We present ab initio simulations of a large system of 2400 particles of molten NaCl to investigate the behavior of collective mode dispersion beyond the hydrodynamic regime. In particular, we aim to explain the unusually strong increase in the apparent speed of sound with wave number, which significantly exceeds the typical positive sound dispersion of 10%-25% observed in simple liquids. We compare dispersions of "bare" acoustic and optic modes in NaCl with ab initio simulations of other ionic melts such as CuCl and LiBr, metallic liquid alloys such as Pb44Bi56 and Li4Tl, and the regular Lennard-Jones KrAr liquid simulated by classical molecular dynamics. Analytical expressions for the "bare" acoustic and optic branches of collective excitations help us to identify the impact of the high-frequency optic branch on the emergence of "fast sound" in binary melts. Our findings show that in ionic melts, the high-frequency speed of sound is much larger than in the simple Lennard-Jones liquids and metallic melts, leading to an observed strong viscoelastic increase in the apparent speed of sound-more than double its adiabatic value.

The influence of oxidation state and crystalline structure on the dissolution mechanisms of both pure iron oxides and combusted iron particles in aqueous oxalic acid (0.5 mol/l) at 60 °C was systematically investigated. Dissolution experiments were carried out in a temperature-controlled, continuous-flow capillary reactor, allowing for the removal of reaction products and thereby suppressing the autocatalytic reaction mechanism. The non-reductive dissolution of α-Fe2O3 was observed through in situ x-ray absorption measurements. In contrast, the dissolution of spinel-type oxides such as γ-Fe2O3 and Fe3O4 proceeded reductively, indicated by gradual changes in characteristic spectral features. Given that γ-Fe2O3 and Fe3O4 share a similar crystal structure but differ in the nominal oxidation state, this implies that the phase composition is decisive for the reductive dissolution. For mixed-phase particles consisting of spinel and rhombohedral phases (maghemite and hematite), the preferential dissolution of the spinel phase was observed. Despite the similar bulk composition of spinel and rhombohedral phases in the combusted iron particles (as confirmed by Mössbauer spectroscopy and x-ray diffraction analysis), dissolution predominantly follows a non-reductive pathway, with no preferential dissolution of the γ-phase. This unique dissolution behavior of combusted iron particles arises from their layered microstructure.

Machine learning potentials (MLPs) are promising for various chemical systems, but their complexity and lack of physical interpretability challenge their broad applicability. This study evaluates the transferability of the deep potential (DP) and neural equivariant interatomic potential (NequIP) models for graphene-water systems using numerical metrics and physical characteristics. We found that the data quality from density functional theory calculations significantly influences MLP predictive accuracy. Prediction errors in transferring systems reveal the particularities of quantum chemical calculations on the heterogeneous graphene-water systems. Even for supercells with non-planar graphene carbon atoms, k-point mesh is necessary to obtain accurate results. In contrast, gamma-point calculations are sufficiently accurate for water molecules. In addition, we performed molecular dynamics (MD) simulations using these two models and compared the physical features such as atomic density profiles, radial distribution functions, and self-diffusion coefficients. It was found that although the NequIP model has higher accuracy than the DP model, the differences in the above physical features between them were not significant. Considering the stochasticity and complexity inherent in simulations, as well as the statistical averaging of physical characteristics, this motivates us to explore the meaning of accurately predicting atomic force in aligning the physical characteristics evolved by MD simulations with the actual physical features.

We theoretically study the evaporation-driven phase separation of a binary fluid mixture in a thin film deposited on a moving substrate, as occurs in meniscus-guided deposition for solution-processed materials. Our focus is on the limit of rapid substrate motion where phase separation takes place far away from the coating device. In this limit, demixing takes place under conditions mimicking those in a stationary film because substrate and film move at the same speed. We account for the hydrodynamic transport of the mixture within the lubrication approximation. In the early stages of demixing, diffusive and evaporative mass transport predominates, consistent with earlier studies on evaporation-driven spinodal decomposition. In the late-stage coarsening of the demixing process, the interplay of solvent evaporation, diffusive, and hydrodynamic mass transport results in several distinct coarsening mechanisms. The effective coarsening rate is dictated by the dominant mass transport mechanism and therefore depends on the material properties, evaporation rate, and time: slow solvent evaporation results in initially diffusive coarsening that for sufficiently strong hydrodynamic transport transitions to hydrodynamic coarsening, whereas rapid solvent evaporation can preempt and suppress hydrodynamic and diffusive coarsening. We identify a novel hydrodynamic coarsening regime for off-critical mixtures, arising from the interaction of the interfaces between solute-rich and solute-poor regions in the film with the solution-gas interface. This interaction induces a directional motion of solute-rich droplets along gradients in the film thickness, from regions where the film is relatively thick to where it is thinner. The solute-rich domains subsequently accumulate and coalesce in the thinner regions.

Recent experiments indicated that nanoparticles (NPs) might efficiently catalyze multiple chemical reactions, frequently exhibiting new phenomena. One of those surprising observations is intra-particle catalytic cooperativity, when the reactions at one active site can stimulate the reactions at spatially distant sites. Theoretical explanations of these phenomena have been presented, pointing out the important role of charged hole dynamics. However, the crucial feature of nanoparticles that can undergo dynamic structural surface rearrangements, potentially affecting the catalytic properties, has not yet been accounted for. We present a theoretical study of the effect of dynamic restructuring in NPs on intra-particle catalytic cooperativity. It is done by extending the original static discrete-state stochastic framework that quantitatively evaluates the catalytic communications. The dynamic restructuring is modeled as stochastic transitions between states with different dynamic properties of charged holes. Our analysis reveals that the communication times always decrease with increasing rates of dynamic restructuring, while the communication lengths exhibit a dynamic behavior that depends on how dynamic fluctuations affect migration and death rates of charged holes. Computer simulations fully support theoretical predictions. These findings provide important insights into the microscopic mechanisms of catalysis on single NPs, suggesting specific routes to rationally design more efficient catalytic systems.

A monolayer of 4-nitrothiophenol adsorbed on an Au substrate was heated by illuminating the substrate with a 19 ps laser pulse of 532 nm wavelength. Within 91 ps, the temperature of the sample increased from room temperature by 113 K. Vibrational sum frequency spectroscopy was used to characterize the adsorption geometry of the molecules in the ordered domains in the monolayer film. Upon heating, the initially ordered monolayer largely lost its structure. While the molecules are initially tilted by about 50° with respect to the surface normal, the analysis indicates that the mean tilt angle increased to 80° with a spread for individual molecules of up to a tilt angle of 40° upon heating. The evolution of this loss of order lagged about 100 ps behind the temperature rise of the substrate.

We present a Lagrangian-based implementation of Ehrenfest dynamics with nuclear-electronic orbital (NEO) theory and real-time time-dependent density functional theory for extended periodic systems. In addition to a quantum dynamical treatment of electrons and selected protons, this approach allows for the classical movement of all other nuclei to be taken into account in simulations of condensed matter systems. Furthermore, we introduce a Lagrangian formulation for the traveling proton basis approach and propose new schemes to enhance its application for extended periodic systems. Validation and proof-of-principle applications are performed on electronically excited proton transfer in the o-hydroxybenzaldehyde molecule with explicit solvating water molecules. These simulations demonstrate the importance of solvation dynamics and a quantum treatment of transferring protons. This work broadens the applicability of the NEO Ehrenfest dynamics approach for studying complex heterogeneous systems in the condensed phase.

Two machine-learning-aided thermodynamic integration schemes to compute the chemical potentials of atoms and molecules have been developed and compared. One is the particle insertion method, and the other combines particle insertion with element substitution. In the former method, the species is gradually inserted into the liquid and its chemical potential is computed. In the latter method, after the particle insertion, the inserted species is substituted with another species, and the chemical potential of this new species is computed. In both methods, the thermodynamic integrations are conducted using machine-learned potentials trained on first-principles datasets. The errors of the machine-learned surrogate models are further corrected by performing thermodynamic integrations from the machine-learned potentials to the first-principles potentials, accurately providing the first-principles chemical potentials. These two methods are applied to compute the real potentials of proton, alkali metal cations, and halide anions in water. The applications indicate that these two entirely different thermodynamic pathways yield identical real potentials within statistical error bars, demonstrating that both methods provide reproducible real potentials. The computed real potentials and solvation structures are also in good agreement with past experiments and simulations. These results indicate that machine-learning surrogate models enabling particle insertion and element substitution provide a precise method for determining the chemical potentials of atoms and molecules.

The water/ethanol system possesses complexities at the molecular level, which render its description a difficult task. For the elucidation of the system's hydrogen bonding features that are the key factors in its complex behavior, we conduct a Density Functional Theory analysis on relevant water/ethanol clusters inside implicit solvent cavities for the determination of the ethanol donor hydrogen bond strength. We record Attenuated Total Reflectance spectra of water/ethanol-OD solutions and utilize our density and refractive index measurements for post-processing. The application of the Badger-Bauer rule reveals a minimum in the strength of the ethanol donor hydrogen bond for a composition of xwater = 0.74. We attempt to analyze further this result by estimating the effect of the implicit solvent on the ethanol donor hydrogen bond strength, finding it to be incremental. A brief analysis of different cluster conformations is carried out to determine the cooperativity conditions that can potentially explain the observed minimum in the ethanol donor hydrogen bond strength. These observations are related to notions of microheterogeneity in water/alcohol mixtures and provide context toward a more elaborate picture of association in heteroclusters.

The structures and chemical bond evolution of ditantalum doped carbon clusters Ta2Cn-/0 (n = 1-7) were studied via size-selected anion photoelectron spectroscopy and theoretical calculations. It is found that Ta2C-/0 has a triangular structure and Ta2C2-/0 has a quasi-rhombus structure with C2v symmetry. Ta2C3- has a quasi-planar structure with a carbon atom and a C2 unit interacting with two tantalum atoms, and the lowest-energy isomer of neutral Ta2C3 has a triangular bipyramid structure with three carbon atoms around the Ta2 unit. Ta2C4-/0 has two C2 units connected with the Ta2 unit in parallel. Two isomers of Ta2C5- are observed, where both isomers have one carbon atom and two C2 units bound to the Ta2 unit in different ways. The most stable structure of neutral Ta2C5 has one carbon atom added on top of the Ta2C4 cluster. The most stable structures of Ta2C6-7-/0 can be viewed as a C2 unit and a C3 unit capping a butterfly like Ta2C4 structure, respectively. Molecular orbital analysis shows that neutral Ta2C3 has a large gap between its highest occupied molecular orbital and lowest unoccupied molecular orbital. Chemical bonding analysis reveals that the Ta-Ta interactions in Ta2Cn-/0 (n = 1-7) clusters are slightly weaker than the Ta-Ta interaction in bare Ta2 due to the participation in forming multicenter bonds.

In a recent paper, Di Lisio et al. [J. Chem. Phys. 159, 064505 (2023)] analyzed a series of temperature down-jumps using the single-parameter aging (SPA) ansatz combined with a specific assumption about density scaling in the out-of-equilibrium system and did not find a good prediction for the largest down-jumps. In this paper, we show that SPA in its original form does work for all their data, including large jumps of ΔT > 20 K. Furthermore, we discuss different approaches to the extension of the density scaling concept to out-of-equilibrium systems.

Carbohydrates constitute one of the key classes of biomacromolecules, yet vibrational spectroscopic studies involving carbohydrates remain scarce as spectra are highly congested and lack significant marker vibrations. Recently, we introduced and characterized a thiocyanate-labeled glucose [Gasse et al., J. Chem. Phys. 158, 145101 (2023)] demonstrating 2D-IR spectroscopy of carbohydrates using vibrational probes. Here, we build on that work and test azide groups as alternative for studies of carbohydrates to expand the available set of local probes. Many common carbohydrates with different azide labeling positions, such as galactose, glucose, or lactose, are readily available due to their application in click chemistry and hence do not require additional complex synthesis strategies. In this work, we have characterized azide-labeled glucose,, galactose, acetylglucosamine and lactose in water using IR and 2D-IR spectroscopy to test their potential for future applications in studies of carbohydrate-protein interactions. Our findings indicate that their absorption profiles and vibrational dynamics are primarily determined by the labeling position on the ring. However, we also observe additional variations between samples with the same labeling position. Furthermore, we demonstrate that their usage remains feasible at biologically relevant concentrations, highlighting their potential to probe more complex biological processes, i.e., enzymatic catalysis.

A four-dimensional potential energy surface (4D-PES) has been constructed for the N2-OCS complex. The PES is achieved by applying the explicitly correlated coupled cluster method, which incorporates single, double, and perturbative triple excitations [CCSD(T)-F12a], along with the augmented correlation consistent triple zeta (aug-cc-pVTZ) basis set. The rovibrational levels are precisely determined and assigned through bound state calculations and wavefunction analysis. The calculated transition frequencies reproduce the experimental observations accurately, achieving an RMSE of 0.0005 cm-1 for the 23 rotational transitions (J ≤ 6, Ka ≤ 2). The R-φ contour plot of the wave function clearly demonstrates the unambiguous delocalization of the dihedral angle, and the averaged geometry of the ground vibrational state is determined to be non-planar with φ = 90°. To obtain a quantitative analysis of this phenomenon, we expanded the 3H-solution model [Guo et al., J. Quant. Spectrosc. Radiat. Transfer 309 (2023) 108711] from a three-dimensional system (Ar-AgF) to a nine-dimensional system (N2-OCS). Based on this model, the tunneling splitting was calculated to be 0.0822 cm-1, which excellently matches the experimental result of 0.0817 cm-1. The excellent agreement between the theoretical and experimental results suggests that the wavefunction delocalization and out-of-plane motion can be attributed to the tunneling effects in the ground vibrational state.

Two-molecule theory refers to a class of microscopic, self-consistent field theories for the radial distribution function in classical molecular liquids. The version examined here can be considered as one of the very few formally derived closures to the reference interaction site model (RISM) equation. The theory is applied to polyethylene liquids, computing their equilibrium structural and thermodynamic properties at melt densities. The equation for the radial distribution function, which is represented as an average over the accessible states of two molecules in an external field that mimics the effects of the other molecules in the liquid, is computed by Monte Carlo simulation along with the intramolecular structure function. An improved direct sampling algorithm is utilized to speed the equilibration. Polyethylene chains of 24 and 66 united atom CH2 units are studied. The results are compared to full, many-chain molecular dynamics (MD) simulations and self-consistent polymer-RISM (PRISM) theory with the atomic Percus-Yevick (PY) closure under the same conditions. It is shown that the two-molecule theory produces results that are close to those of MD and is thus able to overcome defects of PRISM-PY theory and predict more accurate liquid structure at both short and long ranges. Predictions for the equation of state are also discussed.

We report a potential energy surface (PES) development for the F- + SiH3I system to study its gas-phase reactions through quasi-classical dynamics simulations. The PES is represented by a full-dimensional permutationally invariant polynomial fitted to composite coupled cluster energy points obtained at the ManyHF-[CCSD-F12b + BCCD(T) - BCCD]/aug-cc-pVTZ(-PP) level of theory. The development was automated by Robosurfer, which samples the configurational space, manages ab initio calculations, and iteratively extends the fitting set. When selecting the ab initio method, we address two types of electronic structure calculation issues: first, the gold standard CCSD(T)-F12b is prone to occasional breakdown due to the perturbative (T) contribution, whereas CCSD-F12b + BCCD(T) - BCCD, with the Brueckner (T) term, is more robust; second, the underlying Hartree-Fock calculation may not always converge to the global minimum, resulting in highly erroneous energies. To mitigate this, we employed ManyHF, configuring the Hartree-Fock calculations with multiple initial guess orbitals and selecting the solution with the lowest energy. According to the simulations, the title system exhibits exceptionally high and diverse reactivity. We observe two dominant product formations: SN2 and proton abstraction. Moreover, SiH2F- + HI, SiHFI- + H2, SiH2FI + H-, SiH2 + FHI-, SiH2 + HF + I-, and SiHF + H2 + I- formations are found at lower probabilities. We differentiated inversion and retention for SN2, both being significant throughout the entire collision energy range. Opacity- and excitation functions are reported, and the details of the atomistic dynamics are visually examined via trajectory animations.

The development of dielectric capacitors toward high voltage and high power density requires materials with excellent insulation and energy storage performances. In this work, a polymer dielectric with polyetherimide (PEI) as the matrix and calcium barium zirconate titanate (BZCT) coated by barium titanate fiber (BT) as the filler (BZCT@BT) was constructed. The (0.5%-10% BZCT@BT/PEI) polymer dielectric has an excellent discharge energy density (Ue) of 6.66 J/cm3 and maintains an advanced charge/discharge efficiency (η) of 93.29% when the BT content was 0.5% and the BZCT particle content was 10%. The addition of BZCT endows the polymer dielectric with a higher relative dielectric constant (εr), while BT, maintaining a lower εr than BZCT, could reduce the electric field (E) distortion caused by the dielectric mismatch between PEI and BZCT. Oriented fiber fillers increase the breakdown strength of the polymer dielectric, ultimately increasing the performance of energy storage. A new strategy for the design of energy storage polymer dielectrics was provided by this work.

We examine the applicability of the numerically accurate method of time dependent variation with multiple Davydov Ansätze (mDA) to non-Hermitian systems. As illustrative examples, three systems of interest have been studied, a non-Hermitian system of dissipative Landau-Zener transitions, a non-Hermitian multimode Jaynes-Cummings model, and a dissipative Holstein-Tavis-Cummings model, all of which are shown to be effectively described by the mDA method. Our findings highlight the versatility of the mDA as a powerful numerical tool for investigating complex many-body non-Hermitian systems, which can be extended to explore diverse phenomena such as skin effects, excited-state dynamics, and spectral topology in the non-Hermitian field.

Simple and stable generation of monodispersed droplets with volume from picolitre to nanoliter is one of the key factors in high-throughput quantitative microreactors for chemical and biomedical applications. In this work, an efficient method that could realize simple manipulating microflow with a broad operation window for preparing monodispersed droplets with controllable diameter is developed. The microfluidic device is constructed by inserting a capillary with an oblique angle (α) into the continuous phase, named a floating capillary-based open microfluidic device (FCOMD). The transition of droplet-generating mode between dripping and jetting can be achieved by changing capillary number and α. A computational model based on the volume-of-fluid/continuum-surface-force method to explain the controllability of α on the droplet formation regime and droplet breakage, verifying the synergistic effect of ΔP and Fb, facilitates the droplet pinching. A descending order of Pn of capillary with different α is that 45° > 30° > 15° > 60° > 75°, leading to the same order of generated droplet's D. When compared with the traditional capillary co-flow device, the generating throughput of the integrated FCOMD obtained by integrating different numbers of capillaries is at least ten times. Moreover, water in oil, oil in water double-emulsion, colloidal dispersed droplets, and liquid crystal droplets with diameters ranging from 25 to 800 μm are prepared on-demand by the FCOMD, indicating the universality of the microfluidic device. Thus, the FCOMD shows the features of simplicity, practicability, and flexibility, offering valuable guidance for generating controllable droplets with wide size change and showing a great potential application in material science, foods, pharmaceuticals, and cosmetics.

Thermal fluctuations are ubiquitous in mesoscopic and microscopic systems. Take nematic liquid crystals (LCs) as an example; their director fluctuations can strongly scatter light and give rise to random motions and rotations of topological defects and solid inclusions. These stochastic processes contain important information about the material properties of the LC and dictate the transport of the immersed colloidal particles. However, modeling thermal fluctuations of the nematic field remains challenging. Here, we introduce a new Monte Carlo simulation method, namely the Fourier-space Monte Carlo (FSMC) method, which is based on the Oseen-Frank elastic distortion energy model. This method accurately models the thermal fluctuations of a nematic LC's director field. In contrast to the traditional real-space MC method, which perturbs the director locally, the FSMC method samples different eigenmodes of the director distortions in the Fourier space, aligning with the equipartition theorem. We apply FSMC to study defect fluctuations and trajectories in a two-dimensional nematic LC confined to various geometries. Our results show that FSMC can effectively sample degenerate defect configurations and reproduce long-range elastic interactions between defects. In addition, we conduct three-dimensional molecular dynamics simulations using a coarse-grained Gay-Berne potential, which corroborates the findings from FSMC. Taken together, we have developed a new Monte Carlo method to accurately model thermal fluctuations in nematic LCs, which can be useful for searching global free-energy minimum states in nematic, smectic, and other LC mesophases and can also be helpful in modeling the thermal motions of defects and inclusions in LCs.

Enzyme-catalyzed reactions involve two distinct timescales: a short timescale on which enzymes bind to substrate molecules to produce bound complexes and a comparatively long timescale on which the molecules of the complex are transformed into products. The uptake of the substrate in these reactions is the rate at which the product is made on the long timescale. Models often only consider the uptake to reduce the number of chemical species that need to be modeled and to avoid explicitly treating multiple timescales. Typically, the uptake rates cannot be described by mass action kinetics and are traditionally derived by applying singular perturbation theory to the system's governing differential equations. This analysis ignores short timescales by assuming that a pseudo-equilibrium between the enzyme and the enzyme-bound complex is maintained at all times. This assumption cannot be incorporated into current particle-based simulations of reaction-diffusion systems because they utilize proximity-based conditions to govern the instances of reactions that cannot maintain this pseudo-equilibrium for infinitely fast reactions. Instead, these methods must directly simulate the dynamics on the short timescale to accurately model the system. Due to the disparate timescales, such simulations require excessive amounts of computational time before the behavior on the long timescale can be observed. To resolve this problem, we use singular perturbation theory to develop a proximity-based reaction condition that enables us to ignore all fast reactions and directly reproduce non-mass action kinetics at long timescales. To demonstrate our approach, we implement simulations of a specific third order reaction with kinetics reminiscent of the prototypical Michaelis-Menten system.