Kinetic Modeling, Simulation, and Optimization of Pyrolysis

Thermal decomposition of organic matter under inert atmospheric conditions, leading to the release of volatiles and formation of cha is called as pyrolysis. It is also a first step in the biomass gasification. To design of a suitable pyrolysis reactor, understanding of kinetic parameters is essential. In the proposed kinetic model of this study, biomass decomposition is represented as a decomposition of its main three components namely cellulose, hemi cellulose and lignin. Each component is modeled as decomposition by two competing reactions giving gaseous volatiles and solid charcoal is used. Logarithmic Differential Evolution (LDE) is used to find the kinetic parameters by minimizing the square of the error between the reported experimental data of thermogravimetry of hazelnut shell and simulated model predicted values of residual weight fraction. Logarithmic DE, an improved version of simple DE, incorporates logarithmic initialization and logarithmic mutation to take care of wide ranges of variable values. Logarithmic DE is found to yield better kinetic parameters in terms of objective function and gave better fit with experimental data. INTRODUCTION Decomposition of a compound, in the absence of oxygen, by the action of heat alone to produce various organic gaseous products, charcoal and tar, is known as pyrolysis [1]. Pyrolysis is not only an independent process, but also a first step in the gasification or the combustion process. Hazelnut shell is an abundantly available agriculture residue. Dogru et al. [2] used a pilot plant scale downdraft gasifier to investigate gasification potential of hazelnut shells. It is necessary to understand the kinetics of pyrolysis in order to design a suitable pyrolysis reactor. Balci et al. [3] proposed several kinetic models for hazelnut pyrolysis and validated by thermo-gravimetric experiments. In these kinetic models, the rate expression based on first-order decomposition of the reactive solid is defined in terms of fractional conversion. Demirbas [4] performed thermogravimetric experimental runs and presented the weight loss data for different particle sizes of ground hazelnut shell and for various heating rates. Kinetic analysis has also been carried out but the expression for the kinetic constants with respect to temperature is not developed, and the experimental data validation with theoretical models for these experiments is not reported in the literature. Experimental and modeling studies have been conducted on pyrolysis by many researchers [1-10]. In our previous studies [10, 11], a population based search algorithm, Differential Evolution (DE), which is simple and robust and has proven successful record, is employed for estimation of kinetic parameters. However, for some problems simple DE gave poor population distribution for cases where the range of limits was very wide (more than three orders of magnitude). Hence simple DE is modified first by including the logarithmic initialization (LIDE). The algorithm is further improved by incorporating the logarithmic mutation also and named as logarithmic DE (LDE) [12]. In the present study, Biomass decomposition is represented as a sum of decomposition of its components cellulose, hemi-cellulose and lignin. For each component, a kinetic model based on two competing reactions giving gaseous volatiles and solid charcoal is used. To find kinetic parameters of proposed model, an objective function based on least square error between experimental data and simulated results has to be minimized. The objective function minimization is carried out by Logarithmic DE. Model simulation results are validated with the data reported in literature [4]. KINETIC MODELING & SIMULATION The kinetics of thermal decomposition of biomass materials is complicated, as it involves a large number of reactions in parallel and series. Biomass mainly consists of three components: cellulose, hemi-cellulose and lignin. In the present study, the pyrolysis process is described by the independent reactions corresponding to the decomposition of the constituent components cellulose, hemi-cellulose and lignin. The pyrolysis reactions of each component are described by means of the scheme proposed by Koufopanos et al. [8, 9]. This model indicates that the biomass decomposes to volatiles, gases and char. The volatiles and gases may further react with char to produce different types of volatiles, gases and char where the compositions are different. Therefore, the primary pyrolysis products participate in secondary interactions (Reaction 3), resulting in a modified final product distribution. Biomass Component (cellulose/ hemi-cellulose/lignin)(n1 order decay) Reaction 1 Reaction 2 Reaction 3 (Volatiles + Gases)1 + (Char)1 (Volatiles + Gases)2 + (Char)2 (n2 order decay) (n3 order decay) The kinetic equations for the mechanism, shown above are represented by Eq. (1) through Eq. (7).