In recent years, compared with the traditional portland cement, environmentally friendly geopolymers have gained more attention as construction materials. This paper considered volcanic ash (VA) and ground granulated blast furnace slag (GGBFS) in different percentages (0%, 3%, 7%, and 10%) as a replacement for the conventionally used portland cement to stabilize sandy soils. NaOH and in different concentrations (4, 8, and ) and alkali to binder ratios (1, 1.5, 2, and 3) were used as alkali activator solutions to build new geopolymers. Samples were cured at both ambient and oven temperatures and for 1, 7, and 28 days. Unconfined compressive strength (UCS) of samples then was evaluated. Two predictive approaches, artificial neural network (ANN) modeling and the evolutionary polynomial regression technique (EPR), were applied to model UCS of geopolymerized sand samples. Regarding the high value of the coefficient of determination of the proposed ANN, 97%, and acceptable prediction errors, RMS error of 0.0439 and MAE of 0.0336, an 8-5-10-1 ANN was introduced as a more accurate tool for the prediction of UCS. Next, three-dimensional parametrical studies investigated the effects of simultaneous changes in alkali solution, binder, and curing condition parameters on UCS values of geopolymerized samples. Sensitivity analysis based on the cosine amplitude method introduced the Si/Al ratio as the parameter most affecting and VA content as the parameter least affecting the compressive strength of samples. Results were analyzed further using pH and electrical conductivity tests and interpreted based on microstructural investigations using scanning electron microscopy (SEM) images and X-ray diffraction analysis.
1 February 2019 in Tunnelling and Underground Space Technology
The proper design of protective structures may start from improving the characteristics of soils. In order to obtain reasonable safety criteria, several research studies have recently been dedicated to enhancing complex civil engineering structural systems with the use of nanotechnology. Thus, the following paper investigates the effect of nanospheres, including nanosilica (nano‐SiO2) and nano zinc oxide (nano‐ZnO), on lime‐stabilized high‐plasticity clay soil. For this purpose, unconfined compressive strength (UCS) and California bearing ratio (CBR) tests were performed on samples. The results showed that the use of the selected nanospheres greatly increased the UCS of the samples compared to untreated soil. The UCS value of samples containing 6% lime and 1.5% nano‐ZnO after 28 days of treatment increased by 5‐fold compared to the UCS of untreated samples. In addition, the samples containing 6% lime and 2% nano‐SiO2, with similar curing conditions, experienced a 5.3‐fold increase in their UCS value compared to the untreated samples. These compounds were considered as the optimal amounts and showed the highest mechanical strength in both UCS and CBR tests. The same trend was achieved in the CBR test, in which the CBR value for the optimal mixtures containing nano‐ZnO and nano‐SiO2 was 14.8 and 16.6 times higher than that of high‐plasticity clay soil, respectively. Finally, the results obtained from scanning electron microscopy (SEM) analysis revealed that the nanospheres caused a dense and compact matrix to form in the soil, which led to the enhancement of the mechanical strength of the treated samples.
1 March 2014 in Soil Dynamics and Earthquake Engineering