The advent of robotic syntheses assisted by artificial intelligence opens up countless perspectives for the discovery of new nanomaterials, while raising the question of correctly validating these approaches. The goal of this thesis is to discover new nanostructured oxides to make CO2 capture and sequestration energetically efficient. This will require to 1) confirm or disprove that the automated preparation method (mixing robot coupled with a characterization platform by X-ray diffusion and gas analysis) is an approach representative of standard preparation methods, or if the automation is a new preparative approach independent of standard methods, and 2) confirm or disprove that the exploration of the vast space of parameters (nature of oxides, nanostructuring agents, injection laws) makes it possible to exceed the performances of the best current materials.