First-Principle Investigation Of Adsorption And Spin-Polarization Of Graphene

Since the experimental discovery of graphene in 2004 by A. K. Geim and K. S. Novoselov, this one-atom-thick material has attracted considerable attention around the world. The appearance of this two-dimensional (2D) materials have made more and more researchers devote themselves to the investigation of graphene and its derived materials, not only because its successful synthesis broke the common perception that real 2D materials did not exist without a 3D base, but also because it has remarkable mechanical, thermal, optical, electrical and magnetic properties. Therefore, the Royal Swedish Academy of Sciences awards Nobel Prize in Physics 2010 to A. K. Geim and K. S. Novoselov for "groundbreaking experiments regarding the two-dimensional material graphene".Graphene is a single-layer membrane separated from graphite. As one of the carbon (C) allotropes, graphene constitute a huge carbon family with carbon nanotube and fullerene et al. It is also a basic building block for graphitic materials of all other dimensionalities. It can be wrapped up into OD fullerenes, rolled into 1D nanotubes or stacked into 3D graphite. Although graphene looks very simple and basic, it contains exceptional potential. Many studies have indicated that graphene has a Young’s modulus of 1 TPa, a thermal conductivity of 5000 W·m-1·K-1, an absorption of 2.3% of incident light over a broad wavelength range, an electron mobility of 200,000 cm2·V-1·s-1, and a long spin coherence length above 1 μm. Due to the rapid development of information technology, the traditional electronic devices based on silicon will come to a bottleneck soon. As graphene has so many remarkable properties, it can be regarded as a potential candidate in post-silicon era. However, zero band gap of pristine graphene limits its direct applications towards transistors and other semiconducting devices which rely highly on the semiconducting characteristics to achieve the transition between "ON" and "OFF" states. For these reasons, there are more and more studies on derived materials of graphene, and the works in this dissertation are based on two of them.1. Adsorption of GrapheneChemical adsorption of non-carbon atoms on graphene is a good way to open a band gap. When such atoms are adsorbed on a graphene surface, they form covalent bonds with the C atoms. These C atoms change their hybridization from sp2 to sp*, which dramatically alters the electronic structure of graphene. Furthermore, if the numbers of atoms adsorbed on the two honeycomb sublattices are different, the system will carry net magnetic moments. In recent years, adsorbing hydrogen (H) atoms on graphene surface has received the most consideration. On the one hand hydrogenation is the simplest adsorption, which is easily controlled and studied, whereas on the other hand hydrogenation is related to hydrogen storage. However, experimental investigations have shown that hydrogenation of graphene is unstable, which rapidly loses H atoms at moderate temperature. While theoretical works predict that hydrogenated graphene is very stable. The calculated binding energy appears to be rather similar to fluorinated graphene, which has been demonstrated to be stable in experiment, and some even call it a 2D counterpart of Teflon. The formation energy of hydrogenated graphene is lower than that of benzene, which means that it is more stable than benzene. In order to explain the disagreement between present theoretical and experimental investigations on the stability of hydrogenated graphene, we have systematically studied hydrogenated graphene with different configurations from the consideration of single-side and double-side adsorption using first-principles calculations. Both binding energy and formation energy are calculated to characterize the stability of the system. According to our definition, small value corresponds to stable system. The main results are as follows:(1) In experiments, graphene membranes are often set on a substrate, such as SiO2. In this situation, only one side of graphene is accessible for H adsorption. Considering different number of adatoms and positions, we systematically studied the stability of single-side hydrogenated graphene. For the binding energy, it is found that all the values have the similar order of magnitude as that of fluorine adsorption, but the formation energies of hydrogenated graphene are always positive, which is quite different from the negative value of fluorinated graphene. Compared with the binding of a C atom and an H atom, that of two H atoms is energetically favorable, owing to the lower energy of the system. Therefore, all the configuration of single-side hydrogenation of graphene is unstable.(2) In experiment, graphene membranes can also be set as free-standing, so both sides can adsorb. We obtained that double-side hydrogenation of graphene is more stable than single-side hydrogenation. As is well known, adsorption on graphene is a process of breaking π bonds and producing additional σ bonds, and transforming C-C sp2 hybridization to sp3. Comparing the geometric structures and electron distributions of the two kinds of adsorption, we can find that the C-C sp3 hybridization is enhanced by adding H atoms on the other side, leading to double-side hydrogenation more stable. For geometric structures the calculated C-C-C bond angles of double-side are closer to 109.50, and for electron distributions more electrons distribute between C and H atoms and become shared electrons. Both of them correspond to stronger C-H covalent bonds.(3) The comparison among all double-side hydrogenation shows that the stability is related to the numbers of H atoms adsorbed on each side of graphene. The formation energies change from positive to a negative value with the numbers of H atoms adsorbed on each side tending to the same. From our calculations, it is found that not all the configurations of double-side hydrogenated graphene are stable. It means that the stability of hydrogenated graphene with different configurations has essential differences. In theoretical studies, researchers usually focus on hydrogenated graphene with an equal number of hydrogen atoms adsorbed on each side, but in experimental studies, this ideal configuration is hard to realize due to the randomness of adsorption.2. Spin-Polarization of GrapheneCutting graphene into nanoribbons is another good way to open a band gap. Distinguished by the directions of the cutting edges, the fabricated graphene nanoribbons are generally classified into two basic kinds of structures, the zigzag graphene nanoribbons (ZGNR) and the armchair graphene nanoribbons (AGNR). Both kinds can generate a tunable band gap near the Fermi level with the variation of the ribbon width. As graphene nanoribbons retain the integrity and simplicity of the pure-carbon structure and have a tunable band gap, these derived materials of graphene have also attracted considerable attention. Further improvements of graphene nanoribbons based on spin-related devices have been proposed by various physical and chemical modification. Especially, due to the quantum confinement effect, edge states emerge in the ZGNR. These states located at the edge atoms with opposite net spin moments between different edges. Even though the band structure of the ZGNR is spin degenerated at ground state, the electron density is spatially spin-polarized, so breaking the symmetry distribution of spin density is regarded as a good way to realize spin-polarization in graphene. Studies in this field will enrich the application of graphene. Based on the ZGNR, we investigate how to realize spin-polarization in graphene using first-principles calculations. Our method is applying a vertical strain near one edge of the ribbon. The electronic structures of the ZGNR under deformations with different positions, depths and sizes are studied. The main results are as follows:(1) The applied vertical strain near one edge of the ZGNR indeed break the spin degenerated band structure, and the largest spin-splitting is obtained at the bottom of the conduction band. As the electron density of the ZGNR with opposite spin moments is spatially separated, when there is a local deformation at one edge of the ribbon, a localized state will appear, which corresponds to only one kind of spin, and affects the other little.(2) The spin-splitting value is dependent on the choice of the deformation position. For a deformation centered at different sublattices, it is found that the spin-splitting value has a very large difference. However, by changing the position of the local deformation from the edge to the center of the ribbon, no obvious rules can be got, which means that there is no direct relation between spin-splitting and the distance to the edge.(3) The influence of other parameters of the deformation is studied at the optimal position. We find that the spin-splitting value increases with the depth of the deformation. The optimal deformation size is related to the standard for describing the vertical deformation by depth or by force. For a given depth, the spin-splitting value decreases with the size of the deformation, but for a given force, the opposite is true.