The numerical analysis of complete and partial electrocoalescence in the droplet-layer system

employing the sharp interface technique for multiphase-medium simulation

d
Grigorii Yagodin*, Ilia Elagin, Sergei Vasilkov, Vladimir Chirkov

we
St. Petersburg State University, St. Petersburg, 199034, Russia

Abstract. In this paper, the coalescence of a drop of water suspended in oil with a layer of water under the action of a constant electric
field is numerically investigated. Unlike most existing studies, the calculations are based on the application of the arbitrary Lagrangian-
Eulerian method (ALEM), also called the moving mesh method, which belongs to the class of methods for modeling two-phase liquids with
a sharp interface. Using this approach made it possible to avoid a false "escape" of the surface charge from the interface, which often occurs
when using methods with a diffuse interface. Despite the fact that ALEM does not allow describing topology changes by default, a computer
model was implemented in which the calculation is divided into three parts—the convergence of the drop and the layer before the moment of

ie
touch, the manual construction of the bridge at the moment of touch, the union of the drop and the layer. The developed model allowed us to
obtain three possible modes of this process: complete coalescence, partial coalescence and a previously practically not considered mode of
stretching. The dependence of the volume of the detached secondary droplet on the size of the initial droplet and the average intensity of the
applied electric field is obtained. The created model showed good quantitative agreement with experimental studies. It has been shown that

ev
the place of formation of the bridge and the neck in the case of partial coalescence, as a rule, do not coincide. A map of coalescence modes
was obtained, i.e., the dependence of the transition threshold from coalescence to partial coalescence and from partial coalescence to stretching
regime in a wide range of radius of initial droplets and electric field strengths. It has been shown that there is a maximum field strength at
which droplets of any size combine with the layer. This map makes it possible to predict the coalescence regime in electrocoalescer. The
proposed modeling technique can be used to calculate electrocoalescence modes at various values of the main parameters, which will help
optimize the electrocoalescer at the design stage.

rr
Keywords

Electro-coalescence, numerical simulation, partial coalescence, droplet-layer, water-in-oil emulsion, two-phase liquid.

Introduction
ee
When extracting crude oil from a well, methods are used related to adding a large amount of water to the well to
increase the pressure and, as a result, accelerate the pumping of oil. This approach leads to the fact that water disperses
into the smallest droplets in oil and forms a water in oil emulsion [1]. During further processing of oil, it is necessary
to reduce the mass fraction of water. By virtue of Stokes' law, the droplet deposition rate is proportional to 𝑅2, which
p
leads to the fact that the dehydration process by settling the emulsion takes quite a long time. To accelerate this process,
there are several methods: chemical treatment, gravitational and centrifugal separation, pH regulation, heat treatment,
membrane separation and electrocoalescence [2]. The most cost-effective, straightforward and environmentally
friendly method is electrocoalescence.
ot

The electrocoalescence method involves applying strong electric fields to an emulsion, where the dispersed phase is
an electrically conductive liquid. This process causes smaller droplets to merge into larger ones, allowing them to
separate from the dispersion medium (oil) in a relatively short time under the influence of gravity. Devices that use
tn

this technology are known as electrocoalescers.

In all electrocoalescers, the separation process takes place in two stages. In the first stage, water droplets dispersed in
oil are combined under the influence of electrostatic forces. As a result, micrometer-sized droplets merge into larger
ones and start moving downward due to gravity. In the second stage, these larger droplets merge with the water layer
and are removed from the system. To increase the operational volume of the first stage, electrocoalescers use a
rin

horizontal electrode system. The water layer formed at the bottom of the electrocoalescer is grounded, acting as a
collector for the dispersed phase. This setup creates an electric field between the lowest electrode and the surface of
the water layer.

In this configuration, the electric field in the area between the electrode and the layer has both positive and negative
ep

effects. The positive effects include the accelerated movement of droplets towards the layer due to electrophoretic
forces (since droplets in an emulsion usually have a non-zero charge). Additionally, the electric field significantly
reduces the resting time of the droplet on the layer's surface [3-5], even when relatively low voltages are applied.
Pr

*corresponding author at, St. Petersburg State University, St. Petersburg, 199034, Russia

E-mail adress: [email protected] (G. Yagodin)

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 However, when fields are excessively strong, negative effects start to emerge, such as the formation of droplet chains,
partial coalescence, cone-shaped droplet spraying and stretching of a droplet along the interelectrode gap. The use of

d
a pulsating field [6] or a high-frequency alternating field [7] helps to mitigate droplet chain formation by employing
specific optimal parameters, which can vary with changes in the liquid properties. Droplet chain formation is the
primary reason why alternating and pulsating fields are commonly used in electrocoalescers. Nonetheless, direct

we
current (DC) fields can be effective for purifying oil with a low percentage of dispersed phase.

During the coalescence of a drop with a layer, part of the drop may separate, creating a smaller daughter drop. The
new drop has a charge of the same sign as the layer, so after separation, it moves toward the upper electrode. This
mode of interaction between the drop and the layer is known as partial coalescence. Notably, partial coalescence can
occur in systems with or without electric fields. This regime in the absence of electric fields was first discovered and

ie
attempted to be explained in 1960 [8]. However, due to the system's complexity, a satisfactory explanation has yet to
be found.

In hydrodynamic partial coalescence, only surface tension forces are involved. Thus, partial coalescence can occur

ev
only if capillary waves form on the droplet surface [9]. Capillary waves are generated due to the rapid expansion of
the bridge, causing the droplet to stretch vertically while accelerating its horizontal collapse. This dynamic slows down
the transfer of liquid from the drop to the layer, leading to the formation of a neck and, eventually, the separation of
the droplet from the layer. Blanchette and Bigioni [9] noted that surface capillary waves occur only in systems with
low viscosities; otherwise, the waves damp quickly, preventing this effect from occurring.

rr
The phenomenon of partial coalescence in the presence of an electric field was first investigated in 1961 [3]. However,
this study focused on accelerating the thinning process using electric fields with relatively small applied strengths,
resulting in no significant change in hydrodynamic partial coalescence. Another study [5] explored the effect of
electric fields on the volume of the separated droplet, demonstrating that the coalescence time decreases with
ee
increasing electric field strength, irrespective of electrode polarity. Aryafar and Kavehpour [10] revealed that the
critical electric field strength (i.e., the strength at which the transition from complete to partial coalescence occurs)
depends on surface tension, droplet radius, and the dielectric constant. They also observed that, with a substantial
increase in electric field strength, a jet of liquid is formed from a drop that is sprayed at the tip. This type of
electrocoalescence regime has also been reported in other studies [11-13].
p

Explaining the partial coalescence regime as electric field tension increases is challenging. However, according to
Pillai et al. [11], the separation of a daughter droplet involves three factors: Coulomb forces that pull the droplet away
from the layer, surface tension that resists this movement, and the liquid transfer process from the drop to the layer.
ot

The ratio of Coulomb forces to surface tension largely determines whether partial coalescence occurs.
Mousavichoubeh et al. [14] investigated how the volume of the separated droplet depends on the applied electric field,
the initial distance from the layer, and the size of the initial droplet. These studies showed that an increase in any of
these parameters leads to a larger volume of the separated droplet. In another study [15], it was shown that surfactants
tn

can greatly influence the volume of the separated droplet. By altering interfacial tension, they concluded that the
process of liquid transfer from the drop to the layer and the rate of neck formation are key factors in determining the
volume of the daughter droplet, leading to the adoption of a dimensionless number WO=WE×Oh. It was shown that
the volume of the separated droplet, depending on WO, can be fit to a single curve across various system parameters.
rin

An explanation for the independence of the size and charge of the separated droplet from its conductivity was proposed
by Sunder and Tomar [16]. Their numerical modeling showed that the total charge on the curved surface of the layer
remains constant despite changes in conductivity; only the time needed for the charge to stabilize varies. When the
dispersed medium has low conductivity, convection becomes the dominant charge transfer mechanism, reducing the
influence of electrical conductivity. Additionally, variable fields can cause a drop bounce effect [17]. Other factors,
ep

like the presence of ions in water, can also influence partial coalescence. For example, Sun et al. [18] showed that
increasing the concentration of alkali (𝑁𝑎2𝐶𝑂3) raises the threshold for the transition to partial coalescence. Finally,
Yang et al. [19] demonstrated that there is a directly proportional relationship between 𝐸𝑐𝑟𝑖𝑡/ 𝑅 and interfacial
tension.

Due to the high complexity of analyzing this system, numerical methods based on finite element techniques have been
Pr

increasingly used in recent years. Numerical models can be used to obtain data on almost any system, but they present
a significant issue: these models are difficult to verify due to the complexity of both the system under study and the
methods used in these models. Currently, several primary approaches are used to model these tasks. The phase field

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 method and the level set method are predominantly used [20,21]. These methods involve a diffuse interface
description, which poses several problems.

Firstly, the diffuse boundary does not allow us to accurately determine the direction of the normal to this boundary,

d
making it challenging to accurately calculate curvature and, consequently, determine the forces acting on the boundary
[22]. Secondly, without additional modifications, this method can lead to the “escape” of charge from the interface,
as demonstrated in results presented by Li et al. [23]. Additionally, many articles use models with a diffuse interface,

we
which inevitably face the problem of “charge escape,” when describing electrocoalescence in drop-drop systems [24-
29] and drop-layer systems [12,30-33].

A recent approach to address this issue involves replacing the linear relationship between electrical conductivity and
the liquid fraction value with an exponential one [34]. However, such studies are currently limited and mainly pertain
to electrospray and electrohydrodynamic printing [35-37]. Another approach to mitigate these issues is molecular

ie
dynamics modeling [38,39]. However, these models are very resource-intensive and do not allow modeling macro-
sized droplets. As a result, only extremely small system sizes are considered, which significantly complicates the
practical application of the obtained results. For example, in Li et al. [39], a coalescence map is presented for droplets

ev
with a radius of 22–30 nm, whereas macro-sized droplets are more relevant for drop-layer studies. By contrast, the
experimental work by Anand et al. [30] presents a coalescence map for droplets with a radius of 0.3–0.7 mm.

Due to the problems described above, there was a need for an alternative approach to numerical modeling of the
interaction between a drop and a layer at macroscopic scales. A suitable alternative is the arbitrary Lagrangian-
Eulerian method (ALEM), also known as the moving mesh method [40,41], which is a sharp interface approach. This

rr
method is frequently used for numerical modeling of droplet electrodeformation processes and their motion [42-47].
However, applying ALEM to calculate electrocoalescence has been challenging because the default approach does
not allow for describing topology changes. Only recently has an approach been developed and used that allows for
modeling the merging of droplets with the moving mesh method [48-53]. ALEM shows great promise for the
ee
numerical analysis of complete and partial coalescence processes in drop-layer systems.

This work is aimed at:

- implementation of a numerical model of the interaction of a single drop of water suspended in oil with a layer of
water based on ALEM, i.e., a method for describing a two-phase liquid with a sharp interface;
p

- analysis of the outcomes of complete and partial coalescence, as well as cases of droplet stretching along the entire
interelectrode space followed by its spraying.
ot

The main outcome of this study is the development of a coalescence regime map for the droplet-layer system,
illustrating what occurs when individual droplets of varying radius contact a water layer at different electric field
strengths. Specifically, the map delineates the transitions from complete coalescence to partial coalescence, and from
partial coalescence to non-coalescence. This research provides a significant extension and clarification of previously
tn

published findings on the topic of electrocoalescence of water droplets with a layer.

Numerical model
rin

The numerical model is based on a set of Navier-Stokes equations (1)-(2), Poisson equation (3), and the arbitrary
Lagrangian-Eulerian method (moving mesh method) for describing two-phase liquids [51,52]. The following
assumptions were made in the numerical solution of the problem: the Navier-Stokes equations were calculated for an
incompressible isothermal liquid, and the electric field distribution was calculated for a case with equipotential water
volumes (i.e., under the assumption of an ideally electrically conductive liquid), which is quite reasonable due to the
ep

high conductivity of water compared to oil. The initial conditions consider an uncharged droplet. The coupling
between hydrodynamics and electrostatics is achieved through boundary conditions at the interface between the
phases, applying electrostatic pressure corresponding to the Coulomb force effect on the surface charge induced on a
conductive droplet's surface. The high-voltage electrode is located at the top boundary of the model, and zero potential
is set at the surface of the water layer. The boundary conditions are illustrated in Figure 1 and listed in Table 1.
Pr

∂𝑣
𝜌 + 𝜌(𝑣∇)𝑣 = ― ∇𝑝 + ∇( (∇𝑣 + ∇𝑣𝑇)𝜇 ) + 𝜌𝑔
∂𝑡
𝑑𝑖𝑣 𝑣 = 0

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 𝜌𝑒
𝑑𝑖𝑣 𝐸 = 𝜀𝜀0

𝐸 = ― ∇𝜑

d
Here 𝜌 is the liquid density, 𝑣 is the velocity vector, t is the time, 𝑝 is the pressure, 𝑔 is the free-fall acceleration,𝜀 is
the medium electrical permittivity, 𝐸 is the electric field strength, 𝜑 is the electric field potential, and μ is the dynamic

we
viscosity.

ie
ev
rr
ee
Fig. 1 Geometry and boundary conditions.
Table 1. Boundary conditions.
Boundary condition Numeration Mathematical form Symbols meaning
Electrostatics
p
Axial symmetry, zero
1 𝐸∙𝑛=0 𝑛 is the normal to the surface
charge
HV electrode potential 2 𝜑 = 𝑐𝑜𝑛𝑠𝑡
Floating potential 3 𝜑 = 𝑐𝑜𝑛𝑠𝑡, (𝐸 ∙ 𝑛)𝑑𝑆 = 𝑄 = 0 𝛺 is the integration surface
∂𝛺
ot

LV electrode potential 4 𝜑=0

Hydrodynamics
Axial symmetry for
5 𝑣∙𝑛=0
velocity
No slip 6 𝑣=0
𝜇
tn

𝐾𝑛 ― (𝐾𝑛 ∙ 𝑛)𝑛 = ― (𝑣 ― (𝑣 ∙ 𝑛)𝑛)

Navier slip 7 𝛽
𝐾𝑛 = 𝜇(∇𝑣 + (∇𝑣)𝑇)𝑛
𝑛 ∙ (𝑝1 ― 𝑝2) = 𝑓𝑠𝑡 ― 𝑝𝑒
𝑓𝑠𝑡 = 2𝛾𝐻𝑚 ― ∇𝑡 ∙ 𝛾 𝑝1,𝑝2 are pressure inside and outside the droplet, 𝑝𝑒
1
Fluid-fluid interface 8 𝐻𝑚 = (∇𝑡 ∙ 𝑛)𝑛 is the electrostatic pressure, γ is the interfacial
2
rin

1 tension, 𝐻𝑚 is the mean curvature of the surface

𝑝𝑒 = 𝜀𝜀0𝐸2𝑛
2
Moving mesh
Fixed boundary 9 𝑣𝑚𝑒𝑠ℎ = 0 𝑣𝑚𝑒𝑠ℎ is the velocity of mesh nodes
Symmetry 10 𝑣𝑚𝑒𝑠ℎ ∙ 𝑛 = 0
ep

The arbitrary Lagrangian-Eulerian method (ALEM) tracks the position of the boundary by moving the geometric line
that separates the two phases, and thus the finite element mesh [40,45,54,55]. In this method, the mesh elements are
shifted at each iteration step along with the geometric line. As boundary conditions on the interface surfaces, the
equality of velocities between both phases is set, ensuring their synchronized movement. Since ALEM does not allow
for automatic description of topology changes, the solution to this problem was divided into three steps [51-56].
Pr

First, the approximation of the droplet to the layer is simulated. Next, the calculation process is paused, and a bridge
is created between the droplet and the layer when the distance between them becomes sufficiently small. Afterward,
the computation is resumed, running either until the droplet and the layer merge (indicating complete coalescence) or

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 until a narrow neck forms, at which point the droplet separation is expected (indicating partial coalescence). This
process is illustrated in the next section.

The primary parameter affecting the accuracy of solving this problem is the size of the finite element mesh and the

d
way it is stretched when the interface is shifted. For mesh smoothing, the Yeoh smoothing method [57] was used, as
described by Utiugov et al. [51]. The optimal mesh is achieved by splitting the droplet boundary (radius 1 mm) into
200 elements, while the boundary of the water layer is divided into 500 elements (with an element ratio of 20). Further

we
details about mesh distribution and other technical aspects can be found in [56].

The simulation was carried out for olive oil with the following characteristics: density ρo = 910 kg/m3, dynamic
viscosity μo = 0.065 Pa * s, relative permittivity 𝜀 = 2.85. The properties of the drop and layer correspond to water: ρw
= 1000 kg/m3, μw = 0.001 Pa * s. The interfacial tension (IT) for this pair of liquids is set to 𝛾 = 0.016 N/m. However,
in reality, the IT for olive oil (as well as for many other seed oils) depends on the contact time of the two phases [58-

ie
60], which may be due to the presence of amphiphilic, interfacially active compounds in these oils [61]. Therefore, to
calculate each realistic case, it is necessary to use the value of the IT corresponding to the contact time of the two
phases.

ev
Results and discussion
Base results

rr
Figure 2 illustrates the results of the partial coalescence process, divided into three steps as described in the previous
section. A water droplet is inserted into the oil at a distance of 1 mm from the surface of the water layer, which is
situated at the bottom of the model and has a height of 4 mm. If the droplet is far enough from the layer at the initial
time, it deforms to a stable state while approaching the layer. Figure 2 (subplots 1 and 2) shows the transition from its
ee
initial spherical shape to a deformed shape. As the droplet nears the layer's boundary, the Coulomb force significantly
increases, causing the droplet to deform more and become sharper, while the surface of the layer slightly rises toward
the droplet (Fig. 2, subplots 3 and 4). Sometimes this leads to the formation of a small dimple on the upper part of the
layer (Fig. 2, subplot 5), similar to the "cone-dimple" coalescence regime observed in drop-drop systems [62,63].
When the droplet gets to within 1/50 of its radius from the layer, the calculation stops, and a water bridge is built
p

between the droplet and the layer, with a length-to-radius ratio of 1:1 (Fig. 2, subplot 6). The rationale for choosing
this bridge size is partly based on the results from [51]. Afterward, the calculations resume, and the bridge expands
under the influence of surface tension. Depending on the ratio of Coulomb forces to surface tension (i.e., the intensity
of the electric field, assuming other parameters remain constant), the droplet and the layer either completely or partially
ot

merge.
tn
rin
ep
Pr

Fig. 2 Example of performed process R = 1 mm, E = 1.6 kV.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 It is important to analyze the impact of the initial distance between the droplet and the layer when the voltage is turned
on. In most experimental studies [15,30,64,65], this parameter is not explicitly specified. However, Mousavichoubeh
et al. [14] showed that the size of the separating droplet increases with increasing initial distance from 0.1 to 0.2 mm.
In numerical modeling studies, the initial distance is generally indicated, but it typically ranges from 0.1 to 0.2 mm

d
[23,32,33], except for the study by Sunder and Tomar [16], where it is set to 4 mm. None of these studies provide a
clear explanation for this dependence or offer guidance on what distance between the droplet and the layer should be

we
set such that further increases in distance do not affect the outcome of partial coalescence.

To address this issue, a study was conducted to examine the impact of the initial distance H between the droplet and
the water layer on several physical parameters: deformation, the average velocity at the moment of contact, and the
relative volume of the daughter droplet. Figure 3(a) depicts the relationship between relative volume and H. All these
parameters related to coalescence outcomes increase as H rises from 0 to approximately 0.5 mm (for the selected

ie
droplet radius and average electric field strength). Beyond this range, the values of these parameters reach a plateau.
For example, for a droplet with a radius of 1 mm and a field strength of 2 kV/cm, Figure 3 suggests that an initial
distance of about 0.6 mm is sufficient to ensure that further increases in distance do not affect the coalescence outcome.

ev
rr
p ee
Fig. 3 (a) Volume of the detached droplet and (b) z - components of velocity and deformation as a function of initial distance from the drop to the
layer.

The developed numerical model reproduces all outcomes previously observed in experiments. Figure 4 illustrates how
ot

the outcome of the interaction between a 1 mm droplet and a water layer changes as the average electric field strength
increases, ranging from 1 to 3.6 kV/cm. As the voltage increases, complete coalescence, which occurs at 1 kV/cm,
transitions to partial coalescence, with the volume of the separated droplet increasing as the voltage rises. With further
increases in voltage, the droplet stretches along the interelectrode gap. All the images in Figure 4 are presented at the
tn

same scale and for the same droplets, facilitating comparison of the size of the daughter droplet.

Figure 4(a) shows an example of complete coalescence, where the dashed line indicates the boundary of an
undeformed droplet, and the deformed shape of the water layer during its merging process is depicted below. Arrows
schematically represent the movement direction of the water phase. At an electric field strength of 1 kV/cm or less,
rin

complete coalescence is observed, given all other parameters remain constant. When the voltage increases to 1.6
kV/cm, a partial coalescence mode occurs (Figure 4(b)), characterized by the fact that the separation of the droplet
from the layer does not happen at the location of the bridge between the droplet and the layer. As the voltage further
increases (Figure 4(c)), the volume of the separating droplet increases significantly, although there are no fundamental
differences from the previous case. A voltage increases to 3.6 kV/cm results in a different outcome—stretching of the
ep

droplet along the interelectrode gap. This type of non-coalescence has been observed in various experimental and
numerical studies [10,11,19].
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 d
we
a b

ie
ev
rr
c d

Fig. 4 (a) Examples of various outcomes at different electric field strength: a) coalescence (E = 1 kV/cm, plot presented for not the final moment
of calculation), (b) partial coalescence (E = 1.6 kV/cm), (c) partial coalescence (E = 3 kV/cm), (d) droplet stretching along the interelectrode gap
(E = 3.6 kV/cm); the green circle marks the position of the initial bridge surface at the end of the calculation.
ee
In this case, the location of the bridge formation and the point of separation (formation of the neck) generally do not
coincide. In Figure 4, the green circle indicates the initial position of the bridge section, which shifts during the
coalescence process. In the case of partial coalescence (Figures 4(b-c)), the bridge's position does not remain constant;
it moves closer to the neck as the electric field strength increases. In the extreme case of partial coalescence, or when
p
the droplet stretches, the bridge and the neck position may coincide (Figures 4(c-d)).

According to Figure 5(a), the distance between the final position of the bridge and the neck decreases almost linearly
with increasing voltage, eventually reaching a constant value at high voltages. At the same time, when considering
ot

the positions of the bridge and the neck relative to the water layer (Figure 5(b)), the height of the neck remains nearly
unchanged with increasing voltage. This indicates that the droplet separates from the layer at approximately the same
position, while the bridge's position shifts upward as the voltage increases.
tn
rin
ep
Pr

Fig. 5 (a) The average z - coordinate of the bridge (in gray) and the neck (in red) at the final moment of calculation for a drop of R = 1 mm. (b) The
difference between the z - coordinates of the neck and the bridge at the final moment of time. The z - coordinate is considered relative to the initial
level of the water layer.

Experimental verification

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 To validate the reliability of the numerical model, we compare the calculated and experimental relationships of the
volume of the separating droplet during partial coalescence with respect to electric field strength. We consider
parameters from Mousavichoubeh et al. [14]: interfacial tension 𝛾 = 25 mN/m, oil density ρo = 920 kg/m³, oil dynamic
viscosity μo = 0.0465 Pa·s, relative permittivity ε = 4.9, water density ρw = 998 kg/m³, and water dynamic viscosity

d
ρw = 0.001 Pa·s. The partial coalescence case was analyzed at a voltage of 181 V/mm with a droplet diameter (D) of
1196 μm. As shown in [14] and depicted in Figure 3(a), the volume of the separating droplet strongly depends on the

we
initial distance to the water layer (H), which in this case is 111 micrometers.

Consider the dimensionless parameter 𝑊 = 𝑤𝑏𝑟𝑖𝑑𝑔𝑒/𝐷, representing the evolution of the width of the bridge over
time. The bridge width 𝑤𝑏𝑟𝑖𝑑𝑔𝑒 is defined as the narrowest segment perpendicular to the plane of the image, in the
region where the droplet and the layer touch. During the calculation, a non-deformed spherical droplet was placed at
a distance H from the layer, with zero initial velocity distribution. To align the timing with experimental data, the

ie
moment of contact was set to 1.5 ms, so all graphs start from this value.

As a result of the simulation (Fig. 6), the evolution of the bridge behaves similarly to the experiment until t = 4.5 ms,
after which our model indicates a faster formation of the neck compared to the experimental results. This discrepancy

ev
may be due to a decrease in interfacial tension over time, which is a characteristic of most oils [17,47,58-60]. As
surface tension decreases, the dynamics of the process can change. Although Mousavichoubeh et al. [14] did not
mention any change in IT between sunflower oil and water, more detailed experimental studies [59] confirm this
change (Fig. 7).

rr
To determine which IT value should be used, it's necessary to estimate the contact time of a droplet of water with
sunflower oil in the experiment [14]. Based on the droplet radius, viscosity, and density of the contacting liquids, and
using the expression for the sinking rate of a sphere in a viscous fluid, this contact time is estimated to be approximately
50 seconds. The corresponding IT value is 16.1 ± 0.2 mN/m.
p ee
ot
tn
rin

Fig. 6 The evolution of the bridge according to the W parameter. The experimental curve was obtained from the data presented in [14].

Numerical modeling with this value of interfacial tension shows excellent agreement with experimental data, both in
terms of bridge size (Fig. 6) and droplet shape (Fig. 8), almost until the moment of separation of the daughter droplet.
A slight discrepancy in the evolution of the bridge at times over 12 ms can be explained by the fact that when the
ep

droplet is stretched, "fresh" liquid with an increased IT value, corresponding to zero contact time, can be drawn into
the interface area, which increases surface tension and slows the separation process.

The volume of the separating droplets in the experiment and in the model with the initial value of IT differ. In the
experiment, the volume is separated 𝑉exp ~ 10.1 ∗ 10―3 mm3, while the model with IT 25 mN/m gives 𝑉25 = 7.4 ∗ 1
Pr

0―3 mm3. The calculation with an IT equal to 16.1 mN/m (corresponds to a contact time of 50 s) shows that the
volume of the separating droplet 𝑉16.1 = 10.3 ∗ 10―3 mm3, which practically coincides with the experimental value.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 d
ie we
Fig. 7 The dependence of the interfacial tension on the contact time of water with sunflower oil.

ev
rr
p ee

Fig. 8 Comparison of the partial coalescence process from the experiment [14] and a computer model.
ot

Modes of coalescence
The simulation revealed several electrocoalescence modes: complete coalescence, partial coalescence, and a special
tn

type of partial coalescence involving significant stretching of the droplet along the interelectrode gap, known as
stretching. Figure 4 illustrates how these possible outcomes are implemented in the model.

In the complete coalescence mode, the droplet fully merges with the water layer. In partial coalescence, part of the
droplet is absorbed by the layer, while the remaining part separates and moves toward the upper electrode due to a
residual charge. The transition between these coalescence modes depends on both the electric field strength, which
rin

creates additional electrostatic pressure on the droplet surface, and the force of surface tension, which is associated
with the interfacial tension (γ) and mean curvature. An increase in any parameter associated with surface tension raises
the threshold for transitions between coalescence modes, as this force specifically drives the merging of the droplet
with the layer. Thus, as the radius of the initial droplet decreases, resulting in increased surface tension, a stronger
electric field is required to achieve partial coalescence.
ep

Once the droplet and the layer make contact, the droplet's surface gains a charge, with the highest charge density
observed at the top of the droplet, where the field strength is at its maximum (Figs. 9(a) and 10). This leads to the
assumption that the Coulomb force acts primarily in the vertical direction. At the same time, when the droplet makes
contact with the layer, increased pressure is generated in the bridge area due to heightened curvature. The surface
Pr

tension force acts horizontally on the bridge (Fig. 9(b)), causing it to expand. Alongside this process, the droplet moves
toward or away from the electrode, depending on the electric field strength.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 d
ie we
Fig. 9 The surface plot of the pressure distribution inside the water phase after the construction of the bridge. The arrows indicate the Coulomb
forces (red) and the surface tension forces (black). On the right, an enlarged image in the bridge area of the graph on the left. R = 1 mm, E = 3
kV/cm.

ev
rr
ee
Fig. 10 The contour pressure graph at the moment before the droplet separation. The arrows indicate the Coulomb forces (red) and the surface
tension ones (black). R = 1 mm, E = 3 kV/cm.

Consider the case of low electric field strengths that are close to the threshold for partial coalescence. At this threshold,
p

the Coulomb pressure is an order of magnitude weaker than the surface tension force, causing the droplet to move
toward the layer, while the Coulomb force merely slows this movement by holding the upper pole. At this point, the
liquid is transferred from the droplet into the layer, a phenomenon documented in many studies [8,14,66]. The action
of Coulomb forces causes the droplet to stretch slightly along the vertical axis, leading to the formation of a neck,
ot

which eventually results in the separation of the daughter droplet from the layer. The neck is the region near the
inflection of the surface, where it narrows due to a local maximum in the surface tension force. This local maximum
occurs because the sign of one of the curvatures changes as the surface deforms. The formation point of the neck does
tn

not always align with the initial position of the bridge, as mentioned earlier. Generally, the neck forms above the
bridge's initial position, shifting closer to the top of the droplet, which influences the volume of the daughter droplet.

In summary, electrostatic pressure is primarily responsible for the onset of partial coalescence, as it slows the
movement of the droplet's tip, allowing enough time for the neck to form.
rin

At elevated voltages, when the electrostatic pressure exceeds the surface tension forces, the droplet begins to move
toward the upper electrode, and partial coalescence can also be observed. In this situation, the rate of liquid transfer
from the droplet to the layer decreases significantly, and the volume of the daughter droplet approaches that of the
initial one. This characteristic feature of partial coalescence is illustrated in Figure 11, which shows the ratio of the
volume of the daughter droplet (V) to the initial volume (𝑉0) for droplets with radius of 0.5, 1, and 1.5 mm as a function
ep

of field strength. These dependencies were found to increase with field strength, consistent with experimental
observations [15]. For smaller droplets, increasing the voltage to the threshold leads to the volume of the daughter
droplet almost reaching the volume of the initial droplet, indicating that the process of liquid transfer from the droplet
to the layer is minimal. This mode of partial coalescence at high voltages and with small droplets is similar to the
bouncing mode reported by Hamlin et al. [67].
Pr

At the same time, for larger droplets, the maximum volume of the daughter droplet does not exceed 89% and 60% for
droplets with radius of 1 mm and 1.5 mm, respectively. The difference in the separated volumes is due to the fact that
upon contact, larger droplets are less deformed due to lower electric field strengths, which is where partial coalescence
occurs. Therefore, the expansion and contraction process of the bridge takes longer, allowing more liquid to be
transferred to the layer.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 d
ie we
Fig. 11 The dependence of the daughter drop relative volume on the average field strength.

ev
With a further increase in electric field strength (close to the rupture point of a single droplet), the droplet undergoes
significant deformation, as the Coulomb force is so strong that the upper region of the droplet begins to move toward
the upper electrode and sharpens. This sharpening leads to an increase in the electric charge density in that area,
resulting in additional Coulomb force (the distribution of electric field strength is shown in Figure 12). This process

rr
leads to a point where smaller droplets start to separate (spray) from the upper tip of the droplet.

We identify this electrocoalescence mode separately because extreme droplet stretching poses a risk of short-circuiting
the interelectrode gap. The occurrence of this mode is also undesirable due to the potential for electrospraying, which
reduces the average volume of droplets in the emulsion, complicating demulsification.
p ee
ot
tn

Fig. 12 The surface graph of the electric field strength. R = 1 mm, E = 3.5 kV/cm, t = 50 ms. The arrows indicate the velocity of the liquid, the
streamlines are represented by red curves.

To summarize the results, various articles attempt to analyze the dependencies of characteristic dimensionless
rin

parameters. The work of [15] introduces a dimensionless number WO, which is defined by the numbers We and Oh:
2𝑟0.5𝜀𝑜𝑖𝑙𝜀0𝐸2𝜇
𝑊𝑂 = 𝑊𝑒 ∗ 𝑂ℎ = .
𝜌0.5𝜎1.5

In the corresponding experimental work, all the observed dependencies of the separating volume on the dimensionless
ep

parameter WO aligned for systems with different interfacial tensions, droplet volumes, and electric field strengths. In
our case, the transition from the dependency of the separating volume on electric field strength to the dimensionless
number WO did not yield similar values (Figure 13). This discrepancy is linked to the initial distance of the droplet
from the layer. As noted earlier, the volume of the daughter droplet increases with a larger initial distance. Since we
used a large initial distance such that further changes do not affect the coalescence process, we naturally obtained
Pr

higher volumes for the separating droplets.

Moreover, the WO number also does not provide a consistent match across different droplet sizes (Figure 13). For
larger droplets with a radius R ≥ 1 mm, the dependencies align on a single line, while for smaller droplets with a radius
R < 1 mm, there is a noticeable decrease in relative volume. This deviation can be explained by the dominance of the
surface tension force over the Coulomb force due to the high curvature of smaller droplets. Therefore, it can be

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 concluded that the WO number is not a universal parameter for describing the drop-layer system, as it does not account
for the substantial increase in surface tension force with significant reductions in droplet radius.

d
ie we
ev
Fig. 13 The dependence of the relative volume on the number of WO for droplets of different radius.

Coalescence regimes map

rr
Owing to the developed model, we were able to determine the electric field strength at which partial coalescence
modes change for different droplet radius. A series of calculations was performed at various electrode voltages and
different droplet radius. This allowed us to create a coalescence map indicating at what voltage in the system different
outcomes of the droplet-layer interaction will occur. The resulting coalescence map encompasses all possible modes
ee
in the system (Fig. 14). Similar results for the threshold between complete and partial coalescence have been obtained
experimentally [30]. However, the value of our result lies in its reduced resource requirements, enabling modeling
over a wide range of parameters with the rest held constant. Furthermore, the modeling clearly delineates the region
in which the droplet stretches across the entire interelectrode gap after contacting the layer.
p

The voltage step at which the transition occurred was used to estimate the error in the threshold electric field strength.
For smaller droplets, the error increases due to larger voltage steps, which are needed to maintain convergence when
separating daughter droplets with very small volumes.
ot

In Figure 14, curve 1 represents the threshold for the transition from complete to partial coalescence. All system
configurations below this threshold lead to complete merging of the droplet with the layer. The obtained data suggest
that as the droplet radius increases, the electric field strength required for complete coalescence decreases. The
threshold for the transition to partial coalescence appears to reach an approximately constant value for droplet radius
tn

of 2.5 mm or less, indicating that it is possible to select a voltage where droplet coalescence is observed in smaller
droplets.

Curve 2 indicates the transition from partial coalescence to stretching along the interelectrode gap. The nature of this
dependency changes: while the threshold for partial coalescence (curve 1) remains constant for larger radius, the
stretching mode (curve 2) and rupture mode (curve 3) are described by the relationship E ~ 1/ 𝑅 . It can be observed
rin

that the stretching mode occurs in a narrow range of electric field strengths and, as the droplet radius decreases, tends
toward the threshold for the rupture of a single droplet in an external electric field.

Curve 3 represents the maximum electric field intensity at which droplet rupture occurs. A reduction in droplet size
leads to a significant increase in the threshold electric field strengths. This effect is due to an increase in surface tension
ep

forces, indicating that the Coulomb force, which slows the merging process, must be greater to counteract the surface
tension. A similar trend is observed in drop-drop systems [42,68].

The presented map is valid only for the selected oil and water properties. To fully understand how interfacial tension
and oil viscosity affect the transition thresholds to different coalescence modes, additional studies are required.
Pr

Nonetheless, the resulting dependencies should remain qualitatively similar, with more accurate values calculable
from the numerical model described in this article.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 d
ie we
ev
rr
Fig. 14 Coalescence regimes map.
ee
Conclusion

Based on the work completed, several conclusions can be drawn. First, the use of the arbitrary Lagrangian-Eulerian
method (ALEM) shows promise for numerical modeling of electrocoalescence processes in the droplet-layer system
p

and is advisable for further research to advance the theory of partial coalescence. This approach effectively reproduces
outcomes previously found in experimental studies and allows for determining the relationship between the volume
of separating droplets and the electric field strength in the case of partial coalescence. Moreover, this approach is less
resource-intensive than methods based on diffuse interfaces and does not rely on artificial parameters, unlike, for
ot

instance, the phase field method.

The outcome of the interaction between a droplet and the layer depends on the initial distance between them, up to a
tn

point where the distance becomes sufficient to accelerate the droplet and achieve a quasi-steady-state deformation.
For example, the outcome no longer depends on the initial distance when it exceeds 0.5 mm, given the selected
parameters and a droplet radius of 1 mm.

The resulting mode map allows us to identify the thresholds for transitions from complete to partial coalescence and
from partial coalescence to non-coalescence across a wide range of parameter variations. This map indicates the
rin

outcomes for given values of droplet radius and average electric field strength. Despite being obtained for fixed values
of interfacial tension, viscosity, and only DC voltage, it offers a qualitative understanding of these dependencies.
Similar results for other parameter values can be calculated using the developed numerical model. However, using
dimensionless parameters to generalize the results is challenging due to the complexity of internal processes. For
instance, although the dimensionless WO number can yield consistent dependencies for certain parameter ranges,
ep

these trends may diverge when surface tension forces begin to dominate over Coulomb forces.

The analysis of the transition threshold for partial coalescence showed that as the droplet radius increases, there is a
limit to the electric field strength at which coalescence always occurs. This result suggests that for a water-oil
emulsion, it is possible to identify an electric field strength at which a small portion of the droplet does not separate
Pr

from the water surface into the oil, significantly accelerating the process of electrodehydration.

Finally, it is worth noting that the points of bridge formation and neck formation in partial coalescence generally do
not coincide. However, the distance between them tends to zero as the electric field strength increases. Analyzing the

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 behavior of these formations and the forces involved is one of the directions for further research that can be pursued
with the proposed numerical model.

d
Acknowledgements

we
The study was supported by Russian Science Foundation, research project No. 22-79-10078,
https://rscf.ru/en/project/22-79-10078/. The research was performed at the Research park of St. Petersburg State
University “Computing Center,” “Center for Nanofabrication of Photoactive Materials (Nanophotonics),” and “Center
for Diagnostics of Functional Materials for Medicine, Pharmacology, and Nanoelectronics.”

ie
References

[1] Thompson DG, Taylor AS, Graham DE. Emulsification and demulsification related to crude oil production. Colloids and Surfaces

ev
1985;15:175–89. https://doi.org/https://doi.org/10.1016/0166-6622(85)80070-6.
[2] Eow JS, Ghadiri M, Sharif AO, Williams TJ. Electrostatic enhancement of coalescence of water droplets in oil: a review of the current
understanding. Chem Eng J 2001;84:173–92. https://doi.org/10.1016/S1385-8947(00)00386-7.
[3] Allan RS, Mason SG. Effects of electric fields on coalescence in liquid-liquid systems. Trans Faraday Soc 1961;57:2027–40.
https://doi.org/10.1039/TF9615702027.
[4] Brown AH, Hanson C. The effect of oscillating electric fields on the coalescence of liquid drops. Chem Eng Sci 1968;23:841–8.
https://doi.org/10.1016/0009-2509(68)80019-3.

rr
[5] Dong J, de Almeida VF, Tsouris C. Effects of applied electric fields on drop-interface and drop-drop coalescence. J Dispers Sci Technol
2002;23:155–66. https://doi.org/10.1080/01932690208984196.
[6] Midtgård OM. Electrostatic field theory and circuit analysis in the design of coalescers with pulsed dc voltage. Chem Eng J
2009;151:168–75. https://doi.org/10.1016/j.cej.2009.02.014.
[7] Chen TY, Mohammed RA, Bailey AI, Luckham PF, Taylor SE. Dewatering of crude oil emulsions 4. Emulsion resolution by the
ee
application of an electric field. Colloids Surfaces A Physicochem Eng Asp 1994;83:273–84. https://doi.org/10.1016/0927-
7757(93)02653-V.
[8] Charles GE, Mason SG. The mechanism of partial coalescence of liquid drops at liquid/liquid interfaces. J Colloid Sci 1960;15:105–22.
https://doi.org/10.1016/0095-8522(60)90012-X.
[9] Blanchette F, Bigioni TP. Partial coalescence of drops at liquid interfaces. Nat Phys 2006;2:254–7. https://doi.org/10.1038/nphys268.
[10] Aryafar H, Kavehpour HP. Electrocoalescence: Effects of DC electric fields on coalescence of drops at planar interfaces. Langmuir
p
2009;25:12460–5. https://doi.org/10.1021/la902758u.
[11] Pillai R, Berry JD, Harvie DJE, Davidson MR. Electrophoretically mediated partial coalescence of a charged microdrop. Chem Eng Sci
2017;169:273–83. https://doi.org/10.1016/j.ces.2016.07.022.
[12] Vivacqua V, Ghadiri M, Abdullah AM, Hassanpour A, Al-Marri MJ, Azzopardi B, et al. Analysis of partial electrocoalescence by Level-
Set and finite element methods. Chem Eng Res Des 2016;114:180–9. https://doi.org/10.1016/j.cherd.2016.08.019.
ot

[13] Yang D, Sun Y, Ghadiri M, Wu H, Qiao H, He L, et al. Effect of hydrolyzed polyacrylamide used in polymer flooding on droplet–
interface electro-coalescence: Variation of critical electric field strength of partial coalescence. Sep Purif Technol 2019;227.
https://doi.org/10.1016/j.seppur.2019.115737.
[14] Mousavichoubeh M, Ghadiri M, Shariaty-Niassar M. Electro-coalescence of an aqueous droplet at an oil-water interface. Chem Eng
Process Process Intensif 2011;50:338–44. https://doi.org/10.1016/j.cep.2010.09.017.
tn

[15] Mousavichoubeh M, Shariaty-Niassar M, Ghadiri M. The effect of interfacial tension on secondary drop formation in electro-
coalescence of water droplets in oil. Chem Eng Sci 2011;66:5330–7. https://doi.org/10.1016/j.ces.2011.07.019.
[16] Sunder S, Tomar G. Numerical investigation of a conducting drop’s interaction with a conducting liquid pool under an external electric
field. Eur J Mech B/Fluids 2020;81:114–23. https://doi.org/10.1016/j.euromechflu.2020.01.009.
[17] Hellesø SM, Atten P, Berg G, Lundgaard LE. Experimental study of electrocoalescence of water drops in crude oil using near-infrared
camera. Exp Fluids 2015;56. https://doi.org/10.1007/s00348-015-1990-y.
rin

[18] Sun Y, Yang D, He L, Luo X, Lü Y. Influence of alkali concentration, electric waveform, and frequency on the critical electric field
strength of droplet–interface partial coalescence. Chem Eng Sci 2019;208:115136. https://doi.org/10.1016/j.ces.2019.07.054.
[19] Yang D, Sun Y, He L, Luo X, Lü Y, Gao Q. Partial coalescence of droplets at oil–water interface subjected to different electric
waveforms: Effects of non-ionic surfactant on critical electric field strength. Chem Eng Res Des 2019;142:214–24.
https://doi.org/10.1016/j.cherd.2018.12.019.
[20] Lin Y, Skjetne P, Carlson A. A phase field model for multiphase electro-hydrodynamic flow. Int J Multiph Flow 2012;45:1–11.
https://doi.org/10.1016/j.ijmultiphaseflow.2012.04.002.
ep

[21] Sussman M, Puckett EG. A Coupled Level Set and Volume-of-Fluid Method for Computing 3D and Axisymmetric Incompressible
Two-Phase Flows. J Comput Phys 2000;162:301–37. https://doi.org/10.1006/jcph.2000.6537.
[22] Ervik Å, Lervåg KY, Munkejord ST. A robust method for calculating interface curvature and normal vectors using an extracted local
level set. J Comput Phys 2014;257:259–77. https://doi.org/10.1016/j.jcp.2013.09.053.
[23] Li B, Wang Z, Vivacqua V, Ghadiri M, Wang J, Zhang W, et al. Drop-interface electrocoalescence mode transition under a direct current
electric field. Chem Eng Sci 2020;213. https://doi.org/10.1016/j.ces.2019.115360.
Pr

[24] Feng S, Wu S, Li Y, Yang X, Yu Y, Liu Y, et al. Enhanced Coalescence of Fine Droplets by Medium Coalescence under an Electric
Field. J Mar Sci Eng 2023;11:71. https://doi.org/10.3390/jmse11010071.
[25] Hadidi H, Kamali R, Manshadi MKD. Numerical simulation of a novel non-uniform electric field design to enhance the
electrocoalescence of droplets. Eur J Mech B/Fluids 2020;80:206–15. https://doi.org/10.1016/j.euromechflu.2019.10.010.
[26] Liu Z, Wang P, Li C, Li D, Wang Z, Zhang M, et al. Combined effect of charges and external electric field on collision-coalescence of
microns and nanoscale droplets: A numerical simulation perspective. J Mol Liq 2021;328:115376.
https://doi.org/10.1016/j.molliq.2021.115376.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 [27] Santra S, Sen D, Das S, Chakraborty S. Electrohydrodynamic interaction between droplet pairs in a confined shear flow. Phys Fluids
2019;31. https://doi.org/10.1063/1.5088948.
[28] Xu D, Zhu L, Yang Z, Gao J, Jin M. Parameter Optimization of Catering Oil Droplet Electrostatic Coalescence under Coupling Field
with COMSOL Software. Atmosphere (Basel) 2022;13:780. https://doi.org/10.3390/atmos13050780.

d
[29] Adeyemi I, Meribout M, Khezzar L, Kharoua N, AlHammadi K. Influence of constant wave and pulsatile electric field on water droplets
coalescence in crude oil emulsions. In: 2022 International Conference on Electrical and Computing Technologies and Applications
(ICECTA). IEEE; 2022. p. 257–260. https://doi.org/10.1109/ICECTA57148.2022.9990233.
[30] Anand V, Juvekar VA, Thaokar RM. Coalescence, partial coalescence, and noncoalescence of an aqueous drop at an oil-water interface

we
under an electric field. Langmuir 2020;36:6051–60. https://doi.org/10.1021/acs.langmuir.9b03969.
[31] Brik M, Ruscassie R, Saboni A. Droplet deformation and coalescence under uniform electric field. J Chem Technol Metall 2016;51:649–
59.
[32] Li B, Dou X, Yu K, Huang Y, Zhang W, Xu H, et al. Coalescence dynamic response of an aqueous droplet at an oil-water interface
under a steady electric field. Int J Multiph Flow 2021;139. https://doi.org/10.1016/j.ijmultiphaseflow.2021.103628.
[33] Yang D, Sun H, Li M, Li Q, Gao X, Chen C, et al. Effect of electric field strength and droplet diameter on droplet–interface coalescence
mechanism. Chem Eng Sci 2023;282. https://doi.org/10.1016/j.ces.2023.119360.

ie
[34] Chirkov VA, Gazaryan AV, Kobranov KI, Utiugov GO, Dobrovolskii IA. A modification of the phase-field method to simulate
electrohydrodynamic processes in two-phase immiscible liquids and its experimental verification. J Electrostat 2020;107:103483.
https://doi.org/10.1016/j.elstat.2020.103483.
[35] Jamshidian M, Mohammadi K, Moosavi A, Kazemzadeh Hannani S. Pulsed coaxial drop-on-demand electrohydrodynamic printing.

ev
Phys Fluids 2023;35:32110. https://doi.org/10.1063/5.0141214.
[36] Suo X, Zhang K, Huang X, Wang D, Jia H, Yang F, et al. Electrospray beam currents in the cone-jet mode based on numerical simulation.
Phys Fluids 2023;35:13603. https://doi.org/10.1063/5.0131869.
[37] Zhong S, Lu Y, Yang Y, Zhao Y, Yang Y, Peng Y. Numerical investigation of the effect of operating parameters on droplet ejection in
a double ring electrohydrodynamic printing device. J Electrostat 2023;122:103795. https://doi.org/10.1016/j.elstat.2023.103795.
[38] Li L, Cao Q, Liu H, Qiao X. Molecular dynamics study of electrocoalescence of pure water and salty nanodroplets. J Mol Liq
2021;332:115895. https://doi.org/10.1016/j.molliq.2021.115895.

rr
[39] Li N, Pang Y, Sun Z, Wang Z, Sun X, Tang T, et al. Probing the coalescence mechanism of water droplet and Oil/Water interface in
demulsification process under DC electric field. Sep Purif Technol 2023;326. https://doi.org/10.1016/j.seppur.2023.124798.
[40] Hirt CW, Amsden AA, Cook JL. An arbitrary Lagrangian-Eulerian computing method for all flow speeds. J Comput Phys 1974;14:227–
53. https://doi.org/10.1016/0021-9991(74)90051-5.
[41] Anjos GR. Moving mesh methods for two-phase flow systems: Assessment, comparison and analysis. Comput Fluids 2021;228:105053.
ee
https://doi.org/10.1016/j.compfluid.2021.105053.
[42] Chirkov V, Dobrovolskii I, Vasilkov S. The interaction between two electrohydrodynamics phenomena when an electric field affects a
two-phase immiscible liquid. Phys Fluids 2021;33:43310. https://doi.org/10.1063/5.0046386.
[43] Chirkov VA, Reznikova MP, Lashko A V, Dobrovolskii IA. A Method to Determine the Interfacial Tension for the Conductive
Medium/Liquid Dielectric Couple. 2018 IEEE 2nd Int. Conf. Dielectr., IEEE; 2018, p. 1–4. https://doi.org/10.1109/ICD.2018.8468489.
[44] Dobrovolskii IA, Vasilkov SA, Chirkov VA. Electrohydrodynamics of conducting droplets suspended in a low-conducting liquid: The
p
effect of the difference in mobilities of positive and negative ions. J Electrostat 2023;124:103828.
https://doi.org/10.1016/j.elstat.2023.103828.
[45] Raisin J, Reboud JL, Atten P. Electrically induced deformations of water-air and water-oil interfaces in relation with electrocoalescence.
J Electrostat 2011;69:275–83. https://doi.org/10.1016/j.elstat.2011.03.017.
[46] Ridel A, Korobeynikov S, Lyutikova M, Chirkov V, Kostin P. Water droplet deformation at alternating electric field action: experiment
ot

and numerical simulation. Interfacial Phenom Heat Transf 2023;35. https://doi.org/10.1615/InterfacPhenomHeatTransfer.2023049182.

[47] Xia Y, Reboud JL. Hydrodynamic and electrostatic interactions of water droplet pairs in oil and electrocoalescence. Chem Eng Res Des
2019;144:472–82. https://doi.org/10.1016/j.cherd.2019.02.012.
[48] Chirkov VA, Chernykh BA, Utiugov GO. The effect of pulse voltage application on the threshold electric field strength of the transition
from coalescence to non-coalescence. In: 2022 IEEE 4th International Conference on Dielectrics (ICD). 2022. p. 705–8.
tn

https://doi.org/10.1109/ICD53806.2022.9863515.
[49] Chirkov VA, Saifullin DD, Utiugov GO, Samusenko AV. Numerical simulation of non-coalescence regime for two conducting droplets
at high voltage electrostatic separation of water-in-oil emulsion. In: Proceedings of the 2023 5th International Youth Conference on
Radio Electronics, Electrical and Power Engineering, REEPE 2023. Institute of Electrical and Electronics Engineers Inc.; 2023.
https://doi.org/10.1109/REEPE57272.2023.10086869.
[50] Chirkov V, Utiugov G. Features of Quantitative Verification of Numerical Models for Computing Electrohydrodynamic Processes in
rin

Two-phase Immiscible Liquids. 2020 IEEE Ind. Appl. Soc. Annu. Meet., 2020, p. 1–6.
https://doi.org/10.1109/IAS44978.2020.9334733.
[51] Utiugov G, Chirkov V, Reznikova M. Application of the arbitrary lagrangian-eulerian method to simulate electrical coalescence and its
experimental verification. Int J Plasma Environ Sci Technol 2021;15:1–11. https://doi.org/10.34343/ijpest.2021.15.e02009.
[52] Chirkov V., Utiugov G., Kostin P., Samusenko A. Physical correctness of numerical modeling electrohydrodynamic processes in two-
phase immiscible liquids basing on the phase-field and arbitrary Lagrangian–Eulerian methods. Int J Multiph Flow 2024;177:104881.
https://doi.org/10.1016/j.ijmultiphaseflow.2024.104881.
ep

[53] Pang B, Ren Y, Shen Y, Liu HR, Ding H. A conservative sharp interface method for incompressible viscous two-phase flows with
topology changes and large density ratio. Comput Fluids 2024;274. https://doi.org/10.1016/j.compfluid.2024.106212.
[54] Supeene G, Koch CR, Bhattacharjee S. Deformation of a droplet in an electric field: Nonlinear transient response in perfect and leaky
dielectric media. J Colloid Interface Sci 2008;318:463–76. https://doi.org/10.1016/j.jcis.2007.10.022.
[55] Young DL, Jan YJ, Chiu CL. A novel immersed boundary procedure for flow and heat simulations with moving boundary. Comput
Fluids 2009;38:1145–59. https://doi.org/10.1016/j.compfluid.2008.11.010.
Pr

[56] Elagin IA, Yagodin GS, Chirkov VA. Features of numerical modeling electrical coalescence in a droplet-layer system using the arbitrary
Lagrangian-Eulerian method. J Phys Conf Ser 2024;2701:012078. https://doi.org/10.1088/1742-6596/2701/1/012078.
[57] Yeoh OH. Some Forms of the Strain Energy Function for Rubber. Rubber Chem Technol 1993;66:754–71.
https://doi.org/10.5254/1.3538343.
[58] Bahtz J, Knorr D, Tedeschi C, Leser ME, Valles-Pamies B, Miller R. Adsorption of octanoic acid at the water/oil interface. Colloids
Surfaces B Biointerfaces 2009;74:492–7. https://doi.org/10.1016/j.colsurfb.2009.07.041.

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436
 [59] Chirkov V, Utiugov G, Blashkov I, Vasilkov S. The effect of changing interfacial tension on electrohydrodynamic processes in two-
phase immiscible liquids. Int J Plasma Environ Sci Technol 2023;17:e03003. https://doi.org/10.34343/ijpest.2023.17.e03003.
[60] Dopierala K, Javadi A, Krägel J, Schano KH, Kalogianni EP, Leser ME, et al. Dynamic interfacial tensions of dietary oils. Colloids
Surfaces A Physicochem Eng Asp 2011;382:261–5. https://doi.org/10.1016/j.colsurfa.2010.11.027.

d
[61] Xu T, Rodriguez-Martinez V, Sahasrabudhe SN, Farkas BE, Dungan SR. Effects of Temperature, Time and Composition on Food Oil
Surface Tension. Food Biophys 2017;12:88–96. https://doi.org/10.1007/s11483-016-9466-z.
[62] Anand V, Juvekar VA, Thaokar RM. Modes of coalescence of aqueous anchored drops in insulating oils under an electric field. Colloids
Surfaces A Physicochem Eng Asp 2019;568:294–300. https://doi.org/10.1016/j.colsurfa.2019.02.002.

we
[63] Anand V, Roy S, Naik VM, Juvekar VA, Thaokar RM. Electrocoalescence of a pair of conducting drops in an insulating oil. J Fluid
Mech 2019;859:839–50. https://doi.org/10.1017/jfm.2018.849.
[64] Yang D, Ghadiri M, Sun Y, He L, Luo X, Lü Y. Critical electric field strength for partial coalescence of droplets on oil–water interface
under DC electric field. Chem Eng Res Des 2018;136:83–93. https://doi.org/10.1016/j.cherd.2018.05.004.
[65] Yang D, Sun Y, He L, Luo X, Lü Y. Coalescence characteristics of droplets at oil–water interface subjected to different electric
waveforms. Chem Eng Process - Process Intensif 2018;134:28–37. https://doi.org/10.1016/j.cep.2018.10.013.
[66] Mousavi SH, Ghadiri M, Buckley M. Electro-coalescence of water drops in oils under pulsatile electric fields. Chem Eng Sci

ie
2014;120:130–42. https://doi.org/10.1016/j.ces.2014.08.055.
[67] Hamlin BS, Creasey JC, Ristenpart WD. Electrically tunable partial coalescence of oppositely charged drops. Phys Rev Lett 2012;109.
https://doi.org/10.1103/PhysRevLett.109.094501.
[68] Chirkov VA, Kostin PA. Generalization of numerical simulation results on the electrical coalescence threshold for two conducting

ev
droplets based on non-dimensional parameters. J Phys Conf Ser 2024;2701:012075. https://doi.org/10.1088/1742-6596/2701/1/012075.

rr
p ee
ot
tn
rin
ep
Pr

This preprint research paper has not been peer reviewed. Electronic copy available at: https://ssrn.com/abstract=4858436