The structure of synthetic nanodiamond has been characterized by (13)C nuclear magnetic resonance (NMR) spectral editing combined with measurements of long-range (1)H-(13)C dipolar couplings and (13)C relaxation times. The surface layer of these approximately 4.8-nm diameter carbon particles consists mostly of sp(3)-hybridized C that is protonated or bonded to OH groups, while sp(2)-hybridized carbon makes up less than 1% of the material. The surface protons surprisingly resonate at 3.8 ppm, but their direct bonding to carbon is proved by fast dipolar dephasing under homonuclear decoupling. Long-range (1)H-(13)C distance measurements, based on (13)C{(1)H} dipolar dephasing by surface protons, show that seven carbon layers, in a shell of 0.63 nm thickness that contains approximately 60% of all carbons, predominantly resonate more than +8 ppm from the 37-ppm peak of bulk diamond (i.e., within the 45-80 ppm range). Nitrogen detected in (15)N NMR spectra is mostly not protonated and can account for some of the high-frequency shift of carbon. The location of unpaired electrons (approximately 40 unpaired electrons per particle) was studied in detail, based on their strongly distance-dependent effects on T(1,C) relaxation. The slower relaxation of the surface carbons, selected by spectral editing, showed that the unpaired electrons are not dangling bonds at the surface. This was confirmed by detailed simulations, which indicated that the unpaired electrons are mostly located in the disordered shell, at distances between 0.4 and 1 nm from the surface. On the basis of these results, a nonaromatic core-shell structural model of nanodiamond particles has been proposed.

No abstract

The rates for nucleation (I) and crystal growth (U) for a lithium disilicate (Li 2 O⅐2SiO 2 , LS 2 ) glass were determined, as a function of temperature, using a new differential thermal analysis (DTA) technique. This technique requires in situ nucleation and crystal-growth heat treatment of a small amount of powdered sample inside the DTA apparatus, which then are followed by a DTA scan at a constant heating rate. The I and U values that have been determined at selected temperatures for the LS 2 glass are in excellent agreement with those reported in the literature. The technique also has been used to determine the concentration of quenched-in nuclei in LS 2 glasses prepared from melts that have been quenched at different rates, which are in reasonable agreement with those estimated from theoretical considerations. This new DTA technique is less tedious, requires a smaller amount of sample, and is at least 10 times faster than the conventional methods that have been used to measure I and U. Also, no special sample preparation, other than simply grinding and screening the glass to a particle size that is suitable for use, is required in this technique, whereas grinding, polishing, or etching is required in conventional methods. The excellent agreement in the I or U values that have been determined for the LS 2 glass via the present and conventional methods demonstrates the accuracy, validity, and usefulness of this DTA method for rapid determination of the nucleation and crystal-growth rates in glasses.