Header menu link for other important links
X
Experimental Investigations and Prediction of Thermal Behaviour of Ferrosialate-Based Geopolymer Mortars
Yeddula B.S.R,
Published in Springer Science and Business Media LLC
2020
Volume: 45
   
Issue: 5
Pages: 3937 - 3958
Abstract
This paper studies the thermal behaviour of ferrosialate geopolymer mortars. This is done by monitoring various factors influencing the strength gain/loss, weight loss, enthalpy changes, physical and chemical transformations in the ferrosialate geopolymer structure using TG/DT analysis. This study proposed a novel predictive equation for estimating this parameter with the help of gene expression programming (GEP). Fly ash is used as a raw feed for sialate geopolymer, and red mud along with fly ash is used for ferrosialate geopolymer. Till 200 °C, oven-cured samples showed maximum strength results. Whereas in later stages, i.e. after exposure to 400 °C, ambient cured samples surpassed the former by 4.14%. Development of broad amorphous hump in the XRD patterns, presence of thicker geopolymer structure in the SEM images for 400 °C samples, an exothermic peak in the DTA curves at 400 °C and increment in the compressive strength up to 400 °C exposure, all pointing to a conclusion that elevated temperature-favoured ferrosialate geopolymer formation till 400 °C. After exposure to 800 °C, maximum strength loss of 68.57% and 30.3% is observed for sialate and ferrosialate samples dehydroxylation, recrystallization, and melting of unreacted particles are the reasons for diminishing the strength at elevated temperatures. An equation using GEP model (r2 = 0.913) having nine genes is proposed that can predict the residual compressive strength of ferrosialate geopolymer mortars. Though this model is for ferrosialate geopolymer, a similar technique can be easily adapted to other types of geopolymers. © 2020, King Fahd University of Petroleum & Minerals.
About the journal
JournalData powered by TypesetArabian Journal for Science and Engineering
PublisherData powered by TypesetSpringer Science and Business Media LLC
ISSN2193-567X
Open AccessNo