Cellulose nanocrystals (CNCs) are nanomaterials that have shown promise as reinforcement filler materials. Their small size, high modulus, and high aspect ratio makes CNCs good reinforcing materials. CNCs are typically introduced into softer polymer materials, which can have incompatible surface chemistry such as aliphatic chains, leading to aggregation and poor reinforcement of the material. The intrinsic hydrophobicity of the CNC surfaces suggests that dispersal into hydrophobic polymer matrices, which the CNCs could potentially reinforce, represent a significant challenge. Therefore, new non-traditional strategies are needed to introduce CNCs into polymer materials. The hydroxyl groups on the surfaces of CNCs can be functionalized using a variety of chemical techniques to yield materials that can interact better with solvents or polymers. Additionally, surface groups can allow the CNCs to react with environmental stimuli (smart materials).
The primary focus of this work is the incorporation of CNCs in hydrophobic matrices. Herein we introduce a new method of dispersing CNCs in polyethylene (PE), a substance of legendary hydrophobicity that is also the most common synthetic polymer used in consumer packaging. The prospect of increasing the mechanical strength of PE by incorporating CNC materials as fillers may lead to the possibility of using less polymer to obtain the same strength.
This thesis approaches the problem of dispersing CNCs within PE by first functionalizing the CNCs with a catalyst capable of polymerizing ethylene and other α-olefins. The catalyst 1,1'-bis(bromodimethylsilyl)zirconocene dibromide (catalyst 1) is equipped with anchoring groups that are capable of attachment to the surface hydroxyl groups of CNC particles. After immobilizing catalyst 1 onto various CNC samples, introduction of solvent, organoaluminum cocatalyst, and monomer (ethylene alone or ethylene plus 1-hexene) afforded high density polyethylene (HDPE) and linear low-density polyethylene (LLDPE) samples, respectively, containing well-dispersed CNCs as filler materials.
Chapter 2 provided important information on the attachment of catalyst 1 to cellulose nanocrystals and the successful polymerization of ethylene from the cellulose nanocrystals. The resulting composite materials showed a in Young's modulus that was three-fold that of PE samples we tested (1600 ± 100 vs 500 ± 30) and about 10% greater relative to a commercial high modulus PE sample (1450 MPa). The increase in Young's modulus along with the lack of macroscopic aggregates led to the conclusion that we have developed a viable method to disperse CNCs in polyolefin matrices.
Chapter 3 focused on the dispersal of CNCs in a softer, more pliable polyethylene grade known as linear low-density polyethylene (LLDPE). LLDPE incorporates a small fraction of 1-hexene into polyethylene as a randomly inserted comonomer, giving rise to properties suitable for applications in plastic films and bags among other end uses. Catalyst 1 functionalized CNCs were added to a reaction vessel with both ethylene and 1-hexene to afford LLDPE CNC composites. Different loading of catalyst 1 on CNC aerogels afforded the same amount of catalyst in each reaction but allowed for different CNC loadings in each reaction. The composite materials showed increasing Young's modulus with increasing cellulose nanocrystal content.
Chapter 4 describes how CNCs were functionalized with the intention of filling reverse osmosis membrane materials to have surface chemistry that could be impart antibacterial properties and increase flux. CNCs were functionalized with carboxylic acid by 2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO)-mediated oxidation, then amine functionalization by carbodiimide coupling chemistry, and finally functionalized with a zwitterionic group by β-propiolactone ring opening. Amine coupling was confirmed with X-ray photoelectron spectroscopic analysis, and a second carboxylic acid peak was confirmed using infrared spectroscopy. These results were further verified with conductometric titration showing that after each respective reaction there were 1060 mmol kg-1 of carboxylic acid groups, 520 mmol kg-1 of amine groups, and 240 mmol kg-1 of zwitterionic groups. This CNC material was left to undergo future testing for desirable membrane properties.
Chapter 5 assesses the possible value in creating a new composite material using a functionalized polynorbornene, poly(5-triethoxysilyl-2-norbornene) (PTESN). The composites were fabricated by using the solvent casting method, dispersing the CNCs in a toluene solution of polymer and drying. The composite materials showed an increase in Young's modulus with increased loading. The 20 wt% CNC in PTESN had a Young's modulus of 970 MPa, a significant increase over the Young's modulus of the polymer lacking the filler (540 MPa).
In summary, this dissertation advances new techniques for the incorporation of CNCs as fillers in polymer-based nanocomposites. We are confident that further refinement and development of our results will find wide-ranging application.
Doctor of Philosophy
Cellulose nanocrystals (CNCs) are materials that can be added to polymers to form composite materials having increased stiffness. CNCs have the primary advantages over other filler materials of providing significant reinforcement without changing the color or increasing the density of the overall composite. CNCs are therefore good for designing polymer composites that need to be lightweight and aesthetically pleasing. Packaging materials (especially plastic bags and plastic films) are dominated by polyolefin materials such as polyethylene, which is already lightweight and colorless. The challenge of mixing polyethylene and CNCs is that their surface chemistry is incompatible, "like oil and water." To overcome the natural tendency for the CNC filler material to separate from the surrounding polyethylene matrix, a catalyst was attached to the surface of the CNCs and polymerization ensued from that catalyst leading to a composite material in which tiny CNC particles were trapped in the matrix Good dispersal of the component substances in the composite and of excellent overall reinforcement were proven by physical analysis.