FTIR spectral analysis of bituminous binders: reproducibility and impact of ageing temperature

This RILEM round robin study with nine participating laboratories investigated bitumen ageing, its effect on chemical properties and its reproducibility. The impact of temperature used for short-term (RTFOT) binder ageing on the combined short- and long-term (PAV) aged samples was investigated; thereby the effect of reduced mixing temperature such as those relevant for warm mix asphalt technologies on long term ageing was examined. Four 70/100 penetration graded bituminous binders from different sources were selected. In addition to the standard RTFOT temperature of 163 degrees C, two additional temperatures, 143 and 123 degrees C were used. The Fourier transform infrared spectroscopy (FTIR) analysis was carried out using an integration method which considers the area below the absorbance spectrum around a band maximum using baseline and tangential approaches. A statistical investigation into the reproducibility of FTIR spectra analysis based on the accumulated data was done. To assess the reproducibility, the coefficient of variation (CV) was taken as a benchmark parameter. Carbonyl and sulfoxide indices were calculated using different baseline correction methods and tangential and baseline integration, respectively. It was shown that the tangential method was not influenced by the applied baseline correction. However, in all considered cases, the tangential method led to significantly worse reproducibility (CVs ranging from 20 to 120%) compared to the baseline method. The sulfoxide indices calculated by both methods were not affected by the baseline correction method used. Impacts of changes in the short-term ageing temperature on short- or long-term aged samples could not be found whereas differences between different binder sources could be detected. RTFOT temperature and therefore mix production temperature had a stronger impact on the formation of sulfoxide structures than for carbonyl structures. The findings from this study show the most reproducible of all considered methods when more than one laboratory is providing FTIR data.