pubs.acs.org/Langmuir © 2009 American Chemical Society
Photoelectrochemical Characterization of a Robust TiO2/BDD Heterojunction Electrode for Sensing Application in Aqueous Solutions Yanhe Han,†,‡ Shanqing Zhang,*,† Huijun Zhao,† William Wen,† Haimin Zhang,† Hongjuan Wang,†,§ and Feng Peng§ † Griffith School of Environment Gold Coast Campus, Griffith University QLD 4222, Australia, School of Life Science, Liaoning Normal University, Dalian 116029, China, and §School of Chemistry and Chemical Engineering, South China University of Technology, Guangzhou 510640, China
‡
Received September 30, 2009. Revised Manuscript Received November 15, 2009 Titanium dioxide (TiO2) and boron-doped diamond (BDD) are two of the most popular functional materials in recent years. In this work, TiO2 nanoparticles were immobilized onto the BDD electrodes by a dip-coating technique. Continuous and uniform mixed-phase (anatase and rutile) and pure-anatase TiO2/BDD electrodes were obtained after calcination processes at 700 and 450 °C, respectively. The particle sizes of both types of TiO2 film range from 20 to 30 nm. In comparison with a TiO2/indium tin oxide (ITO) electrode, the TiO2/BDD electrode demonstrates a higher photoelectrocatalytic activity toward the oxidation of organic compounds, such as glucose and potassium hydrogen phthalate. Among all the tested TiO2 electrodes, the mixed-phase TiO2/BDD electrode demonstrated the highest photoelectrocatalytic activity, which can be attributed to the formation of the p-n heterojunction between TiO2 and BDD. The electrode was subsequently used to detect a wide spectrum of organic compounds in aqueous solution using a steady-state current method. An excellent linear relationship between the steady-state photocurrents and equivalent organic concentrations was attained. The steady-state oxidation photocurrents of the mixed-phase TiO2/BDD electrode were insensitive to pH in the range of pH 2-10. Furthermore, the electrodes exhibited excellent robustness under strong acidic conditions that the TiO2/ITO electrodes cannot stand. These characteristics bestow the mixed-phaseTiO2/BDD electrode to be a versatile material for the sensing of organic compounds.
1. Introduction Boron-doped diamond (BDD) is one of the most promising advanced electrode materials in the field of electroanalysis with special characteristics such as a very low background current, wide working potential window, resilient mechanical strength, robust resistance against corrosion (even being anodic polarized in acidic solutions) and long-term durability and stability.1-4 The boron doping makes this p-type semiconductor electrically as conductive as common conductors at room temperature.5 In conjunction with n-type semiconductors, such as TiO2, BDD can be a highly desired electrode substrate for the fabrication of a new generation of sensing devices.6 Owing to its excellent photocatalytic activity and superior oxidation ability, TiO2 photocatalyst has been applied in the *To whom correspondence should be addressed: Telephone: þ61-7-5552 8155. Fax: þ61-7-5552 8067. E-mail:
[email protected]. (1) Sires, I.; Brillas, E.; Cerisola, G.; Panizza, M. J. Electroanal. Chem. 2008, 613, 151. (2) Poh, W. C.; Loh, K. P.; Zhang, W. D.; Triparthy, S.; Ye, J.-S.; Sheu, F.-S. Langmuir 2004, 20, 5484. (3) Mitani, N.; Einaga, Y. J. Electroanal. Chem. 2009, 626, 156. (4) Holt, K. B.; Bard, A. J.; Show, Y.; Swain, G. M. J. Phys. Chem. B 2004, 108, 15117. (5) Kalish, R. Carbon 1999, 37, 781. (6) Yu, H.; Chen, S.; Quan, X.; Zhao, H.; Zhang, Y. Environ. Sci. Technol. 2008, 42, 3791. (7) Kikuchi, Y.; Sunada, K.; Iyoda, T.; Hashimoto, K.; Fujishima, A. J. Photochem. Photobiol., A 1997, 106, 51. (8) Sunada, K.; Kikuchi, Y.; Hashimoto, K.; Fujishima, A. Environ. Sci. Technol. 1998, 32, 726. (9) Kominami, H.; Kumamoto, H.; Kera, Y.; Ohtani, B. Appl. Catal., B 2001, 30, 329. (10) Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W. Chem. Rev. (Washington, DC) 1995, 95, 69. (11) Zhao, H.; Jiang, D.; Zhang, S.; Catterall, K.; John, R. Anal. Chem. 2004, 76, 155.
Langmuir 2010, 26(8), 6033–6040
fields of water purification/sterilization,7-9 wastewater treatment,10 and environmental monitoring,11 ever since Fujishima and Honda discovered the photocatalytic splitting of water on titania electrodes.12 A main attraction of the TiO2 photocatalyst is the strong oxidative power of the oxidative species produced under UV illumination, such as photo holes and highly reactive free radicals,13-16 that can oxidize most organic compounds, such as organic pollutant in wastewater. However, this technology still has to overcome a series of hurdles before being utilized for practical applications, such as difficult and costly postseparation of the catalyst from the reaction solution and low photoefficiency. Immobilizing TiO2 onto a conducting substrate not only enables the recycling of the catalyst, eliminating the need of the postseparation process, but also suppresses photoelectron/hole recombination by applying a potential bias to improve the photocatalytic efficiency drastically.17 The general equation for complete mineralization of an organic compound, CyHmOjNkXq, on a TiO2 electrode can be represented by eq 1: Cy Hm Oj Nk Xq þ ð2y -jÞH2 O f yCO2 þ qX - þ kNH3 þ ð4y -2j þ m -3kÞHþ þ ð4y -2j þ m -3k -qÞe -
ð1Þ
where the elements present are represented by their atomic symbols, and X represents a halogen atom, respectively. The (12) Fujishima, A.; Honda, K. Nature 1972, 238, 37. (13) Han, S. T.; Li, J.; Xi, H. L.; Xu, D. N.; Zuo, Y.; Zhang, J. H. J. Hazard. Mater. 2009, 163, 1165. (14) Chen, D.; Yang, D.; Geng, J.; Zhu, J.; Jiang, Z. Appl. Surf. Sci. 2008, 255, 2879. (15) Andronic, L.; Duta, A. Mater. Chem. Phys. 2008, 112, 1078. (16) He, H. Y. Int. J. Environ. Res. 2009, 3, 57. (17) Jiang, D.; Zhang, S.; Zhao, H. Environ. Sci. Technol. 2007, 41, 303.
Published on Web 12/23/2009
DOI: 10.1021/la903706e
6033
Article
Han et al.
stoichiometric ratio of elements in the organic compound is represented by the coefficients y, m, j, k and q. The electron transfer number (n) in the complete mineralization of the organic compound is equal to 4y - 2j þ m - 3k - q. The photocurrent obtained in the oxidation course represents the photoelectrocatalytic activity and effectiveness of oxidation process. Interestingly, it directly relates to the concentration of organic compounds in the reaction media, therefore, it can be used as analytical signal to detect the organic compounds. Additionally, TiO2 nanoparticle exhibits n-type semiconductor properties. Thus, the incorporation of the n-type TiO2 with the ptype BDD could lead to the formation of a p-n heterojunction, which can act as an internal electrostatic potential in the space charge region to facilitate efficient separation of the photoinduced electrons and holes.18 This is an added advantage of using BDD as the conducting substrate over other materials such as indium tin oxide (ITO) and metal (e.g., Pt or Ti). Recently, various pure anatase TiO2/BDD electrodes were fabricated and used to photoelectrocatalytically degrade organic compounds such as ethanol,18 acid orange II and 2,4-dichlorophenol,19 and reactive yellow and hexavalent chromium.6 However, to the best of our knowledge, photoelectrochemical sensing of organic compounds using the mixed-phase (anatase and rutile) TiO2/BDD heterojunction electrode has not been fully investigated and reported. In this work, robust pure-anatase TiO2 and mixed-phase TiO2/ BDD heterojunction electrodes were constructed using a dip coating technique. The preparation, material characterization, and photoelectrochemical characterization of the TiO2/BDD electrodes were investigated systematically. In comparison with TiO2/ITO electrodes, the analytical performance of the TiO2/ BDD electrodes, in terms of sensitivity, linear range, and stability, were evaluated in the photoectrochemical oxidation of numerous organic compounds.
2. Experimental Section 2.1. Materials and Chemicals. BDD electrodes with a
resistivity of 0.1 Ω cm were purchased from CSEM (Switzerland). All chemicals used in this work were of analytical grade from Sigma Aldrich and used as received. All solutions were prepared using high-purity deionized water (>18 MΩ cm). 2.2. Preparation of the Heterojunction Electrodes. Aqueous TiO2 colloid was prepared by hydrolysis of titanium butoxide according to the method described in our previous work.20 The resultant colloidal solution contains ca. 60 g/dm3 of TiO2 solid with particle sizes ranging from 8 to 10 nm. Carbowax (30% w/w based on the solid weight of the TiO2 colloid) was added as a binder to improve the adhesion and increase the porosity of the resultant TiO2 film. The BDD electrode was pretreated with aqua regia and H2O2 solution in sequence, rinsed with distilled water, and dried in air. After the pretreatment, the BDD slides were dipcoated in the TiO2 colloidal solution. The coated electrodes were then calcined at 450 °C in air for 2 h, or sintered at 700 °C in argon atmosphere for 2 h after the carbowax was removed at 450 °C in air for 0.5 h. TiO2/ITO electrodes17 were prepared similarly without the aqua regia pretreatment. 2.3. Apparatus and Measurements. All photoelectrochemical experiments were performed at room temperature in a threeelectrode photoelectrochemical cell with a quartz window for purpose of UV illumination. The working electrode surface area exposed to solution was confined by a circle with a diameter of 6 mm. A Ag/AgCl electrode and a platinum mesh were used as the (18) Manivannan, A.; Spataru, N.; Arihara, K.; Fujishima, A. Electrochem. Solid-State Lett. 2005, 8, C138. (19) Qu, J.; Zhao, X. Environ. Sci. Technol. 2008, 42, 4934. (20) Jiang, D.; Zhao, H.; Zhang, S.; John, R.; Will, G. D. J. Photochem. Photobiol., A 2003, 156, 201.
6034 DOI: 10.1021/la903706e
Figure 1. XRD patterns of BDD (a), pure-anatase TiO2/BDD electrode (b), and mixed-phase TiO2/BDD electrode (c). reference and auxiliary electrodes, respectively. UV illumination was achieved with a 150 W xenon arc lamp light source with focusing lenses (HF-200W-95, Trustech, Beijing, China). To minimize sample heating from the infrared fraction of the Xenon light, the beam was passed through a UV-band-pass filter (UG-5, Schott) prior to illuminating on the electrode surface. Light intensity was measured with a UV-irradiance meter (UVA, Instruments of Beijing Normal University). A voltammograph (CV-27, BAS) was used for application of potential bias in the photoelectrocatalytic and linear potential sweeping experiments. All the steady state currents were obtained at a potential bias of þ0.4 V vs Ag/AgCl and under a light intensity of 6.6 mV/cm2 when the electrode is fixed and the sample solution is static. The surface morphology of the TiO2 samples was observed using scanning electron microscopy (SEM) on a JEOL JSM-6300 F field-emission scanning electron microscope (FESEM, Tokyo, Japan). X-ray diffraction (XRD) was performed with a Philips PW1050 diffractometer using CuKa radiation.
3. Results and Discussion 3.1. Structure Characterization. After the TiO2 sol was dipcoated on conducting substrates such as BDD and ITO electrodes and dried in air, a uniform thin film was formed on the surface of the substrates. The coated substrates were subject to calcination at a high temperature (i.e., 450 or 700 °C) in order to remove the carbowax and improve film adhesion, rigidity, and crystallinity. More importantly, the high temperature calcination facilitates the formation of the TiO2/BDD heterojunction. Our previous work17 reported the formation of a pure anatasephase TiO2 thin film on an ITO substrate by sintering dip-coated TiO2 colloid at 450 °C in air, and mixed antase/rutile phases were obtained by calcination at a higher temperature of 700 °C in air. In this work, we followed a similar procedure to produce a pure anatase-phase TiO2 thin film on a BDD substrate, i.e., sintering at 450 °C in air for 2 h. At 700 °C in air, however, BDD can be damaged because of the oxidation by the oxygen in air. In order to prepare a mixed-phase TiO2/BDD electrode, protective atmosphere (i.e., Ar in this work) must be used to replace the oxidative air. The dip-coated TiO2/BDD electrode was calcinated at 450 °C for 0.5 h in air to eradicate all the carbowax first, and then calcinated in Ar atmosphere at 700 °C for 2 h. Figure 1 shows the XRD patterns of the clean BDD substrate (curve a), and TiO2/BDD electrodes calcinated at 450 °C (curve b) and at 700 °C (curve c). Figure 2 shows their corresponding SEM images. The physicochemical properties of BDD remained intact after the washing process by aqua regia and H2O2, evidenced by the well-defined diamond peak at a 2θ degree of ca. 44° in curve a in Figure 1. This can be also confirmed by the distinguished crackLangmuir 2010, 26(8), 6033–6040
Han et al.
Article
Figure 2. SEM images of BDD (a), pure-anatase TiO2/BDD electrode (b), and mixed-phase TiO2/BDD electrode (c).
free and continuous BDD surface in Figure 2a. The diamond peaks were observed in both curves b and c, suggesting that the BDD crystalline structures were well maintained and have not been altered by the calcination conditions, i.e., 450 °C in air and 700 °C in Ar atmosphere. The diffraction peaks could be indexed to anatase phase (JCPDS No. 89-4921) and rutile-phase (JCPDS No. 89-4920). Only anatase peak (and no rutile peak) is observed for curve (b), while both anatase and rutile peaks were observed for curve (c). This suggests that the TiO2/BDD obtained at 450 °C was mainly the anatase phase, while the electrode calcinated at 700 °C consisted of anatase and rutile phase. Furthermore, the intensity of the diffraction peak of the anatase (101) at 700 °C was significantly stronger than that at 450 °C, suggesting the treatment at 700 °C is in favor of better crystallinity of the TiO2 thin film.17 The electrode obtained at 450 and 700 °C are designated as a pureanatase TiO2/BDD electrode and a mixed-phase TiO2/BDD electrode in the later discussion. Figure 2b,c shows that the pure-anatase TiO2/BDD electrode has similar morphology as the mixed-phase electrode with particles sizes of ca. 20-30 nm, which suggests that the calcination process at 700 °C did not result in the growth of the particle size. The pure-anatase electrode was similar to the one in our previous work where ITO was used as an electrode substrate. The mixedphase electrode, however, is in strong contrast with the electrode in our previous works21 where significant size growth from 20 to 50 nm was observed for the mixed-phase ITO electrode calcinated at 700 °C. In comparison with this work, the substrate for the mixed-phase electrode was ITO aluminosilicate glass, and the calcination process was carried out at 700 °C in air in our previous works. Because the physicochemical properties (such as XRD diffraction and conductivity) of the ITO and the BDD were well maintained in the calcination processes at 700 °C, the difference of the morphology of the mixed-phase TiO2/BDD and TiO2/ITO should not be due to the difference in electrode substrates. It is more likely due to another factor, i.e., the difference in calcination atmosphere, because oxygen in the air can facilitate the rearrangement and growth of TiO2 nanocrystalline.22 In other words, the preservation of the morphology of particle sizes for the (21) Jiang, D. Doctor of Philosophy Thesis, Griffith University, 2004. (22) Walczak, M.; Papadopoulou, E. L.; Sanz, M.; Manousaki, A.; Marco, J. F.; Castillejo, M. Appl. Surf. Sci. 2009, 255, 5267.
Langmuir 2010, 26(8), 6033–6040
Figure 3. LSVs of pure-anatase (a) and mixed-phase (b) TiO2/ BDD electrode in a 0.1 M NaNO3 solution.
TiO2/BDD electrode at 700 °C is mainly due to the use of protective Argon atmosphere. According to the cross-section SEM images analysis (not shown), the thickness of the TiO2 layer on BDD surface is ca. 500 nm. It is almost the same as that on the ITO surface because identical coating TiO2 sol and procedures were used.23 The thickness of the TiO2 layer affects the photocurrent as well as the stability of the TiO2 electrode. Thicker film is beneficial to higher photocatalytic capacity, but becomes less stable especially when it is thicker than 1000 nm. 500 nm is used as an optimum thickness in this work because it possesses reasonable and reproducible photocatalytic activity without sacrificing stability. 3.2. Photoelectrochemical Oxidation of Water. It was well-established that applied potential bias can effectively suppress recombination of photoelectrons and photoholes, and facilitate collecting photoelectrons for the measurement of the photocurrent. The extent of the currents due to the electrochemical oxidation or reduction of water may affect the quality of electrochemical measurements (e.g., analytical signals, detection limits and reproducibility etc.) in aqueous solutions. One of the attractive merits of a BDD electrode is its wide overpotential for water oxidation (2.5 V vs standard hydrogen electrode (SHE)24). In order to confirm whether this merit is maintained, the photoelectrochemical characteristics of the TiO2/BDD electrodes are first studied in the photoelectrochemical oxidation of water. Figure 3 shows a series of linear sweeping voltammograms (LSVs) obtained for the TiO2/BDD electrodes in 0.1 M NaNO3 solution under various UV intensities. In the dark, only a negligible current could be observed for the pure-anatase TiO2/BDD electrode (in Figure 3a) and mixed-phase TiO2/ BDD electrode (in Figure 3b), suggesting that water can not be (23) Zhang, S.; Li, L.; Zhao, H.; Li, G. Sens. Actuators B: Chem. 2009, 141, 634. (24) Panizza, M.; Cerisola, G. Electrochim. Acta 2003, 48, 3491.
DOI: 10.1021/la903706e
6035
Article
electrochemically oxidized in this potential range (-0.3 V to þ0.8 V) and that the BDD electrode retains the properties of wide overpotential for water oxidation after the electrode modification. Under a given UV illumination, the photocurrent increased with the increased potential before leveling off to a saturated photocurrent. The increasing part (i.e., -0.3 to -0.1 V) of photocurrent suggests that electron transport in the TiO2/BDD electrodes was the reaction rate-limiting process, which is similar to the response of a resistor under varying potentials. The saturated part (i.e., -0.1 to þ0.8 V) implies that the potential applied on the TiO2/BDD electrode is capable of removing all the electrons generated in the photocatalytic process at the TiO2 surface.25 Typical saturated photocurrents at both electrodes could be observed when the sweeping potential was greater than -0.1 V, suggesting that the applied potential should be more positive than -0.1 V to maximize the photoelectrocatalytic efficiency of these electrodes. Therefore, to ensure a sufficient potential bias while minimizing the direct electrochemical reaction, the photocurrents of the following experiments were obtained under a constant potential bias of þ0.40 V. Moreover, Figure 3a,b shows that the saturated photocurrents of the mixed-phase TiO2/BDD electrodes are ca. 3 times higher than that of the pure-anatase TiO2/BDD electrode under a given light intensity. The magnitude of the saturated photocurrents represents the maximum reaction rate under a given light intensity,17,25 and can be used to represent the photoelectrocatalytic activity of the electrode. Therefore, it can be concluded that the photoelectrocatalytic activity of the mixed-phase TiO2/BDD electrode is much higher than that of a pure-anatase TiO2/BDD electrode. This may be attributed to the synergetic effect of the mixed-phase TiO2 where the electrons tend to move from a high energy level of anatase phase to a low energy level of rutile phase.17,26 3.3. Photoelectrochemical Oxidation of Organic Compounds. 3.3.1. Photoelectrochemical Oxidation of Weak Adsorbates. Various functional groups of organic compounds result in varying adsorptivity to the TiO2 surface.27-32 They can be classified into weak adsorbates (such as hydroxyl organic compounds, e.g., glucose, phenols, ethanol, etc.) and strong adsorbates (such as organic compounds containing a carboxylic function group, e.g., phathalic acid, glutaric acid, malonic acid, etc.). The photoelectrochemical oxidation of weak adsorbates on TiO2/BDD and TiO2/ITO electrodes was first investigated under an UV illumination intensity of 6.6 mW/cm2. Glucose was selected as a representative of the weak adsorbates. The photocurrent profiles of different glucose concentrations are shown in Figure 4. Under a constant applied potential of þ0.40 V, without the UV illumination, the dark current was approximately zero for all glucose concentrations, which implies that the glucose was not electrochemically oxidized at the surface of all electrodes. Upon illumination, the current increased rapidly, leading to a photocurrent spike. We believe that the spike resulted from the preadsorbed water and glucose, although the adsorbed amount (25) Jiang, D.; Zhao, H.; Zhang, S.; John, R. J. Phys. Chem. B 2003, 107, 12774. (26) Hurum, D. C.; Agrios, A. G.; Gray, K. A.; Rajh, T.; Thurnauer, M. C. J. Phys. Chem. B 2003, 107, 4545. (27) Koeppen, S.; Langel, W. Phys. Chem. Chem. Phys. 2008, 10, 1907. (28) Pettibone, J. M.; Cwiertny, D. M.; Scherer, M.; Grassian, V. H. Langmuir 2008, 24, 6659. (29) Sverjensky, D. A.; Jonsson, C. M.; Jonsson, C. L.; Cleaves, H. J.; Hazen, R. M. Environ. Sci. Technol. 2008, 42, 6034. (30) Zhao, D.; Chen, C.; Wang, Y.; Ji, H.; Ma, W.; Zang, L.; Zhao, J. J. Phys. Chem. C 2008, 112, 5993. (31) Young, A. G.; McQuillan, A. J. Langmuir 2009, 25, 3538. (32) Yu, S.; Ahmadi, S.; Palmgren, P.; Hennies, F.; Zuleta, M.; GoIˆthelid, M. J. Phys. Chem. C 2009, 113, 13765.
6036 DOI: 10.1021/la903706e
Han et al.
Figure 4. Typical photocurrent responses of the mixed phase TiO2/BDD electrode to various concentrations of glucose.
Figure 5. (a) Plot of steady-state photocurrent iss against glucose concentration. (b) Plot of net current inet against glucose concentration, for different electrodes.
of glucose was small. The photocurrents decayed right after the current reached its peak, and subsequently reached steady state, namely, steady state current (iss). Figure 4 also shows a clear trend that iss increased with the increased glucose concentrations. The steady state current iss-glucose concentration profiles of the mixed-phase TiO2/BDD, mixed-phase TiO2/ITO electrodes, pure-anatase TiO2/BDD, and pure-anatase TiO2/ITO electrodes are plotted in Figure 5. For the blank solution, where glucose concentration is zero, the steady-state photocurrent (iblank) resulted mainly from the oxidation of water. Figure 5a indicates clearly that different electrodes have varying iblank. In particular, the mixed-phase TiO2/BDD electrode has a smaller iblank value than the mixed-phase TiO2/ITO; at the same time, the pure-anatase Langmuir 2010, 26(8), 6033–6040
Han et al.
Article
TiO2/BDD electrode has also a smaller iblank than the pureanatase TiO2/ITO. This suggests that TiO2/BDD has a lower background current than ITO when they are used as conducting substrates for TiO2 photocatalyst. In comparison with iblank, iss is the total current of two current components, one from the oxidation of water, which was the same as iblank, and the other from photoelectrocatalytic oxidation of organic compounds (e.g., glucose).11 The net steady state current (inet), i.e., the limiting current, originated from the oxidation of the organics, can be obtained by subtracting iblank from iss in the presence of organic compounds. inet ¼ iss -iblank
ð2Þ
inet can be used to represent the oxidation rate of the corresponding organic compounds (e.g., glucose) and also quantify the concentration of the organic compounds in aqueous samples (e.g., chemical oxygen demand (COD) in wastewater).33 Using eq 2, the inet values of different types of electrodes can be obtained and are plotted against the full range of glucose concentrations in Figure 5b and low glucose concentrations in the inset of Figure 5b. Figure 5b indicated that, for all tested electrodes, inet increased with the increased glucose concentration. More importantly, the inset in Figure 5b shows that, at low glucose concentrations (i.e., lower than 0.3 mM glucose for the pure-anatase electrodes and lower than1.0 mM glucose for the mixed-phase electrodes), the changes of inet values are linear with glucose concentrations. This is mainly because the mass transport of organics to the electrode surface in this concentration range was a limiting step in the photoelectrocatalytic oxidation process.34 This is an important characteristic that can be used for sensing application. The fact that mixed-phase electrodes had a higher linear upper limit (1.0 mM glucose) than the pure-anatase electrode (i.e., 0.3 mM glucose) suggests that the mixed-phase TiO2 electrode has a higher photoelectrocatalytic oxidation capacity to glucose than the pure anatase TiO2 electrode, which is consistent with the aforementioned observation in the photoelectrocatalytic oxidation of water. At higher glucose concentrations, the inet values deviate from the linear relationship. This implies that the photoelectrocatalytic oxidation reaction at the electrode surface becomes a rate-limiting step under the given light intensity.17 Figure 5b also indicates that TiO2/BDD electrodes have higher photocurrents than the corresponding TiO2/ITO electrodes, suggesting that the photoelectrocatalytic activity of TiO2/BDD electrodes was higher than that of the corresponding TiO2/ITO in high glucose concentration range (i.e., >1.0 mM). Overall, in comparison with ITO electrodes, TiO2/BDD electrodes have a lower background photocurrent, similar detection linear range of glucose, and slightly higher photoelectrocatalytic oxidation efficiency to high concentration of glucose. This may be because glucose is an easy oxidizable organic compounds. In order to further differentiate the photocatalytic activity of the resulting TiO2 electrodes, a more persistent organic compound needs to be used as the degradation substrate. 3.3.2. Photoelectrochemical Oxidation of Strong Adsorbates. Potassium hydrogen phthalate (KHP) is a common strong adsorbate that has been studied and reported extensively in the (33) Zhang, S.; Li, L.; Zhao, H. Environ. Sci. Technol. DOI: 10.1021/es901320a. (34) Wen, W.; Zhao, H.; Zhang, S. J. Phys. Chem. C 2009, 113, 10830. (35) Valente, J. P. S.; Padilha, P. M.; Florentino, A. O. Chemosphere 2006, 64, 1128. (36) Zhang, S.; Wen, W.; Jiang, D.; Zhao, H.; John, R.; Wilson, G. J.; Will, G. D. J. Photochem. Photobiol., A 2006, 179, 305.
Langmuir 2010, 26(8), 6033–6040
Figure 6. Typical photocurrent responses of the mixed-phase TiO2/BDD electrode to various concentrations of KHP.
literature.35,36 It is considered a challenging task to degrade KHP because it is chemically stable and it can even poison a TiO2 surface.36,37 Photoeletrocatalytic degradation of KHP was carried out under the same experimental conditions as in the oxidation of glucose in Figure 4. Figure 6 shows photocurrent profiles for the oxidation of various concentrations of KHP at mixed-phase TiO2/BDD electrode. In a low concentration range (i.e., 0-0.04 mM KHP), well-defined photocurrent spikes were observed upon the UV illumination. The photocurrents declined monotonically until steady state was reached, which is similar to the situation of glucose degradation in Figure 4. In strong contrast with Figure 4, the spikes of KHP samples were a lot stronger than those of glucose samples, which is apparently caused by the substantial preadsorption of KHP at the electrode surface. In a high KHP concentration range (i.e., >0.04 mM KHP), the photocurrents spiked, dropped, increased, and finally decayed to steady states, resulting in a shoulder peak-like profile. This suggests that the photoelectrocatalytic degradation of KHP is a lot more sophisticated than that of glucose. This phenomenon was very likely due to accumulation and polymerization of intermediates during the oxidation of aromatic compounds present in high concentrations. The initial decrease in photocurrent after the spike was possibly due to the formation of polymer networks caused by the polymerization of the fragments and free radicals at the electrode surface. These polymer networks may act as a mass transport barrier, diminishing the mobility of KHP and its intermediates, reducing the access of the organic compounds to the electrode surface for the degradation reaction, and therefore resulting in the decreased photocurrent. As the degradation reaction and the consumption of adsorbed KHP and its intermediates at the electrode surface continue, the polymer network (i.e., mass transport barrier) is torn off, amd the normal mass transport of KHP to reaction active sites (i.e., photo holes) is restored, resulting in an increased photocurrent. The subsequent photocurrent decrease and the attainment of the steady states apparently corresponded to the decline in residue intermediates at the electrode surface and diffusion of KHP from bulk solution to the electrode surface, respectively. The appearance time for the shoulder peak-like profile extended with the increased KHP concentration, which was very likely due to the accumulation and polymerization of intermediates. This phenomenon of shoulder peak happened at lower concentration of KHP (i.e., 0.12 mM, see Figure 6) than that of glucose (i.e., 1.0 mM, see Figure 4). This is mainly because KHP is an aromatic compound (37) Calvo, M. E.; Candal, R. J.; Bilmes, S. A. Environ. Sci. Technol. 2001, 35, 4132.
DOI: 10.1021/la903706e
6037
Article
Han et al.
Figure 7. Relationship between the photocurrent and concentration of KHP on different electrodes.
with a chemically stable benzene ring that is more difficult to oxidize, and easier to form stable aromatic intermediates, leading to the polymerization reaction and poisoning effect. Despite the problem of the intermediate polymerization at the electrode surface, steady-state photocurrents can be obtained in a relatively short time period (ca. 1 min) when the KHP concentration was relatively low. Under the same UV light intensity and applied potential bias, the inet values of the TiO2/BDD electrodes and TiO2/ITO electrodes were obtained by subtracting the background current and plotted against KHP concentration as shown in Figure 7. The inet values of all electrodes linearly increased with the increased KHP concentrations in a low concentration range (i.e., 0.15 mM). This is mainly because the inet values were obtained under the mass transport limitation of KHP in the low concentration range. After that, the inet values of different TiO2 electrodes reached various maximum values at different upper linear concentration limits, and then dropped with the increased KHP concentration. The upper limits were found to be ca. 0.15, 0.25, 0.35, and 0.5 mM for pure-anatase TiO2/ITO, pure-anatase TiO2/BDD, mixed-phase TiO2/ITO, and mixed-phase TiO2/BDD electrodes, respectively. When the KHP concentration is higher than the upper linear limits, the KHP supplied by diffusion from solution surpasses the KHP consumed by the photoelectrocatalytic oxidation, which leads to the aforementioned accumulation and polymerization of the intermediates at the electrode surface. Different from the situation of the glucose oxidation, the electrodes will not be able to remove this polymer network, because KHP is a much more persistent organic compound than glucose. This continuous oversupply of KHP will lead to an apparent inhibition effect: photocurrents decrease with the increased KHP concentrations. Electrodes with higher photoelectrocatalytic activities will be able to handle KHP at a faster reaction rate and therefore possess higher upper linear limits. In Figure 7, the upper linear limits of the mixed-phase TiO2/BDD and TiO2/ITO electrodes are higher than that of the pure-anatase TiO2/BDD and TiO2/ITO electrodes, respectively. This demonstrates that the mixed-phase TiO2 electrodes have higher photoelectrocatalytic oxidation power than the pure-anatase electrode due to the aforementioned synergetic effect of rutile and anatase, which is consistent with the observation in glucose oxidation. The upper linear limits of the mixed-phase and pure-anatase TiO2/BDD electrodes were higher than that of TiO2/ITO electrodes, respectively, suggesting that the mixed-phase and pureanatase TiO2/BDD electrodes have higher photoelectrocatalytic activity than the mixed-phase and pure-anatase TiO2/ITO 6038 DOI: 10.1021/la903706e
Figure 8. (a) Relationship between inet and molar concentration CM, (b) relationship between inet and equivalent concentration Ceq, where Ceq (meq) = nCM (mM) in 0.1 M NaNO3 solution (ca. pH 6).
electrodes, respectively. This is very likely due to the formation of TiO2/BDD heterojunction where BDD conducting substrates are used. 3.4. Sensing Application. 3.4.1. Detection of Various Organics. The above investigations clearly demonstrate that the mixed-phase TiO2/BDD has the best photoelectrocatalytic activities among the TiO2/BDD and TiO2/ITO electrodes. The mixedphase TiO2/BDD electrode is therefore used to oxidize various organic compounds in order to demonstrate its suitability of being used for sensing of organic compounds. Figure 8a shows the relationship between the inet values and the concentrations of a group of organic compounds. It was found that the inet values obtained are directly proportional to the molar concentration of the tested organic compounds. The molecular structures and electron transfer numbers (n) during the exhaustive oxidization according to eq 1 for the organics studied in this work are listed in Table 1. The equivalent concentrations (Ceq) can be obtained by multiplying the molar concentrations (CM) with its corresponding electron transfer number, i.e., Ceq = nCM. Subsequently, inet for each organic compound was normalized by replacing CM with Ceq17 and shown in Figure 8b. Meanwhile, the photocurrent from oxidation of a mixture sample containing all the above tested organic compounds with the same equivalent concentration was also displayed in Figure 8b. Excellent linear relationship (R2 = 0.9909) between the inet values and Ceq values of the organic compounds was obtained. This suggests that the mixed-phase TiO2/BDD electrode is able to indiscriminately oxidize various types of organic compounds. This is a significant Langmuir 2010, 26(8), 6033–6040
Han et al.
Article
Table 1. The Molecular Structures and the Electron Transfer Numbers (n) of Selected Organics
Figure 9. Effect of pH on the photocurrents of the mixed-phase TiO2/BDD electrode.
characteristic for sensing application, because this allows the electrode to be used as a universal detector for individual organic compound as well as a specialized detector for a group of organic compounds, such as COD, an important aggregative parameter for wastewater. The mixed-phase TiO2/BDD electrode has excellent stability and reproducibility in the determination of organic compounds. A total of 95-105% of the net currents for the same concentration of organic compound was maintained after the electrode had been used continuously for 2 months. The relative standard deviation (RSD%) value was found to be 0.9% for the 15 replicated measurements of 0.2 mM glucose. 3.4.2. Effect of pH on Photocurrents. Real samples containing organics commonly have various pH. An ideal organic sensor should be sensitive to organic compounds while being insensitive to the property of the detection matrix, such as solution pH. However, the solution pH influences not only the chemical form of target organic compounds but also the speciation of TiO2 surface.38,39 In order to further explore the potential to use the mixed-phase TiO2/BDD electrode for sensing applications and ensure the reliability of its analytical results, the effect of pH on (38) Zhang, F.; Pi, Y.; Cui, J.; Yang, Y.; Zhang, X.; Guan, N. J. Phys. Chem. C 2007, 111, 3756. (39) Kormann, C.; Bahnemann, D. W.; Hoffmann, M. R. Environ. Sci. Technol. 1991, 25, 494.
Langmuir 2010, 26(8), 6033–6040
Figure 10. LSVs of the mixed-phase TiO2/ITO electrode (a) and the mixed-phase TiO2/BDD electrode (b) in 0.1 M NaNO3 solution. The solution was adjusted to pH 1.0 by HNO3 solution.
the photocurrents was also investigated. KHP has been selected as a target analyst because KHP is a commonly used organic standard for analytical measurements, such as COD and TOC analysis.40 Figure 9 shows the effect of pH on the steady state photocurrents of the mixed-phase TiO2/BDD electrode in the 0.1 M NaNO3 solution, and 0.1 M NaNO3 solution with 0.2 mM KHP. When pH is lower than 2, the iss and iblank values for water were slightly lower than the iss and iblank values at pH 2. As a (40) Rand, M. C.; Greemberg, A. G.; Taras, M. J. Standard Methods for Examination of Water and Wastewater, 20th ed.; American Public Health Association: Washington, DC, 1998.
DOI: 10.1021/la903706e
6039
Article
result, the inet for KHP was relatively stable. While pH ranged from 2 to 10, both the iss and iblank were fairly constant, and as a result, the inet values were insensitive to the change in solution pH. For the solution where pH is higher than 11.0, both the iss and iblank dramatically increased, causing the inet to change significantly, which was due to the dramatic increment of water oxidation at high pH,32 evidenced by the production of oxygen and hydrogen bubbles. Furthermore, the reproducibility of the measurements was very poor at such a high pH, which resulted from the mass transport interference at the electrode surface from the produced oxygen and hydrogen bubbles. Overall, the electrodes are suitable for application of sensing organics in the pH range between 2 and 10. 3.4.3. Acidoresistance. One of the driving motives to develop TiO2/BDD electrode is its robustness and stability in an acidic environment, which is an advantage for the electrode to be used as a sensor. The stability of the mixed-phase TiO2/BDD electrode was investigated by comparing with the mixed-phase TiO2/ITO electrode using the LSV method. The investigation was conducted by sweeping the potential from -0.3 to þ0.8 V at a scanning rate of 5 mV/s for 40 cycles in 0.1 M NaNO3 solution. Figure 10a shows that the photocurrent of the mixed-phase TiO2/ITO electrode dropped dramatically with the repeated sweeping cycles, and the photocurrent diminished to zero after 40 cycles. The lost of photoactivity of the TiO2/ITO electrode in acidic solution was attributed to the film damage caused by the dissolution of ITO film in the acid solution,41 which coincides with the fact that the detachment of TiO2 film from the ITO substrate was visibly observed after the experiment. The TiO2/BDD electrodes were tested under identical experimental conditions. Figure 10b shows that the photocurrents of mixed-phase TiO2/BDD electrode were (41) Stotter, J.; Show, Y.; Wang, S.; Swain, G. Chem. Mater. 2005, 17, 4880.
6040 DOI: 10.1021/la903706e
Han et al.
more stable throughout the whole experimental process. The photocurrents were kept between 60 and 65 μA, and, in particular, the LSVs of the mixed-phase TiO2/BDD electrode remained almost identical from the 20th to the 40th cycle. This demonstrates that the mixed-phase TiO2/BDD electrodes possess advantages of greater acid resistance in comparison with the TiO2/ ITO electrode. The photoelectrocatalytic activity toward the oxidation of KHP in the latter experiment was found to be unaffected before and after the acid exposure.
4. Conclusions TiO2/BDD heterojunction photoanodes and TiO2/ITO electrodes were prepared by using a dip-coating method and calcined under different temperature. The photoelectrocatalytic activities of the electrodes were evaluated in the photoelectrocatalytic oxidation of water, glucose, and KHP using photoelectrochemical means. The preliminary results demonstrate that the mixedphase TiO2/BDD electrode exhibits the highest photoelectrocatalytic activity among all the TiO2 electrodes tested. In comparison with the mixed-phase TiO2/ITO electrode, the mixedphaseTiO2/BDD electrode has the advantage of high resistance against acidic corrosion. The mixed-phase TiO2/BDD electrode was subsequently used to detect a wider spectrum of organic compounds in aqueous solution using the steady state current method. The inet was directly proportional to the concentration of organic compounds. The resulting mixed-phase TiO2/BDD heterojunction electrodes have great potential to be used as sensors to detect organic compounds in aqueous solutions. Acknowledgment. The authors acknowledge the financial support of the ARC discovery and ARC linkage grants from the Australian Research Council.
Langmuir 2010, 26(8), 6033–6040