Comparison of the ¹²⁹Xe NMR Chemical Shift with Simulation in Zeolite Y

Abstract

We report quantitative agreement between Monte Carlo simulation and the experimental NMR chemical shift of 129 Xe adsorbed in the supercages of zeolite Y. This agreement supports previous assertions, originating from Ripmeester and Fraissard, that the Xe shift can in principle provide a sensitive measure of the structure imposed on Xe by the three-dimensional potential field provided by the crystal structure of the zeolite. Up to a loading of 7 Xe/cage at 300 K, we verify that the linear rise of shift with loading is due solely to Xe-Xe interaction. We also find excellent agreement at 144 K between simulation and the nonlinear experimental data of Cheung et al. The nonlinear dependence of shift on loading arises both from repulsive Xe-Xe and from repulsive Xe-O interactions. Finally we verify that the effect of temperature on the shift at zero loading can be related to the change of Xe-O pair correlation function at short separations.

Anisotropic Xe Chemical Shifts in Zeolites. The Role of Intra- and Intercrystallite Diffusion

“…

15 Average 129 Xe chemical shifts at 300 K from the GCMC simulations of Xe (□) in NaX (Si/Al = 1.18), and (○) in NaY (Si/Al = 3.00) are compared. These simulations are compared with experimental data (•) in NaX (Si/Al = 1.23) at 293 K from Liu, (·) in NaY (Si/Al = 2.49) at 293 K from Liu, (▴) in NaY (Si/Al = 2.4) at 298 K from Boddenberg et al, and (×) in NaY (Si/Al = 160) at 300 K from Gupta et al Also shown are (+) the experimental data in NaY at various Si/Al ratios ranging from 1.28 to 54.2, at 298 K from Ito et al

…”

“…Pellenq and Nicholson's PN1 potential yielded k H = 3.74 × 10 -2 and 2.90 × 10 -4 mmol/(g Pa) at 195 and 273 K, respectively, whereas our GCMC simulations using parameter set II in the present work, lead to 3.66 × 10 -2 and 2.44 × 10 -4 mmol/(g Pa) at 195 and 273 K, respectively, in good agreement with the results of Pellenq and Nicholson. GCMC simulations have been carried out independently by McCormick and his group for Xe in NaY using a chemical shift that has been parametrized to agree with experiment . In our work, on the other hand, we have used the same chemical shift functions, based on ab initio calculations, for Xe−O, Xe−Na, and Xe−Xe interactions as we have used from the beginning, to interpret the observations for Xe in NaA, Xe + Ar in NaA, Xe + Kr in NaA, Xe in Ca x Na 12 - 2 x A, and in the present work.…”

129Xe NMR Chemical Shifts in Zeolites: Effect of Loading Studied by Monte Carlo Simulations

“…Jameson et al ,,, used grand canonical Monte Carlo simulations and ab initio calculations (justifiably using argon as a model inert gas atom) to study the effect of loading and temperature on the chemical shift of Xe in NaA and found quantitative agreement with the experimental data. In a similar spirit, we recently reported quantitative agreement between Monte Carlo simulation and experiment when the NMR chemical shift of 129 Xe in zeolite Y is modeled by a Drude model …”

“…Fraissard and Ito postulate that δ Xe-Xe is a constant, implying that the shift is linear in loading. In a previous work, we showed for zeolite Y that the implication, that the Xe−Xe pair correlations are constant with loading, is true at low loadings. However, at high loadings, the pair correlation function changes, and the shift can become nonlinear with loading.…”

“…The Xe−pore oxygen interaction and the Xe−Xe interaction were modeled as 6−12 Lennard-Jones interactions. The numerical values of the parameters are given in ref . For channel zeolites, six channel lengths of the structure were used to ensure good statistics; for cage zeolites, eight cages were used.…”