Engineering Structures 118 (2016) 108–124

Contents lists available at ScienceDirect

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

IC debonding failure in RC beams strengthened with FRP: Strain-based

versus stress increment-based models
Julio C. López-González a,⇑, Jaime Fernández-Gómez b, Elena Díaz-Heredia b, Juan C. López-Agüí c,
Paula Villanueva-Llauradó b
a
Tecnologico de Monterrey, Campus Aguascalientes, Eugenio Garza Sada, 1500, 20328 Aguascalientes, Mexico
b
School of Civil Engineering, Technical University of Madrid, Prof Aranguren, s/n, 28040 Madrid, Spain
c
School of Mining Engineering, Technical University of Madrid, Ríos Rosas, 21, 28003 Madrid, Spain

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

Article history: Since the advent of the use of fibre-reinforced polymers (FRP) to strengthen reinforced concrete (RC) and
Received 9 February 2015 its employment in civil engineering in the early nineteen nineties, much of the research has concerned
Revised 6 March 2016 the enhancement of bending capacity. Failure modes are well understood and can be predicted with some
Accepted 8 March 2016
accuracy. The models adopted by the most highly reputed international standards for the specific case of
inter-crack (IC) debonding are based on limiting the strain on the laminate. These methods, an offshoot of
fracture mechanics analysis, are calibrated with the results of point load beam tests. This paper proposes
Keywords:
a new interfacial fracture energy model by characterising the FRP-concrete interface with beam-type
Concrete
FRP
tests. The existing fracture energy models, as well as the proposed model, were applied to an existing
Debonding stress increment-based formulation, in which no calibration was conducted with experimental results
Composite materials and yet proved to predict inter-crack debonding failure moderately better than the models in use. The
Fracture energy performance of each IC debonding model is assessed through an exhaustive statistical analysis with
experimental data of point load RC beams tests gathered from literature. The behaviour of the models
in uniformly loaded RC beams is discussed and the implications of each approach are drawn.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction
The vast majority of IC debonding models are based on fracture
mechanics analysis of pure shear (mode II). In Taljsten’s [1] early
proposals based on linear elastic fracture mechanics and assuming
no deformation in the concrete substrate, the maximum transfer-
able force is given by Eq. (1).
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
F max ¼ bp 2Gf Ep tp
where Gf is the interfacial fracture energy and bp , t p and Ep are
width, thickness and modulus of elasticity of the FRP material
respectively. Shear stress distribution in debonding has been
studied by a number of researchers. Teng et al. [2] analysis is
particularly significant, for it envisages tensile stress acting on both
sides of the laminate between two adjacent cracks, which can be
used as a proxy for crack-induced debonding failure. In that
analysis, expressions for interfacial shear stress, tensile stress on
the FRP laminate and interfacial slip are defined for each stage of
debonding. Shear stress distribution can be readily calculated by
numerical analysis, however, where the FRP-concrete interface is
modeled using spring elements whose mechanical behaviour is
defined by the bond-slip relationship [3]. Teng et al. [2] analyses
assume a linear bond-slip relationship, while the numerical solution
accommodates analysis of different shapes of the descending branch
of this function. Chen et al. [4] conducted a simpler analysis that
ð1Þ addresses the descending branch of the bond-slip relationship only.
That simplification gives an explicit expression for ultimate load: by
assuming no deformation in the concrete substrate the expression
adopts the form of Eq. (1) multiplied by the term which accounts
for the effect of bonded length and load ratio of debonding:
(
pffiffiffiffiffiffiffiffi
1
if Lb P Ls
1t2
bL ¼ sinðkLb Þ
ð2Þ
if Lb < Ls
1t cosðkLb Þ
where

arcos t smax
⇑ Corresponding author. Ls ¼ and k ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð3Þ
k 2Gf Ep tp
E-mail address: julio.logo@gmail.com (J.C. López-González).

http://dx.doi.org/10.1016/j.engstruct.2016.03.024
0141-0296/Ó 2016 Elsevier Ltd. All rights reserved.
 J.C. López-González et al. / Engineering Structures 118 (2016) 108–124 109

0:19
In the above equations, t is the ratio between the lowest and Gf ¼ 0:644f c ð7Þ
highest FRP force, Lb is the bonded length (i.e. crack spacing), Ls pffiffiffiffiffiffiffiffiffi
0:649 Ep tp
is the characteristic softening length and smax is the maximum Leb ¼ 0:095
ð8Þ
shear stress. fc
Comparisons of IC debonding models have shown that strain In the more complex procedure described in fib Bulletin 14 [6],
limitation-based models have greater prediction power [5]. In this the allowable stress increment depends on the lowest value in the
paper, the existing interfacial fracture energy models, as well as area studied. As might be intuitively expected, this procedure calls
the proposed model, were applied to the Chen et al. [4] formula- for iteration and is consequently too complex for practical applica-
tion. In this formulation no calibration was conducted with exper- tions. It consists of three main stages. In the first one, the least
imental results; however, it proved to predict IC debonding failure favourable crack spacing is calculated from the following expres-
moderately better than the models in use. sion, assuming mean bond stress for both the internal steel rein-
forcement and the external strengthening:
2. IC debonding models  
M cr 1
srm ¼ 2 P P ð9Þ
Three major approaches to predicting IC debonding can be zm spm bp þ ssm ds p
defined: limitation of strain in the FRP laminate at the cross-
section where the bending moment is highest; mean bond stress where ds is the diameter of the steel bars. Mcr is the cracking
(expressed as the difference in the tensile forces on the FRP moment, zm is the mean lever arm of internal forces, ssm and spm
laminate divided by the bond area); and the allowable increment are the mean bond stress of the internal steel reinforcement and
in stress on the FRP composite. the external strengthening respectively. These parameters are cal-
culated as follows:
2.1. Strain limitation approach 2
Mcr ¼ 2f ctk;0:95 bh =6 ð10Þ
Earlier studies suggested limiting the strain in the FRP laminate
hEp Ap þ dEs As
to a value ranging from 0.0065 to 0.0085 (fib [6]). However, and zm ¼ 0:85 ð11Þ
Ep Ap þ Es As
since IC debonding depends on a number of variables, a fixed value
could lead to uneconomical solutions. Later methods were based ssm ¼ 1:85f ctm ð12Þ
either on calibrating Eq. (1) (e.g. ACI-440.2R-08 [7], Eq. (4)) or
developing empirical models, as proposed by Said and Wu [5] spm ¼ 0:44f ctm ð13Þ
(Eq. (5)).
sffiffiffiffiffiffiffiffiffi where f ctk;0:95 is the upper bound characteristic tensile strength of
fc concrete, f ctm is the mean value of the tensile concrete strength,
epd ¼ 0:41 6 0:9epu ð4Þ
t p Ep Ap is the cross sectional area of the FRP reinforcement, As is the area
0:2 of the longitudinal steel reinforcement, Es is the modulus of elastic-
fc
epd ¼ 0:23 ð5Þ ity of the steel reinforcement and d, b and h are the effective depth,
ðEp t p Þ0:35 the width and the total depth of the beam respectively.
In the above equations, f c is the concrete compressive strength, In the second stage, the stress on the laminate at each crack is
which is a significant variable of the interfacial fracture energy. found via force equilibrium and strain compatibility, which entails
applying the shift rule to the law of moments as per EC-2 [12]. The
2.2. Mean bond stress approach maximum allowable stress increment is calculated as shown
below:
This approach is described in fib Bulletin 14 [6] as an alternative If rp1 6 rðBÞ
p
method for verifying tangential bond stress, whose mean value, ðAÞ ðBÞ
sbd , is limited to 1:8f ctk =cc . However, inasmuch as this model yields Drp;max  Drp;max
Drpd ¼ DrðAÞ
p;max  rp1 ð14Þ
highly scattered results, and is overly optimistic in most cases rðBÞ
p
[5,8], it has not been included in the present comparative study.
If rp1 > rpðBÞ
2.3. Stress increment limitation approach " rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
qffiffiffiffiffiffiffiffiffiffiffiffiffiffi ! #
2
Drpd ¼ min 1=cc c21 Ep f ck f ctm =t p þ ðrp1 Þ  rp1 ; ðf pu  rp1 Þ
Both Japanese standard [9] and fib Bulletin 14 [6] address
this approach. In the Japanese code, the limitation is defined by ð15Þ
Eq. (1), expressed as stress on the laminate (Eq. (6)), using a design
length of 150–250 mm. In the absence of tests, the standard where
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
qffiffiffiffiffiffiffiffiffiffiffiffiffiffi
suggests using Gf ¼ 0:5 N/mm.
DrðAÞ ¼ ðc =c Þ Ep f ck f ctm =t p ð16Þ
sffiffiffiffiffiffiffiffiffiffiffiffiffi p;max 1 c

2Gf Ep qffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Drpd ¼ ð6Þ rðBÞ
p ¼ c3 Ep =srm  c 4 f ck f ctm srm =4t p ð17Þ
tp rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi !
qffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ðBÞ ðBÞ 2 ðBÞ
Wu and Niu [10] proposed finding the rise in stress in a given length
rp;max ¼ 1=cc c1 Ep f ck f ctm =tp þ ðrp Þ  rp
2
ð18Þ
as the greater of the double of the effective bond length and the dis-
tance between the critical section and the end of the yield region at f ck is the characteristic value of the concrete compressive strength,
the debonding load, Ly . Based on the results from double-face shear cc is the material safety factor for the concrete and rp1 is the lowest
type tests conducted by Nakaba et al. [11], the aforementioned stress in the area studied. The values of constants c1, c2, c3 and c4
authors suggested calculating the interfacial fracture energy and are 0.23, 1.44, 0.185 and 0.285, respectively. These constants are
the effective bond length with Eqs. (7) and (8), respectively. related to the stress-slip relationship as follows:
110 J.C. López-González et al. / Engineering Structures 118 (2016) 108–124

qffiffiffiffiffiffiffiffiffiffiffiffiffiffi
smax ¼ c4 =cc f ck f ctm ð19Þ specimens were prepared, five per each of the three types of con-
crete (C20, C40 and C60). High strength concrete was used so that
du ¼ c3 ¼ c21 =c4 ð20Þ the proposed fracture energy formula was suitable for a wide range
qffiffiffiffiffiffiffiffiffiffiffiffiffiffi
of concrete strengths. The half-beams were fitted with eight strain
Gf ¼ 0:5smax du ¼ 0:5c21 = cc f ck f ctm ð21Þ
gauges to measure strain in the composite with a computer-
controlled data acquisition system that recorded 10 readings per
where du is the ultimate slip.
second to monitor the damage to the FRP-concrete bond.
The strengthening system consisted of 50-mm wide, 1.2-mm
3. Experimental programme for FRP-concrete fracture energy thick pultruded plates. A uniform 2-mm thick layer of adhesive
was applied to emulate standard construction practice. Material
The fracture energy for the FRP-concrete interface was mea- properties are given in Table 1.
sured through a series of beam-type tests similar to the trials used Piecewise polynomial interpolation can be applied to strain
to measure steel-concrete bonds. The specimens were prepared as readings to obtain an estimate of tangential stress and slip from
specified in the Spanish-European UNE-EN-10080 standards. The Eqs. (22) and (23), respectively,
inner edge of each half-beam was bevelled as shown in Figs. 1
dpðxÞ
and 2 to avoid stress over-concentration. A total of fifteen s¼ Ep t p ð22Þ
dx
Z Lb
d¼ pðxÞdx ð23Þ
x
P/2 P/2 where pðxÞ is the piecewise polynomial. In the present study, first
150 100
degree (i.e., linear) polynomials were applied. The graphs in Figs. 3–
5 plot stress against slip at a point located 30 mm from the inner
180 edge in the failed half-beams. The estimated stress-slip relationship
Adhesive
was plotted for each set of specimens, in which the maximum shear
stress, the slip at maximum shear stress, and the ultimate slip were
50 defined as the mean for the group of failed half-beams. The
220 FRP response was fitted using the Popovics [13] equation (Eq. (24)),
300 50 which largely reproduces that behaviour as shown on the graphs.
The graphs also show the respective n parameter and interfacial
Strain gauges fracture energy Gf (area subject to the stress-slip relationship on
the interval ½0; du ).

Fig. 1. Experimental set-up.

τ = 5.96 δ = 0.07 δ = 0.20
max 0 u
10
n = 3.15

8 G = 0.77
f
Stress (MPa)

0
0 0.05 0.1 0.15 0.2 0.25 0.3
Slip (mm)

Fig. 2. Beam test. Fig. 3. Stress-slip relationship in family C20.

Table 1
Mechanical properties.

Material Compressive strength (MPa) Young’s modulus (GPa) Tensile strength (MPa) Elongation (%)
Concrete C20 27.15a – – –
Concrete C40 46.2 a – – –
Concrete C60 68.6 a – – –
FRP plate – 165b 3100b 1.7b
Adhesive – 11.2c 18-21d –
a
On the test day. As per Spanish–European standard UNE-EN12390-3.
b
As per European standard EN 2561.
c
As per international standard ISO 527.
d
As per German standard DIN 53455 (1-day specimens).
 J.C. López-González et al. / Engineering Structures 118 (2016) 108–124 111

τ = 6.65 δ = 0.19 δ = 0.25 4. Existing interfacial fracture energy models

max 0 u

12 n=13.44 Existing interfacial fracture energy models can be found in the

G = 0.97
literature either by explicit expressions [14,15] or by calculating
f
the area beneath the bond-slip curve [11,16–18], both of them cal-
10
ibrated against pure shear tests. The fracture energy Gf and the
parameters of the bond-slip curve are usually expressed as a func-
Stress (MPa)

8
tion of concrete strength. However, the influence of other variables
such as the FRP-to-concrete width ratio [15,16,18] and mechanical
6
properties of the adhesive [14,15,18,19] has been reported. Lu et al.
[15] reported that the bond-slip curve depends little on shear stiff-
4 ness of the adhesive layer (Ga =ta ) for values ranging from 2.5 to
10 GPa/mm. However, Dai et al. [14] reported that interfacial frac-
2 ture energy can be improved through the use of a very soft adhe-
sive layer, being the shear stiffness between 0.14 and 1 GPa/mm.
0 In Savoia et al. [17] the bond-slip curve parameters were obtained
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 from experimental pull-off tests with Ga =ta of 0.29 GPa/mm. The
Slip (mm) shear stiffness of adhesive layer was not reported in Nakaba
et al. [11]; it should be noted, however, that bond-slip curves were
Fig. 4. Stress-slip relationship in family C40.
similar to those reported with very soft adhesives (i.e. lower initial
stiffness and slower softening after the maximum shear stress).
The influence of the mechanical properties of FRP material is barely
τ = 11.13 δ = 0.10 δ = 0.30
max 0 u reported because it is rather small [14]. Explicit interfacial fracture
15
energy expressions and bond-slip models available in existing lit-
n = 4.05 erature are summarised in Table 2.
Gf = 1.73
5. Expression for predicting IC debonding failure
10
Stress (MPa)

The Chen et al. [4] formulation has been expressed in the form
of the stress increment on the laminate in order to comply with the
classification set out in Section 2. From Eqs. (1) and (2) assuming
that the bonded length (i.e. crack spacing) is larger than Ls , the
5
maximum FRP stress at debonding takes the following form:
sffiffiffiffiffiffiffiffiffiffiffiffiffi
2Gf Ep 1
rp2 ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2 ð26Þ
tp
1  rp1 =rp2
0
0 0.1 0.2 0.3 0.4 0.5
where rp1 and rp2 are respectively the lowest and highest tensile
Slip (mm) stresses on the FRP laminate. Eq. (26) can be expressed in a simpler
Fig. 5. Stress-slip relationship in family C60.
form:
2Gf Ep
  r2p2 ¼ þ r2p1 ð27Þ
d n tp
s ¼ smax ð24Þ
d0 ðn  1Þ þ ðd=d0 Þn Let
The C20 family specimens exhibited concrete failure. Concrete Dr2p ¼ r2p2  2rp2 rp1 þ r2p1 ð28Þ
microcracking began at peak stress, which rapidly led to deteriora-
tion and sudden failure due to the low tensile strength of the con- where Drp is the stress increment on the laminate in the bonded
crete. The FRP laminate was essentially covered by the adhesive length (i.e. Drp ¼ rp2  rp1 ). Introducing the necessary terms of
and a roughly even layer of concrete (1–2 mm thick). Since the Eq. (28) into Eq. (27) yields the allowable stress increment on the
FRP-concrete bond strength strongly depends on the concrete ten- FRP laminate between two adjacent cracks:
sile strength, the peak stress and ultimate slip were greater in the sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C60 family, while the greatest slip value (d0 ) reached at peak stress 2Ep Gf
Drpd ¼ þ 2r2p1  2rp1 rp2 ð29Þ
was observed in the C40 specimens. In the bond-slip relationship tp
of the C40 family the downward leg decreased steeply which
was attributed to the fact that the bond between the adhesive
and the concrete was unable to efficiently transfer stress to the 6. Model assessment
concrete. This was deduced from the failure mode in these speci-
mens (i.e. failure at the adhesive-concrete interface) in which the 6.1. Database
FRP was essentially covered by the adhesive, with occasional
patches of concrete. Information on a total of 244 beams and slabs that failed due to
The expression proposed for estimating fracture energy, based IC debonding was gathered to compare the models (Appendix A).
on a linear regression of the experimental results, is shown as Eq. The FRP strengthening consisted of pultruded plates or wet laid-
(25). up laminates. The beams chosen for this study were anchored
0:85 either at one end only or not at all, for the authors deemed that
Gf ¼ 0:0437f c ð25Þ anchoring at both ends could have a substantial effect on the
112 J.C. López-González et al. / Engineering Structures 118 (2016) 108–124
Table 2
Existing fracture energy models.
Reference Gf
smax d0
[16]
2
R1
[11] smax ðd=d0 Þ
0 3 dd
smax du
[18] 2 smax ¼ 1:8bw f ct d0 ¼ 2:5
R1
[17]
max ðd=d0 Þ
2:86
0 s 1:86þðd=d0 Þ
[19] 0:449 0:343
7:554ðGa =t a Þ fc
[14] 0:236
0:514f c
pffiffiffiffiffiffi
[15] 0:308b2w f ct
ultimate beam strength. All the beams were loaded monotonically.
With the exception of eight cases (Zhang et al. [20], specimens A16,
A18, A23, A26, A28, B23, B26 and B28) the strengthening was
bonded into place with no sustained loading. The models were
applied on all specimens of Table A1 appearing in the tables of this
section.
6.2. Analytical method
The parameters used in the comparison were ultimate moment
and a dimensionless parameter defined as l ¼ Mu =bd f c .
Since the comparative study was focused on model goodness, the
strength reduction coefficients for the materials stipulated in the
European code, i.e., cc ¼ 1 and cs ¼ 1, were disregarded. For similar
reasons, the reduction of the FRP contribution to beam strength
specified by the ACI standard (i.e., w ¼ 1) was also omitted.
For the models based on strain or stress increment limitation,
the moment predicted was obtained by applying conventional
force equilibrium analysis and strain compatibility, adopting
the rectangular parabola diagram given in EC-2 [12] for concrete
and a linear diagram for the steel. The initial strain in the
beams strengthened under sustained loading was taken into
consideration.
The mean crack spacing for reinforced concrete elements
strengthened with FRP specified in Ceroni et al. [21] was adopted
for Eqs. (6) and (29) in stress increment calculation. For the former,
however, upper and lower limits of 250 and 150 mm were estab-
lished to comply with JSCE [9] requirements. Furthermore, the
crack spacing specified in EC-2 [12] was adopted in the beams
strengthened under sustained loading. Since the cracking pattern
along the beam is not provided, the critical element between
adjacent cracks was positioned so that the critical crack coincided
at the loading point.
Since the mechanical and/or geometrical properties of the adhe-
sive layer were not reported in a number of cases, the interfacial
fracture energy models that did not include those parameters were
assessed through Eq. (29).
Strictly speaking, when comparing models calibrated with
experimental results, a data set unrelated to the set used to cali-
brate the models studied would have to be applied. Otherwise, a
model calibrated with the same database used in the comparison
would be enormously favoured. One way to avoid that drawback
consists in using the deleted instead of the standard residuals
[22]. Since in the present study the beam results were not used
for calibration, standard residuals could be validly applied in the
comparison. Notwithstanding, it should be stated that some of
the models analysed were calibrated with beams included in the
database used in this study.
Two types of measurements were applied to compare the
models: breakdown of the residual sum of squares and modulus
factors.
Parameters
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
smax ¼ 1:8bw f ct d0 ¼ 0:202bw bw ¼ 1:125 1þbpp=400
2b =bc
h i
3 smax ¼ 3:5f c0:19 d0 ¼ 0:065
2þðd=d0 Þ
  qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ta
þ 50 du ¼ 0:33bw bw ¼
1:5ð2bp =bc Þ
Ea Ec 1þbp =100
h i
smax ¼ 0:19
d0 ¼ 0:051
3:5f c
2:86 dd
qffiffiffiffiffiffiffiffiffiffiffiffiffiffi
smax ¼ 1:5bw f ct bw ¼ 2bp =bc
1þbp =bc
6.2.1. Breakdown of the residual sum of squares
1X 2
^Þ2 þ ðs  s^ Þ2 þ 2ð1  rÞs s^
MPE ¼
y
ei ¼ ðy y y y y ð30Þ
N
where N is the number of observations (i.e. real values), r is the cor-



and y
relation coefficient, y ^ are the means of the experimental and
the predicted values respectively, and sy and sy^ are the variances of
the experimental and the predicted values. The result of Eq. (30) is
the mean predictive error (MPE), in which the first member, the
square of the difference between the means of the real or observed
2 and the predicted values, is here denominated the mean-based pre-
dictive residual (MPR). The second member, the difference between
the variance values, is the variance-based predictive residual (VPR).
The last member is named the incomplete correlation predictive
residual (ICPR). The first residual, which may be positive, negative
or nil, is a measure of the systematic error in the prediction. It
should be noted that MPR provides information on the position of
the centroid of the point cluster on the prediction plot. The second
may also be positive, negative or nil and measures the difference in
variability patterns between the real and predicted values. When
the variances of the predicted and real values are equal, the slope
of the orthogonal regression line is p=4 (i.e. parallel to the bisector
in the prediction plot). The third reflects an imperfect correlation
between predictions and real values, such that r–0. It is always pos-
itive except when r ¼ 1, in which case ICPR is zero. This residual is the
result of the scatter at each side of the orthogonal regression line.
6.2.2. Modulus factors
Although the reduction factors defined in the codes were disre-
garded, three factors were applied to analyse model fit and conser-
vatism in terms of the ratio between the experimental and
forecasted values, y=y ^. The fitting factor (FF) was defined as the
proportion of data conforming to the inequality 0:8 6 y=y ^ 6 1:2;
the safety factor (SF) as the proportion of data conforming to the
inequality y=y^ > 0:9; and the bounded safety factor (BSF) as the
proportion of data conforming to 0:9 6 y=y ^ 6 2.
6.3. Assessment of model goodness
Model behaviour is illustrated in Figs. 6–17, in which the fore-
casted values are shown on the x-axis and the experimental ones
on the y-axis. The hollow circles represent the model predictions
and the solid circles indicate that the ultimate strength was reached
prior to debonding, either due to concrete crushing or FRP rupture.
The average of the experimental to predicted ratio (AVG) as well as
the average absolute error (AAE ¼ jy ^  yj=y) are shown in each figure.
Table 3 gives the breakdown of the residual sum of squares in
mkN and dimensionless values and Table 4 lists the modulus fac-
tors and the coefficients of variation for y=y ^.
The positive value for the MPR residual in most models was a
sign that the point cluster centroid was positioned over the
 J.C. López-González et al. / Engineering Structures 118 (2016) 108–124 113

ACI 440.2R-08 [7] Said and Wu [5]

0.40 0.40

0.35 0.35

0.30 0.30

0.25 0.25

0.20
μue

0.20

μue
0.15 0.15

0.10 0.10

0.05 0.05

0.00 0.00
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
AVG=1.06
μup AVG=1.04
μup
AAE=12% AAE=10%

Fig. 6. Predicted versus experimental value. ACI 440.2R-08 [7] model. Fig. 9. Predicted versus experimental value. Said and Wu [5] model.

JSCE [9] Wu and Niu [10]

0.40 0.40

0.35 0.35

0.30 0.30

0.25 0.25

0.20
μue

0.20
μue

0.15 0.15

0.10 0.10

0.05 0.05

0.00 0.00
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
μup AVG=1.01 μup AVG=1.06
AAE=14% AAE=11%

Fig. 7. Predicted versus experimental value. JSCE [9] model. Fig. 10. Predicted versus experimental value. Wu and Niu [10] model.

fib [6] Eq. (29) - Dai et al. [14]

0.40 0.40

0.35 0.35

0.30 0.30

0.25 0.25

0.20
μue

0.20
μue

0.15 0.15

0.10 0.10

0.05 0.05

0.00 0.00
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
μup AVG=1.53 μup AVG=1.03
AAE=32% AAE=11%

Fig. 8. Predicted versus experimental value. fib [6] model. Fig. 11. Predicted versus experimental value. Eq. (29)-[14].
114 J.C. López-González et al. / Engineering Structures 118 (2016) 108–124

Eq. (29) - Lu et al. [15] Eq. (29) - Neubauer et al. [16]

0.40 0.40

0.35 0.35

0.30 0.30

0.25
0.25
0.20

μue
0.20
μue

0.15
0.15
0.10
0.10
0.05
0.05
0.00
0.00 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 μup AVG=1.19
μup AVG=1.25 AAE=17%
AAE=19%
Fig. 15. Predicted versus experimental value. Eq. (29)-[16].
Fig. 12. Predicted versus experimental value. Eq. (29)-[15].

Eq. (29) - Savoia et al. [17]

Eq. (29) - Brosens and Van Gemert [18] 0.40
0.40
0.35
0.35
0.30
0.30
0.25
0.25
μ ue

0.20
0.20
μue

0.15
0.15
0.10
0.10
0.05
0.05
0.00
0.00 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
AVG=1.15 μup AVG=1.04
μup
AAE=15% AAE=11%

Fig. 13. Predicted versus experimental value. Eq. (29)-[18]. Fig. 16. Predicted versus experimental value. Eq. (29)-[17].

Eq. (29) - Prop.

0.40
Eq. (29) - Nakaba et al. [11]
0.40
0.35
0.35
0.30
0.30
0.25
0.25
μ ue

0.20
μue

0.20
0.15
0.15
0.10
0.10
0.05
0.05
0.00
0.00 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 μ up
μup AVG=1.02 AVG=1.09
AAE=11% AAE=13%

Fig. 14. Predicted versus experimental value. Eq. (29)-[11]. Fig. 17. Predicted versus experimental value. Eq. (29)-Prop.
 J.C. López-González et al. / Engineering Structures 118 (2016) 108–124 115

Table 3
Breakdown of the residual sum of squares.

Models MPE MPR VPR ICPR

2
(mkN) dmsnlss mkN dmsnlss mkN dmsnlss mkN dmsnlss
ACI440.2R-08 [7] 504.76 0.00091 4.94 0.01197 14.30 0.00777 16.61 0.02657
fib [6] 2075.98 0.00539 22.79 0.06452 35.56 0.01607 17.10 0.03113
JSCE [9] 383.70 0.00109 2.22 0.00098 8.33 0.00390 17.59 0.03270
Said and Wu [5] 262.65 0.00078 2.30 0.00712 6.38 0.00534 14.72 0.02653
Wu and Niu [10] 297.02 0.00082 3.53 0.01159 6.80 0.00552 15.44 0.02562
Eq. (29)-[14] 154.53 0.00078 0.84 0.00569 2.01 0.00381 12.24 0.02702
Eq. (29)-[15] 822.76 0.00235 12.63 0.03859 17.30 0.01180 19.08 0.02693
Eq. (29)-[18] 743.21 0.00155 9.44 0.02516 17.41 0.01042 18.74 0.02849
Eq. (29)-[11] 166.45 0.00078 0.25 0.00408 2.47 0.00383 12.66 0.02744
Eq. (29)-[16] 554.55 0.00194 10.18 0.03146 12.75 0.01238 16.98 0.02827
Eq. (29)-[17] 149.19 0.00076 1.40 0.00692 0.63 0.00390 12.12 0.02641
Eq. (29)-Prop. 240.36 0.00103 4.63 0.01532 6.07 0.00559 13.49 0.02763

Table 4
^.
Modulus factors and coefficient of variation for y=y Because of the assumed position of the critical element between
adjacent cracks and the fact that, conceptually, the debonding load
Models FF SF BSF Coefficient of
predicted by Eq. (29) should be lower than the ultimate load of the
variation (%)
beam (i.e. the beam may still bear more loading after the debond-
ACI440.2R-08 [7] 83.61 87.30 87.30 13.63
ing initation), a conservative performance of Eq. (29) is reasonable.
fib [6] 9.84 99.59 88.52 24.74
JSCE [9] 77.05 73.77 73.77 17.53 This behaviour has been exhibited for all interfacial fracture energy
Said and Wu [5] 88.11 84.84 84.84 13.29 models except for those derived from experimental results using
Wu and Niu [10] 82.79 88.52 88.52 13.27 very soft adhesives, which yielded higher values and consequently
Eq. (29)-[14] 87.30 83.61 83.61 14.09 shifted the cluster towards the bisector.
Eq. (29)-[15] 40.57 97.54 97.54 15.04
Recent comparative study of IC debonding models by Elsanad-
Eq. (29)-[18] 56.97 91.39 91.39 15.39
Eq. (29)-[11] 86.89 82.38 82.38 14.23 edy et al. [23] reported different appraisal of the strain-based mod-
Eq. (29)-[16] 52.87 95.08 95.08 15.74 els (i.e. Said and Wu [5], and ACI 440.2R-08 [7]) compared to the
Eq. (29)-[17] 86.07 84.84 84.84 13.95 present study. It should be noted, nonetheless, that experimental
Eq. (29)-Prop. 76.23 88.11 88.11 14.39
databases are quite different, and the comparison shown in the
aforementioned reference was carried out in terms of FRP strain
identity line. The fact that both MPR and residual VPR displayed its values while in this paper it was done using ultimate bending
highest values in the fib model means that the variance was sub- moment. Elsanadedy et al. [23] reported that the three most signif-
stantially lower for the predicted results than for the real ones. icant parameters for predicting IC debonding strain are: axial rigid-
The interfacial fracture energy model proposed by Savoia et al. ity of the FRP composite, yield strain of the internal steel
[17] applied in Eq. (29) presented the smallest absolute variance reinforcement, and the ratio of the width of FRP laminate to the
residual in mkN, while in dimensionless values the same result width of concrete section; further experimental research is needed
was observed in that model, as well as in Nakaba et al. [11] and to take into account the third parameter on the stress increment-
Dai et al. [14] fracture energy models and JSCE model. based model studied in this paper (i.e. Eq. (29)).
When the models were compared on the grounds of ICPR, which In three-point or four-point bending tests, assuming that the
provides information on the scatter on both sides of the orthogonal stress increment in the laminate is proportional to the increase
regression line, similar values were observed in all models, with in the bending moment, factor t can be simplified as follows:
the lowest and highest ones observed for Eq. (29)-[17] and Eq. L c  sr
(29)-[15] in mkN, respectively. t¼ ð31Þ
Lc
Based on MPE, or the sum of squares of the three residuals,
Eq. (29)-[17] yielded the best results in (mkN)2 values, while the where Lc is the shear span and sr is the crack spacing. Consequently,
best performance in dimensionless values was observed for this after substituting Eq. (31) into the Chen et al. [4] formulation, the
model, along with the Said and Wu model, Eq. (29)-[11] and equation for predicting failure in the inter-crack region is:
Eq. (29)-[14]. Finally, the highest value was observed for the fib
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
model. The mean predictive error given by Eq. (29) in mkN2 is u
u 
2Gf
lower than the database-calibrated models (i.e. ACI [7] and Said epd ¼u
t  2  6 epu ð32Þ
and Wu [5]) for Savoia et al. [17], Dai et al. [14], Nakaba et al. Ep tp 1  1  Lsrc
[11] as well as for the proposed fracture energy model.
^ relationship was high in the most
The fitting factor for the y=y Since Eq. (31) does not take into account the contribution of
recent database-calibrated model (Said and Wu [5] model), as it internal steel reinforcement, factor t is higher than the actual
would be expected. Since all the models with the exception of fib value. Hence, the debonding load of Eq. (32) is higher than the
and JSCE exhibited similar coefficients of variation, a comparison one of Eq. (29).
based on that measure of dispersion alone would not suffice for a Table 5 gives the assessment of the fracture energy models
thorough appraisal of the models. All models with the exception through Eq. (32) using the mean crack spacing calculated as in Cer-
of JSCE displayed a safety factor greater than 80, which denotes oni et al. [21]. The existing fracture energy models derived from
their conservative behaviour. The overly conservative performance experimental tests using very soft adhesives are not included in
of the fib model is denoted by the bounded safety factor which is Table 5 because the predominant failure was concrete crushing
lower than the SF. Yet, it should be noted that behaviour of this in the sectional analysis. These fracture energy models were
model can be enhanced through coefficients c1 , c3 and c4 , which assessed through Eq. (32) using the crack spacing calculated as
are related to the stress-slip curve. per EC-2 (Table 6).
116 J.C. López-González et al. / Engineering Structures 118 (2016) 108–124

Table 5
Breakdown of the residual sum of squares of the models assessed through Eq. (32) using the mean crack spacing calculated as in Ceroni and Pecce [21].

Models MPE MPR VPR ICPR

2
(mkN) dmsnlss mkN dmsnlss mkN dmsnlss mkN dmsnlss
Eq. (32)-[15] 169.88 0.00089 3.83 0.01616 2.55 0.00810 12.19 0.02380
Eq. (32)-[18] 166.77 0.00080 0.03 0.00265 1.24 0.00799 12.85 0.02695
Eq. (32)-[16] 198.49 0.00087 1.89 0.00753 4.97 0.00828 13.05 0.02730
Eq. (32)-Prop. 320.25 0.00086 3.29 0.0027 8.52 0.00589 15.39 0.02860

Table 6
Breakdown of the residual sum of squares of the models assessed through Eq. (32) using the crack spacing calculated as per EC-2 [12].

Models MPE MPR VPR ICPR

2
(mkN) dmsnlss mkN dmsnlss mkN dmsnlss mkN dmsnlss
Eq. (32) - [14] 128.65 0.00060 1.64 0.00816 0.55 0.00679 11.21 0.02200
Eq. (32)-[11] 132.46 0.00059 1.05 0.00680 0.66 0.00675 11.44 0.02235
Eq. (32)-[17] 126.50 0.00060 1.98 0.00902 1.06 0.00684 11.02 0.02166
Eq. (32)-Prop. 215.62 0.00078 4.29 0.01461 6.24 0.00705 12.58 0.02277

Based on MPE, the models listed in Table 5, Eq. (32) exhibited Table 7
the best performance for Brosens and Van Gemert [18] model, fol- Predicted results in the example of uniformly loaded beam.
lowed by Lu et al. [15] in mkN2 and the proposed model in dimen- Models Wet laid-up (mkN) Pultruded (mkN)
sionless values. With the exception of the proposed formulation,
ACI440.2R-08 [7] 507 454
the ones listed in Table 5 had better performance in mkN2 and fib [6] 459 420
similar behaviour in the dimensionless parameter as compared Said and Wu [5] 512 468
to strain-based calibrated models. For the models listed in Table 6, Wu and Niu [10] 502 458
the proposed by Savoia et al. [17] shows the lowest value of MPE JSCE [9] 555a,b 540a,b
Eq. (29)-[14] 549 472
in mkN2, followed by Dai et al. [14], Nakaba et al. [11] and the Eq. (29)-[15] 467 449
proposed model. These models performed better than models in Eq. (29)-[18] 501 467
place for both mkN2 and dimensionless values. Lastly, the models Eq. (29)-[11] 554 475
listed in Table 6 exhibited concrete crushing failure in the sectional Eq. (29)-[16] 465 468
Eq. (29)-[17] 546 471
analysis in 33%, 35%, 32% and 23% respectively.
Eq. (29)-Prop. 527 459
In the current study, the calculation process of the sectional
a
analysis assumed that plane sections remain plane after bending Stress increment calculated along 200 mm (average of upper and lower specified
by the JSCE).
(i.e. perfect bond between both the internal steel reinforcement b
Theoretical failure reached in sectional analysis by concrete crushing prior
and external FRP laminate with the adjacent concrete). It is highly debonding.
interesting to examine the behaviour of each model taking into
account that the bond between FRP laminate and the concrete sub-
strate is not rigid, as suggested in Skuturna and Valivonis [24].
Δσp [11]
7. Recommendations for analysing uniformly loaded beams [14] [16]
[15] [17]
The analysis of a uniformly loaded beam entails verification [18] Prop.
along its entire length. The area where debonding is liable to begin 250
is the zone where the stress increment on the laminate is the highest,
i.e., where the inner reinforcement yields. M=527 mkN
The behaviour of Eq. (29) in a uniformly loaded beam was anal- 200
ysed with the example set out in ACI standard 440.2R-08 [7],
applying the two most common strengthening configurations. 150
Δσ (MPa)

The first, described in the example, was two-layer (1.02 mm thick

each) wet laid-up CFRP laminate with a modulus of deformation of
37 GPa. The second was standard market pultruded CFRP plate, 100
100 mm wide and 1.2 mm thick, whose modulus of deformation
and ultimate strain were 165 GPa and 1.5% respectively.
In the present example, the allowable stress increment was 50
obtained considering the crack spacing calculated as EC-2 [12] to
be 276 mm since the external reinforcement is applied under 0
sustained load. The load was varied by trial and error in order to 0 0.5 1 1.5 2 2.5 3 3.5
find the stress increment on the laminate that exceeded the value Distance from left support (m)
calculated with Eq. (29). Results of existing IC models and those
obtained by Eq. (29) are given in Table 7. Figs. 18 and 19 show Fig. 18. Stress increment on wet laid-up laminate attached to the sample beam.
the stress increment on the laminate positioned on the left half
of the beam when the peak value equals Eq. (29) using the pro- Wu model were similar to the values obtained by applying the
posed expression for the interfacial fracture energy (Eq. (25)). strain limitation recommended in the ACI standards. With the
The Wu and Niu model results (e.g. 502 mkN and 458 mkN), exception of fib and JSCE, all models displayed similar predictions
along with the 512 mkN and 468 mkN yielded by the Said and for the beam with pultruded laminate, which means a 5.8%
 J.C. López-González et al. / Engineering Structures 118 (2016) 108–124 117

cracks may cause peak stress increment on the laminate and there-
Δσp [11] fore constitute a potential debonding zone.
[14] [16]
[15] [17] 8. Conclusions
[18] Prop.
In the present paper a new interfacial fracture energy model is pro-
700
posed, characterising the FRP-concrete interface with beam-type tests.
M=459 mkN The existing fracture energy models, as well as the proposed model,
600
were applied to an existing fracture mechanics-based formulation.
500 This formulation applies tensile stress at both ends of the laminate
in order to attain a more realistic forecast of inter-crack debonding fail-
400 ure in reinforced concrete beams strengthened with FRP materials.
Δσ (MPa)

Along with the aforementioned formulation, the predictions set

300
200
100
0 the analysis are listed below.
0 0.5 1 1.5 2 2.5
Distance from left support (m)
Fig. 19. Stress increment on pultruded plate attached to the sample beam.
difference between the lowest and highest ones. For the beam with
wet laid-up laminate, the results exhibited higher variability than
for the beam with pultruded laminate; the difference between
the bound predictions being equal to 19.1%. Lastly, the limitation
proposed by the JSCE was clearly optimistic for the analysed cases.
Due to the complexity associated with stress transfer at the
FRP-concrete interface in cracked beams, calibrated strain-based
models have been largely used because of their simplicity and good
performance. Nevertheless, the stress-increment based model pre-
sents a more realistic approach by taking into account the stresses
in the laminate at two adjacent cracks. This method exhibited good
performance for the point-loaded beams used in the present study
without the need of calibration process. However, compared to
calibrated strain-based models, the stress increment-based formu-
lation (i.e. Eq. (29)) could lead to an optimistic prediction of
debonding moment in uniformly loaded beams if a small crack
spacing is used. For the uniformly loaded beams studied in this
paper, stress increment-based formulation results were higher
than those obtained from strain-based calibrated models for both
the mean and characteristic crack spacing calculated as in Ceroni
and Pecce [21], as it is shown in Table 8.
Since the stress increment on the laminate between two adja-
cent cracks depends on the type of loading and the internal steel
reinforcement, the stress increment-based formulation leads to
higher debonding moment in uniformly loaded beams than in
three or four-point bending tests, in which the maximum shear
force is concomitant with the maximum bending moment. Like-
wise, it should be noted that bars cutoffs between two adjacent
out in the most highly reputed international codes and models
found in the literature were thoroughly compared. The analysis
was conducted by breaking down the residual sum of squares to
identify the origin of the variability between the predicted and real
values. Simple factors were also applied to evaluate the level of
conservatism and goodness of models. The major conclusions of
3 3.5
From the authors’ point of view, stress-increment based models
have received scant attention due to the uncertainty related to the
crack spacing and interfacial fracture energy that should be
applied. The observations of the current study show that the ana-
lytical formulation proposed by Chen et al. [4] (Eq. (29)) calculating
crack spacing as Ceroni and Pecce [21], exhibit less variability than
the methods currently in use for some fracture energy models (i.e.
Savoia et al. [17], Dai et al. [14], Nakaba et al. [11] and the proposed
model), with the advantage of not needing a calibration process.
Unlike strain-based models, Eq. (29) is sensitive to shear-moment
interaction and internal steel reinforcement. Hence, strain-based mod-
els provide a more conservative prediction for uniformly loaded beams
when small crack spacing is adopted for the stress increment-based for-
mulation. In this case, crack spacing calculated as per EC-2 [12] is rec-
ommended for design purposes. Experimental research is needed to
further explicate the debonding process of an FRP-to-concrete bonded
joint where the FRP laminate is subject to tension at both ends.
Based on the observations, from the authors’ point of view, the
test presented in this paper can emerge as an alternative to the
pure shear tests to obtain the interfacial fracture energy. Yet, it
should be noted, that further experimental and numerical research
is still needed to improve the accuracy of the model reported in
this paper and to clarify the influence of normal stresses.
Acknowledgments
This research was funded by the Instituto Técnico de Materiales
y Construcciones (INTEMAC). Special thanks are due to its Central
Laboratory staff for their support in the experimental part of this
study and to BETAZUL for the FRP material used in this research.
Appendix A
See Table A1.

Table 8
Predicted results of Eq. (29) using the mean and the characteristic crack spacing calculated as in Ceroni et al. [21].

Models Wet laid-up Pultruded

Mean crack spacing Characteristic crack spacing Mean crack spacing Characteristic crack spacing
Eq. (29)-[14] 555a 555a 540a 520
Eq. (29)-[15] 548 515 512 487
Eq. (29)-[18] 555a 555a 540a 513
Eq. (29)-[11] 555a 555a 540a 524
Eq. (29)-[16] 546 512 540a 513
Eq. (29)-[17] 555a 555a 540a 518
Eq. (29)-Prop. 555a 555a 531 502
a
Theoretical failure reached in sectional analysis by concrete crushing prior debonding.
Table A1
Experimental database. 118
ID b d fcm fctm As No. fy bp tp Ep epu Lc Mu ACI [7] fib [6] Said and Wu and JSCE [9] Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)-
(mm) (mm) (MPa) (MPa) bars MPa (mm) (mm) (GPa) (%) (m) (mkN) Wu [5] Niu [10] [14] [15] [18] [11] [16] [17] Prop.

Aram et al. [25]

B3 250 120 44.1a 3.28b 3/8 485 50 1.2 214 0.93 0.67 20.94 17.75 7.65 19.36 16.48 17.50 17.89 15.49 19.91 18.10 18.70 17.50 17.11
B4 250 120 46.8a 3.44b 3/8 485 50 1.2 155 1.74 0.67 19.48 16.72 8.30 17.52 15.50 15.52 15.92 14.11 18.12 16.32 16.91 15.71 15.52
Arduini [26]
SM2 320 110 36.00 2.70 2/12 550 300 0.3 235 1.50 0.42 28.14 40.35 16.11 40.91 39.88 36.91 37.50 26.93 27.13 38.12 26.53 36.91 34.11
SM3 320 110 36.00 2.70 2/12 550 300 0.3 235 1.50 0.42 23.10 40.35 16.11 40.91 39.88 36.91 37.50 26.93 27.13 38.12 26.53 36.91 34.11
SM4 320 110 36.00 2.70 2/12 550 300 0.3 235 1.50 0.42 32.76 40.35 16.11 40.91 39.88 36.91 37.50 26.93 27.13 38.12 26.53 36.91 34.11
SM5 320 110 36.00 2.70 2/12 550 140 0.3 235 1.50 0.42 30.45 26.01 11.06 26.30 24.88 23.91 23.10 20.51 21.51 23.30 21.30 22.90 22.10
MM2 160 270 36.00 2.70 2/16 550 150 0.3 235 1.50 0.95 72.20 81.75 64.21 82.32 82.67 77.71 79.35 66.93 69.38 80.15 68.17 78.76 75.57
MM3 160 270 36.00 2.70 2/16 550 150 0.3 235 1.50 0.95 63.65 81.75 64.21 82.32 82.67 77.71 79.35 66.93 69.38 80.15 68.17 78.76 75.57
Beber et al. [27]
VR5 120 214 33.58 2.60b 2/10 565 120 0.44 230 1.48 0.78 40.01 37.34 23.24 39.29 39.83 42.88 40.28 28.47 31.86 41.08 29.25 39.88 36.87
VR6 120 214 33.58 2.60b 2/10 565 120 0.44 230 1.48 0.78 39.38 37.34 23.24 39.29 39.83 42.88 40.28 28.47 31.86 41.08 29.25 39.88 36.87
VR7 120 214 33.58 2.60b 2/10 565 120 0.77 230 1.48 0.78 48.62 43.11 22.06 47.90 45.78 52.06c 49.67 34.66 38.90 50.69 36.05 49.09 45.07
VR8 120 214 33.58 2.60b 2/10 565 120 0.77 230 1.48 0.78 48.55 43.11 22.06 47.90 45.78 52.06c 49.67 34.66 38.90 50.69 36.05 49.09 45.07
VR9 120 214 33.58 2.60b 2/10 565 120 1.1 230 1.48 0.78 50.74 47.48 22.57 54.87 48.82 57.88c 57.49 40.89 44.14 57.88c 42.32 56.48 51.91
VR10 120 214 33.58 2.60b 2/10 565 120 1.1 230 1.48 0.78 53.64 47.48 22.57 54.87 48.82 57.88c 57.49 40.89 44.14 57.88c 42.32 56.48 51.91
Díaz-Heredia [8]
E-1 150 256 26.72 3.00 3/12 509 50 1.2 165 1.70 1.00 55.03 51.38 30.01 55.22 53.81 53.87 52.08 49.47 54.46 52.65 52.46 51.83 49.87
E-2 150 256 26.72 3.00 3/12 509 100 1.2 165 1.70 1.00 65.10 62.72 36.67 69.63 67.77 72.83 69.90 58.24 66.43 71.07 63.44 69.90 63.44
Dong et al. [28]
B3 152.4 253 38.20 2.91b 2/16 410 152 2 73 1.33 0.99 91.07 77.42 53.88 81.92 77.94 94.77c 85.01 57.49 66.26 86.22 61.48 84.03 79.63
Fanning and Kelly [29]
b c
F3 155 203 80.00 4.66 3/12 532 120 1.2 155 1.55 1.10 61.00 75.34 51.38 72.34 68.02 81.01 78.78 55.58 73.31 79.53 68.12 76.78 81.01c
F4 155 203 80.00 4.66b 3/12 532 120 1.2 155 1.55 1.10 65.18 75.34 51.38 72.34 68.02 81.01c 78.78 55.58 73.31 79.53 68.12 76.78 81.01c
Gao et al. [30]
A0 150 162 35.70 2.75b 2/10 531 75 0.22 235 1.79 0.50 20.18 20.25 14.19 20.08 19.74 18.75 18.15 16.35 17.75 18.36 17.35 18.15 17.56
A10 150 162 35.70 2.75b 2/10 531 75 0.22 235 1.79 0.50 19.68 20.25 14.19 20.08 19.74 18.75 18.15 16.35 17.75 18.36 17.35 18.15 17.56
A20 150 162 35.70 2.75b 2/10 531 75 0.22 235 1.79 0.50 21.98 20.25 14.19 20.08 19.74 18.75 18.15 16.35 17.75 18.36 17.35 18.15 17.56
Gao [31]
1N2 150 162 43.10 3.22b 2/10 460 75 0.22 235 1.79 0.50 20.18 19.64 13.43 18.99 18.22 16.24 17.03 15.04 17.03 17.23 16.24 16.83 16.44
3T675-1 150 162 62.10 4.19b 2/10 460 150 0.66 235 1.79 0.50 34.30 45.66 21.96 44.85 35.33 37.44 42.41 26.23 33.42 42.99 30.02 41.23 43.63
3T4100-1 150 162 62.10 4.19b 2/10 460 150 0.44 235 1.79 0.50 32.68 39.69 20.50 37.38 32.33 31.38 35.21 22.20 28.20 35.81 25.41 34.41 36.18
Garden et al. [32]
3U1.0m 100 84 48.60 3.54b 3/6 350 67 0.82 111 1.23 0.34 5.78 7.45 4.20 7.30 6.39 5.39 6.99 4.99 6.99 7.19 6.19 6.79 6.79
4U1.0m 100 84 48.60 3.54b 3/6 350 67 0.82 111 1.23 0.40 6.90 7.45 4.20 7.30 6.39 5.39 7.39 5.19 7.19 7.39 6.59 7.19 7.19
5U1.0m 100 84 48.60 3.54b 3/6 350 67 0.82 111 1.23 0.40 6.92 7.45 4.20 7.30 6.39 5.39 7.39 5.19 7.19 7.39 6.59 7.19 7.19
1U4.5m 145 211 42.30 3.17b 2/12 556 90 1.28 115 1.07 1.53 45.75 43.37 32.72 44.99 44.00 53.13d 51.04 39.45 48.25 51.63 45.42 50.44 48.42
4U 100 84 60.80 4.15b 3/6 350 67 0.82 111 1.23 0.59 9.06 8.18 4.61 7.65 7.41 7.41 8.61 6.21 8.82c 8.81 8.01 8.41 8.81
5U 100 84 60.80 4.15b 3/6 350 67 0.82 111 1.23 0.77 8.75 8.18 5.01 7.65 7.41 8.82c 8.82c 8.82c 8.82c 8.82c 8.81 8.82c 8.82c
Hearing [33]
B120-1.8 200 220 34.50 2.67b 2/10 418 50 1 155 1.55 0.60 29.43 25.24 12.95 27.20 24.07 23.08 23.08 20.72 24.12 23.31 22.72 22.72 21.50
J.C. López-González et al. / Engineering Structures 118 (2016) 108–124

B150-1.5 200 220 34.50 2.67b 2/10 418 50 1 155 1.55 0.60 27.54 25.24 12.95 27.20 24.07 23.08 23.08 20.72 24.12 23.31 22.72 22.72 21.50
B200-1.5 200 220 34.50 2.67b 2/10 418 50 1 155 1.55 0.60 28.44 25.24 12.95 27.20 24.07 23.08 23.08 20.72 24.12 23.31 22.72 22.72 21.50
B200-1.8 200 220 34.50 2.67b 2/10 418 50 1 155 1.55 0.60 27.15 25.24 12.95 27.20 24.07 23.08 23.08 20.72 24.12 23.31 22.72 22.72 21.50
Kim and Sebastian [34]
B5 200 301 45.00 3.80 4/12 504 100 1.2 180 1.78 1.45 102.95 104.15 80.62 109.74 107.06 139.25 116.42 96.34 116.81 118.40 112.19 115.42 112.59
B6 200 301 45.00 3.80 4/12 504 100 1.2 180 1.78 1.45 108.03 104.15 80.62 109.74 107.06 139.25 116.42 96.34 116.81 118.40 112.19 115.42 112.59
Kishi et al. [35]
A200-1 150 210 24.81 1.97b 2/16 378 130 0.138 126.5 1.96 1.05 38.85 34.02c 30.54 34.02c 34.02c 34.02c 34.02c 31.48 32.48 34.02c 31.88 34.02c 33.48
A200-2 150 210 24.81 1.97b 2/16 378 130 0.138 126.5 1.96 1.05 39.90 34.02c 30.54 34.02c 34.02c 34.02c 34.02c 31.48 32.48 34.02c 31.88 34.02c 33.48
A415-1 150 210 24.81 1.97b 2/16 378 130 0.286 126.5 1.96 1.05 43.79 38.07c 31.56 38.07c 38.07c 38.07c 38.07c 33.08 34.67 38.07c 33.89 38.07c 36.69
A623-2 150 210 24.81 1.97b 2/16 378 130 0.429 126.5 1.96 1.05 41.48 41.05 32.37 41.18c 41.18c 41.18c 41.18c 34.49 36.67 41.18c 35.50 41.18c 39.08
A623-1 150 210 24.81 1.97b 2/16 378 130 0.429 126.5 1.96 1.05 42.26 41.05 32.37 41.18c 41.18c 41.18c 41.18c 34.49 36.67 41.18c 35.50 41.18c 39.08
C300-1 150 210 24.81 1.97b 2/16 378 130 0.167 230.5 1.77 1.05 41.58 38.50c 31.56 38.50c 38.50c 38.50c 38.50c 33.28 34.87 38.50c 34.09 38.50c 36.89
C300-2 150 210 24.81 1.97b 2/16 378 130 0.167 230.5 1.77 1.05 39.38 38.50c 31.56 38.50c 38.50c 38.50c 38.50c 33.28 34.87 38.50c 34.09 38.50c 36.89
C445-1 150 210 24.81 1.97b 2/16 378 130 0.248 230.5 1.77 1.05 44.10 41.37 32.56 41.61c 41.61c 41.61c 41.61c 34.87 36.90 41.61c 35.69 41.61c 39.49
C445-2 150 210 24.81 1.97b 2/16 378 130 0.248 230.5 1.77 1.05 43.47 41.37 32.56 41.61c 41.61c 41.61c 41.61c 34.87 36.90 41.61c 35.69 41.61c 39.49
Kishi et al. [36]
A-250-1 150 210 29.60 2.33b 2/16 406 130 0.286 118 1.75 1.05 44.21 41.45c 34.18 41.45c 41.45c 39.76 41.45c 35.55 37.55 41.45c 36.55 41.45c 39.36
A-400-2 150 360 29.60 2.33b 2/16 406 130 0.572 118 1.75 1.05 84.00 81.78 62.51 83.86 81.15 87.31 82.31 65.53 70.67 83.45 68.08 81.68 76.04
Table A1 (continued)
ID b d fcm fctm As No. fy bp tp Ep epu Lc Mu ACI [7] fib [6] Said and Wu and JSCE [9] Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)-
(mm) (mm) (MPa) (MPa) bars MPa (mm) (mm) (GPa) (%) (m) (mkN) Wu [5] Niu [10] [14] [15] [18] [11] [16] [17] Prop.

Kotynia et al. [37]

B-08S 150 261 32.30 2.80 3/12 490 50 1.2 172 1.70 1.40 67.20 53.26 39.99 56.79 55.98 59.08 56.45 50.27 58.44 57.06 55.84 56.05 52.83
B-08M 150 261 37.30 3.50 3/12 490 120 1.4 220 1.24 1.40 98.00 82.10 54.87 92.95 91.36 107.50c 107.50c 77.40 97.24 107.50c 90.46 106.60 100.20
B-083m 150 261 34.40 2.90 3/12 436 150 0.39 230 1.50 1.40 64.40 64.17 46.15 66.23 63.04 75.01c 75.01c 53.16 61.57 75.01c 57.12 75.01c 70.73
Kurihashi et al. [38]
B0-A 150 200 23.93 1.90b 2/13 364 80 0.286 126.5 1.96 1.10 30.86 24.56 19.52 24.87 24.80 25.81 25.21 21.62 23.22 25.61 22.61 25.01 23.02
B40-A 150 200 23.93 1.90b 2/13 364 80 0.286 126.5 1.96 1.10 28.77 24.56 19.52 24.87 24.80 25.81 25.21 21.62 23.22 25.61 22.61 25.01 23.02
B0-C 150 200 23.93 1.90b 2/13 364 80 0.167 230.5 1.77 1.10 30.31 24.75 19.51 25.14 24.79 26.81 25.41 21.81 23.62 25.81 22.82 25.41 23.41
Kurihashi et al. [39]
R7-2 150 210 28.20 2.23b 2/16 378 130 0.572 126.5 1.96 1.45 50.68 44.55 34.18 46.15 45.69 46.51c 46.51c 38.35 42.14 46.51c 40.13 46.51c 45.92
R6-2 150 210 28.20 2.23b 2/16 378 130 0.572 126.5 1.96 1.26 52.04 44.55 34.18 46.15 45.69 46.51c 46.51c 37.73 41.15 46.51c 39.31 46.51c 44.72
R5-2 150 210 28.20 2.23b 2/16 378 130 0.572 126.5 1.96 1.05 48.83 44.55 33.99 46.15 45.69 46.51c 46.51c 36.75 39.94 46.51c 38.14 46.51c 43.34
R4-2 150 210 28.20 2.23b 2/16 378 130 0.572 126.5 1.96 0.84 49.22 44.55 33.80 46.15 44.69 43.73 45.33 35.74 38.72 46.12 37.14 45.13 41.74
R3-2 150 210 28.20 2.23b 2/16 378 130 0.572 126.5 1.96 0.65 50.41 44.55 33.56 46.15 44.69 40.74 43.34 35.56 37.52 44.14 36.31 43.34 40.32
Leung [40] (cited by [10])
B11 300 740 41.50 3.12b 8/20 526 300 0.88 235 1.79 2.40 1221.12 1137.15 1010.32 1176.91 1162.95 1347.9c 1259.10 1024.02 1027.40 1261.65 1020.62 1241.09 1209.52
B12 300 740 41.50 3.12b 8/20 526 300 0.88 235 1.79 2.40 1239.60 1137.15 1010.32 1176.91 1162.95 1347.9c 1259.10 1024.02 1027.40 1261.65 1020.62 1241.09 1209.52
B21 150 370 41.50 3.12b 3/16 535 150 0.44 235 1.79 1.20 164.64 152.98 125.72 154.60 151.50 146.63 154.28 128.02 135.96 155.89 130.66 152.98 149.68
B22 150 370 41.50 3.12b 3/16 535 150 0.44 235 1.79 1.20 163.50 152.98 125.72 154.60 151.50 146.63 154.28 128.02 135.96 155.89 130.66 152.98 149.68
B31 75 170 41.50 3.12b 2/10 599 75 0.22 235 1.79 0.60 19.26 20.76c 17.16 20.76c 20.76c 20.32 20.32 17.53 18.92 20.52 18.12 20.12 19.72
B32 75 170 41.50 3.12b 2/10 599 75 0.22 235 1.79 0.60 19.29 20.76c 17.16 20.76c 20.76c 20.32 20.32 17.53 18.92 20.52 18.12 20.12 19.72
B41 75 185 41.50 3.12b 2/10 599 75 0.22 235 1.79 0.60 20.88 22.17c 18.48 22.17c 22.17c 20.73 21.74 18.74 20.33 21.93 19.53 21.53 21.13
B42 75 185 41.50 3.12b 2/10 599 75 0.22 235 1.79 0.60 22.71 22.17c 18.48 22.17c 22.17c 20.73 21.74 18.74 20.33 21.93 19.53 21.53 21.13
NB1-8 300 740 29.03 2.29b 8/20 519 300 0.88 235 1.79 2.40 1228.80 1044.45 964.38 1100.96 1098.71 1157.0c 1154.94 976.53 976.53 1157.02c 1157.0c 1146.56 1085.05
NB1-16 300 740 29.03 2.29b 8/20 519 300 1.76 235 1.79 2.40 1316.40 1112.45 876.63 1227.60 1231.51 1319.7c 1319.7c 1319.7c 1319.7c 1319.7c 1319.7c 1319.7c 1227.60
NB2-2 150 370 29.03 2.29b 3/16 521 150 0.22 235 1.79 1.20 129.72 127.49c 111.89 127.49c 127.49c 122.92 127.35 113.70 115.96 127.49c 114.51 127.12 122.36
NB2-4 150 370 29.03 2.29b 3/16 521 150 0.44 235 1.79 1.20 143.46 137.72 117.53 142.55c 140.83 137.03 142.02 120.71 123.36 142.55c 121.43 141.76 133.63
NB2-6 150 370 29.03 2.29b 3/16 521 150 0.66 235 1.79 1.20 153.12 144.53 119.91 153.70 150.80 151.90 154.06c 127.51 130.42 154.06c 128.49 154.06 143.94
NB2-8 150 370 29.03 2.29b 3/16 521 150 0.88 235 1.79 1.20 165.54 150.27 112.41 162.90 158.87 163.34c 163.34c 134.63 137.33 163.34c 135.54 163.34c 152.49
NB3-2 75 185 29.03 2.29b 2/10 599 75 0.22 235 1.79 0.60 20.58 19.25c 17.51 19.25c 19.25c 19.25c 19.25c 19.25c 18.81 19.25c 18.21 19.25c 19.21
NB3-4 75 185 29.03 2.29b 2/10 599 75 0.44 235 1.79 0.60 22.32 21.56c 17.52 21.56c 21.56c 21.56c 21.56c 21.56c 20.81 21.56c 20.22 21.56c 21.41
Maalej and Leong [41]
A3 115 120 39.80 3.41 3/10 547 107.8 0.165 235 1.50 0.50 19.38 17.46c 15.86 17.46c 17.46c 17.46c 17.46c 16.33 17.33 17.46c 16.93 17.46c 17.46c
A4 115 120 39.80 3.41 3/10 547 107.8 0.165 235 1.50 0.50 18.88 17.46c 15.86 17.46c 17.46c 17.46c 17.46c 16.33 17.33 17.46c 16.93 17.46c 17.46c
A5 115 120 39.80 3.41 3/10 547 107.8 0.33 235 1.50 0.50 21.85 19.84c 17.27 19.84c 19.84c 19.84c 19.84c 18.13 19.33 19.84c 18.73 19.84c 19.84c
A6 115 120 39.80 3.41 3/10 547 107.8 0.33 235 1.50 0.50 21.45 19.84c 17.27 19.84c 19.84c 19.84c 19.84c 18.13 19.33 19.84c 18.73 19.84c 19.84c
B3 230 240 39.80 3.41 3/20 544 215.6 0.33 235 1.50 1.00 131.75 139.29c 122.52 139.29c 139.29c 137.13 137.94 125.94 128.11 139.13 126.99 137.37 134.32
B4 230 240 39.80 3.41 3/20 544 215.6 0.33 235 1.50 1.00 130.15 139.29c 122.52 139.29c 139.29c 137.13 137.94 125.94 128.11 139.13 126.99 137.37 134.32
B5 230 240 39.80 3.41 3/20 544 215.6 0.66 235 1.50 1.00 147.35 156.98 107.95 158.50c 158.50c 157.12 156.84 140.41 143.02 158.36 141.56 155.44 150.84
B6 230 240 39.80 3.41 3/20 544 215.6 0.66 235 1.50 1.00 142.15 156.98 107.95 158.50c 158.50c 157.12 156.84 140.41 143.02 158.36 141.56 155.44 150.84
C3 368 384 41.00 3.24 3/32 552 368 0.495 235 1.50 1.60 522.32 570.72 477.84 577.66 581.35c 561.46 560.36 517.91 517.91 564.76 516.58 558.13 548.52
C4 368 384 41.00 3.24 3/32 552 368 0.495 235 1.50 1.60 535.44 570.72 477.84 577.66 581.35c 561.46 560.36 517.91 517.91 564.76 516.58 558.13 548.52
C5 368 384 41.00 3.24 3/32 552 368 0.99 235 1.50 1.60 520.08 616.04 390.93 646.14 655.43 661.83c 632.81 580.12 580.12 641.42 577.88 628.92 614.02
Maeda et al. [42]
SP-C 200 165 35.00 2.70b 2/13 360 200 0.167 236 1.75 0.75 29.36 32.25 20.10 31.29 29.98 36.01 33.02 22.21 24.22 33.60 22.81 32.60 30.21
J.C. López-González et al. / Engineering Structures 118 (2016) 108–124

SP-C2 200 165 35.00 2.70b 2/13 360 200 0.334 236 1.75 0.75 40.88 38.90 22.93 40.01 37.97 45.53c 42.61 26.81 30.00 43.43 27.81 42.24 38.82
Matthys [43]
BF2 200 409 36.50 2.80b 4/16 590 100 1.2 159 1.85 1.25 231.25 216.67 162.46 224.60 220.84 216.10 213.22 203.14 208.14 214.67 206.60 211.75 208.14
BF3 200 409 34.90 2.69b 4/16 590 100 1.2 159 1.85 1.25 232.50 214.75 157.08 223.06 219.91 215.31 212.22 202.34 206.95 213.64 205.43 210.79 206.95
BF8 200 409 39.40 2.99b 2/16 590 100 1.2 159 1.85 1.25 139.13 137.01 78.91 144.25 141.75 155.04 136.06 121.51 129.66 137.96 126.09 135.10 129.33
BF9 200 409 33.70 2.61b 2/16 590 100 0.222 233 1.25 1.25 119.75 114.03 97.51 113.94 114.03 108.17 108.36 102.14 105.35 109.12 104.36 107.97 105.35
Mikami et al. [44]
A-140 150 200 23.93 1.90b 2/13 364 80 0.286 126.5 1.96 1.50 30.15 24.56 19.72 24.87 24.80 27.26c 26.61 22.41 24.21 27.01 23.41 26.41 24.01
Niu et al. [45]
c
A1 960 172.7 31.60 2.47 3/19 452 200 1.28 184 1.47 2.10 134.19 102.76 56.00 114.58 115.49 178.39 127.98 108.48 107.55 130.46 109.24 126.66 115.03
A2 960 172.7 33.40 2.59 3/19 452 200 1.21 195 1.27 2.10 136.92 104.37 58.46 115.54 116.61 172.42d 128.99 109.99 109.99 131.51 111.54 127.80 117.22
A3 960 172.7 35.20 2.71 3/19 452 300 1.35 80.4 0.67 2.10 107.84 103.38d 79.31 103.38d 103.38d 103.38d 103.38d 103.38d 101.04 103.38d 103.38d 103.38d 103.38d
A4 960 172.7 34.40 2.66 3/19 452 300 2.55 108.5 1.08 2.10 140.39 130.93 69.65 150.26 152.71 222.92d 185.53 147.79 138.04 189.45 142.06 183.10 168.53
A5 960 172.7 35.90 2.76 3/19 452 200 2.55 108.5 1.08 2.10 112.77 109.61 58.86 122.28 123.71 172.98d 137.63 117.24 118.15 140.22 119.94 136.58 126.20
A6 960 172.7 35.10 2.71 3/19 452 200 1.35 80.4 0.67 2.10 98.39 89.97d 72.99 89.97d 89.97d 89.97d 89.97d 89.66 89.97d 89.97d 89.97d 89.97d 89.97d
B1 960 172.7 35.20 2.71 3/19 452 200 1.28 184 1.47 1.60 114.96 105.81 53.87 116.27 115.80 182.85 119.97 102.96 102.96 121.96 105.33 117.97 109.73
B2 960 172.7 34.50 2.67 3/19 452 300 1.35 80.4 0.67 1.60 90.72 103.31d 73.77 103.31d 103.31d 103.31d 103.31d 99.85 94.45 103.31d 96.60 103.31d 103.31d
B3 960 172.7 34.70 2.68 3/19 452 200 1.35 80.4 0.67 1.60 86.64 89.89d 65.97 89.89d 89.89d 89.89d 89.89d 85.04 85.20 89.89d 86.46 89.89d 89.04
C2 960 176.7 33.30 2.59 7/13 446 200 1.28 184 1.47 2.10 140.49 107.40 74.51 118.69 117.92 185.66c 144.10 121.73 121.73 146.92 125.06 143.25 131.33
C3 960 176.7 34.10 2.64 7/13 446 300 1.35 80.4 0.67 2.10 112.56 106.49d 82.99 106.49d 106.49d 106.49d 106.49d 106.49d 106.49d 106.49d 106.49d 106.49d 106.49d
C4 960 176.7 34.50 2.67 7/13 446 200 1.35 80.4 0.67 2.10 95.03 93.15d 76.78 93.15d 93.15d 93.15d 93.15d 93.15d 93.15d 93.15d 93.15d 93.15d 93.15d
119

(continued on next page)

Table A1 (continued)
ID b d fcm fctm As No. fy bp tp Ep epu Lc Mu ACI [7] fib [6] Said and Wu and JSCE [9] Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)-

120
(mm) (mm) (MPa) (MPa) bars MPa (mm) (mm) (GPa) (%) (m) (mkN) Wu [5] Niu [10] [14] [15] [18] [11] [16] [17] Prop.

Oller [46]
1D2 300 170 35.20 2.76 2/16 550 100 1.4 150 1.67 1.00 55.50 53.69 27.13 57.91 55.83 56.88 54.11 49.30 52.48 55.07 51.12 53.87 50.87
1C1 300 170 35.20 2.76 2/16 550 100 1.4 150 1.67 1.00 52.00 53.69 27.13 57.91 55.83 56.88 54.11 49.30 52.48 55.07 51.12 53.87 50.87
1B1 300 170 35.20 2.76 2/16 550 100 1.4 150 1.67 1.00 50.20 53.69 27.13 57.91 55.83 56.88 54.11 49.30 52.48 55.07 51.12 53.87 50.87
1A 300 170 35.20 2.76 2/16 550 100 1.4 150 1.67 1.00 54.50 53.69 27.13 57.91 55.83 56.88 54.11 49.30 52.48 55.07 51.12 53.87 50.87
2D1 300 170 35.20 2.76 2/20 550 100 1.4 150 1.67 1.00 64.00 70.33 39.38 74.24 72.07 72.12 68.90 66.09 68.08 69.11 67.69 68.68 67.52
2D2 300 170 35.20 2.76 2/20 550 200 1.4 150 1.67 1.00 81.50 87.27 49.25 93.64c 91.03 93.64c 91.33 78.50 79.89 92.68 79.34 90.72 85.55
2C 300 170 35.20 2.76 2/20 550 100 1.4 150 1.67 1.00 71.40 70.33 39.38 74.24 72.07 72.12 68.90 66.09 68.08 69.11 67.69 68.68 67.52
Pham and Al-Mahaidi [47]
S1a 140 224 47.70 3.49b 3/12 551 100 0.352 209 1.87 0.70 52.71 56.32 44.44 55.37 53.78 49.82 51.81 46.19 49.82 52.22 48.21 51.43 51.00
S1b 140 224 47.70 3.49b 3/12 551 100 0.352 209 1.87 0.70 53.21 56.32 44.44 55.37 53.78 49.82 51.81 46.19 49.82 52.22 48.21 51.43 51.00
S2a 140 224 47.70 3.49b 3/12 551 100 0.352 209 1.87 0.70 57.43 56.32 44.44 55.37 53.78 49.82 51.81 46.19 49.82 52.22 48.21 51.43 51.00
S2b 140 224 47.70 3.49b 3/12 551 100 0.352 209 1.87 0.70 53.21 56.32 44.44 55.37 53.78 49.82 51.81 46.19 49.82 52.22 48.21 51.43 51.00
S3a 140 224 47.70 3.49b 2/12 551 100 0.352 209 1.87 0.70 43.07 44.52 32.09 43.48 42.54 38.56 40.35 34.15 38.15 40.75 36.36 39.79 39.38
S3b 140 224 47.70 3.49b 2/12 551 100 0.352 209 1.87 0.70 43.00 44.52 32.09 43.48 42.54 38.56 40.35 34.15 38.15 40.75 36.36 39.79 39.38
Rahimi and Hutchinson [48]
B3 200 120 54.00a 3.40 2/10 575 150 0.4 127 1.21 0.75 20.70 22.96d 15.01 22.37 22.38 22.96d 22.96d 16.61 19.60 22.96d 18.59 22.80 22.96d
B4 200 120 54.00a 3.40 2/10 575 150 0.4 127 1.21 0.75 19.69 22.96d 15.01 22.37 22.38 22.96d 22.96d 16.61 19.60 22.96d 18.59 22.80 22.96d
B5 200 120 54.00a 3.40 2/10 575 150 1.2 127 1.21 0.75 26.14 33.71 17.85 34.06 32.36 38.41 36.22 24.22 29.81 36.80 28.00 35.61 36.22
B6 200 120 54.00a 3.40 2/10 575 150 1.2 127 1.21 0.75 26.10 33.71 17.85 34.06 32.36 38.41 36.22 24.22 29.81 36.80 28.00 35.61 36.22
B7 200 120 54.00a 3.40 2/10 575 150 1.8 36 2.98 0.75 22.16 26.00 15.81 24.35 24.38 26.39 25.41 17.80 21.20 25.80 20.00 24.80 25.20
B8 200 120 54.00a 3.40 2/10 575 150 1.8 36 2.98 0.75 23.10 26.00 15.81 24.35 24.38 26.39 25.41 17.80 21.20 25.80 20.00 24.80 25.20
Reeve [49]
L2 152 229 23.30 1.85b 3/13 429 51 1.4 155 1.80 2.25 49.84 42.40 38.27 46.10 45.87 48.79c 48.79c 43.72 48.33 48.79c 46.14 48.79c 44.94
H1 152 229 23.30 1.85b 3/13 429 25 1.4 155 1.80 2.25 42.39 37.60 35.09 39.66 39.91 40.91 39.52 37.33 39.93 39.93 38.72 39.52 37.52
H2 152 229 23.30 1.85b 3/13 429 51 1.4 155 1.80 2.25 48.98 42.40 38.27 46.10 45.87 48.79c 48.79c 43.72 48.33 48.79c 46.14 48.79c 44.94
H2x1 152 229 23.30 1.85b 3/13 429 51 1.4 155 1.80 2.25 50.78 42.40 38.27 46.10 45.87 48.79c 48.79c 43.72 48.33 48.79c 46.14 48.79c 44.94
H4 152 229 23.30 1.85b 3/13 429 102 1.4 155 1.80 2.25 55.40 51.29 44.06 56.43c 56.43c 56.43c 56.43c 53.35 56.43c 56.43c 56.43c 56.43c 56.43c
Rusinowski and Taljsten [50]
Beam 2 120 135 55.80 3.80 2/10 678 100 1.4 155 1.50 0.95 34.49 33.74 19.55 34.89 34.37 37.93c 37.93c 28.82 34.60 37.93c 31.81 37.39 37.93c
Beam 3 120 135 55.80 3.80 2/10 678 100 1.4 260 0.80 0.95 32.68 39.05 20.92 42.53 37.36 44.35c 44.35c 34.82 42.79 44.35c 39.82 44.35c 44.35c
Beam 4 120 135 55.80 3.80 2/10 678 100 1.4 155 1.50 0.95 32.92 33.74 19.55 34.89 34.37 37.93c 37.93c 28.82 34.60 37.93c 31.81 37.39 37.93c
Beam 6 120 135 55.80 3.80 2/10 678 100 1.4 155 1.50 0.95 33.11 33.74 19.55 34.89 34.37 37.93c 37.93c 28.82 34.60 37.93c 31.81 37.39 37.93c
Beam 7 120 135 40.50 3.60 2/10 678 100 1.4 155 1.50 0.95 27.65 29.45 16.32 31.16c 31.16c 31.16c 31.16c 31.16c 31.16c 31.16c 29.42 31.16c 31.16c
Saadatmanesh and Ehsani [51]
B 205 400 35.00 2.70b 2/25 456 152 6 37 1.07 1.98 247.88 227.97 194.11 243.29 240.23 275.96c 249.69 202.30 217.80 253.45 211.22 247.58 235.79
Seim et al. [52]
b
S11 480 81 33.20 2.58 3/10 462 100 1.19 198 1.20 1.02 20.71 17.85 8.42 20.53 19.62 28.64 23.04 19.04 21.85 23.44 21.25 22.65 20.45
S12 480 81 33.20 2.58b 3/10 462 100 1.19 198 1.20 1.02 21.57 17.85 8.42 20.53 19.62 28.64 23.04 19.04 21.85 23.44 21.25 22.65 20.45
S5 480 81 33.20 2.58b 3/10 462 100 1.19 198 1.20 1.02 21.92 17.85 8.42 20.53 19.62 28.64 23.04 19.04 21.85 23.44 21.25 22.65 20.45
S1m 480 81 33.20 2.58b 3/10 462 100 1.19 198 1.20 1.02 21.26 17.85 8.42 20.53 19.62 28.64 23.04 19.04 21.85 23.44 21.25 22.65 20.45
C12 480 81 33.20 2.58b 3/10 462 480 1.12 63.8 1.20 1.02 41.01 32.59c 19.23 32.59c 32.59c 32.59c 32.59c 28.05 25.05 32.59c 23.65 32.59c 32.59c
C21 480 81 33.20 2.58b 3/10 462 480 1.19 91.5 0.95 1.02 36.18 37.35c 23.23 37.35c 37.35c 37.35c 37.35c 33.85 30.06 37.35c 28.45 37.35c 37.35c
Spadea et al. [53]
A1.1 140 270 30.87a 2.42b 2/16 435 80 1.2 152 1.50 1.80 78.12 62.37 51.77 67.00 65.98 74.49c 71.51 58.66 66.92 72.87 63.71 71.07 66.05
A3.1 140 270 32.04a 2.50b 2/16 435 80 1.2 152 1.50 1.80 67.32 62.95 52.03 67.43 66.14 75.88c 72.27 59.10 67.67 73.45 64.49 71.67 66.87
J.C. López-González et al. / Engineering Structures 118 (2016) 108–124

Takahashi and Sato [54]

F1 200 250 35.80 2.75b 2/19 371 200 0.167 230 1.51 0.70 79.45 75.03 56.13 73.34 71.42 64.43 69.46 57.65 59.43 70.24 58.27 69.04 66.26
F2 200 250 40.20 3.04b 2/19 371 200 0.334 230 1.51 0.70 85.40 87.80 60.79 87.89 82.78 75.93 82.27 63.14 66.13 83.48 64.30 81.66 78.68
F3 200 250 39.00 2.96b 2/19 371 200 0.501 230 1.51 0.70 94.50 94.81 63.79 98.19 89.64 88.78 92.05 67.77 71.02 93.44 69.08 91.06 87.03
F5 200 250 50.30 3.64b 2/19 371 200 0.334 230 1.51 0.70 97.30 93.89 62.73 91.09 84.48 77.50 84.80 64.41 69.37 85.99 66.32 83.71 83.60
F6 200 250 49.50 3.60b 2/19 371 200 0.501 230 1.51 0.70 108.85 102.57 66.59 102.42 92.38 91.59 95.79 69.17 75.41 97.17 71.55 93.94 93.78
Takeo et al. [55]
No.2 160 225 31.30 2.45b 2/13 356 140 0.167 230 1.51 1.00 33.85 34.96 24.34 34.55 33.18 39.23d 36.63 27.03 29.82 37.23 28.44 36.41 33.44
No.3 160 225 31.30 2.45b 2/13 356 140 0.167 230 1.51 0.80 30.68 34.96 24.34 34.55 33.18 36.23 34.82 26.22 28.83 35.42 27.43 34.63 32.02
No.4 160 225 31.30 2.45b 2/13 356 140 0.167 230 1.51 0.70 30.45 34.96 24.11 34.55 33.18 33.21 33.83 25.84 28.23 34.44 27.03 33.63 31.24
No.5 160 225 39.00 2.96b 2/13 356 140 0.167 230 1.51 0.55 36.30 37.13 24.84 35.63 34.42 29.42 33.22 26.02 28.64 33.64 27.43 32.83 31.64
No.6 160 225 39.00 2.96b 2/13 356 140 0.334 230 1.51 1.00 39.30 43.68 27.87 43.90 40.36 57.17c 47.85 32.62 38.67 48.83 36.02 47.23 44.65
No.7 160 225 39.00 2.96b 2/13 356 140 0.501 230 1.51 1.00 42.80 48.61 30.08 50.69 46.41 65.63c 56.05 36.88 44.49 57.04 40.89 55.26 52.04
Tumialan et al. [56]
A1 150 250 51.70 3.72 2/16 428 150 0.165 203 1.67 1.06 36.88 62.33 46.32 58.34 56.15 57.16 58.37 46.96 51.58 58.96 49.57 57.78 57.97
A2 150 250 51.70 3.72 2/16 428 150 0.33 203 1.67 1.06 90.43 71.06 49.54 68.25 65.08 75.19 69.77 51.37 59.02 70.58 55.63 68.80 68.80
A6 150 250 51.70 3.72 2/16 428 75 0.165 203 1.67 1.06 75.11 51.41 42.69 49.31 48.14 46.17 47.37 44.17 47.77 47.37 46.57 46.97 46.97
A7 150 250 51.70 3.72 2/16 428 75 0.33 203 1.67 1.06 91.65 55.73 44.11 54.24 52.13 49.16 51.38 46.36 52.17 51.78 50.38 50.98 51.18
C1 150 250 51.70 3.72 2/16 428 150 0.165 203 1.67 1.06 82.16 62.33 46.32 58.34 56.15 57.16 58.37 46.96 51.58 58.96 49.57 57.78 57.97
Table A1 (continued)
ID b d fcm fctm As No. fy bp tp Ep epu Lc Mu ACI [7] fib [6] Said and Wu and JSCE [9] Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)-
(mm) (mm) (MPa) (MPa) bars MPa (mm) (mm) (GPa) (%) (m) (mkN) Wu [5] Niu [10] [14] [15] [18] [11] [16] [17] Prop.

Wu et al. [57]
RC-1 150 160 30.20 2.37b 2/16 360 140 0.111 230 1.39 0.90 29.25 26.95c 23.02 26.95c 26.95c 26.44 26.95c 24.04 25.24 26.95c 24.44 26.95c 26.83
RC-2 150 160 30.20 2.37b 2/16 360 140 0.111 230 1.39 0.90 31.01 26.95c 23.02 26.95c 26.95c 26.44 26.95c 24.04 25.24 26.95c 24.44 26.95c 26.83
RCS-1 150 160 34.60 2.67b 2/16 360 140 0.222 230 1.39 0.90 33.08 32.93c 24.80 32.93c 32.77 32.78 32.93c 26.59 28.99 32.93c 27.80 32.93c 31.98
Yao et al. [58]
GS1 302 117.9 25.38a 2.01b 2/10 343 89.7 1.27 20.5 1.31 1.00 10.00 10.32 7.27 10.22 10.13 10.42d 10.42d 9.13 9.93 10.42d 9.73 10.42d 9.53
GS2 302.7 112.8 60.93a 4.15b 2/10 343 89.7 1.27 20.5 1.31 1.00 8.95 10.37d 7.79 10.37d 10.37d 10.37d 10.37d 9.98 10.37d 10.37d 10.37d 10.37d 10.37d
CS1 303 115.3 24.12a 1.91b 2/10 343 50 0.165 271 1.37 1.00 8.51 8.90 6.59 9.14 8.98 10.27d 9.18 8.18 8.98 9.18 8.58 8.98 8.18
CS2 301.8 109 68.40a 4.37b 2/10 343 50 0.165 271 1.37 1.00 8.79 10.15d 7.10 9.78 9.79 10.15d 9.59 8.59 10.15d 9.59 10.15d 9.39 9.99
CP1 301.5 117.4 30.42a 2.39b 4/10 343 50 1.2 165 1.70 1.00 19.95 18.37 12.80 20.20 18.89 21.91 20.51 17.91 21.12 20.92 20.11 20.32 18.51
CP2 303.6 111.3 42.39a 3.17b 4/10 343 50 1.2 165 1.70 1.00 17.58 19.45 13.97 20.62 19.45 22.47 20.87 18.47 23.27 21.28 21.87 20.68 19.87
CP3 302.7 108.2 14.22a 1.01b 2/10 343 50 1.2 165 1.70 1.00 13.31 9.70 3.15 12.39 12.42 16.96c 13.04 10.23 10.85 13.43 10.23 13.04 10.04
CP4 304.5 120.2 51.93a 3.74b 2/10 343 50 1.2 165 1.70 1.00 13.54 15.56 9.19 16.22 14.35 21.37 15.99 13.98 18.98 16.37 17.78 15.77 15.99
II-1 303 117.5 28.80a 2.27b 2/10 349 30 0.165 257 1.76 1.00 7.22 8.13 6.72 8.16 8.24 8.24 7.84 7.44 8.04 7.84 7.84 7.84 7.44
II-2 305 116.5 28.44a 2.24b 2/10 332 50 0.165 257 1.76 1.00 8.37 8.98 6.75 9.04 8.89 10.89 8.89 8.09 9.09 9.09 8.69 8.89 8.29
II-3 305 117.5 33.93a 2.63b 2/10 332 70 0.165 257 1.76 1.00 8.87 10.80 7.39 10.61 10.97 13.37d 10.97 9.37 11.18 11.18 10.58 10.97 9.98
II-4 302 118.7 24.66a 1.96b 2/10 332 90 0.165 257 1.76 1.00 10.20 11.17 7.39 11.49 11.97 15.26d 12.38 10.18 11.18 12.78 10.78 12.38 10.78
II-5 320 117.7 23.85a 1.89b 2/10 333 50 0.165 257 1.76 1.00 8.59 8.79 6.58 9.00 8.94 10.95 8.95 7.95 8.75 9.15 8.55 8.95 7.95
II-6 320 119.5 26.28a 2.08b 2/10 338 50 0.165 257 1.76 1.00 7.99 9.17 6.87 9.30 9.14 11.15 9.15 8.35 9.15 9.35 8.75 9.15 8.35
II-8 203.5 114.5 26.82a 2.12b 2/10 364 50 0.165 257 1.76 1.00 8.42 9.20 7.10 9.31 9.19 10.20 9.40 8.39 9.40 9.59 9.00 9.40 8.59
II-9 320 117 24.93a 1.98b 2/10 332 30 0.165 257 1.76 1.00 6.90 7.62 6.35 7.73 7.90 7.90 7.50 7.10 7.50 7.50 7.30 7.30 6.90
III-1 203 121 25.29a 2.01b 2/10 346 50 0.165 257 1.76 1.00 7.52 9.21 6.92 9.37 9.24 10.24 9.64 8.44 9.44 9.64 9.04 9.44 8.64
III-2 199 122.5 23.85a 1.89b 2/10 373 100 0.165 257 1.76 1.00 10.68 12.48 8.58 12.89 12.75 15.10c 14.55 11.15 12.36 14.75 11.96 14.35 12.55
III-4 150.1 122 25.20a 2.00b 2/10 351 50 0.165 257 1.76 1.00 9.18 9.14 7.13 9.29 9.25 10.25 9.65 8.26 9.45 9.85 9.05 9.65 8.65
Zarnic et al. [59]
1 200 270 25.00 1.98b 3/12 450 50 1.2 150 1.60 0.96 56.06 49.26 30.61 53.16 51.46 49.52 49.32 45.89 48.73 49.91 47.12 49.13 46.30
2 800 105 25.00 1.98b 3/12 450 100 1.2 150 1.60 0.96 30.24 24.38 11.53 27.45 26.33 31.36 26.15 22.95 24.35 26.76 23.96 25.96 22.95
Zhang et al. [20]
A10 120 211 20.70a 1.63b 2/12 335 120 0.111 235 1.42 0.75 23.51 22.42 16.72 22.71 22.61 22.63 22.76c 17.84 18.83 22.76c 18.03 22.76c 20.63
A16 120 211 20.70a 1.63b 2/12 335 120 0.111 235 1.42 0.75 23.78 22.17c 16.51 22.17c 21.62 22.17c 21.03 17.43 18.03 21.24 17.63 20.83 19.24
A18 120 211 20.70a 1.63b 2/12 335 120 0.111 235 1.42 0.75 23.93 21.96c 16.52 21.96c 21.62 21.64 21.04 17.44 18.03 21.23 17.63 20.83 19.23
A20 120 211 20.70a 1.63b 2/12 335 120 0.222 235 1.42 0.75 28.43 25.37 17.94 26.95 26.62 27.12c 23.83 18.83 21.25 24.24 19.04 23.63 21.44
A23 120 211 20.70a 1.63b 2/12 335 120 0.222 235 1.42 0.75 29.93 25.38 17.53 26.60c 25.61 26.60c 23.85 18.65 19.64 24.23 18.85 23.63 21.24
A26 120 211 20.70a 1.63b 2/12 335 120 0.222 235 1.42 0.75 28.39 25.41 17.34 26.11c 24.61 26.11c 23.63 18.43 19.44 24.24 18.64 23.63 21.05
A28 120 211 20.70a 1.63b 2/12 335 120 0.222 235 1.42 0.75 27.45 25.42 17.34 25.77c 24.61 25.77c 23.63 18.45 19.44 24.24 18.65 23.63 21.03
B10 120 209 20.70a 1.63b 2/16 335 120 0.111 235 1.42 0.75 30.90 28.92c 26.13 28.92c 28.92c 28.92c 28.92c 26.85 27.65 28.92c 27.06 28.92c 28.65
B20 120 209 20.70a 1.63b 2/16 335 120 0.222 235 1.42 0.75 31.91 32.01c 27.14 32.01c 32.01c 32.01c 32.01c 28.05 29.45 32.01c 28.26 32.01c 30.45
B23 120 209 20.70a 1.63b 2/16 335 120 0.222 235 1.42 0.75 36.98 31.45c 26.92 31.45c 31.45c 31.45c 31.45c 27.86 28.85 31.45c 28.06 31.45c 30.25
B26 120 209 20.70a 1.63b 2/16 335 120 0.222 235 1.42 0.75 35.96 30.86c 26.73 30.86c 30.86c 30.86 30.86c 27.67 28.65 30.86c 27.86 30.86c 30.04
B28 120 209 20.70a 1.63b 2/16 335 120 0.222 235 1.42 0.75 35.06 30.45c 26.73 30.45c 30.45c 30.45c 30.45c 27.66 28.66 30.45c 27.87 30.45c 30.05
Zhang et al. [60]
A-AT 150 210 31.50 2.46b 2/16 407 130 0.378 78.5 2.99 1.05 45.78 41.44c 34.33 41.44c 41.02 40.03 41.44 35.64 37.43 41.44c 36.64 41.23 39.44
A-AK 150 210 31.50 2.46b 2/16 407 130 0.286 118 1.75 1.05 44.42 42.46c 34.51 42.37 42.01 41.03 42.23 35.84 38.04 42.46c 37.04 42.03 40.03
B-AT 150 360 31.50 2.46b 2/16 407 130 0.756 78.5 2.99 1.05 86.42 81.42 62.59 82.46 80.57 84.60 80.66 65.41 70.24 81.80 67.80 80.38 75.27
B-AK 150 360 31.50 2.46b 2/16 407 130 0.572 118 1.75 1.05 84.05 83.03 63.23 84.62 82.49 88.66 83.03 66.26 71.68 84.41 68.82 82.60 77.43
B-C1 150 360 31.50 2.46b 2/16 407 130 0.334 230 1.48 1.05 82.37 84.75 64.06 87.14 84.51 92.58 85.48 67.30 73.10 87.02 70.31 84.99 79.43
J.C. López-González et al. / Engineering Structures 118 (2016) 108–124

Zhang et al. [61]

A-1 150 300 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.30 82.36 68.84 52.71 70.13 67.75 79.39c 72.58 56.64 61.97 73.61 59.22 71.81 67.39
A-2 150 300 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.30 82.55 68.84 52.71 70.13 67.75 79.39c 72.58 56.64 61.97 73.61 59.22 71.81 67.39
A-3 150 300 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.30 82.03 68.84 52.71 70.13 67.75 79.39c 72.58 56.64 61.97 73.61 59.22 71.81 67.39
A-4 150 300 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.30 85.54 68.84 52.71 70.13 67.75 79.39c 72.58 56.64 61.97 73.61 59.22 71.81 67.39
A-5 150 300 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.30 80.80 68.84 52.71 70.13 67.75 79.39c 72.58 56.64 61.97 73.61 59.22 71.81 67.39
A-6 150 300 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.30 80.73 68.84 52.71 70.13 67.75 79.39c 72.58 56.64 61.97 73.61 59.22 71.81 67.39
B-2 150 210 31.50 2.46b 2/16 407 130 0.167 230 1.48 1.15 46.52 43.54c 34.72 43.29 43.01 43.54c 43.54c 36.65 39.04 43.54c 37.85 43.54c 41.43
B-3 150 210 31.50 2.46b 2/16 407 130 0.286 118 1.74 1.15 48.42 42.46c 34.52 42.37 42.02 42.04 42.46c 36.23 38.43 42.46c 37.24 42.46c 40.63
B-4 150 210 31.50 2.46b 2/16 407 130 0.378 79 2.99 1.15 47.21 41.44c 34.33 41.44c 41.02 41.04 41.44c 35.84 37.84 41.44c 36.84 41.44c 39.64
B-6 150 360 31.50 2.46b 2/16 407 130 0.334 230 1.48 1.15 89.87 84.76 63.92 87.15 83.55 98.58 87.27 67.86 73.90 88.45 71.25 86.44 80.45
B-7 150 360 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.15 91.54 83.05 63.24 84.64 81.54 93.59 84.43 66.88 72.62 85.79 69.67 84.01 78.45
B-8 150 360 31.50 2.46b 2/16 407 130 0.756 79 2.99 1.15 89.82 81.42 62.48 82.46 79.51 89.65 81.99 65.63 70.87 83.02 68.23 81.42 76.27
C-1 150 210 31.50 2.46b 2/16 407 130 0.572 118 1.74 0.75 56.21 46.26 35.72 47.09c 46.01 42.05 45.04 37.23 39.85 45.64 38.45 44.84 42.43
C-2 150 210 31.50 2.46b 2/16 407 130 0.572 118 1.74 0.75 56.96 47.39 36.13 48.27 46.99 43.04 46.26 38.04 40.45 46.83 39.06 45.85 43.25
C-4 150 210 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.15 52.04 46.26 35.91 47.09c 46.01 47.09c 47.09c 38.63 41.83 47.09c 40.25 47.09c 45.04
C-5 150 210 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.15 54.28 47.39 36.54 48.27 46.99 49.26c 49.26c 39.23 42.85 49.26c 41.04 49.26c 46.26
C-6 150 205 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.15 55.78 48.06 36.52 49.06 49.18 51.54c 51.04 39.24 43.25 51.54c 41.42 50.81 47.04
C-7 150 210 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.55 53.32 46.26 36.11 47.09c 46.01 47.09c 47.09c 40.05 43.85 47.09c 42.05 47.09c 47.09c
C-8 150 210 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.55 52.70 47.39 36.72 48.27 46.99 49.26c 49.26c 40.84 44.85 49.26c 42.85 49.26c 49.04
C-9 150 205 31.50 2.46b 2/16 407 130 0.572 118 1.74 1.55 54.87 48.06 36.73 49.06 49.18 51.54c 51.54c 41.04 45.63 51.54c 43.43 51.54c 50.23
121

(continued on next page)

Table A1 (continued)
ID b d fcm fctm As No. fy bp tp Ep epu Lc Mu ACI [7] fib [6] Said and Wu and JSCE [9] Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- Eq. (29)- 122
(mm) (mm) (MPa) (MPa) bars MPa (mm) (mm) (GPa) (%) (m) (mkN) Wu [5] Niu [10] [14] [15] [18] [11] [16] [17] Prop.

Zhang et al. [62]

A-1-1 150 210 33.70 2.61b 2/16 365 130 0.572 118 1.75 0.75 56.51 44.37 32.95 44.93 43.31 39.32 43.34 34.74 37.94 43.92 36.34 42.94 40.72
A-1-2 150 210 33.70 2.61b 2/16 365 130 0.572 118 1.75 0.75 57.19 45.56 33.39 46.16 44.28 41.31 44.53 35.13 38.73 45.14 36.92 44.32 41.92
A-2-1 150 210 33.70 2.61b 2/16 365 130 0.572 118 1.75 1.15 51.98 44.37 33.36 44.93 43.31 46.82c 46.82c 36.34 40.33 46.82c 38.53 46.74 43.73
A-2-2 150 210 33.70 2.61b 2/16 365 130 0.572 118 1.75 1.15 54.63 45.56 33.77 46.16 44.28 49.23c 48.54 37.13 41.31 49.23c 39.34 48.14 45.14
A-2-3 150 210 33.70 2.61b 2/16 365 130 0.572 118 1.75 1.15 56.12 47.10 34.58 47.74 46.28 52.63c 50.71 37.94 42.75 51.52 40.52 50.14 46.93
A-3-1 150 210 33.70 2.61b 2/16 365 130 0.572 118 1.75 1.55 53.44 44.37 33.36 44.93 43.31 46.82c 46.82c 38.11 42.54 46.82c 40.52 46.82c 46.54
A-3-2 150 210 33.70 2.61b 2/16 365 130 0.572 118 1.75 1.55 53.52 45.56 34.00 46.16 44.28 49.23c 49.23c 38.93 43.74 49.23c 41.54 49.23c 48.14
A-3-3 150 210 33.70 2.61b 2/16 365 130 0.572 118 1.75 1.55 55.07 47.10 34.80 47.74 46.28 52.63c 52.63c 40.13 45.15 52.63c 42.75 52.63c 50.14
B-1-1 150 210 27.50 2.17b 2/13 376 130 0.286 118 1.75 1.15 39.27 30.88 22.79 30.78 30.53 33.51c 33.14 25.15 27.54 33.51c 26.16 32.94 29.94
B-1-2 150 210 27.50 2.17b 2/16 379 130 0.286 118 1.75 1.15 46.69 39.11c 31.96 39.11c 39.11c 39.11c 39.11c 33.70 35.50 39.11c 34.48 39.11c 37.69
B-1-3 150 210 27.50 2.17b 2/19 383 130 0.286 118 1.75 1.15 57.16 46.53c 42.44 46.53c 46.53c 46.53c 46.53c 43.66 45.07 46.53c 44.28 46.53c 46.53c
B-2-1 150 210 27.50 2.17b 2/13 376 130 0.572 118 1.75 1.15 45.66 35.25 24.78 36.80 35.53 40.76c 40.76c 28.96 32.57 40.76c 30.56 40.56 36.14
B-2-2 150 210 27.50 2.17b 2/16 379 130 0.572 118 1.75 1.15 50.95 43.79 33.76 45.24 44.66 45.27c 45.27c 36.70 39.52 45.27c 38.09 45.27c 42.89
B-2-3 150 210 27.50 2.17b 2/19 383 130 0.572 118 1.75 1.15 66.99 51.32c 44.26 51.32c 51.32c 51.32c 51.32c 46.08 48.49 51.32c 47.08 51.32c 51.06
C-1-1 150 210 29.60 2.33b 2/16 406 130 0.252 78.5 2.99 1.15 47.21 37.96c 33.38 37.96c 37.96c 37.74 37.96c 34.55 35.75 37.96c 35.15 37.96c 37.15
C-1-2 150 210 29.60 2.33b 2/16 406 130 0.286 118 1.75 1.15 48.70 41.45c 34.18 41.45c 41.45c 41.45c 41.45c 35.76 37.75 41.45c 36.76 41.45c 39.76
C-1-3 150 210 29.60 2.33b 2/16 406 130 0.167 230 1.48 1.15 46.46 42.47c 34.57 42.47c 42.47c 42.47c 42.47c 36.15 38.34 42.47c 37.35 42.47c 40.75
C-2-1 150 360 29.60 2.33b 2/16 406 130 0.504 78.5 2.99 1.14 89.58 75.90 60.42 75.71 74.18 76.28 74.69 62.70 66.48 75.65 64.44 74.44 70.26
C-2-2 150 360 29.60 2.33b 2/16 406 130 0.572 118 1.75 1.14 91.52 81.78 62.90 83.86 81.15 92.31 83.66 66.14 71.49 84.89 68.67 83.04 77.26
C-2-3 150 360 29.60 2.33b 2/16 406 130 0.334 230 1.48 1.14 89.75 83.40 63.72 86.21 83.16 97.27 86.33 67.34 73.06 87.71 69.90 85.87 79.15
C-3-2 150 210 27.50 2.17b 2/13 376 130 0.286 118 1.75 1.15 39.27 30.88 22.79 30.78 30.53 33.51c 33.14 25.15 27.54 33.51c 26.16 32.94 29.94
C-3-3 150 210 27.50 2.17b 2/13 376 130 0.43 118 1.75 1.15 43.87 33.27 23.77 34.02 33.52 37.50c 37.34 27.15 30.16 37.50c 28.55 37.14 33.35
C-3-4 150 210 27.50 2.17b 2/13 376 130 0.572 118 1.75 1.15 45.66 35.25 24.78 36.80 35.53 40.76c 40.76c 28.96 32.57 40.76c 30.56 40.56 36.14
C-3-5 150 210 27.50 2.17b 2/13 376 130 0.858 118 1.75 1.15 53.88 38.51 26.61 41.62 40.51 45.97c 45.97c 32.19 36.57 45.97c 33.98 45.97c 41.38

a
Cylinder compressive strength of concrete was converted from cube strength applying a factor that is taken equal to 0.9 when f cm;cub < 60 MPa; 0.95 when 60 MPa 6 f cm;cub < 80 MPa and 1.0 when f cm;cub P 80 MPa.
b 2=3
Tensile strength of concrete was estimated using f ctm ¼ 0:3f ck when f ck 6 50 MPa; otherwise f ctm ¼ 2:12 lnð1 þ f cm =10Þ. Where f ck ¼ f cm  8 MPa.
c
Theoretical failure reached in sectional analysis by concrete crushing prior debonding.
d
Theoretical failure reached in sectional analysis by FRP rupture prior debonding.

r
L

^


^

y
y
y
y
b

h
ei

sr
N

ta
fc

sy^
Ls

sy

tp
Lc

cs
fy
Es

cc
ds

Ly
Ec
bc

Lb
Gf

bL
As

du
d0
Ea

Ep
bp
Ap

f ct

Ga

zm
f ck

Leb
f pu

epd
epu
srm
Mu
f cm

Mcr
f ctm
F max

Drpd
f ct;0:95
Appendix B

f ctm

concrete
beam width

shear length

ultimate slip
crack spacing
J.C. López-González et al. / Engineering Structures 118 (2016) 108–124

bonded length

predicted value
the yield region
equivalent to f cm

total beam depth

cracking moment

ultimate FRP strain

experimental value
steel yield strength
FRP laminate width
effective beam depth

effective bond length

correlation coefficient

FRP laminate thickness

beam bending capacity
number of observations

FRP strain at debonding

clear length of the beam

mean of predicted values

FRP modulus of elasticity
diameter of the steel bars

shear modulus of adhesive

width of the concrete prism

variance of predicted values

mean value of crack spacing
maximum transferable force

FRP-concrete fracture energy

mean of experimental values

slip at maximum shear stress

the Chen et al. [4] formulation

thickness of the adhesive layer

residual of the i-th observation

variance of experimental values

mean lever arm of internal forces

mean tensile strength of concrete

FRP stress increment at debonding

modulus of elasticity of the concrete
modulus of elasticity of the adhesive
The following symbols are used in the paper

material safety factor for the concrete

design value of the FRP tensile strength
cross sectional area of the FRP laminate
total area of longitudinal steel reinforcement

compressive strength of concrete, considered

upper bound, characteristic tensile strength of

characteristic compressive strength of concrete
modulus of elasticity of the steel reinforcement

and load ratio in the Chen et al. [4] formulation

mean cylinder compressive strength of concrete

material safety factor for the steel reinforcement

distance between the critical section and the end of

term which accounts for the effect of bonded length

characteristic softening length of the bonded joint in
tensile strength of concrete, considered equivalent to
 J.C. López-González et al. / Engineering Structures 118 (2016) 108–124 123

rp FRP tensile stress [27] Beber AJ, Campos Filho A, Campagnolo JL. Flexural strengthening of R/C beams
with CFRP sheets. In: Proceedings of the conference structural faults and
rp1 lowest tensile stress on the FRP laminate between two repair.
specific sections [28] Dong Y, Zhao M, Ansari F. Failure characteristics of reinforced concrete beams
rp2 highest tensile stress on the FRP laminate between repaired with CFRP composites. Proceeding of the third international
conference on composites in infrastructure 2002:126–40.
two specific sections [29] Fanning PJ, Kelly O. Ultimate response of RC beams strengthened with CFRP
smax maximum shear stress plates. J Composite Constr 2001;5:122–7.
spm mean bond stress of the external strengthening [30] Gao B, Kim J, Leung CKY. Experimental study on RC beams with FRP strips
bonded with rubber modified resins. Compos Sci Technol 2004;64:2557–64.
ssm mean bond stress of the internal steel reinforcement [31] Gao B. FRP strengthened RC beams: taper design and theoretical analysis. PhD
t ratio between the lowest and the highest FRP force of thesis; 2005.
an FRP-to-concrete bonded joint with the FRP [32] Garden H, Quantrill R, Hollaway L, Thorne A, Parke G. An experimental study of
the anchorage length of carbon fibre composite plates used to strengthen
tensioned at both ends reinforced concrete beams. Constr Build Mater 1998;12:203–19.
[33] Hearing BP, Buyukozturk O. Delamination in reinforced concrete retrofitted
with fiber reinforced plastics; 2000.
[34] Kim D, Sebastian WM. Parametric study of bond failure in concrete beams
References externally strengthened with fibre reinforced polymer plates. Mag Concr Res
2002;54:47–59.
[35] Kishi N, Mikami H, Sato M, Matsuoka K. Bending bond strength of RC beams
[1] Täljsten B. Plate bonding: strengthening of existing concrete structures with
strengthened with FRP sheet. Proc Jpn Concr Inst 1998;20:515–20.
epoxy bonded plates of steel or fibre reinforced plastics; 1994.
[36] Kishi N, Mikami H, Zhang G. Numerical analysis of debonding behavior
[2] Teng JG, Yuan H, Chen JF. FRP-to-concrete interfaces between two adjacent
of FRP sheet for flexural strengthening RC beams. Proc Jpn Soc Civ Eng
cracks: theoretical model for debonding failure. Int J Solids Struct
2003:255–72.
2006;43:5750–78.
[37] Kotynia R, Baky HA, Neale KW, Ebead UA. Flexural strengthening of RC beams
[3] López J, Fernández J, González E. Influencia de la relación tensión-
with externally bonded CFRP systems: test results and 3D nonlinear FE
deslizamiento en el comportamiento de la interfase FRP-Hormigón. In:
analysis. J Compos Constr 2008;12:190–201.
Proceedings of the IX Congreso Nacional de Materiales Compuestos.
[38] Kurihashi Y, Kishi N, Mikami H, Matsuoka K. Flexurally bonding property of
[4] Chen JF, Yuan H, Teng JG. Debonding failure along a softening FRP-to-concrete
FRP sheet on RC beam with two-point loading. Proc Jpn Concr Inst
interface between two adjacent cracks in concrete members. Eng Struct
1999;21:1555–60.
2007;29:259–70.
[39] Kurihashi Y, Kishi N, Mikami H, Matsuoka K. Sheet volume effects on flexural
[5] Said H, Wu Z. Evaluating and proposing models of predicting IC debonding
bonding property of RC beams strengthened with FRP sheet. Proc Jpn Concr
failure. J Compos Constr 2008;12:284–99.
Inst 2000;22:481–6.
[6] FIB C. Externally bonded FRP reinforcement for RC structures. Bulletin 2001;14.
[40] Leung CKY. Personal communication; 2004.
[7] ACI 440.2R-08. 440.2 R-08:Guide for the design and construction of externally
[41] Maalej M, Leong KS. Effect of beam size and FRP thickness on interfacial shear
bonded FRP systems for strengthening concrete structures. Farmington Hills,
stress concentration and failure mode of FRP-strengthened beams. Compos Sci
Michigan: American Concrete Institute; 2009. p. 113.
Technol 2005;65:1148–58.
[8] Heredia D. Influencia de diferentes variables en el comportamiento a flexión de
[42] Maeda T, Komaki H, Tsubouchi K, Murakami K. Strengthening behavior of
secciones de hormigón armado reforzadas con materiales compuestos; 2007.
carbon fiber sheet using flexible layer. Proc Jpn Concr Inst 2001;23:817–22.
[9] JSCE. Recommendations for upgrading of concrete structures with use of
[43] Matthys S. Structural behaviour and design of concrete members strengthened
continuous fiber sheets. JSCE concrete engineering series; 2001.
with externally bonded FRP reinforcement. PhD thesis; 2000.
[10] Wu Z, Niu H. Prediction of crack-induced debonding failure in R/C structures
[44] Mikami H, Kishi N, Sato M, Kurihashi Y. Effect of strengthened area on bending
flexurally strengthened with externally bonded FRP composites. Doboku
capacity of RC beams adhered with FRP sheet. Proc Jpn Concr Inst 1999;21:
Gakkai Ronbunshuu E 2007;63:620–39.
1549–54.
[11] Nakaba K, Kanakubo T, Furuta T, Yoshizawa H. Bond behavior between fiber-
[45] Niu H, Vasquez A, Karbhari VM. Effect of material configuration on
reinforced polymer laminates and concrete. ACI Struct J 2001;98:359–67.
strengthening of concrete slabs by CFRP composites. Compos Part B – Eng
[12] CEN E. Eurocode 2: Design of concrete structures. European Committee for
2006;37:213–26.
Standardization; 2004.
[46] Oller E. Peeling failure in beams strengthened by plate bonding. A design
[13] Popovics S. A numerical approach to the complete stress–strain curve of
proposal. PhD thesis. Directed by Cobo D., Universitat Politecnica de Catalunya;
concrete. Cem Concr Res 1973;3:583–99.
2005.
[14] Dai J, Ueda T, Sato Y. Development of the nonlinear bond stress-slip model of
[47] Pham HB, Al-Mahaidi R. Prediction models for debonding failure loads of
fiber reinforced plastics sheet-concrete interfaces with a simple method. J
carbon fiber reinforced polymer retrofitted reinforced concrete beams. J
Compos Constr 2005;9:52–62.
Compos Constr 2006;10:48–59.
[15] Lu XZ, Teng JG, Ye LP, Jiang JJ. Bond–slip models for FRP sheets/plates bonded
[48] Rahimi H, Hutchinson A. Concrete beams strengthened with externally bonded
to concrete. Eng Struct 2005;27:920–37.
FRP plates. J Composite Constr 2001;5:44–56.
[16] Neubauer U, Rostasy FS. Bond failure of concrete fiber reinforced polymer
[49] Reeve BZ. Effect of adhesive stiffness and CFRP geometry on the behavior of
plates at inclined cracks—experiments and fracture mechanics model. Spec
externally bonded CFRP retrofit measures subject to monotonic loads. Masters
Pub 1999;188:369–82.
thesis; 2005.
[17] Savoia M, Ferracuti B, Mazzotti C. Non linear bond-slip law for FRP-concrete
[50] Rusinowski P, Taljsten B. Intermediate crack induced debonding in concrete
interface. In: Proceedings fibre-reinforcement polymer: reinforcement for
beams strengthened with CFRP plates – an experimental study. Adv Struct Eng
concrete structures, vols. 1 and 2. 2003. p. 163–72.
2009;12:793–806.
[18] Brosens K, Van Gemert D. Anchorage design for externally bonded carbon fiber
[51] Saadatmanesh H, Ehsani MR. RC beams strengthened with GFRP plates I:
reinforced polymer laminates. Fourth international symposium on fiber
experimental study. J Struct Eng 1991;117:3417–33.
reinforced polymer reinforcement for reinforced concrete structures, vol.
[52] Seim W, Horman M, Karbhari V, Seible F. External FRP poststrengthening of
188. p. 635.
scaled concrete slabs. J Compos Constr 2001;5:67–75.
[19] Dai JG, Ueda T. Local bond stress slip relations for FRP sheets-concrete
[53] Spadea G, Swamy RN, Bencardino F. Strength and ductility of RC beams
interfaces. In: Proceedings fibre-reinforcement polymer: reinforcement for
repaired with bonded CFRP laminates. J Bridge Eng 2001;6:349–55.
concrete structures, vols. 1 and 2. 2003. p. 143–52.
[54] Takahashi Y, Sato Y. Flexural behavior of RC beams externally reinforced with
[20] Zhang A, Jin W, Li G. Behavior of preloaded RC beams strengthened with CFRP
carbon fiber sheets. In: Proc., FRPRCS-6—fibre-reinforced polymer
laminates. J Zhejiang Univ – Sci A 2006;7:436–44.
reinforcement for concrete structures. p. 237–46.
[21] Ceroni F, Pecce M. Design provisions for crack spacing and width in RC
[55] Takeo K, Matsushita H, Sagawa Y, Ushigome T. Experiment of RC beam
elements externally bonded with FRP. Compos Part B – Eng 2009;40:17–28.
reinforced with CFRP adhesive method having variety of shear-span ratio. Proc
[22] López-Agüí JC, Fernández-Gómez J, Martínez-Lebrusant R, Rojas-Henao LM,
Jpn Concr Inst 1999;21:205–10.
López-González JC, Sánchez-Amillátegui F, et al. Modelos lineales aplicados al
[56] Tumialan JG, Belarbi A, Nanni A. Reinforced concrete beams strengthened with
hormigón estructural. In: Asociación Científico-Técnica del Hormigón
CFRP composites: failure due to concrete cover delamination. Center for
Estructural (ACHE); 2012.
Infrastructure Engineering Studies; 1999.
[23] Elsanadedy H, Abbas H, Al-Salloum Y, Almusallam T. Prediction of
[57] Wu ZS, Matsuzaki T, Fukuzawa K, Kanda T. Strengthening effects on RC beams
intermediate crack debonding strain of externally bonded FRP laminates in
with externally prestressed carbon fiber sheets. J Mater Concr Struct Pavement
RC beams and one-way slabs. J Compos Constr 2014;18:04014008.
2000;46:153–65.
[24] Skuturna T, Valivonis J. Design method for calculating load-carrying capacity
[58] Yao J, Teng JG, Lam L. Experimental study on intermediate crack debonding in
of reinforced concrete beams strengthened with external FRP. Constr Build
FRP-strengthened RC flexural members. Adv Struct Eng 2005;8:365–96.
Mater 2014;50:577–83.
[59] Zarnic R, Gostic S, Bosiljkov V, Bokan-Bosiljkov V. Improvement of bending
[25] Aram MR, Czaderski C, Motavalli M. Debonding failure modes of flexural FRP-
load-bearing capacity by externally bonded plates. Proc Creating Concr
strengthened RC beams. Compos Part B Eng 2008;39:826–41.
1999:433–42.
[26] Arduini M. Behavior of precracked RC beams strengthened with carbon FRP
sheets. J Composite Constr 1997;1:63.
124 J.C. López-González et al. / Engineering Structures 118 (2016) 108–124

[60] Zhang GF, Kishi N, Mikami H. Influence of material properties of FRPS
on strength of flexural strengthened RC beams. In: Proceedings of
FRPRCS-6. p. 327–36.
[61] Zhang GF, Kishi N, Mikami H, Komuro M. A numerical prediction method for
flexural behavior of RC beams reinforced with FRP sheet. In: Proceedings of the
international symposium on bond behaviour of FRP in structures (BBFS 2005).
p. 215–20.
[62] Zhang GF, Kishi N, Mikami H. Effects of design parameters on flexural behavior
of RC beams reinforced with FRP sheet. In: Proceedings of the 2nd
international congress of the international federation for structural concrete.