Session F19: Glasses, Nanocomposites, and Dynamics

Universal relaxation equation for disordered and complex systems

Slower-than-exponential relaxations often occur in disordered systems such as stress relaxation in polymers.  Lacking a microscopic theory, these are commonly fit to the empirical Kohlrausch-Williams-Watts (KWW) stretched exponential or the Curie-von Schweidler (CvS) power-law algebraic decay. In this work an anomalous-diffusion limited, mixed second-order reaction equation is used to unify the above relaxation laws as different limits of the same overall behavior. Here, relaxation is modeled as a mixed second-order reaction between a concentration of reactants that undergo anomalous diffusion. The resulting general expression uses four parameters: the minority-to-majority reactant ratio 0 < m < 1, the anomalous power-law exponent 0 < β < 1, the characteristic relaxation time τ, and the relaxation amplitude f_Δ. The m = 0 and m = 1 limits correspond to the KWW and CvS expressions, respectively, and the intermediate m values represent a new class of previously unrecognized relaxation functions. A fitting algorithm is introduced that identifies confidence intervals for each of the four experimental parameters. Prominent examples of disordered systems from polymer stress relaxation, biomechanics, energy storage, and dielectric relaxation show excellent fits to the proposed relaxation expression.   

Effects of Print Direction on the Mechanical Properties of 3D Printed Polymer Composites

For polymeric additive manufacturing (AM), print direction is an important processing parameter for the mechanical reliability of the inkjet 3D printed multi-material. To fully exploit AM potential in fabricating polymeric composites, a solid understanding of this factor during printing processes is in an urgent need.  The purpose of this study is to evaluate the influences of printing direction on the mechanical properties of 3D printed multi-material polymeric composites. The printing direction is defined via the angle of the specimens with respect to the movement of the print head. Specimens of stratified composites with various layer thickness are designed and fabricated via a multi-material 3D printer (Stratasys Connex3). Mechanical experiments are conducted to evaluate how the layer thickness and printing directions jointly influence the overall mechanical properties of the designs. Digital Image Correlation (DIC) is used to track the local and overall strain of the specimens during deformation.  This project also studied and quantified the variation in mechanical quality across multi-material 3D printed samples interfaces for different building direction using instrumented indentation testing (nanoindentation).  The findings in this work present an important step towards understanding the microstructural mechanics and properties of such 3D printed parts, which allow designers and engineers to better predict and model these materials.   

Weibull statistical evaluation of the properties of 3D-printed nanographene-containing ABS

Novel Acrylonitrile Butadiene Styrene (ABS) polymers melt-blended with 5 wt.% and 10 wt.% XGnP™ M5 graphene nanoplatelets were fabricated into 1.75 mm diameter filament for 3D printing (i.e., MatEx) to determine how the strength and toughness of 3D-printed ABS may be affected by M5 addition. An ABS with a bimodal molecular weight distribution (MWD) with a prominent portion of the distribution at the low MW end was selected with the intention to foster early-stage molecular interpenetration and interfilamentous adhesion of the filaments during printing. The ABS samples were thoroughly characterized for molecular weight, styrene, and acrylonitrile (AN) level, rubber content, and melt rheology. Scanning electron microscopy (SEM) revealed that the M5 nanoplatelets were very well dispersed in the final specially fabricated filaments. A two parameter Weibull distribution was used to quantify failure tendency and part lifetime. To this end, tensile bars were fabricated each with a 3 mm diameter hole at its center to provide for a defined stress concentrator, and subsequently milled to the shape of ASTM tensile bars. Two 3D printing patterns (0 deg. axial and 45-45 deg. bead raster angles) were selected to determine their influences on the Weibull analyses of the tensile results. As reflected in the Weibull moduli, graphene additions resulted in a decrease in reliability, whereas the defined hole geometry and the use of a 45 - 45 raster angle printing configuration resulted in an increase in Weibull reliability.   

Rational design of multicomponent nanocomposites toward hierarchical assemblies with design flexibility and structural fidelity

Current successes on directed self-assembly heavily rely on precision in building block design, composition, and pair interactions. These requirements impose inherent limitations to developing materials beyond nanoscale. In contrast,  biological blends and high-entropy alloys can readily accommodate composition variations while still achieving their intended structures. We hypothesize that diversified chemical complexity and increased composition variety are the key principles for the unique phase behavior. The entropic energy gains will enhance inter-phase miscibility, weaken the dependence on specific pair interactions and enable long-range cooperativity. The hypothesis is validated in complex blends containing small molecules, block copolymer-based supramolecules, and nanoparticles. We obtained hierarchically structured composites with formulation flexibility in the filler size selection and blend composition. Each component is distributed to locally mediate unfavorable interactions, cooperatively mitigate composition fluctuations, and retain structural fidelity. These systematic studies provided a viable pathway to release multiple constraints in the composite design, developed processing conditions to access structural control beyond nanoscale, and demonstrated an entropy-driven behavior in organic/inorganic composites.   

Effect of Polymer-Nanoparticle Interactions on the Fracture Toughness of Polymer-infiltrated Nanoparticle Films

Polymer-infiltrated nanoparticle films (PINFs) are a novel class of polymer nanocomposites that can have extremely high volume fractions of nanoparticles and thus unique mechanical properties. While our prior work focused on understanding the effect of confinement on the fracture toughness of these PINFs, the effect of interactions between polymer and nanoparticles has not been explored. Here we experimentally investigate the role of interfacial effects on the fracture toughness of PINFs prepared by capillary rise infiltration (CaRI) of polymer into silica (SiO₂) NP packings. We tune the polymer-NP interaction strength by using poly(styrene) which is known to have weak interactions with SiO₂, poly(2-vinyl pyridine) (P2VP) that is capable of forming hydrogen bonds with SiO₂, and trimethyl silane-passivated  SiO₂ that weakens interactions between the polymer and the nanoparticles. Interfacial effects play a role in PINFs composed of small NPs by changing the sliding friction at polymer-NP interface. In contrast, PINFs composed of large NPs are more affected by polymer-polymer interactions than polymer-NP interactions.   

Effects of Attractive Interfacial Interactions on Local, Nanoscale Stiffness of Supported Copolymer Adhesive Films

Research has shown that many key properties of nanoconfined polymers exhibit significant deviations from bulk responses. A major issue in nanoscience concerns the underlying cause of these nanoconfinement effects. Past studies have generally attributed the stiffness-confinement effect in supported films to the disparity between the rigidity of polymers and substrates; however, interfacial interactions can also be responsible for this behavior. Here, we have isolated the effects of interfacial interactions from substrate rigidity and studied how interfacial interactions affect the stiffness of supported polymer films near glass substrates. We have used a non-contact, self-referencing fluorescence method that is sensitive to local caging during the several hundred nanosecond excited-state lifetime of the fluorescence dye and thus provides a measurement related to a high-frequency modulus. We found that the stiffness of supported poly(styrene/n-butyl acrylate) (P(S/nBA)) random copolymer films is enhanced near the substrate interface, and the length scale over which the substrate perturbations propagate inside the film is highly dependent on the strength of the attractive hydrogen-bonding interactions between the hydroxyl groups on glass surface and the ester groups in nBA units. On hydrophilic glass, the length scale associated with the substrate perturbations increases from ~80 nm to ~180 nm when the nBA content in copolymers is increased from 59 mol% to 95 mol%. When the hydroxyl groups on a glass surface are partially or completely removed, the perturbation length scale decreases accordingly, with stronger impact being observed with nBA-richer systems. Thus, in addition to the disparity in polymer and substrate rigidities, our study shows that interfacial effects are another important underlying cause of stiffness-confinement effects of supported polymer films.   

Jamming on Deformable Surfaces

Jamming is a transition to rigidity that occurs as particulate media are compressed from a freely flowing state to a solid state. Jammed media exhibit many remarkable properties: In contrast to crystalline solids, jammed materials lack translational order and are fragile, offering little or no resistance to shear deformation, and exhibit other unusual elastic properties if the particles themselves are deformable. 

Recently, a number of experiments have been conducted that involve solidification of particulate media driven by a deformable or moving interface involving Pickering emulsions, bacterial systems, bijels etc. These situations do not neatly fit into the scenarios envisioned in prior theoretical work, because these only consider particle-particle interactions and take place in Euclidean space subject to deformable boundaries. Here, the rigidification takes place not only with respect to particle degrees of freedom, but also with respect to the shape of the interface itself. Further, the non-Euclidean geometry of the interface means that particles in different locations may experience location dependent states of stress depending on the local shape of the interface.

Here, we propose a new jamming category, which we refer to as metric jamming, that refers to structures jammed both with respect to particle degrees of freedom and surface degrees of freedom. Examples of metric jammed structures are produced through computer simulation. By adjusting the relative influence of surface and particle energies characterized by a single dimensionless parameter metric jammed structures can continuously be tuned from isostatic with vanishing shear modulus, i.e. similar to jammed states in flat space, to states that resemble conventional elastic solids. Other properties of these new structures will be discussed.    

Equilibrium States Corresponding to Targeted Nonequilibrium Pair Statistics

Nonequilibrium many-body systems are ubiquitous in physical, chemical and biological phenomena. Zhang and Torquato [Phys. Rev. E 101, 032124 (2020)] recently conjectured that any realizable pair correlation function $g_2({\bf r})$ or structure factor $S({\bf k})$ corresponding to either a translationally invariant equilibrium or nonequilibrium system can be attained by an equilibrium ensemble at positive temperature $T$ involving only (up to) effective pair interactions. To test this conjecture, we apply a new inverse methodology for two different nonequilibrium 2D models: the nonhyperuniform random sequential addition (RSA) process and a hyperuniform "perfect glass". We show that these nonequilibrium systems

are indeed achievable by equilibrium many-body systems at positive temperature involving up to two-body potentials. We also demonstrate that unlike typical liquids (e.g. high-density Lennard-Jones fluids), both of these equilibrium systems exhibit highly disordered inherent structures upon quenching. Our findings provide further support to the Zhang-Torquato conjecture and demonstrate the capability of equilibrium systems to realize properties conventionally associated with nonequilibrium systems.   

Tunability of the dispersion and rheology in PNCs by non-linear polymers at the polymer-NP interface

Attractive interactions between polymer chains and NPs form a bound layer in the PNCs. Previous studies on grafted and linear adsorbed chains on the NPs show that the bound layer has significant impact on dispersion of NPs and rheology of the PNCs. In this study, we used poly(ethylene oxide) with varying architectures (linear, stars and hyper-branched) to explore the effects of modified surfaces by ‘non-linear’ polymers on dispersion state of NPs and bulk rheology of the PNCs. We observed agglomeration in composites with unentangled non-linear polymers at the interface due to the decoupling of NPs and the linear matrix, while individual dispersion is a norm for PNCs with entangled adsorbed polymers. Rheology results in the linear viscoelastic region shows the tunability of the mechanical reinforcement in PNCs by modifying the bound layer through varying functionality, arm lengths, and the number of end group hydrogen bonds in non-linear polymers. In addition, large amplitude oscillatory shear tests show a strong relation between the extend of the linear viscoelastic region and the compactness of the adsorbed polymers. Overall, this study shows that the architecture of the bound polymers can be a powerful new parameter for controlling the rheological properties of the PNCs.   

Initial Curing Process between Epoxy and Amine at a Solid Interface

A chemical reaction between epoxy and amine compounds at a solid interface is of pivotal importance for the manifestation of the adhesion strength.  Here we examined the initial curing reaction of epoxy phenol novolac (EPN) and 4,4'-diaminodiphenyl sulfone (DDS) at a quartz interface using sum-frequency generation spectroscopy.  The reaction rate constant (k) at the quartz interface was much larger than in the bulk.  The Arrhenius analysis of k revealed that while the activation energy was identical to each other, the frequency factor was much larger at the interface than in the bulk.  The faster curing reaction at the interface was also confirmed for different system of diglycidyl ether of bisphenol A (DGEBA) and DDS.  The results might be explained in terms of a better orientation ordering and/or the densification of reactants at the interface, facilitating the encounter of the reactants there.     

Molecular inhomogeneities in thermal expansion and their impact on relaxation time and densification

Structural relaxation in liquids is considered to be directly controlled by density, a link which is the basis of most theoretical approaches to the glass transition. Yet, none of the theoretical concepts can describe the phenomenon comprehensively; particularly hydrogen bonding liquids defy a proper description. To investigate densification on the molecular scale, Infrared spectroscopy is carried out on a series of polyalcohols at temperatures ranging from far above to far below their respective glass transition temperatures. From specific molecular vibrations, the thermal expansion of intramolecular covalent bonds and strong intermolecular hydrogen bridges is quantified. Pronounced differences between intra- and intermolecular expansion verify the dominance of the latter. Surprisingly, the overall thermal expansion (i.e. the cube root of inverse density) is even bigger than that of the strong hydrogen bridges. This suggests that weak hydrogen bridges dominate thermal expansion while the strong hydrogen bridges clearly control the glass transition. These results demonstrate that inhomogeneities on intra- and intermolecular scale can play distinct roles in densification and glassy solidification and require a careful consideration in a comprehensive theoretical description.   

Predicting Several Seconds of the Relaxation Dynamics of an Entangled Polymer Melt from a Few Nanoseconds of Atomistic Molecular Dynamics

The rheology of high-molecular-weight polymer melts is determined by the entanglement dynamics. Polymer melts with hundreds of entanglements can have relaxation times on the order of several seconds. Predicting those relaxation dynamics from first principles requires bridging about eight orders of magnitude of time. This is a problem where coarse-graining is not only desirable but also necessary. The discrete slip-link model (DSM) is a hierarchy of strongly connected coarse-grained models that have great success predicting the linear and nonlinear rheology of high-molecular-weight polymers. Three of the four parameters of the most detailed DSM can be extracted from primitive path analysis. We have also recently shown how to extract the remaining friction parameter from atomistic simulations [J. Rheol. 64, 1035-1043 (2020)]. To illustrate the procedure, an available quantum chemistry-based force field for polyethylene oxide (PEO) was used to perform 100 ns of molecular dynamics (MD) simulations of a 12 kDa melt (8 entanglements). Then using the MD-extracted parameters, the DSM was used to predict several seconds of the relaxation dynamics of a 256 kDa PEO melt (171 entanglements). The predictions were compared to experiments performed at the same temperature.