Search

This research project investigated the physical aging of glassy polymers in confined environments. Many recent studies of aging in glassy polymers have observed that aging behavior is often strongly affected by confinement. Understanding aging in confined environments (e.g., thin polymer films and nanocomposites) is vital for predicting long-term performance in applications that use confined glassy polymers, such as gas separation membranes and advanced nanocomposite materials. Aging in bulk and layered films produced via layer-multiplying co-extrusion was studied using gas permeability measurement and differential scanning calorimetry (DSC). The layered films consisted of polysulfone (PSF) and a rubbery co-layering material, with PSF layers ranging in thickness from ~185 nm to ~400 nm. Gas permeation aging studies at 35 °C revealed that the PSF layers in layered films aged in a manner that was similar to bulk PSF and independent of layer thickness. This finding differs from what was observed previously in freestanding PSF films, in which aging depended strongly on thickness and was accelerated relative to bulk. Isothermal aging studies at 170 °C and cooling rate studies were performed on both bulk and layered samples using DSC. The aging of the PSF layers was similar to aging in bulk PSF for films having PSF layer thicknesses of ~640 nm and ~260 nm, while the film with 185 nm PSF layers showed a slightly higher aging rate than that of bulk PSF. The results of the DSC studies generally support the conclusions of our gas permeation aging studies. The absence of strong thickness dependence in aging studies of layered films tends to support the idea that the effect of film thickness on physical aging stems from interfacial characteristics and not merely thickness per se. The physical aging of thin polystyrene (PS) films at 35 °C was also investigated using gas permeation techniques. PS films of 400 nm and 800 nm did not exhibit aging behavior that was highly accelerated relative to bulk or strongly dependent on film thickness. At the thicknesses and aging temperature considered, the aging of PS shows much weaker thickness dependence than that seen in polymers like PSF and Matrimid.

The relative rates of the diffusional and relaxational processes during the absorption of penetrant molecules in glassy polymers determine the nature of the transport process and lead to Fickian, Case II, and anomalous absorption behavior. While previous models account for anomalous behavior, there is still a disconnect between theory and experiment, as data must be fit to the model with previously determined independent parameters. With trends leading to smaller device scales and increasingly complex polymer structures, there is a need for a quantitative understanding of the manner in which a polymer’s network structure alters both the rate and the mode of penetrant transport. To this end, samples of glassy poly(methyl methacrylate), poly(2-hydroxyethyl methacrylate), and poly(vinyl alcohol) were synthesized primarily by an iniferter-mediated, thermally-initiated free radical polymerization procedure. The thermal and mechanical properties of these polymers, as well as the polymer network structure, were varied through crosslinking and confirmed by detailed characterization. The dynamics of small molecule penetrant transport were examined in each polymer, with an emphasis on the occurrence of non-Fickian and Case II transport. The degree of crosslinking and choice of crosslinking molecule were shown to be powerful tools in tuning the observed penetrant transport process. For instance, the transport dynamics were altered from Fickian to Case II by increasing the degree of crosslinking and from Case II to Fickian by increasing the crosslinking interchain bridge length. Within the purely Case II regime, the rate of penetrant transport, or the Case II front velocity, was shown to scale with the square root of the degree of crosslinking in all systems investigated. A novel procedure for the in situ examination of penetrant transport in glassy polymers was developed utilizing high-resolution X-ray computed tomography. This completely nondestructive technique was used to visualize features in the interior of opaque solid objects and obtain digital information on their 3-D structure and properties. In this manner, the time-dependent penetrant concentration profiles throughout a swelling polymer were determined and analyzed.

Polycarbonate(PC), like most glassy polymers when undergoing large deformations exhibits a very complex thermo-mechanical response. We look at PC as a model glassy polymer and examine its response to changes in loading rate, loading direction and temperature. We show that plastic flow in compression is accompanied by a change in the elastic response of PC from isotropic to anisotropic. This provides the necessity to introduce a new modeling method that can capture these changes. A finite deformation plasticity like thermodynamically consistent model is developed to capture this and the observed rate and temperature dependence. For this modeling method, we need to provide expressions for the elastic response, flow of the plastic deformation gradient and also provide methods to calculate the back stress. In an effort to mathematically capture the observed response of PC under a broad range of mechanical loads and thermal conditions, a protocol of experiments has been developed that can be used to systematically evaluate the parameters needed to characterize PC, and with potential for application to other glassy polymers. This experimental protocol combines traditional mechanical testing, ultrasonic wave speed measurements, dynamic kolsky bar methods and standard calorimetric studies to systematically construct a model for PC. Polycarbonate (PC) is a tough transparent glassy polymer that is used for many structural applications, particularly as components of window and protective armor. Below its glass transition temperature of about 150°C, PC plastically flows under load, showing a rate dependent response and will not fully recover after removal of load, showing elements of plastic flow. As a result, the response of PC has been modeled using a rate dependent type model. When plastically deformed below the glass transition temperature, unlike PMMA, which is also a glassy polymer, PC will show very strong anisotropic elastic response. This can be seen from ultrasonic measurement of elastic wave-speed moduli after different levels of plastic compression. This is the first indication that traditional modeling methods based on plasticity will not work for capturing the response of PC if they are based on stress response models that only depend on the elastic deformation gradient. One needs to include additional history defining parameters in the stress response function, such as the plastic deformation gradient, to capture this development of anisotropy with plastic flow.