PhD defence Frank Wiggers

surface science studies of epitaxial silicene 

Silicene is a two-dimensional (2D) material, consisting of an atomically buckled, honeycomb lattice of Si atoms. This material is predicted to host charge carriers that behave as massless Dirac fermions, and exhibit the quantum spin Hall effect. This latter property imparts silicene with the behavior of a 2D topological insulator. A topological insulator is predicted to have special properties that could be useful for applications ranging from spintronics to quantum computation.

The main experimental challenges for silicene are the synthesis of quasi free-standing silicene, and the encapsulation of these layers. Encapsulation is necessary, because silicene is chemically reactive and would, for example, quickly oxidize in air. If silicene layers are to be integrated in electronic devices, then they will have to be covered with an insulating layer, in order to protect them. In addition, encapsulation layers that are conductive but which can be easily desorbed, e.g. by an annealing treatment, can also be useful, enabling simplified transport of silicene layers between ultra-high vacuum systems.

The ZrB2 substrate is used, because it is one of a limited number of substrates on which epitaxial silicene has been synthesized, and because it spontaneously forms a self-terminating, single phase, silicene monolayer upon annealing, which does not require Si deposition. We regard this epitaxial silicene layer as a model system of quasi free-standing silicene on which we performed experiments in order to find suitable materials and methods for its encapsulation. In addition, the results of this research are also useful in the development of encapsulation layers for other 2D materials or ultra-thin films.

In chapter two through five, experiments were discussed that involve the adsorption of materials on epitaxial silicene, with the aim of forming an encapsulation layer. These experiments, including the one discussed in chapter six, all used synchrotron-based high-resolution photoelectron spectroscopy, and low-energy electron diffraction. Further, for the experiments in chapter seven, scanning tunneling microscopy (STM) and synchrotron-based angle-resolved photoelectron spectroscopy were also used.

In chapter 2, we studied the interaction between epitaxial silicene and Al atoms, O2 molecules, and a combination of the two. The deposition of Al atoms onto silicene, up to a coverage of about 0.4 Al atom per Si atom, has little effect on the chemical state of the Si atoms, the surface reconstruction and the buckling of silicene, indicating weak interactions between Al and silicene. The silicene-terminated surface is also hardly affected by an exposure to 4500 L O2. In contrast, when Al-covered silicene is exposed to the same O2 dose, a large fraction of the Si atoms become oxidized. This is attributed to dissociative chemisorption of O2 molecules due to the presence of Al atoms, producing reactive atomic oxygen species that cause the oxidation. It is concluded that aluminum oxide overlayers prepared in this fashion are not suitable for encapsulation, since they do not prevent but actually enhance the degradation of silicene on ZrB2(0001) thin films.

In chapter 3, the feasibility of forming an AlN encapsulation layer on top of epitaxial silicene was investigated, using trimethylaluminum (TMA) and ammonia (NH3) as precursor gases. Exposing the sample held at 300 °C to TMA molecules results in a self-limiting dissociative chemisorption process of TMA, which gives rise to various reaction products on the surface, in particular Al–CH3 and Si–CHx species. Upon subsequent exposure to NH3 at 400 °C, nitridation of the surface is observed involving the formation of Al–N and Si–N bonds. While some loss of C atoms is observed in this process, most probably due to the formation of volatile CH4 in reactions between NH3 and Al–CH3 species, the majority of surface carbon remains. While an AlN-related layer can indeed be grown, silicene reacts strongly with both precursor molecules resulting in the formation of Si–C and Si–N bonds, such that the use of these precursors does not allow for the protective AlN encapsulation that leaves the electronic properties of silicene intact.

In chapter 4, we studied the interaction between NaCl and epitaxial silicene. The deposition of NaCl took place with the substrate at room temperature. This caused the appearance of Si 2p intensity overlapping with the expected binding energies of Si–Clx (x = 1-3) species, as a result of dissociative chemisorption of NaCl. With increased NaCl deposition, the majority of the epitaxial silicene reacted to form Si–Clx species.

In chapter 5, the interaction of elemental Se with epitaxial silicene was investigated. The deposition of Se at room temperature resulted in chemical bonding between the deposited Se and Si from silicene. This caused the silicene-related Si 2p peaks to disappear, while new components appeared with chemical shifts of n × 0.51 ± 0.04 eV (n = 1-4), suggesting the formation of SiSe2 and intermediate species. This demonstrates that capping the silicene monolayer, without affecting its structural and electronic properties, is not possible with Se. The annealing treatments up to 470 °C that followed, caused the desorption of Se and Si, resulting in the etching of the Si atoms formerly part of the silicene layer. This resulted in the formation of bare ZrB2(0001) surface area coexisting with a novel surface reconstruction, attributed to Se-terminated ZrB2(0001)-(√7×3)R40.9°. The Se was removed by annealing to 600 °C, which also resulted in the simultaneous restoration of the silicene monolayer by Si originating from the Si substrate.

In chapter 6, the chemical reaction of NH3 with epitaxial silicene on ZrB2 is investigated. Dissociative chemisorption of NH3 on silicene was found to take place at 400 °C and nitrided the surface. At 300 °C, no chemical reaction between silicene and NH3 was observed which demonstrates that the silicene-terminated surface is less chemically reactive towards NH3 molecules than a Si(111) or Si(100) surface for which dissociative chemisorption of NH3 already occurs at room temperature. Furthermore, annealing the nitrided surface up to 830 °C resulted in a solid-phase reaction with the substrate that formed a new epitaxial surface layer with hexagonal symmetry and a single in-plane crystal orientation, involving B, N, and Si atoms.