1.Introduction

Clays form a group of minerals with different characteristics and multiple technological applications. It is used for the elaboration of traditional ceramics and new materials useful today, too ^1,2. Its high plasticity is because water forms an envelope over the lamellar particles, producing a lubricating effect that facilitates the sliding of some particles over others when a strain is exerted on it. Clay minerals, when subjected to heating processes, undergo a series of transformations from dehydration to the appearance of a glassy phase.

The present work aims to analyze the structural changes, qualitatively and quantitatively, produced in clays because of temperature ^3-5. The structural evolution of the phases in each sample, a consequence of the heat treatment, was accompanied by measurements of X-ray diffraction (XRD) after each treatment ^6,7. The qualitative and quantitative analysis of the XRD measurements was performed by applying analytical methods and by structural refinement using the Rietveld method and the TOPAS program, which is particularly useful for the analysis of complex structures as in this case ^8,10.

2.Theoretical framework

3.Materials and methods

The natural sample was collected in the Cerro de Pasco region, in a place usually explored by local artisans. This sample was prepared, pulverized, and separated into 15 sub-samples to treat them thermally at different temperatures. The thermal treatment was carried out by using a three-stage oven, Model PT 1600.

The samples were treated at temperatures from 150 to 800 at intervals of 50. The treatment time was 2 hours, with a rate of 50/min., in an air environment. Individual samples were used for each temperature and coded according to heat treatment, e.g., M-ST (sample without heat treatment); M-150 (sample at 150), M-200 (sample at 200), etc. In total, there were 15 samples investigated: the sample without heat treatment (used as a reference for analysis) and the other 14 samples treated thermally.

The XRD measurements were performed by using a Bruker diffractometer model D8 Focus in θ-2θ geometry. This diffractometer operated with Cu (Kα) radiation, 0.1 mm receiver slit, 1-mm divergence slit, 2.5 soller slit, a detector of scintillation, a voltage of 40 kV, and current intensity of 40 mA. The angle range was from 4 to 65 with an angular step of 0.02 and 4 s per step.

4.Results and discussion

The XRD measurements were performed before and after each heat treatment. Figure 1 and Table I show the results of the phase identification for the sample without heat treatment (M-ST), obtained by computational methods utilizing the PDF (Powder Diffraction File) database. This identification was confirmed by structural refinement applying the Rietveld method.

Figure 1 X-ray diffractogram showing the phases identified in the sample without heat treatment (M-ST). 

Figures 2a and 2b show the enlarged sections of the original diffractograms corresponding to the samples treated up to 300 and 500, respectively. Thus, Figure 2a shows the changes in the intensity of the (001) peak of Ca-montmorillonite, at 2θ ≈ 6.12, as temperature increases. This intensity decreases until it disappears at 250, while the intensity of the peak at 2θ ≈ 8.9 increases, which could be due to the contribution of the (001) peak of Ca-montmorillonite in its new angular position. The change in position is due to the structural change of this phase, produced by its dehydration during the heat treatment, by losing water molecules lodged in the interlaminar region, causing a decrease in the interlaminar distance, which is expected in this clay due to its expansion and contraction properties ¹⁴. These premises were considered in the analysis by structural refinement applying the Rietveld method.

Figure 2 a) Decrease in intensity of the Ca-montmorillonite (001) peak as temperature increases. b) Decrease in intensity of the kaolinite (001) peak caused by the increase in temperature. 

Figure 2b shows the gradual decrease in the intensity of the kaolinite (001) peak at 2θ ≈ 12.43, as the temperature increases, until it collapses between 400^oC and 500 ^oC ^5,15. This phase loses water lodged in its structure as the temperature increases, transforming into metakaolinite through dehydroxylation.

Figure 3 shows the results of the Rietveld refinement for samples before and after heat treatment. The former corresponds to M-ST and the latter to M-150, M-250, M-450, and M-800. The respective diffractograms were selected because they show the important mineral changes that occurred due to heating.

Figure 3 The Rietveld refinement corresponding to the samples before and after heat treatment at temperatures from 150^oC to 800^o C. 

Table II presents the weight percentages of the phases identified in each sample, obtained by structural refinement. There is a high percentage of quartz (> 44%) in all samples.

The Ca-montmorillonite structural refinement was carried out by considering its expansive nature. The change in the angular position of the Ca-montmorillonite (001) peak would be due to a decrease in the interlaminar distance d associated with the (001) plane, a consequence of the heating, directly affecting the parameter c of the unit cell. This parameter was refined for each sample treated at different temperatures, obtaining the values d shown in Table 3 and Fig 4.

Figure 4 Ca-montmorillonite interlaminar distance as a function of temperature. 

As can be seen in Table III, the distance of 14.67 Å before heat treatment is reduced to around 10.00 Å after heating until 250^oC, showing a decrease of around 4.67 Å due to loss of water molecules, particularly in the interlaminar region. According to the literature, calcium montmorillonites with interlaminar distances around 15 Å, as in this case, have a double layer of water molecules in the interlaminar region ¹⁶. Since the average size of a water molecule is around 2.02 Å, the decrease in the interlaminar distance would be due to the removal of two layers of water, generating the dehydrated phase of Ca-montmorillonite and the modification of its structure. This last fact is evidenced in the 2θ position change of the (001) peak, from ≈ 6.12 to ≈ 8.9. Figure 4 graphically shows this change.

In the literature, there are other results related to the structural behavior of Ca-montmorillonite with increasing temperature. Specifically, these results were obtained in previous studies for clay samples from the Nazca and Chulucanas towns, in Peru ^3-5. Concerning the clays from Nazca , the value of d 001 for Ca-montmorillonite (15.37 Å before the thermal treatment) showed a slight decrease (< 0.6 Å) with treatment up to 200. In the interval from 200 to 250, this decrease was much greater (> 5.0 Å) and the distance was reduced to 10.14 Å. As regards clays from Chulucanas ^4,5, the value of d₀₀₁ for Ca-montmorillonite (14.86 Å before heat treatment) showed a decrease of around 5.0 Å at 250; the distance was reduced to 9.92 Å.

In the case of kaolinite, the behavior is different. Table IV and Fig. 5 show the results of the interlaminar distance, which does not show significant changes with increasing temperature. This behavior is expected because it is a non-expansive clay, in which the union between all its layers is solid enough to prevent the penetration of water molecules, thus avoiding swelling and, consequently, contraction with heating. Furthermore, Table II shows that the structure collapses after the treatment at 450^oC; other studies present similar results ⁴.

Figure 5 Kaolinite interlaminar distance as a function of temperature. 

In the case of illite, Table II shows small variations in the weight percentage in the applied temperature range; the refinement did not report alterations in its structure. Regarding quartz, the results did not present structural changes due to heat treatment up to 800; however, slight increases in its weight percentage are observed with heating (Table II). This could be interpreted as a greater crystallization of this phase with increasing temperature, producing an increase in the intensity of the diffraction peaks.

Concerning the indicators involved in each refinement, Table V shows the results for R_wp (weighted profile residual), R exp (expected profile residual), and S (goodness-of-fit). This last indicator determines the quality of the refinement.

5.Conclusions

In the samples before heat treatment, the analyses revealed the presence of clayey phases such as Ca-montmorillonite (1.14 %), kaolinite (28.40 %), illite (24.09 %), and a large amount of quartz (46.37 %). After the respective heat treatment, structural changes were detected in the samples, particularly in the Ca-montmorillonite expansive clay that presented a change in its interlaminar distance d associated with the (001) plane, from 14.67 Å (room temperature) to 9.99 Å (250^oC), due to the removal of the water molecules contained mainly in the interlaminar region, a characteristic of expansive clays. This observation explains, by observing the corresponding diffractogram, the change in the 2θ angular position of the peak associated with the (001) plane of Ca-montmorillonite (from ≈ 6.12 to ≈ 8.9). In the cases of kaolinite and illite, these structural changes were not observed.

Furthermore, the gradual decrease of the kaolinite weight percentage could be observed, interpreting this result as the process towards the collapse of its structure between 450 ^oC and 500 ^oC. Illite presented greater thermal stability with small variations in its weight percentage in the applied temperature range, but without compromising its structure. The quartz phase did not show structural changes with heating, but it had slight increases in its weight percentage with increasing temperature. Consequently, we can conclude that the quartz phase had a greater crystallization.