Enhanced mechanical strength of micro-porous ceramics through the removal of alkaline earth carbonates from Moroccan red clay for membrane support application

3.1.1 Chemical composition

The comprehensive chemical profile of the unprocessed clay is detailed in Table 1. This profile predominantly encompasses silicon dioxide (SiO₂), aluminum oxide (Al₂O₃), calcium oxide (CaO), iron(III) oxide (Fe₂O₃), and potassium oxide (K₂O), while the presence of magnesium oxide (MgO), sodium oxide (Na₂O), titanium dioxide (TiO2), and phosphorus pentoxide (P₂O₅) is comparatively less. The highest concentration of oxides, namely SiO₂ and Al₂O₃, are majorly linked with the composition of quartz and clay minerals, including illite, mullite, and kaolinite, which are ubiquitously found in clay. Furthermore, an elevated SiO₂/Al₂O₃ ratio points towards a considerable presence of quartz and/or amorphous silica (Achiou et al., 2016).

The quantification of calcite content was carried out employing the Bernard calcimeter method, which divulged that raw clay comprises 11 wt% of calcium carbonates (CaCO₃). It can be inferred that a significant proportion of the detected CaO content via X-ray fluorescence (XRF) analysis is embodied as dolomite and calcite. The recorded loss of ignition at a temperature of 1100 °C is attributable to the breakdown of carbonates, burning of organic substances, and removal of hydroxyl groups from clay minerals (Teklay et al., 2014).

3.1.2 FTIR analysis

The Fourier-transform Infrared (FTIR) analysis results for raw clays (CC), as depicted in Fig. 1, reveal distinctive bands at 470 and 535 cm⁻¹. These bands are associated with the stretching vibrations of Si-O-Mg, and Si-O-Al (Miyah et al., 2017). The Si-O stretching vibration exhibits a pronounced and well-defined band at 1036 cm⁻¹. Concurrently, Al₂-OH stretching and bending bonds, intrinsic to phyllosilicates, give rise to a band at 3624 cm⁻¹ (Mudrinić et al., 2015). The FTIR band at 3436 cm⁻¹ is likely attributable to –OH stretching mode. A bending vibration band indicative of water hydration is also visible around 1640 cm⁻¹. An absorption band noted at 1425 cm⁻¹ is characteristic of carbonates, and the bands at 690 and 778 cm⁻¹ serve as markers for quartz (Thuc et al., 2010). Contrastingly, the FTIR spectra for purified Clay (NCC), presented in Fig. 1, demonstrate the notable absence of the characteristic band at 1425 cm⁻¹. This absence strongly suggests that purified clays are devoid of carbonates, a finding which aligns seamlessly with prior X-ray Diffraction (XRD) characterizations and Bernard calcimetry results. This concordance provides a positive indication of the purification's efficacy.

Close

Fig. 1

FTIR spectrum of calcareous clay (CC), non-calcareous clay (NCC) and tea waste (TW).

Export to PPT

On the other hand, the FTIR spectrum of tea waste is illustrated in Fig. 1. An extensive absorption band ranging from 3010 to 3705 cm⁻¹, which is often associated with the O–H and N–H stretching vibrations seen in polyphenols, as highlighted in previous studies (Loo et al., 2012; Senthilkumar and Sivakumar, 2014). Saif et al. (Saif et al., 2016) suggested that the substantial absorption band detected at 3310 cm⁻¹, can be linked to the presence of hydroxyl functional groups (OH), which are common in alcohols and phenolic compounds.

Moreover, a prominent band at 1636 cm⁻¹ is recognized, typically corresponding to the stretching vibrations of C = C bonds in aromatic rings and C = O bonds in polyphenolic compounds (Dubey et al., 2017; Senthilkumar and Sivakumar, 2014). Vibrational bands for the C–H stretch in alkane and the O–H stretch in carboxylic acid are apparent at 2926 cm⁻¹ and 2854 cm⁻¹ respectively (Dubey et al., 2017). A notable band at 1027 cm⁻¹ is assigned to the C-O stretching in amino acids (Dubey et al., 2017; Senthilkumar and Sivakumar, 2014).

J.Cai et al. (Cai et al., 2015) identified similar FTIR bands across multiple tea types, including oolong, green, and black tea. Previous research has mapped the infrared bands of polyphenols in tea at 3387 cm⁻¹, 1635 cm⁻¹, and 1027 cm⁻¹; these are correlated with N–H/O–H, C = C, and C–O–C stretches, respectively (Dubey et al., 2017; Senthilkumar and Sivakumar, 2014; Weng et al., 2013). The FTIR spectrum thus reveals that the dominant functional groups in tea residue include carboxylic acid, polyphenols, and amino acids.

3.1.3 XRD analysis

The X-Ray Diffraction (XRD) patterns for both the calcareous clay (CC) and the non-calcareous clay (NCC) are depicted in Fig. 2. The reflection spectra from the calcareous clay (CC) are attributed to the prevalent phases of quartz, illite, calcite, and dolomite. However, the XRD patterns of the purified clay (NCC) show a different composition. The main components of this purified clay include quartz, kaolinite, illite, and chlorite, indicating that these are the primary clay minerals present. Notably, there is an absence of any detectable peaks for calcite and dolomite, implying that these impurities have been fully eliminated. Thus, according to the XRD findings, it can be inferred that the purified clay exhibits an enrichment of clay minerals and a significant reduction in impurities such as calcite and dolomite.

Close

Fig. 2

DRX of calcareous clay (CC) and non-calcareous clay (NCC).

Export to PPT

3.1.4 Thermal analysis

The organic phases (Tea waste) must be eliminated entirely when sintering the green ceramic support in order to generate the necessary porosity ceramics without causing any flaws. ATD/TG analysis was performed to determine a heating schedule, and the outcome is depicted in Fig. 3. It presents the analysis of a mixture comprising 20 wt% of tea waste and 80 wt% of raw clay (CC). With escalating temperatures, a progressive yet uneven weight reduction is noticed, totaling a weight loss of 35.32 wt%. At temperatures less than 100 °C, there's an initial weight reduction of 5.31 wt%, ascribed to the endothermic evaporation of unbound water. A subsequent 10.41 wt% weight loss in the temperature range of 100–320 °C may be due to the elimination of adsorbed water and the expulsion of volatile organic compounds. In the temperature bracket of 320–560 °C, an additional weight loss of 14.20 wt% is identified, which could be traced to the disintegration of remaining organic compounds with firmly bonded carbonyl and hydroxyl groups. This is also attributed to the dehydroxylation of clay minerals such as illite and kaolinite, and the transformation of α-quartz into β-quartz (Mouiya et al., 2022; Vizcayno et al., 2010).

Close

Fig. 3

ATD/TG of raw red clay (CC) mixed with 20 wt% of tea waste.

Export to PPT

Around 705 °C, an endothermic peak is recorded alongside a 5.40 wt% weight reduction, suggesting the thermal decomposition of mineral carbonates. Post 900 °C, no considerable decrease in weight is discernible. The thermal evaluation advocates that the sintering of ceramic membranes should ideally occur at temperatures exceeding 900 °C.

3.1.5 Dilatometric analysis

Due to thermal agitation, heating a substance increases its internal energy. The expansion coefficient is a proportionality factor that correlates with the increase in average interatomic distances with temperature. Measuring a material's expansion typically only allows us to characterize its linear expansion. The coefficient of linear expansion (α) may be calculated at each temperature using the relationship (3), where L is the length of the specimen.

The thermal expansion behaviour depicted in Fig. 4 exhibits a pronounced curve between heating and cooling. This graph depicts three noteworthy shifts in behavior as the temperature increases to 1100 °C. The initial expansion occurring between 155 and 230 °C can be attributed to the release of loosely bound and accumulated water from the interlayer regions of the specimen. The significant secondary expansion ranging from 330 to 588 °C aligns with the breakdown of the organic components present in TW, and around 573 °C, a slight expansion singularity highlights the α to β transition of quartz, which is the so-called quartz dilatometric anomaly (Mouiya et al., 2022). In addition, at this point clay minerals lose structural water, such as lepidocrocite (FeO(OH), which, following dehydration, produces hematite (Fe₂O₃), (Eq. (4)) (Morris et al., 1998) and kaolinite (Al₂Si₂O₅(OH)₄) is transformed to metakaolin through the so-called kaolinite network collapse, in which the Si–O links stay intact but the Al–O network is reorganized (Eq. (5)) (Kakali et al., 2001). The substantial shrinkage between 855 and 1100 °C was attributed to the presence of Si site rearrangement processes that produced cristobalite (SiO₂) and mullite (3SiO₂Al₂O₃) (Eq (6)) (Faieta-Boada and McColm, 1993). The preceding assignments were validated by XRD (Fig. 5). The coefficient thermal expansion (CTE) of the green compact between 230 and 330 °C was calculated to be 5.75 × 10^-6 °C⁻¹, and 7.14 × 10^-6 °C⁻¹ between 588 and 720 °C⁻¹ assuming linear expansion.

Close

Fig. 4

Thermal expansion behavior of raw red clay containing 20 wt% of tea waste during sintering up to 1100 °C.

Export to PPT

Close

Fig. 5

XRD diagrams of CC and NCC samples fired at (a) 900 °C, (b) 1000 °C, (c) 1100 °C and 1150 °C.

Export to PPT

3.3.1 Weight loss, size and color variation caused by sintering at different temperatures

Table 2 provides a comprehensive analysis of weight losses and dimensional variations (diameter and thicknesses) in pressed ceramic supports derived from CC and NCC samples, which were subjected to increasing firing temperatures ranging from 900 °C to 1150 °C. The results demonstrate significant changes in weight losses for the CC sample as the firing temperature escalates. The experimental data showcases a range starting from 19.3 ± 0.45 wt% at 900 °C / 2 h and reaching a maximum value of 33.73 ± 0.18 wt% at 1150 °C /2h. Comparatively, the weight losses observed in the NCC sample are slightly lower than those of the CC sample. The weight loss values exhibit a similar trend, with a minimum of 17.88 ± 0.27 wt% at 900 °C /2h and a maximum of 31.03 ± 0.22 wt% at 1150 °C / 2 h. As shown in Fig. 3 (ATD-TG), the observed weight variations in the fired ceramic supports can be predominantly ascribed to the removal of organic substances, dehydroxylation of clay minerals such chlorite and illite, and decarbonation of calcite within the CC sample.

Fig. 6 displays a collection of chosen test specimens obtained from the samples that underwent firing from 900 to 1150 °C for a duration of 2 h. The firing process induces noticeable changes in dimensions, accompanied by a color transition from red to a deep shade of dark red. This color alteration can be attributed to the presence of Fe₂O₃ in the samples, as indicated by the chemical analysis presented in Table 1. The color transformation in ceramics is a multifaceted process influenced by various physico-chemical and mineralogical factors (Raigón-Pichardo et al., 1996). In a recent study by Harrati et al. (Harrati et al., 2022), Clayey samples including kaolinite, illite, quartz, calcite and chlorite were studied. The authors suggested that a higher content of Fe₂O₃ (>3 wt%), similar to the conditions in our study, could result in a reddish coloration after firing. This observation aligns with the visual examination of the ceramic supports derived from CC and NCC samples, as depicted in Fig. 6.

Close

Fig. 6

Image depicting the ceramic supports manufactured from the samples CC and NCC at various temperatures.

Export to PPT

3.3.2 Bulk density, open porosity and water absorption

Open porosity and bulk density analyses, is showed in Fig. 7, were conducted on two supports ceramic samples: one incorporated carbonate (CC) while the other (NCC) was devoid of this component. A series of tests were conducted at varying temperatures, unveiling intriguing correlations and dependencies on the presence of carbonates.

Close

Fig. 7

Open porosity (A) and bulk density (B) results for CC and NCC fired samples.

Export to PPT

At 900 °C, the density of the carbonate-inclusive sample (CC) was 1.39 g/cm³, contrasted by a slightly higher 1.47 g/cm³ in the carbonate-free sample (NCC). Concurrently, the open porosity in the carbonate-bearing ceramic was 48.51 %, compared to 41.82 % in the sample devoid of carbonates. This disparity points to the potential role of carbonates in fostering greater porosity, possibly by contributing to the formation of pores during the thermal treatment.

At 1000 °C and 1100 °C, the presence of carbonates corresponded with a lower density and a higher porosity compared to the carbonate-free counterpart. At 1000 °C, the carbonate-inclusive sample showed a density of 1.48 g/cm³ and a porosity of 43.3 %. In contrast, the sample without carbonates displayed a density of 1.53 g/cm³ and a porosity of 38.44 %. Similarly, at 1100 °C, the densities were 1.55 g/cm³ (carbonate-inclusive-CC) and 1.59 g/cm³ (carbonate-free-NCC), with porosities of 39.35 % and 33.15 % respectively. These trends suggest that carbonate presence may be influencing the ceramic microstructure, thereby affecting the porosity and density.

However, at 1150 °C, a dramatic divergence was observed. Both samples experienced a significant increase in density (carbonate-inclusive: 2.01 g/cm3, carbonate-free: 2.09 g/cm3) and a corresponding decrease in open porosity (carbonate-inclusive-CC: 18.87 %, carbonate-free-NCC: 17.11 %). It suggests a substantial structural transition potentially due to a phase transformation, sintering progressively at higher temperatures results in the production of a denser ceramic body (see SEM, in Fig. 9-D). This densification is achieved through the creation of a glassy phase that effectively fills the material's pores. According to Semiz's study (Semiz, 2017), it is probable that the rise in bulk density during firing is linked to the development of a prominent glassy phase. This leads to an increase in the formation of a liquid phase and consequently a reduction in the material's porosity. The carbonate-inclusive sample still maintained lower density and higher porosity than the carbonate-free one.

These findings underscore the significant impact of carbonates on the density and open porosity of supports ceramic. The presence of carbonates seems to enhance the porosity and reduce the bulk density at lower temperatures, while even at high temperatures where densification occurs, carbonates appear to limit this densification to an extent. The exact mechanisms by which carbonates influence these properties would be a compelling subject for further exploration.

It should be noted that the influence of carbonate content on the pore size in ceramics, and consequently on their potential as filtration media, presents a complex interplay of benefits and potential drawbacks that need to be evaluated critically. The presence of carbonates has been observed to increase open porosity, which in turn increases pore size. While at first glance this might seem advantageous for filtration applications, as larger pores would allow for higher flow rates, this effect should not be considered in isolation.

The size and distribution of pores in a filter medium directly affect its performance. Larger pores might indeed facilitate greater flow rates, reducing the pressure needed for filtration and increasing the filter's overall capacity. However, they could also compromise the filter's ability to capture and retain small particles, thus reducing the filter's efficiency in terms of particle removal. If the aim is to filter out microscopic contaminants, an increase in pore diameter induced by carbonates might not be beneficial, but rather detrimental to the performance of the filter.

Furthermore, increased porosity and larger pores could lead to decreased mechanical strength, as larger pores can act as stress concentrators, leading to premature failure under mechanical or hydraulic pressure. This is a crucial factor for ceramic filters, which often operate under significant pressure differentials.

in addition, the water absorption capacity is a useful metric for analyzing the densification process and the quality of ceramic supports, which is determined by the volume of open pores that can absorb water in the fired specimens. The water absorption capacities for CC and NCC samples in Fig. 8 show that the former's capacity declines from 33.2 ± 0.51 % at 900 °C for 2 h to 23.6 ± 0.48 % at 1100 °C for the same duration. It further drops sharply to 12.9 ± 0.40 % at 1150 °C for 2 h. The NCC samples demonstrate a similar trend. A variation of approximately 20 % in the water absorption capacity is recorded for both sample types when heated from 900 to 1150 °C/2h. In a study conducted by Semiz et al. (Semiz, 2017), the water absorption capacity of illite–smectite-kaolinitic clays varied from 20 % at 700 °C to 2 % at 1200 °C. They posited that this shift is likely associated with the emergence of a glassy phase.

Close

Fig. 8

Water absorption results for CC and NCC fired samples.

Export to PPT

Close

Fig. 9

SEM micrographs of sample NCC fired at 900 °C (A), 1000 °C (B), 1100 °C (C) and 1150 °C (D).

Export to PPT

Furthermore, the results illustrated in Fig. 8 demonstrate that the presence of carbonates in CC samples leads to an increase in the water absorption of ceramic supports, independently of firing temperature. This enhancement is to be expected, as the carbonates present in the raw clay (CC) decompose into CO₂ as the firing process proceeds, the open porosity of the ceramic support samples increases (Arslan et al., 2021; Ngayakamo et al., 2020; Njeumen Nkayem et al., 2016).

3.3.3 SEM microstructural analysis of fired samples

The SEM analysis focused on investigating the changes in microstructure of NCC samples after firing. Fractured surfaces were examined to observe these microstructural variations. In this section, we provide specific SEM micrographs of NCC samples that have been subjected to firing within the temperature range of 900 to 1150 °C. The SEM evolution of the NCC samples is illustrated in Fig. 9. Initially, up to a temperature of 900 °C/ 2 h, the microstructure of the material remains unaltered. This can be attributed to the fact that the pressed and fired samples still retain the layered structure of clay minerals (particularly illite and chlorite), which remains discernible. However, the illite undergoes dehydroxylation, while the chlorite decomposes. This observation aligns with previous studies and SEM investigations on the thermal treatment of clay minerals (Sanchez-Soto and Perez-Rodriguez, 1989). Moreover, the micrographs reveal the presence of a porous structure, with an open porosity of 41.82 % (as depicted in Fig. 7). This porosity is a consequence of the dehydroxylation process occurring in the clay minerals. Additionally, the presence of compact crystalline formations of varying sizes in the fired samples can be linked to quartz grains that are integrated within the matrix.

When subjected to a temperature of 1000 °C for a duration of 2 h, the early phases of the vitrification process become. This process manifests in the formation of minute pores, along with the potential merging of the layered structures, such as dehydroxylated illite and decomposed chlorite. Despite this coalescence, layered structures remain identifiable on the fracture surface. The sample's open porosity is approximately 38.44 %, lower than the porosity observed in the sample fired at 900 °C/2h.

At a temperature of 1100 °C for a duration of 2 h, a significant change occurs in the SEM of the fracture surface of the fired samples. This change is characterized by a reduction in the size of pores, indicating a diminution in porosity. The open porosity of the samples shows a slight decrease (33.15 %, Fig. 7) compared to samples fired at lower temperatures of 900 °C/ 2 h and 1000 °C/2h (as shown in Fig. 7). Moreover, the matrix contains small crystalline particles, specifically associated with quartz.

At a temperature of 1150 °C maintained for 2 h, the degree of vitrification in the fired sample is higher and more widespread, resulting in a distinct microstructure compared to earlier observations (as seen in Fig. 9, the SEM of samples fired from 900 to1100 °C for 2 h). Indeed, the existence of alkaline and alkaline-earth oxides plays a significant role in developing the vitreous phase, which becomes liquid at high temperatures and is discernible in the SEM of the samples generated through firing. SEM reveals a more intricate microstructure. Additionally, there's a decrease in open porosity to around 17.11 % (as depicted in Fig. 7) and an increase in the bulk density (∼2.09 g/cm3, as shown in Fig. 7).

3.3.4 Mechanical strength

3.3.4.1

3.3.4.1 Flexural and indirect tensile strength

The outcomes pertaining to the flexural and tensile strength of both fired samples, namely CC and NCC, are illustrated in Figs. 10 and 11 correspondingly. In the case of the CC sample, the highest flexural strength value recorded is 42.47 ± 3.24 MPa, achieved through firing at 1150 °C for 2 h. Similarly, for the NCC sample, a similar trend is observed, with the flexural strength reaching 44.84 ± 2.86 MPa under the same firing conditions of 1150 °C for 2 h. Broadly speaking, the results indicate that subjecting the samples to firing temperatures higher than 900 °C for 2 h leads to an elevation in their flexural strength. This enhancement is attributed to the gradual development of a glassy phase, effectively filling the minute pores within the samples. This finding aligns with the SEM (Fig. 9-D), which exhibits a substantial, compact surface area characterized by the complete absence of tiny pores. The presence of crystalline solid phases (diopside and anorthite) contributed to the enhancement of flexural strength. In the case of samples sintered at 900 and 1000 °C (depicted in Fig. 10-B), a linear progression with heightened elasticity prior to failure was observed, attributed to their exceptional elasticity. However, when examining the tensile load–displacement curves of the supports sintered at 1150 °C, two distinct behaviors emerged. Initially, the material exhibited limited plasticity, succeeded by a substantial increase leading to sudden failure. This crucial observation underscores that the elasticity of the red clay material diminishes beyond 1150 °C during high-temperature sintering.

Close

Fig. 10

Indirect tensile and Flexural strengths of fired sample CC.

Export to PPT

Close

Fig. 11

Indirect tensile and Flexural strengths of fired sample NCC.

Export to PPT

On the other hand, the results reveal that the flexural strength values increase significantly in the decarbonized samples (NCC) at all sintering temperatures. This increase can be explained by a slight reduction in open pores, as shown in Fig. 7, due to the absence of carbonates, which leads to a considerable increase in flexural strength.

Fig. 10 illustrates the plot of the indirect tensile strength values for the sample CC, the indirect tensile strength is 5.03 ± 1.1 MPa at a firing temperature of 900 °C for 2 h, which represents the minimum value. As the firing temperature increases, the indirect tensile strength values also increase, reaching a maximum of 17.66 ± 2.75 MPa at 1150 °C for 2 h. At this maximum point, the water absorption capacity is 12.9 ± 0.64 % (Fig. 8) and the open porosity is 18.87 ± 0.56 % (Fig. 7). The most significant change in indirect tensile strength occurs starting from 1100 °C for 2 h and continues at higher firing temperatures.

For the sample NCC in Fig. 11, the values of indirect tensile strength increase almost linearly, starting from 6.65 MPa at a firing temperature of 900 °C for 2 h and reaching similar values at 900–1100 °C for 2 h. Beyond 1100 °C for 2 h, the indirect tensile strength of the sample exhibits a swift growth, peaking at 19.52 MPa at a temperature of 1150 °C, over a duration of 2 h. At this pinnacle, the water absorption capacity stands at 12.1 ± 0.36 % (as shown in Fig. 8), while the open porosity reaches 17.11 ± 0.51 % (as depicted in Fig. 7). Nevertheless, the maximum value recorded for the samples NCC on the same firing temperature (19.52 ± 0.81 MPa) better than that achieved by the samples CC (17.66 ± 0.53 MPa). Hence, the decline in open porosity (as illustrated in Fig. 7) and the reduction in water absorption capacity (as depicted in Fig. 8) starting from 1100 °C for a duration of 2 h actively supports the augmentation of indirect tensile strength. Pérez-Villarejo et al. (Pérez-Villarejo et al., 2015), in their examination of typical clays and the integration of waste materials into clay bricks, proposed that the elevation in indirect tensile strength within their fired specimens might be ascribed to the decrease in porosity.

3.3.4.2

3.3.4.2 Relationship between indirect tensile strength and porosity of supports ceramic

Initially, the focus was on exploring the potential associations between mechanical strength, specifically indirect tensile strength, and porosity. This correlation was assessed in the heat-treated specimens of this red clay by employing the Ryshkevitch-Duckworth equation (RYSHKEWITCH, 1953) as indicated by equ (7).

(7)

$$\sigma ={\sigma }_0.e^{-bp}$$

In this equation, σ denotes the indirect tensile strength and σ₀ signifies the indirect tensile strength when porosity is zero. The constant b stands for the preexponential constant, indicative of the bonding capacity, whereas P is a measure of porosity. This formula has been successfully utilized across an extensive array of porous materials, including but not limited to ceramics. Therefore, porosity is inversely proportional to the logarithm of indirect tensile strength, allowing an extrapolation of the indirect tensile strength when porosity is zero. In this specific study, the values σ and P were experimentally measured and reported as averages from three separate measures (refer to Fig. 7,10 and 11). If the parameters σ₀ and b are given, then one can predict the indirect tensile strength for a powdered, fired, compacted specimen. Fig. 12 illustrates the Ryshkevitch-Duckworth graphical interpretation for the CC and NCC samples examined in this study. The observed relationship between open porosity and indirect tensile strength under these experimental conditions is apparent. The resulting straight lines suggest that indirect tensile strength correlates with open porosity. Nevertheless, this correlation is stronger for the NCC sample, as indicated by the higher correlation coefficient (R² = 0.9302), when compared to the CC sample (R² = 0.9117). The constant b and indirect tensile strength at zero porosity (σ₀) can be extrapolated from these findings.

Close

Fig. 12

Relationship between indirect tensile strength (results presenteded in Figs. 10 and 11) and porosity (Fig. 7) of the samples CC and NCC based on the Ryshkewitch-Duckworth equation.

Export to PPT

To start with, the observed discrepancies from the linear trend in accordance with the Ryshkevitch-Duckworth equation (7) for both CC and NCC samples could be attributed to factors such as irregular shapes, defects, and imperfections related to open pores, as suggested in a prior study (Martínez-Martínez et al., 2022). An essential factor to consider, is the potential existence of closed pores in the fired samples resulting from progressive sintering at elevated temperatures commencing at 1100 °C. Such a presence could negatively impact the indirect tensile strength. Indeed, reviewing the indirect tensile strength data featured in Fig. 12 reveals two distinct patterns: (i) the first pattern falls within the 900–1100 °C temperature range (refer to Figs. 10 and 11), where the experimental points cluster, demonstrating a nearly linear relationship between indirect tensile strength and porosity; (ii) the second pattern becomes evident from 1100 to 1150 °C/2h, where a rapid divergence in the values of indirect tensile strength and open porosity can be observed, signaled by a change in the slope of the theoretical straight line (highlighted in Fig. 12). This shift takes place when the fired samples have undergone more extensive sintering (see Figs. 7 and 8). This conclusion is corroborated by the SEM image.