Search

fff; International audience; Polymeric carbon was deposited over Siralox supported cobalt catalysts by an ethylene treatment at 230 °C and 260 °C. Cobalt catalysts with cobalt particle sizes of 10 and 14 nm were studied. Temperature programmed hydrogenation and Raman spectroscopy analyses of the deposited species were found to be similar to those formed during Fischer-Tropsch synthesis. The cobalt catalysts were tested in high-throughput 16 parallel reactors at 20 bars and 220 °C, using a H2/CO ratio of 2.12. Data were collected at different space times to compare activities and selectivities at iso-conversion levels. Ethylene treated catalysts showed lower activity that corresponded roughly to the loss of CO adsorption sites. Additionally, a decrease of the heavy products selectivity, an increase in the methane selectivity and a decrease of the olefin to paraffin ratio was observed after the ethylene treatments. Unlike the loss in activity, the deselectivation level was found to depend on the amount of deposited carbon. This phenomenon can be explained by a steric effect of the deposit, but an electronic effect cannot be excluded, but is challenging to prove.

International audience; The emergence of high-throughput experimentation gives new opportunities for accurate and rapid data acquisition for a wide variety of chemical reactions in different fields of application such as hydrocracking, isomerization and syngas conversion. Fischer–Tropsch synthesis is a challenging reaction to carry out due to its exothermicity, its wide range of products including heavy waxes, catalyst sensitivity to deactivation and the large number of parameters influencing the kinetics. Despite the accuracy gain brought by the most recent high-throughput technologies, numerous parameters need to be controlled for accurate data acquisition and correct comparison of catalyst performances. The present study shows that the high-throughput test strategy plays a major role and can affect significantly the accuracy and coherence of the generated data. For instance, when catalyst performances are measured for kinetic purposes, a commonly used method to vary the contact time (Wcat/FCO) is by loading the reactors with different amounts of catalyst. However, this method is not suited for catalysts that are difficult to reduce, since it has been shown that hydrogen space velocity during catalyst reduction has an influence on the degree of the reduction of the catalyst. Still this method might be preferred over the method of varying the syngas flowrate for a constant catalyst weight. The latter is time consuming and introduces deviations because of catalyst deactivation with time on stream. This issue becomes particularly important when liquid composition analysis is required, since even with a miniaturized high-throughput setup, we observed significant transitory periods on the liquid composition, due to wax accumulation in post-reactor volumes or catalyst evolution such as deactivation or surface reconstruction. Deactivation could, however, be limited by operating at low conversion levels or by running successive short tests.

International audience; Fischer-Tropsch cobalt catalyst deselectivation was studied using accelerated aging treatments dedicated to the carburization phenomenon. Cobalt carbide was successfully obtained by CO treatments, with limited simultaneously carbon deposition nor sintering. Particle sizes were found to influence the catalyst sensitivity to carburization, likely because of a possible core–shell mechanism. An important loss of both activity and selectivity to heavy products was observed after the treatments, making carburization one of the potential mechanisms responsible for deselectivation in Fischer-Tropsch synthesis. Side products were also impacted by cobalt carbide formation, with an increase of CO2 production and a decrease of the selectivity to olefins. The extent of carburization was found to directly dictate the level of the selectivity shift.