Opening the black box of loss on ignition of agricultural-based ashes as SCM: Carbon type characterization and its effect on cement hydration and rheology

Abstract

[EN] Despite decades of research on rice husk ash (RHA) and sugarcane bagasse ash (SCBA) as pozzolanic materials, no study has systematically addressed the nature of residual carbon in these ashes. Several standards limit the loss on ignition (LOI) of pozzolans to <= 6%, a criterion that excludes many residual RHAs and SCBAs from potential use. This study demonstrates that LOI is neither a reliable indicator of the actual carbon content nor of its influence on cement hydration. A comprehensive characterization combining FTIR, Raman spectroscopy, and thermogravimetric analysis coupled with mass spectrometry identified three distinct forms of residual carbon: volatile organic fraction (VOF), amorphous carbon, and graphitized carbon. Amorphous carbon, the dominant carbon fraction in agro-industrial ashes, was associated with hydration acceleration, whereas graphitized carbon showed a negligible influence on hydration. In contrast, VOF, composed primarily of residual lignin and cellulose, was identified as the key factor responsible for hydration retardation and rheological modification. The presence of hydroxyl and carboxyl functional groups, analogous to those found in lignin/cellulose-based dispersants, explains the delayed hydration kinetics and the improved paste workability. This important identification of amorphous carbon was only possible through the use of chemical methods, which can discriminate reactive amorphous silica from alumina, which contrasts from Rietveld-based amorphous phase quantification, where amorphous carbon and vitreous inorganic phases are incorporated into the total amorphous fraction estimated. Multiple linear regression analysis confirmed the statistically significant influence of volatile organic fraction and specific surface area on both hydration kinetics and rheological parameters, revealing antagonistic effects whose combinations lead to distinct behaviors.