Limits...
Multiscale investigation of graphene layers on 6H-SiC(000-1).

Tiberj A, Huntzinger JR, Camassel J, Hiebel F, Mahmood A, Mallet P, Naud C, Veuillen JY - Nanoscale Res Lett (2011)

Bottom Line: At the same scale, electron diffraction reveals a significant rotational disorder between the first graphene layer and the SiC surface, although well-defined preferred orientations exist.This is confirmed at the nanometer scale by scanning tunneling microscopy (STM).The most striking result is that the FLGs experience a strong compressive stress that is seldom observed for graphene grown on the C face of SiC substrates.

View Article: PubMed Central - HTML - PubMed

Affiliation: Groupe d'Etude des Semiconducteurs, UMR5650 CNRS-Université Montpellier II, cc074, Place Eugène Bataillon, 34095 Montpellier Cedex 5, France. Antoine.Tiberj@univ-montp2.fr.

ABSTRACT
In this article, a multiscale investigation of few graphene layers grown on 6H-SiC(000-1) under ultrahigh vacuum (UHV) conditions is presented. At 100-μm scale, the authors show that the UHV growth yields few layer graphene (FLG) with an average thickness given by Auger spectroscopy between 1 and 2 graphene planes. At the same scale, electron diffraction reveals a significant rotational disorder between the first graphene layer and the SiC surface, although well-defined preferred orientations exist. This is confirmed at the nanometer scale by scanning tunneling microscopy (STM). Finally, STM (at the nm scale) and Raman spectroscopy (at the μm scale) show that the FLG stacking is turbostratic, and that the domain size of the crystallites ranges from 10 to 100 nm. The most striking result is that the FLGs experience a strong compressive stress that is seldom observed for graphene grown on the C face of SiC substrates.

No MeSH data available.


Related in: MedlinePlus

STM images of the edges of the graphene islands. (a) 23 × 23 nm2 STM image of a monolayer on top of the (3 × 3) SiC-reconstructed surface. This image was recorded at a high tunnel voltage (-2.5 V) with a current of 0.2 nA. With such high voltage, only the (3 × 3) SiC surface can be imaged on the bare SiC and through the monolayer. One can still distinguish that the edges of the monolayer are brighter than the center. It corresponds to a height difference of 40 pm. (b) To get a more detailed image of the edges, a 6.6 × 4.6 nm2 STM image was recorded at low tunnel voltage (+10 mV) with a current of 0.1 nA. In these experimental conditions, the graphene lattice can be probed with an atomic resolution. The graphene edge is folded and bent towards the SiC surface. The height of the fold is 50 pm. This particular shape is consistent with the compressive stress deduced from the Raman spectra. This compressive stress is occured by the thermal stress because of the cooling down of the sample after the growth and because the graphene edges seem to be bound to the SiC surface.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3211224&req=5

Figure 5: STM images of the edges of the graphene islands. (a) 23 × 23 nm2 STM image of a monolayer on top of the (3 × 3) SiC-reconstructed surface. This image was recorded at a high tunnel voltage (-2.5 V) with a current of 0.2 nA. With such high voltage, only the (3 × 3) SiC surface can be imaged on the bare SiC and through the monolayer. One can still distinguish that the edges of the monolayer are brighter than the center. It corresponds to a height difference of 40 pm. (b) To get a more detailed image of the edges, a 6.6 × 4.6 nm2 STM image was recorded at low tunnel voltage (+10 mV) with a current of 0.1 nA. In these experimental conditions, the graphene lattice can be probed with an atomic resolution. The graphene edge is folded and bent towards the SiC surface. The height of the fold is 50 pm. This particular shape is consistent with the compressive stress deduced from the Raman spectra. This compressive stress is occured by the thermal stress because of the cooling down of the sample after the growth and because the graphene edges seem to be bound to the SiC surface.

Mentions: Using these relationships, it can be estimated from the G band that the strain ranges from -0.2 to -0.7% with an average of -0.5%. The corresponding stress values range from -3 to -8.7 GPa with an average of -6 GPa. From the 2D band, a strain comprised between -0.3 and -0.5% with an average of -0.4% is found. The corresponding stress ranges from -3.7 up to -6 GPa with an average of -5 GPa. The strain/stress derived from the 2D band is slightly smaller than the one deduced from the G band. It can arise from the uncertainty on the graphene/graphite Grüneisen parameters [3] and maybe from a small FLG doping that can induce a small shift of both bands. Nevertheless, it can be concluded from this high up shift of the D, G, and 2D bands that FLG are subjected to a high compressive strain (stress) with an average from around -0.4 to -0.5% (-5 to -6 GPa). The most probable origin is the thermal stress due to the cooling down of the sample after the growth. This compressive stress might be considered as being in contradiction with the weak interaction between graphene layers and the underlying substrate that was shown by LEED and STM experiments. The most likely hypothesis is that graphene crystallite edges are bound to the SiC surface. This bonding would induce this strong thermal stress, but let graphene layers free to grow with the wide distribution of disorientation angles revealed by STM and LEED results. To check this hypothesis, STM images of the graphene edges were recorded and are shown in Figure 5. Unfortunately, the SiC surface and the graphene lattice cannot be imaged simultaneously since graphene is only clearly resolved at low bias within the gap of the SiC(3 × 3) surface [18]. In Figure 5a, the SiC surface is probed at high tunnel voltage (-2.5 V), and the bare (3 × 3) SiC surface can be distinguished from the (3 × 3) SiC surface with a graphene monolayer on top. It can also be stated that the edges of the graphene island are higher than the core by 40 pm. At lower tunnel voltage (+10 mV), a detailed image of the graphene edges was recorded with atomic resolution on the graphene lattice. On this image, it can be clearly observed that the graphene sheet is folded and that its edges are bent towards the SiC surface. The fold is 50 pm higher than the island core at this low bias, which is similar to the high bias value and thus should reflect the topography of the graphene edge. This step height is much smaller than the wrinkles that are usually seen on graphene grown on the C face. However, the fold and the bending can indicate that the graphene edges are bound to the SiC substrate. This edge bonding can also explain the thermal compressive stress of these graphene crystallites after cooling down the sample. Usually on the C face, the elastic energy associated with the stress is often relaxed by the formation of wrinkles. It is not the case here because of the small graphene crystallite sizes. The authors observed indeed only one wrinkle on Figure 2d. The compressive stress is not relaxed because the elastic energy to be relaxed is proportional to the domain area and is not high enough to compensate the energy cost of the wrinkle formation.


Multiscale investigation of graphene layers on 6H-SiC(000-1).

Tiberj A, Huntzinger JR, Camassel J, Hiebel F, Mahmood A, Mallet P, Naud C, Veuillen JY - Nanoscale Res Lett (2011)

STM images of the edges of the graphene islands. (a) 23 × 23 nm2 STM image of a monolayer on top of the (3 × 3) SiC-reconstructed surface. This image was recorded at a high tunnel voltage (-2.5 V) with a current of 0.2 nA. With such high voltage, only the (3 × 3) SiC surface can be imaged on the bare SiC and through the monolayer. One can still distinguish that the edges of the monolayer are brighter than the center. It corresponds to a height difference of 40 pm. (b) To get a more detailed image of the edges, a 6.6 × 4.6 nm2 STM image was recorded at low tunnel voltage (+10 mV) with a current of 0.1 nA. In these experimental conditions, the graphene lattice can be probed with an atomic resolution. The graphene edge is folded and bent towards the SiC surface. The height of the fold is 50 pm. This particular shape is consistent with the compressive stress deduced from the Raman spectra. This compressive stress is occured by the thermal stress because of the cooling down of the sample after the growth and because the graphene edges seem to be bound to the SiC surface.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3211224&req=5

Figure 5: STM images of the edges of the graphene islands. (a) 23 × 23 nm2 STM image of a monolayer on top of the (3 × 3) SiC-reconstructed surface. This image was recorded at a high tunnel voltage (-2.5 V) with a current of 0.2 nA. With such high voltage, only the (3 × 3) SiC surface can be imaged on the bare SiC and through the monolayer. One can still distinguish that the edges of the monolayer are brighter than the center. It corresponds to a height difference of 40 pm. (b) To get a more detailed image of the edges, a 6.6 × 4.6 nm2 STM image was recorded at low tunnel voltage (+10 mV) with a current of 0.1 nA. In these experimental conditions, the graphene lattice can be probed with an atomic resolution. The graphene edge is folded and bent towards the SiC surface. The height of the fold is 50 pm. This particular shape is consistent with the compressive stress deduced from the Raman spectra. This compressive stress is occured by the thermal stress because of the cooling down of the sample after the growth and because the graphene edges seem to be bound to the SiC surface.
Mentions: Using these relationships, it can be estimated from the G band that the strain ranges from -0.2 to -0.7% with an average of -0.5%. The corresponding stress values range from -3 to -8.7 GPa with an average of -6 GPa. From the 2D band, a strain comprised between -0.3 and -0.5% with an average of -0.4% is found. The corresponding stress ranges from -3.7 up to -6 GPa with an average of -5 GPa. The strain/stress derived from the 2D band is slightly smaller than the one deduced from the G band. It can arise from the uncertainty on the graphene/graphite Grüneisen parameters [3] and maybe from a small FLG doping that can induce a small shift of both bands. Nevertheless, it can be concluded from this high up shift of the D, G, and 2D bands that FLG are subjected to a high compressive strain (stress) with an average from around -0.4 to -0.5% (-5 to -6 GPa). The most probable origin is the thermal stress due to the cooling down of the sample after the growth. This compressive stress might be considered as being in contradiction with the weak interaction between graphene layers and the underlying substrate that was shown by LEED and STM experiments. The most likely hypothesis is that graphene crystallite edges are bound to the SiC surface. This bonding would induce this strong thermal stress, but let graphene layers free to grow with the wide distribution of disorientation angles revealed by STM and LEED results. To check this hypothesis, STM images of the graphene edges were recorded and are shown in Figure 5. Unfortunately, the SiC surface and the graphene lattice cannot be imaged simultaneously since graphene is only clearly resolved at low bias within the gap of the SiC(3 × 3) surface [18]. In Figure 5a, the SiC surface is probed at high tunnel voltage (-2.5 V), and the bare (3 × 3) SiC surface can be distinguished from the (3 × 3) SiC surface with a graphene monolayer on top. It can also be stated that the edges of the graphene island are higher than the core by 40 pm. At lower tunnel voltage (+10 mV), a detailed image of the graphene edges was recorded with atomic resolution on the graphene lattice. On this image, it can be clearly observed that the graphene sheet is folded and that its edges are bent towards the SiC surface. The fold is 50 pm higher than the island core at this low bias, which is similar to the high bias value and thus should reflect the topography of the graphene edge. This step height is much smaller than the wrinkles that are usually seen on graphene grown on the C face. However, the fold and the bending can indicate that the graphene edges are bound to the SiC substrate. This edge bonding can also explain the thermal compressive stress of these graphene crystallites after cooling down the sample. Usually on the C face, the elastic energy associated with the stress is often relaxed by the formation of wrinkles. It is not the case here because of the small graphene crystallite sizes. The authors observed indeed only one wrinkle on Figure 2d. The compressive stress is not relaxed because the elastic energy to be relaxed is proportional to the domain area and is not high enough to compensate the energy cost of the wrinkle formation.

Bottom Line: At the same scale, electron diffraction reveals a significant rotational disorder between the first graphene layer and the SiC surface, although well-defined preferred orientations exist.This is confirmed at the nanometer scale by scanning tunneling microscopy (STM).The most striking result is that the FLGs experience a strong compressive stress that is seldom observed for graphene grown on the C face of SiC substrates.

View Article: PubMed Central - HTML - PubMed

Affiliation: Groupe d'Etude des Semiconducteurs, UMR5650 CNRS-Université Montpellier II, cc074, Place Eugène Bataillon, 34095 Montpellier Cedex 5, France. Antoine.Tiberj@univ-montp2.fr.

ABSTRACT
In this article, a multiscale investigation of few graphene layers grown on 6H-SiC(000-1) under ultrahigh vacuum (UHV) conditions is presented. At 100-μm scale, the authors show that the UHV growth yields few layer graphene (FLG) with an average thickness given by Auger spectroscopy between 1 and 2 graphene planes. At the same scale, electron diffraction reveals a significant rotational disorder between the first graphene layer and the SiC surface, although well-defined preferred orientations exist. This is confirmed at the nanometer scale by scanning tunneling microscopy (STM). Finally, STM (at the nm scale) and Raman spectroscopy (at the μm scale) show that the FLG stacking is turbostratic, and that the domain size of the crystallites ranges from 10 to 100 nm. The most striking result is that the FLGs experience a strong compressive stress that is seldom observed for graphene grown on the C face of SiC substrates.

No MeSH data available.


Related in: MedlinePlus