Structural and thermal properties of inorganic–organic montmorillonite: Implications for their potential environmental applications

Journal of Colloid and Interface Science 459 (2015) 17–28

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science journal homepage: www.elsevier.com/locate/jcis

Structural and thermal properties of inorganic–organic montmorillonite: Implications for their potential environmental applications Suramya I. Rathnayake, Yunfei Xi, Ray L. Frost, Godwin A. Ayoko ⇑ Discipline of Nanotechnology and Molecular Science, School of Chemistry, Physics and Mechanical Engineering, Queensland University of Technology (QUT), GPO Box 2434, Brisbane, Queensland 4001, Australia

h i g h l i g h t s

g r a p h i c a l a b s t r a c t

 Inorganic–organic montmorillonite

were prepared using three intercalation methods.  Laser ablation technique was used for elemental analysis.  Surfactant loading depends on the intercalation method and surfactant concentration.  Aluminium-pillars are fixed within the interlayers by calcination.  Calcined interlayers cannot change by subsequent surfactant introduction. Mt

Mt

Al13

ODTMA

Al-Mt

Mt

a r t i c l e

i n f o

Article history: Received 3 June 2015 Revised 25 July 2015 Accepted 30 July 2015 Available online 31 July 2015 Keywords: Montmorillonite Octadecyltrimethylammonium bromide Organoclay Hydroxy aluminium polycation Aluminium-pillared clay Inorganic–organic clay

O-Mt

400oC

Al13

C-Al-Mt

ODTMA Al13

O-Al-Mt

ODTMA

C-O-Al-Mt

O-Al-coMt

a b s t r a c t Inorganic–organic clays (IOCs), clays intercalated with both organic cations such as cationic surfactants and inorganic cations such as metal hydroxy polycations have the properties of both organic and pillared clays, and thereby the ability to remove both inorganic and organic contaminants from water simultaneously. In this study, IOCs were synthesised using three different methods with different surfactant concentrations. Octadecyltrimethylammonium bromide (ODTMA) and hydroxy aluminium ([Al13O4 (OH)24(H2O)12]7+ or Al13) are used as the organic and inorganic modifiers (intercalation agents). According to the results, the interlayer distance, the surfactant loading amount and the Al/Si ratio of IOCs strictly depend on the intercalation method and the intercalation agent ratio. Interlayers of IOCs synthesised by intercalating ODTMA before Al13 and IOCs synthesised by simultaneous intercalation of ODTMA and Al13 were increased with increasing the ODTMA concentration used in the synthesis procedure and comparatively high loading amounts could be observed in them. In contrast, Al/Si decreased with increasing ODTMA concentration in these two types of IOCs. The results suggest that Al-pillars can be fixed within the interlayers by calcination and any increment in the amount of ODTMA used in the synthesis procedure did not affect the interlayer distance of the IOCs. Overall the study provides valuable insights into the structure and properties of the IOCs and their potential environmental applications. Ó 2015 Elsevier Inc. All rights reserved.

1. Introduction ⇑ Corresponding author. E-mail address: [email protected] (G.A. Ayoko). http://dx.doi.org/10.1016/j.jcis.2015.07.071 0021-9797/Ó 2015 Elsevier Inc. All rights reserved.

Many industrial and municipal wastewaters generally contain both inorganic and organic pollutants, which essentially require

18

S.I. Rathnayake et al. / Journal of Colloid and Interface Science 459 (2015) 17–28

treatment before they are released into the environment due to their harmful impacts on human health and ecosystems [1,2]. Hence there is an urgent need to find an efficient and inexpensive method to remove inorganic and organic contaminants simultaneously from polluted water. Naturally abundant clay minerals are inexpensive precursor that can be used as low cost adsorbents due to their unique properties. Smectite clays (e.g. montmorillonite) show high cation exchange capacity (CEC) (80–120 meq/g), high surface area, micro and meso-porosity and swelling properties [3,4]. Montmorillonite exhibits one octahedral layer made with octahedral [AlO3(OH)3]6 sandwich between two tetrahedral layers made with tetrahedra [SiO4]4 . Several octahedra and tetrahedra are connected by sharing three oxygen atoms to form octahedral and tetrahedral sheets, respectively. The clay layers are negatively charged due to the isomorphous substitution of Si4+ and Al3+ by cations with lower charges, and the negative charge is counterbalanced by absorbing exchangeable cations, usually alkaline or alkaline earth metals on the edges and in between the layers [5,6]. Negatively charged hydrophilic clay layers and broken edges attract species with opposite charges such as organic cations and heavy metal ions [7]. But clay minerals are ineffective for the adsorption of non-ionic and non-polar organic compounds (NOCs) in aqueous solutions [8– 10]. The nature of the clays can be modified by replacing the exchangeable cations with organic cations, most commonly quaternary alkyl ammonium cations which make clay surfaces hydrophobic and increase the adsorption capacity towards non-ionic and NOCs. Smectites, being strongly hydrophilic, have significant adsorption capacities towards harmful ionic inorganic water contaminants [11–13] which can be increased by intercalating smectites with pillaring agents such as metal oxide cations to produce pillared clays. The most commonly used inorganic pillaring agent is hydroxy aluminium polycation (Al13) and Al-pillared clays (Al-Mt) are obtained by the intercalation of Al13 cations which exchange the interlayer cations of clay followed by dehydration and dehydroxylation upon calcination to give Al2O3 like-clusters [14]. They act as pillars separating adjacent silicate layers and increasing the basal spacing and creating permanent porosity [15]. Al-pillared clays are more thermally stable and more porous than organoclays, and can be used to remove heavy metals [12,16] and oxyanionic contaminants such as chromate (Cr(VI)) and phosphate [17,18] from water. Studies have been conducted in modifying clays with both inorganic modifiers such as metal hydroxy polycations and organic modifiers such as cationic surfactants in order to obtain hydrophobic/organophilic clays with high thermal stability and porosity as early as 1990 and the resultant clays are named as inorganic–organic clays (IOCs) [19]. According to the literature, they have the ability to remove both inorganic and organic contaminants simultaneously from water. Simultaneous adsorption of dodecane (hydrocarbon) and metal cations (Pb2+, Cu2+, Zn2+, Ni2+, Cd2+) [20], para-nitrochlorobenzene and Cr(VI) [21], phenol and phosphate [22], naphthalene and phosphate [23] have been investigated and the IOCs show high adsorption capacity towards a wide range of organic and inorganic contaminants. The most commonly used organic modifier and inorganic modifier in the preparation of IOCs is hexadecyltrimethylammonium bromide (HDTMA) and Al13 [22,24–27]. IOCs have a wide variety of applications as adsorbents in wastewater treatment because the structure (e.g. interlayer distance, pore size) and properties (hydrophobicity, surface charge) can be adjusted to suit the application by varying the organic modifier (usually alkyl ammonium cations with different chain lengths and different number of substituents) and/or inorganic modifier (different metal hydroxy polycations changing the pillar size and

the distance between the pillars). For example, hydroxy polycations of Fe(III), Cr(II) and Ti(II) can be used as inorganic modifier and the pillaring agent in IOCs instead of hydroxy aluminium, and the resultant IOCs are effective adsorbents towards nitrophenol, basic yellow 28 dye [28], Supranol yellow 4 GL [29], herbicide diuron [30] and dye sulfacid brilliant pink [31]. Also, mixed metal (Fe/Al) pillared IOCs with improved structure and properties shows promising adsorption capacity towards both inorganic water contaminants (e.g. Cu2+) and organic water pollutants (e.g. phenol) [32]. Phenol and its chloro and nitro derivatives are most common water pollutants present in all types of wastewaters and IOCs shows high affinity towards these water pollutants [24,31,33,34]. According to the literature, the structure and properties of IOCs strongly depend on the intercalation method (the sequence of introducing inorganic and organic modifier) and the ratio of inorganic and organic modifiers (concentration of surfactant and the metal hydroxy polycation used in the synthesis procedure) [21,23]. Three different methods of synthesising IOCs have been reported; (i) cationic surfactant intercalation into montmorillonite before metal hydroxy polycation intercalation (ii) cationic surfactant intercalation into montmorillonite after metal hydroxy polycation intercalation and (iii) simultaneous cationic surfactant and metal hydroxy polycation intercalation into montmorillonite [21,23,35,36]. Recently, the influences of the calcination temperature of the pillared clays and the used dosage of surfactant on the interlayer structure and the thermal stability of IOCs synthesised by intercalating Al13 first and then HDTMA were investigated [27]. To best of our knowledge, ODTMA has not been used in IOCs synthesis and no comprehensive studies have been conducted to determine the effect of surfactant concentration (the ratio of surfactant/metal hydroxy polycation) used in the synthesis procedure on the structure and the properties of the IOCs synthesised by the three methods mentioned above. Although ODTMA modified organoclays have been reported [37–39] more studies are required for the clear understanding of the structure and properties of IOCs synthesised using ODTMA as the organic modifier and Al13 as the inorganic modifier. Therefore, the aim of the study is to determine the effect of surfactant concentration used in the synthesis process (the organic and inorganic modifier ratio) on the structure and the properties of IOCs and to explore the probable applications of the resultant IOCs for the removal of both recalcitrant inorganic and organic pollutants from wastewater. In order to fulfil these aims, a series of IOCs were synthesised using Al13 and ODTMA as inorganic and organic modifiers, respectively. ODTMA concentration used in the synthesis procedure was varied and three different methods were employed in the synthesis of IOCs. The structure and properties of the resultant IOCs were investigated and compared with that of organoclays with different ODTMA loadings and Al-pillared clays. Changes in the interlayer distance of these modified clays due to intercalation of ODTMA and Al13 were determined by the X-ray diffraction (XRD) technique. The thermal stability of IOCs synthesised using different intercalation methods at different ratios of ODTMA/Al13 was evaluated by thermogravimetric analysis (TG) and derivative thermogravimetric analysis (DTG). Furthermore, the TG/DTG results can be used to determine the different environments of the surfactant and thermal stability within the clay layers. Laser ablation inductively coupled plasma mass spectroscopy (LA-ICP-MS) was employed to determine the intercalated Al13 into IOCs. Although, X-ray fluorescence spectroscopy (XRF) is the most commonly used method for this purpose [16,27,32], in this study the applicability of LA-ICP-MS to determine Al/Si ratio of Al pillared clays and IOCs was explored.

S.I. Rathnayake et al. / Journal of Colloid and Interface Science 459 (2015) 17–28

19

Texas Montmorillonite STx-1 (denoted as Ca-Mt) used in this study was supplied by the Clay Minerals Society and was used without further purification. The cation exchange capacity of this montmorillonite is 84.4 meq/100 g (according to the specifications of its producer) and the major exchange cation is Ca2+ The surfactant selected for this study is analytical grade octadecyltrimethylammonium bromide (denoted as ODTMA, C21H46BrN, FW: 392.50), purchased from Sigma–Aldrich and used without further purification. All the other chemicals, including sodium carbonate anhydrous and aluminium chloride hexahydrate were of analytical grade and used without further purification.

Ultrasonic Cleaner 1.8L 60 W heated digital instrument. A drop of ammonia (Conc. 10–35%) was added to the clay suspension as the dispersant agent and it was shaken for 30 s. The clay was allowed to settle for 5 min and then an aliquot from the top of the suspension containing small clay particles was transferred to a silicon wafer using a Pasteur pipette. A water free thin layer of clay was obtained on the silicon wafer by air drying for 12 h. X-ray diffraction (XRD) patterns were collected using Co Ka radiation (k = 1.78897 Å) on a PANalytical X’Pert Pro MPD Power diffractometer operating at 40 kV and 40 mA with 0.5° divergence slit, 1° anti-scatter slit, between 2° and 42° (2h) at a step size of 0.0167°. Then, the clay samples were treated with ethylene glycol and kept in a desiccator for 1 h to expand the clay layers. Finally, the XRD patterns of ethylene glycol treated clay samples were collected using the same method [38].

2.2. Synthesis of IOCs

3.2. Thermogravimetric analysis

Three different methods were used to prepare IOCs: (1) ODTMA was first intercalated into the galleries of clay (Mt) to synthesise organoclay, and then Al13 is further used to intercalate the organoclay [O-Al-Mt], (2) Al13 was first intercalated into the clay and calcined to synthesise pillared clay, and then ODTMA was used to intercalate the pillared clay [C400-Al-O-Mt], and (3) ODTMA and Al13 were first mixed, and then they were simultaneously used for the intercalation [O-Al-coMt] [23]. For the purpose of comparison, the Al13 pillared montmorillonite (Al-Mt) and calcined Al13 pillared montmorillonite (C400-Al-Mt) and ODTMA intercalated montmorillonites (Ox-Mts) were also synthesised. For each case, the amounts of Al13 were fixed at 10 mmol Al/g montmorillonite [23] while those of ODTMA were selected as 0.2, 0.5, 1.0 and 2.0 CEC to obtain different ODTMA/Al13 ratios. The IOCs obtained from the three methods were denoted as Ox-Al-Mt, C400-Al-Ox-Mt and Ox-Al-coMt respectively (where ‘‘x’’ indicates the amounts of ODTMA used for synthesising IOCs, which were 0.2, 0.5, 1.0 or 2.0 CEC). To prepare pristine hydroxy aluminium solutions, 0.5 M Na2CO3 solution was added slowly to 1.0 M AlCl3 solution under vigorous stirring and stirring kept for 18 h at 60 °C. The final hydrolysis ratio of OH /Al+13 was kept at 2.4 and the solution was subsequently aged for 18 h at 60 °C [23,27]. Pillaring of Al13 into clays was carried out by adding Mt or Ox-Mt into hydroxy aluminium solution and the mixture was stirred for 24 h, and then the solution was aged for 24 h at 60 °C. Calcined derivatives of pillared clays were prepared by heating pillared clay samples in a furnace at 400 °C (using 15 °C/min temperature gradient) for 8 h and the obtained products were denoted as C400-Al-Mt or C400-Al-Ox-MT. Intercalation of ODTMA into Mt was accomplished by the following method: a desired amount of surfactant was dispersed in deionised water by stirring at 60 °C for 0.5 h, and then Mt or C400-Al-Mts were added to the surfactant solution. The mass ratio of water/clay was 20:1 and the mixture was stirred for 24 h at 60 °C [27]. All the synthesised materials were filtrated and washed with deionized water until the washings were free of chloride or bromide ions as determined by the use of the AgNO3, and then freeze dried for 48 h. The dried modified clays were ground in an agate mortar, and stored in a vacuum desiccator.

High-resolution TG was performed on a Q500 TGA Pegasus instrument operating at ramp 5 °C/min with resolution 6.0 °C from room temperature to 1000 °C in a high purity flowing nitrogen atmosphere (40 cm3/min). 30–40 mg of the finely ground sample was heated each time in an open platinum crucible. The derivative thermogravimetric (DTG) curves were derived from the TG curves automatically.

2. Experimental 2.1. Materials

3. Characterisation methods

3.3. Laser ablation inductively coupled plasma mass spectroscopy (LA-ICP-MS) Pellets with a diameter of 6 mm were made with 40–50 mg of Ca-Mt and modified clays and a pressure of 3.5 tons. The pellets and NIST 610, as reference material, were introduced into the ESI New Wave Truline laser ablation cell with a helium atmosphere flow rate of 450 ml/min. The laser cell is designed as a cup and the helium flow in it was set to 50 ml/min. Each sample was line rastered three times across the sample surface at 500 lm/min using an excimer laser with a pulsing rate of 10 Hz and a circular spot size of 65 lm. The laser was connected to an Agilent 8800 ICP-MS, a single collector instrument, which was run in single quad mode. Sample from the laser cell in He was mixed with Ar carrier gas flow of 100 ml/min and introduced to the plasma via TygonÒ tubing. The ICP-MS settings were: RF power = 1350 W, the X-Lens settings were optimised for high sensitivity with a ThO/Th of <0.03%. 14 isotopes were measured for a dwell time of 0.02s and a mass sweep was completed in 0.3175 s. The line rasters were 60 s long and gas backgrounds of 40 s were collected between each laser on interval. The data obtained were processed using IOLITE. The Al/Si ratio obtained for Ca-Mt using LA-ICP-MS method was compared with the results obtained from X-ray fluorescence (XRF) spectroscopy.

3.4. Organic carbon content (foc) 1 mg of Ca-Mt and modified clays were examined with LECOÒ TruSpec Micro organic carbon analyser. Sulfamethazine was used in the instrumental calibration and ODTMA loading amounts of the solid samples were calculated according to the obtained organic carbon content (foc).

3.1. X-ray diffraction

3.5. X-ray fluorescence spectroscopy

A sample of 10 mg of Ca-Mt and modified clay samples was dispersed in 20 ml of deionized water and sonicated for 10 min using

The major elements present in Ca-Mt were determined by using SPECTRO XEPOS R4 benchtop XRF Spectrometer.

20

S.I. Rathnayake et al. / Journal of Colloid and Interface Science 459 (2015) 17–28

4. Results and discussion 4.1. XRD The XRD patterns obtained for ethylene glycol treated and untreated organoclays and IOCs are presented in Fig. 1(a–h) and the corresponding d001 values are listed in Table 1. The interlayer distance of the organo-montmorillonite and IOCs synthesised using the first and third methods increased gradually with increasing ODTMA concentration used in the synthesis procedure. This confirms the entry of cationic surfactant into the interlayer space of the Ca-Mt. However, the interlayer distance of IOCs synthesised using the second method was not affected by changing the ODTMA concentration. This means that the surfactant concentration used in the preparation procedure greatly affects the interlayer distance of the organoclays and IOCs differently when the synthesis method is changed. The interlayer distance of the unmodified Ca-Mt, 1.54 nm, resulted from the presence of hydrated exchangeable cations such as Ca2+ and Na+ between clay layers and the interlayer space environment is hydrophilic. In O0.2-Mt, the interlayer distance decreased to 1.47 nm. This is due to the replacement of hydrated exchangeable cations with surfactant cations, depleting the exchangeable cations [40]. Upon treatment with ethylene glycol, 001 reflection of smectite clays shifts to the left, giving an increased basal spacing around 1.70 nm [38]. According to the results, the interlayer spacing of Ca-Mt, O0.2-Mt and O0.5-Mt increased from 1.54, 1.47 and 1.69 nm to 1.71, 1.73 and 1.77 nm, respectively, after treating with ethylene glycol (Fig. 1(a) and (b)). This is evidence of ethylene glycol entering into the interlayers of these organoclays. The interlayers are not saturated at relatively low ODTMA concentrations; therefore there is enough space for ethylene glycol to be intercalated into the interlayer space, increasing the interlayer distance accordingly [41]. At 1 CEC ODTMA concentration, the ODTMA modified montmorillonite without ethylene glycol treatment gives a reflection at 2.09 nm. Ethylene glycol treated O1.0-Mt shows an additional reflection at 4.29 nm, apart from the reflection at 2.07 nm, which remained constant compared to that of O1.0-Mt (2.09 nm). Clay layers with lower charge densities are saturated with ODTMA at 1 CEC concentration and their hydrophobic properties are increased, preventing the intercalation of hydrophilic ethylene glycol into the interlayers. In contrast, clay layers with higher charge densities which are capable of adsorbing more surfactant are unsaturated at 1 CEC ODTMA concentration. This allows further intercalation of ethylene glycol into the interlayers increasing basal spacing up to 4.29 nm. When ODTMA concentration increased to 2 CEC, two reflections at 2.10 nm and 4.31 nm are observed for the untreated O2.0-Mt (similar to that of ethylene glycol treated O1.0-Mt). These two reflections did not increase significantly after treatment with ethylene glycol meaning that the interlayers of the clay with both higher and lower layer charges are saturated by the surfactant at 2 CEC concentration of ODTMA. In the XRD pattern of ED-O2.0-Mt, the reflection at higher and lower 2 theta positions did not disappear giving single reflection around 1.70 nm, upon ethylene glycol treatment. This supports the idea of heterogeneous basal spacing due to changes in the layer charge densities in this clay mineral [23]. According to previous studies, at high surfactant concentration, ethylene glycol could not be intercalated into the clay interlayers saturated with the surfactant, instead, ethylene glycol adsorbs onto the outer surface of the clay layers [41]. This may have happened in ethylene treated O2.0-Mt, where some of the externally adsorbed ODTMA molecules were replaced by ethylene glycol and the

discharged ODTMA, acting like a solid crystalline phase, is responsible for the additional d001 reflections of at 2.86 nm, 1.43 nm of O2.0-Mt (Fig. 1(b)). These reflections are similar to the reflections in the XRD pattern of pure ODTMA shown in Appendix 1. Surfactant molecules with long alkyl chain such as ODTMA intercalated into the interlayer space of the clay form a hydrophobic partition medium which functions as a bulk organic phase, making interlayer space hydrophobic and they show a great potential towards the adsorption of hydrophobic non-ionic organic molecules [40,42]. The molecular arrangement of the surfactant cations can be deduced from d001 values. The organophilic clays intercalated with long chain surfactant cations forms not only monolayers but bilayers, pseudo-trimolecular arrangements and paraffin-type structures within the interlayer space of the clay minerals [10,40,43–45]. According to the literature, the TOT layer (octahedral + 2 tetrahedral) thickness of montmorillonite is 0.97 nm and the height of the ODTMA cation (head group) varies from 0.67 nm to 0.51 nm according to its molecular arrangement [40]. A lateral-monolayer arrangement of ODTMA in the interlayer space of montmorillonite is implied for 1.47 nm d-spacing value of O0.2-Mt according to the dimensions of ODTMA and layer thickness mentioned above. At 0.5 CEC concentrations, according to the basal spacing (1.69 nm), it is predicted that molecular arrangement varying from lateral monolayer to lateral bilayer arrangements demonstrating property variation (e.g., surface charge density) of montmorillonite layers. Interlayer surfactant in O1.0-Mt takes pseudo-trimolecular layer arrangement according to its d001 value, 2.09 nm. The reflection at 4.31 nm of O2.0-Mt corresponds to a paraffin-type bilayer arrangement since the basal spacing for a fully vertical 18C chain is at most as long as 3.78 nm (2.81 + 0.97 nm) where 2.81 nm is the full length of a 18C chain [40]. Heterogeneous charge distribution on the clay layers leads to the heterogeneous distribution of the intercalation agents within the same layers or different clay layers [23]. This heterogeneous distribution of intercalating agents leads to different interlayer distances in EG-O1.0-Al-Mt (at 2.11, 3.54 and 4.25 nm) and EG-O1.0-Al-coMt (2.13, 3.63 and 4.46 nm) upon treatment with ethylene glycol (Fig. 1(d) and (h)). At 2 CEC ODTMA concentration, IOCs synthesised using the first and third methods show two reflections around 4.30 and 2.10 nm (similar to that of O2.0-Mt) and upon ethylene glycol treatment the reflection at 4.30 nm slightly increased. However, additional peaks could not be observed which correspond to crystalline ODTMA in these two IOCs. This is because there are less ODTMA molecules adsorbed onto the outer surface of the clay layers than that of O2.0-Mt, allowing ethylene glycol to adsorb onto the outer surface of the clay. In contrast to organoclays and IOCs synthesised using the first and third methods, interlayer distance of IOCs synthesised using the second method, did not increase significantly with increasing ODTMA concentration and their basal spacings are similar to that of the precursor calcined Al-pillared clay (C400-Al-Mt). This means that firstly introduced Al13 cations are fixed within the interlayers upon calcination and block the interlayer space reducing subsequent intercalation and adsorption of ODTMA into the interlayer space. Therefore, ODTMA adsorbed onto the outer surface of the clay increased significantly [23]. The interlayer distance increased from 1.74 nm to 1.88 nm with increasing ODTMA concentration up to 1 CEC, indicating ODTMA intercalation into the interlayer space of the clay (Fig. 1(e) and (f)). At 1 CEC ODTMA concentration, an additional set of reflections with high intensity could be seen at 2.87 and 1.42 nm in IOCs synthesised using the second method (Fig. 1(e)). They are similar to the ones appearing in ethylene glycol treated O2.0-Mt (Fig. 1(b)) and pure ODTMA (Appendix 1). However, the intensity of the reflections at 2.87 nm significantly

S.I. Rathnayake et al. / Journal of Colloid and Interface Science 459 (2015) 17–28

21

Fig. 1. XRD patterns of (a) and (b) organoclay and ethylene glycol treated organoclay; (c) and (d) IOCs synthesised using first method and ethylene glycol treated IOCs synthesied using the first method; (e) and (f) IOCs synthesised using second the method and ethylene glycol treated IOCs synthesied using the second method; (g) and (h) IOCs synthesised using the third method and ethylene glycol treated IOCs synthesised using the third method.

22

S.I. Rathnayake et al. / Journal of Colloid and Interface Science 459 (2015) 17–28

Fig. 1 (continued)

23

S.I. Rathnayake et al. / Journal of Colloid and Interface Science 459 (2015) 17–28 Table 1 Basal spacing, Al/Si ratio and ODTMA content of the samples.

a b c

Sample

d001 (nm)

d001 (nm) (ethylene glycol treated)

ODTMA amount (mmol/g)a

ODTMA amount (mmol/g)b

Al/Si (w/w%)c

Ca-Mt O0.2-Mt O0.5-Mt O1.0-Mt O2.0-Mt Al-Mt C400-Al-Mt O0.2-Al-Mt O0.5-Al-Mt O1.0-Al-Mt O2.0-Al-Mt C400-Al-O0.2-Mt C400-Al-O0.5-Mt C400-Al-O1.0-Mt C400-Al-O2.0-Mt O0.2-Al-coMt O0.5-Al-coMt O1.0-Al-coMt O2.0-Al-coMt

1.54 1.47 1.69 2.09 2.10, 4.31 1.73 1.80 1.55 1.81 2.04 2.12, 4.25 1.74 1.83 1.88 1.82 1.70 1.85 2.20 2.12, 4.22

1.71 1.73 1.77 2.07, 2.17, 1.85 1.77 1.70 1.89 2.11, 2.21, 1.82 1.83 1.86 1.79 1.64 1.91 2.13, 2.13,

0 0.16 0.36 0.64 1.05 0 0 0.24 0.39 0.62 0.94 0.14 0.29 0.48 0.49 0.22 0.37 0.63 1.00

0 0.16 0.41 0.74 1.17 0 0 0.15 0.37 0.69 1.08 0.14 0.30 0.45 0.49 0.14 0.37 0.68 1.13

0.28 – – – – 0.45 0.45 0.44 0.39 0.30 0.28 0.46 0.46 0.46 0.46 0.45 0.42 0.32 0.29

4.29 4.54

3.54, 4.25 4.52

3.63, 4.46 4.36

The amount of loaded ODTMA was calculated from the thermogravimetric analysis data. The ODTMA loading amount calculated from organic carbon content obtained from organic carbon analyser (LECOÒ TruSpec Micro carbon analyser). The ratio was calculated from LA-ICP-MS technique.

is reduced and all the other reflections which correspond to ODTMA disappeared after treating with ethylene glycol (Fig. 1(f)). This means that ethylene glycol, could not enter into the interlayers or adsorbed onto the surface of the clay layers but dissolved some of the crystalline ODTMA mixed with the modified clay decreasing the intensity of the reflections.

4.2. Thermogravimetric analysis The high-resolution thermogravimetric analysis (HRTG) of ODTMA modified montmorillonite (organoclays) and IOCs are presented in Figs. 2–5. Four major mass loses were observed. The initial mass loss was observed in the 30–150 °C temperature range, corresponding to the evaporation of free water adsorbed between silicate layers and water molecules around the exchangeable cations such as Ca2+ and Na+ (dehydration). The second major mass loss from 150 to 400 °C corresponds to the thermal decomposition of the surfactant, ODTMA which occurs in several steps. Mass losses in the temperature range of 170–500 °C are solely attributed to the decomposition of the surfactant in organoclays, because montmorillonite does not undergo thermal changes in this range. But in the IOCs synthesised using first and third methods, there were some contributions to the mass loss from dehydroxylation of Al13 cations within the temperature range of 200–450 °C. This is especially the case for IOCs with high intercalated Al13 cations such as O0.2-Al-Mt and O0.2-Al-coMt (Table 1). More accurate values for ODTMA loading amounts in organoclays and IOCs can be calculated from the organic carbon contents (foc) obtained from organic carbon analyser and presented in Table 1. The third mass loss occurred between 400 and 500 °C in which the slight mass loss is related to the removal of OH groups bound to the edges of the clay laminates (dehydroxylation). The final or the fourth mass was observed between 500 and 900 °C, which corresponds to the total dehydroxylation of the hydroxyl groups in the aluminosilicate structure [38,42]. The mass losses corresponding to the adsorbed water for Ca-Mt modified with ODTMA and all IOCs were lower than those of Ca-Mt and pillared clays and they gradually decreased with the increasing surfactant loading (Table 2). This means the surface affinity has changed from hydrophilic to hydrophobic with increasing surfactant loading.

The loading amounts of ODTMA in organoclays and IOCs obtained from thermogravimetric analysis are presented in Table 1. According to the results, ODTMA loading amount strongly depend on the intercalation method employed and the ODTMA/Al13 ratio. IOCs synthesised using the first and the third methods have high ODTMA loading amounts which are similar to organoclays and surfactant loading amount increased significantly with increasing surfactant concentration, there by ODTMA/Al13 ratio in the synthesis procedure. The surfactant loading is lowest in IOCs synthesised using the second method and did not increase significantly after 1CEC ODTMA concentration. The decomposition temperature, hence the thermal stability changes according to the molecular environment of the surfactant. The loaded surfactant is in three different environments in the organoclays and IOCs synthesised using the first and the third methods as represented in the TGA/DTG plots (Figs. 2, 3 and 5) and the ODTMA loaded amounts (%) are presented in Table 2. Similar observations were reported for surfactant modified clays in previous studies [37,45]. At low ODTMA loading, only two decomposition peaks can be seen in the DTG plots. The peak with highest decomposition temperature represents the decomposition of intercalated ODTMA cations into the interlayer space via electrostatic forces and surfactant molecules (ionic pairs) located within the interlayer spaces. They have the highest thermal stability. Decomposition of surfactant molecules in the interparticle pores, forming van der Waals interaction between alkyl chains with second highest thermal stability, is represented by the second peak. With increasing surfactant loading amount, a third peak appeared at low decomposition temperature with the lowest thermal stability. It represents the decomposition of surfactant molecules adsorbed on to the outer surface of the clay layers, which significantly increased with increasing surfactant concentration from 1 CEC to 2 CEC [23,37,42]. However, unlike ODTMA modified organoclays and IOCs synthesised by the first and third methods, two peaks were observed due to the decomposition of surfactant in the TGA/DTG plot of IOCs synthesised through the second method (Fig. 4). This suggests that there are differences in the structures of the IOCs and the molecular environment, and the thermal stabilities of ODTMA molecules within the IOCs synthesised using the second method

24

S.I. Rathnayake et al. / Journal of Colloid and Interface Science 459 (2015) 17–28

Fig. 2. TG and DTG patterns of Ca-Mt modified with ODTMA.

Fig. 3. TG and DTG patterns of IOCs synthesised using first method.

S.I. Rathnayake et al. / Journal of Colloid and Interface Science 459 (2015) 17–28

Fig. 4. TG and DTG patterns of Al-pillared clays and IOCs synthesised using second method.

Fig. 5. TG and DTG patterns of IOCs synthesised using third method.

25

26

S.I. Rathnayake et al. / Journal of Colloid and Interface Science 459 (2015) 17–28

Table 2 Data from thermogravimetric analysis of Ca-Mt modified with ODTMA and IOCs. Clay

Ca-Mt Al-Mt C400-Al-Mt O0.2-Mt O0.5-Mt O1.0-Mt O2.0-Mt O0.2-Al-Mt O0.5-Al-Mt O1.0-Al-Mt O2.0-Al-Mt C400-Al-O0.2-Mt C400-Al-O0.5-Mt C400-Al-O1.0-Mt C400-Al-O2.0-Mt O0.2-Al-coMt O0.5-Al-coMt O1.0-Al-coMt O2.0-Al-coMt

Dehydration

1st ODTMA mass loss

2nd ODTMA mass loss

3rd ODTMA mass loss

Mass loss (%)

Temperature (°C)

Mass loss (%)

Temperature (°C)

Mass loss (%)

Temperature (°C)

Mass loss (%)

Temperature (°C)

16.59 12.51 11.49 8.83 5.57 3.53 3.30 7.07 5.88 3.97 4.26 6.97 3.71 2.75 2.76 7.63 5.59 4.77 2.74

35.0, 115.0 38.5 34.5 33.0, 113.5 71.5, 106.5 36.5 41.5 34.0 35.5 35.5 39.5 35.5 41.5 40.0 153.0 42.0 38.0 39.0 41.0

– – – – – 9.91 26.72 – – 7.06 23.49 – – 9.11 9.87 – – 9.49 25.9

– – – – – 223.0 184.0 – – 213.0 176.0 – – 202.5 195.5 – – 208.5 174.0, 207.0

– – – 1.36 2.81 4.95 5.95 2.69 3.90 7.09 5.41 – 1.96 – – 2.87 3.43 6.59 5.62

– – – 263.0 285.5 284.5 288.0 235.0 280.5 269.0 263.5 – 240.0 – – 243.5 294.0 259.5 271.0

– – – 4.97 11.33 10.43 8.41 4.97 9.73 9.33 7.63 5.57 9.32 9.61 9.44 4.24 9.54 7.79 8.44

– – – 370.0 370.0 378.0 384.0 378.5 356.0 375.0 372.5 353.0 353.5 354.5 355.0 378.0 347.0 370.0 372.0

compared to the other IOCs. Only one peak was observed in the DTG plot of C400-Al-O0.2-Mt. This peak was observed at 350 °C which is lower than the temperature observed for intercalated ODTMA cations into the cation exchange sites replacing original exchangeable cations (380 °C). This is because no intercalation of ODMTA into exchange sties via electrostatic interactions is possible since the exchange sites are filled by firstly intercalated Al13 cations. Only possible adsorption of ODTMA is into the pores between aluminium oxide pillars in the interlayer space forming weak van der Waals interactions between alkyl chains of the surfactant molecules. In IOCs synthesised using the second method, the decomposition temperature for the first DTG peak did not change hence the thermal stability was constant with increasing ODTMA concentration. A second peak appeared in the DTG plot of C400-Al-O0.5-Mt at 240 °C and this mass loss corresponds to the decomposition of ODTMA molecules supported onto the external surface of the clay layers forming interactions with the clay surface. The decomposition temperature (thermal stability) of this peak decreased and finally became very close to 196 °C which is the decomposition temperature of pure ODTMA while the corresponding mass loss (the ODTMA loaded amount) slightly increased with increasing ODTMA concentration used in the preparation procedure. This means that ODTMA exists with C400-Al-O1.0-Mt and C400-Al-O2.0-Mt as another solid crystalline phase on the outer surface of the clay. Overall, the decomposition temperature of intercalated ODTMA is around 380.0 °C (Table 2) indicating the high thermal stability of the ODTMA molecules in organoclays and IOCs (This suggests that the thermal stability of ODTMA has increased due to the intercalation and adsorption.)

Total ODTMA amount (%)

– – – 6.33 14.14 25.29 41.08 7.66 13.63 23.48 36.53 5.57 11.28 18.72 19.31 7.11 12.97 23.87 39.96

increasing ODTMA concentration the ratio gradually decreased to 0.28, which is similar to that of Ca-Mt. At low ODTMA concentration such as 0.2 CEC, only a few exchange sites are filled by ODTMA and there are available exchange sites for introducing hydroxy-aluminium cations. A large number of exchange sites are filled firstly by the introduced ODTMA when the ODTMA concentration used in the synthesis procedure is high (e.g. 1 CEC and 2 CEC). Therefore, less number of exchangeable sites is available for secondly introducing hydroxy aluminium cations. As a result, Al/Si ratio decreased gradually with increasing ODTMA loading amount (Table 1). Consequently, in O2.0-Al-Mt Al/Si ratio is similar to that of Ca-Mt (0.28). IOCs synthesised using the second method have the Al/Si ratio similar to that of the precursor, C400-Al-Mt (0.46), regardless of the changes in the ODTMA concentration used in the synthesis procedure. This proves that the aluminium oxide is fixed in-between the clay layers and no amount of ODTMA can replace them from the interlayers. This supports the constant d001 values obtained in IOCs synthesised using the second method. IOCs synthesised using the third method shows a similar pattern in Al/Si ratios to that of IOCs synthesised using the first method. Al13 cation with higher charge (+7) has higher exchange capacity with exchangeable interlayer cations of the clay. Therefore, Al13 cations shows the highest selectivity towards the exchange sites of the clay in a mixture of ODTMA and Al13 cation solution at low ODTMA loading amount (e.g. 0.2 CEC) [27]. This is reflected in the fact that Al/Si ratios of IOCs synthesised at 0.2 CEC ODTMA concentration are the highest and are equal to that of the precursor, Al-Mt (0.45). The Al/Si ratio in IOCs synthesised using the first and the third methods decreased with increasing ODTMA concentrations (Table 1).

4.3. LA-ICP-MS 5. Conclusions In this study, LA-ICP-MS is used to determine the Al/Si ratio of Al-pillared clays. The Al/Si ratio for Ca-Mt obtained from LA-ICP-MS method is comparable with that obtained from XRF spectroscopy (Appendix 2) proving the validity of the method. Successful intercalation of Al13 cation into the interlayer space of montmorillonite is confirmed by the increased Al/Si ratio (0.45) in Al pillared clays compared to that of Ca-Mt (0.28) (Table 1). At low ODTMA concentration, Al/Si ratio of IOCs synthesised using the first method is similar to that of Al-Mt (0.45) however, with

In conclusion, octadecyltrimethylammonium bromide (ODTMA) loading amount and intercalated hydroxy aluminium (Al13) amount, and consequently, the structure and the properties of the inorganic–organic clays (IOCs) strongly depend on the intercalation method employed and the ratio of ODTMA/Al13. The IOCs synthesised by intercalation of ODTMA before Al13 (first method) and IOCs synthesised by simultaneous intercalation of ODTMA and Al13 (third method) behave in a similar pattern with increasing

S.I. Rathnayake et al. / Journal of Colloid and Interface Science 459 (2015) 17–28

ODTMA/Al13 ratio. Higher ODTMA loading amounts were observed on these IOCs, indicating that most of the added ODTMA in the synthesis procedure could be intercalated into Ca-Mt with these two methods. However, higher ODTMA concentrations did not favour the intercalation of Al13 into IOCs synthesised using first method and third method. As a result, these IOCs have superior hydrophobicity due to the superficially adsorbed ODTMA molecules on the outer surface of the clays and they are potentially effective adsorbents of hydrophobic non-polar organic pollutants in wastewater and groundwater [20,22,35]. These externally adsorbed surfactants may reverse the charge of the clay layers from negative to positive enabling the adsorption of anionic contaminants from water [21]. Al13 dehydroxylate and form aluminium oxide pillars upon calcination which are fixed in between layers preventing the replacement of surfactant molecules even at the high surfactant concentration used in the synthesis procedure (IOCs synthesised using the second method). These pre intercalated aluminium oxide pillars block the interlayer space, as a result the lowest surfactant loading amounts can be seen in IOCs synthesised using the second method even at higher ODTMA concentrations. Since the amount of surfactant molecules adsorbed onto the outer surface is significantly lower than the other IOCs these IOCs are less hydrophobic and show less adsorption capacities towards hydrophobic non-polar organic contaminants. The optimum preparation procedure for the successive introduction of metal polycations and cationic surfactant on montmorillonite appears to be the second method. Studies on the simultaneous adsorption of recalcitrant inorganic and organic water pollutants by the synthesised IOCs are underway. Acknowledgment The authors are grateful to the Central Analytical Research Facilities (CARF) at Queensland University of Technology (QUT) for the provision of technical assistance and research infrastructure for this work. One of us (SIR) is also thankful to QUT for a QUT Postgraduate Research Award.

Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2015.07.071. References [1] T. Hillie, M. Hlophe, Nanotechnology and the challenge of clean water, Nat. Nanotechnol. 2 (2007). [2] E. Iakovleva, M. Sillanpaa, The use of low-cost adsorbents for wastewater purification in mining industries, Environ. Sci. Pollut. Res. Int. 20 (2013) 7878– 7899. [3] E. Ruiz-Hitzky, P. Aranda, M. Darder, G. Rytwo, Hybrid materials based on clays for environmental and biomedical applications, J. Mater. Chem. 20 (2010) 9306. [4] M.O. Adebajo, R.L. Frost, J.T. Kloprogge, O. Carmody, S. Kokot, Porous materials for oil spill cleanup: a review of synthesis and absorbing properties, J. Porous Mater. 10 (2003) 159–170. [5] P. Bankovic´, A. Milutinovic´-Nikolic´, Z. Mojovic´, N. Jovic´-Jovicˇic´, M. Perovic´, V. Spasojevic´, D. Jovanovic´, Synthesis and characterization of bentonites rich in beidellite with incorporated Al or Al–Fe oxide pillars, Microporous Mesoporous Mater. 165 (2013) 247–256. [6] K. Ohtsuka, Preparation and properties of two-dimensional microporous pillared interlayered solids, Chem. Mater. 9 (1997) 2039–2050. [7] C.H. Zhou, J. Keeling, Fundamental and applied research on clay minerals: from climate and environment to nanotechnology, Appl. Clay Sci. 74 (2013) 3–9. [8] S. Dultz, J. Bors, Organophilic bentonites as adsorbents for radionuclides II. Chemical and mineralogical properties of HDPy-montmorillonite, Appl. Clay Sci. 16 (2000) 15–29. [9] Y.H. Shen, Phenol sorption by organoclays having different charge characteristics, Colloids Surf. A: Physicochem. Eng. Aspects 232 (2004) 143– 149.

27

[10] Y. Park, G.A. Ayoko, E. Horvath, R. Kurdi, J. Kristof, R.L. Frost, Structural characterisation and environmental application of organoclays for the removal of phenolic compounds, J. Colloid Interf. Sci. 393 (2013) 319–334. [11] S. Babel, T.A. Kurniawan, Low-cost adsorbents for heavy metals uptake from contaminated water: a review, J. Hazard. Mater. B 97 (2003) 219–243. [12] K.G. Bhattacharyya, S.S. Gupta, Adsorption of a few heavy metals on natural and modified kaolinite and montmorillonite: a review, Adv. Colloid Interf. Sci. 140 (2008) 114–131. [13] F. Fu, Q. Wang, Removal of heavy metal ions from wastewaters: a review, J. Environ. Manage. 92 (2011) 407–418. [14] J.T. Kloprogge, R. Fry, R.L. Frost, An infrared emission spectroscopic study of the thermal transformation mechanims in Al13 pillared clay catalysts with and without tetrahedral substitutions, J. Catal. 184 (1999) 157–171. [15] J.Q. Jiang, Z. Zeng, Comparison of modified montmorillonite adsorbents Part II: the effects of the type of raw clays and modification conditions on the adsorption performance, Chemosphere 53 (2003) 53–62. [16] D. Karamanis, P.A. Assimakopoulos, Efficiency of aluminum-pillared montmorillonite on the removal of cesium and copper from aqueous solutions, Water Res. 41 (2007) 1897–1906. [17] T. Kasama, Y. Watanabe, H. Yamada, T. Murakami, Sorption of phosphates on Al-pillared smectites and mica at acidic to neutral pH, Appl. Clay Sci. 25 (2004) 167–177. [18] V. Lenoble, O. Bouras, V. Deluchat, B. Serpaud, J.C. Bollinger, Arsenic adsorption onto pillared clays and iron oxides, J. Colloid Interf. Sci. 255 (2002) 52–58. [19] K.R. Srinivasan, H.S. Fogler, Use of inorgano-organo-clays in the removal of priority pollutants from industrial wastewaters: adsorption of benzo(a)pyrene and chlorophenols from aqueous solutions, Clays Clay Miner. 8 (1990) 287– 293. [20] C. Ouellet-Plamondon, R.J. Lynch, A. Al-Tabbaa, Comparison between granular pillared, organo- and inorgano–organo-bentonites for hydrocarbon and metal ion adsorption, Appl. Clay Sci. 67–68 (2012) 91–98. [21] B. Hu, H. Luo, H. Chen, T. Dong, Adsorption of chromate and paranitrochlorobenzene on inorganic–organic montmorillonite, Appl. Clay Sci. 51 (2011) 198–201. [22] L. Zhu, R. Zhu, Simultaneous sorption of organic compounds and phosphate to inorganic–organic bentonites from water, Sep. Purif. Technol. 54 (2007) 71–76. [23] R. Zhu, T. Wang, F. Ge, W. Chen, Z. You, Intercalation of both CTMAB and Al13 into montmorillonite, J. Colloid Interf. Sci. 335 (2009) 77–83. [24] R. Wibulswas, D.A. White, R. Rautiu, Adsorption of phenolic compounds from water by surfactant-modified pillared clays, Process Saf. Environ. Protect. 77 (1999) 88–92. [25] S. Al-Asheh, F. Banat, L. Abu-Aitah, Adsorption of phenol using different types of activated bentonites, Sep. Purif. Technol. 33 (2003) 1–10. [26] B. Hu, H. Luo, Adsorption of hexavalent chromium onto montmorillonite modified with hydroxyaluminum and cetyltrimethylammonium bromide, Appl. Surf. Sci. 257 (2010) 769–775. [27] L. Ma, Q. Zhou, T. Li, Q. Tao, J. Zhu, P. Yuan, R. Zhu, H. He, Investigation of structure and thermal stability of surfactant-modified Al-pillared montmorillonite, J. Therm. Anal. Calorim. (2013). [28] F. Zermane, O. Bouras, M. Baudu, J.P. Basly, Cooperative coadsorption of 4nitrophenol and basic yellow 28 dye onto an iron organo-inorgano pillared montmorillonite clay, J. Colloid Interf. Sci. 350 (2010) 315–319. [29] Z. Bouberka, A. Khenifi, H. Ait Mahamed, B. Haddou, N. Belkaid, N. Bettahar, Z. Derriche, Adsorption of Supranol Yellow 4 GL from aqueous solution by surfactant-treated aluminum/chromium-intercalated bentonite, J. Hazard. Mater. 162 (2009) 378–385. [30] O. Bouras, J.-C. Bollinger, M. Baudu, H. Khalaf, Adsorption of diuron and its degradation products from aqueous solution by surfactant-modified pillared clays, Appl. Clay Sci. 37 (2007) 240–250. [31] O. Bouras, T. Chami, M. Houari, H. Khalaf, J.C. Bollinger, M. Baudu, Removal of sulfacid brilliant pink from an aqueous stream by adsorption onto surfactantmodified Ti-pillared montmorillonite, Environ. Technol. 23 (2002) 405–411. [32] J.Q. Jiang, C. Cooper, S. Ouki, Comparison of modified montmorillonite adsorbents Part I: preparation, characterization and phenol adsorption, Chemosphere 47 (2002) 711–716. [33] O. Bouras, M. Houari, H. Khalaf, Using of surfactant modified Fe-pillared bentonite for the removal of pentachlorophenol from aqueous stream, Environ. Technol. 22 (2001) 69–74. [34] P.X. Wu, Z.W. Liaob, H.F. Zhangc, J.G. Guo, Adsorption of phenol on inorganic– organic pillared montmorillonite in polluted water, Environ. Int. 26 (2001) 401–407. [35] R. Zhu, L. Zhu, J. Zhu, F. Ge, T. Wang, Sorption of naphthalene and phosphate to the CTMAB-Al13 intercalated bentonites, J. Hazard. Mater. 168 (2009) 1590– 1594. [36] H. Khalaf, O. Bouras, V. Perrichon, Synthesis and characterization of Al-pillared and cationic surfactant modified Al-pillared Algerian bentonite, Microporous Mater. 8 (1997) 141–150. [37] Y. Xi, W. Martens, H. He, R.L. Frost, Thermogravimetric analysis of organoclays intercalated with the surfactant octadecyltrimethylammonium bromide, J. Therm. Anal. Calorim. 81 (2005) 91–97. [38] Y. Xi, R.L. Frost, H. He, T. Kloprogge, T. Bostrom, Modification of Wyoming montmorillonite surfaces using a cationic surfactant, Langmuir 21 (2005) 8675–8680. [39] D.S. Tong, C.H. Zhou, Y. Lu, H. Yu, G.F. Zhang, W.H. Yu, Adsorption of acid red G dye on octadecyl trimethylammonium montmorillonite, Appl. Clay Sci. 50 (2010) 427–431.

28

S.I. Rathnayake et al. / Journal of Colloid and Interface Science 459 (2015) 17–28

[40] Y. Xi, R.L. Frost, H. He, Modification of the surfaces of Wyoming montmorillonite by the cationic surfactants alkyl trimethyl, dialkyl dimethyl, and trialkyl methyl ammonium bromides, J. Colloid Interf. Sci. 305 (2007) 150–158. [41] S. Zhu, H. Peng, J. Chen, H. Li, Y. Cao, Y. Yang, Z. Feng, Intercalation behavior of poly(ethylene glycol) in organically modified montmorillonite, Appl. Surf. Sci. 276 (2013) 502–511. [42] Y. Park, G.A. Ayoko, R.L. Frost, Application of organoclays for the adsorption of recalcitrant organic molecules from aqueous media, J. Colloid Interf. Sci. 354 (2011) 292–305.

[43] F. Bergaya, G. Lagaly, Handbook of clay science, part A: fundamentals, Second ed., 2013. [44] H. He, Y. Ma, J. Zhu, P. Yuan, Y. Qing, Organoclays prepared from montmorillonites with different cation exchange capacity and surfactant configuration, Appl. Clay Sci. 48 (2010) 67–72. [45] W.H. Yu, Q.Q. Ren, D.S. Tong, C.H. Zhou, H. Wang, Clean production of CTABmontmorillonite: formation mechanism and swelling behavior in xylene, Appl. Clay Sci. 97–98 (2014) 222–234.