Chemoinformatic Investigation of the Chemistry of Cellulose and Lignin Derivatives in Hydrous Pyrolysis

Abstract

We present a data-driven approach to identifying the reaction network of the dominant chemistry in complex mixtures using model compounds representative of cellulose and lignin chemistry that are processed using hydrous pyrolysis. We present two methods for the identification of pseudocomponents: self-modeling multivariate curve resolution, which is a non-negative matrix factorization method, and Bayesian hierarchical clustering. The pseudocomponents are identified from spectroscopic data from two sources: Fourier transform infrared spectroscopy and 1H NMR spectroscopy. The data from the two sources is combined using a simple data combination method. Once pseudocomponents have been identified, Bayesian networks are used to identify directed pathways between the components, resulting in a proposed hypothesis for the reaction network or mechanism. We validate the methods by showing consistency of the derived reaction networks with the known chemistry of cellulose, lignin, and their derivatives and demonstrate the importance of data fusion in developing believable reaction networks.

Data‐driven hypotheses of reaction networks for thermochemical conversion of a physical mixture of levoglucosan and 2‐phenoxyethyl benzene

“…The methods used in this work have been explained in detail in our previous work [1,9,10] ; consequently, we provide only brief descriptions of the specific techniques used here. Note that the headings of the subsections provide both the names of the methods and their purpose.…”

“…The same 27 conditions of temperature, catalyst, and reaction time were used as in our previous work on LG and PEB, and further details of the microreactor and experimental conditions may be obtained there. [1] Subcritical water was used, and three separate sets of conditions: uncatalyzed, acid catalyzed (sulphuric acid) and base catalyst (sodium hydroxide), were used.…”

“…This follows the procedure outlined in our previous work and was performed in MATLAB version 2018b. [1]…”

“…In previous work, we have demonstrated the applicability of data-driven methods in generating hypotheses for reaction networks for the thermochemical conversion of model compounds related to cellulose and lignin. [1] Specifically, we had proposed reaction networks based on the fusion of data from multiple spectroscopic sources (Fourier transform infrared spectroscopy, FTIR, and proton nuclear magnetic resonance spectroscopy, 1 H-NMR) using Bayesian hierarchical clustering or self-modelling multivariate curve resolution coupled with Bayesian belief networks. The methods were applied to propose a reaction network for the hydrous pyrolysis of levoglucosan (LG), and another for the hydrous pyrolysis of 2-phenoxyethyl benzene (PEB).…”

Construction of thermoplastic cellulose esters matrix composites with enhanced flame retardancy and mechanical properties by embedding hydrophobic magnesium hydroxide

“…It is more environmentally sustainable to develop biodegradable composites with renewable biomass as matrix material. Cellulose, a typical kind of renewable and naturally degradable biomass, has the advantages of abundant sources, low cost, lightweight, high specific strength, and high specific modulus 3–5 . Cellulose‐based composites are expected to become the most promising candidate for petroleum‐based composites 6 .…”