Phase junction enhanced photocatalytic activity of Ga2O3 nanorod arrays on flexible glass fiber fabric

Ga2O3 nanostructures hold great potential applications in photocatalytic fields due to their stability, high efficiency and environmental friendliness. The construction of phase junction has been proved to be one of the most effective strategies for enhancing Ga2O3 photocatalytic activity. However, the influence of the formation process at the interface of the phase junction on the photocatalytic activity of Ga2O3 nanostructures is far less well understood. In this work, for the first time, large-area Ga2O3 nanorod arrays (NRAs) with controllable α/β phase junction were prepared in situ on a flexible glass fiber fabric by a facile and environmentally friendly three-step method. The α/β-Ga2O3 phase junction NRAs exhibit an ultra-high photocatalytic degradation rate of 97% during Ultraviolet (UV) irradiation for 60 min, which is attributed to a unique phase junction promoting efficient charge separation. However, the photocatalytic activity of α/β-Ga2O3 phase junction NRAs is not evident in the early phase transition, possibly due to the presence of defects acting as charge recombination centers.


Introduction
With the continuous development of the social economy, environmental pollution has become an increasingly serious problem, prompting humans to continuously explore new solutions. [1][2][3][4][5][6] Photocatalytic reaction, as a simple, efficient and cost-effective method, has promising applications in the removal of environmental pollutants and attracted wide attention. [7][8][9][10][11][12][13] Recently, various metal oxides with d 10 (In 3+ , Ga 3+ , Ge 4+ , Sn 4+ ) congurations have been reported as effective photocatalysts for photodegradation of various organic pollutants. 14,15 Ga 2 O 3 is a typical representative among them. 16 With a wide bandgap (4.2-4.9 eV) and excellent physical and chemical properties, Ga 2 O 3 is recognized as one of the most promising semiconductors of this century. [17][18][19][20][21][22] It has extensively been applied to power devices, 23-25 solar-blind ultraviolet (UV) photodetectors, [26][27][28][29] gas sensors, 30 solar cells 31 and photocatalysis. 32,33 For photocatalysis applications, related studies claim that Ga 2 O 3 can theoretically exhibit better and more stable photocatalytic activity than commercial TiO 2 and realize the degradation of refractory pollutants. 34,35 This is attributed to the extraordinary redox capability of photogenerated electron-hole pairs. 1,36,37 Furthermore, Ga 2 O 3 is also widely accepted as an environmentally friendly material with low cost and high chemical stability. 38 Many methods have been investigated to further improve the photocatalytic activity of Ga 2 O 3 , including morphology controlling, doping, surface modication and semiconductor coupling. [39][40][41][42][43][44] Nitu Syed et al. reported a two-step method for the synthesis of porous a-Ga 2 O 3 nanosheets from liquid metal gallium, explaining that the excellent photocatalytic activity of a-Ga 2 O 3 originated from the narrowed bandgap caused by trap states. 1 Han et al. revealed that the modication of in situ Ag nanoparticles can effectively improve the photocatalytic property of Ga 2 O 3 for hydrogen evolution. 39 Zhang et al. incorporated solvothermally synthesized Ga 2 O 3 nanoparticles into liquid metal/metal oxide frameworks to form enhanced photocatalytic systems. 41 Xu et al. fabricated twodimensional TiO 2 -Ga 2 O 3 p-n heterostructures, demonstrating the contribution of heterostructures in enhancing photocatalytic activity. 45 Furthermore, the construction of appropriate phase junction structure in Ga 2 O 3 can also signicantly enhance photocatalytic activity. [46][47][48] Liu et al. demonstrated that the mesopores and heterojunction in the mixed-phase Ga 2 O 3 are responsible for enhancing photocatalytic activity. 7 However, the inuence of the formation process at the interface of the phase junction on the photocatalytic activity of Ga 2 O 3 nanostructures has not been fully understood. For example, phase transformation is a process from the surface to the bulk, and different thicknesses of phase interface may result in various photocatalytic activities. 49 Therefore, an in-depth understanding of junction-related issues will aid in the design and preparation of efficient Ga 2 O 3 nanostructured photocatalysts. On the other hand, almost all of the reported Ga 2 O 3 nanostructured photocatalysts are currently applied in suspension systems. The disadvantages of photocatalysts, such as agglomeration, inadequate illumination and difficulty in recovery, restrict their large-scale practical applications. Glass ber fabric as a support for in situ growth of Ga 2 O 3 nanostructures is expected to effectively overcome this difficulty. 18 To the best of our knowledge, there are no reports of in situ preparation of Ga 2 O 3 nanostructures on glass ber fabric for the application of photocatalytic degradation.
Herein, for the rst time, we reported a facile and environmentally friendly three-step method for in situ preparation of large-area Ga 2 O 3 nanorod arrays (NRAs) with controllable a/ b phase junction on a exible glass ber fabric. The as-prepared a/ b-Ga 2 O 3 phase junction NRAs exhibited excellent photocatalytic activity for the degradation of Rhodamine B (RhB) aqueous solution. In addition, the mechanism of photocatalytic activity enhancement was discussed and compared with related literature.

Sample preparation
The preparation of Ga 2 O 3 NRAs involves the following three steps. In the rst step, a SnO 2 thin lm was fabricated by radio frequency magnetron sputtering on the surface of the cleaned glass ber fabric, which was used as a growth seed layer of Ga 2 O 3 . The growth temperature and Ar gas pressure were xed at 550 C and 0.8 Pa, respectively. The second step is to prepare GaOOH nanorod precursor by hydrothermal method. Here, 0.20 g of Ga(NO 3 ) 3 -$nH 2 O was dissolved in 30 mL of DI water to prepare a growth solution. Then the substrates glass ber fabric completed in the rst step was placed in the growth solution and transferred separately into a 50 mL Teon-lined stainless steel autoclave for hydrothermal treatment at 150 C for 12 h. Aer the solution was naturally cooled down to room temperature, the precipitates were ltered and washed with DI water, then dried in air at 80 C for 2 h to obtain GaOOH NRAs precursors. The last step, the as-prepared precursors were annealed at 400 C for 4 h in air to obtain a-Ga 2 O 3 NRAs. The detailed synthesis is schematically demonstrated in Fig. 1. Further, other samples were obtained by annealing a-Ga 2 O 3 NRAs in air at 700 C for different times from 20 min to 120 min.

Characterization
The crystal structure of samples was analyzed by a Bruker D8 DISCOVER X-ray diffractometer (XRD). UV-Raman spectra were recorded on a Jobin-Yvon T64000 triple-stage spectrograph with spectral resolution of 2 cm À1 . The thermal behavior of the GaOOH nanorod was investigated by thermal gravimetric analyzer (Pyris1 TGA). For the morphological and microstructural analysis, a Hitachi S-4800 eld-emission scanning electron microscope (SEM) equipped and a JEOL JEM-2100 transmission electron microscopy (TEM) were utilized. The ultraviolet-visible (UV-vis) absorption spectra were taken using a Hitachi U-3900 UV-vis spectrophotometer. The chemical composition of samples was characterized by a Thermo Scientic K-Alpha X-ray photoelectron spectroscopy (XPS).

Photocatalytic experiments
In this experiment, the glass ber fabric with Ga 2 O 3 NRAs were dropped into 50 mL of RhB aqueous solution (2 Â 10 À5 M) and placed in the dark for 30 min to ensure adsorption-desorption equilibrium was reached. Then irradiated reaction solution with a 10 W UV light lamp (l ¼ 254 nm). The light intensity of the UV lamp was always maintained at 1.0 mW cm À2 . During the process, about 3 mL of solution was withdrawn from the reaction system at a given time interval (10 min) for absorbance testing by UV-vis spectrophotometry.

Mott-Schottky measurement
For Mott-Schottky measurements, 5 mg Ga 2 O 3 NRAs powder was scraped from the glass ber fabric and dispersed in 2 mL of absolute ethanol, followed by the addition of 20 mL of 0.5% Naon. Aer the mixed solution was sonicated for 1 h, 0.5 mL was transferred onto a FTO conductive glass. The resulting electrodes were dried in air and further heated at 150 C for 1 h under a N 2 gas ow. The electrochemical measurements were performed in a three-electrode conguration system using a CHI 760E electrochemical workstation (CH Instruments, China), including the as-prepared FTO working electrodes (with an active area of 1.0 cm 2 ), Pt foil as the counter electrode and saturated calomel electrode (SCE) as the reference electrode. 0.5 M Na 2 SO 4 aqueous solution was used as the electrolyte. Fig. 2(a) shows the XRD patterns of as-synthesized GaOOH and a-Ga 2 O 3 NRAs. All the peaks can be indexed to the orthorhombic GaOOH phase (JCPDS no. 06-0180) except the diffraction peak of the SnO 2 seed layer. Aer annealing GaOOH at 400 C for 4 h, the observed diffraction peaks occur at new locations, indicating that the GaOOH is completely converted into a-Ga 2 O 3 of the corundum structure (JCPDS no. 06-0503). 50 The phase transition of a-Ga 2 O 3 at 700 C for various times was also analyzed by XRD, and the corresponding results are shown in Fig. 2(b). With a-Ga 2 O 3 annealed at 700 C for 30 min, a diffraction peak corresponding to the (111) plane attributed to monoclinic b-Ga 2 O 3 is detected, and it becomes stronger with the further increase of annealing time. Ga 2 O 3 with different phase structures can be obtained during annealing for 30-90 min. No diffraction peak assigned to a-Ga 2 O 3 is observed through annealing for 120 min, suggesting that the a-Ga 2 O 3 is totally transformed into b-Ga 2 O 3 at this point.

Results and discussion
UV Raman spectroscopy was also used to monitor the a to b phase transformation of Ga 2 O 3 . As shown in Fig. 2(c), it is worth noting that the typical characteristic Raman bands of b-Ga 2 O 3 at 198 cm À1 and 414 cm À1 can be clearly observed aer annealing a-Ga 2 O 3 at 700 C for 20 min, in addition to the The TG/DTG curve of the as-prepared GaOOH NRAs precursor by heating from room temperature to 800 C in air atmosphere is shown in Fig. 2(d). A major weight loss of 11.2% can be noticed in the temperature range of approximately 260-480 C, with the fastest weight loss rate occurring at 438 C, which is attributed to transformation of GaOOH into a-Ga 2 O 3 by thermal dehydration. A weak weight loss of 2% is also noted at the range of 500-630 C, indicating the conversion of a-Ga 2 O 3 to b-Ga 2 O 3 . With further prolonged heating up to 800 C, there is no weight loss.
A typical SEM image of as-synthesized a-Ga 2 O 3 NRAs, as presented in Fig. 3(a), which reveals the uniform and dense growth of the sample on each ber rod. High-magnication SEM images of a-Ga 2 O 3 NRAs and other Ga 2 O 3 NRAs obtained by annealing (Ga 2 O 3 -60 and b-Ga 2 O 3 ) are also shown in Fig. 3(bd), respectively. Further revealing that the diameter of all nanorods ranged from 100 to 400 nm and the tips are all diamond-shaped. The annealing process has not signicantly changed the morphology of Ga 2 O 3 NRAs. 51 Fig. S1. † The decrease of the characteristic peak at 554 nm during illumination suggests RhB decomposition. 1,43 Fig. 4(a) and S2 † reveals the comparison of photocatalytic activities of different Ga 2 O 3 NRAs. Among them, the Ga 2 O 3 -60 NRAs exhibits the best photocatalytic with a degradation rate of 97%, which can be attributed to the a/b-Ga 2 O 3 phase junction promoting the separation of photogenerated electrons and holes. 47 Furthermore, the photocatalytic degradation process of these Ga 2 O 3 NRAs were tted using the rst-order kinetic curve according to the Langmuir-Hinshelwood model. 32,52 As shown in Fig. 4(b), the value of the reaction rate constant (K) are estimated to be 0.0232, 0.0301, 0.0265, 0.0245, 0.0589, 0.0546 and 0.0418 min À1 , corresponding to the a-Ga (ethylenediamine tetraacetic acid, h + trapping agents), IPA (isopropyl alcohol, cOH trapping agents) and BQ (benzoquinone, cO 2 À trapping agents), respectively. As shown in Fig. 4(c), the photocatalytic activity of the Ga 2 O 3 -60 NRAs is affected slightly with the addition of EDTA, indicating that h + is not the main factor in this system. In contrast, the introduction of IPA or BQ greatly suppressed the photocatalytic activity of the Ga 2 O 3 -60 NRAs, indicating that cOH and cO 2 À acted as dominating reactive species in the reaction system. Moreover, the cycling stability of the Ga 2 O 3 -60 NRAs was evaluated by conducting ve consecutive cycle degradation experiments. As shown in Fig. 4(d), the degradation ratio of RhB is not obviously reduced during the repeated experiments, indicating the remarkable stability of the Ga 2 O 3 -60 NRAs. XRD patterns (Fig. S3 †) also indicates that no structural difference can be observed between the Ga 2 O 3 -60 NRAs before and aer photocatalytic degradation of RhB solution.
For the proposed photocatalytic degradation mechanism of the system, the band structures of a-Ga 2 O 3 and b-Ga 2 O 3 were characterized by Mott-Schottky measurements and XPS. As shown in Fig. 5(a and b), the at band potential of a-Ga 2 O 3 is calculated to be À1.26 eV (vs. SCE), which is more negative than the À0.96 eV (vs. SCE) of b-Ga 2 O 3 , and the valence band potential of b-Ga 2 O 3 is 3.05 eV, which is more positive than the 2.92 eV of a-Ga 2 O 3 . Further combined with the band gap of Ga 2 O 3 reported in our previous work, 51 a schematic illustration of photocatalytic reaction process and charge separation transfer of a/b-Ga 2 O 3 phase junction under UV light irradiation is shown in Fig. 5(c). Under UV light irradiation, the internal electric eld of the a/b-Ga 2 O 3 phase junction could drive the photogenerated charge transfer, promoting the photogenerated electrons transfer from the a phase to the b phase, and the photogenerated holes transfer from the b phase to the a phase. Following that, the photogenerated electrons react with O 2 to generate cO 2 À and the photogenerated holes oxidize OH À to cOH, which together involve in RhB degradation. Efficient charge separation inhibits their recombination, resulting in improved photocatalytic degradation performance. 47,49 In addition, it should be mentioned that phase junction can form on both the surface of Ga 2 O 3 and in the bulk. Although almost no phase junctions were observed on the surface of the Ga 2 O 3 -60 NRAs, they still function as charge separation centers in the bulk. The separated carriers eventually diffuse to the surface of the sample to participate in the photocatalytic reaction. 4 On the other hand, for Ga 2 O 3 as a photocatalytic degradation material, the performance of the b phase is generally better than that of the a phase, 7,43 which is conrmed in Fig. 4(a) of this research. Therefore, in addition to the efficient charge separation due to the phase junction in the bulk, the excellent photocatalytic activity of the Ga 2 O 3 -60 NRAs is also derived from the inherently high activity of the b surface phase. Interestingly, the Ga 2 O 3 -20, Ga 2 O 3 -30 and Ga 2 O 3 -40 NRAs did not exhibit excellent photocatalytic activity despite the formation of phase junctions on various surfaces. The phase transformation of Ga 2 O 3 is a surface-preferred process, which is accompanied by the formation of defects. 54 These defects may become the recombination center of photogenerated electron-hole pairs, reducing the number of efficient carriers on the surface. 55 As  a result, the photocatalytic activity of the initially annealed Ga 2 O 3 phase junction NRAs was not satisfactory.
To obtain more insight into the effect of defects on photocatalytic activity, the Ga 2 O 3 -20 NRAs and Ga 2 O 3 -60 NRAs were selected as typical samples for XPS analysis. As shown in Fig. 6, the O 1s spectra could be divided into two peaks: I and II, representing lattice oxygen ions and oxygen ions in the oxygen vacancies region, respectively. 28 The peak ratio (II/I) of the Ga 2 O 3 -20 NRAs is 1/3, which is higher than that of the Ga 2 O 3 -60 NRAs (1/5), indicating the presence of more oxygen vacancies. Obviously, the Ga 2 O 3 -20 NRAs exhibits poor photocatalytic activity due to the existence of abundant oxygen vacancy defects.
The comparison of the photocatalytic degradation activity of a/b-Ga 2 O 3 phase junction NRAs in this work and other previously reported Ga 2 O 3 related materials is listed in Table 1. Although the comparison of photocatalytic activity is not absolutely reasonable due to the different light source conditions and pollutants used in each experiment, the photocatalytic activity of a/b-Ga 2 O 3 phase junction NRAs in this work is signicantly superior to almost all previous reports on Ga 2 O 3 related materials. This method has realized the large-area growth of a/b-Ga 2 O 3 phase junction NRAs on the exible glass ber fabric and obviously improved its photocatalytic performance, which is of great signicance in the future research in the eld of photocatalysis.

Conclusions
In summary, large-area Ga 2 O 3 NRAs with controllable a/b phase junction were rstly prepared in situ on a exible glass ber fabric by a facile and environmentally friendly three-step method. Photocatalytic degradation experiments showed that the a/b-Ga 2 O 3 phase junction NRAs synthesized by annealing a-Ga 2 O 3 NRAs at 700 C for 60 min exhibited remarkable performance for RhB, with a degradation rate of 97% in 60 min under UV light. The enhanced photocatalytic activity can be attributed to the unique phase junction promoting efficient charge separation and inhibiting the recombination of photogenerated electron-hole pairs. Additionally, the glass ber fabric can realize large-area growth of the Ga 2 O 3 NRAs, effectively solve the trouble of difficult recovery and reuse of photocatalysts, as well as the insufficient absorption of light. A facile environmentally friendly and inexpensive synthesis route will open new avenues for the development of efficient photocatalysts.

Conflicts of interest
There are no conicts to declare.