多源沥青基因特征及其性能构效机制分析

Analysis of genetic characteristics and structure-performance mechanisms in multi-source asphalt

摘要:

为揭示多源基质沥青自身“基因”特征与其性能之间的内在构效关系,选取10种不同油源的常用基质沥青,通过对基质沥青的宏观性能（三大指标等）、化学组分（饱和分、芳香分、胶质、沥青质含量）及元素组成进行系统分析,结合组成分试验与相关性分析,揭示了胶质沥青质比I_R/A与135 ℃旋转黏度极强相关,而胶质R/芳香分A只与老化后软化点存在强负相关关系等等。进一步结合凝胶渗透色谱测试结果,发现分子量分布特征与四组分间协同作用显著影响沥青流变行为,其中重均分子量与老化后延度存在强的负相关性。分布宽度指数与针入度指数PI、老化后软化点、标准黏度和135 ℃旋转黏度存在强或极强的正负相关性。这表明分子量分布越宽,沥青的温度敏感性和黏度越高,而延展性降低。通过回归分析进一步验证了四组分、分子量参数对宏观性能的影响,并得出了这些“基因”特征与沥青性能之间的构效模型,通过“基因”特征指标预测相应的性能指标。研究结果为将来实现沥青材料的精准设计与性能预判提供了理论依据。

Abstract:

To reveal the intrinsic structure-property relationships between the inherent "genetic" characteristics of multi-source base asphalt and its performance, this study selected ten commonly used base asphalts from different crude oil sources. Through systematic characterization of the macroscopic properties (such as the three major indicators), chemical composition (saturates, aromatics, resins, and asphaltene content), and elemental composition of the base asphalts, combined with compositional analysis and correlation analysis, it was revealed that the resin-to-asphaltene ratio (I_R/A) exhibits a very strong correlation with the rotational viscosity at 135 ℃, while the resin-to-aromatics ratio (I_R/A) only shows a strong negative correlation with the softening point after aging. Furthermore, combined with gel permeation chromatography test results, it was found that the synergistic effect between molecular weight distribution characteristics and the four fractions significantly influences the rheological behavior of asphalt. Among these, the weight-average molecular weight demonstrates a strong negative correlation with ductility after aging. The polydispersity index exhibits strong or very strong positive or negative correlations with the penetration index (PI), softening point after aging, standard viscosity, and rotational viscosity at 135 ℃. This indicates that a broader molecular weight distribution leads to higher temperature susceptibility and viscosity of asphalt, while reducing its ductility. Through partial least squares regression analysis, the influence of the four fractions and molecular weight parameters on macroscopic properties was further validated, and structure-property equations between these "genetic" characteristics and asphalt performance were established. To some extent, the constructed structure-property relationship model allows for the prediction of corresponding performance indicators based on genetic characteristic parameters. This study provides a theoretical foundation for the precise design and performance prediction of asphalt materials in the future.