Projected q-shells retained more smooth-function information and produced lower one-electron and molecular benchmark errors than standard nested gausslets in the reported calculations.
A computational study of carbon dioxide reacting with surface hydrogen on Cu(111) predicts that rotational excitation may influence reaction probability more strongly than bending vibration, while leaving the mechanism to be tested experimentally.
A model-free graph analysis reconstructed 158 HDO energy levels from 686 initially unassigned transitions. Broadband tests in acetylene and methane also measured line-position repeatability and identified unusual spectral features.
OrbGNN represents molecular orbitals as graph nodes and uses their connections to predict correlation energies and spin-state gaps in three computational tests.
The method gave finite model clusters an open-boundary response, but the tests were limited to idealized calculations and the authors say it is not yet a complete embedding framework.
A model tracked published lithium and cobalt leaching data reasonably well in lower-acid conditions, but its peroxide predictions ran high at the strongest acid level.
A benchmark of molecular ionization potentials found that Dyson and non-Dyson ADC(3) calculations differed by about 0.1 electronvolt on average, while a screened method produced a narrower spread of errors.
A laboratory study finds that sulfur core-valence spectra resemble valence spectra after resolution differences are considered, while fluorine has extra high-energy features.
A preprint reports that isolated D2O forms D2+ through an indirect multistage pathway, while simulations identify direct, roaming and delayed dissociation branches.
Energy transfer in a cross-linked APC complex was fastest around 30 to 50 kelvin, while simulations linked the turnover to temperature-dependent low-frequency environmental dynamics.
A preprint on one rod-shaped gold nanocluster reports that two vibrational motions appeared in sequence, while catalytic turnover was not directly measured.
A laboratory study used potential-modulated iSCAT to follow voltage-synchronized optical signals from nanoholes and map electrical connectivity across patterned ITO.
Calculations of carbon wires between aluminum electrodes found that orbital treatment altered predicted conductance patterns, but did not reproduce every cited experimental trend.
A computational preprint reports a subsystem method that treats selected protons quantum mechanically, matching its NEO-DFT reference closely at some water-dimer displacements while reaching larger water-cluster systems.
A computational chemistry preprint presents a time-dependent method for simulating resonance Raman spectra. It matched numerically exact calculations in a harmonic model and agreed well with an experimental anthracene spectrum, but the approach remains tied to harmonic assumptions and missed one peak associated with Herzberg–Teller coupling.
An arXiv preprint reports a computational method that reuses one excited-state trajectory to calculate molecular vibronic spectra at different temperatures. It captured hot bands, broadening and peak structure in four molecular examples, but the authors state that the approach is not applicable when tunnelling or wavepacket splitting dominates the spectral shape.
A modeling preprint develops a reduced-order way to connect particle swelling, electrode stress and whole-cell constraints to lithium-ion battery electrochemistry. In deterministic simulations, the full multiscale model’s voltage contribution reaches approximately 15 mV at full charge, while a clamped scenario reaches about 16 mV.
A time-dependent unitary coupled-cluster formulation tracked exact reference calculations for several observables in a six-site model. Smaller excitation spaces introduced truncation error, while the treatment of dipole-dipole correlations and realistic systems remains unresolved.
A modeling study links symmetry-selected molecular vibrations with different proton-transfer products. In high-temperature simulations, p-DAPA-CN transferred faster than p-DAPA-CF3, but the result remains computational.
The study presents a step-by-step MRM workflow and shows that standardization choice can alter both statistical results and biological interpretation in a piglet-liver validation dataset.
Journal of biomolecular techniques : JBT4 min read