• Tidak ada hasil yang ditemukan

Mesoscopic simulation of droplet dynamics in confinements and porous media flows

N/A
N/A
Protected

Academic year: 2023

Membagikan "Mesoscopic simulation of droplet dynamics in confinements and porous media flows"

Copied!
224
0
0

Teks penuh

Moreover, most studies available in the literature considered the two-dimensional domain for the analysis. LIST OF FIGURES xxi 5.15 The shape of the droplet iny−z plane in double grooves at x = 40.

Lattice Boltzmann method

In this field, the understanding of the fluid transport inside the reservoir is thus fundamental to find out the amount of unrecovered oil and the technical parameters to be used in the design of the extraction system. In this case, an important effect is given by the wetting conditions of the porous medium and also by other dimensionless parameters.

Figure 1.1: Different scale models, (a) Macroscopic scale, (b) Mesoscopic scale, (c) Microscopic scale
Figure 1.1: Different scale models, (a) Macroscopic scale, (b) Mesoscopic scale, (c) Microscopic scale

Comparison of LBM with N-S equations solver

Literature review

  • Review on lattice Boltzmann method
  • Review on droplet displacement
  • Review on droplet moving past a solid cylinder
  • Review on flow through porous media

It was also observed that the effect of viscosity ratio on the displacement of the drop also depends on capillary number. The surface features play a major role in the displacement or spread of the droplet on the channel wall.

Motivations and research objectives

The researchers have tried in the past to replicate this irregularity in terms of small grooves on the surface of the canal wall. A critical review of the literature on multiphase and multicomponent flows shows that the multiphase and multicomponent flow can be found extensively in porous media flows or through pores in a channel.

Thesis outline

The effect of low viscosity ratio on the dynamic behavior of the drop displacement on a wet wall of the three-dimensional channel is presented in Chapter 4. In Chapter 6, an attempt is made to remove the fluid that is trapped inside the. groove of the three-dimensional channel wall.

Introduction

Lattice structure

If the length of the domain is L and there are N lattice units along its entire length, the unit space is defined as δx=L/N. Therefore, the discrete time unit can be given as δt= δcx, where the denominator cis is the mesh speed.

Figure 2.1: D3Q19 lattice structure
Figure 2.1: D3Q19 lattice structure

Boltzmann transport equation

  • Background of LBM
  • Boltzmann transport equation
  • BGK model equation
  • Streaming and collision steps
  • Equilibrium distribution function

The collision operator (Ω) gives the rate of change between the final and initial states of the distribution function. The above equation gives that the total rate of change of the distribution function is equal to the collision velocity.

Figure 2.2: Position and velocity vector before and after applying force
Figure 2.2: Position and velocity vector before and after applying force

Boundary conditions in LBM

  • Periodic boundary condition
  • Bounce back boundary condition
  • Mirror/symmetry boundary condition
  • Free-slip boundary condition

The periodic boundary condition is the simplest condition in LBM used to isolate repetitive flow conditions. For this situation, the periodic boundary condition in terms of distribution functions are as follows.

Figure 2.5: Periodic boundary condition in LBM
Figure 2.5: Periodic boundary condition in LBM

LB models for multiphase flow

Shan and Chen model

In the interaction force, these aspects are modeled using the Green function. The molecules of the same substance interact with each other, then gkk 6= 0, and also with those of other substances gkk′ 6= 0.

Wettability on solid surfaces

Validation of LBM code

When a static droplet is formed, the contact angle is evaluated from the final steady-state values ​​of the droplet radius R, the droplet height a0, and the wetted droplet length b0, as shown in the figure. A contact angle greater than 90° is formed. when g2w is positive; indicating that phase 2 is non-wettable and the wettability state is characterized as hydrophobic.

Figure 2.10: Static droplet after steady state
Figure 2.10: Static droplet after steady state

Summary

Most researchers analyzed the coalescence of droplets without considering the wetting effect of the channel walls. The main objective of the present study is to investigate the dynamic behavior and initiation time of the coalescence of two droplets on the wettable channel wall.

Problem specification

The time taken by the two droplets for different parameters to merge is also investigated.

Results and discussion

Effect of centre distance on the coalescence

Looking at the wetted area and the wetted length, we see that both the wetted area and the wetted length initially increase up to the point where two droplets merge and after this point starts to decrease in the case of the hydrophilic surface. It is observed that the coalescence time of two drops is shorter in the mixed wettable surface compared to the uniform hydrophilic surface.

Effect of the capillary number on coalescence

The wetted area and wetted length increase gradually as the capillary number increases in the case of a hydrophilic surface, as shown in Fig. It is also observed that the increase in wetted area is less than the increase in wetted length.

Figure 3.3: The effect of center distance between the droplets (C d ) at Ca=0.35 on hydrophilic surface (a) wetted area, A/A 0 (b) wetted length, b/b 0
Figure 3.3: The effect of center distance between the droplets (C d ) at Ca=0.35 on hydrophilic surface (a) wetted area, A/A 0 (b) wetted length, b/b 0

Summary

Figures 3.6(a)-(b) show the wetted area and wetted length on the mixed wettable surface for three capillary numbers. It is also observed that the extent of the wetted region is larger in the uniform hydrophilic case compared to the mixed wettability case.

Figure 3.5: The effect of capillary number (Ca) at C d = 51 lu on hydrophilic surface (a) wetted area, A/A 0 (b) wetted length, b/b 0
Figure 3.5: The effect of capillary number (Ca) at C d = 51 lu on hydrophilic surface (a) wetted area, A/A 0 (b) wetted length, b/b 0

Introduction

Dynamic drop behavior on wetted flat and grooved surface of channel for low viscosity ratio.

Problem specification

Results and discussion

Effect of viscosity ratio on droplet displacement behaviour

It is found that the viscosity ratio initially does not greatly affect the displacement and deformation of the droplet. A clear picture related to the contact surface of the drop on the wall is in Fig.

Figure 4.3: The shape of the droplet in y − z plane at x = 40 lu for Ca = 0.35, g 2w = 0.05, lattice Time=8 and viscosity ratio (a) M = 0.7, (b) M = 0.8, (c) M = 0.9
Figure 4.3: The shape of the droplet in y − z plane at x = 40 lu for Ca = 0.35, g 2w = 0.05, lattice Time=8 and viscosity ratio (a) M = 0.7, (b) M = 0.8, (c) M = 0.9

Effect of the capillary number on droplet displacement behaviour 66

The height of the groove is a parameter that plays a key role in the movement of the droplet on the channel wall. The results are presented at the second time when the drop passes through the groove on the wall.

Summary

The large fraction of the drop liquid that adhered to the upper surface of the groove and remaining liquid moves down the lower surface of the channel wall. It is also found that after the break-up, a small amount of drop liquid in the form of drop sits on the bottom surface of the groove.

Results and discussion

A very small portion of the droplet liquid remains attached to the upper surface of the groove and spreads completely over the surface. When the droplet reaches the groove opening, in the case of the hydrophobic wetted groove as shown in Fig.

Figure 5.1: Schematic diagram of computational domain (a) single groove (b) double grooves
Figure 5.1: Schematic diagram of computational domain (a) single groove (b) double grooves

Summary

When the groove is located in the bottom of the channel, i.e. at 40-60 lu in z-direction as shown in Fig. In the case of two grooves on the wall, the trapped liquid is almost double that of the single groove case.

Figure 6.1: Schematic diagram of computational domain
Figure 6.1: Schematic diagram of computational domain

Problem specification

Results and discussion

Droplet dynamics on a grooved wall

The effect of this gravitational force can be seen in the change of shape of the droplet as well as the liquid trapped within the groove (Fig. At the same instant a weak velocity vortex is also generated within the previously trapped liquid on the lower surface of the groove.

Figure 6.4: The zoomed view of vector fields inside channel in y − z plane at x=40 lu for Ca = 0.25, groove height, H=70 lu at various lattice time
Figure 6.4: The zoomed view of vector fields inside channel in y − z plane at x=40 lu for Ca = 0.25, groove height, H=70 lu at various lattice time

Effect of wettability on the sweep out behaviour

It is important to mention here that both the quantities, i.e. wet area and wetting length, are calculated on the surface of the channel wall on which the groove is created. However, in the case of hydrophobic groove, as the droplet reaches the groove and passes through the groove surface, the wetted area and the wetted length of the channel wall decrease.

Figure 6.6: The shape of the droplet in y − z plane at x=40 lu for neutral (g 2w = 0.0) groove surface at Ca = 0.25, groove height, H=70 lu and at various lattice time
Figure 6.6: The shape of the droplet in y − z plane at x=40 lu for neutral (g 2w = 0.0) groove surface at Ca = 0.25, groove height, H=70 lu and at various lattice time

Effect of the capillary number and groove height

At later stages, when the coalesced droplet is translated to the channel wall below the groove, the droplet itself breaks into two parts and hence the wetted area and wetted length are found to decrease as shown in Fig.

Figure 6.8: The shape of the droplet in case of hydrophobically wetted groove in
Figure 6.8: The shape of the droplet in case of hydrophobically wetted groove in

Summary

On the other hand, the study of the point passing a solid target has recently been an active field of research for its great importance in many fields. The main objective of this study is to investigate numerically the dynamic behavior of the point moving under gravity in a three-dimensional channel.

Problem specification

But the study is lacking in the more detailed study in three-dimension; the study was only focused on the two-dimensional assumption. Therefore, the study also considers the direct impact of droplet and obstruction, as well as the distribution of droplet liquid over the solid obstruction.

Results and discussion

Effect of capillary number

The drop liquid film breaks at the upper surface of the obstruction due to the curvature of the obstruction as shown in Fig. However, a small portion of droplet fluid remains attached to the upper surface of the obstruction.

Figure 7.7: Effect of the capillary number on passing time and fraction of deposited volume on the upper surface of obstruction at Ca=0.407 and θ = 120 ◦
Figure 7.7: Effect of the capillary number on passing time and fraction of deposited volume on the upper surface of obstruction at Ca=0.407 and θ = 120 ◦

Effect of surface wettability on droplet dynamics

The investigation also shows that droplet liquid deposition occurs in the case of a hydrophobic wetting surface on the upper surface. It is interesting to see that the droplet liquid deposition on the top surface of the barrier is zero in the case of neutral wetting (θ = 0◦).

Figure 7.9: The y − z sliced view (at mid plane of x) of the time evolution of the droplet past a spherical obstruction at Ca=0.407, M =1 and θ = 60 ◦
Figure 7.9: The y − z sliced view (at mid plane of x) of the time evolution of the droplet past a spherical obstruction at Ca=0.407, M =1 and θ = 60 ◦

Effect of the size of the obstruction

As a result, the dissolution of the liquid film on the upper surface of the barrier occurs rapidly. Because of this, the large surface area allows the drop liquid to deposit more liquid on the top surface of the barrier in the case of r=60 lu (compared to tor=30 lu).

Figure 7.16: The isosurface plots of the time evolution of the droplet past a spherical obstruction at Ca=0.407, M =1 and θ = 120 ◦ for spherical obstruction radius, (a) r=30 lu, (b) r=50 lu
Figure 7.16: The isosurface plots of the time evolution of the droplet past a spherical obstruction at Ca=0.407, M =1 and θ = 120 ◦ for spherical obstruction radius, (a) r=30 lu, (b) r=50 lu

Effect of obstruction structure

In the case of hydrophobic wettability, θ = 120◦, the deposition of the droplet liquid is significantly increased for r=50 lu andr=60 lu, as shown in figure. However, it is observed that upon obstruction of the SC structure, the droplet fluid becomes trapped at the upper center of the obstruction due to the presence of a throat region, as shown in Fig.

Figure 7.20: The isosurface plots of the time evolution of the droplet past an ob- ob-struction at Ca=0.407, M =1 and θ = 120 ◦ for spherical obstruction structure, (a) SC, (b) BCC, and (c) FCC
Figure 7.20: The isosurface plots of the time evolution of the droplet past an ob- ob-struction at Ca=0.407, M =1 and θ = 120 ◦ for spherical obstruction structure, (a) SC, (b) BCC, and (c) FCC

Summary

For a better understanding of droplet liquid deposition on the barrier surface, the deposited volume fraction was measured and found to be equal to 15.50% for SC, BCC, and FCC structures, respectively. The porosity of a porous medium is defined as the fraction of the total volume of the medium that is occupied by void space.

Problem specification

It should be noted that all the solid walls are considered wetted and three different wetting conditions are taken for the study. The relaxation parameters for fluid 1 (WP) and fluid 2 (NWP) are considered equal to 1 and 1.42 giving a dynamic viscosity ratio approximately equal to 2.

Figure 8.1: Schematic diagram of immiscible two-fluid displacement experiment setup
Figure 8.1: Schematic diagram of immiscible two-fluid displacement experiment setup

Results and discussion

Effect of wettability on liquid transport

150 Mesoscopic pore-scale study of fluid flow through porous medium particles can be seen in Fig. Since the pore surface is non-wet, it does not allow the liquid to spread over it and block the path of further penetrating liquid.

Figure 8.3: The structure of the pore created between spherical particles
Figure 8.3: The structure of the pore created between spherical particles

Effect of porosity on liquid transport

Effect of pressure gradient on liquid transport

Summary

The validation of this code was found to be in good agreement with the available literature. The present thesis presents a particularly detailed insight into the interactions of capillarity and wettability on the movement and spreading of droplets on channel walls and the flow of an immiscible liquid through porous media with a ball pack.

Figure 8.12: Effect of pressure gradient on the liquid transport for hydrophobic wettability, g 2w =0.05
Figure 8.12: Effect of pressure gradient on the liquid transport for hydrophobic wettability, g 2w =0.05

Scope of future work

D3Q19 lattice structure

Position and velocity vector before and after applying force

Pre-streaming and post-streaming conditions in LBM

Pre-collision and post-collision conditions in LBM

Periodic boundary condition in LBM

Bounce back boundary condition in LBM

Mirror boundary condition in LBM

Free-slip boundary condition in LBM schematic

Contact angle of a liquid droplet wetted to a solid surface

Static droplet after steady state

Variation of contact angle with g 2w

Gambar

Figure 2.2: Position and velocity vector before and after applying force
Figure 3.3: The effect of center distance between the droplets (C d ) at Ca=0.35 on hydrophilic surface (a) wetted area, A/A 0 (b) wetted length, b/b 0
Figure 3.4: The effect of center distance between the droplets (C d ) at Ca=0.35 on mixed wettable surface (a) wetted area, A/A (b) wetted length, b/b
Figure 3.5: The effect of capillary number (Ca) at C d = 51 lu on hydrophilic surface (a) wetted area, A/A 0 (b) wetted length, b/b 0
+7

Referensi

Dokumen terkait

co w Sf Velocity slip factor T∞ Free stream fluid temperature Sg Rotational slip factor u fluid velocity in the x direction Sh Sherwood number U0 Characteristic fluid velocity Shx