On the Deactivation of Cobalt-based Fischer-Tropsch Catalysts

Abstract

The Fischer-Tropsch Synthesis (FTS) process is an attractive way to obtain synthetic liquid fuel from alternative energy sources such as natural gas, coal or biomass. However, the deactivation of the catalyst, consisting of cobalt nanoparticles supported on TiO2, currently hampers the industrial application of the process. Despite many years of research, we still lack the fundamental insights into the mechanism of catalyst deactivation necessary to develop the next generation of FTS catalysts. This lack of knowledge can at least partly be blamed on the lack of studies under reaction conditions. In this PhD Thesis, we have developed new techniques to study the deactivation of FTS catalysts. First, we developed a set-up for combined X-ray microscopy and spectroscopy under realistic reaction conditions (temperature of 250 °C and pressure of 10 bar). Applying these techniques to our FTS catalyst, we found that the catalyst is pretty much stable during the first 10 hours of FTS reaction. There was no sign of the traditional deactivation mechanisms such as bulk oxidation or metal-support compound formation, even on a nanometer scale. We also did not detect any changes in the morphology of the catalyst particle. Then, we decided to compare the fresh catalyst to the thoroughly deactivated catalyst. For this purpose, we developed a novel combination of three different combined microscopy and spectroscopy techniques, bridging four orders of magnitude in length scales – i.e., spanning the 50 μm - 0.5 nm range. Using this methodology we found interesting differences in the distribution of cobalt over the support that happen during FTS. First, on the microscale, cobalt nanoparticles redistribute from aggregates of nanoparticles to a more homogeneous distribution (without aggregates). It is remarkable that the nanoparticles move over such large distances (several micrometers), relative to their size (several nanometers). Second, on the nanoscale, we found that a thin layer (thickness of 1 – 2 nm) of cobalt is formed around the support particles. This layer is probably caused by atomic transport of cobalt. We also found that there is carbon present in the cobalt layer. Both findings suggest that there might be new catalyst deactivation mechanism taking place during FTS. Finally, a new set-up was developed for combined X-ray diffraction, Raman spectroscopy and gas chromatography under reaction conditions (elevated temperatures and pressures). A major improvement over previous comparable set-ups is that this one can be used with a laboratory X-ray source, instead of a synchrotron. This allowed us to study the FTS catalyst for an extended period (about 200 hours). We did not see any cobalt oxidation or sintering. However, the formation of cobalt carbide (Co2C) was clearly seen after about 150 hours of FTS reaction, but only during FTS at 10 bar. At atmospheric pressure, no cobalt carbide was detected. Furthermore carbonaceous deposits (in the form of coke) were found in the Raman spectra. We suspect that both cobalt carbide formation and coke formation play a role in the deactivation of the catalyst.