BDD Heterojunction Thin Film for

Jan 12, 2016 - Semiconductor photocatalysis driven by electron/hole has begun a new era in the field of solar energy conversion and storage. Here we r...
11 downloads 22 Views 3MB Size
Letter www.acsami.org

Charge Separation in TiO2/BDD Heterojunction Thin Film for Enhanced Photoelectrochemical Performance Chiaki Terashima,*,† Ryota Hishinuma,†,‡ Nitish Roy,† Yuki Sugiyama,†,‡ Sanjay S. Latthe,† Kazuya Nakata,†,‡ Takeshi Kondo,†,‡ Makoto Yuasa,†,‡ and Akira Fujishima† †

Photocatalysis International Research Center, Research Institute for Science & Technology, and ‡Faculty of Science and Technology, Tokyo University of Science, 2641 Yamazaki, Noda, Chiba 278-8510, Japan S Supporting Information *

ABSTRACT: Semiconductor photocatalysis driven by electron/hole has begun a new era in the field of solar energy conversion and storage. Here we report the fabrication and optimization of TiO2/BDD p-n heterojunction photoelectrode using p-type boron doped diamond (BDD) and n-type TiO2 which shows enhanced photoelectrochemical activity. A p-type BDD was first deposited on Si substrate by microwave plasma chemical vapor deposition (MPCVD) method and then n-type TiO2 was sputter coated on top of BDD grains for different durations. The microstructural studies reveal a uniform disposition of anatase TiO2 and its thickness can be tuned by varying the sputtering time. The formation of p-n heterojunction was confirmed through I−V measurement. A remarkable rectification property of 63773 at 5 V with very small leakage current indicates achieving a superior, uniform and precise p−n junction at TiO2 sputtering time of 90 min. This suitably formed p-n heterojunction electrode is found to show 1.6 fold higher photoelectrochemical activity than bare n-type TiO2 electrode at an applied potential of +1.5 V vs SHE. The enhanced photoelectrochemical performance of this TiO2/BDD electrode is ascribed to the injection of hole from p-type BDD to n-type TiO2, which increases carrier separation and thereby enhances the photoelectrochemical performance. KEYWORDS: BDD, TiO2, heterojunction electrode, photocurrent, water splitting, photocatalysis, renewable energy

P

diamond or BDD lies near +1.5 V vs SHE, thereby making it a poor photooxidant catalyst, unlike TiO2 (VB position at +3 V vs SHE and thereby strong photooxidant candidate).11 Boron is adequately used as p-type doping material to synthesize BDD because of its small covalent radius. BDD can act as a p-type semiconductor electrode with minute background current, high thermal conductivity, wide potential window, excellent longterm stability against mechanical damage and corrosive solutions.13 Therefore, very high stability of p-type BDD could be utilized by preparing p−n junction with a very strong photooxidant n-type TiO2 to enhance its photoelectrochemical water splitting activity. In a recent review article,14 Wang et al. have thoroughly discussed about the design and development of various efficient and stable semiconductor heterojunction photocatalyst for hydrogen production, pollutant degradation and photocatalytic disinfection. An ideal heterojunction system should have high rectification ratio with small leakage current. The rectification property/ratio greatly depends on nature of the semiconductors and charge transfer at heterojunction. Uniform, stable and proper contact at the junction is desirable and expected to fabricate the good quality p−n junction devices. Few studies on p−n junction with p-type BDD and n-type

hotocatalytic properties of TiO2 semiconductor are continuously and broadly investigated especially after the pioneering discovery made by two Japanese scientists, Fujishima and Honda in 1972. They successfully achieved hydrogen generation through photoelectrochemical water splitting using TiO2 as a photoanode.1 Apart from its excellent photocatalytic properties, TiO2 is primarily used in the paint industry, medical implement of teeth, cosmetics due to its biocomatilibility, chemical stability, and easy fabrication and proven to be no harm to the environment.2 It has enormous emergent applications which deal with photocatalytic dye degradation, as antifogging or self-cleaning agent, dye sensitized solar cells, and superhydrophilic properties.3−5 TiO2 under light illumination with energy greater than its bandgap creates charge carriers (electrons and holes) that undergo a fast recombination process and so provide very small quantum efficiency.6 To reduce the charge recombination rate and therefore to increase quantum efficiency of TiO2, making composites with the plasmonic materials for enhanced light scattering properties, heterojunction, faceting different crystal planes or preparing two mixed phases were studied.7−10 TiO2 is an n-type wide bandgap material (∼3.2 eV) with conduction band (CB) just above the hydrogen evolution potential with respect to standard hydrogen electrode (SHE).11 On the other hand, boron doped diamond (BDD) is a p-type material (bandgap ∼5.5 eV) with very high CB minima (−4 V vs SHE).12 In spite of the very high CB position, the valence band (VB) of © 2016 American Chemical Society

Received: November 14, 2015 Accepted: January 12, 2016 Published: January 12, 2016 1583

DOI: 10.1021/acsami.5b10993 ACS Appl. Mater. Interfaces 2016, 8, 1583−1588

Letter

ACS Applied Materials & Interfaces semiconductor materials are reported. But all of these are found to exhibit either poor amplifying activity or high leakage current. For example, Han et al. report the TiO2/BDD heterojunction formation with leakage current −0.03 mA/cm2 and rectification ratio ∼10 at 5 V.15 Yu et al. show a very high leakage current with 0.01 mA and a rectification ratio of ∼4 at potential 4 V with TiO2/BDD heterojunction.16 Li et al. report a comparatively high rectification property of ∼120 with leakage current ∼0.1 μA at potential 5 V with ZnO/BDD heterojunction.17 Therefore, reliable design of TiO2/BDD p−n junction is still lacking.18 In this study, we report the fabrication of reliable and stable p-n heterojunction by radio frequency (RF) sputter uniform TiO2 thin film on microwave plasma chemical vapor deposition (MPCVD) grown BDD with very small leakage current and high amplification ratio of 63773 and at 5 V. In addition to the optimized heterojunction fabrication, electronic structure, carrier density, depletion region and photoelectrochemical performance of the heterojunction thin films were measured. An enhanced photoelectrochemical activity of the suitably formed p−n junction TiO2/BDD photoelectrode than the bare TiO2 photoelectrode is also demonstrated. A BDD film was first deposited on single-crystalline Si substrate by MPCVD system (AX6500, Cornes Technologies). Si substrate was shaved by nanodiamond and washed by ultrasonic bath of acetone, ethanol and water, respectively. The boron source, boron trioxide (B2O3), was dissolved in acetonemethanol solution (9:1 v/v) with a B/C atomic ratio of 1000 ppm. The mixture was fed into a reactor by hydrogen gas. The typical depositing process lasted for 4 h. TiO2 was then coated by reactive sputtering on a BDD film for different durations and other optimized experimental parameters were kept fixed. A RF sputtering machine (VTR-150M/SRF, ULVAC) was used for TiO2 deposition. Metal Ti target was sputtered by Ar/O2 (Ar:O2 = 1:1 (v/v)) atmosphere. Sputtering conditions such as background pressure, working pressure, Ar and air flow ratio, RF power, sputtering time, and substrate temperature were optimized as shown in Table S1. Li et al. also demonstrated thin and uniform epitaxial growth of ZnO nanorods on diamond grains through thermal vapor transport method without significantly disturbing the original microstructure of the diamond grains.16 Han et al. dip-coated a uniform TiO2 nanoparticle film with a thickness of around 300 nm on the BDD electrode by completely hiding the grain structure. A photocurrent as well as stability of the photoelectrode is entirely depends on the film thickness. Better photocatalytic performance can be achieved using high thickness films (>1 μm); however, stability is reported to be poor. Hence, Han et al. suggested a TiO2 film thickness of less than 0.5 μm on BDD to achieve efficient and reproducible photoactivity without losing stability.19 Therefore, TiO2 film was deposited for three different time intervals of 45, 90, and 180 min at temperature 300 °C to control and optimize the thickness of RF sputter TiO2 thin films. Figure 1 shows the field emission scanning electron microscope (FE-SEM, JEOL, JEM-3100F) images of the TiO2/BDD heterojunction thin films fabricated at different RF sputtering time. Figure 1a shows an FEM-SEM image fabricated at 45 min of sputtering time indicating the morphology is identical with bare BDD (Figure S1a, b). Indeed, high magnification FE-SEM analysis (marked by a rectangle in Figure 1a) as shown in Figure 1b suggests the deposition of a very thin TiO2 film. With an increase in sputtering time to 90 and 180 min, the thickness of the RF

Figure 1. FESEM images of TiO2/BDD thin films at RF sputtering time (a, b) 45, (c) 90, and (d) 180 min, respectively. RF sputtering produces uniform TiO2 deposition on BDD grains and thickness of the RF sputter TiO2 increases with increase in sputtering time.

sputter TiO2 increases as clearly visible from the morphology analysis (Figure 1, Figure S1). It is noted that BDD grains are coated by RF sputter TiO2 thin films in all three different sputtering times and the approximate maximum thickness of the 90 min RF sputter TiO2 is measured to be ∼660 nm (Figure S1f). Phase analysis of the as-deposited TiO2, BDD, and TiO2/ BDD was confirmed by powder X-ray diffractometer (XRD, Rigaku Ultima IV, Cu Kα radiation, λ = 1.54 Å) and Raman analysis (JASCO, NRS-5100). Figure 2(a) shows the XRD of RF sputter TiO2 at 90 min, MPCVD grown BDD and heterojunction TiO2/BDD substrate obtained at sputtering time 90 min, respectively. The Bragg’s peak of BDD and TiO2 thin films are found to match with the theoretical values which confirm the anatase TiO2 and diamond with no other crystalline phases present in the as synthesized thin films. For TiO2, Bragg’s angles at 25.24 and 48.06° correspond to the (101) and (200) planes of anatase phase while for BDD, Bragg’s peaks at 43.7 and at 75.1° match with (111) and (220) planes of diamond. The comparison of peak intensities between RF TiO2 and MPCVD grown BDD suggests the formation of smaller crystallite of the RF sputter TiO2 than the MPCVD grown BDD. Raman study was also employed for phase analysis of the BDD and TiO2 based samples. Raman spectra of the corresponding TiO2, BDD and TiO2/BDD heterojunction are shown in Figure 2b. RF sputter TiO2 thin film obtained at 90 min is found to be three strong peaks at around 394, 514, and 636 cm−1, are typically originated from B1g, A1g, and Eg modes of anatase phase TiO2 as shown in Figure 2b, respectively.20,21 The B1g, A1g, and Eg peaks are mainly originated correspondingly from symmetric bending vibration, antisymmetric bending vibration and symmetric stretching vibration of O−Ti−O in TiO2.22 Diamond shows a characteristic strong Raman peak at ∼1330 cm−1, due to sp3 carbon while upon doping with boron, this peak was reported to be decreased gradually and a new peak around at 500 cm−1 appeared. Therefore, the two Raman peak at 1332 and ∼500 cm−1 indicate the BDD is doped with boron. A Raman spectrum of TiO2/BDD heterojunction film includes all the peaks corresponding to TiO2 and BDD suggesting no phase transformation of BDD and TiO2 in heterojunction (Figure 2b). 1584

DOI: 10.1021/acsami.5b10993 ACS Appl. Mater. Interfaces 2016, 8, 1583−1588

Letter

ACS Applied Materials & Interfaces

Figure 2. (a) XRD and (b) Raman spectra of RF sputter TiO2 (90 min), MPCVD grown BDD, and TiO2/BDD heterojunction obtained at RF sputtering 90 min, respectively.

Figure 3. (a) I−V characteristic of TiO2/BDD electrode fabricated at different RF sputtering time showing the amplification property of the p-n heterojunction with a remarkable rectifying ratio of 63773 at potential 5 V. (b) Photocurrent of the different electrodes under light irradiation (1 SUN) and dark conditions as a function of applied potential in 0.5 M Na2SO4. Red arrow showing the onset potential of hole injection from BDD to VB of photoexcited TiO2, and it becomes flat at around 1.6 V vs SHE.

Table 1. Comparison of BDD-Based Heterojunction Materials with Different Methods, Thickness, Forward Current (IF), Reverse Current (IR), and Rectification Ratio (IF/IR) n-type

method

n-type thickness (nm)

voltage (V)

IF (mA/cm2)

IR (μA/cm2)

IF/IR

ref

TiO2 TiO2 TiO2 TiO2 TiO2 TiO2 ZnO ZnO

sputter sputter sputter liquid phase deposition dip-coat MOCVD thermal transport hydrothermal

∼300 ∼500 ∼1000 2000 ∼300 150 1000−2000 800

5 5 5 5 5 4 5 6

3.43 3.38 2.88 0.4 0.25 0.04 0.012 2.72

34.94 0.053 1.74 400 30 10 0.1 134

98 63773 1655 1 ∼8 4 120 20

this work this work this work ref 13 ref 15 ref 16 ref 17 ref 18

with the literature reports. These p-n heterojunction TiO2/ BDD electrodes were then tested for photoelectrochemical water splitting reaction as photoanode and Pt as counter electrode. The photocurrent/current was measured using a potentiostat (HOKUTO DENKO/HZ-7000) and solar simulator (1 SUN intensity at the photoelectrode), Ag/AgCl saturated with KCl as reference electrode, Pt net as counter electrode, and the counter and working electrode is separated by a proton exchange nafion membrane. Figure 3b shows the linear sweep voltammogrmas (LSVs) of TiO2 and TiO2/BDD electrodes obtained at RF sputtering 90 min in 0.5 M Na2SO4 aqueous solution. TiO2 and TiO2/BDD electrodes show negligible current in the potential range of −0.1 V to +1.6 V vs SHE, implying no oxidation reaction under dark condition. Indeed, under 1 SUN, current density is found to increase with applied potential for TiO2 and TiO2/BDD electrodes (Figure 3b, Figure S3). Photocurrent is kept increasing in the potential range from −0.1 V to +1.5 V and becomes saturated thereafter (Figure 3b). It is noted that up to +0.43 V vs SHE, photocurrent density is almost same for both TiO2 and TiO2/BDD obtained at 90 min, but it starts superior for TiO2/ BDD over TiO2 (Figure 3b) in the potential region > + 0.43 V.

A room-temperature current−voltage (I−V) plot of the TiO2/BDD electrodes obtained at different RF sputtering time is measured to ensure the fabrication of good quality p−-n heterojunction. Figure 3a shows the I−V curve passing through zero point confirming p-n heterojunction formation between the RF sputtered TiO2 and MPCVD grown BDD.15 In all three samples of TiO2/BDD fabricated at different RF sputtering time, the I−V plots show an exponential increase in current density in forward bias. The current density is found to increase more for TiO2/BDD electrode obtained at RF sputtering time 90 min than the others (Figure 3a). Though, in reverse bias, a significant amount of leakage current is observed for TiO2/ BDD obtained at RF sputtering of 45 and 180 min, TiO2/BDD fabricated at 90 min shows almost no leakage current (inset of Figure 3a, Table 1). The onset potential for leakage current of TiO2/BDD obtained at 45 min sputtering time is around −1.1 V, whereas it is −4.3 V for TiO2/BDD fabricated at 180 min. The characteristic of I−V plots reveals that the optimum RF sputtering time is 90 min to fabricate a uniform and good quality TiO2/BDD p-n heterojunction in our system. Table 1 summarizes the forward current, reverse current (leakage current), rectification ratio and film thickness, and compared 1585

DOI: 10.1021/acsami.5b10993 ACS Appl. Mater. Interfaces 2016, 8, 1583−1588

Letter

ACS Applied Materials & Interfaces

band bending takes place at the p−n junction of TiO2/BDD heterosystem to maintain the Fermi level position, we calculated the depletion width to be 71 nm as shown in Figure 4b. The detailed calculation of depletion width, electronic valence study is shown in Supporting Information (Figure S2). Therefore, there is a scope to diffuse the electrons from TiO2 and holes from BDD. Under dark conditions a negligible amount of water oxidation current is obtained, due to very poor carrier transportation in the potential region 0 to +1.6 V (Figure 3a, b). Indeed, in the presence of light and forward bias, extent of photoexcited carrier transportation increases remarkably resulting in a high photocurrent for water oxidation. It is noted that TiO2/BDD electrode obtained at 90 min RF sputtering shows superior water oxidation to TiO2 at bias potential ≥0.43 V vs SHE (Figure 3b, Figure 4d). This potential is due to the difference in VB positions of TiO2 and BDD. BDD has sufficient hole concentration very near to its VB. Therefore, at bias potential ≥0.43 V, VB of BDD becomes equal or higher than TiO2 (Figure 4d) and injection of holes from BDD to VB of photoexcited TiO2 becomes feasible. At lower bias potential, (V vs SHE 0 to ≤0.43), such hole injection from BDD to VB of photoexcited TiO2 is prevented due to lower energy of holes present very near to VB of BDD than VB TiO2 and hence both the electrodes exhibit similar kind of photocurrent (Figure 3b, Figure 4a, c). Perhaps at sufficiently higher bias potential, for example ≥1.5 V, the photocurrent becomes saturated/flat indicating the maximum limit of hole injection from BDD to photoexcited TiO2 and therefore no increase in photocurrent thereafter (Figure 3b). In case of TiO2/BDD photoelectrode fabricated at 45 min RF sputtering, no such suitable p−n junction formation (Figure 3(a)) and hence no excess hole injection form BDD to TiO2 (Figure S3). For the case TiO2/BDD photoelectrode obtained at 180 min, the more perturbed p−n junction electrode is obtained compared to 90 min sputter electrode and thereby more overpotential (+0.79 V) is required for hole injection (Figure S3). The charge separation was further confirmed by incorporation of hole scavenger. Dotan et al. showed the charge separation at the electrode/electrolyte interface using H2O2 as the hole scavenger.24 This is due to the fact that H2O2 has low oxidation potential compared to water oxidation and has higher affinity toward hole (>10 times higher affinity than water). Therefore, charge separation was also tested using H2O2 as the hole scavenger and found that suitably formed TiO2/BDD electrode showed 152% higher charge separation than TiO2 electrode at an applied potential of +1.5 V (Figure S4). Cross-sectional schematic of carrier transport by a suitable TiO2/BDD heterojunction photoelectrode is shown in Figure 5. BDD is a p-type semiconductor with sufficient hole concentration (1.79 × 1020 cm−3) and TiO2 is an n-type selfdoped semiconductor with electron concentration of 5.84 × 1017 cm−3 at 90 min sputtering time. Taking advantage of the scopes of p-type BDD for electron accommodation from n-type TiO2 and water oxidation property of TiO2, we show a facile photoexcited charge carrier separation could take place between these two materials when suitable p−n junction is formed. In this study, at RF sputtering time of 90 min, a good quality p−n junction obtains between TiO2 and BDD as revealed by the I− V plot and found to inject hole from BDD to VB of photoexcited TiO2 at potential ≥0.43 V. This charge transfer (hole from BDD to VB of TiO2, and electron transfer from TiO2 to BDD and finally to counter electrode via external

At + 1.5 V, n-type TiO2 electrode exhibits a photocurrent density of 0.08 mA/cm2, whereas heterojunction electrode obtained at 90 min sputtering time exhibits a density of 0.13 mA/cm2 which is 1.6 fold higher. The improved photoelectrochemical activity of TiO2/BDD heterojunction photoanode fabricated at 90 min is due to the formation of suitable and good quality p−n junction which enhances photoexcited charge carrier separation in the potential region ≥ +0.43 to +1.5 V (Figure 3b).23 Photocurrent for the TiO2/BDD electrode fabricated at RF sputtering 45 min is half (0.04 mA/cm2) than that of the RF sputter TiO2 (Figure S3) at a bias potential of +1.5 V. This poor photoelectrochemical activity could be due to the not formation of good quality p-n heterojunction as revealed by the I−V plot (high leakage current with TiO2/BDD fabricated at 45 min sputtering, inset of Figure 3a). On the other hand, TiO2/BDD obtained at RF sputtering time 180 min is found to inject holes at higher overpotential of +0.79 V vs SHE (Figure S3). The higher overpotential for hole injection in this case could be due to the small leakage current (indicating the p−n junction is not as good as fabricated at 90 min of sputtering time) obtained at −4.3 V (Figure 3a) revealing the reason behind hole injection at higher potential compared to heterojunction electrode obtained at 90 min of RF sputtering (Figure S3). CB and VB positions of BDD and TiO2 individuals have quite large difference as shown in Figure 4a. VB maxima of TiO2 and BDD in TiO2/BDD fabricated at 90 min sputtering are found to be 7.1 and 5.9 eV, respectively (Figure S2), whereas carrier concentrations of MPCVD grown BDD and 90 min RF sputter TiO2 are calculated to be 1.79 × 1020 and 5.84 × 1017 cm−3, respectively. In addition, Fermi levels of TiO2 and BDD lie just below and above the respective CB and VB. As

Figure 4. (a) Electronic band positions of TiO2 and BDD with respect to vacuum level and SHE. (b) TiO2/BDD suitable p-n heterojunction under equilibrium condition. Depth of depletion region is calculated to be 71 nm in the TiO2/BDD heterojunction obtained at 90 min sputtering time. (c) Under 1 SUN light illumination and bias potential ≤0.43 V, no spontaneous hole injection from BDD to VB of photoexcited TiO2 due to difference in VB positions and hence no excess water oxidation. (d) hole injection from BDD to VB of photoexcited TiO2 under a forward bias potential of ≥0.43 V vs SHE, which increases the charge carrier separation and thereby enhanced water oxidation. Dashed line/curve shows the respective Fermi level position. 1586

DOI: 10.1021/acsami.5b10993 ACS Appl. Mater. Interfaces 2016, 8, 1583−1588

Letter

ACS Applied Materials & Interfaces Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by JST ACT-C. N.R. is thankful to Japan Science and Technology for the fellowship. The authors acknowledge technical assistance from Y. Shibano with H2 production experiments.



(1) Fujishima, A.; Honda, K. Electrochemical Photolysis of Water at a Semiconductor Electrode. Nature 1972, 238, 37−38. (2) Li, C.; Koenigsmann, C.; Ding, W.; Rudshteyn, B.; Yang, K. R.; Regan, K. P.; Konezny, S. J.; Batista, V. S.; Brudvig, G. W.; Schmuttenmaer, C. A.; Kim, J.-H. Facet-Dependent Photoelectrochemical Performance of TiO2 Nanostructures: An Experimental and Computational Study. J. Am. Chem. Soc. 2015, 137, 1520−1529. (3) Nakata, K.; Fujishima, A. TiO2 Photocatalysis: Design and Applications. J. Photochem. Photobiol., C 2012, 13, 169−189. (4) Chen, X.; Mao, S. S. Titanium Dioxide Nanomaterials: Synthesis, Properties, Modifications, and Applications. Chem. Rev. 2007, 107, 2891−2959. (5) Bai, Y.; Mora-Sero, I.; De Angelis, F.; Bisquert, J.; Wang, P. Titanium Dioxide Nanomaterials for Photovoltaic Applications. Chem. Rev. 2014, 114, 10095−10130. (6) Sanz, R.; Romano, L.; Zimbone, M.; Buccheri, M. A.; Scuderi, V.; Impellizzeri, G.; Scuderi, M.; Nicotra, G.; Jensen, J.; Privitera, V. UVblack Rutile TiO2: An Antireflective Photocatalytic Nanostructure. J. Appl. Phys. 2015, 117, 074903. (7) Chandrasekharan, N.; Kamat, P. V. Improving the Photoelectrochemical Performance of Nanostructured TiO2 Films by Adsorption of Gold Nanoparticles. J. Phys. Chem. B 2000, 104, 10851−10857. (8) Roy, N.; Sohn, Y.; Pradhan, D. Synergy of Low-Energy {101} and High-Energy {001} TiO2 Crystal Facets for Enhanced Photocatalysis. ACS Nano 2013, 7, 2532−2540. (9) Liu, G.; Yang, H. G.; Pan, J.; Yang, Y. Q.; Lu, G. Q.; Cheng, H. − M. Titanium Dioxide Crystals with Tailored Facets. Chem. Rev. 2014, 114, 9559−9612. (10) Schneider, J.; Matsuoka, M.; Takeuchi, M.; Zhang, J.; Horiuchi, Y.; Anpo, M.; Bahnemann, D. W. Understanding TiO2 Photocatalysis: Mechanisms and Materials. Chem. Rev. 2014, 114, 9919−9986. (11) Roy, N.; Sohn, Y.; Leung, K. T.; Pradhan, D. Engineered Electronic States of Transition Metal Doped TiO2 Nanocrystals for Low Overpotential Oxygen Evolution Reaction. J. Phys. Chem. C 2014, 118, 29499−29506. (12) Zhang, L.; Zhu, D.; Nathanson, G. M.; Hamers, R. J. Selective Photoelectrochemical Reduction of Aqueous CO2 to CO by Solvated Electrons. Angew. Chem. 2014, 126, 9904−9908. (13) Yuan, J.; Li, H.; Gao, S.; Lin, Y.; Li, H. A Facile Route to n-type TiO2-nanotube/p-type Boron-doped-diamond Heterojunction for Highly Efficient Photocatalysts. Chem. Commun. 2010, 46, 3119− 3121. (14) Wang, H.; Zhang, L.; Chen, Z.; Hu, J.; Li, S.; Wang, Z.; Liu, J.; Wang, X. Semiconductor Heterojunction Photocatalysts: Design, Construction, and Photocatalytic Performances. Chem. Soc. Rev. 2014, 43, 5234−5244. (15) Han, Y.; Qiu, J.; Miao, Y.; Han, J.; Zhang, S.; Zhang, H.; Zhao, H. Robust TiO2/BDD Heterojunction Photoanodes for Determination of Chemical Oxygen Demand in Wastewaters. Anal. Methods 2011, 3, 2003−2009. (16) Yu, H.; Chen, S.; Quan, X.; Zhao, H.; Zhang, Y. Fabrication of a TiO2−BDD Heterojunction and its Application As a Photocatalyst for the Simultaneous Oxidation of an Azo Dye and Reduction of Cr(VI). Environ. Sci. Technol. 2008, 42, 3791−3796. (17) Li, H.; Sang, D.; Cheng, S.; Lu, J.; Zhai, X.; Chen, L.; Pei, X. Epitaxial Growth of ZnO Nanorods on Diamond and Negative

Figure 5. Cross-sectional schematic of TiO2/BDD obtained at RF sputtering 90 min on Si substrate and their direction of charge transportation at potential ≥0.43 V to split water.

circuit) increases with increase in applied potential. At sufficiently high potential, V ≥ 1.6 V vs SHE, enough hole are present at the VB of photoexcited TiO2 and thereby water oxidation becomes independent of bias potential rather it is then solely dependent on photoexcitation. Hence no increase in photocurrent is observed thereafter (Figure 3b). To valid this argument, we measured H2 at an applied potential +1.5 V in the Pt counter electrode while illuminating the photoanode under 1 SUN. H2 in Pt chamber is measured to be 61.84 μmole for TiO2/BDD and 29.27 μmole for TiO2 photoelectrode, respectively, after 3 h of illumination (Figure S5). Therefore, ∼2 fold higher H2 amount for TiO2/BDD obtained at 90 min sputtering than TiO2 only is due to the higher charge separation between p-type BDD and n-type TiO2 at the heterojunction. In conclusion, the successful fabrication and optimization of TiO2/BDD p-n heterojunction electrode is achieved by MPCVD grown BDD with TiO2 deposition by RF sputtering at different durations. Microstructure analysis suggests that RF sputtering results in a uniform and thin TiO2 deposition over BDD. TiO2 film thickness and p−n junction quality can be optimized by varying the RF sputtering time. I−V plots suggest the suitable p−n junction is achieved at RF sputtering time 90 min keeping other fabrication parameters fixed. TiO2/BDD p− n junction fabricated at 90 min RF sputtering shows a remarkable rectification ratio of 63773 at 5 V. Photoelectrochemical activity of the p−n junction electrode fabricated at 90 min RF sputtering is found to exhibit 1.6 fold higher than 90 min RF sputter TiO2 thin film at 1.5 V. The enhanced photoelectrochemical activity is ascribed to the suitably formed p−n junction and thereby hole injects from BDD to TiO2 in forward bias of V ≥ 0.43.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.5b10993. Experimental conditions, additional FESEM images, photoemission spectra, calculation of band bending, photocurrent, relative charge separation, and H 2 production (PDF)



REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. 1587

DOI: 10.1021/acsami.5b10993 ACS Appl. Mater. Interfaces 2016, 8, 1583−1588

Letter

ACS Applied Materials & Interfaces Differential Resistance of n-ZnO Nanorod/p-diamond Heterojunction. Appl. Surf. Sci. 2013, 280, 201−206. (18) Yu, Q.; Li, H.; Wang, Q.; Cheng, S.; Jiang, L.; Zhang, Y.; Ai, T.; Guo, C. S. Hydrothermal Synthesis, Characterization and Properties of Boron-doped ZnO Sheets Grown on p-diamond Film. Mater. Lett. 2014, 128, 284−286. (19) Han, Y.; Zhang, S.; Zhao, H.; Wen, W.; Zhang, H.; Wang, H.; Peng, F. Photoelectrochemical Characterization of a Robust TiO2/ BDD Heterojunction Electrode for Sensing Application in Aqueous Solutions. Langmuir 2010, 26, 6033−6040. (20) Roy, N.; Park, Y.; Sohn, Y.; Leung, K. T.; Pradhan, D. Green Synthesis of Anatase TiO2 Nanocrystals with Diverse Shapes and their Exposed Facets-Dependent Photoredox Activity. ACS Appl. Mater. Interfaces 2014, 6, 16498−16507. (21) Balachandran, U.; Eror, N. G. Raman Spectra of Titanium Dioxide. J. Solid State Chem. 1982, 42, 276−282. (22) Ramasamy, P.; Lim, D.-H.; Kim, J.; Kim, J. A General Approach for Synthesis of Functional Metal Oxide Nanotubes and their Application in Dye-sensitized Solar Cells. RSC Adv. 2014, 4, 2858− 2864. (23) Chen, Y.; Crittenden, J. C.; Hackney, S.; Sutter, L.; Hand, D. W. Preparation of a Novel TiO2-Based p−n Junction Nanotube Photocatalyst. Environ. Sci. Technol. 2005, 39, 1201−1208. (24) Dotan, H.; Sivula, K.; Gratzel, M.; Rothschild, A.; Warren, S. C. Probing the Photoelectrochemical Properties of Hematite (α-Fe2O3) Electrodes Using Hydrogen Peroxide as a Hole Scavenger. Energy Environ. Sci. 2011, 4, 958−964.

1588

DOI: 10.1021/acsami.5b10993 ACS Appl. Mater. Interfaces 2016, 8, 1583−1588