On the expansion of turbulent Bose-Einstein condensates

Abstract

Quantum Fluids provide an exquisitely tuneable playground to study numerous areas of Physics. One such area is that of turbulence. Turbulence was famously touted by Feynman as the last unsolved problem in classical physics and remains a vastly studied topic to this day. Bose-Einstein condensates - a type of quantum fluid - are dictated by quantum mechanics. This means that phase defects are quantised, allowing a simplified system to understand the nature of turbulence compared to their classical counterparts. We begin by touching upon the creation and detection of solitons within one-dimensional condensates. Here, we learn the basics of computational methods in the simulation of Bose-Einstein condensates. Using this knowledge, we then move onto more complicated three-dimensional systems, where phase defects are vortices; long strings of empty space that stretch though a condensate. We work to answer three questions in these three-dimensional systems. Firstly, we investigate how the nature of a condensate affect its time-of-flight expansion. Condensate experiments are notoriously difficult to image; experimentalists often release the condensate from its trap and take images as it rapidly expands out. These time-of-flight images are known to be affected by the phase defects within them. Here, we develop a computational technique to have a remeshing computational domain so that sizes of expanded condensates previously impossible to simulate are now reachable. We use this to study how vortices affect an expanding condensate. We then follow on by modelling the only current experimentally viable method of creating quantum turbulence in a Bose-Einstein condensate. The strong granulation and small vortex rings created give power laws not observed previously; we prove that precursory predictions of a Kolmogorov or Vinen type turbulence in these driven systems do not accurately describe our observed dynamics and further name our observed turbulence as “strong turbulence”. Such a result naturally makes one question how one can experimentally create Kolmogorov or Vinen turbulence in a condensate. We thus propose a method using an imaging device which one can tune to realise a range of turbulence types. We also study how experimental results would look from such a setup before finally postulating on further work needed to understand how one can apply this method in an experiment.