Title:
Pore Solution Based Rapid and Innovative Approach to Determine Fly Ash and Natural Pozzolan Dosages for ASR Mitigation
Author(s):
Anol Mukhopadhyay
Publication:
Web Session
Volume:
Issue:
Appears on pages(s):
Keywords:
DOI:
Date:
10/23/2022
Abstract:
The current tests to determine optimum fly ash (FA) dosage to suppress/mitigate alkali silica reaction (ASR) in concrete require testing for a wide range of replacement levels, which is time-consuming and impractical. Moreover, the alkali boosted test conditions of the current tests (e.g., ASTM C1567 and C1293) make these tests insensitive on detecting the effects of soluble alkalis from certain fly ashes on ASR expansion. The current empirical prediction models (e.g., chemical index and extended chemical index) to determine fly ash dosage for mitigating ASR ignore the role of fly ash soluble alkali contribution on ASR evaluation. Therefore, a rapid screening tool (ST) was developed to determine optimum fly ash dosage to mitigate ASR in concrete mixes in the present work. The ST uses an innovative approach to estimate the concrete pore solution alkalinity (PSA) at different fly ash replacement levels considering the combined effect of soluble alkali contribution from cement & water-soluble alkali (WSA) from fly ashes. A modified ASTM C114 procedure was used to measure WSA from the tested fly ashes. Additionally, a non-linear regression model was also developed to estimate WSA from fly ashes. The ST predicts optimum fly ash dosage for ASR mitigation based on the concrete PSA vs. aggregate threshold alkalinity (THA) relationship, i.e., PSA = THA to suppress ASR. Based on the evaluation of wide variety, type, and composition of fly ashes in the current study, the ST predictions of optimum FA dosage demonstrated a mean absolute error (MAE) (= ± 6-9%) compared with the current prediction models. ST was found to be effective to detect the limitations of ASTM C1567 (e.g., underestimation) for certain fly ashes and short out the ashes that need further validation by reliable concrete testing (e.g., ACCT with 90 days and C1293 with 2 years). Notably, ST demonstrated dosage predictions with higher reliability (= 87%) for the unconventional ashes.