Journal of Building Engineering 49 (2022) 104112

Contents lists available at ScienceDirect

Journal of Building Engineering

journal homepage: www.elsevier.com/locate/jobe

Seismic behaviors of steel truss-embedded steel-concrete

composite shear walls
Kai-Yuan Xu a, Dan Xu a, Xiao-Ting Wang a, *, Tao Wang a, Jiachuan Yan c, d,
Jia-Bao Yan b, **
a
Key Laboratory of Earthquake Engineering and Engineering Vibration, Institute of Engineering Mechanics, CEA, Harbin, 150080, China
b
School of Civil Engineering, Tianjin University, Tianjin, 300350, China
c
Key Lab of Structures Dynamic Behaviour and Control of the Ministry of Education, Harbin Institute of Technology, Harbin, 150090, China
d
Key Lab of Smart Prevention and Mitigation of Civil Engineering Disasters of the Ministry of Industry and Information Technology, Harbin Institute
of Technology, Harbin, 150090, China

A R T I C L E I N F O A B S T R A C T

Keywords: This study firstly proposed a novel type of steel truss-embedded steel-concrete composite shear
Mega-frame structure wall (STSCW) for super-tall buildings. The embedded steel truss in this shear wall provides lateral
Steel truss-embedded steel-concrete composite resistance even after concrete crushing, delays concrete failure, and serves as an effective
shear walls connection to the external truss beam in mega-frame structures. Four cyclic loading tests were
Seismic behavior conducted on the proposed STSCWs to investigate their seismic behaviors. In this experimental
Cyclic tests
program, the investigated parameters included the axial force ratio and the steel-concrete
Embedded steel truss
interface with/without studs. The results of the cyclic loading tests demonstrated that all the
Coordinate deformation
STSCWs failed in flexure, and exhibited the characteristics of local buckling of faceplates, vertical
fracture of the welds at corners, tensile fracture of the steel side plates, concrete crushing, chord
necking, or buckling of the embedded steel truss. The axial force ratio was found to significantly
affect the energy dissipation and deformation capacities of the STSCWs. In addition, the defor­
mation of embedded steel truss worked well with the steel-concrete-steel sandwich wall (SCSSW)
before the yielding point. Welding the studs on the flange surface of the embedded steel truss
improved the deformation compatibility between the embedded steel truss and SCSSW wall.
Moreover, the embedded steel truss failed in flexure via chord member buckling or necking.

1. Introduction
Mega-frame structures have been widely used as the sub-frames in super-tall buildings, and their lateral force resistance system
typically consists of outriggers and mega-columns, as shown in Fig. 1. To guarantee the effective transfer of lateral forces produced by
wind and earthquakes, the outrigger is usually inserted into the core shear walls; while this allows the composite joint to sustain large
bending, compression, and even tensile force, it increases the complexity of on-site construction. One effective solution to overcome
this difficulty is to embed a steel truss in the shear wall and connect it directly to the outrigger. Thus, the lateral force can be uniformly
distributed throughout the shear wall.
Reinforced concrete (RC) shear walls with embedded steel trusses have been researched and applied in super-high-rise buildings

* Corresponding author.
** Corresponding author.
E-mail addresses: [email protected] (X.-T. Wang), [email protected] (J.-B. Yan).

https://doi.org/10.1016/j.jobe.2022.104112
Received 25 November 2021; Received in revised form 10 January 2022; Accepted 22 January 2022
Available online 25 January 2022
2352-7102/© 2022 Elsevier Ltd. All rights reserved.
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

[1]. Jiang et al. [2] and Li et al. [3] researched the seismic behavior of steel truss-embedded RC shear walls via both experimental and
numerical methods; the test results revealed that the embedded steel trusses had a noticeable effect on the seismic performance of the
core RC walls, including their load-carrying capability, energy dissipation, and damage mode. Wu et al. [4–6] studied the seismic
performance of RC shear walls with embedded steel trusses, and found that the embedded steel trusses could significantly enhance the
seismic performance; however, local concrete crushing was observed in the RC shear wall, and the widely distributed cracks were
difficult to repair. Recently, steel-concrete-steel (SCS) composite shear walls have been proposed and experimentally investigated
[7–16], and the results have proven that SCS composite shear walls are characterized by low damage and high deformation capacity.
Among these studies, Ji et al. [13] and Yan et al. [14] investigated composite shear walls (CSWs) with infilled steel tubes in the
boundary elements, and their test results demonstrated that the embedded steel tubes improved the seismic behavior of CSWs, e.g., the
deformation and load-carrying capacity, and also prevented concrete crushing at the corner region of the SCS shear walls. Based on
these studies, a novel steel truss-embedded steel-concrete composite shear wall (STSCW) in which a steel truss is embedded into an SCS
composite shear wall is proposed in the present work. In this proposed STSCW, the embedded steel truss can provide lateral resistance
after concrete crushing that delays failure, and serves as an effective connection to the external truss beam in mega-frame structures.
Numerous investigations were carried out to thoroughly understand the seismic performance of this novel STSCW.
In this research, four specimens were designed and tested under quasi-static loading to investigate the seismic behavior of the
proposed STSCW. In this testing program, the two investigated parameters were the axial force ratio and the steel-concrete interface
without studs. The test results provide detailed information on the seismic behavior of the STSCW, and the key influencing parameters
are also discussed. Finally, conclusions are drawn from the experimental results and discussions.

2. Experimental program
2.1. Depiction of specimens
Each STSCW specimen contained a top concrete beam, two steel faceplates, two steel side plates, a concrete core, an embedded steel
truss, tie bolt connectors, and a foundation beam, as shown in Fig. 2. As presented in Fig. 3, the representative fabrication process of the
STSCW specimens included the following: (1) fabricating the chord member and web member, (2) welding the embedded steel truss,
(3) welding steel truss to end plate according to design drawing, and welding end plate to the steel faceplate, assembling the external
steel faceplates and steel truss to form the steel frame, (4) placing the reinforcement for the foundation beam, (5) placing the rein­
forcement for the top beam, and (6) the casting of concrete.

Fig. 1. Outrigger trusses in super high-rise buildings.

2
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

Fig. 2. Section of steel truss-double steel plate-concrete composite shear wall.

Four specimens in total were prepared for the cyclic loading tests, namely STSCW1-4. These four STSCWs were designed to
represent tubular shear walls connected to outriggers in mega-frame buildings, and, considering the load capacity of the testing fa­
cilities, they were scaled by a factor of 1/8. Fig. 2 presents the geometric details of a representative specimen, namely STSCW1. The
height (H) of STSCW1 was 2000 mm, and its cross-section was rectangular with a size of 1000 × 125 mm2 (width × depth). According
to the Chinese Code JGJ／T 380–2015 [22],tie bolt connectors were used to connect two steel faceplates with horizontal and vertical
spacings of 150 mm and 120 mm, respectively. The steel faceplates and side plates were linked by groove welding. The aspect ratio of
each specimen was 2.15 to ensure their flexural failure mode. The reinforced concrete foundation beam measured 2500 × 900 × 800
mm3 (length × width × height). The dimensions of the RC top beam for the STSCW specimens, excluding specimen STSCW3, were the
same, namely 300 × 300 mm2 (width × depth). Because specimen STSCW3 was subjected to tensile force, its cross-section was
designed as 400 × 500 mm2 (width × depth). Specimen STSCW2 was designed with studs welded on the flange of the embedded steel
truss to investigate their influence on the seismic performance of the STSCW. The welding spacing of the studs, which had a diameter of
10 mm and a height of 40 mm, was 150 mm.
The axial force ratios obviously affect the load-carrying capacity of steel truss-embedded steel-concrete composite shear wall. The
STSCWs in building could sustain large bending, compression, and even tension. Thus, three different working cases i.e., axial tension,
normal axial compression, large axial compression were selected. To study its influence, specimens STSCW1, STSCW3, and STSCW4
were subjected to different axial force ratios of 0.3, − 0.1, and 0.7, respectively (the negative ratio denotes tension, and more details are
provided in Table 1). According to the Chinese Code JGJ138-2016 [19], the axial force ratio for the composite wall is defined as
follows:
N
n= (1)
fc Ac + fy As

where n represents the axial force ratio, N is the axial load imposed on the specimen, fc represents the concrete prism axial compressive
strength of the specimen, fy denotes the yield strength of the steel plates, and As and Ac respectively represent the gross cross-sectional
areas of the steel plate and concrete in the specimen.
The yield strength ratio describes the contribution of the embedded steel truss to the lateral shear resistance of the STSCW, and is
defined as follows:
Fu1
ξ= (2)
Fu2

where Fu1 is the yield strength of the embedded steel truss, Fu2 denotes the yield strength of the composite shear wall, and ξ represents
the yield strength ratio.
The yield strength of the embedded steel truss is obtained according to the yield strength ratio. The failure mode of the embedded
steel truss is also flexural, meaning that the chord members would yield while all web members stay elastic. Therefore, the detailed
geometry of the steel truss was obtained, as reported in Table 1.

3
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

Fig. 3. Fabrication process on STSCW wall with embedded steel truss.

Table 1
Details of STSCWs.

Specimen No. H (mm) tw(mm) Size (mm4) Llw (mm) Ldw (mm) nd Ry Welding stud fy (MPa) fu (MPa) Es (GPa)

STSCW1 2000 125 66 × 50 ×8× 8 784 889 0.3 0.3 N-stud 340 450 214
STSCW2 2000 125 66 × 50 ×8× 8 784 889 0.3 0.3 Y-stud 340 450 214
STSCW3 2000 125 62 × 50 ×6× 6 788 900 − 0.1 0.3 N-stud 340 450 214
STSCW4 2000 125 60 × 50 ×5× 5 790 898 0.7 0.3 N-stud 340 450 214

H denotes height of STSCW; tw is the thickness of the composite shear walls; Size represents cross-section size of chord and web member; Ldw denotes length of diagonal
web member; Llw represents length of lateral web member; nd is axial force ratio; Ry is the ratio of yield strength. fy and fu denotes yield and ultimate strength of steel
plate, respectively; Es is elastic modulus of steel plate.

2.2. Material properties

C30 concrete was adopted as the core material for all specimens. The four test specimens were cast with the same consignment of
concrete. The cubic concrete compressive strength (fcu) was obtained via compression tests of five 150 × 150 × 150-mm3 cubes.
According to the Chinese Code GB50010-2015 [18], the standard value of the concrete prism axial compressive strength fc is equal to
0.76fcu. The average cubic compressive strength of C30 is 33.8 MPa. Tensile tests [20] were carried out to determine the material
properties of these steel products. The steel faceplates of all specimens were fabricated with a 4-mm-thick mild steel plate (Q235), the
yield and ultimate strengths of which are listed in Table 1. The embedded steel trusses of the STSCW specimens were fabricated by
Q235 mild steel plates with different thicknesses, e.g., 8-mm-thick for specimens STSCW1-2, 6-mm-thick for specimen STSCW3, and
5-mm-thick for specimen STSCW4; their yield strengths were respectively 393, 395, and 390 MPa, and their ultimate strengths were

4
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

respectively 580, 560, and 560 MPa. The elastic moduli of the 4-, 5-, 6-, and 8-mm-thick plates were respectively 221, 213, 223, and
214 GPa. The strength grade of the M10 tie bolts was 8.8, and their yield and ultimate strengths were respectively 798.4 and 936.0
MPa.

2.3. Test setup and loading procedure

Fig. 4(a) presents the setup for the lateral cyclic loading test. The foundation beam was fixed to the reaction floor by four anchoring
rods. Axial load was imposed on the top of the specimen via a vertical hydraulic actuator. Between the top end of the vertical hydraulic
actuator and the loading beam, a horizontal sliding support was prepared to tolerate the lateral displacements of the specimens.
Because specimen STSCW3 was subjected to axial tension, the steel beam was connected to the actuator by eight bolts (M30, strength
grade of 8.8). The applied axial load was transferred to the STSCW specimens through a steel loading beam. Moreover, a set of out-of-
plane restraining I-beams was provided to avoid out-of-plane deformation. A horizontal cyclic load in the displacement mode was
applied to the specimen via a 2000-kN hydraulic jack. Fig. 5 depicts the lateral cyclic loading histories for the tests. In each loading
cycle, the push force was defined as positive loading, while the pull force was defined as negative loading. The effective height of the
STSCW was 2150 mm, and the aspect ratio of each specimen was 2.15. Once the lateral force of the STSCW specimen was reduced to
85% of its peak load, the test was terminated.
The lateral and vertical displacements of the STSCW specimens at different locations were measured by 18 linear varying
displacement transducers (LVDTs). The lateral top displacements of the STSCW specimens were monitored by LVDT H1, as shown in
Fig. 4(b). The lateral displacements along the height of the STSCW specimens were measured by five LVDTs (H2–H6) attached at the
centerline of the specimens. The shear deformation of the STSCW specimens was measured by two pairs of LVDTs (S1–S2 and S3–S4).
The horizontal slippage and rotation of the foundation beam relative to the reaction floor were measured by LVDT H7 and LVDTs
V6–V7. Five LVDTs, namely V1–V5, were fixed at the base of the STSCW specimens to measure the local vertical deformation. In
addition, strain gauges were mounted at different locations of the steel plates to measure the local strain. These strain gauges were
located at 50 mm and 250 mm above the base of the specimen, as presented in Fig. 4(c). The strains at the critical locations of the

Fig. 4. Test setup and measurements.

5
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

Fig. 5. Loading history curve.

Fig. 6. Failure modes of the typical specimen STSCW1.

6
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

embedded steel truss members were also measured, and the measurement locations are presented in Fig. 4(d). The strain data were
recorded by a computer data acquisition system.

3. Experimental phenomena and failure modes

3.1. General behavior
The composite shear wall samples exhibited similar failure modes; successive failures of the local buckling of the steel plate,
concrete crushing at the base of the STSCWs, vertical weld fracture, the tensile fracture of the steel plates, and the chord necking or
buckling of the embedded steel truss were observed in the four specimens. All the STSCWs exhibited flexure-dominated behavior.
Specimen STSCW1 was selected as a typical example to illustrate the failure process of the STSCWs.
There were three working stages for the failure process, i.e., the elastic, damage development, and recession stages. No local
buckling of the steel plates occurred during the initial loading stage; specimen STSCW1 remained elastic, and the lateral resistance
increased linearly with the increase of the lateral displacement. Fig. 6(a) and (b) present the local buckling of the steel faceplate
initiated at the base of specimen STSCW1 as the drift ratio reached 0.01 rad. Moreover, Fig. 6(c) and (d) present the development of the
local buckling of the steel plate from the outer edge to the central region, followed by the local buckling of the side plate. Specimen
STSCW1 achieved its peak load as the drift ratio reached about 2%. Moreover, the in-filled concrete was crushed, and the local
buckling of the steel faceplate propagated through the cross-section. Vertical fracture took place at the corner, as depicted in Fig. 6(e).
The vertical weld fractures developed progressively at the corner due to the expansion of the crushed concrete. The steel faceplate
suffered excessive tensile deformation, and the horizontal tensile fracture of the steel plates was observed, as depicted in Fig. 6(f). At
the second cycle of the drift ratio of 2.86%, the lateral force decreased to below 85% of the peak load of the STSCW, and the test was
terminated.
The damage processes of specimens STSCW2-4 were similar to that of specimen STSCW1, as shown in Fig. 7. For specimen STSCW2,
the test phenomenon of each stage took place relatively earlier as compared with those of specimen STSCW1, as listed in Table 2. The
studs welded on the flange of the steel truss had a quite marginal influence on the lateral shear resistance of the STSCW, but had
negative effects on its deformation capacity. This was because the interface area between the steel truss and the concrete was increased
due to the welded studs on the flange surface of the embedded steel truss; consequently, the local concrete was subjected to excess
compression or tension from the studs. Therefore, the infilled concrete was separated from the steel plate, which resulted in concrete
crushing occurring earlier in specimen STSCW2 than in specimen STSCW1.
Specimen STSCW3 performed differently because it was subjected to axial tension rather than compression. At the drift ratio of 4%,
the lateral force of specimen STSCW3 reached the peak value. With the continuation of loading, the filled concrete in compression was
seriously damaged due to the large lateral deformation. The excessive vertical deformation also resulted in the shear-off failure of the
tie bolts. Furthermore, the side steel plate was completely fractured, the horizontal fracture of the faceplate propagated horizontally
due to excessive tensile deformation, and the weld fracture length even reached about 110 mm, as shown in Fig. 7(b).
The largest axial compression was imposed on specimen STSCW4. The local buckling of the steel faceplate occurred much earlier
than that for the other specimens. Moreover, vertical welding fracture at the corner did not take place in specimen STSCW4. At the final
stage, the vertical deformation of specimen STSCW4 was noticeable, and the tie bolt was fractured. This is attributed to the severe
buckling of the steel faceplate, and the occurrence of concrete crushing at the base of the specimen. Moreover, during the loading cycle,
the load-carrying capacity of the STSCW dropped quickly, and the specimen failed, as shown in Fig. 7(c).

3.2. Concrete damage after the tests

After the tests, the steel faceplates at the bottom of the specimens were cut off to observe the damage of the infilled concrete. From
Fig. 8, it can be seen that the tensile cracks of the concrete occurred at the heights of 0 and 400 mm from the base of the specimen,
concrete crushing was observed at the corner regions, and horizontal concrete cracks ran through the width of the specimens.
Furthermore, the chord of the embedded steel truss tended to yield under tension, and flange necking occurred. For specimen STSCW2,

Fig. 7. Failure modes of STSCWs.

7
 K.-Y. Xu et al.
Table 2
Experimental results of STSCWs.

Specimen Load direction θb (rad) θc (rad) θf (rad) K0 (kN/mm) Py (kN) Δy (mm) Pm (kN) Δm (mm) Δu (mm) θu (%) μ=Δu/Δy Pme Pme/Pm
No.

STSCW1 + 0.010 0.017 0.020 61.0 570.6 20.9 687.9 42.1 61.3 2.85 2.93 680.5 0.99
– − 0.010 − 0.020 − 0.029 53.4 − 591.4 − 21.0 − 695.2 − 42.1 − 60.3 − 2.80 2.87 680.5 0.98
STSCW2 + 0.010 0.020 0.029 63.2 599.6 19.2 688.9 43.2 61.3 2.85 3.19 680.5 0.99
− 0.010 − 0.013 − 0.017 94.7 − 587.9 − 16.2 − 709.6 − 35.5 − 43.8 − 2.04 2.70 680.5 0.96
8

–
STSCW3 + 0.013 0.017 0.020 37.2 480.2 26.4 550.7 62.6 100.3 4.67 3.80 588.6 1.07
– − 0.017 − 0.020 − 0.029 52.6 − 456.2 − 28.7 − 566.8 − 80.3 − 83.4 − 3.88 2.91 588.6 1.04
STSCW4 + 0.005 0.098 – 135.1 569.0 11.6 684.1 20.2 21.0 0.98 1.81 491.3 0.72
– − 0.005 − 0.098 – 73.6 − 614.5 − 12.1 − 671.5 − 15.7 – – – 491.3 0.74
Mean 0.93
STDEV 0.13

Note: θb denotes drift ratio of the local buckling of faceplate; θc is drift ratio of the concrete crushing; θf represents drift ratio of the welding fracture; Pm denotes peak load of the specimen; Pme is predicted peak load.

Journal of Building Engineering 49 (2022) 104112

K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

Fig. 8. Experimental phenomena of specimens after cutting steel plate.

the damaged region became much larger than that of specimen STSCW1, and an obvious horizontal concrete crack at the bottom of
specimen STSCW2 was observed along the width direction. This is due to the inclusion of welded studs on the flange surface of the
embedded steel truss of specimen STSCW2, which resulted in the increase of the steel truss-concrete interface area. Specimen STSCW3
was tested under axial tension; as it reached its peak load, the corresponding load displacement was much larger than that of the other
specimens. Therefore, the concrete crushing at the corner region was more serious, and more horizontal cracks developed in the
concrete core as compared with the other specimens. Specimen STSCW4 underwent the largest axial force of 2632 kN, and it exhibited
brittle failure at the end of the test. This was due to the sudden occurrence of the crushing of the concrete core. Furthermore, the chord
of the embedded steel truss buckled locally due to compression.

Fig. 9. Hysteretic load versus top displacement curves.

9
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

4. Discussion of the experimental results

4.1. Load-displacement relationship
Fig. 9 exhibits the hysteresis curves of the relationship between the lateral force and top displacement (P-Δ) of specimens STSCW1-
4. The cyclic P-Δ curves of specimens STSCW1-4 were found to be plump with a slight pinching effect. Different failure modes at
varying points are marked on the curves. Before the yield point, the cyclic P-Δ curves of the STSCWs exhibited linear behavior, and the
residual lateral deformations of the four specimens were very small, about 3.5 mm. After the yield point, the cyclic P-Δ curves exhibited
nonlinear characteristics; this was due to the occurrence of the local buckling of the steel faceplate and the yielding of the chord of the
embedded steel truss. Moreover, the cyclic stiffness decreased and the strength steadily increased. After the peak load, the residual
deformation of the STSCWs became more obvious. Compared with that of specimen STSCW1, the resistance of specimen STSCW3
decreased more slowly, and the peak lateral load of specimen STSCW3 remained stable with a large lateral deformation, which implies
more ductile behavior. The lateral load-carrying capacity of specimen STSCW4 exhibited a sharp decrease after the peak load, which
was caused by the concrete crushing and local buckling of the steel plate running through the width of the specimen; this resulted in
similar ultimate displacement and peak displacement values. This observation implies that the axial force ratio is the governing
parameter influencing the deformation capacity of the STSCW.
The shear deformation was measured by the two pairs of diagonal LVDTs shown in Fig. 4(b), and is defined as follows:
D
Δs = [(S4 − S3 ) + (S2 − S1 )] (3)
2W

where D is the length of the diagonal LVDTs S1-4, S1–S4 represent the diagonal displacements measured by LVDTs S1–S4, respectively,
and W represents the width of the diagonal LVDT.
Fig. 11 presents the relationship between the shear deformation and total deformation. The shear deformation accounted for about
15% of the total deformation. This implies that the failure mode of the STSCW specimens was governed by their flexure under hor­
izontal cyclic loads.

4.2. Lateral load-carrying and deformation capacity

Table 2 lists the cyclic initial stiffness (K0), yield load (Py), peak load (Pm), yield displacement Δy, peak displacement Δm, ultimate
displacement Δu, ultimate drift ratio θu (θu = Δu/H, where H is the height of LVDT H1 from the bottom of the specimen), and ductility
ratio μ (μ = Δu/Δy). The envelope P-Δ curves of specimens STSCW1-4 are exhibited in Fig. 10, according to which the yield load can be
calculated by the energy method [8]. The influences of n and the steel-concrete interface without studs on the seismic behavior of the
STSCW are depicted in Fig. 12, from which the following two observations can be made.
(1) As reflected in Fig. 12(a) and (b), a larger axial force ratio was found to have a marginal influence on the load-carrying capacity
of the STSCW, but had a negative effect on the deformation capacity. In contrast, the specimen withstanding axial tension had a
reduced load-carrying capacity, but the ductility was improved. As the largest axial force was applied to the specimen, the
increase of the value of n from 0.3 to 0.7 resulted in decrements of Py and Pm by − 1.8% and 2.0%, respectively. Moreover, the
cyclic initial stiffness K0 was found to be significantly improved by 82.5% with the increase of the value of n from 0.3 to 0.7. This
is attributed to the change of the neutral axis position caused by the axial force ratio, which resulted in different moments of
inertia and lateral stiffnesses. However, the values of Δy, Δm, and Δu were respectively found to be decreased by 43.1%, 59.6%,
and 65.5% as the value of n increased from 0.3 to 0.7. As the specimen was tested under axial tension, with the decrease of n
from 0.3 to − 0.1, the Py and Pm values were respectively decreased by 19.4% and 19.2%, and the value of K0 was decreased by
21.5%. However, with the decrease of n from 0.3 to − 0.1, the values of Δy, Δm, and Δu were respectively increased by 31.5%,
69.7%, and 51.1%.

Fig. 10. Envelope curves of lateral force versus top displacement of specimen.

10
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

Fig. 11. Comparison between the shear deformation and total deformation.

Fig. 12. Influences of different parameters on seismic behavior of STSCW.

(2) As reflected in Fig. 12(c) and (d), specimens STSCW1 and STSCW2 exhibited similar lateral peak resistances, which indicates
that welding the studs on the flange surface of the steel truss had a quite marginal influence on the load-carrying capacity of the
STSCW. This is attributed to the local concrete was subjected to excess compression or tension from the studs during the lateral
cyclic loading. Therefore, the in-filled concrete damage of specimen STSCW2 is more serious than that of specimen STSCW1.
Thus, the specimen STSCW2 reached peak load relatively earlier as compared with specimen STSCW1. However, the cyclic
initial stiffness of specimen STSCW2 was much higher than that of specimen STSCW1. This is because the steel truss-concrete
interface area was increased, which resulted in a much higher moment of inertia of the cross-section. However, the values of Δy
and Δm were decreased by 15.7% and 6.4%, respectively. The ultimate drift ratio of specimen STSCW1 was 2.83%, which was
much higher than that of specimen STSCW2 (2.45%). This implies that welding the studs on the flange surface of the steel truss
had a negative effect on the deformation capacity of the specimen.

11
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

4.3. Stiffness and strength degradation

The cyclic stiffness of the STSCW is defined as follows:
∑
n
Pij
i=1
Kj = ∑n (4)
Δij
i=1

where Kj denotes the lateral stiffness at the jth displacement level, Δji is the maximum displacement at the ith cycle of the jth load
displacement level, Pij denotes the maximum force of the ith cycle at the jth displacement level, and n represents the total number of
loading cycles at displacement level j.
Fig. 13 exhibits the relationship between the cyclic stiffness and displacement of the STSCWs. The positive lateral stiffness was
either higher or lower than the negative lateral stiffness, which was because the embedded steel truss was not symmetric. At the initial
loading stage, as the concrete cracked, the stiffness of the specimens decreased rapidly. After that, the rigidity of the specimens
decreased slowly with the increase of the displacement. Fig. 13 also indicates the following: (1) specimen STSCW4 exhibited a faster
degradation of its stiffness than the other specimens due to the brittle failure of the concrete; (2) specimen STSCW2 exhibited a more
rapid decline in its stiffness than specimen STSCW1. This is because the concrete in specimen STSCW2 cracked earlier than that in
specimen STSCW1, which was caused by the welded studs on the flange surface of the embedded steel truss in specimen STSCW2.
The strength degradation of the specimen is reflected by the strength degradation ratio η, and it is defined as follows:
Pni
ηi = (5)
P1i

where ηi is the strength degradation ratio, P1i represents the peak load of the first loading cycle at loading level i, and Pni denotes the
peak load of the nth cycle at loading level i.
Fig. 14 presents the relationship between the strength degradation ratio and the lateral displacement of the STSCWs. Before the
load displacement achieved 40 mm, the strength degradation ratio was within the range of 0.90–1.1. Moreover, the strength degra­
dation ratio declined as the lateral displacement increased. As the specimen failed, the ratio η decreased to the residual value of about
0.85.

4.4. Energy dissipation

Fig. 15 exhibits the curves of the dissipated energy of each cycle and the relationship between the cumulative dissipated energy and
the drift ratio of the STSCWs. From the figure, it can be observed that the energy dissipation capacity for each loading cycle of the four
STSCW specimens increased with the increase of the lateral load displacement. The figure also reveals the following. (1) Before the
drift ratio reached 1%, the energy dissipation capacity of specimen STSCW4 was better than that of the other specimens. This is
because specimen STSCW4 experienced the damage development state as the other specimens were in an elastic state. After the drift
ratio reached 1%, the energy dissipation capacities of specimens STSCW1-3 increased rapidly. Moreover, at the ultimate drift ratio (θu),
the cumulative dissipated energy of specimen STSCW3 was higher than that of the other specimens. (2) The energy dissipation capacity
of specimen STSCW2 was close to that of specimen STSCW1. This implies that welding the studs on the flange surface had only a slight
effect on the energy capacity of the specimen.

4.5. Strains
Fig. 16 depicts the hysteresis curves of the relationship between the lateral load and the strain of the steel faceplate. Before the yield

Fig. 13. Rigidity degradation of STSCW.

12
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

Fig. 14. Strength degradation versus displacement curves of STSCW.

Fig. 15. Energy dissipated each loading cycle and accumulated dissipated energy versus loading displacement curves of STSCW1~4.

Fig. 16. Hysteretic curves of lateral load versus strain of steel faceplates.

strain, the strain exhibited linear behavior as the drift ratio increased. The load-strain curves of specimen STSCW1 were found to be
plumper than those of specimen STSCW4, which implies that specimen STSCW1 experienced a long period of damage development.
After the yield strain was reached, it was observed that the tensile strains were larger and faster than the compressive strains. This is
because the steel faceplate suffered excessive tensile deformation, and the horizontal tensile fracture of the steel plates occurred under
tension. For specimen STSCW4, after the yield strain was reached, the strain gauges installed at the external edge failed and the strain
values quickly increased. This is attributed to the severe buckling of the steel faceplate and the sudden occurrence of concrete crushing
at the base of the specimen.
Fig. 17 shows the hysteresis curves of the relationship between the lateral force and the vertical deformation of specimens STSCW1

13
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

Fig. 17. Vertical deformation at the bottom of specimen.

and STSCW4. The LVDTs measured the averaged vertical shortening of the specimens at the heights of 0 and 150 mm, and the
measurements included steel faceplate buckling and concrete crushing. Thus, the measurements of the LVDTs perfectly reflect the
overall vertical shortening of the specimens after the yield point. It was observed that the vertical compressive shortening of the
specimen was noticeably larger than the tension after the yield point. This is attributed to the severe buckling of the steel faceplate and
the occurrence of concrete crushing at the base of the specimen. Fig. 18 exhibits the strain distribution of the bottom cross-section of
specimen STSCW1 at four loading levels. Moreover, it was determined that strain distribution nearly met the plane-section assumption
until the specimen achieved its peak resistance.

4.6. Deformation compatibility

Fig. 19 presents the relationships of the strains of the steel plate and chord member of specimens STSCW2 and STSCW3 with time.
Before the yield point, the curve of the strain versus time history of the steel plate agreed well with that of the chord member, which
implies that the deformation of the embedded steel truss work well with the wall. For specimen STSCW2, the embedded steel truss and
steel faceplate exhibited similar load displacement versus strain behavior before achieving the peak load capacity. This is attributed to
the welded studs on the flange surface of the embedded steel truss of specimen STSCW2. After the peak point, the strain gauge data of
the steel faceplate were significantly larger than those of the chord member due to the severe local buckling of the steel faceplate.
However, the embedded steel truss and concrete of specimen STSCW3 separated after the yield point. It was observed that the strain of
the chord member was significantly larger than that of the steel faceplate. This is because the axial tension was fully borne by the
embedded steel truss after concrete crushing.
Fig. 20 depicts the relationship between the lateral load and strain of the chord members. The strain characteristics of the chord
were found to be very similar to those of the steel faceplate. As the lateral load reached its peak, the strain values of the chord members
were in a state of yield. For specimen STSCW4, the strain values of the chord members increased sharply after the peak point. This is
attributed to the severe buckling of the embedded steel truss caused by the largest axial force. Furthermore, the embedded steel trusses
of the other specimens exhibited similar strain characteristics. The embedded steel truss failed in flexure with chord member buckling
or necking, which was in accordance with the experimental design.

5. Evaluation of lateral load-carrying capacity of STSCWs

5.1. Analysis on peak lateral loads of STSCWs
The lateral deformation of composite walls is governed by their flexure. Thus, their lateral shear resistance is determined by their
cross-sectional bending resistance. Several assumptions were made to evaluate the bending resistance, i.e., (1) the plane section
remained plane until reaching the peak load resistance, (2) the tension of the concrete tension zone was ignored, and the concrete
compression zone was calculated by the equivalent rectangular area, and (3) full plasticity was developed in the cross-section of the
steel plates. (4) chord of embedded steel truss yielded in tension and compression, and its ultimate strength of steel was adopted. The
web member of embedded steel truss was ignored. Fig. 21 illustrates the cross-sectional stress and strain distribution of STSCW. Then,
the following equations were established:
N = NC + Tsy + Tcy − Tcl − Tsl (6)

Mp = 0.5(L − βx)NC + 0.5(L − x)Tsy + 0.5xTsl + 0.5LTcy + 0.5LTcl

(7)
+Asc fu (L − Lc )

NC = βxtc fc (8)

Tsy = 2fy1 xts (9)

14
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

Fig. 18. Strain distribution of the bottom cross section for Specimen STSCW1.

Fig. 19. Strain of steel plate and chord member versus time.

Fig. 20. Hysteretic curves of lateral load versus strain of the chord members.

Tsl = 2fy (L − x)ts (10)

Tcy = fy1 Acy (11)

Tcl = fy Acl (12)

15
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

Fig. 21. Stress distribution for cross-sectional flexural resistance of STSCW.

( )
fy1 = min σcr , fy (13)

(t )2
π 2 Es
(14)
s
σcr = k
12(1 − ν s 2)

where, N is the axial compression of the STSCW; Mp denotes the bending resistance of the STSCW; Nc represents the compressive
reaction forces of the concrete; Tsy and Tsl represent the resistances of the steel faceplate in compression and tension, respectively; Tcy
and Tcl denote the reaction forces of the side plate in compression and tension, respectively; x is the neutral axis depth; β, which is equal
to 0.8 [17], denotes the depth reduction factor of compressive concrete,; fy denotes the yield strength of the steel plate; tt represents the
thickness of the steel plate; tc is the depth of the concrete core; L is the width of the STSCW; Asc is the cross-sectional area of chord of the
embedded steel truss; σcr denotes critical compressive buckling stress of faceplate and side plate [21].
Thus, the peak lateral force of STSC can be determined as the following;
Mp
Pme = (15)
H

where, H is the height of LVDT H1 from the bottom of the specimen.

5.2. Validations
Table 2 lists the predictions on peak lateral loads of STSCWs by the proposed formulas and the comparison with the test results. The
prediction (Pme) to test (Pm) ratios of specimens ranges from 0.72 to 1.07, with a standard deviation (STDEV) of 0.13. The proposed
design method provided reasonable prediction on the peak lateral resistances. However, the deviation of predictions on peak lateral
load of STSCW4 achieved 28%. This is because that the severe buckling of the steel faceplate of STSCW4 took place after the peak point.
Thus, the yield strength of steel could be used in future specimen design.

6. Conclusions
This work proposed a novel composite wall, namely the steel truss-embedded steel-concrete composite shear wall (STSCW). To
investigate the seismic performance of the proposed composite shear wall, cyclic loading tests were conducted on four STSCWs under
vertical axial compression or tension and lateral cyclic shear loads. The conclusions of test results are summarized as follows;

16
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

(1) All the STSCWs failed in flexure, which was characterized by the local buckling of the steel faceplate, the vertical fracture of the
welds at the corners, the tensile fracture of the steel side plates, the concrete crushing of the STSCWs, and the chord necking or
buckling of the embedded steel truss. The tensile cracks of concrete occurred at the bottom of the specimens, and horizontal
concrete cracks running through the width of the specimens were observed.
(2) The specimens under axial tension were characterized by a reduced lateral load-carrying capacity, but their ductility was
improved. As the value of n decreased from 0.3 to − 0.1, the Py and Pm values of the STCSW were respectively decreased by
19.4% and 19.2%, and the K0 value was decreased by 21.5%. However, the values of Δy, Δm, and Δu were respectively increased
by 31.5%, 69.7%, and 51.1%.
(3) The increase of the value of n from 0.3 to 0.7 reduced the deformation capacity, but had only a marginal effect on the lateral
load-carrying capacity of the STSCW. As the n value increased from 0.3 to 0.7, the Py and Pm values were respectively decreased
by − 1.8% and 2.0%; however, the cyclic initial stiffness K0 was significantly improved by 82.5%. Moreover, the values of Δy,
Δm, and Δu were respectively decreased by 43.1%, 59.6%, and 65.5%.
(4) Welding the studs on the flange surface of the embedded steel truss had a quite limited influence on the lateral shear resistance
of the STSCW. However, the values of Δy and Δm were respectively decreased by 15.7% and 6.4%. Moreover, welding the studs
on the flange surface increased its initial stiffness, as well as the energy dissipation capacity.
(5) The deformation of the embedded steel truss was found to work well with the wall before the yield point. Welding the studs on
the flange surface of the embedded steel truss improved the deformation compatibility capacity between the embedded steel
truss and the SCS wall. Moreover, the embedded steel truss failed in flexure with chord member buckling or necking, which was
in accordance with the experimental design.

Author contribution statement

Kai-Yuan Xu: Conceptualization, Methodology, Data curation, Writing - original draft. Dan Xu: Data curation. Xiao-Ting Wang:
Data curation, Project administration. Tao Wang: Conceptualization, Writing - review & editing, Formal analysis. Jiachuan Yan:
Formal analysis. Jia-Bao Yan: Conceptualization, Data curation, Writing - review & editing, Project administration.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Acknowledgement
This research was funded by Scientific Research Fund of Institute of Engineering Mechanics, China Earthquake Administration
(2019D03, 2020EEEVL0501, 2020EEEVL0504), National Science Foundation For Distinguished Young Scholars (52125806), National
Science Foundation of China (52178494), Heilongjiang Province Outstanding Youth Natural Science Foundation (JC2018018) and
Heilongjiang Province Touyan Innovation Team Project. Any opinions, findings and conclusions or recommendations expressed in this
paper are those of the authors and do not necessarily reflect the views of the sponsors.

References
[1] X.Z. Zhao, B. Wang, Y.Y. Chen, H. Qin, Monotonic static tests on outrigger truss and its connection with mega column and core tube of the Shanghai Tower,
J. Build. Struct. 34 (2) (2013) 20–28 (in Chinese）.
[2] B. Wang, H.J. Jiang, X.L. Lu, Experimental and numerical investigations on seismic behavior of steel truss reinforced concrete core walls, Eng. Struct. 140 (2)
(2017) 164–176.
[3] Y.S. Li, G.Q. Li, D.G. Cui, Experimental study on seismic behavior of steel reinforced concrete low-rise shear wall, Earthq. Eng. Eng. Vib. 29 (4) (2009) 92–102
(in Chinese）.
[4] Y. T Wu, W. Yang, Y. Xiao, Seismic behavior of high strength concrete composite walls with embedded steel truss, J. Constr. Steel Res. 118 (3) (2016) 180–193.
[5] Y. T Wu, J. Zhang, Y.B. Yang, P. Ye, Y. Xiao, Seismic performance of reinforced concrete squat walls with embedded cold-formed and thin walled steel truss, Eng.
Struct. 132 (2) (2017) 714–732.
[6] Y. T Wu, D.Y. Kang, Y.B. Yang, Seismic performance of steel and concrete composite shear walls with embedded steel truss for use in high-rise buildings, Eng.
Struct. 125 (6) (2016) 39–53.
[7] J.H. Han, G.P. Shu, Y. Qin, G.G. Zhou, X.L. Zhou, Experimental seismic behavior of double skin composite wall with steel truss, J. Constr. Steel Res. 180 (2)
(2021) 1–13.
[8] J.G. Nie, H.S. Hu, J.S. Fan, M.X. Tao, S.Y. Li, F.J. Liu, Experimental study on seismic behaviour of high-strength concrete filled double-steel-plate composite
walls, J. Constr. Steel Res. 88 (6) (2013) 206–219.
[9] F.M. Bu, J.G. Nie, J.S. Fan, Experimental study on seismic behavior of medium and high shear-span ratio composite shear wall with double steel plates and infill
concrete under high axial compression ratio, J. Build. Struct. 34 (4) (2013) 91–98 (in Chinese）.
[10] H.S. Hu, J.G. Nie, M.R. Eatherton, Deformation capacity of concrete-filled steel plate composite shear walls, J. Constr. Steel Res. 103 (12) (2014) 148–158.
[11] J.G. Nie, X.W. Ma, M.X. Tao, J.S. Fan, F.M. Bu, Effective stiffness of composite shear wall with double plates and filled concrete, J. Constr. Steel Res. 99 (8)
(2014) 140–148.
[12] K.M. Anwar Hossain, H.D. Wright, Experimental and theoretical behaviour of composite walling under in-plane shear, J. Constr. Steel Res. 60 (1) (2004) 59–83.
[13] X.D. Ji, F.M. Jiang, J.R. Qian, Seismic behavior of steel tube–double steel plate–concrete composite walls: experimental tests, J. Constr. Steel Res. 86 (4) (2013)
17–30.
[14] J.B. Yan, Z.X. Li, T. Wang, Seismic behaviour of double skin composite shear walls with overlapped headed studs, Construct. Build. Mater. 191 (12) (2018)
590–607.
[15] L.H. Guo, Y.H. Wang, S.M. Zhang, Experimental study of rectangular multi-partition steel-concrete composite shear walls, Thin-Walled Struct. 130 (2018)
577–592.

17
K.-Y. Xu et al. Journal of Building Engineering 49 (2022) 104112

[16] T.S. Eom, H.G. Park, C.H. Lee, J.H. Kim, I.H. Chang, Behavior of double skin composite wall subjected to in-plane cyclic loading, J. Struct. Eng. ASCE. 135 (10)
(2009) 1239–1249.
[17] J.B. Yan, Y.Y. Yan, T. Wang, Z.X. Li, Seismic behaviours of SCS sandwich shear walls using J-hook connectors, Thin-Walled Struct. 144 (2019) 1–15.
[18] GB 50010-2015, Code for Design of Concrete Structures, China Ministry of Construction, Beijing, 2015 (in Chinese）.
[19] JGJ 138-2016, Code for Design of Composite Structures, China Ministry of Construction, Beijing, 2016 (in Chinese）.
[20] GB/T228.1-2010, Metallic Materials: Tensile Testing, China Planning Press, Beijing, 2010 (in Chinese).
[21] GB 5517-2017, Standard for Design of Steel Structures, China Ministry of Construction, Beijing, 2017 (in Chinese).
[22] JGJ/T 380-2015, Technical Specification for Steel Plate Shear Walls, China Ministry of Construction, Beijing, 2015 (in Chinese).

18