Search

We have developed a multi-input E(n) equivariant graph convolution-based model designed for the prediction of chemical properties that result from the interaction of heterogeneous molecular structures. By incorporating spatial features and constraining the functions learned from these features to be equivariant to E(n) symmetries, the interactional-equivariant graph neural network (IEGNN) can efficiently learn from the 3D structure of multiple molecules. To verify the IEGNN’s capability to learn interactional properties, we tested the model’s performance on three molecular data sets, two of which are curated in this study and made publicly available for future interactional model benchmarking. To enable the loading of these data sets, an open-source data structure based on the PyTorch Geometric library for batch loading multigraph data points is also created. Finally, the IEGNN’s performance on a data set consisting of an unknown interactional relationship (the frictional properties resulting between monolayers with variable composition) is examined. The IEGNN model developed was found to have the lowest mean absolute percent error for the predicted tribological properties of four of the six data sets when compared to previous methods.

MoSDeF-GOMC is a python interface for the Monte Carlo software GOMC to the Molecular Simulation Design Framework (MoSDeF) ecosystem. MoSDeF-GOMC automates the process of generating initial coordinates, assigning force field parameters, and writing coordinate (PDB), connectivity (PSF), force field parameter, and simulation control files. The software lowers entry barriers for novice users while allowing advanced users to create complex workflows that encapsulate simulation setup, execution, and data analysis in a single script. All relevant simulation parameters are encoded within the workflow, ensuring reproducible simulations. MoSDeF-GOMC’s capabilities are illustrated through a number of examples, including prediction of the adsorption isotherm for CO2in IRMOF-1, free energies of hydration for neon and radon over a broad temperature range, and the vapor–liquid coexistence curve of a four-component surrogate for the jet fuel S-8. The MoSDeF-GOMC software is available on GitHub at https://github.com/GOMC-WSU/MoSDeF-GOMC.

Concentrated ionic solutions present a potential improvement for liquid electrolytes. However, their conductivity is limited by high viscosities, which can be attenuated via cosolvation. This study employs a series of experiments and molecular dynamics simulations to investigate how different cosolvents influence the local structure and charge transport in concentrated lithium bis(trifluoromethane-sulfonyl)imide (LiTFSI)/acetonitrile solutions. Regardless of whether the cosolvent’s dielectric constant is low (for toluene and dichloromethane), moderate (acetone), or high (methanol and water), they preserve the structural and dynamical features of the cosolvent-free precursor. However, the dissimilar effects of each case must be individually interpreted. Toluene and dichloromethane reduce the conductivity by narrowing the distribution of Li+–TFSI–interactions and increasing the activation energies for ionic motions. Methanol and water broaden the distributions of Li+–TFSI–interactions, replace acetonitrile in the Li+solvation, and favor short-range Li+–Li+interactions. Still, these cosolvents strongly interact with TFSI–, leading to conductivities lower than that predicted by the Nernst–Einstein relation. Finally, acetone preserves the ion–ion interactions from the cosolvent-free solution but forms large solvation complexes by joining acetonitrile in the Li+solvation. We demonstrate that cosolvation affects conductivity beyond simply changing viscosity and provide fairly unexplored molecular-scale perspectives regarding structure/transport phenomena relation in concentrated ionic solutions.

Concentrated solutions of Li salts in acetonitrile are promising alternative electrolytes for the next generation of Li batteries as they may exhibit superior electrochemical properties. However, the reduced mobility of the chemical species is a barrier yet to be overcome, and for this, we explore the utilization of acetone as a cosolvent. Although acetone is a polar compound, we find that its addition to the LiTFSI/acetonitrile solution does not follow the trends expected for a simple dilution process. At a low concentration, acetone subtly shifts acetonitrile from the first to extended solvation sheaths of the ions. Still, most of the original structure of the solution is preserved, and mobile high-concentration clusters are formed in the solution. At higher concentrations, the cosolvation promotes cation–anion interactions but with a different nature from those in the original solution and still allows for a further increase in conductivity. Additionally, the non-coordinating fraction of acetonitrile acquires features resembling the pure solvent, which is a possible additional facilitating factor for ionic diffusion.

We present molecular-simulation-based calculations of the Van Hove correlation function (VHF) of water using multiple modeling approaches: classical molecular dynamics with simple three-site nonpolarizable models, with a polarizable model, and with a reactive force field; density functional tight-binding molecular dynamics; and ab initiomolecular dynamics. Due to the many orders of magnitude difference in the computational cost of these approaches, we investigate how small and short the simulations can be while still yielding sufficiently accurate and interpretable results for the VHF. We investigate the accuracy of the different models by comparing them to recently published inelastic X-ray scattering measurements of the VHF. We find that all of the models exhibit qualitative agreement with the experiments, and in some models and for some properties, the agreement is quantitative. This work lays the foundation for future simulation approaches to calculating the VHF for aqueous solutions in bulk and under nanoconfinement.

Solvent-in-salt electrolytes (SISEs) are a promising alternative to the electrolytes currently used in commercial devices. Despite the SISEs’ advantages, their utilization is not yet realized due to the poor mobility of their chemical species. We explore this problem by adding chloroform to a SISE formed by acetonitrile and a Li-salt. First, we performed illustrative cycling experiments to highlight the potential of this approach. Then, we focused on the description of the microscopic dynamics of the electrolytes and exposed the relevant aspects to be considered for their optimal performance. While the conductivity at low temperatures may be enhanced by the addition of chloroform, only subtle changes occur at room temperature. As revealed by molecular dynamics simulations and quasielastic neutron scattering (QENS) experiments, this effect is related to the preservation of the structure expected for a highly concentrated solution and promotion of the formation of ionic aggregates. These outcomes occur despite the increase in the overall mobility of the chemical species. The dynamics of the electrolytes in porous carbon was also investigated using QENS. In these circumstances, low concentrations of chloroform lead to diffusivities of the molecular species higher than those observed for the bulk electrolytes. As chloroform’s concentration increases, no further changes in the diffusivities are observed. Nonetheless, chloroform is mostly immobilized on the carbon surfaces and this behavior may be intensified at compositions closer to the eutectic mixture.

Most polar solvent molecules are unstable toward electrode materials used in Li-based batteries. Solid electrolytes and ionic liquids are far more stable; however, they have relatively low conductivity, and therefore electrical energy storage devices based on them would suffer from low power. Solvent-in-salt (SIS) systems combine chemical stability with relatively high conductivity. Here, we show how the nature of the employed anion affects the structure and dynamics of SIS systems. The transport of ions in lithium bis(fluorosulfonyl)imide (Li-FSI) systems was determined to be always faster than that in lithium bis(trifluoromethanesulfonyl)imide (Li-TFSI) systems. Moreover, we found that viscosity does not solely control conductivity and that the lower conductivity of TFSI–solutions is related to their stronger interaction with the solvent. This restricts solvent dynamics and slows down ion motions compared to that of FSI–. Interestingly, the TFSI–solvent interaction also leads to better charge separation (weaker ion–ion correlations) and a higher transference number for Li. Our results suggest that the ability to tune the solvent network formed around the anions may further improve electrolyte conductivity and Li transference number for safer and more efficient energy storage devices.

We demonstrate how the recently developed Python-based Molecular Simulation and Design Framework (MoSDeF) can be used to perform molecular dynamics screening of functionalized monolayer films, focusing on tribological effectiveness. MoSDeF is an open-source package that allows for the programmatic construction and parametrization of soft matter systems and enables TRUE (transferable, reproducible, usable by others, and extensible) simulations. The MoSDeF-enabled screening identifies several film chemistries that simultaneously show low coefficients of friction and adhesion. We additionally develop a Python library that utilizes the RDKit cheminformatics library and the scikit-learn machine learning library that allows for the development of predictive models for the tribology of functionalized monolayer films and use this model to extract information on terminal group characteristics that most influence tribology, based on the screening data.

Room-temperature ionic liquids (RTILs) hold promise for applications in electric double layer capacitors (EDLCs), owing to a much wider potential window, lower vapor pressure, and better thermal and chemical stabilities compared to conventional aqueous and organic electrolytes. However, because the low diffusivity of ions in neat RTILs negates the EDLCs’ advantage of high power density, the ionic liquids are often used in mixture with organic solvents. In this study, we measured the diffusivity of cations and anions in RTIL, 1-butyl-3-methylimidazolium bis(trifluoromethylsulfonyl) ([BMIM+][TFSI–]), mixed with 10 organic solvents, by using the pulsed-field gradient NMR method. The ion diffusivity was found to follow that of neat solvents and in most studied solvents showed an excellent agreement with the predicted values reported in the recent molecular dynamics (MD) study [Thompson, M. W.; J. Phys. Chem. B2019, 123, 1340−1347]. In two solvents consisting of long-chain molecules, however, the MD simulations predictions slightly underestimated the ionic diffusivities. The degree of ion dissociation was also estimated for each solvent by comparing the ionic conductivity with the molar conductivity derived from the diffusion measurements. The degree of ion dissociation and the hydrodynamic radius of ions suggest that the ions are coordinated by ∼1 solvent molecule. The scarcity of solvent–ion interactions explains the fact that the diffusivity of ions in the mixture significantly depends on the viscosity of the solvent.

The dissolution of room temperature ionic liquids (RTILs) in organic solvents has been shown to enhance ion dynamics. We previously used molecular dynamics (MD) simulations to study the ionic liquid ([BMIM+][Tf2N–]) in 22 unique solvents over a wide range of concentrations. By screening over a large parameter space, we reached several conclusions: (1) ion diffusivity increases monotonically as a function of increasing ionic liquid composition, (2) pure solvent diffusivity strongly correlates with ion diffusivity, and (3) conductivity predicted by the Nernst–Einstein (NE) equation has a maximum at intermediate compositions of ionic liquid. Building off this work, we now utilize the same parameter space to study the structure of ([BMIM+][Tf2N–]) solvated in organic solvents. We explore ion correlations through a number of structural and thermodynamic properties, including liquid densities, pair correlation functions, ion pairing and ion caging lifetimes, and free energy calculations. Through these analyses, we find that some solvents are much more effective at screening ion–ion interactions than others and that these differences impact the ion dynamics in these mixtures. In general, the strong pairing of ionic liquids negatively impacts transport properties, but some solvents can robustly screen these interactions, resulting in greatly enhanced ion dynamics. These results uncover trends connecting ionic liquid structure to transport, which can help in the design of new electrolytes for energy storage devices, such as electrical double layer capacitors and batteries.

We use molecular dynamics simulations to investigate the effects of solvent concentration on the bulk properties of an ion liquid electrolyte and the electrochemical performance on carbon-based electrodes, including pristine graphene, oxidized graphene, graphene armchair edge, graphene zigzag edge, onion-like carbon, and slit-pore carbon. We find that diluting the electrolyte reduces the number of ion pairs in the bulk and improves ion dynamics. The capacitance of the two-edge electrodes decreases monotonically as the solvent concentration increases, while the capacitance of other nonedge electrodes exhibits nonmonotonic behavior and a capacitance maximum is observed. Further analyses on the electric double layer reveals two competing factors: solvation reduces the charge overscreening effect, but it also causes the dilution of absorbed ion concentration. While the former increases the capacitance in the low dilution regime, the latter decreases the capacitance in the high dilution regime. In addition, the dilution also significantly improves the ion dynamics at the interface. Our simulation results demonstrate that diluting an ionic liquid electrolyte could potentially boost the power density while maintaining or even slightly increasing the energy density with a careful selection of solvent concentrations on a nonedge carbon electrode.

We present the complex microscopic dynamics of 1-butyl-3-methyl-imidazolium bis(trifluromethylsulfonyl)imide, [Bmim+][TFSI–], ionic liquid mixed with organic solvents to improve its properties for energy storage applications. To probe the microscopic dynamics on different length and time scales, we have used different neutron scattering spectrometers for quasielastic neutron scattering measurements to compare the effects of solvation in several organic solvents with nearly the same dipole moment but broadly varying bulk diffusivity. The ionic liquid–solvent mixtures show a nanoscopic phase separation into an ionic liquid-rich and a solvent-rich phase, as revealed by the model-free dynamic susceptibility data and further corroborated by the cluster histogram analysis of molecular dynamics (MD) simulation. In both phases, we observe a long-range translational mobility of the [Bmim+] cation, which scales with the bulk diffusivities of the organic solvents; this correlation becomes stronger in the solvent-rich phase. Additionally, various localized motion modes of the [Bmim+] cation are observed. A combination of neutron scattering and MD simulations reveals the parameters governing solvent-controlled diffusivity in an ionic liquid, which helps formulate new electrolytes optimized for efficient energy storage devices.

We investigated the influence of surface oxidization of planar graphene electrodes on charge storage and ion dynamics of supercapacitors. Our approach compared two distinct ionic liquid (IL) electrolytes: tetraethylammonium tetrafluoroborate solvated in acetonitrile (TEA-BF4/ACN) and 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide solvated in acetonitrile (EMIm-TFSI/ACN). Both experimental electrochemical tests and molecular dynamics (MD) simulations showed positive, electrolyte-specific influences of hydroxyl-free electrode interfaces on capacitance. In the EMIm-TFSI/ACN system, the hydroxylated surface, because of its strong interaction with anions, impeded surface charge storage. On the other hand, in the case of TEA-BF4/ACN, the distribution and orientation of ACN across the system exerted vital influence on the capacitance, especially on the positive hydroxyl-free electrode. Furthermore, MD simulations of ion mobility with respect to the electrode surface in the lateral and perpendicular directions revealed significantly slower diffusion performance on the oxidized surface. Our efforts enhanced the level of fundamental understanding of the effects of hydroxyl groups on electrode–electrolyte interfaces and resulting supercapacitor performance.

Despite the ubiquity of nanoparticles in modern materials research, computational scientists are often forced to choose between simulations featuring detailed models of only a few nanoparticles or simplified models with many nanoparticles. Herein, we present a coarse-grained model for amorphous silica nanoparticles with parameters derived via potential matching to atomistic nanoparticle data, thus enabling large-scale simulations of realistic models of silica nanoparticles. Interaction parameters are optimized to match a range of nanoparticle diameters in order to increase transferability with nanoparticle size. Analytical functions are determined such that interaction parameters can be obtained for nanoparticles with arbitrary coarse-grained fidelity. The procedure is shown to be extensible to the derivation of cross-interaction parameters between coarse-grained nanoparticles and other moieties and validated for systems of grafted nanoparticles. The optimization procedure used is available as an open-source Python package and should be readily extensible to models of non-silica nanoparticles.

We describe a method for modeling constant-potential charges in heteroatomic electrodes, keeping pace with the increasing complexity of electrode composition and nanostructure in electrochemical research. The proposed “heteroatomic constant potential method” (HCPM) uses minimal added parameters to handle differing electronegativities and chemical hardnesses of different elements, which we fit to density functional theory (DFT) partial charge predictions in this paper by using derivative-free optimization. To demonstrate the model, we performed molecular dynamics simulations using both HCPM and conventional constant potential method (CPM) for MXene electrodes with Li-TFSI/AN (lithium bis(trifluoromethane sulfonyl)imide/acetonitrile)-based solvent-in-salt electrolytes. Although the two methods show similar accumulated charge storage on the electrodes, the results indicated that HCPM provides a more reliable depiction of electrode atom charge distribution and charge response compared with CPM, accompanied by increased cationic attraction to the MXene surface. These results highlight the influence of elemental composition on electrode performance, and the flexibility of our HCPM opens up new avenues for studying the performance of diverse heteroatomic electrodes including other types of MXenes, two-dimensional materials, metal–organic frameworks (MOFs), and doped carbonaceous electrodes.

Room-temperature ionic liquids (RTILs) are a class of organic salts that are liquid at room temperature. Their physiochemical properties, including low vapor pressure and wide electrochemical stability window, have driven their use as electrolytes in many electrochemical applications; however, the slow transport properties of many RTILs have limited their utility in some applications. This issue is often mitigated by solvating ionic liquids in neutral organic solvents. To date, however, solvent interactions have only been explored for a small number of solvents, particularly acetonitrile and propylene carbonate, at only a few compositions. In this work, we use molecular dynamics simulations in the context of a computational screening approach to study mixtures of ionic liquids in many different solvents at a range of concentrations. Building on prior work, we again find that ionic liquid diffusivity increases monotonically with greater solvent concentration. In contrast to prior work, we find that pure solvent diffusivity, not polarity, is the most influential solvent property on mixture behavior. We also explore the concentration dependence of ionic conductivity and find maxima at intermediate concentrations. Experimental conductivity measurements, inspired by the computational screening study, support this observation with qualitatively consistent results. These results can further guide the selection of solvents for electrochemical applications of RTILs.

Slurry of activated carbon particles mixed with an aqueous electrolyte solution has been used as "flowable electrode" in a few recent electrochemical systems, e.g., electrochemical flow capacitors (EFCs) for energy storage, and flow-electrode capacitive deionization (FCDI) for water treatment. In these applications, the porous carbon particles with very large specific surface area adsorb ions from the electrolyte and meanwhile store electrical charges when a voltage source is added in the charging process. Under the flow condition, the motion of the particles and their mutual contact form a dynamically varying electrical network for the charge transport through the bulk material. We introduce a novel particle-based computational model using the Stokesian dynamics to simulate the hydrodynamic interaction of the carbon particles and the charge transport. An analogous electrical circuit model is developed by approximating the particles with many interconnected resistor-capacitor units, and the circuit's topology is temporally changing depending on the instantaneous particle configuration. The Stokesian model and the circuit model are solved simultaneously to study how the hydrodynamic interaction and cluster formation affect the charge transport process of the slurry. The presence of the stationary current collector can be included to incorporate the near-wall effects on particle mobility. In the simulation, we vary the particle concentration as well as the ratio of the particle charging time to the hydrodynamic interaction time. The results shows that the charge transport in the carbon slurry is enhanced by increasing the concentration of the particles and faster particle charging. In addition, cluster formation of the particles plays an important role for the electronic transport process. After scaling, the transient electrical current from the present study generally agrees with that from previous experimental and modeling studies.

Present and future electrochemical devices employing advanced electrode and electrolyte materials are expected to operate in diverse environments, where they are exposed to variable conditions, such as changing humidity levels. Such conditions can possibly alter the microscopic mechanisms that influence the electrochemical performance. Here, using quasi-elastic neutron scattering and molecular dynamics simulations, we investigate the influence of humidity exposure on a room-temperature ionic liquid, [EMIm+][Tf2N–], in Ti3C2TxMXene. Absorbed water enhances the microscopic mobility of confined [EMIm+][Tf2N–], even though the ionic liquid itself is not very hygroscopic. The absorbed water molecules predominantly reside on the termination groups of the more hydrophilic MXene layers, thereby displacing the ions from the surface and facilitating their motions in the MXene matrix.

We investigate the influence of surface chemistry and ion confinement on capacitance and electrosorption dynamics of room-temperature ionic liquids (RTILs) in supercapacitors. Using air oxidation and vacuum annealing, we produced defunctionalized and oxygen-rich surfaces of carbide-derived carbons (CDCs) and graphene nanoplatelets (GNPs). While oxidized surfaces of porous CDCs improve capacitance and rate handling abilities of ions, defunctionalized nonporous GNPs improve charge storage densities on planar electrodes. Quasi-elastic neutron scattering (QENS) and inelastic neutron scattering (INS) probed the structure, dynamics, and orientation of RTIL ions confined in divergently functionalized pores. Oxidized, ionophilic surfaces draw ions closer to pore surfaces and enhance potential-driven ion transport during electrosorption. Molecular dynamics (MD) simulations corroborated experimental data and demonstrated the significance of surface functional groups on ion orientations, accumulation densities, and capacitance.

The structure and properties of the electrical double layer in ionic liquids is of interest in a wide range of areas including energy storage, catalysis, lubrication, and many more. Theories describing the electrical double layer for ionic liquids have been proposed; however, a full molecular level description of the double layer is lacking. To date, studies have been predominantly focused on ion distributions normal to the surface; however, the 3D nature of the electrical double layer in ionic liquids requires a full picture of the double layer structure not only normal to the surface, but also in plane. Here we utilize 3D force mapping to probe the in plane structure of an ionic liquid at a graphite interface and report the direct observation of the structure and properties of topological defects. The observation of ion layering at structural defects such as step-edges, reinforced by molecular dynamics simulations, defines the spatial resolution of the method. Observation of defects allows for the establishment of the universality of ionic liquid behavior vs. separation from the carbon surface and to map internal defect structure. These studies offer a universal pathway for probing the internal structure of topological defects in soft condensed matter on the nanometer level in three dimensions.

We studied via molecular dynamics the link between the strain-driven hydration free-energy changes in the association process involving finite-size graphene surfaces, the resulting water–graphene interfacial tension, and the combined effect of the surface strain and confinement on the thermodynamic response functions and the dynamics of confined water. We found that an in-plane biaxial tensile strain ε = 10% enhances significantly not only the water–graphene hydrophobicity with respect to that of the unstrained counterpart but also the confinement effect on the thermodynamic response functions and slowing down of the dynamics of water over those of the corresponding bulk counterpart. The interfacial behavior of water in contact with strained-graphene plates resembles that observed for “pp” corrugated-plate configuration, as reported earlier [Chialvo et al. J. Phys. Chem. C2013, 117, 23875], exibiting a significant enhancement of the fluid–surface hydrophobicity and response functions relative to those of the unstrained surface. In contrast, the slowing down of the dynamics of the confined water does not show any differentiation with respect to the type of surface.

As molecular dynamics simulations continue to grow in size and complexity, new techniques are needed to rapidly identify regions of data likely to benefit from further analysis. Audibilization, the conversion of data to sound, facilitates this task by taking advantage of the user’s innate ability to identify anomalies in patterns of sound. Audibilization also complements visualization of a molecular simulation by allowing the user to easily correlate changes in numerical quantities with changes in the overall structure of the molecular system. Here we present three examples highlighting the utility of audibilization in the analysis of three different molecular simulations. First, we present a simulation of liquid water in which the lengths of the O–H bonds are calculated at each time step and audibilized. Interestingly, we find that anomalies in the pattern of bond vibration are due to intermolecular interactions but do not correlate with the formation of hydrogen bonds. Next, we present a simulation of the rupture of a gold nanowire. Here we audibilize the nanowire potential energy and illustrate that sharp changes in this value coincide with important structural events such as the formation of monatomic chains and dislocations. Finally, we present a simulation of single-stranded DNA passing through a nanogap. Here the bond angle is audibilized and used to illustrate the conformational changes of each base as it passes through the nanogap. This simulation also illustrates the use of more advanced audibilization techniques such as the multiplexing of audibilized signals and the weighting of certain segments of data relative to others.

The electrical double layer (EDL) structure formed by ions at a charged surface, which is the key to determining the performance of supercapacitors, has been extensively studied for many monocationic ionic liquids (MILs). However, it is not known what effect replacing MILs with dicationic ionic liquids (DILs) will have on the EDL structure. In this work, the interfacial structure and electrochemical performance of DILs [Cn(mim)2](BF4)2(n= 3, 6, 9) and [C6(mim)2](Tf2N)2were investigated using classical molecular dynamics (MD) simulation for comparison with their monocationic counterparts. Different EDL structures formed by DILs and MILs near an onion-like carbon (OLC) electrode were observed. The interfacial orientation of the imidazolium plane in dications was verified to be similar to that of monocations. Moreover, the dissimilar sizes of the cation/anion and the specific ion adsorption on OLC were found to contribute to the distinctive shape of the differential capacitance–electric potential (C–V) curves, which were also dependent on the type of anions. Increased capacitance in BF4-containing DILs was not observed in comparison with their counterpart MILs, whereas dicationic [C6(mim)2](Tf2N)2yielded higher differential capacitance in contrast to monocationic [C6mim][Tf2N], which was substantiated by cyclic voltammetry measurements as well. This work provides molecular insights into the EDL structure and C–V curves of imidazolium-based DILs in OLC-based supercapacitors.

We studied the potential and time-dependent changes in the electric double layer (EDL) structure of an imidazolium-based room temperature ionic liquid (RTIL) electrolyte at an epitaxial graphene (EG) surface. We used in situ X-ray reflectivity (XR) to determine the EDL structure at static potentials, during cyclic voltammetry (CV) and potential step measurements. The static potential structures were also investigated with fully atomistic molecular dynamics (MD) simulations. Combined XR and MD results show that the EDL structure has alternating anion/cation layers within the first nanometer of the interface and that these structures are distinct at the most positive and negative static potentials (1.0 and −0.4 V, respectively) applied in this study. The dynamical response of the EDL to potential steps has a slow component (>10 s) and the RTIL structure shows hysteresis during CV scans (e.g., at 100 mV/s scan rate). Our results reveal that both the slow kinetics and hysteresis are due to the reorganization of the distinct EDL structures found at the extreme potentials.

We perform a large-scale statistical analysis (>2000 independent simulations) of the elongation and rupture of gold nanowires, probing the validity and scope of the recently proposed ductile-to-brittle transition that occurs with increasing nanowire length [Wu et al. Nano Lett.2012, 12, 910–914]. To facilitate a high-throughput simulation approach, we implement the second-moment approximation to the tight-binding (TB-SMA) potential within HOOMD-Blue, a molecular dynamics package which runs on massively parallel graphics processing units (GPUs). In a statistical sense, we find that the nanowires obey the ductile-to-brittle model quite well; however, we observe several unexpected features from the simulations that build on our understanding of the ductile-to-brittle transition. First, occasional failure behavior is observed that qualitatively differs from that predicted by the model prediction; this is attributed to stochastic thermal motion of the Au atoms and occurs at temperatures as low as 10 K. In addition, we also find that the ductile-to-brittle model, which was developed using classical dislocation theory, holds for nanowires as small as 3 nm in diameter. Finally, we demonstrate that the nanowire critical length is higher at 298 K relative to 10 K, a result that is not predicted by the ductile-to-brittle model. These results offer practical design strategies for adjusting nanowire failure and structure and also demonstrate that GPU computing is an excellent tool for studies requiring a large number of independent trajectories in order to fully characterize a system’s behavior.

We carried out a systematic molecular simulation study of the behavior of a pair of finite-size graphene plates immersed in water at isobaric–isothermal conditions to provide insights into the nature of the water–graphene (corrugated) surface interactions. The goal was to address the link between the corrugation-driven hydration free energy changes in the association process involving graphene plates and the resulting water–graphene interfacial tension, to interrogate the effect of the surface corrugation and confinement on the thermodynamic response functions and the dynamics of confined water and to put the observed behavior in the context of Wenzel’s modification of Young’s equation. We found that graphene confinement induces a significant increase in the isothermal compressibility and isobaric thermal expansivity as well as a pronounced slowdown of the dynamics of water over that of the corresponding bulk counterpart, whose magnitudes depend on the type of surface corrugation involved. Our simulation results for different types of corrugated graphene plates involving identical surface areas do not support the meaning of the “r”-factor underlying Wenzel’s equation for corrugated nanoscale surfaces.

Density functional theory (DFT) is commonly used to treat the adsorption of molecules in carbon and silica pores of various geometries. In this work, we develop a DFT with an accurate molecular-based equation of state to calculate thermodynamic properties using fundamental measure theory (FMT), which is a rigorous approach for the treatment of homogeneous and nonhomogeneous hard-sphere fluids. A theoretical framework results with adsorbing molecules treated as hard-sphere chains with square-well attractive interactions. The Mansoori–Carnahan–Starling–Leland and Carnahan–Starling–Boublik equations of state are used for the hard-sphere interactions, and a version of the statistical associating fluid theory for potentials of variable range (SAFT-VR) is used to describe the square-well fluid. First- and second-order perturbative attractive terms are included in the theory. Theoretical predictions are in good agreement with published results for Monte Carlo simulations of the adsorption of chain molecules. In an extension to published studies, the impact of potentials of increasing complexity are compared using square-well and Lennard–Jones wall potentials. For more realistic adsorption of molecules of varying chain length, full slit-shaped pores are modeled using a 10–4 wall potential. The new approach provides accurate predictions of adsorption of chain molecules in model systems. It should be useful for predicting adsorption on flat surfaces and in slit-shaped pores as well as for analyzing experimental data.

The distinctive structural organization of dicationic ionic liquids (DILs) with varying alkyl linkage chain lengths is systematically investigated using classical molecular dynamics (MD) simulations. In comparison with their counterparts, monocationic ionic liquids (MILs) with free alkyl chains, the DILs with short linkage chains exhibit almost identical structural features regardless of anion types, whereas the long-chain DILs display a relatively insignificant prepeak and low heterogeneity order parameter (HOP), which are accompanied by the less evident structural heterogeneity. Moreover, the predominant role of anion type in the structure of DILs was verified, similar to what is observed in MILs. Finally, the different nanoscale organizations in DILs and MILs are rationalized by the relatively unfavorable straight and folded chain models proposed for the nanoaggregates in DILs and the favorable micelle-like arrangement for those in MILs.

Molecular dynamics (MD) simulations of supercapacitors with single-walled carbon nanotube (SWCNT) electrodes in room-temperature ionic liquids were performed to investigate the influences of the applied electrical potential, the radius/curvature of SWCNTs, and temperature on their capacitive behavior. It is found that (1) SWCNTs-based supercapacitors exhibit a near-flat capacitance–potential curve, (2) the capacitance increases as the tube radius decreases, and (3) the capacitance depends little on the temperature. We report the first MD study showing the influence of the electrode curvature on the capacitance–potential curve and negligible dependence of temperature on capacitance of tubular electrode. The latter is in good agreement with recent experimental findings and is attributed to the similarity of the electrical double layer (EDL) microstructure with temperature varying from 260 to 400 K. The electrode curvature effect is explained by the dominance of charge overscreening and increased ion density per unit area of electrode surface.

Classical molecular dynamics (MD) simulations were employed to investigate the adsorption behaviors of arginine–glycine–aspartate (RGD) tripeptide onto the negatively charged hydroxylated/nonhydroxylated rutile (110) surfaces, mediated by biologically important cations (Na+or Ca2+). The simulation results indicate that the inherent nature of the cation determines its binding strength, thereby regulating the adsorption geometry of the peptide. The sparse hydroxyl groups on the nonhydroxylated rutile diminish the probability of H-bond formation between RGD and the surface, resulting in an early desorption of the peptide even with a mediating Na+ion. In contrast, the negatively charged aspartate (Asp) side chain is bridged to the negatively charged hydroxylated rutile by an inner-sphere Na+ion, in coordination with the Asp–rutile hydrogen bonds at the anchoring sites. The inner- and outer-sphere Ca2+ions are demonstrated to be capable of “trapping” RGD on both hydroxylated and nonhydroxylated rutile, in the absence of hydrogen bonds with the surface. The strongly bound inner-sphere mediating Ca2+ion exerts a “gluing” effect on the Asp side chain, producing a tightly packed RGD–rutile complex, whereas a less localized distribution density of the outer-sphere mediating Ca2+ion results in a higher mobility of the Asp side chain. The intramolecular interaction is suggested to facilitate the structural stability of RGD adsorbed on the negative rutile in a “horseshoe” configuration.

The nanoscale interactions of room temperature ionic liquids (RTILs) at uncharged (graphene) and charged (muscovite mica) solid surfaces were evaluated with high resolution X-ray interface scattering and fully atomistic molecular dynamics simulations. At uncharged graphene surfaces, the imidazolium-based RTIL ([bmim+][Tf2N–]) exhibits a mixed cation/anion layering with a strong interfacial densification of the first RTIL layer. The first layer density observed viaexperiment is larger than that predicted by simulation and the apparent discrepancy can be understood with the inclusion of, dominantly, image charge and π-stacking interactions between the RTIL and the graphene sheet. In contrast, the RTIL structure adjacent to the charged mica surface exhibits an alternating cation–anion layering extending 3.5 nm into the bulk fluid. The associated charge density profile demonstrates a pronounced charge overscreening (i.e., excess first-layer counterions with respect to the adjacent surface charge), highlighting the critical role of charge-induced nanoscale correlations of the RTIL. These observations confirm key aspects of a predicted electric double layer structure from an analytical Landau–Ginzburg-type continuum theory incorporating ion correlation effects, and provide a new baseline for understanding the fundamental nanoscale response of RTILs at charged interfaces.

We examined the influence of external electric field on the vapor–liquid coexistence curve and on structural properties of TIP4P/2005 water confined in hydrophobic and hydrophilic pores using all-atom molecular dynamics simulations. While the electric field increases the critical temperature of bulk water, the effect is contrary on the confined water. We clearly observe that the critical temperature of confined water decreases with increasing field strength. The current work strongly indicates that using electric field results in a decrease in the saturated liquid phase density and increase in the saturated vapor phase density of the confined water and can be used to modulate the evaporation rate of water under confinement. The effect of an electric field on the confined fluids is more pronounced at higher temperature. We also report that the critical density of water behaves differently in hydrophobic and hydrophilic pores. With increasing electric field, the critical density increases in hydrophobic pores; however, it is found to decrease in hydrophilic pores. We analyze the results using pair correlation functions and orientational, tetrahedral, and hydrogen bond distributions. Our investigation indicates that the presence of an electric field enhances the coordination number N(r) of the bulk phase. In contrast, the presence of an electric field reduces N(r) of the confined fluid. This is clearly reflected in the behavior of the critical temperature of bulk and confined water. Our structural analysis reveals that the application of an external field induces orientational order of dipole vector parallel to the field direction in bulk water, whereas its effects on dipole orientation is much less in confined systems. We also report that the hydrogen-bonding behavior in the vapor phase is responsible for the difference in critical density of water confined in hydrophobic and hydrophilic pores.

Using an updated simulation tool, we examine molecular junctions composed of benzene-1,4-dithiolate bonded between gold nanotips, focusing on the importance of environmental factors and interelectrode distance on the formation and structure of bridged molecules. We investigate the complex relationship between monolayer density and tip separation, finding that the formation of multimolecule junctions is favored at low monolayer density, while single-molecule junctions are favored at high density. We demonstrate that tip geometry and monolayer interactions, two factors that are often neglected in simulation, affect the bonding geometry and tilt angle of bridged molecules. We further show that the structures of bridged molecules at 298 and 77 K are similar.

Recent experiments have revealed that onion-like carbons (OLCs) offer high energy density and charging/discharging rates when used as the electrodes in supercapacitors. To understand the physical origin of this phenomenon, molecular dynamics simulations were performed for a room-temperature ionic liquid near idealized spherical OLCs with radii ranging from 0.356 to 1.223 nm. We find that the surface charge density increases almost linearly with the potential applied on electric double layers (EDLs) near OLCs. This leads to a nearly flat shape of the differential capacitance versus the potential, unlike the bell or camel shape observed on planar electrodes. Moreover, our simulations reveal that the capacitance of EDLs on OLCs increases with the curvature or as the OLC size decreases, in agreement with experimental observations. The curvature effect is explained by dominance of charge overscreening over a wide potential range and increased ion density per unit area of electrode surface as the OLC becomes smaller.

We report comprehensive high-accuracy molecular dynamics simulations using the ReaxFF force field to explore the structural changes that occur as Au nanowires are elongated, establishing trends as a function of both temperature and nanowire diameter. Our simulations and subsequent quantitative structural analysis reveal that polytetrahedral structures (e.g., icosahedra) form within the “amorphous” neck regions, most prominently for systems with small diameter at high temperature. We demonstrate that the formation of polytetrahedra diminishes the conductance quantization as compared to systems without this structural motif. We demonstrate that use of the ReaxFF force field, fitted to high-accuracy first-principles calculations of Au, combines the accuracy of quantum calculations with the speed of semiempirical methods.

The dynamics of a single tripeptide Arg-Gly-Asp (RGD) adsorbing onto negatively charged hydroxylated rutile (110) surface in aqueous solution was studied using molecular dynamics (MD) simulations. The results indicate that the adsorbed Na+ions play an important role in determining the binding geometry of RGD. With an initial “horseshoe” configuration, the charged side groups (COO–and NH2) of the peptide are able to interact with the surface through direct hydrogen bonds (H bonds) in the very early stage of adsorption. The Na+ions approach the positively charged Arg side chain, competing with the Arg side chain for adsorption to the negatively charged hydroxyl oxygen. In coordination with the structural adjustment of the peptide, the Arg residue is driven to detach from the rutile surface. In contrast, the Na+ions in close proximity to the negatively charged Asp side chain contribute to the binding of the COO–group on the surface, helping the carboxyl oxygen not involved in COO–-surface H bonds to orientate toward the hydroxyl hydrogens. Once both carboxyl oxygens form enough H bonds with the hydroxyl hydrogens, the redundant ions move toward a more favorable adsorption site.

Using molecular dynamics simulations, we study the impact of physisorbing adsorbates on the structural and mechanical evolution of gold nanowires (AuNWs) undergoing elongation. We used various adsorbate models in our simulations, with each model giving rise to a different surface coverage and mobility of the adsorbed phase. We find that the local structure and mobility of the adsorbed phase remain relatively uniform across all segments of an elongating AuNW, except for the thinning region of the wire where the high mobility of Au atoms disrupts the monolayer structure, giving rise to higher solvent mobility. We analyzed the AuNW trajectories by measuring the ductile elongation of the wires and detecting the presence of characteristic structural motifs that appeared during elongation. Our findings indicate that adsorbates facilitate the formation of high-energy structural motifs and lead to significantly higher ductile elongations. In particular, our simulations result in a large number of monatomic chains and helical structures possessing mechanical stability in excess of what we observe in vacuum. Conversely, we find that a molecular species that interacts weakly (i.e., does not adsorb) with AuNWs worsens the mechanical stability of monatomic chains.

We report all-atom molecular dynamics simulations of water confined in graphite and mica slit pores of variable size ranging from 10 to 60 Å. For each pore size, we demonstrate that the confinement not only reduces the critical temperature of the water but also introduces inhomogeneity in the system that, in turn, results in different vapor–liquid coexistence densities at different layers of the pore. We report, in detail, the contribution of different layers toward the vapor–liquid phase diagram of the confined water in graphite and mica slit pores. We also present the hydrogen bonding (HB) distribution in various layers and the ordering of water molecules near the surface of pore. Bond orientational order calculations of water near the surface of the pores indicate that water molecules tend to order near the mica surface whereas the ordering is absent for the case of graphite pores.

The inherently small system sizes involved imply that, in the absence of large applied fields designed to overwhelm them, fluctuations will play a major role in determining the response and functionality of nanoscale systems. Theoretical advances over the past two decades have provided fresh insight into fluctuations and their role at the nanoscale, even in the presence of arbitrarily large applied external fields. In contrast to traditional engineered systems, Nature’s approach to nanotechnology is to embrace and to exploit fluctuations and noise to create adaptable, persistent, optimized functional architectures. We describe some of the mechanisms by which Nature exploits noise, with the goal of applying these lessons to engineered physical and chemical nanosystems. In particular, we emphasize the critical role of the tails of distributions of properties in both physical and biological nanosystems and their impact on system behavior.

Cell motility is a fundamental process with relevance to embryonic development, immune response, and metastasis. Cells move either spontaneously, in a nondirected fashion, or in response to chemotactic signals, in a directed fashion. Even though they are often studied separately, both forms of motility share many complex processes at the molecular and subcellular scale, e.g., orchestrated cytoskeletal rearrangements and polarization. In addition, at the cellular level both types of motility include persistent runs interspersed with reorientation pauses. Because there is a great range of variability in motility among different cell types, a key challenge in the field is to integrate these multiscale processes into a coherent framework. We analyzed the motility of Dictyosteliumcells with bimodal analysis, a method that compares time spent in persistent versus reorientation mode. Unexpectedly, we found that reorientation time is coupled with persistent time in an inverse correlation and, surprisingly, the inverse correlation holds for both nondirected and chemotactic motility, so that the full range of Dictyosteliummotility can be described by a single scaling relationship. Additionally, we found an identical scaling relationship for three human cell lines, indicating that the coupling of reorientation and persistence holds across species and making it possible to describe the complexity of cell motility in a surprisingly general and simple manner. With this new perspective, we analyzed the motility of Dictyosteliummutants, and found four in which the coupling between two modes was altered. Our results point to a fundamental underlying principle, described by a simple scaling law, unifying mechanisms of eukaryotic cell motility at several scales.

The bonding geometry at the metal−molecule interface plays an important role in determining the conductance behavior of metal−molecule−metal junctions. This bonding geometry has to be determined a priori in quantum mechanical current−voltage (I−V) calculations. To identify the detailed metal−molecule bonding configurations, we perform classical molecular simulations by combining grand canonical Monte Carlo (GCMC) sampling with molecular dynamics (MD) to explore the dynamic elongations of gold nanowires in the presence of benzenedithiol (BDT) molecules. A specific multitime-scale double reversible reference system propagator algorithm (double-RESPA) has been designed for the metal−organic complex in MD simulations to improve the simulation efficiency. We investigate the variations of bonding sites and bonding angles of BDT molecules on a stretched Au nanowire at a constant chemical potential. The density of BDT and the number of bonded and nonbonded BDT molecules in the simulation box is monitored during the entire elongation process. Simulation results show that BDT molecules can form a denser monolayer on Au nanowires than at the Au (111) surface, owing to the many atomic steps on curved surfaces. Moreover, the chemical bonding of BDT on the Au nanowire significantly effect the elongation behavior of Au nanowires compared with those in vacuum. Our present results will be valuable to the understanding of the broken junction mechanism in molecular electronics conductance measurements.

We have developed an off-lattice hybrid discrete-continuum (OLHDC) model of tumor growth and invasion. The continuum part of the OLHDC model describes microenvironmental components such as matrix-degrading enzymes, nutrients or oxygen, and extracellular matrix (ECM) concentrations, whereas the discrete portion represents individual cell behavior such as cell cycle, cell-cell, and cell-ECM interactions and cell motility by the often-used persistent random walk, which can be depicted by the Langevin equation. Using this framework of the OLHDC model, we develop a phenomenologically realistic and bio/physically relevant model that encompasses the experimentally observed superdiffusive behavior (at short times) of mammalian cells. When systemic simulations based on the OLHDC model are performed, tumor growth and its morphology are found to be strongly affected by cell-cell adhesion and haptotaxis. There is a combination of the strength of cell-cell adhesion and haptotaxis in which fingerlike shapes, characteristic of invasive tumor, are observed.

In this article, we review the sorption of multivalent cations on rutile (alpha-TiO2) powder surfaces in aqueous 1:1 electrolyte media from room temperature to 250 degrees C. All cations are shown to occupy 'inner sphere' sorption sites in contact with surface oxygens and hydroxyl groups, as well as the diffuse portion of the electrical double layer (EDL). Sorption is shown to increase strongly with increasing temperature, and the sorption affinity is strongly-related to cation radius and charge. Macroscopic powder pH-titration results obtained with ORNL's high temperature hydrogen electrode concentration cells can be rationalized with Gouy-Chapman-Stern models of the EDL, augmented by atomic-scale structural and proton affinity data from synchrotron X-ray studies and computational modeling approaches.

Atomistic molecular dynamics simulations are performed for up to 20ns to monitor the formation and the stability of complexes composed of single- or double-strand DNA molecules and C60 in aqueous solution. Despite the hydrophobic nature of C60, our results show that fullerenes strongly bind to nucleotides. The binding energies are in the range −27 to −42kcal/mol; by contrast, the binding energy of two fullerenes in aqueous solution is only −7.5kcal/mol. We observe the displacement of water molecules from the region between the nucleotides and the fullerenes and we attribute the large favorable interaction energies to hydrophobic interactions. The features of the DNA-C60 complexes depend on the nature of the nucleotides: C60 binds to double-strand DNA, either at the hydrophobic ends or at the minor groove of the nucleotide. C60 binds to single-strand DNA and deforms the nucleotides significantly. Unexpectedly, when the double-strand DNA is in the A-form, fullerenes penetrate into the double helix from the end, form stable hybrids, and frustrate the hydrogen bonds between end-group basepairs in the nucleotide. When the DNA molecule is damaged (specifically, a gap was created by removing a piece of the nucleotide from one helix), fullerenes can stably occupy the damaged site. We speculate that this strong association may negatively impact the self-repairing process of the double-strand DNA. Our results clearly indicate that the association between C60 and DNA is stronger and more favorable than that between two C60 molecules in water. Therefore, our simulation results suggest that C60 molecules have potentially negative impact on the structure, stability, and biological functions of DNA molecules.

Abstract Grand canonical Monte Carlo simulations are performed to study the adsorption of water in single-walled carbon nanotubes (SWCNs). At room temperature the resulting adsorption isotherms in (10:10) and wider SWCNs are characterized by negligible amount of water uptake at low pressures, sudden and complete pore filling once a threshold pressure is reached, and wide adsorption/desorption hysteresis loops. The width of these loops decreases as pore diameter narrows. Adsorption/desorption hysteresis loops are not observed for water adsorption in (6:6) SWCNs. When the nanotubes are doped with small amounts of oxygenated sites it is possible to obtain adsorption isotherms in which the water uptake increases gradually as the pressure increases. Simulated X-ray diffraction patterns for confined water are also reported.

Abstract The adsorption of water in graphitic carbons is usually simulated via a weighted average of the adsorption isotherms simulated in carbon-slit pore of different widths. By following this procedure, details about pore morphology and pore connectivity may be overlooked. Towards a better match between virtual and real experiments, we present simulated adsorption isotherms for SPC/E model water in porous carbons composed by interconnected carbon-slit pores. The pores are separated from each other by one graphene layer. Imperfections (lack of carbon atoms) in the graphene layers result in interconnections between pores. The grand canonical Monte Carlo algorithm is used here to simulate water adsorption. Our results show that while the qualitative features obtained in the simulation of independent slit-shaped pores are reproduced when interconnected pores are considered, the adsorption isotherms rise more gradually and the adsorption/desorption hysteresis loops are narrower in the latter case.

Alt’s three-dimensional cell balance equation characterizing the chemotactic bacteria was analyzed under the presence of one-dimensional spatial chemoattractant gradients. Our work differs from that of others who have developed rather general models for chemotaxis in the use of a non-smooth anisotropic tumbling frequency function that responds biphasically to the combined temporal and spatial chemoattractant gradients. General three-dimensional expressions for the bacterial transport parameters were derived for chemotactic bacteria, followed by a perturbation analysis under the planar geometry. The bacterial random motility and chemotaxis were summarized by a motility tensor and a chemotactic velocity vector, respectively. The consequence of invoking the diffusion-approximation assumption and using intrinsic one-dimensional models with modified cellular swimming speeds was investigated by numerical simulations. Characterizing the bacterial random orientation after tumbles by a turn angle probability distribution function, we found that only the first-order angular moment of this turn angle probability distribution is important in influencing the bacterial long-term transport.

The molecular-level orientation and structure of ionic liquids (ILs) at liquid–solid interfaces are significantly different than in the bulk. The interfacial ordering influences both IL properties, such as dielectric constants and viscosity, and their efficacy in devices, such as fuel cells and electrical capacitors. Here, we report the layered structures of four ILs on unbiased, highly ordered pyrolytic graphite (HOPG) and Pt(111) surfaces, as determined by atomic force microscopy. The ILs investigated are 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([emim][Tf2N]), 1-ethyl-3-methylimidazolium perfluorobutylsulfonate ([emim][C4F9SO3]), 7-methyl-1,5,7-triazabicyclo[4.4.0]dec-5-ene bis(trifluoromethylsulfonyl)imide ([MTBD][Tf2N]), and 7-methyl-1,5,7-triazabicyclo[4.4.0]dec-5-ene perfluorobutylsulfonate ([MTBD][C4F9SO3]). Molecular dynamics simulations provide complementary information on the position and orientation of the ions. These ILs form a cation layer at the IL–solid interface, followed by a layer of anions. [Emim]+and [MTBD]+have similar orientations at the surface, but [MTBD]+forms a thinner layer compared to [emim]+on both HOPG and Pt(111). In addition, [Tf2N]−shows stronger interactions with Pt(111) surfaces than [C4F9SO3]−.

Ligands can control the surface chemistry, physicochemical properties, processing, and applications of nanomaterials. MXenes are the fastest growing family of two-dimensional (2D) nanomaterials, showing promise for energy, electronic, and environmental applications. However, complex oxidation states, surface terminal groups, and interaction with the environment have hindered the development of organic ligands suitable for MXenes. Here, we demonstrate a simple, fast, scalable, and universally applicable ligand chemistry for MXenes using alkylated 3,4-dihydroxy-l-phenylalanine (ADOPA). Due to the strong hydrogen-bonding and π-electron interactions between the catechol head and surface terminal groups of MXenes and the presence of a hydrophobic fluorinated alkyl tail compatible with organic solvents, the ADOPA ligands functionalize MXene surfaces under mild reaction conditions without sacrificing their properties. Stable colloidal solutions and highly concentrated liquid crystals of various MXenes, including Ti2CTx, Nb2CTx, V2CTx, Mo2CTx, Ti3C2Tx, Ti3CNTx, Mo2TiC2Tx, Mo2Ti2C3Tx, and Ti4N3Tx, have been produced in various organic solvents. Such products offer excellent electrical conductivity, improved oxidation stability, and excellent processability, enabling applications in flexible electrodes and electromagnetic interference shielding.