Synthesis of alkali-activated specimen consisted of mechanically mixing a solution of 15 M NaOH with a blend of GBFS and RHA solids for 2 min using a mechanical mixer at a constant liquid to solid (L/s) equivalent to 0.4 (Azad and Samarakoon, 2021). BFS particle size was reduced to −100 µm before synthesis to enhance the degree of reaction. RHA mass% content in the blend varied from 0% to 15%, producing 3 samples for each mix design. Further increase in RHA is associated with low workability (Bayraktar and Fello, 2023). The homogeneously mixed paste was cast into 50 × 50 × 50 mm³ cubic moulds and kept for 24 h to consolidate at ambient temperature. Subsequently, the consolidated specimens were de-moulded and placed in the oven at 80°C for 4 days to ensure the completion of the reaction. High curing temperature accelerates the hydration process of precursors (Shi et al., 2023). The moulds containing the AAM specimen were sealed with a steal board so as to minimize loss of moisture or capillary water, which enhances drying shrinkage and cracks (Mo et al., 2014). Lastly, synthesized specimens were sealed in polyethene bags to inhibit carbonation and prevent a decrease in alkalinity. Table 2 summarizes the parameters of the mixing design used in the study.

The mechanical characteristics measured using Matest compression test Machine of synthesized specimens with varied RHA content are presented in Figure 2, whereby it is observed that higher content decreases the strength of materials.

Static alkali leaching experiments were conducted to investigate RHA content’s effect on the extent of alkali leaching. First, synthesized specimens with various RHA content were comminuted and screened. The solid feed consisted of a fraction of alkali-activated GBFS-RHA fragments with particle sizes varying from 2.36 to 4.75 mm (Saludung et al., 2021). Solid to liquid was equivalent to 1:20. Aliquots from leaching vessels were collected after 3 h, 24 h, 72 h, 7 days, 14 days and 28 days. There were two replicates for each mix design. Visual and direct observation were primarily used to observe efflorescence on the specimen. Specimens were placed in open Pyrex glasses which were partially filled with 5 mL of deionized water. The water level was adjusted up to 5 mm for all the efflorescence tests. Efflorescence images were captured after 24 h and 72 h. After 72 h, the change in the observed efflorescence was relatively insignificant.

Determination of water absorption capacity and porosity of specimens was carried out by soaking the dried specimen at 20°C for 24 h according to ASTM C62 standard. Before submersion, synthesized cubic specimens were dried at 80°C until no fluctuation in the weight of dried specimens was observed. Each reported water absorption capacity is the mean value of 3 specimens synthesized under the same condition. Water absorption capacity % (W_AC%) is calculated according to Equation 4 whereby m i and m f are the masses of the specimen before and after, immersion respectively. W A   % = 100 × m f − m i m i (4)

Characterization of synthesized specimens (in a powder form) was carried out through x-ray fluorescence (XRF), x-ray diffraction (XRD), scanning electron microscope-energy dispersive spectrometry (SEM-EDS), Fourier transforms infrared spectroscopic (FTIR) and thermal analyses. Chemical compositions of raw materials were analyzed using an automatically calibrated Rigaku model Primus IV spectrometer operating at 40 mA and 50 V. For mineralogical analysis, specimens were scanned at 0.5°C/min from 5^o to 90° (2 theta) with a Rigaku model Ultima IV XRD spectrometer using a copper K* x-ray source, operating at 30 mA and 40 V and equipped with a PXDL software program. An EDS Vega 3 SEM with a high voltage setting and beam intensity of 20 V and 11 W/cm2 was used to determine carbon-coated specimens’ morphology, microstructure, and elemental composition. The study of chemical bonds within the specimen was carried out using a Thermofisher FTIR spectroscope. Specimen were scanned from 400 to 4,000 cm⁻¹ with resolution set at 4 cm⁻¹ and the number of scans equivalent to 32 scans. A thermal decomposition study based on thermal gravimetric (TGA) and differential thermal analyses (DTA) was carried out from room temperature to 950°C with a heating rate of 20^°C/min. The surface area and micropore volume of the samples from different specimens were determined using a BET surface analyser Tri Star II Plus 3. Before analysis, the samples were degassed at 100°C for 6 h. N₂ at 80.2 K was used for adsorption-desorption tests.

Figure 3 shows the efflorescence evaluation after a period of 24 and 72 h. The results primarily based on direct visual observation of alkali activated GBFS based specimens synthesized with and without RHA additives. It has been demonstrated that the extent of efflorescence decreases with the rise in RHA content. Observed efflorescence on 100% GBFS - 0% RHA specimen after 24 and 72 h is caused by the excess of unreacted sodium hydroxide, thus able to migrate from the pore solutions to the surface and subsequently react with dissolved atmospheric carbon dioxide to generate carbonate salts in the form of NaHCO₃. xH₂O and Na₂CO₃. xH₂O. The x-ray diffractogram of salt deposited on 100% GBFS-0% RHA specimen after 72 h is shown in Figure 4, whereby the high predominance of a sodium carbonate monohydrate polymorph referred to as thermonatrite (DB number: 01-076-0910) has been observed. As aforementioned, efflorescence is associated with the unaesthetic appearance of the alkali-activated material with no direct detrimental impact on the integrity of AAM. Nevertheless, efflorescence can be exacerbated by high moisture content drawbacks leading to the spalling of the manufactured AAM. However, as RHA content increases, the efflorescence can be retarded or hindered over time, as observed in the current study.

Previous studies by Sun et al. (2020) and Saludung et al. (2021) plainly explain the underlying mechanism through which a highly reactive silica exhibit its inhibitory influence on the efflorescence of alkali-activated materials. The underlying mechanism lies in the fact that reactive silica increases the degree of reaction through which a gel is possibly formed and subsequently fills the pore solution within the matrix of the AAMs. With the addition of reactive silica, the Na + diffusion rate from the pore solution to the AAMs surface decreases to some extent (Liaudat et al., 2020).

In conjunction with previous studies, the analogy between the intrinsic properties of RHA and those of previously studied sources of highly reactive silica can be established. The specific area of RHA varies between 20 and 200 m²/g whilst the range of the particular area of silica fume is from 18 to 23 m²/g, and that of nano-SiO₂ has been reported to be 296 m²/g. Consequently, amorphous silica from RHA is as reactive as the SF and nano-SiO₂. Moreover, it can also be inferred that the aforementioned underlying mechanism through which the decrease of the extent of efflorescence is achieved correlates with the inhibitory influence of RHA on the efflorescence formation on the surface of synthesized specimens in the current study.

The effect of RHA content on the alkali leaching extent is illustrated in Figures 5A, B. Figure 5A indicates that the extent of alkali leaching initially rises and remains relatively constant over time. Without the addition of RHA, Na⁺ ions concentration in the eluate is high due to the excess of unreacted Na⁺ ions in the form of Na(H₂O)_n ⁺ or hydroxyl ions in the pore solution within the synthesized specimen. These alkali hydroxyl ions form weaker bonds with the pore solution than alkali Na⁺ ions (Sun et al., 2020). Consequently, alkali hydroxyl ions weakly bound to the calcium aluminosilicates are readily subjected to hydrolysis and increase the concentration of Na⁺ ion concentration of the eluate. Conversely, with the addition of RHA, it is observed that the Na⁺ concentrations of each specimen in water drop with the increase in the RHA content in the alkali-activated GBFS-based specimen over the immersion period of 28 days. However, the addition of RHA does not permanently inhibit alkali leaching but rather retard the extent of alkali leaching over time. In Figure 5B, results demonstrate that the addition of RHA leads to a significant decrease in the extent of total alkali leaching from 29.9% to 9.41%. In other words, the alkali leaching extent has been reduced by 69.56% with the addition of 15% mass of RHA.

In Table 3, the dispersive energy analysis ascertains an increase of up to 15% RHA stabilizes Na⁺ ions from leaching away in contact with water. Hence, 15% RHA (mass%) was the optimum. Compared to the previous study by (Saludung et al., 2021), the observed drop in the alkali leaching extent with the silica addition of silica fume (SF) has been reported from about 67% to 56%. The decrease in alkali leaching is due to the Na/Si ratio (Saludung et al., 2021) which increases with the increase in silica content. Additionally, in the current study, it could be stipulated that forming a gel-like structure able to stabilize Na⁺ ions is favored with the addition of RHA that contains reactive amorphous silica and increases the reaction rate. Moreover, at high, pH, reactive silica from RHA and aluminum ions reacts with OH- to form Si-O-Si and Al-O-Al. Subsequently substitution of Si⁴⁺ with Al³⁺ is more prominent as results of similarity in atomic size, leading to a negative charge imbalance in the crystal lattice. In the later stage, the negative charge imbalance is neutralized by the alkalis. The bonds alkalis between the Si-O-Al structure are characterized by a strong electrostatic force. Consequently, the transport properties of Na⁺ ions are altered so that fewer Na⁺ ions migrate to the water.