Single carbon nanotube diode. (a) An I-V characteristic of the device without illumination, showing a rectifying behavior, characteristic for a diode. A schematic of the device geometry is shown in the inset. +8/−8 V was applied to the trench gates to create the p-n junction. (b) Scanning photocurrent microscopy image of the single carbon nanotube p-n junction. The excitation laser (532 nm) was raster scanned over the device, generating a spatial image of the location of the optical response. The scale bar is 2 μm.
Photoresponse in a single carbon nanotube diode. (a) Photocurrent spectroscopy on the carbon nanotube p-n junction, showing the E11 optical resonance at 0.84 eV with the polarization aligned along the nanotube axis. The inset shows the polarization dependence of the photocurrent at the E11 transition. (b) I-V characteristics recorded without illumination (black), with 0.84 eV excitation (red arrow in (a)) and 0.98 eV excitation (blue arrow in (a)). The incident laser power was 64 μW. We clearly see the increased photoinduced current in the device at the E11 energy (red) compared to off resonance (blue).
Internal power conversion efficiency on and off resonance as a function of excitation power. The internal power conversion efficiency was extracted as a function of excitation power for 0.84 eV (red circles) and 0.98 eV (blue triangles). A clear enhancement in efficiency can be observed on the E11 as well as a peak efficiency around 6 μW. The decrease for large optical power is due to the fact that the junction becomes saturated and not all of the generated electrons and holes have time to escape the junction before recombining.
Power dependence of the internal power conversion efficiency on the E22 at 1.36 eV (orange circles), compared to off resonance excitation at 0.98 eV (blue triangles). An increased internal power conversion efficiency of the diode can be observed also on the E22.
Article metrics loading...
Full text loading...