Date of Award
Fall 2016
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Mechanical Engineering
First Advisor
Singer, Simcha L.
Second Advisor
Allen, Casey
Third Advisor
Bowman, Anthony
Abstract
Coal currently supplies 40% of the world’s electricity needs, and is one of the most important energy sources. As the initial stage of coal combustion, pyrolysis is a thermal decomposition process which converts coal into light gases and tars, which are subsequently consumed in combustion reactions, as well as solid char. Recently there has been interest in using slow pyrolysis as a stand-alone process for the production of chemicals and fuels from large (mm-scale) coal particles. Simulations can be used to efficiently study the impact of pyrolysis conditions on gas, tar and char yields, as well as gas and tar species compositions, which are an important output for a coal-to-chemicals process. In order to simulate pyrolysis of large coal particles, the Chemical Percolation Devolatilization (CPD) model, which predicts the mass fractions of char, tar and light gas, has been modified and improved. A transient multicomponent vaporization sub-model has been developed to predict the partitioning of heavy species into gaseous tar and liquid metaplast. The Direct Quadrature Method of Moments (DQMoM) is introduced as a computationally efficient method to solve for the evolution of the distribution of tar species as a function of molar mass, and the full discrete tar species distribution can be reconstructed by a novel delumping procedure. Finally, a heat transfer model that can predict temperature gradients within the particles has been incorporated using the finite volume method to discretize the energy equation, with the improved CPD model implemented at every position within the particle. The results show the necessity of resolving large particles spatially, due to the impact of the local temperature evolution on tar and gas mass fractions and the production of certain species. Higher pyrolysis temperatures result in increased yields of gas and especially large tar species, while decreasing pressures also increase the production of heavier tar species. The agreement between the full discrete species model, which solves differential equations for every tar species, and DQMoM with delumping, which solves many fewer equations, is excellent, while yielding a large improvement in computational efficiency.