We report on room temperature detection of terahertz radiation by means of antenna-coupled field effect transistors (FETs) fabricated using epitaxial graphene grown on silicon carbide. The achieved photoresponsivity (∼0.25 V/W) and noise equivalent power (∼80 nW/$\sqrt{\text{Hz}}$ ) result from the combined effect of two independent detection mechanisms: over-damped plasma wave rectification and thermoelectric effects, the latter ascribed to the presence of carrier density junctions along the FET channel. The calculated plasmonic and thermoelectric response reproduces qualitatively well the measured photovoltages; the experimentally observed sign-switch demonstrates the stronger contribution of plasmonic detection compared to the thermoelectric one. These results unveil the potential of plasmonic detectors exploiting epitaxial graphene on silicon carbide for fast large area imaging of macroscopic samples.

One of the major issues in graphene-based optoelectronics is to scale-up high-performing devices. In this work, we report an original approach for the fabrication of efficient optoelectronic devices from scalable tungsten disulfide (WS₂)/graphene heterostructures. Our approach allows for the patterned growth of WS₂ on graphene and facilitates the realization of ohmic contacts. Photodetectors fabricated with WS₂ on epitaxial graphene on silicon carbide (SiC) present, when illuminated with red light, a maximum responsivity R ∼220 A W⁻¹, a detectivity D* ∼2.0 × 10⁹ Jones and a −3 dB bandwidth of 250 Hz. The retrieved detectivity is 3 orders of magnitude higher than that obtained with graphene-only devices at the same wavelength. For shorter illumination wavelengths we observe a persistent photocurrent with a nearly complete charge retention, which originates from deep trap levels in the SiC substrate. This work ultimately demonstrates that WS₂/graphene optoelectronic devices with promising performances can be obtained in a scalable manner. Furthermore, by combining wavelength-selective memory, enhanced responsivity and fast detection, this system is of interest for the implementation of 2d-based data storage devices.