The organic carbon content was obtained through wet oxidation of soil matter (Walley-Black method) with potassium dichromate in a sulfuric medium.

0.5 g of sample was added to a 250 ml Erlenmeyer flask. 10 mL of 0.5 mol/L K₂Cr₂O₇ solution were added, followed by 20 mL of 98% H₂SO₄, manually shaken for one minute and left to rest for 30 minutes. After this period, 100 mL of deionized water, 10 mL of 85% H₃PO₄, and six drops of 1% diphenylamine were added. Then, it was titrated with a 0.5 mol・L⁻¹ ammoniacal ferrous sulfate solution until the dark blue color changed to green. From the spent volume, the “organic matter” content of the soil sample was calculated, as described by (EMBRAPA, 1997) [25] .

The oxidation of organic matter present in the sample of each soil was performed using sodium hypochlorite (NaOCl), as described by (Lavkulich et al., 1970) [26] , seeking to avoid possible changes in minerals or the formation of calcium oxalate by the treatment with hydrogen peroxide, as reported by (Dowdy, 1994) [27] .

Initially, the clay fraction samples from the four soils (LVdf, NVef, MTf and VGo) were subjected to treatment with 6% sodium hypochlorite pH 9.5 at a ratio of 1:3 and remained at rest for 72 hours, with exchange of supernatant every 24 hours and kept in an oven at 65˚C. Afterwards, the samples were washed with deionized distilled water until electrical conductivity was close to that of water (Lavkulich et al., 1970) [26] . Afterwards, they were dried in an oven at 65˚C for

72 hours. Soil samples with organic content were defined with (CMO) and without organic content (SMO). Afterwards, the samples were sieved in a 0.84 mm mesh and saved for FTIR spectroscopic analysis.

The soil samples referring to the clay fraction were treated with dithionite-citrate-sodium bicarbonate (DCB), according to Mehra and Jackson (1960) [28] .

In 100 mL test tubes, 2.0 g of soil sample referring to the clay fraction, 40 mL of 0.3 mol・L⁻¹ sodium citrate, 5 mL of 1.0 mol・L⁻¹ sodium bicarbonate and 1.0 g were added of powdered sodium dithionite. The mixture was kept in a thermostat bath for 30 minutes at 65˚C. After this period the sample received another 1.0 g portion of sodium dithionite, under periodic agitation using a vortex, at a temperature of 80˚C, for complete dissolution of the iron oxides and hydroxide observed by the gray coloration of the residual solid material. After dissolution, the solid material and extract were separated by centrifugation. The material was washed with 100 mL of water, flocculated and lyophilized for further analysis.

For the FTIR analyses, 1.0 mg of the soil sample of the clay fraction, as already defined, (DCB), (CMO) and (SMO) and 100 mg of dry KBr were used, where the samples were macerated in an agate mortar, to preparation of the tablets. The spectra were obtained with a resolution of 4 cm⁻¹, in the range of 4,000 - 400 cm⁻¹ and a speed of 21 scan/min in a Fourier transform infrared (FTIR) spectrophotometer, Bomem MB 105 model.

To plot the FTIR spectra outside the equipment, the Microcal Origin program was used with data imported from the analysis, where the vibration bands of the connections and functional groups were evaluated, as described by Benites, et al. (1999) [22] and Ahlrichs (1968) [21] .

In this region, absorptions corresponding to the stretching of -OH and C-H bonds are observed.

Absorptions relative to the -OH stretch of organic content are always presented as a broad and intense band, due to the -OH groups that form intra and intermolecular hydrogen bonds, with absorptions in the ranges of 3600 - 3440 and 3550 - 3200 cm⁻¹, respectively.

Absorptions relative to the -OH stretch of the inner surface of kaolinite appear at 3694, 3669, 3653 and 3620 cm⁻¹, while for smectites the structural OH stretch occurs at 3632 and 3627 cm⁻¹.

It can be observed, in general, that the band close to 3630 cm⁻¹ is due to structural OH groups. Another stretch that follows arises from the vibration also of the OH group of the water, a wide band at 3430 cm⁻¹ that can be attributed to the H-O-H that stretches in the system referring to the water released when heating the sample. This band reflects the water adsorbed onto the structure or between layers of the clay mineral.

The 2950 and 2850 cm⁻¹ vibration bands are attributed to stretching of C-H bonds of aliphatic and -OH groups of carboxylic acids, respectively. This observation is evidenced in samples with the presence of organic matter.

The characteristic band observed at 2400 cm⁻¹ is related to CO₂.

The absorption band close to 1630 cm⁻¹ is due to the deformation of the H-O-H bonds related to the water molecules of crystallization and the stretching of -C=C- of aliphatic and aromatic groups in 1615 cm⁻¹.

There is a relevant observation in the coordination of charges of carboxylic origin with metals by covalent bonds which cause the displacement of the resonant frequency of the axial deformation of the C=O bond of the carboxylic group in 1712 cm⁻¹ for a range of 1630 - 1615 cm⁻¹, according to the metallic species involved and the form of coordination, as described by Vinkler, et al. (1976) [34] .

The band that appears in a strong way, close to 1115 cm⁻¹, is due to the longitudinal vibration of the Si-O bonds and the absorption related to the Si-O perpendicular stretch occurs at approximately 1100 cm⁻¹. The band close to 1045 cm⁻¹, which appears strongly, is related to the Si-O-Si system bonds present in phyllosilicate minerals, more precisely related to smectites, resulting from stretching in the Si-O plane. As for kaolinites, this stretching in the Si-O system plane occurs at 1033 and 1011 cm⁻¹.

The bands observed at 920, 885 and 842 cm⁻¹ are due to deformations of the bonds between AlAl-OH, AlFe-OH and Mg-Al-OH, respectively.

The bands close to 885 and 842 cm⁻¹ are of low intensity. Another weak band near 800 cm⁻¹ is attributed to quartz and silica vibrations, as the weak band near 778 cm⁻¹ is attributed to quartz vibrations. The absorption band at 620 cm⁻¹ is due to Al-O and Si-O coupling, out of plane and the bands 524 and 466 cm⁻¹ are due to deformation of the Al-O-Si and Si-O-Si bonds, respectively.

In Figure 2 and Figure 3, the FTIR spectra of soil samples from the toposequence derived from basalt, where the interval between the 450 and 550 cm⁻¹ absorption bands can be observed, which can be attributed to angular deformation or combinations of Si-OM bonds, where (M = Al, Fe and Mg). In the present study, these vibrations are observed at 470 and 528 cm⁻¹.

It is also possible to observe the presence of kaolinite from the 3698, 3624 and 913 cm⁻¹ bands, which are attributed to the OH groups connected to the outer octahedron layer, with well-defined peaks, as also observed by Madejová and Komadel (2001) [35] .

Although there is no band at 1720 cm⁻¹ in the treatment of samples (CMO), referring to the indication of carbonyl stretch of the carboxylic group, it does not mean that there is no organic matter in the respective fractions, rather indicating the low content of organic content present in the samples, which can be confirmed by the presence of aliphatic CH and CH₂ bonds in 2930 cm⁻¹ and by the broadening of the band referring to structural OH, in addition to the stretching of the -C=C- bond, when compared with the same samples without the organic material (SMO) which can be seen in Figure 3.

The band observed at 2400 cm⁻¹ refers to CO₂ released during the heating of the sample from the IR laser, being of greater intensity in the samples (CMO), indicating its origin in the organic matter present in the soil. In the samples (SMO), a slight band signal is observed, which confirms the efficiency of the organic matter removal process, as described by Lavkulich et al. (1970) [26] .

The change in the absorption band from 1725 to 1625 cm⁻¹, characteristic of a carboxyl group, is due to the adsorption of organic matter on goethite and gibbsite, as described by Kaiser et al. (1997) [36] . These binding forms are quite common in tropical soils, being considered the main binding mechanisms in the formation of organomineral complexes (Varadachari et al., 1997) [37] .

The FTIR spectra of Figure 2 and Figure 3 obtained for the clay fraction (DCB) and fraction (CMO) phyllosilicates (SMO), respectively, demonstrate characteristic peaks in common regions, varying in intensity, but with little variation in absorbance frequencies, indicating that the dissolution and separation processes are suitable for evaluating the absorption bands.

There is a large absorption range around a wave number of 3400 cm⁻¹, due to the high concentration of OH groups from gibbsite, montmorillonite, kaolinite and water. Gibbsite can be observed from the absorption bands at 3528 and 3452 cm⁻¹ and weak bands between 3380 and 3398 cm⁻¹, as described by Russell and Fraser (1994) [38] .

Figure 4 shows the FTIR spectra of phyllosilicates, such as montmorillonite-Na (SWy-2), montmorillonite-Ca (SAz-1) and kaolinite (KGa-1b), which

are classified as an international standard due to their physical properties―chemical and composition.

It was also used to analyze the FTIR spectra, the area ratio calculation technique (in mm²) with the aid of the Softwear tool Microcal Origin, the absorption bands and the kaolinite peak (KGa-1b) in 1042 cm as a standard. -1 as a reference, due to its larger area, compared to the montmorillonites SWy-2 and SAz-1.

Area 1 (A1) was called the peak of the interval [3700 - 3550 cm⁻¹] referring to the structural stretching of the -OH groups of the inner surface of kaolinite, area 2 (A2) was the peak of the interval [3500 - 3250 cm⁻¹] referring to the stretching of internal structural -OH groups and water, and area 3 (A3) the peak of the interval [1300 - 840 cm⁻¹] referring to Si-O stretching in the plane and Al-M-type deformations OH (where M can be Al, Fe and Mg) and area 4 (A4) to the standard peak of the silicates, in the evaluation of kaolinite (KGa-1b) with 66.84 mm² and absorption in (1042 cm⁻¹), as shown in Figure 4. The results of the area list are shown in Table 7.

The phyllosilicates of the clay group present an intense absorption band at 1042 cm⁻¹ that characterize them. For montmorillonite, this band should be evident at 1100 cm⁻¹. As also observed by Besson et al. (1987) [39] , this shift to low frequencies is due to the Al³⁺ ions that took the place of the Si atoms in the tetrahedron structure, a process called isomorphic substitution.

In Table 7, it is observed that the band ratios are higher for the LVdf and NVef classes in the treatment (DCB) for the A horizon, with similar values, indicating the presence of a high amount of kaolinite, which corroborates the results obtained by TGA and XRD, which is associated with the high degree of weathering of these classes, as also observed by Peternele et al. (2014) [30] .

For the LVdf class in the treatment (DCB) horizon B, that is, the horizon frequently used in the assessment of the mineralogy of the soil class, there was also

(*) A4 ? Ratio Kaolinite (KGa-1b) ? area = 66.84 mm² (1042 cm⁻¹).

a greater presence of kaolinite, indicating the class as the most developed in the toposequence, that is, compared to the other three grades, this is the most oxidized grade.