Synthesis of p-Co₃O₄/n-TiO₂ Nanoparticles for Overall Water Splitting under Visible Light Irradiation.

p-Co₃O₄/n-TiO₂ nanoparticles (~400 nm) for photocatalysis were prepared via carbon assisted method and sol-gel method in this work. The paper also studied the application of visible light illuminated p-Co₃O₄/n-TiO₂ nanocomposites cocatalyst to the overall pure water splitting into H₂ and O₂. In addition, the H₂ evolution rate of the p-Co₃O₄/n-TiO₂ nanocomposites is 25% higher than that of the pure Co₃O₄ nanoparticles. Besides, according to the results of the characterizations, the scheme of visible light photocatalytic water splitting is proposed, the Co₃O₄ of the nanocomposites is excited by visible light, and the photo-generated electrons and holes existing on the conduction band of Co₃O₄ and valence band of TiO₂ have endowed the photocatalytic evolution of H₂ and O₂ with higher efficiency. The optimal evolution rate of H₂ and O₂ is 8.16 μmol/h·g and 4.0 μmol/h·g, respectively.

1. Introduction

The shclass="Chemical">arclass="Chemical">p increase in global energy consumclass="Chemical">pan class="Chemical">ption makes efficient utilization of solar energy more urgent [1]. Therefore, overall water splitting under visible light has received much attention for production of renewable hydrogen from water [2]. To improve the efficiency of the hydrogen production, researchers work hard on modifying the nanomaterials [3,4,5,6]. Moreover, doping rare metals on semiconductor nanomaterials, changing the morphology of the nanomaterials and synthesis of complex nanomaterials are hot means that can be employed to improve the photocatalytic activity [7,8,9,10,11,12,13,14,15,16,17,18]. Maeda studied the photocatalytic activity of Rutile TiO2 doped by Ru, Rh, Ir, Pt or Au, with the results showing that the most water splitting amount of H2 and O2 for 4 h is 56.6 μmol and 26.5 μmol, respectively, when Pt doping amount is at 1 wt. % [7]. The water splitting amount of H2 for 8 h is 2750 μmol, which is photocatalyzed by 3D ZnO microspheres prepared by Guo [8]. Co3O4 Quantum Dots show excellent performance on photocatalytic water splitting that the water splitting amount of H2 and O2 is 0.79 μmol/h and 0.4 μmol/h, respectively [18]. Meanwhile, some nanomaterials (La2Ti2O7, PbTiO3, SrTiO3, etc.) also play important roles in water splitting [9,10,11,12,13,14,15,16,17]. The water splitting amount of H2 is 166.67 μmol/h·g when Au and reducing graphene oxide doped La2Ti2O7 act as photocatalyst [11]. However, the methods of synthesizing the photocatalyst mentioned above are complex, along with high cost of doping Pt. Therefore, this paper aims to design a photocatalyst with relatively high activity via easy methods and low cost.

Many reseclass="Chemical">arches focus on enhancing the visible light absorclass="Chemical">pan class="Chemical">ption of TiO2, but the main problems of these methods are high cost, complex processes and compositions [19,20,21,22]. Band gap of the Co3O4 nanomaterials is very close to the wavelength of visible light, so Co3O4 nanomaterials can be excited by visible light. However, typically, the p type Co3O4 semiconductor cannot be used on overall water splitting [23]. Some studies researched on changing the band gap of Co3O4; nevertheless, the constraints of these methods are high cost and difficult controlling processes [12,13,14,15,16,17,24]. Therefore, a new p-Co3O4/n-TiO2 photococatalyst is designed for overall pure water splitting under visible light irradiation, and the Co3O4 and TiO2 nanoparticles is, to the best of our knowledge, combined in nano-scale for the first time [25,26,27]. According to the band edge positions of TiO2-Co3O4 nanocomposites, the excited electrons on the conduction band of the p-type Co3O4 transfer to that of the n-type TiO2, and simultaneously holes on the valence band of n-TiO2 can be transferred to that of p-Co3O4 under the potential of the band energy difference. The migration of photogenerated carriers can be promoted by the internal field, which results in existence fewer barriers. Therefore, the electron–hole pairs recombination of can be reduced, and the photocatalytic reaction can be improved greatly [28,29].

Many efforts have been devoted to synthesizing class="Chemical">Co3O4 with well-controlled dimensionality, sizes, and crystal structure [30,31,32,33]. Wang et al. reclass="Chemical">ported a class="Chemical">pan class="Chemical">carbon-assisted carbothermal method to synthesize the single-crystalline Co3O4 octahedral cages with tunable surface aperture [30]. Moreover, with the carbon spheres obtained through hydrothermal carbonization as the sacrificial template, Du et al. have successfully synthesized Co3O4 hollow spheres by a one-pot calcinations method [31]. Furthermore, Zhang et al. described the synthesis of high purity octahedral Co3O4 with the help of carbon materials using one-step microwave reaction [33]. Based on the researches mentioned above, p-Co3O4 nanoparticles are prepared through a more facile and environment-friendly carbon-assisted method using degrease cotton, which have been reported by our group in 2015 [23]. TiO2 nanoparticles were composed via gol-sel method [34]. Besides, the paper presented the study on the application of visible light illuminated p-Co3O4/n-TiO2 nanocomposites cocatalyst to the overall pure water splitting into H2 and O2, and the H2 evolution rate of the p-Co3O4/n-TiO2 nanocomposites is 25% higher than that of the pure Co3O4 nanoparticles. In addition, the scheme of visible light photocatalytic water splitting is proposed based on the results of the characterizations.

2. Experimental Section

2.1. Synthesis

All chemicals were reagent grade and used without further purification. class="Chemical">Cobalt nitrate, class="Chemical">pan class="Species">tetra-butyl ortho-titanate (TBOT), ethanol, hydrochloric acid and nitric acid were purchased from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China). Commercial degreasing cotton (Pagoda Medical Devices Co., Ltd., Dingzhou, China) was used as the reactant. Deionized water of 18.25 MΩ was purified through an ultra-pure (UPR) system (Xi'an You Pu Equipment Co., Ltd., Xi'an, China).

The class="Chemical">Co3O4 nanoclass="Chemical">pclass="Chemical">pan class="Chemical">articles (0.06 mol) were prepared via environmentally friendly carbon-assisted method, as reported in the previous article [23]. Specifically, immersed into 20 mL Co(NO3)2 pink solution (3 mol/L), 1.5 g degreasing cotton was then kept in an ultrasonic bath for 10 min in order to get a good dispersion of Co2+ on the surface of degreasing cotton. Then, the treated degreasing cotton was collected and transferred into a quartz petri dish in the tube furnace (OTF-1200X-III, Hefei Ke Jing Materials Technology Co., Ltd., Hefei, China) and kept at 600 °C for 2 h in the air. Besides, the Co3O4 powders were cooled to the room temperature naturally.

The class="Chemical">Co3O4 nanoclass="Chemical">pclass="Chemical">pan class="Chemical">articles were composed with 0.06 mol TiO2 through the following steps: The Co3O4 powders were put into a beaker with solation whose volume ratio of TBOT, nitric acid, alcohol and deionized water were 30.3:0.6:12.5:1 with continuous stirring. Moreover, a multifunctional magnetic stirrer (MPL-CJ-88) was used to stir and heat the solution (Jintandadi Automation Factory, Jintan, China). With the help of ultrasonic bath (KQ-2500DE) from Kunshan Shumei Ultrasonic Instrument Co. Ltd. (Kunshan, China), the beaker of solution and 10 g of particles (the obtained TiO2 solution is about 150 mL) were ultra-sonicated, after which it was stirred for 30 min. The wet particles were heated at 500 °C for 2 h after being reacted for three hours. After milling with an agate mortar, the TiO2-Co3O4 composite was finally obtained.

2.2. Characterization

The crystal structure of the sample was measured through X-ray diffraction (XRD, a Bruker D8, λ = 1.5406 Å) (Bruker (Beijing) Technology Co., Ltd., Beijing, China) in the 2θ of 10°–80° with a scan rate of 10°/min and Cu Kα radiation, at 40 KV. Chemical composition analysis was cclass="Chemical">arried out using X-ray class="Chemical">photoelectron sclass="Chemical">pectroscoclass="Chemical">py (XPS). This XPS was collected using an ESCALAB 250Xi sclass="Chemical">pectrometer (Shanghai Hu Yueming scientific instruments Ltd., Shanghai, China) with a standclass="Chemical">pan class="Chemical">ard Al Kα radiation which was provided with the binding energies calibrated based on the contaminant carbon (C1s = 284.6 eV). The morphology was observed by transmission electron microscopy (TEM, JEOL JEM-2011) (Guangzhou Office of Japan Electronics Co., Ltd., Guangzhou, China). Furthermore, with the presence of BET (Brunauer-Emmet-Teller) and BJH (Barrett-Joyner-Halenda), specific surface areas and pore size distributions were computed from the results of N2 physisorption at 77 K (Micromeritics ASAP 2020) (Micromerics (Shanghai) Instrument Co., Ltd., Shanghai, China). A Cary 300 Scan Ultraviolet–visible (UV-Vis) spectrophotometer (Shanghai Precision Instrument Science Co., Ltd., Shanghai, China) was employed to record the UV-Vis diffuse reflectance spectra (DRS) in a region of 200 to 800 nm.

2.3. Photocatalysis

The photocatalytic activity for the overall pure class="Chemical">water sclass="Chemical">plitclass="Chemical">pan class="Chemical">ting into H2 and O2 was estimated under visible light condition. The diagram of visible light water splitting system is shown in Figure 1.

Figure 1

Diagram of visible light water splitting system.

Typically, 0.02 g photocatalyst was added into the solution (~200 mL) containing 200 mL pure class="Chemical">water solution. After class="Chemical">put in an ultrasonic stirring for 20 min, class="Chemical">purged by class="Chemical">pan class="Chemical">Ar gas for 20 min, the mixture was then irradiated under visible light with magnetic stirring. A 300 W Xe arc lamp (LSH-A500, KaifengHxsei Science Instrument Co. Ltd., Kaifeng, China) with UV cut-off filters (420 nm) was used as the light source. In addition, the hydrogen produced was analyzed by a gas chromatography (GC-9890B, Shanghai Linghua Instrument Co. Ltd., Shanghai, China) equipped with a thermal conductivity detector and a stainless steel column packed with molecular sieve (5 A). Ar gas (99.999%) was used as the carrier gas.

3. Results and Discussion

The XRD patterns of the class="Chemical">Co3O4 class="Chemical">pan class="Chemical">are shown in Figure 2a. All peaks have a good agreement with the standard spinel cubic Co3O4 spectrum (JCPDS No. 42-1467), while there are no impurity peaks found in the XRD patterns. The result suggests that the well-crystallized Co3O4 with high purity sample is produced. According to Scherrer’s formula, D = 0.89λ/(Bcosθ) (where D is the average dimension of crystallites; λ is the X-ray wavelength; θ is the Bragg Angle; and B is the pure diffraction broadening of a peak at half-height, which is calculated according to the data of XRD spectrum), the crystalline size of Co3O4, calculated from the strongest peak, locating at (311) plane, are estimated to be 51.64 nm.

Figure 2

X-ray diffraction (XRD) pattern of the: (a) Co3O4 sample; (b) TiO2 sample; and (c) TiO2-Co3O4 sample. a.u.: any unit.

Figure 2b demonstrates that anatase (JCPDS No. 21-1272) with high purity sample is obtained via gol-sel method. According to Scherrer’s formula mentioned above, the crystalline size of class="Chemical">TiO2 is estimated to be 22.48 nm. The wider class="Chemical">peaks in Figure 2c have shown bigger comclass="Chemical">posites sizes. To confirm all of the nanoclass="Chemical">pclass="Chemical">pan class="Chemical">articles contain both Co3O4 and TiO2, the TiO2 are compounded after the obtaining of bigger size Co3O4. In addition, avoiding the visible light absorber Co3O4 being coated completely, the mol ratio of these two materials is controlled strictly as 1:1. As demonstrated in Figure 2c, there are no facets existing other than Co3O4 and TiO2. However, the Co3O4 peaks are much weaker than that of TiO2, with the possible reason being that the Co3O4 is partially coated by TiO2.

To identify the chemical state of the nanocomposite, the X-ray photoelectron spectroscopy (XPS) was measured (Figure 3). By using adventitious class="Chemical">carbon at 284.8 eV, the XPS sclass="Chemical">pectra were corrected for samclass="Chemical">ple chclass="Chemical">pan class="Chemical">arging. The Co2p orbital showed splitting peaks at 794.7 and 779.6 eV, representing Co2p 3/2 and Co2p 1/2 [35]. The Ti2p orbital showed peaks at 464.0 and 458.2 eV, thus indicating the Ti2p 3/2 and Ti2p 1/2 [36,37]. As for the XPS of O1s at 531.6 and 529.5 eV, it would indicate the presence of adsorbed water and oxygen in the near-surface region [13]. The strong peak of C1S centers at 284.5 eV can be assigned to elemental carbon, which has given rise to the incomplete burning of degreasing cotton; in contrast, the other two peaks appeared at 285.8 and 288.6 eV, respectively, which are ascribed to the O=C–O bonds and (CHO) from insufficient combustion residual degreasing cotton [38]. No Co–C or Ti–C band is found in the spectra, which means that the conjunctions of the nanocomposites are not affected by C. Therefore, according to the results of the XPS spectra, the TiO2-Co3O4 sample is composed only by TiO2, Co3O4 and residual degreasing cotton, which is consisted with the XRD results.

Figure 3

X-ray photoelectron spectroscopy (XPS) spectra of the TiO2-Co3O4 sample.

The morphology and structure of the samples class="Chemical">are observed via TEM, with the results class="Chemical">presented in Figure 4. Figure 4a shows that the class="Chemical">pan class="Chemical">Co3O4 nanospheres with a relatively uniform size of 60 nm are obtained. It can be seen that 20 nm TiO2 nanoparticles compose the big group in Figure 4b, which is consistent with the calculation from XRD mentioned above. The Figure 4c is the TEM image of the TiO2-Co3O4 sample. The darker part in Figure 3c is Co3O4 while the lighter part is TiO2, which can be proved by both the synthetic procedures and the results of High Resolution Transmission Electron Microscopy (HRTEM) in Figure 4d–f. The TiO2 accumulates around the Co3O4 incompletely, while the TiO2 and the Co3O4 connected closely at the interface of the composite. The result of Figure 2, Figure 3 and Figure 4 improves the elements and framework of the p-Co3O4/n-TiO2 nanocomposites.

Figure 4

Transmission electron microscopy (TEM) images of: (a) the Co3O4 sample; (b) TiO2 sample; and (c) TiO2-Co3O4 sample. High Resolution Transmission Electron Microscopy (HRTEM) images of: (d) the Co3O4 sample; (e) TiO2 sample; and (f) TiO2-Co3O4 sample.

The class="Chemical">Co3O4 is class="Chemical">preclass="Chemical">pclass="Chemical">pan class="Chemical">ared firstly due to its clear edge and bigger size, while the mesoporous structure of the TiO2 (which is shown as Figure 5) is another reason for the preparation of its designing on step two. Larger area of the conjunction and larger surface area of the composite for higher photocatalytic efficiency can be obtained in this way. Brunauer-Emmett-Teller (BET) surface areas and Barrett-Joyner-Halenda (BJH) pore size distributions of the TiO2-Co3O4 are shown in Figure 5. According to the International Union of Pure and Applied Chemistry (IUPAC) classification, the isotherms exhibit type IV. As for the increase in the uptake of N2 at intermediate pressure, which suggests the existence of mesoporous resulted from the interparticle space in the samples, it can facilitate the water accessibility to nanoparticles. According to the corresponding BJH pore size distribution curve, the pore size distribution has a relatively intense peak at ~10 nm. The BET surface area is calculated to be 39.64 m2/g and the average pore size is 3.83 nm. These results and the results of the TEM demonstrate the mesoporous existence on the TiO2 layer of the composites. Furthermore, the relative higher surface area and the mesoporous structure will play a very important role in improving the water splitting efficiency of the composite.

Figure 5

(a) Nitrogen adsorption–desorption isotherms; and (b) Barrett-Joyner-Halenda (BJH) pore size distributions of TiO2-Co3O4 sample.

Investigation was conducted on the oclass="Chemical">ptical absorclass="Chemical">pan class="Chemical">ption properties of the TiO2 nanoparticles, Co3O4 nanoparticles and TiO2-Co3O4 heterostructures at room temperature by UV-Vis spectroscopy, as shown in Figure 6a. Broad background absorption in the visible light region can be observed for Co3O4 (Figure 6a, curve 1), while there is no absorption invisible light region for TiO2 (Figure 6a curve 2). However, the absorption of the TiO2-Co3O4 composite in the visible light region can achieve great improvement owing to the Co3O4 (Figure 6a curve 3). This result illustrates that the TiO2-Co3O4 composite can be irradiated by visible light. Importantly, Figure 6b shows the valence-band XPS spectra of TiO2 and Co3O4, clearly indicating that the valence band maximum of TiO2 and Co3O4 are 2.4 and 0.6 eV. In addition, according to that the UV-Vis spectroscopy of Co3O4, the bandgap value calculated is 1.9 eV (Figure 6c) while the bandgap Eg value of TiO2 is 3.2 eV (Figure 6d). Based on the above results, the schematic diagram of the water splitting reaction of the TiO2-Co3O4 heterostructures is shown as Figure 6e. When Co3O4 semiconductor absorbs visible light photons, electrons in the valence band are excited to the conduction band. As a result, excited electrons and holes are generated in the conduction and valence bands of the composite, respectively. These photogenerated carriers drive reduction and oxidation reactions. The reduction of water to hydrogen and oxidation of reduced redox mediators occurs on Co3O4 and TiO2 concurrently with the reduction of oxidized redox mediators and oxidation of water to oxygen on the TiO2 [2]. According to the band edge position, the excited electrons on the conduction band of the p-type Co3O4 transfer to that of the n-type TiO2, and simultaneous holes on the valence band of n-TiO2 can be transferred to that of p-Co3O4 under the potential of the band energy difference. The migration of photogenerated carriers can be promoted by the internal field, so barely any barrier exists. Therefore, the electron–hole pairs recombination can be reduced, and the p-n junction has a significant impact on the efficiency of photocatalytic water splitting.

Figure 6

(a) Ultraviolet–visible (UV-Vis) spectra; and (b) Valence-band XPS spectra of TiO2 and Co3O4 nanoparticles. (Ahv)2–hv curve of: (c) Co3O4 nanoparticles; and (d) TiO2 nanoparticles. (e) Schematic diagram of the water splitting reaction of the TiO2-Co3O4 heterostructures.

The results of the measurements of class="Chemical">H2 evolution through direct class="Chemical">photocatalytic class="Chemical">pan class="Chemical">water splitting with Co3O4 nanocomposites under visible light are shown as Figure 7. The H2 evolution rates are 6.5 μmol/h·g, as shown in Figure 7.

Figure 7

Photocatalytic H2 evolution on Co3O4 nanocomposites under visible-light irradiation using 0.02 g photocatalyst suspended in 200 mL pure water solution in a Pyrex glass cell.

The measurements results of the class="Chemical">H2 and class="Chemical">pan class="Chemical">O2 evolution through direct photocatalytic water splitting with Co3O4-TiO2 nanocomposites under visible light are shown in Figure 8. The H2 and O2 evolution rates are 8.16 μmol/h·g and 4.0 μmol/h·g, respectively, as shown in Figure 8a. The approximated 2:1 of the H2 and O2 generation ratio demonstrate the Co3O4-TiO2 nanocomposites capability for the overall water splitting. Due to the photoreduction of O2 via a reverse reaction during water splitting, the slight deviation of the H2:O2 ratio from the ideal stoichiometric value could thus be acquired. The hydrogen peroxide was formed as an oxidation product, and molecular oxygen (another oxidation product) adsorbed too intimately on the surface of photocatalyst so as to desorb the gas phase, thus resulting in the lack of O2 [7,19]. Through the comparison between Figure 7 and Figure 8a, it can be seen that the H2 evolution rate of the p-Co3O4/n-TiO2 nanocomposites is 25% higher than that of the pure Co3O4 nanoparticles. This result illustrates the mechanism in the way that the electron–hole pairs recombination can be reduced due to the structure of the p-Co3O4/n-TiO2 nanocomposites, while the p-n junction has a significant impact on the efficiency of photocatalytic water splitting. Figure 8b shows that the H2 and O2 evolution rates are 7.67 μmol/h·g and 3.92 μmol/h·g after being recycled five times. The slight reduction of the gas generation is resulted by the quality loss caused by powders cleaning and drying; additionally, the mesoporous structure of the nanomaterials makes the water molecules difficult to be removed thoroughly. Together, the capability of the Co3O4-TiO2 nanocomposites for the overall water splitting is demonstrated by the measurements of H2 and O2 evolution under visible light.

Figure 8

(a) Photocatalytic H2 and O2 evolution on Co3O4-TiO2 nanocomposites under visible-light irradiation using 0.02 g photocatalyst suspended in 200 mL pure water solution in a Pyrex glass cell; and (b) cycling measurements of H2 and O2 evolution through direct photocatalytic water splitting with Co3O4-TiO2 nanocomposites under visible light.

The gas generation rate compclass="Chemical">arison result is not class="Chemical">precise becclass="Chemical">pan class="Chemical">ause the light source and the sacrificial agent are different according to different researches. Therefore, a comparison was made of our results with some of the other works on pure water splitting under visible light. As concluded in Table 1, the gas generation rate of our nanocomposites is in the medium level. However, the preparation method of Co3O4 Quantum Dot [12] and Pt-TiO2 [7] are hard to control, and, needless to say, high cost. Besides, the materials of p-LaFeO3/Fe2O3 [39] are uncommon, and the treating temperature of Co-TiO2 preparation is extremely high (~1100 °C) [19]. The simple preparation method of p-Co3O4/n-TiO2 nanocomposites used in this work is environmentally friendly. In conclusion, the p-Co3O4/n-TiO2 nanocomposites presented in this paper has a relatively high photocatalytic activity in terms of its facilitative methods.

Table 1

Research status of photocatalytic water splitting.

Reference	Materials	Light Source	H2 Amount	O2 Amount	Sacrificial Agent
[8]	3D ZnO microspheres	Visible light	343.75 μmol/h		Methanol, ethanol, formaldehyde
[12]	Co3O4 Quantum Dot	Visible light	0.79 μmol/h	0.4 μmol/h	none
[7]	Pt-TiO2	Ultraviolet light	56.6 μmol/h	26.5 μmol/h	none
[11]	Au@Pt-NLTO/rGO	Visible light	166.67 μmol/h·g		ethanol
[39]	p-LaFeO3/Fe2O3	Visible light	80 μmol/h	40 μmol/h	none
[19]	Co-TiO2	Visible light	11 μmol/h·g	5 μmol/h·g	none
[23]	Co3O4	Visible light	42.5 μmol/h·g		ethanol
This work	Co3O4	Visible light	6.5 μmol/h·g		none
This work	Co3O4-TiO2	Visible light	8.16 μmol/h·g	4.0 μmol/h·g	none

4. Conclusions

In this work, class="Chemical">carbon assisted method and sol-gel method were used to obtain the nanocomclass="Chemical">posite class="Chemical">pan class="Chemical">p-Co3O4/n-TiO2 photocatalyst. Based on the XRD and XPS results, no other phase is generated other than Co3O4 and TiO2. The formation of the p-Co3O4/n-TiO2 conjunction is proven by the TEM and HRTEM investigations of Co3O4, TiO2 and Co3O4-TiO2 nanoparticles. The BET surface area and the average pore size of themes porous structure nanocomposite are 39.64 m2/g and 3.83 nm, respectively, which can be concluded through the Nitrogen adsorption-desorption curve. The schematic of visible light photocatalytic water splitting is surmised through the results of UV-Vis spectra, (Ahv)2–hv curve and valence-band XPS spectra of TiO2 and Co3O4 nanoparticles. The p-Co3O4/n-TiO2 composites have a significant impact on the efficiency of photocatalytic water splitting. Finally, the results of visible light irradiating overall water splitting reaction can prove the photocatalytic activity of the p-Co3O4/n-TiO2 nanocomposite. The optimal evolution rate of H2 and O2 is 8.16 μmol/g·L and 4.0 μmol/g·L, respectively. In addition, the H2 evolution rate of the p-Co3O4/n-TiO2 nanocomposites is 25% higher than that of the pure Co3O4 nanoparticles.