Modeling of Spray/Wall Interactions: Based on Droplet Morphology Dynamics

• Main Author: Ribeiro, Rúben Filipe Torres
• Publication Date: 2019
• Format: Master thesis
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10400.6/8588
• topic: Corona Splash; Filme de Líquido; Impacto de Spray; Leidenfrost; Prompt Splash; Ressalto; Superfície Quente; Impacto de Gotas

The present work has the objective of perfecting our knowledge related to spray impact, which is of paramount importance for the optimization of a wide variety of investigation areas, such as combustion systems, coating and cooling processes, and also pollutant emissions. This last referred area has been gaining more and more importance due to the obvious environmental concerns that we face in our age. For these reasons,a remarkable effort by the scientific community has been made in order to deepen the understanding of the mechanisms underlying the spray impingement process. In this dissertation, and through numerical analysis, our in-house code was adapted to reflect the impingement conditions and secondary atomization treatment proposed by Ma et al. [41]. The complex relations between incident spray and the corresponding impact surface are yet far from being duly elucidated, whereby this paper aims to bring us closer to that objective. Evidently, an extensive bibliographic review was performed about theoretical and computational concepts. There are numerous computational models in literature that intend to portray the relation between the impinging spray and the impact surface. Although, not all of these models display the complexity necessary to represent different types of conditions, such as the presence of liquid film or even the existence of a temperature so high that prevents the contact between spray and wall through the generation of a vapor layer. This phenomenon is commonly known as ”Leidenfrost effect” and is usually neglected. One of the first to emerge was proposed by Naber and Reitz, employing the KIVA code, and proposed a single threshold to determine if splash occurred or not. At first glimpse, this model was obviously flawed by way of not accounting for the conditions of occurrence of each impingement regime. Later on, Senda presented a model of their own that was able to predict not only secondary atomization and liquid film formation resulting from the impinging droplets, but also the heat transfer process present in such situation. Sendas’s model despite presenting moderate accuracy, lacked the adaptability to a wider spectrum of applications. Bai and Gosman, using the " model for the gas phase and a stochastic Lagrangian method for the spray, tried to solve this lack of adaptability by modelling the effect of wall conditions and introducing several new regimes. The results translated in improvements describing the secondary droplets, mainly through fitting secondary droplets in a chi-squared distribution and by including surface energy and film dissipation in the conservation equations. Despite these satisfactory results, this model also failed to attain general applicability. Taking into account recent literature alterations, parameters such as saturation temperature and liquid film thickness were utilized to establish more detailed boundary conditions with the intent to represent a more extended range of possible scenarios. In the application of this model a distinction was made between corona splash and prompt splash due to the fact that secondary droplets present different characteristics for each case. Questions such as expansion of the lamella, crown formation and propagation, as well as splashed film mass or transformed mass from crown to secondary droplets became of paramount importance during all the stages of the identified regime and were all detailed in this model. The size and velocity of secondary droplets depend strongly on the initial conditions of the spray at the injector exit, as well as the interaction between incident droplets, crossflow, liquid film, evaporation rate, and interposed hot wall. All these parameters are considered in this macroscopic model of the spray/wall interactions. This dissertation allows us to obtain a detailed analysis about the properties of secondary droplets. In what concerns this subject, a new regime was implemented to a specific gap of boundary conditions and denominated ”uncertain region”. This regime quantifies the probability of splash or rebound occurrence through a uniform distribution since the available information for these conditions is very scarce. Moreover, simulations are carried out for predicting the outcome of flows, including liquid film formation, droplet breakup, and spray evaporation. The numerical results are then compared against experimental data available in open literature to ascertain the predictions capabilities and validate the model.