Phenomenological models of cellulose pyrolysis

Abstract

Using yields vs. residence time and temperature from 50 to 1000 ms and 650–900°C, measured with the ultra pyrolysis system at the University of Western Ontario (UWO) we establish an approximate total gaseous yield function Y(t, T). With UWO data, we also establish approximate correlations between individual gaseous yields (CO, CO 2, C 2H 4, CH 4, C 2H 2, C 2H 6 and C 3H 6) and the total gaseous yield that could be used to give Y i(t, T) for individual gases. We further extend Y(t, T) using shock tube pyrolysis measurements from 0.3 to 2 ms and 900 to 2100°C made at Kansas State University (KSU). In doing so, we develop a global decay model that gives analytical time and temperature dependencies for cellulose, activated cellulose, tar, prompt total gas and late total gas. We next examine the impact of heating rates and heat transfer upon pyrolysis of cellulose using slow pyrolysis data obtained by thermogravimetric analysis at the Colorado School of Mines (CSM). In this effort, we first develop an accurate general relationship for Boltzmann integrals. Then using an analytically convenient Arrhenius reaction rate (ARR) we examine data taken at varying heating rates and with three Biot numbers. We find some phenomenological analytical relationships giving ARR parameter dependencies on heating rate and particle size that appear indicative of heat transfer impacts. If adequate data becomes available these relationships might be applied to hemicellulose and lignin. Then the pyrolysis rates of any plant species might be predicted in terms of the pyrolytic characteristics of their cellulose, hemicellulose and lignin components.