Colloquium

The efficiency of a thermoelectric (TE) device depends on the extent to which, in response to a given temperature gradient, its electron/hole transport symmetry at the Fermi level is broken. This requirement makes molecular junctions highly promising for TE applications due to their non-linear transmission properties. Yet, in the absence of an efficient method to tune the position of the Fermi level within the transmission landscape of these junctions, the Seebeck values of metal-molecules-metal junctions are typically |S|≤50μV/K, while based on their electrical and thermal conductance, it should be |S|≥1mV/K to be relevant for applications. I will describe our effort to reach this goal, which recently has culminated in molecular junctions with the semimetal Bismuth (Bi) as one of their leads and with |S| in the required mV/K range. Unlike the conventional approach to tweak the transmission properties by modifying the structure of the molecules, here the high Seebeck is a result of molecularly induced deterministic changes in the density of states within the Bi lead in the form of quantized 2D interfacial states, that in turn result in highly non-linear transport properties. I will argue that this effect is just one glimpse into the very rich and complex terra incognita of molecular layers on semimetals.

This talk tells my outlook on the development of electric-field-mediated-chemistry/biochemistry and predicts a vision of its future state.1 The talk discusses applications of oriented electric-fields (OEFs) to chemical and biochemical reactions e.g., Diels  Alder reactions, and reactions of the enzyme Cytochrome P450. As shall be demonstrated, the orientation of the OEF controls reaction-rate (acceleration/inhibition), chemo-selectivity, enantio-selectivity, and solvent effect. This will be followed by showing relevant experimental verifications of the impact of OEF on structure and reactivity. Subsequently, the talk will outline other ways of generating OEFs, e.g. by use of; pH-switchable charges, ionic additives, water droplets, and so on. I shall further describe the application of static vs. oscillating OEFs to decompose peptide plaques (e.g., Amyloid Plaques in Alzheimer’s disease). The second part of the talk consists of conceptual principles for understanding and predicting OEF effects, e.g., the “reaction-axis rule”, the capability of OEFs to act as tweezers that orient reactants and accelerate their reaction, etc. Finally, I shall discuss the prospects of up-scaling applications of various OEF-sources to Molar concentrations. The talk ends with the vision that, in the forthcoming years, OEF usage will change chemical education, if not also the art of making new molecules.

Learning from data has led to paradigm shifts in a multitude of disciplines, including web, text and image search and generation, speech recognition, as well as bioinformatics. Can machine learning enable similar breakthroughs in understanding (quantum) molecules and materials? Aiming towards a unified machine learning (ML) model of molecular interactions in chemical space, I will discuss the potential and challenges for using ML techniques in chemistry and physics. ML methods can not only accurately estimate molecular properties of large datasets, but they can also lead to new insights into chemical similarity, aromaticity, reactivity, and molecular dynamics. For example, the combination of reliable molecular data with ML methods has enabled a fully quantitative simulation of protein dynamics in water (https://arxiv.org/abs/2205.08306). While the potential of machine learning for revealing insights into molecules and materials is high, I will conclude my talk by discussing the many remaining challenges.

atom-probe tomograph (APT) can dissect a nanotip shaped specimen (radius