Date of Award

8-2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Mechanical Engineering

Committee Chair/Advisor

Zhaoxu Meng

Committee Member

Hongseok Choi

Committee Member

Zhen Li

Committee Member

Lihua Lou

Abstract

Biopolymer blends and nanocomposites have attracted growing interest as sustainable alternatives to conventional petroleum-based polymers. However, the performance of these materials is governed not only by the macroscopic properties of each component, but also by molecular-level factors such as backbone stiffness, conformational flexibility, intermolecular interactions, component ratios, confinement effects, and processing history. A mechanistic understanding of how these factors collectively regulate processing–structure–property relationships remains limited. This dissertation employs atomistic and coarse-grained molecular dynamics simulations to elucidate the molecular mechanisms governing miscibility, chain mobility, deformation response, electrospinnability, crystallization, and phase behavior in biopolymer blends and nanocomposites. First, atomistic molecular dynamics (MD) simulations were used to systematically investigate the miscibility and mechanical modulation mechanisms of poly (3-hydroxybutyrate-co-3-hydroxyvalerate/Chitosan (PHBV/CS) blends. Flory-Huggins interaction parameter reveals favorable miscibility at both 300 K and 500 K, particularly at asymmetric compositions. Steered molecular dynamics simulations reveal that CS chains possess substantially higher backbone stiffness and torsional resistance compared to PHBV, underscoring the mechanical modulation potential of PHBV. Two representative compositions (10:90 and 90:10 PHBV/CS) are further examined to evaluate chain mobility, mechanical response, and conformational energetics. Results show that incorporation of

PHBV enhances the mobility and ductility of CS under tensile deformation, while CS imposes modest confinement on the dynamics of PHBV. Conformational energetic analysis further confirms that PHBV facilitates conformational transitions in CS chains, lowering the energetic barriers for deformation. These effects persist even in phase-separated morphologies, indicating that local molecular interactions and conformational flexibility play critical roles in mechanical modulation. Second, the processability of CS is examined through blending with polycaprolactone (PCL), with emphasis on the molecular mechanisms underlying enhanced electrospinnability. Experimental observations have demonstrated that PCL improves CS electrospinnability and promotes stable nanofiber formation. Atomistic molecular dynamics simulations further reveal that PCL enhances molecular mobility within the CS matrix, while CS imposes constraints on PCL chain motion, leading to composition-dependent mechanical and viscoelastic behavior. Small-amplitude oscillatory shear simulations further reveal that increasing PCL content leads to a more viscous-dominated response and improved viscoelastic stability, which are critical for sustaining continuous electrospinning filaments. Molecular-level jet simulations demonstrate that extrusion resistance is strongly dependent on blend ratio, geometry, and deformation rate, while being relatively insensitive to wall–polymer interactions. PCL-rich systems consistently exhibit lower resistance to extrusion and elongation, indicating improved deformability under confined flow conditions. Dihedral energy analysis confirms that PCL lowers the torsional energy barrier for CS backbone rotation, enabling more efficient conformational rearrangement during processing.

Third, to overcome the timescale and computational limitations of atomistic simulations, coarse-grained (CG) MD simulations are employed to examine crystallization of a model semicrystalline polymer, polyvinyl alcohol (PVA), under nanoconfinement. The results show that nanoconfinement can promote the crystallization process, especially at the early stage, and the interfaces between nanoconfinement and polymer can function as crystallite nucleation sites. In general, the final degree of crystallinity increases with the level of nanoconfinement. Region-dependent analysis further reveals higher crystallinity near the confinement interfaces, indicating a heterogeneous crystallization process. These findings demonstrate that nanoconfinement can significantly enhance PVA crystallization, even under ultra-high cooling rates. Fourth, we employ CG MD simulations to systematically investigate morphology and phase behavior in model polymer blends with distinct chain stiffness. By varying the intermolecular interactions strength, polymer chain stiffness, temperature, and chain length, the resulting morphological and phase behaviors is quantified using the peak intensity of the structure factor, 𝑆(𝑞∗), and mixing fraction of different beads type, 𝜙(mix). The results reveal that stronger intermolecular repulsive interactions strength promotes the phase separation, whereas increasing the chain stiffness of one polymer component in blends suppresses the phase separation even under strong interactions. Higher temperature enhances phase evolution due to increased bead mobility, while shorter chains exhibit more pronounced phase separation. The normalized density contrast map shows the consistent phase behaviors trend with calculated results obtained by 𝑆(𝑞∗) and 𝜙(mix). Overall, this dissertation provides important insights into how molecular characteristics govern the structure and properties of biopolymer blends and nanocomposites by employing different computational and modeling tools. Rather than treating blending as a simple strategy for combining desirable properties, this work clarifies how specific molecular features—including backbone stiffness, conformational flexibility, interaction strength, composition, and nanoconfinement—can be leveraged to tune miscibility, processability, crystallization, morphology, and mechanical response. These mechanistic insights provide guidance for the rational design of sustainable polymer materials with improved processing behavior and performance across a broad range of applications.

Available for download on Tuesday, August 31, 2027

Share

COinS