Ambient-pressure X-ray photoelectron spectroscopy (APXPS) and high-pressure scanning tunneling microscopy (HPSTM) were used to study the structure and chemistry of model Cu(100) and Cu(111) catalyst surfaces in the adsorption and dissociation of CO2. It was found that the (100) face is more active in dissociating CO2 than the (111) face. Atomic oxygen formed after the dissociation of CO2 poisons the surface by blocking further adsorption of CO2. This "self-poisoning" mechanism explains the need to mix CO into the industrial feed for methanol production from CO2, as it scavenges the chemisorbed O. The HPSTM images show that the (100) surface breaks up into nanoclusters in the presence of CO2 at 20 Torr and above, producing active kink and step sites. If the surface is precovered with atomic oxygen, no such nanoclustering occurs.

The control of nanocrystal structures at will is still a challenge, despite the recent progress of colloidal synthetic procedures. It is common knowledge that even small modifications of the reaction parameters during synthesis can alter the characteristics of the resulting nano-objects. In this work we report an unexpected factor which determines the structure of cobalt nanoparticles. Nanocrystals of distinctly different sizes and shapes have resulted from stock solutions containing exactly the same concentrations of [Co{N(SiMe(3))(2)}(2)(thf)], hexadecylamine, and lauric acid. The reduction reaction itself has been performed under identical conditions. In an effort to explain these differences and to analyze the reaction components and any molecular intermediates, we have discovered that the rate at which the cobalt precursor is added to the ligand solution during the stock solution preparation at room temperature becomes determinant by triggering off a nonanticipated side reaction which consumes part of the lauric acid, the main stabilizing ligand, transforming it to a silyl ester. Thus, an innocent mixing, apparently not related to the main reaction which produces the nanoparticles, becomes the parameter which in fine defines nanocrystal characteristics. This side reaction affects in a similar way the morphology of iron nanoparticles prepared from an analogous iron precursor and the same long chain stabilizing ligands. Side reactions are potentially operational in a great number of systems yielding nanocrystals, despite the fact that they are very rarely mentioned in the literature.