Available online at www.sciencedirect.

com

Procedia Engineering 14 (2011) 1871–1878

The Twelfth East Asia-Pacific Conference on Structural Engineering and Construction

Experimental Study on the Seismic Behaviour of Strengthened

Concrete Column-Beam Joints by Simulated Earthquake
LU Zhoudao a, SU Lei b*, YU Jiangtao b
a
State Key Laboratory of Disaster Reduction in Civil Engineering, Tongji University, Shanghai 200092, China
b
Research Institute of Structural Engineering and Disaster Reduction, Tongji University, Shanghai 200092, China

Abstract

A total of five repaired column-beam joints (C-BJs) strengthened by basalt fiber reinforced polymer (BFRP) had
been pre-damaged under cyclic loads in order to study on the seismic behavior. The experimental study including
pre-damage, rehabilitation and re-test under cyclic loads concentrated on the effect on different levels of pre-damage
and epoxy injection. Through the experimental data, ultimate load-carrying capacity, ultimate displacement, ductility,
hysteresis curves and dissipative ability, the following conclusion was drawn: it is indispensable for the load-carrying
capacity of frame beams to contribute to slab. The repaired C-BJs can reach or even exceed the level of their ultimate
load-carrying capacity and original seismic performance before pre-damage. The load-carrying capacity and bond-
slip behaviors can be improved by the BFRP in the repaired C-BJs. Concrete columns can be effectively confined to
the expansion and deformation. The failure mode of strong beam and weak column was changed into strong column
and weak beam after rehabilitations. The study indicated that the method of BFRP-reinforced pre-damaged concrete
column-beam joints was reasonable and effective.

© 2011 Published by Elsevier Ltd. Open access under CC BY-NC-ND license.

Keywords: seismic damage; seismic performance; 3D column-beam joint; BFRP; rehabilitation of damage

1. Introduction

A large number of reinforced concrete frame structure building earthquake damaged in the Wenchuan
earthquake, where the failure mechanism of "strong column and weak beam" seldom appeared. The
deformation of the frame structure is closely related to its failure mechanism in the earthquake. The

* Corresponding author
E-mail address: sulei0518@sina.com

1877–7058 © 2011 Published by Elsevier Ltd. Open access under CC BY-NC-ND license.
doi:10.1016/j.proeng.2011.07.235
1872 LU Zhoudao et al. / Procedia Engineering 14 (2011) 1871–1878

column-beam joints are the weakest and most vulnerable parts of the framework. Investigations were
carried out to study the effect of yield mechanisms. The steel in the framework would actually increase
the flexural capacity of 20 to 30%. Inclined core area of cross crack was often appeared in the low-cycle
test of two-dimensional joints. If the orthogonal beam constraints were considered on the role of the joints,
the damage of the three-dimensional beam, plate and column joints may be different from the two-
dimensional joints.
To improve the seismic performance of concrete structures, many researchers have been looking for
different reinforcement materials, which have been recognized at different times such as paste steel, fiber
composite (FRP). FRP has some unique advantages because of its lightweight, high strength, corrosion
resistance and easy construction. The mechanical properties of basalt fiber is a little worse, but much
cheaper than carbon fiber. However, the reinforced effect of basalt fiber after earthquake has not been
studied thoroughly.
Survey of existing constructions reveals that this experiment is necessary. On the basis of the
traditional study, it takes more realistic three-dimensional frame joints to simulate pre-damage,
rehabilitation and re-test under cyclic loads and test ultimate load-carrying capacity, ultimate
displacement, ductility, hysteresis curves. The seismic performance of the damaged joints obtained
experimentally is discussed along with failure types.

2. Experiment work

2.1. Specimen details

Five same types of C-BJs with BFRP have been cast together for experimental verification named J0-
J4 compared with the joints with CFRP test by our team in a previous period [18]. The same size and
concrete grade as reference [15] were used in this experiment (Fig.1). The ultimate strength of BFRP was
3200MPa and the modulus was 90GPa. In addition, the compression strength of concrete wais 24MPa
and the modulus was 26000MPa.

(a) (b)

Fig.1˖Details of the C-BJ in the major direction

 LU Zhoudao et al. / Procedia Engineering 14 (2011) 1871–1878 1873

2.2 Rehabilitated specimens

The reinforced methods and volume were completely identical in order to explore the factors of the
damaged degree and pouring depth (Tab.1). J1 and J2 carried out epoxy injection. The details of
reinforced methods were as follows: (1) BFRP sheets are applied in L shape to upgrade the joints with
two layers. One side was along with the steel bar in the column and the other side was within the 300 mm
or 500mm in the frame beams. In this step the BFRP has been applied on the top and bottom concrete
surfaces. (2) BFRP wraps were provided over the inner layers, which was perpendicular to the axis of the
members. The anchorage length was 200mm.

Table 1: The loading regime and strengthened scheme of the C-BJ

Pre-damaged Displacement of 0.85Pmax

name Axial compression ratio Reinforced method
displacement (mm) (mm)
J0 0.5 üü üü 90
J1 0.5 33 BFRP 130
J2 0.5 26 BFRP 134
J3 0.5 26 BFRP 134
J4 0.5 üü BFRP 120

2.3 Test program

The column was fixed at its ends on a loading frame subjected to the axial compression ratio of 0.5
and a constant axial load of 200kN which simulated the axial force of the lower column as far as possible.
Horizontal loads were imposed by force and displacement method. Each load step had two cycles back
and forth controlling by force before yielding of steel bars. While it had three times per cycle in amplitude
level controlled by the displacement after yielding and then increased the load point displacement into the
next load.
As the Reference block, J0 was directly loaded to the destruction without reinforcement. Cyclic tests
from J1 to J3 was carried out to simulate the formation of earthquake and reinforced with basalt fiber
directly to the destruction of test. J4 reinforced with basalt fiber was directly loaded to the destruction
without cyclic tests. The experimental setup truly reflected the deformation and P-ǻ effect of the structure.
Electro-hydraulic servo actuator fixed on the back strength wall applied the low-cycle reversed horizontal
load on the bottom of test fixture. Taking 0.85Pmax as the terminated test load, the experimental data were
collected by the computerized data acquisition system where the horizontal load-displacement curve has
been drawn and the yield of steel bar and pre-damaged displacement could be determined.

3. Results and discussion

3.1 Test phenomena

J0 was directly loaded to the destruction without reinforcement. The failure process was discussed here.
When the displacement of column cap reached 24mm, it cracked at the junction of beam and column;
when it reached 36mm, penetrating cracks appeared at the edge of the beam. Concrete, at the junction of
plate and column, became loose and peeling and transverse cracks, at the bottom of beam near the column,
appeared; when it reached 48mm, concrete was continuously spalling at the bottom of column.
Transverse cracks appeared at all the four corners of the plate along with the beams at the same time;
when it reached 60mm, concrete stripped and exposed aggregate at the junction of beam bottom and
1874 LU Zhoudao et al. / Procedia Engineering 14 (2011) 1871–1878

column; when it reached 72mm, concrete stripped at the core area of the column and the cracks appeared
at the bottom of the beam; when it reached 84mm, concrete broke off and steel bar was exposed from the
column; when it reached 90mm, the capacity of the joints decreased significantly and the test ended. The
failure mode was flexural failure of column tip and shear failure of the core area simultaneously.
Specimens from J1 to J4 were reinforced with basalt fiber directly to the destruction of test. The test
phenomena were similar. Taking J1 for example, the failure process was discussed as follows. When the
horizontal load reached 30kN, it cracked and sounded brittle at the weak parts of colloid. Then it was
controlled by the displacement. When the displacement of column cap reached 24mm, the sheet sounded
brittle in the core area and the cracks appeared at the bottom of plate along with the beams; when it
reached 36mm, the sheet sounded brittle densely and violently in the core area and cracks cut through the
plate; when it reached 60mm, the cracks appeared and parts of sheets tore at the junction of beam bottom
and column; when it reached 96mm, parts of the sheets plumped up and divorced from the concrete at the
top of beams and parts of the sheets broke at the bottom of beam and on the side of column; when it
reached 108mm, it sounded severe and torsion cracks appeared on the side of beam; when it reached
120mm, the concrete crisped and broke off severely in the core area and the test ended. The concrete at
the top of column destructed and the beam got away from the joint after test. The failure mode was
flexural failure of beam tip. (Fig 2)When the displacement reached 36mm, the plate cracked. The number
of cracks increased as the displacement increased. The facts above showed that the contribution of the
plate to the capacity of structure could not be ignored. Joint constrained by plate in tension increased
when steel bar was fully utilized and came into the hardening stage.

3.2 Capacity analysis

Ultimate capacity and displacement of each specimen on the column are summarized in Table 2.
Compared J0 with J4, capacity of the joint reinforced by BFRP increases limitedly, but the displacement
increases significantly. Compared J1 with J4, the pre-damaged displacement has little effect on the
ultimate capacity of concrete structure but has great effect on ultimate displacement. Compared J2 with J3,
epoxy injection has little effect on both the ultimate capacity and displacement. The fact that the ductility
coefficient without reinforcement is less than that with reinforcement shows BFRP can increase the
seismic performance abundantly.

Table 2: Ultimate capacity and displacement of each specimen on the column

Specimens name J0 J1 J2 J3 J4
capacity˄kN˅ 71.4 75.8 76.3 72.6 79.6
Increased level to J-0 — 6.2% 6.9% 1.7% 11.5%
Displacement of 0.85Pmax˄mm˅ 90 130 134 134 120
Increased level to J-0 — 66.5% 71.6% 71.6% 53.6%
ductility coefficient of displacement 2.60 4.06 4.62 4.45 3.64

3.3 Ductile analysis

Hysteresis and skeleton curves of the specimens is showed in Fig.3. In this case, good elastic is
maintained and energy dissipation capacity is small before the displacement reaches 8mm. When it
reaches 24mm, the pinching phenomena of hysteresis curves emerged. The phenomena are more and
more obvious as the displacement increases. It indicates that the stiffness of the components falls down
because of the opening cracks of tensile in the early period of reverse loading after yielding of the steel
 LU Zhoudao et al. / Procedia Engineering 14 (2011) 1871–1878 1875

bar in the core area where the bond deterioration becomes more obvious. After the cracks close in the
beam tip, the concrete of compression zone comes to work and the stiffness is increased. The number of
cycles increases and the stiffness falls down under the same step. The skeleton curve does not have
obvious yielding step. The failure of the joint is both the bending failure of column tip and shearing
failure of core area, which is consistent with the experimental conclusion.
Good elastic is maintained in the beginning and then the pinching appears more and more obvious as
the displacement increases. The inverted S shaped of hysteresis curves dumps gradually to the
displacement axis and the stiffness is smaller and smaller. When the four specimens load to the ultimate
load of J0, conspicuous yield plateau appears and ultimate displacement of column tip is significantly
increased, which indicates good ductile performance of the joint reinforced by BFRP. The beam gets
away from the joint and meanwhile the concrete of the column is loss of capacity, which shows the failure
is the mode of strong beam and weak column.
Spindle with better energy quickly transit to the worse with the inverted S shaped and the phenomena
of pinching is serious, which is the same characteristic of hysteresis curves. In this case the steel bar
sliding, shearing deformation and concrete cracking should be special attention. The phenomena of
pinching and no slip with reinforcement achieve better improvement than that without improvement.
Energy dissipation and seismic performance are improved significantly after reinforcement. Compare
with J0, the load of J4 increases rapidly. The capacity achieves further improvement when the steel bar
slides and comes to the hardening stage. Skeleton curve slope among J1 to J3 becoming larger indicates
that the BFRP sheets involve to increased stiffness in the period of loading.
To summarize, the seismic performance of the joints damaged by cyclic loads and then reinforced by
BFRP can be improved significantly. The reinforcement method has great efficient on controlling the
bending damage of column tip, shearing failure of the core area and anchor damage of joint. The BFRP
sheets is equivalent to provide additional transverse reinforcement to increase shearing capacity of the
core area, limit the concrete spalling and longitudinal reinforcement buckling, and prevent the joint to
damage and slide so that the core area can come to the ideal ductile failure on the beam tip.
load˄kN)

80

60

40

20

0
-100 -50 0 50 100
-20 displacement˄mm)

-40

-60

-80
Fig.2: Typical failure of J1~J4 (a) Joint of J0
1876 LU Zhoudao et al. / Procedia Engineering 14 (2011) 1871–1878

load˄kN)
100
100

load˄kN)
80 80

60 60

40 40
20
20
0
0
-200 -100 -20 0 100 200
-150 -100 -50 0 50 100 150
-20 displacement(mm)
displacement(mm)
-40
-40
-60
-60 -80

-80 -100

(b) Joint of J1 (c) Joint of J2

100
load(kN)

100

load(kN)
80
80
60 60
40 40

20 20

0
0
-150 -100 -50 0 50 100 150
-200 -150 -100 -50 0 50 100 150 200 -20
-20 diaplacement(mm)
displacement(mm) -40
-40
-60
-60 -80

-80 -100
(d) Joint J3 (e) Joint of J4
Fig.3: Hysteresis and skeleton curves of the specimens

8 stiffness ratio

J0
J1 7
J0
J2
 J3 J1
6
J4 J2
J3
 5 J4

4

decreased
coefficient of 3

 capacity
2
displacement˄mm)
 displacement˄mm)
1
                       

Fig.4: The degradation factor of strength Fig.5: The degradation factor of stiffness
 LU Zhoudao et al. / Procedia Engineering 14 (2011) 1871–1878 1877

3.4 Stiffness analysis

The load capacity of each specimen decreases as the displacement of column tip increases (Fig4). The
capacity of J0 decreases the most seriously of all, which is similar to characteristics of brittle failure on
the column tip and the joint. The trend of the other four specimens is consistent with J0 and then the
descending trend is significantly slowed down or even rebound. When the displacement reaches 120mm,
the descending trend slows down regain. Ductile failure is showed obviously because passive restraint
supplied by BFRP plays an important role in the joint when it has a greater distortion. In late loading
stage, the increase of capacity degradation shows the gradual involvement of reinforcement material.
Although measures were taken to pre-damaged component, Paste can not reach the depths of the small
cracks in components. Epoxy injection and reinforcement with BFRP can only improve the seismic
performance to some extent (Fig5). When the displacement is more than 84mm, the stiffness of J0
decreased rapidly, but the others slowly. BFRP makes an important role in limiting the cracks and
delaying the structural stiffness attenuation. The joint may maintain the proper capacity in a larger
deformation to improve the ductility. Therefore the reinforced method is effective to improve the seismic
performance of 3D column-beam joint.

4. Conclusions

Five repaired column-beam joints strengthened by BFRP have been tested including pre-damage,
rehabilitation and re-test under cyclic loads concentrated on the effect on different levels of pre-damage
and epoxy injection in order to study on the seismic behavior. Through the experimental data and analysis,
the conclusion can be listed as follows:
(1) The hysteresis and skeleton curves of the joints reinforced by BFRP become more ideal. In the
early loading period the contribution of BFRP to the capacity and stiffness is small. When the
displacement exceeds 36mm, the degradation of the capacity and stiffness to the reinforced joint is
less than that without reinforcement. The BFRP sheets are equivalent to provide additional
transverse reinforcement to confine the concrete and make the joint ductile failure.
(2) Ultimate capacity and displacement of the joint with reinforcement have been increased. The fact
that ultimate capacity has been increased by 11.5% and meanwhile ultimate displacement has been
increased by 71.6% shows that the reinforcement by BFRP can improve the ductility effectively. In
addition, epoxy injection has little effect on ultimate capacity and displacement.
(3) The seismic performance of the joints damaged by cyclic loads and reinforced by BFRP can be
improved significantly. From the experimental data such as capacity, stiffness and ductility the
seismic performance of the joints with reinforcement has reached or even exceeded the levels that
were not damaged before. The failure mode of strong beam and weak column is changed into
strong column and weak beam after rehabilitations and the method of BFRP-reinforced pre-
damaged concrete column-beam joints is reasonable and effective.

Acknowledgements

The authors would like to acknowledge the state key laboratory of disaster reduction in civil
engineering of Tongji University. The work described in this paper was fully supported by the Ministry of
Science and Technology of China, Grant No. SLDRCE09-D-02.
1878 LU Zhoudao et al. / Procedia Engineering 14 (2011) 1871–1878

References

[1] Ye Lieping, Qu Zhe, Ma Qianli, et al. Study on ensuring the strong column-weak beam mechanism for RC frames based on
the damage analysis in the Wenchuan Earthquake. Building Structure, 2008, 38(11): 52-59. (in Chinese)
[2] Wang Yayong. Lessons learnt from building damages in the Wenchuan earthquake-seismic concept design of buildings.
Journal of Building Structures, 2008, 29(4), 20-25. (in Chinese)
[3] Leon R.T. The Effect of Floor Member Size on the Behavior of Reinforced Concrete Beam-Column Joints. Proceedings of
8th World Conference on Earthquake Engineering, San Francisco, Calif., July, 1984: 445-452.
[4] Paulay T. Developments in the Design of Ductile Reinforced Concrete Frames-. Bulletin of the New Zealand National
Society for Earthquake Engineering, 1979, 12(1): 35-43.
[5] Park Y J, Ang A H-S. Mechanistic Seismic Damage Model for Reinforced Concrete. Journal of Structural Engineering,
ASCE. Vol.ċ, No.4, 1985.
[6] Jiang Yongsheng, Chen Zhongfan, Zhouxuping. et al. Aseismic Research on RC Frame Joint of Monolithically Casted Slab.
Journal of Building Structures, 1994, 12(3), 11-16. (in Chinese)
[7] French C.W. and Moehle J.P. Effect of Floor Slab on Behavior of Slab-Beam-Column Connections. Design of Beam-
Column Joints for Seismic Resistance, SP-123, American Concrete Institute, Farmington Hills, Mich., 1991: 225~258.
[8] Ammerman O.V. and Wolfgram-French C. R/C Beam-Column-Slab Subassemblages Subjected to Lateral Loads. Journal of
Structural Engineering, 1989, 115(6): 1298-1308.
[9] Azadeh Parvin, Shanhong Wu. Ply angle effect on fiber composite wrapped reinforced concrete beam–column connections
under combined axial and cyclic loads. Composite Structures, 2008, 82(2): 532-538.
[10] Abhijit Mukherjee, Mangesh Joshi. FRPC reinforced concrete beam-column joints under cyclic excitation. Composite
Structures, 2005, 70(9): 185-199.
[11] M.A. Adin, D.Z. Yankelevsky and D.N. Farhey, Cyclic behavior of epoxy repaired reinforced concrete beam–column joints.
Struct J, ACI. 1993, 90(2): 170-179.
[12] Hong tao. A Study on Performance of Injured Interior Column-beam-slab Subassemblies Rehabilitated by Pasted Carbon
Fiber Reinforced Plastics . Shanghai: Tongji University, 2002. (in Chinese)
[13] M. J. N. Priestley, F. Seible, Y. Xiao, R. Verma; Steel Jacket Retrofitting of Reinforced Concrete Bridge Columns for
Enhanced Shear Strength [M]. Part 2: Test Results and Comparison with Theory, ACI Structure Journal, 1994.
[14] XiaoY, Wu H, Matin G R. Prefabricated Composite Jacketing of RC Columns for Enhanced Shear Strength [J]. Journal of
Structural Engineering, ASCE, 1999, 125(3), 1357-1364.
[15] Jongsung Sim,Cheolwoo Park,Do Young Moon. Characteristics of basalt fiber as a strengthening material for concrete
structures[J], J.Simetal./Composites: 2005 (36) Part B:504-512.