Experimental investigation and statistical analysis of performance of ceramic fiber-reinforced warm asphalt mixtures containing reclaimed asphalt pavement

Abstract

Asphalt recycling can help conserve natural resources while reducing associated costs. However, the production of hot mix asphalt (HMA) with reclaimed asphalt pavement (RAP) generates a high heat rate, leading to the stiffening of the asphalt binder and the release of toxic gases. Warm Mix Asphalt (WMA) is a cost-effective technique that softens and delays the aging of the asphalt binder while producing asphalt mixtures at lower temperatures, resulting in decreased energy usage and harmful pollutants. However, increasing the RAP content in WMA mixtures could have a detrimental impact on several asphalt mixture properties, such as fatigue and moisture resistance. Consequently, increasing the RAP content in WMA mixes could require devising a strategy to offset the drawbacks of RAP material. Fibers are recognized as one of the additives that are incorporated directly into the mixes and improve the fatigue and moisture resistance of asphalt mixtures. Therefore, this study is conducted to evaluate the impact of ceramic fibers (CFs) on the performance of WMA-RAP mixtures. The response surface methodology (RSM) was employed to find out how the CFs changed the responses of WMA-recycled asphalt mixes at low and intermediate temperatures. These responses included Marshal stability, rutting resistance, moisture susceptibility, and fatigue life resistance. RSM was used to find the best amounts of factors like CFs, WMA additive (Sasobit), and RAP using central composite design (CCD). The results of the analysis of variance (ANOVA) showed a significant interaction between the CFs, RAP, and Sasobit. The CFs significantly improved every studied attribute, particularly the fatigue life at 5 °C and 20 °C. The excellent value of the adjusted coefficient of determination (adjusted R2) of all responses indicates a excellent correlation between the model and experimental results. The outcomes of the validation test demonstrate that every response has a percentage error of less than 5 %, exhibiting strong agreement and the accuracy of the model.