Enhanced cellulose nanofiber mechanical stability through ionic crosslinking and interpretation of adsorption data...

Area of Science:

• Materials Science
• Nanotechnology
• Chemical Engineering

Background:

• Cellulose nanofibers (CNF) are promising nanomaterials with tunable properties.
• Developing advanced functionalized CNF is crucial for novel applications.
• Ionic crosslinking offers a pathway to enhance CNF mechanical and adsorption characteristics.

Purpose of the Study:

• To fabricate and characterize ionically crosslinked cationic functionalized cellulose nanofibers (z_c-CNF).
• To evaluate the mechanical properties and adsorption capacity of the modified nanofibers.
• To develop and compare machine learning models for predicting the performance of z_c-CNF.

Main Methods:

• Fabrication of cationic functionalized cellulose nanofibers (c-CNF) with 0.13 mmol/g ammonium content.
• Ionic crosslinking of c-CNF using a pad-batch process.
• Characterization via infrared spectroscopy, tensile strength testing, and adsorption studies (Thomas model).
• Development and comparison of 23 classical machine learning models and deep neural networks using Pycaret.

Main Results:

• Tensile strength of z_c-CNF increased from 3.8 MPa to 5.4 MPa compared to c-CNF.
• Adsorption capacity of z_c-CNF reached 158 mg/g, fitting the Thomas model.
• Random Forests regression achieved 92.6% accuracy, while a deep neural network reached 96% accuracy.

Conclusions:

• Ionic crosslinking significantly enhances the mechanical properties of c-CNF.
• The developed z_c-CNF exhibits high adsorption capacity, suitable for various applications.
• Deep neural networks provide highly accurate predictive models for functionalized nanomaterial performance.