RSM-Based Preparation and Photoelectrocatalytic Performance Study of RGO/TiO 2 NTs Photoelectrode

: In this paper, reduced graphene oxide (RGO) was prepared by a modiﬁed Hummers method and chemical reduction method, and an RGO/TiO 2 NTs (RGO/TiO 2 nanotubes) photoelectrode was prepared by the electrochemical deposition method. The as-prepared RGO/TiO 2 NTs were analyzed by scanning electron microscopy (SEM) and X-ray diffraction (XRD), and their photocatalytic activities were investigated by measuring the degradation of methylene blue (MB) under simulated solar light irradiation. The SEM and XRD results indicated that the original tubular structure of TiO 2 -NTs was not changed after RGO modiﬁcation. The surface of the TiO 2 NTs photoelectrode was covered with a non-uniform, ﬂake-shaped reduced graphene oxide ﬁlm. The thickness of the RGO/TiO 2 NTs was increased to about 22.60 nm. The impedance of the RGO/TiO 2 NTs was smaller than that of the TiO 2 NT photoelectrode. The optimal preparation conditions of RGO/TiO 2 NT photoelectrodes were investigated by using a single factor method and response surface method. The best preparation conditions were as follows: deposition potential at 1.19 V, deposition time of 10.27 min, and deposition temperature at 24.94 ◦ C. The precision was 15.193, more 4, and proved rea- sonable; the was 2.64%, less than 10%, indicating the high reliability and accuracy of the test. The analysis results showed that the regression equation model was in good agreement with the actual test, which can effectively predict the test results. It is recom- mended to use this model to predict the optimal conditions for the preparation of NTs.


Introduction
In recent years, the discharge of a large amount of dye wastewater has caused a huge threat to the environment, due to the toxicity, mutagenicity and carcinogenicity of dyes [1,2]. Among various dyes, methylene blue is one of the most common pollutants to prevent wastewater [3]. In terms of ecological environment, MB blocks sunlight from penetrating the water body, thereby posing a threat to aquatic life; in terms of public health, MB may burn eyes and cause irreversible damage [4]. Therefore, it is extremely necessary to remove dyes in wastewater. Commonly used methods for removing dyes include biosorption, electrocoagulation, redox, photocatalysis, etc. [5][6][7][8][9]. Photoelectrocatalysis (PEC) is an electrochemical advanced oxidation process (EAOP) combining photocatalysis with electrolysis [10]. Considering the renewable source of solar energy, research on photocatalysis has been rapidly expanding in recent years [11]. However, compared with photocatalysis, PEC can effectively solve the problem of the recombination of photogenerated electronhole pairs in photocatalysis, so it has been widely used in the removal of pollutants in wastewater in recent years [12,13].
In 1972, Fujishima and Honda et al. [14] found that water could be decomposed into H 2 and O 2 by the TiO 2 photoelectrode under light conditions. Since then, TiO 2 has been thought to be an ideal semiconductor close to photocatalysis due to its high stability, low cost, harmlessness, high electron mobility, and excellent photoactivity [15][16][17]. However, TiO 2 has a wide band gap (3.0~3.2 eV) [18]. Therefore, it is slightly difficult to use sunlight as an external light source for catalytic titanium dioxide. This limitation can be overcome by changing the form of titanium dioxide or doping with other materials. TiO 2 nanotubes have been proven to have great research potential in photoelectric decomposition, Processes 2021, 9, 1492 3 of 12 Instruments: Desktop high-speed centrifuge (Shanghai Luxiang Instrument Co., Ltd., Harbin, China), Raman spectrometer (HORIBA Jobin Yvon, Harbin, China).

Synthesis of GO
Graphene oxide (GO) was prepared by a modified Hummers method. Firstly, graphite powder (1 g) was treated with 80 mL H 2 SO 4 , 1 g K 2 S 2 O 4 , 1 g P 2 O 5 and slowly heated to 80 • C for 5 h. Then, a large amount of deionized water was used to wash repeatedly until it was neutral to obtain pre-graphite oxide, and the pre-oxidized graphite was dried at 6 • C for 12 h. A mixture of 40 mL H 2 SO 4 and graphite was added to the ice water bath in the flask. Secondly, 4 g KMnO 4 was slowly added to the solution, heated to 35 • C to react for 2 h. Then, deionized water was added and heated to about 98 • C. After 15 min, deionized water and 10 mL H 2 O 2 were added and then centrifuged when the color of the solution gradually changed into golden yellow. The obtained solid was washed with dilute hydrochloric acid and deionized water to remove residual acid on the surface. Finally, the GO sample was dried by vacuum at 60 • C for 12 h.

Synthesis of RGO
RGO was prepared by the chemical reduction method. Graphene oxide (GO) was added to deionized water and ultrasonically treated for 1 h. Ammonia was added and the pH was adjusted to 10. A total of 250 mL hydrazine hydrate (35%) was added to the treated GO and heated at 95 • C for 4 h. The black solution was filtered, followed by washing with dilute hydrochloric acid and deionized water to remove residual acid on the surface. RGO was obtained by vacuum drying at 70 • C for 12 h.

Fabrication Method of RGO/TiO 2 NTs
In this study, a TiO 2 NT photoelectrode grown on Ti foils was prepared by the anodic oxidation method, and the synthesis progress and experiment parameters are provided in our previous papers [36]. The RGO/TiO 2 NTs photoelectrode was prepared by electrochemical deposition. The electrodeposition solution was a mixture of 100 mg·L −1 RGO solution and 0.5 mol·L −1 Na 2 SO 4 . A three-electrode system was adopted with the TiO 2 NTs electrode as the anode, Pt as the cathode, and SEC as the reference electrode. Electrodeposition was conducted at a certain voltage for a period of time. After the reaction was completed, the sample was repeatedly rinsed with deionized water and then vacuum dried at 70 • C for 4 h. Finally, the RGO/TiO 2 NTs was obtained.

Photoelectric Catalytic Activity Tests
The photoelectric catalytic property of RGO/TiO 2 NT photoelectrode was evaluated by photoelectric catalytic oxidation degradation of 5 mg/L methylene blue aqueous solution. At room temperature, 80 mL methylene blue solution was added to a 100 mL quartz beaker. The mercury lamp was used as the external light source and placed parallel to the Ti sheet, with an interval of 10 cm. The 35 W external light source and the 15 V DC regulated power supply were turned on. At regular intervals of 30 min, a 5 mL sample was taken and the absorbance was checked at 664 nm through a UV-vis spectrophotometer. The degradation efficiency (η) was calculated using the following equation: where C 0 is the initial concentration of MB (mg/L) and C t is the instantaneous concentrations (mg/L) in the solution.

Response Surface Experimental Design
RSM was used to regulate and control the photoelectric catalytic performance of RGO/TiO 2 NTs photoelectrodes. Box-Behnken with the software Design Expert 8.0 was employed to evaluate the combined effects of the three independent variables: deposition potential, deposition time, and deposition temperature. The coded and actual values of the three independent variables together with the responses are shown in Table 1.

Analysis of Single Factor Test Results
The RGO/TiO 2 NTs photoelectrode was prepared by depositing for 10 min with different deposition potentials and then used for photoelectric catalytic degradation of MB, as shown in Figure 1. It was found that the catalytic effect of the photoelectrode on MB was significantly improved, showing a trend of first increasing and then decreasing. When the voltage was 1.2 V, the photoelectrode had the best degradation effect on MB, and the degradation rate reached 65.7%. This may be due to the high deposition potential and to a large amount of RGO deposited on the surface of the nanotubes, which reduces the light absorption performance (Figure 1).
When the deposition potential is 1.2 V, the photoelectric catalytic degradation effect of RGO/TiO 2 NTs photoelectrode on MB is studied under different times. It can be seen from Figure 2 that when the deposition time is 10 min, the photoelectric catalytic degradation effect is the best, and the MB degradation rate reaches 66.1%. However, a longer deposition time causes too much RGO to accumulate on the surface of the nanotubes, which reduces the light absorption performance, resulting in a decrease in the degradation rate of MB ( Figure 2). When the deposition potential is 1.2 V, the photoelectric catalytic degradation effect of RGO/TiO2 NTs photoelectrode on MB is studied under different times. It can be seen from Figure 2 that when the deposition time is 10 min, the photoelectric catalytic degradation effect is the best, and the MB degradation rate reaches 66.1%. However, a longer deposition time causes too much RGO to accumulate on the surface of the nanotubes, which reduces the light absorption performance, resulting in a decrease in the degradation rate of MB ( Figure 2). When the deposition potential was 1.2 V and the deposition time was 10 min, the photoelectrodes were prepared under the conditions of different deposition temperatures. As shown in Figure 3, the photoelectric catalytic performance of the photoelectrode is proportional to the deposition temperature. When the deposition temperature was 25 °C, the photoelectric catalytic performance was at its best, and the degradation rate of MB reached 66.8%. However, when the temperature rises further, the degradation of MB decreases. This may be due to a large amount of RGO accumulated on the surface of the nanotubes caused by the high temperature, which affects the photocatalytic activity ( Figure 3).   When the deposition potential is 1.2 V, the photoelectric catalytic degradation effect of RGO/TiO2 NTs photoelectrode on MB is studied under different times. It can be seen from Figure 2 that when the deposition time is 10 min, the photoelectric catalytic degradation effect is the best, and the MB degradation rate reaches 66.1%. However, a longer deposition time causes too much RGO to accumulate on the surface of the nanotubes, which reduces the light absorption performance, resulting in a decrease in the degradation rate of MB ( Figure 2). When the deposition potential was 1.2 V and the deposition time was 10 min, the photoelectrodes were prepared under the conditions of different deposition temperatures. As shown in Figure 3, the photoelectric catalytic performance of the photoelectrode is proportional to the deposition temperature. When the deposition temperature was 25 °C, the photoelectric catalytic performance was at its best, and the degradation rate of MB reached 66.8%. However, when the temperature rises further, the degradation of MB decreases. This may be due to a large amount of RGO accumulated on the surface of the nanotubes caused by the high temperature, which affects the photocatalytic activity ( Figure 3).  When the deposition potential was 1.2 V and the deposition time was 10 min, the photoelectrodes were prepared under the conditions of different deposition temperatures. As shown in Figure 3, the photoelectric catalytic performance of the photoelectrode is proportional to the deposition temperature. When the deposition temperature was 25 • C, the photoelectric catalytic performance was at its best, and the degradation rate of MB reached 66.8%. However, when the temperature rises further, the degradation of MB decreases. This may be due to a large amount of RGO accumulated on the surface of the nanotubes caused by the high temperature, which affects the photocatalytic activity ( Figure 3).

Optimization Conditions of RGO/TiO2 NTs Production
In the Equation (1)

Optimization Conditions of RGO/TiO 2 NTs Production
In the Equation (1), Y is the degradation rate of MB (%), and A, B, C are the code values of deposition potential, deposition time, and deposition temperature. The positive and negative signs before each item represent synergistic and antagonistic effects. The model was evaluated based on the correlation coefficient R 2 The R 2 of the model is 0.9761, which is close to 1, indicating that the predicted value of the model was similar to the measured value. Table 1 showed the initial conditions of fermentation and the actual and predicted responses.
In the ANOVA results of RSM, shown in Table 2, those model terms with a "p-value" less than 0.05 were significant ( Table 2). The model p-value was less than 0.0001, which was extremely significant. The lack of fit was 0.0762 (>0.05), which was not significant, indicating that the model fit well within the regression area. As the F-values in Table 2 show, the order of the impact of three conditions on RGO/TiO 2 NTs production was: B > A > C. Moreover, the error statistical analysis of the regression equation showed that the R 2 value was 0.9454 and R adj 2 was 0.6900, indicating that the 69% change in response value could be explained by the model. The precision was 15.193, more than 4, and proved reasonable; the CV was 2.64%, less than 10%, indicating the high reliability and accuracy of the test. The analysis results showed that the regression equation model was in good agreement with the actual test, which can effectively predict the test results. It is recommended to use this model to predict the optimal conditions for the preparation of RGO/TiO 2 NTs. The regression equation can be effectively represented by the response surface threedimensional graph. The relationship between the response value of each variable and the test value is shown in Figure 4, which further evaluates the relationship between each variable and the best conditions. According to the analysis of Design Expert 8.0 software, the best preparation conditions for loading RGO onto TiO 2 NTs photoelectrodes were: deposition potential at 1.19 V, deposition time of 10.27 min, and deposition temperature at 24.94 • C, and the predicted value of MB degradation rate was 65.43%. Considering the actual situation, the optimal preparation conditions were identified as: deposition potential at 1.1 V, deposition time of 10 min, and deposition temperature at 25 • C (Figure 4). test value is shown in Figure 4, which further evaluates the relationship between each variable and the best conditions. According to the analysis of Design Expert 8.0 software, the best preparation conditions for loading RGO onto TiO2 NTs photoelectrodes were: deposition potential at 1.19 V, deposition time of 10.27 min, and deposition temperature at 24.94 °C, and the predicted value of MB degradation rate was 65.43%. Considering the actual situation, the optimal preparation conditions were identified as: deposition potential at 1.1 V, deposition time of 10 min, and deposition temperature at 25 °C (Figure 4).    Figure 6a is the XRD diffractogram of the photoelectrode prepared under the optimal preparation conditions. Cu Kα rays (λ was 0.15418 nm), the characteristic diffraction peak of anatase phase A (101), appeared when 2θ = 25.3°, and the characteristic diffraction peak of rutile phase R (101) appeared when 2θ = 27.4°. It can be seen from the figure that there were three crystal phases at the same time, and the diffraction peak position was the same  test value is shown in Figure 4, which further evaluates the relationship between each variable and the best conditions. According to the analysis of Design Expert 8.0 software, the best preparation conditions for loading RGO onto TiO2 NTs photoelectrodes were: deposition potential at 1.19 V, deposition time of 10.27 min, and deposition temperature at 24.94 °C, and the predicted value of MB degradation rate was 65.43%. Considering the actual situation, the optimal preparation conditions were identified as: deposition potential at 1.1 V, deposition time of 10 min, and deposition temperature at 25 °C (Figure 4).  NTs photoelectrode was covered with a sheet of RGO film, which covered most of the nanotube surface. Part of the wall of the RGO/TiO2 nanotube became thicker: measured by nano measurer 1.2.5, the wall was 22.60 nm. Therefore, the RGO/TiO2 NTs photoelectrode had a better photoelectrocatalytic performance with RGO than without it ( Figure 5).  Figure 6a is the XRD diffractogram of the photoelectrode prepared under the optimal preparation conditions. Cu Kα rays (λ was 0.15418 nm), the characteristic diffraction peak of anatase phase A (101), appeared when 2θ = 25.3°, and the characteristic diffraction peak of rutile phase R (101) appeared when 2θ = 27.4°. It can be seen from the figure that there were three crystal phases at the same time, and the diffraction peak position was the same  Figure 6a is the XRD diffractogram of the photoelectrode prepared under the optimal preparation conditions. Cu Kα rays (λ was 0.15418 nm), the characteristic diffraction peak of anatase phase A (101), appeared when 2θ = 25.3 • , and the characteristic diffraction peak of rutile phase R (101) appeared when 2θ = 27.4 • . It can be seen from the figure that there were three crystal phases at the same time, and the diffraction peak position was the same as that of TiO 2 NTs. After RGO modification, the characteristic diffraction peak of graphene (002) appeared at 2θ = 24.9 • , indicating that RGO had been successfully modified [37]. In order to further test whether RGO was successfully modified on the surface of the TiO 2 NTs photoelectrode, Raman spectroscopy was used. In the Raman spectrum of TiO 2 , the Raman peaks at 148, 198, 397, 515, and 638 cm −1 correspond to the anatase-induced A1g +2 B1g +3 Eg mode [38]. As shown in Figure 6b, the four Raman peaks caused by RGO are located at 1365, 1585, 2749, and 2938 cm −1 . This was caused by the D band related to the defects in the hexagonal graphite sheet and the G band related to the E2g mode. In addition, the G band was also related to the sp2 hybridized carbon atoms [39]. Therefore, it is further proved that RGO has been successfully modified on TiO 2 NTs (Figure 6).
the Raman peaks at 148, 198, 397, 515, and 638 cm −1 correspond to the anatase-induced A1g +2 B1g +3 Eg mode [38]. As shown in Figure 6b, the four Raman peaks caused by RGO are located at 1365, 1585, 2749, and 2938 cm −1 . This was caused by the D band related to the defects in the hexagonal graphite sheet and the G band related to the E2g mode. In addition, the G band was also related to the sp2 hybridized carbon atoms [39]. Therefore, it is further proved that RGO has been successfully modified on TiO2 NTs (Figure 6). In order to study the performance of RGO/TiO2 NT photoelectrodes prepared under the best conditions, the photoelectrocatalytic MB (80 mL, 5 mg/L) degradation test was carried out under three different conditions, and the reaction time was 120 min. As shown in Figure 7, when the electrode was illuminated by a 35 W mercury lamp, the photocatalytic degradation rate of MB was 17.2%; when the electrode was protected from light under a 15 V bias voltage, the photocatalytic degradation efficiency of MB was 32%; when the electrode was irradiated by a 35 W mercury lamp and a 15 V bias voltage was applied, the MB degradation efficiency was significantly increased to 64% (Figure 7). In order to study the performance of RGO/TiO 2 NT photoelectrodes prepared under the best conditions, the photoelectrocatalytic MB (80 mL, 5 mg/L) degradation test was carried out under three different conditions, and the reaction time was 120 min. As shown in Figure 7, when the electrode was illuminated by a 35 W mercury lamp, the photocatalytic degradation rate of MB was 17.2%; when the electrode was protected from light under a 15 V bias voltage, the photocatalytic degradation efficiency of MB was 32%; when the electrode was irradiated by a 35 W mercury lamp and a 15 V bias voltage was applied, the MB degradation efficiency was significantly increased to 64% (Figure 7).
it is further proved that RGO has been successfully modified on TiO2 NTs (Figur In order to study the performance of RGO/TiO2 NT photoelectrodes prepare the best conditions, the photoelectrocatalytic MB (80 mL, 5 mg/L) degradation carried out under three different conditions, and the reaction time was 120 min. A in Figure 7, when the electrode was illuminated by a 35 W mercury lamp, the ph lytic degradation rate of MB was 17.2%; when the electrode was protected from der a 15 V bias voltage, the photocatalytic degradation efficiency of MB was 32% the electrode was irradiated by a 35 W mercury lamp and a 15 V bias voltage was the MB degradation efficiency was significantly increased to 64% (Figure 7). It can be seen from Figure 8 that the MB degradation rate of TiO 2 NTs photoelectrode under mercury lamp irradiation is 54%, and the MB degradation rate under the catalysis of RGO/TiO 2 NTs photoelectrode is 63%, indicating that the degradation efficiency of the RGO modified TiO 2 NTs photoelectrode has been significantly improved (Figure 8). Processes 2021, 9, x FOR PEER REVIEW of RGO/TiO2 NTs photoelectrode is 63%, indicating that the degradation efficienc RGO modified TiO2 NTs photoelectrode has been significantly improved ( Figure   Figure 8. Photocatalytic degradation of methylene blue on RGO/TiO2 NTs and TiO2 NT p lectrode.

Analysis of Photoelectrocatalytic Mechanism
In the process of RGO/TiO2 NTs photoelectrode photoelectrocatalysis of MB tive intermediates participating in the reaction mainly include ·OH, h+ and ·O2 − T the catalytic mechanism of RGO/TiO2 NTs photoelectrode to MB, we conducted f ical scavenging experiments. It can be seen from Figure 9 that ·OH and h+ are th active substances in the photoelectric catalytic reaction. In order to study the stability of RGO/TiO2 NTs photoelectrode photoelectroca a series of cyclic experiments were carried out. The result, as shown in Figure 10, w the RGO/TiO2 NTs photoelectrode has excellent stability: after repeated use for fiv the degradation efficiency of MB was still above 58%. Figure 11 shows the catalytic activity mechanism of RGO/TiO2 NTs photoele When light is irradiated, electrons in the valence band of TiO2 are excited to trans the conduction band, leaving a hole in the valence band. At the same time, electron valence band of RGO are excited to transition to the conduction band, also leavin in the valence band. Due to its good conductivity of RGO, it can promote the flow togenerated electrons from the conduction band of TiO2 to the conduction band o These electrons can react with O2 to produce ·O2 − . In addition, the holes in the con

Analysis of Photoelectrocatalytic Mechanism
In the process of RGO/TiO 2 NTs photoelectrode photoelectrocatalysis of MB, the active intermediates participating in the reaction mainly include ·OH, h+ and ·O 2 − To study the catalytic mechanism of RGO/TiO 2 NTs photoelectrode to MB, we conducted free radical scavenging experiments. It can be seen from Figure 9 that ·OH and h+ are the main active substances in the photoelectric catalytic reaction.

Analysis of Photoelectrocatalytic Mechanism
In the process of RGO/TiO2 NTs photoelectrode photoelectrocatalysis of MB tive intermediates participating in the reaction mainly include ·OH, h+ and ·O2 − T the catalytic mechanism of RGO/TiO2 NTs photoelectrode to MB, we conducted f ical scavenging experiments. It can be seen from Figure 9 that ·OH and h+ are t active substances in the photoelectric catalytic reaction. In order to study the stability of RGO/TiO2 NTs photoelectrode photoelectroc a series of cyclic experiments were carried out. The result, as shown in Figure 10, the RGO/TiO2 NTs photoelectrode has excellent stability: after repeated use for fiv the degradation efficiency of MB was still above 58%. Figure 11 shows the catalytic activity mechanism of RGO/TiO2 NTs photoele When light is irradiated, electrons in the valence band of TiO2 are excited to tran the conduction band, leaving a hole in the valence band. At the same time, electro valence band of RGO are excited to transition to the conduction band, also leavi in the valence band. Due to its good conductivity of RGO, it can promote the flow togenerated electrons from the conduction band of TiO2 to the conduction band These electrons can react with O2 to produce ·O2 − . In addition, the holes in the con band of TiO2 and RGO can react with H2O to generate ·OH. In this way, elect transferred to the conduction band of RGO to react with O2, and holes react w effectively inhibiting the recombination of electrons and holes. Moreover, the In order to study the stability of RGO/TiO 2 NTs photoelectrode photoelectrocatalysis, a series of cyclic experiments were carried out. The result, as shown in Figure 10, was that the RGO/TiO 2 NTs photoelectrode has excellent stability: after repeated use for five times, the degradation efficiency of MB was still above 58%.
voltage accelerates the migration rate of photo-generated electrons in the system greatly improves the photocatalytic effect. In addition, this method not only has e degradation effects but also has a simple process and is easy to characterize co with photocatalytic oxidation reactions.

Conclusions
Through single factor experiments, the optimal preparation conditions for RG NTs were found to be: deposition voltage at 1.1 V, deposition time of 10 min, and tion temperature at 25 °C. According to the analysis and optimization of the respo face experiment design, the significance of the catalytic activity of RGO/TiO2 NT electrode was in the order of deposition time > deposition voltage > deposition t ture. The optimal preparation conditions obtained by optimization were: deposit age at 1.19 V, deposition time of 10.27 min, and deposition temperature at 24.94 predicted value of MB degradation rate was 65.43%. Compared with the actu (63%), the error value was 2.43%, indicating that the preparation parameters op by RSM have practical application value. The results provide a reference for th scale application of RGO/TiO2 NTs photoelectrodes in engineering. Through S XRD characterization analysis, it can be seen that after RGO modification, the tubular structure of TiO2 nanotubes was not changed. The surface of the TiO2 NT electrode was covered with a non-uniform flake-shaped reduced graphene oxide f impedance of RGO/TiO2 NTs was smaller than that of TiO2 NTs photoelectrode.   Figure 10. Stability study on RGO/TiO 2 NT photoelectrode. Figure 11 shows the catalytic activity mechanism of RGO/TiO 2 NTs photoelectrodes. When light is irradiated, electrons in the valence band of TiO 2 are excited to transition to the conduction band, leaving a hole in the valence band. At the same time, electrons in the valence band of RGO are excited to transition to the conduction band, also leaving holes in the valence band. Due to its good conductivity of RGO, it can promote the flow of photogenerated electrons from the conduction band of TiO 2 to the conduction band of RGO. These electrons can react with O 2 to produce ·O 2 − . In addition, the holes in the conduction band of TiO 2 and RGO can react with H 2 O to generate ·OH. In this way, electrons are transferred to the conduction band of RGO to react with O 2 , and holes react with H 2 O, effectively inhibiting the recombination of electrons and holes. Moreover, the applied voltage accelerates the migration rate of photo-generated electrons in the system, which greatly improves the photocatalytic effect. In addition, this method not only has excellent degradation effects but also has a simple process and is easy to characterize compared with photocatalytic oxidation reactions. ses 2021, 9, x FOR PEER REVIEW 10 of voltage accelerates the migration rate of photo-generated electrons in the system, whic greatly improves the photocatalytic effect. In addition, this method not only has excellen degradation effects but also has a simple process and is easy to characterize compare with photocatalytic oxidation reactions.

Conclusions
Through single factor experiments, the optimal preparation conditions for RGO/TiO NTs were found to be: deposition voltage at 1.1 V, deposition time of 10 min, and depos tion temperature at 25 °C. According to the analysis and optimization of the response su face experiment design, the significance of the catalytic activity of RGO/TiO2 NTs photo electrode was in the order of deposition time > deposition voltage > deposition temper ture. The optimal preparation conditions obtained by optimization were: deposition vol age at 1.19 V, deposition time of 10.27 min, and deposition temperature at 24.94 °C. Th   Figure 11. Photoelectric catalytic mechanism diagram.

Conclusions
Through single factor experiments, the optimal preparation conditions for RGO/TiO 2 NTs were found to be: deposition voltage at 1.1 V, deposition time of 10 min, and deposition temperature at 25 • C. According to the analysis and optimization of the response surface experiment design, the significance of the catalytic activity of RGO/TiO 2 NTs photoelectrode was in the order of deposition time > deposition voltage > deposition temperature. The optimal preparation conditions obtained by optimization were: deposition voltage at 1.19 V, deposition time of 10.27 min, and deposition temperature at 24.94 • C. The predicted value of MB degradation rate was 65.43%. Compared with the actual value (63%), the error value was 2.43%, indicating that the preparation parameters optimized by RSM have practical application value. The results provide a reference for the large-scale application of RGO/TiO 2 NTs photoelectrodes in engineering. Through SEM and XRD characterization analysis, it can be seen that after RGO modification, the original tubular structure of TiO 2 nanotubes was not changed. The surface of the TiO 2 NTs photoelectrode was covered with a non-uniform flake-shaped reduced graphene oxide film. The impedance of RGO/TiO 2 NTs was smaller than that of TiO 2 NTs photoelectrode.