1. Introduction

As part of an ongoing study on metal aluminate phosphors activated with Eu²⁺ we have investigated the luminescence of the phosphor series Ba_1-xSr_xAl₂O₄, Sr_1-xCa_xAl₂O₄ and Ba_1-xCa_xAl₂O₄ doped with Eu²⁺. Herein we shall describe the results for Sr_1-xCa_xAl₂O₄, the results for Ba_1-xSr_xAl₂O₄ and Ba1_-xCa_xAl₂O₄ will be published separately. In consequence of their rather high photoluminescence (PL) efficiency the alkaline earth (Ca, Sr or Ba) aluminates doped with Eu²⁺ have been extensively studied [1]. When excited in the near UV, the fluorescence emission colour of SrAl₂O₄:Eu²⁺ is green, that of CaAl₂O₄:Eu²⁺ is blue and that of BaAl₂O₄:Eu²⁺ is cyan [1–7]. This colour variation is due to the sensitivity of the crystal field components of the 4f⁶5d excited state configuration of the Eu²⁺ ion in the three crystal structures.

The synthesis and luminescence properties of MAl₂O₄:Eu²⁺ (M = Ba, Ca, Mg and Sr) phosphor powders were almost simultaneously reported in 1968 by Blasse et al. [2,3] and Palilla et al. [4]. Blasse and Bril reported that the PL efficiency of these phosphors is rather high, whereas the cathodoluminescence (CL) efficiency is only 1-3% [2]. In a study on the optical and electrical properties of SrAl₂O₄:Eu²⁺ in 1971 by Abbruscato [5] it was concluded that the method of excitation of Eu²⁺ under UV- and electron beam excitation is predominantly of the charge transfer type. Abbruscato also found that the luminescence of SrAl₂O₄:Eu²⁺ increased substantially by changing the SrO:Al₂O₃ ratio from the stoichiometric value to 5 mol% excess Al₂O₃. As well as having short decay times, the luminescence decay of the Eu²⁺ doped alkaline earth aluminates, MAl₂O₄:_:Eu²⁺ (M = Ca, Sr, Ba), can also be adapted using co-activators (particularly Dy³⁺) to manifest luminescence with very long lifetimes and the same characteristic luminescence in the visible spectrum [7–12] These investigations have led to the widespread use of especially SrAl₂O₄:Eu²⁺,Dy³⁺ as a persistent luminescence material.

From the first decade of the last century onwards, alkaline earth aluminates have been intensively studied for their use in cements. These are refractory materials and were traditionally prepared by solid state reactions at high temperature (1200-1600°C) [13,14]. For phosphor applications with Eu²⁺ as dopant, the reactions need to be carried out in reducing atmospheres so that the europium activator is present as the divalent cation [1–7]. BaAl₂O₄, CaAl₂O₄ and SrAl₂O₄ have stuffed tridymite crystal structures, which can be represented as layers of rings of vertex sharing AlO₄ tetrahedra, in which Ba, Ca or Sr respectively occupy tricapped trigonal anti-prismatic cavities. The ideal undistorted structures of MAl₂O₄ (M = Ca, Sr or Ba) at high temperature for SrAl₂O₄ and BaAl₂O₄ are represented by the P6₃22 space group [15–18]. At room temperature this ideal structure is distorted, either to monoclinic structures as in the case of CaAl₂O₄ and SrAl₂O₄, or to a lower symmetry hexagonal structure (P6₃) for BaAl₂O₄. SrAl₂O₄ has a monoclinic crystal structure at ambient temperature and pressure [15], the hexagonal structure is the stable phase at a temperature > 675°C, then it is again monoclinic at even higher temperatures [16]. BaAl₂O₄ has attracted special attention due to its phase change from the ferroelectric (space group P6₃) to the paraelectric state (space group P6₃22), which takes place at about 400 K [19,20]. Aitasalo et al. [21] reported the existence of hexagonal CaAl₂O₄:Eu²⁺, which can only be prepared by a sol-gel method.

As introduced above, the efficient PL of the alkaline earth aluminates doped with Eu²⁺ has stimulated much research, especially for aluminates that contain Sr [2–12,22–27]. The primary reason for the interest in SrAl₂O₄:Eu²⁺ is its rather intense green luminescence compared to the less bright luminescence of CaAl₂O₄:Eu²⁺ and BaAl₂O4:Eu²⁺. The PL spectrum of SrAl₂O₄:Eu²⁺ at low temperature has attracted much attention since the assignment of the high energy band at 440 nm of the emission spectrum by Poort et al. [6]. There are two suggested explanations to account for the origin of this high energy band; the first is that the high energy peak (at 440 nm) can be assigned to an Eu²⁺ ion at one of the two Sr sites in monoclinic SrAl₂O₄ [6,24,26], while the main emission band at 520 nm is attributed to Eu²⁺ at the other Sr site. This explanation ignores the asymmetric character of this emission band. The second explanation acknowledges the asymmetric character of the main emission band and concludes that it consists of two components, which are attributed to the emission of Eu²⁺ ions at the two Sr sites [25,28]. Clabau et al. [25] assumed that the high energy peak at 440 nm arises from charge transfer from oxygen to residual Eu³⁺ that takes place upon UV irradiation and is associated with hole trapping at Sr²⁺ vacancies.

Figure 1 presents the composition diagram of the ternary systems CaO-SrO-Al₂O₃, which is largely based on the data presented in Shuklás thesis [29], the literature mentioned therein and the work of Ptáček [13] and Ropp [14]. In Fig. 1 the notation of the cement chemistry has been adopted, in which A stands for Al₂O₃, B stands for BaO, C stands for CaO and S stands for SrO. These abbreviations will also be used in this paper. The red lines SA-CA and BA-SA indicate the compositions that were investigated and are described herein. The compounds in Fig. 1 emphasized with a red mark that are not positioned on one of the red lines could be present as a byproduct of the all-solid state reactions, carried out in the this investigation.

 

Fig. 1. Composition diagram of CaO-SrO-Al₂O₃. The cement chemistry notation has been adopted to denote the compounds in this ternary system. The red line indicates the compositions that were studied.

Download Full Size | PDF

The stable compounds in the vicinity of the line CA-SA in Fig. 1 are: CA₂ (calcium di-aluminate or grossite), C₁₂A₇ (mayenite, C₂SA (di-calcium strontium aluminate), S₃A (tri-strontium aluminate), S₂A₃ (di-strontium tri-aluminate), S₄A₇ (tetra-strontium hepta-aluminate) and SA₂ (strontium di-aluminate), which has the similar structure as grossite (CA₂).

The nature and concentration of the crystallographic phases present in the ternary systems shown in Fig. 1 depend critically on factors such as firing temperatures, use of fluxes and firing times, calcination procedures, etc.[13,14,29]. Furthermore, phases may change as a function of temperature as in the case of SrAl₂O₄ (monoclinic-hexagonal). Literature referring to phosphor performance in relation to preparation conditions has largely concentrated on the Sr-containing aluminates [30–32]. Cordoncillo et al. [31] reported that the hexagonal and monoclinic phases were confirmed in SrAl₂O₄:Eu²⁺; after firing at 1000°C they obtained a hexagonal phase, which transformed into the monoclinic phase at lower temperatures, in agreement with the findings of Avdeev et al. [16]. Since the publications of Blasse and his co-authors [2,6] on BaAl₂O₄:Eu²⁺, it is known that the measuring conditions, especially the temperature, have a substantial effect on the spectra. The influence of the processing conditions on the luminescence of SrAl₂O₄:Eu²⁺ nanoparticles that were produced by combustion methods were also reported [32].

As mentioned above, herein we shall focus on only a limited part of the ternary system of C-S-A, namely on the compounds formed on the line CA-SA (Sr_1-xCa_xAl₂O₄, abbreviated CSA) in S-C-A. Ju et al. [33] recorded the CL spectra of CSA doped with Eu²⁺. They measured a blue shift in going from x = 0 (pure Sr) to x = 0.1 and x = 0.3. The system CSA was also studied by Prodjosantoso and Kennedy [34] using X-ray diffraction; they found monoclinic CSA at the Ca-rich side and another monoclinic phase at the Sr-rich side. At 0.2 < x < 0.5 they also found a hexagonal phase based on the structure of BaAl₂O₄. Pöllmann and Kaden [35] also found 3 phases in CSA upon changing the ratio between Ca and Sr; their results confirmed largely the work of Prodjosantoso and Kennedy. Recently, the PL spectra of the system CSA doped with 5 mol% Eu²⁺ and 5 mol% Dy³⁺ have been published by Xie et al. [36]. The phases they found largely confirmed the results of Prodjosantoso and Kennedy and Pöllmann and Kaden.

In view of the scarcity of literature on the luminescent properties of CSA doped with only Eu²⁺ and the scientific debate over the origin of the two luminescence bands of SrAl₂O₄:Eu²⁺, we decided to investigate the mixed aluminates doped with Eu²⁺ in more detail.

2. Experimental

2.1. Materials

Starting materials were: strontium carbonate (Sigma Aldrich, UK, 99.9%), calcium carbonate (Sigma Aldrich, UK, 99.9%), aluminum oxide (SASOL Inc., USA), europium oxide (Ampere Industrie, France, 99.99%), and concentrated hydrochloric acid (Sigma Aldrich, UK, 37%). All materials were used as supplied without further purification. The final annealing of the powders was made in Al₂O₃ crucibles at high temperatures in H₂ (10%)/N₂ (90%) gas.

2.2. Synthesis

Solid state synthesis methods were used to prepare Sr_0.99-xEu_0.01Ca_xAl₂O₄ with x varying between 0 and 0.99 in steps of 0.1. The samples were prepared by calcining mixtures of an appropriate molar ratio of SrCO₃, CaCO₃, γ-Al₂O₃ and EuCl₃ powders in a flow of 90% N₂–10% H₂. After calcination the powders were carefully ground by ball milling (Al₂O₃) for 3 hours. The final annealing of the Sr_0.99-xEu_0.01Ca_xAl₂O₄ samples was at 1400°C under H₂/N₂ for 2 hours. In the wake of the work of Abbruscato [5] we also varied the molar ratio between SrO and Al₂O₃ in SrAl₂O₄ and found maximum PL at 5% excess Al₂O₃. For that reason it was decided to prepare all samples of Sr_0.99-xEu_0.01Ca_xAl₂O₄ with 5% excess Al₂O₃ with respect to the sum of alkaline earth moles. The molar ratio x_Ca in this article refers to the molar ratio between Ca and Sr in the compounds.

2.3. Characterisation

The crystalline phases of the products were determined by X-ray powder diffraction at room temperature using Bruke’s D8 Advance X-ray diffraction (XRD) equipment fitted with a nickel-filtered copper source, CuKα at λ=1.5406 Å, and a LynxEye silicon strip detector. Diffraction peaks were recorded at 5° < 2θ < 100° and 25°C. The diffractometer was calibrated using an Al₂O₃ line position standard from Bruker and a LaB₆ NIST SRM 660a line profile standard. Diffractograms were collected using the fired powders in a conventional holder. The emission of the X-ray source and hence the instrumental line broadening was determined by fitting the NIST standard using Bruke’s Topas (version 5) software package. Crystalline phases in the prepared samples were identified from the XRD-patterns by peak-search matching using the ICCD PDF-2 data files. The identifiable phases were refined according to the Rietveld procedures using the Topas package. The fundamental parameters approach was used. Fits were made taking into consideration the easily identified phases and taking into account how the phase ratios vary according to the observed data across the calcium / strontium range studied. Hence, phases were included in the Rietveld refinements even at low concentrations according to this procedure. R-Bragg values were fairly constant for each phase identified and R_wp values were in the range of 8 to 16.

Variable temperature X-ray powder diffraction measurements were performed on a Rigaku Oxford Diffraction (ROD), SuperNova, Dualflex, AtlasS2 X-ray diffractometer equipped with an Oxford Cryosystems Cobra plus variable temperature device. The X-ray source was a fine-focus, CuKα, Enhance microsource, (λ = 1.54184 Å) and the detector an AtlasS2 CCD. A small amount of powder was distributed on the tip of a 50 µm MiTeGen mount and X-ray powder diffraction patterns were accumulated using the “Powder Power Tool” application from the CrysAlis PRO 1.171.39.44a operating software (Rigaku OD, 2015), which controls both the data collection and data reduction process. Data were accumulated using four θ settings for the detector, two positive and two negative, with different exposure times for the high and low angle segments and merged to give a final powder pattern. Data were collected at 290 K and 100 K for undoped SrAl₂O₄.

Morphology and particle size assessment of the phosphor powders were conducted in a scanning field emission electron microscope (FESEM), Supra 35 VP (Carl Zeiss, Germany). Some samples were also investigated in a transmission electron microscope (TEM), (2100F, JEOL, Japan) equipped with a Schottky-type field emission gun. The TEM was equipped with a Vulcan CL detector, Gatan, USA, for imaging and spectroscopic purposes. This system used a Czerny–Turner spectrometer with back-illuminated CCD and gratings with 1200 or 2400 grooves per mm for collection of CL emission spectra. A small cryostat connected to the sample holder enabled cooling of the samples in the TEM down to 103 K (-170°C); adjustment of the sample temperature anywhere between 103 K and 303 K could be made.

PL excitation and emission spectra of the samples were recorded using a Bentham phosphor spectrometer system (Bentham Instruments Ltd., Reading, UK.), configured with M300 excitation and emission monochromators, which were equipped with 0.2 mm slits. The absolute wavelength calibration of this emission monochromator had maximal error of 0.4 nm; however, relative wavelength values were accurate within 0.05 nm.

3. Results

3.1. Electron microscope

The particle size of the alkaline earth aluminates after the high temperature annealing process was in all cases rather large and varied from about 1 to 6 µm. Figures 2a-2d are electron microscope images of samples after the final annealing step.

Figure 2(a) shows that crystallites sinter and form agglomerates. At temperatures >1400°C there is more sintering and the agglomerates grow in size. For this reason we have limited the annealing temperatures to 1400°C. Figs. 2(c) and 2(d) are STEM images of the particle shown in Fig. 2(b). Figure 2(d) is a panchromatic image; it indicates that the CL from phosphor particles depends on the thickness. At the edge the particle is rather thin and the CL is much weaker.

 

Fig. 2. (a) TEM image at 200 keV of SrAl₂O₄:1%Eu²⁺ annealed at 1400°C. (b) TEM image at 200 keV of CaAl₂O₄:1%Eu²⁺. (c) STEM image (high-angle annular dark-field) of CaAl₂O₄:1%Eu²⁺, same particle as in (b), slightly rotated. (d) Panchromatic image of the same particle (STEM).

Download Full Size | PDF

3.2. X-ray diffraction and crystal structure

Figures 3(a) and 3(b) present powder XRD-diffractograms of Sr_0.99-xCa_xEu_0.01Al₂O₄ (0 ≤ x ≤ 0.99). By comparing the XRD-diffraction patterns of the Sr_0.99-xCa_xEu_0.01Al₂O₄ series in Fig. 3(a) with powder diffraction files (PDF) of the compounds that are located in the vicinity of the SA-CA line in Fig. 1, we have identified five different phases in this phosphor series. These are listed in Table 1. Atomic coordinates and isotropic thermal parameters were not refined. Figure 11 in the appendix shows a typical example of the quality of the fit between the experimental a calculated XRD pattern for Ca_0.99Eu_0.01Al₂O₄. The difference between these two patterns is also indicated.

 

Fig. 3. Powder XRD-patterns of Sr_0.99-xCa_xEu_0.01Al₂O₄ for 0 ≤ x_Ca ≤ 0.99. (a) 10° ≤ 2θ ≤ 80°. (b) Enlarged pattern for the range 28° ≤ 2θ ≤ 32°. The diagrams for 0.02 ≤ x_Ca<0.1 have not been indicated.

Download Full Size | PDF

Table 1. Observed phases in Sr_0.99-xCa_xEu_0.01Al₂O₄ phosphor series

View Table | View all tables in this article

Cell parameters of compounds that are present in concentrations >10 mol % are listed in the appendix. Unlike monoclinic CA and monoclinic CA₂, which are present over a wide concentration range of Ca, the monoclinic SA and hexagonal phases can only exist at a limited Ca concentration range.

The compounds CA₂ (3) and C₁₂A₇ (5) are also known under the mineralogy names grossite and mayenite respectively. Cubic mayenite is present at high Ca content in small concentration only and is therefore not listed in the appendix. The cell constant found for this material agreed with the published value [38]. The comparison with literature data on the cell parameters in non-mixed alkaline earth aluminates is also indicated in the appendix tables (Tables 2–5). The phases 1 and 2 in Table 1, monoclinic SrAl₂O₄ and monoclinic CaAl₂O₄ respectively, are abbreviated as MCSr and MCCa in this work.

Figure 4(a) illustrates the compositions of the phases that we found in the CSA system, while Fig. 4(b) depicts the phases and compositions found by Prodjosantoso and Kennedy [34]. The eye-catching difference between Figs. 4(a) and 4(b) is the presence of large quantities of grossite, (CA₂+SA₂), in Fig. 4(a), whereas Prodjosantoso and Kennedy [34] and Pöllmann and Kade [35] did not find any grossite. As mentioned in section 2.2 we made our samples with 5% excess of Al₂O₃. We assumed that this was the main cause for the presence of grossite. This hypothesis was checked by repeating the synthesis of Sr_0.49Eu_0.01Ca_0.5Al₂O₄ without excess of Al₂O₃. In this case we did not find grossite either. From Fig. 4(a) it can be derived that the rather large quantity of Al₂O₃ stored in grossite at x_Ca=0.49 needs to be compensated by CaO + SrO in other phases to account for the mass balance. These phases were not present in crystalline form, but likely in the amorphous state. We tried to fit the back ground curve of the XRD-patterns at low 2θ-values to the quantity of amorphous material, but this procedure was unsuccessful. Nevertheless, it is assumed that CaO and SrO in amorphous state make up the mass balance.

 

Fig. 4. Composition diagram of the phases in Sr_0.99-xCa_xEu_0.01Al₂O₄ as determined from the XRD-measurements at room temperature. (a) This work. (b) Diagram based on the data published by Prodjosantoso and Kennedy [34].

Download Full Size | PDF

MCSr has two sites for the alkaline earth cations, Sr(1) and Sr(2), which are located in two different channels formed by the AlO₄ tetrahedra along the c-axis [15,16]. Both cation sites have nine nearest neighbour O^2- ions, which belong to six different AlO₄ tetrahedra. Schulze and Müller-Buschbaum [15] commented on the curious distribution of the Sr-O distances for both Sr-sites: three of these distances are > 3 Å and cannot be considered to create a strong Sr-O bond. This conclusion is based on the radii of Eu²⁺, Ca²⁺, Sr²⁺, Ba²⁺ and O^2- ions in six coordination arrangement, which are 1.31 Å, 1.14 Å, 1.32 Å, 1.49 Å and 1.26 respectively [41]. The other six Sr-O bonds have an average distance of about 2.62 Å, which is only slightly more than the sum of the O^2- and Sr²⁺ radii. Schulze and Müller-Buschbaum [15] evaluated the effective coordination numbers (N), which are 6.07 and 6.23 for Sr(1) and Sr(2) respectively. Only Schulze and Müller-Buschbaum [15] published a list with all nine Sr-O bond distances for both the Sr(1) and the Sr(2) sites in SrAl₂O₄. When comparing their results with our data, we found a perfect match for the Sr-O bond distances of Sr(1), but for Sr(2)-O distances we observed a difference that is much larger than the error bar. We used PDF number 76-7488 in the refinements, which is based on the work of Avdeev et al. [16]. We found also differences between the average Sr(2)-O distance of Dutczak et al. [24] and Clabau et al. [25] and the average Sr(2)-O distance determined in this work, namely 2.72 Å for N = 9 and 2.60 Å for N = 7, whereas our result agrees with the average Sr(2)-O distance for N = 9 published by Botterman et al. [26]. The issue for this determination of the crystal structure is (1): the structures with two Sr²⁺ cation sites and 6 independent oxygen atom positions were calculated from a small number of XRD-patterns, and (2): most of the electron density is arising from the heavier Sr²⁺ cations, so the calculated positions of the oxygens show a rather large spread. After analysis of the various crystallographic information files (CIF) in the ICSD database for monoclinic SrAl₂O₄ (P2₁), we conclude that the refinement result depends on the choice for the CIF, which leads to slightly different atomic coordinates for the oxygen atoms: in other words, this choice causes the disagreement.

The next phase at x_Ca ≥ 0.2 in Fig. 4(a) is the monoclinic phase of calcium di-aluminate: the grossite structure. This structure has one cation site, which is seven-coordinated. Although this structure has only one cation site, it is important considering where the Sr²⁺ is located, since the Eu²⁺, responsible for PL, is closer in size to Sr²⁺ than to Ca²⁺. Since the Eu²⁺ ion is slightly smaller in size than Sr²⁺, it will predominantly occupy Sr²⁺-sites and can only be forced into the smaller cation sites in the presence of an excess of Sr²⁺. From the Rietveld refinements it was concluded that the Sr²⁺ cation occupancy in grossite is constant between x = 0.3 and 0.6, while the composition in the samples did not change significantly either. It is thus to be expected that this structure will not strongly affect the luminescence in this concentration range, apart from an effect on the band broadening. Figure 4(a) indicates that MCCa is the dominant phase in Sr_0.99-xCa_xEu_0.01Al₂O₄. Its unit cell contains 12 CaAl₂O₄ formula units and has three alkaline earth cation sites, which are labelled Ca(1), Ca(2) and Ca(3), having equal multiplicity. The first two sites are six-coordinated, in which the O^2- anions belong to four different AlO₄ tetrahedra [17]. These cationic sites are located in one of the channels formed by the rings made by the AlO₄ tetrahedra. The cation site Ca(3) is different, since it is located in a second set of channels and it has nine O^2- anion nearest neighbours belonging to five different AlO₄ tetrahedra. Six of these nine O^2- anions are shared by two Ca(3) cations forming an elongated octahedron (thereby forming a continuous chain of the type Ca(3)-O-Ca(3)). The other three O^2- anions are in a plane perpendicular to this chain. We found that the Sr²⁺ cations in Sr_1-xCa_xAl₂O₄:Eu²⁺ prefer the Ca(3) sites at 0.4 < x < 0.9 and it is only around x_Ca = 0.2 that the three sites become equally populated with Sr²⁺. Both MCCa and MCSr in Sr_0.99-xCa_x Eu_0.01Al₂O₄ show a monotonous decrease of their cell volumes upon increasing x_Ca.

Figure 4(a) furthermore indicates that a hexagonal phase exists in the Sr_0.99-xCa_xEu_0.01Al₂O₄ series at 0.1 < x < 0.6. Prodjosantoso and Kennedy [34] found that the hexagonal phase in Sr_1-xCa_xAl₂O₄ was present at 0.35 < x < 0.65 with its maximum occurring at x = 0.4 as shown in Fig. 4(b). The hexagonal phase in Sr_0.99-xCa_xEu_0.01Al₂O₄ is isostructural with BaAl₂O₄: both having space group P6₃. It should be realized that BaAl₂O₄ shows a phase transition from the paraelectric state (space group P6₃22) at high temperature to the ferroelectric state (space group P6₃) by cooling down below ≈450 K [19,20]. At this phase transition the cell parameter “a” doubles from 5.22 Å to 10.44 Å [20] while the positions of the Ba and Al ions are slightly displaced [18,19]. In the case of hexagonal BA in the Ba_0.97-xSr_xEu_0.03Al₂O₄ series we actually find that a = 10.4452 Å, while in the hexagonal phase of Sr_0.99-xCa_xEu_0.01Al₂O₄ it is 10.1943 Å at x_Ca=0.27, which is about 2% smaller, “apparently” reflecting the decrease of cation size in going from Ba²⁺ to Sr²⁺. We use the adverb apparently as the change in cell parameter “a” size is in fact not as large as would be expected from the change in ionic radius from Sr²⁺ to Ba²⁺ and this must indicate that the network of connected AlO₄ tetrahedra is the dominating building block of the lattice and the cations are sitting in spaces that are not too constrictive. The assignment of very weak reflections in the XRD-pattern, from which the space group for the hexagonal phase of pure BaAl₂O₄ can be identified [18–20], is particularly difficult in the Sr_0.99-xCa_xEu_0.01Al₂O₄ samples with contaminating compounds such as mayenite and grossite present.

3.3. PL and CL spectra

In Fig. 5 the PL spectra of the phosphor series Sr_0.99-xCa_xAl₂O₄:1%Eu²⁺ have been plotted for various values of the molar fraction of one of the alkaline earth ions.

 

Fig. 5. Photoluminescence spectra of Sr_0.99-xCa_xAl₂O₄:1%Eu²⁺ at various values of x_Ca. (a) Emission spectra. (b) Excitation spectra. For clarity reasons only a limited number of spectra are shown.

Download Full Size | PDF

The PL spectra illustrated in Figs 5(a) and 5(b) agree nicely with the spectra published by Xie et al. [36]. A feature of the emission spectra in Fig. 5(a) is the asymmetric shape of the main emission band, viz. the band show a tail at the long wavelength side. This is even the case in a wavenumber representation of the spectra, which reduces the asymmetry slightly. This kind of asymmetry is usually observed in the fluorescence spectra of Eu²⁺ doped phosphors [1,24]. The standard explanation of this phenomenon is the presence of more than one site for the Eu²⁺ ion in the lattice: different sites create different electrostatic fields, so the presence of two or more sites would explain the asymmetry. However, the presence of more than one site for the Eu²⁺ does not explain that the tail is at the long wavelength side in the vast majority of Eu²⁺ doped phosphors. Another explanation of the asymmetry may be due to electron-phonon coupling, which can be described in terms of the Huang-Rhys parameter, the lattice vibrational energy and the line width of vibronic transitions [42]. From Figs. 5(a) and 5(b), using also the deconvolution of the excitation spectrum presented in Fig. 10(a) hereafter, the Stokes shift for the 520 nm band of SrAl₂O₄:Eu²⁺ can be estimated as 4200 cm^-1, which is close to the estimate (4000 cm^-1) made by Poort et al. [6] and larger than the value (3500 cm^-1) reported by Botterman et al. [26]. The Huang-Rhys parameter S is calculated to be 5.1 (assuming an average lattice vibration of 450 cm^-1) for SrAl₂O₄:Eu²⁺; then, according to the calculations presented by de Jong et al. [42], asymmetric broadening of a rare earth electronic transition will be unlikely. For this reason we decided to analyse the asymmetry of the fluorescence bands by deconvolution and to assign the profiles to Eu²⁺ ions positioned at alkaline earth sites in the lattice (whenever possible). This will be discussed in the next sections.

The emission spectra presented in Fig. 5(a) show that single alkaline earth aluminates have higher spectral radiances than aluminates with two alkaline earth ions. The spectra in Fig. 5(a) feature broadening of the emission band at mole fractions between 0.3 and 0.7. The spectra also indicate that the spectral radiance at x_Ca>0.47 is largely determined by the CaAl₂O₄ content in the samples. λ₀ in these spectra is close to 445 nm. Figure 5(a) manifests furthermore the large difference between the maximum spectral radiance at x_Ca=0 (Sr_0.99Eu_0.01Al₂O₄) and x_Ca=0.1 (Sr_0.89 Ca_0.1Eu_0.01Al₂O₄). This substantial decrease in spectral radiance occurred largely between x_Ca is 0 and 0.04 and it indicates that the luminescence of monoclinic SrAl₂O₄ is very sensitive to relatively small changes of its structure as found by Abbruscato in 1971 [5]. As mentioned above, we have confirmed Abbruscatós result [5] by synthesizing SrAl₂O₄:Eu²⁺ phosphors with various SrO/Al₂O₃ molar ratios, close to unity, and investigating their PL spectra and XRD-patterns. We found that the spectral radiance at 515 nm increased by 40% by lowering the SrO/Al₂O₃ molar ratio to 0.95, while the structure remained essentially MCSr. The spectra presented in Fig. 5(a) also show the largest colour change: from green at the Sr-rich side to deep blue at the Ca-rich side; however, the rather low spectral radiance at 0.1 < x_Ca<0.5 makes this phosphor system not particularly attractive for colour tuning in display or lighting devices. We shall describe the analyses the PL and CL spectra of the system CSA in the next two sections. The first is focussing on SrAl₂O₄:Eu²⁺ and the MCSr phases with a small quantity of Ca, the second is focussing on CaAl₂O₄:Eu²⁺ and MCCa with a small amount of Sr.

3.3.1. SrAl₂O₄:Eu²⁺

In Fig. 6 we present deconvolutions of PL-spectra of the MCSr phases of the Sr_0.99-xCa_xAl₂O₄:1%Eu²⁺ series that are presented in Fig. 5(a), viz. the spectra at x_Ca=0 and 0.1. The fitting of the deconvoluted spectrum to the experimental spectrum was carried out using a least squares algorithm in a wavenumber (cm^-1) representation as described previously [43,44]. In the deconvolutions we have taken the minimum number of profiles that gave a good fit with the experimental spectra. The radiance R of a (Gaussian) profile is the area under the curve, often indicated in the literature as integrated intensity.

The PL spectrum of SrAl₂O₄:Eu²⁺ at room temperature as shown in Fig. 6(a) can be decomposed into two Gaussian profiles, p1 and p2. From the PL spectra of SrAl₂O₄:Eu²⁺,Dy³⁺ recorded between 100 K and 500 K by Ueda et al. [45], it can be concluded that their main emission band also consisted of two profiles. The wavelengths of p1 and p2 in Fig. 6(a) are 545 nm and 513 nm respectively; the radiance R of p1 is smaller: R_p2/R_p1=1.54. Ngaruiya et al. [28] published the deconvoluted PL spectrum of SrAl₂O₄:Eu²⁺,Dy³⁺. They also found that the emission band can be described by two Gaussian profiles at λ₀= 528 nm and 567 nm respectively, which deviates from our result presented in the table inserted in Fig. 6(a). This difference is likely caused by another excitation wavelength (325 nm) and their choice for a wavelength base in executing the deconvolution. When a small quantity of Ca is introduced in the MCSr phase, a third profile emerges at the high wavenumber side as shown in Fig. 6(b). The radiance of this peak increases when increasing the quantity of Ca, however, the radiance of the p3 profile is small compared to the radiances of the p1 and p2 profiles. Figure 7 illustrates the deconvolution of a CL spectrum recorded at -169.5°C of Sr_0.99Eu_0.01Al₂O₄. The radiance (CL) of p3 is almost equal to the radiances measured by Clabau et al. [25] and Botterman et al. [26] at 100 K and excitation with UV radiation at 310 nm and 370 nm respectively.

 

Fig. 6. Deconvolution of PL spectra of Sr_0.99Eu_0.01Al₂O₄ (a) and Sr_0.89Ca_0.1Eu_0.01Al₂O₄ (b) with 2 or 3 Gaussian profiles respectively. The inserts show the wavenumber and wavelength at the maximum spectral radiance of the profiles.

Download Full Size | PDF

 

Fig. 7. (a) Deconvolution of CL spectrum of Sr_0.99Eu_0.01Al₂O₄ recorded at 200 keV and -169.5°C. The labelling of the profiles is equal to that in Fig. 6. (b) Normalized radiance of p3 versus temperature.

Download Full Size | PDF

The normalized radiance of the p3 profile has been plotted versus temperature in Fig. 7(b): at room temperature this profile has virtually disappeared, which agrees with the data of Botterman et al. [26].

In Fig. 8 the normalised radiance of p3 has been plotted versus the mole fraction of Ca in Sr_0.99-xCa_xEu_0.01Al₂O₄. The normalized radiance ρ_pi of profile pi is:

 

Fig. 8. Percentage of MCCa and normalised radiance of p3 versus x_Ca. The straight line and the curves have been fitted to the data points to guide the eye.

Download Full Size | PDF

Figure 8 provides the evidence that the p3 profile can be assigned to the emission of the MCCa phase with space group P2₁/n. This assignment is obvious at x_Ca>0.5, where the concentration of the MCCa phase is increasing, while the concentration of CA₂ (grossite) is strongly decreasing (compare Fig. 4(a)). At x_Ca<0.2 the radiance of p3 and the concentration of MCCa behave identically. Although the grossite phase (CA₂) also increases at 0 < x_Ca<0.2, we may exclude this latter phase as a candidate for the p3 profile because of the behaviour of ρ_p3 at x_Ca>0.5.

Clabau et al. [25] and Ngaruiya et al. [28] assigned the two bands p1 and p2 in Figs. 6 and 7 to Eu²⁺ ions that are positioned at the Sr(1) and Sr(2) cation sites in monoclinic SrAl₂O₄. According to Poort et al. [6], Botterman et al. [26] and Dutczak et al. [24] this is not the case: p3 and the combination p1-p2 should be assigned to Eu²⁺ at Sr(2) and Sr(1) respectively. The work of Botterman et al. presents an extensive study of the PL from SrAl₂O₄:Eu²⁺ and supports this conclusion. The conclusion has recently been confirmed in a theoretical study by Ning et al. [45]. They found that the p3 profile of SrAl₂O₄:Eu²⁺ at 440 nm is due to the 5d-4f emission of Eu²⁺ situated at the Sr(2) site. Unfortunately, these recent studies do no address Clabaús point [25]: by combining p1 and p2 profiles to one emission band, the band gets asymmetric. This asymmetry was not considered in the studies mentioned above. Ueda et al. [46] and Nazarov et al. [47] explained the disappearance of p3 at high temperature by energy transfer (down conversion) from the Eu²⁺ at the Sr(1) to Eu²⁺ at Sr(2). These Eu²⁺ ions differ in preferential orientation of the 5d orbitals of Eu²⁺ at Sr(1) and Sr(2). Nevertheless, we agree with the conclusion of Nazarov et al. that there is still no satisfactory explanation in the literature to account for the origin of the p3 profile.

Based on the similarity of the PL and CL measurements presented in Figs. 6(b) and 7(b) respectively, here we would like to discuss an alternative mechanism, namely a phase transition of the MCSr structure to MCCa at low temperature. The three alkaline earth sites in MCCa, viz. Ca(1), Ca(2) and Ca(3), when occupied by Eu²⁺ ions, explain the occurrence of p1, p2 and p3 at low temperature in SrAl₂O₄:Eu²⁺ and at small Ca²⁺ mole fractions in Sr_1-xCa_xAl₂O₄:Eu²⁺. Hence, what occurs in SrAl₂O₄ at high temperature, a transition from the monoclinic to the hexagonal phase, takes place in reversed sense at low temperature: transition from the MCSr phase (space group P2₁) to the slightly more distorted MCCa phase (space group P2₁/n) with three cation sites. This phase transition does not occur at one temperature but takes place gradually: upon lowering the temperature the MCCa quantity will increase, which explains the increase of the p3 signal at about 440 nm, being the emission maximum of CaAl₂O₄:Eu²⁺ (Fig. 5(a)). As mentioned above, XRD-data on SrAl₂O₄ at cryogenic temperatures have not been found in the literature. Nevertheless, we found support for the alternative hypothesis presented above by comparing the excitation spectra of CaAl₂O₄:Eu²⁺,Dy³⁺ and SrAl₂O₄:Eu²⁺,Dy³⁺ at cryogenic temperatures by Ueda et al. [48] and Botterman et al. [26] respectively. The excitation spectrum of the latter compound, when monitored at λ₀ of the p3 profile (440 nm), is almost similar to the excitation spectrum of CaAl₂O₄:Eu²⁺,Dy³⁺ monitored at 440 nm. A second consideration refers to the measurement of the decay time of the green and blue bands of SrAl₂O₄:Eu published by Botterman et al. [26]. These authors found that the decay curve of the green emission band can be described with two exponentials. This does not contradict with the alternative presented here: their result could also be explained in terms of a double emission centre with Eu²⁺ ions at the Sr(1) and Sr(2) sites. Botterman et al. [26] and Bierwagen et al. [49] determined the 1/e-value of the decay curve of the blue band at 0.1-0.2 µs at room temperature, which is (almost) equal to the decay constant of the emission band of CaAl₂O₄:Eu²⁺ [1]. Finally, Bierwagen et al. presented an energy transfer model from Eu²⁺ positioned at Sr(2) (blue emission) to Eu²⁺ at the Sr(1) site (green emission). This model does not easily comply with our alternative hypothesis; however, on the other hand it is not obvious either to explain the increase of the p3 profile by doping SrAl₂O₄:Eu with a small quantity of Ca in terms of an energy transfer model. It is worth commenting on the fact that the increase of the p3 profile by doping SrAl₂O₄:Eu with a small quantity of Ca as reported herein is contrary to what would have been expected by an energy transfer model. Such a model predicts better energy transfer if the lattice decreased in size as the Eu²⁺ cations would be closer together, which is the case here, so it would not be expected that the p3 band would be observed at room temperature. The fact that the p3 profile is seen (Fig. 6(b)) we take as evidence that a smaller lattice stabilizes the small proportion of the low temperature phase (MCCa phase) so that it does not completely convert back at higher temperature. This explanation would be in keeping with the fact that the phase at room temperature switches over to the MCCa phase at just a little higher Ca concentration. The results on the 440 nm luminescence

A second issue is the discrepancy between the alternative hypothesis, presented above, and the assignment of the 440 nm band to Eu²⁺ at Sr(2) by Ning et al. [45] based on a multi-configurational ab initio study. It would be interesting to extend these calculations to Eu²⁺ positioned at the three alkaline earth sites in the P2₁/n lattice (CaAl₂O4 lattice) for SrAl₂O₄. Although the alternative hypothesis put forward here cannot (elegantly) explain the results of Bierwagen et al. [49], we suggest that XRD-analyses of SrAl₂O₄ at low temperatures should be done to resolve the issues discussed above.

It is clear that the hypothesis about the nature of the low temperature p3 band in SrAl₂O₄ presented above can easily be tested by XRD at low temperature. We have recorded XRD-spectra of undoped SrAl₂O₄ annealed at 1350°C in H₂/N₂ at 100 K and 290 K. The result was that there was hardly any difference between these patterns and the conclusion must be that the hypothesis regarding SrAl₂O₄ without Ca has been shown to be false. This conclusion does not affect the result presented in Fig. 8, which presents a fair correspondence between the MCCa content and the radiance of the p3 profile in the Sr_0.99-xCa_xEu_0.01Al₂O₄ compounds. In other words, although the similarity between the p3 profiles in Figs. 6(b) and 7(a) is striking, the nature of the two bands at 440 nm is different.

In a recent article about the luminescence of undoped SrAl₂O₄ Vitola et al. [50] attribute the 440 nm low temperature band to an F-centre as present in α-Al₂O₃. We believe that this is a more reasonable explanation and that as the temperature increases the energy in the F-centres transfers to the Eu²⁺ cations, where it is emitted as part of the emission bands. We will present this in more detail in another paper. We have previously studied energy transfer from intrinsic emission bands in Y₂O₃ caused by cathodoluminescence to Eu³⁺ cations depending on the concentration of the rare earth cation and believe a similar phenomenon is taking place here [51,52]. From this previous work it is perfectly reasonable to see emission from two different phenomena in the same material depending on temperature. In fact, the quantitative analysis of Bierwagen et al. [49] on the energy transfer in SrAl₂O₄:Eu²⁺ is equally valid for the transference of energy from an F-centre (at 440 nm) to the Eu²⁺ cations (at 520 nm). Moreover we note that the lack of an isosbestic point in the temperature dependence of the spectra in the paper of Bierwagen et al. [49] is evidence that more than two species are present and this would fit with our interpretation of an F centre and two Eu²⁺ cation sites. An analysis of the occurrence of isosbestic points in the emission spectra of solids was published by Greger et al. [53]. We note that the absence of an isosbestic point was not addressed [49]. An examination of the data in Table 1 in the Bierwagen et al. [49] paper shows that the lifetime of the blue band decreases as the Eu²⁺ concentration increases; a fact that was not commented on either in the paper of Bierwagen et al. This is exactly what would be expected if energy were transferring from an F-centre to Eu²⁺ cations. The F-centre itself would have a slow decay [54], but as the energy is transferred the decay would be measured faster and would depend on the number of Eu²⁺ cations present to which the energy could be transferred.

Further evidence for our assignment of the optical emission arises by considering the implications of the paper written by Denault et al. [55] for our results. A reviewer kindly suggested we should consider this paper. This paper considers “Average and Local Structure, Debye Temperature, and Structural Rigidity in Some Oxide Compounds Related to Phosphor Hosts”; one of the structures they studied was SrAl₂O_4. They refer to average Sr-0 bond lengths for Sr(1) of 2.680 Å assuming the site is 7 co-ordinate and for Sr(2) the average is 2.749 Å assuming this site is 8 co-ordinate. These values are larger than those we discussed in section 3.2 of this paper but are based on different coordination numbers and so are in reasonable agreement with our earlier discussion. More significantly Denault et al. analysed the Debye temperature of the lattice from a number of different methods and considered it relative to both local structure and longer range structure. They found that SrAl₂O₄ has a high Debye temperature and commented that the higher the value the more rigid the lattice. This is in keeping with the fact that this structure has all its AlO₄ tetrahedra fully three-dimensionally connected by the oxygen atoms. The authors stated that the higher degree of connectivity in SrAl₂O₄ leads to the higher Debye temperature observed. The results of this study provide evidence that the two Sr sites in the SrAl₂O₄ would not be expected to have very different thermal behaviour, and thus do not lend support to suggestions that the thermal dependence of emission from these two sites should be very different. In other words, the substantial differences between the behaviour of the 440 nm band and the main band at 520 nm in SrAl₂O₄ might be explained in terms of originating from different independent mechanisms for these bands.

3.3.2. CaAl₂O₄:Eu²⁺

In Fig. 9 we present deconvolutions of PL and CL spectra of CaAl₂O₄:Eu²⁺ and two samples of Sr_1-xCa_xAl₂O₄:Eu²⁺.

 

Fig. 9. Deconvolutions of PL and CL spectra of CaAl₂O₄:Eu²⁺ and CSA with three (or two) Gaussian profiles. The inserts show the profiles q0 and q1 at an enlarged scale. (a) PL spectrum of Ca_0.97Al₂O₄:Eu_0.03, exc. 340 nm. (b) CL spectrum Ca_0.97Al₂O₄:Eu_0.03, recorded at -169°C and 200 keV. (c) PL spectrum of Sr_0.49Ca_0.5Al₂O₄:Eu_0.01, exc. 365 nm. (d) PL spectrum of Sr_0.19Ca_0.8 Al₂O₄:Eu_0.01, exc. 365 nm.

Download Full Size | PDF

The main emission band shown in Figs. 9(a), 9(b) and 9(d) is asymmetric and cannot be represented by a single Gaussian profile. Two Gaussian profiles can be fitted rather well to the main emission bands: these have about equal radiance and are close (0.1 ± 0.02 eV) to each other. However, the main band of Sr_0.49Ca_0.5Al₂O₄:Eu_0.01 in Fig. 9(c) can be well described by a single Gaussian profile. This latter sample consists predominantly of two phases, grossite and MCCa as indicated in Fig. 4(a): this could be one of the reasons that the asymmetry in the MCCa band has disappeared. The other reason, which seems more likely, is that the occupancy of Sr for the Ca(3) site at this mole fraction of Ca is much larger than the occupancies for the other two sites. Since Eu²⁺ has about the same ion size as Sr²⁺, we assume that this symmetric emission band indicates the emission from Eu²⁺ ions situated at Ca(3) in MCCa. The spectra illustrated in Figs. 9(a) and 9(b), which refer to CaAl₂O₄:Eu²⁺ without Sr, have been fitted with three Gaussian profiles respectively. In analysing the CL spectrum at -169°C of Fig. 9(b), we could not accurately determine the profile of q1: the insert of Fig. 9(b) indicates that there must be an emission signal at about 18500 cm^-1. The profile q0 in the CL spectra gradually disappeared upon increasing the temperature to 25°C. The q2 and q3 profiles are assigned to the Eu²⁺ ions at the Ca(3) site in CaAl₂O₄. As discussed in section 3.2, the Ca(3) site in the P2₁/n lattice of CaAl₂O₄ is roomy and can easily lodge the Eu²⁺ ion. From the asymmetry of the 440 nm band it must be concluded that there are two positions for the Eu²⁺ ions at the Ca(3) lattice site. As the radiances of q2 and q3 are almost equal, it is likely that these two Ca(3) sites have equal probabilities of being occupied. This conclusion includes that CaAl₂O₄ would be ferroelectric as in the case of BaAl₂O₄.

The q1 profile is attributed to Eu²⁺ situated at the Ca(2) and Ca(1) sites, while q0 is assigned to europium trapped exciton (ETE) emission [46], which explains that this emission band disappears at room temperature. Ueda et al. [48] studied the PL from CaAl₂O₄:Eu²⁺ at various temperatures and found the profiles q0 and q1 at 650 nm and 540 nm respectively. These values are somewhat larger than our results. A second difference is that their measurements indicated that the luminescence of q1 (540 nm) at room temperature is virtually zero. They also assigned the main emission at 440 nm to Eu²⁺ ions on the Ca(3) sites, but they did not consider the asymmetry of this emission peak. In contrast to the assignment of the 540 nm band by Ueda et al. [48] (who assigned it solely to Ca(2)); here it is assigned to both the Ca(1) and Ca(2) sites as they are both six-coordinated and only a very small amount of Eu²⁺ ions occupy these lattice sites. Finally it should be pointed out that the ETE emission does not need to be connected to one particular Eu²⁺ ion, but may rather be generated at the various sites, where Eu²⁺ ions are sitting, again in contrast to the suggestion of Ueda et al. [48].

3.4. Excitation spectra

We shall now briefly describe an analysis of the excitation spectra shown in Fig. 5(b) of the Sr_0.99-xCa_xEu_0.01Al₂O₄ series: Fig. 10 manifests deconvolutions for a selection of excitation spectra of this system. These deconvolutions are necessary to obtain accurate values for the Stokes shift of electronic transitions, as described in the beginning of section 3.3. Nazarov et al. [47] have assigned the p4 profile in Fig. 10(a) to a charge transfer transition. Since this band persists in Figs. 10(b), 10(c) and 10(d), we assign p4 in these figures also to charge transfer. Figs. 10 show that the largest changes in the spectra occur between 10(a), 10(b) and 10(c). These changes refer largely to the p1 and p2 bands, while the p3 and p4 bands show only moderate changes. The disappearance of p1 in Fig. 10(c) may be caused by changing the monitoring wavelength from 520 nm to 440 nm, because in the excitation spectrum at x_Ca=0.3 monitored at 520 nm the p1 band is still present (not shown). According to Fig. 4(a) we have only monoclinic SrAl₂O₄:Eu²⁺ at x_Ca=0, and at x_Ca=0.2 and x_Ca=0.4 we have three phases: grossite, hexagonal Sr_0.99-xCa_xAl₂O₄:Eu²⁺ and MCCa. We assume that p1 in Fig. 10(a) is attributed to the 4f→5d transition of Eu²⁺ in monoclinic SrAl₂O₄:Eu²⁺, whereas in Fig. 10(b) it is attributed to the 4f→5d transition of Eu²⁺ in the hexagonal phase.

 

Fig. 10. Deconvolution of excitation spectra of Sr_0.99-xCa_xAl₂O₄:1%Eu²⁺ with three or four Gaussian profiles. (a) Sr_0.99Eu_0.01Al₂O₄. (b) Sr_0.79Ca_0.2Eu_0.01Al₂O₄. (c) Sr_0.59Ca_0.4Eu_0.01Al₂O₄. (d) Ca_0.99Eu_0.01Al₂O₄. Note the difference of the abscissae between (a) and (b) on the one hand, and (c) and (d) on the other hand.

Download Full Size | PDF

This assumption is also based on the fact that monoclinic SrAl₂O₄ transfers to the hexagonal phase at 670°C [16]. In Fig. 10(d) the p2 and p3 profiles are assigned to the 4f→5d transition of Eu²⁺ in MCCa; this assignment refers also to Fig. 10(c) because the monitoring wavelength for this spectrum is at 440 nm, the emission maximum of CaAl₂O₄:Eu²⁺. Possible contributions from grossite and hexagonal Sr_0.99-xCa_xAl₂O₄:Eu²⁺ are likely too small to be distinguished in the spectrum of Fig. 10(c). The profiles p2 and p3 in Fig. 10(b) are attributed to the grossite and hexagonal phases in Sr_0.99-xCa_xAl₂O₄:Eu²⁺. Because of the complex character of excitation spectra due to the overlap between CT-bands and 4f-5d transitions, the above consideration is necessarily tentative.

4. Conclusions

Studies on the phases present and their luminescence in the solid state syntheses of the aluminate series CSA doped with Eu²⁺ yielded some expected and unexpected results. In the category of expected results we found and assigned a large number of phases, whereas amongst the unexpected was the finding of a large quantity of grossite. As discussed above, it is assumed herein that the rather large concentration of grossite found in the CSA samples is caused by the presence of the 5% excess of Al₂O₃ with respect to SrO + CaO. Evidence for this arises from the fact that when we did not use excess Al₂O₃ we did not find the presence of grossite.

The analyses of the PL spectra were largely based on deconvolutions of the emission bands. From the results reported here it can be concluded that this analysis technique is rather powerful. All deconvolutions of the spectra have been executed using a wavenumber base, because line broadening mechanisms are largely energy-related. Many deconvolutions of Eu²⁺ emission bands published in the literature are based on wavelength, which yield deviating parameters, particularly for ν₀ and FWHM, for profiles with FWHM > 1000 cm^-1 (>0.12 eV) in the visible range. Hence, in citing and comparing data on deconvoluted spectra from the literature, attention must be paid to the deconvolution algorithm.

From the analyses of the spectra of SrAl₂O₄:Eu²⁺ at low temperature and with small Ca additions, herein we have discussed an alternative mechanism for the origin of the 440 nm emission band observed in SrAl₂O₄:Eu²⁺ at low temperatures, namely that it is due to a phase transition that goes from monoclinic SrAl₂O₄:Eu²⁺ with space group P2₁ to monoclinic SrAl₂O₄:Eu²⁺ with space group P2₁/n, which is the space group of CaAl₂O₄. This phase transition was assumed to take place in SrAl₂O₄:Eu²⁺ upon lowering the temperature. From XRD-measurements of SrAl₂O₄ at low temperature it was concluded that this alternative mechanism cannot be true: apart from the explanations in the literature, which do not account for the asymmetry of the 515 nm band of SrAl₂O₄, we believe that the assignment of the low temperature 440 nm band to an alumina F-centre is the most viable explanation. We intend to investigate this latter hypothesis in more detail and report this work in a future paper.

We also propose herein a new assignment of the bands in the spectrum of CaAl₂O₄:Eu²⁺ based on the asymmetry of the main emission band that was ignored by previous researchers. From this latter asymmetry it also concluded that CaAl₂O₄ is ferroelectric, which has not been tested in this work any further.

Appendix

 

Fig. 11. Topas V5 refinements for CaAl₂O₄:Eu 1%. The blue line is the experimental data, red is the calculated profile and the grey the difference profile.

Download Full Size | PDF

Table 2. Monoclinic SA in Sr_0.99-xCa_xEu_0.01Al₂O₄

View Table | View all tables in this article

Table 3. Monoclinic CA in Sr_0.99-xCa_xEu_0.01Al₂O₄

View Table | View all tables in this article

Table 4. Monoclinic CA₂ (Grossite) in Sr_0.99-xCa_xEu_0.01Al₂O₄

View Table | View all tables in this article

Table 5. Hexagonal (P6₃) CSA in Sr_0.99-xCa_xEu_0.01Al₂O₄

View Table | View all tables in this article

Funding

Engineering and Physical Sciences Research Council (EPSRC) (CONVERTED (JeS no. TS/1003053/1), FAB3D, PRISM (EP/N508974/1), PURPOSE (TP11/MFE/6/1/AA129F; EP-SRC TS/G000271/1)); TECHNOLOGY STRATEGY BOARD, UK (CONVERT).

References

1. W. M. Yen and M. J. Weber, Inorganic Phosphors (CRC Press, 2004).

2. G. Blasse and A. Bril, “Fluorescence of Eu²⁺ -activated alkaline-earth aluminates,” Philips Res. Rep. 23, 201–206 (1968).

3. G. Blasse, W. I. Wanmaker, and J. W. Ter Vrugt, “Some new classes of efficient Eu²⁺ -activated phosphors,” J. Electrochem. Soc. 115(6), 673 (1968). [CrossRef]  

4. F. C. Palilla, A. K. Levine, and M. R. Tomkus, “Fluorescent properties of alkaline earth aluminates of the type MAl₂O₄ activated by divalent europium,” J. Electrochem. Soc. 115(6), 642–644 (1968). [CrossRef]  

5. V. Abbruscato, “Optical and Electrical Properties of SrAl₂O₄:Eu²⁺,” J. Electrochem. Soc. 118(6), 930–933 (1971). [CrossRef]  

6. S. H. M. Poort, W. P. Blokpoel, and G. Blasse, “Luminescence of Eu²⁺ in barium and strontium aluminate and gallate,” Chem. Mater. 7(8), 1547–1551 (1995). [CrossRef]  

7. B. M. Mothudi, O. M. Ntwaeaborwa, J. R. Botha, and H. C. Swart, “Photoluminescence and phosphorescence properties of MAl₂O₄:Eu²⁺, Dy³⁺ (M = Ca,Ba,Sr) phosphors prepared at an initiating combustion temperature of 500°C,” Phys. B 404(22), 4440–4444 (2009). [CrossRef]  

8. T. Matsuzawa, Y. Aoki, N. Takeuchi, and Y. Murayama, “A new long phosphorescent phosphor with high brightness, SrAI₂O₄:Eu²⁺,Dy³⁺,” J. Electrochem. Soc. 143(8), 2670–2673 (1996). [CrossRef]  

9. K. Van den Eeckhout, P. F. Smet, and D. Poelman, “Persistent luminescence in Eu²⁺ -doped compounds: A Review,” Materials 3(4), 2536–2566 (2010). [CrossRef]  

10. J. Kaur, B. Jaykumar, V. Dubey, R. Shrivastava, and N.S. Suryanarayana, “Optical properties of rare earth-doped barium aluminate synthesized by different methods-A Review,” Res. Chem. Intermed. 41(4), 2317–2343 (2015). [CrossRef]  

11. Y. Lin, Z. Tang, Z. Zhang, and C. Nan, “Influence of co-doping different rare earth ions on the luminescence of CaAl₂O₄-based phosphors,” J. Eur. Ceram. Soc. 23(1), 175–178 (2003). [CrossRef]  

12. D. S. Kshatri and A. Khare, “Characterization and optical properties of Dy³⁺ doped nanocrystallineSrAl₂O₄:Eu²⁺ phosphor,” J. Alloys Compd. 588, 488–495 (2014). [CrossRef]  

13. P. Ptáček, Strontium Aluminate - Cement Fundamentals, Manufacturing, Hydration, Setting Behaviour and Application (InTech, 2014), Chap. 1.

14. R. C. Ropp, Encyclopedia of the Alkaline Earth Compounds (Elsevier, 2013), Chap. 6.

15. A.-R. Schulze and H. Müller-Buschbaum, “Compound Formation MeO:M₂O₃. IV. Structure of Monoclinic SrAl₂O₄,” Z. Anorg. Allg. Chem. 475(4), 205–210 (1981). [CrossRef]  

16. M. Avdeev, S. Yakovlev, A.A. Yaremchenko, and V.V. Kharton, “Transitions between P2₁, P6₃($\sqrt 3 A$), and P6₃2 modifications of SrAl₂O₄ by in situ high-temperature X-ray and neutron diffraction,” J. Solid State Chem. 180(12), 3535–3544 (2007). [CrossRef]  

17. W. Hörkner and H. Müller-Buschbaum, “About the crystal structure of CaAl₂O₄,” J. Inorg. Nucl. Chem. 38(5), 983–984 (1976). [CrossRef]  

18. W. Hörkner and H. Müller-Buschbaum, “About the crystal structure of BaAl₂O₄,” Z. Anorg. Allg. Chem. 451(1), 40–44 (1979). [CrossRef]  

19. S. Y. Huang, R. Von Der Muehll, J. Ravez, and M. Couzi, “Phase transition and symmetry in BaAl₂O₄,” Ferroelectrics 159(1), 127–132 (1994). [CrossRef]  

20. A. M. Abakumov, O. I. Lebedev, L. Nistor, G. Van Tendeloo, and S. Amelinkx, “The ferroelectric phase transition in tridymite type BaAL₂O₄ studied by electron microscopy,” Phase Transitions 71(2), 143–160 (2000). [CrossRef]  

21. T. Aitasalo, J. Hölsa, H. Junger, M. Lastusaari, and J. Niittykoski, “Comparison of sol-gel and solid-state prepared Eu²⁺ doped calcium aluminates,” Mater. Sci. 20(1), 15–20 (2002).

22. B. Smets, J. Rutten, G. Hoeks, and J. Verlijsdonk, “2SrO. 3Al₂O₃:Eu²⁺ and 1.29(Ba,Ca)O, 6Al₂O₃:Eu²⁺ Two new blue-emitting phosphors,” J. Electrochem. Soc. 136(7), 2119–2123 (1989). [CrossRef]  

23. D. Wang, M. Wang, and G. Lü, “Synthesis, crystal structure and X-ray powder diffraction data of the phosphor matrix 4SrO.7Al₂O₃,” J. Mater. Sci. 34(20), 4959–4964 (1999). [CrossRef]  

24. D. Dutczak, T. Jüstel, C. Ronda, and A. Meijerink, “Eu²⁺ luminescence in strontium aluminates,” Phys. Chem. Chem. Phys. 17(23), 15236–15249 (2015). [CrossRef]  

25. F. Clabau, X. Rocquefelte, S. Jobic, P. Deniard, M.-H. Whangbo, A. Garcia, and T. Le Mercier, “Mechanism of phosphorescence appropriate for the long-lasting phosphors Eu²⁺-doped SrAl₂O₄ with codopants Dy³⁺ and B³⁺,” Chem. Mater. 17(15), 3904–3912 (2005). [CrossRef]  

26. J. Botterman, J. Joos, and P. F. Smet, “Trapping and detrapping in SrAl₂O₄:Eu,Dy persistent phosphors: Influence of excitation wavelength and temperature,” Phys. Rev. B: Condens. Matter Mater. Phys. 90(8), 085147 (2014). [CrossRef]  

27. G. Blasse and B. C. Grabmaier, Luminescent Materials (Springer-Verlag, 1994), p. 123.

28. J. M. Ngaruiya, S. Nieuwoudt, O. M. Ntwaeaborwa, J. J. Terblans, and H. C. Swart, “Resolution of Eu²⁺ asymmetrical emission peak of SrAl₂O₄:Eu²⁺, Dy³⁺ phosphor by cathodoluminescence measurements,” Mater. Lett. 62(17-18), 3192–3194 (2008). [CrossRef]  

29. A. Shukla, “Development of a critically evaluated thermodynamic database for the systems containing alkaline-earth oxides,” Thesis, University of Montreal, (2012).

30. V. Singh, J-J. Zhu, M. Tiwari, M. Soni, M. Aynayas, S-H. Hyun, R. Narayanan, M. Mohapatra, and V. Natarajan, “Characterization, luminescence and EPR investigations of Eu²⁺ activated strontium aluminate phosphor,” J. Non-Cryst. Solids 355(50-51), 2491–2495 (2009). [CrossRef]  

31. E. Cordoncillo, B. Julian-Lopez, M. Martinez, M. L. Sanjuán, and P. Escribano, “New insights in the structure–luminescence relationship of Eu:SrAl₂O₄,” J. Alloys Compd. 484(1-2), 693–697 (2009). [CrossRef]  

32. X. B. Yu, C. L. Zhou, X. H. He, X. F. Peng, and S. P. Yang, “The influence of some processing conditions on luminescence of SrAl₂O₄:Eu²⁺ nanoparticles produced by combustion method,” Mater. Lett. 58(6), 1087–1091 (2004). [CrossRef]  

33. S. H. Ju, S.G. Kim, J. C. Choi, H. L. Park, S.-I. Mho, and T. W. Kim, “Synthesis and characterization of white light emitting Ca_xSr_1-xAl₂O₄:Tb³⁺,Eu³⁺ phosphor for solid state lighting,” Mater. Res. Bull. 34(12-13), 1905–1909 (1999). [CrossRef]  

34. A. K. Prodjosantoso and B. J. Kennedy, “Synthesis and evolution of the crystalline phases in Ca_1-XSr_xAl₂O₄,” J. Solid State Chem. 168(1), 229–236 (2002). [CrossRef]  

35. H. Pöllmann and R. Kaden, “X-ray investigations of solid solutions of monocalcium aluminate and monostrontium aluminate important phases in cement and phosphorescence materials,” Powder Diffr. 29(02), 141–146 (2014). [CrossRef]  

36. Q. Xie, B. Li, X. He, M. Zhang, Y. Chen, and Q. Zeng, “Correlation of structure, tunable colors, and lifetimes of (Sr, Ca, Ba)Al₂O₄:Eu²⁺, Dy³⁺ phosphors,” Materials 10(10), 1198 (2017). [CrossRef]  

37. D. W. Goodwin and A.J. Lindop, “The crystal structure of CaO.2AI₂O₃,” Acta Cryst. B 26(9), 1230–1235 (1970). [CrossRef]  

38. M. Karmaoui, M.-G. Willinger, L. Mafra, T. Herntrich, and N. Pinna, “A general non-aqueous route to crystalline alkaline earth aluminate nanostructures,” Nanoscale 1(3), 360–365 (2009). [CrossRef]  

39. A. J. Lindop and D. W. Goodwin, “The refined structure of SrO.2A1₂0₃,” Acta Cryst. B 28(8), 2625–2626 (1972). [CrossRef]  

40. H. Boysen, M. Lerch, A. Stys, and A. Senyshyn, “Structure and oxygen mobility in mayenite (Ca₁₂Al₁₄O₃₃): a high-temperature neutron powder diffraction study,” Acta Cryst. B 63(5), 675–682 (2007). [CrossRef]  

41. R. D. Shannon, “Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides,” Acta Cryst. A 32(5), 751–767 (1976). [CrossRef]  

42. M. de Jong, L. Seijo, A. Meijerink, and F. T. Rabauw, “Resolving the ambiguity in the relation between Stokes shift and Huang–Rhys parameter,” Phys. Chem. Chem. Phys. 17(26), 16959–16969 (2015). [CrossRef]  

43. D. den Engelsen, P. G. Harris, T. G. Ireland, G. Fern, and J. Silver, “Symmetry-related transitions in the photoluminescence and cathodoluminescence spectra of nanosized cubic Y₂O₃:Tb³⁺,” ECS J. Solid State Sci. Technol. 4(12), R145–R152 (2015). [CrossRef]  

44. D. den Engelsen, P. G. Harris, T. G. Ireland, G. Fern, and J. Silver, “Symmetry-Related Transitions in the Spectrum of Nanosized Cubic Y₂O₃:Tb³⁺,” ECS J. Solid State Sci. Technol. 4(7), R105–R113 (2015). [CrossRef]  

45. L. Ning, X. Huang, Y. Huang, and P. A. Tanner, “Origin of Green Persistent Luminescence of Eu-Doped SrAl₂O₄ from Multiconfigurational Ab Initio Study of 4f⁷ → 4f⁶5d¹ Transitions,” J. Mater. Chem. C 6(25), 6637–6640 (2018). [CrossRef]  

46. J. Ueda, T. Nakanishi, Y. Katayama, and S. Tanabe, “Optical and optoelectronic analysis of persistent luminescence in Eu²⁺-Dy³⁺ codoped SrAl₂O₄ ceramic phosphor,” Phys. Status Solidi C 9(12), 2322–2325 (2012). [CrossRef]  

47. M. Nazarov, M. G. Brik, D. Spassky, B. Tsukerblat, A. Nor Nazida, and M. N. Ahmad-Fauzi, “Structural and electronic properties of SrAl₂O₄:Eu²⁺ from density functional theory calculations,” J. Alloys Compd. 573(1), 6–10 (2013). [CrossRef]  

48. J. Ueda, T. Shinoda, and S. Tanabe, “Evidence of three different Eu²⁺ sites and their luminescence quenching processes in CaAl₂O₄:Eu²⁺,” Opt. Mater. 41(1), 84–89 (2015). [CrossRef]  

49. J. Bierwagen, S. Yoon, N. Gartmann, B. Walfort, and H. Hagemann, “Thermal and concentration dependent energy transfer of Eu²⁺ in SrAl₂O₄,” Opt. Mater. Express 6(3), 793–803 (2016). [CrossRef]  

50. V. Vitola, D. Millers, K. Smits, I. Bite, and A. Zolotarjovs, “The search for defects in undoped SrAl₂O₄ material,” Opt. Mater. 87, 48–52 (2019). [CrossRef]  

51. D. den Engelsen, G. R. Fern, T. G. Ireland, P. G. Harris, P. R. Hobson, A. Lipman, R. Dhillon, P. J. Marsh, and J. Silver, “Ultraviolet and blue cathodoluminescence from cubic Y₂O₃ and Y₂O₃:Eu³⁺ generated in a transmission electron microscope,” J. Mater. Chem. C 4(29), 7026–7034 (2016). [CrossRef]  

52. D. den Engelsen, G. R. Fern, T. G. Ireland, and J. Silver, “Cathodoluminescence of Y₂O₃:Ln³⁺ (Ln = Tb, Er and Tm) and Y₂O₃:Bi³⁺ nanocrystalline particles at 200 keV,” RSC Adv. 8(1), 396–405 (2018). [CrossRef]  

53. M. Greger, M. Kollar, and D. Volhardt, “Isosbestic points: How a narrow crossing region of curves determines their leading parameter dependence,” Phys. Rev. B 87(19), 195140 (2013). [CrossRef]  

54. Y. Zorenko, T. Zorenko, T. Voznyak, A. Mandowski, Q. Xia, M. Batentschuk, and J. Friedrich, “Luminescence of F⁺ and F centers in Al₂O₃-Y₂O₃ oxide compounds,” IOP Conf. Ser.: Mater. Sci. Eng. 15, 012060 (2010). [CrossRef]  

55. K. A. Denault, J. Brgoch, S. D. Kloß, M. W. Gaultois, J. Siewenie, K. Page, and R. Seshadri, “Average and Local Structure, Debye Temperature, and Structural Rigidity in Some Oxide Compounds Related to Phosphor Hosts,” ACS Appl. Mater. Interfaces 7(13), 7264–7272 (2015). [CrossRef]