nanoscale properties of 2D materials
Rik van Bremen is a PhD student in the research group Physics of Interfaces and Nanomaterials. His supervisor is prof.dr.ir. H.J.W. Zandvliet from the Faculty of Science and Technology.
The class of two-dimensional (2D) materials are materials that are only one atom layer thick. Graphene was the first 2D material to be isolated. Geim and Novoselov were awarded the Nobel prize in Physics in 2010 for this discovery. It was discovered that graphene has many special properties, completely different from those associated with bulk materials. As for its mechanical properties, graphene is the thinnest material possible as it is only one atom
thick, but it is also the strongest material known to date while at the same time being very flexible. But most interestingly are its special electronic properties; electrons in graphene behave as relativistic particles and are not backscattered by defects, a phenomenon called Klein tunneling.
The discovery of graphene opened the field of 2D materials, as other single layer materials were soon also isolated, such as hexagonal boron nitride and the family of transition metal dichalcogenides. All these materials have their own unique properties and advantages. In bulk form, these materials consist of multiple layers that are bonded to each other out of plane via weak van der Waals forces. Once different single layer 2D materials are isolated, these can be re-stacked on top of each other in different combinations and orders to create new heterostructures. Challenges remain for 2D materials to be incorporated in future devices, such as preparing large, high quality 2D materials, characterizing the properties of the 2D materials, and applying contacts to the 2D material to connect them to the rest of the device.
We use scanning probe microscopy (SPM) tools that allow for the characterization of 2D materials at the nanoscale. SPM’s are based on a nanoscale probe that is raster scanned over the surface of a sample. The interaction between probe and sample is recorded during scanning. Depending on the probe and method used, a multitude of different properties can be measured. Due to their high spatial resolution, spectroscopic abilities and surface sensitivity, SPM’s are ideal surface science techniques to characterize 2D materials.
However, SPM characterization of these layered materials is not trivial. The interaction between probe and sample can deform, damage and even exfoliate the top layers of 2D materials. We show that in the scanning tunneling microscope, the top layer can be lifted for hundreds of nanometers from the surface by the probe. This lifting behavior is caused by the applied bias voltage between tip and sample which induces an attractive electrostatic force. In the atomic force microscope the forces necessary to lift the sample can be measured. Spring constants of 1517 N/m and 453 N/m were measured for graphite and molybdenum disulfide, respectively.
In Chapter 3, germanene growth on Ge2Pt clusters is studied by a combination of scanning probe microscopy and electron diffraction. Germanene is the germanium analogue of graphene. Germanene has the advantage over graphene in that a band-gap can be opened in germanene. This will allow germanene to be used as a field effect transistor. However, germanene does not exist in nature and has to be synthesized. One method that is proposed to grow germanene is by use of a solid-liquid-solid growth procedure. Platinum is deposited on a germanium sample. The sample is consequently heated above 1043 K, such that eutectic Ge-Pt droplets form on the surface. Upon cooling down, these droplets crystallize in two phases, Ge2Pt clusters and pure germanium. It has been proposed that some of the germanium forms a germanene layer on the outside facets of the Ge2Pt clusters. With the use of electron backscatter diffraction and scanning probe techniques, the relation between the germanium ad-layer and the cluster crystal structure has been identified. It has also been shown that the multiple phases that have been measured on the clusters can be related to the same crystal lattice.
Chapters 4 and 5 describe local contact properties of nanoscale probes to 2D materials. Metal contacts are necessarily applied to semiconducting 2D materials when incorporated in devices. The use of these contacts has been found to drastically influence the properties of the 2D materials. By use of a conductive AFM, we found that nanoscale metal-like defects are the cause of the reduced Schottky barrier height of molybdenum disulfide, which was previously not understood when measured by macroscopic contacts. The reduction of the barrier height is induced by a strong Fermi level pinning with a pinning factor of ~0.1 at the defects. As a consequence, current is mainly injected at these defect sites.
We applied the same conductive AFM technique to characterize the charge injection mechanism of a metal probe to thin layers of tungsten disulfide. The tungsten disulfide is placed on a mica substrate. During this process, a hydration layer is encapsulated between tungsten disulfide and mica. By changing the humidity of the environment, the state of the hydration layer can be altered. It is shown that the charge carrier injection mechanism depends on the state of the hydration layer. This environmentally controlled process is the consequence of doping due to charge transfer from the underlying hydration layer. In addition, the charge injection mechanism strongly depends on the thickness of the WS2 flake. Thermionic emission was only observed on multilayer WS2 while for single layer, bilayer and few layer WS2, the injection mechanism was due to direct tunneling or Fowler-Nordheim tunneling. For these thinner layers, a switch from unipolar to ambipolar behavior was
observed when decreasing the thickness of the intercalated water from two layers to a single layer. The reversibility of the dewetting process along with its non-invasive character provides a viable alternative for fabricating devices with in situ control of the metal-semiconductor junction characteristics.
In the last chapter, Chapter 6, thermovoltage scanning tunneling microscopy is used to measure Friedel oscillations on gold and graphite. The tunnel current in thermovoltage scanning tunneling microscopy is not induced by a voltage difference, as in normal STM, but rather by a temperature difference between probe and sample. This makes thermovoltage scanning tunneling microscopy extremely sensitive to small changes in the electron density near the Fermi level, such as Friedel oscillations. The Friedel oscillations occur when electron waves scatter of defects and interfere with themselves, resulting in standing wave patterns. The wavelength and direction of the Friedel oscillations depend on the incoming and scattered electron waves. When all scattering events are equally likely, the reciprocal space image of Friedel oscillations correspond to the joint density of states. Any deviations from the reciprocal space image to the joint density of states are due to suppressed scattering events. Thermovoltage scanning tunneling microscopy has been used to image intravalley and intervalley scattering events in graphite. Thermovoltage signals are also measured on a single layer of graphene on boron nitride. This will allow further experiments to investigate Friedel oscillations and scattering events under the application of a gate voltage.