A comparative study of the utilization of synthetic foaming agent and aluminum powder as pore-forming agents in lightweight geopolymer synthesis

2.1 Materials and preparation

The fly ash of class C from Paiton Power Plant was used in this experiment as a raw aluminosilicate source. Table 1 shows the chemical composition of this material measured by X-Ray Fluorescence (XRF). An alkali activator solution was prepared by mixing the prepared 7M sodium hydroxide (NaOH pellet, technical grade) solution with sodium silicate solution (Na₂SiO₃, technical grade). The dissolution of sodium hydroxide is an exothermic reaction; thus, the solution was cooled for 24 hours before it was mixed with sodium silicate.

The synthetic foaming agent and aluminum powder as pore-forming agents were used in this study. Synthetic foaming agent is a chemical reagent containing sodium laureth sulphate that facilitates the formation of foam which would be suitable for the production of lightweight concrete with a density of 1000 kg/m³ and above [11]. The foaming agent was a colorless liquid with a specific gravity of 1.05, whereas the aluminum powder was a silver-colored powder with a specific gravity of 2.7. The aluminum powder is commonly used to obtain hydrogen gas produced from chemical reaction that tends to generate cavity in mortar [12].

2.2 The synthesis of geopolymer

The geopolymer paste was prepared by mixing the alkali activator solution with fly ash, followed by a 5-minute stirring process. The Si/Al ratio of 4 [13] and solid/liquid ratio of 2.3 [14] were used to synthesize the geopolymer. The addition of pore-forming agent was performed separately to produce the expanded paste. The geopolymer paste was then casted in cubical moulds with dimensions of 5 x 5 x 5 cm³ and cured for 28 days at room temperature.

2.3 The measurement of density and compressive strength

The dry density measurement for both the lightweight geopolymer types were performed on the cured specimens according to ASTM C138. The density of all samples was determined by dividing the mass of dry lightweight geopolymer by its volume. The compressive strength of the lightweight geopolymer was measured at 7, 21, and 28 days by using the compressive strength machine test instrument at the Physical Laboratory of PT Semen Indonesia (Persero), Tbk. The evaluation of the mechanical strength was performed according to ASTM C109. Table 3 compares the lightweight geopolymers produced from the synthetic foaming agent with that of aluminum powder.

2.4 Characterization of geopolymer specimens

The microstructure of the lightweight geopolymer was characterized by using the X-Ray Diffraction (XRD) PHILIPS Xpert MPD with Cu-Kα radiation at a scanning range from 10ᵒ to 70ᵒ. Functional groups of geopolymer products were identified by using the Fourier Transform Infrared Spectroscopy (FTIR) Thermo Scientific Nicolet IS10 with the wavenumber range of 500 – 4000 cm^-1. The surface morphology for both the lightweight geopolymer types were characterized by using the Scanning Electron Microscope (SEM) FEI Inspect 850. All types of lightweight geopolymers were coated with gold palladium metal prior to their analysis in order to enhance their material conductivity.

Ethical approval: The conducted research is not related to either human or animal use.

3.1 The effect of different pore-forming agents on geopolymer structure

The geopolymerization was initiated with the dissolution of aluminosilicate source into an alkaline activator solution, resulting in a dissolved alumina-silica species, followed by the initial polimerization which was aided by the alkaline media to produce oligomers. This process was continued to a supersaturated condition. Afterwards, the gel production promotes the structure transformation into an amorphous, solid bound structure [15]. Diffractograms seen in Figure 1(A) and (B) belong to the lightweight geopolymers that were synthesized by utilizing the foaming agent lightweight geopolymer and aluminum powder, respectively.

Figure 1

The diffractogram of (A) geopolymer without and with foaming agent addition and (B) geopolymer without and with the addition of aluminum powder.

The fact that most peaks in the diffractograms are broad, indicated the existence of an amorphous phase rather than sharp configurations. The broad peak at 2θ between 20ᵒ until 40ᵒ confirmed the formation of a polymer structure of the geopolymer [16]. All the diffractogram patterns show two types of phase: quartz and mullite, but with different intensities for both lightweight geopolymer types. The intensity of quartz phase (Iq) and mullite phase (Im) are shown in Table 2 and the quartz phase weight fraction (Wq) was calculated according to Eq. (1).

High intensity of quartz peak at 2θ of 26.55ᵒ confirmed that some part of the aluminosilicate source did not react after geopolymerization. The peak intensity decreased as the ratio of the foaming agent was added into the paste was made larger. This means that more aluminosilicate source was converted into the amorphous phase. Apart from that, the addition of 0.01%wt Al powder to geopolymer paste produces the least value of Wq intensity. The more Al powder added to geopolymer matrix, the higher Wq intensity which indicates that there was more unconverted quartz phase, therefore, not forming amorphous geopolymer matrix. Generally, the amount of aluminum powder added into the geopolymer paste changed the peak intensity rather than affecting the crystal phase of the lightweight geopolymer.

3.2 The effect of pore-forming agent on geopolymer functional groups

The functional groups of the lightweight geopolymers with and without the pore-forming agent additions were characterized by using Infrared Spectroscopy, as shown in Figure 2. Infrared spectra of geopolymers without the pore-forming agent addition (GP) showed typical peaks at 1000 and 600 cm^-1 that are related to the stretching vibration bands of Si-O-Si and O-Si-O, respectively. In addition, another peak could be observed at wavelengths of 970 and 800 cm^-1 due to the bending vibrations of Al-O-Si and asymmetric vibration of Si-O-T, respectively. The other FTIR spectra showed the same vibrational peaks with different intensities, as it could be seen in Table 3. The intensity of FTIR peak was related to the number of functional groups that existed in the synthesized geopolymer. The addition of foaming agent and aluminum powder increased the intensity of several functional groups and, consequently, the amount of soluble silica necessary for polymeric structure formation with amorphous aluminosilicate gel phase. Abdullah et al., obtained a FTIR spectrum of lightweight geopolymer with sharper peak, indicating the formation of a new product after the dissolution of the aluminosilicate in an alkaline environment [10].

Figure 2

FTIR spectra of (A) lightweight geopolymer without and with foaming agent addition with ratio of 1.5:1 and 1.5:2; (B) lightweight geopolymer without and with aluminum powder addition with ratio of 0.01 %wt and 0.02 %wt.

Both types of lightweight geopolymers showed broad and sharp bands at wavenumbers of 3600 – 3000 and 1650 cm^-1 related to stretching and bending vibrations of H-O-H, respectively. The presence of those vibrational peaks were related to the water molecules that were absorbed onto the surface or trapped inside the defect cavity of the lightweight geopolymer [17]. The larger the pores formed on the geopolymer; the more water molecules would be absorbed into the pore structure of the geopolymer. Hence, the increasing intensity of H-O-H peak indicated by the development of the pore structure that results in a lighter structure. However, this would also, inevitably, cause the reduction of the compressive strength. The intensity of the broad peak at the wavenumber of 3600 – 3300 cm^-1 decreased as the aluminum powder weight ratio was made higher than 0.01 %wt. This was the result of aluminum powder addition quickening the reaction that produced hydrogen gas leading to the increase of the collapsing pore structure.

3.3 The effect of pore-forming agent to geopolymer surface morphology

Scanning Electron Microscope (SEM) was used to observe the surface morphology of the geopolymer with and without the pore-forming agent addition as seen in Figure 3. All the geopolymer morphology images were taken with 2500x magnification. The surface morphology of fly ash geopolymer without the addition of the pore-forming agent displayed some rough parts, indicating unreacted fly ash spherical grain. Yunsheng, et al. found that unreacted particles could produce defects, thus reducing the geopolymer compressive strength. Apart from that, some fractional lines could also be observed in the images. This is related to the quick loss of water [18].

Figure 3

SEM of geopolymer sample for (a) surface geopolymer without pore-forming agent addition, (b) cross section of geopolymer - GP, (c) GP+FA 1:1.00, (d) GP+FA 1:0.67, (e) GP+FA 1:0.50, (f) GP+AP0.01 %wt, (g) GP+AP0.03 %wt and (h) GP+AP0.04 %wt.

Geopolymer is an amorphous aluminosilicate polymer that possess some crystalline structure, such as quartz and mullite shown in XRD results. The presence

of this amorphous phase makes the reaction between aluminosilicate in raw materials with alkaline activator solution to form polymeric chains through polymerization possible. Furthermore, the presence of crystalline phases of quartz and mullite hinders some aluminosilicate components in fly ash from being involved in geopolymer formation. They are shown in the SEM image as some round-shaped particles.

The different ways on how bubbles developed and generated different sizes and foam distribution on geopolymer concrete were studied. In this study, the introduction of air bubbles was performed by executing the process of prefoaming which used the foam generator. After the production of air bubbles, the foaming agent was then introduced into the geopolymer paste and voids were consequently formed. The product produced through this kind of method was known as Cellular Lightweight Concrete (CLC). The surface morphology of lightweight geopolymer with low pore-forming agent, both foaming agent (FA) and aluminum powder (AP), additions showed less voids as shown in Figure 3(c) - (e). Contrary to that, bigger voids were observed on the surface when the geopolymer paste content was made lower. The higher number of pores generated on the geopolymer surface resulted in lower compressive strength. The pore distribution on the geopolymer surface also affected the density and compressive strength of the geopolymer.

In this study, aluminum powder was added into the geopolymer paste based on Non-Autoclaved Aerated Concrete (N-AAC) where the aluminum powder was mixed directly without undergoing the autoclave process. Generally, the aluminum powder lightweight geopolymer, as seen in Figure 3(f) - (h) possessed more voids. Increasing aluminum powder weight added into geopolymer paste meant that the air voids produced were bigger. The voids produced when the aluminum powder was used resulted from the generated endogenous gas from the mixture of fine aluminum powder and geopolymer paste. Zhang et al. [15] stated that aluminum was a reactive metal that had reacted with water and hydroxide in the presence of alkaline, i.e. sodium hydroxide, to produce hydrogen gas and hydrolyzed metal complexes. The reaction stated on Eq. (2) took place when aluminum was added into the geopolymer paste.

When lightweight geopolymer was synthesized, aluminum powder as a pore-forming agent was added to a mixture of aluminosilicate source and alkaline activator solution that contains some water and sodium hydroxide (NaOH). Water as alkaline solid solvent and hydroxide ionic alkaline solution would react with aluminum powder to generate aluminum tetrahydroxide ion and hydrogen gas. The presence of hydrogen gas as a product of this reaction led to the formation of the pore structure and expanded the geopolymer paste, promoting higher chances of yielding a lighter material.

3.4 The effect of pore-foaming agent addition on the density and compressive strength of geopolymer

The density and maximum compressive strength of the unfoamed geopolymer were 2255 kg/m³ and 51 MPa after 28 days of curing, respectively. The density and compressive strength of all geopolymer specimens as shown in Table 4. The addition of foaming agents into the geopolymer paste promoted air bubble production that led to pore structure formation; hence, the density of the geopolymer matrix decreased. This, however, resulted in lower geopolymer compressive strength, weakening the formation of polymer cluster linkage and producing cracks of destruction. The foaming agent lightweight geopolymer exhibited the lowest density at the variation of GP+FA (1:2.00), with the lowest compressive strength of 0.2 MPa. Abdullah et al., used foaming agent to produce lightweight geopolymers with a compressive strength of 18.1 MPa after 28 days of curing [10]. In this study, when the ratio of the geopolymer paste to the foaming agent was made 1:1.33, the highest geopolymer compressive strength was observed to be 33 MPa with a density of 1766 kg/m³, while the addition of 0.01 %wt aluminum powder resulted in the highest compressive strength of 42 MPa with a density of 1830 kg/m³. The foaming agent improves the pores formation within the geopolymer; thus, the more amount foaming agent used, the more pores will be formed. Hence, increasing the amount of foaming agent would decrease the geopolymer density since there will be more pores formed within the specimen itself. On the other hand, more pore formation would also result in less compressive strength of the geopolymer since the strength distribution will be non-uniform along the geopolymer surface.

This correlation could be applied to the aluminum powder lightweight geopolymer. While the weight percentage of aluminum powder was increased, both the density and compressive strength of the geopolymer decreased. The lowest density and compressive strength were observed when 0.15 %wt aluminum powder was used. When the composition of the aluminum powder was higher than 0.01 %wt, the formation of hydrogen gas became faster, more pores are produced as the casting and gel hardened [19]. According to SNI 03-3449-2002, the minimum required compressive strength of the structural lightweight construction is 17.24 MPa with a maximum density of 1850 kg/m³. Based on this regulation, the density of the foaming agent lightweight geopolymer (1766 kg/m³) and aluminum powder lightweight geopolymer (1830 kg/m³) is near the requirements, hence they have the potential to be utilized as an alternative to structural binder construction.