Titanium dioxide (TiO2) has been considered as one of the most promising photocatalysts over the past few decades because of its availability, low cost, photostability, and non-toxicity [1-7]. However, the large binding energy of the excitons (130 meV) and the relatively fast electron-hole recombination severely limit its photon-to-electricity conversion efficiency and, consequently, the photocatalytic performance [8-15]. Loading a cocatalyst is a simple and facile way of promoting the separation and transfer of photogenerated charge carriers. Much effort has recently been devoted to seeking low-cost transition-metal cocatalysts, besides the conventional noble metals. Ni compounds such as NiO, NiS, and Ni(OH)2 are the most promising candidates [16-22]. Yoshikawa et al. [23] reported that NiO nanoparticles can largely enhance photocatalytic hydrogen production over mesoporous TiO2. Xu et al. [24] showed that NiS is highly efficient in promoting photocatalytic hydrogen production over CdS nanoparticles. Ni(OH)2 has received more attention owing to the controllable shapes and hydrogen evolution properties [25-30]. Furthermore, as a p-type semiconductor, Ni(OH)2 is also an ideal host material for fabricating p-n type heterostructures at the interfaces of oxide semiconductors, where the formed internal electric field is able to efficiently separate the charge carriers, consequently leading to enhancements in the photon-to-electricity conversion efficiency and solar photocatalytic hydrogen evolution and production [16, 31].
In this work, we report a solvothermal approach combined with oil-bath-based chemical synthesis to prepare a 3D flower-like TiO2@Ni(OH)2 core-shell heterostructure, through which a six-fold increase in hydrogen production is obtained. First, anatase (TiO2) microspheres were prepared via the solvothermal pathway reported in the literature [32], and then, 50 mg of the as-prepared TiO2 microspheres was dispersed in 200 mL of deionized (DI) water and stirred for 30 min. Afterwards, 2.5 mmol of Ni(NO3)2·6H2O, 2.5 mmol of hexamethylenetetramine, and 0.25 mmol of citric acid trisodium salt dehydrate were well dissolved in the above dispersion and continuously stirred at room temperature for 1 h. The resulting solution was then heated at 363 K in an oil-bath for 12 h with slow stirring. After the growth of Ni(OH)2 on TiO2, the products were separated by centrifugation, and washed several times with DI water and ethanol. The 3D flower-like TiO2@Ni(OH)2 samples were finally obtained after being dried at 333 K for 12 h.
The morphology and structure of the as-prepared TiO2@Ni(OH)2 microspheres were characterized by field-emission scanning electron microscopy (FESEM) and transmission electron microscopy (TEM) analyses. The TiO2 microspheres show uniform morphology and sphere sizes of ca. 1–2 μm, based on the insert image shown in Fig. 1a. Fig. 1a and 1b show the typical FESEM images of the as-prepared TiO2@Ni(OH)2 samples that exhibit a perfect 3D flower-like nanostructure with well-arranged Ni(OH)2 nanoflakes on the surfaces and well-knit core-shell morphologies. However, there are no obvious lattice fringes in the HRTEM image, indicating that the Ni(OH)2 nanosheets exhibit weak crystallinities (Fig. 1c and 1d). From the HAADF-STEM image and the corresponding EDS mapping spectra (Fig. 1e–1h), homogeneous distributions of elemental Ni and Ti in the shell layer and the core, respectively, are observed, confirming the perfect formation of the core-shell heterostructure. Meanwhile, the loading of Ni(OH)2 was ca. 18.47%, calculated based on the atomic ratio of Ni to Ti (4.64:23.77), which was determined from the EDS mapping results.
The crystal structure and composition of the as-prepared samples were analyzed by powder XRD, as shown in Fig. 2a. The dominant peaks, centered at 33.4° and 59.7°, correspond to the diffraction peaks of (101) and (110) that are indexed to the characteristic peaks of Ni(OH)2 standard (JCPDS no. 38-0715) [33]. All the peaks of the as-prepared TiO2 indicate that the sample contains the pure tetragonal anatase phase (JCPDS no. 21-1272). After enwrapping the Ni(OH)2 nanosheet shells on the surface of the TiO2 substrate, the intensities of the characteristic diffraction peaks obviously decreased, and two new diffraction peaks, centered at 33.4° and 59.7° and attributed to Ni(OH)2, can be observed, which confirmed the successful synthesis of TiO2@Ni(OH)2. In order to investigate the change in the optical properties, the UV-vis diffuse reflectance spectroscopy (UV-vis DRS) patterns of the TiO2, Ni(OH)2, and TiO2@Ni(OH)2 samples were examined, as shown in Fig. 2b. Pristine TiO2 shows bandgap absorption at ca. 380 nm, corresponding to a bandgap energy of ca. 3.2 eV. It is notable that upon enwrapping the Ni(OH)2 shells, the photoresponse of TiO2 significantly extends up to the visible-light region. As for the TiO2@Ni(OH)2 samples, the absorptions observed in the region ca. 600–800 nm and at ca. 450 nm can be assigned to the Ni(Ⅱ) d-d transition and the direct interfacial charge transfer (IFCT) that occurs from the VB of TiO2 to that of Ni(Ⅱ) [34]. Fig. 3 shows the N2 adsorption/desorption isotherms and pore-size distribution curves of the TiO2 and TiO2@Ni(OH)2 samples. The special surface area of the TiO2@Ni(OH)2 samples is 111.24 m2 g–1, which is attributed to their flower-like structures with radially aligned nanosheets, and is much higher than that of the TiO2 microspheres (49.4838 m2 g–1). After enwrapping the Ni(OH)2 shells on the TiO2 surface, the pore size is ca. 7.35 nm, as obtained from the pore-size distribution curves. These results demonstrate that the Ni(OH)2 shells offer more active sites for photocatalytic hydrogen production.
All the as-prepared photocatalysts were evaluated in terms of their photocatalytic hydrogen production capabilities. The photocatalytic hydrogen evolution experiments were performed in a relatively evacuated system by using an external glass reaction cell. In a typical experiment, 0.1 g of the photocatalyst and 0.5 g of EDTA-Na2, used as a sacrificial reagent, were suspended in 100 mL of an aqueous solution subjected to magnetic stirring and the temperature of the glass reaction cell was maintained at about 278 K by circulating water. Fig. 4a shows the variations in the amounts of hydrogen evolved over Ni(OH)2, TiO2, and TiO2@Ni(OH)2 samples that were stimulated through solar irradiation. The separated Ni(OH)2 photocatalyst shows a small hydrogen production activity (3.66 μmol within 6 h, or 0.61 μmol g–1 h–1). The amount of hydrogen evolved in the case of the TiO2 photocatalyst is ca. 31.8 μmol within 6 h, corresponding to a rate of 5.3 μmol g–1 h–1, whereas a six-fold enhancement (205.8 μmol within 6 h, or 34.3 μmol g–1 h–1) is achieved over the TiO2@Ni(OH)2 sample. It is noteworthy that the rate of hydrogen evolution reveals a gradually increasing trend, suggesting that an activation process is involved in the reaction system. This will be discussed in detail in the section on reaction mechanism. In order to further understand the role of the Ni(OH)2 cocatalyst, a reference experiment was carried out as shown in Fig. 4b. TiO2@NiO sample displays a much higher hydrogen evolution rate (10.3 μmol g–1 h–1) than pristine TiO2, suggesting that the NiO-based material is a favorable cocatalyst for hydrogen production. In order to eliminate the effect of the separated Ni2+, 100 μl of 10 mg mL–1 aqueous Ni(NO3)2 was added into the reaction system. Remarkably, the added Ni2+ did not change the photocatalytic hydrogen production rate. In this system, it is possible that the added Ni2+ precludes the dissolution of Ni2+ from the Ni(OH)2 nanosheets, which can promote the hydrogen evolution. These results confirm that the enhancement in the amount of photocatalytic hydrogen produced can be attributed to the role of the Ni(OH)2 cocatalyst and the unique core-shell nanostructure. Fig. 5 shows the results of the cycling experiments performed for the TiO2@Ni(OH)2 photocatalyst, over which hydrogen is produced under the stimulation of solar irradiation. It clearly indicates that the TiO2@Ni(OH)2 photocatalyst does not reveal a noticeable deactivation even after four cycling tests for 24 h, implying a high stability. It was calculated that the apparent quantum yield (AQY) of the TiO2@Ni(OH)2 sample for photocatalytic hydrogen evolution is ca. 0.356%, which is a six-fold enhancement compared to that (0.055%) of the bare TiO2 sample. The AQY was calculated as AQY = number of reacted electrons/number of incident photons × 100%, or AQY = number of evolved H2 molecules x 2/number of incident photons × 100%.
The overall irradiance of the Xe lamp used was 150 mW cm– 2, and the irradiated area was 38.47 cm–2. The irradiance of the Xe lamp was measured by using a light intensity meter (ILT950).
The transient photocurrent of TiO2@Ni(OH)2 was recorded to investigate the interfacial electronic interaction occurring between the Ni(OH)2 shell and the TiO2 core. Fig. 6a shows the transient photocurrent curves of the TiO2@Ni(OH)2 and TiO2 photoelectrodes with several on-off cycles of the intermittent solar irradiation in 0.2 mol L–1 aqueous Na2SO4 solution. Upon illumination, the TiO2@Ni(OH)2 photoelectrode shows a strong photocurrent response (ca. 6.5 μA), which is nearly 520 times higher than that of TiO2 (ca. 12.2 nA); this confirms that the unique p-n type heterostructure and the Ni(OH)2 cocatalyst can promote the separation of the photogenerated charge carriers and accelerate the interfacial reaction. After several on-off cycles, there is no obvious decay in the photocurrent, further indicating the high stability of TiO2@Ni(OH)2. Nyquist plots of TiO2@Ni(OH)2 and TiO2 were obtained from EIS measurements to characterize the intrinsic electronic transport properties; a simple Randles circuit model was used for fitting the data (Fig. 6b). The elements Rct.bulk and Cbulk are related to the bulk resistance and bulk capacitance corresponding to the charge transfer from the semiconductor to the electrolyte [35]. TiO2@Ni(OH)2 reveals a smaller radius (Rct.bulk = 1.9 kΩ) than TiO2 (Rct.bulk = 2.5 kΩ), indicating its stronger charge-transfer ability, which is in agreement with the results obtained based on their activities.
To reveal the chemical state of the elemental Ni, high-resolution XPS was carried out on the TiO2@Ni(OH)2 sample before (TN-1) and after (TN-2) the photocatalytic reaction; the obtained patterns are shown in Fig. 7. For the TN-1 sample, the measured binding energies of Ni 2p3/2 and Ni 2p1/2 are equal to 855.6 and 873.1 eV, respectively. On the other hand, after the photocatalytic reaction proceeded for 4000 s, the spectrum of TN-2 is significantly changed. The binding energy of Ni 2p3/2 shifts to 854.8 eV, indicating the formation of NiO, which is in good agreement with the results presented in other reports [36]. One possible explanation is that the photogenerated electrons transfer from the CB of TiO2 to the CB of Ni(OH)2 and partially reduce the Ni2+ to Ni0 atoms. However, these Ni atoms are unstable in air and are easily oxidized. Therefore, it is not surprising that NiO is observed in the TN-2 sample after the photocatalytic reaction. The unchanged binding energy of Ni 2p1/2 indicates that Ni(OH)2 still exists in the TN-2 sample.
Based on the XPS patterns and the gradually increasing trend observed for the photocatalytic hydrogen rate, a possible mechanism has been proposed, as shown in Fig. 8. Since the potential of Ni2+/Ni (E0 = –0.23 V) is slight lower than the CB energy level of TiO2 and owing to the driving force of the electric field that is developed internally in the unique p-n type heterostructure, the photogenerated electrons will be easily transferred to Ni(OH)2 from the CB of TiO2. Therefore, when the TiO2 sphere was irradiated by simulated solar light, the photogenerated electrons were rapidly transferred to the surface of the Ni(OH)2 cocatalyst, where the electrons were partially used to reduce the hydrogen protons (process Ⅰ shown in Fig. 8); the remaining electrons were utilized in the reduction of Ni2+ (process Ⅱ).
In summary, this work demonstrated the fabrication of a p-n type TiO2@Ni(OH)2 core-shell heterostructure to achieve a six-fold increase in hydrogen production and provided some useful guidance for the design and synthesis of nanostructured photocatalysts in the domain of photocatalysis and photoelectrocatalysis.
2019,
Vol. 40 Issue (3): 320-325 
PDF
2019,
Vol. 40 Issue (3): 320-325 PDF