Yb2+ and Yb3+ co-activated luminescent material that can cut one photon in ultraviolet and visible region into multi NIR photons could be used as a downconversion luminescent convertor in front of crystalline silicon solar cell panels to reduce thermalization loss of the solar cell. After a direct excitation of Yb2+ ions, an intense Yb3+ luminescence is observed based on a cooperative energy transfer process. The energy transfer process is discussed according to the dependence of Yb3+ luminescence intensity on the excitation power and the ambient temperature.
©2010 Optical Society of America
The modification of solar spectra to improve the efficiency of solar cells have been given persistent fervency since the energy transfer efficiency in the commercial solar cells is just about 15%, which means that most of sunlight energy (over 70%) is lost [1-3]. The main energy loss in the conversion of solar energy to electricity is related to the so-called spectral mismatch: low energy photons are not absorbed by solar cells while high energy photons are not used efficiently because the excess energy is lost as heat during the fast thermalization of the ‘hot’ charge carriers . Current studies of solar spectra modification are mostly focused on up-conversion (UC) and down-conversion (DC) materials. However, UC is a nonlinear process which means sufficiently high-excitation density is needed to achieve high conversion efficiency while DC (a linear process) does not have such demand . Mostly studied DC materials are based on the energy transfer from RE ions to Yb3+ ions [6-12]. The chosen of Yb3+ are based on the following consideration: Yb3+ ion has a single excited state (denoted by the term symbol 2F5/2) approximately 10000 cm−1 above the 2F7/2 ground state. The absence of other energy levels allows Yb3+ to exclusively ‘pick up’ energy packages of 10000 cm−1 from other co-doped lanthanide ions and emits photons at ~1000 nm, which is just above the band edge of crystalline Si and where silicon solar cells exhibit their greatest spectral response .
However, the present studied DC materials are still far from practical application, because the transitions between 4f levels of trivalent lanthanide ions are forbidden transitions. The oscillator strengths of 4f–4f transitions are typically on the order of 10−6; dipole allowed transitions, by comparison, may have oscillator strength of up to unity. Thus, the absorption strength of lanthanide 4f–4f transitions is very weak . To solve the absorption problem a new ion is required to work as a donor which absorbs efficiently in the UV and visible part and transfers the energy to Yb3+ ions. In this paper, we demonstrate the efficient broadband energy transfer from Yb2+ to Yb3+ ions in CaAl2O4 phosphors for the first time. The photoluminescence (PL) and photoluminescence excitation (PLE) spectra show that energy transfer happened between Yb2+ and Yb3+ ions. Through the study of the dependence of luminescence intensity on the excitation power and the ambient temperature, the mechanism of energy transfer involving a two or three photon cutting process is discussed.
CaAl2O4: Yb2+, Yb3+ samples were prepared using high-temperature solid-state reaction method. High purity (99.99%) CaCO3, Al2O3, and Yb2O3 were used as raw materials. Reactant samples after mixed thoroughly in an agate mortar were presintered in air at 900 °C for 6 h. The samples were ground again and sintered under two different ambiences, reducing atmosphere (8 vol% H2 + 92 vol% N2) and air, at 1300 °C for 2 hours. Samples doped with 0 and 4 mol% Yb3+ ions were labeled as sample A0 (B0) and A (B); respectively (A/A0 stands for the samples sintered under reducing atmosphere while B/B0 represents the samples sintered in air). The crystalline phase and crystallinity of the synthesized phosphors were investigated by powder XRD using Rigaku D/MAX-RA diffractometer. The absorption spectra were obtained with a spectrophotometer (Hitachi-4100). The PL and PLE spectra in the infrared region were recorded with a FLS920 fluorescence spectrophotometer.
3. Results and discussion
The XRD patterns of sample A and A0 are shown in Fig. 1 . From the XRD pattern, all the diffraction peaks except for two small ones (shown in Fig. 1 with a “*” in sample A), can be indexed to the pure monoclinic phase (space group: P21/n) of CaAl2O4. Calculated lattice parameters are as follows: a = 8.702 Å, b = 8.092 Å, c = 15.18 Å and Z = 12. The XRD pattern matched perfectly with that reported for CaAl2O4 (JCPDS (23-1036)) . The XRD patterns of sample B and B0 are the same as that of sample A and A0, which can be indexed to the pure monoclinic phase (space group: P21/n) of CaAl2O4. The crystal structure of CaAl2O4 is derivative of the stuffed tridymite structure. In this structure, all atoms are on the general site, with C1 site symmetry. Two of the Ca ions sit in distorted octahedra, the third Ca is housed in a lopsided pentagonal pyramid, whereas Al is in six distorted tetrahedral sites that are corner linked in three dimensions . When Yb2+ and Yb3+ ions are introduced into CaAl2O4 crystal, they will take the place of Ca2+ ions on the same site due to the similar ionic radius of Ca2+, Yb2+ and Yb3+ ions (both 0.099 nm for Ca2+ and Yb3+, and 0.102 nm for Yb2+) .
Figure 2 shows the absorption spectra for sample A0, A, B0 and B, respectively. The peak centered at 950 nm in the near infrared region is due to the absorption of Yb3+ ions. The intensity of the absorption peak in sample B is much stronger than that of sample A due to the reduction of Yb3+ ions sintered under reducing atmosphere . We calculated the integral area of the absorption peak at 950 nm in sample A and B, and estimate the extent of Yb2+ ions. The result shows that about 42 percents of Yb3+ ions were reduced to Yb2+ in a reducing atmosphere. We calculated the difference between the absorption spectra of sample A and A0, and also the difference between the absorption spectra of sample B and B0 (shown in the inset of Fig. 1, curve 2 and 1). For curve 1, in the ultraviolet region, the peak located at 265 nm is due to the charge transfer (CT) transition of Yb3+ ions that involves the transfer of an electron from the ligand anion O2− to the central cation Yb3+ ion . It is known that configurational 4fN –4fN−15d transitions are in general composed of very broad peaks due to a displacement of the vibrational equilibrium position between the ground and excited states . The new absorption peak in the curve 2, which centered at 310 nm, is caused by the 4f-5d transitions of Yb2+ ions . Curve 3 is the calculated difference between curve 1 and curve 2. From the absorption spectra and the difference curves, we can conclude that some Yb3+ ions were reduced to Yb2+ ions during the sintering process in sample A, and this difference in concentration of Yb2+ and Yb3+ ions between samples A and A0, and B and B0 affects their luminescence properties.
In Fig. 3 , PL spectra in the near infrared region and PLE spectra in the ultraviolet and visible region of Yb3+ monitored at 978 nm of three samples (A, A0 and B) are shown. In the Yb3+-free sample, there is no detectable emission peaks in the near infrared region for both samples A0 and B0 (the emission curve of B0 is not shown). For sample B sintered in air, the near infrared region emission peaks correspond to an excitation peak located at 275 nm which is due to the CT transition of Yb3+ ions, as discussed above. For sample A, when excited with a 310 nm UV light, there is an intense emission peak located at 978 nm accompanied by several weak shoulders owing to transitions among different Stark levels of 2FJ (J=5/2, 7/2) of Yb3+ ions. The excitation spectra monitoring the Yb3+ emission of sample A contains two peaks located at 275 and 310 nm, respectively. The 275 nm excitation peak has been discussed above. The 310 nm excitation peak is caused by the energy transfer (ET) from Yb2+to Yb3+ . When irradiated with a 310 nm UV light, the Yb2+ ions were excited from the 4f14 energy level to the 4f135d energy level. The excited Yb2+ ions can transfer their energy to Yb3+ ions, and result in the near infrared emission originated from the 2F5/2 →2F7/2 transition of Yb3+ ions. In the visible region, no Yb2+ ions emission is observed. Similar phenomenon has been reported by Meijerink et al in 1997 in the same phosphors matrix . It is known that the host lattice (composition and structure) has a large influence on the quenching temperature of the f-d luminescence. In this system, the Yb2+ luminescence cannot be detected until down to 4.2 K caused by thermally activated photoionization at relatively low temperature.
The energy transfer process in the studied system can be taken through different pathways, such as multiphonon-assisted first-order process and second-order cooperative DC process. The energy mismatch between the lowest 5d level of Yb2+ and the 2F5/2 level of Yb3+ in the host of calcium aluminates is approximately 20000 cm−1, which is far larger than the vibration energy of the host (usually 1000 cm−1 for aluminate). As we known, the non-resonant energy transfer combined with phonon assistance is described in Ref .5]. The processes we study require at least twenty phonons to balance the energy gaps. So it is reasonable to assume a cooperative DC route for this ET, in which excitation energy at the 5d level of Yb2+ is transferred simultaneously to three nearby Yb3+ ions. However, the three photon cutting process is just an assumption based on the discussion above. To show the direct evidence, we measured the dependence between Yb3+ luminescence intensity and the excitation power at 325 nm, shown in Fig. 4 (a) , which exhibits a sublinear relationship with a slope of 0.43. We used a 325 nm laser as the irradiation source, and several attenuators (0, 50%, 25%, 10%) working in this wavelength region to realize the control of different laser power. The slope should be close to 0.5 for a two photo cutting process, and 0.33 for a three photo cutting process. Our result indicates that the energy of one excited Yb2+ ion is transferred to two or three different Yb3+ ions. Figure 4 (b) shows the Yb3+ emission spectra of sample A of various temperatures. The emission intensity of Yb3+ decreases with increasing temperature, which is different from previous observation. We suggest that the energy of an excitated Yb2+ ion can be transferred to Yb3+ ions through two ways: (1) three photon cutting process, as one Yb2+ ions fall back to the ground state, three Yb3+ ions are excitated to the 2F5/2 energy level considering that the slope of 0.43 is much smaller than the slope for two photon cutting process, as shown in Fig. 5 (a) . (2) phonon-assisted two photon cutting process, cooperative DC process takes place between one Yb2+ ions and two Yb3+ ions with the assistance of phonons considering the increase in luminescence intensity with decreasing temperature caused by the competition between two processes and the prohibition of non-radiation relaxation, as shown in Fig. 5 (b). Another phenomenon should be explained here. In Fig. 5 (b), the phonon-assisted two-photon quantum cutting process needs at least 8 phonons to bridge the energy gap in this process. As we mentioned above, the radiative decay and multi phonon relaxation can compete when the gap is five times the phonon energy. However, the excited states of Yb2+ are easily dissipated in this phosphor matrix according to the reports by A Meijerink et al., which increased the probability of the phonon-assisted two photon cutting process. Besides this energy transfer pathway, the three photon cutting process that needs three Yb3+ ions at the nearest crystal sites also have a very low probability. Comparing with the two energy transfer process, other pathways have an extremely low probability which can be neglected. So, in this studied system, the excited Yb2+ ions can only dissipate the energy in these two ways (1. the phonon-assisted two photon cutting process, 2. the three photon cutting process) in spite of the low probability of these two pathways in other materials.
In summary, CaAl2O4 phosphors doped with 0 and 4 mol% Yb3+ ions were successfully synthesized using solid-state reaction method under two different ambiences. The quantity of Yb2+ sintered in reducing atmosphere is far larger than that sintered in air. Strong infrared emission originated from Yb3+ 2F5/2 → 2F7/2 transition is observed in Yb3+ ions doped powder samples sintered in reducing atmosphere, which is due to the energy transfer from Yb2+ ions to Yb3+ ions. These Yb2+ ions correspond to a broad excitation band in the near-ultraviolet and visible region of 250-450 nm. From the PL and PLE spectra of powder sample sintered under reducing atmosphere, we can conclude that this kind of materials would have potential use in the modification and application of solar spectral through a downconversion process achieved by energy transfer from Yb2+ to Yb3+ ions and may increase the conversion efficiency of c-Si solar cells.
This work was financially supported by the National Natural Science Foundation of China (Grant Nos. 50672807, 50872123 and 50802083), National Basic Research Program of China (2006CB8060007), and the Program for Changjiang Scholars and Innovative Research Team in University (IRT0651).
References and links
1. D. Timmerman, I. Izeddin, P. Stallinga, I. N. Yassievich, and T. Gregorkiewicz, “Space-separated quantum cutting with silicon nanocrystals for photovoltaic applications,” Nat. Photonics 2(2), 105–109 (2008). [CrossRef]
2. B. van der Zwaan and A. Rabl, “Prospects for PV: a learning curve analysis,” Sol. Energy 74(1), 19–31 (2003). [CrossRef]
3. C. Strumpel, M. McCann, G. Beaucarne, V. Arkhipov, A. Slaouic, V. Svrcek, C. del Canizo, and I. Tobias, “Modifying the solar spectrum to enhance silicon solar cell efficiency - An overview of available materials,” Solar Energy Solar Cells. 91(4), 238–249 (2007). [CrossRef]
4. B. M. van der Ende, L. Aarts, and A. Meijerink, “Near-Infrared Quantum Cutting for Photovoltaics,” Adv. Mater. 21(30), 1 (2009). [CrossRef]
5. L. Aarts, B. M. van der Ende, and A. Meijerink, “Downconversion for solar cells in NaYF4:Er,Yb,” J. Appl. Phys. 106(2), 023522 (2009). [CrossRef]
6. S. Ye, B. Zhu, J. Luo, J. Chen, G. Lakshminarayana, and J. Qiu, “Enhanced cooperative quantum cutting in Tm3+- Yb3+ codoped glass ceramics containing LaF3 nanocrystals,” Opt. Express 16(12), 8989–8994 (2008). [CrossRef] [PubMed]
7. X. Liu, Y. Qiao, G. Dong, S. Ye, B. Zhu, G. Lakshminarayana, D. Chen, and J. Qiu, “Cooperative downconversion in Yb3+/-RE3+ (RE=Tm or Pr) codoped lanthanum borogermanate glasses,” Opt. Lett. 33(23), 2858–2860 (2008). [CrossRef] [PubMed]
8. Q. Zhang, G. Yang, and Z. Jiang, “Cooperative downconversion in GdAl3(BO3)4:RE3+, Yb3+ (RE=Pr, Tb, and Tm),” Appl. Phys. Lett. 91(5), 051903 (2007). [CrossRef]
9. D. Chen, Y. Wang, Y. Yu, P. Huang, and F. Weng, “Quantum cutting downconversion by cooperative energy transfer from Ce3+ to Yb3+ in borate glasses,” J. Appl. Phys. 104(11), 116105 (2008). [CrossRef]
12. D. Chen, Y. Yu, Y. Wang, P. Huang, and F. Weng, “Cooperative Energy Transfer Up-Conversion and Quantum Cutting Down-Conversion in Yb3+:TbF3 Nanocrystals Embedded Glass Ceramics,” J. Phys. Chem. C 113(16), 6406–6410 (2009). [CrossRef]
13. S. Ye, B. Zhu, J. X. Chen, J. Luo, and J. Qiu, “Infrared quantum cutting in Tb3+, Yb3+ codoped transparent glass ceramics containing CaF2 nanocrystals,” Appl. Phys. Lett. 92(14), 141112 (2008). [CrossRef]
15. W. Hörkner, “Zur kristallstruktur von CaAl2O4,” J. Inorg. Nucl. Chem. 38(5), 983 (1976). [CrossRef]
16. S. Iftekhar, J. Grins, G. Svensson, J. Loof, T. Jarmar, G. A. Botton, C. M. Andrei, and H. Engqvist, “Phase formation of CaAl2O4 from CaCO3-Al2O3 powder mixtures,” J. Eur. Ceram. Soc. 28(4), 747–756 (2008). [CrossRef]
17. R. D. Shannon, “Revised effective ionic radii and systematic studies of interatomicdistances in halides and chalcogenides,” Acta Crystallogr. 32(5), 751–767 (1976). [CrossRef]
18. J. Oliva, E. De la Rosa, L. A. Diaz-Torres, P. Salas, and C. Ángeles-Chavez, “Annealing effect on the luminescence properties of BaZrO3: Yb3+ microcrystals,” J. Appl. Phys. 104(2), 023505 (2008). [CrossRef]
19. G. Blasse, and B. Grabmaier, Luminescent Materials, Springer-Verlag, 1994.
20. C. Duan and P. A. Tanner, “Simulation of 4f-5d transitions of Yb2+ in potassium and sodium halides,” J. Phys. Condens. Matter 20(21), 215228 (2008). [CrossRef]
21. M. Engholm, L. Norin, and D. Åberg, “Strong UV absorption and visible luminescence in ytterbium-doped aluminosilicate glass under UV excitation,” Opt. Lett. 32(22), 3352–3354 (2007). [CrossRef] [PubMed]
22. S. Lizzo, E. P. Klein Nagelvoort, R. Ersens, A. Meijerink, and G. Blasse, “On the quenching of the Yb2+ luminescence in different host lattices,” J. Phys. Chem. Solids 58(6), 963–968 (1997). [CrossRef]
23. T. Miyakawa and D. L. Dexter, “Phonon sidebands, multiphonon relaxation of excited states, and phonon-assisted energy transfer between ions in solids,” Phys. Rev. B 1(7), 2961–2969 (1970). [CrossRef]