dc.description.abstracteng | The search for novel phenomena that exceeds and interconnects our existing knowledge is at
the heart of every scientific effort. Physics, and particularly solid-state research, is no
exception. If anything, the technological advancements of the last decades have made the
need for new concepts even more apparent. A prototypical example here is the beginning
breakdown of Moore’s law in the semiconductor industry, where electronic components shrunk
down to a few nanometres in size. This greatly enhanced the performance of, e.g., integrated
circuits, but leads to a dead end since the further shrinking of structures is inhibited by their
instability on a nanoscale [1], [2]. Such an issue cannot be solved by engineering efforts alone
since it is the very physics that blocks the way. In the foreseeable future, disruptive
technological advancement will rely on new physical concepts as a basis, and current efforts
in solid-state physics research reflect this, as considerable resources are assigned to the study
of phenomena that are yet described as “exotic”.
One of such novel phenomena is the existence of non-trivial topological phases of matter.
Theoretically proposed by Haldane in 1988 [3], [4], the concept of topology lied mostly dormant
until 2007, when König et al. identified HgTe quantum wells as the first real world system with
non-trivial topology [5], [6]. Besides the interesting implications for fundamental research, the
emergence of the so-called bulk-boundary correspondence (BBC) from non-trivial topology
initiated an ever-growing research field within the physics community. The BBC enforces a
conductive channel at the interface between a topologically trivial and a non-trivial material, in
which electron backscattering is prohibited [3]–[8], yielding robust metallic conductance
independently of defects and local structural instabilities. This potentially opens a pathway for
completely new concepts in (micro-) electronics.
To this day, all experimentally confirmed topologically non-trivial materials either have an
exceedingly small bulk band gap [9], rendering them uninteresting for technological
applications since a substantial share of the overall electric current would flow though the bulk,
diminishing the advantages of the topologically protected interface conductance, or exhibit
microscopic sample sizes like bismuthene on SiC substrates [10]. Additionally, all
experimentally found topological insulators are band-insulators [9], [11].
The honeycomb transition metal oxide sodium iridate (Na2IrO3) is a promising material in this
context, as it exhibits a band gap (Mott gap) comparable to the band gaps of common
semiconductors and a rich variety of proposed and experimentally reported electronic and
magnetic properties [12]–[20]. Among the anticipated properties is the emergence of a non trivial topological phase, which was first proposed in 2009 [21]. This prediction was reinforced
2
several times, with the latest according work being published in 2019 [22]–[25]. The Mott gap
arises from concurrent crystal field splitting, spin-orbit coupling and Hubbard repulsion
apparent on similar energy scales in the material, making Na2IrO3 a strongly correlated Mott
insulator. Non-trivial topology would render Na2IrO3 to be the first topological insulator with a
large energy gap as well as the first strongly correlated topological insulator. Combined with
sample diameters of up to 1 cm, the material would be interesting for technological application.
Yet, after more than a decade of research, the experimental findings on the Na2IrO3 electronic
structure are largely incoherent, where even the reported Mott gap size varies between
0.34 eV and 1.2 eV [18], [26]–[29] and the role of the surface with its different terminations is
poorly understood.
This thesis provides a comprehensive study of the electronic properties of Na2IrO3, focusing
on its surface. For this, scanning probe and transport measurements on freshly cleaved
samples were performed in ultra-high vacuum (UHV). The new findings yield insights to a
variety of disputed properties of Na2IrO3, including the spectral characteristics at the surface
and their interpretation in the context of strong electron correlation as well as macroscopic
transport properties and the physical picture behind them. Most important, the new results
evidence the existence of a highly conductive surface channel as well as states within the Mott
gap at the surface.
In this work, various physical phenomena apparent in Na2IrO3 are covered, ranging from strong
correlation over topology to transport phenomena. To provide a basis for this, a comprehensive
introduction to the predicted and examined properties of Na2IrO3 is given in the chapters 2 and
3. Here, chapter 2 focuses on the Na2IrO3 bulk as well as the related Mott physics, while
chapter 3 covers the surface characteristics as well as physical properties associated with
topology. Additional physical properties found in experiments that do not arise from the Mott
physics or the surface electronic structure are presented in chapter 4. This is followed in
chapter 5 by the presentation of the experimental techniques and sample preparation
procedures used in the scope of this thesis. All measurement setups are home-built and new
techniques were developed specifically for the work on this thesis. For the two central methods,
being scanning tunnelling microscopy and macroscopic transport measurements, an
introduction to the relevant theoretical concepts and to applicable transport phenomena is
given in chapter 5 as well. The scanning probe results are then presented and discussed in
chapter 6. Chapter 7 covers the transport results and their discussion. Finally, a joint discussion
on the Na2IrO3 surface electronic structure, incorporating the findings from both methods, is
given in chapter 8, before the thesis is concluded with a summary and an outlook. | de |