Raman Spectra of Amino Acids and Peptides from Machine Learning Polarizabilities.

Raman spectroscopy is an important tool in the study of vibrational properties and composition of molecules, peptides, and even proteins. Raman spectra can be simulated based on the change of the electronic polarizability with vibrations, which can nowadays be efficiently obtained via machine learning models trained on first-principles data. However, the transferability of the models trained on small molecules to larger structures is unclear, and direct training on large structures is prohibitively expensive. In this work, we first train two machine learning models to predict the polarizabilities of all 20 amino acids. Both models are carefully benchmarked and compared to density functional theory (DFT) calculations, with the neural network method being found to offer better transferability. By combination of machine learning models with classical force field molecular dynamics, Raman spectra of all amino acids are also obtained and investigated, showing good agreement with experiments. The models are further extended to small peptides. We find that adding structures containing peptide bonds to the training set greatly improves predictions, even for peptides not included in training sets.

Identification of Material Dimensionality Based on Force Constant Analysis.

Identification of low-dimensional structural units from the bulk atomic structure is a widely used approach for discovering new low-dimensional materials with new properties and applications. Such analysis is usually based solely on bond-length heuristics, whereas an analysis based on bond strengths would be physically more justified. Here, we study dimensionality classification based on the interatomic force constants of a structure with different approaches for selecting the bonded atoms. The implemented approaches are applied to the existing database of first-principles calculated force constants with a large variety of materials, and the results are analyzed by comparing them to those of several bond-length-based classification methods. Depending on the approach, they can either reproduce results from bond-length-based methods or provide complementary information. As an example of the latter, we managed to identify new non-van der Waals two-dimensional material candidates.

Highly Selective H2S Gas Sensor Based on Ti3C2Tx MXene-Organic Composites.

Cost-effective and high-performance H2S sensors are required for human health and environmental monitoring. 2D transition-metal carbides and nitrides (MXenes) are appealing candidates for gas sensing due to good conductivity and abundant surface functional groups but have been studied primarily for detecting NH3 and VOCs, with generally positive responses that are not highly selective to the target gases. Here, we report on a negative response of pristine Ti3C2Tx thin films for H2S gas sensing (in contrast to the other tested gases) and further optimization of the sensor performance using a composite of Ti3C2Tx flakes and conjugated polymers (poly[3,6-diamino-10-methylacridinium chloride-co-3,6-diaminoacridine-squaraine], PDS-Cl) with polar charged nitrogen. The composite, preserving the high selectivity of pristine Ti3C2Tx, exhibits an H2S sensing response of 2% at 5 ppm (a thirtyfold sensing enhancement) and a low limit of detection of 500 ppb. In addition, our density functional theory calculations indicate that the mixture of MXene surface functional groups needs to be taken into account to describe the sensing mechanism and the selectivity of the sensor in agreement with the experimental results. Thus, this report extends the application range of MXene-based composites to H2S sensors and deepens the understanding of their gas sensing mechanisms.

Low-Frequency Dielectric Response of Tetragonal Perovskite CH3NH3PbI3.

The dielectric properties of tetragonal hybrid perovskite CH3NH3PbI3 are studied through molecular dynamics at a temperature of 300 K in the presence of a finite electric field. The high-frequency dielectric constant ε∞ is found to be 4.5 along the a axis and 4.7 along the c axis. The values of the respective static dielectric constants ε0 are 43 ± 1 and 53 ± 3, much larger than the value of ∼25 pertaining to the orthorhombic phase, in which the organic cations cannot rotate. At frequencies below 3 cm-1, we observe a significant increase in ε0 by ∼23 (a axis) and ∼30 (c axis) compared to a vibrational approach, which does not account for the reorientation of the molecular units. The decomposition shows that the reorientation of the organic cations accounts for an increase of only ∼10. An increase of similar size results from the displacement of the cations within the cages of the lattice. The dominant contribution is found to arise from lattice vibrations coupled to the motion of the organic cations.