Mechanical Properties of Twisted Carbon Nanotube Bundles with Carbon Linkers from Molecular Dynamics Simulations.

The manufacturing of high-modulus, high-strength fibers is of paramount importance for real-world, high-end applications. In this respect, carbon nanotubes represent the ideal candidates for realizing such fibers. However, their remarkable mechanical performance is difficult to bring up to the macroscale, due to the low load transfer within the fiber. A strategy to increase such load transfer is the introduction of chemical linkers connecting the units, which can be obtained, for example, using carbon ion-beam irradiation. In this work, we investigate, via molecular dynamics simulations, the mechanical properties of twisted nanotube bundles in which the linkers are composed of interstitial single carbon atoms. We find a significant interplay between the twist and the percentage of linkers. Finally, we evaluate the suitability of two different force fields for the description of these systems: the dihedral-angle-corrected registry-dependent potential, which we couple for non-bonded interaction with either the AIREBO potential or the screened potential ReboScr2. We show that both of these potentials show some shortcomings in the investigation of the mechanical properties of bundles with carbon linkers.

Introducing artificial MOFs for improved machine learning predictions: Identification of top-performing materials for methane storage.

In the field of materials science, the main objective of predictive models is to provide scientists with reliable tools for fast and accurate identification of new materials with exceptional properties. Over the last few years, machine learning methods have been extensively used for the study of the gas-adsorption in nanoporous materials as an efficient alternative of molecular simulations and experiments. In several cases, the accuracy of the constructed predictive models for unknown materials is extremely high. In this study, we explored the adsorption of methane by metal organic frameworks (MOFs) and concluded that many top-performing materials often deviate significantly from the known materials used for the training of the machine learning algorithms. In such cases, the predictions of the machine learning algorithms may not be adequately accurate. For lack of the required appropriate data, we put forth a simple approach for the construction of artificial MOFs with the desired superior properties. Incorporation of such data during the training phase of the machine learning algorithms improves the predictions outstandingly. In some cases, over 96% of the unknown top-performing materials are successfully identified.

Longitudinal strand ordering leads to shear thinning in Nafion.

Proton-exchange membrane fuel cells (PEMFC) offer a promising energy generation alternative for a wide range of technologies thanks to their ecological friendliness and unparalleled efficiency. At the heart of these electrochemical cells lies the membrane electrode assembly with its most important energy conversion components, the Proton Exchange Membrane. This component is created through the use of printing techniques and Nafion inks. The physicochemical properties of the ink, such as its viscosity under shear, are critical for the finished product. In this work we present non-equilibrium Molecular Dynamics simulations using a MARTINI based coarse-grained model for Nafion to understand the mechanism governing the shear viscosity of Nafion solutions. By simulating a Couette flow and calculating density maps of the Nafion chains in these simulations we shed light on the process that leads to the experimentally observed shear thinning effects of Nafion solutions under flow. We observe rod-shaped Nafion microstructures, 3 nm in size on average, when shear flow is absent or low. Higher shear rates instead break these structures and align Nafion strands along the direction of the flow, resulting in lower shear viscosities. Our work paves the way for a deeper understanding of the dynamic and mechanical properties of Nafion including studies of more complex CL and PEM inks.

A Universal Machine Learning Algorithm for Large-Scale Screening of Materials.

Application of machine learning (ML) methods for the determination of the gas adsorption capacities of nanomaterials, such as metal-organic frameworks (MOF), has been extensively investigated over the past few years as a computationally efficient alternative to time-consuming and computationally demanding molecular simulations. Depending on the thermodynamic conditions and the adsorbed gas, ML has been found to provide very accurate results. In this work, we go one step further and we introduce chemical intuition in our descriptors by using the "type" of the atoms in the structure, instead of the previously used building blocks, to account for the chemical character of the MOF. ML predictions for the methane and carbon dioxide adsorption capacities of several tens of thousands of hypothetical MOFs are evaluated at various thermodynamic conditions using the random forest algorithm. For all cases examined, the use of atom types instead of building blocks leads to significantly more accurate predictions, while the number of MOFs needed for the training of the ML algorithm in order to achieve a specified accuracy can be reduced by an order of magnitude. More importantly, since practically there are an unlimited number of building blocks that materials can be made of but a limited number of atom types, the proposed approach is more general and can be considered as universal. The universality and transferability was proved by predicting the adsorption properties of a completely different family of materials after the training of the ML algorithm in MOFs.

Molecular dynamics investigation of the influence of the shape of the cation on the structure and lubrication properties of ionic liquids.

We present a theoretical study of the influence of the molecular geometry of the cation on the response of ionic liquids (ILs) to confinement and mechanical strain. The so-called tailed model includes a large spherical anion and asymmetric cation consisting of a charged head and a neutral tail. Despite its simplicity, this model recovers a wide range of structures seen in ILs: a simple cubic lattice for small tails, a liquid-like state for symmetric cation-tail dimers, and a molecular layer structure for dimers with large tails. A common feature of all investigated model ILs is the formation of a fixed (stable) layer of cations along solid plates. We observe a single anionic layer for small gap widths, a double anionic layer for intermediate ones, and tail-to-tail layer formation for wide gaps. The normal force evolution with gap size can be related to the layer formed inside the gap. The low hysteretic losses during the linear cyclic motion suggest the presence of strong slip inside the gap. In our model the specific friction is low and the friction force decreases with tail size.

A Robust Machine Learning Algorithm for the Prediction of Methane Adsorption in Nanoporous Materials.

In the present study, we propose a new set of descriptors that, along with a few structural features of nanoporous materials, can be used by machine learning algorithms for accurate predictions of the gas uptake capacities of these materials. All new descriptors closely resemble the helium atom void fraction of the material framework. However, instead of a helium atom, a particle with an appropriately defined van der Waals radius is used. The set of void fractions of a small number of these particles is found to be sufficient to characterize uniquely the structure of each material and to account for the most important topological features. We assess the accuracy of our approach by examining the predictions of the random forest algorithm in the relative small dataset of the computation-ready, experimental (CoRE) MOFs (∼4700 structures) that have been experimentally synthesized and whose geometrical/structural features have been accurately calculated before. We first performed grand canonical Monte Carlo simulations to accurately determine their methane uptake capacities at two different temperatures (280 and 298 K) and three different pressures (1, 5.8, and 65 bar). Despite the high chemical and structural diversity of the CoRE MOFs, it was found that the use of the proposed descriptors significantly improves the accuracy of the machine learning algorithm, particularly at low pressures, compared to the predictions made based solely on the rest structural features. More importantly, the algorithm can be easily adapted for other types of nanoporous materials beyond MOFs. Convergence of the predictions was reached even for small training set sizes compared to what was found in previous works using the hypothetical MOF database.

Influence of confinement on flow and lubrication properties of a salt model ionic liquid investigated with molecular dynamics.

We present a molecular dynamics study of the effects of confinement on the lubrication and flow properties of ionic liquids. We use a coarse-grained salt model description of ionic liquid as a lubricant confined between finite solid plates and subjected to two dynamic regimes: shear and cyclic loading. The impact of confinement on the ion arrangement and mechanical response of the system has been studied in detail and compared to static and bulk properties. The results have revealed that the wall slip has a profound influence on the force built-up as a response to mechanical deformation and that at the same time in the dynamic regime interaction with the walls represents a principal driving force governing the behaviour of ionic liquid in the gap. We also observe a transition from a dense liquid to an ordered and potentially solidified state of the ionic liquid taking place under variable normal loads and under shear.

Gas adsorption meets deep learning: voxelizing the potential energy surface of metal-organic frameworks.

Intrinsic properties of metal-organic frameworks (MOFs), such as their ultra porosity and high surface area, deem them promising solutions for problems involving gas adsorption. Nevertheless, due to their combinatorial nature, a huge number of structures is feasible which renders cumbersome the selection of the best candidates with traditional techniques. Recently, machine learning approaches have emerged as efficient tools to deal with this challenge, by allowing researchers to rapidly screen large databases of MOFs via predictive models. The performance of the latter is tightly tied to the mathematical representation of a material, thus necessitating the use of informative descriptors. In this work, a generalized framework to predict gaseous adsorption properties is presented, using as one and only descriptor the capstone of chemical information: the potential energy surface (PES). In order to be machine understandable, the PES is voxelized and subsequently a 3D convolutional neural network (CNN) is exploited to process this 3D energy image. As a proof of concept, the proposed pipeline is applied on predicting [Formula: see text] uptake in MOFs. The resulting model outperforms a conventional model built with geometric descriptors and requires two orders of magnitude less training data to reach a given level of performance. Moreover, the transferability of the approach to different host-guest systems is demonstrated, examining [Formula: see text] uptake in COFs. The generic character of the proposed methodology, inherited from the PES, renders it applicable to fields other than reticular chemistry.

Relating Dry Friction to Interdigitation of Surface Passivation Species: A Molecular Dynamics Study on Amorphous Carbon.

Friction in boundary lubrication is strongly influenced by the atomic structure of the sliding surfaces. In this work, friction between dry amorphous carbon (a-C) surfaces with chemisorbed fragments of lubricant molecules is investigated employing molecular dynamic simulations. The influence of length, grafting density and polarity of the fragments on the shear stress is studied for linear alkanes and alcohols. We find that the shear stress of chain-passivated a-C surfaces is independent of the a-C density. Among all considered chain-passivated systems, those with a high density of chains of equal length exhibit the lowest shear stress. However, shear stress in chain-passivated a-C is consistently higher than in a-C surfaces with atomic passivation. Finally, surface passivation species with OH head groups generally lead to higher friction than their non-polar analogs. Beyond these qualitative trends, the shear stress behavior for all atomic- and chain-passivated, non-polar systems can be explained semi-quantitatively by steric interactions between the two surfaces that cause resistance to the sliding motion. For polar passivation species electrostatic interactions play an additional role. A corresponding descriptor that properly captures the interlocking of the two surfaces along the sliding direction is developed based on the maximum overlap between atoms of the two contacting surfaces.

Organic Filling Mitigates Flaw-Sensitivity of Nanoscale Aragonite.

Engineering at nanoscale holds the promise of tuning materials with extraordinary properties. However, macroscopic approaches commonly used to predict mechanical properties do not fully apply at nanoscale level. A controversial feature is the presence of nanoflaws in aragonite nacre, as it is expected that flaws would weaken the material, whereas nacre still shows high toughness and rupture strength. Here, we performed molecular dynamics and finite element simulations emulating flaws found in aragonite nacre. Our simulations reveal two regimes for fracture: nacre remains flaw-insensitive only for flaws smaller than 1.2 nm depth, or flaws of a few atoms, whereas larger flaws follow a Griffith-like trend resembling macroscopic fracture. We tested an alternative mechanism for flaw-insensitivity in nacre, and investigated the mechanical effect of organic filling to mitigate fracture. We found that a single nacre protein, perlucin, decreases the stress concentration at the fracture point, producing enhancements of up to 15% in rupture strength. Our study reveals a more comprehensive understanding of mechanical stability at the nanoscale and offers new routes toward hybrid nanomaterials.

Stress-induced long-range ordering in spider silk.

The emergence of order from disorder is a topic of vital interest. We here propose that long-range order can arise from a randomly arranged two-phase material under mechanical load. Using Small-Angle Neutron Scattering (SANS) experiments and Molecular Dynamics based finite element (FE) models we show evidence for stress-induced ordering in spider dragline silk. Both methods show striking quantitative agreement of the position, shift and intensity increase of the long period upon stretching. We demonstrate that mesoscopic ordering does not originate from silk-specific processes such as strain-induced crystallization on the atomistic scale or the alignment of tilted crystallites. It instead is a general phenomenon arising from a non-affine deformation that enhances density fluctuations of the stiff and soft phases along the direction of stress. Our results suggest long-range ordering, analogously to the coalescence of defects in materials, as a wide-spread phenomenon to be exploited for tuning the mechanical properties of many hybrid stiff and soft materials.

Viscous friction between crystalline and amorphous phase of dragline silk.

The hierarchical structure of spider dragline silk is composed of two major constituents, the amorphous phase and crystalline units, and its mechanical response has been attributed to these prime constituents. Silk mechanics, however, might also be influenced by the resistance against sliding of these two phases relative to each other under load. We here used atomistic molecular dynamics (MD) simulations to obtain friction forces for the relative sliding of the amorphous phase and crystalline units of Araneus diadematus spider silk. We computed the coefficient of viscosity of this interface to be in the order of 10(2) Ns/m(2) by extrapolating our simulation data to the viscous limit. Interestingly, this value is two orders of magnitude smaller than the coefficient of viscosity within the amorphous phase. This suggests that sliding along a planar and homogeneous surface of straight polyalanine chains is much less hindered than within entangled disordered chains. Finally, in a simple finite element model, which is based on parameters determined from MD simulations including the newly deduced coefficient of viscosity, we assessed the frictional behavior between these two components for the experimental range of relative pulling velocities. We found that a perfectly relative horizontal motion has no significant resistance against sliding, however, slightly inclined loading causes measurable resistance. Our analysis paves the way towards a finite element model of silk fibers in which crystalline units can slide, move and rearrange themselves in the fiber during loading.