The determination of the curing time of AAS was carried out in accordance with the method specified in the Chinese cement standard GB/T 1346-2001 “Water consumption for standard consistency of cement, setting time, and stability test method”.

The slag, alkali component, water, and Na₂Cr₂O₄.4H₂O were weighed out in a fixed ratio and mixed evenly in a cement slurry mixer. A 40 × 40 × 40 mm mold was used to form a solid specimen, and the molding surface was covered with a plastic film to prevent moisture evaporating. The mold was removed after 24 h at room temperature. In order to prevent the precipitation of Cr⁶⁺ in the specimen and the diffusion of alkali components to the condensed water on the surface of the specimen, which could lead to insufficient hardening of the surface of the cured body, the cured body specimen was placed in a sealed bag and kept in a standard indoor environment (temperature 20 ± 2°C, RH ≥ 90%) for the specified length of time.

The 40 × 40 × 40 mm specimens cured to 3, 7, and 28 d were removed from the standard curing room, and a universal material testing machine was employed to test the compressive strength of 3, 7, and 28 d for different cured bodies according to “Test Method for Strength of Glue Sand” (ISO Method).

The heat detector of a TAM air microcalorimeter (TA Instruments, New Castle, DE, United States) was used for the heat of hydration test. The basic principle of the instrument is as follows: the heats of hydration of the sample and reference sample are measured using a thermocouple (water or the completely hydrated cement stone is usually selected as the reference sample) in a constant temperature environment, and the exothermic rate and cumulative heat release of the sample are calculated. The test temperature was set at 20°C. Measurements were carried out on 4 g samples of the pastes. The solution of Na₂Cr₂O₄ was added in the proportions given in Table 3. Heat development was measured for 72 h, with a step of 50 s between consecutive measurements. Before the experiment, slag was placed in an ampoule, the solution was placed in a syringe, and the ampoule was placed in the calorimeter. After the syringe had been installed, the solution was injected into the ampoule when the temperature had stabilized at 20°C, when measurement began.

According to the national standards GB 5086.1-1997 “Solid Waste Leaching Toxicity Leaching Method Reversal Method” and GB/T 15,555.4-1995 “Solid Waste Determination of Hexavalent Chromium Diphenylcarbazide Spectrophotometer Method,” the molded test piece was broken and a sample passing through a 5 mm sieve was placed in an oven and dried at 60°C, and then soaked in deionized water at a liquid-to-solid ratio (L/g) of 1:10. Subsequently, the mixture was vibrated at 23 ± 2°C for 18 h with a flip-type shaking device at a speed of 30 ± 2 rev/min, and kept for 30 min, and then the leaching solution was collected with a vacuum filter. A dual-beam UV-Vis spectrophotometer TU-190 made by Beijing General Instrument Co. Ltd. was used to measure the concentration of Cr⁶⁺ in the leaching solution.

The alkali-slag-Cr⁶⁺ solidified body pore solution extrusion test piece was prepared using a special molding test mold (Jiangsu Subote Materials Co., Ltd., Nanjing, Jiangsu, China). The test piece was sealed in a plastic bag and then placed in standard storage environment (temperature: 20 ± 2°C; RH ≥ 90%) for the specified time. The hole solution pressing device of cement-based material was then used to squeeze the hole solution, and the dual-beam spectrophotometer was used to test the Cr⁶⁺ concentration in the hole solution.

The sample was broken following curing for the specified time and hydration was immediately terminated using absolute ethanol. The sample was then baked to a constant weight at 60°C and passed through a 0.08 mm sieve. The powder was then sealed in an ampoule for later testing.

A Rigaku D/Max-5A 12 kW rotating target X-ray diffractometer (Rigaku Corporation, Tokyo, Japan) was used for testing, with Cu Kα radiation at a voltage of 40 kV and a scanning speed of 4°/min. The sample was dried to constant weight in a vacuum drying oven at 60°C before the test.

The sample was broken following curing for the specified time and hydration was immediately terminated with absolute ethanol. The sample was then baked to a constant weight at 60°C and sprayed with gold on a cross-section for later testing.

A variable vacuum scanning electron microscope (TESCAN VEGA2, Tescan Brno, s.r.o, Brno, Czech Republic) was adopted for morphological characterization, and an INCA Energy 350X energy dispersive X-ray spectrometer (Oxford Instruments, Oxford, United Kingdom) was employed for elemental analysis.

Figure 1 shows the effect of Na₂Cr₂O₄ on the curing time of AAS. It can be seen that the latter gradually increased as the Na₂Cr₂O₄ content increased. The initial setting time of AAS with 2% Na₂Cr₂O₄ increased from 17 to 35 min, compared with that of the control AAS, and the final curing time increased from 28 to 50 min, both of which were doubled. This is consistent with the results of previous studies (Palomo and Palacios, 2003).

Figure 2 shows the effect of Na₂Cr₂O₄ on the compressive strength of AAS. It can be seen that the latter decreased with increasing concentration of Na₂Cr₂O₄ for the same curing age. The compressive strength of cement with different Na₂Cr₂O₄ content increased with age.

The curing time and strength development of AAS are two important macro indicators that reflect the hydration process (Shi and Roy, 2006; Pacheco-Torgal et al., 2015). As shown in Figures 1, 2, the addition of Na₂Cr₂O₄ played a significant role in the hydration process of AAS, and the effect increased with increasing concentration of Na₂Cr₂O₄. Fang et al. (2020) pointed out that when the chemical substances introduced into the AAS hydration system combine with the Ca²⁺ ions or [SiO₄]⁴⁻ ions generated by the disintegration of the slag and OH⁻ ions in the system under the action of the alkali component, forming small-sized particles of insoluble substances, they may form precipitates on the surface of the slag particles, hindering further hydration reactions with OH⁻ ions and activating components.

Figure 3 shows the concentration of Cr⁶⁺ leaching from the AAS stone. It can be seen that at both 3 and 7 d, Cr⁶⁺ was effectively solidified in the AAS when the Na₂Cr₂O₄ content was less than 0.5%. The concentration of Cr⁶⁺ leaching out increased significantly when the Na₂Cr₂O₄ content was greater than 0.5%, and it rose sharply with increasing amounts of Na₂Cr₂O₄. Therefore, it can be conjectured that the initial leaching concentration of Cr⁶⁺ has a critical value when the concentration of Na₂Cr₂O₄ is 0.5% for this cement mix. In addition, the Cr⁶⁺ leaching concentration for AAS stone at 28 d also appeared to follow a similar pattern, whereas the Na₂Cr₂O4 content for a critical leaching concentration increased from 0.5 to ∼1.0%.

From the previous analysis, it can be seen that the introduction of Na₂Cr₂O₄ led to a low degree of initial hydration of AAS, so that the porosity of the initial cement stone and the leaching rate of Cr⁶⁺ are both increased, leading to a low leaching concentration threshold. The density of the AAS stone gradually increased, and the porosity gradually decreased with increasing age, so that the leaching rate of Cr⁶⁺ was small, and the Na₂Cr₂O₄ content for a critical leaching concentration increased with extension of the hydration time.

From the previous analysis, it can be seen that the solidification of Cr⁶⁺ is closely related to the hydration process of AAS. In order to further study the solidification mechanism of AAS on Cr⁶⁺, the effect of Na₂Cr₂O₄ on the hydration exothermic process of AAS was studied (Figure 4).

Figure 4A shows the exothermic hydration rate of AAS for the Na₂Cr₂O₄ group and the control group, among which the control group was AAS without Cr⁶⁺. In the Na₂Cr₂O₄ group, the Na₂Cr₂O₄ content was 1.0%, and the content of the rest of the material was the same as that of the control group. The figure shows that, compared with the control group, the AAS exothermic hydration rate of the Na₂Cr₂O₄ group was significantly prolonged.

Figure 4B shows the total hydration heat released by AAS in the Na₂Cr₂O₄ group and the control group. It can be seen that the AAS hydration of the Na₂Cr₂O₄ group was always significantly lower than that of the control group and that the variation in the total hydration heat released by the two groups were completely different. It can be seen from Figure 4A that there is an obvious exothermic peak in the early stages for the two groups, which was caused by the destruction of the slag surface by OH⁻ ions and the formation of C-S-H gel on the surface of the slag particles (Shi and Roy, 2006). However, it can be seen from the total quantity of heat released during the corresponding time in Figure 4B that the total quantity of heat released in the Na₂Cr₂O₄ group during this time was significantly lower than that of the control group. It is inferred that Na₂Cr₂O₄ prevents OH⁻ ions from destroying the slag structure or forming C-S-H gel on the slag surface. The Na₂Cr₂O₄ group showed a longer induction period (Figure 4A), and the total quantity of corresponding hydration heat released was much lower than that of the control group (Figure 4B). It can be inferred that Na₂Cr₂O₄ chemically reacts with the ions in the liquid phase to form insoluble substances on the surface of the slag particles, which prevents Ca²⁺ ions from reacting with SiO₄ ^4- in the liquid phase to form C-S-H gel. After entering the acceleration period, the variation in heat released for the two groups was identical, although the heat released by the Na₂Cr₂O₄ group was always lower than that of the control group. The effect of Na₂Cr₂O₄ on the hydration process of AAS was therefore weakened. In summary, the introduction of Na₂Cr₂O₄ significantly changed the hydration process of AAS according to Hess’s law (Gersten and Gersten, 2001). In order to further understand the effect of Na₂Cr₂O₄ on the hydration process of AAS, the corresponding hydration kinetic parameters were calculated from the hydration heat release curve of the alkali-slag cement, the Knudson hydration heat model, and the Jander equation (Table 4).

Table 4 lists the AAS hydration kinetic parameters of the Na₂Cr₂O₄ group and the control group. It can be seen from the data listed in the table that the start time of the acceleration period in the Na₂Cr₂O₄ group was about 2.4 times that of the control group, and the start time of the deceleration period was about 1.7 times that of the control group. Na₂Cr₂O₄ significantly prolonged the curing time of AAS. In Table 4, N represents the order of the chemical reaction, which mainly characterizes the degree of difficulty of the reaction. The latter is controlled by chemical reactions on the particle surfaces and by the dissolution of reactants or the deposition of reaction products when N ≤ 1. The reaction is controlled by the diffusion of reactants through the porous reaction product layer when 1 ≤ N ≤ 2. The reaction is controlled by the diffusion of reactants through the dense product layer when 2 ≤ N (Shi and Roy, 2006). K represents the rate constant of the chemical reaction, which characterizes the speed of the chemical reaction (Gersten and Gersten, 2001).

It can be seen from Table 4 that the reaction rate of AAS in the Na₂Cr₂O₄ group during both the acceleration and deceleration periods was significantly higher than that of the control group, yet the reaction order was significantly lower than that of the control group. According to the AAS hydration process (Zhang et al., 2016), the chemical reaction order N of the whole hydration process was around 2 for the control group, indicating that the hydration process of the control group was more prone to being controlled by the diffusion of reactants through the dense product layer from the acceleration period. The main hydration products of sodium silicate activated in AAS were mainly low calcium-to-silicon ratio C-S-H gel, which has low porosity and a compact structure (Shi and Roy, 2006). Although the reaction rate of Ca²⁺ and SiO₄ ⁴⁻ was rapid in the liquid phase, it is not easy for Ca²⁺ to penetrate the hydration product layer and enter the liquid phase in order to react. For the Na₂Cr₂O₄ group, the reaction order N was closer to 1, although the reaction rate was faster during both the acceleration and deceleration periods, indicating that the hydration process was more prone to being controlled by the diffusion of reactants through the porous reaction product layer from the acceleration period. Based on this, it can be further speculated that Na₂Cr₂O₄ first forms a poorly soluble porous product on the surface of the slag particles with the substance in the liquid phase. After entering the decay period, the reaction rate constants and reaction orders of the two groups were equivalent, and both were controlled by the diffusion of the reactants through the dense product layer, at which time C-S-H gel should have been generated.

In this study, a cement-based material pore solution pressing device (Jiangsu Subote Materials Co., Ltd.) was used to collect different cement paste and pore solutions of the paste solidification body. The solidification effect of AAS on heavy metal Cr⁶⁺ ions during the hydration process was investigated by analyzing the ion concentration of the pore solution and using the analysis of hydration kinetics described above.

Table 5 shows that the total Cr, Cr⁶⁺, and Ca²⁺ concentrations in the liquid phase gradually decreased with the progress of hydration. The Ca²⁺ concentration of the pore solution in the AAS stone depended mainly on the degree of disintegration of the slag glass body and the amount of C-S-H gel hydration product. With the extension of hydration time, OH⁻ ions were continuously consumed, and the formation of hydration products gradually increased. The hydration product layer gradually changed from a porous structure to a dense structure, which reduced the amount available for disintegration of the slag glass structure as well as the concentration of Ca²⁺. The concentrations of total Cr and Cr⁶⁺ in the alkali-slag solidified body were reduced as a function of time due to the reduction of Cr⁶⁺ by S²⁻ or HS⁻ in the slag and the enhanced adsorption of Cr⁶⁺ by the hydration products.

At 1 d, the Ca²⁺ content in the AAS pore solution of the Na₂Cr₂O₄ group was significantly higher than that of the control group. This shows that Na₂Cr₂O₄ can promote the dissolution of slag particles. At 3 and 7 d, the Ca²⁺ content in the AAS pore solution of the Na₂Cr₂O₄ group was higher than that of the control group. The heat of hydration experiment showed that the rates of hydration of the Na₂Cr₂O₄ group in the initial age were lower than those of the control group, indicating that Na₂Cr₂O₄ delayed the Ca²⁺, which dissolved from the slag, from reacting with SiO₄ ⁴⁻ in the liquid phase to form C-S-H gel, which led to a delay in the second hydration reaction peak in the heat of hydration experiment.

Figure 5 shows the XRD pattern of AAS stone at 28 d. It can be seen that a typical “Bailey broad peak” appeared around 30°, which originated from the C-S-H gel. There are two obvious peaks at 32.3° and 35.2°, both originating from CaCrO₄ (Wang and Vipulanandan, 2000; Chen et al., 2009). Combined with the analysis of the hydration kinetic process, it can be known that Na₂Cr₂O₄ hydrolyzes to form CrO₄ ²⁻, which reacts with the dissolved Ca²⁺ in the slag to form CaCrO₄, hence preventing Ca²⁺ from interacting with SiO₄ ⁴⁻ in the liquid phase to generate C-S-H gel. This reduces the strength of AAS.

Figure 6 shows the SEM images of AAS stones in both the Na₂Cr₂O₄ group and control groups at 28 d. It can be seen from the control group that the slag was basically hydrated, forming a continuous dense structure with fewer pores. In addition, it can be seen from the Na₂Cr₂O₄ group that the structure of the hydration products was relatively loose, and needles and rods were formed. The needles and rods (point B) were analyzed using energy dispersive spectroscopy (EDS) (Figure 7). The results showed that the weight percentage of elemental Cr at this point was 3.23%, which was much higher than the original amount in the mix (1%). In addition, a large amount of Ca was enriched here. According to the results of Laforest and Duchesne (2005), the rod-shaped crystals here were CaCrO₄.