Open Access
Add to BibSonomyAbstract
Broadband NIR emission from Er3+ in highly nonlinear zinc-bismuth-borate glasses with full width at half maximum reaching up to 106 nm is reported. The glass compositional effects on thermal, structural, optical and spectroscopic properties have been explored in view of all-optical communication applications. The Zero Dispersion Wavelength of these glasses is found to be varying from 1.46 to 2.26μm. The observed enhancement in fluorescence intensity, lifetime and quantum efficiency of 4I13/2 → 4I15/2 transition with bismuth addition is attributed to reduced multiphonon relaxations with decreased host phonon energy. The gain profile covering C and L-bands of communication windows suggests their potentiality in broad-band amplification.
©2011 Optical Society of America
1. Introduction
Heavy metal oxide based glasses have been intensively investigated over the past few decades for their promising prospective in the development of all-optical technology towards the faster data communication [1–5]. An all-optical technology includes the optical signal generation; optical de-multiplexing of signals as well as optical wavelength conversion of Wavelength Division Multiplexing (WDM) signals. It is believed that an high capacity all-optical communication system can be realized using a nonlinear material as a ultra-fast switching device along with broadband Er3+ emission gain profile for signal amplification of wider range [6–8]. Today’s commercial WDM system is based on optoelectronics technology, where the electronic part is a main constraint in the slower data handling rate. Further, the silica based Er3+ doped fiber amplifiers (EDFA) allows only a limited number of channels owing to the narrow emission bandwidth spanning a mere 45 nm. The rapidly growing internet users and huge data traffic suggests that the present optoelectronic technology is inadequate for this enormous loading and demands faster communication system with signal amplifiers having enhanced number of channels. Heavy metal oxide based glasses are advantageous for such high speed switching applications due to their high non-linear coefficient, especially the third order non-linear susceptibility [3,4]. Additionally, these glasses are also found to exhibit more broadened Er3+ emission profile, which is required for the development of multi-channel broadband amplifiers [8–10].
Telluride glasses were the first reported for exhibiting broadband emission of Er3+ ions with full width at half maximum (FWHM) of around 76 nm [8,11]. The high refractive index of such heavy metal oxide glasses has provided an added advantage in higher stimulated emission cross section, which enhances the gain of amplifier. Sugimoto and associates commercialized the first heavy metal oxide based Er3+ doped fiber amplifier having ultra-broadband characteristics using bismuth-borate host [12]. Further, the same group has reported highly non-linear fibers with refractive index of 2.2 at lasing wavelength for all-optical signal processing [4]. The main advantage of bismuth based fibers over telluride is the ease in splicing with the well-established silica carrier fibers. Thus, a great deal of work has been carried out in different compositions having bismuth oxide as the main constituent due to their potentiality in broadband communication [13–17]. The main concern of all these studies has primarily been limited to the spectroscopic properties of Er3+ ions.
Generally, with the addition of Bi2O3 in oxide glasses, it first starts to modify the network and then actively participates in the network formation at higher concentration. Being a conditional glass former, the glass stability decreases with Bi2O3 contents and such glasses could readily get crystallized. Thus, a detail understanding of thermal behavior of glasses is equally important along with spectroscopic properties in view of their practical applications. Among several multi-component systems, the bismuth-borate glasses are found to be the most suitable for broadband applications [9]. From our earlier work on borate systems, we observed that the binary zinc-borate system (4ZnO-3B2O3) possesses reasonably lower melting temperature and has good optical transparency alongside their having certain favorable spectroscopic properties for a broadband emission, like non-centrosymmetric ligand field around active ions and high rare earth ion solubility [18]. Further, the presence of zinc increases the linear refractive index in comparison with single component borate glasses adding up to the non-linearity of glasses and also additionally it gives stability to the glass system. Keeping all these in view, in the present investigation, binary zinc-borate system has been selected for studying the effect of bismuth inclusion on thermal, structural, optical and spectroscopic properties of Er3+ ions in view of broadband all-optical communication.
2. Experimental
2.1 Glass preparation
Conventional melt quenching technique was employed in the preparation of glasses with compositions in mol% (99.5-x) (4 ZnO – 3 B2O3) – x Bi2O3 – 0.5 Er2O3, where x = 0, 5, 10, 20, 30, 40 and 60. High purity reagent grade chemicals of H3BO3, ZnO (Fluka, 99.9%), Bi2O3 (Aldrich, 99.9%) and Er2O3 (Alpha-Aesar, 99.99%) weighed in appropriate quantities, were first mixed thoroughly for 30 minutes in agate mortar and then sintered overnight at 400°C to make pre-reacted batches. The melting of sintered batches was carried out in pure platinum crucible at 800 – 1200°C temperatures depending upon the glass compositions for 45 minutes with intermittent stirring to ensure melt homogeneity. The cast glasses were then annealed at 320 – 550°C for 1 hour and later cooled slowly to room temperature. It is observed that, the melting and annealing temperatures were decreased with an increase in the bismuth content in the chemical composition. The obtained glasses then were cut and polished in 15 x 15 x 1.5 mm3 dimensions for optical measurements. All the glasses have been labeled based on the presence of bismuth oxide content in host matrix systems studied.
2.2 Characterization
In confirming the amorphous nature of all prepared samples, X-ray diffraction (XRD) patterns were recorded on X’pert Pro MPD diffractometer (PANalytical, Almelo, Netherlands) with an X’Celerator operating at 40 kV potential and 30 mA current using Ni-filtered CuKα radiation of wavelength 1.5418 Å. The data were acquired in the step-scan mode with a step size of 0.05° (2θ) and a step time of 30 sec from 100 to 80°. Differential thermal analysis was carried out to study the effect of Bismuth addition on various thermal events of the glasses in the temperature range of 40°C to 1000°C on a differential thermal analyzer (DTA) (model STA 409, Netzsch-Gera¨ tebau GmbH, Selb, Germany) given a heating rate of 10 K/min. The infrared absorption spectra of the glasses were recorded on a Fourier transform infrared spectrometer (FTIR) (model 1615, Perkin-Elmer, Norwalk, CT) in the wavenumber range 400 – 1500 cm−1 by using KBr pellets. The glass density was measured by Archimedes’ buoyancy principle using water as immersion liquid on Mettler-Tollado digital balance fitted with a density measurement kit. To study the dispersion properties of the prepared glasses, refractive indices at five different wavelengths (473 nm, 532 nm, 633 nm, 1064 nm, and 1552 nm) were recorded with an accuracy of 10−5 on a Prism Coupler refractometer (Model Metricon M-2010, Pennington, NJ) built-in with five different lasers as illuminating sources. The optical absorption spectral measurements were carried out on a UV-Vis-NIR spectrophotometer (Shimadzu, Japan, Model: 3001) in the 200 – 1700 nm wavelength range. The luminescence spectra of Er3+ ions in NIR region were obtained on a Fluorescence spectrophotometer (Model: Quantum Master-enhanced NIR, from Photon Technologies International, USA) having double monochromators on both excitation and emission channels. The instrument is equipped with LN2 cooled gated NIR photo-multiplier tube (Model: NIR-PMT-R1.7, Hamamatsu) as detector for acquiring the data for both steady-state emission and phosphorescence decay. For performing decay measurements, a 60W Xenon flash lamp was employed as an excitation source. All the measurements were carried out by placing the samples at 600 to the incident beam and the signals were collected from the same surface at right angle to the incident beam.
3. Results and discussion
3.1 Thermal and structural properties
The DTA thermograms of all studied glasses are presented in Fig. 1a . Except Bi-0 glass, all other samples show three distinct regions corresponding to glass transition (Tg, first endotherm), crystallization process (exotherms) and melting phenomena (final endoterms) within the examined temperature range. For Bi-0 glass, the event of melting found to be beyond 1000°C temperature. The display of distinct Tg followed by crystallization peaks by all samples indicates their glassy nature. Thermal properties obtained from DTA curves of glasses are listed in Table 1 . It is clear from the data that glass transition (Tg), crystallization onset (Tx), crystallization peak (TP) and melting temperature (Tm) along with the thermal stability factor (Tx – Tg) decrease with bismuth addition. It is to mention here that, though the glass stability factor is decreasing with bismuth content, it is still well above 60°C for the highest bismuth containing glass, Bi-60; manifesting its convenience towards fiber drawing and similar applications. This decrease in glass stability with bismuth addition is due to its conditional forming ability. Although bismuth is non-conventional glass former with fieldstrength less than 1.3; when substituted for boron in borate based glasses, it can form stable glasses in a wide compositional range [19,20] because of its hyper-polarizability. The inclusion of bismuth in the borate glass network brings in structural modifications owing to the differences in ion size as well as field strength. This has been emphasized from the X-ray diffraction pattern of the present zinc-borate glasses shown in Fig. 1b. The binary zinc-borate glass, Bi-0 exhibits broad hollow bands, characteristic of the glassy nature. With an increase in bismuth content, there exists a clear shifting of these bands and are all in sharp. The sharpening of hollow XRD band suggests more ordered local network with additionally expanded oxide coordination spheres, resulting in the shift of band positions due to significant formation of Bi-O related structural units in place of B-O units [21].
Table 1. Thermal Properties of Zinc-Bismuth-Borate Glasses
A more detailed study on the structural modifications by bismuth inclusion has been carried out using the FTIR transmission spectra as shown in Fig. 1c. For binary zinc-borate glass, Bi-0, three major absorptions bands have been observed in the wavenumber ranges of 1500 – 1200 cm−1, 1100 – 800 cm−1 and 800 – 600 cm−1 due to different vibrational transitions of BO3 and BO4 structural units. Among them, the absorption bands at around 1370 cm−1 and 1235 cm−1 can be attributed to the BIII – O – BIII and BIII – O – BIV stretching vibrations respectively. Similarly, the BIV – O – BIV stretching vibration is observed at around 1030 cm−1 with a broad shoulder peaking at around 910 cm−1 due to B – O- non-bridging vibration. The B – O – B bending vibration is observed at around 700 cm−1. On inclusion of bismuth in the network, the higher energy vibration bands gradually shift towards lower energy, which resulted in the complete fading of 1370 cm−1 vibration pertaining to BIII – O – BIII bond in Bi-60 glass along with appearance of a new absorption peak at around 1140 cm−1 accompanying a shoulder at 1235 cm−1 due to BIII – O – BIV stretching. Further, two more absorption transitions emerged in the region of 900 – 800 cm−1 and 600 – 400 cm−1 with increasing in bismuth content. These observations clearly indicate that, there may be conversion of BO3 to BO4 units at the lower bismuth contents and later, the bismuth ions are replacing boron in BO3 structural units, which may be the reason behind disappearance of 1370 cm−1 band at higher bismuth contents. In addition, it also forms BiO6 octahedral units, which constitute the main structural network in bismuthate glasses [22]. The strong absorption in the 400 – 600 cm−1 region is due to the bending vibrations of Bi – O bond in BiO3 pyramid and BiO6 octahedron occurring at ~480 cm−1 and ~570 cm−1 respectively. Further the absorption at around 840 cm−1 arises due to the symmetrical stretching vibrations of Bi – O in BiO3 and BiO6 units [22,23]. It can be seen that the B – O – B bending vibration at 700 cm−1 is becoming weaker in intensity and narrower with increase in bismuth content. This is due to the decrease in overall borate content in glasses. Moreover, there may be overlap of this band with the second harmonic of bending vibrational band at 350 cm−1 due to Bi2O3. The strong vibration band emerging at 1140 cm−1 in high bismuth containing glasses can be attributed to the Bi – O – B stretching vibrations. As the highest Bi – O stretching occurs at 840 cm−1 and at low borate contents, B – O alone cannot exhibit such strong absorption; hence it is anticipated that this band may be a collective effect of both the cations (B and Bi) as it is located in between the major vibrations of Bi – O and B – O bonds. This attribution has further been strengthened from the observation of its becoming more significant with an increase in bismuth content.
3.2 Physical and optical properties
The Er3+ ion concentration in glasses was estimated from the average molecular weight and the measured density of the well annealed glasses. Table 2 enlists some of the important physical and optical properties of present bismuth modified zinc-borate glasses. The average molecular weight and density increase steeply with the addition of bismuth. This enhanced density and hyper-polarizability of bismuth ions have brought out an increase in the linear refractive index and thereby the non-linear optical properties of these glasses. The linear refractive indices at standard wavelengths have been determined from the dispersion curves obtained by the simulation of five coordinate Sellmeier relation [24,25] as shown in Fig. 2 and were used to derive the non-linear optical parameters. From the data of Table 2 it is clear that, the Abbe number has decreased to 11.6 from 49.9 on 60 mol% Bi2O3 addition indicating more dispersive nature of bismuth containing glasses.
Table 2. Physical and Optical Properties of Zinc-Bismuth-Borate Glasses
In view of the signal communication through fibers, the dispersion plays an important role in determining the maximum signal rate [26]. Figure 3 presents the material dispersion of present glasses at standard communication region with varying bismuth contents. It can be seen that the zero material dispersion wavelength (ZDW) is consistently shifting towards the longer wavelength with bismuth inclusion and attains a maximum of 2.261 μm for Bi-60 glass. As a result, the material dispersion at the lasing wavelength or within the C-band region increases continuously attaining the value around −150 ps/nm/km. Nevertheless, this property may not deteriorate the amplification benefits for the cases of short length fibers and if necessary, it can be nullified by applying dispersion shifting fibers or by waveguide engineering.
Further, such highly dispersive and non-linear properties are advantageous for the realization of all-optical WDM communication system components. The non-linear refractive index of Bi-60 glass is 99.6 × 10−13 esu, which is one of the highest in bismuthate glasses suggesting their potentiality for high speed non-linear applications in WDM as well as in optical switching [3,4].
3.3 Absorption spectra and Judd-Ofelt analysis
Figure 4 depicts the UV-visible-NIR absorption spectra of Er3+ doped zinc-borate glasses with varied bismuth contents. On inclusion of bismuth, the baseline is found to be elevated, which has been attributed to the increased reflection loss at sample surfaces. The UV band edge also exhibits a shift towards the longer wavelength with an increase in bismuth content. This red shift occurs typically for heavy metal ions due to the weaker metal-oxygen bond strength. In addition, the strong absorption of bismuth in the UV region further adds up to this band shift reducing band gap energy (Eg) from 4.9 eV to 2.4 eV on 60 mol% Bi2O3 inclusions [18]. This has resulted in the masking of many characteristic absorption bands of Er3+ ions in the UV-visible region with the gradual increase of bismuth content in the glass series. However, seven absorption bands could clearly be seen for all glasses including highest bismuth containing Bi-60 glass as indexed in the figure. These well defined bands have been used for the estimation of three Judd-Ofelt intensity parameters (Ω2,4,6), which could provide intrinsic information regarding the symmetry and bonding of the active ions with ligand field [27,28]. Absorption spectra of glasses were all base glass corrected, in order to obtain accuracy in the parameterization.
Fig. 4 UV-Vis-NIR absorption spectra of zinc-bismuth-borate glasses. (Inset: Plot of (αhν)n versus hν for bandgap calculations. (n = 0.5 and 2)).
Table 3 lists the experimental and calculated line strengths of the transitions considered for Judd-Ofelt analysis in selected glasses. The corresponding Judd-Ofelt intensity parameters are also tabulated in the same table. It can be seen that the values of intensity parameters follow a trend of Ω2 > Ω6 > Ω4 for all the glasses studied. Ω2 is usually related to the degree of covalency of the RE ─ O bond as well as the asymmetry around the rare earth ion. The higher value of Ω2 results from the relatively superior oscillator strength of hypersensitive transitions (4I15/2 → 4G11/2 and 2H11/2) representing strong covalency of RE ─ O bond and the asymmetry in vicinity of the dopant ion. However, on bismuth inclusion in the glass network, the value of Ω2 parameter is found to be decreasing, which may be due to the occurrence of consistent structural changes from glass to glass with dominance in theincreasing orderliness of local environment over the nature of bonding. This is in concord with FTIR and XRD measuremental analysis. Similarly, the Ω4 and Ω6 parameters are also showing a decreasing trend with increasing bismuth. Though these parameters are considered to be insensitive to the structural changes, however they do depend on the radial part of the wave function and mostly on the basicity of the glass host [29]. In the present glass series, the glass optical basicity is increasing as the bismuth content is increasing resulting in the decrease in Ω4 and Ω6 parameters.
Table 3. Experimental and Calculated Line Strengths, S ( × 10−20); Oscillator Strengths, f ( × 10−6) and Judd-Ofelt Intensity Parameters
3.4 Emission and fluorescence decay
Figure 5 represents the infrared luminescence spectra of Er3+ ions in present glasses. The spectra depict a broad emission band due to 4I13/2 → 4I15/2 transition of Er3+ ion with several stark components ranging from 1450 – 1650 nm. The intensity of emission peak increases with the increase in bismuth in glasses along with its broadening. The insets of the figure show deconvoluted emission spectra of Bi-0 and Bi-60 glasses along with stark component energy level diagram. The quality of the curve fit is justified by the best superimposition ofthe fitted curve over the measured emission peak. The deconvolution has displayed five major transitions between energy level stark components as represented in the inset peaking at around 1470 nm (2 – 0), 1497 nm (1 – 0), 1532 nm (0 – 0), 1569 nm (0 – 1) and 1596 nm (0 – 2) respectively. Upon critical observation of emission peak, it can be seen that the relative intensity of (0 – 1) component is higher than that of (0 – 0) component in all glasses indicting there is slight energy re-absorption at the considered dopant ion concentration. Also, component at 1497 nm is becoming significantly stronger compared to others and 1596 nm is red shifted in Bi-60 glass. These changes in the emission profile are growing constantly with bismuth percentage resulting in a more flattened emission spectrum for Bi-60 glass with an enhanced bandwidth of 106 nm compared to a mere 68 nm in Bi-0 glass as seen from Table 4 . The highest energy component of the emission band further becoming prominent with progressive red shift may be attributed to decreased effective phonon energy and increased glass basicity. Similar observations can be made from the fluorescence decay spectra of 4I13/2 → 4I15/2 transition of Er3+ ion in Fig. 6 , where the lifetime is continuously increasing with bismuth contents. The decay profile is single exponential in nature and the measured values of the fluorescence lifetimes have been tabulated in the Table 4.
Fig. 5 Emission spectra of zinc-bismuth-borate glasses. (Inset: Deconvoluted spectra of Bi-0 and Bi-60 glasses).
Table 4. Radiative Transition Rate (Arad), Fluorescence Decay Time (τrad and τexp), Emission Cross Section (σem), Fluorescence Bandwidth (Δλ), Quantum Efficiency (η) and Gain Bandwidth (σem × ∆λ) of 4I13/2 → 4I15/2 Transition of Er3+ Ion
Fig. 6 Fluorescence decay spectra of zinc-bismuth-borate glasses. (Inset: Plot of variation of decay time and quantum efficiency with bismuth oxide contents).
Both emission intensity as well as the fluorescence lifetime of 4I13/2 → 4I15/2 transition of Er3+ ion are found to be increased with the addition of bismuth in the zinc-borate network. The FTIR results have already made it clear that the bismuth ions are actively involved in network formation. Since, bismuth is a heavy metal ion and the highest vibrational energy of Bi - O bonding is around 840 cm−1, it certainly reduces the local phonon energy in thevicinity of active ions compared to the borate network exhibiting strong absorption in the 1300 - 1400 cm−1 due to B - O stretching [21]. For such multi-component glasses containing both bismuthate and borate structural units, it becomes practically difficult to judge the exact phonon energy of the host from the FTIR spectra. The active ions experience strong site-to-site variations in such hosts. With the increase in bismuth contents, more number of active ions may get incorporated in bismuth network and thus experience lower phonon energy, which reduces the multi-phonon relaxation of 4I13/2 population resulting in enhancement of both emission intensity and decay time. Other external factors like hydroxyl impurity, which can act as high energy phonons, may also affect the spectral characteristics of the active ions. But the experimental conditions were maintained identical for all the samples and thus it is assumed insignificant keeping in view of the relativistic study in present case. The hydroxyl contents in glasses were estimated from FTIR spectra and observed to be around 65 ppm.
The Judd-Ofelt intensity parameters have been used in predicting key radiative properties of dopant ion. Table 4 presents important spectroscopic parameters like radiative transition rate and radiative decay time of 4I13/2 → 4I15/2 transition of Er3+ ions in present glasses derived theoretically using JO parameters. The radiative decay rate shows a monotonous enhancement with the bismuth content and thus reducing the radiative lifetime. The quantum efficiency of present glasses has been calculated from the radiative and experimental lifetime values and is found to be increasing with the bismuth contents and reaches to 44% for Bi-60 glass. The obtained 44% quantum efficiency for the highest bismuth containing glasses is comparable to most of the reported glasses for broadband applications. However, further improvement in the quantum efficiency is possible by the elimination of hydroxyl ions in present glasses. The stimulated emission cross section values derived using Judd-Ofelt analysis also increases with the bismuth addition and can be attributed to the enhanced refractive index of the glasses [30]. To confirm the accuracy of Judd-Ofelt analysis, stimulated emission cross sections have been recalculated from experimental absorption cross-section spectra using McCumber Reciprocity (RC) method and are depicted in Fig. 7 [24]. The peak emission cross section values for both theoretical (JO) and experimental (RC) cross sections have also been tabulated in Table 4, which are in close agreement to each other with acceptable deviation.
The most interesting feature of emission profile is the full width at half maximum (FWHM) of the Bi-60 glass, which is superior to any of the reported host till now [1,8]. The well-known broadband glasses such as bismuth-borate and telluride have reported about 70 - 80 nm of FWHM. The site-to-site variations of active ion due to multi-component structural units like BO3, BO4, BiO3 and BiO6 may be the main reason for observed broadband emission in present zinc-bismuth-borate system. However, a more detailed structural understanding is needed for providing a proper reasoning of this observation.
3.4 Gain Spectrum for Broadband Amplifier
A flat and wide gain spectrum covering most of the communication windows is advantageous for a broadband WDM system. For Er3+ ions in glassy host, the gain spectrum typically exhibits an inhomogeneously broadened spectrum with several peaks due to its stark components. Heavy metal oxide glasses possess more flattened gain compared to silica based EDFA as has been discussed in earlier sections. Further flattening of gain curve is also possible by applying optical filters like long-period fiber Bragg gratings. Figure 8 presents the gain coefficient spectrum of Bi-60 glass with different population inversion probabilities, P. It is interesting to see that the gain profile covers the spectrum for S-band (small), C-band (conventional) and L-band (long). Since the emission profile of Er3+ ions in present zinc-bismuth-borate glasses exhibits a Stokes shift towards longer wavelength, the L-band and C-band could easily be amplified with nominal inversion. However, the S-band needs considerably a strong excitation to cause lasing action. Further, the gain-bandwidth of Er3+ emission has also been calculated for present glasses using the emission cross section (σe)and FWHM values and is tabulated in the Table 4 for comparison, which is found to be superior than most of the systems reported in the literature [2,14,15]. A larger gain bandwidth reduces the overall expenses of WDM communication, since a single amplifier can be utilized to amplify all signals being carried out and thus the present Er3+ doped zinc-bismuth-borate glasses are found to be promising materials for efficient broadband amplification.
4. Conclusion
Highly nonlinear Er3+ doped bismuth modified zinc-borate glasses have been synthesized and studied for their thermal, structural, optical and spectroscopic properties with varied bismuth content. The glass stability has been found to be decreasing with bismuth inclusion in the matrix but it is still good enough for fiber drawing applications. The linear and nonlinear refractive index of glasses have significantly been enhanced with bismuth contents, giving a whopping nonlinear refractive index of 99.6 × 10−13 esu for Bi-60 glass, suggesting their potential for high speed non-linear switching in all-optical communication. The properties estimated based on theoretical Judd-Ofelt analysis have been in good agreement with the experimental results. The emission intensity and the fluorescence decay time for 4I13/2 → 4I15/2 transition of Er3+ ions enhanced with the bismuth inclusion in glasses and the effect has been explained through the reduced multi-phonon relaxation of excited level due to lower phonon energy of bismuth network. The full width at half maximum of emission has been found to be increased with bismuth content and that reaches to 106 nm for 60 mol% Bi2O3 containing glass, which is the highest over any reported glass. The gain-bandwidth and gain coefficient profile suggest that these glasses could be found as more promising materials for broadband amplification of both C and L-band signals.
Acknowledgments
Authors would like to thank Prof. I. Manna, Director, CGCRI for his kind encouragement and permission to publish this work that was carried out in an In-house project No. OLP-0288. One of us (Mr.A.D.S.) is thankful to the BRNS-DAE for the award of a Junior Research Fellowship to him.
References and links
1. N. Sugimoto, “Ultrafast optical switches and wavelength division multiplexing (WDM) amplifiers based on bismuth oxide glasses,” J. Am. Ceram. Soc. 85(5), 1083–1088 (2002). [CrossRef]
2. S. Shen, A. Jha, X. Liu, M. Naftaly, K. Bindra, H. J. Bookey, and A. K. Kar, “Tellurite glasses for broadband amplifiers and integrated optics,” J. Am. Ceram. Soc. 85(6), 1391–1395 (2002). [CrossRef]
3. K. Kikuchi, K. Taira, and N. Sugimoto, “Highly nonlinear bismuth oxide-based glass fibers for all-optical signal processing,” Electron. Lett. 38(4), 166–167 (2002). [CrossRef]
4. N. Sugimoto, T. Nagashima, T. Hasegava, S. Ohara, K. Taira, and K. Kikuchi, “Bismuth-based optical fiber with nonlinear coefficiemt of 1360 W−1km−1,” in Proceedings of Optical Fiber Commun, Fibers, 2004 (OFC-2004).
5. S. Tanabe, “Development of rare-earth doped fiber amplifiers for broad band wavelength-division-multiplexing telecommunication,” Photon. Based Wavel. Integr. Manip. IPAP Books 2, 101–112 (2005).
6. I. P. Kaminow, C. R. Doerr, C. Dragone, T. Koch, U. Koren, A. A. M. Saleh, A. J. Kirby, C. M. Ozveren, B. Schofield, R. E. Thomas, R. A. Barry, D. M. Castagnozzi, V. W. S. Chan, B. R. Hemenway, D. Marquis, S. A. Parikh, M. L. Stevens, E. A. Swanson, S. G. Finn, and R. G. Gallager, “A wideband all-optical WDM network,” IEEE J. Sel. Areas Comm. 14(5), 780–799 (1996). [CrossRef]
7. S. Kinoshita, T. Okiyama, and K. Moruya, “Wideband WDM erbium-doped optical fiber amplifiers for 10Gb/s, 32-channel SMF transmission system,” Fujistu Sci. Tech. J. 35, 82–90 (1999).
8. A. Mori, T. Sakamoto, K. Kobayashi, K. Shikano, K. Oikawa, K. Hoshino, T. Kanamori, Y. Ohishi, and M. Shimizu, “1.58-μm broad-band erbium-doped tellurite fiber amplifier,” J. Lightwave Technol. 20(5), 794–799 (2002). [CrossRef]
9. S. Tanabe, N. Sugimoto, S. Ito, and T. Hanada, “Broad-band 1.5 μm emission of Er3+ ions in bismuth-based oxide glasses for potential WDM amplifier,” J. Lumin. 87–89, 670–672 (2000). [CrossRef]
10. J. H. Shin and J. H. Lee, “investigation of signal excited-state absorption in bismuth-based erbium-doped fiber amplifier,” J. Opt. Soc. Am. B 27(7), 1452–1457 (2010). [CrossRef]
11. A. Mori, Y. Ohishi, M. Yamada, H. Ono, Y. Nishida, K. Oikawa, and S. Sudo, “1.5 μm broadband amplification by tellurite-based EDFAs,” in OFC’97, 1997, Paper PD1.
12. N. Sugimoto, K. Ochiai, T. Hirose, S. Ohara, Y. Fukasawa, H. Hayashi, K. Furuki, M. Tojo, and Y. Tanaka, “Ultracompact optical amplifier with Bi2O3-based erbium-doped fiber,” Reports Res. Lab. Asahi Glass Co. Ltd. 54 (2004).
13. S. Q. Man, S. F. Wong, and E. Y. B. Pun, “Erbium-doped potassium bismuth gallate glasses,” J. Opt. Soc. Am. B 19(8), 1839–1843 (2002). [CrossRef]
14. J. Yang, S. Dai, N. Dai, S. Xu, L. Wen, L. Hu, and Z. Jiang, “Effect of Bi2O3 on the spectroscopic properties of erbium-doped bismuth silicate glasses,” J. Opt. Soc. Am. B 20(5), 810–815 (2003). [CrossRef]
15. J. Yang, S. Dai, Y. Zhou, L. Wen, L. Hu, and Z. Jiang, “Spectroscopic properties and thermal stability of erbium-doped bismuth-based glass for optical amplifier,” J. Appl. Phys. 93(2), 977–982 (2003). [CrossRef]
16. L. R. P. Kassab, A. de Olievera Preto, W. Lozano, F. X. de Sá, and G. S. Maciel, “Optical properties and infrared-to-visible upconversion in Er-doped GeO–BiO and GeO–PbO–BiO glasses,” J. Non-crystal Solids 351(43-45), 3468–3475 (2005). [CrossRef]
17. S. Ohara and N. Sugimoto, “Bi2O3 based erbium doped laser with a tunable range over 130 nm,” Opt. Lett. 33(11), 1201–1203 (2008), http://www.opticsinfobase.org/abstract.cfm?URI=ol-33-11-1201. [CrossRef] [PubMed]
18. A. D. Sontakke, A. Tarafder, K. Biswas, and K. Annapurna, “Sensitized red luminescence from Bi3+ co-doped Eu3+:ZnO-B2O3 glasses,” Physica B 404(20), 3525–3529 (2009). [CrossRef]
19. E. M. Levin and C. L. McDaniel, “The system Bi2O3-B2O3,” J. Am. Ceram. Soc. 45(8), 355–360 (1962). [CrossRef]
20. A. Bishay and C. Maghrabi, “Properties of bismuth glasses in relation to structure,” Phys. Chem. Glasses 10, 1–11 (1969).
21. G. Borodi, I. Bratu, and S. Simon, “Formation of layered structure on bismuth-borate glass surface,” Mater. Lett. 61(25), 4715–4717 (2007). [CrossRef]
22. G. E. Rachkovskaya and G. B. Zakharevich, “Properties, structure and application of low-melting lead-bismuth glasses,” Glass Ceram. 61(1/2), 9–12 (2004). [CrossRef]
23. P. Pascuta, L. Pop, S. Rada, M. Bosca, and E. Culea, “The local structure of bismuth borate glasses doped with europium ions evidenced by FT-IR spectroscopy,” J. Mater. Sci. 19, 424–428 (2008).
24. A. D. Sontakke, K. Biswas, A. K. Mandal, and K. Annapurna, “Concentration quenched luminescence and energy transfer analysis of Nd3+ ions doped Ba-Al-metaphosphate laser glasses,” Appl. Phys. B 101(1-2), 235–244 (2010). [CrossRef]
25. R. Adair, L. L. Chase, and S. A. Payne, “Nonlinear refractive index of optical crystals,” Phys. Rev. B 39(5), 3337–3350 (1989). [CrossRef]
26. G. Ghosh, “Sellmeier coefficients and dispersion of thermo-optic coefficients for some optical glasses,” Appl. Opt. 36(7), 1540–1546 (1997). [CrossRef] [PubMed]
27. B. R. Judd, “Optical absorption intensities of rare-earth ions,” Phys. Rev. 127(3), 750–761 (1962). [CrossRef]
28. G. S. Ofelt, “Intensities of crystal spectra of rare earth ions,” J. Chem. Phys. 37(3), 511–520 (1962). [CrossRef]
29. S. Tanabe, T. Ohyagi, N. Saga, and T. Hanada, “Compositional dependence of Judd-Ofelt parameters of Er3+ ions in alkali-metal borate glasses,” Phys. Rev. B 46(6), 3302–3310 (2002).
30. J. R. Ryan and R. Beach, “Optical absorption and stimulated emission of neodymium in yttrium lithium fluoride,” J. Opt. Soc. Am. B 9(10), 1883–1887 (1992). [CrossRef]
