Seminars

Host: TH Department

Learning Chemistry from a Computer: Recent Applications of Automatic Mechanism Generation for the Microkinetics of CO2 Methanation on Nickel

Microkinetic mechanisms in heterogeneous catalysis can be incredibly complex, and the development of these mechanisms raises many problems. How do you determine which species and reactions to include? How do you obtain the corresponding parameters? [more]

Fully Quantum (Bio)Molecular Simulations: Dream or Reality?

The convergence between accurate quantum-mechanical (QM) models (and codes) with efficient machine learning (ML) methods seem to promise a paradigm shift in molecular simulations. Many challenging applications are now being tackled by increasingly powerful QM/ML methodologies. These include modeling covalent materials, molecules, molecular crystals, surfaces, and even whole proteins in explicit water (https://arxiv.org/abs/2205.08306). [more]

DFTB+, the Fast Way of Quantum Mechanical Simulations

The Density Functional Tight Binding (DFTB) method [1] is an approximate Density Functional Theory (DFT) based framework, which allows for quantum mechanicalsimulations of large systems being typically two or three orders of magnitudefaster than comparable ab initio DFT calculations. [more]

The Value of Information: From Statistics to Algorithms

A fundamental question in data science is: how much information can one extract from the data that one collects? [more]

Quantum-Chemical Methods for Large Systems: Low-, Linear-, and Sublinear-Scaling Methods

An overview of our recently developed low-, linear-, and sublinear-scaling methods ranging from HF, DFT, MP2 to RPA is given. These methods allow — also in combination with graphics processing units (GPUs) — for the efficient description of large systems at QM and QM/MM levels, where QM spheres with typically 500-1000 atoms are necessary for reliable studies. [more]

New Concepts in Battery and Solid Electrolyte Design

Solid electrolytes (SEs) are a key component of all-solid-state batteries (ASSBs), which promisehigher energy density along with safer operation compared to commercial Li ion batteries. As theASSB technology matures, research in the field gravitates towards questions regarding stability,scalability, and integration of solid electrolytes into ASSBs with extended cycle life. [more]

Design of Novel Hybrid and Solid State Battery Materials

Next generation of energy storage devices may largely benefit from fast and solid Li+ ceramic electrolyte conductors to allow for safe and efficient batteries and fast data calculation. For those applications, the ability of Li-oxides to be processed as thin film structures and with high control over Lithiation and phases at low temperature is of essence to control conductivity. [more]

Solving Electrochemistry Puzzles by First-Principles Multi-Scale Modeling

Electrochemistry has become the most-promising prospect towards a sustainable energy landscape. Still, most processes have not been optimized to a level that could compete with fossil fuels. Consequently, substantial optimization of electrode materials, electrolytes and electrochemical cells is required on all size scales to quickly reach the industrially desired performance. [more]

The End of Ab Initio MD

A new computational task has been defined and solved over the past 15 years for extended material systems: the analytic fitting of the Born-Oppenheimer potential energy surface as a function of nuclear coordinates under the assumption of medium-range interactions, 5 ~ 10 Å. [more]

Machine Learning (ML) for Simulating Complex Energy Materials with Non-Crystalline Structures

Many materials with applications in energy materials, e.g., catalysis or batteries are non-crystalline with amorphous structures, chemical disorder, and complex compositions, which makes the direct modelling with first principles methods challenging. To address this challenge, we developed accelerated sampling strategies based on ML potentials, genetic algorithms, and molecular-dynamics simulations. [more]

Hot Electrons in Surface Chemistry: From Molecular Scattering to Plasmonic Chemistry

Nonadiabatic effects that arise from the concerted motion of electrons and atoms at comparable energy and time scales are omnipresent in thermal and light-driven chemistry at metal surfaces. [more]

Machine Learning at the Atomic Scale: From Structural Representations to Chemical Insights

When modeling materials and molecules at the atomic scale, achieving a realistic level of complexity and making quantitative predictions are usually conflicting goals. Data-driven techniques have made great strides towards enabling simulations of materials in realistic conditions with uncompromising accuracy. [more]
The interaction of adsorbates on solid surfaces with light is central to surface spectroscopy, surface photochemistry, and non-adiabatic surface science in general. In the present contribution, light-driven molecular adsorbates will be modelled (mostly) by ab initio molecular dynamics. A few examples will be highlighted: [more]

Machine-Learning Potentials: The Accurate, the Fast, and the Applied

Data-driven algorithms ("machine learning") are increasingly used in science and engineering for analysis, prediction, and control, enabling new insights and applications. A promising example are first-principles simulations of the dynamics of atomistic systems. [more]

In Silico Design of Single-Atom- and Highly-Dilute-Alloy Catalysts: Success Stories and Opportunities for Innovation

Catalysis is undoubtedly at the heart of the chemical industry: out of all chemicals manufacturing processes, 85-90% are catalytic and about 80-85% thereof employ heterogeneous catalysts. Yet, developing catalysts for given applications is non-trivial, necessitating empirical and resource-intensive trial-and-error experimentation. Theory and simulation, on the other hand, can provide fundamental insight into the mechanisms underpinning catalytic function, and guide the design of catalytic materials for applications of practical interest. [more]

Data-Enabled Materials Structure-Property-Synthesizability Predictions

The constant demand for new functional energy materials calls for efficient strategies to accelerate the materials discovery. In addressing this challenge, materials informatics deals with the use of data, computations, and machine learning (complementary to experts’ intuitions) to establish the materials structure-property relationships and to make a new functional discovery in a rate that is significantly accelerated. [more]

Computational Understanding of Electrochemical Energy Storage Materials

The complete electrification of the transport sector will require batteries that can be made from abundant chemical species and exhibit significantly greater energy density than current Li-ion batteries. [more]

Blending Old Concepts with Data-driven Approaches to Discover and Classify Homogeneous Catalysts

Sabatier’s principle,[1] developed in the first decades of the 20th century, states that an ideal catalyst should bind a substrate neither too strongly nor too weakly. Today, this simple idea provides the fundamental underpinning for “volcano plots”,[2,3] which are abundantly used in heterogeneous and electrocatalysis.[4] [more]

Theoretical Perspectives on Proton-Coupled Electron Transfer

Proton-coupled electron transfer (PCET) reactions play a vital role in a wide range of chemical and biological processes. This talk will focus on the theory of PCET as well as illustrative applications to catalysis and energy conversion processes. [more]

Accurate Description of Correlated Physics of Bulk Materials using Diagrammatic Methods and Quantum Embedding

This talk will give an overview of continuing work in the Chan group to describe correlated electron materials with high-level quantum chemistry methods. [more]

Towards Predicting the Charge of Electrochemical Interfaces

The charge of an electrochemical interface helps to determine its chemical reactivity and macroscopic electrostatic properties. However, predicting the charge at an electrochemical interface is challenging because it generally requires both the electronic structure of the interface, and the thermodynamically averaged spatial distribution of the electrolyte. Due to this difficulty, interfacial charge prediction remains an open problem in computational electrochemistry. [more]
Go to Editor View