Vibrational Circular Dichroism Spectroscopy with a Classical Polarizable Force Field: Alanine in the Gas and Condensed Phases.

Polarizable force fields are an essential component for the chemically accurate modeling of complex molecular systems with a significant degree of fluxionality, beyond harmonic or perturbative approximations. In this contribution we examine the performance of such an approach for the vibrational spectroscopy of the alanine amino acid, in the gas and condensed phases, from the Fourier transform of appropriate time correlation functions generated along molecular dynamics (MD) trajectories. While the infrared (IR) spectrum only requires the electric dipole moment, the vibrational circular dichroism (VCD) spectrum further requires knowledge of the magnetic dipole moment, for which we provide relevant expressions to be used with polarizable force fields. The AMOEBA force field was employed here to model alanine in the neutral and zwitterionic isolated forms, solvated by water or nitrogen, and as a crystal. Within this framework, comparison of the electric and magnetic dipole moments to those obtained with nuclear velocity perturbation theory based on density-functional theory for the same MD trajectories are found to agree well with one another. The statistical convergence of the IR and VCD spectra is examined and found to be more demanding in the latter case. Comparisons with experimental frequencies are also provided for the condensed phases.

Ultrafast fragmentation of highly-excited doubly-ionized deoxyribose: role of the liquid water environment.

Ab initio molecular dynamics simulations are used to investigate the fragmentation dynamics following the double ionization of 2-deoxy-D-ribose (DR), a major component in the DNA chain. Different ionization scenarios are considered to provide a complete picture. First focusing on isolated DR2+, fragmentation patterns are determined for the ground electronic state, adding randomly distributed excitation energy to the nuclei. These patterns differ for the two isomers studied. To compare thermal and electronic excitation effects, Ehrenfest dynamics are also performed, allowing to remove the two electrons from selected molecular orbitals. Two intermediate-energy orbitals, localized on the carbon chain, were selected. The dissociation pattern corresponds to the most frequent pattern obtained when adding thermal excitation. On the contrary, targeting the four deepest orbitals, localized on the oxygen atoms, leads to selective ultrafast C-O and/or O-H bond dissociation. To probe the role of environment, a system consisting of a DR molecule embedded in liquid water is then studied. The two electrons are removed from either the DR or the water molecules directly linked to the sugar through hydrogen bonds. Although the dynamics onset is similar to that of isolated DR when removing the same deep orbitals localized on the sugar oxygen atoms, the subsequent fragmentation patterns differ. Sugar damage also occurs following the Coulomb explosion of neighboring H2O2+ molecules due to interaction with the emitted O or H atoms.

The trisulfur radical ion S3•- controls platinum transport by hydrothermal fluids.

Platinum group elements (PGE) are considered to be very poorly soluble in aqueous fluids in most natural hydrothermal-magmatic contexts and industrial processes. Here, we combined in situ X-ray absorption spectroscopy and solubility experiments with atomistic and thermodynamic simulations to demonstrate that the trisulfur radical ion S3•- forms very stable and soluble complexes with both PtII and PtIV in sulfur-bearing aqueous solution at elevated temperatures (∼300 °C). These Pt-bearing species enable (re)mobilization, transfer, and focused precipitation of platinum up to 10,000 times more efficiently than any other common inorganic ligand, such as hydroxide, chloride, sulfate, or sulfide. Our results imply a far more important contribution of sulfur-bearing hydrothermal fluids to PGE transfer and accumulation in the Earth's crust than believed previously. This discovery challenges traditional models of PGE economic concentration from silicate and sulfide melts and provides new possibilities for resource prospecting in hydrothermal shallow crust settings. The exceptionally high capacity of the S3•- ion to bind platinum may also offer new routes for PGE selective extraction from ore and hydrothermal synthesis of noble metal nanomaterials.

Quasi-classical simulations of resonance Raman spectra based on path integral linearization.

Based on a linearization approximation coupled with path integral formalism, we propose a method derived from the propagation of quasi-classical trajectories to simulate resonance Raman spectra. This method is based on ground state sampling followed by an ensemble of trajectories on the mean surface between the ground and excited states. The method was tested on three models and compared to a quantum mechanics solution based on a sum-over-states approach: harmonic and anharmonic oscillators and the HOCl molecule (hypochlorous acid). The method proposed is able to correctly characterize resonance Raman scattering and enhancement, including the description of overtones and combination bands. The absorption spectrum is obtained at the same time, and the vibrational fine structure can be reproduced for long excited state relaxation times. The method can also be applied to dissociating excited states (as is the case for HOCl).

Influence of the environment on the infrared spectrum of alanine: An effective mode analysis.

The vibrational spectrum of the alanine amino acid was computationally determined in the infrared range 1000-2000 cm-1, under various environments encompassing the gas, hydrated, and crystalline phases, by means of classical molecular dynamics trajectories, carried out with the Atomic Multipole Optimized Energetics for Biomolecular Simulation polarizable force field. An effective mode analysis was performed, in which the spectra are optimally decomposed into different absorption bands arising from well-defined internal modes. In the gas phase, this analysis allows us to unravel the significant differences between the spectra obtained for the neutral and zwitterionic forms of alanine. In condensed phases, the method provides invaluable insight into the molecular origins of the vibrational bands and further shows that peaks with similar positions can be traced to rather different molecular motions.

How Crystal Symmetry Dictates Non-Local Vibrational Circular Dichroism in the Solid State.

Solid-State Vibrational Circular Dichroism (VCD) can be used to determine the absolute structure of chiral crystals, but its interpretation remains a challenge in modern spectroscopy. In this work, we investigate the effect of a twofold screw axis on the solid-state VCD spectrum in a combined experimental and theoretical analysis of P21 crystals of (S)-(+)-1-indanol. Even though the space group is achiral, a single proper symmetry operation has an important impact on the VCD spectrum, which reflects the supramolecular chirality of the crystal. Distinguishing between contributions originating from molecular chirality and from chiral crystal packing, we find that while IR absorption hardly depends on the symmetry of the space group, the situation is different for VCD, where completely new non-local patterns emerge. Understanding the two underlying mechanisms, namely gauge transport and direct coupling, will help to use VCD to distinguish polymorphic forms.

Machine Learning Yield Prediction from NiCOlit, a Small-Size Literature Data Set of Nickel Catalyzed C-O Couplings.

Synthetic yield prediction using machine learning is intensively studied. Previous work has focused on two categories of data sets: high-throughput experimentation data, as an ideal case study, and data sets extracted from proprietary databases, which are known to have a strong reporting bias toward high yields. However, predicting yields using published reaction data remains elusive. To fill the gap, we built a data set on nickel-catalyzed cross-couplings extracted from organic reaction publications, including scope and optimization information. We demonstrate the importance of including optimization data as a source of failed experiments and emphasize how publication constraints shape the exploration of the chemical space by the synthetic community. While machine learning models still fail to perform out-of-sample predictions, this work shows that adding chemical knowledge enables fair predictions in a low-data regime. Eventually, we hope that this unique public database will foster further improvements of machine learning methods for reaction yield prediction in a more realistic context.

Why local and non-local terms are essential for second harmonic generation simulation?

Second Harmonic Generation (SHG) today represents one of the most powerful techniques to selectively probe all types of interfaces. However, the origin of the SHG signal at a molecular level is still debated since the local dipole contribution, which is strongly correlated to the molecular orientation can be counterbalanced by non-local quadrupole contributions. Here, we propose a method to simulate the SHG signal of a model water/air interface from the molecular response of each contribution. This method includes both local and non-local terms, which are represented, respectively, by the dependency of the polarisability and hyperpolarisability upon the chemical environment of the molecule and by the bulk quadrupole response. The importance of both terms for the sound simulation of the SHG signals and their interpretation is assessed. We demonstrate that the sole dipole term is unable to simulate a SHG signal, even if the dependency of the hyperpolarisability on the local environment is considered. The inclusion of the bulk quadrupole contribution, which largely dominates the dipole contribution, is essential to predict the SHG response, although the accuracy of the prediction is increased when the dependency upon the local environment is considered.

The important role of non-covalent interactions for the vibrational circular dichroism of lactic acid in aqueous solution.

We present a computational study of vibrational circular dichroism (VCD) in solutions of (S)-lactic acid, relying on ab initio molecular dynamics (AIMD) and full solvation with bulk water. We discuss the effect of the hydrogen bond network on the aggregation behaviour of the acid: while aggregates of the solute represent conditions encountered in a weakly interacting solvent, the presence of water drastically interferes with the clusters - more strongly than originally anticipated. For both scenarios we computed the VCD spectra by means of nuclear velocity perturbation theory (NVPT). The comparison with experimental data allows us to establish a VCD-structure relationship that includes the solvent network around the chiral solute. We suggest that fundamental modes with strong polarisation such as the carbonyl stretching vibration can borrow VCD from the chirally restructured solvent cage, which extends the common explanatory models of VCD generation in aqueous solution.

Spontaneous liquid water dissociation on hybridised boron nitride and graphene atomic layers from ab initio molecular dynamics simulations.

Two-dimensional materials such as graphene (G) and hexagonal boron nitride (BN) have demonstrated potential applications in membrane science and in particular for the harvesting of blue energy. Although pure G and BN atomic layers are known to remain inert towards neutral water, one may wonder about the aqueous reactivity of hybridized monolayers formed by joining BN and G sheets in a planar fashion. Here, we perform ab initio molecular dynamics calculations of liquid water in contact with all possible planar heterostructures. Remarkably, we could observe the spontaneous chemisorption and dissociation of the interfacial water molecule into its self-ions at one specific and non-standard one-dimensional border. Our simulations predict that this type of heterostructure is prone to ionize liquid water in the absence of any electrical gating.

Adiabatic motion and statistical mechanics via mass-zero constrained dynamics.

In recent work [Coretti et al., J. Chem. Phys., 2018, 149, 191102], a new algorithm to solve numerically the dynamics of the shell model for polarization was presented. The approach, broadly applicable to systems involving adiabatically separated dynamical variables, employs constrained molecular dynamics to strictly enforce the condition that the force on the fast degrees of freedom, modeled as having zero mass, is null at each time step. The algorithm is symplectic and fully time reversible, and results in stable and efficient propagation. In this paper we complete the discussion of the mechanics of mass-zero constrained dynamics by showing how to adapt it to problems where the fast degrees of freedom must satisfy additional conditions. This extension includes, in particular, the important case of first principles molecular dynamics. We then consider the statistical mechanics of the mass-zero constrained dynamical system demonstrating that the marginal probability sampled by the dynamics in the physical phase space recovers the form of the Born-Oppenheimer probability density. The effectiveness of the approach and the favorable scaling of the algorithm with system size are illustrated in test calculations of solid Na via orbital-free density functional dynamics.

Assessing cluster models of solvation for the description of vibrational circular dichroism spectra: synergy between static and dynamic approaches.

Solvation effects are essential for defining the shape of vibrational circular dichroism (VCD) spectra. Several approaches have been proposed to include them into computational models for calculating VCD signals, in particular those resting on the "cluster-in-a-liquid" model. Here we examine the capabilities of this ansatz on the example of flexible (1S,2S)-trans-1-amino-2-indanol solvated in dimethyl sulfoxide (DMSO). We compare cluster sets obtained from static calculations with results from explicit molecular dynamics (MD) trajectories based on either force field (FF) or first-principles (FP) methods. While the FFMD approach provides a broader sampling of configurational space, FPMD and time-correlation functions of dipole moments account for anharmonicity and entropy effects in the VCD calculation. They provide a means to evaluate the immediate effect of the solvent on the spectrum. This survey singles out several challenges associated with the use of clusters to describe solvation effects in systems showing shallow potential energy surfaces and non-covalent interactions. Static structures of clusters involving a limited number of solvent molecules satisfactorily capture the main effects of solvation in the bulk limit on the VCD spectra, if these structures are correctly weighted. The importance of taking into consideration their fluxionality, i.e. different solvent conformations sharing a same hydrogen bond pattern, and the limitations of small clusters for describing the solvent dynamics are discussed.

Atomistic simulations of molten carbonates: Thermodynamic and transport properties of the Li2CO3-Na2CO3-K2CO3 system.

Although molten carbonates only represent, at most, a very minor phase in the Earth's mantle, they are thought to be implied in anomalous high-conductivity zones in its upper part (70-350 km). Besides, the high electrical conductivity of these molten salts is also exploitable in fuel cells. Here, we report quantitative calculations of their properties, over a large range of thermodynamic conditions and chemical compositions, which are a requisite to develop technological devices and to provide a better understanding of a number of geochemical processes. To model molten carbonates by atomistic simulations, we have developed an optimized classical force field based on experimental data of the literature and on the liquid structure issued from ab initio molecular dynamics simulations performed by ourselves. In implementing this force field into a molecular dynamics simulation code, we have evaluated the thermodynamics (equation of state and surface tension), the microscopic liquid structure and the transport properties (diffusion coefficients, electrical conductivity, and viscosity) of molten alkali carbonates (Li2CO3, Na2CO3, K2CO3, and some of their binary and ternary mixtures) from the melting point up to the thermodynamic conditions prevailing in the Earth's upper mantle (∼1100-2100 K, 0-15 GPa). Our results are in very good agreement with the data available in the literature. To our knowledge, a reliable molecular model for molten alkali carbonates covering such a large domain of thermodynamic conditions, chemical compositions, and physicochemical properties has never been published yet.

The MgCO3-CaCO3-Li2CO3-Na2CO3-K2CO3 melts: Thermodynamics and transport properties by atomistic simulations.

Atomistic simulations provide a meaningful way to determine the physicochemical properties of liquids in a consistent theoretical framework. This approach takes on a particular usefulness for the study of molten carbonates, in a context where thermodynamic and transport data are crucially needed over a large domain of temperatures and pressures (to ascertain the role of these melts in geochemical processes) but are very scarce in the literature, especially for the calcomagnesian compositions prevailing in the Earth's mantle. Following our work on Li2CO3-Na2CO3-K2CO3 melts, we extend our force field to incorporate Ca and Mg components. The empirical interaction potentials are benchmarked on the density data available in the experimental literature [for the crystals and the K2Ca(CO3)2 melt] and on the liquid structure issued from ab initio molecular dynamics simulations. Molecular dynamics simulations are then performed to study the thermodynamics, the microscopic structure, the diffusion coefficients, the electrical conductivity, and the viscosity of molten Ca,Mg-bearing carbonates up to 2073 K and 15 GPa. Additionally, the equation of state of a Na-Ca-K mixture representative of the lavas emitted at Ol Doinyo Lengai (Tanzania) is evaluated. The overall agreement between the MD results and the existing experimental data is very satisfactory and provides evidence for the ability of the force field to accurately model any MgCO3-CaCO3-Li2CO3-Na2CO3-K2CO3 melt over a large T-P range. Moreover, it is the first report of a force field allowing us to study the transport properties of molten magnesite (MgCO3) and molten dolomite [CaMg(CO3)2].

Effect of puckering motion and hydrogen bond formation on the vibrational circular dichroism spectrum of a flexible molecule: the case of (S)-1-indanol.

The influence of flexibility and hydrogen bond formation on the IR absorption and vibrational circular dichroism (VCD) spectrum of a floppy protic molecule, namely, (S)-1-indanol, is studied in both non-polar CCl4 and polar DMSO solvents. The experimental IR absorption and VCD spectra obtained by Fourier transform spectroscopy are interpreted using both static density functional theory (DFT) calculations and first principles molecular dynamics (FPMD) within DFT, using the nuclear velocity perturbation theory (NVPT). Simulation of the spectra based on static optimised geometries is not sufficient in CCl4 and going beyond static calculations is mandatory for satisfactorily reproducing the VCD spectra. The FPMD results obtained in DMSO indicate that (S)-1-indanol is hydrogen-bonded to one DMSO molecule. As a result, static "cluster-in-the-bulk" DFT calculations in which the solute-solvent interaction is modeled as the most stable (S)-1-indanol:DMSO complexes in a DMSO continuum yield satisfactory agreement with the experiment. Correspondence between experimental and simulated spectra is slightly improved when the VCD spectrum is calculated as the summed contributions of snapshots extracted from FPMD trajectories, due to better sampling of the potential-energy surface. Finally, NVPT calculations further improve the description of experimental spectra by taking into account higher-energy structures, which are not necessary local minima.

Proton Collision on Deoxyribose Originating from Doubly Ionized Water Molecule Dissociation.

In this work, we studied the fragmentation dynamics of 2-deoxy-d-ribose (DR) in solution that arises from the double ionization of a water molecule in its primary hydration shell. This process was modeled in the framework of ab initio molecular dynamics. The charge unbalanced in the solvent molecules produces a Coulomb explosion with the consequent release of protons with kinetic energy in the few electronvolts range, which collide with the surrounding molecules in solution inducing further chemical reactions. In particular, we observe proton collisions with the solute molecule DR, which leads to a complete ring opening. In DNA, damage to the DR moiety may lead to DNA strand breaking. This mechanism can be understood as one of the possible steps in the radiation-induced fragmentation of DNA chains.

Sulfur radical species form gold deposits on Earth.

Current models of the formation and distribution of gold deposits on Earth are based on the long-standing paradigm that hydrogen sulfide and chloride are the ligands responsible for gold mobilization and precipitation by fluids across the lithosphere. Here we challenge this view by demonstrating, using in situ X-ray absorption spectroscopy and solubility measurements, coupled with molecular dynamics and thermodynamic simulations, that sulfur radical species, such as the trisulfur ion S3(-), form very stable and soluble complexes with Au(+) in aqueous solution at elevated temperatures (>250 °C) and pressures (>100 bar). These species enable extraction, transport, and focused precipitation of gold by sulfur-rich fluids 10-100 times more efficiently than sulfide and chloride only. As a result, S3(-) exerts an important control on the source, concentration, and distribution of gold in its major economic deposits from magmatic, hydrothermal, and metamorphic settings. The growth and decay of S3(-) during the fluid generation and evolution is one of the key factors that determine the fate of gold in the lithosphere.

Chiral Crystal Packing Induces Enhancement of Vibrational Circular Dichroism.

We demonstrate that molecular vibrations with originally low or zero intensity in a vibrational circular dichroism (VCD) spectrum attain chirality in molecular crystals by coordinated motion of the atoms. Ab initio molecular dynamics simulations of anharmonic solid-state VCD spectra of l-alanine crystals reveal how coherent vibrational modes exploit the space group's chirality, leading to non-local, enhanced VCD features, most significantly in the carbonyl region of the spectrum. The VCD-enhanced signal is ascribed to a helical arrangement of the oscillators in the crystal layers. No structural irregularities need to be considered to explain the amplification, but a crucial point lies in the polarization of charge, which requires an accurate description of the electronic structure. Delivering a quantitative atomic conception of supramolecular chirality induction, our ab initio scheme is applicable well beyond molecular crystals, for example, to address VCD in proteins and related compounds.

Ca2+ -Cl- Association in Water Revisited: the Role of Cation Hydration.

We investigate the dissociation of a Ca2+ -Cl- pair in water using classical molecular dynamics simulations with a polarizable interaction potential, parameterized from ab initio calculations. By computing the potential of mean force as a function not only of the interionic distance but also of the coordination numbers by water molecules, we show that it is necessary to use a collective variable describing the cation hydration in order to capture the dissociation mechanism. In the contact ion pair, the Ca2+ cation has a first coordination sphere containing 5 or 6 water molecules. The minimum free-energy path for dissociation involves a two-step process: First one or two additional water molecules enter the cation coordination shell, increasing the coordination number up to 7 with an almost fixed interionic distance. Then the dissociation of the ionic pair occurs at this fixed coordination number.

Microscopic flow around a diffusing particle.

We report here on the computation of the microscopic flow induced by the motion of a small tagged particle in a fluid from molecular dynamic simulations. It is found that the hydrodynamical Stokes solution with slip boundary conditions is recovered at only a few diameters away from the tagged particle. However, fluctuations of the diffusing particle itself induce a renormalization of the bath viscosity and, more strikingly, an apparent violation of the non-penetrability of the particles in the laboratory frame. The expected zero normal velocity at contact is satisfied only in the particle frame, or for heavy particles. Further evidence of this generalized boundary condition is given by the evaluation of the flow in a granular gas using data from particle tracking experiments.