A major part of this paper is taken up by the calculation of the first axisymmetric surface oscillation frequency and the (2, 0) mode of an ellipsoidal bubble, in order to compare with experimentally obtained values of bubbles rising in water. First, an energy method is used for an ellipsoid oscillating in stagnant water. Interestingly, with help of a paper by Bjerknes [12] at 1873!. The results compare poorly with experiments. Agreement improves when the oscillation of the rise velocity is taken into account. The remaining difference between the results of theory and experimental values is ascribed to deviation of the bubble shape from an ellipsoid. Finally, volume oscillations of ellipsoidal bubbles are calculated with the energy method and the results compare well with those of an earlier work, based on an electrical analogon.
IntroductionWith pleasure we contribute to this Festschrift for Professor Wilhelm Schneider at the occasion of his 70th birthday. Wilhelm Schneider and one of us (L.v.W) worked together many years in the Scientific Council of CISM in Udine, Italy.We share an interest in classical hydrodynamics and I hope that this contribution will please him. We wish him many years to come in good health. This paper is on gas bubbles rising in clean water, so clean that there are no surfactants. Then, the boundary condition on the interface between gas and liquid is that the tangential shear stress must vanish. Rising in water under buoyancy, very small bubbles (diameter of order of 0.1 mm) remain spherical, but with diameters of 1 mm and larger they assume a shape which is approximately an oblate ellipsoid [1] with the short axis pointing in vertical direction. The ratio of the larger axis to the minor one, indicated here with χ , depends on the speed of rising which in turn depends on the size and on liquid properties. The terminal speed of rise, U T , is often expressed in terms of the Reynolds number Re , defined aswhere D eq is the equivalent bubble diameter and ν the kinematic viscosity of the liquid. The axes ratio depends on the Weber number We defined aswhere ρ and σ are the water density and air-water surface tension, respectively. Below D eq = 1.8 mm (Re ∼ 600), bubbles rise rectilinearly, above that value they describe a spiraling or zigzagging path [2,3], but