Introduction

With the development of the industrialization, the environment pollution has become one of the major concerns of the international community. Photocatalytic (PC) oxidation is considered as a promising way to degrade environment pollutant due to its high oxidation ability and broad compound applicability [1, 2]. As an enhanced photocatalysis technique, photoelectrocatalytic (PEC) oxidation has been proven to be an efficient method that can greatly enhance the PC oxidation activity by exploiting the synergism between PC oxidation and electro-oxidation [3,4,5,6,7,8]. Moreover, photocatalyst immobilized on a conductive substrate as an electrode is more practical than that in a powdery form for industrial application.

TiO2 is the most widely used photocatalyst due to its high oxidation ability, high resistance to photocorrosion, high chemical stability, non-toxic and low cost [9,10,11]. However, the low quantum efficiency, poor visible light response and separation problems greatly limit the large-scale industrial application of TiO2 nanoparticles [9]. Many kinds of TiO2 films on various substrates were used to replace TiO2 nanoparticles as photocatalyst. Compared with TiO2 film, TiO2 nanowire arrays as one-dimensional (1D) nanostructures are expected to have higher photocatalytic activity due to their large surface-to-volume ratios, rapid charge transport and recyclable characteristics [12,13,14]. To extend the application of TiO2 nanowire arrays, considerable efforts have been devoted to improve the photogenerated charge separation efficiency and extend the light response range, such as element doping [15, 16], noble metal deposition [17, 18] and semiconductor coupling [19,20,21,22,23,24].

Coupling TiO2 nanowires with another narrow bandgap semiconductor is an effective way to utilize visible light and enhance photocatalytic activity. Currently, chalcogenide materials (such as CdS, CdSe and ZnSe) are preferred and mostly used narrow bandgap semiconductor to combine with TiO2 nanowires due to their visible light absorption ability and suitable band position [19,20,21,22]. However, chalcogenide materials always suffer photocorrosion and instability in the PEC reaction, leading to a leakage of heavy metal, which is harmful to environmental remediation. Therefore, it is urgent to develop a visible-light-responsive, efficient and environmental-friendly electrode for PEC degradation application. Graphitic carbon nitride (g-C3N4) with a bandgap of 2.7 eV is the most stable allotrope of carbon nitride and is a metal-free semiconductor [25, 26]. It has attracted extensive scientific interest due to its high chemical stability and visible light absorption ability. G-C3N4 has been widely employed in the degradation of dye, phenol and other pollutants [27, 28]. Coupling g-C3N4 with wide bandgap semiconductors with more positive conduction band is an effective way to fabricate efficient and visible-light-responsive photocatalysts. Many g-C3N4-based composites such as ZnO/C3N4 [29], Bi2WO6/C3N4 [30], ZnWO4/C3N4 [31] have been reported and shown enhanced photocatalytic performance. The combination of TiO2 and g-C3N4 may be an ideal system to achieve an enhanced charge separation and transfer due to the matched band alignment between TiO2 and g-C3N4. Moreover, the visible-light-responsive capacity, chemical stability and environmental-friendly structure make g-C3N4 an excellent candidate in PEC degradation application. Actually, some TiO2/g-C3N4 combination works were developed and shown promising results [32,33,34,35]. Zhu and co-workers synthesized a C3N4/TiO2 hybrid photocatalyst with highly improved photocatalytic activity via a ball milling method [34]. Wang and co-workers synthesized a macro-/mesoporous g-C3N4/TiO2 composite and found that g-C3N4/TiO2 presented a higher photocatalytic activity than pure TiO2 and g-C3N4. The enhanced photocatalytic performance was attributed to the heterojunction between TiO2 and g-C3N4, as well as the high surface area [32]. However, to the best of our knowledge, there is no report regarding the fabrication and application of TiO2/g-C3N4 core–shell nanowire arrays electrode in the PEC degradation although it has many unique merit in environmental application. Moreover, the core–shell structure with proper band alignment and well-matched interface can achieve the selective isolation of charge carriers by spatial separation of electrons and holes across the core–shell interface, thus suppressing the charge recombination.

Herein, we report our efforts in fabricating TiO2/g-C3N4 (denoted as TiO2/C3N4) core–shell nanowire arrays and exploring the application of this TiO2/C3N4 electrode in PEC degradation of bisphenol A (BPA). These TiO2/C3N4 core–shell nanowire arrays were synthesized via a hydrothermal method followed by a chemical vapor deposition (CVD) method. TiO2 nanowire arrays provide a direct electron pathway, and g-C3N4 serves as a photo-sensitizer; the type-II band alignment and the well-matched core–shell structure of TiO2 and g-C3N4 can facilitate the photogenerated charge separation and transfer. BPA was chosen as a model pollutant to investigate the PC and PEC activity of as-prepared samples. BPA is one of the priority endocrine disrupting compounds which should be monitored and controlled due to its large-scale production and extensive applications. Many research reports point out that even very low level of BPA may cause infertility as well as breast, ovarian and testicular cancer of human and wildlife [36, 37]. TiO2/C3N4 core–shell nanowire arrays show superior BPA degradation activity than pure TiO2 nanowire arrays. The effect of external potential on PEC degradation of BPA, the identification of BPA degradation intermediate products and the possible degradation pathway were systematically investigated.

Experimental

Preparation of the TiO2 nanowire arrays and TiO2/C3N4 core–shell nanowire arrays

Preparation of TiO2 nanowire arrays: The TiO2 nanowire arrays on a carbon cloth were prepared using a hydrothermal method according to a previous report [38]. Firstly, the carbon cloth was cleaned by sonication in acetone and methanol, subsequently rinsed with deionized (DI) water and finally dried in a nitrogen stream. The carbon cloth was placed within a sealed Teflon reactor (200 mL), containing 100 mL of toluene, 10 mL of tetrabutyl titanate, 10 mL of titanium tetrachloride (1 M in toluene) and 10 mL of hydrochloric acid (37 wt%). The Teflon reactor was kept in an oven at 180 °C for 22 h. After cooling down to room temperature, the carbon cloth was taken out, then rinsed with ethanol and DI water, and finally annealed in air at 500 °C for 2 h.

Preparation of TiO2/C3N4 core–shell nanowire arrays: Deposition of g-C3N4 onto the TiO2 nanowire was performed by a CVD process, and melamine was used as a precursor. Firstly, 2.0 g melamine was added in the bottom of a ceramic crucible with a cover, and a carbon cloth with pre-grown TiO2 nanowire was placed on the upper position of the crucible. Then, the crucible was heated at 550 °C for 4 h with a temperature rise rate at 5 °C/min in a muffle furnace. After cooling down to room temperature, the as-prepared samples were rinsed with ethanol and DI water to remove the g-C3N4 attaching to the carbon cloth. In the bottom of the crucible, g-C3N4 powders can also be obtained.

Characterization

X-ray diffraction patterns were measured by X-ray diffraction with Cu Ka radiation (λ = 1.5406 Å) on a Shimadzu XRD 6000 diffractometer. Morphologies of the as-prepared photocatalysts were characterized by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). SEM images were investigated with a FEI Quanta 200F microscope at an accelerating voltage of 30 kV. TEM and high-resolution transmission electron microscopy (HRTEM) were performed on a JEOL JEM-2100 microscope with an accelerating voltage of 120 kV. The UV–vis diffuse reflectance spectroscopy (DRS) was obtained by a UV–vis spectrophotometer (Hitachi U-4100) with the integration sphere diffuse reflectance attachment, using BaSO4 as the reference. X-ray photoelectron spectroscopy (XPS) spectra were carried out on a K-Alpha XPS instrument. The concentration of BPA during the reaction was examined by a high-performance liquid chromatography (HPLC) on a Agilent 1200 UPLC system with the following conditions: column, 100 mm BEH C18 column; detector, UV detector operated at 280 nm; mobile phase, 1 mL/min of methanol and water (70/30, v/v); injection volume of BPA, 5 μL. The intermediate products of the PEC degradation of BPA were identified by gas chromatography–mass spectrometry (GC/MS) on a Bruker (ACION TQ, USA) combined with an Agilent DB (30 × 0.25 mm, 0.25 μm) quartz capillary column and an electron impact (EI) detector (70 eV). The initial temperature of column oven was 40 °C and held for 1 min; then, it increased up to 280 °C with a heating rate of 15 °C/min and held for 8 min. The temperatures were set as follows: injection port, 260 °C; interface, 250 °C; ion source, 230 °C. Helium was used as a carrier gas with a flow rate of 1 mL/min.

Photoelectrocatalysis experiments

The photoelectrochemical measurements and PEC reactions were performed in a rectangular reactor using a CHI 660E (Shanghai, China) electrochemical workstation. A standard three-electrode cell system was used in the photoelectrochemical measurements and PEC reactions, including a Pt counter electrode, a working electrode (TiO2/C3N4 core–shell nanowire arrays, active area of 2 × 2 cm2) and a reference electrode (a saturated calomel electrode, SCE). The TiO2/C3N4 core–shell nanowire arrays electrode was irradiated with visible light obtained from a 300 W Xe lamp (PLS-SXE300C/300CUV, Perfect Light, Beijing) with 420-nm cutoff filter. The average visible light intensity was 100 mW cm−2. An electrolyte solution with a concentration of 0.1 mol L−1 Na2SO4 was used in the photoelectrochemical measurements. Photocurrent responses were investigated at 0.0 V (with reference to the SCE) under light-on and light-off cycles. The electrochemical impedance spectra (EIS) were performed in the frequency range of 100 kHz–0.01 Hz with an amplitude of the sinusoidal wave of 5 mV. The PC and PEC performance of the as-prepared samples was measured by the removal of BPA (with an initial concentration of 10 mg L−1) in 200 mL Na2SO4 (0.1 mol L−1) solution. The main oxidative species in the PEC process of the TiO2/C3N4 core–shell nanowire arrays were detected through trapping experiment by addition of hole scavenger (methanol), hydroxyl radical scavenger (tert-butyl alcohol, t-BuOH) and superoxide radical scavenger (benzoquinone, BQ).

Results and discussion

Catalyst characterization

Figure 1a shows the two-step fabrication process of TiO2/C3N4 core–shell nanowire arrays grown on carbon cloth. Firstly, the TiO2 nanowire arrays were grown on the carbon cloth substrate by hydrothermal method. Then, the C3N4 shell was deposited on the surface of TiO2 nanowires via CVD method using melamine as a precursor. The XRD patterns of TiO2, C3N4 and TiO2/C3N4 core–shell nanowire arrays are shown in Fig. 1b. Diffraction peaks appearing in the pure TiO2 nanowires can be well indexed to the rutile TiO2 (JCPDS.21-1276). Pure C3N4 shows two characteristic peaks at 12.9 and 27.3°, which can be attributed to the interplanar separation (100) and the stacking of the conjugated aromatic system (002) (JCPDS 87-1526) [39, 40]. The TiO2/C3N4 core–shell nanowire arrays exhibit all the characteristic peaks of rutile TiO2, suggesting that the crystal phase of TiO2 does not change after deposited with C3N4 shell. It is worth to point out that the peak of (002) planar of g-C3N4 at 27.3° overlaps with the (110) planar peak of rutile TiO2. Nevertheless, the characteristic peaks of g-C3N4 at 12.9° can be observed in the XRD pattern of TiO2/C3N4 core–shell nanowire arrays, indicating that the g-C3N4 was successfully deposited on the TiO2 nanowires. The optical properties of the TiO2, C3N4 and TiO2/C3N4 core–shell nanowire arrays were examined by UV–vis DRS spectra, and the results are shown in Fig. 1c. Pure TiO2 nanowires show a sharp absorption edge rise at around 410 nm, which corresponds to a band gap of 3.0 eV. The pure C3N4 displays a characteristic absorption rise at around 550 nm, corresponding to a band gap of 2.3 eV. Compared with that of pure TiO2, the absorption edge of TiO2/C3N4 nanowires was extended to the visible light range due to the presence of C3N4, which is beneficial for the solar energy utilization.

Figure 1
figure 1

a Fabrication process of TiO2/C3N4 core–shell nanowire arrays; b the XRD patterns of TiO2, C3N4 and TiO2/C3N4 core–shell nanowire arrays; c the UV–vis DRS spectra of TiO2, C3N4 and TiO2/C3N4 core–shell nanowire arrays

Full size image

The structural morphology of TiO2 and TiO2/C3N4 nanowire arrays was examined by SEM analysis (Fig. 2). Figure 2a, b shows the SEM images of TiO2 nanowire arrays grown on the carbon cloth. The inset of Fig. 2a is pure carbon cloth. As can be seen, the high-density and orderly TiO2 nanowire arrays were grown uniformly on the carbon cloth. These TiO2 nanowires with smooth surface are about 30–50 nm in diameter (Fig. 2b). The CVD method was performed to deposit the C3N4 shell on the pre-grown TiO2 nanowires. Figure 2c shows a large-scale SEM image of the TiO2/C3N4 core–shell nanowire arrays, which indicates that the structure of TiO2 nanowires do not obviously change after C3N4 deposition. From the high-magnification SEM image of TiO2/C3N4 nanowire (Fig. 2d), a shell deposited on the surface of the TiO2 nanowire can be clearly observed. The surfaces of the TiO2 nanowires become rough, and the diameters of the TiO2 nanowires increase up to the range of 80–100 nm after deposited with C3N4. As can be seen, the high-density and orderly TiO2/C3N4 core–shell nanowires were successfully synthesized.

Figure 2
figure 2

SEM images of TiO2 nanowire arrays (a, b) and TiO2/C3N4 core–shell nanowire arrays (c, d); inset of (a): the pure carbon cloth

Full size image

The HRTEM images of TiO2/C3N4 core–shell nanowires confirm that TiO2 nanowire is covered by a thin layer of C3N4 (Fig. S1). The diameter of TiO2 nanowire is about 30 nm, which is consistent with the SEM results. And the measured interplanar spacing of TiO2 is 0.325 nm, which matches well with the rutile TiO2 (110) plane. The measured interplanar spacing of 0.320 nm in the deposition layer is correlated with the (002) crystal planes of C3N4, further confirming the deposition of C3N4.

The XPS measurement was performed to investigate the surface composition and chemical states of TiO2 and TiO2/C3N4 nanowire arrays. Figure 3 shows the XPS survey spectra and magnified spectra of N 1s, C 1s, Ti 2p and O 1s. As can be seen from Fig. 3a, TiO2 nanowire arrays are composed of C, Ti, O, whereas TiO2/C3N4 core–shell nanowire arrays are composed of C, Ti, O and N, indicating the formation of C3N4 in the TiO2/C3N4 nanowire arrays. In the magnified N 1 s spectra of TiO2/C3N4 nanowire arrays (Fig. 3b), a broad peak extending from 395 to 406 eV can be observed. Three peaks can be distinguished to be centered at 398.3, 400.3, 404.0 eV, respectively. The main peak at 398.3 eV can be attributed to the sp2-hybridized nitrogen (N–C=N), indicating the presence of sp2-bonded C3N4. The other two peaks at 400.3 and 404.0 eV can be assigned to the nitrogen in tertiary N–(C)3 groups and the charging effects [41, 42]. Figure 3c shows the C 1s features of TiO2/C3N4 core–shell nanowire arrays. The peak at 284.1 eV is assigned to the C–C group due to the carbon contamination. And the peak at 288.0 eV is attributed to the N–C=N group in the graphitic C3N4 [39, 40]. From the Ti 2p spectrum of TiO2/C3N4 nanowire arrays (Fig. 3d), two peaks at binding energy of 458.3 eV (Ti 2p3/2) and 464.0 eV (Ti 2p1/2) can be observed [43]. As can be seen from Fig. 3e, the O 1s peak at binding energy of 529.6 and 531.2 eV can be attributed to the O2− in the sample and the –OH group of water molecules on the surface of the sample, respectively [44]. The XPS spectra further confirm the presence of the C3N4.

Figure 3
figure 3

XPS survey spectra of TiO2 and TiO2/C3N4 nanowire arrays (a); magnified spectra of b N 1s peaks, c C 1s peaks, d Ti 2p peaks and e O 1s peaks of TiO2/C3N4 core–shell nanowire arrays

Full size image

Photoelectrochemical properties

Photoelectrochemical experiments were carried out in a standard three-electrode system. The photocurrent responses of TiO2 nanowire arrays and TiO2/C3N4 core–shell nanowire arrays were investigated with light-on and light-off cycles at a bias potential of 0 V versus SCE electrode, and the results are shown in Fig. 4a. Fast and steady photocurrent responses are observed for each light-on and light-off cycle in TiO2 and TiO2/C3N4 electrodes, indicating that the electrodes have good chemical stability and photo-stability. The photocurrent response of carbon cloth was also investigated for comparison, and no photocurrent response can be observed. Under visible light irradiation, the photocurrent density of the pure TiO2 is 0.0025 mA/cm2. Meanwhile, the photocurrent density of the TiO2/C3N4 is 0.0075 mA/cm2, which is three times of that of pure TiO2. The photocurrent enhancement of the TiO2/C3N4 electrode suggests a higher charge separation efficiency, which is beneficial for the BPA PC and PEC degradation process. EIS is a powerful approach to investigate the resistance of the electrodes, which indicates the charge separation process in electrodes [9, 45]. Figure 4b displays the EIS Nyquist plots of the TiO2 and TiO2/C3N4 electrodes with and without visible light irradiation. The radius of the arc in the EIS spectra indicates the charge separation and transfer resistance at the electrode; the smaller the radius, the higher the charge separation efficiency [9, 30, 45]. The TiO2/C3N4 electrodes show smaller radius of the arc than pure TiO2 with and without visible light irradiation, suggesting an improved charge separation and transfer process in TiO2/C3N4. The results of photoelectrochemical characterization indicate that the C3N4 modification can effectively enhance the photogenerated charge separation efficiency of the TiO2 nanowires.

Figure 4
figure 4

Photocurrent responses (a) and EIS Nyquist plots b of TiO2 nanowire arrays and TiO2/C3N4 core–shell nanowire arrays under visible light irradiation, λ ≥ 420 nm

Full size image

Enhancement of PC and PEC activity

The BPA PC and PEC degradation experiment was performed under visible light irradiation (λ ≥ 420 nm). Figure 5a shows the PC and PEC activity of TiO2, C3N4 and TiO2/C3N4, and the apparent rate constant k is shown in the inset graph. As is well known, the PC and PEC degradation processes obey pseudo-first-order kinetics; the slope of the linear line in the plots of ln(C/C0) of BPA versus irradiation time (t) represents the apparent rate constant k [46, 47]. The electro-oxidation of BPA at 1.0 V versus SCE electrode and visible light photolysis (in the absence of the photocatalyst) were performed as references. As can be seen, the visible light photolysis (in the absence of the photocatalyst) has no obvious effect on BPA removal. The TiO2/C3N4 shows much higher PEC activity than pure TiO2 and C3N4. The rate constant k of PEC degradation of TiO2/C3N4 under 1.0 V external potential is 0.00411 min−1, which is one time higher than that of pure TiO2 (0.00203 min−1). Moreover, the reaction rate constant k of PEC degradation of TiO2/C3N4 is larger than the sum of the electro-oxidation and PC degradation, suggesting a synergetic effect between electro-oxidation and PC degradation in the PEC process. The effect of the external potential on the PEC degradation of BPA of TiO2/C3N4 was investigated, and the results are shown in Fig. 5b. As can be seen, the PEC activity first increases and then decreases with the increasing external potential. The TiO2/C3N4 core–shell nanowire arrays present highest PEC activity (apparent rate constant k = 0.01041 min−1) at external potential of 2 V. When the external potential is higher than 2 V, the high external potential may induce fast electro-polymerization of BPA, which would block the current transfer and destroy the synergetic effect of the system, leading to a decreased PEC activity.

Figure 5
figure 5

Comparison of electro-oxidation, PC and PEC degradation rate of BPA a (λ ≥ 420 nm, external potential = 1.0 V), and comparison of the degradation rate of BPA over TiO2/C3N4 core–shell nanowire arrays under different external potentials (λ ≥ 420 nm)

Full size image

Identification of intermediate products and the possible degradation pathway

The identity of the intermediate products from the degradation of BPA was monitored by HPLC and GC–MS. The intermediate products are shown in Table S1, and the possible BPA degradation pathway in PEC process is proposed in Fig. 6. Five main intermediates were identified as 3-pentanone, p-isopropenyl phenol, 1-(4-benzyl alcohol) ethanone, (1,1′-phenyl) ethylene and (2-methyl-1-phenyl-1-propenyl) benzene. Based on these intermediates, the possible degradation pathway of BPA was proposed, including dehydroxylation, cleavage of C–C bonds, elimination reactions and oxidation [48, 49]. As can be seen from Fig. 6, the degradation of BPA in PEC process can be divided into two possible pathways according to the bonds broken and the previously reported results [48, 49]. The first one is dehydroxylation; the bonds connecting two hydroxyl groups with aromatic rings in BPA are first cleaved, leading to the formation of (2-methyl-1-phenyl-1-propenyl) benzene. Then, (2-methyl-1-phenyl-1-propenyl) benzene is subsequently oxidized to form (1, 1′-phenyl) ethylene. Furthermore, the C–C bond connecting the two aromatic rings is vulnerable because of the two electron-donating hydroxyl groups. Therefore, the second pathway is the cleavage of the C–C bond connecting two aromatic rings in BPA, resulting in the formation of p-isopropenyl phenol. The p-isopropenyl phenol can be degraded to p-hydroxyacetophenone. These intermediates are further oxidized to generate 3-pentanone, which can be further oxidized by radicals.

Figure 6
figure 6

Proposed BPA degradation pathway by TiO2/C3N4 core–shell nanowire arrays

Full size image

Mechanism of enhancement of PEC activity

The identification of main oxidative species is important to reveal the mechanism of the synergistic effect of TiO2/C3N4. Generally, three oxidative species, hydroxyl radical, hole and superoxide radical, are hypothesized to be involved in photocatalysis reaction. The main oxidative species were detected through the trapping experiments by adding hole scavenger (methanol) [50], hydroxyl radical scavenger (t-BuOH) [29, 51] and superoxide radical scavenger (BQ) [43], respectively. The results are shown in Fig. 7. The addition of a scavenger of superoxide radicals (BQ) obviously decreases the PEC activity of TiO2/C3N4, indicating that superoxide radicals are the main oxidative species in the BPA degradation. The addition of methanol and t-BuOH reduces the PEC activity of TiO2/C3N4 in a certain extent, suggesting that hole and hydroxyl radical are part of oxidative species.

Figure 7
figure 7

Plots of photogenerated carriers trapping on PEC process by TiO2/C3N4 under 2.0 V external potential and visible light irradiation (λ ≥ 420 nm)

Full size image

Based on the above results, a schematic diagram of the possible mechanism of the charge transfer process of the TiO2/C3N4 core–shell nanowire array is shown in Fig. 8. The position of valance band (VB) and conduction band (CB) of C3N4 is higher than that of TiO2 according to the reported results [33]. The suitable valence band and conduction band position of TiO2 and C3N4 can form a type-II structure, which can lead to a spatial separation of the electrons and holes on TiO2 and C3N4 heterojunction [52], resulting in an enhanced charge separation efficiency. The TiO2 itself cannot be excited by visible light, but C3N4 can absorb visible light to generate electrons and holes. Therefore, under visible light irradiation, the excited electrons of the CB of C3N4 would inject into the CB of TiO2 due to the type-II structure. The electrons would be subsequently driven to the counter electrode by the external potential, leaving the holes on the C3N4 shell, resulting in an improved charge separation and transfer process. The electrons and holes would subsequently react with water and oxygen to produce active superoxide and hydroxyl radicals. The holes and radicals would participate in the degradation of BPA and its intermediate products, leading to a visible light activity of TiO2/C3N4 core–shell nanowire arrays. Moreover, the synergistic effect of PC and electro-oxidation can further enhance the activity of TiO2/C3N4. The applied external potential not only can enhance the charge separation efficiency by driving the electrons to the counter electrode, but also can directly degrade the BPA.

Figure 8
figure 8

Schematic diagram of the possible mechanism of the charge transfer process of the TiO2/C3N4 core–shell nanowire arrays under visible light irradiation

Full size image

Conclusions

In summary, novel TiO2/C3N4 core–shell nanowire arrays were successfully prepared via a hydrothermal method followed by a CVD method. TiO2/C3N4 core–shell nanowire arrays served as a working electrode, which was applied to PEC degradation of BPA. TiO2 nanowire arrays provide a direct electron pathway, g-C3N4 serves as a stable and environmental-friendly visible light sensitizer, and effective charge spatial separation can be achieved across the well-matched core–shell interface. The TiO2/C3N4 core–shell nanowire arrays exhibit higher PC and PEC performance than TiO2 nanowire arrays. The PEC performance of TiO2/C3N4 core–shell nanowire arrays is enhanced by one time after C3N4 modification. The enhancement of PEC BPA degradation performance of TiO2 nanowire arrays after C3N4 modification originates from the photoabsorption enhancement and efficient charge separation induced by the type-II band alignment. The synergetic effect between electro-oxidation and PC degradation in PEC process can further improve the degradation of BPA. Two possible degradation pathways of BPA were proposed, including dehydroxylation, cleavage of C–C bonds and oxidation. We believe that this contribution might provide a promising way to construct visible-light-driven TiO2-based heterostructure photocatalysts.