Experimental study of strength behavior of soft soil stabilised with alkali activated copper slag

Abstract

In this work, Unconfined compression test (UCS) were used to study the effects of an industrial by-product copper slag (CS) on the geotechnical properties of soil in treated and untreated condition. Soil specimens were created with five different percentages of CS (i.e., 0 %, 10%, 20%, 30%, and 40%) by weight to that of dry soil mass. The samples were compressed into a cylindrical specimen and then kept for processing for 7, 14, and 28 days of curing. The presence of CS with alkali activator had a substantial effect on the unconfined compressive strength (UCS) of the given soft soil according to the test results. The effects of the CS on the UCS were more noticeable in 10 M alkali activated specimens as compared to unstabilized specimens. Furthermore, in the alkali activated specimen, the UCS value was increased by more than 5times as the CS content raised by up to 30%. Addition of copper slag reduces the optimum moisture content (OMC) and also increases the maximum dry density (MDD) of the soil because of its higher specific gravity. In addition, an artificial neural network (ANN) model created with eight input parameters: CS content, Alkakli Activator (AA) content, dry density, moisture content, plastic limit, liquid limit, clay content, and curing age. The optimum design for predicting the compressive strength of mixes was determined by a neural network consisted of 10 neurons in the ANN hidden layer, with good similarity between the observed or predicted data and test results. The results of this study showed that the suggested model may be used to estimate the unconfined compressive strength of the CS-stabilized soft soil.