INTERNATIONAL JOURNAL OF MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES

Flow and heat transfer over an unsteady

stretching sheet in a micropolar fluid with
prescribed surface heat flux
N. Bachok, A. Ishak, and R. Nazar

Abstract The unsteady laminar flow of an incompressible

micropolar fluid over a stretching sheet with prescribed surface heat
flux is investigated. The governing partial differential boundary layer
equations are first transformed into ordinary differential equations
before being solved numerically by a finite-difference method. The
effects of the unsteadiness parameter, material parameter and Prandtl
number on the flow and heat transfer characteristics are studied. It is
found that the surface shear stress and the heat transfer rate at the
surface are higher for micropolar fluids compared to Newtonian
fluids.

KeywordsUnsteady flow, Heat transfer, Stretching sheet,

Micropolar fluid, Fluid mechanics.
I.

INTRODUCTION

HE fluid dynamics due to a stretching surface is important

in manufacturing processes. Examples are numerous and
they include the aerodynamic extrusion of plastic sheets, the
boundary layer along a liquid film in condensation processes,
paper production, glass blowing, metal spinning and drawing
plastic films. The thermal fluid flow problems have been
extensively studied numerically, theoretically as well as
experimentally (see [1,2]). The quality of the final product
depends on the rate of heat transfer at the stretching surface.
Crane [3] first obtained an elegant analytical solution to the
boundary layer equations for the problem of steady twodimensional flow due to a stretching surface in a quiescent
incompressible fluid. Gupta and Gupta [4] extended the
problem posed by Crane [3] to a permeable sheet and obtained
closed-form solution, while Grubka and Bobba [5] studied the
thermal field and presented the solutions in terms of Kummers
functions. The 3-dimensional case has been considered by
Wang [6]. Chen [7] studied the case when buoyancy force is
taken into consideration, and Magyari and Keller [8]
considered exponentially stretching surface. The heat transfer
over a stretching surface with variable surface heat flux has
Manuscript received July 28, 2010. This work was supported in part by the
IPTA Fundamental Research Grant.
N. Bachok is with the Department of Mathematics and Institute for
Mathematical Research, Universiti Putra Malaysia, 43400 UPM Serdang,
Selangor, Malaysia (phone: 603-8946-6849; fax: 603-8943-7958; e-mail:
norfifah@math.upm.edu.my).
A. Ishak and R. Nazar are with the School of Mathematical Science,
Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor, Malaysia (email: anuarishak@yahoo.com, rmn72my@yahoo.com).

Issue 3, Volume 4, 2010

been considered by Char and Chen [9], Lin and Cheng [10],
Elbashbeshy [11] and very recently by Ishak et al. [12].
All of the above mentioned studies dealt with stretching
sheet where the flows were assumed to be steady. The
unsteady flows due to a stretching sheet have received less
attention; a few of them are those considered by Devi et al.
[13], Andersson et al. [14], Nazar et al. [15], and very recently
by Ishak et al. [16]. In Ref. [15], the similarity transformation
introduced by Williams and Rhyne [17] was used, which
transformed the governing partial differential equations with
three independent variables to two independent variables,
which are more convenient for numerical computations.
Motivated by the above investigations, in this paper we
present the unsteady flow and heat transfer characteristics
caused by a stretching sheet immersed in a micropolar fluid.
The governing partial differential equations are transformed
into ordinary ones using similarity transformation, before
being solved numerically by the Keller-box method. The
results obtained are then compared with those of Elbashbeshy
[11] and the series solution for the steady-state flow case to
support their validity.
II. PROBLEM FORMULATION
Consider an unsteady, two-dimensional laminar flow of an
incompressible micropolar fluid over a stretching sheet. At
time t = 0 , the sheet is impulsively stretched with velocity
U w ( x, t ) along the x -axis, keeping the origin fixed in the fluid
of ambient temperature T . The stationary Cartesian
coordinate system has its origin located at the leading edge of
the sheet with the positive x -axis extending along the sheet,
while the y -axis is measured normal to the surface of the
sheet. The boundary layer equations may be written as [12,16]
u v
+
=0,
x y

u
u
u + 2 u N
,
+u
+v
=
+

t
x
y y 2 y
N

N
N
2 N
u
j
+u
+v
2N + ,
=
2

y
y
y

T
T
T
2T
+u
+v
= 2
t
x
y
y

167

(1)
(2)
(3)
(4)

INTERNATIONAL JOURNAL OF MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES

1
K
3

1 + h + f h f h K ( 2h + f ) S h + h = 0
2
2
2

1
1

+ f f S + = 0
Pr
2

subject to the boundary conditions

q
u
T
u = U w , v = 0, N = m
,
= w
y
y
k
u 0, N 0, T T as y ,

at

y = 0,

(5)

where m is the boundary parameter with 0 m 1 [18], u

and v are the velocity components in the x - and y directions, respectively, T is the fluid temperature in the
boundary layer, N is the microrotation or angular velocity, and
j , , , , , and are the microinertia per unit mass, spin
gradient viscosity, dynamic viscosity, vortex viscosity, fluid
density and thermal diffusivity, respectively. It is assumed that
the stretching velocity U w ( x, t ) and the surface heat flux

qw ( x, t ) are of the forms

ax
bx
U w ( x, t ) =
, q w ( x, t ) =
1 ct
1 ct

where a , b and c are constants with a > 0 , b 0 and c 0

(with ct < 1 ), and both a and c have dimension time 1 . It
should be noted that at t = 0 (initial motion), Eqs. (1) (4)
describe the steady flow over a stretching surface. These
particular forms of U w ( x, t ) and qw ( x, t ) have been chosen in
order to be able to devise a new similarity transformation,
which transforms the governing partial differential equations
(1) (4) into a set of ordinary differential equations, thereby
facilitating the exploration of the effects of the controlling
parameters (see Andersson et al. [14]).
As was shown by Ahmadi [19], the spin-gradient viscosity
can be defined as

1/ 2

y, = ( xU w )

1/ 2

1/ 2

U
N = Uw w
x

f ( ) ,

(8)

k ( T T ) U w
h ( ) , ( ) =

qw
vx

where is the similarity variable. The transformed nonlinear

ordinary differential equations are:
1
(1 + K ) f + ff f 2 + Kh S f + f = 0

Issue 3, Volume 4, 2010

f ( ) 0, h ( ) 0, ( ) 0 as .

(12)

The quantities of physical interest are the values of f (0) and

f ( ) = 1 e ,

( ) =

(13)

1 Pr M ( Pr 1, Pr + 1, Pr e
e
Pr
M ( Pr 1, Pr, Pr )

),

(14)

where M ( a, b, z ) denotes the confluent hypergeometric

function [21], with

an z n
,
n =1 bn n !

M ( a , b, z ) = 1 +

an = a ( a + 1)(a + 2) (a + n 1) ,
bn = b(b + 1)(b + 2) (b + n 1) .
By using Eqs. (13) and (14), the skin friction coefficient
f (0) and the surface temperature (0) can be shown to be
given by

f (0) = 1 ,

(0) =

(15)

1 M ( Pr 1, Pr + 1, Pr )
.
Pr M ( Pr 1, Pr, Pr )

(16)

The nonlinear ordinary differential equations (9) (12) have

been solved numerically by a finite-difference scheme known
as the Keller-box method, as described in the book by Cebeci
and Bradshaw [22], which is very familiar to the present
authors (see Bachok et al. [23,24] and Bachok and Ishak
[25,26]).

12

f (0) = 1, h(0) = mf (0), (0) = 1,

f (0) = 0,

The continuity equation (1) is satisfied by introducing a

stream function such that u = / y and v = / x .
The momentum, angular momentum and energy equations can
be transformed into the corresponding ordinary differential
equations by the following transformation:
Uw

where primes denote differentiation with respect to , Pr = /

is the Prandtl number and S = c / a is the unsteadiness
parameter. The boundary conditions (5) now become

(7)

where K = / is the dimensionless viscosity ratio and is called

the material parameter. Relation (6) is invoked to allow the
field of equations predicts the correct behavior in the limiting
case when the microstructure effects become negligible and
the total spin N reduces to the angular velocity [19,20].

(11)

1 (0) which represent the skin friction coefficient and the

heat transfer rate at the surface, respectively. Thus, our task is
to investigate how the governing parameters S , m, K and Pr
influence these quantities.
We note that when K = 0 (viscous fluid) and S = 0 (steady
flow), the problem under consideration reduces to a steadystate flow, where the closed-form solution for the flow field
and the solution for the thermal field in terms of Kummers
functions are respectively given by

(6)

= ( + / 2 ) j = (1 + K / 2 ) j ,

(10)

III. SOLUTION PROCEDURE

(9)

(i) Finite-difference method

168

INTERNATIONAL JOURNAL OF MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES

1 t j t j 1
t + ( f t ) j 1 2 ( p s ) j 1 2
Pr h j

To solve the transformed differential Eqs. (9) (11)

subjected to the boundary conditions (12), Eqs. (9) (11) are
first converted into a system of seven first-order equations, and
the difference equations are then expressed using central
differences. For this purpose, we introduce new dependent
variables p( ) , q( ) , g ( ) = h ( ) , n( ) , s( ) = ( ) and
t ( ) so that Eqs. (9) (11) can be written as
f = p,
p = q,
g = n,
s = t,

Rearranging of expressions (26)-(32) gives

(17)
(18)

1
f j f j 1 h j ( p j + p j 1 ) = 0,
2
1
p j p j 1 h j ( q j + q j 1 ) = 0,
2
1
g j g j 1 h j ( n j + n j 1 ) = 0,
2
1
s j s j 1 h j ( t j + t j 1 ) = 0,
2

(19)
(20)

1
(1 + K ) q + fq p 2 + Kn S p + q = 0 ,

1
K
3
1 + n + f n p g K ( 2 g + q ) S g + n = 0 ,
2
2
2

1
1

t+ f t p s S s + t = 0 .
Pr
2

(21)
(22)

p (0) = 1,

g (0) = mq (0), t (0) = 1,

g ( ) 0,

p ( ) 0,

s ( ) 0 as .

(25)

hj
p j p j 1
hj
g j g j 1
hj
s j s j 1
hj

(1 + K )

p j + p j 1
2

=
=

q j + q j 1
2
n j + n j 1
2

t j + t j 1
2

q j q j 1
hj

= p j 1 2 ,

(26)

= q j 1 2 ,

(27)

= n j 1 2 ,

(26)

= t j 1 2 ,

+ ( fq ) j 1 2 ( p 2 )

j 1 2

1 3
1

h j ( g j + g j 1 ) + ( n j + n j 1 ) = 0
2 2
2

1
1
t j t j 1 + h j f j + f j 1 t j + t j 1
Pr
4
1
h j p j + p j 1 s j + s j 1
4
1
1

S h j s j + s j 1 + t j + t j 1 = 0.
2
2

K 2 ( g ) j 1 2

(37)

(38)

(39)

Equations (33)-(39) are imposed for j = 1, 2, 3, ..., J , and the

(29)

transformed boundary layer thickness J is to be sufficiently

large so that it is beyond the edge of the boundary layer. The
boundary conditions are
f 0 = 0, p0 = 1, g 0 = mq0 , t0 = 1,
(40)
pJ = 0, g J = 0, s J = 0.

(30)

(ii) Newtons method

To linearize the nonlinear system (33)-(39), we use
Newtons method, by introducing the following expression:

f j( k +1) = f j( k ) + f j( k ) , p (jk +1) = p (jk ) + p(jk ) ,

q (j

(31)

k +1)

= q (j ) + q (j ) , g (j
k

k +1)

= g (j ) + g (j ) ,
k

n(jk +1) = n(jk ) + n (jk ) , s (jk +1) = s (jk ) + s (jk ) ,

1
3

+ ( q ) j 1 2 S ( g ) j 1 2 + ( n ) j 1 2 = 0

2
2

Issue 3, Volume 4, 2010

)(

)(

+ K ( n ) j 1 2

S ( p ) j 1 2 + ( q ) j 1 2 = 0
2

K n j n j 1
+ ( f n ) j 1 2 ( p g ) j 1 2
1 +
hj
2

(36)

number that indicates the coordinate location. The finitedifference approximation equations (17)-(23) are written for
the midpoint j 1 2 of the segment j 1 j . This procedure

(35)

1
K
1 + n j n j 1 + h j ( f j + f j 1 )( n j + n j 1 )
2
4

1
h j ( p j + p j 1 )( g j + g j 1 )
4
1
K h j 2 ( g j + g j 1 ) + ( q j + q j 1 )
2

(24)

where h j is the - spacing and j = 1, 2,..., J is a sequence

gives
f j f j 1

(34)

1
h j ( f j + f j 1 )( q j + q j 1 )
4
2
1
1
h j ( p j + p j 1 ) + h j K ( n j + n j 1 )
4
2
1
1

h j S ( p j + p j 1 ) + ( q j + q j 1 ) = 0
2
2

We now consider the segment j 1 j , with j 1 2 as the

midpoint, which is defined as below:
0 = 0, j = j 1 + h j , J = ,

(33)

(1 + K ) q j q j 1 +

(23)

In terms of the new dependent variables, the boundary

conditions (12) are given by

f (0) = 0,

(32)

S ( s ) j 1 2 + ( t ) j 1 2 = 0.
2

t (j

169

k +1)

= t (j ) + t (j
k

k)

(41)

INTERNATIONAL JOURNAL OF MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES

( c1 ) j

where k = 0,1, 2, ... . We then insert the left-hand side

expressions in place of f ( k ) , p ( k ) , q ( k ) , g ( k ) , n( k ) ,

( c2 ) j

s ( k ) and t ( k ) . This procedure yields the following linear

( c4 ) j

system (the superscript k is dropped for simplicity):

hj

( p
2

f j f j 1

+ p j 1 ) = ( r1 ) j 1 2 ,

(42)

+ q j 1 ) = ( r2 ) j 1 2 ,

(43)

+ n j 1 ) = ( r3 ) j 1 2 ,

(44)

hj

( q
2

p j p j 1

hj

( n
2

g j g j 1
s j s j 1

hj

( t
2

+ t j 1 ) = ( r4 ) j 1 2 ,

( a1 ) q j + ( a2 ) q j 1 + ( a3 ) f j +
( a4 ) f j 1 + ( a5 ) p j + ( a6 ) p j 1 +
( a7 ) n j + ( a8 ) n j 1 = ( r5 ) j 1 2 ,
( b1 ) n j + ( b2 ) n j 1 + ( b3 ) f j +
( b4 ) f j 1 + ( b5 ) p j + ( b6 ) p j 1 +
( b7 ) g j + ( b8 ) g j 1 + ( b9 ) q j +
( b10 ) q j 1 = ( r6 ) j 1 2 ,
( c1 ) t j + ( c2 ) t j 1 + ( c3 ) f j +
( c4 ) f j 1 + ( c5 ) p j + ( c6 ) p j 1 +
( c7 ) s j + ( c8 ) s j 1 = ( r7 ) j 1 2 ,

( c6 ) j
( c8 ) j
and

( r1 ) j 1 2 = f j + f j 1 + h j p j 1 2 ,
( r2 ) j 1 2 = p j + p j 1 + h j q j 1 2 ,
( r3 ) j 1 2 = g j + g j 1 + h j n j 1 2 ,
( r4 ) j 1 2 = s j + s j 1 + h j t j 1 2 ,

(45)

(46)

( r5 ) j 1 2 = (1 + K ) ( q j q j 1 ) h j ( fq ) j 1 2 +
hj ( p2 )
(47)

( r6 ) j 1 2
(48)

where

( a1 ) j

( r7 ) j 1 2
= 1+ K +

1
S
h j f j 1 2 h j ,
2
4

( a2 ) j = ( a1 ) j 2 (1 + K ) , ( a3 ) j
( a4 ) j = ( a3 ) j , ( a5 ) j

= h j p j 1 2

1
h j q j 1 2 ,
2
S
hj ,
2

( b4 ) j = ( b3 ) j , ( b5 ) j
( b6 ) j = ( b5 ) j ,

(50)

Sh j ( p ) j 1 2 + ( q ) j 1 2 ,
2

K
= 1 + ( n j n j 1 ) h j ( fn ) j 1 2 +
2

h j ( gp ) j 1 2 + Kh j 2 ( g ) j 1 2 + ( q ) j 1 2 +

S
h j 3 ( g ) j 1 2 + ( n ) j 1 2 ,

2
1
= ( t j t j 1 ) h j ( ft ) j 1 2 +
Pr

h j ( ps ) j 1 2 + Sh j ( s ) j 1 2 + ( t ) j 1 2 .
2

The boundary conditions (40) become

f 0 = 0, p0 = 0, g0 = 0, t0 = 0,
pJ = 0, g J = 0, sJ = 0,

1
3
= h j p j 1 2 Kh j Sh j ,
2
4
K
= ( b7 ) j , ( b9 ) j = h j , ( b10 ) j = ( b9 ) j ,
2

(51)

which just express the requirement for the boundary

conditionts to remain constant during the iteration process.
(iii) Block-elimination method
The linearized difference equations (42)-(48) can be solved
by the block-elimination method as outlined by Cebeci and
Bradshaw [22], since the system has block-tridiagonal
structure. Commonly, the block-tridiagonal structure consists
of variables or constants, but here an interesting feature can be
observed that it consists of block matrices. In a matrix-vector
form, Eqs. (42)-(48) can be written as

1
= h j g j 1 2 ,
2

Issue 3, Volume 4, 2010

Kh j ( n ) j 1 2 +

(49)

( b8 ) j

j 1 2

1
( a6 ) j = ( a5 ) j , ( a7 ) j = h j K , ( a8 ) j = ( a7 ) j ,
2
K 1
S
( b1 ) j = 1 + + h j f j 1 2 h j ,
2 2
4
K
1
( b2 ) j = ( b1 ) j 2 1 + , ( b3 ) j = h j n j 1 2 ,
2
2

( b7 ) j

1 1
S
+ h j f j 1 2 h j ,
Pr 2
4
2
1
= ( c1 ) j , ( c3 ) j = h j t j 1 2 ,
Pr
2
1
= ( c3 ) j , ( c5 ) j = h j s j 1 2 ,
2
1
S
= ( c5 ) j , ( c7 ) j = h j p j 1 2 h j ,
2
2
= ( c7 ) j ,

A =r
where

170

(52)

INTERNATIONAL JOURNAL OF MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES

[ A1 ]

[ B2 ]

A=

[C1 ]
[ A2 ] [C2 ]

0
[ BJ ] =
0

0
0

[CJ 1 ]
[ AJ ]

[ BJ 1 ] [ AJ 1 ]
[ BJ ]

[1 ]
[ r1 ]

[ 2 ]
[ r2 ]
= and r = .

[ J 1 ]
[ rJ 1 ]
[ ]
[r ]
J
J
The elements of the matrices are as follows:
0
0
0
1
h1
0
0
2

1
0
h1
0
2

[ A1 ] =
1
0
h1
0
2

( a2 )1 ( a8 )1
0

0
( b10 )1 ( b2 )1
0
0
c
(
2 )1

1
0
0
0

0
1
h1
2

0
0

( a3 )1 ( a1 )1 ( a7 )1
( b3 )1 ( b9 )1 ( b1 )1
0
( c3 ) 1 0

1
hJ
2

1
hJ
2

( b8 ) J
0

( a10 ) J ( a3 ) J ( a1 ) J
0
( b3 ) J ( b9 ) J
c
0
( 8 ) J ( c3 ) J

0 0

0 0

1
hJ
2

0 0

0 0
0 0
0 0

1
h1
2
0

0
( c1 ) 1

1
2 hJ

1
0
[CJ ] = 0
a
( 5 )J
( b5 ) J

( c5 ) J

(53)

0
1
hJ
2

( a4 ) J ( a2 ) J
( b4 ) J ( b10 ) J
0
( c4 ) J

0
0

( a8 ) J
( b2 ) J
0

1
hJ
2
0

0
( c2 ) J

(55)

( a7 ) J
( b1 ) J
0

0
0

1
( a9 ) J

( b7 ) J
0

( c7 ) J

0 0 0 0

0 0 0 0
0 0 0 0

0 0 0 0
0 0 0 0
0 0 0 0

0 0 0 0

(56)

2 j J,

q0
n
0
s0

[1 ] = f1 ,
q1

n1
t
1

2 j J,
1
2 hJ

0
[ AJ ] =
0

(a )
6 J
( b6 ) J

( c6 ) J

1 j J 1,

1
h1
2

0 0

0
(54)
1
hJ
2
0

0
( c1 ) J

q j 1
n
j 1
s j 1

j = f1
qj

nj
t
j

(57)

and 1 j J ,

( r1 ) j 1 2

( r2 ) j 1 2

( r3 ) j 1 2

rj = ( r4 ) j 1 2 .
r

( 5 ) j 1 2
( r )

6 j 1 2
( r7 )

j 1 2

(58)

To solve Eq. (52), we assume that A is nonsingular and it can

be factorized as

A = LU ,

Issue 3, Volume 4, 2010

171

(59)

INTERNATIONAL JOURNAL OF MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES

[ i ] W j = rj B j W j 1 ,

where

[1 ]

[ 2 ]
L=

[ 2 ]

[ J 1 ]
[J ]

[ J ]

2 j J.

(68)

When the elements of W have been found, Eq. (65) gives the
solution for in which the elements are found from the
following relations:

[ J ] = [WJ ] ,
[ i ] = W j j j +1 ,

(69)

1 j J 1.

(70)

and

[ I1 ]

U =

[1 ]
[ I1 ]

[ I1 ]

Once the elements of are found, Eqs. (42)-(48) can be

used to find the ( k + 1) th interation in Eq. (41). These

[ J 1 ]
[ I1 ]

calculations are repeated until the convergence criterion is

satisfied. In laminar boundary layer calculation, the wall shear
stress parameter q ( 0 ) is commonly used as the convergence
criterion (Cebeci and Bradshaw [22]). This is probably
because in boundary layer calculations, it is found that the
greatest error usually appears in the wall shear stress
parameter. Thus, this convergence criterion is used in the
present study. Calculations are stopped when

where [ I ] is a 7 7 identity matrix, while [ i ] and [ i ] are

7 7 matrices in which elements are determined by the
following equations:

[ i ] = [ A1 ] ,
[ A1 ][1 ] = [C1 ] ,
[i ] = [ A1 ] B j [ J 1 ] , j = 2, 3, ..., J ,
[ i ] j = C j , j = 2, 3, ..., J 1.

q0( k ) <1 ,

where 1 is a small prescribed value. In this study,

1 = 0.00001 is used, which gives about four decimal places
accuracy for most of the predicted quantities as suggested by
Bachok and Ishak [25,27] and Ali et al. [28,29].
The present method has a second-order accuracy,
unconditionally stable and is easy to be programmed, thus
making it highly attractive for production use. The only
disadvantage is the large amount of once-and-for-all algebra
needed to write the difference equations and to set up their
solutions.

(61)
(62)
(63)

Substituting Eq. (59) into Eq. (52), we obtain

r.

(64)

W,

(65)

LU

(71)

(60)

If we define

IV. RESULTS AND DISCUSSION

The step size in , and the position of the edge of the
boundary-layer have to be adjusted for difference values of
the parameters to maintain the necessary accuracy. In this
study, the values of between 0.001 and 0.1 were used,
depending on the values of the parameters considered, in order
that the numerical values obtained are mesh independent, at
least to four decimal places. However, a uniform grid of
= 0.01 was found to be satisfactory for a convergence

Eq. (64) becomes

====
r,

LW

(66)

where

====

[W1 ]

[W2 ]
,

[WJ -1 ]
[W ]
J

criterion of 105 which gives accuracy to four decimal places,

in nearly all cases. On the other hand, the boundary-layer
thickness between 5 and 50 was chosen where the infinity
boundary condition is achieved. The results are given to carry
out a parametric study showing the influences of the
unsteadiness parameter S , material parameter K and Prandtl
number Pr . For validation of the numerical method used in
this study, the case when S = 0 (steady-state flow) has also
been considered and the results are compared with those of
Elbashbeshy [11], as well as the series solution given by Eq.

and W j are 7 1 column matrices. The elements of W can

be determined from Eq. (65) by the following relations:

[1 ][W1 ] = [ r1 ] ,

Issue 3, Volume 4, 2010

(67)

172

INTERNATIONAL JOURNAL OF MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES

(16). The quantitative comparison is shown in Table 1 and it is

found to be in a very good agreement.
The velocity profiles for various values of S and K are
presented in Figs. 1 and 2. Figure 2 shows that the velocity
gradient at the surface is larger for larger values of S , which
produces larger skin friction coefficient f ( 0 ) . We note that
the parameter Pr have no influence on the flow field, which is
clear from Eqs. (9)-(11). It is evident from Figure 2 that the
boundary layer thickness increases with K . The velocity
gradient at the surface f ( 0 ) decreases as K increases.
Thus, micropolar fluids show drag reduction compared to
viscous fluids.
Figures 3-6 show the temperature profiles for selected
values of parameters. The temperature profiles are found to
subside monotonously to zero as increases. These curves
represent the physically realistic case. As can been seen from
Figs. 3-6, the surface temperature (0 ) decreases with

increases when S , K or Pr increases. Figure 7 shows that

the surface temperature (0) decreases with increasing values
of K. Thus, the heat transfer rate at the surface 1 (0) is higher
for a micropolar fluid ( K > 0 ) compared to a Newtonian fluid
(K = 0). On the other hand, for a fixed value of K , the surface
temperature (0) decreases when Pr is increased, i.e. the heat
transfer rate at the surface 1 (0) increases with Pr . This is
because the higher Prandtl number fluid has a lower thermal
conductivity (or a higher viscosity) which results in thinner
thermal boundary layer and hence, higher heat transfer rate at
the surface (see Fig. 5).
The effect of m on the angular velocity, when the other
parameters are fixed to unity is presented in Fig. 8. As
expected, the microrotation at the surface h ( 0 ) is more
dominant for larger values of m . Finally, Figs. 1-6 show that
the far field boundary conditions (12) are satisfied
asymptotically, thus supporting the numerical results obtained.

increasing S , K and Pr . Thus, the local Nusselt number

1 (0) , which represent the heat transfer rate at the surface

TABLE 1 Variations of (0 ) for different values of S and Pr

Pr

Elbashbeshy [11]

Eq. (16)

Numerical results

0.72
1

1.2253
1.0000

1.236657472
1.000000000

1.2367
1.0000

0.72
1

0.9116
0.8591

1
Pr = 1, m = 0.5, K = 1

0.8

0.8

0.7

0.7

0.6

0.6

0.5
S = 0, 0.5, 1, 2

0.4

0.5
0.4

0.3

0.3

0.2

0.2

0.1

0.1

0
0

Pr = 1, m = 0.5, S = 1

0.9

f ()

f ()

0.9

0
0

FIG. 1 Velocity profiles f ( ) for some values of S when

10

12

FIG. 2 Velocity profiles f ( ) for some values of K when

Pr = 1, m = 0.5 and K = 1

Issue 3, Volume 4, 2010

K = 0, 0.5, 1, 2, 3, 4

Pr = 1, m = 0.5 and S = 1

173

INTERNATIONAL JOURNAL OF MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES

1.2

1
Pr = 0.7, m = 0.5, K = 2

Pr = 1, m = 0.5, K = 1

0.9

0.8
0.7

0.8

()

( )

0.6

0.6

0.5
0.4

0.4

S = 0, 0.5, 0.9

0.3

S = 0, 0.5, 0.7, 0.9

0.2

0.2

0.1

0
0

0
0

10

10

FIG. 3 Temperature profiles ( ) for some values of S

FIG. 4 Temperature profiles ( ) for some values of S

when Pr = 0.7 , m = 0.5 and K = 2

when Pr = 1, m = 0.5 and K = 1

1.2

Pr = 1, m = 0.5, S = 0.1

0.9

S = 0.1, m = 0.5, K = 1

0.8
0.7

0.8

( )

()

0.6
0.6

0.5
0.4

Pr = 0.7, 0.8, 0.9, 1

0.4

0.3
K = 0, 1, 5

0.2
0.2

0.1
0
0

0
0

FIG. 6 Temperature profiles ( ) for some values of K

FIG. 5 Temperature profiles ( ) for some values of Pr

when Pr = 1, m = 0.5 and S = 0.1

when S = 0.1, m = 0.5 and K = 1

Issue 3, Volume 4, 2010

174

10

INTERNATIONAL JOURNAL OF MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES

1.5
1.4
1.3

0.8

1.2

0.7
Pr = 0.7

0.6

h( )

(0)

1.1

0.9

Pr = 1.0

0.5
0.4

0.8

m = 0, 0.5, 1, 2, 5, 10

0.3

0.7

0.2

0.6
0.5
0

S = 1, Pr = 1, K = 1

0.9

S = 0.1

0.1
1

0
0

FIG. 7 Variations of ( 0 ) with K for Pr = 0.7 and

FIG. 8 Angular velocity profiles h ( ) for some values of

Pr = 1.0 when S = 0.1

m when S = 1, Pr = 1 and K = 1

V. CONCLUSIONS

[7]

C. H. Chen, Laminar mixed convection adjacent to vertical,

continuously stretching sheets, Heat and Mass Transfer, vol. 33, pp.
471-476, 1998.

[8]

E. Magyari and B. Keller, Heat and mass transfer in the boundary

layers on an exponentially stretching continuous surface, J. Phys. D:
Appl. Phys, vol. 32, pp. 577-585, 1999.

[9]

M. I. Char and C. K. Chen, Temperature field in non-Newtonian flow

over a stretching plate with variable heat flux, Int. J. Heat Mass
Transfer, vol. 31, pp. 917-921, 1988.

We have theoretically studied the similarity solutions of the

unsteady boundary layer flow and heat transfer due to a
stretching surface. A new similarity solution has been devised,
which transform the time-dependent governing equations to
ordinary differential equations. We discussed the effects of the
governing parameters S, K and Pr on the fluid flow and heat
transfer characteristics. The numerical results compared very
well with previously reported cases, as well as the series
solution for the steady-state flow. We found that the heat
transfer rate at the surface 1 (0) increases with S, K and Pr.
Further, the heat transfer rate at the surface is higher for a
micropolar fluid compared to a Newtonian fluid.

[11] E. M. A. Elbashbeshy, Heat transfer over a stretching surface with

variable surface heat flux, J. Phys. D: Appl. Phys. Vol. 31, pp. 19511954, 1998.

REFERENCES

[12] A. Ishak, R. Nazar and I. Pop, Heat transfer over a stretching surface
with variable heat flux in micropolar fluids, Phys. Lett. A, vol. 372, pp.
559-561, 2008.

[10] C. R. Lin and C. K. Chen, Exact solution of heat transfer from

stretching surface with variable heat flux, Heat Mass Transfer, vol. 33,
pp. 477-480, 1998.

[1]

K. Abdell and F. Magpantay, Approximate analytic solutions for

mixed and forced convection heat transfer from an unsteady no-uniform
flow past a rotating cylinder, WSEAS Transactions on Heat and Mass
Transfer, vol. 2, pp. 6 16, 2007.

[2]

T. Al-Azab, Unsteady mixed convection heat and mass transfer past

an infinite porous plate which thermophoresis effect, WSEAS
Transactions on Heat and Mass Transfer, vol. 4, pp. 23-33, 2009.

[3]

L.J. Crane, Flow past a stretching plate, J. Appl. Math. Phys. (ZAMP),
vol. 21, pp. 645 647, 1970.

[15] R. Nazar, N. Amin and I. Pop, Unsteady boundary layer flow due to
stretching surface in a rotating fluid, Mech. Res. Commun., vol. 31,
pp. 121-128, 2004.

[4]

P. S. Gupta and A. S. Gupta, Heat and mass transfer on a stretching

sheet with suction or blowing, Can. J. Chem. Eng, vol. 55, pp. 744746, 1977.

[16] A. Ishak, R. Nazar an, I. Pop, Heat transfer over an unsteady stretching
permeable surface with prescribed wall temperature, Nonlinear Anal.
RWA, vol. 10, pp. 2909-2913, 2009.

[5]

L. J. Grubka and K. M. Bobba, Heat transfer characteristics of a

continuous stretching surface with variable temperature, ASME J.
Heat Transfer, vol. 107, pp. 248-250, 1985.

[6]

C. Y. Wang, The three-dimensional flow due to a stretching surface,

Phys. Fluids, vol. 27, pp. 1915-1917, 1984.

Issue 3, Volume 4, 2010

[13] C. D. S. Devi, H. S. Takhar and G. Nath, Unsteady mixed convection

flow in stagnation region adjacent to a vertical surface, Heat Mass
Transfer, vol. 26, pp. 71-79, 1991.
[14] H. I. Andersson, J. B. Aarseth and B. S. Dandapat, Heat transfer in a
liquid film on an unsteady stretching surface, Int. J. Heat Mass
Transfer, vol. 43, pp. 69-74, 2000.

[17] J.C. Williams III and T.B. Rhyne, Boundary layer development on a
wedge impulsively set into motion, SIAM J. Appl. Math., vol. 38, pp.
215-224, 1980.

175

INTERNATIONAL JOURNAL OF MATHEMATICAL MODELS AND METHODS IN APPLIED SCIENCES

[25] N. Bachok and A. Ishak, The effects of suction and injection on a

moving flat plate in a parallel stream with prescribed surface heat flux,
WSEAS Transactions on Heat and Mass Transfer, vol. 5, pp. 73-82,
2010.

[18] A. Ishak, R. Nazar and I. Pop, Mixed convection stagnation point flow
of a micropolar fluid towards a stretching sheet, Meccanica, vol. 43,
pp. 411-418, 2008.
[19] G. Ahmadi, Self-similar solution of incompressible micropolar
boundary layer flow over a semi-infinite plate, Int. J. Eng. Sci., vol.
14, pp. 639-646, 1976.

[26] N. Bachok and A. Ishak, Flow and heat transfer over a stretching
cylinder with prescribed surface heat flux, Malaysian Journal of
Mathematical Sciences, vol. 4, pp. 159-169, 2010.

[20] A. Ishak, R. Nazar and I. Pop, Moving wedge and flat plate in a
micropolar fluid, Int. J. Eng. Sci., vol. 44, pp. 1225-1236, 2006.

[27] N. Bachok and A. Ishak, Flow and heat transfer characteristic on a

moving flat plate in a parallel stream with prescribed surface heat flux,
Proceedings of the 9th WSEAS Int. Conf. on Appl of Comp Eng, pp. 115118, 2010.

[21] M. Abramowitz and I.A. Stegun, Handbook of Mathematical

Functions, Dover, New York, 1965.
[22] T. Cebeci and P. Bradshaw, Physical and Computational Aspects of
Convective Heat Transfer, Springer, New York, 1988.

[28] F. M. Ali, R. Nazar and N. M. Arifin, MHD viscous flow and heat
transfer due to a permeable shrinking sheet with prescribed surface heat
flux, Proceedings of the 9th WSEAS Int. Conf. on Appl of Comp Eng,
pp. 260-263, 2010.

[23] N. Bachok, A. Ishak and I. Pop, Mixed convection boundary layer

flow near the stagnation point on a vertical surface embedded in a
porous medium with anisotropy effect, Trans. Porous Med, vol. 82,
pp. 363-373, 2010.

[29] F. M. Ali, R. Nazar and N. M. Arifin, MHD viscous flow and heat
transfer induced by a permeable shrinking sheet with prescribed surface
heat flux, WSEAS Transactions on Mathematics, pp. 365-375, 2010.

[24] N. Bachok, A. Ishak and I. Pop, Boundary-layer flow of nanofluids

over a moving surface in a flowing fluid, Int. J. of Thermal Sciences,
vol. 49, pp. 1663-1668, 2010.

Issue 3, Volume 4, 2010

176