1577
Recent Advances on Electrically Induced Light Emission from Carbon Nanotubes
The approach consists of operating the nanotubes transistors in different conditions of thermalisation and of exploring for each condition the different signatures in the emission, either thermal or electroluminescent, using an analysis of the resulting near-IR spectrum. The spectra were acquired in the energy range between 0.5 and 1.3 eV using the Spectrometer Infrared of Montreal (SIMON) mounted with a HgCdTe detector and on a probe station.
Three different device configurations were measured. The first is a sub-monolayer network of nanotubes interconnected together to form a percolating semiconducting layer. [5] The numerous tube-tube junctions in the networks limit the current and prevent temperature raise by joule heating at higher bias. Electroluminescence from exciton recombination is found to dominate in all conditions, evidenced by well-defined emission peaks (~200meV in width) in the near-IR (Figure 1). The emission is ascribed to a radiative relaxation of the first excitonic transition. Even at very large biases, the main component of the emission from network transistors is electroluminescence, not incandescence. The second device geometry consists of a thick (~250 nm) and suspended carbon nanotube film having low resistance and reduced thermalization. This configuration favors joule heating when operating at large source-drain voltage and allows to significantly increase the nanotube temperature. The resulting incandescence was measured and used to calibrate SIMON for blackbody emission from a thermal source of pure nanotubes. A third set of devices was finally prepared with individual SWNT and DWNT transistors operating in vacuum, which conditions is used to prevent convection and for reducing the cooling capability of the substrate. The spectra were then compared with the different spectral signatures ascribed either to electroluminescence or thermal radiation depending on the biasing conditions. The experiments show a transition from electro-luminescence at low source-drain voltage to incandescence at higher voltage, with a regime of both in-between.
These results will be discussed in terms of the dominant dissipation channels and carrier scatterings in nanotubes. Last, the best conditions to maximize light emission while avoiding heating and progress towards our effort to build electroluminescent transistors at 1,55 μm wavelength using sorted SWNTs will be discussed.
Figure 1 Left: SEM and infrared micrographs of a large SWNT network transistor. Right: A carbon nanotube network transistor with the electroluminescence spectra at different biases. [Adapted from Ref. 5]
REFERENCES [1] J. A. Misewich, R. Martel, Ph. Avouris, J. C. Tsang, S. Heinze, and J. Tersoff, Science, 300, 783 (2003). [2] J. Chen, V. Perebeinos, M. Freitag, J. Tsang, Q. Fu, J. Liu, and P. Avouris, Science, 310, 1171 (2005). [3] L. Marty, E. Adam, L. Albert, R. Doyon, D. Ménard and R. Martel, Phys. Rev. Lett. 96, 136803 (2006). [4] D. Mann, K. K., A. Kinkhabwala, E. Pop, J. Cao, X. Wang, L. Zhang, Q. Wang, J. Guo, and H. Dai, Nat. Nano, 2, 33 (2006). [5] E. Adam, C. M. Aguirre, L. Marty, B. C. St-Antoine, F. Meunier, P. Desjardins, D. Menard, and R. Martel, Nano Lett. 8, 2351 (2008).
This work was done in collaboration with : Elyse Adam1, Pierre Lévesques2, Étienne Gaufrès2, Vincent Aymong2, David Ménard1 : RQMP and 1École Polytechnique de Montréal, Département de génie physique, Montréal, Québec H3C 3A7, Canada 2Université de Montréal, Département de chimie, Montréal, Québec H3T1J4, Canada