Naiping Hu

The molecular basis of CO2 interaction with polymers containing fluorinated groups: computational chemistry of model compounds and molecular simulation of poly[bis(2,2,2-trifluoroethoxy)phosphazene]

Abstract Ab initio molecular orbital calculations of CO2 and model compounds have been used to identify the nature of specific interactions between CO2 and the fluorinated substituent groups of polymers such as poly(trifluoropropyl methyl siloxane) and poly[bis(2,2,2-trifluoroethoxy)phosphazene] (PTFEP) that exhibits high CO2 permeability and permselectivity. Second-order Moller–Plesset (MP2) perturbation calculations (6-311++G∗∗ basis set) were used to obtain energies, dipole and quadrupole moments, and polarizabilities of CO2, three alkanes (CH4, CH3CH3, and CH3CH2CH3), and three fluoroalkanes (CF4, CH3CF3, and CH3CH2CF3). Results of energy calculations of three CO2–alkane and three CO2–fluoroalkane dimers indicate that CO2 forms a favorable quadrupole–dipole interaction with the fluoroalkyl groups. The maximum quadrupole–dipole interaction energy obtained was −11.5 kJ mol−1 for CO2–CF3CH2CH3. This value is less than interaction energies typical for hydrogen bonding but greater than the London dispersion values reported for the interaction of CO2 with the carbonyl group of poly(methyl methacrylate) (PMMA). Electrostatic potential distributions indicate a small redistribution of electron density to the fluorine atoms of the trifluoroalkanes and to the oxygen atoms of CO2 in the CO2–CF3CH3 and CO2–CF3CH2CH3 dimers. Sorption and molecular association of CO2 with PTFEP has been investigated by molecular simulation of an amorphous cell using the COMPASS molecular mechanics force field. CO2 sorption isotherms obtained by Grand Canonical Monte Carlo (GCMC) simulation indicate an upward deviation from the linear relationship between log S and the Lennard–Jones potential well depth parameter, e/k, in agreement with reported permeability data. Pair-correlation analysis obtained from molecular dynamics simulation show strong correlation of CO2 with the trifluoromethyl group of PTFEP in agreement with the MP2 results showing an association of CO2 with both CH3CF3 and CH3CH2CF3.

A nanoscale-modified LaMer model for particle synthesis from inorganic tin–platinum complexes

The size-tunable structure and properties of Pt nanoparticles at the atomic length scale have attracted significant attention across a wide variety of fields including magnetics, electrocatalysis, optics, and gas-phase synthesis. Mechanisms responsible for the formation Pt nanoparticles remain unclear because of the difficulty generating in situ data for the time-evolution of size, shape, distribution, volume fraction, particle number density, and oxidation state from the starting complexes. We here demonstrate the use of simultaneous small- and wide-angle X-ray scattering combined with UV-vis spectroscopy to measure these key synthesis metrics for the reduction of Pt(IV) by Sn(II) in aqueous solution. This synthesis approach has been previously shown to permit continuous control over Pt nanoparticle size from 0.9 to 2.6 nm to within 10% standard deviation. Such fine control led to the discovery of densely packed amorphous structures at ca. 1.7 nm with substantially enhanced electrocatalytic oxygen reduction relative to nanocrystals and commercial electrocatalysts. Ex situ UV-vis and in situ X-ray scattering are here shown to reveal four distinct stages during synthesis: (1) autoreduction of a ligand/noble metal complex with a unique structure that depends on the Sn(II)/Pt(II) ratio, (2) generation of Pt primary particles and the formation of Pt nuclei at a rate that depends on the structure of the initial complex, (3) nanoparticle growth via LaMers diffusion of these primary particles to the nuclei, and (4) growth termination due to capping from a stabilizing, two-layer ligand shell. We derive a set of consecutive rate equations and associated kinetic parameters that describe each step. The kinetics of ligand rearrangement has been previously found to limit the rate of nanoparticle growth. We incorporate this phenomenon into LaMers classic diffusion-limited growth scheme to extend it to the nanoscale regime. This new model provides detailed understanding of how metal ligands serve as both reducing and stabilizing agents and allow for unprecedented, continuous control over both size and distribution. Systematic variation of temperature permits detailed time resolution at the very onset of Pt primary particle formation, as well as a means to determine temperature sensitivity of nanoparticle growth.

Structural Basis of the 1H-Nuclear Magnetic Resonance Spectra of Ethanol–Water Solutions Based on Multivariate Curve Resolution Analysis of Mid-Infrared Spectra

The 1H-nuclear magnetic resonance (NMR) chemical shifts of ethanol and water hydroxyl groups show a pattern change at a critical ethanol concentration. Below the critical value (20 mol% at 400 Hz), only one hydroxyl peak appears due to fast proton exchange, whereas above the critical concentration, the ethanol hydroxyl peak splits from the water peak emerging as an individual chemical shift. The structural basis of the NMR pattern change was interpreted by a multivariate curve resolution–alternating least squares (MCR-ALS) analysis of the mid-infrared (mid-IR) spectra obtained for ethanol–water solutions. Results suggest that the NMR pattern change is due to the formation of ethanol–ethanol clusters. Below the critical concentration, no ethanol–ethanol clusters exist. Therefore, the NMR does not detect the ethanol environment. Above the critical ethanol concentration, ethanol–ethanol clusters first appear such that a distinct ethanol hydroxyl peak emerges. The basis for the dependence of the critical concentration on working frequency is also interpreted. High frequency NMR measurements are more sensitive to ethanol content, resulting in a lower critical ethanol concentration.

Multilevel morphology of complex nanoporous materials

This work exploits gas adsorption and small-angle X-ray scattering (SAXS) to determine the morphology of complex nanoporous materials. We resolve multiple classes of porosity including previously undetected large-scale texture that significantly compromises the canonical interpretation of gas adsorption. Specifically, a UVM-7 class mesoporous silica was synthesized that has morphological features on three length scales: macropores due to packing of 150 nm globules, 1.9 nm radius spherical mesopores inside the globules, and >7 nm pockets on and between the globules. The total and external surface areas, as well as the mesopore volume, were determined using gas adsorption (αs-plot) and SAXS. A new approach was applied to the SAXS data using multilevel fitting to determine the surface areas on multiple length scales. The SAXS analysis code is applicable to any two-phase system and is freely available to the public. The total surface area determined by SAXS was 12% greater than that obtained by gas adsorption. The macropore interfacial area, however, is only 30% of the external surface area determined by the αs-plot. The overestimation of the external surface area by the αs-plot method is attributed to capillary condensation in nanoscale surface irregularities. The discrepancy is resolved assuming that the macropore-globule interfaces harbor fractally distributed nooks and crannies, which lead to gas adsorption at pressures above the mesopore filling pressure.