Rohan V's Senior Project Blog

Project Title: Into the noise: developing a new change point detection algorithm to interpret single-molecule conductance BASIS Advisor: John Goodwin Internship Location: LabMonti, University of Arizona  Onsite Mentor: Oliver Monti

Project Abstract

Modern semiconductor technology lies at a critical junction as exponentially rising costs prompt the development of alternative technologies (Ivie et. al). Specifically, the electronics industry is turning towards organic, or carbon-based, semiconductors (OSCs), which have the potential to bypass the efficiency limits of silicon-based technology. When OSCs are taken small enough - to the ‘ultimate size limit’ of a single molecule - they can function in new circuits that use quantum mechanical effects to enhance conductivity. In the last fifty years, researchers have developed novel methods to measure the conductive properties of such single-molecule organic semiconductors. One method, known as the mechanically-controlled break junction (MCBJ), involves placing a molecule between two metal electrodes, passing current through the system and recording molecular conductance. Because of the small scale, experiments must be repeated thousands of times to produce reliable results. This, in turn, confers a high degree of variability to the resulting datasets, rendering them difficult to interpret. Researchers have come up with many approaches for analyzing experimental datasets, with most using advanced machine-learning algorithms to identify the “signature” conductance of a molecule. My research project lies in more deeply exploring how molecular conductance changes over time; I hypothesize that assigning a single conductance value to a molecule overlooks significant conductance ‘substructures’ that affect how the molecule functions in a device. Through applying change-point detection methodology in MATLAB, I aim to create a program that can find and classify changes in conductance as the junction evolves. Accurate characterization of molecular properties will, in turn, pave the way for implementation of OSCs in electronics and solar cells, increasing energy efficiency at an essential time for the globe.