Transparent fluorogermanate glass-ceramics singly doped with 0.25 mol.%Tm3+ have been prepared through melt quenching and sequential thermal treatment. The structure and composition of the glass-ceramics have been characterized by means of X-ray diffraction, Raman spectroscopy and transmission electron microscopy. Efficient three-step sequential three-photon near-infrared (NIR) quantum cutting has been demonstrated, where an absorbed blue photon at 468 nm could be cut into three NIR photons at 1190, 1462 and 1800 nm, respectively. The underlying mechanism has been analyzed in terms of the static and dynamic spectra measurements. Based on the experimental data and theoretical consideration, an internal quantum yield has been estimated to be about 160%. Further development of such a triply-cutting material might explore a way to design high efficient photonic devices, which harvest more photons emitted than absorbed in the excitation process.
© 2013 Optical Society of America
Efficient luminescent materials with quantum yield higher than unit could be playing a significant role in the modern lighting and photoelectric display devices [1,2]. In theory, Dexter earlier interpreted that, for system doped with a couple of ions and a single activator ion, an absorbed vacuum ultraviolet (VUV) photon can be downconverted to two visible photons via energy transfer “sensitized luminescence” and “cascade” radiative transitions , respectively. The underlying physical mechanism is well known as photon cascade emission (PCE), quantum splitting (QS), quantum cutting (QC), or downconversion. Experimentally visible quantum cutting (VIS-QC) was first achieved in VUV-excited Pr3+-doped fluorides (YF3 and α-NaYF4) [4,5]. Especially the recent demonstration of VIS-QC in Gd3+-Eu3+ couple with an appreciable quantum yield exceeding 190% greatly stimulate the development of VIS-QC materials for the potential application in the ultra-efficient lighting and display devices [6,7]. In the excitation process, the feature of quantum yield greater than unit could be efficiently achieved with an intermediate level, where, through it, the absorbed energy in a couple can be stepwise transferred to activators from a sensitizer or in a single activator can be de-excited to the ground state by several “cascade” emissions [3–7]. For the potential application in the enhancement of solar cell efficiency, near-infrared (NIR) QC has been widely studied in various materials codoped by RE3+/Yb3+ (RE = Tb, Tm, Pr, Er, Nd and Ho) [8–16], where an ultraviolet-to-blue photon absorbed by RE3+ donor ion can be efficiently downconverted to two NIR photons re-emitted at about 1000 nm by Yb3+ acceptor ions. On the other hand, the NIR-QC system also has been realized in RE3+ (RE = Dy, Er, Ho and Tm) singly doped polycrystalline phosphors [17–20]. Particularly the three-step sequential three-photon NIR-QC of Tm3+ reported by Yu et al.  provides an interesting route to improve the photo-response of certain photoelectronic devices such as germanium solar cells. Unlike the polycrystalline phosphors, oxyfluoride glass ceramics (GCs) exhibits many advantages such as high transparency, low phonon energy, and stable physical and chemical performance [21–23]. Most importantly, due to the higher transparency for visible-to-NIR sunlight, rare earth ions doped GCs of NIR-QC is more preferable to be spectral downconverters in the front of solar cells [1,12,16]. In this paper, we fabricated a new species of fluorogermanate GC containing LaF3:Tm3+ nanocrystals. Upon excitation of a blue photon around 468 nm, a sequential three-step three-photon NIR-QC of Tm3+ can efficiently take place with an internal quantum yield of about 160%. The underlying mechanism involved has been investigated in details.
Oxyfluoride precursor glasses (PGs) were prepared by the nominal molar composition of 50GeO2-20Al2O3-15LaF3-15LiF-0.25TmF3-xYbF3 (x = 0, 1). The starting materials of high-purity GeO2, LaF3, TmF3 and YbF3 (99.99%), and analytical reagent grade Al2O3 and LiF were mixed with the specified stoichiometric ratio and further grounded completely. The batches of homogeneous mixtures (15 g) were melted in a covered corundum crucibles at 1350 °C for 1 h under air atmosphere. The melts were quenched on a preheated stainless steel plate and then annealed at 520 °C for 2 h. Here the obtained PG doped with Tm3+ ion and Tm3+/Yb3+ couple are donated as PG:Tm3+ and PG:Tm3+/Yb3+, respectively. To fabricate transparent GCs, the PG:Tm3+ sample was cut into several small pieces and went through the further thermal treatment at 570 °C for 4, 6 and 8 h, respectively. Accordingly, the prepared GC:Tm3+ samples were labeled as GC@5704h, GC@5706h and GC@5708h relative to the different thermal treatment time. Afterwards all the PG and GC samples were polished for optical measurements.
To identify the precipitated crystalline, X-ray diffractometer measurement of the crushed powders of the samples before and after heat-treatment was carried out by means of an X-ray powder diffractometer (XRD; Philips PW1830, Cu Kα). The microstructures of GC@5708h were characterized by transmission electron microscope (TEM; JEM-2010, Tokyo, Japan) combined with the selected area electron diffraction (SAED). Raman spectra of PG:Tm3+ and GC@5704h, GC@5706h and GC@5708h were detected by LabRAM Aramis micro Raman spectrometer (HORIBA, Jobin Yvon, France). Absorption spectra were measured on a Perkin-Elmer Lambada 900 UV/VIS/NIR spectrophotometer (Cary Model 5000, Varian Incorporation, Paloalto, CA) in the wavelength range of 250-2700 nm with the resolution of 1 nm. Steady and dynamic fluorescence spectra were recorded on a high-resolution Edinburgh FLSP920 spectrophotometer (Edinburgh Instruments, Livingston, UK) equipped with continuous wavelength 450 W xenon (Xe) lamp and microsecond Xe flash lamp as excitation sources, respectively. For FLSP920 system, a red-sensitive R928 photomultiplier tube (PMT. Hamamatsu, Japan) and a liquid nitrogen cooled R5509-72 PMT were applied for the 400-850 nm visible photon detection and the 800-1600 nm NIR photon detection, respectively. Additionally, upon pump of an 808 nm laser diode (LD) and a 980 nm LD, NIR to middle-IR (MIR) photoluminescence (PL) spectra measurements were performed on a computer-controlled Triaxial 320 spectro-flourimeter (Jobin-Yvon Inc., Paris, France) with PbSe-020 detector (operating wavelength region in 1000-4500 nm) assembled with a Standford SR510 lock-in amplifier. For comparison, all the optical measurements were operated under the identical conditions for each series of test.
3. Results and discussions
3.1 Structural characterization
Figure 1(a) shows the measured XRD patterns of the samples before and after heat-treatment. It can be seen that the XRD pattern of PG:Tm3+ exhibits typical amorphous with one broad diffuse humps, while, the XRD patterns of heat-treated samples exhibit several intense diffraction peaks, which can be readily indexed to the hexagonal LaF3 phase (JCPDS Card No. 076-0510) . This result indicates that LaF3 nanocrystals have been successfully precipitated among glass matrix after heat treatment. According to the Scherrer formula, the average crystal diameters are evaluated to be about 14, 19 and 25 nm for GC@5704h, GC@5706h and GC@5708h, respectively. On the other hand, TEM micrograph in Fig. 1(b) reveals the microstructure of GC with irregular 10-20 nm LaF3 crystallites distributing among the glass matrix. Moreover, SAED pattern in the inset of Fig. 1(b), as well as the high-resolution TEM (HRTEM) image in Fig. 1(c), further reveals the detailed crystal structure of a LaF3 nanocrystal in the TEM micrograph.
3.2 Raman spectra and UV-VIS optical absorption
Raman spectra of PG:Tm3+, GC@5704h, GC@5706h and GC@5708h are comparatively shown in Fig. 2.There exhibit two broad vibrational frequency bands in the region of 100-1250 cm−1 with one band centered at 545 cm−1 in the range of 200-700 cm−1, and another one centered at 875 cm−1 in the range of 700-1100 cm−1. By Gaussian fitting, the Raman spectra of PG sample can be subdivided into five sub-peak with the peak centered at 367, 510, 587, 786 and 874 cm−1 (Fig. 2), respectively. In actual, the low frequency band at 367 cm−1 is due to the bending vibration of Ge-O-Ge bridges, and the band at 510 and 587 cm−1 can be assigned to the symmetric stretching of Ge-O-Ge bonds. The high frequency band at 786 cm−1 is related to the Q2 stretching vibrations, and the band at 874 cm−1 is due to the Q3 stretching vibrations [25,26]. Interestingly, for the high vibrational frequency band in 700-1100 cm−1, as the heat treatment time increases, the frequency peak position shifts from 875 to 844 cm−1, which suggests that the maximum phonon energy of the GC sample becomes lower than the PG sample due to the precipitation of LaF3 nanocrystals. Moreover, for the low vibrational frequency band in 200-700 cm−1, there emerges one intense band located at 461 cm−1 and three weak shoulders at 226, 286 and 362 cm−1 after heat treatment. The frequency bands at 226, 286, 362 and 461 cm−1 all can be indexed to the typical vibrational frequency bands of LaF3 crystal lattice [27,28], which provides additional proof for the successful precipitation of LaF3 crystallites among glass matrix after the corresponding heat-treatment regime.
Figure 3 exhibits the absorption spectra of PG:Tm3+, GC@5704h, GC@5706h and GC@5708h in the wavelength region of 250-3000 nm. All the absorption spectra consist of several absorption bands centered at 356, 468, 685, 790, 1209 and 1670 nm , typically ascribed to the electronic transitions from the 3H6 ground state to the 1D2, 1G4, 3F2,3, 3H4, 3H5 and 3F4 excited states of Tm3+, respectively. This observation confirms that Tm3+ ions are well distributed into the as-prepared PG and GC samples. Because of the Rayleigh scatting caused by the LaF3 nanocrystals, the ultraviolet absorption edge of GC samples obviously shifts to longer wavelength with the prolongation of heat treatment time, which is probably induced by the growth of LaF3 nanocrystals . It is should be noticed that, following the heat-treatment at 570 °C for several hours, all the obtained GC samples still keep high transparence, as compared in the inset of Fig. 3.
3.3 Static-to-dynamic photoluminescence of Tm3+ and luminescence mechanisms
Shown in Fig. 4 are the visible-to-NIR PL spectra of GC@5708h and PG:Tm3+/Yb3+. Under excitation of 468 nm monochromatic light, not only the visible 650 and 792 nm emissions but also the NIR emissions at 1190, 1462 and 1610 nm can be intensely detected likely due to the electronic transitions of 1G4 → 3F4, 3H4 → 3H6, 1G4 → 3H4, 3H4 → 3F4 and 3F4 → 3H6 of Tm3+, respectively [10,19], as shown in Fig. 4(a). For a rigorous investigation into the NIR PL mechanisms, the additional PL spectra of GC@5708h are measured under excitation of 790 nm light and an 808 nm LD (Fig. 4(b)), and that of PG:Tm3+/Yb3+ are further obtained upon a 980 nm LD excitation (Fig. 4(c)). As GC@5708h excited at 790 nm, only the two emission bands around 1462 and 1610 nm can be detected because of 3H4 → 3F4 and 3F4 → 3H6 , respectively. Notably, for detection wavelength beyond 1600 nm, the spectral response reduces sharply for R5509-72 PMT but does inversely for the NIR-MIR PbSe-020 photoconductor. As a result, the typical 3F4 → 3H6 of Tm3+ at 1800 nm measured by R5509-72 (solid lines in Figs. 4(a-b)) performs the weak PL intensity and extreme distortion relative to that measured by PbSe-020 detector (dotted lines in Figs. 4(b-c)) . Upon excitation with an 808 nm LD, besides the 3F4 → 3H6 at 1800 nm of GC@5708h, the NIR PL bands around 1462 nm (3H4 → 3F4) can be also detected by PbSe-020 photoconductor (Fig. 4(b)). Whereas, only the NIR emission peaked at 1800 nm can be recorded due to 3F4 → 3H6 in PG:Tm3+/Yb3+ as Tm3+ excited by Stokes energy transfer from Yb3+ pumped by a 976 nm LD [19,30].
Figure 5 shows PL excitation (PLE) spectra of GC@5708h in the wavelength range of 320-850 nm. A series of PLE bands centered at 356, 468, 684 and 790 nm can be easily assigned to the electronic transitions from the 3H6 ground state to the 1D2, 1G4, 3F2,3 and 3H4 excited state of Tm3+ [10,19], respectively. However, it can be found that, by monitoring the PL wavelength of 650 and 1190 nm, only the 356 and 468 nm PLE bands can be observed due to 3H6 → 1D2 and 3H6 → 1G4 of Tm3+, respectively, while, by monitoring the emissions at 792, 1462 and 1610 nm, respectively, additional PLE bands can be detected at 684 nm (3H6 → 3F2,3) and 790 nm (3H6 → 3H4). These observations suggest that the NIR emission at 1190 nm only can be induced from the excited 1G4 state and its above excited state, but the 1462 and 1610 nm emissions can originate from the 3H4 state or it’s below state. Furthermore, decay curves are measured for the 650, 792, 1190, 1462 and 1610 nm emissions, as comparatively shown in Fig. 6. It is of great interest that the 650 and 1190 nm emissions exhibit the same decay cures with the almost identical lifetime of ~280 μs, the 792 and 1462 nm emissions also perform the similar decay curves with the average lifetime around 450 μs, while the 1610 nm emission just feature an independent decay curve with lifetime of 4.04 ms. These results clearly confirm that the 1190 nm NIR emission does originate from the same electronic level to the typical visible 650 nm PL from the 1G4 level of Tm3+, the 1462 nm NIR emission comes from the another same level to the typical visible 792 nm PL from the 3H4 level of Tm3+, and the NIR 1610 nm emission is just from the 3F4 level. Hence, it can be concluded that the simultaneous appearance of 1190, 1462 and 1610 nm NIR PL bands just originate from the “cascade” radiative transitions of Tm3+ with 3H4 and 3F4 acting as the corresponding intermediate levels .
Energy level diagram in Fig. 7 illustrates the three-step sequential three-photon NIR-QC of Tm3+. Under excitation of 468 nm into the 1G4 state of Tm3+, besides the radiation of 1G4 → 3F4 (650 nm) and 3H4 → 3H6 (792 nm), the energy in excited 1G4 state can be decayed to the 3H4 state by the first 1190 nm photon emitting, subsequently the populated 3H4 state can be further relaxed to the 3F4 state via the second 1462 nm photon radiating, and finally the excited 3F4 state gives out the third 1800 nm photon to populate the 3H6 ground state. Based on the energy gap law , the energy differences of 1G4 → 3F2,3 (~6000 cm−1), 3H4 → 3H5 (~4000 cm−1) and 3F4 → 3H6 (~5600 cm−1) are all determined to be large enough for maximum phonon energy ~461 cm−1 of LaF3 crystal lattice, enabling the radiative transitions from the excited 1G4, 3H4 and 3F4 state of Tm3+ ions dominate over the phonon-assisted nonradiative relaxation. In comparison, as the 3H4 level of Tm3+ is directly excited at 790 nm or by an 808 nm LD (Fig. 7(b)), the sequential two-step transitions of 3H4 → 3F4 and 3F4 → 3H6 do take place efficiently with 3F4 acting as an intermediate level, just emitting the second 1462 nm photon and the third 1800 nm photon for the aforementioned three-photon NIR-QC of Tm3+ ((Fig. 7(a)). Moreover, as the 3H5 state of Tm3+ excited by Stokes energy transfer from the 2F5/2 state of Yb3+ pumped with a 976 nm LD (Fig. 7(c)), only the third-step of 3F4 → 3H6 at 1800 nm occurs following the energy in 3H5 state nonradiatively depopulated to 3F4 of Tm3+.
For the sake of exploring more luminescence properties of the three-step sequential three-photon NIR-QC of Tm3+, visible and NIR time resolved emission spectra of GC@5708h were further measured upon 468 nm pulsed excitation with a μF900 Xe flash lamp. As shown in Fig. 8(a), in an initial delay time around 8 μs, only the radiation of 1G4 → 3F4 at 650 nm is observed clearly. Till the delay time about 10 μs, the additional 3H4 → 3H6 transition at ~790 nm emerges intensely, meanwhile, the PL intensity of 650 nm enhances obviously. Moreover, with the increase of delay time to 1000 μs, the PL intensity of 790 nm is shown to improve dramatically relative to that of 650 nm. These results reveal that, as delay time increases, the energy in excited 1G4 state could be stepwise decayed to the 3H4 state likely by the 1G4 → 3H4 transition at 1190 nm. On the other hand, as 1G4 excited directly, only the 1G4 → 3H4 at 1190 nm first emerges at the delay time about 8 μs, and then the relatively feeble 3H4 → 3F4 transition at 1462 nm presents in delay time region of 10-20 μs due to the sequential energy population of 3H4 state of Tm3+ (Fig. 8(b)). Furthermore, PL intensity of 1462 nm rises obviously relative to that of 1190 nm, which is resulted from the more energy population of 3H4 state following the efficient occurrence of 1G4 → 3H4 at longer delay time. These observations demonstrate that, upon excitation of 468 nm, the excited 1G4 state can de-excite stepwise to the 3H4 and 3F4 intermediate states, first emitting an 1190 nm photon and subsequently another 1462 nm photon. In principle, the populated 3F4 level can further generate the third 1800 nm photon by the third-step of 3F4 → 3H6 (Fig. 4). However, due to the feeble spectral response of R5509-72 PMT at wavelength more than 1600 nm, the 1800 nm emission cannot be observed dynamically [19,20].
In theory, luminescence quantum yield,, is defined as the ratio of the number of re-emitted photons to that of absorbed photons. On the basis of the absorption spectra (Fig. 3), Judd-Ofelt analysis was carried out to predict the radiative properties of Tm3+ in PG and GC samples. By applying least-squares fit method, the Judd-Ofelt intensity parameters (Ω2, Ω4, Ω6) and the radiative transition probability () of Tm3+-doped PG and GC samples can becalculated rationally [10,24], where the corresponding Judd-Ofelt intensity parameters and the root-mean-squared (RMS) deviation δrms are summarized in Table 1.By considering the sequential three-step energy transfer mechanisms from the 1G4 state of Tm3+, the total can be evaluated by the following formula [4,5,19]:4,5,17,24],1,11], the calculated total quantum yield on the basis of luminescence branching ratio exhibit similar values in Tm3+ doped PG, GC, and the NaYF4 polycrystalline phosphors . However, GC always feature higher luminescence intensity due to its lower phonon energy, the less defects and impurities in LaF3 crystal lattice, along with advantages such as high transparency, easy fabricate, and stable physical and chemical properties are more preferable for promising applications in downconverting layers [10,12,21–24].
In summary, efficient three-step sequential three-photon NIR QC has been demonstrated in transparent fluorogermanate GC containing LaF3:Tm3+ nanocrystals. Upon excitation of a blue photon at 468 nm, three NIR photons could be obtained at 1190, 1462 and 1800 nm, respectively. In terms of steady and dynamic luminescence spectra, the underlying mechanism has been investigated theoretically and experimentally. The calculated total quantum yield of the three-step sequential three-photon NIR-QC is about 160%. Such multiphoton NIR-QC of Tm3+ singly doping would provide a new approach to design the high efficient NIR luminescence materials.
This work is financially supported by NSFC (Grant Nos. 51125005 and U0934001), Department of Education of Guangdong Province (Grant No. cxzd1011), and the Fundamental Research Funds for the Central Universities.
References and links
1. Q. Y. Zhang and X. Y. Huang, “Recent progress in quantum cutting phosphors,” Prog. Mater. Sci. 55(5), 353–427 (2010). [CrossRef]
2. R. T. Wegh, E. V. D. Van Loef, and A. Meijerink, “Visible quantum cutting via downconversion in LiGdF4:Er3+,Tb3+ upon Er3+ 4f11 → 4f105d excitation,” J. Lumin. 90(3–4), 111–122 (2000). [CrossRef]
3. D. L. Dexter, “Possibility of luminescent quantum yields greater than unity,” Phys. Rev. 108(3), 630–633 (1957). [CrossRef]
4. J. L. Sommerdijk, A. Bril, and A. W. de Jager, “Two photon luminescence with ultraviolet excitation of trivalent praseodymium,” J. Lumin. 8(4), 341–343 (1974). [CrossRef]
5. W. W. Piper, J. A. de Luca, and F. S. Ham, “Cascade fluorescent decay in Pr3+-doped fluorides: Achievement of a quantum yield greater than unity for emission of visible light,” J. Lumin. 8(4), 344–348 (1974). [CrossRef]
7. B. Liu, Y. Chen, C. Shi, H. Tang, and Y. Tao, “Visible quantum cutting in BaF2:Gd,Eu via downconversion,” J. Lumin. 101(1–2), 155–159 (2003). [CrossRef]
8. P. Vergeer, T. J. H. Vlugt, M. H. F. Kox, M. I. Den Hertog, J. van der Eerden, and A. Meijerink, “Quantum cutting by cooperative energy transfer in YbxY1−xPO4:Tb3+,” Phys. Rev. B 71(1), 014119 (2005). [CrossRef]
9. Q. Y. Zhang, C. H. Yang, and Y. X. Pan, “Cooperative quantum cutting in one-dimensional (YbxGd1−x)Al3(BO3)4:Tb3+ nanorods,” Appl. Phys. Lett. 90(2), 021107 (2007). [CrossRef]
10. Q. Zhang, B. Zhu, Y. X. Zhuang, G. R. Chen, X. F. Liu, G. Zhang, J. R. Qiu, and D. P. Chen, “Quantum cutting in Tm3+/Yb3+-codoped lanthanum aluminum germanate glasses,” J. Am. Ceram. Soc. 93(3), 654–657 (2010). [CrossRef]
11. B. M. van der Ende, L. Aarts, and A. Meijerink, “Near-infrared quantum cutting for photovoltaics,” Adv. Mater. 21(30), 3073–3077 (2009). [CrossRef]
13. J. J. Eilers, D. Biner, J. T. V. Wijngaarden, K. Kramer, H. U. Gudel, and A. Meijerink, “Efficient visible to infrared quantum cutting through downconversion with the Er3+-Yb3+ couple in Cs3Y2Br9,” Appl. Phys. Lett. 96(15), 151106 (2010). [CrossRef]
14. J. M. Meijer, L. Aarts, B. M. van der Ende, T. J. J. Vlugt, and A. Meijerink, “Downconversion for solar cells in YF3:Nd3+,Yb3+,” Phys. Rev. B 81(3), 035107 (2010). [CrossRef]
15. D. C. Yu, S. Ye, X. Y. Huang, and Q. Y. Zhang, “Enhanced three-photon near-infrared quantum splitting in β-NaYF4:Ho3+ by codoping Yb3+,” AIP Adv. 2(2), 022124 (2012). [CrossRef]
17. D. C. Yu, S. Ye, M. Y. Peng, Q. Y. Zhang, J. R. Qiu, J. Wang, and L. Wondraczek, “Efficient near-infrared downconversion in GdVO4:Dy3+ phosphors for enhancing the photo-response of solar cells,” Sol. Energy Mater. Sol. Cells 95(7), 1590–1593 (2011). [CrossRef]
18. X. B. Chen, J. G. Wu, X. L. Xu, Y. Z. Zhang, N. Sawanobori, C. L. Zhang, Q. H. Pan, and G. J. Salamo, “Three-photon infrared quantum cutting from single species of rare-earth Er3+ ions in Er0.3Gd0.7VO4 crystalline,” Opt. Lett. 34(7), 887–889 (2009). [CrossRef] [PubMed]
19. D. C. Yu, S. Ye, M. Y. Peng, Q. Y. Zhang, and L. Wondraczek, “Sequential three-step three-photon near-infrared quantum splitting in β-NaYF4:Tm3+,” Appl. Phys. Lett. 100(19), 191911 (2012). [CrossRef]
20. D. C. Yu, J. P. Zhang, Q. J. Chen, W. J. Zhang, Z. M. Yang, and Q. Y. Zhang, “Three-photon near-infrared quantum cutting in Tm3+-doped transparent oxyfluoride glass ceramics,” Appl. Phys. Lett. 101(17), 171108 (2012). [CrossRef]
21. M. Mortier, “Between glass and crystal: Glass–ceramics, a new way for optical materials,” Philos. Mag. B 82(6), 745–753 (2002).
22. M. Mortier, A. Monteville, G. Patriarche, G. Mazé, and F. Auzel, “New progresses in transparent rare-earth doped glass-ceramics,” Opt. Mater. 16(1–2), 255–267 (2001). [CrossRef]
23. M. J. Dejneka, “The luminescence and structure of novel transparent oxyfluoride glass-ceramics,” J. Non-Cryst. Solids 239(1–3), 149–155 (1998). [CrossRef]
24. W. J. Zhang, D. C. Yu, J. P. Zhang, Q. Qian, S. H. Xu, Z. M. Yang, and Q. Y. Zhang, “Near-infrared quantum splitting in Ho3+:LaF3 nanocrystals embedded germanate glass ceramic,” Opt. Mater. Express 2(5), 636–643 (2012). [CrossRef]
25. D. D. Martino, L. F. Santos, A. C. Marques, and R. M. Almeida, “Vibrational spectra and structure of alkali germanate glasses,” J. Non-Cryst. Solids 293–295(1–2), 394–401 (2001).
26. L. Baia, T. Iliescu, S. Simon, and W. Kiefer, “Raman and IR spectroscopic studies of manganese doped GeO2–Bi2O3 glasses,” J. Mol. Struct. 599(1–3), 9–13 (2001). [CrossRef]
27. H. H. Caspers, R. A. Buchanan, and H. R. Marlin, “Lattice Vibrations of LaF3,” J. Chem. Phys. 41(1), 94 (1964). [CrossRef]
28. R. P. Bauman and S. P. S. Porto, “Lattice vibrations and structure of rare-earth fluorides,” Phys. Rev. 161(3), 842–847 (1967). [CrossRef]
29. Z. Pan, A. Ueda, M. Hays, R. Mu, and S. H. Morgan, “Studies of Er3+ doped germanate-oxyfluoride and tellurium-germanate-oxyfluoride transparent glass-ceramics,” J. Non-Cryst. Solids 352(8), 801–806 (2006). [CrossRef]
30. Q. Y. Zhang, T. Li, Z. H. Jiang, X. H. Ji, and S. Buddhudu, “980 nm laser-diode-excited intense blue upconversion in Tm3+/Yb3+-codoped gallate–bismuth–lead glasses,” Appl. Phys. Lett. 87(17), 171911 (2005). [CrossRef]
31. J. M. F. van Dijk and M. F. H. Schuurmans, “On the nonradiative and radiative decay rates and a modified exponential energy gap law for 4f-4f transitions in rare-earth ions,” J. Chem. Phys. 78(9), 5317–5325 (1983). [CrossRef]