Studies sites (Ambatta 1, Ambatta 2 and Siguété) are located in the South-East of N’Djamena Chad ( Figure 1 ).

The material that was the subject of all these analyses is essentially the soil powders taken from the Ambatta 1 (Z1), Ambatta 2 (Z2), and Siguété (Z3) sites, presented in Figure 2 , to which must be added the different experimental devices in the laboratory. Several samples were taken from all the different site areas. This is why Z1, Z2, and Z3 samples are fully representative of the entire study area and take into account all the specific variations relating thereto. Densities (Kg/m³) of Z1, Z2, and Z3 soils are 25.30, 24.52, and 25.1, respectively. The particle size distribution of these same soils is presented as follows: for Z1, 25% of clay, 40% of silt, and 35% of sand; for Z2, 56% of clay, 36% of silt, and 8% of sand and finally for Z3, 48% of clay, 34% of lemon and 18% of sand.

This second part of the study aims to present the manipulation techniques and experimental protocols that made it possible to carry out all the analyses. Z1, Z2, and Z3 samples are representative of the study areas. Indeed, a Z1 sample represents the synthesis of several other samples from the same study locality. This is the same case for Z2 and Z3.

The chemical composition was measured by X-ray fluorescence spectroscopy (XRF). The experiment was carried out using a Bruker S8 Tiger 4 kW spectrometer installed at the Cameroon cement plant (Cimencam of the Lafarge group), where 1 g of each sample was liquefied in 8 g of Na₂B₄O₇₁₀H₂O.

XRD is used here to verify the structural information of the three powder samples studied, Z1, Z2, and Z3.

The analysis is carried out using a Bruker diffractometer whose characteristics are Cu Kα1 anticathode with variation in wavelength at λ = 1.5418 Å, voltage V = 40 kV, and intensity I = 30 mA. The angle is used on 2˚ < 2θ < 70˚ with steps of 0.02˚ and time-frequency in steps of 2 s, in the Bragg-Brentano θ/θ Configuration. The maximum intensities of the models used are both qualitative and semi-quantitative [24] [25] .

Infrared spectroscopy is used to identify bonds formed in mineral species based on vibrational frequencies. Each bond is made up of atoms or groups of atoms with their own vibrational frequencies. The experiment was carried out on a Bruker Alpha-P spectrometer (Thermo Electron) in absorbance mode. Then, the spectra were recorded in the spectral range (250 - 4000) cm⁻¹ with a resolution of 4 cm⁻¹.

TDA/TGA thermal analyses are used to detect and identify molecules by their mass measurement. Its principle lies in the gas phase separation of charged molecules (ions) according to their mass/charge ratio (m/z), then evaluating the heat released during exothermic and endothermic reactions. These analyses are carried out using a SetaramLABevo TG-DSC 1600˚C device, working under an Argon flow. Temperatures vary from room temperature up to 1200˚C in a ramp of 10 to 40˚C∙min⁻¹. DSC data is obtained from the airflow of an Alumina crucible.

SEM is used on these ores to examine their morphology on a microscopic scale using a Philips model XL30 microscope (Liège, Belgium) equipped with electron beams accelerated under high voltage. Our samples are regularly placed on an appropriate support maintained in a vacuum of 5 × 10 cm⁻² Pa.

The XRF chemical analyzes give the quantities of the different major chemical elements of the clay materials studied ( Table 1 ). Observation of these results shows overall that samples studied are mainly made up of SiO₂, Al₂O_3, and Fe₂O₃ and a small proportion of MgO, MnO, CaO, Na₂O, K₂O, and P₂O₅.

Samples from the Ambatta 1 site contain 73.8% of SiO₂ and 8.35% of Al₂O₃. The Fe₂O₃ content is approximately 3.79%. The presence of alkalis and alkaline earth is remarkable: K₂O (3.07%), TiO₂ (0.81%), Na₂O (1.01%), CaO (3.33%) and MgO (0. 47%). Constituents such as manganese Mn₂O₃ at 0.07% and phosphorus P₂O₅ at 0.05%, although present, are negligible. The loss due to fire of these soils is, on average 8.84%. This value is relatively low.

The sample from the Ambatta 2 site also consists of SiO₂ (49.03%), Al₂O₃ (17.34%), and Fe₂O₃ (10.90%). The quantity of silica at 49.03% remains dominant over the other chemical elements on the site. The alkaline earth is similar to those of the first site: K₂O (2.57%), TiO₂ (1.41%), Na₂O (1.01%), CaO (1.38%), and MgO (1.21%) remain present. Other constituents, such as manganese (0.14%) and Phosphorus (0.06%), are insignificant. The loss due to fire is 2.83% and significantly lower than that of Ambatta 2 soils.

Analogous to the two previous sites, the results from the Siguété site present the same constituents with a few easy variations ( Table 1 ). According to these results, we can affirm that the clay materials of these three sites are essentially composed of alumina silicate with an excess of quartz silica. These results are in agreement with the work of Kagonbé et al. [16] , where the soils of Sudano-Sahelian zones with contrasting seasons are mainly composed of silica. The high quantities of alumina (Al₂O₃) observed in Z2 and Z3 samples show that these studied soils are made up of clay minerals such as kaolinite. This permanent presence of kaolinite alongside these minerals highlights this geochemical process called monosialization followed by bissialization [26] .

The relatively high level of Fe₂O₃ comes from hematite identified by XRD and decomposition of clay minerals in the area. Alkalis and alkaline earths are relatively modest in these studied samples. The low values of CaO and MgO can be explained by the low content of carbonate minerals observed in most samples. The potassium and sodium oxide contents observed in the clay feedstock could likely be attributed to illite.

The positioning of the geochemical data of the studied clays in the triangular diagram (SiO₂-Al₂O₃-Fe₂O₃) showed that all samples were localized on the SiO₂-Al₂O₃ axis, toward the SiO₂ pole in line with high SiO₂/Al₂O₃ ratio ( Figure 3 ). The ratio of SiO₂/Al₂O₃ is > 2.8 and ranges from 5.17 to 8.84, indicating excess silica content. The alumina and iron oxide values can be correlated with the presence of clay minerals [27] . The loss on fire is generally between 5.26% and 14.95 %. The highest rate of loss on fire is that of the Ambatta 2 locality. It follows that these soils would have undergone dehydroxylation reactions of clay minerals associated with the combustion of organic matter and the decomposition of the carbonate [28] . They also suggest, for the materials studied, the presence of other melting minerals such as illite and smectite. The search for components capable of ensuring strong bonds must be explored to determine the thermomechanical properties of said materials.

Figure 4 presents Z1, Z2, and Z3 samples diffractograms. These diffractograms show that the mineral procession of the studied soils consists of primary minerals (quartz, feldspars) and secondary minerals (kaolinite, smectites, illites, amphibole). The presence of illite is identified by the basal reflection around 9.55 Å, 10.05. Å, 10.08 Å matched with all these Z1, Z2 and Z3 samples. The amphibole is identifiable by the reflection at 8.52 Å. The kaolinite is found at the reflection at 7.26 Å, 7.08 Å, 7.00 Å ( Figure 3 ). Quartz, which is the majority mineral, is identified by the reflections at 4.25 Å, 3.34 Å, 2.45 Å, 2.13 Å, 1.98 Å, 1.81 Å, 1.67 Å, 1.54 Å, 1.38 Å. The latest element is Feldspars at 4.03 Å, 3.78 Å, 3.55 Å, 3.24 Å. All ores obtained appear entirely at all three sites. These ores are almost the same. In addition, we note the notable presence of amphibole on the Ambatta 2 site. Its presence indicates a sedimentary deposit that has not yet transformed. Another type of ore appears plagioclase. It is a mineral from the feldspar family and tectosilicates with oblique cleavages. Its chemical composition varies from one crystal to another. We have, on the one hand, the albite with the chemical formula NaAlSi₃O₈ and, on the other hand, the anorthite CaAl₂Si₂O₈. Plagioclases are also called calcosodium feldspars. For comparison, the XRD of three samples shows basal reflections in the bumps form ranging from 7.00 to 15.00 Å, indicating the large amount of quartz compared to smectite, illite and kaolinite ( Figure 4 ). The remaining reflection peaks are thin in size, which is characteristic of crystallites. This reflection for all of these samples indicates that the smectite has been characterized as a dioctahedral smectite; kaolinite at 7.26 Å is a mineral phase associated with smectite. In all samples, quartz (3.34 Å) and K-feldspars (2.24 - 4.01 Å) coexist as accessory minerals. We deduce that quartz is the source of free silica in these soils, which is confirmed by the calculation of the SiO₂/Al₂O₃ ratio. This quartz content, with the addition of kaolinite is an asset in the formulation of compressed earth bricks.

The elongated shape of the characteristic peaks of these minerals shows that they are crystallized. The intensity of the quartz peaks proves that they are the majority throughout the mineralogical sphere. The variations observed in the mineralogy of these clay soils can be linked to the climatic conditions of the area, in particular the seasonal contrast, namely the long dry season and the short rainy season and explain their differential erodibility. This seasonal phenomenon is marked by the different phases of deposits observed both in the field. The presence of smectite associated with kaolinite suggests that the main crystallochemical processes involved are bissialitisation and monosiallitisation. Indeed, soils whose mineralogy is impacted by the presence of 2:1 type clay, such as smectite and kaolinite are sensitive to runoff and variable erosion as soon as they are saturated with water. This physical character will have an impact on the mechanical properties of materials from said soils, especially their resilience to the presence of water.

A comparative analysis is carried out with similar work in localities around the Lake Chad basin. Table 2 brings out the quintessence.

Relatively high silica contents specify very plastic sedimentary clays (Reeves et al., 2006), while the relative alumina (Al₂O₃) contents confirm the presence of smectites.

Infrared spectroscopy is used here in addition to XRD. It is an excellent means of detecting organic compounds and mineral structures. The recording interval of our samples goes from 400 to 4000 cm⁻¹ and presents vibrational and deformation bands specific to the minerals found.

The infrared spectrum of the Z1 sample ( Figure 5 ) made it possible to highlight the assignments of the OH groups and those of other related networks. Generally, interval bands 4000 - 3500 cm⁻¹ are those of hydroxyl groups. The elongation vibration bands ranging from 3697 cm⁻¹ to 3649 cm⁻¹, characteristic of clay, are attributed to the inter-layer hydroxyl groups of kaolinite. Their intensities and positions are generally sensitive to the intercalation of organic molecules. The characteristic band at 3620 cm⁻¹ is specifically attributed to external hydroxyls with sheets insensitive to intercalations. The presence of these significant vibrational bands supposes that this kaolinite extracted from this clayey soil has a disordered structure [31] . The weak bands ranging from 1900 to 2300 cm⁻¹ belong to the amine groups, which represent impurities for this work. The band at 1633 cm⁻¹ is specific to the vibration of the OH groups of the hydration water of the soil sample studied; this same band shows the presence of smectite. Between 1000 and 900 cm⁻¹ the angular vibration bands are those of Si-O of Kaolinite. Likewise, the Al-OH and Al-O-Si deformation vibration bands appear from 911.23 cm⁻¹ [32] - [34] . The presence of quartz is identified at 775.82 cm⁻¹. Around 550 to 400 cm⁻¹, the Si-O-M bonds (M designate the metals Al, Mg and Fe located in an octahedral position) are recorded.

Some similarities are recorded between the IR spectra of Z2 and Z1 samples ( Figure 6 ). As for Z1, hydroxyl bands at 3694.43 and 3620.46 cm⁻¹ are characteristic of kaolinite. The weak band at 1630.12 cm⁻¹ materializes the aluminosilicate in the amorphous phase. The vibration bands at 994.19 cm⁻¹ and 909.07 cm⁻¹ are those of Si-O Kaolinite. They also correspond to deformation vibrations of Al-OH-Al and Al-O-Si, which will disappear to give way to broadband with a sharp peak at 994.19 cm⁻¹, belonging to the Si-O-Si bonds of the kaolinite. This modification would be due to the formation of an amorphous structure, as already mentioned during XRD analyses ( Figure 4 ). The weak peaks around 686.60 cm⁻¹ simultaneously justify the presence of quartz and kaolinite in the SiO₄ group. These vibrations correspond to Al-O bonds. The peak band at 744.96 cm⁻¹ is those of the C-O bonds of calcite. Like Z1, in the interval 550 - 400 cm⁻¹, it is the Si-O-M bonds (M designate Al, Mg, and Fe metals located in an octahedral position) that are recorded. Bands at 2170.98 and 2095.12 cm⁻¹ are respectively attributed to the (δ + ν) Al₂-OH and 2νAl₂-OH modes of the aluminous phyllosilicate’s family, including smectite, kaolinite, and muscovite [35] . Although indicated by XRD, quartz and feldspars absorb very little radiation in the visible or near-infrared range, hence their few apparent peaks.

Z1 and Z2 sample bands study allows us to deduce the composition and identification of the peaks observed on Z3 ( Figure 7 ) because of the similar bands observed. This is how the two peaks of the bands at 3695.78 and 3620.14 cm⁻¹ are attributable to the vibrations of the OH groups of the kaolinite already specified previously. The interpretation is identical to that of site Z1 except that here, the area appears more stable. The peaks located at the interval 3300 to 3500 cm⁻¹ specify gibbsite bands in these soils [36] - [38] . The band at 1635.08 cm⁻¹ belongs to the group of aluminosilicates in the amorphous phase. The rest of the bands, ranging from 1000 to 400 cm⁻¹ are specified in samples Z1 and Z2.

We note on all three sites a similarity in the bands corresponding to the minerals observed. The Ambatta 2 site seems closer to Siguété in terms of constituents; this is because of the direct depression of Siguété towards Ambatta 2 before arrival towards Ambatta 1. That is why the Ambatta 1 soil is full of all the deposits of “impurities” characterized by constituents belonging to the amine group and other non-minerals.

The three samples are submitted as taken from the study sites to avoid the amorphization of these ores and the preservation of the material structure. The morphological configuration of Z1, Z2, and Z3 soil samples from the Ambatta 1, Ambatta 2, and Siguété sites, respectively, are presented in Figure 8 . We can give these different structures two interpretation types.

The particles that appear on the surface as shiny sand grains are mostly rounded ( Figure 8 _Z1_Z2_Z3): it is the presence of quartz. The micro textures of the quartz grains on the surface of these samples can exhibit both mechanical and chemical characteristics ( Figure 8 _Z1). The shapes presented in their micro textures have inhomogeneous surfaces with a topography of conchoidal fractures ( Figure 8 _Z2), provided with multiform concavities ( Figure 8 _Z1_Z2). These shapes are adapted to particle rearrangement during the consolidation of the material structure [39] . The superposition of small, more or less wavy clusters in platelets marks the presence of smectite already indicated by the XRD of Z2, while the planar crystalline forms are attributable to kaolinite. The small shiny bulbs observed on all three samples mark the presence of kaolinite on all three sites. The very varied texture with cavities and porous openings, associated with scattered aggregates of circular and somewhat complex shape, is characteristic of the presence of kaolinite particles. This morphology of the kaolinite particles is visible throughout Figure 8 . The presence of crystals stacked on top of each other in clusters with irregular grain sizes is characteristic of illite. This is also noticeable in all three samples.

The presence of oxides obtained by geochemistry is found in SEM images by the multitude of clusters scattered across all three samples with a high content of SiO₂, Al₂O_3, and FeO. These elements are characteristic of crystallized kaolinite because of their irregular morphology [40] . The bright crystals formed in agglomeration illustrate carbonate formations, which are specific to quartz. The superposition of wavy clusters reveals the presence of smectite ( Figure 8 _Z2).

The morphology of the balls observed is generally kaolinite ( Figure 8 _Z1_Z2). The clusters are mostly stacked on top of each other, accompanied by crystals formed on the surface. The irregularity of this granular topography is characteristic of the illite, which is grafted onto all three samples.

Figures 9, 10, and 11 are TGA curves giving the mass losses of Z1, Z2, and Z3 samples. The analysis of these curves leaves similar characteristics. These mass losses recorded on the TGA curves can be presented in three phases. The first difference in height on these curves is attributable to the water loss for each sample. The second drop corresponds to the decomposition of organic matter and hydroxides. The last drop in level materializes mineralogical transformations.

The Z1 Sample presents a mass loss of 1.88%, followed by Z3 with 1.98%, and finally, Z2 for a value of 4.14%. This means that Z1 and Z3 accumulate fewer ambient water molecules than Z2. This mass loss is less than 2% for a temperature oscillating around 250˚C for Z1 and Z3, while for Z2 we are at more than 4% mass loss for a temperature hovering around 200˚C. For the three samples, the second elevations generally show that these materials withstand a large amount of heat. The temperatures range from around 200˚C to 500˚C for Z2 and from 250˚C to 500˚C for Z1 and Z3. The resulting mass losses are 2.83% for Z1, 2.91% for Z3 and 4.95% for Z2. The third difference in height corresponds to the phase where the mass loss is the lowest for the three samples: 1.31% for Z1, 1.46% for Z3 and 2.43% for Z2. It appears that Z1 retains more than 98% of its constituents in the first temperature range located between 0˚C and 250˚C, between 250˚C and 500˚C it retains even more than 95.2% and from 500˚C to 950˚C the TGA curve tends asymptotically while retaining almost 94% of its mineralogical constituents. Z2 retains more than 95% of its constituents at the first temperature range between 0˚C and 200˚C, then between 200˚C and 540˚C it retains more than 90%, and finally, its TGA curve tends asymptotically to more than 88% for a temperature range of up to 950˚C. Sample Z3 has a thermal behavior similar to Z1. Z3 retains more than 93.5% for a temperature reaching 1000˚C.

Each of the phases observed on the three TGA curves involves endo and exothermal transformations on the joint ATD curves ( Figure 9, 10, 11 ). This is how the differential thermal analyses (DTA) present variations in the heat flux for each of the three samples. In Figure 9, 10, and 11 , three endothermic peaks are visible for samples Z1 and Z3 and two for sample Z2. See also exothermic peaks.

The mass loss recorded between 0˚C and 200˚C for the three samples Z1, Z2, and Z3, corresponding to the hygroscopic departure of water within the illite sheets [41] , materialized specifically by the first endothermic peaks on the TDA curves. The second, more perceptible endothermic peaks on these three curves appear between 480˚C and 540˚C. This temperature range corresponds to the hydroxylation of kaolinite and later to its transformation into meta kaolinite. It is also at this temperature range that we witness the release of hydroxide formed during the crosslinking of the bonds made up of oxide networks on the ores responsible for the presence of kaolinite, illite, and quartz. [13] [23] [42] .

The rise in the curve (slight exothermic peak) around 480˚C for Z1, around 490˚C for Z2, and 510˚C for Z3 marks the hydroxylation phase of illite and smectite for these three samples. The second rise, marked by a slight bump around 620˚C for Z1 and 650˚C for Z3 (but non-existent for Z2), characterizes the beginning of the transformations relating to the structural reorganization of oxides of kaolinite, magnesium, illite, and many others.

The third exothermic peak in the form of barely perceptible bumps is observable on all three samples on the DTA curves. The temperatures corresponding to this phase vary from 950˚C to 1000˚C. These bumps mark the beginning of the organic matter and hydroxyl decompositions, the transformation of quartz, and the loss of carbonates. It should be said that all three study areas could withstand high temperatures.

Indeed, it should be noted that before 900˚C, several phenomena are observable. Endothermic peaks recorded at 500˚C for Z1, at 509˚C for Z2, and then 492˚C for Z3 are attributable to the dihydroxylation of the OH groups of kaoline on the one hand and to their transformation into metakaolinite on the other. It should be added that the dihydroxylation temperature of the kaolinite of the three samples is lower than 600˚C, which means that these types of kaolinites are disordered [43] . It could contain stacking faults on their crystallographic axis [44] . In addition, the endothermic peaks around 573˚C for Z1 and 574˚C for Z3 reflect the transformation of quartz α into β. The exothermic phenomena at 945˚C for Z1, 924˚C for Z2, and 955˚C for Z3 are due to the metakaolinite structural reorganization.