2024 AIChE Annual Meeting
Statistical Analysis of Environmental Factors on SAM Systems
By 2036, continuous resolution increase in CMOS chips in computers will become technologically demanding, leading to an invalidation of Moore’s Law. As an alternative, Organic Molecular Electronic Devices could potentially achieve Angstrom-scale resolution. Self-Assembled Monolayers (SAMs) derived from organic molecules, coupled with liquid metal EGaIn electrodes, provide convenient sample preparations and electrical measurement capable of delineating angstrom scale electrical properties at ambient. Due to its small scale, however, minor disturbances to the SAMs system can result in major errors in measurement. Thus, a large number of current density measurements are needed to statistically account for stochastic perturbations. To examine such environmental effects, these measurements are collected over various seasons, times, and locations leading to complexity of the errors. After, various statistical plots are used to correlate performance and reliability with various metadata categories. Heatmaps of current density and voltage measurements tell of consistency and performance while scatterplots of circuit shorts tell of device reliability and error. With these plots, it was found that the season and time greatly affected the performance and reliability of alkanethiol SAMs irrespective of the substrate or molecular length. We hypothesize that humidity differences between seasons cause the above phenomena due to the interaction between water, electrodes, and SAMs. A regression-based Machine Learning algorithm was designed based on these trends to designate the most optimal season and time based on the SAM system chosen. In the future, the optimization of the algorithm along with experimental testing of the device against predictions will be performed. This research was supported by the National Science Foundation DMR-2150360 "REU Site: Materials Research with Data Science (MAT-DAT)".