ILASS–Europe 2017, 28th Conference on Liquid Atomization and Spray Systems, 6-8 September 2017, Valencia, Spain

http://dx.doi.org/10.4995/ILASS2017.2017.****

Numerical Investigation of Droplet Impact on Smooth Surfaces with

Different Wettability Characteristics: Implementation of a dynamic
contact angle treatment in OpenFOAM
Vontas K.1,2., Andredaki M.1, Georgoulas A.1, Nikas K.-S.2, Marengo M.1
1
Advanced Engineering Centre, School of Computing Engineering and Mathematics,
University of Brighton
2
Mechanical Engineering Department, Piraeus University of Applied Sciences
*Corresponding author: A.Georgoulas@brighton.ac.uk

Abstract
The “Direct Numerical Simulations” (DNS) of droplet impact processes is of great interest and importance for a
variety of industrial applications, where laboratory experiments might be difficult, costly and time-consuming.
Furthermore, in most cases after validated against experimental data, they can be utilised to further explain the
experimental measurements or to extend the experimental runs by performing “virtual” numerical experiments. In
such “DNS” calculations of the dynamic topology of the interface between the liquid and gas phase, the selected
dynamic contact angle treatment is a key parameter for the accurate prediction of the droplet dynamics. In the
present paper, droplet impact phenomena on smooth, dry surfaces are simulated using three different contact
angle treatments. For this purpose, an enhanced VOF-based model, that accounts for spurious currents
reduction, which has been previously implemented in OpenFOAM CFD Toolbox, is utilised and further enhanced.
Apart from the already implemented constant and dynamic contact angle treatments in OpenFOAM, the dynamic
contact angle model of Kistler, that considers the maximum advancing and minimum receding contact angles, is
implemented in the code. The enhanced VOF model predictions are initially compared with literature available
experimental data of droplets impacting on smooth surfaces with various wettability characteristics. The constant
contact angle treatment of OpenFOAM as well as the Kistler’s implementation show good qualitative and
quantitative agreement with experimental results up to the point of maximum spreading, when the spreading is
inertia dominated. However, only Kistler’s model succeeds to accurately predict both the advancing and the
recoiling phase of the droplet impact, for a variety of surface wettability characteristics. The dynamic contact angle
treatment fails to predict almost all stages of the droplet impact. The optimum version of the model is then applied
for 2 additional series of parametric numerical simulations that identify and quantify the effects of surface tension
and viscosity, in the droplet impact dynamics.

Keywords
Droplet impact, dynamic contact angle treatment, VOF, OpenFOAM

Introduction
In the last 20 years, many investigations of droplet impact have taken place experimentally as well as numerically.
Wettability constitutes an important controlling parameter in the dynamics of droplet impact, as it can completely
alter the impact characteristics and output [1]. The “Direct Numerical Simulations” (DNS) of droplet impact
processes is of great interest and importance for a variety of industrial applications, where laboratory experiments
might be difficult, costly and time-consuming. Furthermore, in most cases after validated against experimental
data, they can be utilised to further explain the experimental measurements or to extend the experimental runs by
performing “virtual” numerical experiments. In such “DNS” calculations of the dynamic topology of the interface
between the liquid and gas phase, the selected Dynamic Contact Angle (DCA) treatment is a key parameter for
the accurate prediction of the droplet dynamics, since it underpins the wettability characteristics of the simulated
phenomenon. The droplet impact of water on a flat, solid surface has been studied by Pasandideh-Fard et al. [2].
A numerical solution of the Navier-Stokes equations, using a modified SOLA-VOF method was utilised to model
the interface deformation. Yokoi et al. [3] investigated liquid droplet impact behaviour onto a dry and flat surface
numerically and compared their results with experimental data, indicating the significant role of the DCA modelling
in reproducing the droplet impact behaviour. Their numerical method consists of a Coupled Level Set and VOF
framework (CLSVOF), a volume/surface integrated average based multi-moment method, and a continuum
surface force model. The experimental work of Antonini et al. [4] is focused in the understanding of the effect of
surface wettability on impact characteristics of water drops, onto solid dry surfaces. Their results indicated the role
of advancing contact angle and contact angle hysteresis as fundamental wetting parameters. They also found
that, generally, if Reynolds (Re) and Weber numbers (We) are high enough, the spreading drop can be

This work is licensed under a Creative Commons 4.0 International License (CC BY-NC-ND 4.0).
EDITORIAL UNIVERSITAT POLITÈCNICA DE VALÈNCIA
ILASS – Europe 2017, 6-8 Sep. 2017, Valencia, Spain

subdivided into two main regions: a radially spreading lamella and an almost circular rim appearing due to
capillary forces and viscosity [4]. A novel numerical implementation for the adhesion of liquid droplets impacting
normally on solid dry surfaces was presented by Malgarinos et al. [5]. The benefit of this new approach,
compared to most existing models, is that the DCA forming during the surface wetting process is not inserted as a
boundary condition, but is derived implicitly by the induced fluid flow characteristics and the adhesion physics of
the gas-liquid-solid interface, starting only from the advancing and receding equilibrium contact angles. Zhang et
al. [6] investigated the phenomenon of spray impinging on in-cylinder walls, a phenomenon that has important
impact on combustion processes as well as harmful emissions for internal combustion engines. Droplet impact
with a numerical methodology for modelling contact line motion in a Dual–Grid Level–Set method (DGLS), on
hydrophobic and super-hydrophobic surfaces was implemented by Patil et al. [7], using a quasi - DCA model
based on experimental inputs. The accuracy of the partially refined DGLS method is close to that of the fine–grid
based LS method, at a computation cost which is close to that of the coarse–grid based LS method. Furthermore,
the DGLS method is demonstrated as an improved LS method for computational multi-fluid dynamics simulations,
involving contact line motion.
In the present paper, droplet impact phenomena on smooth, dry surfaces are simulated utilising and comparing
three different numerical treatments for the contact angle at the solid-liquid-gas triple contact line against existing,
literature available, experimental measurements. For this purpose, an enhanced VOF-based model that accounts
for spurious currents reduction, which has been previously implemented in OpenFOAM CFD Toolbox, and it was
validated and applied for the case of adiabatic bubble dynamics [8], is further improved. In more detail, apart from
the already implemented Constant Contact Angle (CCA) and DCA treatments in OpenFOAM, the DCA treatment
of Kistler, that considers the limiting advancing and receding contact angles, is also implemented in the code. In
the first part of the paper, the predictions of the three different contact angle treatments in the utilised enhanced
VOF models are compared with literature reported experimental data of droplets impacting on smooth surfaces
[2]. The CCA and Kistler’s DCA models show good qualitative and quantitative agreement with the experimental
results reported by Pasandideh-Fard et al. [2] up to the point of maximum spreading, when the spreading is inertia
dominated. However, only Kistler’s model succeeds to accurately predict the recoiling phase of the droplet impact.
The original DCA model of OpenFOAM fails to predict almost all stages of the simulated droplet impact case.
Then Kistler’s treatment implementation, as it performs better, is further validated by the numerical reproduction of
three additional experiments with varying wettability characteristics [3,7]. The proposed cases vary from
hydrophilic to hydrophobic. The optimum version of the proposed numerical framework is then applied for two
additional series of parametric numerical simulations (virtual experiments) that isolate, identify and quantify the
effects of surface tension and viscosity, in the droplet impact dynamics. The effect of the varied parameters on the
droplet spreading factors with time is identified and quantified and comparisons with the theoretical correlation by
Roisman for the maximum droplet spreading factor are also conducted [9]. The proposed correlation is given by
Equation 1,

𝛽"#$ = 0.87𝑅𝑒 ,/. − 0.4𝑅𝑒 1/. 𝑊𝑒 3,/1 (1)

Numerical Method
With the VOF approach, the transport equation for the volume fraction, 𝛼, of the secondary (dispersed) phase is
solved simultaneously with a single set of continuity and Navier–Stokes equations for the whole flow field. The
corresponding volume fraction of the primary phase is simply calculated as 1 − 𝛼 . The main underlying
assumptions are that the two fluids are Newtonian, incompressible, and immiscible. The governing equations can
be written as:

∇∙𝑈 =0 (2)

𝜕𝜌; 𝑈 (3)
+ 𝛻 ∙ 𝜌; 𝑈𝑈 = −𝛻𝑝 + 𝛻 ∙ 𝜇; 𝛻𝑈 + 𝛻𝑈 A + 𝜌; 𝑓 + 𝐹D
𝜕𝑡

𝜕𝑎
+ ∇ ∙ 𝛼𝑈 − ∇ ∙ (𝛼 1 − 𝛼 𝑈G ) = 0 (4)
𝜕𝑡
where the bulk fluid properties are calculated as weighted averages of the individual phase properties as follows,

This work is licensed under a Creative Commons 4.0 International License (CC BY-NC-ND 4.0).
EDITORIAL UNIVERSITAT POLITÈCNICA DE VALÈNCIA
ILASS – Europe 2017, 6-8 Sep. 2017, Valencia, Spain

𝜌; = 𝜌𝛼 + 𝜌(1 − 𝛼) (5)

𝜇; = 𝜇𝛼 + 𝜇 + 𝜇(1 − 𝛼) (6)

In the VOF method, 𝛼 is advected by the velocity field. For the case of incompressible flow, this is equivalent to
volume fraction conservation, which makes the method mass conservative. Interface sharpening is very important
in simulating two-phase flows of two immiscible fluids. In OpenFOAM the sharpening of the interface is achieved
artificially by introducing an extra compression term in the advection equation of α (last term in the left-hand side
of Eq. 4). 𝑈G is the artificial compression velocity which is calculated from the following relationship,

P P
𝑈G = 𝑛K 𝑚𝑖𝑛 𝐶O , 𝑚𝑎𝑥 (7)
DQ DQ

Finally, the surface tension force is modelled as a volumetric force using the Continuum Surface Force (CSF)
method by Brackbill et al. [10], applying the following equations:

𝐹D = 𝛾ĸ ∇𝛼 (8)

∇𝛼
ĸ=∇∙ (9)
∇𝛼

where 𝛾 is the surface tension coefficient and ĸ is the curvature of the interface. As mentioned in the introduction
section of the present paper the utilized numerical framework constitutes an enhanced version of the original
VOF-based solver of OpenFOAM [8], that suppresses numerical artefacts of the original model, known as
“spurious currents”. The proposed enhancement involves the calculation of the interface curvature ĸ using the
smoothed volume fraction values 𝛼, which are obtained from the initially calculated 𝛼 field, smoothing it over a
finite region near the interface. All other equations are using the initially calculated (non-smoothed) volume
fraction values of 𝛼. Further details on the proposed numerical modelling framework can be found in [8].
In OpenFOAM, there are two ways to predict the evolution of the contact angle between the liquid-gas interface
and the solid wall boundary. The simplest approach is to assign a constant value equal to the equilibrium contact
angle, and therefore neglecting the contact angle hysteresis. This is usually known as static or constant contact
angle treatment. A more complicated approach involves the application of a contact angle that varies with respect
to the instantaneous flow quantities. Such treatments are known as dynamic contact angle treatments (DCA). The
original distribution of OpenFOAM includes both a CCA treatment as well as a DCA treatment.
For the purposes of the present investigation, after the satisfactory predictions in a similar investigation by
Criscione et al. [11], the adopted, enhanced, VOF-based solver from the work of Georgoulas et al. [8], is further
improved by implementing an additional DCA treatment in the solver. The proposed treatment, is known as
Kistler’s model [12], and it calculates the DCA, 𝜃X , using the Hoffman function, 𝑓YZKK , as follows:

3,
𝜃X = 𝑓YZKK 𝐶# + 𝑓YZKK (𝜃[ ) (10)
]^_`
where 𝜃\ is the equilibrium contact angle. The capillary number, 𝐶# , is calculated as 𝐶# = and 𝑈CL, is the
O
3,
spreading velocity of the contact line. 𝑓YZKK is the inverse function of “Hoffman’s” empirical function which is
given in the following form [11].

$ n.onp
𝑓YZKK = 𝑎𝑐o𝑠 1 − 2𝑡𝑎𝑛ℎ 5.16 (11)
,j,.k,$ l.mm

Validation of Numerical Method

o
All simulations presented in the present paper constitute 2D axisymmetric runs. The computational domain is a 5
wedge, with 5 mm width and 8 mm height. The computational mesh consists of 1.6 million hexahedral cells
(1000x1600x1). The dimensions of the computational domain and the total number of computational cells, were

This work is licensed under a Creative Commons 4.0 International License (CC BY-NC-ND 4.0).
EDITORIAL UNIVERSITAT POLITÈCNICA DE VALÈNCIA
ILASS – Europe 2017, 6-8 Sep. 2017, Valencia, Spain

selected after an initial sensitivity analysis and a mesh independency study, respectively. The computational
domain, mesh and boundary conditions are illustrated in Figure 1. To validate the adopted, enhanced, VOF-based
solver for droplet impact cases and compare the numerical predictions with the previously mentioned, contact
angle treatments, four different, literature available, experiments on droplet impact are numerically reproduced. In
the first case, all three contact angle models are tested, while in the rest three cases only Kistler’s dynamic
contact angle is used, to reproduce the proposed experimental runs. The initial conditions and the wetting
characteristics of the proposed experimental runs are summarized in Table 1.

Figure 1. Computational domain, mesh and boundary conditions for numerical simulations

As it can be seen for the axisymmetric simulations of the present paper, a structured computational mesh,
consisting of hexahedral and prismatic elements was used, with grid clustering towards the bottom left corner of
the computational domain (centre of droplet impact). At the solid walls, a no-slip velocity boundary condition was
used with a fixed flux pressure boundary condition for the pressure values and a contact angle condition,
according to the selected in each case treatment, for the volume fraction values. These contact angle boundary
conditions are used to correct the surface normal vector, and therefore adjust the curvature of the interface near
the wall, in relation to the prescribed wettability characteristics. At the outlet, a fixed valued (atmospheric)
pressure boundary condition and a zero-gradient boundary condition for the volume fraction were used, while for
the velocity values a special (combined) type of boundary condition was used that applies a zero-gradient when
the fluid mixture exits the computational domain and a fixed value condition to the tangential velocity component,
in cases that fluid enters the domain. Further details regarding the utilised boundary conditions can be found in
OpenFOAM Documentation [13].

Table 1. Initial conditions and wetting characteristics of validation experiments

o o o
D0 [m] U0 [m/s] We Re θe [ ] θ a[ ] θr [ ]
Experiment I [2] 0.002 1 27 2000 90 110 40
Experiment II [3] 0.00228 1 32 2280 90 107 77
Experiment III [7] 0.0017 0.34 2.7 578 147 161 132
Experiment IV [7] 0.002 0.44 5.3 880 158 165 142

As it can be observed from Table 1, the selected validation cases constitute experiments with various impact as
well as wettability characteristics. Experiments I and II constitute droplet impacts in hydrophilic surfaces, while in
experiments III and IV hydrophobic and super-hydrophobic surfaces are used for the impacts, respectively.
As mention previously, in the case of Experiment I all three different contact angle treatments are used to
numerically reproduce the considered droplet impact. A macroscopic comparison of the numerical predictions for
the droplet evolution with the corresponding experimental snapshots at the same time instances, for each
treatment, is illustrated in Figure 2. For a more quantitative comparison, the contact diameter of the droplet with
the solid surface with respect to time is plotted for each of the three numerical simulations as well as for the
experimental measurements, in the diagram of Figure 3.

This work is licensed under a Creative Commons 4.0 International License (CC BY-NC-ND 4.0).
EDITORIAL UNIVERSITAT POLITÈCNICA DE VALÈNCIA
ILASS – Europe 2017, 6-8 Sep. 2017, Valencia, Spain

Figure 2. Droplet evolution with time - Comparison of numerical predictions for all three contact angle treatments (present
investigation) with corresponding experimental snapshots from [2]. The velocity magnitude field in a central vertical section of
the droplet, is also shown in the numerical snapshots.

Figure 3. Contact diameter with time - Comparison of numerical predictions (present investigation) for all three contact angle
treatments with corresponding experimental data from [2].

After examining Figures 2 and 3, it is obvious that both the CCA and DCA treatments implemented in the original
distribution of OpenFOAM fail to predict the experimental data, while Kistler’s DCA treatment follows the
experimental points well, both before and after the maximum spreading. However, in order to further check the
validity of the Kistler’s DCA treatment for cases with varying wettability characteristics, three additional, literature
available experimental runs that are reported in [3] and [7], are numerically reproduced. In more detail, another
hydrophilic (Experiment II), a hydrophobic (Experiment III) and a super-hydrophobic case (Experiment IV) are
selected for this purpose. A macroscopic comparison of the numerical predictions for the droplet evolution with
the corresponding experimental snapshots, is illustrated in Figure 4. For a more quantitative comparison, the
contact diameter of the droplet with the solid surface with respect to time, is plotted for each case in the diagrams
of Figure 5.

Figure 4. Droplet evolution with time - Comparison of numerical predictions (present investigation) using Kistler’s dynamic
contact angle treatment with corresponding experimental snapshots reported in [3] (top) and [7] (middle) and (bottom). The
velocity magnitude field in a central vertical section of the droplet, is also shown in the numerical snapshots.

This work is licensed under a Creative Commons 4.0 International License (CC BY-NC-ND 4.0).
EDITORIAL UNIVERSITAT POLITÈCNICA DE VALÈNCIA
ILASS – Europe 2017, 6-8 Sep. 2017, Valencia, Spain

(a) (b)

(c)

Figure 5. Contact diameter with respect to time - Comparison of numerical predictions (present investigation) using Kistler’s
dynamic contact angle treatment with corresponding experimental snapshots reported in [3] (a) and [7] (b) and (c).

As it can be observed, Kistler’s DCA treatment successfully predicts the spatial and temporal evolution of the
droplet in all stages of the considered impacts. Therefore, it can be concluded that the utilised enhanced VOF
framework in conjunction with the implemented DCA treatment of Kistler, can safely be applied for the
investigation of droplet impacts through “virtual” numerical experiments, as it successfully predicts the droplet
impact stages in hydrophilic, hydrophobic and super-hydrophobic cases.

Parametric Analysis – Effect of fluid viscosity and surface tension

In the present section of the paper, the optimum version of the VOF-based numerical framework that is presented
and validated against experimental data in the previous sections, is further applied for the conduction of two
additional series of parametric numerical simulations. The proposed numerical experiments mainly aim to identify
and quantify the effects of viscosity and surface tension on the spatial and temporal evolution of the droplets after
their impact on a hydrophilic surface. Furthermore, the validity of a widely used theoretical correlation by Roisman
[9] (Equation 1) for the case of “virtual” fluids is assessed. For this purpose, the validation case reproducing
Experiment II is selected as the base case for the proposed parametric numerical investigations. In the first
parametric investigation 5 additional simulations are conducted “virtually” varying only the liquid viscosity, by
factors of 0.5, 1.5, 2.0, 3.0 and 3.5, while keeping all the other properties and impact characteristics constant and
equal to the base case. In the second parametric investigation, again 5 additional simulations are conducted
virtually varying only the surface tension coefficient, by factors of 0.5, 1.5, 2.0, 3.0 and 3.5, while keeping all the
other properties and impact characteristics constant and equal to the base case. The evolution of the spreading
factor 𝛽 with dimensionless time 𝑡 ∗ for each case is plotted. The spreading factor is calculated as the contact
r ^l
diameter over the initial droplet diameter while the dimensionless time 𝑡 ∗ is calculated as 𝑡 . The effects of
rl rl
the variation of viscosity and surface tension on the spreading factor β with respect to dimensionless time t* are
given in Figures 6(a) and 6(b). In each case the corresponding maximum spreading factor from Equation 1
(Roisman correlation [9]) is plotted for comparison purposes.

This work is licensed under a Creative Commons 4.0 International License (CC BY-NC-ND 4.0).
EDITORIAL UNIVERSITAT POLITÈCNICA DE VALÈNCIA
ILASS – Europe 2017, 6-8 Sep. 2017, Valencia, Spain

(a) (b)
Figure 6. (a) Spreading factor vs dimensionless time – effect of viscosity variation (b) spreading factor vs dimensionless time –
effect of surface tension variation. Dotted lines correspond to the maximum spreading factor as predicted using the theoretical
correlation of Roisman [9] (Equation 1).

As it can be observed from Figure 6 (a), as viscosity increases both the spreading as well as the recoiling stages
of the droplet are dumped. It is characteristic however that the proposed effect, is mor intense around the
maximum spreading stage of the drople,t while it appears to be negligible during the initial spreading and the later
recoiling stages. It can also be seen that the theoretical correlation of Roisman, in each case slightly over-predicts
the resulting maximum spreading factors from the numerical simulations, by 2% up to 4%. Examining Figure 2 (b),
it can be observed that as surface tension increases, the maximum spreading is significantly reduced and the
recoiling stage is strengthened. The proposed effect is quite evident from the first stages of the spreading and it
progressively increases towards the last stages of the recoiling phase. It is characteristic that a variation of the
surface tension coefficient by a factor of just 0.5 can significantly alter the post-impact dynamics of the droplet. As
for the comparison with the theoretical correlation of Roisman an increasing deviation from the numerically
predicted maximum spreading factor is observed with the corresponding increase of the surface tension. The
deviation between the theoretical and numerical maximum spreading factors in this case varies from 0.85% up to
55%.
The results from the present parametric investigation illustrate that the correlation proposed by Roisman [9] can
be safely used for the prediction of the maximum spreading of viscous fluids with viscosities up to 3.5 times higher
than water, but it significantly fails to predict the maximum spreading for fluids with surface tension more than two
times the surface tension of water, such as metal fluids.

Conclusions
In the present paper, Kistler’s DCA model has been implemented in a previously improved version of the VOF-
based solver of OpenFOAM. The performance of the proposed contact angle treatment is compared with the
existing contact angle treatments of OpenFOAM’s original distribution, through comparison with literature reported
experimental measurements on water droplets impacting on hydrophilic surfaces. To further test the revealed
accuracy of the proposed DCA implementation, three additional experimental runs are numerically reproduced; a
second hydrophilic case as well as a hydrophobic and a super-hydrophobic case. In all cases, Kistler’s model
implementation in the utilised, enhanced VOF-based solver, successfully predicts the spreading, recoiling and
rebounding stages of the droplet impact. Further application of the numerical model for the conduction of two
additional series of parametric numerical experiments identifies and quantifies the effects of viscosity and surface
tension variation in the post-impact stages of the droplet evolution. Comparison of the numerical predictions with
a widely accepted theoretical correlation [9], indicate that the proposed correlation can be safely applied to predict
the maximum spreading of fluids with higher viscosity than water but not for fluids with more than two times the
surface tension of water. Finally, from the overall presentation and analysis of the results it is obvious that the
proposed enhanced VOF framework can be safely used to further examine the effects of a variety of important
controlling parameters to the post-impact characteristics of droplets impinging on solid surfaces with various
wetting characteristics, ranging from hydrophilic to super-hydrophobic cases.

Nomenclature
𝛽"#$ maximum spreading factor [-] (contact droplet diameter / initial droplet diameter)
𝑅𝑒 Reynolds Number [-]
𝑊𝑒 Weber Number [-]
-1
𝑈 velocity vector [ms ]
2 2
𝜌; 𝑓 volumetric representation of the gravitational force [kg/m s ]
𝑝 pressure [Pa]

This work is licensed under a Creative Commons 4.0 International License (CC BY-NC-ND 4.0).
EDITORIAL UNIVERSITAT POLITÈCNICA DE VALÈNCIA
ILASS – Europe 2017, 6-8 Sep. 2017, Valencia, Spain

-3
𝜌s bulk density [kgm ]
-1 1
𝜇; bulk viscosity [kgm s- ]
𝛼 volume fraction [-]
-3
𝜌 density of phase 1 [kgm ]
-3
𝜌 density of phase 2 [kgm ],
-1 1
𝜇 viscosity of phase 1[kgm s- ]
-1 1
𝜇 viscosity of phase 2 [kgm s- ]
-1
𝛾 surface tension [Nm ]
-1
ĸ curvature of the interface [m ]
-2 -2
𝐹D volumetric representation of the surface tension force [kgm s ]
-1
𝑈G artificial compression velocity [m s ]
𝑛K cell surface normal vector
−1 −2
𝜑 mass flux [ kg s m ]
2
𝑠K surface area of the cell [m ]
𝐶O interface compression coefficient [-]
𝛼 smoothed volume fraction [-]
0
𝜃[ equilibrium contact angle [ ]
0
𝜃# advancing contact angle [ ]
0
𝜃G receding contact angle [ ]
0
𝜃X dynamic contact angle [ ]
𝑓YZKK Hoffman function [-]
𝐶# capillary number [-],
𝐷n initial droplet diameter [m],
-1
𝑈n impact velocity [ms ],

Subscripts
𝒃 bulk
𝒑 cell
𝒇 face

References
[1] Marengo, M., Antonini, C., Roisman, I. V., and Tropea, C., 2011, “Drop Collisions with Simple and
Complex Surfaces,” Curr. Opin. Colloid Interface Sci., 16(4), pp. 292–302.
[2] Pasandideh-Fard, M., Y. M. Qiao, Y. M., Chandra, S., and Mostaghimi, J., 1996, “Capillary Effects during
Droplet Impact on a Solid Surface,” Phys. Fluids, 8(3).
[3] Yokoi, K., Vadillo, D., Hinch, J., and Hutchings, I., 2009, “Numerical Studies of the Influence of the
Dynamic Contact Angle on a Droplet Impacting on a Dry surfaceNo Title,” Phys. Fluids, 21(7), p. 72102.
[4] Antonini, C., Amirfazli, A., and Marengo, M., 2012, “Drop Impact and Wettability: From Hydrophilic to
Superhydrophobic Surfaces,” Phys. fluids, 24(10), p. 102104.
[5] Malgarinos, I., Nikolopoulos, N., Marengo, M., Antonini, C., and Gavaises, M., 2014, “VOF Simulations of
the Contact Angle Dynamics during the Drop Spreading: Standard Models and a New Wetting Force
Model,” Adv. Colloid Interface Sci., 212, pp. 1–20.
[6] Zhang, Z., Liu, H., Zhang, F., and Yao, M., 2016, “Numerical Study of Spray Micro-Droplet Impinging on
Dry/wet Wall,” Appl. Therm. Eng., 95, pp. 1–9.
[7] Patil, N. D., Gada, V. H., Sharma, A., and Bhardwaj, R., 2016, “On Dual-Grid Level-Set Method for
Contact Line Modeling during Impact of a Droplet on Hydrophobic and Superhydrophobic Surfaces,” Int.
J. Multiph. Flow, 81, pp. 54–66.
[8] Georgoulas, A., Koukouvinis, P., Gavaises, M., and Marengo, M., 2015, “Numerical Investigation of
Quasi-Static Bubble Growth and Detachment from Submerged Orifices in Isothermal Liquid Pools: The
Effect of Varying Fluid Properties and Gravity Levels,” Int. J. Multiph. Flow, 74, pp. 59–78.
[9] Roisman, I., 2009, “Inertia Dominated Drop Collisions. II. An Analytical Solution of the Navier–Stokes
Equations for a Spreading Viscous Film,” Phys. fluids, 21(52104).
[10] Brackbill, J. ., Kothe, D. ., and Zemach, C., 1992, “A Continuum Method for Modeling Surface Tension,” J.
Comput. Phys., 100(2), pp. 335–354.
[11] Criscione, A., Rohrig, R., Opfer, L., Roisman, I., and Jakirlic, S., 2011, “Numerical Investigation of
Impacting Water Drops in Air Cross-Flow,” ILASS—Europe 2011, 24th European Conference on Liquid
Atomization and Spray Systems, Estoril, Portugal.
[12] Kistler, S., 1993, “Hydrodynamics of Wetting,” Wettability, 6(Marcel Dekker, New York), pp. 311--430.
[13] CFD_Direct, 2013, “The OpenSource CFD Toolbox, User Guide.”

This work is licensed under a Creative Commons 4.0 International License (CC BY-NC-ND 4.0).
EDITORIAL UNIVERSITAT POLITÈCNICA DE VALÈNCIA