As a commonly used
filler, CaCO3 frequently
finds its
way into recycled polypropylene (rPP) as a contaminant during the
mechanical recycling process. Given the substantial impact of CaCO3 on the properties of PP materials, close monitoring of their
content is important to ensure the quality of rPP. In the present
work, Raman spectrometry was employed to develop a rapid, accurate,
and convenient method for determining CaCO3 content in
rPP. Partial least-squares (PLS) regression was used to construct
prediction models. Various spectrum pretreatment methods, including
multivariate scatter correction (MSC), standard normal variate transformation
(SNV), smoothing, and first derivative, were investigated to improve
the model performance. In independent validation, the optimal PLS
model reached an R
2 of 0.9735 and a root-mean-square
error of prediction (RMSEP) of 2.7786 CaCO3 wt %. Furthermore,
linear and second-order polynomial regressions, utilizing the intensity
ratios of characteristic CaCO3 and PP Raman peaks, were
conducted. The most effective quadratic regression curve demonstrated
superior independent validation performance with an R
2 of 0.9926 and an RMSEP of 1.6999 CaCO3 wt
%. Validation with recycled PP samples confirmed that the quadratic
regression was more accurate and reliable to quantify CaCO3 in rPP. The observed quadratic relationship between the CaCO3 and PP Raman peak intensity ratio and the CaCO3 wt % can be attributed to the significant difference in the densities
of the two components. The outcomes of this research will help to
facilitate the proper recycling of PP materials.