Exploring Electron Bubbles in Liquid Helium using Cavitation

Abstract

An electron entering liquid helium experiences a repulsive potential of 1 eV. This originates in the interaction between the injected electron and electrons of the closed shells of helium atoms through Pauli exclusion principle. If the energy of the electron exceeds 1 eV, it can penetrate the liquid, form a cavity free from helium atoms and subsequently localize itself within the cavity. This is known as a single electron bubble (SEB). In another configuration, a system of electrons of energy less than 1 eV can form a floating charged layer above the surface of liquid helium: a two-dimensional system that has been studied in great detail over last few decades. If the number of electrons in this layer exceeds a critical value of 2 × 10^13electrons/m^2, an electrohydrodynamic instability sets in, giving rise to multielectron bubbles (MEBs), referring to micron to mm sized cavities containing a large number of electrons.

In the experiments to be discussed in this thesis, the primary technique is based on cavitation of liquid helium using pulsed ultrasound. After a brief introduction of the experimental technique, we will present the main results obtained during my Ph.D., as follows:

The first result is related to the phenomenon of cavitation in superfluid helium. We have observed that after cavitation, the bubble is pushed out of the acoustic focus because of the radiation pressure and can grow up to a size as large as a millimetre. The growth and collapse of these bubbles can be understood through Rayleigh-Plesset equation at low temperatures, and by condensation of vapour (limited by the thermal diffusivity of helium) above lambda but this description fails near the lambda transition. We suspect this is related to the large density of vortices nucleated near the bubble surface during the growth of the cavitating bubble. Second, we have observed a new species which cavitates at a negative pressure approximately 80 and 70 % lower magnitude than SEBs. We conclude that these are multielectron bubbles with small (<20) number of electrons. We will be presenting various evidence supporting this claim and compare our results with related experiments previously reported. We believe that the application of sound to a charged helium surface or a strong discharge from the tungsten tip can render the surface unstable and thereby facilitates the formation of FEBs. Interestingly, FEBs can get trapped on the vortex lines and can act as tracer particles for visualizing vortices. Apart from these two species of FEBs, we have also observed cavitation due to electrons created by the Penning Ionization of diatomic helium molecules (dimers), which occurs when two helium molecules in excited states combine with each other. Thirdly, we have shown howa charged helium surface can be rendered unstable using ultrasound, in the presence of small electric fields such as to create MEBs. An indentation is formed on the charged surface through mechanical impact, and that can lead to a significant increase in the local surface charge density and thereby generation of MEBs. We estimated the initial charge density of the bubbles above lambda point to be close to 1013 electrons/m2, which is significantly higher than what has been achieved before. With gradual condensation of vapor and corresponding reduction in the bubble sizes, we expect to achieve 2DES with strong quantum correlations, while the MEBs would still be observable with standard imaging systems. Finally, we have trapped MEBs using ultrasound pulses above lambda, by balancing the radiation force, electric forces and buoyancy. We observe that we can trap MEBs for as long as 30 𝑚𝑠, only limited by the heating in the system. We propose a method to reduce this heating and increase the duration of the trap. We have also seen an unusual increase in size of the MEBs while it is trapped; this anomalous increase in size can be linked with the convection flow in the liquid. Finally, we discuss the strength and shortcoming of this trap and compare it with previous trapping methods.