Chemometrics to connect feedstock quality, process settings and calorific value of hydrochar through infrared spectra

Abstract

The current urgent need for clean energy has sparked interest in hydrothermal carbonization (HTC) as a sustainable avenue to convert high moisture biomass wastes into an efficient bioenergy source. Without destructive methodologies and offering distinct advantages (e.g. real-time analysis) over traditional assessments, chemometric methods coupled to Fourier Transform Infrared Spectroscopy (FTIR) is presented as an innovative approach for a rapid assessment of the higher heating value (HHV) of hydrochars. With the help of advanced chemometric techniques such as Partial Least Squares (PLS) or Sequentially Orthogonalized PLS (SO-PLS) further insights are gained into the underlying chemical transformations during HTC treatment of a wide variety of biodegradable wastes. Our model predicts the HHV of the diverse hydrochars with an impressive R² value of 0.961 and an RMSE of 0.845 MJ/kg. The scores and loading plots point out the pivotal functional groups that distinguish the various hydrochars, while also unveiling the inherent similarities among them. A second model has been further devised to predict the HHV of hydrochar as a function of feedstock composition and HTC operation conditions, thus allowing process optimization. This challenging approach can be extrapolated to quickly evaluate other physicochemical parameters of hydrochars designed for various applications.