Physico-Chemical and Mineralogical Characterizations of Two Togolese Clays for Geopolymer Synthesis

Geopolymers are an alternative to Portland cement, well known for their contribution to greenhouse gas emissions. Finding materials that can validly replace Portland cement is a challenge. It is in this logic that this work was undertaken with the objective of characterizing two local clay resources of Togo as raw materials for geopolymers. The physico-chemical properties of these clays were determined by characterization using X-ray diffraction (XRD), Fourier transform infrared (FTIR), thermogravimetric (TGA) and elemental analysis (ICP-OES). The results show that these clays contain kaolinite and therefore can be used in the formulation of geopolymers. The cha-racterized clays underwent heat treatments transforming the crystalline phases into more reactive amorphous phases and then were activated by an alkaline solution in order to formulate the geopolymer materials. These ela-borated materials were analyzed by Fourier transform infrared to identify the types of bonds formed. The results of these analyses show that these two local clays are well suited to be used in synthesizing geopolymers. Our future work will focus on the constraints of consolidation as well as the mechanical properties of these geopolymer materials.

important use of clay in Togo remains traditional pottery, yet the country has sufficient clay reserves to supply a ceramic industry [9].
The present study aims at valorizing local clays in ecological materials. More specifically, it will involve carrying out a physicochemical and mineralogical characterization of two clay resources and their use in geopolymer synthesis.

Clay Samples
The clays come from two different sites. The first clay, called clay A comes from a site located in southern Togo at Afagnan (6˚29'39"N; 1˚37'56"E) and the second clay called clay B from a site in northern Togo at Bandjeli (9˚25'05"N; 0˚37'12"E). The samples were dried at 105˚C for 24 hours then ground to 75 µm and stored for analysis.

Clay Characterization
Physico-chemical and mineralogical characterizations were carried out on the clay samples. Structural characterization is carried out by X-ray diffraction (XRD) using the Bruker D8 diffractometer. The range of analysis is from 5˚ to 70˚ and the crystalline phases presented in the clay samples were identified with QualX software version 2.24 using the COD 1906 INO database.
Fourier transformed infrared (FT-IR) was performed using Bruker VERTEX K. M. Anove et al.
70 spectrometer. The samples were finely ground, mixed with 95% KBr by weight and pressed to obtain transparent pellets in the infrared. The range of absorption bands is between 400 and 4000 cm −1 . Thermal analysis (TGA) of the clay samples was performed by a TG 209 F1 ASC-Netzsch apparatus in a dry environment with a heating rate of 10˚C/min from room temperature to 1000˚C.

Geopolymer Synthesis and Characterization
The mass ratio of the solid material and liquid activator for the geosynthetic reaction is optimized as 1.25. The solid material is calcined clay at 750˚C and the liquid activator is 12 M NaOH solution. Thus, the activator was added to the calcined clay powder according to above mentioned ratio and stirred for 5 min.
The mixture was then cast into cylindrical mold at room temperature to allow the fabrication of the molded samples. The molded samples were allowed to mature at room temperature for 24 hours and cured at 70˚C for 24 hours in an oven to increase their mechanical strength. FTIR and mechanical strength were measured on final samples named geopolymer A (GPA) and geopolymer B (GPB).

X-Rays Diffraction
The mineralogical composition of clay samples is determined by XRD analysis of the bulk clay samples. The results are shown on Figure 1 and Figure 2. The diffractogram of clay A ( Figure 1) shows mainly kaolinite mineral (more than 90%) and a minor phase of anatase (TiO 2 ). The clay B ( Figure 2) is composed of a predominance of quartz and the presence of illite (19%) and kaolinite (6%).

Chemical Composition
The chemical composition of clay samples and their loss on ignition (LOI) are listed in Table 1 [11]. The presence of potassium K 2 Ο in clay B indicates the existence of illite.
The sum of exchangeable bases (CaO, K 2 O, Na 2 O and MgO) in clay B was higher (6.89) than Clay A sample (0.63%), indicating that clay B is richer in fluxing agents than clay A.
The loss on ignition determined on the clays shows a greater loss in clay A (15.7%) than in clay B (6.6%). This difference can be related to the organic matter combustion and to the dehydroxylation of the clay minerals. The loss on ignition is explained by the TGA analysis. This chemical composition is confirmed by the mineralogical composition.

Thermal Analysis
The thermal analysis of the different clays can help to determine the rate of kaolinite from the dehydroxylation reaction and the favorable temperature ranges for the geopolymerization reaction.
The thermogravimetric analysis of the studied clays are shown on Figure 3 and

Formulation of Geopolymers and Infrared Characterization
The FTIR of clay A ( Figure 5) confirmed the presence of kaolinite mineral. The bands detected at 3700 -3625 cm −1 are related to the stretching of hydroxyl groups in the kaolinite [11] [13] [14]. Bands at 1115 and 1000 cm −1 correspond to the vibration of Si-O-Al bonds [15] [16], and the peak at 903 cm −1 to the vi-bration of Al-OH bond [17]. In the FTIR of clay B (Figure 6), in addition to the bands characteristic of kaolinite mineral, peaks at 785 and 702 cm −1 which are attributable to the O-Si-O bonds vibrations of Quartz [13] and peak at 3620 cm −1 attributable to hydroxyl groups of illite are observed [18]. The clay samples calcined at 750˚C (AC, BC) and the formulated geopolymers (GPA, GPB) also were analyzed by FTIR.
The FTIR of clay A calcined AC at 750˚C (Figure 7) showed the disappearance of the two peaks of the hydroxyl group. The characteristic peaks of Si-O-Si bonds initially located between 1000 and 1050 cm −1 seem to transform into a single broad peak located around 1080 cm −1 characteristic of silica sites inducing the formation of metakaolinite [12]. It's also observed, in clay B calcined at 750˚C BC, the disappearance of the Si-O bonds between 950 and 1200 cm −1 . The band located at 798 cm −1 highlights the persistence of quartz after the heat treatment ( Figure 8).    It's also observed peaks at 777 cm −1 and at 738 cm −1 on FTIR of geopolymer GPB which shows the presence of quartz after the synthesis indicating the non-reactivity of quartz.
Furthermore, the peaks at 1440 cm −1 (GPA) and at 1452 cm −1 (GPB) are due to the asymmetric stretching mode of carbonates O-C-O [21]. During hardening, the geopolymers were normally exposed to CO 2 from the atmospheric air, which leads to the formation of carbonate [21] in alkaline medium.

Conclusion
The chemical, mineralogical and thermogravimetric characterization of the clays showed that clay A is mainly constituted of kaolinite and clay B is a mixture of quartz, illite and kaolinite. The presence of kaolinite in these clays makes them raw materials for the synthesis of geopolymers. The synthesis of geopolymers with these clays calcined at 750˚C shows that the studied clays can be transformed into ecological cement. Further study should improve the physico-chemical and mechanical properties of the formulated geopolymers.