Structure-based sampling and self-correcting machine learning is used for precise representation of molecular potential energy surfaces and calculating vibrational levels with spectroscopic accuracy (errors less than 1 cm−1 relative to the reference ab initio spectrum) decreasing the number of required quantum mechanical calculations by up to 90%.
(more…)
A series of the substituted two-electron acceptors with a dicyanomethylene-bridged acridophosphine scaffold has been prepared and compared with the nitrogen-containing counterpart using various spectroscopic, electrochemical and theoretical methods.
(more…)
A stable axially chiral radical cation of dithia-bridged hetero[4]helicene has been synthesized and analyzed using experimental and theoretical methods.
(more…)
The stability of odd- vs even-electron ions derived from pyridine-substituted N-heterotriangulenes has been investigated by both experiment and theory.
(more…)
OMx methods have once again been shown to give as accurate results as DFT methods, but substantially faster. The OM2 method has outperformed other semi-empirical methods and has essentially the same accuracy as BLYP for the distribution coefficient part of the SAMPL5 challenge.
Erlangen colleagues synthesized a series of N-heterotriangulenes, which can be used for dye-sensitized solar cells and potentially for fluorescent pH sensors.
What is the best semiempirical method to use for your system? Find out in the most extensive benchmark study.
Details about theory and implementation of up-to-now most advanced semiempirical quantum-chemical methods are published.
Δ-ML approach drastically reduces mean absolute error (MAE) in atomization enthalpies for PM7 and B3LYP methods in respect to accurate G4MP2 values
A highlight by Jan Jensen about the Δ-ML approach proposed by us [1] was the most viewed highlight in Computational Chemistry Highlights in 2015. This marks a pleasant ending of the last-year research on improving accuracy of computationally less demanding methods for computational chemistry. The Δ-ML approach along with the automatic parametrization technique (APT) [2] corrects faults of both lower level quantum mechanics (QM) and ML approaches by combining them in hybrid QM/ML techniques. The highlighted technique uses ML to correct errors of quantum chemical calculations after they were performed, while APT uses ML to improve quantum chemical method before running calculations with it.
1. Pavlo O. Dral, O. Anatole von Lilienfeld, Walter Thiel, Machine Learning of Parameters for Accurate Semiempirical Quantum Chemical Calculations. J. Chem. Theory Comput. 2015, 11, 2120–2125. DOI: 10.1021/acs.jctc.5b00141.
2. Raghunathan Ramakrishnan, Pavlo O. Dral, Matthias Rupp, O. Anatole von Lilienfeld, Big Data meets Quantum Chemistry Approximations: The Δ-Machine Learning Approach. J. Chem. Theory Comput. 2015, ASAP. DOI: 10.1021/acs.jctc.5b00099.
arXiv:1503.04987 [physics.chem-ph].
Wikipedia is treasure trove of free knowledge shared by volunteers. While English Wikipedia is the most known and has the largest number of articles, you may find information in more than 200 different languages. Obviously some topics are covered better in specific languages. For instance, geography of Germany is way better described in German Wikipedia than in English.
