Euler-Lagrangian Simulations of Multiphase Plumes for Environmental Engineering

Ruo-Qian Wang and E Eric AdamsDepartment of Civil and Environmental Engineering, Massachusetts Institute of Technology

The present project is motivated by several issues associated with

the dumping process, and in particular the initial plume formation

phase. Through numerical simulations, we are exploring:

1) Buoyancy effects on rate of penetration and flow patterns with

plumes.

2) Distribution of the dispersed phase within the plume

3) Development of improved models describing sediment plume

behavior to guide engineering practices.

4) Providing a bridge between analytical models and experimental

results.

Fig.1 Multi-phase plumes from oil seepage and fall out (http://convergence.ucsb.edu/news/ucsb-scientists-document-fate-huge-oil-slicks-seeps-coal-oil-point)

The present study is aiming to track sediment fate in the operation of land reclamation to

minimize sand loss through suspension in the water column, i.e. turbidity, during the dumping

process. It is critical to conserve the sediment resources as well as ensure the accurate placement

of the dumped material. The reduction of turbidity will also avoid excessive environmental

impact to the surrounding ambient waters.

Multiphase buoyant intrusions are common in

natural and engineered systems and span

multiple time and length scales. Typically, the

plumes contain solid particles (e.g. volcanic

eruptions, sediment disposal in open waters),

liquid droplets (e.g. piston engine fuel

injection), bubbles (e.g. lake aeration), or their

mixtures (e.g. oil spill in oceans).

Fig.2 Sketch of dredged material dumping in open waters

I. Motivation

II. Numerical Methods

D

28.2D

30D

Free slipInflow

outflow

Grids and boundary conditions

Orifice

Governing equations

Continuous phase: Large-Eddy Simulations with the sub-grid model of Dynamic Mixed

Model, built-in filter functions, and inter-phase coupling terms.

Dispersive phase:

- Lagrangian Tracking:

- FG=gravity and buoyancy force; FS=fluid stress; FD=drag; FL=lift; FA =added mass

- Empirical coefficients:

Drag coefficient:

Lift coefficient: CL=0.05 Virtual mass Coefficient: CA=0.5

- Inter-phase coupling:

ALDSG

pFFFFF

u

dt

dmp

DALcoupling FFFF

Computing environment

The code is developed based on an open-source Computational Fluid Dynamics software

OpenFOAM

Parallel calculations are made at the High Performance Computing cluster in Nanyang

Technological University, Singapore

III. Steady State PlumesNozzle diameter

D (mm)Particle diameter

Dp (mm)Inflow Velocity

U0 (m/s)Volume Fraction

a0

5.08 500 1.66 2.4%

Centerline Characteristics

Cross-sectional Characteristics

IV. Particle Clouds

V. Conclusions and Future Plans

Comparison between simulations and experiments along the centerline

(left: volume fraction; middle: particle velocity; right: fluid velocity)*

t= 1.0 s 2.0 s 3.0 s 4.0 s

Comparison between simulations and experiments of particle cloud patterns

(upper: numerical simulations; lower: experimental photos)**

The research described in this project was funded in whole or in part by the Singapore National Research Foundation

(NRF) through the Singapore-MIT Alliance for Research and Technology (SMART) Center for Environmental

Sensing and Modeling (CENSAM).

Acknowledgement

References[1] R Parthasarathy and G Faeth, “Structure of particle-laden turbulent water jets in still water”, International Journal of Multiphase

Flow 13, 699-716 (1987).

[2] Zhao, B., Law, A. W. K., Adams, E. E., Shao, D. D., & Huang, Z. H. (2011). Effect of air release height on the formation of

sediment thermal in water. Proceedings, 34th IAHR World Congress, Brisbane, Australia.

-The present numerical method can convincingly reproduce dilute particle-laden jets.

-The reproduced dense particle cloud is comparable to the experiment.

-CFD-DEM will be used to incorporate the particle-particle collision and other sophisticated

models in order to achieve a more accurate result.

* PF refers to reference [1]

ac/a

o

wp

c/w

o

wc/w

o

z/D z/D z/D

r/z r/z r/z

a/a

c

wp/w

c

w/w

c

He

igh

t o

f th

e c

en

ter

(cm

)

Ve

locity o

f th

e c

en

ter

(cm

/s)

Ra

diu

s (

cm

)

Time (s)

Comparison between simulations and experiments of particle cloud evolution

(left: height of the cloud center; middle: settling velocity of the cloud center; right: horizontal

extent of the cloud)**

Time (s) Time (s)

** The “exp” here refers to reference [2]

Comparison between simulations and experiments along cross-sections

(left: volume fraction; middle: particle velocity; right: fluid velocity)*