Performance of soft ground improved by floating bottom ash columns

Abstract

The stone column approach has been utilized over the past 60 years to stabilise the soft ground by partially replacing the compressible soil with more stable materials such as aggregate and sand. In practice, the granular material with a diameter of 20 mm to 75 mm is used as column filler material. In this research, the bottom ash material was used as a substitute material in the granular columns instead of the natural aggregate. A series of small-scale 1g physical modelling tests were carried out to investigate the behaviour of soft clay after being treated with a group of bottom ash columns beneath a rigid footing. A parametric study was performed to examine the effect of key design parameters such as; area replacement ratio and height penetration ratio on the bearing capacity and settlement characteristics. Whereas, a total of 19 physical model tests were conducted. A set of strain-controlled loading tests were performed to determine the ultimate bearing capacity, while a dead load method was adopted to examine the settlement characteristics. The physical modelling tests were carried out on the untreated and treated ground with bottom ash columns. Three area replacement ratios of 13 %, 20 %, and 26 % and three-column height of 50 mm, 100 mm, and 150 mm were adopted. The deformation and failure mechanism of the ground model was observed through capturing images during the loading test. Then, the Particle Image Velocimetry technique (PIV), GeoPIV: MATLAB software was used to analyse the collected images. The results clearly proved that the bearing capacity of the soft ground improves significantly with the incorporating of the bottom ash columns. Moreover, a higher area replacement ratio with longer columns demonstrated better load capacity enhancement. Also, it was found that the magnitude of total settlement reduced as the area replacement ratio increased. In addition, the total settlement of the reinforced ground showed a decreasing trend with the increase of the column height. In parallel to the physical modelling tests, three-dimensional numerical analysis was conducted via Plaxis 3D foundation software. This method was adopted to validate the experimental outcomes, since it is more economical and takes less time to complete in comparison to the full-scale model. Two different types of constitutive models were used to simulate the soft ground and bottom ash columns namely; the Soft Soil model and Mohr-Coulomb model. Comparisons between the results obtained from the physical model test and numerical simulation were made considering the different area replacement ratios and column height penetration ratios. The finite element analysis results were used to verify the experimental findings and a good agreement was found between the two methods since the difference is less than 20 % and is considered acceptable. The results revealed that the area replacement ratio and column height penetration ratio significantly influenced the overall performance of the treated ground. Whereas, the stiffness, load-bearing capacity, and settlement characteristics of the reinforced ground improved by increasing the area replacement ratio and column height penetration ratio; a 172 % enhancement in bearing capacity was attained with 26 % area replacement ratio and 0.75 column penetration ratio. A similar observation was obtained for LECA- treated ground under a constant rate of loading. The relationships between ultimate bearing capacity and area replacement ratio or column height to diameter ratio were plotted. From the relationships, six proposed design equations were developed for practical use to estimate the ultimate bearing capacity of the reinforced ground under a rigid footing. Another six design equations were also established to predict the normalised bearing capacity factor using the same parameters (area replacement ratio or column height to diameter ratio). Furthermore, the rationality of the proposed design equations was successfully verified using the finite element results.