13028
J. Phys. Chem. C 2007, 111, 13028-13032
Formation of Vertically Oriented TiO2 Nanotube Arrays using a Fluoride Free HCl Aqueous Electrolyte Nageh K. Allam† and Craig A. Grimes*,†,‡ Department of Materials Science and Engineering and Department of Electrical Engineering, The PennsylVania State UniVersity, UniVersity Park, PennsylVania 16802 ReceiVed: May 21, 2007; In Final Form: June 13, 2007
Described is the synthesis of TiO2 nanotube array thin films by anodization of Ti foil in an aqueous HCl electrolyte. This process represents an alternative electrolyte that can be utilized instead of fluoride-containing electrolytes. Nanotube arrays up to 300 nm in length, 15 nm inner pore diameter, and 10 nm wall thickness were obtained by using a 3 M HCl aqueous electrolyte for anodization potentials between 10 and 13 V. An anodization voltage of up to 20 V could be used if it was increased stepwise with a resulting nanotube length of approximately 600 nm; however, the resulting nanotubes are not as well-ordered as those fabricated by a constant anodization voltage. The addition of a low concentration of H3PO4 as a buffering medium in the concentration range 0.01-0.1 M expands the anodization voltage up to 14 V but the architectures formed more closely resemble rods than tubes.
1. Introduction Anodization of metals is a commonly developed surface treatment process used in a variety of applications.1-3 In 2001 Gong and co-workers4 reported the fabrication of vertically oriented, highly ordered TiO2 nanotube arrays up to approximately 500 nm length by potentiostatic anodization of titanium in an HF aqueous electrolyte. Since then the material architecture has been found to possess outstanding charge transport properties enabling a variety of advanced applications, including their use in sensors,5-7 self-cleaning photocatalytic surfaces and devices,8-9 dye sensitized solar cells,10-14 and in the hydrogen generation by water photoelectrolysis.3,14-18 The utility of the material architecture motivated subsequent studies to further refine and extend the synthesis techniques. Cai and co-workers,19 and subsequently Macak and co-workers,20 adjusted the pH of both KF and NaF aqueous electrolytes to reduce the chemical dissolution of the oxide, increasing the nanotube length to over 6 µm with a growth rate of approximately 0.25 µm/h. Recent reports by Grimes and coworkers describe the fabrication of TiO2 nanotube arrays up to 720 µm in length,14,17,21,22 with growth rates of up to 15 µm, using a variety of polar organic electrolytes that include ethylene glycol (EG), formamide (FA), N-methylformamide (NMF), and dimethyl sulfoxide (DMSO) in combination with HF, KF, NaF, NH4F, Bu4NF, or BnMe3NF to provide fluoride ions. The formation of nanotube arrays appears to be the result of three simultaneously occurring processes: (1) field assisted oxidation of Ti metal to form titanium dioxide, (2) field assisted dissolution of Ti metal ions in the electrolyte, and (3) chemical dissolution of Ti and TiO2 due to etching.23 The same three processes govern the anodic formation of several other selforganized nanoporous metal oxides, including alumina,24 hafnia,25 niobia,26 iron oxide,27 and tungsten oxide.28 Materials with aligned porosity in the submicron regime are of great interest for application in organic electronics, microf* Address correspondence to this author. E-mail:
[email protected]. † Department of Materials Science and Engineering. ‡ Department of Electrical Engineering.
luidics, molecular filtration, drug delivery, and tissue engineering.29,30 The nanotube array architecture allows for the precise design and control of the geometrical features, allowing one to achieve a material with specific light absorption and propagation characteristics,31,32 with the aligned porosity, crystallinity, and oriented nature of the nanotube arrays making them attractive electron percolation pathways for vectorial charge transfer between interfaces.10,12,33,34 In recent work Richter and co-workers35 describe the use of NH4Cl (0.15 to 0.4 M) in combination with either oxalic, formic, or sulfuric acid for fabrication of nondiscrete, thin-wall (∼4 nm) nanotube array bundles ranging from 5 to 50 µm in length. The as-anodized tube-bundles are amorphous, containing approximately 20 atom % carbon due to the organic acid electrolyte. Herein, we report the formation of TiO2 nanotube array films, quite similar to the geometry reported by Gong et al.,4 by potentiostatic anodization of titanium using an HCl aqueous electrolyte. We believe our discovery is significant in that it allows one to work with an arguably safer electrolyte, and presents a previously unexplored synthesis avenue for controlling the architectural features of the nanotube arrays. 2. Experimental Section Pure titanium foil (0.25 mm thick) was purchased from Sigma Aldrich. Prior to anodization samples were cleaned with acetone followed by a deionized (D.I.) water rinse. The anodization was performed with use of a two-electrode cell with titanium foil as the working electrode and platinum foil as the counter electrode, under constant applied voltage at room temperature (approximately 22 °C). Ti-foil sample sizes were 1 cm × 2 cm, with 1 cm × 1.5 cm immersed in the electrolyte. Electrolyte HCl concentrations ranging from 0.15 to 3.0 M with and without low concentrations of H3PO4 were studied. The time-dependent anodization currents were recorded by using a computer controlled Keithley 2000 multimeter. The as-anodized samples were ultrasonically cleaned in D.I. water for 30 s to remove any surface debris. The morphology of the anodized samples was examined by using a JEOL JSM-6300 field emission
10.1021/jp073924i CCC: $37.00 © 2007 American Chemical Society Published on Web 08/23/2007
Formation of Vertically Oriented TiO2 Nanotube Arrays
Figure 1. (a) Illustrative FESEM image of a titania surface achieved by using an electrolyte of 0.5 M HCl at 10 V. (b) The corresponding anodization current-time response of the sample.
J. Phys. Chem. C, Vol. 111, No. 35, 2007 13029
Figure 2. (a) Illustrative FESEM top-surface image of a nanoporous sample, obtained by anodization of Ti foil at 9 V in 3 M HCl. (b) The corresponding anodization current-time response of the sample.
scanning electron microscope (FESEM). The crystalline phases were detected and identified by a glancing angle X-ray diffractometer (GAXRD) on a Philips X’pert MRD PRO X-ray diffractometer (Almelo, The Netherlands). 3. Results and Discussion 3.1. Anodization in Pure HCl Electrolyte. With HCl concentrations below 3 M we were unable to achieve either porous or nanotubular titania surfaces over the voltage range 5 to 60 V. Figure 1a shows an illustrative FESEM image of a surface achieved by using an electrolyte of 0.5 M HCl at 10 V. The corresponding anodization current-time response of the sample, Figure 1b, shows a characteristic diminishment with increasing time as the oxide thickness steadily increases. This is in agreement with Jaszay and co-workers,36 who reported that 2 M HCl was not sufficient to initiate pitting corrosion on titanium. Figure 2a is an illustrative FESEM top-surface image of a nanoporous sample, obtained by anodization of Ti foil at 9 V in 3 M HCl. Inspection shows characteristic rings on the surface indicating that a nanotubular morphology is near. Figure 2b shows the corresponding anodization current-time response of the sample; notable are unusual spike-like oscillations in the current amplitude similar to those recorded for meta-stable pitting.37 Nanotube arrays have been achieved in 3 M HCl between 10 and 13 V; at higher voltages the sample simply corrodes. For example, well-developed nanotube arrays with thicknesses up to 300 nm are observed upon anodizing titanium at 13 V in
Figure 3. Well-developed nanotube arrays are observed upon anodizing titanium at 13 V in 3 M HCl for 10 min. A surface image is shown in the outer ring of the figure illustrating the uniform wide area film obtained; the inner image shows the individual tubes freed upon mechanical fracture of the sample.
3 M HCl for only 10 min, see Figures 3 and 4. The corresponding current-time behavior, see Figure 5, is essentially identical with that seen when nanotube arrays are achieved via anodization of Ti in aqueous HF solutions.38,39 Nanotube array films could be achieved in as little as 5 min, while anodization periods greater than approximately 20 min result in sample corrosion. We were able to increase the anodization voltage up to 20 V by stepwise increments, from 5 to 20 V at a rate of 1 V/min; however, the resulting nanotubes, up to ∼600 nm in
13030 J. Phys. Chem. C, Vol. 111, No. 35, 2007
Allam and Grimes
Figure 6. Illustrative FESEM top-surface image of nanorods/tubes obtained by anodization of Ti foil at 14 V, in 3 M HCl + 0.1 M H3PO4.
The low free energy of formation of TiO240 (-820 kJ/mol) accounts for its stability in aqueous media. The above reaction involves more elementary steps:
Figure 4. Illustrative cross-sectional images of a TiO2 nanotube array sample fabricated by anodizing titanium at 13 V in 3 M HCl for 10 min.
Ti + H2O ) TiO2+ + 2H+ + 4e-
(2)
TiO2+ + H2O ) TiO2 + 2H+
(3)
This oxide layer leads to a dramatic decrease in the recorded current due to its poor electrical conductivity. Note that the current decreased drastically (Figure 5) from an initial value of about 500 mA to about 50 mA within a few seconds. After that, TiO2 starts to dissolve forming pores, leading to the observed slight increase in current with time. This can be explained on the basis of the high field model (HFM)41 and its modified form;42 under sufficient applied voltage magnitude the electric field will be strong enough to migrate the titanium ions leaving behind some voids in the interpore areas43 which in turn will separate one pore from another leading to the formation of discrete tubes oriented vertically to the substrate. According to the HFM,41,42 the current under high field conditions during the formation of the oxide layer takes the simple form
i ) Re βF
(4)
where i represents the current, R is the jump probability of a cation interstitial and is given by eq 5, β is calculated for Ti by eq 6, and F is the electric field strength (V/cm):
R) Figure 5. The current-time behavior seen during anodization of Ti foil sample at 13 V, 3 M HCl; the behavior is essentially identical with that seen when nanotube arrays are achieved via anodization of Ti in aqueous HF solutions.38,39
length, are not nearly as well-organized as those fabricated by using a constant anodization voltage. Initially, a compact layer of TiO2 is formed through hydrolysis of titanium, i.e.,
Ti(s) + 2H2O(l) ) TiO2(s) + 4H+ + 4e-
(1)
0.24ekT -φo /kT e a2φoτ
(5)
4ae 8kT
(6)
β)
In the above equations e is the electron charge, k is Boltzmann’s constant, T is the absolute temperature, a is onehalf of the jump distance, φo is the minimum potential barrier, and τ is the time of vibration. Note here that the charge associated with formation of a titanium cation is 4e. The electric field linearly decreases with thickness, while from eq 4 we see that current will decay exponentially with film thickness. 3.2. Effect of the Addition of H3PO4. In an approach analogous to that reported by Cai and co-workers,19 we tried
Formation of Vertically Oriented TiO2 Nanotube Arrays
J. Phys. Chem. C, Vol. 111, No. 35, 2007 13031 We believe the narrow processing window is attributed to the complexity of the reaction of Ti in HCl. It was shown by Frayert and co-workers35,45,46 that the reaction of Ti in concentrated HCl includes six rate-determining steps with three possible reaction pathways. Also, hydrogen ions resulting from the dissociation of HCl could be considered to decrease the surface energy and thus increase the film instability.44 It was shown that the activity of H+ ions is related to the activity of both HCl and Cl- ions via36
1 1 1 ) + δaHCl δaH+ δaCl-
(7)
The formation of TiO2 in HCl was shown to proceed via the following steps:45 Figure 7. The anodization current-time response upon anodizing a Ti foil sample at 15 V in 3 M HCl + 0.07 M H3PO4.
Ti + H2O ) Ti(OH) + H+ + e-
(8)
Ti(OH) + H2O ) Ti(OH)2 + H+ + e-
(9)
Ti(OH)2 + H+ + Cl- ) [Ti(OH)Cl]+ + H2O + e[Ti(OH)Cl]+ + OH- ) TiO2 + H2O
(10) (11)
while reaction of TiO2 with HCl is supposed to proceed in a similar way to that in HF:
TiO2 + 4H+ + 6Cl- ) TiCl62- + 2H2O
(12)
A possible stepwise mechanism can be represented as:
Figure 8. The 2° glancing angle X-ray diffraction (GAXRD) pattern of a 500 °C annealed sample (3 h in oxygen with a ramp rate of 2 deg per min). Since the nanotube arrays are so thin the signature of the underlying Ti metal is readily apparent. The resulting nanotubes are anatase, separated from the underlying Ti metal by a rutile barrier layer.
increasing the processing window for obtaining TiO2 nanotube arrays in the HCl electrolyte by addition of H3PO4 as a buffering medium in the concentration range 0.01-0.1 M by which we found that we can only expand the anodization voltage up to 14 V with the 3 M HCl requirement remaining in place. Samples corroded at anodization voltages above 14 V. Samples anodized for 20 to 25 min at 14 V in 3 M HCl containing increasing amounts of H3PO4, 0.01-0.1 M, showed a corresponding change in the morphology from tubes to rods. Figure 6 shows an FESEM micrograph of a Ti sample anodized for 20 min at 14 V in 3 M HCl containing 0.1 M H3PO4. Figure 7 depicts the anodization current-time response upon anodizing the sample at 15 V in 3 M HCl + 0.07 M H3PO4. The current decayed for an initial period of 50 s reaching a steady-state value, resulting in formation of barrier layer oxide, up to 450 s beyond which an increase in current was seen. Similar behavior is reported by Raja and co-workers44 for Ti anodized in 0.5 M H3PO4 with incremental addition of HF to the anodization bath; with longer anodizing times perturbation of the steady state occurs with the current starting to increase. This behavior may be attributed to localized breakdown of the barrier layer and possibly nucleation of secondary oxide particles.44 Raja and co-workers44 related the instability of the barrier layer formed during anodization of Ti to two competing forces, namely, [1] surface energy (a stabilizing force) and [2] an increase in strain energy due to electrostriction, electrostatic, and re-crystallization stresses (a destabilizing force).
TiO2 + H+ + Cl- ) TiO(OH)Cl
(13)
TiO(OH)Cl + 3H+ + Cl- ) TiCl22+ + 2H2O
(14)
TiCl22+ + 4Cl- ) TiCl62-
(15)
This is supported by the high stability of TiCl62- where its free energy of formation is in the range -5.12 to -10.90 kcal/ mol depending on the type of cation.47 Like fluoride electrolyte obtained nanotube array films, the as-anodized HCl fabricated nanotube arrays are amorphous, crystallized by a high-temperature anneal.48 Figure 8 shows the XRD of a 500 °C annealed sample, 3 h in oxygen with ramp rates of 2 deg per min. Since the nanotube arrays are so thin the signature of the underlying Ti metal is readily apparent. The tubes themselves are anatase, with a rutile barrier layer separating the tubes from the underlying metal foil. 4. Conclusions We report on the formation of vertically oriented, TiO2 nanotube array films via potentiostatic anodization in an aqueous HCl electrolyte. Our study shows that the formation of TiO2 nanotubes depends on the applied potential as well as the electrolyte concentration. Arrays of discrete, vertically oriented TiO2 nanotubes with thicknesses up to 300 nm, 15 nm inner pore diameter, and 10 nm wall thickness were formed between 10 and 13 V in 3 M HCl, with anodization durations ranging from 5 to 20 min. We were unable to achieve nanotube arrays at lower or higher HCl concentrations, over a voltage span of 5 to 60 V. With the use of a stepwise increase in anodization voltage we were able to increase the anodization voltage up to 20 V and the nanotube length up to 600 nm; however, the resulting nanotubes are not well-organized. The addition of a
13032 J. Phys. Chem. C, Vol. 111, No. 35, 2007 low concentration of H3PO4 as a buffering medium in the concentration range 0.01-0.1 M enables an increase in the anodization voltage up to 14 V; however, the resulting architecture is closer to that of rods than tubes. Considering the great utility of the vertically oriented, highly ordered nanotube arrays5-15 the subject calls for more systematic studies at various concentrations of HCl in the presence of other additives at different anodization potentials to gain more insight into the prevailing mechanisms. Acknowledgment. Support of this work by the Department of Energy under grant DE-FG02-06ER15772 is gratefully acknowledged. N.K.A. is supported by a postgraduate fellowship from the Ford Foundation (IFP program). The authors are grateful to the Reviewers for their helpful suggestions and comments. References and Notes (1) Zwilling, V.; Darque-Ceretti, E.; Boutry-Forveille, A.; David, D.; Perrin, M. Y.; Aucouturier, M. Surf. Interface Anal. 1999, 27, 629-637. (2) La Flamme, K.; Mor, G.; Gong, D.; La Tempa, T.; Fusaro, V. A.; Grimes, C. A.; Desai, T. Diabetes Technol. Ther. J. 2005, 7, 684-694. (3) Mohapatra, S.; Misra, M.; Mahajan, V. K.; Raja, K. S. J. Phys. Chem. C 2007, 111, 8677-8685. (4) Gong, D.; Grimes, C. A.; Varghese, O. K.; Hu, W. C.; Singh, R. S.; Chen, Z.; Dickey, E. C. J. Mater. Res. 2001, 16, 3331-3334. (5) Varghese, O. K.; Yang, X.; Kendig, S.; Paulose, M.; Zeng, K.; Palmer, C.; Ong, K. G.; Grimes, C. A. Sens. Lett. 2006, 4, 120-128. (6) Yoriya, S.; Prakasam, H. E.; Varghese, O. K.; Shankar, K.; Paulose, M.; Mor, G. K.; Latempa, T. J.; Grimes, C. A. Sens. Lett. 2006, 4, 334339. (7) Grimes, C. A.; Ong, K. G.; Varghese, O. K.; Yang, X.; Mor, G.; Paulose, M.; Dickey, E. C.; Ruan, C.; Pishko, M. V.; Kendig, J. W.; Mason, A. J. Sensors 2003, 3, 69-82. (8) Mor, G. K.; Carvalho, M. A.; Varghese, O. K.; Paulose, M.; Grimes, C. A. J. Mater. Res. 2004, 19, 628-634. (9) Carneiro, J. O.; Teixeira, V.; Portinha, A.; Magalha˜es, A.; Coutinho, P.; Tavares, C. J.; Newton, R. Mater. Sci. Eng. B 2007, 138, 144-150. (10) Mor, G. K.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Nano Lett. 2006, 6, 215-218. (11) Mor, G. K.; Varghese, O. K.; Paulose, M.; Shankar, K.; Grimes, C. A. Sol. Energy Mater. Sol. Cells 2006, 90, 2011-2075. (12) Zhu, K.; Neale, N. R.; Miedaner, A.; Frank, A. J. Nano Lett. 2007, 7, 69-74. (13) Jiu, J.; Isoda, S.; Adachi, M.; Wang, F. J. Photochem. Photobiol., A 2007, 189, 314-321. (14) Shankar, K.; Mor, G. K.; Prakasam, H. E.; Yoriya, S.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Nanotechnology 2007, 18, no. 065707. (15) Mor, G. K.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Nano Lett. 2005, 5, 191-195. (16) Radecka, M. Thin Solid Films 2004, 451, 98-104. (17) 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.
Allam and Grimes (18) Mor, G. K.; Prakasam, H. E.; Varghese, O. K.; Shankar, K.; Grimes, C. A. Nano Lett. 2007, 7, 2356-2364. (19) Cai, Q. Y.; Paulose, M.; Varghese, O. K.; Grimes, C. A. J. Mater. Res. 2005, 20, 230-236. (20) Macak, J. M.; Tsuchiya, H.; Schmuki, P. Angew. Chem., Int. Ed. 2005, 44, 2100-2104. (21) Shankar, K.; Mor, G. K.; Fitzgerald, A.; Grimes, C. A. J. Phys. Chem. C 2007, 111, 21-26. (22) Prakasam, H. E.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. J. Phys. Chem. C 2007, 111, 7235-7241. (23) Mor, G. K.; Varghese, O. K.; Paulose, M.; Mukherjee, N.; Grimes, C. A. J. Mater. Res. 2003, 18, 2588-2593. (24) Parkhutik, V. P.; Shershulsky, V. I. J. Phys. D 1992, 25, 12581263. (25) Shobha, T.; Sarma, C. S. N.; Sastry, K. S.; Anjaneyulu, C. Bull. Electrochem. 2001, 17, 519-523. (26) Lu, Q.; Hashimoto, T.; Skeldon, P.; Thompson, G. E.; Habazaki, H.; Shimizu, K. Electrochem. Solid State Lett. 2005, 8, B17-B19. (27) Prakasam, H. E.; Varghese, O. K.; Paulose, M.; Mor, G. K.; Grimes, C. A. Nanotechnology 2006, 17, 4285-4291. (28) Mukherjee, N.; Paulose, M.; Varghese, O. K.; Mor, G. K.; Grimes, C. A. J. Mater. Res. 2003, 18, 2296-2299. (29) Popat, K. C.; Leary-Swan, E. E.; Desai, T. A. Biomaterials 2005, 26, 1969-1975. (30) Grimes, C. A. J. Mater. Chem. 2007, 17, 1451-1457. (31) Ong, K. G.; Varghese, O. K.; Mor, G. K.; Grimes, C. A. J. Nanosci. Nanotechnol. 2005, 5, 1801-1808. (32) Ong, K. G.; Varghese, O. K.; Mor, G. K.; Shankar, K.; Grimes, C. A. Sol. Energy Mater. Sol. Cells 2007, 91, 250-257. (33) Tenne, R.; Rao, C. N. R. Philos. Trans. R. Soc. London, Ser. A 2004, 362, 2099-2102. (34) Frank, A. J.; Kopidakis, N.; van de Lagemaat, J. Coord. Chem. ReV. 2004, 248, 1165-1172. (35) Richter, C.; Wu, Z.; Panaitescu, E.; Willey, R. J.; Menon, L. AdV. Mater. 2007, 19, 946-950. (36) Jaszay, T.; Caprani, A.; Priem, F.; Frayret, J. P. Electrochim. Acta 1988, 33, 1093-1100. (37) See for example: Moayed, M. H.; Newman, R. C. Corros. Sci. 2006, 48, 1004-1018. (38) Mor, G. K.; Varghese, O. K.; Paulose, M.; Grimes, C. A. AdV. Funct. Mater. 2005, 15, 1291-1296. (39) Mor, G. K.; Varghese, O. K.; Paulose, M.; Ong, K. G.; Grimes, C. A. Thin Solid Films 2006, 496, 42-48. (40) Bard, A. J.; Parsons, R.; Jordan, J. Standard Potential in Aqueous Solutions, IUPAC; Marcel Dekker Inc.: New York, 1985. (41) Verwey, J. W. Physica 1935, 2, 1059-1062. (42) Cabrera, N.; Mott, N. F. Rep. Prog. Phys. 1949, 12, 163. (43) Macdonald, D. D. J. Electrochem. Soc. 1993, 140, L27-L30. (44) Raja, K. S.; Misra, M.; Paramguru, K. Electrochim. Acta 2005, 51, 154-165. (45) Frayret, J. P.; Caprani, A. Electrochim. Acta 1982, 27, 391-399. (46) Caprani, A.; Frayret, J. P.; Pointeau, R.; Marya, S. K.; Le Maitre, F. Electrochim. Acta 1983, 28, 141-147. (47) Chassaing, E.; Basile, F.; Lorthioir, G. J. Less-Common Met. 1979, 68, 153-158. (48) Varghese, O. K.; Gong, D.; Paulose, M.; Grimes, C. A.; Dickey, E. C. J. Mater. Res. 2003, 18, 156-165.
