Pozzolanic potential of sustainable precursors for engineered geopolymer composites (EGC)

As concerns about environmental damage and the depletion of natural resources have grown, so has the demand for sustainable building materials. Developing pozzolanic alternatives to Ordinary Portland Cement (OPC) has become imperative in response to these difficulties. According to ACI-116R-00 [1], pozzolan (natural or artificial) is a siliceous or silico-aluminous substance that can chemically react with calcium hydroxide in the presence of water to form compounds with cementitious characteristics. Recently, Engineered Geopolymer Composites (EGC), a group of strain-hardening composites, produced by the reaction between aluminosilicate precursors and alkaline activators has been a viable solution in this regard [2, 3]. EGC are formulated based on the similar micromechanical principles of Engineered Cementitious Composites (ECC), exhibiting enhanced strain hardening behaviour viz., Pseudo Strain Hardening (PSH) with minimal fibre ratio [4]. Moreover, EGC are sustainable alternatives to conventional cement based concrete and composites, exhibiting improved mechanical and strain behaviour than the latter [5]. Also, EGC being highly durable are considered environmentally efficient by involving the pozzolans as precursors and fine aggregate substitutes by fully replacing OPC in their counterpart ECC [6, 7]. Pozzolans are gaining popularity since their application lessens the adverse effects on the environment and improves the physical characteristics of the resulting geopolymer composites (GP) [8]. Selecting appropriate precursors is one of the main determinants of the performance of EGC.

Pozzolans are usually variably sourced and available as minerals or residues of industries. Unlike OPC, they might not follow the similar heats of hydration in contact with water. The hydration mechanism followed for pozzolans differs in forming the resulting hydration products [9, 10]. The feasibility of a pozzolan is measured based on its lime removal capacity, which can be related to the Ca(OH)₂ absorption as a result of OPC hydration and forming a more stable compound, calcium silicate hydrate (C-S-H) gel, and making the binder highly impermeable as in the case of blended OPC (with pozzolan) [11]. The measure of pozzolanicity of the test pozzolans (FA, BOF slag and IOT) is considered crucial for enhanced geopolymerization and strength optimization in the proposed EGC. The chemical reactions of the test pozzolans with the alkaline activator forms the alumino-silicate gel, owing to the overall pozzolanic reaction [12, 13]. This primary dense and robust gel formation is facilitated by the amorphous Si and Al proportions present in the test pozzolans. This gel structure undergoes polycondensation to form three dimensional Al-Si network, which imparts enhanced strength and durability to the proposed EGC [14] and thereby conforming the direct relationship of the pozzolanic activity with the degree of polymerization and in turn the strength properties of the proposed EGC [15]. Additionally, the pozzolanic reaction effectively fills the pores of the EGC matrix by the formation of secondary C-S-H and C-A-H gel phases, thereby making the matrix denser upon reducing the porosity [16]. Moreover, these gel structures improve the bonding between the geopolymer matrix and the fine aggregates, enhancing the overall strength of the EGC [17].

The test methods available for assessing the pozzolanicity can be grouped as direct and indirect methods [9–11]. In this, the presence of lime % being monitored and their subsequent reduction with the pozzolanic reaction is grouped as direct method [9]. This is analytically followed by two main methods viz., Thermogravimetric analysis (TGA) and x-ray diffraction (XRD) [10, 11]. Following this reaction principle on a blend of OPC and pozzolan, the Frattini test, one of the direct methods, measures the concentrations of Ca²⁺ and OH⁻ by conventional titration methodology. A more simplified form of the Frattini method is the saturated lime test (SLT), where a saturated lime solution replaces the pozzolanic blend prepared by OPC. The dissolved residual calcium content assesses the amount of lime removal. It is to be noted that both these methods have been extensively implemented for various types of pozzolans, viz., fly ash [18–20], metakaolin [18, 21], sludge waste, silica fume [22, 23], etc.

Contrarily, indirect methods measure any particular physical characteristics of the test pozzolan, denoting their pozzolanicity. Usually, the measurement involves parameters viz., strength (compression), conductivity (electrical) [24–27], or performing conduction calorimetry by assessing the heat evolution [28, 29]. In this, the most common practice has been assessing the compressive strength attributes, which has been reported in past studies for various pozzolans viz., fly ash [30], silica fume [29, 31], ash based on sewage sludge [32, 33]. However, direct test methods are more precise and confirm the final pozzolanic activity upon comparison with the indirect method [23, 34, 35].

In this study, the Frattini test [36] and the saturated lime test are the two direct methods, while the indirect test involves the measurement of compressive strength by the strength activation index (SAI) test [37] employed to examine the pozzolanicity of the wastes produced by the thermal, steel, and iron ore mining industries, specifically fly ash (FA), basic oxygen furnace (BOF) slag, and iron ore tailings (IOT), respectively alongside M-Sand. The amount of calcium hydroxide utilized during the pozzolanic reaction is measured as part of the Frattini test to assess the precursor's reactivity. Contrarily, the saturated lime test evaluates the ability of the precursors to produce calcium silicate hydrate (C-S-H) gel, which determines the strength of EGC. While previous studies have investigated the pozzolanicity of various industrial residues individually or in combination by direct [18–23] and indirect methods [24–35], no relevant research findings have been reported on assessing and comparing the pozzolanic activity of the proposed specific combinations of FA, BOF slag, and IOT simultaneously through any of the direct and indirect methods. Most existing studies focus on utilizing these waste materials as partial replacements for cement in concrete [38–40]. However, the potential of involving BOF slag and FA as precursors in geopolymer based composites has been explored assessing strength characteristics [41, 42]. Nonetheless, the measure of pozzolanic activity for the proposed specific combination of FA, BOF slag and IOT is relatively explored less [43–45]. Moreover, existing literatures have mentioned the use of different industrial wastes in developing EGC, geopolymer concrete (GP), and blended cement and assessing their pozzolanic activity therein [24–26, 28, 46–49].

However, there is a need to investigate the pozzolanic activity on the combination of FA and BOF slag as precursors, while IOT being replaced partially with fine aggregate (M-sand) in the development of EGCs, is relatively unexplored. Thereby, the research gap owing to this research lies in the limited investigation pertaining to the feasibility by pozzolanic potential assessment of FA, BOF slag and IOT individually. The proposed test materials (pozzolans) have not been explored extensively for their pozzolanic potential to contribute as a combination when utilized as precursors and fine aggregates in the development of EGC.

To maximize resource efficiency, the current study aims to address these research gaps by evaluating and comparing the pozzolanicity of FA, BOF slag, IOT and M-sand through direct (Frattini test, saturated lime test) and indirect (strength activation index test) methods, and the corresponding correlation between those tests has been analyzed. This correlation analysis is considered essential as it aids to assess the consistency and reliability of the proposed methods towards the measurement of pozzolanic activity of the test pozzolans. A strong correlation between any two methods would indicate that they are consistently in measuring the pozzolanicity, thereby adding validation to the results. Contrarily, a weak correlation may indicate difference in aspects of pozzolanic activity, making them unreliable [50–52]. Also, this relation allows the selection of suitable methods, depending on the time and resource availability alongside improving the intended test methods [23, 50, 53]. The study also examines the potential of utilizing these industrial wastes, especially the combination of FA and BOF slag, as sustainable precursors for producing EGC. The findings will be compared to OPC as a control sample, to further comprehend their behaviour. Overall, the pozzolanic activity of FA, BOF slag, and IOT indicated by the Frattini test, SLT, and SAI can significantly contribute and directly relate towards the enhancement of geopolymerization and strength optimization, thereby assessing their effectiveness as precursors for the development of EGC [54]. Consequently, the output of the research will provide useful insights regarding effective utilization of industrial wastes from various sources for sustainable construction and building materials.

2.1. Materials and characterization

The four different test pozzolans were: viz., Fly ash (FA) (class F, NLC India Limited), BOF slag (sourced from JSW steel plant), IOT (supplied by Gogga minerals and chemicals), and M-sand (crushed granite stone sourced from Triveni M-sand Ltd, Andhra Pradesh; sieved to <150 μm). The chemical composition of the test materials, as received was determined using XRF analysis. The loss on ignition (LOI) was measured, subjecting the samples (∼3 g) to oven drying (∼100 °C) until attaining a steady mass prior to calcining at around 770 °C for about an hour, followed by cooling and again measuring weight (table 1). XRF analysis was not performed on the M-sand, considering that most of its content is composed of silicon dioxide (SiO₂). The pH was measured by mixing the test pozzolan in deionized water in the ratio (1:5), following the guidelines as per BS 7755–3.2 [55], while their corresponding specific gravity values were measured by pycnometry. The specific surface area was determined involving the Brunauer–Emmett–Teller (BET) method. All the physical properties of the test materials are summarised in table S1, and the cumulative particle size distribution data by dynamic light scattering (DLS) method is given in figure 1.

Table 1. Chemical composition of test materials—XRF analysis.

Materials	Percentage composition (%)
	SiO2	Al2O3	Fe2O3	MgO	CaO	K2O	Ti	P2O5	SO3	Mn	Ba	Zr	Zn	LOI
BOF	27.8	17.2	0.3	11.5	16.6	0.3	0.2	0.04	0.8	0.09	0.08	0.01	0.006	0.8
FA	57.3	34.0	3.7	1.6	1.3	0.9	0.6	0.4	0.1	0.04	0.03	0.02	0.01	4.2
IOT	24.9	23.9	32.8	0.6	0.2	1.03	0.4	0.08	0.01	0.08	0.01	0.01	0.008	< 1.0

Zoom In Zoom Out Reset image size

2.2. Pozzolanicity test methods

2.2.1. Fratini test and saturated lime test (SLT)

EN 196–5 [36] was followed for the Frattini test, where 20 grams of test samples were prepared by mixing 80% OPC and 20% test pozzolan in 100 ml of distilled water. Samples were sealed in plastic bottles and kept in an oven at 40 °C for 8 days. After 8 days, samples were vacuum filtered through a 2.7 μm filter paper and cooled in sealed Buchner or Hirsch funnels (figure S1). The filtrate was titrated with dilute HCl and methyl orange (MO) indicator to measure [OH⁻], and Ethylenediaminetetraacetic acid (EDTA) to measure [Ca²⁺] involving Patton and Reeders indicator. Results were plotted with (Ca²⁺ - expressed as CaO equivalent, mmol l⁻¹) on the y-axis and (OH⁻ mmol l⁻¹) on the x-axis. Additionally, the solubility curve of Ca(OH)² was also plotted and compared to the OPC (control) mix to ensure this result lies on the curve. Also, results below the line indicate pozzolanic activity, while those on the line suggest no pozzolanic activity, and those above the line indicate no pozzolanic activity. Also, this procedure assumes no source of calcium (Ca) being present, as their presence sourced elsewhere would invalidate this procedure. Also, during conventional cement hydration, portlandite is formed as Ca(OH)₂. When a pozzolan is present in this case, it reacts with Ca(OH)₂, causing more of it to dissolve until it gets depleted or the portlandite is consumed.

Conversely, the SLT [19–21] utilizes a fixed amount of Ca(OH)₂ in the solution. It involves adding 1 g of pozzolan to a plastic bottle with 75 ml of saturated lime solution. The solution is prepared by dissolving 2 g of hydrated lime in 1 l of distilled water. The bottles were sealed and oven-treated (40 °C) for 1, 3, 7 and 28 days (figure S2). They were then filtered and titrated for [OH⁻] and [Ca²⁺] as per the Frattini test. As the initial quantity of Ca²⁺ ions is known, and the same is reactive only with the test pozzolan or water, the amount of lime absorbed by the materials can be calculated therein. Results are expressed as mmol CaO or in % of CaO fixed per gram of test pozzolan. Notably, portlandite's solubility decreases as temperature rises due to its negative heat of solution, causing few dissolved Ca(OH)₂ at nominal room temperature to initially precipitate at 40 °C before re-dissolving. Despite the reaction with the pozzolan under consideration, Ca(OH)₂ quantity remains fixed in the system, with elevated temperature ensuring a rapid reaction.

2.2.2. Strength activity index (SAI)

The method involved preparing control mortar blocks as per the guidelines based on BS 3892 [56] by mixing OPC with sand and water in a suitable mortar mixer. The preparation of test samples was similar, with an addition of 20% pozzolan as a replacement to the OPC. The overall mix proportions formulated are shown in table 2. The concentration of Ca(OH)₂ was determined based on the weight of test pozzolan considered. The weight ratio between Ca(OH)₂ and the test pozzolan is an essential parameter and in this SAI test, the ratio is 1: 1. In order to attain the same flow characteristics as the control mortar, flowability tests were performed on pastes made in accordance with EN 1015–3 [57] by adjusting the water-to-binder ratio. 50 mm cube specimens (6 No.s) of mortar paste were cast for 7- and 28-day strength periods, upon further mixing for an additional 30 to 40 s. The specimens, upon being demoulded, were subjected to water curing. After surface drying, they were tested for compressive strength (CS) for 7 or 28 days. The three test's average findings for strength are shown as a percentage of strength in relation to the control mortar. The SAI reported therein was as per (equation (1))

$$\begin{eqnarray}&&SAI=\displaystyle \frac{A}{B}\times 100\end{eqnarray} \tag{ 1 }$$

Where, A being the unconfined CS of the test pozzolan specimen and B is the unconfined CS of the control specimen (MPa).

Table 2. Mix morphology for the pozzolanicity tests involved.

Test method	Test pozzolan	OPC (g)	Ca(OH)2 (ml)	Fine aggregate (g)	Pozzolan (g)	Water (ml)
Frattini	Control	20	N.A	N.A	0	100
	M-Sand	16	N.A	N.A	4	100
	FA	16	N.A	N.A	4	100
	BOF	16	N.A	N.A	4	100
	IOT	16	N.A	N.A	4	100
SLT	Control	N.A	75	N.A	0	N.A
	M-Sand	N.A	75	N.A	1	N.A
	FA	N.A	75	N.A	1	N.A
	BOF	N.A	75	N.A	1	N.A
	IOT	N.A	75	N.A	0	N.A
SAI	Control	450	N.A	1350	0	230
	M-Sand	360	N.A	1350	90	230
	FA	360	N.A	1350	90	239
	BOF	360	N.A	1350	90	280
	IOT	360	N.A	1350	90	300

N.A—Not Applicable

It is noteworthy that an indication of positive pozzolanicity is verified upon attaining the 28-day SAI values (> 0.80) for a pozzolan replaced with 30% OPC [56]. Contrarily, ASTM C618 [58] highlights the SAI value (>0.75) at both 7 and 28 days.

3.1. Frattini test

Figure 2 shows pozzolanic activity in the test pozzolan (80% OPC with 20% pozzolan), FA, BOF slag, and IOT samples exhibit pozzolanic activity, whereas M-Sand was inactive. These results need to be quantified to correlate with the results of SLT. In order to measure the same, distances of data points from both the lime solubility curve and the zero point on the vertical axis (at the provided OH⁻) are to be considered. However, the standard code of practice [36] provides a suitable bound for the obtained results (figure 2) by providing the data pertaining to Ca(OH)₂ at 40 °C and accordingly when [OH⁻] ranges between 35 and 90 mmol l⁻¹. The theoretical maximum [CaO] concentration between 35 and 90 mmol l⁻¹ [OH⁻] can be calculated using the standard formula (equation (2)), thereby mapping the lime solubility curve.

$$\begin{eqnarray}&&{Theoretical}{\rm{Max}}[{CaO}]=\frac{350}{\left[{OH}\right]-15}\end{eqnarray} \tag{ 2 }$$

Zoom In Zoom Out Reset image size

The sample calcium concentration (CaO) is compared to the theoretical maximum (CaO), and their difference is expressed as a percentage of the theoretical maximum being removed (table 3). As mentioned, OPC samples were not pozzolanic. The % CaO removal for OPC was unexpectedly negative, denoting a much higher percentage of calcium in the solution. This can be attributed to suspended Ca(OH)₂ or C–S–H gel being passed through the filter or any erroneous volumetric analysis. Theoretically, the presence of Ca(OH)₂ in any solid crystalline form will be retained on the filter. However, this theory has been disagreed, owing to the results above the solubility curve. Negative results should be standardized to an equivalent of 0% CaO removal. From the Frattini test, BOF had 90% lime removal, FA had 47%, and IOT had 30%, indicating high, considerable, and reduced pozzolanic activity, respectively. These results are found to be in correlation with previous findings [59, 60], which have reported similar degrees of pozzolanicity for these materials.

Table 3. Frattini test results for the test pozzolans.

Test Pozzolan	OH mmol−1	CaO mmol−1	Max. (CaO) mmol−1	(CaO) removal %
Control sample	56.5	8.3	8.4	1.59
M-sand	52.6	10	9.3	−7.4
FA	53.5	4.8	9.1	47.2
BOF	38	1.5	15.2	90.1
IOT	35.4	12	17.2	30.1

3.2. Saturated lime test (SLT)

Figure 3 depicts the SLT test results. The SLT is simple to quantify, as a fixed amount of lime being added initially, followed by measuring residual lime at the end. The difference arises due to fixed lime by the test pozzolans and subsequently being retained on the filter paper. The solubility of lime exceeded by 0.3 g/3.0 mmol l⁻¹ as opposed to their initial value on added lime (2 g or 20.45 mmol l⁻¹). The solubility of lime is considered significant in the SLT. This can be influenced by various factors viz., temperature, pressure, and the presence of other ions which leads to excess precipitation of lime in the control specimens and thus increasing the lime solubility [61]. However, this is in contrast to the theoretical concept on the OPC (control) sample, comprising only lime solution and sand particles, which is supposed to have exhibited zero results for a fixed lime. The excess precipitated lime in control specimens was extracted by the filter paper, despite results stating that they were removed as a result of reaction with the sand. This makes the control sample to be considered as null activity (baseline), which is a limitation to SLT method. Also, the control baseline value need not be deducted for samples with positive lime fixation. Lime fixation indicates the amount of lime being consumed during the pozzolanic reaction. The rate of lime fixation can be depended by various factors viz., specific surface area and chemical composition of the pozzolanic material. This is because the lime reservoir of starting solution being depleted when the test pozzolan fixes the first 3 mmol l ⁻¹ lime. As per the control baseline, the test results indicate that M-sand was not pozzolanic with lime. It is to be noted that BOF slag and IOT have rapid lime fixation, owing to their possible higher specific surface area and varied chemical compositions to that of FA, which had no removal above the baseline [62]. Upon completion of 7 days, FA, BOF, and IOT all fixed around 90% of available lime, with no significant differences in their lime fixation.

Zoom In Zoom Out Reset image size

3.3. Strength activity index (SAI)

The average compressive strength measured for the test mixes is shown in figure 4 The strength plot depicts that all the test pozzolans experience reduced compressive strength after 7-day period relative to the conventional mix, which reported 40 and 49.5 MPa at 7 and 28 days, respectively. Except for BOF samples, the test results persistently showed a substantial difference in strength between the control and test specimens after 28 days. The diluting effect should result in a 20% reduction in strength development if test pozzolans were entirely inert (M-sand). However, factors like permeability, porosity, and the kinetics of the hydration reaction have an impact on compressive strength development [63]. The study assumes that sand samples are an inert control and do not exhibit any pozzolanic activity. Considering the same, the test results show that FA and IOT possess low pozzolanicity, while BOF slag has considerable pozzolanicity. This demonstrates the use of BOF slag as a primary precursor in the development of EGC to enhance the mechanical characteristics of the proposed EGC [59].

Zoom In Zoom Out Reset image size

Assessing pozzolanicity requires consideration of the test methodology as a mandatory aspect. Multiple methods are often used, with a minimum of one exhibiting Ca(OH)₂ consumption over time. Comparison of methods requires considering the sample's curing temperature and time. The Frattini test was performed for 8 days (40 °C), and the SAI test was up to 28 days (23 °C). Whereas the SLT can be done anytime, as hydration reactions are unnecessary. Possibly, the key factor in comparing the direct and indirect tests is the ratio between the lime and the pozzolan mass ratio (lime: pozzolan). Assessing the amount of Ca(OH)₂ developed during OPC hydration is usually challenging. However, by rule, 25% of the initial cement mass is generally available as Ca(OH)₂ upon full hydration. It is to be noted the Frattini and SAI tests thereby had a 1:1 mass ratio of pozzolan to cement (2 g pozzolan: 8 g OPC). While the SLT had a considerably lower ratio (0.15:1). This reduced ratio makes the SLT more likely to demonstrate pozzolanic activity than the Frattini test. Findings from the Frattini test imply that the hydration mechanism of the proposed pozzolans varies, unlike OPC hydration, indicating the difference in the saturated lime solution and the pore fluid of OPC [64–66]. It is to be noted that the hydration reaction of pozzolans, by the Frattini test, involves alkali metals in the form of sodium (Na⁺) and potassium (K⁺) ions alongside Ca(OH)₂ [67–69]. Contrarily, these ions restrain Ca(OH)₂ solubility in OPC fluids based on the alkali activity [70, 71]. The Frattini test can thereby be associated with a similar methodology as in the case of blended cement (OPC + Pozzolan), where, as the hydration proceeds, initial Ca(OH)₂ present dissolves and aids in the forming of stable compounds [67–69].

4.1. Frattini test and SLT

While the Frattini test measures the ability of a material to reduce the pH of a saturated calcium hydroxide solution, the SLT quantifies the amount of lime fixed by the material [50]. After 7 days, the SLT had reached equilibrium, with all materials displaying similar pozzolanic activity except M-sand, which remained inert, thereby avoiding the probability of Ca(OH)₂ removal from solution being an adsorption phenomenon. Except for M-sand, all materials had high pozzolanic activity in the SLT after 7 days. As with the Frattini test, all the test pozzolans (FA, BOF, and IOT) except M-sand exhibited pozzolanic activity and had notable differences among the same. Thus, the SLT and Frattini tests were unrelated (figure 5).

Zoom In Zoom Out Reset image size

4.2. SAI and frattini

To correlate these two tests, 20% OPC replacement was chosen, the common factor in both tests. The Frattini test provides a comprehension of the possible Ca(OH)₂ concentration corresponding to the SAI specimens. BOF had the highest pozzolanicity irrespective of these tests. This can be associated with the reaction of BOF slag with the Si and Al of the EGC matrix to form calcium silicate hydrates, owing to their high calcium content, thereby imparting enhanced strength properties [59]. Also, the excess removal of Ca(OH)₂ from the Frattini test upon attaining 8 days suggests that, upon attaining 28 days, BOF and the Ca(OH)₂ pore solution react, forming C–S–H gel phases, which are considered to impart strength from the SAI test. Consequently, a considerable correlation exists between the Frattini and SAI tests (figure 6). Nonetheless, the Frattini test values recorded for BOF Slag, FA, and IOT in this study were consistent with the known degrees of their respective pozzolanicity [23, 27, 53].

Zoom In Zoom Out Reset image size

4.3. SAI and SLT

The SLT showed similar pozzolanic activity for BOF slag, IOT, and FA with that of SAI. Nevertheless, the SAI test showed reduced values of pozzolanic activity for M-Sand, average acceptable values for FA and IOT, and the highest for BOF slag, indicating a negligible correlation between these two tests (figure 7). The dissimilarity among the tests can be attributed to difference in their chemical reactivity mechanism. The pozzolanic reaction, which is the main reaction involved in these tests, occurs between Ca(OH)₂, and silicic acid (H₄SiO₄). Silica-rich precursors with no cementitious properties are converted into calcium silicates with excellent cementitious characteristics by this reaction. As no irreversible molecular bonds are formed during the process of physical surface adsorption, it is not considered part of the pozzolanic activity. The unique characteristics of BOF slag, IOT, FA, and M-Sand may have affected the rate and extent of the pozzolanic reaction, contributing to the difference in results of SAI and SLT. Nevertheless, the pozzolanic reaction rate is determined by its intrinsic characteristics, including its specific surface area, chemical composition, water content, hydration stoichiometry and active phase concentration [72]. (i) Specific surface area plays a crucial role in the rate of pozzolanicity and have a direct relationship with the pozzolans. For instance, it is to be noted that, the higher pozzolanic rate of BOF slag in the SAI could be associated to their higher specific surface area in comparison with other materials, facilitated by the additional reactive sites for the enhanced pozzolanic reaction [72, 73]. (ii) The chemical composition: upon onset of a pozzolanic reaction, the presence of Si and Al, being considered crucial for the formation of calcium silicate hydrates, can significantly influence the pozzolanicity [74]. Nonetheless, the varied chemical composition of BOF slag, IOT, FA, and M-Sand contributes to the potential differences in their observed pozzolanic rate from both SAI and SLT [72, 75, 76]. (iii) Water content: the consistency and strength of the resulting composite relies on this parameter, with inverse relationship. Excess water causes porosity and thereby reduces the strength [77]. Unlike SAI and Frattini tests, SLT involves zero cement in its mechanism, thereby avoiding forming any solid phases. Additionally, SAI results are affected by the water requirement of various mixtures (table 2), showing a maximum requirement for BOF, followed by IOT and FA to attain a standard consistency mix. Moreover, the control OPC specimens (SAI) involved a water usage of ∼230 ml, inconsistent with the hydration stoichiometry, which suggests a water-to-cement ratio (w/c) requirement of 0.23 (∼103.5 ml) for OPC hydration [78]. This deviation in water content possibly impacts the hydration process and in turn the rate of pozzolanic reaction [77]. (iv) Hydration stoichiometry: this particular parameter in geopolymer based composites is considered complex involving the reaction of alumino-silicate sources with the alkaline activators to form a stable geopolymer gel structure, denoting the properties of the composite [79]. Any discrepancies in this hydration stoichiometry can possibly affect the pozzolanicity and thereby resulting in reduced strength characteristics [78].

Zoom In Zoom Out Reset image size

From the comparison analysis between the available direct and indirect methods (figures 5–7), it is to be noted that a considerable correlation is achieved between the Frattini and SAI test results (R² = 0.84). Moreover, there wasn't any relation between SLT and the Frattini test or between SLT and the SAI, upon considering results of SLT at various stages at 1,3,7 and 28 days.

The pozzolanic activity of the precursors plays a crucial role in the development of EGC and their potential applications towards sustainable construction. Overall, the study contributes to the growing body of knowledge on the utilization of FA, BOF slag, and IOT by providing valuable insights into their pozzolanic activity involving various methods and the corresponding correlation between them.

• All the test pozzolans considered exhibited pozzolanic activity. BOF slag and IOT were identified as the most reactive pozzolans from the 1-day SLT results. Also, SLT results showed that BOF, IOT, and FA possessed high pozzolanic properties after 7 days.
• Quantification of the Frattini test results can be done within the range 35–90 mmol l⁻¹ [OH⁻].
• SLT and Frattini tests did not correlate. In the SLT, the 7-day period was unduly long for the differentiation of pozzolanic activity between BOF, FA, and IOT. Due to the low activator-to-pozzolan ratio and uncertainty in the amount of Ca(OH)₂, the SLT has significant limitations and thereby Frattini methods is recommended.
• The Frattini test and SAI results established a significant correlation (R² = 0.84).
• No correlation persisted between the SLT and the Frattini or SAI tests which was influenced by various factors viz., specific surface area, chemical composition, water content, and hydration stoichiometry of the test pozzolans.
• Conforming the pozzolanic potential of the test materials enhances the geopolymerization process alongside optimizing the strength characteristics of the proposed EGC.
• In summary, FA, BOF slag, and IOT can be beneficial and highly compatible as sustainable precursors. Specifically, BOF slag can potentially enhance the mechanical properties of the proposed EGC owing to its high pozzolanicity.

The Frattini and SAI tests correlate with each other. Contrarily, the SLT has two main drawbacks: the activator to pozzolanic ratio is much lower than other methods, and the amount of Ca(OH)₂ in each pozzolan is considered inaccurate. It is suggested that the Frattini and SAI tests are more accurate and provide sound results on the pozzolanicity of materials when combined with any thermal methodology to determine the Ca(OH)₂.

Not applicable.

No new data were created or analysed in this study.

Saravanan Subramanian performed writing—original draft and conceptualization; Reviewing and Editing. Robin Davis and Blessen Skariah Thomas performed Supervision; conceptualization.

The authors have no competing interests to declare that are relevant to the content of this article.

No funding was received for conducting this study.

The authors declare that they have no conflict of interest.