Trace Amounts of Co₃O₄ Nano-Particles Modified TiO₂ Nanorod Arrays for Boosted Photoelectrocatalytic Removal of Organic Pollutants in Water

Abstract

Trace amounts of Co₃O₄ modified TiO₂ nanorod arrays were successfully fabricated through the photochemical deposition method without adding any nocuous reagents. The Co₃O₄/TiO₂ nanorod arrays fabricated in acid solution had the highest photo-electrochemical activity. We elaborated on the mechanism of Co₃O₄-TiO₂ fabricated in different pH value solutions. The Co₃O₄-TiO₂ had a more remarkable photo-electrochemical performance than the pure TiO₂ nanorod arrays owing to the heterojunction between Co₃O₄ and TiO₂. The degradation of methylene blue and hydroquinone was selected as the model reactions to evaluate the photo-electrochemical performance of Co₃O₄-TiO₂ nanorod arrays. The Co₃O₄/TiO₂ nanorod arrays had great potential in waste water treatment.

1. Introduction

From the comprehensive point of view, semiconductor metal oxides have been widely used as stable photo-catalysts for the cosmopolitan energy crisis [1,2,3,4,5]. TiO₂ is the most extensively used semiconductor photo-catalyst owing to its exceptional properties, such as high photo-catalytic activity, chemical stability, environmental-friendliness, and low cost [3,4,6,7,8]. However, because of the large bandgap of TiO₂ (3.2 eV), the practical applications are hampered by its low electrical conductivity, strong reflection, and weak light-harvesting ability [9,10], as well as the rapid combination of photo-generated electron and hole pairs. Various strategies have been utilized to improve the photo-catalytic efficiency of TiO₂ materials, such as tuning their crystallite size and structure, sensitizing them by organic dye and quantum dots [8,11,12], and modifying them with metals (e.g., Pt, Ru, Ag, Au, Rh, Pd, Ni, and Co) [3,10,13,14,15,16,17] or transition metal oxides (e.g., Co₃O₄, CoO, Cu₂O, and Fe₂O₃) [18,19,20,21] with a narrow band gap semiconductor. The formation of hetero-junctions between metal oxides and semiconductors is a useful strategy to suppress the recombination of photo-generated electrons and holes in TiO₂, and extend photon absorption into the visible regime, which can enhance the photochemical efficiency of TiO₂ nanomaterials.

Semiconductor-based hetero-junctions are able to facilitate fast charge separation and enhance the photo-catalytic efficiency of TiO₂ nanomaterials. It is an effective strategy to construct TiO₂-based hetero-junction structures with transition metal oxides. Heterojunction structures have the potential to facilitate electron–hole separation. TiO₂ is a n-type semiconductor and, combined with a p-type semiconductor in a suitable band gap position to form a p-n heterojunction, it is an effective tactic to expand light absorption, enhance the separation effects of electrons and holes, prolong the lifetime of the electron and hole, and heighten the photocatalytic activity. Transition metal oxides, such as Ag₂O, Cu₂O, CuO, and Co₃O₄, have been used to form a p-n junction to promote an interfacial electron transfer process and increase the separation effect [18,22,23,24]. Cobalt oxides have received attention because of their excellent photo-catalytic activity in carbon dioxide reduction, oxygen reduction, and environment restoration [25,26]. Based on its properties of outstanding photocatalytic activity and low cost, cobalt oxide becomes a feasible material to fabricate the heterojunction structure with other semiconductor photocatalysts. In this work, the Co₃O₄-TiO₂ nanorod arrays were synthesized by photochemical deposition (a green method). The performance of Co₃O₄-TiO₂ nanorod arrays could be controlled by adjusting the pH value and the concentration of the Co precursor.

2. Experimental

2.1. Preparation of TiO₂ Nanorod Arrays

The TiO₂ nanorod arrays on Fluorine doped tin oxide (FTO) were fabricated through the modified hydrothermal method [7,8]. Typically, 30 mL HCl (6 mol/L) was mixed with 0.4 mL tertrabutyl titanium by strong stirring for 10 min. Then, the above solution was transferred into a Teflon pot, in which an FTO glass electrode with the coated layer facing down was placed against the wall of the Teflon pot. The hydrothermal synthesis was carried out at 150 °C for 6 h in an oven. Then, the TiO₂ nanorod arrays was washed with high pure water, and dried by high pure N₂.

2.2. Fabrication of Ultra Small Co₃O₄ Coated TiO₂ Nanorod Arrays

The Co₃O₄-coated TiO₂ nanorod arrays were prepared by photo-chemical deposition in a 1:1 ethanol/water solution containing different concentrations of Co(NO₃)₂ under Xe lamp with a powder intensity of 100 mW/cm². The driving force of the Co deposition on the TiO₂ nanorod arrays was high energetic photons, which could photo-excite electrons from the valance band of TiO₂ to the conductor band and leave holes on the valance band. The holes were depleted by the hole receptor of ethanol. The electrons can reduce the cobalt ions onto the surface of TiO₂ nanorod arrays. Nano cobalt metal was stable and easily oxidized to Co₃O₄.

2.3. Photo-Electrochemical Studies of the Co₃O₄ Modified TiO₂ Nanorod Arrays

To study the photo-electrochemical response of the Co₃O₄-modified TiO₂ nanorods and pure TiO₂ nanorod arrays (1.0 cm × 1.0 cm), open circuit potential (OCP) and the ampere-metric method (I–T) were conducted in 0.1 mol/L Na₂SO₄ solution at room temperature after being deoxidized by high pure N₂ for 15 min.

2.4. The Photo-Electrochemical Degradation Research

Methylene blue and hydroquinone were used to test the photo-electrochemical activity of Co₃O₄ modified TiO₂ nanorod arrays and pure TiO₂ nanorod arrays. The Co₃O₄ modified TiO₂ and pure TiO₂ nanorod arrays with a geometric area of 1.0 cm × 1.0 cm were used as the working electrode, as well as platinum as the counter electrode and saturated calomel electrode (SCE) as the reference electrode. A 500 W Xe lamp was used to simulate sunlight with a powder intensity of 100 mW/cm². A solution containing 10 mg/L methylene blue or hydroquinone, 0.1 mol/L Na₂SO₄, and 10 mmol/L H₂O₂ was used as the investigated subject. During the photo-electrochemical degradation process, the electrode was added with 1.0 V bias potential (vs. SCE) and was light-illuminated to degrade methylene blue or hydroquinone. To avoid the effects of solution temperature, the reaction system was placed in a constant temperature system with a circulating water device (Beijing LabTech Instruments Co., Ltd., Beijing, China). The photo-electrochemical degradation process of methylene blue and hydroquinone was measured by UV/vis spectrum.

3. Results and Discussions

3.1. XRD Analysis

The Co₃O₄-TiO₂ and TiO₂ nanorod arrays were investigated by X-ray diffraction measurement (XRD) using copper target to identify the phase of all samples. The Co₃O₄-TiO₂ and TiO₂ samples in Figure 1 show the rutile phase of TiO₂ (JCPDS No1-1292). No new XRD peaks of Co₃O₄ were observed for the Co₃O₄-TiO₂ nanorod arrays, which might because of the small amount of cobalt oxide in the Co₃O₄-TiO₂ hybrid catalyst.

3.2. Scanning Electron Microscopy (SEM) and Transmission Electron Microscopy (TEM) Measurement

The morphologies and microstructures of the TiO₂ nanorod arrays and Co₃O₄-TiO₂ nanorod arrays were characterized by scanning electron microscopy (SEM). Figure 2 shows the top view SEM images of the TiO₂ and Co₃O₄-TiO₂ nanorod arrays. Highly ordered and large scale TiO₂ nanorod arrays are vertically aligned on both pure TiO₂ and Co₃O₄-TiO₂ nanorod arrays in Figure 2.

The average diameter of TiO₂ nanorod is about 180 nm with a rectangular cross section. There are almost no different features between Co₃O₄-TiO₂ nanorod arrays and pure TiO₂ nanorod arrays, which might because of the small amount of Co element in the Co₃O₄-TiO₂ nanorod arrays. The small amount of Co element could be confirmed by Energy Dispersive X-ray (EDX) in Supplementary Materials Figure S1. Compared with the super strong signal of Ti and O element, the signal of Co element was very low, which means the cobalt content was very low.

Figure 3a–d show the TEM images of TiO₂ nanorods and Co₃O₄-TiO₂ nanorod arrays, respectively. It can be clearly seen that the TiO₂ nanorod arrays have the dimension of 4 nm with a clearly lattice structure. Figure 3b is the high resolution of TEM image of TiO₂. Compared with the clear lattice structure of pure TiO₂ nanorods, the surface of Co₃O₄-TiO₂ nanorod arrays (Figure 3c) was covered with something similar to fog, which makes the lattice structure of Co₃O₄-TiO₂ nanorods not obvious. High resolution TEM of Co₃O₄-TiO₂ nanorod in Figure 3d showed that there were some amorphous Co₃O₄ on the surface of TiO₂.

3.3. X-ray Photoelectron Spectroscopy (XPS) Analysis

To confirm the composition of Co₃O₄-TiO₂ nanorod arrays sample, the XPS method was used to study the chemical composition and valence state of Co₃O₄-TiO₂. The survey spectra illustrated in Supplementary Materials Figure S2 demonstrates the existence of Co, Ti, and O elements. Figure 4a shows the XPS spectrum of O _1s. Figure 4b shows the XPS spectrum of Ti 2p obital of the Co₃O₄-TiO₂ nanorod array. The Ti 2p_3/2 and 2p_1/2 located at 458.4 eV and 464.1 eV can be assigned to Ti⁴⁺, which coincided with TiO₂. The band energy of 780.50 eV and 797.43 eV in Figure 4c corresponded to Co 2p_3/2 and Co 2p_1/2, respectively. The peaks are the typical signature of Co₃O₄ and are consistent with the previous literature [27]. Compared with the strong intensity of the Ti and O element, the XPS spectra strength of cobalt was very weak, which meant the amount of cobalt element in Co₃O₄-TiO₂ nano-materials was small.

The experimental data were in accordance with the EDX results in Supplementary Materials Table S1. The amount of Co was small, because of the high activity of cobalt metal. Cobalt element is more vivacious than hydrogen element, which means there is a competitive reaction between cobalt ions and hydrogen ions during photochemical deposition. The competition between cobalt ions and hydrogen ions decreases the amount of cobalt deposition on the TiO₂ surface. Furthermore, the Co metal easily dissolved into Co ions under acid circumstance. Therefore, the competitive reaction and the instability of cobalt lead to the ultra-small amount of Co on the TiO₂ nanorod arrays.

3.4. Photocurrent Test

The transient photocurrent was further used to confirm the generation, transfer, and separation processes of the photo-induced electrons and holes on both Co₃O₄ modified TiO₂ nanorod arrays and the pure TiO₂ nanorod arrays. To illustrate the effect of Co₃O₄ on photocatalytic activity of Co₃O₄-TiO₂ nanorod array, the photocurrent response of the TiO₂ nanorod arrays and Co₃O₄-TiO₂ nanorod arrays was measured by chopping light. The curves of both Co₃O₄-TiO₂ and TiO₂ samples in Figure 5a had outstanding responses to chopping light cycles. The current value of Co₃O₄-TiO₂ is five times higher than pure TiO₂ nanorod arrays. This means that the Co₃O₄-TiO₂ nanorod arrays had higher photoelectrocatalytic activity than the pure TiO2 nanorod arrays.

Compared with pure TiO₂ nanorod arrays, the Co₃O₄-TiO₂ nanorod arrays exhibited an obviously higher photocurrent, which indicated that Co₃O₄-TiO₂ nanorod arrays had the higher photo-electrochemical activity. Under light illumination, the electron in the valence band was excited to the conductor band and left a hole in the valence band. The electrons were accumulated on the conductor band and holes were assembled on the valence band by persistent light illumination [28]. The relative OCP value of the samples was measured to compare the performance in a different semiconductor. In comparison, the relative OCP value of Co₃O₄-TiO₂ was higher than the pure TiO₂ nanorod arrays, which demonstrated that separation of e-h⁺ pairs in the Co₃O₄-TiO₂ heterojunction is significantly improved by the addition of Co₃O₄ (Supplementary Materials Figure S3).

3.5. The Formation Mechanisms of Co₃O₄ Nanoparticles

Further observation found that Co₃O₄-TiO₂ nanorods fabricated in different pH values had diverse responses to visible light, as shown in Figure 5b. The Co₃O₄-TiO₂ nanorods fabricated under pH 4.12 had the highest photo-electrochemical response compared with others fabricated at pH 7.82 and 7.00. The formation mechanisms of Co₃O₄ nanoparticles in different pH solutions were different, which led to a different response to visible light. In neutral solution, the Co₃O₄ were formed under photochemical deposition.

$$3\mathrm{Co}-{8\mathrm{e}}^{-}{+4\mathrm{H}}_2{\mathrm{O}=\mathrm{Co}}_3{\mathrm{O}}_4{+8\mathrm{H}}^{+}$$

(1)

Co nanoparticles were easily deposited onto the semiconductor in alkaline solution. Such as Co-ZnO was fabricated, which had high catalytic activity in oxygen production [29]. And the high catalytic activity of Ni-CdS nanorods was fabricated through photochemical deposition in NaOH solution containing Ni ions and methanol [30]. Holes have high energy, which could oxidize methanol adsorbing on the semiconductor. Considering the high concentration of ethanol, ethanol was oxidized to formaldhyde and holes were decomposed. It is very difficult to deposit transition metal or metal oxide onto semiconductors in the acid solution because cobalt metal is more active than the hydron element. There were two different competitive reactions during the photodeposition reaction. They are listed below:

$${\mathrm{Co}}^{2+}{+2\mathrm{e}}^{-}=\mathrm{Co}$$

(2)

$$2{\mathrm{H}}^{+}+{2\mathrm{e}}^{-}={\mathrm{H}}_2$$

(3)

Furthermore, the high activity of transition metals meant they easily dissolved in acid solution. The Co metal was not stable because of oxidization reaction and photocorrosion. Considering the above reasons, the amount of Co is very small.

$$\mathrm{Co}-2{\mathrm{e}}^{-}={\mathrm{Co}}^{2+}$$

(4)

$$3\mathrm{Co}-8{\mathrm{e}}^{-}+4{\mathrm{OH}}^{-}={\mathrm{Co}}_3{\mathrm{O}}_4+{4\mathrm{H}}^{+}$$

(5)

In alkaline solution, the Co ions mainly existed through Co(OH)₄²⁻, which was reduced to Co metal by electrons photo-excited on the TiO₂ surface. The Co metal was not stable and could easily be oxidized to Co₃O₄ by oxygen or OH radical produced in photochemical deposition.

$$\mathrm{Co}{\left(\mathrm{OH}\right)}_4{}^{2-}+{2\mathrm{e}}^{-}={\mathrm{Co}+4\mathrm{OH}}^{-}$$

(6)

The amount of Co₃O₄ can be controlled through regulating the concentration of Co ions from 0.1 mmol/L to 1 mmol/L. We found that Co₃O₄-TiO₂ nanorod array fabricated in 0.5 mmol/L cobalt ions had the highest photocatalytic activity. It can be clearly seen that the photocurrent increased with the concentration of Co ions increasing from 0.1 mmol/L to 0.5 mmol/L (in Figure 6). However, the photo-electrochemical currents decreased when the Co ions’ concentrations were further increased, which may be because of the formation of a thick Co₃O₄ layer and increase in the carrier recombination rate.

3.6. Photo-Electrochemical Activity of Co₃O₄ Modified TiO₂ Nanorod Arrays

Methylene blue was used as the probe to evaluate the photo-electrochemical activity of pure TiO₂ nanorod arrays and Co₃O₄ modified TiO₂ nanorod arrays. As shown in Figure 7, the TiO₂ nanorod array exhibits a moderate catalytic performance for the photo-electrochemical degradation of methylene blue, which could have contributed to the ordered arrays effect. The Co₃O₄ modified TiO₂ nanorod arrays exhibit excellent photo-electrochemical degradation activity to methylene blue. TiO₂ nanorod arrays modified with Co₃O₄ had good photochemical catalytic activity, which was attributed to the disjunction of Co₃O₄-TiO₂ and the charge transfer between Co₃O₄ and TiO₂ nanometer rod components. Furthermore, the isolated Co₃O₄ islands on the surface of TiO₂ nanorod arrays acted as reactive sites to enhance the photo-electrochemical activity. The degradation of methylene blue by Co₃O₄ modified TiO₂ was greater than that of pure TiO₂ nanorod arrays. Methylene blue was totally degraded on Co₃O₄ modified TiO₂ nanorod arrays. Compared with Co₃O₄-TiO₂ nanorod arrays, only 60% methylene blue was degraded on pure TiO₂ nanorod arrays.

The TiO₂ nanorod arrays only absorbed ultraviolet photons to generate e-h⁺ pairs to degrade the methylene blue molecule, while Co₃O₄-TiO₂ nanorod arrays provide a p-n junction between Co₃O₄ nanoparticles and TiO₂ nanorod arrays. The enhancement of photo-electrochemical activity of Co₃O₄/TiO₂ heterojunction samples could have contributed to the formation of the type-II p-n hetero-junction between Co₃O₄ and TiO₂. TiO₂ is an n-type wide band gap semiconductor with the conduction band at 0.14 V, and Co₃O₄ is a p-type narrow band gap with band energy of 2.07 eV [31]. When Co₃O₄ nanoparticles are deposited onto the surface of TiO₂ nanorod arrays, a p-n heterojunction can be formed at the surface of TiO₂ nanorod arrays, and the electrons can be transferred from the Co₃O₄ to TiO₂ nanorod arrays until their Fermi levels are equal [32]. The equilibrium can be broken by methylene blue, which acted as a holes receptor. With the photo-electrochemical reaction going on, methylene blue was degraded. The positive potential was loaded onto the Co₃O₄-TiO₂ electrode to further increase the efficiency of the separation of hole and electrons. The Co₃O₄ has Co²⁺ and Co³⁺ valence state. Co²⁺ and Co³⁺ can be easily oxidized to Co⁴⁺, and Co⁴⁺/Co³⁺ had high activity to oxidize methylene blue to water and carbon dioxide. Thus, Co₃O₄-TiO₂ had higher photo-electrochemical activity than pure TiO₂ nanorod arrays.

Hydroquinone was used to further study the photo-electrochemical activity of Co₃O₄ modified TiO₂ and pure TiO₂ nanorod arrays. It can be seen clearly from Figure 8a,b that, compared with pure TiO₂, the Co₃O₄ modified TiO₂ nanorod arrays had higher photo-electrochemical activity. Only 30% hydroquinone was degraded on pure TiO₂ nanorod arrays, and hydroquinone displayed total degradation on Co₃O₄ modified TiO₂ nanorod arrays.

In order to clearly measure the photo-electrochemical activity of Co₃O₄ modified TiO₂ nanorod arrays and pure TiO₂ nanorod arrays, the corresponding kinetic constant was computed through fitting the experimental degradation of hydroquinone by the following equation.

$$-\ln \left(\frac{C_t}{C_0}\right)=kt$$

(7)

where C_t is the concentration of hydroquinone at a certain reaction time, C₀ is the original concentration, k is the apparent first rate, and t is photo-electrochemical time. The model was suitable for the photo-electrochemical hydroquinone degradation process. The k-values for the Co₃O₄-TiO₂ and pure TiO₂ nanorod arrays are 0.91745 and 0.1206, respectively. The k-value of Co₃O₄-TiO₂ is almost eight times that of the pure TiO₂ nanorod arrays, which further confirmed that Co₃O₄ addition greatly enhanced the photo-electrochemical activity.

3.7. Photo-Electro-Catalytic Degradation Mechanism

The photo-electrochemical activity of the photo-catalyst was mainly determined by light absorption, charges holes separation, and charge transfer from the inner to the surface of catalyzer. The band gap of the semiconductor was the key factor, which had a great influence on the photo-activity of the catalyst. The bandgap of Co₃O₄-TiO₂ nanorod arrays and pure TiO₂ nanorod arrays was determined from UV/vis diffuse reflectance spectra using the Tauc function, as shown in Figure 9a. The bandgap of TiO₂ nanorod arrays is 3.2 eV and is very close to that from the former literature [33]. The bandgap of Co₃O₄-TiO₂ is 2.85 eV, which could have contributed to the intrinsic narrow band gap of Co₃O₄ hybrid with TiO₂. The addition of Co₃O₄ reduced the band gap of the nanomaterial and then enhanced its catalytic activity.

The influence of Co₃O₄ on the energy level of the photo-catalyst was studied through the Mott–Schottky electrochemical method in 0.1 mol/L Na₂SO₄ with 1000 Hz for Co₃O₄-TiO₂ and pure TiO₂ nanorod arrays. Both TiO₂ and Co₃O₄-TiO₂ had positive slopes, meaning that both pure TiO₂ and Co₃O₄-TiO₂ are n-type semiconductors and electrons as the majority carriers. It can be seen that the addition of Co₃O₄ did not change the semiconductor type of TiO₂, but greatly changed the Fermi level and the flat band potential of TiO₂. This phenomenon is consistent with other reported work that the addition of cobalt would change the band energy of the nanomaterial [34]. The flat-band potential (V_b) and the carrier density of nano-materials can be calculated according to the following equation:

$${\mathrm{N}}_{\mathrm{D}}=\frac{2}{e\epsilon 0\epsilon }\left(\frac{dE}{d\left(\frac{1}{C^2}\right)}\right)$$

(8)

where $e_0$ is the dielectronic constant of the material, $\epsilon$ is the permittivity of the vacuum, e is the element charge, N_D is the donor density, and C is the capacitance. From the slope in the plot of 1/C² versus V in the Figure 9b, the smaller slope for the Co₃O₄-TiO₂ reflects a higher electron donor density. The higher N_D means lower resistance, faster charge transfer, and higher electrochemical activity.

The Co₃O₄-TiO₂ nanorod arrays had greatly enhanced photo-electrochemical performance, which could be attributed to the synergetic effects of the formation of a p-n junction between Co₃O₄ and TiO₂. The catalytic performance of semiconductor nanomaterials depends on the bandgap of the semiconductor nanomaterial, the separation of electrons and holes, and the lifetime of electrons and holes generated by photo exciting. The hole and electron can move the surface to react with the adsorbed reactant. However, the electron and hole could recombine easily in a short time, which greatly abates the activity of the catalyst. Therefore, the catalyst’s activity can be greatly influenced by the life-time of the photo-induced electron-holes pairs. The p-n junction could greatly enhance the life-time of the electron and holes. The longer life-time of holes and electrons greatly enhances the activity of catalysts. Co₃O₄ is a p-type semiconductor with a band gap of 2.19 eV [31], while TiO₂ is an n-type semiconductor with a band gap of 3.2 eV [1]. Thus, Co₃O₄ participating in the TiO₂ nanorod arrays could change the structure of TiO₂ in three aspects: (1) broaden the absorption range from ultraviolet light to visible light; (2) form a p-n junction to enhance the life-time of the electron; or (3) the disjunct Co₃O₄ nanoparticles act as an activation point to improve the photo-electrochemical activity. The conduction band (CB) position of TiO₂ is more anodic than Co₃O₄, so the excited electrons on the CB of TiO₂ could not transfer to Co₃O₄, while the holes could transfer from TiO₂ to Co₃O₄. The recombination of electron and hole could be reduced just as shown in Scheme 1. At the hetero-junction in thermal equilibrium, the p-type and n-type regions have completely opposite charges, and the n-type regions become positive, while the p-type region becomes negative. When the n-p hetero-junction semiconductor is excited by visible light with high energy to band gap, the photo-generated electrons can move to the CB of the n-type TiO₂ and holes can move to the VB of the n-type semiconductor for the formation of the inner electric field in the Co₃O₄/TiO₂ sample, which effectively impedes the recombination of electron–hole pairs. The biased voltage could further restrain the photo-excited electrons and holes recombination through the following mechanism. The positive bias voltage depletes the electrons and, as a result, the holes can be excluded to the surface. Then, the absorbed molecules on the surface of Co₃O₄ can react with the holes to form a series of radicals, such as OH radical and other radicals. These radicals have great energy to oxidize the organic waste to water and carbon dioxide.

$${\mathrm{Co}}_3{\mathrm{O}}_4+\mathrm{visible}\mathrm{light}\to {\mathrm{Co}}_3{\mathrm{O}}_4\left({\mathrm{e}}^{-}\right)+{\mathrm{Co}}_3{\mathrm{O}}_4\left({\mathrm{h}}^{+}\right)$$

(9)

$${\mathrm{TiO}}_2+\mathrm{bias}\mathrm{voltage}\to {\mathrm{TiO}}_2\left({\mathrm{h}}^{+}\right)$$

(10)

$${\mathrm{Co}}_3{\mathrm{O}}_4+\mathrm{bias}\mathrm{voltage}\to {\mathrm{Co}}_3{\mathrm{O}}_4+\mathrm{holes}$$

(11)

$${\mathrm{H}}_2{\mathrm{O}}_2+\mathrm{hole}\to 2\mathrm{OH}$$

(12)

$$\mathrm{MB}+2\mathrm{OH}\to {\mathrm{CO}}_2+{\mathrm{H}}_2\mathrm{O}$$

(13)

According to the above results and discussion, consequently, under external bias voltage, electrons transfer along the external wires to auxiliary electrode and leave holes on the surface of Co₃O₄, which could oxidize the organic waste to water and carbon dioxide, as the schematic shows.

4. Conclusions

Ultra small amounts of Co₃O₄-modified TiO₂ nanorod arrays were successfully fabricated thorough green photochemical deposition methods without adding any nocuous reagents. The Co₃O₄/TiO₂ nanorod arrays fabricated in acid solution had the highest photo-electrochemical activity. We elaborated on the mechanism of Co₃O₄-TiO₂ fabricated in different pH value solutions. The amount of Co₃O₄ could be controlled by adjusting the concentration of Co ions. The small amount of Co₃O₄ made many disjunct active points, which acted as active sites to mineralize organic wastes during photoelectrochemical degradation. The Co₃O₄/TiO₂ nanorod arrays had higher photo-electrochemical activity to degrade organic waste than pure TiO₂ nanorod arrays, which had great potential in waste water treatment.