![Lateral-torsional-buckling (a) numerical (this research), (b) tested beam [29]](https://www.researchgate.net/profile/Kambiz-Narmashiri/publication/271830431/figure/fig6/AS:668701472874502@1536442263356/Lateral-torsional-buckling-a-numerical-this-research-b-tested-beam-29_Q320.jpg)
Content uploaded by Kambiz Narmashiri
Author content
All content in this area was uploaded by Kambiz Narmashiri on Feb 04, 2015
Content may be subject to copyright.
200 | PART A. APPLIED AND NATURAL SCIENCES
INTERNATIONAL JOURNAL
o
f
ACADEMIC RESEARCH
Vol.
7
. No.
1. January
, 201
5
Library of Congress Classification: TA1-2040, TA630-695
FLEXURAL STRENGTHENING OF DEFICIENT
STEEL BEAMS USING CFRP
Omid Yousefi1, Kambiz Narmashiri2*, Ali Ghods3
1,3Department of Civil Engineering, College of Engineering,
Zahedan Branch, Islamic Azad University, Zahedan,
2Assistant Professor of Structural Engineering, Department of Civil Engineering, College of Engineering,
Zahedan Branch, Islamic Azad University, Daneshgah Street, Zahedan 98168 (IRAN)
*Corresponding author: narmashiri@iauzah.ac.ir, narmashiri@yahoo.com
DOI: 10.7813/2075-4124.2015/7-1/A.31
Received: 29.10.2014
Accepted: 11.01.2015
ABSTRACT
There are a large number of steel buildings and bridges deteriorated due to aging and corrosion and need to be
repaired. Carbon fibre reinforced polymer (CFRP) plates are proposed to fill the lack of strength lost. This research presents
the flexural behavior of deficient steel beams refurbished with CFRP strips. The various deficiencies size are intentionally
created by notching one side and both sides of the tension flange of the beams. Fourteen specimens were examined to
evaluate the static performance of the repaired beams with emphasis on load bearing capacity and the CFRP–steel interface
stress. A three-dimensional finite element analysis (FEA) by ABAQUS is utilized. Results show that CFRP-repair cause a
recovery of static load-carrying capacity of the deficient beams. By using CFRP within 17% of beam length, it indicates
performance as a non-deficient beam. Besides, local buckling, vertical deflection and the deficiencies influences on the
beams were investigated.
Key words: Steel beam; CFRP; Deficiency; Strengthening
1. INTRODUCTION
Structures may need to be rehabilitated for different reasons, e.g. design fault, material degradation, and change in
the load acting on the structures. Some steel structures may suffer problems related to fatigue, decay, and humidity and
consequently their load bearing capacity may be significantly reduced. Standard rehabilitation techniques make use of
welded or bolted steel plate to strengthen the damaged elements but the problem of steel corrosion still remains and
difficulties in fitting complex profiles can arise. Fiber reinforced polymer (FRP) materials and in specific carbon fiber
reinforced polymer (CFRP) materials may be used to strengthen steel structures because of their well-known high
mechanical properties and high strength to weight ratio. Retrofitting of current beams by bonding a CFRP plate to the tension
flange may produce an increase of both the flexural capacity and the local rigidity.
Normally, the FRP for flexural strengthening are installed to the bottom (tensile) flange. Edberg et al. [1] presented
an experimental study in which five different configurations of glass (GFRP) and carbon (CFRP) fibre reinforced polymers
were bonded to the tensile flange of small scale steel wide flange beams using adhesive bonding. Also, a similar study was
carried out by Ammar [2]. In addition, Tavakkolizadeh and Saadatmanesh [3] tested small scale steel beams in four-bending
test method. All these aforementioned researches showed that it is feasible to flexural strengthen using CFRP plates.
Identification of CFRP failure modes in flexural strengthening of steel I-beams is useful in order to overcome or retard these
failures. Deng et al. [4] had highlighted an important feature of the reinforced steel beam which is the significant stress
intensity on the adhesive at the tip of the CFRP plate due to discontinuity by the abrupt termination of the CFRP plate.
Buyukozturk et al. [5] reviewed the achievements in the strengthening of both reinforced concrete and steel members. They
concluded that failures of FRP flexural strengthened reinforced concrete (RC) and steel members occur due to different
mechanisms, and it is dependent on the parameters of strengthening. They found that shear failure takes place when the
shear capacity of the beam is not able to accommodate the increment of the flexural capacity due to flexural strengthening.
They indicated that the following are the failure modes of an FRP steel member: (a) buckling of top flange in compression,
(b) buckling of web in shear, (c) FRP rupture, and (d) FRP debonding.
FRP has great advantages as a structural material which restore the lost capacity of damaged structures [6, 7]. FRP
sheets/strips are also effective in strengthening of steel structural elements to improve their fatigue life time and reduce crack
propagation [8, 9], galvanic corrosion is prevented and adequate bond is provided [10, 11]. Recent strengthening projects in
Japan, the United States and United Kingdom indicated that there was great potential for using CFRP to upgrade steel
structural members. The Action Bridge in England [12] was strengthened with applying CFRP elements to bottom of the
girders and stress reduction on the original materials was observed, and the fatigue life was enhanced. Also, study on
increasing fatigue life for both analytical and numerical investigations on small scale specimens [13-16] and full scale
specimens [17-20] were carried out. Prestressing was also used to cause an everlasting compressive stress in bolted details
in order to prevent the fatigue crack growth [21]. Kim and Brunell [22] and Kim and Harries [23] have studied the interaction
between CFRP and an initial damage applied at the mid-span of wide-flange steel beams in a three-point setup under fatigue
loading.
The major conducted investigations are about fatigue life under cyclic loading by initial minor damage, in order to
create crack propagation. By increasing of cross-section damage, it tends to damage propagation instead of creating crack
under static loading.
Baku, Azerbaijan | 201
INTERNATIONAL JOURNAL
of
ACADEMIC RESEARCH
Vol.
7
. No.
1. January
, 201
5
This study examines the structural behaviors and larger dimensions of the notched steel I-beams strengthened using
CFRP in the deficiency area. The numerical simulation utilized using ABAQUS software. Firstly, the simulation was validated
with the experimental researches available in the literature, then the investigations took place.
2. MATERIALS AND METHODS
In order to investigate the CFRP behavior to strengthening deficient beams and approach to the best CFRP length
which has a performance as a non-deficient beams, fourteen beams with various deficiencies dimensions and different
CFRP strips length are chosen. Figs 1, 2 show the schematics of the beams and deficiencies.
Fig. 1. Steel beams dimensions and location of deficiency region
Fig. 2. Schematic of deficiencies in bottom flange
2.1. Materials
The steel I-sections with the mild-steel property are chosen. Table 1 shows the dimensions and material properties
of the selected steel I-sections.
The CFRP materials have high tensile strength which can develop the structural behavior of structures. Normally,
CFRP is produced in the form of a strip (plate) or a sheet (wrap). In this research, only CFRP strips are used. According to
the one directional behavior of CFRP strips, they are grouped in the orthotropic materials [24]. CFRP strips are made in
different widths and thicknesses and classified based on the strength and modulus of elasticity. In this research, the following
types of CFRP strips is used: normal tensile strength–intermediate modulus (IM). The material properties of this type of
CFRP strip is shown in Table 2. It has cut in 82 mm width. The chosen thickness of CFRP strips is 1.4 mm. For the deficient
beams strengthened using CFRP, different bonded lengths are selected i.e. 400 mm, 500 mm, and 700 mm and 1000 mm.
The dimensions of the CFRP strips are shown in Table 2.
The engineering epoxy (structural adhesive) for installing the CFRP strips on the steel structures must be strong
sufficient to transfer the interfacial stress between the common surfaces [25, 26]. A structural adhesive is chosen to be
applied as it is widely used [27, 28]. This adhesive is a two-part epoxy resin (resin and hardener, in 3:1 proportions) that
must be mixed with 1% in weight of ballotini (1 mm thickness) to ensure a uniform thickness of the bond line. Table 3 shows
the dimensions and material properties of the selected adhesive.
Table 1. Dimensions and material properties of steel I-section
Steel I
-
section
–
mild steel IPE
-
160
Steel I
-
section dimensions (mm)
E
-
modulus
(N/mm2) Mean
value
Stress (N/mm
2
)
Strain
Width
High
Flange thick
Web thick
Yielding (F
y
)
Ultimate (F
u
)
Yielding (
ɛ
y
) %
Ultimate (
ɛ
u
) %
82
160
7.4
5.0
200,000
250
370
0.12
13.5
Table 2. Dimensions and material properties of CFRP strips
Dimensions (mm)
Elasticity modulus
(N/mm2)
Tensile strength (N/mm
2
)
Strain at
break
Width
Thickness
length
82
1.4
4
00, 500, 700, 1000
210,000
2400
1.35%
202 | PART A. APPLIED AND NATURAL SCIENCES
INTERNATIONAL JOURNAL
o
f
ACADEMIC RESEARCH
Vol.
7
. No.
1. January
, 201
5
Table 3. Dimensions and material properties of adhesive
Adhesive
Dimensions (mm)
Compressive strength
(N/mm2)
Tensile strength (N/mm
2
)
Shear strength
(N/mm2)
Bond strength (N/mm
2
)
Width
Thick.
Length
E
-
modulus
St
rength
(7 days)
E
-
modulus
Strength
(7 days)
Strength
(7 days)
Mean value
Min.
value
82
1.0
Var.
5485
80
-
110
11,520
22.7
31.7
20
>15
2.2. Specifications of the beams
Specifications, load bearing capacities, and deficient dimension of the beams are shown in Table 4. As it indicates
different CFRP lengths are selected to investigate the effects of the deficiencies dimensions and CFRP lengths on the load
capacities. Dimensions of steel I-beams are the same for all specimens and the CFRP strips thicknesses are same with 1.4
mm, but CFRP and deficiencies dimensions are various for each beam that applied at the tensile flange. The specimens S1
is non-deficient and not strengthened which used as the control beam. The beams S2, S3, S4, S5, and S6, are deficient with
20×30 mm2 (the first digit is length and second one is width for all deficiencies dimensions) that strengthened with the CFRP
strip varied lengths (400 mm, 500 mm, 700 and 1000 mm). The beams S7, S8, S9, and S10, are deficient with varied
dimensions (30×30 mm2, 30×35 mm2, 35×30 mm2) which S10, is strengthened with 400 mm in length of CFRP. The beams
S11, S12, S13, and S14, are deficient with 30×30 mm2 in dimension at one-side of tensile flange, that S12, and S14, are upgraded
with 400 mm and 2×400 mm and S13 with 500 mm in length of CFRP respectively. All beams were numerically simulated.
Table 4. Specifications, load bearing capacities, and deficient dimension of the beams
No.
Specimen
CFRP length
(mm)
Deficiencies dimensions (mm)
*
Load bearing capa
city
Load
(kN)
Increase/
decrease (%)
1
S
1
N/A
N/A
127
-
2
S
2
N/A
30×20
89
-
30
3
S
3
1000
137
+7
4
S
4
700
133
+4
5
S
5
500
129
+2
6
S
6
400
127
-
7
S
7
N/A
30×30
87
-
32
8
S
8
N/A
30×35
87
-
32
9
S
9
N/A
35×30
78
-
39
10
S
10
400
127
-
11
S
11
N/A
One-Side
30×30
109
-
15
12
S
12
400
127
-
13
S
13
500
129
+2
14
S
14
2×400
127
-
*Deficiencies dimensions: the first digit is length and the second one is width of deficiency
2.3. Numerical simulation
To model the specimens, the full 3D simulation using ABAQUS software was performed.
The steel I-sections, steel stiffeners, CFRP plates, and adhesive were simulated by using the 3D solid triangle
elements (ten-nodes.) The interface of common surfaces was defined between the steel I-beam, adhesive, and CFRP plates.
Debonding, and splitting occurred when the plastic strains exceeded the ultimate strain. Non-linear static analysis was
carried out to achieve the failures. Linear and non-linear properties of materials were defined. The CFRP plate material
properties were defined as linear and isotropic because CFRP materials have linear properties. The steel beams and
adhesive were defined as the materials having non-linear properties. Due to verify, the specimens, F1 (non-strengthened and
non-deficient) and F4 (strengthened with CFRP without any deficiency) from Narmashiri et al. [29] with good agreement were
validated. The results of vertical deformation versus load (Fig. 3) showed that the full-3D modelling case and non-linear
analyses applied in this research, had high accuracy with the experimental tests. (Fig. 3). Also, Fig. 4 makes a comparison
between results of lateral deformation of the beam in numerical simulation used in this research and experimental test
available in the literature [29]. This comparison indicates high accuracy of modelling. After this validation, the investigation on
the beams tabulated in Table 4 can be taken place.
Fig. 3. Validation of vertical displacement at the mid-span (experimental and numerical).
Baku, Azerbaijan | 203
INTERNATIONAL JOURNAL
of
ACADEMIC RESEARCH
Vol.
7
. No.
1. January
, 201
5
Fig. 4. Lateral-torsional-buckling (a) numerical (this research), (b) tested beam [29]
3. RESULTS AND DISCUSSIONS
3.1. Load bearing capacity
In order to approach appropriate CFRP length, deficiencies dimension within 30×20 mm, with the CFRP strip varied
lengths (400 mm, 500 mm, 700 and 1000 mm) were modelled. The deformation in the deficiency area is shown in Fig. 5. As
Fig. 6 indicates the notched deficiency on non-strengthened beam led to reduce load carrying capacity up to 30 %. Due to
recovery and reach to non-deficient beam (control beam) load carrying capacity, 1000 mm, 700 mm, 500 mm, and 400 mm
CFRP length were used, which increase of 37%, 34%, 32%, and 30% were observed, respectively. It found out that 400 mm
is the best choice to reform deficient beam which perform such as control beam and proposed for specimens with bigger
deficiencies. Also, Fig. 7 presents the failures of the strengthened specimens at the deficiency region. End-debonding of
CFRP at the tip of plate and stress intensity on CFRP at the deficiency area are observed. End-debonding occurs due to the
high stress and strain intensity on adhesive at the CFRP tips [8,9]. High strain intensity occurred on adhesive at the CFRP tip
which is shown in Fig. 9
Fig. 5. Deficiency Propagation in tensile flange of non-strengthened beam
Fig. 6. Vertical displacement at the mid-span for deficiencies by 30×20 mm2
204 | PART A. APPLIED AND NATURAL SCIENCES
INTERNATIONAL JOURNAL
o
f
ACADEMIC RESEARCH
Vol.
7
. No.
1. January
, 201
5
Fig. 7. End-debonding of CFRP at the tip of plate and stress intensity on CFRP at the deficiency area
Fig. 8. CFRP end-debonding. Fig. 9. Strain intensity on adhesive at the CFRP tip.
3.2. Effects of deficiencies dimensions
The effects of deficiencies dimensions on the load bearing capacity were studied on the specimens S7, S8, and S9
which include the deficiencies dimensions of 30×30mm2, 30×35 mm2, and 35×30 mm2 respectively. The deficiency
propagation of non-strengthened beams (Fig. 5) was rapid, however deficiency levels was affected by load extent.
Deficiencies (Fig. 10) with 30×30mm2 reduced load carrying capacity by 32%, which is equal for specimen by 30×35 mm2. It
indicates that increasing of width by 5 mm does not effect on ultimate load. Also, in case of S9 with 35×30 mm2 of deficiencies
dimensions decreased by 39%. Thus, it obtained which enhance of length has considerable reduction in ultimate load,
because of vicinity to the web occur. In order for strengthening specimen S9 which is the most critical, 400 mm CFRP length
could recover deficient beam, appropriately. The stress range at deficiency propagation and plastic strain of the repaired
beams were rehabilitated. Near the deficiency, local plasticity was not improved proportionally. Except at the deficiency
location where significant concentrations were observed, the strain of the CFRP strip along the beam span gradually
increased. (Fig. 7)
Fig. 10. Investigation of deficiencies dimensions on Force-Displacement
3.3. Effects of deficiencies locations
In order to investigate one-side deficiency in tensile flange, beams namely S11, S12, S13, and S14 with 30×30 mm2 of
deficiency dimensions was examined. In the case of S11, the deficiency caused reduction in ultimate load by 15% which in
compare with two-side deficiencies is half (Fig. 11). Also, beam with one-side deficiency includes local buckling (Fig. 12).
Strengthening with 400 mm, 500 mm, and 2×400 mm length of CFRP strips for S12, S13, and S14 were applied, respectively.
Using shorter length of CFRP plate caused premature debondong, and applying longer CFRP plate resulted in more flexible
behaviour. However 400 mm, and 500 mm, length recovered ultimate load by 15%. Due to local buckling as a result of one
side deficiency, specimen S14 was strengthened by two strips with 400 mm, length which was settled under deficiency in
bottom flange and on the web adjacent to deficiency for preventing local buckling properly. (Fig. 13)
Baku, Azerbaijan | 205
INTERNATIONAL JOURNAL
of
ACADEMIC RESEARCH
Vol.
7
. No.
1. January
, 201
5
Fig. 11. Vertical displacement for one-side deficiency by 30×30 mm2
Fig. 12. Local buckling for non-strengthened beam which has one-side deficiency
Fig. 13. One-side deficient beam which is strengthened with 2×400 mm2
lengths of CFRP. (a) After analysis, (b) before analysis
4. CONCLUSIONS
This research shows the possibility of using CFRP to recover the strength lost due to deficiency. The stress range at
deficiency propagation and plastic strain of the repaired beams were rehabilitated. Vicinity of the deficiencies, local plasticity
was not improved proportionally. Except at the deficiency location where significant concentrations were observed, the strain
of the CFRP strip along the beam span slowly increased. The deficiency propagation of non-strengthened beams was rapid,
however deficiency levels was affected by load magnitude. The local debonding of the CFRP strip as a result of deficiency
caused a sudden increase in the bond stress and consequently the strip slip. By increasing load, CFRP strains near the
deficiency location and debonding strains were noticed. The length of CFRP affects the flexural behaviour of steel beam.
Using shorter length of CFRP plate caused premature debondong, and applying longer CFRP plate resulted in more flexible
behaviour. Under the deficiency region, strengthened beams indicated a local strain concentration on the CFRP strip, while it
had a uniform strain distribution along the CFRP strip and the local stress concentration on the CFRP was created by the
adhesive layer which can slow the deficiency propagation. CFRP-repair cause a recovery of static load-carrying capacity of
the deficient beams, so that by using CFRP by 17% of beam length, it indicates performance as a non-deficient beam.
Increase length of deficiency has more affected to reduce ultimate load capacity in contrast with increase width. In the case
of two-side deficiencies, reduction of strength and load bearing capacity in comparison by one-side deficiency was two times
approximately. In specimen with one-side deficiency, local buckling as result of asymmetric section occurred. In order to
avoid local buckling, additional CFRP strip was used adjacent to deficiency on the web. This method was effective to reduce
local buckling, adequately.
REFERENCES
1. W. Edberg, D. Mertz, J. J. Gillespie. Rehabilitation of steel beams using composite materials. In: fourth
materials engineering conference. (1996)
2. N. Ammar. Rehabilitation of steel bridge girders with graphite pultrusion. University of, Delaware (CCM
research report 96-26). (1996)
206 | PART A. APPLIED AND NATURAL SCIENCES
INTERNATIONAL JOURNAL
o
f
ACADEMIC RESEARCH
Vol.
7
. No.
1. January
, 201
5
3. M. Tavakkolizadeh, H. Saadatmanesh. Repair of cracked steel girder using CFRP sheet. In: A, editor. Creative
systems in structural and construction engineering. Rotterdam: Balkema; (2001)
4. J. Deng, M. M. K. Lee, S. S. J. Moy. Stress analysis of steel beams reinforced with a bonded CFRP plate.
Composite Structures. 65:205–15 (2004)
5. O. Buyukozturk, O. Gunes, and E. Karaca. Progress on understanding debonding problems in reinforced
concrete and steel members strengthened using FRP composites. Construction and Building Materials.18:9–
19 (2004)
6. Y. J. Kim and P. J. Heffernan. Fatigue behavior of externally strengthened concrete beams with fiber-
reinforced polymers: state of art. Journal of Composites for Construction.12:246-56 (2008)
7. L. Van Den Einde, L. Zhao, and F. Seible. Use of FRP composites in civil structural applications. Construction
and building materials. 17:389-403 (2003)
8. N. G. Tsouvalis, L. S. Mirisiotis, andD. N. Dimou. Experimental and numerical study of the fatigue behaviour of
composite patch reinforced cracked steel plates. International journal of fatigue. 31:1613-27 (2009)
9. Q. Yu, T. Chen, X. Gu, X. Zhao, Z. Xiao. Fatigue behaviour of CFRP strengthened steel plates with different
degrees of damage. Thin-Walled Structures. 69:10-17 (2013)
10. A. Shaat, D. Schnerch, A. Fam, and S. Rizkalla. Retrofit of steel structures using fiber- polymers (FRP): state-
of-theart. In Proceedings of the Transportation Research Board (TRB) Annual Meeting. CD-ROM (04-4063)
(2004)
11. K. A. Harries, A. Varma, S. El-Tawil, J. Liu, Y. Xiao, S. Moy, et al. Steel-FRP composite structural systems. In
Proceedings of the 6th International Conference on Composite Construction in Steel and Concrete. 703–716
(2011)
12. S. Moy. Early age curing under cyclic loading- an investigation into stiffness development in carbon fibre
reinforced steel beams. In Advanced Polymer Composites for Structural Applications in Construction.
Proceeding of the First International Conference, Held at Southampton University, UK, April (2002)
13. P. Colombi. Plasticity induced fatigue crack growth retardation model for steel elements by composite patch.
Theoretical and Applied Fracture Mechanics. 43:63–76 (2006)
14. H. Liu, R. Al-Mahaidi, and X.-L. Zhao. Experimental study of fatigue crack growth behaviour in adhesively
reinforced steel structures. Composite Structures.90:12–20, (2009)
15. H. Liu, Z. Xiao, X. L. Zhao, R. Al-Mahidi. Prediction of fatigue life for CFRP strengthened steel plates. Thin-
Walled Structures. 47:1069–77 (2009)
16. H. Nakamura, W. Jiang, H. Suzuki, K. Maeda, T. Irube. Experimental study on repair of cracks at welded web
gusset joint using CFRP strips. Thin-Walled Structures. 47:1059–68 (2009)
17. M. Dawood, S. Rizkalla, Sumner E. Fatigue and overloading behaviour of steel concrete flexural members
strengthened with high modulus CFRP materials. Journal of Composite Construction. 11(6):659–69 (2007)
18. S. C. Jones, S. A. Civjan. Application of fibre reinforced polymer overlays to extend steel fatigue life. Journal of
Composite Construction. 7(4):331–8 (2003)
19. K. Nozaka, C. K. Shield, J. F. Hajjar. Effective bond length of carbon-fiber-reinforced polymer strips bonded to
fatigue steel bridge I-girders. J Bridge Eng (ASCE).10(2):195–205 (2005)
20. M. Tavakkolizadeh and H. Saadatmanesh, “Fatigue strength of steel girders strengthened carbon fiber
reinforced polymer patch. Journal of Structural Engineering. 129(2):186–96 (2003)
21. B. Taljsten, CS. Hansen, JW. Schmidt. Strengthening old metallic structures in fatigue with prestressed and
non-prestressed laminates. Constr Build Mater. 23:1665–77 (2009)
22. Y. J. Kim, G. Brunell. Interaction between CFRP-repair and initial damage of wideflange steel beams subjected
to three-point bending. Composite Structures. 93(8):1986–96 (2011)
23. Y. J. Kim, K. A. Harries. Fatigue behavior of damaged steel beams repaired with CFRP strips. Engineering
Structures. 33(5):1491–502 (2011)
24. D. Linghoff, R. Haghani, M. Al-Emrani. Carbon-fibre composites for strengthening steel structures. Thin-Wall
Structures. 47:1048–58 (2009)
25. D. Schnerch, K. Stanford, E. Sumner, and S. Rizkalla. Bond behavior of CFRP steel bridges and structures. In
Proceedings of the International Symposium on Bond Behaviour of FRP in Structures (BBFS ’05), International
Institute for FRP in Construction, (2005)
26. D. Schnerch, M. Dawood, S. Rizkalla, E. Sumner, and K. Stanford. Bond behavior of CFRP strengthened steel
structures. Advances in Structural Engineering. 9: 805–817, (2006)
27. M. Bocciarelli. Response of statically determined steel beams reinforced by CFRP plates in the elastic plastic
regime. Engineering Structures. 31:956–67.(2009)
28. J. Deng and M. M. K. Lee, Fatigue performance of metallic beam strengthened with a bonded CFRP plate,
Composite Structures. 78: 222–231 (2007).
29. K. Narmashiri, N. H. Ramli Sulong, M. Z. Jumaat. Failure analysis and structural behavior of CFRP
strengthened steel I-beams. Construction and Building Materials. 30:1-9 (2012)
