Polyethylene production from the polymerization of ethylene is an industrial process of great importance. Ethylene stream for the polymerization of polyethylene is produced by the steam cracking of a wide range of hydrocarbon feedstock and usually contains acetylene impurities (1%) which poison the polymerization catalyst. The ethylene steam has to be purified by the selective semi-hydrogenation of acetylene which requires a catalyst with high selectivity to hydrogenate acetylene to ethylene. The intermetallic compound Al13Fe4 was introduced in 2012 by Armbuster et al. as an active and selective catalyst for the semi-hydrogenation of acetylene for polyethylene production. It has a crystal structure with high average inter-atomic distances Fe-Fe and a low coordination number of iron atoms, which falls under the concept of "site isolation principle". This compound is also attractive because of its low cost (without any noble metals compared to Pd/Al2O3 industrial catalysts) and low toxicity. However, it has been produced in the form of unsupported powder by the Czochralski method which limits its use in catalytic engineering. In this context, supporting the catalyst presents many advantages as the ease of separation of the heterogeneous catalyst from the reaction mixture. In contrast to homogeneous catalysts in which separation is often costly and difficult, separating the supported heterogeneous catalyst can be achieved by a variety of methods such as filtration for example. Another advantage of supported catalysts is the higher surface area of the catalyst. Since catalysis is a surface reaction, consequently, maximizing the surface area of a catalyst by distributing it over the support will enhance/optimize the catalytic activity.Chemical synthetic routes such as Metal Organic Chemical Vapor Deposition (MOCVD) and Metal Organic Deposition (MOD) referred as “Chimie douce” process are reputed to be flexible and economically competitive methods to prepare nanoparticles or thin films. Our work is thus aimed at developing Al13Fe4 as supported films or nanoparticles by MOCVD and/or MOD. The first step to meet our objective is the development of compatible molecular precursors of metallic aluminum and iron followed by MOCVD or MOD of those precursors to form the intermetallic compound in the good stoichiometry. Among the numerous aluminum MOCVD precursors used in the literature, dimethyl ethylamine alane (DMEAA, [AlH3(NMe2Et)]) is used due to its properties such as high vapor pressure and low deposition temperatures. Moreover, the absence of Al-O and Al-C bonds leads to the production of carbon and oxygen free films. However, iron molecular precursors for the MOCVD of pure iron films are scarce and less developed. Apart from iron pentacarbonyl that produces pure iron films, amidinates and guanidinates are used as iron precursors. However, oxygen and carbides impurities are present in high percentages. Thus the main objective of this Ph-D work is to design and synthesize novel and original iron molecular complexes that serve as precursors for the low temperature MOCVD of iron films. In this Ph-D work, nanoparticles of the intermetallic complex were prepared via solution reduction of novel Fe triazenide precursors and Al metal. Supported films were also prepared via sequential MOCVD by using DMEAA and Fe(CO)5 as molecular precursors. Its catalytic properties have been explored and showed that it is very little active in the hydrogenation reaction of acetylene. Regeneration under hydrogen or oxygen was not very successful and only some activity restored. The catalytic tests have been further extended to Al13Fe4 powder prepared by solution reduction as well as to commercial Al13Fe4 and found that Al13Fe4 was inactive catalytically in all forms (in our conditions of reactions)