The kaolinitic clays come from two villages in the Ketou region. These are the villages of Adakplamè and Adjozoumè. They are sampled and ground in a corn mill. The material passing through the 80-micron sieve is recovered for testing.

The cement used for our tests comes from the company “Nouvelle Cimenterie du Benin” SA (NOCIBE SA) and is of type CEM II/B-LL 32.5R with a specific density of 3.2 t/m³ and complies with standard NF EN 197-1.

The sand used is locally available natural sand. Its characteristics are presented in Table 1 .

These characteristics show that the sand has a spread granulometry as Cu = 4.07 > 2. Its granularity is tight as Cc = 1.04. The sand equivalent value shows that the sand used is a clean sand with a low percentage of clay fines. It can therefore be used to produce good quality mortar or concrete.

The two kaolinitic clays are ground in a corn mill and the 80-micron passing are collected for calcination. An initial cycle of manual grinding was carried out to gradually break down the material. After each grinding cycle, the product obtained was sieved through a 1mm mesh sieve. The fractions that passed through the 1mm sieve were then subjected to a second, more intensive grinding using a mechanical mill of the “corn mill” type. This grinding was also carried out cyclically. After each pass through the mill, the ground product was sieved using an 80μm mesh sieve. The particles passing through this fine sieve were retained as the final product of this stage. This processing step is necessary as fineness plays an important role in reactivity. The passing are then calcined in a laboratory furnace. The furnace is programmed to a ramp rate of 10˚C·min⁻¹ up to 700˚C. The temperature rise is monitored by the furnace’s built-in PID controller. The ramp time to reach 700˚C is approximately 70 minutes. Once 700˚C is reached, the temperature is maintained constant for 60 minutes (soak time) to complete dehydroxylation and ensure homogeneous transformation throughout the entire powder volume. At the end of the soak, the furnace is shut down and the sample cools naturally in situ, with the door closed, to avoid thermal shocks that could induce micro-cracks in the agglomerates formed or alter the texture. Once the temperature has dropped below ≈100˚C, the crucibles are removed and the sample is left at room temperature until it has cooled completely. A 2-theta X-ray diffraction (XRD) test is used to analyze the mineralogy and principal oxides of the calcined clays.

Three types of cement mortar were formulated: control mortars with no cement replacement and mortars with 20% and 25% cement replaced by the metakaolin produced. The ratio of water to cement (W/C) used follows the recommendation of ASTM C311. A ratio sand to cement (S/C) of 2.75 is used.

Test specimens measuring 4 × 4 × 16 cm³ are produced after mixing sand, cement, water, and/or metakaolin in a mortar mixer. The molds are then placed in the impact table and vibrated in accordance with ASTM C311. After demolding, the test specimens are stored in water until the crushing date. The 4 × 4 × 16 test specimens are first crushed in three-point bending, and the half-cubes are then crushed in compression ( Photo 1 ).

The strength activity index is calculated in accordance with ASTM C618 as the ratio of the 28-day compressive strength of the specimen with 20% and 25% cement replacement by MK to that of the control specimen. When this index cumulatively exceeds 75% (in the case of 20% replacement) and the main oxide content exceeds 70%, the replacement material is pozzolanic [18] .

Chemical analysis was used to estimate the proportion of main oxides in the two metakaolins. The results are presented in the following Table 2 . This chemical analysis reveals that metakaolin from Adjozoumè and Adakplamè respectively have a main oxide content (silica, alumina, iron oxide) of 94.5% and 95%, far exceeding the minimum required value. This means they fully comply with the first requirement of ASTM C618. These rates are higher than those of metakaolins produced by other researchers, [19] - [21] .

Compressive strengths are measured on specimens at various times. The results obtained for control specimens (MT), specimens containing adjozoumè metakaolin (MK1) and those containing adakplamè metakaolin (MK2) are shown in the graphs in Figure 5 and Figure 6 .

The results show a continuous increase in compressive strength for all specimens. This demonstrates a pozzolanic reaction between Portland cement and metakaolin [22] . The evolution of compressive strength shows a good consumption of Portlandite to form additional silicate phases. The calcination temperature as well as the fineness due to the sieving with an 80 micron sieve contributed to the development of the pozzolanic reaction as indicated in the works of Geu [23] and Morh [24] .

The activity indices per strength of the metakaolins produced from Adjozoumè and Adakplamè clays are therefore 97.5% and 96.26% respectively. These values are well above those required by ASTM C618. These values are also higher than the values obtained for metakaolins studied in the work of other researchers [5] [20] [21] .

The drop in resistance when going from a substitution rate of 20% to 25% could be explained by a deficit of portlandite resulting from the hydration of the cement.