Graphène est un matériau bidimensionnel unique physique, chimique et les propriétés mécaniques. Il pourrait être prometteur pour de nouvelles applications, mais le contrôle nm échelle de traitement de graphène défis la technologie actuelle, en particulier dans le traitement du plasma, empêchant ainsi le développement de la technologie à base de graphène à l'échelle industrielle. / Graphene is a two-dimensional material with unique physical, chemical and mechanical properties. It could be promising for novel applications, but the nm-scale control of graphene processing challenges current technology, especially in plasma treatment, thus preventing the development of graphene based technology at industrial scale. The main issue associated with plasma/graphene processes is the atomic thickness of the material: graphene is easily damaged upon exposure to reactive plasma. One critical question to answer then: is it possible to use conventional plasma technologies to pattern/clean/dope graphene layers, as is done for other materials in the microelectronic industry?Hydrogen plasmas have been shown to be promising for graphene treatment with minimal damages, but little is known about the fundamental mechanisms involved in graphene etching. Thus, in our work, we applied classical molecular dynamics (MD) simulations of H2 plasma/graphene interaction to assist the development of three important processes. First, MD allowed us to explain the lateral etching mechanisms of graphene nanorribons (GNR) in downstream H2 plasmas, which is an important technological step to produce GNR with a width<10 nm. Second, we show that H2 plasmas can be used to clean polymeric residues from the graphene surface (selective removal of PMMA/photo-resist residues or atmospheric contaminant from its surface). Modeling results combined with experimental work shows very promising results in this application, which is demanded by the entire graphene community. Third, MD simulations were also used to assist the development of multilayer graphene processing by Atomic Layer Etching. Although irreversible damages of graphene are observed when the ion bombarding energy is in the 5-50 eV range, MD predicts a very interesting phenomenon at 20-25eV range: the implantation of hydrogen atoms and subsequent formation of H2 gas sandwiched between first two layers. This causes a pressure rise, which leads to a lift-off of the entire top graphene layer. This result from modeling suggests that H2 plasmas can be used to etch graphene layer by layer in a controlled way through an entirely new mechanism. However, in order to avoid damages of underneath layers during the processing, additional investigations should be provided.In conclusion, several novel and unexpected results were obtained during the present PhD study and MD simulations have proven to be a powerful tool to assist plasma process development. Indeed, based on this fundamental research work an ANR project was launched to develop cleaning, doping and etching processes of graphene in the ICP reactors available in the LTM laboratory, Grenoble, France. MD calculation developed during this PhD will therefore continue to be used to assist further the development of innovative processes.The current PhD project was held in LTM etching group Grenoble, France under supervision of Gilles Cunge and Emilie Despiau-Pujo in the framework of the Chair of Excellence 2010 of Prof. David Graves and financial support of Nanoscience Foundation. We would like to acknowledge collaboration with several groups from Institute Neel (Vincent Bouchiat, Laurence Magaud and Johann Coraux) and our colleagues from CEA-Grenoble, France (Okuno Hanako).