The Journal of Physical Chemistry A最新文献

The photodissociation of the simplest diazirine, 3H-diazirine (cyclo-CH₂N₂), was studied in the gas phase following excitation of S₁ levels with one quantum of C-N symmetric stretching (ν₆, 802 cm^-1) or two quanta of C-N asymmetric stretching (2ν₉, 216 cm^-1). The angular and velocity distributions of the products were measured using 9.9 eV single photon vacuum ultraviolet photoionization and 70 eV electron impact ionization. Preferential scattering of products perpendicular to the laser polarization axis indicates that the transition is ¹B₁ ← ¹A₁ with dissociation occurring on subpicosecond time scales. From photofragment anisotropy measurements, initial asymmetric parent vibrational excitation results in shorter dissociation time scales as compared to symmetric stretching, suggesting that dissociation is initiated by asymmetric stepwise ring-opening. However, the final product translational energy distributions were nearly identical for each level, suggesting similar later-time dissociation dynamics. We observed no evidence for formation of ground state CH₂ (X̃ ³B₁). The CH₂ + N₂ products are formed with a most probable total internal energy of ∼2 eV. Although the translational energy distributions are consistent with production of highly vibrationally excited CH₂ (ã ¹A₁) following passage through a conical intersection, a significant yield of CH₂ (b̃ ¹B₁) is possible.

In this study, a high-dimensional neural network potential for the smectite pyrophyllite clay has been developed from density functional theory (DFT) data, including correction for dispersion interactions. The data set has been built from the adaptive learning approach, resulting in a diverse and very concise set of selected structures comprising only representative ones. Two neural network potential (NNP) data sets have been constituted from sets of energies and forces computed at two different levels of DFT accuracy. Validation tests show very good accuracy for the computed energies and forces of various systems differing by their size and simulation conditions. The developed potentials are able to reproduce structural parameters with excellent agreement with DFT values as well as experimental data and are the first NNPS able to reproduce clay layers' properties held together via van der Waals interactions. The NNP constructed from data of higher DFT levels shows better results for extreme condition simulations. In addition, elastic properties, exfoliation energies, and vibrational density of state are also well reproduced, showing better performances than standard force fields at a fraction of DFT computation time.

The photodissociation of the simplest diazirine, 3H-diazirine (cyclo-CH₂N₂), was studied in the gas phase following excitation of S₁ levels with one quantum of C–N symmetric stretching (ν₆, 802 cm^–1) or two quanta of C–N asymmetric stretching (2ν₉, 216 cm^–1). The angular and velocity distributions of the products were measured using 9.9 eV single photon vacuum ultraviolet photoionization and 70 eV electron impact ionization. Preferential scattering of products perpendicular to the laser polarization axis indicates that the transition is ¹B₁ ← ¹A₁ with dissociation occurring on subpicosecond time scales. From photofragment anisotropy measurements, initial asymmetric parent vibrational excitation results in shorter dissociation time scales as compared to symmetric stretching, suggesting that dissociation is initiated by asymmetric stepwise ring-opening. However, the final product translational energy distributions were nearly identical for each level, suggesting similar later-time dissociation dynamics. We observed no evidence for formation of ground state CH₂ (X̃ ³B₁). The CH₂ + N₂ products are formed with a most probable total internal energy of ∼2 eV. Although the translational energy distributions are consistent with production of highly vibrationally excited CH₂ (ã ¹A₁) following passage through a conical intersection, a significant yield of CH₂ (b̃ ¹B₁) is possible.

Knowledge of the redox properties of cytosine (C), uracil (U), and their natural derivatives is essential for a deeper understanding of DNA damage, repair, and epigenetic regulation. This study investigates the one-electron oxidation potential (E_ox, V) using DFT (B3LYP-D3) and DLPNO-CCSD(T) methods with explicit/implicit (SMD) solvation model. Calculations in the gas phase and aprotic solvents such as acetonitrile showed a high correlation with experimental data (0.96-0.98). In aqueous solutions at pH 7, oxidation potentials are significantly influenced by deprotonation equilibria, as acidic molecules like 5caC become easier to oxidize upon deprotonation. The resulting oxidation potentials reflect a complex interplay of substituent effects, acidity, and protonation states. A pH-dependent model based on the Nernst equation for aqueous solutions demonstrated a correlation coefficient of 0.93. The calculated E_ox values for cytosine epigenetic derivatives in water, accounting for deprotonation effects, follow the trend: d_5caC < 5mC < 5caC < 5hmC < C < 5dhmC < 5fC, where "d_" deprotonated, "5ca" 5-carboxy, "5m" 5-methyl, "5hm" 5-hydroxymethyl, "5dhm" 5-dihydroxymethyl, "5f" 5-formyl.

Hot exciton materials have been recently studied toward improving the efficiency of fluorescent organic light-emitting diodes (OLEDs). The improvement is achieved by harvesting triplet excitons through high-lying reverse intersystem crossing (hRISC), and for its success, it is necessary to suppress internal conversion (IC) from the spin-converting high-lying triplet state to any lower triplet states. Kasha’s rule dictates that such a process is not highly likely, and indeed, there is no direct evidence on inhibited triplet IC. Here, we suggest spin conversion in polaron pairs (PPs) as another channel that can also enhance singlet exciton generation. In our model, the spin states may interconvert by hyperfine coupling and the singlet exciton yields can be influenced by the relative rates of charge recombination of singlet and triplet PPs. We calculate the rate constants of the charge recombination, IC, ISC, and hRISC processes of hot exciton molecules and apply them to generate a kinetic picture via the kinetic master equation, toward examining changes in singlet exciton yields. The results show increases in the singlet exciton generation when the recombination within triplet PP is slower than within singlet PP and when that recombination occurs at a rate comparable to or slower than the hyperfine coupling-induced spin conversion. Additionally, correlation analyses demonstrate that although electronic coupling predominantly determines charge recombination rates, the energy barriers still contribute significantly, manifesting the need of considering both coupling and energy barriers during charge recombination processes in PPs.

In the present study, we have computed the heat of formation (HOF) for over 500 C-, H-, N-, O-, F-, S-, Cl-, Br-containing molecules in the NIST Chemistry Webbook with a previously established methodology [from the highest- to lowest-level methods, W1X-2, CCSD(T)-F12b, DSD-PBEP86, and ωB97M-V, with the lower levels calibrated against higher levels for the atomic energies, see: J. Phys. Chem. A 2022, 126, 4981–4990]. We find a reasonable level of agreement between the computed and NIST values for the present set of species. However, the set of F-containing compounds shows considerably larger discrepancies, which can in part be attributed to dubious experimental values, as we have demonstrated in some cases. With our highest-level computed HOFs, we validated the lower-level methods used in our protocol. Specifically, CCSD(T)-F12b yields chemically accurate (±4.2 kJ mol^–1) values for all types of molecules, while DSD-PBEP86 and ωB97M-V yield similar levels of accuracy for most systems, with key exceptions being molecules with numerous electron-withdrawing F and NO₂ groups. Our results further support the use of the protocol for the computation of HOFs, particularly for systems with few reliable reference values.

In the present study, we have computed the heat of formation (HOF) for over 500 C-, H-, N-, O-, F-, S-, Cl-, Br-containing molecules in the NIST Chemistry Webbook with a previously established methodology [from the highest- to lowest-level methods, W1X-2, CCSD(T)-F12b, DSD-PBEP86, and ωB97M-V, with the lower levels calibrated against higher levels for the atomic energies, see: J. Phys. Chem. A 2022, 126, 4981-4990]. We find a reasonable level of agreement between the computed and NIST values for the present set of species. However, the set of F-containing compounds shows considerably larger discrepancies, which can in part be attributed to dubious experimental values, as we have demonstrated in some cases. With our highest-level computed HOFs, we validated the lower-level methods used in our protocol. Specifically, CCSD(T)-F12b yields chemically accurate (±4.2 kJ mol^-1) values for all types of molecules, while DSD-PBEP86 and ωB97M-V yield similar levels of accuracy for most systems, with key exceptions being molecules with numerous electron-withdrawing F and NO₂ groups. Our results further support the use of the protocol for the computation of HOFs, particularly for systems with few reliable reference values.

Mass spectrometry (MS) is a fundamental tool for chemical identification. The current in-silico prediction tools can handle broad instrument conditions, large molecular libraries or fragment structures only on a very limited level. In this work, we propose a dual-model machine learning strategy that can solve this problem by jointly a classification model for fragment identification and noise filtering, and a regression model for spectral prediction. With the help of attention mechanism, our method outperforms other algorithms in accuracy and efficiency, providing a deeper understanding of the molecular fragmentation behavior in mass spectra. Our method can facilitate the large-scale in-silico spectra calculations and the analysis of unknown molecular structures, which may promote wider applications for MS.

The self-energy operator of the ab initio Dyson quasiparticle equation generates orbital-relaxation and differential-correlation corrections to Koopmans predictions of electron binding energies. Among the most important corrections are terms that may be expressed as ring and ladder diagrams. Inclusions of such terms in all orders of the fluctuation potential constitute renormalizations. The ability of several renormalized self-energies to predict molecular ionization energies has been tested versus reliable computational and experimental standards. These results reveal the superior accuracy and efficiency of several new-generation electron-propagator methods. They also demonstrate the strengths and weaknesses of self-energies that include ring or ladder renormalizations only and of methods that allow interactions between these terms. Whereas a simplified ladder method produces useful results, its simplified ring counterpart is more computationally efficient, but less accurate. New-generation alternatives to both methods are more accurate and efficient. No adjustable parameters are included in the generation of reference orbitals or in the formulation of the self-energy approximations examined in this work.

In this contribution we present the first local density-fitted multicomponent density functional theory implementation and assess its use for the calculation of anharmonic zero-point energies. Four challenging cases of molecular aggregates are reviewed: deprotonated formic acid trimer, diphenyl ether-tert-butyl alcohol conformers, anisole/methanol and anisole/2-naphtol dimers. These are all cases where a mismatch between the low-temperature computationally predicted minimum and the experimentally determined structure was observed. Through the use of nuclear-electronic orbital energies in the thermodynamic correction, the correct energetic ordering is recovered. For the smallest system, we compare our results to vibrational perturbation theory anharmonically corrected zero-point energy, with perfect agreement for the lower-lying conformers. The performance of the newly developed code and the density fitting errors are also analyzed. Overall, the new implementation shows a very good scaling with system size and the density fitting approximations exhibit a negligible impact.