Langmuir 2007, 23, 9501-9504
9501
Use of High-Purity Metal-Catalyst-Free Multiwalled Carbon Nanotubes To Avoid Potential Experimental Misinterpretations Craig P. Jones,† Kerstin Jurkschat,‡ Alison Crossley,‡ Richard G. Compton,§ Bill Logan Riehl,| and Craig E. Banks*,† Chemistry, School of Biomedical and Natural Sciences, Nottingham Trent UniVersity, Clifton Campus, Nottingham NG11 8NS, Department of Materials, UniVersity of Oxford, Parks Road, Oxford OX1 3PH, and Physical and Theoretical Chemistry Laboratory, Oxford UniVersity, South Parks Road, Oxford OX1 3QZ, United Kingdom, and Phoenix NanoSystems, LLC, 2000 Composite DriVe, Kettering, Ohio 45420 ReceiVed May 24, 2007. In Final Form: June 22, 2007 Carbon nanotubes, even after extensive posttreatment, contain metallic impurities which may produce misleading results, giving rise to false claims of the properties of carbon nanotubes. To overcome this, we report on high-purity catalyst-free multiwalled carbon nanotubes which have been explored with transmission electron microscopy, X-ray photoelectron spectroscopy, and cyclic voltammetry using the electrochemical oxidations of hydrazine and potassium ferrocyanide. The multiwalled carbon nanotubes are approximately 150 nm in length and consist of 6-10 graphite layers. Due to the definitive absence of metallic impurities, experimentalists using these carbon nanotubes can avoid potential misinterpretations of their results.
Introduction Carbon nanotubes are at the forefront of nanotechnological research due to their reported unique structural and physical properties and receive extensive attention in a plethora of areas.1-8 Carbon nanotubes can be fabricated in a variety of ways such as via arc discharge,9 laser ablation,10 and, most commonly, chemical vapor deposition (CVD).11 Such fabrication processes inevitably utilize a metal catalyst and result in the production of carbon nanotubes which contain metal impurities and require extensive posttreatment.12 However, technological advancements rely on the high purity of the carbon nanotube material. A key example of the problems associated with catalyst metal particles and carbon nanotubes is in the case of hydrogen storage. * To whom correspondence should be addressed. E-mail: craig.banks@ ntu.ac.uk. † Nottingham Trent University. ‡ Department of Materials, University of Oxford. § Physical and Theoretical Chemistry Laboratory, University of Oxford. | Phoenix NanoSystems, LLC. (1) Heller, D. A.; Jeng, E. S.; Yeung, T.-K.; Martinez, B. M.; Moll, A. E.; Gastala, J. B.; Strano, M. S. Science 2006, 311, 508-511. (2) Guldi, D. M.; Rahman, G. M. A.; Prato, M.; Jux, N.; Qin, S.; Ford, W. Angew. Chem., Int. Ed. 2005, 44, 2015-8. (3) Wu, Z.; Chen, Z.; Du, X.; Logan, J. M.; Sippel, J.; Nikolou, M.; Kamaras, K.; Reynolds, J. R.; Tanner, D. B.; Hebard, A. F.; Rinzler, A. G. Science 2004, 305, 1273-1277. (4) Liu, Y.; Wu, D.-C.; Zhang, W.-D.; Jiang, X.; He, C.-B.; Chung, T. S.; Goh, S. H.; Leong, K. W. Angew. Chem., Int. Ed. 2005, 44, 4782-5. (5) Hinds, B. J.; Chopra, N.; Rantell, T.; Andrews, R.; Gavalas, V.; Bachas, L. G. Science 2004, 303, 62-65. (6) Baughman, R. H.; Zakhidov, A. A.; de Heer, W. A. Science 2002, 297, 787-792. (7) Wu, W.; Wieckowski, S.; Pastorin, G.; Benincasa, M.; Klumpp, C.; Briand, J.-P.; Gennaro, R.; Prato, M.; Bianco, A. Angew. Chem., Int. Ed. 2005, 44, 635862. (8) Katz, E.; Willner, I. ChemPhysChem 2004, 5, 1084-1104. (9) Mathur, R. B.; Seth, S.; Lal, C.; Rao, R.; Singh, B. P.; Dhami, T. L.; Rao, A. M. Carbon 2007, 45, 132-140. (10) Moravsky, A. P.; Wexler, E. M.; Loutfy, R. O. Carbon Nanotubes 2005, 65-97. (11) Ivanov, V.; Nagy, J. B.; Lambin, Ph.; Lucas, A.; Zhang, X. B.; Zhang, X. F.; Bernaerts, D.; Van Tendeloo, G.; Amelinckx, S.; Van Landuyt, J. Chem. Phys. Lett. 1994, 223, 329-35. (12) Park, T.-J.; Banerjee, S.; Hemraj-Bennya, T.; Wong, S. S. J. Mater. Chem. 2006, 16, 141-154.
Dillon et al.13 reported the promising storage of hydrogen in single-walled carbon nanotubes (SWCNTs), which received considerable interest. However, Tibbets et al.14 have reported that they believe claims of more than 1 wt % hydrogen at room temperature for carbon nanotubes are erroneous and due simply to experimental errors.14 Hirscher et al.15 have shown that posttreatment of single-walled carbon nanotubes influence the adsorption capacity, and Costa et al.16 have shown that residual catalyst metal particles strongly influence the hydrogen storage capacity;16 hence, the presence of metal particles in carbon nanotubes, either introduced during posttreatment15 or from residual catalyst particles,16 provide a means for hydrogen adsorption, rather than the carbon nanotubes themselves, and allowed false claims to the properties of carbon nanotubes to be made. Another area where this has been highlighted is in the case of using multiwalled carbon nanotubes (MWCNTs) as electrode materials, which have the possibility of producing nanoelectrochemical sensors, for example, which may be planted under the skin to monitor, in real time, glucose levels. Recently we have challenged the general belief that carbon nanotubes are electrocatalytic.17,18 We have shown that catalyst iron impurities trapped in MWCNTs can, in certain cases, such as in sensing hydrogen peroxide, which is one of the most important analytes ever determined with carbon nantoubes, dominate the electrochemical response.17,18 Consequently, these trapped metallic impurities have resulted in misinterpretation of the electrochemical origins of heterogeneous charge transfer. This has implications in producing sensors based on carbon nanotubes since the amount (13) Dillon, A. C.; Jones, K. M.; Bekkedahl, T. A.; Kiang, C. H.; Bethune, D. S.; Heben, M. J. Nature (London) 1997, 386, 377-379. (14) Tibbetts, G. G.; Meisner, G. P.; Olk, C. H. Carbon 2001, 39, 2291-2301. (15) Hirscher, M.; Becher, M.; Haluska, M.; Dettlaff-Weglikowska, U.; Quintel, A.; Duesberg, G. S.; Choi, Y.-M.; Downes, P.; Hulman, M.; Roth, S.; Stepanek, I.; Bernier, P. Appl. Phys. 2001, A 72, 129-132. (16) Costa, P. M. F. J.; Coleman, K. S.; Green, M. L. H. Nanotechnology 2005, 16, 512-517. (17) Banks, C. E.; Crossley, A.; Salter, C.; Wilkins, S. J.; Compton, R. G. Angew. Chem., Int. Ed. 2006, 45, 2533-2537. (18) Sljukic, B.; Banks, C. E.; Compton, R. G. Nano Lett. 2006, 6, 15561558.
10.1021/la701522p CCC: $37.00 © 2007 American Chemical Society Published on Web 07/26/2007
9502 Langmuir, Vol. 23, No. 18, 2007
Jones et al.
of metallic impurity can vary from sample to sample, giving a large deviation in the electrochemical response from one sensor to another. All current methods of producing carbon nanotubes result in catalyst impurities which need extensive purification strategies.12 The assumption that a homogeneous material is obtained is often a false one, and even a small amount of heterogeneity in the sample can produce potentially erroneous results. In fact, quantifying the amount of heterogeneity can be problematic.19 In this paper we report on novel multiwalled carbon nanotubes which are fabricated via a solid-phase growth mechanism. This process produces high-purity, metal-impurity-free multiwalled carbon nanotubes which provide clear advantages over other currently commercially available carbon nanotubes.
Results and Discussion The carbon nanotubes were obtained from Phoenix NanoSystems (Kettering, OH), which utilizes a catalyst-free growth process based on a solid-state production method. This means that high-purity (99.99+%) multiwalled carbon nanotubes can be produced without the worry of catalyst impurities and alleviating the need for lengthy posttreatment of the carbon nanotubes. Transmission electron microscopy (TEM) was used to examine the structure of the multiwalled carbon nanotubes. Figure 1A shows that the multiwalled carbon nanotubes exist in clusters which have a unique 3-D nanoporous structure; this has obvious implications for scaleup and related applications. Close examination of the clusters reveals that they consist of multiwalled carbon nanotubes (Figure 1B) which are typically 150 ((50) nm in length and are made up of between 6 and 10 graphite layers. Raman scattering is known to be a very powerful technique for determining the structure, chirality, electronic properties, and diameter of carbon nanotubes. It has gained great acceptance for use with SWCNTs in particular, but is viewed by many as suspect for MWCNTs, largely due to the great variation of diameter in commonly available materials. With these materials, the relatively weak radial breathing mode (RBM) can be lost in the noise even though it may be present. A spectrum of the carbon nanotube material is shown in the Supporting Information. RBMs are shown at 111, 163, and 264 cm-1, indicating the presence of carbon nanotubes with diameters of 2.3, 1.5, and 0.9 nm, respectively, as calculated via the equation provided in ref 20. The graphitic peak at 1576 cm-1 was observed, as well as the G* peak at 2608 cm-1. Of notable interest is the large defect, or D, peak observed at 1306 cm-1. The high intensity of this peak corresponds to a high level of “defects” in the tubes, in this case the large number of “kinks” and other features that lead to the high edge plane character of the tubes. One of the unique facets of the MWCNTs is the high amount of defect sites, which are not typically shown in MWCNT Raman spectra. Next X-ray photoelectron spectroscopy (XPS) was used to examine the MWCNTs, as shown in Figure 2. Analysis of the XPS data indicates the MWCNTs to be 91.5 atom % carbon, 6 atom % oxygen, and 2.5 atom % silicon. This again demonstrates the true absence of any metallic impurities. We note that silicon is one of the most common contaminants in processing, and since silicon is not used in the fabrication process of the MWCNTs, we conclude that this arises due to postproduction. Due to the relative amount (atom %) of oxygen present on the MWCNTs, (19) Itkis, M. E.; Perea, D. E.; Jung, R.; Niyogi, S.; Haddon, R. C. J. Am. Chem. Soc. 2005, 127, 3439-3448. (20) Rols, S.; Righi, A.; Alvarez, L.; Anglaret, E.; Almairac, R.; Journet, C.; Bernier, P.; Sauvajol, J. L.; Benito, A. M.; Maser, W. K.; Munoz, E.; Martinez, M. T.; De la Fuente, G. F.; Girard, A.; Ameline, J. C. Eur. Phys. J. B: Condens. Matter Phys. 2000, 18, 201-205.
Figure 1. Transmission electron microscopy of the multiwalled carbon nanotubes.
Figure 2. X-ray photoelectron spectra of the multiwalled carbon nanotubes.
the XPS spectra could not be deconvoluted to determine the exact presence of oxygen functionalities, and there likely is a combination of O-O, C-O, and CdO functionalities present. However, these functionalities serve to promote a means of binding absorbed species. The multiwalled carbon nanotubes were studied via cyclic voltammetry by immobilization on a basal plane pyrolytic graphite
Use of High-Purity Metal-Catalyst-Free MWCNTs
Figure 3. Voltammetric profiles of (A) bamboo multiwalled carbon nanotubes and (B) new multiwalled carbon nanotubes in a solution containing 1 mM hydrazine in pH 7.1 phosphate buffer.
electrode. This electrode is prepared so that the heterogeneous charge-transfer kinetics of the species in solution is relatively slow. This is achieved by careful preparation of the electrode, which involves polishing the surface on carborundum paper and then pressing Cellotape onto the basal plane pyrolytic graphite electrode surface, which removes generally attached graphite layers. This preparation method, which is described in more detail in the Experimental Section, produces a low coverage of edge plane sites/defects on the highly ordered pyrolytic graphite (HOPG) surface. Note that, depending on the preparation of the basal plane electrode, the peak-to-peak separation at the basal plane electrode can be up to 500 mV, depending on the percentage coverage of edge plane sites on the electrode surface. Conversely, edge plane sites/defects can be introduced by roughening of the electrode surface, simply induced by polishing the basal plane electrode with alumina on a soft lapping pad. After the basal plane pyrolytic graphite electrode is freshly prepared, the multiwalled carbon nanotubes, which are dispersed in a suitable organic solvent (see Experimental Section), are then pipetted onto the electrode surface. Thus, any subtle changes in the voltammetric response can be attributed solely to the carbon nanotubes. Note that if the nanotubes were put onto an electrode surface which exhibited electron transfer at approximately the same rate as the carbon nanotubes, deconvolution would be difficult. The electrochemical response of the multiwalled carbon nanotubes is compared to that of existing commercial “bamboo” multiwalled carbon nanotubes using the electrochemical oxidation of hydrazine. This electrochemical probe is highly sensitive to metallic impurities, with an electrochemical signal only observed when metallic impurities are present and not the carbon nanotubes themselves (viz., edge plane like sites/defects) due to the sluggish heterogeneous charge-transfer kinetics on the latter. Figure 3A shows the voltammetric profiles resulting from the bamboo multiwalled carbon nanotubes where a large electrochemical oxidation wave is observed at ∼+0.46 V (vs SCE), while in contrast, no electrochemical oxidation wave is observed on the new multiwalled carbon nanotubes (Figure 3B), indicating the definitive absence of any metallic impurities; this clearly demonstrates advantages of the new metal-catalyst-free multiwalled carbon nanotubes since with carbon nanotubes fabricated with metal catalysts, one has to always exercise caution in interpretation of the experimental response.17,18 Next the response of the multiwalled carbon nanotubes in a 1 mM solution of potassium ferrocyanide/0.1 M KCl was explored, as shown in Figure 4. The peak-to-peak separation of 60 mV indicates fast heterogeneous charge-transfer kinetics, which compares well with the 66 mV previously reported using bamboo multiwalled carbon nanotubes.21 Increasing the quantity (21) Ji, X.; Banks, C. E.; Crossley, A.; Compton, R. G. ChemPhysChem 2006, 7, 1337-1344.
Langmuir, Vol. 23, No. 18, 2007 9503
Figure 4. Cyclic voltammetric profiles of the multiwalled carbon nanotubes immobilized onto a basal plane pyrolytic graphite electrode in a 1 mM solution of potassium ferrocyanide/0.1 M KCl. The dotted line represents the bare basal plane pyrolytic graphite electrode. Curves A-D result from increasing amounts, 20-80 µg, of multiwalled carbon nanotubes immobilized onto the basal plane pyrolytic graphite electrode surface (see the Experimental Section).
of multiwalled carbon nanotubes on the surface of the basal plane pyroltyic graphite electrode, as shown in Figure 4, results in the voltammetric peak height increasing in magnitude. Note that comparison of the voltammetric profile in Figure 4A with those of Figure 4B-D reveals that the profile of the voltammogram changes in shape as a result of increasing the amount of multiwalled carbon nanotubes on the surface of the basal plane pyrolytic graphite electrode. This difference in the voltammetric profile reflects the fact that as more multiwalled carbon nanotubes are put on the basal plane pyrolytic graphite electrode surface a nanotube film results which is potentially “porous”, leading to “thin layer” behavior and loss of the diffusional tail. Given the above electrochemical responses, the impurity of silicon, as found via XPS, does not appear to affect the overall electrochemical activity. The fast heterogeneous charge transfer, and increments in the peak height from increasing the quantity of multiwalled carbon nanotubes on the surface of the basal plane pyrolytic graphite electrode, indicates a high proportion of edge plane like sites/defects22 occur on the multiwalled carbon nanotubes; note that in comparison to bamboo multiwalled carbon nanotubes and other carbon nanotubes we can rule out the possibility of metallic impurities contributing to the electrochemical activity17,18 and we can attribute the sole origin of the electrochemical activity to be due to edge plane like sites/defects, since these novel multiwalled carbon nanotubes are fabricated without a catalyst.
Summary In summary we have presented a new class of multiwalled carbon nanotubes, which due to the definitive absence of metallic impurities will allow experimentalists to avoid any potential experimental misinterpretations, not only in electrochemistry and electrocatalysis, but also in related applications. These new high-purity metal-catalyst-free multiwalled carbon nanotubes have implications for technological advancements. Experimental Section Chemicals. All chemicals were purchased from Aldrich, obtainable at the highest grade available, and used directly without further purification. All experiments were carried out at a temperature of 295 ( 3 K. Aqueous electrolyte solutions were prepared with KCl (99.5%) (Aldrich) using ultrapure water from a Vivendi UHQ grade water system (Vivendi, U.K.) with a resistivity of not less than 18.2 MΩ cm-1. All the solutions were degassed with oxygen-free nitrogen (BOC Gases, Guildford, Surrey, U.K.). (22) Banks, C. E.; Moore, R. R.; Davies, T. J.; Compton, R. G. Chem. Commun. 2004, 16, 1804-1805.
9504 Langmuir, Vol. 23, No. 18, 2007 Instrumentation. Electrochemical experiments were performed using a µ-Autolab type III potentiostat (Eco-Chemie, Utrecht, The Netherlands) controlled by General Purpose Electrochemical Systems v.4.7 software. For all electrochemical experiments carried out in the electrolyte, the working electrode used was a basal plane pyrolytic graphite electrode (4.9 mm diameter, Le Carbone, Ltd., Sussex, U.K.). The counter electrode was a bright platinum wire with a large surface area, with a saturated calomel reference electrode completing the circuit. The basal plane pyrolytic graphite electrode was prepared by first polishing the electrode surface on carborundum paper and then pressing Cellotape onto the cleaned basal plane pyrolytic graphite surface, removal of which also removes generally attached graphite layers.23 Before use the electrode was then cleaned in acetone to remove any adhesive. Onto the desired freshly prepared electrode surface were cast the multiwalled carbon nanotubes, which had been suspended in methanol, 15 mg in 15 mL of methanol. The suspension was placed into an ultrasonic bath for 1 min, after which microliter aliquots, 20, 40, 60, and 80 µL, were pippetted onto the electrode surface, which correspond to curves A, 20 µg, B, 40 µg, C, 60 µg, and D, 80 µg, respectively, in Figure 4. This was allowed to volatize (23) Moore, R. R.; Banks, C. E.; Compton, R. G. Anal. Chem. 2004, 76, 2677-2682.
Jones et al. at room temperature, producing a presumed random distribution of carbon nanotubes on the electrode surface. For comparison, bamboo multiwalled carbon nanotubes were obtained from NanoLab (Brighton, MA) which were 30 ((15) nm in diameter and 5-20 µm in length. These were grown via CVD on a silica wafer which supports the iron metal catalyst. These carbon nanotubes were again dispersed in a suitable organic solvent and pippetted onto the electrode surface, as described above. XPS was performed in an ion-pumped UHV chamber equipped with a VG nine-channel CLAM4 electron energy analyzer (base pressure 5 × 10-10 Torr), and 250 W Mg X-ray (1253.6 eV) excitation was used. The analyzer was operated at a constant pass energy of 100 eV. Data were obtained using the VGX900-W operating system. TEM micrographs were taken on a JEOL 3000FEG (TEM) instrument, which was equipped with an Oxford Instruments energydispersive X-ray spectrometer with a superatmospheric thin window (SATW).
Supporting Information Available: Raman spectra of the MWCNTs. This information is available free of charge via the Internet at http://pubs.acs.org. LA701522P
