Photochromic materials have been widely investigated because of their potential technological applications in displays and large-area optical devices^[1,2].

Recently another kind of photochromic material, i.e., rare-earth (RE) compounds, has also attracted interest. The color change for RE-based compounds is due to the light absorbed or reflected by the compounds, and from neither an excitation/luminescence emission process nor a photo-induced change in the material microstructure^[3]. It is well-known that RE ions have unique and extraordinary luminescence properties such as narrow emission lines covering the short wavelength (ultraviolet) to visible region, until the near-infrared region^[4]. Therefore, trivalent RE (i.e., RE3+)-ion-doped luminescent materials receive much attention due to their extensive applications^[5]. Considering that many RE ions also have discrete absorption bands in the visible range, photochromism of RE-based compounds is also possible under illumination from discrete and broad spectral distribution light sources, especially tri-phosphor fluorescent light and sunlight.

Among the RE ions, Ho3+ is promising for photochromic applications, because it exhibits several important electronic transitions in the visible range^[6]. Recently an investigation has also suggested that HoPO4 crystals can display different colors under exposure to the Sun and a tri-phosphor fluorescent lamp^[3]. Therefore, HoPO4 is likely to be developed into a photochromic material for practical applications as a security pigment and a decoration material if the photochromic properties are further improved.

The effect of Li+ doping on the luminescence properties has been investigated in many phosphors such as in SrTiO3:Pr3+^[7] and Gd2−xYxO3:Eu3+^[8]. This fact reveals that Li+ addition remarkably affects the crystallinity, morphology, as well as the photoluminescent efficiency of the phosphors. In this work, Li+ ions are introduced into the HoPO4 in an attempt to improve its color-change properties. The Li+ ions induced changes of the crystal structure, and the absorption properties of HoPO4 have been investigated in detail.

HoPO4:xLi (x=0.0, 0.1, 0.2, and 0.3 in mole fraction) crystals was prepared using the co-precipitation method followed by heat treatment. Ho(NO3)3·6H2O, NH4H2PO4, and LiCl·H2O aqueous solutions were dissolved in distilled water according to the designed mole ratio. The dissolved Ho(NO3)3·6H2O, NH4H2PO4, and LiCl·H2O aqueous solutions were marked as Solutions A–C, respectively. The mole ratio of Ho(NO3)3·6H2O to NH4H2PO4 was kept as 1∶1.1 while different mole contents of LiCl·H2O were introduced according to the designed mole ratio during the co-precipitation process. Earthy yellow precipitates were obtained by dumping slowly Solution A and then Solution C into Solution B under constant stirring with a glass rod. The precursors without washing were stewed for 48 h. After drying at 150°C for 24 h, earthy yellow pigments were obtained. Then they were sintered at 1000°C for 4 h to get the final products. To explore the influence of doped Li+ ions on the spectral characteristics of Ho3+ in HoPO4, Eu3+-doped YPO4 powders of different Li+ content were also prepared with the same method, so as to probe changes of the local environment of Ho3+ in the lanthanum phosphate host.

The structure of the samples were identified by X-ray diffraction (XRD) on a Bruker D8 Advance instrument using a Cu tube with Kα radiation of 0.15406 nm in the 2θ range of 10°–80°. The microstructure was analyzed using a JSM-6610 scanning electron microscope (SEM). The absorption spectra were recorded using diffuse reflectance spectroscopy (a Varian Cary 10 UV-Vis spectrophotometer with a DRA-CA-30I diffuse reflectance accessory). All experiments were carried out at room temperature. The Eu3+ probing spectra were measured using a Perkin–Elmer LS-50 luminescence spectrometer under an excitation of 396 nm produced by a xenon flash lamp.

Figure 1 shows the XRD patterns of the HoPO4 crystals doped with different Li+ concentrations. Figure 1 reveals that all the diffraction peaks can be readily indexed to the tetragonal structure of HoPO4 according to JCPDS-76-1533, indicating that Li+ ions has been embedded into the crystal lattice of HoPO4. The cell parameters of the samples are calculated using the XRD data (Fig. 1, inset). The inset in Fig. 1 reflects that the values of lattice parameters initially decrease with the increasing Li+ ion concentration, and then increase at higher Li+ ion concentration. It is believed that Li+ ions usually occupy interstitial sites in the lanthanum phosphate host. In our work, however, Li+ ions of relatively high concentration are introduced. This means that some of the Li+ ions occupy Ho3+ sites, causing a substantial contraction of the unit cell. As a result, the lattice parameters decrease with the increasing Li+ ion concentration in our work. At higher Li+ ion concentrations (>20mol%), the Li+ ions no longer occupy Ho3+ sites and only occupy interstitial regions because of the large charge imbalance. Therefore, the values of the lattice parameters begin to increase.

Fig. 1. XRD patterns of HoPO4:xLi+ crystals (x=0.0, 0.1, 0.2, and 0.3); inset, Li+ concentration dependence of unit cell parameters.

下载图片 查看所有图片

At present, great effort has been devoted to investigating the internal impact of metal-ion doping on nanocrystallite size and shape. It has been reported the Ti-doped CeO2 makes nanoparticles become more perfect (spherical) with increasing TiO2 content^[9]. This may be an efficient method to modify the shape, size, and phase of crystals by metal ion doping^[10]. Figure 2 shows the SEM images of HoPO4 powders with different concentrations of Li+ ion doping. Notably, after Li+ ion doping, the shape and size of the HoPO4 crystals were remarkably modified. Figure 2(a) shows the shape is irregular and the size is very small. With Li+ ion doping, the particles have a large size variation and tend to form flake particles. This phenomenon indicates that the introduction of Li+ ions reduces the eutectic melting point and thereby promotes crystal growth and aggregation of the particles, resulting in an increase of the grain size of the powder.

Fig. 2. SEM images of HoPO4:xLi+ powder; (a) x=0.0; (b) 0.1; (c) 0.2; (d) 0.3.

下载图片 查看所有图片

The photochromic phenomenon of HoPO4 is due to its efficient light absorption in a discrete wavelength range. As is known to researchers, sunlight is a light source of a broad spectral distribution, while a tri-phosphor fluorescent lamp has a discrete spectral distribution in the blue, green, and red region. The color is the same when a common material is set aside under the sun and under the tri-phosphor fluorescent lamp. Unlike common materials, HoPO4 powders have strong and discrete absorption bands in the visible region. Thus under illumination of a specific light source, part of the light will be absorbed and the rest will be reflected. The wavelength distributions of the sun and the tri-phosphor fluorescent lamp are different. Consequently, the wavelengths reflected by the HoPO4 crystals will also be different. As a result, the HoPO4 powder will exhibit different colors under illumination of the two different light sources.

Figure 3(a) is the diffuse reflection spectra of HoPO4 powder. It exhibits several stronger absorption bands in the visible region (from 360 to 760 nm). The formation of the absorption bands of HoPO4 powders is attributed to the 4f−4f transitions of Ho3+ ions from the ground state to different excited states. The four main absorption bands are centered at 419.5, 457, 541, and 661 nm, which are assigned I85 to (G5,G3)5, G65, S25(F54), and F55 transitions, respectively^[11]. Figure 3(b) shows a typical spectrum of the tri-phosphor fluorescent lamp. From the spectrum we can see that there are three main and strong emission peaks, which are located at 435, 545, and 610 nm, respectively. The other emissions have a relatively weak intensity.

Fig. 3. (a) Diffuse reflection spectrum of HoPO4 powder; (b) emission spectrum of the tri-phosphor fluorescent lamp.

下载图片 查看所有图片

The HoPO4 powder presents a pink color under a tri-phosphor fluorescent lamp and then earthy yellow under sunlight. The color change of HoPO4 powder under illumination of a tri-phosphor fluorescent lamp can be easily understood by a comparison of the emission spectrum of the tri-phosphor fluorescent lamp with the diffuse reflection spectrum of HoPO4 powder in Fig. 3. Figure 3 shows that the HoPO4 powder has a strong absorption band at 541 nm, of which the reflective absorption coefficient is almost 67%. This means that most of the green light at 545 nm that is emitted by the tri-phosphor fluorescent lamp can be absorbed. In the blue region, a part of the blue light emitted by the tri-phosphor fluorescent lamp can also be absorbed. However, the red emission centered at 611 nm is hardly absorbed because of the mismatch between the absorption wavelength of HoPO4 and the emission wavelength of the tri-phosphor fluorescent lamp in the red region. Therefore, the reflected weak blue light plus the strong red light will present a pink color when the HoPO4 powder is exposed to the tri-phosphor fluorescent lamp. The color illuminated by sunlight can also be analyzed by the same method and it is found that the HoPO4 powder presents an earth yellow after part of the light is absorbed.

Obviously, the photochromic effect of HoPO4 powder relies on its absorption efficiency. The larger the absorption efficiency, the more remarkable the photochromic effect will be. Thus in order to improve the photochromic effect of HoPO4 powder, we attempted to introduce Li+ ions into the HoPO4 powder. Figure 4 is the diffuse reflection spectra of the HoPO4:xLi+ crystals (x=0.0, 0.1, 0.2, and 0.3). From the spectra, we can see the absorption intensity increases with x increasing from 0 to 0.1, then decreases at higher Li+ ion doping concentrations. By comparison, it is obvious to see that doping by using Li+ ions can improve the absorption efficiency of HoPO4.

Fig. 4. Diffuse spectrum of HoPO4:xLi+ (x=0.0, 0.1, 0.2, and 0.3).

下载图片 查看所有图片

To explore Li+-ion-induced spectral properties and thus the photochromic properties, powders of Eu0.05Y0.95PO4:xLi+ were prepared with the same method, in order to detect the variation of local environment by Eu3+ spectroscopy. As researchers know, yttrium orthophosphate^[12] and holmium orthophosphate both have a tetragonal structure and their space groups are the same [I41/amd]. Therefore, the Li+-doping-induced variation of the local environment in HoPO4 can be reflect by Eu3+ spectroscopy in a Li+-doped YPO4 host to some extent. The emission spectra (normalized at 592 nm) of Eu0.05Y0.95PO4 and Eu0.05Y0.95PO4:0.1Li+ under 396 nm excitation are shown in Fig. 5, inset. The emission spectra of Eu0.05Y0.95PO4:xLi+ mainly consist of a sharp peak at approximately 592 nm along with a weaker peak at approximately 617 nm. The 592 and 617 nm emissions are attributed to the D05→F17 and D05→F27 transitions of Eu3+, respectively.

Fig. 5. Ratio of D05→F17/D05→F27 as a function of Li+ content; inset, emission spectra of Eu0.05Y0.95PO4:xLi+ (x=0.0 and 0.1) under 396 nm excitation.

下载图片 查看所有图片

The D05→F27 transition of Eu3+ ion is a hypersensitive electronic transition, which is markedly affected by the coordination environment of Eu3+, whereas the D05→F17 transition is the magnetic dipole transition and is much less sensitive to the coordination environment^{[13–16" target="_self" style="display: inline;">–16]}. Therefore, the relative intensity ratio T termed the symmetry ratio, which is defined as D05→F17/D05→F27, gives a measure of the degree of distortion from the inversion symmetry of the local environment surrounding the Eu3+ ions in the host matrix^[12,17]. A smaller value of T means a lower local symmetry of the luminescence center and thus possibly higher luminescence intensity. The T value as a function of Li+ concentration is shown in Fig. 5. The result suggests the local symmetry of luminescence centers in lattice decreases with x increasing from 0 to 0.1, then increases at higher Li+-ion-doping concentration. Clearly the doping of Li+ will change the symmetry of the local environment of Eu3+^[12,18]. Considering the same crystalline structure between YPO4 and HoPO4, the Li+-ion-induced decrease of local symmetry will also take place in HoPO4. Therefore, it is reasonable to believe that the doped Li+ ions in HoPO4 of suitable concentration decrease the local symmetry of the luminescence centers in the lattice.

This suggests that when Li+ ions are doped into HoPO4, Li+ ions may occupy sites in or diffuse to various interstitial sites in the HoPO4 lattice, making the change of Coulomb interactions and Ho–O bond slightly distorted. The Ho–O bond distortion alters the local symmetry of the crystal field around Ho3+. The intra-4f electronic transitions that are parity-forbidden in principle become partially allowed, increasing the intra-4f electronic transition probability of the RE3+ ions^[19]. As a result, there is enhancement of the absorption intensity of Ho3+ ions in accordance with Li+ doping.

Li+ ions might occupy both interstitial sites and Ho3+ sites, which depends on the Li+-ion-doping concentration. Both kinds of Li+ ion in HoPO4 will give rise to the formation of a defective structure^[18] and thus decrease the local symmetry environment of the Ho3+ ions. Thus the doped Li+ ions of proper concentration will enhance the electronic dipole transition probability of Ho3+ and intensify its absorption cross section, which has been proved by Eu3+ probing spectroscopy. However, the absorption intensity of the powder tends to decrease when the doped Li+ concentration is beyond 10%. Presumably, the doped Li+ ions induced an enhancement of the electronic dipole transition probability of Ho3+ by changing its local environment, which is no longer significant at a Li+ doping concentration over 10%. As shown in Fig. 5, the symmetry ratio of Eu3+ no longer decreases but slight increases when the Ho3+ concentration is over 10%. These results mean that the absorption cross section of Ho3+ will not effectively be enhanced when the Ho3+ concentration is over 10%. The Ho3+ concentration per unit volume will decrease with the increase of the Li+ ion content, leading to the decrease of the absorption intensity. As a result, the photochromic effect begins to deteriorate at higher Li+ doping concentration.

In conclusion, a series of HoPO4:xLi+ powders are prepared using the co-precipitation method. It is believed that the doped Li+ ions of proper concentration arouse the variation of the asymmetry and covalency at the Ho3+ sites, leading to the enhancement of the 4f−4f transition of Ho3+ and therefore improvement of the photochromic effect. This work indicates that the HoPO4:xLi+ powders at x=0.1 presents the most significant color change under illumination of sunlight and tri-phosphor fluorescent light. The improved HoPO4 powder has potential wide applications for use as a photochromic pigment in the fields of security and decoration.