Determination of asphalt cement properties by near infrared spectroscopy and chemometrics

In the present study, near infrared (NIR) spectroscopy and chemometrics techniques have been combined to predict some properties used to characterize and to classify asphalt cement grade. Principal Components Regression (PCR) and Partial Least Squares Regression (PLSR) were employed to develop linear models to predict penetration value, absolute viscosity at 60degreesC, kinematic viscosity at 135degreesC and flash point of asphalt cements. The good correlation between NIR pretreated spectra and the properties investigated, measured by values of correlation coefficients and root mean squared errors of both calibration and prediction are here discussed. It was concluded that the combination of both techniques can be used to predict properties of asphalt cement to be delivered after on-line blending of hard and soft binders stocked in refineries.