Quantum mechanical/molecular mechanical trajectory surface hopping molecular dynamics simulation by spin-flip time-dependent density functional theory.

Analytic energy gradients of individual singlet and triplet states with respect to nuclear coordinates are derived and implemented for the collinear mixed-reference spin-flip time-dependent density functional theory (MRSF-TDDFT), which eliminates the problematic spin-contamination of SF-TDDFT. Dimensional-transformation matrices for the singlet and triplet response spaces are introduced, simplifying the subsequent derivations. These matrices enable the general forms of MRSF-TDDFT equations to be similar to those of SF-TDDFT, suggesting that the computational overhead of singlet or triplet states for MRSF-TDDFT is nearly identical to that of SF-TDDFT. In test calculations, the new MRSF-TDDFT yields quite different optimized structures and energies as compared to SF-TDDFT. These differences turned out to mainly come from the spin-contamination of SF-TDDFT, which are largely cured by MRSF-TDDFT. In addition, it was demonstrated that the clear separation of singlet states from triplets dramatically simplifies the location of minimum energy conical intersection. As a result, it is clear that the MRSF-TDDFT has advantages over SF-TDDFT in terms of both accuracy and practicality. Therefore, it can be a preferred method, which is readily applied to other "black-box" type applications, such as the minimum-energy optimization, reaction path following, and molecular dynamics simulations.

A combination of X-ray absorption spectroscopy (XAS) measurements and quantum mechanical/molecular mechanical (QM/MM) molecular dynamics (MD) simulations has been applied to elucidate detailed information on the hydration structures of Ca(2+) and Cl(-). The XAS spectra (extended X-ray absorption fine structure, EXAFS, and X-ray absorption near-edge structure, XANES) measured from aqueous CaCl(2) solution were analyzed and compared to those generated from snapshots of QM/MM MD simulations of Ca(2+) and Cl(-) in water. With regard to this scheme, the simulated QM/MM-EXAFS and QM/MM-XANES spectra, which correspond to the local structure and geometrical arrangement of the hydrated Ca(2+) and Cl(-) at molecular level show good agreement with the experimentally observed EXAFS and XANES spectra. From the analyses of the simulated QM/MM-EXAFS spectra, the hydration numbers for Ca(2+) and Cl(-) were found to be 7.1 +/- 0.7 and 5.1 +/- 1.3, respectively, compared to the corresponding values of 6.9 +/- 0.7 and 6.0 +/- 1.7 derived from the measured EXAFS data. In particular for XANES results, it is found that ensemble averages derived from the QM/MM MD simulations can provide reliable QM/MM-XANES spectra, which are strongly related to the shape of the experimental XANES spectra. Since there is no direct way to convert the measured XANES spectrum into details relating to geometrical arrangement of the hydrated ions, it is demonstrated that such a combined technique of XAS experiments and QM/MM MD simulations is well-suited for the structural verification of aqueous ionic solutions.

The photoinduced nonadiabatic decay dynamics of 9H-adenine (hereafter, adenine) in aqueous solution were investigated by surface-hopping simulations within a quantum mechanical/molecular mechanical (QM/MM) framework. The QM subsystem (adenine) was treated at the semiempirical OM2/MRCI level, whereas the MM solvent (water) was described by the TIP3P force field model. Classical molecular dynamics (MD) simulations were used to generate snapshots with different solvent configurations and geometries. For a representative number of these snapshots, the energy minima of the lowest electronic states and the most important conical intersections were located by QM/MM geometry optimization. Surface-hopping QM/MM MD simulations were performed for all selected snapshots to study the nonadiabatic dynamics after photoexcitation, including the two lowest excited singlet states, which are both populated in the initial photoexcitation due to strong vibronic coupling in the Franck-Condon region. The simulations yield ultrafast S(2)-S(1) decay within 40 fs and S(1)-S(0) internal conversion to the ground state within 410 fs, which is consistent with recent experimental results from time-resolved spectroscopy.

Using a combination of photoelectron spectroscopy measurements and quantum chemistry calculations, we have identified competing electron emission processes that contribute to the 350-315 nm photoelectron spectra of the deprotonated green fluorescent protein chromophore anion, p-hydroxybenzylidene-2,3-dimethylimidazolinone. As well as direct electron detachment from S0, we observe resonant excitation of the 2(1)ππ* state of the anion followed by autodetachment. The experimental photoelectron spectra are found to be significantly broader than photoelectron spectrum calculated using the Franck-Condon method and we attribute this to rapid (∼10 fs) vibrational decoherence, or intramolecular vibrational energy redistribution, within the neutral radical.

We present a series of new implementations that we recently introduced in COBRAMM, the open‐source academic software developed in our group. The goal of these implementations is to offer an automatized workflow and interface to simulate time‐resolved transient absorption (TA) spectra of medium‐to‐big chromophore embedded in a complex environment. Therefore, the excited states absorption and the stimulated emission are simulated along nonadiabatic dynamics performed with trajectory surface hopping. The possibility of treating systems from medium to big size is given by the use of time‐dependent density functional theory (TD‐DFT) and the presence of the environment is taken into account employing a hybrid quantum mechanics/molecular mechanics (QM/MM) scheme. The full implementation includes a series of auxiliary scripts to properly setup the QM/MM system, the calculation of the wavefunction overlap along the dynamics for the propagation, the evaluation of the transition dipole moment at linear response TD‐DFT level, and scripts to setup, run and analyze the TA from an ensemble of trajectories. Altogether, we believe that our implementation will open the door to the easily simulate the time‐resolved TA of systems so far computationally inaccessible.

The previously introduced workflow to achieve an energetically and structurally optimized description of frontier bonds in quantum mechanical/molecular mechanics (QM/MM)-type applications was extended into the regime of computational material sciences at the example of a layered carbon model systems. Optimized QM/MM link bond parameters at HSEsol/6-311G(d,p) and self-consistent density functional tight binding (SCC-DFTB) were derived for graphitic systems, enabling detailed investigation of specific structure motifs occurring in graphene-derived structures quantum-chemical calculations. Exemplary molecular dynamics (MD) simulations in the isochoric-isothermic (NVT) ensemble were carried out to study the intercalation of lithium and the properties of the Stone-Thrower-Wales defect. The diffusivity of lithium as well as hydrogen and proton adsorption on a defective graphene surface served as additional example. The results of the QM/MM MD simulations provide detailed insight into the applicability of the employed link-bond strategy when studying intercalation and adsorption properties of graphitic materials.

Computational studies of ultrafast photoinduced processes give valuable insights into the photochemical mechanisms of a broad range of compounds. In order to accurately reproduce, interpret, and predict experimental results, which are typically obtained in a condensed phase, it is indispensable to include the condensed phase environment in the computational model. However, most studies are still performed in vacuum due to the high computational cost of state-of-the-art non-adiabatic molecular dynamics (NAMD) simulations. The quantum mechanical/molecular mechanical (QM/MM) solvation method has been a popular model to perform photodynamics in the liquid phase. Nevertheless, the currently used QM/MM embedding techniques cannot sufficiently capture all solute-solvent interactions. In this Perspective, we will discuss the efficient ΔSCF electronic structure method and its applications with respect to the NAMD of solvated compounds, with a particular focus on explicit quantum mechanical solvation. As more research is required for this method to reach its full potential, some challenges and possible directions for future research are presented as well.

Molecular dynamics (MD) simulations were carried out to study the conformational rearrangement induced by deprotonation of the fluorescent chromophore in GFP, as well as the associated changes in the hydrogen-bonding network. For both the structures with either a neutral or an anionic chromophore, it was found that the beta-barrel was stable and rigid, and the conformation of the chromophore was consistent with the available x-ray structure. The conformational change in Thr203 due to deprotonation was also found to be consistent with the three-state isomerization model. Although GFP is highly fluorescent, denatured-GFP is nonfluorescent, indicating that the environment of the protein plays an important role in its fluorescence behavior. Our MD simulations, which explore the effect of the protein shell on the conformation of the chromophore, find the flexibility of the central chromophore to be significantly restricted due to the rigid nature of the protein shell. The hydrogen-bonding between the chromophore and neighboring residues was also shown to contribute to the chromophore rigidity. In addition to the MD studies, quantum mechanics/molecular mechanics (QM/MM) ONIOM calculations were carried out to investigate the effect of the beta-barrel on the internal rotation in the chromophore. Along with providing quantitative values for torsional rotation barriers about the bridging bond in the chromophore, the ONIOM calculations also validate our MD force field parameters.

We report comparisons between energy-based quantum mechanics/molecular mechanics (QM/MM) and buffered force-based QM/MM simulations in silica. Local quantities-such as density of states, charges, forces, and geometries-calculated with both QM/MM approaches are compared to the results of full QM simulations. We find the length scale over which forces computed using a finite QM region converge to reference values obtained in full quantum-mechanical calculations is ∼10 Å rather than the ∼5 Å previously reported for covalent materials such as silicon. Electrostatic embedding of the QM region in the surrounding classical point charges gives only a minor contribution to the force convergence. While the energy-based approach provides accurate results in geometry optimizations of point defects, we find that the removal of large force errors at the QM/MM boundary provided by the buffered force-based scheme is necessary for accurate constrained geometry optimizations where Si-O bonds are elongated and for finite-temperature molecular dynamics simulations of crack propagation. Moreover, the buffered approach allows for more flexibility, since special-purpose QM/MM coupling terms that link QM and MM atoms are not required and the region that is treated at the QM level can be adaptively redefined during the course of a dynamical simulation.

Aggregation-induced emission (AIE) spectra accompanied by excited state intramolecular proton transfer (ESIPT) for two triphenylamine salicylaldehyde derivatives (namely, TS and TS-OMe) are investigated by performing molecular spectral and dynamics simulations associated with the hybrid quantum mechanics/molecular mechanics (QM/MM) at the quantum level of the time-dependent density functional theory. The simulated emission spectral peaks and Stokes' shifts are in good agreement with the experimental results for both TS and TS-OMe. Furthermore, the AIE spectral mechanisms are well explained to be associated with the ESIPT processes for both TS and TS-OMe monomers in the aggregated crystal state, while the AIE spectra mechanism for the TS-OMe (TS) dimer is accompanied by intermolecular charge-transfer excitation process. Besides, the TS dimers also contributed to the AIE mechanisms in the crystal with the intermolecular charge-transfer from one monomer to another. In addition, the TS dimers are contributed to the AIE mechanisms in the crystal with the intermolecular charge-transfer from one monomer to another. On the other hand, simulated emission spectra for both the TS and TS-OMe monomers in acetonitrile solution are involved in mixed emission with and without the ESIPT process, as interpreted by nonadiabatic molecular dynamics simulation. It is also briefly addressed that the emission spectra in the solution are weak and enhanced in the crystal. The present study provides a great physical insight into the design of highly efficient AIE compounds.

Protein splicing is a post-translational process in which a biologically inactive protein is activated by the release of a segment denoted as an intein. The process involves four steps. In the third, the scission of the intein takes place after the cyclization of the last amino acid of the segment, an asparagine. Little is known about the chemical reaction necessary for this cyclization. Experiments demonstrate that two histidines (the penultimate amino acid of the intein, and a histidine located 10 amino acids upstream) are relevant in the cyclization of the asparagine. We have investigated the mechanism and determinants of reaction in the GyrA intein focusing on the requirements for asparagine activation for its cyclization. First, the influence that the protonation states of these two histidines have on the orientation of the asparagine side chain is investigated by means of molecular dynamics simulation. Molecular dynamics simulations using the CHARMM27 force field were carried out on the three possible protonation states for each of these two histidines. The results indicate that the only protonation state in which the conformation of the system is suitable for cyclization is when the penultimate histidine is fully protonated (positively charged), and the upstream histidine is in the His(ε) neutral tautomeric form. The free energy profile for the reaction in which the asparagine is activated by a proton transfer to the upstream histidine is presented, computed by hybrid quantum mechanics/molecular mechanics (QM/MM) umbrella sampling molecular dynamics at the SCCDFTB/CHARMM27 level of theory. The calculated free energy barrier for the reaction is 19.0 kcal mol(-1). B3LYP/6-31+G(d) QM/MM single-point calculations give a qualitatively a similar energy profile, although with somewhat higher energy barriers, in good agreement with the value derived from experiment of 25 kcal mol(-1) at 60 °C. QM/MM molecular dynamics simulations of the reactant, activated reactant and intermediate states highlight the importance of the Arg181-Val182-Asp183 segment in catalysing the reaction. Overall, the results indicate that nucleophilic activation of the asparagine for its cyclization by the upstream histidine acting as the base is a plausible mechanism for the C-terminal cleavage in protein splicing.

Structure and dynamics of the Li+ hydrates: A comparative study of conventional QM/MM and ONIOM-XS MD simulations

α-l-Arabinofuranosidase (EC 3.2.1.55) plays a crucial role in the degradation and processing of arabinoxylan and arabinoxylo-oligosaccharides (AXOS), making it a promising target for industrial biomass conversion and biotechnological applications. Despite its ability to cleave α(1→2), α(1→3), and α(1→5) glycosidic bonds, the origin of its preference for specific AXOS linkages remains unclear. Using quantum mechanics (QM) calculations and quantum mechanics/molecular mechanics (QM/MM) molecular dynamics (MD) simulations, we investigated the catalytic mechanism and selectivity of hydrolysis and transglycosylation in a GH51 α-l-arabinofuranosidase from Thermobacillus xylanilyticus. We assessed various synthetic substrates and mono- and disubstituted AXOS (pNP-Araf, A3, A2X, A3X, XA2X, XA3X and XA2+3X), predicting their binding modes and substrate specificity. Glu176 (acid/base) and Glu298 (nucleophile) facilitate AXOS hydrolysis via a two-step, double-displacement mechanism, with the second step (deglycosylation) being the rate-limiting step. Based on combined QM/MM MD and QM(ONIOM) studies, His240 is most likely protonated to maintain the positioning of the Glu176 residue within the Asp297-His240-Glu176 catalytic triad, thereby enabling efficient catalysis via an optimal electrostatic environment. Consistent with the experiment, the enzyme efficiently hydrolyzes pNP-Araf and A3, with free-energy barriers of 14.8-15.4 kcal mol-1 (compared to 17.9-29.3 kcal mol-1 for AXOS); it selectively cleaves Araf-α(1→3) over Araf-α(1→2) linkages of donor substrates and favors xylobiose as an acceptor over aryl α-d-xylopyranoside in transglycosylation due to lower reaction barriers and effective transition-state stabilization. Our study provides a theoretical foundation for the engineering of this GH51 enzyme for efficient hydrolysis and synthesis of valuable xylan-based products.

Quantum mechanics/molecular mechanics (QM/MM) molecular dynamics (MD) simulations have been developed to simulate molecular systems, where an explicit description of changes in the electronic structure is necessary. However, QM/MM MD simulations are computationally expensive compared to fully classical simulations as all valence electrons are treated explicitly and a self-consistent field (SCF) procedure is required. Recently, approaches have been proposed to replace the QM description with machine-learned (ML) models. However, condensed-phase systems pose a challenge for these approaches due to long-range interactions. Here, we establish a workflow, which incorporates the MM environment as an element type in a high-dimensional neural network potential (HDNNP). The fitted HDNNP describes the potential-energy surface of the QM particles with an electrostatic embedding scheme. Thus, the MM particles feel a force from the polarized QM particles. To achieve chemical accuracy, we find that even simple systems require models with a strong gradient regularization, a large number of data points, and a substantial number of parameters. To address this issue, we extend our approach to a Δ-learning scheme, where the ML model learns the difference between a reference method (density functional theory (DFT)) and a cheaper semiempirical method (density functional tight binding (DFTB)). We show that such a scheme reaches the accuracy of the DFT reference method while requiring significantly less parameters. Furthermore, the Δ-learning scheme is capable of correctly incorporating long-range interactions within a cutoff of 1.4 nm. It is validated by performing MD simulations of retinoic acid in water and the interaction between S-adenoslymethioniat and cytosine in water. The presented results indicate that Δ-learning is a promising approach for (QM)ML/MM MD simulations of condensed-phase systems.

The absolute intrinsic hydration free energy GH+,wat◦ of the proton, the surface electric potential jump χwat◦ upon entering bulk water, and the absolute redox potential VH+,wat◦ of the reference hydrogen electrode are cornerstone quantities for formulating single-ion thermodynamics on absolute scales. They can be easily calculated from each other but remain fundamentally elusive, i.e., they cannot be determined experimentally without invoking some extra-thermodynamic assumption (ETA). The Born model provides a natural framework to formulate such an assumption (Born ETA), as it automatically factors out the contribution of crossing the water surface from the hydration free energy. However, this model describes the short-range solvation inaccurately and relies on the choice of arbitrary ion-size parameters. In the present study, both shortcomings are alleviated by performing first-principle calculations of the hydration free energies of the sodium (Na+) and potassium (K+) ions. The calculations rely on thermodynamic integration based on quantum-mechanical molecular-mechanical (QM/MM) molecular dynamics (MD) simulations involving the ion and 2000 water molecules. The ion and its first hydration shell are described using a correlated ab initio method, namely resolution-of-identity second-order Møller-Plesset perturbation (RIMP2). The next hydration shells are described using the extended simple point charge water model (SPC/E). The hydration free energy is first calculated at the MM level and subsequently increased by a quantization term accounting for the transformation to a QM/MM description. It is also corrected for finite-size, approximate-electrostatics, and potential-summation errors, as well as standard-state definition. These computationally intensive simulations provide accurate first-principle estimates for GH+,wat◦, χwat◦, and VH+,wat◦, reported with statistical errors based on a confidence interval of 99%. The values obtained from the independent Na+ and K+ simulations are in excellent agreement. In particular, the difference between the two hydration free energies, which is not an elusive quantity, is 73.9 ± 5.4 kJ mol-1 (K+ minus Na+), to be compared with the experimental value of 71.7 ± 2.8 kJ mol-1. The calculated values of GH+,wat◦, χwat◦, and VH+,wat◦ (-1096.7 ± 6.1 kJ mol-1, 0.10 ± 0.10 V, and 4.32 ± 0.06 V, respectively, averaging over the two ions) are also in remarkable agreement with the values recommended by Reif and Hünenberger based on a thorough analysis of the experimental literature (-1100 ± 5 kJ mol-1, 0.13 ± 0.10 V, and 4.28 ± 0.13 V, respectively). The QM/MM MD simulations are also shown to provide an accurate description of the hydration structure, dynamics, and energetics.