Performance and Prediction Modeling of Environmentally Friendly Straw Retarder for Geopolymer Cement

Le résumé fourni par la source

To address the balance between sustainability and performance in conventional geopolymer retarders, this study innovatively employs a dual-waste valorization strategy using mineral powder activated corn straw as a high-efficiency retarder. Under invariant grinding conditions with a 30-min duration, corn straw–based retarders incorporating silica fume exhibited significantly higher specific surface areas compared with those modified by fly ash. At an optimized silica fume-to-corn straw ratio of 1∶4 and a composite retarder dosage between 2% and 3%, carbohydrates derived from corn straw effectively regulated the setting behavior via Ca2+ complexation, complying with current Chinese standards. Simultaneously, silica fume improves straw dispersion and supplies active silicon sites, contributing to a compressive strength of 87.3 MPa. Furthermore, a back-propagation neural network (BPNN) model was developed for the first time to optimize corn straw–based retarders, achieving prediction errors below 8% for both setting time and compressive strength. The model identifies straw composition as the dominant factor governing retardation effectiveness-a finding that transcends conventional trial-and-error approaches and facilitates the rational design of waste-sourced admixtures. This strategy simultaneously delivers environmental and economic benefits.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.