.png){kind=icon}
.png){kind=icon}
Return
Ambient-Cured alkaline geopolymer concrete with industrial by-products and micro-binders: Mix optimization and TiPNN-Based strength prediction
.png){kind=icon}
PRE
AI
N
T
S
DOI：10.1016/j.compositesa.2026.110138.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Geopolymer concrete (GPC) is a sustainable alternative to ordinary Portland cement (OPC) concrete because of its lower carbon footprint and enhanced durability. However, the absence of standardized mix design guidelines and limited early-age strength data for ambient-cured systems continue to restrict its practical adoption, motivating the present study. This study investigates ambient-cured alkaline geopolymer concrete through nine mix designs containing fly ash, GGBS, metakaolin, micro-silica, and micro-alumina. Compressive strength (CS), elastic modulus, water absorption, chloride penetration, sulphate resistance, and acid resistance were evaluated through standardized laboratory testing. The optimized F60G30MS10 mix achieved the highest 7-day compressive strength (CS) of 48.31 MPa, the lowest water absorption of 2.77%, and the lowest RCPT charge passed of 823C. SEM analysis confirmed that the inclusion of micro-silica promotes a denser, more uniform geopolymeric matrix, directly explaining the superior mechanical and durability performance. A Temporal Inductive Path Neural Network (TiPNN) model was developed as a supplementary predictive tool, trained on the experimental dataset using Leave-One-Out Cross-Validation (LOOCV), achieving an R2 of 0.9544, an RMSE of 1.306 MPa, and an MAE of 0.992 MPa for CS prediction. These findings highlight the potential of optimized ambient-cured GC as a sustainable and high-performance material for modern infrastructure.
Journal
IF:
8.9
Papers:
8.6K
Citations:
4.5W
Organization
Cited Papers