Modelagem matemática e simulação da reforma a seco do metano catalisada por Ni/Si-MCM-41

Abstract

Methane dry reforming process (RSM) emerges as a promising technology for generating clean and renewable energy, using greenhouse gases (CH4 and CO2) to create a synthesis gas predominantly composed of H2 and CO. This method requires high temperatures (873-1473 K), making the use of catalysts essential to optimize energy efficiency. In the literature, (Ni)-based catalysts stand out for their high catalytic activity, economic accessibility and natural occurrence. Expansion to larger reactors and significant scale production of hydrogen require analysis of reaction kinetics and the application of mathematical modeling. Recent research highlights the performance of the Ni/Si-MCM-41 catalyst, achieving conversions greater than 98% for CH4 and CO2 at 1023 K. In this context, the present study investigated six rate law models, covering irreversible and reversible power approaches, in addition to the Langmuir-Hinshelwood (LH) and Eley-Rideal (ER) models. To achieve this objective, RSM reaction experiments were performed in a fixed bed reactor. The study covered a temperature range of 873-973 K, volumetric flow rates of the mixture varying from 300 to 600 ml/min, maintaining feed composition (CH4/CO2, 1:1) and constant pressure (1 bar). Among the models evaluated, the reversible LH model demonstrated the best performance, evidenced by coefficients of determination greater than 0.9. A pseudo-homogeneous mathematical model was developed to describe the RSM in fixed-bed reactors. The simulations carried out with the pseudo-homogeneous model and the estimated parameters for kinetics showed that the model describes well the dry reforming of methane, especially when coupled to the reverse water-gas-shift reaction. In addition to these results, DRMSimulator was developed, a software implemented in Java programming language, focused on the pseudo-homogeneous model. Using this tool, it was possible to simulate methane conversion under experimental conditions, achieving coefficients of determination greater than 0.9 for the power models.