X-ray absorption spectroscopy is used to determine valence and site symmetry for all the 3d transition metals in fused quartz prepared by the flame hydrolysis method. The results are compared with optical data on the same samples with generally good agreement.

We have measured transition-metal ion (Ti, V, Cr, Mn, Fe, Co, Ni, Cu) spectra and their effects on Nd fluorescence quenching in Nd-doped phosphate and silicate glasses. Our purpose was to determine the maximum allowable impurity content given particular limits on the absorption loss at 1053 nm and the Nd fluorescence quenching rate. To keep the absorption loss <0.1 m−1 the transition-metal impurity content should be kept below 0.5 ppmw. To keep the increase in the Nd fluorescence decay rate below 1%, the impurity content should be <3 ppmw. We have also found that the Nd quenching rates do not scale as predicted by the Forster- Dexter dipole-dipole energy transfer theory if we assume that the dominant variation with transition metal is the overlap integral of the Nd fluorescence spectrum and the transition-metal absorption. We suggest that phonon-assisted energy transfer to transition metals is effective in quenching Nd. We find that quenching rates increase 1.5 to 4 times as the Nd concentration increases from 0.5 to 10 × 1020 cm−2.

Optical absorption spectra of d-d transitions can be interpreted through crystal field theory as evidence for at least four kinds of species: traditional network modifier ions (Mn2+, Fe2+, and Ni2+), octahedral complexes (Cr3+, Mn3+), tetrahedral complexes (Co2+, Fe3+), and network-forming ions (Ti4+, Fe3+). Absorption spectra measure only majority species since crystal field bands are broad and overlapping. Luminescence spectra reveal minority species if these species happen to be efficient emitters under laser excitation. Doublet-quartet emission bands (Cr3+) are independent of crystal field but allow the distinction between species because of small differences in bonding which shifts the Racah B parameter. Quartet-sextet emission (Mn2+, Fe3+) is strongly dependent on coordination number and bond lengths. Qualitative, but not quantitative, changes in speciation can be followed as glass composition is varied. Conceptually, the complexation of transition metal ions in insulator glasses results from a competition between the transition metal and the glass-forming framework for p-orbital electrons of non-bridging oxygens. Complexation is further enhanced in clustered glasses where the ions segregate into the more depolymerized units.

We have used time resolved fluorescence line narrowing to study optical dephasing and energy transfer among Yb3+ ions doped in inorganic glasses. We observe for the first time in any rare earth doped glass system a change in the power of the temperature dependence of the homogeneous linewidth of the 2F5/2(1)-2F7/2(1) transition of Yb, 3+ as well as an efficient one-phonon assisted energy transfer process occuring among Yb3+ ions at higher concentration due to a quadrupole-quadrupole coupling of the ions. In this paper we review past studies in rare earth doped gla s systems of these topics and compare them to our observations of Yb3+ in silicate and phesphate glasses.

We have investigated the electronic properties of films of arsenic triselenide into which the transition metal nickel has been introduced by a co-sputtering technique. Measurement of d.c. conductivity, thermopower, optical absorption and photomobility have been performed to characterise the material. These measurements give new information concerning the influence of nickel on the optical gap, the position of the Fermi-level, and other aspects of transport in As2 Se3. In the most heavily doped sample the d.c. conductivity is increased by 11 orders of magnitude and the position of the Fermi-level is shifted by ΔEF∼0.6 eV. The results are explained in terms of a model in which an acceptor level is introduced into the valence band tail of a background density of states (DOS) obtained earlier for the undoped material. In contrast to the suggestion by previous workers that the incorporation of high concentrations of nickel can produce a transition to n-type conduction, we argue that the Fermi-level remains in the valence band tail and that hopping conduction of carriers close the Fermi-level yields a negative sign in thermopower.

Color centers formed in the core and cladding of optical fiber waveguides by exposure to nuclear radiation can greatly increase the attenuation in the infrared spectral region of interest for optical communications. The radiation-induced paramagnetic defect centers in pure silica, silica doped with Ge, P, or B, and heavy metal fluoride glasses have been identified and thoroughly characterized by electron spin resonance (ESR) techniques. This paper will review the results of recent studies of color centers in optical fiber waveguide materials and their Identification via correlations of the radiation-induced optical absorptions and defect centers elucidated by ESR.

Paramagnetic color centers have been observed in germanium silicate and germanium phosphosilicate multimode optical fiber exposed to broadband ultraviolet light (2–5 eV). These centers are characterized by an ESR and optical absorption similar to 1 meV and 100 keV radiation induced defects and show an apparent saturation as the UV dose approaches 100 J/cm2. The UV induced ESR spectra are not identical to that induced by 60Co radiation however, similar Ge(2) and Ge(3) germanium defect signatures are apparent. For both compositions these centers are characterized by a rapidly increasing loss from 1.0 to 0.5 µm with an additional broad absorption peak at 1.5 µm for the phosphorus containing cores. We suggest that the UV induced optical absorptions for both compositions in the short wavelength range are due in part to the Ge(2) germanium substitutional sites and expect that the 1.5 µm absorption is due to the P1 phosphorus oxygen vacancy.

The formation of paramagnetic centers upon γ-irradiation was examined for the silica based waveguides doped with P2O5 by using ESR and optical absorption. The centers inherent in SiO2 glasses such as Si-E′ and OHC were found to be replaced with P-related centers such as phosphorus oxygen hole centers and phosphorus electron center by introducing a small amount of P2O5. New type of POHC was detected especially in the glass with [P2O5] of 1.3 mol % at 77K, which was assumed to be a precursor of the POHC stable at room temperature. The correlation between γ-induced loss-increase and the formation of these defects was examined.

ESR hyperfine structure produced by 73Ge of the drawing induced center in 10GeO2:90SiO2 or 30GeO2:70SiO2 fibers was detected and analyzed by taking into account site to site fluctuation in the glasses. The current assignment of the center to Ge-E′ in which an unpaired spin occupies one of sp3 hybrids of Ge was confirmed by the observation. The stress assisted radical cleavage of Ge-O bond was proposed to explain the enhancement of Ge-E′ formation in the fibers under quenching process. The hyperfine structure of the Y-induced center was also detected and similarly analyzed. New model structure was proposed for γ-induced center in which an electron is trapped on a Ge atom having four oxygen neighbors.

Electron paramagnetic resonance (EPR) spectroscopy has been used to identify paramagnetic intrinsic bonding defects and impurities in as-deposited thin solid SiO2 films. Thin films grown by E-beam vacuum deposition, RF sputtering, thermal oxidation of polysilicon, plasma enhanced chemical vapor deposition (PECVD), and low pressure chemical vapor deposition (LPCVD) techniques have been examined. Some of the growth techniques yield films that have paramagnetic centers similar to those found in bulk radiation-damaged vitreous SiO2. A new temperature dependent EPR center was observed in PECVD SiO2 films and has been assigned to trapped NO2. Slow-motional EPR lineshape theory was used to analyze the temperature dependent spectra.

The columnar structure in obliquely evaporated silicon oxide layers was investigated by transmission electron microscope (TEM). For TEM studies of these layers, samples were made by low angle ion-beam thinning of cross-sections, the planes of which were determined by the normal of the film and the direction of evaporation. Increasing the angle of evaporation from 5° to 30° (measured from the plane of the substrate), a change from a well–defined columnar structure to a striated structure was observed, for layers evaporated both under “low-rate” and “high-rate” conditions. There is a clear-cut dependence of the orientation of columns (αc) upon the angle of evaporation (α), however deviating from the “tangent rule” (tanαc=2tanα).

Doping effects of metal ions in chalcogenide glass films were studied. Doped impurities were found to act as donor or acceptor centers, when impurity ions were introduced in chalcogenide glass films below the glass transition temperature. Rectifying effects were realized in chalcogenide glass films doped with Al, Cd and Mg, so it is certain that the p-n junction was formed by the impurity doping. The tunneling of electrons through gap states, however, appeared to increase the reverse current of heavily doped diodes, so that large rectifying effects were not attained. A high density of defect states especially in the surface region would be responsible for low temperature diffusion of metal ions.

Defects in gel-derived glasses were reviewed for the three principal stages of the sol-gel process: solution, porous gel, and consolidated glass. The most prominent defects in gel-derived SiO2 are Raman active defects characterized by sharp bands at 490 cm−1(D1) and 608 cm−1(D2) 18O enrichment, experimental estimates of heats of formation, and structural studies of cyclic model compounds were used to identify the structural origins of D1 and D2 in gels. We conclude that D1 and D2 in gel-derived SiO2 are the oxygen breathing modes of 4- and 3-fold siloxane rings, respectively, as originally proposed by Galeener to explain these Raman-active defects in conventional vitreous silica (v-SiO2). During heat treatments near Tg (defined by n = 1013.5 poises), the gel-derived SiO2 structure rapidly approaches that of conventional v-SiO2.

The 9.5 GHz ESR-spectra of gamma-ray irradiated 94 SiO2 6 GeO2 (mol%) aero gels heat treated at 462°C were recorded at 300 K. The ESR-spectrum of this gel consisted of an asymmetric line (g1 = 2.001, g2 = 1.995, and g3 = 1.994) attributed to the Ge(3)-center and other components which have not yet been identified. The 94 SiO2 6 GeO2 gel was treated at 1000 and 1280°C and then exposed to gamma-ray radiation. The Ge(1)- and Ge(3)-centers were identified in the 94 SiO2 6 GeO2 gel with TH = 1000°C. At TH = 1280°C the signal of the Ge(1)-center dominated the spectrum of this sample. The spectra of the 58 SiO2 42 GeO2 gels with TH = 1200 and 1400°C can be described by the Ge(3)-center. The Ge(1)-center was of minor importance in these samples.

SiO2-gels produced with 4,8, and 16 mol water per mol TEOS were densified up to 1000°C and then exposed to gamma-ray irradiation in order to produce paramagnetic defect centers. Various paramagnetic states were detected during densification of these gels, the formation of which depends on the mol water per mol TEOS ratio. In the gels O−2 ions were detected. The paramagnetic centers detected in the n = 16 gel with TH = 900°C were attributed to a non-bridging oxygen and an E'1-center. Three different O−2 ions, probably in interstitial positions, were detected in the n = 16-gel with TH = 1000°C. The CO−2 center was observed in the n = 8 gel with TH = 900°C. A variety of new components occur at TH = 1000°C in the n = 8 gel, the origin of which remains unresolved at present. The CH3, CO−2, and E1′-centers were detected in the n = 4 gel with TH = 700°C. The CO−2 and E′1 centers were preserved after heat treating the n = 4 gel at TH = 1000°C.

All but two of the fifteen physical and chemical mechanisms which are necessary to explain all the varied causes of color apply in one way or another to glass. These fifteen causes of color derive from a variety of physical and chemical mechanisms and are summarized in five groups with concentration on those mechanisms that apply to glass and the related glazes and enamels. Vibrations and simple excitations explain the colors of incandescence (e.g. flames, hot glass), gas excitations (neon tube, aurora), and vibrations and rotations (blue ice, water, glasses based on water). Ligand field effect colors are seen in transition metal compounds (turquoise, chrome oxide green, glasses based on copper sulfate) and impurities (ruby, emerald, many doped glasses). Molecular orbitals explain the colors of organic compounds (indigo, chlorophyll, organic glasses) and charge transfer compounds (blue sapphire, lapis lazuli, “beer-bottle” brown and chromate glasses). Energy bands are involved in the colors of metals and alloys (gold, brass, glassy metals), of semiconductors (cadmium yellow, vermillion, chalcogenide glasses), doped semiconductors (blue and yellow diamond), and color centers (amethyst, topaz, irradiated glass). Geometrical and physical optics are involved in the colors derived from dispersive refraction (rainbow, green flash, glass prism spectrum), scattering (blue sky, blue eyes, red sunset, ruby gold and opal glasses), interference (soap bubbles, iridescent beetles, cracks in glasses, interference filters), and diffraction (the corona aureole, diffraction grating spectrum).

The thermal conductivities of beryllium fluoride glass and a fluoroberyllate glass have been measured in the 2–100K temperature range. These results are compared with earlier results on fluorozirconate glasses. The “plateau” region is well-defined for both materials. The plateau-center conductivity for BeF2 glass is the lowest yet recorded for any glass; it is some three times lower than the fluoroberyllate glass and four times lower than the fluorozirconates. A preliminary analysis of the data show a strong correlation between plateau-center conductivity and average cation mass of the material.

This study was conducted to investigate the possible occurrence of non-equilibrium defects and their influence on glass transition in a relatively depolymerized silicate structure. The glass of interest has an average bridging/non-bridging oxygen ratio of 1:2. It was prepared by fast quenching and was examined in the alumina pushrod dilatometer at heating rates of 5 and 15 ° C/min. The thermal expansion decreases with increase in the rate of heating although, the expansion coefficient is independent of heating rate at temperatures below the transition region. In the supercooled liquid region the expansion coefficient is higher at the higher heating rate. In the transition region it is both heating rate and temperature dependent. Three distinct temperatures are revealed in the transition region from temperature dependence of expansion rate. Their dependence on heating rate is described by a low activation energy of about 3Kcal/mol for transition. The single parameter fictive temperature is not in agreement with kinetics of relaxation due to a wide temperature interval of 600 to 750° C for the transition region. The transition kinetics is perhaps enhanced by the presence of defects whose annealing is impedded with increase in heating rate. Thus under the influence of higher defect concentration a spiked change in relaxation rate occurs and points to the presence of a critical temperature in the transition region. In the neighborhood of this temperature the relaxation rate of supercooled liquid is significantly lower than its solid analog. For short relaxation times in the supercooled liquid thermal expansion coefficient is relatively higher at the higher heating rate.

Due to the limited amount of structural information that can be obtained using traditional techniques of solid state physics (e.g., x-ray or neutron diffraction), the structure of oxide glasses has remained a controversial topic since the introduction of the “random network theory” in 1932. While there are several experimental methods that can provide detailed information about the local structural properties of glasses (e.g., NMR, EXAFS, optical absorption, Mössbauer absorption, and EPR spectroscopy), very little information is available regarding the distribution of intermediate size (∼10–500 Å) structural units.

The lack of general applicability of the many theoretical models for the mixed alkali effect (MAE) in glasses is briefly reviewed. Although the MAE appears to be related to the bonding affinity of alkali ions to charge compensating centers in the glass network, experimental scatter and the difficulty of comparing different glass networks have prevented systematic tests of this relationship. The present paper discusses why the mixed alkali germanate glasses should provide a glass system where the concentration and strength of charge-compensating centers can be systematically varied and the relationship to the MAE tested. Such tests are only possible if the ionic conductivity of a series of mixed alkali germanate glasses can be measured in a reproducible manner. The measurements of the ionic conductivity of two series of X(Na,Rb)2O:(1-X)GeO2 glasses, with X = 0.19 and X = 0.29, respectively, suggest the necessary reproducibility can be attained.