Synthesis and Property Examination of Er2FeSbO7/BiTiSbO6 Heterojunction Composite Catalyst and Light-Catalyzed Retrogradation of Enrofloxacin in Pharmaceutical Waste Water under Visible Light Irradiation

A new photocatalyst, Er2FeSbO7, was prepared by solid phase sintering using the high-temperature synthesis method for the first time in this paper. Er2FeSbO7/BiTiSbO6 heterojunction (EBH) catalyst was prepared by the solvent thermal method for the first time. Er2FeSbO7 compound crystallized in the pyrochlore-type architecture and cubelike crystal system; the interspace group of Er2FeSbO7 was Fd3m and the crystal cellular parameter a of Er2FeSbO7 was 10.179902 Å. The band gap (BDG) width of Er2FeSbO7 was 1.88 eV. After visible light irradiation of 150 minutes (VLGI-150min) with EBH as a photocatalyst, the removal rate (RR) of enrofloxacin (ENR) concentration was 99.16%, and the total organic carbon (TOC) concentration RR was 94.96%. The power mechanics invariable k toward ENR consistency and visible light irradiation (VLGI) time with EBH as a photocatalyzer attained 0.02296 min−1. The power mechanics invariable k which was involved with TOC attained 0.01535 min−1. The experimental results showed that the photocatalytic degradation (PCD) of ENR within pharmaceutical waste water with EBH as a photocatalyzer under VLGI was in keeping with the single-order reactivity power mechanics. The RR of ENR with EBH as a photocatalyzer was 1.151 times, 1.269 times or 2.524 times that with Er2FeSbO7 as a photocatalyst, BiTiSbO6 as a photocatalyst, or N-doping TiO2 (N-TO) as a photocatalyst after VLGI-150min. The photocatalytic activity, which ranged from high to low among above four photocatalysts, was as follows: EBHP > Er2FeSbO7 > BiTiSbO6 > N-TO. After VLGI-150min toward three periods of the project with EBH as a photocatalyst, the RR of ENR attained 98.00%, 96.76% and 95.60%. The results showed that the stability of EBH was very high. With appending trapping agent, it could be proved that the oxidative capability for degrading ENR, which ranged from strong to weak among three oxidic radicals, was as follows: superoxide anion > hydroxyl radicals (HRS) > holes. This work provides a scientific basis for the research and oriented leader development of efficient heterojunction catalysts.


Introduction
Antibiotics were found to be stubborn pollutants in pharmaceutical wastewater and could not be completely eliminated after conventional treatment [1][2][3][4], such as biodegradation, electrochemical process, adsorption method and flocculation settling method [5]. Enrofloxacin (ENR) was one of the common antibiotics which could effectively kill Grampositive and Gram-negative bacteria and mycoplasma and be widely used to cure various diseases in animals [6][7][8][9][10]. However, a large amount of ENR in wastewater could promote and spread antibiotic resistance and had a negative impact on biota. Therefore, an efficient method for degrading ENR was urgently needed [11][12][13][14].
In recent years, various methods which could be used to degrade aqueous pollutants (especially microbiotic and organic dyestuff) had been developed, including chemical oxidation, biotechnology, electrochemical treatment, photodegradation and advanced oxidation. Due to the low cost and environmental benefits, photodegradation by photocatalyst was considered to be a promising technology. Therefore, many researchers focused on the design and synthesis of high-efficiency photocatalysts, which could convert the inexhaustible solar energy into the driving force of photodegradation. During the photodegradation process, antibiotics were effectively oxidized by superoxide radicals, hydroxyl radicals (HRS) or holes which derived from separated photoinduced electrons and photoinduced holes. This made the suitable band structure and effective electron hole separation particularly important for the photodegradation of antibiotics. Among various strategies (such as defect engineering, morphological variation element doping and heterostructure construction), heterogeneous junction engineering was considered to be the most promising method for achieving appropriate conduction band, valence band and charge separation capability at the same time. In particular, the heterostructure scheme system which was inspired by natural photosynthesis retained high redox capacity and was conducive to the formation of holes and free radicals [15].
Photocatalytic technology had developed rapidly since Fujishima and Honda discovered photocatalytic reaction in 1972, which attracted the attention of a large number of scholars in the scientific community. In 1976, Carey et al. studied the photocatalytic oxidation of PCBs and extended the photocatalytic technology to the field of eliminating environmental pollution. Since then, various new semiconductor materials, such as SrTiO 3 , BiVO 4 , Ag 3 PO 4 , TaON, Ta 3 N 5 , g-C 3 N 4 , CdS, MoS 2 , and their nanoparticles, had been directly applied for exploiting solar energy by various photocatalytic reactions (PLR) [16].
However, most of the traditional photocatalytic materials owned problems, such as low quantum efficiency [17][18][19][20][21], low visible light (VL) utilization and poor stability, which greatly restricted their further development. For example, commercial TiO 2 could only absorb ultraviolet light (UV-light), which accounted for 4% of the solar energy. Therefore, light energy could not be fully utilized [22]. For the purpose of improving the catalytic activity of photocatalysts, the ion doping method, photosensitization and heterojunction construction were proved to be very effective [23][24][25][26][27][28][29]. It was well known that small changes in the inner structure of semi-conductor photocatalysts might promote the separation efficiency of photogenerated electrons and photogenerated holes, so as to increase the photocatalytic activity. Many scientists had proven that the photocatalysts with heterojunction construction possessed preferable light utilization efficiency, a long carrier life, high photocatalytic performance and high chemical stability [30][31][32][33][34][35][36]. Among the photocatalysts with heterojunction construction, p-n heterojunction architecture had been proven by many scholars to be a valid method for separating electron and hole pairs [37]. Swain et al. fabricated a AuS 2 /ZnIn 2 S 4 p-n heterojunction photocatalyst which was prepared for the degradation of phenol in industrial sewage. From the experimental results, the degradation rate of phenol over AuS 2 /ZnIn 2 S 4 heterojunction photocatalyst was about 15 times higher than those over pure ZnIn 2 S 4 [38]. Liao and his colleagues constructed VL-driven BiFeO 3 /TiO 2 p-n heterojunction composites by the simple hydrolysis precipitation method. Moreover, BiFeO 3 /TiO 2 improved photocatalytic property owing to the ferroelectric effect of BiFeO 3 and the effect of the internal electric field of the BiFeO 3 /TiO 2 p-n heterojunction. As a result, the recombination of carriers was obviously inhibited, and above factors greatly promoted the degradation rate of tetracycline under visible light irradiation (VLGI) [39].
In our previous work [46][47][48], we found that Sm 2 FeSbO 7 had a pyrochlore architecture. As a photocatalyzer under the condition of VLGI, the anatomical prettification of Sm 2 FeSbO 7 seemed to have the potential for increasing the light-catalyzed liveness. In the light of above analyses, we could presume that the superseding of Sm 3+ by Er 3+ in Sm 2 FeSbO 7 could improve the carriers' concentration [49]. The results showed that the electronical transmittal and optical physics characteristics of the late model Er 2 FeSbO 7 compound was changed and improved and might have advanced photocatalytic properties. Moreover, the catalyst material which was prepared in this study was a new compound which had not been reported. In addition, a series of experiments with Er 2 FeSbO 7 /BiTiSbO 6 heterojunction (EBH) as a photocatalyst for the degradation of ENR were designed and compared with the previously reported composites for the degradation of ENR, reflecting the novelty of EBH. Li [50]. In addition, Chen et al. found that graphene-like boron nitride, which was modified by bismuth phosphate material, could be used for promoting the photocatalytic degradation rate of ENR. The experimental results showed that under sunlight irradiation of 120 minutes, ENR was photodegraded by BiPO 4 , while the removal rate of ENR was 79.2%, 91.5% or 83.8% by 0.5 wt% BN/BiPO 4 , 1 wt% BN/BiPO 4 or 2 wt% BN/BiPO 4 , respectively. It could be found that 1 wt% graphene-like BN/BiPO 4 had the highest degradation rate of ENR [51]. However, although this improved the degradation rate of ENR, the results were not ideal. Therefore, the new catalyst material which was studied in this paper had a higher degradation efficiency in the degradation process of ENR, which made its work more meaningful, reflecting the advantages of the new material.
In this article, X-ray diffractometer (XRD), scanning electron microscope-X ray energy dispersive spectra (SEM-EDS), X-ray photoelectron spectrograph (XPS), synchrotronbased ultraviolet photoelectron spectroscope (UPS) and UV-Vis diffuse reflectance spectrophotometer (DRS) were used for resolving the anatomical characteristics of pure phase Er 2 FeSbO 7 and pure phase BiTiSbO 6 , which were prepared by power mechanics controlment quomodo and elevated temperature solid phase fritting quomodo. Furthermore, the removal rate (RR) of ENR under VLGI with pure phase Er 2 FeSbO 7 , BiTiSbO 6 , N-doping TiO 2 (N-TO) or EBH as a photocatalyst was detected. In this investigation, our aim was to prepare new-type heterogenous junction (HJ) catalysts which could remove ENR within pharmaceutic waste water under VLGI. Since the energy gap (ENG) of Er 2 FeSbO 7 was 1.88 eV, which was less than the energy of incoming VL, it was likely to segregate the photo-induced electrons and photo-induced holes which were efficiently disjunct and difficult to be recombined. Therefore, Er 2 FeSbO 7 was a VL responsive-type catalyst with high light-catalyzed liveness. Meanwhile, the conduction band (CB) electric potential of Er 2 FeSbO 7 was −0.71 eV, which was more subtractive than −0.33 eV. Thus, the photoinduced electrons on the CB of Er 2 FeSbO 7 were more easily to interreact with dissolved oxygen (DO) in water for producing superoxide anion. Superoxide anion had a powerful oxidated impact and could oxidate ENR firsthand. The valence band (VB) electric potential of BiTiSbO 6 was 2.995 eV, which was more nonnegative than (2.38) eV. Therefore, holes in the VB of BiTiSbO 6 could oxidate H 2 O or OHinto HRS for degrading ENR in water. HRS had a powerful oxidated impact, whereupon the contamination of ENR in water could be immediately and efficiently oxidated by HRS. At the same time, the ENG of BiTiSbO 6 was 2.32 eV. Thus, BiTiSbO 6 was also a VL responsive-type catalyst with high light-catalyzed liveness. Both Er 2 FeSbO 7 and BiTiSbO 6 could produce photo-induced electrons and photo-induced cavities, which were difficult to be recombined under VLGI, and owned powerful light-catalyzed liveness. Thus, Er 2 FeSbO 7 and BiTiSbO 6 could structure an ideal HJ in conjunction.
The innovation of our work was that a new type of Er 2 FeSbO 7 catalyst and Er 2 FeSbO 7 / BiTiSbO 6 heterojunction photocatalyst (EBHP) were synthesized by solid phase fritting using the elevated temperature synthesis method for the first time. Further, we showed that Er 2 FeSbO 7 and EBHP were VL responsive photocatalysts with higher photocatalytic activity for removing ENR effectively. The removal of organic pollutants within pharmaceutic wastewater by using EBHP was more efficient and safer.

X-ray Diffractometer (XRD) Analysis
The architecture of the Er 2 FeSbO 7 results, which were examined by XRD, are shown in Figure 1. Materials Studio (MS) software was applied to acquire the quantitative data, which took Rietveld analysis quomodo as the basis. The results show that Er 2 FeSbO 7 was single phase and the lattice parameters of the new-type photocatalyst Er 2 FeSbO 7 were 10.179902 Å. At the same time, the ultimate refinement for Er 2 FeSbO 7 showed that there was good consistency between the observed strength and the surveyed strength of the pyrochlore structure, which was a cubelike crystal system with a interspace group of Fd3m (O atoms were included in the mold). Moreover, the total XRD data of Er 2 FeSbO 7 could be successfully indexed to space group Fd3m, according to the lattice constant. Table 1 shows the atomistic co-ordinates and architecture parameters of Er 2 FeSbO 7 . Figure S1 shows the atomistic architecture of Er 2 FeSbO 7 . The conclusion can be drawn from Figure 1 that Er 2 FeSbO 7 crystallized, and the pyrochlore-type architecture was achieved. Based on the results of refinement, the R P = 15.33% with space group Fd3m was achieved.  Be aware that the x co-ordinate of the O (1) atom could be regarded to be an index mark of the variance of the transistorization architecture on the pyrochlore-type A 2 B 2 O 7 compounds (cubic, space group Fd3m), and was equivalent to 0.375 when the six A-O (1) bond distances were equal to that of the two A-O (2) bond distances [52]. Therefore, information on the deformation of the MO 6 (M = Fe 3+ and Sb 5+ ) octahedra could be obtained from the x value [50]. The x value was shifted off x = 0.375 [50], meaning that the deformation of the MO 6 (M = Fe 3+ and Sb 5+ ) octahedra presented visibly in the crystalloid architecture of Er 2 FeSbO 7 . Electric charge disassociation was demanded for the photocatalytic degradation (PCD) of ENR under VLGI with the view of averting the regrouping of the photo-induced sigma electrons and photo-induced cavities. Inoue [53] and Kudo [54] showed that the partial deformation of the MO 6 octahedra, which was gained from several photocatalyzers-for instance, BaTi 4 O 9 and Sr 2 M 2 0 7 (M = Nb 5+ and Ta 5+ )-was highly momentous for averting the electric charge regrouping, and contributed to increasing the photocatalytic activity. Therefore, the distortion of the MO 6 (M = Fe 3+ and Sb 5+ ) octahedra in the crystal structure of Er 2 FeSbO 7 could also be considered effectual for enhancing the photocatalytic activity. Er 2 FeSbO 7 included a tri-dimensional (3D) meshwork architecture of corner-sharing MO 6 (M = Fe 3+ and Sb 5+ ) octahedra. The MO 6 (M = Fe 3+ and Sb 5+ ) octahedra were annexed into catena by an Er 3+ [52]. It revealed that the included angles were between the public angle MO 6 (M = Fe 3+ and Sb 5+ ) octahedra. For instance, the M-O-M BDA for Er 2 FeSbO 7 were significant for affecting the light-catalyzed liveness of Er 2 FeSbO 7 . If the M-O-M BDA was close to 180 degrees, the mobility of the photo-induced electrons and photo-induced cavities would be great [52]. The mobilities of the photoinduced electrons and photoinduced holes affected the probability of electrons and cavities for reaching reaction sites on the catalyzer skin layer, which ultimately led to the influence on the photocatalytic activity [52].
In addition, the Sb-O-Sb BDA of Er 2 FeSbO 7 was larger, thus, the photocatalytic activity of catalyst Er 2 FeSbO 7 was improved. As for Er 2 FeSbO 7 , Er is a 4p-block rare-earth (RE) metallic element, Fe is a 4f-block metallic element and Sb is a 5p-block metallic element. Based on the above analysis, the impact of retrograding ENR under VLGI with Er 2 FeSbO 7 as a photocatalyst could be credited chiefly to its crystal structure and electronic structure. Figure S2 shows the X-ray diffraction pattern of BiTiSbO 6 . We marked the singlehanded diffractive apices in Figure S2. The architecture of BiTiSbO 6 was trialed by the X-ray diffraction technique. MS software was used for obtaining the achieved quantitative data based on a Rietveld analysis quomodo. The verdict could be made that BiTiSbO 6 was single-phase and the lattice parameters of BiTiSbO 6 were 9.87708 Å. In the light of the refinement result, we proved that BiTiSbO 6 crystallized with a tetragonal spinel structure and space group I41/A. The BDG of BiTiSbO 6 was calculated to be 2.32 eV. Figure 2 shows the XRD diffraction pattern of Er 2 FeSbO 7 /BiTiSbO 6 HJ photocatalyzer before the reaction. From Figure 2, it can be seen that the pure single crystal photocatalyzer Er 2 FeSbO 7 and single-phase photocatalyzer BiTiSbO 6 existed. Every diffraction maximum of Er 2 FeSbO 7 and every diffraction maximum of BiTiSbO 6 were resoundingly labeled, and impurity was otherwise not discovered in Figure 2. Figure 2 also shows that the XRD image of the HJ Er 2 FeSbO 7 /BiTiSbO 6 was a single-phase with perfect crystallinity, and the defects were not found inside the crystal. Thus, the photoinduced electrons and photoinduced holes which were generated during the reaction would not aggregate and compound at the defects inside the crystal catalyst, therefore, prolonging the service life of photo-induced sigma electrons and photo-induced cavities and improving the catalytic activity of the photocatalyst. Figure 2. The X-ray diffraction spectrum of Er 2 FeSbO 7 /BiTiSbO 6 HJ before reaction. Figure S3 shows the XRD diffraction pattern of Er 2 FeSbO 7 /BiTiSbO 6 HJ photocatalyst after the reaction. A comparison of Figures 2 and S3 found that the XRD images of the HJ Er 2 FeSbO 7 /BiTiSbO 6 before the reaction and after the reaction were consistent and unchanged. As a result, the HJ Er 2 FeSbO 7 /BiTiSbO 6 had good stability.

UV-Vis Diffuse Reflectance Spectra
The absorption spectra of the Er 2 FeSbO 7 sample are listed in Figure 3a,b. The absorption edge (ABE) of this new photocatalyst Er 2 FeSbO 7 was discovered to be at 458 nm, which was within the visual range of the optical spectrum. The band gap of energy (B-GE) of the crystal semi-conductors could be certain by the point of crossing between light quantum energy capacity hν axes and the line calculated from the linear section of the ABE, also known as the Kubelka-Munk function (1) (called re-emission function) [55,56].
where S was the dispersion facient, R d was the spread reflection and α was the delegated absorbability modulus of radialization. The optic absorbability near the energy band edge of the crystalline semiconductor conforms to Equation (2) [57,58]: Here, A, α, Eg and ν represent the ratable numeric constant, absorbability modulus, BDG and light frequency, separately. In this equation, n determines the properties of transitions in semiconductors. Eg and n could be reckoned by the lower row process: (1) draw ln (αhν) versus ln (hν − Eg), supposing a hemiidentic number of Eg; (2) infer the number of n tasks based on the rate of the grade in the Figure; (3) by drawing (αhν) 1/n and hν to refine the number of Eg and extrapolating the graph to (αhν) 1/n = 0. According to the above method, the Eg value of Er 2 FeSbO 7 was calculated to be 1.88 eV. The estimated value of n was about 2 and obliquely licensed the optical transition of Er 2 FeSbO 7 .
The B-GE of Er 2 FeSbO 7 was 1.88 eV, the B-GE of Bi 3 O 5 I 2 was 2.02 eV [55], and the B-GE of co-doping ZnO was 2.39 eV. All the B-GE of these three chemical compounds (CMPDs) were less than 2.78 eV, which implied that above three catalyzers had VL-responsive properties and held distorted latent energy for displaying high light-catalyzed liveness under VLGI. Figure S4a,b show the UV-Vis diffuse reflectance spectrum of BiTiSbO 6 . In light of the above sequences and Figure S4a,b, the number of Eg for BiTiSbO 6 was calculated as 2.32 eV. The cursory value of n was about 2, which indicated discretely indirect optical lambda transition of BiTiSbO 6 .
The absorption spectra of the EBH specimen are ranked in Figure 4a,b. The ABE of this new photocatalyzer EBH was discovered to be at 620 nm, which was in the VL range of the sunlight spectra. The ABE of the photocatalyst Er 2 FeSbO 7 was at 640 nm and that of the photocatalyst BiTiSbO 6 was at 534 nm, which was located in the VL area of the sunlight optical spectrum. The B-GE of Er 2 FeSbO 7 was 1.88 eV, the B-GE of BiTiSbO 6 was 2.32 eV and the B-GE of EBH was 1.99 eV. Every B-GE of the above three CMPDs was less than 2.32 eV, which implies that the above three catalyzers had VL-range characteristics and possessed gigantic latent force for displaying high light-catalyzed liveness under VLI.

Property Characterization of Er 2 FeSbO 7 /BiTiSbO 6 Heterojunction Photocatalyst
For the purpose of obtaining the skin layer chemical constitution and the quantivalency ungerade states of each constituent of EBHP, the X-ray photoelectron spectrum (XPS) was executed. Figure 5 shows the XPS investigation spectra of EBHP. Figure 6 shows the XPS spectrogram of O 2− , Er 3+ , Fe 3+ , Bi 3+ , Ti 4+ and Sb 5+ which derived from EBHP. In light of the XPS full spectrogram, which is shown in Figure 5, the compound EBHP subsumed the elementary substance of Er, Fe, Bi, Ti, Sb and O. In light of the XPS analysis effects, which are exhibited in Figures 5 and 6, the oxidation state of Er, Fe, Bi, Ti, Sb or O ion was +3, +3, +3, +4, +5, or −2, separately. Due to the above analysis effects, it could be summarized that the chemical formula of the new CMPD was Er 2 FeSbO 7 /BiTiSbO 6 . It can be seen from Figure 6 that multifarious elemental apices with unique binding energies were attained. In Figure 6, the O1s peak of O elementary substance is situated at 530.54 eV; the Er 4d and Er 4d5/2 peak of Er elementary substance is situated at 164. 47   The reason for the high oxygen value is that there was a lot of adsorbed oxygen on the skin layer of EBHP. Obviously, neither shoulders nor widening in the XPS apices of EBHP was observed, meaning that the as-prepared compound was pure phase. Figure 7 shows the XPS survey spectrum of the EBHP after the reaction. Figure 8 shows the XPS spectra of Er 3+ , Bi 3+ , O 2− , Sb 5+ , Fe 3+ and Ti 4+ , which derived from the EBHP after the reaction. Figure 9 shows a SEM picture of EBHP. Figure 10 shows the EDS elemental plotting of EBHP (Er, Fe, Sb, O from Er 2 FeSbO 7 and Bi, Ti, Sb, O from BiTiSbO 6 ). Figure 11 shows the EDS spectra of EBHP. It can be found from Figures 9 and 10 that the larger tetragonal particulates remained with BiTiSbO 6 and the smaller circular shaped particles belonged to Er 2 FeSbO 7 . From Figures 9 and 10, the particulates of BiTiSbO 6 were encircled by the smaller particulates of Er 2 FeSbO 7 , and these two particulates were snugly associated, which manifested the triumphant fusion of EBHP. BiTiSbO 6 possessed rhombic dodecahedron-like morphology. As we all know, different surface energy of crystal sections gain command of the anatomical upgrowth of a photocatalyst. The professors discovered the order of the skin layer energy of dissimilar sides of silver phosphate (SP) (111) < SP (100) < SP (110). In mean solar time, the SP (110) facet showed higher skin layer energy than the (111) facet, which was the reason why it haphazardly took the form of SP architectures along (110) crystallographic orientation, leading to the formulation of a diamond dodecahedrons-like morphology for SP [59]. Thus, BiTiSbO 6 had diamond dodecahedron-like morphology that could be interpreted as being above description. The experimentation upshots which are shown in Figure 9 signified that Er 2 FeSbO 7 had an inerratic sphere-like morphology and a well-proportioned particulates repartition. The particulate size of Er 2 FeSbO 7 was surveyed to be around 250 nm, while the larger particulate dimension of BiTiSbO 6 was surveyed to be about 1800 nm. Figure 6. (a) XPS spectra of O 2− which derives from the EBHP before the reaction; (b) XPS spectra of Er 3+ which derived from the EBHP before the reaction; (c) XPS spectra of Fe 3+ which derived from the EBHP before the reaction; (d) XPS spectra of Sb 5+ (Sb 3d) which derived from the EBHP before the reaction; (e) XPS spectra of Bi 3+ which derived from the EBHP before the reaction; (f) XPS spectra of Ti 4+ which derived from the EBHP before the reaction; (g) XPS spectra of Sb 5+ (Sb 4d) which derived from the EBHP before the reaction.     was obtained as 41:100. Based on the above effects, we suggest that the EBHP were of a high purity quotient following our experimental environment and prerequisites. Figure 12 shows the consistence change curve of ENR during the PCD of ENR with EBH, Er 2 FeSbO 7 , BiTiSbO 6 , or N-TO as a photocatalyzer under VLGI. From Figure 12, we first adsorbed enrofloxacin in the dark with EBH, Er 2 FeSbO 7 , BiTiSbO 6 , or N-TO as a photocatalyst. It can be seen from Figure 12 that after 45 min of adsorption under dark conditions with EBH as a catalyst, the adsorption rate of ENR in drug wastewater was 3.88%. All the other experiments followed the same dark conditions for 45 min. When Er 2 FeSbO 7 was used as a photocatalyst, the adsorption rate of ENR was 3.6%. Using BiTiSbO 6 as a photocatalyst, the adsorption rate of ENR in pharmaceutical wastewater was 3.2%. Using N-TO as a photocatalyst, the adsorption rate of ENR was 12.12%. As can be seen from Figure 12, when EBH, Er 2 FeSbO 7 , BiTiSbO 6 , or N-TO was used as a photocatalyst to degrade ENR, the concentration of ENR in drug waste water continually decreased with the increase in VLGI time. The effects that are attained from Figure 12 indicate that the RR of ENR within pharmaceutical waste water reached 99.16%. The velocity of the reaction was 2.75 × 10 −9 mol/L/s and the photonic efficiency (PE) was 0.0578% with EBHP as a catalyzer after a VLGI of 150 min (VLGI-150min). All the other experiments followed the same VLGI-150min. When Er 2 FeSbO 7 was used as a photocatalyzer, the RR of ENR reached 86.12%, the rate of the reaction was 2.39 × 10 −9 mol/L/s and the PE was 0.0502%. The RR of ENR within pharmaceutical waste water reached 78.12%, the rate of reaction was 2.17 × 10 −9 mol/L/s and the PE was 0.0456% with BiTiSbO 6 as a photocatalyzer. The RR of ENR reached 39.28%, the velocity of the reaction was 1.09 × 10 −9 mol/L/s and the PE was 0.0229% with N-TO as a photocatalyzer. The effects that are attained from Figure 12 indicate that the RR of ENR within pharmaceutical waste water reached 77.56%, and the velocity of the reaction was 3.23 × 10 −9 mol/L/s with EBH as a photocatalyzer after VLGI of 100 min. Moreover, the results that are attained from Figure 12 represent that the RR of ENR within pharmaceutical waste water reached 99.16%, and the velocity of the reaction was 2.75 × 10 −9 mol/L/s with EBH as a catalyst after VLGI-150min. It can be seen from the above results that the concentration of ENR decreased and the removal rate of ENR increased with prolonging the VLGI time during the PCD of ENR with EBH as the photocatalyst.

Photocatalytic Activity
Moreover, we can summarize from the above effects that the photodegradation efficiency (PGE) of ENR in the presence of EBHP was the highest. Simultaneously, the PGE of ENR with Er 2 FeSbO 7 as a photocatalyzer was higher than that with BiTiSbO 6 as a photocatalyzer or with N-TO as a photocatalyzer. Ultimately, the PGE of ENR with BiTiSbO 6 as a photocatalyzer was higher than that with N-TO as a photocatalyzer, manifesting that the VL light-catalyzed liveness of EBHP was maximally contrasted with that of Er 2 FeSbO 7 , BiTiSbO 6 or N-TO. The above effects show that the RR of ENR with EBHP was 1.151 times, 1.269 times or 2.524 times higher than that with Er 2 FeSbO 7 , BiTiSbO 6 , or N-TO as a photocatalyzer after VLGI-150min. In the previous reports, Chen et al. found that graphene-like white graphite modified bismuth phosphate material could be used to promote the photocatalytic degradation rate of ENR. The experimental results show that the removal rate of ENR was 79.2%, 91.5% or 83.8% by 0.5 wt% BN/BiPO 4 , 1 wt% BN/BiPO 4 or 2 wt% BN/BiPO 4 under sunlight irradiation of 120 min. It could be found that 1wt% graphene-like BN/BiPO 4 had the highest degradation rate of ENR [51]. However, our experimental results show that the removal rate of ENR was 95.75% with EBH as a photocatalyst under a VLGI of 120 min. Therefore, the EBHP which was studied in this paper possessed higher degradation efficiency during the degradation process of ENR compared with 1 wt% graphene-like BN/BiPO 4 .
Using Er 2 FeSbO 7 or EBH as a photocatalyzer separately, the consistence of null valence Sb or Sb 5+ in the water before the PCD of ENR was zero. After VLGI-150 min for the PD of ENR, the contents of null valence Sb or Sb 5+ in the water contents was also zero. Regarding the specific surface, the surface area of Er 2 FeSbO 7 was 4.21 m 2 /g, the surface area of BiTiSbO 6 was 3.94 m 2 /g, and the surface area of EBHP was 4.52 m 2 /g. In addition, the average pore diameter of Er 2 FeSbO 7 , BiTiSbO 6 or EBHP was calculated to be 13.69 nm, 14.12 nm or 13.14 nm. From Figure 12, it can be seen that the catalytic coefficient of performance of Er 2 FeSbO 7 /BiTiSbO 6 , Er 2 FeSbO 7 and BiTiSbO 6 after a VLGI of 100 min was 77.56%, 50.20% and 42.52%, respectively. Therefore, it could be concluded that the photocatalyst which possessed a larger specific surface area (SSA) had higher photocatalytic activity. Our PLR were expounded within several Figures by the photocatalyzer, which was arranged by the hydrothermal synthesis quomodo (HSQ), thus, it would not conduce disparity in light-catalyzed liveness. XRD mensuration was carried out to indagate the chemical composition (CC) and phase architecture of the preliminary materials. All sharp diffraction peaks of the materials which were prepared by HSQ showed that the samples were well crystallized, and an impurity peak was not observed, which indicated that the purity of the product was very high.
The size of the photocatalyst particle morphology would affect the SSA of the photocatalyst. The larger SSA of the catalyst would result in more active sites on the catalyst surface, thus, the photocatalytic activity would be stronger. When EBH was used as a catalyst, the RR of ENR could reach 99.16% after VLGI-150min. This research work was carried out to develop VL-responsive nano materials with low cost, high catalytic activity and full use of 43% VL in the solar spectrum. Ultimately, the toxic organic pollutants from pharmaceutical wastewater should be deeply purified and removed, and a safe, hygienic and pollution-free water environment should be obtained. Figure 13 shows the consistence change curves of total organic carbon (TOC) during the PCD of ENR within pharmaceutical waste water with EBH, Er 2 FeSbO 7 , BiTiSbO 6 , or N-TO as a photocatalyzer under VLGI. The consistence of ENR continually decreased with increasing VLGI time. As can be seen in Figure 13, the RR of TOC within pharmaceutical waste water attained 94.96%, 83.94%, 74.45% or 32.83%, respectively, after VLGI-150min when EBHP, Er 2 FeSbO 7 , BiTiSbO 6 or N-TO was used for degrading ENR. In summary, we can conclude from the above effects that the RR of TOC during the degradation of ENR in EBHP was higher than that of Er 2 FeSbO 7 , BiTiSbO 6 or N-TO. The above results also indicate that the RR of TOC during the degradation of ENR in Er 2 FeSbO 7 was much higher than that in BiTiSbO 6 or N-TO, which signifies that EBHP had the maximum mineralization rate during ENR degradation compared with Er 2 FeSbO 7 , BiTiSbO 6 or N-TO.  Figure 14 shows the concentration change curve of ENR during PCD with EBH as a photocatalyzer under VLGI for TCDT. It can be seen from Figure 14 that the RR of ENR reached 98%, 96.76% or 95.6%, respectively, after VLGI-150min with EBH as a photocatalyst, by accomplishing recursion retrogradation tests for degrading ENR. Figure 15 shows the consistency change crooked line of TOC during the PCD of ENR with EBH as a photocatalyzer under VLGI for TCDT. From Figure 15, it can be seen that the RR of TOC was 93.31%, 91.87% or 90.38%, respectively, after VLGI-150min with EBHP, by accomplishing TCDT for degrading ENR. The experimental effects which were attained from Figures 14 and 15 show that the EBHP had a high anti-whip ability.   Figure 16 shows the first-order kinetics plots of PCD for ENR under VLGI with EBH, Er 2 FeSbO 7 , BiTiSbO 6 or N-TO as a photocatalyzer. As can be seen from Figure 16, the power mechanics invariable k which was obtained from the dynamical crooked line toward ENR consistence and VLGI time with EBH, Er 2 FeSbO 7 , BiTiSbO 6 or N-TO as a photocatalyzer reached 0.02296 min −1 , 0.00882 min −1 , 0.00699 min −1 or 0.00254 min −1 , separately. The dynamic numeric constant k, which hails from the namic profile toward TOC consistence, reached 0.01535 min −1 , 0.00786 min −1 , 0.00607 min −1 or 0.00192 min −1 with EBH, Er 2 FeSbO 7 , BiTiSbO 6 or N-TO as a photocatalyzer. The fact that the value of K TOC for deteriorating ENR was lower than the number of K C for deteriorating ENR by using the same catalyzer indicated that the photodegradation intermediates of ENR might appear in the PCD of ENR under VLGI. Meanwhile, contrasted with the other photocatalyzer, EBHP had a better mineralization workpiece ratio for ENR retrogradation. Figure S5 shows the single-order kinetics of the PCD of ENR under VLGI with EBH as a photocatalyst for three cycle degradation tests (TCDT). In the light of the results from Figure S5, the power mechanics invariable k which came from the dynamical crooked line toward ENR consistence and VLGI time with EBHP for TCDT reached 0.01786 min −1 , 0.01536 min −1 or 0.01354 min −1 . The power mechanics invariable k which derived from the dynamical crooked line toward TOC consistence and VLGI time with EBHP for TCDT reached 0.01352 min −1 , 0.01206 min −1 or 0.01079 min −1 . In the light of the experimental effects from Figures 16 and S5, the PCD of ENR within pharmaceutical waste water with EBHP under VLGI conformed to the single-order reaction dynamics.
It can be seen from Figure S5 that the RR of ENR lessened by 2.4% with EBH as a photocatalyzer under VLGI after TCDT, and the RR of TOC lessened by 2.93%. It did not show a significant difference for the retrogradation efficiency above TCDT, and the photocatalyzer architecture of EBHP was jarless. Figure 17 presents the impact of diverse free radical scavengers (FRSs), for instance, benzoquinone (BQ),isopropanol (IPA) or ethylenediamine tetraacetic acid (EDTA) on the RR of ENR with EBH as a photocatalyzer under VLGI. At the start of the photocatalytic experiment, different FRSs were added to ENR resolution for determining the active species in the retrogradation process of ENR. We used isopropanol (IPA) to capture HRS ( • OH); benzoquinone (BQ), which we utilized for captured superoxide anions ( • O 2 − ); and ethylenediaminetetraacetic acid (EDTA), which we used for captured holes (h + ). The scheming IPA consistence, BQ consistence or EDTA consistence was 0.15 mmol/L, and the added amount of IPA, BQ or EDTA was 1 mL. As for the selection of free radical scavenger consistence, five concentrations of free radical scavengers (0.05 mmol/L, 0.1 mmol/L, 0.15 mmol/L, 0.2 mmol/L and 0.25 mmol/L, respectively) were used to participate in the reaction. According to the experimental results, the concentration of capture agent was taken as the abscissa, meanwhile, [99.16%-C] was taken as the ordinate. Ultimately, five corresponding curves were obtained. As for [99.16%-C], 99.16% was the removal rate of ENR at 150 min in the blank experiment, and C was the removal rate of ENR after adding the corresponding concentration of capture agent for 150 min. At this time, the curve had a maximum value, which meant that the corresponding radicals were completely captured by the capture agent. At this time, the highest point was determined as the concentration of free radical scavenger which was required in the experiment. In light of Figure 17, while the IPA, BQ or EDTA was added into ENR solution, the RR of ENR decreased by 59.87%, 70.89% or 29.15%, respectively, compared with the RR of ENR which came from the matched troop. Thus, it could be predicated that • OH, h + and • O 2 − were all active free radicals in the process of ENR retrogradation. It could be seen from Figure 17 that • OH in the ENR result played an overriding position when ENR was degraded with EBH as a photocatalyzer under VLGI. Through the medium of the project of the appending trapping agent, it was assumed that superoxide anion possessed maximal oxidation removal ability for removing ENR within pharmaceutical waste water, relative to hydroxyl radicals or holes. The oxygenation elimination capacity for degrading ENR, which was from high to low among three oxygenation radicals, was as follows: superoxide anion > hydroxyl radicals > holes. A Nyquist impedance plot measurement was the other significant qualitative method which showed the photo-induced sigma electrons and photo-induced cavities transfer procedure of the preliminary photocatalyzer at solid/electrolyte limiting surfaces. The smaller the arc radius (ARs) was, the higher the transport workpiece ratio of the photocatalyst was. Figure 18 shows the homologous Nyquist impedance plots of the preliminary EBHP, Er 2 FeSbO 7 or BiTiSbO 6 photocatalyzer. It was easily found from Figure 18 that the size of the ARs was in the order: BiTiSbO 6 > Er 2 FeSbO 7 > EBHP. Above effects indicated that the preliminary EBHP presented a more expeditious disassociation of photo-induced sigma electrons and photo-induced holes, and a faster interfaced charge transfer capability. After VLGI of 100 min, the removal rate of ENR with EBH, Er 2 FeSbO 7 or BiTiSbO 6 as a catalyst was 77.56%, 50.2% or 42.52%, respectively, which was consistent with the intercomparable results of the curvature radius of the catalysts in the electrochemical impedance experiment. After a VLGI of 150 min, the removal rate of ENR by EBH, Er 2 FeSbO 7 or BiTiSbO 6 was 99.16%, 86.12% or 78.12%, respectively, which was also consistent with the intercomparable results of the curvature radius of the catalysts in the electrochemical impedance experiment. Therefore, it could be concluded that the smaller curvature radius of the catalyst led to a higher removal rate of ENR and a higher catalytic activity of the catalyst.  Figure 19 shows the potential PCD machine process of ENR with EBH as a photocatalyst under VLGI. The potentials of VB and CB for a semi-conductor catalyzer could be calculated based on Equations (3) and (4) [59]:

Possible Degradation Mechanism Analysis
where E g is the BDG of semi-conductor, X is the electronegativity of the semi-conductor, and E e is the energy of free electrons on the hydrogen scale (about 4.5 eV). Based on the above equations, the VB potential (VBP) or the CB potential (CBP) for Er 2 FeSbO 7 was imputed to be 1.17 eV or −0.71 eV, separately. For BiTiSbO 6 , the VBP or the CBP was imputed to be 2.995 eV and 0.31 eV, separately. It could be discovered that both Er 2 FeSbO 7 and BiTiSbO 6 could absorb VL and internally generated electrons-holes pairs when the EBHP was search lighting by VL. Due to the redox potential position of the CB of Er 2 FeSbO 7 (−0.71 eV), it was more subtractive than that of BiTiSbO 6 (0.31 eV), and the photo-induced sigma electrons on the CB of Er 2 FeSbO 7 could migrate to the CB of BiTiSbO 6 . the redox potential position of the VB of BiTiSbO 6 (2.995 eV) was more nonnegative than that of Er 2 FeSbO 7 (1.17 eV), and the photo-induced cavities on the VB of BiTiSbO 6 could migrate to the VB of Er 2 FeSbO 7 . Figure 19. Possible PCD mechanism of ENR with EBH as photocatalyzer under VLGI.
Therefore, combining Er 2 FeSbO 7 and BiTiSbO 6 realizes a new heterojunction photocatalyst EBHP that can powerfully lessen the recombination rate of photo-induced sigma electrons and photo-induced cavities, reduce the essential resistance, extend the service life of photo-induced sigma electrons and photo-induced cavities, and improve the interface charge migration efficiency. Therefore, more oxyradicals such as • OH or • O 2 − could be used for increasing the retrogradation workpiece ratio of ENR. Furthermore, the CBP of Er 2 FeSbO 7 was −0.71 eV, which was more subtractive than that of O 2 / • O 2 − (−0.33 V), indicating that the electrons within the CB of Er 2 FeSbO 7 could assimilate O 2 for caused • O 2 − , which could degrade ENR (as shown as path 1 in Figure 19). Simultaneously, the VBP of BiTiSbO 6 was 2.995 eV, which was more positive than that of OH − / • OH (2.38 V), indicating that the holes in the VB of BiTiSbO 6 could oxidize H 2 O or OH − into • OH for degrading ENR, which is shown as path 2. Ultimately, the photo-induced cavities in the VB of Er 2 FeSbO 7 or BiTiSbO 6 could directly oxidize ENR and degrade ENR due to the high oxidizing ability, and this is shown as path 3. In conclusion, EBHP had good photocatalytic activity for the degradation of ENR, which was mainly due to the efficient electron and hole separation efficiency induced by EBHP.
With the view of studying the retrogradation mechanism of ENR, the in-between products yield which were caused during the retrogradation process of ENR were checked by LC-MS. The midterm offspring which were attained during the PCD of ENR were identified as C 17  According to the above checked in-between products yield, the PCD paths of ENR were proposed. Figure 20 shows the PCD pathway scheme recommended by ENR under VLGI with EBH as a photocatalyzer. It can be discovered from Figure 20 that an oxidizing reaction and hydroxylation reaction were sensed during the PCD process of ENR. Finally, ENR was transformed into a micromolecule organic compound, and combined with oxidative active free radicals to transform into CO 2 and H 2 O.

Preparation Method of Er 2 FeSbO 7
A novel photocatalyzer, Er 2 FeSbO 7 , was compounded by the high temperature solidstate sintering method. Er 2 O 3 , Fe 2 O 3 and Sb 2 O 5 (Sinopharm Group Chemical Reagent Co., Ltd., Shanghai, China) with a purity of 99.99% were used as rough stock without further purification. All powders (n(Er 2 O 3 ):n(Fe 2 O 3 ):n(Sb 2 O 5 ) = 2:1:1) were synthesized after drying at 200 • C for 4 h. In order to prepare Er 2 FeSbO 7 , the precursor was stoichiometric blended, then pushed into a small column and fitted into a compalox crucible (Shenyang Crucible Co., LTD, Shenyang, China). After scorifying at 400 • C for 2 h, the primary materials and the small pillars were taken out of the galvanical stove. The compounded materials were ground and then fitted into the galvanical furnace (KSL 1700X, Hefei Kejing Materials Technology CO., LTD, Hefei, China). Finally, calcination was carried out in an electric smelter at 1100 • C for 36 h.
A total of 0.30 mol/L of Er(NO 3 ) 3 ·5H 2 O, 0.15 mol/L of Fe(NO 3 ) 3 and 0.15 mol/L of SbCl 5 were compounded and stirred for 20 h. This was transferred to an autoclave lined with polyfluortetraethylene and heated at 200 • C for 15 h. Subsequently, the attained powder was scorified at 800 • C for 10 h in a tube furnace at a velocity of 8 • C/min under an atmosphere of N 2 . Finally, Er 2 FeSbO 7 powder was also attained by HSQ.

Preparation Method of BiTiSbO 6
A novel photocatalyzer, BiTiSbO 6 , was compounded by the high temperature solidstate sintering method. Bi 2 O 3 , TiO 2 and Sb 2 O 5 (Sinopharm Group Chemical Reagent Co., Ltd., Shanghai, China) with a purity of 99.99% were used as rough stock without further purification. All powders (n(Bi 2 O 3 ):n( TiO 2 ):n(Sb 2 O 5 ) = 1:2:1) were synthesized after drying at 200 • C for 2 h. In order to prepare BiTiSbO 6 , the precursor was stoichiometric blended, then pushed into a small column and fitted into an compalox crucible (Shenyang Crucible Co., Ltd., Shenyang, China). After scorifying at 400 • C for 4 h, the raw material and small cylinder were removed from the galvanical furnace. The compounded materials were ground and then fitted into the galvanical furnace (KSL 1700X, Hefei Kejing Materials Technology CO., Ltd., Hefei, China). Finally, calcination was carried out in an electric smelter at 1010 • C for 25 h.
A total of 0.15 mol/L of Bi(NO 3 ) 3 ·5H 2 O, 0.15 mol/L of TiCl 4 , and 0.15 mol/L of SbCl 5 was compounded and stirred for 20 h. The solution was transferred into a Teflon-lined autoclave and scorified at 200 • C for 15 h. Afterwards, the achieved powder was scorified at 780 • C for 10 h in a tube-type furnace at a velocity of 8 • C/min under an air envelope of N 2 . BiTiSbO 6 powder was finally achieved.

Synthesis of N-Doping TiO 2
Nitrogen-doped titania (N-doping TiO 2 ) catalyst was prepared by sol-gel modus with tetrabutyl titanate as a precursor and ethanol as a solvent. The procedure was as follows: first, 17 mL of tetrabutyl titanate and 40 mL of absolute ethyl alcohol were combined to serve as solution A; 40 mL of absolute ethyl alcohol, 10 mL of glacial acetic acid and 5 mL of double distilled water were blended to be solution B; subsequently, solution A was added dropwise into solution under vigorous magnetic stirring condition, and then a transparent colloidal suspension was shaped. Then, ammonial solution with an N/Ti ratio of 8 mol% was fitted into the transparent tremellose soliquid under magnetical mix round prerequisite for 1 h. Dry gel was then generated after 2 days of aging. The xerogels were ground into powder, dead-burned at 500 degrees for 2 h, then ground on the powder and screened by a vibrating screen to obtain N-TO powder.

Synthesis of Er 2 FeSbO 7 /BiTiSbO 6 Heterojunction Photocatalyst
The maximal calcination temperature (MCT) of Er 2 FeSbO 7 which was prepared by solid state sintering modus was 1100 • C, and the insulated time was 36 h. The MCT of BiTiSbO 6 which was preliminary by solid state sintering modus was 1010 • C, and the insulated time was 25 h. The MCT of BiTiSbO 6 which was preliminary by HSQ was 780 • C, and the insulated time was 10 h. The MCT of Er 2 FeSbO 7 which was preliminary by HSQ was 800 • C, and the insulated time was 10 h. The higher the MCT, the greater the power dissipation energy, which would decrease and expend the operational life proof cycle of the stove. The longer insulated time and the higher maximal sintering temperature would cause the larger particle size of BiTiSbO 6 or Er 2 FeSbO 7 . That being so, the SSA of BiTiSbO 6 or Er 2 FeSbO 7 would be reduced and the light-catalyzed liveness of BiTiSbO 6 or Er 2 FeSbO 7 would be accordingly lessened. For the purpose of increasing the light-catalyzed liveness, reducing energy consumption and increasing the instrument operational life proof cycle of high-temperature incinerator, we used HSQ for preparing BiTiSbO 6 and Er 2 FeSbO 7 in the process of preparing HJ. Er 2 FeSbO 7 and BiTiSbO 6 were prepared by HSQ, which mainly used the dissolution recrystallization mechanism to dissolve Er (NO 3 ) 3 ·5H 2 O; Fe(NO 3 ) 3 ; SbCl 5 ; Bi(NO 3 ) 3 ·5H 2 O; and TiCl 4 in hydrothermal medium, and then the above materials entered the solution in the form of ion groups and molecular groups. Strong convection, which was caused by a temperature difference in autoclave, would prompt these ions and molecules to transport to the growth area, which contained seed crystal. Ultimately, the saturated solution was formed and crystallized. The prepared Er 2 FeSbO 7 and BiTiSbO 6 powders were prepared by the solvothermal method. The powders of Er 2 FeSbO 7 or BiTiSbO 6 were prepared by dissolving in octanol organic solvent in an autoclave. At this time, under liquid phase and supercritical conditions, the reactants were dispersed in the solution and became more active. Therefore, the reactants were dissolved and dispersed, meanwhile, the reaction occurred, and the product was synthesized slowly.
A handy solvothermal modus was used for synthesizing new EBHP in this report. EBHP was preliminary by compounding 525.36 mg of Er 2 FeSbO 7 with 30 wt% (974.64 mg) of BiTiSbO 6 in 300 mL of octanol (C 8 H 18 O) and then dispersed in an ultrasonic bath for 1 h. Then, it was tepefied and channeled back at 140 • C for 2 h under acute whisking condition to increase the adhesion of Er 2 FeSbO 7 on the surface of BiTiSbO 6 nanoparticles to form EBHP. After cooling to indoor temperature, the product was attained by the centrifuge method and rinsed several times with a mixture of n-hexane/ethanol. The purified powder was dried in a vacuum oven at 60 • C for 6 h and stored in a dryer for further use. Finally, EBHP was successfully prepared.

Characterizations
The pure crystals of the preliminary patterns were checked by the powder X-ray diffractometer (XRD, Shimadzu, XRD-6000, Cu Kα radiation, λ = 1.54184 Å, sampling pitch of 0.02 • , preset time of 0.3 s step −1 , Kyoto, Japan). The morphology and microstructure of the preliminary patterns were characterized by using a scanning electronic microscope (SEM, FEI, Quanta 250), and the elementary composition which derived from above prepared samples was captured by energy dispersive spectroscopy (EDS). The diffuse reflectance spectrum of the above prepared sample was obtained by using an UV-Vis spectrophotometer (UV-Vis DRS, Shimadzu, UV-3600, Kyoto, Japan). Surface CC and states of the prepared sample were analyzed by X-ray photoelectron spectrograph (XPS, UlVAC-PHI, PHI 5000 VersaProbe) with an Al-kα X-ray source.

Photoelectrochemical Experiments
Electrochemical impedance spectroscopy (EIS) project was implemented by a CHI660D electrochemical station (Chenhua Instruments Co., Shanghai, China) with a normative 3-electrode. The 3-electrode comprised a working electrode (as-prepared catalysts); counter electrode (platinum plate); and reference electrode (Ag/AgCl electrode). The electrolyte was Na 2 SO 4 aqueous solution (0.5 mol/L). The lamphouse for the project was a 500 W Xe lamp with an UV cut-off filter. We placed the pattern (0.03 g) and chitosan (0.01 g) in dimethyl formamide (0.45 mL) and then used ultrasonication for an hour to obtain the well-proportioned suspending liquid. Afterward, they were trickled on indium tin oxide (ITO) conducting glass (10 mm × 20 mm). Finally, the working electrode was dried at 80 • C for 10 min.

Experimental Setup and Procedure
The projects were implemented in a photocatalytic catalyst case (XPA-7, Xujiang Electromechanical Plant, Nanjing, China) and the temperature of the response setup was 20 • C, which was mastered by rotative cooling water. Imitated solar light search-lighting was offered by a 500 W xenon with a 420 nm cut-off filter. There were 12 of the same silica tube, among which the bulk of a single response solution was 40 mL, and the total response bulk for pharmaceutic waste water was 480 mL. The doses of Er 2 FeSbO 7 , BiTiSbO 7 or EBHP were 0.75 g/L. Furthermore, the consistence of ENR was 0.025 mmol/L. The consistence of ENR was the remaining consistence after biologic degradation for practical pharmaceutical waste water which contained an ENR of 1.0 mmol/L (Taihu Lake, Wuxi, China). During the response, 3 mL suspension was withdrawn periodically, whereafter the percolation (0.22 µm PES polyethersulfone filter membrane) was realized for removing the activator. Finally, the remaining consistency of ENR in the solution was determined by the UV-visible spectrophotometer (UV-2550, Shimadzu Corporation, Kyoto, Japan). The absorption wavelength (detecting wavelength) of ENR was 276 nm. The absorbancy normative curvilineal of ENR at dissimilar consistency was accomplished under UV-irradiation search-lighting in the area of 220-320 nm with a visible UV-irradiation wedge photometer. The relation between the consistency of ENR and the absorbancy number at 276 nm should be counted. The absorbancy of ENR in the solution was surveyed at the absorptive wavelength of 276 nm, and the specification curve of ENR was drawn and a linear regression quomodo was used for the quantification of ENR. Prior to VLGI, the suspending, which contained the photocatalyzer and ENR, was magnetically mixed in darkness for 45 min to ensure the establishment of an adsorption/desorption equilibrium among the photocatalyzer, ENR and atmospheric oxygen. During VL illumination, the suspending liquid was stirred at 500 rpm.
The mineralization project data of ENR within the response solution were surveyed by using a TOC analyzer (TOC-5000 A, shimadzu Corporation, Kyoto, Japan). For the purpose of examining the consistency of TOC during the PCD of ENR, potassium acid phthalate (KHC 8 H 4 O 4 ) or natrium carbonicum calcinatum was used as normative reactant. Normative solutions of KHC 8 H 4 O 4 with a known carbon consistency (in the range of 0-100 mg/L) were preliminary for calibration. Six samples which contained 45 mL of response solution were used for measuring TOC consistence every time.
The discrimination and survey of ENR and its in-between retrogradation offspring were carried out by liquid chromatography-mass spectrometry (LC-MS, Thermo Quest LCQ Duo, Thermo Fisher Scientific Corporation, Bay State Waltham, MA, USA. Beta Basic-C18 HPLC column: 150 × 2.1 mm, ID of 5 µm, Thermo Fisher Scientific Corporation, MA, USA). Here, 20 µL of solution attained after the photocatalytic response was emptied voluntarily into the LC-MS system. The mobile phase contained 60% methyl alcohol and 40% UPW, and the current velocity was 0.2 mL/min. MS conditions contained an electrojet electric dissociation limiting surface, a capillary temperature of 27 • C with an electric tension of 19.00 V, a spray electric tension of 5000 V and an invariable sheath gas flow velocity. The action spectrum was attained in the anion scan mode and the m/z sweep interval from 50 to 600.
For the purpose of survey, the photon intensity of incoming ray, the strainer which was 7 cm in length and 5 cm in breadth was chosen to be irradiated by incident single pitch of waves VL of 420 nm. In light of the equation of υ = c/λ and hv which deputized the photon energy, Avogadro's number N A , Planck constant h, photonic frequency υ, incoming ray pitch of waves λ and speed of light c were used to attain the mole number of the total photons or the reactive photons which passed through the gross area of above strainer per unit time. To adopt adjusting the length between the photoreactor and the xenon arc lamp, the incident photon gamma flux on the photoreactor was varied.
The incident photon gamma flux Io surveyed by a radiometer-type receiver (Model FZ-A, Photoelectric Instrument Factory Beijing Normal University, Beijing, China) was determined to be 4.76 × 10 −6 Einstein L −1 s −1 under VLGI (pitch of waves area of 400-700 nm). The incident photon gamma flux on the photoreactor was varied by adjusting the length between the photoreactor and the Xe arc lamp.
The PE was obtained based on the following Equation (5): where φ was the PE (%), R was the retrogradation velocity of ENR (mol L −1 s −1 ), and I o was the incident photon gamma flux (Einstein L −1 s −1 ).

Conclusions
Er 2 FeSbO 7 was prepared by solid-state modus. For the first time, EBHP was prepared by a facile solvothermal modus. The optical physics characteristics of the single phase Er 2 FeSbO 7 and EBH were researched and verified with SEM, XRD, UV-Vis DRS and XPS tests. The main conclusion is that Er 2 FeSbO 7 is a pure phase which crystallizes in a pyrite architecture that maintains a cubic crystal system with the space group Fd3m. The crystal parameter or the BDG of Er 2 FeSbO 7 were a = 10.179902 Å or 1.88 eV. EBHP was circumstantiated to be an efficient photocatalyzer for exterminating ENR in the waste water. After VLGI-150min, the RR of ENR or TOC reached 99.16% or 94.96%. The RR of ENR with EBH as a photocatalyzer was 1.151 times, 1.269 times or 2.524 times higher than that with Er 2 FeSbO 7 , BiTiSbO 7 or N-TO as a catalyzer. The photocatalytic activity, which ranged from high to low among the above four photocatalysts, was as follows: EBHP > Er 2 FeSbO 7 > BiTiSbO 6 > N-TO. It could be proved that the oxidation removal ability for degrading ENR, which ranged from high to low among three oxidation radicals groups, was as follows: superoxide anion > HRS > holes. Therefore, it can be concluded that using EBH as a photocatalyst might be a potent method for treating pharmaceutical wastewater that is polluted by ENR. Finally, the possible photodegradation pathway for ENR was speculated.