Thermo-Mechanical and Thermal Performance of Rice Husk Ash-Based Lightweight Geopolymer Concrete Activated by a Dual-Alkali System Incorporating Expanded Perlite and Pumice Aggregates

Table 2. Physical properties of rice husk ash (RHA), expanded perlite (EP), and pumice

Table 3. Mix proportions for rice husk ash (RHA)-based lightweight geopolymer concrete (GPC)

The results presented in this section are the mean values of the three samples. The observed pattern was very regular with a small amount of scatter on the graph, indicating that the experimental results were reliable and accurate.

As the complete substitution of EP by pumice in the lightweight GPC mix design progressed, the data on the compressive strength in Table 4 and Figure 1 showed an increase in the values. The lowest value of the compressive strength measured in Mix M1 with 100% perlite is significant at 18.6 MPa, showing how perlite cannot withstand any amount of load as a coarse aggregate material.

Table 4. Mechanical and physical results of the geopolymer concrete (GPC) mixtures incorporating expanded perlite (EP) and pumice

Figure 1. Compressive strength of rice husk ash (RHA)-based lightweight geopolymer concrete (GPC) mixes with varying ratios of pumice to expanded perlite (EP)

Note: Error bars represent ±1 standard deviation (n = 3).

Porous perlite, an expansive perlite with a very loose structure, has a better insulating capacity than its compressive strength [14]. It has less mechanical interlocking with a low compressive strength and is an expansive perlite with a looser structure.

The compressive strength values were observed to fall drastically as the percentage of pumice increased, with the maximum value at 31.2 MPa for Mix M5 with 100% pumice. This tendency conforms to the known mechanical behavior of pumice-based lightweight aggregates, which exhibit greater angularity, better aggregate interlock than perlite [15], and greater strength. Stronger bonding with the geopolymer gel, made possible by the rough surface texture of the pumice, also helps obtain denser and more coherent microstructures [16]. Furthermore, the strength growth critically depends on the binder matrix. Effective dissolution of the high silica content of RHA was made possible by a binary alkaline activator system (NaOH + NaAlO₂), which helped generate a strong sodium–alumino–silicate–hydrate (N-A-S-H) gel [2]. Additional Al³⁺ ions from NaAlO₂ improve crosslinking in the gel network, thereby producing a denser and more mechanically stable structure [2]. Several studies have found that the Si/Al ratio plays an important role in the formation of the structure and properties of geopolymer gels. Chen et al. [17] and El Alouani et al. [18] have particularly described the controlled synthesis of N-A-S-H gels of different Si/Al ratios and their thermodynamic properties and found a strong correlation between the properties of the gels and the available Al content. The addition of NaAlO₂ as an additional source of alumina has also been found to significantly alter the chemistry of the gels in a major way by increasing the concentration of Al in the gels. This, in turn, increased the three-dimensional connectivity of the N-A-S-H geopolymer structure. The dense three-dimensional structure of the geopolymer matrix makes it easier to bind with reactive siliceous aggregates. Recent studies on the mechanism of geopolymerization have also emphasized the importance of the formation of Si-O-Al bonds in the process of polycondensation in determining the structure and mechanical properties of geopolymer gels [17, 19]. In mixes with pumice, which already supported better physical bonding, the interaction at the aggregate–binder interface was enhanced. These findings are consistent with those of previous studies on lightweight geopolymers made from porous pebbles. Federowicz et al. [20] discovered that the compressive strength of lightweight concrete systems can be significantly enhanced by replacing lightweight weakening aggregates with lightweight structurally resilient aggregates, including pumice. Correspondingly, Almadani et al. [21] demonstrated that the type of aggregate and its combination with the binder matrix directly influence the mechanical strength of geopolymer composites, and this effect emerges in systems where the geopolymer gel benefits from both silica and alumina sources. Similar trends were recently reported by these studies [22, 23] for aerated polystyrene-based lightweight concretes, showing that the properties of lightweight aggregates and their compatibility with the binder play a significant role in the development of their strength and structural properties. Although mixes with a high concentration of perlite (M1 and M2) showed lower strength values, they still met the lowest structural requirements for lightweight applications, as stipulated in ACI 213R (> 7 MPa). These combinations are appropriate for non-load-bearing and thermally efficient parts. Although maintaining the advantages of being lightweight and providing better thermal insulation, the pumice-dominated mixtures (M4 and M5) had favorable mechanical properties, which can also find applications in other structural areas. Finally, especially when the alkaline activator solution was at an optimal ratio, the trend clearly indicated that pumice was more effective than perlite in improving the compressive strength of RHA GPC. These results confirm the applicability of this hybrid form to the structural conditions found in buildings.

The unique physical and chemical properties of pumice gave it greater compressive strength, which was noted in mixes with pumice, especially in M4 and M5. The angular particle shape and rough surface texture resulted in a denser and more cohesive microstructure, enhanced mechanical interlocking, and formed stronger bonds in the ITZ. In addition, the reactive aluminosilicate phases present in the pumice can contribute to secondary geopolymerization, which provides more support to the matrix. However, perlite is more porous and smoother; therefore, the bonding is poorer, and the stress transfer at the ITZ is low, reducing the compressive strength. The tendencies of the thermal conductivity also had a strong resemblance to the aggregate porosity and ITZ properties. Because of the higher air trapping in the highly porous perlite particles and surrounding matrix, which reduced the rate of heat transfer, mixtures with higher perlite contents (M1 and M2) had lower thermal conductivities. However, the pumice-based mixes (M4 and M5) had slightly better thermal conductivities owing to the better packing of the particles and reduced porosity, which allowed heat to be conducted by the continuous solid structure. This is supported by the findings of previous studies [24], which showed that the aggregate porosity and densification of the ITZ are relevant for equalizing the mechanical and thermal properties of lightweight GPCs. These results underscore the need to balance the increased ratio of perlite to pumice depending on the nature of the intended structural or nonstructural use to obtain the best balance between mechanical integrity and thermal insulation efficiency.

As the percentage of pumice replaced the EP, the hardened density values presented in Table 4 and Figure 2 indicate a continuous increase in the bulk density of the geopolymer concrete (GPC) mixtures. Mix M1 has the lowest density figure recorded—1354 kg/m³—which can be ascribed to the exclusive use of EP, a very porous and ultra-lightweight volcanic aggregate with a bulk density of roughly 145 kg/m³ [24]. The density was also increased by the addition of pumice, which is denser than perlite. Mix M5, with a composition of 100% pumice, had a density of 1689 kg/m³. From the graph, the type of aggregate is shown to affect the general mass-density relationship for GPC. Although pumice offers superior compactness and packing efficiency owing to its more solid matrix and granular texture [25], the cellular structure of EP significantly reduces the bulk mass but enhances void content. The slow change from M1 to M5 shows consistent densification of the matrix; each step produces a more solid structure without appreciably changing the binder phase. According to ASTM C330, with a maximum density of 1850 kg/m³, all measured densities fall within the approved limits for lightweight structural concrete. This validates the appropriateness of all mixes, particularly M2–M5, for structural use, where a low dead load is a major design issue. Zahabi and Said [26] stated that concrete with a density of less than 2000 kg/m³ is light and may offer both performance and durability benefits, particularly in high-rise and seismic zone structures. Furthermore, these findings are consistent with those of previous studies on RHA-based geopolymers. Particularly in the case of low-density pozzolanic binders, such as RHA, Chanda and Guchhait [27] found that aggregate density and pore structure have a significant effect on hardened density in alkaline-activated materials. The results of this investigation demonstrate that although EP is still the best choice in terms of saving weight and providing superior insulation, the addition of pumice produces concrete with a higher density, which performs better in terms of mechanical integration. In GPC, the type of aggregate must ultimately be a balance between density, strength, and thermal efficiency, depending on its use. The results showed that GPCs with customized density profiles could be produced using a targeted combination of EP and pumice, depending on the structural or envelope purpose of their application in sustainable buildings.

As the percentage of EP decreased and the pumice content increased in the concrete mixes, it was clear from the water absorption results presented in Table 4 and Figure 3 that there was a trend of decreasing absorption capacity in the concrete. With 100% EP, Mix M1 exhibited the highest water absorption value of 18.4%. Mix M5, which is made completely of pumice, recorded the lowest absorption rate at 10.1% instead. This is a significant decrease, which emphasizes the direct influence of the aggregate type on the pore structure and moisture transport behavior of lightweight GPC. Owing to the highly porous cellular structure of EP, a large amount of water is absorbed; thus, the total absorption rate of the concrete matrix is increased [24]. Although porous, pumice has a dense, more stable surface shape that reduces capillary absorption and enhances aggregate-binder-interaction, thus reducing total permeability [28]. Therefore, mixes with higher pumice content showed better resistance to water intrusion. The results obtained were in good agreement with those of previous investigations on lightweight concrete structures. Ma et al. [29] claimed that the inherent porosity of the aggregates and the quality of the ITZ between the aggregates and binder mostly control the water absorption behavior of GPC. In particular, in mixes with better packing (M4 and M5), the addition of NaAlO₂ to the activator used in this investigation contributed to greater matrix formation through a more developed N-A-S-H gel structure, which also resulted in a denser pore network. Moreover, the binder, which was finely ground with RHA, acted as a microfiller and reactive pozzolan, thereby improving the pore structure. By decreasing the open porosity and water ingress, the pozzolanic process burns calcium hydroxide and forms other binding phases [3]. Therefore, the geopolymer matrix in the pumice-rich blends exhibited a synergistic effect, resulting in a combination of lower absorption and higher strength. Even though the blends presented higher absorption than regular dense traditional concrete (usually approximately 8%), this is not a problem in the context of lightweight and nonstructural concrete, particularly if the insulation properties and self-weight are the dominant requirements. The proportion of the optimum ratio of EP to pumice should be adjusted to tailor the specific ratio according to the water absorption and thermal requirements.

Figure 2. Dry density of rice husk ash (RHA)-based lightweight geopolymer concrete (GPC) mixes with varying ratios of pumice to expanded perlite (EP)

Note: Error bars represent ±1 standard deviation (n = 3).

Figure 3. Water absorption of rice husk ash (RHA)-based lightweight geopolymer concrete (GPC) mixes with varying proportions of pumice and expanded perlite (EP)

Note: Error bars represent ±1 standard deviation (n = 3).

As the amount of EP decreased and the amount of pumice increased, the results presented in Table 4 and Figure 4 show that the thermal conductivity of the GPC mixes was enhanced. The lowest thermal conductivity, 0.189 W/m·K, was exhibited by Mix M1, while the highest thermal conductivity, 0.331 W/m·K, was exhibited by Mix M5 containing 100% pumice. This difference reflects the general characteristics of the thermal conductivities of both perlite and pumice and how their amount affects the thermal conductivity of the concrete mixture. As mentioned earlier, perlite has an exceptionally high insulation value because of the presence of cellular structures that trap replenished air pockets, thereby acting as superior thermal barriers [30]. Because of this characteristic of perlite, mixes M1 and M2, which contained perlite, exhibited lower thermal conductivities. Although pumice is lightweight and porous in nature, it has a higher rigidity in the form of minerals and is characterized by a low air entrapment content; thus, it has a higher thermal conductivity than perlite [31]. The marginal increment in thermal conductivity from M1 to M5 agrees with earlier observations made by Güneri [31]. This study clearly proved that concrete mixes containing higher amounts of pumice display higher heat conductivity because of increased connectivity and less internal air space entrapment [31]. It has also been clarified that Przybek et al. [32] demonstrated that the extent of thermal conductivity of GPCs largely depends on how lightweight materials behave and what form they possess. It should be between 1.4 and 1.7 W/m·K. The value of the thermal conductivity measured in this study was well within the range of thermally good-performing concrete and lower than that of normal-weight concrete [33]. These findings verify that, particularly for use in energy-efficient building envelopes, wall panels, and prefabricated insulating blocks, all the evaluated mixes provide significant thermal-insulation advantages. Mix M1 provides the best insulating properties, but its lower compressive strength and greater water absorption could restrict its use in structural projects. However, mixes M3 and M4 appear to be the optimum compromise between thermal efficiency, strength, and durability; hence, they are perfect options for non-load-bearing structural components that require both mechanical integrity and energy-saving characteristics. The addition of EP significantly increased the thermal insulating ability of RHA-based GPC, whereas pumice improved its mechanical performance. The optimal mix can be selected based on the desired balance between strength and thermal behavior, thereby enhancing the adaptability of this system for use in thermally efficient buildings (Table 5).

The compressive strength, density, and thermal conductivity performance trends were consistent with those of EP and pumice aggregates, whose porosity and surface properties have been thoroughly examined in previous studies [31]. These earlier results corroborate the interpretations made in this study, especially regarding the effects of aggregate porosity and interfacial bonding on the mechanical and thermal performance.

Correspondingly, mix M5 contains much more Al (11.05 at.%) than mix M1 does (10.62 at.%), but less Si. Thus, the Al/Si atomic ratio increased from approximately 0.367 in M1 to 0.414 in M5, which suggests greater incorporation of aluminum in the aluminosilicate network as well as increased cross-linking. This constitutes concrete compositional evidence that NaAlO₂ is an efficient supplier of reactive Al species that participate in geopolymerization. The mineralogical nature of the pumice accounts for the somewhat higher concentrations of minor elements (Fe, Mg, and Ca) in M5. The higher mechanical performance and reduced water absorption of M5 in the macroscopic tests are fully compatible with its higher Al/Si ratio and better polymerized gel network.

Figure 4. Thermal conductivity of rice husk ash (RHA)-based lightweight geopolymer concrete (GPC) mixes with varying proportions of pumice and expanded perlite (EP)

Note: Error bars represent ±1 standard deviation (n = 3).

Table 5. Energy-dispersive X-ray spectroscopy (EDS) atomic composition of mixes M1 and M5

LCA studies indicated that the developed RHA-based GPC could reduce the emission of CO₂ by as much as 80%. The amount of CO₂ produced per ton of cement in a conventional production of OPC is about 0.85-0.95 tons, mostly as a result of the substances and steps of the high-temperature kiln and the calcination of limestone. Conversely, with geopolymer binders manufactured using waste products like RHA, the process of clinkerization at elevated temperatures is not necessary, and the process of alkaline activation generates much less CO₂—about 0.15 to 0.20 tons per ton of binder [7]. It was estimated that the total carbon footprint reduction relative to the case of lightweight concrete, relative to OPC, was between 75% and 82%, depending on the energy source and transportation effects, because in this research, 100% RHA was used as a binder in the mix designs. These results indicate the possibility of low-carbon GPC as a green alternative in the construction sector and are not beyond the scope of recent life-cycle assessment (LCA) studies.

The findings of this study demonstrate the influence of the number of lightweight aggregates used in the mixture as well as the chemistry of the activating solution and the resulting N-A-S-H gel on the performance of RHA lightweight GPC. The use of a two-alkali activating solution of NaOH and NaAlO₂ was found to be highly significant for the alteration of the Al/Si ratio and the promotion of the formation of a highly cross-linked N-A-S-H gel.

The various properties of pumice and perlite were found to influence the trade-off between strength and insulation properties. Siliceous pumice, which is more reactive, enhances the strength of the geopolymer gel owing to its better reactivity, whereas perlite, which is highly porous, is mainly responsible for the enhancement of thermal efficiency. This highlights the need for the reactivity of the aggregate, not just the density, to be tailored when preparing lightweight geopolymer composites.

From an engineering and environmental perspective, the proposed method is a good approach for producing low-carbon and thermally efficient structural materials by converting agricultural and volcanic wastes to useful products while retaining good mechanical strength. These findings provide a mechanistic basis for the systematic development of lightweight GPCs, in which the chemistry of the activator, gel structure, and aggregate reactivity can be simultaneously tailored to meet specific requirements.