Poster: Influence of Comonomer Type on the Calibration of Filter-based IR Detector in GPC-IR

Alberto Ortín. Polymer Char, Valencia, Spain.

Olivier Boyron. University of Lyon, France.

Abstract:

Linear Low Density Polyethylene (LLDPE) is an important type of polyolefin obtained by copolymerization of ethylene with α-olefins. In the manufacturing process of LLDPE, different α-olefins are used as comonomer by different producers, and both average comonomer weight fraction and its distribution are used to control the product’s properties.

Regarding comonomer distribution – or Chemical Composition Distribution (CCD) – both homogeneous and heterogeneous copolymers are produced nowadays. Techniques such as TREF, CRYSTAF, CEF, and more recently interactive Liquid Chromatography, are used to characterize the Chemical Composition Distribution. Besides the CCD, the way the comonomer is incorporated into the polyethylene chains as a function of chain length, or molar mass, is also an imporant microstructure property of LLDPE. This also applies to High Density Polyethylene (HDPE) products, in which a small amount of comonomer is added to the higher molar mass chains to enhance their end-use properties.

High-temperature Gel Permeation Chromatography with Infrared detection (HT-GPC-IR) is used to analyse the average chemical composition variations across the molar mass distribution in polyolefin copolymers. Filter-based multiple band IR detectors can be calibrated with copolymers of known comonomer content, and this calibration is greatly simplified when using homogeneous copolymers. A series of homogeneous ethylene/α-olefin copolymers with different level of comonomer have been used to develop calibration curves for different comonomer types, in order to elucidate the influence of the side-chain length in the detector’s sensitivity and calibration.

The calibration curves obtained for the different comonomers are compared using objective statistical criteria to establish when the detector sensitivity is significantly different, or when the observed differences fall within the method’s experimental uncertainty.