Machine learning (ML) has been flourishing in various fields, including image recognition, natural language processing and protein structure analysis. In recent years, it is also attracting considerable attention in the field of optoelectronics. Researchers not only use ML tools to improve the optoelectronic device properties but also develop new optoelectronic materials and devices to build neuromorphic computers. Neuromorphic computers can run ML algorithms efficiently by simultaneous processing and storing the data. In this dissertation, both directions are presented.
Recently, thin-film perovskite solar cells (PSCs) are gaining considerable attention due to their high power conversion efficiency, solution processability and low cost. In the first part of this dissertation, ML is used to predict perovskite material properties and the performances of PSCs. Additionally, several strategies to design high-performing PSCs, including the selection of electron and hole transporting materials, are discussed based on the predictions from ML models. New materials compositions were also developed, validating the predictions from ML.
Second, new types of optoelectronic synapses are fabricated as potential building blocks of neuromorphic computers. By following a heterogeneous nucleation approach to grow perovskite quantum dots (PQDs) on multi-wall carbon nanotubes (MWCNTs) and graphene, new hybrid materials are synthesized and used to fabricate optoelectronic synapses. The potentiation in the synapses is realized by light pulses to obtain photonic memory and the depression is accomplished by electrical pulses to erase the memory. With the assistance of simulations using the device properties, the performances of these devices as the potential building blocks of optoelectronic neuromorphic computers are tested.
Organic electrochemical transistors (OECTs) are attractive for biosensing, building blocks of neuromorphic computers and the human-machine interface. Finally, the properties of OECTs incorporating plasmonics are investigated. An in-house developed inexpensive nanoimprinting method is used to fabricate plasmonic substrates to fabricate plasmonic OECTs. Using glucose sensing as proof, the developed plasmonic OECTs provided sensitivity under the light at a given concentration while regular OECTs did not show any response. This makes them a potential candidate for biosensing, human-machine interfaces and optogenetic inspired neuromorphic computers.
Major: Optics and Photonics
Educational Career:
BS: 2015, Optical Information Science and Technology, Beijing Institute of Technology
MS: 20204, Optics and Photonics, College of Optics and Photonics, University of Central Florida
Committee in Charge:
Dr. Jayan Thomas (Chair)
Dr. Winston V. Schoenfeld
Dr. Aravinda Kar
Dr. C. Kyle Renshaw
Dr. Tania Roy
Approved for distribution by Dr. Jayan Thomas , Committee Chair, on November 1, 2020.
The public is welcome to attend.
Join Zoom Meeting
https://ucf.zoom.us/j/91483721883?pwd=dVZEaFBtU3pSN1RnYS91ZXcvc1RKZz09
Meeting ID: 914 8372 1883
Passcode: 690910
Read More