NMR and conductivity studies of the mixed glass former effect in lithium borophosphate glasses.

Alkali ion charge transport has been studied in a series of mixed glass former lithium borophosphate glasses of composition 0.33Li(2)O + 0.67[xB(2)O(3) + (1 - x)P(2)O(5)]. The entire concentration range, 0.0 ≤ x ≤ 1.0, from pure glassy Li(2)P(4)O(11) to pure glassy Li(2)B(4)O(7) has been examined while keeping the molar fraction of Li(2)O constant. Electrical conductivity measurements and nuclear magnetic resonance techniques such as spin relaxometry, line shape analysis, and stimulated-echo spectroscopy were used to examine the temperature and frequency dependence of the Li(+) ion motion over wide ranges of time scale and temperature. By accurately determining motional time scales and activation energies over the entire composition range the ion dynamics and the charge transport are found to be fastest if the borate and the phosphate fractions are similar. The nonlinear variation of the charge conduction, the most notable feature of the mixed glass former effect, is discussed in terms of the composition dependence of network former units which determine the local glass structure.

Spectral editing based on scalar spin-spin interactions: new results on the structure of metathiophosphate glasses.

The local structure of glassy NaPS(3) and AgPS(3) was analyzed based on quantitative (31)P MAS-NMR spectroscopy. The glasses contain some oxide impurities, which could be quantified from the NMR spectral analysis. Four discrete resonances are observed in both glasses, which were assigned to four distinct types of phosphate groups P((n)), where n is the number of P-S-P bridges(i.e., P((0)), P((1)), P((2)), and P((3)) units, respectively) with the help of 2D homonuclear J-resolved and INADEQUATE methods. Based on the results obtained, the interpretations of previous spectra obtained at low spinning speeds on lithium and silver thiophosphate glasses (Chem. Mater. 2 (1990), 273, and J. Am. Chem. Soc. 114 (1992), 5775) need to be revised. Contrary to the situation in alkali phosphate glasses, the corresponding sulfide analogs are characterized by a wide P((n)) species distribution close to that predicted by a statistical charge distribution. INADEQUATE experiments fail to detect P((n))-P((n-1)) connectivities, suggesting that the structure of these glasses is rather inhomogeneous, possibly featuring the different P((n)) species in segregated domains.