Influence of hydrophilicity on adsorption of caffeine onto montmorillonite

Abstract

Some types of montmorillonite containing different interlayer ions were prepared and the changes in the interlayer spacings, the hydrophilicity, and the characteristics of adsorption of caffeine in solution were observed. Ion exchange treatments were performed using Li, Na, K, Rb, Cs, Mg, Ca, Sr, or Ba. As a result, Li- and Na-type montmorillonite showed larger interlayer distance (1.31–1.53 nm), than K, Rb, and Cs-type montmorillonite (1.23–1.26 nm). In the measurement of hydrophilicity using a pulse NMR-based particle interface analyzer, Li- and Na-type montmorillonite showed higher hydrophilicity. In addition, K_Lang, which indicates the interaction with caffeine, was 0.25–0.32 l/mmol, which is lower than K-, Rb-, and Cs-type montmorillonite (1.14–1.60 l/mmol). It is possible that adsorption of water molecules inhibits caffeine from adsorbing. Because of the difficulty of exchange between caffeine and water molecules in interlayer of the Li- and Na-type montmorillonite, the interaction with caffeine decreased. Alternatively, another possibility is that when highly hydrophilic montmorillonite retains many water molecules, the caffeine adsorption sites are blocked by water molecules. In either case, hydrophilicity has a large influence on the adsorption of caffeine onto montmorillonite.

Keywords

Montmorillonite caffeine adsorption hydrophilic ion exchange

Introduction

Montmorillonite (Mont) is one of smectite clay minerals mined naturally at many places in the world. Mont has been known to have various properties such as swelling, thickening, adsorptivity, and plasticity. Furthermore, because of its low cost and high stability, Mont is an industrially useful material and has been used in various industries including agriculture, civil engineering, pharmaceuticals, cosmetics, and food (Geng et al., 2013; Lee and Fu, 2003). Mont has a lamella structure in which three layers mainly composed of silica and aluminum repeat. In a silica or aluminum layer isomorphous replacement of Si by Al or of Al by Mg or Fe occurs and the layers then bear permanent charges called layer charges. Layer charges produced by isomorphous replacement are electrically neutralized and stabilized by holding exchangeable cations in the interlayer spaces. Such exchangeable cations are also called interlayer ions. Because the interlayer ions can be easily exchanged, Mont has been used as a cation exchange material (Karamanis and Assimakopoulos, 2007; Sylvester et al., 1999). Furthermore, techniques that impart further functions to Mont by exchanging interlayer ions for a compound having a desired function has been generally used. An example is controlling an affinity for organic solvents or viscosity by making Mont containing organoammonium ions (Jordan and Organophilic Bentonites, 1949; Lagaly, 1979).

Various studies on the adsorptivity of Mont have been reported and studies of adsorption of inorganic ions, such as Pb and Cd, and organic compounds, such as amines and pyrazines, have been conducted (Abollino et al., 2003; Barbier et al., 2000; Ogawa et al., 1989, 1991; Theng, 1974). Cesium adsorptivity, in particular, has been receiving much attention in recent years. Several studies on such behavior, including the adsorption of radioactive ¹³⁷Cs in soil and the use of smectites to remove it have been reported (Sato et al., 2013; Yamaguchi et al., 2012). It is thought that because smectites, including Mont and saponite, contain exchangeable cations, adsorption of these inorganic ions is mainly caused by cation exchange reactions. On the other hand, two reaction mechanisms are proposed for the adsorption of organic compounds. One is a reaction caused by a dipole moment, which occurs when an adsorbate is a nonpolar molecule and the other is a cation exchange reaction that occurs when an adsorbate is an ionic organic cation (Ogawa et al., 1995, 2003). Studies in which the adsorbate is an amine or imine suggest that electrons localized on nitrogen atoms in organic compounds or specific functional groups such as aromatic rings interact with and adsorb onto oxygen atoms at the surfaces of the clay (Bradley, 1945; MacEwan, 1948). In the case of compounds having CN bonds such as acrylonitrile, it is reported that a dipole moment exists between CN groups and interlayer ions (Yamanaka et al., 1974, 1975). In the case of some organic compounds such as aminotriazole, it is suggested that not only is adsorption caused by dipole moments but also adsorption can be due to interactions between amino groups and interlayer ions of Mont and the resultant formation of complexes (Morillo et al., 1997; Serratosa, 1968). The sum of these studies show that not only interactions with the clay structure but also interactions with interlayer ions has an influence on the adsorption characteristics of organic compounds onto Mont and that there is a possibility that control of adsorption characteristics can be achieved by controlling the interlayer ions.

When focusing on adsorption of compounds in solution, water molecules are considered to be one of important factors affecting adsorption. In the case of the interaction between Mont and water molecules, in addition to interaction of water with the Mont structure itself, interaction between intercalated cations and water molecules has to be taken into consideration. According to a report by Schwarzenbach et al. (2002) investigating the interaction between the surface of a Mont layer and water molecules, the free energy of water molecules adsorbed onto oxygen atoms on the surface of clay or onto highly polar functional groups, typified by a hydroxy group, is higher than free energy of water molecules adsorbed onto organic compounds such as alcohols and ketones. Hence, when a molecule other than water molecules adsorbs onto Mont onto which water molecules have already been adsorbed, an energy higher than adsorption energy of water molecules is required. Furthermore, the interaction between interlayer ions and water molecules is expected to be largely affected by the properties of the intercalated cations. Although a study by Norrish (1954) has been reported which showed that swelling property of Mont varies according to interlayer ions, there have been a limited number of reports focusing on the hydrophilicity of Mont.

A pulse nuclear magnetic resonance (NMR) technique has been used to evaluate the particle size and surface area of clay minerals and the state of a specific adsorbed compound (Aepuru and Panda, 2014; Karpovich et al., 2016; Laxman et al., 2015; Sapsford et al., 2011). When Mont suspended in water is measured by the pulse NMR method, the relaxation time of water molecules is obtained and the number of water molecules adsorbed can be evaluated quantitatively. As the pulse NMR technique enables us to measure a property of an adsorbent onto which water adsorbs, that is hydrophilicity, the present study employed pulse NMR to investigate the hydrophilicity of Mont.

In this study, we selected caffeine as the target adsorbate to discuss the adsorptivity of Mont. Caffeine (Figure 1) is a weak-basic purine base extracted from plants such as tea and coffee. Caffeine has been used in various fields such as food and pharmaceuticals and also used as a reagent chemical due to its stability and preservability. Therefore, a technique to separate caffeine from solid or solution containing caffeine and purify it industrially and efficiently has been desired. In addition, caffeine is contained in wastewater in recent years and selective removal techniques of caffeine have been studied as measures to control environmental pollution (Cabrera-Lafaurie et al., 2012, 2015; Kowalska et al., 1994).

Figure 1.

Molecular structures of caffeine.

In this study interlayer ions of Mont were exchanged with alkali metal cations or alkaline earth metal cations and their influence on the hydrophilicity of Mont and adsorption characteristics of caffeine was investigated. Because highly hydrophilic Mont adsorbs many water molecules, much energy will be required for the desorption of water molecules and subsequent adsorption of other organic compounds. On the other hand, low hydrophilicity Mont, which adsorbs fewer water molecules, is expected to more easily adsorb organic compounds. Thus, we proposed and examined a hypothesis that enhancement of hydrophilicity prevents Mont from adsorbing organic compounds.

Materials and methods

Cation exchange of Mont

The Mont used in this study is produced in Yamagata Prefecture (Mizusawa Industrial Chemicals, Ltd), which samples were not purified before this treatment. The reagents used are from Wako Pure Chemical Industries, Ltd, unless otherwise specified. Cation exchange was conducted by adding Mont to an ionic solution containing a specific cation. Table 1 shows the concentration and pH of the ionic solutions used for cation exchange. Each ionic solution was prepared using a chloride salt and the concentration was adjusted so that the ionic strength was more than the cation exchange capacity of Mont (CEC: 91.5 cmol(+)/g). To 3.0 g of Mont 30 ml of ionic solution was added and the solution was shaken gently for 30 min. The solution was centrifuged at 3420 g for 10 min (KUBOTA 5930, KUBOTA Corporation) to separate solid and liquid. To the pellet obtained, additional ionic solution was added again and the same method was repeated four times to obtain a pellet. To the pellet obtained, 30 ml of ultrapure water was added to suspend the pellet. After complete suspension, the suspension was centrifuged at 3420 g for 10 min to again separate solid and liquid. To the pellet obtained, 30 ml of ultrapure water was added again and the same method was repeated again to wash the pellet. Then the pellet was washed once with ethanol once and then once with acetone. The pellet washed was air dried at 40℃ for a day. The solid obtained was pulverized into powder by a ceramic mill and ion-exchanged Mont powder was obtained. Ion-exchanged Mont obtained as above will be described hereafter as X-Mont, where X denotes the interlayer ions. In the subsequent analyzes, X-Mont was used without purification.

Table 1.

Cations and ionic solution examined.

Cation	Ionic solution
Cation	Concentration (mol/l)	pH
Lithium	1.0	5.54
Sodium	1.0	5.52
Potassium	1.0	5.53
Rubidium	0.10	5.30
Cesium	1.0	6.19
Magnesium	1.0	9.32
Calcium	1.0	6.27
Strontium	1.0	4.97
Barium	0.050	5.40

Measurement of the amount of interlayer ions in ion-exchanged Mont

Decomposition of Mont with using a microwave was conducted by a partially modified method based on that of Imai (1987) and Torigai et al. (1997). Ion-exchanged Mont (100 mg) was taken in a resin container and hydrofluoric acid (1 ml, Ultrapur, Kanto Chemical), hydrogen peroxide (2 ml, Ultrapur, Kanto Chemical), and nitric acid (7 ml, PlasmaPURE, GL Sciences) were added to the container in that order. After being slightly stirred, the solution was treated with a 2450 MHz microwave (TOPwave, analytik jena) to dissolve the material at 220℃ for 22 min. The solution obtained was diluted with nitric acid to an appropriate concentration and used as a sample for elemental analysis by inductively coupled plasma atomic emission spectrometry (ICP-AES, iCAP 6500Duo, Thermo) and inductively coupled plasma mass spectrometry (7500 ce with Octopole Reaction System, Agilent). Conditions of ICP-AES analysis were as follows: Nebulizer: Miramist, RF power: 1150 W, Coolant gas flow: 12 l/min, auxillary gas flow: 0.5 l/min, nebulizer gas flow: 0.65 l/min, plasma view: axial. Detection wavelengths for respective elements were set as follows: Li: 670.7 nm, Na: 818.3 nm, K: 766.4 nm, Rb: 780.0 nm, Mg: 285.2 nm, Ca: 317.9 nm, Sr: 421.5 nm, Ba: 233.5 nm, Rh: 343.4 nm. Conditions of ICP-AES analysis were as follows: nebulizer: microflow PFA, interface: platinum sampling cone, platinum skimmer cone with copper base, RF generator: 1500 W, carrier gas flow: 0.75 l/min, makeup gas flow: 0.3 l/min, sampling depth: 8 mm, nebulizer pump: 0.1 rps, reduction gas (H₂) flow: 4.5 ml/min. The target element was ¹³³Cs. The elements used for cation exchange were measured and their concentrations were quantified based on calibration curves constructed from internal standards. The internal standard used for quantification is Rh or Pt and calibration curves were prepared in the range from 0.001 to 5 mg/l. The amount of interlayer ions contained in ion-exchanged Mont was calculated based on concentration of each element in the sample solution.

Structural analysis of ion-exchanged Mont

Powder XRD patterns of ion-substituted Mont were obtained by a Rigaku RINT-Ultima PC diffractometer (monochromatic Cu Kα radiation), operated at 40 mA and 40 kV. The XRD measurement was done at room temperature, and the humidity was not adjusted during measurement. The basal spacing was calculated on the basis of d001 diffraction peaks and Bragg’s (1929) law.

Caffeine adsorption experiment

Ion-exchanged Mont (400 mg) was reacted with caffeine solution (40 ml) at room temperature for 1 h with gentle shaking. The concentration of caffeine solution was prepared in the range from 0.0515 to 25.7 mmol/l. After that, the solution was centrifuged at 3420 g for 10 min to obtain a supernatant which was filtered through a membrane filter (Millex GP 0.22 µm, PES, Millipore). The filtrate was diluted with ultrapure water to an appropriate concentration to be used for the analysis of caffeine. This analysis used high performance liquid chromatography coupled with UV–Vis spectroscopy (HPLC–UV/Vis, Prominence 20 A series, Shimadzu Corporation). The amount of sample solution injected was 10 µl and the absorbance was measured at a wavelength of 275 nm. Separation was conducted with using ODS columns (Shim-pack FC-ODS, 75 X 4.6 mm, Shimadzu Corporation) at 40℃. Solutions of 0.1 v/v% formic acid (A) and 50:50 (v/v) methanol and acetonitrile (B) were used as mobile phases. Gradient conditions were 90% A and 10% B from 0 to 10 min, 10% A and 90% B from 10 to 15 min, and 90% A and 10% B from 15 to 20 min. The flow rate was 1.0 ml/min. From the measured values, the equilibrium concentration of caffeine (C_eq: mmol/l) in the caffeine solution after reacting with ion-exchanged Mont was obtained. From the caffeine concentration in the caffeine solution before reacting with ion-exchanged Mont and C_eq, the amount of caffeine adsorbed onto ion-exchanged Mont per unit weight (Q_eq: mmol/g) was calculated and the maximum adsorption capacity (Q_Max) and equilibrium constants (K_Lang) were then calculated with using Langmuir adsorption isotherm (equation (1))

Q_{eq} = \frac{K_{Lang} \cdot Q_{max} \cdot C_{eq}}{1 + K_{Lang} \cdot C_{eq}}

(1)

Measurement of relaxation time coefficients

The hydrophilicity of ion-exchanged Mont was measured using a pulse NMR-based particle interface analyzer (Acorn area, XiGo Nanotools). The transverse relaxation time (T₂) was measured by the Carr-Purcell-Meiboom-Gill method (Beek et al., 1991). Ion-exchanged Mont (100, 500, and 1000 mg) was suspended in ultrapure water (10 ml) and fully mixed. A proper amount of the suspension was taken in glass tubes as samples for the measurement. The sample was placed in the analyzer until the temperature of the sample reached equilibrium and was then shaken to mix fully. After that, the measurement of the transverse relaxation time was performed. The measurement was performed three times and the average was taken. Ultrapure water was used as a blank sample. The coefficients of the relaxation time R_2sp were calculated from the relaxation rate of samples (R₂) and relaxation rate of the blank sample (R_2b) (equation (2))

R_{2 sp} = \frac{R_{2}}{R_{2 b}} - 1

(2)

Results and discussion

Figure 2 shows the XRD patterns of ion-exchanged Mont. Figure 3 shows the adsorption isotherms of ion-exchanged Mont. Table 2 shows the amount of interlayer ions, interlayer spacings, and Langmuir parameters of ion-exchanged Mont analyzed. The amounts of Na, K, Mg, and Ca contained in the interlayer space of raw Mont before ion exchange treatment were measured, resulting in 0.061 mmol/g for Na, 0.036 mmol/g for K, 0.18 mmol/g for Mg, and 0.19 mmol/g for Ca.

Figure 2.

XRD patterns of ion-exchanged montmorillonite: XRD patterns of (a) untreated montmorillonite, (b) montmorillonite intercalating alkali metal ions (Li, Na, K, Rb, Cs-Mont), (c) montmorillonite intercalating alkaline earth metal ions (Mg, Ca, Sr, Ba-Mont). Mont: montmorillonite; XRD: X-ray diffraction.

Figure 3.

Caffeine adsorption isotherms of ion-exchanged montmorillonite. (a) Untreated montmorillonite, (b) montmorillonite intercalating alkali metal ions (Li, Na, K, Rb, Cs-Mont), (c) montmorillonite intercalating alkaline earth metal ions (Mg, Ca, Sr, Ba-Mont). Mont: montmorillonite.

Table 2.

Analysis result of ion-exchanged montmorillonite.

X-Mont	Amount of interlayer ions	d001	Interlayer spacings	Q_Max	K_Lang	r²
X-Mont	(mmol/g)	(°)	(nm)	(mmol/g)	(L/mmol)	r²
Untreated	–	5.72	1.55	0.50	1.40	0.971
Li	0.64 ± 0.02	5.76	1.53	0.56 ± 0.11	0.25 ± 0.05	0.949–0.983
Na	0.56 ± 0.13	6.74	1.31	0.62 ± 0.26	0.32 ± 0.13	0.924–0.985
K	0.39 ± 0.01	7.10	1.25	0.50 ± 0.13	1.21 ± 0.36	0.860–0.935
Rb	0.31 ± 0.05	7.16	1.23	0.40 ± 0.15	1.60 ± 0.74	0.864–0.963
Cs	0.54 ± 0.02	7.04	1.26	0.37 ± 0.07	1.14 ± 0.25	0.867–0.925
Mg	0.70 ± 0.22	5.60	1.58	0.55 ± 0.06	0.87 ± 0.12	0.951–0.982
Ca	0.25 ± 0.01	5.60	1.58	0.54 ± 0.07	1.38 ± 0.28	0.933–0.986
Sr	0.23 ± 0.04	5.64	1.57	0.53 ± 0.06	1.62 ± 0.36	0.902–0.985
Ba	0.24 ± 0.02	5.60	1.58	0.49 ± 0.03	1.63 ± 0.12	0.972–0.985

Mont: montmorillonite.

Note: The caffeine adsorption test was performed five times (N = 5) and Q_Max and K_Lang were expressed by the averages of the five results and their standard deviations.

Although a method in which exchangeable interlayer ions are exchanged with cations such as ammonium ion (Borden and Giese, 2001) is well known as a method of measuring the amount of interlayer ions, we adopted a method of measuring the content of interlayer ions after dissolving Mont in acid. Mont after ion exchange contained enough exchanged cations (Table 2) to prove that successful exchange of interlayer ions had occurred. Based on the amount of interlayer ions, the interlayer distance, and Langmuir parameters obtained, the untreated Mont used in this test was thought to be a type similar to Ca-Mont. The total amount of interlayer ions was equivalent in the untreated Mont and the ion-exchanged Mont. This suggests that Mont containing ions of interest was successfully prepared by the ion exchange treatment in this study. Furthermore, Mont containing alkaline metal ions, in particular, in the interlayer space showed remarkably different adsorption characteristics.

It was also demonstrated that interlayer spacings differ according to the interlayer ions present (Table 2). Smectite clay minerals including Mont can adsorb various compounds in their interlayer spaces and have different interlayer distances in accordance with the type and amount of adsorbed compounds (Greesh et al., 2008; Koh and Dixon, 2001). Mont can also adsorb water molecules (Hensen et al., 2001; Sheng et al., 2002) and the interlayer distances vary according to the state of water molecule adsorption. It has been reported that interlayer spacings of Mont change stepwise according to interlayer ions and the state of intercalated water (Devineau et al., 2006; Zheng et al., 2011). The interlayer spacing without intercalated water is 1.0 nm and it expands with increase in intercalated water molecules. One of the factors affecting the state of water molecule adsorption is the types of intercalated ions in Mont. When the interlayer ion is a Na ion, the interlayer spacing with one and two layers of water molecules will be 1.26 and 1.52 nm (Sato et al., 1992), respectively, and when the interlayer ion is Ca ion, those will be 1.17 and 1.5 nm (Bray et al., 1998), respectively. Table 2 suggests that Na, K, Rb, and Cs-Mont have a layer of water molecules, and Li, Mg, Ca, Sr, and Ba-Mont have two layers of water molecules. For the alkali metal interlayer Mont ions, the interlayer spacings of Li and Na-Mont were larger than those of K, Rb, and Cs-Mont. The reason is thought to be that Li and Na-Mont are hydrated in the air during the drying process of cation exchange. This fact shows that Li and Na-Mont have a high adsorptivity to water.

The adsorption isotherm of X-Mont (Figure 3), especially Li-Mont and Na-Mont showed a tendency to increase without saturation even at high equilibrium concentration conditions. When approximated by the adsorption isothermal equation of Langmuir the result obtained for r² is high enough for this result to be considered of Langmuir type. Therefore, it is estimated that the change in Q_eq reaches equilibrium at an even higher condition of C_eq. In addition, it is probably that the value of Q_Max of Li-Mont and Na-Mont became higher than those of Rb-Mont and Cs-Mont is caused by this tendency. As for the adsorption characteristics of caffeine, Q_Max, that of Cs-Mont in particular was lowered remarkably. This is considered to be caused by a decrease in the interlayer spacing. It is reported that in Cs-intercalated Mont, Cs ions bond with the surface of the Si layer at a specific site (Siloxane Ditrigonal Cavity) (Sato et al., 2013; Yamaguchi et al., 2012). In addition, because Cs-Mont is less hydrated than Na-Mont (Numata et al., 2012), Cs ions tend to bring Si layers closer together in caffeine solution and therefore the surface of the layers was less exposed to the solution. As a result, the surface layer that is considered to be an adsorption site of caffeine (Yamamoto et al., 2016) is relatively less exposed to the caffeine solution and therefore Q_Max decreased more as compared to other ion-exchanged Mont.

In addition, K_Lang, which indicates the interaction with caffeine, shows an increasing trend as atomic numbers increase in alkali metal ion-exchanged and alkaline earth metal-exchanged Monts. This result shows that interlayer ions of Mont influence the adsorption of caffeine. It has been known for some time that the interlayer spacings play an important role in caffeine adsorption. The authors found from the XRD of caffeine-adsorbed Mont that the interlayer distance changed due to caffeine adsorption (Yamamoto et al., 2016). Adsorption of organic compounds other than caffeine, glycine for example, in the interlayer space has also been reported (Ramos and Huertas, 2013). The results of this study suggested that caffeine molecules are adsorbed in the interlayer space and that cations held in the interlayer space play an important role in caffeine adsorption.

A study investigating adsorption of benzonitrile onto Mont intercalated with different ions showed that polarization by intercalated cations changes the coordination state of benzonitrile and adsorbates can be stabilized in Mont by intercalating highly polarizing cations (Tsunashima et al., 1978). A similar examination was undertaken for the results obtained in this study. Figure 4 shows the relation between the interaction with caffeine and the polarizing power (z/r²) of interlayer ions (Pauling, 1927). It is thought that K_Lang is an indicator to express Q_eq, particularly under conditions of low C_eq, and can be considered to reflect the affinity of Mont as adsorbent and caffeine as adsorbate. As a result, there is a trend in which interaction with caffeine becomes low when the interlayer ion is a cation with high polarizing power. Cations having high polarizing power, such as Li and Na, interacted with water molecules strongly. While, to the contrary, Cs ions interacted less with hydrated water molecules, as compared to Li ions. Although a correlation between the adsorption capacity of caffeine (Q_Max) and polarizing power was examined by the same method as is in Figure 4, no obvious trend was found. Concerning ion-exchanged Mont, for those with alkaline earth metal ions in particular (Mg-, Ca-, Sr-, and Ba-Mont), the Q_Max values were almost same regardless of the degree of polarizing power. In other words, there was no relationship between Q_Max and polarizing power. This suggests that the amount of caffeine adsorption sites is constant, regardless of the kinds of ions held in the interlayer space.

Figure 4.

Relation between interaction with caffeine and polarizing power of interlayer ions. Black circles: montmorillonite intercalating alkali metal ions (Li, Na, K, Rb, Cs-Mont), white circles: montmorillonite intercalating alkaline earth metal ions (Mg, Ca, Sr, Ba-Mont). Polarization power of ions is from Pauling (1927). Mont: montmorillonite.

From the above results, we can suggest that because Mont interlayer cations which have a low polarizing power showed a larger interaction with caffeine, then the exchange reaction between water coordinated to interlayer ions and caffeine has an important role in caffeine adsorption. Because Li, Na, and Mg ions have a strong interaction with coordinated water molecules, exchange between coordinated water and caffeine is difficult when these cations are intercalated in Mont, and therefore the interaction with caffeine decreases. On the other hand, exchange between coordinated water and caffeine occurred easily when cations (K, Cs, and Ba ions) having a weaker interaction with coordinated water molecules are intercalated in Mont.

As the index of hydrophilicity of ion-exchanged Mont, we have showed a linear regression equation was obtained from a plot of R_2sp against the concentrations of ion-exchanged Mont suspensions and a slope of the regression equation was calculated (Beek et al., 1991). Figure 5 shows the relation between K_Lang and the obtained slope as the index of hydrophilicity of ion-exchanged Mont in the present work. K_Lang shows a good negative correlation (R = 0.62) against the increase in hydrophilicity. The results shown in Figures 4 and 5 indicate that degree of hydration of Mont influenced by interlayer ions play a large role in caffeine adsorption. The hydrophilicity of Mont intercalated by alkali metal ions, that is Li, Na, K, Rb, and Cs-Mont, showed a decreasing tendency as atomic numbers increased. On the other hand, the hydrophilicity of alkaline earth metal intercalated Mont, that is Mg, Ca, Sr, and Ba-Mont, showed no obvious change. From these results, we propose that Mont intercalating ions with higher polarizing power tend to be more hydrated and accordingly Li and Na-Mont showed a remarkably high hydrophilicity. It has been reported that the hydration phenomenon of Mont is influenced by ions held between layers (Boek et al., 1995). According to a simulation using the Monte Carlo method, it has been shown the interactions of a clay surface of Mont with interlayer ions, the interaction between a clay surface and water molecules, and interactions between interlayer ions and water molecules are all involved in hydration phenomenon of Mont. In particular, interaction between interlayer ions and water molecules is considered to be important for the degree of hydration of clay particles (Liu and Lu, 2006; Salles et al., 2008). Regarding Mont with different cations held between the layers, the energy states of the water molecules trapped between the layers are also mutually different (Meleshyn and Bunnenberg, 2005). Mont having Li and Na in its interlayer space is reported to swell easily (Norrish, 1954), and the reason is thought to be that Mont holds a lot of water molecules in its interlayer space in solution. Because Mont with a high hydrophilicity, that is Mont intercalated by ions with high polarizing power shows a decrease in its interaction with caffeine, we can assert that possession of water molecules by interlayer ions has a negative influence on caffeine adsorption. Although a correlation between the adsorption capacity of caffeine (Q_Max) and hydrophilicity was examined by the same method as is in Figure 5, no obvious trend was found.

Figure 5.

Relation between interaction with caffeine and hydrophilicity of ion-exchanged montmorillonite. The horizontal axis shows the index of hydrophilicity of ion-exchanged montmorillonite which is a slope of the regression equation obtained the plot of R_2sp against the concentrations of ion-exchanged montmorillonite suspensions. The ion exchange treatment of montmorillonite was performed five times (N = 5) and the hydrophilicity test and the caffeine adsorption test were performed for all the samples of X-Mont obtained. In this figure averages of the five values are plotted. Standard deviations are shown by error bars and approximate lines are represented by broken lines. Mont: montmorillonite.

From the results described above, the following two mechanisms of caffeine adsorption onto Mont are considered: (1) Adsorption of caffeine occurs by exchange of caffeine molecules with water molecules coordinated to interlayer ions. When cations strongly bound to coordinated water are intercalated, exchange between coordinated water and caffeine is difficult and, as a result, the interaction between Mont and caffeine is lowered. (2) Adsorption of caffeine occurs at sites created by the structure of Mont, regardless of interlayer ions. However, access of caffeine to the adsorption site is inhibited if many water molecules coordinate to interlayer ions. As a result, the interaction between Mont and caffeine is lowered.

A study on caffeine adsorption using organo-Mont suggests that caffeine adsorption occurs by exchange of caffeine for interlayer water molecules (Okada et al., 2015). Unlike our study using inorganic materials, Okada et al. used organic substances as intercalation compounds. However, the results of their study are in good agreement with those of this study.

This study showed that caffeine adsorption is influenced by water coordinated to interlayer ions. On the other hand, identification of the actual adsorption site in Mont is required to further understand the caffeine adsorption phenomenon. Further surface analysis studies of caffeine adsorption on Mont, for example, are necessary.

Conclusion

This study examined the influence of the hydrophilicity of Mont on adsorption of caffeine. The main findings concerning the influence of intercalated cations are as follows:

In Mont having Cs ions as the interlayer ion the maximum adsorption capacity for caffeine decreased. This is thought to be because Cs ions tend to bring the layers closer together and therefore the surface of the layers was less exposed to the solution. This fact suggested that adsorption of caffeine onto Mont occurs on the layer surface.

The relationship between the hydrophilicity of Mont and caffeine showed a negative correlation. There exists the possibility that, in Mont having a high hydrophilicity, water molecules coordinated to interlayer ions are strongly held and therefore interaction with caffeine may be decreased because of coordinated water blocking the layer surface or by inhibition of the exchange between water molecules and caffeine.

To understand the interaction between the surface of Mont and caffeine, a more detailed investigation of the adsorption site where interaction occurs is required. As a correlation between adsorption characteristics and hydrophilicity was observed, it may be possible to develop an effective adsorbent suited to a specific adsorbate by controlling the hydrophilicity of the adsorbent.

Footnotes

Acknowledgements

We thank members of our research groups for valuable discussions, especially Keiji Deuchi, Aruto Yoshida, Takahiro Eguchi.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Abollino

Aceto

Malandrino

et al. (2003) Adsorption of heavy metals on Na-montmorillonite. Effect of pH and organic substances. Water Research 37(7): 1619–1627.

Aepuru

Panda

(2014) Adsorption of charge carriers on radial zinc oxide and the study of their stability and dielectric behavior in poly(vinylidene fluoride). The Journal of Physical Chemistry C 118(33): 18868–18877.

Barbier

Duc

Petit-Ramel

(2000) Adsorption of lead and cadmium ions from aqueous solution to the montmorillonite/water interface. Colloids and Surfaces A: Physicochemical and Engineering Aspects 166(1–3): 153–159.

Beek

GPVd

Stuart

MAC

Cosgrove

(1991) Polymer adsorption and desorption studies via proton NMR relaxation of the solvent. Langmuir 7(2): 327–334.

Boek

Coveney

Skipper

(1995) Molecular modeling of clay hydration: A study of hysteresis loops in the swelling curves of sodium montmorillonites. Langmuir 11(12): 4629–4631.

Borden

Giese

(2001) Baseline studies of the clay minerals society source clays: Cation exchange capacity measurements by the ammonia-electrode method. Clays and Clay Minerals 49(5): 444–445.

Bradley

(1945) Molecular associations between montmorillonite and some polyfunctional organic liquids1. Journal of the American Chemical Society 67(6): 975–981.

Bragg

(1929) The diffraction of short electromagnetic waves by a crystal. Scientia 23(45): 153.

Bray

Redfern

SAT

Clark

(1998) The kinetics of dehydration in Ca-montmorillonite; an in situ X-ray diffraction study. Mineralogical Magazine 62(5): 647–656.

10.

Cabrera-Lafaurie

Román

Hernández-Maldonado

(2012) Transition metal modified and partially calcined inorganic–organic pillared clays for the adsorption of salicylic acid, clofibric acid, carbamazepine, and caffeine from water. Journal of Colloid and Interface Science 386(1): 381–391.

11.

Cabrera-Lafaurie

Román

Hernández-Maldonado

(2015) Single and multi-component adsorption of salicylic acid, clofibric acid, carbamazepine and caffeine from water onto transition metal modified and partially calcined inorganic–organic pillared clay fixed beds. Journal of Hazardous Materials 282: 174–182.

12.

Devineau

Bihannic

Michot

et al. (2006) In situ neutron diffraction analysis of the influence of geometric confinement on crystalline swelling of montmorillonite. Applied Clay Science 31(1–2): 76–84.

13.

Geng

Nakamura

et al. (2013) Unusually stable ∼100-fold reversible and instantaneous swelling of inorganic layered materials. Nat Communications 4: 1632.

14.

Greesh

Hartmann

Cloete

et al. (2008) Adsorption of 2-acrylamido-2-methyl-1-propanesulfonic acid (AMPS) and related compounds onto montmorillonite clay. Journal of Colloid and Interface Science 319(1): 2–11.

15.

Hensen

EJM

Tambach

Bliek

et al. (2001) Adsorption isotherms of water in Li–, Na–, and K–montmorillonite by molecular simulation. The Journal of Chemical Physics 115(7): 3322–3329.

16.

Imai

(1987) Multielement analysis of stream sediment by ICP-AES. Bunseki Kagaku 36(4): T41–T45.

17.

Jordan

Organophilic Bentonites

(1949) Swelling in organic liquids. The Journal of Physical and Colloid Chemistry 53(2): 294–306.

18.

Karamanis

Assimakopoulos

(2007) Efficiency of aluminum-pillared montmorillonite on the removal of cesium and copper from aqueous solutions. Water Research 41(9): 1897–1906.

19.

Karpovich

Vlasova

Sapronova

et al. (2016) Determination of dimensions of exfoliating materials in aqueous suspensions. MethodsX 3: 19–24.

20.

Koh

S-M

Dixon

(2001) Preparation and application of organo-minerals as sorbents of phenol, benzene and toluene. Applied Clay Science 18(3): 111–122.

21.

Kowalska

Güler

Cocke

(1994) Interactions of clay minerals with organic pollutants. Science of the Total Environment 141(1): 223–240.

22.

Lagaly

(1979) Crystalline silicic acids and their interface reactions. Advances in Colloid and Interface Science 11(2): 105–148.

23.

Laxman

Myint

MTZ

Abri

et al. (2015) Desalination and disinfection of inland brackish ground water in a capacitive deionization cell using nanoporous activated carbon cloth electrodes. Desalination 362: 126–132.

24.

Lee

(2003) Effect of montmorillonite on the swelling behavior and drug-release behavior of nanocomposite hydrogels. Journal of Applied Polymer Science 89(13): 3652–3660.

25.

Liu

X-D

X-C

(2006) A thermodynamic understanding of clay-swelling inhibition by potassium ions. Angewandte Chemie International Edition 45(38): 6300–6303.

26.

MacEwan

DMC

(1948) Complexes of clays with organic compounds. I. Complex formation between montmorillonite and halloysite and certain organic liquids. Transactions of the Faraday Society 44: 349–367.

27.

Meleshyn

Bunnenberg

(2005) Swelling of Na/Mg-montmorillonites and hydration of interlayer cations: A Monte Carlo study. The Journal of Chemical Physics 123(7): 074706.

28.

Morillo

Rodriguez

JLP

Rubio

et al. (1997) Interaction of aminotriazole with montmorillonite and Mg-vermiculite at pH 4. Clay Minerals 32(2): 307–313.

29.

Norrish

(1954) The swelling of montmorillonite. Discussions of the Faraday Society 18: 120–134.

30.

Numata

Sato

Fujimoto

(2012) Indication of cesium adsorption into angstrom-scale open spaces in saponite clay mineral. International Journal of Nanoscience 11(6): 1240034.

31.

Ogawa

Hagiwara

Handa

et al. (1995) Solid-state ion exchange reactions between homoionic-montmorillonites and organoammonium salts. Journal of Porous Materials 1(1): 85–89.

32.

Ogawa

Hashizume

Kuroda

et al. (1991) Intercalation of 2,2′-bipyridine and complex formation in the interlayer space of montmorillonite by solid-solid reactions. Inorganic Chemistry 30(3): 584–585.

33.

Ogawa

Ishii

Miyamoto

et al. (2003) Intercalation of a cationic azobenzene into montmorillonite. Applied Clay Science 22(4): 179–185.

34.

Ogawa

Kuroda

Kato

(1989) Preparation of montmorillonite-organic intercalation compounds by solid-solid reactions. Chemistry Letters 18(9): 1659–1662.

35.

Okada

Oguchi

Yamamoto

et al. (2015) Organoclays in water cause expansion that facilitates caffeine adsorption. Langmuir 31(1): 180–187.

36.

Pauling

(1927) The sizes of ions and the structure of ionic crystals. Journal of the American Chemical Society 49(3): 765–790.

37.

Ramos

Huertas

(2013) Adsorption of glycine on montmorillonite in aqueous solutions. Applied Clay Science 80–81: 10–17.

38.

Salles

Beurroies

Bildstein

et al. (2008) A calorimetric study of mesoscopic swelling and hydration sequence in solid Na-montmorillonite. Applied Clay Science 39(3–4): 186–201.

39.

Sapsford

Tyner

Dair

et al. (2011) Analyzing nanomaterial bioconjugates: A review of current and emerging purification and characterization techniques. Analytical Chemistry 83(12): 4453–4488.

40.

Sato

Fujimoto

Dai

et al. (2013) Molecular mechanism of heavily adhesive Cs: Why radioactive Cs is not decontaminated from soil. The Journal of Physical Chemistry C 117(27): 14075–14080.

41.

Sato

Watanabe

Otsuka

(1992) Effects of layer charge, charge location, and energy change on expansion properties of dioctahedral smectites. Clays and Clay Minerals 40(1): 103–113.

42.

Schwarzenbach

Gschwend

Imboden

(2002) Environmental Organic Chemistry, 2nd ed. New Jersey: John Wiley & Sons Inc.

43.

Serratosa

(1968) Infrared study of benzonitrile (C6H5- CN)-montmorillonite complexes. American Mineralogist 53: 1244–1251.

44.

Sheng

Johnston

Teppen

et al. (2002) Adsorption of dinitrophenol herbfrom water by montmorillonites. Clays and Clay Minerals 50(1): 25–34.

45.

Sylvester

Clearfield

Diaz

(1999) Pillared montmorillonites: Cesium-selective ion-exchange materials. Separation Science and Technology 34(12): 2293–2305.

46.

Theng

BKG

(1974) The Chemistry of Clay-Organic Reactions, New Jersey: John Wiley & Sons Inc.

47.

Torigai

Ouyang

Iwashima

et al. (1997) Studies on microwave digestion procedures for the simultaneous multielement determination of arsenic, antimony, chromium, cadmium, nickel, and lead in municipal waste incineration fly ash by ICP-AES. Bunseki Kagaku 46(5): 401–406.

48.

Tsunashima

Tachiki

Kodaira

et al. (1978) Complexes of benzonitrile and montmorillonites saturated with various interlayer cations. Bulletin of the Faculty of Engineering 91: 87–94.

49.

Yamaguchi

Takata

Hayashi

et al. (2012) Behavior of radiocaesium in soil-plant systems and its controlling factor: A review. Bulletin of National Institute for Agro-Environmental Sciences 31: 75–129.

50.

Yamamoto

Shiono

Matsui

et al. (2016) Changes the structure and caffeine adsorption property of calcined montmorillonite. International Journal of Geomate 11(24): 2301–2306.

51.

Yamanaka

Kanamaru

Koizumi

(1974) Role of interlayer cations in the formation of acrylonitrile-montmorillonite complexes. The Journal of Physical Chemistry 78(1): 42–44.

52.

Yamanaka

Kanamaru

Koizumi

(1975) Orientation of acrylonitrile adsorbed on interlamellar surfaces of montmorillonites. The Journal of Physical Chemistry 79(13): 1285–1288.

53.

Zheng

Zaoui

Shahrour

(2011) A theoretical study of swelling and shrinking of hydrated Wyoming montmorillonite. Applied Clay Science 51(1–2): 177–181.