Magnetic Nature of Intrinsic Carbon Defects

Magnetism is a phenomenon that has been known for a very long time. Iron, cobalt, and nickel are known ferromagnetic materials. It is less known, probably because it is so unexpected, that even carbon can have ferromagnetic behaviour. Experimentally this has been confirmed on many occasions within the last decade. Ferromagnetic behaviour of carbon provides an example of the fact that magnetism is not well understood at the atomic scale. One of the aims of this thesis is to study and understand possible sources of ferromagnetism in carbon systems, thereby creating a possible foundation for the next generation of ferromagnets. Carbon itself is a very interesting substance with numerous interesting properties. In the late 1980s and early 1990s new carbon allotropes were found, such as fullerenes (bucky balls) and nanotubes (cylinders), next to the old ones (graphite and diamond). Especially nanotubes have been considered as candidates for several future applications. Whatever the fabrication process, all allotropes of carbon will have intrinsic defects, and in this thesis the role of these defects in carbon magnetism is investigated in detail. Studying magnetism requires “state-of-the-art”-methods due to the demand of high accuracy because energy differences between non-magnetic and magnetic cases are usually very small. Ab initio methods are usually the best for such studies, especially methods based on the density functional theory. Here, a state-of-art method which is based on the density functional theory and implementing projector augmented waves to model the properties of carbon is used. Adatoms and vacancies are found to have magnetic moments of 0.5 μB and 1.0 μB, respectively. In practice, however, the high mobility of adatoms on graphene at room temperature would suggest that many of them recombine with vacancies or cluster together, destroying their magnetism. Despite the indications that a barrier to vacancy-interstitial pair recombination exists, efficient recombination seems to be confirmed by He-irradiation experiments. The magnetic signal was small despite the fact that the amount of defects created by the He ions is large. Also, the effect of the changing electronic structure on the magnetic moments of adatoms and vacancies is studied with the help of nanotubes. On nanotubes, the magnetism of an adatom decreases because of the curvature and differences in electronic structures while the magnetic moment of a vacancy in all but strongly metallic tubes is destroyed. The experimental demonstration of induced ferromagnetism by proton irradiation on graphite indicates a promising direction for creating a magnetic carbon system in a controllable way. Simulations indicate that this is due to a combination of a hydrogen atom trapping at vacancies and pinning of mobile adatoms, producing magnetic C-H complexes and uncompensated vacancies.