With technology manufacturers pushing for smaller and smaller devices, there is an increasing demand for smaller and smaller electrical materials. In response to this demand, materials scientists are making great efforts to control the shapes of materials on the scale of their molecule building blocks. With this level of shape control, we can imagine electrical wires with diameters 1/100,000th of a piece of hair, circuitry that could fit on the tip of a needle, and a dramatic reduction in the size of modern electronics.

Our research uses theoretical methods to predict how to control the shapes of materials on the molecular scale. We employ techniques from theoretical chemistry (mainly statistical mechanics and quantum chemistry) and applied mathematics (mainly statistics, machine learning, and probability), and develop our own techniques whenever needed. We work in close collaboration with several experimental groups, and aim for tangible realizations of our theoretical predictions.

Inquiries are most welcome (replace AT with @)



Bayesian Optimization for Materials Science is now available from Springer. This short monograph is intended for materials scientists who are interested in machine learning techniques, and has been published as part of the series SpringerBriefs in the Mathematics of Materials

Article published in the Japanese chemistry magazine Kagaku (English title: Prediction of graphene nanoribbon assembly via a mathematical model - the counter-intuitive role of entropy)

Daniel Packwood speaks at Mathematics Summer (数理サマー), a public event organized by RIKEN iTHEMS, Kyoto University RIMS, and Kyoto University KUIAS.

Paper on dye-sensitized solar cells published in ChemSusChem (collaboration led by the Imahori group from iCeMS)

Daniel Packwood speaks at TedXKyotoUniversity

Guoxi Zhang from the Kyoto University Graduate School of Informatics joins as a summer student.

Paper on Bayesian optimisation for surface-adsorbed molecules published in Applied Physics Express.

Paper on surface-assisted molecular self-assembly published in Nature Communications