Résumé : Every year millions of tons of lignin, a complex biopolymer present in plants that naturally contains aromatic subunits, are produced as a by-product of industries like the food sector and paper sector. Until now it has been considered as a waste but proper valorisation through optimized depolymerization techniques would allow to recover of high-added value fine chemicals and bulk commodities. Vanadium(V) triphenolamine complexes (VO-TPA), developed and studied in the group of Giulia Licini (University of Padova, IT), are of interest in this context as they are efficient catalysts for the oxidative cleavage of carbon-carbon bonds typical of those present in lignin. However, the mechanism of this reaction had not yet been elucidated and the catalysts have exclusively been used in organic solvents. This Thesis is a contribution to the further development of these catalysts and takes different Green Chemistry principles into account such as waste prevention, atom economy, renewable feedstocks, catalysis, energetic efficiency and the use of benign solvents. A first part of this thesis is a contribution to the elucidation of the reaction mechanism. The work was undertaken with vicinal diols, which are the simplest model compounds of lignin. Based on a review of the literature, radical trapping experiments, kinetics, 13C KIE experiments, Hammett plots and multi-parametric correlations, the C-C cleavage of a non-oxo or oxo chelate has been identified as the rate determining step. Attempts were made to synthesise the chelate and identify it via MS. Moreover, three hypothetic pathways were posited to reach this intermediate. Isotopic labelling experiments, determination of reaction activation parameters via an Eyring plot and computations of the different pathways along EPR elucidation of reduced species, have been used to discriminate the operative mechanism. The second part of this thesis is devoted to the transfer of the reaction into water, which can be considered a benign solvent. The VO-TPA complexes were successfully transferred to water using both non-ionic TPGS-750-M and zwitterionic DPC micelles. The latter maintained the catalyst stability over a broad pH range and temperature range which was not the case with the non-ionic micelles. The substrate scope was evaluated with the DPC micelles and a strong selectivity was observed for hydrophobic substrates. After scaling up the process to reach standards used in the literature, mixed DPC/TPGS micelles were characterized and used as a simple solution to combine the pH stability of the former and extractive properties of the latter. This allowed to completely recycle the micellar phase while maintaining good yields.