Droplet Wetting and Evaporation: From Pure to Complex Fluids

1806;LXIII.

Section I

Wetting and Spreading

Outline

Chapter 1 Liquid Spreading

Chapter 2 Contact Angles and the Surface Free Energy of Solids

Chapter 3 Triple Line Motion and Evaporation

Chapter 4 Precursor Films and Contact Line Microstructures

Chapter 5 Superamphiphobic Surfaces

Chapter 6 Superspreading

Chapter 7 Leidenfrost Drops

Chapter 1

Liquid Spreading

Odile Carrier and Daniel Bonn,    Institute of Physics, University of Amsterdam, Amsterdam, The Netherlands

This section addresses the physics of liquid spreading from an energetics point of view. The different modes of spreading are present and several fluid natures (pure or complex) are investigated.

Keywords

Total spreading; surface tension; contact angle; surface energy

This section addresses the physics of liquid spreading from an energetics point of view. The different modes of spreading are presented and several fluid natures (pure or complex) are investigated.

1.1 Theoretical concepts: equilibrium surface thermodynamics

Capillary phenomena have been studied for centuries, and the key quantity to consider in equilibrium situations is the excess free energy per unit area associated with the formation of an interface between two phases. In order to define the excess free energy, or surface tension, it is common to work in the grand canonical ensemble, fixing the chemical potential rather than the total number of molecules present in the system. This is done because the interface can be viewed as an open system; molecules can move between the bulk and interfacial regions, so the number of particles in the interfacial region is not fixed. The particles in the interfacial region must, in addition, be in equilibrium—at a fixed chemical potential—with the particles of the same species in either of the bulk phases (Figure 1.1).

Figure 1.1 Droplets on plant leaves. Because the plant leaves clearly repel water, it is difficult to get the droplets to stick to the leaves so, for example, pesticides can be applied.

, with F the Helmholtz free energy, μ the chemical potential, and N the number of molecules. The surface tension σ follows from the differential:

(1.1)

with A the surface area, S the entropy, T the temperature, p the pressure, and V .

Consider, now, a liquid drop on a solid substrate, as depicted in Figure 1.2. Because three different phases are present, three surface tensions must be considered: solid–liquid σsl, liquid–vapor σlv, and solid–vapor σsv. The relation between the cosine of the so-called contact angle θ and the three surface tensions is given by the Young equation (1805):

(1.2)

Figure 1.2 Force balance at the three-phase contact line. The mechanical equilibrium for this system is equivalent to the Young equation.

As shown in Figure 1.2, this equation can be interpreted as a force balance for the line of three-phase contact. It simply expresses the mechanical equilibrium of the contact line. The unit of the surface tension (J m−2) is an energy per unit area, but it can also be interpreted as a force per unit length (N m−1) that acts on the three-phase contact line.

Wetting can be viewed as a direct consequence of the Young equation. The situation where a droplet is present on the surface is commonly called the partial wetting state, because the surface is only partially covered with the liquid, as depicted in Figure 1.3A. If we now change some thermodynamic variable—say, the temperature—the three surface tensions will change, and, consequently, the contact angle will also change. At a certain point, it may be that the sum of the solid–liquid and the liquid–vapor surface tensions becomes equal to the solid–vapor interfacial tension. The contact angle will then be zero, and the system will be in equilibrium when a uniform film covers the whole solid surface. In the usual terminology, we say that the liquid completely wets the surface. This is depicted in Figure 1.3B, where the complete wetting state is shown.

Figure 1.3 Wetting of a liquid drop on a solid substrate. (A) Partial wetting state. (B) Complete wetting state. (C) Complete drying state.

Analogously, it may be that, by changing the temperature, at a certain point the contact angle reaches 180°—that is, the sum of the solid–vapor and the liquid–vapor equals the solid–liquid interfacial tension. Then, the liquid does not wet the solid and, in equilibrium, remains separated from it by a film of vapor. This situation is depicted in Figure 1.3C and is, quite logically, called the complete drying state. Because complete drying is in fact completely analogous to complete wetting—but with the roles of the liquid and vapor reversed—in the following we will just consider wetting.

The wetting transition is a transition between a partial state and a complete wetting state. If the system is studied as a function of temperature, the transition temperature is called the wetting temperature Tw. The order of the wetting transition is determined—in the same manner as for a bulk phase transition—by the discontinuities in derivatives of the surface free energy. If a discontinuity occurs in the first derivative of the free energy, the transition is said to be of first order and will take place in a discontinuous way. On the other hand, if the first derivative of the free energy is continuous through a phase transition point, then that indicates that it is a higher-order or critical phase transition.

The distinction between the different wetting states is usually made by considering the spreading coefficient S. The spreading coefficient is thus negative for the partial wetting state. Because the sum of the solid–liquid and the liquid–vapor tensions is larger than the solid–vapor surface tension, it is unfavorable to replace the solid–vapor interface with the two others.

An important remark is that for complete wetting, the equilibrium spreading coefficient is zero. This is because, at complete wetting, the solid–vapor interface, at equilibrium, is covered with a macroscopically thick wetting layer, so its tension is equal to the sum of the solid–liquid and liquid–vapor surface tensions. Simply said, the solid–vapor interface ceases to exist, because it is no longer stable thermodynamically. Consequently, positive spreading coefficients, sometimes reported in the literature, by definition consider nonequilibrium situations. A much-studied example of this is the spreading of a droplet (see, for instance, the review of de Gennes, 1985, and references therein). Although the initial spreading coefficient may be positive in this situation, because it is the driving force of the spreading, once the droplet has spread to cover the whole surface, the equilibrium spreading coefficient will be zero.

It is important to note here that the partial wetting state does not correspond to the complete absence of molecules in the fluid that does the wetting besides the droplet. This is, of course, because of entropy: the presence of a few molecules of the wetting fluid at the substrate—for instance, as a two-dimensional gas—lowers the free energy by kBT per molecule. Even if the interactions do not favor the presence of many of these molecules at the surface, as is the case for partial wetting, a few molecules will always be present (although it should be kept in mind that it is possible to have a depletion from the surface and thus a negative surface excess density); therefore, we speak of the presence of a microscopic film.

1.2 Spreading dynamics: pure fluids

If a drop is placed on a solid surface, it will in general be far from its equilibrium state—that is, in terms of the extent to which the free energy in the initial spreading coefficient is different from the equilibrium one. In this case, a hydrodynamic flow occurs until the equilibrium contact angle is reached, provided the drop is not trapped in some metastable state. The hydrodynamics of this problem has been studied extensively. The simplest case is that of complete wetting (Seq=0); here, the drop will spread almost forever, presumably until it reaches a thickness set by its own volume and possibly also the van der Waals forces. If the equilibrium contact angle is finite but smaller than the one just after deposition of the drop, the initial stage of the dynamics is similar: the drop will spread out but will eventually go to its equilibrium state given by the Young equation with a finite contact angle.

The dynamics of droplets have attracted a great deal of attention over the past few years and have been studied by many different groups. The spreading of drops is a prototypical singularity problem: a hydrodynamic singularity is at the moving contact line, where the viscous stresses diverge and need to be regularized in the lubrication equations in order to be able to predict the spreading rate. The problem is complicated by the fact that the form of the droplet during the spreading is a priori unknown. For a perfectly wetting liquid on a perfectly flat surface, the effects of the anchoring of the contact line are small. When you deposit a drop, because the liquid is perfectly wetting, the drop starts to spread on the surface and its radius increases in time. Figure 1.4 shows a few typical curves of the radius as a function of time for different alkanes spreading. These experiments were done by gently depositing drops on a glass surface and photographing the time evolution of the drop’s radius with Kruss Easydrop, which performs automated image recognition and fits the drop image to a spherical cap. The volume of the droplets was about 1 µL. The largest radius these droplets attained was about 2 to 3 mm, somewhat larger than the capillary length (1 mm).

Figure 1.4 Spreading of 1 µL nonane (black squares) and dodecane (red dots) droplets on a glass microscope slide. The lines are fits to a power-law following t⁰.¹.

The key observation from this (and a lot of earlier) experiment is that the increase in drop radius is very small and is strongly sublinear in time. One might have (naively) anticipated a linear dependence of the radius in time. A simple dimensional analysis would (wrongly) assume that the spreading coefficient is the driving force, which is countered by viscous forces. This would lead to a linear increase of the drop radius in time, with a slope on the order of the capillary speed, which is on the order of ~10 m s−1 for low-viscosity liquids such as the alkanes discussed above.

Clearly, this is not the case, and the spreading is much, much slower. This happens for two reasons. First, the driving force is not the spreading coefficient (or, more precisely, the difference between the initial and equilibrium one), because this surface energy is lost in the development of the microscopic film discussed in Section 1.1. The existence of this film, called precursor film for the case of spreading, changes notably the solid–vapor surface energy. It does so in such a way that if, for instance, the equilibrium state is complete wetting (zero contact angle), there is no driving force for the spreading from the free energy difference between the partial and complete wetting states. The driving force for the spreading is then only the (much weaker) capillary pressure within the drop, given by the surface tension of the liquid–vapor interface multiplied by the curvature of that same