Synthesis, structure and upconversion luminescence of Yb³⁺, Ho³⁺ co-doped Gd₃Al₅O₁₂ garnet phosphor prepared by the Pechini sol–gel method

Abstract

A pure-phase phosphor of Gd₃Al₅O₁₂ garnet (GAG) co-doped with Yb³⁺ and Ho³⁺ ions has been synthesized by the Pechini sol–gel method. Phase evolution upon the calcination temperature has been studied and the structural parameters of the garnet phase have been refined by the Rietveld method based on powder XRD measurements. The FE-SEM micrographs show that the phosphor crystalized in spherical shaped particles with mean diameter smaller than 200 nm when the calcination temperature was not higher than 1200 °C. After the doping of Yb³⁺ and Ho³⁺, on the one hand the GAG phase shows very good thermal stability and on the other hand, yellow-green upconversion luminescence with color coordinates of (0.355, 0.626) was generated under the pumping with a LD centered at 976 nm. Results have indicated that the phosphor has potential applications in some upconversion based lighting and displays.

---

Introduction

Rare earth aluminate garnets with a general formula Re₃Al₅O₁₂ (Re = rare earth) comprise a category of well-known cubic oxides owing to their excellent chemical and physical stability, high creep resistance, optical isotropy, and particularly the ability to accept substantial trivalent Re³⁺ for diverse optical functionalities.¹ Amongst these, Y₃Al₅O₁₂ (YAG) and Lu₃Al₅O₁₂ (LuAG) crystals, two of the most promising laser materials, have also been investigated as candidate hosts for upconversion materials by which the conversion of near infrared (NIR) light to visible light can be easily achieved.^2,3 These materials have attracted considerable interest due to their potential applications in lighting, flat panel displays, and other relevant optoelectronic devices.^4–8 The Gd₃Al₅O₁₂ garnet (GAG) which is not a well-known material as a host compared to YAG and LuAG, has also been adopted as a host for phosphors such as Yb³⁺/Er³⁺ co-doping for upconversion,⁹ Dy³⁺, Ce³⁺, or Eu³⁺ doping for lighting.^10–12 However, rare earth aluminate garnet phosphors have traditionally been produced by solid-state reaction at very high temperature (1500–1700 °C). Some researchers have previously reported that the thermal stability of the GAG phase was not so good at high temperature.^13,14 It is essential to develop a suitable synthetic method and solve the problem of phase decomposition before further exploiting of new applications for this material.

It would be a good choice to prepare such material by the sol–gel method which has been earlier used for preparation of garnet phosphors.^15,16 The aqueous method offers a number of advantages including homogeneous mixing on the atomic scale, accurate stoichiometry control, short processing time, much lower energy costs, and easy manufacturing. By this method the materials can be prepared at much lower temperature (<1000 °C) in comparison with the temperature required for the traditional solid-state reaction at high temperature. Meanwhile, some researchers have reported the way of replacing the Al³⁺ or Gd³⁺ by a certain of other similar elements via forming solid solution to stabilize the garnet lattice of GAG. It has been found that the replacing Al³⁺ with Ga³⁺ or replacing Gd³⁺ with Lu³⁺ would be good choices for the phase stabilization.^14,17

In this work, we report the synthesis results of Yb³⁺ and Ho³⁺ co-doped in GAG phosphor by means of the Pechini sol–gel method and calcining at various temperatures. The pure GAG phase was obtained and the structural parameters were refined by the Rietveld method based on the XRD measurement. The doping of rare earth ions, on the one hand has overcome the decomposition of the GAG phase at temperature as high as 1600 °C and on the other hand, served as sensitizer/activator to generate yellow-green upconversion luminescence in the phosphor.

Experimental

The phosphor of Gd₃Al₅O₁₂ garnet co-doped with 10 at% Yb³⁺ and 1 at% Ho³⁺ (hereafter YH : GAG) was synthesized by a Pechini sol–gel method. The starting chemicals of Gd₂O₃ (purity 99.99%), Yb₂O₃ (purity 99.99%), Ho₂O₃ (purity 99.95%), and Al(NO₃)₃·9H₂O (AR grade) were accurately weighted according to their stoichiometric amounts of the target compound. Firstly, the rare earth oxides of Gd₂O₃, Yb₂O₃, and Ho₂O₃ were dissolved in the diluted nitride acid. Clear solution was obtained after stirring at 70 °C for about 3 hours. In the following step the required amount of Al(NO₃)₃·9H₂O and citric acid monohydrate (citric acid–metal ion = 2 : 1) were added and dissolved in the above solution, the mixture was stirred for 30 minutes. Afterward, about 2.0 g polyethylene glycol (PEG, M.W. = 20 000, AR grade) was added as cross-linking agent and the pH value of the mixed solution was adjusted to about 3.0 with aqueous ammonia. After several hours of heating and vigorous stirring, the wet gel was obtained. Secondly, the wet gel was dried at 100 °C for 24 hours in a vacuum oven. Then the obtained xerogel was preheated at 400 °C for 5 hours in a muffle furnace and the precursor after preheat treatment was later ground into fine powders in an agate mortar. Finally, the powders of precursor were calcined at various temperatures from 700 to 1600 °C for 5 hours in air and the final homogeneous powder products with white color were obtained for measurements.

Powder XRD measurement was carried out on a Shimadzu XRD-6100 X-ray diffractometer with Cu-Kα radiation produced at 40 kV, 30 mA. The micrographs of the phosphors were taken by using a JEOL JSM-7800F field emission scanning electron microscope (FE-SEM). Upconversion spectrum was measured at room temperature by using a PerkinElmer LS-55 spectrometer with a laser diode (LD) centered at 976 nm as pump source. The emission intensity was corrected by taking the spectral sensitivity of the device into account.

Results and discussion

Phase evolution of the powder products upon calcination temperature has been investigated. Fig. 1(a) shows the XRD patterns of the products obtained at various calcination temperatures. It can be seen that the precursor after calcination at 700 °C are essentially amorphous. The garnet phase of GAG (space group Ia3̄d, no. 230) was firstly emerged from the amorphous precursor at a temperature as low as 750 °C, and this is the lowest reported temperature for the GAG crystallization in comparison with the other works.^11,12,14,18 Further increasing the temperature, in addition to the GAG phase the other intermediate phase which can be identified to be the hexagonal GdAlO₃ perovskite (H-GAP, space group P6₃/mmc, no. 194) will emerge together.¹⁹ Fig. 1(b) has shown the partial detail of the XRD patterns of the products obtained after calcination at temperatures from 750 to 950 °C. Comparing the maximum diffraction peaks of the two coexisted phase, i.e., the (102) peak of H-GAP and the (420) peak of GAG, it can be seen that the relative amount of the H-GAP was significantly increased until the temperature being raised up to 900 °C. Thereafter the H-GAP quickly disappeared in the product obtained after calcination at 950 °C. Finally, only the GAG phase could be detected in the product obtained after calcination at the temperate of 1000 °C or above. It can be inferred that, the phase of H-GAP should directly transfer to the GAG phase via reaction with the residual Al–O contained amorphous precursor in the calcination process. The target crystalline phase of GAG can be produced through the following two coexisting reaction routes:

Fig. 1 XRD patterns of the products obtained after calcination at various temperatures of 700–1600 °C (a), the partial detail of the XRD patterns for the products obtained after calcination at 750–950 °C (b), and XRD patterns of the non-doped products obtained after calcination at temperatures of 1200 and 1600 °C (c). The standard XRD patterns of GAG (PDF#73-1371), H-GAP, and O-GAP (PDF#46-0395) are given as bars for comparison.

Such finding is some different from the results reported in other literatures in which the H-GAP phase is rarely generated and another phases of monoclinic Gd₄Al₂O₉ (GAM) and orthorhombic GdAlO₃ perovskite (O-GAP) are most common intermediate products when heating at low temperature.^12,14,16,18 As also shown in the XRD patterns, calcinations at even higher temperature up to 1600 °C have only yielded stronger and sharper XRD peaks but without any change in phase purity of the product. This indicates that the structure of GAG is very thermally stable after the incorporation of Yb³⁺ and Ho³⁺. As mentioned above the pure GAG phase is not stable and it will decompose into the GdAlO₃ and Al₂O₃ when heat at elevated temperature (≥1500 °C).¹³ In this work a comparative study without Yb³⁺ and Ho³⁺ additions were conducted. Fig. 1(c) shows the XRD patterns of the products obtained after calcination at temperatures of 1200 and 1600 °C. It is worth noting that beside the cubic GAG phase, there was another second phase indexed to be the O-GAP being observed in the calcined non-doped products and the amount was increased while the calcination temperature was elevated. It is clearly that the co-doping of Yb³⁺ and Ho³⁺ has successfully suppressed the thermal decomposition of GAG at high temperature. The stabilization effect is similar to the rare earth of Lu³⁺ via replacing the 10% of the Gd³⁺ sites at least in the GAG structure.¹⁴

It is known that the pure phase is favorable for luminescent property of phosphor. The XRD peaks of the products obtained after calcination above 1000 °C are all well matched with the standard pattern of pure cubic phase of GAG. In order to determine the structural parameters of YH: GAG, a typical structural refinement by using the Rietveld method was performed to the phosphor obtained at 1200 °C by the Fullprof program.²⁰ Fig. 2 shows the observed, calculated, and the difference XRD patterns of the YH: GAG. The refined crystal structural parameters and the reliability factors (R-factors) for the refinement were summarized in Table 1. The quality of structural refinement data was checked by the goodness fit factor of χ², which was found to be 1.70. It is indicated a good agreement between the observed and calculated XRD patterns. The YH: GAG crystallizes in space group of Ia3̄d with 8 formula units in the cubic unit cell (Z = 8). The refined lattice parameters were a = b = c = 12.0937(3) Å, α = β = γ = 90°. Considering the ionic radium of Gd³⁺(1.053 Å), Yb³⁺(0.985 Å), and Ho³⁺(1.015 Å), the Yb³⁺ and Ho³⁺ ions would enter the matrix Gd³⁺ sites and cause a slight volume shrinkage (∼0.5%) in comparison with the pure GAG (a = b = c = 12.113 Å, α = β = γ = 90°). This effect of volume shrinkage should be one of the most important reasons for the structural stabilization.

Fig. 2 The observed, calculated, and the difference XRD patterns for the Rietveld refinement of the YH: GAG.

Table 1 Structural parameters of YH: GAG after Rietveld refinement

Atom	Wyckoff position	Point symmetry	Coordinates (x, y, z)
Al1	16a	S6	0	0	0
Al2	24d	S4	3/8	0	1/4
Gd/Yb/Ho	24c	D2	1/8	0	1/4
O	96h	C1	0.96420	0.15016	0.00023
Space group: Ia3̄d (230) – cubic; a = b = c = 12.0937(3) Å; α = β = γ = 90°
R-Factors: Rp = 7.59, Rwp = 10.7, Rexp = 8.19; χ^2 = 1.70

---

Fig. 3(a) illustrates the schematic of unit cell of YH: GAG which was modeled according to the atomic parameters by using the Diamond software.²¹ In the YH: GAG structure, the Al atoms take up two different kinds of site positions. One is the octahedral sites (16a) occupied by Al1 atoms with point symmetry of S₆ and the other is the tetrahedral sites (24d) occupied by Al2 atoms with point symmetry of S₄, as shown in Fig. 3(b–c). The rare earth Gd/Yb/Ho were distributed over only one kind of site position (24c) and were coordinated 8-fold by oxygen with point symmetry of D₂, as shown in Fig. 3(d). The garnet structure can be viewed as a framework built up via corner sharing of the Al–O polyhedra, with the rare earth ions residing in dodecahedral interstices. The shortest distance of neighboring rare earth ions which should be very important for achieving high efficient energy transfer between sensitizer and activator ions is about 3.70 Å in the structure. This is shorter than those found in some other oxides such as CaWO₄(3.87 Å),²² YPO₄(3.77 Å),²³ and Lu₂TeO₆(3.94 Å),²⁴ which have also been used as host crystals for generating efficient upconversion luminescence.^25–27

Fig. 3 Schematic representation of the unit cell of YH: GAG (a) and the [AlO₆] (b), [AlO₄] (c), and [Gd/Yb/HoO₈] (d) polyhedra.

The FE-SEM micrographs of some typical samples are presented in Fig. 4(a–c). From which it can be seen that the effect of calcination temperature on particle morphology is very significant. The precursor after calcination at 1000 °C was almost fully crystalized in uniformly spherical shape and the mean particle size is estimated about 50 nm in diameter. Rising the calcination temperature the crystal particle would undergo considerable growth and coarsening and the particle had grown up to 100–200 nm after calcining the precursor at 1200 °C. Further rising the temperature the boundaries between the neighboring particles would become increasingly blurred. Fig. 4(c) has shown that the particle was almost combined one by one and the average size is increased larger than 200 nm for the product obtained after calcination at 1600 °C. Taking into account the crystallization and morphology, it should be reasonable for setting the calcination temperature in the range of 1000–1200 °C for the phosphor preparation. A particle (marked by red-cross) obtained after calcination at 1200 °C was selected for EDS analysis and the spectrum was shown in Fig. 4(d). The atomic percentage of the majority atom of Gd, Al, and O were measured to be 15.05%, 26.40%, and 58.55%, respectively, which is very close to the theoretical atomic ratio of the GAG garnet phase (Gd : Al : O = 3 : 5 : 12).

Fig. 4 FE-SEM micrographs of the precursor after calcination at 1000 °C (a), 1200 °C (b), and 1600 °C (c), respectively, and the EDS spectrum for selected YH: GAG particle obtained after calcination at 1200 °C (d).

Upconversion luminescence from the YH: GAG was measured by using a 976 nm LD to pump the Yb³⁺ ions. Fig. 5(a) shows the spectrum in the wavelength region from 400 to 800 nm. According to the energy level diagrams all the recorded emission bands should be ascribed to the transitions of Ho³⁺ ions. The intensive bands of green (∼550 nm) and red (∼650 nm) emissions should be respectively assigned to the ⁵F₄, ⁵S₂ → ⁵I₈ and ⁵F₅ → ⁵I₈ transitions, while the other weak bands of the blue (∼470 nm) and NIR (∼760 nm) emissions should be respectively assigned to the ⁵F_2,3 → ⁵I₈ and ⁵F₄, ⁵S₂ → ⁵I₇ transitions. The dependence of upconversion emission intensity on the pump power is essential to investigate the upconversion mechanism. The slope (n) obtained from the relation between the integrated upconversion intensity (I_UC) and the pump intensity (I_P), I_UC ∝ I_Pⁿ, in a double logarithmic scale, indicates the number of photons involved in the transition process to populate the upconversion emitting level.²⁸ As shown in Fig. 5(b), the linear fitted slopes for the green and red emissions were 1.99 and 1.80, respectively, which indicated that both of the two upconversion emissions should be ascribed to two photon processes. The possible upconversion mechanism for the Yb³⁺, Ho³⁺ co-doped system have been discussed in many other works.^29–33 As described in the energy level diagram shown in Fig. 5(c), when pumped at 976 nm the Yb³⁺ ions at ground state of ²F_7/2 absorbed the excitation energy effectively and were populated to the excited level of ²F_5/2. After this, Ho³⁺ ions at the ⁵I₈ ground state can be populated to the ⁵F₄, ⁵S₂ levels from the ⁵I₆ due to two consecutive energy transfers from Yb³⁺ to Ho³⁺. The major part of Ho³⁺ ions returns to the ⁵I₈ ground state by producing the strong green emission. In the meantime, small part of Ho³⁺ ions on ⁵F₄, ⁵S₂ levels can also depopulate to the ⁵I₇ level by emitting the weak NIR emission. Then, Ho³⁺ ions on ⁵I₇ level can again be excited to ⁵F₅ level via energy transfer and the ⁵F₅ level acts as the emitting level for generating the red emission via ⁵F₅ → ⁵I₈ transition. The cooperative sensitization (CS) involving energy transfer from two excited Yb³⁺ ions to a Ho³⁺ ions at ground state could be responsible for the Ho³⁺ ion populating on the ⁵F_2,3 levels. The weak blue emission was produced via ⁵F_2,3 → ⁵I₈ transition and the similar energy transfer process has been previously demonstrated in the Yb³⁺, Ho³⁺ co-doped BaCa₂Al₈O₁₅ as well as the Yb³⁺, Tb³⁺ and Yb³⁺, Eu³⁺ co-doped garnet systems by other researchers.^34,35 The color coordinates of the upconversion luminescence was also calculated to be (0.355, 0.626) based on the upconversion spectrum and plotted in the 1931 CIE chromaticity diagram, as shown in Fig. 5(d). The coordinates is located in the yellow-green region which can also be seen in the digital image shown in the inset. It may indicate that the YH: GAG was a potential phosphor for generating upconvertion light emission and could be find applications in lighting and displays.

Fig. 5 Upconversion spectrum of YH: GAG phosphor pumped with a 976 nm LD (a), the integrated intensities upon the pump intensities in double logarithmic diagram for the green and red emissions (b), the schematic representation of the mechanisms responsible for the generation of upconversion emissions (c), and the color coordinates in the 1931 CIE chromaticity diagram with the inset of upconversion digital image (d).

Conclusions

In conclusion, Yb³⁺ (10 at%) and Ho³⁺ (1 at%) co-doped GAG phosphor with high stability was successfully synthesized by Pechini sol–gel method. Phase evolution of the precursor upon calcination temperature has been investigated. The garnet phase was initially crystalized from the precursor at temperature as low as 750 °C. And then the H-GAP phase occurred with increasing the calcination temperature. The pure garnet phase with spherical shape can be obtained after calcining the precursor above 1000 °C and the mean particle diameter can be easily controlled within 200 nm. The structural parameters of the YH: GAG crystal have been confirmed by the Rietveld refinement. The Yb³⁺ and Ho³⁺ substituting the Gd³⁺ ions occupy the dodecahedral sites with point symmetry of D₂. This substituting has caused some volume shrinkage for the garnet unit cell and inhibited the decomposition of GAG phase at elevated temperature (1600 °C). When the YH: GAG phosphors were excited by a NIR LD centered at 976 nm, the upconversion emission spectrum has shown two dominate emission bands of Ho³⁺ ions corresponding to the ⁵F₄, ⁵S₂ → ⁵I₈ and ⁵F₅ → ⁵I₈ transitions. The color coordinates was calculated to be (0.355, 0.626) which was located at the yellow-green region in the CIE 1931 diagram. Results have shown that the pure Yb: GAG phosphor has some potential application in lighting and display. The next study of further optimizing the Yb³⁺ and Ho³⁺ concentrations is very necessary to achieve the best upconversion performance.

Acknowledgements

This work has been supported by the National Natural Science Foundation of China (grant 51302228) and Fundamental Research Funds for the Central Universities of Southwest University (grant XDJK2013C089, XDJK2015B019, and XDJK2016E008).

References

1. J.-G. Li and Y. Sakka, Sci. Technol. Adv. Mater., 2015, 16, 014902.
2. M. Liu, S. W. Wang, J. Zhang, L. Q. An and L. D. Chen, Opt. Mater., 2007, 29, 1352–1357.
3. D. N. Patel, B. R. Reddy and S. K. Nash-Stevenson, Appl. Opt., 1999, 38, 3271–3274.
4. H. A. Höppe, Angew. Chem., Int. Ed., 2009, 48, 3572–3582.
5. E. Downing, L. Hesselink, J. Ralston and R. Macfarlane, Science, 1996, 273, 1185–1189.
6. Z. Lu, J. Wang, X. Xiang, R. Li, Y. Qiao and C. M. Li, Chem. Commun., 2015, 51, 6373–6376.
7. F. Wang, D. Banerjee, Y. Liu, X. Chen and X. Liu, Analyst, 2010, 135, 1839–1854.
8. J. Zhou, Q. Liu, W. Feng, Y. Sun and F. Li, Chem. Rev., 2015, 115, 395–465.
9. H. Choi, J. H. Kim, J. Y. Park, B. K. Moon and J. H. Jeong, J. Korean Phys. Soc., 2011, 59, 2329–2333.
10. L. Tian, L. Wang, L. Zhang, Q. Zhang, W. Ding and M. Yu, J. Mater. Sci.: Mater. Electron., 2015, 26, 8507–8514.
11. J. Y. Park, H. C. Jung, G. Seeta Rama Raju, B. K. Moon, J. H. Jeong, S.-M. Son and J. H. Kim, Opt. Mater., 2009, 32, 293–296.
12. J. Li, J.-G. Li, J. Li, S. Liu, X. Li, X. Sun and Y. Sakka, J. Solid State Chem., 2013, 206, 104–112.
13. T. Shishido, K. Okamura and S. Yajima, J. Am. Ceram. Soc., 1978, 61, 373–375.
14. J. Li, J.-G. Li, Z. Zhang, X. Wu, S. Liu, X. Li, X. Sun and Y. Sakka, J. Am. Ceram. Soc., 2012, 95, 931–936.
15. C. Sun, W. Lü, F. Yang and C. Tu, J. Alloys Compd., 2012, 512, 160–164.
16. E. Garskaite, N. Dubnikova, A. Katelnikovas, J. Pinkas and A. Kareiva, Collect. Czech. Chem. Commun., 2007, 72, 321–333.
17. C.-C. Chiang, M.-S. Tsai and M.-H. Hon, J. Electrochem. Soc., 2007, 154, J326–J329.
18. J. Li, J.-G. Li, S. Liu, X. Li, X. Sun and Y. Sakka, J. Mater. Chem. C, 2013, 1, 7614–7622.
19. F. Bertaut and J. Marescha, Comptes Rendus Hebdomadaires des Seances de l'Academie des Sciences, 1963, 257, 867–870.
20. J. Rodríguez-Carvajal, Phys. B, 1993, 192, 55–69.
21. W. Pennington, J. Appl. Crystallogr., 1999, 32, 1028–1029.
22. A. Zalkin and D. H. Templeton, J. Chem. Phys., 1964, 40, 501–504.
23. I. KrstanoviĆ, Z. Kristallogr., 1965, 121, 315–316.
24. J. F. Tang, J. Gou, G. N. Li, H. He, Y. Li and C. M. Li, J. Alloys Compd., 2016, 672, 1–6.
25. W. Xu, H. Zhao, Y. Li, L. Zheng, Z. Zhang and W. Cao, Sens. Actuators, B, 2013, 188, 1096–1100.
26. R. Balakrishnaiah, D. W. Kim, S. S. Yi, S. H. Kim, K. Jang, H. S. Lee, B. K. Moon and J. H. Jeong, Thin Solid Films, 2010, 518, 6145–6148.
27. J. F. Tang, G. N. Li, C. Yang, J. Gou, D. H. Luo and H. He, CrystEngComm, 2015, 17, 9048–9054.
28. M. Pollnau, D. R. Gamelin, S. R. Lüthi, H. U. Güdel and M. P. Hehlen, Phys. Rev. B, 2000, 61, 3337–3346.
29. J. F. Tang, C. H. Cheng, Y. J. Chen and Y. D. Huang, J. Alloys Compd., 2014, 609, 268–273.
30. J. C. Boyer, F. Vetrone, J. A. Capobianco, A. Speghini and M. Bettinelli, Chem. Phys. Lett., 2004, 390, 403–407.
31. A. Pandey and V. K. Rai, Appl. Phys. B, 2012, 109, 611–616.
32. T. Grzyb, S. Balabhadra, D. Przybylska and M. Węcławiak, J. Alloys Compd., 2015, 649, 606–616.
33. A. S. Gouveia-Neto, L. A. Bueno, A. C. M. Afonso, J. F. Nascimento, E. B. Costa, Y. Messaddeq and S. J. L. Ribeiro, J. Non-Cryst. Solids, 2008, 354, 509–514.
34. A. Kumari and V. K. Rai, Mater. Res. Bull., 2015, 72, 168–175.
35. R. Martín-Rodríguez, R. Valiente, S. Polizzi, M. Bettinelli, A. Speghini and F. Piccinelli, J. Phys. Chem. C, 2009, 113, 12195–12200.

---