Introduction. The development of autoclaved lime-sand binders is characterized by the synthesis of calcium silicate hydrates (C–S–H) serves as a cornerstone of modern environmentally sustainable building materials. These materials are based on the hydrothermal reaction between calcium hydroxide and reactive silica to form crystalline phases, such as tobermorite and xonotlite, which provide the structural matrix with high compressive strength and durability. However, the introduction of impurities, especially aluminum oxide (Al2O3), significantly alters the reaction kinetics and the mineralogical composition of the binder [1–5]. The use of high-alumina tropical clays for the production of modern silicate materials is often limited by the inherent mineralogical complexity of the raw material. Although these clays are rich in aluminum, the presence of impurities and the specific bonding states of silicon and aluminum atoms within the clay matrix can hinder the process of hydrothermal transformation [6, 7]. The high alumina content in raw materials, such as the clay deposits found in the State of Eritrea, exerts a detrimental effect on the hydrothermal synthesis of these binders [8–12]. When the molar ratio of Al2O3/SiO2 exceeds specific threshold values, the formation of well-crystallized tobermorite is suppressed. In this environment, alumina promotes the preferential formation of hydrogarnets, which are generally less effective at providing the interconnected crystalline structure required for high load-bearing capacity. The presence of alumina disrupts the growth centers of the desired calcium silicate phases, leading to the formation of a porous and weaker microstructure. To improve the characteristics of lime-clay binders under hydrothermal conditions (typically from 170 °C to 200 °C at saturated steam pressure), the primary challenge is to shift the chemical equilibrium toward silica-rich phases. Mitigating the negative influence of aluminum oxide mainly involves adjusting the chemical composition of the raw mixture to optimize the Al2O3/SiO2 molar ratio. To achieve this, one can utilize pozzolanic materials, such as volcanic ash, which contain reactive silica and possess a low Al2O3/SiO2 ratio [8, 9]. Volcanic ash, characterized by its glassy, non-crystalline structure, undergoes rapid dissolution in the alkaline environment created by lime [13–16]. This dissolution provides an influx of reactive silica, which compensates for the high alumina content in the clay. Moreover, the pozzolanic reaction consumes the excess Ca(OH)2, which might otherwise contribute to the formation of undesirable phases, effectively "locking" the alumina into a more stable matrix. The structural integrity of silicate brick under hydrothermal conditions depends significantly on the hydrothermal treatment cycle [17–21]. A slow increase in pressure to the maximum level ensures the controlled nucleation of tobermorite. If the temperature rises too rapidly, the swift reaction of aluminum oxide can lead to the formation of amorphous, porous phases. Maintaining a high molar ratio of CaO/SiO2 (typically in the range of 0.8 to 1.0) is crucial for the formation of a dense crystalline structure of tobermorite. The purpose of this study is to investigate the influence of volcanic ash on the phase formation process in a lime-sand-clay mixture during hydrothermal synthesis. Furthermore, the work is aimed at optimizing the binder composition to achieve high physical and mechanical characteristics, which will allow for the use of high-alumina tropical clays in the production of efficient autoclaved building materials. Materials and Methods. For the research, the following materials from the State of Eritrea were used: clay rock from the vicinity of the village of Tselot in the Maekel administrative zone (Maekel clay), volcanic ash (VA) sampled in the area of Mount Alid, as well as sand sampled in the Debub administrative zone. Lime from JSC 'STROIMATERIALY', Belgorod, was used in the research work. The clay from the Maekel deposit is characterized by a plasticity index of  Ip=13, which classifies it as a moderately plastic material. The rock exhibits a light brown coloration. Chemically, this clay is distinguished by a high aluminum oxide content (Al2O3) of 30.17 wt. % and a moderate silicon dioxide (SiO2) content of 55.49 wt. % (tab.1). Mineralogically, the primary constituents of the rock are quartz and kaolinite. Structurally, the material is defined as a loose composite featuring a matrix microstructure, wherein sandy grains are distributed chaotically within a continuous clayey matrix, lacking direct contact between the grains themselves [8].   Table 1 Chemical composition of the raw materials Content of oxides, wt. % Components SiO2 Al2O3 Fe2O3 TiO2 CaO MgO SO3 P2O5 Na2O K2O Ʃ Clay Maekel 55,49 30,17 8,42 0,57 1,27 1,24 0,05 0,21 0,43 1,88 99,73 VA 68,82 15,49 2,56 0,21 2,22 0,72 0,12 0,05 5,04 4,23 99,73 Sand 74,10 15,42 1,26 0,15 1,18 0,60 – 0,04 3,37 3,70 99,82     The binder was prepared by co-grinding clay, volcanic ash, and lime to a specific surface area of 240–250 m2/kg. The resulting binder was mixed with sand in predetermined proportions, moistened with the required amount of water, and kept in airtight containers to allow for the slaking of the lime. After the lime had slaked, samples were molded from the raw mixtures at a moisture content of 9–10 % under pressure of 20 MPa. The resulting samples were steamed in a laboratory autoclave at a saturated steam pressure of 1MPa according to the following cycle: pressure rise – 1.5 hours, isothermal holding time – 6 hours, and steam pressure release – 1.5 hours. Before testing, the samples were kept for 2 days at room temperature to stabilize the material's structure prior to conducting physical-mechanical tests. X-ray phase analysis of the studied samples was performed on a DRON-2 diffractometer using the powder diffraction method. A NETZSCH STA 449 F3 Jupiter derivatograph was used for differential thermal analysis. The microstructural characteristics of the prepared samples was studied using a TESCAN MIRA 3 LMU scanning electron microscope. Results and discussion. The influence of the Maekel clay content in the composition of a lime-clay binder on the physical and mechanical properties of silicate materials (fig. 1) has been studied.                                                                   a                                                                              b Fig.1. The dependence of compressive strength (a), average density (b) on clay content     Increasing the clay content to 40 wt.% results in a marginal improvement in the compressive strength of the samples (see fig. 1, a). Beyond this threshold, strength declines, likely due to a reduction in average density (see fig. 1, b) and a subsequent loss of packing efficiency. Furthermore, exceeding 40 wt.% clay disrupts the optimal oxide balance—specifically by elevating the Al2O3/SiO2 molar ratio—which leads to the formation of a less durable cementing matrix. To stabilize the microstructure and mitigate strength degradation, the incorporation of corrective additives with lower alumina content is recommended, consistent with standard practices in the manufacturing of autoclaved materials. The primary reason for the less effectiveness of the clay Maekel to improve the compressive strength of the samples with lime-silica binder is attributed to the high molar ratio of Al2O3/SiO2 of the rock, which stands at 0.319 (tab. 2). According to established materials science research, the optimal value for this ratio in silicate-based construction materials should be less than 0.2 [8]. An excess of aluminum oxide hinders the crystallization of high-strength calcium hydrosilicate phases, instead promoting the formation of less stable aluminosilicate structures. To rectify this, it is necessary to introduce additional siliceous components with low aluminum content, which helps balance the reaction environment and facilitates the formation of a dense, durable stone microstructure. A promising direction for solving this problem is the use of volcanic ash, which has a molar ratio of Al2O3/SiO2 at the level of 0.133. The use of volcanic ash in a lime-clay system allows for an effective reduction in the average value of the molar ratio, contributing to the development of a dense binder structure under hydrothermal synthesis conditions. Thus, the creation of a complex binder based on a lime-clay mixture with the addition of volcanic ash is a scientifically grounded method for improving the operational characteristics of silicate materials. Table 2 The molar ratio of Al2O3/SiO2 and the sum of moles of fluxes ΣR2O + RO + Fe2O3 of the rocks Components Al2O3 (mole) SiO2 Σ R2O+RO+ Fe2O3 (mole) Clay Maekel 0,319 0,134 VA 0,133 0,20   To select rational compositions of a raw material mixture based on clay from the Maekel deposit and volcanic ash, the method of orthogonal central composite design of experiment (OCCDE) was applied. Within the framework of this study, the factors: clay content (x1), volcanic ash (x2), and lime (x3) were varied at three levels: basic, lower, and upper. The step of factor variation was denoted as Δxi (tab. 3).   Table 3 Experiment planning conditions Factors Levels of variation The range of variation –1 0 +1 Content of clay Debub, wt. % (х1) 10 25 40 15 Content of VA, wt. % (х2) 5 15 25 10 Content of CaOact. (х3) 6 8 10 2     Regression equations have been obtained that quantitatively describe the influence of the content of Maekel clay, volcanic ash, and active lime on the compressive strength (RC), bulk density (ρ), water absorption (ω), and the softening coefficient (KS) of silicate materials.     RC = 23,62 + 0,296x1 + 2,895x2 + 0,286x3 – 0,518x1x2 + 0,605x1x3                               (1) ρ = 1855,3 – 44,5x1 – 23,5x2 – 38x3 – 14,3x12 – 14,3x22 –16,8x32 – 28,75x1x2 – 5,0x1x3 + 1,25x2x3       (2) ω = 10,29 + 0,855x1 – 0,581x2 + 0,681x3 + 0,101x12 + 0,281x22 + 0,40x32 + 0,566x1x2 – 0,261x1x3 + 0,34x2x3 (3) KS = 0,818 – 0,0086x1 – 0,033x2 + 0,035x3 – 0,079x12 + 0,07x22 +0,0002x32 – 0,029x1x2 – 0,004x1x3 – 0,04x2x3 (4)     Using the obtained equations, calculations were performed and graphs were constructed that clearly demonstrate the functional dependences of the physico-mechanical characteristics on the variation in the composition of the initial components (fig. 2). Regression analysis (Equation 1) shows that volcanic ash is the most significant positive factor influencing strength, due to its pozzolanic activity. Reactive amorphous silica in the composition of volcanic ash reacts with lime, forming high-strength cementing binders. Clay and lime also contribute to the matrix; however, their effectiveness directly depends on the presence of reactive silica supplied by volcanic ash. The synergistic effect between clay and lime, expressed through the interaction 0.605x1x3, significantly improves the structural integrity of the material. The optimal content of active calcium oxide for achieving maximum compressive strength is 8 wt.%, while a further increase in the proportion of volcanic ash up to 25 wt.% allows for an increase in the material's strength by 1.3–1.5 times compared to the values at 5 wt.% volcanic ash content (see fig. 2a). Compressive strength ranges from 23.32 to 24.74 MPa when using 8 wt.% active CaO, 10–25 wt. % of clay, and 15–25 wt. % volcanic ash. The regression equation for the average density (Equation 2) shows that all primary linear coefficients are negative, which indicates that an increase in the proportion of any of these three components—within the studied range—tends to decrease the overall density of the resulting material. Negative quadratic terms (−14.3x12, −14.3x22, and −16.8x32) further suggest that the density reduction is non-linear and accelerates as the content of these components moves away from the central point of the experimental design. Moreover, a significant negative interaction term between clay and volcanic ash (−28.75x1x2) highlights a synergistic effect, where the combination of these two materials leads to a more pronounced reduction in density than could be predicted based on their individual contributions. Such behavior is characteristic of silicate systems, where the formation of calcium silicate hydrates (C-S-H) and the specific packing density of the matrix are altered by the pozzolanic activity of volcanic ash and the structural role of clay minerals during the hydrothermal synthesis process. The average density is in the range of 1817 to 1885 kg/m³ with a content of 8% active CaO, 10–25 wt.% clay, and 15–25 wt.% volcanic ash (see fig. 2, b).   Fig. 2. Dependence of the physical and mechanical properties of autoclaved silicate materials on the component composition: a – compressive strength, b – average density, c – water absorption, d – softening coefficient.     The regression equation (Equation 3) showed that the content of Maekel clay is the primary factor in increasing water absorption, likely due to the inherent porosity and hydrophilicity of clay minerals when processed under autoclave conditions. Conversely, the inclusion of volcanic ash contributes to a reduction in water absorption, suggesting that the products of the pozzolanic reaction effectively fill capillary voids, thereby densifying the matrix. Lime positively influenced the absorption index, reflecting its role in the formation of calcium silicate hydrates, which, while providing structural strength, can affect pore structure and connectivity in the hardened material. Interaction conditions, particularly the positive coefficient for x1x2, indicate that the combined effect of clay and volcanic ash can lead to non-linear changes in porosity, which requires precise dosing to optimize material durability and its resistance to moisture penetration. The range of water absorption for samples with 8% active CaO and within the range of 10 to 25 wt.% clay and 15 to 25 wt.% volcanic ash is between 8.7% and 10.3% (see Fig. 2, c). In the range of 25 wt.% clay, 5 wt.% volcanic ash, and 10 wt.% lime, the material demonstrated high water resistance. As the composition shifts toward a higher content of volcanic ash, the interaction term −0.04x2x3 (Equation 4) and the negative influence of the volcanic ash coefficient significantly reduce the softening coefficient; however, its value exceeds the established threshold (>0.8) (see fig. 2d). The apparent contradiction between the increase in compressive strength and the simultaneous decrease in water resistance can be explained by the evolution of the pore structure and the chemical nature of the hydration products at a higher volcanic ash content. Thus, the rational composition corresponds to a content of active CaO of 8 wt.%, Maekel clay of 10–25 wt.%, and volcanic ash of 15–25 wt.%. Such a combination ensures that the pozzolanic activity of the volcanic ash is sufficient to bind lime into stable calcium silicate hydrates (C-S-H), while the clay content is limited to prevent excessive water absorption, thereby maintaining high mechanical strength and durability of the materials. In a mixture containing 10 wt.% clay, 5 wt.% volcanic ash, and 10 wt.% active CaO, X-ray diffraction (XRD) analysis confirms the successful formation of tobermorite (Ca5Si6O16(OH)2.4H2O, with a characteristic basal spacing of 11.65Å), which serves as the primary binder providing structural integrity to the autoclaved product (see fig.3a). The presence of CSH(B) (detected at 3.040Å) indicates an intermediate stage in the development of calcium silicate hydrates, while the identification of hydrogarnet (at 2.738Å) demonstrates the incorporation of alumina from the clay into the binder matrix. Residual minerals identified include quartz (4.258Å and 3.345 Å) and feldspar (3.241 Å), the latter originating from the volcanic ash component. Furthermore, the detection of Ca(OH)2 (at 2.631Å) suggests that a portion of the lime remains unreacted or has undergone partial carbonation, a factor that can influence the long-term durability of the material matrix by altering its pore structure.     Fig. 3. X-ray diffraction patterns of autoclaved silicate materials: a – Maekel clay 10 wt. %, volcanic ash 5 wt. %, CaOact 10 %; b – Maekel clay 40 wt. %, volcanic ash 25 wt. %, CaOact 10 %.   In the second formulation, characterized by a content of 40 wt.% clay, 25 wt.% volcanic ash, and 10% active CaO, the increased proportion of aluminosilicate precursors significantly alters the mineralogical balance (fig.3b). The tobermorite peak at 11.191 Å indicates a slightly different structural spacing compared to the first sample, which is likely associated with a higher level of aluminum substitution in the tobermorite chains. Notably, the retention of kaolinite (7.242 Å) suggests that the hydrothermal conditions were insufficient for the complete decomposition of the increased clay content, resulting in a significant portion of unreacted mineral phase remaining. The presence of hydrogarnet (2.765 Å) is more pronounced here, reflecting the increased availability of aluminum oxide. Furthermore, the detection of Ca(OH)2 (2.765 Å) alongside quartz and feldspar indicates that, although the system is chemically active, the high clay-to-lime ratio limits the efficiency of the pozzolanic reaction. Carbonation remains a critical factor, as the high specific surface area of unreacted clay minerals can facilitate the diffusion of atmospheric CO2, potentially leading to the formation of calcium carbonates, which can either densify or degrade the matrix depending on the specific autoclaving parameters. Differential thermal analysis (DTA) showed that in the presented composition, which includes 10 wt. % clay, 5 wt. % volcanic ash, and 10 wt. % active calcium oxide, a complex sequence of endothermic and exothermic effects is observed, reflecting the dehydration and structural rearrangement of calcium silicate hydrates (CSH). The endothermic peak at 77.9 °C is associated with the removal of adsorbed water, while the peak at 440.7 °C indicates the dehydroxylation of residual clay minerals or the decomposition of low-crystallinity CSH phases (see fig.4a). The effect at 575 °C corresponds to the α-β inversion of quartz, which confirms the presence of unreacted quartz aggregate. The high-temperature endothermic effect at 698 °C is most likely due to the decomposition of calcium carbonate (carbonation products) or the breakdown of CSH(B) phases, while the exothermic peak at 883.6 °C indicates the recrystallization of amorphous silicate phases into wollastonite or the formation of new anhydrous crystalline phases. The second composition (40 wt.% clay, 25 wt.% volcanic ash, 10% active CaO) demonstrated a thermal profile, reflecting a high degree of pozzolanic reactivity and the formation of complex hydration products. The endothermic peak at 322.6 °C indicates the dehydration of hydrogarnet or aluminum-containing tobermorite, which are frequently encountered in lime-activated systems with high clay content (fig. 4b). On the DTA curve of materials with low clay and volcanic ash content, an endothermic effect of Ca(OH)2 dehydration is recorded at 440.7 °C, which completely disappears when the content of these components is increased to 40 and 25 wt.%, respectively, indicating complete lime binding during the autoclave treatment process. The peak at 574.7 °C again marks the quartz inversion, although its intensity varies because a higher clay content consumes more silica. The endothermic peak at 674.7 °C suggests the decomposition of calcium silicate hydrate phases, in particular, the loss of structural water from structures similar to tobermorite. Finally, the endothermic peak at 930.7 °C indicates the final destruction of the silicate framework and the subsequent formation of high-temperature phases, such as gehlenite or anorthite, which is facilitated by the significant alumina content provided by clay and volcanic ash. Scanning electron microscopy (SEM) reveald the presence of tobermorite in the form of plate-like crystals that radiate outward from a central nucleation point, eventually aggregating into complex radial clusters (see fig.5a). This specific growth pattern is particularly pronounced in environments containing aluminum, as aluminum ions substitute for silicon within the bridging tetrahedra of the silicate chains, significantly influencing the mineral's structural development and habit. Moreover, the intermediate stages of this mineralogical progression which are characterized by semi-crystalline phases exhibiting a unique, 'blooming cabbage-like' morphology, widely attributed to the presence of calcium silicate hydrate (C-S-H) are observed. This gel-like structure serves as a precursor to the more ordered tobermorite lattice, representing a critical phase in the hydration kinetics of clay- and volcanic ash-based cementing systems. Scanning electron microscopy in backscattered electron (BSE) mode further revealed the presence of dense, spherical formations, often referred to as "globules" (fig.5b). These globules typically represent localized regions of high-density C-S-H gel or incipient nucleation sites where the concentration of calcium and silica has reached a critical supersaturation threshold, leading to rapid, isotropic growth before the transition to the characteristic plate-like, radial clusters of mature tobermorite.     Fig. 4. Differential thermal analysis of autoclaved silicate materials: a – Maekel clay 10 wt. %, volcanic ash 5 wt. %, active CaO 10 %; b – Maekel clay 40 wt. %, volcanic ash 25 wt. %, active CaO 10 %   Fig. 5. SEM images of samples containing  clay Maekel 40 wt. %, volcanic ash 25 wt. % and active CaO 10 wt.%: a) – ×50000; b) ×20000   Conclusion. High-alumina clay Maekel does not have a positive effect on the physical and mechanical properties of lime-sand materials due to a high molar ratio of Al2O3/SiO2 equal to 0.319, which prevents the preferential synthesis of calcium silicate hydrates that provide high strength. It has been established that the inclusion of volcanic ash in a lime-clay binder significantly improves the physical and mechanical properties of silicate materials. This synergistic effect is associated with the formation of a more complex matrix of cementing material, in which low-basicity calcium silicate hydrates and aluminum-containing tobermorite predominate. Due to the high alumina content in the raw material, hydrogarnets are also formed. Such a composition of new formations ensures a rational microstructure of composite materials based on high-alumina clays and, accordingly, high strength properties. It has been demonstrated that the rational composition of the raw material mixture includes 8 wt.% active CaO, 10–25 wt.% Maekel clay, and 15–25 wt.% volcanic ash. At this optimal composition, the compressive strength reaches 23.32–24.74 MPa, which is 1.3–1.5 times higher than that of samples containing only 5 wt.% volcanic ash. Furthermore, the average density reaches 1817.5–1885.5 kg/m3, water absorption remains within the range of 8.7–10.3%, and the softening coefficient exceeds 0.8, indicating the water resistance of the resulting material. In the production of wall materials based on high-alumina clay rocks in the State of Eritrea, priority should be given to the use of volcanic ash as a corrective additive to stabilize the binder matrix and optimize the physical-mechanical properties of autoclaved silicate materials.