Red mud (RM) as a highly alkaline solid waste generated during alumina production presents significant environmental challenges due to its massive annual output and stockpiling, including land occupation and pollution. The utilization of RM in combination with other solid wastes or minerals to produce geopolymer cementitious materials becomes one of the effective approaches for its resource recovery. However, Bayer-processed RM exhibits a low Si/Al molar ratio and a poor reactivity, making it difficult to form a stable aluminosilicate polymer structure. Most RM-based geopolymers require mechanical/thermal activation pretreatment or high-temperature curing to enhance early strength. The existing studies typically involve either low RM incorporation ratios or result in a low geopolymer strength at high RM dosages. Based on previous research and the concept of multi-source solid waste synergy, this study was to introduce highly reactive ground granulated blast-furnace slag (GGBS) and calcium-rich carbide slag (CS) into an RM-Class Ⅱ fly ash (FA) geopolymer system. The influence of GGBS and CS composite dosages on the mechanical properties of the geopolymer was investigated. Furthermore, the microstructure of the geopolymer was analyzed from multiple perspectives to elucidate the synergistic reaction mechanism and strength development under the co-incorporation of GGBS and CS.

In this study, RM, FA, GGBS, and CS were used as raw materials at a RM/FA mass ratio of 1.0/1.0. A mixed alkaline activator solution consisting of NaOH solution (NH) and water glass (WG) was used at an NH/WG mass ratio of 1.0/2.5. The NaOH solution had a molarity of 10 mol/L, and the liquid-to-solid ratio (L/S) was 0.65. Different dosages of GGBS and CS were used as the experimental variables.

The RM, FA, GGBS, and CS were manually mixed for 30 s, followed by blending the solid powder mixture with alkaline activator in a cement paste mixer for 3 min. The resulting slurry was then cast into triple molds with internal dimensions of 70.7 mm × 70.7 mm × 70.7 mm. The specimens were cured at 20 ℃ and 95% relative humidity for 24 h before demolding, after they continued to cure under standard conditions.

The compressive strength tests were conducted at different curing ages of 7, 14, 28, 45 d, and 60 d. The collected fragments taken from the center of the specimens were stored in vials and immersed in anhydrous ethanol to halt further reaction. The chemical composition, microstructure, and mineral phases on the geopolymer's strength were determined by X-ray diffraction (XRD), Fourier-transform infrared spectroscopy (FTIR), and scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS). Prior to testing, the samples were dried at 60 ℃ for 24 h and ground for subsequent characterization.

All the specimens exhibit a strength development trend with increasing curing age. Among them, the geopolymer RFGC-1 with CS as sole additive shows the lowest strength, reaching only 1.8 MPa at 60 d. In contrast, RFGC-5 with GGBS as sole additive demonstrates the highest strength among the five formulations, achieving 11.6 MPa at 60 d, indicating that GGBS is more effective than CS in enhancing the strength of RM-Class Ⅱ FA-based geopolymers.

For the samples RFGC-2, RFGC-3, and RFGC-4 incorporated both GGBS and CS, the 60-d compressive strengths are 5.8, 8.8 MPa, and 6.5 MPa, respectively. The strength initially increases and then decreases with increasing GGBS content and decreasing CS content, indicating that the relative proportions of GGBS and CS must be carefully controlled to optimize the geopolymer strength. At a total additive content of ≤12%, the RM-FA geopolymer with GGBS alone exhibits a superior strength, compared to those with CS alone or GGBS/CS combinations.

The microstructural characterization reveasl that, in addition to the original mineral phases (i.e., hematite (Fe₂O₃) and quartz (SiO₂)), the primary cementitious products in the geopolymers include calcium aluminate hydrate (C-A-H), calcium silicate hydrate (C-S-H), calcium aluminosilicate hydrate (C-A-S-H), ettringite (AFt), aluminosilicate oligomers (—Si—O—Al—Si—O—)_n, and hydrotalcite. For RFGC-1 (CS only), the microstructure is relatively loose, with the limited formation of amorphous C-S-H and insufficient conversion into C-A-S-H or (—Si—O—Al—Si—O—)_n, resulting in a low strength. As the GGBS content increases and CS decreases, intermediate products (C-S-H, C-A-H) further react to form C-A-S-H, AFt, and (—Si—O—Al—Si—O—)_n. The gel phases envelop adjacent particles, forming interconnected coatings and enhancing the aluminosilicate and SiO₄ tetrahedral networks, thereby strengthening the geopolymer matrix.

The incorporation of GGBS alone was more conducive to strength development in the RM-Class Ⅱ FA based geopolymers, compared to CS alone. In the case of co-incorporating GGBS and CS, the primary cementitious products included C-A-H), C-S-H, C-A-S-H monomers, as well as ettringite, aluminosilicate oligomers (—Si—O—Al—Si—O—)_n, and hydrotalcite, which collectively contributed to the strength of the geopolymers. The addition of GGBS and CS to the RM-Class Ⅱ FA system modulated the contents of Ca, Si, and Al in the composite system, significantly influencing geopolymer strength development. At a higher content of Ca, C-S-H became the dominant product. Conversely, at a higher content of Si or Al, these components first dissolved in the alkaline environment to form gels such as C-S-H, C-A-H, and C-A-S-H. These gels subsequently underwent dissolution and polycondensation under strong alkaline conditions, forming long-chain polymers like Si—O—Al—O and Si—O—Al—Si—O. For overtime, water was expelled, leading to coagulation and polymerization, ultimately resulting in the formation of dense and stable geopolymer structures.