The possibility to engineer novel devices based on molecular nanostructures with a “bottom-up” approach relies on the clarification, at the atomistic level, of the processes responsible for the reaction of molecules and their assembly on template surfaces.

In our laboratory we deploy a combined experimental and theoretical strategy with the aim of designing novel technically accessible systems with desired properties.

Within this strategy, computer simulations performed on powerful clusters and supercomputers can efficiently explain the experiments, improve models and algorithms, and suggest new systems to be realized and measured.

Our research is based on the combination of different theory levels such as density functional theory (DFT), semi-empirical and classical approaches, applied to atomistic models representative of a real system. We run efficient parallel codes (http://cp2k.berlios.de) on our local cluster ipazia and at the Swiss supercomputing center (CSCS), exploiting up to 2048 cpus at the same time.

The results obtained in synergy with the experimental groups allow to better define the field of application of our computational strategy.

It is in particular very promising what we have obtained in the field of graphene nanoribbons and nanographene and derived structures.

Our goal is to achieve a deep understanding of the bottom-up production process of graphene-like networks by using theoretical simulation concepts in connection with the experiment. Our fellow experimentalists are innovative in the field of new methods for the production of graphene-like networks on surfaces. Therefore, the direct connection to the experiment is granted to provide the success of this comparative study.

The process of cyclodehydrogenation is fundamental in the self-assembly formation of those structures.

Molecular networks are grown on metallic surfaces, and the simulation of the complex system adsorbate+substrate requires the most advanced computational resources and algorithms to be efficiently performed.

The substrate shows often a selectivity towards adsorption and a catalytic effect for the chemical reactions among different species leading to the formation of the extended networks.

Given a set of precursor molecules suggested by experimentalists our calculations are focused on

1. Computing stable adsorption sites of the molecules on different metal substrates (Au, Cu and Ag).
2. Characterizing the reaction processes necessary to create a network from the adsorbed molecules (dehalogenation/dehydrogenation, surface diffusion, molecule-molecule reactions.
3. Describing the electronic properties of the nano-networks.
4. Electronic properties of graphene nanoribbons
5. Compute band gaps as a function of doping and ribbon size
6. Interplay between doping and ribbon size