Droplet-resolved direct numerical simulation of fuel droplet evaporation

Name: Abhishek Jain

Abstract:
Evaporation of fuel droplets and mixing of fuel vapor with the oxidizer is the driving force for combustion reactions in many combustion devices. Since the flow in most practical combustion devices is turbulent, an understanding of the interactions among turbulence, mixing and reactions is necessary to improve fuel efficiency and reduce pollutant emissions. However, the effects of small-scale turbulence on the dynamics of evaporation and the resultant fuel vapor field are relatively less known. Fluctuating relative velocity and inhomogeneities in the fuel vapor field around the droplets have been observed to affect the droplet vaporization rates. Apart from the droplet-turbulence interactions, droplet-droplet interactions and turbulent mixing of the fuel vapor with the oxidizer are other aspects critical to the formation of a combustible mixture. The present work aims to contribute towards a physical understanding of the droplet-turbulence and droplet-droplet interactions as well as the turbulent mixing of fuel vapor by performing droplet-resolved direct numerical simulation (DNS). The effects of turbulence on the evaporation of droplets larger than the Kolmogorov length scale are investigated using droplet-resolved DNS. The DNS is performed using a numerical method based on the immersed boundary method (IBM) that is developed here to perform efficient DNS with multiple droplets. Firstly, an improved IBM for a general particulate flow is developed. The displaced forcing and the predictor-corrector forcing are proposed and validated for the prediction of drag and scalar gradients on the immersed body in incompressible flows. The method is extended for application to variable density flows using a low Mach formulation by carefully considering the phase change. The predictions of evaporation rates are validated using a correlation for both stationary and moving droplets. The droplet-resolved DNS considers droplets in the dilute regime of spray atomization and targets operating conditions relevant to internal combustion engines. The effects of key factors such as the droplet diameter, the Kolmogorov length scale, the ratio of droplet velocity to the fluctuating gas velocity and pressure are studied. The temporal statistics are studied in terms of model predictions using a commonly employed correlation. The evaporation and scalar mixing for clustered droplets is studied to investigate the droplet-droplet interactions and assess their relevance on top of the droplet-turbulence interactions experienced by a droplet in isolation. The small-scale mixing is investigated in terms of the scalar dissipation and diffusion rates. The effects of the unity Lewis assumption, a frequently used simplification, on the evaporation and mixing are also studied.

Zoom Link (or alternative) - if available

Committee Members
Associate Professor Seung Hyun Kim
Professor Datta Gaitonde
Professor Sandip Mazumder
Professor Jeffrey Sutton