Numerical Study on the Flow of a

Non-Newtonian Fluid Through an

Masahide Nakamura
Assistant of Mechanical Engineering. Axisymmetric Stenosis
Tadashi Sawada The laminar steady flow of non-Newtonian fluid (biviscosity fluid) through an axi-
symmetric stenosis is calculated using the finite element methods. The flow pattern,
Professor of Mechanical Engineering. the separation and reattachment points, and the distributions of pressure and shear
stress at the wall are obtained. Then, the axial force acting on the stenosis is
Akita University, Akita, Japan evaluated. It is suggested by the authors that this force can become one of the causes
of post-stenotic dilatation. Calculated results show that the non-Newtonian proper-
ty of blood weakens the distortion of flow pattern, pressure and shear stress at the
wall associated with the stenosis and that the non-Newtonian property of blood
decreases the axial force acting on the stenosis.

Introduction
It is well known that the strain and stress distributions in the blood. But in this study, biviscosity model [10],[11] is used as
vicinity of an arterial stenosis play an important role in the a constitutive equation of blood. This model is written as:
development of vascular lesions. As abnormal strain and '2{\l.BJ!-py/4lTs)elj, 7T>XC
stress distributions near the arterial stenosis are caused mainly (1)
by the abnormal fluid force acting on the arterial wall, it is
very important to analyze the blood flow through an arterial
stenosis. A number of studies [l]-[5] on the blood flow The following quantity is introduced as a nondimensional
through an arterial stenosis have been performed, but the non- parameter including wc.
Newtonian property of blood is not taken into consideration & = fx,B4l^c/py (2)
in these studies. As blood shows the remarkable non-
The value of /3 denotes the upper limit of apparent viscosity
Newtonian property in low shear rate and the shear rate is low
coefficient. As the upper limit of apparent viscosity coefficient
in the downstream side of the stenosis, it is considered that the
of blood is expected to be very high, very small value of (3
analysis of the flow pattern near the stenosis should include
should be used. In this study, j3 is set at 0.01. If j8 becomes
the non-Newtonian property of blood. Moreover, some ex-
twice as large as this value, calculated results change for about
perimental results [6]-[8] exist, which show the importance of
1 percent. Therefore, the calculated results are not affected
the non-Newtonian effects of blood.
very much by the value of |8.
In this paper, blood is considered to be a non-Newtonian The reasons to adopt this model are explained subsequently.
fluid, that is, a biviscosity fluid. The reason to adopt a At first, comparison of the biviscosity model with the
biviscosity model as a constitutive equation of blood is Casson model is performed. Flow curves of these models are
discussed later. The laminar steady flow through an axisym- shown in Fig. 1, where D = shear rate, T = shear stress,
metric stenosis is analyzed and the flow pattern, the separation £ 0 = 5 x l 0 - 3 P a , fc, = 4 x l 0 3 P a . s , pB = 4 x l O ^ P a - s ,
and reattachment points, and the distributions of pressure and py = 4x 10~ 2 Pa. Figure 2 shows the apparent viscosity coeffi-
shear stress at the wall are obtained. As the torque generated cient ( = T/D) of these models. These figures show that two
by the axial force acting on the stenosis can be one of the constitutive equations resemble each other except the very low
causes of post-stenotic dilatation as suggested by the authors shear rate region. So, it is suggested there is no significant dif-
[9], this force is also calculated, and on the basis of these ference between the Casson model and the biviscosity model.
results, the effect of non-Newtonian property of blood on the We should note that the yielding stress of biviscosity model is
development of vascular lesions, especially post-stenotic much greater than the one of Casson model. In this case, the
dilatation, are discussed. ratio of yielding stress is about 8. Usually, the value of this
ratio is order of 10.
Constitutive Equation and Its Verification Next, the calculated velocity distributions on basis of
Usually, Casson model is used as a constitutive equation of biviscosity model are compared with the measured ones. The
measurement of velocity distributions in vivo are performed in
recent years [6], [7]. The velocity distributions shown in Einav
Contributed by the Bioengineering Division for publication in the JOURNAL OF
BIOMECHANICAL ENGINEERING. Manuscript received by the Bioengineering Divi- et al. [6] indicate the existence of flatness region near the
sion July 1, 1987; revised manuscript received February 27, 1988. center of artery and indicate the non-Newtonian effects. These

Journal of Biomechanical Engineering MAY 1988, Vol. 110/137

Copyright © 1988 by ASME

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/22/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use
 1.0

a 1.0 9 - - - - °~
a. o:/T= fkl^fk^D

bi-viscosity
Newtonian
0.5
bi-viscosity

o : Einav et al.

R„=47.5 |im

0.5
Fig. 1 Flow curves of Casson model and biviscosity model,
k n = 5 x 1 0 - 3 Pa, fc, = 4 x 1 0 ~ 3 Pa-s r/R r
Fig. 3 Comparison of the calculated velocity distributions with the
measured one [6], f?0 = 47.5 pm

x10~

6
QL R
o-ix= fko+Jk^D S
s O \

\
\ \ \\
bi-viscosity
- Newtonian -
/\ V
0.5 \\ \\
- bi-viscosity \ \
\\o
\ \
o: Einav et al. \ \\
\
M
50 100 |0.m
i i i

Fig. 2 Apparent viscosity coefficients of Casson model and biviscosity 0.5

model, k0 = 5 x 1 0 - 3 Pa, k-, = 4 x 1 0 - 3 Pa-s r/Rn

Fig. 4 Comparison of the calculated velocity distributions with the

measured one [6], R0 = 39 i<m
characteristics of velocity distributions are clear specially in
case of D0 = 95 nm in Einav et al. [6]. We try to evaluate the
yielding stress of biviscosity model based on the velocity 8/R0 = 0.02 after Charm [12] and the viscosity of plasma layer
distribution of D0 = 95 /xm, which shows that the relative plug is assumed as 1 x 10~3Pa»s. Results are:
radius is about 0.2. Blood is treated as a biviscosity model. By
using the following values: viscosity = 4x 10~3Pa«s, densi- Re = 0.107, py = 0.21 Pa
ty =1050 kg/m3, the Reynolds number Re and the yielding As the length of entrance region 1 is very short
stress py are calculated. As the diameter of this artery is 95 (1 =0.065Re«D0 = 0.66 /mi), we can neglect the entrance
fim, the effect of plasma layer should be considered. In this length effects. Moreover, as the Womersley number is about
calculation, the thickness of plasma layer S is assumed as 0.05, the unsteadiness effect can be neglected.

Nomenclature
D = shear rate in simple shear
flow p* = nondimensional pressure 0 = \/2irc'tiB/py, see equa-
D0 = diameter of the artery far Py = yielding stress tions (1),(2)
away from the stenosis R = inner radius of the HB = plastic viscosity
ft axial force acting on the stenosis, see equation (27) T = shear stress in simple
stenosis R0 = D0/2 shear flow
T
n nondimensional axial
force acting on the
Re = Reynolds number,
Re = pD0v/iiB
,j = (ixf) component of the
deviatoric stress
stenosis u,v = radial and axial com- TW = shear stress at the wall
He Hedstrom number, ponents of the velocity T* = nondimensional shear
Ue = 4PpyR§/n2B V = mean axial velocity far stress at the wall
I = stenosis length away from the stenosis ir = e^, where etj is the (ij)
Ma,Np = interpolation function, r,0,z = coordinate component component of the defor-
a = l , 2 , . . ,6, 0=1,2,3 (dp/dz)0 = pressure gradient far mation rate
p = pressure away from the stenosis p = density

138/Vol. 110, MAY 1988 Transactions of the ASME

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/22/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 One may think that this yielding stress is different very u, v, p.
much from the one which is widely noted, but we can under-
stand this value by considering the fact that the yielding stress
v
of biviscosity model is much greater than the one of Casson ^ 5 5
model and the uncertainty of the measured one. The
calculated velocity distributions based on the biviscosity U
u 3V3P3
model and py = 0.21 Pa is shown in Fig. 3. For reference, a 6 v6
graph of py = 0.0 Pa (Newtonian fluid) is given in Fig. 3,
where vm= maximum axial velocity. Comparison of the U
calculated velocity distributions based on the biviscosity A VA
model with the other data shown in Einav et al. [6] is per-
formed in Fig. 4. Of course, calculation is performed by using u2 v 2 p 2
the same values to the calculation of Fig. 3. These figures
show that the biviscosity model can be used as a constitutive Fig. 5 Finite element
equation of blood and show the importance of non-
Newtonian effects. 1 d dv
+
= 0 (4)
It is clear that the value of 5 has some uncertainty. So, the r dr dz
calculation should be performed by using the other value of
5/R0. For example, the calculated yielding stress of biviscosity
model becomes about 0.19 Pa if the value of 8/R0 is set at / Py\ dU / Py\ ( dU dV \
0.04. This value does not conflict with the results mentioned
above. Therefore, it is expected that the uncertainty of the
thickness of plasma layer does not produce the results which
conflict with the results mentioned above. / py\ dv ( py\ u
Figures 1 and 2 show that the difference between the Casson
model and the biviscosity model exists in a very low shear rate
region. But Figs. 3 and 4 show that the calculated velocity (5)
distributions based on the biviscosity model agree with the
measured one pretty well. So, it is expected that this difference 7r>7Tc

is not so significant.
7 T < 7TC
The velocity distributions of frog are measured by (6)
Horimoto et al. [7] and they point out that the measured The boundary conditions are:
velocity distributions differ from the calculated ones based on (i) inlet:
the Newtonian model. This difference resembles the difference
between the biviscosity model and the Newtonian model w=0
shown in Figs. 3 and 4. (7)
Next, the advantage of the biviscosity model compared with 1 / dp \ ^--Rl
the Casson model is discussed. The advantage of the biviscosi- (r-R0) + (—^-) /•;</•</?<,
ty model appears when the numerical calculation is perform- Ma v °' 4 V dz / o MB
ed. The shear stress can be written explicitly if the biviscosity
model is adopted [ 10], [ 11 ]. On the other hand, it is well known
that the form of basic equation of Casson model is changed — ( dzSo— ) M l + 0 - 1 ) + Vn r<r,
very much by the stress distributions in a fluid and the shear 4 \ dz So
stress cannot be written explicitly at a very low shear rate where, (8)
region. So, numerical calculation based on the Casson model
2py
is very difficult compared with the biviscosity model. y -U + ffl.
Therefore, the biviscosity model is used as a constitutive equa- idp/dz\
tion of blood in this study.
Lastly, we should add some opinion to the biviscosity v0=——(r;-R0)
model. Of course, we do not think that the biviscosity model is
the most excellent model that reveals the characteristics of the
blood. At a very low shear rate region, there is a possibility (9)
that the adequacy of the biviscosity model is poor. So, the 4 V dz / o I aB
PB u/ ^B((Il++I/S3"- 11 )) -J
study on a constitutive equation should be carried on in the (ii) outlet:
future. However, as the mean shear rate of this calculation is M = 0, /, = 0 (10)
about 20-150 s ~ \ the biviscosity model is expected to be ef-
fective within the present calculation.
p=0 (11)
Basic Equations and Numerical Calculation Methods where, tz = axial component of the boundary force,
(iii) wall:
Basic equations for the axial symmetric incompressible flow
model are written as: u=v=0 (12)

dp drrr drrz T r r -T 9 , (iv) center:

/ bu du \ | | |

dr dr dz r H = 0, dv/dr = 0 (13)
Equations (7) and (8) represent a fully developed flow of
/ dv dv \ dp dr. dTrz T
rz
biviscosity fluid in a pipe. The outlet boundary conditions
V dr dz / dz dz dr r
given by equation (10) are similar to the ones which are used in
the calculation of entry flow [13] and equation (11) is used for
(3) the convenience of numerical calculation.

Journal of Biomechanical Engineering MAY 1988, Vol. 110/139

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/22/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 A number of numerical calculation methods have been pro-
posed and applied to the viscous flow. The most popular in/ 11111111111111111111
methods among them may probably be the finite difference , D n /4
methods. But it has become clear that the finite element
method has the advantage over the finite difference method
when solving the viscous flow in complex geometries. Con-
sidering this advantage, the finite element method is chosen in 7777 777777777 777777777"
this study.
A finite element used in this study is shown in Fig. 5, which Fig. 6 Stenosis model
is the most popular type of element applied to the viscous
flow. In the element, the pressure is written as:
p=paNa (14)
When deriving these equations, the value of g is approx-
where, pa = pressure at node (a = 1,2,3) The axial and radial imated by the mean value in an element which is denoted by
components of velocity are given by: ge. When the variation of velocity in an element is small, the
adequacy of this approximation is clear, but when the varia-
u = u0Ml3, v=vl3Mp (15) tion of velocity in an element becomes violent, the adequacy
where uB, Vg = radial and axial component of velocity at node of this approximation is doubtful. It is, however, expected
0 ( 0 = 1 , 2 , 3 , . . . ,6) that the error of calculated results due to this approximation
By integrating the product of equation (3) and the second in the letter is small, because the ratio of py/ge to ixB is small.
order interpolation function Ma over one element, the follow- The integral of equations (18)-(23) can easily be performed
ing equations are obtained: except Kap. The integrand of Ka0 contains a term which
diverges at r = Q. This trouble can easily be avoided by con-
E^u^+Ei^v^+IT^Pg +Kalluli+LalJV(t = 0 (16)
sidering that the value of u at r=Q are determined from the
boundary conditions and the values of KaB that contains a
Eral3y ue vy + E%By vp vy + HzapP& +RaliUp + Zal3 vp = 0 (17) term which diverges at r = 0 are not necessary for the calcula-
where, tion of velocity distributions. Therefore, the integrals required
for the calculation of velocity distributions can be performed
by means of such an ordinary methods as Gaussian
Erapy = P})Ve MaM0—My2irrdrdz, E\^ quadrature.

Calculated Results
-p\\ MaMB-—M„2irrdrdz (18) The basic equations and the numerical calculation methods
v
JiVe dz mentioned above are applied to the laminar steady flow of
biviscosity fluids through an axisymmetric stenosis. The
calculation conditions are written as:
H'aB = j J ^ Ma—NB2irrdrdz, H^
Re: 0-100
(25)
He:
-II Ve
Ma—-Ng2irrdrdz
OZ
(19) 0, 67.3, 376.0

Hedstrom number is used as the nondimensional parameter

dMa dMg which denotes the non-Newtonian effects and this parameter
^ = 2(^)Jjve Flirrdrdz, F=
dr
-—-1-
or
is defined as:

He = 4pp R20/ix2B (26)

1 dMa dMp MaMB
-| - • « 1- - (20) Some additional explanation for the calculation conditions
dz dz
of equation (25) are presented. As our main target is a
peripheral artery, the range of Reynolds number can be
1 dM„ dM,
+ V-2-xrdrdz (21) understood easily. It is easily expected that there are many dif-
^ = 4- f-)IL 2 dz dr ferences between individuals in the diameter of artery or in the
blood velocity. So, we cannot fix the value of Hedstrom
1 dMe dM„ number and it is very important to evaluate the upper limit of
p (22) Hedstrom number. At first, the range of the diameter of
ve 2 dr - dz 2irrdrdz artery is evaluated. Reference [3] shows the Reynolds number
and the diameter of each artery. It is found that the Reynolds
1 dMa number shown in this reference is low comparing with the
1 G2irrdrdz, G= Reynolds number shown in the other literature under the same
J Ve dr
diameter of artery. So, the upper limit of diameter of this
calculation is set at 5.Ox 10~3m after [3]. By using the follow-
dMB dMa dMg (23) ing value: />= 1050 kg/m 3 , /xfl = 3.5x 10" 3 Pa«s,Pj, = 0.27 Pa
dr + _ • _dz
dz (result of Fig. 3 or Fig. 4), the upper limit of Hedstrom
number is calculated as:
In these equations, Ve shows the range of the integration (e is
the index of element) and Se shows the boundary of Ve. (He) upper Hmit = 580
In the same manner, the following equation can be obtained Our calculations are performed within this upper limit.
from equation (4). Moreover, we assume that the effect of plasma layer can be
neglected. This assumption can be understood by considering
H^Up + HzaS Vs = 2irrNa(udz + vdr) (24) the Reynolds number in this study.
J Se

140/Vol. 110, MAY 1988 Transactions of the ASME

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/22/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 100 -
/ / ••"'

/ / /
i
He=0 |\\
75
/ / , -

\ / / / -
50
/ / , - • - " Re = A0 -'//r\
cc - 60 -/// 1 /
. / , / He = 0
25
,-'"' H e = 6 7.3
80 - y u without

1 00 -> stenosis
H e = 376.
I 1 I
-60 i 1
0.5 1.0 1.5 2.0 -1.0 -0.5 0 0.5 1.0
z/D 0 2/D0
Fig. 7 Variation of the separation and reattachment points with Fig. 10 Change of the nondimensional pressure (Newtonian fluid)
Reynolds number

/— z'Do=0.5 /- z / D o = 0.5

=~~-^ \// / — 2.0 \ / / - 2.0 -

/— 4.0 / / - 4.0 H e = 376.
r / ^- CO
~~~\ \/\//—
,
6.0
CO
^~~~"~\ \/

^-^\ -
N^.
Re = 40 -* ) \
- Vs\ov
He = 0 \ \J^ik> He = 0 \ ^ V " 6 0 -1 H
80 -1 1
Re=l00 Re = 40
1 00 -1 \ / without
0.5 1.0 0 0.5 stenosis
r/R0 r/R0
-1.0 -0.5 0 0.5 1.0
Fig. 8 Velocity distribution (Newtonian fluid)
z/D0
Fig. 11 Change of the nondimensional pressure (biviscosity fluid)
y— Z/D 0 =0.5 /— Z / D 0 = 0.5

\y /— 2.0
\ /'/— 2.0
/ — 40 /— 4.0
which the flow separation occurs is greater than the one of
\A s— 5.0
A / /— °° water. Present results do not conflict with their results
\/— °° qualitatively.
=
" ^ ^ ^ / / / ^
The calculated velocity distributions in the downstream side
- ^ K ^
- V**%. of the stenosis are shown in Figs. 8 and 9, where t; = mean ax-
He=376
\^~5V He=376 ial velocity far away from the stenosis. The dominating non-
dimensional parameters are Reynolds number Re and the
Re=l00 Re=40 Hedstrdm number He. One of the main purposes of this study
is to evaluate the effect of Hedstrom number and this effect
r/R0 r/Rr should be evaluated at constant Reynolds number. So, the
comparison of velocity distributions are performed at con-
Fig. 9 Velocity distribution (biviscosity fluid)
stant Reynolds number. The existence of reverse flow near the
wall is recognized at (Re,He) = (40,0),(100,0),(100,376) and it
is found that the decrease in Hedstrom number and the in-
The shape of axisymmetric stenosis used in this calculation crease in Reynolds number facilitate the occurrence of reverse
is the same one as has been used by Lee and Fung [2] and the flow. It is also found that a dip appears in a velocity profile at
inner radius of stenosis is given by: z/D0 = 0.5 and Re =100, which does not conflict with the
results of Forrester and Young [5].
i? = tf0[1.0-0.5.exp(-16z2 /£>§)] (27) On the other hand, the calculated velocity distributions
First, the calculated results concerned with the flow pattern become closer to the one of fully developed flow with the in-
are shown. The separation and reattachment points are very crease in z/D0. The increasing Reynolds number increases the
important results from a biomechanical viewpoints. The varia- entrance length and the dependence of Hedstrom number on
tions in separation and reattachment points with the Reynolds the entrance length is weak in the range of present calculation.
number are shown in Fig. 7. The curves in figure indicate that In Fig. 9, it is found that the velocity distributions far away
the non-Newtonian effect increases the critical Reynolds from the stenosis has a flatness region near the center of
number for the initiation of the separation and decreases the artery. It is expected that the blood becomes solid within this
length between the separation and reattachment points. An ex- region. This velocity distributions show the characteristics of
periment on the flow through a mild axisymmetric stenosis non-Newtonian fluid with the yielding stress.
was performed by Forrester and Young [8]. They found that Next, the pressure distributions and shear stress at the wall
the length of the separated flow region of blood is smaller than are calculated. The following nondimensional quantities are
the one of water and the critical Reynolds number of blood introduced:

Journal of Biomechanical Engineering MAY 1988, Vol. 110/141

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/22/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 j
1 5 -
Re=AO Re=«0
1 0 He = 0 ! Re=IOO He=376 Re=100

' JI
// |
5
"
-Js
0

I 1 i i r
-1.0 -05 0 0.5 1.0 -1.0 0 0.5
z/D0 z/D0
Fig. 12 Change of the nondimensional shear stress at the wall (Newto- Fig. 13 Change of the nondimensional shear stress at the wall
nian fluid) (biviscosity fluid)

P"=- w (28)
(dp/dz)0D0' (dp/dz)Q-D0/4
where, (dp/dz) 0 = pressure gradient far away from the
stenosis.
The calculated nondimensional pressure p* on the wall are
shown in Figs. 10 and 11, where the values of p* at z/D0 = - 1
are set at 0 and a dotted line shows the results of p* without
stenosis. The increasing Hedstrom number has an effect which
decreases the pressure drop in the converging part of stenosis.
The calculated nondimensional shear stress at the wall are
shown in Figs. 12 and 13. The increase in Hedstrom number
decreases the peak value of T£. The point where the maximum 100
wall shear stress occurs is on the upstream side of stenosis for
all cases. Fig. 14 Nondimensional axial force acting on the stenosis
Lastly, the axial force acting on the stenosis is obtained. The
following nondimensional quantity is introduced.
rz=fz/^D0rS (29)
flow holds. This result does not conflict with the other studies
where, fz = axial force acting on the stenosis, T0 = shear stress [14]-[17].
at the wall far away from the stenosis, / = stenosis length = D0 Lastly, the effects of unsteadiness are discussed briefly. It is
(see Fig. 6). well known that the blood flow in an artery has a pulsation.
Calculated results are shown in Fig. 14. The increasing The effects of unsteadiness has some intrinsic interest.
Reynolds number and the decreasing Hedstrom number in- Because the change of wall shear stress and the change of the
crease the nondimensional axial force acting on the stenosis. length of separated flow region affect the damage of arterial
wall. Womersley number or Strouhal number are widely used
Discussion as a nondimensional parameter which denotes the effects of
unsteadiness. The value of these parameters are different in
The results mentioned above indicate that the non- each artery [3]. As the main target of this study is a peripheral
Newtonian effects weaken the distortion of flow pattern, artery, it is expected that the effects of unsteadiness are weak.
pressure distribution and shear stress at the wall associated But we cannot definitely conclude that the effects of
with the stenosis. Therefore, it is expected that the non- unsteadiness are weak. So, unsteadiness analysis should be
Newtonian property of blood can be one of the factors which needed in the future. This analysis is now going on and the
inhibit the lesions of vascular, which can not be definitely results will be reported at a later date.
concluded.
A number of uncertainties are still left in the field of
One of the purposes of this study is to evaluate the effects of hemorheology although many studies have been performed.
the non-Newtonian property of blood on the development of Therefore, it is impossible to make quantitative predictions
post-stenotic dilatation. One answer to this problem is in- from the present calculations. More detailed theoretical and
volved in Fig. 14. As seen in Fig. 14, the non-Newtonian prop- experimental studies should be performed in the future.
erty of blood works to decrease the axial force acting on the
stenosis. Therefore, if the hypothesis that this axial force is
one of the causes of post-stenotic dilatation proposed by the References
authors is reasonable, the non-Newtonian property of blood 1 Young, D. F., "Fluid Mechanics of Arterial Stenosis," ASME JOURNAL OF
has an effect which inhibits the post-stenotic dilatation. In ad- BlOMECHANICAL ENGINEERING, V o l . 1 0 1 , 1979, p p . 157-175.
dition, Fig. 14 indicates that the increasing Reynolds number 2 Lee, J. S., and Fung, Y. C , "Flow in Locally Constricted Tubes at Low
Reynolds Numbers," ASME Journal ofApplied Mechanics, Vol. 37, 1970, pp.
increases the nondimensional axial force ,/J. This fact means 9-16.
that the increasing Reynolds number facilitates the generation 3 Fukushima, T., Azuma, T., and Matsuzawa, T., "Numerical Analysis of
of post-stenotic dilatation. Therefore, it is expected that the Blood Flow in the Vertebral Artery," ASME JOURNAL OF BIOMECHANICAL
post-stenotic dilatation easily occurs under the high Reynolds ENGINEERING, Vol. 104, 1982, pp. 143-147.
4 Skalak, R., Keller, S. R., and Secomb, T. W., "Mechanics of Blood
number where transition to the turbulence is easy to occur Flow," ASME JOURNAL OF BIOMECHANICAL ENGINEERING, Vol. 103, 1981, pp.
compared with the low Reynolds number where the laminar 102-115.

142/Vol. 110, MAY 1988 Transactions of the ASME

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/22/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

 5 Forrester, J. H., and Young, D. F., "Flow Through a Converging-Diverg- Pulsatile Flow of Slurries," Journal of Non-Newtonian Fluid Mechanics, Vol.
ing Tube and Its Implications in Occlusive Vascular Stenosis-I, Theoretical 22, 1987, pp. 191-206.
Development," Journal of Biomechanics, Vol. 3, 1970, pp. 297-316. 12 Charm, S. E., and Kurland, G. S., "Blood Rheology," Cardiovascular
6 Einav, S., Berman, H. J., Fuhro, R. L., DiGiovanni, P. R., Fine, S., and Fluid Dynamics, Vol. 2, ed., D. H. Bergel, Academic Press, N.Y., 1972, pp.
Fridman, J. D., "Measurement of Velocity Profiles of Red Blood Cells in the 157-203.
Microcirculation by Laser Doppler Anemometry (LDA)," Biorheology, Vol. 13 Crochet, M. J., Davies, A. R., and Walters, K., Numerical Simulation of
12, 1975, pp. 207-210. Non-Newtonian Flow, Elsevier, Amsterdam, 1984, pp. 234-240.
7 Horimoto, M., Koyama, T., Mishina, H., and Asakura, T., " Pulsatile 14 Roach, M. R., "An Experimental Study of the Production and Time
Blood Flow in Arteriole of Frog Web," Biorheology, Vol. 16, 1979, pp. Course of Poststenotic Dilatation in the Femoral and Carotid Arteries of Adult
163-170. Dogs," Circulation Research, Vol. 13, 1965, pp. 537-551.
8 Forrester, J. H., and Young, D. F., "Flow Through a Converging-Diverg- 15 Roach, M. R., "Changes in Arterial Disintensibility as a Cause of
ing Tube and Its Implications in Occlusive Vascular Stenosis-II, Theoretical and Poststenotic Dilatation," American Journal of Cardiology, Vol. 12, 1963, pp.
Experimental Results and Their Implications," Journal of Biomechanics, Vol. 802-815.
3, 1970, pp. 307-316. 16 Holman, E., "The Obscle Physiology of Post-Stenotic Dilatation: Its
9 Nakamura, M., and Sawada, T., to be published. Relation to the Development of Aneurysms,'' Journal of Thoracic Surgery, Vol.
10 Keentok, M., Milthorpe, J. F., and O'donovan, E., "On the Shearing 28, 1954, pp. 109-133.
Zone around Rotating Vanes in Plastic Liquids: Theory and Experiment," 17 Roach, M. R., "Poststenotic Dilatation in Arteries," Cardiovascular
Journal of Non-Newtonian Fluid Mechanics, Vol. 17, 1985, pp. 23-25. Fluid Dynamics, ed., D. H. Bergel, Vol. 2, Academic Press, N.Y., 1972, pp.
11 Nakamura, M., and Sawada, T., "Numerical Study on the Laminar 111-156.

First National Conference on Biomechanics

Indian Institute of Technology, New Delhi-110016
December 17-19, 1988

The conference aims to deal with the following topics:

Soft and hard tissue mechanics, Biofluid dynamics, Physiological systems mechanics,
Modelling in biomechanics, Transport phenomena in living systems, Occupational hazard,
Injury, safety and rehabilitation, Orthotics and prosthetics, Artificial Organs, Computers
in biomechanics, Mechanics of biomaterials, Sports biomechanics, Ergonomics, Mechanics
of reproductive system and instrumentation.

For further information please contact: Dr. K. B. Sahay, Secretary, First National Con-
ference on Biomechanics, Centre for Biomedical Engineering, Indian Institute of
Technology, New Delhi 110016, India.

Journal of Biomechanical Engineering MAY1988, Vol.110/143

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 01/22/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use