As shown in Table 5, compared to the unreinforced RA, the cement-silica fume slurry reinforcement increased the apparent density by 0.39%, raised the water absorption rate by 1.9%, and reduced the crushing value by 4.5%. This improvement is attributed to the finer particle size of silica fume, which enables more effective filling of pores and cracks.

SEM-EDS tests were conducted on RA reinforced with a cement-silica fume slurry at a water-to-cement ratio of 1.0. Figures 2 and 3 present SEM images of the RA surface before and after reinforcement. The RA surface is covered with a significant amount of hardened mortar, exhibiting an uneven morphology with numerous pores, which results in high water absorption, low density, and relatively weak strength [30]. After reinforcement with cement-silica fume, the aggregate structure becomes denser, displaying a crystalline skeletal structure. Hydration products of cement and silica fume cover the RA surface, filling the voids. The EDS results for RA before and after reinforcement are shown in Fig. 4. Tables 5 and 6 provide the elemental contents of RA before and after cement-silica fume reinforcement. Due to the addition of silica fume and cement hydration reactions, the contents of Ca, Si, and S in the RA significantly increase.

By comparing the results before and after reinforcement, it is found that the silicon content in the aggregate significantly increases after reinforcement, rising by 52%, and the calcium content also significantly increases, rising by 83%. Based on their mass ratios, it can be inferred that a large amount of calcium silicate is generated on the surface of the treated aggregate, filling the tiny cracks and pores in the aggregate [31–33].

As shown in Fig. 5, when the RA replacement rate is consistently maintained at 70% or 100% and regardless of whether the RA undergoes cement-fly ash modification, increasing the SSR dosage from 9% to 15% results in a continuous increase in the compressive strength of the concrete at 7d, 28d, and 60d. This trend is particularly pronounced in promoting the long-term strength at 28d and 60d in cement-fly ash modified RAC. When the SSR dosage is fixed and RA is not modified with cement-silica fume, increasing the RA replacement rate from 70% to 100% significantly reduces compressive strength at all ages. However, in cement-fly ash-modified RAC, the adverse effect of 100% replacement on strength was effectively mitigated. Combined with a 15% SSR dosage, the 60-day compressive strength of 100% replacement concrete still reached 32.2 MPa, fully demonstrating the synergistic strengthening effect between cement-fly ash-treated RA and SSR cementitious materials. Furthermore, the compressive strength of TR-70-15 surpassed that of cement-based RAC processed under identical conditions, demonstrating that this synergistic effect enables solid waste-based concrete to achieve superior mechanical properties.

As shown in Fig. 6, with the increase in the incorporation amount of SSR, the macroscopic pores inside the RAC gradually decrease, and a more uniform failure area is formed on the flexural fracture surface. From the flexural strength results in Fig. 7, SSR, as a cementitious material composed of steel slag, slag, and fly ash, when its incorporation amount increases, more active components such as the glassy phase in the slag and calcium-magnesium minerals in the steel slag participate in the hydration reaction, thus providing a material basis and structural support for the development of flexural strength [34–35]. It can be seen from the strength growth trend from 7d to 60d that with the prolongation of the curing age, the flexural strength exhibits significant time dependence. The flexural strength of TR-70-15 reaches 6.5MPa at 60d, which is more than 2.6 times higher than that at 7d.

In the group of unmodified RA, the flexural strength of concrete with a 100% replacement rate is generally slightly lower than that with a 70% replacement rate. However, in the TR series modified with cement-silica fume, the gap in flexural strength between the 100% replacement rate and the 70% replacement rate is significantly narrowed. Moreover, with a 5% SSR incorporation amount, the flexural strength of TR-100-15 at 60d reaches 6.2MPa, which is only 4.6% lower than that of TR-70-15. This indicates that cement-silica fume modification can effectively mitigate the adverse impact of a high RA replacement rate on the flexural performance of concrete. The synergistic effect of cement-silica fume modification and SSR cementitious material can achieve the optimization of the flexural performance of solid waste-based concrete.

As illustrated in Fig. 8, when the RA treatment method and replacement rate are fixed, as the SSR incorporation amount increases from 9% to 15%, the overall decline rate of the corrosion resistance coefficient of the concrete slows down, and its resistance to sulfate attack is enhanced. With a 70% RA replacement rate, the corrosion resistance coefficient of TWR-70-15 still reaches 0.65 after 75 cycles, which is significantly higher than the 0.61 of TWR-70-9 after 45 cycles. In the TR series reinforced with cement-silica fume, the corrosion resistance coefficient of TR-70-15 is 0.74 after 75 cycles, higher than the 0.69 of TR-70-9 after 45 cycles. Moreover, the groups with a high SSR incorporation amount can still maintain a relatively high corrosion resistance coefficient after a greater number of cycles. This is because a high incorporation amount of SSR can involve more active components in the reaction, generating denser hydration products and impeding the invasion of sulfate ions.

In terms of the RA replacement rate, in the TWR series with unreinforced RA, the corrosion resistance coefficient decreases relatively faster under a 100% RA replacement rate, and its resistance to sulfate attack is slightly inferior to that of the 70% replacement rate group. However, in the TR series reinforced with cement-silica fume, the gap in the corrosion resistance coefficient between the 100% replacement rate and the 70% replacement rate is significantly narrowed. The corrosion resistance coefficient of TR-100-15 reaches 0.70 after 75 cycles, which is relatively close to the 0.74 of TR-70-15. This indicates that cement-silica fume modification can effectively mitigate the adverse impact of a high RA replacement rate on the concrete's resistance to sulfate attack by repairing surface defects of RA and strengthening the interfacial transition zone between the aggregate and the cementitious matrix. It works synergistically with the SSR cementitious material to optimize the resistance to sulfate attack of solid waste-based concrete.

Figure 9 shows the mass loss rate of concrete under sulfate dry-wet cycles. As the dosage of SSR—a cementitious material composed of steel slag, blast furnace slag, and fly ash—increases, the generated calcium silicate hydrate (C-S-H) gel and other products fill pores at the microscopic scale and optimize the interfacial transition zone [34]. This provides structural support for maintaining compressive strength and corrosion resistance coefficient while simultaneously hindering sulfate ion penetration due to the dense structure. Consequently, leaching of concrete constituents is reduced, leading to a lower mass loss rate [13]. At 75 cycles, TR-70-15 achieved a corrosion resistance coefficient of 0.74 and a mass loss rate of 74%, outperforming TR-70-9 with a lower SSR content. Regarding RA replacement rate, unreinforced RA exhibits inherent microcracks, pores, and weak interfaces. At high replacement rates, sulfate erosion readily penetrates along these defects, causing rapid corrosion coefficient decline and increased mass loss rate [36–41]. TWR-100-9 recorded a corrosion coefficient of only 0.61 and a mass loss rate of 61% after 45 cycles, inferior to TWR-70-9. Conversely, cement-silica fume modification effectively “compensates” for the adverse effects of high replacement ratios by repairing aggregate defects and strengthening interfaces: TR-100-15 exhibited a corrosion resistance coefficient of 0.70 and a mass loss rate of 70% after 75 cycles, significantly narrowing the gap with TR-70-15. This approach brings the sulfate resistance of high-replacement-ratio RACs close to that of the 70% replacement ratio group.

Through the analysis of mechanical and durability properties, it was ultimately decided to conduct microscopic tests on the TWR-100-12 and TR-100-12 groups. Figure 10 shows the SEM images of TWR-100-12 and TR-100-12 before and after sulfate attack. Before the attack, the RA surface in TWR-100-12 had residual micro-cracks, pores, and a weak mortar layer. Meanwhile, the C-S-H gel generated from the hydration of the SSR cementitious material was sparsely distributed, and there were many unhydrated particles remaining in the matrix. In contrast, after being reinforced with cement-silica fume, in TR-100-12, the C-S-H gel and the products of the silica fume's pozzolanic effect filled the surface defects of the aggregate. Moreover, the hydration products in the reinforcement layer could promote the secondary hydration of the active components in SSR, supplemented by the filling of micropores by silica fume, resulting in a more continuous network of hydration products in the matrix [42–44]. After sulfate attack, the loose interface in TWR-100-12 provided a rapid pathway for the invasion of sulfate ions. The ettringite and plate-like gypsum generated from the reaction between sulfate ions and hydration products caused the matrix to expand, exhibiting a honeycomb-like loose structure [45–47]. The cement-silica fume reinforcement layer in TR-100-12 hindered the rapid invasion of sulfate ions. The C-S-H gel network did not undergo global destruction, and the overall degree of deterioration was much less severe than that of TWR-100-12.

By referring to the XRD results in Fig. 11, the differences in the phase composition and evolution between TWR-100-12 and TR-100-12 can be identified. Essentially, these differences stem from the regulatory effect of RA modification on the types of hydration products, the residual amount of unhydrated minerals, and the formation of sulfate attack products. Prior to sulfate attack, since TWR-100-12 lacks cement-silica fume activation, the SSR cementitious material predominantly undergoes low-calcium hydration. In the XRD pattern, the broadened and dispersed peak of C-S-H gel dominates, while the peaks of calcium hydroxide and AFt are weak. Additionally, the characteristic peaks of unhydrated slag and steel slag are prominent. In contrast, the cement-silica fume reinforcement layer in TR-100-12 promotes high-calcium hydration in coordination with the SSR reaction. This results in an increase in the intensity of the C-S-H dispersed peak, a significant enhancement of the CH and AFt peaks, and a substantial reduction in the peaks of unhydrated minerals due to the complete hydration of SSR triggered by alkaline activation. After sulfate attack, in TWR-100-12, the intensity of the gypsum peak surges, while the C-S-H dispersed peak and CH peak nearly disappear, indicating severe degradation of the original products. In TR-100-12, however, due to the reinforcement layer's obstruction of SO₄²⁻ intrusion, the gypsum peak only appears weakly, and the C-S-H and CH peaks only experience a slight reduction. The phase stability of TR-100-12 is significantly superior to that of TWR-100-12, fully demonstrating the optimizing effect of cement-silica fume modification on the phase evolution of SSR-based concrete under sulfate attack.

As shown in Fig. 12, prior to sulfate attack, in TWR-100-12, the hydration of SSR generates C-S-H. The Si-O stretching vibration peak is located in the range of 950–1000 cm⁻¹ and is weak and broadened. The characteristic peak of CO₃²⁻ is weak, and the -OH stretching vibration peak is weak, with the -OH peak of CH (3640 cm⁻¹) being invisible, indicating low contents of bound water and CH. In contrast, the cement-silica fume reinforcement layer in TR-100-12 promotes the formation of high-calcium C-S-H. The Si-O peak shifts to the range of 1000–1050 cm⁻¹ and becomes strong and sharp. The intensity of the CO₃²⁻ peak increases, the intensity of the -OH peak significantly rises, and the -OH peak of CH is clearly visible, indicating an enrichment of bound water and CH.

After sulfate attack, in TWR-100-12, the intensity of the Si-O peak drastically decreases and the peak shape becomes fragmented. The intensity of the -OH peak drops sharply, and the peak of adsorbed water disappears, indicating a significant loss of water. In TR-100-12, however, due to the reinforcement layer's obstruction of SO₄²⁻ intrusion, the Si-O peak only experiences a slight decrease and its position remains stable. The CO₃²⁻ peak rises gradually, the -OH peak decreases slightly, and the peak of adsorbed water remains stable. This fully demonstrates the optimizing effect of cement-silica fume modification on the chemical structural stability of SSR-based concrete under sulfate attack.

As illustrated in Fig. 13, the essential difference in the mechanisms of TWR-100-12 and TR-100-12 under sulfate attack lies in the regulatory discrepancies of RA modification on the sulfate ion transport efficiency and the erosion resistance stability of hydration products.

In TWR-100-12, due to the presence of micro-cracks and a weak mortar layer remaining on the unreinforced RA surface, a wide and porous interfacial transition zone (ITZ) is formed with the SSR cementitious material. Additionally, a significant amount of unhydrated particles remains in the SSR cementitious matrix, enabling SO₄²⁻ to rapidly penetrate into the interior [39–41]. Meanwhile, its hydration products primarily consist of low calcium-to-silicon ratio (C/S) C-S-H gel and a trace amount of calcium hydroxide (CH). The low-calcium C-S-H has a low degree of polymerization in its silicate framework, making it susceptible to destruction by SO₄²⁻. CH rapidly reacts with SO₄²⁻ to form expansive gypsum. The expansion causes interfacial delamination and matrix cracking, ultimately leading to rapid structural deterioration [45].

In contrast, the cement-silica fume reinforcement layer in TR-100-12 fills the surface defects of RA, forming a dense transition shell and promoting the generation of high C/S ratio C-S-H in SSR. This significantly retards the migration of SO₄²⁻ [46]. The continuous cement hydration supplements CH, maintaining a high alkalinity in the system. As a result, only a small amount of gypsum is formed on the surface layer by SO₄²⁻. The expansive stress is buffered by the dense matrix, causing only micro-cracks on the surface layer while keeping the internal structure intact [47].