HDS kinetics experiments were conducted in a pilot plant trickle-bed hydrotreater packed with a commercial NiMo/Al2O3 hydrotreating catalyst and a commercial sample of fluid catalytic cracking light cycle oil (LCO). Speciation of dibenzothiophenic compounds in the feed and hydrotreated effluents was conducted using a GC-AED method developed in-house. It was found that at high temperature (above 385 degreesC) the hydrogenation/dehydrogenation equilibrium effect became significant and the HDS kinetics of each individual sulphur compound was no longer of pseudo-first order. Therefore, a two-route HDS kinetics model was used to model the hydrogenation/dehydrogenation equilibrium effect by assuming first-order reactions with respect to sulphur, and zero-order with respect to hydrogen for the four reactions involved. The kinetics parameters for each reaction were determined for 14 dibenzothiophenic compounds based on experimental data. As expected, the activation energy for dehydrogenation was always higher than that for hydrogenation. The averaged difference between these two activation energies was about 43.5 kJ/mol. This difference explained the lower sensitivity of the HDS rate to temperature at higher temperatures. The kinetics parameters showed that the hydrogenolysis route contributed less than the hydrogenation route to the overall HDS, which is consistent with observations reported in other similar studies. (C) 2004 Elsevier B.V. All rights reserved.