A statistical study of the effect of manufacturing methods on the mechanical properties of high-density polyethylene/layered double hydroxide composites

Abstract

Polymer-clay composites have applications in numerous sectors such as packaging, automotive manufacturing and agriculture. A primary focus of polymer composite research is to improve the performance of these composites while also reducing costs. Adding clay to the polymer can enhance the strength and stiffness of the composite. However, adding too much clay can degrade the ductility, hence reducing the usefulness of the material system.

In historical exploratory studies of polymer-clay composites, conducted at the University of Pretoria, it was observed that the mechanical properties were not enhanced as expected by the addition of clay. In fact, the variability observed in the mechanical properties was greater than the effect of increasing the clay weight loading. This could possibly be attributed to manufacturing methods. If polymer-clay composite properties are much more sensitive to manufacturing methods than has been recognised, this is concerning, since bulk manufacturing processes will generally be less tightly controlled than laboratory investigations. By gaining more insight and understanding into the effects of manufacturing variation on the final composite properties it is possible to reduce the issues which inevitably arise when scaling a manufacturing process from a laboratory to an industrial setting.

This study therefore aims to investigate the effects of different manufacturing methods on the statistical variation of the mechanical properties of polymer-clay composites. The material system studied is high-density polyethylene (HDPE) filled with layered double hydroxide (LDH), a synthetic clay. A multi-site collaborative study between University of Pretoria (UP, South Africa), Tshwane University of Technology (TUT, South Africa) and Leibniz Institute for Polymer Research (IPF, Germany) was designed. This study is fundamentally interdisciplinary, combining knowledge from polymer materials science, manufacturing, mechanical characterisation and statistics.

A statistical design of experiments was developed using the insights gained from a statistical analysis of historical data and from an in-depth systematic literature review on the effects of manufacturing variation on mechanical properties of HDPE-clay composites. The systematic literature review followed Preferred Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Statistical design of experiments is a robust method to draw reasonable conclusions about the influence of multiple contributing factors on experimental results. The design considered the influence of manufacturing method (compression and injection moulding) and site (UP, TUT and IPF) in addition to the clay type and clay weight loading material parameters. Due to the limited mould availability at the South African site, a sub-study considering the influence of the injection moulded tensile sample mould type was also included.

A statistical analysis of the experimental results indicated that the moulding method, sample mould type and site (i.e., machine variation) had a larger effect on the mechanical properties then the clay type and clay loading. The effect due to moulding method was expected as it has been documented in literature. The influence of site and sample mould type are important results from this study. The same processing conditions were used at the different sites for both the compression moulding and injection moulding comparisons. The core manufacturing process should not change even when the equipment used is not exactly the same. However, the experimental results demonstrated significant variability as a result of compression moulding on different equipment. The influence of the tensile sample mould type was not expected as the mechanical properties are normalised to the sample geometry, and the comparison was between ISO standard moulds. This raises concerns about the validity of applying experimental test results to predict bulk material performance. This thesis has therefore demonstrated the importance of the consideration of manufacturing variability in studies of polymer-clay composites.