Search

Although potentially powerful, molecular oxygen is an inert oxidant due to the triplet nature of its ground state. Therefore, many enzymesse various metal cations (M) to produce singlet active species MO. In this communication we investigate the topology of the Electron Localization Function (ELF) within five biomimetic complexes which are representative of the strategies followed by metalloenzymes to activate O. Thanks to its coupling to the constrained DFT methods the ELF analysis reveals the tight connection between the spin state of the adduct and the spatial organization of the oxygen lone pairs. We suggest that enzymes could resort to spin state control to tune the regioselectivity of substrate oxidations. © 2010 Wiley Periodicals, Inc. J Comput Chem, 2011 [ABSTRACT FROM AUTHOR]

Ni‐CeO2 nanoparticles (NPs) are promising nanocatalysts for water splitting and water gas shift reactions due to the ability of ceria to temporarily donate oxygen to the catalytic reaction and accept oxygen after the reaction is completed. Therefore, elucidating how different properties of the Ni‐Ceria NPs relate to the activity and selectivity of the catalytic reaction, is of crucial importance for the development of novel catalysts. In this work the active learning (AL) method based on machine learning regression and its uncertainty is used for the global optimization of Ce(4‐x)NixO(8‐x) (x = 1, 2, 3) nanoparticles, employing density functional theory calculations. Additionally, further investigation of the NPs by mass‐scaled parallel‐tempering Born‐Oppenheimer molecular dynamics resulted in the same putative global minimum structures found by AL, demonstrating the robustness of our AL search to learn from small datasets and assist in the global optimization of complex electronic structure systems. [ABSTRACT FROM AUTHOR]

Reinforcement learning (RL) methods have helped to define the state of the art in the field of modern artificial intelligence, mostly after the breakthrough involving AlphaGo and the discovery of novel algorithms. In this work, we present a RL method, based on Q‐learning, for the structural determination of adsorbate@substrate models in silico, where the minimization of the energy landscape resulting from adsorbate interactions with a substrate is made by actions on states (translations and rotations) chosen from an agent's policy. The proposed RL method is implemented in an early version of the reinforcement learning software for materials design and discovery (RLMaterial), developed in Python3.x. RLMaterial interfaces with deMon2k, DFTB+, ORCA, and Quantum Espresso codes to compute the adsorbate@substrate energies. The RL method was applied for the structural determination of (i) the amino acid glycine and (ii) 2‐amino‐acetaldehyde, both interacting with a boron nitride (BN) monolayer, (iii) host‐guest interactions between phenylboronic acid and β‐cyclodextrin and (iv) ammonia on naphthalene. Density functional tight binding calculations were used to build the complex search surfaces with a reasonably low computational cost for systems (i)–(iii) and DFT for system (iv). Artificial neural network and gradient boosting regression techniques were employed to approximate the Q‐matrix or Q‐table for better decision making (policy) on next actions. Finally, we have developed a transfer‐learning protocol within the RL framework that allows learning from one chemical system and transferring the experience to another, as well as from different DFT or DFTB levels. [ABSTRACT FROM AUTHOR]

Genetic algorithms (GAs) are stochastic global search methods inspired by biological evolution. They have been used extensively in chemistry and materials science coupled with theoretical methods, ranging from force‐fields to high‐throughput first‐principles methods. The methodology allows an accurate and automated structural determination for molecules, atomic clusters, nanoparticles, and solid surfaces, fundamental to understanding chemical processes in catalysis and environmental sciences, for instance. In this work, we propose a new genetic algorithm software, GAMaterial, implemented in Python3.x, that performs global searches to elucidate the structures of atomic clusters, doped clusters or materials and atomic clusters on surfaces. For all these applications, it is possible to accelerate the GA search by using machine learning (ML), the ML@GA method, to build subsequent populations. Results for ML@GA applied for the dopant distributions in atomic clusters are presented. The GAMaterial software was applied for the automatic structural search for the Ti6O12 cluster, doping Al in Si11 (4Al@Si11) and Na10 supported on graphene (Na10@graphene), where DFTB calculations were used to sample the complex search surfaces with reasonably low computational cost. Finally, the global search by GA of the Mo8C4 cluster was considered, where DFT calculations were made with the deMon2k code, which is interfaced with GAMaterial. [ABSTRACT FROM AUTHOR]

Born-Oppenheimer molecular dynamics (BOMD) in combination with density functional theory, augmented with a damped empirical dispersion term, has been used to study the behavior of dilauroyl phosphatidylcholine (DLPC) isomers with temperature. In contrast to dimyristoyl phosphatidylcholine (DMPC), the BOMD results show the presence of various conformations at different temperatures. The molecular order-disorder process, quantified by the distance-fluctuation criterion as a function of temperature, is characterized by two transitions. This is in line with the known two-phase transitions of DLPC bilayers, involving trans to gauche conformational changes in the alkane chains. The different temperature dependence of the DLPC and DMPC molecules suggests that the experimentally observed unusual dynamics of DLPC bilayers compared to that of longer chain lipids is governed to a large extent by the intramolecular dynamics. A first-principles methodology applied at the molecular level can thus be an appropriate tool for microscopic analysis of the order-disorder transitions, which are related to the molecular structural transformations within large assemblies. © 2012 Wiley Periodicals, Inc. [ABSTRACT FROM AUTHOR]

Sum-over-states density functional perturbation theory (SOS-DFPT) (Malkin, V. G.; Malkina, O. L.; Casida, M. E.; Salahub, D. R. J Am Chem Soc 1994, 116, 5898) has been successful as a method for calculating nuclear magnetic resonance (NMR) chemical shifts. The key to this success is the introduction of an ad hoc correction to the excitation energies represented by simple orbital energy differences in uncoupled density functional theory. It has been suggested (Jamorski, C.; Casida, M. E.; Salahub, D. R. J Chem Phys 1996, 104, 5134) that the good performance of this methodology could be partly explained by the resemblance of the corrected excitation energy to the orbital energy difference given by time-dependent density functional theory (TDDFT). In fact, according to exact (wave function) time-dependent perturbation theory, both magnetic and electric perturbations may be described using essentially the same simple SOS expression. However in adiabatic TDDFT, with no explicit relativistic or current density functional dependence, the functional is approximate and so the magnetic and electric SOS expressions are different. Because TDDFT (neglecting relativistic and current density functional dependence) is formally exact for electric perturbations but not magnetic perturbations and because the two SOS expressions should have the same form, we propose that the SOS expression for electric perturbations should also be used for magnetic perturbations. We then go on to realize our theory by deriving a “Loc.3” approximation that is explicitly designed by applying the electric field SOS expression to magnetic fields within the two-level model and Tamm–Dancoff approximation. Test results for 13 small organic and inorganic molecules show that the Loc.3 approximation performs at least as well as the “Loc.1” and “Loc.2” approximations of SOS-DFPT. © 2002 Wiley Periodicals, Inc. Int J Quantum Chem, 2003 [ABSTRACT FROM AUTHOR]

Abstract: Exemplars of the state of the art in modeling enzymes are reviewed through a selection of works from leading schools using QM‐only cluster models, QM/MM models and QM/MM/MD models. A computational protocol is proposed, intended to guide neophytes through the myriad of methodological choices that are necessary for the successful study of chemical reactions in the complex enzymatic environment. [ABSTRACT FROM AUTHOR]

The efficient conversion of carbon dioxide into useful products is a prime challenge to modern chemistry. An alternative route to address this challenge based on a rhodium-substituted human carbonic anhydrase is described that can be considered the first cofactor-independent reductase. This artificial enzyme is able to convert CO2 into formic acid, with potential applications in renewable energy. Our quantum mechanical investigation (QM/QM′ method), which considers the entire catalytic pocket (390 atoms), provides evidence that the catalytic process is governed by an energetically favored σ-bond-metathesis mechanism and the rate-limiting step is the release of formic acid (11.7 kcal mol−1). Water molecules are found to play an active role during the chemical process by contributing to reduce dramatically the energy of the rate-limiting step and favoring an efficient regeneration of the catalyst. [ABSTRACT FROM AUTHOR]

Standard density functional theory (DFT) is augmented with a damped empirical dispersion term. The damping function is optimized on a small, well balanced set of 22 van der Waals (vdW) complexes and verified on a validation set of 58 vdW complexes. Both sets contain biologically relevant molecules such as nucleic acid bases. Results are in remarkable agreement with reference high-level wave function data based on the CCSD(T) method. The geometries obtained by full gradient optimization are in very good agreement with the best available theoretical reference. In terms of the standard deviation and average errors, results including the empirical dispersion term are clearly superior to all pure density functionals investigated—B-LYP, B3-LYP, PBE, TPSS, TPSSh, and BH-LYP—and even surpass the MP2/cc-pVTZ method. The combination of empirical dispersion with the TPSS functional performs remarkably well. The most critical part of the empirical dispersion approach is the damping function. The damping parameters should be optimized for each density functional/basis set combination separately. To keep the method simple, we optimized mainly a single factor, sR, scaling globally the vdW radii. For good results, a basis set of at least triple-ζ quality is required and diffuse functions are recommended, since the basis set superposition error seriously deteriorates the results. On average, the dispersion contribution to the interaction energy missing in the DFT functionals examined here is about 15 and 100% for the hydrogen-bonded and stacked complexes considered, respectively. © 2006 Wiley Periodicals, Inc. J Comput Chem 28: 555–569, 2007 [ABSTRACT FROM AUTHOR]