Journal of Solid State Chemistry
Volume 183, Issue 1, January 2010, Pages 120-127
MAS-NMR studies of lithium aluminum silicate (LAS) glasses and glass–ceramics having different Li2O/Al2O3 ratio
A. Ananthanarayanana, G.P. Kothiyala, L. Montagneb and B. Revelb
Keywords: Glass; Glass–ceramics; Silicates; Crystallization; MAS-NMR; XRD
Emergence of phases in lithium aluminum silicate (LAS) glasses of composition (wt%) xLi2O–71.7SiO2–(17.7−x)Al2O3–4.9K2O–3.2B2O3–2.5P2O5 (5.1≤x≤12.6) upon heat treatment were studied. 29Si, 27Al, 31P and 11B MAS-NMR were employed for structural characterization of both LAS glasses and glass–ceramics. In glass samples, Al is found in tetrahedral coordination, while P exists mainly in the form of orthophosphate units. B exists as BO3 and BO4 units. 27Al NMR spectra show no change with crystallization, ruling out the presence of any Al containing phase. Contrary to X-ray diffraction studies carried out, 11B (high field 18.8 T) and 29Si NMR spectra clearly indicate the unexpected crystallization of a borosilicate phase (Li,K)BSi2O6, whose structure is similar to the aluminosilicate virgilite. Also, lithium disilicate (Li2Si2O5), lithium metasilicate (Li2SiO3) and quartz (SiO2) were identified in the 29Si NMR spectra of the glass–ceramics. 31P NMR spectra of the glass–ceramics revealed the presence of Li3PO4 and a mixed phase (Li,K)3PO4 at low alkali concentrations.
J. Phys. Chem. A, 2010, 114 (18), pp 5743–5751
A System for NMR Stark Spectroscopy of Quadrupolar Nuclei
Matthew R. Tarasek and James G. Kempf
Electrostatic influences on NMR parameters are well accepted. Experimental and computational routes have been long pursued to understand and utilize such Stark effects. However, existing approaches are largely indirect informants on electric fields, and/or are complicated by multiple causal factors in spectroscopic change. We present a system to directly measure quadrupolar Stark effects from an applied electric (E) field. Our apparatus and applications are relevant in two contexts. Each uses a radiofrequency (rf) E field at twice the nuclear Larmor frequency (2ω0). The mechanism is a distortion of the E-field gradient tensor that is linear in the amplitude (E0) of the rf E field. The first uses 2ω0 excitation of double-quantum transitions for times similar to T1 (the longitudinal spin relaxation time). This perturbs the steady state distribution of spin population. Nonlinear analysis versus E0 can be used to determine the Stark response rate. The second context uses POWER (perturbations observed with enhanced resolution) NMR. Here, coherent, short-time (T2, the transverse relaxation rate) excitation at 2ω0 is synchronized with an NMR multiple-pulse line-narrowing sequence. Linear analysis of the Stark response is then possible: a quadrupolar multiplet with splitting proportional to E0. The POWER sequence converts the 2ω0 interaction from off-diagonal/nonsecular to the familiar diagonal form (Iz2) of static quadrupole interactions. Meanwhile, background contributions to line width are averaged to zero, providing orders-of-magnitude resolution enhancement for correspondingly high sensitivity to the Stark effect. Using GaAs as a test case with well-defined Stark response, we provide the first demonstration of the 2ω0 effect at high-field (14.1 T) and room temperature. This, along with the simplicity of our apparatus and spectral approach, may facilitate extensions to a wider array of material and molecular systems. The POWER context, which has not previously been tested, is detailed here with new design insights. Several key aspects are demonstrated here, while complete implementation is to be presented at a later time. At present, we (1) account for finite pulse times in pulse sequence design, (2) demonstrate two-channel phase coherence for magnetic (ω0) and electric (2ω0) excitation, and (3) provide line narrowing by a factor of 103. In addition, we find that certain anomalous contributions to the line shape, observed in previous low-field (250 mT) applications, are absent here.