Synthesis and characterization of micro/nano-structured BaHPO₄/Ba₃(PO₄)₂/Ba₅(PO₄)₃OH phases and their luminescence

Abstract

Microspheres covered with microcuboids/nanorods and nanoparticles of BaHPO₄/Ba₃(PO₄)₂/Ba₅(PO₄)₃OH phases have been successfully synthesized by a facile hydrothermal (HT) method using the citric acid as a surfactant at different pH values. X-ray diffraction (XRD), field emission scanning electron microscopy (FE-SEM) and fluorescence spectrometry were used to characterize the samples. It was found that the pH value was a crucial factor for the phase formation and shape determination of the final products, which were discussed in detail. Attractively, the as-prepared BaHPO₄/Ba₃(PO₄)₂/Ba₅(PO₄)₃OH samples emitted an intense blue light in a broad band from 380 to 625 nm, for which the mechanism was complex ions luminescence originating from the transition of ³T₁ → ¹A₁ in PO₄³⁻. Meanwhile, an obvious red shift for the emission band was observed between nano- and bulk-Ba₃(PO₄)₂ synthesized by HT and conventional solid-state (CSS) reactions, respectively, which was due to the effect of the product being nanosized. The same effect was also revealed by the fact that the decay time of the latter was about 2.5 times that of the former. Moreover, the decay mode of Ba₅(PO₄)₃OH was different from those of BaHPO₄ and Ba₃(PO₄)₂, which was ascribed to the effect of the substitution of three OH⁻ for one PO₄³⁻ on their electronic structures.

---

1. Introduction

In the past decades, the orthophosphate of barium, Ba₃(PO₄)₂, has been reported to be a good host for rare earth-doped luminescence materials. As early as 1979, Tale et al. reported that Eu²⁺ doped Ba₃(PO₄)₂ was an efficient luminophor for X-ray screens.¹ Subsequently, in 1996, Poort et al. reported that the emission spectrum of Ba₃(PO₄)₂:Eu²⁺ presented two blue bands ascribed to two different positions of Eu²⁺ in the host.² In 2003, Liang et al. systematically investigated the optical spectroscopic properties of Ce³⁺, Sm³⁺, Eu³⁺, Eu²⁺ and Tb³⁺ doped Ba₃(PO₄)₂.³ In further research, in 2011, Chen et al. studied the luminescence properties of Ba₃(PO₄)₂ co-doped with Ce³⁺ and Dy³⁺ ions.⁴ They found that the introduction of a moderate amount of Dy³⁺ ions could cause a red shift in the Ce³⁺ emission band and enhance its emission intensity. However, the conventional solid-state (CSS) reaction route has been the only preparation method for all these phosphors. As we all know, their block-shape and poor particle distribution are prominent disadvantages of CSS products, and may hinder the effectiveness of the phosphors. In addition, the long sintering time and high sintering temperature in the CSS process greatly increase the cost. To improve the current situation, soft chemical synthetic methods, including sol–gel, precipitation, hydrothermal (HT) synthesis and so on,^5,6 have been rapidly developed and certified to be some of the most effective and convenient approaches to yielding homogeneous and shape-controllable products with micro/nanostructures in recent years.^7,8 In 2012, Cheng et al. successfully synthesized Eu²⁺ and Tb³⁺ doped Ba₃(PO₄)₂ nanowires by the precipitation method for the first time, but anodic aluminum oxide was used as a template in the precipitation process, which involved a complicated process and may have introduced impurities owing to the incomplete removal of the template.^5,9,10 As such, template-free strategies, of which HT synthesis is a typical one, should be more popular. In HT synthesis, a unique trait is that the liquid or gas is usually used as the reaction medium in a closed environment to form high temperature and high pressure surroundings, which can promote the formation of well dispersed and shaped crystals with complete crystallization in one step.^11,12 Many experiments have shown that the shape and phase structure of products depended on the HT conditions to a certain degree, including the pH value of the initial solution, type and amount of surfactant, reaction time and temperature, concentration of the reaction solution, etc.^8,13–17 Therefore, this present work focuses on the HT process in the synthesis of Ba₃(PO₄)₂ materials. It is exciting that different phases, existing as spheres made up of distinct unit elements, were formed at different pH values, and all of the as-obtained host materials were observed to show an intense blue emission for the first time. Moreover, some nanostructures were generated in the HT process, which demonstrated different luminescence properties to those of the CSS products. This finding, that all of the as-synthesized non-rare earth doped BaHPO₄/Ba₃(PO₄)₂/Ba₅(PO₄)₃OH materials emit a blue light in our work, may contribute to the field of near-ultraviolet (n-UV) conversion tri-color phosphors and save rare earth resources to some degree.

2. Experimental

2.1. Synthesis

To begin with, Ba₃(PO₄)₂ powder was expected to be obtained using the HT method. The raw materials were Ba(NO₃)₂ (A.R.) and NH₄H₂PO₄ (A.R.). In a typical procedure, the stoichiometric amount of Ba(NO₃)₂ and NH₄H₂PO₄, and moderate citric acid (CA) acting as a surfactant, were weighed and dissolved together in deionized water under continuous stirring. Subsequently, ammonia (NH₃·H₂O) was added to adjust the pH to different values. After additional agitation for 30 minutes, the mixed solution was transferred to an autoclave with 30% filling capacity, and HT treatment was performed at 180 °C for 12 h. The as-obtained samples were washed several times using deionized water and anhydrous ethanol, then dried in an oven at 80 °C. As a comparison, the Ba₃(PO₄)₂ samples were also synthesized using a CSS reaction, with 0 and 5 mass% H₃BO₃ (labelled as CSS0 and CSS5). As raw materials, the stoichiometric amounts of BaCO₃ and NH₄H₂PO₄ were weighed and mixed well in an agate mortar. The mixture was then transferred to an alumina crucible and calcined at 1300 °C for 4 h in an oxygen atmosphere. Finally, the as-obtained samples were slowly cooled to room temperature for subsequent characterization.

2.2. Characterization

The crystal structure of the phosphors was characterized using X-ray diffraction (XRD) (Bruker D8 Focus) with Ni-filtered Cu-Kα (λ = 1.540598 Å) radiation at 40 kV tube voltage and 40 mA tube current. The XRD data were collected in a 2θ range from 10° to 70°, with the continuous scan mode at a speed of 0.05 s per step with a step size of 0.01°. The morphology and microstructure were characterized using field emission scanning electron microscopy (FE-SEM) (Japan SU8010) at 15 kV. Excitation and emission spectra were recorded using a fluorescence spectrometer (FLUOROMAX-4P, Horiba Jobin Yvon, New Jersey, USA) equipped with a 150 W xenon lamp as the excitation source. The spectral step length of both the excitation and emission spectra was set up to be 1.0 nm with the width of the monochromator slits adjusted to 0.50 nm. The lifetime was recorded on a spectro-fluorometer (HORIBA, JOBIN YVON FL3-21), and pulse laser radiation (nano-LED) was used as the excitation source. The other measurement conditions were kept consistent from sample to sample during the measurements. All the measurements were carried out at room temperature.

3. Results and discussion

3.1. Phase identification, structure and calculation of crystallite size

Fig. 1 shows the XRD spectra of the as-obtained phases produced by the HT process at 180 °C for 12 h with the different pH values of 6.0, 7.0, 8.0, 9.0 and 10.0. For comparison, XRD standard data patterns of the Joint Committee on Powder Diffraction Standards (JCPDS) card no. 72-1370 and 78-1141 are also given. It could be found that BaHPO₄ was obtained with the pH value adjusted to 6.0, 7.0, 8.0 and 9.0, respectively, and that further elevating the pH value to 10.0 caused Ba₅(PO₄)₃OH to be formed. These results suggest that the pH value has a great effect on the phase formation while the other experimental conditions are the same. Thus, we assume that Ba₃(PO₄)₂ should be obtained by adjusting the pH value to 9.0–10.0, as shown in Fig. 2, in which the XRD data of the as-synthesized CSS0 and CSS5 samples prepared through a high temperature solid state process at 1300 °C for 4 h under an oxygen atmosphere, together with the spectrum from JCPDS card no. 85-0904, are shown. As expected, the Ba₃(PO₄)₂ phase was obtained under feasible conditions, and was exactly indexed to the pure trigonal phase belonging to the space group R3̄m (no. 166) with cell parameters of a = b = c = 7.711 Å, V = 570.28 ų, and Z = 1.

Fig. 1 XRD data of the as-synthesized samples produced by the HT method under 180 °C for 12 h at different pH values, together with the standard data of BaHPO₄ (JCPDS card no. 72-1370) and Ba₅(PO₄)₃OH (JCPDS card no. 78-1141) as references.

Fig. 2 (a) XRD data of the as-synthesized samples produced via the CSS route at 1300 °C for 4 h under an oxygen atmosphere and the HT method at 180 °C for 12 h at different pH values. The standard data for Ba₃(PO₄)₂ (JCPDS card no. 85-0904) were used as a reference. (b) Partially enlarged details of the XRD patterns to show the differences in the major peaks.

Among the as-synthesized Ba₃(PO₄)₂ samples, the relative intensities of the diffraction peaks obtained under HT conditions were stronger than those obtained under CSS conditions. Moreover, comparing CSS0 and CSS5, the relative intensities of the former were weaker than those of the latter. The degree of crystallization under different reaction conditions may be responsible for these results.¹⁶ As mentioned above, the surroundings in a closed environment under HT conditions are better than those in CSS for the generation of highly crystallized products. And in the CSS process, H₃BO₃, a powdered solid at room temperature, will melt into liquid at about 169 °C, thus a liquid surrounding is created, which can provide more contact room for the solid-state raw materials BaCO₃ and NH₄H₂PO₄ so that they are sufficiently mixed and react better.^18–20 And finally, in comparison to CSS with 0% H₃BO₃ (CSS0), CSS5 could produce powders with better crystallization.

On the other hand, for the three typical samples obtained by the HT method, it was observed that the differences in the relative intensities based on (110) and (205) were especially salient, and were due to the different preferential growth orientations caused by the different pH values.^5,21,22 A similar phenomenon could also be found in the as-prepared samples using the CSS method, which, however, was probably ascribed to the introduction of H₃BO₃.

Also note that the full width at half maximum (FWHM) values of the diffraction peaks for the three typical Ba₃(PO₄)₂ samples synthesized under HT conditions were widened compared with those obtained under CSS conditions, as illustrated in Fig. 2(b). Generally speaking, when a crystallite’s size is larger than 200 nm, the FWHM values of diffraction peaks will remain constant and just depend on the instrument (the Bruker D8 Focus XRD, in our work), conversely, they will be broadened due to the inconsistent scattering. Herein, the phenomenon shown in Fig. 2(b) indicates that the average crystallite sizes of as-prepared Ba₃(PO₄)₂ samples produced under HT conditions are on the nanometer scale, smaller than 200 nm, which can be theoretically estimated by the Scherrer equation^23–25 as follows:

where k is a constant of 0.89, λ is the wavelength of the X-rays (0.15405 nm), β(2θ) is FWHM of the considered diffraction peak in radians after correcting for the peak’s broadening due to the instrument, θ is the Bragg angle (half of the diffraction angle 2θ), and D_hkl is the size of the crystallite, which is perpendicular to the plane (hkl). For the samples synthesized at the pH values of 9.3 (HT9.3), 9.5 (HT9.5) and 9.7 (HT9.7), the average crystallite sizes were estimated to be 71, 66 and 63 nm, respectively. This suggests that the average crystallite size gradually dwindles with an increase in the pH value, which will be verified by SEM experimental results.

3.2. Phase formation, morphology and growth process

It has been found that the pH value plays an important role in the phase structure of the as-formed samples produced via the HT route, which may also have an effect on their morphologies. Fig. 3 demonstrates the SEM images of the final samples obtained at 180 °C for 12 h at different pH values. On the whole, they are micro/nano-structured materials with some differences in morphology. The XRD results indicate that the BaHPO₄ phase can be obtained under HT conditions with pH values of 6.0, 7.0, 8.0 and 9.0, and that the morphologies are all microspheres with different sizes and structural components. Fig. 3(A) and (B) show the shapes of BaHPO₄ samples obtained at a pH value of 6.0. The homogeneous flower-like microspheres were petalled by a large number of uneven microcuboids, which coincidentally spread outward from the center of the flower-like structures. As the pH value was gradually increased to 7.0, 8.0 and 9.0, a series of microspheres with different sizes was observed, as shown in Fig. 3(C), (E) and (G), respectively. From the corresponding higher magnification SEM images (D), (F) and (H), it could be seen that the petals of the flower-like microspheres seemed to vary with respect to those shown in image (B). Orderly slender nanorods, circular planes formed with aggregated nanorods, and messy and aggregated nanorods were respectively observed to be the basic constituent elements of those microspheres. Increasing the pH value to 10.0 caused the as-formed nanoparticles to be in the Ba₅(PO₄)₃OH phase. Further accurately adjusting the pH value between 9.0 and 10.0 yielded the expected phase Ba₃(PO₄)₂. Fig. 4-1 exhibits the SEM images of Ba₃(PO₄)₂ samples prepared with the pH values at 9.3 (A and B), 9.5 (C and D) and 9.7 (E and F). As depicted in Fig. 4-1(A), (C) and (E), the as-prepared Ba₃(PO₄)₂ samples were composed of well distributed nanoparticles, whose size gradually decreased as the pH value increased from 9.3 to 9.7. Comparing the higher magnification images (B), (D) and (F), it was revealed that the shapes of particles tended to change from round to long, which confirmed the phenomenon of different preferential growth orientations. Moreover, all of these nanoparticles presented a very smooth surface indicating their high crystallinity, as the XRD results had shown.²⁶ For all the as-prepared samples, the possible phase structure formation mechanism, accompanied by morphology architecture details, is shown in Scheme 1. It highlights the roles of both the pH value and CA in the phase formation and morphologies of the final products. The whole process may be described with the following equations.

3Ba²⁺ + 2Cit³⁻ → 3Ba²⁺– 2Cit³⁻ (citrate complex) (2)

HPO₄²⁻ + H₃O⁺ ↔ H₂PO₄⁻ + H₂O ↔ H₃PO₄ + OH⁻ (3)

(add NH₃·H₂O into the above solution system)

NH₃·H₂O ↔ NH₄⁺ + OH⁻ (4)

pH = 6.0–9.0,

H₂PO₄⁻ + OH⁻ ↔ HPO₄²⁻ + H₂O (5a)

3Ba²⁺– 2Cit³⁻ + 3HPO₄²⁻ → 3BaHPO₄ + 2Cit³⁻ (5b)

pH = 9.0–10.0,

H₂PO₄⁻ + 2OH⁻ ↔ PO₄³⁻ + 2H₂O (6a)

3Ba²⁺– 2Cit³⁻ + 2PO₄³⁻ → Ba₃(PO₄)₂ + 2Cit³⁻ (6b)

pH = 10.0,

3H₂PO₄⁻ + 7OH⁻ ↔ [(PO₄)₃OH]¹⁰⁻ + 6H₂O (7a)

5(3Ba²⁺– 2Cit³⁻) + 3[(PO₄)₃OH]¹⁰⁻ → 3Ba₅(PO₄)₃OH + 10Cit³⁻ (7b)

Fig. 3 SEM images of as-synthesized samples produced under HT conditions at 180 °C for 12 h with the pH value at (A and B) 6.0: flower-like microspheres petalled with microcuboids; (C and D) 7.0: microspheres covered with orderly nanorods; (E and F) 8.0: microspheres/hemispheres constructed by a series of planes formed of aggregated nanorods; (G and H) 9.0: microspheres structured with messy and aggregated nanorods; and (I and J) 10.0: nanoparticles.

Fig. 4 (1) SEM images of as-prepared Ba₃(PO₄)₂ samples produced by the HT method at 180 °C for 12 h at different pH values: (A and B) non-uniform round granular nanoparticles at pH = 9.3; (C and D) mixture of long and round granular nanoparticles at pH = 9.5; and (E and F) long granular nanoparticles at pH = 9.7. (2) SEM images of as-obtained Ba₃(PO₄)₂ samples produced by the CSS method with (G and H) 0 mass% H₃BO₃ (CSS0): heterogeneous, irregular and shaggy structures; and (I and J) 5 mass% H₃BO₃ (CSS5): less heterogeneous, regular and smooth structures.

Scheme 1 Schematic illustration of the possible process for forming the as-obtained HT products at different pH values at 180 °C for 12 h. (a) The reaction process, and (b) the details of the formation of the phases and morphologies. The whole process can be summarized as (1) the formation of the Ba²⁺–citrate complex, (2) ion-exchange between PO₄³⁻ and citrate, while OH⁻ ions come into effect, and (3) the assembly (pH value at 6.0–9.0) of the unit elements.^13,26

The citrate complex (3Ba²⁺– 2Cit³⁻) is temporarily formed by an adsorption process when cationic Ba²⁺ meets with CA, which then waits for the right anion to take the Ba²⁺ ions away. It is known that H₂PO₄⁻, H₃PO₄ and HPO₄²⁻ exist synchronously in NH₄H₂PO₄ aqueous solution, as shown in eqn (3), and that adding a moderate amount of NH₃·H₂O will change the balance of the system. The OH⁻ in NH₃·H₂O aqueous solution can restrain the right side of the reaction while accelerating the left one, resulting in less and less H₃PO₄. In our case, when the pH value was adjusted to 6.0–9.0, HPO₄²⁻ became the selective anion that conducted an ion-exchange reaction with citrate ions to generate the BaHPO₄ phase, as represented in eqn (5a) and (5b). Likewise, when an appropriate OH⁻ concentration existed in the reaction system, the phases of Ba₃(PO₄)₂ and Ba₅(PO₄)₃OH, could be obtained, respectively, exactly as eqn (6a), (6b), (7a) and (7b) showed. Throughout the process, various morphologies of products were formed, which are illustrated in Scheme 1(b). Under different pH conditions, Cit³⁻ exhibited different adsorption/desorption abilities that could control the reaction rates to yield products with various morphologies and sizes in the same limited reaction time.^13,27 On the other hand, CA has three carboxylic groups and each of them has a unique spatial orientation, which will be highly sensitive to the surface’s molecular structure. Hence, it could assemble the pre-grown elementary structures (microcuboids, nanorods and nanoparticles) together by binding side-on to each and, in view of growth energies, the new crystal nucleus prefers to develop on the existing growth steps to form the final flower-like structures, spheres and aggregated round/long particles.^8,13,27–29 Taking into consideration the fact that the Ba₃(PO₄)₂ crystallite size decreases as the pH value increases from 9.3 to 9.7, the citrate complex 3Ba²⁺– 2Cit³⁻ and OH⁻ have comprehensive actions. As eqn (6a) shows, the increasing OH⁻ concentration will destroy the balance of the reaction system and promote the right reaction to produce more PO₄³⁻, which will immediately react with 3Ba²⁺– 2Cit³⁻, as demonstrated in eqn (6b), to form more and more Ba₃(PO₄)₂ crystal nuclei before the initially formed Ba₃(PO₄)₂ crystal nuclei grow. According to the growth route, Ba₃(PO₄)₂ nanoparticles with a smaller size can be obtained when the pH value is increased.

Moreover, the SEM images of Ba₃(PO₄)₂ samples prepared by the CSS method, with 0 mass% (G and H) and 5 mass% (I and J) H₃BO₃, are shown in Fig. 4-2. These indicate that regular and smooth blocks with good distribution could be achieved with the additional 5 mass% H₃BO₃ added in the raw materials, while irregular and shaggy blocks were observed without H₃BO₃, which implied that a proper amount of H₃BO₃ could work as a “shape modifier”, like an organic additive to improve the crystallization, shape and surface smoothness. The possible influencing mechanism is illustrated in Scheme 2. In the growth process, H₃BO₃ melts into a liquid phase, promoting the mobility and homogeneity of solid reactants to gain products with fewer surface defects on the one hand, and to eliminate the aggregation of solid–solid contact for the formation of well dispersed particles on the other hand.^18,19,31

Scheme 2 Schematic for the formation of the morphologies of the as-prepared Ba₃(PO₄)₂ samples by the CSS method with (a) 0 mass% H₃BO₃: direct solid–solid boundaries, aggregation nucleation to form irregular and shaggy blocks without liquid surroundings; and (b) 5 mass% H₃BO₃: sufficient mixing, fewer direct solid–solid boundaries and less homogeneous nucleation to form smooth and well dispersed particles in a liquid environment.^18,30

Comparing Fig. 4-1 with 4-2, it should be emphasized that the particle sizes of as-synthesized Ba₃(PO₄)₂ samples using the different methods were on different scales. Obviously, the size of products obtained by HT was on the nanoscale, while that of products obtained by CSS was in the micron range. This difference in size may have a big influence on the luminescence properties of Ba₃(PO₄)₂ as previously reported in the case of ZnO.^32,33

3.3. Luminescence properties and mechanism

Fig. 5 exhibits the photoluminescence excitation and emission spectra of the series of as-synthesized samples produced via the CSS and HT methods, respectively. As can be seen in Fig. 5(a) and (b), the Ba₅(PO₄)₃OH, Ba₃(PO₄)₂ and BaHPO₄ samples, which were obtained using the HT method at pH values of 10.0, 9.5 and 8.0, respectively, demonstrated similar profiles in both their excitation and emission spectra with different peaks and relative intensities. Moreover, according to the phase order of Ba₅(PO₄)₃OH, Ba₃(PO₄)₂ and BaHPO₄, blue shifts of 2–5 nm could be observed in their emission spectra, with peaks at 447, 444 and 442 nm, respectively. In essence, all of these differences may result from the different phase structures.

Fig. 5 (a) Excitation spectra of the Ba₃(PO₄)₂ sample prepared by the CSS method with 5 mass% H₃BO₃, and samples of Ba₅(PO₄)₃OH, Ba₃(PO₄)₂ and BaHPO₄ obtained by the HT route at pH values of 10, 9.5 and 8, respectively. (b) Emission spectra of Ba₅(PO₄)₃OH, Ba₃(PO₄)₂ and BaHPO₄. (c) Emission spectra of a series of Ba₃(PO₄)₂ samples synthesized by the CSS (with 0 and 5 mass% H₃BO₃) and HT (pH = 9.3, 9.5 and 9.7) methods, respectively. (d) Energy scheme for the luminescence process of a PO₄³⁻ molecule.

As for the as-formed Ba₃(PO₄)₂ phase samples produced via the CSS and HT methods, respectively, there are many differences in their photoluminescence properties. The excitation spectra in Fig. 5(a) illustrate that the excitation spectrum of Ba₃(PO₄)₂ prepared by CSS was composed of two broad bands from 225 to 400 nm with the intense peak at 320 nm, while that synthesized by HT presented only a broad band from 250 to 420 nm centred at 376 nm, between which an obvious red shift of 56 nm existed. At the same time, a similar phenomenon could also be observed in their emission spectra, as shown in Fig. 5(c), which indicate that the emission peaks of the as-obtained Ba₃(PO₄)₂ samples produced by the CSS and HT routes were at 414 and about 444 nm, respectively. The photoluminescence spectra show that the as-prepared blue emission Ba₃(PO₄)₂ phosphor can be applied to the white light emitting method of n-UV conversion tri-color phosphors. Also note that the FWHM values of the emission peaks for the Ba₃(PO₄)₂ samples produced by HT have been broadened and the relative intensities were higher in comparison to those produced by CSS, which is consistent with the XRD results. Moreover, a new sharp emission peak at 403 nm could be observed in the emission spectrum of Ba₃(PO₄)₂ synthesized by the HT route. All these phenomena may be ascribed to their size being on the nanometer scale. As we all know, the small size effect, surface and interface effect, quantum size effect, macroscopic quantum tunnel effect and dielectric confinement effect are the five main specific effects of nanomaterials, among which the first two are the most investigated for their effects on luminescence properties.^34,35 Generally, the occurrence of red shifts can probably be attributed to the surface and interface effect. In more detail, the surface tension of nanoparticles will increase due to the surface and interface effect, which could increase the crystal field leading to changes in the energy levels or the narrowing of the band gap. And the small size effect can generate two results. With a decrease in the size of nanomaterials, the crystal periodic boundary conditions will be destroyed, leading to the introduction of an unordered phase resulting in the broadening of emission peaks on the one hand. And on the other hand, the relevant specific surface area will increase promptly, which puts numerous surface atoms in an exposed state, so that many surface defects are inevitable. As a result, the spectral composition may change, leading to, for example, the observed new sharp absorption peak in our case. At the nanometer scale, a material’s properties are highly sensitive to the particle sizes. With an increase in the pH value from 9.3 to 9.7 under HT conditions, there was a gradual increase in the relative intensities and narrowing in the FWHM for the Ba₃(PO₄)₂ samples, as listed in Table 1. From the results of XRD in Fig. 2, it could be seen that the diffraction intensities, especially along the planes (110) and (205), slowly increased according to the sample sequence of HT9.3, HT9.5, and HT9.7, which implied that better and better crystallinity of the particles was achieved as the pH value was increased from 9.3 to 9.7. Combining the above results, it can be concluded that particles with better crystallinity show higher emission intensities in our current work, which can also be used to interpret the phenomenon that CSS5 (with 5 mass% H₃BO₃) shows a stronger relative intensity than CSS0 (without H₃BO₃) when produced via CSS. At the same time, all of these phenomena can be supported by the SEM results showing surface smoothness and good graininess.

Table 1 A summary of emission peak positions, FWHM values, relative intensities and particle sizes for the as-synthesized Ba₃(PO₄)₂ samples produced by the CSS and HT routes

Ba3(PO4)2 samples prepared under CSS and HT conditions	Emission peak (nm)	FWHM (nm)	Relative intensity (a.u.)	Particle size (nm)
CSS 0% H3BO3	414	33.44	0.2329	—
CSS 5% H3BO3	414	34.52	0.4095	—
HT pH = 9.3	435	78.53	0.3778	71
HT pH = 9.5	444	76.35	0.6745	66
HT pH = 9.7	443	72.91	1.0015	63

---

The photoluminescence mechanism of the as-prepared BaHPO₄, Ba₃(PO₄)₂ and Ba₅(PO₄)₃OH hosts can be described as similar to that of (MoO₄²⁻),³⁶ (WO₄²⁻),³⁷ and (VO₄³⁻).^38,39 This involves using the model of the MO₄ⁿ⁻ (n = 1, 2, 3 and 4) complex, due to the common closed-shell electronic configurations, which is one of the four common kinds of luminescence centers, namely, ns² type, transition metal, lanthanide metal and complex ions luminescence centers.⁴⁰ Actually, the outer electron configuration of the P atom is 3s²3p³, and in our case, four unequal sp³ hybrid orbitals are formed when bonding with the O atom, of which the one occupied by a pair of electrons has a lower energy and the other three, each occupied by a single electron, have a higher energy, so that a coordination bond and three covalent bonds come into being between P and O. Therefore, the [PO₄] group is a deformation tetrahedron with P as the center. Based on the above discussion, the P⁵⁺ ion has a closed-shell electronic configuration with an empty 3d orbital in PO₄³⁻. Corresponding to the electron charge–transfer transition process from the O2p orbital (t₁ symmetry in T_d) to the 3d orbital (e and t₂ symmetry) of the P⁵⁺ ion, a broad blue band ranging from 380 to 625 nm was observed with excitation. To make the luminescence process clearer, the molecular orbital theory⁴¹ is usually used. A molecular orbital calculation leads to e and t₁ states for the lowest unoccupied molecular orbital (LUMO) and the highest occupied molecular orbital (HOMO), respectively. By taking the e → t₁ transition into account, the excited electronic states of the t₁e electronic configuration in T_d symmetry are found to consist of ³T₁ ≦ ³T₂ < ¹T₁ < ¹T₁, in the order of increasing energy, the ground state being the ¹A₁ state. The orbital triplets (³T₁, ³T₂) have degenerate levels in the spectral region of 250 to 500 nm. Herein, Fig. 5(d) gives the three-level energy scheme for the luminescence process of a PO₄³⁻ molecule. Since the photoluminescence properties of nano-samples produced under HT conditions will have some differences, the CSS-yielded Ba₃(PO₄)₂ sample was used to illustrate the luminescence process of PO₄³⁻, which can be described in detail as follows. Combining Fig. 5(a), (c) and (d), when excited at a wavelength of 271/320 nm, electrons can be excited from the ground state ¹A₁ to the excited states ¹T₂/¹T₁, following a non-radiative (NR) transition process among the excited states of PO₄³⁻ the electrons are then transferred to ³T₁, and finally electrons return to the ground state and give the emission of the PO₄³⁻ ion centred at 414 nm.

Fig. 6 depicts the decay curves of the as-prepared hosts BaHPO₄, Ba₃(PO₄)₂ and Ba₅(PO₄)₃OH. The decay curves of BaHPO₄ and Ba₃(PO₄)₂ were well fitted with a typical single-order exponential decay mode (eqn (8))⁴² while that of Ba₅(PO₄)₃OH was second-order (eqn (9)).⁴³

I(t) = I₀ exp(−t/τ) (8)

I(t) = I₀ + A₁ exp(−t/τ₁) + A₂ exp(−t/τ₂) (9)

τ* = (A₁τ₁² + A₂τ₂²)/(A₁τ₁ + A₂τ₂) (10)

where I₀ expresses the initial emission intensity immediately after being excited, I(t) is the luminescence intensity at time t; A₁ and A₂ are the fitting constants, and τ₁ and τ₂ are the decay times for the corresponding components. Based on the above eqn (8) and (9), the average decay times were determined to be 13.05, 12.25 and 31.54 ns for BaHPO₄, HT-Ba₃(PO₄)₂ and CSS-Ba₃(PO₄)₂ (HT/CSS-Ba₃(PO₄)₂ denotes the sample generated by the HT/CSS route), respectively, and that of Ba₅(PO₄)₃OH could be calculated by eqn (10),³⁷ which was 169.64 ns. Firstly, the decay time of CSS-Ba₃(PO₄)₂ is about 2.5 times that of HT-Ba₃(PO₄)₂. That is, the decay time of nano-Ba₃(PO₄)₂ is dramatically shortened compared with that of bulk-Ba₃(PO₄)₂, which may be ascribed to the increased quenching centers on the nanomaterial’s surface resulting from the interface and surface effect.^35,43,44 Secondly, we can note that the decay times of BaHPO₄ and HT-Ba₃(PO₄)₂ are almost the same, while that of Ba₅(PO₄)₃OH is about 12 times larger than them, even though the spectral compositions are the same with respect to nano-BaHPO₄, -Ba₃(PO₄)₂ and -Ba₅(PO₄)₃OH. These phenomena demonstrate that the change in the anionic groups can accordingly lead to a change in the decay times, further certifying that the main luminescence center comes from the [PO₄] group for the three hosts, ignoring their different structures. It is known that the trigonal Ba₃(PO₄)₂, orthorhombic BaHPO₄ and hexagonal Ba₅(PO₄)₃OH in our work have different electronic structures, so that electrons should accordingly have different transition probabilities from the excited state to ground state, which result in various lifetimes. It is clear that the change in decay mode and the big difference in lifetime between nano-Ba₃(PO₄)₂ and -Ba₅(PO₄)₃OH indicate the big impact of the substitution of three OH⁻ for one PO₄³⁻ on the electronic structures or transition probability, and that this needs to be explored further.

Fig. 6 Decay curves of host emissions for samples obtained via the HT method (BaHPO₄, Ba₃(PO₄)₂ and Ba₅(PO₄)₃OH) and CSS method (Ba₃(PO₄)₂), respectively.

4. Conclusions

In summary, this work has revealed a facile HT route for the synthesis of a series of micro/nanomaterials of BaHPO₄, Ba₃(PO₄)₂ and Ba₅(PO₄)₃OH. By precisely adjusting the pH values, micro/nanospheres of BaHPO₄, nanoparticles of Ba₃(PO₄)₂ and Ba₅(PO₄)₃OH with different sizes could be obtained. A particular mechanism of action governed by pH values was described in the phases and morphologies forming process, meanwhile, the different adsorption/desorption abilities of CA under different pH conditions and its molecular space-specific influences were considered to have a large effect. Interestingly, the as-synthesized three hosts were all found to emit blue light in a broad band from 380 to 625 nm for the first time, for which the complex ions luminescence mode was proposed. For comparison, Ba₃(PO₄)₂ was also prepared by CSS, and the photoluminescence spectrum showed a blue shift. The small volume effect, and the interface and surface effect of nanomaterials were used to interpret this shift. These phenomena are conducive to facilitating the potential application of the Ba₃(PO₄)₂ host in n-UV conversion tri-color phosphors. Furthermore, the decay lifetimes of BaHPO₄, Ba₅(PO₄)₃OH, HT-Ba₃(PO₄)₂ and CSS-Ba₃(PO₄)₂ were determined to be 13.05, 169.64, 12.25 and 31.54 ns, respectively. The differences in the decay lifetimes of HT-Ba₃(PO₄)₂ and CSS-Ba₃(PO₄)₂ were discussed to come from the quenching centers on the surface of the nanomaterial HT-Ba₃(PO₄)₂, while those among nano-BaHPO₄, -Ba₃(PO₄)₂ and -Ba₅(PO₄)₃OH were ascribed to their different electronic structures and therefore the varying transition probabilities from the excited state to the ground state.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant nos 21171152 and 21301162), the Guangdong Province Enterprise-University-Academy Collaborative Project (no. 2012B091100474), and by Public Service Project of the Chinese Ministry of Land and Resources (201311024).

Notes and references

1. I. Tale, J. Lumin., 1979, 20, 343–347.
2. S. Poort, H. Reijnhoudt, H. Van der Kuip and G. Blasse, J. Alloys Compd., 1996, 241, 75–81.
3. H. Liang, Y. Tao, Q. Zeng, H. He, S. Wang, X. Hou, W. Wang and Q. Su, Mater. Res. Bull., 2003, 38, 797–805.
4. F. Chen, X. Yuan, X. Xiong and L. Liu, J. Chin. Rare Earth Soc., 2011, 29, 450–454.
5. G. Li, C. Peng, C. Li, P. Yang, Z. Hou, Y. Fan, Z. Cheng and J. Lin, Inorg. Chem., 2010, 49, 1449–1457.
6. D. Yue, W. Lu, C. Li, X. Zhang, C. Liu and Z. Wang, Nanoscale, 2014, 6, 2137–2145.
7. J. P. Liu, X. T. Huang, Y. Y. Li, K. M. Sulieman, X. He and F. L. Sun, Cryst. Growth Des., 2006, 6, 1690–1696.
8. S. P. Garcia and S. Semancik, Chem. Mater., 2007, 19, 4016–4022.
9. C. Mu and J. He, Mater. Lett., 2012, 70, 101–104.
10. G. Li, C. Peng, C. Zhang, Z. Xu, M. Shang, D. Yang, X. Kang, W. Wang, C. Li and Z. Cheng, Inorg. Chem., 2010, 49, 10522–10535.
11. G. Demazeau, Z. Naturforsch., B: Chem. Sci., 2010, 65, 999–1006.
12. L. A. Ramajo, F. Rubio-Marcos, A. D. Campo, J. F. Fernández, M. S. Castro and R. Parra, Ceram. Int., 2014, 40, 14701–14712.
13. M. Yang, H. You, Y. Zheng, K. Liu, G. Jia, Y. Song, Y. Huang, L. Zhang and H. Zhang, Inorg. Chem., 2009, 48, 11559–11565.
14. Q. Zhu, J. G. Li, R. Ma, T. Sasaki, X. Yang, X. Li, X. Sun and Y. Sakka, J. Solid State Chem., 2012, 192, 229–237.
15. C. Zhang, L. Zhang, C. Song, G. Jia, S. Huo and S. Shen, J. Alloys Compd., 2014, 589, 185–191.
16. G. Jia, H. You, M. Yang, L. Zhang and H. Zhang, J. Phys. Chem. C, 2009, 113, 16638–16644.
17. S. Som, P. Mitra, V. Kumar, V. Kumar, J. J. Terblans, H. C. Swart and S. K. Sharma, Dalton Trans., 2014, 9860.
18. Y. Yu, H. Wang, L. Li, Y. Chen and R. Zeng, Ceram. Int., 2014, 40, 14171–14175.
19. T. T. Lai, C. C. Chang, C. Y. Yang, S. Das and C. H. Lu, Ceram. Int., 2013, 39, 159–163.
20. S. Som, A. K. Kunti, V. Kumar, V. Kumar, S. Dutta, M. Chowdhury, S. K. Sharma, J. J. Terblans and H. C. Swart, J. Appl. Phys., 2014, 115, 193101.
21. Y. Wu, Q. Gu, G. Ding, F. Tong, Z. Hu and A. M. Jonas, ACS Macro Lett., 2013, 2, 535–538.
22. T. Reimer, I. Paulowicz, R. Roder, S. Kaps, O. Lupan, S. Chemnitz, W. Benecke, C. Ronning, R. Adelung and Y. K. Mishra, ACS Appl. Mater. Interfaces, 2014, 6, 7806–7815.
23. V. Uvarov and I. Popov, Mater. Charact., 2013, 85, 111–123.
24. L. Zhou and B. Yan, J. Phys. Chem. Solids, 2008, 69, 2877–2882.
25. A. K. V. Raj, P. Prabhakar Rao, T. S. Sreena, S. Sameera, V. James and U. A. Renju, Phys. Chem. Chem. Phys., 2014, 16, 23699–23710.
26. H. Lai, Y. Du, M. Zhao, K. Sun and L. Yang, Ceram. Int., 2014, 40, 1885–1891.
27. C. Li, J. Yang, P. Yang, H. Lian and J. Lin, Chem. Mater., 2008, 20, 4317–4326.
28. Y. Xu, S. Xu, S. Wang, Y. Zhang and G. Li, Dalton Trans., 2014, 479–485.
29. Y. K. Mishra, S. Kaps, A. Schuchardt, I. Paulowicz, X. Jin, D. Gedamu, S. Wille, A. Kovalev, O. Lupan and R. Adelung, KONA Powder Part. J., 2014, 31, 92–110.
30. Z. He, Y. Ueda and M. Itoh, J. Cryst. Growth, 2006, 297, 1–3.
31. L. Cheng, H. Zhong, J. Sun, X. Zhang, Y. Peng, T. Yu and X. Zhao, J. Rare Earths, 2008, 26, 211–214.
32. N. Pan, B. Wang, X. P. Wang and J. G. Hou, J. Mater. Chem., 2010, 20, 5567–5581.
33. Y. K. Mishra, S. Kaps, A. Schuchardt, I. Paulowicz, X. Jin, D. Gedamu, S. Freitag, M. Claus, S. Wille, A. Kovalev, S. N. Gorb and R. Adelung, Part. Part. Syst. Charact., 2013, 30, 775–783.
34. S. D. Zhu, G. S. Zhou, R. Cai, Y. Han and W. Tian, China. Heat Treat. Technol., Equip., 2010, 31, 1–5.
35. J. L. Zhang and G. Y. Hong, Chin. J. Lumin., 2005, 26, 285–293.
36. T. Skibiński, S. M. Kaczmarek, G. Leniec, T. Tsuboi, Y. Nakai, M. Berkowski, Z. Kowalski and W. Huang, J. Cryst. Growth, 2014, 401, 802–806.
37. X. C. Chu, Y. N. Xu and L. T. Li, J. Mater. Sci.: Mater. Electron., 2014, 25, 2601–2604.
38. A. Barros, R. Deloncle, J. Deschamp, P. Boutinaud, G. Chadeyron, R. Mahiou, E. Cavalli and M. G. Brik, Opt. Mater., 2014, 36, 1724–1729.
39. V. Kumar, A. K. Bedyal, J. Sharma, V. Kumar, O. M. Ntwaeaborwa and H. C. Swart, Appl. Phys. A, 2014, 116, 1785–1792.
40. Z. Xia, Y. Liang, W. Huang, D. Yu, M. Zhang and M. Tong, Ceram. Int., 2013, 39, 7097–7100.
41. C. J. Ballhausen and A. D. Liehr, J. Mol. Spectrosc., 1958, 2, 342–360.
42. Z. Xia, Y. Liang, D. Yu, M. Zhang, W. Huang, M. Tong, J. Wu and J. Zhao, Opt. Laser Technol., 2014, 56, 387–392.
43. A. I. Becerro, S. Rodríguez-Liviano, A. J. Fernández-Carrión and M. Ocaña, Cryst. Growth Des., 2013, 13, 526–535.
44. D. Li, S. Z. Lu, J. S. Zhang, J. Y. Liu, H. Y. Wang, L. D. Sun and S. H. Huang, Chin. J. Lumin., 2002, 21, 134–138.

---