New Insights into the Cs Adsorption on Montmorillonite Clay from 133Cs Solid-State NMR and Density Functional Theory Calculations.

The adsorption sites of Cs on montmorillonite clays were investigated by theoretical 133Cs chemical shift calculations, 133Cs magic-angle-spinning nuclear magnetic resonance (MAS NMR) spectroscopy, and X-ray diffraction under controlled relative humidity. The theoretical calculations were carried out for structures with three stacking variations in the clay layers, where hexagonal cavities formed with Si-O bonds in the tetrahedral layers were aligned as monoclinic, parallel, alternated; with various d-spacings. After structural optimization, all Cs atoms were positioned around the center of hexagonal cavities in the upper or lower tetrahedral sheets. The calculated 133Cs chemical shifts were highly sensitive to the tetrahedral Al (AlT)-Cs distance and d-spacing, rather than to the Cs coordination number. Accordingly, three peaks observed in our theoretical spectra were interpreted to be adsorbed Cs around the center of hexagonal cavity with or without AlT and on the surface in the open nanospace. In a series of 133Cs MAS NMR spectral changes for partial Cs substituted samples, the Cs atoms are preferentially adsorbed at sites near AlT for low Cs substituted montmorillonites. The presence of nonhydrated Cs was also confirmed in partially Cs substituted samples, even after being hydrated under high relative humidity.

Molecular Dynamics Simulation of Water Confinement in Disordered Aluminosilicate Subnanopores.

The porous structure and mass transport characteristics of disordered silicate porous media were investigated via a geometry based analysis of water confined in the pores. Disordered silicate porous media were constructed to mimic the dissolution behavior of an alkali aluminoborosilicate glass, i.e., soluble Na and B were removed from the bulk glass, and then water molecules and Na were introduced into the pores to provide a complex porous structure filled with water. This modelling approach revealed large surface areas of disordered porous media. In addition, a number of isolated water molecules were observed in the pores, despite accessible porous connectivity. As the fraction of mobile water was approximately 1%, the main water dynamics corresponded to vibrational motion in a confined space. This significantly reduced water mobility was due to strong hydrogen-bonding water-surface interactions resulting from the large surface area. This original approach provides a method for predicting the porous structure and water transport characteristics of disordered silicate porous media.

Ab Initio Molecular Dynamics Simulations and GIPAW NMR Calculations of a Lithium Borate Glass Melt.

The atomic structure of a molten 0.3Li2O-0.7B2O3 glass at 1250 K was investigated using ab initio molecular dynamics (AIMD) simulations. The gauge including projector augmented wave (GIPAW) method was then employed for computing the chemical shift and quadrupolar coupling constant of (11)B, (17)O, and (7)Li from 764 AIMD derived structures. The chemical shift and quadrupolar coupling constant distributions were directly estimated from the dynamical structure of the molten glass. (11)B NMR parameters of well-known structural units such as the three-coordinated ring, nonring, and four-coordinated tetrahedron were found to be in good agreement with the experimental results. In this study, more detailed classification of B units was presented based on the number of O species bonded to the B atoms. This highlights the limitations of (11)B NMR sensitivity for resolving (11)B local environment using the experimentally obtained spectra only. The (17)O NMR parameter distributions can theoretically resolve the bridging and nonbridging O atoms with different structural units such as nonring, single boroxol ring, and double boroxol ring. Slight but clear differences in the number of bridging O atoms surrounding Li that have not been reported experimentally were observed in the theoretically obtained (7)Li NMR parameters.

First-principles molecular dynamics simulation and conductivity measurements of a molten xLi2O-(1- x)B2O3 system.

The electronic properties and atomic structure of a molten xLi2O-(1 - x)B2O3 system were investigated by measuring conductivity and using first-principles molecular dynamics (MD) simulations. The conductivities obtained were converted to a Li self-diffusion coefficient Dσ, using the Nernst-Einstein equation to assess charge transfer mechanisms. Dσ was compared with a Li self-diffusion coefficient, DNMR, which we measured in a previous study using high-temperature pulsed field gradient NMR. The DNMR/Dσ of xLi2O-(1 - x)B2O3 (0.2 ≤ x ≤ 0.5) at 1250 K ranged from 2.5 to 3.2, following the same trend as room temperature ionic liquids. First-principles MD simulations were performed using our own finite element density functional theory code, FEMTECK (finite element method-based total energy calculation) for molten xLi2O-(1 - x)B2O3 systems at 1250 K. We found that the O-B-O angle distribution functions were characterized by a peak at approximately 120°. Although the electron number from the electronic radial distribution function was arbitrary with regard to the cutoff distance, the net Li charge calculated from the integrated electron number surrounding Li was approximately 0.9 at 0.085 nm. The mean square displacement (MSD) of Li as a function of time was evaluated from the atomic configuration. Li self-diffusion coefficients calculated from the MSD were in better agreement with experimental results than they were using classical MD.

Deconvolution and Estimation of Water Diffusion in Sulfonated Polyethersulfone Membranes Using Diffusion-Weighted Inversion Recovery.

An NMR method was applied for the deconvolution of specific water in sulfonated polyether sulfone membranes, where sulfonated polyethersulfone is a proton-conducting polymer in polymer electrolyte membrane fuel cells. The distribution of (1)H longitudinal relaxation times obtained by the inverse Laplace method was utilized to estimate the volume fraction of proton species as a function of relative humidity (RH). The relaxation time distribution clearly revealed two distinguished peaks on the order of 10(-3) and 10(-2) s, which corresponded to water in the larger and smaller channels for proton transports, respectively. We applied a pulse sequence to understand the water species by diffusion-weighted inversion recovery, which led to individual self-diffusion coefficients for deconvoluted water by using the longitudinal relaxation time. At 30% RH, the diffusion coefficient of water in small-sized channels is greater than that in large-sized channels. On the other hand, the diffusion coefficients of protons with smaller and larger water channels are almost the same at 50, 70, and 90% RH.