# Immersed quantum systems [Elektronische Ressource] : a sodium Bose-Einstein condensate for polaron studies / put forward by Jens Appmeier

-

English
108 Pages

Description

Dissertationsubmitted to theCombined Faculties for the Natural Sciences and for Mathematicsof the Ruperto-Carola University of Heidelberg, Germanyfor the degree ofDoctor of Natural SciencesPut forward byDiplom-Physiker: Jens AppmeierBorn in: Leipzig, GermanyOral examination: Juli 13, 2010Immersed Quantum Systems:A Sodium Bose-Einstein Condensatefor Polaron StudiesReferees: Prof. Dr. Markus K. OberthalerProf. Dr. Selim JochimZusammenfassungIn der vorliegenden Arbeit wird der Aufbau eines neuen Experiments zur Unter-suchung ultrakalter bosonischer und fermionischer Quantengase beschrieben. Bose-23Einstein Kondensation von Na Atomen wird in zwei verschiedenen Magnetfallenkon- gurationen, der plugged\ Quadrupolfalle und der Kleeblattfalle erreicht. Au erdem" 23werden beide Fallentypen bezuglic h ihrer Eignung fur Gemischexperimente mit Na6und Li verglichen. In einem solchen Gemisch sollte es moglich sein Polaronen zu unter-suchen. Diese Quasiteilchen entstehen, sobald eine Komponente des Natrium-LithiumGemisches nur noch in einer sehr geringen Konzentration vorliegt. Der Grenzfall eineseinzelnen Teilchens in einem bosonischen Hintergrund wird theoretisch betrachtet undanhand einer numerischen Simulation untersucht.AbstractThe subject of this work is the setup of an experiment to study immersed quantum23 6 23systems using bosonic Na and fermionic Li.

Subjects

##### Physik

Informations

Report a problem

Dissertation
submitted to the
Combined Faculties for the Natural Sciences and for Mathematics
of the Ruperto-Carola University of Heidelberg, Germany
for the degree of
Doctor of Natural Sciences
Put forward by
Diplom-Physiker: Jens Appmeier
Born in: Leipzig, Germany
Oral examination: Juli 13, 2010Immersed Quantum Systems:
A Sodium Bose-Einstein Condensate
for Polaron Studies
Referees: Prof. Dr. Markus K. Oberthaler
Prof. Dr. Selim JochimZusammenfassung
In der vorliegenden Arbeit wird der Aufbau eines neuen Experiments zur Unter-
suchung ultrakalter bosonischer und fermionischer Quantengase beschrieben. Bose-
23Einstein Kondensation von Na Atomen wird in zwei verschiedenen Magnetfallenkon-
gurationen, der plugged\ Quadrupolfalle und der Kleeblattfalle erreicht. Au erdem
" 23werden beide Fallentypen bezuglic h ihrer Eignung fur Gemischexperimente mit Na
6und Li verglichen. In einem solchen Gemisch sollte es moglich sein Polaronen zu unter-
suchen. Diese Quasiteilchen entstehen, sobald eine Komponente des Natrium-Lithium
Gemisches nur noch in einer sehr geringen Konzentration vorliegt. Der Grenzfall eines
einzelnen Teilchens in einem bosonischen Hintergrund wird theoretisch betrachtet und
anhand einer numerischen Simulation untersucht.
Abstract
The subject of this work is the setup of an experiment to study immersed quantum
23 6 23systems using bosonic Na and fermionic Li. Bose-Einstein condensation of Na
has been achieved in two di erent magnetic trap con gurations, namely the plugged
quadrupole trap and the cloverleaf trap. Both are compared with respect to their suit-
ability for a two-species experiment using this particular isotopes. In such a mixture,
it should be possible to investigate polarons, which are quasiparticles, emerging when
one of components of the mixture has only a very rare concentration. Furthermore,
a theoretical study of the polaron will be discussed. A mean- eld calculation has
been carried out in order to simulate the impurity behavior in the presence of a large
bosonic background gas.Contents
1. Introduction 9
I. BEC Impurity as a Polaron 13
2. Theory of Ultracold Atomic Gases 15
2.1. BEC: Weakly Interacting Case . . . . . . . . . . . . . . . . . . . . . . . 15
2.1.1. Basic Scattering Theory . . . . . . . . . . . . . . . . . . . . . . 15
2.1.2. Tuning Interactions: Feshbach Resonances . . . . . . . . . . . . 17
2.1.3. Mean-Field Approximation: Gross-Pitaevskii-Equation . . . . . 19
2.1.4. Excitations of the Interacting BEC: Bogoliubov Transformation 21
2.2. Degenerate Fermions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
2.3. Polaron Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.3.1. Condensed Matter Treatment . . . . . . . . . . . . . . . . . . . 24
2.3.2. Cold Atomic Gases Perspective . . . . . . . . . . . . . . . . . . 27
3. Numerical Methods 29
3.1. Two-Component GPE . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.2. E ective Mass Determination . . . . . . . . . . . . . . . . . . . . . . . 32
3.3. Validity Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4. Mean Field Simulation: 2 Component GPE 35
4.1. Single Impurity in a BEC Background . . . . . . . . . . . . . . . . . . 35
4.1.1. Impurity Localization . . . . . . . . . . . . . . . . . . . . . . . . 35
4.1.2. Density Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.1.3. The Coupling Parameter . . . . . . . . . . . . . . . . . . . . . . 39
4.1.4. Central Density . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
4.1.5. In uence of the Impurity Potential . . . . . . . . . . . . . . . . 42
4.1.6. E ective Mass Computation . . . . . . . . . . . . . . . . . . . . 43
4.1.7. Which is the Better Impurity: Na or Li? . . . . . . . . . . . . . 45
4.2. Many Impurity Case . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
7II. Bose-Einstein Condensation of Sodium 49
5. Trapping and Investigation of Cold, Neutral Atoms 51
5.1. Optical Dipole Traps . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
5.2. Magnetic Trapping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
5.2.1. Plugged Quadrupole Trap . . . . . . . . . . . . . . . . . . . . . 54
5.2.2. Cloverleaf Trap . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
5.2.3. Special Requirements in a Two Species Design . . . . . . . . . . 59
5.3. Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
5.4. Deducing the Temperature . . . . . . . . . . . . . . . . . . . . . . . . . 63
6. Experimental Setup 65
6.1. Vacuum System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
6.2. Sodium Laser System . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
6.3. Magneto-Optical Trap . . . . . . . . . . . . . . . . . . . . . . . . . . . 70
6.4. Spin Polarizing and Puri cation . . . . . . . . . . . . . . . . . . . . . . 72
6.5. BEC in the Plugged Quadrupole Trap . . . . . . . . . . . . . . . . . . 75
6.6. NaLi Unplugged: BEC in Cloverleaf Trap . . . . . . . . . . . . . . . . . 75
6.7. Dipole Trap . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
7. Conclusion and Outlook 83
A. Sodium and Lithium Line Data 87
B. Cloverleaf Design 91
C. Interlock 95
C.1. Service Water Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
C.2. Clean Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
Bibliography 106
81. Introduction
Theorie und Experiment
geh oren zusammen, eines ohne
das andere bleibt unfruchtbar.
(Max Planck)
The development of quantum mechanics in 1925 released a great discourse how these
new ideas of describing the physics at very small scales modi es our understanding of
nature. The outcome of an experiment is open to di erent interpretations, which is in
contradiction to the ideas of the classical understanding of physics. Several interpre-
tations have been developed to gain information about \what’s going on there...", the
most popular of which is the Copenhagen interpretation, formulated by N. Bohr and
W. Heisenberg. Its key feature is the probabilistic interpretation of the wavefunction,
describing the state of a particle. A measurement of the state of the particle, for in-
stance its position, causes the wavefunction to collapse to the value of this observable
de ned by the measurement itself. The question about the position of the particle
before thet is thus meaningless.
This interpretation of the wavefunction { being the probability amplitudes of the
particle { puts aside the discussion about the nature of light and matter, dating
back to the 1600s to the competing theories of light by C. Huygens and I. Newton.
With the Copenhagen interpretation at hand, the wave-particle duality is a central
concept of quantum mechanics. Light and matter can both be described as particles
or waves. Young’s double-slit experiment has been performed for both matter and
light successfully.
The success of quantum mechanics started a new era in physics, leading to a deeper
understanding of nature. New phases of matter, apart from solids, uids and gases
were predicted, namely the Bose-Einstein condensate and the degenerate Fermi gas.
Each of these are speci c to the two fundamental classes of particles, bosons and
fermions. The distinction between both becomes important only on the quantum
scale, i.e. when the gas of particles can no longer be described by the classical Maxwell-
Boltzmann distribution. The reason for this deviation from the classical picture is that
the niteness of the available, discrete energy states of the particles becomes important
{ quantum mechanics comes into play.
Prominent examples of those two quantum phases is the Bose-Einstein condensation
of Cooper pairs, responsible for type I superconductivity or the conduction electrons
in solids that are already degenerate at room temperature. But these states also
occur at large scale, for instance the neutron stars are thought to consist mostly of
9Fermi-degenerate neutrons.
The availability of these states of matter in dilute, cold atomic gases since the mid
1990s [1, 2], triggered a whole range of experiments, addressing fundamental physical
questions, which were not accessible in experiments before.
The reason for the success of the cold atomic gases experiments is, that they o er
1very clean systems with a wide experimental control over the systems parameters like
atom number, density, temperature and trapping geometry. Furthermore the scatter-
ing properties of the atoms can be manipulated via the use of Feshbach resonances,
a degree of freedom that is not accessible in other systems, e.g. solids. This freedom
of tuning parameters allows to study a wide range of physical questions, e.g. some
condensed matter phenomena with an extended range of tunable parameters, one of
which is the polaron problem, addressed in this thesis.
Some achievements of cold atoms gases are the rst experimental studies of quantum
e ects like matterwave interference [3] or the tunneling processes [4] on a macroscopic
scale. Using optical lattices in the experiments, has lead to the observation of the
super uid to Mott-insulator transition [5], which can not be studied in solids. This
phase transition is not driven by temperature as a classical phase transition, but rather
by quantum uctuations, which still occur at T = 0.
Also degenerate Fermi gases have been studied extensively, revealing the transition
from the Bardeen-Cooper-Schrie er (BCS) state of Cooper pairs to the formation
of compound molecules [6] and their Bose-Einstein condensation [7, 8, 9]. Feshbach
resonances that allow to tune the interaction, have been found in single component
Bose gases [10] and in mixtures of two di erent species [11, 12]. E ects of fermionic
atoms on bosonic atoms in a 3D optical lattice have been studied [13, 14]. Experiments
succeeded to involve di erent spin states of an ultracold atom gas as an additional
free parameter of the system [15].
The aim of the experiment presented here is to study immersed quantum systems,
where a species with a small concentration (impurities) is interacting with a degenerate
background gas. A similar system is known from condensed matter physics. If an
electron in the conduction band is moving through the ionic crystal lattice it can
interact with the lattice excitations (phonons), thereby forming a quasiparticle|the
polaron.
In the crystal lattice, the electron-phonon interaction mediates a weak coupling
between electrons near the Fermi edge. Cooper showed that even an arbitrarily weak
attractive interaction in these systems will lead to the occurrence of weakly bound
pairs (the Cooper-pairs), responsible for type I superconductivity [16].
However, type II superconductors are not explained by the BCS theory. Here the
critical temperature for superconductivity is close to the Debye temperatureT . EvenD
at these temperatures, the electron-phonon interaction should still be large enough,
such that every electron forms a polaron. In this limit the polaron-polaron inter-
1The techniques used to achieve Bose-Einstein condensation of dilute atomic gases are highly isotope
selective, such that any kind of depletion of the system with a di erent species or isotope requires
a high experimental e ort. This is in contradiction to condensed matter systems where a very
high e ort is necessary to produce samples with a negligible concentration of dopants.
10

en expand_more