Skip to content

Speaker Abstracts

Speakers & Abstracts

(Click on each tab to reveal title and abstract)

Quantum Chemical Alchemy: Dynamical Gold from Born-Oppenheimer Lead

The Born-Oppenheimer (BO) approximation is the basis for the description of a vast array of chemical processes that largely occur on a single potential energy surface. The BO electronic states adjust their character to nuclear position and continuously change from reactants to products as the nuclei move. However, many processes involve systems evolving on more than one potential surface (e.g. electron transfer reactions) and these reactions are more easily described by states that retain their electronic character as the nuclei move. The Generalized Mulliken-Hush method is a flexible approach for extracting diabatic states and surfaces for electron transfer processes from BO data. The method and its advantages will be discussed and several applications will be presented.

Understanding the Structure of Structureless Regions in the Genome and the Transcriptome

While the genome is organized primarily in canonical double-stranded form, DNA is exposed as single-stranded (ss) structures during many parts of the life cycle of cell. The diverse conformations of these transitory structureless stretches of ssDNA are characterized by thermal ensembles, but structural information about them is lacking. I will describe theoretical and computational approaches to understanding these structureless ssDNAs and their transitions back to folded forms, emphasizing the physical chemical forces that drive DNA and RNA structures. At the core of these driving forces, entropy plays surprisingly key roles in many different ways.

Accurate electronic excitations with stochastic many-body methods

I will present stochastic approaches to many-body perturbation theory, which accelerate calculations of quasiparticle energies and increase their accuracy. A new concept of sparse stochastic compression is used to speed up computations and leads to a decrease of statistical errors in large (finite and periodic) systems. It is thus feasible to treat systems with more than 10,000 electrons and simulate their photoemission spectra. Predictions of quasiparticle energies are improved by a simplified self-consistency approach, which can be implemented at zero additional cost irrespective of the system size. In the last part of my talk, I will introduce the most recent developments of beyond-GW approaches with stochastic vertex corrections.

Green function methods for optoelectronics and molecular dynamics

We discuss Green function methods as conveninet theoretical tools for description of open nonequilibrium molecular systems. In particular, we focus on the standard nonequilibrium Green function (NEGF) methodology and its many-body flavors: the pseudoparticle (PP-) and Hubbard NEGF. For a system characterized by strengths of intra-molecular ($U$) and molecule-baths ($Gamma$) couplings, the NEGF is convenient in the limit $UllGamma$, while the PP- and Hubbard NEGF are most suitable for the $UggGamma$ range of parameters. We also discuss intermediate regime, $UsimGamma$, where the auxiliary master equation-dual fermion (auxDF) method is a convenient alternative. As an application of the Green function methods we focus on optical spectroscopy and on molecular dynamics in single molecule junctions. Optical spectroscopy of open nonequilibrium systems is a natural meeting point for (at least) two research areas: nonlinear optical spectroscopy and quantum transport, each with its own theoretical toolbox. We argue that theoretical approaches of the quantum transport community (and in particular, the Green function based considerations) yield a convenient tool for optoelectronics when the radiation field is treated classically, and that differences between the toolboxes may become critical when studying the quantum radiation field in junctions. A crucial part of formulating molecular dynamics in junctions is the definition of nuclear forces induced by the nonequilibriium electronic subsystem. We present general first-principles derivation of the expression for current-induced forces. The expression is applicable in nonequilibrium molecular systems with arbitrary intra-molecular interactions and for any electron−nuclei coupling. We derive results of previous considerations (and in particular the celebrated Head-Gordon and Tully expression for electronic friction) as limiting cases of our general expression and discuss effective ways to evaluate the friction tensor in single molecule junctions. Finally, we consider nonadiabatic molecular dynamics (NAMD) at molecule-metal interfaces. Utilizing many-body flavor of the NEGF we derive a set of equations for the nuclear dynamics in the presence of nonadiabatic electronic transitions between different molecular charge states. The equations are shown to reproduce the surface-hopping formulation in the limit of small metal−molecule coupling (where broadening of the molecular energy levels can be disregarded) and Ehrenfest dynamics (motion on the potential of mean force) when information on the different charging states is traced out.

Dynamics of O2 formation in hyperthermal collisions of CO2 with Au surfaces

The intramolecular conversion of CO2 to molecular oxygen is an exotic reaction, rarely observed even with extreme optical or electronic excitation means. Here we show that this reaction occurs readily when CO2+ scatters from solid surfaces in a two-step sequential collision process at hyperthermal incidence energies. The produced O2 is preferentially ionized by charge transfer from the surface over the predominant atomic oxygen product, leading to direct detection of both O2+ and O2-. First-principles simulations of the collisional dynamics reveal that O2 production proceeds via strongly-bent CO2 configurations, without visiting other intermediates. Bent CO2 provides dynamic access to the symmetric dissociation of CO2 to C+O2 with a calculated yield of 1 to 2% depending on molecular orientation. This unexpected collision-induced transformation of individual CO2 molecules provides an accessible pathway for generating O2 in astrophysical environments and may inspire plasma-driven electro- and photo-catalytic strategies for terrestrial CO2 reduction.

Computational chemistry meets photochemistry – Applications from photobiology to atmospheric chemistry

The recent theoretical developments of electronic structure methods for electronically excited states have opened the door for computational chemistry to study photochemical reactions. For instance time-dependent density functional theory (TDDFT) and more accurate approximate coupled cluster methods (CC2) allow to study excited states of large molecular systems. In this talk I will review theoretical developments to study the dynamics of excited state reactions. I will show how we use methods to unravel experimental findings in the photochemistry of vitamin D, photochemical switches, and atmospheric chemistry. A major focus is set on the simulation of ultrafast pump-probe experiments, and the prediction of electronic spectra and product quantum yields. Besides gas phase photochemical reactions, I will also discuss methods to include the chemical environment of a solvent.

Effect of surfaces and osmolytes in modulating peptide assembly

Intrinsically disordered peptides are a special class of proteins that do not fold to a unique three-dimensional shape. These proteins play important roles in the cell, from signaling to serving as structural scaffolds. Under pathological conditions, these proteins are capable of self-assembling into structures that are toxic to the cell, and a number of neurodegenerative diseases, such as Alzheimer’s disease and Parkinson’s disease, are associated with this self-assembly process. In this talk, I will discuss the effect of surfaces and the osmolytes urea and TMAO in regulating the structure and assembly of intrinsically disordered peptides. I will focus on two model systems, the mussel foot protein implicated in underwater adhesion of mussels to rocks, and the Tau peptide implicated in Alzheimer’s Disease.
Dynamic nature of heterogeneous catalysts The determination of the structure of heterogeneous catalytic systems is a key point for a detailed understanding of the nature of active sites and for the rational design of efficient catalysts. However catalysts are not static but dynamic, fluxional, metatstable and they strongly evolve under reaction conditions, creating new active sites, not present for the as prepared catalysts. The lecture will discuss two systems. The first one will focus on the modelling of small Pt cluster (Pt13) under hydrogen pressure and on their reactivity for methane activation.1 The approach combines Density Functional Theory, high-dimensional Neural Networks and evolutionary techniques. The bare Pt clusters shows a large number of low energy isomers (49 in 0.5 eV). Hydrogenated clusters adopt different geometries and appear more rigid, with a smaller number of low energy isomers. These metastable isomers nevertheless play a major role for the catalytic reactivity of the hydrogen covered cluster, which cannot be described by considering the most stable structure alone. Fluxionality and accessible metastable structure are hence key characteristics for the catalytic properties of small Pt clusters. The second part will deal with single-atom catalysts, that are widely investigated heterogeneous catalysts. We will consider Rh single atoms on TiO2 as a generic example, investigating the optimal structure of the single atoms under H2 reduction, CO adsorption and its reactivity for reverse water gas shift (RWGS) reaction. The combination of theoretical and experimental studies clearly demonstrates that Rh single atoms change their structure and adapt their catalytic site under reaction conditions.

Modeling of protein-DNA binding with a multi-module deep learning framework

Transcription factor (TF)-DNA binding is a fundamental component of gene regulatory processes, but how these proteins recognize their target sites in the genome is still not completely understood. TFs can recognize their binding sites by having a surface that is physicochemically complementary to the physicochemical signature of DNA, forming a series of contacts between the protein and the base pairs. These contacts include direct hydrogen bonds, water-mediated hydrogen bonds, and hydrophobic contacts. X-ray crystallographic structures of TF-DNA complexes provide essential insights into TF–DNA binding mechanisms. However, experimentally determined structures are available for relatively few TFs and are typically limited to complexes where a protein or its DNA binding domain binds to a single DNA sequence. In the last decade, several high-throughput technologies have been developed for a better understanding of TF-DNA binding mechanisms by quantitatively measuring the binding affinities of a TF against thousands or even millions of different DNA sequences in vitro. These methods provide an alternative path to infer TF-DNA binding mechanisms without requiring time-consuming structural biology experiments. Here we present DeepRec (Deep Recognition for TF-DNA binding), a multi-module deep learning framework capable of building precise predictive models for TF-DNA binding based on large-scale in vitro experimental data. The method also integrates a forward perturbation-based interpretation approach to highlight the important patterns for binding. The approach enabled accurate predictions of DNA binding specificities and unraveled important structural insights into readout mechanisms.

Condensed phase quantum chemistry