Lateral heterostructures, alloys, and the Kirkendall effect in transition metal dichalcogenide monolayers

This dissertation analyzes two-dimensional structures and effects involving Mo, Ta, and S on Au(111), using scanning tunneling microscopy (STM). The structures presented include monolayer dislocation-free concentric lateral heterostructures (CLHs) between MoS2 and TaS2, alloys and a kagome structure. All of these are created utilizing epitaxial growth methods, including a mix of physical vapor and chemical vapor deposition. Physical vapor deposition supplies the transition metals Ta and Mo, and chemical vapor deposition using H2S, supplies the S components. By fine tuning of the growth parameters the differing structures could be realized with precision and consistency.
The analyzed CLHs consist of a core of MoS2 and a surrounding envelope of TaS2, which represents a Schottky contact with maximally reduced dimensionality in the z-direction. Structures of this kind are of great interest as model systems for device manufacturing with reduced dimensionalities. While a growth process was already established in previous works, the nature of the interconnection of these two materials, which exhibit a lattice constant mismatch of ∼4.5 %, was left unresolved. This thesis shows that each CLHs presents two types of dislocation-free interfaces, compensating lattice constant mismatch not via dislocations but interface-type dependent deformations. The interface types have been identified from the perviously unknown stacking registries of the CLH with the substrate. These were determined from the distribution of the orientations of CLHs within the hcp- and fcc-stacked areas of the Au(111) substrate, in comparison to free-standing islands of MoS2. Additionally, the edge terminations of the CLHs could be determined from these results. The majority of the deformation was shown to accumulate in the TaS2 envelope. This envelope contracted in both interfaces to match the MoS2 core, which only showed little expansion. In one interface a relaxation of the envelope lattice constant with distance to the interface was observed. Notably, with significant distance, even an extension beyond the equilibrium lattice constant was detected. Empirical potentials and finite elements simulations were performed, showing that most of the effects were explainable by simple mechanics.
Modifying the growth process, originally designed to grow CLHs, Mo_(1-x)Ta_(x)S2 alloys were observed. The dependence of alloy formation on the new growth parameters has been analyzed and, utilizing a combination of analysis methods, the ranges of mixtures inside alloys with different growth parameters were estimated. A continuous change from CLHs to alloys with increase in annealing temperature has been found. The investigation of these alloys was further expanded upon by the alteration of the growth process, resulting in the observation of additional structures and variations.
Another new structure was discovered as a byproduct of various growth processes. A structure crystallizing in the distinctive kagome lattice, aptly called kagome structure. Topography analysis was used to calculate a first estimation of the lattice constant of the structure. Additionally, a synthesization method to grow large areas of the material was developed.
Lastly, Kirkendall holes, the 2D equivalent of the 3D Kirkendall voids, were discovered as a result of the Kirkendall effect. The detection took place after a special growth process which resulted in the formation of Mo_(1-x)Ta_(x)S2 alloys with gradual rates of mixture. Local changes of alloy lattice constants and coverages rates of holes and alloys, in comparison to CLHs, were used to analyze the resulting structures. A model of the growth process was proposed with the help of DFT calculations and a Monte Carlo simulation, showing that the Kirkendall effect can be transferred from 3D to 2D applications.