Friday, January 07, 2011

J. Chem. Phys.

Electrical and ionic conductivity effects on magic-angle spinning nuclear magnetic resonance parameters of CuI

James P. Yesinowski, Harold D. Ladouceur, Andrew P. Purdy, and Joel B. Miller
We investigate experimentally and theoretically the effects of two different types of conductivity, electrical and ionic, upon magic-angle spinning NMR spectra. The experimental demonstration of these effects involves 63Cu, 65Cu, and 127I variable temperature MAS-NMR experiments on samples of γ-CuI, a Cu+-ion conductor at elevated temperatures as well as a wide bandgap semiconductor. We extend previous observations that the chemical shifts depend very strongly upon the square of the spinning-speed as well as the particular sample studied and the magnetic field strength. By using the 207Pb resonance of lead nitrate mixed with the γ-CuI as an internal chemical shift thermometer we show that frictional heating effects of the rotor do not account for the observations. Instead, we find that spinning bulk CuI, a p-type semiconductor due to Cu+vacancies in nonstoichiometric samples, in a magnetic field generates induced AC electric currents from the Lorentz force that can resistively heat the sample by over 200 °C. These induced currents oscillate along the rotor spinning axis at the spinning speed. Their associated heating effects are disrupted in samples containing inert filler material, indicating the existence of macroscopic current pathways between micron-sized crystallites. Accurate measurements of the temperature-dependence of the 63Cu and 127I chemical shifts in such diluted samples reveal that they are of similar magnitude (ca. 0.27 ppm/K) but opposite sign (being negative for 63Cu), and appear to depend slightly upon the particular sample. This relationship is identical to the corresponding slopes of the chemical shifts versus square of the spinning speed, again consistent with sample heating as the source of the observed large shift changes. Higher drive-gas pressures are required to spin samples that have higher effective electrical conductivities, indicating the presence of a braking effect arising from the induced currents produced by rotating a conductor in a homogeneous magnetic field. We present a theoretical analysis and finite-element simulations that account for the magnitude and rapid time-scale of the resistive heating effects and the quadratic spinning speed dependence of the chemical shift observed experimentally. Known thermophysical properties are used as inputs to the model, the sole adjustable parameter being a scaling of the bulk thermal conductivity of CuI in order to account for the effective thermal conductivity of the rotating powdered sample. In addition to the dramatic consequences of electrical conductivity in the sample,ionic conductivity also influences the spectra. All three nuclei exhibit quadrupolar satellite transitions extending over several hundred kilohertz that reflect defects perturbing the cubic symmetry of the zincblende lattice. Broadening of these satellite transitions with increasing temperature arises from the onset of Cu+ ion jumps to sites with different electric field gradients, a process that interferes with the formation of rotational echoes. This broadening has been quantitatively analyzed for the 63Cu and 65Cu nuclei using a simple model in the literature to yield an activation barrier of 0.64 eV (61.7 kJ/mole) for the Cu+ ion jumping motion responsible for the ionic conductivity that agrees with earlier results based on 63Cu NMR relaxation times of static samples


Communication: Critical dynamics and nuclear relaxation in lipid bilayers

Harden McConnell
Membrane composition fluctuations affect deuterium nuclear magnetic relaxation in lipid bilayers. The time dependence of the fluctuations depends on lipid diffusion. Near a miscibility critical point this diffusion involves an advective hydrodynamic coupling to the aqueous phase. The corresponding diffusion coefficient depends on both the critical length and the fluctuation wavelength. We calculate the effects of these dynamics on transverse deuterium nuclear relaxation in the 0.1o–10o range above the critical temperature.

No comments: