1 Discretization

1.1 Discretizing governing equation with central difference approximation

Starting equation  
DN O d dCN O
r + RN O = 0 (1)
r dr dr
Simplify
d2 CN O DN O dCN O
DN O + = −RN O (2)
dr2 r dr
Compare to form
u00 + P (r)u0 = F (r) (3)
1 RN O
u = CN O , P (r) = , F (r) = − (4)
r DN O
Taylor expansion
1
ui+1 = ui + u0i ∆r + u00i ∆r2 ,
2
1
ui−1 = ui − u0i ∆r + u00i ∆r2
2
Central difference
ui+1 − ui−1
u0i = ,
2∆r
ui+1 − 2ui + ui−1
u00i =
∆r2
Sub back into Eq. (3)
ui+1 − 2ui + ui−1 ui+1 − ui−1
+ Pi = Fi , h = ∆r (5)
h2 2h
where index i starts at 1.
Rearrange, let ai = h2 Pi
(1 − ai ) ui−1 + (−2)ui + (1 + ai ) ui+1 = h2 Fi (6)
2
(−2)ui = − (1 − ai ) ui−1 − (1 + ai ) ui+1 + h Fi (7)
1
ui = − − (1 − ai ) ui−1 − (1 + ai ) ui+1 + h2 Fi


(8)
2

1.2 Discretization schemes for boundary condition

1.2.1 Use imaginary node and correction term
u2 − u0
= 0 → u0 = u2 (9)
2h
uend+1 − uend−1
= 0 → uend+1 = uend−1 (10)
2h
Substitute into Eq. (3) when i = 1
1
(−2)u1 = −2u2 + h2 F1 (11)
2
Substitute into Eq. (3) when i = end

(−2)uend = − (1 − aend ) uend−1 − (1 + aend ) uend+1 + h2 Fend (12)

(−2)uend = −2uend−1 + h2 Fend (13)

1.2.2 Use second-order one-sided forward/backward approximation

From [1]:
−3u1 + 4u2 − u3
= 0 → −3u1 + 4u2 − u3 = 0 (14)
2h
3uend − 4uend−1 + uend−2
= 0 → 3uend − 4uend−1 + uend−2 = 0 (15)
2h
2 Solution algorithm
2.1 Gauss-Seidel
From [2]:
1
uk+1 k+1
− (1 + ai ) uki+1 + h2 Fi


i =− − (1 − ai ) ui−1 (16)
2

2.2 Successive overrelaxation method

 
2 k+1 2
− u = − (1 − ai ) uk+1 k 2
i−1 − (1 + ai ) ui+1 + h Fi − − 2 uki (17)
ω i ω
Alternative formula [3]:
 −1    
k+1 1 1 k
u = D+L −1 D−U u +f (18)
ω ω

3 Governing equation in individual compartment

3.1 Nitric oxide
RBC core r0 < r < r1  
DN O d dCN O
r − λcore CN O = 0
r dr dr
λcore ui
Fi =
DN O
CFL r1 < r < r2  
DN O d dCN O
r =0
r dr dr
0
Fi =
DN O
EC r2 < r < r3  
DN O d dCN O
r + RN O = 0
r dr dr
RN Omax PO 2
Fi = −
DN O PO2 + Km,eN OS
VW r3 < r < r4  
DN O d dCN O
r − λvw CN O = 0
r dr dr
λvw ui
Fi =
DN O
T r4 < r < r5  
DN O d dCN O
r − λt CN O = 0
r dr dr
λt ui
Fi =
DN O

3.1.1 NO matrix
     

 u1   0 
   
 −3 4 −1 ··· ··· ··· 0
 
    
     
..
     
2
u h F
 
1 − a2 −2 1 + a2 0 ··· ··· .  
 2 
 
 2 

     
 
..
   
 ..  
u
 
h 2
F

 0

1 − a3 −2 1 + a3 . .

 

3 



3 

 
. .
   
 .. ..
 
.. .. .. ..   .
. 
  .
. 

A= .
 . . . . . 
 , ũ =



, F =

  (19)
     
 . ..    
2

 . .

u
 
h F

 . 1 − an−2 −2 1 + an−2 0 i i
    
    
 
. .
   
 ..
 
  .
.
  .
.

 . ··· ··· 0 1 − an−1 −2 1 + an−1 
   
     
     
     
 2
u h F
  
 0 ··· ··· ··· 1 −4 3 n−1 n−1
     
    
     
un 0
3.2 Oxygen
RBC core r0 < r < r1
PO2 = 70
CFL r1 < r < r2  
DO2 d dPO2
α r =0
r dr dr
0
Gi =
αDO2
EC r2 < r < r3  
DO2 d dPO2
α r − RN O = 0
r dr dr
RN Omax PO2
Gi =
αDO2 PO2 + Km,eN OS
VW r3 < r < r4  
DO2 d dPO2 PO 2
α r − QO2 max V W =0
r dr dr PO2 + appKm
 
CN O
appKm = Km 1 + (20)
Cref
QO2 max V W PO 2
Gi =  
αDO2 PO2 + Km 1 + ui
Cref

T r4 < r < r5  
DO2 d dPO2 PO 2
α r − QO2 max T =0
r dr dr PO2 + appKm
 
CN O
appKm = Km 1 + (21)
Cref
QO2 max T PO2
Gi =  
αDO2 P + K ui
O2 m 1+ Cref

3.2.1 O2 matrix
 
   
1 0 0 ··· ··· ··· 0
 
 
   v   70 
   r1   
     
1 − ar1 +1 −2 1 + ar1 +1 0 ··· ··· 0     
     2 
  vr1 +1  h Gr1 +1 
 .. ..     

0 1 − ar1 +2 −2 1 + ar1 +2 . .
 
 . 
 
.

 ..  ..
   
   
.. ..
     
 .. .. .. ..     
A=
 . . . . . .  , ṽ = 
 
, B =  2
  
 (22)
   vi   h Gi 
 .. ..     

. . 1 − an−2 −2 1 + an−2 0
 
 . 
 
..

 .. 
   
   . 
..
     
     

 . ··· ··· 0 1 − an−1 −2 1 + an−1 
 




 2


   vn−1   h Gn−1 
     
0 ··· ··· ··· 1 −4 3
     
 
  vn 0
4 Velocity profile
4.1 Parameter manipulation
Ie is the second invariant of the rate of deformation tensor.
Velocity gradients:
   
∂u ∂u ∂u

∂v   ∂x ∂y ∂z  0 0 0
Lij = =  ∂v ∂v ∂v = (23)
  ∂v 
∂v 
∂x  ∂x ∂y ∂z   ∂x 0 ∂z 
∂w ∂w ∂w 0 0 0
∂x ∂y ∂z
∂v
as u = 0, w = 0,∂y= 0.
Rate of deformation tensor:
     
∂v ∂v
1 1 0 0 0  0 ∂x 0 0 ∂x 0
T  1  ∂v
Dij = (Lij + Lij ) =  ∂v  +  =  (24)
 ∂v   
∂v 
2 2  ∂x 0 ∂z  0 0 0
 2  ∂x 0 ∂z 
∂v ∂v
0 0 0 0 ∂z 0 0 ∂z 0
Second invariant of the rate of deformation tensor
1
Ie = [(Tr Dij )2 − Tr(D2ij )]
2
= D11 D22 + D22 D33 + D11 D33 − D12 D21 − D23 32 − D13 31
 2  2
1 ∂v 1 ∂v (25)
=0+0+0− − −0
4 ∂x 4 ∂z
"   2 #
2
1 ∂v ∂v
=− +
4 ∂x ∂z
Conversion from Cartesian derivatives to cylindrical derivatives [4]
"   2 #
2
1 ∂v ∂v
Ie = − +
4 ∂x ∂z
"  2 #
1 2 2 ∂v
= − (cos θ + sin θ)
4 ∂r (26)
 2
1 ∂v
=−
4 ∂r
1
= − γ̇ 2
4

4.2 Axisymmetric
γ̇ µ∞m √ √
Pg R1
= [r − α(1 + q) r + α2 + ( r − α)Cr ] (27)
µ∞
4µ∞m
dV Pg R1 µ∞m √ √
− = [r − α(1 + q) r + α2 + ( r − α)Cr ] (28)
dR 4µ∞m µ∞
where √ √
τ0 + µ∞ Λ
α= √
λτw
√ √
τ0 − µ∞ Λ
q=√ √ (29)
τ0 + µ∞ Λ
√
q
Cr = r − 2αq r + α2
where
µp
µ∞ = 1 2
(1 − 2 k∞ H)
2
1 − 12 k0 H

Λ = γc (30)
1 − 12 k∞ H
1 2
2 H(k0 − k∞ )
τ0 = µp γc
4
1 − 12 k∞ H
where
2.0
k0 = 0.275363 +
0.100158 + H
k∞ = exp {1.3435 + H[−2.803 + H(2.711 − 0.6479H)]} (31)
γc = exp {−6.1508 + H[27.923 + H(−25.6 + 3.697H)]}
5 Physiological parameters

Table 1: Systemic parameters and NCFL widths [5]

NCFL (%)
Aggregating conditions Hct (%) Diameter (µm) Outer Inner
Non 44.0 ± 1.6 52.5 ± 4.7 14.4 ± 2.1 11.1 ± 1.1
Normal 42.8 ± 1.7 50.0 ± 4.7 21.3 ± 3.4 13.5 ± 1.2
Hyper 42.2 ± 1.6 51.8 ± 4.4 23.6 ± 2.7 15.3 ± 1.7

5.1 CFL widths from mass conservation

Hematocrit measures proportion of volume of red blood cells (RBC) to total blood volume (RBC and plasma) [6]. Cell volume
fraction or tube hematocrit Ht is calculated as:
Nc Vef f
Ht = (32)
Vt
where Nc is the number of RBCs in the tube volume Vt = πR2 L, R is the tube radius, and L is the tube length [7].
An empirical relation between Ht and Hd is given by:
Ht
= Hd + (1 − Hd ) 1 + 1.7e−0.35D − 0.6e−0.01D


(33)
Hd
where D is the tube diameter in micrometers, and discharge hematocrit Hd is equal to systemic hematocrit [8].
Since systemic hematocrit in both cases (non-bifurcating and bifurcating flow) is the same, Ht is the same according to Eq.
33. Since Vt is the same due to constant vessel diameter, Nc assumed (?) to be the same, Vef f must be the same (Eq. 32).
Ignoring gravity effect and assuming symmetry, we need to find a egg / ellipse shaped graph with the same area as the circle
in non-bifurcating flow.
Non-bifurcating

Bifurcating

Assumptions: Define normal CFL width as mean of inner and outer CFL width.
N CF Linner + N CF Louter
N CF Lnormal = (34)
2
References
[1] M. R. Flynn. (2011) Some common difference approximation. [Online]. Available: http://websrv.mece.ualberta.ca/mrflynn/
finite difference equations.pdf
[2] F. Gounelas. (2005) Solving the 1D boundary value problem. [Online]. Available: http://www2.mathematik.hu-berlin.de/
∼gounelas/projects/Solving The 1D-Boundary Value Problem.pdf

[3] J.-L. Liu. (2011) Successive overrelaxation method (SOR). [Online]. Available: http://www.nhcue.edu.tw/∼jinnliu/teaching/
nde07/Lecture5.pdf
[4] S. K. L. Sjue. (2009) The laplacian operator from cartesian to cylindrical to spherical coordinates. [Online]. Available:
http://skisickness.com/2009/11/20

[5] Y. C. Ng, B. Namgung, H. L. Leo, and S. Kim, “Erythrocyte aggregation may promote uneven spatial distribution of
NO/O2 in the downstream vessel of arteriolar bifurcations,” Journal of Biomechanics, pp. –, 2015. [Online]. Available:
http://www.sciencedirect.com/science/article/pii/S0021929015006934
[6] H. H. Billett, Hemoglobin and Hematocrit. Boston: Butterworths, 1990, ch. 151.

[7] D. A. Fedosov, B. Caswell, A. S. Popel, and G. E. Karniadakis, “Blood flow and cell-free layer in microvessels,”
Microcirculation, vol. 17, no. 8, pp. 615–628, 2010. [Online]. Available: http://dx.doi.org/10.1111/j.1549-8719.2010.00056.x
[8] K. A. Lamkin-Kennard, D. Jaron, and D. G. Buerk, “Impact of the Fåhraeus effect on NO and O2 biotransport: a computer
model,” Microcirculation, vol. 11, no. 4, pp. 337–349, 2004.