A Theoretical Model For Blood Flow in Small Vessels: V. P. Srivastava
A Theoretical Model For Blood Flow in Small Vessels: V. P. Srivastava
A Theoretical Model For Blood Flow in Small Vessels: V. P. Srivastava
http://pvamu.edu/pages/398.asp
ISSN: 1932-9466
Mathematics
(AAM):
Vol. 2, Issue 1 (June 2007) pp. 51 – 65 An International Journal
(Previously Vol. 2, No. 1)
Received April 20, 2006; revised March 20, 2007; accepted March 26, 2007
Abstract
A two-fluid model consisting of a core region of suspension of all the erythrocytes (particles)
in plasma (fluid) assumed to be a particle-fluid mixture and a peripheral layer of cell-free
plasma (Newtonian fluid), has been proposed to represent blood flow in small diameter tubes.
The analytical results obtained in the proposed model for effective viscosity, velocity profiles
and flow rate have been evaluated numerically for various values of the parameters available
from published works. Quantitative comparison has shown that present model suitability
represents blood flow at hematocrit ( 40%) and in vessels up to 70m in diameter. Using
experimental values of the parameters, the flow rate for normal and diseased blood has been
computed and compared with corresponding values obtained from a well known
experimentally tested model in the literature.
Keywords: Erythrocytes, plasma, hematocrit, peripheral layer, velocity profile, flow rate,
effective viscosity
1. Introduction
The study of blood flow through mammalian circulatory system has been the subject of
scientific research for about a couple of centuries. Like most of the problems of nature and
life sciences, it is complex one due to the complicated structure of blood, the circulatory
system and their constituent materials. The experimental studies and the theoretical
treatments of blood flow phenomena are very useful for the diagnosis of a number of
cardiovascular diseases and development of pathological patterns in human or animal
physiology and for other clinical purposes and practical applications.
A large number of theoretical and experimental efforts have been made in the literature to
explain the blood flow behavior when it flows through the vessels of circulatory system of
living beings. To account for the new evidences uncovered through improved experimental
theories of blood flow from the numerous relevant and important contributions of Bayliss
(1952), Womersley (1954,1955,1957,1958), Muller(1951,1959), McDonald (1960),
51
52 V. P. Srivastava
Whitmore (1963), Copley and Stainsby (1960), Attinger (1964), Fung (1964), Lew and Fung
(1970) and many others, mathematical modeling of blood flow has been subject to constant
changes and modifications. Above listed investigators have used single-phase homogeneous
Newtonian viscous fluid, a classical approach that does not account for the presence of red
cells (one of the main constituents of blood, which is responsible for many of the blood
properties and diseases (Srivastava, 1995)) in blood while flowing through the circulatory
system. Although, this approach provides satisfactory tools to describe certain aspects of
blood flow in aorta and large arteries, it fails to give an adequate representation of flow field,
especially in the vessels of small diameter (2400 – 8m, Srivastava and Srivastava (1983)).
Several researchers (Casson, 1959; Haynes, 1959; Charm and Kurland, 1964; Eringen, 1964;
Gupta et al., 1982; Chaturani and Upadhya, 1981) have proposed various representative
models for blood in small vessels and narrow capillaries. A number of investigators including
Haynes and Burton (1959), Merrill et al. (1963), Charm and Kurland (1965), Hershey et al.
(1964), Cokelet (1972) and Lih (1975) have pointed out that blood being a suspension of
corpuscles, behaves like a non-Newtonian fluid at low shear rates. In particular, Hershey et
al. (1966) and Huckaba et al. (1968) have shown that blood flowing through a tube of
diameter less than 0.2 mm and at low shear rate less than 20/s, behaves as a power law- fluid
while Casson (1959), Reiner and Blair (1959), Charm and Kurland (1965,1974) and Merrill
et al. (1964) have suggested that blood inhibits yield stress and behaves as a Casson model
fluid at a shear rate equal to 0.1/s.
blood in small diameter tubes by a two-layered model assuming peripheral and core fluids as
Newtonian fluids of different viscosities. Following the theoretical study of Haynes (1960)
and experimentally tested model of Bugliorello and Sevilla (1970), two-fluid modeling of
blood flow has been discussed and used by a good number of researchers. Shukla et al.
(1980) applied a two-fluid model to discuss the flow of blood through a stenosis. Chaturani
and Upadhya (1979, 1981) addressed the flow of blood in small diameter tubes using the two-
layered model of micropolar and couple stress fluids, respectively. Pralhad and Scultz (1988)
used a two-fluid model of polar fluid to analyze the flow of blood through stenotic arteries.
Two-fluid model analyses have been carried out by Srivastava (2000, 2002) to observe the
effects of a non-symmetrical stenosis on blood flow characteristics. Sharan and Popel
suggested a modification on the models of Haynes(1960) and Bugliarello and Sevilla (1970)
assuming the viscosity in the peripheral layer to be higher than that of plasma due to
additional dissipation of energy caused by the red cells motion near the cell-free layer. Wang
and Bassingthwaighte (2003) applied the two-layered models of Haynes (1960) and Sharan
and Popel (2001) to discuss the flow of blood in narrow curved tubes, etc.
The studies mentioned just above on two-fluid modeling have represented blood either by a
single-phase Newtonian or non-Newtonian fluid in the core region. With increasing interest
in two-phase flows and its applications to blood flow problems, it is however regretted that
no rigorous effort, at least to the author’s knowledge, has been made in the literature to
represent blood as a two-phase system (i.e., erythrocytes and plasma mixture) in the core
region. The purpose of this paper is therefore to investigate the flow of blood in small vessels
involving a two-fluid model. The mathematical model considers a two-layered model of
blood, consisting of a core region of suspension of all the erythrocytes (small spherical non-
flexible particles), assumed to be a particle-fluid suspension (i.e., a suspension of red cells in
plasma) and a peripheral layer of plasma (Newtonian fluid). The study thus presents a
theoretical model for blood, seems to be the only one of its kind which enables one to observe
the simultaneous effects of hematocrit and the peripheral layer on the flow characteristics
while flowing through small vessels.
Consider the axisymmetric flow of blood in an uniform circular tube of radius R. Blood is
represented by a two-fluid model consisting of a core region (central layer) of suspension of
all the erythrocytes assumed to be a particle-fluid mixture (i.e., a suspension of red cells in
plasma) of radius R1 and a peripheral layer of plasma (Newtonian fluid) of thickness (R-R1)
as shown in Fig. 1. Under the simplified assumptions stated in Srivastava and Srivastava
(1983), the equations governing the flow are therefore written (Srivastava, 1995; Drew, 1974)
as
dp
= o (r ) uo , R1 < r < R, (1)
dz r r r
dp (C )
(1– C) = (1 – C) s (r ) uf + C S (up – uf), 0 r R1, (2)
dz r r r
dp
C = C S (uf – up), 0 r R1, (3)
dz
54 V. P. Srivastava
where (r, z) are (radial, axial) coordinates, (uf, up) are the axial velocity of (fluid, particle) in
the core region (0 r R1), (o, uo) are the (viscosity , fluid velocity) in the peripheral region
( R1 r R), s s (C) is the suspension viscosity in the core region, C denotes the
constant volume fraction density of the particles (called hematocrit) and S is the drag
coefficient of interaction between the two phases (fluid and particle). The expression for the
drag coefficient of interaction S and empirical relation for the viscosity of suspension s may
be selected (Charm and Kurland, 1974; Srivastava, 1995) as
S=
9 μo 4 3 (8C 3C 2 1/ 2
) 3C , (4)
2 a o2 (2 3C) 2
o 1107 0 K
s ; q = 0.07 exp[2.49 C + exp(-1.69C)], (5)
1 qC T
where ao is the radius of a particle and T is measured in absolute temperature.
The boundary conditions are the standard no-slip conditions of velocities and shear stresses at
the tube wall and the interface, and are given as
uo = 0 at r = R, (6)
uo = uf and o = f at r = R1, (7)
uf up
= = 0 at r = 0, (8)
r r
with 0 0 u0 / r and f (1 C) s uf / r as the shear stresses of the peripheral
and central layers, respectively.
The expressions for the velocities u0, uf and up obtained as the solutions of equations (1) - (3),
subject to the boundary conditions (6) - (8), are given as
R 2 dp
u0 = –
4 0 dz
1 (r / R )2 , R1 < r < R, (9)
dp R1 r R1 2
2 2
R2
uf = – '
+ (1– C) 1– ,
4 (1 C ) 0 dz R R R
0 < r < R1, (10)
dp R1
2 2
R2 r
up = – '
4 (1 C ) 0 dz R R
R1 2 4 (1 C ) 0
+(1 – C) 1– + , 0 < r < R1 , (11)
R SR 2
with ' 0 / s .
AAM: Intern. J., Vol. 2, Issue 1 (June 2007) [Previously, Vol. 2, No. 1] 55
R R1
Where, Q0 = 2 ruo dr, Qf = 2 (1– C) ruf dr, and
R 0
1
R1
Qp = 2 C rup dr.
0
Using equations (9) – (11) into equation (12), one obtains the expression for flow flux as
R 4 dp
Q=- (1-C) 1 ( R1 / R ) 4 ' (R1 / R ) 4 + 2 (R1 / R )2 , (13)
8 (1 C) 0 dz
with 2 = 8 C (1– C) 0/SR2, a non-dimensional suspension parameter. The use of the fact
that total flux is equal to the sum of the fluxes across the two regions (peripheral and core)
R1 = R. (14)
An application of relation (14) into equation (13), yields the following expression for the
effective (apparent) viscosity as
(1 – C) o
e = . (15)
(1 C) (1 4 ) ' 4 2 2
When R1 = R (i.e., in the absence of the peripheral layer), above results reduces to the case of
a single layered model of a particle-fluid suspension as
(1 C) 0
es = . (16)
' 2
It is worth mentioning that in the absence of the particles (i.e., C = 0), the core mixture
changes to the same fluid as in the peripheral region and thus the role of the peripheral layer
automatically disappears. In addition, when core mixture behaves as a single-phase fluid of
constant viscosity (i.e., s = 1 o), one obtains the same expression for effective viscosity
derived from steady Newtonian fluid model of Bugliarello and Sevilla (1970) as
56 V. P. Srivastava
0
eb = . (17)
1 4 ' 4
Equation (17) recovers the result obtained in Haynes (1960) when o = 1cp.
In order to discuss the results of the theoretical model proposed in the study quantitatively
and to point out its biological relevance, computer codes are developed to evaluate the
analytical results for effective viscosity, velocity profiles and flow rate obtained in equations
(9) - (15) for various values of the parameters involved. For the purpose of comparison, the
corresponding results obtained in the theoretical model of Haynes (1960) considering a two-
phase fluid in the core region and experimentally tested steady flow model of Bugliarello and
Sevilla (1970) using a single- phase fluid (blood) of constant viscosity for a given hematocrit
have been evaluated for the experimental values of the parameters available from the
published literature of (Bugliarello and Sevilla,1970;, Sud and Sekhon,1985; Srivastava et al.,
1994) at the temperature of 25.50 C. The value of is calculated from the relation: = 1-
/R, in which (C) denotes the peripheral layer thickness for a given hematocrit (Haynes,
1960).
Owing to the significance of viscosity, the effective viscosity at 20% and 40% hematocrit
(red cell concentration) has been computed for different size blood vessels. The results
obtained are arranged in Table 1 and compared with the corresponding theoretical values of
Haynes (1960), Chaturani and Upadhya (1979) and experimental values of Bugliarello and
Sevilla (1970). For numerical evaluation of the result for effective viscosity given in equation
(15), the mixture viscosity s has been computed using empirical relation (5) for two values
of the pressure gradient, – dp/dz = 67.5 dyne/mm3and 76.0 dyne /mm3. An inspection of the
Table 1 reveals that effective viscosity computed from the proposed model is in agreement
within the acceptable range to the corresponding values of the effective viscosity obtained in
the theoretical models of Haynes (1960), Bugliarello and Sevilla (1970) and Chaturani and
Upadhya (1979), particularly at low concentration of red cells (20% hematocrit) and in
Vessels of diameter 40m. The effective viscosity deviates from experimental value with
increasing hematocrit and also with the vessel size. However, one notices that present model
exhibits Fahraeus – Lindqvist effect (i.e., apparent viscosity of blood decreases with
decreasing diameter of blood vessel).
The axial velocity profiles (uf, up and uo), computed from the present theory (equations (9) -
(11)), the corresponding model derived (i.e., using erythrocytes-plasma suspension to
represent blood in the core region similar to the present proposed model) from Haynes (1960)
and the steady flow model of Bugliarello and Sevilla (1970) at 20% and 40% hematocrit are
displayed graphically in Figs. 2 and 3 respectively. To evaluate the results obtained for
velocity profiles in Haynes (1960) and Bugliarello and Sevilla (1970), the mixture viscosity
(or blood viscosity) has been taken to be 2.18 cp and 3.10 cp for 20% and 40% hematocrit,
respectively from published literature (Bugliarello and Sevilla, 1970; Sud and Sekhon, 1985;
Srivastava et al., 1994). One observes that erythrocyte velocity at the tube axis assumes
AAM: Intern. J., Vol. 2, Issue 1 (June 2007) [Previously, Vol. 2, No. 1] 57
higher magnitude than the plasma velocity but the difference in their magnitudes decreases
with increasing radial coordinate r towards the interface and at the interface the plasma
velocity (in present and Haynes, 1960 models) coincides with the blood velocity obtained in
Bugliarello and Sevilla (1970).
Table 1.
The volumetric flow rate Q vs pressure gradient –dp/dz computed from the proposed model
(equation (13)) and the experimentally tested model of Bugliarello and Sevilla (1970), at 20%
and 40% hematocrits have been plotted in Fig 4. It may be noted that the magnitudes of the
flow rate Q obtained in the proposed theory are in reasonable agreement with the
corresponding value obtained in Bugliarello and Sevilla (1970), particularly for low pressure
gradients. To emphasize further on the study presented above, flow rate Q vs pressure
gradient –dp/dz for normal and diseased blood (Hb SS, plasma cell dycrasias,
hypertension(controlled), hypertension (uncontrolled) and polycythemia) in a 70m diameter
vessel using the present theoretical approach and the experimental data available from
published literature of shu (1982) and Bugliarello and Sevilla (1970), has been shown in Fig.
5 and compared with the corresponding values obtained in Bugliarello and Sevilla (1970).
The various values of the parameters used for the purpose are arranged in Table 2.
Table 2. Experimental data for diseased and normal blood in a 70 m diameter vessel
(Bugliarello and Sevilla, 1970; Shu, 1982).
Disease Hematocrit (%) 0 (cp) s (cp)
Hb SS (Sickle cell) 24.80 1.30 5.10 0.795
Plasma cell dycrasias 28.00 3.09 5.43 0.816
Normal blood 42.60 1.24 4.03 0.920
Hypertension (uncontrolled) 43.25 1.52 5.15 0.925
It is clearly visible that the flow rates obtained in the present analysis are in good agreement
with those obtained in Bugliarello and Sevilla (1970) for relatively low pressure gradients,
58 V. P. Srivastava
particularly for Hb SS (24.8% hematocrit) and plasma cell dycrasias (28% hematocrit). The
flow rates obtained from the study deviate from those obtained in Bugliarello and Sevilla
(1970), with increasing values of the pressure gradient and also with increasing hematocrit.
4. Conclusions
Acknowledgements:
Author gratefully acknowledges the comments and the suggestions by the Editor and the
Reviewers of the journal. I express my sincere thanks to Prof. Dr. (Mrs.) Mala Tandon,
Northern India Engineering College, Lucknow for her encouragement and help in many ways
during the course of the present work.
REFERENCES
Attinger, E.O. (1964). Pulsatile Blood Flow, McGraw – Hill, New York.
Bayliss, L. E. (1952). Rheology of Blood and Lymph, In: Deformation of flow in Biological
System. A. Fry – Wyssling (Ed.), North Holland Publ. Co. Amsterdam,
Bugliarello, G. and Sevilla, J. (1970). Velocity distribution and other characteristics of steady
and pulsatile blood flow in fine glass tubes, Biorheology 7, pp. 85 – 107.
AAM: Intern. J., Vol. 2, Issue 1 (June 2007) [Previously, Vol. 2, No. 1] 59
Casson, N. A. (1959). Flow equation of pigment oil suspension of printing ink type: In
Rheology of Disperse Systems (Ed. C. C. Mill), pp. 84 – 120, Pergamon Press, London.
Charm, S. E. and Kurland, G. S. (1962). Tube flow behaviour and shear stress shear strain
characteristics of canine blood, American Journal of Physiology 203, pp. 417 – 421.
Charm, S. E. and Kurland, G. S. (1965). Viscometry of human blood for shear rate of 0 – 100,
000 sec-1, Nature 206, pp. 617 – 618.
Charm, S. E. and Kurland, G. S. (1974). Blood Flow and Microcirculation, Wiley, New
York.
Chaturani, P. and Upadhya, V. S. (1979). On miropolar fluid model for blood flow through
narrow tubes, Biorheology 16, pp. 419-428.
Chaturani, P. and Upadhya, V. S. (1981). A two-fluid model for blood flow through small
diameter tubes with non-zero couple stress boundary condition at the interface, Biorheology
18, pp. 245-253.
Cokelet, G. R. (1972). The rheology of human blood: In Biomechanics (Ed. Y.C. Fung et al.),
Prentice Hall, Englewood Cliffs, New Jersey.
Copley, A. L. and Stainsby, G. (1960). Flow Properties of Blood. Pergamon press, Oxford.
Drew, D. A. (1974). Low concentration of two-phase flow near a stagnant point, Physics
Fluids 17, pp.1688-1691.
Fringe, A. C. (1964). Simple micro polar fluids, International Journal of Engineering Science,
pp. 205 – 217.
Fung, Y. C. (1964). Biomechanics, its scope, history and some problems of continuum
mechanics in physiology, Applied Mechanics Review 21, 1-20.
Gupta, B. B., Enigma, K.M. and Jaffrey, M. Y. (1982). A three layer semi – empirical model
for blood flow and particulate suspension through narrow tubes, ASME Journal of
Biomechanical Engineering 104, pp. 129 – 135.
Haynes, R. H. (1960). Physical basis of the dependence of blood viscosity on tube radius,
American Journal of Physiology 198, 1193 – 1205.
Hershey, D., Byrnes, R. E., Deadens, R. L. and Roam, A. M. (1964). Blood theology:
temperature dependent of the power law model. Presented at the A.I.CH.E. Meeting, Boston.
Hershey, D. and Chow, S. J. (1966). Blood flow in rigid tubes: thickness and slip velocity of
plasma film at the wall, Journal of Applied Physiology 21, 27 – 36.
60 V. P. Srivastava
Lew, H. S. and Fung, Y. C. (1970). Entry flow into blood vessels at arbitary Reynolds
number, Journal of Biomechanics 3, pp. 23 – 30.
Merril, E. W., Cokelet, G. R., Britten, A. and Wells, R. E. (1963). Non- Newtonian rheology
of human blood – effect of fibrinogen deduced by subtraction. Circulation Research 13, pp.48
– 56.
Merrill, E. W., Cokelet, G. R., Britten, A. and Wells, R. E. (1964). Rheology of human blood
and red cells plasma membrane. Biophysiology Anat. 4, pp. 51 – 63.
Muller, V.A. (1959). Die mehrschichtige rohrwand als model fur die aorta. Helv. Physiology
Pharmacology Acta 17, pp.131 – 145.
Pralhad, R. N. and Schultz, T. J. (1988). Two-layered blood flow in stenosed tubes for
different diseases. Biorheology 25, pp. 715-726.
Reiner, M. and Scott Blair, G. W. (1959). The flow of blood through narrow tubes, Nature.
Sharan, M. and Popel, A. S. (2001). A two-phase model for flow of blood in narrow tubes
with increased effective viscosity near the wall. Biorheology 38, pp. 415-428.
Shukla, J. B., Parihar, R. S. and Gupta, S. P. (1980). Effects of peripheral layer viscosity
on blood flow through an artery with mild stenosis. Bulletin of Mathematical Biology 42,
pp. 797-805.
Srivastava, V. P. (1995). Particle – fluid suspension model of blood flow through stenotic
vessels, International Journal of Biomedical Computing 38, pp.141 – 154.
Srivastava, V. P. (2000). Blood flow through stenosed vessels with a peripheral plasma
layer and applications, Automedica 18, pp. 271-300.
Srivastava, V. P. (2002). Particulate suspension blood flow through stenotic arteries: effect
of hematocrit and stenosis shape. Indian Journal of Pure & Applied Mathematics 33,
pp. 1353-1360.
Srivastava, V. P. (2003). Flow of a couple stress fluid representing blood flow through
stenotic vessels with a peripheral layer, Indian Journal of Pure & Applied
Mathematics 34, pp. 1727-1740.
Sud, V. K. and Sekhon, G. S. (1985). Arterial flow under periodic body acceleration, Bulletin
of Mathematical Biology 47, pp.35 – 52.
Thurston, G. B. (1989). Plasma release-cell layering theory for blood flow, Biorheology 26,
pp.199 – 214.
Wang, C. Y. and Bassingthwaighte, J. B. (2003). Blood flow in small curved tubes, Journal
of Biomechanical Engineering 125, pp. 910-913.
Womersley, J. R. (1954). Method for the calculation of velocity, rate of flow and viscous drag
in arteries when the pressure gradient is known, J. Physiol. 127, pp.38 – 51.
Womersley, J. R. (1955). Oscillatory motion of a viscous liquid in a thin walled elastic tube,
I. The linear approximation for long waves. Philosophical Magzine 46, pp. 199 – 221.