Porous NiTiO3/TiO2 nanostructures for photocatatalytic hydrogen evolution

Porous NiTiO 3 /TiO 2 nanostructures for photocatatalytic hydrogen evolution† Congcong Xing, ab Yongpeng Liu, c Yu Zhang, a Junfeng Liu, a Ting Zhang, d Pengyi Tang, d Jordi Arbiol, de Llu´ ıs Soler, b Kevin Sivula, c N´ estor Guijarro, c Xiang Wang, a Junshan Li, a Ruifeng Du, a Yong Zuo, a Andreu Cabot * ae and Jordi Llorca * b We present a strategy to produce porous NiTiO 3 /TiO 2 nanostructures with excellent photocatalytic activity toward hydrogen generation. In a first step, nickel-doped TiO 2 needle bundles were synthesized by a hydrothermal procedure. Through the sintering in air of these nanostructures, porous NiTiO 3 /TiO 2 heterostructured rods were obtained. Alternatively, the annealing in argon of the nickel-doped TiO 2 needle bundles resulted in NiO x /TiO 2 elongated nanostructures. Porous NiTiO 3 /TiO 2 structures were tested for hydrogen evolution in the presence of ethanol. Such porous heterostructures exhibited superior photocatalytic activity toward hydrogen generation, with hydrogen production rates up to 11.5 mmol h 1 g 1 at room temperature. This excellent performance is related here to the optoelectronic properties and geometric parameters of the material. Introduction Since the rst reports on photocatalytic water splitting, TiO 2 has been considered one of the most attractive photocatalysts. 1–3 Among others, its interest resides on its excellent optoelectronic properties, high resistance to photocorrosion, natural abundance, low-cost and safety. 4–6 However, TiO 2 optimization for solar photocatalysis requires maximizing its surface area. 7 It also requires the introduction of dopants, additional phases or structural defects that allow this wide band gap semiconductor to take advantage of a larger spectral range of the solar radiation. 8–13 To maximize surface area, the production of hollow and porous TiO 2 nanostructures has recently drawn growing attention. 14–16 Besides, Ti-based mixed oxides and particularly perovskite titanates MTiO 3 (M ¼Ba, Sr, Ca and Ni) have been pointed out as a new exciting class of photoactive materials. 17–19 Additionally heterostructured oxides, such as SrTiO 3 /TiO 2 and NiTiO 3 /TiO 2 have been demonstrated to promote chargeseparation, hole-transportation and visible-light-driven photocatalytic performance over TiO 2 . 20–23 To cite some examples in this direction, Wu et al. demonstrated an enhancement of the charge separation and hole transportation in TiO 2 /SrTiO 3 nano-heterostructures to be at the origin of an improved photoelectrochemical performance. 20 When combined with TiO 2 , several other oxides and chalcogenides have demonstrated to be able to enhance its photocatalytic activities. 23–26 In particular, Ni-based materials have been proven especially effective to improve TiO 2 photocatalytic activity. Rawool et al. demonstrated pn heterojunctions in NiO/ TiO 2 produced by sol gel to promote photocatalytic H 2 generation. 27 Wei et al. showed Ni(HCO 3 ) 2 produced through a hydrothermal method to be an excellent co-catalyst to boost photocatalytic hydrogen evolution of mesoporous TiO 2 . 28 Enhanced photocatalytic H 2 production was also reported with the introduction of Ni(OH) 2 clusters to commercially available TiO 2 (ref. 29) and in Ni/NiO/N–TiO 2x heterostructures produced from a NH 3 plasma treatment of sol–gel derived Ti–Ni precursors. 30 Besides, Qu et al. prepared porous NiTiO 3 nanorods with enhanced photocatalytic performance toward the degradation of nitrobenzene. 21 While excellent results have been obtained to date, the synthesis of such mixed oxides and oxide heterostructures in the form of high surface area materials remains as an extremely challenging endeavor, what has limited further advances in this direction. 31–33 To overcome this limitation, we present here a simple, high throughput and scalable hydrothermal-based strategy to a Catalonia Institute for Energy Research (IREC), Sant Adri` adeBes ` os, 08930 Barcelona, Spain. E-mail: [email protected] b Institute of Energy Technologies, Department of Chemical Engineering and Barcelona Research Center in Multiscale Science and Engineering, Universitat Polit` ecnica de Catalunya, EEBE, 08019 Barcelona, Spain. E-mail: [email protected] c Laboratory for Molecular Engineering of Optoelectronic Nanomaterials (LIMNO), ´ Ecole Polytechnique F´ ed´ erale de Lausanne (EPFL), Station 6, CH-1015 Lausanne, Switzerland d Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC, BIST, Campus UAB, Bellaterra, 08193 Barcelona, Catalonia, Spain e ICREA, Pg. Lluis Companys 23, 08010 Barcelona, Spain †Electronic supplementary information (ESI) available: Additional TEM, HRTEM, EELS analysis and other characterizations, photocatalytic hydrogen production measurement. See DOI: 10.1039/c9ta04763h Cite this: J. Mater. Chem. A,2019,7, 17053 Received 7th May 2019 Accepted 28th June 2019 DOI: 10.1039/c9ta04763h rsc.li/materials-a This journal is © The Royal Society of Chemistry 2019 J. Mater. Chem. A,2019,7,17053–17059 | 17053 Journal of Materials Chemistry A PAPER Published on 29 juni 2019. Downloaded by Universitat Politecnica de Catalunya on 12.12.2019 18:45:00. View Article Online View Journal | View Issue produce porous NiTiO 3 /TiO 2 heterostructures. This strategy is based on the synthesis of needle bundles of Ni-doped TiO 2 (TiO 2 : Ni) and their posterior sintering in air to produce highly crystalline NiTiO 3 /TiO 2 heterostructures with a signicant level of porosity reminiscent of the large density of interfaces within the needle bundles. We further demonstrate here that these preliminarily optimized materials are excellent photocatalysts for hydrogen generation. We choose hydrogen generation as test reaction owing to the excellent performances of Ni-based photocatalysts in this direction. 27–30,34–36 Experimental Chemicals Titanium(IV) isopropoxide (97%, Sigma-Aldrich), nickel(II)nitrate hexahydrate (98%, Fluka), ethanol (96%, PanReac AppliChem), hexadecylamine (HDA, 90%, Sigma-Aldrich), potassium chloride (Sigma-Aldrich), and ammonium hydroxide solution (28–30%, Sigma-Aldrich) were used without further purication. Synthesis of NiTiO 3 /TiO 2 heterostructures A solution of KCl (1.9 mg) in MiliQ water (0.99 g) was added dropwise to a HDA (0.45 g) solution in ethanol (20 mL) under stirring at room temperature. To this solution, a proper amount of Ni(NO 3 ) 2 $6H 2 O was added (0 mg, 22.5 mg, 45 mg, 90 mg, and 225 mg to reach 0, 0.5, 1, 2, and 5 mol% concentrations, respectively). Then, 4.7 mL of titanium(IV) isopropoxide was added. Aer stirring for 1 h at room temperature, we kept the solution 16 h without stirring, and aerward the solution was washed with ethanol three times using centrifugation to recover the product. This product was then dissolved in a solution containing 20 mL of MiliQ water, 40 mL of ethanol and 2.2 mL of ammonium hydroxide. This solution was stirred for 1 h and it was subsequently transferred to a 100 mL Teon-lined hydrothermal reactor. The reactor was heated at 170 C and maintained at this temperature for 17 h. Aerward, it was cooled down to room temperature and the product was washed three times with ethanol. Dried materials were nally annealed in a tube furnace under synthetic air ow at 650 C for 8 h. Characterization X-ray diffraction analyses (XRD, 2q:20–80; scanning rate: 5min 1 ) were carried out on a Bruker AXS D8 Advance X-ray diffractometer with Ni-ltered Cu-Karadiation (l¼0.15406 ˚ A), operating at 40 mA and 40 kV. The morphology and size of the particles were characterized by transmission electron microscopy (TEM, ZEISS LIBRA 120), working at 120 kV and eld-emission scanning electron microscopy (SEM, Zeiss Auriga) operating at 5.0 kV. High resolution transmission electron microscopy (HRTEM) images and scanning transmission electron microscopy (STEM) studies were conducted on a FEI Tecnai F20 eld emission gun microscope operated at 200 kV with a point-to-point resolution of 0.19 nm, which was equipped with high angle annular dark eld (HAADF) and a Gatan Quantum electron energy loss spectroscopy (EELS) detectors. Elemental analysis was carried out using an Oxford energy dispersive X-ray spectrometer (EDX) combined with the Zeiss Auriga SEM working at 20.0 kV. X-ray photoelectron spectroscopy (XPS) was examined on a SPECS system equipped with a Phoibos 150 MCD-9 detector, working at 150 mW with an Al anode XR50 source. Fourier transform infrared spectroscopy (FTIR, Alpha Bruker) was carried with a platinum attenuated total reectance single reection module. Nitrogen sorption measurements were collected on a MicroActive 3.00 at 77.3 K. Photocatalytic hydrogen evolution tests In a typical experiment, a cellulose paper impregnated with 2.0 mg of the photocatalyst was placed inside a photocatalytic reactor that was equipped with a UV LED (Fig. S1†). An Ar gas stream was saturated with a water : ethanol vapour mixture (90 : 10 ratio on a molar basis) by bubbling dry Ar gas at a ow rate of 20 mL min 1 through a saturator (Dreschel bottle) containing a liquid mixture of 87.5 g of H 2 O and 9.92 g of ethanol. 37,38 This Ar stream was introduced into the photoreactor and passed through the cellulose paper loaded with the photocatalyst. The photoreactor effluent was monitored on-line every 4 min using gas chromatography (GC) (Agilent 3000A MicroGC) using three columns: MS 5 ˚ A, Plot U and Stabilwax. The LED UV light source (from SACOPA, S.A.U.) consisted of four LEDs at 365 5 nm and a synthetic quartz glass cylindrical lens that transmitted the light to the photocatalyst. Light irradiation was measured directly with a UV-A radiation monitor from Solar Light Co. and was 79.1 0.5 mW cm 2 at the sample position. At the beginning of each experiment, the UV light was off, and the reaction system was purged by entering 20 mL min 1 of saturated Ar gas with the water–ethanol vapour mixture, to remove oxygen in the line. Aer 30 min, the UV light was turned on and we monitored all photoreaction products during ca. 20–40 min by GC. Control experiments were carried out with only the cellulose paper support and no photoactivity was measured. Electrochemical impedance spectroscopy (EIS) EIS analyses were performed using a Bio-Logic SP-300 potentiostat in a 3-electrode conguration employing a Cappuccinotype PEC cell, wherein the photocatalyst was deposited as a thin lm in a conductive substrate (FTO) to be used as working electrode, whereas a Ag/AgCl electrode and a Pt wired were used as a reference and a counter electrode, respectively. The applied potential on the working electrode was sinusoidally modulated with an amplitude of 25 mV using frequencies ranging from 100 kHz to 50 mHz. The analysis of the impedance response was undertaken using ZView (Scribner Associates). The apparent quantum yield (AQY) was estimated using the following equation: AQY ¼2nH2 np 100 ¼nNA ET=Ep 100 where n H 2 is the number of molecules of H 2 generated and n p is the number of incident photons reaching the catalyst. The number of incident photons can be calculated by n p ¼E T /E p , where E T is the total energy reaching the catalyst and E p is the 17054 |J. Mater. Chem. A,2019,7,17053–17059 This journal is © The Royal Society of Chemistry 2019 Journal of Materials Chemistry A Paper Published on 29 juni 2019. Downloaded by Universitat Politecnica de Catalunya on 12.12.2019 18:45:00. View Article Online energy of a photon. E T ¼PSt, where P(W m 2 ) is the power density of the incident monochromatic light, S(m 2 ) is the irradiation area and t(s) is the duration of the incident light exposure. E p ¼hc/l, where his the Planck's constant, cthe speed of light and l(m) is the wavelength of the incident monochromatic light. The number of hydrogen molecules can be calculated as n H 2 ¼nN A , where nare H 2 moles evolved during the time of light exposure (t), and N A is the Avogadro constant. 39–41 In our experimental conditions, the wavelength of the incident light was l¼365 nm, the power density of the incident light at the paper surface was P¼79.1 mW cm 2 and the irradiation area was S¼2.27 cm 2 . Results and discussion Characterization of the photocatalysts Ni-doped TiO 2 (TiO 2 : Ni) nanostructures were produced from the hydrothermal decomposition of titanium(IV) isopropoxide in the presence of proper amounts of nickel(II) nitrate hexahydrate, HDA as a shape-directing agent and KCl to control the ionic strength of the solution (see experimental section for details and Fig. S2†for SEM images). 42 As shown by TEM characterization, the size and shape of the nanostructures obtained from this hydrothermal reaction strongly depended on the amount of Ni introduced (Fig. S3†). In the absence of Ni, nanoparticles with an average size around 50 nm and mostly irregular shapes, although some elongated, were obtained. When introducing increasingly higher amounts of Ni, more and more elongated nanostructures were obtained. In the presence of a nominal 5% of Ni, most structures resembled nanorods with a length in the range from 100 nm to 200 nm and thickness of 30–60 nm (Fig. 1b, 2b and S3†). A closer look to these nanorods allowed discerning that they consisted of needleshaped nanostructures assembled in bundles (Fig. 2b). XRD analyses showed the crystal structure of these needle bundles to match that of anatase TiO 2 .Diffraction peaks shied to lower angles with the introduction of increasing amounts of Ni (Fig. 1a), which pointed at the presence of Ni ions within the TiO 2 anatase lattice. Besides anatase TiO 2 , no additional crystallographic phase could be discerned from XRD patterns. EDX analysis showed the Ni atomic concentration to be systematically higher than the nominal amount introduced, pointing at a higher yield of reaction of the nickel(II) nitrate than the titanium(IV) isopropoxide under the hydrothermal conditions used for the synthesis (Table S1†). TiO 2 : Ni nanopowders were annealed under synthetic air ow at 650 C during 8 h. Upon this thermal treatment, the color of the material changed from the original blueish, characterizing TiO 2 : Ni samples, to yellow (Fig. S4†). TEM micrographs of the annealed powder showed it to still consist on elongated nanostructures. However, on the contrary to the starting nanoneedle bundles, the annealed nanostructures displayed a crystallographic continuity (Fig. 2c) and presented numerous holes all over them (Fig. 2d). In contrast, undoped TiO 2 nanopowders annealed in the same conditions did not present such porous structure (Fig. S5†). XRD analysis showed that the porous nanostructures obtained aer the annealing of TiO 2 : Ni in air consisted mainly on anatase TiO 2 (Fig. 2a). With the annealing process, the shiof the XRD peaks observed in the precursor TiO 2 : Ni material disappeared, pointing at the outward diffusion of the Ni ions from the TiO 2 lattice. Besides the anatase phase, XRD patterns displayed the presence of a second crystal structure, which was Fig. 1 (a) XRD patterns of TiO 2 and TiO 2 : Ni (1%, 2%, 5%) nanopowders. (b) TEM image of TiO 2 : Ni (5%) nanopowder. Fig. 2 (a) XRD patterns of NiTiO 3 /TiO 2 (0, 1%, 2%, 5%) nanopowders obtained by annealing in air. (b) TEM micrograph of TiO 2 : Ni (5%) needle bundles. (c and d) TEM micrographs of NiTiO 3 /TiO 2 (5%) porous structures obtained after annealing in air. (e) HAADF micrograph of a NiTiO 3 /TiO 2 (5%) rod and corresponding EELS elemental maps for Ti, O, Ni and Ti–O–Ni. This journal is © The Royal Society of Chemistry 2019 J. Mater. Chem. A,2019,7,17053–17059 | 17055 Paper Journal of Materials Chemistry A Published on 29 juni 2019. Downloaded by Universitat Politecnica de Catalunya on 12.12.2019 18:45:00. View Article Online identied as NiTiO 3 . 43 The relative intensity of the peaks associated to this NiTiO 3 phase clearly increased with the nominal amount of Ni (Fig. 2a). Rietveld analysis showed the relative amount of the NiTiO 3 phase to correspond to a 4.6 mol%, 9.0 mol% and 15 mol%, for samples obtained from nominal Ni concentration of 1 mol%, 2 mol% and 5 mol%, respectively (Fig. S6 and Table S1†). 44 Within its experimental error, EDX analysis showed the Ni molar concentrations to be maintained aer the annealing process (Table S1†). HRTEM analysis conrmed the annealed material to contain mostly anatase TiO 2 (Fig. S7†). 45 EELS elemental maps of the annealed materials showed Ni, Ti and O to be evenly distributed throughout the nanostructures (Fig. 2e). The lack of HRTEM evidences of the presence of large NiTiO 3 crystal domains and the homogeneous distribution of Ni throughout the material observed by EELS and EDX analyses pointed at the growth of small NiTiO 3 domains as a discontinuous shell on the surface of the porous TiO 2 nanostructures. 46 XPS analyses of the annealed samples displayed only one chemical state for Ti, assigned to Ti 4+ , and one for Ni, assigned to Ni 2+ , consistently with XRD, HRTEM, EELS and EDX characterization (Fig. S8 and S9†). 47 XPS analyses also showed the amount of Ni to be almost a twofold of that measured by EDX, pointing at its preferential surface location (Table S1†). When Ni-doped materials were annealed under Ar instead of air, the sample color changed from blue to black, instead of yellow (Fig. S4†), but the elongated geometry of the initial structures was conserved. In contrast to the annealing in air, the nanostructuration of the original nanoneedle bundles partially persisted aer annealing in argon. The particles annealed in argon did not show full crystallographic continuity and contained a lower density of holes compared with the material obtained from the annealing in air (Fig. S10†). XRD patterns of materials annealed under argon showed anatase TiO 2 as the main phase, with no evidence of the presence of NiTiO 3 , but with a small XRD peak at 2q¼28.6that we associated to a NiO x phase (Fig. S11†). The black nanopowder annealed under argon showed certain magnetization, which further pointed toward the formation of a NiO x phase (Fig. S4†). 48 We hypothesize that the porous geometry of the NiTiO 3 / TiO 2 structures obtained aer annealing TiO 2 :Niinairwas created by the sintering of the nanoneedle bundles. The large density of interfaces of the initial bundles translated in the formation of a large density of holes on the sintered material, which could be aided by the simultaneous outward diffusion of Ni ions to form NiTiO 3 domains. On the contrary, annealing in argon did not allow a similar level of sintering of the nanostructures, which was promoted in air by the ubiquitous presence of oxygen. Thus, the material obtained from the annealing process in argon maintained smaller crystal domains and presented a less porous structure. The BET specic surface areas of NiTO 3 /TiO 2 and NiO x /TiO 2 were 40.8 and 28.0 m 2 g 1 , respectively, supporting the higher porosity of the former (Fig. S12†). With the annealing process in air, pure TiO 2 nanopowders did not result in porous materials as those obtained from TiO 2 : Ni because they did not present a proper initial nanostructuration. Photogeneration of hydrogen To test their photocatalytic properties toward hydrogen evolution, the annealed materials were supported on a conventional lter paper and placed on a reactor containing a UV LED light source. An Ar ow saturated with a water–ethanol vapor mixture was owed through this reactor. Exhaust gases were analyzed using GC, being acetaldehyde and hydrogen the main reaction products (Fig. S1†). Fig. 3a shows the accumulated hydrogen production during a 32 min reaction for NiTiO 3 /TiO 2 hetero-nanostructures having different nominal Ni concentrations. Data for bare TiO 2 and pure NiTiO 3 (Fig. S13 and S14†) prepared following the same procedure (see experimental section for details†) were also plotted in Fig. 3a as reference. NiTiO 3 /TiO 2 heteronanostructures with a nominal 1% of Ni systematically provided the highest hydrogen production rates, with values up to 11.5 mmol h 1 g 1 and an AQY of 11.6% at 365 nm, well above those previously reported for other Ni–Ti–O systems (Table S2†). The hydrogen production rate of NiTiO 3 /TiO 2 (1%) was almost a fourfold higher than that of bare TiO 2 and 60 times larger than that of pure NiTiO 3 (Fig. 3b, S15†). The presence of larger amounts of Ni did not improve the evolution rate with respect to NiTiO 3 /TiO 2 (1%). We associate this experimental evidence with the blocking of the TiO 2 surface active sites when too large amounts of NiTiO 3 were grown on the TiO 2 surface. SEM characterization showed the morphology of the NiTiO 3 / TiO 2 hetero-nanostructures to remain unmodied aer the catalytic test (Fig. 3c). We believe the enhancement of the photocatalytic properties of TiO 2 with the NiTiO 3 introduction to be related with an improvement of three fundamental parameters: (i) surface area; (ii) light absorption; and (iii) charge separation. We estimate that the introduction of Ni had associated a factor 3 increase of Fig. 3 (a) H 2 production of TiO 2 , NiTiO 3 /TiO 2 (1%, 2%, 5%), NiO x /TiO 2 (1%) and NiTiO 3 . (b) H 2 production rate of TiO 2 , NiTiO 3 /TiO 2 (1%, 2%, 5%), NiO x /TiO 2 (1%) and NiTiO 3 . (c) Optical and SEM images of NiTiO 3 / TiO 2 (5%) before and after 30 min photocatalytic test. 17056 |J. Mater. Chem. A,2019,7,17053–17059 This journal is © The Royal Society of Chemistry 2019 Journal of Materials Chemistry A Paper Published on 29 juni 2019. Downloaded by Universitat Politecnica de Catalunya on 12.12.2019 18:45:00. View Article Online the surface area with respect to pure TiO 2 through the formation of the porous NiTiO 3 /TiO 2 hetero-nanostructures during the annealing process. Additionally, the presence of NiTiO 3 provided a higher light absorption coefficient at the used excitation wavelength. In case of using solar radiation, the presence of NiTiO 3 would provide an even larger improvement with respect to pure TiO 2 , as it would allow absorbing a larger portion of the solar spectra due to its smaller band gap (Fig. S16†). Besides, the proper band alignment of TiO 2 and NiTiO 3 allowed an efficient spatial charge separation that signicantly reduced recombination of the photogenerated charge carriers before reaction. To gain additional insight into the band structure alignment and the interfacial charge carrier dynamics at the NiTiO 3 /TiO 2 heterostructure, EIS analyses were carried out on TiO 2 , NiTiO 3 / TiO 2 (1%) and NiTiO 3 thin lms in dark conditions. Representative impedance responses in the form of Nyquist plot for TiO 2 and NiTiO 3 /TiO 2 (1%) are displayed in Fig. 4a. Responses typically feature a single semicircle. Interestingly, the size of the semicircle obtained for the TiO 2 sample was reduced with the incorporation of Ni. A more quantitative description of the EIS data can be provided by modelling the electrical response with an equivalent circuit, such as a Randles circuit (Fig. 4a inset), which has been previously reported to successfully describe nanostructured TiO 2 and other photocatalysts. 22,50,54 This circuit consists of a resistor R S in series with a bulk capacitance C bulk , attributed to the space charge region, and a second resistor R ct,bulk that represents the charge transport resistance, both in parallel. At a rst glance, the decrease in the size of the semicircle with the incorporation of Ni in TiO 2 suggests the reduction in the charge transport resistance, 49,50 which is further conrmed with the decrease of R ct,bulk from 15.4 kUto 9.8 kU pointing out the amelioration of the charge transport process. In addition, the electron lifetime s n can be calculated as the inverse of the angular frequency at the maximum of the Nyquist plot (2pf max ), 22 highlighted in Fig. 4a. Results reveal that s n increases from 4.5 ms to 6.6 ms when the heterostructure is formed. In general, s n is considered a metric of electron recombination, i.e., the longer the value the lesser the charge recombination. The lengthening of s n when TiO 2 is interfaced with NiTiO 3 underpins the effective charge carrier separation across the n–n heterojunction and thus the enhanced photocatalytic activity of the composite with respect TiO 2 . Similar behavior was reported by Wei and co-workers on NiTiO 3 /TiO 2 composite for inorganic sensitized solar cells. 22 In an attempt to further interrogate the energy band positions, Mott–Schottky (M–S) plots were constructed from C bulk values extracted from EIS measurements on TiO 2 and NiTiO 3 and further analyzed using the M–S equation: Cbulk 2¼2 qA2330NDVVfb kT q where qis elementary charge, Ais taken as the geometric area (assuming a low surface roughness spin-coated thin lm), 3is relative permittivity (taken to be 55 and 15 for TiO 2 (ref. 51) and NiTiO 3 , 52 respectively), 3 0 is vacuum permittivity, N D is donor density, Vis applied potential, V  is at band potential, kis Boltzmann constant, and Tis absolute temperature. The positive slope in the linear region (dashed line in Fig. 4b) unambiguously conrmed the n-type character of both NiTiO 3 and TiO 2 , whereas V  values of 0.186 V and 0.276 V vs. RHE were obtained for TiO 2 and NiTiO 3 , respectively. Likewise, N D shows values of 1.1 10 18 cm 3 for TiO 2 and 5.5 10 17 cm 3 for NiTiO 3 . Similar values have also been reported by Thimsen et al. on nanocrystalline anatase TiO 2 , 51 and by Trari and co-workers on nanostructured NiTiO 3 . 53 We note that the depletion width is estimated to be 20 nm in the linear region of M–S plot for both TiO 2 and NiTiO 3 so that M–S analysis can be reasonably applied in the case of not fully depleted feature. 54 Considering the density-of-state effective mass for electrons m de for nanocrystalline anatase TiO 2 is 9.1 10 30 kg, 51 and for nanostructured NiTiO 3 is 1.9 10 30 kg, 55 the effective density of states in the conduction band N C is: 54,56 NCh22pmdekT h23 2 where his Planck constant. N C is estimated to be 7.7 10 20 cm 3 for TiO 2 , and 7.5 10 19 cm 3 for NiTiO 3 . Since the ratio between N D and N C is in all cases less than 0.05, both TiO 2 and NiTiO 3 could be treated as nondegenerate semiconductors, and therefore, the Boltzmann statistics can be applied: 54,56 Fig. 4 (a) Representative impedance response of TiO 2 (black sphere) and NiTiO 3 /TiO 2 (1%) (orange sphere) with the corresponding fit (solid lines) in Nyquist plot. The maximum point of the Nyquist plot is highlighted (hollow), and the equivalent circuit used is included as an inset. (b) Mott– Schottky plot of TiO 2 (black sphere) and NiTiO 3 (red sphere) including a linear fit (dashed lines). (c) Schematics of the electronic band structures of a Z-scheme NiTiO 3 /TiO 2 heterojunction. This journal is © The Royal Society of Chemistry 2019 J. Mater. Chem. A,2019,7,17053–17059 | 17057 Paper Journal of Materials Chemistry A Published on 29 juni 2019. Downloaded by Universitat Politecnica de Catalunya on 12.12.2019 18:45:00. View Article Online E CB E F ¼kT ln(N C /N D ) where E CB is the bottom of the conduction band (CB) and E F is the Fermi level position. E CB is found to be around 168 mV and 126 mV above V  for TiO 2 and NiTiO 3 , respectively. Based on the aforementioned results, Fig. 4c displays a scheme of the electronic band structures. The relative band alignment between NiTiO 3 and TiO 2 allows the formation of a direct Z-scheme heterojunction. In such Ni–TiO 2 system, while photogenerated electrons in the CB of TiO 2 recombine with photogenerated holes in the valance band (VB) of NiTiO 3 , photogenerated electrons with strong reduction ability remain in the CB of NiTiO 3 ready to reduce water into hydrogen and photogenerated holes with strong oxidation abilities in the VB of TiO 2 could drive ethanol dehydrogenation to acetaldehyde and hydrogen. Therefore, the formation of a direct Z-scheme NiTiO 3 /TiO 2 heterojunction can promote photogenerated carriers with strong redox abilities to drive photocatalytic reaction. 57 As a result, NiTiO 3 /TiO 2 composites at optimal NiTiO 3 loading condition shows a signicantly increased photocatalytic activity. Similar Z-scheme photocatalytic systems have also been reported such as TiO 2 /Rh, 58 TiO 2 /CdS, 59 and CdS/ cobalt-benzimidazole. 60 Conclusions Porous NiTiO 3 /TiO 2 heterostructures were prepared from the annealing in air of TiO 2 : Ni nanoneedle bundles obtained from a hydrothermal route. Pores were generated during the sintering of the nanostructured bundles due to the presence of a large density of interfaces on the precursor material. NiTiO 3 /TiO 2 heterostructures were used for photocatalytic hydrogen generation from ethanol–water solutions at room temperature. NiTiO 3 /TiO 2 hetero-structures provided unprecedented H 2 production rates up to 11.5 mmol h 1 g 1 and an AQY of 11.6%. This excellent performance was associated to the improvement of three parameters, surface area, light absorption and charge separation. We estimated a factor 3 increase of the surface area with the introduction of Ni through the formation of porous NiTiO 3 /TiO 2 hetero-nanostructures during the annealing in air. Additionally, the presence of NiTiO 3 , with a lower band gap energy, provided a higher light absorption coefficient at the excitation wavelength. Besides, TiO 2 and NiTiO 3 presented a proper band alignment that allows an efficient spatial charge separation of the photogenerated charge carriers before reaction, as conrmed by EIS analysis. Author contributions The manuscript was prepared through the contribution of all authors. A. C. and J. L. guided the project and supervised the work. C. C. X., Y. P. L., Y. Z., Y.$P. L., J. L., and A. C. conceived and prepared the manuscript. L. S. and J. L. designed the hydrogen production experiments. C. C. X., Y. Z., and J. F. L. produced the samples. Y. P. L., K.$S., N. G., and Y. Z. performed the electrochemical measurements and analyzed the results. J. S. L and R. F. D performed XRD measurements and discussed these results. T. Z., P.$Y. T., and J. A. performed HRTEM and STEM-EELS and discussed these results. X. W., performed BET results. L. S. and J. L. performed XPS measurements and discussed these results. The manuscript was corrected and improved by all authors. Conflicts of interest There are no conicts of interest to declare. Acknowledgements This work was supported by the European Regional Development Funds and by the Spanish Ministerio de Econom´ ıay Competitividad through projects ENE2016-77798-C4-3-R and ENE2015-63969-R. Y. L. and N. G. thank the Swiss National Science Foundation (SNF) for funding under the Ambizione Energy grant no. PZENP2_166871. J.L. is a Serra Hunter fellow and is grateful to ICREA Academia program and GC 2017 SGR 128. T. Z., P.$Y. T. and J. A. acknowledge funding from Generalitat de Catalunya 2017 SGR 327 and the Spanish MINECO project ENE2017-85087-C3. ICN2 acknowledges support from the Severo Ochoa Programme (MINECO, Grant no. SEV-20130295) and is funded by the CERCA Programme/Generalitat de Catalunya. Part of the present work has been performed in the framework of Universitat Aut` onoma de Barcelona Materials Science PhD program. C. C. X., Y. Z and T. Z. thank the China Scholarship Council for scholarship support. Notes and references 1 X. Wu, G. Q. Lu and L. Wang, Energy Environ. Sci., 2011, 4, 3565–3572. 2 D. P. Wang and H. C. Zeng, Chem. Mater., 2009, 21, 4811– 4823. 3 Q. Xiang, J. Yu and M. Jaroniec, J. Am. Chem. Soc., 2012, 134, 6575–6578. 4 Y. Ma, X. Wang, Y. Jia, X. Chen, H. Han and C. Li, Chem. Rev., 2014, 114, 9987–10043. 5 D. Su, S. Dou and G. Wang, Chem. Mater., 2015, 27, 6022– 6029. 6 J. Liu, H. Bai, Y. Wang, Z. Liu, X. Zhang and D. D. Sun, Adv. Funct. Mater., 2010, 20, 4175–4181. 7 B. Liu and H. C. Zeng, Chem. Mater., 2008, 20, 2711–2718. 8 N. O. Gopal, H. H. Lo and S. C. Ke, J. Am. Chem. Soc., 2008, 130, 2760–2761. 9 C. Burda, Y. Lou, X. Chen, A. C. S. Samia, J. Stout and J. L. Gole, Nano Lett., 2003, 3, 1049–1051. 10 M. Chiodi, C. P. Cheney, P. Vilmercati, E. Cavaliere, N. Mannella, H. H. Weitering and L. Gavioli, J. Phys. Chem. C, 2012, 116, 311–318. 11 R. Bouchet, A. Weibel, P. Knauth, G. Mountjoy and A. V. Chadwick, Chem. Mater., 2003, 15, 4996–5002. 12 M. Murdoch, G. I. N. Waterhouse, M. A. Nadeem, J. B. Metson, M. A. Keane, R. F. Howe, J. Llorca and H. Idriss, Nat. Chem., 2011, 3, 489–492. 17058 |J. Mater. Chem. A,2019,7,17053–17059 This journal is © The Royal Society of Chemistry 2019 Journal of Materials Chemistry A Paper Published on 29 juni 2019. Downloaded by Universitat Politecnica de Catalunya on 12.12.2019 18:45:00. View Article Online 13 G. I. N. Waterhouse, A. K. Wahab, M. Al-Ou, V. Jovic, D. H. Anjum, D. Sun-Waterhouse, J. Llorca and H. Idriss, Sci. Rep., 2013, 3, 2849. 14 J. H. Pan, X. Zhang, A. J. Du, D. D. Sun and J. O. Leckie, J. Am. Chem. Soc., 2008, 130, 11256–11257. 15 S. C. Yang, D. J. Yang, J. Kim, J. M. Hong, H. G. Kim, I. D. Kim and H. Lee, Adv. Mater., 2008, 20, 1059–1064. 16 J. Y. Hwang, H. M. Kim, S. K. Lee, J. H. Lee, A. Abouimrane, M. A. Khaleel, I. Belharouak, A. Manthiram and Y. K. Sun, Adv. Energy Mater., 2016, 6,1–7. 17 H. Fan, H. Li, B. Liu, Y. Lu, T. Xie and D. Wang, ACS Appl. Mater. Interfaces, 2012, 4, 4853–4857. 18 B. Inceesungvorn, T. Teeranunpong, J. Nunkaew, S. Suntalelat and D. Tantraviwat, Catal. Commun., 2014, 54, 35–38. 19 X. Wang, C.-N. Xu, H. Yamada, K. Nishikubo and X.-G. Zheng, Adv. Mater., 2005, 17, 1254–1258. 20 F. Wu, Y. Yu, H. Yang, L. N. German, Z. Li, J. Chen, W. Yang, L. Huang, W. Shi, L. Wang and X. Wang, Adv. Mater., 2017, 29, 1701432. 21 Y. Qu, W. Zhou, Z. Ren, S. Du, X. Meng, G. Tian, K. Pan, G. Wang and H. Fu, J. Mater. Chem., 2012, 22, 16471–16476. 22 Y.-Y. Li, J.-G. Wang, H.-H. Sun and B. Wei, ACS Appl. Mater. Interfaces, 2018, 10, 11580–11586. 23 M. Dahl, Y. Liu and Y. Yin, Chem. Rev., 2014, 114, 9853–9889. 24 C. Tang, W. Xu, C. Chen, Y. Li and L. Xu, Adv. Mater. Interfaces, 2017, 4, 1700361. 25 J. Tian, P. Hao, N. Wei, H. Cui and H. Liu, ACS Catal., 2015, 5, 4530–4536. 26 D. Barreca, G. Carraro, A. Gasparotto, C. Maccato, M. E. A. Warwick, K. Kaunisto, C. Sada, S. Turner, Y. G¨ on¨ ull¨ u, T. P. Ruoko, L. Borgese, E. Bontempi, G. Van Tendeloo, H. Lemmetyinen and S. Mathur, Adv. Mater. Interfaces, 2015, 2,1–11. 27 A. Rawool, M. R. Pai, A. M. Banerjee, A. Arya, R. Ningthoujam, R. Tewari, R. Rao, B. Chalke, P. Ayyub and A. Tripathi, Appl. Catal., B, 2018, 221, 443–458. 28 Y. Wei, G. Cheng, J. Xiong, F. Xu and R. Chen, ACS Sustainable Chem. Eng., 2017, 5, 5027–5038. 29 J. Yu, Y. Hai and B. Cheng, J. Phys. Chem. C, 2011, 115, 4953– 4958. 30 S. Hu, F. Li, Z. Fan and J. Gui, J. Power Sources, 2014, 250,30– 39. 31 S. Liu, S. Xue, W. Zhang, J. Zhai and G. Chen, J. Mater. Chem. A, 2014, 2, 18040–18046. 32 J. H. Pan, C. Shen, I. Ivanova, N. Zhou, X. Wang, W. C. Tan, Q. H. Xu, D. W. Bahnemann and Q. Wang, ACS Appl. Mater. Interfaces, 2015, 7, 14859–14869. 33 Y. Xin, J. Jiang, K. Huo, T. Hu and P. K. Chu, ACS Nano, 2009, 3, 3228–3234. 34 S. Peng, Y. Yang, J. Tan, C. Gan and Y. Li, Appl. Surf. Sci., 2018, 447, 822–828. 35 W. Zhang, Y. Li and S. Peng, J. Mater. Chem. A, 2017, 5, 13072–13078. 36 S. Peng, X. Zeng and Y. Li, Int. J. Hydrogen Energy, 2015, 40, 6038–6049. 37 E. Molins, M. Benito, I. Mata, L. Mart´ ınez, L. Soler and J. Llorca, MRS Adv., 2017, 2, 3499–3504. 38 L. Mart´ ınez, L. Soler, I. Angurell and J. Llorca, Appl. Catal., B, 2019, 248, 504–514. 39 Y. Li, P. Han, Y. Hou, S. Peng and X. Kuang, Appl. Catal., B, 2019, 244, 604–611. 40 Y. Li, H. Li, Y. Li, S. Peng and Y. H. Hu, Chem.–Eng. J., 2018, 344, 506–513. 41 Y. Li, Y. Hou, Q. Fu, S. Peng and Y. H. Hu, Appl. Catal., B, 2017, 206, 726–733. 42 R. Imani, R. Dillert, D. W. Bahnemann, M. Pazoki, T. Apih, V. Kononenko, N. Repar, V. Kralj-Igliˇ c, G. Boschloo, D. Drobne, T. Edvinsson and A. Igliˇ c, Small, 2017, 13, 1700349. 43 K. Huo, Y. Li, R. Chen, B. Gao, C. Peng, W. Zhang, L. Hu, X. Zhang and P. K. Chu, Chempluschem, 2015, 80, 576–582. 44 B. H. Toby and R. B. Von Dreele, J. Appl. Crystallogr., 2013, 46, 544–549. 45 P. D. Cozzoli, A. Kornowski and H. Weller, J. Am. Chem. Soc., 2003, 125, 14539–14548. 46 H. Bin Wu, H. H. Hng and X. W. D. Lou, Adv. Mater., 2012, 24, 2567–2571. 47 J. Huang, Y. Jiang, G. Li, C. Xue and W. Guo, Renewable Energy, 2017, 111, 410–415. 48 J. H. Kim, K. Zhu, Y. Yan, C. L. Perkins and A. J. Frank, Nano Lett., 2010, 10, 4099–4104. 49 Y. Zuo, Y. Liu, J. Li, R. Du, X. Yu, C. Xing, T. Zhang, L. Yao, J. Arbiol, J. Llorca, K. Sivula, N. Guijarro and A. Cabot, ACS Appl. Mater. Interfaces, 2019, 11, 6918–6926. 50 X. Zhu, N. Guijarro, Y. Liu, P. Schouwink, R. A. Wells, F. Le Formal, S. Sun, C. Gao and K. Sivula, Adv. Mater., 2018, 30, 1801612. 51 E. Thimsen, A. B. F. Martinson, J. W. Elam and M. J. Pellin, J. Phys. Chem. C, 2012, 116, 16830–16840. 52 S. H. Chuang, M. L. Hsieh and D. Y. Wang, J. Chin. Chem. Soc., 2012, 59, 628–632. 53 L. Boudjellal, A. Belhadi, R. Brahimi, S. Boumaza and M. Trari, Mater. Sci. Semicond. Process., 2018, 75, 247–252. 54 Y. Liu, F. Le Formal, F. Boudoire, L. Yao, K. Sivula and N. Guijarro, J. Mater. Chem. A, 2019, 7, 1669–1677. 55 M. A. Ruiz Preciado, A. Kassiba, A. Morales-Acevedo and M. Makowska-Janusik, RSC Adv., 2015, 5, 17396–17404. 56 M. Rudan, Physics of Semiconductor Devices, 2015. 57 Q. Xu, L. Zhang, J. Yu, S. Wageh, A. A. Al-Ghamdi and M. Jaroniec, Mater. Today, 2018, 21, 1042–1063. 58 E. N. K. Glover, S. G. Ellington, G. Sankar and R. G. Palgrave, J. Mater. Chem. A, 2016, 4, 6946–6954. 59 J. Low, B. Dai, T. Tong, C. Jiang and J. Yu, Adv. Mater., 2019, 31, 1802981. 60 W. Yu, S. Zhang, J. Chen, P. Xia, M. H. Richter, L. Chen, W. Xu, J. Jin, S. Chen and T. Peng, J. Mater. Chem. A, 2018, 6, 15668–15674. This journal is © The Royal Society of Chemistry 2019 J. Mater. Chem. A,2019,7,17053–17059 | 17059 Paper Journal of Materials Chemistry A Published on 29 juni 2019. Downloaded by Universitat Politecnica de Catalunya on 12.12.2019 18:45:00. View Article Online