Using high-temperature annealing conditions with a graphite cap covering the C-face of, both, on axis and 8 degrees off axis 4H-SiC samples, large and homogeneous single epitaxial graphene layers have been grown. Raman spectroscopy shows evidence of the almost free-standing character of these monolayer graphene sheets, which was confirmed by magneto-transport measurements. On the best samples, we find a moderate p-type doping, a high-carrier mobility and resolve the half-integer quantum Hall effect typical of high-quality graphene samples. A rough estimation of the density of states is given from temperature measurements.

We review the results of growing few layer graphene on alpha-SiC with different surface orientations. To this end we have used successively a pure {000-1} C face, a 8 degrees-off one, a nonpolar {11-20} surface and, finally, a 8 degrees-off Si face and a pure (exactly {0001}-oriented) Si one. Without any need for complex hydrogen annealing, we have shown that a pure C or a 8 degrees-off C or even a pure {11-20} surface allows the growth of single layer epitaxial graphene sheets that are strain-free, p-type, low doped and exhibit reasonably high carrier mobility. This is not the case on pure Si or 8 degrees-off Si faces on which strained material is constantly found with heavy n-type doping and low carrier mobility.

Micro-Raman and micro-transmission imaging experiments have been done on epitaxial graphene grown on the C-and Si-faces of on-axis 6H-SiC substrates. On the C-face it is shown that the SiC sublimation process results in the growth of long and isolated graphene ribbons (up to 600 mu m) that are strain-relaxed and lightly p-type doped. In this case, combining the results of micro-Raman spectroscopy with micro-transmission measurements, we were able to ascertain that uniform monolayer ribbons were grown and found also Bernal stacked and misoriented bilayer ribbons. On the Si-face, the situation is completely different. A full graphene coverage of the SiC surface is achieved but anisotropic growth still occurs, because of the step-bunched SiC surface reconstruction. While in the middle of reconstructed terraces thin graphene stacks (up to 5 layers) are grown, thicker graphene stripes appear at step edges. In both the cases, the strong interaction between the graphene layers and the underlying SiC substrate induces a high compressive thermal strain and n-type doping.

Graphene has emerged recently as a new material with outstanding electronic properties1. This includes mass-less Dirac fermions, ballistic transport properties at room temperature and good compatibility with silicon planar technology2. Different techniques have been developed in the last six years to fabricate mono or bi-layer graphene. They range from exfoliated graphite, either mechanically1 or in a liquid-phase solution3 to chemical vapor deposition on a metal surface4, and, more recently, to substrate-free synthesis when passing ethanol into an argon plasma5. The method investigated in this work consists in a controlled sublimation of few atomic layers of Si from a single crystal SiC substrate6. Such epitaxial growth of graphene (EG) seems to be the most suitable option for industrial applications but, for easy control, it necessitates either a large and homogeneous sheet of monolayer graphene (MLG) or few layers graphene (FLG) covering the full wafer surface. Basically, on both the Si and C faces of any SiC substrate, graphene grows selectively on some reconstructed parts of the surface. Controlling the growth means then controlling locally the surface reconstruction. At low pressure conditions (below 10-6 Torr), it remains challenging to grow FLG with homogeneous domains larger than few hundred nanometers on both faces7. The homogeneity can be increased by lowering the sublimation rate. It has been demonstrated on the Si face by working at high pressure under a noble gas atmosphere such as argon8,9. In this work10, the surface reconstruction of the C face during the Si sublimation is modified by covering the SiC substrate with a graphite cap. It leads to a strongly step-bunched morphology with on few selected terraces the growth of long anisotropic graphene ribbons (5 μm wide and up to 600 μm long). Since the Raman fingerprint of Bernal stacked FLG depends strongly of the number of graphene layers11 and the absorbtance of FLG is almost independent of the wavelength and proportional to the number of graphene layers12, we combine micro-Raman spectroscopy with micro-transmission measurements to study the quality and thickness uniformity of these ribbons. We find that most of these ribbons are homogeneous monolayers or bilayers of graphene and that the thermal stress between the graphene layer and the 6H-SiC substrate is relaxed by the formation of wrinkles. This combination of techniques is especially useful to discriminate without any ambiguity between a monolayer graphene and a misoriented bilayer because their Raman fingerprint are identical. The spectra and extinction coefficient of a monolayer, a Bernal stacked bilayer noted AB, and a misoriented bilayer noted AA' are shown in Figure 1.

We have investigated graphene layers grown on 6H-SiC(000-1) under Ultra High Vacuum and under controlled atmosphere (600mb AR pressure, 2000°C). The UHV growth gives a thin system with an averaging thickness given by Auger spectroscopy between one and two graphene planes. The growth under AR pressure shows multilayer samples with an average thickness larger than 5 graphene planes. Already from the first layer, a significant rotational disorder is revealed by electron diffraction (LEED), although well defined preferred orientations exist. For the multilayer samples, STM studies show the presence of “ponds” (with lateral size <100 nm) separated by “ridges” of height 0.7 nm. Combined micro-differential transmission and micro-Raman spectroscopy confirm these results. The analysis of the so-called D, G and 2D bands give a good estimation of the thickness, the quality of the films and the stacking of the multilayer samples. Transport measurements have been performed using Van der Pauw design of 2.8 mm by 2.8 mm and hall bar structure of 1μm by 6μm. The mobility of the multilayer samples (around 0.8 m2V-1s-1) is larger than the one of the almost monolayer but seems to reveal a relative electronic coupling between the different graphene planes. At low temperature (4k), the low field magneto resistance is consistent with the weak localisation correction.

The unique versatility of micro-Raman spectroscopy (μRS) in semiconductor physics remains in Raman imaging. Numerous applications cover the whole development of modern electronic and optoelectronic devices: from semiconductor growth to advanced device inspection tools. Raman mapping enables first to check the crystalline quality, the composition, the doping and the uniformity of the grown semiconductor. One on the most popular application in microelectronics is strain measurements either at the device or at the full wafer scale. It can also be used for final device inspection through the temperature mapping of operating devices (FETs, lasers, actuators). In this talk, we will review these application through a wide variety of semiconductors (SiC, GaN, SiGe, strained Si, sSOI, SGOI) and will focus on recent results obtained on graphene as it is a promising candidate for future carbon based electronics.

Epitaxial graphene was grown on the surface of on-axis and off-axis SiC (0001) by solid state graphitization at high temperatures (2000 degrees C) in Ar ambient. The effect of the miscut angle on the lateral uniformity of the few layers of graphene (FLG) was investigated by combined application of micro-Raman spectroscopy and Torsion Resonance Conductive Atomic Force Microscopy, the latter method enabling a quantification of the FLG coverage on SiC with submicrometer lateral resolution. While the on-axis samples result in uniform coverage by thin (similar to 3 monolayers) FLG, the coverage for off-axis samples is much less uniform, following closely the step bunching morphology of the SiC surface.