Journal of Chemical Physics最新文献

The Effective Fragment Potential (EFP) method, a polarizable quantum mechanics-based force field for describing non-covalent interactions, is utilized to calculate protein-ligand interactions in seven inactive cyclin-dependent kinase 2-ligand complexes, employing structural data from molecular dynamics simulations to assess dynamic and solvent effects. Our results reveal high correlations between experimental binding affinities and EFP interaction energies across all the structural data considered. Using representative structures found by clustering analysis and excluding water molecules yields the highest correlation (R2 of 0.95). In addition, the EFP pairwise interaction energy decomposition analysis identifies critical interactions between the ligands and protein residues and provides insight into their nature. Overall, this study indicates the potential applications of the EFP method in structure-based drug design.

We have quantum chemically analyzed the trends in bond dissociation enthalpy (BDE) of H3C-XHn single bonds (XHn = CH3, NH2, OH, F, Cl, Br, I) along three different dissociation pathways at ZORA-BLYP-D3(BJ)/TZ2P: (i) homolytic dissociation into H3C∙ + ∙XHn, (ii) heterolytic dissociation into H3C+ + -XHn, and (iii) heterolytic dissociation into H3C- + +XHn. The associated BDEs for the three pathways differ not only quantitatively but, in some cases, also in terms of opposite trends along the C-X series. Based on activation strain analyses and quantitative molecular orbital theory, we explain how these differences are caused by the profoundly different electronic structures of, and thus bonding mechanisms between, the resulting fragments in the three different dissociation pathways. We demonstrate that the nature and strength of a chemical bond are only fully defined when considering both (i) the molecule in which the bond exists and (ii) the fragments from which it forms or into which it dissociates.

Understanding the ultrafast vibrational relaxation following photoexcitation of molecules in a condensed phase is essential to predict the outcome and improve the efficiency of photoinduced molecular processes. Here, the vibrational decoherence and energy relaxation of a binuclear complex, [Pt2(P2O5H2)4]4- (PtPOP), upon electronic excitation in liquid water and acetonitrile are investigated through direct adiabatic dynamics simulations. A quantum mechanics/molecular mechanics (QM/MM) scheme is used where the excited state of the complex is modeled with orbital-optimized density functional calculations while solvent molecules are described using potential energy functions. The decoherence time of the Pt-Pt vibration dominating the photoinduced dynamics is found to be ∼1.6 ps in both solvents. This is in excellent agreement with experimental measurements in water, where intersystem crossing is slow (>10 ps). Pathways for the flow of excess energy are identified by monitoring the power of the solvent on vibrational modes. The latter are obtained as generalized normal modes from the velocity covariances, and the power is computed using QM/MM embedding forces. Excess vibrational energy is found to be predominantly released through short-range repulsive and attractive interactions between the ligand atoms and surrounding solvent molecules, whereas solute-solvent interactions involving the Pt atoms are less important. Since photoexcitation deposits most of the excess energy into Pt-Pt vibrations, energy dissipation to the solvent is inefficient. This study reveals the mechanism behind the exceptionally long vibrational coherence of the photoexcited PtPOP complex in solution and underscores the importance of short-range interactions for accurate simulations of vibrational energy relaxation of solvated molecules.

Membrane properties are determined in part by lipid composition, and cholesterol plays a large role in determining these properties. Cellular membranes show a diverse range of cholesterol compositions, the effects of which include alterations to cellular biomechanics, lipid raft formation, membrane fusion, signaling pathways, metabolism, pharmaceutical therapeutic efficacy, and disease onset. In addition, cholesterol plays an important role in non-cellular membranes, with its concentration in the skin lipid matrix being implicated in several skin diseases. In phospholipid membranes, cholesterol increases the tail ordering of neighboring lipids, decreasing the membrane lateral area and increasing the thickness. This reduction in the lateral area, known as the cholesterol condensing effect, results from cholesterol-lipid mixtures deviating from ideal mixing. Capturing the cholesterol condensing effect is crucial for molecular dynamics simulations as it directly affects the accuracy of predicted membrane properties, which are essential for understanding membrane function. We present a comparative analysis of cholesterol models across several popular force fields: CHARMM36, Slipids, Lipid17, GROMOS 53A6L, GROMOS-CKP, MARTINI 2, MARTINI 3, and ELBA. The simulations of 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC) and 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) membranes with varying cholesterol concentrations were conducted to calculate the partial-molecular areas of cholesterol and other condensing parameters, which are compared to the experimental data for validation. While all tested force fields predict small negative deviations from ideal mixing in cholesterol-DOPC membranes, only all-atom force fields capture the larger deviations expected in DMPC membranes. United-atom and coarse-grained models under-predict this effect, condensing fewer neighboring lipids by smaller magnitudes, resulting in too small deviations from ideal mixing. These results suggest that all-atom force fields, particularly CHARMM36 or Slipids, should be used for accurate simulations of cholesterol-containing membranes.

Heteropolar two-dimensional materials, including hexagonal boron nitride (hBN), are promising candidates for seawater desalination and osmotic power harvesting, but previous simulation studies have considered bare, unterminated nanopores in molecular dynamics (MD) simulations. There is presently a lack of force fields to describe functionalized nanoporous hBN in aqueous media. To address this gap, we conduct density functional theory (DFT)-based ab initio MD simulations of hBN nanopores surrounded by water molecules. The results reveal a high propensity for hydrogen (H) and hydroxyl (OH) functionalization at boron edges, while nitrogen edges are functionalized with H and occasionally with oxygen (O), highlighting a route to tune membranes. We demonstrate the role of the Grotthuss mechanism during the functionalization of hBN edges in water. We develop high-fidelity force fields for H- and OH-functionalized hBN nanopores using potential energy surface fitting based on DFT calculations. The nonbonded parameters for H functionalization are obtained by training a force field for borazine (B3N3H6). We find that the proposed force field enables stable MD simulations of water/ion transport through B- and N-terminated hBN nanopores. Our results also indicate that previous studies that considered bare nanopores without functional groups overestimated the water flux and underestimated the ionic rejection of nanoporous hBN. Overall, our work is expected to enable the realistic modeling of edge-functionalized hBN in aqueous media for various application areas.

Time-resolved spectroscopy is an important tool for probing photochemically induced nonequilibrium dynamics and energy transfer. Herein, a method is developed for the ab initio simulation of vibronic spectra and dynamical processes. This framework utilizes the recently developed nuclear-electronic orbital time-dependent configuration interaction (NEO-TDCI) approach, which treats all electrons and specified nuclei quantum mechanically on the same footing. A strategy is presented for calculating time-resolved vibrational and electronic absorption spectra from any initial condition. Although this strategy is general for any TDCI implementation, utilizing the NEO framework allows for the explicit inclusion of quantized nuclei, as illustrated through the calculation of vibrationally hot spectra. Time-resolved spectra produced by either vibrational or electronic Rabi oscillations capture ground-state absorption, stimulated emission, and excited-state absorption between vibronic states. This methodology provides the foundation for fully ab initio simulations of multidimensional spectroscopic experiments.

The crystal and electronic structure of ZrxTi1-xSe2 (0 < x < 1) compounds and their electrical resistivity have been studied in detail for the first time. A combination of soft x-ray spectroscopic methods (XPS, XAS, and ResPES) was used to investigate the electronic structure. The lattice parameters as a function of the metal concentration x obey Vegard's law. It was shown that the substitution of Ti by Zr results in an increase in the Fermi energy, attributed to the lower binding energy of Zr 4d compared to Ti 3d in the ZrxTi1-xSe2 valence band. Given that the oxidation states of both Ti and Zr are +4, and the concentration of free charge carriers remains unchanged upon substitution, the observed effect is explained by a reduced density of electronic states near the Fermi level. The influence of temperature on the Ti 2p-3d and Zr 3p-4d ResPES spectra is interpreted in terms of pseudodoping occurring with the substitution of Ti by Zr.

Identifying informative low-dimensional features that characterize dynamics in molecular simulations remains a challenge, often requiring extensive manual tuning and system-specific knowledge. Here, we introduce geom2vec, in which pretrained graph neural networks (GNNs) are used as universal geometric featurizers. By pretraining equivariant GNNs on a large dataset of molecular conformations with a self-supervised denoising objective, we obtain transferable structural representations that are useful for learning conformational dynamics without further fine-tuning. We show how the learned GNN representations can capture interpretable relationships between structural units (tokens) by combining them with expressive token mixers. Importantly, decoupling training the GNNs from training for downstream tasks enables analysis of larger molecular graphs (that can represent small proteins at all-atom resolution) with limited computational resources. In these ways, geom2vec eliminates the need for manual feature selection and increases the robustness of simulation analyses.

The detection of HC(S)CN in TMC-1 suggests that HC(S)NC may also exist. To aid in its possible detection, HC(S)NC and its deuterated isotopologue DC(S)NC were investigated via high-level ab initio methods, specifically CCSD(T) and CCSD(T)-F12. By utilizing multidimensional potential energy surfaces derived from explicitly correlated coupled-cluster calculations, we analyzed their geometrical parameters, vibrational frequencies, rotational constants, and a comprehensive set of spectroscopic constants generated via the vibrational second-order perturbation theory, vibrational self-consistent field, and vibrational configuration interaction theory(VCI) approaches. HC(S)NC is thermodynamically stable relative to the HCS + NC dissociation limit, with a predicted bond dissociation energy of 4.1 eV. The calculated vibrational frequencies are characterized by two bright modes that correspond to CN stretching. Finally, HC(S)NC shows a significant dipole moment, predicted to be 1.9 D, making its detection via rotational spectroscopy plausible.

We derive a new expression for the strength of a hydrogen bond (VHB) in terms of the elongation of the covalent bond of the donor fragment participating in the hydrogen bond (ΔrHB) and the intermolecular coordinates R (separation between the heavy atoms) and θ (deviation of the hydrogen bond from linearity). The expression includes components describing the covalent D-H bond of the hydrogen bond donor via a Morse potential, the Pauli repulsion, and electrostatic interactions between the constituent fragments using a linear expansion of their dipole moment and a quadratic expansion of their polarizability tensor. We fitted the parameters of the model using ab initio electronic structure results for six hydrogen bonded dimers, namely, NH3-NH3, H2O-H2O, HF-HF, H2O-NH3, HF-H2O, and HF-NH3, and validated its performance for extended parts of their potential energy surfaces, resulting in a mean absolute error ranging from 0.07 to 0.31 kcal/mol. The derived expression describes the energy-structure relationship in terms of a single structural parameter, namely, the elongation of the donor's covalent bond (ΔrHB), and suggests the novel relationship of 8.0 kcal/mol pm-1 (or 0.8 kcal/mol per 0.001 Å elongation). This structural parameter is easily obtained from theory and can serve as the single descriptor of the strength of individual hydrogen bonds.