(Invited) The Electroluminescence of Graphene

Abstract

Electroluminescence is the phenomenon by which a material emits light in response to the passage of an electric current. In solids, it is the prerogative of semiconductors and related organic materials, and it results from the radiative recombination of electrons and holes. Today, electroluminescent devices represent the majority of lighting and display devices and are a building block of the global information and telecommunication network.We report on two observations signaling the electroluminescence of high-mobility graphene field-effect transistors (FET). The semi-metallic nature of graphene a priori forbids electroluminescence: Indeed, in semiconductors, the bandgap energetically insulates the electrons from the holes. Such protection doesn’t exist in graphene which is gapless. Nonetheless, electroluminescence is possible, because (i) of the remarkable inefficiency of the non-radiative carrier relaxation in graphene, and (ii) thanks to an original carrier injection mechanism specific to 2D semimetals: the Zener-Klein (ZK) tunnel conductance. [1]Our first observation consists in the thermometry of graphene’s electrons with applied bias, see figure 1a. Using electronic noise thermometry, we observe that, after an initial warm up with the deposited Joule power, the electron gas cools down, starting at a doping-dependent threshold. The turning point is concomitant with the onset of electron-hole injection by Zener-Klein (ZK) tunneling. The cooling mechanism involves the radiative emission, due to the electron-hole recombination, in electromagnetic near-field modes of the hBN-graphene heterostructure. It is reminiscent of far-field electroluminescent cooling that can be observed in regular LEDs [2] except that it’s much more efficient (~10 mW vs. ~10 pW, i.e., 9 orders of magnitude larger). [3]Our second observation consists in the detection of the middle-wave infrared radiation (MIR) emitted by a high-mobility graphene FET operating in ambient conditions, see figure 1b. We observe a sizeable optical emission in the range 80–250 meV, starting at the ZK tunneling threshold bias. The spectral analysis of the emitted light indicates that it originates from the scattering of the hBN-graphene heterostructure near-fields, as observed in reference [4] by surface near-field optical microscopy. Bibliography [1] W. Yang et al, A graphene Zener-Klein transistor cooled by a hyperbolic substrate. Nat. Nanotechnol. 2018, Vol. 47, 13.[2] P. Santhanam et al, Room temperature thermo-electric pumping in mid-infrared light-emitting diodes. Appl. Phys. Lett. 2013, Vol. 103, 18, p. 183513.[3] E. Baudin et al, Hyperbolic Phonon Polariton Electroluminescence as an Electronic Cooling Pathway. Adv. Funct.. Mater. 2020, Vol. 30, 8, p. 1904783.[4] S. Dai et al, « Graphene on hexagonal boron nitride as a tunable hyperbolic metamaterial ». Nat. Nanotech. 2015, Vol. 10, 8, p. 682. Figure 1