Paul F. McMillan

Relaxation in glassforming liquids and amorphous solids

The field of viscous liquid and glassy solid dynamics is reviewed by a process of posing the key questions that need to be answered, and then providing the best answers available to the authors and their advisors at this time. The subject is divided into four parts, three of them dealing with behavior in different domains of temperature with respect to the glass transition temperature, Tg , and a fourth dealing with ‘‘short time processes.’’ The first part tackles the high temperature regime T.Tg ,i n which the system is ergodic and the evolution of the viscous liquid toward the condition at Tg is in focus. The second part deals with the regime T;Tg , where the system is nonergodic except for very long annealing times, hence has time-dependent properties ~aging and annealing!. The third part discusses behavior when the system is completely frozen with respect to the primary relaxation process but in which secondary processes, particularly those responsible for ‘‘superionic’’ conductivity, and dopart mobility in amorphous silicon, remain active. In the fourth part we focus on the behavior of the system at the crossover between the low frequency vibrational components of the molecular motion and its high frequency relaxational components, paying particular attention to very recent developments in the short time dielectric response and the high Q mechanical response.

A Raman spectroscopic study of glasses along the joins silica-calcium aluminate, silica-sodium aluminate, and silica-potassium aluminate

Aluminosilicate glasses with compositions along the joins silica-calcium aluminate, silica sodium aluminate and silica-potassium aluminate have been prepared by conventional and solar melting techniques and studied by Raman spectroscopy. The Raman spectra of crystalline calcium aluminate, anorthite and silica polymorphs are discussed in relation to their crystal structures, and compared with the spectra of the corresponding glasses. The glass and crystal spectra are generally comparable, suggesting similar vibrational structures. These crystals have structures based on tetrahedral aluminosilicate frameworks, and a similar molecular structure is suggested for the glasses, although it is noted that the Raman spectra do not directly characterize the aluminate polyhedra. Within the three glass series, our interpretation of the unresolved high-frequency bands shows the appearance of discrete bands near 1120, 1000, 930 and 890 cm−1 as the silica content is decreased. This is compared with the behaviour of high-frequency bands in simple silicate systems, and used to suggest that the four bands in the aluminosilicate systems are due to stretching vibrations of silicate tetrahedra bound to one, two, three and four aluminium atoms. The spectra of calcium, sodium, potassium and lithium aluminosilicate glasses with similar silica contents are compared, and interpreted by the above model. This is used to construct a simple model for the effect of metal cation on aluminosilicate molecular groups in the glass structure, consistent with the results of calorimetric studies on similar systems.

Pressure-induced amorphization and an amorphous–amorphous transition in densified porous silicon

Crystalline and amorphous forms of silicon are the principal materials used for solid-state electronics and photovoltaics technologies. Silicon is therefore a well-studied material, although new structures and properties are still being discovered. Compression of bulk silicon, which is tetrahedrally coordinated at atmospheric pressure, results in a transition to octahedrally coordinated metallic phases. In compressed nanocrystalline Si particles, the initial diamond structure persists to higher pressure than for bulk material, before transforming to high-density crystals. Here we report compression experiments on films of porous Si, which contains nanometre-sized domains of diamond-structured material. At pressures larger than 10 GPa we observed pressure-induced amorphization. Furthermore, we find from Raman spectroscopy measurements that the high-density amorphous form obtained by this process transforms to low-density amorphous silicon upon decompression. This amorphous–amorphous transition is remarkably similar to that reported previously for water, which suggests an underlying transition between a high-density and a low-density liquid phase in supercooled Si (refs 10, 14, 15). The Si melting temperature decreases with increasing pressure, and the crystalline semiconductor melts to a metallic liquid with average coordination ∼5 (ref. 16).

The tetrahedral framework in glasses and melts — inferences from molecular orbital calculations and implications for structure, thermodynamics, and physical properties

Results of ab initio molecular orbital (MO) calculations provide a basis for the interpretation of structural and thermodynamic properties of crystals, glasses, and melts containing tetrahedrally coordinated Si, Al, and B. Calculated and experimental tetrahedral atom-oxygen (TO) bond lengths are in good agreement and the observed average SiO and AlO bond lengths remain relatively constant in crystalline, glassy, and molten materials. The TOT framework geometry, which determines the major structural features, is governed largely by the local constraints of the strong TO bonds and its major features are modeled well by ab initio calculations on small clusters. Observed bond lengths for non-framework cations are not always in agreement with calculated values, and reasons for this are discussed in the text. The flexibility of SiOSi, SiOAl, and AlOAl angles is in accord with easy glass formation in silicates and aluminosilicates. The stronger constraints on tetrahedral BOB and BOSi angles, as evidenced by much deeper and steeper calculated potential energy versus angle curves, suggest much greater difficulty in substituting tetrahedral B than Al for Si. This is supported by the pattern of immiscibility in borosilicate glasses, although the occurrence of boron in trigonal coordination is an added complication. The limitations on glass formation in oxysulfide and oxynitride systems may be related to the angular requirements of SiSSi and Si(NH)Si groups.Although the SiO and AlO bonds are the strongest ones in silicates and aluminosilicates, they are perturbed by other cations. Increasing perturbation and weakening of the framework occurs with increasing ability of the other atom to compete with Si or Al for bonding to oxygen, that is, with increasing cation field strength. The perturbation of TOT groups, as evidenced by TO bond lengthening predicted by MO calculations and observed in ordered crystalline aluminosilicates, increases in the series Ca, Mg and K, Na, Li. This perturbation correlates strongly with thermochemical mixing properties of glasses in the systems SiO2-M1n/n+AlO2 and SiO2-Mn+On/2 (M=Li, Na, K, Rb, Cs, and Mg, Ca, Sr, Ba, Pb), with tendencies toward immiscibility in these systems, and with systematics in vibrational spectra. Trends in physical properties, including viscosity at atmospheric and high pressure, can also be correlated.

The structures and vibrational spectra of crystals and glasses in the silica-alumina system

Solar furnace melting and fast-quench techniques have been used to prepare SiO2Al2O3 glasses to high alumina content (near 60 mol% Al2O3), which have been studied by Raman spectroscopy. These spectra may not be simply interpreted. The structures of crystalline compounds in the SiO2Al2O3 system are discussed in relation to their vibrational spectra. On the basis of this discussion and other considerations, a structural model for the silica-alumina glass system is proposed, which is consistent with the stable or metastable immiscibility suggested along this join. The essential features of this model include a modified silica structure at low alumina content, and “structure-broken” regions at high alumina compositions, with silicon in tetrahedral coordination, but aluminium assuming a variety of bonding geometries. These are proposed to include aluminate tetrahedra with higher polymerization than simple corner-sharing, and less well-defined polyhedra of higher average coordination number.

Raman spectroscopy of calcium aluminate glasses and crystals

Solar furnace melting and fast-quench techniques have been used to prepare calcium aluminate glasses from 75 mol% CaO to 82 mol% Al2O3, which have been studied by Raman spectroscopy. The CaAl2O4 glass spectrum may be interpreted in terms of a fully-polymerized network of tetrahedral aluminate units, which is depolymerized on addition of CaO component analogous to binary silicate systems. The spectra of glasses with higher alumina content than CaAl2O4 may not be simply interpreted and a structural model is proposed which would be consistent with the glass spectra and with observed crystal structures along the CaAl2O4Al2O3 join. This model suggests formation of highly condensed aluminate tetrahedral on initial addition of alumina, with the appearance of aluminate polyhedra of higher average coordination at higher alumina content. Similar high coordination polyhedral are also suggested for a limited composition range along the CaOCaAl2O4 join. These interpretations are compared with the results of a previous study in the SiO2Al2O3 glass system.

Vibrational spectroscopy of silicate liquids and glasses

The application of vibrational spectroscopy to the study of silicate liquids and glasses is described, and new Raman data for K2Si2O5 and K2Si4O9 compositions are presented. The timescale of the vibrational spectroscopic experiments relative to relaxation timescales in the melts is discussed. Silicate systems are usually described as “liquid” or “glassy” based on experimental measurements of viscosity or heat capacity. These have a much longer characteristic measurement timescale than the vibrational spectroscopic experiments. Because of the long structural relaxation times for silicate frameworks over the normal laboratory temperature range, silicate “glasses” and “liquids” always show the same, unrelaxed response to the vibrational spectroscopic experiment. This is one reason for the observed close similarity between “glass” and “melt” spectra. Vibrational spectroscopy can readily be used to investigate structural changes which occur within supercooled silicate liquids due to structural relaxation on the laboratory timescale, above the glass transition temperature, Tg. The vibrational spectroscopies are complementary to other spectroscopic methods, including nuclear magnetic resonance, for this type of study. Our new Raman spectroscopic results on K-disilicate and -tetrasilicate glasses and liquids show effects due to structural relaxation above Tg. The spectra for K2Si2O5 show evidence for an increase in the concentration of Q2 silicate species with increasing temperature. We have determined the enthalpy change for the 2Q3  Q2 + Q4 speciation reaction in K2Si2O5 to be ∼ 20 kJ mol−1, of the same order of magnitude as those obtained for liquids near Na2Si2O5 composition by previous workers. For K2Si4O9 glass, the Raman data show evidence for a different type of structural relaxation. The intensity of a peak near 590 cm−1 increases with increasing temperature above Tg, which is interpreted as an increase in the proportion of three-membered siloxane rings in the liquid. The enthalpy change for formation of these three-membered rings is also 20 kJ mol−1, consistent with the results of a previous study on SiO2 glass.

Icosahedral packing of B12 icosahedra in boron suboxide (B6O)

Objects with icosahedral symmetry (Ih) bear a special fascination; natural examples are rare, but include radiolaria and virus particles (virions). The discovery of C60, a molecule in the shape of a truncated icosahedron with Ih symmetry, has aroused widespread interest. In 1962, Mackay described a radiating packing of spheres in Ih symmetry, in which the centres of successive shells of spheres lie on the surfaces of icosahedra. There has been extensive investigation of the conditions under which such packing might be realized in assemblies of atoms or of molecules such as C60 (ref. 5). Here we report the preparation, at high temperatures and pressures, of boron suboxide (B6O) in which the preferred form of the material is as macroscopic, near-perfect, regular icosahedra, similar to the multiply-twinned particles observed in some cubic materials. A major difference is that B6O has a rhombohedral structure that nearly exactly fits the geometrical requirements needed to obtain icosahedral twins. These icosahedral particles have a structure that can be described as a Mackay packing of icosahedral B12 units, and thus has long-ranged order without translational symmetry.

Structural controls on the solubility of CO2 in silicate melts: Part I: bulk solubility data

Abstract CO 2 solubility data are presented for a wide range of melt compositions in the following systems: SiO 2 –Na 2 O–Al 2 O 3 +CO 2 (SNA) at 2.0 GPa and 1600°C; SNA+CaO (SNAC) and SNA+MgO+CaO (MSNAC) at 1.5 GPa and 1275–1400°C; and for several “natural” magma compositions (Mg- and Ca-rich melilitites, andesite and phonolite) at 1.2–2.7 GPa and 1300–1600°C. At a given pressure and temperature, the solubility is found to be a strong function of the “non-bridging oxygen” (NBO) content of the melt, expressed as the NBO/T ratio, where T represents tetrahedral network-forming cations. The NBO/T ratio, calculated from the melt composition, thus provides a useful parameter for expressing and predicting the CO 2 solubility. In highly polymerised melts, other dissolution mechanisms involving bridging oxygens become important and NBO/T is no longer the exclusive control on solubility. In Fe-bearing systems, the best correlation between CO 2 solubility and NBO/T is found when both Fe 3+ and Fe 2+ are assumed to be tetrahedral (T), indicating that these cations should be considered in a polymerising role in the melt, with respect to CO 2 dissolution. There is also evidence that some fraction of the Mg 2+ in the melt should be assigned to a polymerising role.

Al Coordination Changes in High-Pressure Aluminosilicate Liquids

Understanding the effect of pressure on aluminosilicate glass and liquid structure is critical to understanding magma flow at depth. Aluminum coordination has been predicted by mineral phase analysis and molecular dynamic calculations to change with increasing pressure. Nuclear magnetic resonance studies of glasses quenched from high pressure provide clear evidence for an increase in the average coordination of Al with pressure.