LiP-CVD growth of multi-shape monolayer WS2: Determination and investigation of defect domains

Abstract

Throughout human history, the discovery and application of new materials have been crucial drivers of civilization's advancement. From ancient epochs to the present day, the utilization of specific materials has often defined eras, emphasizing the profound impact of material innovation on societal progress. Group 4A elements, including silicon, germanium, tin, and lead have played a pivotal role in this narrative, shaping the technological landscape in significant ways. Silicon, in particular, has become synonymous with electronic applications, epitomized by Silicon Valley's status as a global hub of innovation. The inclusion of carbon within Group 4A further underscores the group's importance, highlighting its role as a cornerstone of modern technology. In 2004, Novolesov and Geim's achieved a groundbreaking feat by isolating one-atom-thick graphene using the scotch tape method, a milestone in materials science. This discovery, honored with the Nobel Prize in 2010, propelled graphene into prominence as a highly promising material with broad potential across diverse sectors, including electronics and medicine. The exceptional mechanical, electrical, and optical properties of graphene have prompted scientists to explore new two-dimensional (2D) materials. Transition Metal dichalcogenides (TMDs) are among the most intriguing 2D layered structures. When reduced to few-layer or monolayer thicknesses from their bulk counterparts, they exhibit direct bandgaps, high photoluminescence, and spin-orbit splitting. These unique properties make them promising materials for electronic and optoelectronic applications. In this thesis, we focus on WS2, a remarkable member of TMDs known for its exceptional properties, such as strong photoluminescence, ~ 2 eV direct optical bandgap, and large spin-orbit splitting at the K point, enabling spin-polarized light-matter transduction. These remarkable attributes position WS2 as a promising candidate for advanced technology applications, including field-effect transistors (FETs), photodetectors, and spintronics. In the literature, numerous production techniques for synthesizing TMDs have been documented, each with its drawbacks, such as lack of thickness control and limited scalability. In our work, we opted for Chemical Vapor Deposition (CVD), which offers advantages such as the ability to precisely control thickness and dimensions, produce diverse morphologies, and enable large-scale growth. The flexibility to tailor growth parameters during CVD facilitates the synthesis of WS2 flakes tailored precisely to meet specific requirements and applications. We used liquid precursors (LiP) to overcome the limitations posed by solid precursors, which often require high-temperature CVD processes and tend to yield smaller flakes. LiP-CVD is a synthesis technique which provides several fundamental advantages, including scalability, ease of transfer, cost-effectiveness and the use of precursors with low toxicity. Employing a W-containing precursor in liquid phase allows the reaction with sulfur to occur directly on the growth substrate, thereby reducing the required amount of reactant compared to standard co-evaporation techniques. In addition, the use of a growth promoter – i.e. NaOH – helps to minimize precursor consumption while ramping-up the temperature and to lower the reaction energy with S. This allows to achieve large-size crystals in a relatively short time and at relatively low temperatures, compared to the solid precursor CVD process. We conducted research on CVD synthesis methodology by fine-tuning the growth parameters. Our experimental findings confirmed the alignment of the CVD-grown monolayer WS2 with existing literature. In the first part of the thesis, Raman, PL, and AFM analyses were employed in our preliminary investigations to assess the properties of CVD-grown WS2. The insights gleaned from the analysis results helped us deepen our understanding of critical parameters in CVD, enabling the synthesis of WS2 flakes with diverse characteristics. Previous studies have demonstrated control over WS2 shape by adjusting the amount of sulfur relative to that of W precursor. They found that triangular WS2 flakes were formed within a sulfur-rich environment, while hexagonal WS2 flakes developed in a sulfur-deficient condition. In analogy with those results, we reduced the amount of sulfur reacting with the W precursor by decreasing the Ar background pressure during the growth. Raman and PL measurements indicate the existence of distinct domains within each crystal type, and via selected-area X-ray photoemission spectroscopy (µXPS) measurements, we identified the chemical nature of the defects on each domain, corroborating the recent findings in the literature. Using PL and XPS techniques, distinct domains were distinguished and labeled as VW (dark) and VS (bright), identifying regions with W- and S-rich vacancies characterized by low and high PL intensity, respectively. Kelvin probe force microscopy (KPFM) investigations and micro angle-resolved photoemission spectroscopy (µARPES) analyses were employed to unveil the electronic properties (i.e., electronic affinity and band structure) within chemically different domains. A higher surface potential was observed in the VS domain which also exhibited a deeper binding energy compared to the VW domain. Through gate-dependent PL experiments, we identified the quasiparticle composition of the PL spectrum and proposed a model to hypothesize the relaxation mechanism, suggesting that the presence of W vacancies screens the excitons from the influence of an external gate. In the CVD process, defects are inevitably formed, and their nature, concentration, and distribution profoundly affect the optical and electronic properties of the crystal. In the second part of the thesis, we assessed the influence of the precursor on WS2 flakes by varying the ratio between the metal precursor and the promoter in the growth solution. Additionally, we examined the shape evolution of CVD-grown WS2 based on the component ratio in the growth solution. We observed dendritic structures, which are likely a result of the growth conditions. Through Raman and PL measurements, we found that the field transition observed during shape evolution, similar to previous growth instances, did not significantly manifest in the 2D maps. Notably, the PL intensity demonstrated varying intensities across different regions of the flakes, without exhibiting a dramatic decrease. Apart from the dendritic structure, which prompted consideration of stress-strain within the crystal during CVD, we also observed E2g redshift in some flakes. To preserve the intrinsic properties of the WS2 crystals for long-term research, we carefully managed their storage conditions, particularly with regard to exposure to atmospheric conditions and natural light. Upon exposure to daily atmospheric conditions, including humidity, we noted a gradual enlargement of residual promoter particles on and underneath the substrate, resulting in increased surface roughness over time. Additionally, we explored the significant impact of natural light on hexagonally shaped WS2 crystals, especially those exhibiting VW and VS domains. Our findings revealed that VW domains were more sensitive to light compared to VS domains.