SUBSTANTIATION OF USING GEOPOLYMER IN DEEP MIXING METHOD: AN EXPERIMENTAL INVESTIGATION

Abstract

Soft soils encounter a serious problem for the stability of the superstructure, and this has been a crucial issue since the beginning of the history. Among various ground improvement options, the Deep Mixing Method is one of the promising methods that can alter the soil matrix at close spacing and resemble pile-like responses that can provide sufficient reactions to the superimposed loads. Cement is generally mixed with problematic soils and has proven to be effective in many applications; however, environmental concerns have become substantial in cement-treated ground improvement options. Alternately, waste materials, mostly Fly Ash or Slag, have been considered as replacements of cement for possible applications. Fly Ash is being produced in high amount globally as an industrial byproduct; its disposal is a big concern and, hence, attracted concern of the geotechnical experts for its utilization especially in ground improvement. To date, fly ash has been mostly considered for shallow ground improvement, while its justification in deep mixing is seldom found in technical writings. The available applications mostly focus on treating granular materials. The present study, as such, aims to explore the possibility of utilizing fly ash in the deep mixing method. To enhance the bonding, a chemical additive, commonly known as an alkali activator, has been added to the fly ash to form a geopolymer that has been added with the parent soils at different fly ash proportions ranging from 10% to 25% of the dry weight of soils. In addition to the variation of geopolymer, three different moisture levels of soils, at the liquid limit and on both sides of the liquid limit, have been considered in this research. A number of samples of each considered case has been prepared at three different curing ages: 14 days, 28 days, and 56 days, in order to find the vi effectiveness of the fly ash-based geopolymer treated deep mixing method in ground improvement. The Unconfined Compressive Strength has been found to increase with the increase of fly ash content in all cases. The increase of strength has been found to depend on the optimum dosage of both alkali activator and precursor binder (fly ash). Moisture sensitivity has been seen as distinctive, and geo polymerization of soil in the wet phase has no portrayed considerable strength development. In this study, the activator-to-precursor binder ratio of 0.5 with fly ash contents of 20-25% has shown the highest strength gain in approximately all scenarios. Based on the discrete datasets, predictive models, keeping the aim of estimating preliminary compressive strength, have been developed considering the non-linearity and influence of different properties. Several prediction models (both equation based and Machine Learning models) have been developed, and model performances have been compared with the experimental data. The highest accuracy has been observed from Exponential Gaussian Process Regression Model with an R2 value of 0.97 which indicates that the model fits excellently with an ability to explain 97% of the experimental data. Furthermore, Mineralogical bond formation of stabilized soil has been observed from X-ray diffraction (XRD) pattern and Scanning Electron Microscope (SEM) image. Densification and void reduction due to geo polymerization process in soil-geopolymer mix as well as the surface texture having no cracks in micro level have been observed from SEM image. XRD pattern analysis provided the information about the compounds present in the mix by which susceptible elements can be identified. Overall, this research provides a base for the use of waste-based geopolymer in deep mixing methods in alluvial deposition.