After last summer semester’s successful debut, the “Seminar Afternoon on Atomistic Modeling” returns this winter semester!

Harald Oberhofer is Professor of Theoretical Physics at the Faculty of Mathematics, Physics and Computer Science at the University of Bayreuth. His research combines first‑principles simulations with data‑driven analysis to uncover the microscopic mechanisms that govern charge‑carrier behavior in a wide range of materials. By integrating statistical methods, data‑mining, and machine learning, they extract general design principles that support and guide experimental materials development.

Computational science has evolved from a primarily explanatory tool into a proactive driver of discovery, enabled by modern data‑science methods and explainable machine learning. In his presentation on “Computing and Predicting Properties of Energy Materials”, Oberhofer will highlight new quantitative modeling approaches and machine‑learning techniques that reveal structure–function relationships in materials through a bottom‑up, multiscale framework.

Dr. Íñigo Iribarren Aguirre is postdoctoral researcher in the group of Prof. Alessio Gagliardi. 
In his talk on “A Computational Workflow to Unravel the Structural Dynamics of Supramolecular Metallacages in Solution”, he will introduce a computational framework to analyze the structural dynamics of self‑assembled [Pd₂L₄]⁴⁺ metallacages in solution, benchmarking classical force‑field and machine‑learning potentials to compare accuracy and cost. This workflow enables reliable, scalable simulations that connect quantum‑level detail with the time and length scales needed to investigate host–guest behavior relevant to drug‑delivery applications.

Date: Friday, January 30, 2026

Agenda:

Location: TUM Entrepreneurship Research Institute (Lichtenbergstr. 6, Garching)

Prof. Harald Oberhofer:

Abstract:
With the advent of ever more powerful computers and more accurate, yet efficient algorithms, computational science has by now been widely accepted as a valuable and equal contribution to both pure theory and experiment. Traditionally, computation thereby played the roles of elucidating microscopic properties and mechanisms of given systems and reaction pathways, leading to numerous breakthroughs not otherwise possible. The choice of system or reaction, thereby was mostly guided by experiment or the intuition and experience of the researcher. Recently, though, modern data science approaches such as data-mining and explainable machine learning allowed computation to take on a role of proactive exploration and design, supplementing the traditional roles of computational materials science.

In my presentation, I will outline some of our research regarding method development and application in this field. First, I will present our recent extension of the famous Newns Anderson model, converting it from a qualitative tool aimed at gaining a rough understanding of charge transfer problems, to a fully convergent and quantitative method that can be used to predict experimental results.

Second, I will show how the use of explainable machine learning models not only allows us to predict properties of materials, but also to extract the materials' characteristics leading to these properties. Thus, we show how such an approach can yield useful structure-function relationships. Thereby, all my work is based on a bottom up approach, where specifically developed methods on the electronic structure level, aided by machine learned datapoint selection and machine learned or classical force-fields inform larger scale effective, embedded or kinetic models, which in turn yield the descriptors necessary for data-driven and machine learning based analyses of more general problems.

Dr. Íñigo Iribarren Aguirre:

Abstract:

The structural dynamic of self-assembled metallacages is important because it determines their function and stability in different applications involving encapsulation and release of a guest molecule. We present here an integrated computational (1) to study the dynamic behavior of selected [Pd₂L₄]⁴⁺ metallacages (2) in explicit solvents (water and DMSO) and benchmark them for future in silico investigation of their host-guest chemistry, pivotal to their application as drug delivery systems. Two different pathways for the Molecular Dynamics (MD) simulations of the systems are explored, namely classical force field (FF) and Machine Learning Interatomic Potential (MLIP), to assess the conformational changes of two cage systems in solution, enabling evaluation of the performance vs computational cost for both methodologies. The proposed workflow offers a versatile framework to computationally assess the structural dynamics of supramolecular systems in solution, effectively bridging the gap between quantum-level accuracy and the temporal and spatial scales needed for simulations of different functional applications like studying the host-guest interactions with different compounds (3). 

(1) Stebani, J. A.; Iribarren Aguirre, I.; Siddiqui, G. A.; Wragg, D.; Gagliardi, A.; Casini, A. Computational Workflow to Unravel the Structural Dynamics of Supramolecular Metallacages in Solution. J. Chem. Theory Comput. 2025, 21 (23), 12278–12288. doi.org/10.1021/acs.jctc.5c01465. 
(2) Casini, A.; Woods, B.; Wenzel, M. The Promise of Self-Assembled 3D Supramolecular Coordination Complexes for Biomedical Applications. Inorg. Chem. 2017, 56 (24), 14715–14729. doi.org/10.1021/acs.inorgchem.7b02599. 
(3) Stebani, J. A.; Iribarren, I.; Wragg, D.; Gagliardi, A.; Casini, A. A Dynamic Evaluation of Cisplatin Encapsulation into [Pd₂L₄]⁴⁺ Metallacages in Solution. ChemRxiv December 14, 2025. doi.org/10.26434/chemrxiv-2025-7m9mj.