A flexible silicon barrier diode was fabricated by the transfer printing method. Micro-line patterned p-type single crystalline silicon membranes were created from a silicon on insulator wafer. The dark current of our device was very low, about 1 pA for reverse bias voltages up to 5 V, and showed rectifying behavior with an ideality factor of 1.05. The photo-response and the responsivity was 32 and 0.3 A/W, respectively, for light intensity of 1.2 mW/cm2. Also, the current of the photodetector changed under compressive stress or tensile stress. Our device is functional as the piezotronic sensor as well as the photodetector.

This paper presents low temperature solution-processed fabrication techniques for modern thin-film transistors (TFTs). We have investigated the electrical performance of aqueous solution-processed amorphous indium oxide (a-In2O3) TFTs prepared using different annealing temperatures. Even though the a-In2O3 TFTs were annealed at 200 °C, electrical characteristics of aqueous solution-processed a-In2O3 TFTs were obtained. High performance such as a saturation mobility of 8.6 cm2 V-1 s-1 and an on/off current ratio of over 106 was exhibited by a-In2O3 TFTs annealed at 250 °C.

No abstract

Deep learning is a rapidly growing technology repeating epoch-making development in the field of voice/text/image cognition. Its basic principle is to systematize information and let users find the pattern for themselves through the neural network using lots of layers. Technological core is anticipation by classification. This thesis uses SNS and webpage scrapping data and GIS data for consumer needs. Data will then be extracted by accurate classification for the purpose of spatial information data with deep learning algorithm. It is necessary to call shapefiles to R, improve the accessibility to data, and cross one data set to other data set areas. This thesis intends to analyze data of various environments with data analysis tool, R, and design the process combining data of spatial information and visualizing it based on deep learning algorithm

The introduction of GaAs grown at low MBE growth temperatures has spurred considerable activity in attempts to understand conduction mechanisms and optical properties. In LT GaAs, the formation of microscopic As precipitates dominates the conductivity, producing electronic transport mainly by variable range hopping conduction. The resulting high resistivity and short carrier lifetimes have enabled the use of LT GaAs in FET buffer layers as well as in ultra-fast optical switches. An extension to AlInAs has also been performed and it was seen that variable range hopping is also present in the Asbased ternary.We report the optical and electronic properties of InP grown at low temperatures in a gas source MBE using dimeric phosphorus produced from cracked phosphine. The conductivity is higher than the equivalent GaAs LT material and does not have the same temperature dependence. The conditions under which growth occurs ie, substrate temperatures, V/ill ratios and annealing is explored. The structural properties, temperature dependence of the conductivity, deep level structure and the photoluminescence properties of the material are also investigated.