The Synthesis of a Core-Shell Photocatalyst Material YF3:Ho3+@TiO2 and Investigation of Its Photocatalytic Properties

In this paper, YF3:Ho3+@TiO2 core-shell nanomaterials were prepared by hydrolysis of tetra-n-butyl titanate (TBOT) using polyvinylpyrrolidone K-30 (PVP) as the coupling agent. Characterization methods including X-ray diffraction (XRD), transmission electron microscopy (TEM), energy-dispersive X-ray spectroscopy (EDS) under TEM, X-ray photoelectron spectroscopy (XPS), fluorescence spectrometry, ultraviolet-visible diffuse reflectance spectroscopy, and electron spin resonance (ESR) were used to characterize the properties and working mechanism of the prepared photocatalyst material. They indicated that the core phase YF3 nanoparticles were successfully coated with a TiO2 shell and the length of the composite was roughly 100 nm. The Ho3+ single-doped YF3:Ho3+@TiO2 displayed strong visible absorption peaks with wavelengths of 450, 537, and 644 nm, respectively. By selecting these three peaks as excitation wavelengths, we could observe 288 nm (5D4→5I8) ultraviolet emission, which confirmed that there was indeed an energy transfer from YF3:Ho3+ to anatase TiO2. In addition, this paper investigated the influences of different TBOT dosages on photocatalysis performance of the as-prepared photocatalyst material. Results showed that the YF3:Ho3+@TiO2 core-shell nanomaterial was an advanced visible-light-driven catalyst, which decomposed approximately 67% of rhodamine b (RhB) and 34.6% of phenol after 10 h of photocatalysis reaction. Compared with the blank experiment, the photocatalysis efficiency was significantly improved. Finally, the visible-light-responsive photocatalytic mechanism of YF3:Ho3+@TiO2 core-shell materials and the influencing factors of photocatalytic degradation were investigated to study the apparent kinetics, which provides a theoretical basis for improving the structural design and functions of this new type of catalytic material.


Introduction
With the development of industry, organic matter such as drugs, pesticides, surfactants, and raw chemical materials cause an increasing amount of pollutants in surface water, groundwater, sewage, and drinking water. It is even worse that most of these contaminants are complex and non-biodegradable, and therefore traditional water treatment methods cannot completely remove them. In recent years, photocatalysis has gained increasing attention due to the discovery of water splitting on a semiconductor electrode [1][2][3][4]. Over the past 20 years, photocatalysis has become a the degradation efficiency of the catalyst. The YF 3 :Ho 3+ @TiO 2 photocatalyst was prepared in this work by the hydrolysis of tetrabutyltitanate (TBOT) using PVP as the coupling agent. The influences of different dosages of TBOT on the materials' morphology, size, and photocatalysis efficiency were investigated. In addition, the photocatalysis mechanism of YF 3 :Ho 3+ @TiO 2 and the apparent kinetics of RhB degradation are discussed in details.

Preparation of YF 3 :Ho 3+ @TiO 2 Photocatalyst
In this paper, YF 3 nanoparticles were first prepared based on Jun's study [42,43], and then the YF 3 :Ho 3+ @TiO 2 photocatalyst was prepared by hydrolysis of TBOT using PVP as the coupling agent based on Qin's work. In preparation processes, TBOT (6.0 mL) was first dissolved in ethanol (30.0 mL) and CH 3 COOH (2.0 mL), and then the solution was vigorously stirred for 30 min to form precursor A; YF 3 :Ho 3+ nanoparticles (0.02 g) were dispersed in ethanol (20.0 mL) and H 2 O (4.0 mL) to form precursor B. After that, precursor B was added dropwise into precursor A at a rate of 1 mL/min while stirring for 1 h. After standing for 24 h, the resulting nanoparticles were dried at 105 • C, and then calcined by a heating rate of 2 • C/min to 400 • C for 2 h. The experimental parameters of the TBOT dosage and the hydrolysis reaction time are shown in Table 1 below.

Photocatalytic Activity Measurements
In this research, RhB and phenol were used to test the photocatalytic activity of YF 3 :Ho 3+ @TiO 2 . The photocatalytic activity of YF 3 :Ho 3+ @TiO 2 was evaluated via degradation of RhB and phenol under the irradiation of a 500 W Long arc xenon lamp with a UV cutoff filter (λ > 420 nm) under laboratory conditions using a Hitachi U-3010 UV-Vis spectrophotometer (Hitachi Corp., Tokyo, Japan).
For specific test procedures, first add 0.15 g of photocatalyst material in 500 mL of 5 mg/L solution of rhodamine B (RhB) and 500 mL of 5 mg/L solution of phenol respectively for a dark-reaction for half an hour, so as to achieve adsorption-desorption equilibrium between the pollutants and photocatalyst. Then, place the reaction system 30 cm away from the light source to be irradiated for 10 h. Take out 8 mL of the samples once every 2 h. Finally, perform centrifugation treatment for the samples and then test the absorbance of RhB at 552 nm using UV-Vis. Test the absorbance of phenol at 510 nm using 4-amino antipyrine as the chromogenic reagent under UV-Vis.

X-ray Diffraction (XRD) Pattern Analysis
The phase structures of the materials were characterized by XRD measurements. The XRD patterns of pure TiO 2 , YF 3 , and YF 3 :Ho 3+ @TiO 2 with different TBOT dosages are shown in Figure 1. All the diffractions of the YF 3 :Ho 3+ @TiO 2 could be assigned to the anatase TiO 2 (JCPDS No. 21-1272). As we all know, among all the crystal types of TiO 2 , the anatase TiO 2 has the highest photocatalysis efficiency. Hence, the prepared materials have excellent photocatalytic capacity.
In addition, the XRD shows that only YF 3 :Ho 3+ @TiO 2 doped with 0.1 mL TBOT coincides weakly with the YF 3 standard (JCPDS No. 74-0911), while the others only show the diffraction peak of anatase TiO 2 . This is mainly because when the dosage of TBOT was too high, the content of TiO 2 in the material would be relatively high, making the YF 3 :Ho 3+ content lower than its detection limit. It may also be due to the fact that when YF 3 :Ho 3+ was covered by TiO 2 , the strong diffraction peak of TiO 2 would hide the diffraction peak of YF 3 :Ho 3+ , so that the diffraction peak of YF 3 :Ho 3+ would show up only when the dosage of TBOT was reduced to a certain degree.
From the XRD results, we can see that the photocatalyst material YF 3 :Ho 3+ @TiO 2 is synthesized by the hydrolysis of TBOT. Changing the dosage of the TBOT cannot change the phase and crystal of the UCL material, but can only affect the relative proportion between anatase TiO 2 and YF 3 :Ho 3+ , thereby affecting the degradation efficiency.

X-ray Diffraction (XRD) Pattern Analysis
The phase structures of the materials were characterized by XRD measurements. The XRD patterns of pure TiO2, YF3, and YF3:Ho 3+ @TiO2 with different TBOT dosages are shown in Figure 1. All the diffractions of the YF3:Ho 3+ @TiO2 could be assigned to the anatase TiO2 (JCPDS No. 21-1272). As we all know, among all the crystal types of TiO2, the anatase TiO2 has the highest photocatalysis efficiency. Hence, the prepared materials have excellent photocatalytic capacity.
In addition, the XRD shows that only YF3:Ho 3+ @TiO2 doped with 0.1 mL TBOT coincides weakly with the YF3 standard (JCPDS No. 74-0911), while the others only show the diffraction peak of anatase TiO2. This is mainly because when the dosage of TBOT was too high, the content of TiO2 in the material would be relatively high, making the YF3:Ho 3+ content lower than its detection limit. It may also be due to the fact that when YF3:Ho 3+ was covered by TiO2, the strong diffraction peak of TiO2 would hide the diffraction peak of YF3:Ho 3+ , so that the diffraction peak of YF3:Ho 3+ would show up only when the dosage of TBOT was reduced to a certain degree.
From the XRD results, we can see that the photocatalyst material YF3:Ho 3+ @TiO2 is synthesized by the hydrolysis of TBOT. Changing the dosage of the TBOT cannot change the phase and crystal of the UCL material, but can only affect the relative proportion between anatase TiO2 and YF3:Ho 3+ , thereby affecting the degradation efficiency.

TEM
The TEM images of YF 3 :Ho 3+ @TiO 2 with different dosages of TBOT are shown in Figure 2a-g, from which we can see that TiO 2 nanoparticles with particle sizes of about 10 nm stick to the UCL material. YF 3 :Ho 3+ @TiO 2 is overall homogeneous, but in an agglomerated state. The amount of TiO 2 doped on the YF 3 :Ho 3+ increases with the increase of the TBOT dosage. When the dosage of TBOT is small, TiO 2 cannot evenly coat the YF 3 :Ho 3+ , making a part of the YF 3 :Ho 3+ still exposed. With the increase of the TBOT dosage, YF 3 :Ho 3+ is coated by the TiO 2 particles gradually. Figure 2h shows the high resolution image of TiO 2 shell coated on the surface of the YF 3 . Figure 2j shows the EDS line scanning profiles which are recorded along the white line as presented in Figure 2i. The EDX elementary line scanning was used to further determine the composition of the composite, and to prove whether the synthesized materials were of a core-shell structure or not. As shown in Figure 2j, there are signal detections of both YF 3 :Ho 3+ and TiO 2 at point A where the scanning starts. As the scanning goes outside and comes close to point B, the signal of Ti drops while Y increases. At point C, Ti gradually reduces to the minimum level while Y grows to the maximum level. When the scanning reaches the other end of YF 3 :Ho 3+ at point D, Y begins to decrease while Ti gradually increases. This is obvious evidence to prove that the synthesized materials have a core-shell structure and the TiO 2 is strongly coupled on YF 3 :Ho 3+ .

TEM
The TEM images of YF3:Ho 3+ @TiO2 with different dosages of TBOT are shown in Figure 2a-g, from which we can see that TiO2 nanoparticles with particle sizes of about 10 nm stick to the UCL material. YF3:Ho 3+ @TiO2 is overall homogeneous, but in an agglomerated state. The amount of TiO2 doped on the YF3:Ho 3+ increases with the increase of the TBOT dosage. When the dosage of TBOT is small, TiO2 cannot evenly coat the YF3:Ho 3+ , making a part of the YF3:Ho 3+ still exposed. With the increase of the TBOT dosage, YF3:Ho 3+ is coated by the TiO2 particles gradually. Figure 2h shows the high resolution image of TiO2 shell coated on the surface of the YF3. Figure 2j shows the EDS line scanning profiles which are recorded along the white line as presented in Figure 2i. The EDX elementary line scanning was used to further determine the composition of the composite, and to prove whether the synthesized materials were of a core-shell structure or not. As shown in Figure 2j, there are signal detections of both YF3:Ho 3+ and TiO2 at point A where the scanning starts. As the scanning goes outside and comes close to point B, the signal of Ti drops while Y increases. At point C, Ti gradually reduces to the minimum level while Y grows to the maximum level. When the scanning reaches the other end of YF3:Ho 3+ at point D, Y begins to decrease while Ti gradually increases. This is obvious evidence to prove that the synthesized materials have a core-shell structure and the TiO2 is strongly coupled on YF3:Ho 3+ .

Chemical States Investigation by X-ray Photoelectron Spectroscopy (XPS)
X-ray photoelectron spectroscopy (XPS) was used to examine the chemical states of the elements on the surface of the YF3:Ho 3+ @TiO2 core-shell materials. Figure 3a shows the full survey spectrum which reveals the co-presence of Ti, O, Y, F, and Ho. In Figure 3b, the binding energy of 462.98 and 457.28 eV, which are respectively labeled as Ti 2p1/2 and Ti 2p3/2, are consistent with the typical values reported for TiO2 [44]. According to the asymmetric profile of O 1s shown in Figure 3c, it can be seen

Chemical States Investigation by X-ray Photoelectron Spectroscopy (XPS)
X-ray photoelectron spectroscopy (XPS) was used to examine the chemical states of the elements on the surface of the YF 3 :Ho 3+ @TiO 2 core-shell materials. Figure 3a shows the full survey spectrum which reveals the co-presence of Ti, O, Y, F, and Ho. In Figure 3b, the binding energy of 462.98 and 457.28 eV, which are respectively labeled as Ti 2p 1/2 and Ti 2p 3/2 , are consistent with the typical values reported for TiO 2 [44]. According to the asymmetric profile of O 1s shown in Figure 3c, it can be seen that more than one kind of oxygen species exists. It was reported that when the binding energy was around 528.58 eV, the peak corresponded to the characteristic peak of Ti-O-Ti; while if the binding energy was around 530.08 eV, the peak was attributed to H-O. According to Figure 3d, the three peaks at 156.38, 158.88, and 161.08 eV all corresponded to Y 3d. Element F shows two peaks at 682.78 and 685.28 eV, respectively, which correspond to F 1s (see Figure 3e). In addition to the main elements in the YF 3 nanoparticles, the doping elements in these nanoparticles were also detected. The peaks at 156.28, 158.78, and 160.88 eV (see Figure 3f) are attributed to the Ho 3+ ions. XPS results show that rare earth ions have been successfully incorporated into the YF 3 host matrix.
Materials 2017, 10, 302 6 of 14 that more than one kind of oxygen species exists. It was reported that when the binding energy was around 528.58 eV, the peak corresponded to the characteristic peak of Ti-O-Ti; while if the binding energy was around 530.08 eV, the peak was attributed to H-O. According to Figure 3d, the three peaks at 156.38, 158.88, and 161.08 eV all corresponded to Y 3d. Element F shows two peaks at 682.78 and 685.28 eV, respectively, which correspond to F 1s (see Figure 3e). In addition to the main elements in the YF3 nanoparticles, the doping elements in these nanoparticles were also detected. The peaks at 156.28, 158.78, and 160.88 eV (see Figure 3f) are attributed to the Ho 3+ ions. XPS results show that rare earth ions have been successfully incorporated into the YF3 host matrix.

UV-Vis Diffuse Reflection Spectroscopy
To analyze the optimal absorption wavelength of the synthesized YF3:Ho 3+ @TiO2 material, the UV-Vis diffuse reflection spectroscopy was investigated. Figure 3 shows the representative spectra of YF3:Ho 3+ @TiO2 and YF3:Ho 3+ . From the spectrum of YF3:Ho 3+ @TiO2, we can observe a light absorption edge before 400 nm, which is overlapped with that of TiO2. Moreover, it confirms that the YF3:Ho 3+ @TiO2 material can absorb the light with wavelengths between 300-700 nm, which is shown in the spectrum of YF3:Ho 3+ . There are three absorption peaks in the visible light region (450 nm, 538 nm, 644 nm), where the intensity of the 450 nm peak is relatively higher. As we can see from Figure 4, for one wavelength, the stronger the absorption peak's light absorption ability is, the more suitable that wavelength is for excitation. Hence, 450 nm was selected as the excitation wavelength of YF3:Ho 3+ @TiO2, which is consistent with the goal of utilizing visible light as the excitation source for UCL materials.

UV-Vis Diffuse Reflection Spectroscopy
To analyze the optimal absorption wavelength of the synthesized YF 3 :Ho 3+ @TiO 2 material, the UV-Vis diffuse reflection spectroscopy was investigated. Figure 3 shows the representative spectra of YF 3 :Ho 3+ @TiO 2 and YF 3 :Ho 3+ . From the spectrum of YF 3 :Ho 3+ @TiO 2, we can observe a light absorption edge before 400 nm, which is overlapped with that of TiO 2 . Moreover, it confirms that the YF 3 :Ho 3+ @TiO 2 material can absorb the light with wavelengths between 300 and 700 nm, which is shown in the spectrum of YF 3 :Ho 3+ . There are three absorption peaks in the visible light region (450 nm, 538 nm, 644 nm), where the intensity of the 450 nm peak is relatively higher. As we can see from Figure 4, for one wavelength, the stronger the absorption peak's light absorption ability is, the more suitable that wavelength is for excitation. Hence, 450 nm was selected as the excitation wavelength of   Figure 5 shows the fluorescence emission spectra of YF3:Ho 3+ @TiO2 under the visible light excitation at 450 nm, from which we can see that all the prepared samples share the similar upconversion luminescence properties as YF3:Ho 3+ . There is a strong emission peak at 288 nm, which resulted from the transition of the Ho 3+ ion from 5 D4→ 5 I8. When YF3:Ho 3+ is doped with TiO2, the upconversion luminescence capacity becomes weaker. This may be due to the fact that the UV light emitted by the UCL after absorbing visible light would be absorbed by TiO2, and moreover, with the increase of the TBOT dosage, the TiO2 doped on the YF3:Ho 3+ would hinder the excitation light from arriving at the YF3:Ho 3+ , resulting in reduced excitation light energy for YF3:Ho 3+ [45]. In addition, a large number of TiO2 loaded on the YF3 surface would also absorb most of the UV light and decrease the excitation intensity of YF3:Ho 3+ .   Figure 5 shows the fluorescence emission spectra of YF 3 :Ho 3+ @TiO 2 under the visible light excitation at 450 nm, from which we can see that all the prepared samples share the similar upconversion luminescence properties as YF 3 :Ho 3+ . There is a strong emission peak at 288 nm, which resulted from the transition of the Ho 3+ ion from 5 D 4 → 5 I 8 . When YF 3 :Ho 3+ is doped with TiO 2 , the upconversion luminescence capacity becomes weaker. This may be due to the fact that the UV light emitted by the UCL after absorbing visible light would be absorbed by TiO 2 , and moreover, with the increase of the TBOT dosage, the TiO 2 doped on the YF 3 :Ho 3+ would hinder the excitation light from arriving at the YF 3 :Ho 3+ , resulting in reduced excitation light energy for YF 3 :Ho 3+ [45]. In addition, a large number of TiO 2 loaded on the YF 3 surface would also absorb most of the UV light and decrease the excitation intensity of YF 3 :Ho 3+ .  Figure 5 shows the fluorescence emission spectra of YF3:Ho 3+ @TiO2 under the visible light excitation at 450 nm, from which we can see that all the prepared samples share the similar upconversion luminescence properties as YF3:Ho 3+ . There is a strong emission peak at 288 nm, which resulted from the transition of the Ho 3+ ion from 5 D4→ 5 I8. When YF3:Ho 3+ is doped with TiO2, the upconversion luminescence capacity becomes weaker. This may be due to the fact that the UV light emitted by the UCL after absorbing visible light would be absorbed by TiO2, and moreover, with the increase of the TBOT dosage, the TiO2 doped on the YF3:Ho 3+ would hinder the excitation light from arriving at the YF3:Ho 3+ , resulting in reduced excitation light energy for YF3:Ho 3+ [45]. In addition, a large number of TiO2 loaded on the YF3 surface would also absorb most of the UV light and decrease the excitation intensity of YF3:Ho 3+ . Figure 5. Upconversion luminescence spectra of YF3:Ho 3+ @TiO2 photocatalysts with different TBOT dosages. Figure 5. Upconversion luminescence spectra of YF 3 :Ho 3+ @TiO 2 photocatalysts with different TBOT dosages.

Photocatalysis Mechanism of RhB Degradation
According to Jun's study [43], there are two upconversion mechanisms for YF 3 :Ho 3+ , of which one is a two-photon upconversion fluorescence mechanism and the other one is a three-photon upconversion fluorescence mechanism.
For the capacity of generating radicals, such as the DMPO-hydroxyl radical (·OH) and DMPO-superoxide radical (·O 2 − ), 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) has been generally used for trapping radicals. According to the results shown in Figure 6, we can see that the signals of ·OH and ·O 2 − are obvious and clear. The intensities of these two radicals' signals increase considerably after irradiation for 6 min. Hence, the ·OH and ·O 2 − are two main oxidative species for the YF 3 :Ho 3+ @TiO 2 system. Moreover, RhB is speculated to react with these two radicals during its photocatalytic degradation, and the response equation is shown as follows.
YF 3 :Ho 3 + + visible light → YF 3 :Ho 3 + + UV (R1) Besides, based on the response equation, this paper also theorizes a possible reaction process shown in Figure 7. Firstly, the UCL material YF 3 :Ho 3+ emits UV light after absorbing visible light (R1). Then, TiO 2 trapped on the UCL surface will generate electron-hole pairs through UV excitation (R2

Photocatalysis Mechanism of RhB Degradation
According to Jun's study [43], there are two upconversion mechanisms for YF3:Ho 3+ , of which one is a two-photon upconversion fluorescence mechanism and the other one is a three-photon upconversion fluorescence mechanism.
For the capacity of generating radicals, such as the DMPO-hydroxyl radical (·OH) and DMPOsuperoxide radical (·O2 − ), 5,5-dimethyl-1-pyrroline-N-oxide (DMPO) has been generally used for trapping radicals. According to the results shown in Figure 6, we can see that the signals of ·OH and ·O2 − are obvious and clear. The intensities of these two radicals' signals increase considerably after irradiation for 6 min. Hence, the ·OH and ·O2 − are two main oxidative species for the YF3:Ho 3+ @TiO2 system. Moreover, RhB is speculated to react with these two radicals during its photocatalytic degradation, and the response equation is shown as follows. Besides, based on the response equation, this paper also theorizes a possible reaction process shown in Figure 7. Firstly, the UCL material YF3:Ho 3+ emits UV light after absorbing visible light (R1). Then, TiO2 trapped on the UCL surface will generate electron-hole pairs through UV excitation (R2). The excited electrons on the TiO2 surface react with oxygen to form ·O2 − and HO2, and then the resulted HO2 combines with H + to form hydrogen peroxide later (H2O2; R4). When ·O2 − meets H2O2, it will generate ·OH (R5); meanwhile, the photogenerated holes at (R2) will react with H2O to form ·OH (R3). For the conduction band of TiO2 located above the RhB redox potential, the oxygen species (·OH, ·O2 − , and H2O2) could oxidize the RhB to realize the purpose of RhB's degradation.    Figure 8a,b show the influences of YF3:Ho 3+ @TiO2 with different dosages of TBOT on RhB and phenol degradation, respectively. All the samples have almost zero adsorption efficiency both for RhB and phenol, wherein the most significant one is less than 1%. For the RhB and phenol, the photocatalytic degradation efficiency first increases then decreases with the increase of the TBOT dosage. When the dosages of TBOT are 6 mL, it can obtain the highest degradation efficiency (67%) of RhB as well as the highest degradation efficiency (34.6%) of phenol. According to the analyses above, this could be due to the fact that when the TBOT dosage increased, more TiO2 would be doped on the YF3:Ho 3+ , and thus more UV light would be absorbed, resulting in more excited electron-hole pairs and an improved degradation efficiency of RhB and phenol. However, when the dosage of TBOT continued to increase to 8 mL, the degradation efficiency would be decreased for both RhB and phenol. According to the TEM of YF3:Ho 3+ in Figure 3, when the dosage reaches 8 mL, the YF3:Ho 3+ is covered by TiO2 completely, which will affect the absorption of visible light and the conversion from visible light to UV light, thus reducing the photocatalytic activity of YF3:Ho 3+ @TiO2. This paper also tested the photocatalytic efficiency of P25, TiO2 (prepared using a similar method as YF3:Ho 3+ @TiO2 without YF3:Ho 3+ ), and BiVO4, respectively. For RhB degradation, the results showed that the efficiency of P25 was 12.2%, the efficiency of TiO2 was 17.8%, and the efficiency of BiVO4 was 22.6%. For phenol degradation, the results showed the efficiency of P25 and TiO2 was nearly zero, and the efficiency of BiVO4 was 9.7%, all of which were less than that of YF3:Ho 3+ @TiO2. Therefore, it can be concluded that the YF3:Ho 3+ @TiO2 material can make up for the defects of TiO2 (which is unable to have a photocatalytic reaction under visible light irritation), and has a higher photocatalytic efficiency than the common visible light photocatalyst BiVO4.   Figure 8a,b show the influences of YF 3 :Ho 3+ @TiO 2 with different dosages of TBOT on RhB and phenol degradation, respectively. All the samples have almost zero adsorption efficiency both for RhB and phenol, wherein the most significant one is less than 1%. For the RhB and phenol, the photocatalytic degradation efficiency first increases then decreases with the increase of the TBOT dosage. When the dosages of TBOT are 6 mL, it can obtain the highest degradation efficiency (67%) of RhB as well as the highest degradation efficiency (34.6%) of phenol. According to the analyses above, this could be due to the fact that when the TBOT dosage increased, more TiO 2 would be doped on the YF 3 :Ho 3+ , and thus more UV light would be absorbed, resulting in more excited electron-hole pairs and an improved degradation efficiency of RhB and phenol. However, when the dosage of TBOT continued to increase to 8 mL, the degradation efficiency would be decreased for both RhB and phenol. According to the TEM of YF 3 :Ho 3+ in Figure 3, when the dosage reaches 8 mL, the YF 3 :Ho 3+ is covered by TiO 2 completely, which will affect the absorption of visible light and the conversion from visible light to UV light, thus reducing the photocatalytic activity of YF 3 :Ho 3+ @TiO 2 .

Photocatalysis Application
This paper also tested the photocatalytic efficiency of P25, TiO 2 (prepared using a similar method as YF 3 :Ho 3+ @TiO 2 without YF 3 :Ho 3+ ), and BiVO 4 , respectively. For RhB degradation, the results showed that the efficiency of P25 was 12.2%, the efficiency of TiO 2 was 17.8%, and the efficiency of BiVO 4 was 22.6%. For phenol degradation, the results showed the efficiency of P25 and TiO 2 was nearly zero, and the efficiency of BiVO 4 was 9.7%, all of which were less than that of YF 3 :Ho 3+ @TiO 2 . Therefore, it can be concluded that the YF 3 :Ho 3+ @TiO 2 material can make up for the defects of TiO 2 (which is unable to have a photocatalytic reaction under visible light irritation), and has a higher photocatalytic efficiency than the common visible light photocatalyst BiVO 4 .  Figure 8a,b show the influences of YF3:Ho 3+ @TiO2 with different dosages of TBOT on RhB and phenol degradation, respectively. All the samples have almost zero adsorption efficiency both for RhB and phenol, wherein the most significant one is less than 1%. For the RhB and phenol, the photocatalytic degradation efficiency first increases then decreases with the increase of the TBOT dosage. When the dosages of TBOT are 6 mL, it can obtain the highest degradation efficiency (67%) of RhB as well as the highest degradation efficiency (34.6%) of phenol. According to the analyses above, this could be due to the fact that when the TBOT dosage increased, more TiO2 would be doped on the YF3:Ho 3+ , and thus more UV light would be absorbed, resulting in more excited electron-hole pairs and an improved degradation efficiency of RhB and phenol. However, when the dosage of TBOT continued to increase to 8 mL, the degradation efficiency would be decreased for both RhB and phenol. According to the TEM of YF3:Ho 3+ in Figure 3, when the dosage reaches 8 mL, the YF3:Ho 3+ is covered by TiO2 completely, which will affect the absorption of visible light and the conversion from visible light to UV light, thus reducing the photocatalytic activity of YF3:Ho 3+ @TiO2. This paper also tested the photocatalytic efficiency of P25, TiO2 (prepared using a similar method as YF3:Ho 3+ @TiO2 without YF3:Ho 3+ ), and BiVO4, respectively. For RhB degradation, the results showed that the efficiency of P25 was 12.2%, the efficiency of TiO2 was 17.8%, and the efficiency of BiVO4 was 22.6%. For phenol degradation, the results showed the efficiency of P25 and TiO2 was nearly zero, and the efficiency of BiVO4 was 9.7%, all of which were less than that of YF3:Ho 3+ @TiO2. Therefore, it can be concluded that the YF3:Ho 3+ @TiO2 material can make up for the defects of TiO2 (which is unable to have a photocatalytic reaction under visible light irritation), and has a higher photocatalytic efficiency than the common visible light photocatalyst BiVO4.

Influencing Factors of the Photocatalytic Degradation Reaction
The influencing factors of photocatalytic reaction include photocatalyst dosage (m cata ), substrate concentration (C 0 ), and irradiation intensity (E), which are also the three major factors that affect the k a coefficient of the photocatalytic kinetics equation, according to the Langmuir-Hinshelwood equation (other factors are not considered). This paper has studied a series of experiments to figure out how the apparent photocatalytic degradation kinetics change with these three factors.
3.7.1. The effect of m cata Figure 9 shows how different dosages of YF 3 :Ho 3+ @TiO 2 affect the photocatalytic efficiency. For the experiment, this paper changes the dosages of the photocatalyst from 0.05 to 0.25 g. And the result shows that with the increasing YF 3 :Ho 3+ @TiO 2 dosage, the degradation efficiency of RhB first increases and then decreases. This may be due to the fact that the increased photocatalyst dosage not only improved the photon efficiency to generate more photogenerated radicals, but also increased effluent turbidity, causing light scattering to decrease photon efficiency. After fitting, an equation can be obtained as shown below.

Influencing Factors of the Photocatalytic Degradation Reaction
The influencing factors of photocatalytic reaction include photocatalyst dosage (mcata), substrate concentration (C0), and irradiation intensity (E), which are also the three major factors that affect the ka coefficient of the photocatalytic kinetics equation, according to the Langmuir-Hinshelwood equation (other factors are not considered). This paper has studied a series of experiments to figure out how the apparent photocatalytic degradation kinetics change with these three factors.
3.7.1. The effect of mcata Figure 9 shows how different dosages of YF3:Ho 3+ @TiO2 affect the photocatalytic efficiency. For the experiment, this paper changes the dosages of the photocatalyst from 0.05 to 0.25 g. And the result shows that with the increasing YF3:Ho 3+ @TiO2 dosage, the degradation efficiency of RhB first increases and then decreases. This may be due to the fact that the increased photocatalyst dosage not only improved the photon efficiency to generate more photogenerated radicals, but also increased effluent turbidity, causing light scattering to decrease photon efficiency. After fitting, an equation can be obtained as shown below.

The effect of C0
This paper tested different concentrations of substrate (RhB) for photodegradation. Figure 10 shows how different concentrations of substrate affect the photocatalytic efficiency. The degradation efficiency of RhB decreases with the increase of the substrate concentration. When the concentration of the substrate is 4 mg/L, the photocatalytic efficiency reaches the highest value, which is 76%, and when it is 8 mg/L, the photocatalytic efficiency reaches the lowest value, which is 56%. This is mainly because when the concentration of the substrate is relatively low, the photocatalyst YF3:Ho 3+ @TiO2 is in excess, and then the photocatalytic efficiency of RhB will be relatively high. On the contrary, when the concentration of the substrate is high, the photocatalyst YF3:Ho 3+ @TiO2 is no longer sufficient, and all of the photocatalyst needs to take part in the photodegradation. At this time, while the photodegradation rate is at its highest value, the apparent degradation efficiency will decrease with the increase of the substrate concentration, and thus  Figure 10 shows how different concentrations of substrate affect the photocatalytic efficiency. The degradation efficiency of RhB decreases with the increase of the substrate concentration. When the concentration of the substrate is 4 mg/L, the photocatalytic efficiency reaches the highest value, which is 76%, and when it is 8 mg/L, the photocatalytic efficiency reaches the lowest value, which is 56%. This is mainly because when the concentration of the substrate is relatively low, the photocatalyst YF 3 :Ho 3+ @TiO 2 is in excess, and then the photocatalytic efficiency of RhB will be relatively high. On the contrary, when the concentration of the substrate is high, the photocatalyst YF 3 :Ho 3+ @TiO 2 is no longer sufficient, and all of the photocatalyst needs to take part in the photodegradation. At this time, while the photodegradation rate is at its highest value, the apparent degradation efficiency will decrease with the increase of the substrate concentration, and thus photodegradation efficiency will be worse. However, when the concentration of the substrate is too high, it will reduce the light transmittance of the solution, thus reducing the photocatalytic activity.
After fitting, an equation can be obtained as shown below.  Figure 11 shows how different irradiation intensities affect the photocatalytic efficiency. When the irradiation intensity reaches 141,500 lx, each factor is at its best level, the RhB is degraded completely, and the process is not a zero order reaction. Therefore, in this paper, only the irradiation intensities of 87,100 lx, 52,300 lx, 43,500 lx were considered. With the increase of irradiation intensity, reactions between the photocatalyst and photon will increase and the rate of the photocatalysis reaction will be faster, thus leading to an increased efficiency. After fitting, an equation can be obtained as shown below.  Figure 11 shows how different irradiation intensities affect the photocatalytic efficiency. When the irradiation intensity reaches 141,500 lx, each factor is at its best level, the RhB is degraded completely, and the process is not a zero order reaction. Therefore, in this paper, only the irradiation intensities of 87,100 lx, 52,300 lx, 43,500 lx were considered. With the increase of irradiation intensity, reactions between the photocatalyst and photon will increase and the rate of the photocatalysis reaction will be faster, thus leading to an increased efficiency. After fitting, an equation can be obtained as shown below. The apparent kinetics model of RhB degradation is C A = C 0 -2.19967 × 10 -6 m 0.31835 C 0 Et (0.05 g ≤ m ≤ 0.15 g, 4.0 mg/L ≤ C0 ≤ 8.0 mg/L, 43500lx ≤ E ≤ 87100lx) C A = C 0 -4.97 × 10 -7 m -0.56254 C 0 Et (0.15 g ≤ m ≤ 0.25 g, 4.0 mg/L ≤ C0 ≤ 8.0 mg/L, 43500lx ≤ E ≤ 87100lx) Figure 11. Degradation rate of RhB with different irradiation intensities. Figure 11. Degradation rate of RhB with different irradiation intensities.

Conclusions
To solve the problem of TiO 2 having nearly no photocatalytic efficiency under visible light irradiation, a composite photocatalyst material YF 3 :Ho 3+ @TiO 2 was prepared in this paper using upconversion luminescence technology. Through analyzing the morphology and composition, crystal structure, and optical spectra of YF 3 :Ho 3+ @TiO 2 , it was found that this material had high photocatalytic efficiency under visible light irradiation. In addition, this paper also investigated the influences of different dosages of TBOT on the properties of the photocatalyst.
In summary, YF 3 :Ho 3+ @TiO 2 can be successfully prepared using a simple hydrothermal method. By analyzing the XRD images, we found that almost all the samples showed the diffraction peak of anatase TiO 2 and that the crystal structure of the material did not change with TBOT dosage. TiO 2 was uniformly doped on UCL, and the particle size was about 10 nm. The rice-shaped UCL material had good dispersion, of which the particle size was about 100 nm. The change of TBOT dosage would not cause the change of the material morphology, but would cause the change of the amount of TiO 2 doped on UCL, resulting in an impact on the photocatalytic activity of YF 3 :Ho 3+ @TiO 2 . The composite material prepared in this paper shared the same upconversion luminescence property with the UCL material prepared by Jun. It confirmed that the material prepared in this paper could absorb 450 nm visible light and emit UV light, with energy transferred from YF 3 :Ho 3+ to anatase TiO 2 . With the increase of the TBOT dosage, more TiO 2 would be doped on the YF 3 :Ho 3+ , therefore the excitation light was obstructed, resulting in a lower energy of exciting light. The photocatalytic properties of the YF 3 :Ho 3+ @TiO 2 was evaluated by the degradation of RhB and compared with those of traditional photocatalysts such as P25, TiO 2 , and the visible light photocatalyst BiVO 4 . The results showed that the prepared composite material exhibited better photocatalytic properties as compared with the other three photocatalysts. With the increase of the TBOT dosage, the photocatalytic efficiency of composite YF 3 :Ho 3+ @TiO 2 first increased and then decreased. When the TBOT dosage was 6 mL, the photocatalytic efficiency reached the highest value, which was 67%. The results of this paper indicated that Ho 3+ -single-doped hexagonal YF 3 could absorb visible light and emit UV light via UC processes. Under UV light irradiation, the composite material YF 3 :Ho 3+ @TiO 2 could exhibit better photocatalytic properties than that of anatase TiO 2 , therefore the prepared composite material YF 3 :Ho 3+ @TiO 2 has promising applications in photocatalysis.