Sustainable utilization of aluminum waste in geopolymer concrete: Influence of alkaline activation on microstructure and mechanical properties

1.1 Importance of study

Literature evaluation has shown a need for more research on the use of GPC, particularly with the incorporation of Al-waste. Moreover, a literature review indicates that the limited investigation on GPC has mostly focused on curing techniques. Research on the use of both ambient and oven-curing techniques in GPC is lacking. Additional research is required owing to the distinct advantages and challenges linked to GPC. This study is significant because it investigates the impact of different compounds on GPs, given that the curing process is a crucial factor. Thus, in this study, the effect of different curing methods (ambient and oven) on the strength of GPC was investigated, and mechanical properties were evaluated. Another aim of this study is to expand the knowledge of this material by investigating the machinability and strength of GPC produced by partially replacing slag with FA. The experiment detailed here aims to address these disparities by examining the impacts of Al-waste and the percentage of slag on various mechanical properties, durability, and fresh properties of FA-based GPC specimens. Since ambient curing of GPC is considered a feasible alternative for ordinary Portland concrete (OPC), efforts in this field are encouraged. This research was conducted to assess the influence of the granulated blast furnace slag ratios 0, 5, 10, and 20%.

3.1 Penetrability and setting time

The penetrability of the mixture was reduced with the increase in the M of NaOH. This behavior was observed in numerous studies in the literature. Nevertheless, in this investigation, it was found that the addition of Al-waste led to a reduction in the slump values and improved the penetrability of the material. It was noted that the time of setting was around 10 min with 1, 2, and 3% Al-waste addition and 40 min with 4% addition. The consequences were consistent with similar investigations in the literature. Based on Yip and Van Deventer [38], there is evidence that the presence of calcium throughout the mixture of FA-based GPC enhances the likelihood of forming compounds such as calcium silicate and calcium aluminate hydrates. This information comes from studies that have been conducted [36]. Alkali-activated concrete might be compressed well on a vibrating table even for comparatively low slump values. Consequently, the penetrability of alkali-activated concrete is categorized and established on the form of compaction [39].

3.2 Slump test

Based on studies collected from prior examinations and experience obtained during the procedure, it was observed that permeability dramatically reduces as both the M of NaOH increased. The results indicated that the slump values declined as M increased. Conversely, maintaining a constant M allowed for the investigation of how the amount of Al-wastes affected the slump. While the NaOH content was selected as 11 M, it was revealed that the slump values decreased as the Al-waste content proportion increased from 0 to 1%, indicating a correlation between the two parameters. The same observation was made for 13 M when considering it as the AL percentage. It was revealed that the value of the slump grew when the percentage of Al-waste contents improved from 0 to 1%. When the Al content increased from 1 to 4%, a reduction in slump values was observed. The same observation was made for 16 M when the Al proportion was altered. It was found that the slump values increased as the Al-waste content proportion improved from 0 to 1%.

While the amount of Al-waste increased by 0–4%, there was a significant decrease in slump values. Specifically, the slump decreased to 40–60 mm with 1% Al, 15–40 mm with 2% Al, and 5–10 mm with 3–4% Al. Al reacts with water in an alkaline environment, especially in systems containing NaOH, to produce hydrogen gas (H₂). This reaction causes gas bubbles to form in the concrete, causing the mix to expand in volume. At the same time, this outgassing destabilizes the consistency of the mixture, reducing cohesion and workability. The irregular, sharp, wiry geometry of the chip particles and their low specific gravity reduce the cohesion of the concrete and require more water. However, since no additional water is added, the slump decreases. As the slag ratio increases in the mixture, an increase in slump values is observed, especially in Al-free or low Al ratio samples. However, this positive effect disappears in specimens containing slag and high Al-waste. Slag is a reactive binder that dissolves particularly well in alkaline media, forming binder gels such as calcium-aluminosilicate-hydrate (C–A–S–H) and sodium-aluminosilicate-hydrate (N–A–S–H). This generally enhances the workability. However, in the presence of Al-waste, the positive effect of slag is suppressed because the resulting hydrogen gas and rapid reactions destabilize the mixture.

Examining the chemical composition of the slag in Table 2 reveals a particularly high amount of calcium oxide (CaO). This indicates that, in addition to the dominant N–A–S–H gel phase found in the traditional low-calcium GP systems, other phases, such as C–A–S–H, may also form. These hybrid structures have a twofold effect on long-term stability. First, a rapid increase in mechanical strength occurs at early ages due to the presence of C–A–S–H phases. This significantly improves the performance of the binder, especially in systems cured at ambient temperature. Conversely, the possibility of carbonation and other chemical deterioration in alkaline environments increases with higher calcium content, which may limit long-term durability. The study emphasizes that the slag additive facilitates ambient-temperature curing. However, it does not present measurements such as the strength activity index (SAI) or thermal reactivity profiles based on isothermal calorimetry, which would quantitatively reveal the effectiveness of slag in the binder system. Therefore, additional characterizations are needed to clarify whether the slag additive creates a synergistic binder structure with the GP matrix. Future studies can perform tests such as separating heat flows in hydration and geopolymerization processes by isothermal calorimetry.

3.3 A thorough examination of the compressive test

As can be seen from Table 3, the properties are highly dependent on the relative amounts of NaOH and Al-waste ratio, in addition to the temperature at which they are cured [40–43]. In the context of this inquiry, it is important to observe the required curing temperature and NaOH concentration. The CS was in GPC that was prepared by combining various amounts of aluminum hydroxide and NaOH. These GPCs were structured such that the concentrations of both variables varied. It was discovered that there was a pattern in which the strength of the GPC enhanced as the concentration of NaOH increased. As shown in Figure 4, the concentrations of 11, 13, and 16 M NaOH produced the highest strength in the CS when the CS contained 0% of Al-waste. The M of NaOH was fixed at 11, while the proportion of slag ratio was adjusted to 0, 5, 10, and 20% in the GPC mixture. In addition, the ratio of this factor was changed from 0 to 1% in the calculation of the mixture modified with the ratio of Al-waste aggregate, which included FA. This was done to reflect that the ratio was changed. At the end of this procedure, the ratio of variation in CS was determined to be 17.66% (from 20.28 to 16.697 MPa for 5% slag ratio). While the percentage in the mixture was enhanced, it was observed that this quantity of variation reduced to approximately 15.90% (from 18.86 to 15.86 MPa for 10% slag ratio). While the percentage in the mixture was enhanced, it was observed that this quantity of variation reduces to approximately 9.66% (from 17.59 to 15.89 MPa for 20% slag ratio). Additionally, the M of NaOH was selected repeatedly at 13 M, and the percentage of slag ratio was changed to 0, 5, 10, and 20% in the GPC mixture. At this mixture, the percentage of variation in CS at 5% slag ratio was reduced as 18.9% (from 22.33 to 18.10 MPa). While the amount of Al-waste in this mixture increased to 4%, the amount of CS in the mixture decreased by around 42.32% (from 22.33 to 12.88 MPa). Similarly, while the slag ratio in the mixture was increased from 0 to 20%, it was discovered that there was a 27.7% reduction in the CS of the mixture (from 26.59 to 19.228 MPa). The mixture’s CS had dropped by 46.3%, while the levels of Al-waste in the mixture increased from 0 to 4%, constituting 20% of the mixture’s total volume.

Figure 4

Results of the CS test.

In addition to this, the NaOH concentration was determined to be 16 M, supporting consistent findings. While the slag ratio in the mixture was improved from 0 to 20%, the CS of the mixture was observed to decrease by 30.93% (from 30.264 to 20.9 MPa). While the amount of Al-waste in the combination increased from 0 to 1%, resulting in a significant relative increase, the mixture’s CS was determined to have decreased by 33.79% (from 28.82 to 19.08 MPa).

The outcomes of the CS test are provided in Table 3 as well. During these investigations, the percentage of Al-waste aggregate that was mixed with FA was calculated, and the molar concentration of NaOH in the mixture was varied to 11, 13, and 16 M to observe the effects of these changes. The results of these investigations are summarized in Table 3. As observed in Table 3, the calculation of the amount of the Al-waste aggregate with FA was incrementally occupied at 0, 1, 2, 3, and 4%, while the percentage of M of NaOH was increased incrementally from 11 to 13 M and then to 16 M in the mixture of concrete. It was observed that this percentage of variance tends to increase to 10.10% (from 11 to 13 M) and 42.11% (from 11 to 16 M) for Al-waste containing 0 and 5% slag ratios. When the Al-waste ratio increased by 1%, it was observed that this percentage of variance tends to increase to 8.40% (from 11 to 13 M) and to 14.27% (from 11 to 16 M) for Al-waste containing 1 and 5% slag ratios.

This may be because larger concentrations of NaOH can melt silica and alumina together into a homogeneous slurry paste [20]. Furthermore, it was found in earlier studies that while the GP contained increased Al-waste content, the strength was somewhat lower than that of 0 wt% [20]. Adding Al-waste influenced the alumina component in the matrix. The findings of this study are consistent with previous research findings. In another study, it was also observed that additional NaOH was the reason for the CS reduction, attributable to the aluminosilicate gel precipitation [43]. It was found that higher concentrations of NaOH retained the Na ions in GPC. Furthermore, the concentration of Ca decreased concomitantly with increasing NaOH concentration. This statement showed the Na component’s inclination to exchange with the Ca component in the GP matrix [21].

3.4 Examination of the splitting tensile strength (STS)

The consequences of STS persistently follow the parallel trend as the CS, as shown in Figure 5. As shown in Table 4, the M of NaOH was consistently chosen at 11 M, whereas the inclusion of slag ratio in the concrete mixture was varied to 0, 5, 10, and 20% for the Al-waste aggregate. In the experimental design, the composition of the mixture was modified by increasing the proportion of Al-waste ratio from 0 to 1%. In addition, a consistent M value of 11 for NaOH was used, while the amount of Al-waste aggregate mixed with FA was varied at 0, 5, 10, and 20%. For 11 M and 0% Al-waste rate, the decrease in STS values (from 0 to 20% slag ratio) is as follows. In this configuration, the proportion of variability in STS was 1.20, 4.12, and 9.45%. Moreover, a constant M ratio of NaOH at 13 was used, while the fraction of Al-waste aggregate mixed with FA in the GPC mixture was enhanced to 0, 5, 10, and 20%. In this instance, the percentage of variance in STS at a significant level of 5% was determined to be 6.44% (from 9.31 to 8.71 MPa). It was found that the STS of the mixture dropped by 35.18% as the slag ratio in the mixture increased from 5 to 20%. It was detected that the STS of the mixture decreased by 61.19%, while the quantities of Al-waste in the mixture were enhanced from 0 to 4% for 5% slag ratio and 13 M. Additionally, the M of NaOH was designated as 16 M. While the proportion of slag in the mixture was enhanced from 0 to 20%, it was detected that the STS of the mixture decreased by 5.46% (from 10.43 to 9.86).

Figure 5

STS test results.

During these investigations, the proportion of the Al-waste aggregate that included FA was maintained at a constant level, while the amount of NaOH in the mixture was varied to 11, 13, and 16 M, as shown in Table 4. Furthermore, as observed from Table 4, the percentage of Al-waste aggregate with FA was altered repetitively as 0, 1, 2, 3, and 4%, and the amount of M of NaOH was kept as 11, 13, and 16 M the concrete mixture. STS decreased as the M of the mixture increased. In Figures 5 and 6, STS test results are given.

Figure 6

STS via CS test results.

3.5 A comprehensive analysis of flexural behavior

The strength of samples that have been subject to flexure may be used as the tensile capability of GPC, when they are combined. Besides this, the flexural strength (FS) is generally greater than the indirect STS [44]. As detected from Table 5, 11 M was used as the M percentage of NaOH, and the ratio of slag ratio in the concrete mix was adjusted from 0 to 5%, 10%, and 20%, respectively. The percentage was improved from 0 to 20% in the design of the mixture, which was altered by adjusting the slag ratio. This modification occurred in the design of the mixture. Additionally, the value of M of NaOH was set as 11. While the proportion in the mix was enhanced, it was observed that this amount of variation tends to decrease. In this setup, the proportion of variation in FS was recognized to be 2.33, 6.61, and 10.11%.

The percentage was improved from 0 to 1% in the design of the mixture, which was altered by adjusting the Al-waste ratio. This modification occurred in the design of the mixture. Additionally, the value of M of NaOH was set as 11. While the proportion in the mix was enhanced, it was observed that this amount of variation tends to decrease. In this setup, the proportion of variation in FS was recognized to be 12.74, 17.08, and 21.21%.

Moreover, the percentage of M of NaOH was set as 13, and the percentage of Al-waste with FA was varied from 0 to 4%, with 5% in the GPC mix. In this mixture, the amount of variation in FS was recognized to be 47.25% for 5% slag ratio, 56.29% for 10% slag ratio, and 62.04% for 20% slag ratio. When the amount of Al-waste in this mixture was chosen as 1%, and the amount of slag ratio in the mixture was enhanced from 0 to 20%, it was noticed that the FS of the mix decreased by 25.55%. It was noticed that the FS of the mix decreased significantly.

Moreover, the M of NaOH was consistently chosen as 16. While the percentage of slag in the mix was improved from 0 to 20%, it was noticed that the FS of the mix decreased by 18.53% (from 3.13 to 2.55). It was noticed that the FS of the mix reduced by 14.11% when the amounts of Al-waste in the mix increased from 0 to 1% and the slag ratio was chosen as 20%. In this investigation, the percentage of Al-waste aggregate with FA was kept constant, and the M of NaOH in the mix was altered to 11, 13, and 16 M, as shown in Table 5. The results of this study are presented in Table 5. In addition, as can be seen from Table 5, the percentage of Al-waste aggregate with FA was altered many times to obtain values of 0, 1, 2, 3, and 4%, while the M of NaOH was maintained at 11 M, 13, and 16 M in the concrete mix. FS reduced as the M of the mix rises, as presented in Figure 7. The results show that increasing the heat curing temperature increases the CS and FS of almost all samples. This is similar to the studies conducted in the literature [45]. Extending the heat curing duration and elevating the temperature provide a more conducive environment for the development of GP gels [46]. Because of the non-dissolved particles in the gelatinous matrix of materials, a high alumina concentration will induce fractures and a more heterogeneous microstructure, which will lower the strength [47].

Figure 7

FS test consequences.

3.6 Mechanical property correlations of Al-waste

Concrete specimens’ CS to STS and CS to FS are shown graphically in Figure 8. Figure 8 shows the results of a linear regression study comparing the compressive-STSs of two samples: 0% Al-waste M11 and 1% Al-waste M11. The correlation values for the two samples are R = 0.9213 and R = 0.9845, respectively. It was determined that the correlation relationship between 0% Al-waste M61 and 1% Al-waste M16 was R = 0.9624 and R = 0.9839. The correlation association between 0% Al-waste M11 and 1% Al-waste M11 was determined to be R = 0.9913 and R = 0.9992, respectively, when looking at the linear regression study between CS and FTS, as shown in Figure 8. The correlation relationship between 0% Al-waste M16 and 1% Al-waste M16 was determined to be R = 0.9912 and R = 0.9802, respectively, when the linear regression analysis between CS and FTS is shown in Figure 8. Within the scope of this investigation, there exist significant correlations that exceed 95% for the predicted links.

Figure 8

Relation between CS and STS (a) and between CS and FTS (b).

3.7 Thermal gravimetric analysis (TGA)

TGA is an analytical method that monitors the change in weight of materials as a result of temperature changes. Thanks to this technique, information about the composition and properties of the material is obtained. Through TGA, thermal behaviors of materials such as degradation, thermal decomposition, water loss, and gas absorption are generally observed. The temperature for the sample to be analyzed is increased within a predetermined program. As the sample heats up, changes such as evaporation, decomposition, and oxidation occurring on the material can cause mass loss or gain. The weighing device continuously records the weight changes of the sample. Thus, the obtained data give information about the thermal behavior of the material when correlated with temperature.

The results presented in Figure 9 illustrate the behavior of the provided sample as it undergoes thermal changes from 0 to 600°C. The blue line depicts the alteration in mass. As the temperature of the initial 9.528 g sample reached 600°C, its mass decreased to approximately 94%, resulting in a 6% loss of mass. In the TGA analyses conducted by Chouksey et al. [48] and Aliabdo et al. [49] for GPC, mass losses of 6–8% and 9–13% were observed upon heating to 800 and 900°C, respectively. This comparison leads to the inference that the achieved outcomes can be more favorably appraised due to their reduced mass loss [48,49].

Figure 9

TGA of the GPC sample.

It was emphasized that the mass loss detected between 100 and 200°C is primarily attributed to the release of physically and chemically bound water. In the 200–600°C range, the weight reduction is considered to result from the decomposition of residual organics and weakly bonded phases [50,51]. Mass changes occurring above 600°C were interpreted as being indicative of partial structural degradation. The differential thermal analysis (DTA) outcomes are illustrated by the green line in the chart. This leads to a peak within the DTA curves. The peak around 500°C in the DTA thermogram is indicative of a phase transition. Derivative thermogravimetry (DTG) involves calculating the rate of change of a material’s temperature with respect to time. Essentially, it quantifies how swiftly the material’s temperature changes. Within this technique, the alterations in temperature as a material’s temperature is systematically elevated are documented. Fundamentally, its purpose is to determine the material’s thermal transitions between phases. As depicted by the red line in the diagram, the DTG outcome demonstrates the points of lowest and highest temperature fluctuations in the sample as time progresses [48].

3.8 Examination with a scanning electron microscope

To evaluate the microstructural properties of the GPC specimens, scanning electron microscopy (SEM) analysis was performed. The SEM images presented in Figure 10a–f provide a detailed illustration of the effects of Al-waste addition on the internal structure. Figure 10a shows the microstructure of the control specimen without Al-waste. Numerous pores and unreacted FA particles are observed. This indicates incomplete geopolymerization and insufficient formation of binding gel, which may negatively affect the mechanical strength. In Figure 10b, distinct N–A–S–H formation is evident, along with the presence of a GP gel. This structure suggests the development of a hybrid binding system resulting from both geopolymerization and the presence of calcium-containing components. However, the existence of microcracks may be attributed to internal stresses or volumetric changes during setting or curing, potentially compromising long-term durability. Figure 10c and d corresponds to specimens containing Al-waste. A more porous structure is evident in these images. The observed pores are considered to result from the foaming effect caused by the release of hydrogen gas when Al-waste reacts in the alkaline environment. Increased porosity, particularly at higher Al contents, can adversely affect mechanical properties such as strength and density. Figure 10e displays a magnified view of Al fibers. The distribution and orientation of these fibers within the matrix can positively influence mechanical performance by inhibiting crack propagation and improving impact resistance. However, the effectiveness of this contribution depends on the dosage and the fiber–matrix interfacial bonding. Additionally, the formation of C–A–S–H positively affects the strength gained by the concrete after it hardens. Figure 9f focuses on the aggregate–matrix interface. As shown in Figure 10f, the incorporation of Al shavings into the GP matrix affects the interfacial transition zone (ITZ) formed between the aggregate and the binder phase. According to the literature, metals can form either strong or weak ITZ structures with the binder, depending on their surface reactivity and passivation tendencies within the GP environment. In this context, the uneven distribution of Al particles, surface oxidation, and corrosive reactions in an alkaline environment can lead to regions of low bond strength. Consequently, cracks are more likely to initiate and propagate at these weak points under mechanical load [52]. The microstructural integrity of this region is critical for effective load transfer and has a direct impact on the overall mechanical performance [28,53–55]. The high CaO content slag leads to the formation of C–S–H and C–A–S–H gel phases alongside the conventional N–A–S–H gel within the GP matrix. This hybrid gel system enhances early-age strength and contributes to a denser and more durable microstructure. In particular, C–A–S–H gels improve water impermeability and ductility. However, potential long-term issues such as alkali carbonation, thermal instability, and phase separation may arise. Therefore, the quantity and stability of these gel phases in slag-blended GPs should be carefully optimized [56,57].

Figure 10

SEM micrographs of the obtained samples.

3.9 Evaluation with energy dispersive X-ray (EDX) analysis

EDX analysis, also known as or electron scattering X-ray analysis, is a material characterization technique used to detect the composition present on the surface of a sample. By directing a thin beam of electrons onto or into the sample, various interactions are initiated in the sample. EDX analysis uses these emitted X-rays to understand the chemical composition of the elements present in the sample. Each element exhibits a distinct X-ray energy spectrum. EDX analysis detects these energy spectra to determine the type and abundance of elements in the sample [32,58,59]. In Figure 10, the contents of elements Al, Si, O, Ca, Fe, Na, K, and C resulting from EDX analysis are presented in terms of weight and percentage. The presence of elements in Al-containing GPC has been demonstrated within the scope of this EDX study.

In the sample (Figure 11a), Al(K): 9.56% (weight). In Figure 11b, Al(K): 6.14%. The increase in Al content shows that metallic Al doping in chip form is directly reflected in EDX analysis. This suggests that adding Al strengthens the aluminosilicate structure during geopolymerization reactions by increasing the formation of Al–O–Si bonds, which improves the matrix integrity. Regarding silicon (Si) content, Si(K) in sample A: 19%; Si(K) in sample B: 20.55%. A significant increase in Si content is observed with the addition of Al-waste. This indicates that the interactions between alumina and silica in the matrix are more intense and favor the formation of C–A–S–H or N–A–S–H-like geopolymeric gel phases. The presence of sodium (Na) among the other elements reflects the effect of the activators (NaOH and Na₂SiO₃). A slight decrease in the calcium (Ca) and iron (Fe) ratios was observed with the addition of Al-waste. This suggests that the matrix is concentrated on Al and Si, and the influence of other elements is reduced. From the SEM images, a denser and more compact microstructure is observed in sample B, suggesting tighter interparticle bonding. This finding supports the hypothesis that the addition of Al-waste improves matrix integrity.

Figure 11

EDX spectra of the obtained samples. (a) EDX-1 and (b) EDX-2.

The Al CNC waste used in this study had various effects on the GP system with high alkali content at both the chemical and physical levels. The following reaction occurs when metallic Al interacts with strong bases, such as NaOH: this reaction forms sodium aluminate [NaAl(OH)₄] solutions and releases a significant amount of hydrogen gas.

This reaction results in the formation of sodium aluminate (NaAl(OH)₄) solutions and the release of a significant amount of hydrogen gas. This accelerates the geopolymerization mechanism by increasing the ambient pH, contributing to the formation of Si–O–Al bonds at an earlier stage and shortening the setting process. Additionally, Al particles are quickly surrounded by GP gel phases, creating a passivation effect on their surfaces. This passivation prevents both further reaction of the particles and excessive porosity in the system. At the same time, these surface layers act as nucleation centers that promote the growth of gel phases and homogeneous matrix formation. This can positively affect strength increase, especially at early stages. Particle size distribution also plays a decisive role in these interactions. The exothermic reactions observed due to Al doping caused local temperature increases and accelerated gel phase formation, reducing the setting time. However, uncontrolled hydrogen gas outgassing can lead to pore formation, resulting in a decrease in mechanical strength. Therefore, control mechanisms related to particle size, distribution, and surface reactions are critical for system performance. Further studies on this subject can be conducted. In conclusion, SEM images and EDX analyses demonstrate that Al doping is homogeneous within the GP matrix. In conclusion, SEM and EDX analysis demonstrate that Al doping can be homogeneously dispersed in the GP matrix and integrated with silicate-based phases. Under appropriate conditions, the Al sawdust additive can accelerate geopolymerization processes, shorten the setting time, and increase the final mechanical strength.