Open Access
Add to BibSonomyAbstract
Nd3+-doped bismuth titanate, (Bi4Ti3O12) (BiT) containing transparent glass-ceramics has been prepared by the melt-quenching followed by heat-treatment in the SiO2-K2O-Bi2O3-TiO2 (SKBT) system. The structural properties of the heat-treated GC are investigated using XRD, FE-SEM, TEM and FT-IR reflectance spectroscopy. The emission intensity of the 4F3/2→4I11/2 transition at 1065 nm is enhanced by 9-fold due to entering of Nd3+ ions (r = 0.98 Å) into the low-phonon crystalline phase (BiT), possibly by replacing Bi3+ ions (r = 1.02 Å) of the same valency. An increase of εr value (from 18.5 to 32.93) was observed with increase in heat-treatment time.
© 2014 Optical Society of America
1. Introduction
Nano/micro-crystallites dispersed glass–ceramics comprising of ferroelectric materials have been of immense interest from the dielectric application’s point of view [1]. The ferroelectric glass- ceramics are valued primarily for their high breakdown strengths (BDSs) which provide unique attribute for energy storage application [2]. Non-volatile memory devices are so named because they retain information when power is interrupted; thus they are important computer components. In this context, there has been considerable interest in developing non-volatile memories that use ferroelectric thin films—'ferroelectric random access memories', or FRAMs—in which information is stored in the polarization state of the ferroelectric material. To realize a practical FRAM, the thin films should satisfy the following criteria: compatibility with existing dynamic random access memory technologies, large remnant polarization (Pr) and reliable polarization-cycling characteristics. Early work focused on lead zirconia titanate (PZT) but, when films of this material were grown on metal electrodes, they generally suffered from a reduction of Pr ('fatigue') with polarity switching [3]. Moreover, use of PZT and other lead containing materials have been restricted because of existence of large amounts of toxic lead. These ferroelectric random-access memories are expected to replace magnetic core memory, magnetic bubble memory systems, and electrically erasable read-only memory for many applications [4].
For memory applications two varieties of ferroelectric oxides are considered, those are lead zirconia titanate (PZT) and bismuth layered structured materials (BLSM). Bismuth titanate, (Bi4Ti3O12) (BiT) is a bismuth layered structured material; its structure and properties have been described in details in our earlier papers [5,6] and elsewhere [7]. It has triple layers of Ti–O octahedra sandwiched between the (Bi2O2)2+ layers. The (Bi2O2)2+ layers make BiT fatigue-free because the layers have net electrical charges, and so their positioning in the lattice is self-regulated to compensate for space charges [3].
Ferroelectric BiT ceramics are fabricated using various techniques, such as, solid-state reaction process [8–10], co-precipitation [11–13] and molten salt synthesis [14]. These methods have disadvantages, such as, particle coarsening and aggregation, which result in inferior microstructure and poor properties of the BiT ceramics [15]. Mechanochemical synthesis process was also reported by Kong et.al [15] and others [7]. Nevertheless, one of the major difficulties of ceramics in piezoelectric applications is their high electrical conductivity, which interferes with the poling process [16,17]. Extensive studies were reported on the crystallization and dielectric behavior of ferroelectric GC, specifically PbTiO3 and NaNbO3 [18]. More recently, researchers reported the crystallization behavior, microstructure and dielectric properties of various ferroelectric GC such as, lead strontium titanate [19,20], PZT [21], barium strontium titanate [2,22,23] and Sr2TiSi2O8 [24].
Rare earth doped bismuth titanate thin films, such as (Bi,La)4Ti3O12 and (Bi,Nd)4Ti3O12, are regarded as one of the most potential materials for nonvolatile ferroelectric random access memories because of their high fatigue resistance, good retention and high remanent polarization value and fast switching speed. (Bi,Nd)4Ti3O12 thin films have been reported to exhibit excellent optical properties, such as large optical nonlinearity and high optical transparency in the visible wavelength region, which are attractive for the potential uses in optic and optoelectronic devices [25]. Highly transparent (Bi,Nd)4Ti3O12 ferroelectric thin films were prepared on indium-tin-oxide (ITO)-coated glass substrates and this multifunctional material has application potential in transparent electro optic devices [26].
However, very limited work has been reported [27–29] on the fabrication and characterization of BiT crystal dispersed GC in bulk quantity, despite its wide spread applications. Among the BLSM, BiT is a high temperature ferroelectric material (Tc = 675 °C) with useful properties for optical memory, piezoelectric, and electro-optic devices [30]. Because of this high transition temperature (ferroelectric to paraelectric), BiT ceramics are good candidates for high-temperature piezoelectric applications [31]. Shankar and Varma [27] reported the sharp piezoelectric resonances exhibited by the strontium tetraborate glasses containing bismuth titanate glass-ceramics in the 185 to 200 kHz frequency range which suggest that there is scope for exploitation of these materials for piezo-resonant electro-optic modulator applications, however, high temperature piezoelectric application of such glass-ceramics might be restricted because of their low Tg. Particularly, its relatively high remnant polarization (Pr ~25 μC/cm2) is favorable for nonvolatile random access memory devices and microwave applications [32]. Moreover, in BiT, bismuth is present as Bi3+ and as a trivalent cation it does not contribute to high ionic conductivity and thus do not interfere with the poling process [33]. It must be noted that some rare earth elements such as Nd, Pr, and Er in BiT GC, on the one hand, act as a structural modifier which greatly improves the electrical properties of material [34,35]; on the other hand, these elements can also act as the activator ions of luminescent materials [36,37].
However, to the best of our knowledge, the fabrication of Nd3+ - doped bismuth titanate glass-ceramic nanocomposites in bulk quantity by conventional process (melt quenching and ceramization) and its mechanical, microstructure, thermal, optical, dielectric and photoluminescence (PL) properties have not been reported so far. In this work, transparent Neodymium (Nd+3) doped bismuth titanate glass-ceramic nanocomposites have been prepared in bulk quantity by melt quenching followed by ceramization and their microstructure, thermal, optical, dielectric and photoluminescence properties have been evaluated and correlated with their ceramization conditions.
2. Experimental procedure
Glass with the molar composition of 55KS2-45BiT (where KS2 is potassium disilicate (K2O.2SiO2) and BiT corresponds to 2 Bi2O3. 3 TiO2 (Bi4Ti3O12)) doped with 0.5 mole % of Nd2O3 (in excess) were prepared from high-purity chemicals of SiO2 (99.8%, Sipur A1 Bremtheler Quartzitwerk, Usingen, Germany), potassium carbonate anhydrous, K2CO3 (GR, 99.9%, Loba Chemie, Mumbai, India), titanium (IV) oxide, TiO2 (99.3%, Merck KGaA, Darmstadt, Germany) and bismuth (III) oxide, Bi2O3 (99.0%, Loba Chemie, Mumbai, India) and Nd2O3 (99.9%, Alfa Aesar,Ward Hill, MA) by the conventional melt-quench technique. About 50 g of glass batch was mixed thoroughly by agate-mortar and melted in a platinum crucible in an electric furnace at 1050 °C for 2 h in air with intermittent stirring for homogenization of the glass melt. The glass melt was poured onto a preheated mould made of cast iron, followed by annealing at 450°C for 2 h to remove the internal stresses of the glass, and then slowly cooled down @1°C/min to room temperature. The as-prepared glass block (BiTNd0h) was shaped into desired dimensions by cutting and optically polished to carry out various characterization experiments. Transparent BiT GC has been derived from these glasses by controlled heat-treatment at 470°C for 5 h for nucleation and at 500°C with varying time of 5, 10, 15, 20, 25 and 30 h for growth of BiT crystals and samples are designated as BiTNd5h, BiTNd10h, BiTNd15h, BiTNd20h, BiTNd25h and BiTNd30h respectively.
The Archimedes’s principle was used to determine densities of the glasses and GC with water as the immersion liquid on a single pan electrical balance to the nearest 0.0001 g. The error in density measurement is estimated to be ± 0.004 gm/cm3. Differential Scanning Calorimetry (DSC) of the precursor BiTNd glass powders was carried out at temperatures up to 1000°C and at different heating rates of 5, 10, 15, and 20°C/min to determine the activation energy of crystallization (NETZSCH Model STA 449 Jupiter F3, NETZSCH-Gerätebau GmbH, Selb, Germany) as well as glass transition temperature (Tg), crystallization on-set and peak temperature (Tp). Tg, co-efficient of thermal expansion (CTE) and deformation temperature (Td) of the precursor glass sample has been evaluated using a horizontal dilatometer, NETZSCH DIL 402 PC (NETZSCH-Gerätebau GmbH, Germany).
XRD data were recorded using an Xpert-Pro MPD diffractometer (PANalytical, Almelo, The Netherlands) with the anchor scan parameters wavelength CuKα = 1.5406 Å at 25 °C, having a source power of 40 kV and 30 mA, for identifying the developed crystalline phases of the heat-treated glass–ceramics. A high-resolution FESEM (Gemini Zeiss Suprat 35 VP model of Carl Zeiss Microimaging GmbH, Berlin, Germany) was employed to examine the microstructure of the heat-treated glass–ceramics after etching in HF solution and coating with a thin carbon film. The TEM images of the powdered glass–ceramic samples were obtained from an FEI (Model Tecnai G2 30ST, FEI Company, Hillsboro, OR) instrument. The FTIR transmittance spectra were recorded using a FTIR spectrometer (Model 1615, Perkin Elmer, Norwalk, CT), in the wavenumber range of 400–2000 cm−1 at a resolution of ± 2 cm−1 after 16 scans. The refractive index (n) of the precursor glass was measured by a Prism Coupler (Model 2010/M, Metricon Corporation, New Jersey, USA) at five different wavelengths of 473, 532, 632.8, 1064 and 1552 nm equipped with respective laser sources. Further, these data have been used to estimate n at other wavelengths employing Cauchy dispersion fittings. The resolution of the prism coupler is ± 0.0005. Several optical properties such as Abbe number (υD), molar refractivity RM, Optical polarizability (αm) and third order susceptibility has been estimated. The optical absorption spectra were recorded on a Perkin Elmer UV-Vis-NIR spectrophotometer (Model Lambda 20, Perkin-Elmer Corporation, Waltham, MA, USA) in the wavelength range of 400-1100 nm with an accuracy of ± 1%. The fluorescence emission and excitation spectra were measured using a bench-top modular spectrofluorometer (QuantaMaster, Photon Technology International, Birmingham, NJ) attached with a Xe arc lamp as the excitation source. All the measurements were carried out by placing the samples at 60° to the incident beam and the signals were collected at right angle to the incident beam from same surface. Capacitance and dielectric loss of the glasses and glass–ceramic nanocomposites were measured at room temperature using a Hioki LCR meter (Model 3532-50 Hitester, Hioki, Ueda, Nagano, Japan) at 1 MHz frequency after coating the surfaces with a conductive silver paint (here silver acts as an electrode) followed by drying at 140°C for 1 h. The dielectric constants (εr) were calculated from the dimensions of the sample and the measured capacitance. The accuracy of the LCR meter is 0.08%.
The hardness and indentation fracture toughness properties of the GC samples were measured by taking micro-indentation on the polished surface of the samples. A micro indentation hardness testing system (Clemex CMT, Canada) equipped with a conical Vickers indenter was used for glasses and GCs at a load of 50 g. Ten indents were taken for each sample under identical loading conditions. The diagonals of the Vickers indents were measured using an optical microscope attached to the equipment and subsequently, the hardness was calculated using the standard equation for the Vickers geometry.
Where HV is the Vickers hardness number (VHN) in kg/mm2, P is the normal load in kg, and d is the average diagonal length of the indentation in mm. The following equation, proposed by Antis et al. [38], was used for calculation of the fracture toughness (KIc) from the cracks generated during indentation:Here, P is the applied load and C is the mean length of the two radian cracks. E and H are the Young's modulus and hardness of the sample, respectively, measured at the same applied load P at which the KIc for the sample is measured. For Vickers indenters the value was taken as 0.016.3. Results and discussion
3.1 Thermal and physical properties
A DSC study is important for determining the heat-treatment conditions required to convert a glass system into a glass-ceramic with the desired crystalline phases. DSC trace for Nd3+ - doped BiT glass powders were obtained at four different heating rates i.e. 5, 10, 15 and 20 and is provided in Fig. 1.Glass transition (Tg) temperatures were found to vary from 461 to 474°C when heating rates were varied from 5 to 20 °C/min. Tg obtained through the DSC is found identical with the Tg (470°C) determined using a dilatometer. The Tg temperatures so evaluated through these experiments are found to increase with respect to the un-doped BiT glass (450°C) [5] which is attributed to the higher melting temperature (2233°C) of the dopant, Nd2O3. Only, one prominent exothermic peak has been observed in the DSC curve which confirms the crystallization of a single phase.
From the DSC trace it has also been observed that melting of glass powder starts at around 764°C (Tm). Coefficient of thermal expansion (CTE) over the temperature range of 50-300°C and dilatometric softening temperature (Td) of the glass was determined from the dilatometric curve as shown in Fig. 2 and found as 11.90 x 10−6/°C and 500°C respectively. Over all, the glass transition temperature (Tg) and the dilatometric softening temperature (Td) observed for the BiT glass is low due to the presence of large amount (18 mol %) of low melting Bi2O3 (M. P. 824°C). The densities of BiT glasses and GC studied in the present work are recorded as 4.783 for the precursor glass and 4.790 - 4.796 g. cm−3 for the GC heat-treated for 5-30 h at 500°C. With an increase in heat-treatment duration, the density of the GC increases monotonically by a small amount which is attributed to the densification of the GC with progression of crystallization (see Table 1). The high densities of these glass and GC (4.788 - 4.796 g.cm−3) are because of the existence of large amount of high density Bi2O3 (8.90 g. cm−3).
Table 1. Some measured and calculated properties of Nd3+:BiT precursor glass and glass-ceramics
3.2 Kinetic study
The activation energy for crystallization of glass was calculated using different models which were proposed by various researchers. Some of these popular models are provided below:
Kissinger equation [39],
Augis-Bennett equation [40],
Ozawa equation [41],
where, α is rate of heating, Tp is crystallization peak temperature, R is gas constant, Ec is the activation energy of crystallization and C is a constant.A plot of ln (α/Tp22) versus 1000/Tp2 (K−1), based on the Kissinger model, is shown in Fig. 3.The corresponding activation energy has been estimated as 438 kJ/mol. From the Augis-Bennett model plot of ln (α/Tp2) versus 1000/Tp2 (K−1) and the Ozawa model plot of ln α versus 1000/Tp2 (K−1), activation energies of 446 kJ/mol and 453 kJ/mol were determined respectively. These activation energies match closely to the dissociation energy of Ti-O (435 kJ/mol) [42] which support the formation of BiT in the resultant GCs.
Fig. 3 (a) Plot of ln (α/Tp2) as a function of 1000/Tp (K−1), Kissinger model, (b) ln α as a function of 1000/Tp (K−1), Augis & Bennett model, and (c) ln (α/Tp) vs. 1000/Tp (K−1), Ozowa model.
The kinetics of crystallization is driven by nucleation and growth steps. Nucleation and crystal growth rates can be measured from microscopic images but DTA/DSC study is also valuable for the quantitative evaluation of crystallization kinetics in different glassy systems [43]. The glass stability (ΔT = Tx-Tg) factor has also been estimated using DSC data. The ΔT values at different heating rates are about 100°C, indicating that these glasses are reasonably stable.
3.3 XRD analysis
XRD studies were performed on the Nd3+ - doped BiT glass powders heat treated at 500°C with varying time of 5, 10, 13, 16, 20 and 30 h to determine the crystal phases evolved after ceramization. XRD patterns of the heat-treated glass-ceramic samples have been shown in Fig. 4.The diffraction patterns are compared and found matching with the JCPDS file card no. 35-0795 which is corresponding to the bismuth titanate (Bi4Ti3O12). It has been determined that with increase in heat-treatment time there was no signature for development of any additional peaks. XRD peaks are observed broader for all the samples, which is an indication of formation of crystals of nano size. However, sharp peaks are observed for the GC heat-treated for 50 and 100 h. The GC heat-treated for more than 10 h has been found to be opaque in appearance (please see Fig. 8). This indicates that large sized crystals are generated in the samples subjected to longer heat-treatment duration. To confirm these findings further studies were conducted using FE-SEM and TEM which are discussed in the subsequent section.
Fig. 4 XRD pattern of Nd3+ - doped BiT glass-ceramics heat-treated at 470°C for 5 h and at 500°C for different duration of 5 – 100 h.
3.4 FT-IR reflectance spectroscopy analysis
Figure 5 shows the FTIR reflectance spectra of the as-prepared glass and the glass-ceramic samples, heat-treated at 500°C for varying time durations in the wave number range of 500–4000 cm−1. The bands observed at 900-1000 cm−1 of reflectance spectra is due to the presence of Si-O-Si bonds. It has also been observed that with increase in BiT crystal content in the GC, this Si-O-Si band intensity is reduced and this is attributed to the fact that the amount of SiO2 in the GC decreases with increase in BiT content, that is, with increase in duration of heat-treatment time. It is also evident that with increase in BiT content, the intensity of reflectance at around 500 cm−1 increases. The presence of absorption peaks due to the stretching vibrations of Bi-O bonds in strongly distorted BiO6 octahedral units at around 630 cm−1 in the FTIR absorption spectra were reported by Ardelean et. al [44]. This study evidently indicate that, this occurrence may be attributed to the fact that with augment in BiT content, the extent of Bi2O3 in the GC increases, which leads to increase in the intensity of reflectance at around 630 cm−1. The reflection band at around 740 cm−1 is due to Ti-O bond or TiO4 tetrahedra. The bands of FT-IR reflectance spectra have been assigned and provided in Table 2.
Fig. 5 FT-IR reflectance spectra of Nd3+ - doped BiT precursor glass and glass-ceramics heat-treated at 470°C for 5 h and at 500°C for different duration of 16 and 30h.
Table 2. Band assignment of FT-IR spectra
3.5 FE-SEM and TEM analysis
Field emission scanning electron microscopy (FE-SEM) and transmission electron microscopy (TEM) studies were carried out to investigate the nucleation and crystallization process. FE-SEM micrographs of Nd3+ - doped BiT GC heat-treated at 500°C for varying time of 5-30 h are shown in Fig. 6.The GC was etched with 10% hydrofluoric acid for 60 seconds prior to FE-SEM observations. The etching procedure resulted in the solution of a few microns of the glassy matrix in the samples, leaving the crystals protruding from the surface and giving a high contrast. The glass-ceramic displayed a fine microstructure with many BiT crystals of uniform size homogeneously dispersed in the glass matrix. The crystals are spherical in nature for all the samples. However, it is clear from the images that with increase in heat-treatment duration both the crystal volume fractions as well as the size of the crystals are increased substantially. FE-SEM images of GC heat-treated for shorter time span of 5 h show that crystals are not distinctly visible may be due to the presence of lesser volume of crystals. However, with increase in heat-treatment time i.e. 10 h or more crystals become more distinctly visible and are bigger in size. The average size of the crystals are about 200-300 nm for the samples heat-treated for the period of 10 and 15 h. Such nano-sized crystals explain the transparency of the GC when heat-treated for shorter time duration (<10 h). Whereas, the samples heat-treated for more than 15 h at 500°C resulted in higher crystal sizes and content, this affected the transparency of the GC, and becomes opaque. Researchers reported sheaf-shaped, rod like [45], platelets [46] crystals of dimension in the micrometer range, crystals of platy morphology [47] with the dimension in the nanometer range has also been reported for this glass system.
Fig. 6 FE-SEM images of Nd3+ - doped BiT glass-ceramics heat-treated at 470°C for 5 h and at 500°C for different duration of (a) 5, (b) 10, (c) 15, (d) 20, (e) 25, and (f) 30 h.
In our earlier paper [5], nano-rods morphology was reported. However, in the present study only spherical morphology of crystals are evident where crystals are homogeneously distributed throughout the glass matrix.
Bright field TEM images of BiT GC along with respective SAED patterns are shown in Fig. 7.These images reveal presence of nano crystallites of average size 10-15 nm in GC heat treated for 10, 15, 20 and 25 h. The BiT crystallites are generated by employing heat-treatment on the precursor glass. The thermodynamic driving force of the glass-crystal transition is the chemical potential or free energy between the separated phase and the crystal. In the present case, heterogeneous nucleation was catalyzed using refractory TiO2 (melting point = 1850°C) particles. Two distinct phases were evolved during nucleation, one is TiO2 and Bi2O3 rich phase and other one is SiO2 and K2O rich phase [5]. During heat-treatment, TiO2 and Bi2O3 rich phase was grown to form Bi4Ti3O12 polycrystals out of the BiT crystallites. The nano-crystalline nature of the material is confirmed from the diffused hallow of SAED pattern obtained from the glass-ceramic. Further, indexing of the SAED pattern (shown as inset in Fig. 7) confirms that crystals developed after ceramization for 20 h are BiT. The major (hkl) planes obtained from the pattern are (171) and (280) and these are matching with the JCPDS file no. 35-0795 which corresponds to BiT crystals.
Fig. 7 TEM images and SAED patterns of Nd3+ - doped BiT glass-ceramics heat-treated at 470°C for 5 h and at 500°C for different duration of (a) 10, (b) 15, (c) 20, and (d) 25 h.
3.6 Optical properties
The images of the as prepared glass and heat-treated glass-ceramics have been presented in Fig. 8. The refractive index (n) of the glass and glass-ceramic samples are measured and presented in Table 1. The plot of n vs. wavelength for the precursor glass has been shown in Fig. 9. RI of the samples measured at 632.8 nm was found to be very high and increases from 1.93502 for the precursor glass to 1.93529 for the glass-ceramics heat-treated for 30 h. This increase in RI with increasing Bi2O3 and TiO2 (i.e. BiT) content in the glasses occurs mainly because of the more dense structure of the glass obtained and also due to the incorporation of more polarisable as well as more ionic refraction of Bi and Ti ions present in the glass.
To understand better the effect of ceramization on the RI, the empirical relation derived by Lorentz [48] and Lorenz [49] is used to estimate the molar refraction (RM).
Where, ρ and M are the density and molecular weight of the glass respectively.Molar refraction, RM is related to the structure of glass and directly proportional to the optical polarizability [27] of the material, αm, according to the following relationship:
where N is the number of polarizable ions per mole, assumed to be equal to the Avogadro number.The values of RM and αm were estimated using Eqs. (6) and 7 and are provided in Table 1. It shows that with increase in duration of ceramization (i.e., heat-treatment) the optical polarizability of the glass changes marginally as there is hardly any variation in the n of the samples, since there is no composition change. These values are in agreement with findings of the previous researchers [47]. The linear and nonlinear optical properties of the glasses are determined by the bond polarizabilities [50]. This is due to the interaction of the propagating light with the electronic charge distribution in the glass structural units. The Bi3+ ion belongs to the group of ions with one lone-pair electron configuration [45]. This loan-pair electron exists at the apex of the pyramidal structure and easily polarizes on applying electric field and or in interaction with light. Researchers [50] showed with Raman spectroscopy study that Ti does not behave as simple network modifiers but rather behave as intermediates. They also established that the optical properties of multi-component oxide glasses are more influenced by the concentration of the transition metal cations rather than by the number of non-bridging oxygen. The nonlinear index in the present system is controlled by the nonlinear bond polarizability of the Ti-O bonds. It has also been reported that ions with an empty or unfilled d shell (e.g. Ti) contribute greatly to the linear and nonlinear polarizabilities [50].
The Abbe number (υd) obtained for the glass and glass-ceramic samples demonstrates that with increase in ceramization time, the υd of the glass is decreases. Abbe numbers were calculated from a well-known empirical relation using RI of light at different wavelengths, the values thus obtained are presented in Table 1. It is evident that glass and glass-ceramics studied here belong to the family of high index and high dispersion comparable to the high lead bearing glasses.
In treating nonlinear optical phenomena in glasses a distinction can be made between phenomena such as the nonlinear refractive index, multiphoton absorption, and stimulated Raman and Brillouin scattering, where glass itself is responsible for the nonlinearity (intrinsic effect), and those cases such as the optical nonlinearities in glasses containing rare earth ions, semiconductor microcrystals, or organic dyes, where the nonlinearity is associated with dopant and the glass acts as a host (extrinsic effect). For intrinsic effects the structure and composition of the glass are of primary importance in controlling the nonlinearity, whereas for extrinsic dopant effects the influence of the glass matrix is secondary. Hence, the selection of glass composition and dopants provides a wide range of possibilities for controlling and tailoring nonlinear optical properties for a multitude of applications.
In order to understand the role of dopants, phases present and matrix glass, an attempt was made to estimate the third order susceptibility, χ3, using an indirect method. According to Boling’s [51] semi empirical equation, χ3 of a materials is strongly dependent on both the nonlinear refractive index (n2) and linear refractive index (n). χ3 is expressed by the following relation, according to Vogel et al. [50].
where n is refractive index of glass at 587.6 nm and n2 is non-linear refractive index estimated from an empirical relation derived by Vogel et al. [50]. The χ3 values obtained for the precursor glass and glass-ceramics are presented in Table 3 and values for other glasses are also presented in the table for the sake of comparison. The susceptibility values estimated from the above Eq. (8) shows that with increase in BiT content in the glass, the χ3 values increases steadily from 4.82 for the glass to 6.589 for the glass-ceramics heat-treated for 30 h. About 37% increase in χ3 values could be observed. As the dopant (Eu2O3) concentration and the matrix glass composition remained unaltered, the increase in χ3 values is attributed to the increase in BiT crystal phase. The values thus obtained were significantly higher than the value reported for tellurite glasses by Prakash et al. [52] and Takebe et al. [53] and lower than Nb2O5 doped tellurite glasses reported by Canioni et al. [54] and much lower than the chalcogenide and semiconductor doped glasses. In silicates, phosphates, tellurites, and halides glass where systematic changes in the number and type of network modifier ions are used to increase or decrease optical nonlinearities. Chalcogenide glasses and inorganic glasses containing organic dye molecules and semiconductor microscrystals are examples of materials exhibiting much larger nonlinearities (Table 3). The materials exhibit nonlinear properties due to the distortion of electronic structure of any atom, complex, or molecule by the application of an electric field to produce net dipole moment that is linearly or nonlinearly proportional to that field.Table 3. Calculated third-order nonlinear susceptibilities (χ3) of Nd3+:BiT precursor glass and glass-ceramics and other glasses reported in the literature
The room temperature measured absorption spectra of the Nd3+-doped precursor glass (a) and heat-treated glass-ceramic samples in the visible-NIR range have been presented in Fig. 10.The spectra reveal absorption peaks due to the 4f 3-4f 3forced electric dipole transitions from the ground 4I9/2 state to different excited states of Nd3+ ion in 4f 3 configuration. All the peaks 4I9/2→ 4G9/2 (512 nm), 2K13/2 + 4G7/2 + 4G9/2 (526 nm), 4G5/2 + 2G7/2 (583 nm), 2H11/2 (626 nm), 4F9/2 (679 nm), 4F7/2 + 4S3/2 (739 nm), 4F5/2 + 2H9/2 (806 nm) and 4F3/2 (880 nm) are assigned in accordance with Carnall’s convention [31,32]. From this figure it is noticed that the base line of absorption spectra of heat-treated samples has been elevated significantly with the diminishing intensities of the absorption peaks. This uplifting can be attributed to scattering of short wavelength light by the crystals [33,34]. Since the crystallites (10-15 nm) are smaller than the visible wavelength, a Rayleigh scattering model should be applicable [35]. According to this model, the scattering loss, τ is given by
where d is the particle diameter, λ the wavelength of light, n the refractive index, N the number density of particles, and V the volume of the particle. With progression of heat-treatment, the number and sizes of nano crystallites developed in the glassy matrix increase and hence the scattering centre and scattering loss increase that corresponds to a decrease in the visible transparency of the glass-ceramics samples as evident from Fig. 8. There is no significant difference in the shapes of absorption bands for glass and crystallized sample. During heat-treatment of the glasses, the hypersensitive transition 4I9/2 → 4G5/2, 2G7/2 (obeying selection rule: ΔJ ≤ 2, ΔL ≤ 2 and ΔS = 0 with large value, 0.898 of squared reduced matrix element, |U(2)|2) of Nd3+ ion lying around 590 nm has been selected to understand the environment change closely around the rare-earth ion which is shown in Fig. 10.Fig. 10 Absorbance spectra of Nd3+ - doped BiT precursor glass and glass-ceramics heat-treated at 470°C for 5 h and at 500°C for varying duration of 5-30 h with the interval of 5h.
3.7 NIR-excited NIR fluorescence and lifetime
The infrared fluorescence spectra (λex = 590 nm) of the samples around 1065 nm are shown in Fig. 11.Nd3+ ion having the higher absorption cross-section at 590 nm wavelength and moreover, the band gap of Bi4Ti3O12 is around ~2.4 eV, so we have chosen the 590 nm as excited wavelength to understand the contribution of these crystals in the enhancement of Nd3+ emission at 1065. The emission band intensity around 1065 nm increases with heat-treatment time. The spectra of the Nd3+-doped glass and the GCs exhibit three emission peaks at 906, 1065 and 1340 nm. These three emission peaks can be attributed to the 4F3/2→4I9/2, 11/2, 13/2 transitions of Nd3+. The emission intensity of the 4F3/2→4I11/2 transition is enhanced by an increase in heat-treatment time and increases about 9-fold in the GC samples heat-treated for 30 h compared to the precursor glass. The increased emission intensity suggests that the Nd3+ ions entered into the low-phonon crystalline phase (BiT), possibly by replacing Bi3+ ions of the same valency. The Nd3+ ion prefers to replace Bi3+ rather than Ti4+ due to the similarity of their ionic radii (Nd3+ = 0.98 Å, Bi3+ = 1.02 Å and Ti4+ = 0.61 Å) and charges. In the present case, the Nd3+-Nd3+ ionic separation (Ri) in the precursor glass is found to be about 21.66 Å which was calculated using the relation [60]:
where NNd3+ is the Nd3+ ion concentration as already provided in Table 1.Fig. 11 Photoluminescence emission spectra (λex = 590 nm) of Nd3+ - doped BiT precursor glass and glass-ceramics heat-treated at 470°C for 5 h and at 500°C for varying duration of 5-30 h with the interval of 5h.
The room temperature fluorescence decay curves of the emission transition (4F3/2→4I11/2) at 1065 nm with an excitation at 590 nm for Nd3+ ions in as-prepared glass and glass-ceramic nanocomposites have been depicted in Fig. 12.The measured curves demonstrate a single exponential decay. The excited state lifetime (τ) for all the samples has been estimated from these decay curves and the results are shown in the inset of Fig. 12. It is seen that the excited state (4F3/2) lifetime (τ) increases marginally with increase in heat-treatment duration. The increase in life time can be attributed to incorporation of Nd+3 in the low phonon energy crystal sites.
Fig. 12 Decay curves for the 4F3/2→4I11/2 transition of Nd3 + ion at 1065 nm under excitation at 590 nm of Nd3+ - doped BiT precursor glass and glass-ceramics heat-treated at 470°C for 5 h and at 500°C for varying duration of 5-30 h with the interval of 5h. (Inset shows the life time of the individual samples).
3.8 Dielectric property
Glass and glass–ceramics have certain advantages as dielectric materials because of their high dielectric strength. However, the disadvantages of glass are a low permittivity (εr = 4–15) and a low thermal conductivity [61]. In the present investigation, the dielectric constant (εr) has been measured for as-prepared BiT glass is 18.5 which is relatively higher than common glasses such as vitreous silica (3.8), soda-lime silicate (7.2) or borosilicate glasses (4.1–4.9) [62] and similar to that of the Li2O–Ta2O5–SiO2–Al2O3 glasses [63] studied recently in our group. The change in dielectric constant of GC heat-treated for varying time duration is depicted in Fig. 13.It is observed that with increase in heat-treatment time the dielectric constant of the BiT GC increases. A very marginal increase of εr value (from 18.5 to 20.68) was observed for the corresponding GC heat-treated at 500°C for 5 h and 30 h. However, when the GC is heat-treated for longer durations of 50 and 100 h, a huge increase in εr values is observed which is 29.01 and 32.93 respectively. This increase in εr value confirms that with increase in heat-treatment time at an elevated temperature the quantity of Bi4Ti3O12 crystals in the GC increases. Hence, this increase in εr values with increase in heat-treatment time at a particular crystallization temperature is expected as the crystals developed after heat-treatment is (Bi4Ti3O12) crystals and with increase in durations of heat-treatment time, quantity of such crystals are increased. This increase in εr value can be anticipated because polycrystalline Bi4Ti3O12 ceramics has higher εr (~150) values compared to their base glass (18.5) [64]. Hence, when the glass is subjected to a heat-treatment for ceramization and a fair amount of glass is converted in to GC containing Bi4Ti3O12 phase, the εr value is likely to increase. This is due to the very high ionic refraction (Bi = 30.5 and Ti = 19.0) and polarizability [65] (Bi = 1.31 and Ti = 0.46) of Bi3+ and Ti4+ ions present in the material. The increase in εr with increasing BiT content is attributed mainly to the increase in interfacial polarization caused by the presence of BiT crystallites in the glassy matrix.
Fig. 13 Dielectric constant (εr) of BiT glass and glass-ceramics heat-treated at 470°C for 5 h and at 500°C for varying duration of 5-100 h.
3.9 Hardness and fracture toughness
The mechanical properties i.e., hardness and indentation fracture toughness of the Nd2O3 doped glasses and GCs have been evaluated with a Vickers indenter at indent loads of 50 g using the average diagonal lengths of the hardness impressions. A few representative images of indentation impressions are provided in Fig. 14.At 50 g load, cracks were generated both in the glass and glass-ceramic samples. These crack lengths were used to estimate fracture toughness values. The average hardness value obtained for the BiTNd glass is 4.34 GPa. Hardness values obtained for the GCs heat-treated for varying durations of 5, 10, 15, 20, 25 and 30 h are 4.42, 4.52, 4.63, 4.67, 4.75 and 4.68 GPa respectively. The hardness values of the GCs increase marginally with increase in heat-treatment time.
Fig. 14 Representative images of the indentation impressions taken on Nd3+ - doped BiT precursor glass and glass-ceramics heat-treated at 470°C for 5 h and at 500°C for varying duration of 5, 20 and 30 h observed under optical microscope (load = 50 g).
Indentation fracture toughness values were calculated by measuring the average lengths of the cracks using Antis’s equation [38]. This equation is based on the assumption that nucleation of the cracks is due to an elastic stress field formed beneath the indenter. Though crack growth in elastic/plastic contact fields has been established as the basis for fracture studies, there is still sometimes a dominant residual component of the driving force which leads to an increase in the crack length even after completion of the actual indentation process [66].
The sample BiTNd glass had an indentation fracture toughness (K1c) value of 1.2 MPa m0.5 while K1c values between 1.06 and 1.71 MPa m0.5 were found for heat-treated GCs. The average fracture toughness values obtained in the GCs are higher than the precursor glass. These improvements in K1c are attributed to the development of nano crystals in the glass matrix as seen in the FE-SEM micrographs.
4. Conclusions
The effect of heat-treatment duration on Bi4Ti3O12 (BiT) content in the Nd3+ - doped K2O-SiO2-Bi2O3-TiO2 glass system and its various properties have been demonstrated by using various characterization tools such as, DSC, dilatometer, XRD, FE-SEM, TEM, prism coupler, spectrophotometer and FT-IRRS. The important conclusions are summarized as:
- By using DSC experimental data in three different models such as Kissinger, Augis-Bennett and Ozawa, activation energy for crystallization has been estimated as 438, 446 and 453 kJ/mol respectively which is similar to the dissociation energy of Ti-O bonds which confirms the formation of Bi4Ti3O12 crystals.
- The glass-ceramic displayed a fine microstructure with many BiT crystals of uniform size homogeneously dispersed in the glass matrix. The average grain size of the crystals of the heat-treated samples are about 200-300 nm and TEM images reveal nanocrystallites of average size 10-15 nm.
- Systematic increase in refractive index and decrease in optical transmission properties are observed which is attributed to the Rayleigh scattering that is caused due to the increase in nanocrystals content in the system with increase in heat treatment duration.
- Third order susceptibility, χ3 (10−14 esu) values have been found to increase with heat-treatment duration from 4.82 to 6.59 for the precursor glass and glass-ceramics heat-treated for 30 h indicating an increase in non-linearity in the glass-ceramics.
- The emission intensity of the 4F3/2→4I11/2 at 1065 nm transition is enhanced by 9-fold in the GC samples heat-treated for 30 h compared to the precursor glass. The increased emission intensity suggests that the Nd3+ ions entered into the low-phonon crystalline phase (BiT), possibly by replacing Bi3+ ions of the same valency and due to the similarity of their ionic radii (Nd3+ = 1.12 Å, Bi3+ = 1.02 Å and Ti4+ = 0.61 Å) and charges.
- A marginal increase of εr value (from 18.5 to 20.68) was observed for the corresponding precursor glass and GC heat-treated at 500°C for 30 h. This increase in εr value confirms that with increase in heat-treatment time at an elevated temperature the quantity of Bi4Ti3O12 crystals in the GC increases.
- The hardness and fracture toughness values are found to increase in the glass-ceramics.
Acknowledgment
The authors thank Shri Kamal Dasgupta, Director CSIR-CGCRI and Dr. Ranjan Sen, Head, Glass Division for their encouragement to carry out this work and financial support (OLP 0369). The authors would like to thank Dr. K. Annapurna and Dr. K. Biswas, Scientists of CSIR-CGCRI for their help in recording the photoluminescence and refractive indices respectively. The authors also thankfully acknowledge the XRD and Electron Microscope Sections of this institute for their help in recording XRD patterns and microscopic images respectively.
References and links
1. J. Chen, Y. Zhang, C. Deng, and X. Dai, “Crystallization kinetics and dielectric properties of barium strontium titanate based glass–ceramics,” Mater. Chem. Phys. 121(1-2), 109–113 (2010). [CrossRef]
2. J. Huang, Y. Zhang, T. Ma, H. Li, and L. Zhang, “Correlation between dielectric breakdown strength and interface polarization in barium strontium titanate glass ceramics,” Appl. Phys. Lett. 96(4), 042902 (2010). [CrossRef]
3. B. H. Park, B. S. Kang, S. D. Bu, T. W. Noh, J. Lee, and W. Jo, “Lanthanum-substituted bismuth titanate for use in non-volatile memories,” Nature 401(6754), 682–684 (1999). [CrossRef]
4. J. F. Scott and C. A. Paz de Araujo, “Ferroelectric memories,” Science 246(4936), 1400–1405 (1989). [CrossRef] [PubMed]
5. A. R. Molla, A. Tarafder, and B. Karmakar, “Synthesis and properties of glasses in the K2O–SiO2–Bi2O3–TiO2 system and bismuth titanate (Bi4Ti3O12) nano glass–ceramics thereof,” J. Mater. Sci. 46(9), 2967–2976 (2011). [CrossRef]
6. A. R. Molla, A. Tarafder, S. Mukherjee, and B. Karmakar, “Transparent Nd3+-doped ferroelectric bismuth titanate glass-ceramic nanocomposites: fabrication and properties,” J. Am. Ceram. Soc. 95(10), 3056–3063 (2012). [CrossRef]
7. Z. Ž. Lazarević, B. D. Stojanović, M. J. Romčević, and N. Ž. Romčević, “Mechanochemical activation assisted synthesis of bismuth layered-Perovskite Bi4 Ti4O12,” Sci. Sin. 41(1), 19–26 (2009). [CrossRef]
8. H. S. Shulman, M. Testorf, D. Damjanovic, and N. Setter, “Microstructure, electrical conductivity and piezoelectric properties of bismuth titanate,” J. Am. Ceram. Soc. 79(12), 3124–3128 (1996). [CrossRef]
9. M. Villegas, A. C. Caballero, C. Moure, P. Duran, and J. F. Fernandez, “Factors affecting the electrical conductivity of donor-doped Bi4Ti3O12 piezoelectric ceramics,” J. Am. Ceram. Soc. 82(9), 2411–2416 (1999). [CrossRef]
10. N. Zhu and A. R. West, “Formation and Stability of Ferroelectric BaTi2O5,” J. Am. Ceram. Soc. 93(1), 295–300 (2010). [CrossRef]
11. A. Q. Jiang, H. G. Li, and L. D. Zhang, “Dielectric study in nanocrystalline prepared by chemical doprecipitation,” J. Appl. Phys. 83, 4878–4883 (1998). [CrossRef]
12. S. H. Hong, J. A. Horn, S. Trolier-McKinstry, and G. L. Messing, “Dielectric and ferroelectric properties of Ta-doped bismuth titanate,” J. Mater. Sci. Lett. 19(18), 1661–1664 (2000). [CrossRef]
13. J. A. Horn, S. C. Zhang, U. Selvaraj, G. L. Messing, and S. Trolier-McKinstry, “Templated grain growth of textured bismuth titanate,” J. Am. Ceram. Soc. 82(4), 921–926 (1999). [CrossRef]
14. T. Takeuchi, T. Tani, Y. Saito, and T. TakNdchi, “Unidirectionally textured CaBi4Ti4O15 ceramics by the reactive templated grain growth with an extrusion,” Jpn. J. Appl. Phys. 39(Part 1, No. 9B), 5577–5580 (2000). [CrossRef]
15. L. B. Kong, J. Ma, W. Zhu, and O. K. Tan, “Preparation of Bi4Ti3O12 ceramics via a high-energy ball milling process,” Mater. Lett. 51(2), 108–114 (2001). [CrossRef]
16. A. Fouskova and L. E. Cross, “Dielectric properties of bismuth titanate,” J. Appl. Phys. 41(7), 2834–2838 (1970). [CrossRef]
17. M. Villegas, A. C. Caballero, and J. F. Fernandez, “Modulation of electrical conductivity through microstructural control in Bi4Ti3O12-based piezoelectric ceramics,” Ferroelectrics 267(1), 165–173 (2002). [CrossRef]
18. T. Kokubo and M. Tashiro, “Dielectric properties of fine grained PbTiO3 crystals precipitated in a glass,” J. Non-Cryst. Solids 13(2), 328–340 (1974). [CrossRef]
19. C. R. Gautam, D. Kumar, and O. Parkash, “Crystallization behavior and microstructural analysis of aead-rich (PbxSr1−x)TiO3 glass ceramics containing 1 mole% La2O3,” Adv. Mater. Sci. Eng. 2011, 1–11 (2011). [CrossRef]
20. O. P. Thakur, D. Kumar, O. Parkash, and L. Pandey, “Effect of K2O adiition on crystallization and microstructural behavior of strontium titanate borosilicate glass ceramic system,” J. Mat. Lett. 23(4-6), 253–260 (1995). [CrossRef]
21. S. Golezardi, V. K. Marghussian, A. Beitollahi, and S. M. Mirkazemi, “Crystallization behavior, microstructure and dielectric properties of lead titanate glass ceramics in the presence of Bi2O3 as a nucleating agent,” J. Eur. Ceram. Soc. 30(6), 1453–1460 (2010). [CrossRef]
22. Y. Zhang, T. Ma, X. Wang, Z. Yuan, and Q. Zhang, “Two dielectric relaxation mechanisms observed in lanthanum doped barium strontium titanate glass ceramics,” J. Appl. Phys. 109(8), 084115 (2011). [CrossRef]
23. D. Zhang, T. W. Button, V. O. Sherman, A. K. Agantsev, T. Price, and D. Iddles, “Effects of glass additions on the microstructure and dielectric properties of barium strontium titanate (BST) ceramics,” J. Eur. Ceram. Soc. 30(2), 407–412 (2010). [CrossRef]
24. N. Maury, F. Cambier, and M. Gonon, “Bulk crystallisation of (00l) oriented fresnoite Sr2TiSi2O8 in GC of the Sr–Ti–Si–K–B–O system,” J. Non-Cryst. Solids 357(3), 1079–1084 (2011). [CrossRef]
25. B. Gu, Y. H. Wang, X. C. Peng, J. P. Ding, J. L. He, and H. T. Wang, “Giant optical nonlinearity of a Bi2Nd2Ti3O12 ferroelectric thin film,” Appl. Phys. Lett. 85(17), 3687–3689 (2004). [CrossRef]
26. K. Ruan, X. Chen, T. Liang, G. Wu, and D. Bao, “Photoluminescence and electrical properties of highly transparent (Bi,Nd)4Ti3O12 ferroelectric thin films on indium-tin-oxide-coated glass substrates,” J. Appl. Phys. 103(7), 074101 (2008). [CrossRef]
27. M. V. Shankar and K. B. R. Varma, “Crystallization, dielectric and optical studies on strontium tetraborate glasses containing bismuth titanate,” Mater. Res. Bull. 33(12), 1769–1782 (1998). [CrossRef]
28. K. Gerth and C. Rüssel, “Crystallization of Bi4Ti3O12 from glasses in the system Bi2O3/TiO2/B2O3,” J. Non-Cryst. Solids 221(1), 10–17 (1997). [CrossRef]
29. S. Kojima, A. Hushur, F. Jiang, S. Hamazaki, M. Takashige, M. S. Jang, and S. Shimada, “Crystallization of amorphous bismuth titanate,” J. Non-Cryst. Solids 293–295, 250–254 (2001). [CrossRef]
30. M. Villegas, T. Jardiel, A. C. Caballero, and J. F. Fernández, “Electrical Properties of Bismuth Titanate Based Ceramics with Secondary Phases,” J. Electroceram. 13, 543–548 (2004). [CrossRef]
31. M. Villegas, A. C. Caballero, C. Moure, P. Dura’n, and J. F. Ferna’ndez, “Factors affecting the electrical conductivity of donor-doped Bi4Ti3O12 piezoelectric ceramics,” J. Am. Ceram. Soc. 82(9), 2411–2416 (1999). [CrossRef]
32. R. E. Melgarejo, M. S. Tomar, S. Bhaskar, P. S. Dobal, and R. S. Katiyar, “Large ferroelectric response in Bi4-xNdxTi3O12 films prepared by sol–gel process,” Appl. Phys. Lett. 81(14), 2611–2613 (2002). [CrossRef]
33. C. Hirayama and E. C. Subbarao, “Dielectric properties of bismuth borate glasses,” Phys. Chem. Glasses 3, 111–115 (1962).
34. G. D. Hu, S. H. Fan, and X. Cheng, “Anisotropy of ferroelectric and piezoelectric properties of Bi3.15Pr0.85Ti3O12 thin films on Pt (100)/Ti/SiO2/Si substrates,” J. Appl. Phys. 101(5), 054111 (2007). [CrossRef]
35. X. J. Zheng, L. He, Y. C. Zhou, and M. H. Tang, “Effects of europium content on the microstructural and ferroelectric properties of Bi4−xNdxTi3O12 thin films,” Appl. Phys. Lett. 89(25), 252908 (2006). [CrossRef]
36. M. E. Villafuerte-Castrejón, F. Camacho-Alanís, F. González, A. Ibarra-Palos, G. González, L. Fuentes, and L. Bucio, “Luminescence and structural study of Bi4−xNdxTi3O12 solid solution,” J. Eur. Ceram. Soc. 27(2-3), 545–549 (2007). [CrossRef]
37. K. Aizawa and Y. Ohtani, “Ferroelectric and luminescent properties of Nd-doped SrBi{2}Ta{2}O{9} films,” Jpn. J. Appl. Phys. 46(10B), 6944–6947 (2007). [CrossRef]
38. G. R. Antis, P. Chantikul, B. R. Lawn, and D. B. Marshall, “A critical evaluation of indentation technique for measuring fracture toughness: I, direct crack measurements,” J. Am. Ceram. Soc. 64(9), 533–538 (1981). [CrossRef]
39. H. E. Kissinger, “Variation of peak temperature with heating rate in differential thermal analysis,” Res. Nat. Bur. Stand. 57(4), 217–221 (1956). [CrossRef]
40. J. A. Augis and J. E. Bennett, “Calculation of the avrami parameters for heterogeneous solid state reactions using a modification of the Kissinger method,” J. Therm. Anal. 13(2), 283–292 (1978). [CrossRef]
41. T. Ozawa, “A new method of analysing thermogravimetric data,” Bull. Chem. Soc. Jpn. 38(11), 1881–1886 (1965). [CrossRef]
42. A. K. Varshneya, Fundamentals of Inorganic Glasses (The Society of Glass Technology, 2006).
43. A. A. Al-Ghamdi, M. A. Alvi, and S. A. Khan, “Non-isothermal crystallization kinetic study on Ga15Se85−xAgx chalcogenide glasses by using differential scanning calorimetry,” J. Alloy. Comp. 509(5), 2087–2093 (2011). [CrossRef]
44. I. Ardelean, S. Cora, and D. Rusu, “EPR and FT-IR spectroscopic studies of Bi2O3–B2O3– CuO glasses,” Physica B 403(19-20), 3682–3685 (2008). [CrossRef]
45. K. Gerth and C. Rüssel, “Crystallization of Bi4Ti3O12 from glasses in the system Bi2O3/TiO2/B2O3,” J. Non-Cryst. Solids 221(1), 10–17 (1997). [CrossRef]
46. T. Jardiel, M. A. de la Rubia, and M. Peiteado, “Control of functional microstructure in WO3-doped Bi4Ti3O12 ceramics,” J. Am. Ceram. Soc. 91(4), 1083–1087 (2008). [CrossRef]
47. M. V. Shankar and K. B. R. Varma, “Crystallization of ferroelectric bismuth vanadate in BiO –VO –SrBO glasses,” J. Non-Cryst. Solids 226(1-2), 145–154 (1998). [CrossRef]
48. H. A. Lorentz, “Ueber die beziehung zwischen der fortpflanzungsgeschwindigkeit des lichtes und der koreperdichte,” Ann. Phys. 245(4), 641–665 (1880). [CrossRef]
49. R. Lorenz, “Ueber die refractionsconstante,” Ann. Phys. 247(9), 70–103 (1880). [CrossRef]
50. E. M. Vogel, M. J. Weber, and D. M. Krol, “Nonlinear optical phenomena in glass,” Phys. Chem. Glasses 32, 231–254 (1991).
51. N. L. Boling, A. J. Glass, and A. Owyoung, “Empirical relationships for predicting nonlinear refractive index changes in optical solids,” IEEE J. Quantum Electron. 14(8), 601–608 (1978). [CrossRef]
52. D. W. Hall, M. A. Newhouse, N. F. Borreli, W. H. Dumbaugh, and D. L. Weidman, “Nonlinear optical suceptibilities of high index glasses,” Appl. Phys. Lett. 54(14), 1293–1295 (1989). [CrossRef]
53. I. Thomazeau, J. Etchepare, G. Grillon, and A. Migus, “Electronic nonlinear optical susceptibilities of silicate glasses,” Opt. Lett. 10(5), 223–225 (1985). [CrossRef] [PubMed]
54. G. V. Prakash, D. N. Rao, and A. K. Bhatnagar, “Linear optical properties of niobium-based tellurite glasses,” Solid State Commun. 119(1), 39–44 (2001). [CrossRef]
55. H. Nasu, K. Kubodera, M. Kobayashi, M. Nakamura, and K. Kamiya, “Third harmonic generation from some chalcogcnidc glasses,” J. Am. Ceram. Soc. 73(6), 1794–1796 (1990). [CrossRef]
56. M. J. Weber, D. Milam, and W. L. Smith, “Nonlinear refractive index of glasses and crystals,” Opt. Eng. 17, 463–475 (1987).
57. I. Thomazeau, J. Etchepare, G. Grillon, and A. Migus, “Electronic nonlinear optical susceptibilities of silicate glasses,” Opt. Lett. 10(5), 223–225 (1985). [CrossRef] [PubMed]
58. H. Nasu, Y. Ibara, and K. Kubodera, “Optical third-harmonic generation from some high-index glasses,” J. Non-Cryst. Solids 110(2-3), 229–234 (1989). [CrossRef]
59. R. K. Jain and R. C. Lind, “Degenerate four wave mixing in semiconductor-doped glasses,” J. Opt. Soc. Am. 73(5), 647–653 (1983). [CrossRef]
60. T. Hiramichi, F. Dhigeru, and M. Kanji, “Refractive-index dispersion of tellurite glasses in the region from 0.40 to 1.71 μm,” J. Am. Ceram. Soc. 77(9), 2455–2457 (1994). [CrossRef]
61. K. Pátek, Glass Lasers (Butterworth & Co. 1970).
62. A. J. Moulson and J. M. Herbert, Electroceramics: Materials, Properties, Applications (Chapman & Hall, 1990).
63. A. Tarafder, K. Annapurna, R. S. Chaliha, V. S. Tiwari, P. K. Gupta, and B. Karmakar, “Processing and properties of Nd3+:LiTaO3 transparent glass–ceramic nanocomposites,” J. Am. Ceram. Soc. 92(9), 1934–1939 (2009). [CrossRef]
64. H. Watanabe, T. Kimura, and T. Yamaguch, “Sintering of platelike bismuth titanate powder compacts with preferred orientation,” J. Am. Ceram. Soc. 74(1), 139–147 (1991). [CrossRef]
65. M. B. Volf, Chemical Approach to Glass (Elsevier, 1984).
66. K. K. Bamzai and P. N. Kotru, “Fracture mechanics, crack propagation and microhardness studies on flux grown ErAlO3 single crystals,” J. Mater. Sci. Technol. 16, 405–411 (2000).
