Comparison of carbon nanotube modified electrodes for photovoltaic devices
Introduction
Carbon nanotubes (CNTs) are unique materials that hold promise for applications in the electronics industry. This is due to the high conductivity found in CNTs, caused by the delocalised π electrons allowing for ballistic conduction along the tube length [1]. As well as being excellent conductors, carbon nanotubes also possess a complex density of states, allowing them to exist as both metallic conductors and semi-conductors for single-walled carbon nanotubes (SWCNT) and predominantly metallic conductors in the case of multi-walled carbon nanotubes (MWCNT) depending on chirality and tube diameter. One application that makes use of both the conductivity and complex electronic properties of CNTs is photovoltaics. There have been several publications about the response of carbon nanotubes, both SWCNTs [2], [3], [4], [5], [6], [7], [8], [9], [10], [11] and MWCNTs [12], [13], to visible light. These have shown that CNTs by themselves, without any further chemical functionalisation, can be used to generate a photovoltaic response. The general consensus in the literature is that photocurrent generation is caused by electronic transitions between van Hove singularities present in CNTs which then leads to exciton creation upon illumination [6], [12]. SWCNT are generally thought to be the most applicable because the semiconducting nature allows for electron–hole separation, whilst metallic MWCNT provide routes for easy recombination reducing the generated photocurrent [14], although it has been recently shown that MWCNT are excellent hole carriers and may be able to provide charge separation with further chemical modification [15]. This previous work on the photoresponse of CNTs has used varied methods for producing the CNT working electrode, ranging from individual nanotubes [4], [5], [7], [15], growth of CNTs by chemical vapour deposition [13], deposition from solution [9], [16], deposition by electrophoresis [6], [8], [17], [18] and chemical attachment of nanotube arrays to electrodes [11], [19]. Previously, it has been shown that chemical attachment of CNTs to surfaces provides enhanced electron transport when used in electrochemistry compared to unmodified electrode surfaces [20], [21], [22], [23], [24]. This enhanced electron transfer has also been shown to be dependent upon the orientation of the CNTs used, either vertically or horizontally aligned [25]. The use of chemical attachment to create CNT arrays is expected to be advantageous for applications in electronic devices due to the increased electron transfer rates and increased conductivity provided by a direct covalent bond, as opposed to a physisorption process present in grown CNT arrays.
We have previously performed extensive work on creating chemically attached vertically aligned SWCNT arrays on substrates for use as electrochemical sensors [22], [23], [26], [27], field emission sources [28] and photovoltaics [11], [29]. The work presented here allows a direct comparison between chemically attached CNTs, both multi and single-walled, and grown CNTs, again both single and multi-walled for applications such as electrochemical sensors and photovoltaic devices.
Section snippets
SWCNT modified FTO
All chemicals were purchased from Sigma–Aldrich and used as received unless otherwise described. To produce SWCNT modified electrodes a previously published method was used, and shown in Fig. 1 [11]. It is described here in brief. Firstly, purified low functionality single-walled carbon nanotubes (P2-SWCNT, Carbon Solutions, Inc., USA) were sonicated at 0 °C in a 3:1 (v/v) mixture of H2SO4:HNO3 for 8 h. The resulting functionalized nanotube solution was then filtered through a 0.4 μm membrane
Scanning electron microscopy
The morphology of the CNT arrays formed on each sample was analysed using SEM. Fig. 3A shows a top down SEM image of the CNTs grown by tCVD on silicon, showing a dense coverage, the inset shows the side view showing a forest of vertical CNTs with a height of around 20–30 μm, whilst Fig. 3B shows the PECVD modified surface. In Fig. 3B no discernible structures can be seen and other techniques, such as Raman microscopy, must be used to determine the presence of CNTs as is discussed in Section 3.2.
Conclusions
The work presented here has allowed for a direct comparison between chemically attached carbon nanotube arrays and those grown with chemical vapour deposition and also compares single-walled to multi-walled carbon nanotubes. Using Raman spectroscopy it was possible to first identify and verify each nanotube modification had occurred successfully. Cyclic voltammetry results showed that all of the modification of surfaces with carbon nanotubes provided an increased electrode performance over the
Acknowledgement
This work was supported by the Australian Microscopy and Microanalysis Research Facility (AMMRF).
References (44)
- et al.
Size effects on the photoelectrochemical activities of single wall carbon nanotubes
Electrochim Acta
(2008) - et al.
Single walled carbon nanotube network electrodes for dye solar cells
Sol Energy Mat Sol C
(2010) Nanostructuring electrodes with carbon nanotubes: a review on electrochemistry and applications for sensing
Electrochim Acta
(2005)- et al.
Chemically immobilised carbon nanotubes on silicon: stable surfaces for aqueous electrochemistry
Electrochim Acta
(2010) - et al.
Electron-transfer characteristics of ferrocene attached to single-walled carbon nanotubes (SWCNT) arrays directly anchored to silicon(1 0 0)
Electrochim Acta
(2007) - et al.
The effects of the lengths and orientations of single-walled carbon nanotubes on the electrochemistry of nanotube-modified electrodes
Electrochem Commun
(2007) - et al.
Patterned attachment of carbon nanotubes to silane modified silicon
Carbon
(2007) - et al.
Electroless plated gold as a support for carbon nanotube electrodes
Electrochim Acta
(2009) - et al.
Raman spectroscopy of carbon nanotubes
Phys Rep
(2005) - et al.
The big picture of Raman scattering in carbon nanotubes
Vib Spectrosc
(2007)