Structural and Luminescence Behavior of Nanocrystalline Orthophosphate KMeY(PO₄)₂: Eu³⁺ (Me = Ca, Sr) Synthesized by Hydrothermal Method

Abstract

KMeY(PO₄)₂:5% Eu³⁺ phosphates have been synthesized by a novel hydrothermal method. Spectroscopic, structural, and morphological properties of the obtained samples were investigated by X-ray, TEM, Raman, infrared, absorption, and luminescence studies. The microscopic analysis of the obtained samples showed that the mean diameter of synthesized crystals was about 15 nm. The KCaY(PO₄)₂ and KSrY(PO₄)₂ compounds were isostructural and they crystallized in a rhabdophane-type hexagonal structure with the unit-cell parameters a = b ≈ 6.90 Å, c ≈ 6.34 Å, and a = b ≈ 7.00 Å, c ≈ 6.42 Å for the Ca and Sr compound, respectively. Spectroscopic investigations showed intense ⁵D₀ → ⁷F₄ transitions connected with D₂ site symmetry of Eu³⁺ ions. Furthermore, for the sample annealed at 500 °C, europium ions were located in two optical sites, on the surface of grains and in the bulk. Thermal treatment of powders at high temperature provided better grain crystallinity and only one position of dopant in the crystalline structure. The most intense emission was possessed by the KSrY(PO₄)₂:5% Eu³⁺ sample calcinated at 500 °C.

1. Introduction

In recent times, inorganic phosphors doped with lanthanide ions have attracted much attention due to their potential applications as phosphors, laser materials, or solid-state lighting [1,2,3,4,5,6]. The current white-light-emitting diodes (W-LEDs) suffer from a lack of red light, therefore, red phosphors are sought to fill deficiency. Generally, W-LEDs are fabricated by combining the yellow emission of YAG: Ce³⁺ with the blue light of the diode [7]. The second method is to mix the blue, green, and red emissions generated from three or two blended powders. Rare-earth-doped (RE) phosphates are a good candidate for the formation of white LEDs due to their good physical and chemical stability, low cost of substrates, and reduced preparation temperatures compared to silicates, aluminosilicates, or nitrates [8,9,10,11,12,13].

One of the most promising phosphors are KMeRE(PO₄)₂ phosphates, where Me = Ca or Sr, and have a hexagonal structure with the unit cell parameters a = b ≈ 7.0 and c ≈ 6.4 Å [14,15,16,17,18]. The structure is closely related to rhabdophane (REPO₄·nH₂O), in which half of the cationic sites of RE³⁺ are filled by Me²⁺ (Figure 1). Simultaneously, the electrostatic charge is compensated by the placement of K⁺ in free spaces inside of channels running parallel to the c axis and formed by (RE, Ca)-PO₄-(RE, Ca) chains [14]. According to the available literature, the simultaneous replacement of RE³⁺ with K⁺ and Ca²⁺ reduces the group symmetry of PO₄ and RO₈ (R = Ca or RE) from D₂ to C₂ [17].

The KMeRE(PO₄)₂ phosphates show good luminescence and hence can be applied as efficient luminophores [19,20]. Recently, a series of KMeRE(PO₄)₂ (Me = Ca, Sr; RE = Y, La, Lu, Gd) have been investigated carefully [4,21,22,23,24]. Wang et al. successfully synthesized KCaY(PO₄)₂:Dy³⁺, Eu³⁺ as a nanosized phosphor-generated warm-white light with a high color rendering index (CRI) [23]. Basic information about KCaY(PO₄)₂:Eu³⁺ has been reported in 2006 by Wang et al. [24]. Furthermore, the energy transfer from Eu²⁺ to Mn²⁺ in KCaY(PO₄)₂ was studied by Liu and co-workers [4]. Up until now, strontium ternary phosphates were activated by RE ions, such as Ce³⁺, Tb³⁺, Eu³⁺, and Mn²⁺ [21,22]. Samples doped with Ce³⁺, Tb³⁺, and Mn²⁺ exhibited high quantum efficiency (60%) and were obtained by a Pechini method [22]. Compounds are mostly obtained via a conventional ceramic method (sintering at high temperatures), however, results presented in [25,26,27] show that obtaining single-phase KMeRE(PO₄)₂ is difficult because of the limited thermal stability of such ternary phosphates at temperatures exceeding 1100–1200 °C. A relatively mild hydrothermal process (short-term heating carried out at 240 °C) has been applied for the synthesis of phase-pure KCaCe(PO₄)₂, which resulted in obtaining nanocrystalline powders [27].

The aim of the presented study was to describe the structural and luminescence properties of nanocrystalline KMeY(PO₄)₂ (Me = Ca, Sr) powders doped with Eu³⁺ and to discuss the obtained results from the point of view of the powders’ morphology.

2. Materials and Methods

The powders of KMeY(PO₄)₂:5% Eu³⁺ (Me = Ca or Sr) were obtained by a short-lasting hydrothermal technique. In the first step, Y₂O₃ (99.99% POCh, Gliwice, Poland) and Eu₂O₃ (99.9%, POCh, Gliwice, Poland) were digested in dilute nitric acid. The obtained clear solutions were dried for several days over solid NaOH until solid nitrates crystallized. The starting sols for hydrothermal processes were prepared as follows. Rare earth nitrates and Ca(NO₃)₂·4H₂O (≥99.0%, POCh, Gliwice, Poland) or Sr(NO₃)₂ (≥99.0%, Sigma-Aldrich, Steinheim, Germany) were dissolved in 10 cm³ of distilled water, then K₃PO₄·H₂O (≥94%, Fluka, Steinheim, Germany) was added to the solution, and the obtained mixture was diluted to 50 cm³ volume. The molar ratio of K:Me:Y:Eu metals in the obtained sols was 68:1:0.95:0.05. After one-day ageing, the sols were used in a hydrothermal process performed in a Magnum II autoclave (Ertec, Wrocław, Poland) with heating of the sols by microwaves of 600 W maximal power. The syntheses were controlled by temperature measurements performed at a bottom part of the reactor made of PTFE. The syntheses were carried out for 2 h at a maximal temperature of 250 °C and a pressure of 30 bar. As they were obtained, fine precipitates were centrifuged, washed several times with distilled water, and dried for 24 h at 120 °C. Further heating of the samples was carried out at 500 °C and 1000 °C for 10 h.

Phase analysis of obtained powders was performed using the XRD technique and a Siemens D5000 diffractometer (Munich, Germany) equipped with a copper radiation cathode. The measurements were performed at a 2θ angle range of 5–60° with a 0.02° step and at least 4 s per step. The indexing of powder patterns was made using the computer program DICVOL [28]. Unit cell lattice constants of KMeY(PO₄)₂:5%Eu were refined using the Checkcell software (LMGP, Grenoble, France) [29].

Simultaneous TGA/DSC measurements were carried out using a SETSYS^TM (TG-DSC 1500; Setaram, Geneva, Switzerland) in the temperature range of 20–1200 °C (sample mass 40 mg, platinum crucibles, heating rate 10 °C min⁻¹, argon atmosphere).

Excitation spectra were measured on a McPherson spectrometer (McPherson Inc. Chelmsford, MA, USA) equipped with a 150 W Xe lamp and 150 W deuterium lamp (for VUV part) as excitation sources, and a Hamamatsu R928 photomultiplier (Hamamatsu Photonics, Hamamatsu, Japan) as a detector. The normalization of the XRD and excitation spectra was performed at the highest point of the spectrum. The emission spectra were measured using a Hamamatsu PMA12 multichannel analyzer (Hamamatsu Photonics, Hamamatsu, Japan) as a detector, and 395 nm line from tunable pulsed Ti: Sapphire laser pumped by second harmonic (λ = 532 nm) of Nd: YAG laser as an excitation source (Lotis TII, Minsk, Belarus). Decay time measurements were performed using the same excitation as for emission spectra, while decay curves were collected on a Tektronix MSO-3052 digital oscilloscope (Tektronix, Beaverton, OR, USA). The low temperature measurement was carried out at 77 K using a quartz dewar.

3. Results and Discussion

3.1. Structure and Morphology

The analysis of XRD patterns of obtained phosphates confirmed the crystalline characteristics of the powders. All of the samples were isostructural to the rhabdophane-type KCaY(PO₄)₂ phosphate (ICDD card No. 51-1632) (Figure 2). The XRD pattern of dried samples showed pure structures after the hydrothermal process (Figure S1). Due to the presence of hydroxyl groups originating from zeolitic water molecules, which can easily quench luminescence, samples were annealed at 500 °C and 1000 °C. The annealing process does not change the structure of prepared phosphates. Minor reflections of a secondary phase were registered in the XRD pattern of KCaY(PO₄)₂:5%Eu³⁺ sintered at 500 °C. They most likely derive from the YPO₄ xenotime phase; however, due to the low intensity of reflections, the presence of the phase can be not confirmed with reliability. The refined hexagonal unit cell parameters of the phosphates sintered at 1000 °C are listed in

Table 1. The unit cell parameters and the cell volume of the KMeY(PO₄)₂:5% Eu³⁺ powders sintered at 1000 °C.

Sample	Unit Cell Parameters (Å)		Cell Volume (Å3)
	a = b	c
KCaY(PO4)2:5% Eu3+	6.8873 ± 0.0045	6.3380 ± 0.0004	260.4
KSrY(PO4)2:5% Eu3+	6.9911 ± 0.0015	6.4186 ± 0.0002	271.7

.

Figure 3 and Figure 4 present the TEM and SEAD representative images of KCaY(PO₄)₂:5% Eu³⁺ and KSrY(PO₄)₂:5% Eu³⁺ samples, respectively. As can be seen, the obtained compounds possess rounded grains with an average particle length size (maximum particle dimension) D[1, 0] equal to 12 ± 2 nm and 18 ± 4 nm for Ca and Sr phosphates, respectively. The powders consist of single grains, as well as bigger objects created by the clumping of a few grains. Agglomerates have an irregular shape made from connected spherical single nanocrystals. The replacement of calcium ions by the strontium one does not change the shape of obtained nanopowders.

The analysis of TGA heating curves of samples obtained hydrothermally showed a 3–4% mass loss of powders in the temperature range of 100–350 °C (Figure 5). Mass loss is characteristic of hydrated samples. A relatively high temperature of the effect suggests the presence of zeolitic-bound water. No other thermal effects were observed on TGA/DSC heating curves.

3.2. Raman and IR Studies

Figure 6a,b present infrared spectra in the middle (MIR) and far (FIR) infrared range, and Figure 7 shows the Raman spectra. The spectra measured for the strontium derivative are identical, so we decided to only show spectra for the calcium compound as the representative. They were also similar to those reported earlier by Tie et al. [17]. In general, the bands observed in these spectra agree with those reported earlier for other orthophosphates [30,31,32,33]. The particular types of vibrations appear in the typical range. The ν₃(F₂) = ν_as(PO₄) vibrations were observed at 1130 and 1090 cm⁻¹, where a strong multiplet consisting of 2–3 bands appeared. In the range characteristic for ν₁(A₁) = ν_s(PO₄) vibrations, a strong single Raman band was observed. All bending δ(PO₄) vibrations were observed in the range of 350–650 cm⁻¹, in which nine Raman and five IR bands could be identified. The IR bands of average to strong intensities in the ranges of 250–350 cm⁻¹ and 150–250 cm⁻¹ correspond to the T’(K⁺) and T’(Ca²⁺) translations, respectively. The analysis of these vibrations was performed in Reference 4 on the basis of the D₂ site symmetry of this unit, although factor group splitting was not taken into account.

According to the above-cited references [14,15,16,17,18], KMeY(PO₄)₂ phosphates (Me = Ca, Sr) crystallize in the hexagonal P6₂22 (D₆⁴) structure with Z = 3. Since the occupation of Ca²⁺, Nd³⁺, and K⁺ cations was equal to 0.5, a proper chemical formula can be written as (Ca_1.5Y_1.5)K_1.5(PO₄)₃ with Z = 1 [14].

Its primitive unit cell contains three Ca²⁺/Y³⁺ ions equally sharing the 3c (D₂ site) position, and three PO₄³⁻ ions with P and O atoms in the 3d (D₂) and 12k (C₁) positions, respectively. The K⁺ ions could not be used for calculations because their occupancy in the primitive cell was partial. For this reason, only a partial factor group analysis can be presented.

The three PO₄³⁻ ions and the Ca²⁺/Y³⁺ ions in the primitive cell give rise to 54 zone–center degrees of freedom described by the irreducible representation: Γ_v = 3A₁ + 5A₂ + 5B₁ + 5B₂ + 10E₁ + 8E₂ (54). Three of them, Γ_T = A₂ + E₁ (3), describe the acoustic phonons, and the remainder, Γ_O = 3A₁ + 4A₂ + 5B₁ + 5B₂ + 9E₁ + 8E₂ (51), correspond to the optical modes. The optical modes can be further subdivided by the vibrations of the respective components of the primitive cell, i.e., translations of Ca²⁺/Y³⁺ ions: Γ_T(Ca²⁺/Y³⁺) = B₁ + B₂ + E₁ + E₂ (6), internal vibrations of PO₄³⁻ ions: Γ_v(PO₄) = 3A₁ + 2A₂ + 2B₁ + 2B₂ + 4E₁ + 5E₂ (27), as well as translations and librations of the PO₄³⁻ ions, Γ_T/L(PO₄) = A₂ + B₁ + B₂ + 2E₁ + E₂ (6).

In the discussion of the IR and Raman spectra, the correlation between the molecular T_d, site D₂, and factor D₆ symmetry should be used. The following schemes related to the correlation T_d → D₂ → D₆: for the ν₁ mode A₁ → A → A₁ + E₂; for the ν₂ mode E → 2A → 2A₁ + 2E₂; and for the ν₃ and ν₄ modes F₂ → B₁ + B₂ + B₃ → A₂ + B₁ + B₂ + 2E₁ + E₂. Since the A₂ and E₁ modes were IR-active, A₁, E₁, and E₂ were Raman-active, and B₁ and B₂ were silent for P6₂22 symmetry, at least 13 (4A₂ + 9E₁) IR bands were expected in this spectrum, and similarly, 20 (3A₁ + 9E₁ + 8E₂) bands should be observed in the Raman spectrum. These data fit well to the number of bands observed in the IR and Raman spectra, i.e., 14 and 20, respectively (Figure 6 and Figure 7).

3.3. Optical Studies

The excitation spectra of KMeY(PO₄)₂:5% Eu³⁺ were measured by observing the transition at λ_em = 612 nm (Figure 8). The spectra consisted of a broad and intense band centered at 235 nm, which was assigned to the charge transfer transition from oxygen to europium ions (CTS—charge-transfer state) [34]. Besides the CTS, there were also sharp and narrow bands, which derived from the 4f–4f transition of Eu³⁺ [35]. The bands were assigned to the ⁷F₀ → ⁵H_3–7 at 319 nm, ⁷F₀ → ⁵D₄ at 361.5 nm, ⁷F₀ → ⁵G_2–6, ⁵L₇ at 379 nm, ⁷F₀ → ⁵L₆ at 393 nm, ⁷F₀ → ⁵D₃ at 413.5 nm, ⁷F₀ → ⁵D₂ at 464 nm, and ⁷F₀ → ⁵D₁ at 532 nm. All spectra were normalized to the intensity of the charge transfer band. In the sample sintered at 1000 °C, the ratio between the CTS and 4f–4f bands increased.

The room temperature emission spectra measured with excitation at λ = 393 nm consist of bands characteristic for Eu³⁺ ions’ emission (Figure 9a) [36]. They can be assigned to the transition from the ⁵D₀ excited level to ⁷F_J (J = 0, 1, 2, 3, 4) ground levels. The emission peak attributed to the ⁵D₀ → ⁷F₀ was located with the maximum at 578 nm, while the ⁷F₁ was at 587 nm, ⁷F₂ was at 612 nm, and ⁷F₃ and ⁷F₄ were at 653 nm and 704 nm, respectively. The highest emission intensity was exhibited by the KSrY(PO₄)₂:5% Eu³⁺ sample annealed at 500 °C (Figure 9b). The Ca²⁺ analogous phosphate possessed lower emission intensity. Furthermore, heat treatment of samples at 1000 °C decreased the overall emission intensity for both KCaY(PO₄)₂:5% Eu³⁺ and KSrY(PO₄)₂:5% Eu³⁺. This was probably connected with more efficient non-radiative processes. When the samples were annealed at a higher temperature, all Eu³⁺ ions only occupied one position in the crystal structure of the investigated phosphate. As a consequence, the distance between Eu³⁺–Eu³⁺ ions decreased causing concentration quenching. The enhancement of non-radiative processes with annealing temperatures confirmed registered decay curves (see Figures S4–S7). As shown in Figure 9a, the most intense was the electric dipole transition (ED) ⁵D₀ → ⁷F₂. However, it should the emphasized that the intensity of ⁷F₄ bands was anomalously intense. Similar behavior was reported for the first time in 1988 by Blasse [37]. He linked this situation with the D_4h site symmetry of Eu³⁺ ions. The same observation has been presented by Ferreira in Na₉[EuW₁₀O₃₆]∙14H₂O [38]. However, in the investigated phosphates, Eu³⁺ ions occupy Ca²⁺/Sr²⁺ sites with D₂ symmetry. Bettinelli in 2011 explained that the distortion of EuO₈ polyhedra to D₂ symmetry was responsible for the enhancement of the ⁵D₀ → ⁷F₄ transition [39].

No significant changes were detected in the emission spectra recorded at 77 K (Figure S2). With the temperature decreasing, emission bands become narrower; however, even at low temperatures, the peaks overlapped and they were not well separated. The ⁵D₀→⁷F₀ transition of the sample annealed at 500 °C consisted of two peaks coming from two optical positions of dopant in the host of the compounds under study. On the other hand, for samples obtained at 1000 °C, only one ⁷F₀ band was presented. The localization of Eu³⁺ ions on the surface of nanocrystals and inside the bulk can cause the occurrence of two dopant sites, and has been reported many times in the literature [40]. Presented emission and decay times also confirmed this hypothesis. The ⁵D₀ → ⁷F₀ transition of samples annealed at 500 °C consisted of two separate bands, while only one band was observed for samples prepared at 1000 °C. The magnification of the ⁵D₀ → ⁷F₀ transition of all samples is presented in Figure S3. These observations were in line with measured emission decay curves (see below).

For europium ions, the magnetic dipole transition (⁷F₀ → ⁵D₁) was almost independent of the local symmetry and type of matrix, while the electric dipole transition (⁷F₀ → ⁵D₂) was very sensitive to any change in the local environment. When europium ions occupied a site with a center of symmetry, the only allowed transition was the magnetic dipole, while for europium ions at non-centrosymmetric sites, the most intense transition was the electric dipole. Therefore, the ratio between these transitions is called the asymmetry parameter R and is defined as follows:

$$R=\frac{{\displaystyle \int {}^{5}D_0\to {}^{7}F_2}}{{\displaystyle \int {}^{5}D_0\to {}^{7}F_1}}$$

The higher the value of the R parameter, the more non-centrosymmetric the site of Eu³⁺ was. Calculated values of the R parameter are listed in

Table 2. The decay time and the R parameter for the KMeY(PO₄)₂:5% Eu samples.

Sample	τ1 (ms)	τ2 (ms)	R
KCaY(PO4)2:5% Eu3+/500 °C	1.4	5.4	1.52
KCaY(PO4)2:5% Eu3+/1000 °C	4.7	---	1.14
KSrY(PO4)2:5% Eu3+/500 °C	1.7	4.6	1.65
KSrY(PO4)2:5% Eu3+/1000 °C	3.8	---	1.35

and Figure 9c. Increasing the annealing temperature of samples lowered the R parameter, which meant that distortion of the local surrounding of Eu³⁺ ions decreased, and for KCaY(PO₄)₂:5% Eu³⁺ annealed at 1000 °C, it was close to being ideally octahedral.

Room temperature decay curves measured for all samples are presented in Figures S4–S7, while the calculated decay times are listed in Table 2. The decay curves of samples annealed at 500 °C have biexponential characteristics, which confirmed the localization of europium ions on the nanograins surface and inside the bulk. The calculated values of decay time were typical for Eu³⁺ ions [35,41,42,43]; for example, the decay time of the Eu³⁺ ions in the Y₂O₃ compound equaled 2 ms, while for YAsO₄:Eu³⁺ it was around 4 ms. Increasing the annealing temperature caused a decrease in decay time, which was in good agreement with the observed behavior of emission intensity. The enhancement of non-radiative processes with annealing temperatures confirmed registered decay curves. The lifetime of the KCaY(PO₄)₂:5% Eu³⁺ decreased from 5.4 ms to 4.7 ms with increasing annealing temperatures from 500 °C to 1000 °C, respectively.

4. Conclusions

Nanocrystalline powders of the KMeY(PO₄)₂:5% Eu³⁺ (Me = Ca, Sr) composition were obtained at relatively mild hydrothermal conditions. The XRD study confirmed the crystal structure of the investigated ternary phosphates, and TEM images showed that irregular agglomerates consisted of particles with an average size below 20 nm. Careful study of the IR and Raman spectra allowed all the vibrations, translations, and librations to be assigned. K(Ca, Sr)Y(PO₄)₂ activated with 5% of Eu³⁺ ions showed intense red emission, with a dominance of electric dipole transitions from the ⁵D₀ to ⁷F₂ level. The local symmetry of Eu³⁺ ions in the investigated phosphates KMeY(PO₄)₂:5% Eu³⁺ (Me = Ca, Sr) was D₂, which caused the abnormally high intensity of the ⁵D₀ → ⁷F₄ transition. For samples annealed at the temperature of 500 °C, Eu³⁺ ions were on the surface of the nanocrystals and inside the bulk. Increasing the annealing temperature to 1000 °C resulted in better crystallinity of the sample and only one location of doping ions. Due to this, the distance between Eu³⁺–Eu³⁺ ions was reduced, which caused concentration quenching. For this reason, the most intense emission was observed for the samples heat-treated at 500 °C. These results were consistent with the recorded decay times.