Fast Pyrolysis of Wood for Biofuels: Spatiotemporally Resolved Diffuse Reflectance In situ Spectroscopy of Particles

Abstract

Fast pyrolysis of woody biomass is a promising process capable of producing renewable transportation fuels to replace gasoline, diesel, and chemicals currently derived from nonrenewable sources. However, biomass pyrolysis is not yet economically viable and requires significant optimization before it can contribute to the existing oil-based transportation system. One method of optimization uses detailed kinetic models for predicting the products of biomass fast pyrolysis, which serve as the basis for the design of pyrolysis reactors capable of producing the highest value products. The goal of this work is to improve upon current pyrolysis models, usually derived from experiments with low heating rates and temperatures, by developing models that account for both transport and pyrolysis decomposition kinetics at high heating rates and high temperatures (>400 °C). A new experimental technique is proposed herein: spatiotemporally resolved diffuse reflectance in situ spectroscopy of particles (STR-DRiSP), which is capable of measuring biomass composition during fast pyrolysis with high spatial (10 μm) and temporal (1 ms) resolution. Compositional data were compared with a comprehensive 2D single-particle model, which incorporated a multistep, semiglobal reaction mechanism, prescribed particle shrinkage, and thermophysical properties that varied with temperature, composition, and orientation. The STR-DRiSP technique can be used to determine the transport-limited kinetic parameters of biomass decomposition for a wide variety of biomass feedstocks.

Sources of Biomass Feedstock Variability and the Potential Impact on Biofuels Production

“…Even more promising, these models can be continuously updated and expanded to increase predictive utility as more biomass samples are added. Recently, Paulsen et al have obtained in situ biomass structural information using spatiotemporally resolved diffuse reflectance spectroscopy in a reactive fast pyrolysis system for woody feedstocks [59]. This new technique could allow for the monitoring of lumped carbohydrate content during pyrolysis.…”

Pyrolysis of whole wood chips and rods in a novel ablative reactor

“…Aston University has claimed that whole wood chips up to 50 mm can be processed in their ablative reactor, but no experimental data or details on the reactor design are available [26]. Recently, Paulsen et al [27] designed a micro ablative pyrolysis apparatus in order to develop a robust wood particle pyrolysis model that accounts for both transport and pyrolysis decomposition kinetics at high temperatures and high heating rates. In the reactor, pyrolysis of 1 mm thick wood particle was conducted by direct ablation with a heated surface, and the subsequent changes of wood composition were characterized by the spatiotemporally resolved diffuse reflectance in situ spectroscopy.…”

Application of machine learning to pyrolysis reaction networks: Reducing model solution time to enable process optimization

“…This is especially true for those models that use CFD approaches to move beyond single particles and model whole reactors. (Anca-Couce, 2016;Anca-Couce & Zobel, 2012;Mehrabian et al, 2012;Moghtaderi, 2006;Papadikis et al, 2009;Paulsen et al, 2014) Dynamic mathematical models are commonly used in process engineering across disciplines as key tools for model-based control, product and process design, and optimization. In particular, detailed models based on the underlying physics of a scientific phenomenon can serve as tools that complement, and, in some cases, replace physical experiments.…”