Thin-Walled Structures 112 (2017) 149–158

Contents lists available at ScienceDirect

Thin–Walled Structures
journal homepage: www.elsevier.com/locate/tws

Full length article

Axisymmetric post-buckling behavior of saturated porous circular plates

⁎
MARK
M.R. Feyzi, A.R. Khorshidvand
Department of Mechanical Engineering, Islamic Azad University, South Tehran Branch, 11365/4435 Tehran, Iran

A R T I C L E I N F O A B S T R A C T

Keywords: This study aimed to investigate axisymmetric post-buckling behavior of a circular plate made of porous material
Post-buckling under uniformly distributed radial compression with simply supported and clamped boundary conditions. Pores
Porous material are saturated with fluid and plate properties vary continuously in the thickness direction. Governing equations
Circular plate are obtained based on classical plate theory and applying Sanders nonlinear strain-displacement relation.
Shooting method
Shooting numerical method is used to solve the governing equations of problem. Effects of porosity coefficient,
pore distribution, pore fluid compressibility, thickness change and boundary conditions on the post-buckling
behavior of the plate are investigated. The results obtained for post-bucking of homogeneous/isotropic plates
and critical buckling load of porous plates are compared with the results of other researchers.

1. Introduction
Porous materials are composed of two components; a solid matrix
and fluid within matrix pores that can be liquid or gas. Porous materials
exist in nature such as stone, wood and bone and may be made
artificially such as metal, ceramic and polymer foams and they are used
as structural components in various industries such as aerospace,
transportation, building, etc.
Biot [1] who is the pioneer in developing the theory of poroelas-
ticity, studied buckling of a fluid-saturated porous slab under axial
compression and showed that critical buckling load is proportional to
pore compressibility. Magnucki and Stasiewicz [2] investigated buck-
ling of a simply supported porous beam and showed porosity effect on
the strength and buckling load of the beam. Buckling and bending of a
rectangular porous plate with varying properties across the thickness
and under in-plane compression and transverse pressure were studied
by Magnucki et al. [3]. Magnucka-Blandzi [4] examined buckling of a
circular porous plate and showed that the critical load linearly decrease
with increase porosity of the plate; he also studied dynamic stability of
a circular plate made of metal foam and showed porosity effect on
critical loads with numerical results [5]. He continued his research in
this field and investigated rectangular sandwich plate with metal foam
core and simply supported boundary condition. In this study, he
considered the middle plane of the plate as symmetry plane and by
numerical method obtained critical buckling loads for a set of sandwich
plates [6]. Jasion et al. [7] obtained global and local buckling for
sandwich beam and plate with metal foam core by experimental,
numerical and analytical methods, and compared the obtained results.
Wen [8] obtained an analytical solution for saturated porous plate with
an incompressible fluid and showed that there is a significant interac-
tion between the solid and flow. Closed-form solution for buckling of
porous circular plate saturated with fluid and with three different types
of pore distribution in thickness direction including nonlinear nonsym-
metric, nonlinear symmetric and monotonous and based on classical
plate theory (CPT) under mechanical and thermal loads was obtained
by Jabbari et al. [9,10]. Buckling of porous circular plates integrated
with piezoelectric layers was investigated under mechanical and
thermal loads and based on CPT by [11–13] and under thermal load
and based on first-order shear deformation plate theory by [14].
Buckling analysis of porous plates with functional properties is
similar to functionally graded material (FGM) plates to some extent.
Woo and Meguid [15] obtained an analytical solution in terms of
Fourier series for the coupled large deflection of FG plates and shallow
shells under transverse mechanical loads and a temperature field.
Closed-form solution for the critical buckling temperature of a rectan-
gular FG plate was obtained under different types of thermal loads and
based on classical and higher-order shear deformation plate theories by
Javaheri and Eslami [16,17]. They showed that classical plate theory
overestimates buckling temperatures. They also examined Buckling of
FG Plates under in-plane Compression and based on CPT [18].
Najafizadeh and Eslami [19] presented buckling of a circular plate
with functional properties under uniform radial compression. Ma and
Wang [20] investigated axisymmetric post-buckling of an FG circular
plate under uniformly distributed radial compressive load. They also
[21] studied bending and thermal post-buckling of an FG circular plate
based on classical nonlinear von Karman plate theory. The governing

⁎
Corresponding author.
E-mail address: ar_khorshidvand@azad.ac.ir (A.R. Khorshidvand).

http://dx.doi.org/10.1016/j.tws.2016.11.026
Received 10 June 2016; Received in revised form 24 November 2016; Accepted 30 November 2016
0263-8231/ © 2016 Elsevier Ltd. All rights reserved.
M.R. Feyzi, A.R. Khorshidvand Thin-Walled Structures 112 (2017) 149–158

equations were solved with the shooting method. Effects of material

constant and boundary conditions on temperature distribution, non-
linear bending, critical buckling temperature and thermal post-buckling
behavior were indicated. They [22] studied the relationships between
axisymmetric bending and buckling of FG circular plates based on
third-order plate theory (TPT) and CPT. They compared TPT solutions
with first-order plate theory (FPT) and CPT solutions and showed that
TPT solutions are almost the same as FPT solutions and FPT is sufficient
to consider shear deformation effect on the axisymmetric bending and
buckling of FG plate. Post-buckling of FG circular plate with geometric
imperfection under transverse mechanical load and transversely non-
uniform temperature rise was studied by Li et al. [23]. Effects of
geometric imperfections on buckling and post-buckling of FG plates
were studied by many researchers such as [24–26]. Post-buckling of an
FG circular plate under asymmetric transverse and in-plane loadings
was studied by Fallah et al. [27].
To the authors' knowledge, there is no study in previous works Fig. 1. Variation of shear modulus through the dimensionless thickness for nonlinear
regarding post-buckling of porous circular Plates. Therefore in this nonsymmetric pore distribution for different values of e1.
study, mechanical post-buckling of a saturated porous circular plate is
investigated. It is assumed that mechanical properties vary continu-
ously in the thickness direction. Based on classical plate theory and
Sanders assumption, governing equations of the problem are obtained
as a system of differential equations and shooting method is used to
solve them. Both clamped and simply supported boundary conditions
are considered. The effects of geometric parameters, poroelastic
material parameters and boundary conditions on the post-buckling
behavior of plate are investigated.

2. Governing equations

2.1. Mechanical properties of poroelastic plates

In this study, a circular plate with radius b and thickness h is

considered, which is made of porous material and its pores are
saturated with fluid. Cylindrical coordinate axes are located on the
mid-plane of the plate and z axis is in the thickness direction. Plate
properties vary continuously along the thickness. For pore distribution Fig. 2. Variation of shear modulus through the dimensionless thickness for nonlinear
in the thickness direction, three different cases are considered [2,9,10]. symmetric pore distribution for different values of e1.
At first case, pore distribution is nonlinear symmetric and the middle
plane of plate is its symmetry plane and moduli of elasticity, which extensional strains, but in the symmetrical and monotonous cases, such
depend on pore distribution, are as follows couplings are neglected [14]. Variations of shear modulus with porous
⎡ distribution in the thickness direction are shown in Fig. 1 for nonlinear
⎛ z ⎞⎤ ⎡ ⎛ z ⎞⎤ E G
E (z ) = E0⎢1−e1cos⎜π ⎟⎥G(z ) = G0⎢1−e1cos⎜π ⎟⎥e1=1− 1 =1− 1 nonsymmetric distribution, in Fig. 2 for nonlinear symmetric distribu-
⎣ ⎝ h ⎠⎦ ⎣ ⎝ h ⎠⎦ E0 G0 (1) tion and in Fig. 3 for monotonous distribution.
where e1 is the porosity coefficient of the plate (0 < e1<1); E1 and E0 are
Young's moduli at the middle plane (z = 0) and the upper and lower
surfaces of the plate (z = ± h /2), respectively; G1 and G0 are shear
moduli at (z = 0) and (z = ± h /2), respectively. The relationship be-
tween elastic modulus and shear modulus is Ej =2Gj (1 + ν ),j = 0,1 and
Poisson's ratio (ν ) is assumed to be constant across the plate thickness.
At second case, pore distribution is nonlinear nonsymmetric and moduli
of elasticity are expressed as follows
⎧
⎪
⎡⎛ π ⎞⎛ h ⎞⎤⎫ ⎪
⎧ ⎪
⎡⎛ π ⎞⎛ h ⎞⎤⎫ ⎪

E (z ) = E0⎨1−e1cos⎢⎜ ⎟⎜z + ⎟⎥⎬G(z ) = G0⎨1−e1cos⎢⎜ ⎟⎜z + ⎟⎥⎬

⎩
⎪
⎣ ⎝ 2h ⎠ ⎝ 2 ⎠⎦⎭ ⎪
⎩
⎪
⎣⎝ 2h ⎠⎝ 2 ⎠⎦⎭ ⎪

(2)
in which E1 and E0 are Young's moduli at the lower surface (z = − h /2)
and the upper surface (z = + h /2) of the plate, respectively; G1 and G0
are shear moduli at the lower and upper surfaces, respectively. At third
case, pore distribution is monotonous and moduli of elasticity in this
case, are:
E (z ) = E0(1−e1)G (z ) = G0(1−e1) (3)
In the asymmetrical case, there is mechanical coupling between Fig. 3. Variation of shear modulus through the dimensionless thickness for monotonous
extensional forces and curvatures and between bending moments and pore distribution for different values of e1.

150
M.R. Feyzi, A.R. Khorshidvand Thin-Walled Structures 112 (2017) 149–158

2.2. Stress-strain constitutive relation (Nr − Nθ )

Nr , r + =0(rNr W , r − Mθ ) + (rMr ), rr =0
r ,r (13)
Biot [1] theory of linear poroelasticity has two features: (1) an where N and M are force and moment components, respectively and
increase of pore pressure induces a dilation of pore, and (2) compres- are defined as follows
sion of the pore causes a rise of pore pressure. Stress-strain relation for
h /2 h /2
poroelastic materials is expressed as follow [28]: (Nr , Nθ ) = ∫−h/2 (σr, σθ )dz(Mr , Mθ ) = ∫−h/2 (σr, σθ )zdz (14)
2Gνu
σij =2Gεii + εkk δij − αpδij Substitution of Eqs. (8) and (11) into Eqs. (14), yields
1 − 2νu (4)
⎧ Nr ⎫ ⎡ A2 B2 ⎤⎧ εr ⎫ ⎡ A3 B3 ⎤⎧ kr ⎫⎧ Mr ⎫ ⎡ A3 B3 ⎤⎧ εr ⎫
where ⎨ ⎬=⎢ ⎥⎨ ⎬ + ⎢ ⎥⎨ ⎬⎨ ⎬ = ⎢ ⎥⎨ ⎬
⎩ Nθ ⎭ ⎣ B2 A2 ⎦⎩ εθ ⎭ ⎣ B3 A3 ⎦⎩ kθ ⎭⎩ Mθ ⎭ ⎣ B3 A3 ⎦⎩ εθ ⎭
p = M (ξ − αεkk ) (5) ⎡ A B ⎤⎧ k ⎫
+ ⎢ 4 4 ⎥⎨ r ⎬
2G(νu − ν ) ⎣ B4 A4 ⎦⎩ kθ ⎭ (15)
M=
α 2(1 − 2νu )(1 − 2ν ) (6) where
αB(1 − 2ν ) h /2 h /2
ν+ 3 (A2 , A3 , A4 ) = ∫−h/2 A1(1,z, z2)dz(B2, B3, B4) = ∫−h/2 B1(1,z, z2)dz
νu = αB(1 − 2ν )
(16)
1− 3 (7)
By substituting Eqs. (12) and (15) in Eqs. (13), the governing
where p is pore fluid pressure; M is Biot's modulus; νu is undrained equilibrium equations of the problem in terms of displacement compo-
Poisson's ratio (0 < ν<νu<0.5); α is Biot coefficient of effective stress nents are obtained
(0 < α < 1); B is Skempton pore pressure coefficient (marker of pore ⎛ 2 2⎞
fluid properties); ξ is variation of fluid volume content; and εkk is dU 1 dU U d 2W dW 1 ⎛ dW ⎞ ⎟
A2 ⎜⎜ 2 + − 2 + + ⎜ ⎟⎟
volumetric strain. By simplifying Eq. (4) to plane-stress condition in ⎝ dr r dr r 2
dr dr 2r ⎝ dr ⎠ ⎠
cylindrical coordinates and under undrained condition (ξ = 0), follow- ⎛
⎛ d 3W 1 d 2W 1 dW ⎞ ⎛ ⎞2 ⎞
ing relations are obtained [10] + A3 ⎜ − 3 − + ⎟ + B ⎜ − 1 ⎜ dW ⎟ ⎟=0
2 ⎜ ⎟
⎝ dr r dr 2 r 2 dr ⎠ ⎝ 2r ⎝ dr ⎠ ⎠ (17)
σr = A1εr + B1εθ σθ = A1εθ + B1εr (8)
A2 ⎡⎢ dU
2⎤
p = M (−αεkk ) (9) 1 ⎛ dW ⎞ ⎥⎛ d 2W ⎞ B ⎛ U ⎞⎛ d 2W ⎞ A ⎛ U ⎞⎛ 1 dW ⎞
+ ⎜ ⎟ ⎜ 2 ⎟ + 2 ⎜ ⎟⎜ 2 ⎟ + 2 ⎜ ⎟⎜ ⎟
A4 ⎢⎣ dr 2 ⎝ dr ⎠ ⎥⎦⎝ dr ⎠ A4 ⎝ r ⎠⎝ dr ⎠ A4 ⎝ r ⎠⎝ r dr ⎠
where the constants A1 and B1 in terms of the constants C1 and C2 are:
B2 ⎡⎢ dU
2⎤
1 ⎛ dW ⎞ ⎥⎛ 1 dW ⎞ B ⎛ 3 d 2W dW ⎞
⎡ νu νu − ν ⎤ + + ⎜ ⎟ ⎜ ⎟ + 3 ⎜− ⎟
C1 = 2⎢1 + + ⎥G (z ) A4 ⎢⎣ dr 2 ⎝ dr ⎠ ⎥⎦⎝ r dr ⎠ A4 ⎝ r dr 2 dr ⎠
⎣ 1 − 2νu (1 − 2νu )(1 − 2ν ) ⎦
⎡ ⎛ d 2W ⎞2 ⎤ B2A3 ⎛ 1 d 2W dW ⎞
A3 ⎢ 1 ⎛ dW ⎞
2
⎡ (ν − ν )(1+νu ) ⎛ C ⎞⎤
2
)⎢1 + νu + u ⎜1 − 2 ⎟⎥G ( z ) + − 2⎜ ⎟ −⎜ 2⎟⎥+ ⎜ ⎟
C2 = C1 − 2G (z )A1=( 2 ⎢
1 − νu ⎣ 1 − 2ν ⎝ C1 ⎠⎥⎦ A4 ⎢⎣ r ⎝ dr ⎠ ⎝ dr ⎠ ⎥⎦ A4 A2 ⎝ r dr 2 dr ⎠

⎡ (ν − ν )(1+νu ) ⎛ C ⎞⎤ ⎛ A3 A3 ⎞⎛ d 4W 2 d 3W 1 d 2W 1 dW ⎞
2 = ⎜1− ⎟⎜ 4 + ⎟
B1=( )⎢(1+νu )νu + u ⎜ 1 − 2 ⎟⎥G (z ) − 2 + 3
2 ⎢
1 − νu ⎣ 1 − 2ν ⎝ C1 ⎠⎥⎦ (10) ⎝ A4 2 ⎠⎝ dr
A r dr 3
r dr 2
r dr ⎠ (18)
The continuity and symmetry conditions at the center of plate
(r = 0) are

2.3. Strain-displacement relations dW ⎛ d 3W 1 d 2W ⎞

U =0, =0, W =Finite, lim⎜ 3 + ⎟=0
dr r→0⎝ dr r dr 2 ⎠ (19)
Based on classical plate theory and according to [29], strain
components at distance z from the middle plane are determined by The plate edge in r direction is movable and in z direction is
following relations clamped or simply supported. Boundary conditions at the edge of the
plate (r = b ) for clamped case, are
εr = εr + zkrεθ = εθ + zkθ (11)
dW
W =0, =0, Nr =−P
where εr , εθ are engineering strain components in the middle plane dr (20)
and kr , kθ are curvatures. These strains and curvatures, in terms of and boundary conditions at the edge of the plate (r = b ) for simply
displacement components and according to the Sanders assumption are supported case, are
[30]
W =0, Mr =0, Nr =−P (21)
1 U 1
εr = U ,r + (W , r )2 εθ = kr = −W , rrkθ =− W , r For making the governing equations and boundary conditions
2 r r (12)
dimensionless, following dimensionless parameters are defined [10,20]
where W and U are displacement components of a point on the mid-
plane of the plate in z and r directions, respectively and Comma in r W Ub A3 h A2 h 2
x= , w= , u= , f1 = , f2 = ,
subscript denotes differentiation. b h h2 A4 A4
A3 B h2 B3h B2 B
f3 = f4 = 2 , f5 = , f6 = ,f = 4 ,
2.4. Governing equilibrium equations A2 h A4 A4 A2 7 A4
B3 b 2P P b 2Mr
Governing equilibrium equations of a thin circular plate under f8 = λ= 2 , P* = , Mr* =
A2 h h A2 G0b hA4 (22)
radial uniform compressive load and with polar symmetry condition
utilizing principle of minimum total potential energy are presented as Dimensionless governing equations and boundary conditions are
follows [10] derived as follows:

151
M.R. Feyzi, A.R. Khorshidvand Thin-Walled Structures 112 (2017) 149–158

d 2u 1 du u d 2w dw 1 ⎛ dw ⎞
2 ⎛ d 3w 1 d 2w 1 dw ⎞ 2 1 1 ⎧ ⎛ 1 ⎞ ⎛1 ⎞ ⎛1 ⎞
+ − 2 + + ⎜ ⎟ − f3 ⎜ 3 + − 2 ⎟ ω=− y4 + 2 y3 − 3 y2 + ⎨f4 ⎜y6 + y22⎟y3 + f2 ⎜ y5y3⎟ + f4 ⎜ y5y2⎟
dx 2
x dx x 2
dx dx 2x ⎝ dx ⎠ ⎝ dx x dx 2 x dx ⎠ x x x ⎩ ⎝ 2 ⎠ ⎝x ⎠ ⎝x ⎠
⎛ ⎛ ⎞2 ⎞ ⎛ 1 ⎞⎛ 1 ⎞ ⎛1 ⎞ ⎛1 ⎞⎫
1 dw + f2 ⎜y6 + y22⎟⎜ y2⎟ + (f2 f3 − 3f5 )⎜ y3y2⎟ − f1 ⎜ 2 y22 + y32⎟⎬
− f6 ⎜⎜ ⎜ ⎟ ⎟⎟=0 ⎝ 2 ⎠⎝ x ⎠ ⎝x ⎠ ⎝x ⎠⎭
⎝ x ⎝ dx ⎠ ⎠
2 (23)
/ (1 − f1 f3 ) (31)
⎡ 2⎤
du 1 ⎛ dw ⎞ ⎛ d 2w ⎞ ⎛ u ⎞⎛ d 2w ⎞ ⎛ u ⎞⎛ 1 dw ⎞
f2 ⎢ + ⎜ ⎟ ⎥⎜ 2 ⎟ + f4 ⎜ ⎟⎜ 2 ⎟ + f2 ⎜ ⎟⎜ ⎟ 1 1 1 2 ⎛ 1 1 ⎞ ⎛1 ⎞
⎢⎣ dx 2 ⎝ dx ⎠ ⎥⎦⎝ dx ⎠ ⎝ x ⎠⎝ dx ⎠ ⎝ x ⎠⎝ x dx ⎠ μ=− y6 + 2 y5 − y3y2 − y2 + f3 ⎜y4 + y3 − 2 y2⎟ + f6 ⎜ y22⎟
x x 2x ⎝ x x ⎠ ⎝ 2x ⎠ (32)
⎡ 2⎤ ⎡ ⎛ ⎞2 ⎛ 2 ⎞2 ⎤
du 1 ⎛ dw ⎞ ⎛ 1 dw ⎞ 1 dw dw
+ f4 ⎢ + ⎜ ⎟ ⎥⎜ ⎟ − f1 ⎢ 2 ⎜ ⎟ + ⎜ 2 ⎟ ⎥ for given load case
⎢⎣ dx 2 ⎝ dx ⎠ ⎥⎦⎝ x dx ⎠ ⎢ x ⎝ dx ⎠ ⎝ dx ⎠ ⎥⎦
⎣ ⎡0 1 0 0 0 0⎤ ⎧ 0⎫
⎛ 1 d 2w dw ⎞ ⎛ d 4w ⎢ ⎥ ⎪ ⎪
2 d 3w 1 d 2w 1
B0 = ⎢ 0 0 Δx 1 0 0 ⎥ , b0 = ⎨ 0 ⎬
+ (f2 f3 −3f5 )⎜ ⎟ = (1−f1 f3 )⎜ 4 + − 2 2 ⎪ ⎪
⎝ x dx dx ⎠
2
⎝ dx x dx 3
x dx ⎢⎣ 0 0 0 0 1 0 ⎥⎦ ⎩0⎭
(33)
1 dw ⎞
+ 3 ⎟ ⎡1 0 0 0 0 0⎤
x dx ⎠ ⎢ ⎧0⎪ ⎫
(24)
B1 = ⎢ 0 1 0 0 0 0 ⎥, b = ⎪
⎨ 0 ⎬,(Clamped)
y2 f8 f
⎥ 1 ⎪ ⎪

dw ⎛ d 3w 1 d 2w ⎞ dw ⎢⎣ 0 − −f3 0 6 1 ⎥⎦ ⎩− λ ⎭
u=0, =0, w = δ, lim ⎜ + ⎟=0, at x =0 w=0, =0,
2 x x (34)
dx x →0⎝ dx 3 x dx 2 ⎠ dx
⎡1 0 0 0 0 0⎤
Nr =−λ,atx =1 (Clamped)w=0, Mr*=0, ⎢ f1 y2 f7 f ⎥ ⎧
⎪
0⎫ ⎪
B1 = ⎢ 0 2
− x
−1 0 5 f1⎥ , b = ⎨
x 0⎬ ,(Simply Supported)
Nr =−λ , at x =1 (Simply supported) (25) ⎢ ⎥ 1 ⎪ ⎪
⎢0
y2 f8 ⎩ − λ ⎭
1 ⎥⎦
f6
⎣ − −f3 0
where δ is a dimensionless parameter to determine deflection at the 2 x x (35)
center of the plate and Nr , Mr* are for given deflection case
1 ⎛ dw ⎞
2 ⎛ d 2w ⎞ ⎛u⎞ ⎛ 1 dw ⎞ ⎡1
Nr =
du
+ ⎜ ⎟ − f3 ⎜ 2 ⎟ + f6 ⎜ ⎟ − f8 ⎜ ⎟M * 0 0 0 0 0⎤ ⎧δ ⎫
2 ⎝ dx ⎠ ⎝x⎠ ⎝ x dx ⎠ r ⎢0 1 0 0 0 0⎥ ⎪0⎪
dx ⎝ dx ⎠
B0 = ⎢ ⎥, b0 = ⎨ ⎬
⎛ 2⎞ ⎢0 0 1 1 0 0⎥ ⎪0⎪
du 1 ⎛ dw ⎞ d 2w ⎛u⎞ ⎛ 1 dw ⎞
⎢⎣
Δx
⎥
= f1 ⎜⎜ + ⎜ ⎟ ⎟⎟ − + f5 ⎜ ⎟ − f7 ⎜ ⎟ 0 0 0 0 1 0⎦ ⎩0⎭
(36)
⎝ dx 2 ⎝ dx ⎠ ⎠ dx 2 ⎝ x ⎠ ⎝ x dx ⎠ (26)
⎡ ⎤ ⎧ ⎫
This problem can be expressed in two different ways; in the first B1 = ⎢ 1 0 0 0 0 0 ⎥ , b1 = ⎨ 0 ⎬,(Clamped)
case, deflection of the plate center (δ ) is given and Nr =−λ is removed ⎣0 1 0 0 0 0⎦ ⎩0⎭ (37)
from boundary conditions in Eqs. (25). In the second case, applied load
⎡1 0 0 0 0 0⎤ ⎧ ⎫
on the edge of the plate (λ ) is given and w = δ is removed in Eqs. (25). B1 = ⎢ f1 y2 f7 f ⎥ , b = ⎨ 0 ⎬,(Simply Supported)
⎢⎣ 0 − −1 0 5 f1⎥⎦ 1 ⎩ 0 ⎭
2 x x (38)
3. Numerical method
Initial value problem (IVP) related to the boundary value problem is
In order to solve the governing differential equations, as a boundary expressed as
value problem (BVP), shooting method [20,21,31,32] is used. The dZ
= H (x , Z )
governing equilibrium equations of the problem as a system of two dx (39)
coupled differential equations for numerical solution, are written as a
T
system of six first-order differential equations. In order to avoid the Z(Δx ) = { z1 z2 z3 z 4 z5 z 6 } = I (40)
singularities at x = 0 , the small quantity ∆x (∆x > 0) which is close to T
zero is considered, thus the problem is solved on the interval [∆x, 1]. I (D) = { d1 0 d 2 −d 2 /Δx 0 d3} ,
Boundary value problem is rewritten in the following standard form (λGiven )I (δ,D) = { δ 0 d1 −d1/Δx 0 d 2 }T , (δGiven ) (41)
using the vector/matrix notations
where I is initial values vector and D is unknown initial values vector
dY and is defined as
= H (x , Y ) ,
dx (27)
T
D = { d1 d 2 d3 } , (λGiven )D = { d1 d 2 }T , (δGiven ) (42)
B0 Y (∆x ) = b0 , B1Y (1) = b1 (28)
The aim is to find values for D so that Eq. (43) be satisfied, these
where
values are indicated with D*.
⎧ y1 ⎫ ⎧ w ⎫
⎪ y ⎪ ⎪ dw / dx ⎪ B1Z(1; D*) − b1 = 0 , (λGiven )B1Z(1; δ, D*) − b1 = 0 , (δGiven ) (43)
⎪ 2⎪ ⎪ 2 2⎪
⎪y ⎪ ⎪ ⎪ D* is assumed as the root of Eq. (43) and a Newton-Raphson numerical
Y = ⎨ y3 ⎬ = ⎨ d 3w / dx 3 ⎬
⎪ 4 ⎪ ⎪ d w / dx ⎪ method is used to obtain it. In order to solve the IVP (Eqs. (39) and
⎪ y5 ⎪ ⎪ u ⎪ (40)), the fourth-order Runge-Kutta method is used.
⎪y ⎪ ⎪ ⎪
⎩ 6 ⎭ ⎩ du / dx ⎭ (29) In this case, the solution of IVP with initial values I in x = ∆x is the
same as the solution of BVP (Eqs. (27) and (28)), since such solution
⎧ y2 ⎫ ⎧ dw / dx ⎫ satisfies boundary conditions in the edge of the plate (x = 1), Thus
⎪ y ⎪ ⎪ d 2w / dx 2 ⎪
⎪ ⎪ ⎪
⎪y ⎪
3
⎪ ⎪ 3 3⎪ Y (x ) = Z(x; D*) , (λGiven )Y (x; δ ) = Z(x; δ, D*) , (δGiven ) (44)
H (x, Y ) = ⎨ 4 ⎬ = ⎨ d 4w / dx 4 ⎬
⎪ ω ⎪ ⎪ d w / dx ⎪ In order to achieve the post-buckling equilibrium paths of plate, in
⎪ y6 ⎪ ⎪ du / dx ⎪
⎪ μ⎪ ⎪ ⎪ given deflection case, a small value is considered for δ and after solving
⎩ ⎭ ⎩ 2
d u / dx 2 ⎭ (30) BVP, applied load proportional to the deflection of the plate center, in

152
M.R. Feyzi, A.R. Khorshidvand Thin-Walled Structures 112 (2017) 149–158

Table 1
Comparisons of the results of the present numerical method with results obtained from
analytical method by [10], for critical buckling load (P* × 10 5 ) . (ν = 0.3 , B = 0 ).

Materials e1 h /b Clamped plate Simply supported plate

Reference Present Reference Present

Homogenous & 0 0.01 0.3496 0.3496 0.1 0.0999

isotropic 0.02 2.7971 2.7965 0.8005 0.7996
0.03 9.4404 9.4381 2.7018 2.6986
Porous/ nonlinear, 0.3 0.01 0.2852 0.2851
nonsymmetrical 0.02 2.2815 2.2808
distribution 0.03 7.6999 7.6978
0.5 0.01 0.2364 0.2364
0.02 1.8916 1.8910
0.03 6.3842 6.3822
0.7 0.01 0.1794 0.1793
0.02 1.4353 1.4346
0.03 4.8442 4.8425

Fig. 4. Post-buckling paths of a saturated porous plate for clamped boundary condition
and for different values of e1, and comparison of homogeneous/isotropic case (e1=0 ) with
equilibrium state, is obtained. Then, with continuation method δ is
results of [20]. (B = 0, ν = 0. 3 ).
increased step by step and results of each step are used as an initial
guess for next step.

4. Numerical results and discussions

In this paper, post-buckling behavior of a saturated porous circular

plate was studied, and effects of geometric parameters, poroelastic
material parameters and boundary conditions on the post-buckling
behavior were investigated.
Comparisons of obtained results from the present numerical method
with obtained results from the analytical method [10] for critical
buckling load are shown in Tables 1, 2. In Table 1, critical buckling load
for a saturated porous plate under clamped boundary condition and a
homogeneous/isotropic plate under both clamped and simply sup-
ported boundary conditions is shown for different values of porosity
coefficients and thickness ratios. In Table 2, critical buckling loads for a
porous plate under clamped boundary condition and for different
values of Skempton coefficients are presented. Also, obtained results
for the homogeneous/isotropic plate are compared, in Figs. 4 and 5 for
Fig. 5. Post-buckling paths of a saturated porous plate for simply supported boundary
post-buckling paths and in Fig. 8 for post-buckling configurations, with
condition and for different values of e1, and comparison of homogeneous/isotropic case
results of [20]. There are good agreements between obtained results (e1=0 ) with results of [20]. (B = 0, ν = 0. 3 ).
from the present numerical method with existing results in the
literature that confirms validity of the present numerical method.
conditions, are shown. It is observed that the post-buckling strength
of the homogeneous/isotropic plate (e1=0 ), is more than the porous
4.1. Porosity plate. Increasing the porosity coefficient decreases the post-buckling
strength. The greater the value of deflection of the plate, the more
In this section, the effect of porosity coefficient on post-buckling considerable the effect of porosity coefficient on plate strength reduc-
behavior of a saturated porous plate with nonlinear nonsymmetric pore tion. Also, with comparison of two figures it can be found that for each
distribution is investigated. specified value of dimensionless central deflection (W (0)/ h ), the corre-
In Figs. 4 and 5 post-buckling equilibrium paths for different sponding post-buckling load for the clamed case in comparison with the
porosity coefficients and clamped and simply supported boundary simply supported case is more than double.
Fig. 6 is composed of four magnified views of Fig. 4. As it can be
Table 2 seen, the homogeneous/isotropic plate has a strictly increasing post-
Comparisons of the results of the present numerical method with results obtained from buckling path, but in the porous plate, the post-buckling load does not
analytical method by [10], for critical buckling load (P* × 107 ) and for clamped boundary
increase monotonically with increasing the central deflection and post-
conditions. (ν = 0. 3,e1=0. 5, h /b = 0. 01) .
buckling loads for small deflections are lower than critical buckling
Materials B Reference Present load. In other words, in the porous plate after the buckling, applied load
decreases and then increases with deflection increasing. From Fig. 6, it
Porous/nonlinear, symmetric distribution 0.3 29.80428 29.3065
can be concluded that post-buckling load reduction value (in small
0.5 30.52959 29.7422
0.7 31.21832 30.168 deflections) decreases with decreasing of porosity coefficient and in
Porous/nonlinear, nonsymmetrical distribution 0.3 24.65218 24.6483 homogeneous/isotropic plate tends to zero.
0.5 25.28168 25.2777 Fig. 7 is a magnified view of Fig. 5. According to this figure, it can
0.7 25.88022 25.8762 be found that in the porous plate with the simply supported boundary
Porous/monotonous distribution 0.3 18.20553 18.7322
0.5 18.65789 19.521
condition, unlike the clamped case, buckling does not occur but from
0.7 19.08871 20.2751 the beginning of loading the plate starts bending. Buckling behavior of

153
M.R. Feyzi, A.R. Khorshidvand Thin-Walled Structures 112 (2017) 149–158

Fig. 6. Zoomed views of Fig. 4.

Fig. 8. Post-buckling configurations of a saturated porous plate for clamped boundary
condition and for different values of e1, and comparison of homogeneous/isotropic case
(e1=0 ) with results of [20]. (B = 0, ν = 0. 3 ).

Fig. 7. Zoomed view of Fig. 5.

homogeneous/isotropic plate under simply supported boundary condi-

tion is the same as clamped boundary condition. Therefore, in the
circular plate with simply supported edge, buckling occurs only for the
case e1=0 (i.e. homogeneous/isotropic plate) and for the case e1≠0 (i.e.
porous plate) bending occurs.
Fig. 9. Post-buckling configurations of a saturated porous plate for simply supported
In Figs. 8 and 9, for each of loads λ = 20,100 , the post-buckling boundary condition and for different values of e1. (B = 0, ν = 0. 3 ).
configurations of the saturated porous plate are shown for different
porosity coefficients and for clamped and simply supported boundary
Fig. 12 is a magnified view of Fig. 10 for the small values of applied
conditions, respectively. As it can be seen, deflection of porous plate is
load. This figure shows that in the porous plate with clamped boundary
more than homogeneous/isotropic plate, and increasing the porosity
condition, bending moment at the center of the plate has a negative
coefficient increases the plate deflection. Also, deflection of simply
value in the beginning of buckling, and after the buckling when the
supported plate is more than that of clamped plate. In addition, as
post-bucking load reduces to its minimum value, central bending
shown in these figures, with increase of applied load, curvatures of the
moment tends to zero, then both of them (i.e. post-buckling load and
post-buckling configurations curves decrease near the plate center and
central bending moment) increase with increasing the deflection; But
increase near the plate edge.
homogeneous/isotropic plate has a strictly increasing post-buckling
Variations of the bending moments at the center of the plate versus
path and load reduction does not occur after the buckling, therefore
applied load on the edge of the plate for different values of porosity
central bending moment does not have a negative value.
coefficient are studied in Figs. 10 and 11 for clamped and simply
Magnified view of Fig. 11 is plotted for the small values of applied
supported boundary conditions, respectively. It can be observed that, as
load in Fig. 13. As it can be seen, in the porous plate with simply
the plate starts buckling, bending moment at the plate center increases
supported edge from beginning of loading, central bending moment
rapidly and then decreases slowly with increase of the applied load.
increases slowly and then increases rapidly with increasing the post-
This decline occurs because of the decline of curvature near the plate
buckling load. Homogeneous/isotropic plate with simply supported
center (as mentioned above).
boundary condition behaves like clamped boundary condition thus

154
M.R. Feyzi, A.R. Khorshidvand Thin-Walled Structures 112 (2017) 149–158

Fig. 10. Central bending moments Vs. applied load for clamped plate and for different Fig. 13. Zoomed view of Fig. 11.
values of e1. (B = 0, ν = 0. 3 ).

Fig. 14. Variations of applied load (P*) Vs. thickness to radial ratio for clamped plate and
Fig. 11. Central bending moments Vs. applied load for simply supported plate and for for different values of e1 for each of two cases W (0 )/h = 3, 7 . (B = 0, ν = 0. 3 ).
different values of e1. (B = 0 , ν = 0. 3 ).

Fig. 15. Variations of applied load (P*) Vs. thickness to radial ratio for simply supported
Fig. 12. Zoomed view of Fig. 10.
plate and for different values of e1 for each of two cases W (0 )/h = 3, 7 . (B = 0, ν = 0. 3 ).

starts buckling in a certain critical load, then it's bending moment

4.2. Thickness
increases.
In Figs. 14 and 15 for each of the dimensionless central deflections
W (0)/ h=3,7, curves of post-buckling load versus thickness to radial ratio

155
M.R. Feyzi, A.R. Khorshidvand
Fig. 16. Post-buckling configurations of a saturated porous plate for clamped boundary
condition and for different pore distributions with e1=0. 3 and comparison with
homogeneous/isotropic plate. ( λ = 100, B = 0, ν = 0. 3 ).
are shown for a saturated porous plate with nonlinear nonsymmetric
pore distribution and for different porosity coefficients and for clamped
and simply supported boundary conditions, respectively. According to
the results presented in these figures, increasing the thickness increases
the post-buckling strength; in other words, to create a certain deflec-
tion, by increasing the plate thickness, more load must be applied on
the edge of the plate. It is also clear that, increasing the porosity
coefficient, decreases the post-buckling strength of the plate.
4.3. Pore distribution
In Figs. 16 and 17, the effect of pore distribution on post-buckling
behavior of a saturated porous plate is investigated for clamped and
simply supported boundary conditions and is compared with a homo-
geneous/isotropic plate. The plates are subjected to the post-buckling
load λ = 100 . These figures show that the porous plate with any pore
distribution has lower post-buckling strength than the homogeneous/
isotropic plate. Also, it can be seen that the post-buckling strength of
the porous plate is highest in the nonlinear symmetric distribution case,
because near the upper and lower surfaces of the plate, porosity tends to
zero. For example, this phenomenon is also observed in wide-flange
beams (W-beams) that a large proportion of cross section is in farthest
distance from the neutral axis and near the upper and lower surfaces.
Fig. 17. Post-buckling configurations of a saturated porous plate for simply supported
boundary condition and for different pore distributions with e1=0. 3 and comparison with
homogeneous/isotropic plate. ( λ = 100, B = 0, ν = 0. 3 ).
156
Thin-Walled Structures 112 (2017) 149–158
Fig. 18. Variation of applied load ( λ ) with coefficient of porosity (e1) for a plate with
clamped boundary condition and for different cases of pore distributions.
(W (0 )/h = 7 , B = 0, ν = 0. 3 ).
Fig. 19. Variation of applied load ( λ ) with coefficient of porosity (e1) for a plate with
simply supported boundary condition and for different cases of pore distributions.
(W (0 )/h = 7 , B = 0, ν = 0. 3 ).
Effect of porous change on the post-buckling behavior of the porous
plate for different pore distributions and for clamped and simply
supported boundary conditions is shown in Figs. 18 and 19, respec-
tively; It can be found that with increasing porosity coefficient, the
post-buckling strength of the porous plate with any pore distributions
decreases almost linearly.
4.4. Pore fluid properties
Under the undrained condition (ξ=0), that fluid within porous solid
cannot escape, Skempton pore pressure coefficient depends on pore
fluid compressibility and lies in the interval (0,1). So that, if pore fluid
compressibility increases (B → 0), saturated porous plate behaves like a
porous plate without fluid under the drained condition, and if pore fluid
compressibility decreases (B → 1), plate behavior will be close to a rigid
body [10].
In order to investigate the effect of pore fluid compressibility on
post-buckling behavior of a saturated porous plate with different pore
distributions, Figs. 20 and 21 are studied. It can be observed that by
decreasing the pore fluid compressibility, the post-buckling strength of
plate increases. With the aim of increasing the strength of the porous
plate versus post-buckling load, appropriate distribution of pores in the
thickness direction, is more effective than decreasing the pore fluid
compressibility.
M.R. Feyzi, A.R. Khorshidvand
Fig. 20. Variation of applied load ( λ ) with Skempton coefficient ( B ) for a saturated
porous plate for clamped boundary condition and different pore distributions.
(e1=0. 5, W (0 )/h = 7 , ν = 0. 3 ).
Fig. 21. Variation of applied load ( λ ) with Skempton coefficient ( B ) for a saturated
porous plate for simply supported boundary condition and different pore distributions.
(e1=0. 5, W (0 )/h = 7 , ν = 0. 3 ).
5. Conclusions
Post-buckling of a saturated porous circular plate was studied under
uniform radial compression. It was assumed that mechanical properties
vary continuously in thickness direction. Pore distribution in the
thickness direction was considered by three different functions. Based
on classical plate theory and Sanders assumption, governing equili-
brium equations of the problem in terms of the displacement compo-
nents and as a system of coupled nonlinear differential equations were
obtained. Boundary conditions were considered clamped and simply
supported. In order to solve this boundary value problem, a shooting
method in conjunction with a Newton-Raphson method was employed.
The effects of boundary conditions, geometrical parameters and
poroelastic material parameters on post-buckling of the plate were
expressed. The conclusions of this study are presented as follows:
1) By decreasing the porosity coefficient, the post-buckling behavior
of porous plate will be close to the homogeneous/isotropic plate.
2) By increasing the porosity coefficient, the post-buckling strength of
plate with different pore distributions decreases almost linearly.
3) Increasing deflection of the plate increases the effect of porosity
coefficient on the post-buckling strength reduction.
4) A plate with symmetric pore distribution in the thickness direction,
157
Thin-Walled Structures 112 (2017) 149–158
in which porosity value decreases near the top and bottom
surfaces, shows more post-buckling strength compared to other
pore distributions.
5) In a porous plate with clamped boundary condition, after buckling,
applied load decreases in small deflections and then increases with
increasing the deflection.
6) Load reduction after buckling is directly proportional to the
porosity coefficient.
7) In a porous plate with the simply supported boundary condition
buckling does not occur, unlike the clamped boundary condition,
but from the beginning of loading the plate starts bending so
deflection increases slowly and then increases rapidly with in-
creasing the post-buckling load.
8) Bending moment at the center of plate increases rapidly in
beginning of buckling and then decreases slowly with increase of
post-buckling load.
9) In order to form a specified deflection in the plate, the required
post-buckling load for the clamped case in comparison with the
simply supported case is more than double.
10) By increasing plate thickness, post-buckling strength increases
nonlinearly.
11) Increasing pore fluid compressibility reduces post-buckling
strength.
12) In order to increase post-buckling strength of plate, appropriate
distribution of pores in the thickness direction is more effective
than fluid compressibility decline.
Acknowledgement
This work was supported by the Islamic Azad University, South
Tehran Branch, Tehran, Iran.
References
[1] M.A. Biot, Theory of buckling of a porous slab and Its thermoelastic analogy, J.
Appl. Mech. 31 (1964) 194–198.
[2] K. Magnucki, P. Stasiewicz, Elastic buckling of a porous beam, J. Theor. Appl. Mech.
42 (2004) 859–868.
[3] K. Magnucki, M. Malinowski, J. Kasprzak, Bending and buckling of a rectangular
porous plate, Steel Compos. Struct. 6 (2006) 319–333.
[4] E. Magnucka-Blandzi, Axi-symmetrical deflection and buckling of circular porous-
cellular plate, Thin-Walled Struct. 46 (2008) 333–337.
[5] E. Magnucka-Blandzi, Dynamic stability of a metal foam circular plate, J. Theor.
Appl. Mech. 47 (2009) 421–433.
[6] E. Magnucka-Blandzi, Mathematical modelling of a rectangular sandwich plate with
a metal foam core, J. Theor. Appl. Mech. 49 (2011) 439–455.
[7] P. Jasion, E. Magnucka-Blandzi, W. Szyc, K. Magnucki, Global and local buckling of
sandwich circular and beam-rectangular plates with metal foam core, Thin-Walled
Struct. 61 (2012) 154–161.
[8] P.H. Wen, The analytical solutions of incompressible saturated poroelastic circular
Mindlin's plate, J. Appl. Mech. 79 (2012) (051009-051009).
[9] M. Jabbari, M. Hashemitaheri, A. Mojahedin, M.R. Eslami, Thermal buckling
analysis of functionally graded thin circular plate made of saturated porous
materials, J. Therm. Stress. 37 (2014) 202–220.
[10] M. Jabbari, A. Mojahedin, A. Khorshidvand, M. Eslami, Buckling analysis of a
functionally graded thin circular plate made of saturated porous materials, J. Eng.
Mech. 140 (2013) 287–295.
[11] M. Jabbari, E.F. Joubaneh, A. Khorshidvand, M. Eslami, Buckling analysis of porous
circular plate with piezoelectric actuator layers under uniform radial compression,
Int. J. Mech. Sci. 70 (2013) 50–56.
[12] M. Jabbari, A. Mojahedin, E.F. Joubaneh, Thermal buckling analysis of circular
plates made of piezoelectric and saturated porous functionally graded material
layers, J. Eng. Mech. 141 (2014) 04014148.
[13] A. Khorshidvand, E.F. Joubaneh, M. Jabbari, M. Eslami, Buckling analysis of a
porous circular plate with piezoelectric sensor–actuator layers under uniform radial
compression, Acta Mech. 225 (2014) 179–193.
[14] M. Jabbari, E.F. Joubaneh, A. Mojahedin, Thermal buckling analysis of porous
circular plate with piezoelectric actuators based on first order shear deformation
theory, Int. J. Mech. Sci. 83 (2014) 57–64.
[15] J. Woo, S.A. Meguid, Nonlinear analysis of functionally graded plates and shallow
shells, Int. J. Solids Struct. 38 (2001) 7409–7421.
[16] R. Javaheri, M.R. Eslami, Thermal buckling of functionally graded plates, AIAA J.
40 (2002) 162–169.
[17] R. Javaheri, M.R. Eslami, Thermal buckling Of functionally graded plates based On
higher order theory, J. Therm. Stress. 25 (2002) 603–625.
M.R. Feyzi, A.R. Khorshidvand
[18] R. Javaheri, M.R. Eslami, Buckling of functionally graded plates under In-plane
compressive loading, ZAMM - J. Appl. Math. Mech./Z. für Angew. Math. Und Mech.
82 (2002) 277–283.
[19] M.M. Najafizadeh, M.R. Eslami, Buckling analysis of circular plates of functionally
graded materials under uniform radial compression, Int. J. Mech. Sci. 44 (2002)
2479–2493.
[20] L. Ma, T.J. Wang, Axisymmetric post-buckling of a functionally graded circular
plate subjected to uniformly distributed radial compression, Mater. Sci. Forum
Trans. Tech. Publ. (2003) 719–724.
[21] L.S. Ma, T.J. Wang, Nonlinear bending and post-buckling of a functionally graded
circular plate under mechanical and thermal loadings, Int. J. Solids Struct. 40
(2003) 3311–3330.
[22] L.S. Ma, T.J. Wang, Relationships between axisymmetric bending and buckling
solutions of FGM circular plates based on third-order plate theory and classical plate
theory, Int. J. Solids Struct. 41 (2004) 85–101.
[23] S.-R. Li, J.-H. Zhang, Y.-G. Zhao, Nonlinear thermomechanical post-buckling of
circular FGM plate with geometric imperfection, Thin-Walled Struct. 45 (2007)
528–536.
[24] B.A.S. Shariat, R. Javaheri, M.R. Eslami, Buckling of imperfect functionally graded
plates under in-plane compressive loading, Thin-Walled Struct. 43 (2005)
1020–1036.
158
Thin-Walled Structures 112 (2017) 149–158
[25] H.-S. Shen, Postbuckling of FGM plates with piezoelectric actuators under thermo-
electro-mechanical loadings, Int. J. Solids Struct. 42 (2005) 6101–6121.
[26] J. Yang, K.M. Liew, S. Kitipornchai, Imperfection sensitivity of the post-buckling
behavior of higher-order shear deformable functionally graded plates, Int. J. Solids
Struct. 43 (2006) 5247–5266.
[27] F. Fallah, M.K. Vahidipoor, A. Nosier, Post-buckling behavior of functionally graded
circular plates under asymmetric transverse and in-plane loadings, Compos. Struct.
125 (2015) 477–488.
[28] E.Detournay, A.Cheng, Fundamentals of Poroelasticity, Vol. II of Comprehensive
Rock Engineering: Principles, Practice & Projects, chapter 5, in: Pergamon Press,
1993pp. 113–171.
[29] J.N. Reddy, Mechanics of Laminated Composite Plates and Shells: Theory and
Analysis, CRC Press, New York, 2004.
[30] D.O. Brush, B.O. Almorth, Buckling of Bars, Plates, and Shells, McGraw-Hill, New
York, 1975.
[31] S.R. Li, Y.H. Zhou, Shooting method for non-linear vibration and thermal buckling
of heated orthotropic circular plates, J. Sound Vib. 248 (2001) 379–386.
[32] S.R. Li, Y.H. Zhou, X. Song, Non-linear vibration and thermal buckling of an
orthotropic annular plate with a centric rigid mass, J. Sound Vib. 251 (2002)
141–152.