Flow in bifurcations
Nico A.M. Schellart, Dept.
of Med. Physics, AMC
Principle

Fig. 1 Bifurcation
As Fig. 1
shows, a bifurcation changes the circular symmetric parabolic fluid flow (of a
liquid or gas) in a circular mother tube to a skewed profile in the daughter
branches. The radius ratio of mother and daughter branches, rd/rm,
is energetically optimal for a ratio of 2–1/3 = 0.79. This gives a
daughters/mother area ratio of

Fig. 2 Model of the flow profile of air in the
trachea bifurcation
Application
Vascular and
airways flow, and flow in the kidney. They play an important role in the study
of the origination and grow of arteriosclerosis and aneurisms.
In the
airways tree of the lungs some 26 generations (orders) of bifurcations, from
trachea to the most remote alveoli can be distinguished. In mammals, the
bifurcation ratio rd/rm, is on average 0.78 and the
bifurcation angle 67 o, 8o less than the optimum. Human
midrange generations show higher angles: 79 o. The two semi-bifurcation
angles are more similar for higher generations. From the 2nd to the
10th generation the human bifurcation ratio rd/rm
is very close to 0.79, but for higher generations the ratio slowly increases
and the highest generations hardly show a diameter decrease from mother to
daughter. The first generation bifurcation shows daughter diameters which are
much smaller than the optimum. However, above optimizations holds for
Poisseuille flow (see Poisseuille’s
Law), and this does not hold for the human trachea.
In the
vascular system of mammals an averaged value of 0.68 is found for rd/rm.
In the myocardium, rd/rm goes from 0.79 (the optimum) for
the capillaries to about 0.72 for arteries of some mm diameter. Nevertheless,
the optimization is poor since the variation is very large with many much too
small or large daughters. Also the bifurcations are rather asymmetric, going
from 0.8 (= rsmall daughter/rlarge daughter) of
capillaries to about 0.2 of mm-sized arteries. For energetic reasons, asymmetry
in diameter is accompanied by asymmetry in the semi-bifurcation angle. The
human brain the mean bifurcation angle is 74o with tens of degrees
of variation. Branching angles in muscles can deviate substantial from their
fluid dynamic minimum cost optimum. Classically, minimum cost according to
(rd1/rm)3+
(rd2/rm)3 = 1.
With rd1
= rd1 the smallest minimum (with the ratio 0.79) is obtained. This
solution is about the same as that on the basis of calculations with the Womersley number
for laminar flow with minimal local wave reflection as criterion. For a given
radius ratio, the optimal angles can be calculated.
The
experimentally obtained exponents of
More info
Fig. 3
illustrates in more detail the flow patters in a bifurcation.

Fig. 3 a. The dashed profile in the upper daughter
branch is the velocity profile perpendicular on the plane of the bifurcation.
b. the dashed line is in the plane of the bifurcation (the symmetry plane). c. Stream lines with points of separation and
reattachment indicated. d Streamline
of flow impinging onto the apex.
When the
direction of flow is inverted, in the mother tube four secondary loops can
arise (Fig. 4).
Secondary
flows are characterized by a swirling, helical component superimposed on the
main streamwise velocity along the tube axis, see also Flow in a bended
tube). A
complicating factor is wall compliance, especially of importance with pulsatile
flow, another factor which makes the pattern more complicated.

Fig. 4 Inversed flow in a bifurcation.
Literature
Canals M et. al.
A simple geometrical
pattern for the branching distribution of the bronchial tree, useful to
estimate optimality departures. Acta
Biotheor. 2004;52:1-16.
Frame MD,
Sarelius IH.Energy optimization and bifurcation angles in
the microcirculation. Microvasc
Res. 1995 Nov;50(3):301-10.
Pedley TJ
et. al, Gas flow and mixing in the airways, In: West, J.B. (ed.) Bioengineering Aspects of the Lung.
VanBavel E,
Spaan JA.
Branching patterns in the porcine coronary arterial tree. Estimation of flow
heterogeneity. Circ Res. 1992;71:1200-12.
http://www.vki.ac.be/research/themes/annualsurvey/2002/biological_fluid_ea1603v1.pdf