5162
J. Phys. Chem. B 2009, 113, 5162–5167
Connectivity and Proximity between Quadrupolar Nuclides in Oxide Glasses: Insights from through-Bond and through-Space Correlations in Solid-State NMR Sung Keun Lee,* Michael Deschamps,‡,§ Julien Hiet,‡,§ Dominique Massiot,‡,§ and Sun Young Park† School of Earth and EnVironmental Sciences, Seoul National UniVersity, Seoul, 151-742 Korea, CNRS, UPR3079 CEMHTI, 1D aVenue de la Recherche Scientifique, 45071 Orle´ans cedex 2, France, and Faculte´ des Sciences, UniVersite´ d’Orle´ans, AVenue du Parc Floral, BP 6749, 45067 Orle´ans cedex 2, France ReceiVed: December 4, 2008; ReVised Manuscript ReceiVed: February 6, 2009
The connectivity and proximity among framework cations and anions in covalent oxide glasses yields unique information whereby their various transport and thermodynamic properties can be predicted. Recent developments and advances in the reconstruction of anisotropic spin interactions among quadrupolar nuclides (spin > 1/2) in solid-state NMR shed light on a new opportunity to explore local connectivity and proximity in amorphous solids. Here, we report the 2D through-bond (J-coupling) and through-space (dipolar coupling) correlation NMR spectra for oxide glasses where previously unknown structural details about the connectivity and proximity among quadrupolar nuclides (27Al, 17O) are determined. Nonbridging oxygen peaks in Ca-aluminosilicate glasses with distinct connectivity, such as Ca-O-Al and Al-O-(Al, Si) are well distinguished in {17O}27Al solid HMQC NMR spectra. Both peaks shift to a lower frequency in direct and indirect dimensions upon the addition of Si to the Ca-aluminate glasses. The 2D 27Al double quantum magic angle spinning NMR spectra for Mg-aluminoborate glasses indicate the preferential proximity between [4]Al and [5]Al leading to the formation of correlations peaks such as [4]Al-[4]Al, [4]Al-[5]Al, and [5]Al-O-[5]Al. A fraction of the [6]Al-[6]Al correlation peak is also noticeable while that of [4,5]Al-[6]Al is missing. These results suggest that [6]Al is likely to be isolated from the [4]Al and [5]Al species, forming [6]Al clusters. The experimental realization of through-bond and through-space correlations among quadrupolar nuclides in amorphous materials suggests a significant deviation from the random distribution among framework cations and a spatial heterogeneity due to possible clustering of framework cations in the model oxide glasses. Introduction The atomic- and nanoscale structures in amorphous oxides control their diverse macroscopic properties (e.g., refs 1-3). Among the many structural controls of thermodynamic and transport properties, the connectiVity (or degree of polymerization) and proximity between the constituent atoms in glasses and the corresponding melts are of great importance.2,3 The network connectiVity in oxide glasses is often described in terms of the fractions of bridging oxygen (BO: oxygen that links framework cations such as Si and Al, e.g., Si-O-Si, Al-O-Al, or Al-O-Si) and nonbridging oxygen (NBO: depolymerized by network modifying cations, e.g., Ca-O-Si, Ca-O-Al, etc.) with a strong dependence on the pressure, temperature, and composition.2,3 A slight increase in the NBO fraction can lead to a decrease of several orders of magnitude in the viscosity of alkali silicate melts;4 this phenomenon yields insights into magma dynamics and emplacement.2,5-7 Despite its fundamental geophysical implications and technological applications, the nature of melt connectivity among framework cations and anions has not yet been fully understood. While the connectivity in oxide glasses refers to how the cations and anions are connected through mostly covalent bonds, * To whom correspondence should be addressed. E-mail: sungklee@ snu.ac.kr. Tel.: 822-880-6729. Fax: 822-871-3269, Address: School of Earth and Environmental Sciences, Seoul National University, Seoul, Korea 151742. † Seoul National University. ‡ CNRS, UPR3079 CEMHTI. § Universite´ d’Orle´ans.
the proximity among constituent atoms in amorphous oxides denotes the degree of spatial closeness among constituent atoms that are not necessarily connected via sharing electrons (mostly manifested as framework cation-oxygen distance and thus short-range order) but include the atomic arrangements up to several coordination spheres, often characterized by intermediate-range order, including through-space distribution of Al cation in glasses. The degree of proximity thus affects the configurational entropy, enthalpy, activity coefficients of oxide component, and heat capacity of oxide glasses (e.g., refs 8-10). Synchrotron or conventional X-ray scattering has often been used to determine a pair correlation (or distribution) function and thus the degree of proximity and connectivity among constituent atoms in oxide glasses.11,12 However, it has limited usefulness in probing the structural details of silicates or borates due to the similar scattering factors of Si and Al, and the low atomic scattering factors for these framework cations. While the neutron scattering can potentially overcome this difficulty, the overlap among correlation functions increases significantly with the number of components, and this hampers the quantitative analysis. Further, detailed information about the connectivity and proximity of melts cannot be determined based on the chemical composition of glasses and melts alone. Along with diverse element-specific local structural probes, recent advances in solid-state NMR have enabled the determination of essential structural information that can be used to reconstruct the atomic connectivity and proximity.5,13 Two highresolution NMR approaches are used to probe the connectivity
10.1021/jp810667e CCC: $40.75 2009 American Chemical Society Published on Web 03/18/2009
Quadrupolar Nuclides in Oxide Glasses and proximity in oxide glasses. First, 17O NMR, particularly 17 O 3QMAS NMR,14,15 is effective for probing the indirect connectivity between framework cations (e.g., Si, Ti, B, and Al) and nonframework cations (e.g., Na, Ca, and K) by analyzing the fractions and the topology around BOs including Si-O-Al, Si-O-Ti, B-O-Al, Na-O-Si, and NBO clusters:16-23 as the signal intensity in 3QMAS NMR depends on the magnitude of quadrupolar interactions and other experimental conditions (e.g., rf power, spinning speed, etc.), calibration of the intensity is essential to yield the quantitative oxygen site fractions using numerical simulations (see ref 24 and references therein for more details). The effect of pressure and temperature on the local connectivity has also been recently studied by 17O NMR, and the results of these studies have confirmed that the degree of network connectivity increases with pressure (e.g., refs 25-28) and NBO cluster fraction in some Ca and Na aluminosilicate glasses also depends on temperature.29,30 Alternatively, the connectivity among nuclides in oxide glasses can be directly probed by exploring intrinsic nuclear spin interactions such as J-coupling (through-bond correlation) and dipolar coupling (through-space correlation). These methods are routinely being used to determine the connectivity in organic molecules in liquid-state NMR. Furthermore, solid-state NMR can be used to determine the proximity among spin 1/2 nuclides (e.g., 29Si, 13C, 31P) in crystals31-34 and glasses (e.g. refs 35, 36). While more than two-thirds of the NMR active nuclides are quadrupolar nuclides (spin number > 1/2), similar progress in solid-state NMR among those nuclides remains difficult due to the prevalent second-order broadenings in solids (see refs 37, 38 and reference therein). It is only in recent years that the reconstruction of anisotropic spin interactions, particularly J-coupling in solids among quadrupolar nuclides has made it possible to determine the local connectivity among quadrupolar nuclides in solids. For example, a solid-state HMQC (heteronuclear multiple quantum correlation) NMR study for Al-rich Ca-aluminate crystals and glasses revealed a through-bond correlation spectrum with direct dimensions for 27Al (spin 5/2) correlated with indirectly observed 17O (spin 5/2) under magic angle spinning (MAS) conditions where an Al-O-Al cluster and a tricluster oxygen ([3]O) can be resolved.39 It would be extremely useful to extend this technique to explore the nature of connectivity in Ca-aluminosilicate glasses, one of the important model systems of silicate magmas and complex multicomponent oxide glasses that have industrial applications.2 Upon the addition of an SiO2 component to binary Ca-aluminate glasses comprising Al-O-Al (BO) and Ca-O-Al (NBO), it would be interesting to observe whether Si increases the polymerization of the melts (increasing BO fraction) forming Si-O-Al or decreases the network polymerization, leading to the formation of Ca-O-Si. The previous 17O 1D MAS and 3QMAS NMR studies revealed the bonding preference of Ca-O-Si over Ca-O-Al.40 {17O}27Al HMQC NMR experiments can provide additional details by directly probing the Al-O connectivity in Ca-aluminosilicate glasses. Homonuclear through-space correlations (dipolar coupling) between spin 1/2 nuclides have recently been investigated using solid-state double quantum (DQ) MAS NMR for crystalline and amorphous oxides. For example, a detailed proximity among varying Q species in alkali silicate glasses was reported recently.35,36 Homonuclear DQ MAS NMR has been successfully applied to provide details about the proximity among quadrupolar nuclides (27Al) in crystalline materials including natural and synthetic zeolites.41-43 The proximity among quadrupolar
J. Phys. Chem. B, Vol. 113, No. 15, 2009 5163 nuclides in oxide glasses, however, has not yet been investigated due to the intrinsic broadening of NMR spectra arising from diverse topological and configurational disorders. Aluminoborate glasses have applications in the glass-ceramic industry and environmental applications in nuclear waste sequestrations.44 While recent careful experiments using REDOR, HETCOR, and multinuclear 3QMAS NMR studies provided much improved insights into their local structures,38,45 the proximity among framework Al remains elusive. Since magnesium aluminoborate glass has three partially resolved Al coordination environments ([4]Al, [5]Al, and [6]Al) at 1 atm under 1D MAS NMR,38,46 it is an ideal model system for testing the connectivity among different Al species using the first application of the 27Al DQ MAS NMR technique. Here, we report a {17O}27Al solid-state HMQC NMR spectra for calcium-rich calcium aluminosilicate and aluminate glasses where two types of NBO environments can be distinguished to provide detailed information about the framework of cation connectivity. We also present the first 27Al DQ MAS NMR spectra for amorphous oxides that show unexpected proximity among framework cations via through-space correlation among quadrupolar nuclides (27Al) in Mg-aluminoborate (MAB) glasses. The results and methods indicate the possibility of studying the local connectivity and proximity among quadrupolar nuclides in a variety of amorphous oxides and other disordered materials. Experiments Sample Preparation. 17O enriched Ca-aluminate and Caaluminosilicate glasses [CA (CaO:Al2O3 ) 65:35) and CAS10 (CaO:Al2O3:SO2 ) 60:30:10)] were synthesized from CaCO3 and 40% 17O-enriched Al2O3 and SiO2. Approximately 0.2 wt % of cobalt oxide was added to reduce the spin-lattice relaxation time. 17O enriched Al2O3 was synthesized by hydrolyzing Al-trietoxide40 and then heated at 1100 °C for 1 h under an Ar environment. These mixtures were decarbonated at 800 °C and then fused for 1 h at 1630 °C in an Ar environment before being quenched. 17O enriched Mg aluminoborate glass (with nominal composition of MgO:Al2O3:B2O3 ) 2:1:2, MAB) was synthesized by fusing mixtures of MgO, B2O3 glass (synthesized right before mixing with other oxides), and 17O enriched Al2O3 at 1500 °C in an Ar environment. MgO powder was preheated at 300 °C for 2 days. A few percent of weight loss was observed for MAB glass possibly due to hygroscopic nature of MgO and/or B2O3. The melts were quenched into glasses by manually lowering the Pt crucible into water. NMR Spectroscopy. The NMR spectra for oxide glasses at 17.6 T collected on a Bruker Avance 750 spectrometer at the CEMHTI laboratory of the CNRS-Orleans include {17O}27Al HMQC, 17O MAS for CA and CAS10 glasses, and 27Al MAS and 27Al double quantum MAS for MAB glass at spinning speed of 23 kHz. The NMR spectra such as 27Al MAS, 3QMAS, and 17 O 3QMAS for MAB glass at 9.4 T were obtained using a Varian Solid NMR 400 system at Seoul National University (at spinning speed of 15 kHz). 27 Al MAS NMR spectra were collected with single pulse acquisition for a pulse length of 0.3 µs, recycle delay of 1 s at 17.6 T at a Larmor frequency of 195.9 MHz. The spectrum was fitted (using DMFIT software47) using a Gaussian isotropic model distribution (GIM, case of d ) 5 of the more general Czjzek distribution) which assumes a statistical distribution of charges around the observed nuclei (see refs 48, 49 for more detailed discussion on GIM and the detailed parametrization to simulate the quadrupolar line shape with disorder and thus the distribution of EFG tensors.). 17O MAS NMR spectra for CA
5164 J. Phys. Chem. B, Vol. 113, No. 15, 2009
Lee et al.
Figure 1. (A) 17O MAS NMR spectra for Ca-aluminate and Ca-aluminosilicate glasses [CA (CaO:Al2O3 ) 65:35) and CAS10 (CaO:Al2O3:SO2 ) 60:30:10) glasses at 17.6 T. (B) {17O}27Al HMQC NMR spectra for CA (C) and CAS10 (D) glasses at 17.6 T. Contour lines are drawn at 5% intervals from relative intensities of 13% to 93%, with added lines at the 4%, 6.5%, and 9% levels to better represent low-intensity peaks. Pulse sequence and coherence pathway for the HMQC NMR experiment are also shown.
and CAS10 glasses were obtained by single pulse acquisition for a pulse length of 0.3 µs, recycle delay of 1 s at 17.6 T at a Larmor frequency of 101.8 MHz for 17O with a 3.2-mm ZrO2 rotor in a Bruker triple-resonance MAS probe. {17O}27Al HMQC spectra were obtained at 17.6 T (with 23 kHz spinning speed). To enhance the signal-to-noise ratio, double frequency sweep (DFS) and acquisition of the full echo were implemented (see ref 39 for more details): DFS on the starting nucleus increases the S/N by a factor of at least 2 for 27Al,50 and acquisition and processing of the full echo, instead of the second half of the FID, also leads to an enhancement of S/N ratio by a factor of about 1.4.51,52 The 90° pulses for the selective excitation of the central transitions were used with a relaxation delay of approximately 1 s (see ref 39 for more details). A high-field (17.6 T) condition also effectively reduces the second-order quadrupolar interactions in 27Al and 17O. The Gaussian window function with 250 Hz line broadening was used in direct dimension while no apodization was used in the indirect dimension. 27 Al DQ MAS NMR spectra for MAB glass were obtained using a pulse sequence employing rotary resonance conditions similar to that described earlier (see ref 43 for details); selective 90° pulses of 12 µs were used with a phase cycle ensuring selection of double quantum coherence between two spins. This
sequence was successfully tested on a crystalline aluminum borate sample, Al20B4O36, (or 9Al2O3 · 2B2O3), using the same experimental conditions, and all the expected correlation peaks were observed, as predicted from the structure, indicating that this NMR sequence was suitable to probe the proximities among [4] Al, [5]Al, and [6]Al atoms.33 The spectrum was sheared such that it had the same scales in both single and double quantum dimensions. 17O and 27Al 3Q MAS NMR spectra for MAB glass at 9.4 T were obtained using shifted-echo pulse sequences, as described previously (5 µs-delay-1.5 µs-delay-20 µs). The recycle delay for 17O MAS and 3QMAS NMR at 9.4 T is 1 s and a magic angle sample spinning speed of 15 kHz was employed. All spectra at 17.6 and 9.4 T were referenced to tap water (for 17O) and AlCl3 (27Al). Results and Discussion Figure 1A shows the 17O MAS NMR spectra for CA and CAS10 glasses at 17.6 T where two types of NBOs and BOs (at approximately 50-80 ppm) were well resolved, as labeled; the obtained spectra were consistent with those obtained in a previous 17O NMR study for similar glasses at a lower field of 14.1 T.40 The peak at approximately 160 ppm for CA glass corresponds to Ca-O-[4]Al. Because the peak at approximately 110-120 ppm develops only in the case of CAS10 glasses, it
Quadrupolar Nuclides in Oxide Glasses was assigned to Ca-O-[4]Si.40 The observed BO peak at approximately 70-80 ppm corresponds to [4]Al-O-[4]Al (in the case of CA glass). The addition of an SiO2 component to the CA glasses leads to the formation of an Al-O-Si BO peak ([4]Al-O-[4]Al + [4]Al-O-[4]Si), as indicated by the shift of the BO peak position in the case of CA glass.40,27Al MAS NMR spectra confirm that the [4]Al species is dominant in both glasses (not shown). A double resonance {17O}27Al HMQC experiment can confirm the above assignments and thus the nature of polymerization in model aluminosilicate glasses by directly probing the connectivity between 27Al and 17O (e.g., ref 39). Figure 1B shows {17O}27Al HMQC spectra for the CA and CAS10 glasses. Two peaks reflecting an [4]Al-O bond in [4] Al-O-[4]Al and Ca-O-[4]Al cluster (labeled) are observed in the indirect dimension (17O) in CA glass. Note that these peaks overlap in the direct 27Al dimension (Figure 1A) and single resonance 27Al MAS NMR spectrum (not shown here). The peak positions in the indirect 17O dimension are roughly consistent with 17O MAS NMR. The Ca-O-[4]Al peak intensity is apparently smaller than that predicted from the 1D MAS data: the peak intensities in the HMQC spectra are not entirely quantitative but instead depend on the spin-spin relaxation time and relative magnitude of J-coupling for each site. Well-resolved oxygen peaks are also observed in the CAS10 glass, which can be attributed to Ca-O-[4]Al (at around 160 ppm in the indirect dimension) and [4]Si-O-[4]Al + [4]Al-O-[4]Al (at round 60-80 ppm), as suggested by the 17O MAS NMR spectra. Note that there is no correlation peak at around 120 ppm in the indirect dimension, confirming the assignment of the peak (Ca-O-Si) using 17O MAS NMR (Figure 1A).40 The difference in peak maxima for Ca-O-[4]Al in CA (approximately 73 ppm) and CAS10 (approximately 68 ppm) glasses are also observed in the direct 27Al dimension. As the chemical shift in 27Al decreases with increasing Si content in aluminosilicate glasses,53 it is likely that [4]Al in Ca-O-[4]Al in CAS10 has [4]Si as the next nearest neighbor. The peak maximum for BO peak (CAS10) in the indirect dimension also moves to a lower frequency (more negative chemical shift) of approximately 6 ppm because of the formation of [4]Si-O-[4]Al. These results demonstrate the usefulness of the J-coupling correlation among quadrupolar nuclides in amorphous oxides for resolving the connectivity among the atomic sites as well as for probing detailed changes in atomic structures in both dimensions without overlap among these BO and NBO sites. Information about the proximity and connectivity among cations is essential to reconstruct the structure-property relations in amorphous glasses and melts from atomic structures. Figure 2A shows the 27Al 3QMAS NMR spectra for MAB glass at 9.4 T, where varying Al coordination environments are well resolved, consistent with earlier 27Al 3QMAS NMR.54 The Al species are also partially resolved in 27Al MAS NMR spectra at a high field of 17.6 T (Figure 2B). The fractions of [5]Al and [6] Al are approximately 26% and 10%, which are similar to those reported from the earlier 1D MAS NMR spectra for similar glass at 11.7 T.38 Information about the relative proximity and indirect connectivity among cations can be obtained from 17O 3QMAS NMR. Figure 2C shows 17O 3QMAS NMR spectra for MAB glasses at 9.4 T where various oxygen sites such as Al-O-Al (at approximately -20 ppm in isotropic dimension) and B-O-B (at approximately -65 ppm dimension) are partially resolved. The peak at approximately -50 ppm in the isotropic dimension arises from Mg-O-B + Al-O-B, while the detailed peak assignments have not yet been investigated.55 The presence of Al-O-Al suggests the proximity among Al cations
J. Phys. Chem. B, Vol. 113, No. 15, 2009 5165
Figure 2. (A) 27Al 3QMAS spectrum at 9.4 T (* refers to the spinning sideband), (B) 27Al MAS NMR spectrum (thick) at 17.6 T and its fits (thin), and (C) 17O 3QMAS NMR spectrum for magnesium aluminoborate glass {MgO:Al2O3:B2O3 ) 2:1:2}. In 17O 3QMAS NMR spectra, contour lines are drawn at 5% intervals from relative intensities of 8% to 93%, with added lines at the 4% level.
in the glasses. While previous quantum chemical calculations and experiments suggested that the 17O isotropic chemical shift of [4]Si-O-[n]Al increases with the coordination number n of Al,56 the correlation for [4]Al-O-[n]Al has not yet been established, which makes it difficult to confirm whether the peak at approximately -20 ppm arises solely from [4]Al-O-[4]Al or [5,6]Al-O-[5,6]Al. The aforementioned 27Al DQ MAS NMR can yield improved data for understanding the proximity among Al cations in MAB glass. Figure 3 shows the 27Al DQ MAS NMR spectra for MAB glasses where partially resolved correlation peaks are clearly
5166 J. Phys. Chem. B, Vol. 113, No. 15, 2009
Lee et al. configurational entropy in the melts. The methods and results presented here can be extended to other complex oxide glasses and diverse disordered materials with varying temperature and pressure, providing greater possibilities for determining the structure of glasses, and in particular, revealing the nature of nonrandomness in the atomic arrangement of oxide glasses. Acknowledgment. S.K.L. was supported by Korea Science & Engineering Foundation through the National Research Laboratory Program (2007-000-20120). We thank J. F. Stebbins and Allwardt for their useful discussions, and two anonymous reviewers for careful and constructive suggestions. We finally thank P. Grandinetti for providing us with the RMN software for 2D NMR data processing. D.M., J.H., and M.D. acknowledge financial support by Re´gion Centre and RMNSOLIDEHR-HC ANR-05-BLAN-0317 ANR contract.
Figure 3. 27Al DQ MAS NMR spectra for magnesium aluminoborate glass {MgO:Al2O3:B2O3 ) 2:1:2} where the several Al-Al correlation peaks are shown (labeled). Open squares denote expected peak positions for other correlation peaks as labeled. The spectra are sheared such that they have equal dimension in both single and double quantum dimensions. Twenty contour lines were drawn from intensities of 99% to 3.6% in such a way that previous contour level is divided by 1.18 (i.e., 93, 83.9, 71.1, 60,..., 4.3, 3.6%). Total projection of the 2D spectra is also shown.
observed in each dimension. The features and peaks in the spectra are due to the proximity among the Al cations with varying coordination states, such as [4]Al-[4]Al, [5]Al-[5]Al, [5] Al-[4]Al, and [6]Al-[6]Al (as labeled). While a random distribution of Al framework units should indicate the formation of noticeable fractions of [5]Al-[6]Al and [4]Al-[6]Al pairs (approximately 15%), such correlation peaks have not been observed (the expected peak positions of these peaks were shown with open square in Figure 3). If the observed “proximity” among Al cations in the MAB glass may stem from the short-range atomic arrangements (i.e., Al-O-Al), the results suggest the presence of [4,5]Al-O-[4,5]Al and [6]Al-O-[6]Al. However, [6]Al-O-[6]Al is a high energy cluster considering bond valence theory and is not likely to form (see ref 3 and references there in). Therefore, the presence of [6]Al-[6]Al in the Al-27 DQMAS spectrum may not be accounted for the short-range atomic arrangement only. This suggests a preferential proximity between [5]Al and [4]Al. Furthermore, an [6]Al cluster is likely to be spatially separated from [5]Al or [4]Al, suggesting another complexity in Mg-aluminoborate glasses. While further experimental study and theoretical quantum chemical calculations are certainly necessary, the results also indicate a significant deviation from the random mixing among Al cations ([4]Al, [5]Al, and [6]Al), and thus, the nature of mixing among framework cations, particularly Al-27 is apparently much more complex than is expected from simple statistical distributions. Note that the 27Al DQ may not be quantitative, and the results obtained here suggest a rather semiquantitative trend. We note that earlier Al-27 DQ NMR study of La-aluminate glasses showed the correlation among diverse Al coordination environments.57 The current NMR results based on correlations among quadrupolar nuclides exhibit nonrandomness in the distributions of cations, and thus, the connectivity and proximity, allowing us to better explain the atomistic origin of configurational results.8,58 The manifested deviation from randomness in the CAS and MAB glasses contributes to a decrease in the
Note Added after ASAP Publication. This paper was published ASAP on March 18, 2009. A new reference was added. The updated paper was reposted on April 9, 2009. References and Notes (1) Greaves, G. N.; Sen, S. AdV. Phys. 2007, 56, 1. (2) Mysen, B. O.; Richet, P. Silicate Glasses and Melts: Properties and Structure (DeVelopments in Geochemistry); Elsevier: Amsterdam, 2005. (3) Lee, S. K. Geochim. Cosmochim. Acta 2005, 69, 3695. (4) Mazurin, O. V. Handbook of Glass Data: Silica Glass and Binary Silicate Glasses; Elsevier Science: New York, 1983. (5) Stebbins, J. F. Dynamics and structure of silicate and oxide melts: nuclear magnetic resonance studies. In Structure, Dynamics, and Properties of Silicate Melts; Stebbins, J. F., McMillan, P. F., Dingwell, D. B., Eds.; Mineralogical Society of America: Washington, D.C., 1995; Vol. 32, pp 191. (6) Lee, S. K.; Lin, J. F.; Cai, Y. Q.; Hiraoka, N.; Eng, P. J.; Okuchi, T.; Mao, H. K.; Meng, Y.; Hu, M. Y.; Chow, P.; Shu, J. F.; Li, B. S.; Fukui, H.; Lee, B. H.; Kim, H. N.; Yoo, C. S. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 7925. (7) Lee, S. K.; Eng, P. J.; Mao, H. K.; Meng, Y.; Newville, M.; Hu, M. Y.; Shu, J. F. Nat. Mater. 2005, 4, 851. (8) Richet, P. Geochim. Cosmochim. Acta 1984, 48, 471. (9) Lee, S. K.; Stebbins, J. F. Am. Mineral. 1999, 84, 937. (10) Lee, S. K.; Stebbins, J. F. Geochim. Cosmochim. Acta 2002, 66, 303. (11) Brown, G. E., Jr.; Farges, F.; Calas, G. X-ray scattering and x-ray spectroscopy studies of silicate melts. In Structure, Dynamics, and Properties of Silicate Melts; Stebbins, J. F., McMillan, P. F., Dingwell, D. B., Eds.; Mineralogical Society of America: Washington, D.C., 1995; Vol. 32, pp 317. (12) Petkov, V.; Billinge, S. J. L.; Shastri, S. D.; Himmel, B. Phys. ReV. Lett. 2000, 85, 3436. (13) Eckert, H. Structural studies of non-crystalline solids using solid state NMR. New experimental approaches and results. In Solid-State NMR IV. Methods and Applications of Solid-State NMR; Blich, B., Ed.; SpringerVerlag: Berlin, 1994; pp 127. (14) Frydman, I.; Harwood, J. S. J. Am. Chem. Soc. 1995, 117, 5367. (15) Wu, G.; Dong, S. J. Am. Chem. Soc. 2001, 123, 9119. (16) Amoureux, J. P.; Fernandez, C.; Frydman, L. Chem. Phys. Lett. 1996, 259, 347. (17) Ashbrook, S. E.; Berry, A. J.; Wimperis, S. J. Phys. Chem. B 2002, 106, 773. (18) Dirken, P. J.; Kohn, S. C.; Smith, M. E.; Vaneck, E. R. H. Chem. Phys. Lett. 1997, 266, 568. (19) Lee, S. K.; Stebbins, J. F. J. Non-Cryst. Solids 2000, 270, 260. (20) Lee, S. K.; Stebbins, J. F J. Phys. Chem. B 2003, 107, 3141. (21) Stebbins, J. F.; Oglesby, J. V.; Xu, Z. Am. Mineral. 1997, 82, 1116. (22) Stebbins, J. F.; Xu, Z. Nature 1997, 390, 60. (23) Sen, S.; Youngman, R. E J. Phys. Chem B 2004, 108, 7557. (24) Lee, S. K.; Lee, B. H. J. Phys. Chem. B 2006, 110, 16408. (25) Allwardt, J. R.; Schmidt, B. C.; Stebbins, J. F. Chem. Geol. 2004, 213, 137. (26) Lee, S. K.; Cody, G. D.; Fei, Y.; Mysen, B. O. Geochim. Cosmochim. Acta 2004, 68, 4189. (27) Xue, X.; Stebbins, J. F.; Kanzaki, M. Am. Mineral. 1994, 79, 31. (28) Yarger, J. L.; Smith, K. H.; Nieman, R. A.; Diefenbacher, J.; Wolf, G. H.; Poe, B. T.; McMillan, P. F. Science 1995, 270, 1964. (29) Stebbins, J. F.; Dubinsky, E. V.; Kanehashi, K.; Kelsey, K. E. Geochim. Cosmochim. Acta 2008, 72, 910.
Quadrupolar Nuclides in Oxide Glasses (30) Dubinsky, E. V.; Stebbins, J. F. Am. Mineral. 2006, 91, 753. (31) Fayon, F.; King, I. J.; Harris, R. K.; Gover, R. K. B.; Evans, J. S. O.; Massiot, D. Chem. Mater. 2003, 15, 2234. (32) Fayon, F.; Le Saout, G.; Emsley, L.; Massiot, D. Chem. Commun. 2002, 1702. (33) Fyfe, C. A.; Grondey, H.; Feng, Y.; Kokotailo, G. T. J. Am. Chem. Soc. 1990, 112, 8812. (34) Brouwer, D. H.; Darton, R. J.; Morris, R. E.; Levitt, M. H. J. Am. Chem. Soc. 2005, 127, 10365. (35) Feike, M.; Jager, C.; Spiess, H. W. J. Non-Cryst. Solids 1998, 223, 200. (36) Olivier, L.; Yuan, X.; Cormack, A. N.; Ja¨ger, C. J. Non-Cryst. Solids 2001, 293-295, 53. (37) Mali, G.; Kaucic, V.; Taulelle, F. J. Chem. Phys. 2008, 128, 204503. (38) Chan, J. C. C.; Bertmer, M.; Eckert, H. J. Am. Chem. Soc. 1999, 121, 5238. (39) Iuga, D.; Morais, C.; Gan, Z. H.; Neuville, D. R.; Cormier, L.; Massiot, D. J. Am. Chem. Soc. 2005, 127, 11540. (40) Allwardt, J.; Lee, S. K.; Stebbins, J. F. Am. Mineral. 2003, 88, 949. (41) Eden, M.; Annersten, H.; Zazzi, A. Chem. Phys. Lett. 2005, 410, 24. (42) Eden, M.; Grinshtein, J.; Frydman, L. J. Am. Chem. Soc. 2002, 124, 9708. (43) Mali, G.; Fink, G.; Taulelle, F. J. Chem. Phys. 2004, 120, 2835.
J. Phys. Chem. B, Vol. 113, No. 15, 2009 5167 (44) Xu, Z.; Maekawa, H.; Oglesby, J. V.; Stebbins, J. F. J. Am. Chem. Soc. 1998, 120, 9894. (45) Du, L. S.; Stebbins, J. F. J. Non-Cryst. Solids 2005, 351, 3508. (46) Bunker, B. C.; Kirkpatrick, R. J.; Brow, R. K.; Turner, G. L.; Nelson, C. J. Am. Ceram. Soc. 1991, 74, 1430. (47) Massiot, D.; Fayon, F.; Capron, M.; King, I.; Le Calve, S.; Alonso, B.; Durand, J. O.; Bujoli, B.; Gan, Z. H.; Hoatson, G. Magn. Reson. Chem. 2002, 40, 70. (48) Le Caer, G.; Brand, R. A. J. Phys. Condens. Matter 1998, 10, 10715. (49) Neuville, D. R.; Cormier, L.; Massiot, D. Geochim. Cosmochim. Acta 2004, 68, 5071. (50) Iuga, D.; Kentgens, A. P. M. J. Magn. Reson. 2002, 158, 65. (51) Grandinetti, P. J.; Baltisberger, J. H.; Llor, A.; Lee, Y. K.; Werner, U.; Eastman, M. A.; Pines, A. J. Magn. Reson., Ser. A 1993, 103, 72. (52) Massiot, D.; Touzo, B.; Trumeau, D.; Coutures, J. P.; Virlet, J.; Florian, P.; Grandinetti, P. J. Solid State Nucl. Magn. Reson. 1996, 6, 73. (53) Lee, S. K.; Stebbins, J. F. J. Phys. Chem. B 2000, 104, 4091. (54) Baltisberger, J. H.; Xu, Z.; Stebbins, J. F.; Wang, S.; Pines, A. J. Am. Chem. Soc. 1996, 118, 7209. (55) Wang, S.; Stebbins, J. F. J. Non-Cryst. Solids 1998, 231, 286. (56) Lee, S. K. J. Phys. Chem. B 2004, 108, 5889. (57) Lo, A. Y. H.; Eden, M. Phys. Chem. Chem. Phys. 2008, 10, 6635– 6644. (58) Adam, G.; Gibbs, J. H. J. Chem. Phys. 1965, 43, 139.
JP810667E
