Supercritical Hydrogenation of Vegetable Oils

Doktorsavhandling, 2001

Vegetable oils were hydrogenated in a supercritical single-phase mixture with propane and hydrogen. The reaction was carried out in a small (0.5 ml) fixed-bed reactor and the influence of different process parameters was investigated. In this system both the oil and the hydrogen are dissolved in near-critical or supercritical propane, which changes the reaction mixture around the catalyst particles from two-phase (gas/liquid) into single-phase. Hence the hydrogen transport to the catalyst surface is greatly improved. The investigated oils were palm oil, rapeseed oil, and rapeseed fatty acid methyl esters (FAME). They were hydrogenated at low concentrations (1.0 - 2.4 wt.%), using different palladium catalysts. Due to the high reaction rates obtained, very short residence times were required (100 - 660 ms for FAME, 400 - 2000 ms for triglycerides, TG). In the FAME experiments the formation of trans fatty acids could be substantially reduced. At an iodine value (IV) of 70 a trans content as low as 4 % was achieved, but at the cost of selectivity. The catalyst used in the palm and rapeseed studies was more selective, but formed high amounts of trans, which is likely a consequence of its highly microporous structure. Evidence for transport limitation of the substrate was found as an unexpected result. It proved that the hydrogen concentration at the catalyst surface had indeed been raised to such a high level, that it was sufficient at all times, while the substrate molecules had to compete for adsorption sites. Selectivity and product composition were then only depending on the catalyst and the degree of hydrogenation, not on the other parameters like temperature and hydrogen concentration. This made the catalyst the most crucial factor for the selectivity, whereas the temperature mainly affected the reaction rate. Unfortunately, the catalyst deactivated rapidly in the triglyceride experiments. There are indications that the used catalyst suffered from substrate accumulation in the small pores, which radically reduced its active surface. It remains to prove whether single-phase conditions and sufficient hydrogen transport can be maintained at higher oil concentrations, which are required to make the process economically feasible. Also, the feed-related deactivation of the catalyst requires further investigation.