High-Density Vertically Aligned ZnO Rods with a Multistage Terrace Structure and Their Improved Solar Cell Efficiency Rong Zhang,† Sachin Kumar,‡ Shouzhong Zou,*,‡ and Lei L. Kerr*,† Department of Paper and Chemical Engineering and Department of Chemistry and Biochemistry, Miami UniVersity, Oxford, Ohio 45056
CRYSTAL GROWTH & DESIGN 2008 VOL. 8, NO. 2 381–383
ReceiVed August 30, 2007; ReVised Manuscript ReceiVed October 11, 2007
ABSTRACT: This work reports a solution-based approach to grow high-density vertically aligned ZnO rod arrays with a multistage terrace structure (HDVAMT) on Au nanoparticle covered glass substrates. The solar cell performance of HDVAMT is 10 times better than that of the disordered ZnO due to the much higher short circuit current (Jsc). This significant improvement of Jsc is attributed to two factors: the vertically aligned structure, which provides smoother electron transport channels, and the high density of the rods with the multistage terrace structure, which provides a higher surface area for the dye adsorption. A tentative growth mechanism is proposed. Solar energy is increasingly viewed as an unlimited, abundant, clean, and sustainable energy source. Current conventional inorganic materials such as GaAs, CdTe, CuInGaSe2, and silicon solar cells exhibit a cell efficiency of 10–25%.1 The major limitation of these materials is their high cost. Low cost materials are needed to make solar energy more economic. One low cost alternative is hybrid organic/inorganic material such as dye-sensitized solar cells (DSSCs). DSSC efficiency depends on the carrier transport. Ordered onedimensional structure such as rods or wires of metal oxides can greatly improve DSSCs efficiency by enhancing the electron transport.2,3 TiO2 nanoparticle DSSCs have shown the highest efficiency (11%) among all semiconductor anodes, but it is difficult to grow TiO2 anisotropically to obtain ordered structures. The band gap energy and the electron affinity of ZnO are nearly identical to those of TiO2, and the wurtize structure of typical ZnO favors the formation of anisotropic structures such as rods. Moreover, ZnO, which is an environmentally friendly material, has shown better electron transport compared with TiO2 films.4 These advantages make ZnO an ideal alternative for metal oxide-based DSSCs, and it has attracted much attention recently.5–8 The vertically ordered ZnO structure is crucially important in developing ZnO rod-based dye-sensitized solar cells9–12 because it provides smoother electron transport channels and a longer electron diffusion length. Consequently, a higher solar cell efficiency compared to the disordered structures can be obtained.10 In addition, organized arrays contain a higher density of rods, which can significantly increase the dye adsorption surface area and therefore improves the solar cell efficiency. Here we present a simple method to synthesize high-density vertically aligned ZnO rods with multistage terrace structure (HDVAMT) on Au nanoparticle array covered indium tin oxide (ITO) glass slides by chemical bath deposition (CBD). These aligned ZnO rods are perpendicular to the substrate, which provide the smooth electron transport channels. In addition, the highdensity rods with multistage terrace structure of HDVAMT ZnO film provide more surface area for dye loading than a simple hexagonal structured ZnO rod, and consequently a higher solar cell efficiency.10 To verify that the vertically aligned structure is the contributing factor to the improvement in the solar cell performance, disordered ZnO films were also prepared. Comparison of the ZnO film morphology on both ITO/Au and ITO substrates also provides insights on the film growth mechanism. The Au nanoparticles were fabricated following the approach developed by Spatz et al., using polystyrene-b-poly(2-vinylpy* Corresponding author. (L.L.K.) Tel: 513-529-0768. Fax: 513-529-0761. E-mail:
[email protected]. † Department of Paper and Chemical Engineering. ‡ Department of Chemistry and Biochemistry.
Figure 1. TEM image of the Au nanoparticles served as the nucleation sites.
ridine) (PS-b-P2VP) as the template.13 Briefly, 25 mg of PS-bP2VP was dissolved in 5 mL of dry toluene under vigorous mechanical stirring for at least 5 h. Toluene is a selective solvent for the PS block; therefore, the polymer forms inverse spherical micelles with the P2VP block as the core and the PS block as the corona. Hydrogen tetracholoroaurate trihydrate was used as the metal precursor for forming Au nanoparticles. The metal salt was loaded to the micelles by continuous stirring for at least 48 h. The molar ratio between HAuCl4 and the P2VP unit was 0.2. A monolayer of metal salt loaded micelles on the ITO substrate was formed by dip coating with an immersion rate of 30 mm min-1 and a pulling rate of 10 mm min-1. The metal salt loaded micelle monolayer covered substrate was subjected to argon plasma (100 W, 0.3 Torr) for 10 min. During plasma treatment the metal salt was reduced to form particles, and the polymer template was completely removed as revealed by X-ray photoelectron spectroscopy.13 It should be noted that during the plasma treatment the substrate temperature does not exceed 100 °C.13 The dip coating and subsequent plasma treatment were repeated two more times to form higher density Au nanoparticles. TEM image shows that this approach produces arrays of 3-6 nm Au particles (Figure 1), and the particles are well separated from each other. The aqueous solution for ZnO thin film deposition was prepared by mixing 50 mL of 0.4 mol/L zinc acetate dihydrate (Zn (CH3COO)2 · 2H2O) and 5 mL of 0.5 mol/L ethylenediaminetetraacetic acid (EDTA). Ammonia hydroxide (NH4OH, 5 mol/L) was added dropwise to the above solution to adjust the solution pH to 10, which yielded a clear solution. The substrates were suspended vertically in the solution for ZnO film growth. The reaction temperature was kept at 80 °C for 7 h, after which the glass substrates as well as the wall of the reactor were fully coated with white ZnO films. Finally, the glass substrates were withdrawn from
10.1021/cg700825v CCC: $40.75 2008 American Chemical Society Published on Web 12/20/2007
382 Crystal Growth & Design, Vol. 8, No. 2, 2008
Communications
Figure 2. SEM images of the HDVAMT and disordered ZnO films. Table 1. Dye Adsorption and I-V Characteristics of DSSCs with HDVAMT and Disordered ZnO Films Jsc (mA/cm2)a Voc (V)b fill factor efficiency (%) HDVAMT ZnO rods disordered ZnO rods a
8.0 0.6
0.42 0.52
0.27 0.28
0.9 0.09
Short-circuit current density (mA/cm2). b Open-circuit voltage (V).
Figure 3. XRD patterns of the HDVAMT and disordered ZnO films.
the solution, rinsed with deionized water, and dried at room temperature for about 5 h. The film morphologies were examined by a Zeiss Supra 35 VPFEG scanning electron microscope (SEM) with a 2 keV electron beam. SEM images of three different magnifications are shown for each ZnO film in Figure 2. As seen from the images, ZnO film grown on the ITO/Au substrate shows high density of rods that are perpendicular to the substrate. These rods present a terrace structure at the top. Each terrace-tower consists of a hexagonal center rod surrounded by several smaller and shorter rods. The rods with different lengths in a terrace-tower form the multistage structure. For the ZnO film grown on the ITO without the Au nanoparticles, randomly oriented rods are observed, and the film is not as compact as the HDVAMT film. However, an individual rod also exhibits the stage-like structure. The as-prepared ZnO thin films were characterized by X-ray diffraction (XRD) as shown in Figure 3. Several sharp diffraction peaks are observed, which are in agreement with the typical wurtzite structure ZnO diffraction pattern (hexagonal phase, space group P63mc, JCPDS No. 36-1451). The sharp diffraction peaks, especially the (002) peak shown in Figure 3, of the aligned ZnO rods indicate good crystallinity of these ZnO films, and the favorable growth direction of ZnO crystals is (002). The ZnO/ITO/glass was immersed into a dye solution for dye adsorption. The solar cell dye used in the measurement was
Figure 4. I-V curves for DSSC cells based on the HDVAMT and disordered ZnO rods.
RuL2(NCS)2:2TBA (L ) 2,2′-bipyridyl-4,4′-dicarboxylic acid, and TBA ) tetrabutylammonium, N719). Then the dye adsorbed ZnO electrode was assembled into DSSC with the Pt-coated counter electrode. A redox organic electrolyte which contains iodide ions (I-) and iodine in the form of triiodide (I3-I3) was injected between the electrodes. Solar cell efficiency of both ZnO films was obtained by the current–voltage measurement. All of the samples have an exposed area of 0.20 cm.2 Details about the solar cell efficiency measurement have been described earlier by Tan and Wu.2 The results are summarized in Table 1. The I-V curves are shown in Figure 4. Although the open circuit potential is lower for the HDVAMT, the higher short circuit current density makes its overall solar cell efficiency 10 times higher than that of the disordered ZnO film. This increasing efficiency verifies our hypothesis that high-density vertically aligned ZnO rods increase dye adsorption and have smoother electron transport. As discussed above, HDVAMT ZnO film has a much higher solar cell efficiency than the disordered ZnO rods film. Comparing the morphology of the two samples (Figure 2), the key factors responsible for the improved solar cell efficiency are the vertically aligned structure and the high-density rods with multistage terrace
Communications
Crystal Growth & Design, Vol. 8, No. 2, 2008 383
Figure 5. Schematic diagrams depicting the formation mechanism of ZnO rods on ITO substrate (a) with Au nanoparticles, and (b) without Au nanoparticles.
structure. We believe that the HDVAMT structure is controlled by the precoated Au nanoparticles on the ITO substrate. Although the exact growth mechanism is not clear at present, we offer a tentative explanation below. As shown in the TEM image (Figure 1), the Au nanoparticles distribute uniformly on the ITO substrate. These particles serve as the nucleation sites for the subsequent ZnO dot growth during the incipient stage (Figure 5a). Assisted by these dots, ZnO crystals begin to grow rapidly along the preferred (002) direction14,15 to form the aligned and dense ZnO rod arrays. The growth along other directions forms the typical hexagonal shape rods and brings each rod close to one another. In the combining stage, several nearby ZnO rods form one big hexagonal rod, and the tower structure is created. With the dense HDVAMT structure, the surface area for dye loading is large, and the vertically aligned rod arrays are beneficial to smoother electron transport. Both factors contribute to the improved solar cell efficiency. Without the uniform and dense Au nanoparticles as the nucleation sites (Figure 5b), ZnO dots spread randomly and sparsely on the ITO substrate during the incipient stage, which form the random ZnO rods in the subsequent growing stage. Although the shape of individual ZnO terrace-tower is similar to the HDVAMT film, the lower density and the random orientation of the ZnO terrace-towers in the disordered ZnO film lead to the lower solar cell efficiency. It is worth noting that the vertically aligned structure plays a more significant role than the rod density because the HDVAMT ZnO solar cell efficiency is about 10 times greater than that of the disordered ZnO, while the rod density of the HDVAMT ZnO film is only about two to three times higher than that of the disordered ZnO film as seen from the SEM images (Figure 2). The significance of this work lies in the simple approach to grow the high-density vertically aligned ZnO rods that can enhance solar cell efficiency. With the synthesized high-density HDVAMT ZnO film, the short circuit current is improved more
than an order, which results in a 10 times increase in the overall solar cell efficiency.
Acknowledgment. We would like to thank Dr. Bing Tan, and Professor Yiying Wu of Department of Chemistry at Ohio State University, for the solar cell efficiency measurement and helpful discussion. We are also grateful to Dr. Richard E. Edelmann of the Electron Microscopy Facility at Miami University for SEM operation training, and the staff of the Instrumentation Laboratory at Miami University for their technical support in fabricating the chemical bath reactor.
References (1) Surek, T. J. Cryst. Growth 2005, 275, 292–304. (2) Tan, B.; Wu, Y. J. Phys. Chem. B 2006, 110, 15932–15938. (3) Baxter, J. B.; Walker, A. M.; Ommering, K. van.; Aydil, E. S. Nanotechnology 2006, 17, S304–S312. (4) Keis, K.; Magnusson, E.; Lindström, H.; Lindquist, S.-E.; Hagfeldt, A. Sol. Energy Mater. Sol.Cells 2002, 73, 51–58. (5) Yoshida, T.; Tochimoto, M.; Schlettwein, D.; Wöhrle, D.; Sugiura, T.; Minoura, H. Chem. Mater. 1999, 11, 2657–2667. (6) Vayssieres, L.; Keis, K.; Hagfeldt, A.; Lindquist, S.-E. Chem. Mater. 2001, 13, 4395–4398. (7) Hara, K.; Horiguchi, T.; Kinoshita, T.; Sayama, K.; Sugihara, H.; Arakawa, H. Sol. Energy Mater. Sol. Cells 2000, 64, 115–134. (8) O’Regan, B.; Schwartz, D. T.; Zakeeruddin, S. M.; Grätzel, M. AdV. Mater. 2000, 12, 1263–1267. (9) Cao, B.; Teng, X.; Heo, S. H.; Li, Y.; Cho, S. O.; Cai, W. J. Phys Chem C 2007, 111, 2470–2476. (10) Law, M.; Greene, Lee; Johnson, J. C.; Saykally, R.; Yang, P. Nat. Mater. 2005, 4, 455–459. (11) Könenkamp, R.; Boedecker, K.; Lux-Steiner, M. C.; Poschenrieder, M.; Zenia, F.; Levy-Clement, C.; Wagner, S. Appl. Phys. Lett. 2000, 77, 2575–2577. (12) Greene, L. E.; Law, M.; Tan, D. H.; Montano, M.; Goldberger, J.; Somorjai, G.; Yang, P. Nano Lett. 2005, 5, 1231–1236. (13) Spatz, J. P.; Mössmer, S,; Hartmann, C.; Möller, M.; Herzog, T.; Krieger, M.; Boyen, H.-G.; Ziemann, P.; Kabius, B. Langmuir 2000, 16, 407–415. (14) Feng, X.; Kang, J.; Inami, W.; Yuan, X.; Terauchi, M.; Sekiguchi, T.; Tsunekawa, S.; Ito, S.; Sakurai, T. Cryst. Growth Des. 2007, 7, 564– 568. (15) Zhang, R.; Kerr, L. L. J. Solid State Chem. 2007, 180, 988–1004.
CG700825V
