A Z-scheme BiYO3/g-C3N4 heterojunction photocatalyst for the degradation of organic pollutants under visible light irradiation

Photocatalysis is one of the fascinating fields for the wastewater treatment. In this regard, the present study deals with an effective visible light active BiYO3/g-C3N4 heterojunction nanocomposite photocatalyst with various ratios of BiYO3 and g-C3N4 (1:3, 1:1 and 3:1), synthesised by a wet chemical approach. The as-synthesised nanocomposite photocatalysts were investigated via different physicochemical approaches like Fourier transform infrared (FT-IR), X-ray diffraction (XRD), scanning electron microscopy (SEM), energy dispersive spectroscopy (EDS), transmission electrons microscopy (TEM), UV–vis diffuse reflectance spectroscopy (DRS), photoluminescence (PL) and photoelectrochemical studies to characterise the crystal structure, morphology, optical absorption characteristics and photoelectrochemical properties. The photocatalytic degradation ability of the prepared photocatalytic samples was also analysed through the degradation of RhB in the presence of visible light irradiation. Of all the synthesised photocatalysts, the optimised CB-1 composite showed a significant photocatalytic efficiency (88.7%), with excellent stability and recyclability after three cycles. O2•− and •OH radicals were found to act a major role in the RhB degradation using optimised CB-1 composite, and it possessed ~ 1 times greater photocurrent intensity than the pristine g-C3N4 and BiYO3. In the present work, a direct Z-scheme heterojunction BiYO3/g-C3N4 with a considerably improved photocatalytic performance is reported.


Introduction
Recently, the rapid industrial development led to severe environmental pollution, due to the organic pollutants especially azo dyes discharges from various industries like textile, printing, pharmaceutical, and pesticides, which are very noxious and non-biodegradable in character (Saravanan et al. 2021;Bavani et al. 2021a;Priya et al. 2020a, b;Danish et al. 2020). Hence, the removal of these effluents from the industries is a challenging task to retain the aquatic environment. The most commonly used wastewater treatment technologies have the complications of incomplete removal and the creation of harmful secondary by-products with high-cost implementation (Bavani et al. 2022a, b;Zhao et al. 2019;Malathi et al. 2017).
The photocatalytic technology with semiconductors has been achieved with light irradiation and can be used in the hydrogen evolution, carbon dioxide reduction, organic pollutant degradation, disinfection and so on (Senthil et al. 2019a;Preeyanghaa et al. 2022a;Liu et al. 2021a, b;Danish et al. 2020). But, most of the semiconductor metal oxide photocatalysts exhibit their excellent activity only in the presence of ultraviolet (UV) illumination due to their wide bandgap significantly restricting its large-scale application (Geng et al. 2021;Liu et al 2022;Sun et al. 2020;Wu et al. 2021;Ullah and Dutta 2008;Mondal and Sharma 2014;Tayyab et al. 2022a, b;Liu et al. 2021a, b). Hence, there are rising demands in the improvement of visible-light-driven photocatalysts for highly effective utilisation of solar energy.
From a global perspective, visible light-driven semiconductor photocatalysis is considered to be an optimistic and positive approach to focus on the prevailing environmental and energy issues. Large numbers of photocatalysts with excellent photocatalytic properties have been reported by the researchers (Vinitha et al. 2022). Graphitic carbon nitride (g-C 3 N 4 ) is a polymeric semiconductor with a suitable bandgap energy (E g ) value (2.7-3.0 eV); in addition to this, the preparation of g-C 3 N 4 is made from different inexpensive precursors like urea, melamine, cyanamide and dicyanamide. Among many semiconductor materials, polymeric metal-free g-C 3 N 4 has been comprehensively reconnoitered to address the environmental pollution issues as an efficient, photochemically stable, cost-effective, non-hazardous and eco-friendly photocatalyst. Though, the pure g-C 3 N 4 is not a proficient photocatalyst owing to its rapid reconnection of photo-produced e − /h + pairs. Hence, doping with metals and nonmetals and coupling with different semiconductors were reported to enhance the photocatalytic expertise (Senthil et al. 2017;Liu et al. 2015;Che et al. 2018;Wang et al. 2016). For instance, Sb 2 MoO 6 /g-C 3 N 4 and g-C 3 N 4 /Bi 2 MoO 6 Z-scheme heterojunctions were fabricated, substantially exaggerated the photocatalytic activity on increasing visible light absorption (Zhao et al. 2020;Lv et al. 2015).
Among several other photocatalysts derived from yttrium, BiYO 3 with suitable E g value (~ 2.6 eV) explored their excellent photocatalytic ability in the organic pollutant degradation (Qin et al. 2009;Wongli et al. 2017;Qin et al. 2020;Sirimahachai et al. 2017). Nevertheless, there are various shortcomings that restricted its practical applications. The efficient charge separation and the photocatalytic ability of BiYO 3 could be enhanced by creating heterojunction in photocatalyst (Lian et al. 2021;Preeyanghaa et al. 2022b;Bashir et al. 2022;Priya et al. 2020a).
Since the g-C 3 N 4 possesses certain demerits in the photocatalytic activity, BiYO 3 was selected to fabricate the BiYO 3 /g-C 3 N 4 nanocomposite to create the Z-scheme heterojunction. The BiYO 3 /g-C 3 N 4 Z-scheme heterojunction photocatalyst for the dye pollutant degradation was not yet reported in the literature. Here, we have constructed a BiYO 3 /g-C 3 N 4 heterojunction photocatalyst by using a wet chemical process. The photocatalytic performance of the prepared g-C 3 N 4 and BiYO 3 composite was assessed through the RhB degradation in the presence of visible light exposure. Additionally, to this, suitable photocatalytic degradation mechanism for BiYO 3 /g-C 3 N 4 composite with more relevant scavengers has also been explored.

Synthesis of BiYO 3
A simple co-precipitation technique was used to synthesise the pure BiYO 3 . In this synthesis, 2.5 mmol of Bi(NO 3 ) 3 ·5H 2 O and 2.5 mmol of (Y(NO 3 ) 3 ·6H 2 O were added to a beaker containing 0.1 M HNO 3 following constant stirring for 60 min. Subsequently, 1 M KOH was slowly dripped into the above mixture for altering the pH of the medium to 9.0. Then, the solution was transferred into a Teflon-lined autoclave and maintained at 180 °C for 24 h. Later on, the attained precipitate was filtered and then subjected to washing with water and ethanol several times. Finally, it was dried at 80 °C for 3 h followed by calcination at 750 °C for 2 h.

Synthesis of g-C 3 N 4
The pure g-C 3 N 4 was directly synthesised by the self-condensation of melamine. To accomplish the above process, melamine was directly heated to synthesise the g-C 3 N 4 (CN) powders in a semi-closed system by a reported method (Li et al. 2013). Approximately 15 g of melamine was kept in a crucible with a lid, which was heated at 550 °C for 4 h with a rate of 15 °C/min. After reaching the room temperature, the resulting yellow coloured g-C 3 N 4 was collected and ground well as a fine powder.

Synthesis of BiYO 3 /g-C 3 N 4 nanocomposite
The BiYO 3 /g-C 3 N 4 nanocomposite was fabricated via a facile wet-chemical process. In this method, different 1:3, 1:1 and 3:1 ratios of pure BiYO 3 and g-C 3 N 4 were added into a beaker with 25 mL of methanol. Afterward, the beaker was kept under ultra-sonication for 30 min for uniform dispersal of photocatalysts. Then, methanol was vaporised by following constant stirring at 80 °C. Later, the attained product was dried in an oven for about 5 h. The BiYO 3 /g-C 3 N 4 nanocomposites with different ratios 1:3, 1:1 and 3:1 are denoted as CB-1, CB-2 and CB-3 in the upcoming discussions.

Characterisation of photocatalysts
The obtained crystal phase of the materials was characterised by an X-ray diffractometer (XRD) (Rigaku Miniflex ii) with Cu Kα wavelength of 0.154 nm with a scan rate of 3°/ min. Then, the Fourier transform infrared (FT-IR) spectral analysis was carried out through a Perkin-Elmer FT-IR spectrometer. The UV-Visible DRS had been examined using a UV-Visible spectrophotometer (UV-DRS-SHIMADZU, UV 3600 PLUS). The morphological structure and elemental composition of the materials were attained from the EDX with scanning and high-resolution transmission electron spectroscopy (JEOL-6330 HR-TEM).

Photocatalytic degradation experiments
The photocatalytic activity of the synthesised photocatalysts was measured by degrading RhB under visible light irradiation (VLI) by utilising a 250 W tungsten halogen lamp as a source of illuminating light. Here, 75 mg of the photocatalyst was added in 75 mL of RhB (1 × 10 −5 M) solution and kept in dark with constant stirring for 30 min to attain sorption equilibrium. During the light illumination, at regular time intervals (10 min), 5 mL of aliquots was collected, and the photocatalyst was removed by centrifugation. Subsequently, the UV-vis spectrometer (JASCO-630, Japan) was employed to monitor the RhB concentration at an absorption maximum (λ max ) of 554 nm.

Photoelectrochemical experiments
The CHI608-electrochemical work station (CHI, USA) consists of three distinctive electrodes to study and analyse the photoelectrochemical properties of the prepared photocatalysts. Here, Pt and Ag-AgCl are used as the counter and reference electrodes individually. On the other hand, the FTO-coated photocatalysts, a working electrode, were prepared by mixing 5 mg of photocatalyst, 10 µL of Triton X-100 and 20 µL of DI water. The obtained slurry was layered over the FTO plate with an area about 0.5 × 0.5 cm 2 followed by heating at 80 °C for 6 h. Figure 1A illustrates the FTIR spectra of g-C 3 N 4 , BiYO 3 and BiYO 3 /g-C 3 N 4 nanocomposite photocatalysts. From Fig. 1A, the g-C 3 N 4 shows bands between 689 and 881 cm −1 corresponding to tri-S-triazine mode. In addition, the C-N stretching peaks at 1205 and 1317 cm −1 and C = N stretching at 1388-1520 cm −1 are also observed. The bands appeared at 1621 and 3264 cm −1 are owing to O-H bending and stretching respectively. Moreover, N-H stretching is found at 3071 cm −1 (Jia et al. 2019). Besides, BiYO 3

Fig. 1 A FTIR spectra and B
XRD pattern of (a) pristine BiYO 3 , (b) pristine g-C 3 N 4 and (c-e) BiYO 3 /g-C 3 N 4 composites Bi-O band's peak at 530 cm −1 and a Y-O band allocated at 881-1033 cm −1 for Y-O are observed. The peaks at 1590 and 3496 cm −1 can be ascribed to the characteristic H-O-H bending and O-H stretching vibrations individually (Wongli et al. 2017). For BiYO 3 /g-C 3 N 4 composites, the increasing amount of BiYO 3 seems to slightly influence the peak intensity of the g-C 3 N 4 as observed in the spectra. Figure 1B depicts the crystalline pattern of the pristine g-C 3 N 4 , pristine BiYO 3 and BiYO 3 /g-C 3 N 4 nanocomposites. From Fig. 1B, the XRD for bare g-C 3 N 4 shows two diffraction peaks were observed at 2θ = 12.9°, 27.6° for (100) and (002) (111), (200), (220), (311), (222) and (400) diffraction planes respectively. These peaks seem to be consisted with the cubic phase structure of BiYO 3 [JCPDS No.27-1047] (Qin et al. 2020). Both the peaks of the BiYO 3 and g-C 3 N 4 are observed in the BiYO 3 /g-C 3 N 4 nanocomposites. Then, the peak intensity of the g-C 3 N 4 at 2θ = 27.6° seemed to decrease with the increasing content of BiYO 3 . At the same time, the BiYO 3 peaks at 2θ = 31.8° and 45.7° have shifted from its normal position along with a rise in the peak intensity of 55.6° with the increasing amount of BiYO 3 in the BiYO 3 /g-C 3 N 4 composites. These results evidently supported the formation of BiYO 3 /g-C 3 N 4 heterojunction.
In order to explore the catalysts' structure, the morphological structure and elemental composition of BiYO 3 , g-C 3 N 4 and optimised CB-1 composite photocatalytic materials were investigated through TEM and EDS analysis. Figure 2a demonstrates pure BiYO 3 exhibited nanorodlike structure and accumulated 2D nanosheet-like structure for g-C 3 N 4 as shown in Fig. 2b. Furthermore, the optimised CB-1 composite presents 2D nanosheets of g-C 3 N 4 wrapped with 1D-BiYO 3 nanorods, thus revealing the closer surface interaction between BiYO 3 and g-C 3 N 4 (Fig. 2c). Figure 2d presents the HRTEM image of the optimised CB-1 composite and the lattice fringes with interplanar distances of 0.295 and 0.336 nm for (002) and (200) planes of pure g-C 3 N 4 and BiYO 3 with tetragonal and cubic phase structure. Furthermore, the SAED pattern of the optimised CB-1 composite photocatalyst is also shown in Fig. 2e-g. Figure 3a and b displays the EDX spectra of the pure BiYO 3 and g-C 3 N 4 containing Bi, Y and O and C and N respectively. Furthermore, the Fig. 3c depicts the existence of Bi, Y, O, C and N elements observed in the EDX image of the optimised CB-1 composite, thus indicating the presence of no other impurities in the prepared BiYO 3 /g-C 3 N 4 nanocomposite photocatalysts. In accordance with EDX, Fig. 4 shows the elemental mapping images of Bi, Y, O, C and N elements.
The absorption capability of light and E g values of the photocatalyst are significant factors to determine the photocatalytic degradation behaviour (Palanisamy et al. 2021). The optical absorption behaviour of the pure g-C 3 N 4 , pure BiYO 3 and optimised CB-1 composite photocatalysts was characterised through UV-vis DRS, and the results are exposed in Fig. 5a. From Fig. 5a, the pristine g-C 3 N 4 , Fig. 2 a-c TEM images of pristine BiYO 3 , pristine g-C 3 N 4 and CB-1 composite. d HRTEM image of optimised CB-1 composite. e-g SAED patterns for pure BiYO 3 , pure g-C 3 N 4 and CB-1 composite photocatalysts pristine BiYO 3 and CB-1 show the absorption edges at 428, 487 and 465 nm respectively. In addition, the following Kubelka-Munk function is used to calculate the E g values of photocatalysts (Bavani et al. 2021b).  Here, E g , A, h, v and α are the band gap energy value, constant, Planck's constant, light absorption frequency and absorption co-efficient respectively. Figure 5b displays the αhν vs. photon energy (hv) plot. Here, the intercept on X-axis shows the corresponding E g values. In Fig. 5b, the pure g-C 3 N 4 , pure BiYO 3 and CB-1 composite has E g values 2.84, 2.77 and 2.81 eV respectively. The acquired results show that the CB-1 composite has lower E g value than the pristine g-C 3 N 4 , owing to the addition of BiYO 3 in the BiYO 3 /g-C 3 N 4 composite. The Brunauer-Emmett-Teller (BET) surface area of the optimised CB-1 composite photocatalyst was carefully examined via N 2 adsorption-desorption analysis, and the obtained outcomes are presented in Fig. S1. In Fig. S1, the type-IV isotherm which exposes H3 hysteresis loop indicates mesopore nature of the optimised photocatalyst. The BET specific surface area of the optimised CB-1 composite was calculated to be 14.8 m 2 g −1 .
Photoluminescence (PL) emission spectra of the prepared photocatalyst expose the separation and transfer efficiency of the photo-produced e − /h + pairs. In general, the smaller PL emission intensity reveals the lower recombination of charge carriers. As seen in Fig. 6, comparatively, the BiYO 3 /g-C 3 N 4 composite exhibits lower PL emission intensity than that of pure BiYO 3 , thus revealing the reduced recombination of e − /h + pairs. Obviously, the addition of BiYO 3 in BiYO 3 /g-C 3 N 4 composite have considerably influenced the PL emission peak intensity. The separation and migration ability of the pristine g-C 3 N 4 , pristine BiYO 3 and CB-1 composite photocatalysts were characterised by photoelectrochemical studies as presented in Fig. 7a and b. From Fig. 7a, the transient photocurrent study indicates the higher photocurrent intensity of optimised CB-1 composite and has greater separation efficiency than the pure g-C 3 N 4 and BiYO 3 . From Fig. 7a, the photocurrent intensity of the optimised CB-1 photocatalyst is 1.048 and 1.16 times higher than that of the pure BiYO 3 and g-C 3 N 4 , which reveals the more charge transfer efficiency of the prepared BiYO 3 /g-C 3 N 4 heterojunction. In electrochemical impedance spectroscopy (EIS) displayed in Fig. 7b, the smaller semicircular radius of the Nyquist plot indicates the lower resistance. From Fig. 7b, the optimised CB-1 composite possesses smaller arc radius than the bare g-C 3 N 4 and BiYO 3 , which represents the lower charge transfer resistance. These outcomes show the creation of BiYO 3 /g-C 3 N 4 heterojunction has feasibly enhanced the separation and transfer of e − /h + pairs on the interface of the heterojunction.

Photocatalytic degradation studies
Photocatalytic degradation activity of a series of BiYO 3 /g-C 3 N 4 composite photocatalysts was studied via the RhB degradation under VLI. Figure 8a displays the poor photocatalytic decomposition ability of the pristine BiYO 3 and g-C 3 N 4 when compared with BiYO 3 /g-C 3 N 4 composite photocatalysts. Here, the photocatalytic degradation Kubelka-Munk plot of pristine BiYO 3 , pristine g-C 3 N 4 and CB-1 composite Fig. 6 PL spectra of pristine BiYO 3 , pristine g-C 3 N 4 and BiYO 3 /g-C 3 N 4 composite efficiency of the pure g-C 3 N 4 , BiYO 3 , CB-1, CB-2 and CB-3 are 49.3, 16.2, 87.9, 46.7 and 33.5%, respectively (Fig. 8b), towards the degradation of RhB in 60 min. Among the different ratio of BiYO 3 /g-C 3 N 4 composites, the CB-1 shows greater photocatalytic efficiency than the other prepared photocatalysts; it is mainly accredited by the formation of BiYO 3 /g-C 3 N 4 heterojunction. Here, the photocatalytic degradation ability of the BiYO 3 /g-C 3 N 4 composite seems to increase with the increasing content of BiYO 3 , but the excessive amount of BiYO 3 possibly affected the formation of the heterojunction between the two semiconductors and also impeded the transfer of charges. In addition, the kinetic study of the prepared materials for the decomposition of RhB was analysed by the fitting with first-order kinetics equation (Theerthagiri et al. 2014;Long et al. 2006;Yin et al. 2017).
where, k, t, C 0 and C are the apparent rate constant (min −1 ), irradiation time (min), concentration of dye at initial and different time intervals correspondingly. Figure 8c and d represents the k values of pure BiYO 3 , pure g-C 3 N 4 and CB-1, CB-2 and CB-3 are 0.0021, 0.0125, 0.0319, 0.0096 and 0.00658 min −1 respectively. The optimised CB-1 heterostructure shows 15 and 2 times greater rate constant value than pure BiYO 3 and g-C 3 N 4 . The obtained outcomes reveal the introduction of the BiYO 3 in (2) lnC 0 ∕C = kt Fig. 7 a Transient photocurrent and b Nyquist plot of pristine BiYO 3 , pristine g-C 3 N 4 and BiYO 3 /g-C 3 N 4 composite Fig. 8 a Photocatalytic degradation plot, b photocatalytic degradation efficiency, c pseudo first-order kinetics plots and d rate constant values of pristine BiYO 3 , pristine g-C 3 N 4 and BiYO 3 /g-C 3 N 4 composites towards the RhB degradation under VLI (all of these experiments carried triplicate) the BiYO 3 /g-C 3 N 4 composite has effectively increased the photocatalytic degradation efficiency. The Table 1 represents the comparative photocatalytic degradation study of different BiYO 3 and g-C 3 N 4 nanocomposite photocatalysts for the degradation of different organic pollutants with various experimetnal conditions.
The effect of catalyst amount of the optimised CB-1 composites on the RhB degradation under VLI is shown in Fig. 9. Figure 9 depicts the effect of optimised CB-1 catalyst amount from 0.5 to 2.0 g/L on the degradation of RhB over composite photocatalyst under identical reaction conditions. The obtained results reveal the increasing trend of photocatalytic degradation while increasing the amount of catalyst from 0.5 to 1.0 g/L, but the later additions of catalyst exhibited reduced photocatalytic activity, attributing to the colloidal formation that tends to scatter the light on the surface of catalyst. Thus, the excessive amount of catalyst is found not to be suitable for the photocatalytic RhB degradation. At the same time, recyclability and stability of the photocatalysts are also the vital factors for practical applications. The recycling experiment was conducted by collecting the photocatalyst at the end of every cycle and several times washing it with water then dried at 80 °C for 6 h. Figure 10a displays the optimised CB-1 composite resuming its initial stability with negligible changes after three cycles. Furthermore, the XRD pattern of the optimised catalyst CB-1 before and after three recycling experiments is shown in Fig. S2. As seen, the crystalline phase structure of CB-1 composite is remained with negligible changes thereby endorsing the stability of the photocatalyst. According to the acquired results, eventually, it is revealed that the optimised CB-1 composite possess a reasonable reusability characteristics along with photostability features after three cycling runs. Furthermore, to identify the major reactive species in the RhB degradation over CB-1 composite, radical scavenging experiments were preferred. A variety of trapping agents like AO, IPA and BQ were used to trap h + , • OH and O 2 •− radicals individually [3]. Figure 10b represents the BQ and IPA apparently display the inhibitory effect in the RhB degradation under VLI. But the addition of AO showed no obvious effects on the rate of RhB degradation. These experimental results proved that the O 2 •− and • OH are playing the vital roles in RhB degradation by the optimised CB-1 composite under VLI. Figure 11 represents the Z-scheme mechanism for the RhB degradation over BiYO 3 /g-C 3 N 4 in the presence of VLI. Here, the values of conduction band (CB) and valence band (VB) potentials are calculated by using the Butler-Gingley Eqs. (3) and (4) respectively (Priya et al. 2018;Sun et al. 2020;Senthil et al. 2019b;Bavani et al. 2022a, b), here, E g, E e , X, E CB and E VB are the band gap energy, energy of a free electron on the hydrogen scale (~ 4.5 eV), absolute electronegativity and conduction and valence band potentials of a semiconductor correspondingly. In the presence of VLI, both the BiYO 3 and g-C 3 N 4 are excited to produce the photo-generated e − /h + pairs. Simultaneously, the electrons on the CB of BiYO 3 are transferred Fig. 11 Pictorial representation of mechanism for Z-scheme BiYO 3 /g-C 3 N 4 heterojunction towards the RhB degradation under VLI to the VB of g-C 3 N 4 , and holes on the VB of the g-C 3 N 4 are transferred into the CB of BiYO 3 . At the time, the CB potential of g-C 3 N 4 is more negative than the standard reduction potential O 2 /O 2 •− E°, here the O 2 is reduced to of 4 O 2 •− (0.33 eV vs. NHE). Similarly, the • OH is created on the oxidation of H 2 O, due to the more positive VB potential of BiYO 3 than H 2 O/ • OH (+ 2.40 eV vs. NHE). Therefore, both the photo reduction and oxidation result in the degradation of organic pollutant into harmless and biodegradable products. Successfully, the generated BiYO 3 /g-C 3 N 4 Z-scheme heterojunction greatly improved the photocatalytic efficiency by effectively enhancing the charge separation and transfer efficiency and limiting the reconnection of e − /h + pairs. The following equations depict the photocatalytic degradation of BiYO 3 /g-C 3 N 4 Z-scheme heterojunction towards the RhB degradation under VLI.

Conclusions
The 1D/2D BiYO 3 /g-C 3 N 4 composite was synthesised via simple wet chemical method and characterised through FT-IR, XRD, UV-DRS, PL, SEM, EDX and TEM analyses. The photocatalytic degradation activity of the BiYO 3 /g-C 3 N 4 composite is comparatively greater than that of pristine BiYO 3 and g-C 3 N 4 . The optimised CB-1 composite possess greater photostability and reusability even after three cycles. The optimised CB-1 photocatalyst is 1.048 and 1.16 times higher than pure BiYO 3 and g-C 3 N 4 . The radical trapping studies confirmed the O 2 •− and OH radicals are majorly involved in the RhB degradation under VLI. A feasible mechanism of Z-scheme BiYO3/g-C3N4 heterojunction is accounted for the RhB degradation. From the above results, the prepared Z-scheme BiYO3/g-C3N4 heterojunction photocatalyst could act as a promising photocatalyst for the wastewater remediation.