† Corresponding author. E-mail:
We investigate the interactions between two symmetric monovacancy defects in graphene grown on Ru (0001) after silicon intercalation by combining first-principles calculations with scanning tunneling microscopy (STM). First-principles calculations based on free-standing graphene show that the interaction is weak and no scattering pattern is observed when the two vacancies are located in the same sublattice of graphene, no matter how close they are, except that they are next to each other. For the two vacancies in different sublattices of graphene, the interaction strongly influences the scattering and new patterns’ emerge, which are determined by the distance between two vacancies. Further experiments on silicon intercalated graphene epitaxially grown on Ru (0001) shows that the experiment results are consistent with the simulated STM images based on free-standing graphene, suggesting that a single layer of silicon is good enough to decouple the strong interaction between graphene and the Ru (0001) substrate.
Graphene, carbon in the form of two-dimensional hexagonal lattice, which exhibits amazing optical, electrical, thermal and mechanical properties,[1–5] has become one of the most inspiring research topics since it was discovered in 2004. Because of its high mobility and near-ballistic transport at room temperature,[6] graphene has potential applications in the nanoelectronics. However, these super excellent electronic and transport properties are affected by various defects.[7–10] For example, vacancies, adatoms, Stone-Wales defects, substitutional impurities or topological defects are inevitably formed during the growth of graphene.[11,12] The defects commonly present in graphene are a limiting factor for electronic transport and device performance through charged impurities[13] or resonant scatters.[14] To improve device performance and utilize the full potential of graphene, it is crucial to identify these defects, particularly how they interact with each other and how they affect the electronic properties of graphene.
One of the most common defects in graphene is atomic vacancy, which is expected to be of fundamental importance regarding the electron transport properties of graphene-based devices. Atomic vacancies lead to sharp electronic resonances at the Fermi energy, which significantly limit the mobility of carriers in graphene and can be associated with the formation of local magnetic moments.[15] Recently, the π magnetism of a single carbon vacancy in graphene has been confirmed by using a scanning tunneling microscope.[16] It is known that both the symmetric monovacancy defect and its reconstructed configuration (asymmetric monovacancy) have been found in a graphene system, where the asymmetric configuration is the most stable.[17] The interaction between the asymmetric monovacancy defects in graphene has been investigated by first-principles calculations.[18] However, there is no report about the interaction between the symmetric monovacancy defects in graphene.
In this paper, by combining first-principles calculations with scanning tunneling microscopy (STM), we investigate the characterization of symmetric monovacancy defects in graphene and their interactions. First-principles calculation results show that when two symmetric monovacancy defects are in the same sublattice of graphene, scattering of each vacancy keep intact at arbitrary distances except when they are neighboring. When two vacancies are in different sublattices, they present new and complex quantum interference patterns. Our experiments on the silicon intercalated graphene on Ru (0001) prove that the interference between two monovacancy defects appears only if they are in the different sublattices.
All our calculations were performed within density functional theory, as implemented in the Vienna ab-initio Simulation Package (VASP) with the projector augmented wave (PAW) method.[19,20] Local density approximation (LDA) in the form of Perdew–Zunger was adopted for the exchange-correlation functional.[21] The energy cutoff of the plane-wave basis sets was 300 eV. To confirm the independence of the point defect, a 20 × 20 supercell of graphene and an 8 Å vacuum layer were used. In our calculations, all the carbon atoms were fixed in the same plane, and they were fully relaxed in geometric optimizations until the residual forces were smaller than 0.02 eV/Å. Due to the calculation limitation, a Gamma point K-sampling was employed to investigate the Brillouin zone matrix. The experimental data were acquired with an ultrahigh-vacuum (UHV) system with a base pressure of 1×10−10 mbar (1 bar = 105 Pa), which is equipped with an Omicron STM, low energy electron diffraction (LEED), silicon evaporators, and an electron beam heater.[22,23] The Ru (0001) surface has been prepared by argon-ion sputtering and annealing to the 1100 K, and was exposed to oxygen at 1500 K to remove the residual carbon. Monolayer graphene was grown on the surface of Ru (0001) by thermal decomposition of ethylene at 1100 K. Silicon atoms was evaporated to the graphene surface and then annealed at 800 K, producing an intercalated Si layer between the graphene and Ru.[24]
In Fig.
With these three kinds of configurations, we further change the distance between these two monovacancy defects, and simulate the corresponding STM images (corresponding to the local density of states), where the results are shown in Fig.
To confirm our theoretical characterization of vacancies and their interactions, we perform the corresponding experimental characterization of these kinds of vacancies in graphene. Monolayer graphene is grown on the surface of Ru (0001) and then Si layer is intercalated between the graphene and Ru. The intercalated silicon layer has proved to be silicene.[25] This silicene intercalated graphene is decoupled from its substrate as demonstrated by the ARPES measurements, which indicates that the silicene-intercalated graphene possesses the same linear dispersion as that of the free-standing graphene sheet.[24] In our STM results, most regions of the sample exhibit the perfect honeycomb lattice of monolayer graphene, and we could only find a few monovacancy defects. The STM images of the G/Si/Ru containing local defects are shown in Figs.
We also explore the interactions between two monovacancy defects which are in different sublattices by STM. The STM images of these local defects with different sublattices in G/Si/Ru sample are shown in Figs.
In this work, the interactions between two symmetric monovacancy defects in graphene are investigated based on first-principles calculations combined with scanning tunneling microscopy (STM). When the two monovacancy defects are located in the same sublattice of graphene, there are no new states exhibiting. But when the two monovacancy defects are in different sublattices of graphene, the interactions strongly influence the scattering and new patterns exist, which are determined by the relative position and distance between the two monovacancy defects. Our calculation results accord well with the experimental observations on silicene intercalated graphene epitaxially grown on Ru (0001).
This work was done in Key Laboratory of Vacuum Physics, Chinese Academy of Sciences.
[1] | |
[2] | |
[3] | |
[4] | |
[5] | |
[6] | |
[7] | |
[8] | |
[9] | |
[10] | |
[11] | |
[12] | |
[13] | |
[14] | |
[15] | |
[16] | |
[17] | |
[18] | |
[19] | |
[20] | |
[21] | |
[22] | |
[23] | |
[24] | |
[25] | |
[26] | |
[27] | |
[28] |