Introduction.
One of the most important aims of nanotechnology today is the development of new composite materials. At first, the improvement of the mechanical properties of such materials was sought after, but their possible applications have widened due to, for example, their ability to avoid electrostatic discharge and provide electromagnetic interference protection. Their applications now range from daily life to high technology such as mobile phone, computers, electronic-, aeronautic-, car-, aerospace industry, etc. The fabrication of this type of composites can be achieved by mixing conductive additive to the polymer matrix. Among the materials that can be used as additives (metal fibers, carbon black, carbon nanofibers), carbon nanotubes are promising due to their unique characteristics. Their electric conductivity and high aspect ratio (length to diameter ratio) are of great importance to provide a three dimensional conductive network trough the polymer with a low percolation threshold. Their low density provides a low weight polymer-based composite, with significantly improved mechanical properties.
However the formation of a homogeneous composite with as-grown carbon nanotubes has a technological difficulty, due to their non-reactive nature, the presence of amorphous carbon at the nanotube surface and to the natural agglomeration of nanotubes into bundles. To overcome these issues, modification of the carbon nanotubes by changing their surface chemical composition is necessary. Covalent chemical functionalisation of carbon nanotubes can now be achieved by several methods such as electrochemical and chemical functionalisation, mechano-chemical treatment, and plasma treatment. Non-covalent functionalisation can be obtained by π-stacking with aromatic molecules, polymer- or DNA-wrapping, doping, or endrohedral filling with e.g. C60 molecules. Among these techniques, plasma functionalisation has the important advantage to be non polluting, which is not negligible for an industrial production, it can be easily scaled up to produce large quantities necessary for commercial use, reaction times are much lower as compared to other chemical modification methods and it provides the possibility to add a wide range of functional groups depending on the plasma parameters.
Our research consists of the theoretical analysis of chemical properties of carbon nanotubes through a range of semi-empirical and ab initio methods, in close collaboration with different experimental research groups. Below a general outline is given for the different research areas explored, together with the experimental findings.
Plasma functionalisation of carbon nanotubes is based on inductive coupled RF-plasma at 13.56 MHz. This technique can be easily used to tailor the chemical composition of carbon nanotubes by attaching a large variety of functional groups at their surface: i.e. oxygen-, nitrogen- and fluorine- containing groups have been successfully grafted. The functionalisation of the carbon nanotubes can be varied by modifying the plasma conditions (power, type of gas, treatment time, pressure, position of the nanotube sample inside the chamber), as was demonstrated by XPS analysis, and the concentration and type of functional groups added are both in close connection with the plasma conditions. Distinct parameters were varied in order to find the optimal conditions for addition of a whole range of functional groups (hydroxide, carbonyl, carboxyl, amine, fluorine,...). These results are compared to interaction energies predicted by ab initio calculations for different functional groups under consideration. These results show for example that the functional groups with lower interaction energy are obtained in higher concentration using oxygen plasma functionalisation.
Carbon nanotube-based gas sensors can be key devices in the call for the development of low-cost sensing elements and low-power consumption gas sensors. Chemical gas sensors have now been developed with an active layer of functionalised multi-walled carbon nanotubes (see Figure 1). Functionalisation was performed using RF-Plasma under different reaction conditions and plasma composition (oxygen, ferrocene, acrylic acid), and films were prepared by drop coating functionalised nanotubes dispersed in glycerol. XPS analysis demonstrates that the chemical composition depends on the plasma conditions, and SEM micrographs show the morphology of the films to depend on the type of functional groups.
Detection of hazardous gases was investigated for controlled concentrations of different gases (NO2, CO, NH3, ...). Room temperature experiments indicated that the best response was obtained for sensors with oxygen functionalised nanotubes. These results were compared with theoretical ab initio calculations, showing the different adsorption properties for the gases under consideration. Interaction of gas molecules with different functional groups on pristine and on defectuous carbon nanotubes was analysed. In particular, distinct interaction for gas molecules with oxygen functionalised defect sites was demonstrated (see Figure 2). These results predict the presence of these functionalised defect sites after plasma treatment and show their interaction with gas molecules.
Of the available methods for chemical functionalisation of carbon nanotubes, fluorination is emerging as an important process for modifying the nanotube surface and chemically activating carbon nanotubes. It can be achieved with elemental fluorine or F2 gas with HF as catalyst, or with plasma functionalisation (see above), and enables coverage up to C2F. Fluorinated nanotubes can be used as precursors for further functionalisation, and are soluble in a variety of common solvents, aiding tube bundle separation and purification. Defluorination can be achieved using sonication with anhydrous hydrazine as a reagent. Potential applications for fluorinated nanotubes include use in Lithium-ion batteries, supercapacitors and as lubricants.
Fluorination of carbon nanotubes falls into two distinct regimes with a transition between 200 °C and 250 °C, with an increase in bulk sample resistance and changes in observed FTIR. We have now demonstrated via density functional calculations of partially fluorinated nanotubes that this transition corresponds to the barrier for fluorine pairs to migrate to next neighbour sites on the nanotube surface. This barrier blocks close packing and nanotube fluorine banding at lower temperatures, leading instead to a superlattice of 3rd neighbour fluorine corresponding to a maximum 25% surface coverage. Fluorine bonding at different coverages have now been identified and characterised as ionic, semi-ionic and covalent bonding types, depending on whether the fluorine is close packed or not. XPS simulations confirm that these correspond to experimentally observed covalent and semi-ionic fluorination respectively.
Migration barriers and fluorination processes have been determined, and the 200-250 °C transition in behaviour is associated with specific changes in surface diffusion behaviour. The role of HF in catalysing gas fluorination has been analysed and explained through the formation of HF2¯ surface species. Also the effect of solvation on fluorination behaviour is explored. Transition from low to high fluorine coverage has been analysed, and the banded fluorine patterns experimentally observed by STM for high C2F coverage are explained by a more stable contiguous axial addition giving a circumferential banded structure, rather than circumferential addition, due to the higher radial deformation of the latter (see Figure 3). Also the role of structural defects in the carbon nanotube lattice on fluorine addition patterns is currently under further investigation.
In order to fabricate carbon nanotube junctions and networks with novel electronic properties, it is necessary to understand the nanotube coalescence mechanism from an atomistic point of view. Nanotube-nanotube materials could be achieved by controlling the growth of carbon nanotubes using chemical vapour deposition methods in conjunction with organometallic compounds containing Fe and S. Alternatively, it may be possible to achieve coalescence or connect carbon nanotubes by controlling the surface chemical activity between nanotubes. In this context, carbon nanotubes could be joined at defect sites, but a better control of surface defects and the introduction of novel types of imperfections are still needed to catalyse the formation of covalent nanotube interconnections.
Novel possibilities capable of achieving carbon nanotube coalescence and nanotube junctions have now been explored using high-resolution transmission electron microscopy (HRTEM) and resonance Raman studies. In particular, the role played by different types of hetero-atoms in substitution within the lattice or as interstitials between nanotubes is analysed. In order to gain insight in the mechanisms involved, a theoretical analysis was performed using AM1 Molecular Dynamics (MD) simulations at different temperatures, as well as ab initio calculations. The electronic properties of the new systems have also been studied.
In order to perform a systematic analysis of chemical addition and the analysis of addition paths, a modular code has been developed called SACHA (see Figure 4). From any given starting structure, it constructs a range of isomers with different possible chemical addends for single, pair-wise addition or [2+1] addition, atom substitution or atom growth. Input data files for electronic structure calculations are then constructed for structural optimisation with a range of external available codes, from simple inter-atomic potentials such as the Brenner potential, semi-empirical calculations using AM1 or PM3, through to full Hartree-Fock or Density Functional Theory calculations using Gaussian03 or AIMPRO. Input structures can be pre-processed by adapting the bond lengths, angles or pyramidalisation angles of the addends for faster structural optimisation. The lowest energy isomer is chosen from these optimised structures at the chosen level of theory, and used as a starting structure for the next addition. All obtained isomers are outputted at each step in lowest-energy ordering, and if desired as input files for treatment at higher-level calculations up to a given energy cut-off. The extension of the addition from the first addition site can be adapted by modifying the surface distance cut-off radius, in order to restrict the number of isomers treated at each step.
As this code permits the construction of a large number of isomers without further manual interaction, this results in an important gain of time. It can be easily adapted for use with a large range of available external codes and for different types of addends. It has been applied in the study of nitrogen substitution, hydrogenation and fluorination of C60 (see Figure 5), and for various studies of carbon nanotube functionalisation.
Dr. Carla Bittencourt, Laboratoire Interdisciplinaire de Spectroscopie Electronique (LISE), Facultés Universitaires Notre-Dame de la Paix (FUNDP), Namur, Belgium.
Prof. Alessandro De Vita, Universita' di Trieste, Italy.
Dr. Christopher P. Ewels, Laboratoire de Physique du Solide, Orsay, France.
Drs. Alexandre Felten, Laboratoire Interdisciplinaire de Spectroscopie Electronique (LISE), Facultés Universitaires Notre-Dame de la Paix (FUNDP), Namur, Belgium.
Prof. Paul Geerlings, Department of General Chemistry (ALGC), Vrije Universiteit Brussel (VUB), Brussels, Belgium.
Prof. Eduard Llobet, Universitat Rovira i Virgili, Spain.
Prof. Jean-Jacques Pireaux, Laboratoire Interdisciplinaire de Spectroscopie Electronique (LISE), Facultés Universitaires Notre-Dame de la Paix (FUNDP), Namur, Belgium.
Prof. Mauricio Terrones, IPICYT, San Luis Potosi, Mexico.
Drs. Filippo Zuliani, Universita' di Trieste, Trieste, Italy.
Convention Région Wallone n° 02/1/5225; Programme"Nanotechnologies" ENABLE: Procédé novateur de blindage électromagétique par recours à des nanotubes de carbone.
Fonds National de la Recherche Scientifique (FNRS - Wallonia).
"Pôle d'Attraction Interuniversitaire" P5/01 on "Quantum size effects in nanostructured materials" (UCL-FUNDP-KUL-RUCA-UIA) 2002-2006.
European Network of Excellence (N° NMP3-CT-2004-500159)"FAME : Functionalized Advanced Materials and Engineering : Hybrids and Ceramics " 2004-2008.