Perturbation

June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
International Journal of Biomathematics

Vol. 3, No. 2 (June 2010) 255–275
c World Scientific Publishing Company
DOI: 10.1142/S1793524510000970
PERISTALTIC FLOW OF A COMPRESSIBLE

NON-NEWTONIAN MAXWELLIAN
FLUID THROUGH POROUS
MEDIUM IN A TUBE
by UNIV OF WISCONSIN-MILWAUKEE on 01/30/15. For personal use only.
Int. J. Biomath. 2010.03:255-275. Downloaded from www.worldscientific.com
ISLAM M. ELDESOKY∗ and A. A. MOUSA†

Department of Basic Engineering Science
Faculty of Engineering, Menofia University
Shebin El-Kom, Menofia, Egypt
∗eldesokyi@yahoo.com
†A mousa15@yahoo.com
Received 11 April 2009

Revised 14 September 2009
This work is concerned with the peristaltic transport of a Newtonian and non-Newtonian
Maxwellian fluid in an axisymetric cylindrical tube filled with a homogenous porous
medium, in which the flow is induced by traveling transversal waves on the tube wall.
Like in peristaltic pumping, the traveling transversal waves induce a net flow of the
liquid inside the tube. The viscosity as well as the compressibility of the liquid is taken
into account. Modified Darcy’s law has been used to model the governing equation. The
present theoretical model may be considered as mathematical representation to the case
of gall bladder and bile duct with stones and dynamics of blood flow in living creatures.
The Navier–Stokes equations for an axisymmetric cylindrical tube are solved by means
of a perturbation analysis, in which the ratio of the wave amplitude to the radius of
the tube is small parameter. In the second order approximation, a net flow induced by
the traveling wave is calculated for various values of the compressibility of the liquid,
relaxation time and the permeability parameter of porous medium. The calculations
disclose that the compressibility of the liquid, the permeability parameter of porous
medium and non-Newtonian effects in presence of peristaltic transport have a strong
influence of the net flow rate. Finally, the graphical results are reported and discussed
for various values of the physical parameters of interest.
Keywords: Peristaltic transport; Maxwell fluid; compressible liquid; traveling waves;

porous medium; modified Darcy’s law; non-Newtonian effects.
1. Introduction
Peristaltic transport is a form of fluid transport that occurs when a progressive
wave of area contraction or expansion propagates along the length of an extensible
tube containing a liquid. It appears to be major mechanism for urine transport in
ureter, food mixing and chyme movement in intestinal, transport of spermatozoa in
cervical canal, transport of bile in bile ducts and so on. Technical roller and finger
pumps also operate according to this rule.
255
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
256 I. M. Eldesoky & A. A. Mousa
To understand peristaltic action in different situations, several theoretical and

experimental attempts have been made since the first investigation of Latham
[1]. The literature on peristalsis is by now quite extensive. Important recent con-
tributions to the topic include the works of Hayat et al. [2–6], Elshehawey and
El-Sebaei [7], Mekheimer [8–12].
It is well known that flow through a porous medium has practical applications
especially in geophysical fluid dynamics. Examples of natural porous media are
beach sand, sandstone, limestone, rye bread, wood, the human lung, bile duct, gall
bladder with stones and in small blood vessels. In the arterial systems of humans
or animals, it is quite common to find localized narrowings, commonly caused by
intravascular plaques. These stenoses disturb the normal pattern of blood flow
through the artery. Acknowledge of flow characteristics in the vicinity of stenosis

may help to further the understanding of some major complications which can
arise such as, an ingrowth of tissue in the artery, the development of a coronary
thrombosis, the weakening and bulging of the artery downstream from stenosis, etc.
The investigations of blood flow through arteries are of considerable importance
in many cardiovascular diseases, particularly atherosclerosis. In some pathological
situations, the distribution of fatty cholesterol and artery clogging blood clots in
the lumen of coronary artery can be considered as equivalent to a porous medium.
Recently, Kothandapani and Srinivas [13] studied the effect of elasticity of
the flexible walls on the MHD peristaltic flow of a Newtonian fluid in a two-
dimensional porous channel with heat transfer under the assumptions of long wave-
length and low-Reynolds number. The analytical solution has been obtained for
the stream function, temperature and heat transfer coefficient. Also, Kothandapani
and Srinivas [14] studied the peristaltic transport of an incompressible viscous fluid
in an inclined asymmetric channel through a porous medium under long wave-
length and low-Reynolds number assumptions. The flow is examined in a wave
frame of reference moving with the velocity of the wave. Vajravelu et al. [15] stud-
ied the interaction of peristalsis with heat transfer for the flow of a viscous fluid
in a vertical porous annular region between two concentric tubes. Long wavelength
approximation (that is, the wavelength of the peristaltic wave is large in comparison
with the radius of the tube) is used to linearize the governing equations. Using the
perturbation method, the solutions are obtained for the velocity and the tempera-
ture fields. Manoranjan and Ramachandra [16] studied the peristaltic transport in
a two-dimensional channel, filled with a porous medium in the peripheral region
and a Newtonian fluid in the core region. The fluid flow is investigated in the
wave frame of reference moving with the velocity of the peristaltic wave. Brinkman
extended Darcy equation is utilized to model the flow in the porous layer. The
interface is determined as a part of the solution using the conservation of mass
in both the porous and fluid regions independently. Elshehawey et al. [17] studied
peristaltic transport in an asymmetric channel through a porous medium. The flow
is investigated in a wave frame of reference moving with velocity of the wave under
the assumptions of long-wavelength and low-Reynolds number. Hayat et al. [18]
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
Peristaltic Flow of a Compressible Non-Newtonian Maxwellian Fluid 257
studied a mathematical model of peristalsis in tubes through a porous medium.

Hayat et al. [19] studied Hall effects on the peristaltic flow of a Maxwell fluid in a
porous medium. This work has been concerned with the peristaltic transport of an
incompressible, electrically conducting Maxwell fluid in a planar channel. The flow
in the porous space is due to a sinusoidal wave traveling on the channel walls. The
Hall Effect is taken into account and permeability of porous medium is considered
uniform.
Recently, the motion of non-Newtonian fluids has been an important subject
in the field of chemical, biomedical and environmental engineering and science.
Undoubtedly the mechanics of non-Newtonian fluids presents special challenges to
engineers, physicists, modelers, numerical simulates and mathematicians. This is
due to the fact that nonlinearity manifests itself in a variety of ways. The flows of
non-Newtonian fluids are not only important because of their technological signifi-
cance but also in the interesting mathematical features presented by the equations
governing the flow. Moreover, the elastic properties of real fluids can be deter-
mined and measured. There are few studies available about peristaltic flow of non-
Newtonian fluids. Recently, Abd Elnaby and Haroun [20] discussed the peristaltic
flow of a viscous fluid in a channel having compliant walls.
In all the above mentioned studies, no compressibility of the liquid has been
taken into account. But it is well known that the compressibility of the liquid has
a strong influence on the net flow induced by the peristaltic motion. A recent few
papers have been made in this filed. Aarts and Ooms [21] studied the net flow of
compressible liquids induced by traveling waves in porous media. They found that,
the compressibility of the liquid has a strong influence on the net flow rate, and the
Reynolds number plays a more significant role in the net flow of a compressible liquid
than of an incompressible one. Also, the net flow rate is decreasing with increasing
the wave number in the case of incompressible liquid but when the compressibility
of the liquid has been taken into account the net flow rate increases with increas-
ing the wave number. Tsiklauri and Bresenev [22] studied non-Newtonian effects
in the peristaltic flow of a Maxwell fluid. They analyzed the effect of viscoelastic-
ity on the dynamics of fluids by studying the flow of Maxwell fluid in a circular
tube taking the compressibility of the liquid into account. They have found that
in the extreme non-Newtonian regime, there is a possibility of a fluid flow in the
direction opposite to the propagation of the wave traveling on the tube wall. Elshe-
hawey et al. [23] studied the peristaltic transport of a compressible viscous liquid
through a tapered pore. Later, Elshehawey et al. [24] studied slip effects on the
peristaltic flow of a non-Newtonian Maxwellian fluid. They analyzed the effect of
slip boundary conditions on the dynamics of fluids by studying the flow of a New-
tonian and non-Newtonian Maxwellian fluid in an axisymmetric cylindrical tube,
taken the compressibility of the liquid into account. The calculations disclose that
the compressibility of the liquid, Knudsen number of slip flow and non-Newtonian
effects in presence of peristaltic transport have a strong influence of the net
flow rate.
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
To the best of the author’s knowledge, no attempt has been made yet to discuss
the peristalsis of non-Newtonian Maxwellian fluid through a porous medium taking
the compressibility of the liquid into account. Therefore the main purpose of the
present paper is to present a theoretical analysis of the peristaltic flow of a non-
Newtonian Maxwellian fluid through porous medium with constant permeability in
a cylindrical tube. Modified Darcy’s law for a Maxwell fluid has been used for the
modeling. The analysis has been carried out using a perturbation method in which
the ratio of the wave amplitude to the radius of the tube is the small parameter. The
net flow was obtained in explicit form. Moreover, the net flow induced by the
traveling waves in presence of porous medium is calculated for various values of
the compressibility number, relaxation time and the permeability parameter of
porous medium. Actually, we extend the analysis of Yin and Fung [25] by taking
the permeability of porous medium into account. It is also an extension of the
later work, Aarts and Ooms [21], in which the compressibility has been taken into
account and of Tsiklauri and Bresenev [22], where the non-Newtonian effects have
been incorporated.
2. Basic Equations and Position of the Problem

We consider an axisymmetric cylindrical tube of radius R and length L. We assume
that an elastic wave induces a traveling wave on the wall (boundary) of the tube
with the displacement of the following form:

2π
W (z, t) = R + a cos (z − ct) , (2.1)
λ
where a is the amplitude of the traveling wave, while λ and c are its wave length and
velocity, respectively. We note that the z-axis of the (r, φ, z) cylindrical coordinate
system is directed along the axis of the tube.
It is well known that in an unbounded porous medium, Darcy’s law holds for a
Newtonian fluid at low speed. This law provides a relation between pressure drop
and velocity. According to this law the pressure drop induced by the frictional drag
is directly proportional to the velocity. There are some studies [26, 27] on Stock’s
problem involving viscoelastic fluid through a porous medium. There is no study on
peristalsis dealing viscoelastic fluid in a porous medium. On the basis of Oldroyd’s
model [28–32] the following law has been suggested

∂ µϕ
1 + tm ∇p = − (1 + λν )ν , (2.2)
∂t k
where k is the permeability, λν is the retardation time, tm is the relaxation time, µ
is the dynamic viscosity, p is the hydrostatic pressure, v is the velocity vector and
ϕ is the porosity of the porous medium.
It is known that constitutive equation for Maxwell fluid can be obtained from
the constitutive equation of an Oldroyd-B fluid by letting λν = 0. Since we have
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
an interest in Maxwell fluid in this paper, the filtration law for Maxwell fluid can
be inferred from Eq. (2.2) as follows:

∂ µϕ
1 + tm ∇p = − ν . (2.3)
∂t k
Note that for tm = 0, the above equation yields Darcy’s law. Since the pressure
gradient in Eq. (2.3) can also be interpreted as a measure of the resistance to flow in
the bulk of the porous medium and f is a measure of the flow resistance offered by
the solid matrix. Therefore, f can be inferred from Eq. (2.3) to satisfy the following
equation [19]

∂ µϕ
f = − ν .
1 + tm (2.4)
∂t k
Under the assumptions mentioned above, the continuity and momentum equa-
tions governing the flow of a non-Newtonian Maxwellian fluid (taking its compress-
ibility into account) through a porous medium in a cylindrical tube can be written
as:
The equations that govern the flow are the balance of mass
∂ρ
+ ∇ · (ρ ν ) = 0, (2.5)
∂t
and the momentum equation
∂ν + f,
ρ + ρ(ν · ∇)
ν = −∇p
− ∇τ̃ (2.6)
∂t
where ρ is the liquid density, p the pressure, v the velocity vector and τ̃ represents
the viscous stress tensor. We describe the viscoelastic properties of the fluid using
Maxwell’s model [33], which assumes that
∂ τ̃ ν − µ∇
tm = −µ∇ · ν − τ̃ . (2.7)
∂t 3
We further, assume that the following equation of state holds [34]:
1 ∂ρ
= k∗ , (2.8)
ρ ∂p
where k ∗ is the compressibility of the liquid. We also assume that the fluid’s velocity
has only r and z components. The solution of this equation for the density as a
function in the pressure is given by
∗
ρ = ρo e[k (p−po )]
, (2.9)
where ρo is the constant density at the reference pressure po .

We make use of “no-slip” boundary condition at the boundary of the tube, i.e.
∂W
νr (W, z, t) = , νz (W, z, t) = 0. (2.10)
∂t
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
Equation (2.7) can be rewritten in the following form:

∂ v − µ∇
1 + tm τ̃ = −µ∇ · ν . (2.11)
∂t 3
∂
Further, we apply the operator 1 + tm ∂t ) to the momentum equation (2.6) and
eliminate τ̃ in it using Eq. (2.11).

∂ ∂ν ∂ 2 ν
1 + tm ρ + ρ(ν · ∇)ν = − 1 + tm
∇p + µ∇
∂t ∂t ∂t

µ ∂
+ ∇( ∇ · ν ) + 1 + tm f . (2.12)
3 ∂t
In cylindrical coordinates, the balance of mass equation (2.5) reads

∂ρ ∂ρ ∂ρ ∂vr vr ∂vz
+ vr + vz +ρ + + = 0, (2.13)
∂t ∂r ∂z ∂r r ∂z
while the Navier–Stokes Eq. (2.12) becomes

∂ ∂νr ∂νr ∂νr
1 + tm ρ + ρ νr + νz
∂t ∂t ∂r ∂z
2
∂ ∂p ∂ νr 1 ∂νr νr ∂ 2 νr
= − 1 + tm +µ + − +
∂t ∂r ∂r2 r ∂r r2 ∂z 2

µ ∂ ∂νr νr ∂νz µϕ
+ + + − νr , (2.14)
3 ∂r ∂r r ∂z k

∂ ∂νz ∂νz ∂νz
∂t ∂t ∂r ∂z
2
∂ ∂p ∂ νz 1 ∂νz ∂ 2 νz
= − 1 + tm +µ + +
∂t ∂z ∂r2 r ∂z ∂z 2

µ ∂ ∂νr νr ∂νz µϕ
+ + + − vz . (2.15)
3 ∂z ∂r r ∂z k
It would be expedient to simplify these equations by introducing non-
dimensional variables. We have a characteristic velocity c and have characteris-
tic lengths a, λ, and R. The following variables based on c and R could this be
introduced.
W νr νz ρo p
W = , νr = , νz = , ρ= , p= ,
R c c ρ ρ o c2
po ct Q k
po = , t= , Q= , k= .
ρ o c2 R cR2 ϕR2
Here the over par denotes the dimensionless quantity.
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
The amplitude ratio ε, the wave number α, the Reynolds number Re, and the
compressibility number χ are defined by:
a 2πR ρo cR
ε= , α= , Re = and χ = k ∗ ρo c2 .
R λ µ
Under the above assumptions Eqs. (2.9) and (2.13)–(2.15) can be rewritten in the
non-dimensional form after dropping the bars, as:

∂ρ ∂ρ ∂ρ ∂νr νr ∂νz
+ νr + νz +ρ + + = 0, (2.16)
∂t ∂r ∂z ∂r r ∂z

∂ ∂νr ∂νr ∂νr
∂t ∂t ∂r ∂z

∂ ∂p 1 ∂ 2 νr 1 ∂νr νr ∂ 2 νr
= − 1 + tm + + − +
∂t ∂r Re ∂r2 r ∂r r2 ∂z 2

1 ∂ ∂νr νr ∂νz 1
+ + + − νr , (2.17)
3Re ∂r ∂r r ∂z Re k

∂ ∂νz ∂νz ∂νz
∂t ∂t ∂r ∂z

∂ ∂p 1 ∂ 2 νz 1 ∂νz ∂ 2 νz
= − 1 + tm + + +
∂t ∂z Re ∂r2 r ∂z ∂z 2

1 ∂ ∂νr νr ∂νz 1
+ + + − νz , (2.18)
3Re ∂z ∂r r ∂z Re k
ρ = eχ(p−po ) . (2.19)
Also the boundary conditions Eq. (2.10) becomes
∂η(z, t)
νr ((1 + η), z, t) = , νz ((1 + η), z, t) = 0, (2.20)
∂t
where
η(z, t) = ε cos α(z − t). (2.21)
3. Perturbation Solution
To illustrate the nature of the solution we shall consider the important case
of no flow in absence of the peristaltic wave. It is clear that we have to
choose very small amplitude ratio ε 1 (this occurs only when the wave amplitude
is small compared with the radius of the tube) because we used the perturbation
method with the amplitude ratio ε as a parameter [35]. Following [21], we seek the
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
solution of the governing equations in a form
p = p0 + εp1 (r, z, t) + ε2 p2 (r, z, t) + · · · ,

vr = εu1 (r, z, t) + ε2 u2 (r, z, t) + · · · ,
(3.1)
vz = εv1 (r, z, t) + ε2 v2 (r, z, t) + · · · ,
ρ = 1 + ερ1 (r, z, t) + ε2 ρ2 (r, z, t) + · · ·
To illustrate the nature of the solution, we shall consider the important

case of no flow in absence of the peristaltic wave. Then, doing a usual
perturbative analysis using the latter expansions, we can obtain a closed set of
governing equations for the first (ε) and second (ε2 ) order as the following:

∂ ∂u1 ∂ ∂p1 1 ∂ 2 u1 1 ∂u1 u1 ∂ 2 u1
1 + tm = − 1 + tm + + − 2 +
∂t ∂t ∂t ∂r Re ∂r2 r ∂r r ∂z 2

1 ∂ ∂u1 u1 ∂ν1 u1
+ + + − , (3.2)

∂ ∂ν1 ∂ ∂p1 1 ∂ 2 ν1 1 ∂ν1 ∂ 2 ν1
1 + tm = − 1 + tm + + +
∂t ∂ ∂t ∂z Re ∂r2 r ∂r ∂z 2

1 ∂ ∂u1 u1 ∂ν1 ν1
+ + + − , (3.3)
∂ρ1 ∂u1 u1 ∂ν1

+ + + = 0, (3.4)
∂t ∂r r ∂z
ρ1 = χp1 . (3.5)

∂ ∂u2 ∂u1 ∂u1 ∂u1
1 + tm + ρ1 + u1 + ν1
∂t ∂t ∂t ∂r ∂z

∂ ∂p2 1 ∂ 2 u2 1 ∂u2 u2 ∂ 2 u2
= − 1 + tm + + − 2 +
∂t ∂r Re ∂r2 r ∂r r ∂z 2

1 ∂ ∂u2 u2 ∂ν2 u2
+ + + − , (3.6)

∂ ∂ν2 ∂ν1 ∂ν1 ∂ν1
1 + tm + ρ1 + u1 + ν1
∂t ∂t ∂t ∂r ∂z

∂ ∂p2 1 ∂ 2 ν2 1 ∂ν2 ∂ 2 ν2
= − 1 + tm + + +
∂t ∂z Re ∂r2 r ∂r ∂z 2

1 ∂ ∂u2 u2 ∂ν2 ν2
+ + + − , (3.7)
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970

∂ρ2 ∂ρ1 ∂ρ1 u2 ∂u2 ∂ν2 ∂u1 u1 ∂ν1
+ u1 + ν1 + + + + ρ1 + + = 0, (3.8)
∂t ∂r ∂z r ∂r ∂z ∂r r ∂z
1
ρ2 = χp2 + χ2 p21 . (3.9)
2
Expanding Eq. (2.20) by Taylor expansion around r = 1 and substituting from
Eq. (3.1), we get the following boundary conditions:
iα iα(z−t)
u1 (1, z, t) = − (e − e−iα(z−t) ), (3.10)
2
1 ∂u1
u2 (1, z, t) + (eiα(z−t) + e−iα(z−t) ) (1, z, t) = 0, (3.11)
2 ∂r
ν1 (1, z, t) = 0, (3.12)
1 ∂ν1
ν2 (1, z, t) + (eiα(z−t) + e−iα(z−t) ) (1, z, t) = 0. (3.13)
2 ∂r
Further, following [21, 22], we seek the solution of the liner problem in the form
u1 (r, z, t) = U1 (r)eiα(z−t) + U1 (r)e−iα(z−t) ,

ν1 (r, z, t) = ν1 (r)eiα(z−t) + V1 (r)e−iα(z−t) ,
(3.14)
p1 (r, z, t) = P1 (r)eiα(z−t) + P1 (r)e−iα(z−t) ,
ρ1 (r, z, t) = χP1 (r)eiα(z−t) + χP1 (r)e−iα(z−t) .
Here and in the following equations, the bar denotes a complex conjugate.
On the other hand, we seek the second (ε2 ) order solution in the form
u2 (r, z, t) = U20 (r) + U2 (r)e2iα(z−t) + U2 (r)e−2iα(z−t) ,

ν2 (r, z, t) = V20 (r) + V2 (r)e2iα(z−t) + V1 (r)e−2iα(z−t) ,
(3.15)
p2 (r, z, t) = P20 (r) + P2 (r)e2iα(z−t) + P2 (r)e−2iα(z−t) ,
ρ2 (r, z, t) = D20 (r) + D2 (r)e2iα(z−t) + D2 (r)e−2iα(z−t) .
The latter choice of solution is motivated by the fact that the peristaltic flow is
essentially a nonlinear (second-order) effect [21], and adding a nonoscillatory term
in the first order gives only a trivial solution. Thus, we can add nonoscillatory terms,
such as U20 (r), V20 (r), P20 (r), and D20 (r) which do not cancel out in the solution
after the time averaging over the period, only in the second and higher orders.
Substituting from Eq. (3.14) in Eqs. (3.2)–(3.5) and Eqs. (3.10)–(3.13), we
obtain the following system of equations:

1 U1 U1 2 1
− iα(1 − iαtm )U1 = −(1 − iαtm )P1 + U1 + − 2 − α + U1
Re r r k

1 d U1
+ U1 + + iαV1 , (3.16)
3Re dr r
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970

1 V 1
−iα(1 − iαtm )V1 = −iα(1 − iαtm )P1 + V1 + 1 − α2 + V1
Re r k

iα U1
+ U1 + + iαV1 , (3.17)
3Re r
U1
U1 + + iαV1 = iαχP1 , (3.18)
r
iα
U1 (1) = − , (3.19)
2
V1 (1) = 0. (3.20)
Here, the prime denotes a derivative with respect to r.

Further, we rewrite the system of Eqs. (3.16)–(3.18) in the following form

U U1
− γP1 + U1 + 1 − 2 − β 2 U1 = 0, (3.21)
r r

i V
−γP1 − V1 + 1 − β 2 V1 = 0, (3.22)
α r
where the complex parameters γ and β are given by
iαχ 1
γ = (1 − iαtm )Re − , β 2 = α2 + − iα(1 − iαtm )Re. (3.23)
3 k
Note that, Eq. (3.23) is similar to [21, Eq. (3.12)] except that γ and β are
modified by substitution Re → (1 − iαtm )Re and the term k1 in β.
Eliminating V1 (r) by using Eq. (3.18), we rewrite Eq. (3.22) as

iχ P 1 iαγ
− P1 + − β2 + P1
α r χ

1 d 1 U U1
+ 2 + U1 + 1 − 2 − β 2 U1 = 0. (3.24)
α dr r r r
Differentiating Eq. (3.24) with respect to r and using Eq. (3.21), we are led to
the following equation
2
iχ d 1 d 1 iαγ U U1
− 2
+ − − β 2
+ U1 + 1 − 2 − β 2 U1
αγ dr r dr r2 χ r r
2
1 d d 1 U1 U1
+ 2 + − U 1 + − 2 − β 2 U1 = 0. (3.25)
α dr2 dr r2 r r
This equation is rewritten, after multiplication by α2 , as
2 2
iχα d 1 d 1 2 d d 1 2
1− + − − υ + − − β U1 = 0. (3.26)
γ dr2 r dr r2 dr2 dr r2
From Eq. (3.26), we obtain the master equation for U1 (r) and find its general
solution as
U1 (r) = C1 I1 (υr) + C2 I1 (βr), (3.27)
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
where I1 is the modified Bessel function of the first kind of order one. Note that,
Eq. (3.27) is similar to [21, Eq. (3.18)] except that C1 and C2 are complex constants
(calculated using Eqs (3.19) and (3.20)) and defined by
iαβυI0 (β)
C1 = ,
2[α2 I1 (β)I0 (υ) − βυI1 (υ)I0 (β)]
(3.28)
−iα3 I0 (υ)
C2 = .
2[α2 I1 (β)I0 (υ) − βυI1 (υ)I0 (β)]
where
iχ
(1 − χ)(1 − iαtm )Re − (4/3)iαχ −
υ 2 = α2 αk
. (3.29)
(1 − iαtm )Re − (4/3)iαχ
Note that, Eq. (3.29) is similar to [21, Eq. (3.17)], except that υ is modified by
iχ
substitution Re → (1 − iαtm )Re and the term − αk .
Here, I0 is the modified Bessel function of the first kind of order zero.
We also obtain the general solution for V1 (r):
iαC1 iβC2
V1 (r) = I0 (υr) + I0 (βr), (3.30)
υ α
and the general solution for P1 (r):
C1 (υ 2 − β 2 )
P1 (r) = I0 (υr). (3.31)
γυ
To solve the system of second order of ε, inserting Eq. (3.15) into Eqs. (3.6)–(3.9)
and Eqs. (3.10)–(3.13), we obtain the following set of equations:
D20 = χP20 + χ2 P1 P1 , (3.32)

U20 P1 U1 P1 U1
U20 + = −χ P1 U1 + P1 U1 + + + U1 P1 + U1 P1 , (3.33)
r r r
iαχP1 U1 − iαχP1 U1 + U1 U1 + U1 U1 + iαU1 V1 − iαU1 V1

1 U20 U20
= −P20 + U20 + − 2
(1 − iαtm )Re r r

1 d U20 U20
+ U20 + − , (3.34)
3(1 − iαtm )Re dr r k Re

1 V 20 V20
iαχP1 V1 − iαχP1 V1 + U1 V1 + U1 V =
V20 + − ,
(1 − iαtm )Re r k Re
(3.35)
1
U20 (1) + U1 (1) + U1 (1) = 0, (3.36)
2
1
V20 (1) + V (1) + V1 (1) = 0. (3.37)
2 1
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
It will be seen that, as far as the net flow is considered only the functions
U20 , V20 , P20 and D20 contribute to the net flow as long as terms up to O (ε2 ) are
retained. Thus, the functions U2 , V2 , P2 and D2 do not contribute to the net flow,
and therefore, we shall not write down the equations that these functions satisfy or
solve them. We continue with the solutions for U20 , V20 , P20 and D20 [21]. To that
end, the second-order solution U20 (r) can also be found in a way similar to the one
used in the first order as follows:
D1
U20 (r) = − χ[P1 (r)U1 (r) + P1 (r)U1 (r)], (3.38)
r
where D1 is a complex constant (follows from the boundary condition Eqs. (3.35)
and (3.36)) defined by D1 = 0 and U20 (r) have the final form as
U20 (r) = −χ[P1 (r)U1 (r) + P1 (r)U1 (r)]. (3.39)
We also obtain the general solution for V20 (r)as follows

1
V20 (r) = D2 − (1 − iαtm )Re [V1 (ζ)U1 (ζ) + V1 (ζ)U1 (ζ)]dζ, (3.40)
r
where D2 is a complex constant defined by
−1
D2 = (V (1) + V1 (1)), (3.41)
2 1

where the values of V20 (1) and V1 (1) are defined by:

iαC1 C1
V20 (1) = (1 − iαtm )Re I0 (υ)I1 (υ)
υ
iαC1 C2 iβC2 C1 iβC2 C2
+ I0 (υ)I1 (β) + I0 (β)I1 (υ) + I0 (β)I1 (β)
υ α α
iαC1 C1 iαC2 C1
− I0 (υ)I1 (υ) − I0 (υ)I1 (β)
υ υ
iβC1 C2 iβC2 C2
− I0 (β)I1 (υ) − I0 (β)I1 (β) , (3.42)
α α
and
iβ 2 C2
V1 (1) = iαC1 [υI0 (υ) − I1 (υ)] + [βI0 (β) − I1 (β)]. (3.43)
α
The dimensionless fluid flow rate Q can be calculated as [21]
1 1
Q(z, t) = 2π ε v1 (r, z, t)r dr + ε2 v2 (r, z, t)r dr + O(ε3 ) .
0 0
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
Next, the net flow is considered over one period of time. The average of a variable
G over one period T of time t as
T
1
G = G(r, z, t)dt, (3.44)
T 0
2π
at T = α . Consequently, the mean net axial velocity Vz reads
Vz = ε2 V20 (r). (3.45)
3
Under neglect of O (ε )-terms, while the net flow rate Q is given by
1
Q = 2πε2 V20 (r)r dr. (3.46)
0
Under neglect of O (ε3 )-terms. Thus, the traveling wave induces a net flow of the
liquid, of which the (dimensionless) rate is expressed by Eq. (3.46). Hence, the net
flow and the mean net axial velocity are an effect of order ε2 .
4. Numerical Results and Discussion

To study the behavior of the net flow rate, numerical calculations for several values
of α, k, tm and χ are carried out. We concentrate on the solution of the dimensionless
problem as described in the previous section. It is clear that we have to choose
ε 1 because we used the perturbation method with the amplitude ratio ε as a
parameter [35]. Also, for to be perturbation method valid and accurate we must
have (εα2 Re 1) accordingly to [36].
We consider the net flow rate Q given by Eq. (3.46). After one integration by
parts Q can be expressed as
1
Q = πε2 D2 − (1 − iαtm )Re r2 [V1 (r)U1 (r) + V1 (r)U1 (r)]dr , (4.1)
0
where the solution of Eq. (3.40) for V20 (r) is used.

A numerical code has been written to calculate Q according to Eq. (49).
In order to check the validity of our code, we run it for the parameters similar
to the ones used by other authors. For instance, for ε = 0.15, Re = 10000.0,
α = 0.20, χ = 0.0, tm = 0.0 and k → ∞, we obtain Q = 0.2709, which equal
(if we keep four digits after the decimal point) to the result of the authors of [21]
who actually obtained Q = 0.2709. Note that, while much of the hemodynam-
ics in a healthy human body has low Reynolds number, resulting in laminar flow,
relatively high Reynolds number flow is observed at some specific locations, which
can cause transition to turbulence, for example, the Reynolds number in the human
aorta [37, 38]. Further, we have made several runs of our code for different val-
ues of the parameter k at tm = 0.0 (Newtonian regime). We note again that k
and tm enter the equations because we have included the flow through porous
medium into non-Newtonian Maxwell’s model. Equation (4.1) will be identical to
the similar [21, Eq. (4.1)], if we set tm = 0.0 and k → ∞ in all our equations.
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970

Fig. 1. The dimension less flow rate Q versus χ at ε = 0.001, Re = 10000.0, tm = 0.0 and
α = 0.001.
Further, Eq. (4.1) will be identical to the similar [22, Eq. (16)], if we set k → ∞ in
all our equations.
First, we investigate the effect of the permeability of porous medium k in the
case of a Newtonian (tm = 0.0) Maxwellian fluid. The results of our calculations
are presented in Fig. 1, where we investigate the dependence of Q on the com-
pressibility parameter χ for various values of k. In order to compare our results
with the ones from [21], we have plotted Q for the following set of parameters:
ε = 0.001, Re = 10000.0, α = 0.001, tm = 0.0 and k → ∞. We note that the curve
is identical to the corresponding curve (solid curve) in [21, Fig. 2]. This obviously
corroborates the validity of our numerical code. Further, to investigate the depen-
dence of the flow rate Q on k. We perform the calculation for a few values of k.
When k = 0.05, we observe that the net flow rate Q reaches to a maximum value
Q = 2.3361 × 10−5 and occurs at χ = 0.1. Further, when k = 0.1, we notice that
the maximum value of Q = 2.2099 × 10−5 and occurs at χ = 0.2. Moreover, the
maximum value of Q = 2.4561 × 10−5 occurs at χ = 0.4 when k = 0.05. From the
above discussion, we notice that Q attains a maximum for a certain value of χ
and there is shifting the maximum value of Q towards higher values of χ’s with
increasing k. Furthermore, we notice that at any value of χ within the range of
0.0 < χ < 0.2 (approximately) the net flow rate decreases with increasing k while
at any value of χ within the range of 0.2 < χ < 1.0 (approximately) the net flow
rate Q increases with increasing k.
In Fig. 2, we investigate the behavior of the net flow rate Q on the param-
eter α, which is the tube radius measured in wavelengths. Again, to check for
the consistency of our numerical results with the ones of [21], and also investigate
phenomena brought about by the introduction of permeability of porous medium
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970

Fig. 2. The dimensionless net flow rate Q versus α at ε = 0.001, Re = 10000.0, tm = 1000 and
χ = 0.6.
into the model (non-Newtonian Maxwell fluid), we plot the net flow rate Q versus
α for the following set of parameters: ε = 0.001, Re = 10000.0, χ = 0.6, tm = 0.0 and
k → ∞. If we compare this curve in our Fig. 2 with the dashed curve in [21, Fig. 3],
we will note no difference, which again corroborates the validity of our numerical
code. We then set k to various values and investigate the changes induced by porous
Fig. 3. The dimensionless net flow rate Q versus χ at ε = 0.001, Re = 10000, tm = 1000 and
α = 0.001.
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
medium effects. We note that the net flow rate Q increases with increasing the
wave number α. Moreover, we notice that at any value of α within the range of
0.0 < α < 0.004 (approximately) the net flow rate increases with increasing k while
at any value of α within the range of 0.004 < α < 0.01 (approximately) the net
flow rate Q decreases with increasing k.
Second, we investigate the effect of the permeability of porous medium in the
case of non-Newtonian Maxwellian fluid. It is known that the viscoelastic fluids,
described by the Maxwellian fluid, have different flow regimes depending on the
value of the parameter De = tν /tm , which is called the Deborah number. In effect,
the Deborah number is a ratio of the characteristic time of viscous effects tν =
ρR2 /µ to the relaxation time tm . As noted in [33], the value of the parameter De
(which the authors of [33] actually call α) determines in which regime the system
resides. Beyond a certain critical value (De = 11.64), the system is dissipative, and
conventional viscous effects dominate. On the other hand, for small De (De < Dec)
the system exhibits viscoelastic behavior.
The results of our calculations are presented in Fig. 3, where we investigate the
dependence of Q on the compressibility parameter χ for various values of k. In
order to compare our results with the ones from [22], we have plotted Q for the
following set of parameters: ε = 0.001, Re = 10000.0, α = 0.001, tm = 1000 and
k → ∞. We note that the curve is identical to the corresponding curve in Fig. 1
(dashed-dotted curve with asterisks) from [22]. This obviously corroborates the
validity of our numerical code. Further, to investigate the dependence of the flow
rate Q on k. We perform the calculation for a few values of k. When k = 0.01
we can observe that the flow rate decreases with increasing the compressibility
parameter χ. Furthermore, when k = 0.05, we observe that the net flow rate reaches
to a maximum value Q = 1.7052 × 10−5 at χ = 0.1. Further, when k = 0.1,
we notice that the maximum value of Q = 1.5413 × 10−5 occurs at χ = 0.4.
Moreover, the maximum value of Q = 2.4918 × 10−5 occurs at χ = 0.8 when
k → ∞. From the above discussion, we notice that Q decreases with increasing
the compressibility parameter χ at k = 0.01 and with increasing k, Q attains
a maximum for a certain value of χ and there is shifting the maximum value of
Q towards higher values of χ’s with increasing k. Furthermore, we notice that
at any value of χ, within the range of χ between 0.0 and 0.4 (approximately), the
net flow rate decreases with increasing k while at any value of χ within the range
of χ between 0.4 (approximately) and 1.0 the net flow rate Q increases with
increasing k.
In Fig. 4, the dimensionless net flow rate Q is plotted versus α at the following
set of parameters: ε = 0.001, Re = 10000.0, χ = 0.6 and tm = 1000. We investigate
the behavior of the net flow rate Q on the parameter α at various values of k
within the range of 0.0 ≤ α ≤ 0.01. We note that at k → ∞ the curve in our Fig. 4
is the same curve (dashed-dotted curve with asterisks) in [22, Fig. 2]. The net flow
rate Q is weakly affected by k when α < 0.0015. At any value of α within the
range 0.0015 < α < 0.01, the net flow rate Q decreases with increasing k.
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
×10−5
2.2 k=0.1
2.0 k=0.5
1.8 k=1.0
1.6 k →∞
1.4
<Q>
1.2
1.0
0.8
0.6
0.4
0.000 0.002 0.004 0.006 0.008 0.010

α
Fig. 4. The dimensionless net flow rate Q versus wave number α at ε = 0.001, Re = 10000.0,
tm = 1000 and χ = 0.6.
Fig. 5. The dimensionless net flow rate Q versus χ at ε = 0.001, Re = 10000, tm = 10000.0
and α = 0.001.
In Fig. 5, we investigate the behavior of the flow rate Q on the compressibility
parameter χ at tm = 10000 (deeply non-Newtonian regime). Again, to check for the
consistency of our numerical results with the ones of [22], we first plot Q versus χ
for the following set of parameters: ε = 0.001, Re = 10000.0, α = 0.001 and k → ∞.
If we compare the curve in our Fig. 5 with the (dashed curve with empty squares)
in [22, Fig. 1], we will note no difference, which again corroborates the validity of
our numerical code. We then set k to various values and investigate the effect of k
at deeply non-Newtonian regime. In this deeply non-Newtonian regime, we notice
that, the net flow rate Q increases with increasing χ. Also, at any value of χ, the
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970

Fig. 6. The dimensionless net flow rate Q versus α at ε = 0.001, Re = 10000.0, tm = 10000.0
and χ = 0.6.
net flow rate Q decreases with increasing k. Furthermore, at high compressibility
(χ = 1.0) the net flow rate is weakly affected by k since the values of Q lie within
the range of (7.2595 × 10−5 –7.9463 × 10−5 ).
In Fig. 6, the net flow rate Q is plotted versus α, for the following set of
parameters: ε = 0.001, Re = 10000.0, χ = 0.6 and tm = 10000 (deeply non-
Newtonian regime) and various values of Kn within the range of 0.0 ≤ α ≤ 0.01.
We note that at k → ∞ the curve in our Fig. 6 is the same curve in [22, Fig. 2]. We
note from this figure that in this deeply non-Newtonian regime, Q becomes high
oscillatory, but what is unusual again is that we observe the negative flow rates
for certain values of α. Oscillatory behavior (appearance of numerous of maxima
in the behavior of a physical value) in the deeply non-Newtonian regime is not
new [22, 24]. Furthermore, we can observe that there is no effect (approximately)
of k on the net flow rate Q in the deeply non-Newtonian regime.
5. Conclusions
No attempt has been made yet to discuss the peristalsis of non-Newtonian
Maxwellian fluid through a porous medium taking the compressibility of the liq-
uid into account. Therefore, the main purpose of the present paper is to present
a theoretical analysis of the peristaltic flow of a non-Newtonian Maxwellian fluid
through porous medium with constant permeability in a cylindrical tube. Modi-
fied Darcy’s law for a Maxwell fluid has been used for the modeling. The present
theoretical model may be considered as mathematical representation to the case of
gall bladder and bile duct with stones and dynamics of blood flow in living crea-
tures. The analysis has been carried out using a perturbation method in which the
ratio of the wave amplitude to the radius of the tube is the small parameter. We
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
have found that the compressibility number χ has a significant influence on the net
flow rate as shown in [21, 22, 24] and the permeability of porous medium k plays
a more significant role in the net flow of a compressible liquid than of an incom-
pressible one. The main conclusions of the present paper may be summarized as
follows:
• For Newtonian fluid (relaxation time tm = 0.0), the net flow rate Q attains
a maximum for a certain value of the compressibility parameter χ and there is
shifting the maximum value of Q towards higher values of χ’s with increasing
the permeability parameter k. The net flow rate Q decreases with increasing
the permeability parameter k within 0.0 < χ < 0.2 while the net flow rate Q
increases with increasing the permeability parameter k within 0.2 < χ < 1.0.
• For Newtonian fluid (relaxation time tm = 0.0), the net flow rate Q increases
with increasing the wave number α. The net flow rate Q increases with
increasing the permeability parameter k within 0.0 < α < 0.004 while the
net flow rate Q decreases with increasing the permeability parameter k within
0.004 < α < 0.01.
• For non-Newtonian fluid (relaxation time tm = 1000.0), the net flow rate Q
attains a maximum for a certain value of the compressibility parameter χ and
there is shifting the maximum value of Q towards higher values of the com-
pressibility parameter χ’s with increasing the permeability parameter k. The net
flow rate Q decreases with increasing the permeability parameter the perme-
ability parameter k within 0.0 < χ < 0.4 while the net flow rate Q increases
with increasing k within 0.4 < χ < 1.0.
• For non-Newtonian fluid (relaxation time tm = 1000.0), the net flow rate Q
is weakly affected by the permeability parameter k when the wave number
α < 0.0015. The net flow rate Q decreases with increasing the permeability
of porous medium k within the range 0.0015 < α < 0.01.
• In deeply non-Newtonian regime (relaxation time tm = 10000.0), the net flow
rate Q increases with increasing χ. Also, the net flow rate Q decreases with
increasing the permeability parameter k. Furthermore, at high compressibility
(the compressibility parameter χ = 1.0) the net flow rate is weakly affected by
the permeability parameter k since the values of Q lie within the range of
(7.2595 × 10−5 –7.9463 × 10−5 ).
• In deeply non-Newtonian regime (relaxation time tm = 10000.0), Q becomes
high oscillatory, but what is unusual again is that we observe the negative flow
rates for certain values of α. Oscillatory behavior (appearance of numerous of
maxima in the behavior of a physical value) in the deeply non-Newtonian regime
is not new [22, 24]. Furthermore, we can observe that there is no effect (approxi-
mately) of the permeability parameter k on the net flow rate Q in the deeply
non-Newtonian regime.
• The results of [22] have been recovered by taking the permeability parameter
k → ∞.
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
• The results of [21] have been recovered by taking relaxation time tm = 0.0 and
the permeability parameter k → ∞.
• The results of [7] can be recovered by choosing the compressibility parameter
χ = 0.0 and relaxation time tm = 0.0.
• The results of [25] have been recovered by taking the compressibility parameter
χ = 0.0 relaxation time tm = 0.0 and the permeability parameter k → ∞.
References
[1] T. W. Latham, Fluid motion in peristaltic pumping, Master’s thesis (Massachusetts,
MIT Cambrige, 1966).
[2] T. Hayat, A. Nida and N. Ali, Peristaltic mechanism of Maxwell fluid in an asym-
metric channel, Nonlinear Anal. 9 (2008) 1474–1490.
[3] T. Hayat, Q. Hussain and N. Ali, Influence of partial slip on the peristaltic flow in a
porous medium, Phys. A 387 (2008) 3399–3409.
[4] T. Hayat, N. Ali and S. Asghar, Peristaltic motion of Burger’s fluid in a planner
channel, App. Math. Comput. 1 (2007) 309–329.
[5] T. Hayat and N. Ali, On mechanism of peristaltic flows for power-law fluids, Phys.
A 371 (2006) 188–194.
[6] T. Hayat, N. Ali, S. Asghar and A. M. Siddiqui, Exact peristaltic flow in tubes with
an endoscope, App. Math. Comput. 182 (2006) 359–368.
[7] E. F. El-Shehawey and W. El-Sebaei, Peristaltic transport in a cylindrical tube
through a porous medium, Int. J. Math. Math. Sci. 24 (2000) 217–230.
[8] Kh. S. Mekheimer, Nonlinear peristaltic transport through a porous medium in an
inclined planar channel, J. Porous Media 6 (2003) 189–201.
[9] Kh. S. Mekheimer, E. F. El-Shehawey and A. M. Alwa, Peristaltic motion of a particle
fluid suspension in planar channel, Int. J. Theor. Phys. 37 (1998) 2895–2919.
[10] Kh. S. Mekheimer and A. N. Abd-El-Wahab, Effect of wall properties on compressible
viscous fluid through a flexible pore, in The Proceeding of the Second International
Conference on MATHEMATICAL: Trends and Developments, organized by Egyptian
Mathematical Society ETMS, Vol. II, Bio. & Compl. Math. 27–30 Dec. Cairo Egypt
(2007), pp. 391–404.
[11] Kh. S. Mekheimer and Y. Abd Elmaboud, Peristaltic flow through a porous medium
in an annulus: Application of an endoscope, J. Appl. Math. Inform. Sci. 2(1) (2008)
103–111.
[12] Kh. S. Mekheimer and Y. Abd Elmaboud, Peristaltic flow of a couple stress fluid in
an annulus: Application of an endoscope, Phys. A 387 (2008) 2403–2415.
[13] M. Kothandapani and S. Srinivas, On the influence of wall properties in the MHD
peristaltic transport with heat transfer and porous medium, Phys. Lett. A 372 (2008)
4586–4591.
[14] M. Kothandapani and S. Srinivas, Non-linear peristaltic transport of a Newtonian
fluid in an inclined asymmetric channel through a porous medium, Phys. Lett. A 372
(2008) 1265–1276.
[15] K. Vajravelua, G. Radhakrishnamacharya and V. Radhakrishnamurty, Peristaltic
flow and heat transfer in a vertical porous annulus, with longwave approximation,
Int. J. Non-Linear Mech. 42 (2007) 754–759.
[16] M. Manoranjan and R. A. Ramachandra, Peristaltic transport in a channel with a
porous peripheral layer: Model of a flow in gastrointestinal tract, J. Biomech. 38
(2005) 779–789.
June 8, 2010 15:49 WSPC S1793-5245 242-IJB
S1793524510000970
[17] E. F. Elshehawey, N. T. Eldabe, E. M. Elghazy and A. Ebaid, Peristaltic transport in

an asymmetric channel through a porous medium, Appl. Math. Comput. 182 (2006)
140–150.
[18] T. Hayat, M. Khan, S. Asghar and A. M. Siddiqui, A mathematical model of peri-
stalsis in tubes through a porous medium, J Porous Media 9 (2006) 55–67.
[19] T. Hayat, N. Ali and S. Asghar, Hall effects on peristaltic flow of a Maxwell fluid in
a porous medium, Phys. Lett. A 363 (2007) 397–403.
[20] M. A. Abd Elnaby and M. H. Haroun, A new model for study the effect of wall
properties on peristaltic transport of a viscous fluid, to appear in Comm. Non-linear
Sci. Numer. Simm.
[21] A. C. T. Aarts and G. Ooms, Net flow of compressible viscous liquids induced by
traveling waves in porous media, J. Engin. Math. 34 (1998) 435–450.
[22] D. Tsiklauri and I. Beresnev, Non-Newtonian effects in the peristaltic flow of a
Maxwell fluid, Phys. Rev. E. 64 (2001) 036303.

[23] E. F. Elshehawey, A. El-Saman, M. El-Shahed and M. Dagher, Peristaltic transport
of a compressible viscous liquid through a tapered pore, Appl. Math. Comput. 169
(2005) 526–543.
[24] E. F. El-Shehawy, N. T. El-Dabe and I. M. Eldesoky, Slip effects on the peristaltic
flow of a non-Newtonian Maxwellian fluid, Acta Mech. 186 (2006) 141–159.
[25] F. Yin and Y. C. Fung, Peristaltic waves in circular cylindrical tubes, J. Appl. Mech.
36 (1969) 579–587.
[26] W. C. Tan and T. Masuoka, Stokes’ first problem for an Oldroyd-B fluid in a porous
half space, Phys. Fluid 17 (2005) 023101–023107.
[27] W. C. Tan and T. Masuoka, Stokes’ first problem for a second grade fluid in a porous
half-space with heated boundary, Int. J. Non-Linear Mech. 40 (2005) 515–522.
[28] C. Fetecau, Analytical solutions for non-Newtonian fluid flows in pipe-like domains,
Int. J. Non-Linear Mech. 39 (2004) 225–231.
[29] C. Fetecau and C. Fetecau, The first problem of Stokes for an Oldroyd-B fluid, Int.
J. Non-Linear Mech. 38 (2003) 1539–1544.
[30] T. Hayat, A. M. Siddiqui and S. Asghar, Some simple flows of an Oldroyd-B fluid,
Int. J. Engin. Sci. 39 (2001) 135–147.
[31] T. Hayat, Kh. Masood and M. Ayub, Exact solutions of flow problems of an
Oldroyd-B fluid, Appl. Math. Comput. 151 (2004) 105–119.
[32] S. Asghar, S. Parveen, S. Hanif, A. M. Siddiqui and T. Hayat, Hall effects on the
unsteady hydromagnetic flows of an Oldroyd-B fluid, Int. J. Engin. Sci. 41 (2003)
609–619.
[33] J. A. Del Rio, M. L. de Haro and Whitaker S., Enhancement in the dynamic response
of a viscoelastic fluid flowing in a tube, Phys. Rev. E. 58 (1998) 6323–6327.
[34] J. D. Anderson, Modern Compressible Flow with Historical Perspective (McGraw-Hill
Publishing Company, 1990).
[35] A. H. Nayfeh, Perturbation Methods (John Wiley & Sons, Inc., 1973).
[36] S. Takabatake, K. Ayukawa and A. Mori, Peristaltic pumping in circular cylindrical
tubes: A numerical study of fluid transport and its efficiency, J. Fluid Mech. 193
(1988) 267–283.
[37] K. J. Hutchison and E. Karpinski, In vivo demonstration of flow recirculation and
turbulence downstrea stenoses in canine arteries, J. Biomech. 18 (1985) 285–296.
[38] D. N. Ku, Blood flow in arteries, Annu. Rev. Fluid Mech. 29 (1997) 399–434.

Perturbation

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Perturbation

Uploaded by

Copyright:

Available Formats

June 8, 2010 15:49 WSPC S1793-5245 242-IJB

International Journal of Biomathematics

PERISTALTIC FLOW OF A COMPRESSIBLE

ISLAM M. ELDESOKY∗ and A. A. MOUSA†

Received 11 April 2009

Keywords: Peristaltic transport; Maxwell ﬂuid; compressible liquid; traveling waves;

256 I. M. Eldesoky & A. A. Mousa

To understand peristaltic action in diﬀerent situations, several theoretical and

through the artery. Acknowledge of ﬂow characteristics in the vicinity of stenosis

Peristaltic Flow of a Compressible Non-Newtonian Maxwellian Fluid 257

studied a mathematical model of peristalsis in tubes through a porous medium.

258 I. M. Eldesoky & A. A. Mousa

2. Basic Equations and Position of the Problem

Peristaltic Flow of a Compressible Non-Newtonian Maxwellian Fluid 259

where ρo is the constant density at the reference pressure po .

260 I. M. Eldesoky & A. A. Mousa

Equation (2.7) can be rewritten in the following form:

In cylindrical coordinates, the balance of mass equation (2.5) reads

Peristaltic Flow of a Compressible Non-Newtonian Maxwellian Fluid 261

Also the boundary conditions Eq. (2.10) becomes

η(z, t) = ε cos α(z − t). (2.21)

262 I. M. Eldesoky & A. A. Mousa

solution of the governing equations in a form

p = p0 + εp1 (r, z, t) + ε2 p2 (r, z, t) + · · · ,

To illustrate the nature of the solution, we shall consider the important

∂ρ1 ∂u1 u1 ∂ν1

Peristaltic Flow of a Compressible Non-Newtonian Maxwellian Fluid 263

u1 (r, z, t) = U1 (r)eiα(z−t) + U1 (r)e−iα(z−t) ,

u2 (r, z, t) = U20 (r) + U2 (r)e2iα(z−t) + U2 (r)e−2iα(z−t) ,

264 I. M. Eldesoky & A. A. Mousa

Here, the prime denotes a derivative with respect to r.

Peristaltic Flow of a Compressible Non-Newtonian Maxwellian Fluid 265

iαχP1 U1 − iαχP1 U1 + U1 U1 + U1 U1 + iαU1 V1 − iαU1 V1

266 I. M. Eldesoky & A. A. Mousa

U20 (r) = −χ[P1 (r)U1 (r) + P1 (r)U1 (r)]. (3.39)

We also obtain the general solution for V20 (r)as follows

where D2 is a complex constant deﬁned by

Peristaltic Flow of a Compressible Non-Newtonian Maxwellian Fluid 267

4. Numerical Results and Discussion

where the solution of Eq. (3.40) for V20 (r) is used.

268 I. M. Eldesoky & A. A. Mousa

Peristaltic Flow of a Compressible Non-Newtonian Maxwellian Fluid 269

270 I. M. Eldesoky & A. A. Mousa

Peristaltic Flow of a Compressible Non-Newtonian Maxwellian Fluid 271

0.000 0.002 0.004 0.006 0.008 0.010

272 I. M. Eldesoky & A. A. Mousa

Peristaltic Flow of a Compressible Non-Newtonian Maxwellian Fluid 273

274 I. M. Eldesoky & A. A. Mousa

Peristaltic Flow of a Compressible Non-Newtonian Maxwellian Fluid 275

[17] E. F. Elshehawey, N. T. Eldabe, E. M. Elghazy and A. Ebaid, Peristaltic transport in

Maxwell ﬂuid, Phys. Rev. E. 64 (2001) 036303.

You might also like