Cement based materials are the most widely used in the building industry around the world [1] . However, cement manufacture requires significant energy consumption [2] and consequently produces a large quantity of carbon dioxide (CO₂, a greenhouse gas) [3] . This gas comes essentially from the decarbonation of the limestone (CaCO₃) used to make the clinker that is the main component of cement. To reduce CO₂ emissions during cement production, it would be wise to use alternative cementitious materials to partially replace clinker. By combining clinker with industrial co-products such as silica fume, blast furnace slag and fly ash, it is possible to achieve the technical and environmental performance sought in cementitious materials. However, the availability of these industrial co-products may be limited both globally and locally. The use of calcined clay now appears to be a promising alternative solution, as these raw materials are relatively abundant and available in many parts of the world (particularly Côte d'Ivoire). They are obtained by calcination (at temperatures below 1000˚C), a process that consists of dehydroxylation, without emission of CO₂ [4] .

Several studies carried out on calcined clays have shown that they can act as pozzolans or supplementary cementing materials [5] . It has also been shown that, after calcination, kaolin has the best pozzolanic activity of all clay minerals [6] . Consequently, cements with added calcined kaolin or metakaolin have better mechanical properties and durability [7] [8] .

The present work is devoted to the physico-chemical characterization of two clay raw materials from the Man region (west of Côte d'Ivoire) with a view to their use in low-carbon cements. The aim is to determine the chemical and mineralogical composition of these two clays.

Table 1 shows the results of the chemical analysis of the two raw clay materials.

The results obtained show that these clays are mainly composed of silica, alumina and iron oxide. The high silica and alumina contents indicate that these clays can be classified in the aluminosilicate group [9] . The SiO₂/Al₂O₃ ratio, which is an indicator of the clay mineral content of a sample, is 4.8 and 4.4 respectively for ME1 and ME2 instead of between 1 and 2 for pure kaolinite [10] . This difference in value can be explained by the presence silica in free form in significant quantities in both samples [11] . The iron oxide content is in both clays provides for the presence of oxy-hydroxides, namely goethite (α-FeOOH), lepidocrocite (γ-FeOOH), and/or oxides such as hematite (α-Fe₂O₃) and maghemite (γ-Fe₂O₃) [12] . Low levels of K₂O and Na₂O are also observed. The high alumina content and low alkaline element content indicate the possibility of using these clays as a raw material for making refractory product [13] .

Figure 1 shows the X-ray diffractograms of the ME1 and ME2 clays obtained on a fraction of grain size less than or equal to 100 μm. The main reflections were indexed with kaolinite (Si₂Al₂O₅(OH)₄), quartz (SiO₂), goethite (FeOOH) and hematite (Fe₂O₃) crystallographic phases.

The DTA-TG thermograms of the two samples ME1 and ME2 shown in Figure 2(a) and Figure 2(b) respectively exhibit six thermal phenomena, four of which are endothermic and two exothermic.

The endothermic peak around 96˚C with a mass loss of around 1.8% represents the evaporation of adsorbed hygroscopic water. The departure of this water does not modify the structure of the material [14] .

Around 300˚C, the exothermic peak observed reflects the combustion and/or oxidation of the organic matter present in the samples.

Between 345˚C and 360˚C, another endothermic peak is observed, associated with a mass loss of around 1.15%. This peak corresponds to the transformation of goethite into hematite [15] according to the Equation (2):

$\underset{\text{Goethite}}{2\text{FeOOH}}\Rightarrow \underset{\text{Hematite}}{{\text{Fe}}_2{\text{O}}_3}+{\text{H}}_2\text{O}$ (2)

The intense endothermic peak at around 500˚C with a mass loss of between 3.44% and 3.50% characterizes kaolinite dehydroxylation. Dehydroxylation is due to the departure of hydroxyls from the kaolinite structure in the form of water, leading to the formation of “metakaolin”, which has a disorganized structure [15] . This reaction is expressed according to Equation (3):

$\underset{\text{Kaolinite}}{{\text{Si}}_2{\text{Al}}_2{\text{O}}_5{\left(\text{OH}\right)}_4}\Rightarrow \underset{\text{Metakaolinite}}{{\text{Si}}_2{\text{Al}}_2{\text{O}}_7}+2{\text{H}}_2\text{O}$ (3)

The endothermic peak at around 566˚C corresponds to the allotropic transition of quartz (α→β).

Finally, the exothermic peak without mass loss at around 900˚C reflects the structural reorganization of metakaolinite into a spinel phase and the formation of amorphous silica [16] according to the following Equation (4):

$\underset{\text{Metakaolinite}}{2\left[{\text{Si}}_2{\text{Al}}_2{\text{O}}_7\right]}\Rightarrow \underset{\text{Spinelle}}{{\text{Si}}_3{\text{Al}}_4{\text{O}}_{12}}+\underset{\text{Silica}}{{\text{SiO}}_2}$ (4)

The chemical analysis results combined with those of X-ray diffraction were used to determine the mineralogical composition of the samples. The results are recorded in Table 2 as well as those for specific surface areas.

The results show that both samples are rich in kaolinite and quartz. It can be deduced that quartz is the source of free silica, which is in agreement with the calculation of the SiO₂/Al₂O₃ ratio carried out above. The high quartz content, which favours the use of these raw materials in ecomaterial formulations (bricks, tiles, etc...), also plays the role of degreasing kaolinite [17] . The high kaolinite is an advantage for obtaining metakaolin. Consequently, they could be used in eco-cement.

The specific surface area value obtained for sample ME1 is between 10 and 30 m²/g, which corresponds to the range for the specific surface of kaolinite-type clays [18] , while the ME2 is slightly above the upper limit of this range. This difference can be explained by the slightly higher iron content in ME2 than in ME1 ( Table 1 ). Indeed, according to some authors [19] - [21] , the specific surface area of kaolinitic clays increases with the amount of iron.

Figure 3 shows the infrared spectra of ME1 and ME2 samples. These show broadly the same vibrational bands.

The presence of bands in the region from 3700 to 3600 cm⁻¹ confirms the presence of clay minerals in our two samples. These correspond to the elongation vibrations of kaolinite’s OH bonds [22] . The absorption bands at 1097 and 1032 cm⁻¹ are attributed to the Si-O and Si-O-Si vibrations of kaolinite, respectively [1] . The absorption near 912 cm⁻¹, observed represents to Al-OH stretching vibrations of kaolinite [1] . The bands around 779 and 536 cm⁻¹ on the two spectra are attributed with the Si-O-Al bonds of the kaolinite [1] . The presence of quartz in our samples is indicated by bands around 685 and 469 cm⁻¹ corresponding to the elongation of Si-O-Si bonds. The band at 1610 cm⁻¹ is due to absorbed hygroscopic water [1] .

The particle size distribution curves for the two samples are shown in Figure 4 . These show an order of magnitude in the order of the meshes corresponding to 90%, 50% and 10% of the passage of the particles respectively noted d90, d50 and d10 ( Table 3 ). Analysis of the table shows that our samples are characterized by relatively clean grains, with 90% of the particles passing through having an average diameter of less than 56 microns and 50% having an average diameter of less than 5 microns. Figure 4 also shows the presence of three populations in our samples, centered on 3, 9 and 39 microns.

The scanning electron microscope (SEM) microstructure of our samples is shown in Figure 5 . This shows a porous microstructure and a disordered morphology. The latter is explained by the high quartz content in our samples, which can screen the kaolinite platelets [23] .

Using a material requires knowledge of its physico-chemical properties, we therefore set out to characterize two clay minerals from west Côte d’Ivoire with a view to their use in low-carbon cements. Mineralogical and structural analyses showed that the samples contain kaolinite, quartz, goethite and hematite. The particle size analysis of our samples showed that they were very fine. The specific surface values obtained confirm that our samples belong to the kaolinite family of clay minerals.

At the end of our study, we can say that samples ME1 and ME2 belong to the kaolinite family and, after calcination, could have good pozzolanic activity with a view to substituting clinker in the manufacture of low-carbon cement.

For the continuation of this work, we count: