A comprehensive investigation on Cr3+ → Ho3+ energy transfer has been comparatively carried out in Ho3+-doped, Cr3+-doped and Cr3+/Ho3+ co-doped fluorogermanate glasses. Steady and dynamic luminescence spectra of the glass samples are detected under excitation of static/microsecond-pulsed xenon lamps and an 808 nm laser diode, respectively. Through strong sensitization of Cr3+, the 2.0 μm emission of Ho3+: 5I7 → 5I8 can be efficiently achieved in an extremely extended excitation band of 325-830 nm. The energy transfer mechanisms involved are rationally discussed in detail for Cr3+/Ho3+ co-doping. On the basis of theoretical and experimental data, the energy transfer efficiency of Cr3+ → Ho3+ is appropriately calculated to be 26.2%. This novel Cr3+/Ho3+ codoping system could provide the experimental basis for obtaining 2.0 μm laser flexibly pumped by the xenon lamp or the 808 nm laser diode.
© 2014 Optical Society of America
Due to the promising applications in remote sensing, “eye-safe” laser surgery, etc [1,2], significant efforts have been devoted to the development of 2.0 μm solid state laser activated by Ho3+ ions. However, owing to the lack of well-matched absorption bands, Ho3+ can’t be pumped by the commercially available 808 and 980 nm laser diodes (LDs). Accordingly, rare-earth (RE) ions characterized by settled photo-excitation and emission bands in narrow bandwidth have been widely utilized as sensitizers to improve the optical pumping efficiency of LDs [3–5]. On the contrary to RE ions, the d-d transitions of Cr3+ are very sensitive to the change of local environment, and thus typically feature broad photo-excitation and emission bands . Meanwhile, it is noted that the broad near-infrared (NIR) emission of Cr3+-doped glass also overlaps well with the NIR absorption of Ho3+. Therefore, in principle, Ho3+ might be also efficiently activated by the direct energy transfer (ET) from Cr3+. Nonetheless, up to now, the ET mechanism of Cr3+ → Ho3+ is not clear at all.
As an excellent sensitizer, Cr3+ is well known for its broadband absorption in the ultraviolet (UV)-to-visible region that can efficiently couple to the conventional flashlamp pumping sources. However, in the Cr3+-doped glasses, non-radiative transitions usually dominate the relaxations of excited states, thus easily leading to the optical inactivity of amorphous hosts. In practice, the majority of previous researches about Cr3+ in amorphous materials are realized in silicate glass ceramics and glasses . Considering the similarity in the network structure with silicate glass, germanate glasses might likewise be available Cr3+-activated hosts, which provide extra advantages like low phonon energy, excellent draw ability and good infrared transmissivity for 2 μm laser.
In the present paper, the spectroscopic properties of Cr3+-doped and Cr3+/Ho3+ co-doped fluorogermanate glasses are studied in detail. The underlying Cr3+ → Ho3+ ET mechanisms are rationally demonstrated according to the static-to-dynamic luminescence spectra. The intense sensitization of Cr3+ to Ho3+ provides an efficient pumping approach with the flashlamp and commercial LDs, which would greatly boost the potential application of Cr3+/Ho3+ co-doped fluorogermanate glasses for fiber laser.
Glass samples with nominal molar composition of 50GeO2-20Al2O3-15LaF3-15LiF-xHoF3-yCr2O3 (x = 0, 1; y = 0, 0.1) were prepared by the conventional melt-quenching method. Analytical reagents of Al2O3 and LiF, and LaF3, GeO2, HoF3 and Cr2O3 (all 99.99%) were applied as raw materials. The well ground stoichiometric chemicals were put into a covered corundum crucible and melted at 1350 °C for 1 h with the controlled prepared condition (covered corundum crucibles and a mount of NH3HF3 were used). Subsequently, the melts were casted onto a stainless steel plate, and further annealed at 500 °C for 2 h. Then the prepared glass samples were cut into small pieces and polished for the optical measurements. To check the composition of the glass samples, energy dispersive spectra (EDS) were taken with a Philips XL-30FEG scanning electron microscope equipped with an EDAX DX4i EDS. The results reveal that a very limited loss (< 3 mol.%) due to evaporation of fluorine while melting, which might be due to the low glass melting temperature and duration with the controlled prepared condition.
Absorption spectra were measured on Perkin-Elmer Lambda 900 UV/VIS/NIR spectrophotometer. Excitation and emission spectra were performed on an Edinburgh FLS920 spectrophotometer equipped with a continuous wave 450 W Xe lamp as excitation source, a liquid nitrogen cooled R5509-72 photomultiplier (PMT) for NIR emission, and an InSb detector for mid-infrared (MIR) emission. Fluorescence decay curves were further recorded on FLS920 system with μF900 microsecond Xe lamp excitation source. Additionally, the photoemission spectra pumped by an 808 nm LD were carried out on a Triax 320 type spectrofluorometer (Jobin-Yvon Corp.) fitted with an InGaAs detector (830-1450nm) and an InAs detector (1800-2250nm).
3. Results and discussions
Figure 1 shows the absorption spectra of Cr3+-doped, Ho3+-doped and Cr3+/Ho3+ co-doped fluorogermanate glasses in the wavelength range of 250-2200 nm, respectively. For Ho3+ singly doped glass, a series of sharp absorption peaks are observed clearly because of the typical electronic transitions from the 5I8 ground state to the corresponding excited states of Ho3+. For Cr3+-doped sample (Fig. 1), three broad absorption bands centered at 360, 430 and 630 nm can be obtained definitely, where the UV absorption band around 360 nm should be caused by the strong charge transfer of Cr6+ , but the bands around 430 and 630 nm are readily attributed to the d-d electronic transitions from the 4A2 ground state to the 4T1 and 4T2 excited states of Cr3+ , respectively. Moreover, for Cr3+/Ho3+ codoped glass sample, the various absorption bands of Ho3+ and Cr3+ all can be detected obviously, and the sharp absorption bands of Ho3+ are well superimposed on the broad absorption bands of Cr3+, as depicted in Fig. 1. These results indicate that Ho3+ and Cr3+ ions have been well incorporated into the as-prepared fluorogermanate glasses. It should be noted that, the intense broadband absorption of Cr3+ nearly covers the near UV-to-NIR wavelength range likely of 300-850 nm both in Cr3+-doped and Cr3+/Ho3+ co-doped glass samples, which perhaps benefits the feasible pumping of a Xe lamp greatly if the ET process of Cr3+ → Ho3+ occurs efficiently. Additionally, the daylight photographs of all glass samples are shown in the inset of Fig. 1, all featuring good transparency even with Cr3+ or/and Ho3+ dopants.
According to the optical absorption transitions , the ligand parameter Dq/B here is calculated to be 2.02, which means that the 4T2 level lies below the 2E level on account of the Tanabe-Sugano diagram for Cr3+ in the octahedral symmetry. Related parameters in various Cr3+-doped hosts are enumerated and compared in Table 1. It is obvious that Cr3+ ions experience a typical strong crystal field (Dq/B>2.3) in the ruby, where the first laser in the world was obtained. As for the glass hosts except for lead borate glass, they usually exhibits a weak crystal field (Dq/B≤2.3), which might be ascribed to the loosely packed glass network.
To investigate the effect of Cr3+ ions on spectral properties of Ho3+ ions, NIR emission spectra are measured for all samples excited at 600 nm that is just located in the absorption band range of Cr3+ but outside that of Ho3+, as comparatively shown in Fig. 2(a).It can be seen that the Ho3+-doped glass sample does not yield any fluorescence, while the Cr3+-doped glass sample exhibit an intense broad emission band centered at 850 nm, readily ascribed to the Cr3+: 4T2 → 4A2 transition . Interestingly, the 850 nm emission band of Cr3+ in 700-1300 nm well overlaps with the absorption band of 5I8 → 5I5 at about 890 nm and that of 5I8 → 5I6 at 1150 nm of Ho3+, as pictured in Figs. 1 and 2(a), which suggests there might exist feasible ET from the excited Cr3+ to Ho3+ ions. Correspondingly, as Ho3+ codoped into Cr3+-doped sample, the peak intensity of 850 nm emission decreases greatly, and meanwhile another typical NIR emission of Ho3+ at 1200 nm obviously emerges as a shoulder due to the 5I6 → 5I8 electronic transition, as shown in Fig. 2(a). This observation directly confirms that the ET from Cr3+ to Ho3+ does take place in Cr3+/Ho3+ co-doped fluorogermanate glass. Meanwhile, among various speciations of chromium in glass hosts, tetrahedrally coordinated Cr4+ ions usually show an emission band centered around 1400 nm in the glass host, while the Cr3+ emission band is observed at around 900 nm in the silicate glass . Hence, it is deduced that chromium ions might exist mainly as Cr3+ in this fluorogermanate glass.
On the other hand, MIR emission spectra of Cr3+-doped, Ho3+-doped and Cr3+/Ho3+ co-doped glass samples are further detected under excitation of 600 nm, as shown in Fig. 2(b). As expected, the Cr3+-doped and Ho3+-doped sample both cannot give rise to the 2.0 μm fluorescence, but only the Cr3+/Ho3+ co-doped glass can efficiently yield the broad 2.0 μm emission with a full width at half maximum of 175 nm due to Ho3+: 5I7 → 5I8 electronic transition. This result additionally demonstrates that, upon excitation of Cr3+ at 600 nm, the ET process from Cr3+ to Ho3+ does take place effectively in the Cr3+/Ho3+ co-doped fluorogermanate glass.
Figure 2(c) depicts the excitation spectra of Ho3+-doped and Cr3+/Ho3+ co-doped glasses monitored at 1200 nm (Ho3+: 5I6 → 5I8). For Ho3+-doped sample, a series of excitation bands are gained typically due to the electronic transitions from the 5I8 ground state to the corresponding excited states of Ho3+, but no photo-absorption excitation band appears at about 600 nm. Whereas, for Cr3+/Ho3+ co-doped sample, the broad excitation bands due to 4A2 → 4T1 and 4A2 → 4T2 of Cr3+ are strongly detected except the typical excitation bands of Ho3+, which just provides another evident proof for the presence of ET from the excited Cr3+ ions to Ho3+ ions in Cr3+/Ho3+ co-doping system. Noted that, as shown in Fig. 2(c), the resulted excitation bands of Cr3+/Ho3+ co-doped glass nearly covers the complete near-UV-to-NIR region. It also means that the intriguing 2.0 μm fluorescence of Ho3+ can be efficiently obtained by using excitation source with wavelength output roughly in 325-830 nm (Fig. 1 and Fig. 2(c)).
To further confirm the existing Cr3+ → Ho3+ ET, and calculate the related ET efficiency, luminescence decay curves of 850 nm (Cr3+: 4T2 → 4A2) are measured in Cr3+-doped and Cr3+/Ho3+ co-doped samples under pulsed light excitation of 600 nm with μF900 microsecond Xe lamp, as pictured in Fig. 2(d). Herein, lifetimes of excited Cr3+: 4T2 were determined from the first e-folding time of emission intensities in the decay curves. It can be found that the lifetime of 4T2 level is 23.7 μs in Cr3+ singly doped glass, while it rapidly reduces to 17.5 μs with the introduction of Ho3+, readily caused by the efficient ET from Cr3+ to Ho3+ ions. Generally, the total ET efficiency (η) of Cr3+ → Ho3+ can be evaluated as following:
From the absorption (Fig. 1) and excitation (Fig. 2(c)) spectra of Cr3+/Ho3+ codoped glass, we can observe that the excitation intensity starts to decrease rapidly at wavelength beyond 750 nm, and even become weakest as wavelength exceeds 800 nm in the whole excitation region. To check the availability of this full near-UV-to-NIR excitation (325-830 nm) in Cr3+/Ho3+ co-doping system, a cheap commercial 808 nm semiconductor laser is utilized as excitation source. As comparatively shown in Fig. 3(a), no NIR emission is obtained in Ho3+-doped glass due to the lack of intra-4f absorption transition associated with an 808 nm LD. However, similar to the phenomena in Figs. 2(a)-2(b), the broad luminescence band of Cr3+: 4T2 → 4A2 is intensely recorded in 850-1400 nm for Cr3+-doped sample pumped by the 808 nm LD, but the intensity of Cr3+: 4T2 → 4A2 decreases clearly as Ho3+ codoped into the Cr3+-doped sample, which also accompanies with a superimposed fluorescence peak of Ho3+: 5I6 → 5I at 1200 nm. It is of great interest that, in comparison with Fig. 2(a), the emission peak shifts from 850 to 1000 nm. This apparent shift could be ascribed to the various excitation wavelengths for different powers , which favor different ions within the crystal-field distribution in the glass matrix, thereby leading to the distinct luminescence characteristics. Moreover, the different detectors of R5509-72 PMT and InGaAs photoconductor also make a varied contribution because of their distinct photoresponse performance. On the other hand, Fig. 3(b) shows the MIR spectra of Ho3+-doped, Cr3+-doped, and Cr3+/Ho3+ co-doped glasses. It is found that, under excitation of an 808 nm LD, Cr3+-doped and Ho3+-doped samples do not generate any apparent fluorescence, but the Cr3+/Ho3+ co-doped sample features a typical 2.0 μm emission with a full width at half maximum of 165 nm because of Ho3+: 5I7 → 5I8 transition. These results confidently reveal that, even for the rather weak excitation intensity in 325-830 nm, the available 2.0 μm emission can be easily obtained through the Cr3+ → Ho3+ ET in Cr3+/Ho3+ co-doped fluorogermanate glass.
According to the aforementioned results and Tanabe-Sugano diagram given for d3 electron configuration, the possible ET mechanism between Cr3+ and Ho3+ are proposed in Fig. 4. In glass hosts, the effects of spin-orbit coupling and electron-vibrational interaction would admix 4T2 and 2E levels together, just as shown in Fig. 4. Addition of Cr2O3 into Ho3+-doped glass results in effective absorption around 600 and 808 nm, readily corresponding to the broad absorption band from 4A2 to 4T2 level. After fast non-radiative relaxations to lower 4T2 levels, ET processes of 4T2 (Cr3+) → 5I5 (Ho3+) and 4T2 (Cr3+) → 5I6 (Ho3+) can efficiently take place except for that partial Cr3+ ions de-excite to the 4A2 ground state by emitting NIR photons. The ET brings about Ho3+ ions being excited from the 5I8 ground state to the 5I5 and 5I6 excited states and subsequently populating the 5I6 and 5I7 states with the assistance of multiphonon relaxations. Eventually, 1.2 and 2.0 μm emissions are efficiently induced from the 5I6 → 5I8 and 5I7 → 5I8 transitions of Ho3+, respectively.
Under excitation of a conventional Xe lamp or an 808 nm LD, we successfully obtain the intense MIR 2.0 μm luminescence of Ho3+ in Cr3+/Ho3+ codoped fluorogermanate glass. Through the efficient sensitization of Cr3+, the excitation wavelength range for getting 2.0 μm luminescence of Ho3+ is greatly enlarged for the whole near-UV-to-NIR area of about 325-830 nm. In terms of absorption, excitation and photoemission spectra, and time-resolved spectra, the resonant ET mechanisms of 4T2 (Cr3+) → 5I5 (Ho3+) and 4T2 (Cr3+) → 5I6 (Ho3+) are rationally demonstrated. Furthermore, the ET efficiency of Cr3+ → Ho3+ is appropriately calculated to be 26.2%. The further development of this Cr3+/Ho3+ system would provide the available 2.0 μm laser material efficiently excited by flashlamps or various commercial lasers.
This work is financially supported by NSFC (Grant Nos. 51125005 and 51302086), and the Fundamental Research Funds for the Central Universities, SCUT.
References and links
1. J. P. Zhang, W. J. Zhang, J. Yuan, Q. Qian, and Q. Y. Zhang, “Enhanced 2.0 μm emission and lowered upconversion emission in fluorogermanate glass-ceramic containing LaF3: Ho3+/Yb3+ by codoping Ce3+ ions,” J. Am. Ceram. Soc. 96(12), 3836–3841 (2013). [CrossRef]
2. Q. C. Sheng, X. L. Wang, and D. P. Chen, “Enhanced broadband 2.0 μm emission and energy transfer mechanism in Ho–Bi co-doped borophosphate glass,” J. Am. Ceram. Soc. 95(10), 3019–3021 (2012). [CrossRef]
3. W. J. Zhang, Q. Y. Zhang, Q. J. Chen, Q. Qian, Z. M. Yang, J. R. Qiu, P. Huang, and Y. S. Wang, “Enhanced 2.0 microm emission and gain coefficient of transparent glass ceramic containing BaF2: Ho3+,Tm3+ nanocrystals,” Opt. Express 17(23), 20952–20958 (2009). [CrossRef] [PubMed]
Z. H. Jiang and Q. Y. Zhang, “The structure of glass: a phase equilibrium diagram approach,” Prog. Mater. Sci. 61, 144–215 (2014). [CrossRef]
4. R. R. Xu, J. J. Pan, L. L. Hu, and J. J. Zhang, “2.0 μm emission properties and energy transfer processes of Yb/Ho codoped germanate glass,” J. Appl. Phys. 108(4), 043522 (2010). [CrossRef]
5. J. Yuan, S. X. Shen, D. D. Chen, Q. Qian, M. Y. Peng, and Q. Y. Zhang, “Efficient 2.0 μm emission in Nd3+/Ho3+ co-doped tungsten tellurite glasses for a diode-pump 2.0 μm laser,” J. Appl. Phys. 113(17), 173507 (2013). [CrossRef]
6. M. J. Weber, “Chromium—rare earth energy transfer in YAlO3,” J. Appl. Phys. 44(9), 4058–4064 (1973). [CrossRef]
7. B. T. Wu, S. F. Zhou, J. Ruan, Y. B. Qiao, D. P. Chen, C. S. Zhu, and J. R. Qiu, “Energy transfer between Cr3+ and Ni2+ in transparent silicate glass ceramics containing Cr3+/Ni2+ co-doped ZnAl2O4 nanocrystals,” Opt. Express 16(4), 2508–2513 (2008). [CrossRef] [PubMed]
8. F. Rasheed, K. P. O’Donnell, B. Henderson, and D. B. Hollis, “Disorder and the optical spectroscopy of Cr3+-doped glasses: I. Silicate glasses,” J. Phys. Condens. Matter 3(12), 1915–1930 (1991). [CrossRef]
9. H. Y. Fu, S. Cui, Q. Luo, X. S. Qiao, X. P. Fan, and X. H. Zhang, “Broadband downshifting luminescence of Cr3+/Yb3+-codoped ﬂuorosilicate glass,” J. Non-Cryst. Solids 358(9), 1217–1220 (2012). [CrossRef]
10. W. A. Pisarski, J. Pisarska, G. Dominiak-Dzik, and W. Ryba-Romanowski, “Transition metal (Cr3+) and rare earth (Eu3+, Dy3+) ions used as a spectroscopic probe in compositional-dependent lead borate glasses,” J. Alloy. Comp. 484(1-2), 45–49 (2009). [CrossRef]
11. W. M. Fairbank Jr, G. K. Klauminzer, and A. L. Schawlow, “Excited-state absorption in ruby, emerald, and MgO: Cr3+,” Phys. Rev. B 11(1), 60–76 (1975). [CrossRef]
13. L. J. Andrews, A. Lempicki, and B. C. McCollum, “Spectroscopy and photokinetics of chromium (III) in glass,” J. Chem. Phys. 74(10), 5526–5538 (1981). [CrossRef]
14. F. Rasheed, K. P. O’Donnell, B. Henderson, and D. B. Hollis, “Disorder and the optical spectroscopy of Cr3+-doped glasses. II. Glasses with high and low ligand fields,” J. Phys. Condens. Matter 3(21), 3825–3840 (1991). [CrossRef]
15. S. Tanabe and X. Feng, “Temperature variation of near-infrared emission from Cr4+ in aluminate glass for broadband telecommunication,” J. Appl. Phys. 77(6), 818 (2000).
16. D. L. Russell, K. Holliday, M. Grinberg, and D. Hollis, “Broadening of optical transitions in Cr3+-doped aluminosilicate glasses,” Phys. Rev. B 59(21), 13712–13718 (1999). [CrossRef]