Erbium doped oxyfluoride glass was synthesized from the molar composition 10.1% Na2CO3 −20.2% PbO −33.7% GeO2 −33.6% TeO2 −2.4% ErF3 by melt quenching technique. The Judd-Ofelt intensity parameters were estimated as Ω2 = 10.8 × 10−20, Ω4 = 1.17 × 10−20, and Ω6 = 4.32 × 10−20 cm2. Radiative transition probabilities and lifetimes were also calculated. Differential scanning calorimetry (DSC) was used for thermal analysis of the sample. Nanocrystals were induced in the glass by heat-treatment. Strong room temperature upconversion emissions were observed at 415, 540, 554 and 667 nm from Er3+ doped sample under 972 nm Ti-sapphire laser excitation. X-ray diffraction (XRD) measurements revealed the presence of NaErF4 crystallites 35 nm in the glassy matrix. The concentration of nanocrystals is found to be low in the middle of the sample and higher close to the surface. Time correlated single photon counting (TCSPC) was used to measure Er3+ lifetimes.
©2013 Optical Society of America
Oxide glasses are easy to prepare and possess good optical, thermal, chemical and mechanical characteristics, however, the phonon frequencies of oxide glasses are much higher when compared to those of halides. In general rare-earth dopants possess higher radiative transition probabilities in fluorides. If fluoride nanostructures are prepared in oxide glasses then one can take advantage of the bulk properties of oxide glasses and luminescence properties of fluorides [1–4]. Such glass ceramics will have potential applications for the development of lasers, optical amplifiers, and many other luminescent devices [5,6]. Transparent oxyfluoride glass ceramics containing LaF3 nanocrystals were investigated in the past [7–9]. PbO-GeO2-TeO2 glass was a low-loss infrared transmitting glass  and Pb2+ luminescence enhancement occurred when such a glass was co-doped with silver nanoparticles . The triply ionized erbium is a promising ion for frequency upconversion due to its numerous energy levels spanning the electromagnetic region from ultraviolet (UV) to infrared (IR) region. Accordingly upconversion emissions were observed in Er3+ -doped fluoroindate , oxide , borate , fluoride  and tellurium-germanium based oxyfluoride transparent glass ceramics . Upconversion efficiency depends on the guest host combination and hence numerous materials are investigated for such phenomena. Sodium-lead-germanate glasses possess good thermal stability . We have successfully fabricated sodium-lead-germano-tellurite based transparent glass ceramics containing NaErF4 nanocrystals and report the following information for the first time to our knowledge. In this paper we report on the structural changes in the absorption and emission spectra, and the observation of strong upconversion emission due to two and three photon absorption, under 972 nm laser excitation in the erbium doped glass ceramic. This material may be useful as a gain medium for efficient, low cost fiber optical amplifiers and lasers.
2. Experimental details
Rare-earth doped oxyfluoride glass samples were prepared from reagent grade compounds of GeO2, TeO2, PbO, Na2CO3 and ErF3. The glass was made by the melt quenching technique using a procedure described elsewhere . The chemicals were melted at 1150°C for 1.5 before allowing it to cool to room temperature in ambient air. The resulting glass was of good optical quality with slight pink color due to the presence of erbium. The glass was subsequently annealed below the glass transition temperature for an hour to relinquish it from internal stress. Later on it was polished for spectroscopic studies. Differential scanning calorimetry measurements were performed using NETZSCH model DSC 404C Pegasus® to measure the glass transition temperature (Tg) and the crystallization peak temperature (Tc). About 25 mg of glass sample was heated in a platinum pan at a constant heating rate 10°C/min under argon atmosphere.
Refractive index of the glass was measured using Brewster angle method. Absorption spectrum of the sample was recorded at room temperature using a CARY 3E spectrophotometer over a spectral range of 200-900 nm with a spectral band width of 1nm. The sample was excited with 488 nm Ar+ laser and 972 nm Ti-sapphire laser and the resulting luminescence spectra were recorded in the 200-900nm region with a medium resolution spectrometer (Acton Research Corporation SpectraPro 500). The fluorescence signals were detected with a Hamamatsu model R928 photomultiplier tube (PMT) whose output was acquired by a computer for further processing. For lifetime measurements the PMT output was amplified and given to Stanford Research Systems multichannel scaler (SR 430). Excited state lifetimes were derived by fitting the decay curves to a single exponential function. Some of the lifetimes were measured by using time correlated single photon counting technique (TCSPC).
Crystallites were induced by heat treating the glass above the glass transition temperature at 475 °C for 2 hours. There was a change in the color of the sample from dark pink to light pink after heat treatment. To identify the crystalline phase and determine the mean crystallite size X-ray diffraction (XRD) measurements were performed using PANalytical X'pert PRO MPD with X'celerator detector using CuKα radiation (λ = 0.154 nm) at 45 kV and 40 mA. The 2θ scan range was 20 to 60° with a step size 0. 02° and a scan speed 0.2° min−1.
3. Results and discussion
Figure 1 shows a typical DSC curve of erbium doped oxyfluoride glass. The onset temperature for the glass transition is 385.4°C, the mid-point temperature is 405.3°C and the ΔCp between the two states (after and before the glass transition) is 0.134 J/(g•K). Two exothermic peaks observed at 667.5 (Tc1) and 756.8 °C (Tc2) could be due to the crystallization of the sample. A sharp endothermic peak at 803°C is attributed to melting (Tm). The difference between glass transition and crystallization temperature (∆T) is 282.1 °C. Thus the synthesized glass sample is relatively stable against devitrification. Glass ceramics are subsequently obtained by heat-treating the precursor glass above the glass transition temperature at 475 °C for 2 h in an electrical furnace. The XRD patterns of as-made glass and the heat-treated sample are shown in Fig. 2. The as-made glass is completely amorphous with no diffraction peaks (curve a). The heat treated sample showed intense diffraction peaks (curve b) which can be attributed to NaErF4. These identifications are based on the standard ICDD data (file # 01-077-2041). The crystallite sizes were calculated using the Scherrer equation ,Fig. 2 indicate crystallization of the sample and the calculated size of the crystal at various peaks is in the range of 25-35 nm, however, optical microscope images also reveal microstructures.The absorption spectra of heat-treated and as-made glass are shown in Fig. 3. The ground state of Er3+ is 4I15/2. The absorption peaks are centered at 366 (4G9/2), 379 (4G11/2), 407 (2H9/2), 444 (4F3/2), 452 (4F5/2), 489 (4F7/2), 522 (2H11/2), 545 (4S3/2), 652 (4F9/2) and 799nm (4I9/2), where the designations in the parentheses are those of excited states. The measured refractive index of the sample at 632.8nm is 1.85. From the absorption spectral intensities, Judd-Ofelt intensity parameters were estimated as Ω2 = 10.8 × 10−20, Ω4 = 1.17 × 10−20, and Ω6 = 4.32 × 10−20 cm2 for the as-made glass. By using these intensity parameters and the matrix elements  transition probabilities are estimated for different transitions using the standard formulae . The transition probability of 4G11/2→4I15/2 (73.8 × 103 s−1) is nearly 3.3 × that of 2H11/2→4I15/2 (21.9 × 103 s−1) and both the transitions are hypersensitive, so their spectral features are very sensitive to the local environment. Also, when erbium compound crystallizes, its environment (site symmetry) is different from that in the oxide phase. Hence there is a change in the absorption spectral intensity of the hypersensitive transition at 379 nm (4G11/2) upon heat treatment (Fig. 3). This is due to small changes in the crystal field environment around Er3+ ions. Judd-Ofelt intensity parameters were estimated for heat-treated glass as Ω2 = 16.5 × 10−20, Ω4 = 3.2 × 10−20 and Ω6 = 13.28 × 10−20 cm2. These values are slightly higher than those of as-made glass because some of the ions are in crystalline environment, after heat-treatment. A partial energy level diagram was constructed (Fig. 4) by using the absorption peak energies. Under 375 or 388 nm pulsed diode excitation the sample revealed emissions at 406-416, 545-555, and 650-670 nm. Intrinsic lifetime of a given level is inversely related to the total relaxation rate, which is the sum of radiative and non-radiative relaxation rates. Intrinsic lifetimes of the excited states were measured by TCSPC technique for the Er3+-doped glass ceramic. It revealed three different values for 4S3/2 lifetime (Fig. 5), 17.3 (45%), 6.2 (40%) and 35.5 μs (14%); and two different values for 2H9/2 lifetime (Fig. 6), 3.3 (75%) and 24.3 ns(25%). The numbers in the parentheses indicate relative strengths. Such a large variation in lifetimes is due to different sites for Er3+ in the glass ceramic.
When the sample was pumped with 972 nm Ti-sapphire laser that resonantly excites 4I11/2 state (Fig. 4), upconversion signals were observed at 408-415 (2H9/2→4I15/2), 540 (2H11/2→4I15/2), 554 (4S3/2→4I15/2), and 650-670 nm (4F9/2→4I15/2), in both as-made and heat-treated glasses at room temperature (Fig. 7). However, the upconversion signal intensities are much higher in the heat-treated glass. A structural change is noticed at 540 nm (2H11/2→4I15/2) in heat-treated glass (curve b in Fig. 7) because this transition is hypersensitive; consequently its spectral features are sensitive to changes in crystal field. Green upconversion emission intensity enhanced 60% when the laser beam is launched closer to a surface than that observed when it passed through the middle of the sample. This indicates that the density of crystallites is higher near the surface of the sample. When the Er3+ ions are in a crystalline environment the crystal field symmetry is different. This modifies the emission spectra. To reveal these additional spectral features the upconversion spectrum was recorded by setting the spectrometer slit widths to 0.25 mm (Fig. 8). The 2H11/2→4I15/2 transition reveals more peaks because it is a hypersensitive transition; consequently its spectral features are influenced by the crystalline field environment. So the spectra in Fig. 7 look different from that in Fig. 8. The additional sharp peaks in Fig. 8 confirm the existence of crystallites. In order to understand the mechanisms responsible for the short wavelength emissions, we have recorded the power dependence of emission peaks at 415, 554 and 667 nm. The emissions at 554 and 667 nm exhibited nearly quadratic dependence on the pump laser beam (not shown). The pump power dependence of 4S3/2 and 4F9/2 emission suggests that a two photon upconversion process is responsible for the generation of green and red signals. This upconversion process may occur either by excited state absorption (ESA) or by energy transfer upconversion (ETU) interaction between two neighboring excited ions:Fig. 9), and 667 nm signals under 972 nm laser excitation. The measured lifetime of upconversion emissions from 4S3/2 and 4F9/2 are found to be respectively 42 and 140 μs. The decay time of 4S3/2 (42 μs) is higher than all the intrinsic lifetimes measured by TCSPC technique (Fig. 5). This suggests that ETU is occurring simultaneously. However the delay time expected for energy transfer upconversion could not be measured because a mechanical chopper was used to produce pulses. Our measurements suggest that both ESA and ETU are responsible for upconversion. Absorption of a 972 nm photon resonantly excites the 4I11/2 level. The ion in the 4I11/2 level can be excited to 4F7/2 by ESA with a second 972 nm pump photon (Fig. 4). The ions in the 4F7/2 level relax non-radiatively to 2H11/2 and 4S3/2 levels causing green emission at 540 and 554 nm respectively. Ion-ion interaction (or ETU) between two excited ions in the 4I11/2 state also populates 4F7/2, the latter relaxes in cascade to 4S3/2 and 4F9/2 causing them to emit green and red photons. The emission from 2H9/2 at 408-415 nm exhibited nearly cubic dependence with a slope of 2.7 (not shown) indicating that three 972 nm photons are involved in generating the 415 nm upconversion signal. Absorption of a third photon of 972 nm from 4S3/2 excites electrons to 2G7/2 level (Fig. 4). The excited ions relax very fast in cascade to the lower levels causing 2H9/2 to emit blue fluorescence (Fig. 10) at 415 nm (2H9/2→4I15/2). Such a three photon excitation process was not observed under 795 nm laser excitation in the as-made glass .
When the heat-treated glass was excited with 488 nm Ar+ laser it revealed emission at 392 (4G11/2→4I15/2) and 415 nm (not shown). The 488 nm laser resonantly excites 4F7/2 level which undergoes non-radiative relaxation to 4S3/2. An observation of the energy level diagram (Fig. 4) indicates that a two-photon excitation process of the type 4I15/2→4F7/2 and 4S3/2→4D7/2 occurs.
The excited level 4D7/2 relaxes non-radiatively to the lower levels in cascade causing 4G11/2 to emit at 392 (4G11/2→4I15/2), and 2H9/2 at 415 nm (2H9/2→4I15/2. Accordingly, the emission band centered at 392 nm exhibited quadratic dependence on pump power. Whenever rare-earth ions are incorporated into fluorine environment, non-radiative relaxation rate between any two states is much smaller than that in oxide glasses because of smaller cut-off phonon frequency. Hence, in our observations, there is an enhancement in the upconversion emission intensity in a heat-treated glass.
A new erbium doped transparent oxyfluoride glass ceramic was successfully fabricated. The designed glass has good thermal stability and is stable against devitrification. Judd-Ofelt intensity parameters and transition probabilities are estimated for the Er3+ transitions. XRD measurements revealed the presence of NaErF4 nanocrystals in the heat-treated glass. Both the precursor glass and the heat-treated glass ceramic exhibit good optical quality and high transparency. We have observed relatively more efficient upconversion emission in the Er3+ -doped glass ceramic than that in the as-made glass, at room temperature. Structural changes observed in the absorption and emission spectra confirm the presence of crystallites in the heat-treated glass.
This research was supported by NSF grant HRD 0927644 and U.S. Army Research Office Grant No. W911NF-08-1-0425.
References and links
1. T. Suzuki, S. Masaki, K. Mizuno, and Y. Ohishi, “Synthesis and luminescent properties of transparent oxyfluoride glass-ceramics containing Er3+:YLiF4 nanocrystals,” Appl. Phys. Express 3(7), 072601 (2010). [CrossRef]
2. L. A. Bueno, A. S. Gouveia-Neto, E. B. da Costa, Y. Messaddeq, and S. J. L. Ribeiro, “Structural and spectroscopic study of oxyfluoride glasses and glass-ceramics using europium ion as a structural probe,” J. Phys. Condens. Matter 20(14), 145201 (2008). [CrossRef]
3. M. Beggiora, I. M. Reaney, A. B. Seddon, D. Furniss, and S. A. Tikhomirova, “Phase evolution in oxy-fluoride glass ceramics,” J. Non-Cryst. Solids 326-327, 476–483 (2003). [CrossRef]
4. F. Auzel, D. Pecile, and D. Morin, “Rare earth doped vitroceramics: new, efficient, blue and green emitting materials for infrared up-conversion,” J. Electrochem. Soc. 122(1), 101–107 (1975). [CrossRef]
5. M. J. Digonnet, Rare-Earth Doped Fiber Lasers and Amplifiers (Dekker, 1993).
6. S. Tanabe, S. Yoshii, K. Hirao, and N. Soga, “Upconversion properties, multiphonon relaxation, and local environment of rare-earth ions in fluorophosphates glasses,” Phys. Rev. B 45(9), 4620–4625 (1992). [CrossRef]
7. M. J. Dejneka, “The luminescence and structure of novel transparent oxyfluoride glass-ceramics,” J. Non-Cryst. Solids 239(1-3), 149–155 (1998). [CrossRef]
8. G. Jones and S. N. Houde-Walter, “Upconversion mechanisms in an erbium-doped transparent glass ceramic,” J. Opt. Soc. Am. B 22(4), 825–830 (2005). [CrossRef]
9. S. Tanabe, H. Hayashi, T. Hanada, and N. Onodera, “Fluorescence properties of Er3+ ions in glass ceramics containing LaF3 nanocrystals,” Opt. Mater. 19(3), 343–349 (2002). [CrossRef]
10. S. Nielsen, W. D. Lawson, and A. F. Fray, “Some infrared transmitting glasses containing germanium dioxide,” Infrared Phys. 1(1), 21–26 (1961). [CrossRef]
11. C. B. de Araújo, L. R. P. Kassab, R. A. Kobayashi, L. P. Naranjo, and P. A. Santa Cruz, “Luminescence enhancement of Pb2+ ions in TeO2-PbO-GeO2 glasses containing silver nanostructures,” J. Appl. Phys. 99(12), 123522 (2006). [CrossRef]
12. G. S. Maciel, C. B. de Araújo, Y. Messaddeq, and M. A. Aegerter, “Frequency upconversion in Er3+-doped fluoroindate glasses pumped at 1.48 μm,” Phys. Rev. B 55(10), 6335–6342 (1997). [CrossRef]
13. B. R. Reddy and P. Venkateswarlu, “Infrared to visible energy upconversion in Er3+-doped oxide glass,” Appl. Phys. Lett. 64(11), 1327–1329 (1994). [CrossRef]
14. S. Tanabe, T. Ohyagi, N. Soga, and T. Hanada, “Compositional dependence of Judd-Ofelt parameters of Er3+ ions in alkali-metal borate glasses,” Phys. Rev. B Condens. Matter 46(6), 3305–3310 (1992). [CrossRef] [PubMed]
15. K. Hirao, S. Todoroki, and N. Soga, “CW room temperature upconversion lasing in Er3+-doped fluoride glass fiber,” J. Non-Cryst. Solids 143, 40–45 (1992). [CrossRef]
16. 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]
17. S. Yang, S. Xu, S. Dai, J. Yang, L. Hu, and Z. Jiang, “Thermal stability and optical transition of Er3+ in sodium-lead-germanate glasses,” J. Mater. Sci. 39(11), 3641–3646 (2004). [CrossRef]
18. S. Bommareddi, M. Dokhanian, and B. R. Reddy, “Energy upconversion in erbium doped sodium lead germano tellurite glass,” Phys. Status Solidi C 6(S1), S67–S70 (2009). [CrossRef]
19. X. S. Qiao, X. P. Fan, J. Wang, and M. Q. Wang, “Judd-Ofelt analysis and luminescence behavior of Er3+ ions in glass ceramics containing SrF2 nanocrystals,” J. Appl. Phys. 99(7), 074302 (2006). [CrossRef]
20. W. T. Carnall, H. Crosswhite, and H. M. Crosswhite, “Energy level structure and transition probabilities of the trivalent lanthanides in LaF3,” Argonne National Laboratory Rep. ANL-78–95 (Argonne National Laboratory, 1979).
21. M. J. Weber, “Probabilities for radiative and nonradiative decay of Er3+ in LaF3,” Phys. Rev. 157(2), 262–272 (1967). [CrossRef]