Two differential constitutive equations, i.e. Giesekus model and Johnson-Segalman model were employed here to predict the time-dependent viscoelastic behavior of an LDPE melt in thixotropy-loop experiments and step shear rate experiment. Multiple relaxation modes were adopted, and the parameters used to describe the nonlinear viscoelasticity in the two models were obtained by fitting the shear-thinning viscosity. The predictions on those transient shear characteristics by the two models are found in qualitative agreement with our previous experiments. JohnsonSegalman model predicts oscillation behavior in the thixotropy-loop and step shear rate experiments, whereas Giesekus model does not. Both models predict higher shear stresses than the experimental data in the case of long time shearing, implying that both models are not able to completely characterize the time-dependent shear stress of the melt at high shear rate.
The time-dependent viscoelastic behaviors of an Low-Density-Pdyethyene（LDPE） （PE-FSB-23D022） melt were experimentally investigated using two types of thixotropy-loop test i.e., the triangular- and the trapezoidal-loop shear test. The shear strengthening was noted in the thixotropy-loop tests not only at high shear rate but also at low shear rate, and it was found that the dimensionless stress amplitude of shear strengthening at low shear rate is larger than that at high shear rate. Moreover, the shear strengthening is a time-dependent viscoelastic behavior that is observed during the initial stage of shearing, and it cannot be found when the time interval corresponding to the increase in shear rate is sufficiently long, The remarkable shear-weakening phenomenon was noticed at higher shear rate when shearing time is prolonged. Five types of the triangular thixotropy-loop results were marked in terms of the configurations of the stress-rate curves, which indicate the complex features of the time-dependent viscoelasticity. A simple model was used to describe the thixotropy-loop tests, and the fitted results showed good agreement with the experimental results.
A new simple thixotropy model was proposed in the present paper to characterize the thixotropy-loop experiments and the start-up experiment of an LDPE （PE-FSB23D0221Q200） melt. The thixotropy model is a combination of a viscoelastic-component and a postulated kinetics process of structure change, which is constituted in terms of the indirect microstructural approach usually adopted in the characterization of thixotropy. The descriptions of the thixotropy model on both the thixotropy-loop tests and the startup test show good agreement with the experimental values, indicating the good capability of the model in characterizing the time-dependent nonlinear viscoelastic. The stress overshoot phenomenon and the stress relaxation after cessation of the thixotropy loop test can be described well by the model, whereas both of the typical viscoelastic phenomena could not be described in our previous work with a variant Huang model.
The theoretical characterizations on the triangular-form thixotropy-loop tests of an LDPE melt （PE-FSB- 23D022/Q200） were conducted in the present paper by using a new thixotropy model, which is constituted by the upper convected Maxwell model and a rate-type kinetic equation. The new thixotropic Maxwell model can partially describe well three reported thixotropy-loop experiments by comparison with the previous calculations of the variant form of the thixotropy-type Huang model. It is noted that the stress deviations between the experiments and the predictions of the new thixotropic Maxwell model are much slighter than those deviations obtained by using the variant Huang model at the same condition, although both models include five parameters. The constitution of the new thixotropic Maxwell model is more reasonable than that of the variant Huang model.
