Abstract of the thesis

	The laboratory evaluation of FCC catalysts requires especially designed units due to the high deactivation rate of the catalyst. This work involved the construction, optimization and application to FCC catalyst testing of a new laboratory unit called MicroDowner, which uses a transported bed and short catalyst residence time. Its principal application is FCC catalyst testing, but it can also be used for other processes that require short residence time like selective oxidation of alkanes.
	First, the influence of operating conditions on product yields was studied and a standard protocol was established. Then, the yields given by our new unit were compared with those given by MAT unit and a DCR pilot plant. It was shown that the results given by the MicroDowner were very similar to those of the DCR pilot plant, and reproduced the well documented differences between pilot plant and MAT unit, especially referring to olefins and coke yields.
	The Microdowner unit was used to evaluate the potential of naphtha cracking in the FCC unit, with the production of short olefins as main objective. While LSR naphtha only cracks appreciably under severe reaction conditions (high temperature and high space velocity), FCC naphtha was encountered much more reactive even at moderate temperature (450ºC and coked catalyst). The light olefins yield increases with the temperature while selectivity decreases, and is boosted by the addition of ZSM5 zeolite. In all the processing schemes considered, a high reduction of olefins in gasoline fraction was achieved.
	The study of the activity of a coked catalyst showed that coke has only a small influence on product yields, being mainly an increase of products olefinicity. Nevertheless, working in partial regeneration mode in order to generate coked catalyst can have beneficial effects for removing NOx in regenerator flue gas.
	Finally, a kinetic model has been developed for the Microdowner unit. It has been shown that coke-on-catalyst accounts for less than a half of the deactivation at short catalyst residence time, while the other part of deactivation can be represented by strongly adsorbed components, which desorbs during stripping. The use of a refractory fraction generated from gasoil feed during the cracking reaction allowed the representation of gasoil conversion data with first order kinetics compared with the traditional second order kinetics.