The rational usage of energy resources and the reduction of environmental pollution become more important every day. There is a claim for the production of better quality fuels and the increase of the efficiency of automotive vehicles. Considering these two aspects, the use of diesel fuel is advantageous when compared to gasoline, because diesel engines allow greater compression ratios, resulting on better thermodynamic efficiencies. It is then possible to consume less fuel, which leads to reduction of CO2 emissions and consequently to reduction of global warming effects.
Regarding the specifications of diesel fuels, large reduction of sulfur contents was imposed in the past. Nowadays, the allowed sulfur content keeps diminishing, while additional restrictions, which include the aromatic content, are being defined.
The hydrodearomatization (HDA) process is an alternative to meet the recent diesel fuels specifications. In general, HDA studies are performed on bench-scale differential plug flow reactors, using mixtures of key aromatic compounds as models to analyze reaction rates. In this work, however, actual refinery products are used to build a mathematical pseudo-phenomenological kinetic model, which is employed to simulate and optimize the HDA process. The elaborated methodology allows the prediction of mono-, di- and polyaromatics concentration on products as functions of the feedstock composition and of the reactor operation conditions.
Experimental Six real industrial feedstocks obtained from distinct oil sources and refinery processes were used in this work: kerosene, light gas oil (LGO) and heavy gas oil (HGO) from atmospheric distillation; light and middle gas oil (LGO/MGO) from delayed coking; light vacuum gas oil (LVGO) from vacuum distillation; and light cycle oil (LCO) from fluid catalytic cracking. Commercial Ni-Mo-S/ Al2O3 catalysts were used for HDA tests. Catalysts were sulphided in-situ on the pilot plant, by passing a stream of hydrogenated spindle doped with CS2 (20000 ppmw) for 8 h. The reactor was heated by six independently controlled electric furnaces, which guaranteed the isothermal operation over the whole catalyst bed. The reactor was loaded with 310 cm3 of catalysts and an up-flow arrangement was used to minimize back-mixing problems. The liquid product was collected into a refrigerated recipient for analysis. The gas product was analyzed in-line with a wet gas meter (WGM) and a process chromatographer. A total of 79 tests were performed in a completely automated pilot plant. Replicates and blank tests were carried out to guarantee that catalyst remained active. 22 tests were used for independent validation of the model and were not used for pa
