晶圆级石墨烯的红外光谱

We report on spectroscopy results from the mid- to far-infrared on wafer scale graphene, grown either epitaxially on silicon carbide, or by chemical vapor  deposition. The free carrier absorption (Drude peak) is simultaneously obtained with the  universal optical conductivity (due to interband transitions), and the wavelength at which  Pauli blocking occurs due to band filling. From these the graphene layer number, doping  level, sheet resistivity, carrier mobility, and scattering rate can be inferred. The mid-IR  absorption of epitaxial two-layer graphene shows a less pronounced peak at 0.37 ±0.02 eV  compared to that in exfoliated bilayer graphene. In heavily chemically-doped single layer  graphene, a record high transmission reduction due to free carriers approaching 40% at  250 μm (40 cm-1) is measured in this atomically thin material, supporting the great  potential of graphene in far-infrared and terahertz optoelectronics.

As pristine graphene is gapless and a tunable, moderate band gap can be engineered using  symmetry breaking schemes, it exhibits particularly strong potential in far-IR and  terahertz optoelectronics.15 Characterization of wafer-scale graphene in the IR and  terahertz range is a crucial step in the development of graphene optoelectronic devices.  Previously, broadband optical absorption measurements of few and multilayer epitaxial  graphene down to the terahertz range provided useful information of layer number as  well as doping levels.Large area graphene samples produced by chemical vaporphase deposition (CVD) were also studied.The free carrier response of back-gated  CVD graphene devices in the far-IR was carefully examined by Horng et al. 18 for carrier  densities below 71012  cm -2. A reduction of Drude weight was observed.

Finally, a few comments should be made to the carrier scattering rate . The two  epitaxial samples have similar scattering rate of ~270 cm-1, which corresponds to a  scattering time of 20 femto-seconds. The CVD graphene has smaller scattering rate of  ~100 cm-1 (scattering time of ~50 femto-seconds) and it doesn’t increase with the  presence of chemical absorbates from the chemical doping. The difference for the  scattering rate between epitaxial and CVD graphene can be due to the different sample  quality, the surface morphology, and the doping concentration. The scattering times  measured here are consistent with transport measurements.

Fig2

The red curve in far-IR is the fitting based on Drude model using parameters presented in  the main text. Less noise in the far-IR spectrum than that in Fig. 2 is due to the higher  light source power used. Inset: the enlarged view of the spectrum in mid-IR region with a sketch for the interband transitions. Red curve is the fitting based on Pauli blocking in  mid-IR. The blue line is the universal interband absorption for single layer graphene on  quartz. 2EF is indicated by the grey arrow. The Dirac cone of graphene is also shown in  the inset. Photons with energy smaller than 2EF can not be absorbed by graphene due to  Pauli blocking.

To conclude, we demonstrate that infrared spectroscopy is an excellent technique to  characterize wafer-scale graphene in a non-invasive manner. We show that through the  analysis of the absorption spectra in mid- and far-IR ranges simultaneously, layer number,  doping, sheet resistivity, carrier mobility, and scattering rate can be inferred. For  chemically doped CVD graphene, the reduction of transmission in far-IR in this  atomically thin film approaches 40%, demonstrating the great potential of this novel two  dimensional material in far-IR and terahertz applications, such as terahertz detectors and  imagers.