29Si and 27Al MAS NMR Characterization of the Structural Evolution of a Lateritic Clay under Acidic and Alkaline Treatments

In the present work the modifications induced by the geomimetic processing on the environment of silicon and aluminum atoms are finely characterized by solid state MAS NMR. The raw clay (Lat), a lateritic clay from Yaoundé (Cameroon), contains kaolinite, quartz, hematite and goethite as major mineral phases. This material is treated under acidic conditions during 24 h (LatAF) and then under alkaline conditions during 18 days, leading to the final consolidated “geomimetic” product (LatAFCH). The samples have been characterized by 29Si and 27Al solid state NMR at each step of the process. The NMR spectra obtained for the starting clay indicate the presence of AlIV and AlVI populations together with silicon Q2(0 or 1Al) and Q3(0 or 1Al) environments located at δiso=-83 ppm and δiso=-91 ppm respectively. The acidic reaction during 24 h, does not significantly affect the silicon-rich layers. It induces a conversion of part of AlIV population into AlVI environment mainly related to the occurrence of six-fold organo-aluminum complexes when using fulvic acid. With inorganic acid, Al dissolution prevails. The neutralization (using lime) of the medium, followed by ageing for 18 days, promotes clay interactions with available calcium ions. The newly formed phases are cementitious CSH, CASH and CAH phases. *Corresponding author: Gisèle L Lecomte-Nana, GEMH: Groupe d’Etude des Matériaux Hétérogènes, Centre Européen de la Céramique – GEMH, 12 rue Atlantis, 87068 Limoges Cedex, France, Tel: +33 (0) 587 50 25 59; Fax: +33 (0) 587 50 23 01; E-mail: gisele.lecomte@unilim.fr Received November 22, 2017; Accepted December 26 , 2017; Published January 03, 2018 Citation: Goure-Doubi H, Montouillout V, Lecomte-Nana GL, Naît-Ali B, Koffi Konan L, et al. (2018) Effect of Durable Superhydrophobic FS/PS Using DCTES on Carbon Steel. J Material Sci Eng 7: 409. doi: 10.4172/2169-0022.1000409 Copyright: © 2018 Goure-Doubi H, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.


Introduction
Clays are natural raw materials characterized by a layered structure and therefore belonging to the phyllosilicate family. There are two groups of phyllosilicates depending on the constitution of an elementary sheet resulting from a pilling up of octahedral (Al and/ or Mg cations) and tetrahedral (Si cations) layers along the c-axis [1][2][3][4]: 1:1 or Te-Oc phyllosilicate group that exhibits an octahedral layer combined to a tetrahedral layer, typically as in kaolinite where the characteristic d-spacing of the elementary sheet is 0.714 nm and the 2:1 phyllosilicate group characterized by an elementary sheet composed of one octahedral layer sandwiched between two tetrahedral layers to form an elementary sheet. Due to the occurrence of various cation substitutions within the octahedral and/or tetrahedral layers, the global sheet is often negatively charged and balanced by compensating cations present within the interlayer space [4][5][6][7][8]. The d-spacing is in the range of 0.9 to 1.4 nm in general and typical clay minerals of this group are montmorillonites, vermiculites, talc, illite, mica … These raw materials are readily available and can be easily recycled, they are therefore involved in sustainable and environmentally friendly applications. Clays commonly include, together with clay minerals, various amounts of associated phases or secondary phases such as quartz, feldspar, iron oxides or oxy-hydroxides, carbonates as well as titanium oxides. In tropical regions [9][10][11][12][13][14][15], the major type of associated phase is iron oxide and/or oxy-hydroxide. These associated iron phases are responsible for the yellowish-reddish to dark-brown colors of related clays. In literature, clay is considered as lateritic clay when the amount of iron phases is in the range of 15-25 mass% in association with near to 50% mass clay minerals. Such raw materials are mainly used for road and building purposes. The consolidation of lateritic clay products is generally performed through firing, compaction or stabilization using hydraulic binders (cements, lime). However, a processing including acidic and alkaline consolidating reactions to obtain "geomimetic" materials has recently been proposed [16]. Different characterization (XRD and DTA-TGA) were carried out on these geomimetics materials products to understand the strengthening mechanisms during the key processing steps [6,17,18]. Results clearly indicate the occurrence of new crystalline phases that may result from dissolution-precipitation mechanisms during the 24 h acidic reaction followed by alkaline neutralization and ageing at 60°C for 18 days under saturated water atmosphere. There are still some misunderstands regarding the potential occurrence of amorphous binding phases. The present work is aiming to contribute to clear up the latter point by the investigation of the evolution of the local environment of the main constitutive atoms namely Si and Al using solid state MAS NMR analyses.

Materials and Methods
The raw clay (labeled as Lat) is a lateritic clay from Yaoundé town (Cameroon, Awae deposit, South-east of Yaoundé, 3°49" to 3°51" North latitude and 11°33" to 11°35" East longitude). It is naturally rich in iron compounds. A KGa-2 kaolin (labeled K) from the clay mineralogy society in Georgia is also used in this work. This reference kaolin contains a poorly-crystalline kaolinite as in the raw lateritic clay (similar crystallinity, R2 index).
Additional reagents are fulvic acid (an organic macromolecule having several functional groups Figure 1 [19] purchased from Bois Valor in France and lime (99.9 mass% of Ca(OH) 2 ) provided by Aldrich.

Chemical analysis
The chemical composition of clays was determined using Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES). Prior to analysis, dried sample powders were dissolved using microwaves under acidic (hydrofluoric and nitric acids) and high pressure conditions. A CEM MARS 5 microwave was used and the dissolution was achieved after a 45 min cycle including a maximum temperature and pressure of 180°C and 3 MPa, respectively.

X-ray diffraction (XRD)
X-ray diffraction (XRD) analyses were conducted on powdered samples with a Bruker-AXS D5000 powder diffractometer using K 1 radiation of Cu and a graphite back-monochromator. XRD experiments have been achieved in step-scan mode from 3° to 45° (2θ) with a counting time of 10.1 s per 0.02° step. Crystalline phases were identified by comparison with PDF standards (Powder Diffraction Files) from ICDD (the International Center for Diffraction Data).

Thermal analysis (DTA-TGA)
Differential thermal and thermogravimetric analyses (DTA and TGA) were carried out between 25°C and 1200°C at 5°C/min under dried-air atmosphere using SETSYS Evolution DTA-TGA equipment from SETARAM. The samples were previously crushed and sieved at 100 µm, then oven-dried at 40°C for at least one week or until the achievement of no mass variation. Alumina powder, previously heated at 1500°C for one hour, was used as a reference material. All analyses were performed using 60 mg aliquots of sample and reference.

Scanning electron microscopy (SEM) and Specific surface area measurements
The observation of some samples microstructure have been performed by scanning electron microscopy (SEM) using a Stereoscan 260 apparatus equipped with a PGT Prism energy dispersive spectrometry analyzer. Prior to their observation, the samples were dried at 60°C for one week and metallization was conducted through the sputtering of a nanometric layer of platinum onto the sample surface.
The Brunauer Emmet and Teller method was used to determine the specific surface area of the samples, using TriStar II equipment from Micromeritics. Measurements were carried out after a 16 hours degassing step at 150°C, on dried samples previously crushed and sieved to 100 µm.

Solid state magic angle spinning nuclear magnetic resonance (SSMAS NMR)
High-resolution 29 Si MAS NMR spectra were acquired at room temperature on a Bruker 400 spectrometer operating at a Larmor frequency of 79.5 MHz and using a spinning rate of 10 kHz. Each spectrum necessitated the sum of 10000 to 20000 transients accumulated with a recycling delay of 10 s insuring complete relaxation of the magnetization. 27 Al MAS NMR experiments were conducted on an AVANCE 750 spectrometer operating at 195 MHz with very high MAS conditions (spinning rate of 60 kHz). The spectra were acquired after a single short pulse ( /12) ensuring a quantitative excitation and quantification of 27 Al central transition. The recycling delay was fixed at 1 s.
The chemical shift were referenced relative to TMS and Al(NO 3 ) 3 1 M solution for 29 Si and 27 Al respectively. Quantitative analyses of the 29 Si and 27 Al MAS NMR spectra were performed with the Dmfit program [20].
In the spectra interpretation, we used the Q n notation, commonly used to describe the structure of silicate anions: Q represents a silicon atom bonded to four oxygen atoms, forming a SiO 4 tetrahedron; the superscript n indicates the connectivity, i.e., the number of other Q units attached to the SiO 4 tetrahedron. Q 0 denotes the monomeric orthosilicate anion (SiO 4 ) 4-, Q 1 represents the end groups of chains, while Q 2 , Q 3 , and Q 4 are used respectively for the middle groups of chains or cycles, groups at chain-branching sites, and threedimensionally cross-linked groups. Dealing with alumino-silicates, some of the neighboring Si atoms in the tetrahedral layer could be Al. In this case, the number of AlO 4 tetrahedral bound to the central silicon of a Q n unit is given in parentheses; namely, Q n (mAl) indicates a SiO 4 group connected to mAl and (n -m) other Si atoms via oxygen bridges, where n is in the range 0 to 4 and m ≤ n.
The processing of geomimetic materials ( Figure 2) was described in previous papers [6,18]. Prior to elaboration, the raw clay (Lat sample) was ground and sieved under 100 micrometers. The acidic solutions were prepared by a homogeneous mixing during 5 min of fulvic acid (V AF ) and deionized water (V W ) to obtain a final pH value of 2. This acid solution was noted AF. The Lat sample was then added and mixed with AF for 24 h. Finally, lime, noted CH, was added and the mixture is left to cure at 60°C for 18 days. The solid part, S, consists of 80 mass% of Lat clay (Lat) and 20 mass% of lime (CH). The quantities were calculated in order to obtain a final mass ratio, water over solid, equal to 0.29. After 24 h of reaction with acid, the sample was labeled as LatAF. The final consolidated "geomimetic" material was labeled as LatAFCH.

Mineralogical and structural characterization
The chemical and mineralogical compositions of this lateritic clay (Lat) ( Table 1) indicate the presence of aluminum, silicon, and iron oxides as major constituents. The density and BET surface area of the starting powder are 2.8 g/cm 3 and 48 m 2 /g respectively. The particles are stacked together and appear as aggregates with a mean size of 1 µm [6]. Such high specific surface area may arise from the presence of iron surface of kaolinite platelets, into the metakaolinite network, affecting the structural reorganization [24].

Solid state MAS NMR characterization
To allow a better understanding of the structural evolution of materials after acidic and alkaline treatment, a systematic 27 Al and 29 Si NMR analyses was performed on four series of samples. The first series was composed of the Lat, LatAF and LatAFCH samples studied so far. The second series was based on a KGa-2 reference kaolin that was subjected to the same treatment (KGa-2, KGa-2AF, and KGa-2AFCH samples) and finally the third and fourth series were obtained from the same Lateritic and KGa-2 reference kaolin, treated in the first step using nitric acid (HNO 3 ), the second step remaining unchanged (Lat, LatAN, LatANCH and KGa-2, KGa-2AN, KGa-2ANCH samples). 27 Al and 29 Si MAS NMR spectra are shown on Figures 5 and 6 respectively.
The 27 Al MAS spectrum of the raw clay, Lat sample, exhibits two symmetric signals centered at 70.6 ppm and 5.7 ppm assigned to 4-fold coordinated (Al IV ) and 6-fold coordinated aluminum [25] respectively ( Figure 5A). Despite the quadrupolar nature of 27 Al (I=5/2), both signals appear narrow (FWHM are 1.2 and 1.1 kHz respectively), symmetric and can be unambiguously simulated using Gaussian line shape. Taken into ccount that the spectrum was acquired at quite a high magnetic field (17.6 Tesla), the observed line shapes indicate a relatively symmetric environment for aluminum characterized by a low electric field gradient, and then a very low quadrupolar interaction.
phase nodules onto the surface of clay platelets as well as from the high alteration (natural leaching) of this type of lateritic soil.
The X-ray diffraction patterns obtained for Lat and LatAFCH are shown on Figure 3. Kaolinite, goethite, hematite and quartz are present in both the raw clay (Lat) and the consolidated material (LatAFCH). However, two new phases were detected for LatAFCH: the katoite (Ca 3 Al 2 (SiO 2 )(OH) 8 ), a cementitous phase, i.e calcium aluminum silicate hydrate (noted CASH, where C, A, S and H stand respectively for CaO, Al 2 O 3 , SiO 2 and H 2 O), and the richterite (K 0.9 Na 1.6 Ca 0.5 Fe 5 Si 8 O 22 (OH) 2 ), an hydroxylated phase containing silicon, iron, sodium, potassium and calcium. The presence of katoite suggests that during the consolidation process, siliceous and aluminous ions contained in the lateritic clay would dissolve and re-precipitate in the form of hydrated cementitous phases such as CASH, CAH or amorphous CSH [21]. The formation of richterite appears to be the result of dissolution and precipitation reactions involving iron compounds and clay platelets.
Thermal analysis of the different materials gives complementary information about the nature of the mineral phases present in the starting lateritic clay or in the consolidated clays. The thermal transformations occurring during the heat treatment of the starting lateritic clay and the resulting "geomimetic" materials are shown in Figure 4. These DTA-TGA curves exhibit many differences. In the 100-200°C range, the endothermic peak might be related to the departure of physisorbed water. This peak is associated with a mass loss of 1.05% for the lateritic clay (Lat) and with a mass loss of 2.2% for the consolidated material (LatAFCH). The increase of the mass loss and also the temperature range could be the signature of the thermal decomposition of CSH phases [21].
In the 200-400°C range, while the lateritic clay shows an endothermic phenomenon, the "geomimetic" material presents an exothermic peak. This phenomenon can be due to the organic matter decomposition, to the transformation of ferrihydrite into hematite or to the oxidation of organic matter [22,23].
In the 900-1000°C range, the exothermic peak, characteristic of structural reorganization in the lateritic clay is significantly reduced in the case of the consolidated material (LatAFCH). This observation can be explained by the diffusion of iron, from iron nodules present on the    The deconvolution of the spectrum leads to the estimation of Al IV and Al VI species to 3% and 97% respectively, consistent with the presence of small fraction of a 2:1 clay mineral together with the major 1:1 clay mineral phase (kaolinite) [26][27][28][29][30][31] detected on XRD diagram ( Figure  3A). The 2:1 clay mineral has not been identified by XRD, probably due to a content less than 5 mass%. It should be noted, that the observed 27 Al NMR signals does not seem affected by the presence the high iron content (23.4 mass% of Fe 2 O 3 ), neither signal broadening nor very short relaxation times can be observed. This was explained by the location of iron which is essentially in goethite, hematite and ferrihydrite phases and then by a very low iron substitution in clay mineral phases. However, we cannot completely exclude that a slight part of aluminum signal is unobservable, due to paramagnetic interaction with substituted iron cations.
The 27 Al NMR spectrum ( Figure 5A) of the sample treated under acidic conditions (LatAF) is quite completely equivalent, except a slight increase of the Al VI /Al IV ratio. At this stage, it is difficult to distinguish whether the Al environment has been highly modified. Interestingly, the treatment under alkaline conditions induces more distinguishable differences on aluminum environment. The 27 Al spectrum of LAFCH exhibits additional signals, both in the Al IV and in the Al VI chemical shift ranges. Without going more ahead in details, aluminum is now at 18% four-fold coordinated and 82% six-fold coordinated. This observation, together with the results of XRD and DTA-TGA analyses, translates a strong interaction between alkaline cations and the Al provided by Lat. 29 Si MAS NMR spectra of the raw clay (Lat), LatAF and LatAFCH are shown in Figure 6A. The Lat and LatAF spectra are completely equivalent, composed of two distinguishable narrow peaks centered at -83 ppm (FWHM=200 Hz), and -91 ppm (FWHM=300 Hz). The signal at -83 ppm can be attributed to Q 2 (1Al), and Q 2 (0Al) while the signal at -91 ppm may correspond to Q 3 (1Al) or Q 3 (0Al). These results are in agreement with the presence of kaolinite as expected in lateritic clay. Furthermore, the Q 2 (1Al) and Q 3 (1Al) environments are consistent with the presence of 2:1 clay mineral as also noted regarding the Al IV detected on the 27 Al NMR spectrum ( Figure 5A) of Lat sample. The Q 2 (1Al),environments are characteristic of Si-contained units at the edges of clay minerals. The spectrum of the final "geomimetic" product, LatAFCH, seems also quite similar. However, we can distinguish the presence of an additional narrow signal centered at -86 ppm (FWHM=200Hz) but also a large contribution under the main resonances. The slight signal can be assigned to a Q 2 (1Al) environment of silicon atoms and the large contribution could probably be related to the new Al IV species observed on 27 Al spectrum of LatAFCH. The substitution of aluminum in the silicate layer induce the existence of Qn(mAl) species and then the overlapping of numerous 29 Si signal. This is in agreement with the formation of new binding silicate phases during the strengthening of the "geomimetic" products.
Two additional remarks: first we observed on LatAF and LatAFCH spectra, the presence of a very small signal at -107 ppm. This signal could be related to the quartz phase detected by XRD. Its amount is then certainly under-estimated on NMR analysis due to the spectral acquisition conditions (recycling delay too short to ensure complete relaxation of this specific signal). The second remark concerns the effect of iron presence. As for aluminum, the spectra do not show any paramagnetic effect, however it cannot be excluded that part of the signal is undetectable. 27 Al and 29 Si high resolution MAS NMR spectra obtained on the raw kaolin (KGa-2) and upon subsequent reactions (KGa-2AF and KGa-2AFCH), as described with the Lat sample, are shown in Figures 5B and 6B respectively. 27 Al spectrum exhibits a main signal centered at 5.9 ppm corresponding to six-fold aluminum species, and a slight signal, corresponding to less than 1% of the total intensity, is detected in the chemical range of four-fold aluminum. 29 Si spectrum is very simple, constituted of only one narrow signal centered at -91 ppm (FWHM=123 Hz) corresponding to Q 3 (0 or 1Al). As for lateritic clay, the environment of aluminum and silicon do not seem affected by the acidic treatment: 27 Al and 29 Si spectra of KGa-2 and KGa-2AF are similar. After the alkaline treatment, the 29 Si environment remains unaffected (no new signal is detected), and the 27 Al spectrum of KGa-2AFCH exhibits only a new shoulder on the left side of the main signal corresponding to additional Al VI species. The structural evolution of KGa-2 is minimal under these conditions. Finally in the case of the series which have undergone the same process but with nitric acid, no modification is observed on 27 Al and 29 Si spectra after the first step (not shown) either for lateritic clay or KGa-2 reference kaolin. On the contrary, some differences are noticed on the consolidated products, LatANCH and KGa-2ANCH 27 Al spectra ( Figures 5B and 6B). For both of them, we observed a new sharp resonance (δ iso =11 ppm, FWHM=80 Hz) indicating the presence of a new 6-fold Al species. This species was not detected in the sample treated with the fulvic acid. Considering XRD characterization, this signal can be attributed to the highly soluble calcium aluminum nitrate. The XRD diffraction of the material LatANCH shows ( Figure 7) the presence of the phases detected in the starting clay namely: kaolinite, goethite, hematite and quartz. In addition, a new phase appears in the material LatANCH and is identified as a hydrated calcium aluminum nitrate (Ca 6 Al 2 O 6 (NO 3 ) 6 .xH 2 O). This newly formed nitrate belongs to the family of compounds called layered double hydroxides of aluminum and calcium. Its crystalline structure is typical of that of a calcium sulfoaluminate type phase (noted AFm). Moreover, in the case of LatANCH sample, we observe the same new Al IV signal as in the case of LatAFCH in a smaller quantity (the final Al VI /Al IV ratio are ~15 and ~4 respectively). No new Al iv species are detected for KGa-2ANCH. In parallel, the 29 Si spectrum of LatANCH consolidated product exhibits the same large contribution and the additional narrow signal centered at -86 ppm as observed for LatAFCH, once again in lower quantities. These signals are not present on KGa-2ANCH spectrum. These observations confirm our assumption that these silicon species belong to the same phase than the new tetrahedral aluminum species. These results seem to indicate on the one hand that the inorganic acid is more aggressive with the lateritic clay than the organic acid, which tends to form organic-mineral complexes [32], and in the other hand that the lateritic clay is more reactive that the KGa-2 kaolin.

Summarized considerations
The acidic reaction seems to have a minor influence on lateritic clay structure. The NMR results indicate only a low difference in the Al VI / Al IV ratio, and no observable difference on silicon environment. Some ICP -AES analysis of the filtrate recovered after 24 h reaction of Lat with fulvic acid indicate that together with iron, 0.62 mass% of initial Al 2 O 3 is dissolved during this acidic attack. One can then imagine that in our "geomimetic" process the 2:1 clay mineral (where aluminum occurs in tetrahedral environment) is partially dissolved and re-precipitate forming an Al-fulvic acid complexes in which aluminum atoms are now six-fold coordinated. This dissolution-reaction is unfortunately too scarce to induce the formation of a new distinguishable Al VI site, but is coherent with the increase of the Al VI /Al IV ratio.
On the contrary, the alkaline neutralization and curing at 60°C for 18 days under saturated water atmosphere induced a strong modification of the lateritic clay. NMR results on LatAFCH sample indicate a significant modification of both Al and Si environment. The occurrence of new Al IV sites indicates the formation new Alcontaining phases, as probably the expected amorphous hydrated calcium aluminum silicate (C-A-S-H), similar to cementitious binding phases. The katoite identified by XRD characterization of the LatAFCH contains tetrahedrally coordinated as well as octahedrally coordinated aluminum. According to literature [33,34], unhydrated katoite contains mainly Al IV sites while hydrated and partially Si-substituted katoite exhibits mainly Al VI sites located at 12.4 ppm and 5.3 ppm. Our "geomimetic" sample seems to contain a hydrated Si-containing katoite. The presence of the 29 Si peak at -86 ppm is in agreement with a concomitant formation of pure hydrated calcium silicate (CSH) gels [35,36] as suggested from DTA-TGA thermograms ( Figure 5).
It can therefore be assumed that the governing reactions during the acidic stage are summarized by eqns. (1)-(3) (R stands for the organic part of the macromolecule). In fact, with organic acid, a sixfold organic-aluminum complex is formed.
After the alkaline neutralization (pH > 11), the 18 days of ageing enhances subsequent reactions leading to the consolidated products. These reactions, involves both silicon and aluminum environments modifications. Considering a given clay mineral, the various reactions are described through eqns. (4)- (8). Partially solubilized silicon-rich and aluminium -rich layers are combining to available calcium cations to form the different cementitious CSH, CASH and CAH like phases. The stoichiometry of these species can vary depending on the ageing period. Nevertheless, the involved reaction kinetic is too slow at room temperature and the detected phases after the optimized strengthening can be considered stable upon a year.

Conclusion
"Geomimetic" materials, obtained using raw lateritic clay, under acidic and alkaline treatments have been finely characterized using 29 Si and 27 Al solid state MAS NMR.
Both Al IV and Al VI species have been evidenced in the raw clay (Lat). The acidic reaction seems to lead to the conversion of a fraction of the four-fold coordinated aluminum into six-fold coordinated aluminum, in agreement with a dissolution-precipitation reaction. The alkaline condition tends to increase the later reaction since new Al IV and Al VI species appear at 61.6 ppm and 12.5 ppm respectively.
The silicon atoms within the Lat clay are related to Q 2 (0Al) or Q 2 (1Al) located at δ iso =-83 ppm, and to Q 3 (0Al) and Q 3 (1Al) environments located at δ iso =-91 and -90 ppm. The enlargement of Q 3 (0Al) peak at the expense of the for the former peaks located at δ iso =-83 ppm (Q 2 or Q 3 ) noted for LatAF, was consistent with an amorphization of the initial clay minerals. A new peak is observed at -86 ppm and is assigned to a Q 2 (1Al) environment within LatAFCH. The latter indicates a greater alteration and redistribution of silicon atoms under alkaline conditions together with curing.
The observed trends together with XRD and DTA-TGA characterizations allow to comfort the assumption of the occurrence of consolidating crystalline (katoite and richterite) and cementitious gel-like CSH (hydrated calcium silicate), CSAH (hydrated calcium and aluminum silicate) and CAH (hydrated calcium aluminate) phases. Furthermore, it appears that the use of fulvic acid partially inhibits the alkaline interaction and promotes the formation of aluminum hydroxide phase that might be responsible of the surface aspect observed with as-obtained "geomimetic" products.