Experimental and Simulation Studies on Biomass Torrefaction and Gasification

Abstract

The potential for bioenergy in Norway is significant. This potential can be realized by improving the properties of biomass and making it a convenient and competitive alternative to other fuels. Torrefaction is the most promising biomass pretreatment technique to date, improving its effectiveness as a fuel in various thermochemical processes. Torrefaction considerably reduces moisture content but increases the heating value, hydrophobicity and grindability of biomass. Torrefaction is influenced by many parameters, including biomass composition, temperature, holdup time and particle size. To evaluate the feasibility of torrefaction in a particular region, locally available biomass resources should be investigated. This approach forms the basis of the present study. To improve the viability of bioenergy in Norway, I undertook fundamental research on the torrefaction of Norwegian woody biomass and evaluated the behavior of torrefied biomass in thermochemical processes. Starting with a detailed literature review on the topic, torrefaction behavior of Norwegian Birch and Spruce was experimentally investigated. Torrefaction experiments were performed in a macro-TGA reactor with provisions for continuous measurement of volatiles. Process temperature (225 and 275 °C), holdup time (30 and 60 minutes) and sample size (10 and 40 mm cubes) were varied. Fuel characterization, derivative thermogravimetric (DTG) curves, product yields, hydrophobicity tests, grinding energies and particle size distributions are discussed. Temperature had the strongest effect on the properties of torrefied biomass of all the studied parameters. Overall, considerable improvements in grindability and hydrophobicity were obtained in torrefied biomass from both feedstocks. To obtain information on the intrinsic kinetics of torrefaction, the pyrolysis kinetics of Norwegian spruce and birch wood was investigated in another study. Micro-TGA was employed with nine different heating programs, including linear, stepwise, modulated and constant reaction rate (CRR) experiments. The 18 experiments on the two feedstocks were evaluated simultaneously using the method of least squares. Part of the kinetic parameters could be assumed common for both woods without a considerable worsening of the fit quality. Three pseudocomponents were assumed. Two of them were described using distributed activation energy models (DAEM), while the decomposition of the cellulose pseudocomponent was described using self-accelerating kinetics. In another approach, all three pseudocomponents were described using n-order reactions. A table was calculated to provide guidance about the extent of devolatilization during torrefaction at various temperatures and residence times. For understanding torrefied biomass reactivity in oxidative conditions, another micro- TGA study was conducted with four torrefied wood samples and their original feedstocks (birch and spruce) at slow heating rate programs. Particularly low sample masses were employed to avoid self-heating of the samples due to heat of combustion. Linear, modulated and CRR temperature programs were employed in TGA experiments under gas flows of 5 and 20% O2. The kinetic model consisted of two devolatilization reactions and a subsequent char burn-off reaction. Cellulose decomposition in the presence of oxygen has selfaccelerating (autocatalytic) kinetics. Decomposition of the non-cellulosic components of the biomass was described using a distributed activation model. The char burn-off was approximated by power-law (n-order) kinetics. Each of these reactions has its own dependence on oxygen concentration, which was also expressed using power-law kinetics. The model contained 15 unknown parameters for a given biomass. Certain of these parameters could be assumed to be identical for the six samples without a substantial worsening of fit. Lastly, the behavior of torrefied biomass in a gasification process was evaluated. A twostage biomass gasification model was selected using Aspen Plus as the simulation and modeling tool. The model included minimization of the Gibbs free energy of the produced gas to achieve chemical equilibrium, constrained by mass and energy balances for the system. Air and steam were used as the oxidizing agents with both untreated and torrefied biomass as feedstocks. Three process parameters were studied: equivalence ratio (ER), Gibbs reactor temperature and steam-to-biomass ratio (SBR). A total of 27 cases were included in the analysis, operating the system below the carbon deposition boundary with all carbon in the gaseous form in the product gas. Product gas composition [hydrogen (H2), carbon monoxide (CO), carbon dioxide (CO2) and nitrogen (N2)] was analyzed together with cold gas energy and exergy efficiencies for all cases. Torrefied biomass gave higher H2 and CO contents in the product gas, as well as higher energy and exergy efficiencies, than untreated biomass. The overall efficiency of an integrated torrefaction-gasification process depends on the mass yield of torrefaction. The results were validated using a C-H-O ternary diagram combined with results from similar studies.