Seminar by Daria Galimberti: 'DFT-MD & Vibrational Spectroscopy for inhomogeneous materials: state-of-the-art and future developments with Graph Theory' (Lecture)
- Monday 10 February 2020Add to my calendar
- from 09:15
Daria Galimberti - Institut fur Chemie, Humboldt-Universitat zue Berlin (Germany)
In the last few years, quantum chemical calculations have shown high accuracy for the prediction of the structural and spectroscopic properties of molecular systems, allowing for the characterization of inhomogeneous materials, such as flexible organic molecules, clusters, water solutions, and interfaces. In particular, the use of DFT-MD simulations to compute IR, Raman, or SFG spectra allow for combining the accuracy of DFT (being able to describe reactive surfaces) and the broad sampling of free energy surfaces by MD.
I will show examples of my works on how advanced such state-of-the-art computational techniques have been successfully applied for the characterization of the structural and vibrational properties of gas phase molecules and interfaces.
Our newly developed theoretical method for the assignment of (anharmonic) vibrational modes from MD simulations, based on Graph Theory coupled to Atomic Polar Tensors (APT) weighted-internal coordinates velocities DOS spectra, will also be presented. As an example of the possibilities offered by this new road, I will show the results obtained in collaboration with Dr A.M. Rijs (FELIX laboratory at Radboud University) for the 3D structures of (Ac-Phe-OMe)1,2 peptides, especially using the far-IR/THz signatures (100-800 cm-1).
I will then discuss the future possibilities to devise a protocol that combines experimental vibrational spectroscopy with computational chemistry and chemo-informatics to predict 3D images/structures of molecular assemblies, also as they evolve/react in time. This will be done by combining Game Theory and Deep Learning, starting from the graph representation of the spectra.
Prof. Herma Cuppen