Short-Time Hydrothermal Synthesis of CuBi2O4 Nanocolumn Arrays for Efficient Visible-Light Photocatalysis

In this article, a short-time hydrothermal method is developed to prepare CuBi2O4 nanocolumn arrays. By using Bi(NO3)3·5H2O in acetic acid and Cu(NO3)2·3H2O in ethanol as precursor solutions, tetragonal CuBi2O4 with good visible light absorption can be fabricated within 0.5 h at 120 °C. Tetragonal structured CuBi2O4 can be formed after 15 min hydrothermal treatment, however it possesses poor visible light absorption and low photocatalytic activity. Extending the hydrothermal treatment duration to 0.5 h results in a significant improvement invisible light absorption of the tetragonal CuBi2O4. The CuBi2O4 obtained through 0.5 h hydrothermal synthesis shows a band gap of 1.75 eV and exhibits the highest photocatalytic performance among the CuBi2O4 prepared with various hydrothermal time. The removal rate of methylene blue by the 0.5 h CuBi2O4 reaches 91% under visible light irradiation for 0.5 h. This study proposes a novel strategy to prepare photoactive CuBi2O4 nanocolumn arrays within 0.5 h at a moderate temperature of 120 °C. The hydrothermal method provides a facile strategy for the fast synthesis of metal-oxide-based photocatalysts at mild reaction conditions.

In this study, we have successfully prepared tetrahedral CuBi 2 O 4 nanocolumn arrays through a rapid hydrothermal method at mild reaction conditions. Tetrahedral CuBi 2 O 4 with a band gap of 1.75 eV can be obtained within 0.5 h hydrothermal treatment at a moderate temperature of 120 • C. The

Material Preparation
The CuBi 2 O 4 nanocolumn arrays were prepared by the hydrothermal method. Typically, 0.9701 g Bi(NO 3 ) 3 ·5H 2 O (99.999% Alfa Aesar (China) Chemical Co. Ltd., Shanghai, China) was dissolved in 5 mL acetic acid, and 0.2416 g Cu(NO 3 ) 2 ·3H 2 O (99.8% Alfa Aesar (China) Chemical Co. Ltd., Shanghai, China) was dissolved in 25 mL ethyl alcohol [22,23]. Then, the two solutions were mixed and NaOH solution was added under magnetic stirring at room temperature until pH 14 was reached. Subsequently, the above mixture was transferred into a 100 mL Teflon-lined steel autoclave, and then maintained in an oven at 120 • C for 5 min, 10 min, 15 min, 0.5 h, 1 h, 2 h, 8 h, and 12 h, respectively (the samples were denoted as CBO-5 min, CBO-10 min, CBO-15 min, CBO-0.5 h, CBO-1 h, CBO-2 h, CBO-8 h, and CBO-12 h, correspondingly). The obtained products were washed with ethanol and distilled water three times respectively and then dried in a vacuum drying oven for 12 h at 80 • C.

Characterization
The crystal structure was analyzed by an X-ray diffractometer (XRD, Bruker D 8 Advance, Karlsruhe, Germany) using Cu kα 1 as the radiation source. The morphologies of the CBO were characterized using a scanning electron microscopy (SEM, HITACHI S-3700N, Tokyo, Japan). X-ray photoelectron spectroscopy (XPS, ESCALAB 250XI, Thermo Fisher Scientific, Waltham, MA, USA) was carried out to investigate the stoichiometry and chemical bonding states of the samples. The UV-vis absorption spectra of the samples were collected by a UV-2800 spectrophotometer (Shanghai, China).

Photocatalytic Activity Measurement
First, 40 mg of the photocatalyst was dispersed into 50 mL methylene blue dye solution (0.02 mM). Then, 0.05 mL H 2 O 2 was added and stirred in the dark for 30 min to reach the adsorption equilibrium. A 300 W xenon lamp with a 420 nm cut filter was used as the light source. Then, 2 mL aliquots were extracted and centrifuged every 5 min. The concentration of the remaining methylene blue was then analyzed by measuring the absorption of the supernate at 664 nm (the maximum absorption wavelength of methylene blue) via a UV-vis spectrophotometer. The degradation rate was calculated by the following formula: where C 0 and C refer to the absorbance of the solution at 664 nm before and after the test.

Synthetic Procedures of CuBi 2 O 4 Nanocolumn Arrays
Different from traditional hydrothermal methods, the Bi(NO 3 ) 3 ·5H 2 O was dissolved in acetic acid to avoid the use of strong acid, and the Cu(NO 3 ) 2 ·3H 2 O was dissolved in ethanol instead of water in order to create high vapor pressure at moderate hydrothermal temperature. Those precursor solutions were previously used for the spray pyrolysis of CuBi 2 O 4 thin films by Wang et al. [22]. It has been verified experimentally that both solutions can remain stable for at least 1 year at room temperature without stirring. As illustrated in Figure 1, a blue transparent solution was obtained when the colorless and transparent Bi(NO 3 ) 3 solution was mixed with the blue Cu(NO 3 ) 2 solution. When the NaOH solution was added dropwise to the mixture, the blue transparent solution first turned into a blue-white suspension, then a yellow-green suspension. After hydrothermal reaction for 0.5 h, a brownish black powder was finally obtained. The chemical reaction during the reaction can be expressed by the Formulas (2) to (5) [14]. The crystallization mechanism of CuBi 2 O 4 particles can be described as a "dissolution-crystallization" process [14], during which the amorphous precipitate BiOOH and Cu(OH) 2 are attacked by mineralizers in an alkaline environment to form ion aggregates. Subsequently, CuBi 2 O 4 particles are produced through nucleation, precipitation, dehydration, and crystal growth in a supersaturated solution.
Bi 3+ + 3OH − → Bi(OH) 3 (2) Nanomaterials 2019, 9, x FOR PEER REVIEW 3 of 10 turned into a blue-white suspension, then a yellow-green suspension. After hydrothermal reaction for 0.5 h, a brownish black powder was finally obtained. The chemical reaction during the reaction can be expressed by the formulas (2) to (5) [14]. The crystallization mechanism of CuBi2O4 particles can be described as a "dissolution-crystallization" process [14], during which the amorphous precipitate BiOOH and Cu(OH)2 are attacked by mineralizers in an alkaline environment to form ion aggregates. Subsequently, CuBi2O4 particles are produced through nucleation, precipitation, dehydration, and crystal growth in a supersaturated solution.
Bi 3+ + 3OH -→ Bi(OH) 3 (2)   [23][24][25]. For the samples obtained under the hydrothermal times longer than 15 min, similar XRD patterns were obtained. It can be obviously noticed that the peak width at half-height of the CBO-10 min was wider than those with longer hydrothermal times, indicating small crystal size at the beginning of the crystal growth. To further investigate the surface chemical state of CuBi2O4, the CBO-10 min, CBO-15 min, and CBO-0.5 h samples were analyzed by XPS. The full XPS spectra of CBO samples displayed in Figure S1a confirmed the existence of Cu, Bi, and O elements. The C 1s high-resolution spectra of different samples were shown in Figure S1b. The binding energies were calibrated using the C 1s peak at 284.8 eV. It was observed that the three samples displayed similar elemental characteristic peaks. As shown in Figure 2b, the Cu 2+ peaks of the CBO-0.5 h shifted to lower binding energy of 933.65 eV and 953.35 eV, which corresponded to the Cu 2p3/2 and Cu 2p1/2 binding energies of Cu 2+ , respectively. Two shake-up satellites appeared at about 962 eV and 941 eV [1,2,26,27]. The high-resolution Bi 4f spectrum of pristine CBO-0.5 h in Figure 2c can be divided into two main peaks at 164.05 eV and 158.8 eV, which originated from Bi 4f5/2 and Bi 4f7/2 binding energies of Bi 3+ , respectively. The O 1s spectrum of CBO-0.5 was shown in Figure 2d, where the peak at 529.55 eV can be ascribed to the lattice oxygen [4,28,29]. As for the CBO-10 min and CBO-15 min samples, new peaks centered at about 530.9 eV and 531.4 eV in the XPS spectra of O 1s emerged (Figure 2d), which may represent the presence of subsurface oxygen vacancy defects [30].   [23][24][25]. For the samples obtained under the hydrothermal times longer than 15 min, similar XRD patterns were obtained. It can be obviously noticed that the peak width at half-height of the CBO-10 min was wider than those with longer hydrothermal times, indicating small crystal size at the beginning of the crystal growth. To further investigate the surface chemical state of CuBi 2 O 4 , the CBO-10 min, CBO-15 min, and CBO-0.5 h samples were analyzed by XPS. The full XPS spectra of CBO samples displayed in Figure S1a confirmed the existence of Cu, Bi, and O elements. The C 1s high-resolution spectra of different samples were shown in Figure S1b. The binding energies were calibrated using the C 1s peak at 284.8 eV. It was observed that the three samples displayed similar elemental characteristic peaks. As shown in Figure 2b, the Cu 2+ peaks of the CBO-0.5 h shifted to lower binding energy of 933.65 eV and 953.35 eV, which corresponded to the Cu 2p 3/2 and Cu 2p 1/2 binding energies of Cu 2+ , respectively. Two shake-up satellites appeared at about 962 eV and 941 eV [1,2,26,27]. The high-resolution Bi 4f spectrum of pristine CBO-0.5 h in Figure 2c can be divided into two main peaks at 164.05 eV and 158.8 eV, which originated from Bi 4f 5/2 and Bi 4f 7/2 binding energies of Bi 3+ , respectively. The O 1s spectrum of CBO-0.5 was shown in Figure 2d, where the peak at 529.55 eV can be ascribed to the lattice oxygen [4,28,29]. As for the CBO-10 min and CBO-15 min samples, new peaks centered at about 530.9 eV and 531.4 eV in the XPS spectra of O 1s emerged (Figure 2d), which may represent the presence of subsurface oxygen vacancy defects [30]. Nanomaterials 2019, 9, x FOR PEER REVIEW 4 of 10 To investigate the morphologies of CuBi2O4 with different hydrothermal time, SEM was conducted. When the hydrothermal time was only 5 min (Figure 3a), no obvious appearance was found except for some aggregated nanoparticles. As the hydrothermal time increased to 10 min (Figure 3b), some spherical particles were formed, which were dispersed and arranged in a random manner. After 15 min hydrothermal treatment (Figure 3c), hierarchical CuBi2O4 nanocolumn arrays were created. This is consistent with the XRD results in Figure 2d, where the CBO-5 min and CBO-10 min were found to be amorphous while the CBO-15 min showed well defined tetragonal structure. When the hydrothermal time increased from 0.5 h to 12 h, the hierarchical microstructures continued to grow. As shown in Figure 3d-f, the size of the nanorods increased and they became closer to each other, but in general the CuBi2O4 still appeared as nanorod arrays without substantial change in morphology. The proposed fabrication mechanism of CuBi2O4 nanocolumn arrays via the rapid hydrothermal synthesis route was illustrated in Scheme 1. Initially, the CuBi2O4 cores were created, based on which the nanorods started to grow. Then, the nanocolumns were obtained with the continuous generation and growth of the nanorods. Finally, the nanocolumn arrays were fabricated through the self-assembling of the nanocolumns. The repaid nucleation and high crystal growth rate of CuBi2O4 greatly shortened the hydrothermal reaction time and reduce the energy consumption. To investigate the morphologies of CuBi 2 O 4 with different hydrothermal time, SEM was conducted. When the hydrothermal time was only 5 min (Figure 3a), no obvious appearance was found except for some aggregated nanoparticles. As the hydrothermal time increased to 10 min (Figure 3b), some spherical particles were formed, which were dispersed and arranged in a random manner. After 15 min hydrothermal treatment (Figure 3c), hierarchical CuBi 2 O 4 nanocolumn arrays were created. This is consistent with the XRD results in Figure 2d, where the CBO-5 min and CBO-10 min were found to be amorphous while the CBO-15 min showed well defined tetragonal structure. When the hydrothermal time increased from 0.5 h to 12 h, the hierarchical microstructures continued to grow. As shown in Figure 3d-f, the size of the nanorods increased and they became closer to each other, but in general the CuBi 2 O 4 still appeared as nanorod arrays without substantial change in morphology. The proposed fabrication mechanism of CuBi 2 O 4 nanocolumn arrays via the rapid hydrothermal synthesis route was illustrated in Scheme 1. Initially, the CuBi 2 O 4 cores were created, based on which the nanorods started to grow. Then, the nanocolumns were obtained with the continuous generation and growth of the nanorods. Finally, the nanocolumn arrays were fabricated through the self-assembling of the nanocolumns. The repaid nucleation and high crystal growth rate of CuBi 2 O 4 greatly shortened the hydrothermal reaction time and reduce the energy consumption.

Optical Absorption Properties
UV-vis diffuse reflectance spectra of CuBi2O4 with different hydrothermal time were investigated and displayed in Figure 4a. Tauc plots were applied to determine the band energy of different CuBi2O4 samples, as illustrated in Figure 4b. The CBO-15 min showed poor visible light absorption and its exact band gap could hardly be determined from the Tauc plot. In comparison with the CBO-15 min, the samples with longer hydrothermal time exhibited significantly enhanced visible light absorption. The CBO-0.5 h sample showed a direct band gap of 1.75 eV, which lies in the band gap range of CuBi2O4 as previously reported [22,23,28,31]. Further hydrothermal treatment for more than 1 h did not lead to improvement in visible light absorption.

Photocatalytic Activity of CuBi2O4
As showed previously in the XRD results, amorphous CuBi2O4 was obtained when the hydrothermal treatment time was shorter than 15 min. Therefore, the CuBi2O4 samples with treatment time over 15 min were selected to evaluate the photocatalytic performance via the degradation of methylene blue under visible light. As shown in Figure 5a, the CBO-15 min sample showed low photocatalytic performance, which can be ascribed to its poor visible light absorption (Figure 5a). CBO-0.5 h showed significantly higher photocatalytic activity than CBO-15 min, which can be explained by its good visible light absorption as shown in Figure 4. Among all the samples with different hydrothermal treatment durations, the CBO-0.5 h exhibited the highest photocatalytic performance, of which a removal rate of 91% was achieved within 30 min in the presence of a trace amount of H2O2 (about 0.05 mL in 50 mL solution). The photocatalytic degradation efficiency of the CBO-0.5 h is comparable to the previously reported CuBi2O4 prepared by traditional method with much longer hydrothermal treatment duration [6]. Further extending the hydrothermal time to 1 h or longer did not result in significant change in the photocatalytic performance. In general, photocatalysts with high specific surface areas and large pore volumes are beneficial for the enhancement of photocatalytic performance. To elucidate the effect of specific area, BET measurements were carried out. Figure S3 showed the nitrogen adsorption-desorption isotherms of CuBi2O4 with different hydrothermal time. It can be seen that all the samples had isotherms of type IV, indicating the presence of mesopores (2-50 nm). The isotherms exhibited H3 hysteresis loops at a

Photocatalytic Activity of CuBi 2 O 4
As showed previously in the XRD results, amorphous CuBi 2 O 4 was obtained when the hydrothermal treatment time was shorter than 15 min. Therefore, the CuBi 2 O 4 samples with treatment time over 15 min were selected to evaluate the photocatalytic performance via the degradation of methylene blue under visible light. As shown in Figure 5a, the CBO-15 min sample showed low photocatalytic performance, which can be ascribed to its poor visible light absorption (Figure 5a). CBO-0.5 h showed significantly higher photocatalytic activity than CBO-15 min, which can be explained by its good visible light absorption as shown in Figure 4. Among all the samples with different hydrothermal treatment durations, the CBO-0.5 h exhibited the highest photocatalytic performance, of which a removal rate of 91% was achieved within 30 min in the presence of a trace amount of H 2 O 2 (about 0.05 mL in 50 mL solution). The photocatalytic degradation efficiency of the CBO-0.5 h is comparable to the previously reported CuBi 2 O 4 prepared by traditional method with much longer hydrothermal treatment duration [6]. Further extending the hydrothermal time to 1 h or longer did not result in significant change in the photocatalytic performance. In general, photocatalysts with high specific surface areas and large pore volumes are beneficial for the enhancement of photocatalytic performance. To elucidate the effect of specific area, BET measurements were carried out. Figure S3 showed the nitrogen adsorption-desorption isotherms of CuBi 2 O 4 with different hydrothermal time. It can be seen that all the samples had isotherms of type IV, indicating the presence of mesopores (2-50 nm). The isotherms exhibited H3 hysteresis loops at a high relative pressure range from 0.8 to 1.0, suggesting the presence of slit-like pores. As shown in Table S1, no significant difference in BET surface areas and pore volumes was found among the CBO samples with different hydrothermal time, therefore the difference in degradation rate should not be attributed to the specific surface area of samples. In addition, the optimization experiments of sodium hydroxide concentration showed that the concentration of the sodium hydroxide had no significant effect on the structure of the CuBi 2 O 4 and 6.5 M sodium hydroxide was used in this study ( Figure S2). The photocatalytic performance of CuBi2O4 samples with different hydrothermal temperatures was also investigated. As shown in Figure 5b, CuBi2O4 sample with the hydrothermal temperature of 120 °C could achieve comparable photocatalytic performance as the one treated at a much higher temperature of 200 °C . When the temperature exceeded 120 °C , the (312) and (420) peaks appeared, nevertheless the emerged crystal planes of (312) and (420) had no promoting effect on the photocatalytic activity. From an energetic point of view, hydrothermal treatment at low temperature and low pressure is highly expected. Therefore, 120 °C was selected as a suitable hydrothermal temperature in this study. Eventually, we successively prepared the CuBi2O4 nanocolumn arrays via a facile hydrothermal method with much shorter reaction time at lower temperature compared to the traditional hydrothermal synthesis routes [4,7,14,25,30,32,33]. The time-dependent UV-vis absorption spectra of methylene blue (MB) under light irradiation for CBO-0.5 h at the hydrothermal temperature of 120 °C were presented in Figure 5d. The intensity of the absorption peak gradually decreased and eventually diminished as the irradiation time increased from 0 to 30 min.

Stability of the CuBi2O4 Nanocolumn Arrays
The reusability of the CuBi2O4 samples was tested by repeating the photodegradation experiment under visible light irradiation. After each cycle of the photocatalysis experiment, the dye solution was adjusted to the initial concentration and the separated photocatalysts were washed and used again. As illustrated in Figure 6a, the decomposition rate of MB by the CBO-0.5 h slightly decreased by only 6.6% after five cycles of experiments, demonstrating the high stability of the asprepared CuBi2O4 photocatalyst. We note that the decrease may be partially due to the loss of the sample powder during the separation and cleaning processes. The XRD patterns before and after 5 The photocatalytic performance of CuBi 2 O 4 samples with different hydrothermal temperatures was also investigated. As shown in Figure 5b, CuBi 2 O 4 sample with the hydrothermal temperature of 120 • C could achieve comparable photocatalytic performance as the one treated at a much higher temperature of 200 • C. When the temperature exceeded 120 • C, the (312) and (420) peaks appeared, nevertheless the emerged crystal planes of (312) and (420) had no promoting effect on the photocatalytic activity. From an energetic point of view, hydrothermal treatment at low temperature and low pressure is highly expected. Therefore, 120 • C was selected as a suitable hydrothermal temperature in this study. Eventually, we successively prepared the CuBi 2 O 4 nanocolumn arrays via a facile hydrothermal method with much shorter reaction time at lower temperature compared to the traditional hydrothermal synthesis routes [4,7,14,25,30,32,33]. The time-dependent UV-vis absorption spectra of methylene blue (MB) under light irradiation for CBO-0.5 h at the hydrothermal temperature of 120 • C were presented in Figure 5d. The intensity of the absorption peak gradually decreased and eventually diminished as the irradiation time increased from 0 to 30 min.

Stability of the CuBi 2 O 4 Nanocolumn Arrays
The reusability of the CuBi 2 O 4 samples was tested by repeating the photodegradation experiment under visible light irradiation. After each cycle of the photocatalysis experiment, the dye solution was adjusted to the initial concentration and the separated photocatalysts were washed and used again. As illustrated in Figure 6a, the decomposition rate of MB by the CBO-0.5 h slightly decreased by only 6.6% after five cycles of experiments, demonstrating the high stability of the as-prepared CuBi 2 O 4 photocatalyst. We note that the decrease may be partially due to the loss of the sample powder during the separation and cleaning processes. The XRD patterns before and after 5 cycles of photodegradation experiments (Figure 6b) showed that the CBO-0.5 h maintained the tetragonal structure.

Conclusions
In this study, we developed a facile hydrothermal method for the fast preparation of tetragonal CuBi2O4 nanorod arrays with good light absorption and high photocatalytic performance. The use of Bi(NO3)3·5H2O in acetic acid and Cu(NO3)2·3H2O in ethanol as precursor solutions allowed for the rapid hydrothermal synthesis of CuBi2O4 nanorod arrays at mild reaction conditions. Amorphous CuBi2O4 was obtained when the hydrothermal treatment time was shorter than 15 min. Tetragonal structured CuBi2O4 can be achieved when the hydrothermal treatment time reached 15 min, whereas the CBO-15 min presented poor visible light absorption and low photocatalytic performance. Extending the hydrothermal treatment duration to 0.5 h resulted in significantly improvement in visible light absorption and photocatalytic activity. Further prolonging the hydrothermal time to 1 h or even longer did not result in significant change in the photocatalytic performance. The CuBi2O4 with a hydrothermal treatment time of 0.5 h exhibited a degradation efficiency up to 91% for MB degradation, which is comparable to the CuBi2O4 prepared by traditional methods that commonly consume 12-36 h. Our hydrothermal method provides new insights for the rapid preparation of metal-oxide-based photocatalysts at moderate temperature and pressure.
Author Contributions: F.W. and Q.X. conceived and designed the experiments. F.C. performed the experiments. P.G., Y.L. and Y.W. provided guidance for the general idea of the manuscript. All authors analyzed the data. All

Conclusions
In this study, we developed a facile hydrothermal method for the fast preparation of tetragonal CuBi 2 O 4 nanorod arrays with good light absorption and high photocatalytic performance. The use of Bi(NO 3 ) 3 ·5H 2 O in acetic acid and Cu(NO 3 ) 2 ·3H 2 O in ethanol as precursor solutions allowed for the rapid hydrothermal synthesis of CuBi 2 O 4 nanorod arrays at mild reaction conditions. Amorphous CuBi 2 O 4 was obtained when the hydrothermal treatment time was shorter than 15 min. Tetragonal structured CuBi 2 O 4 can be achieved when the hydrothermal treatment time reached 15 min, whereas the CBO-15 min presented poor visible light absorption and low photocatalytic performance. Extending the hydrothermal treatment duration to 0.5 h resulted in significantly improvement in visible light absorption and photocatalytic activity. Further prolonging the hydrothermal time to 1 h or even longer did not result in significant change in the photocatalytic performance. The CuBi 2 O 4 with a hydrothermal treatment time of 0.5 h exhibited a degradation efficiency up to 91% for MB degradation, which is comparable to the CuBi 2 O 4 prepared by traditional methods that commonly consume 12-36 h. Our hydrothermal method provides new insights for the rapid preparation of metal-oxide-based photocatalysts at moderate temperature and pressure.