Nonlinear Micromechanics-Based Finite Element Analysis of

FRP-Rapped Concrete Columns Subjected to Axial Load

Abdel Baky (Hussien.El.Baky@USherbrooke.ca)

Department of Civil Engineering, Helwan University, Egypt
Demers & Yahiaoui & Neale (Kenneth.Neale@USherbrooke.ca>)
Department of Civil Engineering, University of Sherbrooke, Quebec, Canada




ABSTRACT In this research work, new modifications for the microplane constitutive law are first presented to
precisely simulate concrete behaviour under various levels of lateral confinement pressure. A key feature of the suggested
formula is to make the microplane theory valid to represent concrete behaviour under both low and high lateral confining
pressures. An analysis is then carried out using two different numerical procedures. The first procedure is the use of a
three-dimensional finite element analysis with the modified microplane formulations as a user-supplied subroutine into
the FE commercial software ADINA to model the concrete behaviour under various stress and strain histories. Here, the
sub-objective is to validate the formulated microplane approach that is proposed in the first phase of the research
program. In the second analysis; the confinement behaviour is simulated using an in-house code. The code uses the
proposed formulations for the microplane theory to represent the concrete characteristics to predict the stressstrain
relationships of FRP-wrapped concrete columns up to failure. An accurate equation correlating the axial stiffness of the
FRP laminates and the lateral strain of the concrete columns to the confining pressure is incorporated in the in-house
code. The two analyses give almost the same predictions, with minor discrepancies due to some numerical aspects. An
experimental program consisting of testing thirty eight concrete cylinders under various lateral confinement pressures is
carried out to assess the accuracy of the numerical predictions. In the experimental program, the lateral pressure is
designed to give almost the same lateral confinement behaviour as that resulting from applying FRP sheets. The
numerical predictions are finally compared to experimental results for FRP-wrapped and un-wrapped concrete columns
under various levels of lateral pressure. This paper gives an overview for the whole research program with a special
emphasis on the theoretical part.
KEY WORDS
1 INTRODUCTION
A survey of the literature reveals that nonlinear finite
element analyses of FRP-wrapped concrete columns are
generally restricted to simulate the axial stressaxial
strain and axial stresslateral strain of FRP-wrapped
columns. These analyses adopt two concrete constitutive
laws; namely, nonlinear elastic, and elastoplastic
formulations (Deniaud and Neale 2006; Malver et al
2004; Mirmiran et al. 2000; Parent and Labossire 2000;
Spoelstra and Monti 1999; Toutanji 1999). Both of these
approaches have been used, whether simulating the
hardening behaviour of confined concrete or the
softening branch of unconfined columns. Recently, the
microplane model has come to be a powerful
constitutive model that simulates three-dimensional
concrete behaviours This is attributed to the fact that the
microplane theory is capable of capturing both the
hardening and softening behaviour of the concrete
(Caner, F.C. and Baant 2000, Neale et al. 2008).
However it has been concluded, through several
numerical simulations, that the microplane theory has
inconsistent predictions for the concrete behaviour when
applying low confinement lateral pressures, as is the
case of FRP-wrapped concrete cylinders (Ghazi et al.,
2002). Ghazi et al. (2002) observed that the microstructure
coefficients of the M4 model do not suitably predict the
behaviour of concrete with different uniaxial compressive
strengths and different confinement levels. To provide
better simulations for an axially loaded FRP-wrapped
concrete column using the microplane theory, Ghazi et
al. (2002) proposed a new formula for the adjustable
micromechanics variables (
1
k to
4
k ) and for the fixed
micromechanics constant (
10
c ) as a function of the
level of the confinement lateral pressure (Ghazi et al.,
2002). They introduced the factors affecting the
post-peak steepness and the stress and strain at the peak
of the stressstrain curve of confined concrete as
functions of the concrete strength and confinement
levels to give a better fit between analytical results and
CICE 2010 - The 5th International Conference on FRP Composites in Civil Engineering
September 27-29, 2010, Beijing, China
L. Ye et al. (eds.), Advances in FRP Composites in Civil Engineering
Tsinghua University Press, Beijing and Springer-Verlag Berlin Heidelberg 2011
September 2729, 2010, Beijing, China

627
experimental data. These adjustments in the coefficients
of the M4 model were very successful to show excellent
agreement between numerical predictions and
experimental data of axially loaded FRP-wrapped
columns. However, their new proposed coefficients
restricted the microplane theory to represent the
hardening behaviour of axially loaded concrete cylinders.
With that the microplane with the new Ghazi et al.
formula has a new challenge to capture the softening
behaviour of unconfined concrete cylinders. Furthermore,
these adjustments cannot be looked as a unified
constitutive law that can describe the characteristics of
FRP-wrapped concrete columns under a wide range of
complex stress and strain histories. For example, in the
finite element analysis of eccentrically loaded FRP-
rapped concrete cylinders, the lateral confinement
pressure is not constant for all nodes in the finite
element mesh, depending on their relative horizontal
and vertical locations from the neutral axis. In this paper
and the research work at the University of Sherbrooke,
we aim to develop consistent modifications of the M4
model to successfully simulate the concrete behaviour
under low and high lateral confinement pressures and
maintain the consistency with the existing microplane
model for un-confined concrete.
In this research work, new modifications are first
proposed for the microplane constitutive law to
precisely simulate the concrete behaviour under various
levels of lateral confinement pressure. Then the new
formulations are implemented to represent the stress
train behaviour of FRP-wrapped concrete columns up to
failure using nonlinear finite element analysis. The
existing microplane model M4 is not suitable for
concrete behaviour under low confinement pressures.
The analysis results show that an additional nonlinear
function should be introduced in the existing deviatoric
boundaries of the M4 model. In relation to the dilation
behaviour of the microplane model, the nonlinearity in
the shear boundary of the M4 theory affects the
accuracy of predicting the lateral strain of the concrete.
The proposed new formulations for the microplane
model successfully describe the hardening behaviour of
the concrete subjected to low confinement pressure, and
it is consistent with an existing microplane model to
capture softening responses.
The analyses in this research work were carried out
using two different numerical procedures. The first is
the use of a three-dimensional finite element analysis
with the modified microplane formulations as a
user-supplied subroutine in the finite element package
ADINA. In the second analysis, the confinement
behaviour is simulated using an in-house code. The code
uses the proposed modified formulations for the
microplane model to represent the concrete characteristics
and the arc-length iterative technique for the numerical
implementations to predict the stressstrain relationships
of columns up to failure. An accurate equation
correlating the axial stiffness of the FRP laminates and
the lateral strain of the concrete columns to the confining
pressure was incorporated in the in-house code.
2 XPERIMENTAL PROGRAM
The experimental program consisted of testing three
different specimen groups. The objective of the first
group is to calculate the parameters of the microplane
concrete theory. For that, uniaxial compressive tests
(
2 3
0 o o = = ) and hydrostatic tests (
1 2 3
c c c = = ) were
carried out. The second set of specimens aimed to
validate the accuracy of the numerical results. In this set,
FRP-wrapped and steel-encasted concrete cylinders
were tested. The last experimental group was tested
under applied lateral confinement pressures. The lateral
pressures were applied manually and it was designed to
have approximately the same pressure resulting from the
confining FRP sheets used in the second part of the
experimental program.
The experimental program was carried out on
concrete cylinders of circular cross-section of 152 mm
diameter and 300 mm height. Two different concrete
batches were used in the experimental program with an
average actual cylinder compressive strengths after
28-days of 35 MPa and 45 MPa. The experimental
program was carried out on thirty eight specimens
divided into two main groups according to their uniaxial
compressive strength. Each group included testing three
uniaxially loaded concrete cylinders, six specimens
subjected to three different lateral confinement pressures,
two specimens under hydrostatic compressive stresses,
three steel-encasted concrete cylinders and two FRP-
wrapped concrete cylinders as shown in Table 1.
Table 1 Configuration and characteristics of the tested specimens
Confinement
pressure
level
Hydrostatic
test
Steel-confine
ment
cylinders
FRP-wrapped
cylinders
Unconfined
cylinders
0.1
c
f ' t =1/4 in
0.2
c
f ' t =1/2 in
0.3
c
f '
1 2 3
c c c = =
t =3/4 in
73GPa
FRP
E =
0.154mm t =
Three
tested
specimens

In the third specimen group of the experimental
program, the lateral confinement pressures (first column
in the Table 1) were applied manually using a hydraulic
pump. First the axial displacement was applied on the
specimens until reaching an applied load level
corresponding to 80% of the uniaxial concrete
compressive strength then the applied lateral pressure
was activated. The detailed experimental results will be
published in future manuscripts.
Proceedings of the 5th International Conference on FRP Composites in Civil Engineering

628
3 THEORETICAL STUDY
The M4 Baant formulations for the microplane model
accurately represent the mechanical characteristics of
concrete under high confining states of stress (Caner and
Baant 2000). However, as mentioned in the literature,
various problems occurred in the simulations of concrete
subjected to low confining lateral pressures. The
numerical inconsistency of the Baant formulations
occurs under low lateral confining states of stress, where
the M4 model could not capture the hardening branch of
the uniaxial stressstrain profiles.
We found that this problem arises from the function
of the deviatoric boundaries (
D
F
+
,
D
F

) and the shear

boundary
T
F . The deviatoric boundaries are used to
characterize the behaviour of lateral cracks normal to
the microplane, while the shear boundary physically
represents the friction between these cracks. The key
problem is the degree of the nonlinearity of these
boundaries function and their dependency on the low
lateral confining pressure and the associated volumetric
stresses. To overcome this problem and to introduce
unified formulations that are able to simulate concrete
behaviour under various states of stress, we multiplied
these boundaries by a nonlinear function ( )
v
f o . This
function was expressed as a function of the volumetric
stress
v
o and the material adjustable parameters (
1
k
to
4
k ). These modifications are:
3.1 Deviatoric boundaries (spreading, splitting):
( )
( ) ( )
1 5
2
5 6 1 1 18 7
0 :
( )
1 / 1
b
D D D D
v
D v
F
Ek c
f
c c k k c c g
o o c
o
c o
+
> = =
( + < > +

(1)
where
1
1
1
0.1 3.755 1
1
3.755
( ) 1
3.755
2.75
e
2.75
v
v v
v
v
k E v
k E
f
k E
k E
k E
o
o o
o
o
o
< + >
< > < + >
= +
< >
(2)
and
( )
1
1 2.75
v
v v
v
g k E
o
o o
o
< >
= + < > (3)
( ) ( )
1 8
2
8 9 1 1 7
For 0 : ( )
1 / 1
b
D D D D
D v
F
Ek c
c c k k c g
o o c
c o

s =
=
( + < > +

(4)
3.2 Frictional yield surface (shear boundary):
( ) ( )
( ) ( )
0
1 2 10 1
0
1 2 10 1
0 1 11
12
( )
1
,
1
1
b
T T N
T V N N
T V N N
T
N
v
F
E k k c f
E k k c f
E k c
c
o o
o o o
o o o
o
c
=
+ < + >
=
+ + < + >
=
+ < >
(5)
where:
( )
1 1 1
0.37/(3.755 ) 3.755
v
v v
v
f k E k E
o
o o
o
< >
= < + >

(6)
3.3 Numerical implementations using the new
formulation
Four-node tetrahedral solid elements with three
translational degrees of freedom per node were used to
represent the concrete, FRP laminate and adhesive
layers in the finite element simulations. Due to the
geometrical and loading symmetries, only one eighth of
the FRP-wrapped column was modelled. Symmetrical
boundary conditions were placed along the planes of
symmetry. A displacement-controlled numerical procedure
was used in the finite element analysis to capture the
softening branch of the loaddisplacement curve and the
post-debonding behaviour. The finite element mesh of
one eighth of a concrete cylinder is shown in Figure 1.
All details concerning the in-house code and the related
formulations will be presented in future publications.

Figure 1 Finite element mesh
The comparison between experimental results and
numerical predictions is depicted in Figures 2 and 3 for
the application of one and six CFRP sheets, respectively.
The experimental results were taken from the research
work of Lahlou et al. (1992). From Figures 2 and 3, it
was concluded that the microplane approach with the
new modifications accurately simulate the stressstrain
relationships of concrete under both low (one CFRP
sheet) and high (six CFRP sheets) lateral confinement
0
10
20
30
40
50
60
70
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0
Strain %
A
x
i
a
l

s
t
r
e
s
s

(
M
P
a
)
Exp.
M4 (before modification)
M4 (proposed

Figure 2 Comparisons between experimental results and numerical
predictions using one sheet of CFRP
September 2729, 2010, Beijing, China

629
0
30
60
90
120
150
0 1 2 3 4 5
Strain (%)
A
x
i
a
l

s
t
r
e
s
s

(
M
P
a
)
Exp.
M4 (proposed
modification)

Figure 3 Comparisons between experimental results and numerical
predictions using six sheets of CFRP
pressures. The M4 model of Bazant without including
the proposed modification has an obvious difficulty to
simulate the hardening behaviour of CFRP-wrapped
concrete cylinders with one sheet (Figure 2).
4 CONCLUSION
In part of this study, an experimental program has been
carried out to investigate the behaviour of concrete
cylinders subjected to low confining lateral pressures.
The lateral pressures were designed to give the same
lateral and axial responses resulting from using FRP
wraps. In another part of the paper, nonlinear
micromechanics-based finite element analysis have been
developed using the microplane concrete theory to
simulate the behaviour of FRP-wrapped concrete
columns. The analysis was carried out based on our
proposed modifications of the original M4 model.
Nonlinear functions were multiplied to the deviatoric
and shear boundaries functions to overcome the numerical
problems encountered when using the old formulations.
In the application of FRP-wrapped columns, the finite
element analysis with the microplane concrete theory
after considering the proposed modifications was shown
to accurately represent the stressstrain relationships
of the concrete under both low and high lateral
confinement pressures. The key advantage of the
proposed formulations is the ability to adequately
represent the concrete behaviour under low lateral
confinement pressure and maintaining the consistency
with the existing microplane model for un-confined
concrete.
REFERENCES
Caner F. & Baant Z. 2000. Microplane model M4 for concrete. II:
Algorithm and calibration. Journal of Engineering Mechanics,
ASCE, 126(9), 954-961.
Deniaud C. & Neale K.W. 2006. An assessment of constitutive
models for concrete columns confined with fibre composite
sheets. Composite Structures, 73(3), 318-330.
Ghazi M. & Attard M. & Foster S. 2002. Modelling triaxial
compression using the microplane formulation for low
confinement. Computers and Structures, 919-934.
Lahlou K. & Aitcin P. & Chaallal C. 1992. Behaviour of high
strength concrete under confined stresses. Cement and Concrete
Composites, 14, 185-193.
Malver J. & Morrill K. & Crawford J. 2004. Numerical modelling
of concrete confined by fiber-reinforced composites. Journal
of Composites for Construction, ASCE, 8(4), 315-322.
Mirmiran A. & Zagers K. & Yuan W. 2000. Nonlinear finite
element modelling of concrete confined by composites, Finite
Element Analysis and Design, 35(1), 79-96.
Neale K.W. & Abdel Baky H. & Yahiaoui A. & Ebead U. 2008.
Micromechanical modelling of FRP-strengthened concrete
structures. Sixth International Conference on Analytical Models
and New Concepts in Concrete and Masonry Structures
(AMCM), Kotynia, R. & Gawin, D. (Eds.) Lodz, Poland,
109-124.
Parent S. & Labossire P. 2000. Finite element analysis of
reinforced concrete columns confined with composite materials.
Canadian Journal of Civil Engineering, 27(3), 400411.
Spoelstra M. & Monti G. 1999. FRP-confined concrete model.
Journal of Composites for Construction, ASCE, 3(3), 143150.
Toutanji H. 1999. Stressstrain characteristics of concrete columns
externally confined with advanced fibre composite sheets.
ACI Materials Journal, 96(3), 397-404.