mathematical and artificial neural network modeling and simulation of catalytic fixed bed reactor for producing ethyl benzene from ethanol

Ethyl benzene is the raw material of producing styrene monomer and is produced via benzene alkylation in presence of ethanol and ethylene. benzene alkylation process for production of ethylbenzene consists of three steps: alkylation step for the reaction of benzene with ethylene. transition step in which polyethyl-benzenes (mostly diethyl benzene and triethyl benzene) are converted to ethylbenzene in reverse-alkylation process in presence of benzene. separation step where non-reacted, polyethyl-benzenes and other compounds are separated and thylbenzene with high purity is obtained. the amount of ethylbenzene in crude oil is very low and it is also used for production of diethyl benzene, cellulose acetate, etc. as well as production of styrene. in this study, benzene alkylation in presence of ethanol in a catalytic fixed bed reactor in steady-state, unsteady-state, adiabatic and nonadiabatic conditions was modeled using fundamental laws and artificial neural network assuming one dimensional pseudo-homogeneous system. based on the results of this study, at steady-state condition, percent conversion of benzene to ethylbenzene was reduced by increasing the amount of feed flow.in both steady-state and unsteady-state conditions, concentrations of products increased, as inlet temperature increased. modeling of the system by artificial neural network showed that the best network with 7 neurons in hidden layer, estimated the output results of fundamental modeling with minimum error value of about 0.001.