arXiv subjects

Carolin Müller

Publications and source records attributed to Carolin Müller.

3 recordsLinked to original sources

arXiv2024

Thermomechanical properties of zero thermal expansion materials from theory and experiments

Origin and composition dependence of the anisotropic thermomechanical properties are elucidated for Ba1-xSrxZn2Si2O7 (BZS) solid solutions. The high-temperature phase of BZS shows negative thermal expansion (NTE) along one crystallographic axis and highly anisotropic elastic properties characterized by X-ray diffraction experiments and simulations at the density functional theory level. Ab initio molecular dynamics simulations provide accurate predictions of the anisotropic thermal expansion in excellent agreement with experimental observations. The NTE considerably decreases with increasing Sr content x. This is connected with the composition dependence of the vibrational density of states (VDOS) and the anisotropic Grüneisen parameters. The VDOS shifts to higher frequencies between 0-5 THz due to substitution of Ba with Sr. In the same frequency range, vibrational modes contributing most to the NTE are found. In addition, phonon calculations using the quasi-harmonic approximation revealed that the NTE is mainly connected with deformation of four-membered rings formed by SiO4 and ZnO4 tetrahedra. The thermomechanical and vibrational properties obtained in this work provide the basis for future studies facilitating the targeted design of BZS solid solutions as zero or negative thermal expansion material.

cond-mat.mtrl-sci↗

arXiv2023

Analysis of a Deep Learning Model for 12-Lead ECG Classification Reveals Learned Features Similar to Diagnostic Criteria

Despite their remarkable performance, deep neural networks remain unadopted in clinical practice, which is considered to be partially due to their lack in explainability. In this work, we apply attribution methods to a pre-trained deep neural network (DNN) for 12-lead electrocardiography classification to open this "black box" and understand the relationship between model prediction and learned features. We classify data from a public data set and the attribution methods assign a "relevance score" to each sample of the classified signals. This allows analyzing what the network learned during training, for which we propose quantitative methods: average relevance scores over a) classes, b) leads, and c) average beats. The analyses of relevance scores for atrial fibrillation (AF) and left bundle branch block (LBBB) compared to healthy controls show that their mean values a) increase with higher classification probability and correspond to false classifications when around zero, and b) correspond to clinical recommendations regarding which lead to consider. Furthermore, c) visible P-waves and concordant T-waves result in clearly negative relevance scores in AF and LBBB classification, respectively. In summary, our analysis suggests that the DNN learned features similar to cardiology textbook knowledge.

eess.SP↗

arXiv2019

Imaging the Renner-Teller effect using laser-induced electron diffraction

Structural information on electronically excited neutral molecules can be indirectly retrieved, largely through pump-probe and rotational spectroscopy measurements with the aid of calculations. Here, we demonstrate the direct structural retrieval of neutral carbonyl disulfide (CS$_2$) in the B$^1$B$_2$ excited electronic state using laser-induced electron diffraction (LIED). We unambiguously identify the ultrafast symmetric stretching and bending of the field-dressed neutral CS$_2$ molecule with combined picometer and attosecond resolution using intrapulse pump-probe excitation and measurement. We invoke the Renner-Teller effect to populate the B$^1$B$_2$ excited state in neutral CS$_2$, leading to bending and stretching of the molecule. Our results demonstrate the sensitivity of LIED in retrieving the geometric structure of CS$_2$, which is known to appear as a two-center scatterer.

physics.chem-ph↗