Annealing effects of ZnO nanorods on dye-sensitized solar cell efficiency

https://doi.org/10.1016/j.physb.2010.03.041Get rights and content

Abstract

Dye-sensitized solar cells (DSSCs) were fabricated using ZnO nanorod arrays vertically grown on fluorine-doped tin oxide (FTO) glass using a low-temperature hydrothermal method. When the ZnO seed layer was annealed, greater DSSC efficiency was obtained. This may be attributed to the improvement of adhesion between the FTO and the seed layer and the corresponding effective growth of the ZnO nanorods. The DSSCs fabricated using ZnO nanorods which underwent annealing were more efficient than those that did not undergo annealing. The ZnO nanorods which were annealed in N2/H2 or O2 had increased dye loadings due to higher OH concentrations on the hydrophilic surface, which contributed to the improved DSSC efficiency. The fill factor increased after the annealing of the ZnO nanorods, potentially due to the improved crystallinity of the ZnO nanorods. In this study, annealing of both the seed layer and the ZnO nanorods resulted in the greatest DSSC efficiency.

Introduction

A dye-sensitized solar cell (DSSC) is a promising option for solar cells due to relatively low manufacturing costs [1], [2]. A DSSC fabricated using TiO2 nanoparticles exhibits 11% cell efficiency [3]. The DSSC has a fast electron injection from the excited dye to the conduction band of the TiO2, with an injection rate constant of 1013 s−1. This is greater than that resulting from the recombination with the oxidized dye, characterized by a back reaction rate constant of 1012 s−1 [4]. When a transition metal complex is used as the dye, the injection rate is greater than the back reaction rate constant by three orders of magnitude [5].

Alternatively, injected electrons from the excited dye experience trapping and detrapping events at the TiO2 nanoparticle grain boundary and the nanoparticles/electrolyte interface. The injected electrons experience 106 times the number of detrapping events because it takes 10 ms to pass through a nanoparticle anode [6], [7], [8]. As a result, the electron diffusion coefficient is significantly low for the TiO2 anode of DSSC, 5×10−5 cm2/s [9], [10]. It has been reported that the recombination of electrons with I3 also reduces cell efficiency [11], [12]. In addition, the diffusion length of the electron in TiO2 under 1000 W/m2 is as short as 4–12 μm [11], [13], [14], [15], which has to be greater than the anode thickness. In addition, the anode thickness must be thicker than the light absorption length in order to harvest light in a wide range of wavelengths [16]. As a result, it is difficult to reach theoretical cell efficiency. Therefore, an alternative anode material is required to achieve a faster charge transfer, lower recombination rate, and longer diffusion length.

Zinc oxide is a promising candidate for an alternative anode material. The band gap energy and conduction band edge of ZnO are similar to those of TiO2 [17], [18]. In addition, it has been reported that the electron diffusion coefficient of the ZnO anode of DSSC is 1.2×10−4 cm2/s [19]. Specifically, DSSCs fabricated using ZnO nanocrystalline particles for the anode exhibited 5.4% cell efficiency [20]. A significantly fast electron injection time below 300 fs and a high incident photon-to-charge carrier efficiency are observed if the aggregation of dye with Zn2+ is avoided [18]. The recombination with O2 and I3 in the electrolyte is negligible [21].

Mesoporous nanoparticles are used as anodes because their larger inner surface areas absorb dye. However, the nanoparticles have slower electron transport due to the multi-trapping events. One way of addressing this problem is to introduce a nanorod structure to the anode. Zinc oxide has an additional advantage in that nanorods can be easily synthesized using a hydrothermal method [22]. Zinc oxide nanorods have an internal electrical field in the direction of the c-axis to facilitate charge transport and suppress the recombination of injected electrons [23]. Furthermore, the diffusion coefficient of the electrons in the ZnO nanorod is approximately two orders of magnitude greater than that in the ZnO nanoparticle [23]. This implies that use of ZnO nanorods in DSSCs may improve cell efficiency compared to DSSCs containing ZnO particles.

Recently, ZnO nanorod-based DSSCs have been investigated. For example, the strategy to increase the surface area was investigated because ZnO nanorods have a surface area that is one-fifth of TiO2 particles [23]. The effects of nanorod growth conditions, such as hydrothermal synthesis time [24], [25] and growth temperature in metal-organic chemical vapor deposition (MOCVD), on the performance of DSSCs were studied [26]. Nanorod length and cell efficiency have a linear relationship because of the large amount of dye loading on the enlarged surface area. Additionally, the synthesis of ZnO nanorods in the hydrothermal method resulted in enhanced DSSC performance compared to the MOCVD [27]. The branched nanorod structure significantly improved the cell efficiency compared to the simple nanorod structure [28], and the nanoparticle/nanorod composite structure improved efficiency from 0.45% to 1.27% [29].

In particular, it is suggested that the surface condition of the ZnO nanostructure is a key factor in determining cell performance because dye and dissolved Zn2+ on the ZnO surface result in aggregation that decreases injection efficiency [30], [31], and poor interaction between ZnO and dye decreases the light harvesting efficiency [32]. In addition, the crystallinity of ZnO nanorods may be an important parameter because the electron transfer can be determined by grain size. One effective method to modify the surface condition and crystallinity of ZnO nanorods is annealing, which will change the diameter of nanorods [33] and release some functional groups, such as carbon and nitrogen [34]. Thermal annealing also improves the crystallinity of ZnO nanorods by decreasing the oxygen vacancy concentration [35] and deep level defects or surface defect recombination [34]. Nevertheless, the various effects of the surface condition and crystallinity of ZnO nanorods on the DSSC have not yet been systematically elucidated. The aim of this study is to determine the effect of the annealing of ZnO nanorods on DSSC cell efficiency and to make a correlation between cell efficiency and the amount of dye loading and crystallinity of ZnO nanorods.

Section snippets

Synthesis of ZnO nanorod arrays

Before synthesizing ZnO nanorods, the ZnO seed layer was deposited onto fluorine-doped tin oxide (FTO) glass (Hartford Glass Co.) using radio frequency (RF) sputtering. The sputtering was performed at room temperature at 1 mTorr using a ZnO (99.99% purity) target in an Ar (99.999% purity) gas atmosphere at RF power of 80 W. The ZnO seed layers on the FTO glasses were optionally annealed in a quartz tube furnace at 300 °C and 1 Torr in O2 or N2/H2 (N2:H2 =9:1) ambient condition at a flow rate of 20 

Performance of the DSSC

The performance and efficiency of the DSSCs fabricated under different conditions were measured. The annealing conditions and measurement results are summarized in Table 1. Two important trends were observed. When the ZnO seed layer was annealed, the efficiency of the DSSCs increased whether or not the ZnO nanorods were annealed (e.g., A→B, A→C, A→D). Annealing of the ZnO nanorods in N2/H2 or O2 ambient condition increased the efficiency of the DSSCs with an annealed seed layer (e.g., B→C, B→D).

Conclusions

In summary, DSSCs were fabricated using vertically aligned ZnO nanorod arrays on FTO glasses. Annealing of the ZnO seed layer and nanorods in N2/H2 or O2 ambient condition was conducted, and the effect on ZnO nanorod properties and the efficiency of DSSCs were studied. The DSSCs with an annealed ZnO seed layer exhibited greater cell performance than those that were not annealed. It is noted that annealing of the seed layer improved adhesion between the FTO and the seed layer, and ZnO nanorods

Acknowledgments

This work is an outcome of a Manpower Development Program for Energy & Resources supported by the Ministry of Knowledge and Economy (MKE). This work was also supported by Priority Research Centers Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2009-0093823).

References (53)

  • H. Gao et al.

    Mater. Res. Bull.

    (2008)
  • G. Hua et al.

    Mater. Lett.

    (2008)
  • D.-I. Suh et al.

    Chem. Phys. Lett.

    (2007)
  • P. Falaras

    Sol. Energy Mater. Sol. Cells

    (1998)
  • K. Murakoshi et al.

    J. Electroanal. Chem.

    (1995)
  • L. Wu et al.

    Opt. Mater.

    (2006)
  • K. Kalyanasundaram et al.

    Coord. Chem. Rev.

    (1998)
  • L.H. Quang et al.

    J. Cryst. Growth

    (2006)
  • S. Greg et al.

    Sol. Energy Mater. Sol. Cells

    (1994)
  • G.L. Roger et al.

    Prog. Photovoltaics

    (1997)
  • Y. Chiba et al.

    Jpn. J. Appl. Phys.

    (2006)
  • N.J. Cherepy et al.

    J. Phys. Chem. B

    (1997)
  • A. Hagfeldtt et al.

    Chem. Rev.

    (1995)
  • J. van de Lagemaat et al.

    J. Phys. Chem. B

    (2001)
  • G. Boschloo et al.

    J. Phys. Chem. B

    (2005)
  • G. Schlichthörl et al.

    J. Phys. Chem. B

    (1997)
  • L. Dloczik et al.

    J. Phys. Chem. B

    (1997)
  • N. Kopidakis et al.

    Phys. Rev. B

    (2006)
  • T. Oekermann et al.

    J. Phys. Chem. B

    (2004)
  • A.C. Fisher et al.

    J. Phys. Chem. B

    (2000)
  • J. Krüger et al.

    J. Phys. Chem. B

    (2003)
  • J.R. Jennings et al.

    J. Phys. Chem. C

    (2007)
  • F. Fabregat-Santiago et al.

    J. Phys. Chem. C

    (2007)
  • M. Grtäzel

    Inorg. Chem.

    (2005)
  • Å Petersson et al.

    J. Phys. Chem. B

    (2000)
  • C. Bauer et al.

    J. Phys. Chem. B

    (2001)
  • Cited by (85)

    • Comparative study of nickel selenide, iron selenide and platinum on triiodide reduction for dye-sensitized solar cells

      2022, Optical Materials: X
      Citation Excerpt :

      Many of these techniques are to be performed at relatively high temperature and complicated synthetic routes. Then, hydrothermal emerged as a simple yet effective synthetic technique for metal-chalcogenides [15,16]. In this work, we synthesized nickel selenide (NiSe2) and iron selenide (FeSe2) via a mild hydrothermal reduction route.

    • Effect of morphology and surface treatment on the performance of ZnO nanorod-based dye-sensitized solar cells

      2019, Journal of Alloys and Compounds
      Citation Excerpt :

      Due to these excellent optoelectronic properties and rich structural verities, ZnO has also been used in different photoelectrical devices like: as an output enhancement in InGaN light-emitting diodes [7], as component in ZnO/Cu2O solar cell [8] and as an electron transport and as a photoresponse enhancement in CsPbBr3 based perovskite photodetector [9]. In addition, ZnO possesses a suitable band alignment with well-established indoline dyes [4–6,10–15], which results in efficient electron injection from organic molecules into its conduction band [3]. However, in spite of the numerous studies over the past few years, the reported efficiencies of ZnO-based solar cells are still significantly lower than those of their TiO2–based counterparts.

    • ZnO-based dye-sensitized solar cells

      2019, Nanomaterials for Solar Cell Applications
    View all citing articles on Scopus
    View full text