Effect of viscosity on droplet-droplet collision outcome: Experimental study and numerical comparison

  • Cecilie Gotaas
    Norwegian University of Science and Technology (NTNU) Department of Chemical Engineering, , Sem Sælandsvei 2, N-7491 Trondheim, Norway
  • Pavel Havelka
    Norwegian University of Science and Technology (NTNU) Department of Chemical Engineering, , Sem Sælandsvei 2, N-7491 Trondheim, Norway
  • Hugo A. Jakobsen
    Norwegian University of Science and Technology (NTNU) Department of Chemical Engineering, , Sem Sælandsvei 2, N-7491 Trondheim, Norway
  • Hallvard F. Svendsen
    Norwegian University of Science and Technology (NTNU) Department of Chemical Engineering, , Sem Sælandsvei 2, N-7491 Trondheim, Norway
  • Matthias Hase
    University of Stuttgart Institute of Aerospace Thermodynamics (ITLR), , Pfaffenwaldring 31, 70569 Stuttgart, Germany
  • Norbert Roth
    University of Stuttgart Institute of Aerospace Thermodynamics (ITLR), , Pfaffenwaldring 31, 70569 Stuttgart, Germany
  • Bernhard Weigand
    University of Stuttgart Institute of Aerospace Thermodynamics (ITLR), , Pfaffenwaldring 31, 70569 Stuttgart, Germany

説明

<jats:p>The influence of viscosity on droplet-droplet collision behavior at ambient conditions was studied experimentally and numerically. N-decane, monoethyleneglycol (MEG), diethyleneglycol (DEG), and triethyleneglycol were used as liquid phase providing viscosities in the range from 0.9to48mPas. Collision Weber numbers ranged approximately from 10 to 420. A direct numerical simulation code, based on the volume-of-fluid concept, was used for the simulations. Experimentally, observations of two droplet streams using a modified stroboscopic technique (aliasing method) were used to investigate the whole range of impact parameters during one experimental run. The experimental method has previously been verified for the water/air system [C. Gotaas et al., Phys. Fluids 19, 102105 (2007)]. In the present work, it was tested and validated for the n-decane/air system. Measured data agree well with those published in the literature. Well-defined regions of stretching separation and coalescence were identified, while reflexive separation regions were not found by using a single sinusoidal disturbance. However, the onset of reflexive separation was identified for MEG and DEG using an amplitude modulation technique. The results show that the criteria for onset of reflexive separation for viscous fluids provided by Y. I. Jiang et al. [J. Fluid Mech. 234, 177 (1992)] are not valid. This is consistent with the results given by K. D. Willis and M. Orme [Exp. Fluids 34, 28 (2003)]. A new empirical correlation for the onset of reflexive separation for high viscosity fluids is presented. The borders between coalescing and stretching separation were shifted toward higher Weber numbers with increasing viscosity. The lack of occurrence of reflexive separation for the single sinusoidal disturbance (small droplets), as well as the stretching separation boundary shift, can be explained by dissipation of collision kinetic energy in viscous flows inside the merged droplet after collision. Results from numerical simulations for MEG, DEG, and TEG correlated well with experimental data for the same fluids.</jats:p>

収録刊行物

被引用文献 (2)*注記

もっと見る

詳細情報 詳細情報について

問題の指摘

ページトップへ