When an oil droplet is placed on a quiescent oil bath, it eventually collapses into the bath due to gravity. The resulting coalescence may be eliminated when the bath is vertically vibrated. The droplet bounces periodically on the bath, and the air layer between the droplet and the bath is replenished at each bounce. This sustained bouncing motion is achieved when the forcing acceleration is higher than a threshold value. When the droplet has a sufficiently low viscosity, it significantly deforms : spherical harmonic oldmath{$Y_{ell}^m$} modes are excited, resulting in resonant effects on the threshold acceleration curve. Indeed, a lower acceleration is needed when $ell$ modes with $m=0$ are excited. Modes $m
e 0$ are found to decrease the bouncing ability of the droplet. When the mode $ell=2$ and $m=1$ is excited, the droplet rolls on the vibrated surface without touching it, leading to a new self-propulsion mode.
When a droplet is gently laid onto the surface of the same liquid, it stays at rest for a moment before coalescence. The coalescence can be delayed and sometimes inhibited by injecting fresh air under the droplet. This can happen when the surface of the bath oscillates vertically. In this case the droplet basically bounces on the interface. The lifetime of the droplet has been studied with respect to the amplitude and the frequency of the excitation. The lifetime decreases when the acceleration increases. The thickness of the air film between the droplet and the bath has been investigated using interference fringes obtained when the system is illuminated by low-pressure sodium lamps. Moreover, both the shape evolution and the motion of the droplet center of mass have been recorded in order to evidence the phase offset between the deformation and the trajectory. A short lifetime is correlated to a small air-film thickness and to a large phase offset between the maximum of deformation and the minimum of the vertical position of the center of mass.
The bouncing of air bubbles of about 1 mm in diameter on an air/oil interface is reported. We compare the behavior of the bubble when the interface is at rest or when it is vibrated. The results obtained for the interface at rest allowed extracting the key parameter responsible for the outcome of the bubble collision, i.e. bounce or coalescence (bubble rupture). This parameter is the deformation of the bubble. When the bubble deformation is sufficient (according to the oil viscosity and to the bubble velocity and size), the bubble bounces because the intervening oil film has a large radius and the bubble–interface contact time is much shorter than the time needed for the film drainage to its critical thickness of rupture. This hypothesis is confirmed by the experiments of the bubble collisions with vertically vibrated air/oil interface. It was shown that it is possible to significantly increase the bubble lifetime: the bubble bounces at the interface when the interface is vibrated, and this occurs when the vibration amplitude is above a threshold. The amplitude threshold depends on the frequency, on the size of the bubble and on the viscosity of the surrounding oil. For bouncing, the vibrated interface must provide enough energy to the bubble to deform in such a way that the drainage characteristic time of the oil film becomes longer than the contact time between the bubble and the interface.
Experiments on the coalescence time of submillimeter bubbles colliding with a distilled water/air interface either being at rest (undisturbed) or vibrating vertically (with controlled amplitude and frequency) were carried out. It was found that the outcome of the bubble collision (coalescence or bounce) depends on impact velocity and size of the bubble, i.e. the parameters determining the bubble deformation degree. With the surface at rest, when the deformation of the bubble was sufficiently high, bubble bouncing was observed. It was caused by the fact that the radius of the intervening liquid film formed between the colliding bubble and water/air interface was large enough to prevent the liquid layer from reaching its thickness of rupture within the time of bubble–interface contact. Coalescence occurred in a consecutive collision if the bubble deformation was below a threshold value, as a result of dissipation of the kinetic energy associated with the bubble motion. The hypothesis about the crucial role of the bubble deformation and size of the liquid film formed in the bouncing mechanism was confirmed in a series of experiments where the bubble collided with a vibrating water/air interface. It was shown that when the kinetic energy was properly re-supplied from an external source (interface vibrations), the spectacular phenomenon of “immortal” bubbles, dancing indefinitely at the water/air interface, was achieved. It was shown that “immortal” bubble formation is a consequence of a similarly high degree of the bubble shape deformation and consequently a large enough radius of the liquid film formed.