European Journal of Mechanics / B Fluids 83 (2020) 42–57

Contents lists available at ScienceDirect

European Journal of Mechanics / B Fluids

journal homepage: www.elsevier.com/locate/ejmflu

Modeling and simulation of blood flow with magnetic nanoparticles as

carrier for targeted drug delivery in the stenosed artery
∗
Sreeparna Majee, G.C. Shit
Department of Mathematics, Jadavpur University, Kolkata 700032, India

article info a b s t r a c t

Article history: A systematic study on targeted drug delivery is carried out in an unsteady flow of blood infused with
Received 3 September 2019 magnetic nanoparticles with an aim to understand the flow pattern and nanoparticle aggregation in
Received in revised form 6 March 2020 a diseased arterial segment having atherosclerosis. The magnetic NPs(nanoparticles) are supervised
Accepted 9 April 2020
by a magnetic field, which is significant for the therapeutic treatment of arterial diseases, tumors,
Available online 14 April 2020
and cancer cells and removing blood clots. Coupled thermal energy equation has been modeled by
Keywords: considering the dissipation of energy that encounters due to the application of the magnetic field
Magnetic nanoparticles and the presence of high viscosity of blood. The simulation technique used to solve the mathematical
Blood flow model is vorticity-stream function formulations in the diseased artery. An elevation in SLP (Specific
Targeted drug delivery loss power) is noted in the aortic bloodstream when the agglomeration of nanoparticles is higher. This
Atherosclerosis phenomenon has potential applications in the treatment of hyperthermia. The study focuses on the
Hyperthermia lowering of WSS (wall shear stress) with increasing particle concentration at the downstream of the
stenosis, which depicts the vigorous flow circulation zone. These low shear stress regions prolong the
residing time of the nanoparticles carrying drugs, which soaks up the LDL (Low-Density Lipoprotein)
deposition. Moreover, an increase in NP concentration enhances the Nusselt number, which marks
the increase of heat transfer from the arterial wall to the surrounding tissues to destroy tumor and
cancer cells without affecting the healthy cells. The results have a significant influence on the study
of medicine to treat arterial diseases such as atherosclerosis without the need for surgery, which can
minimize the expenditures on cardiovascular treatments and post-surgical complications in patients.
© 2020 Elsevier Masson SAS. All rights reserved.

Targeted drug delivery is a mechanism of delivering medi- delivery system, which combines drugs and magnetic nanopar-
cation to an affected organism in a pattern that enhances the ticles to affect the target area. Drochon et al. [1] presented a
aggregation of the medication in the targeted tissue at the cellular review on the analytical solutions of the magnetohydrodynamic
or subcellular level. The medium of delivery of drugs largely flow of blood with an aim to provide biomedical applications
depends on nanomedicine, which employs nanoparticles, loaded in Magnetic Resonance Imaging (MRI), magnetic drug targeting,
with drugs, and transported to specific parts of the body (tar- tissue engineering, and blood pulse energy harvesting etc. In a
geted area) where the disease persists, avoiding interaction with similar context, metachronal wave propagation of magnetized
healthy tissue. The significance of the targeted drug delivery particle–fluid suspension was studied by Abdelsalam et al. [2]
system has an aim to confine, protract, and protected drug in- for MHD pumping applications and Waqas et al. [3] studied for
teraction with the diseased tissue. The targeted release system is bioconvection flow of a magnetized nanofluid for beneficial in
primarily effective for minimizing the frequency of drug dosages manufacturing processes and the enhancement of transport of
taken by the patient and its fluctuation in drug levels. This energy.
method acts as a therapeutic agent for a prolonged period to a Magnetic micro and nanoparticles are widely used in the fields
of microbiology, biomedicine, and biotechnology for the applica-
targeted diseased area within the body and prevents damage to
tion in targeted therapeutic drugs [4–7]. They are either encap-
the healthy tissue due to the drug inclusion. Several scientists are
sulated into a magnetic micro or nanosphere or injected into the
currently interested in the targeted drug delivery system due to
arterial path to the affected tissues [8–10]. Lübbe et al. [11] first
its benefits of capturing efficiency in the target area. Magnetically
clinically tested magnetically controlled drug targeting by bond-
controlled drug targeting is the most widely used in the drug
ing anticancer agents with ferrofluids and observed that the drug
releases from the ferrofluid and enfolds the tumor. Magnetic par-
∗ Corresponding author. ticles are becoming hugely accepted due to its non-toxic nature.
E-mail address: [email protected] (G.C. Shit). This is used for treating various cardiovascular diseases involving

https://doi.org/10.1016/j.euromechflu.2020.04.004
0997-7546/© 2020 Elsevier Masson SAS. All rights reserved.
 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57 43

thrombosis, stenosis, aneurysm, atherosclerosis etc. Moreover, simulations. Computational simulations were also developed by
drug targeting to cancerous tissues [12] and hyperthermia treat- Rotariu and Strachan [31] that evaluated the flow of magnetic
ment to malignant tumor cells are some significant application nanoparticles targeted at a tumor site deep inside the body.
areas of magnetic particles. It is ideal for probing and manipulat- Sharma et al. [32,33] studied the capturing efficiency of magnetic
ing biological particles and systems by enabling it for appropriate nanoparticles in the targeted area under the influence of the ex-
surface treatments and coating, which enhances the ability for ternal magnetic field. After that, a mathematically modeled path
specific binding with the target biomaterial. Furthermore, mag- trajectory of a cluster of magnetic nanoparticles in a blood vessel
netic nanoparticles can be superparamagnetised by an external under the magnetic effect is generated by Sharma et al. [34].
magnetic field to detect and target specific diseased areas [13]. They have also optimized the position of outside magnet for more
The use of magnetic drug targeting (MDT) is expanding with effective drug targeting at the tumor site. Nacev et al. [35] studied
functionalized magnetic nanoparticles, which has various use in the use of a magnetic drug delivery system to deal with magnetic
chemo and gene therapy at tumor site [14]. The behavior of nanoparticle transportation in the blood vessel. However, an in-
magnetic nanoparticles transport for MDT in the vascular level crease in the thermal conduction was noted by Sheikholeslami
is reviewed in many works [15–17]. The most commonly used
and Shehzad [36] when magnetic nanoparticles are augmented.
magnetic nanoparticles include iron oxides (Fe2 O3 and Fe3 O4 ),
Magnetic hyperthermia has promising prospects in numer-
manganese ferrite (MnFe2 O4 ), cobalt ferrite (CoFe2 O4 ) etc. Berry
ous applications as electromagnetic energy can be converted to
and Curtis [18] observed that the magnetite nanoparticle (Fe3 O4 )
thermal energy using magnetic nanoparticles at the nanoscale
is generally preferred in MDT due to its biocompatibility and large
level. The most common form is the destruction of malignant
magnetization. Yokoyama [19] analyzed the present and future
cells by heating them up to their apoptosis threshold [37,38].
prospective drug targeting concept in blood flow and has ob-
Another emerging strategy for cancer treatment is heating pro-
served a difference in viscosity of blood having magnetic particles
and in the hydrodynamic case under a magnetic environment. tocol induced by an alternating magnetic field, which in turn
Riegler et al. [20] first carried out the experiment of cell tar- release drug molecules into the surrounding tissues of the tar-
geting using Magnetic Resonance Imaging (MRI) scanner, which geted area by heating at nanoscale [39]. Several experiments
demonstrated that living human cells are labeled with different were performed to demonstrate the significance of hyperthermia
iron oxide particles within MRI scanner, can be targeted in a associated with chemotherapy and radiation for the treatment
vascular vessel under magnetic field effect. Moreover, simulation of cancer [40]. Experimental evaluation for an increase in mi-
of magnetic capturing of drug carriers in the brain vascular sys- crowave heating by iron oxide nanoparticles has been studied by
tem is investigated by Kenjeres and Righolt [21]. They have also McWilliams et al. [41] for improving microwave hyperthermia.
observed that magnetically targeted drug delivery sufficiently Haase and Nowack [42] illustrated that the magnetic dipolar in-
enhances the particle capturing efficiency in the targeted region. teractions between the nanoparticles cause a decrease in heating
The magnetic nanoparticles are loaded with drugs to stay for power when there is an increase in NP concentration in the
a longer time and work on the diseased part of the targeted ferrofluid. This phenomenon is known as Specific Loss Power
region. Statins are long known drugs to absorb cholesterol that (SLP). Mehdaoui et al. [43] simulated that similar NP dipolar
are built up as plaques containing LDL (Low-Density Lipoprotein) interactions with low anisotropy enhance the SLP. This proves
in the arteries. However, a new study has been published, in that the interparticular distance within the NPs in the flow regime
which plaques that have been built up in the heart vessel are has a significant influence on their heating efficiency. Highly con-
more effectively shrunk by PCSK9 inhibitor evolocumab com- centrated nanoparticles in liposomes increase the SLP [44], while
bined with a statin. The U.S. Food and Drug Administration (FDA) it decreases when confined in a polystyrene matrix [45]. How-
approved this first new class of cholesterol-lowering drugs. This ever, de Sousa et al. [46] stated a decrease in heating efficiency
drug prevents cholesterol to form and deposit in the blood vessel exhibited by SLP while examining the dispersion state of NPs
and, moreover, soaks up LDL that is already deposited in the functionalized with citrate ligands. Hamid et al. [47] simulated
arteries. Cohen et al. [22] established that PCSK9 is likely an hydromagnetic flow and heat transfer analysis with magnetic
attractive therapeutic target. These drugs can then be embedded nanoparticles to investigate the dual numerical solution for such
with magnetic nanoparticles, which are targeted to release the flow.
drugs for effectively work on the cholesterol depositions in the
The present paper focuses on modeling and simulation of
arteries.
targeted drug delivery in a stenosed artery by using magnetic
A mathematical model was proposed by Furlani and Furlani
nanoparticles having an external influence of a magnetic field.
[23] in which they considered blood behaved as a Newtonian fluid
The blood is considered to be Newtonian in the aorta due to
and observed that magnetic targeting of multifunctional carrier
high shear stress. The trajectory of a cluster of magnetic nanopar-
particles that deliver therapeutic agents to malignant tissue in
ticles is developed which is injected in the blood vessel near
vivo. This work is extended by Shaw et al. [24] and Shaw and
the stenosed area and is controlled by an external magnetic
Murthy [25] by considering Herschel–Bulkley fluid as blood for
both permeable and impermeable microvessels. Later on, Mon- field applied perpendicular to the flow of blood. The simulation
dal and Shit [26] examined transport phenomena of magnetic technique used to solve the mathematical model is the vorticity-
nanoparticles during electro-osmotic flow in a microvessel in stream function formulations and the results are obtained im-
the presence of external magnetic field and electrokinetic effect. portant varying parameters such as nanoparticle concentration
Moreover, Shaw and Murthy [27] and Shaw et al. [28] considered and nanoparticle size. The rise in temperature using magnetic
two-phase Casson’s fluid model and observed capture condition nanoparticles in the form of hyperthermic treatment is also per-
for the carrier particle. Larimi et al. [29] numerically investigated formed in the stenosed area and observed an increasing SLP.
MDT technique by particle tracking in the presence of a magnetic This study has the potential to treat arterial diseases such as
field in a bifurcation vessel. They have considered blood as a atherosclerosis without the need for surgery, which can minimize
Newtonian fluid and used magnetic iron oxide nanoparticles as the cost of treatment and post-surgical complications. Further-
drug carriers. Simulations on drug targeting infused with mag- more, this proposition has enormous applications in the treat-
netic carriers are developed by Ritter et al. [30], by employing ment of different genres of diseases such as tumors, infections,
a high gradient magnetic separation technique using FEMLAB and removing blood clots.
44 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57

1. Mathematical model has been neglected from the present analysis because of the fact
that in comparison to the applied magnetic field strength, the
1.1. Governing equations induced magnetic field strength is very small as the electrical
conductivity of blood is very low (σ = 0.5 s/m). The values of
In this paper, an unsteady two dimensional and fully de- these physical variables in dimensional form are collected from
veloped blood flow (at the entry) through a stenosed artery is the study [23,48].
considered (cf. Fig. 1), into which magnetic nanoparticles are
injected in the blood stream and captured them at a specific
location under the influence of an externally applied magnetic 1.2. Fluidic force formulation
field. The cylindrical polar co-ordinates (r ∗ , θ ∗ , x∗ ) along x∗ as the
axis of symmetry of the artery is chosen and assumed that all the
flow characteristics are independent of θ ∗ due to axi-symmetry The motion of a spherical shaped magnetic nanoparticle is
of flow. Motivated from the study of Drochon [48] on the mag- considered in the flow of blood under the influence of applied
netohydrodynamic flow of blood, the governing equations for magnetic field. In this problem, we have neglected the Brownian
unsteady two-dimensional MHD fluid flow along with the ther- motion by restricting our attention to low Reynolds number flow
mal energy equation and magnetic nanoparticle velocity [13,23, regimes where the magnetic and viscous forces are dominated.
33] are taken as Therefore, the equations of motion of nanoparticle using the
∂ u∗ v∗ ∂v ∗ classical Newtonian dynamics is given by:
+ + = 0, (1)
∂ x∗ r∗ ∂r∗ ∂vpx
∗
m = Fmx + Ffx + Fgx , (6)
∂t∗
∂u ∗ ∂u ∗ ∂u
( ∗ ∗ ∗
)
ρ +u +v ∂vpy
∗
∂t∗ ∂ x∗ ∂r∗
m = Fmy + Ffy + Fgy , (7)
∂p ∗
∂ u
2 ∗
1 ∂ u∗ ∂ 2 u∗ ∂t∗
( )
=− ∗ +µ + +
∂x ∂ r ∗2 r∗ ∂r∗ ∂ x∗2 where m = 34 π R3p ρp is the mass of the nanoparticle, ρp is the
+ Ks N vpx − u∗ − σ B20 u∗ , density of the nanoparticle, Rp (= 0.5 µm) is the radius of the
( ∗ )
(2)
nanoparticle, Fmx , Fmy , Ffx , Ffy , Fgx , Fgy are the magnetic, fluidic and
∂v ∗ gravitational forces along x− and y− directions respectively. The
∗ ∂v ∗ ∂v
∗ ∗
( )
ρ + u + v expression for magnetic force is Fm = µ0 Vp (Mp · ∇ )H, ⃗ where
∂t∗ ∂ x∗ ∂r∗
µ0 (= 4π × 10−7 ) is the permeability of the free space, Mp the
∂p ∂ v 1 ∂v ∗ ∂ 2 v∗ v∗
∗
( 2 ∗ )
=− ∗ +µ + ∗ ∗ + ∗2 − ∗2 magnetization of the nanoparticle, Vp = 34 π R3p the volume of
∂r ∂ r ∗2 r ∂r ∂x r ⃗ is the magnetic field intensity vector. We
the nanoparticle and H
+ Ks N vpy − v ,
( ∗ ∗
)
(3) have assumed the induced magnetic field is negligible due to low
magnetic Reynolds number and hence the force components of
∂T ∗ ∗ ∂T
∗
∗ ∂T
∗ magnetic field on the nanoparticles along x− and y− directions
( )
ρ Cp + u + v are vanish (Fmx = 0, Fmy = 0). The gravitation force can be
∂t∗ ∂ x∗ ∂r∗
defined as Fgx = 0 and Fgy = −Vp (ρp − ρ )g with g = 9.8 m/s2 , in
∂ T 1 ∂T ∂ 2T ∗
( 2 ∗ ∗
)
=k + + + σ B20 u∗2 + φ ∗ , (4) which the horizontal gravitational force is neglected [49,50].
∂ r ∗2 r∗ ∂r∗ ∂ x∗ 2 When the inertial term m ∂ tpx
∂v ∗ py ∂v ∗
∗ and m ∂ t ∗ is ignored due to the
where negligible mass of the nanoparticles, the particle motion satisfies
the following simple force balance Fmx + Ffx + Fgx = 0 and
[ ( )2 )2 )2 ]
∂v ∗ ∂v ∗ ∂ u∗ ∂ u∗
( (
φ =µ 2
∗
+ + +2 (5) Fmy + Ffy + Fgy = 0.
∂r∗ ∂ x∗ ∂r∗ ∂ x∗
The fluidic force exerted by a nanoparticle is obtained by using
is the viscous dissipation term. Stokes’ law for the viscous drag on the sphere as
In the above, the velocity of blood components u∗ and v ∗ as
well as nanoparticles velocity components vpx ∗
and vpy
∗
are re- Ffx = −6πµRhyd,p fDp (vpx
∗
− u∗ ) , (8)
spectively taken along the axial and radial directions, p∗ denotes
the blood pressure, µ the coefficient of dynamic viscosity (µ = Ffy = −6πµRhyd,p fDp (vpy
∗
− v ∗ ). (9)
4 × 10−3 Pa s), σ the electrical conductivity, ρ the density of
blood (ρ = 1050 kg m−3 ), B0 the applied magnetic field strength, In this paper, we ignored gravity effects on submicron nanopar-
Ks the Stokes constant (Ks = 3.17 × 10−17 ), N the number of ticles and focus instead on the magnetic and fluidic forces. When
magnetic nanoparticles per unit volume (N = 3.78 × 1020 ), T ∗ the force on the nanoparticle is fluidic, the Eqs. (6) and (7) reduce
the temperature variable, Cp the specific heat at constant pressure to
(Cp = 0.94 J/Kg K) and k represents the thermal conductivity
∂vpx
∗
of blood (k = 0.492 ± 0.009 W/mK). The last term of Eq. (2) m = Ks (u∗ − vpx
∗
), (10)
−
→ − → ∂t∗
depicts the magnetic body force ( J × B ) per unit volume, in
−
→ −
→ − → − →
which J = σ [ E + V × B ] is the current density due to ∂vpy
∗
−
→ −
→ m = Ks (v ∗ − vpy
∗
), (11)
Ohm’s law, B (B0 , 0, 0) is the magnetic field vector and E is the ∂t∗
induced electric field. The second term on the right hand side of
Eq. (4) is the dissipation of energy due to applied magnetic field where Ks = 6πµRhyd,p fDp in which Rhyd,p is the effective hydro-
and the last term φ ∗ represents the dissipation of energy due to dynamic radius of the nanoparticle (Rhyd,p = 2.5 × 10−7 ) and
the presence of blood viscosity. A magnetic field of strength B0 fDp is the drag coefficient. The drag coefficient can be defined as
2F
is uniformly applied perpendicular to the axial direction of blood fDp = ρ u2DA where FD (drag force of blood) ≈ 7 × 10−13 Newtons
flow throughout the artery. The effect of induced magnetic field (cf. [35]).
 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57 45

Fig. 1. Schematic diagram of the model geometry with arterial stenosis.

∂u ∂u ∂u ∂p 1 ∂ 2u 1 ∂u ∂ 2u
( ) ( )
1.3. Boundary conditions
+ u +v =− + + +
∂t ∂x ∂r ∂x Re ∂ r 2 r ∂r ∂ x2
The boundary conditions for blood flow through a stenosed
Rs Ha2
artery are taken as + (vpx − u) − u, (19)
Re Re
∂ u∗ ∂T ∗
= v ∗
= =0 along the central axis of the artery,
∂r∗ ∂r∗ ∂v ∂v ∂v ∂p 1 ∂ 2v 1 ∂v ∂ 2v v
( ) ( )
+ u +v =− + + + −
i.e., at r = 0,
∗
(12) ∂t ∂x ∂r ∂r Re ∂ r 2 r ∂r ∂ x2 r2
Rs
u∗ = v ∗ = 0 , T ∗ = Tw at the arterial wall, i.e., at r ∗ = R∗ (x∗ ), + (vpy − v ), (20)
Re
(13)
∂θ ∂θ ∂θ ∂ θ 1 ∂θ ∂ 2θ
( ) ( 2 )
1
+ u +v = + + 2
[ )2 ]
r∗
(
∂t ∂x ∂r RePr ∂ r 2 r ∂r ∂x
u∗ = 2U0 1 − , v ∗ = 0,
R∗ (x∗ ) EcHa2 Ec
+ u2 + φ, (21)
∗
T = Tw + 2(1 − r )(T∞ − Tw ) ∗2
at the inlet i.e, at x = 0,
∗ Re Re
(14) where,
(( )2 )2 ) )2
and ∂v ∂u ∂v ∂u
( (
∂ u∗ ∂v ∗ ∂T ∗ φ=2 + + + . (22)
= = ∗ =0 at the outlet i.e, at x∗ = R0 L. (15) ∂r ∂x ∂x ∂r
∂ x∗ ∂x ∗ ∂x
The non-dimensional parameters appearing in Eqs. (19)–(21) are
In the above, R∗ (x∗ ) represents the geometry of the stenosis U R
defined as the Reynolds number Re = 0ν 0 , Hartmann number
which can be put mathematically as
µCp
√
σ
[ ( ∗ )2 ]
x
Ha = B0 R0 ρν
, Prandtl number Pr = k
, Eckert number Ec =
−α
R (x ) = R0 1 − δ e
∗ ∗ R0
, (16) U02 Ks NR20
and particle concentration parameter Rs =
Cp (T∞ −Tw ) µ
.
where R0 is the radius of the normal portion of the artery, δ is Moreover, Eqs. (10) and (11) become,
the maximum height of the stenosis and α represents the shape ∂vpx
parameter of the stenosis, Tw is the temperature of the arterial G = (u − vpx ), (23)
∂t
wall, T∞ is the ambient temperature, U0 is the characteristic
velocity and the length of the artery is taken as L times the radius and
of the artery. ∂vpy
G = (v − vpy ), (24)
The following non-dimensional variables are introduced to ∂t
make all the flow variables dimensionless: mU
where G = K R0 is the particle mass parameter.
x∗ r∗ u∗ v∗ p∗ t ∗ U0 s 0
x= ,r = ,u = ,v = ,p = ,t = , The non-dimensional stream function ψ is defined by
R0 R0 U0 U0 ρ U02
R0
(17) 1 ∂ψ 1 ∂ψ
vpx
∗
vpy
∗
T ∗ − Tw u= , v=− (25)
vpx = , vpy = ,θ = . r ∂r r ∂x
U0 U0 T∞ − Tw that satisfies the continuity equation (18) and the non-
Using these dimensionless variables, Eqs. (1)–(4) become, dimensional vorticity ω defined by
∂u v ∂v ∂v ∂u
+ + = 0, (18) ω= − . (26)
∂x r ∂r ∂x ∂r
46 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57
( )2 )
η2 ∂R ∂ 2θ 2η ∂ R ∂ 2 θ
( ]
Now, from Eqs. (19) and (20), we obtain the vorticity-stream 1
function equations by eliminating the pressure terms as + + 2 −
R2 R ∂x ∂η2 R ∂ x ∂η∂ x
∂ω 1 ∂ψ ∂ω 1 ∂ψ ∂ω ω ∂ψ
[ ]
2
Ha2 Ec 2 1 ∂v
[ ( ( )
+ − + 2 Ec
∂t r ∂r ∂x r ∂x ∂r r ∂x + u + 2
Re Re R2 ∂η
1 ∂ 2ω 1 ∂ω ∂ 2ω ω
( )
)2 ) (( )2 ]
∂u η ∂R ∂u ∂v η ∂ R ∂v 1 ∂u
( )
= + + 2 − 2
Re ∂ r 2 r ∂r ∂x r + − + − + .
∂x R ∂ x ∂η ∂x R ∂ x ∂η R ∂η
Rs ∂vpy ∂vpx Ha2 ∂ 2 ψ 1 ∂ψ
( ) ( )
+ − −ω + − , (27) (33)
Re ∂x ∂r Rer ∂r2 r ∂r
and Similarly, the boundary conditions (12)–(15) are transformed
to
∂ ψ
2
1 ∂ψ ∂ ψ 2
− + = −ωr . (28) ∂θ
∂ x2 r ∂r ∂r2 ψ =ω= =0 at η = 0, (34)
∂η
In the next section, we proceed to present solution techniques of ( ( )2 ) )
∂ 2ψ ∂R ∂ 2ψ
(
Eqs. (27), (28) and (21) subjected to boundary conditions (12)– 1 1 1
ψ= , ω=− + 2 1 + η2 ,
(15). 2 ηR ∂x 2 R ∂x ∂η2

2. Coordinate transformation θ =0 at η = 1, (35)

η2 4η
( )
The physical geometry is transformed into a rectangular do-
ψ = η2 R2 1 − , ω= , θ = 2(1 − η2 ) at x = 0,
main in order to apply the computational scheme, we use a 2 R
suitable coordinate transformation given by
(36)
r
η(x, r) = , x = x, (29) and
R(x)
∂ψ ∂ω ∂θ
where = = =0 at x = L. (37)
∂x ∂x ∂x
R∗ (x∗ ) 2
R(x) = = 1 − δ e−αx , (30) Using the boundary conditions (34)–(37), Eqs. (31)–(33) are
R0 solved fully numerically that follow in the next section.
is the non-dimensional form of the geometry of the stenosis
shown in Fig. 1. 3. Computational scheme
Use of Eq. (29) into Eqs. (27) and (28), we obtain,
The transformed governing equations as presented in the
∂ω ω ∂ψ ω ∂ R ∂ψ 1 ∂ψ ∂ω 1 ∂ψ ∂ω
[ ]
+ 2 2 − 3 + 2 − 2 previous section are solved by using finite difference method.
∂t η R ∂x ηR ∂ x ∂η ηR ∂η ∂ x ηR ∂ x ∂η Meshes are produced by subdividing the rectangular domain
1 ∂ 2ω
[
obtained by drawing parallel straight lines to the coordinate axes.
= The solutions of ω(x, η, t), ψ (x, η, t) and θ (x, η, t) at any mesh
Re ∂ x2
( ( )
2
) point (i, j, k) in a computational domain are symbolized by ωik,j ,
ω 3η ∂ R η ∂ 2R 1 ∂ω ψik,j and θik,j respectively. Let xi = i∆x, (i = 0, 1, . . . , m); ηj =
− 2 2 + − + 2
η R R2 ∂ x R ∂ x2 ηR ∂η j∆η, (j = 0, 1, . . . , n); tk = k∆t, (k = 0, 1, . . .,) represent
( ( )
2
) the position of any mesh points of which ∆x, ∆η and ∆t are
η2 ∂ R 1 ∂ 2ω the spacing lengths in x, η and t directions respectively. The
+ +
R2 ∂ x R2 ∂η2 maximum number of mesh points in x and η directions are m
and n; k denotes the number of time steps.
2η ∂ R ∂ ω
2
Rs ∂vpy ∂R ∂vpx
] [ ( ) ]
1
− + − η +1 −ω The system of Eqs. (31)–(33), along with the boundary condi-
R ∂ x ∂η∂ x Re ∂ x R ∂x ∂η tions (34)–(37) are solved numerically by developing Peaceman–
Ha2 ∂ ψ ∂ψ
( 2 )
Rachford ADI (Alternating Direction Implicit) method and as Eq.
+ 2 η 2 − , (31) (32) is a linear equation in ψ , it has been solved by Point Suc-
η ReR3 ∂η ∂η
cessive Over Relaxation (SOR) method, thereafter discretizing it
( ( )2 ) explicitly, we obtain
∂ 2ψ 1 ∂R 2 ∂ R
2
∂ψ
+ 2 3η 2
−η R 2 −1 (
ψik+1,j + ψik−1,j
∂x 2 ηR ∂x ∂x ∂η ψ
k+1
= −ωik,j ηj Rj −
i,j
( ( )
2
) ∆ x2
η 2
∂R 1 ∂ ψ
2
2ηj ∂ R ψi+1,j+1 − ψi−1,j+1 − ψi+1,j−1 + ψi−1,j−1
( )
k k k k
+ + 2
R 2 ∂x R ∂η2 +
Ri ∂ x 4∆x∆η
2η ∂ R ∂ 2 ψ
= −ωηR, ψik,j+1 − ψik,j−1
( ( )2 )( )
− (32) 1 ∂R ∂ 2R
R ∂ x ∂η∂ x − 3ηj2
− ηj Ri 2 − 1
2
ηj R2i ∂x ∂x 2∆η
and the thermal energy equation given in Eq. (21) is transformed (
to ( )2 )
1 ∂R
∂θ
[ (
∂θ η ∂ R ∂θ
)
v ∂θ
] − 2 1 + ηj2
+ u − + Ri ∂x
∂t ∂x R ∂ x ∂η R ∂η ( ( ) )⎞
)) ⎛ 2 ∂R 2
ψi,j+1 + ψi,j−1 η
(
( ( ) ) k k 2 1 + ∂x
⎝− 2 −
2
∂ θ 3η ∂ R η ∂ 2R ∂θ j
[ 2
1 1 ⎠ . (38)
= + − + 2 ∆η 2 ∆x2 R2i ∆η2
RePr ∂ x2 R2 ∂ x R ∂ x2 ηR ∂η
 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57 47

A relaxation parameter β has been introduced for faster con- where

vergence of the numerical schemes and modified the solution in ∆t ∆t
the stream function ψ as A1ki,j = − uki,j − ,
4∆x 2RePr ∆x2
k+1
ψ k+1
i,j = βψ i,j + (1 − β )ψ . k
i,j (39) ∆t ∆t ∆t
B1ki,j = 1 + , C 1ki,j = uki,j − ,
We have observed that the value of β = 1.5 is a good choice RePr ∆x2 4∆x 2RePr ∆x2
for successive over relaxation scheme to converge the numerical [ k ∂R
∆t ui,j ηj ∂ x − vi,j
k
scheme faster.
D1ki,j = θik,j +
Now Eqs. (31) and (33) are discretized by finite difference 2 Ri
scheme using Alternating Direction Implicit (ADI) method which ( ∂ R )2
3ηj θi,j+1 − θik,j−1
] k
consists of two time steps. The first half time step (kth step to ∂x ηj ∂ 2 R 1
+ − +
k + 1/2 time step) uses central difference formula for developing RePrR2i RePrRi ∂ x2 RePr ηj R2i 2∆η
the scheme sweep in x-direction, which leads to the following
θ θ θik,j−1
( ( ) )
2 k k
tridiagonal system of Eq. (31) for ω, ∆t ∂R i,j+1 − 2 i,j +
+ η 2
+ 1
k+1/2 k+1/2 k+1/2 2RePrR2i
j
∂x ∆η2
Aki,j ωi−1,j + Bki,j ωi,j + Cik,j ωi+1,j = Dki,j , (40)
where ∆tHa2 Ec ∆t ηj ∂R
+ u2i,j −
2Re 4RePrRi ∆η∆x ∂ x
ψik,j+1 − ψik,j−1
( )
∆t ∆t
Aki,j = − − , (θik+1,j+1 − θik−1,j+1 − θik+1,j−1 + θik−1,j−1 )
2Re∆x2 4∆xηj R2i 2∆η )2
vik,j+1 − vik,j−1
[ ((
∆tEc uki+1,j − uki−1,j
(
+ 2 +
∆t ∂ R ψ −ψ
( )
k k
∆t i,j+1 i,j−1 2Re 2∆η 2∆x
Bki,j = 1 + −
Re∆x 2
2ηj Ri ∂ x
3
2∆η
ηj ∂ R uki,j+1
− uki,j−1
)2 )
−
ψi+1,j − ψi−1,j
( )
∆t
k k
∆t ∆tRs Ri ∂ x 2∆η
+ 2 2 + + ,
2∆x vi+1,j − vik−1,j ηj ∂ R vi,j+1 − vi,j−1
(( k )
2ηj Ri 2Reηj2 R2i 2Re k k
+ −
2∆x Ri ∂ x 2∆η
ψik,j+1 − ψik,j−1
( )
∆t ∆t
k
Ci,j = − + , k k
1 ui,j+1 − ui,j−1
2
) ]
2Re∆x2 4∆xηj R2i 2∆η + .
Ri 2∆η
( ∂ R )2
ψik+1,j − ψik−1,j
[( )
∆t 3ηj ∂x
Furthermore, Eqs. (23) and (24) can be discretized as
Dki,j =ω + k
i,j +
4∆η 2∆xη 2
ReR2i ∆tui,j ∆t
( ) ( )
j Ri k+1/2
vpx i,j = vpx ki,j + / 1+ , (44)
ηj ∂ 2 R G G
]
1
− + (ωik,j+1 − ωik,j−1 ) (
∆t vi,j
) (
∆t
)
ReRi ∂ x2 Reηj R2i vpy i,j
k+1/2
= vpy ki,j + / 1+ . (45)
∆t ηj ∂R k G G
− (ω − ωik−1,j+1 − ωik+1,j−1
4ReRi ∆η∆x ∂ x i+1,j+1 The central difference scheme for sweep in η direction is per-
( ( )
2
) formed in the second half time step (time step k + 1/2 to k + 1)
∆t ∂R
+ ωi−1,j−1 ) +
k
ηj
2
+1 and the discretized form of Eq. (31) in next half time step from
2ReR2i ∆η2 ∂x k + 1/2 to k + 1 is constituted in a tri-diagonal system that takes
in the form
∆tRs vpy i+1,j − vpy i−1,j
[ k k
(ωik,j+1 − 2ωik,j + ωik,j−1 ) + k+1/2
Pi,j ωik,+j−11 + Qi,j
k+1/2
ωik,+j 1 + Ei,j
k+1/2
ωik,+j+11 = Si,j
k+1/2
, (46)
4Re ∆x
v v
) k k
∂R − where
( ]
1 px i,j+1 px i,j−1
− η +1 k+1/2 k+1/2
∂x ∆η
( ∂ R )2
ψi+1,j − ψi−1,j
[( )
R k+1/2 ∆t 3ηj ∂x
Pi,j = +
ψi,j+1 − 2ψik,j + ψik,j−1
[ ( k )
Ha2 ∆t 4∆η 2∆xηj R2i ReR2i
+ 2 3 ηj
2ηj Ri Re ∆η2 ηj ∂ 2 R 1
]
− +
ψik,j+1 − ψik,j−1 ReRi ∂ x2 Reηj R2
( )]
− . ( (i ) )
2∆η ∆t ∂R
2
− ηj2
+1 ,
After having determined ω and ψ numerically, u and v are 2ReR2i ∆η2 ∂x
obtained using Eq. (25) as
( ( ) )
2
k+1/2 ∆t ∂ R
Qi,j = 1+ ηj2 +1
1 ψi,j+1 − ψi,j−1
k k
ReR2i ∆η2 ∂x
uki,j = , (41)
ηj R2i 2∆η ( k+1/2 k+1/2
∆t ∂ R ψi,j+1 − ψi,j−1
)
and −
2ηj R3i ∂ x 2∆η
1 ∂ R ψi,j+1 − ψi,j−1 1 ψi+1,j − ψi−1,j
k k k k
( k+1/2 k+1/2
vik,j = . ψi+1,j − ψi−1,j
)
− (42) ∆t ∆t ∆tRs
R2i ∂ x 2∆η ηj R i 2∆x + 2 2 + + ,
2ηj Ri 2∆x 2Reηj2 R2i 2Re
The tri-diagonal system corresponding to Eq. (33) in the first half
k+1/2 k+1/2 ( )2
ψi+1,j − ψi−1,j 3ηj ∂∂Rx
( )
time step yields ∆t
[
k+1/2
k+1/2 k+1/2 k+1/2 Ei,j =− +
A1ki,j θi−1,j + B1ki,j θi,j + C 1ki,j θi+1,j = D1ki,j , (43) 4∆η 2∆xηj R2i ReR2i
48 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57

Fig. 2. Variation of axial velocity u at the downstream of the stenosis for Fig. 3. Variation of axial velocity u at the downstream of the stenosis for
different values of the Hartmann number Ha, when Re = 300, Pr = 21, t = 2, different values of the concentration parameter Rs , when Ha = 1, Re = 300,
Rs = O(102 ). Pr = 21, t = 2.

ηj ∂ 2 R
]
1
− +
ReRi ∂ x2 Reηj R2
( (i ) )
2
∆t ∂R
− ηj2
+1 ,
2ReR2i ∆η2 ∂x
k+1/2 k+1/2 ∆t k+1/2 k+1/2 k+1/2
Si,j = ωi,j − (ψi,j+1 − ψi,j−1 )(ωi+1,j
8ηj R2i ∆η∆x
k+1/2 ∆t
− ωi−1,j ) +
2Re∆x2
k+1/2 k+1/2 k+1/2 ∆t ηj ∂R k+1/2
(ωi+1,j − 2ωi,j + ωi−1,j ) − (ωi+1,j+1
4ReRi ∆η∆x ∂ x
k+1/2 k+1/2 k+1/2
− ωi−1,j+1 − ωi+1,j−1 + ωi−1,j−1 )
[ k+1/2 k+1/2
∆tRs vpy i+1,j − vpy i−1,j
+
4Re ∆x
) k+1/2 k+1/2 ]
∂R vpx i,j+1 − vpx i,j−1
(
1
− η +1
R ∂x ∆η
( k+1/2 k+1/2 k+1/2
ψi,j+1 − 2ψi,j + ψi,j−1
)
Ha2 ∆t
[
Fig. 4. Variation of velocity (Vpx ) of nanoparticles at the downstream of the
+ 2 3 ηj stenosis for different values of the Hartmann number Ha, when Re = 300,
2ηj Ri Re ∆η 2 Pr = 21, t = 2, Rs = O(102 ).
( k+1/2 k+1/2 ]
ψi,j+1 − ψi,j−1
)
− ,
2∆η
( )2 )
∆t ∂R
(
and the discretized Eq. of (33) in η direction from time step k+1/2
Q 1i,j =1+ η 2
+1 ,
k + 1/2 to k + 1 gives rise to RePrR2i ∆η2
j
∂x
k+1/2 k+1 k+1/2 k+1 k+1/2 k+1 k+1/2
P1i,j θi,j−1 + Q 1i,j θ
i,j + E1i,j θ
i,j = S1i,j , (47) [ k+1/2
∆t vi,j
k+1/2
− ui,j ηj ∂∂Rx
( )2
3ηj ∂∂Rx
k+1/2
E1i,j = −
where 4∆η Ri RePrR2i
[ k+1/2 k+1/2 ηj ∂ 2 R ∆t
]
− ui,j ηj ∂∂Rx
( )2
k+1/2 ∆t vi,j 3ηj ∂∂Rx + −
1
−
P1i,j =− − RePrRi ∂ x 2
RePr ηj Ri
2
2RePrR2i ∆η2
4∆η Ri RePrR2i ( ( ) )
2
ηj ∂ R
2
∆t ∂R
]
1
+ − − ηj2 +1 ,
RePrRi ∂ x 2
RePr ηj Ri
2
2RePrR2i ∆η2 ∂x
( ( ) )
2
∂R k+1/2
θi+1,j − θi−1,j
k+1/2
ηj2 +1 , k+1/2 k+1/2 ∆t k+1/2
∂x S1i,j =θ i,j − ui,j
2 2∆x
 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57 49

Fig. 5. Variation of velocity (Vpx ) of nanoparticles at the downstream of the Fig. 7. Variation of non-dimensional temperature θ at the downstream of the
stenosis for different values of the concentration parameter Rs , when Ha = 1, stenosis for different values of concentration parameter Rs , when Ha = 1,
Re = 300, Pr = 21, t = 2. Re = 300, Pr = 21, t = 2.

k+1/2 k+1/2 )2 ]
1 ui,j+1 − ui,j−1
+ ,
Ri 2∆η
Similarly, Eqs. (23) and (24) can be discretized as
∆t ∆tui,j
( ) ( )
k+1/2
v k+1
px i,j = v +
px i,j / 1+ , (48)
G G
∆t vi,j ∆t
( ) ( )
k+1/2
vpy ki,+j 1 = vpy i,j + / 1+ . (49)
G G
Accordingly the expressions for ui,j and vi,j are used as presented
in the previous time step. The tri-diagonal systems are solved by
using the well known Thomas algorithm.
After determining u, v and θ numerically, we can obtain the
skin-friction (dimensionless wall shear stress) Cf i , the Nusselt
number Nui and volumetric flow rate Qi at any cross-section of
the artery can be obtained from the following difference schemes:
uki,n − uki,n−1
( )
1
Cf i =
ReRi∆η
τw ∂ u∗
where Cf = with τw = µ ∗ . (50)
Fig. 6. Variation of non-dimensional temperature θ at the downstream of the ρ U02
∂r
stenosis for different values of Hartmann number Ha, when Re = 300, Pr = 21,
θ θ
( k k )
t = 2, Rs = O(102 ). 1 i,n−1 − i,n
Nui =
Ri ∆η
k+1/2 k+1/2 k+1/2
qw R0 ∂T ∗
∆t θi+1,j − 2θi,j + θi−1,j ∆tHa2 Ec 2 where Nu = with qw = −k . (51)
+ + ui,j k(T∞ − Tw ) ∂r∗
2RePr ∆ x2 2Re ∫ 1
∆t ηj ∂ R k+1/2 k+1/2 k+1/2 Qi = 2π (Rki )2 uki,j ηj dη (52)
− (θ − θi−1,j+1 − θi+1,j−1
4RePrRi ∆η∆x ∂ x i+1,j+1 0

[ (( k+1/2 k+1/2 2
vi,j+1 − vi,j−1
)
∆tEc 4. Computational results and discussion
k+1/2
+ θi−1,j−1 ) + 2
2Re 2∆η
A numerical model for diseased arterial segment has been
( k+1/2 k+1/2 k+1/2 k+1/2 )2 ) developed in this study to observe the effects of cholesterol
ui+1,j − ui−1,j ηj ∂ R ui,j+1 − ui,j−1
+ − absorbing drugs infused with magnetic nanoparticles. These mag-
2∆x Ri ∂ x 2∆η netic nanoparticles are targeted to the specific parts by using
(( k+1/2 k+1/2 k+1/2 k+1/2
vi+1,j − vi−1,j ηj ∂ R i,j+1 − vi,j−1
v
)
external magnetic field which work solely on the diseased section
+ − of the artery without affecting the healthy tissues. Using the
2∆x Ri ∂ x 2∆η
standard second order Peaceman–Rachford ADI method which is
a two time-step algorithm, along with implicit finite difference
50 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57

Fig. 8. Streamline (ψ ) contours for different values of Hartmann number, (a) Ha = 0, (b) Ha = 2, (c) Ha = 4, (d) Ha = 6, when Re = 300, Pr = 21, t = 2, Rs = O(102 ).
 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57 51

Fig. 9. Streamline (ψ ) contours for different values of Reynolds number, (a) Re = 100, (b) Re = 300, (c) Re = 500, when Ha = 1, Pr = 21, t = 2, Rs = O(102 ).

scheme, we have computed the stream function and vorticity in The numerical computation has been carried out by using the
a long circular tube by considering the length to be 22 times the following data within a range of haemodynamical flows [48–54]:
radius of the artery such that flow becomes fully developed at Re = 100, 300, 500; Ec = 0.002; Pr = 21; β = 1.5; σ = 0.8;
the downstream. The resulting approximations are second order δ = 0.3; G = 0.02; Rs = O(10), O(102 ), O(103 ); Ha = 0 to 1.
in both time and space. The method used in this study is un- Moreover, we have also examined our study for wide range of Ha
conditionally stable due to the consideration of implicit scheme. up to 10 to validate our model.
We used uniform grids with different mesh sizes along the space Figs. 2 and 3 correspond to the variation of axial velocity along
and time directions. Moreover, we considered ∆t = 0.001, ∆x = with the radial direction of the artery at the downstream of the
0.025 and ∆η = 0.0125 for the entire simulation process, since stenosis. Fig. 2 gives the variation of axial velocity at the immedi-
the stability condition on the time-step is ∆t ∼
= O(∆x2 ). It is also ate downstream of the stenosis with different values of Hartmann
checked that further decrease in ∆t does not show any significant number Ha. As the magnetic field interacts with blood flow, a
change in the numerical results for all flow quantities. resistive force called Lorentz force arises which has a tendency to
52 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57

Fig. 10. Streamline (ψ ) contours for different values of concentration parameter, (a) Rs = 10, (b) Rs = 100, (c) Rs = 1000, when Ha = 1, Pr = 21, t = 2.

reduce the blood flow. It is observed that as Hartmann number to Brownian and thermophoretic forces and makes the velocity
increases, the velocity of blood tends to decrease and flattens field to be stronger [55].
in the central region, whereas the blood velocity accelerates in Figs. 4 and 5 are the axial component of velocity profiles
the vicinity of the arterial wall. More interestingly a reverse flow of the magnetic nanoparticles along radial direction. Both of
takes place near the arterial wall, which tends to vanish under the these profiles show that axial velocity is maximum at the axis.
influence of certain strength of magnetic field. Fig. 3 represents It is observed from Fig. 4 that the velocity of nanoparticles
the variation of axial velocity with varying coefficient of nanopar- decreases with a rise in the magnetic field strength. This obser-
ticle concentration Rs . It is observed that as the concentration vation leads to the capturing efficiency of nanoparticles bound
of nanoparticles increases, velocity also increases. This is due to with drug at a specific location. Fig. 5 shows that the velocity
the fact that, nanoparticles are not uniformly distributed, rather, of magnetic nanoparticles decreases near the arterial wall with
they are concentrated more in the vicinity of the centerline of the increasing concentration parameter Rs. Thus more augmenta-
artery than the wall. This causes lack of uniform distribution due tion of nanoparticles into the blood stream has more capturing
 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57 53

Fig. 11. Isotherms (temperature contour) for different values of concentration parameter, (a) Rs = 10, (b) Rs = 100, (c) Rs = 1000, when Ha = 1, Re = 300, Pr = 21,
t = 2.

efficiency. Moreover, this defines the flow behavior of the nanopar- concentration parameter Rs . At the vicinity of the wall, the tem-
ticles which coincides with the blood flow behavior in macro perature tends to increase with increasing concentration of the
level. nanoparticles, but as it goes deeper into central line, this phe-
Figs. 6 and 7 give the variation of temperature profile, along nomenon tends to reverse. This is due to specific loss power (SLP)
the radial direction. Fig. 6 illustrates the variation of temperature generation. However, well dispersed nanoparticles show highest
profile for different values of Hartmann number Ha. It is observed SLP. This is due to increase in magnetic interaction between the
that the temperature profiles tend to increase as the magnetic nanoparticles results in certain conditions which decreases the
field strength increases. The results presented in this figure has hyperthermia efficiency.
important implication in hyperthermic treatment of tumor. The Fig. 8(a)–(d) exhibit blood flow pattern via streamline con-
application of an external magnetic field to blood flow causes tours for different Hartmann number Ha. From these figures we
enhancement of temperature to the surrounding tissues. Fig. 7 observe the formation of vortices which detaches from the wall
represents the variation of temperature profile with nanoparticle and rolls up into a ring like vortex structure at the downstream of
54 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57

the stenosis which get reduced and finally vanishes as magnetic

field strength is sufficiently increased. This confirms the idea of
flow circulation zone and the position of flow reversal that takes
place at the downstream of the stenosis. Fig. 9(a)–(c) represent
the streamline contours for different Reynolds number Re. The
streamlines near the downstream of the stenosis rolls up and
forming a ring like vortex structure which interacts and collide
with the arterial wall. When Reynolds number increases, we
observe more vigorous vortex dynamics, resulting in increasing
the intensity of the vortices which approaches to the arterial
wall at higher speed and produces further ring like vortex struc-
ture. The streamlines are observed parallel in the central region
due to axisymmetricity of the artery. In real physiological sys-
tem, flow of blood is considered to be laminar when Reynolds
number is in between 100 to 500 for large arteries such as in
aorta. Fig. 10(a)–(c) depicts the streamline contours with varying
nanoparticle concentration parameter Rs . It is observed that as
the aggregation of nanoparticles in the blood stream increases,
the blood velocity increases, enhancing the intensity of the vor-
tices at the downstream of the stenosis. This may prolong the
residing time of the magnetic nanoparticles near the stenosis
and increases capturing efficiency to treat clinically the arterial Fig. 12. Trajectories of magnetic nanoparticles in a blood vessel at various
disease. Fig. 11(a)–(c) describes the isothermal lines with varying positions in the radial direction η, when Ha = 1, Re = 300, Pr = 21, Rs = O(102 ).
nanoparticle concentration parameter Rs . It is observed that as
the nanoparticle concentration increases in the blood, the dis-
turbance in the temperature contour near the wall enhances.
As we discussed earlier, the enhancement in SLP due to the
agglomeration of nanoparticles causes the disturbances in the
isotherm contours.
The trajectories of the magnetic nanoparticles infused in the
blood stream at various radial positions are shown in Fig. 12.
The trajectories are presented for a range of initial positions (−2,
0.0), (−2, 0.2), (−2, 0.4), (−2, 0.6), (−2, 0.8) and (−2, 1.0). It
is observed that all particles are captured in the lower arterial
wall due to the presence of a bell-shaped stenosis starting from
the position (−2, 1) in the upper wall. The nanoparticles near
the vicinity of the arterial wall close to the onset of the stenosis
are observed to have early capture than the nanoparticles in the
central region of the artery.
Fig. 13 shows the variation of dimensionless wall shear stress
(skin-friction) along the length of the arterial segment with vary-
ing concentration of nanoparticles in the blood stream. It is ob-
served that at the throat, wall shear stress increases and grows to
a peak and falls sharply at the downstream of the stenosis. At the
throat of the stenosis, wall shear stress increases with increasing
concentration of nanoparticles but the opposite trend is followed Fig. 13. Distribution of wall shear stress for different values of concentration
parameter Rs , when Ha = 1, Re = 300, Pr = 21, t = 2.
from the immediate downstream of the stenosis. In this figure, we
have compared the shear stress with infused magnetic nanopar-
ticles and without mixing nanoparticles in the blood stream. The
shear stresses in both cases have similar profile which shows that of concentration of the nanoparticles does not impose extra force
infusion of magnetic nanoparticles in the arteries does not cause on the arterial wall to damage it.
any abnormalities in the flow behavior. Moreover, the region Fig. 16 illustrates the variation of Nusselt number along the
of negative shear stress shows the radius of flow circulation axial direction with varying concentration of nanoparticles. It is
eddies. This low shear stress regions have pathological signifi- observed that the rate of wall heat transfer increases to its peak
cance and increasing concentration prolong the residing time of at the throat of the stenosis and gradually falls in the immediate
blood constituents that eventually flows alongside the arterial downstream of the stenosis. Furthermore, increasing concentra-
wall, which may cure arterial diseases. Figs. 14 and 15 depict the tion of nanoparticles in the blood increases the Nusselt number
variation of wall shear stress in function of Hartmann number Ha Nu significantly, precisely in the downstream of the stenosis. This
with different Reynolds number Re and concentration parameter increase in wall heat transfer due to enhanced agglomeration of
Rs respectively. This can be perceived that shear stress has an particles can transfer heat to the surrounding tissues enabling the
enhancing nature with increasing Hartmann number and concen- cure of tumor/ cancerous cells. Fig. 17 gives the Nusselt number
tration parameter whereas it has reducing effect with increasing variation with Hartmann number Ha for different concentration
Reynolds number at the throat of the stenosis. Moreover, Fig. 15 parameter. It is observed that as the concentration of particles
reveals that increasing magnetic field strength does not bear any increases, less magnetic field strength is required to enhance the
significant effect on WSS when nanoparticle concentration is very Nusselt number. The rate of heat transfer at the wall increases
high in the blood stream. This confirms us the maximum range with both concentration parameter and higher magnetic strength.
 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57 55

Fig. 14. Distribution of wall shear stress at the wall in function of Hartmann
number Ha for different values of Reynolds number Re, when Pr = 21, t = 2. Fig. 16. Variation of Nusselt number Nu for different values of concentration
parameter Rs , when Ha = 1, Re = 300, Pr = 21, t = 2.

Fig. 15. Distribution of wall shear stress at the wall in function of Hartmann
number Ha for different values of concentration parameter Rs , when Re = 300,
Pr = 21, t = 2. Fig. 17. Variation of Nusselt number Nu along with the Hartmann number Ha
for different values of concentration parameter Rs , when Pr = 21, t = 2.

Thus transferring more heat to the surrounding tissues with the

injection of magnetic nanoparticles to the blood stream is one of 5. Conclusion
the pioneering results of the present study.
Fig. 18 gives the variation of volumetric flow rate along the The primary focus of the study is drug targeting to a specific
length of the artery for different Hartmann number Ha. The figure diseased location in the artery with the help of external magnetic
depicts that the volumetric flow rate falls sharply before the field. Modeling and simulation of blood flow and heat transfer
throat of the stenosis and grows to a peak value at the immediate have been carried out with magnetic nanoparticles as carrier
downstream of the stenosis. One can observe that the flow rate for targeted drug delivery in the stenosed artery. The govern-
decreases when magnetic strength is increased. This is due to ing equations along with the suitable boundary conditions are
the fact that when magnetic field strength is increased, velocity simulated by adapting vorticity stream-function formulation and
of blood reduces due to the existence of Lorentz force. Fig. 19 thereafter solved fully numerically by implicit finite difference
gives the plot for flow rate with Reynolds number for differ- technique using P–R (Peaceman–Rachford) Alternating Direction
ent concentration parameter of nanoparticles infused in blood Implicit (ADI) method. The study emphasizes on minimizing the
stream. It is observed that the flow rate decreases with increasing LDL cholesterol levels deposited at the endothelial cell of the
aggregation of nanoparticles while it increases along the Reynolds arterial wall by injecting cholesterol absorbing drugs mixed with
number. Moreover, at higher ranges of Reynolds number, flow bio-compatible magnetic nanoparticles. Moreover, determination
rate coincides for moderate and low nanoparticle concentration. of Specific Loss Power (SLP) in the blood due to the presence of
56 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57

drug infused blood constituents to be able to absorb the choles-

terol deposition. Moreover, comparison of WSS of blood infused
with or without nanoparticles, show similar profile and makes
this method safe. However, when nanoparticle concentration in
blood is high, increase in magnetic strength does not have any
significant effect on the WSS that gives us a specific range of
the nanoparticle concentration to be used. Increase in wall heat
transfer due to particle concentration can be useful for treatment
of tumor and cancer cells in the surrounding tissues by allowing
more arterial heat to pass through the arterial wall. Thus this
study convey potential application in the field of drug delivery
system, hyperthermia treatments, biomedical technologies and
bioengineering.

Declaration of competing interest

The authors declare that they have no known competing finan-

cial interests or personal relationships that could have appeared
to influence the work reported in this paper.

Fig. 18. Variation of volumetric flow rate Q for different values of Hartmann Acknowledgments
number Ha, when Re = 300, Pr = 21, t = 2, Rs = 75.
The authors are grateful to the esteemed reviewers for their
comments and suggestions based on which the manuscript has
been improved. Research support from INSPIRE, DST, India fel-
lowship (IF140754), Department of Science and Technology, Gov-
ernment of India is gratefully acknowledged.

References

[1] A. Drochon, M. Beuque, D.A. Rodriguez, A review of some reference analytic

solutions for the magnetohydrodynamic flow of blood, Appl. Math. 9
(2018) 1179–1192.
[2] S.I. Abdelsalam, M.M. Bhatti, A. Zeeshan, A. Riaz, O.A. Bég, Metachronal
propulsion of a magnetised particle-fluid suspension in a ciliated channel
with heat and mass transfer, Phys. Scripta 94 (2019) 115301.
[3] H. Waqas, S.U. Khan, M. Imran, M.M. Bhatti, Thermally developed Falkner–
Skan bioconvection flow of a magnetized nanofluid in the presence of
a motile gyrotactic microorganism: Buongiorno’s nanofluid model, Phys.
Scripta 94 (2019) 115304.
[4] Q.A. Pankhurst, N.K.T. Thanh, S.K. Jones, J. Dobson, Progress in applications
of magnetic nanoparticles in biomedicine, J. Phys. D: Appl. Phys. 42 (2009)
224001.
[5] Q.A. Pankhurst, J. Connolly, S.K. Jones, J. Dobson, Applications of magnetic
nanoparticles in biomedicine, J. Phys. D: Appl. Phys. 36 (2003) R167–R181.
[6] C.C. Berry, Progress in functionalization of magnetic nanoparticles for
Fig. 19. Variation of volumetric flow rate Q in function of Reynolds number Re applications in biomedicine, J. Phys. D: Appl. Phys. 42 (2009) 224003.
for different values of concentration parameter Rs , when Ha = 1, Pr = 21, t = 2. [7] M. Arruebo, R. Fernández-Pacheco, R.M. Ibarra, J. Santamaría, Magnetic
nanoparticles for drug delivery, Nanotoday 2 (2007) 22–32.
[8] A. Nacev, C. Beni, O. Bruno, B. Shapiro, The behaviors of ferromagnetic
nanoparticles in and around blood vessels under applied magnetic fields,
magnetic nanoparticles, leads to the study of bio-heat transfer J. Magn. Magn. Mater 323 (2011) 651–668.
that gives us to attempt to study further. [9] A.D. Grief, G. Richardson, Mathematical modelling of magnetically targeted
drug delivery, J. Magn. Magn. Mater 293 (1) (2005) 455–463.
The study reveals that the axial velocity has increasing ef-
[10] P. Guo, X.X. Li, P. Xiong, J.S. He, Local magnetic nanoparticle delivery in
fect on concentration of nanoparticles that are infused in the microvasculature, Chin. Phys. Lett. 26 (2009) 018703–018706.
blood and its velocity has similar profile with the velocity of [11] A.S. Lübbe, C. Alexiou, C. Bergemann, Clinical applications of magnetic drug
blood. The study has captured that the temperature profile has targeting, J. Surg. Res. 95 (2001) 200–206.
[12] C. Alexiou, A. Schmidt, R. Klein, P. Hulin, C. Bergemann, W. Arnold,
a reducing trend as the concentration of the nanoparticles in-
Magnetic drug targeting: biodistribution and dependency on magnetic field
creases due to increase in SLP which is a significant study for strength, J. Magn. Magn. Mater. 252 (2002) 363–366.
biological applications regarding hyperthermia. In the streamline [13] E.P. Furlani, Magnetic biotransport: Analysis and applications, Materials 3
plots, we can observe ring like vortex structure near the vicinity (2010) 2412–2446.
of the arterial wall at the outset of the stenosis. This vortex [14] H. Chen, A.D. Ebner, M.D. Kaminski, A.J. Rosengart, J.A. Ritter, Analysis
of magnetic drug carrier particle capture by a magnetizable intravascular
structures decreases its intensity as magnetic field strength is stent-2: parametric study with multi-wire two dimensional model, J. Magn.
induced but yields more vigorous structure as Reynolds number Magn. Mater. 293 (1) (2005) 616–632.
and concentration parameter is increased in the laminar flow [15] Q. Cao, X. Han, L. Li, Enhancement of the efficiency of magnetic targeting
range. This large vortex rings followed by internal shear layers for drug delivery: development and evaluation of magnet system, J. Magn.
Magn. Mater 323 (2011) 1919–1924.
facilitate the generation of wall shear stress which decreases with
[16] A. Nacev, A. Komaee, A. Sarwar, R. Probst, S.H. Kim, M.E. Buck, B. Shapiro,
increasing agglomeration of nanoparticles after the downstream Towards control of magnetic fluids in patients, IEEE Control Syst. Mag. 32
of the stenosis. This low WSS regions prolong the residing time of (2012) 32–74.
 S. Majee and G.C. Shit / European Journal of Mechanics / B Fluids 83 (2020) 42–57 57

[17] B. Polyak, G. Friedman, Magnetic targeting for site- specific drug delivery: [37] C. Kumar, F. Mohammad, Magnetic nanoparticles for hyperthermia- based
Applications and clinical potential, Expert Opin. Drug Deliv. 6 (2009) therapy and controlled drug delivery, Adv. Drug Deliv. Rev. 63 (2011)
53–70. 789–808.
[18] C.C. Berry, A.S.G. Curtis, Functionalization of magnetic nanoparticles for [38] R.E. Rosensweig, Heating magnetic fluid with alternating magnetic field, J.
applications in biomedicine, J. Phys. D: Appl. Phys. 36 (2003) R198–R206. Magn. Magn. Mater. 252 (2002) 370–374.
[19] M. Yokoyama, Drug targeting with nano-sized carrier systems, J. Artif. [39] A. Riedinger, P. Guardia, A. Curcio, M.A. Garcia, R. Cingolani, L. Manna,
Organ. 8 (2005) 77–84. T. Pellegrino, Subnanometer local temperature probing and remotely con-
[20] J. Riegler, J.A. Wells, P.G. Kyrtatos, A.N. Price, Q. Pankhurst, M. Lythgoe, Tar- trolled drug release based on azo-functionalized iron oxide nanoparticles,
geted magnetic delivery and tracking of cells using a magnetic resonance Nano Lett. 13 (2013) 2399–2406.
imaging system, Biomaterials 31 (2010) 5366–5371. [40] B. Ziaie, A. Baldi, M. Lei, Y. Gu, R.A. Siegel, Hard and soft micromachining
[21] S. Kenjeres, B.W. Righolt, Simulations of magnetic capturing of drug for BioMEMS: review of techniques and examples of applications in
carriers in the brain vascular system, Int. J. Heat Fluid Flow 35 (2012) microfluidics and drug delivery, Adv. Drug Deliv. Rev. 56 (2004) 145–172.
68–75. [41] B.T. McWilliams, H. Wang, V.J. Binns, S. Curto, S.H. Bossmann, P. Prakash,
[22] J.C. Cohen, E. Boerwinkle, T.H. Mosley, H.H. Hobbs, Sequence variation in Experimental investigation of magnetic nanoparticle- enhanced microwave
PCSK9, low LPL, and protection against coronary heart disease, New Engl. hyperthermia, J. Func. Biomater. 8 (2017) 21.
J. Med. 354 (2006) 1264–1272. [42] C. Hasse, U. Nowak, Role of dipole-dipole interactions for hyperthermia
[23] E.J. Furlani, E.P. Furlani, A model for predicting magnetic targeting of heating of magnetic nanoparticle ensembles, Phys. Rev. B 85 (4) (2012)
multifunctional particles in the microvasculature, J. Magn. Magn. Mater. 045435.
312 (2007) 187–193. [43] B. Mehdaoui, R.P. Tan, A. Meffre, J. Carrey, S. Lachaize, B. Chaudret, M.
[24] S. Shaw, P.V.S.N. Murthy, S.C. Pradhan, Effect of non-Newtonian character- Respaud, Increase of magnetic hyperthermia efficiency due to dipolar
istics of blood on magnetic drug targeting in the impermeable microvessel, interactions in low anisotropy magnetic nanoparticles: Theoretical and
J. Magn. Magn. Mater. 322 (2010) 1037–1043. experimental results, Phys. Rev. B 87 (17) (2013) 174419.
[25] S. Shaw, P.V.S.N. Murthy, Magnetic targeting in the permeable blood [44] G. Béalle, R.D. Corato, J. Kolosnjaj-Tabi, V. Dupuis, O. Clément, F. Gazeau, C.
vessel- The effect of blood rheology, Trans. ASME, J. Nanotechnol. Eng. Wilhelm, C. Ménager, Ultra magnetic liposomes for MR imaging, targeting
Med. 1 (2) (2010) 021001-(1-11). and hyperthermia, Langmuir 28 (2012) 11834–11842.
[26] A. Mondal, G.C. Shit, Transport of magneto-nanoparticles during electro- [45] M.E. Sadat, R. Patel, J. Sookoor, S.L. Bud’ko, R.C. Ewing, J. Zhang, H. Xu,
osmotic flow in a micro-tube in the presence of magnetic field for drug Y. Wang, G.M. Pauletti, D.B. Mast, D. Shi, Effect of spatial confinement on
delivery application, J. Magn. Magn. Mater. 442 (2017) 319–328. magntic hyperthermia via dipolar interactions in Fe3O4 nanoparticles for
[27] S. Shaw, P.V.S.N. Murthy, Magnetic targeting in the impermeable micro- biomedical applications, Mater. Sci. Eng. 42 (2014) 52–63.
vessel with two phase fluid model- non-Newtonian characteristics of [46] M.E. de Sousa, M.B. Fernández van Raap, P.C. Rivas, P. Zélis, P. Girardin,
blood, Microvasc. Res. 80 (2010) 209–220. G.A. Pasquevich, J.L. Alessandrini, D. Muraca, F.H. Sánchez, Stability and
[28] S. Shaw, P.V.S.N. Murthy, P. Sibanda, Magnetic drug targeting in a relaxation mechanisms of citric acid coated magnetite nanoparticles for
permeable microvessel, Microvasc. Res. 85 (2013) 77–85. magnetic hyperthermia, J. Phys. Chem. C 117 (2013) 5436–5445.
[29] M.M. Larimi, A. Ramiar, A.A. Ranjbar, Numerical simulation of magnetic [47] A. Hamid, Hashim, A. Hafeez, M. Khan, A.S. Alshomrani, M. Alghamdi, Heat
nanoparticles targeting in a bifurcation vessel, J. Magn. Magn. Mater. 362 transport features of magnetic water–graphene oxide nanofluid flow with
(2014) 58–71. thermal radiation: Stability test, Eur. J. Mech. B Fluids 76 (2019) 434–441.
[30] J.A. Ritter, A.D. Ebner, K.D. Daniel, K.L. Stewart, Applications of high [48] A. Drochon, Sinusoïdal flow of blood in a cylindrical deformable vessel
gradient magnetic separation principles to magnetic drug targeting, J. exposed to an external magnetic field, Eur. Phys. J. Appl. Phys. 73 (2016)
Magn. Magn. Mater. 280 (2004) 184–201. 31101.
[31] O. Rotariu, N.J.C. Strachan, Modelling magnetic carrier particle targeting in [49] E.P. Furlani, Magnetophoretic separation of blood cells at the microscale,
the tumor microvasculature for cancer treatment, J. Magn. Magn. Mater. J. Phys. D: Appl. Phys. 40 (2007) 1313–1319.
293 (2005) 63. [50] E.P. Furlani, Y. Sahoo, Analytical model for the magnetic field and force
[32] S. Sharma, A. Gaur, U. Singh, V.K. Katiyar, Capture efficiency of magnetic in a magnetophoretic microsystem, J. Phys. D: Appl. Phys. 39 (2006)
nanoparticles in a tube under magnetic field, Proc. Mater. Sci. 10 (2015) 1724–1732.
64–69. [51] D.F. Young, Fluid mechanics of arterial stenoses, J. Biomech. Eng. 101
[33] S. Sharma, U. Singh, V.K. Katiyar, Magnetic field effect on flow parameters (1979) 157–175.
of blood along with magnetic particles in a cylindrical tube, J. Magn. Magn. [52] S.C. Ling, H.B. Atabek, A non-linear analysis of pulsatile flow in arteries, J.
Mater. 377 (2015) 395–401. Fluid Mech. 55 (1972) 493–511.
[34] S. Sharma, V.K. Katiyar, U. Singh, Magnetic modelling for trajectories of [53] J.C. Misra, G.C. Shit, H.J. Rath, Flow and heat transfer of a MHD vis-
magnetic nanoparticles in a blood vessel under magnetic field, J. Magn. coelastic fluid in a channel with stretching walls: Some applications to
Magn. Mater. 379 (2015) 102–107. haemodynamics, Comput. Fluids 37 (2008) 1–11.
[35] A. Nacev, C. Beni, O. Bruno, B. Shapiro, Magnetic nanoparticle transport [54] H.J. Lim, Y.J. Lee, J.H. Nam, S. Ching, S. Shin, Temperature-dependent
within flowing blood and into surrounding tissue, Nanomedicine (Lond) 5 thresshold shear stress of red blood cell aggregation, J. Biomech. 43 (2010)
(9) (2010) 1459–1466. 546–550.
[36] M. Sheikholeslami, S.A. Shehzad, Magnetohydrodynamic nanofluid convec- [55] M. Hangi, M. Bahiraei, A two-phase simulation for ferrofluid flow between
tion in a porous enclosure considering heat flux boundary condition, Int. two parallel plates under localized magnetic field by applying Lagrangian
J. Heat Mass Transfer 106 (2017) 1261–1269. approach for nanoparticles, Eur. J. Mech. B Fluids 47 (2019) 252–259.