Progressive deconvolution of biomass thermogram to derive lignocellulosic composition and pyrolysis kinetics for parallel reaction model. (1st September 2022)

• Record Type: Journal Article
• Authors: Kim, Heeyoon; Yu, Seunghan; Kim, Minsu; Ryu, Changkook
• Is Part Of: Energy. Volume 254:Part C(2022)
• Publication Date: 2022-09-01
• Subjects: Biomass -- Lignocellulosic composition -- Pyrolysis -- Parallel reaction model -- Thermogravimetric analysis; Power resources -- Periodicals; Power (Mechanics) -- Periodicals; Energy consumption -- Periodicals; 333.7905
• DOI: 10.1016/j.energy.2022.124446
• Languages: English
• ISSNs: 0360-5442

Abstract

Abstract: The pyrolysis of land biomass incorporates the characteristic behaviors of three main carbohydrates, namely hemicellulose, cellulose, and lignin. The three-parallel-reaction model (TPRM) assumes independent decomposition of the components and has been shown to accurately predict the pyrolysis kinetics with rate parameters acquired by the deconvolution of a differential thermogram (DTG). However, the nonlinearity of mathematical rate expressions involving several parameters, such as kinetic constants and lignocellulosic composition, makes it difficult to obtain optimal values. In this study, a new method was proposed to resolve the nonlinearity by a stepwise deconvolution of the DTG curve for TPRM with n-th order reaction rates, without requiring initial values for the model parameters. Based on the characteristic pyrolysis behavior of each component, the kinetic constants were determined in the order of lignin, cellulose, and hemicellulose; next, the lignocellulosic composition was obtained using multiple linear regression. For four woody biomasses, the method predicted the lignocellulosic compositions within a 5.1% deviation and reproduced the thermogravimetric analysis curve within a 1.78% deviation. When tested for different biomass data available in the literature, the proposed method achieved an accuracy comparable to that of existing methods of DTG deconvolution employing complex mathematical expressions. Highlights: A new method was proposed to deconvolve differential thermogram of biomass stepwise. Rate constants of three biomass components were derived for parallel reaction model. Lignocellulosic composition was determined together with pyrolysis kinetics. The method was validated for woody and herbaceous biomasses with reasonable accuracy. The accuracy was comparable to existing methods employing more complex models.