Liu, Muyuan (2017)
Numerical Study of Head-on Binary Droplet Collisions: Towards Predicting the Collision Outcomes.
Technische Universität Darmstadt
Ph.D. Thesis, Primary publication
|
Text
Thesis_final.pdf - Accepted Version Copyright Information: CC BY-NC 4.0 International - Creative Commons, Attribution NonCommercial. Download (9MB) | Preview |
Item Type: | Ph.D. Thesis | ||||
---|---|---|---|---|---|
Type of entry: | Primary publication | ||||
Title: | Numerical Study of Head-on Binary Droplet Collisions: Towards Predicting the Collision Outcomes | ||||
Language: | English | ||||
Referees: | Bothe, Prof. Dr. Dieter ; Tropea, Prof.Dr. Cameron | ||||
Date: | 2017 | ||||
Place of Publication: | Darmstadt | ||||
Date of oral examination: | 17 October 2017 | ||||
Abstract: | Binary droplet collision plays an important role in nature and in many technical processes involving sprays. The modeling of the collision outcomes, namely bouncing, coalescence, separation after temporary coalescence, and spatter (also called ‘shattering’ and ‘splashing’), establishes the basis for the investigation of the atomization processes on larger length scales. The aim of this thesis is to develop numerical methods that are employed in the prediction of the collision outcomes and the numerical investigation of the phenomena in binary droplet collisions which affect the collision outcomes. The in-house code Free Surface 3D (FS3D), which is based on the Volume of Fluid (VOF) method, is employed for the numerical simulations. The numerical investigations are restricted to head-on collisions. Spatter occurs at high energetic collisions, resulting in a thin liquid lamella that ruptures artificially in standard numerical simulations. In order to simulate spatter, an improved lamella stabilization algorithm has been developed and extensively validated. By means of properly chosen white noise disturbances of the initial velocity field, the instability of the rim of the collision complex is triggered and the spatter is successfully reproduced in the simulations. Very good agreements between the simulation results and the experiments are achieved. Based on the simulation results, the development of the rim instability is considered as an amplification of disturbances via a signal amplification system that is subdivided into three sequential connected subsystems. It is confirmed that the development of the rim instability in the linear phase of the instability can be predicted by the Rayleigh-Plateau instability theory. The influence of the droplet viscosity is studied numerically and it is shown that the collision outcome tends to be spatter when the droplet viscosity is reduced. This dependency decreases with the decrease of the droplet viscosity. The droplet viscosity influences the development of the rim instability mainly through varying the geometrical evolution of the rim. A successful elucidation of the mechanism of rim instability builds the foundation for the prediction of the occurrence of spatter and the prediction of the size distribution of the secondary droplets arising in spatter. The investigation of the mechanism of the rim instability in the context of binary droplet collisions is of general importance because the ejection of secondary droplets from an unstable rim also emerges in collisions of a droplet on a solid substrate or on a liquid film. Binary droplet collisions result in bouncing or coalescence at relatively small Weber numbers. The simulations of bouncing and coalescence have been successfully conducted by switching the boundary conditions on the collision plane. The simulation results are in good agreement with corresponding experiments. However, the simulations are not predictive because the collision outcome must be specified in advance. The difficulty of the prediction of bouncing versus coalescence lies in the fact that the thin gas film between the colliding droplets cannot be resolved in feasible simulations and that a physically meaningful coalescence criterion is missing in the numerical method. In order to facilitate the predictive simulation, a multi-scale simulation concept has been developed. In addition to the main solver FS3D, which solves the flow on the macroscopic scale, the multi-scale simulation concept consists of three parts: (1) A sub-grid-scale (SGS) model is integrated within the main solver FS3D. (2) Coalescence is numerically suppressed before a suitable coalescence criterion is contingently satisfied. (3) A numerical coalescence criterion is applied. Based on the lubrication theory, the SGS model is derived which accounts for the rarefied flow effect. The SGS model is implemented in FS3D and extensively validated. For the integration of the SGS model, the pressure in the gas film, which is solved by the SGS model, applies as a pressure boundary condition on the collision plane. Employing the first intersection of PLIC-surfaces with the collision plane as coalescence criterion, the collision outcome in the simulation can be both bouncing and coalescence. The predicted collision outcome, however, depends on the grid resolution. Employing zero gas film thickness (in algorithm tolerance) as coalescence criterion, the simulations result only in bouncing. It is shown that various possible corrections of the velocity field, which decides the transport of the liquid phase, have not led to a meaningful prediction of the transition between coalescence and bouncing. Further developments, e.g. the volume-averaged Volume of Fluid (VA-VOF) method, which takes into account the velocity difference within a computational cell, shall be implemented in future work to increase the accuracy of the transport of the fluid phase. By means of the multi-scale simulation it is qualitatively shown that the collision outcome tends to be coalescence at higher rarefaction in the gas phase. |
||||
Alternative Abstract: |
|
||||
URN: | urn:nbn:de:tuda-tuprints-70187 | ||||
Classification DDC: | 500 Science and mathematics > 500 Science 500 Science and mathematics > 510 Mathematics 500 Science and mathematics > 530 Physics 600 Technology, medicine, applied sciences > 600 Technology |
||||
Divisions: | 16 Department of Mechanical Engineering | ||||
Date Deposited: | 13 Dec 2017 06:39 | ||||
Last Modified: | 09 Jul 2020 01:56 | ||||
URI: | https://tuprints.ulb.tu-darmstadt.de/id/eprint/7018 | ||||
PPN: | 423934449 | ||||
Export: |
View Item |