Nucleation sequence on the cation exchange process between Y_0.95Eu_0.05PO₄ and CePO₄ nanorods

Abstract

Nanorods of Y_0.95Eu_0.05PO₄@CePO₄ (Y_0.95Eu_0.05PO₄ phase was nucleated first and then a CePO₄ phase was nucleated) and CePO₄@Y_0.95Eu_0.05PO₄ (CePO₄ phase was nucleated first and then Y_0.95Eu_0.05PO₄ phase was nucleated) were prepared at a relatively low temperature of 140 °C in ethylene glycol medium. Based on XRD, TEM and Raman studies it has been inferred that Y_0.95Eu_0.05PO₄@CePO₄ sample consists of a mixture of bigger (length around 800–1000 nm and width around of 80–100 nm) and smaller (length around 70–100 nm and width around 10–20 nm) nanorods, having monoclinic CePO₄ and tetragonal YPO₄ structure, whereas CePO₄@Y_0.95Eu_0.05PO₄ sample consists of mainly small nanorods having a single phase CePO₄ structure. From the detailed luminescence studies it has been established that there exists significant incorporation of Y³⁺/Eu³⁺ ions in the CePO₄ phase in CePO₄@Y_0.95Eu_0.5PO₄ sample. This has been attributed to the cation exchange taking place between Ce³⁺ ions in CePO₄ host and Eu³⁺ and Y³⁺ ions in solution during the synthesis stage. Unlike this, such an exchange is not possible for Y_0.95Eu_0.05PO₄@CePO₄ sample synthesized under identical conditions due to the higher solubility product (K_sp) value of YPO₄ compared to CePO₄. Incorporation of Eu³⁺ in the CePO₄ lattice of CePO₄@Y_0.95Eu_0.5PO₄ sample is confirmed by the significant reduction in the lifetime of ⁵D₀ level of Eu³⁺ and the luminescence intensity from Eu³⁺, arising due to the electron transfer between the Ce³⁺/Ce⁴⁺ and Eu³⁺/Eu²⁺ species. These results are further supported by the non-radiative decay rates and quantum yields calculated from the emission spectrum.

---

1. Introduction

Lanthanide phosphates (LnPO₄) are robust materials for various technological applications involving lasers due to their high thermal and chemical stability, extremely low solubility in water and improved solubility for rare earth ions like Nd³⁺.^1–6 They exist mainly in two crystalline modifications, known as monazite and xenotime forms. Monazite structure is formed by lanthanide phosphates with relatively larger ionic radii lanthanide ions (i.e. La³⁺ - Gd³⁺) and xenotime structure is formed by smaller ionic radii lanthanide ions (i.e. Tb³⁺-Lu³⁺). YPO₄ crystallizes in the xenotime structure and belongs to the tetragonal crystal system whereas CePO₄ crystallizes in monazite structure and belongs to the monoclinic crystal system.⁷ An essential difference in both structures is that in the xenotime structure, Y³⁺ ions exist as regular YO₈ polyhedra whereas in the monazite structure, Ce³⁺ ions exist as CeO₉ polyhedra. Lanthanide phosphates have got an important application as inter layers in high temperature oxide-oxide ceramic matrix composites (CMCs).^8–10 They are known to improve the toughness in ceramic composites by providing crack deflection at the fiber–matrix interface. This property mainly depends on the thermal expansion coefficient of the material. Generally xenotime structures have a lower thermal expansion coefficient as compared to monazite structures.^11–13 Thus the solid solution of lanthanide phosphates with xenotime and monazite structure can give rise to a material with a tunable thermal expansion coefficient. Another approach to make materials with a tunable thermal expansion coefficient is to form materials with core-shell geometry. The core or the shell can be the materials with either monozite or xenotime structures. If the core-shell materials have nano size dimensions, it will be an advantage as it can be used to distribute uniformly in a matrix.

A number of studies are available on the low temperature synthesis and characterization of lanthanide phosphate nano-materials with core-shell geometry.^14–18 For example, Kompe et al.¹⁷ employing co-precipitation in presence of coordinating ligands like tributyl phosphate (TBP) and trihexyl amine, prepared CePO₄ : Tb³⁺ core- and CePO₄ : Tb³⁺-LaPO₄ core-shell nanoparticles in water free environments at 200 °C. The authors have used the luminescence technique to confirm the core-shell formation. They observed that the luminescence properties of CePO₄ : Tb³⁺ nanorods/nanoparticles get significantly improved when covered by a shell of LaPO₄. Buissette et al.¹⁸ prepared highly luminescent core/shell particles with rhabdophane structures and it was observed that such particles have an improved stability against chemical and thermal oxidation.

Although, the core-shell formation has been reported in various lanthanide based systems, the existence of core and shell materials with well defined demarcation is questionable, particularly in lanthanide based inorganic nano-materials. In a recent study, Dong et al.¹⁹ have demonstrated that when GdF₃ nanoparticles dispersed in water are exposed to excess of La³⁺ ions, there exists a rapid exchange of Gd³⁺ ions from a GdF₃ lattice by La³⁺ ions in aqueous solution, resulting in the formation of LaF₃ nanoparticles and aqueous solution of Gd³⁺ ions. The extent of exchange has been found to depend on the time duration involved in the reaction. It is observed that lanthanide ions from the left of the periodic table (for example La³⁺) can replace the ones from the right (for example Gd³⁺) present in a lattice. Such rapid ion exchange process will be a hindrance to form materials with core-shell structures; at the same time it will be helpful for forming solid solutions with a range of compositions.

In our previous studies^20,21 we have shown that lanthanide phosphate nano-materials with high dispersability in water can be easily prepared at low temperatures by using ethylene glycol as a solvent as well as a stabilizing ligand. In the present study we investigate the ion exchange process taking place between Ce³⁺ and Y³⁺&Eu³⁺ ions from CePO₄ and Y_0.95Eu_0.05PO₄ phases, respectively, by changing the sequence/order in which nucleation of CePO₄ and Y_0.95Eu_0.05PO₄ phases is taking place in ethylene glycol medium. The Eu³⁺ ion has been used as a probe to understand this, as its luminescence is very sensitive to the nature of the environment around the ion. Raman and X-ray diffraction (XRD) techniques have been used to further substantiate the luminescence results. To the best of the authors’ knowledge this is the first time that such studies are being carried out for two different LnPO₄ nanorods by using luminescence and Raman spectroscopic techniques.

2. Experimental details

2.1 Materials

Chemicals used for the synthesis were AR grade Y₂O₃, Eu₂O₃, Ce₂(CO₃)₃·5H₂O and NH₄H₂PO₄ (all 99.9% pure). All chemicals were used as obtained, without further purification.

2.2 Synthesis of YPO₄, Y_0.95Eu_0.05PO₄ and CePO₄ nano-materials

Around 0.5 g of Y₂O₃ and a stoichiometric amount of Eu₂O₃ (0 and 5 at% of Y³⁺) or 0.5 g of Ce₂(CO₃)₃·5H₂O were dissolved in dilute HCl and the excess of acid was removed by repeated evaporation by adding water. This was then mixed with 25 ml of ethylene glycol and heated first to 100 °C. At this temperature ∼0.14 g of NH₄H₂PO₄ dissolved in 10 ml of ethylene glycol was added dropwise and the temperature was raised to 140 °C and maintained at this value for a period of two hours until the reaction was complete. The precipitate thus obtained was washed many times with acetone and ethanol and dried under ambient conditions. The powder samples are readily dispersible in solvents like methanol and water.

2.3 Preparation of Y_0.95Eu_0.05PO₄@CePO₄ and CePO₄@Y_0.95Eu_0.05PO₄ samples

For preparing Y_0.95Eu_0.05PO₄@CePO₄ samples, initially Y_0.95Eu_0.05PO₄ phase was nucleated in ethylene glycol medium as described above. The nucleation of the Y_0.95Eu_0.05PO₄ phase is marked by the appearance of turbidity in the reaction medium. Five minutes after the nucleation of the Y_0.95Eu_0.05PO₄ phase, Ce³⁺ solution in a minimum amount of dilute HCl and a required amount of NH₄H₂PO₄ dissolved in 5 ml ethylene glycol were added and the temperature was maintained at this value for 1 more hour. After the reaction, the precipitate was washed several times with acetone and ethanol and dried under ambient conditions. The powder samples are readily dispersible in solvents like methanol and water. A similar procedure was employed for the CePO₄@Y_0.95Eu_0.05PO₄ sample.

2.4 Synthesis of bulk cerium phosphate and yttrium phosphate

Bulk YPO₄ and CePO₄ were prepared using the precipitation method in water followed by heating in air at 800 °C. Initially two separate solutions, containing 0.5 g of Y₂O₃/Ce₂(CO₃)₃·5H₂O dissolved in 20 ml of dilute HNO₃ and stoichiometric amount of NH₄H₂PO₄ dissolved in distilled water, were made. Yttrium nitrate/cerium nitrate solution was added dropwise into a solution of NH₄H₂PO₄ with vigorous stirring and the pH of the medium was raised to a value of 10 by adding NH₄OH. The precipitate obtained was then filtered, washed, dried and calcined at 800 °C for 5 h.

2.5 Characterization

X-Ray powder diffraction patterns were recorded using Ni filtered Cu Kα radiation (λ = 1.54178 Å) from a Diano Worburn, X-ray diffractometer attached with a graphite monochromator, operating at 35 kV and 20 mA. The lattice parameters were calculated from the diffraction peaks using the least squares fit method. TEM investigations were carried out using a JEOL JEM 3000F Microscope. Samples for TEM investigations were prepared by dispersing the samples in methanol and a drop of the above solution was put on a carbon coated holey TEM grids and allowed to dry. Raman spectra were recorded using a 532 nm line from a diode-pumped Nd-YAG laser (power 15 mW) focused to a spot size of about 20 μ. The scattered light was analysed using a home-built 0.9 m single monochromator coupled with a super notch filter and detected by a cooled charge couple device (CCD, Andor technology). The entrance slit was kept at 50 mm, which gave a resolution limited linewidth of 3 cm⁻¹.

All steady-state luminescence and lifetime measurements were carried out at room temperature by using an Edinburgh Instruments' FLSP 920 system, having a 450W Xe lamp, 60 W microsecond flash lamp and nanosecond Hydrogen flash lamp as the excitation sources. Around 20 mg sample was dispersed in 5 ml methanol prior to luminescence measurements. Emission spectra were corrected for the detector response and the excitation spectra for the lamp profile. All emission measurements were carried out with a resolution of 3 nm.

3. Results and discussion

Fig. 1 shows the X-ray diffraction patterns for Y_0.95Eu_0.05PO₄@CePO₄ and CePO₄@Y_0.95Eu_0.05PO₄ samples. The XRD pattern of Y_0.95Eu_0.05PO₄@CePO₄ sample shows peaks characteristic of both tetragonal yttrium phosphate and monoclinic CePO₄ phases.^22–24 Unlike that, the CePO₄@Y_0.95Eu_0.05PO₄ sample shows only the monoclinic cerium phosphate phase. Table 1 gives the unit cell parameters obtained from the least square fitting of the corresponding X-ray diffraction patterns. For the purpose of comparison, the lattice parameters are calculated for undoped CePO₄ and YPO₄ samples prepared by the identical methods and are also given in the table. There is a slight reduction in the unit cell volume for CePO₄ phase and a slight increase in the unit cell volume of Y_0.95Eu_0.05PO₄ phase for Y_0.95Eu_0.05PO₄@CePO₄ sample. A similar extent of decrease in the unit cell volume of CePO₄ phase is also observed for CePO₄@Y_0.95Eu_0.05PO₄ sample compared to undoped CePO₄. The increase in the unit cell volume of Y_0.95Eu_0.05PO₄ phase compared to YPO₄ has been attributed to the higher ionic radii of Eu³⁺ (1.07 Å) compared to Y³⁺ (1.02 Å).²⁵ Similarly, a decrease in the unit cell volume of CePO₄ phase in both samples indicates a partial replacement of Ce³⁺ ions (ionic radius = 1.14 Å) by Y³⁺/Eu³⁺ ions. The absence of peaks corresponding to Y_0.95Eu_0.5PO₄ phase in the diffraction pattern of CePO₄@Y_0.95Eu_0.05PO₄ sample has been attributed to the significant extent of ion exchange between Ce³⁺ ions in the CePO₄ lattice by Eu³⁺ and Y³⁺ ions present in the reaction medium. The reason for the observed ion exchange process is the lower solubility product for CePO₄ phase (4.2 × 10⁻¹²) compared to YPO₄ phase (1.4 × 10⁻¹¹)²⁶ (assuming the free energy for the formation of Ce³⁺/Y³⁺ ions and CePO₄/YPO₄ phases in aqueous solution is the same as that in ethylene glycol medium). On the other hand, when the nucleation sequence is reversed, the conditions under which the Y_0.95Eu_0.05PO₄ phase precipitates, the CePO₄ phase also precipitates and ion exchange will be difficult between two different solid matrices. In order to confirm the above and to understand the observed variation in the lattice parameters, detailed studies have been carried out on these samples using transmission electron microscopy, Raman and luminescence spectroscopic techniques and the results are described below.

Table 1 Lattice parameters of Y_0.95Eu_0.05PO₄@CePO₄ and CePO₄@Y_0.95Eu_0.05PO₄ samples along with undoped CePO₄ and YPO₄ samples

Sample	Lattice	Lattice parameters/Å				Volume/Å^3
		a	b	c	β
CePO4@ Y0.95Eu0.05PO4	Monoclinic	6.766(7)	6.997(4)	6.453(4)	103.56(6)	296.892
Y0.95Eu0.05PO4@ CePO4	Tetragonal	6.890(4)		6.019(8)		285.75
	Monoclinic	6.760(6)	6.983(3)	6.462(5)	103.59(6)	296.544
^22undoped CePO4	Monoclinic	6.779(1)	7.006(3)	6.470(2)	103.60(1)	298.68
^23undoped YPO4	Tetragonal	6.867(1)		6.014(2)		283.59

---

Fig. 1 XRD patterns for (a) Y_0.95Eu_0.05PO₄@CePO₄ and (b) CePO₄@Y_0.95Eu_0.05PO₄ samples. The peaks marked with * correspond to Y_0.95Eu_0.05PO₄ phase. Also the Miller indices of the planes corresponding to Y_0.95Eu_0.05PO₄ phase are marked red in colour.

Fig. 2 (a and b) shows the TEM images of Y_0.95Eu_0.05PO₄@CePO₄ and CePO₄@Y_0.95Eu_0.05PO₄ samples. The corresponding images from single phase Y_0.95Eu_0.05PO₄ and CePO₄ samples are shown in Fig. 1 of the ESI.† The Y_0.95Eu_0.05PO₄@CePO₄ sample consists of a mixture of bigger and smaller nanorods. The bigger nanorods have a length in the range 0.8–1 μm and width in the range of 80–100 nm, whereas for the smaller nanorods the lengths are in the range of 70–100 nm and width around 10–15 nm. A closer look of the image revealed that some of the bigger nanorods also exist in the form of aggregates (bundles). However in CePO₄@Y_0.95Eu_0.05PO₄ sample, only small nanorods having length in the range of 80–130 nm and width in the range of 10–15 nm are seen. The above results along with the TEM image of CePO₄ nanorods shown in Fig. 1 of the ESI† suggest that Y_0.95Eu_0.05PO₄@CePO₄ sample consists of both CePO₄ and YEuPO₄ phases whereas CePO₄@Y_0.95Eu_0.05PO₄ sample mainly consists of single phase nanorods having CePO₄ structure. These results agree well with that obtained from XRD studies discussed above. Fig. 2 (c and d) shows HRTEM images and a selected area electron diffraction (SAED) pattern acquired from a collection of Y_0.95Eu_0.05PO₄@CePO₄ and CePO₄@Y_0.95Eu_0.05PO₄ nano-crystals, respectively. The lattice spacing of 3.2 and 3.5 Å matches well with the (100) and (111) planes of Y_0.95Eu_0.05PO₄ and CePO₄ lattices. The ring pattern observed in the SAED images of CePO₄@Y_0.95Eu_0.05PO₄ nano-crystals (Fig. 2d) suggests a polycrystalline nature of the sample. Thus, the TEM observations, in addition to XRD results show that reversing the order of nucleation changes the dimensions and crystallinity of the nano-rods.

Fig. 2 TEM images of (a) Y_0.95Eu_0.05PO₄@CePO₄ (b) CePO₄@Y_0.95Eu_0.05PO₄ samples. Corresponding HRTEM images and SAED pattern for Y_0.95Eu_0.05PO₄@CePO₄ and CePO₄@Y_0.95Eu_0.05PO₄ samples, respectively, are shown in Fig. 2 (c and d).

As the PO₄ stretching vibrations are Raman active, the Raman spectrum of the sample can give a valuable information on the interaction of tetragonal Y_0.95Eu_0.05PO₄ and monoclinic CePO₄ phases. Hence the Raman spectrum of both Y_0.95Eu_0.05PO₄@CePO₄ and CePO₄@Y_0.95Eu_0.05PO₄ samples was recorded and is shown in Fig. 3. For the purpose of comparison Raman spectra were also recorded for bulk and nano forms of CePO₄ and YPO₄ phases. Fig. 3 (a) shows the Raman spectrum CePO₄ nanorods and YPO₄ nanoparticles along with that of bulk CePO₄ and YPO₄. The internal modes corresponding to PO₄ vibration in the bulk samples appear at 1001 cm⁻¹ in YPO₄ and 971 cm⁻¹ in CePO₄ (Fig. 3(a)). In YPO₄ nanoparticles the frequency of the PO₄ stretching mode was found to be less by about 5 cm⁻¹ as compared to that in bulk, besides, there is an asymmetric broadening of the peak. For CePO₄ nanorods the peak maximum with respect to bulk CePO₄ did not show any detectable variation. However the line widths are found to be more in nanorods compared to bulk sample. The broadening of the peak of YPO₄ nanoparticles and CePO₄ nanorods can be attributed to the combined effect of strain and phonon confinement as the dimensions get smaller. A similar effect has been observed in the Raman spectroscopic studies on the nanoparticles of CeO₂.²⁷ The Raman spectra of Y_0.95Eu_0.05PO₄@CePO₄ and CePO₄@Y_0.95Eu_0.05PO₄ samples are shown in Fig. 3(b). For Y_0.95Eu_0.05PO₄@CePO₄ sample, the peaks corresponding to the PO₄ stretching vibrations from both CePO₄ (973 cm⁻¹) and Y_0.95Eu_0.05PO₄ (992 cm⁻¹) phases can be clearly seen. Deconvolution of the patterns revealed that their intensity ratio is 1 : 1; suggesting that in Y_0.95Eu_0.05PO₄@CePO₄ sample CePO₄ and Y_0.95Eu_0.05PO₄ exist as separate phases without having any interaction. Unlike this, for CePO₄@Y_0.95Eu_0.05PO₄ sample, mainly a broad asymmetric peak at around 973 cm⁻¹ along with a weak shoulder around 993 cm⁻¹ is observed. The peak at 973 cm⁻¹ is characteristic of the symmetric stretching vibrations of the PO₄ unit of monoclinic CePO₄ and that around 993 cm⁻¹ is due to tetragonal YEuPO₄. The deconvolution of the spectrum revealed that their intensity is in a ratio of 9 : 1. A comparison of the band at 973 cm⁻¹ with that of pure CePO₄ nanorods revealed that both the peak maximum and line shape have changed in the CePO₄@Y_0.95Eu_0.05PO₄ nanorods as compared to pure CePO₄ nanorods. This has been attributed to the incorporation of Y³⁺ and Eu³⁺ in the lattice of CePO₄. The very weak intensity of the peak corresponding to PO₄ stretching of YEuPO₄ phase indicate that most of the Y³⁺ and Eu³⁺ions are getting incorporated in the CePO₄ lattice.

Fig. 3 Raman spectrum for (a) bulk and nano phases of CePO₄ and YPO₄ samples. The corresponding patterns for Y_0.95Eu_0.05PO₄@CePO₄ and CePO₄@Y_0.95Eu_0.05PO₄ samples are shown in Fig. 3(b).

Fig. 4(a) shows the emission spectrum for as prepared Y_0.95Eu_0.05PO₄ nanoparticles dispersed in methanol, obtained after exciting the samples at 230 nm. The peaks around 591 and 615 nm are characteristic of ⁵D₀ → ⁷F₁ (magnetic dipole allowed) and ⁵D₀ → ⁷F₂ (electric dipole allowed) transitions respectively of Eu³⁺ ions. The ratio of the relative intensity of ⁵D₀ → ⁷F₂ to ⁵D₀ → ⁷F₁ transition known as the asymmetric ratio of luminescence, is measure of the extent of distortion around the Eu³⁺ ion. The value is found to be 1.4 for Y_0.95Eu_0.05PO₄ nanoparticles. In crystalline YPO₄, Y³⁺ ions have D_2h symmetry and hence in Y_0.95Eu_0.05PO₄ nanoparticles, Eu³⁺ ions will also have D_2h symmetry²² which favors both electric and magnetic dipole transitions. The excitation spectrum corresponding to 591 nm emission is shown as an inset in Fig. 4(a). The broad peak centered around 230 nm is characteristic of the Eu–O charge transfer process in YPO₄ host. In addition to this, sharp peaks due to intra 4f transitions of Eu³⁺ ions can also be clearly seen in the figure. As expected, strong Eu³⁺ emission has been observed from the sample on exciting at the charge transfer peak maximum (230 nm). The decay curve corresponding to 590 nm emission from the nanoparticles is shown in Fig. 4(b). The decay is found to be bi-exponential with a faster decay component of 0.7 ms (14%) and a slower decay component of 2.7 ms (86%). The faster decay component has been attributed to the Eu³⁺ ions at the surface of the nanoparticles and the slower to the ones at the bulk of nanoparticles.

Fig. 4 Emission spectrum (a) and decay curve corresponding to ⁵D₀ level of Eu³⁺ ions (b) for as prepared Y_0.95Eu_0.05PO₄ nanoparticles dispersed in methanol (excitation wavelength used was 230 nm for recording the emission spectrum and 394 nm for recording the decay curve). The inset in Fig. 4(a) shows an excitation spectrum corresponding to 591 nm emission.

Fig. 5 (a and b) shows the emission spectrum and decay curves corresponding to the ⁵D₀ level of Eu³⁺ from Y_0.95Eu_0.05PO₄@CePO₄ sample. The emission spectra of Y_0.95Eu_0.05PO₄ nanoparticles and Y_0.95Eu_0.05PO₄@CePO₄ sample are similar and for the latter the asymmetric ratio of luminescence is also calculated to be 1.4. The same asymmetric ratio of luminescence for both Y_0.95Eu_0.05PO₄ nanoparticles and Y_0.95Eu_0.05PO₄@CePO₄ sample very clearly demonstrates that Eu³⁺ environment is the same in both samples. This is further supported by the comparable lifetime values, namely 0.46 ms (12%) and 2.3 ms (88%), corresponding to the ⁵D₀ level of Eu³⁺ ions in Y_0.95Eu_0.05PO₄@CePO₄ sample. This result is also in agreement with the XRD and Raman results, as it reveals the coexistence of both CePO₄ and Y_0.95Eu_0.05PO₄ phases. The above lifetime values also confirm that a core-shell structure with Y_0.95Eu_0.05PO₄ core and CePO₄ shell does not exist in the sample, as it would have resulted in a significant change in the ⁵D₀ lifetime values of Eu³⁺ ions, which is not observed in the present study. Even though the asymmetric ratio of luminescence and lifetime of ⁵D₀ level of Eu³⁺ are comparable for both Y_0.95Eu_0.05PO₄ nanoparticles and Y_0.95Eu_0.05PO₄@CePO₄ sample, the relative intensity of ⁷F₀ → ⁵L₆ and Eu³⁺-O²⁻ charge transfer transitions are quite different in both samples. This has been attributed to the fact that in the case of Y_0.95Eu_0.05PO₄ sample only one type of Eu³⁺ ion is present with the lattice and all the Eu³⁺ ions can undergo Eu³⁺-O²⁻ charge transfer processes. Unlike this, in the case of Y_0.95Eu_0.05PO₄@CePO₄ sample some of the Eu³⁺ ions present on the surface of the Y_0.95Eu_0.05PO₄ phase can interact with the Ce³⁺ ions of CePO₄ phase. This interaction can have a competing effect with the Eu³⁺ and O²⁻ charge transfer process and thereby reduce the intensity of the charge transfer band in the excitation spectrum corresponding to Eu³⁺ emission from the Y_0.95Eu_0.05PO₄@CePO₄ sample. Detailed Ce³⁺ luminescence studies to further substantiate this aspect have been carried out on this sample as well as on CePO₄@Y_0.95Eu_0.05PO₄ sample and are discussed below.

Fig. 5 Emission spectrum (a) and decay curve (b) corresponding to the ⁵D₀ level of Eu³⁺ from Y_0.95Eu_0.05PO₄@CePO₄ sample. (The excitation wavelength used was 230 nm for recording the emission spectrum and 394 nm for recording the decay curve.) The inset in Fig. 5(a) shows an excitation spectrum corresponding to 591 nm emission.

Fig. 6 (a and b) shows the emission spectrum and decay curve corresponding to Ce³⁺ emission from the Y_0.95Eu_0.05PO₄@CePO₄ sample obtained after 296 nm excitation. The corresponding excitation spectrum is shown as an inset. The peak around 350 nm is characteristic of transition forming the lowest excited state of 5d¹ electronic configuration (²T₂ and ²E levels) of Ce³⁺ to the ²F_5/2 and ²F_7/2 levels of its ground state 4f¹ configuration. The excited state corresponding to the Ce³⁺ emission has been found to decay bi-exponentially with decay time of 10.5 ns (54%) and 2.8 ns (46%). In bulk CePO₄, Ce³⁺ ions are having a distorted environment with a local point group symmetry of C_s.²⁸ Hence a single exponential decay is not expected from the sample. However at this stage it is difficult to pinpoint the origin of the bi-exponential nature of the decay curves.

Fig. 6 The emission spectrum (a) and decay curve (b) corresponding to Ce³⁺ emission from the Y_0.95Eu_0.05PO₄@CePO₄ sample obtained after 296 nm excitation. The corresponding excitation spectrum monitored at 350 nm emission is shown as an inset in Fig. 6(a).

Fig. 7(a) shows the emission spectrum from CePO₄@Y_0.95Eu_0.05PO₄ sample obtained after 230 nm excitation. The emission spectrum is too noisy with a very poor signal to noise ratio indicating that Eu³⁺ emission is significantly quenched in this sample. This is further supported by the lifetime values of the ⁵D₀ level of Eu³⁺ ions in the sample, shown in Fig. 7(b). This is because in this sample Eu³⁺ ions must be occupying both Ce³⁺ and Y³⁺ sites due to the formation of solid solution between CePO₄ and Y_0.95Eu_0.05PO₄ phases brought about by the ion exchange of Ce³⁺ ions from CePO₄ host by Y³⁺/Eu³⁺ ions in solution at the synthesis stage. Eu³⁺ ions, which are near to Ce³⁺ sites, have got a tendency to get converted to Eu²⁺ ions.²⁹ This is associated by the oxidation of Ce³⁺ to Ce⁴⁺.²⁹ The electron transfer process is schematically represented in Fig. 8. The excited electron at the 5d level of Ce³⁺ ions can jump to Eu³⁺ level resulting in the formation of Eu²⁺ and Ce⁴⁺. However as the occupied Eu²⁺ ground state level is higher than the unoccupied Ce³⁺ ground state level, the electron can get transferred back to Ce⁴⁺. No photon is emitted in this process³⁰ and all the absorbed energy gets non-radiatively decayed. This results in the formation of ground state Ce³⁺ and Eu³⁺ ions in the sample, a situation before the absorption of energy. The absence of EPR signals characteristic of Eu²⁺ species from the sample even at liquid nitrogen temperatures also confirmed that the above mechanism is taking place in the sample. It is to be noted that this process can take place only for Ce³⁺ and Eu³⁺ ions which are in the near proximity. One of the reasons for the faster decay time of ⁵D₀ level of Eu³⁺ ions has been attributed to the non-radiative decay of the excited state of Eu³⁺ ion brought about by the above mentioned electron transfer processes between Ce³⁺/Ce⁴⁺ and Eu³⁺/Eu²⁺ species taking place in the lattice. Such electron transfer process and non radiative decay of Eu³⁺ excited state are also expected to have a competing effect on the Eu³⁺-O²⁻ charge transfer process. This has been attributed to the reason for the change in the relative intensities of Eu³⁺-O²⁻ charge transfer peak and ⁷F₀ → ⁵L₆ transition peak of Eu³⁺ ions in Y_0.95Eu_0.05PO₄@CePO₄ sample compared to Y_0.95Eu_0.05PO₄ nanoparticles.

Fig. 7 Emission spectrum (a) and decay curve for 590 nm emission (b) corresponding to CePO₄@Y_0.95Eu_0.05PO₄ sample obtained after 230 nm excitation. The corresponding excitation spectrum for 615 nm emission is shown as an inset in Fig. 7(a).

Fig. 8 Energy level scheme showing the electron jumps (represented as dotted lines) taking place between Eu³⁺ and Ce³⁺ ions in CePO₄ host.

Ce³⁺ emission from CePO₄@Y_0.95Eu_0.05PO₄ sample is shown in Fig. 9 along with its decay curve. The emission and excitation peak characteristic of transition between 4f and 5d states can be very clearly seen. However the lifetime values are shorter compared to the ones corresponding to Y_0.95Eu_0.05PO₄@CePO₄ sample. The reason for the quenching effect is the non-radiative process associated with the Ce³⁺ ions mentioned above taking place in the sample. Thus the luminescence studies confirm the fact that, in the case of CePO₄@Y_0.95Eu_0.05PO₄ sample Eu³⁺/Y³⁺ get incorporated into the CePO₄ lattice. This is associated with a significant reduction in the luminescence intensity from Ce³⁺ and Eu³⁺ ions. Unlike this, for Y_0.95Eu_0.05PO₄@CePO₄ sample, there exist separate phases of CePO₄ and Y_0.95Eu_0.05PO₄ with interaction between Ce³⁺ and Y³⁺ ions existing only at the surface of Y_0.95Eu_0.05PO₄ phase. The reduction in the Eu³⁺ luminescence intensity can also be attributed to the reduction in the Eu³⁺ concentration in CePO₄@Y_0.95Eu_0.05PO₄ sample compared to Y_0.95Eu_0.05PO₄@CePO₄ sample. The above inferences are further confirmed by Raman and XRD results. In order to further substantiate the Eu³⁺ incorporation in CePO₄ phase of CePO₄@ Y_0.95Eu_0.05PO₄ samples, the intensity parameters (Ω_λ) where λ = 2 and 4 along with the radiative (A_rad) and non-radiative (A_nrad) rates for Eu³⁺ ions in all three samples were calculated from the emission spectrum and are described in the following section.

Fig. 9 Emission spectrum (a) and decay curve (b) corresponding to Ce³⁺ emission from the CePO₄@Y_0.95Eu_0.05PO₄ sample obtained after 296 nm excitation. The corresponding excitation spectrum monitored at 350 nm emission is shown as an inset in Fig. 9(a).

The intensity parameters (Ω_λ) where λ = 2 and 4 can be calculated either theoretically from the structural information or experimentally from the emission/absorption spectrum. The Ω_λ value is a measure of the extent of co-valency around Eu³⁺ ions. Further the value also depends on the nature of the forced electric dipole transitions as well as the dynamic coupling mechanisms operating in the sample. In the present study we have calculated the Ω_λ values (where λ = 2 and 4) by considering the ⁵D₀ → ⁷F₁ transition as the reference. Emission intensity can be expressed as I = NħωA, where N is the population of the emitting level (⁵D₀), ħω is the transition energy, and A is Einstein's coefficient of spontaneous emission.³¹A is related to Ω_λ values through the following expression (eqn (1))

where “ν” is the frequency, “ℏ” is the Planck constant, “e” is the charge of the electron, “1/4πε₀” is a constant and is equal to 8.99 × 10⁹ N m² C⁻², “χ” is Lorents field correction and is equal to n₀ (n₀² + 2)²/9, where n₀ is the refractive index of the material. The term 〈⁵D₀‖U^λ‖⁷F_J〉² is the square of the reduced matrix elements. The values of 〈⁵D₀‖U²‖⁷F₂〉² and 〈⁵D₀‖U⁴‖⁷F₄〉² are taken as 0.0032 and 0.0023, respectively³² by considering the coordination number around Eu³⁺ ions in the YPO₄ lattice. The values of the spontaneous emission cross-section can be calculated from the emission spectrum using the relation A_0−λ = (ν_0-1/ν_0−λ) (I_0-λ/I₀₋₁) A_0-1,³¹ where ν_0−λ and I_0−λ represents, respectively, the frequency and intensity corresponding to the respective transition in the emission spectrum. A₀₋₁ is taken as a constant^31,33 and is equal to 50 s⁻¹. As the matrix elements U⁽⁴⁾, U⁽⁶⁾ and U⁽²⁾, U⁽⁶⁾ for the ⁵D₀ → ⁷F₂ and ⁵D₀ → ⁷F₄ transitions, respectively, are very small (negligible), the values of Ω₂ and Ω₄ can be calculated from eqn (1) by determining A_0–λ directly from the emission spectrum. The sum of all the A_0−λ values gives the radiative decay rate (A_rad) and the inverse of the experimentally measured lifetime (1/τ_measured) gives the sum of non-radiative and radiative decay rates (A_rad + A_nrad). The Ω_λ values along with the radiative and non-radiative decay rates and quantum yields (A_rad/(A_rad + A_nrad)) calculated by the above procedure are given in Table 2. From this table it is clear that the Ω₄ components are higher for all the three samples as compared to Ω₂ values, indicating that the higher rank components of the crystal field influence more than the lower rank components under such co-ordination environments around Eu³⁺ ions. The values of Ω₂ and Ω₄ and the radiative rates are comparable for all samples, indicating that the extent of polarizability around Eu³⁺ ion is the same in all samples. However the non-radiative rate for CePO₄@Y_0.95Eu_0.05PO₄ sample is significantly higher as compared to Y_0.95Eu_0.05PO₄@CePO₄ and Y_0.95Eu_0.05PO₄ samples. This is also reflected in the quantum yield values. These results indicate that the Eu³⁺ ions in YPO₄ host is only luminescent and the remaining Eu³⁺ ions which have occupied the CePO₄ lattice, undergo significant quenching due to electron exchange taking place between Ce³⁺/Ce⁴⁺ and Eu³⁺/Eu²⁺ species present in the sample. Hence, based on the above inferences it is confirmed that a significant extent of Eu³⁺ ions get incorporated into the CePO₄ lattice of CePO₄@Y_0.95Eu_0.05PO₄ samples.

Table 2 Radiative (A_rad), non-radiative (A_nrad), total (A_total) rates of the ⁵D₀ level of Eu³⁺ ions along with experimental intensity parameters Ω₂ and Ω₄ for different samples

Compounds	A rad/s^−1	A nrad/s^−1	A total/s^−1	Ω 2/10^−20 cm^2	Ω 4/10^−20 cm^2	Quantum yields (%)
Y0.95Eu0.05PO4	257	156	413	1.7	4.9	62
Y0.95Eu0.05 PO4@CePO4	280	202	482	1.7	5.7	58
CePO4@Y0.95 Eu0.05PO4	242	3612	3854	1.5	4.0	6

---

4. Conclusions

Based on XRD, Raman and TEM studies it has been concluded that in CePO₄@Y_0.95Eu_0.05PO₄ sample, Y³⁺/Eu³⁺ ions get incorporated in the CePO₄ lattice due to ion exchange process taking place during the synthesis stage. Unlike this, such ion exchange is not taking place with Y_0.95Eu_0.05PO₄@CePO₄ sample and the lower solubility product values of CePO₄ compared to YPO₄ is the reason responsible for this. The asymmetric ratio of luminescence, ⁵D₀ lifetime values of Eu³⁺ ions, along with the excited state lifetime values of Ce³⁺ in these samples further support the incorporation of Y³⁺/Eu³⁺ in the CePO₄ host of CePO₄@Y_0.95Eu_0.05PO₄ sample and the existence of separate CePO₄ and Y_0.95Eu_.05 PO₄ phases in Y_0.95Eu_0.05PO₄@CePO₄ sample. The calculated values of non-radiative decay rates and quantum yields from the emission spectrum of these samples also confirm the significant incorporation of Eu³⁺ in the CePO₄ phase of CePO₄@Y_0.95Eu_0.05PO₄ sample. The poor Eu³⁺ luminescence from CePO₄@Y_0.95Eu_0.05PO₄ sample is due to the electron exchange taking place between the Ce³⁺/Ce⁴⁺ and Eu³⁺/Eu²⁺ species present in the sample.

Acknowledgements

The authors thank Dr A. K. Suri, Director, Materials Group, BARC, Dr T. Mukherjee, Director, Chemistry Group, BARC, Dr D. Das, Head, Chemistry Division, BARC, Dr I. G. Sharma, Head, Materials Processing Division, BARC and Dr N. Krishnamurthy, Head, Advanced Ceramics Section, BARC, for their encouragement during this work.

References

1. G. Y. Adachi and N. Imanaka, Chem. Rev., 1998, 98, 1479–514.
2. A. H. Peruski, L. H. Johnson and L. F. Peruski, J. Immunol. Methods, 2002, 263, 35–41.
3. J. A. Capobianco, J. C. Boyer, F. Vetrone, M. A. Speghini and M. Bettinelli, Chem. Mater., 2002, 14, 2915–21.
4. Y. Guo, P. Woznicki, A. Barkatt, E. E. Saad and I. G. Talmy, J. Mater. Res., 1996, 11, 639–649.
5. F. H. Firsching and S. N. Brune, J. Chem. Eng. Data, 1991, 36, 93–95.
6. A. Rouanel, J. J. Serra, K. Allaf and V. P. Orlovskii, Inorg. Mater., 1981, 17, 76–81.
7. R. Gratz and W. Heinrich, Am. Mineral., 1997, 82, 772–780.
8. J. B. Davis, J. P. A. Lofvander, A. G. Evans, E. Bischoff and M. L. Emiliani, J. Am. Ceram. Soc., 1993, 76, 1249–1257.
9. D. B. Marshall, P. E. D. Morgan, R. M. Houley and J. T. Cheung, J. Am. Ceram. Soc., 2005, 81, 951–956.
10. R. J. Kerans, R. S. Hay, T. A. Parthasarathy and M. K. Cinibulk, J. Am. Ceram. Soc., 2004, 85, 2599–2632.
11. Y. Hikichi, T. Ota, K. Daimon, H. Keiji, T. Hattori and M. Mizuno, J. Am. Ceram. Soc., 2005, 81, 2216–2218.
12. Y. Hikichi and T. Nomura, J. Am. Ceram. Soc., 1987, 70, C252–C253.
13. E. C. Subbarao, D. K. Agrawal, H. A. McKinstry, C. W. Sallese and R. Roy, J. Am. Ceram. Soc., 1990, 73, 1246–1252.
14. M. Cao, C. Hu, Q. Wu, C. Guo, Y. Qi and E. Wang, Nanotechnology, 2005, 16, 282–286.
15. L. Zhu, X. Liu, Q. Li, J. Li, S. Zhang, J. Meng and X. Cao, Nanotechnology, 2006, 17, 4217–4222.
16. M. Guan, J. Sun, M. Han, Z. Xu, F. Tao, G. Yin, X. Wei, J. Zhu and X. Jiang, Nanotechnology, 2007, 18, 415602–7.
17. K. Kompe, H. Borchert, J. Storz, A. Lobo, S. Adam, T. Möller and M. Hasse, Angew. Chem., Int. Ed., 2003, 42, 5513–5516.
18. V. Buissette, M. Moreau, T. Gacoin and J. P. Boilot, Adv. Funct. Mater., 2006, 16, 351–355.
19. C. Dong and F. C. J. M. van Veggel, ACS Nano, 2009, 3, 123–130.
20. K. Srinivasu, R. S. Ningthoujam, V. Sudarsan, R. K. Vatsa, A. K. Tyagi, P. Srinivasu and A. Vinu, J. Nanosci. Nanotechnol., 2009, 9, 3034–3039.
21. A. K. Gulnar, V. Sudarsan, R. K. Vatsa, A. K. Tyagi, R. C. Hubli, U. K. Gautam and A. Vinu, Cryst. Growth Des., 2009, 9, 2451–2456.
22. Y. Ni, J. M. Hughes and A. N. Mariano, Am. Miner., 1995, 80, 21–26.
23. PC-PDF Card No. 84–0335.
24. PC-PDF Card No. 84–0247.
25. R. D. Shanon, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1976, 32, 751–767.
26. Z. S. Cetiner, S. A. Wood and C. H. Gammons, Chem. Geol., 2005, 217, 147–169.
27. E. Jonathan, S. Richard, D. Robinson, F. Zhang, S. W. Chan and I. P. Herman, Phys. Rev. B: Condens. Matter Mater. Phys., 2001, 64, 245407–8.
28. D. F. Mullica, D. A. Grossie and L. A. Boatner, J. Solid State Chem., 1985, 58, 71–77.
29. G. Blasse and B. C. Grabmaier, Luminescent Materials, Springer-Verlag, Berlin, 1994, p. 87–88.
30. P. Dorenbos, in Phosphor hand book, ed. W. M. Yen, S. Shionoya and H. Yamamoto, CRC press, Boca Raton, 2007, pp. 142–143.
31. D. Hreniak, W. Strek, J. Amami, Y. Guyot, G. Boulon, C. Goutaudier and R. Pazik, J. Alloys Compd., 2004, 380, 348–351.
32. W. T. Carnall, H. Crosswhite and H. M. Crosswhite, Energy Structure and Transition Probabilities of the Trivalent Lanthanides in LaF₃, Argonne National Laboratory Report, USA, unnumbered, 1977.
33. C. A. Kodaira, H. F. Brito, O. L. Malta and O. A. Serra, J. Lumin., 2003, 101, 11–21.

---

Footnote

† Electronic supplementary information (ESI) available: TEM images and SAED patterns from (a) CePO₄ nanorods and (b) Y_0.95Eu_0.05PO₄ nanoparticles. See DOI: 10.1039/c0nr00334d

---