Extracting the P: Two novel amorphous allotropes of elemental phosphorus that consist of polymeric chains (see example depicted) were obtained in a pure form by extraction from their copper iodide adducts with an aqueous solution of potassium cyanide. ³¹P NMR spectra and high-resolution TEM images suggest that the structure of these phosphorus nanorods is unchanged after extraction.

The crystallization mechanism of a high-strength lithium disilicate glass–ceramic in the SiO₂–Li₂O–P₂O₅–Al₂O₃–K₂O–(ZrO₂) system, used as restorative dentistry material, has been examined on the basis of quantitative ²⁹Si magic angle spinning (MAS) and ²⁹Si{⁷Li} rotational echo double resonance (REDOR) NMR spectroscopy. Crystallization occurs in two stages: near 650 °C a significant fraction of the Q⁽³⁾ units disproportionates into crystalline Li₂SiO₃ and Q⁽⁴⁾ units. Upon further annealing of this glass–ceramic to 850 °C the crystalline Li₂SiO₃ phase reacts with the Q⁽⁴⁾ units of the softened residual glass matrix, resulting in the crystallization of Li₂Si₂O₅. The NMR experiments provide detailed insight into the spatial distribution of the lithium ions suggesting the absence of lithium ion clustering in the residual glassy component of the final glass–ceramic. ³¹P MAS-NMR spectra indicate that phosphate acts as a lithium ion scavenger, resulting in the predominant formation of orthophosphate (P⁽⁰⁾) and some pyrophosphate (P⁽¹⁾) groups. Crystallization of Li₂SiO₃ occurs concomitantly with the formation of a highly disordered Li₃PO₄ phase as evidenced from strong linebroadening effects in the ³¹P MAS-NMR spectra. Well-crystallized Li₃PO₄ is only formed at annealing conditions resulting in the formation of crystalline lithium disilicate. These results argue against an epitaxial nucleation process previously proposed in the literature and rather suggest that the nucleation of both lithium metasilicate and lithium disilicate starts at the phase boundary between the disordered lithium phosphate phase and the glass matrix.

The structural organization of sodium borophosphate glasses with composition (Na₂O)_x(BPO₄)_1−x (0.25 ≤ x ≤ 0.55) has been investigated by differential scanning calorimetry, Raman spectroscopy, X-ray photoelectron spectroscopy (XPS), as well as single- and double resonance ¹¹B and ³¹P magic-angle spinning (MAS) NMR. ¹¹B MAS-NMR data indicate the dominance of anionic four-coordinated boron units, and ³¹P MAS NMR reveals the successive transformation of neutral P⁽³⁾ into singly charged P⁽²⁾ units and their further transformation into doubly charged P⁽¹⁾ units at high Na₂O contents. The quantification of these units provides detailed insight into the competition of the network formers borate and phosphate for the network modifier oxide. At low modifier content (x < 0.35), the anionic species are almost exclusively borate (B⁽⁴⁾) units, whereas at higher sodium concentrations, large numbers of anionic phosphate (P⁽²⁾ and P⁽¹⁾) species are formed. O-1s XPS data provide a quantitative distinction between B–O–B, B–O–P, and P–O–P linkages as well as non-bridging oxygen atoms, and comparable numbers can be extracted from ¹¹B and ³¹P MAS-NMR experiments. Both XPS as well as ³¹P{¹¹B} and ¹¹B{³¹P} rotational echo double resonance (REDOR) NMR results reveal strong interactions between the two network formers boron oxide and phosphorus oxide, resulting in a preferred formation of B–O–P linkages. For higher Na₂O contents, however, the successive network modification diminishes this preference, resulting in close-to-statistical network connectivities. Compositional trends of T_g in the Na₂O–B₂O₃–P₂O₅ glass forming system can be correlated with the overall network connectedness, expressed by the total number of bridging oxygen atoms per network former species. However, separate linear correlations are observed for different compositional lines, indicating also the relevance of the type of network former linkages present.

Vitreous samples were prepared in the (100 − x) NaPO3–x WO3 (0 ≤ x ≤ 70) glass forming system using conventional melting–quenching methods. The structural evolution of the vitreous network was monitored as a function of composition by thermal analysis, Raman spectroscopy and high resolution one- and two-dimensional 31P solid state NMR. Addition of WO3 to the NaPO3 glass melt leads to a pronounced increase in the glass transition temperatures, suggesting a significant increase in network connectivity. At the same time Raman spectra indicate that up to about 30 mol% WO3 the tungsten atoms are linked to some non-bridging oxygen atoms (W–O− or W[double bond, length as m-dash]O bonded species), suggesting that the network modifier sodium oxide is shared to some extent between both network formers. W–O–W bond formation occurs only at WO3 contents exceeding 30 mol%. 31P magic angle spinning (MAS)-NMR spectra, supported by two-dimensional J-resolved spectroscopy, allow a clear distinction between species having two, one, and zero P–O–P linkages. The possible formation of some anionic tungsten sites suggested from the Raman data implies an average increase in the degree of polymerization for the phosphorus species, which would result in diminished 31P/23Na interactions. This prediction is indeed confirmed by 31P{23Na} and 23Na{31P} rotational echo double resonance (REDOR) NMR results, which indicate that successive addition of WO3 to NaPO3 glass significantly diminishes the strength of phosphorus–sodium dipole–dipole couplings.