Study on aging behavior and prediction of SBS modified asphalt with various contents based on PCA and PLS analysis

• Published in: Construction & building materials 2020-12, Vol.265, p.120732, Article 120732
• Main Authors: Xu, Meng, Zhang, Yuzhen, Zhao, Pinhui, Liu, Cheng
• Format: Article
• Language: English
• Subjects: Aging process; Chemical structure; IR analysis; PCA; PLS; SBS modified asphalt
• ISSN: 0950-0618
• DOI: 10.1016/j.conbuildmat.2020.120732

[Display omitted] •SBS 4.5 had higher aging susceptibility than SBS 3.0 and 3.8.•Peak area@1455cm-1, ICO, G” and penetration distinguished most of tested properties in PCA analysis.•PLS algorithm based on FTIR analysis results can predict rheological properties of asphalt effectively during aging time. The present study was meant to investigate aging behavior of different SBS content modified asphalt pre and post aging. The conventional, rheological properties, chemical structure of aged matrix and SBS modified asphalt at various degrees were tested in this study. Principle component analysis (PCA) was used to eliminate the mutual influence between the evaluation indicators. Partial Least Squares method (PLS) was used to build model for predicting aging behaviors of SBS modified asphalt. Drawing the results of experimental tests and statistical analysis, main conclusions are as follows: Carbonyl groups of matrix and different SBS modified asphalt increased obviously with fatty side chain decreased during aging process according to X-ray photoelectron spectroscopy (XPS) results and Proton nuclear magnetic resonance tests (H-NMR). According to carbonyl and sulfoxide index results of the Fourier Transform Infrared Spectroscopy (FTIR) analysis, rheological and conventional properties aging index results, SBS modified asphalt (4.5% SBS addition) (SBS 4.5 for short) showed lower high-performance and anti-aging properties than SBS modified asphalt (3.0% SBS addition) (SBS 3.0 for short) after the RTFOT and PAV aging. Based on PCA results, properties peak area@1455 cm−1, carbonyl index ICO, viscosity modulus G″ and penetration distinguished most in asphalt aging behaviors. The PLS aging simulation model had good robustness for predicting the physical and rheological properties changes during the SBS modified asphalt aging process. From measurement IR analysis data, the expected rheological property δ can be predicted through PLS model best with correlation coefficient >0.90.