The hydrogenation of soybean oil consists of highly complicated consecutive and simultaneous steps and always includes isomerization reactions.
Therefore, the control of these reactions is not so easy and the factors such as hydrogenation rate, selectivity, and the controlled trans-isomers formation must be considered.
On the other hand, the above-mentioned three factors are dictated by the reaction parameters such as type of catalyst, quality of feeds, and such reaction conditions as temperature, pressure, agitation rate, and catalyst concentration.
In this study, first, we presented the basic data for a guide to the selection of a set of conditions to give each desired products by investigating the effects of the reaction conditions on reaction rate, selectivity, trans-isomer formation under consistent catalyst and feeds.
It was found that the typical hydrogenation conditions in our bench scale experiments were as follows ;
Reaction Temperature : 180℃
Reaction pressure : 15 Psig
Agitation Rate : 1200 Rpm
Catalyst Concentration : 0.05 Wt % Ni
Secondly, we also studied the feasibility of introducing the "Gulftronic Separator" in separating the Ni-catalyst particles from the hydrogenated oil (Black oil) at the end of hydrogenation process. From the results of tests using a laboratory scale single module cylinderical "Gulftronic Separator" (4 inches O.D., 13.5 inches high), we could conclude that the separator is much more efficient than the conventional filtration system in terms of control, operation, and separation performance --- the catalyst contents was reduced from 0.05 wt % to below 1 ppm at a flow rate of 50 cc/min and 30 KV voltage. Then, the plant was scaled up for an 18,000 t/yr processing plant, based on the typical conditions obtained, and its economic feasibility study against the conventional filtration process was done. A total incremental investment was estimated to be $506,760 with annual savings of $128,940, which corresponds to 25.4% annual return on incremental investment and appears to be economically acceptable.
