Modeling, simulation and kinetic parameter estimation for diesel hydrotreating. (1st December 2017)

• Record Type: Journal Article
• Authors: Novaes, Leandro da Rocha; de Resende, Neuman Solange; Salim, Vera Maria Martins; Secchi, Argimiro Resende
• Is Part Of: Fuel. Volume 209(2017)
• Page Start: 184
• Page End: 193
• Publication Date: 2017-12-01
• Subjects: Hydrotreating -- Modeling -- Parameter estimation -- Experimental planning -- Internal diffusion; Fuel -- Periodicals; Coal -- Periodicals; Coal; Fuel; Periodicals; 662.6
• DOI: 10.1016/j.fuel.2017.07.092
• Languages: English
• ISSNs: 0016-2361

Abstract

Highlights: Kinetic parameters for HDS and HDN reactions are accurately estimated from experimental data. Deviations between experimental data and model predictions are below 4% for both reactions. Statistical results show the good quality of the estimated kinetic parameters. Intraparticle mass transfer limitation is present in HDS/HDN reactions in commercial catalyst. Abstract: The modeling and simulation of hydrotreating (HDT) reactors can provide relevant information about optimal operating conditions, reactor design and formulation of catalysts to achieve cleaner fuels. This study developed a phenomenological model to estimate the kinetic parameters for the HDT process using the design of experiments technique. The catalytic tests were carried out in a bench-scale unit with diesel as feedstock under similar conditions to those of the industrial units and commercial nickel-molybdenum on alumina (NiMo/γ-Al2 O3 ) as catalyst. The reactor temperature was varied from 330 to 380 °C and liquid hourly space velocity (LHSV) from 0.5 to 1.5 h −1, keeping constant hydrogen to oil ratio (H2 /oil) and pressure at 800 std m 3 /m 3 and 90 bar, respectively. The unknown parameters are estimated from experimental data by minimization of the unweighted least-squares function, using a non-linear optimization method (complex algorithm). The proposed model was able to reproduce different operating conditions with good adjustment and accuracy. Relative deviations below 4% for the experimental data and model prediction were achieved. Statistical parameters, p values, and confidence intervals confirmed the good quality of the estimated kinetic parameters. Internal diffusion resistances were present during hydrodesulfurization (HDS) and hydrodenitrogenation (HDN) reactions. The hydrogen consumption for HDS was higher than for HDN reaction. Finally, the proposed experimental design with 12 catalytic tests was satisfactory to obtain the desired results in a reliable way.