CFD-DEM Simulation of Flocculation

and Sedimentation of Cohesive Fine

Particles

Liu-chao Qiu, Jia-jie Liu, Yi Liu, Peng-zhi Lin and Yu Han

Abstract A numerical model coupling the discrete element method (DEM) and the
computational fluid dynamics (CFD) was presented and applied to simulate particle
flocculation and sedimentation processes. The volume-averaged continuity and
momentum equations are solved for fluid motions, while the particle motion is
modeled by the DEM with considering the inter-particle interactions based on
mechanics of contact. The van der Waals force induced cohesive attractions
between particles were taken into account in our study. The predicated results
demonstrated that the proposed method is capable of simulating the particle sedi-
mentation and flocculation process.

1 Introduction

It is important to remove the fine particles from liquid suspensions due to the wide
applications in natural and industrial processes. Flocculation is often applied to let
the particles approach enough to collide. A collision causes the formation of an
aggregate when added flocculants suppress inter-particle repulsive forces and the
inertia of approaching particles overcomes the viscous resistance of fluids. The
flocculation process improves the agglomeration of particles to form aggregates
with irregular shape that can be filtered or settled. These aggregates are then
removed in downstream separation processes. It is more important today than in the

L. Qiu (&)  J. Liu  Y. Han

College of Water Resources and Civil Engineering, China Agriculture University, Beijing
100083, China
e-mail: [email protected]
Y. Liu
State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China
Institute of Water Resources and Hydropower Research, Beijing 100038, China
P. Lin
State Key Laboratory of Hydraulics and Mountain River Engineering, Sichuan University,
Chengdu 610065, China

© Springer Science+Business Media Singapore 2017 537

X. Li et al. (eds.), Proceedings of the 7th International Conference
on Discrete Element Methods, Springer Proceedings in Physics 188,
DOI 10.1007/978-981-10-1926-5_55

[email protected]
538 L. Qiu et al.

past to understand and design/optimise the flocculation processes due to the

requirements for removing particulates have become increasingly urgent [1]. The
main difficulty in simulating flocculation and sedimentation can be attributed to
their complex flow behaviour. The flows with fine particles are different greatly
from the larger and non-adhesive particulate flows. For example, the attractive
forces between colliding particles with micron or sub-micron size are larger than the
corresponding inertial forces, which bring fine particles to close to adhere upon
collision and form particle agglomerates by hydrodynamic or Brownian motion. In
this study, we present a numerical model by coupling DEM and CFD to take into
account the inter-particle contacts and the interactions between particles and fluids.
The DEM by Cundall and Strack [2] is a Lagrangian method for modelling granular
mechanics and now widely used in flows with particles. The van der Waals force
induced cohesive attractions between particles were taken into account in our study.

2 Mathematical Formulation

The translational and rotational motion of particle i is governed by:

dvi
mi ¼ f ci þ f fpi þ mi g ð1Þ
dt
dxi
Ii ¼ Ti ð2Þ
dt

in which m is the mass and I is the inertia moment. v and ω are the linear and
angular velocity, T is the torque, g is the gravity acceleration, fci is the inter-particle
contact force, the fluid-particle interaction force ffpi is given by [3]:

f fpi ¼ vpi r  p þ vpi r  sf þ ef di ð3Þ

where vpi is the particle volume, e is the void fraction, p is the fluid pressure, sf is
viscous stress, and f di is the drag force and calculated using Di Felice [4]
formulation.
The governed equations for fluids are the followed averaged continuity and
momentum equations [3]:

@ðeqf Þ
þ r  ðeqf uÞ ¼ 0 ð4Þ
@t
@ðeqf uÞ
þ r  ðeqf uuÞ ¼ r  p þ r  sf  Ffp þ eqf g ð5Þ
@t

[email protected]
CFD-DEM Simulation of Flocculation and Sedimentation … 539

where u is velocity and qf is the density. Ffp is given by

Pn c
i¼1 f fpi
Ffp ¼ ð6Þ
DVc

in which nc is the particle numbers in a fluid element with a volume of DVc .

The Johnson-Kendall-Roberts (JKR) Cohesion model [5] was used to calculate
the cohesion force between two spherical particles due to Van der Waals force.
Cohesion force between two spherical particles is expressed as:

3
Fco ¼ pRW ð7Þ
2

where R is the equivalent radius, W is the work of cohesion (J/m2).

3 Results and Discussion

In this study, the particulate flow in a 2D lid-driven square cavity is simulated. The
problem geometry shown in Fig. 1 consists of a two-dimensional L × L cavity,
where L = 1.2 mm. In this study, two types of flow state are simulated. In simu-
lation of flow at rest all walls are fixed with a no slip wall boundary condition. For
moving case, the top wall moves with a constant velocity U = 1.2 mm/s.
Gravitation is considered with an acceleration of 9.81 m/s2. The square cavity is
fully filled with water and 500 particles are randomly distributed in the water,
among which 250 particles have the diameter of 0.04 mm and the other 250 particle
have the diameter of 0.02 mm. The rest calculation conditions are given in Table 1.
The simulated particle configurations in static flow at different time are shown in
Fig. 2. In which, the sedimentation and flocculation states of particle near the

Fig. 1 The problem

geometry

[email protected]
540 L. Qiu et al.

Table 1 Calculation Particle

conditions
Density 1500 kg/m3
Elastic modulus 1 × 108 Pa
Poisson ratio 0.33
Friction coefficient 0.3
Work of cohesion 0.01 J/m2
Water
Density 1000 kg/m3
Dynamic viscosity 0.001 Pa.s
Solution
Time step 0.0001 s
Computational cell sizes 0.12 mm

Fig. 2 The states of particle in static flow at time = 0.0 s (a), 0.2 s (b), 0.6 s (c), 1.0 s (d), 1.4 s
(e), and 2.0 s (f)

bottom are presented. Particle collisions occur in settling process and aggregates
with irregular shape are formed. The aggregates fall faster than single particles due
to the aggregates obtained larger settling velocity than a single particle. Figure 3
show calculated transient behavior of the particles in a lid-driven cavity flow. The
motions of particles are caused by the flow of the fluid phase. The calculated result
indicates that more aggregates are formed in lid-driven cavity flow rather than in

[email protected]
CFD-DEM Simulation of Flocculation and Sedimentation … 541

Fig. 3 The states of particle in lid-driven cavity flows at time = 0.0 s (a), 0.2 s (b), 0.6 s (c), 1.0 s
(d), 1.4 s (e), and 2.0 s (f)

static flows. The particle aggregates arrive early at the bottom and the number of
floating particles in lid-driven cavity flow has become less than that in static flow
because of higher settling velocity due to particle agglomeration. The flow driven
particle movements becomes large and the collision opportunity of particles
increase in the case of lid-driven cavity flow.

4 Conclusion

A two-way coupled numerical model was applied to simulate the flocculation and
sedimentation of cohesive fine particles. In the proposed method, the particle
motions are calculated by using DEM while the fluid motions are solved by using
CFD. The flocculation and Sedimentation process of cohesive fine particle in static
flow and lid-driven cavity flow were simulated. In this study, the JKR model was
applied to model the van der Waals force induced cohesive forces acting on par-
ticles. Numerical results show that particle collision, flocculation and sedimentation
can be well reproduced. Comparison of the results show the number of particle
aggregates in lid-driven cavity flow is larger than that in static flow.

[email protected]
542 L. Qiu et al.

Acknowledgments This research work is supported, in part, by the National Natural Science
Foundation of China (Grant No. 11172321), Open Fund of the State Key Laboratory of
Hydroscience and Engineering of China, Tsinghua University (Grant No. sklhse-2015-C-03),
Open Fund from the State Key Laboratory of Simulation and Regulation of Water Cycle in River
Basin, China Institute of Water Resources and Hydropower Research (Grant
No. IWHR-SKL-201505), and Open Fund from the State Key Laboratory of Hydraulics and
Mountain River Engineering, Sichuan University (SKHL1425).

References

1. Bratby, J.: Coagulation and Flocculation in Water and Wastewater Treatment. IWA Publishing,
London (2006)
2. Cundall, P., Strack, O.: A discrete numerical model for granular assemblies. Gotechnique 29,
47–65 (1979)
3. Anderson, T.B., Jackson, R.: A fluid mechanical description of fluidized beds: equations of
motion. Ind. Eng. Chem. Fundam. 6, 527–539 (1967)
4. Di Felice, R.: The voidage function for fluid-particle interaction systems. Int. J. Multiph. Flow
20(1), 153–159 (1994)
5. Johnson, K.L.: Contact Mechanics. Cambridge University Press, Cambridge (1987)

[email protected]