Atomically resolved imaging of ion-Implanted two-dimensional transition metal dichalcogenides

Monolayer-thick transition metal dichalcogenides (TMDs) with the chemical formula MX2 (M = Mo, W; X = S, Se) constitute a new class of direct band gap semiconductors. They exhibit several favourable material properties, such as high mechanical flexibility, high photoresponsivity, and impressive electrical performance. Monolayer TMDs can also be incorporated into electrically driven devices, which in turn can be coupled to optical microcavities or photonic circuits. In order to continue the development of such technologies, however, new methods of material modification must be developed and tailored for application to these materials. One such technique is ultra-low energy ion implantation, which allows for highly pure, clean and selective substitutional incorporation of dopants.This method is compatible with standard semiconductor processing and offers an expanded selection of possible dopants compared to the other popular doping mechanisms.

In this thesis, results of ultra-low energy (10 - 25 eV) ion implantation of mono-layer TMDs are reported. Successful implantation of foreign ions was achieved using the ADONIS mass-selected ion beam deposition system at the University of Göttingen . Transition metal (Cr) and chalcogen (S, Se) dopants were introduced into the MoS2 and MoSe2 crystal lattice. Atomic resolution high angle annular dark field (HAADF) scanning transmission electron microscopy (STEM) and core loss electron energy loss spectroscopy (EELS) were used to identify the sites of individual dopants in the host lattice and examine the atomic structure of the defects and dopants in the monolayers on the sub-atomic scale.

Analysis of experimental HAADF STEM data was carried out using open source Python libraries to determine the percentage of ions that cause substitutional implantation, create adatoms, and create vacancies. The Model Refiner class from the TEMUL Toolkit Python package was developed to automatically identify and characterise each atomic site in the HAADF STEM image. Accurate atom statistics can be obtained from non-ideal STEM imaging conditions through post-processing and avoidance of contaminated areas when extracting atom sites. Furthermore, by simulating images based on the atomic arrangements determined via the software package, and by comparing these and achieving agreement with the experimental images, the individual atom positions of the dopants can be revealed with sub-Angstrom accuracy.

This thesis constitutes part of a proof-of-principle study concerning the possibility and procedure of incorporating non-classical single photon emitting diodes into monolayer TMDs. The development of such devices has far-reaching implications for quantum science and technology, with applications in the fields of quantum cryptography and quantum metrology.