Presentation Type
Poster
Presentation Type
Submission
Department
Physics
Major
Physics
Abstract
Transition metal dichalcogenides (TMDCs) like molybdenum disulfide (MoS2) possess unique electronic and optical properties, making them promising materials for nanotechnology. Photoluminescence (PL) is a key indicator of MoS2 crystal quality. This study aimed to develop a machine-learning model capable of predicting the peak PL wavelength of single MoS2 crystals based on micrograph analysis. Our limited ability to consistently synthesize high-quality MoS2 crystals hampered our ability to create a large set of training data. The project focus shifted towards improving MoS2 crystal synthesis to generate improved training data. We implemented a novel approach utilizing low-pressure chemical vapor deposition (LPCVD) combined with established synthesis techniques to produce MoS2 monolayers with superior quality and surface area. This method achieved high reproducibility, facilitating the collection of improved training data.
Faculty Mentor
Dr. John Mann
Funding Source or Research Program
Keck Scholars Program
Location
Waves Cafeteria
Start Date
22-3-2024 1:30 PM
End Date
22-3-2024 2:30 PM
Included in
Condensed Matter Physics Commons, Data Science Commons, Semiconductor and Optical Materials Commons
Machine Learning Prediction of Photoluminescence in MoS2: Challenges in Data Acquisition and a Solution via Improved Crystal Synthesis
Waves Cafeteria
Transition metal dichalcogenides (TMDCs) like molybdenum disulfide (MoS2) possess unique electronic and optical properties, making them promising materials for nanotechnology. Photoluminescence (PL) is a key indicator of MoS2 crystal quality. This study aimed to develop a machine-learning model capable of predicting the peak PL wavelength of single MoS2 crystals based on micrograph analysis. Our limited ability to consistently synthesize high-quality MoS2 crystals hampered our ability to create a large set of training data. The project focus shifted towards improving MoS2 crystal synthesis to generate improved training data. We implemented a novel approach utilizing low-pressure chemical vapor deposition (LPCVD) combined with established synthesis techniques to produce MoS2 monolayers with superior quality and surface area. This method achieved high reproducibility, facilitating the collection of improved training data.