3.1. X-Ray Fluorescence (XRF)
In
Table 2, the results of the chemical composition, expressed as weight percentages, of the analyzed samples are presented. Previous studies have highlighted that the pozzolanic reaction is activated when the ash contains an adequate concentration of silicon dioxide (SiO
2), aluminum oxide (Al
2O
3), and iron oxide (Fe
2O
3) [
37,
38]. As shown in
Table 2, this increase in the concentration of SiO
2 as the calcination temperature rises implies a higher potential for the ashes to be used as a partial substitute for cement in the production of high-strength and durable concrete, positioning POFA as a viable option in terms of sustainability and environmental impact reduction.
The chemical composition results of each evaluated sample were compared with the requirements of the ASTM C 618 standard for pozzolans. The analyzed samples meet the chemical composition requirements for Class N pozzolans according to the standard [
39]. This indicates that they are either raw or calcined natural pozzolans and materials that require calcination to induce satisfactory properties. According to the [
39], the chemical composition requirement for basic oxides—silicon dioxide (SiO
2), aluminum oxide (Al
2O
3), and iron oxide (Fe
2O
3)—is a minimum of 70% for Class N pozzolans to be used in concrete. Upon analyzing the results obtained, it can be observed that, for all the tested samples, including sample 0 (MO-T100C-C-t24 h-c-m), this requirement is met. Additionally, the maximum allowed percentage of sulfur trioxide (SO
3) for Class N pozzolans is 4%, according to the standard [
39]. And all the analyzed samples are within this limit, confirming their compliance with the quality standards. It should be noted that the limits on the content of these chemical compounds arise because the specifications allow the use of ashes produced from coal combustion with co-combustion materials, as is the case with oil palm kernel shell ash. Consequently, elements such as free P
2O
5 and CaO in this type of supplementary cementitious materials are known to cause significant setting delay and expansion, respectively [
40]. However, the content of these elements in all the studied samples is below 5%, suggesting that their presence will not be detrimental when incorporating these ashes into cementitious matrices.
Figure 6 illustrates the variation in the percentage of silicon dioxide (SiO
2) in the evaluated samples as a function of the calcination temperature. It is observed that the uncalcinated sample (M0, dried at 100 °C) exhibits the highest silica content, exceeding 78%, indicating that the oil cake has a high potential as a pozzolanic material, given that the presence of silica in pozzolanas is crucial for its reaction with Portland cement and the subsequent formation of CSH gel. For the sample M17A-c-m-T500C-t1.5h-tr1h subjected to 500 °C, a decrease in the silica content is observed, reaching 69.51%, suggesting the combustion of silica-rich particles. However, the sample calcined at 600 °C (M18A-c-m-T600C-t1.5h-tr1h) shows an increase in silica content, peaking at 71.24%. Therefore, this temperature is considered optimal, since, at 700 °C, a decrease in the silica content occurs, and temperatures above 800 °C cause a phase change from amorphous to crystalline, which is undesirable for pozzolanas added to concrete [
30,
41]. Thus, it is observed that combustion at 700 °C promotes the burning of silica-rich particles, which reduces the SiO2 content at this temperature compared to 600 °C and 800 °C.
Previous studies have shown that, in untreated samples, the silica content may appear higher due to the presence of organic and carbonaceous material that has not been removed [
38,
42]; this can retain fine silica particles in their natural state. However, this silica content may not be fully available for the pozzolanic reaction due to the interference of these organic compounds [
43]. Therefore, although the SiO
2 percentage may appear high, its reactivity is limited until the material undergoes a calcination process [
44].
Based on the results indicated in
Table 2 and the graph obtained from the comparison of these results with respect to the temperature (
Figure 6), it becomes evident that, for the calcined samples, as the temperature increased, the silica content also increased. However, the temperature of 600 °C is considered optimal, as it is higher than both 500 °C and 700 °C, and it is noted that, according to the literature, temperatures above 800 °C promote the formation of crystals, causing silica to transition from amorphous to crystalline [
40], which can alter the material’s composition [
45]. This transformation is critical, as crystalline silica is less reactive as a pozzolan, reducing its effectiveness in cementitious applications [
46]. Additionally, it has been experimentally demonstrated that a temperature of 600 °C allows for a higher silica content and exhibits lower carbon contents [
30]. This temperature was selected due to its potential for use as pozzolans and because it yields a greater amount of pozzolana than temperatures higher than 600 °C.
In general, palm oil kernel ash (POFA) has a silica (SiO
2) content comparable to or even higher than other agricultural ashes analyzed, especially at higher calcination temperatures. Although POFA presents a favorable chemical profile, other types of ashes may offer similar or superior properties depending on their specific composition and the thermal treatment applied. Furthermore, the improvement in the compressive strength of concrete incorporating these ashes, compared to control mixes, highlights their potential as a partial cement substitute, offering a sustainable alternative in construction [
47].
The result of the chemical analysis of palm oil kernel ash (POFA) was compared with various studies that have investigated the potential use of ashes from different agricultural products in concrete, such as rice husk ash (RHA) [
48], olive seed ash (OSA) [
49], ceniza de biomasa de astillas de madera (BA) [
50], biomass ash from wood chips (BA) [
9,
51], and bagasse ash from sugarcane (SCBA) [
52,
53]. The results of this comparison indicated POFA as a good pozzolanic material due to its high silica content, as well as WSA and SBA. The results of the RHA showed a silica content higher than that of POFA [
48,
54].
Below is a comparison between POFA-FB, which refers to the fruit bunch ash of the oil palm, POFA-K, understood as the ash from the oil palm kernel shell, and G-POFA, corresponding to the ground ash of the oil palm kernel. Additionally, a comparison is made with studies addressing U-POFA, which refers to the ultrafine ash of the oil palm kernel, T-POFA, characterized as the treated ash of the oil palm kernel, MT-POFA, referring to the treated and modified oil palm kernel ash, and POCP, corresponding to the powder of the oil palm kernel.
The results of the study [
55]: The results of the study for POFA-FB and POFA-K show a SiO
2 content of 64.72% and 64.07%, respectively, at calcination temperatures of 60 °C and 150 °C, as detailed in
Table 3. These results do not meet the minimum requirement of 70% in the sum of SiO
2, Al
2O
3, and Fe
2O
3 contents to be considered effective pozzolans according to ASTM C618. In contrast, the palm kernel ash in this study meets the regulatory requirement when calcined at higher temperatures, indicating that a higher calcination temperature improves the silica availability and, therefore, the pozzolanic quality of the ashes, as long as the maximum temperature limit is not exceeded.
The palm kernel ash (POFA) samples at temperatures between 500 °C and 1000 °C show a SiO2 content ranging from 68.47% to 76.80%, thus meeting the minimum requirement of 70% for the sum of SiO2, Al2O3, and Fe2O3. In terms of comparison with previous studies, a clear difference in SiO2 content values is observed. The results reported in the literature for samples calcined at 600 °C reflect a significant difference in the methodology applied and its implications on the chemical composition of the ashes. Thus, the existing literature reports significantly lower silica contents, with most values below 65.01%, whereas, in this study, the samples characterized experimentally using XRF exhibit much higher silica content, showing a notable disparity with these results. For the ultrafine sample (U-POFA) and treated modified samples (MT-POFA) calcined at 600 °C, these silica values were higher, approaching the values from the present study. The contrast can be explained by the additional processes applied in this research, such as cutting, crushing, and grinding, which improve the fineness of the material and, therefore, increase the reactive surface area, optimizing the release of silica.
In this research, the additional processes of cutting, crushing, and grinding played a key role in improving the pozzolanic properties. Crushing and grinding help reduce the particle size of POFA, increasing the specific surface area available to react during cement hydration. Additionally, it promotes a greater release of silica, as the internal layers of the material are exposed to the calcination and crushing processes. This study demonstrates that the application of cutting, crushing, and grinding processes, combined with calcination at 600 °C, maximizes the availability of silica, improving POFA’s properties as a pozzolanic material and suggesting that these methodologies should be followed to achieve optimal results in cementitious applications.
In
Table 3, it is observed that the U-POFA samples U-POFA [
56,
57], the T-POFA sample [
37,
58], and the MT-POFA sample [
61] meet the minimum requirement of 70% according to ASTM C618 for the sum of the oxides SiO
2, Al
2O
3, and Fe
2O
3, with the MT-POFA sample, calcined at a higher temperature, exhibiting the highest silica content. This indicates that both the calcination temperature and the material fineness are critical factors that must be carefully controlled to achieve optimal results. It can also be observed that the two samples that do not have calcination data do not meet the criteria to be considered pozzolanic materials. Therefore, it would be beneficial to investigate the impact of different grinding methods, as, as observed, the fineness of the material improves its pozzolanic efficiency by increasing the reactive surface area. Hence, a combination of calcination at optimal temperatures (around 600 °C) and advanced grinding techniques can maximize the performance of POFA ashes as a partial cement replacement material.
It can also be observed, upon making the comparison, that the samples that do not meet the minimum requirement of 70% for the sum of SiO
2, Al
2O
3, and Fe
2O
3 have only one grinding or crushing process, whereas those that do meet the requirement mostly have more than one process. For example, U-POFA and T-POFA have an additional grinding process, and the samples from this study undergo three processes. As a future recommendation, it is urged to conduct a detailed evaluation of the properties of the soil in which the oil palm is grown to determine its impact on the silica content of the kernel ashes. Soil quality can significantly influence the silica content and other compounds in the ashes, which would, in turn, affect their pozzolanic reactivity [
68], since the soil pH affects the solubility of silicon in it, facilitating the absorption of silica by the crops [
69]. Similarly, the texture and the presence of organic matter in the soil can affect the availability of silica contained in it, and salinity can impact the absorption of nutrients by the crops [
68].
3.2. Loss on Ignition (LOI)
The calcination of biomass ashes such as POFA becomes necessary primarily to reduce the loss on ignition (LOI), which corresponds to the content of unburned carbon and other organic impurities that do not contribute to pozzolanic activity. The current standards, represented by ASTM C-618 [
39], set a maximum LOI limit of 10% for Class N pozzolans to be suitable for use in concrete, as higher values negatively affect the reactivity of the material and, consequently, the mechanical properties of the concrete. A high LOI value is indicative of unburned carbon, which interferes with the hydration reaction and increases the water demand, affecting the final properties of the concrete. In the case of POFA, the calcination process helps reduce these undesirable components, improving the purity of the material and activating SiO
2 for the formation of calcium silicate hydrates (C-S-H).
The procedure of ASTM C311 standard [
45]: It sets a temperature of 750 °C ± 50 °C for natural pozzolans as the standard to ensure consistent results in the LOI measurement without significantly degrading other important components of the material. If the temperature is too high, it may affect the composition of the remaining materials, altering the number of oxides and other critical components in the analysis.
Based on the literature reviewed and referenced in
Table 3, the samples that do not present calcination data [
38,
42] showed a loss on ignition of 20.7% and 15%, respectively, well above the limit specified by ASTM, suggesting a significant amount of carbonaceous and non-reactive materials. As the calcination temperature increases, a significant reduction in LOI is observed, with values ranging from 1.8% at 500 °C to 6.19% from 800 °C to 1000 °C for the POFA samples with various grinding processes [
37,
56,
64]. These results confirm that calcination is necessary not only to remove unburned carbon but also to improve the reactivity of SiO
2 and ensure the effectiveness of the material as a pozzolan in concrete production.
Table 4 shows the relationship between the calcination temperatures of the POFA samples calcined at temperatures ranging from 500 °C to 1000 °C and the losses associated with the organic material, primarily represented by the carbon present in the samples.
Below is a graphical representation of the variation in organic material content losses for the different samples analyzed, as show in
Figure 7. It can be observed that sample M0, which undergoes drying at 100 °C instead of calcination, exhibits a moisture content of 14.5%. In contrast, sample M17A, subjected to calcination at 500 °C, experiences organic material losses reaching 72.60%. Notably, there is a decrease in organic material losses for the sample calcined at 600 °C (M18A), dropping to 70.36%, making it the sample with the lowest recorded losses. From this point onward, a linear relationship between calcination temperature and organic material loss percentage becomes evident, indicating that higher temperatures result in greater losses.
As can be seen, the percentage of weight loss due to organic material content is very high. This can be verified by the color of the samples. (
Figure 4); the samples are black in color, indicating a high carbon content. Additionally, it should be noted that the sample
M18A-c-m-T600C-t1.5h-tr1h has the lowest percentage of organic material losses, establishing that this temperature allows for the burning of a higher carbon content compared to the other samples evaluated. However, the percentage of organic material losses remains very high. Therefore, it was decided to subject the sample to calcination once again.
For this process, the optimal temperature of 600 °C, previously established based on the results obtained through XRF, was considered. Therefore, the sample initially calcined at 500 °C (considered the temperature at which the combustion of the oil palm seed begins) was subjected to a second calcination at 600 °C for 2 h, with stirring to incorporate the material, followed by an additional 30 min of calcination, resulting in the final sample
MDEF-c-m-T500C-t1.5h-tr1h-T600C-t2h. It should be mentioned that the samples were saved after the second combustion to obtain a homogeneous particle size. After this process, a color difference can be observed (
Figure 8), compared to the samples before this process (
Figure 9).
Below is a comparative image between the samples M18A-c-m-T600C-t1.5h-tr1h and the final sample MDEF-c-m-T500C-t1.5h-tr1h-T600C-t2h, where it can be observed that with the second calcination at 600 °C, the characteristic black color associated with high carbon content is eliminated.
The LOI was calculated according to the ASTM C311 standard for the final sample, which underwent cutting, grinding in a jaw mill, burning at 500 °C, crushing in a ball mill, and subsequent calcination at 600 °C (
MDEF-c-m-T500C-t1.5h-tr1h-T600C-t2h). To eliminate residual organic material, for the LOI determination, the final sample was heated to 750 °C, as it is natural pozzolan (Equation (1), loss on ignition (LOI)).
where
A = Mass loss between 105 °C and 750 °C
B = Mass of the sample without moisture used.
In this way, 15 g of the final POFA samples were taken (
MDEF-c-m-T500C-t1.5h-tr1h-T600C-t2h) with calcination at a temperature of 500 °C followed by 600 °C, and the sample was subjected to a muffled furnace at 750 °C for 1 h. The final weight of the sample was 14.5 g, resulting in a loss on ignition (LOI) of 3.33%, significantly decreasing. This LOI value is within the limits allowed by ASTM C618, indicating that the oil palm seed ashes obtained through the three applied physical processes (cutting, grinding, and crushing) and the two thermal processes (calcination at 500 °C and subsequent calcination at 600 °C) can potentially be used as pozzolanic material in concrete. The obtained results highlight the importance of calcination, particularly at this temperature, as thermal processes allow for the removal of organic material. The significance of removing organic material is that it does not react properly with cement during the hydration process. Therefore, the decomposition of organic materials within the concrete can release gases or leave voids, leading to the formation of cavities that weaken the structure and may increase permeability [
70]. This prevents the proper formation of hydration products, such as calcium silicate hydrates (C-S-H), which are responsible for the mechanical strength of concrete.
It should be mentioned that, in addition to the calculation of the loss on ignition at 750 °C as stated by the ASTM C311 standard, losses at 650 °C were also determined, resulting in a loss on ignition percentage of 4.4%. This confirms that the two calcination processes contribute to the removal of most of the organic material present in the sample.
3.3. Scanning Electron Microscopy (SEM) (SEM EDS/EDX)
A scanning electron microscopy (SEM) and energy-dispersive X-ray (EDX) analysis was performed on the sample M18A-c-m-T600C-t1.5h-tr1h. This technique allows for high-resolution images of a sample’s surface using an electron beam. It is used to analyze nanoparticles of various materials and can enhance manufacturing methods, purification systems, and make improvements in the medical industry, among others [
71]. SEM allows for the detailed study of morphology and composition of materials. After performing this analysis, the sample M18A-c-m-T600C-t1.5h-tr1h showed a predominant peak of carbon (C) compared to the other elements, with a significant content of oxygen (O). (
Table 5 and
Figure 10b). This indicates that, with just a single calcination process at 600 °C, there is still a considerable amount of carbonaceous material and residues that have not been fully eliminated, which affects the quality of the ashes as pozzolan.
In
Figure 10b, the clear difference in the intensity of the carbon peaks compared to silica (Si), potassium (K), magnesium (Mg), and calcium (Ca) is observed after performing the SEM analysis on the sample M18A-c-m-T600C-t1.5h-tr1h. This visually reveals the high organic material content present in this sample and highlights the need for a new process to eliminate it, thereby aiming to improve POFA for replacing ordinary cement. In
Figure 10a, the spectrum obtained through the SEM analysis of the sample M18A-c-m-T600C-t1.5h-tr1h is presented.
The sample M18A-c-m-T600C-t1.5h-tr1h has an average carbon content greater than 82% (
Table 5); this suggests a high presence of unburned organic material. It indicates that the complete elimination of volatile compounds has not been achieved. Once the process was standardized and it was determined that the oil palm seed ash should be obtained as a potential pozzolanic material for concrete through the following processes: cutting, grinding, burning in a muffle furnace for 1.5 h at 500 °C, crushing in a ball mill, and burning in a muffle furnace for 2 h at 600 °C, SEM tests were performed again on this sample (MDEF-c-m-T500C-t1.5h-tr1h-T600C-t2h) as well as SEM tests for samples subjected to subsequent temperatures of 650 °C and 750 °C for loss on ignition determination. The SEM analyses showed the elimination of carbon in the sample and an increase in silica content
Table 6 and
Figure 11b.
In
Figure 11a, the spectrum of the sample with the standardized process is shown. When compared to the spectrum of the sample M18A-c-m-T600C-t1.5h-tr1h (
Figure 10a), there is a noticeable difference in particle size and uniformity achieved with the standardized process. It can be established that the second calcination allows us to produce smaller ash particles, facilitating their incorporation and mixing in the concrete. This results in better filling of voids, reducing concrete porosity and, consequently, enhancing its resistance to the attack of aggressive substances, such as chlorides and sulfates.
Similarly, the SEM analysis was also performed on the samples calcined at 650 °C and 750 °C after the standardized process for LOI determination. This last test was conducted to observe the behavior of the POFA based on the standardized process and its increase or decrease in carbon and silica content.
In
Table 7 and
Figure 12b, the weight percentage and the intensity graph of elements for the standardized sample are observed (
MDEF-c-m-T500C-t1.5h-tr1h-T600C-t2h), calcined at 650 °C, to evaluate the loss on ignition at this temperature. In
Figure 12a, the spectrum of this sample is shown.
When comparing the chemical composition of the standardized sample
MDEF-c-m-T500C-t1.5h-tr1h-T600C-t2h with that obtained for the same sample subject to an additional temperature process at 650 °C for LOI evaluation, an increase in silica content is observed. This confirms that the variation in silica concentrations indicates that the thermal conditions during the calcination process have a direct effect on the redistribution and reaction of the silica components present in the samples [
30]. In
Table 8, the content of the elements in the sample with the standardized process, subsequently calcined at 750 °C to evaluate LOI, is shown in weight percentage. In
Figure 13a,b, the intensity of the elements and the spectrum are shown, respectively, for this sample.
By comparing the elemental analysis for the sample with the standardized process and the samples with the standardized processes followed by burning at 650 °C and 750 °C, it can be observed that the silica content increased, with percentages of 49.49%, 54.45%, and 62.67%, respectively. Additionally, no carbon content is observed for these three samples. The following figure more clearly shows this variation, indicating that for the sample with a second calcination (
MDEF-c-m-T500C-t1.5h-tr1h-T600C-t2h), transitioning from 500 °C to 600 °C, no carbon content is exhibited, in contrast to M18A (
M18A-c-m-T600C-t1.5h-tr1h), which underwent a single calcination process at 500 °C and shows a carbon peak of 86.26%, as show in
Figure 14.
These results are consistent with the LOI results, which showed that as the calcination temperature increases (up to 750 °C), the LOI percentage decreases, indicating a reduction in carbon content. It can also be stated that due to the increase in silica, the reactivity of these samples improves, which further enhances their properties as potential pozzolans. Following these considerations, it is important to mention that as the carbon content decreases with the different thermal processes, the silica (Si) content increases because of the combustion of organic material.
Finally, the XRF analysis results obtained for the sample
MDEF-c-m-T500C-t1.5h-tr1h-T600C-t2h are presented in
Table 9.
Then, from the table above, it is evident that the sample with the standardized process meets the requirements of the ASTM C618 standard, achieving a content of basic oxides of 75%, demonstrating its potential as a pozzolanic material. Thus, the importance and need to continue exploring this material to demonstrate its pozzolanicity through physical and chemical characterization techniques is clear. It should be mentioned that although the silica (SiO2) content in the final sample (MDEF-c-m-T500C-t1.5h-tr1h-T600C-t2h) decreases by 2.94% and 4.67% compared to the values obtained for samples M17A-c-m-T500C-t1.5h-tr1h and M18A-c-m-T600C-C-t1.5h-tr1h, the basic oxide content in the final sample still meets the ASTM C618 requirements, confirming its potential pozzolanic use. It is crucial to note that despite this slight decrease in SiO2 content, the reduction in organic material losses is significant and highly favorable. Therefore, it can be stated that this second calcination process at 600 °C is necessary to confer pozzolanic properties to the oil palm kernel shell ash in compliance with the current standard.