Investigation of spectroscopic properties, structure and luminescence spectra of Sm3+ doped zinc bismuth silicate glasses

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 101 (2013) 74–81

Contents lists available at SciVerse ScienceDirect

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa

Investigation of spectroscopic properties, structure and luminescence spectra of Sm3+ doped zinc bismuth silicate glasses I. Pal a, A. Agarwal a, S. Sanghi a,⇑, M.P. Aggarwal b a b

Department of Applied Physics, Guru Jambheshwar University of Science and Technology, Hisar 125 001, Haryana, India Department of Physics, Guru Nanak Khalsa College, Yamunanagar 135 001, Haryana, India

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

3+ doped zinc bismuth silicate glasses have been studied. 4 6 " Transitions G5/2? H7/2 and 4 G5/2?6H9/2 are responsible for orange luminescence. 3+ " Bi cations are incorporated in the glass network as [BiO6] octahedral units.

Fluorescence spectra of ZBSS5 glass samples (inset: energy levels depicting the various lasing transitions of Sm3+ ions (4G5/2 ? 6H7/2, 9/2, 11/2)) 4

G5/2→6H9/2

.

Fluorescence Intensity (a.u.)

" Optical properties of Sm

Different lasing emission transitions of Sm3+ in ZBSS glasses 4

G5/2→6H7/2

4

G5/2→6H11/2

λ excitation = 450 nm 590

a r t i c l e

i n f o

Article history: Received 19 April 2012 Received in revised form 7 September 2012 Accepted 20 September 2012 Available online 28 September 2012 Keywords: Glasses Spectroscopy Judd–Ofelt theory Sm3+ doped glasses Optical properties

610

630 650 Wavelength (nm)

x = 50 670

690

a b s t r a c t The glasses with compositions 20ZnO(79.5  x)Bi2O3xSiO20.5Sm2O3 (10 6 x 6 50, mol%) have been synthesized using normal melt-quench technique. Optical absorption and fluorescence spectra of the glasses were recorded at ambient temperature. Judd–Ofelt (J–O) theory has been successfully applied to characterize the absorption and luminescence spectra of these glasses. From the measured intensities of absorption bands of these glasses, the Judd–Ofelt parameters, Xk (k = 2, 4, 6) have been evaluated. The variation of X2 with Bi2O3 content has been attributed to changes in the asymmetry of the ligand field at the rare earth (RE) ion site (due to structural change) and to changes in RE–O covalency, whereas the variation of X6 is found to be strongly dependent on nephlauxetic effect. The shift of the hypersensitive band shows that the covalency of the RE–O decreases with decrease in Bi2O3 content in the host glass. Also, using J–O theory various radiative properties like spontaneous emission probability (Arad), radiative life time (sr), fluorescence branching ratio (br) and stimulated emission cross-section (r) for various emission bands of these glasses in the visible spectral region have been determined. A close correlation is observed between the Bi2O3 content and the spectroscopic, radiative and structural properties of the prepared glasses. The values of radiative properties indicated that 4G5/2 ? 6H7/2 and 4G5/2 ? 6H9/2 transitions responsible for orange luminescence might be used in the development of materials for LED’s and other optical devices in the visible region. Ó 2012 Elsevier B.V. All rights reserved.

⇑ Corresponding author. Tel.: +91 1662 263385; fax: +91 1662 276240. E-mail address: [email protected] (S. Sanghi). 1386-1425/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.saa.2012.09.047

I. Pal et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 101 (2013) 74–81

Introduction Glasses containing rare earth ions have attracted considerable attention because of their wide applications in laser, up-conversion fluorescence, high-density optical storage, solar concentrators, wave-guide lasers and in other photo-electronic fields. From the observed values of absorption and luminescence properties of Sm3+ ions, it has been predicted that the optical properties of these rare earth ions can be affected by varying the host glass composition [1,2]. Trivalent samarium [3–5] and trivalent erbium [6–8] ions doped in glasses exhibit strong luminescence in the orange spectral region at 600 nm and red spectral region at 670 nm respectively. The radiative properties of Er3+ red luminescence have been extensively studied and used in the development of materials for light emitting diodes (LED’s) and are highly useful in the visible spectral region and electroluminescent devices [9]. But only a few attempts have been made to explore the possibility of using radiative properties of orange–red luminescence of Sm3+ ions [5,10,11] in the developments of LED’s in the visible spectral region and other visible optical devices such as visible lasers and fluorescent devices. The main reason for not carrying on much spectral studies of Sm3+ ions doped in glasses is the [12] complicated structure of 4f6 configuration of this ion. A large number of energy levels lying close to each other make the interpretation of the absorption spectrum of this ion rather difficult for the determination of meaningful intensity parameters needed in the calculation of various radiative properties which otherwise requires a suitable and skillful calculation technique [13]. This could be made by properly grouping the closely observed transitions and carefully determining their intensities. Recently, Ratnakaram et al. [14] have reported the luminescence properties of Sm3+ doped lithium cesium mixed alkali borate glasses. Lin et al. [11] studied the radiative properties Sm3+ ions doped in alkali–barium–bismuth–tellurite glasses; Agarwal et al. [5] have also reported the spectroscopic and radiative properties of Sm3+ ions doped zinc bismuth borate glasses to explore the suitability of these glasses for various optical devices. Luminescence intensity in the visible spectral region can be increased by doping the glass with Bi2O3 content. Recently, authors [5,8,15,16] have made spectral studies of Sm3+, Pr3+ and Er3+ ions doped zinc/cadmium bismuth borate glasses in visible and NIR region. It is reported that by varying the amount of Bi2O3 in the host glass there is a considerable increase in the luminescence intensity and radiative properties of Sm3+ and Er3+ ions in the red spectral region. This may be due to intermixing of the electronic levels of rare earth ions with conventional modifier Bi2O3, resulting an increase in the population of the excited states of rare earth ions. In general, optical transitions of rare earth ions greatly depend on the kind of host material. Oxide glasses, particularly bismuth silicate glasses, are one of the most adequate hosts for obtaining efficient luminescence in the rare earth ions. Tellurite and heavy metal oxide glasses have drawn much attention because their maximum phonon energy is lower in comparison to borate, phosphate, silicate and germanate glasses. In rare-earth doped glasses, the emission quantum efficiency depends strongly on the phonon energy of the host medium. Therefore, it is expected that the non-radiative loss to the lattice will be small [17,18]. Since such studies may also be used in the development of materials for LED’s especially in the orange– red spectral region, we have prepared Sm3+ ions doped zinc bismuth silicate (ZBSS) glasses. The luminescence spectra in the visible region and the absorption spectra in the UV–VIS–NIR region have been recorded to investigate the optical and radiative properties. The intensity of the transitions for the rare earth ion can be calculated by using the Judd–Ofelt theory [19,20]. This theory define a set of three intensity parameters, Xk (k = 2, 4, 6) which are sensitive to the environment of the rare earth ion. These intensity parameters

75

are used to calculate radiative transition probability for spontaneous emission, radiative life time of the excited state, branching ratios (which predict the fluorescence intensity of laser transitions) and induced emission cross-section of the various emission lines because in the developing materials for LED’s and other optical devices all these data are essentially required. Theoretical details The experimental oscillator strength (fexpt) of the absorption bands in the VIS–NIR spectral region of Sm3+ ions doped ZBSS glasses can be calculated using the relation:

fexpt: ¼ 4:318 

Z

eðmÞdm

ð1Þ

where e(m) is the molar extinction coefficient at average energy (m) in cm1, to be evaluated from the Beer–Lambert law. Under Gaussian approximation, using Beer–Lambert law, the observed oscillator strength (fexpt) of the absorption bands have been experimentally calculated [10] using the modified relation:

fexpt: ¼ 4:6  109 

  1 I0 log  Dm1=2 cl I

ð2Þ

  where log II0 is the optical density (OD), Dm1/2 is the half band width, l is the path length and c is the molar concentration of the absorbing rare earth ions per unit volume. According to the f–f intensity model of the J–O theory [19,20], the calculated oscillator strengths (fcal) for induced electric dipole transition between the initial |4fn(S, L) J> level and the terminal |4fn(S0 , L0 ) J0 > level of the rare earth ion is given by the equation:

#  " 2 8p2 mc ðn þ 2Þ2  SðJ; J 0 Þ fJJ0 ¼  3hð2J þ 1Þe2 k 9n 

ð3Þ

where m is the mass of electron, c is the velocity of light in vacuum, h is the Planck’s constant, n is the refractive index of glass and S(J, J0 ) is the line strength given by:

SðJ; J 0 Þ ¼ e2

X

Xk h4f n ðS; LÞJkU ðkÞ k4f n ðS0 ; L0 ÞJ0 i2

ð4Þ

k¼2;4;6

In the above equation Xk (k = 2, 4, 6) are the J–O intensity parameters which contain the effect of the odd-symmetry crystal field terms, radial integrals and energy denominators. kU ðkÞ k are the doubly reduced matrix elements of the unit tensor operator of the rank k = 2, 4 and 6 which are calculated from the intermediate coupling approximation. These are almost insensitive to the ion environment in the glass. We have used the values of the matrix elements given by Carnall et al. [21] and Xk parameters have been calculated (from the absorption measurements) using the least square fitting method. The Xk parameters thus obtained are used to calculate a number of radiative properties viz, spontaneous emission probability (Arad), radiative life time (sr), fluorescence branching ratio (br) and stimulated emission cross-section (r). The spontaneous emission probability from an initial manifold |4fn(S, L) J> to a final manifold |4fn(S0 , L0 ) J0 > is given by:

" # nðn2 þ 2Þ2  SðJ; J 0 Þ 3 9 3hc ð2J þ 1Þ 64p4 m3 e2

Arad ½ðS; LÞJ; ðS0 ; L0 ÞJ 0  ¼

ð5Þ

For Sm3+ ion J = 5/2. The fluorescence branching ratio for the transition from an initial manifold |4fn(S, L) J> to a final manifold |4fn(S0 , L0 ) J0 > is given by:

A½ðS; LÞJ; ðS0 ; L0 ÞJ 0  br ½ðS; LÞJ; ðS0 ; L0 ÞJ 0  ¼ X A½ðS; LÞJ; ðS0 ; L0 ÞJ 0  0

0

ðS ;L ÞJ

0

ð6Þ

76

I. Pal et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 101 (2013) 74–81

The total radiative transition probability (AT) of an excited state is given by the sum of the all possible terminal levels and is related to radiative life time (sr) as:

sr ¼ X

1 A½ðS; LÞJ; ðS0 ; L0 ÞJ 0 

¼

1 AT

ð7Þ

Fluorescence measurements The emission spectra (in visible region) of polished samples were recorded on Fluro-Max-3 Fluorimeter with a Xe-arc lamp (450 W) as the excitation source at a wavelength of 450 nm.

ðS0 ;L0 ÞJ0

The peak stimulated emission cross-section, r(kp), is an essential parameter to predict the laser performance of the glass and can be determined using the following expression:

"

#  k4p  A ðS; LÞJ; ðS0 ; L0 ÞJ 0 rðkp Þ ¼ 8pcn2 Dkeff

ð8Þ

where kp is the wavelength of the fluorescence peak and Dkeff is the line width obtained by dividing the area of the emission band by its average height. Therefore, the large value of stimulated emission cross-section, high quantum efficiency (low radiative life time), high branching ratio and high figure of merit (r  sr) determine the suitability of the good materials for optical devices.

IR transmission measurements IR transmission spectra of the glasses were recorded at room temperature using KBr pellet technique on a Shimadzu FTIR 8001PC spectrometer in the range 400–4000 cm1. The samples investigated were fine particles mixed with pulverized KBr in the ratio of 1:20 (glass powder to KBr). The weighed mixture was then subjected to a pressure of 7–8 tons to make clear homogenous disks. The infrared transmission spectra were recorded immediately after preparing the desired disks. Results and discussion Absorption spectra and Judd–Ofelt intensity parameters

Experimental details Glass synthesis The glasses having composition 20ZnO(79.5  x)Bi2O3xSiO2 0.5Sm2O3, (10 6 x 6 50, x in mol%) were prepared using normal melt-quenching technique. Analytical reagents of ZnO, Bi2O3, SiO2, and Sm2O3 were used to synthesize the samples. 15g batches were taken in a crucible and were put in an electric furnace at a temperature of 1150 °C for 40 min. The melt was stirred frequently for homogeneous mixing of all the constituents. The glass samples were obtained by pouring and quenching the melt in-between two stainless steel plates kept at room temperature (RT). Clear, bubble free and transparent glasses were obtained. Disk shaped samples were polished with different grades of emery powder. Density measurement The density (D) of each glass sample was measured by Archimedes’ principle using Xylene as immersing liquid (Dxylene = 0.8645 g cm3). All measurements were repeated two to three times per sample. The accuracy of the results in density measurement is ±0.001 g cm3. The molar volume was calculated from the relation VM = M/D where M is the molar mass of the glass. Refractive index measurement Polished glass samples were used for the measurement of refractive index (n). Brewster angle technique was employed to measure n using He–Ne laser (632 nm) as a source. X-ray diffraction technique The amorphous nature of the prepared samples was confirmed by recording X-ray diffraction patterns using Miniflex-II (Rigaku) X-ray diffractometer. UV–VIS–NIR absorption measurements The optical absorption spectra of all the polished glass samples (thickness 1.10–1.78 mm) were recorded at room temperature in the wavelength range 300–3200 nm using a Varian–Carry (5000) spectrophotometer. The absorption coefficient as a function of wavelength, a(k), was calculated by dividing the measured absorbance by sample thickness.

The values of density (D), molar volume (VM), number density of samarium ions (N), mean rare-earth ion separation (ri), ionic radius (rp), refractive index (n), dielectric constant (e), and reflection loss (RL) of the prepared glasses are calculated from the conventional formulae using experimental data and are presented in Table 1. From Table 1 it is observed that D and VM decrease with decrease in Bi2O3 content in the glass. Other physical parameters (n, e, RL, ri, rp) follow the same trend except the number density of samarium ions (N). Density and molar volume of the prepared glasses decrease with decrease in Bi2O3 content in the host because the higher mass of Bi2O3 (465.98 gm) is replaced by lighter mass of SiO2 (60.08 gm). It is also observed from IR spectra (Section 4.3) that the successive replacement of Bi2O3 by SiO2 leads to the formation of large number of SiO4 tetrahedral units and the number of non-bridging oxygen’s also increases. Other physical parameters, viz., mean rare earth separation, ionic radius, refractive index, dielectric constant and reflection loss decrease with decrease in Bi2O3 in the host glass while number density of samarium ions increases because ‘N’ is directly related to density. The oscillator strength of Sm3+ ions are arranged in two groups [3]; one being the ‘‘low energy’’ group corresponding to transitions up to 10,700 cm1 and the other is the ‘‘high energy’’ group corresponding to transitions up to energy range 17,600–32,000 cm1. Therefore, the Judd–Ofelt equation (Eq. (3)) is applicable to cases where the high f-splitting is small as compared to the f–d energy gap since it is not appropriate to use the high-energy levels for the calculation of Xk. Thus, in the present study, the Xk parameters are determined only for the low energy region. Also, in the high energy region, due to strong ultraviolet absorption by Bi3+ ions, Sm3+ peaks are not well resolved. All the transitions of Sm3+ are electric dipolar with significant intensities observed in the range 1020– 1620 nm (Fig. 1). The absorption peaks corresponding to the transitions from the ground state 6H5/2 to the various excited states are marked in the spectrum. Due to the strong absorption by the host glasses in the ultraviolet range, the absorption bands below 420 nm could not be distinguished. The absorption spectra for all other glasses are qualitatively similar except for the slight variation in the shape and the absorption peaks. The experimental oscillator strengths (fexpt) for absorption bands: 6H5/2 ? 6F9/2, 6F7/2, 6F5/2, 6 F3/2, 6F1/2 and 6H15/2 are determined with known values of the Sm3+ ion concentration, sample thickness, peak position and peak areas by using Eq. (2) for all the sample under study. The values of fexpt and fcal are listed in Table 2. The J–O intensity parameters Xk (k = 2, 4, 6) have been determined by using the experimentally

77

I. Pal et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 101 (2013) 74–81

Table 1 Density (D), molar volume (VM), number density of Sm3+ ions (N), inter-ionic distance (ri), ionic radius (rp), refractive index (n), dielectric constant (e) and reflection loss (RL) of Sm3+ ions doped ZBSS glasses. Sample code

x (mol%)

D (g/cm3)

VM (cm3/mol)

N (1020 ions/cm3)

ZBSS1 ZBSS2 ZBSS3 ZBSS4 ZBSS5

10 20 30 40 50

6.89 6.64 6.54 6.46 6.28

48.45 44.27 38.98 33.81 29.09

1.86 2.04 2.31 2.67 3.10

6

F7/2

Transitions from ground state 6H5/2

6

F5/2

6

F9/2

Optical density (a.u.)

6

F3/2

6

6

x = 10 1020

1120

F1/2

1220

1320

1420

H15/2

1520

1620

Wavelength (nm) Fig. 1. Optical absorption spectra of ZBSS1 glass sample.

measured oscillator strengths and least square fitting analysis method. In order to determine the accuracy of the intensity parameters, the root-mean-square deviation (dRMS) is also calculated using the relation:

dRMS ¼

"P #1=2 ðfcal  fexpt Þ2 P 2 fexpt

ð9Þ

where summation is taken over the bands used to calculate Xk parameters. The values of dRMS for all the glasses samples are also given in Table 2. The small values of dRMS indicate a good fitting

0

0

ri (Å A)

rp (Å A)

8.12 7.86 7.55 7.21 6.85

3.27 3.17 3.05 2.90 2.76

n

e (n2)

RL (%) (n  1/n + 1)2

2.22 2.18 2.16 2.13 2.08

4.93 4.75 4.66 4.53 4.32

14.35 13.76 13.47 13.03 12.29

between experimental and calculated oscillator strengths. In the present glasses Sm3+ has a 4f5 electron configuration which is characterized by 198 2S+1LJ free ion levels [22]. In the presence of lowsymmetry crystal field, these free-ion manifolds split into a total of 1001 crystal field’s levels (Kramers doublet). In case of glasses the crystal field fine structure is not resolved due to inhomogeneous line broadening and only absorption bands between 2S+1LJ manifolds are observed. Only a limited number of 198 2S+1LJ manifolds can be observed experimentally. Moreover many of the manifolds are lying so close to each other that it is not possible to determine their positions correctly. Therefore, in the absorption spectrum of Sm3+ ions in zinc bismuth silicate glasses only six absorption bands originating from ground state 6H5/2 could be observed in NIR (energies < 10,000 cm1) regions belonging to a total of 33 2S+1LJ levels. These levels have been assigned by comparing their band positions with the energy level scheme of LaF3: Sm3+ and Sm3+ (aquo) published by Carnall et al. [21,23]. All the transitions in the spectrum originate from induced electric dipole (ED) interactions with the selection rule DJ 6 6. The high intensity of these transitions is expected because the transitions are spin allowed (DS = 0). The intensity of these transitions is a function of host glass composition. As a consequences of this absorption, samarium doped zinc bismuth silicate glasses look pale yellow in color. The Judd–Ofelt intensity parameters, Xk are important for investigating structure and bonding of doped rare earth ions with the surrounding glass matrix. All the three J–O parameters (O2, O4, O6) found with the help of Eq. (4) have been presented in Table 3 and follow the order O2 > O4 > O6. Similar order in the J–O parameters has been observed by Boehm et al. [3] in oxide and in LKBB tellurite [11] glasses. Reported values of Ok parameters for borate [3], phosphate [3], silicate [10], LKBB tellurite [11], LCN borate [14], ZBLAN [24], fluoride [25], and ZFBP [26] glasses have been given in Table 3 for comparison. In the present Sm3+ doped zinc bismuth silicate (ZBSS) glasses it is observed that O2 decreases whereas O4 and O6 show increasing behavior with Bi2O3 content in the host glass. The variation in Judd–Ofelt intensity parameters with Bi2O3 content is shown in Fig. 2. It has been reported that the X2 parameter is related to the asymmetry of the ligand field near the rare earth ions, i.e., higher the asymmetry, the larger the X2 [27]. Therefore, it is estimated from the variation

Table 2 Oscillator strength for some transitions from the indicated levels to the ground level 6H5/2, and their root mean square deviation (dRMS), which indicates the fit quality of theoretical and experimental results for Sm3+ ions doped ZBSS glasses. Oscillator strength (108)

Transitions from ground level

ZBSS1

ZBSS2

ZBSS3

ZBSS4

ZBSS5

6

k (nm)

fexpt

fcal

fexpt

fcal

fexpt

fcal

fexpt

fcal

fexpt

fcal

6

1584 1530 1480 1376 1230 1078

113 95 157 189 277 198 3.65

132 77 170 204 238 221

98 66 172 164 258 178 3.15

105 76 163 188 218 201

89 42 93 148 237 159 3.68

91 33 99 167 196 183

76 40 92 143 207 128 3.80

88 32 94 159 172 148

58 46 77 147 127 131 3.67

65 32 87 134 143 122

H5/2?

H15/2 6 F1/2 6 F3/2 6 F5/2 6 F7/2 6 F9/2 dRMS (107)

78

I. Pal et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 101 (2013) 74–81 Table 3 Judd–Ofelt intensity parameters (X2, X4, X6) of Sm3+ ions doped ZBSS glasses. Glass

X2 (1020 cm2)

X4 (1020 cm2)

X6 (1020 cm2)

X4/X6

References

ZBSS1 ZBSS2 ZBSS3 ZBSS4 ZBSS5 Borate Phosphate ZBS Silicate LKBB tellurite LCN borate ZBP ZBLAN Fluoride ZFBP

3.17 2.75 2.46 1.96 1.56 6.36 4.31 1.93 1.94 4.73 0.84 1.30 2.06 1.44 2.68

1.32 1.61 1.67 1.92 2.19 6.02 4.28 1.87 2.63 2.78 4.00 3.29 2.55 2.87 3.12

1.25 1.38 1.59 1.72 1.77 3.51 5.78 1.79 1.63 1.77 5.02 2.13 1.63 1.44 1.70

1.05 1.16 1.05 1.11 1.23 1.71 0.74 1.04 1.62 1.57 0.80 1.54 1.56 1.99 1.84

Present Present Present Present Present [3] [3] [5] [10] [11] [14] [15] [24] [25] [26]

transitions of Sm3+ ions (6H5/2 ? 6F1/2; 1530 nm to 1521 nm) with decrease in Bi2O3 content in the host glass indicates the decreasing covalency of RE–O bond in the glass matrix. Thus X2 and ‘nephlauxetic effect’ are good parameters to evaluate the degree of covalency and both of them also interfere in the shifting of hypersensitive band position. X2 is related with the symmetry of the glass while X6 is inversely proportional to the covalency of the Sm–O bond. The covalency of the Sm–O bond is assumed to be related with the local basicity around the rare-earth sites, which can be adjusted by the composition or structure of the glass host. With the substitution of SiO2 for Bi2O3, the number of the tetrahedral units [SiO4] and non-bridging oxygen’s are increasing in the network structure of the host glass. Therefore, more and more non-bridge oxygen ions, which tend to coordinate with Sm3+, will contribute to coordinate with glass former cations. On the other hand, according to theory of electronegativity, smaller the difference of electronegativity between cation and anions,

Ω2

3

Intensity parameters (10-20 cm2)

Ω4 Ω6

2.5

work work work work work

2

1.5

1 25

35

45

55

65

75

Bi2O3 (mol%) Fig. 2. Variation of intensity parameters (X2, X4, and X6) with Bi2O3 concentration (lines are guides for the eye).

of X2 in Fig. 2 that the polarity and asymmetry of the rare earth sites and the bond strength between Sm3+ and O2 decreases with decrease in Bi2O3 content in the host glass. The spectroscopic quality factor (SQF = O4/O6) is found to be greater than one in the present study. It is higher than phosphate [3], ZBS [5] and LCN borate [14] glasses and lower than other glasses presented in Table 3. Since O4 and O6 are structure dependent parameters, therefore SQF > 1 indicates that present glasses are more stable and rigid as compared to phosphate, ZBS and LCN borate glasses. Ratnakaram et al. [28] have also reported similar conclusion in mixed alkali borate glasses. The position and intensity of certain electric dipole transition of rare earth ions are found to be very sensitive to the environment of the rare earth ion. Such transitions are known as hypersensitive transitions [29]. It has been proposed that the positions of hypersensitive bands change due to ‘‘nephlauxetic effect’’. The ‘‘nephlauxetic effect’’ [30] arise from the partially filled f-shell. Thus, this effect can measured the covalency of RE–O bond in the host glass. With increase in the overlap of the oxygen orbitals and 4f-orbitals, the energy level structure of the rare earth ion contracts, and leads to shift in the hypersensitive bands transitions. The blue shift in the peak wavelength of the hypersensitive

Fig. 3. Fluorescence spectra of ZBSS5 glass samples (Inset: Energy levels depicting the various lasing transitions of Sm3+ ions (4G5/2 ? 6H7/2, 9/2, 11/2)).

79

15.99

7.63 6.74 1.62 0.413 0.481 0.105

17.42 19.17

13.29 12.94 33.78

3.05 2.03 11.12

sr (ms)

21.10

*

75 77 29 Fluorophosphate glass [13] H7/2 596 6 H9/2 639 6 H11/2 703 6

327 491 89

Arad (s1) kp (nm)

599 645 713 H7/2 6 H9/2 6 H11/2*

6

Not observed experimentally. Predicted values are reported.

3.70 5.80 3.02 0.391 0.401 0.154

5.66 11.41 – 0.304 0.457 0.084

r (1022cm2) br (%)

24.06 LKBB tellurite glass [11]

r (1022cm2) br (%) Arad (s1)

67 78 17 162 6.17 8.51 7.04 1.87 0.426 0.462 0.111 73 79 19 171 5.84 9.15 8.06 1.96 0.409 0.478 0.112 77 90 21 188 5.32 9.95 8.90 2.25 0.410 0.480 0.110 82 96 22 200 5.00 11.27 9.98 2.81 0.409 0.474 0.114 88 102 25 215 4.38 597 633 679 H7/2 H9/2 H11/2 AT (s1) sr (ms) rt (1022cm2) 6

6

6

ZBSS5

r (1022cm2) br (%) Arad (s1)

ZBSS4

r (1022cm2) br (%) Arad (s1)

ZBSS3

r (1022cm2) br (%) Arad (s1)

ZBSS2

r (1022cm2) br (%) ZBSS1

Arad (s1)

Fig. 3 shows the fluorescence spectrum of (ZBSS5) glass recorded at RT at excitation wavelength 450 nm. This figure shows three broad peaks near 597, 633 and 679 nm and the peaks on either side of most intense peak (at 633 nm) have approximately same intensities. All the fluorescence peaks arise due to 4 G5/2 ? 6H7/2,9/2 and 11/2 transitions, respectively. The most intense peak at 633 nm gives the orange color of luminescence in the present glasses. This orange luminescence can be excited by using powerful pumping [11] sources like commercial UV and blue laser diodes, blue and bluish-green LED’s and Ar+ optical laser. All the peaks shown in Fig. 3 seem to have two components. These components are most likely due to the Stark splitting of the ionic levels. The Stark splitting reduces with increase in Bi2O3 content in the host glass. Judd–Ofelt parameters (Table 3) and fluorescence data (Table 4) have been used to calculate radiative parameters: spontaneous emission probability (Arad), radiative life time (sr), fluorescence branching ratio (br) and stimulated emission cross-section (r) using Eqs. (5)–(8). These values have been presented in Table 4 for all the observed fluorescence transitions. Table 4 also includes the radiative properties of the electric dipole transitions responsible for orange luminescence in Sm3+ ion doped tellurite [11] and fluorophosphate [13] glasses. Oxide glasses are chemically stable and cheaper than fluoride or fluorophosphate glasses, yet they are not good choice as materials for LED’s because of high nonradiative transition probabilities in the pure oxide glasses. Tellurite glasses do not have this problem because of low vibrational phonon frequencies. These glasses have high stimulated emission cross-section [31] but these are still not good choice for LED’s because these are not free from scattering and dispersion losses. Materials for this purpose should have minimum dispersion losses and fluoride and fluorophosphate glasses fullfil these required condition for LED’s. Sm3+ ions doped bismuth silicate co-doped with ZnO have nearly same (Table 4) radiative properties as that of Sm3+ doped LKBB tellurite and fluorophosphate glasses. In the present glass two transitions viz., 4G5/2 ? 6H7/2 and 4G5/2 ? 6H9/2 (Fig. 4) responsible for orange luminescence have branching ratio 0.5 and also have energy separation 6000 cm1 or more between the energy level under consideration and terminating level, so the rate of multi-phonon emission is negligible. The branching ratios are evaluated for each transition and a probable lasing transition of the rare earth ion is shown as an insert of Fig. 4. In addition to this, the present glass is easy to prepare with low dispersion [32] in the visible region, non-hygroscopic in nature and low in cost. With a little compositional change in it, core and cladding glasses with matching thermo dynamical properties [33] can be prepared and fibers of required dimensions can easily be drawn with high precision [34]. Therefore, Sm3+ ions doped zinc bismuth silicate glasses may be used as materials for LED’s and other optical devices in the visible region.

kp (nm)

Fluorescence spectra

Transitions from 4G5/2?

stronger will be the covalency of the bond. The values of electronegativity for Bi, Si, and O elements are about 1.8, 1.9, and 3.5, respectively. As a result, the covalency of the Si–O bond is stronger than that of the Bi–O bond. It is expected that the influence of the Si–O bond on the local ligand environments around Sm3+ increases with an increase of SiO2 content. Consequently, the covalency of the Sm–O bond decreases, and the values of X6 and X4 increase accordingly. Here, the close packing of structure takes place with the formation of SiO4 and BiO6 units and results in increased interaction between the rare earth and charged non-bridging oxygens (NBO’s). Thus the covalency of the RE–O bond is decreased as is also evident by the decrease in the intensity of the 6H5/2 ? 6F1/2 transition.

Table 4 The peak wavelength (kp), radiative transition probability (Arad), branching ratio (br), stimulated emission cross-section (r), total radiative transition probability (AT), radiative life time (sr) and the total emission cross-section (rt) of Sm3+ ion doped ZBSS glasses.

I. Pal et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 101 (2013) 74–81

80

I. Pal et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 101 (2013) 74–81

Fig. 4. Infrared transmission spectra of 20ZnO (79.5  x)Bi2O3xSiO20.5Sm2O3 glasses.

Infrared transmission spectra The effect of substitution of unconventional glass former (Bi2O3) by the conventional glass former (SiO2) on the structural properties of zinc bismuth silicate glasses was investigated by recording their Fourier transform infrared (FTIR) spectrum in the range 400– 4000 cm1. With high Bi2O3 content, it is expected to reflect the effect of high mass cations (Bi3+) on the infrared absorption spectra. In the present work the FTIR spectrum is given in the range 400–1400 cm1 for better clarity (Fig. 4). The positions of absorption bands in the mid near infrared spectra of the present glasses are slightly different from the spectra usually obtained for the traditional silicate glasses and crystal [35] due to abundance of the heavy metal bismuth oxide (Bi2O3) and modifier cation (ZnO). Infrared spectroscopy is one of the most useful experimental techniques available for easy structural studies of glasses [36]. As this technique leads to structural aspects related to both local units constituting the glass network and the anionic sites hosting the modifying cations, infrared is a powerful tool for the structural studies of glasses modified by metal oxides. The mid-region extending from 400 to 2000 cm1 is characterized by the appearance of the characteristic absorption bands of the network forming groups [35]. It is accepted that the main vibrational modes appeared above 400 cm1 in mid-infrared range are associated with structural change in the glass network [37,38]. These network modes are well separated from the metal ion site vibrational modes active in the far infrared region [38,39]. In the IR spectra of the present glasses a broad but strong band at 496–508 cm1 was observed and it shifts towards lower wave number with decrease in Bi2O3/SiO2 ratio. It has been predicted by various authors [40,41] that this band originates from Bi–O bonds in BiO6 octahedra and the shifting to lower wave number (496 cm1) is due to decrease of the degree of the distortion [40]. Dimitriev et al. [42] have also attributed the shift in band around 482–520 cm1 to the vibration in the local symmetry of highly distorted BiO6 polyhedra and the same was also observed in IR spectra of other bismuth based glasses [43]. Some authors observed that the band in the wave number range 440–470 cm1 is due to the symmetric oxygen bending-rock made (R) BO’s bonding [44–46]. The band near 496–508 cm1 may compose of two bands but most of its intensity comes from bending vibrations of the silicate network as there is practically no change in the integrated intensity ratio between the band near 1000 cm1 (stretching mode of

silicate network) and the band near 496–508 cm1 for the studied glasses in spite of decrease in the concentration of Bi2O3. Therefore, shift of this band is due to depolymerization of the network and change in the Si–O–Si angle. The second prominent infrared band is observed nearly around 750–1150 cm1 centered at about 900 cm1 and is accompanied by a broad shoulder centered at 1200 cm1. The center of this shoulder shifts to higher wave number with decrease in Bi2O3 content (Fig. 4). The spectrum of vitreous SiO2 also shows, in addition to the band at 450 cm1, a weak band at about 800 cm1 and a strong one at about 1080 cm1; the later being accompanied by a broad shoulder centered at 1200 cm1. These bands are attributed to different vibrational modes of Si–O–Si links [47]. It is also reported that [48,49] in the presence of modifier oxide the bands associated with glass network shifts towards lower wave number and are broadened. This is due to build up of SiO4 tetrahedral units bearing a progressively higher number of non-bridging oxygens [48,49]. Thus, the blue shift of this band (750–1150 cm1) with the successive replacements of Bi2O3 by SiO2 suggests an increase in Si–O–Si bond angle. Further, the broadening of bands may also be attributed to the local mode formation caused by intrinsic point defects induced by intermolecular interaction (SiO2/Bi2O3) in the glass network [50]. A sharp peak at 670 cm1 was observed in the present glasses. Betch et al. [51] reported detailed data on IR and Raman spectra of a-Bi2O3 and bismuthate phases Bi12SiO20 having the silenite structure. With these investigations, band at 670, 620, 580 and 390 cm1 in the spectrum of a-Bi2O3 were interpreted as vibrations of Bi–O bonds of different lengths in the distorted BiO6 polyhedra.

Conclusions The optical properties of Sm3+ ions doped zinc bismuth silicate glasses have been studied. Effect of bismuth oxide on the absorption and emission spectra was investigated with the help of Judd– Ofelt theory. The variation of intensity parameters are discussed and correlated to the structural changes in the glass network. In the present glasses it has been observed that order of intensity parameters is O2 > O4 > O6. The shift of the hypersensitive bands of Sm3+ ions (6H5/2 ? 6F1/2; 1530 nm) shows that the covalency of the RE–O bond increase with increase of Bi2O3 content, due to the increased interaction between rare earth ions and non-bridging oxygens. The radiative properties viz., spontaneous emission probability, radiative life time, branching ratio and stimulated emission cross-section for the prepared glasses have been determined. From emission spectra it is observed that two transitions 4G5/2 ? 6H7/2 and 4G5/2 ? 6H9/2 responsible for orange luminescence have branching ratio 0.5. Radiative properties indicated the suitability of the present glasses for their use in the development of LED’s and other optical devices in the visible region. IR spectra revealed that Bi3+ cations are incorporated in the glass network as [BiO6] octahedral units. The shifting of band around 496–508 cm1 to lower wave number and of the band at 750–1150 cm1 towards higher wave number suggests the increase in Si–O–Si bond angle. The presence of distorted [BiO6] octahedral units in the glass network was observed and the degree of distortion was found to decrease on progressive substitution of Bi2O3 by SiO2 ions. References [1] [2] [3] [4] [5] [6]

R. Reisfeld, A. Bornstein, L. Boehm, J. Solid State Chem. 14 (1975) 14–19. M. Canalejo, R. Cases, R. Alcala, Phys. Chem. Glasses 29 (1988) 187–191. L. Boehm, R. Reisfeld, N. Sepctor, J. Solid State Chem. 28 (1979) 75–78. Q. Zeng, N. Kilah, M. Riley, K.H. Risen, J. Lumin. 104 (2003) 65–76. A. Agarwal, I. Pal, S. Sanghi, M.P. Aggarwal, Opt. Mater. 32 (2009) 339–344. B.T. Stone, K.L. Bray, J. Non-Cryst. Solids 197 (1996) 136–144.

I. Pal et al. / Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 101 (2013) 74–81 [7] 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. [8] I. Pal, A. Agarwal, S. Sanghi, M.P. Aggarwal, Mater. Chem. Phys. 133 (2012) 151–156. [9] A.R. Zanata, L.A.O. Nunes, Appl. Phys. Lett. 72 (1998) 3127–3129. [10] K. Annapurna, R.N. Dwivedi, A. Kumar, A.K. Chaudhari, S. Buddhudu, Spectrochim. Acta Part A 56 (1999) 103–109. [11] H. Lin, X.Y. Wang, L. Lin, D.L. Yang, T.K. Xu, J.Y. Yu, E.Y.B. Pun, J. Lumin. 116 (2006) 139–144. [12] R. Reisfeld, C.K. Jorgensen, Lasers and Excited States of Rare Earths, Springer, Berlin, 1977. [13] R.V. Deun, K. Binnemans, C. Gorller-Walrand, J.L. Adam, J. Alloys Compd. 283 (1999) 59–65. [14] Y.C. Ratnakaram, N.D. Thirpathi, R.P.S. Chakaradhar, J. Non-Cryst. Solids 352 (2006) 3914–3922. [15] I. Pal, A. Agarwal, S. Sanghi, M.P. Aggarwal, J. Alloys Compd. 509 (2011) 7625– 7631. [16] S. Sanghi, I. Pal, A. Agarwal, M.P. Aggarwal, Spectrochim. Acta Part A 83 (2011) 94–99. [17] Z. Pan, S.H. Morgan, K. Dyer, A. Ueda, H. Liu, J. Appl. Phys. 79 (1996) 8906– 8913. [18] S. Tanabe, X. Feng, T. Hanada, Opt. Lett. 25 (2000) 817–819. [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, P.R. Fields, K. Rajnak, J. Chem. Phys. 49 (1968) 4424–4442. [22] W.G. Wybourne, Spectroscopic Properties of Rare Earths, Interscience Publications John Wiley and Sons Inc., New York, 1965. [23] W.T. Carnall, H. Crosswhite, H.M. Crosswhite, Energy level Structure and Transition Probabilities of the Trivalent Lanthanides in LaF3, Argonne National Lab. Report, Argonne, Illnois, 1977. [24] R.V. Deun, K. Binnemans, C. Gorller-Walrand, J.L. Adam, SPIE 3622 (1999) 175– 181. [25] R. Cases, M.A. Chamarro, J. Solid State Chem. 90 (1991) 313–319. [26] Y.K. Sharma, S.S.L. Surana, R.P. Dubedi, V. Joshi, Mater. Sci. Eng. B 119 (2005) 131–135. [27] S. Tanabe, T. Ohyagi, N. Soga, T. Hanada, Phys. Rev. B 46 (1992) 3305–3310.

81

[28] Y.C. Ratnakaram, A.V. Kumar, N.D. Thirpathi, R.P.S. Chakaradhar, K.P. Ramesh, J. Lumin. 110 (2004) 65–77. [29] C.K. Jorgensen, B.R. Judd, Mol. Phys. 8 (1964) 281–290. [30] C.K. Jorgensen, Modern Aspects of Ligand Field Theory, North-Holland, Amsterdam, 1971. [31] R.A.H. El-Mallawany, Tellurite Glasses Handbook-Physical Properties and Data, Boca Raton, FL, 2001. [32] C.F. Rapp, Optical Materials in Handbook of Laser Science & Technology, CRC Press, Boca Raton, 1987. [33] E. Snoeks, G.N.V. Hoven, A. Polman, J. Appl. Phys. 73 (1993) 8179–8183. [34] W.J. Miniscalco, Optical and electronic properties of rare earth ions in glasses, in: M.J.F. Digonnet (Ed.), Rare Earth doped Fiber Lasers and Amplifiers, Marcel Dekker, New York, 1993. [35] A. Witkowska, J. Rybicki, A.D. Cicco, J. Alloys Compd. 401 (2005) 135–144. [36] J. Wong, C.A. Angell, Glass Structure by Spectroscopy, Marcel Dekker Inc., New York, 1967. p. 409. [37] K.M. ElBradry, F.A. Moustaffa, M.A. Azooz, F.H. ElBatal, Indian J. Pure Appl. Phys. 38 (2000) 41–45. [38] J. Krogh-Moe, Phys. Chem. Glasses 6 (1965) 46–54. [39] C.I. Merzbacher, W.B. White, J. Non-Cryst. Solids 130 (1991) 18–34. [40] L. Baia, R. Stefan, W. Kiefer, J. Popp, S. Simon, J. Non-Cryst. Solids 303 (2002) 379–386. [41] F.H. ElBatal, Nucl. Inst. Methods Phys. Res. B 254 (2007) 243–253. [42] Y. Dimitriev, V. Mihallova, in: A. Dufan, F. Navarro (Eds.), Proc. Int. Cong. On Glass, vol. 3, Madrid, 1992, p. 293. [43] S. Bale, M. Purnima, Ch. Srinivasu, S. Rahman, J. Alloys Compd. 457 (2008) 545– 548. [44] J. Bell, P. Dean, Discuss. Faraday Soc. 50 (1970) 55–61. [45] P.N. Sen, M.F. Thorpe, Phys. Rev. B. 15 (1977) 4030–4038. [46] F.L. Galeener, Phys. Rev. B 19 (1979) 4292–4297. [47] J. Wong, in: Borate Glasses: Structure, Applications, Plenum Press, New York, 1977, p. 297. [48] I. Simon, H.O. McMahon, J. Am. Ceram. Soc. 36 (1953) 160–166. [49] Y. Kim, A.E. Clark, L.L. Hench, J. Non-Cryst. Solids 113 (1989) 195–202. [50] A.M. Efimov, Optical Constants of Inorganic Glasses, CRC Press, New York, 1995. [51] R. Betsch, W. White, Spectrochim. Acta 34A (1977) 505–514.