Alexandria Engineering Journal (2016) 55, 2023–2033

H O S T E D BY
Alexandria University

Alexandria Engineering Journal

www.elsevier.com/locate/aej
www.sciencedirect.com

ORIGINAL ARTICLE

Effect of heat transfer on unsteady MHD

flow of blood in a permeable vessel in the presence
of non-uniform heat source
A. Sinha a,*, J.C. Misra b, G.C. Shit a

a
Department of Mathematics, Jadavpur University, Jadavpur 700032, India
b
Indian Institute of Technology, Kharagpur, India

Received 17 June 2015; revised 8 July 2016; accepted 11 July 2016

Available online 24 August 2016

KEYWORDS Abstract This paper presents a theoretical analysis of blood flow and heat transfer in a permeable
Permeable capillary; vessel in the presence of an external magnetic field. The unsteadiness in the coupled flow and tem-
Unsteady stretching motion; perature fields is considered to be caused due to the time-dependent stretching velocity and the sur-
Non-uniform heat source/ face temperature of the vessel. The non-uniform heat source/sink effect on blood flow and heat
sink; transfer is taken into account. This study is of potential value in the clinical treatment of cardiovas-
Velocity-slip; cular disorders accompanied by accelerated circulation. The problem is treated mathematically by
Temperature-jump reducing it to a system of coupled nonlinear differential equations, which have been solved by using
similarity transformation and boundary layer approximation. The resulting nonlinear coupled ordi-
nary differential equations are solved numerically by using an implicit finite difference scheme.
Computational results are obtained for the velocity, temperature, the skin-friction coefficient and
the rate of heat transfer in the vessel. The estimated results are compared with another analytical
study reported earlier in scientific literatures. The present study reveals that the heat transfer rate
is enhanced as the value of the unsteadiness parameter increases, but it reduces as the space-
dependence parameter for heat source/sink increases.
Ó 2016 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V. This is an
open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction mathematical models have also been developed, which have

been applied to cryopreservation, as well as cryosurgery and
Although heat transfer is primarily an important domain of laser surgery. These models have also found useful
Thermal Engineering, attempts have been successfully made applications in various modern techniques of physiotherapy,
by several researchers to apply its principles to explore a vari- that involve heating the affected portion of the human body.
ety of information on how the body transfers heat [1]. Several Moreover, as discussed in our previous communication [2],
the therapeutic procedure of electromagnetic hyperthermia,
* Corresponding author. where the cancerous tissues are exposed to a thermal environ-
E-mail address: [email protected] (A. Sinha). ment of 42 °C, while maintaining the surrounding normal
Peer review under responsibility of Faculty of Engineering, Alexandria tissues at a suitable temperature, formulation and analysis of
University.
http://dx.doi.org/10.1016/j.aej.2016.07.010
1110-0168 Ó 2016 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
2024 A. Sinha et al.

Nomenclature

g dimensionless distance K thermal slip factor

u velocity of blood along the axis of the capillary M Hartmann number of the blood mass
v transverse velocity of blood in the capillary S unsteadiness parameter
T temperature of blood at any point within the cap- k buoyancy parameter
illary Sf non-dimensional velocity-slip factor
us velocity-slip at the wall St non-dimensional thermal slip factor
Uw stretching velocity Pr Prandtl number
Vw injection/suction velocity A injection/suction parameter
Tw surface temperature q density of blood
Ts thermal slip r electrical conductivity of blood
T0 a constant denoting the reference temperature m coefficient of kinematic viscosity of blood
g acceleration due to gravity b coefficient of thermal expansion
BðtÞ time-dependent magnetic field intensity cp specific heat at constant pressure
q000 non-uniform heat generation/absorption AH space-dependent heat generation/absorption
k1 ðtÞ time-dependent permeability parameter of blood BH temperature-dependent heat generation/absorp-
k2 constant permeability of the medium tion
k3 non-dimensional permeability parameter Cf skin-friction coefficient
k thermal conductivity of blood Nux local Nusselt number
N velocity-slip factor

mathematical models on heat transfer during blood flow have temperature of the body must always remain consistent. For
been found to be very useful. this purpose, it is necessary that if any excess heat is generated
In the human circulatory system, blood flowing through in the body due to any reason, it must have to be dissipated out
different branches of the arterial/venous tree, heat is dis- of the body as early as possible. The manner in which the
tributed through tissues evenly to the whole body and so heat transfer of heat in the living human body takes place is some-
cannot accumulate in any part of the tissue medium. The dif- what complicated, owing to the fact here it is a combination of
ferent components for calculating the total quantity of heat thermal conduction, convection and metabolic heat produc-
that blood carries when it flows through blood vessels include tion. Conduction (also called diffusion) takes place in tissues,
blood velocity, vessel diameter, thickness of blood, tempera- while convection occurs in the case of blood and other body
ture of the surrounding tissues and heat transfer coefficient fluids.
of blood. But since the first three factors are included in Rey- Convective heat transfer during blood flow in small tubes
nolds number, one can quantify the heat carried by blood only was studied analytically by Wang [5] in the case of fully devel-
through the consideration of temperature of the tissues, oped flow with constant heat flux.
together with the heat transfer coefficient of blood and the Pennes [6] formulated a simplified bio-heat transfer model
Reynolds number. in order to illustrate the effects of blood perfusion and meta-
Measurement of blood flow rate is of great interest to clin- bolic heat generation in living tissues. Although this model
icians, medical scientists and biomedical engineers. It is very bears the potential to describe the effect of blood flow on tissue
useful in the diagnosis of various arterial/venous diseases, temperature, it has some serious shortcomings, because he
e.g., atherosclerosis, atheroma. These diseases deter the normal assumed uniform perfusion rate and did not account for the
flow of blood. direction of blood flow. Moreover, in his model he considered
Variation in blood flow due to changes in the temperature only the stream of venous blood as the fluid stream equili-
was investigated by Barcroft and Edholm [3]. Barozzi and brated with tissue. Subsequently, several other models were
Dumas [4] used a numerical method to study the convective proposed by other researchers, in which some of the shortcom-
heat transfer in blood vessels of the circulatory system. They ings were removed. Nakayama and Kuwahara [7] presented a
observed that the rheological behavior of blood does not sig- mathematical analysis of the bio-heat transfer problem by
nificantly affect the heat transfer rate in small blood vessels. using volume averaging theory. Some of the shortcomings of
It is to be noted that the main source of heat produced in Pennes model [6] could be removed in this study. Keeping in
the human body is the process of metabolism of nutrients that view that large blood vessels can produce steep temperature
takes place in a ceaseless manner. This supplies energy to the gradients in heated tissues leading to inadequate tissue temper-
various organs of the body. It has been reported in the scien- atures during hyperthermia, Colios et al. [8] used numerical
tific literatures that in a human body at rest, about two- techniques to solve a problem concerning the cooling of large
thirds of the total heat is generated by the metabolic activities blood vessels in a heated tissue medium. Chato [9] investigated
of the internal organs in the thorax, the abdomen and the heat transfer to individual blood vessels in three different con-
brain; the share of the heat generated in the brain being around figurations: a single vessel, two vessels in counter flow, and a
16% of the total heat generated in the whole body. However, in single vessel near the skin surface. Heat transfer in perfused tis-
order that the human body can function properly, the internal sues in the presence of a vessel was studied by Rai and Rai [10],
Effect of heat transfer on unsteady MHD flow of blood 2025

considering metabolic heat generation and blood perfusion in of the arteries from the surrounding tissues. Although pulsatile
the tissue as temperature-dependent. flow through permeable walls is important in understanding
Nilsson [11] made an attempt to determine the relationship blood flow in the entire circulatory system, not much work
between skin blood flow and temperature. Another study on has been done in this direction.
the relationship between blood circulation and peripheral tem- Oka and Murata [24] made a theoretical study on the
perature was conducted both experimentally and numerically steady motion of blood through a permeable vessel obeying
by He et al. [12]. He et al. [13] reported another study of Starling’s law. Mariamma and Majhi [25] analyzed the steady
thermo-fluid model of blood circulation. In this study, the laminar flow of blood as a single phase of homogeneous New-
influence of bending stiffness of the vessel and that of blood tonian fluid in tubes with permeable walls. A discussion rele-
viscosity on blood pressure and temperature were analyzed. vant to the boundary conditions between a porous medium
The heat convection coefficient in large blood vessels was eval- and a viscous fluid was made by Beavers and Joseph [26]. Their
uated mathematically by Consiglieri et al. [14] to show that analysis and experiment were, however focused on a rectangu-
heating a large portion of blood vessel reduces convective heat lar channel with a porous wall in which fluid flows under the
loss. A physiology-based algorithm for simulating of a realistic action of a pressure gradient. Saffman [27] developed a gener-
vascular tree containing all thermally significant vessels in a tis- alized Darcy equation for non-uniform flows in non-
sue medium was suggested by Baish [15]. Some experimental homogeneous porous media. The boundaries were considered
investigations on the role of blood as heat source/sink in as discontinuous changes of properties of the porous media.
human organs and temperature variation along artery-vein The flow over a stretching surface has received attention of
pairs were carried out respectively by Ducharme and Tikuisis several researchers in the past. Misra and Shit [28,29] devel-
[16] and He et al. [17]. Srinivas et al. [18] theoretically studied oped mathematical models for investigating the flow of a vis-
an unsteady hydromagnetic heat and mass transfer of blood in coelastic fluid over a stretching sheet as well as in a channel
a time-dependent porous blood vessel over a permeable with stretching walls under the action of an externally applied
stretching sheet in the presence of thermal radiation, chemical magnetic field generated by a magnetic dipole. Misra et al. [30]
reaction and slip conditions. also investigated the flow and heat transfer of an MHD vis-
In the recent years, the study of magnetohydrodynamic coelastic fluid in a channel with stretching walls.
(MHD) flow of blood through arteries has gained considerable It is known that a viscous fluid normally sticks to a bound-
interest because of its important applications in physiology. It ary, i.e., there is no slip of the fluid relative to the boundary.
is known that blood is a suspension of various cells in plasma, Misra et al. [31] conducted a study concerning blood flow
the main bulk of the cells being erythrocytes. Since erythro- through a stenosed arterial segment, considering no-slip condi-
cytes have negative charge (although small), an applied mag- tion at the vessel wall. For many fluids, such as particulate flu-
netic field can influence the motion of erythrocytes; thus, ids, the motion is still governed by the Navier-Stokes
blood flow is affected due to the action of an external magnetic equations, but the usual no-slip condition at the boundary
field. The potential use of the magnetohydrodynamic princi- should be replaced by the slip condition [32]. Some authors
ples in the prevention and rational therapy of the arterial [33,34] suggested the presence of a red blood cell slip at the ves-
hypertension was explored by Vardanyan [19], who found that sel wall. Misra and Kar [35] developed a momentum integral
for steady flow of blood in an artery of circular cross section, a method to investigate the problem of blood flow through a ste-
uniform transverse magnetic field alters the flow rate of blood. nosed vessel by taking into consideration the slip velocity at
Sud and Sekhon [20] numerically studied the effect of magnetic the wall.
field on blood flow though the human arterial system. Haldar Modeling of hyperthermia-induced temperature distribu-
and Ghosh [21] investigated the effect of magnetic field on tion requires an accurate description of the mechanism of heat
blood flow through an indented tube in the presence of ery- transfer. It is reported in [36] that blood flow is affected by the
throcytes. According to the investigation reported by Bar- thermal response of living tissues. The rate of heat exchange
nothy [22], the heat rate decreases when a biological system between living tissues and the arterial network through blood
is exposed to an external magnetic field. The electrocardio- flows depends on the geometry of the blood vessels and the
graphic (ECG) pattern taken in the presence of a magnetic flow variation of blood. Misra and Sinha [37] theoretically
field can not only provide information on blood flow, but also investigated the effect of magnetic field and thermal radiation
offers a new non-invasive method for studying the cardiac per- on the flow and heat transfer in capillary blood vessels. They
formance. A three-parameter rheological model was used by reported that velocity of blood can be controlled by suitably
Kirkovskii et al. [23] in order to study the influence of a vari- regulating the intensity of the external magnetic field. Craci-
able magnetic field on the rheological properties of blood of unescu and Clegg [38] studied the effect of oscillatory flow
patients suffering from rheumatoid arthritis. The parameters on the resulting temperature distribution of blood and convec-
of the model are related to hydrodynamic properties of ery- tive heat transfer in rigid vessels. The study of heat generation
throcytes suspension. They observed that out of the three or absorption in moving fluids is also very much important in
parameters considered in their study, only one has the capacity the dynamics of thermo-fluids. Heat generation can alter the
to alter the rheological properties of blood to an appreciable temperature distribution in the fluid mass. Consequently, the
extent. particle deposition rate will be affected. Such studies bear the
The major activity of the entire cardiovascular system is to promise of important applications in the design of various
supply blood to tissues under sufficient pressure gradient to biomedical appliances, as well as in nuclear reactors, electronic
exchange materials through arterial wall. Small arteries are chips and semi-conductor wafers.
thin-walled and consist of endothelial cells. They contain In this study, we are interested in the unsteady fluid flow
ultra-microscopic pores through which substances of various and heat transfer of a fluid mass through a permeable vessel
molecular sizes can penetrate inside and pass into the lumen executing stretching motion under the action of an externally
2026 A. Sinha et al.
a
applied magnetic field. The problem investigated here involves the positive xdirection and the effective stretching rate 1ct
the effect of non-uniform heat source/sink. The thermal slip increase with time as 0 6 c < 1.
accompanied by velocity-slip at the capillary wall has been Here blood is considered as magnetohydrodynamic fluid
accounted for in the study. Because of the permeability of cap- properties independent of space and time at temperature T1 .
illary blood vessel walls, consideration of the slip-velocity at The walls of the artery are assumed to have non-uniform inter-
the wall makes the study closer to the reality. The analysis of nal heat source/sink and at the lower wall temperature Tw is
the model involves the solution of a set of nonlinear partial dif- assumed to vary with the distance x from the slit at time t as
ferential equations subject to the boundary conditions appro-
ax2
ð1  ctÞ2 ;
3
priate to the physical problem. In order to solve the Tw ¼ T1 þ T0 ð2Þ
boundary value problem, we have made an endeavor to 2m
develop a suitable numerical method that involves the use of where T0 is taken as a constant reference temperature and m is
finite differences, together with the method of discretization, the kinematic coefficient of viscosity of blood.
where the matrix of unknown physical variables is diagonally The expression of the lower wall temperature Tw ðx; tÞ given
dominant. The numerical results of various flow characteristics by (2) represents a situation in which the lower wall tempera-
are presented graphically. The results indicate very clearly the ture increases from T1 at the slit in proportional to x2 and
threshold value of the magnitude of the magnetic field strength such that the amount of temperature enhancement along the
that should be applied to the human body. The study will be of sheet decreases with time.
profound interest to medical surgeons in regulating blood flow The velocity-slip as well as the thermal slip will be taken
during surgery. into account.
The governing conservation equations of mass, momentum
2. Mathematical formulation and energy at unsteady state can be expressed as
@u @v
þ ¼ 0; ð3Þ
Let us consider the unsteady two-dimensional incompressible @x @y
laminar flow of blood through a permeable capillary under
the action of a time dependent magnetic field BðtÞ, which acts @u @u @u @ 2 u rB2 ðtÞ m
in a direction transverse to that of the flow. It is assumed that þu þv ¼m 2 u u  gbðT  T1 Þ;
@t @x @y @y q k1 ðtÞ
the magnetic Reynolds number is much less than unity so that
ð4Þ
the induced magnetic field is negligible in comparison with
applied magnetic field. The basic flow in the medium is entirely and
due to the buoyancy force caused by temperature difference
@T @T @T k @2T 1 000
between the wall and the inside medium. For the mathematical þu þv ¼ þ q : ð5Þ
analysis that follows we use Cartesian coordinates ðx; yÞ, where @t @x @y qcp @y2 qcp
x-axis is taken along the lower wall of the capillary, while y- The definition of all the symbols involved in the equations is
axis is along in the transverse direction (see Fig. 1). The flow included in the Nomenclature.
is driven by the stretching motion of the vessel wall, such that The last term on the right hand side of the Eq. (4)
the velocity of the stretching sheet Uw ðx; tÞ can be written as represents the influence of thermal buoyancy force and the
ax signs ‘±’ indicate the flow assisting and flow opposing
Uw ðx; tÞ ¼ ; ð1Þ
1  ct respectively.
The associated boundary conditions are
where a and c are both positive constants and ct < 1.
The expression (1) for the stretching velocity of the vessel @u
u ¼ Uw þ Nl ¼ Uw þ us ; v ¼ Vw ;
emphasizes that the vessel, which is fixed at a point (taken as @y
the origin of coordinates) is stretched by applying a force in @T
T ¼ Tw þ K ¼ Tw þ Ts at y ¼ 0 ð6Þ
@y
and
y
u ! 0; T ! T1 at y ! 1: ð7Þ
In Eq. (6), Vw represents the blood velocity at the vessel walls,
T=T u= 0 given by
rffiffiffiffiffiffiffiffiffiffiffiffi
ma
Vw ¼  fð0Þ; ð8Þ
1  ct
where Vw > 0 in the case of injection and Vw < 0 in the case of
g suction.
v=vw In Eq. (4), k1 ðtÞ ¼ k2 ð1  ctÞ represents the time-dependent
T=T w +Ts 1
x permeability parameter and in Eq. (6), N ¼ N0 ð1  ctÞ2
o 1
denotes the velocity-slip factor, and K ¼ K0 ð1  ctÞ2 stands
u=U w +u s for the thermal slip factor.
B(t) B(t)
The time-dependence of the applied transverse magnetic
Figure 1 Physical sketch of the problem. field is considered to be of the form
Effect of heat transfer on unsteady MHD flow of blood 2027

BðtÞ ¼ B0 ð1  ctÞ2 ;
1
ð9Þ 2.2. Skin-fiction coefficient and local Nusselt number
where B0 is a constant, representing the magnetic field at t ¼ 0.
The non-uniform heat generated or absorbed per unit vol- The physical quantity of most interest in a problem such as the
ume is taken as present one is the skin-friction coefficient (Cf ), which is defined
by the relation:
kUw  H  sw
q000 ¼ A ðTw  T1 Þeg þ BH ðT  T1 Þ ; ð10Þ Cf ¼ ; ð16Þ
xm qU2w =2
where AH and BH are parameters of space-dependent and where the skin-friction on the flat plate sw is given by
temperature-dependent heat generation/absorption (cf. [39]).  
AH and BH are both positive in the case of internal heat source @u
sw ¼ l : ð17Þ
and negative in the case of internal heat sink. Thus qm > 0 in @y y¼0
the case of heat generation; it is negative in the case of heat
Using (11) and (17), we have from (16):
absorption.
00
Cf ¼ 2Re
1
x f ð0Þ: ð18Þ
2

2.1. Similarity solution for momentum and heat transfer

Another important physical parameter of the current problem
equations
is the wall heat transfer coefficient (also called the local Nusselt
number), which is defined as
The governing Eqs. (4) and (5) admit a self-similar solution of
qw x
the form Nux ¼ ; ð19Þ
kðTw  T1 Þ
rffiffiffiffiffiffiffiffiffiffiffiffi rffiffiffiffiffiffiffi
ax 0 ma Uw T  T1 where the heat flux qw on the stretching sheet is defined as
u¼ f ; v¼ f; g ¼ y and h ¼ ;
1  ct 1  ct mx Tw  T1  
@T
ð11Þ qw ¼ k : ð20Þ
@y y¼0
where f is the dimensionless stream function and g is the sim-
ilarity variable. Again using (11) and (20) in (19), we get
Nux ¼ Re2x h0 ð0Þ:
1
It is evident that the non-dimensional variables satisfy the ð21Þ
continuity Eq. (3).
Substituting these non-dimensional variables in Eq. (4), we
obtain the following third-order nonlinear ordinary differen- 3. Computational procedure
tial equation:
The system of coupled nonlinear ordinary differential Eqs. (12)
1 1
f 000 þ ff 00  f 02  Sðf 0 þ gf 00 Þ  M2 f 0  f 0 þ kh ¼ 0; ð12Þ and (13) along with the boundary conditions (14) and (15) has
2 k3 been solved numerically by employing a finite difference
qffiffiffiffi
where M ¼ qa r
B0 is the Hartmann number, k3 ¼ akm 2 is the per- scheme. We use Newton’s linearization method to linearize
the discretized equations. The essential feature of this tech-
meability parameter, S ¼ ac is the unsteadiness parameter, nique is that it is based on a finite difference scheme, which
gbðTw T1 Þx 3
k ¼ Re
Grx
2 is the buoyancy parameter, Grx ¼ m
is the local has better stability characteristics and is quite efficient. It is
x

Grashof number and Rex ¼ Umw x is the local Reynolds number. found to yield sufficiently accurate results. Application of the
finite difference technique leads to a system, which is tri-
Here it is worthwhile to mention that k > 0 in the flow
diagonal and therefore has speedy convergence. Moreover, it
assisting region, k < 0 in the flow opposing region and k ¼ 0
brings about economical memory space to store the coeffi-
represent the case when the buoyancy forces are absent.
cients. The computational work has been carried out by taking
Using the transformation (11), we can obtain the trans-
g ¼ 0:0125. We have examined that further reduction in dg
formed energy equation from (5) as
does not produce any significant change. This ensures the sta-
 
1 bility of our numerical scheme. The finite difference scheme
h00  Pr Sð3h þ gh0 Þ þ 2f 0 h  h0 f þ AH eg þ BH h ¼ 0; ð13Þ that we have developed is briefly described below.
2
lc
Writing f 0 ¼ P in (12), we get
in which Pr ¼ kp is the Prandtl number.  
1 1
The boundary conditions (6) and (7) give rise to P00 þ fP0  P2  S P þ gP0  M2 P  P þ kh ¼ 0: ð22Þ
2 k3
fð0Þ ¼ A; f 0 ð0Þ ¼ 1 þ Sf f 00 ð0Þ and hð0Þ ¼ 1 þ St h0 ð0Þ ð14Þ
Also the boundary conditions (14) and (15) will now be of the
and form
f0 ðgÞ ! 0; hðgÞ ! 0 as g ! 1: ð15Þ fð0Þ ¼ A; Pð0Þ ¼ 1 þ Sf P0 ð0Þ; Pð1Þ ! 0: ð23Þ
In Eq. (13), A < 0 and A > 0 correspond to injection and suc- Using the central difference scheme for derivatives with respect
tion respectively. The non-dimensional velocity-slip factor Sf to g, we can write
and the non-dimensional thermal slip factor St are defined by
pffiffiffiffiffi qffiffi Giþ1  Gi1
Sf ¼ N0 q am and St ¼ K0 bm. ðG0 Þi ¼ þ OððdgÞ2 Þ ð24Þ
2dg
2028 A. Sinha et al.

and finite difference scheme with the help of Newton’s linearization

Giþ1  2Gi þ Gi1 method described in the previous section.
ðG00 Þi ¼ þ OððdgÞ2 Þ; ð25Þ For numerical solution it is necessary to assign some
ðdgÞ2
numerical values to the parameters involved in the problem
where G stands for P or h; i is the grid index in g-direction with under consideration. We find that the magnetic parameter M
gi ¼ i  dg; i = 0, 1, . . ., m; dg is the increment along the g-axis. is approximately 4 when the system is under the influence of
Newton’s linearization method can then be applied to lin- a strong magnetic field of strength B0 ¼ 8 T (Tesla), the blood
earize the discretized equations as described below. density q ¼ 1050 kg/m3 and the electrical conductivity of
When the values of the unknown functions at the n th iter- blood, r ¼ 0:8 s/m.
ation are known, the corresponding values of these variables at Experimental reports reveal that like most other fluids, for
the next iteration are obtained by using the equation human blood the viscosity l, specific heat at constant pressure
cp and the thermal conductivity k are temperature-dependent.
Gnþ1
i ¼ Gni þ ðDGi Þn ; ð26Þ
On the basis of their experimental studies, Valvano et al. [40]
n
in which ðDGi Þ represents the error at the n th iteration, and Chato [9] reported the following data for human blood
i ¼ 0; 1; 2; . . . ; n. It is worthwhile to mention here that the error at a temperature T ¼ 310 K:
ðDGi Þn at the boundary is zero, because the values of Gi at the l ¼ 3:2103 kg=ms; cp ¼ 14:65 J=kgK; and k ¼ 2:2103 J=ms K:
boundary are known. Using (26) in (22) and dropping the
quadratic and higher order terms in ðDGi Þn , we get a system Using these data, we find that the value of the Prandtl num-
of block tri-diagonal equations. To solve this tri-diagonal sys- lc
ber Pr ¼ kp for human blood is 21.
tem of equations, we have used the ‘‘Tri-diagonal matrix algo-
Numerical investigation was carried out with an aim to
rithm”, referred to as ‘‘Thomas algorithm”.
examine the variation of various physical variables for differ-
In the process of determination of the distribution of the
ent cases, in the range: 0 6 M 6 5; S ¼ 0; 0:3; 2:0; 5:0; A ¼
function fðgÞ, the accuracy can be defined as the difference
0; 0:5; 1:5; 2:0; k ¼ 15; 5; 5; 15; k3 ¼ 0:1; 0:2; 0:3; 0:4; Pr ¼
between the calculated values of fðgÞ at two successive opera-
18; 20; 21; 22; 23; 24; 27; Sf ¼ 0:25; St ¼ 0; 0:5; 1:0; 2:0; 0 6 AH
tions, say (n + 1)th and nth.
In the present study, the error  is equal to 6 1 and BH ¼ 0:5; 0:3; 0:05; 0:3; 0:5. The computational
work has been carried out by taking dg ¼ 0:0125 with 401 grid
 ¼ jf nþ1 ðgÞ  f n ðgÞj points. The results computed have been presented graphically
in Figs. 2–12. Fig. 13 establishes the validity of our results.
and is estimated to be less than 106 .
Distribution of non-dimensional axial velocity f0 for differ-
ent values of Hartmann number M, buoyancy parameter k,
4. Numerical estimates and related discussion
velocity slip factor Sf , porosity permeability k3 and unsteadi-
ness parameter S is depicted in Figs. 2–6. The presence of a
It is not possible to determine the exact solution of Eqs. (12) magnetic field in a direction normal to the flow in an electri-
and (13) because of the nonlinearity of the momentum and cally conducting fluid (e.g. blood) introduces a Lorentz force,
thermal boundary layer equations. Thus we had to obtain which has a tendency to resist the flow. The resistive force
the solution of the problem numerically. Appropriate similar- tends to slow down the flow and hence the blood velocity
ity transformation has been adopted to transform the govern- decreases with the increase in the magnetic field strength, as
ing partial differential equations of flow and heat transfer into observed from Fig. 2. This figure further indicates that blood
a system of nonlinear ordinary differential equations. The velocity in the vessel is reduced with a increase in the distance
resulting boundary value problem is solved by developing a from the lower wall. Fig. 3 gives an idea of the velocity

Figure 2 Velocity distribution for different values of M with Figure 3 Velocity distribution for different values of k, when
S ¼ 0:3; Sf ¼ 0:25; St ¼ 1:0; k3 ¼ 0:2; A ¼ 0:5; k ¼ 5:0; Pr ¼ 21:0; S ¼ 0:3; Sf ¼ 0:25; St ¼ 1:0; k3 ¼ 0:2; A ¼ 0:5; M ¼ 2:0; Pr ¼ 21:0;
AH ¼ 0:05 and BH ¼ 0:05. AH ¼ 0:05 and BH ¼ 0:05.
Effect of heat transfer on unsteady MHD flow of blood 2029

Figure 4 Velocity distribution for different values of Sf in a

situation, where S ¼ 0:3; M ¼ 2:0; St ¼ 1:0; k3 ¼ 0:2; A ¼ 0:5; Figure 6 Velocity distribution for different values of S, when
k ¼ 5:0; Pr ¼ 21:0; AH ¼ 0:05 and BH ¼ 0:05. M ¼ 2:0; Sf ¼ 0:25; St ¼ 1:0; k3 ¼ 0:2; A ¼ 0:5; k ¼ 5:0; Pr ¼ 21:0;
AH ¼ 0:05 and BH ¼ 0:05.

Figure 5 Velocity distribution for different values of k3 , when Figure 7 Temperature distribution for different values of Pr, in
S ¼ 0:3; Sf ¼ 0:25; St ¼ 1:0; M ¼ 2:0; A ¼ 0:5; k ¼ 5:0; Pr ¼ 21:0; the situation when M ¼ 2:0; Sf ¼ 0:25; St ¼ 1:0; k3 ¼ 0:2; A ¼ 0:5;
AH ¼ 0:05 and BH ¼ 0:05. k ¼ 5:0; S ¼ 0:3; AH ¼ 0:05 and BH ¼ 0:05.

distribution for different values of the buoyancy parameter k. monotonically with the velocity-slip parameter Sf . So for large
It is seen that blood velocity increases along the axis of the ves- values of Sf , the frictional resistance between the blood and the
sel, as the buoyancy parameter increases in the flow assisting vessel wall can be eliminated. Fig. 5 shows that blood velocity
region (k > 0), while it decreases as the buoyancy parameter increases along the vessel-axis, with a rise in permeability.
increases in the flow opposing region (k < 0). Physically, Moreover, for small permeability (k3 ¼ 0:1), the axial velocity
k > 0 means heating the fluid or cooling the stretching sheet, reduces faster than when the permeability is higher
while k < 0 implies cooling the fluid or heating the sheet, (k3 ¼ 0:2; 0:3; 0:4). Fig. 6 gives the distribution of axial velocity
and k ¼ 0 corresponds to the free convection. This figure fur- for different values of the unsteadiness parameter S in the case
ther shows that while cooling the walls of the vessel, the of cooling of the vessel wall. It may be observed that the influ-
momentum boundary layer thickness increases, whereas when ence of S on axial velocity is similar when compared with
blood is cooled, there will occur a reduction in the boundary Fig. 2.
layer thickness. It is also seen that in the case of heating of Figs. 7–9 give some characteristic temperature profiles for
the vessel walls, for large values of k, velocity of blood vanishes different values of Prandtl number (Pr), heat source/sink
at two different points. These points are the so-called points of parameter (BH ) and temperature-jump parameter (St ). Fig. 7
inflexion. Fig. 4 describes the effects of the velocity-slip factor demonstrates the effect of Prandtl number (Pr) on temperature
Sf on axial velocity in the case of cooling of the stretching wall. profile in the boundary layer. It may be observed that with the
Clearly velocity in the axial direction reduces when the slip- rise in Prandtl number, temperature of the boundary layer
velocity increases. In fact, the quantity 1  f0 ð0Þ increases diminishes. This may be attributed to the fact that the thermal
2030 A. Sinha et al.

Figure 10 Variation of skin-friction with M for different values

Figure 8 Temperature distribution for different values of BH ,
of suction parameter A, when St ¼ 1:0; Sf ¼ 0:25; Pr ¼ 21:0;
when M ¼ 2:0; Sf ¼ 0:25; St ¼ 1:0; k3 ¼ 0:2; A ¼ 0:5; k ¼ 5:0;
k3 ¼ 0:2; k ¼ 5:0; S ¼ 0:3; AH ¼ 0:05 and BH ¼ 0:05.
S ¼ 0:3; AH ¼ 0:05 and Pr ¼ 21:0.

Figure 9 Temperature distribution for different values of St , Figure 11 Variation of heat transfer coefficient with M for
with M ¼ 2:0; Sf ¼ 0:25; Pr ¼ 21:0; k3 ¼ 0:2; A ¼ 0:5; k ¼ 5:0; different values of Pr, in a situation when with
S ¼ 0:3; AH ¼ 0:05 and BH ¼ 0:05. St ¼ 1:0; Sf ¼ 0:25; A ¼ 0:5; k3 ¼ 0:2; k ¼ 5:0; S ¼ 0:3; AH ¼ 0:05
and BH ¼ 0:05.

that the temperature reduces when thermal slip enhances

boundary thickness decreases with a rise in Prandtl number.
(when BH < 0).
This figure further indicates that the temperature gradient at
Fig. 10 shows the variation of skin-friction with the Hart-
the surface increases with a rise in Prandtl number. This
mann number for different values of suction parameter A.
implies that an increase in Prandtl number is accompanied
by an enhancement of the heat transfer rate at the wall of From Eq. (18), it follows that the variation in f00 ð0Þ can rep-
the blood vessel. When blood attains a higher Prandtl number, resent the variation of the magnitude of the skin-friction coef-
its thermal conductivity is lowered down and so its heat con- ficient. It is seen that skin-friction increases as the suction
duction capacity diminishes. Thereby the thermal boundary parameter A increases. From the same figure, it may also be
layer thickness gets reduced. Thus the heat transfer rate at observed that for any value of suction parameter, the skin-
the capillary wall is increased. The temperature profiles for dif- friction increases with increase in the value of the magnetic
ferent temperature-dependent parameters for heat source/sink parameter M.
are presented in Fig. 8. In the case of heat generation (when Figs. 11 and 12 present the variation of local Nusselt num-
ber for different values of Prandtl number (Pr) and unsteadi-
BH > 0), temperature of blood increases with increasing BH ,
ness parameter S respectively. Fig. 11 shows that local
while the reverse trend is observed in the case of heat absorp-
Nusselt number increases as the value of Prandtl number
tion. Fig. 9 illustrates the changes that take place in the tem-
increases. It may be noted that the local Nusselt number
perature distribution when the values of thermal slip factor
diminishes as the magnetic field intensity is enhanced. Fig. 12
(St ) are changed. It is important to observe from this figure
Effect of heat transfer on unsteady MHD flow of blood 2031

1
fðgÞ ¼ f  egf ; so that f0 ðgÞ ¼ egf
f
In their analysis, f0 ¼ f  1f (with f > 0), and 0 < f < 1 and
f > 1 correspond to injection and suction respectively, while
in the analysis of our model, S stands for the injection/suction
parameter which is taken as 1.0.
One may observe from this figure that the results of the pre-
sent study are in good agreement with those of the previous
study [41].

5. Summary and conclusion

The problem of magneto-hydrodynamic flow of blood as well

as the heat transfer in a permeable blood vessel, having a heat
source/sink has been investigated here. The study is particu-
larly applicable to a situation, where the lumen of the blood
Figure 12 Variation of heat transfer coefficient with AH for vessel has turned into a porous structure owing to a patholog-
different values of the unsteadiness parameter S, when ical condition. The vessel is considered to execute an unsteady
M ¼ 2:0; A ¼ 0:5; St ¼ 1:0; Sf ¼ 0:25; Pr ¼ 21:0; k3 ¼ 0:2; k ¼ 5:0 stretching motion. The velocity-slip at wall and the thermal
and BH ¼ 0:05. slip have been duly accounted for. The nonlinear governing
equations are solved numerically by developing a suitable
numerical technique with the help of Newton’s linearization
method.
The study bears the potential to explore some important
information regarding the complex flow behavior of blood
in situations where all the ten physical parameters
M; A; S; k; k3 ; Sf ; St ; Pr; AH and BH play prominent roles in
the hemodynamical flow and heat transfer in small blood ves-
sels. The results presented here should be useful to clinicians,
hematologists and biomedical engineers, because they serve
as useful estimates, which are capable of throwing some light
toward the understanding of genesis of pathological states, like
arteriosclerosis as well as the mechanism of gaseous exchange
that takes place within tissues and blood vessels. The following
conclusions can be drawn on the basis of the numerical results
reported in the preceding section.

(i) By the application of sufficiently strong magnetic field,

blood velocity can be diminished. This is accompanied
Figure 13 Velocity distribution function f0 ðgÞ: comparison of
by a reduction in the velocity gradient at the vessel wall,
our results with those of Ishak et al. [41] for
whereby the local skin-friction can be reduced. The
M ¼ S ¼ Sf ¼ k ¼ 0:0; A ¼ 1:0 and k3 ¼ 100:0.
results presented should be of significant interest to sur-
geons who usually want to keep the blood flow rate at a
indicates that the heat transfer rate is reduced as the space- desired level during the entire surgical procedure.
dependent parameter for heat source AH increases, while the (ii) The buoyancy parameter has a dual effect on the axial
heat transfer rate increases with the increase in the unsteadi- velocity. The axial velocity increases in the flow assisting
ness parameter S. region (k > 0) and decreases in the flow opposing region
In Fig. 13, results of a previous investigation [41] have been (k < 0).
presented along with the results of the present study. For the (iii) The velocity of blood along the axis of the capillary
purpose of comparison, both the studies have been naturally increases with a rise in permeability.
brought to the same platform, by considering a vanishing (iv) By increasing the Prandtl number Pr, it is possible to
value for the magnetic field parameter and also by disregarding bring about a reduction in the thermal boundary layer
the slip effects for the present study. For comparison, we also thickness.
neglect the heat transfer effect, by setting k ¼ 0. Further, we (v) With a rise in heat generation, thermal boundary layer
assumed that the flow is steady, by taking the unsteadiness thickness increases, but a reverse trend is observed in
parameter S ¼ 0 and we considered in our study a large value the case of heat absorption. This observation is of par-
of the permeability parameter (k3 ¼ 100:0). ticular importance in the therapeutic procedure of elec-
The analytical solution for fðgÞ presented by Ishak et al. [41] tromagnetic hyperthermia used in the treatment of
is given by cancer, because the therapy involves rising the tempera-
ture of the cancerous tissues above 42 °C. An increase in
2032 A. Sinha et al.

temperature brings down blood viscosity. While apply- [16] M.B. Ducharme, P. Tikuisis, Role of blood as heat source or
ing the procedure of hyperthermia, it is necessary to sink in human limbs during local cooling and heating, J. Appl.
see that blood viscosity does not increase, because in Physiol. 76 (1994) 2084–2094.
that case there may arise various problems in respect [17] Q. He, D.E. Lemons, S. Weinbaum, L. Zhu, Experimental
measurements of the temperature variation along artery-vein
of circulation of blood.
pairs from 200 to 1000 lm diameter in rat hind limb, J. Biomech.
Eng. 124 (2002) 656–661.
The present study can also be applied to deal with other [18] S. Srinivas, P.B.A. Reddy, B.S.R.V. Prasad, Effects of chemical
problems that deal with different types of nonlinear variations reaction and thermal radiation on MHD flow over an inclined
of the velocity of the stretching surface subject to the action of permeable stretching surface with non-uniform heat source/sink:
a magnetic field, under the preview of the assumptions made An application to dynamics of blood flow, J. Mech. Med. Biol.
here. Of course, the input values of the different physical 14 (2014) 1450067:1–1450067:24.
parameters need to be changed as and when necessary. [19] V.A. Vardanyan, Effects of magnetic field on blood flow,
Biofizika 18 (1973) 491–496.
[20] V.K. Sud, G.S. Sekhon, Blood flow through the human arterial
Acknowledgment
system in the presence of a steady magnetic field, Physiol. Med.
Biol. 34 (1989) 795–805.
One of the authors (A. Sinha) is grateful to the NBHM, DAE, [21] K. Haldar, S.N. Ghosh, Effect of a magnetic field on blood flow
Mumbai for the financial support of this investigation. through an indented tube in the presence of erythrocytes, Ind. J.
Pure Appl. Math. 25 (1994) 345–352.
References [22] M.F. Barnothy, Biological Effects of Magnetic Fields, Plenum
Press, New York, 1964, pp. 1964–1969.
[1] Y.A. Cengel, A.J. Ghajar, Heat and Mass Transfer: [23] V.V. Kirkovskii, V.A. Mansurov, N.P. Mit’kovskaya, Y.
Fundamentals and Applications, fourth ed., McGraw-Hill, Mukharskaya, Influence of a variable magnetic field on the
2010. rheological properties of blood in treatment of rheumatoid
[2] J.C. Misra, A. Sinha, G.C. Shit, Flow of a biomagnetic arthritis, J. Eng. Phys. Therm. Phys. 76 (2003) 708–714.
viscoelastic fluid: application to estimation of blood flow in [24] S. Oka, T. Murata, A theoretical study of the flow of blood in a
arteries during electromagnetic hyperthermia, a therapeutic capillary with permeable wall, Jpn. J. Appl. Phys. 9 (1970) 345–
procedure for cancer treatment, Appl. Math. Mech. – Engl. 352.
Ed. 31 (2010) 1405–1420. [25] N.K. Mariamma, S.N. Majhi, Flow of Newtonian fluid in a
[3] H. Barcroft, O.G. Edholm, The effect of temperature on blood blood vessel with permeable wall – a theoretical model, Comp.
flow and deep temperature in human forearm, J. Physiol. 102 Math. Appl. 40 (2000) 1419–1432.
(1943) 5–20. [26] G.S. Beavers, D.D. Joseph, Boundary conditions at a naturally
[4] G.S. Barozzi, A.J. Dumas, Convective heat transfer coefficients permeable wall, J. Fluid. Mech. 30 (1967) 197–207.
in the circulation, J. Biomech. Eng. 113 (1991) 308–310. [27] P.G. Saffman, On the boundary condition at the surface of a
[5] C.Y. Wang, Heat transfer to blood flow in a small tube, J. porous medium, Stud. Appl. Math. 2 (1971) 93–101.
Biomech. Eng. 130 (2008) 024501, http://dx.doi.org/10.1115/ [28] J.C. Misra, G.C. Shit, Biomagnetic viscoelastic fluid flow over a
1.2898722. stretching sheet, Appl. Math. Comput. 210 (2009) 350–361.
[6] H.H. Pennes, Analysis of tissue and arterial blood temperatures [29] J.C. Misra, G.C. Shit, Flow of a biomagnetic viscoelastic fluid in
in the resting forearm, J. Appl. Physiol. 1 (1948) 93–122. a channel with stretching walls, J. Appl. Mech. 76 (2009) 061006.
[7] A. Nakayama, F. Kuwahara, A general bioheat transfer model [30] J.C. Misra, G.C. Shit, H.J. Rath, Flow and heat transfer of an
based on the theory of porous media, Int. J. Heat Mass Transfer MHD viscoelastic fluid in a channel with stretching walls: some
51 (2008) 3190–3199. applications to hemodynamics, Comput. Fluids 37 (2008) 1–11.
[8] M.C. Colios, M.D. Sherar, J.W. Hunt, Large blood vessel [31] J.C. Misra, M.K. Patra, S.C. Misra, A non-Newtonian fluid
cooling in heated tissues: a numerical study, Phys. Med. Biol. 40 model for blood flow through arteries under the stenotic
(1995) 477–494. conditions, J. Biomech. 26 (1993) 1129–1141.
[9] J.C. Chato, Heat transfer to blood vessels, J. Biomech. Eng. 102 [32] W.A. Ebart, E.M. Sparrow, Slip flow in rectangular and annular
(1980) 110–118. ducts, J. Basic Eng. 87 (1965) 1018–1024.
[10] K.N. Rai, S.K. Rai, Effect of metabolic heat generation and [33] P. Brunn, The velocity slip of polar fluids, Rheol. Acta 14 (1975)
blood perfusion on the heat transfer in the tissues with a blood 1039–1054.
vessel, J. Biomech. Eng. 35 (1999) 75–79. [34] Y. Nubar, Blood flow, slip and viscometry, J. Biophys. 11 (1971)
[11] A.L. Nilsson, Blood flow, temperature and heat loss of skin 252–264.
exposed to local radiative and convective cooling, J. Invest. [35] J.C. Misra, B.M. Kar, Momentum integral method for studying
Dermatol. 88 (1987) 586–593. flow characteristics of blood through a stenosed vessel,
[12] Y. He, H. Liu, R. Himeno, M. Shirazaki, Numerical and Biorheology 26 (1989) 23–25.
experimental study on relationship between blood circulation [36] M.W. Dewhirst, T.V. Samulski, Hyperthemia in the Treatment
and peripheral temperature, J. Mech. Med. Biol. 5 (2005) 39–53. for Cancer, Upjhon, Kalamazoo, MI, 1988.
[13] Y. He, H. Liu, R. Himeno, A one-dimensional thermo-fluid [37] J.C. Misra, A. Sinha, Effect of thermal radiation on MHD flow
model of blood circulation in the human upper limb, Int. J. Heat of blood and heat transfer in a permeable capillary in stretching
Mass Transfer 47 (2004) 2735–2745. motion, Heat Mass Transfer 49 (2013) 617–628.
[14] L. Consiglieri, I. dos Sanotosh, D. Haemmerich, Theoretical [38] O.I. Craciunescu, S.T. Clegg, Pulsatile blood flow effects on
analysis of the heat convection coefficient in large vessels and the temperature distribution and heat transfer in rigid vessels, J.
significance for thermal ablative therapies, Phys. Med. Biol. 48 Biomech. Eng. 123 (2001) 500–505.
(2003) 4125–4134. [39] R. Tsai, K.H. Huang, J.S. Huang, Flow and heat transfer over
[15] J.W. Baish, Formulation of a statistical model of heat transfer in an unsteady stretching surface with non-uniform heat source,
perfused tissues, J. Biomech. Eng. 116 (1994) 521–527. Int. Commun. Heat Mass Transfer 35 (2008) 1340–1343.
Effect of heat transfer on unsteady MHD flow of blood 2033

[40] J.W. Valvano, S. Nho, G.T. Anderson, Analysis of the [41] A. Ishak, R. Nazar, I. Pop, Heat transfer over an unsteady
Weinbaum-Jiji model of blood flow in the canine kidney stretching permeable surface with prescribed wall temperature,
cortex for self-heated thermistorsrom, J. Biomech. Eng. 116 Nonlinear Anal.: Real World Appl. 10 (2009) 2909–2913.
(1994) 201–207.