Flow topologies in primary atomization of liquid jets: a direct numerical simulation analysis. (26th November 2018)

• Record Type: Journal Article
• Authors: Hasslberger, Josef; Ketterl, Sebastian; Klein, Markus; Chakraborty, Nilanjan
• Is Part Of: Journal of fluid mechanics. Volume 859(2019)
• Page Start: 819
• Page End: 838
• Publication Date: 2018-11-26
• Subjects: aerosols/atomization, -- gas/liquid flow, -- topological fluid dynamics; Fluid mechanics -- Periodicals; 532.005
• DOI: 10.1017/jfm.2018.845
• Languages: English
• ISSNs: 0022-1120

Abstract

Abstract : The local flow topology analysis of the primary atomization of liquid jets has been conducted using the invariants of the velocity-gradient tensor. All possible small-scale flow structures are categorized into two focal and two nodal topologies for incompressible flows in both liquid and gaseous phases. The underlying direct numerical simulation database was generated by the one-fluid formulation of the two-phase flow governing equations including a high-fidelity volume-of-fluid method for accurate interface propagation. The ratio of liquid-to-gas fluid properties corresponds to a diesel jet exhausting into air. Variation of the inflow-based Reynolds number as well as Weber number showed that both these non-dimensional numbers play a pivotal role in determining the nature of the jet break-up, but the flow topology behaviour appears to be dominated by the Reynolds number. Furthermore, the flow dynamics in the gaseous phase is generally less homogeneous than in the liquid phase because some flow regions resemble a laminar-to-turbulent transition state rather than fully developed turbulence. Two theoretical models are proposed to estimate the topology volume fractions and to describe the size distribution of the flow structures, respectively. In the latter case, a simple power law seems to be a reasonable approximation of the measured topology spectrum. According to that observation, only the integral turbulent length scale would be required as an input for the a priori prediction of the topology size spectrum.