Composites Part B 127 (2017) 166e174

Contents lists available at ScienceDirect

Composites Part B
journal homepage: www.elsevier.com/locate/compositesb

A new approach for nonlinear buckling analysis of imperfect

functionally graded carbon nanotube-reinforced composite plates
Pham Toan Thang, Tan-Tien Nguyen, Jaehong Lee*
Department of Architectural Engineering, Sejong University, 209 Neungdong-ro, Gwangjin-gu, Seoul 05006, Republic of Korea

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

Article history: This paper presents an analytical approach to analyze the nonlinear static buckling of imperfect func-
Received 20 August 2016 tionally graded carbon nano-reinforced composite (FG-CNTRC) plates subjected to axial compression.
Received in revised form The material properties of the FG-CNTRC are assumed to be graded through the thickness direction
11 November 2016
according to several linear distributions of the volume fraction of carbon nanotubes. The theoretical
Accepted 4 December 2016
formulations are based on the classical plate theory (Kirchhoff plate) with von Karman-type of nonlin-
Available online 8 December 2016
earity and the initial geometrical imperfection. The approximate solution is developed for simply sup-
ported and freely movable boundary conditions. By applying the traditional Galerkin method and the
Keywords:
Carbon nanotube-reinforced composite
Airy stress function, the results for the critical load are obtained in closed-form solutions, which are
Nonlinear static buckling analysis convenient to be used in engineering design. Some illustrative examples are also presented in details to
Airy stress function investigate the effects of the imperfection, carbon nanotubes, and geometrical parameters on the
Imperfection nonlinear static behavior the plates.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction presented a discussion about the interfacial bonding properties,

The carbon nanotubes (CNTs) have been considered to be one
the hottest research topics in various fields of research and tech-
nology over the past decade. Most journal papers and technical
reports on CNTs cite Iijima's work [1] as the first discovery of carbon
nanotube. The outstanding electrical, mechanical, and thermal
characteristics of carbon nanotubes (CNTs) [2,3] have made them
among the most promising materials in a wide range of applica-
tions such as nano-sensors and atomic transportation [4]. In addi-
tion, the excellent mechanical properties of CNTs, such as ultra-
high Young's modulus around 1 TPa [5] and tensile strengths
varying from 11 to 63 GPa [6], are promising ultra-high-strength
reinforcements in high-performance polymer matrix composites.
In recent years, many works have been carried out to study the
constitutive models and mechanical properties of CNTRC. By using
a combination of computational and experimental methods, Gou
et al. [7] investigated the interfacial bonding of single-walled
nanotube (SWCNT) reinforced epoxy composites. The recent re-
searches and applications on carbon nanotubes and nanotube
composites are also reviewed by Lau and Hui [8]. These authors
* Corresponding author.
E-mail addresses: phamtoanthang1991@gmail.com (P.T. Thang), tien.ntan@
gmail.com (T.-T. Nguyen), jhlee@sejong.ac.kr (J. Lee).
mechanical performance and reliability of nanotube/polymer
composites. In view of the morphology of the composites and the
anisotropic nature of layered silicate, formulas for the overall
moduli of composite materials reinforced with transversely
isotropic spheroids were proposed by Wang and Pyrz [9]. Han and
Eliott [10] reported the classical molecular dynamics (MD) simu-
lations of model polymer/CNT composites constructed by
embedding a single wall (10,10) CNT into two different amorphous
polymer matrices: poly (methyl methacrylate) (PMMA) and poly
(m-phenylene-vinylene)-co-[(2,5-dioctoxy-p-phenylene) vinylene]
(PmPV), respectively. Joshi and Upadhyay [11] studied the elastic
response of multiwalled carbon nanotube (MWCNT) reinforced
composite for different interphase property between matrix and
multiwalled CNT. The stress distribution of CNTs embedded within
polyimide matrix subjected to applied loading was investigated by
Tzeng and Tsai [12] when using molecular dynamics simulation.
Using the finite element (FE) method, the role of the interphase on
mechanical performance of glassy polymer/single-walled CNTRC
has been investigated by Banerjee et al. [13]. Tarfaoui et al. [14]
presented a report about the mechanical properties of carbon
nanotubes based polymer composites. The objective of this work
was to understand the effect of carbon nanotubes (CNT) additives
on the elastic behaviors of textile based composites.
With regard to the analysis of CNTRC structural elements, the
number of publications has considerably increased in recent years

http://dx.doi.org/10.1016/j.compositesb.2016.12.002
1359-8368/© 2017 Elsevier Ltd. All rights reserved.
 P.T. Thang et al. / Composites Part B 127 (2017) 166e174
with plenty of newly results. Wuite and Adali [15] presented a
multiscale analysis of the deflection and stress behavior of CNTRC.
Arani et al. [16] investigated the buckling analysis of laminated
composite plates reinforced by SWCNTs. Through the molecular
dynamics simulation method, Wong and Vijayaraghavan [17]
studied the buckling characteristics of several curved forms of
SWCNTs. Using the Eshelby-Mori-Tanaka approach, Formica et al.
[18] investigated the vibrational properties of CNTRC by employing
an equivalent continuum model. Wang and Shen [19] studied the
large amplitude vibration and the nonlinear bending of a sandwich
plate with CNTRC face sheets. The static and free vibration analyses
of CNTRC plates using FEM with was carried out by Zhu et al. [20].
Based on the first and third order beam theories, Lin and Xiang [21]
analyzed the vibration of CNTRC beams. The nonlinear forced vi-
bration behavior of CNTRC plates using a numerical approach was
also investigated Ansari et al. [22]. Baykasoglu et al. [23] analyzed
the equivalent nonlinear fracture models for pristine and recon-
structed one- and two-atom vacancy defected single wall carbon
nanotubes by using the molecular mechanics based models. The
vibration analysis of two orthogonal slender single-walled carbon
nanotubes with a new insight into continuum-based modeling of
van der Waals forces was investigated by Kiani [24].
It is well-known that the applications of CNTs to nano-
composites have been hindered due to the weak interfacial bonding
between CNTs and matrix. Functionally graded materials (FGMs)
are inhomogeneous composites characterized by smooth and
continuous variations in both compositional profile and material
properties and have found a wide range of applications in many
industries [25e27]. By using concept of FGM, Shen and Wang [28]
carried out the thermal buckling and postbuckling behavior for FG
nanocomposite plates reinforced by SWCNTs. Based on three-
dimensional theory of elasticity, the bending behavior of FG-
CNTRC rectangular plate subjected to thermo-mechanical loads
Fig. 1. Configurations of CNT reinforced functionally graded plates.
167
was examined by Alibeigloo and Liew [29]. Based on Reissners
mixed variational theorem (RMVT), Wu and Chang [30] developed
a unified formulation of finite layer methods (FLMs) for the three-
dimensional (3D) buckling analysis of simply-supported, FG-CNTRC
plates with surface-bonded piezoelectric actuator and sensor
layers. Based on the first order shear deformation plate theory
(FSDT), Jalali and Heshmati [31] investigated the buckling behavior
of circular sandwich plates with tapered cores and FG-CNTRC face
sheets under uniform radial compression. Using Mindlina ^™s strain
gradient theory, shahriari et al. [32] performed the vibration
analysis of FG-CNTRC nanoplates. As a first endeavor, the vibration
behavior of FG-CNTRC truncated conical panels was studied by
Jooybar et al. [33]. Shen [34] analyzed thermal postbuckling anal-
ysis for FG-CNTRC cylindrical shells reinforced by SWCNTs sub-
jected to a uniform temperature rise. Recently, Garca-a-Macas et al.
[35] provided some insight into the linear buckling analysis of FG-
CNTRC cylindrical curved panels under compressive and shear
loading.
In the abovementioned studies, nonlinear static buckling of
imperfect FG-CNTRC plates based on analytical solution has not
been yet considered. In this work, an analytical approach is
developed successfully to analyze the nonlinear static buckling of
imperfect functionally graded carbon nano-reinforced composite
plates with simply supported boundary conditions and subjected to
axial compression. The material properties of functionally graded
carbon nano-reinforced composite (FG-CNTRC) plates are assumed
to be graded through the thickness direction according to several
linear distributions of the volume fraction of carbon nanotubes. The
governing equations for thin FG-CNTRC plates are derived by using
the classical plate theory (Kirchhoff plate) with von Karman geo-
metric nonlinearity and are solved by the traditional Galerkin
method and the Airy stress function to obtained the critical buck-
ling load and the load-deflection relations in closed-form solutions.
168 P.T. Thang et al. / Composites Part B 127 (2017) 166e174

The influence of initial geometrical imperfection of the plate is also

 wCNT
investigated. A detailed parametric study is conducted to gain and VCNT ¼
 ; (9)
insight into the effects of CNTs, imperfection size, geometrical pa- rCNT
wCNT þ rm ð1  wCNT Þ
rameters on the nonlinear buckling of FG-CNTRC plates.
The outline of the paper is organized as follows. In Section 2, we
present the material properties of FG-CNTRC plates. The theoretical in which wCNT is the mass fraction of the CNTs. rCNT and rm are the
formulations for the imperfect FG-CNTRC plate are derived in densities of CNT and matrix, respectively. nCNT
12 and nm are the
Section 3. The boundary condition and solutions are proposed in Poisson's ratios of the CNTs and polymer matrix, respectively.
Section 4. Section 5 gives some illustrative examples. Finally, the Note that overall mass fraction and volume of the CNT in
main conclusions are given in Section 6. different kinds of the FG-CNTRC plates are equal.

2. Material properties of FG-CNTRC plates 3. Theoretical formulations for FG-CNTRC plates

Consider a rectangular FG-CNTRC plate of length a in the x di- This work is studied by applying thin plate theory [39] and
rection, width b in the y direction and thickness h in the z direction introducing:
as shown in Fig. 1. The uniform distribution is represented by UD.
FG-O and FG-X represent the functionally graded distributions of  Von Karman approximation, for which all displacements are
carbon nanotubes in the thickness direction of the composite infinitesimal except the displacement w normal to mid-plane.
plates. For the FG-X type, the top and the bottom surface of CNTRC  Kirchhoff hypothesis, for which the normals to mid-plane
plate are CNT-rich. And in case of FG-O, the CNT-rich zone is in the remain normal to deformed mid-plane.
middle of the CNTRC plate. Unlike the isotropic properties of
polymer matrix, the CNTs represent anisotropic behavior. The Under these assumptions, forces and moments per unit length
effective material properties of this mixture can be predicted by on mid-plane fNg and fMg, and strains and curvatures of mid-
Mori-Tanaka scheme [36] or the rule of mixture [37]. In Ref. [36], a plane fε0 g and fkg, are related as shown Eq. (10):
comparative study between Mori-Tanaka scheme and the rule of
mixtures was presented. Based on the extended rule of mixture, the 8 9 2 38 0 9
> Nx > A A12 0 B11 B12 0 > εx >
>
> >
> 6 11 >
> >
effective properties of the FG-CNTRCs can be expressed as [38]. >
> Ny >> 0 7 > ε0 >>
>
< >
= 66 A12 A22 0 B12 B22 7>>
< y0 >
>
=
Nxy 0 0 A66 0 0 7
B66 7 g
E11 ¼ h1 VCNT E11
CNT
þ Vm Em ; (1) ¼6
6 B11 xy ; (10)
>
> Mx >
> 6 B12 0 D11 D12 0 77>> kx >>
>
> > 4 > >
>
> My >>
> B12 B22 0 D12 D22 0 >5 >
> ky >
>
>
: ; : ;
n2 VCNT Vm Mxy 0 0 B66 0 0 D66 kxy
¼ CNT þ m; (2)
E22 E22 E
where Aij are called extensional stiffness matrix, Dij are terms of
n3 V Vm flexural stiffness matrix, and Bij are terms of bending-extensional
¼ CNT þ m ; (3)
G12 GCNT G coupling stiffnesses, are given as
12

CNT , E CNT are the Young's modulus of the CNTs in x and y

where E11 Zh=2 Zh=2
22 E11 E22 v12
A11 ¼ dz; A12 ¼ dz;
directions, and GCNT m
12 indicates shear modulus of the CNTs. E and 1  n12 n21 1  n12 n21
m
G are the corresponding properties for the matrix. It should be h=2 h=2

noted that E11 ; E22 and G12 are the corresponding properties of the Zh=2 Zh=2
FG-CNTRCs. Due to the presence of small scale effect, the CNT ef- E22
A22 ¼ dz; A66 ¼ G12 zdz;
ficiency parameters, hi (ði ¼ 1÷3Þ) are introduced in Ref. [38]. VCNT 1  n12 n21
and Vm are the volume fractions of the CNTs and matrix, respec- h=2 h=2

tively and the relation among these volume fractions is given by Zh=2 Zh=2
E11 E22 v12
A11 ¼ dz; A12 ¼ dz;
VCNT þ Vm ¼ 1: (4) 1  n12 n21 1  n12 n21
h=2 h=2
Functionally graded material profile of CNTs is taken to be of
three types namely, UD, FG-O and FG-X. The volume fraction VCNT Zh=2 Zh=2
E11 E22 v12
are assumed to have B11 ¼ zdz; B12 ¼ zdz; (11)
1  n12 n21 1  n12 n21
 h=2 h=2
UD : VCNT ðzÞ ¼ VCNT : (5)
Zh=2 Zh=2

 E22
2jzj  B22 ¼ zdz; B66 ¼ G12 zdz;
FG  O : VCNT ðzÞ ¼ 2 1  VCNT : (6) 1  n12 n21
h h=2 h=2

Zh=2 Zh=2
4jzj  E11 E22 v12 2
FG  X : VCNT ðzÞ ¼ V : (7) D11 ¼ z2 dz; D12 ¼ z dz;
h CNT 1  n12 n21 1  n12 n21
h=2 h=2
and this case, the Poisson's ratio can be obtained as
  Zh=2 Zh=2
 E22
n12 ¼ VCNT nCNT
12 þ 1  n12
CNT
nm ; (8) D22 ¼ z2 dz; D66 ¼ G12 z2 dz:
1  n12 n21
h=2 h=2
where
 P.T. Thang et al. / Composites Part B 127 (2017) 166e174 169

Applying Hamilton's principle, the nonlinear equilibrium

equations of the FG-CNTRC plate based on the Kirchhoff plate A1 ¼ ½B11 C12 þ B12 C22 ;
theory can be obtained A2 ¼ ½B11 C11 þ 2B12 C12 þ B22 C22 þ B66 C31 ;
A3 ¼ ½B12 C11 þ B22 C12 ;
A4 ¼ ½B11 C13 þ C12 C23  D11 ;
vNx vNxy A5 ¼ ½B11 C14 þ B12 C24  2D12 þ B12 C13 þ B22 C23 þ B66 C32  2D66 ;
þ ¼ 0; (12) A6 ¼ ½B12 C14 þ B22 C24  D22 :
vx vy
(21)

vNxy vNy Eq. (20) includes two dependent unknowns, w and F. To obtain
þ ¼ 0; (13) a second equation relating these two unknowns, the compatibility
vx vy
equation may be used. The strains must be relative in the defor-
mation compatibility equation [40].

v2 Mx v2 Mxy v2 My v2 w v2 w v2 w !2
v2 ε0x v εy v gxy
2 0 2 0
þ2 þ þ Nx 2 þ 2Nxy þ Ny 2 ¼ 0: v2 w v2 w v2 w
vx 2 vxvy vy 2 vx vxvy vy þ 2  ¼  : (22)
vy 2 vx vxvy vxvy vx2 vy2
(14)
The equation of equilibrium are identically satisfied by the stress Substituting Eqs. (15)e(18) into Eq. (19) leads to
function, Fðx; yÞ, introduced by G.B. Airy, related to the stresses as
v4 F v4 F v4 F v4 w
follows C22 4
þ ð2C12  C31 Þ 2 2
þ C11 4
þ C23
vx vx vy vy vx4
!2
v2 Fðx; yÞ v2 Fðx; yÞ v2 Fðx; yÞ v4 w v4 w v2 w v2 w v2 w
Nx ¼ ; N ¼ ; N ¼  : (15) þðC13 þ C24  C32 Þ þ C14 ¼  :
vy2
y
vx2
xy
vxvy 2
vx vy 2
vy 4 vxvy vx2 vy2
From first three equations in the constitutive relations Eq. (10), (23)
one can write inversely Assuming that jðx; yÞ is a known function representing initial
small imperfection of the plate. The initial imperfect of the plate
considered here can be seen as a small derivation of the plate
v2 F v2 F v2 w v2 w
ε0x ¼ C11 þ C12 2 þ C13 2 þ C14 2 ; (16) middle surface from the perfect shape. It is assumed that the initial
vy 2 vx vx vy deflection is much smaller compared with the plate dimensions.
The deflection wðx; yÞ can be replaced by the sum of wðx; yÞ þ jðx; yÞ
. This assumption is referred as in Refs. [41,42]. As a result, Eqs. (20)
v2 F v2 F v2 w v2 w and (23) have the following form
ε0y ¼ C12 þ C 22 þ C 23 þ C 24 ; (17)
vy2 vx2 vx2 vy2
v4 F v4 F v4 F v4 w v4 w
A1 4
þ A2 2 2
þ A3 4
þ A4 4
þ A5
vx vx vy vy vx vx2 vy2
v2 F v2 w ! !
g0xy ¼ C31 þ C32 ; (18) v4 w v2 F v2 w v2 j v2 F v2 w v2 j
vxvy vxvy þA6 þ þ 2 þ (24)
vy4 vy2 vx2 vx2 vxvy vxvy vxvy
where !
v2 F v2 w v2 j
þ 2 þ 2 ¼ 0:
" # " # vx vy2 vy
A22 A12
C11 ¼  ; C12 ¼ ;
A212  A11 A22 A212  A11 A22 v4 F v4 F v4 F v4 w
" # " # C22 4
þ ð2C12  C31 Þ 2 2
þ C11 4
þ C23
A12 B12  A22 B11 A12 B22  A22 B12 vx vx vy vy vx4
C13 ¼ ; C14 ¼ ; !2
A212  A11 A22 A212  A11 A22 v4 w v4 w v2 w
" # " # (19) þðC13 þ C24  C32 Þ þ C14 ¼ (25)
A11 A11 B12  A12 B11
2
vx vy 2
vy 4 vxvy
C22 ¼ 2 ; C23 ¼ ;
A12  A11 A22 A212  A11 A22 v2 w v2 w v2 w v2 j v2 w v2 j v2 w v2 j
" #  2 þ 2   2 :
A11 B22  A12 B12 1 2B vx vy2 vxvy vxvy vx2 vy2 vy vx2
C24 ¼ ; C31 ¼  ; C ¼ 66 :
A212  A11 A22 A66 32 A66 Eqs. (24) and (25) are the basic equations used to analyze the
nonlinear buckling of the imperfect FG-CNTRC plates under axial
Substituting once again Eqs. (16)-(18) into the expression of Mx ,
compression.
My , Mxy in Eq. (10) and then into Eq. (14) leads to
4. Boundary condition and solutions
v F
4
v F
4
v F
4 4
v w 4
v w
A1 þ A2 þ A3 þ A4 þ A5 Depending on the in-plane behavior at the edges, the edges are
vx4 vx2 vy2 vy4 vx4 vx2 vy2
(20) simply supported and freely movable (FM). The associated
v4 w v2 F v2 w v2 F v2 w v2 w v2 w boundary conditions are [42].
þA6 þ 2 þ ¼ 0:
vy4 vy2 vx2 vxvy vxvy vx2 vy2 w ¼ Nxy ¼ Mx ¼ 0; Nx ¼ Nx0 at x ¼ 0; a:
(26)
w ¼ Nxy ¼ My ¼ 0; Ny ¼ Ny0 at y ¼ 0; b:
where
170 P.T. Thang et al. / Composites Part B 127 (2017) 166e174

where Nx0 , Ny0 are the pre-buckling force resultants in x and y di-
rections, respectively. 1 A1 a3 ð3F3 abab þ256F1 Þ

To solve two Eqs. (24) and (25) for unknowns w and F with the 12 b
consideration of the boundary condition in Eq. (26), we assume the
1 A3 b3 ð3F3 abab þ256F2 Þ 1 2
following approximate solutions as [42].  þ A2 F3 aba2 b
12 a 4
1 
w ¼ WsinðaxÞsinðbyÞ; (27) 2 4
þ A4 a4 þA5 a2 b þA6 b abW
4 " #
1 18F2 abab2 ðW þ mhÞþ9abaðW þ mhÞNx0
F ¼ F1 cosð2axÞ þ F2 cosð2bxÞ þ F3 cosð2axÞcosð2bxÞ  a
(28)
36 64F3 bðW þ mhÞ
1 1 " #
þ Nx0 y2 þ Ny0 x2 ;
2 2 8 1 18F1 aba2 bðW þ mhÞþ9abbðW þ mhÞNy0
 F3 abðW þ mhÞ b
9 36 64F3 aðW þ mhÞ
in which
¼0:
mp np
a¼ ;b ¼ ; m; n ¼ 1; 2; 3; … (29) (32)
a b
The simply supported FG-CNTRC plates with freely movable
and W is the amplitude of central deflection of the FG-CNTRC plate; edges is assumed to be under in-plane compressive loads Px and Py
m; n are the number of half waves in x and y directions, respectively. (in Pascals) with x¼0 and y¼0, respectively [40].
In Eq. (28), the coefficients Fj ðj ¼ 1; 2; 3Þ are determined by
substitution of Eqs. (27) and (28) into Eq. (25) as Nx0 ¼ Px h; Ny0 ¼ 0: (33)

Combining Eq. (32) and Eq. (33), we can obtain the equation for
b2
F1 ¼ WðW þ 2mhÞ; investigating the closed form solution as
32C22 a 2

a2 W WðW þ 2mhÞ
F2 ¼ WðW þ 2mhÞ; (30) Px ¼ P1 W þ P2 þ P3 þ P4 WðW þ 2mhÞ;
2 ðW þ mhÞ ðW þ mhÞ
32C11 b
(34)
C a4 þ C14 b4 þ ðC13 þ C24  C32 Þa2 b2
F3 ¼  23 W:
C22 a4 þ C11 b4 þ ð2C12  C31 Þa2 b2 where

h i
4 2 2
176b C23 a þ C14 b þ ðC13 þ C24  C32 Þa b
4
P1 ¼ h i ;
9aabh C22 a4 þ C11 b4 þ ð2C12  C31 Þa2 b2
8 h i9
>  C23 a4 þ C14 b4 þ ðC13 þ C24  C32 Þa2 b2 >
>
> i >
>
> A1 a þ A3 b þ A2 a b
4 2
4 2 h >
1 < C22 a4 þ C11 b4 þ ð2C12  C31 Þa2 b2
=
P2 ¼ 2 ;
ha >>
>
>
>
> (35)
>
:   >
;
 A4 a4 þ A5 a2 b2 þ A6 b4


8b A1 A
P3 ¼ þ 3 ;
3abha C22 C11
!
1 a4 b4
P4 ¼ þ :
16ha2 C11 C22

The imperfection of the FG-CNTRC plate is assumed as follows In order to verify the results of present study, let us consider a
[41e43]. special case of m ¼ 0 (perfect plate) plate is made of functionally
graded material. Therefore, the equation result leads to

jðx; yÞ ¼ mhsinðaxÞsinðbyÞ; (31)  2

Dp2 m2 B2a þ n2
where m is the imperfection size of the FG-CNTRC plate with Pxb ¼  ; (36)
0  m  1. B2h m2 B2a þ zn2
By substituting Eqs. (27),(28) and (31) into Eq. (24) and then
applying the Galerkin procedure for the resulting equation yield where
 P.T. Thang et al. / Composites Part B 127 (2017) 166e174 171

.  agreement between the results can be seen in Fig. 2.

Ba ¼ b=a; Bh ¼ b=h; D ¼ D h3 ; z ¼ Py Px ; (37)
5.2. Nonlinear buckling analysis of imperfect FG-CNTRC plates
in which
Next, to investigate the influences of CNT distributions, imper-
E1 E3  E22 fection size and geometric parameters, in this subsection presents
D¼  ; (38)
E1 1  n2 the nonlinear buckling behavior of FG-CNTRC plates under axial
compression. According to the rule of mixture, the effective mate-
and rial properties of CNTs and isotropic polymer matrix can be ob-
tained with the following properties: Em ¼ 2.1 (GPa), n ¼ 0.34,
CNT ¼ 5.6466 (TPa), ECNT ¼ 7.08 (TPa), GCNT ¼ 1.9445 (TPa). The
Zh=2 Zh=2 Zh=2 E11 22 12
E1 ¼ Edz; E2 ¼ Ezdz; E3 ¼ Ez2 dz: (39) (10,10) SWCNTS with the geometrical parameters R ¼ 0.68 (nm),
L ¼ 9.26 (nm), h ¼ 0.067 (nm) are selected as the reinforcements.
h=2 h=2 h=2
The mechanical properties of the CNTs and the matrix are assume
where E is the Young's modulus of the plate; n is the Poisson's ratio with temperature room (T ¼ 300 K). The CNT efficiency parameter
of the plate. in Eqs. (1)e(3) are obtained by the extended rule of mixture the
Eq. (36) has been presented by Javaheri and Elasmi [44] when molecular simulation results [10]. For instance, the detailed mate-
authors investigate the linear buckling of perfect FGM plates under rial properties of SWCNTs for present analysis are reported by Shen
in-plane compressive load. and Zhang [28], and are tabulated in Table 1. For convenience, the
In this study, the critical buckling load is obtained for values of m following dimensionless quantities are defined:
and n that make the preceding expression a minimum.
 Dimensionless central deflection

5. Results and discussion W

W¼ : (41)
h
5.1. Verification

In this subsection, to validate the results of present study for the  Dimensionless buckling load (Zhang et al. [46])
perfect and imperfect plates under uniaxial compression, we
compare the present results with the results given by Shen [45] for Nx a2
Px ¼ : (42)
buckling of isotropic thin plates. In this case, assume that the plate Em h3
is simply supported on all edges. The comparison of dimensionless
In order to analyze the influence of imperfect size, Figs. 3e5
load-deflection curves of an imperfect isotropic plate are shown in
show the relationship between the dimensionless load and
Fig. 2. Geometry parameters of plate are taken to be b=a ¼ 1,
b=h ¼ 40, ðm; nÞ ¼ ð1; 1Þ. The Possion's ratio is assumed v ¼ 0:326.
Before presenting the results, let us consider the dimensionless Table 1
buckling load and dimensionless deflection as Young's modulus for the PMMA/CNT composite reinforced by (10, 10) SWCNTs [28].

VCNT Rule of mixture h1 h2 h3
Px
Pxcr ¼ ; (40) E11 (GPa) E22 (GPa)
Pcr
0.12 94.78 2.9 0.137 1.022 0.715
where Px is the uniaxial compression with simply supported 0.17 138.68 4.9 0.142 1.626 1.138
0.28 224.50 5.5 0.141 1.585 1.109
boundary conditions; Pcr is the critical buckling load. Clearly, good

Fig. 2. Comparison of nonlinear buckling curves for isotropic thin plate under unaxial Fig. 3. Effect of geometrical imperfection on load-deflection curves for FG-CNTRC
compression. plates with UD distribution.
172 P.T. Thang et al. / Composites Part B 127 (2017) 166e174

Fig. 4. Effect of geometrical imperfection on load-deflection curves for FG-CNTRC Fig. 6. Effect of CNT distributions on load-deflection curves for FG-CNTRC plates.
plates with FG-O distribution.

Fig. 7. Effect of volume fraction of CNT on load-deflection curves for FG-CNTRC plates
Fig. 5. Effect of geometrical imperfection on load-deflection curves for FG-CNTRC with UD distribution.
plates with FG-X distribution.

central deflection for the simply supported FG-CNTRC plates with

the different values of the initial imperfection (m ¼ 0, 0.01, 0.05, 0.1,
0.2) under uniaxial compression. In this case, UD, FG-O, FG-X types
of CNT distributions are taken into account, respectively. Geometry
properties of the plate are a=b ¼ 1, b=h ¼ 100. The CNT volume

fraction VCNT is taken to be 12%. At the first case, it can be concluded
that the curves decrease when the imperfection size of the plate
increases with the dimensionless deflection stands at approxi-
mately from 0 to 0.7 (UD), 0 to 0.55 (FG-O), 0 to 0.9 (FG-X),
respectively. As can be observed, an inverse trend occurs when the
deflection exceeds a specific value as 0.7 (UD), 0.55 (FG-O), 0.9 (FG-
X), the curves become higher when m is increased. Therefore, the
initial geometrical imperfection makes plates more unstable under
mechanical load.
To further investigate the effect of UD, FG-O and FG-X distri-
butions, Fig. 6 show the on non-dimensional central deflection and
and-shorting of various types of perfect and imperfect FG-CNTRC
plates under axial compression. Among these distributions, UD
represents uniformly distributed, and FG-O and FG-X denote the
Fig. 8. Effect of volume fraction of CNT on load-deflection curves for FG-CNTRC plates
other two types of FG distributions of CNTs. Geometry parameters with FG-O distribution.
 P.T. Thang et al. / Composites Part B 127 (2017) 166e174 173

Fig. 9. Effect of volume fraction of CNT on load-deflection curves for FG-CNTRC plates Fig. 11. Effect of width-to-length ratio a=b on load-deflection curves for FG-CNTRC
with FG-X distribution. plates with FG-O distribution.

are selected with a=b ¼ 1, b=h ¼ 100, and the volume fraction of CNT

is fixed with VCNT ¼ 12%. It can be found that the central deflection
increases slowly in the pre-buckling stage and increases very fast in
the postbuckling stage. For three types of distributions of SWCNT in
the plates, it can be seen that FG-O plate has the lowest post-
buckling strength, while FG-X plate has the highest postbuckling
strength among the three types of CNTR-FG plates, and the post-
buckling path of UD plate lies between FG-X and FG-O. It is clear
that CNTs distributed close to top and bottom surfaces are more
efficient in increasing the stiffness and postbuckling strength of the
FG-CNTRC plates than CNTs distributed near the mid-plane.
In order to examine the effectiveness of volume fraction of CNTs
 , Figs. 7e9 present the relationship between the dimensionless
VCNT
buckling load and central deflection for UD, FG-O, and FG-X plates
with different values of nanotube volume fraction. Three different
values of CNT volume fraction are considered as VCNT  ¼ 12%,

VCNT 
¼ 18% and VCNT ¼ 27%. The width-to-length ratio a=b of plate is
fixed at 1.0. It can be found that the larger the carbon nanotube
volume fraction, the greater is the degree of postbuckling strength Fig. 12. Effect of width-to-length ratio a=b on load-deflection curves for FG-CNTRC
observed. This is to be expected, because the increase of the carbon plates with FG-X distribution.

nanotube volume fraction yields an increase of postbuckling

strength and the FG-CNTRC plate becomes stiffer.
Finally, Figs. 10e12 show the impacts of geometric parameters
on load-deflection curves of imperfect FG-CNRTC plates for three
types of the CNTs distribution (UD, FG-O and FG-X), respectively.
The plate is considered with four different values of the width-to-
length ratio a=b (a=b¼1, 1.25, 1.5, 2.0) for two cases of perfect
(m ¼ 0) and imperfect (m ¼ 0:1) plates. The volume fraction of CNT

distribution is chosen to be VCNT ¼ 12%. From these figures, with the
width-to-length ratio a=b of plates changing from 1.0 to 2.0, the
postbuckling strength is slightly increased. These results are useful
to examine the influence of different ratio a=b on the behavior of
the plate.

6. Conclusions

In this paper, an analytical solution is developed to analyze the

nonlinear static buckling of imperfect functionally graded carbon
Fig. 10. Effect of width-to-length ratio a=b on load-deflection curves for FG-CNTRC nano-reinforced composite plates with simply supported boundary
plates with UD distribution. conditions and subjected to axial compression. The material
174 P.T. Thang et al. / Composites Part B 127 (2017) 166e174
properties of functionally graded carbon nano-reinforced com-
posite (FG-CNTRC) plates are assumed to be graded through the
thickness direction according to several linear distributions of the
volume fraction of carbon nanotubes. Fundamental equations for
thin FG-CNTRC plates have been obtained using the classical plate
theory (Kirchhoff plate) with von Karman-type of nonlinearity for
large deflections. The influence of initial geometrical imperfection
of plate was investigated. Applying the traditional Galerkin method
and the Airy stress function, the results for the critical buckling and
the load-deflection relations are obtained in closed-form solutions,
which are convenient to be used in engineering design. It was
discerned that the CNT volume fraction and its distribution type
have significant effect on the buckling behavior of the imperfect
plate. Increase of the volume fraction of the reinforcement makes
the plate stiffer which then results in the larger deflection. Further,
it was found that the FG-O plate is of the smallest stiffness so, it has
the lowest load capacity when compare to the other types of CNTRC
plate. Moreover, the result is also shown that imperfection size
significantly influence on the nonlinear static behavior of the
plates.
Acknowledgements
This research was supported by a Grant (NRF-
2015R1A2A1A01007535) from NRF (National Research Foundation
of Korea) funded by MEST (Ministry of Education and Science
Technology) of Korean government. The support is gratefully
acknowledged.
References
[1] Iijima S. Helical microtubules of graphitic carbon. Nature 1991;354(6348):
56e8.
[2] Wilder1998 Wilder JWG, Venema LC, Rinzler AG, Smalley RE, Dekker C.
Electronic structure of atomically resolved carbon nanotubes. Nature
1998;391(6662):59e62.
[3] Yu MF, Lourie O, Dyer MJ, Moloni K, Kelly TF, Ruoff RS. Strength and breaking
mechanism of multiwalled carbon nanotubes under tensile load. Science
2000;287(5453):637e40.
[4] Wang Q. Atomic transportation via carbon nanotubes. Nano Lett 2009;9(1):
245e9.
[5] Kt Lau, Gu C. A critical review on nanotube and nanotube/nanoclay related
polymer composite materials. Compos Part B Eng 2006;37(6):425e36.
[6] Coleman JN, Khan U, Blau WJ, Gun’ko YK. Small but strong: a review of the
mechanical properties of carbon nanotube“polymer composites. Carbon
2006;44(9):1624e52.
[7] Gou J, Minaie B, Wang B, Liang Z, Zhang C. Computational and experimental
study of interfacial bonding of single-walled nanotube reinforced composites.
Comput Mater Sci 2004;31(3):225e36.
[8] Lau AKT, Hui D. The revolutionary creation of new advanced materi-
als^a”carbon nanotube composites. Compos Part B Eng 2002;33(4):263e77.
[9] Wang J, Pyrz R. Prediction of the overall moduli of layered silicate-reinforced
^”part I: basic theory and formulas. Compos Sci Tech
nanocompositesa
2004;64(7):925e34.
[10] Han Y, Elliott J. Molecular dynamics simulations of the elastic properties of
polymer/carbon nanotube composites. Comput Mater Sci 2007;39(2):315e23.
[11] Joshi P, Upadhyay S. Effect of interphase on elastic behavior of multiwalled
carbon nanotube reinforced composite. Comput Mater Sci 2014;87:267e73.
[12] Tzeng SH, Tsai JL. Investigating the stress distribution of single walled carbon
nanotubes embedded in polyimide nanocomposites. J Reinf Plast Compos
2011;30(11):922e31.
[13] Banerjee D, Nguyen T, Chuang TJ. Mechanical properties of single-walled
carbon nanotube reinforced polymer composites with varied interphase's
modulus and thickness: a finite element analysis study. Comput Mater Sci
2016;114:209e18.
[14] Tarfaoui M, Lafdi K, El Moumen A. Mechanical properties of carbon nanotubes
based polymer composites. Compos Part B Eng 2016;103:113e21.
[15] Wuite J, Adali S. Deflection and stress behaviour of nanocomposite reinforced
beams using a multiscale analysis. Compos Struct 2005;71(3):388e96.
[16] Arani AG, Maghamikia S, Mohammadimehr M, Arefmanesh A. Buckling
analysis of laminated composite rectangular plates reinforced by SWCNTs
using analytical and finite element methods. J Mech Sci Technol 2011;25(3):
809e20.
[17] Wong C, Vijayaraghavan V. Nanomechanics of imperfectly straight single
walled carbon nanotubes under axial compression by using molecular dy-
namics simulation. Comput Mater Sci 2012;53(1):268e77.
[18] Formica G, Lacarbonara W, Alessi R. Vibrations of carbon nanotube-reinforced
composites. J Sound Vib 2010;329(10):1875e89.
[19] Wang ZX, Shen HS. Nonlinear vibration and bending of sandwich plates with
nanotube-reinforced composite face sheets. Compos Part B Eng 2012;43(2):
411e21.
[20] Zhu P, Lei Z, Liew K. Static and free vibration analyses of carbon nanotube-
reinforced composite plates using finite element method with first order
shear deformation plate theory. Compos Struct 2012;94(4):1450e60.
[21] Lin F, Xiang Y. Vibration of carbon nanotube reinforced composite beams
based on the first and third order beam theories. Appl Math Modell
2014;38(15):3741e54.
[22] Ansari2015 Ansari R, Hasrati E, Faghih Shojaei M, Gholami R, Shahabodini A.
Forced vibration analysis of functionally graded carbon nanotube-reinforced
composite plates using a numerical strategy. Phys E 2015;69:294e305.
[23] Baykasoglu C, Kirca M, Mugan A. Nonlinear failure analysis of carbon nano-
tubes by using molecular-mechanics based models. Compos Part B Eng
2013;50:150e7.
[24] Kiani K. Vibration analysis of two orthogonal slender single-walled carbon
nanotubes with a new insight into continuum-based modeling of van der
Waals forces. Compos Part B Eng 2015;73:72e81.
[25] Koizumi M. Functionally gradient materials the concept of FGM. Ceram Trans
1993;34:3e10.
[26] Obata Y, Noda N. Transient thermal stresses in a plate of functionally gradient
material. Ceram Trans Funct Graded Mater 1993;34:403e10.
[27] Koizumi M. FGM activities in Japan. Compos Part B Eng 1997;28(1e2):1e4.
[28] Shen HS, Zhang CL. Thermal buckling and postbuckling behavior of func-
tionally graded carbon nanotube-reinforced composite plates. Mater Des
2010;31(7):3403e11.
[29] Alibeigloo A. Static analysis of functionally graded carbon nanotube-
reinforced composite plate embedded in piezoelectric layers by using the-
ory of elasticity. Compos Struct 2013;95:612e22.
[30] Wu CP, Chang SK. Stability of carbon nanotube-reinforced composite plates
with surface-bonded piezoelectric layers and under bi-axial compression.
Compos Struct 2014;111:587e601.
[31] Jalali S, Heshmati M. Buckling analysis of circular sandwich plates with
tapered cores and functionally graded carbon nanotubes-reinforced com-
posite face sheets. Thin Walled Struct 2016;100:14e24.
[32] Shahriari B, Karamooz Ravari M, Zeighampour H. Vibration analysis of func-
tionally graded carbon nanotube-reinforced composite nanoplates using
Mindlin's strain gradient theory. Compos Struct 2015;134:1036e43.
[33] Jooybar N, Malekzadeh P, Fiouz A. Vibration of functionally graded carbon
nanotubes reinforced composite truncated conical panels with elastically
restrained against rotation edges in thermal environment. Compos Part B Eng
2016;106:242e61.
[34] Shen HS. Thermal buckling and postbuckling behavior of functionally graded
carbon nanotube-reinforced composite cylindrical shells. Compos Part B Eng
2012;43(3):1030e8.
[35] García-Macías E, Rodriguez-Tembleque L, Castro-Triguero R, S aez A. Buckling
analysis of functionally graded carbon nanotube-reinforced curved panels
under axial compression and shear. Compos Part B Eng 2017;108:243e56.
[36] Sobhani Aragh B, Nasrollah Barati A, Hedayati H. Eshelbya ^“Mori^a“Tanaka
approach for vibrational behavior of continuously graded carbon nanotube-
reinforced cylindrical panels. Compos Part B Eng 2012;43(4):1943e54.
[37] Fidelus J, Wiesel E, Gojny F, Schulte K, Wagner H. Thermo-mechanical prop-
erties of randomly oriented carbon/epoxy nanocomposites. Compos Part A
Appl Sci Manuf 2005;36(11):1555e61.
[38] Shen HS. Nonlinear bending of functionally graded carbon nanotube-
reinforced composite plates in thermal environments. Compos Struct
2009;91(1):9e19.
[39] Reddy J. Mechanics of laminated composite plates and shells: theory and
analysis. CRC Press; 2004.
[40] Don O Brush, Almroth BO. Buckling of bars, plates and shells. McGraw-Hill,
Inc; 1975.
[41] Shariat BS, Javaheri R, Eslami M. Buckling of imperfect functionally graded
plates under in-plane compressive loading. Thin Walled Struct 2005;43(7):
1020e36.
[42] Thang PT, Nguyen-Thoi T, Lee J. Closed-form expression for nonlinear analysis
of imperfect sigmoid-FGM plates with variable thickness resting on elastic
medium. Compos Struct 2016;143:143e50.
[43] Samsam Shariat B, Eslami M. Thermal buckling of imperfect functionally
graded plates. Int J Solids Struct 2006;43(14e15):4082e96.
[44] Javaheri R, Eslami MR. Buckling of functionally graded plates under in-plane
compressive loading. ZAMM - J Appl Math Mech/Zeitschrift für Angewandte
Math und Mech 2002;82(4):277e83.
[45] Shen HS. Postbuckling of FGM plates with piezoelectric actuators under
thermo-electro-mechanical loadings. Int J Solids Struct 2005;42(23):6101e21.
[46] Zhang L, Liew K, Reddy J. Postbuckling of carbon nanotube reinforced func-
tionally graded plates with edges elastically restrained against translation and
rotation under axial compression. Comput Meth Appl Mech Eng 2016;298:
1e28.