Engineering Structures 56 (2013) 897–909

Contents lists available at SciVerse ScienceDirect

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

Experimental investigation on reinforced concrete interior

beam–column joints rehabilitated by ferrocement jackets
Bo Li a, Eddie Siu-shu Lam a,⇑, Bo Wu b, Ya-yong Wang c
a
Dept. of Civil & Environmental Engineering, The Hong Kong Polytechnic Univ., Hong Kong, China
b
School of Civil Engineering & Transportation, South China Univ. of Technology, Guangzhou, China
c
Institute of Earthquake Engineering, China Academy of Building Research, Beijing, China

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

Article history: In this study, a method for rehabilitating reinforced concrete interior beam–column joints using ferroce-
Received 24 August 2012 ment jackets with embedded diagonal reinforcements is proposed. It improves seismic performance of
Revised 22 May 2013 substandard beam–column joints and repairs deteriorated concrete cover without increasing the dimen-
Accepted 24 May 2013
sions of the joints. Ferrocement, comprising mortar and wire mesh, was applied to replace concrete cover
to enhance shear strength of the joints. Diagonal reinforcements were installed to reduce the forces trans-
ferred to the joint core. Four 2/3 scale interior beam–column joints, including one control specimen and
Keywords:
three strengthened specimens, were prepared and tested under quasi-static cyclic loading. Three types of
Reinforced concrete
Beam–column joints
mortars were considered for each strengthened specimen. Test results have indicated that the proposed
Rehabilitation rehabilitation method can improve seismic performance of interior beam–column joints using ferroce-
Shear strength ment with high strength mortar. Strength of mortar is the vital factor affecting the performance of
Ferrocement jackets strengthened specimens. Anchor bolts installed at the interface between ferrocement and concrete sub-
Diagonal reinforcements strate improve bonding and overall performance. Finally, a method for predicting the shear strength of
joints rehabilitated by ferrocement jackets with embedded diagonal reinforcements is proposed.
Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction
In reinforced concrete structures, beam–column joints are crit-
ical members for transferring forces and moments between beams
and columns. Due to the moments reversal across beam–column
joints when subjected to seismic action, higher joint shear forces
are formed in the joint cores. As a result, beam–column joints
are designed to have sufficient strength to maintain the stability
and integrity of the structures. However, buildings in areas of
low to moderate seismic risk, such as those in Hong Kong and Sin-
gapore [1,2], were traditionally designed without seismic provi-
sions, i.e. designed to gravity load and wind load only. Such
beam–column joints become vulnerable members in moment
resisting structures and display poor performance under seismic
action according to post-earthquake investigations [3,4] and
experimental study [5]. Furthermore, substandard beam–column
joints are common in old buildings of which are now facing the
problem of aging or approaching the end of their design life span.
This increases the probability of failure of beam–column joints and
ultimately collapse of structures under seismic action [6]. Therefore,
it is necessary to rehabilitate existing substandard beam–column
⇑ Corresponding author. Tel.: +852 2766 6058; fax: +852 2334 6389.
E-mail address: cesslam@polyu.edu.hk (E.S.-s. Lam).
joints for enhancing their seismic performance and extending their
design life span.
Several rehabilitation methods for beam–column joints, includ-
ing concrete jacketing, steel jacketing and fiber reinforced polymer
(FRP) wrapping, etc., have been proposed. Concrete jacketing is
effective for upgrading beam–column joints due to compatibility
with the original structure [5,7,8]. However, it requires labor-
intensive procedures to construct, increases the member sizes
and reduces the usable floor areas. Besides, dynamic properties
of strengthened members may be altered [9]. Steel jacketing [10],
FRP wrapping [11–17] and a combined method using both steel
jacketing and FRP wrapping [18] have the advantage of high
strength and eliminate some limitations of concrete jacketing.
However, they have poor fire resistance due to strength degrada-
tion of resin under moderate temperature. Proper insulation is re-
quired [19]. Further, FRP wrapping is subjected to limitations on
construction (e.g. obstructed by beams and slabs) causing difficul-
ties in providing sufficient anchorage [9]. All in all, the above are
unattractive to building structures by reasons of complicated to
construct, increased member sizes and/or poor resistance to fire.
With due consideration on simplicity and constructability, a
rehabilitation method for beam–column joints using ferrocement
jackets with embedded diagonal reinforcements is proposed. Fer-
rocement was applied to replace the concrete cover in the joint
region to enhance shear strength. It is defined as ‘‘a type of thin

0141-0296/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.engstruct.2013.05.038
898 B. Li et al. / Engineering Structures 56 (2013) 897–909

wall reinforced concrete commonly constructed of hydraulic ce-

ment mortar reinforced with closely spaced layers of continuous
and relatively small size wire mesh’’ [20]. The use of uniform and
close dispersed wire mesh in both directions equips ferrocement
with homogenous and isotropic properties. As a thin-walled struc-
ture within thickness compatible to the concrete cover, ferroce-
ment does not increase the member sizes. It can be easily
applied without formwork and to any shape [21,22]. Furthermore,
geometry and mass of existing structures are unmodified, dynamic
characteristics of the structures are unchanged.
Tests on reinforced concrete columns and beams strengthened
by ferrocement have shown significant enhancement in strength
[22–24]. In addition, diagonal reinforcements were embedded in
beam–column joint to reduce the force transferred to the joint
core. Among others, Tsonos et al. [25] and Chalioris et al. [26] have
demonstrated that joints with crossed and inclined reinforcements
exhibit improved seismic performance.
To assess the effectiveness of the proposed rehabilitation
method, four 2/3 scale beam–column joints, including one control
specimen and three strengthened specimens, were tested under
quasi-static cyclic loading. Three types of mortars, including ce-
ment–sand mortar, cementitious mortar and epoxy-based mortar,
were used for strengthening. Seismic performances in terms of Fig. 2. Dimensions and reinforcement details of specimens.

ultimate loading capacity, ductility, energy dissipation, stiffness

and joint shear strength were evaluated. Finally, a method is pro-
reinforcements (4T16) are provided as top and bottom longitudinal
posed for predicting the shear strength of joints strengthened by
reinforcements (or 1.35% longitudinal reinforcement ratio) in the
the proposed rehabilitation method.
beams. Transverse reinforcements in both columns and beams
are R8 ties at 150 mm spacing and are reduced to 100 mm spacing
at 400 mm from the ends. The transverse reinforcement ratios are
2. Experimental program
0.22% for both beams and columns. All specimens were designed
without transverse reinforcements in the joints.
2.1. The specimens
Specimens were casted using ready-mixed concrete. Measured
cubic concrete strengths estimated on the day of testing are given
Four 2/3 scale beam–column joints, named as C1, S1, S2 and S3,
in Table 1. Measured yield strength and ultimate strength of longi-
were prepared. Specimen C1 was the control specimen while the
tudinal reinforcements are 540 MPa and 642 MPa respectively.
other three specimens were strengthened by ferrocement jackets
Measured yield strength and ultimate strength of transverse rein-
with embedded diagonal reinforcements. The specimens represent
forcements are 410 MPa and 521 MPa respectively.
beam–column joints with smaller-size transverse beams (Fig. 1).
Three types of mortars were used in the ferrocement, including
For simplicity, slabs are not modeled. Nevertheless, tests on ideal-
cement–sand mortar, cementitious mortar and epoxy-based mor-
ized plane beam–column joints can help us to understand effec-
tar. Measured cubic strength of each mortar is given in Table 1.
tiveness of the proposed rehabilitation method.
Weight proportion of the cement–sand mortar was 1 part of ce-
Dimensions and reinforcement details of specimens are shown
ment to 2.5 parts of sand and with 15% of cement replaced by
in Fig. 2. Ends of columns and beams are points of contra-flexure.
PFA. The other two types of mortar were commercial products
The columns are 2385 mm high with a cross section of 300 mm
available in the market. Welded square mesh had averaged wire
by 300 mm. Overall lengths of the beams are 2700 mm with a cross
diameter of 1.45 mm and spacing at 13.23 mm in both directions.
section of 300 mm by 400 mm. Main reinforcements in the col-
Measured ultimate strength of single wire is 350 MPa in both
umns are 12T16 (or 2.7% longitudinal reinforcement ratio). Same
directions. Ultra high strength steel bars were used as diagonal
reinforcements with measured yield strength at 800 MPa.

2.2. Rehabilitation scheme

The proposed rehabilitation method aims to increase shear

resistance by ferrocement and to diverge part of the forces trans-
ferred into joint core by diagonal reinforcements. Fig. 3 shows
the proposed rehabilitation method. Concrete cover within the
joint region and the plastic hinge zones was removed and surface

Table 1
Compressive strength of concrete and mortar.

Specimen C1 S1 S2 S3
Concrete (MPa) 46.1 47.4 47.5 49.3
Mortar (MPa) N/A 34.0 34.7 71.6
Types of mortar for N/A Cement– Cementitious Epoxy-
ferrocement sand based
Fig. 1. Interior beam–column joint with smaller-size transverse beam.
 B. Li et al. / Engineering Structures 56 (2013) 897–909 899

Wire mesh

Wire mesh
Ferrocement

(a) Ferrocement wrapping

Diagonal reinforcements welded

to main reinforcements of beam

Diagonal reinforcements

(b) Diagonal reinforcements

Fig. 3. Rehabilitation scheme.

150 150 150 150

120 120

Column

Ferrocement
50 140 110

25

Detail (A) 50 140 110

Anchor
60 110 110 60

40
115 110 115

Beam
Concrete
110 140 50

Detail (A)

60 120 60 60 120 60

Fig. 4. Arrangement and details of anchors.

of concrete was roughened for improved bonding. Two layers of from the column edges and the beam edges to protect the critical
wire mesh properly folded were installed. 30 mm thick ferroce- regions and to enhance the overall performance of beam–column
ment was applied manually to replace the concrete cover. Ferroce- joints. It is worth noticing that cementitious mortar and epoxy-
ment was applied to 300 mm (least dimensions of the members) based mortar could be easily applied due to higher viscosity as
900 B. Li et al. / Engineering Structures 56 (2013) 897–909

Fig. 5. Test setup for beam–column joint.

compared with cement–sand mortar. Holes were drilled through
the joints. Four diagonal reinforcements (two in each direction)
were installed and were 10 mm, 12 mm and 12 mm in diameter
for specimens S1, S2 and S3 respectively. Ends of the diagonal rein-
forcements were welded to the main reinforcements of the beams
as shown in Fig. 3b.
As premature delamination of ferrocement was observed in
specimen S1, anchors were installed in specimens S2 and S3 to im-
prove bonding behavior and to facilitate fixing of wire mesh as
shown in Fig. 4. They were arranged in accordance with CECE-
242 [27]. The anchors had a total length of 65 mm with 40 mm
embedded into concrete as shown in Fig. 4.
2.3. Test setup and loading sequence
Fig. 5 shows the test setup for interior beam–column joints.
The specimens were tested by a displacement control multi-
purpose testing system with loading capacity of 10,000 kN in
vertical direction and 1500 kN in horizontal direction. The bot-
tom column was fixed to strong floor by a hinge. Ends of the
beams were supported by rollers to allow free horizontal move-
ment. Loading was applied at the end of the upper column by
two actuators acting in horizontal and vertical directions. The
vertical actuator was free to move horizontally in response to
the horizontal displacement.
Each specimen was tested under reversed horizontal displace-
ment under constant axial load. Firstly, axial load was applied at
Wong [29] on buildings in Hong Kong concluded that axial load
ratio of columns under ultimate load condition could be up to
0.629 based on loading schedules of columns retrieved from the ap-
proved drawings.
As shown in Fig. 6, quasi-static cyclic displacements were ap-
plied to the upper column. Proper loading sequence is needed to
capture member capacity and seismic demand. General comments
on the adopted loading sequence is referred to [9,13]. To reflect the
cumulative damage, inelastic deformations were repeated twice at
each displacement ductility factor. Detail of the loading sequence
is summarized below.
 The specimen was subjected to progressive increase in horizon-
tal load up to 75% of the nominal moment capacity of the
beams. Yield displacement Dy was first determined with due
consideration of tip displacement at the top column, and yield-
ing of the reinforcements in beams and columns.
 Afterwards, horizontal displacement was applied to the speci-
men. Each displacement cycle was repeated twice with dis-
placement ductility factor l (ratio of applied displacement to
yield displacement) advancing from 1, 2, 3, etc.
7 60
6 50
5
Horizontal displacement (mm)

the upper column at 0:6Ag fc0 and was kept constant throughout 40
4
the test, where Ag is the gross cross-sectional area of column and 30
3
fc0 is the cylinder strength of concrete. fc0 is assumed to be 0.75fcu, 20
Ductility factor

2
where fcu is the cubic strength of concrete. 1 10
In this study, high axial load is adopted. According to a local de- 0 0
sign code [28], ultimate loading capacity Nult of column can be cal- -1 -10
culated using Eq. (1). -2 -20
-3
Nutl ¼ 0:4f cu Ac þ 0:75f y Ast ð1Þ -30
-4
-40
-5
where Ac, Ast and fy are area of concrete section, area of longitudinal -50
-6
reinforcements respectively and yield strength of longitudinal rein- -60
-7
forcements. At different concrete strengths and reinforcement ra- 0 1 2 3 4 5 6 7 8 9 10 11 12 13
tios, axial load ratio ranges from 0.5 to 0.8 at ultimate limit state.
Cycle number
High axial load ratio is obtained with low concrete strength and
high reinforcement ratio. For instance, a recent survey by Su and Fig. 6. Displacement history.
 B. Li et al. / Engineering Structures 56 (2013) 897–909 901

(a) (b)

(c) (d)
Fig. 7. Crack patterns of specimens (a) C1, (b) S1, (c) S2 and (d) S3 after test.

3. Test results
3.1. General observations and failure modes
Crack patterns and failure modes of specimens are shown in
Figs. 7 and 8 respectively. Vertical deflections of columns for each
specimen at the end of each displacement cycle are shown in Fig. 9.
3.1.1. Specimen C1
Fig. 7a shows the crack pattern of control specimen C1. Before
reaching 0.8% drift, flexural cracks were observed in the beams
and two vertical cracks were observed in the joint. Occurrence of
vertical cracks in the joint is attributed to deformation of longitu-
dinal reinforcements with inadequate confinement under high ax-
ial load. At 1.0% drift, diagonal cracks appeared in the joint. After
formation of first diagonal crack in the joint, there was no further
propagation of cracks in beams. With progressive increase in drift,
diagonal cracks propagated and widened gradually. Finally, cracks
were connected together as shown in Fig. 8a. With opening and
closing of the diagonal cracks, spalling of concrete cover in the joint
was observed at the 9th cycle achieving a 1.5% drift. Throughout
the displacement cycles, crack was not observed in the columns.
Deterioration of joint strength (after the 8th cycle at 1.5% drift) re-
duced the axial loading capacity of the columns and caused failure
of beam–column joint. This suggests that confinement of
beam–column joint is critical at high axial load. Vertical deflection
of column in specimen C1 increased gradually during displacement
cycles as shown in Fig. 9. Failure of the control specimen is due to
joint shear failure accompanied by buckling of longitudinal rein-
forcements in the column.
3.1.2. Specimen S1
Fig. 7b shows the crack pattern of specimen S1 strengthened by
ferrocement using cement–sand mortar. At early stage of displace-
ment cycles, formation of cracks was not observed due to better
ability on crack control offered by ferrocement. Cracks first ap-
peared at the interface between the beams and the joint at 0.36%
drift. At 0.8% drift, diagonal cracks were formed in the joint with

Fig. 8. Failure modes of specimens (a) C1, (b) S1, (c) S2 and (d) S3.
902 B. Li et al. / Engineering Structures 56 (2013) 897–909

Drift ratio (%) ability of shifting severe cracking from the joint cores to the beams.
0.0 0.5 1.0 1.5 2.0 2.5 3.0 At 0.3% drift, cracks were first observed at the interface between
8.0 beams and the joint, and in the beams. At an interface between
Specimen C1 beam and the joint, a single crack opened and closed without cre-
7.5 Specimen S1
ating new cracks until the 7th cycle at 1.1% drift. Meanwhile, more
Vertical displacement (mm)

Specimen S2
7.0 cracks were observed at 1.1% drift at the interface between the col-
Specimen S3
umns and the beams. At 1.8% drift, first diagonal crack was ob-
6.5 served. At 2.5% drift, diagonal cracks were observed in other
directions. Diagonal cracks developed with increase in both crack
6.0
width and crack length in the following cyclic loading. Cracks in
5.5 the joint were ‘‘X’’-shape. When the diagonal cracks connected to-
gether throughout the joint, bulging of ferrocement occurred. As
5.0 shown in Fig. 9, vertical deflection of column increased rapidly
4.5
after 2.2% drift. Similar to specimen S2, joint shear failure triggered
pull-out of anchor bolts installed in the joint. In other strengthened
4.0 areas, anchor bolts worked well. Delamination of ferrocement pri-
0 10 20 30 40 50 60 70 mary occurred within the joint as shown in Fig. 8d. Failure of spec-
Horizontal displacement (mm) imen S3 was caused by combination of joint shear failure and
buckling of longitudinal reinforcements in the columns. All in all,
Fig. 9. Vertical deflections of columns for each specimen.
specimen S3 displayed the best crack control ability.

several inclined cracks in the upper column. With increase in drift, 3.2. Hysteretic behavior
more cracks were observed in the joint and crack width in the
upper column widened. Spalling of mortar was observed in the Fig. 10 shows relationships of horizontal load against displace-
upper column and the joint at 1.5% and 2.2% drift respectively. Fur- ment at the upper columns for each specimen. Summary of the test
ther damage was observed in the joint only. When approaching results is given in Table 2. Loading capacities of strengthened spec-
failure, ferrocement in the joint deformed laterally, i.e. bulging. It imens S1 and S3 were increased by 8% and 16.6% respectively as
ceased to offer confinement to the joint. As shown in Fig. 9, vertical compared with that of control specimen. This is primarily contrib-
deflection of column increased rapidly after 1.8% drift. Fig. 8b uted to enhancement in shear resistance and confinement of joint
shows condition of the joint after the test. Delamination of ferroce- resulting from ferrocement. However, specimen S2 shows no
ment and bulging of longitudinal reinforcements in the column improvement on loading capacity. Also, horizontal displacement
were observed. Failure of specimen S1 is due to joint shear failure corresponding to peak strength of specimen S2 is smaller than oth-
with buckling of longitudinal reinforcements in the column. ers. This is mainly due to premature failure of ferrocement possibly
caused by low strength mortar and inconsistent deformation be-
tween mortar and concrete. Vice versa, specimen S1, strengthened
3.1.3. Specimen S2 using ferrocement with low strength mortar, exhibited improved
As shown in Fig. 7c, the crack pattern of specimen S2, strength- loading capacity due to consistent deformation between mortar
ened by ferrocement using cementitious mortar, is similar to that and concrete. Comparing the hysteretic performance between
of specimen S1. After applying axial load, cracks appeared at ferro- specimens S2 and C1, diagonal reinforcements in the joint are inef-
cement in the joint and in the beams. This is attributed to low ten- fective if ferrocement cannot provide sufficient confinement to the
sile strength of mortar and inconsistent deformation between joint.
ferrocement and concrete. At less than 0.4% drift, cracks formed When high strength mortar was used for strengthening in spec-
along the longitudinal reinforcements in the joint. At 0.8% drift, imen S3, largest increase in loading capacity and deformation
diagonal cracks were observed. At the next displacement cycle, capacity is obtained. Improvement in loading capacity of specimen
cracks propagated rapidly in the joint. Subsequently, cracks were S3 is twice than that of specimen S1. Displacement corresponding
extended to the columns. In general, cracks were concentrated in to peak load of specimen S3 is about 1.5 times than that of control
the joint and were short and uniform. Ferrocement began to de- specimen. The proposed rehabilitation scheme can improve joint
form laterally, i.e. bulging, at the 11th cycle with 1.8% drift. At shear strength and deformation capacity when using high strength
the same time, there was rapid increase in vertical deflection of mortar.
column as shown in Fig. 9. Further into the displacement cycles,
bulging of ferrocement in the joint aggravated, which finally 3.3. Displacement ductility
caused spalling of mortar. After the test, delamination of ferroce-
ment was found in the joint. Anchor bolts were pulled out from Table 3 comprises displacement ductility of each specimen. The
concrete and remained intact with ferrocement. This suggests pos- estimations are based on envelops of hysteretic curves of each spec-
sible joint shear failure prior to delamination of ferrocement. imen as shown in Fig. 11. Displacement values of each specimen
Fig. 8c shows buckling of longitudinal reinforcements in the col- are averaged from positive and negative envelops. Displacement
umns. Similar to specimen S1, failure of specimen S2 is due to joint ductility factor l is defined as the ratio of ultimate displacement
shear failure with buckling of longitudinal reinforcements in the Du to yield displacement Dy. Ultimate displacement is defined as
column. displacement corresponding to 15% drop of loading capacity. Yield
displacement of specimens is assessed using two different meth-
3.1.4. Specimen S3 ods as follows.
For specimen S3, strengthened by ferrocement using epoxy- The first method is based on balance of energy as shown in
based mortar, relatively lesser number of cracks was observed in Fig. 12a [30]. A secant line passing the origin Point O and a Point
the strengthened area as shown in Fig. 7d. Vice versa, more cracks I on the curve intersects the peak strength at Point A. By adjusting
were formed at the un-strengthened area of the beams. This indi- the secant line so that area A1 equals to area A2, Point A is esti-
cates that the proposed ferrocement jacketing technique has the mated to represent the yield displacement.
 B. Li et al. / Engineering Structures 56 (2013) 897–909 903

Drift ratio (%) Drift ratio (%)

-3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0
120 120
100 100
80 80
60 60
Horizontal load (kN)

Horizontal load (kN)

40 40
20 20
0 0
-20 -20
-40 -40
-60 -60
-80 -80
Specimen C1 Specimen S1
-100 -100
-120 -120
-70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70
Horizontal displacement (mm) Horizontal displacement (mm)

Drift ratio (%) Drift ratio (%)

-3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 -3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0
120 120
100 100
80 80
60 60
Horizontal load (kN)

Horizontal load (kN)

40 40
20 20
0 0
-20 -20
-40 -40
-60 -60

-80 -80
Specimen S2 Specimen S3
-100 -100

-120 -120
-70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70
Horizontal displacement (mm) Horizontal displacement (mm)

Fig. 10. Column tip load–displacement relationships.

Table 2 Strengthened specimens show significant improvement in dis-

Summary of test results.
placement ductility. The increase in displacement ductility is 17%
Specimen Ultimate Corresponding Improvement in for specimen S1, 28% for specimen S2 and 34% for specimen S3.
load (kN) displacement (mm) loading capacity Ferrocement and diagonal reinforcements have preserved the joint
C1 Pull 84.3 25.40 0 from shear failure. Although premature failure of ferrocement oc-
Push 84.7 25.35 0 curred in specimen S2 with the lowest peak strength, there is
S1 Pull 92.4 23.20 +9.6%
improvement in ductility. Comparing averaged ultimate displace-
Push 90.1 17.40 +6.4%
S2 Pull 82.1 17.89 2.6% ment, performance of specimens S1 and S2 is similar to that of
Push 79.6 17.50 6.0% specimen C1. Using high strength mortar, specimen S3 shows the
S3 Pull 101.0 35.05 +19.8% best improvement on ultimate displacement at 27% and ductility
Push 96.0 34.85 +13.3% at 34%.

3.4. Energy dissipation

The second method is based on general yielding as shown in
Fig. 12b. A line for elastic behavior intersects the peak strength
at Point H. Vertical line passing Point H intersects the load–
displacement curve at Point I. Secant line passing Point O and Point
I intersects peak strength at Point A which becomes the yield
displacement.
As shown in Table 3, similar yield displacements are obtained
from two methods. Finally, displacement ductility is estimated
by averaging the values obtained from two methods.
Energy dissipation at each cycle is calculated from enclosed
area within load–displacement loop at this cycle. Cumulative en-
ergy dissipation is computed by summating energy dissipated in
previous cycles. Fig. 13 shows cumulative energy dissipation of
each specimen against horizontal displacement.
Energy dissipation capacity of beam–column joint is enhanced
by the proposed rehabilitation method. With the same demand
on energy dissipation, specimens S1, S2 and S3 produce smaller
904 B. Li et al. / Engineering Structures 56 (2013) 897–909

Table 3 Drift ratio (%)

Summary of displacement ductility. 0.0 0.5 1.0 1.5 2.0 2.5 3.0
50000

Cummulative energy dissipation (kNmm)

Spe. Du Ratio Balance of General Averaged Ratio Specimen C1
(mm) of Du energy yielding l of l 45000 Specimen S1
Dy l Dy l 40000 Specimen S2
(mm) (mm) Specimen S3
35000
C1 39.25 1.00 15.01 2.61 15.32 2.56 2.58 1.00
S1 40.29 1.03 13.24 3.06 13.42 3.01 3.03 1.17 30000
S2 39.10 1.00 11.77 3.34 12.00 3.29 3.31 1.28
25000
S3 49.94 1.27 14.90 3.35 14.06 3.55 3.45 1.34
20000
15000

Drift ratio (%) 10000

-3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 5000
120
100 0
0 10 20 30 40 50 60 70
80
Horizontal displacement (mm)
60
Horizontal load (kN)

40 Fig. 13. Cumulative energy dissipation for each specimen.

20
0
secant stiffness is defined as the slope of a line passing through
-20
peak loads at both directions. It represents the ability to resist
-40 deformation. Plots of stiffness degradation against horizontal dis-
Specimen C1
-60 placement are shown in Fig. 15.
Specimen S1
-80 Specimen S2 Strengthened specimens exhibit larger stiffness at the initial
-100 Specimen S3 stage of loading. Due to premature failure of ferrocement, stiffness
-120 of specimen S2 decreases rapidly as compared with other strength-
-70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70 ened specimens, especially after passing 0.8% drift. Stiffness of
Horizontal displacement (mm) specimen S1 is higher than that of control specimen but is lower
than that of specimen S3 at the same drift. Specimen S3 strength-
Fig. 11. Envelops of hysteretic loops for each specimen. ened using high strength mortar exhibits higher stiffness and
slower reduction in stiffness. In general, stiffness of strengthened
specimens increases with increasing strength of mortars. Loss of
horizontal displacement as compared with control specimen.
stiffness can be attributed to deterioration of concrete and ferroce-
Strengthened specimens are able to sustain larger energy dissipa-
ment in the joint.
tion capacity when the horizontal displacement increases. Speci-
men S3 strengthened by epoxy-based mortar shows the best
ability to dissipate energy. Comparing energy dissipation of speci- 3.6. Effectiveness of diagonal reinforcements
mens S1 or S2 with specimen S3, capacity of energy dissipation in-
creases with increasing strength of mortar. Fig. 16 shows strains of diagonal reinforcements against the
loading cycles. Here, strains induced from the axial load are ig-
3.5. Stiffness degradation nored. Similar strain increments are observed on the diagonal rein-
forcements of different strengthened specimens. With increasing
Stiffness degradation is assessed using secant stiffness deter- horizontal displacement, diagonal reinforcements of specimen S1
mined from each complete hysteresis loop. As shown in Fig. 14, have the largest strain. This is due to the use of smaller diameter

(a) (b)
Fig. 12. Definitions of yield displacement using (a) energy balance method, and (b) general yielding method.
 B. Li et al. / Engineering Structures 56 (2013) 897–909 905

Drift ratio (%)

0.0 0.5 1.0 1.5 2.0 2.5 3.0
6000
Specimen S1
Specimen S2
5000
Specimen S3

4000

Strain (micron)
3000

2000

1000

0
0 10 20 30 40 50 60 70

Fig. 14. Definition of stiffness.

Horizontal displacement (mm)

Fig. 16. Strains of diagonal reinforcements in tension versus horizontal

displacement.
Drift ratio (%)
0.0 0.5 1.0 1.5 2.0 2.5 3.0
10 ultimate displacement and dissipated energy using the following
Specimen C1 equation.
9
Specimen S1
Z
8 Specimen S2 dM b
Specimen S3 D¼ þ dE ð2Þ
7 du Q y du
Stiffness (kN/mm)

6
where dM is the maximum deflection attained during seismic load-
5 ing; du is the ultimate deflection capacity under monotonic load; b
4 is a model parameter that depends on shear force, axial force,
amount of longitudinal and confinement reinforcements; Qy is the
3
calculated yield strength; and dE is the incremental dissipated
2 energy.
1 Values of dM, Qy and dE are obtained from the experimental re-
sults. For beam–column joints, du comprises contributions
0
0 10 20 30 40 50 60 70 from deformations of beams, columns and joint. Deformations of
beams and columns are calculated based on the empirical formula
Horizontal displacement (mm)
for the evaluation of ultimate drift ratio according to CEN Eurocode
Fig. 15. Stiffness degradation versus horizontal displacement. 8 [32].
 0:225
1  maxð0:01; x0 Þ 0:35
reinforcements. As the tensile force in the diagonal reinforcements du ¼ 0:016  ð0:3P Þ fc ðL Þ
cel maxð0:01; xÞ
is similar to that of the other specimens, increasing the diameter of  
f
diagonal reinforcements may not affect magnitude of the forces aqs fytc
 25 ð1:25Þ100qd ð3Þ
transferred through the joint.
Diagonal reinforcements reduce strains of longitudinal rein- where cel is equal to 1.0 in this study; P is the axial load index and
forcements of beam in the joints. Fig. 17 shows strains of longitu- 0
is equal to P=bhf c ; b and h are width and height of the cross-section
dinal reinforcements in beams near the joint against horizontal 0
respectively; x and x are the ratio of tension and compression
displacement. The data are obtained from the first cycle at each reinforcements respectively; fc is the concrete strength; L is the
drift and are based on measurements obtained from strain gauges shear span index and is equal to M/Vh; qs is the transverse rein-
at 25 mm away from column face in tensile reinforcements. Rein- forcement ratio and is equal to Asx/bsh; sh is the spacing of stirrups;
forcement strains were reduced in the strengthened specimens as fyt is the yield strength of transverse reinforcement; qd is the diag-
compared with that in control specimen. This phenomenon is more onal reinforcement ratio; and a is the confinement effectiveness
obvious in the advanced loading stages. Participation of diagonal factor derived from the following equation.
reinforcements to joint shear resistance increases with reducing
contribution from ferrocement. Strains of bottom reinforcements    P 2!
sh sh bi
in beams are larger than that of top reinforcements in beams for a¼ 1 1 1 ð4Þ
2b0 2h0 6b0 h0
all specimens. Axial compression failure in the joint induced addi-
tional tension force in the bottom reinforcements of beams.
where b0 and h0 are the dimensions of confined concrete core to the
centerline of the stirrup; and bi is the centerline spacing of longitu-
3.7. Damage indices dinal reinforcements (index by i) laterally restrained by a stirrup
corner or a cross-tie along the perimeter of the cross-section. In re-
Damage achieved by each specimen is evaluated using Park and spect of joint shear deformation capacity, it is determined based on
Ang’s damage model [31] as applied to reinforced concrete beam– experimental results. The model parameter b is taken to be 0.15 for
column joints [9]. It is expressed as a linear combination of rehabilitated specimens as recommended in [33] and 0.25 for the
906 B. Li et al. / Engineering Structures 56 (2013) 897–909

3000 3000

2500 2500

2000 2000

Strain (micron)
Strain (micron)

1500 1500

1000 1000
Specimen C1 Specimen C1
500 Specimen S1 500 Specimen S1
Specimen S2 Specimen S2
Specimen S3 Specimen S3
0 0
0 10 20 30 40 50 0 10 20 30 40 50
Horizontal displacement (mm) Horizontal displacement (mm)

3000 3000

2500 2500

2000 2000
Strain (micron)
Strain (micron)

1500 1500

1000 1000
Specimen C1 Specimen C1
500 Specimen S1 500 Specimen S1
Specimen S2 Specimen S2
Specimen S3 Specimen S3
0 0
0 10 20 30 40 50 0 10 20 30 40 50
Horizontal displacement (mm) Horizontal displacement (mm)

Fig. 17. Strains of longitudinal reinforcements of beam at interface of beam to column.

control specimen, i.e. beam–column joints without seismic details
[34].
The damage indices for specimens are illustrated in Fig. 18.
Rehabilitated specimens show lower damage indices at each drift
ratio as compared with control specimen. This indicates that the
proposed rehabilitation method is effective for upgrading rein-
forced concrete beam–column joints.
joint with confinements even without transverse reinforcement
in the joint. In the above, joint shear strength of
strengthened specimens includes contributions from the diagonal
reinforcements.
4.2. Principal tensile stress

Nominal principal tensile stress rt in a beam–column joint is

4. Joint shear strength
calculated using Eq. (5).
4.1. Joint shear stress sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
rp r2p
rt ¼ þ þ v 2jh ð5Þ
Joint shear stress is improved through enhancing shear 2 4
resistance and reducing the forces transferred into the joint core.
Horizontal joint shear force is calculated from equilibrium of a where rp is the axial compression stress in the joint (i.e. axial
free body diagram at mid-height of the joint plane [35]. Taking force divided by area of cross-section); and vjh is the joint shear
into account the P–D effect as shown in Fig. 5, beam shear force stress.
becomes Vb = (Vclc + PD)/lb. Horizontal joint shear stresses are esti- Fig. 19 shows nominal principal tensile stresses against hori-
mated from horizontal joint shear force divided by the prespective
ffiffiffiffi zontal displacement for each specimen. Maximum nominal princi-
effective joint area and are 1.03, 1.11, 1.01 and 1:26 fc0 MPa for pal tensile stresses of specimen S1 and S3 are higher than that of
specimens C1, S1, S2 and S3 respectively. All specimens possess specimen C1. Specimen S2 exhibits lower nominal principal tensile
inherent
pffiffiffiffi stresses larger than the allowable joint shear stress at stress. This agrees with experimental observations that specimen
0:83 fc0 MPa in ASCE/SEI 41 [36]. The proposed rehabilitation S2 failed prematurely. In fact, nominal
pffiffiffiffi principal tensile stress of
method has enhanced joint shear stresses of specimens S1 and specimen S2 was less than 0:29 fc0 MPa (i.e. joint cracking
S3. According to ACI-ASCE 352R [37], joint shear stresspffiffiffiffi of speci- strength). This also indicates that the proposed rehabilitation
men S3 reaches the nominal joint strength ð1:25 fc0 MPaÞ for method is effective with high strength mortar.
 B. Li et al. / Engineering Structures 56 (2013) 897–909 907

Drift ratio (%) Shear strength of concrete in the joint can be calculated using
0.0 0.5 1.0 1.5 2.0 2.5 3.0 Eq. (7), as suggested by ACI-ASCE 352R [37].
1.4 qffiffiffiffi
Specimen C1 V ec ¼ c fc0 bj hc ð7Þ
1.2 Specimen S1
Specimen S2 where c is joint shear strength factor, bj is the effective joint width
1.0 Specimen S3 determined according to ACI-ASCE 352R [37] and hc is the depth of
Damage indices

column in the direction of joint shear. Here, shear strength factors c

0.8 is equal to 0.83 [36], i.e. no transverse reinforcement being provided
in the joint. Using a value of 0.83 is too conservative. As shown in
0.6 Table 4, c = 1.0 produces better agreement with experimental data.
Contribution of ferrocement to joint shear strength is evaluated
0.4 by a semi-empirical equation proposed by [38] as shown in Eq. (8).
It takes into account the influence of wire mesh on shear strength.
0.2
It is noticed that ferrocement failed completely when strengthened
specimens reached their peak strengths.
0.0
 pffiffiffiffiffiqffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 
0 10 20 30 40 50 60 70 bf hc 2
Horizontal displacement (mm) V f ¼ 0:935 2
0:249 fm ða=hc Þ þ 1 þ qm fym ð8Þ
ða=hc Þ þ 1
Fig. 18. Comparison of damage indices for each specimen.
where a and hc are shear span and depth of ferrocement respec-
tively; bf is the thickness of ferrocement; fm is the cubic mortar
strength; qm is the longitudinal wire mesh ratio in ferrocement;
Drift ratio (%)
and fym is the yield strength of wire mesh.
-3.0 -2.5 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 2.5 3.0 As shown in Fig. 16, shear force balanced by diagonal reinforce-
3.0
ments in a joint is mainly related to the drift and regardless of the
diameter of diagonal reinforcements. Ratio of tensile force in diag-
2.5
Principle tensile stress (MPa)

onal reinforcements to tensile force in longitudinal reinforcements

of beams varies linearly with drift as shown in Fig. 20. When pre-
2.0 dicting contribution of diagonal reinforcements to joint shear
0.29 fc'
force, drift at peak strength is applied. An empirical formula based
1.5 on measured strains of diagonal reinforcements is proposed in Eq.
(9).
1.0
Specimen C1
Specimen S1
0.5
Specimen S2
40
Specimen S3 Linear fit
0.0 Measured
-70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70
Horizontal displacement (mm) 30

Fig. 19. Comparison of nominal principal tensile stresses in joint cores.

Tdiag/T (%)

20
4.3. Predicted joint shear strength

Shear strength of a strengthened beam–column joint can be

predicted by summating shear forces taking by concrete, ferroce- 10
ment and diagonal reinforcements in the form of

V n ¼ V ec þ V f þ V diag ð6Þ 0
0.0 0.5 1.0 1.5 2.0

where Vec is shear force taking by concrete in the joint; and Vf and Drift ratio (%)
Vdiag are shear forces taking by ferrocement and diagonal reinforce- Fig. 20. Relationship between force ratio of diagonal reinforcements (Tdiag) to beam
ments respectively. reinforcements (T) and drift ratio.

Table 4
Comparison of measured and predicted joint shear forces for specimens.

Spe. (1) Vjh.exp (kN) Predicted shear force taking by each component (6) Vjh2 (7) Vjh2 Vjh.exp/Vjh1 Vjh.exp/Vjh2
(2)a Conc. (3)b Conc. (4) Ferro. (5) Diag. reinf. (2) + (4) + (5) (kN) (3) + (4) + (5) (kN) (1)/(6) (1)/(7)
C1 554.2 439.2 529.2 0.0 0.0 439.2 529.2 1.26 1.05
S1 583.0 356.3 429.3 78.3 65.1 499.7 572.7 1.17 1.02
S2 543.0 356.7 429.7 78.7 65.1 500.5 573.5 1.08 0.95
S3 689.0 363.4 437.8 94.4 109.4 567.2 641.6 1.21 1.07
Mean 1.16 1.01
a
c = 0.83 in Eq. (7).
b
c = 1.0 in Eq. (7).
908 B. Li et al. / Engineering Structures 56 (2013) 897–909
V diag ¼ T diag cos hd ¼ kT cos hd
where Tdiag is the tensile force of diagonal reinforcements; T is the
tensile force of longitudinal reinforcements; hd is the inclination
of diagonal reinforcements to horizontal direction; and k is propor-
tional factor which is related to story drift. Based on measured
strains of longitudinal reinforcements and diagonal reinforcements,
k is equal to 0.35 for specimen strengthened with high strength
mortar (e.g. specimen S3) and 0.15 for specimen strengthened with
normal strength mortar (e.g. specimens S1 and S2).
Shear forces of joints rehabilitated by ferrocement jackets
embedded with diagonal reinforcements can be predicted by
superposition using Eq. (6)–(9). Comparison of predicted and mea-
sured joint shear forces is summarized in Table 4. Predicted shear
forces of strengthened specimens agree well with test results. With
due consideration on structural safety, c is recommended to be
0.83 for evaluating the contribution of concrete to joint shear resis-
tance. Further studies are necessary to assess the shear forces of
strengthened joints under low axial load.
5. Conclusions
An experimental program was conducted to evaluate the effec-
tiveness of the proposed rehabilitation method for reinforced con-
crete interior beam–column joints. The proposed rehabilitation
method is based on increasing joint shear resistance and reducing
forces transferred into joint core. Based on the observation and test
results, the following conclusions can be drawn.
(1) Beam–column joints under high axial load ð0:6Ag fc0 Þ are
prone to failure at the joints with joint shear stress larger
than the maximum allowable stress as specified in ASCE/SEI
41 and buckling of longitudinal reinforcements.
(2) The proposed rehabilitation method using ferrocement jack-
ets with embedded diagonal reinforcements enhances seis-
mic performance of interior beam–column joints using
ferrocement with high strength mortar. Damage levels of
rehabilitated specimens are reduced as compared with that
of control specimen.
(3) Bonding between ferrocement and concrete can be
improved by installing anchor bolts at the interface. Anchor
bolts are effective to prevent delamination of ferrocement
and facilitate fixing of wire mesh for easy construction.
(4) Increasing the strength of mortar in ferrocement is a vital
factor for achieving higher strength, larger ductility, better
energy dissipation, and higher stiffness.
(5) Diagonal reinforcements installed in the joint reduce strains
in the longitudinal reinforcements. This has the benefit of
reducing joint shear stress and preventing bonding failure
of longitudinal reinforcements.
(6) A method to predict shear capacities of joints strengthened
by ferrocement jackets with embedded diagonal reinforce-
ments is proposed. The proposed method is limited to spec-
imens under high axial load.
As axial load ratio is the factor affecting the performance of
beam–column joints and effectiveness of the rehabilitation meth-
ods, further tests will be conducted on joints with low axial load.
Acknowledgements
The authors are grateful to the financial support from The Hong
Kong Polytechnic University and the Research Grants Council of
Hong Kong (RGC No: PolyU 5206/08E). The authors would like to
thank for the technical assistance provided by Structural
ð9Þ Engineering Research Laboratory and Concrete Technology Labora-
tory of The Hong Kong Polytechnic University.
References
[1] Lam SSE, Xu YL, Chau KT, Wong YL, Ko JM. Progress in earthquake resistant
design of buildings in Hong Kong. In: Structural engineering world congress,
October 9–12, Yokohama, Japan; 2002.
[2] Li B, Kai Q. Seismic behavior of reinforced concrete interior beam-wide column
joints repaired using FRP. J Compos Constr 2011;15:327–38.
[3] Dogangun A. Performance of reinforced concrete buildings during the May 1,
2003 Bingol earthquake in Turkey. Eng Struct 2004;26:841–56.
[4] Zhao B, Taucer F, Rossetto T. Field investigation on the performance of building
structures during the 12 May 2008 Wenchuan earthquake in China. Eng Struct
2009;31:1707–23.
[5] Hakuto S, Park R, Tanaka H. Seismic load tests on interior and exterior beam–
column joints with substandard reinforcing details. ACI Struct J
2000;97:11–25.
[6] Choi H, Kim J. Progressive collapse-resisting capacity of RC beam–column sub-
assemblage. Mag Concr Res 2011;63:297–310.
[7] Wang YC, Hsu K. Shear strength of RC jacketed interior beam–column joints
without horizontal shear reinforcement. ACI Struct J 2009;106:222–32.
[8] Tsonos ADG. Performance enhancement of R/C building columns and beam–
column joints through shotcrete jacketing. Eng Struct 2010;32:726–40.
[9] Karayannis CG, Chalioris CE, Sirkelis GM. Local retrofit of exterior RC beam–
column joints using thin RC jackets – an experimental study. Earthq Eng Struct
D 2008;37:727–46.
[10] Ghobarah A, Aziz TS, Biddah A. Rehabilitation of reinforced concrete frame
connections using corrugated steel jacketing. ACI Struct J 1997;94:283–94.
[11] El-Amoury T, Ghobarah A. Seismic rehabilitation of beam–column joint using
GFRP sheets. Eng Struct 2002;24:1397–407.
[12] Ghobarah A, Said A. Shear strengthening of beam–column joints. Eng Struct
2002;24:881–8.
[13] Karayannis CG, Sirkelis GM. Strengthening and rehabilitation of RC beam–
column joints using carbon-FRP jacketing and epoxy resin injection. Earthq
Eng Struct D 2008;37:769–90.
[14] Pantelides CP, Okahashi Y, Reaveley LD. Seismic rehabilitation of reinforced
concrete frame interior beam–column joints with FRP composites. J Compos
Constr 2008;12:435–45.
[15] Tsonos AG. Effectiveness of CFRP-jackets and RC-jackets in post-earthquake
and pre-earthquake retrofitting of beam–column subassemblages. Eng Struct
2008;30:777–93.
[16] Lee K, Le-Trung K, Lee J, Lee DH, Woo S. Experimental study of RC beam–
column joints strengthened using CFRP composites. Compos Part B – Eng
2010;41:76–85.
[17] Niroomandi A, Maheri A, Maheri MR, Mahini SS. Seismic performance of
ordinary RC frames retrofitted at joints by FRP sheets. Eng Struct
2010;32:2326–36.
[18] Sasmal S, Novak B, Rmanjaneyulu K, Roehm C, Srinivas V, Lakshmanan N, et al.
Upgradation of gravity load designed sub-assemblages subjected to seismic
type loading. Compos Struct 2011;93:1561–73.
[19] Ahmed A, Kodur V. The experimental behavior of FRP-strengthened RC beams
subjected to design fire exposure. Eng Struct 2011;33:2201–11.
[20] ACI 549R-97. Report on ferrocement. Detroit: American Concrete Institute;
1997. p. 68.
[21] Naaman AE. Ferrocement and laminated cementitious composites. Ann Arbor
(MI): Techno Press 3000; 2000. p. 372.
[22] Ho IFY, Lam ESS, Wu B, Wang YY. Monotonic behavior of reinforced concrete
columns confined with high-performance ferrocement. J Struct Eng – ASCE
2013;139:574–83.
[23] Abdullah, Takiguchi K. An investigation into the behavior and strength of
reinforced concrete columns strengthened with ferrocement jackets. Cem
Concr Compos 2003;25:233–42.
[24] Paramasivam P, Lim CTE, Ong KCG. Strengthening of RC beams with
ferrocement laminates. Cem Concr Compos 1998;20:53–65.
[25] Tsonos AG, Tegos IA, Penelis GG. Seismic resistance of type-2 exterior beam–
column joints reinforced with inclined bars. ACI Struct J 1992;89:3–12.
[26] Chalioris CE, Favvata MJ, Karayannis CG. Reinforced concrete beam–column
joints with crossed inclined bars under cyclic deformations. Earthq Eng Struct
D 2008;37:881–97.
[27] CECS-242. Technical specification for strengthening concrete structures with
grid rebar and mortar. China Association for Engineering Construction
Standardization. China Planning Press; 2008.
[28] CopConcrete-2004. Code of practice for structural use of concrete. Buildings
Department; 2004. p. 180.
[29] Su RKL, Wong SM. A survey on axial load ratios of structural walls in medium-
rise residential buildings in Hong Kong. HKIE Trans 2007;14:40–6.
[30] Lam SSE, Wu B, Wong YL, Wang ZY, Liu ZQ, Li CS. Drift capacity of rectangular
reinforced concrete columns with low lateral confinement and high-axial load.
J Struct Eng – ASCE 2003;129:733–42.
[31] Park YJ, Ang AHS. Mechanistic seismic damage model for reinforced concrete. J
Struct Eng – ASCE 1985;111:722–39.
[32] CEN Eurocode 8: Design of structures for earthquake resistance – part 3:
assessment and retrofitting of buildings, Brussels; 2004.
 B. Li et al. / Engineering Structures 56 (2013) 897–909
[33] Cosenza E, Manfredi G, Ramasco R. The use of damage functionals in
earthquake engineering – a comparison between different methods. Earthq
Eng Struct D 1993;22:855–68.
[34] Altoontash A. Simulation and damage models for performance assessment of
reinforced concrete beam–column joints. PhD thesis. Stanford University;
2004.
[35] Kim J, LaFave JM. Key influence parameters for the joint shear behaviour of
reinforced concrete (RC) beam–column connections. Eng Struct
2007;29:2523–39.
909
[36] ASCE/SEI 41. Seismic rehabilitation of existing buildings. American Society of
Civil Engineers, Reston, VA; 2007.
[37] ACI-ASCE 352R-02. Recommendations for design of beam–column
connections in monolithic reinforced concrete structures. American Concrete
Institute, Farmington Hills, MI; 2002.
[38] Desayi P, Nandakumar N. A semi-empirical approach to predict shear strength
of ferrocement. Cem Concr Compos 1995;17:207–18.