Real-time monitoring and accurately understanding the continuous occurring psychical-chemical characterization during the hardening process of alkali-activated materials present a formidable challenge. In this study, the early-age reaction process of alkali-activated slag-waste brick powder paste (ASWP) was in-situ and continuous monitored by using acoustic emission (AE) system and temperature. A novel machine learning method was employed to categorize the evolution of AE signals into three distinct stages: reaction acceleration, reaction stabilization and volume shrinkage stage.During reaction acceleration stage (0-2h), the waste brick powder (WBP) increased AE counts rate (4000-7300 counts/h), with RT values exceeding 200μs, indicating the settlement of WBP. Additionally, WBP reduced the high r-value AE activity in volume shrinkage stage, effectively mitigated the drastic volume shrinkage, and delayed its occurrence time by approximately 4 hours. Semi-quantitative analyses of FTIR and TG-DTG indicate that WBP reduced gel phase formation in the early-age stage, decreasing the relative areas in Q2 site by 16% before 36 hours. However, WBP positively impacted ASWP hydration at long ages, increasing Q2 site d areas by 60%-100%. AE technique offers a new perspective for in-depth understanding of the reaction mechanism and microstructural development of alkali-activated materials in situ and continuously.
Read full abstract