NTEC 1

Lecture CFD-3
2014 31

CFD modelling of multiphase flows

Simon Lo
CD-adapco
Trident House, Basil Hill Road
Didcot, OX11 7HJ, UK
simon.lo@cd-adapco.com

NTEC 2
Multiphase models and applications
2014 31
•  VOF
–  Free surface flows
•  LMP
–  Droplet flows
–  Liquid film
•  DEM
–  Particle flows
•  EMP
–  Particle flows
–  Bubbly flows
–  Population balance
–  Boiling heat and mass transfers
–  Interphase mass transfer

1
NTEC 3
Volume of Fluid (VOF) model
2014 31

•  Solve equation for volume fraction (based on conservation of mass) to

identify location of gas and liquid.

∂α i
+ ∇.(α i u ) = m! i
∂t
•  Momentum equation for the gas-liquid mixture:
∂
(ρmui ) + ∂ (ρmu j ui − τ ij ) = − ∂p + ρm gi + M
∂t ∂x j ∂xi
•  Properties of mixture:
N Gas flow
ρ m = ∑ (α i ρ i )
i =1

N
µ m = ∑ (α i µi )
i =1
Liquid flow

NTEC 4
Slug flow in interconnected subchannels
2014 31
Calculation grid 204,512 cells
18.7 mm
20 mm

1.7 mm

Water Inlet 0.23 m/s

150 mm

Air Inlet 2.0 m/s Air Inlet 0.5 m/s

2
NTEC 5
Slug flow in interconnected subchannels
2014 31

NTEC 6
Effects of surface tension
2014 31

S. Tension = 0.072 S. Tension = 0.06

3
NTEC 7
Free surface flow in drink carton
2014 31

NTEC 8
Lagrangian model for particle flows
2014 31

•  Equation of motion for individual particle:

du dx
md d = F ud = d
dt dt
•  F = force acting on particle

Particle
Gas flow

4
NTEC 9
Spray modelling
2014 31

•  Modified Han et al atomisation model

Charalambos (2002)!

NTEC 10
Droplet collision and coalescence
2014 31

5
NTEC 11
Spray coating
2014 31

NTEC 12
Fluid film boiling
2014 31

Simulation of films and their evaporation on high temperature surfaces

•  Walls above saturation temperature of droplet fluid
•  Required multi-component film

6
NTEC 13
DEM (Discrete Element Method)
2014 31

•  Linear momentum of particle:

d vi
mi = F Drag + F Contact + F Other
dt

•  Angular momentum:

dω i k
! !
Ii
dt
[
= ∑ τ ij + M ij ]
j =1
! ! !
M ij = − µ roll FContact ωi
!
•  M ij = rolling torque
•  µ roll = rolling friction coefficient.

NTEC 14
Pneumatic conveying of particles in pipe
2014 31

7
NTEC 15
Spreading of particles by conveyor belts
2014 31

NTEC 16
Eulerian multiphase model
2014 31

•  Conservation of mass of phase k:

N
∂
(α k ρk ) + ∇.(α k ρk uk ) = ∑ (m! jk − m! kj )
∂t j =1

•  Conservation of momentum of phase k:

∂
(α k ρ k uk ) + ∇.(α k ρ k uk uk )
∂t
t
= −α k ∇p + α k ρ k g + ∇.α k (τ k + τ k ) + M k

Particle Gas flow

8
NTEC 17
Forces on a particle
2014 31

•  Forces acting on a particles: L, T

–  Buoyancy, B.
–  Drag, D. uc
–  Lift, L.
D V
–  Virtual mass, V. ud
–  Turbulent dispersion, T.
–  Basset force.
g
–  And others. B

•  Buoyancy and drag are the dominant ones.

ud B
•  Basset force is complicated and almost always
ignored. Lift, virtual mass and other forces will
be considered later. g

NTEC 18
Slurry flow in horizontal pipe
2014 31
Uniform inlet Measurement plane
g 1m

V D

L=10m
Liquid velocity

Inlet Middle Outlet

Particle volume fraction

9
NTEC 19
Slurry flow horizontal pipe
2014 31
Uniform inlet Measurement plane
g 1m

V D

L=10m
d=270 µm, vf=0.2, d=480 µm, vf=0.203,
d=90 µm, vf=0.19,
D=51.5mm V=5.4 m/s D=51.5mm V=3.41 m/s
D=103mm, V=3 m/s

d=165 µm, vf=0.189, d=165 µm, vf=0.0918

D=51.5mm, V=4.17 m/s D=51.5mm V=3.78 m/s

d=165 µm, vf=0.273,

D=495mm V=3.46 m/s

NTEC 20
Particle suspension in stirred vessel
2014 31

Liquid Velocities
0.5
0.4 Uz Experimental

0.3 Uz Star-CCM+
0.2 Ur Experimental
0.1 Ur Star-CCM+
U/Utip

0 Utheta
-0.1 Experimental
-0.2
-0.3
-0.4
-0.5
0 0.2 0.4 0.6 0.8 1
r/R
Measurement
Solids Velocities
0.5
0.4
Uz Experimental
0.3
Uz Star-CCM+
0.2
0.1 Ur Experimental
U/Utip

0 Ur Star-CCM+
-0.1 Utheta
Experimental
-0.2
-0.3
-0.4
-0.5
0 0.2 0.4 0.6 0.8 1
r/R

10
NTEC 21
Gas dispersion in stirred vessel
2014 31

Gas superficial velocity:

U=0.0184 m/s U=0.0448 m/s U=0.0835 m/s U=0.1175 m/s U=0.201 m/s

NTEC 22
Comparison of gas holdup
2014 31

50
45
40
35
G as  holdup  [% ]

30
25 E xperimental
SSTAR-CCM+
tar-­‐C C M+
20
15
10
5
0
0 0.05 0.1 0.15 0.2 0.25
S uperfic ial   veloc ity   [m/s ]

11
NTEC 23
Adaptive MUSIG Model
2014 31

An Eulerian population balance method for poly-disperse multiphase flows.

NTEC 24
Adaptive MUSIG Model
2014 31

Mass and number density are redistributed between neighbour groups

so that each group has the same mass but new diameters.

1 2 3 4 5 6 7 8

1 2 3 4 5 6 7 8

• d

• d

12
NTEC 25
Droplet breakup through an orifice
2014 31

Sauter Mean Diameter

Breakup Rate (log scale)

Turbulent-induced breakup

Shear-induced breakup

NTEC 26
Boiling heat and mass transfers
2014 31

•  Conservation of energy for phase k:

⎡ ⎛ ⎞⎤
∂
(α k ρk hk ) + ∇.(α k ρk uk hk ) − ∇.⎢α k ⎜⎜ λk ∇Tk + µt ∇hk ⎟⎟⎥ = Qk
∂t ⎣ ⎝ σh ⎠⎦

•  Wall heat flux is modelled by three mechanisms:

q!Tʹ′ʹ′ = q!cʹ′ʹ′ + q! qʹ′ʹ′ + q!eʹ′ʹ′

Evaporation
Convective Quenching
heating

q!cʹ′ʹ′ q! qʹ′ʹ′ q!eʹ′ʹ′

13
NTEC 27
Bartolomei (1982) – 147 bar experiments
2014 31
•  D = 0.012 m
•  L = 2 m
•  P = 147 bar Water + steam
•  Tsat = 613 K
•  Q = 0.42 - 2.21 MW/m2
•  G = 1878 - 2012 kg/m2s
•  ΔTsub = 16 - 145 K
g

L
Wall heat
flux

Sub-cooled water Void fraction Condensation rate

NTEC 28
Bart 22-26 : Comparison of axial void profiles
2014 31

14
NTEC 29
Multi-component multiphase model
2014 31

•  General species transport equation for phase k is:

⎡ ⎛ ⎞ ⎤
∂
(α k ρkYk ) + ∇.(α k ρk ukYk ) − ∇.⎢α k ⎜⎜ Dk + µt ⎟⎟∇Yk ⎥ = Sk
∂t ⎣ ⎝ σY ⎠ ⎦

Y =mass fraction of species or other scalar quantity,

D =diffusion coefficient,
σ =Schmidt number,
S =sources.

NTEC 30
Oxygen transfer in aeration tank
2014 31

Volume fraction of air Oxygen in air

Water

Air injector

Oxygen in water Oxygen level at water surface and exit

15
NTEC 31
Summary
2014 31
•  VOF
–  Free surface flows
•  LMP
–  Droplet flows
•  DEM
–  Particle flows
•  EMP
–  Particle flows
–  Bubbly flows
–  Population balance
–  Boiling heat and mass transfers
–  Interphase mass transfer

16