PHYSICAL REVIEW E 89, 013023 (2014) Droplet-air collision dynamics: Evolution of the film thickness L. Opfer, 1 I. V. Roisman, 1 J. Venzmer, 2 M. Klostermann, 2 and C. Tropea 1 1 Institute of Fluid Mechanics and Aerodynamics, Technische Universit¨ at Darmstadt, Petersenstraße 17, 64287 Darmstadt, Germany 2 Evonik Industries AG, Goldschmidtstraße 100, 45127 Essen, Germany (Received 29 April 2013; revised manuscript received 6 November 2013; published 28 January 2014) This study is devoted to the experimental and theoretical investigation of aerodynamic drop breakup phenomena. We show that the phenomena of drop impact onto a rigid wall, drop binary collisions, and aerodynamic drop deformation are similar if the correct scaling is applied. Then we use observations of the deforming drop to estimate the evolution of the film thickness of the bag, the value that determines the size of the fine child drops produced by bag breakup. This prediction of film thickness, based on film kinematics, is validated for the initial stage by direct drop thickness measurements and at the latest stage by the data obtained from the velocity of hole expansion in the film. It is shown that the film thickness correlates well with the dimensionless position of the bag apex. DOI: 10.1103/PhysRevE.89.013023 PACS number(s): 47.55.D−, 47.55.Ca I. INTRODUCTION The fragmentation of bulk liquid in small drops is of great importance for a wide range of applications such as fuel injection in internal combustion engines and gas turbines, surface coatings, and the application of pesticides in agriculture. The characteristics of the spray are crucial for the quality and efficiency of the overall process. For situations in which the relative velocity between the drops and the air is high, the outcome of aerodynamic drop breakup determines the size distribution of the atomized spray, e.g., in air-blast atomization. For low-viscosity liquids, the outcome of the liquid-gas interaction can be described with respect to the Weber number, which defines the ratio of deforming aerodynamic forces and opposing surface tension forces, We = ρ a D 0 U 2 a /σ , where ρ a is the gas density, D 0 is the initial drop diameter, U a is the relative velocity between liquid drop and gas flow, and σ is the surface tension coefficient. The influence of viscosity on the liquid-gas interaction can be expressed by the Ohnesorge number Oh = μ l / √ D 0 ρ l σ , where μ l and ρ l are the dynamic viscosity and density of the liquid, respectively. Early investigations of bag breakup that revealed the morphology of drops during the fragmentation process can be found in [1,2]. Depending on the Weber and the Ohnesorge number, different deformation and breakup modes have been identified in the literature [3,4]: vibrational deformation (and breakup), bag breakup, multimode breakup, and shear breakup. Recent comprehensive reviews on aerodynamic fragmentation of drops can be found in [5,6]. Two experimental approaches for investigating such breakup processes have been used: drop breakup by a steady uniform air flow [7,8] in a wind tunnel and by the propagation of a shock wave [9–11] in a shock tube. While shock tube experiments require more extensive apparatus, the use of such a device avoids the problem that the drop first has to pass through a boundary layer before it enters the homogeneous flow field. The availability of digital high-speed cameras during the past decade revealed quantitative time-resolved information about the breakup processes. The effect of Rayleigh-Taylor wave numbers on the breakup morphology is studied in [12]. In [13] laser-induced fluorescence is used to visualize not only the contour of the drop but the structure of the liquid surface during the fragmentation. The authors conclude that pure shadowgraph visualizations might lead to misinterpretations concerning the physics of high We breakup events. Flock et al. [14] successfully used particle image velocime- try in order to measure velocity fields in the gas phase around the fragmenting drop in the bag and shear breakup regime. Their results also reveal quantitative information about the backflow in the wake of the drop. Among the fragmentation regimes described above, the bag breakup regime has the widest range of applications since it occurs at relatively low Weber numbers. In [15] bag breakup is shown to be responsible for distinct peaks in the drop size distributions of high-flow-rate industrial sprays. The drop size distribution of rain close to the ground can also be related to the occurrence of bag breakup events during the free fall of drops [16]. The bag breakup process can not only be observed during the fragmentation of drops, but it is also responsible for the breakup of liquid jets injected into a crossflow [17]. It is interesting to note that the morphology of drop impact onto a deep pool surface is very similar to that of a bag breakup. However, in this case the liquid and gas phases are inverted and the expanding film consists of the gas entrained by the drop impact. This phenomenon is visualized and studied in [18]. The present study is focused on measuring and modeling the main morphological parameters of drops deformed by a continuous air flow in a wind tunnel. The aim is to estimate the film thickness in the bag. This thickness determines the outcome of the breakup, i.e., final drop size, but cannot yet be directly measured. II. EXPERIMENTALMETHODS AND OBSERVATIONS A horizontal open-circuit wind tunnel has been designed and manufactured for the experimental investigation of drop deformation and breakup due to aerodynamic forces. A sketch of this wind tunnel is depicted in Fig. 1. A radial blower (a) generates the pressure gradient necessary to accelerate the air. The blower is connected to the settling chamber (c) by an adapter (b), which continuously transforms the cross section of the blower to the square cross section of the wind tunnel. The cross-sectional area of the settling chamber is 150 × 150 mm 2 . 1539-3755/2014/89(1)/013023(6) 013023-1 ©2014 American Physical Society