Turkish J. Eng. Env. Sci.

32 (2008) , 289 294.

ITAK

c TUB


The Eect of Injection/Suction on Flow past Parallel Plates with

Transpiration Cooling
Pawan Kumar SHARMA
Department of Applied Mathematics, Amity School of Engineering and
Technology, 580, Delhi Palam Vihar Road, Bijwasan, New Delhi-110061, INDIA
e-mail:drpawanksharma@yahoo.com
Baljinder Singh SAINI
Plot No. 6/G-5, Dhamoli Road, Rajpura Town, Distt. Patiala, Punjab-140401, INDIA

Received 13.05.2008

Abstract
This study investigated the eect of injection/suction between 2 horizontal parallel porous at plates,
with transverse sinusoidal injection of uid at the stationary plate and its corresponding removal by periodic
suction through the plate in motion, assuming the sinusoidal injection at the lower plate and its corresponding
removal by the upper plate in motion. The approximate solutions were obtained for the ow eld, pressure,
skin-friction, temperature eld, and rate of heat transfer, and are discussed with the help of graphs and
tables.
Key Words: Incompressible Fluid, Injection/Suction, Skin-Friction, Heat Transfer.

Introduction
Transpiration cooling methods have been developed
in an attempt to protect structural elements in turbojet and rocket engines from the inuence of hot
gases, such as combustion chamber walls, exhaust
nozzles, and gas turbine blades. They have many
engineering applications in the development of missiles, satellites, and spacecraft. In view of this Eckert (1958) obtained an exact solution of the plane
Couette ow with transpiration cooling. The problem remained 2 dimensional due to uniform injection
and suction applied at the porous plates. Flow and
heat transfer along a plane wall with periodic suction velocity has been studied by Gersten and Gross
(1974). Eects of such a suction velocity on various ow and heat transfer problems along at and
vertical porous plates have been studied by Singh et
al. (1978a, 1978b) and Singh (1993). Recently, the
problem of transpiration cooling with the application
of the transverse sinusoidal injection/suction velocity has been studied by Singh (1999). Chaudhary
and Jain (2007) studied exact solutions of incom-

pressible Couette ow with constant temperature

and constant heat ux on walls in the presence of
radiation. Recently, Sharma et al. (2007) discussed
the unsteady free convection oscillatory Couette ow
through a porous medium with periodic wall temperature. Hence, the aim of the present study was to examine the eects of injection/suction on the Couette
ow between 2 horizontal parallel porous plates, with
transverse sinusoidal injection of uid at the stationary plate and its corresponding removal by periodic
suction through the plate in motion. The governing
equations were solved for small amplitude oscillation
() of injection/suction velocity. The solution was
obtained by regular perturbation, in terms of . The
eect of various parameters on ow characteristics
were examined and are discussed with the help of
graphs and tables.

Formulation of the Problem

We considered the Couette ow of a viscous incompressible uid between 2 parallel at porous plates,
289

SHARMA, SAINI

with transverse sinusoidal injection of the uid at

the stationary plate and its corresponding removal
by periodic suction through the plate in average motion (U). Let x*-z* plane lie along the plates and
y*-axis be taken normal to the free stream velocity.
The distance (d) is taken between the plates. The
lower and upper plates are assumed to be at constant
temperature T0 and T1 , respectively, with T1 > T0 .
The plates are considered innite in the x direction.
Hence, all physical quantities will be independent of
x because an asymptotic ow was selected, assuming sinusoidal injection/suction velocity [V*(z*) =
V (1 + cos z*/d)] at the lower and upper plates.
The physical conguration of the problem is visualized in Figure 1 (Singh, 1999). Denoting the velocity
components u, v, and w in the x, y, and z directions,
respectively, and the temperature by , the problem
is governed by the following non-dimensional equations:
vy + wz = 0
(1)

where f stands for any of u, v, w, p, and . When

= 0, the problem is reduced to the well-known 2dimensional ows with constant injection and suction at both plates (Eckert, 1958). The solutions of
this 2-dimensional problem are

vuy + wuz = [uyy + uzz ]/

(2)

u0 (y) = (ey 1)/(e 1), (8)

vvy + wvz = py + [vyy + vzz ]/

(3)

vwy + wwz = pz + [wyy + wzz ]/

(4)

vy + wz = (yy + zz )/P r

(5)

where y = y*/d , z = z*/d, u = u*/U, v = v*/V, w

= w*/V, p = p / V2 , Pr (Prandtl number) = /,
(injection/suction parameter) Vd/, = (T T0 )
/ (T1 T0 ).
y*

All physical variables are dened in the Nomenclature. The (*) stands for dimensional quantities.
The corresponding boundary conditions in the dimensionless form are:
y = 0 : u = 0, v(z) = 1 + cos z, w = 0, = 0,
y = 1 : u = 1, v(z) = 1 + cos z, w = 0, = 1 (6)
Solution of the Problem
Since the amplitude of the injection/suction velocity
(<<1) is very small, we now assume the solution
of the following form:
f(y, z) = fo (y) + f1 (y, z) + 2 f2 (y, z) + . . ..

o (y) = (eP ry 1)/(eP r 1), v0 = 1, w0 = 0, p0 =

constant.
When = 0, substituting (7) in Eq. (1) to (5)
and comparing the coecient of , neglecting those
of 2 , 3 , etc., the following equations are obtained
with the help of solution (8):
v1y + w1z = 0

(9)

v1 uoy + u1y = (u1yy + u1zz )/

(10)

v1y = p1y + (v1yy + v1zz )/

(11)

w1y = p1z + (w1yy + w1zz )/

(12)

v1 oy + 1y = (1yy + 1zz )/P r

(13)

y = 0 : u1 = 0, v1 = cos z, w1 = 0, 1 = 0

V* (z* )

y = 1 : u1 = 0, v1 = cos z, w1 = 0, 1 = 0
d

T0
o

z*

V* (z* )

x*

Figure 1. The physical conguration of the problem.

290

(8)

The corresponding boundary conditions reduce to

T1

(7)

(14)

This is the set of linear partial dierential equations

that describe the ow. To solve these equations we
assume v1 , w1 , p1 , u1 , and 1 of the following form:

u1 (y, z) = u11 (y) cos z

v1 (y, z) = v11 (y) cos z

/
(15)
w1 (y, z) = {v11 (y)sinz}/

p1 (y, z) = p11 (y) cos z

1 (y, z) = 11 (y) cos z

SHARMA, SAINI

11yy P r11y 2 11 = P rv11 0y

where primes denote dierentiation with respect to

y. The expressions for v1 (y,z) and w1 (y,z) were chosen so that the equation of continuity (9) is satised.
Substituting (15) in (10) to (13) we obtain the following equations
u11yy u11y 2 u11 = v11 u0y

(16)

v11yy v11y 2 v11 = p11y

(17)

v11yyy v11yy 2 v11y = p11 2

(18)

Corresponding boundary conditions are:

y = 0 : u11 = 0, v11 = 1, v11y = 0, 11 = 0,

y = 1 : u11 = 0, v11 = 1, v11y = 0, 11 = 0
(20)
Solving Eqs. (16) to (19) under the boundary conditions (20) and using Eqs. (15), we get the solutions
for v1 , w1 , p1 , u1 , and 1 as

v1 (y, z) = A1 [A1 ey + A2 ey A3 ey A4 ey ] cos z.. . .(21)

1

w1 (y, z) = (A)

[A1 e

+ A2 e

p1 (y, z) = A
y

u1 (y, z) = [Ee

(+)y

+ C1 {A1 e

+ Fe

sy

1 (y, z) = [M e
1

+A2 ( + P r)
where

(+P r)y

.e

[A3 e

+ Ne

A3 e

+ A4 e

/2 + A2 e

(+)y

ty

(19)

+ A4 e

+ C2 {A1 ( + P r)

A3 (P r)

.e

(22)
(23)

(+)y

/2 A3 e

(+P r)y

]sinz

] cos z
1

(21)

()y

/ + A4 e

/}] cos z

(24)

(+P r)y

.e

+ A4 (P r)1 .e(P r)y }] cos z

(25)

A = 2( )(1 + e ) + ( + 2 )(e+ + e )
+( + 2 )(e + e+ ),
A1 = 2(1 + e ) ( 2 + )(e + e ) + ( 2 )(e+ + e ),
A2 = 2(1 + e ) + ( 2 + )(e + e ) ( 2 )(e+ + e ),
A3 = ( )(e+ + e ) ( + )(e + e ) + ( + )(e + e ),
A4 = ( )(e+ + e ) ( )(e+ + e ) + ( )(e+ + e ),
= [ + (2 + 4 2 )1/2 ]/2, = [ (2 + 4 2 )1/2 ]/2,
s = [P r + (2 P r 2 + 4 2 )1/2 ]/2, t = [P r (2 P r 2 + 4 2 )1/2 ]/2,
C1 = [A(e 1)]1 , C2 = 2 P r 2 [A(eP r 1)]1 ,
C3 = [A(e 1).(e e )]1 , C4 = 2 P r 2 [A(eP r 1)(es et )]1 ,
E = C3 [A1 (e e+ )/2 + A2 (e e+ )/2 A3 (e e+ )/ + A4 (e e )/],
F = C3 [A1 (e+ e )/2 + A2 (e+ e )/2 A3 (e+ e )/ + A4 (e e )/],
M = C4 [A1 ( + P r)1 (et e+P r ) + A2 (
+P r)1 (et e+P r ) A3 (P r)1 (et e+P r ) + A4 (P r)1 (et eP r )],
N = C4 [A1 ( + P r)1 (e+P r es ) + A2 (
+P r)1 (e+P r es ) A3 (P r)1 (e+P r es ) + A4 (P r)1 (eP r es )].

Now, after knowing the velocity eld, we can calculate skin-friction components xx and zz in the main and
transverse directions, respectively, as

xx = dxx
/U = (du0 /dy)y=0 + (du11 /dy)y=0 cos z

(26)

xx = (e 1)1 + [E + F + C1 {A1 ( + )/2 + A2 ( + )/2 A3 ( + )/ + A4 ( )/}] cos z (27)

291

SHARMA, SAINI

zz = dzz
/V = (w1 /y)y=0
1

zz = (A)

(28)

[A1 + A2 A3 A4 ]sinz
2

(29)

From the temperature eld we can obtain the heat transfer coecient, in terms of the Nusselt number as

/[(T0 T1 )] = (d0 /dy)y=0 + (d11 /dy)y=0 cos z

N u = dqw

(30)

N u = P r(P r 1)1 + [M s + N t + C2 {A1 ( + P r).( + P r)1

+A2 ( + P r).( + P r)1 A3 ( + P r).(P r)1 + A4 (P r ).(P r)1 }] cos z

It was observed that xx and zz decreased with increasing . It is also clear from Table 2 that the
values of zz were much lower than those of xx .

Results and Discussion

Main ow velocity proles are presented in Figure
2. This graph indicates that the main ow velocity
decreased with the injection/suction parameter ().
The transverse velocity component is presented in
Figure 3. It was observed that forward ow developed from y = 0 to about y = 0.5, and then onwards
there was backward ow. This was due to the fact
that the dragging action of the faster layer exerted
on the uid particles in the neighborhood of the stationary plate was sucient to overcome the adverse
pressure gradient and, hence, there was forward ow.
The dragging action of the faster layer exerted on the
uid particles reduced due to the periodic suction at
the upper plate and, hence, this dragging action was
insucient to overcome the adverse pressure gradient; therefore, there was backward ow. Furthermore, it is evident from Figure 3 that velocity w1
decreased with increasing in the forward and back
ow. Pressure values are reported in Table 1; pressure decreased with the injection/suction parameter
(). The decrease in pressure was suciently large
for small uid injection/suction.

w1 (transverse velocity)

0.2

0.8

u (velocity)

Figure 2. The velocity proles for = 0.2 and z = 0.

The values of skin-friction xx and zz in the main

and transverse ow directions are given in Table 2.
292

0.2

0.4

0.6

0.8

-0.1

y
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

0
0.6

Table 1. Pressure values of (p1 ) for z = 0.

0.2

Figure 3. The transverse velocity for z = 0.5.

0.4

0.4

0.05

-0.2

0.6

0.2

0.1

-0.15

=0
= 0.5
= 1.0

= 0.5
= 1.0
=0

0.15

-0.05

1
0.8

(31)

p1 ( = 0.2)
22.468
15.752
10.604
6.5114
3.0665
0.0732
3.2202
6.6877
10.820
16.030
22.835

p1 ( = 0.5)
8.8780
6.2182
4.1773
2.5520
1.1807
0.0773
1.3342
2.7281
4.3934
6.4960
9.2450

Nusselt number (Nu) values are shown in Figure

4. It was observed that Nu decreased with increasing in both situations [Pr = 0.71 (air) and Pr = 7
(water)]. It is also clear from Figure 4 that Nu was
much lower in the case of water (Pr = 7) than air
(Pr = 0.71).

SHARMA, SAINI

Nu (Nusselt number)

small values ( < 0.5) of the injection/suction parameter, while the reverse eect was observed for
> 0.5. The present analysis gave a better result,
as we considered the injection/suction velocity variable at both plates, because in actual practice injection/suction cannot be uniform in all cases.

0.8
0.6
0.4
Pr = 0.71
Pr = 7.0

0.2

Acknowledgements

0
0

0.2

0.4
0.6
0.8
(Injection/Suction parameter)

Figure 4. The Nusselt number for = 0.2 and z = 0.

Table 2. Skin-friction component (xx and zz ) values for

= 0.2.

0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0

xx (z = 0)
0.9508
0.9033
0.8575
0.8133
0.7707
0.7298
0.6905
0.6528
0.6166
0.5820

zz (z = 0.5)
0.3562
0.3529
0.3495
0.3462
0.3429
0.3395
0.3363
0.3330
0.3297
0.3265

Conclusions
On the basis of the above discussion we conclude
that the main ow velocity and skin friction components in the main and transverse ow directions
decreased with increases in the injection/suction parameter. Additionally, the dimensionless coecient
of heat transfer (Nusselt number) decreased with
the injection/suction parameter and the transverse
velocity component increased with increasing from

One of the authors, Dr. Pawan Kumar Sharma,

wishes to express sincere thanks to Prof. B.P Singh
(Senior Director) and Prof. Prem Prakash (Director)
for providing the necessary facilities for completion
of the work. We also humbly acknowledge the referees and Prof. S.C. Rajvanshi for their valuable
suggestions.
Nomenclature
d
distance between plates
Nu
Nusselt number
p
dimensionless pressure
pressure
p
Pr
Prandtl number
qw
constant heat ux per unit area
T
temperature of uid
T
temperature of free stream
U
average velocity
V
injection/suction velocity
u ,v ,w components of velocity
u,v,w
dimensionless velocity components
x ,y ,z
Cartesian coordinates
x,y,z
dimensionless Cartesian coordinates

thermal diusivity

amplitude of injection/suction velocity

(<< 1)

thermal conductivity

injection/suction parameter

viscosity

kinematics viscosity

dimensionless temperature

density

References
Chaudhary R.C. and Jain Preeti, Exact Solution of Incompressible Couette Flow with Constant Temperature
and Constant Flux on Walls in the Presence of Radiation. Turkish J. Eng. Env. Sci., 31, 297-304, 2007.
Eckert E.R., Heat and Mass Transfer, McGraw-Hill,
New York, 1958.

Gersten K. and Gross J.F., Flow and Heat Transfer

along a Plane Wall with Periodic Suction, J. Appl.
Math. Phys. (ZAMP), 25, 399-408, 1974.
Sharma Pawan Kumar, Sharma B.K and Chaudhary
R.C., Unsteady Free Convection Oscillatory Couette
Flow through a Porous Medium with Periodic Wall
Temperature, Tamkang Journal of Mathematics, 38 1,
93-102, Spring 2007

293

SHARMA, SAINI

Singh K.D., Three-Dimensional Couette Flow with

Transpiration Cooling, J. Appl. Math. Phys. (ZAMP),
50, 661-668, 1999.
Singh K.D., Three-Dimensional Viscous Flow and
Heat Transfer Along a Porous Plate, J. Appl. Math.
Mech. (ZAMM), 73, 58-61, 1993.

294

Singh P., Sharma V.P. and Misra U.N., ThreeDimensional Fluctuating Flow and Heat Transfer along
a Plate With Suction, Int. J. Heat Mass Transfer, 21,
1117-1123, 1978a.
Singh P., Sharma V.P. and Misra U.N., ThreeDimensional Free Convection Flow and Heat Transfer
along A Porous Vertical Wall, Appl. Sci. Res., 34, 105115, 1978b.