Data-driven many-body potentials from density functional theory for aqueous phase chemistry

Efficient charge-transfer (CT) phosphorescence in the near-IR (NIR) spectral region is reported for four substituted Ru-(R-dipyrrinato) complexes, [Ru(bpy)2(R-dipy)](PF6), where bpy is 2,2'-bipyridine and the substituent R is phenyl (ph), 2,4,6-trimethylphenyl, 4-carboxyphenyl (HOOC-ph), or 4-pyridinyl. The experimentally determined phosphorescence efficiency, ιem(p) = kRAD(p)/(νem(p))3 (where kRAD(p) and νem(p) are the phosphorescence rate constant and the phosphorescence frequency, respectively), of the [Ru(bpy)2(R-dipy)]+ complexes was approximately double that of [Ru(bpy)(Am)4]2+ complexes (Am = ammine ligand) in the NIR region. Density functional theory (DFT) modeling indicated two strikingly different electronic configurations of the triplet emitting state (Te) in the two types of complexes. The Te of [Ru(bpy)2(R-dipy)]+ complexes shows a CT-type corresponding to the metal-to-ligand charge transfer (MLCT)-(Ru-(R-dipy)) and the ππ*-(R-dipy) moiety configurations, and the Te state in the [Ru(bpy)(Am)4]2+ complexes corresponds to an approximately MLCT excited state consisting of mostly MLCT-(Ru-bpy) with a minimal ππ*(bpy) contribution. DFT modeling also indicated that the low-energy singlet excited states in the Te geometry (Sn(T)) of the [Ru(bpy)2(ph-dipy)]+ complex consist of numerous CT-Sn(T)-type states of the Ru-dipy and Ru-bpy moieties, whereas the [Ru(bpy)(Am)4]2+ ions show quite simple MLCT-Sn(T)-type states of the Ru-bpy moiety. Based on experimental observations, DFT modeling, and the plain spin-orbit coupling (SOC) principle, we conclude that the remarkably high ιem(p) amplitudes of the [Ru(bpy)2(R-dipy)]+ complexes relative to those of [Ru(bpy)(Am)4]2+ complexes can be attributed to the relatively substantial contribution of intrinsic SOC-mediated intensity stealing from the numerous low-energy CT-type Sn(T) states.

We present OrbNet Denali, a machine learning model for an electronic structure that is designed as a drop-in replacement for ground-state density functional theory (DFT) energy calculations. The model is a message-passing graph neural network that uses symmetry-adapted atomic orbital features from a low-cost quantum calculation to predict the energy of a molecule. OrbNet Denali is trained on a vast dataset of 2.3 × 106 DFT calculations on molecules and geometries. This dataset covers the most common elements in biochemistry and organic chemistry (H, Li, B, C, N, O, F, Na, Mg, Si, P, S, Cl, K, Ca, Br, and I) and charged molecules. OrbNet Denali is demonstrated on several well-established benchmark datasets, and we find that it provides accuracy that is on par with modern DFT methods while offering a speedup of up to three orders of magnitude. For the GMTKN55 benchmark set, OrbNet Denali achieves WTMAD-1 and WTMAD-2 scores of 7.19 and 9.84, on par with modern DFT functionals. For several GMTKN55 subsets, which contain chemical problems that are not present in the training set, OrbNet Denali produces a mean absolute error comparable to those of DFT methods. For the Hutchison conformer benchmark set, OrbNet Denali has a median correlation coefficient of R2 = 0.90 compared to the reference DLPNO-CCSD(T) calculation and R2 = 0.97 compared to the method used to generate the training data (ωB97X-D3/def2-TZVP), exceeding the performance of any other method with a similar cost. Similarly, the model reaches chemical accuracy for non-covalent interactions in the S66x10 dataset. For torsional profiles, OrbNet Denali reproduces the torsion profiles of ωB97X-D3/def2-TZVP with an average mean absolute error of 0.12 kcal/molfor the potential energy surfaces of the diverse fragments in the TorsionNet500 dataset.

We present an automated procedure for computing stacking fault energies in random alloys from large-scale simulations using moment tensor potentials (MTPs) with the accuracy of density functional theory (DFT). To that end, we develop an algorithm for training MTPs on random alloys. In the first step, our algorithm constructs a set of $\ensuremath{\sim}10$ 000 or more training candidate configurations with 50--100 atoms that are representative for the atomic neighborhoods occurring in the large-scale simulation. In the second step, we use active learning to reduce this set to $\ensuremath{\sim}100$ most distinct configurations---for which DFT energies and forces are computed and on which the potential is ultimately trained. We validate our algorithm for the MoNbTa medium-entropy alloy by showing that the MTP reproduces the DFT $\frac{1}{4}[111]$ unstable stacking fault energy over the entire compositional space up to a few percent. Contrary to state-of-the-art methods, e.g., the coherent potential approximation (CPA) or special quasi-random structures (SQSs), our algorithm naturally accounts for relaxation, is not limited by DFT cell sizes, and opens opportunities to efficiently investigate follow-up problems, such as chemical ordering. In a broader sense, our algorithm can be easily modified to compute related properties of random alloys, for instance, misfit volumes, or grain boundary energies. Moreover, it forms the basis for an efficient construction of MTPs to be used in large-scale simulations of multicomponent systems.

To evaluate the accuracy of density functional theory (DFT) for describing van der Waals interaction between rare‐gas atoms, we performed coupled‐cluster theory restricted to single, double, and noniterative triple excitations [CCSD(T)], fourth‐order perturbation theory including single, double, and quadruple substitutions [MP4(SDQ)], and several DFT calculations for He, Ne, Ar, and Kr dimers. The results clarified that (1) CCSD(T) and MP4(SDQ) with an “adjusted” diffuse basis‐set reproduce the experimental trend in binding energies as well as equilibrium distances of these dimers, (2) MP3 and MP2 underestimate the binding energies, (3) Becke's three‐parameter exchange functional and the gradient‐corrected functional of Lee, Yang, and Paar (B3LYP) and B3PW91 completely fail to describe the binding, and (4) Perdew–Wang exchange‐correlation functional (PW91PW91) and its modified ones (MPWPW91 and MPW1PW91) can provide the binding states, although the calculated binding energies cannot reproduce the experimental trend in binding energies. Therefore, we conclude that the above DFT functionals cannot quantitatively describe the van der Waals interaction between rare‐gas atoms and that the Perdew‐Wang–type exchange functional needs further improvement to describe the interaction. © 2003 Wiley Periodicals, Inc. Int J Quantum Chem, 2003

The last decade has seen a continued development of better experimental techniques to measure equation-of-state (EOS) for various materials. These improvements of both static and shock-compression approaches have increased the accuracy of the EOS and challenged the complimentary theoretical modeling. The conventional modeling of EOS, at least at pressure and temperature conditions that are not too extreme, is founded on density-functional theory (DFT). Naturally, there is an increased interest in the accuracy of DFT as the measurements are becoming more refined and there is a particular interest in the robustness and validity of DFT at conditions where experimental data are not available. Here, we consider a broad and large set of 64 elemental solids from low atomic number Z up to the very high Z actinide metals. The intent is to compare DFT with experimental zero-temperature isotherms up to 1 Mbar (100 GPa) and draw conclusions regarding the theoretical (DFT) error and quantify a reasonable and defensible approach to define the theoretical uncertainty. We find that in all 64 cases the DFT error at high pressure is smaller than or equal to the DFT error at lower pressures which thus provides an upper bound to the error at high compression.

In the beginning of quantum mechanical Density-Functional Theory (DFT), there was the Thomas-Fermi (TF) model, which uses the electron density p(r) (a function of only 3 coordinates) as the only physical variable. Calculations with this model were inexpensive but yielded poor numerical results due to a lack of understanding of exchange-correlation effects and the kinetic-energy density functional. Many years later, Hohenberg and Kohn (HK) established the formal foundation for DFT; Kohn and Sham (KS) devised a practical implementation and brought DFT into mainstream calculations of electronic structure. Although the KS formulation allows exact evaluation of the KS kinetic energy (Ts [p]), the one-electron orbitals introduced by the KS scheme inevitably encumber the formulation in three ways: (i) 3N (vs. 3) degrees of freedom, (ii) orbital orthonormalization, and (iii) Brillouin-zone (k-point) sampling in condensed phases. Given the accuracy of DFT with present exchange-correlation density functionals, it is logical to conclude that the last frontier in DFT is a better representation of the kinetic energy solely in terms of the density. If this is true, KS orbitals will be completely eliminated from DFT formulation, and the density can be solved directly from the TF-HK equation. This is certainly superior to the KS scheme because all energy terms can be computed in momentum space with an effectively linear scaling, O(MlnM), where M is the integration grid size. This work reviews major ideas in the design of such optimal orbital-free kinetic-energy density functionals and their applications.

In contrast to the ground state, the calculation of the infrared (IR) spectroscopy of molecular singlet excited states represents a substantial challenge. Here, we use the structural IR fingerprint of the singlet excited states of a range of coumarin dyes to assess the accuracy of density functional theory based methods for the calculation of excited state IR spectroscopy. It is shown that excited state Kohn-Sham density functional theory provides a high level of accuracy and represents an alternative approach to time-dependent density functional theory for simulating the IR spectroscopy of singlet excited states.

For a long time, the prediction of molecular crystal structures from first principles has been one of the most challenging computational problems. Because of the large number of degrees of freedom, methods with low computational effort have to be used. The price for this efficiency is, however, a loss of accuracy. In recent years, important progress has been made in the development of dispersion corrections [1-3], which raise the accuracy of density functional theory (DFT) calculations to a much higher level. Based on these developments, the software package GRACE [4] was the first program ever to correctly predict all four crystal structures at the 4th crystal structure prediction blind test, organized by G. M. Day and the Cambridge Crystallographic Data Centre in 2007 [5]. GRACE uses dispersion-corrected density functional theory calculations (DFT-D) to generate reference data to which a tailor-made force field (TMFF) is fitted for each respective molecule [6]. Crystal structures are generated with a Monte-Carlo parallel-tempering algorithm that uses the TMFF for the evaluation of lattice energies. Only the most stable crystal structures according to the TMFF are then re-optimized and re-ranked using DFT-D. The DFT calculations are performed using an interface to the two common ab-inito programs VASP [7] and QuantumESPRESSO [8]. In the most recent, 5th blind test, organized in 2010, the excellent agreement between experimental and predicted structures for small, neutral molecules was confirmed. However, it turned out that the crystal structure prediction of molecular salts and hydrates is still a challenging task. The inability of DFT without exact exchange to describe anions properly, as recently published by Jensen [9], suggests that the inclusion of the Hartree-Fock exchange should improve the accuracy for molecular salts and hydrates. Therefore, we will present an assessment of a variety of current density functionals, in which the dispersion-correction parameters were fitted with respect to a minimization of the deviation between experimental and computed crystal structures.

The magnetisation density of different iridium PNP pincer-type complexes depends on the chosen quantum chemistry method.

The terahertz absorption spectra of aspirin and benzoic acid have been measured in the range 200-500 cm-1 (6-15 THz). Density-functional theory (DFT) modeling has assigned fundamental vibrational modes to the observed absorption bands. Hydrogen bonds between the crystalline planes of aspirin resulted in better agreement between the experimental and modeled spectra than for benzoic acid. The similar structure of these two molecules suggests a similar absorption spectrum, which indeed was obtained experimentally. However, the detailed crystal structure and molecular differences result in some of the apparently common absorption bands being assigned to different vibrational modes through the DFT modeling. Thus, our study importantly reveals that even though crystalline forms of two similar molecules may have similar experimental terahertz spectra, the resemblance may be superficial rather than fundamental.

Establishing the fundamental possibility of the existence of the heteroligand macrotetracyclic complexes of vanadium, chromium, manganese, and iron-containing in the inner coordination sphere phthalocyanine, oxygen (O2-) and fluorine (F-) ions and having general [MPc(O)F] formula (M= V, Cr, Mn, Fe), by using of quantum-chemical calculation of parameters of their molecular/electronic structures and thermodynamical characteristics. The molecular and electronic structures of the above-mentioned heteroligand macrotetracyclic chelates of 3d-elements (M) of the type [MPc(O)F] (M= V, Cr, Mn, Fe) which are unknown at present, were theoretically investigated. Standard thermodynamic parameters of formation (standard enthalpy ΔH0 f, 298, entropy S0 f, 298, and Gibbs’s energy ΔG0 f, 298) for these macrocyclic compounds were calculated, too. Identifying details of molecular and electronic structures of compounds indicated above. Density functional theory (DFT) model chemistries (B3PW91/TZVP and OPBE/TZVP) with a combination of the D3 version of Grimme’s dispersion. The data on the geometric parameters of the molecular structure of these complexes are presented; it was shown that MN4 chelate nodes, all metal-chelate and 6-membered non-chelate rings in each of these macrocyclic coordination compounds, are practically planar with a small deviation from coplanarity (not more 3o); nonetheless, N4 grouping from donor nitrogen atoms and 5-membered non-chelate rings are strictly planar. Wherein, the bond angles between two donor nitrogen atoms and M atom are not equal to 90o; a similar situation occurs for the bond angles between donor atoms N, M, and O or F (notwithstanding the bond angles formed by M, O, and F atoms are exactly 180°). Also, NBO analysis data and the values of the standard enthalpy, entropy, and Gibbs energy of the formation of these compounds were presented. Specific features of DFT calculated molecular and electronic structures of the heteroligand metal macrocyclic compounds have been discussed. It has been shown that good agreement between the parameters of molecular structures obtained by two various DFT model chemistries takes place. Also, it has been noted that predicting the possibility of the existence of exotic coordination compounds and modeling their molecular/electronic structures using modern quantum chemical calculations (and, in particular, using DFT of various levels) is a very useful tool for solving problems associated with such synthesis.

Density functional theory (DFT) calculations allow us to reproduce the SERS (surface-enhanced Raman scattering) spectra of molecules adsorbed on nanostructured metal surfaces and extract the most information this spectroscopy is potentially able to provide. The latter point mainly concerns the anchoring mechanism and the bond strength between molecule and metal as well as the structural and electronic modifications of the adsorbed molecule. These findings are of fundamental importance for the application of this spectroscopic technique. This review presents and discusses some SERS–DFT studies carried out in Italy as a collaboration between the universities of Modena and Reggio-Emilia and of Florence, giving an overview of the information that we can extract with a combination of experimental SERS spectra and DFT modeling. In addition, a selection of the most recent studies and advancements on the DFT approach to SERS spectroscopy is reported with commentary.

The aggregation behaviors of silver atoms modulated by polymorphic DNA templates involving i-motif, G-quadruplex, and the Watson-Crick duplex, were investigated by using the density functional theory (DFT) calculations, combining with the experimental characterizations of CD, UV, fluorescence measurements and TEM, in order to understand the reason in the molecular level that polymorphic DNA templates affect the fluorescence emitting species of Ag nanomaterials. First, the affinity sites of silver ions on different DNA templates were analyzed by using DFT calculations, and the conformational variations of DNA templates caused by silver ions and atoms were disclosed. Second, the aggregation behaviors of silver atoms constrained by the polymorphic DNA templates were studied by DFT modeling, and distinct fluorescence property of nanosilvers templated by polymorphic DNA were evaluated using the time-dependent DFT calculations. It is illustrated that with the DNA template adopting i-motif or the duplex the silver atoms tend to aggregate inside the encapsulated spaces of nucleobases, and the formed silver nanoclusters are positively charged with high fluorescent spectral features; whereas with the template of the G-quadruplex the silver atoms are preferential to aggregate outside of the G-tetrad, which results in the formation of larger silver crystals without fluorescence property. The results obtained here are useful to explore the nucleation and growth mechanism of silver nanomaterials regulated by the structure-specific DNA templates, which is important to rational design of desirable fluorescent emitters for sensing in the field from biology to nanoscience.

The density functional theory (DFT) method has been used to study the electronic communication in strongly interacting oxo-bridged di-{Mo(II,I)(NO(+))}(3+,2+) complexes stabilized by tris(3-methylpyrazol-1-yl)borate, [Tp(Me)](-) (dihydroxy 1' and its modified analogs), having fully localized valences on the two Mo centers (Class I), despite a short (ca. 3.8 A) Mo...Mo distance. Structural and electrochemical (separation between the redox potentials Delta(red/ox)E(1/2)) properties and IR spectra (in particular the nu(NO) frequencies) obtained from the B3LYP calculations for 1' are successfully related to experimental values. Strongly twisted geometry with the (O)N-Mo1...Mo2-N(O) angle close to 90 degrees (confirmed by DFT modeling performed for 1'(-1,0,+1) and X-ray diffraction study of [{Mo(NO)(Tp(Me2))(OH)}(2)(mu-O)](-) (1) presented herein) is a common, though so far not fully understood, structural feature of this class of mu-oxo species, in contrast to the closely related {Mo(V)(=O)}(3+) analogs. This study shows that the orthogonality of the local equatorial planes for the two Mo centers may be rationalized by the electronic structure, namely from the balance between the destabilizing repulsion of the Mo-based (d, pi(x)*)(b) electron pairs versus a favorable but relatively weak electron delocalization. Strongly repelling electron pairs avoid each other, which enforces the twisted geometries and blocks the electron delocalization. Steric hindrance (a nonbonding repulsion of the adjacent Tp(x) ligands and the weak hydrogen-bonding interactions, i.e., OH...ON, OH...OH, and C-H...O((NO/OH))) is shown not to be decisive since neither the removal of the inner 3-Me groups of [Tp(Me)](-) in complex 1' nor the substitution of OH groups by OCH(3) ligands did substantially influence the dihedral twist angle in the minimum energy structure. Yet the relative orientation of the {Mo(NO)}(2+,3+) cores along with the position of the bridging oxygen (significantly bent upon reduction) controls the prospective intramolecular through-bond electron transfer in the mixed valence form. Our DFT modeling demonstrates that a maximum delocalization (via a hole-transfer mechanism) of the unpaired electron in 1'(-), measured as a spin population on the nonreduced Mo2 center, is achieved for the structure with a torsional deflection of 23 degrees, at a cost of 16.5 kcal/mol. These results show that the electron exchange along the Mo-O-Mo array in the originally fully valence-trapped {17e:16e}(-) complexes may be controlled and can be thermally activated (e.g., using a high-boiling solvent or by irradiation at ca. 50-200 cm(-1)).

There has been some confusion in the published literature concerning the structure of Metastudtite (UO2)O2(H2O)2 where differing unit cells and space groups have been cited for this compound. Owing to the absence of a refined structure for Metastudtite, Weck et al. (2012) have documented a first-principles study of Metastudtite using density functional theory (DFT). Their model presents the structure of Metastudtite as an orthorhombic (space group Pnma) structure with lattice parameters of a = 8.45, b = 8.72, and c = 6.75 Å. A Powder Diffraction File (PDF) database entry has been allocated for this hypothetical Metastudtite phase based on the DFT modeling (see 01-081-9033) and aforementioned Dalton Trans. manuscript. We have obtained phase pure powder X-ray diffraction data for Metastudtite and have confirmed the model of Weck et al. via Rietveld refinement (see Figure 1). Structural refinement of this powder diffraction dataset has yielded updated refined parameters. The new cell has been determined as a = 8.411(1), b = 8.744(1), and c = 6.505(1) Å; cell volume = 478.39 Å3. There are only subtle differences between the refined structure and that of the first-principles model derived from DFT. Notably, the b-axis is significantly contracted in the final refinement as compared with DFT. There were also subtle changes to the U1, O1, and O3 atom positions. Tabulated powder diffraction data (d's and I's) for the Metastudtite have been derived from the refined model and these new values can serve to augment the PDF entry 01-081-9033 with a more updated entry based on observed X-ray powder diffraction data.