Nuclear magnetic resonance (NMR) parameters of graphene fragments evaluated computationally

03.11.2009

Recently developed experimental techniques have made it possible to produce two-dimensional sheets of graphene, a carbon nanostructure with many interesting characteristics. Graphene is composed of a single layer of carbon atoms arranged in a honeycomb lattice, and has been shown to have a gapless semiconductor band structure. Its properties include high current-carrying capacity and thermal conductivity, making it a potentially advantageous material for electronic devices.

First-principles electronic structure studies on nuclear magnetic resonance (NMR) properties of increasingly large planar hydrocarbons, which can be related to finite graphene fragments, have been carried out by collaboration between the Molecular Magnetism and NMR Research groups at the Universities of Helsinki and Oulu, respectively, and CSC. NMR provides a powerful tool for characterization of materials. NMR studies of graphene have not yet been conducted experimentally. This study showed that the isotropic and anisotropic chemical shifts and one-, two- and three-bond spin-spin coupling tensors converge rather quickly as the number of carbon atoms in the molecule is increased, which allows extrapolation of the parameters to the large-system limit. Hence, these results constitute a plausible starting point of the analysis of eventual NMR experiments for graphene.

High predictive value of the data was ensured by calibrating the density-functional theoretical methods by correlated ab initio wave functions for small model systems, as well as by using the novel completeness-optimized basis-set paradigm developed earlier by two of the authors. With these novel basis sets, results close to the basis-set limit are attainable with relatively few basis functions: the saving by a factor of three in the dimension of the basis set means a reduction of the computational resources by two orders of magnitude. The calculations carried out in the study would not have been feasible using traditional basis sets, as the number of functions would have been too high for the larger molecules.

Large-scale density-functional linear response calculations using up to 128 parallel cores were carried out in the supercomputers of CSC and the local Linux cluster facilities in Helsinki and Oulu.

The work is part of the Ph.D. project of Suvi Ikäläinen under the supervision of Prof. Juha Vaara. The academic teams belong to the Finnish Center of Excellence in Computational Molecular Science (CMS) funded by the Academy of Finland.

More information

Suvi Ikäläinen, Perttu Lantto, Pekka Manninen and Juha Vaara: NMR tensors in planar hydrocarbons of increasing size, Physical Chemistry Chemical Physics (2009) [DOI:10.1039/B919860A]

CSC

CSC – IT Center for Science Ltd. is a non-profit limited company administered by the Ministry of Education. The center provides IT support and resources for academia and research institutes, including modeling, calculation and data services. Researchers have access to the largest selection of scientific software and scientific databases in Finland and, through Funet data connections, the most efficient supercomputers.