When a liquid drop impacts a liquid surface, it deforms the surface, trapping a thin layer of air between drop and bath. The drop then either bounces off the bath or, if the gas film ruptures, merges with it. We present a model for drop-bath impacts that can accurately predict whether a drop will bounce or merge, and compare it to theory-driven experiments. The model incorporates a lubrication description of the deforming trapped gas film, which accounts for its significant curvature, within a finite-element framework. Capturing this outcome requires micro- and nano-scale physics: van der Waals-driven disjoining pressure and gas-kinetic effects that depend on the mean free path of the gas. Using only independently measured material parameters, we obtain the first quantitative agreement with experiment for the critical impact speed separating bounce from merge. We explore the behaviour of drop-bath impacts at low ambient air pressure, and then investigate the bounce-merge threshold for impacts onto shallow baths and thin films, finding a bounce-merge-bounce-merge transition as the bath depth decreases.
Introduction
Understanding the behaviour of liquid drops that impact liquid surfaces is important to a wide variety of applications, including the processes of air-sea gas exchange when rain drops hit the ocean ([86]), inkjet printing ([54]), and spray cooling of hot electronics components ([52]). Although there are many possible outcomes of such an impact, including coalescence, splashing, or the production of a high-speed jet as first identified by Worthington [87], impacts can be divided into two primary outcomes: either the drop makes contact with the impacted surface, or it does not, because there is a thin layer of the surrounding gas that becomes sandwiched between the drop and the surface. In the latter case, the lubricating pressure from the gas film is strong enough to prevent contact so that either the drop rebounds, a phenomenon observed as early as in Reynolds [70], or floats for a sustained period before eventually coalescing (Walker [83]). Notably, it is now well known that by vibrating the bath a drop can be made to bounce repeatedly (Couder et al. [15]) – with the resulting interaction between the drop and its induced waves producing a quantum analogue system capable of macroscopically reproducing quantum-like behaviour (Bush [6]). If the drop does make contact, and begins coalescing, a huge variety of outcomes are possible, with a long history of study ([69, 90, 44]), including the potential for a daughter drop to be ejected as part of a coalescence cascade (Thoroddsen & Takehara [81]).
In this article, we are focused on the thresholds in speed (or relevant dimensionless number) at which the drop-bath contact behaviour changes. This threshold has normally been characterised in terms of the Weber number, We=ρdRU2/γd, where ρd is the drop’s density, R its radius, U the impact speed and γd its interfacial surface tension coefficient. This quantifies the ratio of kinetic energy to surface energy of the drop, and gives a measure of how ‘violent’ the impact is. As the impact velocity is increased, there is a critical Weber number where the outcome of the impact changes from bouncing to merging. Borrowing the language from the equivalent phenomenon in drop-drop collisions (Pan et al. [63]), we refer to this transition, from bouncing to merging with increasing impact velocity (i.e. increasing We), as the ‘hard’ transition (referred to as hard because it occurs at a higher kinetic energy (that is, higher We), and is therefore more ‘violent’, than other transitions that are observed). The critical Weber number of this transition has been identified in experiments to depend on many different things, including the ambient pressure of the gas (Schotland [72]). Interestingly, at lower impacts speeds, there is another transition, with the slowest drops merging and faster drops bouncing, named the ‘soft’ transition (named due its lower kinetic energy than the ‘hard’ transition), which was identified experimentally by Jayaratne et al. [41]. Recently, Pan & Law [62] and Tang et al. [78] identified that the critical Weber number of the hard transition depends on the depth of the bath, with remarkable transitions from merging-bouncing-merging-bouncing with increasing bath depth.
Following the observations of Walker [83], the ability of a drop to float for an extended period of time was observed using high-speed photography by Thoroddsen & Takehara (2000), who were able to measure the time that the drop was floating for. Klyuzhin et al. [46] showed that reducing the air pressure reduced the time that such drops would float for. More recently, Lo et al. (2017) observed drops floating above very thin liquid films, with small changes in the table angle having a dramatic effect on the floating time. As mentioned, Couder et al. [15] found that by vibrating the bath, drops can be made to float indefinitely.
Critical to the bounce-merge threshold is the dynamics of the gas film that is sandwiched, or ‘trapped’ between the drop and the bath. A key observation is that the gas film can be as thin as tens to hundreds of nanometres before contact occurs (Kavehpour [44]). These nano-scale gas films are not accurately described by the Navier-Stokes equations with no-slip boundary conditions, because the mean free path of the gas is comparable to the thickness of the film. Unconventional models are required, which incorporate gas kinetic effects that go beyond the Navier-Stokes paradigm. The general use of these models in drop-impact problems has been reviewed in Sprittles [74].
In the context of drop-bath impacts, Bach et al. [5], building on the work of Gopinath & Koch [28], used a lubrication model for the gas film based on the Boltzmann equation to show that the soft transition for very small drops was particularly sensitive to the mean free path of the gas. These were pioneering works in the study of drop impacts, but were limited to low impact speeds: their model used potential flow in the liquid domains, neglecting viscosity, and was limited to small deformations of the drop surface, assumptions that we have been able to relax. Lubrication models for the gas film between drop and bath have also been used to study the dynamics of drops within the bouncing regime (Phillips et al. [64]), and in the cushioning of the early stages of high-speed impacts of drops onto smooth solids (Mandre et al. [55]; Hicks & Purvis [33]; Moore 2021).
The problem of drops impacting liquid baths is closely linked to other drop-impact problems, where drops have been shown to bounce rather than merge with, or wet, the impacted surface due to a trapped gas film. A particularly relevant case is impacts of drops onto atomically smooth solids. These drop-solid impacts are much easier to image than drop-bath or drop-drop impacts, and various techniques, such as interferometry, have been used to accurately measure the trapped gas film (Driscoll & Nagel [20]; Kolinski et al. [47]; van der Veen et al. [82]), as reviewed in Josserand & Thoroddsen [43] and Sprittles [74]. More recently, the gas film in a drop-bath impact has also been measured by Tang et al. [79], finding similar dynamics for the evolution of the gas film to those seen for drop on solid impacts, with a “neck” at the edge of the gas film where the film is thinnest.
Notably, using pressure-controlled chambers, it has been observed in drop-bath impacts by Schotland [72] that critical Weber numbers of bounce-merge transitions depend dramatically on the ambient pressure. Such effects are now well-recognised for both drop-drop collisions (Qian & Law [67]; Zhang et al. [92]) and drop-solid ones (Zhang et al. [93]), and represent an important feature which theoretical models should expect to reproduce.
As well as gas kinetic effects, another phenomenon that is missing from conventional models but is critical to understanding how a drop makes contact is the van der Waals force between the drop and the bath (Israelachvili [40]). As the gas layer thins, it is this force, acting as a disjoining pressure, that ultimately drives the liquid surfaces together. This has been discussed as the cause of contact as early as Jayaratne et al. [41], and its importance was highlighted in the context of drop-drop collisions by Pan et al. [63].
In this article, we present a model for drop-bath impacts that is able to predict, for the first time, the transitions between bouncing and merging. This uses a lubrication equation to model the gas film, which incorporates gas kinetic effects and a van der Waals-driven disjoining pressure. This is coupled to the free-surface Navier-Stokes equations for the flow of the drop and the bath. Together these are solved with an arbitrary Eulerian-Lagrangian finite-element method. This builds on our previous models for drop-drop collisions (Lewin-Jones et al. [50]) and drop-solid impacts (Chubynsky et al. [13]), which in turn built on the work of Li [51] for drop-drop collisions.
The study of drop-bath impacts, unlike these previous studies, is more challenging as one cannot rely on planar symmetry, applicable for the head-on collisions of equal-sized drops, so that two distinct liquid domains have to be simultaneously resolved. Furthermore, the gas film geometry becomes significantly curved, so that we are forced to reformulate the lubrication equation with respect to a curved coordinate system that moves with the deforming drop surface, and provides the stress that is applied to the surfaces of the drop and the bath. Only with a curved lubrication model that is embedded with nanoscale physics are we able to accurately predict the bounce-merge threshold.
To confirm the accuracy of our new model, we have also performed new experiments, using the methodology and performed alongside those of Harris et al. [31], to measure the critical Weber number for the hard transition of drop-bath collisions of silicone oil at a range of viscosities and droplet radii.
This article is structured as follows. In § 2, we introduce our model for drop-bath impact, that requires us to formulate a curved lubrication equation that sits within a finite element framework to model the trapped gas film, incorporating the nano-scale van der Waals and gas kinetic effects. This model is then benchmarked in § 2.8 against volume-of-fluid simulations for the case in which the drop does not merge. The experimental methodology is described in § 3. In § 4 we compare simulations of the new model to experiments of impacts onto a deep bath, and then explore beyond the current experimental regime to make predictions of the critical Weber numbers for low speed impacts and understand the behaviour at reduced ambient air pressure. In § 5, we consider impacts onto shallow baths, comparing to the experiments of Tang et al. [78] and explore the limit of vanishing bath depth. Finally, in § 6 we conclude and discuss future directions for this methodology and the investigation of drop impacts.
Methodology: Modelling Drop-Bath Impacts
Consider the impact of a liquid drop onto a liquid bath. A thin layer of the surrounding gas is trapped between them, which we model using lubrication theory, exploiting the small ratio between the thickness of the gas layer and the size of the drop. We assume that there is no drop or surface deformation until the lubrication approximation is valid, and from thereon it is only at the region of the lubrication model where the drop and bath directly interact. In § 2.1 we introduce the models for the drop and the bath. In § 2.2, we describe the model for the gas film, how the (curved) trapped gas film is described and the lubrication equation that is solved, with a full derivation in appendix B. Next, we describe the additional physics, beyond those of classical fluid dynamics, that are required for accurate modelling: gas kinetic effects in § 2.4 and disjoining pressure in § 2.5. We introduce the numerical implementation of our new model, using the finite element method, in § 2.7. In § 2.8 we benchmark our new model against an established volume-of-fluid method, and in § 3 describe the experimental methodology.
Drop and Bath Models
We work in a cylindrical coordinate system (r,z,θ), shown in Figure 1, with axisymmetry always assumed, so there is no dependence of quantities on θ and there is no velocity in the θ-direction. This is a natural setup for such impacts as gravity tends to align drops to fall ‘down’ and ‘flatten’ liquid baths/layers. Unit basis vectors in the r- and z-directions are denoted by er and ez, respectively, with gravity in the −ez direction. We will use subscripts r and z to denote components of vectors in the r and z directions (for example, u=urer+uzez).
Figure 1. Initial conditions of our model, with a drop of radius R, with base at a height of hoffset=0.1R and initial downward speed U impacting an initially stationary bath of depth Hb and radius Rb (not drawn to scale).
Initially, the drop is a sphere of radius R, with uniform velocity −Uez, as shown in Figure 1 (this assumes that any oscillations in the drop, from its formation, have had time to settle). We will use superscripts of d, b and g to denote quantities associated with the drop, bath and gas respectively. The dynamic viscosity of the liquid drop is μd, and its density is ρd. The base of the upper drop starts from height hoffset=0.1R, which is far enough for the gas lubrication pressure to be negligible (see the appendix of Lewin-Jones et al. [50], which discusses the equivalent assumptions for drop-drop collisions). The bath, with dynamic viscosity, μb, and density, ρb, has a solid floor and wall, with depth, Hb, and radius, Rb, and initially is at rest, with a horizontal free surface. Although the model is described in a form that allows different liquid properties in the bath and the drop, all of the results presented in this article use the same liquid for both, leaving such variations (Fudge et al. [23]; Wu & Saha 2022) as an avenue for future work.
The Navier-Stokes equations (1)–(2)) are solved for the velocity ud, and pressure, pd, in the liquid drop at time t, and are given dimensionally as
ρdDtDud=−∇pd+μd∇2ud−ρdgez,(1)
0=∇⋅ud,(2)
where g is the acceleration due to gravity. The corresponding Navier-Stokes equations (3)–(4) are solved for the bath, with superscript b replacing d:
ρbDtDub=−∇pb+μb∇2ub−ρbgez,(3)
0=∇⋅ub.(4)
On the surface of the drop, where nd is the outer unit normal to the drop, the dynamic boundary condition is
Td⋅nd=Tg⋅nd−γdnd∇s⋅nd+Fd,(5)
where Td is the (drop) stress tensor, ∇s⋅ is the surface divergence (see (C 42) in appendix §C.3) and γd is the constant surface tension coefficient of the drop-gas interface. The stress tensor in the drop is given by
Td=−pdI+μd(∇ud+(∇ud)T).(6)
Similarly, Tg is the gas stress tensor. There is also a surface force, Fd which models the van der Waals driven disjoining pressure between the drop and the bath, and will be discussed in §2.5. Correspondingly, the dynamic boundary condition on the bath surface is
Tb⋅nb=Tg⋅nb−γbnb∇s⋅nb+Fb,(7)
with Tb, γb and Fb being the bath stress tensor, bath-gas surface tension coefficient and bath surface force respectively.
At r=0, the boundary conditions for the drop and the are given by symmetry as ur=0 and ∂r∂uz=0, where here (ur,uz) is the velocity of the drop and bath respectively. The bath initially occupies the region from z=0 to z=−Hb, and from r=0 to r=Rb. At z=−Hb (the ‘floor’ of the bath), we apply the no-slip boundary condition, ub=0. At r=Rb (the ‘wall’ of the bath), a no penetration condition, urb=0, is applied, as well as a no shear condition, er⋅Tb⋅ez=0, and a 90° contact angle is prescribed at the boundary of the free surface. This choice of boundary condition on the bath wall allows the free-surface to wet/dewet the side, as waves produced by the impact encounter it, without any moving-contact-line issues. In this article Rb will always be chosen so that the edge of the bath is far from the impacting drop, and does not affect whether the drop bounces or merges with the bath; the effect of this will be discussed in § 4.3 and would be an interesting path of future investigation.
Modelling the Gas Film
Previous research for drop impact onto smooth solids ([13]), heated solids in the Leidenfrost regime ([11]), and the head-on collision of two identical drops ([50]) has successfully used lubrication theory to model the trapped gas film, coupled to the Navier-Stokes equations used to model the dynamics in the drop. Lubrication theory has also been coupled to various reduced-order models for the drop ([5, 55, 33, 64] (Moore 2021)). In all of these cases the film is horizontal, and the lubrication height was measured vertically, in an Eulerian sense, so that the film height is a function of r.
However, for drop-bath impacts, especially at higher impact speeds, the bath is deformed considerably, and we cannot apply the horizontal model, as in Lewin-Jones et al. [50], because the vertical distance between the drop and the bath will not be an accurate measurement of the thickness of the gas film. Among other problems, once the bath deforms significantly a lubrication equation based on the vertical height will no longer preserve mass correctly (as this does not correctly measure change in the volume of gas), and the magnitude of the van der Waals force, which depends strongly on height, will be too small. The solution is to replace the horizontal lubrication equation with a model that treats the film thickness as the distance between the drop and the bath in the direction normal to a free surface rather than vertically, and derive a lubrication equation where the governing equations of the gas film are averaged over that ‘normal’ thickness. This is shown in Figure 2, where hV shows how the film height would be measured vertically, compared to h which is how the model in this article is developed.
Figure 2. Two possible definitions for the height of the gas film. Denoted by h is the height measured in the normal direction to the drop, as used by the model in this article. Denoted by hV is the vertical height of the film, which is unsuitable for this problem due to the large deformation of the bath.
The model we have developed describes a lubricating film between two axisymmetric curved surfaces. The general approach is to stop solving for the gas dynamics in the fixed Eulerian frame (e.g. the pressure is solved as a function of r) and instead fit a coordinate system to the moving surface of the drop. Lubrication theory then provides both the pressure and shear stress terms for the boundary conditions (5) and (7).
We have used the approach of Howell [36], originally developed for thin viscous liquid sheets, to consider the governing equations of the gas in a curved, time evolving coordinate system, and applied lubrication theory, integrating across the film in the normal direction to the drop, to obtain a single partial differential equation for the gas pressure. However, the thin film in Howell [36] is a sheet with free surfaces, rather than our setup consisting of a film trapped between two moving surfaces. To calculate the time evolution the film height, we have adapted the method of Temizer & Stupkiewicz [80], who derived a general lubrication equation between two curved surfaces, without any symmetry assumptions.
The choice of film height definition is not unique, as, for example, we could equally work on the bath surface, and define the film height in terms of the normal projection from the bath. It is also possible to formulate the lubrication based on a ‘lubrication surface’ that lies between the two bounding surfaces, with the film height defined as normal projection from either the boundary surfaces or the intermediate lubrication surface. These possibilities are all compared in Temizer & Stupkiewicz [80], but as long as the surfaces are approximately locally parallel (as they will be for the lubrication approximation to be valid), the differences between these formulations will be asymptotically small.
Curved Lubrication Equation
A full derivation of the lubrication equation is available in appendix § 9, which we summarise here. We parametrise the axisymmetric surface of the drop by curvilinear coordinate s and azimuthal coordinate θ. Axisymmetry means that the position vector at s and θ, at time t, can be written as xd(s,θ,t)=rd(s,t)er+zd(s,t)ez. Note that in Cartesian coordinates er depends on θ, but dependencies on θ will be unstated except where it is important. Correspondingly the bath surface is parametrised by ξ and θ.
The point of intersection of the drop normal nd, with the bath surface is denoted by xb, and we denote its curvilinear surface coordinate by ξˉ, as shown in Figure 3. Throughout this section, symbols with over-bars denote variables evaluated at the point of projection (this is described in more detail in appendix § 9.§ 2). Because of the axisymmetry, xb is at the same azimuthal coordinate θ as xd. The film thickness is then denoted by h=(xb−xd)⋅nd, so that hnd=xb−xd, as shown in Figure 3.
Figure 3. The definition of the film height h at drop position vector xd(s,t). The height is measured in the direction normal to the drop’s free surface. The intersection of the projected normal on the bath surface is denoted by xb, at coordinate ξˉ.
References
[1]ALBERTONI, SERGIO, CERCIGNANI, CARLO & GOTUSSO, LAURA 1963 Numerical Evaluation of the Slip Coefficient. Physics of Fluids 6 (7), 993.DOI
[2]ALVENTOSA, LUKE F. L., CIMPEANU, RADU & HARRIS, DANIEL M. 2023 Inertio-capillary rebound of a droplet impacting a fluid bath. Journal of Fluid Mechanics 958, A24.DOI
[3]ANTHONY, CHRISTOPHER R., WEE, HANSOL, GARG, VISHrUT, THETE, SUMEET S., KAMAT, PRITISH M., WAGONER, BRAYDEN W., WILKES, EDWARD D., NOTZ, PATRICK K., CHEN, ALVIN U., SURYO, RONALD, SAMBATH, KRISHNARAJ, PANDITARATNE, JAYANTA C., LIAO, YING-CHIH & BASARAN, OSMAN A. 2023 Sharp Interface Methods for Simulation and Analysis of Free Surface Flows with Singularities: Breakup and Coalescence. Annual Review of Fluid Mechanics 55 (1), 707–747.DOI
[4]AOKI, KAZUO, TAKATA, SHIGERU, TATSUMI, ERI & YOSHIDA, HIROAKI 2010 Rarefied gas flows through a curved channel: Application of a diffusion-type equation. Physics of Fluids 22 (11), 112001.DOI
[5]BACH, GLORIA A., KOCH, DONALD L. & GOPINATH, ARVIND 2004 Coalescence and bouncing of small aerosol droplets. Journal of Fluid Mechanics 518, 157–185.DOI
[6]BUSH, JOHN W. M. 2010 Quantum mechanics writ large. Proceedings of the National Academy of Sciences of the United States of America 107 (41), 17455–17456.DOI
[7]CAIRNCROSS, RICHARD A., SCHUNCK, P. RANDALL, BAER, THOMAS A., RAO, REKHA R. & SACKINGER, PHILIP A. 2000 A finite element method for free surface flows of incompressible fluids in three dimensions. Part I. Boundary fitted mesh motion. International Journal for Numerical Methods in Fluids 33 (3), 375–403.doi.org/10.1002/1097-0363(20000615)33:3<375::aid-fld13>3.0.co;2-o
[9]CERCIGNANI, CARLO, LAMPIS, MARIA & LORENZANI, SILVIA 2004 Plane Poiseuille Flow with Symmetric and Nonsymmetric Gas-Wall Interactions. Transport Theory and Statistical Physics 33 (5-7), 545–561.DOI
[11]CHAKRABORTY, INDRAJIT, CHUBYNSKY, MYKYTA V. & SPRITTLES, JAMES E. 2022 Computational modelling of Leidenfrost drops. Journal of Fluid Mechanics 936, A12.DOI
[12]CHEN, YINGYING, QUAN, XIAOJUN, WANG, ZHONGYONG, LEE, CHIAHSUN, WANG, ZIZHAO, TAO, PENG, SONG, CHENGYI, WU, JIANBO, SHANG, WEN & DENG, TAO 2016 Stably dispersed high-temperature Fe3O4/silicone-oil nanofluids for direct solar thermal energy harvesting. Journal of Materials Chemistry A 4 (44), 17503–17511.DOI
[13]CHUBYNSKY, MYKYTA V., BELOUSOV, KIRILL I., LOCKERBY, DUNCAN A. & SPRITTLES, JAMES E. 2020 Bouncing off the Walls: The Influence of Gas-Kinetic and van der Waals Effects in Drop Impact. Physical Review Letters 124 (8), 084501.
[15]COUDER, Y., FORT, E., GAUTIER, C.-H. & BOUDAOUD, A. 2005 From Bouncing to Floating: Noncoalescence of Drops on a Fluid Bath. Physical Review Letters 94 (17), 177801.pubmed.ncbi.nlm.nih.gov/15904334
[17]DAI, BING, LEAL, L. GARY & REDONDO, ANTONIO 2008 Disjoining pressure for nonuniform thin films. Physical Review E 78 (6), 061602.pubmed.ncbi.nlm.nih.gov/19256848
[19]DONEA, JEAN, HUERTA, ANTONIO, PONTHOT, J.-Ph. & RODRÍGUEZ-FERRAN, A. 2004 Arbitrary Lagrangian– Eulerian Methods. In Encyclopedia of Computational Mechanics, chap. 14. John Wiley & Sons, Ltd.
[20]Driscoll, Michelle M. & Nagel, Sidney R. 2011 Ultrafast Interference Imaging of Air in Splashing Dynamics. Physical Review Letters 107 (15), 154502.DOI
[22]Fudge, Ben D., Cimpeanu, Radu, Antkowiak, Arnaud, Castrejón-Pita, J. Rafael & Castrejón-Pita, Alfonso A. 2023 Drop splashing after impact onto immiscible pools of different viscosities. Journal of Colloid and Interface Science 641, 585–594.DOI
[23]Fudge, Ben D., Cimpeanu, Radu & Castrejón-Pita, Alfonso A. 2021 Dipping into a new pool: The interface dynamics of drops impacting onto a different liquid. Physical Review E 104 (6), 065102.DOI
[24]Fukui, S. & Kaneko, R. 1988 Analysis of Ultra-Thin Gas Film Lubrication Based on Linearized Boltzmann Equation: First Report—Derivation of a Generalized Lubrication Equation Including Thermal Creep Flow. Journal of Tribology 110 (2), 253–261.DOI
[25]Fukui, Shigehisa, Shimada, Hidekazu, Yamada, Yamane, Kiyomi & Matsuoka, Hiroshige 2005 Flying characteristics in the free molecular region (influence of accommodation coefficients). Microsystem Technologies 11 (8), 805–811.DOI
[26]Gabbard, Chase T., Aguerre, Elvis A., Cimpeanu, Radu, Kuehr, Katharina, Silver, Eli, Barotta, Jack-William, Galeano-Rios, Carlos A. & Harris, Daniel M. 2025 Drop rebound at low Weber number. Journal of Fluid Mechanics 1019, A25.DOI
[27]Galeano-Rios, Carlos A., Cimpeanu, Radu, Bauman, Isabelle A., MacEwen, Annika, Mlewski, Paul A. & Harris, Daniel M. 2021 Capillary-scale solid rebounds: Experiments, modelling and simulations. Journal of Fluid Mechanics 912, A17.
[28]Gopinath, Arvind & Koch, Donald L. 2001 Dynamics of droplet rebound from a weakly deformable gas–liquid interface. Physics of Fluids 13 (12), 3526–3532.DOI
[29]Hamaker, H. C. 1937 The London—van der Waals attraction between spherical particles. Physica 4 (10), 1058–1072.DOI
[30]Harris, D.M., Liu, T. & Bush, J.W.M. 2015 A low-cost, precise piezoelectric droplet-on-demand generator. Experiments in Fluids 56, 1–7.DOI
[31]Harris, Daniel M., Alventosa, Luke, Sand, Oliver, Silver, Eli, Mohammadi, Arman, Sykes, Thomas C., Castrejón-Pita, Alfonso A. & Cimpeanu, Radu 2026 Bouncing to coalescence transition for droplet impact onto moving liquid pools. Journal of Fluid Mechanics 1030, A16.arxiv.org/abs/2510.02220
[32]Heil, Matthias, Hazel, Andrew & Matharu, Puneet 2022 Oomph-lib.
[33]Hicks, Peter D. & Purvis, Richard 2011 Air cushioning in droplet impacts with liquid layers and other droplets. Physics of Fluids 23 (6), 062104.DOI
[34]Hirt, C. W & Nichols, B. D 1981 Volume of fluid (VOF) method for the dynamics of free boundaries. Journal of Computational Physics 39 (1), 201–225.DOI
[35]Hocking, L. M. 1973 The effect of slip on the motion of a sphere close to a wall and of two adjacent spheres. Journal of Engineering Mathematics 7 (3), 207–221.DOI
[36]Howell, P 1994 Extensional thin layer flows. PhD thesis, Oxford.DOI
[37]Huang, Kuan-Ling & Pan, Kuo-Long 2021 Transitions of bouncing and coalescence in binary droplet collisions. Journal of Fluid Mechanics 928, A7.DOI
[38]Ionkin, N. & Harris, D.M. 2018 Note: A versatile 3D-printed droplet-on-demand generator. Review of Scientific Instruments 89 (11).DOI
[40]Israelachvili, Jacob N. 2011 Intermolecular and Surface Forces, 3rd edn. Burlington, MA: Academic Press.
[41]Jayaratre, O. W., Mason, Basil John, Bartlett, J. T. & Blackett, Patrick Maynard Stuart 1964 The coalescence and bouncing of water drops at an air/water interface. Proceedings of the Royal Society of London. Series A. Mathematical and Physical Sciences 280 (1383), 545–565.
[43]Josserand, C. & Thoroddsen, S.T. 2016 Drop Impact on a Solid Surface. Annual Review of Fluid Mechanics 48 (1), 365–391.DOI
[44]Kavehpour, H. Pirouz 2015 Coalescence of Drops. Annual Review of Fluid Mechanics 47, 245–268.DOI
[45]King, J. R. C. & Lind, S. J. 2019 The Kaye effect: New experiments and a mechanistic explanation. Journal of Non-Newtonian Fluid Mechanics 273, 104165.DOI
[46]Klyuzhin, Ivan S., Jenna, Federico, Roeder, Brandon, Wexler, Adam & Pollack, Gerald H. 2010 Persisting Water Droplets on Water Surfaces. The Journal of Physical Chemistry B 114 (44), 14020–14027.DOI
[47]Kolinski, John M., Rubinstein, Shmuel M., Mandre, Shreyas, Brenner, Michael P., Weitz, David A. & Mahadevan, L. 2012 Skating on a Film of Air: Drops Impacting on a Surface. Physical Review Letters 108 (7), 074503.pubmed.ncbi.nlm.nih.gov/22401209
[48]Kreyszig, Erwin 2020 Advanced Engineering Mathematics. John Wiley & Sons.
[49]Lewin-Jones, Peter 2024 Liquid Drop Impacts: When Trapped Gases Nanofilms Lead to Bouncing. PhD thesis, University of Warwick.
[50]Lewin-Jones, Peter, Lockerby, Duncan A. & Sprittles, James E. 2024 Collision of liquid drops: Bounce or merge? Journal of Fluid Mechanics 995, A1.
[51]Li, Jie 2016 Macroscopic Model for Head-On Binary Droplet Collisions in a Gaseous Medium. Physical Review Letters 117 (21), 214502.pubmed.ncbi.nlm.nih.gov/27911562
[52]Liang, Gangtao & Mudawar, Issam 2017 Review of spray cooling – Part 1: Single-phase and nucleate boiling regimes, and critical heat flux. International Journal of Heat and Mass Transfer 115, 1174–1205.DOI
[54]Lohse, Detlef 2022 Fundamental Fluid Dynamics Challenges in Inkjet Printing. Annual Review of Fluid Mechanics 54 (1), 349–382.DOI
[55]Mandre, Shreyas, Mani, Madhav & Brenner, Michael P. 2009 Precursors to Splashing of Liquid Droplets on a Solid Surface. Physical Review Letters 102 (13), 134502.pubmed.ncbi.nlm.nih.gov/19392358
[56]Marcotte, Florence, Michon, Guy-Jean, Séon, Thomas & Josserand, Christophe 2019 Ejecta, Corolla, and Splashes from Drop Impacts on Viscous Fluids. Physical Review Letters 122 (1), 014501.DOI
[57]Moon, Joo Hyun, Choi, Chang Kyoung, Allen, Jeffrey S. & Lee, Seong Hyuk 2018 Observation of a mixed regime for an impinging droplet on a sessile droplet. International Journal of Heat and Mass Transfer 127, 130–135.DOI
[60]Oron, Alexander, Davis, Stephen H. & Bankoff, S. George 1997 Long-scale evolution of thin liquid films. Reviews of Modern Physics 69 (3), 931–980.DOI
[61]Pack, M., Hu, H., Kim, D., Zheng, Z., Stone, H. A. & Sun, Y. 2017 Failure mechanisms of air entrainment in drop impact on lubricated surfaces. Soft Matter 13 (12), 2402–2409.
[62]Pan, K. L. & Law, C. K. 2007 Dynamics of droplet–film collision. Journal of Fluid Mechanics 587, 1–22.
[63]Pan, Kuo-Long, Law, Chung K. & Zhou, Biao 2008 Experimental and mechanistic description of merging and bouncing in head-on binary droplet collision. Journal of Applied Physics 103 (6), 064901.DOI
[64]Phillips, K. A., Cimpeanu, R. & Milewski, P. A. 2025 Modelling two-dimensional droplet rebound off deep fluid baths. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences 481 (2317), 20240956.arxiv.org/abs/2406.16750
[65]Popinet, Stéphane 2015 A quadtree-adaptive multigrid solver for the Serre–Green–Naghdi equations. Journal of Computational Physics 302, 336–358.DOI
[66]Poydenot, Florian & Andreotti, Bruno 2024 Gap in drop collision rate between diffusive and inertial regimes explains the stability of fogs and non-precipitating clouds. Journal of Fluid Mechanics 987, A37.DOI
[67]Qian, J. & Law, Chung K. 1997 Regimes of coalescence and separation in droplet collision. Journal of Fluid Mechanics 331, 59–80.DOI
[68]Ramírez-Soto, Olinka, Sanjay, Vatsal, Lohse, Detlef, Pham, Jonathan T. & Vollmer, Doris 2020 Lifting a sessile oil drop from a superamphiphobic surface with an impacting one. Science Advances 6 (34), eaba4330.
[69]Rein, Martin 1993 Phenomena of liquid drop impact on solid and liquid surfaces. Fluid Dynamics Research 12 (2), 61–93.DOI
[70]Reynolds, Osborne 1881 On the floating of drops on the surface of water. Proceedings of the Literary and Philosophical Society of Manchester 21 (1), 413–414.DOI
[71]Sanjay, Vatsal, Lakshman, Srinath, Chandelot, Pierre, Snoeijer, Jacco H. & Lohse, Detlef 2023 Drop impact on viscous liquid films. Journal of Fluid Mechanics 958, A25.DOI
[72]Schotland, R. M. 1960 Experimental results relating to the coalescence of water drops with water surfaces. Discussions of the Faraday Society 30 (0), 72–77.DOI
[73]Sharma, Praveen K. & Dixit, Harish N. 2021 Regimes of wettability-dependent and wettability-independent bouncing of a drop on a solid surface. Journal of Fluid Mechanics 908, A37.
[74]Sprittles, James E. 2024 Gas Microfilms in Droplet Dynamics: When Do Drops Bounce? Annual Review of Fluid Mechanics 56 (1), 91–118.DOI
[75]Sundararajakumar, R. R. & Koch, Donald L. 1996 Non-continuum lubrication flows between particles colliding in a gas. Journal of Fluid Mechanics 313, 283–308.
[77]Tang, X., Saha, A., Law, C.K. & Sun, C. 2018 Bouncing-to-merging transition in drop impact on liquid film: Role of liquid viscosity. Langmuir 34 (8), 2654–2662.
[78]Tang, Xiaoyu, Saha, Abhishek, Law, Chung K. & Sun, Chao 2016 Nonmonotonic response of drop impacting on liquid film: Mechanism and scaling. Soft Matter 12 (20), 4521–4529.
[79]Tang, Xiaoyu, Saha, Abhishek, Law, Chung K. & Sun, Chao 2019 Bouncing drop on liquid film: Dynamics of interfacial gas layer. Physics of Fluids 31 (1), 013304.
[80]Temizer, İ. & Stupkiewicz, S. 2016 Formulation of the Reynolds equation on a time-dependent lubrication surface. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences 472 (2187), 20160032.
[81]Thorodsen, S. T. & Takehara, K. 2000 The coalescence cascade of a drop. Physics of Fluids 12 (6), 1265–1267.
[82]van der Veen, Roeland C. A., Tran, Tuan, Lohse, Detlef & Sun, Chao 2012 Direct measurements of air layer profiles under impacting droplets using high-speed color interferometry. Physical Review E 85 (2), 026315.
[83]Walker, Jearl 1978 Drops of liquid can be made to float on the liquid. What enables them to do so? Scientific American 238 (6), 151–159.
[84]Wang, Ning, Zhang, Zhenyu, Zhang, Peng & Zhao, Changlu 2025 Towards quantitative prediction of droplet collision outcomes: A dual-VOF approach with rarefied gas effect and augmented van der Waals force. International Journal of Multiphase Flow 188, 105207.
[85]Wang, Ning, Zhang, Zhenyu, Zhang, Peng & Zhao, Changlu 2026 Direct numerical simulation of colliding droplets with intervening nonequilibrium gas film. Acta Mechanica Sinica 42 (2), 325492.
[86]Wanninkhof, Rik, Asher, William E., Ho, David T., Sweeney, Colm & McGillis, Wade R. 2009 Advances in Quantifying Air-Sea Gas Exchange and Environmental Forcing*. Annual Review of Marine Science 1 (Volume 1, 2009), 213–244.pubmed.ncbi.nlm.nih.gov/21141036
[87]Worthington, Arthur Mason 1908 A Study of Splashes. Longmans, Green, and Company.
[92]Zhang, Chenwei, Xu, Zhigang, Zhou, Jinhui & Zhang, Zhenyu 2026 A droplet collision model for bouncing–coalescence boundary of droplets under different ambient pressures. Physics of Fluids 38 (1), 017120.
[93]Zhang, Lige, Soori, Tejaswi, Rokoni, Arif, Kaminski, Allison & Sun, Ying 2021 Air film contact modes of drop impact on lubricated surfaces under reduced pressures. Physics of Fluids 33 (9), 092110.
[94]Zhang, Lige, Soori, Tejaswi, Rokoni, Arif & Sun, Ying 2022b Postcontact droplet spreading and bubble entrapment on a smooth surface. Physical Review Fluids 7 (10), 104003.
[95]Zienkiewicz, O. C. & Zhu, J. Z. 1987 A simple error estimator and adaptive procedure for practical engineering analysis. International Journal for Numerical Methods in Engineering 24 (2), 337–357.DOI
Julia Müller, Jaron Grigat, Robin Glade-Beucke, Sebastian Lindemann, Tiffany Luce, Gnanesh Chandra Madduri, Jens Reininghaus, Marc Schumann, Adam Softley-Brown, Andrew Stevens ·