硫酸盐
材料科学
岩土工程
复合材料
法律工程学
环境科学
工程类
冶金
作者
Qi Wang,Jiangfeng Long,Linglin Xu,Zhen Zhang,Yang Lv,Zhenghong Yang,Kai Wu
标识
DOI:10.1016/j.conbuildmat.2022.128436
摘要
• A green stabilizer was confirmed in improving the properties of softy soil. • Experimental and numerical study was done on sulfate resistance ability of stabilized soil. • Less ettringite and gypsum were formed in the green stabilizer treated soil. • Numerical model was proposed to predict crystallization pressure by ettringite and gypsum. Developing an alternative for traditional cement in soil stabilization has long been an attractive topic in order to improve the overall performance and reduce cement consumption. Considering the long term safety, a novel green soil stabilizer mainly consisting of industrial wastes, i.e., ground granulated blast furnace slag, steel slag, desulfurization gypsum was evaluated by a series of tests involving the unconfined compressive strength (UCS) under wet-dry cycling and sulfate attack. X-ray diffraction and backscatter scanning electron microscope were performed to reveal the deterioration mechanism. The results show that the addition of green stabilizer shows an increment effect on UCS of soil compared with neat cement. The UCS after 15 cycles of wetting–drying and 30 days of external sulfate attack are both highly retained for the soil stabilized with green stabilizer, while the porosity and expansive gypsum and ettringite are less. A computational model was implemented to investigate the deterioration by determining the crystallization pressure contributed by ettringite and gypsum, and then compared with the tensile strength obtained in experimental works. It indicates that the internal stress in the soil stabilized with green stabilizer is less than that with cement, and the former one show a better durability under sulfate attack. The method and materials presented in this work is helpful for the design and application of stabilized soil considering the complex environmental effects.
科研通智能强力驱动
Strongly Powered by AbleSci AI