Oxyfluorotellurite glasses doped by dysprosium ions. Thermal and optical properties

Optical Materials 42 (2015) 538–543

Contents lists available at ScienceDirect

Optical Materials journal homepage: www.elsevier.com/locate/optmat

Oxyfluorotellurite glasses doped by dysprosium ions. Thermal and optical properties Barbara Klimesz a,⇑, Witold Ryba-Romanowski b, Radosław Lisiecki b a b

Department of Physics, Opole University of Technology, ul. Prószkowska 76, 45-758 Opole, Poland Institute of Low Temperature and Structure Research, Polish Academy of Sciences, ul. Okólna 2, 50-395 Wrocław, Poland

a r t i c l e

i n f o

Article history: Received 8 November 2014 Received in revised form 14 January 2015 Accepted 10 February 2015 Available online 26 February 2015 Keywords: Optical properties Optical spectroscopy Rare earth-doped oxyfluoride glasses Thermal properties Glass transitions

a b s t r a c t The paper shows results of investigation of thermal and optical properties of oxyfluorotellurite (65  x)TeO2–20ZnF2–12Pb2O5–3Nb2O5–xDy2O3 (x = 0.5, 2 and 5) glass systems. Thermal stability and the onset of crystallization of the materials were monitored by differential thermal analysis (DTA). It was found that characteristic parameters, namely glass transition temperatures (Tg), onset of crystallization temperatures (Tc) and thermal stability criteria DT and H’ increased with increasing Dy2O3 content indicating that the incorporation of dysprosium ions improves substantially thermal stability of glass system under study. Optical absorption and emission spectra of Dy3+ ions in oxyfluorotellurite glass were investigated at room temperature in the visible (VIS) and near-infrared (NIR) region. Oscillator strengths, phenomenological Judd–Ofelt (JO) intensity parameters X2,4,6, radiative transition probabilities, branching ratios and radiative lifetimes of luminescent levels were determined. Decay curves of the 4F9/2 luminescence of incorporated Dy3+ ions were recorded and analysed. Lifetimes and the luminescence dynamics were studied as a function of the Dy2O3 concentration. It was concluded that good thermal stability combined with desirable spectroscopic parameters of investigated dysprosium-doped oxyfluorotellurite glass point at the suitability of this material for the design of UV-excited visible phosphors. Ó 2015 Elsevier B.V. All rights reserved.

1. Introduction Oxyfluorotellurite glasses combine advantages of the two (fluoride and tellurite) glass systems and that is reason why they were subject of scientific and technological interest [1–6]. Due to their good thermal stability [1–6] and chemical durability [1–5], non hygroscopicity, high refractive index [1–4,6], low phonon energies [1–3,5,6], wide optical transmittance region [1,2,4], good transparency in the mid-infrared [3,5,6] and also high solubility for rare earth ions [3,6] oxyfluorotellurite glasses are good candidates for application and development of optoelectronics (colour displays), photonics (fibre amplifiers), telecommunications (fibreoptic communications, optical amplifiers) or laser techniques (solid-state lasers, laser fibres) [1–6]. Rare earth (RE) ions play an important role in modern technology as active ions in many optical materials. It is common knowledge that the optical properties of rare-earth ions in glasses depend on the chemical composition of the glass matrix, which determines the structure and nature of the bonds [7]. Among different lanthanides, dysprosium ion (Dy3+) is of interest ions ⇑ Corresponding author. Tel.: +48 77 4498837. E-mail address: [email protected] (B. Klimesz). http://dx.doi.org/10.1016/j.optmat.2015.02.012 0925-3467/Ó 2015 Elsevier B.V. All rights reserved.

because it is able to show emissions in the visible (VIS) and near infrared (NIR) regions. The visible fluorescence of Dy3+ ion consists of two strong emission transitions: one in the yellow (560– 600 nm, 4F9/2 ? 6H13/2) and the other one in blue (470–500 nm, 4 F9/2 ? 6H15/2) regions. As a consequence dysprosium-doped glasses are promising for application in solid state visible lasers [8,9], up converters and optical amplifiers [9], commercial display devices [10] and through appropriately combination of yellow to blue emission could result in white light emission [8–11]. Knowledge on their thermal behaviour and spectroscopic features is thus relevant to the practice. In the present paper we report on the synthesis, thermal stability and fundamental spectroscopic properties of Dy3+-doped oxyfluorotellurite glass with chemical composition (65  x)TeO2– 20ZnF2–12PbO–3Nb2O5–xDy2O3 where x = 0.5, 2 and 5 mol%. The thermal stability parameters, oscillator strength, Judd–Ofelt intensity parameters, branching ratios and radiative transition probabilities were calculated and compared with available literature data. As far as we know, thermal and spectroscopic features of this system have not been reported before. Detailed investigation of the undoped glass matrix has been reported by Guihua Liao et al. [2] and the information gathered therein was used as a reference point for the present work.

539

B. Klimesz et al. / Optical Materials 42 (2015) 538–543

2. Experimental

(65-x)%TeO2-20%ZnF2-12%PbO-3%Nb2O5-x%Dy2O3 o

T g= 378,1 C

3. Results and discussion 3.1. Thermal analysis In Fig. 1 the DTA curves recorded for (65  x)TeO2–20ZnF2– 12Pb2O5–3Nb2O5–xDy2O3 samples glasses with 0.5, 2 and 5 mol ratio of Dy2O3 and for oxyfluorotellurite matrix glass 65TeO2– 20ZnF2–12Pb2O5–3Nb2O5 are compared. The characteristic glass temperatures, namely the glass transition temperatures (Tg) and onset of crystallization temperatures (Tc) were estimated in accordance with the Keavney and Eberlin method [12]. It can be seen in Fig. 1 that the value of Tg indicating initiation of the glass softening is affected by incorporated Dy3+ ions. The value of Tg for undoped glass matrix is the lowest and equals to 365.3 °C. In doped samples it grows with increasing Dy3+ concentration and amounts to 368.6 °C, 371.3 °C and 378.1 °C for samples containing 0.5, 2, 5 mol% of Dy2O3, respectively. Evaluated values of the glass crystallization temperature Tc follow the same trend with a more pronounced increase from 552.3 °C for undoped glass to 830.5 °C for a glass sample containing 5 mol% of Dy2O3. For all samples of glasses (65  x)TeO2–20ZnF2–12Pb2O5–3Nb2O5–xDy2O3 with x = 0.5, 2 and 5 mol% of Dy2O3 the exothermal crystallization peak Tpc values are higher than that for undoped glass matrix and appear between 700 and 860 °C.

(as-melted)

x=5 3 T g= 371,

x=2

8,6 T g= 36

o

C

o

C

o

Tpc= 588,5 C *

T g= 36

o 5,3 C

o

x = 0.5

Tc= 830,5 C

DTA (a.u.)

The glass samples were prepared from a mixture of high purity (4 N or 5 N, Alfa Aesar) powders of tellurium (TeO2), niobium (Nb2O5) and lead (PbO) oxides, anhydrous zinc fluoride (ZnF2) and 5 N-purity dysprosium oxide (Dy2O3). Dy3+- doped glasses have the following chemical compositions (in mol%) (65  x)TeO2– 20ZnF2–12PbO–3Nb2O5–xDy2O3 with nominal concentrations x = 0.5, 2 and 5 mol% of Dy2O3. Thoroughly mixed in dry box 10 g batches of the starting substrates were placed in corundum crucible and then melted in a resistance furnace at 830 °C for 30 min in normal atmosphere. The melt was poured onto a preheated steel plate and then was annealed for a few hours below the glass transition temperature (Tg) in order to eliminate internal stresses. Differential thermal analysis (DTA) measurement was performed in atmospheric air under normal pressure using a NETZSCH differential scanning calorimeter DSC 404/3/F with Pt/PtRh DSC measuring head and platinum sample pans. An empty platinum crucible was used as the reference. The heating rates of 10 K/min in the DTA measurements were the same for all samples studied. Optical absorption spectra were recorded with a Varian 5 Absorption Spectrophotometer employing spectral bandwidth of 0.2 nm in UV and visible and of 0.8 nm in near infrared. Emission measurements were carried out in the visible and near infrared spectral range. Luminescence spectra were recorded with an Optron Fluorometer System consisting of 150 W xenon lamp coupled to an excitation monochromator, an emission monochromator with 750 mm focal length equipped with a photomultiplier and InGaAs detector and a signal recovering unit. Luminescence decay curves were recorded following a short pulse excitation provided by an optical parametric oscillator (OPO) pumped by a third harmonic of a Continuum Model Surelite I Nd:YAG laser. Resulting luminescence signal was filtered using a Zeiss model GDM-1000 monochromator, detected by a Hamamatsu R928 photomultiplier and recorded with a Tektronix TDS 3052 oscilloscope. All optical measurements were carried out at room temperature and in atmospheric air under normal pressure. Values of the density q = 5.76 g/cm3 and refractive index n = 1.9813 of the glass have been taken from the literature [2].

x=0 o

Tpc= 700,9 C

T= c

100

200

300

400

500

600

679

* o

Tpc= 848,1 C *

*

o

Tc= 552,3 C

o

Tpc= 863,1 C

o

T = 804,6 C ,4 oC c

700

800

900

1000

o

Temperature ( C) Fig. 1. Differential thermal analysis (DTA) curves recorded for 65TeO2–20ZnF2– 12Pb2O5–3Nb2O5 glass matrix and Dy3+- doped glass samples (65  x)TeO2– 20ZnF2–12Pb2O5–3Nb2O5–xDy2O3 with x = 0.5, 2 and 5.

The increase of all characteristic glass temperatures, especially the glass crystallization temperatures (Tc) could be the result of the increased in number of bonds per unit volume and an increased density with the addition of Dy2O3 from 0.5 to 5 mol% into TeO2–ZnF2–Pb2O5–Nb2O5 glasses. This behaviour may be connected with a fact that Dy3+ ions are capable of breaking existing oxygen bonds (e.g.: Te-O-Te, Te-O-Pb, Te-O-Nb and etc.) linkages and building new bonds with oxygen atoms in the Dy2O3 containing glasses like Te-O-Dy, Dy-O-Pb or Dy-O-Nb. Jyothi et al. hint that the increasing the number of bonds per unit volume should be combined with the substitution of TeO2 by Dy2O3 content to provide two dysprosium and three oxygen ions in the place of one tellurium and two oxygen ions [10]. Dy2O3 content has greater number of cations per mol and average cross-link density and this consequently leads to an increase of the number of bonds per unit volume [13]. Similar behaviour for both the glass transition and the glass crystallization temperatures have been reported for other oxyfluorotellurite glass systems, e.g.: TeO2–BaO–BaF2–La2O3– LaF3, (60  x)TeO2–20ZnO–20LiF–xEr2O3, (75  x)TeO2–10TiO2– 15WO3–xDy2O3, 80TeO2–5TiO2–(15  x)WO3–xAnOm (where AnOm is Nb2O5, Nd2O3, and Er2O3) or (1  x)TeO2–xWO3 doped with Nd2O3, Er2O3, Tm2O3 and Yb2O3 [1–3,10,13,14]. It follows from these data that incorporated rare earth ions influence the ability of glass formation and thermal stability of the glass. The glass stability can be determined qualitatively based on a difference of thermal stability criteria:

DT ¼ T c  T g H0 ¼

S¼

Tc  Tg Tg

Dietzel factor ½15;

ð1Þ

Saad  Poulain factor ½16;

ð2Þ

ðT c  T g Þ ðT pc  T c Þ Tg

Saad  Poulain factor ½16;

ð3Þ

remembering that the larger are values of these parameters the higher is thermal stability of the glass. The DT, H’ and S values for the systems under study, listed in Table 1, imply that the incorporation of Dy2O3 brings about a substantial improvement of thermal stability of the 65TeO2–20ZnF2–12Pb2O5–3Nb2O5 glass. It should be noticed here that in contrast to DT and H’ parameters the S parameter values found do not change monotonously with increasing Dy2O3 content. In our opinion this is due to high incertitude of Tpc values that follow from broad and asymmetric shapes of the matrix crystallization bands. In addition, the Tpc value also depends on

540

B. Klimesz et al. / Optical Materials 42 (2015) 538–543

Table 1 0 Characteristic temperatures (Tg, Tc, Tpc) and the DT, H , S values estimated for 65TeO2–20ZnF2–12Pb2O5–3Nb2O5glass matrix and (65  x)TeO2–20ZnF2–12Pb2O5–3Nb2O5–xDy2O3 (x = 0.5, 2, 5 mol%) glassy systems. 0

Composition (%mol)

Tg (°C)

Tc (°C)

Tpc (°C)

DT (°C)

H

65TeO2–20ZnF2–12Pb2O5–3Nb2O5 64.5TeO2–20ZnF2–12Pb2O5–3Nb2O5–0.5Dy2O3 63TeO2–20ZnF2–12Pb2O5–3Nb2O5–2Dy2O3 60TeO2–20ZnF2–12Pb2O5–3Nb2O5–5Dy2O3

365 369 371 378

552 679 805 830

588 701 848 863

187 310 434 452

0.51 0.84 1.17 1.20

technical parameters of experiment (mass and heating rate of sample) as well as on constructional features of apparatus [17].

Table 2 Experimental (Pexp) and calculated (Pcal) oscillator strengths for the Dy3+ doped oxyfluorotellurite glasses 60TeO2–20ZnF2–12Pb2O5–3Nb2O5–5Dy2O3.

3.2. Analysis of optical absorption spectra Fig. 2 shows optical absorption spectra for 60TeO2–20ZnF2– 12Pb2O5–3Nb2O5–5Dy2O3 glass sample in the UV–VIS and NIR regions. The former part contains weak, albeit well defined bands located between 21,000 and 24,000 cm1 (480–420 nm), corresponding to transitions from the 6H15/2 ground state to the 4F9/2, 4 I15/2 and 4G11/2 quartet states. Absorption bands related to transitions to higher energy levels of Dy3+ are hidden by rising host matrix absorption (presumably colour centre absorption) except for relatively intense bands around 25,800 cm1 (390 nm) and 27,400 cm1 (370 nm) that we assign to the 6H15/2 ? 4M21/2,

S 18.44 18.48 50.30 39.46

Transition from 6 H15/2

Energy m (cm1)

6

5929 9132 11,086 12,450 13,270 21,124 22,065

H11/2 6 F9/2, 6H7/2 6 H5/2, 6F7/2 6 F5/2 6 F3/2, 6F1/2 4 F9/2 4 I15/2

Oscillator strengths P  106 Pexp

Pcal

2.29 4.04 4.90 2.10 0.28 0.21 0.78

2.48 4.04 4.60 2.60 0.50 0.39 0.94

X2 = 2.32  1020 cm2X4 = 0.64  1020 cm2X6 = 4.64  1020 cm2

4

650

600

550

500

450

400 4

40

4

4

P3/2, P5/2, D3/2, I11/2 M21/2, I13/2, F7/2, K17/2

60%TeO2-20%ZnF2-12%PbO-3%Nb2O5

4

-5%Dy2O3

4

30

350

6

4

10 6

H15/2

F9/2

4

4

20

4

I15/2 4G11/2

UV-VIS

0 16000

20000

1800 1500

24000

1200

900 6

H15/2

-1

(cm )

40

6

30

20

6

H9/2, F11/2

6

6

0.5% 2% 5%

H11/2

2

1600

1700

1800

(nm) 6

10

6

4

0

H15/2

28000

H11/2

6

I13/2, 4F7/2, 4K17/2, 4M19/2 and 6H15/2 ? 4P3/2, 6P5/2, 4D3/2, 4I11/2 transitions, respectively. The second part consists of five absorption bands related to transitions from the 6H15/2 ground state to the 6HJ (J = 11/2, 9/2, 7/2, 5/2) and 6FJ (J = 11/2, 9/2, 7/2) excited manifolds in the infrared region between 5900 and 14,000 cm1 (1700–720 nm). Spectra for samples containing 2 and 0.5 mol% of Dy2O3, that were not shown, are similar except for the band intensities, which are proportionally smaller. Recorded optical absorption spectra were used to evaluate experimental oscillator strengths Pexp by numerical integration of absorption bands related to transitions from the ground 6H15/2 manifold to excited states and making use of the relation:

Pexp ¼ 4:318  109

Z

eðmÞdm

where e(m) denotes the molar extinction and m denotes the energy expressed in wavenumbers. All transitions considered are assumed to be electric dipole in nature, except for the 6H15/2 ? 4I15/2 transition which has a magnetic dipole component. Corresponding contribution of magnetic dipole transition Pmd = 0.19  107 was calculated using standard formula [18] and subtracted from Pexp prior to the Judd–Ofelt treatment. When absorption bands related to different excited J manifolds overlap, the unresolved intensity was assigned to the group of levels. The experimental oscillator strength related to selected f–f transition of optically active ion is determined taking into account the Ln3+ concentration. Inset of

6

6 F9/2, H7/2 H5/2, F7/2

6

F5/2

NIR 6

0

6

F3/2, F1/2

Table 3 Xt parameters (1020 cm2) estimated for several glass systems. Glass system

6000

8000

10000

ð4Þ

12000

14000

Fig. 2. Optical absorption spectra for 60TeO2–20ZnF2–12Pb2O5–3Nb2O5–5Dy2O3 glass samples recorded at 300 K in VIS (visible) and NIR (infrared) regions. Inset displays the 6H15/2–6H11/2 absorption bands of dysprosium measured for samples doped with various Dy3+ concentration.

NaPO3–CaF2–DyF3 PbO–B2O3–Al2O3–WO3–Dy2O3 BaO–TeO2:Dy ZBLAN:Dy TeO2–ZnF2–Pb2O5–Nb2O5–Dy2O3

J–O parameters

Refs.

X2

X4

X6

3.36 4.90 3.20 1.86 2.32

2.03 0.94 1.35 1.42 0.64

1.89 2.07 2.47 2.37 4.64

[18] [22] [23] [24] This work

541

B. Klimesz et al. / Optical Materials 42 (2015) 538–543

Fig 2 presents the 6H15/2 ? 6H11/2 absorption bands measured for the samples containing 0.5, 2 and 5 mol% of Dy2O3, respectively. It can be found that the absorption coefficient of considered absorption band properly increases with elevating of dysprosium content. Because of prominent spectroscopic quality of dysprosium absorption bands for materials under study are comparable except of their intensity, the Judd–Ofelt calculation was prepared for one selected glass doped with 5 mol% of Dy2O3. Adopting the Judd–Ofelt procedure [19,20] the evaluated Pexp were equated to corresponding theoretical oscillator strengths: 2 X 8p2 mc ðn2 þ 2Þ  Xt jhf N ½L; SJ kU t kf N ½L0 ;S0  J0 ij2 3hk ð2J þ 1Þ 9n t¼2;4;6

ð5Þ

6

4

500

H

13/2

0.68

F9/2

6 6

H

12000

6 9/2 0.01

H

15/2

0.29 H

11/2 0.02

14000

16000

18000

20000

22000

24000

-1

Wavenumber [cm ]

SLJ

6

H15/2 H13/2 6 H11/2 6 H9/2, 6F11/2 6 F9/2, 6H7/2 6 H5/2, 6F7/2 6 F5/2 6 F3/2, 6F1/2 6

Average wavelength (lm)

Wr (s1)

0.477 0.571 0.658 0.752 0.834 0.996 1.153 1.273

754 1281 107 40 3 7 2 2

bther.

0.34 0.61 0.03 0.02 0.00 0.00 0.00 0.00

bexp.

0.29 0.68 0.02 0.01 – – – –

sr (ls) 453

Fig. 3. Emission spectrum of the glass doped with 5 mol% of Dy2O3 recorded at 300 K upon 393 nm excitation. Numbers indicate experimental branching ratios.

these hypotheses we deduce that among glass systems gathered in Table 3 the rigidity of the system under study is the lowest. 3.3. Emission characteristics Phenomenological parameters X2,4,6, derived from the analysis of optical absorption spectra were used to calculate radiative transition rates:

Wr ¼

2 n ðn2 þ 2Þ X N N Xt jhf ½L0 ;S0 J0 kU t kf ½L; SJij2 ; 9 3hk3 ð2J þ 1Þ t¼2;4;6

64p4 e2

ð6Þ

luminescence branching ratios b defined as:

W r ði; jÞ ¼ sr W r ði; jÞ b¼X W r ði; jÞ

ð7Þ

j

where the summation is over electric dipole transitions to all terminal j states and radiative lifetimes sr:

1

sr

¼

X W r ði; jÞ:

ð8Þ

j

5

e

x = 0.5

4

e

Luminescence intensity [a.u.]

Table 4 Radiative transition rates Wr, luminescence branching ratios b and radiative lifetime sr calculated for the 4F9/2 luminescent level of Dy3+ in 60TeO2–20ZnF2–12Pb2O5– 3Nb2O5 glass.

F9/2 ?

600

N

jhf ½L; S J k U t kf ½L0 ; S0  J0 ij are double reduced matrix elements of unit tensor operators given in Ref. [21]. Next, three phenomenological parameters X2,4,6 were determined by a least square fit between experimental and theoretical oscillator strengths. Results of fitting procedure are gathered in Table 2. Obtained J–O intensity parameters are as follows: X2 = 2.32  1020 cm2, X4 = 0.64  1020 cm2 and X6 = 4.64  1020 cm2. In Table 3 the obtained intensity parameters for our oxyfluorotellurite glasses are compared to those for other glasses doped by dysprosium. It can be seen that the magnitude of X2 for the system under study is clearly higher than X4 but smaller than X6 parameter. Based on JO Xt parameters several important optical qualities may be determined, for instance considering spontaneous emission probability or branching ratio of luminescence. For glass, relation between these spectroscopic parameters and optically active ion’s local positions is inconclusive due to random structure of amorphous material. On the other hand, it is rather clear that Xt parameters depending on glass composition are affected by gain a higher transition probability or more efficient laser performance. It has been pointed out [18] that there is a certain correspondence between magnitude of phenomenological parameters X2,4,6 and the local structure and bonding in the vicinity of incorporated rare earth ions. In particular, high value of the X2 parameter indicates that chemical bonds have significant covalent character. Besides, X2 is very sensitive to the ligand environment of rare earth ion’s coordination sphere. In relation to that, high value of X2 is expected for effectively polarized and asymmetric environment around Ln3+ ions [29]. Data collected in Table 3 demonstrates that asymmetry of Dy3+ local position in our oxyfluorotellurite glass is quite moderate. X4,6 parameters depends less on RE ion environment than X2 [30]. However these parameters may be related to electron density on the oxide ions especially on main component of glass [31]. Furthermore, low value of X6 parameter indicates high rigidity of the glass system. Based on

4

700

60%TeO2-20%ZnF2-12%PbO-3%Nb2O5-5%Dy2O3

where m is the electron mass, c – speed of light, h – Planck’s constant, k is the mean wavelength of transition, (2J + 1) is the degeneracy of the ground state of lanthanide ion and N

800

Luminescence intensity [a.u.]

Pcal ¼

Wavelength [nm] 900

exp

= 367 µs

3

e

2

e

1

e

x=2

0

e

exp

-1

= 141 µs

x=5

e

exp

-2

= 26 µs

e

4

(65-x)%TeO2-20%ZnF2-12%PbO-3%Nb2O5-x%Dy2O3

F9/2

-3

e

0

(as-melted)

100

200

300

400

500

600

700

800

Time [µs] Fig. 4. Luminescence decay curves of the 4F9/2 level recorded at room temperature for (65  x)TeO2–20ZnF2–12Pb2O5–3Nb2O5–xDy2O3 (x = 0.5, 2 and 5) glass samples. See text for explanation of this line.

542

B. Klimesz et al. / Optical Materials 42 (2015) 538–543

Table 5 Comparison of experimental (exp), radiative (rad) lifetime of metastable 4F9/2 Dy3+ level, luminescence quantum efficiency (g), theoretical luminescence branching ratio (bther) for the 4F9/2 ? 6H13/2 transition, critical radius (R0) and dipole–dipole coupling parameter (Cda) assessed for selected glass systems. Host lattice

sexp (ls)

srad (ls)

g (%)

bther

R0 (Å)

Cda (m6s1)

Ref.

B2O3–PbO–TeO2–P2O5–ZnO–BaO–Dy2O3 B2O3–CaF2–CaO–BaO–Al2O3–DyF3 B2O3–PbO–PbF2–Dy2O3 (1 mol%) P2O5–K2O–BaO–BaF2–Al2O3–Dy2O3 (1 mol%) TeO2–ZnF2–Pb2O5–Nb2O5–Dy2O3

468 372 386 660 367

488 873 702 750 453

96 43 55 79 81

0.21 0.62 0.31 0.67 0.61

6.57 4.15 4.40 6.90 7.76

1.26  1052 5.96  1052 1.19  1052 1.14  1052 4.83  1052

[25] [26] [27] [28] This work

Calculated values of the radiative transition rates for the 4F9/2 luminescent state of Dy3+ in 60TeO2–20ZnF2–12Pb2O5–3Nb2O5–5Dy2O3 are presented in Table 4. Examination of survey emission spectrum in Fig. 3 reveals that almost all emitted light intensity is comprised of two bands related to two transitions: 4F9/2 ? 6H13/2 (yellow region of spectrum) and 4F9/2 ? 6H15/2 (blue region of spectrum). There are also significantly weaker emission bands corresponding to the 4F9/2 ? 6H11/2 and 4F9/2 ? 6H9/2, 6F11/2 transitions in the near infrared. From recorded emission spectrum the experimental branching ratios of dysprosium luminescence was determined to be 0.29, 0.68, 0.02 and 0.01 for the 4F9/2 ? 6H15/2, 4F9/2 ? 6H13/2, 4 F9/2 ? 6H11/2 and 4F9/2 ? 6H9/2 transitions, respectively. It can be seen in Table 4 that the agreement between experimental and calculated branching ratios of Dy3+ luminescence is reasonably good. 3.4. Decay curve analysis Experimental 4F9/2 luminescence decay curves recorded with samples containing 0.5, 2 and 5 mol% of Dy2O3 are compared in Fig. 4. The luminescence decay curve for a sample containing 0.5 mol% of Dy2O3 is consistent with a single exponential time dependence with a lifetime value of 367 ls. When Dy3+ concentration increases the decay becomes faster and decay curves deviate from a single exponential time dependence as a consequence of Dy3+ – Dy3+ interaction. In view of this, luminescence decays for samples containing 2 mol% and 5 mol% of Dy2O3 can not be characterized by a definite value of lifetime. Instead, a concept of a ‘‘mean lifetime’’ smean defined as:

smean

R1 t  IðtÞdt R1 ¼ t¼0 IðtÞdt t¼0

ð9Þ

where the I is the intensity of luminescence and t denotes time, is frequently used to assess roughly luminescence decays in such cases. Evaluated smean values of 141 ls and 26 ls for our samples containing 2 mol% and 5 mol% of Dy2O3, respectively indicate substantial importance of self-quenching of the Dy3+ luminescence in systems under study. Examination of luminescence decay curves in question reveals that the decay is initially fast and non exponential, consistent with a Forster-like static decay, and next it tends to approach the single exponential character. Accordingly, we adopt the Inokuti–Hirayama model to determine phenomenological parameters characterizing the Dy3+–Dy3+ interaction. According to Inokuti-Hirayama model [32] the time dependence of the luminescence intensity is expressed by following formula:

IðtÞ ¼ I0 exp

" t

s0

a

 3=s # t

s0

ð10Þ

where I(t) is the luminescence intensity after pulse excitation, s0 denotes intrinsic lifetime of donors in the absence of acceptors, s = 6 for a dipole–dipole interaction, and the a is given by the equation:

4 3



a ¼ pC 1 

 3 N0 R30 s

ð11Þ

where C is the gamma function, N0 is the concentration of acceptors and R0 is the critical radius defined as a distance between donor and acceptor when the rate of energy transfer to the acceptor is equal to the rate of intrinsic decay of the donor. The our calculations revealed that R0 = 7.73 Å for a = 1.53. The dipole–dipole coupling parameter Cda estimated from the relation C da ¼ R60 s1 0 amounts to 4.83  1052 m6s1. It follows from data collected in Table 5 that the system under study shows advantageously high values of quantum efficiency and branching ratio for the 4F9/2 ? 6H13/2 transition. On the other hand the self quenching of Dy3+ emission is rather strong. 4. Conclusions In this paper we present results of investigation of thermal and optical properties of the oxyfluorotellurite glasses (65  x)TeO2– 20ZnF2–12PbO–3Nb2O5–xDy2O3 doped with different (x = 0.5, 2 and 5 mol% of Dy2O3) content of Dy3+ ions. It was found that characteristic parameters, namely glass transition temperatures (Tg), onset of crystallization temperatures (Tc) and thermal stability criteria DT and H’ increase with increasing Dy2O3 content indicating that the incorporation of dysprosium ions improves substantially thermal stability of glass system under study. These findings point at a high resistance of dysprosium-doped oxyfluorotellurite glass to devitrification, thereby facilitating fabrication of optical fibres. Spectroscopic parameters found indicate that this activator-host combination is able to show efficient visible emission distributed mainly in two bands located in blue and yellow region of the spectrum. Efficiency of this emission is adversely affected by Dy3+ concentration, however. Accordingly, good thermal stability combined with desirable spectroscopic parameters point at the suitability of this material for the design of UV-excited visible phosphors provided the Dy3+ concentration is kept low. References [1] Junjie Zhang, Jianbei Qiu, Yoji Kawamoto, Mater. Lett. 55 (2002) 77–82. [2] Guihua Liao, Qiuping Chen, Jianjun Xing, Hrvoje Gebavi, Daniel Milanese, Michael Fokine, Monica Ferraris, J. Non-Cryst. Solids 355 (2009) 447–452. [3] P. Babu, Hyo Jin Seo, C.R. Kesavulu, Kyoung Hyuk Jang, C.K. Jayasankar, J. Lumin. 129 (2009) 444–448. [4] R. Balda, M. Al-Saleh, A. Miguel, J.M. Fdez-Navarro, J. Fernández, Opt. Mater. 34 (2011) 481–486. [5] Gongxun Bai, Jia Ding, Lili Tao, Kefeng Li, Hu Lili, Yuen H. Tsang, J. Non-Cryst. Solids 358 (2012) 3403–3406. [6] P. Babu, Hyo Jin Seo, Kyoung Hyuk Jang, K. Upendra Kumar, C.K. Jayasankar, Chem. Phys. Lett. 445 (2007) 162–166. [7] C. Cascales, R. Balda, J. Fernández, M.A. Arriandiaga, J.M. Fdez-Navarro, Opt. Mater. 31 (2009) 1092–1095. [8] V.B. Sreedhar, D. Ramachari, C.K. Jayasankar, Phys. B 408 (2013) 158–163. [9] Anal Tarafder, Atiar Rahaman Molla, Basudeb Karmakar, J.M. Fdez-Navarro, Opt. Mater. 35 (2013) 1549–1556. [10] L. Jyothi, G. Upender, R. Kuladeep, D. Narayana Rao, Mater. Res. Bull. 50 (2014) 424–431. [11] N. Vijaya, K. Upendra Kumar, C.K. Jayasankar, Spectrochim. Acta A 113 (2013) 145–153. [12] D. Schultze, Differentialthermoanalyse, VEB, Berlin, 1971. [13] R. El-Mallawany, I. Abbas Ahmed, J. Mater. Sci. 43 (2008) 5131–5138. [14] A.E. Ersundu, G. Karaduman, M. Celikbilek, N. Solak, S. Aydin, J. Alloys Compd. 508 (2010). 266-22.

B. Klimesz et al. / Optical Materials 42 (2015) 538–543 [15] A. Dietzel, Glass structure and glass properties, I, Glastechnik 22 (1968) 41–50. [16] M. Saad, M. Poulain, Glass forming ability criterion, Mater. Sci. Forum 19–20 (1987) 11–18. [17] B. Klimesz, G. Dominiak-Dzik, W. Ryba-Romanowski, J. Rare Earths 28 (2010) 893–898. [18] Ch. Görller-Walrand, K. Binnemans, Spectral intensities of f–f transitions, in: K.A. Gschneidner Jr., L. Eyring (Eds.), Handbook on the Physics and Chemistry of Rare Earths, vol. 25, Elsevier Science BV, Amsterdam, Netherlands, 1998, pp. 101–264 (Chapter 167). [19] B.R. Judd, Phys. Rev. 127 (1962) 750–761. [20] G.S. Ofelt, J. Chem. Phys. 37 (1962) 511–520. [21] W.T. Carnall, H. Crosswhite, H.M. Crosswhite, Energy Level Structure and Transition Probabilities of the Trivalent Lanthanides in LaF3, Argonne National Laboratory Report, Argonne, Illinois, 1977. [22] J. Pisarska, J. Phys. Condens. Matter 21 (2009) 285101–285106.

543

[23] J. Hormadaly, R. Reisfeld, J. Non-Cryst. Solids 30 (1979) 337–348. [24] L. Wetenkamp, G.F. West, H. Többen, J. Non-Cryst. Solids 140 (1992) 25–30. [25] S. Selvi, G. Venkataiah, S. Arunkumar, G. Muralidharan, K. Marimuthu, Phys. B 454 (2014) 72–81. [26] J. Suresh Kumar, K. Pavani, A. Mohan Babu, N. Kumar Giri, S.B. Rai, L. Rama Moorthy, J. Lumin. 130 (2010) 1916–1923. [27] I. Arul Rayappan, K. Maheshvaran, S. Surendra Babu, K. Marimuthu, Phys. Status Solidi A 209 (2012) 570–578. [28] S. Surendra Babu, P. Babu, C.K. Jayasankar, Th. Tröster, W. Sievers, G. Wortmann Opt. Mater. 31 (2009) 624–631. [29] C.K. Jørgensen, B.R. Judd, Mol. Phys. 8 (1964) 281–290. [30] R.D. Peacock, Struct. Bond. 22 (1975) 83–122. [31] J.A. Duffy, M.D. Ingram, J. Non-Cryst. Solids 21 (1976) 373–410. [32] M. Inokuti, F. Hirayama, J. Chem. Phys. 43 (1965) 1978–1989.