Machine learning potential for modelling H2 adsorption/diffusion in MOFs with open metal sites.

Metal-organic frameworks (MOFs) incorporating open metal sites (OMS) have been identified as promising sorbents for many societally relevant-adsorption applications including CO2 capture, natural gas purification and H2 storage. This has been ascribed to strong specific interactions between OMS and the guest molecules that enable the MOF to achieve an effective capture even under low gas pressure conditions. In particular, the presence of OMS in MOFs was demonstrated to substantially boost the H2 binding energy for achieving high adsorbed hydrogen densities and large usable hydrogen capacities. So far, there is a critical bottleneck to computationally attain a full understanding of the thermodynamics and dynamics of H2 in this sub-class of MOFs since the generic classical force fields (FFs) are known to fail to accurately describe the interactions between OMS and any guest molecules, in particular H2. This clearly hampers the computational-assisted identification of MOFs containing OMS for a target adsorption-related application since the standard high-throughput screening approach based on these generic FFs is not applicable. Therefore, there is a need to derive novel FFs to achieve accurate and effective evaluation of MOFs for H2 adsorption. On this path, as a proof-of-concept, the soc-MOF-1d containing OMS, previously envisaged as a potential platform for H2 adsorption, was selected as a benchmark material and a machine learning potential (MLP) was derived for the Al-soc-MOF-1d from a dataset initially generated by ab initio molecular dynamics (AIMD) simulations. This MLP was further implemented in MD simulations to explore the H2 binding modes as well as the temperature dependence distribution of H2 in the MOF pores from 10 K to 80 K. MLP-Grand Canonical Monte Carlo (GCMC) simulations were then performed to predict the H2 sorption isotherm of Al-soc-MOF-1d at 77 K that was further confirmed using sorption data we collected on this sample. As a further step, MLP-based molecular dynamics (MD) simulations were conducted to anticipate the kinetics of H2 in this MOF. This work delivers the first MLP able to describe accurately the interactions between the challenging H2 guest molecule and MOFs containing OMS. This innovative strategy applied to one of the most complex molecules owing to its highly polarizable nature, paves the way towards a more systematic accurate and efficient in silico assessment of MOFs containing OMS for H2 adsorption and beyond to the low-pressure capture of diverse molecules.

Selective Water Pore Recognition and Transport through Self-Assembled Alkyl-Ureido-Trianglamine Artificial Water Channels.

In nature, aquaporins (AQPs) are proteins known for fast water transport through the membrane of living cells. Artificial water channels (AWCs) synthetic counterparts with intrinsic water permeability have been developed with the hope of mimicking the performances and the natural functions of AQPs. Highly selective AWCs are needed, and the design of selectivity filters for water is of tremendous importance. Herein, we report the use of self-assembled trianglamine macrocycles acting as AWCs in lipid bilayer membranes that are able to transport water with steric restriction along biomimetic H-bonding-decorated pores conferring selective binding filters for water. Trianglamine [(±)Δ, (mixture of diastereoisomers) and (R,R)3Δ and (S,S)3Δ], trianglamine hydrochloride (Δ.HCl), and alkyl-ureido trianglamines (n = 4, 6, 8, and 12) [(±)ΔC4, (±)ΔC8, (±)ΔC6, and (±)ΔC12] were synthesized for the studies presented here. The single-crystal X-ray structures confirmed that trianglamines form a tubular superstructure in the solid state. The water translocation is controlled via successive selective H-bonding pores (a diameter of 3 Å) and highly permeable hydrophobic vestibules (a diameter of 5 Å). The self-assembled alkyl-ureido-trianglamines achieve a single-channel permeability of 108 water molecules/second/channel, which is within 1 order of magnitude lower than AQPs with good ability to sterically reject ions and preventing the proton transport. Trianglamines present potential for engineering membranes for water purification and separation technologies.

A squarate-pillared titanium oxide quantum sieve towards practical hydrogen isotope separation.

Separating deuterium from hydrogen isotope mixtures is of vital importance to develop nuclear energy industry, as well as other isotope-related advanced technologies. As one of the most promising alternatives to conventional techniques for deuterium purification, kinetic quantum sieving using porous materials has shown a great potential to address this challenging objective. From the knowledge gained in this field; it becomes clear that a quantum sieve encompassing a wide range of practical features in addition to its separation performance is highly demanded to approach the industrial level. Here, the rational design of an ultra-microporous squarate pillared titanium oxide hybrid framework has been achieved, of which we report the comprehensive assessment towards practical deuterium separation. The material not only displays a good performance combining high selectivity and volumetric uptake, reversible adsorption-desorption cycles, and facile regeneration in adsorptive sieving of deuterium, but also features a cost-effective green scalable synthesis using chemical feedstock, and a good stability (thermal, chemical, mechanical and radiolytic) under various working conditions. Our findings provide an overall assessment of the material for hydrogen isotope purification and the results represent a step forward towards next generation practical materials for quantum sieving of important gas isotopes.

Unravelling the pore network and gas dynamics in highly adaptive rubbery organic frameworks.

Rubbery organic frameworks-ROFs have recently emerged as an intriguing class of dynamers by virtue of reversible connections between their building units. Their highly adaptative features at the origin of their spectacular self-healing properties made them also attractive candidates for the development of gas-selective membranes combining high selectivity and fast permeability. So far, little is known on the origin of this unique trait and this clearly hampers the exploitation of this class of dynamers in many areas where stimuli-responsive pore dynamics is of great importance. To address this lack of fundamental knowledge, herein we unravel the self-assembly process of ROFs via the development of an advanced computational methodology combining quantum and force field molecular simulations that enable the description of reversible connections of building units and the long-range organization of the cross-linked ROF network. We demonstrate that both accurate energy barriers associated with the covalent bond formation between the building units and presence of solvent are key parameters to ensure the in silico construction of reliable ROF structure models that are supported by a set of experimental data collected on synthesized ROFs including density, connectivity and porosity. Atomistic insights into the unusual guest-responsive pore dynamics of this intriguing class of dynamers are further gained with a special attention paid to the tunability of this pore flexibility by controlling the chemical composition of the building units. As a further stage, the dynamics of CO2 in these compliance frameworks is scrutinized to shed light on the mechanism at the origin of their promising performance as CO2-selective membranes. We highlight that guest-triggered pore dynamics enables the creation of a diffusion pathway to ensure effective gas transport throughout the whole ROF. This knowledge of the pore structure and its guest-responsive dynamics at the microscopic level is unprecedented in the field of dynamers and it is expected to pave the way towards the optimization of this class of adaptive porous frameworks for many potential applications. Interestingly, this computational approach can be transferable to the exploration of any complex disordered systems showing a high degree of flexibility and guest induced structure/pore reorganization.

How chemical defects influence the charging of nanoporous carbon supercapacitors.

SignificanceNanoporous carbon texture makes fundamental understanding of the electrochemical processes challenging. Based on density functional theory (DFT) results, the proposed atomistic approach takes into account topological and chemical defects of the electrodes and attributes to them a partial charge that depends on the applied voltage. Using a realistic carbon nanotexture, a model is developed to simulate the ionic charge both at the surface and in the subnanometric pores of the electrodes of a supercapacitor. Before entering the smallest pores, ions dehydrate at the external surface of the electrodes, leading to asymmetric adsorption behavior. Ions in subnanometric pores are mostly fully dehydrated. The simulated capacitance is in qualitative agreement with experiments. Part of these ions remain irreversibly trapped upon discharge.

Condensation and growth of amorphous aluminosilicate nanoparticles via an aggregation process.

The precipitation of zeolite nanoparticles involves the initial formation of metastable precursors, such as amorphous entities, that crystallize through non-classical pathways. Here, using reactive force field-based simulations, we reveal how aluminosilicate oligomers grow concomitantly to the decondensation of silicate entities during the initial step of the reaction. Aluminate clusters first form in the solution, thus violating the Loewenstein rule in the first instant of the reaction, which is then followed by their connection with silicate oligomers at the terminal silanol groups before reorganization to finally diffuse within the silicate oligomers to form stable amorphous aluminosilicate nanoparticles that do obey the Loewenstein rule. Our results clearly indicate that aluminate does not serve as the nucleation center for the growth of aluminosilicates in a nucleation-like process but rather proceeds via an aggregation process. The coexistence of aluminosilicate oligomers and small silicate entities induces a phase separation that promotes the precipitation of zeolites with aging.

Quantum Nuclear Dynamics of Protons within Layered Hydroxides at High Pressure.

Studies of the structure of hydroxides under pressure using neutron diffraction reveal that the high concentration of hydrogen is distributed in a disordered network. The disorder in the hydrogen-bond network and possible phase transitions are reported to occur at pressures within the range accessible to experiments for layered calcium hydroxides, which are considered to be exemplary prototype materials. In this study, the static and dynamical properties of these layered hydroxides are investigated using a quantum approach describing nuclear motion, shown herein to be required particularly when studying diffusion processes involving light hydrogen atoms. The effect of high-pressure on the disordered hydrogen-bond network shows that the protons tunnel back and forth across the barriers between three potential minima around the oxygen atoms. At higher pressures the structure has quasi two-dimensional layers of hydrogen atoms, such that at low temperatures this causes the barrier crossing of the hydrogen to be significantly rarefied. Furthermore, for moderate values of both temperature and pressure this process occurs less often than the usual mechanism of proton transport via vacancies, limiting global proton diffusion within layers at high pressure.

Importance of a Fully Anharmonic Treatment of Equilibrium Isotope Fractionation Properties of Dissolved Ionic Species As Evidenced by Li+(aq).

Equilibrium fractionation of stable isotopes is critically important in fields ranging from chemistry, including medicinal chemistry, electrochemistry, geochemistry, and nuclear chemistry, to environmental science. The dearth of reliable estimates of equilibrium fractionation factors, from experiment or from natural observations, has created a need for accurate computational approaches. Because isotope fractionation is a purely quantum mechanical phenomenon, exact calculation of fractionation factors is nontrivial. Consequently, a severe approximation is often made, in which it is assumed that the system can be decomposed into a set of independent harmonic oscillators. Reliance on this often crude approximation is one of the primary reasons that theoretical prediction of isotope fractionation has lagged behind experiment. A class of problems for which one might expect the harmonic approximation to perform most poorly is the isotopic fractionation between solid and solution phases. In order to illustrate the errors associated with the harmonic approximation, we have considered the fractionation of Li isotopes between aqueous solution and phyllosilicate minerals, where we find that the harmonic approximation overestimates isotope fractionation factors by as much as 30% at 25 °C. Lithium is a particularly interesting species to examine, as natural lithium isotope signatures provide information about hydrothermal processes, carbon cycle, and regulation of the Earth's climate by continental alteration. Further, separation of lithium isotopes is of growing interest in the nuclear industry due to a need for pure 6Li and 7Li isotopes. Moving beyond the harmonic approximation entails performing exact quantum calculations, which can be achieved using the Feynman path integral formulation of quantum statistical mechanics. In the path integral approach, a system of quantum particles is represented as a set of classical-like ring-polymer chains, whose interparticle interactions are determined by the rules of quantum mechanics. Because a classical isomorphism exists between the true quantum system and the system of ring-polymers, classical-like methods can be applied. Recent developments of efficient path integral approaches for the exact calculation of isotope fractionation now allow the case of the aforementioned dissolved Li fractionation properties to be studied in detail. Applying this technique, we find that the calculations yield results that are in good agreement with both experimental data and natural observations. Importantly, path integral methods, being fully atomistic, allow us to identify the origins of anharmonic effects and to make reliable predictions at temperatures that are experimentally inaccessible yet are, nevertheless, relevant for natural phenomena.

Efficient Calculation of Free Energy Differences Associated with Isotopic Substitution Using Path-Integral Molecular Dynamics.

The problem of computing free energy differences due to isotopic substitution in chemical systems is discussed. The shift in the equilibrium properties of a system upon isotopic substitution is a purely quantum mechanical effect that can be quantified using the Feynman path integral approach. In this paper, we explore two developments that lead to a highly efficient path integral scheme. First, we employ a mass switching function inspired by the work of Ceriotti and Markland [ J. Chem. Phys. 2013, 138, 014112] that is based on the inverse square root of the mass and which leads to a perfectly constant free energy derivative with respect to the switching parameter in the harmonic limit. We show that even for anharmonic systems, this scheme allows a single-point thermodynamic integration approach to be used in the construction of free energy differences. In order to improve the efficiency of the calculations even further, however, we derive a set of free energy derivative estimators based on the fourth-order scheme of Takahashi and Imada [ J. Phys. Soc. Jpn. 1984, 53, 3765]. The Takahashi-Imada procedure generates a primitive fourth-order estimator that allows the number of imaginary time slices in the path-integral approach to be reduced substantially. However, as with all primitive estimators, its convergence is plagued by numerical noise. In order to alleviate this problem, we derive a fourth-order virial estimator based on a transferring of the difference between second- and fourth-order primitive estimators, which remains relatively constant as a function of the number of configuration samples, to the second-order virial estimator. We show that this new estimator converges as smoothly as the second-order virial estimator but requires significantly fewer imaginary time points.