Experimental experience indicates that even the originally homogeneous array of bubbles rising in a liquid column is slowly reorganized. Even in the homogeneous array of bubbles starts slowly to rearrange bubbles vertically (Ruzicka 2000) and laterally Tomiama et al. 2002, Hibiki and Ishii 2007). Resulting array pattern consists mainly of the separated, essentially horizontal assemblages of the bubbles.

As a first step to the understanding of bubble interactions, we started to investigate behavior of the pairs of identical size bubbles. The experimental setup enabling the levitating bubbles to be observed was applied for the observation of rising bubbles. Here, liquid is pumped downward through the entrance region of the conical diverging channel. It was proved that with exception of a narrow boundary layer, the plug liquid flow occurs in the channel. The experiments were carried out in wide class of transparent liquids. When the volume flow rate is properly adjusted, all bubbles of given size tend to remain in definite vertical plane of the channel space for quite a long period. According to this idea, there may also be a large probability of the bubble collisions and coalescence.

Position of the pair of bubbles was observed by the high speed camera simultaneously from two perpendicular projections (one direct, and one mirror image). With help of tailor made image analysis, time series of vertical and horizontal distances of bubbles were determined. With respect to natural fast oscillation of the rising bubble shape and velocity (5-10 Hz), the frequency of sampling was adjusted to 150 frames per second, and the recorded period was 60 s, which was satisfactory for the statistic evaluation of the data.

It has been proven, that the most probable position of the bubble pair is close to the horizontal alignment (the most probable deviation has been 20-25°), and the bubbles spacing is considerable (in water is the most probable distance around 15 mm). The wake behind the front bubble is evidently a forbidden space for the rear one.

The fact that there is no steady configuration of a pair of bubbles is obvious, because even any single bubble path is not straightforward but essentially zigzag or helical. Probable theoretical explanation, why the rear bubble practically neither enters the wake nor coalesces, can be based on the experience with bubbles rising in the sheared liquid column. When the bubble approaches the wake border, it appears in the sheared liquid; the wake essentially somewhat follows the front bubble velocity. As a result, the small, nearly spherical bubble starts to rotate, which induces the drift force inducing lateral motion. The same occurs with larger, ellipsoidal bubbles. Here, the drift effect of unsteady accelerating rotation is significant as well. The bubble in such a case has likely not enough time to consolidate the steady inclined position, which is typical for sliding by negative drift of large bubbles in shear flow (Tomiyama et al. 2002). Some horizontal shift of the most frequent position of a pair of bubbles is likely due to the optimization of the over all drag coefficient.

The experimental research in our laboratory still continues, namely with the aim to elucidate the effect of various liquid viscosity and surface tension (Reynolds and Eötvös numbers).

Generous support of the present paper by EC project ICT no. ED2.1.00/03.0082 is gratefully acknowledged.

References:

Hibiki T., Ishii M. (2007) Lift force in bubbly flow systems. Chem. Eng. Sci. 62, 6457-6474.

Ruzicka M.C. On bubbles rising in line. (2000) Int.l J. Multiphase Flow 26, 1141-1181.

Tomiyama A., Tamai H., Zun I., Hosokawa S. (2002) Transverse migration of single bubbles in simple shear flows. Chem. Eng. Sci. 57, 1849 1858.