pubs.acs.org/NanoLett
Ta3N5 Nanotube Arrays for Visible Light Water Photoelectrolysis Xinjian Feng,† Thomas J. LaTempa,‡ James I. Basham,‡ Gopal K. Mor,† Oomman K. Varghese,† and Craig A. Grimes*,†,‡ †
Materials Research Institute and ‡ Department of Electrical Engineering The Pennsylvania State University, University Park, Pennsylvania 16802 ABSTRACT Tantalum nitride (Ta3N5) has a band gap of approximately 2.07 eV, suitable for collecting more than 45% of the incident solar spectrum energy. We describe a simple method for scale fabrication of highly oriented Ta3N5 nanotube array films, by anodization of tantalum foil to achieve vertically oriented tantalum oxide nanotube arrays followed by a 700 °C ammonia anneal for sample crystallization and nitridation. The thin walled amorphous nanotube array structure enables transformation from tantalum oxide to Ta3N5 to occur at relatively low temperatures, while high-temperature annealing related structural aggregation that commonly occurs in particle films is avoided. In 1 M KOH solution, under AM 1.5 illumination with 0.5 V dc bias typical sample (nanotube length ≈ 240 nm, wall thickness ≈ 7 nm) visible light incident photon conversion efficiencies (IPCE) as high as 5.3% were obtained. The enhanced visible light activity in combination with the ordered one-dimensional nanoarchitecture makes Ta3N5 nanotube arrays films a promising candidate for visible light water photoelectrolysis. KEYWORDS Hydrogen, water photoelectrolysis, tantalum nitride, tantalum, photolysis
T
highly ordered oxide nanotube arrays have been fabricated by anodization of metals and metal alloys, including Ti,12-15 Ta,16,17 Ti45Nb,18 Fe,19 FeTi,20 and CuTi.21 TiO2 nanotube arrays, obtained via Ti anodization, have been successfully used in numerous solar energy applications including dyesensitized solar cells,22-24 heterojunction and hybrid solar cells,25,26 as a photocatalyst for the conversion of CO2 into hydrocarbon fuels,27 and as a water-splitting photoanode for hydrogen generation under UV illumination where incident photon conversion efficiency (IPCE) values of over 80% are commonly obtained.10 Looking toward a high-performance photocorrosion stable photocatalyst with inherent visible light activity, in this work we report on tantalum nitride (Ta3N5) nanotube arrays with a band gap of approximately 2.07 eV suitable for collecting more than 45% of the incident solar spectrum energy. Ta3N5 demonstrates good hydrolytic stability with a band gap position appropriate for overall water splitting28-35 and has a maximum theoretical solar spectrum photoconversion efficiency of 15.9% making it a viable candidate for the solar generation of hydrogen by water photoelectrolysis.29-32 Herein we report synthesis of well-aligned nanotube arrays of Ta3N5 with lengths up to 10 µm and wall thicknesses as small as 7 nm. The development of the ordered nanotube array architecture in Ta3N5 is expected to improve charge transport properties and reduce charge carrier recombination. Under visible light illumination, Ta3N5 nanotube arrays 240 nm in length with 7 nm wall thickness and in 1 M KOH solution under 0.5 V dc bias, demonstrate IPCE values as high as 5.3%. Experimental Section. A two-step process was used to prepare the Ta3N5 nanotube arrays. First, tantalum oxide
here have been significant efforts over the past thirty years to develop new materials for the solar production of hydrogen by water photoelectrolysis.1-4 However, scale implementation of water photoelectrolysis systems have proven unsuccessful since the majority of photocorrosion stable metal oxides are wide band gap materials, limiting their activation to the ultraviolet region that contains but a small fraction of the incident solar spectrum energy. Motivated by the pioneering work of Fujishima and Honda,1 photocorrosion-stable titanium dioxide (TiO2) has been one of the most commonly investigated semiconductors for water photoelectrolysis. Considerable effort has been spent in trying to reduce the TiO2 band gap (anatase, 3.2 eV) by anionic impurity doping, including nitrogen,5,6 boron,7 or carbon.8 Yet these dopants introduce charge carrier recombination centers,9 consequently only marginal improvements have ever been observed in solar spectrum performance. In addition to intrinsic material properties, the nanoscale architecture of a material plays a critical role in determining the maximum photoconversion efficiency that can be achieved. One-dimensional (1D) semiconductor nanotube arrays constitute a material architecture that offers a large internal surface area with lengths sufficient to effectively capture incident illumination in combination with minimal radial dimensions providing facile separation of photogenerated charge. The oriented nature of the nanotube arrays makes them excellent electron percolation pathways for charge transfer between interfaces.10,11 Vertically oriented,
* To whom correspondence should be addressed. E-mail:
[email protected]. Received for review: 11/20/2009 Published on Web: 02/10/2010 © 2010 American Chemical Society
948
DOI: 10.1021/nl903886e | Nano Lett. 2010, 10, 948–952
FIGURE 1. FE-SEM images of tantalum oxide nanotube arrays. Panels a and b are, respectively, top and bottom-surface images of the as self-organized tantalum nanotube layer. Panels c and d are cross-sectional views of the nanotube arrays at low and higher magnification, respectively. The nanotube arrays shown here were anodized at 15 V for 30 min at room temperature.
nanotube arrays were prepared by anodization of tantalum foil at 15 V for different times at room temperature. Electrolytes composed of a concentrated H2SO4 and HF mixture (9:1 v/v) have previously been reported for the preparation of Ta2O5 nanotubes.16,17 However, the large volume of concentrated HF solution used in earlier experiments generally induced a violent electrochemical reaction with the nanotube arrays separating from the underlying substrate due to interfacial stress between the oxide and metal layers. In the present work, DI water was used to replace 85% of the HF solution allowing the anodization reaction to be readily controlled over anodization durations of several hours with the resulting tantala nanotube arrays homogenously formed over large sample sizes and strongly adhered to the substrate. Ta3N5 nanotube arrays were prepared by annealing the amorphous tantala nanotubes in ammonia at 700 °C for 10 h. Figure 1 panels a-d are typical top, bottom, and crosssectional field emission scanning electron microscope (FESEM, JEOL JSM-6300, Japan) images of the as anodized tantala nanotube arrays prepared by anodization at 15 V for 30 min. It is readily apparent that the tubes are open at the top surface and closed on the bottom layer; these nanotubes have an average outer diameter of 45 nm and a length of 6 µm. The nanotube length increases with anodization duration, but the pore diameter and wall thickness remain constant with applied potential, similar to the formation of anodized nanotube arrays on titanium foil. Glancing angle X-ray diffraction (GAXRD, Scintag Inc., CA), see Figure 2, reveals that the as-formed oxide nanotubes are amorphous. Unlike previous reports on Ta anodization,16,17 it was found © 2010 American Chemical Society
FIGURE 2. Glancing angle X-ray diffraction (GAXRD) patterns of as received tantalum foil (black line), as self-organized tantala nanotube arrays (gray line), Ta2O5 nanotube arrays, and Ta3N5 nanotube arrays after being annealed at 700 C° for 10 h in ammonia atmosphere.
that very stable and homogeneous nanotube arrays films can be prepared over large surface areas, a necessary prerequisite for practical device implementation. Nanotube growth is a result of several competing affects including fieldassisted metal oxidation, chemical dissolution, and fieldassisted dissolution.12-15 The introduction of water and the reduction of HF concentration both impact the growth behavior. When electrolytes containing high HF concentrations were used the violent electrochemical reaction caused the nanotube film to peel off the substrate in less than 120 s, a behavior most likely due to the large amounts of stress between the Ta foil and the rapidly formed tantalum oxide film. The smaller amount of HF solution used in the current experiments makes the nanotube array growth readily 949
DOI: 10.1021/nl903886e | Nano Lett. 2010, 10, 948-–952
FIGURE 3. Panels a and b are, respectively, FE-SEM top and bottomsurface images of the as prepared Ta3N5 nanotube arrays films. Panels c and d are TEM images of the nanotube arrays at low and higher magnification, respectively. The insert in Figure 2d is a selected area electron diffraction pattern from the corresponding region.
controllable with the as-grown film stable even after several hour anodization durations. Figure 3 panels a and b are FESEM images of nanotube arrays after the ammonia annealing treatment, showing that the tubular structure was not affected by the high temperature nitridation process. Figure 3 panels c and d are, respectively, transmission electron microscope (TEM, JEOF 2010 F, Japan) images of the nanotubes at low and high resolution. The nanotubes have an average inner pore diameter ∼22 nm and a wall thickness of about 7 nm. A selected area electron diffraction pattern from the corresponding region is shown in the insert of Figure 3d indicating that the nanotube is highly crystallized. The asprepared nanotube arrays can be classified as Ta3N5 (JCPDF file no. 97-007-6460) from the GAXRD pattern as shown in Figure 2. For films made from particles, it is commonly observed that the small particles will merge together forming a larger one after high temperature. In contrast, no nanotube aggregation was observed with the ammonia annealing. It was found that the temperature required to form Ta3N5 nanotubes in the present experiment is about 150 °C lower than the temperatures previously reported, 850-900 °C, required for conversion of particulate Ta2O5 films to Ta3N5.29-32 In the present experiment, we transformed the amorphous phase tantala directly to Ta3N5, with the substitution of N3- for O2occurring at a relatively low temperature. It has been observed that the conversion of Ta2O5 into Ta3N5 is highly sensitive to the precursor crystallite size with smaller particles, studies have investigated particles down to 20 nm diameter, requiring lower conversion temperatures.33 As can be seen from Figures 1 and 3, the wall thickness of our tantalum oxide nanotubes is less than 10 nm; it © 2010 American Chemical Society
FIGURE 4. (a) Photographs of gray colored Ta2O5 (left) and red colored Ta3N5 (right) films. (b) Diffuse reflectance UV-vis spectra of nanotube films with different thickness after ammonia annealing. The Ta2O5 film was obtained by annealing the as anodized tantala nanotube arrays at 550 °C for 30 min in air.
appears the thin wall structure helps enable the lower temperature conversion. Figure 4a is a photograph showing the sample color before and after ammonia treatment. Figure 4b shows the UV-vis diffuse reflectance spectra of the Ta3N5 nanotube array films as a function of thickness (Perkin-Elmer Lambda 950 spectrophotometer). The absorbance intensity increases with the increase of the film thickness. The absorption edge of Ta3N5 is located at approximately 600 nm, red shifted by ∼300 nm from that of Ta2O5 nanotube array films (obtained by annealing the as-anodized tantala nanotube array film in air at 550 °C for 30 min), in agreement with the band gap energy estimated to be 2.07 eV. The conduction and valence bands of Ta3N5 are, respectively, attributable to the Ta 5d and N 2p orbitals. The substitution of N for O results in the higher negative potential of the valence band and the narrowing of the band gap compared to that of Ta2O5, because N 2p orbitals have higher potential energies than O 2p. Figure 5 shows the incident photon conversion efficiency (IPCE) of Ta3N5 nanotube array films of different thicknesses, calculated using the equation 950
DOI: 10.1021/nl903886e | Nano Lett. 2010, 10, 948-–952
FIGURE 6. Photocurrent action spectra (IPCE vs wavelength) for Ta3N5 nanotube array film electrodes approximately 750 nm long observed in 1 M KOH under AM 1.5 illumination at different applied dc biases.
FIGURE 5. Photocurrent action spectra (IPCE vs wavelength) for Ta3N5 nanotube array film electrodes of different thickness observed in 1 M KOH at an applied dc potential of 0.5 V.
IPCE (%) ) (1240 eV·nm) × (photocurrent density mA/cm2) × 100 (wavelength·nm) × (irradiance mW/cm2)
for Ta3N5, which can be attributed in part to its lower conduction band edge.19 Our results indicate that Ta3N5 nanotube array films, with an aligned 1D nanostructure, visible light activity, and suitable band edges positions, offer an exciting potential as a material architecture for overall water splitting. Conclusions. We report on a simple method for the fabrication of highly oriented Ta3N5 nanotube array films by anodization of tantalum foil samples, followed by ammonia treatment for crystallization and nitridation of the resulting films. Enhanced visible light activity together with a 1D nanoarchitecture makes Ta3N5 nanotube arrays films an exciting candidate for visible light water photoelectrolysis. The thin wall structure of the amorphous nanotube arrays enables transformation into Ta3N5 at significantly, ∼150 °C, lower temperatures than previously reported for tantala particle films without structural aggregation. Incident photon conversion efficiencies as high as 5.3% were achieved for 240 nm long Ta3N5 nanotube arrays (0.5 V bias, 1 M KOH solution). We are currently investigating means for more precise control of the annealing conditions as a means to reduce surface defect and increase the photoconversion efficiency, and as a possible means for obtaining TaON.38-44
The experiment was performed in a two-electrode arrangement36 with the Ta3N5 nanotube array film as the working photoelectrode and platinum foil as a counter electrode at 0.5 V applied dc bias in 1 M KOH solution. An IPCE of 5.3% is obtained at a wavelength of about 450 nm with film thickness of 240 nm. The IPCE values are in good agreement as a function of wavelength with the UV-vis diffuse reflectance spectrum of the corresponding Ta3N5 nanotube films shown in Figure 4b, confirming that the photocurrent occurs as a result of the band gap transition. However, unlike the UV-vis absorbance, the IPCE decreases with increasing film thickness, which can be attributed to the greater distance the electrons (majority charge carriers) travel to the back contact that, in turn, may cause an increase in electron-hole pair recombination; as reported by Maeda and co-workers it may be possible to suppress surface defect formation during nitridation by the addition of ZrO2.37 Figure 6 shows IPCE values corresponding to four different applied voltages for Ta3N5 nanotube array films 750 nm thick. It is worth noting that IPCE and photocurrent values were observed without electrical bias, showing that the conduction band of Ta3N5 is high enough for water reduction with the present experimental conditions. These observations are consistent with electrochemical analyses and ultraviolet photoelectron spectroscopy measurements performed on Ta3N5 particle films.30 The applied photoanode bias assists separation of the electron-hole pairs, thereby enhancing the IPCE; appreciable values are obtained in the visible light range from 400 to 600 nm. In comparison, TiO2 and SrTiO3, which are among the most widely studied semiconductors for water splitting, only respond to UV illumination due to their larger bandgap energies. Hematite (R-Fe2O3) with a band gap energy of ∼2.1 eV is an attractive candidate for visible light water splitting; however, its measured IPCE values are significantly lower than those obtained © 2010 American Chemical Society
Acknowledgment. Support of this work by the Department of Energy, Grant DE-FG36-08GO18074 is gratefully acknowledged. We thank Trevor Clark of the Materials Research Institute, Penn State University, for help with the TEM images as well as helpful discussions. REFERENCES AND NOTES (1) (2) (3) (4)
951
Fujishima, A.; Honda, M. Nature 1972, 238, 37–38. Bak, T.; Nowotny, J.; Rekas, M.; Sorrell, C. C. Int. J. Hydrogen Energy 2002, 27, 991–1022. Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. U.S.A. 2006, 103, 15729–15735. (a) Grimes, C. A.; Varghese, O. K.; Ranjan, S. Light, Water, Hydrogen: The Solar Generation of Hydrogen by Water Photoelectrolysis; Springer: New York, 2008. (b) Grimes, C. A.; Mor, G. K. TiO2 Nanotube Arrays: Synthesis, Properties, and Applications; Springer: New York, 2009. (c) Shankar, K.; Basham, J. I.; Allam, N. K.; Varghese, O. K.; Mor, G. K.; Feng, X. J.; Paulose, M.; Seabold, J. A.; Choi, K. S.; Grimes, C. A. J. Phys. Chem. C 2009, 113, 6327–6359. DOI: 10.1021/nl903886e | Nano Lett. 2010, 10, 948-–952
(5) (6) (7) (8) (9) (10)
(11) (12) (13) (14)
(15) (16) (17) (18) (19) (20) (21)
(22) (23)
Asahi, R.; Morikawa, T.; Ohwaki, T.; Aoki, K.; Taga, Y. Science 2001, 293, 269–271. Burda, C.; Lou, Y.; Chen, X.; Samia, A. C. S.; Stout, J.; Gole, J. L. Nano Lett. 2003, 3, 1049–1051. Zhao, W.; Ma, W.; Chen, C.; Zhao, J.; Shuai, Z. J. Am. Chem. Soc. 2004, 126, 4782–4783. Sakthivel, S.; Kisch, H. Angew. Chem., Int. Ed. 2003, 42, 4908– 4911. Serpone, N. J. Phys. Chem. B 2006, 110, 24287–24293. (a) Mor, G. K.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Nano. Lett. 2005, 5, 191–195. (b) Varghese, O. K.; Paulose, M.; Shankar, K.; Mor, G. K.; Grimes, C. A. J. Nanosci. Nanotech. 2005, 5, 1158–1165. Zhu, K.; Neale, N. R.; Miedaner, A.; Frank, A. J. Nano. Lett. 2007, 7, 69–74. Gong, D.; Grimes, C. A.; Varghese, O. K.; Hu, W.; Singh, R. S.; Chen, Z.; Dickey, E. C. J. Mater. Res. 2001, 16, 3331–3334. Cai, Q.; Paulose, M.; Varghese, O. K.; Grimes, C. A. J. Mater. Res. 2005, 20, 230–236. Paulose, M.; Shankar, K.; Yoriya, S.; Prakasam, H. E.; Varghese, O. K.; Mor, G. K.; Latempa, T. A.; Fitzgerald, A.; Grimes, C. A. J. Phys. Chem. B 2006, 110, 16179–16184. Grimes, C. A. J. Mater. Chem. 2007, 17, 1451–1457. Allam, N. K.; Feng, X. J.; Grimes, C. A. Chem. Mater. 2008, 20, 6477–6481. Ei-Sayed, H. A.; Birss, V. I. Nano. Lett. 2009, 9, 1350–1355. Ghicov, A.; Aldabergenova, S.; Tsuchyia, H.; Schmuki, P. Angew. Chem., Int. Ed. 2006, 45, 6993–6996. LaTempa, T. J.; Feng, X.; Paulose, M.; Grimes, C. A. J. Phys. Chem. C 2009, 113, 16293–16298. Mor, G. K.; Prakasam, H. E.; Varghese, O. K.; Shankar, K.; Grimes, C. A. Nano Lett. 2007, 7, 2356–2364. Mor, G. K.; Varghese, O. K.; Wilk, H. H. T.; Sharma, S.; Shankar, K.; Latempa, T. J.; Choi, K. S.; Grimes, C. A. Nano. Lett. 2008, 8, 1906–1911. Mor, G. K.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Nano. Lett. 2006, 6, 215–218. Shankar, K.; Bandara, J.; Paulose, M.; Wietasch, H.; Varghese, O. K.; Mor, G. K.; Latempa, T. J.; Thelakkat, M.; Grimes, C. A. Nano. Lett. 2008, 8, 1654–1659.
© 2010 American Chemical Society
(24) Varghese, O. K.; Paulose, M.; Grimes, C. A. Nat. Nanotechnol. 2009, 4, 592–597. (25) Mor, G. K.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Appl. Phys. Lett. 2007, 91, 152111. (26) Mor, G. K.; Kim, S.; Paulose, M.; Varghese, O. K.; Shankar, K.; Basham, J.; Grimes, C. A. Nano Lett. 2009, 9, 4250–4257. (27) Varghese, O. K.; Paulose, M.; LaTempa, T. J.; Grimes, C. A. Nano. Lett. 2009, 9, 731–737. (28) Fang, C. M.; Orhan, E.; de Wijs, G. A.; Hintzen, H. T.; de Groot, R. A.; Marchand, R.; Saillard, J. Y.; de With, G. J. Mater. Chem. 2001, 11, 1248–1252. (29) Hitoki, G.; Ishikawa, A.; Takata, T.; Kondo, J. N.; Hara, M.; Domen, K. Chem. Lett. 2002, 7, 736–737. (30) Chun, W. J.; Ishikawa, A.; Fujisawa, H.; Takata, T.; Kondo, J. N.; Hara, M.; Kawai, M.; Matsumoto, Y.; Domen, K. J. Phys. Chem. B , 107, 1798–1803. (31) Hara, M.; Chiba, E.; Ishikawa, A.; Takata, T.; Kondo, J. N.; Domen, K. J. Phys. Chem. B 2003, 107, 13441–13445. (32) Ishikawa, A.; Takata, T.; Kondo, J. N.; Hara, M.; Domen, K. J. Phys. Chem. B 2004, 108, 11049–11053. (33) Zhang, Q. H.; Gao, L. Langmuir 2004, 20, 9821–9827. (34) Nakamura, R.; Tanaka, T.; Nakato, Y. J. Phys. Chem. B 2005, 109, 8920–8924. (35) Murphy, A. B.; Barnes, P. R. F.; Randeniya, L. K.; Plumb, I. C.; Grey, I. E.; Horne, M. D.; Glasscock, J. A. Int. J. Hydrogen Energy 2006, 31, 1999–2017. (36) Varghese, O. K.; Grimes, C. A. Sol. Energy Mater. Sol. Cells 2008, 92, 374–384. (37) Maeda, K.; Terashima, H.; Kase, K.; Higashi, M.; Tabata, M.; Domen, K. Bull. Chem. Soc. Jpn. 2008, 8, 927–937. (38) Brauer, G.; Weidlein, J. Angew. Chem., Int. Ed. 1965, 4, 875–876. (39) Jansen, M.; Letschert, H. P. Nature 2000, 404, 980–982. (40) Guenther, E.; Jansen, M. Mater. Res. Bull. 2001, 36, 1399–1405. (41) Hitoki, G.; Takata, T.; Kondo, J. N.; Hara, M.; Kobayashi, H.; Domen, K. Chem. Commun. 2002, 1698–1699. (42) Ito, S.; Thampi, K. R.; Comte, P.; Liska, P.; Gra¨tzel, M. Chem. Commun. 2005, 268–270. (43) Schilling, H.; Stork, A.; Irran, E.; Wolff, H.; Bredow, T.; Dronskowski, R.; Lerch, M. Angew. Chem., Int. Ed. 2007, 46, 2931–2934. (44) Banerjee, S.; Mohapatra, S. K.; Misra, M. Chem. Commun. 2009, 7137–7139.
952
DOI: 10.1021/nl903886e | Nano Lett. 2010, 10, 948-–952
