Study of non-spherical bubble oscillations under acoustic irradiation in viscous liquid

• Published in: Ultrasonics sonochemistry 2019-06, Vol.54, p.256-273
• Main Authors: Klapcsik, Kálmán, Hegedűs, Ferenc
• Format: Article
• Language: English
• Subjects: GPU accelerated IVP; High viscosity; Keller-Miksis equation; Spectral-collocation method; Spherical stability
• ISSN: 1350-4177; 1873-2828
• DOI: 10.1016/j.ultsonch.2019.01.031

•The effect of dissipation on the spherical stability was investigated.•The exact linear formulation and the boundary layer approximation were compared.•Optimal range of dissipation rate can be defined.•The computations were carried out by a GPU accelerated IVP solver. The effect of dissipation on the shape stability of a harmonically excited bubble is investigated. The employed liquid is the highly viscous glycerine. The rate of the dissipation is controlled through the alteration of viscosity of the liquid by varying its temperature. The mean radius of the bubble during its radial oscillation is described by the Keller-Miksis equation. Two approaches are used to describe the surface oscillations. The first model solves the surface dynamics equations of each mode together with the transport equation of the vorticity in the liquid domain. The second model approximates the transport equation, which is a partial differential equation, with a boundary layer approximation reducing the required computational resources significantly. The comparison of the surface models shows qualitative agreement at low dissipation rate; however, at high viscosity the application of the full transport equation is mandatory. The results show that an increasing rate of dissipation can significantly extend the shape stable domains in the excitation frequency-pressure amplitude parameter plane. Nevertheless, the collapse strength is decreasing due to the highly damped oscillations. It has been found that an optimal range of dissipation rate in terms of temperature can be defined expressing a good compromise between the collapse strength and surface stability. The computations are carried out by an in-house GPU accelerated initial value problem solver.