Search

Energy redistribution after a chemical reaction is one of the few mechanisms to explain the diffusion and desorption of molecules which require more energy than the thermal energy available in quiescent molecular clouds (10 K). This energy distribution can be important in phosphorous hydrides, elusive yet fundamental molecules for interstellar prebiotic chemistry. We studied the reaction dynamics of the \ce{P + H -> PH} reaction on amorphous solid water, a reaction of astrophysical interest, using \emph{ab-initio} molecular dynamics with atomic forces evaluated by a neural network interatomic potential. We found that the exact nature of the initial phosphorous binding sites is less relevant for the energy dissipation process because the nascent PH molecule rapidly migrates to sites with higher binding energy after the reaction. Non-thermal diffusion and desorption-after-reaction were observed and occurred early in the dynamics, essentially decoupled from the dissipation of the chemical reaction energy. From an extensive sampling of reactions on sites, we constrained the average dissipated reaction energy within the simulation time (50 ps) to be between 50 and 70 %. Most importantly, the fraction of translational energy acquired by the formed molecule was found to be mostly between 1 and 5 %. Including these values, specifically for the test cases of 2% and 5% of translational energy conversion, in astrochemical models, reveals very low gas-phase abundances of PH$_{x}$ molecules and reflects that considering binding energy distributions is paramount for correctly merging microscopic and macroscopic modelling of non-thermal surface astrochemical processes. Finally, we found that PD molecules dissipate more of the reaction energy. This effect can be relevant for the deuterium fractionation and preferential distillation of molecules in the interstellar medium., Accepted for publication in Astronomy and Astrophysics

The acquisition of labels for supervised learning can be expensive. In order to improve the sample-efficiency of neural network regression, we study active learning methods that adaptively select batches of unlabeled data for labeling. We present a framework for constructing such methods out of (network-dependent) base kernels, kernel transformations and selection methods. Our framework encompasses many existing Bayesian methods based on Gaussian Process approximations of neural networks as well as non-Bayesian methods. Additionally, we propose to replace the commonly used last-layer features with sketched finite-width Neural Tangent Kernels, and to combine them with a novel clustering method. To evaluate different methods, we introduce an open-source benchmark consisting of 15 large tabular regression data sets. Our proposed method outperforms the state-of-the-art on our benchmark, scales to large data sets, and works out-of-the-box without adjusting the network architecture or training code. We provide open-source code that includes efficient implementations of all kernels, kernel transformations, and selection methods, and can be used for reproducing our results., Comment: Changes in v3: Improvements in writing and other minor changes. Accompanying code can be found at https://github.com/dholzmueller/bmdal_reg

The isomerism of molecules in the interstellar medium and the mechanisms behind it are essential questions in the chemistry of organic molecules in space. In particular, for the simple formic and thioformic acids, the low temperatures found in molecular clouds indicate that cis-trans isomerization in the gas-phase must be impeded. Reactions happening on top of interstellar dust grains may explain the isomer interconversion at low temperatures. We studied the isomerization processes of formic and thioformic acid susceptible to happen on the surface of interstellar dust grains and initiated by H abstraction reactions. Similarly, deuterium enrichment of the acids can occur by the same mechanism. Our objective is to shed light on both topics to increase our understanding of key precursors of organic molecules in space., Comment: Accepted for publication in Astronomy and Astrophysics

A trivial flaw in the utilization of artificial neural networks in interpolating chemical potential energy surfaces (PES) whose descriptors are Cartesian coordinates is their dependence on simple translations and rotations of the molecule under consideration. A different set of descriptors can be chosen to circumvent this problem, internuclear distances, inverse internuclear distances or z-matrix coordinates are three such descriptors. The objective is to use an interpolated PES in instanton rate constant calculations, hence information on the energy, gradient and Hessian is required at coordinates in the vicinity of the tunneling path. Instanton theory relies on smoothly fitted Hessians, therefore we use energy, gradients and Hessians in the training procedure. A major challenge is presented in the proper back-transformation of the output gradients and Hessians from internal coordinates to Cartesian coordinates. We perform comparisons between our method, a previous approach and on-the-fly rate constant calcuations on the hydrogen abstraction from methanol and on the hydrogen addition to isocyanic acid.

We calculated reaction rate constants including atom tunneling of the reaction of dihydrogen with the hydroxy radical down to a temperature of 50 K. Instanton theory and canonical variational theory with microcanonical optimized multidimensional tunneling (CVT/$\mu$OMT) were applied using a fitted potential energy surface [J. Chem. Phys. 138, 154301 (2013)]. All possible protium/deuterium isotopologues were considered. Atom tunneling increases at about 250 K (200 K for deuterium transfer). Even at 50 K the rate constants of all isotopologues remain in the interval $ 4 \cdot 10^{-20}$ to $4 \cdot 10^{-17}$ cm$^3$ s$^{-1}$ , demonstrating that even deuterated versions of the title reaction are possibly relevant to astrochemical processes in molecular clouds. The transferred hydrogen atom dominates the kinetic isotope effect at all temperatures., Comment: title page, 16 pages, followed by 6 pages of SI

The scope of this thesis is the application of quantum machine learning (QML) methods to problems in quantum chemistry and chemical compound space, especially chemical reactions. First, QML models were introduced to improve job scheduling of quantum chemistry tasks on small university and large super computing clusters. Using QML based wall time predictions to optimally distribute the workload on a cluster resulted in a significant reduction of the time to solution by up to 90% depending on the type of calculation studied: Ranging from single point calculations, over geometry optimizations, to transition state searches on a variety of levels of theory and basis sets. The main focus of this thesis remains with the navigation through the chemical reaction space using QML models. To train and test these models large, consistent, and carefully evaluated data sets are required. While extensive data sets with experimental results are available, consistent quantum chemical data sets, especially for reactions, are rare in literature. Thus, a dataset for two competing text book reactions E2 and SN2 was generated, reporting thousands of reactant complexes and transition states with different nucleophiles (-H$^{-}$,-F$^{-}$, -Cl$^{-}$, -Br$^{-}$), leaving groups (-F, -Cl, -Br), and functional groups (-H, -NO$_2$, -CN, -CH$_3$, -NH$_2$) on an ethane scaffold. The geometries were obtained on the MP2/6-311G(d) level of theory with subsequent DF-LCCSD/cc-pVTZ single point calculation. However, limited by computational resources, the data set was incomplete. Therefore, reactant to barrier (R2B) machine learning models were introduced to support the data generation and complete the dateset by predicting ~11'000 activation barriers solely using the reactant geometry as input. Using R2B predictions, design rules for chemical reaction channels were derived by constructing decision trees. Furthermore, Hammond's postulate was investigated, showing the limits for its application on reactants far away from the transition state, e.g. conformers. Finally, the geometry relaxation and transition state search solely using machine learned energies and forces was investigated. Trained on 200 reactions, the QML model was able to find 300 transition states, reaching out of sample RMSD of 0.14Å and 0.4Å for reactant geometries and transition states, respectively. Although, relatively large RMSD for the geometries remain, the out of sample MAE of 26.06$\mathrm{cm^{-1}}$ for the transition state frequencies show a well described curvature of the transition state normal modes in agreement with the MP2 reference.