The International Journal of Advanced Manufacturing Technology (2019) 103:4671–4680

https://doi.org/10.1007/s00170-019-03838-4

ORIGINAL ARTICLE

Finite element analysis of mechanical response of cellulosic

fiber-reinforced composites
Yosra Glouia 1 & Yassine Chaabouni 1 & Asma El Oudiani 1 & Imen Maatoug 1 & Slah Msahli 1

Received: 9 February 2019 / Accepted: 30 April 2019 / Published online: 4 June 2019
# Springer-Verlag London Ltd., part of Springer Nature 2019

Abstract
An investigation is carried out on natural cellulosic fibers which have gained interest in the composite field due to their superior
specific properties as well as their eco-friendly environment character and biodegradability. A multi-scale finite element (FE)
model of natural fiber composite materials is developed. The study is performed for multiple fibers in hexagonal packings with
10, 20, and 30% of fiber mass fractions. The mechanical behavior of the composite, processed by means of hand lay-up method
and examined thanks to scanning electron microscopy (SEM), is simulated at macro-scale as well as mesoscopic scale. In
particular, the response to tensile and three-point bending test is studied. Linear material properties are obtained by using pure
strain assumptions in the implicit analysis of the composite, while the non-linear behavior and viscoelastic parameters require the
explicit dynamic analysis. Simulation is performed thanks to ABAQUS finite element software. Comparison of experimental and
FEM tensile and three-point bending strength shows very good agreement. From the results, it has been found that the prepared
natural fiber composite materials can be used for structural engineering applications.

Keywords Natural fibers . Mechanical behavior . Finite element analysis . Viscoelasticity . Plasticity . ABAQUS

1 Introduction materials, and wonder materials are becoming an essential

part of today’s materials due to their advantages, such as
History is often marked by the materials and technology low weight, corrosion resistance, and high fatigue strength
that reflect human capability. There have been tremendous [2]. They are extensively used as materials in making
strides in engineering materials since the Second World aircraft structures, automotive industry, electronic, pack-
War [1]. Metallurgists from the aerospace and nuclear aging, medical equipment, space vehicle, and home build-
industries have developed a large range of super alloy ing [3]. Despite their interesting performance, these mate-
and heat-resistant material mnemonics, like ceramics and rials are polluting and harmful to nature. Furthermore,
composite materials. Composites are multi-phase considering the increase in environmental constraints,
such as the Kyoto protocol, it was necessary to think of
eco-friendly materials as substituents, for instance natural
* Yosra Glouia
fibers [4].
[email protected]
Natural fibers made up of polymer composites are gaining
Yassine Chaabouni more interest in engineering applications due to their availabil-
[email protected] ity and ease of manufacturability compared to the other fiber
Asma El Oudiani sources. They also provide appreciable structural properties at
[email protected] a relatively low cost, which finds its attraction towards the
field of research. Eco-friendliness character, which can be
Imen Maatoug
[email protected] easily deposited by the end of their lifecycle, also adds upon
its highlights. Some of the natural fibers used for composite
Slah Msahli
[email protected]
manufacturing are jute, flax, banana, coir, sisal, kenaf, Typha,
luffa, alfa, and hemp [5–10]. Currently, natural fiber compos-
1
Laboratory of Textile Engineering, University of Monastir, ite materials are applied in automobile and packing industries
5070 Ksar Hellal, Tunisia used for low load carrying applications [1, 10].

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

4672 Int J Adv Manuf Technol (2019) 103:4671–4680

Table 1 NaOH-treated fibers’ characteristics

Count (tex) Strain (%) Strength (N) Stress (MPa) Initial modulus (GPa)

Mean value 19.20 33.42 6.21 408.27 5.91

cv% 26.25 18.68 26.53 26.53 28.18
IC 1.83 2.27 0.60 39.43 0.61
LPE 95% 9.56 6.80 9.66 9.66 10.26

Composite microstructures are determined by the element and elements of the set of solutions to solve the
physical and mechanical properties of the individual ma- observed problems. Literature reveals that few researchers
terials. Some analytical and numerical techniques have have been interested to viscoelastoplastic behavior law of
been used for prediction and characterization of com- cellulosic fiber-reinforced composites in their finite ele-
posite behavior. Analytical methods provide reasonable ment modeling and simulation studies. They worked rath-
prediction for relatively simple configurations of the er with the assumption of perfectly elastic material, which
phases. Complicated geometries, loading conditions, is not really the case, because natural cellulosic textile
and material properties often do not yield analytical so- fibers possess viscoelastic properties. This necessitates
lutions, due to complexity and the number of equations. carrying out a research work on mechanical response
To determine the macroscopic overall characteristics of using a numerical study taking into account the specificity
heterogeneous media is an essential problem in many of these textile structures.
engineering applications. In this context, the present work aims to simulate the
From the time and cost viewpoints, performing mechanical behavior of natural cellulosic fiber-reinforced
straightforward experimental measurements on a number composites, made by hand layup technique, using finite
of material samples, for various phase properties, volume element modeling method. For this purpose, a macro-
fractions, and loading histories, is a hardly feasible work. scale and a mesoscopic-scale models were implemented
In this case, the numerical methods are used for ap- in a finite element (FEM) code.
proximate solutions, but they still make some simplifying
assumptions about the inherent microstructures of hetero-
geneous multiphase materials, one such method is finite
element analysis. 2 Materials and methods
In previous decades, the finite element method (FEM)
has become the widely accepted numerical method, not 2.1 Natural fiber-reinforced composite manufacturing
only in computational mechanics, but also in many engi-
neering disciplines. The methodology developed in this Two biocomposites were investigated in this work. One
method enables solving simple linear problems in me- was fabricated with sisal fibers (Agave americana L.)
chanics of solid bodies by demanding non-linear problems and thermoset matrix phase, epoxy resin. While, in the
in almost all fields of applied physics, and more recently, second one, an unsaturated polyester matrix is added to
and many other branches of science. Software packages these fibers. Agave americana L. fibers were extracted
used are ABAQUS, ANSYS, ADINA, NASTRAN, and by hydrolyzing leaves at high temperature and then ap-
LUSAS. Finite element method gives an approximate dis- plying a mechanical action. In order to promote com-
tribution of the required size of the observed area, which patibility and ensure a good adhesion between the two
cannot be determined analytically. Approximate solution phases, a chemical treatment of fibrous phase is per-
is achieved by dividing the area (bodies) to the elements. formed by means of 2% of NaOH solution at 60 °C
Then, the corresponding physical laws are applied to each for 120 min.

Table 2 Comparison of characteristics of treated and untreated fibers

Agave fibers Count (tex) Strength(N) Strain (%) Stress (MPa) Initial modulus (GPa)

CV% CV% CV% CV% CV%

Untreated fibers 20.7 29.4 5.1 23.1 31.1 15.2 336 23.1 8.30 17.7
NaOH-treated fibers 19.2 26.3 6.2 26.5 33.4 18.7 408 26.5 5.9 28.9

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Int J Adv Manuf Technol (2019) 103:4671–4680 4673

Fig. 1 Assembly of natural fibers using a metallic structure

The fibers treated with NaOH are then titrated and mechan-
ically tested. The study is carried out on 30 fibers. Table 1
summarizes these results.
Table 2 gives a comparison between the treated and un-
Fig. 3 Distribution of Agave americana L. fibers in the composite
treated fibers.
The average count of treated fibers is 19.2 tex. It is slightly
been mentioned that the reinforcement of the composites
lower than that of the untreated fibers (21.2 tex). This decrease
by Agave americana L. fibers leads to the increase in the
in linear density is predictable because of the delignification of
Young’s modulus in a significant way. Moreover, the use
the fiber by the NaOH solution.
of Agave fibers increases the bending elasticity of the
Strain and maximum stress have, contrariwise, increased.
composite [7].
The maximum stress supported by the fiber increases from
Figure 3 shows that the technical fiber treated with alkaline
about 330 MPa for the untreated fibers to 408.3 MPa for the
solution (NaOH) is well-impregnated in the matrix. The matrix
treated fibers. This increase can be explained by
is clearly visible inside the helices formed by ultimate fibers.
delignification which acts by double action: on the one hand,
We resume in Table 3 the elastic parameters (Young’s mod-
it gives the fiber a greater fineness, and, on the other hand, it
ulus and Poisson’s coefficient) as well as the tensile mechan-
releases the helical structure of the ultimate fibers, thus
ical characteristics (resistance, elongation) of the performed
allowing the technical fiber a greater lengthening before
composites with diverse ratios of fibers.
breaking.
The study of the tensile behavior of unidirectional Agave/
Treated fibers were then dried at 90 °C for 4 h. Resulting
polyester plies as a function of the mass fraction of fibers
fibers are very interesting, thanks to their low density (1.36)
shows the following:
and high strain (49%) compared with the synthetic fibers and
other natural fibers [7, 11]. For these reasons, these fibers are
– An increase in elongation at break of around 400%.
very interesting and promoting for academic as well as indus-
– An increase in the tensile strength of the order of 160%
trial applications. According to El Oudiani’s rheological
compared to that of the matrix.
study, Agave americana L. fibers are viscoelastic [12].
– An almost linear increase in the Young’s modulus of the
A composite lamina reinforced with Agave americana L.
order of 20%.
fibers was manufactured manually by the use of a square
metallic frame to align the fibers, as shown in Fig. 1 [7].
The elongation at break of the polymerized polyester
The obtained composite is shown in Fig. 2.
matrix is 1.6%. This value increases almost linearly with
The manufactured composites are examined by means
of scanning electron microscopy (SEM), and the result is
shown in Fig. 3. Fibers are arranged in hexagonal packing Table 3 Elastic parameters of Agave-based composites
array. This will be useful in the definition of FE model’s
Matrix Agave Initial Poisson’s Stress Strain
geometry. Tensile test and three-point bending test were fraction (%) modulus coefficient (Mpa) (%)
carried out on Agave fiber-reinforced composites. It has (Gpa) (ν)

Polyester 0 0.87 0.32 12 1.6

15 1.48 0.32 28 4.8
24 1.07 0.32 31 9
31 1.164 0.32 33 8.5
Epoxide 0 1.9 0.333 30 1.9
10 2.34 0.333 48 3.2
25 2.317 0.333 24 1.1
30 2.307 0.333 22 1.1
Fig. 2 Composite sample for tensile test

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

4674 Int J Adv Manuf Technol (2019) 103:4671–4680

Table 4 Bending properties of Agave-based composites The study of the unidirectional Agave/epoxy resin tensile
Matrix Polyester Epoxide behavior shows an almost linear increase in Young’s modulus
of about 20% and a decrease in elongation at break. This
Agave fraction (%) 0 15 24 31 0 10 20 30 makes it possible to affirm that Agave fiber makes it possible
MOE (GPa) 1.1 1.6 2.5 2.6 1.9 2.4 2.5 2.8 to reinforce epoxy resin composites. This result confirms the
one obtained by Ben Cheikh [13] performed on this type of
applications.
the weight ratio to reach 8.5% (corresponding to a weight Moreover, we resume in the following table the bending
ratio of 31%). Due to its particular microstructure, the properties of Agave-based composites, in particular the elastic
mechanical properties of the technical Agave fiber give bending modulus (MOE) relative to each weight ratio (Table 4).
to the obtained composite the possibility of elongating The modulus measured in bending increases substan-
in relatively high proportions. Indeed, the use of natural tially in an almost linear manner from a zero weight ratio
fibers as reinforcement of polyester composite generally (polyester matrix) to a load ratio of 30%. It goes from
improves the elongation at break. 1100 to 2600 MPa. The reinforcement of the composite

Fig. 4 Comparison between

experimental and numerical
stress–strain response under 35
tensile test of 15% Agave 30
polyester composite (a), 24%
25
Agave polyester composite (b),
Stress (MPa)

and 30% Agave epoxide 20

composite under tensile test (c), 15
respectively Experimental curve
10
Numerical curve
5
0
0 0.5 1 1.5 2 2.5 3

(a) Strain (%)

35

30

25
Stress (MPa)

20

15
Experimental
10 curve
Numerical curve
5

0
0 1 2 3 4 5 6 7 8 9
(b) Strain (%)

35

30

25
Stress (MPa)

20

15

10 Experimental curve
Numerical curve
5

0
0 1 2 3 4 5 6
(c) Strain (%)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Int J Adv Manuf Technol (2019) 103:4671–4680 4675

the detailed investigation of the complex phenomena. The

mesoscopic heterogeneous model [10, 14–20] instead repro-
duces every single fiber and the matrix, as well as the interac-
tion among these. This approach provides a more complete
awareness of the phenomena, as proven in the comparison
with the homogeneous approach in.
Using the finite element analysis, we have tried in this work
to model the mechanical behavior of Agave americana L.
fibers reinforcing polyester and epoxide-based matrix, respec-
tively. Our study contains two different scales, macro- and
Fig. 5 Agave-reinforced composite bending behavior meso-scale models, because of the heterogeneity in composite
materials when analyzing them at numerous scales.
by the Agave fibers increases the bending elasticity of the
polyester-based composite.
The modulus measured in bending for the case of compos-
ites with the epoxy matrix increases considerably in a quasi- 3 Results and discussion
linear way by passing from a weight ratio of 10 to 30%. It goes
from 2.4 to 2.8 GPa. The strengthening of the composite by 3.1 Macro-scale modeling, tensile test simulation
the Agave fibers increases the bending elasticity of the
epoxide-based composite. Unidirectional Agave-reinforced composite is considered as a
The results of the tensile and three-point bending tests re- three-dimensional plate (rectangular shape) when analyzing it
veal the interest and importance of Agave fibers in the com- at macro-scale. Dimensions of the plate are 100 ∗ 20 ∗ 2 mm3
posite application. It is an original fiber; resistant, light, and as mentioned in Fig. 5.
biodegradable. The global material has an elastoplastic behavior referring
to the experimental results [7]. The elastic properties are
shown in Table 1.
2.2 Finite element modeling
Concerning plastic behavior, we used experimental data
and we calculate true stress and true strain using the following
The choice of the length scale, together with the constitutive
formulas:
material models and the related parameters, is a key point,
especially in the fields of composites, which are made of dif- σtrue ¼ σnom  ð1 þ εnom Þ ð1Þ
ferent materials. Several approaches were used and developed
εtrue ¼ ln ð1 þ εnom Þ ð2Þ
during the last years and are reported in the literature. In par-
ticular, two different techniques of modeling have been εplastic ¼ εtotal true –εElastic true ð3Þ
employed in the literature; one technique does not reproduce εElastic true ¼ σtrue =E ð4Þ
all the fibers and the matrix but considers the layer as a unique
homogenous mean. This approach most likely does not allow where

Fig. 6 Curve bending tests three 400

points of the polyester matrix
(PES) and PES-reinforced Agave 350
fiber composites with different
300
ratios
250
Load (N)

200

150 EPX matrix

EPX 10
100
EPX 20
50 EPX 30

0
0 1 2 3 4 5 6 7
Deflection (mm)

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

4676 Int J Adv Manuf Technol (2019) 103:4671–4680

Fig. 7 Curve bending tests three 350

points of the epoxide matrix PES 15 agave
(EPX) and Agave fiber-reinforced 300 PES 24 agave
EPX composites with different PES 31 agave
ratios 250 PES matrix

200

Load (N)
150

100

50

0
0 1 2 3 4 5 6 7
Deflection (mm)

σnom is the stress value obtained experimentally To simulate the response of these composites to three point-
εnom is the strain value obtained experimentally bending test Fig. 5, we work with the following boundary
E is the Young’s modulus conditions:
σtrue and are stress and strain values implemented in the
εtrue finite element solver, respectively BC1: The supporting cylinder position is as: U1 = U2 =
U3 = UR1 = UR2 = UR3 = 0.
Since we propose to simulate the behavior of the composite
BC2: Each supporting pin is encastred with the compos-
under unidirectional monotonous tensile test, the boundary
ite plate at its partition levels.
conditions will be as follows:
BC3: At the initial step, loading pin’s velocity is equal to
zero.
BC1: Encastre: Ui = 0; Uri = 0; i = 1, 2, 3
BC4: At step one, loading pin’s velocity is V2 = − 6 mm/
BC2: Displace: U1 = 8 mm; U2 = U3 = UR1 = UR2 =
mn, V1 = V3 = 0 and VR1 = VR2 = VR3 = 0.
UR3 = 0
We use surface to surface contact. Tangential behavior
The element type used for the meshing is linear hexagonal
(penalty) friction coefficient is equal to 0.15. Furthermore,
8-node linear beam-reduced integration hourglass control
concerning normal behavior, we suppose hard contact
(C3D8R).
(https://www.youtube.com/watch?v=AdTCGbD4nD8).
Figure 4 represents the results of the simulation of the ten-
Element type used in this case is C3D8R.
sile test of Agave polyester composites as well as Agave ep-
Numerical load–defection curves corresponding to Agave-
oxide composites using various weight ratios.
reinforced composites were first drawn as mentioned in
We draw for each composite the numerical and the exper-
Figs. 6 and 7.
imental curve. Then, we conclude from the two curves the
error generated by the model. The smaller the error percent-
age, the more the convergence of the suggested model is
ensured.
For all composites, experimental and numerical curves Table 5 Experimental and numerical values of flexural modulus
present the same shape with an error in the range of 7–15%. (MOE) of Agave-based composites under 3-point bending test
This error may be due to the assumption of homogeneity of Composite Numerical MOE Experimental MOE
composite material which does not exist concretely. (GPa) (GPa)

PES/15% Agave 1.4 1.6

3.2 Macro-scale modeling, 3 point-bending test PES/24% Agave 2 .12 2.5
simulation PES/31% Agave 2.4 2.6
Epoxyde/10% Agave 2.34 2.4
We consider the same constitutive law for the model used in Epoxyde/20% Agave 2.4 2.5
the simulation of the tensile test since we are working with the Epoxyde/30% Agave 2.7 2.8
same materials already described.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Int J Adv Manuf Technol (2019) 103:4671–4680 4677

Fig. 8 3D composite model at

mesoscopic scale

Figure 6 shows the shape of the PES bending curves and 3.3 Meso-scale modeling, tensile test simulation
PES/Agave fibers. It is noted from this figure that the failure with elastic behavior law
mechanism for the polyester polymer is the same compared
with that of Agave/polyester composite. The material is modeled using certain assumptions and ana-
The curve was clearly linear for the PES and Agave lyzed for mechanical properties with finite element method
fiber-reinforced polyester composites. Besides, the flexur- software (ABAQUS 6.14 version). The composite material
al properties increase with the addition of fibers. Flexural is assigned as unidirectional composite by assuming the fiber
strength rises with about 12% in the case of PES 15 and matrix materials are linearly elastic. Our model is a three-
Agave composite, 44% for PES 24 Agave composite, dimensional deformable body. We created a rectangle that is
and 52% in PES 31 Agave composite. then extruded to form a parallelepiped. The modeling of the
Figure 7 shows the shape of the EPX bending curves fibers is created by their partition in the matrix in order to
and EPX/Agave fibers. It is noted from this figure that the ensure a perfect adhesion. Thus, we can model the composite
failure mechanism for the epoxide polymer is the same on the mesoscopic scale, without having to create different
compared with that of Agave/epoxide composite. The pieces. The number of circles corresponds to the mass fraction
curve was clearly linear for the EPX and Agave fiber- considered in the study (Fig. 8).
reinforced epoxide composites. Besides, the flexural prop- Dimensions of sample are 100 ∗ 20 ∗ 2 mm3. Fiber’s di-
erties increase with the addition of fibers. Flexural ameter is about 0.263 mm. Elastic parameters of Agave fibers
strength rises with about 51% in the case of EPX 10 and matrices are summarized in Table 6.
Agave composite, 52% for EPX 20 Agave composite, In order to apply the boundary conditions, we have coupled
and 57% in EPX 30 Agave composite. the two faces z = 0 and Z = 100 mm to a rigid body with well-
According to three-point bending test simulation results defined reference points (RP1 and RP2).
illustrated in Figs. 6 and 7, it is evident that when the fiber Hexahedral dominated elements shape are employed and
content increases in the polyester or epoxide matrix, the flex- C3D8R elements are the most suitable for meshing the
ural strength increases. composite.
In fact, the polyester/epoxide resin transmits and distrib- The numerical simulation results appear in Fig. 9a,b.
utes the applied stress to the Agave fibers resulting in The curves in Fig. 9a,b present the same shape with a slight
higher strength. As a result, the composite can sustain difference. The difference between experimental and numeri-
higher load before failure compared to the unreinforced cal simulated curves is equal to 10% for the case of polyester/
polyester. agave composite. While, for epoxide Agave composite, the
The flexural modulus increases as the weight ratio of fiber difference between numerical and experimental curves is
increases. Same results were reported by Chaabouni [7]. This about 2%. Hence, the model converges. Numerical and exper-
ascertains the validity of the model. imental results have very good agreement.
Numerical flexural modulus was calculated using the Although the previous model shows good results, it does
curves in Figs. 6 and 7 and applying the following formula: not reflect the real behavior law of Agave fibers as well as
matrices employed in this study. That is why, we propose a
P 1 L3
MOE ¼ ðMPaÞ ð5Þ
4be3 y1
Table 6 Fiber and matrix properties [7] (http://www.materiatech-
where P1 is the maximum load (N), L the range (mm), e the carma.net/html/pdf/GlossaireMateriauxComposites_CARMA.pdf)
specimen thickness (mm), b the specimen width (mm), and y1
E (MPa) ν
the defection (mm).
Results are summarized in Table 5. Agave fibers 4000 0.33
Comparison of experimental and FEM tensile and three- Polyester matrix 870 0.37
point bending strength shows very good agreement since error Epoxide matrix 1924 0.386
average is in the range of 8%.

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

4678 Int J Adv Manuf Technol (2019) 103:4671–4680

Fig. 9 Comparison between

experimental and numerical 3000
responses under tensile test
considering elastic behavior law 2500
of polyester 15% Agave
composite (a) and epoxide 10% 2000
Agave composite (b),

Load (N)
respectively 1500

1000
Experimental curve

500 Numerical curve

0
0 1 2 3 4 5 6
(a) Displacement (mm)

5000
4500
4000
3500
Load (N)

3000
2500
2000
1500 Experimental curve
1000
Numerical curve
500
0
0 1 2 3 4 5 6
(b)
Displacement (mm)

second model, in which we consider a viscoelastic behavior explained by the fact that during the preparation of the model,
law for Agave fibers and an elastoplastic behavior for polyes- we considered and worked with simplifying assumptions,
ter and epoxide matrix, as is mentioned in the literature [21]. which do not exist in the real case. As examples of simplifying
hypothesis, we note the following: absence of porosity, conti-
3.4 Meso-scale modeling, tensile test simulation nuity of materials, absence of voids in fibers and matrix, iso-
with viscoelastoplastic behavior law tropic behavior of Agave fibers … etc.
Difference between experimental- and numerical-
The viscoelastic law of fibers is identified by the load– simulated curves is equal to 5.88% for the case of Agave
extension test added with the relaxation test. These tests have PES composite and 9% for the case of Agave EPX composite.
been then implemented in the ABAQUS finite element pack- Hence, the model converges. Numerical and experimental re-
age via Prony series method. In this method, we conduct an sults have very good agreement. The error value is smaller
explicit analysis. The mechanical property values are the same when considering a viscoelastoplastic behavior law compared
as mentioned in Table 6. Results of simulation under tensile to the error value generated when working at macroscopic
test for Agave polyester and Agave epoxide composites are scale. Hence, this ascertains the importance and the validity
illustrated in Fig. 10. of the mesoscopic-scale model using a viscoelastoplastic be-
In the previous figure, we present the response under ten- havior law of Agave americana L. composite materials.
sile test of PES 15% Agave composite (Fig. 10a), particularly
experimental curve (red) versus numerical one (purple). The
same work is done for EPX 10% Agave composite; the ex-
perimental curve (blue) as well as the numerical one (red) are 4 Conclusions
drawn (Fig. 10b).
Figure 10a,b is very similar and presents the same shape Agave americana L. fibers are green, biodegradable, and eco-
with a slight difference, which corresponds to the gap between friendly fibers. In addition, they are very interesting and pos-
the experimental procedure and the numerical one. This is a sess promising properties, thanks to their low density and high
very frequent phenomenon that appears in all research work flexibility. Their application in composite material contributes
dealing with modeling and simulation [10, 18, 19]. It can be in enhancing Young’s modulus and the global composite’s

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Int J Adv Manuf Technol (2019) 103:4671–4680 4679

Fig. 10 Comparison between 5000

experimental and numerical
4500
responses under tensile test using
explicit analysis and 4000
viscoelastoplastic behavior law of 3500
polyester 15% agave composite 3000
(a) and epoxide 10% agave (b),

Load (N)
2500
respectively
2000 Experimental curve
1500
1000 Numerical curve
500
0
0 2 4 6 8 10 12
(a) Displacement (mm)

4500
4000
3500
3000
Load (N)

2500
2000 Experimental curve
1500 Numerical curve
1000
500
0
0 2 4 6 8 10 12 14 16
(b)
Displacement (mm)

properties, which leads us to conclude their academic and A second model in which we consider a viscoelastic be-
industrial values. havior for Agave fibers and an elastoplastic behavior for poly-
Finite element analysis is performed in two different scales: ester and epoxide matrices was thus performed to remedy to
macroscopic and mesoscopic (scale of fibers). For macro- deficiencies of the first simulation model.
scale modeling, we have predicted the behavior and the re- The error percentage for the second model does not exceed
sponse of polyester Agave composites as well as epoxide 9%, which is evaluated as a very satisfying result.
Agave composites. Numerical and experimental results were The difference in the results, obtained from finite element
globally in agreement, even so the presence of a slight differ- analysis and experimental analysis, could be explained by the
ence which varied depending on the material studied. This fact that some characteristics of the material were not taken
error was in the interval of 7 and 15% concerning tensile test into account in the models. For instance, we note the presence
simulation results. of porosity, discontinuity, and voids in the fiber and matrix
Moreover, we have simulated the behavior of composites and the anisotropic behavior of the Agave fiber. We work
under three-point bending test. In one hand, we have calculat- rather with simplifying assumptions which are very useful in
ed flexural modules. In the other hand, we showed the influ- numerous research studies [18].
ence of fiber addition to polyester and epoxide matrices on The error between experimental and numerical results of
flexural properties of composite materials. Simulation results Agave fiber composites could be also attributed to the non-
were in very good agreement with the literature. uniformity and heterogeneity of the fibers.
Concerning the meso-scale modeling, we have predicted the
behavior and the response of polyester Agave composites as
well as epoxide Agave composites, taking into account the prop-
References
erties of matrix and fibers. We started by a simple elastic behav-
ior law for Agave fibers, polyester, and epoxide. Even though 1. Wertz JL, Richel A (2014) Les biocomposites et composites
this model was in concordance with the experimental results, it polymère-chanvre en particulier. Report, Unité de chimie
did not reflect accurately the behavior law of the materials. biologique industrielle, Liège

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

4680 Int J Adv Manuf Technol (2019) 103:4671–4680

2. Wilson MJ (2003) Finite element analysis of glass fibre reinforced 13. Ben Cheikh R, Ben Brahim S The agave fibers in composite
thermoplastic composites for structural automotive component. materials
Online referencing. http://eprints.nottingham.ac.uk/11773/1/M_ 14. Da Silva LJ, Panzera TH, Christoforo AL, Durão LMP, Rocco Lahr
Wilson_396786.pdf. Accessed Sept. 2003 FA (2012) Numerical and experimental analyses of biocomposites
3. Berreur L, Maillard B Nösperger SL (2002) Industrie française des reinforced with natural fibres. Int J Mater Eng 2(4):43–49
matériaux composites. Report, Nodal Consultants, Paris 15. SarvanaBavan D, Mohan Kumar GC (2013) Finite element analysis
4. Wambua P, Ivens J, Verpoest I (2003) Natural fibres: can they re- of a natural fiber (Maize) composite beam. J Eng 2012:1–7
place glass in fibre reinforced plastics? Compos Sci Technol 63(9): 16. Ghering F (2013) Mechanical behaviour and damage study of short
1259–1264 hemp fibre reinforced thermoplastics: experimental approach and
5. Ben Mlik Y, Jaouadi M, Rezig S, Khoffi F, Msahli S, Durant B modelling. PhD Thesis, University of Lorraine
(2017) Kenaf fibre-reinforced polyester composites: flexural char- 17. Sun CT, Vaidya RS (1996) Prediction of composite properties from
acterization and statistical analysis. J Text Inst 109(6):713–722 a representative volume element. Compos Sci Technol 56(2):171–
6. Rezig S, Khoffi F, Ben Mlik Y, Jaouadi M, Msahli S, Bernar D 179
(2015) Flexural properties of typha natural fiber-reinforced polyes-
18. Kumar J, Praveen DN, Thara R, Irfan G (2016) Experimental &
ter composites. Fibers Polym 16(11):2451–2457
finite element analysis of sisal fibre reinforced composites. IJRTER
7. Chaabouni Y (2005) Microstructure characterisation of Agave
2(7):155–160
americana L. fibre; study of Agave fibre reinforced composites.
PhD thesis, University of Mulhouse 19. Prasad V, Venkatachalam G, Rathi A, Rajakuma S (2014) Finite
8. Ghali L, Msahli S, Zidi M, Sakli F (2011) Effects of fibre weight element analysis of jute fibre made hybrid polymer matrix compos-
ratio, structure and fibre modification onto flexural properties of ite. Appl Mech Mater 592(594):363–367
Luffa–polyester composites. AMPC 1(3):78–85 20. Houshyar S, Shanks RA, Hodzic A (2009) Modelling of polypro-
9. Ben Brahim S, Ben Cheikh R (2007) Influence of fibre orientation pylene fibre-matrix composites using finite element analysis.
and volume fraction on the tensile properties of unidirectional Alfa- Express Polym Lett 3(1):2–12
polyester composite. Compos Sci Technol 67(1):140–147 21. Perrot Y, Baley C, Davies P (2006) Influence of low styrene emis-
10. Rajesh M, Srinag T, PhaniPrasanthi P, Venkataraovenkatrao K sion polyester resins on the aging behavior of composites in a ma-
(2016) Finite element analysis of coir/banana fiber reinforced com- rine environment. Appl Compos Mater 13(1):1–22
posite material. International Journal of Advanced Research in
Mechanical Engineering & Technology 2(4):29–33
11. Msahli S, Chaabouni Y, Sakli F, Drean JY (2007) Mechanical be-
havior of Agave Americana L. fibres: correlation between fine
structure and mechanical properties. J Appl Sci 7(24):3951–3957
12. El Oudiani A (2003) Study of the mechanical behavior of agave Publisher’s note Springer Nature remains neutral with regard to
fibers Americana L. Master’s thesis. Engineering School of jurisdictional claims in published maps and institutional affiliations.
Monastir, University of Monastir

Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:

1. use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
2. use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
3. falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
4. use bots or other automated methods to access the content or redirect messages
5. override any security feature or exclusionary protocol; or
6. share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at

[email protected]