NANO LETTERS
Vertically Aligned Single Crystal TiO2 Nanowire Arrays Grown Directly on Transparent Conducting Oxide Coated Glass: Synthesis Details and Applications
2008 Vol. 8, No. 11 3781-3786
Xinjian Feng, Karthik Shankar, Oomman K. Varghese, Maggie Paulose, Thomas J. Latempa, and Craig A. Grimes* Department of Electrical Engineering and The Materials Research Institute, The PennsylVania State UniVersity, UniVersity Park, PennsylVania 16802 Received July 15, 2008; Revised Manuscript Received September 24, 2008
ABSTRACT Single-crystal one-dimensional (1D) semiconductor architectures are important in materials-based applications requiring a large surface area, morphological control, and superior charge transport. Titania has widespread utility in applications including photocatalysis, photochromism, photovoltaics, and gas sensors. While considerable efforts have focused on the preparation of 1D TiO2, no methods have been available to grow crystalline nanowire arrays directly onto transparent conducting oxide (TCO) substrates, greatly limiting the performance of TiO2 photoelectrochemical devices. Herein, we present a straightforward low temperature method to prepare single crystal rutile TiO2 nanowire arrays up to 5 µm long on TCO glass via a non-polar solvent/hydrophilic substrate interfacial reaction under mild hydrothermal conditions. The as-prepared densely packed nanowires grow vertically oriented from the TCO glass substrate along the (110) crystal plane with a preferred (001) orientation. In a dye sensitized solar cell, N719 dye, using TiO2 nanowire arrays 2-3 µm long we achieve an AM 1.5 photoconversion efficiency of 5.02%.
Largely motivated by the initial work of Fujishima and Honda using TiO2 for hydrogen generation by water photoelectrolysis1 and the development of the dye-sensitized or Gra¨tzel solar cell (DSSC),2 considerable efforts have focused on the development of photoelectrochemical cells using a variety of TiO2 architectures.3-12 To date, the most common photoelectrochemical architecture is a (large surface area) several micron-thick film composed of nanocrystalline TiO2 nanoparticles on a TCO substrate. However, the electron diffusion coefficient of these nanoparticulate films is several orders of magnitude smaller than the value in bulk singlecrystal TiO2, presumably due to electron traps at the contacts between nanoparticles.13-15 Polycrystalline transparent TiO2 nanotube array films16 demonstrate lower recombination17 and have been successfully used in DSSCs;11,18 however, the films are relatively difficult to fabricate requiring Ti film deposition, an anodization step, then crystallization by thermal annealing, which commonly degrades the conductivity of the TCO layer. In ZnO-based DSSCs, the use of a photoelectrode constructed of 1D crystalline nanowire arrays, aligned perpen* Corresponding author. E-mail:
[email protected]. 10.1021/nl802096a CCC: $40.75 Published on Web 10/28/2008
2008 American Chemical Society
dicular to the charge collecting TCO substrate, has been found to improve charge collection efficiency by promoting faster charge transport and faster ion diffusion at the semiconductor-electrolyte interface.19,20 However, in contrast to TiO2, ZnO based DSSCs suffer from poor chemical stability and have shown limited photoconversion efficiencies, approximately 1%. A variety of synthesis techniques have been used to form oriented or disoriented TiO2 nanorods or wires on nontransparent and/or nonconductive substrates including surfactant assisted self-assembly,3,6,7,21 templated sol-gel methods,22 high temperature chemical vapor deposition,23 and high temperature vapor-liquid-solid growth.24 However, prior to this report, synthesis routes for achieving aligned, densely packed single crystal nanowire arrays on TCO coated glass have not been known. Our synthetic process is performed at relatively low temperatures, below 180 °C, with the conductivity of the TCO glass substrate unaffected. Through a nonpolar solvent/hydrophilic solid substrate interfacial reaction under hydrothermal conditions, we achieve single crystal rutile, vertically oriented TiO2 nanowires on a TCO substrate, in our case fluorine-doped tin oxide (FTO) coated glass.
TiO2 nanowire arrays are synthesized on FTO coated glass (TEC-8, 8 Ω per square) initially cleaned by sonication in acetone, 2-propanol, and methanol, subsequently rinsed with deionized (DI) water, and finally dried in a nitrogen stream. The FTO coated glass substrates are placed within a sealed Teflon reactor (23 mL), containing 10 mL of toluene, 1 mL of tetrabutyl titanate, 1 mL of titanium tetrachloride (1 M in toluene), and 1 mL of hydrochloric acid (37 wt %). Nanowire array synthesis is achieved using toluene as the nonpolar solvent and toluene soluble tetrabutyl titanate and titanium tetrachloride as the precursor. Hydrochloric acid is used to avoid hydrolysis of the precursor by ensuring an acidic solution pH. The small amount of polar water from the HCl solution is immiscible with the nonpolar toluene. With an increase in temperature, to minimize system energy water will diffuse away from the high-energy water/toluene interface aggregating on the hydrophilic FTO glass surface. Simultaneously, Ti4+ precursors will hydrolyze with water at the water/FTO interface, resulting in the formation of a crystal nucleus on the substrate. After the formation of the first nanocrystalline layer, a new interface between the hydrophilic TiO2 and toluene is formed, with continuous hydrolysis and subsequent growth-crystallization. A reaction temperature of 180 °C is used, with reaction times lasting from 30 min to 48 h. In our initial synthesis conditions, we find that nanowire growth slows with time, for example, a 2 h reaction results in 2.1 µm nanowires, a 4 h reaction leads to 3.2 µm nanowires, 8 h to 3.8 µm, and 22 h to 4 µm. No increase in nanowire length was achieved for reaction times beyond 22 h. After the reaction period, the nanowire samples are removed, washed with ethanol, then dried in air. We note that in solar cell application, the nanowire array samples grown directly on the FTO coated glass sometimes demonstrated electrical shorting with the redox electrolyte coming into direct contact with the FTO layer. Hence, for solar cell use, before growing the nanowire arrays a compact TiO2 layer approximately 20 nm thick was commonly deposited onto the well-cleaned FTO glass substrate by immersion in a 0.2 M TiCl4 solution for 12 h, then heated in air at 500 °C for 0.5 h.31 Such a layer does not appear to affect nanowire growth while greatly enhancing sample integrity for solar cell use. Figures 1a and b are field-emission scanning electron microscope (FESEM; JEOL JSM-6300, Japan) top-surface images of a typical as-synthesized nanowire array sample at low and high magnification, respectively, showing a highly uniform and densely packed array of nanowires with flat tetragonal crystallographic planes. Figure 1c is a crosssectional view of the same sample, showing that the nanowires grow almost perpendicularly from the substrate. On the basis of length, we grouped the nanowires arrays into three families, namely, nanowires of length 2-3 µm (Group A), 3-4 µm (Group B), and 4-5 µm (Group C). The lateral dimension, or width, of the square nanowire lies in the range 10-35 nm. The nanowire width is a parameter that is a sensitive function of the duration of the hydrothermal treatment with larger duration increasing the nanowire width. From the X-ray diffraction pattern (XRD; Scintag Inc., CA. 3782
Figure 1. FESEM images of vertically oriented self-organized TiO2 nanowire array grown on FTO coated glass at 180 °C for 24 h: (a,b) top-view images; (c) cross-sectional FESEM image of the same array, mechanically fractured.
USA) (see Figure 2a), the nanowires can be classified as tetragonal rutile (JCPDS file no. 21-1276); the enhanced (002) peak indicates the nanowire is well crystallized and grows perpendicular to the substrate. Transmission electron microscope (TEM) and high resolution transmission electron microscope (HR-TEM; JEOL 2010F, Japan) images, and selected-area electron diffraction patterns, Figures 2b-f, confirm that the nanowires are single crystalline. The HRTEM image shows a (110) interplane distance of 0.325 nm; the nanowires grow along the (110) crystal plane with a preferred (001) orientation. The anisotropic growth of our 1D nanocrystals can be understood from shape-control chemistry.6,25,26 With the formation of a TiO2 nanocrystal, a Cl- ion can selectively Nano Lett., Vol. 8, No. 11, 2008
Figure 2. Vertically oriented self-organized TiO2 nanowires on FTO coated glass: (a) X-ray diffraction pattern; (b), (c), and (d) TEM image of the nanowires; (e) HRTEM image; (f) selected-area electron diffraction pattern.
adsorb onto the (110) crystal plane3 suppressing further growth of this plane, resulting in anisotropic growth and hence TiO2 nanowire array formation. We note the precursor immediately hydrolyses when the hydrochloric acid is replaced with DI water (1 mL), in which case only TiO2 nanoparticles are obtained with no observed crystal faceting. Hydrochloric acid is believed to play a dual role here: (1) tailor the pH value of the reaction solution and retard hydrolysis of the precursor in the presence of water at low Nano Lett., Vol. 8, No. 11, 2008
temperatures; (2) reduce the surface energy of the crystal plane side wall, promoting anisotropic growth in the [001] direction.27 The conductivity of the TCO layer underneath the TiO2 nanowire array appears unaffected by the nanowire growth due to the mild synthesis conditions, facilitating photoelectrochemical use. Photoelectrochemical characterization of the samples was performed using a three-electrode configuration (Keithley 2400 source-meter and a CHI 600B potentiostat), 3783
Figure 3. Photocurrent density and photoconversion efficiency of a TiO2 nanowire array electrode as a function of measured potential [vs Ag/AgCl] in 1 M KOH under AM 1.5 illumination. The inset shows measured IPCE.
with TiO2 nanowires on FTO-glass as the working photoelectrode, saturated Ag/AgCl reference electrode, and a platinum foil counter electrode. Figure 3 shows the photocurrent density versus potential of a 2.4 µm long TiO2 nanowire array electrode, nominal wire width 20 nm, sample size 0.5 cm2, measured in 1 M KOH electrolyte under 1.5 AM solar illumination (100 mW/cm2, Spectra Physics Simulator, USA). The potential was scanned at a rate of 20 mV/s. The inset of Figure 3 shows the photon-to-electron conversion efficiency (IPCE) as a function of wavelength for the TiO2 nanowire photoelectrode without bias. The IPCE threshold is at 415 nm and the values reach a maximum of approximately 90% at 380 nm, indicating that the nanowire arrays have broad and high light absorption and low charge recombination. The light-to-chemical energy conversion efficiency of the nanowires was determined in a two electrode configuration with TiO2 nanowires on FTO glass as the working photoelectrode and platinum foil as a counter electrode;28 a photoconversion efficiency of 0.75% is obtained. Holes avoid recombination with electrons by oxidizing solution ions at the surface of the nanowires. The maximum distance holes in the rutile nanowires have to travel to reach the surface (the retrieval length) is equal to onehalf the lateral dimension of the nanowire, which equates to a relatively short distance of ∼10 nm in our samples. The electron mobility of single crystalline rutile is ∼1 cm2 V-1 s-1, over 2 orders of magnitude larger than those for nanoparticulate TiO2 films.29 Hence, the combination of a short hole retrieval length and fast electron transport may account for the excellent charge separation. Unlike nanoparticle-based electrodes that need a positive bias between 0.5 to 1 V (vs reference electrode) to completely separate the light generated electron-hole pairs,28,30 the photocurrent of the TiO2 nanowire array-based electrode increases sharply, reaching saturation at approximately -0.25 V, indicative of both low series resistance and facile separation of the photogenerated charges. For use in DSSCs, nanowire array samples were coated with dye by immersion overnight in a 0.5 mM solution of 3784
Figure 4. Photocurrent density (red), dark current density (green), and power density (blue) of 2.0 µm long TiO2 nanowire array based dye sensitized solar cells under AM 1.5 illumination (100 mW/ cm2): (a) unmodified nanowire array grown directly on TCO substrate; (b) a TiCl4 treatment is used to coat the TCO substrate prior to nanowire array growth and a NbCl5 treatment used on the nanowires prior to device assembly.32,33
commercially available N719 dye (Solaronix Inc., Switzerland). A liquid junction solar cell was prepared by infiltrating the dye coated TiO2 electrode with commercially available redox electrolyte MPN-100 (Solaronix, Inc., Switzerland) containing 100 mM of tri-iodide in methoxypropionitrile. A conductive glass slide sputter-coated with 100 nm of Pt was used as the counter-electrode. Electrode spacing between the nanowire and counter-electrodes was ensured by the use of a 25 µm thick SX-1170 spacer (Solaronix Inc., Switzerland). Photocurrent density and photovoltage of the dye sensitized solar cells were measured with active sample areas of 0.4-0.5 cm2 using AM-1.5 simulated sunlight produced by a 500 W Oriel Solar Simulator. Figure 4a shows the J-V characteristics of one of a typical sample from Group A (2-3 µm nanowire length) grown directly on the FTO coated substrate, under AM 1.5 illumination. An overall photoconversion efficiency of 4.35% is achieved, with an open circuit voltage (Voc) of 0.758 V, short circuit current density (Jsc) of 10.2 mA cm-2, and fill factor (FF) of 0.56. Table 1 includes statistical data related to the performance of solar cells comprising nanowire arrays of different lengths. By using the same batch of N-719 dye, the same redox electrolyte and similar active areas, we sought to isolate the Nano Lett., Vol. 8, No. 11, 2008
Table 1. Statistics of TiO2 Nanowire Array Dye Sensitized Solar Cells Showing the Mean (µ) and Standard Deviation (σ) of Solar Cell Parameters: Short Circuit Photocurrent Density (Jsc), Open Circuit Potential (Voc), Fill-factor (ff), and Efficiency (η) length of nanowires (µm)
typical nanowire width (nm)
# of devices
µJsc
σJsc
µVoc
σVoc
µff
σff
µn
σn
2-3(A) 3-4(B) 4-5
10-15 20-25 30-35
6 6 10
10.34 8.47 7.29
2.52 1.06 2.29
751 742 769
36 21 19
0.57 0.6 0.62
0.05 0.05 0.04
4.34 3.75 3.39
1.09 0.39 0.85
effect of nanowire length on device performance. Normally, the surface area and hence the dye adsorbed increases as a function of length because of which an increase in solar cell performance with nanowire length is expected. However, here we observe that our Group A nanowire array devices exhibit the best performance with an average efficiency of 4.83%, with decreasing performance seen for longer nanowires. This can be understood within the context of nanowire array formation. Longer nanowire arrays result when the duration of the hydrothermal growth is extended; the longer times cause an increase in the lateral dimension as well as the axial dimension of the nanowires. An increase in nanowire width decreases the packing density per unit area of the nanowire arrays, which results in lower effective internal surface area available for dye adsorption despite the higher length. We also note that it appears extended reaction times, corresponding to nanowire arrays longer than 4 µm, weaken the nanowire-FTO interface, with the resulting samples demonstrating poorer photoelectrochemical properties. Smaller nanowire widths are desirable for increased surface areas; currently, nanowire widths of ∼10 nm are achievable only for Group A samples. For nanowires longer than 4 µm (Group C), the basal regions of the nanowires were found to bunch together in about 30% of fabricated nanowires array samples. Because small changes in the width of 10-15 nm nanowires can have a large effect on the surface area of the nanowires arrays, the Group A nanowires exhibited the largest dispersion in performance. Because of the bunching effect at the base of the nanowires and weaker interfaces, the Group C nanowire arrays exhibit significant dispersions in their photocurrents and efficiencies, while Group B (3-4 µm) nanowires arrays provided the most consistent performance. We note that device photoconversion efficiency could be improved in the following manner. To improve the fill factor, after the TiO2 nanowire growth, we subsequently coated an overlayer of niobium oxide onto the nanowires, which has been shown to improve the efficiency by reducing recombination.32,33 To form the Nb2O5 coating, the prepared nanowire samples were dipped in a 5 mM NbCl5 dry ethanol solution, then heated in air at 500 °C for 0.5 h.33 Figure 4b shows the J-V characteristics of a typical Group A sample under AM 1.5 illumination. An overall photoconversion efficiency of 5.02% is achieved, with an open circuit voltage (Voc) of 0.744 V, short circuit current density (Jsc) of 10.84 mA cm-2, and fill factor (FF) of 0.62. We note that in spite of their superior charge transport characteristics, 20 µm long ZnO nanowire arrays exhibit relatively low efficiencies when used in dye sensitized solar cells due to poor dye adsorption.20 We attribute our much higher light-to-electricity conversion efficiency achieved Nano Lett., Vol. 8, No. 11, 2008
using TiO2 nanowire arrays to the ability of the dye to anchor on the (110) crystal plane, which is both stable and has a strong interaction with carboxylate groups of the N719 dye molecule.34 In contrast, the acidic carboxylate groups of the dye dissolve the outer layer of the ZnO surface forming a Zn2+-dye complex layer,35,36 which in turn weakens the interaction of the dye with the electrode surface. DSSC TiO2nanoparticulate electrodes exhibiting the highest performance are typically constructed from films at least 10 µm thick.37 Yet solar cells comprising 2 µm long rutile nanowires synthesized by us exhibit higher efficiencies than rutile nanoparticulate films thicker than 5 µm.38 Efficient vectorial charge transport, enabled by the single crystal nanowires, leads us to believe that significantly higher efficiencies can reasonably be expected through solution of the interface problem and nanowire-bunching presently seen in TiO2 nanowire arrays longer 4 µm, which would enable greater light absorption without a corresponding penalty in charge transfer capabilities. In summary, densely packed single crystal TiO2 nanowire arrays were successfully prepared on FTO coated glass substrates via a low temperature non-polar solvent/hydrophilic substrate interface hydrolysis reaction. The synthesis method is simple, of low cost, and highly reproducible. After a growth reaction, the organic and water phases can be easily separated and recycled for the next reaction, minimizing the environmental impact. The synthesis method appears general, one that could be readily extended to the synthesis of other metal oxide semiconductors in single-crystal nanowire array form, given a suitable precursor and interface reaction. Dye sensitized solar cells fabricated using the nanowire arrays demonstrate very encouraging photoconversion efficiencies, 5.02% for a 2-3 µm long nanowire array when using N719 dye. Finally, we note the relatively low synthesis temperatures needed for achieving the single crystal nanowire arrays are compatible with polymeric substrates; hence, there is exciting potential for realizing flexible devices using nanowire arrays. Acknowledgment. Support of this work by the Department of Energy, Basic Energy Sciences, under grant DEFG02-06ER15772 is gratefully acknowledged. Supporting Information Available: Description of the estimation of the roughness factor of nanowire arrays, UV-vis absorption, diffuse transmittance measured with N-719 coated nanowire array, FTO counter electrode and MPN-100 introduced between clamped electrodes, and measured IPCE of N-719 sensitized 2 µm nanowire array solar cell. This material is available free of charge via the Internet at http://pubs.acs.org. 3785
References (1) Fujishima, A.; Honda, K. Nature 1972, 238, 37. (2) O’regan, B.; Gra¨tzel, M. Nature 1991, 353, 737. (3) Adachi, M.; Murata, Y.; Takao, J.; Jiu, J. T.; Sakamoto, M.; Wang, F. M. J. Am. Chem. Soc. 2004, 126, 14943. (4) Hosono, E.; Fujihara, S.; Kakiuchi, K.; Imai, H. J. Am. Chem. Soc. 2004, 126, 7790. (5) Zhang, Q. H.; Gao, L. Langmuir 2003, 19, 967. (6) Jun, Y. W.; Casula, M. F.; Sim, J. H.; Kim, S. Y.; Cheon, J.; Alivisatos, A. P. J. Am. Chem. Soc. 2003, 125, 15981. (7) Cozzoli, P. D.; Kornowski, A.; Weller, H. J. Am. Chem. Soc. 2003, 125, 14539. (8) Tenne, R.; Rao, C. N. R. Philos. Trans. R. Soc. London, Ser. A 2004, 362, 2099. (9) Xiac, Y. N.; Yang, P. D.; Sun, Y. G.; Wu, Y. Y.; Mayers, B.; Gates, B.; Yin, Y. D.; Kim, F.; Yan, H. Q. AdV. Mater. 2003, 15, 353. (10) Shankar, K.; Bandara, J.; Paulose, M.; Wiletasch, H.; Varghese, O. K.; Mor, G. K.; Latempa, T. J.; Thelakkat, M.; Grimes, C. A. Nano Lett. 2008, 8, 1654. (11) Mor, G. K.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Nano Lett. 2005, 5, 191. (12) Gra¨tzel, M. J. Photochem. Photobiol., A 2004, 164, 3. (13) Solbrand, A.; Lindstro¨m, H.; Rensmo, H.; Hagfeldt, A.; Lindquist, S. E.; So¨dergren, S. J. Phys. Chem. B 1997, 101, 2514. (14) Fisher, A. C.; Peter, L. M.; Ponomarev, E. A.; Walker, A. B.; Wijayantha, K. G. U. J. Phys. Chem. B 2000, 104, 949. (15) van de Lagemaat, J.; Park, N. G.; Frank, A. J. J. Phys. Chem. B 2000, 104, 2044. (16) Mor, G. K.; Varghese, O. K.; Paulose, M.; Grimes, C. A. AdV. Funct. Mater. 2005, 15, 1291. (17) Zhu, K.; Neale, N. R.; Miedaner, A.; Frank, A. J. Nano Lett. 2007, 7, 69. (18) Mor, G. K.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Appl. Phys. Lett. 2007, 91, Art. No. 152111. (19) Baxter, J. B.; Aydil, E. S. Appl. Phys. Lett. 2005, 86, Art. No. 053114. (20) Law, M.; Greene, L. E.; Johnson, J. C.; Saykally, R.; Yang, P. D. Nat. Mater. 2005, 4, 455.
3786
(21) Wang, W.; Varghese, O. K.; Paulose, M.; Grimes, C. A.; Wang, Q.; Dickey, E. C. J. Mater. Res. 2004, 19, 417. (22) Lei, Y.; Zhang, L. D.; Meng, G. W.; Li, G. H.; Zhang, X. Y.; Liang, C. H.; Chen, W.; Wang, S. X. Appl. Phys. Lett. 2001, 78, 1125. (23) Lee, J. C.; Kim, T. G.; Choi, H. J.; Sung, Y. M. Cryst. Growth Des. 2007, 7, 2588. (24) Lee, J. C.; Park, K. S.; Kim, T. G.; Choi, H. J.; Sung, Y. M. Nanotechnology 2006, 17, 4317. (25) Peng, X. G.; Manna, L.; Yang, W. D.; Wickham, E. S.; Kadavanich, A.; Alivisatos, A. P. Nature 2000, 404, 59. (26) Barnard, A. S.; Curtiss, L. A. Nano Lett. 2005, 5, 1261. (27) Yang, H. G.; Sun, C. H.; Qiao, S. Z.; Zou, J.; Liu, G.; Smith, S. C.; Cheng, H. M.; Lu, G. Q. Nature 2008, 453, 638. (28) Grimes, C. A.; Varghese, O. K.; Ranjan, S. Light, Water, Hydrogen: The Solar Generation of Hydrogen by Water Photoelectrolysis;, Springer: New York, 2008; pp 485-516. (29) Hendry, E.; Koeberg, M.; O’Regan, B.; Bonn, M. Nano. Lett. 2006, 6, 755. (30) Oliva, F. Y.; Avalle, L. B.; Santos, E.; Camara, O. R. J. Photochem. Photobiol., A 2002, 146, 175. (31) Mor, G. K.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Nano. Lett. 2006, 6, 215. (32) Zaban, A.; Chen, S. G.; Chappel, S.; Gregg, B. A. Chem. Commun. 2000, 2231. (33) Chen, S. G.; Chappel, S.; Diamant, Y.; Zaban, A. Chem. Mater. 2000, 13, 4629. (34) Diebold, U. Surf. Sci. Rep. 2003, 48, 53. (35) Bauer, C.; Boschloo, G.; Mukhtar, E.; Hagfeldt, A. J. Phys. Chem. B 2001, 105, 5585. (36) Chou, T. P.; Zhang, Q. F.; Cao, G. Z. J. Phys. Chem. C 2007, 111, 18804. (37) Nazeeruddin, M. K.; Kay, A.; Rodicio, I.; Humphry-Baker, R.; Mu¨ller, E.; Lisaka, P.; Vlachopoulos, N.; Gra¨tzel, M. J. Am. Chem. Soc. 1993, 115, 6382. (38) Park, N. G.; van de Lagemaat, J.; Frank, A. J. J. Phys.Chem. B 2000, 104, 8989.
NL802096A
Nano Lett., Vol. 8, No. 11, 2008
