Aside from melt flow rate, density is the most important parameter to describe the properties of polyethylene (PE) materials. PE stiffness, rigidity, and heat resistance increase with higher density. Various testing methods exist for density in PE – the most common is by density balance, measuring the buoyancy in a liquid (ASTM D792). This test is easy to perform, but the method contains a variety of measurement errors sources, such as specimen fixation corrections, temperature changes, or air bubbles within the sample pellets.

Trapped air bubbles formed during polymer pellet production result in lower density values when measured with the buoyancy method. In contrast, near-infrared spectroscopy (NIRS) is a fast analytical technique which shows a low influence on density measurement error if any air bubbles are present in the sample material.

29 different polyethylene samples with varying density were measured on the Metrohm NIRS DS2500 Solid Analyzer (Figure 1) as well as with the buoyancy method described in ASTM D792. All measurements on the DS2500 Solid Analyzer were performed in rotation to average the subsample spectra. This setup with the DS2500 large sample cup reduces influences from the particle size distribution of the polymer pellets. Data acquisition and prediction model development were performed with the software package Vision Air Complete.

The obtained Vis-NIR spectra (Figure 2) were used to create a prediction model for the density value determination in PE pellets. To verify the quality of the prediction model, correlation diagrams were created which display the correlation between the Vis-NIR prediction and primary method values received from the supplier (Figures 3–4).