(164f) Comparison of Rheological Models for Low-Density Polyethylene (LDPE) Produced in High-Pressure Tubular Reactors.

Low-density polyethylene (LDPE) is an important commodity in today’s economy due to its molecular architecture flexibility and low cost. It is usually manufactured by radical polymerization under high-pressure conditions in tubular or autoclave reactors. Frequently, a variety of LDPE grades are produced in a reactor by changing the operating conditions. These changes in the operating procedures modify the shapes of the distributions of properties of the different polymer grades, such as the molecular weight distribution (MWD) and long-chain-branching distribution (LCBD). Since the rheological behavior of the molten polymer depends on the connectivity of the molecules and the degree to which they are entangled with each other, these changes in molecular structure produce variations in the final processing properties of the polymer. It has been reported that even subtle changes in the number of branches every 1000 carbon atoms (LCB/1000C) have the potential of changing engineering properties such as shear viscosity and elastic response [1]. For most of the LDPE samples studied in the literature, several of the processability and final product properties are influenced significantly by the level of long-chain branches and the presence of high molecular weight species in MWD [2]. For example, in the case of LDPE blown film, which is one of the most used polymeric products, information related to the blown film structure and the relationship between molecular structure-end use properties is important to film manufacturers for improving film properties and reducing production costs. For that reason, controlling and designing properties of polymeric melts under processing flow is still one of the fundamental issues in polymer science [3].

To understand the effect of reactor operation conditions on molecular structure and consequently on rheological and processing properties of the molten polymer, it is necessary to describe how the molecules react to the imposition of flow. Several mathematical models have been proposed in the literature to describe the rheological behavior of both linear and branched polymer melts in terms of molecular characteristics, such as the weight-average molecular weight, MWD and LCBD. However, this is still today a topic of continuous research [4]. Linear viscoelastic parameters, such as the relaxation modulus, dynamic moduli and the complex viscosity, are commonly used to characterize polymer melts, because they are experimentally measured under low-frequency rotational tests that ensure that the rheological response obtained is completely due to the molecular architecture of the polymer. Another important rheological property is the shear viscosity as a function of the shear rate. This property, also called flow curve of the polymer, is usually employed as an input for the estimation of useful end-use parameters such as the melt index [5]. The above mentioned rheological properties are measured using different types of rheometers, which are usually expensive and need trained operators. Therefore, it would be very useful for polymer producers to count with a computational tool capable of predicting the rheological properties of the polymer as a function of the reactor operating conditions.

In the present work, two different rheological models for the prediction of the rheological properties: relaxation modulus, dynamic moduli, complex viscosity, and the shear viscosity curve, are compared. The first one is an empirical model developed by Bersted [6] and subsequently adapted by Pedersen and Ram [7] to predict the shear viscosity curve of LDPE. This model uses as input the MWD as well as long chain branching information expressed as the branching index. Bersted [6] developed an empirical model relating the shear dependence of the steady shear viscosity of linear high-density polyethylene (HDPE) melts to the MWD. Pedersen and Ram [7] extended this model to branched LDPE melts using the mean-square radius of gyration instead of the weight-average molecular weight to describe the melt viscosity behavior. The model by Bersted [6] requires a shear-rate dependent parameter, M_c, which separates molecular weights into two classes: molecules with molecular weight below M_c contribute to the viscosity as they do at zero shear rate, and molecules with molecular weight above M_c contribute to the viscosity as if they had the same relaxation times as molecules of molecular weight . Each molecular weight species contributes according to its weight fraction. In the extended model by Pedersen and Ram [7], M_c is replaced by gM_c, introducing the effect of the branches in LDPE trought the branching index g.

The second rheological model studied in this work is based on molecular theories that describe how sufficiently long molecules participate in entanglements with other molecules restricting the motion of the polymer melt. The first theoretical molecular model for predicting rheological properties in terms of the MWD was originally proposed by de Gennes [8] and by Doi and Edwards [9] for linear polymers. The underlying theory of these models represents the molecular environment of an entangled chain employing the concept of a tube, in which the chain relaxes after a strain imposition through the mechanism of reptation. Subsequently, this theory was extended by Tsenoglou [10] and Des Cloizeaux [11], who introduced the concept of double repatriation (DR). This last author proposed a model called time-dependent diffusion reptation (TDD). Several authors have used the TDD model in conjunction with a double reptation model obtaining excellent predictions of the relaxation modulus, dynamic moduli, the complex viscosity, and the shear viscosity curve, for linear and slightly branched polyethylene [12]. In the present work, the TDD model is updated to incorporate the influence of long-chain branches of LDPE on molecular entanglements and in the relaxation mechanism, using information about MWD and LCBD.

Each rheological model was added to a deterministic model of the high-pressure ethylene polymerization in tubular reactors previously developed by the authors [13]. The reactor model predicts the MWD and the branching index, as well as several other molecular properties of LDPE, such as the bivariate molecular weight-long chain branching distribution (MWD-LCBD), the bivariate molecular weight-short chain branching distribution (MWD-SCBD),the weight-average molecular weight, the number-average molecular weight, and LCB/1000C. The calculated MWD and branching index are used as input for the rheological models. Model parameters corresponding to each rheological approach were adjusted using experimental data of shear viscosity of several LDPE samples. The results of both models were compared to evaluate the effectiveness of the predictions concerning the complexity and the computational time of each model.

1. Beer, F., G. Capaccio, and L. Rose, High molecular weight tail and long‐chain branching in SRM 1476 polyethylene. Journal of applied polymer science, 1999. 73(14): p. 2807-2812.
2. Shida, R. and L. Cancio, Correlation of low density polyethylene rheological measurements, with optical and processing properties. Polymer Engineering Science, 1977. 17(11): p. 769-774.
3. Costa, M.C.B., et al., Empirical models for end-use properties prediction of LDPE: application in the flexible plastic packaging industry. Materials Research, 2008. 11(1): p. 23-30.
4. Read, D.J., From reactor to rheology in industrial polymers. Journal of Polymer Science Part B: Polymer Physics, 2015. 53(2): p. 123-141.
5. Seavey, K.C., et al., Quantifying relationships among the molecular weight distribution, non-Newtonian shear viscosity, and melt index for linear polymers. Industrial & engineering chemistry research, 2003. 42(21): p. 5354-5362.
6. Bersted, B., An empirical model relating the molecular weight distribution of high‐density polyethylene to the shear dependence of the steady shear melt viscosity. Journal of Applied Polymer Science, 1975. 19(8): p. 2167-2177.
7. Pedersen, S. and A. Ram, Prediction of rheological properties of well‐characterized branched polyethylenes from the distribution of molecular weight and long chain branches. Polymer Engineering Science, 1978. 18(13): p. 990-995.
8. de Gennes, P.G., Reptation of a polymer chain in the presence of fixed obstacles. The journal of chemical physics, 1971. 55(2): p. 572-579.
9. Doi, M. and S.F. Edwards, The theory of polymer dynamics. Claredon, Oxford. ISBN 0-19-852033, 1986. 6.
10. Tsenoglou, C., Molecular weight polydispersity effects on the viscoelasticity of entangled linear polymers. Macromolecules, 1991. 24(8): p. 1762-1767.
11. Des Cloizeaux, J., Relaxation and viscosity anomaly of melts made of long entangled polymers: Time-dependent reptation. Macromolecules, 1990. 23(21): p. 4678-4687.
12. Van Ruymbeke, E., et al., A sensitive method to detect very low levels of long chain branching from the molar mass distribution and linear viscoelastic response. Journal of rheology, 2005. 49(6): p. 1503-1520.
13. Dietrich, M.L., et al., LDPE Production in Tubular Reactors: Comprehensive Model for the Prediction of the Joint Molecular Weight-Short (Long) Chain Branching Distributions. Industrial Engineering Chemistry Research, 2019. 58(11): p. 4412-4424.