Introduction
Bentonite, consisting predominantly of smectite minerals (montmorillonite, saponite, nontronite, etc.) with a layered structure, is a hydrous aluminum and magnesium silicate with a particle size of <2 μm (Lagaly and Dekany Reference Lagaly, Dekany, Lagaly and Bergaya2013). Bentonite is used for many purposes because of its outstanding characteristics such as large surface area, high CEC, plasticity, swelling ability, biocompatibility, muco-adhesiveness, good mechanical properties, and chemical stability (Yıldız and Kuşcu Reference Yιldιz and Kuşcu2007; Nones et al. Reference Nones, Riella, Trentin and Nones2015; Chenliang et al. Reference Chenliang, Fanfei, Lingyun and Chen2017). Bentonites are generally used as drilling muds, adsorbents in separation processes, drug-delivery systems, catalysis, composite materials, lubricants in emulsions, food additives for prevention of aflatoxicosis, anti-bacterial material in dental infection, and food-packaging materials, and as a raw material in the cement, ceramics, and pharmaceutical industries (Falaras et al. Reference Falaras, Kovanis, Lezou and Seiragakis1999; Carraro et al. Reference Carraro, De Giacomo, Giannossi, Medici, Muscarella, Palazzo, Quaranta and Tateo2014; Yang et al. Reference Yang, Lee, Ryu, Elzatahry, Alothman and Choy2016; Jayrajsinh et al. Reference Jayrajsinh, Shankar, Pharm, Agrawal and Bakre2017). Raw bentonites generally contain gangue (non-clay) minerals such as quartz, opal, calcite, feldspar, zeolite, and illite. Some bentonites containing high levels of gangue minerals are classified as low-quality bentonites (Gong et al. Reference Gong, Liao, Lv and Wang2016). In some applications, such as pharmaceuticals, cosmetics, food, and nanocomposite materials, low-quality bentonites must be purified before use. Purification methods for bentonites reported in the literature have included sedimentation-centrifugation, ion exchange reaction (Shah et al. Reference Shah, Valenzuela, Ehsan, Díaz and Khattak2013), modification-centrifugation (Yeşilyurt et al. Reference Yeşilyurt, Boylu, Çinku, Esenli and Çelik2014), multistage hydrocyclone (Boylu et al. Reference Boylu, Çinku, Esenli and Çelik2010), and chemical decomposition (Veiskarami et al. Reference Veiskarami, Sarvi and Mokhtari2016). A washing process is required when the bentonites are purified by chemical purification methods, particularly for use in sensitive applications such as pharmaceuticals, food, and cosmetics. In addition, solid-liquid separation treatments (filtration, evaporation, or centrifugation) are also required which are either difficult and/or costly processes. Simpler physical methods are needed, therefore, to reduce the cost and to provide cleaner production of pure bentonite. The yield of a purification process is another important issue, and needs to be as high as possible. Purification yields have not been reported in most previous studies, however.
Industrially mined bentonites are rich in expandable clay minerals of the smectite group with negatively charged surfaces resulting in hydrophilic properties (Bergaya et al. Reference Bergaya, Theng and Lagaly2006; Moslemizadeh et al. Reference Moslemizadeh, Aghdam, Shahbazi, Aghdam and Alboghobeish2016). These characteristics mean that they resist the adsorption of anionic, neutral, or hydrophobic species (Ersoy and Çelik Reference Ersoy and Celik2004). Bentonites are generally modified with cationic surfactants to eliminate these problems, and the products prepared are called organo-bentonites (Zhou Reference Zhou2011). Quaternary alkylammonium salts such as CTAB are commonly used as cationic surfactants to prepare organo-bentonite. Organo-bentonites are widely investigated as adsorbents (Fatimah and Huda Reference Fatimah and Huda2013), fillers in polymer nanocomposites with improved mechanical properties (Bergaya et al. Reference Bergaya, Jaber, Lambert and Galimberti2011), and rheology-control agents in drilling fluids, paints, and cosmetics (Yu et al. Reference Yu, Ren, Tong, Zhou and Wang2014; Liang et al. Reference Liang, Long, Yang and Dai2015).
Wettability (or hydrophobicity) and ζ potential are two important material characteristics used in surface-modification studies. They are encountered in many areas such as mining, textile, paper, medicine, plastic, paint, and food processing (Erol et al. Reference Erol, Devrim, Ciftci, Ersoy and Cigerci2017). Wettability is measured by the contact angle between a solid surface and a drop of water. The greater the contact angle, the lower the degree of wettability. For example, a solid having a contact angle of 90o is defined as hydrophobic (non-wettable), and super hydrophobic if >150o. Materials are called hydrophilic (wettable) if the contact angle is close to zero. ζ potential is the measurable electrical charge on the surface of a particle in a liquid. The strength of the electrostatic repulsive forces between the particles and, therefore, the stability of a colloidal system are determined by ζ (Çelik Reference Çelik, Wypych and Satyanarayana2004). In addition, the adsorption capacity of a particle is heavily dependent on ζ potential.
Wettability and ζ potential are also important surface characteristics of organo-clays. Determination of the relationship between applied amounts of cationic surfactant and surface characteristics of organo-clays is necessary for some applications. In the literature, similar studies dealing with the preparation of organo-bentonites using alkylammonium salts are reported; a search of the literature revealed that wettability (contact angle) of organo-bentonites depending on the amount of CTAB was investigated in only two studies, however (Zhang et al. Reference Zhang, Li, Xu and Sun2014; Schampera et al. Reference Schampera, Šolc, Tunega and Dultz2016), and SFE in only one study (Zhang et al. Reference Zhang, Li, Xu and Sun2014). On the other hand, several studies have determined ζ profiles of organo-bentonites. Zhang et al. (Reference Zhang, Li, Xu and Sun2014) reported the neutralization point of CTAB-modified bentonite at 0.6 mmol/g (49.6% CEC) while the maximum ζ potential (~41 mV) was determined at 1.5 mmol/g (123% CEC) concentration. Barany et al. (Reference Barany, Meszaros, Taubaeva and Musabekov2015) showed that the surface of organo-bentonite was neutralized when CTAB concentration reached ~2 mmol/g, and the maximum ζ was determined as 35–40 mV at 7 mmol/g CTAB concentration. Bianchi et al. (Reference Bianchi, Fernández, Pantanetti, Viña, Torriani, Torres Sánchez and Punte2013) prepared organo-montmorillonite samples with three different CTAB concentrations, and maximum ζ was reported as ~35 mV at 100% CEC (1.74 mmol/g) in the pH range of 6.5–7.0. A remarkable decrease in ζ potential (up to 5 mV) at 200% CEC (3.48 mmol/g) concentration was also reported. Such varied or surprising results for the maximum ζ and neutralization points may be due to the use of different CEC values for of the bentonites, different methods of organo-bentonite preparation, and perhaps improper measurements obtained by the use of non-calibrated instruments. The CEC of bentonite is not taken into account when determining the CTAB concentration in most studies.
The objectives of the present study were: (1) to improve the purification of raw bentonites using simple physical methods (sedimentation and centrifugation) in order to obtain high yields; and (2) to investigate the effect of CTAB concentration in organo-bentonite on their wettability (contact angle), ζ potential, and SFE profiles, taking into account the CEC of the bentonite.
Materials and Methods
Materials
Two commercial, raw Na-bentonites were investigated in the present study. The first one (labeled as Bent-A) was collected from the mine site of Samaş Industrial Mines Company located in Reşadiye-Tokat, Turkey, and the second one (Bent-B) was collected from a mine site of Çanbensan Industry and Trade Co. Ltd. located in Kalecik-Ankara, Turkey. The manufacturers are market leaders in Turkey and these bentonites have been investigated in many studies (Boylu et al. Reference Boylu, Hojiyev, Ersever, Ulcay and Çelik2012; Yeşilyurt et al. Reference Yeşilyurt, Boylu, Çinku, Esenli and Çelik2014), including in the preparation and characterization of organo-bentonites (Yeşilyurt et al. Reference Yeşilyurt, Boylu, Çinku, Esenli and Çelik2014; Hojiyev et al. Reference Hojiyev, Ersever, Karaağaçlιoğlu, Karakaş and Boylu2016), preparation of clay-polymer nanocomposites (Hojiyev et al. Reference Hojiyev, Ulcay and Çelik2017), and adsorption behavior (Orucoglu and Haciyakupoglu Reference Orucoglu and Haciyakupoglu2015). Diiodomethane (CH2I2; 99%), used for SFE measurements, and ethanol (C2H5OH; 96%) were purchased from Merck (Boston, Massachusetts, USA), and CTAB (C19H42BrN; 95%) was purchased from Sigma-Aldrich (St. Louis, Missouri, USA).
Purification of Raw Bentonites
Raw bentonite aggregates were crushed to <5 mm particle size by using a laboratory-scale roller mill crusher. Na-bentonite is easily dispersed in water, and, therefore, no further size reduction was applied in order to avoid reduction of the particle size of gangue minerals. Clay dispersion was prepared in a 2000 mL beaker using 1980 mL of distilled water and 20 g of crushed bentonite sample. The prepared dispersion was allowed to stand for 24 h to ensure the swelling and dispersion of the clay minerals as much as possible (Fig. 1a). Subsequently, the dispersion was stirred mechanically for 30 min at 1000 rpm, and left to sediment for 8 h to allow gangue minerals to settle. The settling behavior of particles in a dispersing system is described by Stokes’ law. The necessary settling time (8 h) of gangue minerals having particle sizes >2 μm was calculated according to Stokes’ equation. Finally, the suspended dispersion was decanted at the end of the settling time and stored in a beaker. The dispersions obtained were labeled as “Bent-A-Sed” and “Bent-B-Sed” for Bent-A and Bent-B, respectively. Decantation was continued to a height of 3 cm from the bottom of the beaker to prevent the mixing of the settled minerals to the suspended dispersion as a result of turbulence (Fig. 1a).
Fig. 1 Photographs of a the Ben-A-Raw dispersion, b Bent-A-Sed, and c gangue minerals.
FGS with small modifications was used as another purification method. Experiments were carried out at 300×g centrifugal force and 1.5 L/min feed rate. Fluidizing water was not used, and the water holes in the concentrator bowl were sealed with tape during the experiments. First, the clay dispersion was prepared as in the sedimentation procedure, and then left to stand for 10 min to allow the coarse particles to settle. Subsequently, the dispersion remaining on the settled particles was decanted completely and fed once to the FGS. The products obtained from Bent-A and Bent-B in this step were labeled as “Bent-A-FGS” and “Bent-B-FGS”, respectively.
Surface Modification of Purified Bentonites (Organo-bentonite Preparation)
Organo-bentonites were prepared from Bent-A-Sed and Bent-B-Sed samples. The concentrations of CTAB were prepared by adding amounts equal to various fractions of the CEC of the purified bentonites (Bent-A: 95 meq/100 g and Bent-B: 90 meq/100 g) and ranged from 50 to 200% of CEC. As an example, for Bent-A, 100% CEC indicates that 95 mmol CTAB was added for each 100 g Bent-A purified by sedimentation. In the first step, all samples were dried for 24 h at 80oC, ground using a ring mill, and then 5 g of sample was dispersed mechanically in 500 mL of distilled water. Meanwhile, the amount of CTAB required was dissolved in 50 mL of ethanol with magnetic stirring for 30 min and the solution obtained was added dropwise to the clay dispersion. The final mixture was stirred mechanically at 750 rpm for 90 min to ensure that the adsorption and intercalation of cetyltrimethylammonium (CTA+) ions were completed. Finally, the organo-bentonite obtained was filtered, washed several times with distilled water to remove excessive ions, and then dried for 24 h at 80°C. Products obtained from Bent-A-Sed and Bent-B-Sed were labeled as Organo-Bent-A and Organo-Bent-B, respectively.
Characterization Techniques
Mineralogy and morphology.
Mineralogical analysis was performed using a Shimadzu XRD-6000 instrument with CuKα radiation (λ: 1.54184 Å) at 40 kV. The samples were scanned over the angular range 2–70°2θ with a scan rate of 0.02°2θ/step. The raw and purified samples were analyzed after drying under normal atmospheric conditions. The clay mineralogy was also determined for oriented samples and, for these three samples, were prepared for both bentonite samples. Firstly, dispersions containing clay particles were obtained from raw samples by 15 min of sedimentation at a 1.0 wt.% solid:liquid ratio. A few drops of dispersion were then dropped onto the glass slides and allowed to air dry overnight. One of the dried samples was saturated with ethylene glycol vapor at 60°C, while one was heated at 550°C for 2 h. Subsequently, XRD analyses were performed on air-dried, ethylene glycol-saturated, and heated samples (Brown and Brindley Reference Brown, Brindley, Brindley and Brown1980). A Jeol 2100F transmission electron microscope (TEM) and LEO 1430 VP scanning electron microscope (SEM) were used to examine the morphology of the bentonite samples.
Chemical and thermogravimetric analyses.
The chemical compositions of the samples were determined using a Rigaku ZSX Primus II XRF spectrophotometer. The temperature-related reactions and mass losses were recorded by DTA-TG analysis. For this purpose, Netzsch's differential scanning calorimeter was used and the analysis were carried out from room temperature to 1000°C in argon gas flow with a heating rate of 10°C/min.
Determination of cation exchange capacity (CEC) and free swelling volume (FSV).
CEC is the number of equivalents of exchangeable cations between the clay mineral layers per 100 g of clay. The CECs of bentonites were determined by the methylene blue (MB) test according to the ASTM D 5890-95.
Standard and the procedure can be summarized as follows. A 2 g of sample was dried for 24 h at 105°C and then ground using a ring mill. The sample was then added to a 600 mL beaker containing 300 mL of distilled water and the dispersion was stirred using a magnetic stirrer until complete dispersion. The pH of the prepared dispersion was adjusted to between 2.5 and 3.8 using 2 N H2SO4 solution. Stirring was then continued for 10–15 min and the pH was adjusted again if necessary. While stirring was continued, 5 mL of 10–2 M MB solution (1 mL = 0.01 meq) was added to the dispersion and stirring continued for 1–2 min. One drop of suspension was then taken using a glass rod and left on the filter paper (Sartorius 392). The last two steps were continued until formation of a light blue halo around the drop. When the halo appeared, a drop was taken from the dispersion after 2 min without adding MB, and left on the filter paper. The end point was reached if the halo was formed. If the halo was not observed, further 1-mL aliquots of MB solution were added to the dispersion until the end point was reached. The CEC was calculated using Eq. 1.
where CEC is cation exchange capacity (meq/100 g), E is the miliequivalent value for 1 mL of MB, V is the amount of methylene blue (mL) used in titration, and W is the amount of clay (g).
The MB test is used commonly for determination of the CEC of clays; some limitations have been reported for this method, however. Possible dye–dye interactions can increase the bilayer adsorption of dye cations on the surface of clay minerals. In addition, aggregation of dye molecules can occur on the external surfaces (Lagaly et al. Reference Lagaly, Ogawa, Dekany, Lagaly and Bergaya2013). The method is used widely because of its simple and fast application. In the bentonite industry, smectite contents (product qualities) are often quantified using MB adsorption; comparability is usually based on MB adsorption of a reference material with known smectite content, however. This requires assumption of a similar layer-charge density. Additionally, diluted MB solution should be added gradually to the clay/water dispersion as described above. Otherwise, clay surfaces are quickly covered by large MB molecules and clay aggregates are formed. This prevents the intercalation of the MB molecules (Taylor Reference Taylor1985; Rytwo et al. Reference Rytwo, Serban, Nir and Margulies1991; Kahr and Madsen Reference Kahr and Madsen1995).
FSV is caused by hydration of interlayer surfaces of the clay mineral, resulting in a larger total volume due to the penetration of water molecules between the layers. FSV measurements were performed according to ASTM 837 C standard. A ground sample with a particle size of <150 μm was dried at 105°C for 24 h. 90 mL of distilled water was added to a 100 mL graduated cylinder. Approximately 0.1 g of clay was added slowly to the cylinder each time. The added clay was allowed to settle for a minimum period of 10 min. The two previous steps were repeated until 2 g of dried clay sample had been added to the cylinder. Mineral particles adhering to the cylinder edges were then dispersed into the water by adding more water until the level reached 100 mL. After that, the dispersion was allowed to stand for 16 h. Finally, the volume of hydrated clay mass standing at the bottom of the cylinder was recorded by reading from the graduated scale on the cylinder.
Surface properties and particle size-distributions.
The wettability of the Bent-A-Sed and Bent-B-Sed samples loaded with various concentrations of CTAB was determined by contact-angle measurements. First, 2×25 mm discs were prepared from dried and ground powders under 50 kN pressure using a manual hydraulic press (Specac Co., London, UK). Then, 5 μL of distilled water (a drop) was placed on each disc-shaped sample using a Hamilton injector. The contact angles were then measured with a One Attension theta optical goniometer (Biolin Scientific Co., Espoo, Finland) using the sessile drop method at room temperature. The value obtained at the fifth second immediately after dropping the water drop on the sample surface was recorded as the contact angle. At least three measurements were performed for each sample and the averages of the results recorded. Contact angles of each sample were also measured using diiodiomethane to calculate SFE according to the OWRK/Fowkes method (Fowkes Reference Fowkes1962; Owens and Wendt Reference Owens and Wendt1969; Kaelble Reference Kaelble1970). Once the contact angles for water and diiodiomethane were determined, the SFE calculations were performed automatically by the instrument.
Measurements of ζ were obtained using a Malvern Nano-Z zeta-sizer instrument (Malvern Panalytical Co., Malvern, UK) at room temperature. Dried and ground samples were added to 10–3 M NaCl solutions at 0.01 wt.% solid:liquid ratio, and dispersed for 30 min using a magnetic stirrer. Subsequently, stirring was stopped and the dispersion was left for 5 min to allow coarse particles to settle. One-mL aliquots were then taken from the dispersions and placed in the measuring cell. ζ values were calculated automatically by the instrument software using the classical Smoluchowski formula and the measured electrophoretic mobility values of particles. The measurements were repeated three times and the results having minimum standard deviations (3–5%) were recorded as ζ values.
Particle-size distributions were determined using a Malvern Mastersizer 2000 instrument (Malvern Panalytical Co., Malvern, UK). Dispersions obtained from sedimentation and FGS were used for analysis. Firstly, a 1000 mL beaker containing 800 mL of distilled water was placed in the pump section of the wet analysis apparatus, stirring was started at 2000 rpm, and then the background was measured. Subsequently, an appropriate amount of each sample was homogenized for 1 min at 400 Watt using a Hielscher UP400S sonicator (Hielscher Ultrasound Technology Co., Teltow, Germany) and added slowly to the beaker until the laser obscuration reached 3–6%. Finally, measurements were started and the average values were recorded by repeating each analysis three times.
RESULTS AND DISCUSSION
Characterization of Purified Bentonites
Smectites (montmorillonite, nontronite, saponite, etc.) usually exhibit a morphology that resembles honeycomb, cornflake, leafy, and rosette-like appearance (Keller et al. Reference Keller, Reynolds and Inoue1986). SEM images (Fig. 2a, c) revealed that the bentonite samples consisted of grain aggregates with smooth surfaces, and have leaf-layer morphology with its curved edges. Transmision election microscopy images (Fig. 2b, d) showed that the samples have a multi-layered structure consisting of successive sheets with parallel and flat surfaces.
Fig. 2 a, c SEM and b, d TEM images of the bentonites.
The XRD patterns of the raw bentonites (Fig. 3) revealed that the samples consisted of smectites, traces of kaolinite (in Bent-B), and gangue minerals such as feldspar, calcite, quartz, opal (C-T), clinoptilolite (zeolite), and mica. Bent-A samples other than organo-bentonites showed characteristic reflections (7.07o2θ) with the d 001 basal spacing of 12.4 Å confirming Na+-rich smectites (Brown and Brindley Reference Brown, Brindley, Brindley and Brown1980; Hayakawa et al. Reference Hayakawa, Minase, Fujita and Ogawa2016). Bent-B showed a similar but broader reflection (7.01o2θ) with a d 001 basal spacing of 12.6 Å (Fig. 3b). Additionally, gangue mineral (especially feldspar) contents were greater in Bent-B when considering the reflection intensities. The d 060 values (1.5 Å) of the smectites showed that both bentonites were composed of dioctahedral smectites (montmorillonite, nontronite, beidellite) (Grim Reference Grim1968). Almost all of the gangue minerals were removed from the Bent-A-Raw by sedimentation and FGS. The result for Bent-B was slightly different, and small amounts of quartz and kaolinite remained in both Bent-B-Sed and Bent-B-FGS products as a result of the suspended particles having particle sizes <2 μm. Of course, such small particles may have also remained in the Bent-A-Sed and Bent-A-FGS products, but in very small amounts below the detection limit of XRD. Considering the purification levels, XRD data showed no significant difference between the sedimentation and FGS techniques for both bentonites. The XRD patterns of the oriented mounts of the two bentonites (Fig. 4) revealed that the d 001 basal spacing of both smectites expanded to 16.9 Å after ethylene glycol (EG) treatment. On the other hand, it collapsed to 9.7 Å for Bent-A and 10.0 Å for Bent-B after heating at 550°C. The differences in basal spacings after heating can be attributed to the exchangeable cation diversity between the smectite layers. The XRD results obtained from oriented samples proved that both samples are dominated by smectite minerals. The change in the basal spacings of smectites of oriented samples is based mainly on the type of the exchangeable cations and magnitude and localization of the layer charge (Harward and Brindley Reference Harward and Brindley1965). No chemical pre-treatment was performed other than dispersion in deionized water. Basal reflections of air-dried samples at 12.4–12.6 Å indicate the dominance of Na+ over other exchangeable cations. The broader peak of Bent-B (Fig. 4b) indicates the presence of significant amounts of bivalent Ca2+. Additionally, both EG-treated samples showed d 002 and d 003 values of 8.5 Å and 5.6 Å, which confirmed that the smectites do not contain significant amounts of interstratified minerals (compare e.g. Kaufhold et al. Reference Kaufhold, Chryssikos, Kacandes, Gionis, Ufer and Dohrmann2019)
Fig. 3 XRD patterns of the raw, purified, and organo-bentonites: a Bent-A, and b Bent-B.
Fig. 4 XRD patterns based on oriented mounts of the bentonites: a Bent-A, b Bent-B. AD: air dried, EG: ethylene glycol-treated, and 550: heated at 550oC.
The CEC and FSV data of the raw and purified bentonites (Table 1) revealed that Bent-B-Raw has smaller CEC and FSV values than Bent-A-Raw, indicating greater gangue-mineral contents in the Bent-B-Raw sample. After purification treatments, the CECs of both samples reached similar values. A significant difference was observed between the FSVs of the samples even after purification treatments, and Bent-A samples presented quite large FSVs when compared to Bent-B. This can be explained by the smaller Na+ content of Bent-B-Sed (Na2O = 1.03 wt.%; Table 2) compared to Bent-A-Sed (Na2O = 2.02 wt.%; Table 2) indicating a much larger exchangeable Na content of smectite in Bent-A. This is confirmed by a larger CaO content of Bent-B-Sed (1.04 wt.%; Table 2) compared to Bent-A-Sed (0.74 wt.%; Table 2). CEC and FSV data agreed well with the XRD results, and also successful purification of the raw bentonites.
Table 1 CEC and FSV values of the raw and purified bentonites.
Table 2 Chemical analysis (XRF) data of the raw and purified bentonites.
LOI: loss on ignition
The XRF analysis data (Table 2) showed that the samples had large percentages of silicon and aluminum oxides (≥80 wt.%), indicating the presence of a clay mineral. (Na2O+K2O)/(CaO+MgO) ratios were determined to be 0.67 and 0.58 for Bent-A-Sed and Bent-B-Sed, respectively, and values >0.33 can be attributed to Na-smectites. Ca2+ content in samples decreased significantly after removal of calcite minerals by purification treatments which are in good accordance with XRD data. Compared with that of the raw bentonite, a strong decrease in Na2O content after purification can be explained by the removal of Na-rich feldspar (anorthoclase) minerals.
Particle-size distributions (Fig. 5) revealed that raw bentonites contain large particles, >10 μm, which can be attributed to non-clay (gangue) minerals. Particle-size distributions of all purified samples were the range 0.3–10 μm. This distribution comprises predominantly pure clay minerals. Shah et al. (Reference Shah, Valenzuela, Ehsan, Díaz and Khattak2013) also reported similar results and determined the particle-size distribution of bentonite purified by sedimentation to be between 0.3 and 7.5 μm. The d 10 , d 50 , and d 90 values (Fig. 5 and Table 4) show particle sizes, and 10%, 50% and 90% of the sample amount are smaller than these particle sizes, respectively. For example, d 90 of Bent-A-Sed means that 90% of the sample consists of particles with a size. The majority of the purified bentonite particles have a particle size <5 μm and puts them close to the clay particle-size range (Lagaly and Dekany Reference Lagaly, Dekany, Lagaly and Bergaya2013). The minimum particle size for all samples was measured to be ~0.3 μm. Sedimentation products have narrower particle-size distributions with little difference when compared to FGS products. The data also showed that the difference between the particle size of clay and non-clay minerals is the main effect on the purification methods used. The particle-size analysis was not performed for organo-bentonites because of agglomeration of particles.
Fig. 5 Particle-size distributions of the raw and purified bentonites.
Higher yields for Bent-A and Bent-B (81.3 wt.% and 68.8 wt.%, respectively) were obtained by FGS compared to the sedimentation technique (Table 3), while the purification levels were almost the same for both techniques. Greater gangue-mineral content in the Bent-B sample resulted in lower yield values, which is in accordance with the XRD, CEC, and XRF results.
Table 3 Yields of the purified bentonite products according to separation techniques.
Table 4 Particle-size distributions of the raw and purified bentonites.
Table 5 Summarized surface characteristics of the samples.
Characterization of Organo-bentonite
The XRD patterns of organo-bentonites prepared by CTAB concentration of 100% CEC (Fig. 3) showed that the basal spacing expanded to 19.3 Å (2θ: 4.58o) for Bent-A and 18.9 Å (2θ: 4.69o) for Bent-B as a result of the successful intercalation of CTA+ ions into the interlayer spaces of smectites by the cation exchange process. Similar results for organo-bentonites treated with CTAB at a concentration of 100% CEC were also reported by other research teams (Yılmaz and Yapar Reference Yιlmaz and Yapar2004; Haloi et al. Reference Haloi, Goswami and Das2013). Lee and Kim (Reference Lee and Kim2002) reported the formation of a bilayer arrangement of CTA+ ions (17–18 Å) between bentonite layers when loading from 50% to 150% CEC concentration.
The DTA-TG curves of raw, purified, and organo-modified (at 100% CEC concentration) Bent-A samples revealed an intense endothermic reaction in raw bentonite at 150°C (Fig. 6). This is due to removal of adsorbed water on the external and interlayer surfaces of the smectites (Velde Reference Velde1992). The second endothermic peak observed at 700°C was due to dehydroxylation of the smectite layers. The endothermic peak at 900°C was attributed to phase transformations and degradation of the clay mineral structure (Bulut et al. Reference Bulut, Chimeddorj, Esenli and Çelik2009).
Fig. 6 DTA-TG curves of a the Bent-A-Raw, b Bent-A-Sed, and c Organo-Bent-A.
Further analysis of the raw bentonite TG curve (Fig. 6a) found that the first mass loss was ~7–8% between 80 and 170°C, and the second mass loss was ~4% between 625 and 725°C temperature ranges. The Bent-A-Sed sample compared to Bent-A-Raw showed a lower mass loss of ~3% between 80–170°C and this may be due to the small moisture content. The mass loss was approximately the same in the second step, and this may be due to the small amount of gangue minerals (especially calcite) in the sample.
The DTA curve of organo-bentonite showed (Fig. 6c) the first small endothermic peaks at ~90oC and ~150°C which are attributed to the release of free water molecules due to evaporation. The second endothermic peak at ~310°C is related to decomposition of CTA+ ions and their micelles adsorbed on the surface of the clay (Houhoune et al. Reference Houhoune, Nibou, Chegrouche and Menacer2016). The third peak at ~420°C is attributed to the decomposition of CTA+ molecules intercalated in the interlayer spaces of the smectites (Majdan et al. Reference Majdan, Pikus, Gajowiak, Sternik and Zięba2010). The last two endothermic peaks at ~600°C and ~900°C are associated with dehydroxylation of clay layers and phase transformations of clay minerals, respectively. The TG curve of organo-bentonite (Fig. 6c) showed an additional weight loss (21.7%) between 200oC and 640°C which can be attributed to the degradation of cationic surfactants on the external surfaces and in the interlayer spaces of the smectites.
The ζ profiles of purified bentonites (Fig. 7 and Table 5) revealed that the surface electro kinetics of the bentonites is heavily dependent on CTAB concentration. The 0% CEC points indicate the ζ of the CTAB free purified bentonites, and the measured values (Bent-A-Sed; -38.7 mV and Bent-B-Sed; –35.7 mV) are in good agreement with the literature (Barany et al. Reference Barany, Meszaros, Taubaeva and Musabekov2015; Zhang et al. Reference Zhang, Li, Xu and Sun2014; Barany et al. Reference Barany, Meszaros, Taubaeva and Musabekov2015). Bent-A and Bent-B displayed similar profiles and the ζ of both moved dramatically towards positive values with increasing CTAB concentration. This can be explained by the fact that CTA+ ions were also adsorbed on clay surfaces as they intercalated into the interlayer spaces. In addition, adsorption of CTA+ ions on the external surface is easier and faster than intercalation in the early stages of treatment. The surfaces of organo-bentonites were neutralized at concentrations of 78–86% CEC (0.68–0.77 mmol/g). Maximum values of ζ of Bent-A and Bent-B were measured as 39.8 mV and 38.5 mV at 140% CEC (1.23 mmol/g) and 120% CEC (1.08 mmol/g) concentrations, respectively. No significant change in ζ was observed in either bentonite with more than 120% CTA+ applied. At this applied concentration, monolayer adsorption on the surface was completed, but intercalation continued. In addition, partial bilayer adsorption or micelle formation of the CTA+ ions could have occurred on the surface. The pH of the dispersions prepared for ζ analysis of raw bentonite and organo-bentonite were 9.6 and 7.1, respectively (Fig. 7). The pH of the distilled water was 5.7. The CTAB concentration did not have a significant effect on pH.
Fig. 7 Zeta potential (ζ) profiles of bentonites loaded with CTA+ at various concentrations.
The wettability of the purified bentonites was investigated by measuring the contact angles with distilled water. Significant changes of the contact angle were observed depending on the CTAB concentration, ranging from super hydrophilic to partial hydrophobic (Fig. 8 and Table 5). This is attributed to the neutralization of Lewis base sites (O atoms on the surface) by adsorption of organic cations (Jai Prakash Reference Jai Prakash, Wypych and Satyanarayana2004). The ζ and contact angles initially change rapidly because of the rapid adsorption of the large CTA+ ions onto the external surfaces; intercalation into the smectites interlayer spaces is initially more difficult until some layer expansion occurs. Intercalation is then easier and takes less time. The 0% CEC point indicates the contact angles of the CTAB free Bent-A-Sed and Bent-B-Sed samples to be 12.7° and 14.1°, respectively. Low contact angles (high hydrophilicty or low hydrophobicity) result from the hydration of exchangeable cations. All clay minerals except the uncharged minerals talc and pyrophyllite exhibit such behavior when they come into contact with liquids (Grim Reference Grim1968). The maximum contact angle or Bent-A-Sed was measured to be 71.9o at the CTAB concentration of 110% CEC while it was 72.4o at 80% CEC for Bent-B-Sed. Schampera et al. (Reference Schampera, Šolc, Tunega and Dultz2016) reported the contact angle of Wyoming bentonite (MX-80) modified by cetyltrimethylammonium chloride (CTACl) as being 105o and 52o at 40% and 200% CEC concentrations, respectively. Zhang et al. (Reference Zhang, Li, Xu and Sun2014) determined the maximum contact angle for organo-bentonite to be 105o. The two teams have also determined the decrease in contact angle after reaching a plateau with further increase in CTAB concentration. Different contact angle values obtained from various studies can be caused by variations in the organo-bentonite preparation and on which contact-angle measuring methods were employed. The increase in contact angle is caused by the hydrophobic character of the hydrocarbon tails which are directed away from the smectite surface as a result of adsorption of the polar heads of CTA+ ions on the surface. The contact angles of both samples decreased slightly after 100% CEC concentration, which may be explained by the formation of bilayer adsorption or micelles on the external surfaces after the active sites are completely occupied by CTA+ ions.
Fig. 8 Contact angle (Bent-A-CA, Bent-B-CA) and surface free energy (Bent-A-SFE, Bent-B-SFE) profiles of bentonites loaded with CTA+ at various concentrations.
The surface free energy (γi) of a solid surface (i) is theoretically equal to half of the cohesive energy (ΔG ii) that holds both surfaces together (Jai Prakash Reference Jai Prakash, Wypych and Satyanarayana2004). γi is the sum of the free energy components of Lifshitz-van der Waals (γi LW) and Lewis acid-base (γi AB) interactions as shown in Eq. 2 (van Oss et al. Reference van Oss, Chaudhury and Good1988). γi LW is the surface free energy component emerging with van der Waals (London, Debye, and Keesom) forces resulting from apolar interactions. γi AB is a surface free energy component that results from polar interaction forces on the surface capable of receiving electrons (Lewis acid) or electron-donating (Lewis base).
Interactions in the solid–liquid interface have not yet been fully expressed mathematically and, therefore, the methods of calculating SFE using the contact angle data have different assumptions. As a result, the measured values of the SFE of solids may be different from each other according to the methods used.
In the present study, as expected, the measured SFEs of purified and organo-bentonites exhibited a trend in contrast to the contact angle profiles (Fig. 8). The SFEs of the CTAB-free purified Bent-A and Bent-B were determined to be 78.2 and 78.6 mJ/m2, respectively. SFEs decreased significantly with CTAB concentration increasing up to 80–100% CEC. The SFEs of both samples were then changed slightly and fixed. Minimum SFEs were calculated to be 43.6 mJ/m2 for Bent-A and 41.5 mJ/m2 for Bent-B at 100% CEC concentration. Zhang et al. (Reference Zhang, Li, Xu and Sun2014) also reported a similar SFE profile for organo-bentonites according to the CTAB concentration, giving a minimum SFE of ~35 mJ/m2. As a result, solids having high SFE, such as bentonite, can be wetted by most liquids, and vice versa. In addition, the stabilization of organo-clays is very limited, while unmodified clays can easily form stable suspensions with water (hydrophilic character).
Conclusions
The raw bentonites were purified by sedimentation and FGS separation techniques, both of which techniques are based on differences in particle sizes of clay and non-clay minerals. While small amounts of quartz could not be removed from the Bent-B sample, no gangue (non-clay) mineral in purified Bent-A samples was detected by XRD. Separation of the <5 μm gangue minerals from clay minerals was quite difficult for both techniques under the conditions used. Modified FGS would be a better choice for purifying raw bentonites in a much shorter time and at higher yields; separation of <5 μm gangue minerals from clay minerals is not possible using these techniques, however.
X-ray diffraction and surface characterization analyses indicated that CTA+ ions were adsorbed on external surfaces as well as intercalated into the interlayer spaces of smectites. Zeta potential and contact-angle measurements showed that the electrodynamic and chemical surface properties of clay particles can be altered significantly by adsorption of surfactant molecules on the clay surfaces. Negative surface charges of Na-smectites were converted to positive values up to a certain CTAB concentration. ζ reached a plateau at 39.2 mV (at 130% CEC concentration) and 38.5 mV (at 120% CEC conc.) for Bent-A-Sed and Bent-B-Sed, respectively. No significant change in ζ was then recorded for either bentonite with further increase of CTAB concentrations. Loading of the clay minerals with CTA+ molecules resulted in the formation of a partially hydrophobic surface with maximum contact angles of ~74.3o (at 100% CEC concentration) and 72.4o (at 80% CEC concentration) for Bent-A-Sed and Bent-B-Sed, respectively. Slight decreases in contact angle were seen at higher concentrations. The CTAB concentration of ~100–130% CEC is, therefore, sufficient to prepare organo-bentonites with maximum hydrophobicity and positively charged surfaces.
ACKNOWLEDGMENTS
This study was supported financially by Afyon Kocatepe University, Scientific Research Project (BAP), Project No: 18.FEN.BIL.06 and 18.KARIYER.107. The authors are grateful to the journal's editors and reviewers.
