- Email: [email protected]
Structure and properties of a non-traditional glass containing TeO2, SeO2 and MoO3
Optical Materials 34 (2012) 1781–1787
Contents lists available at SciVerse ScienceDirect
Optical Materials journal homepage: www.elsevier.com/locate/optmat
Structure and properties of a non-traditional glass containing TeO2, SeO2 and MoO3 A. Bachvarova-Nedelcheva a,⇑, R. Iordanova a, K.L. Kostov a, St. Yordanov b, V. Ganev c a
Institute of General and Inorganic Chemistry, Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria Central Laboratory of Solar Energy and New Energy Source, 1784 Sofia, 72, Tsarigradsko Shose Blvd., Bulgaria c Institute of Mineralogy and Crystallography ‘‘Acad. I. Kostov’’, Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria b
a r t i c l e
i n f o
Article history: Received 12 March 2012 Received in revised form 23 April 2012 Accepted 2 May 2012 Available online 26 May 2012 This paper is dedicated to Professor DSc Yanko Dimitriev from University of Chemical Technology and Metallurgy – Sofia on the occasion of his 75th anniversary. Keywords: Multicomponent glass Network formers Structure Building units
a b s t r a c t A glass containing SeO2, TeO2, MoO3 and La2O3 was obtained at high oxygen pressure (P = 36 MPa) using pure oxides as precursors. The real bulk chemical composition of the glass according to LA-ICP-MS analysis is 17SeO250TeO232MoO31La2O3 (wt.%). The glass was characterized by X-ray diffraction, scanning electron microscopy (SEM), differential thermal analysis (DTA), UV–Vis, XPS, IR and EPR spectroscopy. According to DTA the glass transition temperature (Tg) is below 300 °C. By IR and X-ray photoelectron spectroscopy was determined the main building units (TeO3, TeO4, SeO3, Mo2O8) and the existing of mixed bridging bonds only, which build up the amorphous network. It was established by UV–Vis that the glass is transparent above 490 nm. As a result of a lengthy heat treatment, crystallization took place and crystals rich in SeO2 and TeO2 were found incorporated into the amorphous part containing all components. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Both tellurite and selenite glasses are of scientific and technological interest. The first ones provoke interest due to their high refractive indices, low melting temperatures, high dielectric constants as well as their good IR transmissions. It is known that they are considered as promising materials for non-linear optical devices [1–5]. Some tellurite glasses are also reported to be suitable for setting up optical fiber amplifiers [6]. They also possess electronic behavior-notable semiconductivity and electronic switching effects [7]. These special optical properties encourage identifying them as important materials for potential applications in high performance optics, laser technology and optical communication networks [8–11]. On the other hand selenite glasses as a new and exotic class of non-traditional glasses have not been extensively studied up to now. The investigations of glass formation in different two-, three- and multi-component selenite systems show that it is possible to obtain a selenite glass even at low cooling rate, particularly if the second components are also glass-forming oxides [12,13]. The thermal stability of these glasses increases with the number of the components. In our previous investigations different colored selenite glasses with specific optical properties were obtained [14]. Some complicated compositions should be potential ⇑ Corresponding author. E-mail address: [email protected] (A. Bachvarova-Nedelcheva). 0925-3467/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.optmat.2012.05.002
candidates for technological applications as amorphous and superionic semiconductors, infrared transmission components, in nonlinear optical devices, sensors, reflecting windows, soluble microfertilizers, etc. [15–22]. The main advantage of the introduction of SeO2 is its ability to decrease the melting temperature of glass compositions and to modify their optical properties [23–26]. Our experience in obtaining selenite and tellurite glasses as well as the new papers published recently [9,27,28] concerning nontraditional glasses, motivate us to continue the investigations in this direction. A representative glass with nominal composition 23TeO250SeO222MoO35La2O3 (mol%), i.e. 27TeO243SeO2 24MoO35La2O3 (wt.%) containing only non-traditional glass formers was selected and small amount of La2O3 is used for stabilization of the glass formation. The purpose of this study is to verify the local order in the glass, to determine its thermal stability and optical properties in oxidizing and reduction atmosphere. 2. Experimental 2.1. Glass preparation The composition investigated was selected on the basis of our previous results on the glass formation in various model selenite systems [13,29–32]. The problem with these glasses is their high volatility and the sublimation of SeO2 (at atmospheric pressure) as well as the hygroscopicity of the samples. The glass sample
1782
A. Bachvarova-Nedelcheva et al. / Optical Materials 34 (2012) 1781–1787
has been prepared using reagent grade SeO2 (Reachim, chem. pure), TeO2 (Merck, p.a.), MoO3 (Merck, p.a.) and La2O3 (Alfa-Aesar, p.a.) as starting compounds. The weighing and homogenizing of the batches was carried out in a dry box. In order to avoid SeO2 sublimation, the melting was performed in an autoclave at increased oxygen pressure P = 35–36 MPa and temperature up to 650 °C in silica crucibles. The samples were held for 20 min at this temperature and then cooled slowly at the rate of 2–2.5 °C/min. The obtained glass was sliced and polished to a thickness of 1 mm for optical measurements. The glass was heat-treated at 300 °C in a N2 + H2 (80%) gas flow for 4 and 12 h exposure times. 2.2. Glass characterization The thermal stability was determined by differential thermal analysis (LABSYSTM EVO apparatus) with Pt–Pt/Rh thermocouple at a heating rate of 10 K/min in air flow, using Al2O3 as a reference material. The accuracy of the temperature maintenance was determined ±5 °C. The heating of the glasses were limited up to 600 °C. The microstructure of the samples obtained were characterized by scanning electron microscopy – SEM (SEM 525 M). Microprobe analysis (analyze EDAX 9900) was performed on polish samples in order to localize the element distribution. The amorphous state of the samples was established by X-ray phase analysis (Bruker D8 Advance, Cu Ka radiation). The optical transmission and absorbance spectra in the visible region were investigated by an UV– Vis Spectrophotometer (Cary 100 Scan, Varian and Evolution 300) in the wavelength range 400–1000 nm. The transmission spectrum was performed on a polished glass sample with a thickness of about 3 mm. The absorbance test was performed on a powdered sample. Infrared spectroscopy was used for studying the glass structures. The IR spectra of the glasses were recorded in the 1400–400 cm 1 region, using the KBr pellet technique (Nicolet320 FTIR spectrometer). X-ray photoelectron spectroscopy (XPS) measurements were carried out on an ESCALAB Mk II (VG Scientific Ltd) electron spectrometer with base vacuum in the analysis chamber of 10 8 Pa. The XPS spectra were recorded using Mg Ka excitation source with photon energy of 1253.6 eV. The photoemitted electrons were separated, according to their kinetic energy, by a 150° spherical analyzer with a pass energy of 20 eV and 6 mm entrance- and exit-slit widths, which gives a total instrumental resolution of 1.06 eV (as measured with the FWHM of Ag3d5/2 photoelectron line). Energy calibration was performed by normalizing the C1s line of adventitious adsorbed hydrocarbons to 285 eV. The relative concentrations of the different chemical species were determined by normalization of the areas of the corresponding photoelectron peaks by their photoionization cross-sections calculated by Scofield [33]. The thermal conductivity (C – THERM/Thermal Conductivity Analyzer (TCITM) and the density (Micromeritics Accu Pyc 1330 (Gas pycnometer) of the glass sample were measured as well. The real composition of the bulk glass was determined by laser ablation inductively coupled plasma (LA-ICP-MS) measurements. They are performed on PerkinElmer ELAN DRC-e ICP-MS instrument in standard mode integrated with New Wave Research (ESI) UP-193FX excimer laser ablation system in single ablation spot setup (laser energy 7.5 mJ; repetition rate 10 Hz; spot size 50 lm). Optimized dry plasma conditions are obtained by precisely controlled carrier (He) and make-up (Ar) gas flows. The machine drift effect correction is done by NIST 610 double standard measurements bracketing of each five sample unknowns. The second standard NIST612 is used for data verification purposes. A preliminary, determined by EPMA content of molybdenum in the sample, is used as an internal standard. The final glass composition is calculated by offline laser ablation data reduction software SILLS.
The EPR spectra were recorded as the first derivative of the absorption signal of an ERS-220/Q spectrometer and the recording temperatures were 123 K. The g factors were determined with respect to a Mn2+/ZnS standard. The signal intensity was established by double integration of the experimental EPR spectrum.
3. Results and discussion The actual chemical composition of the investigated glass determined by different methods is given in Table 1. There is some difference in the obtained data due to the accuracy and peculiarities of the used analytical methods and devices for analysis. We accept the results obtained by LA-ICP-MS as more realistic for the bulk composition because this method gives the distribution of elements in the sample volume. On the other hand, as is seen from Table 1 there is a good coincidence in the results obtained by XPS and Microprobe analysis methods which are surface sensitive techniques. Despite the fact that the melting was performed at high oxygen pressure (35–36 MPa), the measured SeO2 content by all methods is significantly lower, comparing to that in the initial glass sample. It is in the range 17–28 wt.% depending on the applied analytical method, which means that the SeO2 volatility is difficult to be overcome. Fig. 1 shows the DTA curves of the glass before and after heat treatment in reduction atmosphere. The temperature range 220–300 °C could be related to the relaxation processes in the glass transition region. A strong endothermic effect was observed above 400 °C, which is connected with sharp loss of weight. TG curves indicate that the mass loss (25–30 wt.%) increases drastically when the temperature exceeds 350 °C. Our results are compatible with the data for other tellurite glasses [34–36]. The XRD results (Fig. 3) for the amorphous nature of sample (after 4 h reduction) and the lack of well defined exothermic peak of crystallization indicate the kinetic stability of the glass samples before and after heat treatment. This feature is important for their further applications in optical technology. Formation of well shaped nanorods, rich in tellurium and selenium was observed only after 12 h reduction by SEM (Fig. 2b and c). According to XRD pattern peaks of TeO2 (JCPDS 74-1131), SeO2 (JCPDS 04-0429) and TeSeO4 (JCPDS 01-0704906) were identified (Fig. 3). The measured glass density value 4.49 g/cm3 is lower than that of pure TeO2 glass (5.11 g/cm3) and other multicomponent tellurite glasses (5.18–5.99 g/cm3) [3,36–38]. The thermal conductivity at room temperature is 0.51 W/m K. According to the literature data, this value is lower than that of typical silicate and boro-silicate glasses (1.1–1.46 W/m K) and it is closer to the values of flint glasses (0.53 W/m K) [37] and other tellurite glasses [4].The IR spectral data of the glass melted in oxidizing atmosphere is shown in Fig. 4a. IR data for pure TeO2 [38], SeO2 [39], MoO3 [40] and selenite compounds (Ag2SeO3) [32] gives us additional information for the interpretation of the spectral data (Fig. 4b and c). The assignment of the bands is made in the framework of the local point symmetry approach, which has been used in our previous studies for molybdate [41], tellurite [42] and selenite glasses [43]. The spectrum of the glass is characterized by an intensive absorption region centered at 670 cm 1 and weak bands in the range 960–850 cm 1. The band at 670 cm 1 may be assigned to the asymmetric vibrations (mas ax ) of TeO4 units [44]. However, a vibration of symmetrical TeO3 groups was also observed in the same absorption region (700–670 cm 1) [44]. That is why we assumed the simultaneous existence of both TeO4 and TeO3 groups in the amorphous network. This is in agreement with the photoelectron spectroscopy data reported below (Fig. 8). The weak band at 840 cm 1 could be related to the vibrations of isolated SeO3 groups [32,43]. The bands at 960 and 890 cm 1 are due to the
1783
A. Bachvarova-Nedelcheva et al. / Optical Materials 34 (2012) 1781–1787 Table 1 Oxides quantity (wt.%) according to LA-ICP-MS, Microprobe and XPS analysis. Oxides (wt.%)
Initial composition
LA-ICP-MS
SEM-microprobe analysis
XPS
SeO2 TeO2 MoO3 La2O3
50 23 22 5
17 50 32 1
28 42 29 1
27 41 31 <1
2.0
105
20
initial glass
after 4h reduction
100
TG
1.5
100 10
85 0.0
80
215
90
DTA
-10 85
-0.5
80
455
420
-20
75
75
-30
TG
70
-1.0
100
200
300
400
Weight loss, wt %
0.5
95
0 355
270
90
Heat flow, μV
Heat flow, μV
1.0
Weight loss, wt %
95
DTA
500
100
Temperature, ºC
200
300
400
500
600
Temperature, ºC
Fig. 1. DTA curves of the glass before and after heat treatment.
(a)
(b)
(c)
amorphous part
Crystals TeO2, SeO2 Fig. 2. SEM images of the initial glass (a) and after 12 h heat treatment in a reduction atmosphere (b and c).
A. Bachvarova-Nedelcheva et al. / Optical Materials 34 (2012) 1781–1787
transition, metal particles, etc.) [49]. Further investigations will be performed in order to elucidate this problem. After the reduction (4 and 12 h), samples became darker visually and as a result their transmission decreased (Fig. 5b). The optical absorption spectra (Fig. 5b) were used to determine the optical band gap value (Eopt) of 2.48 eV applying the Tauc equation, [50] and which is similar to the value reported for a 60TeO240ZnO glass composition [51]. The Eopt value for different multicomponent tellurite glasses with higher TeO2 content (70–90 mol %) varies from 2.54 to 2.97 eV depending on composition [46,47,51–53]. Our band gap value is lower than those values, which is in accordance with the observed red shifting of kc. Useful information concerning the connectivity between the building units (bridging and non-bridging bonds) was obtained also from XPS analysis [54,55]. In Fig. 6 the Se3d spectral region of the initial glass sample (a) and after its two-step reduction (b, c) is shown. The most intense features of the spectra can be well fitted with two peaks at binding energies (BE) of 59.4 and 60.3 eV corresponding to 3d5/2- and 3d3/2-electron energy levels to Se4+ ions [56]. These results are in accordance with the previous investigations on multicomponent oxide glasses containing SeO2 where it was established presence only of Se4+ (BE = 59.4 eV) [56,57]. A shoulder at about 60 eV is observed and it cannot be attributed to the other lower binding-energy peaks of Mo4p, Te3d and Mo4s peaks found at about 40.8, 44.8 and 68 eV. For the initial composition and after 4 h reduction this feature can be well fitted with two peaks at 56.8 and 57.7 eV. They are attributed to the valence state of Se lower than Se4+. However the 3d5/2 binding energy of this newly appeared state is higher than that of elementary Se0 found at about 55.5 eV [56]. Indeed, after longer reduction time (12 h) the shoulder becomes much broader and it can be fitted with two pairs of to 3d5/2- and 3d3/2-peaks keeping the same spin–orbital splitting (0.86 eV) between them. The fitting curves of shoulder are shown more clearly in the top spectrum in Fig. 6, magnified six times. The new 3d5/2-peak is located at 55.1 eV in good agreement with the value for the elementary Se0 [33,56]. Therefore, as a conclusion we can state that the reduction of the sample includes the appearance of a lower than Se4+ valence state, most probably Se0.
glass
Intensity, a.u.
∗
ο - TeO2 (JCPDS 74 - 1131)
ο∗
∗
∗ - SeO (JCPDS 04 - 0429) 2
X - TeSeO4 (JCPDS 01-070-4906)
X XX X X
ο
∗ ∗ο
X
(b)
(a) 10
20
30
40
50
60
70
80
2Θ, deg Fig. 3. X-ray diffraction of the sample after 4 and 12 h reduction in N2 + H2 (80%).
720
920
vibrations of Mo2O8 (MoO5) units following the IR data obtained for selenite-molybdate and tellurite-molybdate glasses [26,42, 45]. Moreover, in the IR spectrum of composition 60TeO240SeO2, where MoO3 is absent, there are no bands at 960 and 900 cm 1 (Fig. 4b). These results are in accordance with our previous structural studies, where was shown that V2O5, SeO2, TeO2, preserved the high coordination number of molybdenum (5 and 6) in a wider concentration range [26,45]. As can be seen in the IR spectrum, there is a plateau below 600 cm 1 which could be related to the vibrations of different mixed bonds (containing Te, Se and Mo – Fig. 4b). A transmission spectrum of the orange colored initial glass is presented in Fig. 5a. The glass is transmittant up to 80% and a sharp cut-off near 490 nm (kc) was observed. This value is red shifted compared to the values of other multicomponent tellurite glasses [46–48]. It is difficult to explain the origin of coloration of the initial glass due to the simultaneous presence of several ions, the coloring mechanism of which is different (charge transfer, d–d
880
initial glass
(a)
(b)
820
990
590
870 980
830
Ag2SeO3
760
620
pure TeO2
MoO3
490
Transmittance (a. u.)
520 470 430 700 660 630
770
650
720
Transmittance (a. u.) 670
Transmittance (a. u.)
900 840
830
960
60TeO2.40SeO2
SeO2
430
17SeO2.50TeO2.32MoO3.1La2O3
ο
600 570 530
1784
(c)
1200 1100 1000 900 800 700 600 500 400
1200 1100 1000 900 800 700 600 500 400
1200 1100 1000 900 800 700 600 500 400
wavenumber, cm-1
wavenumber, cm-1
wavenumber, cm-1
Fig. 4. IR spectra of: investigated glass (a); pure TeO2 and sample 60TeO240SeO2 (b); SeO2, MoO3 and Ag2SeO3 (c).
1785
A. Bachvarova-Nedelcheva et al. / Optical Materials 34 (2012) 1781–1787
(a) 80
(b) 1.2
70 1.0
Absorbance, (a. u)
60
T, %
50 40 30 20
0.8 0.6
after reduction at 300º - 12 h
0.4 0.2
10
after reduction at 300º - 4h Initial glass
0.0
0 200
300
400
500
600
700
800
400
wavelength, nm
500
600
700
800
900
1000 1100
wavelength, nm
Fig. 5. UV–Vis spectra of the glass: (a) optical transmittance spectrum of the initial glass; (b) optical absorbance spectra of the glass before and after heat treatment in a reduction atmosphere.
Mo 3d 5/2
Mo 3d3/2
Se 3d x6 c
b
Intensity (cps)
Intensity (arb. units)
c x1
b
a
50
55
60
65
Binding Energy (eV)
a
225
230
235
240
Binding Energy (eV)
Fig. 6. Photoelectron Se3d spectra of the initial glass composition (a) and after reduction heating for 2 h (b) and 12 h (c), respectively. The different peak contributions obtained by the fitting procedure are shown with blue curves and the red ones denote their sum.
Fig. 7. XPS spectra of the Mo3d core-level region of the initial glass composition (a) and after reduction heating in N2 + H2 (80%) gas flow for 2 h (b) and 12 h (c), respectively. The different peak contributions obtained by the fitting procedure are shown with blue curves and the red ones denote their sum.
The Mo3d spectral region (Fig. 7) is fitted with two coupled pairs Mo3d5/2 and Mo3d3/2 peaks. The first set of Mo3d peaks are located at 233.4 and 236.5 eV, respectively and they are related to Mo6+ six coordination (VI) state of the Mo atoms [33,58]. The other two fitted peaks are centerd at lower values 231.8 and 234.9 eV, that could be related to Mo6+O4 or Mo5+. The absence of MoO4 units according to IR analysis and formation of Mo5+ according to EPR analysis allow us to claim that these bands are dominated by Mo5+ [54,59,60]. For binary tellurite-molybdate glasses, Mekki et al. [61] found also presence of Mo5+ ions [61]. We accept the formation of Mo5+ which is confirmed also by the EPR results (see Fig. 10). Hence, the presence of Mo5+ ions along with Se0 and Te0 are the reason for darkening of the sample after reduction.
The Te3d core level spectra show two intense peaks near 577.0 and 586.7 eV, characteristic of the Te3d5/2- and Te3d3/2- peaks, respectively (Fig. 8). The Te3d5/2 peak is fitted with two contributions at 576.8 and 575.0 eV. According to Mekki and Khattak [62] these BE are related to the presence of TeO4 (tbp) and TeO3 (tp) units, respectively. Hence, both TeO4 and TeO3 groups exist in our glass but the TeO4 amount is predominant in the amorphous structure. This result confirmed the IR data which are discussed above. The fitted spectrum of the sample after 12 h in N2 + H2 (80%) gas flow, showed the appearance of a small peak centered at 572.8 eV that could be related to Te0 resulting from the reduction process [33]. Both peaks attributed to TeO3 groups and Te0 are clearly visible by the magnification of spectrum in Fig. 8c (top curve in Fig. 8).
1786
A. Bachvarova-Nedelcheva et al. / Optical Materials 34 (2012) 1781–1787
123 K
c
x1
dP/dB, a.u.
Intensity (arb. units)
x5 4h reduction
initial glass
b
330
340
350
360
370
380
B, mT Fig. 10. EPR spectra of the glass before and after heat treatment.
Te 3d5/2 a 570
575
580
Binding Energy (eV) Fig. 8. Fitted Te3d5/2-peaks characteristic for the initial glass composition (a) and after reduction heating for 2 h (b) and 12 h (c), respectively. The different peak contributions obtained by the fitting procedure are shown with blue curves and the red ones denote their sum.
The O1s photoelectron line of the investigated glass is characterized by a single narrow peak at 531.3 eV (Fig. 9a–c) which may be connected to the existence of bonds with the same character. Moreover, there is no shifting of this peak toward lower energies. It is well known that in the network of oxide glasses the O1s line for BO (bridging oxygen) is about 530–531 eV, while for NBO (nonbridging oxygen) it is near 529 eV [59–62]. The observed
O 1s
Intensity (arb. units)
c
b
single narrow O1s peak is an indication that strong bridging bonds (Te–O–Te, Mo–O–Mo) and mixed bridging oxygen bonds (Te–O–Se, Te–O–Mo, Se–O–Mo) are building up the amorphous network. A weak adsorption of hydroxide groups (532.9 eV) was registered in the XPS spectrum of the sample after 12 h reduction [60,63]. EPR spectroscopy was used in order to determine the local structure of Mo atoms in the investigated glass sample (Fig. 9). The observed signal centered near 1.92 (g\) could be attributed to Mo5+ cations, which is commensurate with the results obtained by other authors [64,65]. As is seen, after reduction the intensity of the EPR signal increased (Fig. 10) which means that Mo5+ content increases also. These obtained results confirmed those obtained by XPS analysis. 4. Conclusions The present study demonstrates that the mixing of several conditional but compatible network formers in an appropriate ratio, favors the formation of a low-melting glass containing SeO2, TeO2, MoO3 and La2O3. For the initial glass the applied analyzing methods show a Se enrichment of the surface region accompanied with a corresponding depletion of Te in comparison to the glass bulk composition. By IR spectroscopy it was proved that the amorphous structure is built up by TeO3, TeO4, SeO3 and Mo2O8 (MoO5) units. The presence of elementary Te0 and Se0 as well as Mo5+ ions, are responsible for the darkening of the glass after heat treatment in a reduction atmosphere. The photoelectron spectroscopy reveals the formation only of bridging oxygen bonds between different structural units. The optical measurements show that the glass possesses a sharp cut-off at about 490 nm. The investigated nontraditional selenite – tellurite – molybdate glass is stable up to 300 °C in air and in reduction atmosphere. A lengthy annealing in reduction conditions leads to the formation of nanocrystals rich in SeO2 and TeO2. Acknowledgements
a 525
530
535
We would like to thank Professor Yanko Dimitriev for his generous support and valuable discussions. The present work is partially supported by RNF 01/0006 and DOO2-104 grant of the Bulgarian National Science Fund.
Binding Energy (eV) References Fig. 9. O1s-photoelectron regions for the initial glass composition (a) and after reduction heating for 2 h (b) and 12 h (c), respectively. The different peak contributions obtained by the fitting procedure are shown with blue curves and the red ones denote their sum.
[1] V. Dimitrov, S. Sakka, J. Appl. Phys. 79 (1996) 1736. [2] T. Yoko, M. Fujita, M. Miyaji, S. Sakka, Chem. Exp. 5 (1990) 549. [3] S. Kim, T. Yoko, J. Am. Ceram. Soc. 49 (1966) 670.
A. Bachvarova-Nedelcheva et al. / Optical Materials 34 (2012) 1781–1787 [4] R. El-Malawany, Tellurite Glasses Physical Properties and Data, CRC Press, 2002. [5] D. Chen, Y. Liu, Q. Zhang, et al., Mater. Chem. Phys. 90 (2005) 78. [6] J. Wang, E. Vogel, E. Snitzer, Opt. Mater. 3 (1994) 187. [7] H. Hirashima, M. Youshida, J. Non-Cryst. Solids 86 (1986) 327. [8] R. El-Mallawany, I. Abbas Ahmed, J. Mater. Sci. 43 (2008) 5131–5138. [9] V. Dorofeev, A. Moiseev, M. Churbanov, et al., Opt. Mater. 33 (2011) 1911– 1915. [10] T. Kosuge, Y. Benino, V. Dimitrov, R. Sato, T. Komatsu, J. Non-Cryst. Solids 242 (1998) 154–164. [11] M. Tatsumisago, S. Lee, T. Minami, Y. Kowada, J. Non-Cryst. Solids 177 (1994) 154–163. [12] L. Lakov, Y. Dimitriev, Phys. Chem. Glasses 22 (1981) 69. [13] Y. Dimitriev, Y. Ivanova, V. Dimitrov, L. Lakov, St. Yordanov, J. Mater. Sci. Lett. 9 (1990) 793. [14] Y. Dimitriev, A. Bachvarova-Nedelcheva, R. Iordanova, St. Yordanov, Phys. Chem. Glasses: Eur. J. Glass Sci. Technol. B 48 (3) (2007) 138–141. [15] A. Bachvarova-Nedelcheva, R. Iordanova, St. Yordanov, Y. Dimitriev, J. NonCryst. Solids 355 (2009) 2027–2030. [16] G. Govindaraj, N. Baskaran, K. Shahi, P. Monoravi, Sol. State Ionics 75 (1995) 1. [17] N. Machida, M. Chusho, T. Minami, J. Non-Cryst. Solids 101 (1988) 70. [18] M. Pal, K. Hirota, Y. Tsujigami, H. Sakata, J. Phys. D: Appl. Phys. 34 (2001) 459. [19] B.V.R. Chowdari, P. Pramoda Kumari, Sol. State Ionics 113–115 (1998) 665. [20] C. Sunandana, A. Bhatnagar, J. Phys. C17 (1984) 467. [21] M. Abe, Y. Benino, T. Fujiwara, T. Komatsu, J. Appl. Phys. 97 (2005) 123516. [22] N. Rangelova, PhD thesis, Sofia 2004. [23] R.J. Short, R.J. Hand, N.C. Hyatt, Symp. II Sci., in: R. Finich, D. Bullen (Eds.), Basis for Nuclear Waste Management XXVI, December 2–5, 2002. [24] L. Lakov, Y. Dimitriev, Phys. Chem. Glasses 23 (1982) 76. [25] Y. Dimitriev, Y. Ivanova, V. Dimitrov, et al., J. Mater. Sci. Lett. 9 (1990) 793. [26] Y. Dimitriev, St. Yordanov, L. Lakov, J. Non-Cryst. Solids 410 (2001) 293–295. [27] B. Deb, S. Bhattacharya, A. Gosh, Lett. J. Exploring Front. Phys. 96 (2011) 37005. [28] A. Moiseev, V. Dorofeev, A. Chilyasov, I. Kraev, et al., Opt. Mater. 33 (2011) 1858–1861. [29] Y. Dimitriev, V. Dimitrov, L. Lakov, Phys. Chem. Glasses 29 (1988) 45. [30] Y. Dimitriev, Y. Ivanova, V. Dimitrov, L. Lakov, J. Mater. Sci. Lett. 21 (1986) 142. [31] A. Bachvarova-Nedelcheva, R. Iordanova, Y. Dimitriev, J. Univ. Chem. Technol. Metall. 40 (4) (2005) 315–318. [32] A. Bachvarova, Y. Dimitriev, R. Iordanova, J. Non-Cryst. Solids 351 (2005) 998– 1002. [33] J.H. Scofield, J. Electr. Spectrosc. Relat. Phenom. 8 (1976) 129. [34] Y. Dimitriev, A. Bankov, Y. Ivanova, et al., Build. Mater. Silic. Ind. 1 (1980) 22– 26 (Bulg). [35] D. Souri, Middle-East J. Sci. Res. 5 (1) (2010) 44–48. [36] E. Lambson, G. Saunders, B. Bridge, et al., J. Non-Cryst. Solids 69 (1984) 117. [37] O. Sergeev, A. Shashkov, Teplofizika opticheskih sred, Nauka i Technika, 1983 (Russ).
1787
[38] M. Dimitrova-Pankova, Y. Dimitriev, M. Arnaudov, V. Dimitrov, Phys. Chem. Glasses 30 (6) (1989) 260–263. [39] G. Gospodinov, L. Sukova, et al., J. Inorg. Chem. 33 (1988) 1970. [40] L. Seguin, M. Figlarz, R. Cavagnat, J. Lassegues, Spectrochim. Acta 51A (1995) 1323. [41] Y. Dimitriev, R. Iordanova, Phys. Chem. Glasses: Eur. J. Glass Sci. Technol. B 50 (2) (2009) 123–132. [42] Y. Dimitriev, J.C.J. Bart, V. Dimitrov, M. Arnaudov, Z. Annorg, Allg. Chem. 479 (1981) 229–240. [43] A. Bachvarova-Nedelcheva, R. Iordanova, Y. Dimitriev, St. Yordanov, J. Opt. Adv. Mater. 9 (10) (2007) 3187. [44] Y. Dimitriev, V. Dimitrov, M. Arnaudov, J. Mater. Sci. 18 (1983) 1353–1358. [45] Y. Dimitriev, St. Yordanov, L. Lakov, J. Non-Cryst. Solids 192&193 (1995) 179– 182. [46] R. El-Mallawany, M. Abdalla, I. Ahmed, Mater. Chem. Phys. 109 (2008) 291– 296. [47] M. Halimah, W. Daud, D. Sidek, et al., Mater. Sci. Poland 28 (1) (2010) 173–180. [48] H. Burger, K. Kneipp, H. Hobert, et al., J. Non-Cryst. Solids 151 (1992) 13. [49] W. Vogel, Chemistry of Glass, American Ceramic Society, Westerville, OH, 1985. [50] J. Tauc, R. GriGorovici, A. Vancu, Phys. Status Solidi 15 (1966) 627. [51] S. Rosmawati, H. Sidek, A. Zainal, et al., J. Appl. Sci. 7 (20) (2007) 3051–3056. [52] Y. Wang, S. Dai, F. Chen, et al., Mater. Chem. Phys. 113 (2009) 407–411. [53] B. Eraiah, Bull. Mater. Sci. 29 (4) (2006) 375–378. [54] L. Holland, Sol–gel Science and Technology, Processing Characterization and Applications, in: S. Sakka (Ed.), Kluwer Acad. Publ., 2005, vol. II, pp. 128. [55] C. Pantano, Experimental Techniques of Glass Science, in: C. Simmons, O. ElBayoumi (Ed.), The Am. Ceram. Soc., 1995, pp. 147. [56] M. Shenasa, S. Sainkar, D. Lichtman, J. Electr. Spectr. Rel. Phenom. 40 (1986) 329–337. [57] St. Yordanov, L. Lakov, Y. Dimitiev, K. Toncheva, in: Proc. 12th Confer. Glass Sci. Ceram., Sofia, Bulgaria, 1997, pp. 331. [58] R. Iordanova, M. Milanova, K.L. Kostov, Phys. Chem. Glasses: Eur. J. Glass Sci. Technol. B 47 (6) (2006) 631–637. [59] R. Iordanova, M. Mancheva, Y. Dimitriev, et al., J. Alloys Compd. 485 (2009) 104–109. [60] M. Milanova, R. Iordanova, K.L. Kostov, J. Non-Cryst. Solids 355 (2009) 379– 385. [61] A. Mekki, G.D. Khattak, L.E. Wenger, J. Non-Cryst. Solids 351 (2005) 2493– 2500. [62] A. Mekki, G. Khattak, Proc. VII Europ. Soc. Glass Sci. Techn. Conf., Athens, Greece, 25–28 April 2004, Phys. Chem. Glasses 46 (2) (2005) 161–164. [63] A. Losev, K. Kostov, G. Tyuliev, Surf. Sci. 213 (1989) 564. [64] M. Rada, S. Rada, P. Pascuta, E. Culea, Spectrochim. Acta Part: A 77 (2010) 832– 837. [65] O. Cozar, D. Magdas, I. Ardelean, J. Non-Cryst. Solids 354 (2008) 1032–1035.
Contents lists available at SciVerse ScienceDirect
Optical Materials journal homepage: www.elsevier.com/locate/optmat
Structure and properties of a non-traditional glass containing TeO2, SeO2 and MoO3 A. Bachvarova-Nedelcheva a,⇑, R. Iordanova a, K.L. Kostov a, St. Yordanov b, V. Ganev c a
Institute of General and Inorganic Chemistry, Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria Central Laboratory of Solar Energy and New Energy Source, 1784 Sofia, 72, Tsarigradsko Shose Blvd., Bulgaria c Institute of Mineralogy and Crystallography ‘‘Acad. I. Kostov’’, Bulgarian Academy of Sciences, 1113 Sofia, Bulgaria b
a r t i c l e
i n f o
Article history: Received 12 March 2012 Received in revised form 23 April 2012 Accepted 2 May 2012 Available online 26 May 2012 This paper is dedicated to Professor DSc Yanko Dimitriev from University of Chemical Technology and Metallurgy – Sofia on the occasion of his 75th anniversary. Keywords: Multicomponent glass Network formers Structure Building units
a b s t r a c t A glass containing SeO2, TeO2, MoO3 and La2O3 was obtained at high oxygen pressure (P = 36 MPa) using pure oxides as precursors. The real bulk chemical composition of the glass according to LA-ICP-MS analysis is 17SeO250TeO232MoO31La2O3 (wt.%). The glass was characterized by X-ray diffraction, scanning electron microscopy (SEM), differential thermal analysis (DTA), UV–Vis, XPS, IR and EPR spectroscopy. According to DTA the glass transition temperature (Tg) is below 300 °C. By IR and X-ray photoelectron spectroscopy was determined the main building units (TeO3, TeO4, SeO3, Mo2O8) and the existing of mixed bridging bonds only, which build up the amorphous network. It was established by UV–Vis that the glass is transparent above 490 nm. As a result of a lengthy heat treatment, crystallization took place and crystals rich in SeO2 and TeO2 were found incorporated into the amorphous part containing all components. Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction Both tellurite and selenite glasses are of scientific and technological interest. The first ones provoke interest due to their high refractive indices, low melting temperatures, high dielectric constants as well as their good IR transmissions. It is known that they are considered as promising materials for non-linear optical devices [1–5]. Some tellurite glasses are also reported to be suitable for setting up optical fiber amplifiers [6]. They also possess electronic behavior-notable semiconductivity and electronic switching effects [7]. These special optical properties encourage identifying them as important materials for potential applications in high performance optics, laser technology and optical communication networks [8–11]. On the other hand selenite glasses as a new and exotic class of non-traditional glasses have not been extensively studied up to now. The investigations of glass formation in different two-, three- and multi-component selenite systems show that it is possible to obtain a selenite glass even at low cooling rate, particularly if the second components are also glass-forming oxides [12,13]. The thermal stability of these glasses increases with the number of the components. In our previous investigations different colored selenite glasses with specific optical properties were obtained [14]. Some complicated compositions should be potential ⇑ Corresponding author. E-mail address: [email protected] (A. Bachvarova-Nedelcheva). 0925-3467/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.optmat.2012.05.002
candidates for technological applications as amorphous and superionic semiconductors, infrared transmission components, in nonlinear optical devices, sensors, reflecting windows, soluble microfertilizers, etc. [15–22]. The main advantage of the introduction of SeO2 is its ability to decrease the melting temperature of glass compositions and to modify their optical properties [23–26]. Our experience in obtaining selenite and tellurite glasses as well as the new papers published recently [9,27,28] concerning nontraditional glasses, motivate us to continue the investigations in this direction. A representative glass with nominal composition 23TeO250SeO222MoO35La2O3 (mol%), i.e. 27TeO243SeO2 24MoO35La2O3 (wt.%) containing only non-traditional glass formers was selected and small amount of La2O3 is used for stabilization of the glass formation. The purpose of this study is to verify the local order in the glass, to determine its thermal stability and optical properties in oxidizing and reduction atmosphere. 2. Experimental 2.1. Glass preparation The composition investigated was selected on the basis of our previous results on the glass formation in various model selenite systems [13,29–32]. The problem with these glasses is their high volatility and the sublimation of SeO2 (at atmospheric pressure) as well as the hygroscopicity of the samples. The glass sample
1782
A. Bachvarova-Nedelcheva et al. / Optical Materials 34 (2012) 1781–1787
has been prepared using reagent grade SeO2 (Reachim, chem. pure), TeO2 (Merck, p.a.), MoO3 (Merck, p.a.) and La2O3 (Alfa-Aesar, p.a.) as starting compounds. The weighing and homogenizing of the batches was carried out in a dry box. In order to avoid SeO2 sublimation, the melting was performed in an autoclave at increased oxygen pressure P = 35–36 MPa and temperature up to 650 °C in silica crucibles. The samples were held for 20 min at this temperature and then cooled slowly at the rate of 2–2.5 °C/min. The obtained glass was sliced and polished to a thickness of 1 mm for optical measurements. The glass was heat-treated at 300 °C in a N2 + H2 (80%) gas flow for 4 and 12 h exposure times. 2.2. Glass characterization The thermal stability was determined by differential thermal analysis (LABSYSTM EVO apparatus) with Pt–Pt/Rh thermocouple at a heating rate of 10 K/min in air flow, using Al2O3 as a reference material. The accuracy of the temperature maintenance was determined ±5 °C. The heating of the glasses were limited up to 600 °C. The microstructure of the samples obtained were characterized by scanning electron microscopy – SEM (SEM 525 M). Microprobe analysis (analyze EDAX 9900) was performed on polish samples in order to localize the element distribution. The amorphous state of the samples was established by X-ray phase analysis (Bruker D8 Advance, Cu Ka radiation). The optical transmission and absorbance spectra in the visible region were investigated by an UV– Vis Spectrophotometer (Cary 100 Scan, Varian and Evolution 300) in the wavelength range 400–1000 nm. The transmission spectrum was performed on a polished glass sample with a thickness of about 3 mm. The absorbance test was performed on a powdered sample. Infrared spectroscopy was used for studying the glass structures. The IR spectra of the glasses were recorded in the 1400–400 cm 1 region, using the KBr pellet technique (Nicolet320 FTIR spectrometer). X-ray photoelectron spectroscopy (XPS) measurements were carried out on an ESCALAB Mk II (VG Scientific Ltd) electron spectrometer with base vacuum in the analysis chamber of 10 8 Pa. The XPS spectra were recorded using Mg Ka excitation source with photon energy of 1253.6 eV. The photoemitted electrons were separated, according to their kinetic energy, by a 150° spherical analyzer with a pass energy of 20 eV and 6 mm entrance- and exit-slit widths, which gives a total instrumental resolution of 1.06 eV (as measured with the FWHM of Ag3d5/2 photoelectron line). Energy calibration was performed by normalizing the C1s line of adventitious adsorbed hydrocarbons to 285 eV. The relative concentrations of the different chemical species were determined by normalization of the areas of the corresponding photoelectron peaks by their photoionization cross-sections calculated by Scofield [33]. The thermal conductivity (C – THERM/Thermal Conductivity Analyzer (TCITM) and the density (Micromeritics Accu Pyc 1330 (Gas pycnometer) of the glass sample were measured as well. The real composition of the bulk glass was determined by laser ablation inductively coupled plasma (LA-ICP-MS) measurements. They are performed on PerkinElmer ELAN DRC-e ICP-MS instrument in standard mode integrated with New Wave Research (ESI) UP-193FX excimer laser ablation system in single ablation spot setup (laser energy 7.5 mJ; repetition rate 10 Hz; spot size 50 lm). Optimized dry plasma conditions are obtained by precisely controlled carrier (He) and make-up (Ar) gas flows. The machine drift effect correction is done by NIST 610 double standard measurements bracketing of each five sample unknowns. The second standard NIST612 is used for data verification purposes. A preliminary, determined by EPMA content of molybdenum in the sample, is used as an internal standard. The final glass composition is calculated by offline laser ablation data reduction software SILLS.
The EPR spectra were recorded as the first derivative of the absorption signal of an ERS-220/Q spectrometer and the recording temperatures were 123 K. The g factors were determined with respect to a Mn2+/ZnS standard. The signal intensity was established by double integration of the experimental EPR spectrum.
3. Results and discussion The actual chemical composition of the investigated glass determined by different methods is given in Table 1. There is some difference in the obtained data due to the accuracy and peculiarities of the used analytical methods and devices for analysis. We accept the results obtained by LA-ICP-MS as more realistic for the bulk composition because this method gives the distribution of elements in the sample volume. On the other hand, as is seen from Table 1 there is a good coincidence in the results obtained by XPS and Microprobe analysis methods which are surface sensitive techniques. Despite the fact that the melting was performed at high oxygen pressure (35–36 MPa), the measured SeO2 content by all methods is significantly lower, comparing to that in the initial glass sample. It is in the range 17–28 wt.% depending on the applied analytical method, which means that the SeO2 volatility is difficult to be overcome. Fig. 1 shows the DTA curves of the glass before and after heat treatment in reduction atmosphere. The temperature range 220–300 °C could be related to the relaxation processes in the glass transition region. A strong endothermic effect was observed above 400 °C, which is connected with sharp loss of weight. TG curves indicate that the mass loss (25–30 wt.%) increases drastically when the temperature exceeds 350 °C. Our results are compatible with the data for other tellurite glasses [34–36]. The XRD results (Fig. 3) for the amorphous nature of sample (after 4 h reduction) and the lack of well defined exothermic peak of crystallization indicate the kinetic stability of the glass samples before and after heat treatment. This feature is important for their further applications in optical technology. Formation of well shaped nanorods, rich in tellurium and selenium was observed only after 12 h reduction by SEM (Fig. 2b and c). According to XRD pattern peaks of TeO2 (JCPDS 74-1131), SeO2 (JCPDS 04-0429) and TeSeO4 (JCPDS 01-0704906) were identified (Fig. 3). The measured glass density value 4.49 g/cm3 is lower than that of pure TeO2 glass (5.11 g/cm3) and other multicomponent tellurite glasses (5.18–5.99 g/cm3) [3,36–38]. The thermal conductivity at room temperature is 0.51 W/m K. According to the literature data, this value is lower than that of typical silicate and boro-silicate glasses (1.1–1.46 W/m K) and it is closer to the values of flint glasses (0.53 W/m K) [37] and other tellurite glasses [4].The IR spectral data of the glass melted in oxidizing atmosphere is shown in Fig. 4a. IR data for pure TeO2 [38], SeO2 [39], MoO3 [40] and selenite compounds (Ag2SeO3) [32] gives us additional information for the interpretation of the spectral data (Fig. 4b and c). The assignment of the bands is made in the framework of the local point symmetry approach, which has been used in our previous studies for molybdate [41], tellurite [42] and selenite glasses [43]. The spectrum of the glass is characterized by an intensive absorption region centered at 670 cm 1 and weak bands in the range 960–850 cm 1. The band at 670 cm 1 may be assigned to the asymmetric vibrations (mas ax ) of TeO4 units [44]. However, a vibration of symmetrical TeO3 groups was also observed in the same absorption region (700–670 cm 1) [44]. That is why we assumed the simultaneous existence of both TeO4 and TeO3 groups in the amorphous network. This is in agreement with the photoelectron spectroscopy data reported below (Fig. 8). The weak band at 840 cm 1 could be related to the vibrations of isolated SeO3 groups [32,43]. The bands at 960 and 890 cm 1 are due to the
1783
A. Bachvarova-Nedelcheva et al. / Optical Materials 34 (2012) 1781–1787 Table 1 Oxides quantity (wt.%) according to LA-ICP-MS, Microprobe and XPS analysis. Oxides (wt.%)
Initial composition
LA-ICP-MS
SEM-microprobe analysis
XPS
SeO2 TeO2 MoO3 La2O3
50 23 22 5
17 50 32 1
28 42 29 1
27 41 31 <1
2.0
105
20
initial glass
after 4h reduction
100
TG
1.5
100 10
85 0.0
80
215
90
DTA
-10 85
-0.5
80
455
420
-20
75
75
-30
TG
70
-1.0
100
200
300
400
Weight loss, wt %
0.5
95
0 355
270
90
Heat flow, μV
Heat flow, μV
1.0
Weight loss, wt %
95
DTA
500
100
Temperature, ºC
200
300
400
500
600
Temperature, ºC
Fig. 1. DTA curves of the glass before and after heat treatment.
(a)
(b)
(c)
amorphous part
Crystals TeO2, SeO2 Fig. 2. SEM images of the initial glass (a) and after 12 h heat treatment in a reduction atmosphere (b and c).
A. Bachvarova-Nedelcheva et al. / Optical Materials 34 (2012) 1781–1787
transition, metal particles, etc.) [49]. Further investigations will be performed in order to elucidate this problem. After the reduction (4 and 12 h), samples became darker visually and as a result their transmission decreased (Fig. 5b). The optical absorption spectra (Fig. 5b) were used to determine the optical band gap value (Eopt) of 2.48 eV applying the Tauc equation, [50] and which is similar to the value reported for a 60TeO240ZnO glass composition [51]. The Eopt value for different multicomponent tellurite glasses with higher TeO2 content (70–90 mol %) varies from 2.54 to 2.97 eV depending on composition [46,47,51–53]. Our band gap value is lower than those values, which is in accordance with the observed red shifting of kc. Useful information concerning the connectivity between the building units (bridging and non-bridging bonds) was obtained also from XPS analysis [54,55]. In Fig. 6 the Se3d spectral region of the initial glass sample (a) and after its two-step reduction (b, c) is shown. The most intense features of the spectra can be well fitted with two peaks at binding energies (BE) of 59.4 and 60.3 eV corresponding to 3d5/2- and 3d3/2-electron energy levels to Se4+ ions [56]. These results are in accordance with the previous investigations on multicomponent oxide glasses containing SeO2 where it was established presence only of Se4+ (BE = 59.4 eV) [56,57]. A shoulder at about 60 eV is observed and it cannot be attributed to the other lower binding-energy peaks of Mo4p, Te3d and Mo4s peaks found at about 40.8, 44.8 and 68 eV. For the initial composition and after 4 h reduction this feature can be well fitted with two peaks at 56.8 and 57.7 eV. They are attributed to the valence state of Se lower than Se4+. However the 3d5/2 binding energy of this newly appeared state is higher than that of elementary Se0 found at about 55.5 eV [56]. Indeed, after longer reduction time (12 h) the shoulder becomes much broader and it can be fitted with two pairs of to 3d5/2- and 3d3/2-peaks keeping the same spin–orbital splitting (0.86 eV) between them. The fitting curves of shoulder are shown more clearly in the top spectrum in Fig. 6, magnified six times. The new 3d5/2-peak is located at 55.1 eV in good agreement with the value for the elementary Se0 [33,56]. Therefore, as a conclusion we can state that the reduction of the sample includes the appearance of a lower than Se4+ valence state, most probably Se0.
glass
Intensity, a.u.
∗
ο - TeO2 (JCPDS 74 - 1131)
ο∗
∗
∗ - SeO (JCPDS 04 - 0429) 2
X - TeSeO4 (JCPDS 01-070-4906)
X XX X X
ο
∗ ∗ο
X
(b)
(a) 10
20
30
40
50
60
70
80
2Θ, deg Fig. 3. X-ray diffraction of the sample after 4 and 12 h reduction in N2 + H2 (80%).
720
920
vibrations of Mo2O8 (MoO5) units following the IR data obtained for selenite-molybdate and tellurite-molybdate glasses [26,42, 45]. Moreover, in the IR spectrum of composition 60TeO240SeO2, where MoO3 is absent, there are no bands at 960 and 900 cm 1 (Fig. 4b). These results are in accordance with our previous structural studies, where was shown that V2O5, SeO2, TeO2, preserved the high coordination number of molybdenum (5 and 6) in a wider concentration range [26,45]. As can be seen in the IR spectrum, there is a plateau below 600 cm 1 which could be related to the vibrations of different mixed bonds (containing Te, Se and Mo – Fig. 4b). A transmission spectrum of the orange colored initial glass is presented in Fig. 5a. The glass is transmittant up to 80% and a sharp cut-off near 490 nm (kc) was observed. This value is red shifted compared to the values of other multicomponent tellurite glasses [46–48]. It is difficult to explain the origin of coloration of the initial glass due to the simultaneous presence of several ions, the coloring mechanism of which is different (charge transfer, d–d
880
initial glass
(a)
(b)
820
990
590
870 980
830
Ag2SeO3
760
620
pure TeO2
MoO3
490
Transmittance (a. u.)
520 470 430 700 660 630
770
650
720
Transmittance (a. u.) 670
Transmittance (a. u.)
900 840
830
960
60TeO2.40SeO2
SeO2
430
17SeO2.50TeO2.32MoO3.1La2O3
ο
600 570 530
1784
(c)
1200 1100 1000 900 800 700 600 500 400
1200 1100 1000 900 800 700 600 500 400
1200 1100 1000 900 800 700 600 500 400
wavenumber, cm-1
wavenumber, cm-1
wavenumber, cm-1
Fig. 4. IR spectra of: investigated glass (a); pure TeO2 and sample 60TeO240SeO2 (b); SeO2, MoO3 and Ag2SeO3 (c).
1785
A. Bachvarova-Nedelcheva et al. / Optical Materials 34 (2012) 1781–1787
(a) 80
(b) 1.2
70 1.0
Absorbance, (a. u)
60
T, %
50 40 30 20
0.8 0.6
after reduction at 300º - 12 h
0.4 0.2
10
after reduction at 300º - 4h Initial glass
0.0
0 200
300
400
500
600
700
800
400
wavelength, nm
500
600
700
800
900
1000 1100
wavelength, nm
Fig. 5. UV–Vis spectra of the glass: (a) optical transmittance spectrum of the initial glass; (b) optical absorbance spectra of the glass before and after heat treatment in a reduction atmosphere.
Mo 3d 5/2
Mo 3d3/2
Se 3d x6 c
b
Intensity (cps)
Intensity (arb. units)
c x1
b
a
50
55
60
65
Binding Energy (eV)
a
225
230
235
240
Binding Energy (eV)
Fig. 6. Photoelectron Se3d spectra of the initial glass composition (a) and after reduction heating for 2 h (b) and 12 h (c), respectively. The different peak contributions obtained by the fitting procedure are shown with blue curves and the red ones denote their sum.
Fig. 7. XPS spectra of the Mo3d core-level region of the initial glass composition (a) and after reduction heating in N2 + H2 (80%) gas flow for 2 h (b) and 12 h (c), respectively. The different peak contributions obtained by the fitting procedure are shown with blue curves and the red ones denote their sum.
The Mo3d spectral region (Fig. 7) is fitted with two coupled pairs Mo3d5/2 and Mo3d3/2 peaks. The first set of Mo3d peaks are located at 233.4 and 236.5 eV, respectively and they are related to Mo6+ six coordination (VI) state of the Mo atoms [33,58]. The other two fitted peaks are centerd at lower values 231.8 and 234.9 eV, that could be related to Mo6+O4 or Mo5+. The absence of MoO4 units according to IR analysis and formation of Mo5+ according to EPR analysis allow us to claim that these bands are dominated by Mo5+ [54,59,60]. For binary tellurite-molybdate glasses, Mekki et al. [61] found also presence of Mo5+ ions [61]. We accept the formation of Mo5+ which is confirmed also by the EPR results (see Fig. 10). Hence, the presence of Mo5+ ions along with Se0 and Te0 are the reason for darkening of the sample after reduction.
The Te3d core level spectra show two intense peaks near 577.0 and 586.7 eV, characteristic of the Te3d5/2- and Te3d3/2- peaks, respectively (Fig. 8). The Te3d5/2 peak is fitted with two contributions at 576.8 and 575.0 eV. According to Mekki and Khattak [62] these BE are related to the presence of TeO4 (tbp) and TeO3 (tp) units, respectively. Hence, both TeO4 and TeO3 groups exist in our glass but the TeO4 amount is predominant in the amorphous structure. This result confirmed the IR data which are discussed above. The fitted spectrum of the sample after 12 h in N2 + H2 (80%) gas flow, showed the appearance of a small peak centered at 572.8 eV that could be related to Te0 resulting from the reduction process [33]. Both peaks attributed to TeO3 groups and Te0 are clearly visible by the magnification of spectrum in Fig. 8c (top curve in Fig. 8).
1786
A. Bachvarova-Nedelcheva et al. / Optical Materials 34 (2012) 1781–1787
123 K
c
x1
dP/dB, a.u.
Intensity (arb. units)
x5 4h reduction
initial glass
b
330
340
350
360
370
380
B, mT Fig. 10. EPR spectra of the glass before and after heat treatment.
Te 3d5/2 a 570
575
580
Binding Energy (eV) Fig. 8. Fitted Te3d5/2-peaks characteristic for the initial glass composition (a) and after reduction heating for 2 h (b) and 12 h (c), respectively. The different peak contributions obtained by the fitting procedure are shown with blue curves and the red ones denote their sum.
The O1s photoelectron line of the investigated glass is characterized by a single narrow peak at 531.3 eV (Fig. 9a–c) which may be connected to the existence of bonds with the same character. Moreover, there is no shifting of this peak toward lower energies. It is well known that in the network of oxide glasses the O1s line for BO (bridging oxygen) is about 530–531 eV, while for NBO (nonbridging oxygen) it is near 529 eV [59–62]. The observed
O 1s
Intensity (arb. units)
c
b
single narrow O1s peak is an indication that strong bridging bonds (Te–O–Te, Mo–O–Mo) and mixed bridging oxygen bonds (Te–O–Se, Te–O–Mo, Se–O–Mo) are building up the amorphous network. A weak adsorption of hydroxide groups (532.9 eV) was registered in the XPS spectrum of the sample after 12 h reduction [60,63]. EPR spectroscopy was used in order to determine the local structure of Mo atoms in the investigated glass sample (Fig. 9). The observed signal centered near 1.92 (g\) could be attributed to Mo5+ cations, which is commensurate with the results obtained by other authors [64,65]. As is seen, after reduction the intensity of the EPR signal increased (Fig. 10) which means that Mo5+ content increases also. These obtained results confirmed those obtained by XPS analysis. 4. Conclusions The present study demonstrates that the mixing of several conditional but compatible network formers in an appropriate ratio, favors the formation of a low-melting glass containing SeO2, TeO2, MoO3 and La2O3. For the initial glass the applied analyzing methods show a Se enrichment of the surface region accompanied with a corresponding depletion of Te in comparison to the glass bulk composition. By IR spectroscopy it was proved that the amorphous structure is built up by TeO3, TeO4, SeO3 and Mo2O8 (MoO5) units. The presence of elementary Te0 and Se0 as well as Mo5+ ions, are responsible for the darkening of the glass after heat treatment in a reduction atmosphere. The photoelectron spectroscopy reveals the formation only of bridging oxygen bonds between different structural units. The optical measurements show that the glass possesses a sharp cut-off at about 490 nm. The investigated nontraditional selenite – tellurite – molybdate glass is stable up to 300 °C in air and in reduction atmosphere. A lengthy annealing in reduction conditions leads to the formation of nanocrystals rich in SeO2 and TeO2. Acknowledgements
a 525
530
535
We would like to thank Professor Yanko Dimitriev for his generous support and valuable discussions. The present work is partially supported by RNF 01/0006 and DOO2-104 grant of the Bulgarian National Science Fund.
Binding Energy (eV) References Fig. 9. O1s-photoelectron regions for the initial glass composition (a) and after reduction heating for 2 h (b) and 12 h (c), respectively. The different peak contributions obtained by the fitting procedure are shown with blue curves and the red ones denote their sum.
[1] V. Dimitrov, S. Sakka, J. Appl. Phys. 79 (1996) 1736. [2] T. Yoko, M. Fujita, M. Miyaji, S. Sakka, Chem. Exp. 5 (1990) 549. [3] S. Kim, T. Yoko, J. Am. Ceram. Soc. 49 (1966) 670.
A. Bachvarova-Nedelcheva et al. / Optical Materials 34 (2012) 1781–1787 [4] R. El-Malawany, Tellurite Glasses Physical Properties and Data, CRC Press, 2002. [5] D. Chen, Y. Liu, Q. Zhang, et al., Mater. Chem. Phys. 90 (2005) 78. [6] J. Wang, E. Vogel, E. Snitzer, Opt. Mater. 3 (1994) 187. [7] H. Hirashima, M. Youshida, J. Non-Cryst. Solids 86 (1986) 327. [8] R. El-Mallawany, I. Abbas Ahmed, J. Mater. Sci. 43 (2008) 5131–5138. [9] V. Dorofeev, A. Moiseev, M. Churbanov, et al., Opt. Mater. 33 (2011) 1911– 1915. [10] T. Kosuge, Y. Benino, V. Dimitrov, R. Sato, T. Komatsu, J. Non-Cryst. Solids 242 (1998) 154–164. [11] M. Tatsumisago, S. Lee, T. Minami, Y. Kowada, J. Non-Cryst. Solids 177 (1994) 154–163. [12] L. Lakov, Y. Dimitriev, Phys. Chem. Glasses 22 (1981) 69. [13] Y. Dimitriev, Y. Ivanova, V. Dimitrov, L. Lakov, St. Yordanov, J. Mater. Sci. Lett. 9 (1990) 793. [14] Y. Dimitriev, A. Bachvarova-Nedelcheva, R. Iordanova, St. Yordanov, Phys. Chem. Glasses: Eur. J. Glass Sci. Technol. B 48 (3) (2007) 138–141. [15] A. Bachvarova-Nedelcheva, R. Iordanova, St. Yordanov, Y. Dimitriev, J. NonCryst. Solids 355 (2009) 2027–2030. [16] G. Govindaraj, N. Baskaran, K. Shahi, P. Monoravi, Sol. State Ionics 75 (1995) 1. [17] N. Machida, M. Chusho, T. Minami, J. Non-Cryst. Solids 101 (1988) 70. [18] M. Pal, K. Hirota, Y. Tsujigami, H. Sakata, J. Phys. D: Appl. Phys. 34 (2001) 459. [19] B.V.R. Chowdari, P. Pramoda Kumari, Sol. State Ionics 113–115 (1998) 665. [20] C. Sunandana, A. Bhatnagar, J. Phys. C17 (1984) 467. [21] M. Abe, Y. Benino, T. Fujiwara, T. Komatsu, J. Appl. Phys. 97 (2005) 123516. [22] N. Rangelova, PhD thesis, Sofia 2004. [23] R.J. Short, R.J. Hand, N.C. Hyatt, Symp. II Sci., in: R. Finich, D. Bullen (Eds.), Basis for Nuclear Waste Management XXVI, December 2–5, 2002. [24] L. Lakov, Y. Dimitriev, Phys. Chem. Glasses 23 (1982) 76. [25] Y. Dimitriev, Y. Ivanova, V. Dimitrov, et al., J. Mater. Sci. Lett. 9 (1990) 793. [26] Y. Dimitriev, St. Yordanov, L. Lakov, J. Non-Cryst. Solids 410 (2001) 293–295. [27] B. Deb, S. Bhattacharya, A. Gosh, Lett. J. Exploring Front. Phys. 96 (2011) 37005. [28] A. Moiseev, V. Dorofeev, A. Chilyasov, I. Kraev, et al., Opt. Mater. 33 (2011) 1858–1861. [29] Y. Dimitriev, V. Dimitrov, L. Lakov, Phys. Chem. Glasses 29 (1988) 45. [30] Y. Dimitriev, Y. Ivanova, V. Dimitrov, L. Lakov, J. Mater. Sci. Lett. 21 (1986) 142. [31] A. Bachvarova-Nedelcheva, R. Iordanova, Y. Dimitriev, J. Univ. Chem. Technol. Metall. 40 (4) (2005) 315–318. [32] A. Bachvarova, Y. Dimitriev, R. Iordanova, J. Non-Cryst. Solids 351 (2005) 998– 1002. [33] J.H. Scofield, J. Electr. Spectrosc. Relat. Phenom. 8 (1976) 129. [34] Y. Dimitriev, A. Bankov, Y. Ivanova, et al., Build. Mater. Silic. Ind. 1 (1980) 22– 26 (Bulg). [35] D. Souri, Middle-East J. Sci. Res. 5 (1) (2010) 44–48. [36] E. Lambson, G. Saunders, B. Bridge, et al., J. Non-Cryst. Solids 69 (1984) 117. [37] O. Sergeev, A. Shashkov, Teplofizika opticheskih sred, Nauka i Technika, 1983 (Russ).
1787
[38] M. Dimitrova-Pankova, Y. Dimitriev, M. Arnaudov, V. Dimitrov, Phys. Chem. Glasses 30 (6) (1989) 260–263. [39] G. Gospodinov, L. Sukova, et al., J. Inorg. Chem. 33 (1988) 1970. [40] L. Seguin, M. Figlarz, R. Cavagnat, J. Lassegues, Spectrochim. Acta 51A (1995) 1323. [41] Y. Dimitriev, R. Iordanova, Phys. Chem. Glasses: Eur. J. Glass Sci. Technol. B 50 (2) (2009) 123–132. [42] Y. Dimitriev, J.C.J. Bart, V. Dimitrov, M. Arnaudov, Z. Annorg, Allg. Chem. 479 (1981) 229–240. [43] A. Bachvarova-Nedelcheva, R. Iordanova, Y. Dimitriev, St. Yordanov, J. Opt. Adv. Mater. 9 (10) (2007) 3187. [44] Y. Dimitriev, V. Dimitrov, M. Arnaudov, J. Mater. Sci. 18 (1983) 1353–1358. [45] Y. Dimitriev, St. Yordanov, L. Lakov, J. Non-Cryst. Solids 192&193 (1995) 179– 182. [46] R. El-Mallawany, M. Abdalla, I. Ahmed, Mater. Chem. Phys. 109 (2008) 291– 296. [47] M. Halimah, W. Daud, D. Sidek, et al., Mater. Sci. Poland 28 (1) (2010) 173–180. [48] H. Burger, K. Kneipp, H. Hobert, et al., J. Non-Cryst. Solids 151 (1992) 13. [49] W. Vogel, Chemistry of Glass, American Ceramic Society, Westerville, OH, 1985. [50] J. Tauc, R. GriGorovici, A. Vancu, Phys. Status Solidi 15 (1966) 627. [51] S. Rosmawati, H. Sidek, A. Zainal, et al., J. Appl. Sci. 7 (20) (2007) 3051–3056. [52] Y. Wang, S. Dai, F. Chen, et al., Mater. Chem. Phys. 113 (2009) 407–411. [53] B. Eraiah, Bull. Mater. Sci. 29 (4) (2006) 375–378. [54] L. Holland, Sol–gel Science and Technology, Processing Characterization and Applications, in: S. Sakka (Ed.), Kluwer Acad. Publ., 2005, vol. II, pp. 128. [55] C. Pantano, Experimental Techniques of Glass Science, in: C. Simmons, O. ElBayoumi (Ed.), The Am. Ceram. Soc., 1995, pp. 147. [56] M. Shenasa, S. Sainkar, D. Lichtman, J. Electr. Spectr. Rel. Phenom. 40 (1986) 329–337. [57] St. Yordanov, L. Lakov, Y. Dimitiev, K. Toncheva, in: Proc. 12th Confer. Glass Sci. Ceram., Sofia, Bulgaria, 1997, pp. 331. [58] R. Iordanova, M. Milanova, K.L. Kostov, Phys. Chem. Glasses: Eur. J. Glass Sci. Technol. B 47 (6) (2006) 631–637. [59] R. Iordanova, M. Mancheva, Y. Dimitriev, et al., J. Alloys Compd. 485 (2009) 104–109. [60] M. Milanova, R. Iordanova, K.L. Kostov, J. Non-Cryst. Solids 355 (2009) 379– 385. [61] A. Mekki, G.D. Khattak, L.E. Wenger, J. Non-Cryst. Solids 351 (2005) 2493– 2500. [62] A. Mekki, G. Khattak, Proc. VII Europ. Soc. Glass Sci. Techn. Conf., Athens, Greece, 25–28 April 2004, Phys. Chem. Glasses 46 (2) (2005) 161–164. [63] A. Losev, K. Kostov, G. Tyuliev, Surf. Sci. 213 (1989) 564. [64] M. Rada, S. Rada, P. Pascuta, E. Culea, Spectrochim. Acta Part: A 77 (2010) 832– 837. [65] O. Cozar, D. Magdas, I. Ardelean, J. Non-Cryst. Solids 354 (2008) 1032–1035.