A new method of polymer classification is described involving dynamic mechanical analysis of polymer properties as temperature is changed. The method is based on the chemometric analysis of the damping factor (tan delta) as a function of temperature. In this study four polymer groups, namely, polypropylene, low density polyethylene, polystyrene and acrylonitrile-butadiene-styrene, each characterised by different grades, were studied. The aim is to distinguish polymer groups from each other. The polymers were studied over a temperature range of -50 degrees C until the minimum stiffness was reached, tan delta values were recorded approximately every 1.5 degrees . Principal components analysis was performed to visualise groupings and also for feature reduction prior to classification and clustering. Several clustering and classification methods were compared including k-means clustering, hierarchical cluster analysis, linear discriminant analysis, k-nearest neighbours, and class distances using both Euclidean and Mahalanobis measures. It is demonstrated that thermal analysis together with chemometrics provides excellent discrimination, representing a new approach for characterisation of polymers.

No abstract

An original instrument for thermo-mechanical polymer testing has been developed. This article describes the process of data acquisition, preprocessing and classification into 11 main polymer groups.The following polymer groups are used: polystyrene, acrylonitrile-butadiene-styrene, polycarbonate, low density polyethylene, polypropylene, high density polyethylene, polyamide 4.6, polyamide 6, polyamide 6/6, polybutylene terephthalate and polyethylene terephthalate. Three pattern recognition techniques of increasing complexities are applied in order to assess their suitability for the automated categorisation of polymer samples: k-nearest neighbours, various combinations of Q-and D-statistics (sometimes referred to as Soft Independent Modelling of Class Analogy, SIMCA) and Backpropagation Neural Networks. It is found that all the three methods categorise the materials into the correct polymer groups irrespective of their complexity. Methods based on the correlation structure in the data prove more beneficial than methods based on distance due to particular characteristics in the data. Best results are obtained using an adequate combination of two coefficients: one based on correlation and another based on distance.