Ethylene is a very important material in petrochemical industries, whose chief application is producing polymers. The steam cracking of naphtha or ethane is usually applied to produce ethylene. A small amount of acetylene is produced in this process. The amount of acetylene in the product stream should not exceed 1 ppm, because it is harmful to polymerization catalysts in downstream units. The acetylene hydrogenation unit is designed for acetylene removal in industrial plants. In this unit, the removal of acetylene up to 1 ppm in the product stream and ethylene’s selectivity are of great importance. In this paper, the dynamic optimization of acetylene hydrogenation reactors of Marun petrochemical complex with considering catalyst deactivation is presented. In this study, the differential evolution (DE) method is used as a powerful method for determination of a dynamic optimal temperature profile to achieve maximum of ethylene’s selectivity in a period of 720 operating days. Then, the optimal results are compared with the condition wherein the inlet temperatures of the reactors are maintained constant at 55 ˚C and with the condition wherein the inlet temperatures of them increase linearly from 55 to 90 ˚C. The results showed when the inlet temperatures are kept 55 ˚C, the outlet acetylene exceed 1 ppm, but the best selectivity is achieved. With a linear increase in the inlet temperatures, the outlet acetylene is below 1 ppm but the selectivity is decreased. An optimal temperature profile maximizes the selectivity when the outlet acetylene is below 1 ppm.

Catalyst deactivation is usually indispensable, although the rate at which it occurs varies greatly. At first, this article discusses the causes of deactivation in a commercial hydrocracking unit called Isomax. Then, a 5-lump kinetic model including catalyst decay function for hydrocracking of vacuum gas oil in a commercial plant is proposed. The model considers vacuum gas oil (VGO) having boiling point higher than 380oC (380+oC), diesel (260380oC), kerosene (150260oC), naphtha (IBP: 150oC), and gas as products. By using selective catalyst decay function in the kinetic model, the effect of the catalyst deactivation on the yield of products over time is studied. The prediction of the model during 1.5 years is in good agreement with the actual commercial data. The average absolute deviation (AAD%) of the model for the strategic products like naphtha, kerosene and diesel are about 1.784%, 1.983% and 1.971%, respectively. Also it is observed that the estimated parameters are consistent with the reported characteristics of amorphous catalysts.

Deactivation of cobalt catalyst supported on carbon nanotubes (CNTs) in Fischer-Tropsch synthesis (FTS) was investigated by considering different deactivation mechanisms. The changes in the cobalt catalyst morphology during FTS reaction were studied using different techniques. The experimental results show that the sintering of cobalt particles is the dominant mechanism in deactivation of Co/CNTs catalyst, because of no interaction between support and active phase. Thus, the growth of cobalt particle size has been evaluated using a novel model in the catalyst deactivation. In this novel method, a new thermodynamically size-dependent kinetic equation is used to investigate the crystal growth during the catalyst deactivation. The difference between experimental FTS results and calculated results using the new model is related to other deactivation mechanisms.

Styrene is an important industrial unsaturated aromatic monomer. Styrene monomer is used for production of polymers. The most important commercial process for production of styrene is based on catalytic dehydrogenation of ethyl benzene. The major problem in this process is the decrease of catalyst life time and catalyst deactivation. Catalyst deactivation occurs by several mechanisms. Coke formation is considered to be the most important factor in catalyst deactivation. Gasification reaction is used for catalyst regeneration. Modeling and determination of kinetic parameters and optimization of the process can improve the final conversion to styrene and postpone catalyst deactivation. From different coke formation modeling levels, deactivation at particle level was selected in this research. In this level, four rival models for the gasification reaction and four deactivation functions for coke formation in the particle level type are  developed. The 16 models obtained are the results of the coupling of these cases. Anova one-way was used for the analysis of models with SPSS software. The reference model is the model with the largest F. coefficient. Kinetic parameters are estimated from a set of experimental data. The reference model describes satisfactorily the experimental results and has the necessary and sufficient validity and reliability. The reference model was compared with the Froment mode1.The error was less than 0.1% in the range of experimental data. This model is easy to use and has acceptable accuracy. Catalyst coke content, rate of coke formation and gasification and net rate of coke formation are considered under various temperatures and hydrogen, steam and aromatics pressures, with respect to the run length and other variables. The results obtained in this work correspond satisfactorily with the published data.  

A kinetic model of a fixed bed tubular reactor incorporating catalyst deactivation was developed for the ISOMAX unit of Arak refinery. The kinetic parameters for the hydrocracking reactions over the commercial catalyst were determined using initial activity plant data i.e. when the catalyst is fresh. Catalyst deactivation was then taken into account by means of deactivation function based on plant data. The catalyst deactivation function is defined in terms of normalized time (BPP) of operation. Effect of catalyst deactivation on the product yield has been investigated. Steady state material and energy balances were then developed for an extended four lumped kinetic network. To determine the effect of reaction types on the rate, we calculate frequency factor for each individual bed with constant activation energy and heat of reaction. Furthermore, we calculate the frequency factor for individual beds, for the first one to estimate the rate of reactions in the different beds. The results show that the reactions in the first and second bed are faster than those in the 3rd and 4th beds. The comparison between model conversion and experimental conversion of the unit indicates that the model is capable of predicting product yield with an error of less than 5%.