Temperature-Program Assisted Synthesis of Novel Z-Scheme CuBi2O4/β-Bi2O3 Composite with Enhanced Visible Light Photocatalytic Performance

Novel Z-Scheme CuBi2O4/β-Bi2O3 composite photocatalysts with different mass ratios and calcination temperatures were firstly synthesized by the hydrothermal method following a temperature-programmed process. The morphology, crystal structure, and light absorption properties of the as-prepared samples were systematically characterized, and the composites exhibited enhanced photocatalytic activity toward diclofenac sodium (DS) degradation compared with CuBi2O4 and β-Bi2O3 under visible light irradiation. The optimal photocatalytic efficiency of the composite, achieved at the mass ratio of CuBi2O4 and β-Bi2O3 of 1:2.25 and the calcination temperature of 600 °C is 92.17%. After the seventh recycling of the composite, the degradation of DS can still reach 82.95%. The enhanced photocatalytic activity of CuBi2O4/β-Bi2O3 is closely related to OH•, h+ and O2•−, and the photocatalytic mechanism of CuBi2O4/β-Bi2O3 can be explained by the Z-Scheme theory.


Introduction
Diclofenac sodium (DS) is a very effective anti-inflammatory drug widely used in many countries [1,2]. Due to the poor absorbability of organisms to DS, most of the DS will be released into the environment after it is taken [1]. However, long-term exposure to a DS-polluted environment will cause serious risk to ecosystems and human health [3,4]. Therefore, effective degradation technologies for DS should be explored to control its persistent contamination.
Semiconductor photocatalytic technology has aroused widespread interest in organic pollution control [5][6][7][8] due to its advantages of rich catalysts, mild reaction conditions, fast reaction speed, and no secondary pollution. In past decades, titanium dioxide (TiO 2 ) has attracted considerable attention for its low cost and chemical inertness [8][9][10]. However, the large band gap of TiO 2 , which is approximately 3.2 eV, limits its absorption of sunlight to the ultraviolet (UV) region [10,11]. Because the visible light occupies 43% of sunlight, whereas UV light occupies only 4%, the development of visible-light responsive semiconductor photocatalysts is becoming a research hotspot [12][13][14][15].

Synthesis of CuBi 2 O 4 /β-Bi 2 O 3
The detailed synthesis pathway of CuBi 2 O 4 /β-Bi 2 O 3 is illustrated in Figure 1. Firstly, CuBi 2 O 4 was prepared through a simple hydrothermal method, and the detailed experiment processes are similar to that reported by our previous paper [16]. Secondly, a core-shell structural CuBi 2 O 4 @C was synthesized, that is, 0.1 g of CuBi 2 O 4 was ultrasonically dispersed into 70 mL of aqueous solution (with 0.3 mL gluconic acid), and then the solution was transferred into a 100-mL sealed Teflon-lined stainless steel autoclave for 4 h at 180 • C. When the autoclave was cooled naturally to room temperature, the precipitate was isolated by centrifugation and washed several times with distilled water. Thirdly, a temperature-programmed method was used to synthesize the CuBi 2 O 4 /β-Bi 2 O 3 . Then 0.1 g of CuBi 2 O 4 @C obtained above was ultrasonically dispersed into 20 mL HNO 3 solution (1 mol/L) to obtain solution A; 0.39 mmol Bi(NO 3 ) 3 ·5H 2 O was completely dissolved into another 20 mL HNO 3 solution (1 mol/L) under the acute agitation to obtain solution B; 2.34 mmol Na 2 CO 3 was dissolved into 40 mL ultrapure water to obtain solution C. After solutions A and B were well mixed, solution C was added into the mixture drop by drop, and a large amount of white precipitate was produced. Then the washed precipitate by ethanol and ultrapure water was placed into a temperature-programmed furnace, and the reaction furnace was set to be heated to 600 • C within 30  were also prepared through the above synthesis process, and the detailed synthesis steps can also be seen from the Figure 1. FOR PEER REVIEW  3 of 12 composition, the pure CuBi2O4 and β-Bi2O3 were also prepared through the above synthesis process, and the detailed synthesis steps can also be seen from the Figure 1. Figure 1. The detailed synthesis pathway of CuBi2O4/β-Bi2O3.

Characterization
Powder X-ray diffraction (XRD) patterns were examined to study the crystal structure and phase composition of the materials using the D/max 2500 PC (Rigaku, Japan) instrument with Cu Kα radiation (40 kV, 100 mA) at a rate of 4.0°/min over a 2θ range of 20°-60°. Morphologies of the samples were characterized by scanning electron microscopy (SEM, JSM-6360LV, JEOL, Tokyo, Japan). Ultraviolet-visible (UV-Vis) diffusive reflectance spectra were obtained by a Shimadzu UV-2550 spectrophotometer (Shimadzu, Kyoto, Japan) over the analysis range from 200 to 800 nm. X-ray photoelectron spectra (XPS) were recorded using the ESCALAB 250 instrument (Thermo Scientific, Waltham, MA, USA) with Al Kα radiation.

Photocatalytic Performance Experiments
The photocatalytic performance experiments were conducted in photoreaction apparatus (BL-GHX-V, Bilang Biological Science and Technology Co., Ltd., Xi'an, China) using a 300 W Xe lamp with an ultraviolet cutoff filter (providing visible light ≥ 400 nm) as the light source. In each experiment, a 25 mg photocatalyst was added to 50 mL of DS solution (8 mg/L). Before irradiation, the solution was magnetically stirred in the dark for 30 min to reach the adsorption-desorption equilibrium. As the reaction time elapsed, the sample was taken out and filtered immediately with 0.45 µm membrane filters, the concentrations of DS and Total Organic Carbon (TOC) are measured by a high performance liquid chromatography (HPLC, Agilent 1260, Santa Clara, CA, USA) and TOC (Shimadzu TOC-VCPH, Kyoto, Japan) analyzer, respectively. In addition, repeated experiments for DS degradation were also conducted to study the stability of the as-prepared photocatalysts, and the operation processes were similar to the photocatalytic experiments. All experiments were repeated twice and the data shown in the article was averaged.

Analysis of Reactive Species
Free radical capture experiments were used to ascertain the reactive species of CuBi2O4/β-Bi2O3 photocatalytic system, and tert-butanol (t-BuOH) was chosen as the hydroxyl radical (OH • ) scavenger, disodium ethylenediamine tetraacetate (EDTA-Na2) was chosen as the hole (h + ) scavenger,

Characterization
Powder X-ray diffraction (XRD) patterns were examined to study the crystal structure and phase composition of the materials using the D/max 2500 PC (Rigaku, Japan) instrument with Cu Kα radiation (40 kV, 100 mA) at a rate of 4.0 • /min over a 2θ range of 20 • -60 • . Morphologies of the samples were characterized by scanning electron microscopy (SEM, JSM-6360LV, JEOL, Tokyo, Japan). Ultraviolet-visible (UV-Vis) diffusive reflectance spectra were obtained by a Shimadzu UV-2550 spectrophotometer (Shimadzu, Kyoto, Japan) over the analysis range from 200 to 800 nm. X-ray photoelectron spectra (XPS) were recorded using the ESCALAB 250 instrument (Thermo Scientific, Waltham, MA, USA) with Al Kα radiation.

Photocatalytic Performance Experiments
The photocatalytic performance experiments were conducted in photoreaction apparatus (BL-GHX-V, Bilang Biological Science and Technology Co., Ltd., Xi'an, China) using a 300 W Xe lamp with an ultraviolet cutoff filter (providing visible light ≥ 400 nm) as the light source. In each experiment, a 25 mg photocatalyst was added to 50 mL of DS solution (8 mg/L). Before irradiation, the solution was magnetically stirred in the dark for 30 min to reach the adsorption-desorption equilibrium. As the reaction time elapsed, the sample was taken out and filtered immediately with 0.45 µm membrane filters, the concentrations of DS and Total Organic Carbon (TOC) are measured by a high performance liquid chromatography (HPLC, Agilent 1260, Santa Clara, CA, USA) and TOC (Shimadzu TOC-V CPH , Kyoto, Japan) analyzer, respectively. In addition, repeated experiments for DS degradation were also conducted to study the stability of the as-prepared photocatalysts, and the operation processes were similar to the photocatalytic experiments. All experiments were repeated twice and the data shown in the article was averaged.

Analysis of Reactive Species
Free radical capture experiments were used to ascertain the reactive species of CuBi 2 O 4 /β-Bi 2 O 3 photocatalytic system, and tert-butanol (t-BuOH) was chosen as the hydroxyl radical (OH • ) scavenger, disodium ethylenediamine tetraacetate (EDTA-Na 2 ) was chosen as the hole (h + ) scavenger, benzoquinone (BZQ) was chosen as the superoxide radical (O 2 •− ) scavenger. The detailed experiment processes were similar to the photocatalytic activity experiments.

Results and Discussion
Figure 2 depicts scanning electron microscope (SEM) images of the as-prepared materials. A well-distributed micro-sphere structure (~5 µm) with smooth surface for pure CuBi 2 O 4 is shown in Figure 2a. The transition material of CuBi 2 O 4 @C, from Figure 2b, shows increased size and relatively rough surface compared with pure CuBi 2 O 4 . Moreover, some dispersed carbon microspheres exist as well. In the formation process of CuBi 2 O 4 /β-Bi 2 O 3 , the control of temperature is an important step. As shown from Figure 2c, the products exhibit a chain-connected spherical structure when the temperature is at 400 • C, and the connection matter is the incompletely burned carbon. That is because the complete combustion temperature of carbon is higher than 500 • C [29]. Therefore, the composition of calcined products just contain CuBi 2 O 4 and β-Bi 2 O 3 when the temperatures are at 600 • C and 800 • C. But the rising of calcination temperature will destroy sphere-structural CuBi 2 O 4 , which can be seen from the SEM images of Figure 2d,e.

Results and Discussion
Figure 2 depicts scanning electron microscope (SEM) images of the as-prepared materials. A well-distributed micro-sphere structure (~5 µm) with smooth surface for pure CuBi2O4 is shown in Figure 2a. The transition material of CuBi2O4@C, from Figure 2b, shows increased size and relatively rough surface compared with pure CuBi2O4. Moreover, some dispersed carbon microspheres exist as well. In the formation process of CuBi2O4/β-Bi2O3, the control of temperature is an important step. As shown from Figure 2c, the products exhibit a chain-connected spherical structure when the temperature is at 400 °C, and the connection matter is the incompletely burned carbon. That is because the complete combustion temperature of carbon is higher than 500 °C [29]. Therefore, the composition of calcined products just contain CuBi2O4 and β-Bi2O3 when the temperatures are at 600 °C and 800 °C. But the rising of calcination temperature will destroy sphere-structural CuBi2O4, which can be seen from the SEM images of Figure [22,25]. Moreover, there are no obvious carbon peaks from the XRD pattern of CuBi 2 O 4 /β-Bi 2 O 3 (600 • C), indicating the complete combustion of carbon in this calcination temperature [29]. The purity and crystallinity of CuBi2O4 and CuBi2O4/β-Bi2O3 calcined at 600 °C were examined with XRD, as depicted in Figure 3. From the XRD pattern of CuBi2O4, the diffraction peaks can be perfectly indexed to the phase of CuBi2O4 (JCPDS No. 84-1969) [30]. In the XRD pattern of CuBi2O4/β-Bi2O3 (600 °C), except for the peaks indicating CuBi2O4, the diffraction peaks located at the 2θ values of 27.66°, 32.86°, 46.32°, 55.44°, and 57.86°, can be indexed to the (201), (220), (222), (421), and (402) crystalline planes of tetragonal β-Bi2O3 (JCPDS No. 27-0050) [22,25]. Moreover, there are no obvious carbon peaks from the XRD pattern of CuBi2O4/β-Bi2O3 (600 °C), indicating the complete combustion of carbon in this calcination temperature [29]. The UV-Vis absorption spectrum of CuBi2O4, β-Bi2O3 (600 °C) and CuBi2O4/β-Bi2O3 (600 °C) are displayed in Figure 4a. Both of the materials exhibit strong absorbance in the UV and visible light regions, and the maximum absorption boundary of CuBi2O4, β-Bi2O3 (600 °C) and CuBi2O4/β-Bi2O3 (600 °C) appear at approximately 800 nm, 500 nm and 725 nm, respectively. The band gap energy (Eg) (600 • C) appear at approximately 800 nm, 500 nm and 725 nm, respectively. The band gap energy (E g ) of CuBi 2 O 4 and β-Bi 2 O 3 (600 • C) can be determined with the classic Tauc approach by using the following equation [31]: αhv = A hv − E g n/2 , where α, h, v, E g and A are the absorption coefficient, the Planck constant, the light frequency, the band gap energy, and a constant, respectively. In the equation, n is a number characteristic of the charge transition in a semiconductor, and n = 1 for a direct transition while n = 4 for an indirect transition [31]. As for CuBi 2 O 4 and β-Bi 2 O 3 (600 • C), n = 4 [32,33]. Therefore, their band gap energy could be elicited from the plot of light energy (αhv) 2 versus energy (hv), shown in Figure 4b, which suggests that the band gap energy of CuBi 2 O 4 and β-Bi 2 O 3 (600 • C) are 1.72 eV and 2.70 eV, respectively, which are very close to previous literature [33][34][35]. In addition, the valence band (VB) and conduction band (CB) edge position of the semiconductors can be estimated according to the following empirical equation: where E VB and E CB are the valence band (VB) and conduction band (CB) edge potentials, respectively; X is the electronegativity of the absolute electronegativity of the constituent atoms, that is 4.59 eV for CuBi 2 O 4 and 6.24 eV for β-Bi 2 O 3 [25,33]; E c is the energy of free electrons on the hydrogen scale (approximately 4.5 eV) [31].  [22,25]. Moreover, there are no obvious carbon peaks from the XRD pattern of CuBi2O4/β-Bi2O3 (600 °C), indicating the complete combustion of carbon in this calcination temperature [29]. The UV-Vis absorption spectrum of CuBi2O4, β-Bi2O3 (600 °C) and CuBi2O4/β-Bi2O3 (600 °C) are displayed in Figure 4a. Both of the materials exhibit strong absorbance in the UV and visible light regions, and the maximum absorption boundary of CuBi2O4, β-Bi2O3 (600 °C) and CuBi2O4/β-Bi2O3 (600 °C) appear at approximately 800 nm, 500 nm and 725 nm, respectively. The band gap energy (Eg) of CuBi2O4 and β-Bi2O3 (600 °C) can be determined with the classic Tauc approach by using the following equation [31]: αhv = A (hv -E g ) n/2 , where α, h, v, Eg and A are the absorption coefficient, the Planck constant, the light frequency, the band gap energy, and a constant, respectively. In the equation, n is a number characteristic of the charge transition in a semiconductor, and n = 1 for a direct transition while n = 4 for an indirect transition [31]. As for CuBi2O4 and β-Bi2O3 (600 °C), n = 4 [32,33]. Therefore, their band gap energy could be elicited from the plot of light energy (αhv) 2 versus  Figure 5c shows the photodegradation efficiency of DS in the system of composites CuBi 2 O 4 /β-Bi 2 O 3 with different mass ratios. The adsorption capacity of composites to DS is less than 10%, suggesting the removal of DS is closely related to photodegradation and photocatalysis. The degradation efficiency of DS is enhanced with increasing β-Bi 2 O 3 in the composite as the mass ratio of CuBi 2 O 4 and β-Bi 2 O 3 changes from 1:0.5 to 1:2.25, but will decrease with a further increase of the content of β-Bi 2 O 3 , i.e., the mass ratio of 1:4. Figure 5e shows the degradation efficiency of DS using CuBi 2 O 4 /β-Bi 2 O 3 calcined at different temperatures. It can be seen that the degradation efficiencies of DS are 84.37%, 92.17% and 76.80%, respectively, and the composites calcined at 400 • C, 600 • C and 800 • C. Figure 5b,d,f describe the degradation rate curves of DS in different photocatalytic systems which derive from the ln(C 0 /C) versus irradiation time and the values of degradation rates are listed in Table 1. The maximum degradation rate of DS can be obtained in the system of composite CuBi 2 O 4 /β-Bi 2 O 3 with the mass ratio of 1:2.25 and the calcination temperature of 600 • C, which is 0.0099 min −1 .  To further evaluate the degradation efficiency of DS in the as-prepared catalyst's system, the removal efficiencies of Total Organic Carbon (TOC) are also explored and the results are shown in Figure 6. In the blank irradiation system, only 18.01% of TOC can be removed. And in the pure β-Bi2O3 and CuBi2O4 photocatalytic systems, the TOC removal efficiencies are 22.52% and 30.59%, respectively. When CuBi2O4 and β-Bi2O3 with the mass ratio of 1:2.25 are mechanically mixed to add into the system, 35.96% of TOC are removed. But for the composites CuBi2O4/β-Bi2O3 with different mass ratios calcinated at various temperatures, the removal efficiencies of TOC can be greatly improved, and the maximum value reaches 57.37%, which is achieved in the photocatalytic system of CuBi2O4/β-Bi2O3 with the mass ratio of 1:2.25 and the calcination temperature of 600 °C. To further evaluate the degradation efficiency of DS in the as-prepared catalyst's system, the removal efficiencies of Total Organic Carbon (TOC) are also explored and the results are shown in Figure 6. In the blank irradiation system, only 18 Figure 7b shows the degradation rate curves of DS over CuBi2O4/β-Bi2O3 (1:2.25, 600 °C) under different recycling runs, and the values of rate constants are described in the illustration. It can be seen that the degradation rate of DS decreases from 0.0099 min −1 to 0.0071 min −1 after the seventh recycle. To further understand the photocatalytic stability of the asprepared CuBi2O4/β-Bi2O3, XRD and Bi 4f XPS spectrums of the recycled composites are conducted seven times and the results are shown in Figure 8. Compared with the XRD spectra of the fresh and reused CuBi2O4/β-Bi2O3 in Figure 8a, there are no obvious changes for the main peaks. But from the Figure 8b, a new peak occur in the composite after recycling expect for the two main peaks existing in both of the two samples. According to reports, the main 4f 7/2 peak at 158.32 eV for a fresh sample and 158.30 eV for a reused sample, and the other main 4f 5/2 peak at 163.60 eV for a fresh sample and 163.49 eV for a reused sample, correspond to the Bi 3+ oxidation state [36,37], which are in accordance to the presence of either CuBi2O4 or β-Bi2O3 phases. The extra peak at 156.63 eV in the reused sample is ascribed to Bi metal [38], indicating that some Bi 3+ in the substance was reduced to be Bi metal after the reaction.   Figure 8a, there are no obvious changes for the main peaks. But from the Figure 8b, a new peak occur in the composite after recycling expect for the two main peaks existing in both of the two samples. According to reports, the main 4f 7/2 peak at 158.32 eV for a fresh sample and 158.30 eV for a reused sample, and the other main 4f 5/2 peak at 163.60 eV for a fresh sample and 163.49 eV for a reused sample, correspond to the Bi 3+ oxidation state [36,37], which are in accordance to the presence of either CuBi 2 O 4 or β-Bi 2 O 3 phases. The extra peak at 156.63 eV in the reused sample is ascribed to Bi metal [38], indicating that some Bi 3+ in the substance was reduced to be Bi metal after the reaction.    Figure 7b shows the degradation rate curves of DS over CuBi2O4/β-Bi2O3 (1:2.25, 600 °C) under different recycling runs, and the values of rate constants are described in the illustration. It can be seen that the degradation rate of DS decreases from 0.0099 min −1 to 0.0071 min −1 after the seventh recycle. To further understand the photocatalytic stability of the as-  Moreover, the free radical capture experiments used to investigate the active species involved in the DS degradation with the CuBi2O4/β-Bi2O3 (1:2.25, 600 °C) photocatalyst are explored and the results are shown in Figure 9. The degradation efficiency of DS decreases to 41.74% when 1 mM EDTA-Na2 was added, and the DS degradation was completely inhibited with the addition of 2 mM EDTA-Na2. But the degradation efficiency of DS decreases to 54.98% and 27.82% when 2 mM t-BuOH or BZQ was added, respectively. The results suggest that the enhanced photocatalytic activity of CuBi2O4/β-Bi2O3 is closely related to OH • , h + and O2 •− , and the contribution order is h + > O2 •− > OH • .
As for the photocatalytic mechanisms of a semiconductor-semiconductor composite catalyst, the heterojunction energy band theory and Z-scheme theory were of concern [8]. Based on the analysis above, a possible Z-scheme photocatalytic mechanism was put forward to explain the enhanced photocatalytic activity of the CuBi2O4/β-Bi2O3 composite, which is illustrated in Figure 10. That is, both CuBi2O4 and β-Bi2O3 could generate the photoinduced electron-hole pairs under visiblelight illumination owing to their good light absorption abilities. Then, the formed Bi metal becomes the recombination center of the photogenerated electron from CB of β-Bi2O3 and the holes from the VB of the CuBi2O4 in the photocatalytic reaction process, leading to the improved charge separation. Besides, the dissolved O2 in the solution can be captured by the photogenerated electrons in the CB of CuBi2O4 to form O2 •− due to the more negative CB level of CuBi2O4 than the potential of O2/O2 •− (E(O2/O2 •− ) = 0.13 eV (vs. NHE) [39]; while the photogenerated holes in the VB of β-Bi2O3 can lead the H2O/OH − to be oxidized to OH • for the higher energy of holes in the VB of β-Bi2O3 than the potential  Figure 9. The degradation efficiency of DS decreases to 41.74% when 1 mM EDTA-Na 2 was added, and the DS degradation was completely inhibited with the addition of 2 mM EDTA-Na 2 . But the degradation efficiency of DS decreases to 54.98% and 27.82% when 2 mM t-BuOH or BZQ was added, respectively. The results suggest that the enhanced photocatalytic activity of CuBi 2 O 4 /β-Bi 2 O 3 is closely related to OH • , h + and O 2 •− , and the contribution order is h  Moreover, the free radical capture experiments used to investigate the active species involved in the DS degradation with the CuBi2O4/β-Bi2O3 (1:2.25, 600 °C) photocatalyst are explored and the results are shown in Figure 9. The degradation efficiency of DS decreases to 41.74% when 1 mM EDTA-Na2 was added, and the DS degradation was completely inhibited with the addition of 2 mM EDTA-Na2. But the degradation efficiency of DS decreases to 54.98% and 27.82% when 2 mM t-BuOH or BZQ was added, respectively. The results suggest that the enhanced photocatalytic activity of CuBi2O4/β-Bi2O3 is closely related to OH • , h + and O2 •− , and the contribution order is h + > O2 •− > OH • .
As for the photocatalytic mechanisms of a semiconductor-semiconductor composite catalyst, the heterojunction energy band theory and Z-scheme theory were of concern [8]. Based on the analysis above, a possible Z-scheme photocatalytic mechanism was put forward to explain the enhanced photocatalytic activity of the CuBi2O4/β-Bi2O3 composite, which is illustrated in Figure 10. That is, both CuBi2O4 and β-Bi2O3 could generate the photoinduced electron-hole pairs under visiblelight illumination owing to their good light absorption abilities. Then, the formed Bi metal becomes the recombination center of the photogenerated electron from CB of β-Bi2O3 and the holes from the VB of the CuBi2O4 in the photocatalytic reaction process, leading to the improved charge separation. As for the photocatalytic mechanisms of a semiconductor-semiconductor composite catalyst, the heterojunction energy band theory and Z-scheme theory were of concern [8]. Based on the analysis above, a possible Z-scheme photocatalytic mechanism was put forward to explain the enhanced photocatalytic activity of the CuBi 2 O 4 /β-Bi 2 O 3 composite, which is illustrated in Figure 10 [39]. Subsequently, the holes in the VB of β-Bi2O3, the formed OH • and O2 •− participate in the DS photodegradation. Figure 10. Supposed photocatalytic mechanism of CuBi2O4/β-Bi2O3 by the Z-scheme theory.

Conclusions
A combined hydrothermal and temperature-programmed method was employed to synthesize the CuBi2O4/β-Bi2O3 composite photocatalysts. The properties of the as-prepared materials were systematically characterized by SEM, XRD, XPS and UV-Vis. Moreover, the composites exhibit enhanced photocatalytic performance toward DS degradation under visible light irradiation, and the optimal photodegradation efficiency of DS is achieved in the catalytic system of CuBi2O4/β-Bi2O3 with the mass ratio of 1:2.25 and the calcination temperature of 600 °C. Besides, the active species in the CuBi2O4/β-Bi2O3 photocatalytic system were discussed through the free radical capture experiments, and the photocatalytic mechanism of CuBi2O4/β-Bi2O3 was also put forward.
Author Contributions: X.C. (Xiaojuan Chen) designed and performed the experiments, and drafted the manuscript. N.L. and R.Z. provided guidance for the general idea of the thesis. S.L., C.Y., W.X., S.X. and X.C. (Xin Chen) carried out the sample preparations, part of the material characterization analysis, and data collection. All authors read and approved the final manuscript.