Kinetics, catalysis and mechanism of methane steam reforming

• Main Author: Liu, James
• Other Authors: Ravindra Datta, Advisor
• Format: Others
• Published: Digital WPI 2007
• Subjects: Methane Steam Reforming; Kinetics; Catalysis
• Online Access: https://digitalcommons.wpi.edu/etd-theses/68; https://digitalcommons.wpi.edu/cgi/viewcontent.cgi?article=1067&context=etd-theses

The search for an alternative clean and renewable energy source has become an urgent matter. One such energy-saving technology is a fuel cell; it uses fuel as the source of energy to produce electricity directly and the byproducts formed are not as voluminous and environmentally harmful. The conventional low temperature fuel cells use hydrogen as the fuel which is produced from conventional fuels via reforming. However, developing reformers for hydrocarbon fuels requires AN understanding of the fundamental mechanisms and kinetics studies. In this study, simple hydrocarbon fuel, namely methane, in external reforming or internal reforming within a solid oxide fuel cell has been studied because of its importance and with the hope that it will ultimately lead to an understanding of reforming of higher hydrocarbons, such as logistic fuels like JP-8. For this purpose, methane was used the starting point and building block for the progressive understanding of reforming of complex hydrocarbons. Methane steam reforming (MSR), CH4 + 2H2O = CO2 + 4H2 is, in fact, the most common method of producing commercial bulk hydrogen along with the hydrogen used in ammonia plants. United States alone produces 9 million tons of hydrogen per year. The overall MSR reaction CH4 + 2H2O = CO2 + 4H2 is in fact composed of two reactions, the water gas shift reaction, CO + H2O = CO2 + H2, which has recently been investigated by a former Ph.D. student in our group, Caitlin Callaghan. Here, the first reaction CH4 + H2O = CO + 3H2, i.e., methane reforming, is analyzed using a reaction route network approach to obtain the overall methane steam reforming network and kinetics. Kinetics providing detailed information of elementary reaction steps for this system, namely micro-kinetics, has not yet been fully addressed. Employing the theory of Reaction Route Network Theory, recently developed by Fishtik and Datta, and using the Unity Bond Index-Quadratic Exponential Potential (UBI-QEP) method of Shustorovich to predict elementary step kinetics coupled with transition-state theory, a detailed microkinetic model of steam and dry reforming of methane has been developed for Rh(111) and Ni(111) in this thesis. While there is extensive literature on it, the standard reference on the mechanism and kinetics of MSR is that of Xu and Froment, who proposed a 13 step mechanism. Based on the assumption of rate limiting steps for these overall reactions, Xu and Froment derived rate expressions for overall kinetics with fitted parameters. Here a more detailed micro-kinetic model of steam reforming of methane has been developed by adding 3 steps pertinent to carbon formation on the catalyst to Xu and Froment's mechanism. The complete set as well as the dominant reaction routes has been identified. This was accomplished first by enumerating the list of reaction routes and drawing this network. A program was written in Maple and was used to assist in creating the list of full routes, empty routes and intermediate nodes. This program reduces the amount of repetitive work that was needed in an earlier Matlab program when computing the list. After drawing the complete reaction network it was than converted into an equivalent electrical circuit and Multisim analysis was performed. Further, the resistances of various reaction steps were compared. From the reduced graph, it was determined that reaction steps pertaining to desorption of carbon dioxide, i.e., step s4, and intermediate methylene forming intermediate methylidyne, s11, are the rate limiting steps. Further, through simulation with Multisim, it was determined that in fact only 2 overall reactions are needed. Adding a third overall reaction results in a nodal balance error. A rate expression was developed based on assuming the above two rate determining steps, with remaining steps at pseudo equilibrium along with the quasi-steady state approximation. The rate expression however produced a substantial error in conversion when compared to the overall microkinetic model. In addition to computing the micro-kinetic model, experimental work for methane steam reforming was conducted. A steam to carbon ratio of 2:1 was fed to the packed bed reactor, where experimental conversion data were obtained. These data points for Ni and Rh catalyst were plotted against the model to see how well the simulation predicted the experimental results. Reasonable agreement was obtained.