Strength characteristics of spent coffee grounds and oyster shells cemented with GGBS-based alkaline-activated materials

This study examined the use of by-product materials such as spent coffee grounds (SCGs), oyster shell (OS), and ground-granulated blast furnace slag (GGBS) for geotechnical applications. To improve the granular skeletons of the highly compressive particles of SCG, the OS were added to a mixture in forms of grinding powder (OS.P) or crushed-granular (OS.G) acting as fill factor in terms of physical interaction, whereas GGBS was only precursor material for alkali-activated materials (AAMs). The OS.P and OS.G were employed at various OS:SCG ratios, by weight. The mixtures were cemented using GGBS-based AAMs activated by a sodium hydroxide solution (SH) and a mixture of sodium hydroxide and sodium silicate (SS). From the test results, several factors that affected the strength development of the samples have been observed, such as the curing time, type of alkaline solution, temperature variation, as well as form and ratio of OS. At the same curing conditions, the SS solution exhibited better strength compared with SH as significant contents of sodium (Na2O) and silicate (SiO2) were provided for the mixture. During the curing period, the temperature variation strongly affected the strength development of the samples. Generally, the strength and stiffness of the samples were improved by the incorporation of SCG with OS.P or OS.G, whereas the slake durability indices were maintained at very high to extremely high values, regardless of the OS type. The maximum unconfined compressive strength (UCS) value reached approximately 2.8 MPa, with optimal ratios of OS.P or OS.G of 0.3 or 0.5, respectively, compared with the SCG weight. At a given ratio, the blend mixture of OS.G and SCG generally exhibited better stiffness and a slightly lower unit weight than OS.P. The mineralogical and microstructure analyses revealed that there was no chemical reaction among SCG, OS.P, and OS.G particles, indicating that these materials act as fillers. In addition, the produced C–S–H gel acts as the dominant binder in this study. Moreover, the excessive addition of OS.P or OS.G caused negative results in strength and durability due to the poor bonding effects between the SCG and OS particles in granular skeletons. Thus, the addition of 50% of OS.G compared with the amount of SCG, by weight, was proposed as the optimal ratio for practical applications.