Er3 +-doped zinc tellurite glasses revisited: Concentration dependent chemical durability, thermal stability and spectroscopic properties

Er3 +-doped zinc tellurite glasses revisited: Concentration dependent chemical durability, thermal stability and spectroscopic properties

Journal of Non-Crystalline Solids 429 (2015) 70–78 Contents lists available at ScienceDirect Journal of Non-Crystalline Solids journal homepage: www...

2MB Sizes 0 Downloads 23 Views

Journal of Non-Crystalline Solids 429 (2015) 70–78

Contents lists available at ScienceDirect

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

Er3 +-doped zinc tellurite glasses revisited: Concentration dependent chemical durability, thermal stability and spectroscopic properties M.R. Dousti a,⁎, R.J. Amjad b, M.R. Sahar c, Z.M. Zabidi d, A.N. Alias d, A.S.S. de Camargo a,⁎ a

Instituto de Física de São Carlos, Universidade de São Paulo, Av. Trabalhador Sãocarlense 400, São Carlos, SP 13566–590, Brazil Department of Physics, COMSATS Institute of Information Technology, Lahore 54000, Pakistan Advanced Optical Materials Research Group, Department of Physics, Faculty of Science, Universiti Teknologi Malaysia, Skudai 81310, Malaysia d Applied Science Department, Universiti Teknologi Mara Cawangan Pulau Pinang, 13500 Permatang Pauh, Pulau Pinang, Malaysia b c

a r t i c l e

i n f o

Article history: Received 21 June 2015 Received in revised form 23 July 2015 Accepted 24 July 2015 Available online xxxx Keywords: Zinc tellurite glass; Erbium; Optical spectroscopy; Chemical durability

a b s t r a c t Tellurite glasses are interesting materials with extensive infrared transmission window, relatively low phonon energy, high refractive indexes and the ability to incorporate reasonably high amount of rare earth ion dopants. These characteristics make them popular candidates for infrared and visible emissions. Particularly, Er3+-doped tellurite glass compositions have been actively studied for broadband near infrared applications where the requirement for low dimension needs to be compensated by higher doping ion concentration. In this work, we revisit Er3+-doped zinc tellurite glasses, which are among the most thermally and chemically stable tellurite compositions. The glasses were prepared by the melt-quenching technique and the favorable effects of increasing dopant concentration on chemical durability, water resistivity and thermal stability (up to 140 °C) are discussed. The photophysical properties of the glasses were studied by absorption and luminescence spectroscopic techniques. The Stokes and anti-Stokes emissions of erbium were analyzed and it was verified that the width of the emission band at 1532 nm strongly depends on Er3+ concentration varying from 60 to 82 nm for 0.5 and 2.5 mol% of Er2O3, respectively. The intensity of green and red upconversion emissions was evaluated and the increased efficiency of red emission with increasing concentration is attributed to energy transfer mechanisms between infrared energy levels. © 2015 Elsevier B.V. All rights reserved.

1. Introduction In the last decades, the impressive technological advances in telecommunications, lighting and lasing devices have pushed a fast development of new luminescent and optical materials, among which glasses occupy a strategic position [1–5]. Contrary to natural and synthetic crystals, glasses can be tailored and obtained faster, at lower costs, and in a much larger variety of sizes and shapes, for the most diverse applications. Besides that, glasses can usually incorporate higher concentration of active dopant ions such as the trivalent rare earth ions (RE). The tremendous commercial advantages that these characteristics represent, associated with the growing demand for more efficient, fastresponsive and multifunctional materials,have motivated the search for new glass compositions. The most commonly explored oxide compositions are based on silicates [6], phosphates [7] and borates [8]. Although these glasses present, in general, good thermal and mechanical

⁎ Corresponding authors. E-mail addresses: [email protected] (M.R. Dousti), [email protected] (A.S.S. de Camargo).

http://dx.doi.org/10.1016/j.jnoncrysol.2015.07.039 0022-3093/© 2015 Elsevier B.V. All rights reserved.

stabilities, their high melting temperatures, hygroscopicity and low RE dispersion hinder desired applications. Furthermore, regular oxide glasses present high phonon energies (~1000 cm−1) which results in larger probability of non-radiative deactivation of RE excited states. As an alternative to overcome that particular disadvantage, fluoride glasses are largely studied, but their mechanical, chemical and thermal stabilities are less to be desired [9]. More recently, special attention has been paid to heavy metal oxide (HMO) glasses such as antimony, bismuth, germanate and tellurite glasses [10]. Particularly, tellurite glasses present promising features such as lower working temperature. One of the earliest reports on the preparation of tellurite glass was by Brady [11]. He noticed that the tellurium dioxide is a poor glass former, requiring the addition of a modifier to trigger the formation of a glass from TeO2 powder. A typical concentration of 10 mol% of any modifier is required to form tellurite glasses. Two crystalline phases exist in pure TeO2; paratellurite (α-form) [12] and tellurite (β-form) [13]. In both forms, four coordinated tellurium ions and four bridging oxygens constitute a completely interconnected network. Incorporation of modifier and intermediate groups breaks the Te–O–Te network and forms the TeO3+ 1 and TeO3 groups. In general, three major different tellurite groups can exist in the glassy form, known as Q44, Q34 and Q23 which

M.R. Dousti et al. / Journal of Non-Crystalline Solids 429 (2015) 70–78

consist of 4, 3 and 2 bridging oxygens, respectively and 0, 1 and 1 nonbridging oxygens. The very good RE solubility, low phonon energies, and high refractive indexes (1.8–2.5) make these glasses promising materials as substitutes for oxide glasses in technological applications, especially in the infrared. The third-order nonlinear susceptibility (χ(3)) of tellurite glass has been measured by Kim et al. as 14.0 × 10−13 esu, which is much higher than those of silicate glasses (0.28 × 10−13 esu) [14], and a non-linear refractive index, n2 = 10.48 for 90TeO2– 10La2O3 decreases to 6.84 for 80TeO2–20La2O3 glass sample [14]. For (100 − x)Te2O–xLi2O glasses, the χ(3) of 24 × 10−13 and 38 × 10−13 esu were measured for x = 20 and 25 mol%, respectively [15]. The structure of zinc tellurite glasses was investigated by Sidek et al. [16], Hoppe et al. [17] and Sahar et al. [18–20]. They showed that the introduction of ZnO to the TeO2 network shifts the Raman band at 626 cm− 1 to 669 cm− 1 for pure TeO2 and 15ZnO–85TeO2 glasses, respectively. The addition of zinc oxide to TeO2 matrix increases the density, refractive index and Urbach energy, while polarizability, molar volume and optical band gap are gradually decreased. Hoppe et al. [17] showed that introduction of ZnO in TeO2 glass system decreases the total Te–O coordination number while TeO4 trigonal bipyramid (tbp) groups transform to TeO3 units. The bond distances for two existing species of Te–O band were measured to be 1.9 Ả and 2.1 Ả. The coordination number changes as follows: addition of second oxide to corner-connected TeO4 in TeO2 network breaks the Te–O–Te bridging and converts two TeO4 units to TeO3 units with one nonbridging oxygen for each. Zinc-tellurite glasses are among those with highest thermal stability (Tx–Tg N 100 °C [20]) but they can be melted below 1000 °C. They show high thermal expansion coefficients (150–200 × 10−6·oC−1) and low deformation temperature, around 250–350 °C [21]. Chemical durability of tellurite glasses were studied by Stanworth [22]. Tellurite glass shows large durability against water, alkaline and acid environments. A weight loss of the order of 5 × 10−7 and 20 × 10−7 g·cm−2·day−1 was measured for 18PbO–82TeO2 and 22PbO–78TeO2 glasses, respectively. The durability of lead-tellurite glass in water is reduced by introduction of modifiers such as lithium oxide and sodium oxide, while this effect is more rigorous by adding the boron oxide in tellurite glass. Tellurite glasses containing BaO, Li2O, Na2O and As2O5 showed heavy attacks and reduction of water durability. Moreover, it has been shown that the chemical durability and water resistance of zinc tellurite glasses increase with increasing rare earth in doped samples [23], and decreases by introduction of silver NPs [24]. Tellurite glasses doped with erbium ions are studied wide and large [25–30]. The yellowish color of the undoped zinc-tellurite glass changes to orange color by addition of erbium ions [31], although green color is also reported for the undoped zinc tellurite glass sample [32]. Er3+ ions show green and red luminescence under various excitation wavelengths (e.g. ~ 376, 480, 800, 980 and 1532 nm). The anti-Stokes upconversion emissions (excitation at 980 and 1532 nm) and the Stokes near-infrared emissions (excitation at 980 nm and emission at around 1532 nm) attracted large attention due to their applications in lasers, sensors, upconverting devices and amplifiers. Moreover, the enhancement of upconversion luminescence of Er3+-doped zinc tellurite glasses embedding metallic nanoparticles has been recently observed [25,31, 33–35]. The NIR broadband emission of the Er3+-doped tellurite glasses is a promising feature of these glass compositions. The FWHM of this band in tellurite glasses is larger (60–80 nm [36–38]) than those reported for soda lime silicate (~19 nm [39]) and aluminum silicate (59.3 nm [40]) glasses. The aim of this work is to revisit the important properties of the erbium doped zinc tellurite glasses prepared by a simple melt-quenching method. The studied composition was (80 − x)TeO2–20ZnO–xEr2O3 (where x = 0, 0.5, 0.7, 1.0, 1.3, 1.5, 1.7, 2.0 and 2.5 mol%). Investigations on chemical durability towards exposure to water and a more aggressive, acidic environment are presented. In addition, thermal stability and optical properties, including Stokes and anti-Stokes emissions and

71

Table 1 Labels and corresponding glass compositions (mol%), theoretical and experimental densities (g.cm-3), ionic densities (ion.cm-3) and characteristic temperatures (°C) [20]. Label

TeO2

ZnO

Er2O3

ρexp

ρth

NEr (×1021)

Tg

Tx

ΔT

ZTEr0.0 ZTEr0.5 ZTEr1.0 ZTEr1.5 ZTEr2.0 ZTEr2.5

80.0 79.5 79.0 78.5 78.0 77.5

20 20 20 20 20 20

0.0 0.5 1.0 1.5 2.0 2.5

5.10[16] 5.12 5.27 5.30 5.35 5.37

5.66 5.67 5.69 5.70 5.71 5.73

0.00 0.59 1.14 1.70 2.25 2.80

325 321 322 – 323 335

439 420 442 – 460 475

114 99 120 – 137 140

excited state lifetimes, are presented as a function of Er3+ ion molar concentration. 2. Experimental The Er3 +-doped glass samples with composition (80 − x)TeO2– 20ZnO–xEr2O3 (where x = 0, 0.5, 0.7, 1.0, 1.3, 1.5, 1.7, 2.0 and 2.5 mol%) were prepared by the conventional melt-quenching technique. The homogeneously mixed powders were melted at 850 °C for 1 h and subsequently poured and annealed on preheated stainless steels molds at 300 °C, for 3 h. The samples were left inside the furnace to cool down to the ambient temperature. Then, the glasses were cut and optically polished for optical measurements. Very fine powders of samples were also prepared for structural and thermal analysis, as well as for photophysical measurements. The glass compositions and corresponding labels are listed in Table 1, in addition to some of the physical properties. The calorimetric and thermal properties of glasses were measured using a differential thermal analyzer (DTA, Model: Pyris Diamond TG-DTA, Japan) by heating with a rate 10 °C/min. The samples were heated from room temperature to 650 °C under N 2 atmosphere with 200 mL/min rate of flow. The fine powders of glasses were mixed with KBr to measure the Fourier Transform Infrared (FTIR) spectra of the samples using a Perkin Elmer spectrometer model Spectrum One. The density values of the glasses were determined by Archimedes' principle taking into account the weight of the samples in air and in water as the immersion liquids. The UV–VIS–NIR absorption spectra were measured in a PerkinElmer spectrophotometer model Lambda 1050. The excitation and emission spectra, as well as the lifetime decay curves were measured in a Horiba Fluorolog fluorimeter, which is equipped with CW and pulsed Xe lamps as excitation sources. The signals were collected by a photodiode detector model PPD-850 in the visible, and by a Hamamatsu photomultiplier in the infrared. The spectra were corrected by the lamp profile and detectors response. The upconversion spectra were measured with excitation by a 976 nm laser with tunable power, DMC model Lasertool. The excited state lifetime values of Er3 + ions were

Fig. 1. Representative differential thermal analysis of the ZTEr1.0 glass sample.

72

M.R. Dousti et al. / Journal of Non-Crystalline Solids 429 (2015) 70–78

Fig. 2. (a) Amount of leached Te4+, Zn2+ and Er3+ ions (mg/L) versus solution pH after 32 days of immersion for sample ZTEr1.5. (b) Plot of pH versus time for various samples exposed to distilled water for 18 days. (The lines are guides for the eye).

evaluated by phosphorimetric technique, using a flashlamp as the excitation source. The chemical durability of the samples was analyzed by immersing each of four pieces of each sample, from the same batch, in four different buffer solutions with pH 4, pH 6, pH 8 and pH 10. The changes in pH of such solutions were recorded for a period of 24 h. Also, the chemical durability was investigated by immersing the glasses in two different solutions; 0.004 M HNO3 (acidic solution) and in 0.004 M NaOH (alkaline solution). Nitric acid (HNO3) was used due to its ability to donate more H+ ions, as well as having relatively higher percentage of oxygen ions as compared to other acidic solutions. Distilled water was

also used as immersion solution to investigate the corroding behavior of the glasses in neutral environment. The quantitative analysis for the cation content that leached out from the glass into the solution was carried out for erbium (Er3+), tellurium (Te4+) and zinc (Zn2+) using Inductively Coupled Plasma Mass Spectroscopy (ICPMS) in a Perkin Elmer equipment model ELAN 6100. Surface morphology studies of the films were investigated by Field Effect Scanning Electron Microscopy (FESEM) using a Zeiss microscope model SUPRA 35VP with magnification of 100 ×. The glass samples were mounted on an aluminum holder before being coated with a thin layer of gold by the sputtering technique.

Fig. 3. Amount of leached ions in HNO3 solution: (a) [Zn2+], (b) [Er3+], and (c) [Te4+]. (d) Leached ion concentration [Zn2+, Er3+ and Te4+] in distilled water for ZTEr1.5. (Lines are guides for the eye).

M.R. Dousti et al. / Journal of Non-Crystalline Solids 429 (2015) 70–78

73

Fig. 4. Cross-sectional FESEM images of samples ZTEr0.0 and ZTEr1.3 before and after immersion in HNO3 (left), and before and after immersion in distilled water (right).

3. Results and discussions 3.1. Structural and thermal analysis The non-crystalline nature of the zinc tellurite glasses were examined by XRD technique as shown in the supplemental material (Fig. S1). The absence of sharp crystallization peaks confirms the amorphous nature of the studied set of glasses. As it can be seen in Table 1, the density of the glasses increases by increasing the concentration of

erbium oxide due to the higher molecular weight of the latter, in comparison to those of zinc and tellurium oxides. However, the increase in the measured density (5.29%) is higher than the theoretical values (~1.23%), for Er2O3 content varying from 0 to 2.5 mol%. The refractive index of the undoped 20ZnO–80TeO2 (mol%) glass sample ZTEr0.0, measured at 632.8 nm is around 1.98 ± 0.01 as reported by Sidek et al. [16]. Fig. 1 shows the DTA profile of the ZTEr1.0 glass, as the representative case. The glass transition temperatures, determined by the intersection method, are around 321–335 °C as also given in Table 1. Two crystallization peaks are observed at around 421 and 452 °C. According to the DTA results, different exothermic peaks indicate the presence of different crystallization phases for all samples. Nukui et al. [41] studied the DTA and high temperature XRD patterns of ZTEr0.0 glass. They observed two peaks around 420 °C and 470 °C and attributed the first to the crystallization of TeO2 and ZnTeO3 phases and the second to the crystallization of Zn2Te3O8 phase. The thermal stability is given as the difference between the first crystallization peak and the glass transition temperature peak (ΔT) and the values are listed in Table 1 for all the samples. By increasing the concentration of Er3+ up to 2.5 mol% Er2O3, the thermal stability increases from 114 to 140 °C (see Table 1). The FTIR spectrum of the ZT glasses was reported by Sahar et al. [20].

Table 2 Experimental and calculated oscillator strengths and J.O. parameters of ZTEr1.0 glass. Transitions

Energy (cm−1)

fexp (×10−6)

fcalc (×10−6)

4

6527 10,204 12,500 15,314 18,382 19,157 20,492

3.06 1.20 0.61 4.59 0.83 12.56 4.45

3.00 1.36 0.71 4.56 1.16 12.55 4.60

I15/2 → 4I13/2 I15/2 → 4I11/2 4 I15/2 → 4I9/2 4 I15/2 → 4F9/2 4 I15/2 → 4S3/2 4 I15/2 → 2H11/2 4 I15/2 → 4F7/2 rmserror = 0.207 × 10−6 4

Fig. 5. Optical absorption spectrum of ZTEr1.0 glass sample. Insets show the magnified spectrum in the UV-edge and the concentration dependent integrated absorption band at 980 nm.

74

M.R. Dousti et al. / Journal of Non-Crystalline Solids 429 (2015) 70–78

It shows the characteristics bands of TeO4 groups (668 cm− 1), TeO3/ TeO3 + 1 groups (747 cm− 1), Te–O–Te linkages (452 cm− 1) and, by increasing Er3 + concentration, the destruction of TeO4 groups and formation of more TeO3/TeO3+ 1 groups was evidenced by the decrease in intensity of the band at 670 cm− 1 and increase of the intensity around 750 cm−1. 3.2. Chemical durability Chemical durability of the glasses was examined in terms of the effect of the solution pH, the glass resistance to the attack and the etching-ability of HNO3 acid. The leached ions were analyzed by inductively coupled plasma mass spectroscopy (ICPMS), the morphology of the glasses after the tests was obtained by FESEM, and the characterization of corroded surface layer was investigated by FTIR and XRD. Fig. 2(a) shows the amount of ions leached from ZTEr1.5 glass pieces, from the same batch, which were exposed during 32 days to various pHs. At pH 4, large amounts of Zn2+ leached out into the solution as compared to Te4+ and Er3+. However at pH 6, 8 and 10, Te4+ leached at a greater rate than Zn2 + and Er3 + ions. Thus, the amount of each leached ionic species strongly depends on the solution pH. The accelerated corrosion process of the glasses in low-pH leaching environment has also been reported by Riley et al. [42]. Comparing Er3+ activity in solutions with pH 6 and pH 4, it can be concluded that its solubility in acidic environment is higher than in the alkali one. Fig. 2(b) presents the variation of pH in time for the distilled water containing the glass series. By increasing the immersing time up to 7 days, the pH descends sharply for glass sample containing smaller Er3 + content. However, for longer time intervals, the variation of pH decreases slowly. This behavior is opposite to that of the corrosion in acidic environment. The occurrence of two decreasing stages indicates that the corrosion in distilled water is controlled by two different mechanisms. First, an ion exchanged process takes place (up to seven days) and later a diffusion process takes over. The final solution is slightly acidic indicating. In Fig. 2(b) it is also seen that rare earth doping increases the resistance of the ZT glasses to water. Same trend was observed for the corrosion in HNO3. Fig. 3(a)–(c) present the leaching amount of different ions versus the immersing period in acidic HNO3 solution. The leaching of Zn2+, Er3+ and Te4+ is proportional to the (time)1/2 which indicates that corrosion occurred by a diffusion mechanism, as discussed by Tammann [43] and Evans [44]. Moreover, it can be inferred that all ions experience the same mechanism, though the amount of leaching is different for each one. The concentration ratio of Zn2+, Te4+ and Er3+ ions that are leached out of glass were obtained by ICPMS measurements. The average proportions Te4+: Zn2+: Er3+ are 1:6:4 and the dependence of Zn2+, Te4+ and Er3+ amounts on (time)1/2 also takes place in two stages, as discussed above. In distilled water, the ratio of leached ions Te4+: Zn2+: Er3+ is 1:5:4. As depicted in Fig. 3(a) and (c), leaching of Zn2+ and Te2+ is much lower for the sample ZTEr0.0 than for the sample ZTEr0.5. However, it is observed that the leaching of these ions begun to decrease again by further addition of Er3+ into the glass network. As shown in Fig. 3(b), leaching of Er3 + initially decreases with increasing erbium content, but for Er2O3 content larger than 1.7 mol%, the leached amount of Er3+ions is increased. It is assumed that the addition of Er3+ changes the glass network through two approaches: (1) Up to 1.7 mol% the rare earth enhances the compactness of the glass structure and improves the resistance of the glasses to the acidic environment; (2) For concentrations of Er2O3 higher than 1.7 mol% the resistance decreases, because Er3+ ions do not participate in the glass network structure. The presence of the ions with large radius and smaller polarizability result in the incompatible distortion of glass matrix structure [45,46]. Fig. 4(a) shows the cross-sectional FESEM images of glass samples ZTEr0.0 and ZTEr1.3 before and after exposure to HNO3. As the immersion period increases, a tellurite-rich corroded layer appears on the glass

Table 3 Judd–Ofelt intensity parameters (×10−20 cm2) and quality factor for Er3+-doped zinc tellurite glass along with those previously reported in literature.

This study, ZTEr1.0 Zinc-tellurite (0.1 mol%) [54] Zinc-tellurite [55] Zinc-tellurite [56] Zinc-tellurite [57] Tungsten-tellurite [56] Lead–zinc-tellurite [58] Zinc-borotellurite [59] Fluoro-tellurite [60] Cadmium phosphate [61] Fluoride phosphate [62] Metaphosphate [62] Fluoroindate [63] YAG ceramic [52] PLZT ceramic [49] Single crystal NaBi(WO4)2[53]

Ω2

Ω4

Ω6

5.02 4.67 5.93 6.58 5.56 6.84 4.97 4.96 4.935 1.11 4.71 6.22 2.17 0.42 3.20 5.50

2.21 1.28 1.50 1.80 1.58 1.33 1.66 2.02 3.148 9.34 1.61 1.52 2.31 0.70 0.35 1.0

2.04 0.79 1.07 1.03 1.32 1.10 0.73 2.39 1.745 7.31 1.62 1.02 0.89 0.47 0.41 0.70

surface. This evidence is in good agreement with the corrosion mechanism in acidic solution as previously discussed. The corroded layer for sample ZTEr1.3 is thicker in comparison to sample ZTEr0.0. Fig. 4(b) shows the cross-sectional FESEM images of glass samples ZTEr0.0 and ZTEr1.3 before and after exposure to distilled water. In this case, there is no corroded layer on the glass surface even after the immersing period was longer than 7 days. However, small amounts of the leached ions from the glass surface contribute to the solution. According to these cross-sectional images, the dissolution of glass in distilled water does not significantly affect the surface and the chemical durability of the zinc tellurite glass systems.

3.3. Optical absorption and luminescence properties Fig. 5 presents the representative optical absorption spectrum of the ZTEr1.0 glass in the range 350–1700 nm. The characteristic bands of Er3+ are observed and assigned to transitions from the ground state to the indicated excited states (4I15/2 → 4G11/2 at 380 nm, 4I15/2 → 2H9/2 at 407 nm, 4I 15/2 → 4 F5/2 , 4F 3/2 at 450 nm, 4 I15/2 → 4 F 7/2 at 488 nm, 4 I15/2 → 2H11/2 at 522 nm, 4I15/2 → 4S3/2 at 544 nm, 4I15/2 → 4F9/2 at 653 nm, 4 I 15/2 → 4 I 9/2 at 800 nm, 4 I 15/2 → 4 I 11/2 at 980 nm, and

Fig. 6. Excitation spectra (λem = 545 nm) of Er3+-doped zinc tellurite glasses, normalized to the band around 520 nm.

M.R. Dousti et al. / Journal of Non-Crystalline Solids 429 (2015) 70–78

75

Fig. 7. Emission spectra of Er3+-doped tellurite glasses: (a) visible (λexc = 377 nm), normalized to the band at 545 nm; (b) infrared (λexc = 980 nm), normalized.

4 I 15/2 → 4 I 13/2 at 1532 nm). Similar spectra are observed for the samples doped with other concentrations of erbium ions. The cut-off wavelength is around 360 nm. As evidenced by the linear dependence of integrated area at 980 nm, as a function of Er2O3 concentration, the studied zinc-tellurite glasses presented large rare earth solubility up to 3.0 mol% without transparency loss or evidences of clustering. The Judd–Ofelt (J.O.) analysis [47,48] is the widespread method used to evaluate the radiative properties of trivalent rare earth ions in different host matrices, through the determination of the phenomenological intensity parameters Ωλ (λ = 2, 4, 6 for Er3+). The procedure is based on equaling the expressions for the theoretical oscillator strength (fcalc), which contain the Ωλ parameters, to the experimental oscillator strengths (fexp) determined from the absorption spectra. This way, a system with as many equations as observed absorption bands is obtained and, by least square fitting, the three Ωλ values that best fulfill the criterion of minimum fitting error are determined [49,50]. These parameters can then be employed in the calculation of transition probabilities and consequently, of excited state lifetime values. Table 2 summarizes the values of experimental and calculated oscillator strengths determined for the representative ZTEr1.0 zinc tellurite glass sample. The small root mean square error (rmserror) evidences the good quality of the fit. The intensity parameters are given in Table 3 in comparison to those of other tellurite and oxide glasses. Although there is always some controversy in the attribution of the Ωλ values, it is usually accepted that the Ω2 value is related to the asymmetry around the rare earth ion and the RE-O covalence strength. On the other hand, the Ω4 and Ω6 parameters are related to the rigidity of the matrix [51]. In most of the cases seen in Table 3, the parameters follow the Ω2 N Ω4 N Ω6 trend except in a few cases such as for YAG:Er3+ ceramics with Ω4 N Ω6 N Ω2[52]. As for the comparison of our results to previously reported zinc-tellurite glasses, the values are quite similar, ranging from 5–6 × 10−20 cm2 and 1.5–2.3 × 10− 20 cm2 to Ω2 and Ω4, respectively. The Ω2 parameter strongly depends on the hypersensitive transition at 522 nm (4I15/ 2 2 2 → H11/2), since the reduced matrix element ||U(2)|| for this transition

is higher than other transitions involved in the calculation: 

f exp

 522 nm

∝0:7125Ω2 þ 0:4125Ω4 þ 0:0925Ω6 :

ð1Þ

Similarly, the Ω6 parameter is mostly affected by the near-infrared transition of Er3 + ions at 1532 nm as the reduced matrix element ||U(6)||2 for this transition is higher than others. 

f exp

 1532 nm

∝0:0195Ω2 þ 0:1173Ω4 þ 1:4316Ω6 :

ð2Þ

Fig. 6 presents the excitation spectrum of ZTEr glasses recorded by monitoring the emission centered at 545 nm. The spectra are normalized to the band at 520 nm. As observed, the normalized excitation spectra show a monotonic enhancement by increasing the concentration of Er3 +. In the range 350–530 nm, the excitation spectra reproduces quite well the absorption spectra, and the bands are labeled accordingly. Fig. 7(a) presents the visible emission spectra of Er3 +doped tellurite glasses under 377 nm excitation. The spectra are normalized to the band at 545 nm. The bands centered at 410, 525, 545 and 655 nm are attributed to transitions from the 2H9/2, 2H11/2, 4 S3/2 and 4F9/2 excited states to the 4I15/2 ground state. In addition, another band at around 820 nm is attributed to the 4S3/2 → 4I13/2 transition. By employing the calculated J.O. intensity parameters, radiative properties such as transition probabilities, branching ratios and the radiative excited states lifetime values were calculated and are presented in Table 4. The luminescence intensity ratios of the Er3 + emissions for each sample were extracted from Fig. 7(a) and are listed in Table 5. The emissions and energy transfer processes involved in the energy level diagram of Er3 + are schematized in Fig. 8. By and large, the ion-ion interactions between the Er3 + ions can result in four mechanisms of energy transfer: cooperative upconversion (CUC), energy migration (EM), cross-relaxation (CR), and excited state absorption (ESA) [64]. In Table 5, the intensity ratio of the band at 545 nm to the band at

Table 4 Spectroscopic properties of the ZTEr1.0 glass sample. Transition

Wavelength (nm)

A (s−1)

β (%)

4

410 520 545 820 670 1532

10,592.05 21,500.85 4656.84 2011.33 4724.46 411.15

79.30 93.39 65.62 28.34 90.95 100

F7/2 → 4I15/2 H11/2 → 4I15/2 4 S3/2 → 4I15/2 4 S3/2 → 4I13/2 4 F9/2 → 4I15/2 4 I13//2 → 4I15/2 2

τrad (μs) 75 43 141 193 2432

Table 5 Luminescence intensity ratio versus concentration of Er3+ ions in studied zinc tellurite glasses.

I545/I520 I545/I670 I545/I820 I545/I410

ZTEr0.5

ZTEr 1.0

ZTEr 1.5

ZTEr 2.0

ZTEr 2.5

4.26 27.62 29.80 19.88

4.41 21.79 20.46 14.32

4.52 15.19 10.44 9.63

4.76 13.5 8.80 6.83

5.41 11.86 7.63 5.52

76

M.R. Dousti et al. / Journal of Non-Crystalline Solids 429 (2015) 70–78

Fig. 8. Schematic energy level diagrams of Er3+ depicting energy transfer mechanisms of excited state absorption (ESA), energy migration (EM), cooperative upconversion (CUC) and cross-relaxation (CR) [55,64,65,67].

520 nm (I545/I520) shows a steady increase with increasing Er3+ concentration, which is due to the activation of the CR-4 process [66,67] (see Fig. 8) which depopulates the upper state 2H11/2. At the same time, the intensity ratio of the former band in the green to the one in the red (I545/I665) decrease monotonically due to activation of CR-1 which depopulates the 4S3/2 state, and CR-2 which favors the population of the red emitting level 4F9/2[65]. Interestingly, the I545/I820 also decrease monotonically which would in principle mean larger probability of 4S3/2 → 4I13/2 and thus be contradictory to the depopulation effect of CR-1 to 4S3/2. However, it ought to be remembered that the emission band at 820 nm has also a contribution of the transition 4I9/2 → 4I15/2 lying at the same energy range. The population of 4I9/2, and therefore the latter transition, are favored by the previously discussed CR-4 mechanism. Finally, the ratio I545/I410 also decreases with increasing Er3+ concentration because of the combined mechanisms CR-1, which depopulates 4S3/2, and the excited state absorption ESA-3 that favors the population of the blue emitting 2H9/2 level. Another possible crossrelaxation process is CR-3 (4I9/2,4I15/2 → 4I13/2,4I13/2) which favors the infrared emission at 1.53 μm [55]. The cooperative upconversion mechanisms CUC-1 and CUC-2 (which is the reverse cross-relaxation process of CR-3) are also proposed by Jackson [66] and Chai et al. [67]. Fig. 9(a) presents the luminescence decay curves of 4S3/2 → 4I15/2 emission (545 nm) excited at 377 nm. As shown in the inset, the excited state lifetime values decrease exponentially with increasing concentration of Er3+ due to the activation of the energy transfer mechanisms as discussed. The NIR normalized emission spectra of the Er3 +-doped ZnTe glasses are presented in Fig. 7(b). The FWHM of this band at 1.53 μm (4I13/2 → 4I15/2 transition) increases from 60.09 nm to 81.82 nm by increasing the dopant concentration from 0.5 to 2.5 mol%. Such broadening is due to the self-absorption (self-trapping) by Er3 + ions, as it has been previously reported by Mattarelli et al. [68,69] and Jaba et al. [58]. This mechanism is based on the energy migration through the sample, due to the large spectral overlap between the emission and absorption spectra of a given level. The collected PL intensity Id(λ) depends on the thickness of the sample, d and the absorption coefficient, α(λ) at any wavelength (λ) as Id(λ) = I0(λ) exp[− d · α(λ)], where I0 is the intensity of initially generated luminescence [68]. This way, the spontaneously emitted energy is reabsorbed in a consecutive mode and the probability of EM is proportional to the Er3 +

Fig. 9. Luminescence intensity decay profiles of ZTEr glasses. (a) Monitoring 4S3/2 → 4I15/2 transition (545 nm) with excitation at 377 nm. (b) Monitoring 4I13/2 → 4I15/2 transition (1.53 μm) with excitation at 980 nm.

concentration and the thickness of samples. In order to eliminate the thickness dependence, the NIR spectra in the current study were collected by preparing a thin layer of powdered tellurite glass samples (~1 mm). Fig. 9(b) presents the NIR luminescence decay curves from which the excited state lifetimes were determined. The values, shown in the graph, vary from 2.72 ms for the ZTEr0.5 sample to 0.87 ms for the ZTEr2.5 sample. Such concentration dependence of lifetime values is not unexpected, owning to the larger probability of energy transfer mechanisms. These results are in agreement with previously reported ones for various concentration of Er3+ ions [68,70]. The emission cross-section of the Er3+-doped tellurite glasses can be evaluated [71] by: σ em ¼

λ4p 8πcn2 Δλeff

A

where λP, Δλ and A are the peak position, effective bandwidth and radiative transition rate of the 4I13/2 → 4I15/2 transition, and n is the refractive index of the host (measured as 2.04 for λexc = 632.8 nm). Although, the refractive index depends on excitation wavelength and can vary by formation of more non-bridging oxygen at higher Er3+ concentrations, we have, by approximation, taken its value as a constant in the analyzed spectral range. However, because the effective bandwidth of the near-infrared transition slightly increases with increasing

M.R. Dousti et al. / Journal of Non-Crystalline Solids 429 (2015) 70–78

77

Fig. 10. (a) Upconversion spectra of ZnTeEr as a function of Er3+ concentration; (b) upconversion luminescence of the ZTEr1.0 glass sample obtained under 976 nm excitation at different pumping powers. Inset shows the log–log plots of luminescence vs. excitation power confirming the two photon absorption ESA-1 and ESA-2 processes.

Er 3 + concentration, the emission cross-sections proportionally decrease from 1.21 to 1.07, 1.02, 1.00 and 0.88 × 10− 20 cm2. These results are within the emission cross-sections of some lead–zinctellurite 0.877 × 10 − 20 cm2 [58], and more than phosphate 0.11 × 10 − 20 cm 2 [70], and bismuth–lead 0.103 × 10 − 20 cm 2 [36] glasses. Fig. 10(a) presents the upconversion emission of the ZnTeEr glasses excited at 976 nm (100 mW laser) normalized to the emission centered at 545 nm. As depicted in Fig. 8, excitation of the emitting levels 4S3/2 and 4F9/2 takes place mainly by ESA-1 and ESA-2 mechanisms. The resulting spectra are similar to those obtained by direct excitation in the visible, displaying three bands at 525, 550 and 670 nm, corresponding to transitions from 2H11/2, 4S3/2 and 4F9/2 excited states to the 4I15/2 ground state. The increase in the red to green ratio with increasing Er3 + concentration corroborates the previous discussion on energy transfer, since the activation of CR-1 causes depopulation of 4 S3/2 in favor of 4F9/2. The inset shows the linear dependence of the red to green emission ratio as a function of Er3 + concentration. Fig. 10(b) shows the excitation power dependence of upconversion intensities for the ZTEr1.0 sample. The inset shows the log I - log P plots for the emissions at 530, 550 and 670 nm. The slope of each fitted line is nearly 2, confirming that the excited state absorptions of two photons (ESA-1 and ESA-2) are the main mechanisms responsible for population of the emitting levels. Similar observations have been reported using 975–980 nm excitation wavelengths [72–76]. The processes involved in the upconversion emissions are schematized in the energy level diagram of Fig. 8. By absorption of 976 nm photons, the Er3 +-ions are first excited to the metastable 4I11/2 state. From there, two paths can happen: i) a second photon is absorbed (ESA-1) promoting excitation of the levels 2H11/2 and 4S3/2 state; ii) the 4I11/2 decays non-radiatively to 4I13/2 from which absorption of a 976 nm photon can also happen (ESA-2) to populate 4F9/2. 4. Conclusions The structural, thermal, chemical and spectroscopic properties of zinc-tellurite glasses with composition 20ZnO–80TeO2 doped with up to 2.5 mol% Er2O3 were studied in detail. The glasses can be easily prepared at relatively low temperature and they display high refractive index, large density, high rare earth solubility, low maximum phonon energy and excellent luminescence properties in the visible and nearinfrared spectral regions. The Judd–Ofelt intensity parameters were calculated for the sample doped with 1 mol% Er2O3 as Ω2 = 5.02 × 10− 20 cm 2, Ω4 = 2.21 × 10−20 cm2 and Ω6 = 2.04 × 10−20 cm2 and are in good agreement with similar studies. The bandwidth of the near-infrared emission of the Er3 + ions in zinc

tellurite glasses increases from 60.09 nm 81.82 nm by increasing the dopant concentration from 0.5 to 2.5 mol% due to self-trapping effect. The red to green upconversion emission intensity ratio of the samples increases by increasing the Er3 + concentration thanks to the energy transfer among neighboring ions. Based on the results we propose that zinc-tellurite glasses are promising low phonon hosts for rareearth elements that can also become potential candidates for the development of solid-state lasers, near infrared sensors, modern lighting devices, optical displays and optical fibers. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.jnoncrysol.2015.07.039.

Acknowledgments The authors are thankful to FAPESP-Brazil (Project Nos. 13/24064-8 and CEPID 13/07793-6),CNPq Universal project (479672/2012-1), Universiti Teknologi Malaysia (Grant Nos. GUP 05H45 and FRGS 4 F319) and Universiti Teknologi Mara Cawangan Pulau Pinang for the grant of Excellence Fund (600-RMIIST/DANA 5/3/Dst(318/2011)).

References [1] J.C. Mauro, E.D. Zanotto, Int. J. Appl. Glas. Sci. 15 (2014) 313–327. [2] J.H. Campbell, J.S. Hayden, A. Marker, Int. J. Appl. Glas. Sci. 2 (2011) 3–29. [3] M. Irannejad, T. Fernandez, P. Joshi, Z. Zhao, J. Lousteau, B. Richards, G. Jose, A. Jha, Int. J. Appl. Glas. Sci. 4 (2013) 202–213. [4] H. Li, M.J. Davis, A.J. Marker, J.S. Hayden, Int. J. Appl. Glas. Sci. 1 (2010) 63–73. [5] K. Richardson, H. Ebendorff-Heidepriem, Int. J. Appl. Glas. Sci. 3 (2012) 287–288. [6] a Kenyon, Recent Developments in Rare-Earth Doped Materials for Optoelectronics, 2002. [7] R.K. Brow, J. Nanophotonics 263–264 (2000) 1–28. [8] E.I. Kamitsos, G.D. Chryssikos, J. Mol. Struct. 247 (1991) 1–16. [9] R. Reisfield, M. Eyal, C.K. Jorgensen, J. Less-Common Met. 126 (1986) 187–194. [10] Y. Guo, Y. Tian, L. Zhang, L. Hu, J. Zhang, J. Non-Cryst. Solids 377 (2013) 119–123. [11] G.W. Brady, J. Chem. Phys. 27 (1957) 300–303. [12] O. Lindqvist, Acta Chem. Scand. 22 (1968) 977–982. [13] O. Noguera, T. Merle-Méjean, A.P. Mirgorodsky, M.B. Smirnov, P. Thomas, J.C. Champarnaud-Mesjard, J. Non-Cryst. Solids 330 (2003) 50–60. [14] S. Kim, T. Yako, S. Saka, J. Am. Ceram. Soc. 76 (1993) 2486–2490. [15] H. Yamamoto, H. Nasu, J. Matsuoka, K. Kamiya, J. Non-Cryst. Solids 170 (1994) 87–96. [16] H.A.A. Sidek, S. Rosmawati, Z.A. Talib, M.K. Halimah, D.M. Daud, J. Appl. Sci. 6 (2009) 1489–1494. [17] U. Hoppe, E. Yousef, C. Rüssel, J. Neuefeind, A.C. Hannon, J. Phys. Condens. Matter 166 (2004) 1645–1663. [18] M.R. Sahar, A.K. Jehbu, M.M. Karim, J. Non-Cryst. Solids 213–214 (1997) 164–167. [19] M.R. Sahar, N. Noordin, J. Non-Cryst. Solids 184 (1995) 137–140. [20] M.R. Sahar, K. Sulhadi, M.S. Rohani, J. Non-Cryst. Solids 354 (2008) 1179–1181. [21] R.A.H. El-mallawany, Tellurite Glasses Handbook: Physical Properties and Data, CRC Press, Boca Raton, 2002. [22] J. Stanworth, Soc. Glas. Technol. 36 (1952) 217T–241T. [23] Z.A.S. Mahraz, M.R. Sahar, S.K. Ghoshal, Chalcogenide Lett. 11 (2014) 453–460.

78

M.R. Dousti et al. / Journal of Non-Crystalline Solids 429 (2015) 70–78

[24] M.R. Dousti, P. Ghassemi, M.R. Sahar, Z.A. Mahraz, Chalcogenide Lett. 11 (2014) 111–119. [25] M.R. Dousti, M.R. Sahar, S.K. Ghoshal, R.J. Amjad, A.R. Samavati, J. Mol. Struct. 1035 (2013) 6–12. [26] H. Lin, G. Meredith, S. Jiang, X. Peng, T. Luo, N. Peyghambarian, E.Y.B. Pun, J. Appl. Phys. 93 (2003) 186. [27] X. Shen, Q. Nie, T. Xu, T. Peng, Y. Gao, Phys. Lett. A 332 (2004) 101–106. [28] X. Shen, Q. Nie, T. Xu, S. Dai, X. Wang, Phys. B Condens. Matter 381 (2006) 219–223. [29] L. Dong, S. Huang, H. Wen, K. Kang, X. Duan, J. Rare Earths 27 (2009) 334–337. [30] A. Miguel, R. Morea, J. Gonzalo, M.A. Arriandiaga, J. Fernandez, R. Balda, J. Lumin. 140 (2013) 38–44. [31] M.R. Dousti, M.R. Sahar, S.K. Ghoshal, R.J. Amjad, R. Arifin, J. Non-Cryst. Solids 358 (2012) 2939–2942. [32] S. Hj, A. Aziz, S.S. Badaron, H.M. Kamari, K.A. Matori, Adv. Mater. Sci. Eng. 2015 (2015). [33] R.J. Amjad, M.R. Sahar, M.R. Dousti, S.K. Ghoshal, M.N.A. Jamaludin, Opt. Express 21 (2013) 14282–14290. [34] A. Awang, S.K. Ghoshal, M.R. Sahar, R. Arifin, F. Nawaz, J. Lumin. 149 (2014) 138–143. [35] M.R. Dousti, M.R. Sahar, S.K. Ghoshal, R.J. Amjad, R. Arifin, J. Mol. Struct. 1033 (2013) 79–83. [36] Y.G. Choi, K.H. Kim, J. Heo, J. Am. Ceram. Soc. 68 (1999) 2762–2768. [37] Y. Ding, S. Jiang, B. Hwang, T. Luo, Opt. Mater. 15 (2000) 123–130. [38] J. Yang, S. Dai, Y. Zhou, L. Wen, L. Hu, Z. Jiang, J. Appl. Phys. 93 (2003) 977–983. [39] G.C. Righini, S. Pelli, M. Brenci, M. Ferrari, C. Duverger, M. Montagna, R. Dall'lgna, J. Non-Cryst. Solids 284 (2001) 223–229. [40] Desurvire, Erbium Doped Fiber Amplifiers, Wiley Interscience, 1994. [41] A. Nukui, T. Taniguchi, M. Miyata, J. Non-Cryst. Solids 293–295 (2001) 255–260. [42] B.J. Riley, B.T. Rieck, J.S. McCloy, J.V. Crum, S.K. Sundaram, J.D. Vienna, J. Nucl. Mater. 424 (2012) 29–37. [43] G. Tammann, Textbook of Metallography, Chemical Catalog Company, Leipzig, 1923. [44] U.R. Evans, The Corrosion and Oxidation of Metals: Scientific Principles and Practical Applications, Edward Arnold (Publishers) Ltd, London, 1960. [45] K. Hamzah, M.A.I. Yassin, M.R. Sahar, S.K. Ghoshal, R. Arifin, M.S. Rohani, Adv. Mater. Res. 501 (2012) 81–85. [46] M. Xianfeng, Z. Qitu, L. Chunhua, X. Zhongzi, J. Rare Earths 25 (2007) 125–128. [47] B.R. Judd, Phys. Rev. 127 (1962) 750–761. [48] G.S. Ofelt, J. Chem. Phys. 37 (1962) 511–520. [49] A.S.S. de Camargo, L.A.O. Nunes, I.A. Santo, D. Garcia, J.A. Eiras, J. Appl. Phys. 95 (2004) 2135.

[50] B.M. Walsh, Judd–Ofelt theory: principles and practices, Adv. Spectrocopy Lasers Sens 2006, pp. 403–433. [51] C.K. Jørgensen, R. Reisfeld, J. Less-Common Met. 93 (1983) 107–112. [52] M. Pokhrel, G.A. Kumar, P. Samuel, K.I. Ueda, T. Yanagitani, H. Yagi, D.K. Sardar, Opt. Mater. Express 1 (2011) 1272. [53] D.K. Sardar, C.C. Russell, R.M. Yow, J.B. Gruber, B. Zandi, E.P. Kokanyan, J. Appl. Phys. 95 (2004) 1180–1184. [54] S. Dai, J. Zhang, C. Yu, G. Zhou, G. Wang, L. Hu, Mater. Lett. 59 (2005) 2333–2336. [55] N. Jaba, A. Kanoun, H. Mejri, A. Selmi, S. Alaya, H. Maaref, J. Phys. Condens. Matter 12 (2000) 4523–4534. [56] Y. Ma, Y. Guo, F. Huang, L. Hu, J. Zhang, J. Lumin. 147 (2014) 372–377. [57] L. Wu, B. Huang, F. Yang, Y. Qi, S. Peng, Y. Zhou, Opt. Mater. 43 (2015) 42–48. [58] Z. Jin, Q. Nie, T. Xu, S. Dai, X. Shen, X. Zhang, Mater. Chem. Phys. 104 (2007) 62–67. [59] Z.A.S. Mahraz, M.R. Sahar, S.K. Ghoshal, M.R. Dousti, J. Lumin. 144 (2013) 139–145. [60] A. Ghosh, R. Debnath, Opt. Mater. 31 (2009) 604–608. [61] K.H. Mahmoud, Z.M. El-Bahy, A.I. Hanafy, J. Non-Cryst. Solids 363 (2013) 116–120. [62] H. Ebendorff-Heidepriem, M. Ehrt, D. Bettinelli, A. Speghini, J. Non-Cryst. Solids 240 (1998) 66–78. [63] A. Florez, Y. Messaddeq, O.L. Malta, M.A. Aegerter, J. Alloys Compd. 227 (1995) 135–140. [64] A. Kenyon, Prog. Quantum Electron. 26 (2002) 225–284. [65] J.L. Ferrari, K. de Oliveira Lima, E. Pecoraro, R.A.S. Ferreira, L.D. Carlos, R.R. Gonçalves, J. Mater. Chem. 22 (2012) 9901. [66] S.D. Jackson, Nat. Photonics 6 (2012) 423–431. [67] G. Chai, G. Dong, J. Qiu, Q. Zhang, Z. Yang, Sci. Rep. 3 (2013) 1598. [68] M. Mattarelli, M. Montagna, L. Zampedri, A. Chiasera, M. Ferrari, G.C. Righini, L.M. Fortes, M.C. Gonçalves, L.F. Santos, R.M. Almeida, Europhys. Lett. 71 (2005) 394–399. [69] N. Jaba, H.B.B. Mansour, B. Champagnon, Opt. Mater. 31 (2009) 1242–1247. [70] F. Rivera-López, P. Babu, L. Jyothi, U.R. Rodríguez-Mendoza, I.R. Martín, C.K. Jayasankar, V. Lavin, Opt. Mater. 34 (2012) 1235–1240. [71] Y. Wang, C. Tu, Z. You, J. Li, Z. Zhu, B. Wu, Opt. Mater. 29 (2006) 257–261. [72] X. Pan, J. Yu, Y. Liu, S. Yoda, M. Zhang, F. Ai, F. Jin, H. Yu, W. Jin, J. Lumin. 132 (2012) 1025–1029. [73] H. Lin, K. Liu, E.Y.B. Pun, T.C. Ma, X. Peng, Q.D. An, J.Y. Yu, S.B. Jiang, Chem. Phys. Lett. 398 (2004) 146–150. [74] A. Pandey, S. Som, V. Kumar, V. Kumar, K. Kumar, V.K. Rai, H.C. Stwart, Sensors Actuators B Chem. 202 (2014) 1305–1312. [75] A.J. Garcia-adeva, Transition 21 (2004) 744–752. [76] A.S.S. de Camargo, J.F. Possatto, L.A.O. Nunes, É.R. Botero, É.R.M. Andreeta, D. Garcia, J.A. Eiras, Solid State Commun. 137 (2006) 1–5.