In ﬂ uence of the ionic and nonionic surfactants mixture in the structure and properties of the modi ﬁ ed bentonite clay ☆

A new surface modi ﬁ cation in calcium bentonite clay (Bent-Ca) was investigated in this work, from the sequen-tialintercalationofthreetypesofsurfactants: cetyltrimethylammoniumbromide(CTAB),cationicsurfactant,so-diumdodecylsulfate(SDS),anionicsurfactant,andlaurylalcoholethoxylated(ALEO),nonionicsurfactant.Three modi ﬁ cationsintheBent-Cawerepreparedinaqueous solution obtaining conventionally theOBent-I withCTAB andOBent-IIwithCTAB+SDS,beyondthenewmodi ﬁ cationOBent-IIIwithCTAB+SDS+ALEO.Thesynergistic effect of the sequential mixture of surfactants used in the surface modi ﬁ cation of Bent-Ca was characterized by X-ray Diffraction (XRD), Fourier Transform by Infrared Radiation (FTIR), Scanning Electron Microscopy (SEM), Energy Dispersion Spectroscopy (EDS), Optical Microscopy (OM), Thermogravimetric Analysis (TG), Contact Angle Test and Swelling Index. The results indicated that OBent-III was the reinforcement material among the analyzed ones that presented the highest values of basal spacing (13 nm), thermal stability (400 °C), super ﬁ cial lipophilicity index (79.76°) and swelling index in polar or apolar medium (6 – 10 mL/g). Therefore, a reinforcing material suitable for various applications was proposed.


Introduction
Bentonite clay is a hydrophilic clay-mineral because it presents a three-phase crystalline structure that results in its intrinsic characteristic, the form disposed in stacked layers of silicate, therefore negative charges on the clay-mineral surface are caused by the presence of oxygen and hydroxyls at its vertices, so, it can establish bonds with the also hydrophilic medium through intermolecular hydrogen bonds, since its stacked silicate layers are stabilized by Van der Waals forces by the presence of inorganic cations.In this condition, the clay is dispersed in the medium in an aggregate and intercalated form, thus, not uniformly performing its barrier property [1,2,29].
In this perspective, clay is a material that can be used in several industrial applications depending on the degree of exfoliation with the medium, such as: viscosifier in drilling fluids [3], adsorbent of organic molecules and heavy metals [4,5,28], antibacterial [30,32], and as a barrier material for biopolymer films [6,7].Indeed, the modification of clay in the presence of surfactants has been widely studied as a method of exfoliation.Such exfoliation, in this work, is intrinsically related to the ability of clay to interact with the medium, in other words, if the surface modification in the clay entails scattering of the silicate layers in both polar and non-polar medium, the greater the tendency of this reinforcement material to be applied in various mediums.
In fact, the exfoliation of the clay-mineral can be obtained from the break of the stacking of the silicate layers by the presence of ionic or ethoxylated surfactants, being the functional groups that constitute the tail of the surfactant the responsible for establishing the compatibility of the clay-mineral in the applied system.The most used surfactants in the preparation of modified clays are ammonium quaternary salts of long chain [8,9].However, these salts have some limitations that the non-ionic does not present, such as high resistance to degradation when disposed in the medium [10,11].Moreover, comparing to the ionic surfactants the anionic have the best thermal stability, but they cannot intercalate the silicate layers alone [12,13].
In this study it was proposed to obtain a bentonite clay modified by synergistic intercalations of ionic and nonionic surfactants, aiming to propose a modification that leads to a high dispersion of the silicate layers with the polar or non polar medium compared to the conventional modifications, considering that the modification of the claymineral by synergistically combined surfactants results in compounds with the individual characteristics grouped together.

Materials
The clay used was Calcium Bentonite (Bent-Ca), gently supplied by ArmilMineração doNordeste (AMN), Parelhas, RN, Brazil, with a mean particle size of 0.074 mm and a cation exchange capacity (CEC) of 90 mmol/100 g (determined by the method of adsorption of methylene blue).The chemical composition of Bent-Ca used was characterized by X-ray fluorescence (XRF) and the result is listed in

Modification in betonite clay (OBent)
Four surface modifications in Bent-Ca (OBent) were prepared distinctly: the first was called OBent-I in the presence of CTAB (1.0 CEC), the second OBent-II due to the synergistic presence of CTAB (1.0 CEC) + SDS (0.2 CEC), the third OBent-III by the synergistic presence of CTAB (1.0 CEC) + SDS (0.2 CTC) + ALEO 23 (0.2 CEC) and the fourth was called OBent-IV due to the synergistic presence of CTAB (1.0 CEC) + SDS (0.2 CEC) + ALEO 2 (0.2 CEC).Each chemical surface modification in the bentonite clay was determined by the methodology proposed by Liao et al. [10], where the percentage of each surfactant in 100 mL of distilled water was established for each modification in relation to the Bent-Ca ion exchange capacity.
The dispersion of the bentonite was 6%w in each system.For this, 6 g of bentonite and 94 g of distilled water were mixed and stirred for 30 min at 80 °C in a thermostat.In each system the previous procedure was continued for a further 2 h after addition of the surfactant solutions, and a thermostatic bath was then used to disperse the particles in the mixture.Finally, each mixture was filtered and washed to remove excess salt.The material trapped in the filtration was brought to an oven at 60 °C and held for 24 h.After drying, they were macerated with mortar and pistil, and then sieved in ABNT number 200 sieve (φ = 0.074 mm).

X-ray diffraction -XRD
The XRD analyzes were performed on a Bruker D2 Phaser diffractometer using Cu-Kα radiation (λ = 1.5406Å) with Ni filter and Lynxeye detector.The scanning rate was 0.01°/s for a 2θ interval of 1.5°to 10°with a current of 10 mA and a voltage of 30 kV.The obtained diffractograms were compared with the JCPDS (Joint Committee on Powder Diffraction Standards).Using HighScore Plus Version 3.0e.Rietveld's refinement of the diffraction data was employed using the Maud software (version 2.55).

Fourier transform by infrared radiation -FTIR
The infrared absorption spectra were obtained with a BRUKER spectrometer, model FT-IR VERTEX 70, with scanning of 4000 to 400 cm −1 , resolution of 4 cm −1 and number of scans 16.The samples were characterized directly by total reflectance attenuated, without any preparation.

Scanning electron microscopy -SEM
The micrographs were performed in microscope bench of brand HITACHI, model TM 3000 top, with detector High-sensitivity semiconductor back scattered electron detector, with magnification of 1500×, operating at 15 kV voltage acceleration with tungsten filament.The powder of the samples was added in aluminum stubs and analyzed.

Dispersion spectroscopy of X-ray energy -DSE
For the quantification of the samples was used equipment of bench of the brand HITACHI, model TM 3000, with detector EDX-7000 of Shimadzu, with measurement range of sodium to uranium.

Contact angle
The contact angle measurements are based on the sessile drop technique, according to Boinovich et al. [14].The apparatus is composed of a mobile base with sample holder, a camera (VP 540 s, Intelbras) and a light source.On each sample 0.05 mL of distilled water was dripped and waited 5 s for image capture.The software used to calculate the contact angle of the water droplet on the powder was Surftens 4.5.

Foster swelling test
The swelling index was tested according to ASTM D 5890-11, with isopropyl alcohol and mineral oil being used as the polar and apolar solvent respectively.

Zeta potential
Zeta potentials of Bent-Ca and OBents in distilled water dispersions were tested on a Zeta analyzer of brand Brookhaven, model ZETAPLUS, following the methodology proposed by Wu et al. [15].The samples for  the tests were prepared in concentration of 0.5 g/100 mL under ultrasound conditions.

Optical microscopy -OM
Optical microscopy observations were performed using an Olympus BX52 microscope, equipped with an Olympus C5050Z camera.Optical micrographs were obtained for each dispersion of Bent-Ca or OBents (procedure prepared in the modification step before centrifugation) using the incident light mode.The samples were carefully poured into an alumina carrier and spread under glass cover, at room temperature, before observations.

Thermogravimetric analyzes -TG
A thermogravimetric analyzer and simultaneous calorimeter, model TG209F1 Libra and manufacturer NETZSCH were used for the analysis.All the tests were carried out obeying the following parameters: alumina crucible; nitrogen purge gas; bleed gas flow rate of 50 mL/min; heating rate of 10 °C/min; final temperature of 900 °C and sample mass of 4 mg.

X-ray diffraction
Fig. 1 shows the diffractograms of each OBent overlapping with that of Bent-Ca.
The basal space between the silicate layers that compose Bent-Ca was investigated by the 2θ displacement of the reflection (001) (Fig. 1) that identifies the elements of its crystalline structure.On the other hand, the reflection (002) evidenced in the diffractograms of the OBent samples (Fig. 1) is related to the ordering of the internal network coming from the extensive chain of surfactants [9].
Therefore, the basal space of Bent-Ca was 3 nm, which was displaced to 8 nm after the addition of 1.0 CEC CTAB (Fig. 1), resulting in an expansion of 5 nm.This may be due to the CTAB replacing the calcium cations present on the Bent-Ca surface, attracted by the bromine anions present in the quaternary salt structure.With this, the functional groups of this one intercalate between the layers of silicate increasing the basal space [8].
Sequentially, the basal space reached 10 nm after addition of 0.2 CEC SDS (Fig. 1), which according to Zhu et al. [2], such increase exists  because dipole-dipole bonds occur between the polar parts of the ionic surfactants and dipole bonds induced between the non-polar parts forming a zwitterionic surfactant, which results in increasing the volume of the functional groups between the silicate layers.Finally, after the addition of 0.2 CECALEO 23 the basal space increased to 13 nm (Fig. 1), and the observed effect may be due to the high hydrophilic-lipophilic balance (HLB)of the nonionic surfactant used, since this allows it to establish intermolecular hydrogen bonds with the clay surface, as ion exchange occurs between the exchangeable ions, breaking the stacking and intercalating the agglomerate of the functional groups between the silicate layers, further increasing the basal space, such agglomeration arises from the affinity between the apolar groups of the three surfactants.
Liao et al. [10] observed a reduction of basal spacing in a sodium montmorillonite modified by three surfactants (cationic, anionic and nonionic), with the nonionic surfactant being low in HLB.In fact, Wu et al. [15] argues that nonionic surfactants with low HLB have a lipophilic character and therefore cannot intercalate between the silicate layers of montmorillonite clay.

Infrared radiation by Fourier transform
Fig. 2 shows the overlap of spectra, in the infrared region, of Bent-Ca clay and of each OBent.
Fig. 2 shows bands near 1029 cm −1 and 3631 cm −1 corresponding to the stretching vibrations of Si\ \O\ \Si and Al\ \O\ \Al groups related to tetrahedral and octahedral leaf, respectively, of the three-phase crystalline structure of each silicate layer which characterizes the stacked form of Bent-Ca.The peak around 2958 cm −1 -2849 cm −1 corresponds to the long hydrocarbons chains from the functional groups of each surfactant.
In addition, sharp peaks at 1470 cm −1 were attributed to N\ \H stretching vibrations present in the samples containing the quaternary ammonium salt.The characteristic absorption band of \ \SO 3 was not observed, of the SDS groups, considering that it was covered by the Si\ \O\ \Si (1029 cm −1 ) stretching vibration, such a band would be found between 1204 cm −1 and 1066 cm −1 .
Zhuang et al. [16] observed the same result when characterizing a modification of a calcium montmorillonite in the presence of SDS and a nonionic surfactant.In addition, the peak at 1250 cm −1 is attributed to the C\ \O\ \C stretch vibration corresponding to the mols of ethylene oxide, which constitutes together with the carbonic chain of the lauryl alcohol the structure of the nonionic surfactant used.
The band attributed to 3426 cm −1 , which refers to the elongation O\ \H of water, decreased in the order OBent-III N OBent-II N OBent-I N Bent-Ca, suggesting that adsorption of surfactants sequentially removed partially the water of Bent-Ca proving an increased hydrophobicity.Similar results can be found in Liao et al. [10] and Ouellet-Plamondon, Stasiak & Al-Tabbaa [17].

Scanning electron microscopy
Fig. 3 shows the comparison of the images made by the SEM of each OBent with Bent-Ca.
Fig. 3(a) shows that Bent-Ca had aggregate particles on its surface characterizing the tightly stacked silicate layers.However, the morphologies shown in Fig. 3 disaggregated particles on each surface was observed superficially, demonstrating the intercalation between the silicate layers of both CTAB (Fig. 3(b)) and its synergistic presence with SDS (Fig. 3(c)) due to the stacking of the silicate layers being more dispersed by the larger volume of intercalated functional groups.
Fig. 3(d) shows that OBent-III presented the highest number of disaggregated particles, characterizing, among the micrographs compared, the greater intercalation of the modifier between the silicate layers.It should be emphasized that the scattered lamellar particles were curved because the clay minerals have plasticity properties [9].Similar results were observed by Sun et al. [18] in a resolution of 1500×.

Dispersion spectroscopy of X-ray energy
Fig. 4 and Table 2 show the results of the elemental quantitative analysis of Bent-Ca and each OBent.
Fig. 4(a) shows the presence of calcium in the natural clay (Bent-Ca) by elemental quantitative analysis; these cations are located in the socalled interlamellar region, responsible for interconnecting the silicate layers by compactibilizing them in a stacked format.With this, the clay in its natural state has the characteristic of being able to modify its surface, since the present cations can be exchanged ionically.
The ion exchange is shown in Fig. 4(b), (c) and (d), since the presence of calcium is not identified, as well as is identified the carbon peak, which refers to the hydrocarbons of the groups of each surfactant.
Table 2 shows that the carbon content was higher in OBent-III, which proves the higher molecular weight of the surfactant (CTAB + SDS + ALEO 23 ).Finally, the intercalation of the organic modifiers between the silicate layers is verified.The result confirms the investigated in the XRD, where the increase of the basal space is related to the volume of the intercalated functional groups.Wang et al. [19] and Karaca, Açişli & Gurses [20] also reached similar conclusions, proving the presence of functional groups of surfactants between the silicate layers of a sodium montmorillonite.

Contact angle
The results of the contact angles measured between the drop of water and the surface of Bent-Ca, OBent-I, OBent-II and OBent-III were illustrated in Fig. 5.
It can be seen in Fig. 5 that the contact angle of Bent-Ca was equal to 10°evidencing its affinity for polar molecules, this is due to isomorphic substitutions of Al 3+ by Si 4+ on the leaf of tetrahedra and Mg 2+ or Fe 2+ by Al 3+ on the octahedron leaf occur in the geological formation of the clays, resulting in a negative charge on the surface of the clays.However, part of this negative charge is compensated by the presence of cations in the interlamellar space [25].Therefore, the contact angle of Bent-Ca increased to 41.67°, 64.34°and 79.76°when modified in the presence of CTAB, CTAB + SDS and CTAB + SDS + ALEO 23 , respectively.This fact demonstrated that the synergistic and sequential presence of the three types of surfactants influenced the Bent-Ca surface properties.
As observed in XRD, the three OBent-I, OBent-II and OBent-III clays presented their modifiers intercalated between the silicate layers, so that, by the result of the contact angle, it was possible to perceive that these modifiers are arranged in an orderly manner both in the interlamellar region and in the surface region.In fact, the synergistic presence of the three surfactants resulted in OBent-III having the highest affinity for apolar molecules, due to the fact that it presents a greater number of functional groups arranged neatly on its surface.The values of the contact angles are close to those observed by Ratkievicius et al. [21] and Yuan et al. [11].

Zeta potential
Fig. 6 shows the zeta potential result of Bent-Ca, OBent-I, OBent-II and OBent-III.
As shown in Fig. 6, the zeta potential of Bent-Ca was −12.072 mV while that of OBent-I, OBent-II and OBent-III were +23.63, +26.99 and +33.36 mV, respectively.In fact, the zeta potential of Bent-Ca was negative, proving the negative surface charges of the natural bentonite.Therefore, positive values of OBent demonstrate that the surface of Bent-Ca adsorbed the organic modifier to become organophilic by the presence of the functional groups so as to become vulnerable to establishing dipole bonds induced with the apolar space and dipole-dipole bonds with the polar space [15,22].Thus, the larger the volume of the functional groups arranged neatly in both the interlamellar region and the surface as observed in the results of the greater contact angle the zeta [12].
Wu et al. [15] states that in general, clay particles having zeta potential values greater than +30 mV or lower than −30 mV are considered stable suspensions.Therefore, OBent-III has the ability to establish more  stable connections with the space, that is, it tends to establish connections in a uniform and non-aggregated way in both a polar and a nonpolar space.

Foster swelling test
Fig. 7 shows the swelling indices of the samples in mineral oil (low polarity) and isopropanol (high polarity) before and after stirring.
The observed can be explained by the fact of OBent-III, besides presenting functional groups that are able to establish induced dipole bonds with the apolar medium, as well as in OBent-I and OBten-II, contains non-ionic surfactants with high HBL, considered hydrophilic and selective with a polar medium.The values showed here corroborate with studies carried out by Sanqin et al. [13] and Silva et al. [24].

Optical microscopy
Fig. 8 shows images made on OBent-Ca, OBent-I, OBent-II and OBent-III clays in an optical microscope.
The optical microscopy images of aqueous dispersion of Bent-Ca, OBent-III and IV are showed in Fig. 8.The purpose of the comparison is to observe the influence of the HLB value of the non-ionic surfactant used in each OBent on the aqueous dispersion of the clay.Fig. 8a shows the stacked silicate layers of Bent-Ca aggregately dispersed and intercalated in water, this occurs because such layers interconnect with Van der Waals forces by gaps called galleries or intermediate layers in which the exchangeable cations Ca 2+ reside, which are electrically fixed and partially neutralize the surface of the Bent-Ca, not allowing the layers to exfoliate in water [25].With the addition to the clay of the synergistic mixture of surfactants, differing only in the HLB value of the non-ionic surfactant used, the following best dispersibility ratio was observed in the water: OBent-III N Bent-Ca N OBent-IV.
In fact, OBent-III showed the best exfoliation of the silicate layers in the water (see Fig. 8b), since it used non-ionic surfactant with high HLB in the synergistic mixture of the modifier, achieving the intercalation between the silicate layers and leaving free ethoxylate groups vulnerable to water binding [10,15].Different from this, it occurred with OBent-IV (see Fig. 8c) that due to the use of non-ionic surfactant with low HLB in the mixture, it could not have sufficient strength to break the stacking of the layers and to intercalate between them, with effect, the modifier agglomerated around the surface of the Bent-Ca forming micelles that made difficult the interaction of the clay with water [15].

Thermogravimetric analysis
Fig. 9 shows the overlapping of thermogravimetric curves of Bent-Ca, OBent-I, OBent-II and OBent-III that relates the percentage of mass loss to a given temperature range.
The TGA/DTGA curves of OBent and Bent-Ca are shown in Fig. 9.In the TGA and DTGA curves simultaneously, in each sample different stages of mass loss in the range 25-900 °C are visualized.In all samples, the first decay refers to the desorption of water from the external surface of the clay, as well as to the dehydration of the interlayer, whereas the second decay shows the degradation of the modifier [23].Bent-Ca presented only the first decay, with mass loss around 7.5%.OBent-I, II and III also presented in sequence the second decay with mass loss of 15, 24 and 30%, respectively.The results described suggest that the mass loss of the bentonite increased with the growth of the molecular weight of the modifier.
This increase in the percentage of mass loss corroborates with the improvement of the thermal stability of the material, since it delayed the first stabilization point of the reinforcing material after decomposition of the modifier [25,31].The Bent-Ca, OBent-I, II and III had stabilization temperatures equal to 100, 290, 350 and 400 °C, respectively, according to Fig. 9a, b, c and d.Therefore, the proposed modification made OBent-III the reinforcement material with better thermal stability among the investigated ones.Based on the above results, the proposed modification mechanism can be completed.The Fig. 10 initially shows step I (synergistic mixing of the three surfactants -CTAB, SDS and ALEO 23 -in aqueous solution to obtain the organic modifier) and step II (Bent-Ca in solution).In effect, the structure of the bentonite soon after surface modification concludes the illustration.Such modification in the bentonite is obtained in the presence of the mixture between the solution containing Bent-Ca and the solution of surfactants.On the other hand, when in contact with the clay, the ALEO for presenting 23 units of ethylene oxide has a hydrophilic character that allows to establish intermolecular hydrogen bonds with the surface of the clay, this, connected to the ionic substitution effected between the exchangeable ions in the interlamellar region, results in the disruption of Van der Waals forces that previously stacked the silicate layers.The hydrophobic part of ALEO 23 , in turn, establishes dipole-induced bonds with the also hydrophobic part of the ionic surfactants, and consequently, the polar part of the ionic surfactants by dipole-dipole forces interconnect.In fact, the tangle of functional groups agglomerates in the interlamellar region of the clay [1,26].The described process can also occur on the outer surface of the natural bentonite.

Conclusion
Sequential modifications were made in Bent-Ca -OBent-I, OBent-II and OBent-IIIfrom three types of surfactants.OBent-III was characterized as the best applicability clay as reinforcement material among those studied, due to: the greater basal spacing between the layers of silicate that was 13 nm, 10 nm more than that observed by Bent-Ca; the greater exfoliation of the silicate layers; the greater tendency of the material to stabilize uniformly in either a medium with negative charges by dipole-dipole bonds or hydrogen bonds, or in a medium with positive charges by induced dipole bonds; to be the only clay which showed satisfactory degree of swelling in both the polar solvent and a nonpolar solvent; and higher thermal stability.One of the possible applications of the proposed material can be seen in Monteiro et al. [27].

Fig. 2 .
Fig. 2. Spectra in the infrared region of Bent-Ca and each OBent.

Fig. 4 .
Fig.3shows the comparison of the images made by the SEM of each OBent with Bent-Ca.Fig.3(a)shows that Bent-Ca had aggregate particles on its surface characterizing the tightly stacked silicate layers.However, the morphologies shown in Fig.3(b) and (c) were different from those observed in Bent-Ca, although they are similar to each other.The presence of

3. 10 .
Fig. 10 shows the proposed mechanism for surface modification of bentonite clay in the presence of three types of surfactants (CTAB, SDS e ALEO 23 ).Based on the above results, the proposed modification mechanism can be completed.The Fig.10initially shows step I (synergistic mixing of the three surfactants -CTAB, SDS and ALEO 23 -in aqueous solution to obtain the organic modifier) and step II (Bent-Ca in solution).In effect, the structure of the bentonite soon after surface modification concludes the illustration.Such modification in the bentonite is obtained in the presence of the mixture between the solution containing Bent-Ca and the solution of surfactants.On the other hand, when in contact with the clay, the ALEO for presenting 23 units of ethylene oxide has a hydrophilic character that allows to establish intermolecular hydrogen bonds with the surface of the clay, this, connected to the ionic substitution effected between the exchangeable ions in the interlamellar region, results in the disruption of Van der Waals forces that previously stacked the silicate layers.The hydrophobic part of ALEO 23 , in turn, establishes dipole-induced bonds with the also hydrophobic part of the ionic surfactants, and consequently, the polar part of the ionic surfactants by dipole-dipole forces interconnect.In fact, the tangle of functional groups agglomerates in the interlamellar region of the clay[1,26].The described process can also occur on the outer surface of the natural bentonite.

Table
. The cationic surfactant used was cetyltrimethyl ammonium bromide (CTAB) of molecular formula C 19 H 42 BrNand molecular weight of 364.45 g/mol, supplied by Proquímios.The anionic surfactant was sodium dodecyl sulfate (SDS) of molecular formula NaC 12 H 25 SO 4 and molecular weight of 364.45 g/mol, supplied by Oxiteno.The nonionic surfactant was the class of lauryl alcohol (ALEO n ) with an ethoxylation number equal to 2 (280 g/mol) and 23 (1290 g/mol), termed as diethylene glycol monododecyl ether (C 16 H 34 O 3 ), supplied by Oxiteno.

Table 1
Chemical composition of Bent-Ca.

Table 2
DSE results of Bent-Ca samples and each OBent.