DFTB+ is a versatile community developed open source software package offering fast and efficient methods for carrying out atomistic quantum mechanical simulations. By implementing various methods approximating density functional theory (DFT), such as the density functional based tight binding (DFTB) and the extended tight binding method, it enables simulations of large systems and long timescales with reasonable accuracy while being considerably faster for typical simulations than the respective ab initio methods. Based on the DFTB framework, it additionally offers approximated versions of various DFT extensions including hybrid functionals, time dependent formalism for treating excited systems, electron transport using non-equilibrium Green’s functions, and many more. DFTB+ can be used as a user-friendly standalone application in addition to being embedded into other software packages as a library or acting as a calculation-server accessed by socket communication. We give an overview of the recently developed capabilities of the DFTB+ code, demonstrating with a few use case examples, discuss the strengths and weaknesses of the various features, and also discuss on-going developments and possible future perspectives.

Machine learning advances chemistry and materials science by enabling large-scale exploration of chemical space based on quantum chemical calculations. While these models supply fast and accurate predictions of atomistic chemical properties, they do not explicitly capture the electronic degrees of freedom of a molecule, which limits their applicability for reactive chemistry and chemical analysis. Here we present a deep learning framework for the prediction of the quantum mechanical wavefunction in a local basis of atomic orbitals from which all other ground-state properties can be derived. This approach retains full access to the electronic structure via the wavefunction at force-field-like efficiency and captures quantum mechanics in an analytically differentiable representation. On several examples, we demonstrate that this opens promising avenues to perform inverse design of molecular structures for targeting electronic property optimisation and a clear path towards increased synergy of machine learning and quantum chemistry.

Molecular adsorbates on metal surfaces exchange energy with substrate phonons and low-lying electron-hole pair excitations. In the limit of weak coupling, electron-hole pair excitations can be seen as exerting frictional forces on adsorbates that enhance energy transfer and facilitate vibrational relaxation or hot-electron mediated chemistry. We have recently reported on the relevance of tensorial properties of electronic friction [Phys. Rev. Lett. 116, 217601 (2016)] in dynamics at surfaces. Here we present the underlying implementation of tensorial electronic friction based on Kohn-Sham Density Functional Theory for condensed phase and cluster systems. Using local atomic-orbital basis sets, we calculate nonadiabatic coupling matrix elements and evaluate the full electronic friction tensor in the Markov limit. Our approach is numerically stable and robust as shown by a detailed convergence analysis. We furthermore benchmark the accuracy of our approach by calculation of vibrational relaxation rates and lifetimes for a number of diatomic molecules on metal surfaces. We find friction-induced mode-coupling between neighboring CO adsorbates on Cu(100) in a c(2x2) overlayer to be important to understand experimental findings.

Adsorption geometry and stability of organic molecules on surfaces are key parameters that determine the observable properties and functions of hybrid inorganic/organic systems (HIOSs). Despite many recent advances in precise experimental characterization and improvements in rst-principles electronic structure methods, reliable databases of structures and energetics for large adsorbed molecules are largely amiss. In this review, we present such a database for a range of molecules adsorbed on metal single-crystal surfaces. The systems we analyze include noble-gas atoms, conjugated aromatic molecules, carbon nanostructures, and heteroaromatic compounds adsorbed on ve dierent metal surfaces. The overall objective is to establish a diverse benchmark dataset that enables an assessment of current and future electronic structure methods, and motivates further experimental studies that provide ever more reliable data. Specically, the benchmark structures and energetics from experiment are here compared with the recently developed van der Waals (vdW) inclusive density-functional theory (DFT) method, DFT+vdW^surf. In comparison to 23 adsorption heights and 17 adsorption energies from experiment we nd a mean average deviation of 0.06 Å and 0.16 eV, respectively. This conrms the DFT+vdW^surf method as an accurate and ecient approach to treat HIOSs. A detailed discussion identies remaining challenges to be addressed in future development of electronic structure methods, for which the here presented benchmark database may serve as an important reference