(698i) A Modelling Pathway and Software Tool Connecting Molecular Structure to Linear Viscoelastic Properties and Melt Flow Index for Polyolefins

Polymer rheological properties are crucial for the processing behaviour as well as the end-use properties of polyolefins. Moreover, they are widely used for polymer quality control. Many different experimental methods exist in order to describe the complexity of polymer rheological behaviour. In industrial practice, frequency sweep and melt index measurement are mainly used. Frequency sweep measurement is typically employed for viscoelastic properties description providing polymer melt viscosity versus frequency diagram (flow curve). On the other hand, melt flow index measurement (MI) is the one that has gained great popularity in the field defined by the mass of polymer, in grams, flowing in ten minutes through a capillary under specified conditions. MI offers a fast, easy to perform and low cost polymer rheological behaviour characterization connected to end-use customer specifications.

Polymer micro-structure is affecting the rheological behaviour of the polymer. The term micro-structure describes fundamental properties of the polymer including molecular weight, co-polymer composition and chain sequence distribution. A clear example describing this connection is the following: high molecular weight polyolefins exhibit improved end-use mechanical properties (e.g., toughness, strength, impact resistance and environmental stress cracking resistance, etc.). At the same time, they have higher melt viscosities and, therefore, it is more difficult to process them. Usually, a balance is sought; a broad or bimodal MWD exhibits improved mechanical properties and good processability levels as well. In this way, rheology acts as a link between molecular structure and final properties of a polymer (Touloupidis V., 2011, ‘Mathematical Modelling and Simulation of an Industrial α-olefins Catalytic Slurry Phase Loop-reactor Series’, PhD thesis, Aristotle University of Thessaloniki). Thus, fundamental understanding of polymer rheology and having a method for the prediction of viscoelastic polymer properties on base of the underlying molecular micro-structure offers a valuable tool for more efficient polymer product development and reverse engineering.

In this work, a modelling pathway and software tool is presented for linking entangled linear polymer molecular properties to linear viscoelasticity and MI values. A two-step approach is followed: (i) a model developed under Matlab environment links molecular properties with flow curves, and then, (ii) a Polyflow model calculates MI values based on the flow curves predicted. The method applies for linear homo-polyethylene grades while it can be additionally extended for ‘slightly’ non-linear co-polymers.

The connection between the polymer molecular micro-structure and the viscoelasticity is achieved through a model combining reptation and Rouse relaxation concepts and basic rheology theory found in literature (van Ruymbeke E., Keunings R., Stephenne V., Hagenaars A. and Bailly C., 2002, ‘Evaluation of Reptation Models for Predicting the Linear Viscoelastic Properties of Entangled Linear Polymers’, Macromolecules 2002, 35, Kiparissides C., Pladis P. and Moen O., 2006, ‘From Polyethylene curves to Molecular Weight Distributions’, Multiscale Modelling of Polymer Properties, Elsevier). Based on molecular weight distribution measurements obtained via size exclusion chromatography (SEC), the developed rheology model is able to predict viscoelastic properties (including relaxation modulus, storage and loss moduli, flow curve, etc.) of linear polymers (e.g., polyethylene, PE, polystyrene, PS, polycarbonate, PC). The strong point of this method is that the predicted flow curve includes the polydispersity effect and the actual shape of the molecular weight distribution as this information. A software tool was developed under Matlab programming language based on the method described as a user-friendly stand-alone application, entitled ‘BorVisc’, presented via a user-friendly graphical user interface environment.

Furthermore, on the basis of the commercial Finite Elements CFD software package Ansys^®Polyflow a model was developed computing the MI value based on the flow curve predicted by BorVisc. The simulation model takes into account geometry details of the MI instrument as well as the specific loading of measurements (e.g., 2.16, 5, 10, 21Kg). In addition to MI values, the model is able to predict in detail pressure, velocity and shear rate profiles of measurements (i.e., barrel and capillary region). Polyflow model together with Borvisc offer the pathway to move from purely molecular micro-structure to the other end of MI application-oriented bulk properties.

The method was thoroughly tested and validated for different low- and high-density polyethylene grades exhibiting great performance. Accuracy levels in the range of ±10% on average is exhibited considering both steps of computation (i) MWD to flow curves and (ii) flow curves to MI. Taking into account that MWD, flow curve and MI experimental measurement deviations are inevitably present, one could note that the model is close to resolution of measurements. These promising results offer a solid ground for using the methodology presented (i.e., BorVisc software and Polyflow model) as a valuable tool to enhance product development towards the direction of end-use polymer bulk properties prediction. Moreover, the method offers also high potential when applied for reverse engineering, computing a molecular microstructure based on desired rheological properties.