Student contribution: Plasmonic systems are efficient in converting optical energy into heat hence show technological significance in solar thermophotovoltaics, nanoparticle manipulation, and photocatalysis, etc. Conventional techniques to characterize plasmonic heaters are mostly thermal camera- and thermographic phosphor (TGP)- based. In this work, we present our results of characterizing a plasmonic heater using thermoreflectance imaging (TRI). The TRI technique presented here outperforms thermal camera-based technique in spatial resolution due to the visible light utilized for illumination, and does not require special sample preparation as in TGP-based technique. We chose to use a gap plasmon structure to maximize the optical absorption, and fabricated structures with various dimensions that exhibit varying optical absorptions at a fixed wavelength of 825 nm, which is the wavelength of pump light used in the TRI measurement. The TRI setup uses a millisecond-modulated continuous-wave pump laser to induce local temperature fluctuation on the sample surface, a 530 nm LED probe light then senses the change in the temperature-dependent material reflectance between high and low temperatures, which combined with a pre-calibrated thermoreflectance coefficient can be used to calculate the temperature rise on each image pixel. This technique grants us a resolution of ~200 nm. The experimentally obtained temperature rise on various gap plasmon structures correlates well with their optical absorption, and we compare the results against a finite element heat transfer model. Using a separate pump-probe thermoreflectance technique, we experimentally obtain the heat transfer dynamics of such gap plasmon structure under laser irradiation with picosecond resolution.
Graphene has been demonstrated to be a promising photodetection material because of its atomic-thin nature, broadband and uniform optical absorption, etc. Photovoltaic and photothermoelectric, which are considered to be the main contributors to photo current/voltage generation in graphene, enable photodetection through driving electrons via built-in electric field and thermoelectric power, respectively. Graphene photovoltaic/photothermoelectric detectors are ideal for ultrafast photodetection applications due to the high carrier mobilities in graphene and ultrashort time the electrons need to give away heat. Despite all the advantages for graphene photovoltaic/photothermoelectric detectors, the sensitivity in such detectors is relatively low, owing to the low optical absorption in the single atomic layer. In the past, our research group has used delicately designed snowflake-like fractal metasurface to realize broadband photovoltage enhancement in the visible spectral range, on SiO2 thin film backed by Si substrates. We have also demonstrated that the enhancement from the proposed fractal metasurface is insensitive to the polarization of the incident light. In this current work, we have carried out experiments of the same fractal metasurface on transparent SiO2 substrates, and obtained higher enhancement factor on the fractal metasurface than that achieved on SiO2/Si substrates. Moreover, the device allows more than 70% of the incident light to transmit during the detection, enabling photodetection in the optical path without any significant distortion. Another possibility to make use of the large portion of transmitted light is to stack multiple such devices along the optical path to linearly scale up the sensitivity.
Graphene has been demonstrated to be a promising photo-detection material because of its ultra-broadband absorption, compatibility with CMOS technology, and dynamic tunability. There are multiple known photo-detection mechanisms in graphene, among which the photovoltaic effect has the fastest response time thus is the prioritized candidate for ultrafast photodetector. There have been numerous efforts to enhance the intrinsically low sensitivity in graphene photovoltaic detectors using metallic plasmonic structures, but such plasmonic enhancements are mostly narrowband and polarization dependent. In this work, we propose a gold Cayley-tree fractal metasurface design that has a multi-band resonance, to realize broadband and polarization-insensitive plasmonic enhancement in graphene photovoltaic detectors. When illuminated with visible light, the fractal metasurface exhibits multiple hotspots at the metal-graphene interface, where the electric field of the incident electromagnetic wave is enhanced and contributes to generating excess electron-hole pairs in graphene. The large metal-graphene interface length in the fractal metasurface also helps to harvest at a higher efficiency the electron-hole pairs by built-in electric field due to metal-graphene potential gradient. To demonstrate the concept, we carried out experiment using Ar-Kr CW laser, an optical chopper, and lock-in amplifier. We obtained experimentally an almost constant ten-fold enhancement of photocurrent generated on the fractal metasurface compared to that generated on the plain gold-graphene edge, at all tested wavelengths (488 nm, 514 nm, 568 nm, and 647 nm). We also observed an unchanged photoresponse with respect to incident light polarization angles, which is a result of the highly symmetric geometry of the fractal metasurface.