An experimental and modeling study on stochastic aspects of microexplosion of binary-fuel droplets

Abstract

Disruptive droplet burning is characterized by microexplosion occurrence during burning. The microexplosion phenomena of binary-fuel droplets are caused by internal boiling that is initiated by the homogeneous nucleation of bubbles in the liquid phase. Since homogeneous nucleation is a random process due to density fluctuation, some stochastic aspects appear in disruptive burning of the droplets. In order to investigate stochastic characteristics in disruptive burning, the microexplosion occurrence was studied experimentally and theoretically. In experiments, the burning droplet of an n -hexane/ n -hexadecanemixture was injected upward, and the induction time for microexplosion occurrence was obtained. Results show that microexplosion induction time is distributed over the quasi-steady vaporization period. The maximum frequency of microexplosion induction time and the occurrence probability of microexplosion increase with the initial droplet diameter. The microexplosion occurrence was modeled considering the homogeneous bubble nucleation rate. The present stochastic model well demonstrated the experimental results. The theory shows that these stochastic characteristics of microexplosion occurrence depend on the fifth power of the initial droplet diameter. The bubble nucleation leading to microexplosion depends on the timescale and the superheated liquid volume. The timescale in droplet combustion is proportional to the second power of the initial droplet diameter, and the volume of superheated liquid is proportional to the third power of the initial droplet diameter. For relatively large droplets, the microexplosion occurrence can be understood in a deterministic way. For smaller droplets, appearing in spray combustors, however, the microexplosion occurrence is more stochastic. The scatter in explosion behavior is also discussed considering the bubble nucleation location inside the droplet.