Tree based machine learning framework for predicting ground state energies of molecules.

We present an application of the boosted regression tree algorithm for predicting ground state energies of molecules made up of C, H, N, O, P, and S (CHNOPS). The PubChem chemical compound database has been incorporated to construct a dataset of 16 242 molecules, whose electronic ground state energies have been computed using density functional theory. This dataset is used to train the boosted regression tree algorithm, which allows a computationally efficient and accurate prediction of molecular ground state energies. Predictions from boosted regression trees are compared with neural network regression, a widely used method in the literature, and shown to be more accurate with significantly reduced computational cost. The performance of the regression model trained using the CHNOPS set is also tested on a set of distinct molecules that contain additional Cl and Si atoms. It is shown that the learning algorithms lead to a rich and diverse possibility of applications in molecular discovery and materials informatics.

Role of electronic localization in the phosphorescence of iridium sensitizing dyes.

In this work we present a systematic study of three representative iridium dyes, namely, Ir(ppy)(3), FIrpic, and PQIr, which are commonly used as sensitizers in organic optoelectronic devices. We show that electronic correlations play a crucial role in determining the excited-state energies in these systems, due to localization of electrons on Ir d orbitals. Electronic localization is captured by employing hybrid functionals within time-dependent density-functional theory and with Hubbard-model corrections within the Δ-SCF approach. The performance of both methods are studied comparatively and shown to be in good agreement with experiment. The Hubbard-corrected functionals provide further insight into the localization of electrons and on the charge-transfer character of excited-states. The gained insight allows us to comment on envisioned functionalization strategies to improve the performance of these systems. Complementary discussions on the Δ-SCF method are also presented in order to fill some of the gaps in the literature.

Transport properties of KTaO3 from first-principles.

The transport properties of the perovskites KTaO3 are calculated using first-principles methods. Our study is based on Boltzmann transport theory and the relaxation time approximation, where the scattering rate is calculated using an analytical model describing the interactions of electrons and longitudinal optical phonons. We compute the room-temperature electron mobility and Seebeck coefficients of KTaO3, and SrTiO3 for comparison, for a range of electron concentrations. The comparison between the two materials provides insight into the mechanisms that determine room-temperature electron mobility, such as the effect of band-width and spin-orbit splitting. The results, combined with the efficiency of the computational scheme developed in this study, provide a path to investigate and discover materials with targeted transport properties.

Searching for high magnetization density in bulk Fe: the new metastable Fe6 phase.

We report the discovery of a new allotrope of iron by first principles calculations. This phase has Pmn2(1) symmetry, a six-atom unit cell (hence the name Fe6), and the highest magnetization density (Ms) among all the known crystalline phases of iron. Obtained from the structural optimizations of the Fe3C-cementite crystal upon carbon removal, Pmn2(1) Fe6 is shown to result from the stabilization of a ferromagnetic FCC phase, further strained along the Bain path. Although metastable from 0 to 50 GPa, the new phase is more stable at low pressures than the other well-known HCP and FCC allotropes and smoothly transforms into the FCC phase under compression. If stabilized to room temperature, for example, by interstitial impurities, Fe6 could become the basis material for high Ms rare-earth-free permament magnets and high-impact applications such as light-weight electric engine rotors or high-density recording media. The new phase could also be key to explaining the enigmatic high Ms of Fe16N2, which is currently attracting intense research activity.

Tuning bad metal and non-Fermi liquid behavior in a Mott material: Rare-earth nickelate thin films.

Resistances that exceed the Mott-Ioffe-Regel limit (known as bad metal behavior) and non-Fermi liquid behavior are ubiquitous features of the normal state of many strongly correlated materials. We establish the conditions that lead to bad metal and non-Fermi liquid phases in NdNiO3, which exhibits a prototype bandwidth-controlled metal-insulator transition. We show that resistance saturation is determined by the magnitude of Ni eg orbital splitting, which can be tuned by strain in epitaxial films, causing the appearance of bad metal behavior under certain conditions. The results shed light on the nature of a crossover to a non-Fermi liquid metal phase and provide a predictive criterion for Anderson localization. They elucidate a seemingly complex phase behavior as a function of film strain and confinement and provide guidelines for orbital engineering and novel devices.

Origin of magnetic interactions and their influence on the structural properties of Ni2MnGa and related compounds.

In this work, we perform first-principles DFT calculations to investigate the interplay between magnetic and structural properties in Ni(2)MnGa. We demonstrate that the relative stability of austenite (cubic) and non-modulated martensite (tetragonal) phases depends critically on the magnetic interactions between Mn atoms. While standard approximate DFT functionals stabilize the latter phase, a more accurate treatment of electronic localization and magnetism, obtained with DFT+U, suppresses the non-modulated tetragonal structure for the stoichiometric compound, in better agreement with experiments. We show that the Anderson impurity model, with Mn atoms treated as magnetic impurities, can explain this observation and that the fine balance between super-exchange RKKY type interactions mediated by Ni d and Ga p orbitals determines the equilibrium structure of the crystal. The Anderson model is also demonstrated to capture the effect of the number of valence electrons per unit cell on the structural properties, often used as an empirical parameter to tune the behavior of Ni(2)MnGa based alloys. Finally, we show that off-stoichiometric compositions with excess Mn promote transitions to a non-modulated tetragonal structure, in agreement with experiments.

Instability of anisotropic cosmological solutions supported by vector fields.

Models with vector fields acquiring a nonvanishing vacuum expectation value along one spatial direction have been proposed to sustain a prolonged stage of anisotropic accelerated expansion. Such models have been used for realizations of early time inflation, with a possible relation to the large scale cosmic microwave background anomalies, or of the late time dark energy. We show that, quite generally, the concrete realizations proposed so far are plagued by instabilities (either ghosts or unstable growth of the linearized perturbations) which can be ultimately related to the longitudinal vector polarization present in them. Phenomenological results based on these models are therefore unreliable.