Available online at www.sciencedirect.com
Journal of Non-Crystalline Solids 354 (2008) 1981–1985 www.elsevier.com/locate/jnoncrysol
Spectroscopic properties of Er3+-doped Na2O–Sb2O3–B2O3–SiO2 glasses Q. Qian, Y. Wang, Q.Y. Zhang *, G.F. Yang, Z.M. Yang, Z.H. Jiang Key Lab of Specially Functional Materials of Ministry of Education, Institute of Optical communication, South China University of Technology, Guangzhou 510641, PR China Received 22 June 2007; received in revised form 3 November 2007 Available online 4 January 2008
Abstract Spectroscopic properties of Er3+-doped Na2O–Sb2O3–B2O3–SiO2 glasses have been investigated for developing 1.5-lm broadband fiber amplifiers. An intense 1.5-lm near infrared emission with a broad full width at half maximum (FWHM) of 88 nm has been obtained for Er3+-doped 5Na2O–20Sb2O3–35B2O3–40SiO2 glass upon excitation with a 980 nm laser diode. The obtained emission cross-section of the 4I13/2 ? 4I15/2 transition and the lifetime of the 4I13/2 level of Er3+ ions are 6.8 1021 cm2 and 0.36 ms, respectively. It is noted that the product of the emission cross-section and the FWHM of the glass, re FWHM, is as great as 598.4 1021 cm2 nm, which is comparable or higher than that of Er3+-doped bismuth-based and tellurite-based glasses. These special optical properties encourage in identifying them as important materials for potential applications in high performance optics and optical communication networks. Ó 2007 Elsevier B.V. All rights reserved. PACS: 42.70.a; 78.20.e; 78.55.Qr Keywords: Optical fibers; Glasses; Luminescence; Borosilicates; Rare-earths in glasses
1. Introduction Over the past several years, wavelength division multiplexing (WDM) system has been adopted to meet rapidly increasing data traffics. An optical amplifier is required to replenish optical signals. Er3+-doped fiber amplifiers (EDFA) comprise an important part in WDM systems because of the practical use for C-band or L-band [1,2]. However, the bandwidth of the conventional SiO2-based EDFA is limited, and an amplifier’s structure with a parallel configuration is complicated [1,3]. It is important to find other glass matrix for Er3+ doping to produce an intrinsically broader gain bandwidth at 1.5-lm than that of SiO2-based EDFA. Many researchers have paid much attention to heavy metal oxide glasses [4,5], phosphate glasses [6,7] and germinate glasses [8–10]. Especially, it *
Corresponding author. Tel./fax: +86 20 87114834. E-mail address:
[email protected] (Q.Y. Zhang).
0022-3093/$ - see front matter Ó 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jnoncrysol.2007.11.005
has been reported that Er3+-doped bismuth-based glass showed large bandwidth, high emission and absorption cross-section [11,12]. Antimony and bismuth are located in the same VA group and usually show the similarity in chemical and physical properties. It is reasonable to infer Er3+-doped antimony-based glass, as Er3+-doped bismuth-based glass, exhibited large bandwidth, high emission and absorption cross-sections, as well as high solubility of erbium ions. In this paper, we report on spectroscopic properties of Er3+-doped antimony-borosilicate glasses. Based on the Judd–Ofelt (JO) theory, the bandwidth of emission spectra, the intensity parameters Xt (t = 2, 4, 6) and the spontaneous emission lifetime of the 4I13/2 state of Er3+ ions have been investigated. The stimulated emission cross-sections of the 4I13/2 ? 4I15/2 transition of Er3+ ions have been calculated according to the McCumber theory. Influence of the glass compositions on the spectroscopic properties of Er3+ ions at 1.5-lm band has also been discussed.
1982
Q. Qian et al. / Journal of Non-Crystalline Solids 354 (2008) 1981–1985
2. Experimental details
0.8 2
3. Results The various properties of the glasses with different compositions are summarized in Table 1. The densities and the refractive indices of these samples increase with increasing Sb2O3 content added in the glasses. Fig. 1 shows the absorption spectra of Er3+-doped Na2O–Sb2O3–B2O3–SiO2 glasses at room temperature. Seven peaks in the absorption spectra correspond to the 4 I15/2 ? 4F7/2, 2H11/2, 4S3/2, 4F9/2, 4I9/2, 4I13/2 transitions, respectively. It is found that the absorption edge of glasses shifts from 430 to 470 nm with increasing Sb2O3 content, while no shift has been observed of the seven absorption peaks. Table 2 summarizes the calculated radiative probabilities, the radiative lifetimes and the branching ratios of Er3+ ions in SBS1 glass by the JO theory. The calculated spontaneous emission lifetime of 4I13/2 level of Er3+ ions
H11/2
0.6
Absorption (a.u.)
Er3+-doped antimony-borosilicate glasses with the composition of 5Na2O–xSb2O3–(55x)B2O3–39SiO2–1Er2O3 (x = 20, 25, 30, 35 mol%, namely SBS1, SBS2, SBS3, SBS4) were prepared by the melting-quenching technique. Analytical reagents Na2CO3, H3BO3, Sb2O3, SiO2 and Er2O3 were selected as raw materials. The batches of 100 g were mixed homogenously in agate mortars, and then melted at 1400 °C in corundum crucibles for 30 min in air. The melts were poured onto a stainless steel plate. After annealing, all the samples were cut and polished to 10 10 1 mm3 size for optical measurements. The refractive indices of the samples were recorded on a Metricon 2010 by means of a prism coupling method. The optical absorption spectra of the samples were measured on a Perkin–Elmer Lambda-900 UV/VIS/NIR spectrophotometer. The luminescence spectra were recorded on a Triax320 spectrophotometer with InGaAs as detector in the wavelength region of 1400–1650 nm under excitation of a 980 nm laser diode (LD). The lifetimes were obtained by exciting the samples with a 980 nm LD. The signals detected by an InGaAs photodetector were recorded by a storage digital oscilloscope. In our experiment, all data in figures and tables were analyzed for error. Generally, random errors were found due to unavoidable factors were expressed in term of a standard deviation. Other, systematic errors are also provided.
SBS1 SBS2
4
F7/2
0.4
4
4
F9/2
S3/2
SBS3 4 4
I9/2
I11/2
4
SBS4
I13/2
0.2 4
F5/2 400
600
800
1000
1200
1400
1600
Wavelength (nm)
Fig. 1. Absorption spectra of Er3+-doped Na2O–Sb2O3–B2O3–SiO2 glasses.
Table 2 Transition levels, wave numbers, transition rates and radiative lifetimes of Er3+ ions in SBS1 glass Transitions
Wave number (cm1)
Aed (s1)
Amd (s1)
b
s (ls)
4
6533 10 284 3751 12 491 5958 15 368 8835 5084 2877 18 484 11 951 8200 5993
203 262 33 208 85 2498 125 118 4 2235 920 71 113
58.725
1.00 0.84 0.16 0.71 0.29 0.91 0.05 0.04 0.00 0.67 0.28 0.02 0.03
3822
I13/2 ? 4I15/2 I11/2 ? 4I15/2 4 I13/2 4 I9/2 ? 4I15/2 4 I13/2 4 F9/2 ? 4I15/2 4 I13/2 4 I11/2 4 I9/2 4 S3/2 ? 4I15/2 4 I13/2 4 I11/2 4 I9/2 4
17.145
3199 3412
364
299
8
The RMS of the fitting procedure is 9.1 10 .
in SBS1 glass is 3822 ls, which is larger than that of Er3+-doped tellurite glass (2950 ls) [13]. The normalized infrared fluorescence spectra of Er3+-doped Na2O– Sb2O3–B2O3–SiO2 glasses in the wavelength range between 1400 and 1650 nm are shown in Fig. 2. Broad emissions at 1.5-lm are assigned to the 4I13/2 ? 4I15/2 transition of Er3+ ions. It is noteworthy that the SBS1 glass exhibits the maximum FWHM of 88 nm in the range of experimental composition.
Table 1 Refractive indices, densities and Er3+ concentrations of Er3+-doped Na2O–Sb2O3–B2O3–SiO2 glasses Glass samples
Sb2O3 (mol%)
Density (g/cm3)
Refractive index
Er3+ concentration (1020 ions/cm3)
SBS1 SBS2 SBS3 SBS4
20 ± 0.043 25 ± 0.054 30 ± 0.065 35 ± 0.076
3.283 ± 0.005 3.497 ± 0.005 3.620 ± 0.005 3.809 ± 0.005
1.787 ± 0.001 1.810 ± 0.001 1.857 ± 0.001 1.881 ± 0.001
3.353 ± 0.050 3.325 ± 0.050 3.232 ± 0.049 3.179 ± 0.048
Normalized emission intensity (a.u.)
Q. Qian et al. / Journal of Non-Crystalline Solids 354 (2008) 1981–1985
x=20 x=25 x=30 x=35
1400
1450
1500
1550
1600
1650
Wavelength (nm)
Fig. 2. Normalized infrared fluorescence spectra in the wavelength range of 1400–1650 nm under excitation of a 980 nm LD.
4. Discussion According to the JO theory [14,15], the line strength (Sed) of the electric dipole component of the 4I13/2 ? 4I15/2 transition of Er3+ ions is the function of glass structure and composition and can be calculated [14,15]. The Sed is expressed by [16]: S ed ½4 I 13=2 ;4 I 15=2 ¼ 0:0188X2 þ 0:1176X4 þ 1:4617X6
ð1Þ
where the three coefficients of Xt (t = 2,4,6) are the reduced matrix elements of the unit tensor operators provided in Ref. [16], and coefficients Xt (t = 2, 4, 6) are the intensity parameters. Evidently, the value of Sed is mainly deter-
5.4
5.0
Ω2
4.8 1.6
Ω6
4.6
Ω2 (10-20 cm2)
Ω4 and Ω6 (10-20cm2)
5.2
Ω4
1.8
4.4 1.4
4.2 20
24
28
32
36
Sb2O3content (mol.%)
Fig. 3. Compositional dependence of Xt (t = 2, 4, 6) parameters of Er3+doped Na2O–Sb2O3–B2O3–SiO2 glasses. Solid lines are drawn as a guide to the eyes.
1983
mined by X6 parameter. Fig. 3 exhibits the compositional dependence of Xt (t = 2, 4, 6) parameters of Er3+-doped Na2O–Sb2O3–B2O3–SiO2 glasses. The value of X2 increases monotonically with increasing Sb2O3 content, while the value of X6 decreases. The value of X4 is nearly invariant with the glass composition. According to the Judd–Ofelt theory, X2 is related with the symmetry of glass while X6 is inversely proportional to the covalency of the Er–O bond. The covalency of the Er–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 [17]. In antimony-borosilicate glasses, along with the substitution of B2O3 for Sb2O3, the number of the borate groups [BO3] trigonal and [BO4] tetrahedral units in the network structure increases. Therefore, much more of the nonbridge oxygen ions, which tend to coordinate with Er3+, will contribute to coordinate with glass former cation ions. On the other hand, based on the electronegativity theory [18], the smaller the difference of electronegativity between cation and anion ions, the stronger the covalency of the bond. The values of electronegativity for Sb, B, and O elements are about 1.9, 2.0, and 3.5, respectively. As a result, the covalency of the Sb–O bond is weak than that of the B–O bond. It is expected that the influence of the B–O bond on the local ligand environments around Er3+ increases with increasing B2O3 content. Consequently, the covalency of the Er–O bond in antimony-borosilicate glasses decreases with increasing B2O3 content, then the values of X6 and Sed increase accordingly. For the transitions between the states which meet the transition selective rules DS = DL = 0, DJ = 0, 1, there exists the contribution of magnetic dipole transitions (Smd) [15,19]. Smd is independent of the ligand fields. The bandwidth of the 4I13/2 ? 4I15/2 transition of Er3+ ions is proportional to the Sed/(Sed + Smd) value. The larger the Sed/(Sed + Smd) value, the larger the bandwidth. Table 3 shows the line strength ratio Sed/(Sed + Smd) of the 4I13/ 4 3+ ions in various glasses. It can 2 ? I15/2 transition of Er be clearly observed that antimony-borosilicate glasses possess larger Sed/(Sed + Smd) values than other glass hosts, which results in the broad FWHM of Er3+-doped antimony-borosilicate glass. Along with the increase of Sb2O3 content from 20 to 35 mol%, FWHM varies from 88 to 71 nm. The large values of FWHM could be beneficial to the WDM system. According to the McCumber theory [20], the emission cross-section (re) of the 4I13/2 ? 4I15/2 transition of Er3+ can be calculated from the measured absorption cross-section (ra), which are related by: re ðkÞ ¼ ra ðkÞ exp½ðe hmÞ=kT
ð2Þ
Table 3 Line strength ratio Sed/(Sed + Smd) of 4I13/2 ? 4I15/2 transition of Er3+ in various glasses Glass
SBS1
Bismuth [22]
Tellurite [23]
Fluoride [21]
Silicate [21]
Phosphate [23]
Sed/(Sed + Smd)
0.787 ± 0.016
0.785
0. 763
0.683
0.675
0.652
where h is the Planck constant, k the Boltzmann constant, and e the net free energy required to excite one Er3+ from the 4I15/2 state to 4I13/2 state at temperature T. The ra and e values can be calculated according to the experimentally obtained absorption spectra and the simplified procedure is provided in Ref. [21]. The measured absorption crosssections and the emission cross-sections of Er3+ ions in Na2O–Sb2O3–B2O3–SiO2 glasses are illustrated in Fig. 4. It is evident that the SBS4 glass (x = 35) exhibits larger absorption and emission cross-section than that of the SBS1 glass (x = 20). Moreover, the line shape of the emission cross-sections in Er3+-doped antimony-borosilicate glasses is broad and flat, so that the gain excursion over the whole spectral width can be minimized in the WDM system. Fig. 5 illustrates the compositional dependence of the peak emission cross-section and the integrated emission cross-section of the 4I13/2 ? 4I15/2 transition of Er3+ ions in Na2O–Sb2O3–B2O3–SiO2 glasses. The data in Fig. 5 suggest that for antimony-borosilicate glasses, the peak emission cross-sections possess large values and have slightly compositional dependence. Previous studies have shown [22,23] that glass composition has an important effect on the absorption and emission cross-sections of rare-earth ions as well as on the shapes of the ion bands. The emission cross-section is expected to increase with increasing refractive index of a glass host, because the emission cross-section that is due to the electric dipole transitions of rare-earth ions increases as the refractive index of the glass host [re (n2 + 2)2/n] increases [24]. With the increase of Sb2O3 content from 20 to 35 mol%, the refractive index
Absorption cross-section Emission cross-section
Cross-sections (10-21cm2)
8
x=35
6
4
x=20
2
0 1400
1450
1500
1550
1600
1650
Wavelength (nm)
Fig. 4. The measured absorption cross-sections and the calculated emission cross-sections of Er3+ ions in Na2O–Sb2O3–B2O3–SiO2 glasses.
60
0.74
58 0.72 56 0.70 54
0.68
52
20
24
28
32
Integrated emission cross section (10-20 cm2nm)
Q. Qian et al. / Journal of Non-Crystalline Solids 354 (2008) 1981–1985
Peak emission cross section (10-20 cm2)
1984
36
Sb2O3 content (mol.%)
Fig. 5. Compositional dependence of the peak emission cross-sections and the integrated emission cross-sections of Er3+ ions in Na2O–Sb2O3–B2O3– SiO2 glasses. Solid lines are drawn as a guide to the eyes.
of the glass hosts increases and the peak emission cross-section increases from 6.6 1021 to 7.3 1021 cm2, while the integrated cross section slightly decreases from 59.75 1020 to 51.95 1020 cm2 nm. The product of the emission cross-section and the lifetime, re FWHM, is an important parameter to characterize laser materials, because the gain bandwidth of an amplifier can be evaluated by re FWHM. Table 4 lists re, FWHMs and re FWHM in various glass hosts. Obviously, the antimony-borosilicate glass is better than other glasses as a host material for Er3+ doping for a broadband amplifier. The lifetime of the 4I13/2 level and quantum efficiency of 4 I13/2 ? 4I15/2 transition of Er3+ ions are directly related to amplifiers’ performance and treated as key parameters in the spectroscopic analysis. Fig. 6 illustrates compositional dependence of the experimental lifetime, the calculated lifetime by the JO theory and the quantum efficiency of the 4 I13/2 level of Er3+ ions in the antimony-borosilicate glasses. Obviously, the experimental lifetime increases monotonically with the increase of Sb2O3 content. The experimental lifetime (sexp) of the 4I13/2 level of Er3+ ions 1 is given by s1 exp ¼ scal þ W mp þ W et [25], where Wmp, Wet and scal denote the multiphonon decay rate, the energy transfer rate between Er3+ ions and the calculated lifetime by the JO theory. Because the Er3+-doping concentrations in all samples are close, the energy transfer rate of Er3+–
Table 4 re, FWHM and re FWHM in various glass hosts Glasses
SBS1
Bismuthate
Tellurite
Silicate
Phosphate
Germanate
re (1021 cm2) FWHM (nm) re FWHM Reference
6.8 ± 0.5 88 ± 0.5 598.4 ± 48 This work
7.0 79 554.0 [22]
7.5 65 487.5 [23]
5.5 40 220.0 [23]
6.4 37 236.8 [4]
5.68 53 301.0 [10]
Q. Qian et al. / Journal of Non-Crystalline Solids 354 (2008) 1981–1985 30 4.0
Lifetime (ms)
25
τcal
3.5
20
0.9
15
Quantum efficiency (%)
η
τexp
10
0.3 25
30
infrared emission at 1.5-lm. FWHM of the 1.5-lm luminescence emission varies from 88 to 71 nm, and the peak emission cross-sections increases from 6.6 1021 to 7.3 1021 cm2 with increasing Sb2O3 content from 20 to 35 mol%, which are comparable or higher than that of Er3+-doped bismuth-based and tellurite-based glasses. These results indicate that the Er3+-doped Na2O–Sb2O3– B2O3–SiO2 is a promising host material for potential broadband optical amplifiers in the WDM system. Acknowledgement
0.6
20
1985
35
Sb2O3 content (mol.%)
Fig. 6. Compositional dependence of the experimental lifetime, the calculated lifetime and the quantum efficiency of the 4I13/2 level of Er3+ ions in Na2O–Sb2O3–B2O3–SiO2 glasses. Solid lines are drawn as a guide to the eyes.
Er3+ ions should also be close for all samples. Therefore, it can be deduced that the compositional dependence of measured lifetime is mainly determined by the joint effect of the radiative decay rate and the multiphonon decay rate. Since the phonon energy of antimonite glass is lower than that of borate glass, the substitution of Sb2O3 for B2O3 leads to prolonging the experimental lifetime by decreasing the phonon energy of the glasses and the multiphonon decay rate in the glass host. Quantum efficiency (g) is another important parameter for a luminescence material. In rare-earths doped luminescent materials, the quantum efficiency of a certain energy level means the ratio of the number of photons generated in a radiative transition to the number of electrons excited to the excited state. The quantum efficiencies of the 4I13/2 level of Er3+ ions in Na2O–Sb2O3–B2O3–SiO2 glasses are calculated by the formula g ¼ sexp =scal [26]. Fig. 6 shows the dependence of quantum efficiency on the glass composition. It is noted that the monotonic increase of experimental lifetime with increasing Sb2O3 content promotes the quantum efficiency from 9.3 to 28.7%, owing to the decrease of the multiphonon decay rate in the glass host. 5. Conclusions In summary, spectroscopic properties of the 1.5-lm emission from 4I13/2 ? 4I15/2 transition of Er3+ ions in Na2O–Sb2O3–B2O3–SiO2 glasses have been investigated for developing broadband fiber amplifier. It is found that the glasses studied exhibits significant broad and flat near
The authors would like to acknowledge the support received from the NSFC (50472053, 50602017), and DSTG (2006J1-C0491). References [1] C.G. Atkins, J.F. Massicott, J.P. Armitage, R. Wyatt, B.J. Ainslie, S.P. Craig-Ryan, Electron. Lett. 25 (1989) 910. [2] J.F. Massicott, J.R. Armitage, R. Wyatt, B.J. Ainslie, S.P. CraigRyan, Electron. Lett. 26 (1990) 1645. [3] Y. Ohishi, A. Mori, M. Yamada, H. Ono, Y. Nishida, K. Oikawa, Opt. Lett. 23 (1998) 274. [4] X. Feng, S. Tanabe, T. Hanada, J. Am. Ceram. Soc. 84 (2001) 165. [5] A. Mori, K. Kobayashi, M. Yamada, Electron. Lett. 34 (1998) 887. [6] S. Jiang, T. Luo, B.C. Hwang, F. Smekatala, K. Seneschal, J. Lucas, N. Peyghambarian, J. Non-Cryst. Solids 263 (2000) 364. [7] B.C. Hwang, S. Jiang, T. Luo, K. Seneschal, G. Sorbello, M. Morrell, F. Smektala, S. Honkanen, J. Lucas, N. Peyghambarian, IEEE Photon. Technol. Lett. 13 (2001) 197. [8] Q.Y. Zhang, T. Li, D.M. Shi, J. Appl. Phys. 99 (2006) 033510. [9] M. Wachtler, A. Speghini, K. Gatterer, H.P. Fritzer, D. Ajo, M. Bettinelli, J. Am. Ceram. Soc. 81 (1998) 2045. [10] K. Soga, H. Inoue, A. Makishima, J. Non-Cryst. Solids 274 (2000) 69. [11] I.I. Oprea, H. Hesse, K. Betzler, Opt. Mater. 28 (2006) 1136. [12] Y. Kuroiwa, N. Sugimoto, K. Ochiai, S. Ohara, Y. Fukasawa, S. Ito, S. Tanabe, T. Hanada, IEEE OFC 2001 TuI5, 2001, p. 1. [13] H. Chen, Y.H. Liu, Y.F. Zhou, J. Non-Cryst. Solids 351 (2005) 3060. [14] B.R. Judd, Phys. Rev. 127 (1962) 750. [15] G.S. Ofelt, J. Chem. Phys. 37 (1962) 511. [16] M.J. Weber, Phys. Rev. 157 (1967) 262. [17] S. Tanabe, S. Yoshii, K. Hirao, N. Soga, Phys. Rev. B 45 (1992) 4620. [18] L. Pauling, J. Am. Chem. Soc. 51 (1929) 1010. [19] S. Tanabe, T. Ohyagi, N. Soga, T. Hanada, Phys. Rev. B 46 (1992) 3305. [20] D.E. McCumber, Phys. Rev. 134 (1964) A299. [21] W.J. Miniscalco, R.S. Quimby, Opt. Lett. 16 (1991) 258. [22] H. Takebe, Y. Nageno, K. Morinaga, J. Am. Ceram. Soc. 78 (1995) 1161. [23] S. Tanabe, T. Ohyagi, S. Todoroki, T. Hanada, N. Soga, J. Appl. Phys. 73 (1993) 8451. [24] J.S. Wang, E.M. Vogel, E. Snitzer, Opt. Mater. 3 (1994) 187. [25] Y.S. Han, J.H. Song, J. Heo, J. Appl. Phys. 94 (2003) 2817. [26] Z.C. Yang, S.C. Hang, S.Z. La, B.J. Cheng, J. Non-Cryst. Solids 343 (2004) 154.
