1. Introduction
Geopolymers have been extensively studied as construction materials. It is a technology that was developed to find a more environmentally friendly option than traditional Portland cement [
1]. Generally speaking, a geopolymer is an inorganic synthetic polymer that is generated by aluminosilicate materials and alkaline agents. After curing, a semi-crystalline amorphous material similar to concrete is produced [
2]. The curing reaction can occur by adding external heat, i.e., curing in an oven and ambient temperature, depending on the composition of the geopolymer [
3]. It should be noted that the curing conditions have a significant effect on the strength development of geopolymers [
3].
A wide variety of aluminosilicate reagents can produce geopolymers [
2]. The most common sources of aluminosilicates used for geopolymer production are metakaolin and by-products from other industries, such as fly ash, mine tailings, red mud, and slag [
4,
5,
6,
7] Within these raw materials, the use of by-products from other industries for the production of geopolymer concrete was recently encouraged since this significantly reduces the amount of CO
2 emitted for the manufacture of these materials [
8]. Geopolymer concrete (GPC) is estimated to reduce the carbon footprint of construction projects by 80% compared with ordinary Portland cement (OPC) [
9].
Geopolymer precursor materials, both in natural and by-product forms, should be rich in alumina (
) and silica (
) contents, preferably in the reactive amorphous form [
5], for good dissolution of these compounds upon contact with the alkaline agent. The function of silica and aluminum is to impart strength and set the cement [
10]. A concern related to aluminosilicate dissolution is the rate at which it occurs and the amount of total amorphous aluminosilicate material available for geopolymerization [
11], as the curing time and the compressive strength subsequently obtained will depend on this [
3].
In addition to the aluminosilicate reagent, an alkaline activator is needed to produce a geopolymer. The alkaline activator causes the dissolution of the raw materials [
12]. The type and concentration of the alkaline activator should be carefully selected because its composition affects the dissolution of the aluminosilicate source and has different impacts on the properties of fresh geopolymer pastes and the development of compressive strength in hardened geopolymers [
13]. The most common activators are alkaline hydroxides and silicate solutions, and within these groups, sodium hydroxide (NaOH) and sodium silicate (
) are the most commonly used, respectively. The ideal concentration of the alkaline activator increases the strength of the geopolymer [
3]. In addition, an increase in the concentration of the alkaline activator leads to a rise in the pH of the activating solution. Different authors recommend working with pH values above 13 for the correct dissolution of the aluminosilicates present in the raw material [
9,
14,
15]. Geopolymers’ strength development depends on the raw materials and the alkali-activating solutions [
3].
When activating the aluminosilicate source with NaOH, the reaction starts with the dissolution of Al and Si, which are precursor particles in the alkali solution, where reactive aluminate and silicate monomers are released [
2]. Then, these monomers interact to form aluminosilicate oligomers, and the latter polymerize in the alkaline environment to form geopolymer gels. At the onset of geopolymerization, an aluminum-rich gel phase is generated, transforming into a final silicon-rich geopolymer gel [
2,
16].
The result based on a geopolymerization reaction is a semi-crystalline amorphous substance composed of solid phases of aluminosilicates assembled on the basis of
and
linkages as tetrahedra forming a 3D structure. The main hydration product of low-calcium or calcium-free geopolymers is an N-A-S-H gel, which possesses a 3D structure [
17].
From another point of view, geopolymers are a manufactured material that offers several advantages, including good mechanical strength (similar or superior to ordinary Portland cement) and the ability to encapsulate hazardous waste [
3,
18], as well as being resistant to water and high temperatures [
19]. The compressive strength of geopolymers is a critical factor in the construction field; for this reason, geopolymers are being widely studied and show promise as a more environmentally friendly alternative to Portland cement concrete [
20].
For the generation of geopolymers, different variables must be considered, such as the source of aluminosilicates, the type of alkaline activator, and the addition of external heat for curing, since these variables have a notable impact on the mechanical properties of the hardened geopolymers.
Therefore, this study sought to analyze the effect of the variation of the alkaline activator on the properties of geopolymers based on copper flotation tailings. For this purpose, two reagents were used, sodium hydroxide (NaOH) and sodium silicate (), where geopolymers activated with 100% NaOH and geopolymers activated with 100% were made, keeping the tailings/activator ratio constant. The purpose of using these reagents alone without incorporating both reagents together is to analyze the effect that each reagent has on the development of the compressive strength of the geopolymers.
3. Methodology
For the formulation of the geopolymeric mixtures, the type of alkaline activator used was varied, formulating two different mixtures: for the first case, geopolymeric mixtures activated with 100% NaOH were used, and for the second case, geopolymeric mixtures activated with 100%
were used, keeping the ratio in weight tailing/alkaline activator constant in both cases to analyze which activating reagent had a more significant influence on the development of the compressive strength of the already cured geopolymers at the same concentration in weight but different molar concentration [
1]. The details of the mixtures made and their compositions are shown in
Table 6.
Figure 6 shows each geopolymer mixture’s chemical compositions (wt.%) in the SiO
2–Na
2O–Al
2O
3 ternary diagram. From this, it can be observed that the composition of the geopolymers varied approximately from 63% silica by weight for the sample activated with sodium hydroxide to 72% silica for the sample activated with sodium silicate due to the addition of soluble silica by this reagent; in addition, the sample activated with sodium hydroxide had a higher percentage of sodium oxide in its composition due to the sodium added by part of the reagent used. On the other hand, both mixtures presented similar alumina amounts in their composition. It should be noted that both mixtures fell in the range of the geopolymeric phase described in the SiO
2–Na
2O–Al
2O
3 ternary phase diagram of the study by Juengsuwattananon et al. [
24]. As expected, both blends formed the geopolymer gel since they both hardened after curing; therefore, both blends were in the correct range for geopolymer formation according to their composition.
As mentioned above, the tailings had to be homogenized in the first instance, which was achieved by forming a tailings pulp of Cp 70%. After this, the homogenized tailings were filtered to obtain a tailings pulp of Cp 83%, which was the solids content required in the method used in this study to manufacture geopolymers (method VII of the study by Castillo et al., [
1]). The use of dry tailings was avoided, as in previous studies carried out in the same line of study [
1], because industrially, the tailing drying process is not very efficient since water has to be added later for the formation of the geopolymeric mixtures. After this, the number of tailings to be used in each mixture was weighed separately and sodium hydroxide and sodium silicate were added in the appropriate amounts to each mixture according to
Table 6. Then, each mixture was mechanically agitated for 15 min to obtain a homogeneous geopolymer paste. After agitation, each mixture was poured into cylindrical steel molds that were 10 cm high and 5 cm in diameter, where vibration was used for 5 min to eliminate the air bubbles trapped inside the mold. The molded mixtures were then cured in an oven for 7 days at 90 °C for hardening. After curing, the specimens were demolded and prepared for testing through different characteristic tests.
To analyze the properties of the fresh geopolymer pastes created, rheology, pH, and temperature tests were performed on them prior to curing since these properties are essential to analyze, for example, the eventual transport of these mixtures by pumps for later use, as well as to analyze whether the mixtures are in the correct range of alkalinity; pH measurements of the fresh geopolymer pastes were carried out using a DFRobot digital pH meter, previously calibrated between pH 7 and 13, at a controlled temperature of 20 °C. At the same time as the pH measurements, temperature measurements were performed on the fresh geopolymer pastes using a digital thermometer to obtain the temperature of the mixtures at all times during agitation. Rheological measurements of the geopolymer pastes were carried out using a Haake RheoStress 6000 rheometer with an FL100 paddle probe using a shear rate of 0.1 1/s for 60 s.
Several tests were performed to characterize the hardened geopolymers, i.e., after curing, including compressive strength, permeability, leaching (TCLP), X-ray diffraction, and automated mineralogy, with the latter carried out to analyze the formation of crystalline phases during geopolymerization. Simple compression tests were performed in Geocontrol’s laboratory under ASTM D7012-10 in a concrete press. The hydraulic conductivity (permeability) was obtained using a flexible wall permeameter at constant load, where the test was performed in Geocontrol’s laboratory under ASTM D5084-00. The characteristic leaching test (TCLP) was carried out to guarantee the safety of using this new material called a geopolymer since one of the characteristics that a geopolymer must have is that it must be non-toxic and non-leachable. Andes Analytical Assay Laboratory performed this test under US EPA Method 1311, where an acetic acid solution with a low pH of 2.8 was used as a leaching solution with an extraction cycle of 18 h.
5. Discussion
According to Glukhovsky [
28], there are three stages in the synthesis of geopolymers, and these stages are:
During the dissolution stage, the high concentration of
in the system is responsible for the breakage of Si-O-Si, Si-O-Al, and Al-O-Al bonds of the soluble glassy aluminosilicate phases, forming Si-OH and Al-OH groups [
29], causing an increase in the pH of the medium, as observed in
Figure 11. The appearance of ≡
-
-
bonds prevents the Si-O-Si bonds from reforming again. The aforementioned alkali silicate bonds can contribute to ion exchange and types of complexes such as Si-O-Ca-OH. With Al-O-Si bonds, the same type of reaction happens and ends up generating predominantly
-type complexes. The -
-
-
complexes are stable in an alkaline medium, thus fulfilling a transport role and allowing the development of a coagulated structure (colloidal phase) based on the abovementioned units.
In the coagulation–condensation stage, due to the high pH > 11, the disintegration of the Si-O-Si bond gives rise to hydroxylated complexes, with as the most stable component, which condenses and forms a new Si-O-Si bond. The ion acts as a catalyst for the reaction. The polyhydroxylan groups are formed by the polymerization of orthosilicic acid and can grow in all directions, leading to the formation of colloidal particles. The formation of the hydroaluminogel depends on the pH of the medium.
For the condensation–crystallization stage, in addition to the microparticles formed from the condensation, the particles of the solid phases coming from the source (copper flotation tailings in the case of this research) indicate the precipitation of products that are dependent on the mineralogy and chemistry of the initial phase, as well as the nature of the alkaline component and the hardening conditions.
According to the study by Jianhe Xie et al. [
30], the simple compressive strength could be improved due to the coupling effect of calcium aluminosilicate gel (C-A-S-H), calcium silicate gel (C-S-H), and sodium aluminosilicate gel (N-A-S-H) in the hydration of alkali-activated fly-ash-based geopolymer products; however, for the present study and given that the percentage of calcium present according to the ICP test was 3.28% in the copper flotation tailings, the same analysis cannot be done. It is suggested to carry out future research by making geopolymeric mixtures with calcium addition and studying their behavior.
The carbonation mechanism in alkali-activated geopolymers is different from that in Portland cement since, in Portland cement pastes, atmospheric CO
2 dissolves in the pore solution and reacts rapidly with portlandite to form CaCO
3 and then with a calcium-silicate-hydrate (C-S-H) gel to form CaCO
3 and silica gel; in contrast, the carbonation of alkali-activated pastes occurs directly in the calcium-aluminosilicate-hydrate (C-A-S-H) gel due to the lack of portlandite, leaving an alumina-containing silica gel remaining, in addition to CaCO
3 [
31,
32]. Relative humidity of 95% in the curing chambers can inhibit initial carbonation in geopolymers [
33].
On the other hand, the dissolution of calcium in sources processed in a low alkalinity system, which was not the case in this research, provides good resistance to simple compression due to the formation of C-S-H gel (calcium-silicon-H
2O) that coexists with the geopolymer gel (N-A-S-H), which complements each other and forms a mostly amorphous structure that provides better mechanical behavior. For aluminosilicate sources, such as copper flotation tailings, very little calcium is dissolved (the percentage of calcium present according to the ICP test was 3.28% in copper flotation tailings), causing a C-S-H gel not to form. The work of Professor Engui Liu [
34] on the use of marine shells goes in this direction of taking advantage of the improved characteristics of Portland cement and geopolymers, increasing their hydration of the C-S-H matrix phases and resistance to simple compression; however, the addition of further treatment, selection, and crushing make this technique more expensive for the traditional mining industry.
In high-alkalinity systems, a higher concentration of NaOH or Na
2SiO
3, as in the case of this research, leads to the formation of N-A-S-H gel being predominant, where the role of calcium is less influential in the final product as it cannot generate a C-S-H gel that contributes to the geopolymer gel or a large amount of crystals that interrupt the amorphous gel structure; therefore, the dissolution of calcium does not impact in a determinant way in the simple compressive strength [
34].
The work of Yip et al. [
35] on geopolymer blends focused on replacing part of the aluminosilicate source with calcium, as did the work of Tian et al. [
36], which replaced part of the alkaline activator (NaOH) with industrially produced calcium oxide (CaO) (since the sources of these authors did not have soluble calcium sources) to form a geopolymer based on copper tailings from China with the same compositions in their studies and analyzed the effects of varying the curing temperature [
37] and fly ash. It was observed that by substituting 20% of the alkaline activator with CaO, this mixture provided a higher simple compressive strength, but slightly reduced the long-term simple compressive strength (28 days of aging at ambient conditions) since the best strength was obtained at 7 days. In addition, they observed that Si, Al, and Na tended to accumulate in the area corresponding to the N-A-S-H gel and the area with the absence of these elements (Si, Al, and Na) was where the concentration of Ca predominates, concluding that there was no interaction of Ca with the rest of the elements to form a C-S-H gel. On the other hand, unreacted sectors were observed with a high concentration of aluminum, which would indicate that much of the aluminum present in the tailings was in the form of non-soluble crystals; therefore, it was not diluted in the solution, as it had a structure that the alkaline solution was unable to dissolve.
The influence of pH on the development of the compressive strength of geopolymers is directly related to the dissolution of the aluminosilicates present in the raw material, which in this case was copper flotation tailings. This dissolution allowed both silicon and aluminum to go into solution and, therefore, were usable during the geopolymerization process for the formation of the N-A-S-H gel, which was ultimately responsible for providing the strength and stiffness of these materials. At this point, future work to be developed could involve the use of aluminum sulfate and/or fluorine gypsum as a possibility to increase the resistance to simple compression obtained in some percentage to be optimized (2% by weight), as in the work of Professor Yilu Wang [
38], enabling optimizations and cost reductions in the generation of geopolymers of copper flotation tailings.
The difference in the rheologies, considering that all of them were performed at the same %Cp, that is, with low water contents, lay mainly in the reagent used, which was higher in the case of sodium hydroxide relative to the sodium silicate rheology; this could have been due to the fact that sodium hydroxide continued to react in the mixture, generating higher coagulations and condensations due to the high pH measured (pH > 14), causing an early solidification of the N-A-S-H gel, as indicated by Lahlou et al. [
39], Favier et al. [
40], and Riffai et al. [
12]. This was not so for the case of Na
2SiO
3, where the pH was between 11 and 12. In addition, as Na
2SiO
3 is soluble, it provides higher apparent viscosity to the tailing particles due to the ease of soluble silica to dissolve and start to polymerize, which allowed for generating a metastable hydrated gel that served as a core for the N-A-S-H gel, as indicated by Favier et al. [
40].
6. Conclusions
According to the methodology used for the creation of the geopolymers, their composition, and the temperatures and curing times, it was deduced that these variables were in the correct ranges because, in both cases, hardened geopolymers with good mechanical properties were obtained, which is essential due to the future applications that can be given to this new material.
The characterization of the copper tailings used was fundamental since it could be demonstrated that it was mainly composed of Si and Al, which are elements that are essential for the formation of geopolymers. In addition, the mineralogical characterization and the degree of grinding of the particles was also an important parameter to consider since this indicated how much material from the aluminosilicate source was available for the geopolymerization reactions. On the other hand, the composition of the geopolymeric mixtures created was within the range studied by previous authors.
After curing the geopolymers, it was found that both presented volumetric contractions due to the loss of water that occurred during the geopolymerization process, where this change in volume was more noticeable in the mixture activated with sodium silicate.
Among the geopolymeric mixtures created, the one activated with NaOH had a more specific rheological behavior, presenting an almost linear shape after its yield point. On the other hand, the sample activated with sodium silicate showed a significant decrease in the stress required to deform this pulp after its high yield point.
It was found that the sample activated with NaOH presented higher pH and temperature values during its agitation due to the exothermic reaction of dissolution of this reagent when in contact with water. On the other hand, the mixture activated with SS had lower pH and temperature values because this reagent was not as alkaline as NaOH, and it did not present exothermic reactions during its dissolution.
Based on the mineralogy of the geopolymers, it was found that the one activated with SS presented mineralogy similar to that of the tailings without the addition of reagents. In contrast, the one activated with NaOH presented significant variation in its mineralogy, from which it can be deduced that this reagent caused more significant changes in the raw material, dissolving a greater amount of crystalline phases. The information obtained from the Qemscan tests corroborated these results.
Based on [
2], it was identified that the final characteristics of the geopolymer depended on its chemical formation, where the chemical elements Na, Al, and Si, along with H
2O, played a fundamental role in the generation of the N-A-S-H gel, where these first elements are the main compounds of plagioclase, and thus, explaining its decrease in comparative mineralogical analyses.
Based on the uniaxial compression results, it was found that the geopolymer that obtained the highest resistance was the one activated with SS, with 34.46 MPa, and then the one activated with NaOH, with approximately 23 MPa compressive strength.
On the other hand, neither geopolymer presented the characteristic of toxicity via leaching; therefore, these created materials are safe for use. Finally, it was found that both geopolymers presented very low permeability levels, raising their tailing permeability due to geopolymerization.