Sample Origin Effect on Chemical Reactivity of Tajogaite Volcanic Ashes for Ancient Mortar Repair

Abstract

Volcanic ashes (VA) ejected by the Tajogaite Volcano were studied to determine their potential as pozzolanic materials for construction applications. A representative number of VA samples (15 in total) were collected from different geolocations and altitudes during and immediately after the volcanic eruption, in order to assess their reactivity as a function of position and environmental exposure. Various analytical techniques—XRD, FTIR, and SEM/EDX—were used to determine the initial microstructural composition of the VA samples. Additionally, saturated lime testing and the Frattini test were performed to evaluate their pozzolanic reactivity for use in historical mortars. The microstructural analyses revealed that the dominant mineral phases are aluminosilicates. The reactivity tests confirmed a good pozzolanic response, with the formation of C-A-S-H gels identified as the main hydration products at the studied curing times.

1. Introduction

Volcanic areas classified as active are those where eruptions have occurred within the last 10,000 years. In Spain, only two regions meet this criterion: the Canary Islands and the Garrotxa region in Girona (Catalonia). On 19 September 2021—the day the eruption began—over 1500 seismic events were recorded in the Cumbre Vieja Natural Park area, where the Teneguía volcano is also located. The Tajogaite Volcano emerged from the eruption that took place on the island of La Palma in 2021. It reaches an elevation of 1120 m above sea level, with a major axis of 172 m and a minor axis of 106 m. This eruption destroyed more than 2000 buildings in the Aridane Valley and devastated over 1000 hectares of land. The emission of lava and ash ceased on 13 December, after 85 days of continuous activity, resulting in a substantial accumulation of volcanic ash. The vast quantity of this material—often referred to as tephra—was rapidly introduced into the environment, raising serious concerns regarding its potential impacts on human health and agriculture.

Current restoration plans for the affected area include the reuse of volcanic ejecta, as well as the reconstruction of houses buried by lava. Among volcanic products, ash is particularly concerning due to its content of potentially toxic elements such as sulfur, aluminum, and occasionally radioactive components [1,2,3]. Volcanic ash (VA) is composed of rock fragments, minerals, and volcanic glass (amorphous silicon oxide) [4]. Its oxide composition typically includes silica (SiO₂), alumina (Al₂O₃), iron oxides (Fe₂O₃/FeO), and, to a lesser extent, oxides of alkali and alkaline earth metals such as CaO, MgO, Na₂O, and K₂O, as well as various soluble salts.

It is well known that materials ejected by erupting volcanoes have been used in mortars and concrete for construction purposes since ancient times [5,6,7,8,9,10]. The composition of volcanic ash also includes fine particles and varying proportions of volcanic glass shards, minerals or crystals, and other rock fragments [11]. These ashes are generated as a result of the rapid cooling of lava droplets expelled into the atmosphere during the degasification process of volcanic eruptions.

Volcanic ash (VA) can be processed for a variety of applications, including ceramics, lightweight aggregates, bricks, and concrete products [9,11]. Previous studies have categorized volcanic ash into two main groups: the first is derived from molten rock and consists primarily of basaltic compositions; the second originates from more explosive pyroclastic flow eruptions and leads to the formation of secondary pozzolanic clays and zeolitic phases [12,13,14].

The significant carbon footprint of conventional cement-based materials has recently driven interest in using volcanic ash as a partial replacement for Portland cement [15,16,17,18,19]. This approach is further supported by the need to manage the large volumes of ash that accumulate in limited areas during volcanic events, which can hinder normal activities. The incorporation of VA into concrete is not a new practice—historical constructions already contain it [12,20,21,22]. Roman builders, for example, used VA as a natural source of aluminosilicates to create strong and durable structures [23].

Not all volcanic ashes can be used as pozzolans [24,25]. It is important to note that the pozzolanic activity of volcanic ash (VA) depends primarily on the amount of reactive silica and the content of the amorphous phase [12]. Djon Li Ndjock et al. [26] concluded that VA can be effective either as a filler in mortars and concretes or for cement synthesis, depending on both the quantity and the molar ratio of SiO₂/Al₂O₃ in its amorphous phase. A low content of reactive silica may result in limited pozzolanic activity, rendering the VA unsuitable for use in blended cements [25].

The valorization of volcanic ash as a supplementary cementitious material aligns with key principles of sustainability in the construction sector. By utilizing a locally available and abundant by-product of the Tajogaite eruption, the environmental impact associated with the extraction and processing of conventional raw materials is significantly reduced. Moreover, the partial replacement of Portland cement with volcanic ash helps to lower CO₂ emissions, as cement production is one of the largest industrial sources of greenhouse gases. This approach supports circular economy strategies by transforming a potentially problematic volcanic residue into a valuable resource for eco-efficient construction materials. The methodology adopted in this study—based on standardized tests and extended durability analysis—reinforces the feasibility of integrating sustainable materials into real-world applications.

To enhance the pozzolanicity of volcanic ash, various methods have been proposed, including increasing the specific surface area by grinding to reduce particle size and enhancing the zeolitic mineral content by removing clay minerals [27]. Shah et al. [28] reported improved pozzolanic activity through mechanical activation using vibratory milling, which increased the amorphous content. Pozzolanic activity is further enhanced when aluminosilicate precursors with high cation exchange capacity are used [29]. Additionally, the reactivity of volcanic ash in blended cements can be improved through treatments such as calcination [30], acid activation [31], or other thermal processes [32]. The aim of this work is to establish the pozzolanic activity of 15 VAs from different geolocations from Tajogaite Volcano, La Palma, Canary Islands, by means of chemical methods: saturated lime and Frattini tests. In addition, before and after the chemical treatment a microstructural, morphologic, and chemical analysis is performed through XDR, FTIR and SEM.

2. Materials and Methods

2.1. Materials

A significant number of samples (15 in total, see

Table 1. Coordinates and altitude of collected VA samples.

Sample		Crater Orientation	Coordinates X	Coordinates Y	Sea Level (m)	Accumulation (cm)
M2		N	28R 219405	3169339	803	60
M3		NW	28R 215674	3170678	302	0–1
M4		NW	28R 217951	3170010	522	0–1
M5		N	28R 218771	3170626	695	1–2
M6		N	28R 219039	3169670	738	12
M7		S	28R 219002	3163586	687	2
M8		SW	28R 216959	3167330	458	3
M9		W	28R 216268	3168801	344	3
M10		W	28R 215816	3168552	286	1–1.5
M11		SW	28R 217687	3167280	542	10
M12	a	SW	28R 218273	3167395	611	7
	b	SW	28R 218273	3167395	611	10
	c	SW	28R 218273	3167395	611	9
M14	a	SW	28R 218725	3168017	661	120
	b	SW	28R 218725	3168017	661	60

) were collected from the volcano during the initial days of the eruption and while the eruption was still ongoing. The sampling was carried out by the Geotechnical Engineering and Geosciences Group of the Universitat Politècnica de Catalunya (UPC). Representative images of the sampling process are shown in Figure 1, Figure 2, Figure 3 and Figure 4. Table 1 and Figure 5 include the geolocation map indicating the sampling sites. The samples were collected along the main lava flow of the volcano at various elevations, ranging from 800 to 200 m above sea level. In some locations, multiple samples were taken to assess the accumulated mass at the same elevation, particularly with respect to the thickness of deposited ash (in cm), as illustrated by samples 12 and 14.

X and Y correspond to UTM coordinates (Zone 28N, ETRS89 datum), expressed in meters. Sampling locations correspond to the colored markers shown in the Figure 5.

2.2. Methodology

The pozzolanic activity of the volcanic ash (VA) was evaluated using two methods: the Saturated Lime Solution (SLS) method and the Frattini test (Ft). Both methods are briefly described below:

Accelerated Chemical Method with Saturated Lime Solution (SLS): This method monitors the reaction between ash and lime over time. One gram of each sample was placed in contact with 75 mL of saturated lime solution inside 100 mL double-cap polyethylene flasks. The flasks were then stored in an oven at 40 ± 1 °C for 3, 7, and 28 days (two flasks for each curing period).

At the end of each period, the solutions were filtered, and the residual CaO content in the filtrate was determined. For this, 20 mL of the filtered solution was titrated using EDTA (0.0178 mol/L) with calcein as the indicator. The amount of fixed lime (mmol/L) was calculated as the difference between the initial concentration of CaO in the saturated lime solution and the concentration found after reaction with the sample [33]. This method has also been applied to assess the pozzolanicity of other industrial residues, including paper sludge waste [34], sugarcane straw ash [35,36], and ferroalloy industry by-products [37].

Frattini test (Ft): The Ft was performed according to the UNE-EN 196-5:2011 standard [38,39]. This test evaluates the reaction between the pozzolanic material and calcium hydroxide in solution over time. It has been previously used to assess the pozzolanic activity of materials such as metakaolin [40], catalytic cracking residues [41], crushed bricks [42], and fly ash [43].

The concentration of Ca²⁺ ions and the alkalinity (OH⁻) of the solution in contact with the paste were measured after 3, 7, and 28 days. The OH⁻ concentration was determined by titration with HCl using methyl orange as an indicator, while the Ca²⁺ concentration was determined using EDTA and calcein, as specified in the standard.

Prior to testing, all samples were subjected to a grinding process to homogenize their particle size distribution, as they originally displayed a wide granulometric range. Grinding was performed using a disc mill for 10 min until a dust-like consistency was achieved. The ground material was then sieved through a 63 μm mesh, and the fraction passing the sieve was used for the experiments (Figure 6).

X-ray diffraction (XRD) patterns for the mineralogical composition of the volcanic ash (VA) were obtained using a D2 PHASER diffractometer (Bruker, Billerica, MA, USA) equipped with a secondary graphite monochromator (CuKα₂, flat sample geometry). The instrument was operated at 40 kV and 50 mA, with a step size of 0.01° and a counting time of 0.5 s per step, across a 2θ range of 5° to 60°. A semi-quantitative analysis of the mineral phases was conducted based on the intensity of the XRD reflections. XRD analysis was performed on samples M1–M14 both before and after the Frattini tests (3, 7, and 28 days). To arrest the pozzolanic reaction, the hardened pastes were crushed and immersed in acetone, followed by filtration and rinsing with ethanol, then stored in a desiccator over silica gel for 24 h [44].

Following this preparation, the samples were analyzed by infrared (IR) spectroscopy to identify changes in amorphous and crystalline phases over time (3, 7, and 28 days). The IR spectra were collected using a Bruker Alpha spectrometer (400–4000 cm⁻¹) equipped with a Platinum ATR module (Bruker, Billerica, MA, USA). Morphological and chemical microanalyses were conducted using Scanning Electron Microscopy (SEM) coupled with Energy Dispersive X–ray Spectroscopy (EDS). This technique provided both qualitative and semi-quantitative elemental analysis at specific points or regions of interest. The ash particles were mounted on metal stubs using conductive adhesive tape and placed in the sample holder. SEM images were acquired using a Phenom XL G5 scanning electron microscope (Thermo Fisher Scientific, Waltham, MA, USA). The measurements were carried out at an acceleration voltage of 15 kV under high vacuum conditions (1 Pa). A cerium hexaboride (CeB₆) filament was used as the electron source. Microstructural observations and images were acquired using backscattered electron detection.

3. Results and Discussion

3.1. Raw Volcanic Ash Characterization

Figure 7 shows the X-ray diffraction (XRD) patterns of the initial volcanic ash (VA) samples. The diffractograms reveal multiple diffraction peaks corresponding to crystalline phases, as well as a broad hump or halo indicating the presence of an amorphous phase, characteristic of disordered atomic structures [45]. The amorphous fraction of volcanic ash typically contains oxides such as SiO₂, Al₂O₃, MgO, Fe₂O₃, and CaO [26].

The crystalline mineralogical composition appears consistent across all samples. The dominant crystalline phases identified are augite ((Ca,Na)(Fe,Mg,Al,Ti)(Si,Al)₂O₆) and albite (NaAlSi₃O₈), both aluminosilicates belonging to the feldspar group. Olivine ((Mg,Fe)₂SiO₄), a neosilicate, was also detected but only in sample M–12C was it absent—replaced instead by actinolite (Ca₂(Mg,Al,Fe²⁺)₅Si₈O₂₂(OH)₂). These crystalline phases, particularly augite and olivine, have been frequently reported in volcanic ashes from other volcanoes [7,46,47].

Augite is the most prominent mineral phase in all samples, followed by albite. Olivine is present in lower proportions. As mentioned, sample M–12C is the only exception, where olivine is absent and actinolite appears instead, although augite remains the predominant phase.

The infrared (IR) spectra of all samples exhibit a similar profile; Figure 8 presents a representative example. The absorption bands obtained are consistent with those reported in previous studies [48,49,50,51]. The main spectral feature is a broad band centered around 1000 cm⁻¹, resulting from the superposition of several vibrational modes. A shoulder at 898 cm⁻¹ is attributed to asymmetric stretching vibrations (νₐₛ–Si–O–Si/νₐₛ–Si–O–Al) [52]. The presence of aluminosilicates is further supported by a weak band at approximately 740 cm⁻¹, corresponding to deformation vibrations (δ–Si–O–Si/δ–Si–O–Al) [53], and a shoulder around 1050 cm⁻¹ associated with vibrations of amorphous aluminosilicate structures [54], in agreement with the XRD results.

The most prominent contribution comes from asymmetric stretching vibrations of Si–O–Si bonds, forming broad absorption bands between 990 and 1000 cm⁻¹ [55]. In some samples, such as M8 and M9, additional bands at 1650 and 1540 cm⁻¹ are observed, corresponding to the bending (δ–H–O–H) vibrations of water molecules bound within the structure [56]. As previously reported [57,58], the fine structure of this band is due to the overlap of individual peaks, which are attributed to asymmetric stretching of T–O–Si (where T = Si or Al) bonds in the amorphous aluminosilicate network. The broad nature of the absorbance band is also related to the variability in bond angles and bond lengths within the tetrahedral framework around the silicon atoms [52].

A representative analysis of the volcanic ashes was conducted using Scanning Electron Microscopy (SEM). The observations revealed a variety of morphologies, with no single dominant shape identified across the samples. Similar morphological features were observed in all samples; Figure 9 presents SEM images of samples M2, M6, M10, M12–A, and M14–A. Samples M2, M6, M10, and M12–A exhibit a compact, bulk-solid morphology with small crystalline particles deposited on their surfaces. At higher magnifications, numerous surface and internal cavities or tunnels become apparent. These structures, resembling zeolite-like morphologies, are particularly distinct in sample M14–A. The texture of the volcanic ash is characterized by numerous vesicles—spherical or elongated cavities formed by gas bubbles trapped within the material during the eruption process [59]. These vesicular structures result from the rapid release of volatile gases during magma degassing, which creates the porous texture typical of pyroclastic materials [60].

The EDX analysis conducted on the volcanic ash samples revealed that their chemical composition is primarily dominated by silicon (Si), aluminum (Al), sodium (Na), iron (Fe), and calcium (Ca). Minor amounts of other elements such as potassium (K), titanium (Ti), and fluorine (F) were also detected. The oxide compositions and the CaO–Al₂O₃–SiO₂ ternary diagram derived from the EDX results are shown in Figure 10. In all analyzed samples, SiO₂ was the most abundant oxide, ranging from 25% to 50% by weight. This was followed by Al₂O₃, typically between 10% and 20%, with CaO values close to those of alumina.

These findings, in agreement with the XRD and FTIR analyses, confirm that aluminosilicates constitute the main mineralogical framework of the volcanic ash, consistent with previous studies [7,42,43]. Notably, moderate amounts of CaO, K₂O, and Na₂O (typically 5–10 wt%) were also identified. Other oxides such as MgO, Fe₂O₃, and SO₂ were present but showed a more irregular distribution among the samples. The CaO–Al₂O₃–SiO₂ ternary diagram places the samples within a well-defined compositional region associated with natural pozzolans and fly ash [61]. This position, characterized by high silica and alumina content with comparatively low calcium levels, supports the suitability of Tajogaite volcanic ash as a pozzolanic material. While the potential substitution of Portland cement is notable, this study specifically highlights the use of volcanic ash as a pozzolanic additive in lime-based mortars. In conclusion, Tajogaite volcanic ash can be classified as a low-calcium silico-aluminous ash, indicating promising pozzolanic reactivity and applicability in sustainable construction materials.

The identification of crystalline phases such as olivine, clinopyroxene, and albite (Figure 10) provides a general mineralogical context for the tephra. However, given the primary objective of this study—evaluating the pozzolanic and cementitious behavior of the material—detailed quantification of the crystalline and amorphous phases was not deemed necessary at this stage. The employed methodologies, specifically the Frattini test and the Saturated Lime Solution (SLS) method, are designed to assess the overall reactivity of powdered samples without requiring precise phase differentiation. Nevertheless, further detailed characterization of the amorphous phase, including its chemical composition and structural features, could provide valuable insights in future research.

The Si/Al and Na/Al molar ratios (Figure 11) were found to range between approximately 2 to 3 and around 0.5, respectively. Previous studies suggest that these ratios can be predictive of the mechanical strength of cementitious materials incorporating volcanic ash [51]. Specifically, Si/Al molar ratios above 2.5 are associated with enhanced strength development. The volcanic ashes analyzed here meet this criterion for the Si/Al ratio, but the Na/Al ratio is relatively low. According to earlier research, iron-bearing phases such as augite may serve as microcrystalline fillers or even participate in geopolymerization reactions under certain conditions [62,63]. The formation of geopolymer structures typically requires high Si/Al ratios to achieve sufficient polymerization and mechanical performance [6]. The dissolution rate of alkali feldspar is primarily influenced by factors such as pH, temperature, and the presence of organic acids. It is likely that the temperature increase during hydration enhances the dissolution of these phases. Based on these considerations, further investigation into the mechanical properties and hydration behavior of Tajogaite volcanic ash is warranted.

It should be noted that the elemental ratios presented in Figure 11 were derived from EDX analyses performed on various sample areas selected based on their morphological features observed under SEM. However, since EDX does not differentiate between amorphous and crystalline phases, these compositional data cannot be conclusively assigned to a specific phase type. While some analyzed regions may correspond to the glassy (amorphous) matrix, this cannot be definitively confirmed using EDX alone.

3.2. Pozzolanic Activity Test

Table 2. Results of Saturated Lime Solution test at 3, 7 and 28 days.

	[CaO]–3 Days		[CaO]–7 Days		[CaO]–28 Days
Sample	mM/L	% vol. Fixed	mM/L	% vol. Fixed	mM/L	% vol. Fixed
M2	1.08	6.11	5.58	31.56	14.77	83.51
M3	1.73	9.79	8.08	45.70	15.98	90.38
M4	3.28	18.55	4.38	24.77	13.58	76.81
M5	4.68	26.47	4.48	25.34	13.78	77.94
M6	2.23	12.61	4.93	27.88	13.23	74.83
M7	2.48	14.03	3.58	20.25	14.43	81.62
M8	1.88	10.63	4.88	27.60	14.98	84.73
M9	2.53	14.31	5.13	29.02	16.03	90.67
M10	2.88	16.29	5.48	31.00	14.18	80.20
M11	2.18	12.33	5.13	29.02	14.58	82.47
M12-A	3.28	18.55	5.33	30.15	14.88	84.16
M12-B	2.73	15.44	4.28	24.21	13.43	75.96
M12-C	2.58	14.59	4.78	27.04	12.63	71.44
M14-A	3.13	17.70	5.43	30.71	13.13	74.26
M14-B	2.33	13.18	5.13	29.02	13.28	75.11

reports the concentration of fixed lime, expressed in mmol/L and percentage, for the three-time intervals studied. The results of the pozzolanic activity determined by the Saturated Lime Solution (SLS) test are presented in Figure 12, which plots fixed lime content (% vol.) against reaction time (days). The figure clearly demonstrates the pozzolanic behavior of the different ashes over time.

At 3 days, all samples exhibit a similar amount of fixed lime, with an average of 14.71%. This value lies between that of silica fume and metakaolin and is higher than that of fly ash [64]. Samples M2 and M3 show a lower fixation capacity, below 10%, while sample M5 fixes more than 20% of the lime. These variations cannot be correlated with the sample location or accumulation thickness, as no significant differences were observed between them.

At 7 days, pozzolanic activity increases, particularly for samples that showed lower fixed lime at 3 days, such as M2 and M3, reaching an average of about 29%. Differences between samples diminish, showing more uniform fixed lime values.

By 28 days, pozzolanic behavior is maximized, as evidenced by the rapid consumption of available lime, with some samples exceeding 90% fixed lime. This level of pozzolanic activity is comparable to that of silica fume [65,66], which highlights the potential of Tajogaite volcanic ash as an active additive in mortars. The high reactivity is likely related to the content of the amorphous phase, similarly to silica fume. Consequently, various blended binders with tailored properties could be developed using this material.

Based on these studies, it can be concluded that the samples share a similar mineralogical composition and exhibit very similar behavior in the Frattini test (Ft). Therefore, different sample types were mixed to create a representative, generic sample for tests requiring larger quantities, such as mechanical testing. To validate the consistency of this representative sample, it was also analyzed using the Frattini test. The Ft results are shown in Figure 13 as points on a graph of [CaO] versus [OH⁻]. In this graph, the calcium isothermal curve separates the pozzolanic region (below the curve) from the non-pozzolanic region (above the curve). At both 7 and 28 days, the results fall within the pozzolanic area, confirming the findings from the Saturated Lime Solution (SLS) test. Good pozzolanic activity is observed at 7 days, with further progress evident at 28 days. The Ft results indicate that the behavior of the representative volcanic ash lies between that of silica fume and metakaolin [39].

Recent work by García-González et al. (2024) has also investigated the pozzolanic potential of volcanic ash from the Tajogaite eruption [67]. Although the general objective is similar, the methodology and scope of both studies differ significantly. In that work, the pozzolanic activity was inferred from the mechanical performance of mortars containing volcanic ash as a cement replacement, with compressive strength tests performed only at 28 days. In contrast, our study applies the Frattini test according to EN 196-5:2011, a standardized chemical method that allows a direct evaluation of the pozzolanic reaction in the early stages and under controlled conditions without the interference of cement hydration. Both approaches are complementary, and together they contribute to a more comprehensive characterization of this volcanic material. Additionally, while García-González et al. report on the amorphous content of the ash, our work focuses on its chemical reactivity and performance in mortar formulations with lime and cement.

3.3. Pozzolanic Reaction Products Characterization

The main hydration reactions of Portland cement transform tricalcium silicate (C₃S) and dicalcium silicate (C₂S) into calcium silicate hydrates (C–S–H) and calcium hydroxide (CH) [reactions 1 and 2]. Pozzolanic materials can fix this CH and generate gels with cementitious properties (GCP). In the presence of water, CH reacts with the added pozzolan, resulting in the formation of additional calcium silicate hydrates (C–S–H), calcium aluminate hydrates (C–A–H), and calcium aluminosilicate hydrates (C–A–S–H) [reaction 3]. The nature and amount of these products depend on the pozzolan’s reactivity, which is influenced by treatment conditions, material purity, and fineness [67]. Volcanic ash was studied at the ages used in the saturated lime test to identify the different pozzolanic reaction products.

C₃S + H₂O → C-S-H + CH

(1)

C₂S + H₂O → C-S-H + CH

(2)

The reaction with lime is produced as follow:

Pozzolan + CH + H₂O → GCP (C-S-H, C-A-H, C-A-S-H)

(3)

Figure 14 presents the X–ray patterns that illustrates the evolution of sample M12–A over time during the saturated lime test, highlighting the formation of crystalline phases due to the reaction with lime. Initial crystalline phases—augite, albite, olivine, and actinolite—remain unchanged, retaining their diffraction peak intensities. As the test progresses, two new crystalline phases emerge: hydrocalumite and hydrotalcite, both increasing in intensity over time. The production of hydrocalumite is more pronounced than that of hydrotalcite. Hydrocalumite typically forms as a product of the pozzolanic reaction between aluminosilicates and portlandite [68]. Thus, the presence of these phases indicates an active reaction between volcanic ash and lime, confirming the pozzolanic nature of the material. Both phases belong to the family of anionic clays, also known as layered double hydroxides (LDHs) [69].

The general formula of hydrocalumite is [Ca₂M³⁺(OH)₆]⁺[${\mathrm{A}}_{1/\mathrm{n}}^{\mathrm{n}-}$·mH₂O], where the hydroxide layer consists mainly of divalent and trivalent cations, typically Ca²⁺ and Al³⁺ [70]. In this study, the hydrocalumite identified by XRD corresponds to [Ca₂Al(OH)₇·H₂O], where Ca²⁺ occupies the divalent cation positions. This is consistent with the high calcium content and the possibility that aluminum participates in other reactions. These lamellar calcium aluminum hydroxide salts have been extensively studied because they occur during cement hydration, leading to the formation of hydrogarnet [71].

Hydrotalcite has a variable composition described by the general formula Mg_1−x(Al, Fe)_x(OH)₂·[Aⁿ⁻]_x/n·mH₂O.

Its structure consists of positively charged brucite-like layers intercalated with anions [An-] and water molecules. This structure can accommodate various cations, interlayer anions such as OH⁻, Cl⁻, ${\mathrm{C}\mathrm{O}}_3^{2-}$ and ${\mathrm{S}\mathrm{O}}_4^{2-}$, and varying amounts of water [72]. Hydrotalcite was identified by XRD and is reported to have the composition [Mg₆Al₂CO₃(OH)₁₆·4(H₂O)]. The carbonate anions ${\mathrm{C}\mathrm{O}}_3^{2-}$ in the interlayer likely arise from excess carbonates in the aqueous environment, facilitating their incorporation into the layered double hydroxide structure. In Portland limestone cements with high magnesium content, hydrotalcite formation is common [73].

On the other hand, it is verified that the halo due to amorphous species found in the X–ray diffractograms of the initial samples is almost imperceptible in this case, which could indicate the high reactivity of these phases.

The progress of the pozzolanic reaction can be followed by infrared spectroscopy. Figure 15 shows the infrared spectra at different ages during the saturated lime test of sample M4. It shows the general behaviour of the formation of reaction products in the tested samples. The spectra are quite similar at the beginning, as well as at 3 and 7 days, with only small changes in the main peaks. At 28 days the reaction is more advanced and the changes in the spectra are more pronounced.

In the first zone between 2400 and 4000 cm⁻¹, the different asymmetric vibrations due to the O–H bonds appear; these vibrations are well known in the existing literature [73,74,75]. A weak band at 3670 cm⁻¹ associated with portlandite is observed only at 28 days. Portlandite was not detected in XRD, probably due to the low intensity; in the infrared spectrum its presence is not high and could be covered by the broad signal of the amorphous phases in XRD preventing its detection. The bands associated with hydrocalumite and hydrotalcite situated at 3550 and 3370 cm⁻¹, respectively, form a small band at 7 days and at 28 days the bands separate, allowing their individual detection. Finally, in this zone there is a broad band centered at 3200 cm⁻¹ associated with signals of polymerization C–S–H like gels.

Between 1300 and 1800 cm⁻¹ there is a time evolution in spectra. A sharp band at 1400 cm⁻¹ associated with asymmetrical vibration (ν as C–O) is growing, possibly due to sample carbonation—carboaluminate phases type. The bands due to symmetric bending (δ H–O–H) change position; for the initial sample they are at 1650 and 1540 cm⁻¹, which disappear, and a new band form at 1651 cm⁻¹. This is due to loss of water molecules and progressive polymerization of gel [74]. Also, these bands can be associated with hydrotalcite formation [76], allowing association of these two bands with different compounds. The increase in intensity and sharpness of these bands confirms the increase of the reaction products as a function of reaction time. Interesting changes occur in the broad band between 800 and 1300 cm⁻¹, with several shoulders due at different vibrations; this band is typical for aluminosilicates and assigned to the asymmetric and symmetric stretching vibration of Si-O-Si and Si-O-Al bonds in [SiO₄]⁴⁻ and [AlO₄]⁵⁻ [77,78]. Over time the band profile changes and signals appear that can be associated with gel polymerization.

At 28 days, a shoulder at 1180 cm⁻¹ appears associated with Si–O stretching vibrations in Q³ sites [79], indicating that polymerization is formed in three-dimensional chains [80,81]. The main contribution of this band due to asymmetrical vibration (ν as Si–O) shifts from 1000 cm⁻¹ to 970 cm⁻¹; this change in position is caused by the polymerization of silicates producing Q² silica units [82,83]. A small shoulder appears at 820 cm⁻¹ associated with Q¹ units; the lower intensity of this signal indicates the presence of large polymerization chains because these units are found at the ends of the chains [79]. This band is related to chain length, having higher intensity for high Ca/Si ratios, indicating the existence of short chains [84].

The EDX results obtained show a higher SiO₂ content and a low CaO content, resulting in a low Ca/Si ratio. This suggests that long polymeric chains will form; the signal due to the Q¹ units is weak, indicating the formation of these long chains.

The inclusion of Al in the polymerization is evidenced by the bands associated with Al–O modes. The band associated with amorphous aluminosilicate vibration at 1070 cm⁻¹ is replaced by a strong shoulder at 1080 cm⁻¹ due to symmetric bending (δ Al–O–H) [85,86]. This can be explained by the loss of aluminosilicate units due to polymerization produced during reaction with lime. The band at about 898 cm⁻¹ has been associated with the stretching vibration (ν as Al–O–Si) (terminal bond) [87]. The bendings associated with Si–O–Si/Al–O–Al at 740 cm⁻¹ are replaced over time by a band at 780 cm⁻¹ associated with vibrations of the Al–O bond in AlO₄ units [88], indicating the breakdown of aluminosilicates.

The curve-fitting analysis of the 800–1250 cm⁻¹ region showed (Figure 16) the changes produced during the pozzolanic reaction. At the initial time, only three contributions are present in the band, one at 1056 cm⁻¹ due to amorphous Si–O–Si/Al–O–Al (peak 1), which is the main contribution indicating the high presence of aluminosilicates. Together with the contribution at 897 cm⁻¹ due to stretching vibration (ν_as Si–O–Si/Al–O–Al) (peak 3), it confirms the high aluminosilicate content forming more than 90% of the band. The other band at 978 cm⁻¹ due to Si–O (peak 1) bonds is in smaller proportion and does not reach 10% of the total band percentage.

At 3 and 7 days, the band shows four contributions that shift from the initial positions. The signal due to Si–O vibrations grows and changes position to 965 cm⁻¹ at 3 days with a 30.51% contribution and to 973 cm⁻¹ at 7 days with a 44.1% contribution (peak 3). This indicates polymerization and the formation of Q² units, which increase with time. The contribution due to amorphous Si–O–Si/Al–O–Al (peak 2) shifts to 1076 cm⁻¹ at 3 days and to 1099 cm⁻¹ at 7 days. This change indicates that the contribution is due to symmetric bending (δ Al–O–H) involving aluminates in the pozzolanic reaction, forming C–A–S–H gel. A new contribution at 1170 cm⁻¹ (peak 1) appears due to Q³ units, indicating three-dimensional polymerization.

At 28 days, a new contribution at 844 cm⁻¹ associated with Q¹ units (peak 5) appears in the final chain with a low percentage, less than 2%, confirming the presence of large polymerization chains. The contribution due to Q² units is the largest at 978 cm⁻¹, with more than 50% of the total band. The positions of the other bands remain, but their contributions decrease due to the high degree of polymerization obtained during the pozzolanic reaction. This fitting confirms that a C–A–S–H gel is obtained, demonstrating the pozzolanic activity of the volcanic ash.

XRD and FTIR results show a clear production of pozzolanic reaction products. Crystalline phases such as hydrocalumite and hydrotalcite were detected, which are characteristic of the reaction between aluminosilicates and portlandite in pozzolanic materials. Several silicate polymerization units were developed, indicating the formation of large and tridimensional chains. Also, aluminums were introduced inside these chains leading to the formation of C-A-S-H gel. The formation of C-A-S-H gel due to the reaction between volcanic ash and lime is the typical pozzolanic reaction that takes place [reaction 3]. Therefore, Tajogaite volcanic ash can be considered a suitable material for use in blended cements, especially in systems where calcium hydroxide is produced. The volcanic ash can fix this calcium hydroxide, producing hydration products with cementitious properties. Their reactivity does not require additional alkaline activation as they can react with lime directly, which is an advantage as it avoids the use of alkaline agents.

4. Conclusions

This study provides a comprehensive evaluation of the pozzolanic potential of volcanic ash derived from the 2021 Tajogaite eruption (La Palma, Canary Islands), with implications for its sustainable application in construction materials. The ashes are rich in aluminosilicates, primarily present in an amorphous phase, with augite as the dominant crystalline component. Elemental analysis revealed a high Si/Al ratio, placing the material compositionally between fly ash and natural pozzolans such as metakaolin. These characteristics indicate a strong predisposition toward pozzolanic reactivity.

The performance of the material in pozzolanic activity tests was consistent and favorable. In lime consumption assays, nearly 80% of the calcium hydroxide was fixed after 28 days, while the Frattini test results placed the volcanic ash in the same range as high-reactivity pozzolans like silica fume and metakaolin. Reaction products such as hydrocalumite, hydrotalcite, and C-A-S-H gels were identified, confirming the occurrence of typical pozzolanic processes. No significant differences in composition or reactivity were found between samples collected from different locations, indicating that the geographic origin within the affected area does not critically influence performance.

Importantly, this work reinforces the potential of Tajogaite volcanic ash as a sustainable supplementary cementitious material. Its use can partially replace Portland cement or lime in mortar and concrete applications, thereby reducing CO₂ emissions and lowering the environmental footprint of construction activities. These findings align with circular economy principles, promoting the valorization of a locally abundant volcanic residue. The demonstrated reactivity and compatibility with standardized pozzolanic tests provide a solid basis for further development of low-carbon, eco-efficient building materials using this volcanic by-product.