Effects of annealing on density, glass transition temperature and structure of tellurite, silicate and borate glasses

Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Journal of Non-Crystalline Solids journal homepage: www.elsevier.com/locate/jnoncrysol

Effects of annealing on density, glass transition temperature and structure of tellurite, silicate and borate glasses ⁎

Amarjot Kaura, Atul Khannaa, , Amandeep Kaura, Hirdesha, Marina Gonzàlez-Barriusob, Fernando Gonzàlezb a b

Sensors and Glass Physics Laboratory, Department of Physics, Guru Nanak Dev University, Amritsar 143005, Punjab, India Department of Chemistry and Process & Recourse Engineering, University of Cantabria, Santander 39005, Spain

A R T I C LE I N FO

A B S T R A C T

Keywords: Oxide glasses Annealing Structural relaxation Density and glass transition temperature Short-range structure

Barium tellurite, lead tellurite, molybdenum tellurite, lead silicate and bismuth borate glasses of the compositions: xBaO-(100-x)TeO2 (x = 10, 15 and 20 mol%), xPbO-(100-x)TeO2 (x = 15 and 20 mol%), 25MoO3-75TeO2, xPbO-(100-x)SiO2 (x = 50, 60 and 65 mol%) and 40Bi2O3-60B2O3 were prepared by melt-quenching. The effects of long duration thermal annealing (below the respective glass transition temperature (Tg) values) on the density, thermal properties and the short-range structure of glasses were studied. Density increases on average by about 0.5% to 1% and Tg values show drastic increase in the range of 8 °C-64 °C after annealing. Raman and FTIR studies were performed on glasses before and after annealing, FTIR studies found that bismuth borate glasses show a small but significant increase in the fraction of tetrahedral borons in the borate network with annealing. Raman studies on lead metasilicate glass showed an increase in the concentration of SiO4 units that contain three bridging oxygens. Raman and FTIR spectra of all annealed barium and lead tellurite glasses were indistinguishable from those of the initial samples, however the FTIR spectra of annealed 25MoO3-75TeO2 glass show changes in the relative intensities of the peaks due to MoeO vibrations along with the growth and sharpening of low frequency Raman bands, furthermore this glass also showed large differences in its crystallization and melting properties during thermal analysis.

1. Introduction Glass is an amorphous (non-crystalline) solid that has atomic structure indistinguishable from that of a liquid. Glass structure is complex, due to the lack of long-range order of the constituent atoms/ ions, but glasses have definite short-range order. The study of nature of glasses and the phenomenon of the glass transition remains a challenging, unresolved and intriguing problem of condensed matter physics [1,2]. When a glass is fabricated, the material is rapidly cooled from its liquid state till its temperature drops below it's melting or freezing point. At this stage, the material is a supercooled liquid, an intermediary, transient phase between the liquid and glass states. To become an amorphous solid or glass, the material is cooled further, below the glass-transition temperature [3]. Moreover, supercooled liquids and glasses are non-equilibrium states of matter; their mechanical, optical, thermal, electrical and other properties are not stationary in time but slowly and gradually approach toward their equilibrium values when glasses are heated at temperatures below the glass transition temperature, Tg. This phenomenon, which changes properties of glasses, is ⁎

known as the structural relaxation or ageing [4]. The relaxation of thermodynamic properties, e.g., volume or enthalpy, with annealing time and temperature is a known as structural recovery. Studies of relaxation phenomena in glasses are profoundly interesting because it reveals modifications in the atomic structure of disordered solids and their correlations with the thermal, optical and mechanical properties [5–7]. For any glass, the value of Tg provides information on the strength of interatomic bonds and on the network connectivity, similar to the relationship of melting temperature with chemical bonding strength in crystals. Ma et al. [8] reported that the enthalpy will slowly relax to the equilibrium value, when tellurium chalcogenide glasses are isothermally annealed at temperature, T, below their Tg values. Thermal studies on Te33Se5Br2 glass (Tg = 71 °C) were carried out on as-formed glass (without annealing) and after several days of annealing treatment at a temperature of 20 °C and it was found that sub-Tg isothermal annealing produces drastic increase in the glass transition temperature. During annealing, the fraction of non-bridging oxygens in oxide glasses is likely to decrease with the simultaneous formation of bridging

Corresponding author. E-mail address: [email protected] (A. Khanna).

https://doi.org/10.1016/j.jnoncrysol.2018.08.035 Received 27 May 2018; Received in revised form 4 August 2018; Accepted 28 August 2018 0022-3093/ © 2018 Published by Elsevier B.V.

Please cite this article as: Kaur, A., Journal of Non-Crystalline Solids, https://doi.org/10.1016/j.jnoncrysol.2018.08.035

Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx

A. Kaur et al.

oxygens which fills the voids as a result of which a more compact or densely packed network is produced. Long duration heat treatment of glasses may possibly decrease the degree of disorder or some voids may be removed from the network [9–11]. By increasing the ageing time, the glass transition endotherm (change in baseline) is reported to become more pronounced [12]. The influence of isothermal annealing was comphrenhensively studied in polymer glasses by Volynskii et al. [13] and in poly DL-latide by Kairong et al. [14], it was found that as the annealing time increases the glass transition-endothermic peak becomes more sharp or prominent. The effects of annealing time at constant annealing temperature shows remarkable and very intriguing changes in the density, and optical properties of glasses. It is well known that annealing treatment of silica glasses profoundly effects the loss factors for optical fiber applications [15,16]. Tellurite, silicate and borate glasses are of great scientific and technological importance for their applications in fiber optics, laser hosts, infrared to visible upconversion, applications in optical data storage, sensors, laboratory glassware, reflecting and transmitting windows, and for non-linear optical devices [17–19]. B2O3 and SiO2 are excellent glass formers which form glassy phases easily even at very low melt-quenching rates, on the contrary, TeO2, is an conditional glass former which forms glass in the pure form at high melt-quenching rate of ~105 K s−1 [20]. TeO2 however forms variety of binary and ternary glasses at moderate quenching rates of 102 to 103 K s−1 when it is mixed with alkali, alkaline-earth, heavy metal and transition metal oxides [21,22]. Borate and tellurite glasses have many attractive properties over the silicate glasses. While silicate glasses have all silicon ions in tetrahedral coordination [22,23], the boron and tellurium ions in the borate and tellurite network respectively, have a dual co-ordination numbers of 3 and 4 with oxygens [24,25]. The addition of modifier metal oxides in borate and tellurite network produces changes in BeO and TeeO speciation and influences the glass forming ability of materials [26,27]. In case of tellurite glasses TeO4 transform into TeO3 units with increase in metal oxide content [28,29]. There is close connection between physical ageing/relaxation and the glass transition phenomenon. To understand the glass transition problem, the full understanding of the structural relaxation (α-relaxation and β-relaxation) and its dynamics are necessary [30]. It is found that the relaxation behavior in amorphous solids can be well described by the Kohlraush-Williams-Watts (KWW) relaxation function above and below Tg [31,32].

ϕ (t ) = exp.[−(t /τ)β]

however on moving toward lower frequencies and with lowering of temperature, the secondary relaxations dominate the dielectric spectra. The α-relaxation is slowed down upon cooling as the glass transition temperature Tg is approached. Secondary relaxation mainly depends upon the thermodynamic properties of glass [34]. Different types of motions that influence the molecular arrangement of glass like localized instability of the whole molecules, or the rotational instability of the side groups, give rise to the secondary relaxation. According to Johari and Goldstein, the secondary relaxation which is a universal feature of glass forming materials occurs in rigid, molecular glass-formers devoid of intramolecular degrees of freedom [35–37]. According to Vyazovkin [38], by increasing the time of annealing, the endothermic glass transition peak shifts to higher temperatures. Chen [39] found that, when metallic glass, Pd48Ni32P20 is annealed for 210 h below Tg, the whole curve for Cp Vs T shifts toward higher temperature relative to the reference one with the simultaneous large increase in the Tg value. There are several reports on the effects of sub-Tg annealing on the properties of chalcogenide, organic, polymer and metallic glasses [8,13,14,35,36,39], however there are only few studies on the effects of annealing/ageing on the structure and properties of oxide glasses which have wide range of applications [5,6,15,16]. It is the objective of the present study to understand the effects of long duration annealing treatment (below the respective Tg values) of some heavy metal oxide tellurite, silicate and borate glasses and correlate the changes in the density and thermal properties with the changes in short-range structure. Raman and Fourier Transform Infrared (FTIR) spectroscopies are used to determine the changes in short-range structure, while thermal properties are analyzed by Differential Scanning Calorimetry (DSC). 2. Experimental Barium tellurite, lead tellurite, molybdenum tellurite, lead silicate and bismuth borate glasses of the composition: xBaO-(100-x)TeO2 (x = 10, 15 and 20 mol%), xPbO-(100-x)TeO2 (x = 15 and 20 mol%), 25MoO3-75TeO2, xPbO-(100-x)SiO2 (x = 50, 60 and 65 mol%) and 40Bi2O3-60B2O3 were prepared by melting appropriate amounts of analytical reagent grade chemicals (PbO, SiO2, Bi2O3, MoO3, BaCO3, H3BO3 and TeO2) in a platinum crucible. The starting chemicals were mixed and ground in an agate mortar pestle for about 30 min and then transferred to a platinum crucible. Disk shaped samples were prepared by normal quenching; in this method, the melt was poured on a brass block and a disk shaped glass was prepared and immediately transferred to another furnace where it was annealed for about half an hour at temperatures below their respective glass transition temperatures to reduce thermal stresses that are generated by rapid cooling. Normally quenched samples are labeled as “N”. Lead metasilicate and bismuth borate glass samples were also fabricated by splat quenching; in this method small quantity of the melt was pressed between two metal plates. The splat-quenched samples are

(1)

where τ is characteristic relaxation time and β = (1-n) is the fractional exponent. The relaxation process is generally studied by mechanical or dielectric spectroscopy in wide temperature ranges. The α-relaxation prominently appears in the dielectric spectra [33]. The α-relaxation is observed above the glass transition temperature at high frequencies,

Table 1 Composition, density, glass transition temperature (Tg) of barium tellurite, lead tellurite and molybdenum tellurite glasses. Samples with label “H” denote the glasses with long duration annealing treatment. Sample code

Composition (mol %)

Annealing temperature (°C)

Annealing time (h)

Density, d (g cm−3)

(Tg) (°C) ( ± 1°C)

10BaTe 10BaTe-H 15BaTe 15BaTe-H 20BaTe 20BaTe-H 15PbTe 15PbTe-H 20PbTe 20PbTe-H 25MoTe 25MoTe-H

10BaO-90TeO2

250

15BaO-85TeO2

250

20BaO-80TeO2

250

15PbO-85TeO2

250

20PbO-80TeO2

250

25MoO3-75TeO2

250

0.5 1920 0.5 1920 0.5 1920 0.5 1626 0.5 1626 0.5 1082

5.582 5.614 5.561 5.587 5.521 5.564 6.042 6.111 6.199 6.268 5.256 5.302

321 331 326 341 335 343 291 306 284 299 320 343

2

± ± ± ± ± ± ± ± ± ± ± ±

0.002 0.001 0.001 0.001 0.001 0.001 0.002 0.002 0.002 0.002 0.004 0.007

Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx

A. Kaur et al.

Table 2 Composition, density, glass transition temperature (Tg) of lead silicate and bismuth borate glasses. Samples with label “H” denote the glasses with annealing treatment. Sample code

Composition (mol %)

Annealing temperature (°C)

Annealing time (h)

Density, d (g cm−3)

(Tg) (°C) ( ± 1 °C)

50PbSi-N 50PbSi-N-H 50PbSi-S 50PbSi-S-H 60PbSi 60PbSi-H 65PbSi 65PbSi-H 40BiB-N 40BiB-N-H 40BiB-S 40BiB-S-H

50PbO-50SiO2

350

50PbO-50SiO2

350

60PbO-40SiO2

350

65PbO-35SiO2

350

40Bi2O3-60B2O3

350

40Bi2O3-60B2O3

350

0.5 1179 0.5 1179 0.5 2055 0.5 2055 0.5 1179 0.5 1179

5.973 6.019 5.970 6.005 6.699 6.694 7.029 7.019 6.367 6.420 6.319 6.418

388 452 387 443 379 388 350 366 438 455 439 454

labeled as “S”. All glass samples were clear and transparent. The composition and density of samples are given in Table 1 and Table 2. Long duration heat treatment was given at 250 °C for barium tellurite, lead tellurite and molybdenum tellurite glasses and at 350 °C for lead silicate and bismuth borate glasses to find out the changes in glass short-range structure, density and thermal properties (glass-transition, crystallization and melting temperatures). The annealing temperatures were kept 30 °C to 90 °C below the respective Tg values of as-formed glasses. All samples were annealed continuously for several days in a chamber furnace in ambient air to study the structural relaxation effects. The details of annealing times and temperatures of the samples are given in Table 1 and Table 2. All the initial (as-formed) and annealed samples were characterized by X-ray diffraction (XRD), density, DSC, FTIR and Raman spectroscopy. The glassy state of samples was confirmed by XRD studies on Bruker D8 Focus X-ray diffractometer with Cu Kα radiation. Densities of the as prepared and well annealed glasses were measured by Archimedes method using dibutylphatalate as the immersion fluid. The error in density was calculated from the precision of measurement of mass by electronic balance (10−4 g). The maximum uncertainty in the density values is in the range: ± 0.001 to 0.013 g cm−3. Splat quenched glasses had larger uncertainties in the density values due to their smaller size and mass. DSC studies were performed on the SETARAM SETSYS 16 TG-DSC system in temperature range of 200–800 °C at a heating rate of 10 °C min−1. Raman studies were performed on Renishaw In-Via Reflex Micro-Raman spectrometer with 514.5 nm Argon laser. The FTIR absorption spectra of samples were recorded on VARIAN FTIR spectrometer using KBr pellet technique in the wave number range of 400 cm−1 to 2000 cm−1 at room temperature [40].

± ± ± ± ± ± ± ± ± ± ± ±

0.005 0.002 0.009 0.005 0.001 0.001 0.002 0.002 0.003 0.002 0.013 0.007

(Sample Code: 20BaTe), density increases from 5.521 ± 0.001 to 5.564 ± 0.001 g cm−3 [Table 1]. In case of two lead tellurite glasses, that were annealed at 250 °C for 1626 h, density measurements done before and after annealing found that, there is an enhancement of the density from 6.042 ± 0.002 to 6.111 ± 0.002 g cm−3, for 15 mol% lead tellurite glass (Sample Code: 15PbTe), density increases from 6.199 ± 0.002 to 6.268 ± 0.002 g cm−3 for 20 mol% lead tellurite glass (Sample Code: 20PbTe). Variation of density and annealing time of barium and lead tellurite glasses have been shown in Fig. 4(a) along with glass transition temperature, Tg. In case of 25MoO3-75TeO2 glass sample, the density increases from 5.256 ± 0.004 to 5.303 ± 0.007 g cm−3 after annealing the sample at 250 °C for 1082 h. The density of normal quenched 50PbO-50SiO2 glass (lead metasilicate composition) increases from 5.973 ± 0.005 g cm−3 (Sample Code: 50PbSi-N) to 6.019 ± 0.002 g cm−3 (Sample Code: 50PbSi-NH), similarly for splat quenched 50PbO-50SiO2 glass density increases from 5.970 ± 0.009 g cm−3 (Sample Code: 50PbSi-S) to 6.005 ± 0.005 g cm−3 (Sample Code: 50PbSi-S-H). In case of lead silicate glass with 60 mol% PbO, density however shows a small decrease, the density before annealing is 6.699 ± 0.001 g cm−3 while after annealing, the density is found to decrease to 6.694 ± 0.001 g cm−3. Similarly 65 mol% lead silicate glass also shows a small decrease in the density from 7.029 ± 0.002 to 7.019 ± 0.002 g cm−3 after annealing treatment. Density of bismuth borate glass prepared by normal quenching increases significantly from 6.367 ± 0.003 to 6.420 ± 0.002 g cm−3 (Sample Code: 40BiB-N) after its annealing at 350 °C, similarly the density increases from 6.319 ± 0.013 to 6.418 ± 0.007 g cm−3 in case of splat quenched 40Bi2O3-60B2O3 glass (Sample Code: 40BiB-S). The enhancement in density and glass transition temperature, Tg with annealing time of normal and splat quenched 50PbO-50SiO2 glasses and bismuth borate glasses are shown in Fig. 4(b). It may be noted that while the initial density of the normal and splat quenched bismuth borate glasses is different (it is expected that glass prepared by splat quenching will have lower initial density), the final density values after annealing are close to each other for the two samples (40BiB-N and 40BiB-S), which indicates that both the borate glasses relax to the same final structural configuration with annealing [Table 2]. Similar increase in density has been reported earlier in glassy polymers subjected to long term ageing by Plazek et al. [41].

3. Results and discussion 3.1. Structure XRD patterns of all the powdered samples show broad humps which confirmed the amorphous nature of the samples before and after long duration annealing treatment (Fig. 1-3). The absence of any sharp peak confirmed that the annealing treatment did not produce any crystallization in the glasses. 3.2. Density

3.3. Thermal properties The annealing treatment produces significant changes in density of all tellurite glasses, for example in case of 10 mol% barium tellurite glass (Sample Code: 10BaTe), density increases from 5.582 ± 0.001 to 5.614 ± 0.001 g cm−3, for 15 mol% barium tellurite glass (Sample Code: 15BaTe) density increases from 5.561 ± 0.001 to 5.587 ± 0.001 g cm−3 and finally for 20 mol% barium tellurite glass

The thermal properties of as-formed and annealed glasses were studied by DSC. The DSC scan of a glass, gives the glass transition temperature (Tg), the crystallization temperature (Tc) and the melting temperatutre (Tm). Increase in Tg after long duration heat treatment is due to the increase in the concentration of bridging oxygens and/or due 3

Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx

A. Kaur et al.

Fig. 1. (a) XRD plots of barium tellurite glasses (b) XRD plots of lead tellurite glasses. H-denotes the annealed samples.

291 °C to 306 °C for 15PbTe while for 20PbTe glass, Tg increases from 284 °C to 299 °C (Fig. 5). Barium tellurite glasses with 10 and 15 mol% BaO show significant shifting of crystallization and melting peaks to lower temperatures with annealing treatment; however barium tellurite glass with 20 mol% BaO and the two lead tellurite glasses do not show any significant changes in their crystallization and melting properties after annealing. Therefore it is concluded that depending upon the glass composition, sub-Tg annealing can also produce modification of the glass crystallization/devitrification properties. Molybdenum tellurite glass (Sample Code: 25MoTe) again shows a large increase in Tg value from 320 °C to 343 °C (Fig. 6). The tendency of crystallization increases significantly in molybdenum tellurite glass with annealing, as evidenced by is two strong exothermic crystallization peaks and melting peak in the DSC scan, on the contrary the original molybdenum tellurite glass sample showed only one crystallization peak and a very weak melting peak (Fig. 6). Nishida et al. [42], correlated the short-range and medium range order of tellurite glasses with glass transition temperature (Tg) and concluded that the value of Tg in potassium, barium and magnesium tellurite glasses increases with annealing due to an increase in the distortion of network forming polyhedra such as TeO3 and TeO4. Normal quenched 50 mol% lead metasilicate (Sample Code: 50PbSiN and 50PbSi-N-H) glass shows very large enhancement in Tg (midpoint

Fig. 2. XRD plots of 25MoO3-75TeO2 glasses.

to enhanced network connectivity. Drastic enhancement of the glass transition temperature (midpoint value) is found in all the glasses after annealing. Fig. 5 displays the DSC scans of three barium tellurite glasses before and after annealing, the value of Tg after annealing increases from 321 °C to 335 °C for 10BaTe, from 326 °C to 343 °C for 15BaTe and from 335 °C to 345 °C for 20BaTe glass, lead tellurite glasses also follow the same trend in the enhancement of Tg; the latter increases from

Fig. 3. (a) XRD plots of 50PbO-50SiO2 glasses (b) XRD plots of 40Bi2O3-60B2O3 glasses (S = splat and N = normal quenched).

4

Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx

A. Kaur et al.

Fig. 4. (a) Variation of density and Tg with annealing time of barium and lead tellurite glasses (b) Variation of density and Tg with annealing time of lead silicate and bismuth borate glasses.

Recently Alderman [43] showed from in situ XRD studies on borate melts that BeO coordination decreases with increase in temperature. This is because as melt temperature enhances, the borate network depolymerizes and non-bridging oxygen (NBOs) are formed by the isomerization reaction: BO4←→BO2O− . Studies carried out on bismuth borate glasses reveal that heat treatment of borate glasses at temperatures lower than Tg causes an increase in the concentration of BO4 units due to the above reaction occurring in the backward direction. The pronounced increase in the glass transition temperature is due to the decrease of fictive temperature (Tf) [8] and can be attributed to the structural rearrangements in the short-range and medium-range order of glasses. The decrease of Tf is the cause for the enhancement of Tg and is explained by Tool-Narayanaswamy equation [44]. The increase in Tg with annealing is a universal and fundamental property of non-equilibrium glassy phase and it occurs in all type of glasses i.e. oxide, chalcogenide, polymer, organic and metallic glasses, although it is not always possible to attribute it to structural modifications in the glass network. A notable feature in the DSC scans of all tellurite, silicate and borate glasses is the sharpening of the glass transition with an endothermic dip just after the Tg, the glass transition shows almost a endothermic peak because the area above the DSC scan (measure of

value) from 388 °C to 452 °C, while the splat quenched 50 mol% lead silicate (Sample Code: 50PbSi-S and 50PbSi-S-H) also shows the similar large increase in Tg value from 387 °C to 443 °C after annealing at 350 °C for 1179 h (Fig. 7(a)). For lead silicate glass with 60 mol% PbO, Tg shows an increase from 385 °C to 394 °C, while lead silicate glass contatining 65 mol% PbO shows large enhancement in Tg value from 350 °C to 366 °C after annealing treatment (Fig. 7(b)). It may be noted that the glass transition behavior is drastically modified in all lead silicate glasses; the latter becomes very sharp and shifts to higher temperature. However the crystallization and melting peak positions in the DSC scans remain same post annealing (Fig. 7a and b). In case of 40 mol% bismuth borate glass, Tg increases significantly from 438 °C to 455 °C for normally quenched glass (Sample Code: 40BiB-N) while it increases from 439 °C to 454 °C in case of splat quenched glass (Sample Code: 40BiB-S) (Fig. 8). The increase in Tg after long duration heat treatment is due to the formation of more bridging oxygens in borate network. On annealing the concentration of nonbridging oxygens decreases and the average boron‑oxygen coordination increases as confirmed by FTIR analysis (discussed below).

Fig. 5. DSC plots of the as formed and annealed barium and lead tellurite glasses. (H– denotes the annealed glasses and successive curves are shifted by 0.4 units for clarity).

Fig. 6. DSC plots of the initial and annealed 25MoO3-75TeO2 glass. 5

Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx

A. Kaur et al.

Fig. 7. (a) DSC plots of the initial and unannealed 50PbO-50SiO2 glasses (S = splat and N = normal) (b) DSC plots of the initial and unannealed of 60PbO-40SiO2 and 65PbO-35SiO2 glasses. Successive curves are shifted by 0.8 unit for clarity.

Fig. 8. DSC plots of 40Bi2O3-60B2O3 glasses (successive curves are shifted by 1 unit for clarity).

Fig. 10. Raman spectra of 25MoO3-75TeO2 glass sample before and after annealing.

Fig. 9. Raman spectra of barium and lead tellurite glasses before and after annealing. Successive curves are shifted by 0.8 units for clarity.

Fig. 11. Raman spectra of normal and splat quenched 50PbO-50SiO2 glasses before and after annealing. Successive curves are shifted by 0.55 units for clarity. 6

Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx

A. Kaur et al.

Fig. 12. Deconvoluted Raman spectra of normal and splat quenched 50PbO-50SiO2 glasses (H-denotes the annealed samples).

Fig. 14. Raman spectra of annealed and unannealed 40Bi2O3-60B2O3 glasses (successive curves are shifted by 0.5 unit for clarity).

3.4. Short range order by Raman study The Raman spectra of all annealed tellurite, borate and silicate glasses are similar to those of the initial samples. Raman spectra of barium and lead tellurite glasses show five distinct modes: three medium frequency peaks around 254, 326, 464 cm−1 and two strong overlapping broad bands at ~663 and 754 cm−1. Raman bands centered around 464 cm−1 are due to Te-O-Te bending vibrations while the band in the range: 550 to 800 cm−1 are due to stretching vibrations of Te-O-Te linkages in TeO4, TeO3+1 and TeO3 units. The band in the range of 550 to 800 cm−1 is composed of at least four peaks at 620, 665, 720 and 760 cm−1. The peaks at 620 and 640 cm−1 are due to TeeO vibrations in TeO4 units while the higher frequency bands at 720 and 760 cm−1 are due to TeeO vibrations in TeO3+1 and TeO3 units [40,45] (Fig. 9). When the concentration of modifiers such as BaO and PbO is increased in barium tellurite and lead tellurite glass series, the

Fig. 13. Raman spectra of annealed and unannealed lead silicate glasses contatining (a) 60 mol% PbO (b) 65 mol% PbO.

enthalpy) must increase to be equal to the enhanced area (measure of enthalpy) below the DSC scan due to the shifting of Tf to lower temperatures. 7

Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx

A. Kaur et al.

Fig. 15. (a) FTIR spectra of barium tellurite glasses (b) lead tellurite glasses before and after annealing (successive curves are shifted by 0.5 units for clarity).

intensity of the band at 665 cm−1 decreases while that of band at 720 cm−1 increases due to the conversion of TeO4 into TeO3 units, this structural transformation is a well-known effect in tellurite glasses [17,39–41]. It should be noted that Raman spectra of all the barium tellurite and lead tellurite glasses are exactly identical to those of their respective initial glass samples; similarly the FTIR spectra of lead and barium tellurite glasses are also quite similar although not identical to those of the initial glasses. Therefore both FTIR and Raman studies indicate that there are little or no changes in the short-range structure of lead and barium tellurite network with annealing. The Raman spectra of one molybdenum tellurite glass (Fig.10) has the strongest peak at 925 cm−1 due to vibrations of MoeO− and Mo]O bonds in MoO4 or MoO6 units [24,46]. The shoulder at 867 cm−1 is due to vibrations of Mo-O-Mo linkages. The band in the wavenumber region: 700 to 820 cm−1 is due to stretching vibrations of TeO3 units, while the band from 570 and 700 cm−1 is due to stretching vibrations of TeO4 tetrahedra. The broad band from 400 to 500 cm−1 is due to bending vibration of Te-O-Te and shoulder at 378 cm−1 is assigned to vibrations in corner-shared MoO6 octahedra units [24,47]. The Raman patterns of 25MoO3-75TeO2 glass is again invariant post-annealing, however the FTIR spectra of this glass shows significant differences in the relative intensities of some of the peaks (discussed below). Fig. 11 shows the Raman spectra of lead metasilicate glass containing 50 mol-% PbO with Raman bands at 57, 93 and 133 cm−1, shoulder at 516 cm−1 and broad hump from 800 to 1200 cm−1. The broad band from 800 to 1200 cm−1 is due to SieO vibrations in five types of SiO4 units that exist in these glasses [48,49], this band (800 to 1200 cm−1) is deconvoluted with peaks centered at 850, 900, 950, 1000 and 1050 cm−1 (Fig. 12). The peak at 850 cm−1 is due to the orthosilicate groups, the peak at 900 cm−1 is due to pyrosilicate, the peak at 950 cm−1 and 1000 cm−1 are due to metasilicate and the peak at 1050 cm−1 is due to disilicate structural groups [50–53]. Fig. 12 shows the deconvolution of Raman bands in the two 50 mol% lead silicate glasses before and after annealing, the deconvoluted peaks are due to vibrations of SieO bonds in SiO4 tetrahedral units having zero, one, two, three and four bridging oxygens that are denoted as Q0, Q1, Q2, Q3 and Q4 respectively [53,54]. Clearly the intensity of Raman band due to Q3 units increases relative to the intensities of peaks due to Q0, Q1 and Q2 units (Fig. 12), this confirms the decrease in the concentration of non-bridging oxygens in the silicate network with annealing and it produces a drastic enhancement of Tg value. The Raman patterns of two PbO-SiO2 glasses containing 60 and 65 mol% PbO

(Fig. 13 (a) and 13(b)) do not show any significant changes in the bands in the wavenumber range 800–1200 cm−1, however these two glasses show some changes in low frequency peaks due to PbeO bond vibrations [50], and therefore these two glasses show structural rearrangements in PbOx polyhedral units with annealing [50]. Fig. 14 is the Raman spectra of two bismuth borate glasses before and after annealing treatment, the band at 265 cm−1 are due to BieO vibrations, the band around ∼366 cm−1 is assigned to Bi-O-Bi stretching vibrations, whereas the band around ∼860 cm−1 band is due to the vibrations of boroxol ring and stretching vibrations of BeO bond of BO4 units, finally the bands in the range of centered from ∼1173 cm−1 to 1440 cm−1, BeO stretching vibrations in BO3 units bismuth borate glass network [55]. The Raman spectra are not useful to determine changes in BeO co-ordination with annealing treatment, the changes in BeO coordination can be determined by FTIR analysis that are discussed below. 3.5. Short range structure by FTIR The FTIR absorption spectra of barium tellurite (Fig.15(a)) and lead tellurite (Fig.15(b)) before and after long duration annealing are very

Fig. 16. FTIR absorption spectra of 25MoO3-75TeO2 glass before and after annealing. 8

Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx

A. Kaur et al.

feature of all types of oxide (borate, tellurite and silicate), chalcogenide, metallic and polymer glasses. While the enhancement in Tg and density can be correlated with increase in BeO co-ordination number in bismuth borate glasses, No such enhancement in TeeO coordination number is observed from the Raman and FTIR studies of tellurite glasses. Lead meta- silicate glasses show maximum increase in the Tg value with annealing, this glass also shows a small but definite increase in the fraction of Q3 units at the expense of Q0, Q1 and Q2 units. It is concluded that enhancement in the glass transition temperature with annealing or ageing is a universal feature or the fundamental property of glasses, however no single mechanism or structural rearrangement can be attributed to an increase in Tg, the latter is determined by changes in co-ordination number of network forming and modifying cations, distortion of structural units besides changes in the concentration of non-bridging oxygens. Acknowledgements Atul Khanna thanks UGC-DAE-CSR Indore and Mumbai Centres, India and IUAC, New Delhi, India for research grants that supported this work.

Fig. 17. FTIR absorption spectra of 40Bi2O3-60B2O3 glasses before and after annealing treatment (successive curves are shifted by 0.5 units for clarity).

References Table 3 BeO co-ordination of as formed and annealed bismuth borate glasses. Sample code

Composition (mol %)

NB-O ( ± 0.01)

40BiB-N 40BiB-N-H 40BiB-S 40BiB-S-H

40Bi2O3-60B2O3 40Bi2O3-60B2O3 40Bi2O3-60B2O3 40Bi2O3-60B2O3

3.28 3.34 3.26 3.33

[1] C.A. Angell, K.L. Ngai, G.B. McKenna, P.F. McMillan, S.W. Martin, Relaxation in glassforming liquids and amorphous solids, J. Appl. Phys. 88 (2000) 3113–3157. [2] E.D. Zanotto, J.C. Mauro, The glassy state of matter: Its definition and ultimate fate, J. Non-Cryst. Solids 471 (2017) 490–495. [3] G. Johari, Effect of annealing on the secondary relaxations in glasses, J. Chem. Phys. 77 (1982) 4619–4626. [4] M. Utz, P.G. Debenedetti, F.H. Stillinger, Atomistic simulation of aging and rejuvenation in glasses, Phys. Rev. Lett. 84 (2000) 1471–1474. [5] A.Q. Tool, J. Valesak, Concerning the annealing and characteristics of glass, J. Frankl. Inst. 189 (1920) 240–241. [6] A.Q. Tool, Relation between inelastic deformability and thermal expansion of glass in its annealing range, J. Am. Ceram. Soc. 29 (1946) 240–253. [7] R.D. Priestley, Physical aging of confined glasses, Soft Matter 5 (2009) 919–926. [8] H. Ma, X. Zhang, J. Lucas, C. Moynihan, Relaxation near room temperature in tellurium chalcohalide glasses, J. Non-Cryst. Solids 140 (1992) 209–214. [9] S. Chaudhuri, S.K. Biswas, A. Choudhury, K. Goswami, Variation of optical gap of thick amorphous selenium film on heat treatment, J. Non-Cryst. Solids 54 (1983) 179–182. [10] M. Di Giulio, D. Manno, R. Rella, P. Siciliano, A. Tepore, Effects of thermal annealing on optical absorption of amorphous indium selenide thin films, Solar energy materials 15 (1987) 209–218. [11] S.A. Kartha, M. Ittyachen, B. Pradeep, M.A. Khadar, Effect of annealing on the optical properties of B2O3-Li2O-PbO glass thin films, J. Mater. Sci. Lett. 22 (2003) 9–11. [12] M.K. Haque, K. Kawai, T. Suzuki, Glass transition and enthalpy relaxation of amorphous lactose glass, Carbohydr. Res. 341 (2006) 1884–1889. [13] A. Volynskii, A. Efimov, N. Bakeev, Structural aspects of physical aging of polymer glasses, Polymer Science Series C 49 (2007) 301–320. [14] K. Liao, D. Quan, Z. Lu, Effects of physical aging on glass transition behavior of poly (DL-lactide), Eur. Polym. J. 38 (2002) 157–162. [15] A.Q. Tool, Effect of heat-treatment on the density and constitution of high-silica glasses of the borosilicate type, J. Am. Ceram. Soc. 31 (1948) 177–186. [16] J.J. Witcher, W.J. Reichman, L.B. Fletcher, N.W. Troy, D.M. Krol, Thermal annealing of femtosecond laser written structures in silica glass, Optical Materials Express 3 (2013) 502–510. [17] J. Stanworth, Tellurite glasses, Nature 169 (1952) 581. [18] K. Tanaka, K. Kashima, K. Hirao, N. Soga, A. Mito, H. Nasu, Second harmonic generation in poled tellurite glasses, Jpn. J. Appl. Phys. 32 (1993) L843. [19] K. Richardson, D. Krol, K. Hirao, Glasses for photonic applications, Int. J. Appl. Glas. Sci. 1 (2010) 74–86. [20] E.R. Barney, A.C. Hannon, D. Holland, N. Umesaki, M. Tatsumisago, R.G. Orman, S. Feller, Terminal Oxygens in Amorphous TeO2, J. Phys. Chem. Lett. 4 (2013) 2312–2316. [21] R.A. El-Mallawany, Tellurite glasses handbook: physical properties and data, CRC Press, 2011. [22] N. Kaur, A. Khanna, M. Gónzález-Barriuso, F. González, B. Chen, Effects of Al3 +, W6 +, Nb5 + and Pb2 + on the structure and properties of borotellurite glasses, J. Non-Cryst. Solids 429 (2015) 153–163. [23] J. Hiet, M. Deschamps, N. Pellerin, F. Fayon, D. Massiot, Probing chemical disorder in glasses using silicon-29 NMR spectral editing, Phys. Chem. Chem. Phys. 11 (2009) 6935–6940. [24] A. Kaur, A. Khanna, F. González, C. Pesquera, B. Chen, Structural, optical, dielectric and thermal properties of molybdenum tellurite and borotellurite glasses, J. NonCryst. Solids 444 (2016) 1–10.

similar and along with the Raman studies confirm that the short-range structural properties of all lead and barium tellurite glasses do not change with annealing. Although the Raman patterns of 25MoO375TeO2 glass is invariant post-annealing, its FTIR absorption pattern shows significant changes in the relative intensity of the bands in the wavenumber ranges: 400–530 cm−1 and 840–980 cm−1 due to MoeO bond vibrations (Fig. 16). The FTIR spectra of Bi2O3-B2O3 glass (Fig. 17) shows four broad bands from 400 to 640 cm−1, 640 to 720 cm−1,720 to 1120 cm−1 and 1120 to 1520 cm−1. The broad band from 720 to 1120 cm−1 is due to asymmetric stretching vibrations of BeO bonds in tetrahedral BO4 units while the band from 1120 to 1520 cm−1 is due to the asymmetric vibrations of BeO bonds in triangularly co-ordinated BO3 units [56–58]. The integrated areas under these two bands was used to calculate the fraction of tetrahedral borons (N4) in the borate glasses and it is found that the N4 and NB-O (boron-oxygen coordination number) increases significantly, by 6% to 7% in the two 40Bi203-60B2O3 glasses, this confirms the structural transformation: BO3➔BO4 takes place with annealing treatment below the glass transition temperature of the borate glass (Table 3). Alderman [43] discussed in detail the temperature dependence of BeO speciation in borate melts. The present study reveals that short range structure of borate glasses, in particular NB-O is also effected by sub Tg annealing treatment to borate glasses. A well annealed borate glass has significantly higher BeO co-ordination number and Tg value. 4. Conclusions The glass transition temperature shows a drastic enhancement in the range of 8 °C to 64 °C for borate, silicate and tellurite glasses with long duration annealing treatment. In the present study, maximum increase in Tg (64 °C) is found in case of lead metasilicate glasses. On comparing the present findings with the reported data in the literature, it is concluded that the enhancement in Tg by annealing or ageing is a universal 9

Journal of Non-Crystalline Solids xxx (xxxx) xxx–xxx

A. Kaur et al.

1111–1116. [42] T. Nishida, M. Yamada, H. Ide, Y. Takashima, Correlation between the structure and glass transition temperature of potassium, magnesium and barium tellurite glasses, J. Mater. Sci. 25 (1990) 3546–3550. [43] O. Alderman, Borate melt structure: a short review, Physics & Chemistry of Glasses: European Journal of Glass Science & Technology Part B 59 (2018). [44] O. Narayanaswamy, A model of structural relaxation in glass, J. Am. Ceram. Soc. 54 (1971) 491–498. [45] Z. Pan, S.H. Morgan, Raman spectra and thermal analysis of a new lead–tellurium–germanate glass system, J. Non-Cryst. Solids 210 (1997) 130–135. [46] M. Dieterle, G. Weinberg, G. Mestl, Raman spectroscopy of molybdenum oxides Part I. Structural characterization of oxygen defects in MoO3−x by DR UV/VIS, Raman spectroscopy and X-ray diffraction, Phys. Chem. Chem. Phys. 4 (2002) 812–821. [47] Y. Dimitriev, V. Dimitrov, J.C.J. Bart, M. Arnaudov, Structure of Glasses of the TeO2 – MoO3 System, Zeitschriftfür anorganische and allgemeine Chemie, 479 (1981), pp. 229–240. [48] F. Fayon, C. Landron, K. Sakurai, C. Bessada, D. Massiot, Pb2+ environment in lead silicate glasses probed by Pb-LIII edge XAFS and 207Pb NMR, J. Non-Cryst. Solids 243 (1999) 39–44. [49] S. Feller, G. Lodden, A. Riley, T. Edwards, J. Croskrey, A. Schue, D. Liss, D. Stentz, S. Blair, M. Kelley, A multispectroscopic structural study of lead silicate glasses over an extended range of compositions, J. Non-Cryst. Solids 356 (2010) 304–313. [50] T. Furukawa, S.A. Brawer, W.B. White, The structure of lead silicate glasses determined by vibrational spectroscopy, J. Mater. Sci. 13 (1978) 268–282. [51] A.M. Zahra, C.Y. Zahra, B. Piriou, DSC and Raman studies of lead borate and lead silicate glasses, J. Non-Cryst. Solids 155 (1993) 45–55. [52] H. Jia, G. Chen, W. Wang, UV irradiation-induced Raman spectra changes in lead silicate glasses, Opt. Mater. 29 (2006) 445–448. [53] A. Kaur, A. Khanna, S. Singla, A. Dixit, G. Kothiyal, K. Krishnan, S.K. Aggarwal, V. Sathe, F. González, M. González-Barriuso, Structure–property correlations in lead silicate glasses and crystalline phases, Phase Transit. 86 (2013) 759–777. [54] H.W. Nesbitt, G.S. Henderson, G.M. Bancroft, R. Sawyer, R.A. Secco, Bridging oxygen speciation and free oxygen (O2−) in K-silicate glasses: Implications for spectroscopic studies and glass structure, Chem. Geol. 461 (2016) 13–22. [55] B. Meera, J. Ramakrishna, Raman spectral studies of borate glasses, J. Non-Cryst. Solids 159 (1993) 1–21. [56] B. Karthikeyan, S. Mohan, Structural, optical and glass transition studies on Nd3+doped lead bismuth borate glasses, Phys. B Condens. Matter 334 (2003) 298–302. [57] H. Doweidar, Y.B. Saddeek, FTIR and ultrasonic investigations on modified bismuth borate glasses, J. Non-Cryst. Solids 355 (2009) 348–354. [58] A. Bajaj, A. Khanna, Crystallization of bismuth borate glasses, J. Phys. Condens. Matter 21 (2009) 035112–035119.

[25] M.A. Pandarinath, G. Upender, K.N. Rao, D.S. Babu, Thermal, optical and spectroscopic studies of boro-tellurite glass system containing ZnO, J. Non-Cryst. Solids 433 (2016) 60–67. [26] S. Rada, M. Culea, E. Culea, Structure of TeO2-B2O3 glasses inferred from infrared spectroscopy and DFT calculations, J. Non-Cryst. Solids 354 (2008) 5491–5495. [27] A. Kaur, A. Khanna, H. Bhatt, M. Gónzález-Barriuso, F. González, B. Chen, M. Deo, B-O and Te-O speciation in bismuth tellurite and bismuth borotellurite glasses by FTIR, 11B MAS-NMR and Raman spectroscopy, J. Non-Cryst. Solids 470 (2017) 19–26. [28] A.K. Yadav, P. Singh, A review of the structures of oxide glasses by Raman spectroscopy, RSC Advances, (2015), pp. 67583–67609. [29] N. Gupta, A. Kaur, A. Khanna, F. Gonzàlez, C. Pesquera, R. Iordanova, B. Chen, Structure-property correlations in TiO2-Bi2O3-B2O3-TeO2 glasses, J. Non-Cryst. Solids 470 (2017) 168–177. [30] K. Ngai, Correlation between the secondary β-relaxation time at Tg with the Kohlrausch exponent of the primary α relaxation or the fragility of glass-forming materials, Phys. Rev. E 57 (1998) 7346–7349. [31] A. Alegria, E. Guerrica-Echevarria, L. Goitiandia, I. Telleria, J. Colmenero, alpha.Relaxation in the Glass Transition Range of Amorphous Polymers. 1. Temperature Behavior across the Glass transition, Macromolecules 28 (1995) 1516–1527. [32] S. Capaccioli, M. Paluch, D. Prevosto, L.-M. Wang, K. Ngai, Many-body nature of relaxation processes in glass-forming systems, J. Physical Chem. Lett. 3 (2012) 735–743. [33] G. Harrison, The dynamic properties of supercooled liquids, London and New York, Academic Press, 1976, 206 p, 1976. [34] T. Egami, Structural relaxation in metallic glasses, Ann. N. Y. Acad. Sci. 371 (1981) 238–251. [35] I.M. Hodge, A.R. Berens, Effects of annealing and prior history on enthalpy relaxation in glassy polymers. 2. Mathematical modeling, Macromolecules 15 (1982) 762–770. [36] R. Böhmer, K. Ngai, C.A. Angell, D. Plazek, Nonexponential relaxations in strong and fragile glass formers, J. Chem. Phys. 99 (1993) 4201–4209. [37] I.M. Hodge, Enthalpy relaxation and recovery in amorphous materials, J. NonCryst. Solids 169 (1994) 211–266. [38] S. Vyazovkin, I. Dranca, Physical stability and relaxation of amorphous indomethacin, J. Phys. Chem. B 109 (2005) 18637–18644. [39] H. Chen, On mechanisms of structural relaxation in a Pd48Ni32P20 glass, J. NonCryst. Solids 46 (1981) 289–305. [40] A. Kaur, A. Khanna, L.I. Aleksandrov, Structural, thermal, optical and photo-luminescent properties of barium tellurite glasses doped with rare-earth ions, J. NonCryst. Solids 476 (2017) 67–74. [41] D. Plazek, K. Ngai, R. Rendell, An application of a unified relaxation model to the aging of polystyrene below its glass temperature, Polym. Eng. Sci. 24 (1984)

10