Increase of long-chain branching by thermo-oxidative treatment of LDPE: Chromatographic, spectroscopic, and rheological evidence

Abstract

Low-density polyethylene was thermo-oxidatively degraded at 170 °C, i.e., degraded in the presence of air, by a one thermal cycle (1C) treatment during times between 30 and 90 min, and by a two thermal cycles (2C) treatment, i.e., after storage at room temperature, an already previously degraded sample was further degraded during times between 15 and 45 min. Characterization methods include gel permeation chromatography (GPC), Fourier transform infrared (FTIR) spectroscopy, as well as linear and nonlinear rheology. A reduction of molar mass was detected for all degraded samples by GPC, as well as an increase of the high molar mass fraction of the 1C sample degraded for the longest time. Intrinsic viscosity measurements indicate also a reduction of molar mass with increasing degradation times for both 1C and 2C samples. Thermo-oxidation is confirmed for 1C and 2C samples by analyzing specific indices in FTIR. Linear viscoelasticity seems to be in general only marginally affected by thermo-oxidative exposure, while the enhanced strain-hardening effect observed in uniaxial extension experiments presents a clear evidence for an increased long-chain branching (LCB) content in both 1C and 2C samples. Elongational viscosity data were analyzed by the molecular stress function (MSF) model as well as the Wagner-I model, and for both models, quantitative description of the experimental data for all samples was achieved by fit of only one nonlinear model parameter. Time-deformation separability was confirmed for all samples degraded, 1C as well as 2C, for cumulative degradation times of up to 90 min. The characterization by GPC was confronted with the characterization obtained from nonlinear rheology. It can be stated that elongational rheology is a powerful method to detect structural changes due to thermo-oxidative degradation, especially the formation of enhanced LCB. It has the further advantage that experimental data can be quantified by a single nonlinear model parameter of constitutive equations like the MSF or the Wagner-I model.