Estimation of Kinetic Parameters and Simulation of Methylacetylene and Propadiene Liquid-Phase Selective Hydrogenation Reactor Considering the Catalyst Deactivation

This paper presents a kinetic study of methylacetylene (MA) and propadiene (PD) liquid-phase selective hydrogenation considering catalyst deactivation. For this purpose, more than 20 000 plant operational data points were collected from two different fixed-bed reactors in an olefin plant during a 14-month run-step. Four different kinetic models were used parametrically in a dynamic reactor simulation. The genetic algorithm (GA) technique was used to estimate the kinetic constant by minimization of the mean relative error (MRE) between the simulation results and plant data. The effects of Tinlet, WHSV, H-2/MAPD, and catalyst deactivation on the reactor performance, propylene (PR) selectivity, PR yield, and other undesired products such as benzene (BZ) and toluene (TL) have been investigated. The average MRE of the reactor temperature, species molar flow, and reactor pressure drop is 3.46, 5.9, and 4.86%, respectively, indicating that the Langmuir-Hinshelwood-Hougen-Watson (LH-HW) type model with hydrogen dissociation fit the experimental data with acceptable accuracy.

Automated summary

This study conducted a kinetic investigation of liquid-phase selective hydrogenation of methylacetylene and propadiene, considering catalyst deactivation. By collecting extensive plant operational data, utilizing genetic algorithm to estimate kinetic constants, and using a dynamic reactor simulation, the study explored the impact of various factors on reactor performance and product yields. The experimental data was effectively fitted with a Langmuir-Hinshelwood-Hougen-Watson type model with hydrogen dissociation.