Yb3+-doped alkali-barium-bismuth-tellurite glasses

Yb3+-doped alkali-barium-bismuth-tellurite glasses

Spectrochimica Acta Part A 65 (2006) 702–707 Optical parameters and upconversion fluorescence in Tm3+/Yb3+-doped alkali-barium-bismuth-tellurite glas...

176KB Sizes 2 Downloads 47 Views

Spectrochimica Acta Part A 65 (2006) 702–707

Optical parameters and upconversion fluorescence in Tm3+/Yb3+-doped alkali-barium-bismuth-tellurite glasses Hai Lin a,b,∗ , Ke Liu c , Lin Lin d , Yanyan Hou a , Dianlai Yang a , Tiecheng Ma a , Edwin Yun Bun Pun c , Qingda An a , Jiayou Yu a , Setsuhisa Tanabe b a

Faculty of Chemical Engineering and Materials, Dalian Institute of Light Industry, Dalian 116034, PR China Graduate School of Human and Environmental Studies, Kyoto University, Sakyo-ku, Kyoto 606-8501, Japan c Department of Electronic Engineering, City University of Hong Kong, Tat Chee Avenue, Kowloon, Hong Kong, PR China d Dalian Medical University, Dalian 116027, PR China b

Received 23 August 2005; received in revised form 17 December 2005; accepted 17 December 2005

Abstract Tm3+ /Yb3+ -doped alkali-barium-bismuth-tellurite (LKBBT) glasses have been fabricated and characterized. Density, refractive index, optical absorption, absorption and emission cross-sections of Yb3+ , Judd–Ofelt parameters and spontaneous transition probabilities of Tm3+ have been measured and calculated, respectively. Intense blue three-photon upconversion fluorescence and near-infrared two-photon upconversion fluorescence were investigated under the excitation of a 980 nm diode laser at room temperature. Wide infrared transmission window, high refractive index and strong blue three-photon upconversion emission of Tm3+ indicate that Tm3+ /Yb3+ co-doped LKBBT glasses are promising upconversion optical and laser materials. © 2005 Elsevier B.V. All rights reserved. Keywords: Alkali-barium-bismuth-tellurite glasses; Rare-earth ions; Optical parameters; Upconversion

1. Introduction Rare-earth doped low-phonon optical glasses are important materials for infrared solid lasers, optical broadband amplifiers, upconversion laser and visible display devices [1–11]. With the rapid increment of demand in high transmission capacity, Pr3+ and Tm3+ -doped tellurite glasses and tellurite-based fiber amplifiers have attracted significant interest in 1.3 and 1.4 ␮m broadband application [12–16]. In the mean time, tellurite glasses also catch much attention in visible upconversion because their maximum phonon energy is lower than those in borate, phosphate, silicate and germanate glasses [17–20]. Inasmuch as the emission quantum efficiency from a given level depends strongly on the phonon energy of the host medium, it can be expected that the non-radiative loss to the lattice will be small and the fluorescence quantum efficiency will be high in tellurite glasses [21,22].



Corresponding author. Tel.: +81 75 753 6817; fax: +86 411 8632 3097. E-mail address: [email protected] (H. Lin).

1386-1425/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.saa.2005.12.034

Due to the excellent application foreground in visible spectral range, the further attempts in optimizing tellurite glasses become more significative. Barium- and bismuth-containing glasses are considered to be the best candidates for optical materials since the barium and bismuth not only improves the glass stability, chemical resistance and optical refractive index, but also causes a further reduction in the phonon frequencies resulting in an obvious improvement of upconversion fluorescence efficiency [23,24]. In this research work, alkali-barium-bismuth-tellurite (LKBBT) glasses are designed as optical glass hosts and Tm3+ ions are chosen to be doping ions due to their seductive blue upconversion fluorescence. Because Tm3+ -doped glasses do not have any absorption band near 980 nm, commercial 980 nm diode laser cannot be used as a powerful pumping source. To overcome the shortcoming, co-doping with Yb3+ is adopted to achieve the strong energy absorption at 980 nm in this case. Density, refractive index and optical absorption in Tm3+ /Yb3+ -doped LKBBT glasses have been measured. Absorption and emission cross-sections of Yb3+ , Judd–Ofelt parameters and spontaneous transition probabilities of Tm3+ were calculated. Intense blue three-photon upconversion and near-infrared (NIR) two-photon upconversion fluorescence were

H. Lin et al. / Spectrochimica Acta Part A 65 (2006) 702–707

703

observed and recorded under the excitation of 980 nm laser at room temperature. With the development of low-temperature drawing, high-power femtosecond laser pulse and ion beam irradiation technologies, the low-melting point glasses become more realizable in fabricating optical fiber and waveguide devices. 2. Experiments Tm3+ and Yb3+ -doped LKBBT glasses were prepared from lithium carbonate (Li2 CO3 ), potassium carbonate (K2 CO3 ), barium carbonate (BaCO3 ), bismuth oxide (Bi2 O3 ), tellurium oxide (TeO2 ), thulium oxide (Tm2 O3 ) and ytterbium oxide (Yb2 O3 ) powders according to the following molar host composition: 7.5% Li2 O, 7.5% K2 O, 5% BaO, 5% Bi2 O3 and 75% TeO2 . 0.5 wt% Tm2 O3 and 1.0 wt% Yb2 O3 was added to dope the glasses. The well-mixed raw materials were first preheated at 200 ◦ C for 2 h, then melted at 840 ◦ C for 20 min using an electric furnace, and finally quenched to room temperature in a cold steel mold. The glasses were subsequently annealed at 300 ◦ C for 10 min, and then at 290 ◦ C for 2 h, after that slowly cooled down to room temperature. For optical measurements, the annealed glass samples were sliced and polished to pieces with two parallel sides. The densities of 0.5 wt% Tm2 O3 single-doped and 0.5 wt% Tm2 O3 and 1 wt% Yb2 O3 co-doped glass samples was measured to be 5.33 and 5.37 g/cm3 , respectively. Thus, the number density of Tm3+ in the former is 8.34 × 1019 /cm3 and the number densities of Tm3+ and Yb3+ ions in the latter are 8.40 × 1019 and 1.64 × 1020 cm−3 , respectively. The refractive indices of 0.5 wt% Tm2 O3 single-doped LKBBT glass were measured by Metricon 2010 prism coupler at two wavelengths, and the values are n1 = 2.0372 at 632.8 nm and n2 = 1.9809 at 1550 nm. The refractive indices of the sample at all other wavelengths can be calculated by Cauchy’s equation n = A + B/λ2 with A = 1.9696 and B = 27053 nm2 . Absorption spectrum was obtained with a Perkin-Elmer Lambda 900 UV–vis/NIR spectrophotometer. Infrared transmission spectrum was recorded by a Spectrum One-B FT-IR spectrometer. Visible luminescence spectrum was recorded with a Spex 500 M monochromator and detected by a photomultiplier. A 980 nm laser from a diode laser was used as pumping source. All measurements were performed at room temperature. 3. Results and discussion An effective broadband absorption around 977 nm has been observed in Tm2 O3 and Yb2 O3 co-doped LKBBT glasses. It is due to the transition of Yb3+ from the ground state 2 F7/2 to the excited state 2 F5/2 . The absorption cross-section (σ abs ) and the stimulated emission cross-section (σ em ) for Yb3+ can be calculated by the following equations [22–24]: σabs =

2.303log[I0 (λ)/I(λ)] 2.303E(λ) ln[I0 (λ)/I(λ)] = = N0 d N0 d N0 d (1)

Fig. 1. Absorption (curve 1) and emission (curve 2) cross-sections of Yb3+ in 0.5 wt% Tm2 O3 and 1 wt% Yb2 O3 co-doped LKBBT glasses.

Zl exp[(Ezl − hcλ−1 )/kT ] Zu    Zl hc 1 1 = σabs (λ) exp − Zu kT λp λ

σem (λ) = σabs (λ)

(2)

where I0 (λ) and I(λ) are the primary optical intensity and the optical intensity throughout the sample, N0 the Yb3+ ion concentration (ion/cm3 ), E(λ) the absorbance and d is the sample thickness. Zl and Zu are the partition functions of the lower and the upper states, respectively, according to the energy-level diagram of Yb3+ . Ezl , λp , h, c, k and T represent the zero-line energy, wavelength of peak absorption, Planck’s constant, velocity of light, Boltzmann’s constant and room temperature, respectively. Because coordination of Yb3+ is the same in various glass hosts, the values of Zl /Zu do not vary greatly and 0.93 can be used in the cross-section calculation [25], i.e.,    hc 1 1 σem (λ) = 0.93σabs (λ)exp − . (3) kT λp λ The calculated absorption and emission cross-sections of Yb3+ in 0.5 wt% Tm2 O3 and 1 wt% Yb2 O3 co-doped LKBBT glasses are shown in Fig. 1 and the maximum values are 1.78 × 10−20 cm2 at 977 nm and 1.69 × 10−20 cm2 at 978 nm, respectively. These values are higher than those in fluoride, fluorophosphate [25] and bismuth borate glasses [26], and close to those in multicomponential tellurite-based glasses [27]. Large absorption and emission cross-sections of Yb3+ are beneficial to obtaining enough pumping energy and transferring considerable energy to Tm3+ in co-doped LKBBT glass system. Fig. 2 shows the absorption spectrum of 0.5 wt% Tm2 O3 single-doped LKBBT glasses. The 1209, 793, 687, 660 and 464 nm absorption bands correspond to the transitions of Tm3+ from the ground state 3 H6 to the excited states 3 H5 , 3 H4 , 3 F3 , 3 F and 1 G , respectively. The radiative transition within the 2 4 4fn configuration of Tm3+ in LKBBT glasses can be analyzed by the Judd–Ofelt theory [28,29] based on the absorption spectrum. According to the Judd–Ofelt theory, the oscillator strength, Pcalc [(S,L)J;(S ,L )J ], of an electric-dipole absorption transition from the initial state |(S,L)J, to the final state |(S ,L )J , depends

704

H. Lin et al. / Spectrochimica Acta Part A 65 (2006) 702–707 Table 1 Measured and calculated oscillator strengths and Judd–Ofelt intensity parameters for Tm3+ in LKBBT glasses Absorption

Energy (cm−1 )

Pexp (10−6 )

Pcalc (10−6 )

Title

→ 3 H5 → 3 H4 → 3 F3 , 3 F2 → 1 G4 Ω2 (10−20 cm2 ) Ω4 (10−20 cm2 ) Ω6 (10−20 cm2 ) Root mean square deviation (10−7 )

8271 12610 14556 21552

2.770 4.601 4.459 1.680

2.297 (Ped ) 4.574 4.408 1.503

0.542 (Pmd )

3H 6 3H 6 3H 6 3H 6

Fig. 2. Absorption spectrum of 0.5 wt% Tm2 O3 -doped LKBBT glasses (sample thickness is 1.9 mm).

5.40 1.47 1.19 1.99

probability is given by A[(S, L)J; (S  , L )J  ] 2

n(n2 + 2) 64π4 e2 = 3hλ3 (2J + 1) 9  × Ωt |(S, L)J||U (t) ||(S  , L )J  |2 ,

on three Ωt parameters (t = 2,4,6) as: Pcalc [(S, L)J; (S  , L )J  ] 2

8π2 mc (n2 + 2) = 9n 3hλ(2J + 1)  2 Ωt |(S, L)J||U (t) ||(S  , L )J  | ×

(4)

t=2,4,6

where λ is the mean wavelength of the transition, m the mass of the electron, c the velocity, n the refractive index, h the Planck constant, Ωt are the Judd–Ofelt parameters. The term |(S,L)J||U(t) ||(S ,L )J |2 is the square of the matrix elements of the tensorial operator, which connects |(S,L)J to |(S ,L )J  states and is considered to be independent of host matrix. The experimental oscillator strengths Pexp of the transitions can be obtained by integrating absorbance for each band and the relationship is mc2 = 2 πe N



mc2 α(¯ν) d¯ν = 2 Pexp πe N  2.303mc2 = E(¯ν) d¯ν πe2 Nd



(6)

t=2,4,6

ln[I0 (¯ν)/I(¯ν)] d¯ν d (5)

where N is the number density of rare-earth ions, e the charge of the electron, ν¯ the wavenumber, E(¯ν) the absorbance, and d is the sample thickness. The Judd–Ofelt intensity parameters Ωt of Tm3+ were derived from the electric-dipole contributions of the experimental oscillator strengths. For 3 H6 → 3 H5 transition of Tm3+ , the magnetic-dipole contribution had been subtracted from the experimental oscillator strength before it was used. The squares of the matrix elements given in [30] were applied in the calculation. The measured oscillator strengths and Judd–Ofelt intensity parameters of Tm3+ are presented in Table 1. Some important radiative properties can be calculated by use of the values of Ωt [31,32]. The spontaneous transition

where the squares of the matrix elements |(S,L)J|| U(t) ||(S ,L )J |2 are cited from [33] and listed in Table 2. The fluorescence branching ratio of transitions from initial manifold |(S,L)J to lower levels |(S ,L )J  is given by A[(S, L)J; (S  , L )J  ] .    S  ,L ,J  A[(S, L)J; (S , L )J ]

β[(S, L)J; (S  , L )J  ] = 

(7)

The radiative lifetime of an emitting state is related to the total spontaneous emission probability for all transitions from this state by ⎧ ⎫−1 ⎨  ⎬ τrad = A[(S, L)J; (S  , L )J  ] (8) ⎩    ⎭ S ,L ,J

Table 2 shows the spontaneous transition probabilities, branching ratios, and lifetimes of the optical transitions in Tm3+ doped LKBBT glasses. The branching ratios for the transitions 1 G → 3 H and 3 H → 3 H are 52.6 and 90.5%, respectively. 4 4 6 6 Thus, it is reasonable to obtain efficient blue and NIR emissions in the glasses under suitable excitation condition. LKBBT glasses exhibit a wide infrared (IR) transmission window. The infrared transmittance spectrum of 0.5 wt% Tm2 O3 -doped LKBBT glass is presented in Fig. 3. The IR edge (transmittance of 5%) of the sample is at 1583 cm−1 (6.32 ␮m) and the wide window indicates that vibrational phonon energy is small compared to those in other traditional oxide glasses [23]. The reduction of phonon energy provides opportunities towards the realization of more efficient upconversion and IR lasers and optical fiber amplifier. Intense blue upconversion fluorescence was observed by naked eye when Tm3+ /Yb3+ co-doped LKBBT glasses was excited by 980 nm diode laser. The upconversion fluorescence spectrum of Tm3+ /Yb3+ co-doped LKBBT glasses is shown in

H. Lin et al. / Spectrochimica Acta Part A 65 (2006) 702–707

705

Table 2 Predicted spontaneous-radiative transition rates and lifetimes of Tm3+ in LKBBT glasses Transition

Energy (cm−1 )

U(2)2

U(4)2

U(6)2

A (s−1 )

τ rad (ms)

β

3F 4

→ 3 H6

5594

0.5589

0.7462

0.2574

477.8

2.093

1

3H 5 3F 4

→ 3 H6

8279 2685

0.1074 0.0915

0.2313 0.1396

0.6382 0.0397

490.5 7.2

2.009

0.986 0.014

3H 4 3F 4 3H 5

→ 3 H6

12583 6989 4304

0.2187 0.1215 0.0152

0.0944 0.1329 0.4669

0.5758 0.2258 0.0153

2625 238.2 38.5

0.345

0.905 0.082 0.013

3F → 3H 3 6 3F 4 3H 5 3H 4

14481 8887 6202 1898

0 0.0031 0.6286 0.0816

0.3163 0.0011 0.3468 0.3545

0.8409 0.1654 0 0.2988

3854 122.9 742.9 7.1

0.212

0.815 0.026 0.157 0.002

3F → 3H 2 6 3F 4 3H 5 3H 4 3F 3

15039 9445 6760 2456 558

0 0.2849 0 0.3120 0.0040

0 0.0548 0.2916 0.1782 0.0738

0.2591 0.0448 0.5878 0.0773 0

1284 1611 390.5 33.2 0.03

0.301

0.387 0.485 0.118 0.010 ∼0

1G 4 3F 4 3H 5 3H 4 3F 3 3F 2

→ 3 H6

20986 15392 12707 8403 6505 5947

0.0452 0.0042 0.0704 0.1511 0.0100 0.0050

0.0694 0.0186 0.0055 0.0046 0.0698 0.0695

0.0122 0.0642 0.5176 0.3750 0.2915 0.0413

2509 314.8 1358 474.4 86.1 23.2

0.210

0.526 0.066 0.285 0.100 0.018 0.005

1D 2 3F 4 3H 5 3H 4 3F 3 3F 2 1G 4

→3 H6

27808 22214 19529 15225 13327 12768 6822

0 0.5792 0 0.1147 0.1637 0.0639 0.1926

0.3144 0.0968 0.0017 0.0138 0.0714 0.3093 0.1666

0.0916 0.0194 0.0164 0.2307 0 0 0.0006

1.084 × 104 2.801 × 104 120.4 2199 1557 1099 254.0

0.023

0.246 0.635 0.003 0.050 0.035 0.025 0.006

Fig. 4. The 477 nm blue, 652 nm red and 809 nm NIR bands can be assigned to 1 G4 → 3 H6 , 1 G4 → 3 F4 and 3 H4 → 3 H6 transitions, respectively. The ratio of radiation energies for the three emission bands [34] can be replaced by the ratio of emission intensities integrated along wavenumber (¯ν), i.e.,  ν2  ν  ν 2 2 ρblue (¯ν) d¯ν : ρred (¯ν) d¯ν : ρNIR (¯ν) d¯ν ν1

ν1

 =

ν2

Iblue (¯ν) d¯ν : ν1



ν1

ν2

ν1

Ired (¯ν) d¯ν :



ν2

ν1

INIR (¯ν) d¯ν,

(9)

Fig. 3. IR transmittance spectrum of 0.5 wt% Tm2 O3 -doped LKBBT glasses (sample thickness is 1.4 mm).

where ρ(¯ν) is energy density. The ratio among the blue, red and NIR bands is obtained to be 72.6:1.8:25.6. The emission energy of 1 G4 → 3 H6 transition occupies 97.6% of the total radiation energy in visible spectral range (400–700 nm), so the upconversion fluorescent color is mainly blue. In an upconversion mechanism, the upconversion emission intensity IUP will be proportional to the mth power of the IR excitation intensity IIR ; i.e., m IUP ∝ IIR ,

(10)

Fig. 4. Upconversion spectrum of 0.5 wt% Tm2 O3 and 1.0 wt% Yb2 O3 codoped LKBBT glasses under the excitation of 980 nm laser.

706

H. Lin et al. / Spectrochimica Acta Part A 65 (2006) 702–707

Actually, the energy gap between Yb(2 F5/2 ) and Yb(2 F7/2 ) is in a wide range and the low energy edge is near the high energy edge of the energy gap between Tm(3 H6 ) and Tm(3 H5 ). Thus, the energy transfer from Yb3+ to Tm3+ becomes efficient. When Tm3+ is excited into 3 H5 level, the non-radiative relaxation will begin by the cooperation of the phonons in LKBBT glasses because the energy gap between Tm(3 H5 ) and Tm(3 F4 ) is only 2685 cm−1 . (2) The same Tm3+ ion is further excited into the 3 F2 level by the second excited Yb3+ ion and relaxes into the lower metastable state 3 H4 : Fig. 5. Dependence of upconversion emission intensity on excitation power under 980nm laser excitation for 0.5 wt% Tm2 O3 and 1.0 wt% Yb2 O3 co-doped LKBBT glasses.

where m is the number of IR photons absorbed per visible photon emitted. A plot of log IUP versus log IIR yields a straight line with slope m. Fig. 5 shows such a plot for the 482 nm blue and 808 nm NIR emissions, and the values of m obtained are 2.45 and 1.67, respectively. The results reveal that three and two photons contribute to the upconversion processes of blue and NIR emission bands, respectively. In order to understand the upconversion mechanism under 980 nm laser radiation, the energy level diagram for Tm3+ /Yb3+ co-doped LKBBT glasses are shown in Fig. 6. When the glass sample is pumped with 980 nm diode laser, Tm3+ cannot absorb 980 nm photons directly due to the lack of the matched energy level, but Yb3+ ions can absorb the near-infrared radiation efficiently and transfer their excitation energies to Tm3+ ions. The Tm3+ ions can be excited into the 1 G4 blue emitting level through three successive energy transfer (ET) steps: (1) A Tm3+ ion is excited into the 3 H5 level by energy transfer from the first excited Yb3+ ions, and this Tm3+ ion then relaxes into its 3 F4 level: Yb(2 F5/2 ) + Tm(3 H6 ) → Yb(2 F7/2 ) + Tm(3 H5 ) → Yb(2 F7/2 ) + Tm(3 F4 ).

Yb(2 F5/2 ) + Tm(3 F4 ) → Yb(2 F7/2 ) + Tm(3 F2 ) → Yb(2 F7/2 ) + Tm(3 H4 ). As the case above, energy transfer in this step from Yb3+ to Tm3+ is considered to be efficient, and the relaxation process from Tm(3 F2 ) to Tm(3 H4 ) will be effective due to the narrow gap between them (2456 cm−1 ). (3) The third excited Yb3+ ion transfers its excitation energy to this Tm3+ ion and excites it from the 3 H4 to the 1 G4 state: Yb(2 F5/2 ) + Tm(3 H4 ) → Yb(2 F7/2 ) + Tm(1 G4 ). This energy transfer process from Yb3+ to Tm3+ in the final excitation process will be also efficient owing to the well matching of the energy gaps. In addition, following the ET excitation processes, the excited state absorption (ESA) from the 3 F4 and the 3 H4 states of Tm3+ are possible to occur in the upconversion excitation, respectively. The measured lifetimes for the two levels have proven to be much longer (∼450 ␮s for 3 F4 and ∼350 ␮s for 3 H4 ) in tellurite glass sample due to its lower phonon energies [35], indicating the nonradiative transition rates of the two levels are limited which is benefit to achieve ESA processes. Anyway, whether in visible or NIR upconversion emission of Tm3+ , Yb3+ has been shown as a dominating role in energy absorption and transfer. Because the three-photon upconversion fluorescence is more sensitive to pumping energy, the intensity of blue band exceeds that of NIR band when the excitation power increase to 30 mW and more (Fig. 5). Intense blue emission has been observed when the excitation power rise to the value above 50 mW. It proven that commercial 980 nm diode laser is a powerful pumping source for upconversion fluorescence in Tm3+ /Yb3+ co-doped LKBBT glasses. 4. Conclusions

Fig. 6. Energy level diagrams for Tm3+ and Yb3+ in LKBBT glasses. Threeand two-photon upconversion mechanisms under 980 nm laser excitation are indicated.

co-doped alkali-barium-bismus-tellurite Tm3+ /Yb3+ (LKBBT) glasses have been fabricated and characterized. Large absorption and emission cross-sections of Yb3+ are obtained and they are beneficial to achieve the efficient energy from Yb3+ to Tm3+ . The Ω intensity parameters, the radiative rates, the branching ratios and the fluorescence lifetimes of Tm3+ were calculated based upon the Judd–Ofelt theory. Under 980 nm laser excitation, intense 477 nm blue upconversion and

H. Lin et al. / Spectrochimica Acta Part A 65 (2006) 702–707

809 nm near-infrared upconversion fluorescence are proven to be due to three- and two-photon absorption processes, respectively. Wide optical transmission window, high refractive index and strong blue three-photon upconversion fluorescence indicate that rare-earth doped LKBBT glasses are promising upconversion optical and laser materials. Acknowledgements We would like to acknowledge the financial supports from Science and Technology Foundation of Liaoning Province (20041067), Science and Technology Foundation of Dalian City (2004166) and Scientific Research Foundation for the Returned Overseas Chinese Scholars of State Education Ministry. References [1] A. Mori, K. Kobayashi, M. Yamada, T. Kanamori, K. Oikawa, Y. Nishida, Y. Ohishi, Electron. Lett. 34 (1998) 887–888. [2] R. Balda, A. Oleaga, J. Fernandez, J.M. Fdez-Navarro, Opt. Mater. 24 (2003) 83–90. [3] R. Paschotta, P.R. Barber, A.C. Tropper, D.C. Hanna, J. Opt. Soc. Am. B 14 (1997) 1213–1218. [4] 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. [5] J.H. Song, J. Heo, S.H. Park, J. Appl. Phys. 93 (2003) 9441–9445. [6] J.L. Doualan, S. Girard, H. Haquin, J.L. Adam, J. Montagne, Opt. Mater. 24 (2003) 563–574. [7] J.S. Wang, D.P. Machewirth, F. Wu, E. Snitzer, E.M. Vogel, Opt. Lett. 19 (1994) 1448–1449. [8] B. Pedersen, W.J. Miniscalco, R.S. Quimby, IEEE Photon. Technol. Lett. 4 (1992) 446–448. [9] F. Roy, F. Leplingard, L. Lorcy, A. Le Sauze, P. Baniel, D. Bayart, Electron. Lett. 37 (2001) 943–945. [10] T. Murata, H. Takebe, K. Morinaga, J. Am. Ceram. Soc. 81 (1998) 249–251. [11] S.G. Grubb, K.W. Bennett, R.S. Cannon, W.F. Humer, Electron. Lett. 28 (1992) 1243–1244.

707

[12] S. Tanabe, T. Kouda, T. Hanada, J. Non-Cryst. Solids 274 (2000) 55–61. [13] S.Q. Man, E.Y.B. Pun, P.S. Chung, Opt. Commun. 168 (1999) 369–373. [14] E.R. Taylor, L.N. Ng, N.P. Sessions, H. Buerger, J. Appl. Phys. 92 (2002) 112–117. [15] H. Yamauchi, G.S. Murugan, Y. Ohishi, J. Appl. Phys. 97 (2005) 043505. [16] S. Shen, A. Jha, S.J. Wilson, E. Zhang, J. Non-Cryst. Solids 326–327 (2003) 510–514. [17] G. Poirier, F.C. Cassanjes, C.B. de Araujo, V.A. Jerez, S.J.L. Ribeiro, Y. Messaddeq, M. Poulain, J. Appl. Phys. 93 (2003) 3259–3263. [18] P.V. dos Santos, M.V.D. Vermelho, E.A. Gouveia, M.T. de Araujo, A.S. Gouveria-Neto, F.C. Cassanjes, S.J.L. Ribeiro, Y. Messaddeq, J. Alloy Compd. 344 (2002) 304–307. [19] H. Lin, S. Jiang, J. Wu, F. Song, N. Peyghambarian, E.Y.B. Pun, J. Phys. D: Appl. Phys. 36 (2003) 812–817. [20] H.T. Amorim, M.V.D. Vermelho, A.S. Gouveia-Neto, F.C. Cassanjes, S.J.L. Ribeiro, Y. Messaddeq, J. Solid State Chem. 171 (2003) 278–281. [21] N. Rakov, G.S. Maciel, M.L. Sundheimer, L. de, S. Menezes, A.S.L. Gomes, Y. Messaddeq, F.C. Cassanjes, G. Poirier, S.J.L. Ribeiro, J. Appl. Phys. 92 (2002) 6337–6339. [22] G. Ozen, A. Aydinli, S. Cenk, A. Sennaroglu, J. Lumin. 101 (2003) 293–306. [23] K.J. Plucinski, W. Gruhn, J. Wasylak, J. Ebothe, D. Dorosz, J. Kucharski, I.V. Kityk, Opt. Mater. 22 (2003) 13–19. [24] H. Sun, S. Xu, S. Dai, J. Zhang, L. Hu, Z. Jiang, Solid State Commun. 132 (2004) 193–196. [25] H. Yin, P. Deng, J. Zhang, F. Gan, Mater. Lett. 30 (1997) 29–33. [26] Y. Chen, Y. Huang, Z. Luo, Chem. Phys. Lett. 382 (2003) 481–488. [27] G. Wang, J. Zhang, S. Xu, S. Dai, L. Hu, Z. Jiang, J. Lumin. 109 (2004) 1–7. [28] B.R. Judd, Phys. Rev. 127 (1962) 750–761. [29] G.S. Ofelt, J. Chem. Phys. 37 (1962) 511–520. [30] W.T. Carnall, P.R. Fields, K. Rajnak, J. Chem. Phys. 49 (1968) 4424– 4442. [31] C. Guery, J.L. Adam, J. Lucas, J. Lumin. 42 (1988) 181–189. [32] S. Tanabe, K. Tamai, K. Hirao, N. Soga, Phys. Rev. B 47 (1993) 2507– 2514. [33] N. Spector, R. Reisfeld, L. Boehm, Chem. Phys. Lett. 49 (1977) 49–53. [34] C.C. Davis, Lasers and Electro-Optics: Fundamentals and Engineering, Cambridge University Press, Cambridge, 1996, pp. 7–10. [35] S. Tanabe, C. R. Chimie 5 (2002) 815–824.