Impact of species-based wood feedstock variability on physicochemical properties of cellulose nanocrystals

Cellulose nanocrystals (CNCs) are renewable, biodegradable and biocompatible nanomaterials that exhibit unique mechanical, optical, rheological and barrier properties useful for a variety of applications. Since CNCs remain an expensive product, coupling its production to pulp and paper products could reduce the production costs. However, pulp and paper mills do not always use a single wood species and so, differences in CNC properties based on the feedstock must be delineated. This work reports on the production and characterization of CNCs from three hardwoods, i.e. white oak, sweetgum and cottonwood, and one softwood, i.e. loblolly pine. Stable aqueous CNC dispersions were obtained from pre-extracted and fully bleached kraft wood pulps. CNC yield (with respect to pulp) ranged between 60 and 80% w/w, and a linear regression model explaining the relationship between CNC yield and pulp cellulose to lignin ratio was found significant (R² = 0.99 and p value = 0.0044). Sweetgum CNCs had the highest yield and purity, indicating the usefulness of this unconventional raw material for scalable production of high-quality CNCs. Pine CNCs were characterized by the largest length (150–200 nm vs 100–150 nm for the rest), whereas cottonwood CNCs displayed the lowest width (6 nm vs 14–17 nm for the rest) and consequently the highest aspect ratio (~ 20 vs ~8–13 for the rest) as well as the highest Young’s modulus (1780 GPa vs 100–300 GPa for the rest). A linear regression model explaining the relationship between Young’s modulus and CNC aspect ratio was found significant (R² = 0.92 and p value = 0.0387). Other physicochemical properties, such as crystallinity index, surface charge, polydispersity index and surface roughness, were similar for CNCs derived from all four species; however, thermal degradation behavior of oak CNCs was slightly different from that of the remaining CNCs (lower T_half and residual weight). Findings in this study indicate the potential to tune the quantity (yield) and quality (purity and properties) of CNCs based on the type of wood species chosen as the feedstock, thus having a direct impact on production economics and end use in different applications.