Chemo-mechanical modeling of static fatigue of high density polyethylene in bleach solution

Abstract

High density polyethylene (HDPE) is increasingly used in infrastructure applications with a design service lifetime of several decades. In many cases, the HDPE member is exposed to a corrosive environment, such as in pipes carrying potable water, where the dissolved bleach selectively attacks the loosely packed amorphous phase of the polymer. The failure mechanism of HDPE transitions from a ductile to a brittle mode as the corrosion level increases. This leads to subcritical crack propagation, which deteriorates the load capacity of the structure. In this study, we develop a coupled chemo-mechanical model to simulate stress corrosion cracking (SCC) in HDPE members exposed to a bleach solution. The mechanical response of the polymer is described by a constitutive model that separately considers the individual deformation and damage behaviors of the amorphous and the crystalline phases. The model accounts for the intermolecular deformation and homogeneous void growth in the crystalline and amorphous phases, along with the resistance of the entangled network and craze damage in the amorphous phase. The embrittlement due to corrosion is captured by relating the constitutive parameters of amorphous phase to the molecular weight of the polymer. The diffusion and chemical reaction of bleach are described by a reduced order kinetics model that links the extent of polymer oxidation to the reduction of the molecular weight. The material constitutive model and diffusion–reaction model are combined in a single finite element (FE) code to investigate the SCC behavior of double edge notched HDPE specimens. The simulation yields the stress-life curves and fracture kinetics under different environments. The predicted stress-life curve qualitatively matches the measured stress-life data of polymer pipes. It is shown that the stress-life curve exhibits different regimes corresponding to distinct failure mechanisms, as indicated by the stress and strain distributions in the specimen.