Computational study of acetylene plasma polymerization using molecular dynamics

Plasma polymerization (glow discharge polymerization) refers to the formation of a polymeric thin film on a specific substrate under the influence of plasma. Although many experimental results have been reported on the plasma polymerization process and properties of plasma polymer films, the theoretical study of the plasma polymerization process could really help to understand the structure of the formed films and its dependence on various physical parameters. Computer simulation methods provide an opportunity to investigate the plasma polymerization process and the resulting polymer films at a fundamental level. In particular, molecular dynamics simulations make it possible to study the effect of a series of structural parameters. Therefore, the main goal of this thesis was to model the growth and characterize plasma polymerized acetylene (PPA) films, using computer simulations based on classical molecular dynamics (MD). The effect of parameters such as the substrate temperature and the nature of the substrate on the growth of the film were also carefully investigated. To do so, we first established an approach to model the acetylene plasma polymerization process. Our protocol involved the creation of a substrate and the deposition of the acetylene precursors (mixture of acetylene molecules and radicals) from a randomly arranged cell, by giving them initial kinetic energy toward the substrate. To mimic the cyclic or continuous flow of the gas toward the surface, successive depositions of the precursor cells were performed. We identified the most proper potential(s) to describe the interaction of hydrocarbons for the formation of PPA films using the MD and density functional theory (DFT) methods. In order to investigate the interactions, the potential energy was calculated as a function of the reacting atom interdistances in the radical and the molecule which were supposed to react and the results were compared to electronic structure calculations. The protocol was carried out with DFT and then repeated with four different MD potentials that cover hydrocarbons reactions. The REBO potential was found to give the most-realistic energy barrier for the first and second steps of polymerization and was therefore selected for the description of the H-H, H-C and C-C interactions during the formation of PPA films. Afterward, the effects of some adjustable parameters of the plasma polymerization process on the growth of PPA films, such as substrate temperature and nature of the substrate were investigated. To characterize the films a series of structural parameters such as total deposited mass, the creation of new bonds, coordination number of carbon atoms and types of species in the films were analyzed. To study the effect of substrate temperature, six PPA films were developed on Ag(111) at the substrate temperatures of 0, 100, 200 300, 400 and 500 Kelvin. Our results showed a high deposited mass at 0 K which decreases with a temperature rising to 500 K, while the of PPA films form on the substrate. The rates of the creation of new C-C and C-H bonds are higher when the substrate is still exposed than when it is covered with polymeric chains. The analysis of the coordination number of carbon atoms shows that the grafting of radicals and molecules on the growing sides of the chains increases with substrate temperature, which results in more carbon atoms with coordination numbers 3 and 4, and less with coordination number 2. Furthermore, the investigation of the number of species in the PPA films at each substrate temperature suggests that 300 K is adequate to develop a fully cross-linked structure. Finally, we investigated the effect of the substrate nature on the growth of PPA films. To do so, two PPA coatings were developed on nonreactive gold and reactive diamond substrates while the substrate temperature was kept at 300 K. During the initial stages of deposition, the PPA films grow in a 2D-like manner on gold, in contrast with a 3D growth on the diamond surface. The H atoms can easily penetrate the depth of the polymeric films, reach the surface of the diamond and react with it. This makes the H/C ratio higher in the films developed on the diamond in comparison with the ones on the gold at early stages of film formation. This difference levels off for the bulk of the films. The time distribution of bonding ratio indicates higher C-C and C-H bonding ratios at early stages of deposition (0.9-8 ns) for the films grown on the reactive diamond than those grown on the non-reactive gold substrate. However, the Diamond surfaces become H-passivated over time and after 18.9 ns, the values of C-C and C-H bonding ratios are similar for both substrates. This behavior indicates that the bonding ratios become independent from the nature of the substrate over time. The coordination number ratio was also calculated for the coatings grown on both substrates. Our results indicate that the nature of the substrate does not affect the coordination number ratios beyond the initial stages of precursors deposition. For all the considered cases, the characterization of the polymerization and cross-linking of the coatings demonstrates that their structures are different in the bulk regions than in the vicinity of the substrates.