Engineering Structures 189 (2019) 11–24

Contents lists available at ScienceDirect

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

Experimental investigation of influences of several parameters on seismic T

capacity of masonry infilled reinforced concrete frame
Hamood Alwashalia, Debasish Sena, , Kiwoong Jinb, Masaki Maedaa
⁎

a
Graduate School of Engineering, Department of Architecture and Building Science, Tohoku University, Sendai, Japan
b
Graduate School of Science and Technology, Department of Architecture, Meiji University, Tokyo, Japan

ARTICLE INFO ABSTRACT

Keywords: This study investigated the in-plane seismic capacity of three ½ scale single story reinforced concrete frames
Masonry infill with masonry infill that were subjected to static cyclic lateral loading. Three main parameters were investigated:
Reinforced concrete buildings the strength of the surrounding reinforced concrete (RC) columns, strength of RC beams, and mortar strength of
Seismic capacity the masonry infill. These parameters were selected because there are large variations between the design codes
Backbone curve
and past research on their importance. Experimental results showed that the infill shear strength greatly in-
Equivalent strut width
creased with the surrounding frame strength. The investigated parameters did not significantly affect the initial
stiffness. The specimen consisting of a masonry panel built with low-strength mortar showed relatively greater
ductility and relatively less damage to the masonry infill panel and surrounding frame. Based on the results, the
frame strength ratio needs to be included as a parameter in the design codes for new buildings because it can
alter the shear strength of the masonry infill and ductility of the whole structure.

1. Introduction openings, mortar characteristics, and frame strength. Out of the afore-
mentioned parameters, the characteristics of the surrounding RC frame,
It is a very common practice to use masonry infill for non-structural i.e., the strength and stiffness, are some of the most important variables
partition walls in reinforced concrete (RC) buildings, particularly in governing the seismic capacity and failure mode of infilled RC frames,
developing countries. However, it has a significant influence on the as shown in several previous experimental studies [3,18–20]. The stu-
structural behavior, which became apparent in recent earthquakes, e.g., dies by Wood [21] and Liauw et al. [22] proposed different failure
2016 Taiwan earthquake, 2016 Ecuador earthquake, and 2015 Nepal mechanisms for infilled RC frames along with strength evaluation
earthquake. In a broad sense, masonry infill enhances the overall in- methods based on the relative strength of the RC frame compared to the
plane capacity, as shown in many previous experimental studies [1–4], masonry infill strength. In a similar study, Saneinejad et al. [23] pro-
which is considered to be a favorable aspect. In contrast, the presence of posed an analytical model to estimate the length of the contact between
masonry infill in an RC frame causes reaction forces on the surrounding the RC frame and masonry, depending on the strength of the RC frame,
frame, which could be followed by unexpected failure modes, as shown which was used to evaluate the in-plane strength of the infill. However,
in several studies using experimental or analytical models [4–14], as a FEMA 306 [8], based on the study by Mainstone [24], emphasized not
result of a higher moment and shear demand in the column. Further- only the relative strength of the frame, but also its relative stiffness
more, masonry infill certainly enhances the stiffness, i.e., reduces the compared to the masonry infill, to define the width of the equivalent
natural period of an RC frame, which might affect the seismic demand. strut used to calculate the stiffness and strength of the infill. ASCE/SEI
Even though masonry infill has been experimentally and analytically 41 [25] suggested the calculation of the in-plane strength of a masonry
studied for decades, design codes and guidelines still contain large infilled frame based on a material test, i.e., shear test of the masonry,
variations in the methods used to evaluate its strength, ductility, and thus ignoring the relative stiffness of the frame compared to the ma-
stiffness, as shown in recent comparative studies by Turgay et al. [15], sonry infill, as proposed by Mainstone [24]. However, the relative
Chrysostomou et al. [16], and Al-Washali et al. [17]. strength ratio of the frame and masonry is used by ASCE/SEI 41 [25] to
There are several parameters that influence the seismic performance estimate the deformation limits. In another experimental study, Fla-
of an infilled frame, such as the brick type, aspect ratio of the panel and nagan et al. [26] showed that the strength and stiffness of the RC frame

Corresponding author.
⁎

E-mail addresses: [email protected] (H. Alwashali), [email protected] (D. Sen), [email protected] (K. Jin),
[email protected] (M. Maeda).

https://doi.org/10.1016/j.engstruct.2019.03.020
Received 22 May 2018; Received in revised form 25 February 2019; Accepted 11 March 2019
Available online 25 March 2019
0141-0296/ © 2019 Elsevier Ltd. All rights reserved.
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

had an insignificant influence on the lateral strength in the case of = Vf / Vinf (1)
compression failure mechanism. Thus, they proposed empirical equa-
where Vf corresponds to the expected lateral capacity of the boundary
tions that ignored the influence of the surrounding frame. MSJC (2016)
frame determined from a cross- sectional analysis of columns, assuming
[27] proposed an evaluation method for the shear strength that in-
plastic hinges at their ends, as in a bare frame (or plastic hinges at the
corporated the simple material properties of a masonry prism without
end of beams in the case of a weak beam and strong column). Vinf is the
considering the influence of the frame strength or stiffness.
expected lateral capacity of the masonry infill as computed using Eq.
In summary, large variations exist among the different codes as the
(2), which is a simple prediction model that shows a fair compatibility
result of an incomplete knowledge of the masonry infill material as well
with an experimental database of 24 specimens, as observed in a pre-
as seismic performance governing parameters that were derived from a
vious study by the authors [29].
relatively limited number of experimental studies. The large variations
in the literature cause confusion when practicing engineers have to Vinf = 0.05fm ·tinf · l inf (2)
select an appropriate design approach. Therefore, this study had three
objectives. The first was to present the experimental results for three ½ where linf is the infill wall length, tinf is the infill wall thickness, and fm is
scale specimens tested under cyclic static loading considering different the compressive strength of a masonry prism. It should be noted that
parameters, i.e., varying the strength of the surrounding RC columns, Eq. (2) is an empirical equation based on the experimental results for
strength of the RC beams, and mortar strength of the masonry infill. The masonry infill within the range of fm = 3–20 MPa and aspect ratio (H/
second was to investigate the seismic capacities of the tested specimens L) = 0.5–1. Further validation is needed for a prism compressive
in order to clarify the impact of the investigated parameters. The third strength outside this range.
was to investigate the applicability of the existing prediction models in An investigation was conducted of the ratio of the expected strength
estimating the seismic capacities of infilled frames, as shown in pre- of the RC frame to the expected strength of the masonry infill in a ty-
vious papers, in order to determine an appropriate method to estimate pical existing RC low-rise building (4 stories) in Bangladesh, that was
the stiffness, strength, and deformation capacity of masonry infill. mentioned in CNCRP [30]. The β index values for three existing
This experimental study was part of a wider-scope ongoing research buildings that were investigated fell in the range of 0.3–0.8 for the
program of a Japanese project called SATREPS [28], which has the goal outer frame, where the masonry wall thickness was typically 250 mm.
of upgrading the seismic evaluation methods for RC buildings in Ban- On the other hand, the β index values for the inner frames were in the
gladesh. A Bangladeshi RC frame and masonry infill were taken as re- range of 0.5–1.5 because the infill thickness in the inner partitions was
ferences when designing the test specimens. However, the design of the 125 mm, which was half of that in the outer frame.
specimens focused more on varying the investigated parameters than Based on these observations, three specimens with different
on reflecting the exact reinforcement details and materials used in strengths for their RC frames, but identical characteristics and sizes for
Bangladesh. their unreinforced masonry infills, were designed to represent the lower
boundary, average, and upper boundary of the β index.
2. Experimental program These specimens were called F-0.4, F-0.6, and F-1.5, where the
numeral represents the β index, which had values of 0.4, 0.6, and 1.5,
2.1. Design of specimens respectively. The beams were designed with a higher capacity than the
columns to reflect actual field conditions, i.e., a strong beam-weak
Five specimens were tested in this experimental program. The cur- column structural frame system, of the decades old buildings of
rent paper presents the results for three specimens (namely F-0.6, WM, Bangladesh. These are the typical conditions in several countries, as
and WB), which were recently tested in the second series. The results discussed in several studies such as Colangelo [3] and Mehrabi et al.
for the other two specimens (namely F-0.4 and F-1.5), which were [18]. Thus, the lateral frame strength, Vf, of series i) mainly depends on
tested in the first year, were presented in a previous paper [17] and will the ultimate flexural capacity of the plastic hinges at the top and bottom
be used in this paper for comparison purposes. The tested specimens of columns. However, in a real case, the positions of the hinges might
were designed based on three main parameters: (i) the ratio of an RC change. The studies of Wood [21], Liauw et al. [22], and Tempestti and
column’s lateral strength to the masonry infill’s lateral strength, (ii) the Stavridis [31] emphasized that the masonry infill force changed the
strength and stiffness of an RC beam (using a very rigid strong beam or position of the plastic hinges formed in RC frames during in-plane lat-
relatively flexible beam), and (iii) the strength of the mortar. The eral loading, based on the relative frame strength. The conventional
specimen design concept for each parameter is explained as follows: assumption of hinges formed at both ends of columns can be altered by
assuming that the lower hinge is formed in the middle or in the upper
2.1.1. Series i) specimen: Varying ratio of RC column to masonry infill third of the column. This results in a failure mechanism similar to the
lateral strength short column effect, which causes the shear failure of a column with
The strength of the surrounding RC frame compared to the masonry insufficient transverse reinforcement. In this research, to allow the full
infill has an influence on the failure mechanism and strength of both the investigation of the flexural behavior of the frame and plastic hinge
RC frame and masonry infill, as indicated in experimental studies positions, the columns were designed with additional transverse re-
conducted by Colangelo [3] and Mehrabi et al. [18]. The experimental inforcement to avoid shear failure in a case where plastic hinges might
observations and mathematical prediction models suggested by several be formed at the middle of the column. Fig. 1 shows the specimens of
researchers as well as the design guidelines, disagree with each other in series i), which had identical characteristics except for differences in
relation to the impact of varying the stiffness and strength ratio on the the column size and reinforcements.
in-plane seismic capacity of infilled masonry. Several studies [20–23]
have emphasized the necessity of using the relative frame strength 2.1.2. Series ii) specimen: Varying strength and stiffness of beam
when computing the infill strength, whereas other studies [25–27] have In the aforementioned specimens of series i), the beams were de-
ignored it. Therefore, in this study, three ½ scale specimens consisting signed with a higher capacity than the columns to reflect the actual
of reinforced concrete frames with different lateral strengths and situation of old buildings in Bangladesh. However, the recent building
identical unreinforced masonry panels were adopted to clarify the in- codes of many countries emphasize designing buildings with strong
fluence of this parameter. To classify the frames into weak and strong columns and weaker beams, with a view to improving the deformation
ones, the β index was used in this study, which can be defined as the capacity of a building and avoiding the story collapse mechanism of a
quotient of the lateral strength of the surrounding RC frame to that of building. Thus, in this study, one specimen called WB (weak beam) was
the infilled masonry panel, as shown in Eq. (1). designed with weak beams and strong columns, in order to study the

12
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

2300
200 200
580 2100 580
200

400

200
Column 200x200mm
(Specimen F-0.4) 60
1400
Main bars; 4-D10 210 100
Stirrup; D6 @100
Concrete cover = 15mm Masonry unit
brick dimensions
400

600

400
a) Specimen F-0.4
2300
300 300
530 2000 530
Beam of all specimens except specimen WB
400

600x400mm
Main bars; 10-D22
Stirrup; D13 @100
Concrete cover = 30mm
1400

300

300
400

b) Specimen F-1.5 Column 300x300 mm

(Specimen F-1.5)
Main bars; 8-D16
2300 Stirrup; D10 @80
200 200
580 2100 580 Concrete cover = 15mm
400
1400

200

200
Column 200x200mm
(Specimen F-0.6)
400

Main bars; 4-D16

Stirrup; D10 @50
c) Specimen F-0.6 Concrete cover = 10mm
Fig. 1. Dimensions and reinforcement of specimens of series i) (all dimensions in mm).

2300
200 200
580 2100 580
200
250

200
250

200
Upper Beam Column
1475

200x250mm 200x200mm
(Specimen WB) (Column identical
Main bars; 6-D13 to Specimen F-0.6
Stirrup; D6 @100 in Figure 1-c)
Concrete cover 35mm
400

Fig. 2. Dimensions and reinforcement of specimen WB with weak beam (all dimensions in mm).

influence of weak beams on the seismic capacity and collapse me- relatively smaller and designed to have a beam plastic moment capacity
chanism of infilled RC frames. Specimen WB was designed to be exactly (Mub) to column plastic moment capacity (Muc) ratio of 0.7 (Mub/
identical to specimen F-0.6, with the exception that the beam was Muc = 0.7). The details of specimen WB are shown in Fig. 2.

13
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

2.1.3. Series iii) specimen: Varying mortar strength of masonry infill wall Table 2
In the previously mentioned specimens of series i), the mortar of the Material properties of concrete.
infill walls was of very good quality, with a compressive strength of Specimen name Compressive strength Elastic modulus Split Tensile
approximately 25 MPa. However, in many countries, masonry infill (MPa) (MPa) strength (MPa)
walls are considered to be non-structural walls, and thus low-quality
F-0.4 24.2 2.31 × 104 2.06
mortar is commonly used.
F-0.6 25.5 2.42 × 104 1.94
There have been numerous masonry infill experiments conducted to F-1.5 28.3 2.71 × 104 2.45
study the effect of changing the brick type and strength, including those WM 25.8 2.42 × 104 2.03
of Colangelo [3], Mehrabi et al. [18], Cavaleri et al. [20], and Zovkic WB 23.6 2.37 × 104 1.96
et al. [32]. However, relatively few infilled masonry experiments, as in
the study of Chiou et al. [33], have investigated the influence of the
mortar strength alone without changing the brick type. Lowering the Table 3
compressive strength of the mortar will reduce the compressive Materials properties of masonry.
strength of the masonry prism (fm). This will reduce the shear strength Specimen Masonry prism Mortar Brick
of masonry infill in the case of a diagonal compressive failure of the name
Compressive Modulus of Compressive Compressive
infill strut. Lowering the strength of the mortar will also affect the co-
strength (MPa) Elasticity strength (MPa) strength (MPa)
hesive characteristics of the mortar bed joint and thus reduce the (MPa)
sliding failure capacity of the masonry infill.
On the other hand, the compressive masonry prism tests conducted F-0.4 17.3 7840 20.2 38.1
by Kaushik et al. [34] showed that a masonry prism with a low mortar F-0.6 19.5 10,230 27.7
F-1.5 18.6 8140 29.2
strength has a much greater strain at the maximum compressive load, WM 13.3 5470 4.8
which indicates a greater deformation capacity. One of the most well- WB 19.5 10,230 27.7
recognized methods for modeling masonry infill is a macro-model that
uses the equivalent diagonal compression strut, which is recommended
by several guidelines such as FEMA 306 [8] and the New Zealand Table 4
seismic assessment guideline [35]. Thus, theoretically, decreasing the Mechanical properties of reinforcement.
strength of the mortar will increase the deformation capacity assuming Bar Nominal strength Yield strength (MPa) Ultimate tensile strength (MPa)
the strut behaves as ideally as a compressive prism in material tests. In
this study, one specimen called WM (weak mortar) was constructed D6 SD345 346 546
D10 SD345 384 576
using masonry infill with a very low compressive strength mortar
D13 SD345 380 568
(4 MPa), in order to study the influence of weak mortar on the seismic D16 SD345 380 563
capacity and collapse mechanism of infilled RC frames. Other than the D22 SD390 447 619
mortar strength in the masonry part, the WM specimen was designed to
be exactly the same as specimen F-0.6.
Table 1 summarizes the values for the three parameters and the individually conducted for each specimen simultaneously with the
main differences between the specimens. frame loading. Although the material properties should have been the
same for all of the specified specimens, there were slight differences in
the compressive strengths of the concrete, masonry prism, and mortar,
2.2. Material properties
which were attributed to the time gap between the loading tests for
different specimens.
The masonry panels were built with 60 × 100 × 210 mm solid
bricks. The RC frame was first constructed. Then, a skilled mason
constructed the infill panel. The wall thickness was 100 mm, and the 2.3. Test setup and instrumentation
mortar joint (e.g., head and bed) thickness was approximately 10 mm.
The height for each masonry row was carefully assigned before the Schematic diagrams of the overall test loading frame are shown in
construction of the infill panel. The gap between the beam and last of Figs. 3 and 4. Two vertical hydraulic jacks were employed to apply and
row bricks was 15 mm and filled with mortar. maintain a vertical load of 200 kN on each of the RC columns during the
Tables 2–4 summarize the mechanical properties of the concrete, lateral loading of the frame. This represented a nominal axial load, N/
masonry, and reinforcing steel, respectively. The steel and concrete (fc⋅Ac) (the axial load divided by the concrete compression capacity of
material tests were performed according to the Japanese standard [36]. the columns), of 0.21 on each column for all of the specimens, with the
The computation of the masonry prism compressive strength was in exception of specimen F-1.5, which had a larger column size, where N/
compliance with ASTM C1314 [37]. The concrete used in all of the RC (fc⋅Ac) was 0.10. Two pantographs were attached next to the vertical
frame specimens had the same mix design. The joint mortar consisted of jacks to restrict any out-of-plane movement or torsional displacement.
cement and sand with a mixing ratio of 1:2.5 for all of the specimens, Two horizontal jacks, attached at the beam level, applied an incre-
except specimen WM, where 1:6 is used. The material tests were mental cyclic lateral load. These horizontal jacks worked together, and
the cyclic loading was regulated by the inter-story drift, which refers to
Table 1 the ratio of the lateral story displacement to the frame height measured
Summary of specimen details and varying parameters. at the center of the beam depth (h = 1600 mm), as shown in Fig. 4. The
lateral load was applied as an incremental cyclic load to obtain peak
Series no. Main varying Specimen β-index Ratio Mortar
* story drift angles of 0.05%, 0.1%, 0.2%, 0.4%, 0.6%, 0.8%, 1%, 1.5%,
parameter name Mub/Muc strength
and 2%, where two cycles for each peak drift were adopted. Specimens
i. Strength of RC F-0.4 0.39 9.7 Strong which did not significantly degrade in strength after the 2nd cycle of
columns F-0.6 0.56 5.9 Strong 2% drift were then pushed monotonically until severe damage or sig-
F-1.5 1.51 3.3 Strong
nificant strength degradation was observed.
ii. Weak beam WB 0.43 0.7 Strong
iii. Weak mortar WM 0.82 5.9 Weak The locations of the LVDTs, linear variable displacement trans-
former, are schematically shown in Figs. 5 and 6. The story drift was
* Mub/Muc is the ratio of beam to column‘s flexural moment capacity. measured using LVDTs attached at the vertical midpoint of the beam.

14
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

Fig. 3. Test setup (all dimensions in mm).

Strain gauge

Interval of
280 mm

Fig. 6. Location of strain gauges attached on reinforcement.

Fig. 4. Photograph of test setup.
descriptions of the other three specimens (i.e., F-0.6, WM, and WB) are
LVDTs were attached diagonally on the infill and columns to measure presented in the following.
the corresponding shear deformation. The curvatures of the columns
and beams were measured using pairs of LVDTs distributed uniformly 3.1. Specimen F-0.6
over the length (as shown in Fig. 5). In addition, strain gauges were
attached to at least two main reinforcing bars of each column and At the early stage of loading and a drift angle of approximately
distributed uniformly at 280 mm intervals, as shown in Fig. 6. Strain 0.05%, the mortar bed joints and bricks at the loading corner responded
gauges were also attached to the rebars of beams at the center of the with the onset of a tiny crack and a few diagonal cracks, respectively,
span and at critical sections near the faces of the surrounding columns. with a crack width of approximately 0.05 mm. At a drift angle of 0.1%,
the cracks became more visible on the masonry infill wall, and the
3. Experimental results maximum crack width was 0.5 mm. At drift angles of 0.6%–0.8%, the
reinforcements in the tension column started to yield at the upper end
Fig. 7(a)–(e) show the lateral load–drift angle curves of all the of the column and adjacent to its mid-height, in the fashion of the short
specimens. The crack patterns and failure modes at a drift angle of 2% column failure mechanism, which was similar to the behavior observed
are shown in Fig. 8(a)–(e). The failure mechanism of specimens F-0.4 in specimen F-0.4 [17]. A schematic drawing showing the locations of
and F-1.5 were presented in our previous study [17]. The failure the hinges is shown in Fig. 9(a). The compression column yielded at the

Fig. 5. Schematic diagram of instrumentation.

15
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

400 400
a) Specimen F-0.4 b) Specimen F-0.6
300 300
200 200

lateral load (kN)

lateral load (kN) 100 100
0 0
-100 -100
-200 Max lateral load -200 Max lateral load
+ loading 285 kN + loading 277 kN
-300 -300
- loading 230 kN - loading 294 kN
-400 -400
-3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 3
Story drift (%) Story drift (%)
800 400
c) Specimen F-1.5 d) Specimen WM
600 300
400 200
lateral load (kN)

lateral load (kN)

200 100
0 0
-200 -100
-400 -200
Max lateral load Max lateral load
-600 + loading 571 kN -300 + loading 272 kN
- loading 582 kN - loading 296 kN
-800 -400
-3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 3
Story drift (%) Story drift (%)

400
e) Specimen WB
300
200
lateral load (kN)

100
0
-100
-200
Max lateral load
-300 + loading 258 kN
- loading 259 kN
-400
-3 -2 -1 0 1 2 3
Story drift (%)
Fig. 7. Lateral load-story drift angle curves: (a) F-0.4, (b) F-0.6, (c) F-1.5, (d) WM, and (e) WB.

top and bottom ends (similar to the bare frame behavior). The shear 0.6%, the longitudinal reinforcement in the upper beam yielded at both
cracks were concentrated at the upper thirds of the columns, as shown ends near the joint. At a drift of 0.8%, the tensile column yielded
in Fig. 8(b), and stirrups yielded at a drift angle of 0.8%. The damage to around the vertical mid-point, showing the short column mechanism
the masonry infill was also concentrated at the top corners, with almost similar to specimen F-0.6, and the compression column formed hinges
no damage at the mid-span. The main difference compared to specimen at the top and bottom ends. A schematic drawing showing the locations
F-0.4 was that after attaining the lateral strength, the strength dete- of these hinges is shown in Fig. 9(b). The damage to the infill panel was
rioration was not abrupt. After a drift of 2%, the shear cracks in the concentrated at the upper half and near the loading corners, with al-
upper regions of the columns were extended, the concrete cover spalled most no damage in the middle lower part of the infill panel. The main
off, and the main reinforcement became visible. However, the columns difference from specimen F-0.6 (an identical specimen but with a strong
continued to carry the lateral load, and the longitudinal bars did not beam) was that the strength degraded at a faster rate. At a drift of 2%,
buckle even after a story drift of 3%. the cracks in the beam and columns widened significantly, with a
maximum crack width of approximately 3 mm, but there was no spal-
3.2. Specimen WB (weak beam) ling of the concrete cover at this stage.

In contrast to the previous specimen (F-0.6) with a strong beam, 3.3. Specimen WM (weak mortar)
relatively more cracks occurred at the beam and infill panel at a story
drift of 0.05%, with widths of approximately 0.2 mm for the infill and In contrast to the previous specimen (F-0.6) with stronger mortar
0.05 mm for the beam. There was a slight and sudden drop in strength and an identical RC frame, cracks did not occur at a drift angle of 0.05%
at a story drift of 0.1%, but the strength started to increase again until it in this specimen with weak mortar, but the onset occurred at a 0.1%
reached the maximum strength at a story drift of 0.8%. At a drift of drift. At drift angles of 0.6–0.8%, both columns yielded at the bottom

16
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

Fig. 8. Crack patterns at story drift of 2%: (a) F-0.4, (b) F-0.6, (c) F-1.5, (d) WM, and (e) WB.

and top sections, which were similar to the hinge locations of specimen main variation from specimen F-0.6 was that the surrounding RC col-
F-1.5 [17]. A schematic drawing showing the locations of the hinges is umns did not form hinges at their mid-points, and there was relatively
shown in Fig. 9(c). The lateral load slightly degraded during the posi- much less damage. A comparison of the damage level and cracks at a
tive loading after 1%. However, there was no significant strength de- drift angle of 2% showed that the damage to the masonry infill and RC
gradation during the negative loading, and the maximum strength was columns was relatively much lower than that for the other specimens,
reached at a drift of 1.5%. Even though the mortar strength and ma- with no signs of shear failure or spalling of the masonry or concrete
sonry prism strength were much weaker, as summarized in Table 3, the cover, as shown in Fig. 8.
maximum lateral strength was similar to that of specimen F-0.6. The

17
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

Main reinforcement yielded at story drift 0.6%

Main reinforcement
yielded story drift
0.7~0.8% Main bars yielded at
story drift 0.7% 0.8

800 mm
(a) Specimen F-0.6

Longitudinal reinforcement in beams yielded at drift 0.6%

Main bars in columns

yielded at story drift
0.8%
800 mm

(b) Specimen WB

Main reinforcement
yielded drift 0.6%

Main reinforcement
yielded drift 1% Main reinforcement
yielded at story drift
300 mm

0.8%~1%

(c) Specimen WM
Fig. 9. Schematic diagram of hinge locations for positive loading direction: (a) F-0.6, (b) WB and (c) WM.

4. Analysis of experimental data

0.1%
300
4.1. Initial stiffness

The initial stiffness was determined using the slope of the experi- 200
mental load–displacement curve between the origin and the point
where cracks on the masonry and frame became visible. A significant 100
Lateral load (kN)

stiffness degradation was also considered when determining this point,

which was estimated to be at a drift angle of 0.1%, as shown in the
0
backbone curves of Fig. 10. Even though the investigated parameters
influenced the strengths and deformation capacities of the specimens,
F-0.4
these parameters did not significantly influence the initial stiffness. -100
There is still significant variation in the existing proposed methods F-0.6
for estimating the initial stiffness. The most common method uses the -200 WB
concept of a diagonal compression strut, but the appropriate strut width
WM
is a controversial topic among many researchers. In this paper, some 0.1%
-300
recent and well-known design codes and methods are briefly introduced -2 -1.5 -1 -0.5 0 0.5 1 1.5 2
and investigated. Story drift (%)

Fig. 10. Lateral load- story drift angle backbone curve of all specimens.
(1) ASCE/SEE 41(2006) [25]

18
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

ASCE/SEE 41 [25] adopts a method to calculate the diagonal respectively. However, the simplified method of Paulay and Priestley
compression strut width at the initial stiffness based on an empirical [6] gave relatively better results than any of the other more complex
relation using the stiffness ratio of the frame to masonry infill, as shown empirical methods, with an average experimental to analytical ratio of
in Eqs. (3) and (4). unity.
Winf = 0.175·( hH ) 0.4 ·d
m (3)
4.2. Strength
1
Einf tinf sin(2 ) 4
h = The load–lateral drift curves show the strength of the entire com-
4Ec Ic h inf (4)
posite structure. To evaluate/compute the strengths of the masonry
where Winf is the equivalent strut width; tinf is the infill thickness; Einf infill and RC frame, it is essential to split the contributions of the
and Ec are the elasticity moduli of the infill and concrete, respectively; bounding RC frame and infill panel to the lateral resisting capacity. In
hinf and H are the clear height of the infill and the story height, re- this study, two methods were used: a simplified approach and detailed
spectively; dm is the panel's diagonal length; θ is the diagonal inclina- analysis approach.
tion angle of the infill; and Ic is the moment of inertia of the RC column.
4.2.1. Simplified approach
(2) New Zealand seismic assessment standard for existing buildings The maximum lateral load taken by the masonry infill itself is as-
(2017) [35] sumed to be equal to the difference between the bare frame capacity
and maximum experimental lateral load, as shown in Eq. (11). The bare
The New Zealand seismic assessment standard [35] uses a modified frame lateral strength was calculated using Eq. (12).
version of Eq. (3), as shown in Eq. (5).
Vinf = Vmax Vf (11)
Winf = 0.18·( h ·H ) 0.25 ·d
m (5)
Vf = 4Mu/H (12)
(3) Masonry Society Joint Committee (MSJC)-2016 [27] where Vmax is the maximum lateral load of the structure and Mu is the
plastic moment capacity of the beam or column, whichever is lower, in
The MSJC [27] proposed a concept similar to that of ASCE/SEE 41 compliance with the AIJ standard [39]. H is the clear height of a
[25], where the strut width is related to the empirical ratio of stiffness column in the RC frame in the case of a weak column and strong beam.
λh in Eq. (4), but proposed another method for the strut width, as shown For a case where the flexural capacity of the beam is larger than that of
in Eq. (6). the column, H is taken as the height up to the centerline of the girder.
Winf = 0.3/ h ·cos (6) In this study, the maximum load occurred at a drift of 0.8%. Before
this drift, the RC columns had already yielded at a drift of approxi-
(4) Paulay and Priestley (1992) [6] mately 0.6% which indicated that the frame was approaching its
maximum strength after the 0.6% drift. Therefore, subtracting the bare
Paulay and Priestley [6] used a simplified assumption, as shown in frame strength from the maximum lateral load, at approximately 0.8%,
Eq. (7), where the strut width at the initial stiffness is not related to the is supposed to be an acceptable assumption and employed herein as a
stiffness or strength of the frame. comparison benchmark. This method has been employed by several
researchers [15,18].
Winf = 0.25·dm (7)
The masonry shear strengths computed using the simplified ap-
proach are summarized in Table 6. The masonry shear strength, τinf, is
(5) Fiorato et al. (1970) [38] calculated by dividing Vinf by the cross-sectional area of the infill
(linf⋅tinf). The shear strengths of all the specimens, except for F-1.5, were
Fiorato et al. [38] proposed a method that considers the structure to in the same range of 0.8–0.9 N/mm2. However, specimen F-1.5 had a
be a composite system, where the beam and columns are the flanges masonry shear strength that was approximately 1.5 times that of the
and the masonry panel is the web. other specimens. Fig. 11 shows the relation between the masonry infill
1 shear strength, τinf (normalized by prism compressive strength, fm), and
Kini =
(1/ Ksh)+(1/ Kfl ) (8) the parameter β index. As shown in Fig. 11, there is a clear relation
between an increase in the β index and an increase in the shear
Ksh = Ainf ·Ginf / h inf (9) strength. The large improvement in shear strength could be attributed
to the greater confinement effect of the stronger surrounding frame
Kfl = 3Ec· I /h3 (10)
owing to the increased length of the contact between the frame and
In Eqs. (8)–(10), when calculating the shear stiffness, Ksh, only the masonry infill (less separation gap), which increased the strut width
masonry panel is considered, where Ainf is the cross-sectional area of the and lateral capacity.
infill, and Ginf is the shear modulus taken as 0.4Einf. For the flexural The failure mechanisms observed in the experimental results were a
stiffness calculation, Kfl, the whole composite section is used. Here, I mixture of both compression and sliding failures, with a different ratio
denotes the equivalent moment of inertia of the transformed section of these for each specimen. It was observed that the failure mechanisms
considering the elastic moduli of the concrete and masonry infill. This were also influenced by the β index, as shown in Fig. 8. Specimen F-1.5,
method does not use the compression strut concept, but it is included in which had a very large β index (a relatively strong and stiff RC frame),
this paper because many recent studies have used it to estimate the had a relatively greater number of diagonal compression failures. It is
initial stiffness. thought that the stronger surrounding frame constrained the movement
The initial stiffness of the RC frame was calculated using the gross and deformation of the infill panel, causing the infill to undergo diag-
concrete section of the RC members in the case of the methods that use onal compression failure. However, the number of specimens with large
the strut mechanism concept. The values for the ratio of the experi- β index values in this study was limited, and further verification is
mental to analytical initial stiffness values are compared in Table 5. needed.
The method of Fiorato et al. [38] overestimates the initial stiffness. Interestingly, there was no significant decrease in the shear strength
On the other hand, the methods of ASCE/SEE 41 [25] and MSJC-2016 of the masonry infill with weak mortar. One explanation for this would
[27] greatly underestimate the initial stiffness by ratios of 1.8 and 2.5, be that the relatively weaker masonry infill (larger β index) had a wider

19
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

Table 5
Comparison of experimental and analytical initial stiffness.
Specimen name Experimental. initial stiffness (KN/ Ratio = Experimental/Analytical
mm)
ASCE/SEI 41-06 New Zealand Standard MSJC (2016) Paulay et al. (1992) Fiorato et al. (1970)
(2007) [25] (2017) [35] [27] [6] [38]

F-0.4 128 1.9 1.6 2.8 1.0 0.3

F-0.6 145 1.8 1.5 2.8 0.9 0.3
F-1.5 166 1.2 1.1 1.3 0.8 0.4
WM 125 2.3 2.0 3.1 1.6 0.4
WB 131 1.7 1.4 2.7 0.8 0.3
Average 1.8 1.5 2.5 1.0 0.3
Standard deviation 0.4 0.3 0.6 0.3 0.1

strut width, allowing the infill panel to carry a greater lateral load. 0.1

τinf /fm (Normalized shear strength )

0.08
4.2.2. Detailed analysis approach
Several researchers in previous studies commonly employed the
0.06
aforementioned simplified approach that considers the strength of the
RC frame to be that of a bare frame. However, the computation of the Specimen F-0.4
0.04
real lateral strength of an infilled RC frame can be more complicated Specimen F-0.6
than merely finding the sum of the shear strength of the bare frame and Specimen F-1.5
masonry infill, because of the flexural and shear deformations by the 0.02
Specimen WB
complex frame-masonry interaction, shifting of the plastic hinge loca- Specimen WM
tions of the frame owing to the forces exerted by the infill panel, and 0
0 0.5 1 1.5 2
internal variation in the axial load on the columns. Therefore, a more β index
detailed investigation of the RC frame and masonry infill was per-
formed, and the results are reported in this section. Fig. 11. Relation of shear strength of infill, normalized by prism compressive
The lateral load carried by the RC frame at a certain drift angle was strength, with β index.
calculated by a sectional analysis using the measured strain values and
curvature of the RC frame. The strains of the reinforcing bars were sectional analysis are included in the Appendix A.
recorded by strain gauges attached to the longitudinal reinforcing bars For all the specimens, the values for the lateral load as a function of
at different intervals along the height of the columns, as shown in the story drift are shown in Fig. 12(a)–(e). Because the hysteresis curves
Fig. 6. At each story drift angle during loading, the recorded strain of all the specimens, as shown in Fig. 7, are almost symmetric, only
values of the longitudinal reinforcements of the columns were analyzed positive cycles are shown in Fig. 12
to calculate the instantaneous bending moment diagram and curvature Similar to the results of the simplified approach, the lateral capacity
distribution along the column height for each column. Subsequently, of the infilled masonry was considerably enhanced with an increase in
the lateral load carried by each RC column was determined, based on the relative shear strength of the bounding RC frame (an increase in the
basic mechanics, from the exerted end moments and effective height of β index). Table 7 summarizes the actual shear stress of an infill panel
the column using Eq. (13). (τinf), which corresponds to the maximum shear force (Vinf) taken by the
infilled panel divided by the infill cross-sectional area. The shear
Mc , top + Mc, bottom
Vc = strengths of all the specimens, except for F-1.5, are in a range of
ho (13) 0.9–1.1 N/mm2. Specimen F-1.5 had a shear capacity that was ap-
proximately 1.8 times the average of the aforementioned range. In
Here, Vc is the lateral load carried by one RC column of the sur-
general, the simplified approach, which assumed that the strength of
rounding frame, Mc is the bending moment exerted on the column (top
the surrounding frame was that of the bare one, gave a slightly lower
or bottom), and ho is the effective height of the column. The effective
shear strength for the masonry infill, which could be considered to be a
height of a column was calculated using the curvature distribution
conservative value. In addition, the masonry infill shear strength τinf
along the column height and the hinge location.
had a range of 0.04–0.08 fm. This range is similar to the results for 24
After evaluating the load carried by the RC frame at a certain drift,
specimens from various researchers with different masonry material
the contribution of the masonry at that drift could be calculated by
types compared in the previous study [29], where the average shear
subtracting the evaluated frame strength from the experimental
strength τinf was approximately 0.05 fm.
strength of the infilled frame. The models and assumptions used in the

Table 6
Shear strength of encased masonry of masonry infilled RC frame (simplified approach).
Specimen name Infilled RC frame Bare frame Masonry infill

Experimental load resistance, Vmax (KN) Lateral capacity, Vf (KN) Lateral load resistance, Vinf (KN) Shear resistance, τinf (MPa)

+ve loading −ve loading – +ve loading −ve loading Avg. +ve loading −ve loading Avg.

F-0.4 285 230 71 214 159 187 1.02 0.76 0.89

F-0.6 277 294 113 164 181 173 0.78 0.86 0.82
F-1.5 571 582 280 291 302 297 1.46 1.51 1.48
WM 272 296 113 159 183 171 0.76 0.87 0.81
WB 258 259 88 170 171 171 0.81 0.81 0.81

20
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

300 300
250 250

Lateral load (kN)

Lateral load (kN)

200 200
150 150
100 100
50 50
0 0
0 0.4 0.8 1.2 1.6 2 0 0.4 0.8 1.2 1.6 2
Story drift % Story drift %
(a) Specimen F-0.4 (b) Specimen F-0.6

600 300
500 250

Lateral load (kN)

Lateral load kN

400 200
300 150
200 100
100 50
0 0
0 0.4 0.8 1.2 1.6 2 0 0.4 0.8 1.2 1.6 2
Story drift % Story drift %
(c) Specimen F-1.5 (d) Specimen WM

300
250
Lateral load (KN)

200
150
Infilled frame (experiment)
100 Masonry wall
50 RC frame
0
0 0.4 0.8 1.2 1.6 2
Story drift (%)
(e) Specimen WB
Fig. 12. Backbone curve of bounding RC frame, masonry infill and infilled frame: (a) F-0.4, (b) F-0.6, (c) F-1.5, (d) WM, and (e) WB.

Table 7
Shear strength of encased masonry of masonry infilled RC frame (detailed analysis approach).
Specimen name Load carried at the peak resistance of masonry (KN) Shear resistance of masonry, τinf (MPa) Average of both directions, τinf (MPa)

Infilled RC frame Bounding RC frame Masonry infill

F-0.4 +ve loading 285 45 240 1.14 0.99

−ve loading 230 56 174 0.83

F-0.6 +ve loading 261 34 227 1.08 1.06

−ve loading 294 75 219 1.04

F-1.5 +ve loading 558 184 374 1.87 1.89

−ve loading 582 200 382 1.91

WM +ve loading 251 59 192 0.91 0.98

−ve loading 296 78 218 1.04

WB +ve loading 258 72 186 0.89 0.91

−ve loading 259 64 195 0.93

21
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

Table 8 SEI 41 [25], where the masonry infill deformation points are de-
Comparison of predicted deformation limits with experimental result. termined based on the β index, but it only considers β index values
Specimen Predicted drift limit (%) Experimental drift Ru (%) (at greater than unity, which is the case when the frame's lateral strength is
name 80% of max strength) larger than the strength of the masonry infill.
ASCE/SEI New Zealand Table 8 compares the experimental results for drift Ru and the drift
41-06 [25] standard [35] point at which a sudden degradation is expected to occur, as suggested
F-0.4 0.33 – 0.9 by ASCE/SEI 41 [25] and the New Zealand seismic assessment standard
F-0.6 0.33 – 1.5 [35]. Based on the experimental results, both ASCE/SEI 41 [25] and the
F-1.5 1.02 0.68 1.5 New Zealand seismic assessment [35] proved to be on the safe side but
WM 0.68 – 2 overly conservative.
WB 0.34 – 1.5

5. Conclusions
4.3. Ductility
This study presented the results of an experimental study of three
In this study, Rcrack, Rmax, and Ru represent the drift angles corre- masonry infilled RC frame specimens, with the following major find-
sponding to the cracking point, maximum lateral load point, and post- ings:
peak degradation point, respectively. The post-peak degradation drift
Ru on the degradation branch was set to be the drift corresponding to (a) Experimental results demonstrated that with an increase in the re-
80% of the maximum strength. Here, Ru does not correspond to the safe lative lateral strength of the frame, compared to masonry, there was
displacement limit for the masonry infilled frame and is considered a considerable improvement in the seismic capacity of the masonry
herein for comparison purposes. infill walls in terms of the strength and ductility of the masonry
Rcrack was estimated to be in the range of 0.1–0.2% for all the spe- infill. This parameter is not taken into account when calculating the
cimens in this study, where small cracks became clearly visible. From strength of masonry using the design codes but should be con-
the computed backbone curves of the infilled masonry panels of all the sidered when designing RC buildings with masonry infill because it
specimens, as shown in Fig. 12(a)–(e), it is evident that irrespective of has the potential to change the brittle behavior into a relatively
the varying parameters, i.e., the relative strength of the surrounding ductile behavior. It should be noted that the number of specimens
frame, capacity of the beam, and joint mortar strength, the infilled in this study was limited, and further investigations are necessary to
masonry panel started to crack at a drift of 0.1–0.2%, which was also account for the influence of the investigated parameters in design
observed in the experiment. The Rcrack range could be assumed to be the codes.
immediate occupancy limit for masonry infilled RC frames. (b) There is a large variation in the methods used by different recent
The Rmax (drift at peak strength) values for all the specimens were codes and researchers to estimate the initial stiffness. A simplified
estimated to be approximately 0.8%, except for specimen WM with assumption of a strut width of 0.25 times the infill diagonal length,
weak mortar in the negative cycle, which was found to be 1.5%. as suggested by Paulay and Priestley [6], gave relatively better
For Ru in specimen F-0.4 (with a very weak column relative to the results than more empirical complex prediction models.
masonry infill), there was a sudden drop in strength at a story drift (c) Interestingly, specimen WM (built with weak mortar) showed
angle of approximately 0.9%. The Ru values of specimens F-0.6, F-1.5, beneficial influences from weak mortar because there was less
and WB were approximately 1.5%. Based on the results, varying the strength degradation and relatively less damage to the masonry
surrounding frame strength had a slight influence on Rcrack and Rmax. infill panel and surrounding frame. However, it should be noted
However, it greatly influenced the strength degradation slope and Ru. that the use of very weak mortar might have an adverse effect on
Specimen WM (weak mortar) showed the beneficial influence of other aspects such as the out-of-plane capacity, which was out of
weak mortar, because there was almost less degradation in the strength the scope of this study. In addition, the in-plane seismic capacity of
and relatively less damage in the masonry infill panel and surrounding masonry infill using very weak mortar below a certain lower limit
frame. However, it should be noted that the use of weak mortar might needs further study.
have adverse effects on other aspects that are outside of the scope of
this study, such as the out-of-plane capacity, which needs further study.
Few of the design codes and published papers explicitly state the Acknowledgements
displacement margin of masonry infill. ASCE/SEI 41 [25] gives the drift
points where a sudden strength degradation of the masonry infill occurs This research is supported by SATREPS-TSUIB project lead by Prof.
based on the frame strength to masonry strength ratio, β index, and Nakano Yoshiaki, Tokyo University and JSPS KAKENHI Grant Number
aspect ratio of the infill panel. The updated seismic assessment guide- JP18H01578 (Principal investigator: Prof. Masaki Maeda, Tohoku
lines for New Zealand [35] use a concept similar to that of ASCE/ University).

Appendix A

A.1. Contribution of bounding frame to lateral load resistance

A.1.1. Estimation of bending moment diagram

The tensile and compressive reinforcement strain values of columns were employed to obtain the instantaneous strain profile for a column section
using the plane section assumption. Based on the concrete strains, the Todeschini continuous curve [40] of concrete stress distribution was used to
estimate the average compressive stress on the concrete using Eqs. (14) and (15). A schematic diagram of the strain and stress distribution on the
column section is shown in Fig. 13. This stress block was further used for the calculation of the resultant compressive force on the concrete.
ln[1 + ( max / o)2]
f¯c = · fc
( max / o ) (14)

22
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

εc’
εc
εt εmax
Fc
T C fc
c

c
a

Fig. 13. Schematic diagram of stress-strain distribution on column section.

1.4
1.2
Height x 103 (mm)

1 ho ≈ 800
0.8
0.6
0.4
0.2
0
-60 -30 0 30 60 -1.0 -0.5 0.0 0.5 1.0
Moment (KN-m) Curvature x 10-4

(a) Tension column

1.4
1.2
Height x 103 (mm)

1
ho ≈1400
0.8
0.6
0.4
0.2
0
-60 -30 0 30 60 -1.0 -0.5 0.0 0.5 1.0
Moment (KN-m) Curvature x 10-4

(b) Compression column

Moment Ultimate moment
Curvature Yield curvature
Fig. 14. Moment and curvature distribution of (a) tensile and (b) compression column at 0.8% drift angle for F-0.6 specimen.

2{( max / o ) tan 1 ( max / o )}

a= 1 2
·c
( max / o ) 1 (15)
Here, f¯c is the average stress on the concrete, fc is the stress at the outer face of the column, εmax is the strain at the outer face of the column, and εo
is the strain at the maximum stress. The symbols a and c are self-explanatory in reference to Fig. 13.
This is followed by the determination of the forces acting on steel bars based on the strain values of the reinforcement considering a bilinear
model. After obtaining the internal concentrated loads on the concrete and steel, along with the length of the compression zone of the stress block,
the bending moments exerted by the lateral loads at the locations where the strain gauges were attached to the reinforcements were calculated. Using
the moment at each section, a bending moment diagram of the entire column at a certain drift angle was estimated. It should be noted that the strain
gauge data were used up to a drift angle of 1%, because the reinforcements yielded at approximately this drift. After this, the strain gauges might not
have provided reliable results.

A.1.2. Estimation of column's effective height

The effective height was estimated from the distribution of the curvature, along with the bending moment diagram along the column height. The
curvature (φ) at a certain height was evaluated using the strains recorded by two strain gauges affixed to the tensile and compression longitudinal

23
H. Alwashali, et al. Engineering Structures 189 (2019) 11–24

bars at that level using Eq. (16).

t c
=
l (16)
Here, εt and εc are the strains of the tensile and compressive reinforcements, respectively, and l is the center-to-center distance of the re-
inforcements.
The moment and curvature distributions, along with the ultimate moment capacity and yield curvature of specimen F-0.6 at the 0.8% drift angle,
are shown in Fig. 14(a) and (b). The moment distribution and curvature distribution along the column height show similar tendencies. Hence, the
curvature distribution was employed to determine the effective height of the column. The yield curvature of column has been set as the criteria to
find out the location of hinge formation. The yield curvature was calculated using the yield strain of the column reinforcement [41]. Utilizing the
aforementioned criteria, the effective clear heights of the tension and compression column of specimen F-0.6 were estimated to be 0.82 m and 1.4 m,
respectively, which resembled the experimental observations discussed earlier. Hence, for the other specimens, the same procedure was followed to
evaluate the effective height of column. It should be noted that in specimen WB, hinges were formed at the beam and bottom of the columns.
Therefore, the effective height was considered to be from the hinge formed in the beam to the bottom of the column.

References Experimental results and simplified modeling. Soil Dyn Earthq Eng
2014;65:224–42.
[21] Wood RH, Plasticity BRE. composite action and collapse design of unreinforced
[1] Pujol S, Fick D. The test of a full-scale three-story RC structure with masonry infill shear wall panels in frames. Proc Inst Civ Eng 1978;65(2):381–411.
walls. Eng Struct 2010;32(10):3112–21. [22] Liauw TC, Kwan KH. Unified plastic analysis for infilled frames. J Struct Eng
[2] Stavridis A, Koutromanos I, Shing PB. Shake-table tests of a three-story reinforced 1985;111(7):1427–48.
concrete frame with masonry infill walls. Earthq Eng Struct Dyn [23] Saneinejad A, Hobbs B. Inelastic design of infilled frames. J Struct Eng
2012;41(6):1089–108. 1995;121(4):634–50.
[3] Colangelo F. Pseudo-dynamic seismic response of reinforced concrete frames in- [24] Mainstone RJ. Summary of paper 7360. On the stiffness and strengths of infilled
filled with non-structural brick masonry. Earthq Eng Struct Dyn frames. Proc Inst Civ Eng 1971;49(2):230.
2005;34(10):1219–41. [25] American Society of Civil Engineers. Seismic rehabilitation of existing buildings.
[4] De Risi MT, Del Gaudio C, Ricci P, Verderame GM. In-plane behaviour and damage Reston, Virginia: ASCE Publications; 2007 [ASCE/SEI 41-06].
assessment of masonry infills with hollow clay bricks in RC frames. Eng Struct [26] Flanagan RD, Bennett RM. In-plane behavior of structural clay tile infilled frames. J
August 2018;168(1):257–75. Struct Eng 1999;125(6):590–9.
[5] Milanesi RR, Morandi P, Magenes G. Local effects on RC frames induced by AAC [27] Masonry Standards Joint Committee (MSJC). Building Code Requirements and
masonry infills through FEM simulation of in-plane tests. Bull Earthq Eng Specification for Masonry Structures and Related Commentaries in the United
2018;16:4053. https://doi.org/10.1007/s10518-018-0353-5. States; 2016.
[6] Paulay T, Priestley MJN. Seismic design of reinforced concrete and masonry [28] Science and Technology Research Partnership for Sustainable Development
buildings. John Wiley & Sons; 1992. (SATREPS). Technical development to upgrade structural integrity of buildings in
[7] Crisafulli FJ. Seismic behaviour of reinforced concrete structures with masonry densely populated urban areas and its strategic implementation towards resilient
infills PhD Thesis New Zealand: University of Canterbury; 1997. cities. https://www.jst.go.jp/global/english/kadai/h2712_bangladesh.html; 2015.
[8] Applied Technology Council (ATC). Evaluation of earthquake damaged concrete [29] Al-Washali H, Suzuki Y, Maeda M. Seismic evaluation of reinforced concrete
and masonry wall buildings; 1998 (FEMA 306). buildings with masonry infill wall. In: The 16th world conference of earthquake
[9] Shing PB, Mehrabi AB. Behaviour and analysis of masonry-infilled frames. Prog engineering, Chile; 2017.
Struct Mat Eng 2002;4(3):320–31. [30] Public Works Department, Bangladesh. Manual for Seismic Evaluation of Existing
[10] Mohammad AF, Faggella M, Gigliotti R, Spacone E. Seismic performance of older R/ Reinforced Concrete Building. CNCRP project; 2015.
C frame structures accounting for infills-induced shear failure of columns. Eng [31] Tempestti M, Stavridis A. Simplified method to assess lateral resistance of infilled
Struct 2016;122:1–13. reinforced concrete frames. In: Accepted 16th World Conference in Earthquake
[11] Mohammad AF, Faggella M, Gigliotti R, Spacone E. Effects of bond-slip and ma- Engineering, Chile; 2017.
sonry infills interaction on seismic performance of older R/C frame structures. Soil [32] Zovkic J, Sigmund V, Guljas I. Cyclic testing of a single bay reinforced concrete
Dyn Earthq Eng 2018;109:251–65. frames with various types of masonry infill. Earthq Eng Struct Dyn
[12] Romano F, Faggella M, Gigliotti R, Zucconi M, Ferracuti B. Comparative seismic loss 2013;42(8):1131–49.
analysis of an existing non-ductile RC building based on element fragility functions [33] Chiou TC, Hwang SJ. Tests on cyclic behavior of reinforced concrete frames with
proposals. Eng Struct 2018;177:707–23. brick infill. Earthq Eng Struct Dyn 2015;44(12):1939–58.
[13] Cavaleri L, Di Trapani F. Prediction of the additional shear action on frame mem- [34] Kaushik HB, Rai DC, Jain SK. Stress-strain characteristics of clay brick masonry
bers due to infills. Bull Earthq Eng 2015;13(5):1425–54. under uniaxial compression. J Mater Civ Eng 2007;19(9):728–39.
[14] Di Trapani F, Shing PB, Cavaleri L. A macro-element model for in-plane and out-of- [35] New Zealand Society for Earthquake Engineering. The seismic assessment of ex-
plane responses of masonry infills in frame structures. J. Struct. Eng. isting buildings, Part C, C7; 2017.
2018;144(2):04017198. [36] Japanese Standard Association. Japanese Industrial Standards. 2010 (No. JIS
[15] Turgay T, Durmus MC, Binici B. Ozcebe G. Evaluation of the predictive models for R1250, JIS Z 2201, JIS A 1108, JIS A 1113) [in Japanese].
stiffness, strength, and deformation capacity of RC frames with masonry infill walls. [37] American Society of Testing and Materials (ASTM). Standard test method for
J Struct Eng 2014;140(10):06014003. compressive strength of masonry prisms; 2011 (ASTM C1314).
[16] Chrysostomou CZ, Asteris PG. On the in-plane properties and capacities of infilled [38] Fiorato AE, Sozen MA, Gamble WL. An investigation of the interaction of reinforced
frames. Eng Struct 2012;41:385–402. https://doi.org/10.1016/j.engstruct.2012.03. concrete frames with masonry filler walls. University of Illinois Engineering
057. Experiment Station. College of Engineering. University of Illinois at Urbana-
[17] Alwashali H, Torihata Y, Jin K, Maeda M. Experimental observations on the in- Champaign; 1970.
plane behaviour of masonry wall infilled RC frames; focusing on deformation limits [39] Architectural Institute of Japan (AIJ). AIJ standard for lateral load-carrying capa-
and backbone curve. Bull Earthq Eng 2018;16(3):1373–97. city calculation of reinforced concrete structures; 2016 [in Japanese].
[18] Mehrabi AB, Benson Shing P, Schuller MP, Noland JL. Experimental evaluation of [40] Wight JK, MacGregor JG. Reinforced concrete- Mechanics and Design. Pearson
masonry-infilled RC frames. J Struct Eng 1996;122(3):228–37. Education Inc.; 2009.
[19] Basha SH, Kaushik HB. Behavior and failure mechanisms of masonry-infilled RC [41] Priestley MN, Seible F, Calvi GM. Seismic design and retrofit of bridges. John Wiley
frames (in low-rise buildings) subject to lateral loading. Eng Struct & Sons; 1996.
2016;111:233–45.
[20] Cavaleri L, Di Trapani F. Cyclic response of masonry infilled RC frames:

24