
loading to peak strength in safe shutdown earthquake shaking.
The predominant performance of squat shear walls is shear rather
than flexure because most of the applied lateral loads are trans-
ferred through the concrete strut action at the wall web to the
base. Furthermore, sliding displacement along the interface
between the wall and base becomes more important at a high level
of inelasticity after the ultimate strength of the squat walls is
reached [5]. Hence, sufficient experimental data on the shear
strength and slip resistance of squat HWC shear walls should be
accumulated for the reliable design of HWC shielding structures.
However, available studies [6,7] on the cyclic shear behavior and
sliding resistance of squat shear walls are very rare, even for
normal-weight concrete (NWC). The ACI 318-14 guideline [8] for
the shear strength of squat shear walls is based on empirical
expressions for the shear transfer capacities of concrete and shear
reinforcements originally established for slender beam tests [9].
The shear transfer capacity of horizontal shear reinforcement is
formulated using the 45 truss analogy derived from the lower

http://dx.doi.org/10.1016/j.engstruct.2017.03.054
0141-0296/ 2017 Elsevier Ltd. All rights reserved.
 K.-H. Yang et al. / Engineering Structures 141 (2017) 596606 597

Notations

Ag cross-sectional area of walls V fl flexural strength of walls

As area of longitudinal reinforcement at the boundary ele- V fr shear friction strength at the base interface
ments Vn shear strength of walls
As1 net area of reinforcing bars V Rd3 diagonal shear tension strength of walls predicted by
As;i area of diagonal bars crossed with the base interface EC8
Av area of vertical shear reinforcement at the wall web V Rd2 diagonal shear compression strength of walls predicted
bw thickness of the wall web by EC8
dr drift ratio of walls V Rd;s sliding shear strength of walls predicted by EC8
Es modulus of elasticity of reinforcing bars Wc work damage indicators calculated up to the failure of
0
fc compressive strength of concrete walls
fy yield strength of longitudinal reinforcement z lever arm between tensile reinforcement and the center
f yh yield strength of shear reinforcement of compression zone
f yh;i yield strength of diagonal bars crossed with the base as shear ratio of walls
interface D lateral displacement measured at the line of action of
f su tensile strength of longitudinal reinforcement the lateral load
hl distance from the base interface of the wall to the lateral ey yield strain of reinforcing bars
load line h angle between the concrete compression strut and the
hw height of walls axis perpendicular to the shear force
li distance between the centers of the axes of diagonal l friction coefficient of concrete
reinforcement at the wall base m1 strength reduction factor for concrete cracked in shear
lw length of walls n normalised neutral axis depth.
M cr initial flexural cracking moment qh horizontal shear reinforcement ratio
Mn nominal momment capacity of walls, ql longitudinal reinforcement ratio
Nu applied axial load qv vertical shear reinforcement ratio
T cfr concrete friction resistance predicted by Salonikios ud nominal diameter of reinforcing bars
Td dowel action resistance of longitudinal reinforcement u inclination of diagonal bars crossed with the base inter-
predicted by Salonikios face
V cd shear transfer capacity of concrete
V cr initial shear cracking strength of walls

bound theorem of concrete plasticity, whereas no specific com-
ments on the shear transfer capacity of vertical shear reinforce-
ment (longitudinal bars in the wall web). Furthermore, the shear
friction design equations specified in ACI 318-14 ignore concrete
cohesion and assume that the applied shear force is entirely trans-
ferred by the friction action of transverse reinforcement. The shear
friction equations of ACI 318-14 are derived from the regression
analysis using test data subjected to pure shear stresses. Overall,
sliding resistance at the base interface of squat walls simultane-
ously subjected to moment, shear, and axial force are still ambigu-
ous in the design specification. Against sliding shear failure of
squat walls, EC8 [10] requires a definite calculation of the slip
resistance, for which the mechanism includes the dowel action of
vertical and diagonal bars and frictional action of concrete at the
base interface. Previous studies [5,6] also indicate that diagonal
shear reinforcement is effective in improving the seismic response
of squat shear walls but is insufficient for preventing a large slip
displacement at the base interface.
Walls commonly have construction joints at the base interface
because of the separate placement of concrete between the wall
and foundation at different ages. Based on push-off tests, Yang
et al. [11] demonstrated that the failure mechanism in construc-
tion interfaces is similar to that in monolithic ones, whereas the
sliding resistance of smooth construction interfaces decreased con-
siderably because of the absence of aggregate interlock and the
reduced concrete cohesion along the interface. This finding implies
that the contribution of slip to the lateral displacement of squat
shear walls would be greater for walls with construction interfaces
than for monolithic walls, particularly at a high inelasticity state of
the wall. However, most of the specimens [1214] prepared to
examine the cyclic shear performance of squat walls were manu-
factured monolithically with foundation, and insignificant exami-
nation of the slip shear displacement along the base interface
was performed. Furthermore, few tests [5,6] on the shear sliding
resistance of squat shear walls were conducted using monolithic
specimens. Experimental work on squat shear walls with construc-
tion joints remains insufficient, particularly where a shear and/or
sliding shear mode dominates, even for NWC shear walls.
This study tested six HWC squat shear walls with construction
joints to examine the effectiveness of different slip-resistance bars
on reducing sliding shear displacement at the base interface under
a high inelastic state of walls. To prepare HWC squat shear wall
specimens with unit weights of 3380 kg/m3, magnetite particles
were used for the concrete aggregates, and the shear ratio of the
wall was fixed at 1.0. The contribution of the sliding shear displace-
ment component to the lateral displacement of the walls was
determined at each lateral displacement increment of the walls.
The measured shear strength was compared with predictions
obtained using the empirical equations of Salonikios [5] and EC8
[10], which were proposed to avoid sliding shear failure at a high
wall inelasticity level.
2. Experimental test program
2.1. Details of wall specimens
Table 1 provides a summary of the test parameters selected for
the squat HWC shear walls in this study. The dimensional details
and configuration of the reinforcing bars in the boundary elements
and web of the wall specimens are also shown in Fig. 1. The main
parameters investigated were the details of the slip-resistance bars
at the interface between the wall and bottom stub. Specimen C had
no slip-resistance bars in the base interface. Specimens X, W, and O
were designed to examine the effect of the configuration of the
598 K.-H. Yang et al. / Engineering Structures 141 (2017) 596606

Table 1
Details of shear wall specimens.
0
Specimen f c (MPa) Shear reinforcement at the web Slip resistance bars (U10) at the base interface
Type qv qh
C 20.8 Individual deformed bars 0.0042 (U10@200) 0.0042 (U10@200)
X X-shaped bars
W W-shaped bars
O O-shaped bars
F Welded wire fabric bars 0.0036 (U7.6@150) 0.0036 (U7.6@150)
FX X-shaped bars

Note: f0 c = concrete compressive strength, qv = vertical shear reinforcement ratio, and qh = horizontal shear reinforcement ratio.

Fig. 1. Specimen details and arrangement of reinforcing bars (all dimensions are in mm).

slip-resistance bars at the base interface, with X-shaped, W-
shaped, and O-shaped bars, respectively, arranged at the base
interface of the specimens. The above four specimens had identical
vertical and horizontal shear reinforcements at their web using 10-
mm-diameter deformed bars arranged with a spacing of 200 mm
in each direction, which formed two curtains with a ratio of
0.0042. The X- and W-shaped bars passed through the boundary
element of the wall and bottom stub for suitable anchorage. The
legs of the O-shaped bars were anchored in the bottom stub. The
intersection point of the X-shaped bars crossed using diagonal bars
with 45 inclination was located at the center of the base interface.
Specimens F and FX were prepared to examine the structural appli-
cability of welded wire fabric (WWF) reinforcement as an alterna-
tive to the individual deformed bars conventionally used for the
shear reinforcement of shear walls. The diameter and mesh size
of the WWF deformed reinforcing bars were 7.6, and 150 mm,
respectively, producing vertical (qv ) and horizontal (qh ) shear rein-
forcement ratios of 0.0036 in each direction. The amount of WWF
reinforcement in the F and FX walls was 15% lower than that of the
conventional shear reinforcement using 10-mm-diameter bars in
wall C. Except for the shear reinforcement details in the web of
the wall, specimens F and FX were identical to the specimens C
and X, respectively, in terms of their dimensions and slip-
resistance bar arrangements. Note that the amount of shear rein-
forcement in the specimens was determined considering the con-
ventional arrangement with the ratios between 0.003 and 0.005
in shear walls governed by shear failure as well as the minimum
requirements and maximum spacing for shear reinforcement spec-
ified in the seismic design guidelines of ACI 318-14 [8].
All of the specimens had the same geometric dimensions, as
shown in Fig. 1. The walls had a barbell-shaped cross section
including the boundary elements with a square section of
250 mm in dimension and the web with a thickness of 170 mm.
The overall length (lw ) and net height (hw ) of the wall were 1200
and 900 mm, respectively. The walls were integrated with a
2200  700  600 mm top stub and a 2200  800  500 mm bot-
tom stub. The distance from the base interface to the central axis
of the top stub was 1200 mm, resulting in a shear ratio (as ) of
1.0. To guide the structural behavior of the walls to be governed
by shear prior to flexural yielding failure, predictions of the flexural
 K.-H. Yang et al. / Engineering Structures 141 (2017) 596606
(V fl ) and shear (V n ) strengths of the walls were compared accord-
ing to the variation of the amount of longitudinal reinforcement
at the boundary elements. The nominal predictions for V fl and V n
of each specimen were determined using the equivalent stress
block described in ACI 318-14 and the strut-and-tie model (STM)
proposed by Mun and Yang [15], respectively. From the compar-
isons of the predictions to induce shear-governed failure, eight
deformed bars with a diameter of 13 mm were symmetrically
arranged for longitudinal reinforcement at each boundary element,
yielding a longitudinal reinforcement ratio of 0.016. The longitudi-
nal reinforcement was confined by lateral hoops using 6-mm
round bars with a spacing of 60 mm. The amounts of shear rein-
forcement at the web and lateral hoops at the boundary element
met the demand for the minimum requirements of ACI 318-14
guidelines. For the slip-resistance bars at the base interface, 10-
mm-diameter deformed bars were used. All of the specimens also
had a smooth construction joint at the base interface. To produce
smooth construction joints, the concrete for the wall elements
was cast with no special treatment after a curing age of the bottom
stub concrete of seven days.
2.2. Material properties
To obtain well-graded heavyweight coarse and fine aggregates,
locally available magnetite ores were crushed and graded into 10
different sizes using a sieve shaker apparatus. The crushed mag-
netite particles had rough and irregular external surface with
fewer voids. Yang et al. [16] noted that the lower porosity volume
of magnetite aggregates results in a lower absorption and provides
a stronger restraint against the shrinkage of concrete. The water
absorption of magnetite particles was recorded to be approxi-
mately 0.6%, which is considerably lower than that of conventional
normal-weight aggregates. Magnetite particles with maximum
sizes of 5 and 25 mm were used for fine and coarse aggregates,
respectively, to produce HWC with a dry unit weight of 3380 kg/
m3. The particle distribution of aggregates used herein showed
continuous grading and satisfied the standard distribution curves
recommended in ASTM C33/C33M [17], yielding a fineness modu-
lus of 6.7 for the coarse aggregates and 3.4 for the fine aggregates.
Ordinary Portland cement conforming to ASTM Type I [17] was
used as the main binder for the HWC mixes. When HWC is used
for shielding, the density and materials costs are of prime impor-
tance because the shielding capacity of a concrete element depend
on member thickness and material density. Furthermore, high-
strength HWC with high cement content and low water-to-
cement ratio may exhibit increased creep and shrinkage, and in a
massive concrete placement could generate high temperatures at
early ages causing undesirable thermal cracking. The code provi-
sions [1,2] specify the minimum compressive strength of concrete
to be 21 MPa as a structural concrete for nuclear facilities. Hence,
the design compressive strength of HWC was selected to be
0
21 MPa. Table 1 gives the compressive strength (f c ) of the concrete
measured using 100 mm  200 mm cylindrical molds at an age of
wall tests.
Table 2
Mechanical properties of reinforcement.
Type /d (mm) As1 (mm2)
Deformed bar 6 28.26
10 71.0
13 127.0
WWF 7.6 45.34
Note: Ud = nominal diameter, As1 = net area, fy = yield strength, fsu = tensile strength, Es = modulus of elasticity, and ey = yield strain.
*
The yield strength of 6 mm bar and 7.6 mm WWF was determined using 0.2% offset method.
599
The mechanical properties of the reinforcing bars and WWF
bars used are given in Table 2. The yield strengths of the 6-mm-
diameter deformed bars and 7.6-mm-diameter bars used for
WWF reinforcement were determined using the 0.2% off set
method. The yield strengths of the deformed bars with diameters
of 6, 10, and 13 mm were 498, 508, and 475 MPa, respectively.
The yield and tensile strengths of the 7.6-mm-diameter bars used
for manufacturing WWF reinforcement were 566 and 611 MPa,
respectively. The modulus of elasticity approached 200,000 MPa
for all the reinforcing bars.
2.3. Test procedure
To simulate the cantilever behavior of a wall with full fixity at
the end, the bottom stub of each wall was fixed to the test bed
using eight steel rock bolts of 50-mm-diameter penetrating the
stub and two steel angles on both sides of the stub, as illustrated
in Fig. 2. All of the specimens were tested under a constant axial
load and cyclic lateral loads controlled according to the predeter-
mined lateral displacement increment of the wall. An axial com-
pressive force was uniformly applied using an axial load transfer
assembly composed of a 1000-kN hydraulic jack, a steel beam of
400  400  1460 mm in dimension, a steel plate with 100 mm
thickness, and two steel bars with built-in hinges at both ends.
0
The applied axial loads (N u ) under a ratio (N u =f c Ag ) of 0.1 for all
the specimens were keep constant until the end of the tests, where
Ag is the cross-sectional area of the wall. After fully applying the
predetermined axial load, the specimen was rapidly coupled with
a lateral load transfer assembly composed of a 2000-kN capacity
actuator with hinges at both ends and two steel blocks. To intro-
duce reversal lateral loads, two steel blocks were fixed at both
sides of the top stub of the wall using four steel roads longitudi-
nally penetrating the top stub. The lateral load reversals were
applied at the center of the top stub. The test was terminated when
the applied lateral load dropped sharply.
The specimens were subjected to the predetermined lateral dis-
placement history shown in Fig. 3. The magnitude of the lateral
displacement at each cycle was calculated in accordance with the
drift ratio (dr D=hl ) increment recommended in FEMA [18] to
estimate the structural performance of precast concrete walls,
where D is the lateral displacement measured at the line of action
of the lateral load and hl is the distance from the base interface of
the wall to the lateral load line. Three cycles were iterated at each
lateral displacement increment. The lateral loadlateral displace-
ment relationship of the walls was determined using the data
recorded from the actuator loading program and a 300-mm capac-
ity linear variable differential transducer (LVDT) attached along the
lateral load line. Five 100-mm-capacity LVDTs (vertical, diagonally
intersectional, and horizontal directions, as shown in Fig. 2) were
also installed at the wall web to determine the shear displacement
component at each loading cycle. The shear slip displacement at
the base interface with construction joints was measured using a
50-mm-capacity LVDT. To avoid including the shear slip displace-
ment along the base interface into the shear displacement at the
f y (MPa) f su (MPa) Es (MPa) ey
*
498 532 202,000 0.0042
508 633 216,000 0.0024
475 594 207,000 0.0023
566* 611 199,809 0.0048
600
Lateral displacement of each wall (mm)
45
25
1/400
1/800 1/600
(0.25%) (0.33%)
5 (0.13%)(0.17%)
-15
-35
-55
0 3 6 9
Fig. 3. Predetermined lateral displacement history for lateral load reversals.
wall web, the five LVDTs were installed at spacing of 100 mm from
the base interface. Hence, the shear displacement at the wall web
and shear slip displacement along the base interface were mea-
sured independently of each other.
3. Test results and discussions
3.1. Crack propagation and failure mode
The crack propagation pattern up to the point of failure of each
wall specimen is shown in Fig. 4. All of the walls exhibited a similar
distribution and propagation of cracks in terms of the flexural
cracking at the bound elements, severe inclined cracking at the
web, and sliding damages at the base interface. The crack propaga-
tion and concrete crushing zone imply that the structural behavior
of the walls is primarily governed by shear action and sliding shear
rather than flexural yielding of longitudinal bars at the boundary
elements. The initial hairline flexural cracks appeared in the lower
zone of the boundary element and then propagated horizontally
toward the wall web. The initial flexural cracking moment (M cr )
was independent of shear reinforcement at the wall web. With
an increase of lateral loads, the flexural crack zone gradually
K.-H. Yang et al. / Engineering Structures 141 (2017) 596606
Fig. 2. Test setup and measurement details for various displacement components.
extended upwards. The first inclined shear cracks at the web sud-
1/25 denly developed in the middle area of the web. For wall C, the
(4%)
inclined shear crack appeared at approximately 476.1 kN, corre-
1/35
(2.86%)
q
0
1/50 sponding to a nominal shear stress of 0:43 f c MPa, which is
1/75 (2%) q
1/100 (1.33%) 0
1/200 1/150 (1%) higher than the concrete transfer capacity (269.1 kN = 0:24 f c )
1/300
(0.5%) (0.67%)
calculated using the equation from the ACI 318-14 guideline. The
inclined shear crack strength of the walls was insignificantly
affected by the amount of shear reinforcement, as given in Table 3.
As the lateral displacement of the wall increased, inclined shear
cracks were spaciously distributed in the web and propagated
toward the corners of the top and bottom interfaces. At the first
cycle of dr = 0.5%, the sliding cracks began to appear along the con-
struction joints. For wall C, the sliding cracks appeared at approx-
12 15 18 21 24 27 30 33 imately 800 kN, corresponding to a nominal shear slip stress of
Loading cycles q
0
0:72 f c MPa. The initial shear slip stresses of the wall were not
significantly affected by the configuration of the slip-resistance
bars. Beyond dr = 1.0%, which is the lateral displacement corre-
sponding to the peak strength, the wall web was mostly covered
by intersected inclined cracks. Beyond dr = 2.0%, crushing of con-
crete began to occur in the most heavily stressed parts of the
boundary elements, and then, the sliding crack zone increased
sharply. At dr = 2.86%, the concrete splitting failure zone near the
base interface gradually extended along with the kinking of the
longitudinal and vertical shear bars, as observed in Fig. 5. However,
the concrete crushing at the boundary element and concrete split-
ting failure along the base interface did not occur for wall W
because the tests for the wall were stopped at the 26th cycle (=dr
of 2.0%) because of a malfunction of the actuator loading program.
The kinking of the longitudinal and vertical shear bars implies that
the sliding shear at the base interface is resisted by the dowel
action of the steel bars. When dowel bars fail with the formation
of plastic hinges in the high bending stress section, the failure
mode is associated with a falling of the dowel bar and splitting
of the surrounding concrete [19].
3.2. Lateral loadlateral displacement relationship
The lateral loadlateral displacement relationship of each wall
is shown in Fig. 6. To ascertain the shear governing behavior of
 K.-H. Yang et al. / Engineering Structures 141 (2017) 596606
(a) Wall C
(d) Wall O
Fig. 4. Crack distribution and concrete crushing zones up to the failure of each wall (the numbers in the figures indicate the number of loading cycles).
Table 3
Summary of test results and comparison with predicted shear strengths.
the walls, the flexural strength (V fl ) and shear friction strength (V fr )
of each wall are also shown in the same figure. The flexural
strength of the walls was calculated using the equivalent stress
block described in ACI 318-14 under the assumption of a linear
strain distribution across the wall section [20]. The shear friction
strength at the base interface with construction joints was
calculated using the generalized equations [11] based on the
upper-bound theorem of concrete plasticity. The value of V fr was
approximately 1.5 times higher than that of V fl . Furthermore, the
shear strength measured for all of the walls was lower than the
value of V fl , indicating that the behavior of the walls was primarily
dominated by the shear capacity of the wall web.
All of the walls exhibited unloading and reloading stiffness
reductions and similar pinching behavior of hysteresis loops as
the cyclic displacement amplitude increased. The degree of pinch-
601
(b) Wall X (c) Wall W
(e) Wall F (f) Wall FX
ing was insignificantly affected by the configuration of the sliding
resistance bars at the base interface, even at the high inelastic state
of the walls. Wall C (Fig. 6a) reached its ultimate shear strength at
dr of 1.0% with severe inclined shear cracking at the web. Beyond
the ultimate shear strength, the lateral displacement steadily
increased without a sharp drop in the applied lateral load, accom-
panied by severe sliding damage at the base interface. At the sec-
ond cycle at dr of 2.86% (corresponding to the 29th cycle), the
lateral load dropped sharply along with the kinking in the longitu-
dinal and vertical shear bars. The lateral loadlateral displacement
curve of wall X (Fig. 6b) is very similar to that measured for wall C,
except for the retention of one more hysteresis loop at dr of 2.86%
in the former specimen. For wall W (Fig. 6c), test data were com-
piled up to the second cycle of dr = 2.0% owing to the malfunction
of the actuator loading system. Wall W exhibited a similar slope of
602 K.-H. Yang et al. / Engineering Structures 141 (2017) 596606

Fig. 5. Typical concrete crushing observed in wall W owing to the kinking of the longitudinal and vertical shear bars.

Wall drift ratio (%) Wall drift ratio (%)

3.3 2.5 1.7 0.8 0 0.8 1.7 2.5 3.3 3.3 2.5 1.7 0.8 0 0.8 1.7 2.5 3.3
2000 2000
Initial inclined shear crack Initial inclined shear crack
1600 Vfr 1600 Vfr
Lateral load (kN)

Sliding crack at the base Sliding crack at the base

Lateral load (kN)

1200 interface 1200 interface

Vfl Vfl
800 Concrete crushing at web 800 Concrete crushing at web

Kinking in vertical shear Kinking in vertical shear
400 400
reinforcement reinforcement
0 0
-400 -400
-800 -800
Vfl Vfl
-1200 -1200
Vfr Vfr
-1600 -1600
-2000 -2000
-50 -40 -30 -20 -10 0 10 20 30 40 50 -50 -40 -30 -20 -10 0 10 20 30 40 50
Total lateral displacement (mm) Total lateral displacement (mm)
(a) Wall C (b)Wall X

Wall drift ratio (%) Wall drift ratio (%)

3.3 2.5 1.7 0.8 0 0.8 1.7 2.5 3.3 3.3 2.5 1.7 0.8 0 0.8 1.7 2.5 3.3
2000 2000
Initial inclined shear crack Initial inclined shear crack
1600 Vfr 1600 Vfr
Lateral load (kN)

Lateral load (kN)

Sliding crack at the base Sliding crack at the base

1200 interface 1200 interface
Concrete crushing at web
Vfl Vfl
800 800 Concrete crushing at web

Kinking in vertical shear Kinking in vertical shear
400 400
reinforcement reinforcement
0 0
-400 -400
-800 -800
Vfl
*Test stop due to the
Vfl

-1200 malfuction of loading -1200
Vfr system Vfr
-1600 -1600
-2000 -2000
-50 -40 -30 -20 -10 0 10 20 30 40 50 -50 -40 -30 -20 -10 0 10 20 30 40 50
Total lateral displacement (mm) Total lateral displacement (mm)
(c) Wall W (d) Wall O

Wall drift ratio (%) Wall drift ratio (%)

3.3 2.5 1.7 0.8 0 0.8 1.7 2.5 3.3 3.3 2.5 1.7 0.8 0 0.8 1.7 2.5 3.3
2000 2000
1600 Initial inclined shear crack 1600 Initial inclined shear crack
Sliding crack at the base Vfr Vfr
Lateral load (kN)

Sliding crack at the base

Lateral load (kN)

1200 interface 1200 interface

800 Concrete crushing at web Vfl 800 Concrete crushing at web Vfl

Kinking in vertical shear Kinking in vertical shear
400 400
reinforcement reinforcement
0 0
-400 -400
-800 Vfl -800 V fl

-1200 -1200
Vfr Vfr
-1600 -1600
-2000 -2000
-50 -40 -30 -20 -10 0 10 20 30 40 50 -50 -40 -30 -20 -10 0 10 20 30 40 50
Total lateral displacement (mm) Total lateral displacement (mm)
(e) Wall F (f) Wall FX

Fig. 6. Lateral loadlateral displacement relationship of each wall.

 K.-H. Yang et al. / Engineering Structures 141 (2017) 596606 603

the ascending branch and post-peak behavior until the test termi-
nation as those of wall X. Compared with the lateral loadlateral
displacement relationship of wall X, the retention of more hystere-
sis loops beyond dr of 2.0% is expected for wall W. The tests for wall
O (Fig. 6d) were terminated at the first cycle of dr = 4.0% along with
severe concrete splitting failure because of the kinking in the ver-
tical shear reinforcement. Wall F (Fig. 6e) with WWF reinforcement
at the web reached its ultimate shear strength at dr of 1.0% and had
a similar decreasing rate of the applied loads at the post-peak
branch as that of the companion wall C. The applied load dropped
sharply at the third cycle of dr = 2.86% (corresponding to the 30th
cycle). Wall FX (Fig. 6f) also exhibited a very similar lateral load
lateral displacement curve as that in the companion wall X. This
finding indicates that for the squat shear walls studied with
as = 1.0, the WWF reinforcement performed comparably to the
conventional vertical and horizontal shear reinforcements in terms
of preventing a rapid decrease in the applied loading after the peak
strength.
3.3. Components of shear slip displacement
Fig. 7 shows the typical lateral loadshear slip displacement
relationship measured in walls C and X. The contribution of shear
slip displacement to the total lateral displacement of the wall is
also shown in Fig. 8 with respect to the drift ratio increment of
the wall. The amount of slip displacement was small enough to
be negligible up to the peak load. At the peak strength of the walls,
the contribution of slip displacement to the total lateral displace-
ment ranged between 5.9% and 9.1% for the walls with the conven-
tional shear reinforcement and 9.2% and 15.9% for the walls with
WWF reinforcement. Because of the smaller amount of WWF rein-
forcement, the contribution of slip displacement was higher for the
tional shear reinforcement. No clear trend was observed for the
effect of the slip-resistance bars on the slope of the ascending
branch of the lateral loadshear slip displacement. After the peak
strength state of the walls, the amount of slip displacement sharply
increased without a noticeable drop of the applied loads, regard-
less of the configuration of the slip-resistance bars. Therefore, the
contribution of shear slip displacement became more notable
beyond dr = 1.0%. At dr of 2.0%, the slip displacement contributions
accounted for 20.1%, 12.2%, 9.5%, and 16.2% for walls C, X, W, and O,
respectively. Up to dr of 2.0%, the lowest slip displacement compo-
nent was obtained for wall W. The X- and W-shaped bars at the
base interface were more effective than O-shaped bars in reducing
slippage at the base interface of squat shear walls with as = 1.0. The
contribution of slip displacement measured in walls with WWF
reinforcement was also comparable to that of the companion spec-
imens (walls C and X) with the conventional shear reinforcement.
The amount of slip displacement significantly depended on the
elongation of the reinforcing bars crossing the interfacial shear
plane because the reinforcing bars develop clamping stresses by
restraining the sliding along the interface [11]. The slip along the
interfacial shear plane was entirely resisted by the dowel action
generated from the reinforcing bars crossing the interface. The
CEB-FIP model [19] estimates the slip amount corresponding to
the shear friction strength as a function of equivalent normal stres-
ses generated from the reinforcing bars crossing the interface and
applied axial stresses. Hence, it can be regarded that the longitudi-
nal bars in the boundary elements and vertical shear reinforcing
bars in the web play important roles as dowel bars in resisting
30
25
Contribution of slip displacement

walls with WWF reinforcement than for the walls with conven-
20
component (%)

1200
900 15

600
C
10
Lateral load (kN)

300
X

0 5
W

-300
O
0
-600 0 0.5 1 1.5 2 2.5 3
-900 Wall drift ratio (%)

-1200
(a) Walls with conventional shear reinforcement
-8 -6 -4 -2 0 2 4 6 8
30
Slip displacement (mm)
(a) Wall C
Contribution of slip displacement

25
1200
component (%)

20
900
600
15
Lateral load (kN)

300
0 10
F
-300
5 FX
-600
-900 0
0 0.5 1 1.5 2 2.5 3
-1200
-8 -6 -4 -2 0 2 4 6 8 Wall drift ratio (%)
Slip displacement (mm)
(b) Walls with WWF reinforcement
(b) Wall X
Fig. 8. Contribution of shear slip displacement component to the total lateral
Fig. 7. Typical lateral loadshear slip displacement curves of the walls. displacement.
604
the sliding shear along the base interface. Although the amount of
slip-resistance bars was considerably lower than that of the longi-
tudinal and vertical shear reinforcing bars, the slip-resistance bars
acted effectively in reducing the sliding displacement along the
base interface with construction joints. At dr of 2.0%, the contribu-
tion of the slip displacement component determined for walls X
and W decreased by half compared with that of the companion
wall C.
3.4. Shear strength
Because the pre-peak shear behavior of all the specimens was
primarily governed by shear at the concrete strut of the wall web
and the slip displacement at the base interface was very small up
to the peak load, the slip-resistance bars at the base interface
insignificantly contributed to the shear strength enhancement of
the squat shear walls with as of 1.0, as observed in Table 3.
Although a higher shear strength was measured for wall W than
for the other walls, the enhanced ratio was quite small. Compared
with the shear strength of companion specimen C, wall X had a
1.5% lower value, whereas wall W had only a 7% higher value.
The shear strength of wall F was approximately 7% lower than that
of the companion wall C, whereas wall FX had a 6% higher shear
strength that of the companion wall X, even though the web rein-
forcement ratio was lower for the WWF reinforcement than for the
conventional shear reinforcement. In terms of the effectiveness of
WWF in enhancing the shear strength of walls, a significant trend
was not observed.
3.5. Work damage indicator
Although all of the wall specimens were governed by shear, the
cumulative work damage indicator (W c ) proposed by Sheikh and
Khoury [21] was calculated at each loading cycle until the failure
of walls to quantitatively examine the effect of the slip-
resistance bars on the energy dissipation of the walls. The W c value
remained nearly constant up to 20 cycles (dr = 1.0%), beyond which
it increased sharply, as shown in Fig. 9. All of the walls exhibited a
consistent result for W c values at each loading cycle, indicating
that the slip-resistance bars had no significant effect on the energy
dissipation of the walls with as = 1.0.
4. Comparison with prediction models
Most empirical equations including the shear guideline of ACI
318-14 provide no specific comments on the shear friction resis-
tance against shear slip displacement at the base interface of walls.
Furthermore, STMs [9,15] proposed for estimating the shear
400
C that the lateral load capacity of the walls depends on the flexural
350
X strength up to a displacement ductility ratio of 2.0, whereas the
Work damage indicator (W)
300 W lateral load capacity corresponds to the sliding shear strength at
O the base interface for a displacement ductility ratio of 2.5. The Sal-
250
F onikios equation for the sliding shear failure mode is summarized
200 FX
150
100
50
0
0 5 10 15
Loading cycle
Fig. 9. Work damage indicator of walls tested.
K.-H. Yang et al. / Engineering Structures 141 (2017) 596606
strength of squat walls do not consider the slip resistance along
the base interface in the assumed load transfer mechanisms.
Meanwhile, against sliding shear failure of squat walls, EC8 [10]
requires an explicit calculation of the resistance of walls being
assumed to be made up of the dowel action of reinforcing bars
crossing the base interface and frictional resistance of concrete.
For RC walls subjected to seismic loading, EC8 specifies different
design equations associated with the modes of shear failure such
as diagonal tension (V Rd3 ), diagonal compression (V Rd2 ), and sliding
shear (V Rd;s ):
V Rd3 V cd qh as  0:3 qv 1:3  as f yh bw z 1
V Rd2 bw zm1 f c =cot h tan h
0
2
V Rd;s V dd V id V fd 3
q
0
where V cd (= 0:06 f c 1:2 40ql bw z) is the shear transfer
capacity of concrete, ql is the longitudinal reinforcement ratio at
the boundary elements, bw is the thickness of the wall web, z is
the internal lever arm between tensile reinforcement and the cen-
ter of compression zone, f yh is the yield strength of shear reinforce-
ment, m1 is strength reduction factor for concrete cracked in shear,
and h is the angle between the concrete compression strut and the
axis perpendicular to the shear force. The sliding shear resistance
given in Eq. (3) includes the contributions of the dowel resistance
(V dd ) of vertical bars, the shear resistance (V id ) of diagonal bars, and
the friction resistance (V fd ) of concrete, as given in the following
form:
q q
0 0
V dd Min1:3Av f c f yh As f c f y ; 0:25Av f yh As f y  4
V id As;i f yh;i cos u 5
V fd MinlAv f yh As f y N u n Mn =z; 0:50:6  0:0024f c f c nlw bw 
0 0
6
where Av is the area of vertical shear reinforcement at the wall web,
As and f y are the area and yield strength, respectively, of longitudi-
nal reinforcement at the boundary elements, As;i , f yh;i , and u are the
area, yield strength, and inclination, respectively, of diagonal bars
crossed with the base interface, l is the friction coefficient of con-
crete, Mn is the nominal momment capacity of walls, and n is the
normalised neutral axis depth.
Salonikios [5] was concerned with the possibility of sliding
shear failure of squat walls designed in accordance with the EC8
recommendations. Based on the test results, he proposed an empir-
ical method for predicting the shear strength of squat shear walls
dominated by the sliding shear failure mode. For squat shear walls
governed by the flexural yielding mechanism, Salonikios noted
below:
V n T cfr T d for walls with conventional reinforcement 7:a
1 0:1li =0:8lw
V n T cfr 0:27T d 2As;i f yh;i cos u
as
for walls with diagonal reinforcement 7:b
20 25 30
Nu
T cfr 2:14 0 0:75Mn =z Nu =2 8
f c Ag as
 K.-H. Yang et al. / Engineering Structures 141 (2017) 596606
f su As
T d 0:385 9
as
where li is the distance between the centers of the axes of diagonal
reinforcement at the wall base, and f su is the tensile strength of lon-
gitudinal reinforcement. Salonikios considers the concrete friction
resistance (T cfr ) and dowel action resistance (T d ) of longitudinal
reinforcement as a primary sliding resistance mechanism of walls.
Table 3 compares the shear strength of the squat HWC shear
walls investigated in this study with predictions calculated using
the above empirical equations. Mun and Yang [15] discussed the
significance and limitations of the previous equations proposed
by ACI 318-14 [8], Wood [12], and Gulec and Whittaker [22] as
well as STMs proposed by Hwang et al. [9] and Siao [23] in their
comparisons using collected test data on 150 squat NWC shear
wall specimens. Hence, the comparisons conducted in this study
focused on estimating the reliability and safety of the equations
of EC8 and those proposed by Salonikios to avoid sliding shear fail-
ure at higher inelasticity levels. In Table 3, the predictions obtained
from EC8 are separately treated concerning the different modes of
shear failure, including the shear strengths governed by diagonal
tension and diagonal compression failures at the wall web and
sliding shear strength at the base interface. According to EC8, the
shear strength of wall C depends on the sliding shear failure at
the base interface, but is predominated by diagonal tension failure
at the wall web for the other walls. However, the shear strength of
squat HWC shear walls was fairly higher than the predictions cal-
culated under the assumption of diagonal tension failure of the
wall web, resulting in the ratios between the experiments and pre-
dictions ranging from 2.23 to 2.63. Furthermore, the shear
strengths of the walls at dr of 2.0% were higher than the predicted
sliding shear strengths of EC8. The equations proposed by Saloni-
kios estimated the sliding shear strengths of the HWC shear walls
to be as much as 2.2 times higher than those of the EC8 equation.
The shear strength of squat HWC shear walls can be conservatively
estimated using the equations proposed by EC8, even for the slid-
ing shear strength at a high wall drift ratio with a large slip dis-
placement component. Note that stiff top and bottom stubs
confined the walls, which could slightly enhance the shear and
flexural capacities of the walls.
The predictions obtained using the equations of Salonikios were
close to the peak strength of the HWC wall specimens, with the
ratios between the experiments and predictions ranging from
0.82 to 1.12, even though the strength of the walls was governed
by the inclined shear cracks at the web rather than the sliding shear
mechanism at the base interface. At dr of 2.0%, the shear strength of
the HWC wall specimens ranged between 80% and 89% of the peak
shear strength. As a result, the shear strength at dr of 2.0% was lower
than the predictions of the Salonikios equations. It is also found that
the ratios between the experiments and predictions were lower for
walls with slip-resistance bars than for wall C with the conven-
tional shear bars. This implies that Salonikios equation given in
Eq. (7) overestimates the shear transfer capacity of diagonal shear
bars crossed with the base interfaces and the boundary longitudinal
reinforcing bars. In particular, the equation assumes that the longi-
tudinal reinforcing bars reach their tensile strength in calculating
the dowel action resistance, as given in Eq. (9). This would be one
of the sources for the overestimation at a high wall drift ratio. Over-
all, the equations proposed by Salonikios for the sliding shear
strength of shear walls tend to overestimate the shear strength of
squat HWC shear walls at a high wall drift ratio.
5. Conclusions
Based on the tests to examine the effect of different slip-
resistance bars on the sliding shear behavior of squat HWC shear
605
walls at a high inelastic state, the following conclusions may be
drawn:
1. The ultimate failure of the walls was associated with kinking of
the longitudinal and vertical shear bars and splitting of the sur-
rounding concrete at the base interface due to the dowel resis-
tance of the bars.
2. The WWF reinforcement performed comparably to the conven-
tional vertical and horizontal shear reinforcements in terms of
enhancing the shear strength of squat shear walls and prevent-
ing a rapid decrease in the applied loading after the peak
strength.
3. No clear trend was observed for the effect of the slip-resistance
bars on the ultimate shear strength of the walls and the slope at
the ascending branch of the lateral loadshear slip displace-
ment curve. However, after the peak load, the slip displacement
components were lower for the walls with slip-resistance bars
than for the wall without such bars. Up to the wall drift ratio
of 2.0%, the lowest slip displacement component was obtained
for the wall with the W-shaped slip-resistance bars.
4. The shear strength of squat HWC shear walls can be conserva-
tively estimated using the equations proposed by EC8, whereas
the equations proposed by Salonikios for the sliding shear
strength of shear walls tends to overestimate the shear strength
of squat HWC shear walls at high wall drift ratios.
Acknowledgement
This work was supported by the National Research Foundation
of Korea (NRF) grant funded by the Korea Government (MSIP
South Korea) (No. NRF-2014R1A2A2A09054557).
References
[1] ACI 349:2006. Code Requirements for Nuclear Safety-Related Concrete
Structures & Commentary. Farmington Hills: American Concr Inst; 2007.
[2] ASCE/SEI 43-05. Seismic Design Criteria for Structures, Systems, and
Components in Nuclear Facilities. American Society Civil Eng; 2005.
_ Properties of heavyweight concrete produced with barite. Cem Concr
[3] Topu IB.
Res 2003;33(6):81522.
[4] Yang KH, Mun JS, Lee H. Workability and mechanical properties of
heavyweight magnetite concrete. ACI Mater J 2014;111(3):27382.
[5] Salonikios TN. Shear strength and deformation patterns of R/C walls with
aspect ratio 1.0 and 1.5 designed to Eurocode 8 (EC8). Eng Struct 2002;24
(1):3949.
[6] Paulay T, Priestley MJN, Synge AJ. Ductility in earthquake-resisting squat shear
walls. ACI J Proc 1982;79(4):25769.
[7] Salonikios TN, Kappos AJ, Tegos IA, Penelis GG. Cyclic load behavior of low-
slenderness reinforced concrete walls: design basis and test results. ACI Struct
J 1999;96(4):64960.
[8] ACI Committee 318. Building Code Requirements for Structural Concrete (ACI
318-14) and Commentary. Farmington Hills: American Concr Inst; 2014.
[9] Hwang SJ, Fang WH, Lee HJ, Yu HW. Analytical model for predicting shear
strength of squat walls. J Struct Eng ASCE 2001;127(1):4350.
[10] CEN Technical committee 250/SC8. Eurocode 8: Earthquake resistant design of
structures-Part 1: General Rules, (ENV 1998 1-1, 1-2, and 1-3). Brussels: ECN;
1995.
[11] Yang KH. Shear stressrelative slip relationship at concrete interface. Adv
Mater Sci Eng 2016;2016:19.
[12] Wood SL. Shear strength of low-rise reinforced concrete walls. ACI Struct J
1990;87(1):99107.
[13] Gulec CK, Whittaker AS, Stojadinovic B. Shear strength of squat reinforced
concrete walls with boundary flanges or barbells. ACI Struct J 2009;106
(3):36877.
[14] Gupta A, Vijaya Rangan B. High-strength concrete (HSC) structural walls. ACI
Struct J 1998;95(2):194204.
[15] Mun JH, Yang KH. Load capacity of squat RC shear walls by strut-and-tie
model. Mag Concr Res 2016;68(24):126577.
[16] Yang KH, Mun JS, Shim HJ. Shrinkage of heavyweight magnetite concrete with
and without fly ash. Constr Build Mater 2013;47:5665.
[17] ASTM C33/C33M, C150/C150M. Annual book of ASTM standards: V.
4.02. ASTM International; 2012.
[18] FEMA. FEMA 356: prestandard and commentary for the seismic rehabilitation
of buildings. Washington DC.: Emergency Management Agency; 2000.
606 K.-H. Yang et al. / Engineering Structures 141 (2017) 596606

[19] CEB-FIP. Structural connections for precast concrete [21] Sheikh SA, Khoury SS. Confined concrete columns with stubs. ACI Struct J
buildings. Switzerland: International Federation for Structural Concrete (fib); 1993;90(4):41431.
2008. [22] Gulec CK, Whittaker AS. Empirical equations for peak shear strength of low
[20] Mun JH, Yang KH. Generalized lateral loaddisplacement aspect ratio reinforced concrete walls. ACI Struct J 2011;108(1):809.
relationship of reinforced concrete shear walls. J Korea Concr Inst 2014;26 [23] Siao WB. Shear strength of short reinforced concrete walls, corbels, and deep
(2):15969. beams. ACI Struct J 1994;91(2):12332.