A updated and complete list of publications can be found here: Google Scholar
Abstract The calculation of heat transfer in fluid flow in simple flat channels is a relatively easy task for various simulations methods. However, once the channel geometry becomes more complex, numerical simulations become a bottleneck in optimizing wall geometries. We present a combination of accurate numerical simulations of arbitrary, non-flat channels and machine learning models predicting drag coefficient and Stanton number. We show that convolutional neural networks can accurately predict the target properties at a fraction of the time of numerical simulations. We use the CNN models in a virtual high-throughput screening approach to explore a large number of possible, randomly generated wall architectures. We find that S-shaped channel geometries are Pareto-optimal, a result which seems intuitive, but was not obvious before analysing the data. The general approach is not only applicable to simple flow setups as presented here, but can be extended to more complex tasks, such as multiphase or even reactive unit operations in chemical engineering.
Abstract The automation of scientific experiments with artificial intelligence has the potential to significantly
accelerate the development and optimization of novel materials with tailored properties .
In a recently published article  in ACS Nano, the authors show that they can use neural networks to draw conclusions about the conductivity of polymer mixtures from comparatively easy to measure optical properties. The conductivity of polymers is more interesting, but also much more difficult to measure than optical properties. This principle can also be transferred to other materials and properties. The use of artificial intelligence methods therefore has the potential to optimize the numerous parameters available for the development of materials in faster and more cost-effective way by replacing complex measurement methods for determining material properties with rapid, e.g. optical experiments, which then use machine learning to predict the actual target property.
Abstract CatalChemical Science titlepageysts are not only used to clean exhaust gases, but are also indispensable in the production of molecules and materials. The design of ever newer molecules and materials would be unthinkable without the simultaneous development of newer and better catalysts. In addition, novel catalysts also play a decisive role in recovering CO2 from the atmosphere and thus in combating climate change. A recent study  published in Chemical Science demonstrates how machine learning can be coupled with highly accurate but expensive simulation methods to predict the efficiency of new catalysts by computational methods and thus significantly accelerate their design. Pascal Friederich and his co-authors show in the article that machine learning is not a black box but can be interpreted to derive design rules for new catalysts that are intelligible to scientists. The proposed method for the virtual design of new iridium catalysts can be transferred as desired to further classes of reactions of homogeneous (and in principle also heterogeneous) catalysis.
My paper about machine learning of dihedral force fields was published in Scientific Reports.
Abstract Computer simulation increasingly complements experimental efforts to describe nanoscale structure formation. Molecular mechanics simulations and related computational methods fundamentally rely on the accuracy of classical atomistic force fields for the evaluation of inter- and intramolecular energies. One indispensable component of such force fields, in particular for large organic molecules, is the accuracy of molecule-specific dihedral potentials which are the key determinants of molecular flexibility. We show in this work that non-local correlations of dihedral potentials play a decisive role in the description of the total molecular energy - an effect which is neglected in most state-of-the-art dihedral force fields. We furthermore present an efficient machine learning approach to compute intramolecular conformational energies. We demonstrate with the example of alphaNPD, a molecule frequently used in organic electronics, that this approach outperforms traditional force fields by decreasing the mean absolute deviations by one order of magnitude to values smaller than 0.37 kcal/mol (16.0 meV) per dihedral angle.
Our paper about the molecular origin of the dye orientation in vapor deposited OLEDs got published in Chemistry of Materials.
Abstract Molecular orientation anisotropy of the emitter molecules used in organic light emitting diodes (OLEDs) can give rise to an enhanced light-outcoupling efficiency, when their transition dipole moments are oriented preferentially parallel to the substrate, and to a modified internal quantum efficiency, when their static dipole moments give rise to a locally modified internal electric field. Here, the orientation anisotropy of state-of-the-art phosphorescent dye molecules is investigated using a simulation approach which mimics the physical vapor deposition process of amorphous thin films. The simulations reveal for all studied systems significant orientation anisotropy. Various types are found, including a preference of the static dipole moments to a certain direction or axis. However, only few systems show an improved outcoupling efficiency. The outcoupling efficiency predicted by the simulations agrees with experimentally reported values. The simulations reveal in some cases a significant effect of the host molecules, and suggest that the driving force of molecular orientation lies in the molecule-specific van der Waals interactions of the dye molecule within the thin film surface. The electrostatic dipole-dipole interaction slightly reduces the anisotropy. These findings can be used for the future design of improved dye molecules.
Our paper about rational in silico design of an organic semiconductor with improved electron mobility got published in Advanced Materials.
Abstract Organic semiconductors find a wide range of applications, such as in organic light emitting diodes, organic solar cells, and organic field effect transistors. One of their most striking disadvantages in comparison to crystalline inorganic semiconductors is their low charge-carrier mobility, which manifests itself in major device constraints such as limited photoactive layer thicknesses. Trial-and-error attempts to increase charge-carrier mobility are impeded by the complex interplay of the molecular and electronic structure of the material with its morphology. Here, the viability of a multiscale simulation approach to rationally design materials with improved electron mobility is demonstrated. Starting from one of the most widely used electron conducting materials (Alq3), novel organic semiconductors with tailored electronic properties are designed for which an improvement of the electron mobility by three orders of magnitude is predicted and experimentally confirmed.