Date

2015

Document Type

Dissertation

Degree

Doctor of Philosophy

Department

Chemical Engineering

First Adviser

Caram, Hugo S.

Other advisers/committee members

Snyder, Mark A.; Acharya, Divyanshu; Herman, Richard G.; Hsu, James T.; McIntosh, Steven

Abstract

This work has been an investigation of the catalytic conversion of syngas into mixed alcohols over molybdenum based catalysts. The primary focus has been on the cesium promoted molybdenum disulfide catalysts supported on activated carbon. The catalyst was selected because of its excellent sulfur tolerance and water gas shift properties. The alcohol synthesis is a possibility for the production of gasoline additives or replacements to cater the growing demand of alcohols as a motor fuel.A catalyst preparation method has been developed for the production of cesium promoted molybdenum sulfide catalysts supported on activated carbon and its application for mixed alcohol synthesis was demonstrated. The basic steps involved in catalyst preparation are formation of crystalline molybdenum dioxide upon thermal decomposition of highly dispersed molybdenum precursor on activated carbon support, followed by transformation to sulfide complexes upon sulfidation, and cesium promotion being the last step. The catalyst composition (cesium to molybdenum ratio), catalyst preparation process parameters (temperature, promotion rate, etc.), alcohol synthesis reaction conditions (reaction temperature, pressure, gas hourly space velocity, and hydrogen to carbon monoxide feed ratio) were optimized with respect to alcohol yields, alcohol selectivity, and carbon monoxide conversion. The catalyst maintains its activity for more than 500 hours under higher alcohol synthesis conditions. The sulfur products and water were not detected in the products during this period.The x-ray diffraction (XRD) and x-ray photoelectron spectroscopy (XPS) analyses were performed after each stage of catalyst preparation and testing process. The combined XRD and XPS studies reveal that, the sulfidation (conversion of molybdenum dioxide to molybdenum disulfide) was not complete at a sulfidation temperature of 723.15 °K. The effect of sulfidation temperature on catalyst performance was investigated. It was found that the alcohol yields, alcohol selectivity, and carbon monoxide conversion increases with increase in suilfidation temperature, up to a maximum sulfidation temperature of 923.15 °K employed for our studies. The increase in catalyst activity was attributed to the increase in molybdenum disulfide phase and decrease in molybdenum dioxide phase in the catalyst. A complete conversion to molybdenum disulfide was achieved at 923.15 °K. Another attempt was made to prepare the catalyst by direct sulfidation of highly dispersed molybdenum precursor on activated carbon support. The catalyst was tested extensively for more than 600 hours under higher alcohol synthesis conditions. The loss of catalyst activity and the presence of water and sulfur compounds were not observed in the product during this period. An increase in alcohol selectivity was observed for this catalyst compared to the catalyst prepared by previous method.Additional experiments involving external injections of methanol and ethanol were performed to understand the reaction pathways involved during higher alcohol synthesis. It was observed that, at least part of the hydrocarbons is formed from alcohol decomposition and the higher alcohols are formed via aldehyde route. The steady-state power-law and Langmuir-hinshelwood type kinetic models were developed based on these observations to demonstrate the effect of reaction temperature, pressure, gas hourly space velocity, and hydrogen to carbon monoxide feed ratio on product yields. The power-law and Langmuir-hinshelwood model requires seven reactions. The singular value decomposition was applied for the first time to a higher alcohol synthesis system. Only two empirical forward reactions are sufficient to describe the catalytic behavior under higher alcohol synthesis conditions. An empirical kinetic model based on these empirical reactions was developed. A genetic algorithm minimization tool was employed to estimate the kinetic parameters associated with the power-law, Langmuir-hinshelwood, and empirical kinetic models. Finally, a non-isothermal reactor model was developed based on the two reaction empirical kinetic model and further extended to incorporate the recycle of un-converted syngas from out let of the reactor.

Share

COinS