- Email: [email protected]
Silylation of montmorillonite surfaces: Dependence on solvent nature
Journal of Colloid and Interface Science 391 (2013) 16–20
Contents lists available at SciVerse ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Silylation of montmorillonite surfaces: Dependence on solvent nature Linna Su a,b, Qi Tao a, Hongping He a,⇑, Jianxi Zhu a, Peng Yuan a, Runliang Zhu a a b
Key Laboratory of Mineralogy and Metallogeny, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, PR China Graduate University of Chinese Academy of Sciences, Beijing 100049, PR China
a r t i c l e
i n f o
Article history: Received 16 June 2012 Accepted 8 August 2012 Available online 11 October 2012 Keywords: Montmorillonite Surface silylation Solvents Swelling property Intercalation
a b s t r a c t Silylation of clay mineral surfaces has attracted much attention due to their extensive applications in materials science and environmental engineering. Silylation of montmorillonite surfaces with 3-aminopropyltriethoxysilane was carried out in polar-protic and nonpolar solvents. The swelling property of the silylated montmorillonites was investigated by intercalating with cetyltrimethylammonium bromide. Silylated montmorillonites prepared in nonpolar solvents showed a larger amount of loaded silane and a higher extent of condensation among different silane molecules, comparing with those prepared in polar-protic solvents with high dielectric constant. Meanwhile, the silylated montmorillonites prepared in nonpolar solvents displayed poor swelling property due to the linkage between silane oligomers and clay layers, that is, the neighboring clay layers were locked by the silane oligomers. The present study demonstrated that the polarity of the solvents used had an important influence on the extent of grafting, interlayer structure, and swelling property of the silylated products. This is of high importance for synthesis and application of silylated clay minerals. Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction Modification of clay mineral surfaces with organic moieties has proven to be an important step for their industrial applications, such as in clay mineral–polymer nanocomposites [1,2], environmental materials [3–6], and drug delivery [7]. The methods used for modifying clay minerals mainly include exchange of the inherent interlayer cations with cationic surfactants [8,9], grafting clay minerals surfaces with silanes [10–12], and using silanes as silicon source in the synthesis of clay minerals [13,14]. Silane grafting, also known as silylation, was proven to be an efficient method to modify clay minerals surfaces. Silanes are anchored onto clay mineral surfaces through a condensation reaction between the hydroxyls in hydrolyzed silanes and the silanol groups on clay mineral surfaces. The covalent bond enables a durable immobilization of the organic moieties, preventing their leaching into the surrounding solutions. Meanwhile, the adsorption selectivity of the silylated products to contaminants could be greatly improved by introducing special functional groups (e.g., –NH2, –SH) via the organosilane agents used [3–6]. On the other hand, the introduced functional groups can react with the polymer matrix, resulting in the formation of a network among clay mineral, silane, and polymer via covalent bonds. This can greatly improve the mechanical property of the resultant clay mineral–polymer nanocomposites and may lead to a breakthrough in synthesis of novel materials. ⇑ Corresponding author. Fax: +86 20 85290130. E-mail address: [email protected] (H. He). 0021-9797/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcis.2012.08.077
As reported in the literature, all internal surface, external surface, and broken edge of expandable clay minerals are possible sites for silane grafting [15–19]. In the cases of internal surface and broken edge grafting, a prominent increase in the basal spacing was observed [15–18], whereas the external surface grafting showed no influence on the basal spacing of clay minerals [19]. The basal spacing of the silylated products strongly depended on the configuration of the silane used [17], the loaded silane amount in the interlayer spaces [18] and the interactions among silane molecules (e.g., intermolecular hydrogen bonding) and between clay mineral and special group of silane (e.g., –NH2 group) [20]. The chemical nature of the solvents used in silylation reaction has significant influences on the amount and distribution of the grafted silanes and the structure of the silylated products [16,21,22]. Shanmugharaj et al. reported that the surface energy of the solvents showed a significant effect on the average basal spacing of the grafted products [16]. Although many successful silylation of clay mineral surfaces have been achieved, little attention was paid to the influence of solvents with different polarity and dielectric constant on the silylation mechanism and the swelling property of the silylated clay minerals. In fact, the swelling property of the silylated clay minerals is a key factor for intercalation of macromolecules and exfoliation of the silylated clay minerals layers [23,24]. In this study, solvents with different polarity and dielectric constant (ethanol, isopropanol, toluene, and cyclohexane) were used in the preparation of silylated montmorillonites. The structure of the silylated products was characterized, and the swelling property
L. Su et al. / Journal of Colloid and Interface Science 391 (2013) 16–20
was investigated by intercalation of surfactant, to elucidate the influence of the solvent nature on the silylation mechanism and structure of the silylated products. The new insights obtained in this study are of high importance for advancing our fundamental understanding of the mechanisms underlying the silylation of clay minerals surfaces and developing an efficient silylating technology. 2. Experimental section 2.1. Materials Ca-montmorillonite (Mt) used in this study was obtained from Neimeng, China, with cationic exchange capacity (CEC) of 110 mequiv/100 g. The trifunctional organosilane agent, 3-aminopropyltriethoxysilane (APTES), was purchased from Aldrich (P98%). The surfactant used was cetyltrimethylammonium bromide (CTAB) with a purity of 99%, obtained from Nanjing Chem. Co., China. Anhydrous ethanol, isopropanol, toluene, and cyclohexane were of analytical purity and provided by Tianjin Chem. Co., China. All of the reagents were used without further purification. 2.2. Silylation with APTES 10.0 g of Mt and 8.0 g of APTES were mixed by stirring in 200 mL of solvent at 80 °C for 10 h. The polar-protic solvents used were ethanol and isopropanol, and the nonpolar solvents were toluene and cyclohexane. The silylated products were separated by centrifugation, washed with the solvent six times, and then dried at 80 °C. 2.3. Intercalation with CTAB A desired amount of CTAB was dissolved in 30 mL of distilled water. Then, 2.0 g of pristine Mt or silylated Mt was added into the prepared solution containing CTAB. The mixtures were stirred at 60 °C for 6 h. All the products were washed with distilled water eight times and dried at 80 °C. The concentrations of CTAB in the intercalation were 0.8 CEC, 1.5 CEC, and 2.0 CEC of Mt’s CEC. 0.8 CEC meant that the added amount of CTAB was 0.8 times CEC of Mt in the suspension. 2.4. Characterization
17
3. Results and discussion A series of APTES silylated Mts were synthesized in polar-protic (ethanol and isopropanol) and nonpolar (toluene and cyclohexane) solvents. An obvious increase in the basal spacing in silylated Mts was observed, from 1.48 nm (pristine Mt) to ca. 1.7 nm (silylated in polar-protic solvents) or to ca. 1.8 nm (silylated in nonpolar solvents) (Fig. 1). This indicates that APTES was readily intercalated into the Mt interlayer spaces, and the polarity of solvents had an influence on the interlayer structure of the silylated products [15]. Compared with FTIR spectrum of pristine Mt (Fig. 2), the occurrence of the antisymmetric stretching vibration of –CH2 at ca. 2934 cm1 and the out-of-plane deformation of –CH at 696 cm1 provided a supporting evidence for the existence of APTES in the silylated products [15,16]. In the FTIR spectra of pristine Mt and silylated products, both the vibration at ca. 1032 cm1 and the shoulder at ca. 1085 cm1 correspond to the stretching vibration of Si–O–Si [25]. A frequency shift of the vibration from 1032 cm1 (pristine Mt) to 1037 cm1 (silylated products) and a weakening of the shoulder are important information for the intercalation of APTES and consequent grafting reaction. As shown by the DTG curves of pristine Mt and the silylated products (Fig. 3), new mass loss stages are observed in the silylated products in the temperature range from 200 to 600 °C. The amount of the loaded silane was determined by calculating the mass loss in the range of 200–600 °C from thermogravimetric curves. The mass losses at ca. 200 °C, 330 °C, and 430 °C are attributed to the silanes physisorbed on Mt surfaces, covalently bonded at the broken edges and/or adsorbed on the Mt surfaces, and intercalated in the interlayer spaces, respectively [15,16,20]. The mass loss at ca. 540 °C is assigned to the decomposition of the covalent bonded silane [15,20,26]. The mass loss corresponding to the dehydroxylation of Mt almost disappeared, partly due to the consumption of hydroxyl groups during silylation reaction [20]. This provides another important evidence for the covalent bond between silane and siloxane surface of Mt. Prominent difference of the silane loaded amount in the silylated Mts prepared in different solvents could be observed as shown in Table 1. The amounts of loaded silane are 13.3% and 15.6%, respectively, in the samples prepared in ethanol and isopropanol (polar-protic solvents), while 17.4% and 17.8% in the samples prepared in toluene and cyclohexane (nonpolar solvents), respectively. The relationship between solvent polarity and silane
X-ray diffraction (XRD) analysis was performed on a Brucker D8 advanced diffractometer with a scan rate of 3.0° (2h) min1. FTIR spectra were obtained with KBr pressed disk technique using a Bruker VERTEX 70 Fourier transform infrared spectrometer. All spectra were collected at room temperature over the range 400– 4000 cm1 with a resolution of 4 cm1 and 64 scans. Thermogravimetric (TG) analysis was carried out on a Netzsch STA 409 PC/PG instrument, employing a heating rate of 10 °C min1 from 30 to 900 °C under a nitrogen flow of 60 mL min1. The percentage of the loaded silane, which corresponds to the percentage of silane molecules with respect to the total inorganic mass, was calculated as follows:
Silane loaded amount ð%Þ ¼
100 W 200—600 100 W 200—600
ð1Þ
where W200–600 corresponds to the mass loss between 200 °C and 600 °C. Solid-state 29Si CP/MAS NMR spectra were acquired using a Bruker Advanced 300 NMR spectrometer operating at 59.63 MHz. The contact time was 5 ms, the recycle delay 3 s, and the spinning rate 5.5 kHz. Tetramethylsilane (TMS) was used as the external reference.
Fig. 1. XRD patterns of pristine Mt (a) and silylated Mts prepared in ethanol (b), isopropanol (c), toluene (d), and cyclohexane (e).
18
L. Su et al. / Journal of Colloid and Interface Science 391 (2013) 16–20
Fig. 2. FTIR spectra of pristine Mt (a) and silylated Mts prepared in ethanol (b), isopropanol (c), toluene (d), and cyclohexane (e).
Fig. 3. DTG curves of pristine Mt (a) and silylated Mts prepared in ethanol (b), isopropanol (c), toluene (d), and cyclohexane (e).
Table 1 Thermogravimetric analysis of silylated Mts prepared in polar-protic and nonpolar solvents.
a b
Solvent
Dielectric constant
Mass loss (%)a
Loaded amount (%)b
Ethanol Isopropanol Toluene Cyclohexane
24.6 18.0 2.4 2.0
11.7 13.5 14.8 15.1
13.3 15.6 17.4 17.8
Mass loss between 200 °C and 600 °C. Determined by using Eq. (1).
loaded amount can be well explained on the basis of the effects of solvation and dielectric properties of solvents [27]. Solvation of reactants could affect reaction rates and mechanisms in organic reactions [28,29]. As polar-protic solvent molecules contain dipoles, they can solvate the silanol groups on Mt surfaces when solvent molecules are close to Mt surfaces. In this case, a solvent cage
is created outside silanols on Mt surfaces, making the silanols less available to react with silane [21,27]. The decrease in the silanol reactivity results in a less amount of loaded silane. For nonpolar solvents, lack of the solvation ability will facilitate the silane grafting reaction and lead to more silane loaded amount. The dielectric constant of the solvent used is another important factor to affect the amount of loaded silane [27]. In the solvent with high dielectric constant, hydrogen bond will be formed between solvent molecules and the amine groups of APTES. The strength of the hydrogen bond strongly depends on the dielectric constant of the solvent used, that is, the higher the dielectric constant, the stronger the hydrogen bond. A strong hydrogen bond will decrease the mobility of silane molecules and hinder the silanes to approach to the surface silanols of Mt and grafting reaction [21,27], resulting in a less amount of silane loaded in the silylated products. This assumption was well verified by a crosscurrent between the dielectric constants of the solvents and the silane loaded amount (Table 1). The present study showed that the silane loaded amounts decreased as: cyclohexane > toluene > isopropanol > ethanol. Swelling property of the silylated products is of high importance for their industrial applications since it controls the intercalation of macromolecules and exfoliation of clay minerals layers. Hence, a series of intercalation experiments with CTAB were conducted on pristine Mt and the silylated products to investigate their swelling property. For pristine Mt, the basal spacing increased with the increase in the intercalated surfactant amount (Fig. 4a), similar to those reported in the literature [9,30]. When the CTAB concentration increased to 2.0 CEC, the maximum basal spacing reached 3.74 nm. However, a prominent difference of the basal spacings was observed between the intercalation compounds from the silylated Mts prepared in polar-protic and nonpolar solvents. For the silylated Mts prepared in ethanol and isopropanol (polarprotic solvents), a dramatic increase in the basal spacing was recorded after CTAB intercalation (Fig. 4b and c), similar to the case of intercalating pristine Mt with CTAB. The maximum basal spacings increased to 3.45 nm and 3.72 nm for silylated Mts prepared in ethanol and isopropanol, respectively, when the CTAB concentration increased to 2.0 CEC. This implies that the silylated Mts layers remain expandable like the pristine Mt, that is, the neighboring layers are not fixed by the grafted silanes or polysiloxane oligomers in the interlayer spaces. However, for the silylated products prepared in nonpolar solvents (toluene and cyclohexane) (Fig. 4d and e), both the d001 values increased to 2.16 nm after intercalating with CTAB at 0.8 CEC. But the basal spacing remained unchanged even increasing the CTAB concentration to 2.0 CEC in the preparation system, very different from the cases of pristine Mt and the silylated Mts prepared in polar-protic solvents. This implies that the neighboring montmorillonite layers were ‘‘locked’’ during silylation, and the polarity of the solvents used has a significant effect on the silylation reaction and the structure of the silylated products. In the media of polar-protic solvents, there is a competitive adsorption between silane and solvent to approach onto the Mt surfaces. Meanwhile, the hydrogen bond between the amino groups of APTES and solvents lowers the tendency to successful silane grafting. On the other hand, the hydrolysis rate of APTES is restrained when ethanol is used as grafting solvent [31,32]. These led to a low density of intercalated silane in the interlayer spaces, as indicated by XRD and TG measurements, and a low extent of condensation among silane molecules. In this case, the montmorillonite layers of silylated products would not be locked by the grafted silane and/or polysiloxane oligomers. That is to say, the silylated Mts prepared in the polar-protic solvents kept the swelling property as pristine Mt. On the contrary, the absence of hydrogen bond between silane molecules and nonpolar solvent molecules leads to
L. Su et al. / Journal of Colloid and Interface Science 391 (2013) 16–20
19
Fig. 4. XRD patterns of pristine Mt, silylated Mts and samples intercalated with CTAB at 0.8, 1.5 and 2.0 CEC, respectively. (a) Pristine Mt. (b) Silylated Mt prepared in ethanol. (c) Silylated Mt prepared in isopropanol. (d) Silylated Mt prepared in toluene. (e) Silylated Mt prepared in cyclohexane.
a higher degree of condensation among silane molecules and between silane molecules and Mt surfaces [21,33]. In this case, a bridge of polysiloxane oligomers was formed between the adjacent layers as depicted in Fig. 5. This lock effect resulted in a loss of swelling property in the silylated Mts and hindered the efficiency of subsequent CTAB intercalation (Fig. S1).
Fig. 5. Schematic illustration of Mt silylation with APTES in various solvents.
Solid-state 29Si MAS NMR spectra gave further evidence for the influence of the solvent polarity on the structure of the silylated products. Tn notations (n = 0, 1, 2, and 3, which represents the
Fig. 6. 29Si NMR spectra of pristine Mt (a) and silylated Mts prepared in ethanol (b) and cyclohexane (c).
20
L. Su et al. / Journal of Colloid and Interface Science 391 (2013) 16–20
number of bridge ‘‘O’’ atoms surrounding the silicon atom) were used to describe the different kinds of siloxane bonds formed between APTES and montmorillonite and among APTES molecules [11,34–36]. As shown in Fig. 6a, pristine Mt displayed a resonance at 93.6 ppm, corresponding to the silicon in Si–O tetrahedral sheet of montmorillonite. After silylation reaction, a new resonance was recorded at 59.6 ppm in 29Si MAS NMR spectrum of the silylated Mt prepared in ethanol, which was attributed to the hydrolyzed bidentate (T2). For the silylated Mt prepared in cyclohexane, a predominant resonance occurred at 67.6 ppm with a shoulder at 62.5 ppm, corresponding to the hydrolyzed tridentate (T3) and hydrolyzed bidentate (T2), respectively. This indicates that the nonpolar solvents with low dielectric constant facilitate hydrolysis and condensation of silane molecules. In this case, polysiloxane oligomers will be formed in the interlayer spaces of montmorillonite and result in a ‘‘lock’’ effect on the silylated products.
4. Conclusions Silylated montmorillonites were prepared in polar-protic and nonpolar solvents. The present study demonstrated that silane was readily intercalated into the interlayer spaces of montmorillonite in both polar-protic and nonpolar solvents. Nonpolar solvents with low dielectric constant facilitate intercalation, hydrolysis, and condensation of silane molecules, resulting in a high loading of silane in the silylated products and formation of polysiloxane oligomers in the interlayer spaces. The polysiloxane oligomers linked the neighboring montmorillonite layers via covalent bond, leading to a ‘‘lock’’ effect and loss of swelling ability of the silylated products. However, in the case of polar-protic solvents with high dielectric constant, the decreased hydrolysis extent and the hydrogen bond between the amino groups of APTES and solvents led to a low loading of silane and a low degree of condensation among silane molecules. The silylated products prepared in polar-protic solvents displayed swelling ability as pristine montmorillonite. The present study showed that the polarity of the solvents used has an important influence on the structure and swelling property of the silylated products. The obtained new insights advance our fundamental understanding of the mechanism underlying the silylation of clay mineral surfaces and are of importance for synthesis and application of silylated clay minerals.
Acknowledgments This is contribution No. IS-1551 from GIGCAS. This work was financially supported by the National Natural Science Foundation of China (Grant Nos. U0933003 and 41002015), the National Key Technology R&D Program (Grant No. 2011BAB03B06), and the Guangdong Natural Science Foundation (Grant No. S2011010003599).
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.2012.08.077. References [1] S.-J. Park, B.-J. Kim, D.-I. Seo, K.-Y. Rhee, Y.-Y. Lyu, Mater. Sci. Eng. A – Struct. 526 (2009) 74. [2] S.R. Ha, K.Y. Rhee, H.C. Kim, J.T. Kim, Colloid Surf. A 313–314 (2008) 112. [3] H. Sayılkan, S. Erdemog˘lu, Sß. Sßener, F. Sayılkan, M. Akarsu, M. Erdemog˘lu, J. Colloid Interface Sci. 275 (2004) 530. [4] A. de Mello Ferreira Guimarães, V.S.T. Ciminelli, W.L. Vasconcelos, Appl. Clay Sci. 42 (2009) 410. [5] I.K. Tonle, E. Ngameni, D. Njopwouo, C. Carteret, A. Walcarius, Phys. Chem. Chem. Phys. 5 (2003) 4951. [6] M.G. da Fonseca, C. Airoldi, Mater. Res. Bull. 36 (2001) 277. [7] L. Zampori, P.G. Stampino, C. Cristiani, P. Cazzola, G. Dotelli, Appl. Clay Sci. 50 (2010) 266. [8] H.P. He, R.L. Frost, T. Bostrom, P. Yuan, L. Duong, D. Yang, Y.F. Xi, J.T. Kloprogge, Appl. Clay Sci. 31 (2006) 262. [9] H.P. He, Y.H. Ma, J.X. Zhu, P. Yuan, Y.H. Qing, Appl. Clay Sci. 48 (2010) 67. [10] E. Ruiz-Hitzky, J.J. Fripiatt, Clays Clay Miner. 24 (1976) 25. [11] A.-Y. Park, H. Kwon, A.J. Woo, S.-J. Kim, Adv. Mater. 17 (2005) 106. [12] P.A. Wheeler, J.Z. Wang, J. Baker, L.J. Mathias, Chem. Mater. 17 (2005) 3012. [13] M.G. da Fonseca, C.R. Silva, C. Airoldi, Langmuir 15 (1999) 5048. [14] S. Letaief, E. Ruiz-Hitzky, Chem. Commun. (2003) 2996. [15] H.P. He, J. Duchet, J. Galy, J.-F. Gerard, J. Colloid Interface Sci. 288 (2005) 171. [16] A.M. Shanmugharaj, K.Y. Rhee, S.H. Ryu, J. Colloid Interface Sci. 298 (2006) 854. [17] N.N. Herrera, J.-M. Letoffe, J.-L. Putaux, L. David, E. Bourgeat-Lami, Langmuir 20 (2004) 1564. [18] N.N. Herrera, J.-M. Letoffe, J.-P. Reymond, E. Bourgeat-Lami, J. Mater. Chem. 15 (2005) 863. [19] M. Park, I.-K. Shim, E.-Y. Jung, J.-H. Choy, J. Phys. Chem. Solids 65 (2004) 499. [20] F. Piscitelli, P. Posocco, R. Toth, M. Fermeglia, S. Pricl, G. Mensitieri, M. Lavorgna, J. Colloid Interface Sci. 351 (2010) 108. [21] O.G. Nik, B. Nohair, S. Kaliaguine, Microporous Mesoporous Mater. 143 (2011) 221. [22] K. Song, G. Sandi, Clays Clay Miner. 49 (2001) 119. [23] K. Wang, L. Wang, J.S. Wu, L. Chen, C.B. He, Langmuir 21 (2005) 3613. [24] A. Di Gianni, E. Amerio, O. Monticelli, R. Bongiovanni, Appl. Clay Sci. 42 (2008) 116. [25] Y.H. Ma, J.X. Zhu, H.P. He, P. Yuan, W. Shen, D. Liu, Spectrochim. Acta A 76 (2010) 122. [26] W. Shen, H.P. He, J.X. Zhu, P. Yuan, R.L. Frost, J. Colloid Interface Sci. 313 (2007) 268. [27] K.K. Sharma, A. Anan, R.P. Buckley, W. Ouellette, T. Asefa, J. Am. Chem. Soc. 130 (2008) 218. [28] T.W. Bentley, J.-P. Dau-Schmidt, G. Llewellyn, H. Mayr, J. Org. Chem. 57 (1992) 2387. [29] Y. Umebayashi, N. Shigeta, Y. Zhang, M. Komiya, Y.-i. Murakami, S.-i. Ishiguro, J. Mol. Liq. 119 (2005) 177. [30] H.P. He, R.L. Frost, F. Deng, J.X. Zhu, X.Y. Wen, P. Yuan, Clays Clay Miner. 52 (2004) 350. [31] Q. Tao, J.X. Zhu, R.L. Frost, T.E. Bostrom, R.M. Wellard, J.M. Wei, P. Yuan, H.P. He, Langmuir 26 (2010) 2769. [32] M.-C.B. Salon, M. Abdelmouleh, S. Boufi, M.N. Belgacem, A. Gandini, J. Colloid Interface Sci. 289 (2005) 249. [33] K.K. Sharma, T. Asefa, Angew. Chem. Int. Ed. 46 (2007) 2879. [34] X. Feng, G.E. Fryxell, L.-Q. Wang, A.Y. Kim, J. Liu, K.M. Kemner, Science 276 (1997) 923. [35] Q. Tao, J.X. Zhu, R.M. Wellard, T.E. Bostrom, R.L. Frost, P. Yuan, H.P. He, J. Mater. Chem. 21 (2011) 10711. [36] M.-C.B. Salon, P.-A. Bayle, M. Abdelmouleh, S. Boufi, M.N. Belgacem, Colloid Surf. A 312 (2008) 83.
Contents lists available at SciVerse ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Silylation of montmorillonite surfaces: Dependence on solvent nature Linna Su a,b, Qi Tao a, Hongping He a,⇑, Jianxi Zhu a, Peng Yuan a, Runliang Zhu a a b
Key Laboratory of Mineralogy and Metallogeny, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, PR China Graduate University of Chinese Academy of Sciences, Beijing 100049, PR China
a r t i c l e
i n f o
Article history: Received 16 June 2012 Accepted 8 August 2012 Available online 11 October 2012 Keywords: Montmorillonite Surface silylation Solvents Swelling property Intercalation
a b s t r a c t Silylation of clay mineral surfaces has attracted much attention due to their extensive applications in materials science and environmental engineering. Silylation of montmorillonite surfaces with 3-aminopropyltriethoxysilane was carried out in polar-protic and nonpolar solvents. The swelling property of the silylated montmorillonites was investigated by intercalating with cetyltrimethylammonium bromide. Silylated montmorillonites prepared in nonpolar solvents showed a larger amount of loaded silane and a higher extent of condensation among different silane molecules, comparing with those prepared in polar-protic solvents with high dielectric constant. Meanwhile, the silylated montmorillonites prepared in nonpolar solvents displayed poor swelling property due to the linkage between silane oligomers and clay layers, that is, the neighboring clay layers were locked by the silane oligomers. The present study demonstrated that the polarity of the solvents used had an important influence on the extent of grafting, interlayer structure, and swelling property of the silylated products. This is of high importance for synthesis and application of silylated clay minerals. Ó 2012 Elsevier Inc. All rights reserved.
1. Introduction Modification of clay mineral surfaces with organic moieties has proven to be an important step for their industrial applications, such as in clay mineral–polymer nanocomposites [1,2], environmental materials [3–6], and drug delivery [7]. The methods used for modifying clay minerals mainly include exchange of the inherent interlayer cations with cationic surfactants [8,9], grafting clay minerals surfaces with silanes [10–12], and using silanes as silicon source in the synthesis of clay minerals [13,14]. Silane grafting, also known as silylation, was proven to be an efficient method to modify clay minerals surfaces. Silanes are anchored onto clay mineral surfaces through a condensation reaction between the hydroxyls in hydrolyzed silanes and the silanol groups on clay mineral surfaces. The covalent bond enables a durable immobilization of the organic moieties, preventing their leaching into the surrounding solutions. Meanwhile, the adsorption selectivity of the silylated products to contaminants could be greatly improved by introducing special functional groups (e.g., –NH2, –SH) via the organosilane agents used [3–6]. On the other hand, the introduced functional groups can react with the polymer matrix, resulting in the formation of a network among clay mineral, silane, and polymer via covalent bonds. This can greatly improve the mechanical property of the resultant clay mineral–polymer nanocomposites and may lead to a breakthrough in synthesis of novel materials. ⇑ Corresponding author. Fax: +86 20 85290130. E-mail address: [email protected] (H. He). 0021-9797/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcis.2012.08.077
As reported in the literature, all internal surface, external surface, and broken edge of expandable clay minerals are possible sites for silane grafting [15–19]. In the cases of internal surface and broken edge grafting, a prominent increase in the basal spacing was observed [15–18], whereas the external surface grafting showed no influence on the basal spacing of clay minerals [19]. The basal spacing of the silylated products strongly depended on the configuration of the silane used [17], the loaded silane amount in the interlayer spaces [18] and the interactions among silane molecules (e.g., intermolecular hydrogen bonding) and between clay mineral and special group of silane (e.g., –NH2 group) [20]. The chemical nature of the solvents used in silylation reaction has significant influences on the amount and distribution of the grafted silanes and the structure of the silylated products [16,21,22]. Shanmugharaj et al. reported that the surface energy of the solvents showed a significant effect on the average basal spacing of the grafted products [16]. Although many successful silylation of clay mineral surfaces have been achieved, little attention was paid to the influence of solvents with different polarity and dielectric constant on the silylation mechanism and the swelling property of the silylated clay minerals. In fact, the swelling property of the silylated clay minerals is a key factor for intercalation of macromolecules and exfoliation of the silylated clay minerals layers [23,24]. In this study, solvents with different polarity and dielectric constant (ethanol, isopropanol, toluene, and cyclohexane) were used in the preparation of silylated montmorillonites. The structure of the silylated products was characterized, and the swelling property
L. Su et al. / Journal of Colloid and Interface Science 391 (2013) 16–20
was investigated by intercalation of surfactant, to elucidate the influence of the solvent nature on the silylation mechanism and structure of the silylated products. The new insights obtained in this study are of high importance for advancing our fundamental understanding of the mechanisms underlying the silylation of clay minerals surfaces and developing an efficient silylating technology. 2. Experimental section 2.1. Materials Ca-montmorillonite (Mt) used in this study was obtained from Neimeng, China, with cationic exchange capacity (CEC) of 110 mequiv/100 g. The trifunctional organosilane agent, 3-aminopropyltriethoxysilane (APTES), was purchased from Aldrich (P98%). The surfactant used was cetyltrimethylammonium bromide (CTAB) with a purity of 99%, obtained from Nanjing Chem. Co., China. Anhydrous ethanol, isopropanol, toluene, and cyclohexane were of analytical purity and provided by Tianjin Chem. Co., China. All of the reagents were used without further purification. 2.2. Silylation with APTES 10.0 g of Mt and 8.0 g of APTES were mixed by stirring in 200 mL of solvent at 80 °C for 10 h. The polar-protic solvents used were ethanol and isopropanol, and the nonpolar solvents were toluene and cyclohexane. The silylated products were separated by centrifugation, washed with the solvent six times, and then dried at 80 °C. 2.3. Intercalation with CTAB A desired amount of CTAB was dissolved in 30 mL of distilled water. Then, 2.0 g of pristine Mt or silylated Mt was added into the prepared solution containing CTAB. The mixtures were stirred at 60 °C for 6 h. All the products were washed with distilled water eight times and dried at 80 °C. The concentrations of CTAB in the intercalation were 0.8 CEC, 1.5 CEC, and 2.0 CEC of Mt’s CEC. 0.8 CEC meant that the added amount of CTAB was 0.8 times CEC of Mt in the suspension. 2.4. Characterization
17
3. Results and discussion A series of APTES silylated Mts were synthesized in polar-protic (ethanol and isopropanol) and nonpolar (toluene and cyclohexane) solvents. An obvious increase in the basal spacing in silylated Mts was observed, from 1.48 nm (pristine Mt) to ca. 1.7 nm (silylated in polar-protic solvents) or to ca. 1.8 nm (silylated in nonpolar solvents) (Fig. 1). This indicates that APTES was readily intercalated into the Mt interlayer spaces, and the polarity of solvents had an influence on the interlayer structure of the silylated products [15]. Compared with FTIR spectrum of pristine Mt (Fig. 2), the occurrence of the antisymmetric stretching vibration of –CH2 at ca. 2934 cm1 and the out-of-plane deformation of –CH at 696 cm1 provided a supporting evidence for the existence of APTES in the silylated products [15,16]. In the FTIR spectra of pristine Mt and silylated products, both the vibration at ca. 1032 cm1 and the shoulder at ca. 1085 cm1 correspond to the stretching vibration of Si–O–Si [25]. A frequency shift of the vibration from 1032 cm1 (pristine Mt) to 1037 cm1 (silylated products) and a weakening of the shoulder are important information for the intercalation of APTES and consequent grafting reaction. As shown by the DTG curves of pristine Mt and the silylated products (Fig. 3), new mass loss stages are observed in the silylated products in the temperature range from 200 to 600 °C. The amount of the loaded silane was determined by calculating the mass loss in the range of 200–600 °C from thermogravimetric curves. The mass losses at ca. 200 °C, 330 °C, and 430 °C are attributed to the silanes physisorbed on Mt surfaces, covalently bonded at the broken edges and/or adsorbed on the Mt surfaces, and intercalated in the interlayer spaces, respectively [15,16,20]. The mass loss at ca. 540 °C is assigned to the decomposition of the covalent bonded silane [15,20,26]. The mass loss corresponding to the dehydroxylation of Mt almost disappeared, partly due to the consumption of hydroxyl groups during silylation reaction [20]. This provides another important evidence for the covalent bond between silane and siloxane surface of Mt. Prominent difference of the silane loaded amount in the silylated Mts prepared in different solvents could be observed as shown in Table 1. The amounts of loaded silane are 13.3% and 15.6%, respectively, in the samples prepared in ethanol and isopropanol (polar-protic solvents), while 17.4% and 17.8% in the samples prepared in toluene and cyclohexane (nonpolar solvents), respectively. The relationship between solvent polarity and silane
X-ray diffraction (XRD) analysis was performed on a Brucker D8 advanced diffractometer with a scan rate of 3.0° (2h) min1. FTIR spectra were obtained with KBr pressed disk technique using a Bruker VERTEX 70 Fourier transform infrared spectrometer. All spectra were collected at room temperature over the range 400– 4000 cm1 with a resolution of 4 cm1 and 64 scans. Thermogravimetric (TG) analysis was carried out on a Netzsch STA 409 PC/PG instrument, employing a heating rate of 10 °C min1 from 30 to 900 °C under a nitrogen flow of 60 mL min1. The percentage of the loaded silane, which corresponds to the percentage of silane molecules with respect to the total inorganic mass, was calculated as follows:
Silane loaded amount ð%Þ ¼
100 W 200—600 100 W 200—600
ð1Þ
where W200–600 corresponds to the mass loss between 200 °C and 600 °C. Solid-state 29Si CP/MAS NMR spectra were acquired using a Bruker Advanced 300 NMR spectrometer operating at 59.63 MHz. The contact time was 5 ms, the recycle delay 3 s, and the spinning rate 5.5 kHz. Tetramethylsilane (TMS) was used as the external reference.
Fig. 1. XRD patterns of pristine Mt (a) and silylated Mts prepared in ethanol (b), isopropanol (c), toluene (d), and cyclohexane (e).
18
L. Su et al. / Journal of Colloid and Interface Science 391 (2013) 16–20
Fig. 2. FTIR spectra of pristine Mt (a) and silylated Mts prepared in ethanol (b), isopropanol (c), toluene (d), and cyclohexane (e).
Fig. 3. DTG curves of pristine Mt (a) and silylated Mts prepared in ethanol (b), isopropanol (c), toluene (d), and cyclohexane (e).
Table 1 Thermogravimetric analysis of silylated Mts prepared in polar-protic and nonpolar solvents.
a b
Solvent
Dielectric constant
Mass loss (%)a
Loaded amount (%)b
Ethanol Isopropanol Toluene Cyclohexane
24.6 18.0 2.4 2.0
11.7 13.5 14.8 15.1
13.3 15.6 17.4 17.8
Mass loss between 200 °C and 600 °C. Determined by using Eq. (1).
loaded amount can be well explained on the basis of the effects of solvation and dielectric properties of solvents [27]. Solvation of reactants could affect reaction rates and mechanisms in organic reactions [28,29]. As polar-protic solvent molecules contain dipoles, they can solvate the silanol groups on Mt surfaces when solvent molecules are close to Mt surfaces. In this case, a solvent cage
is created outside silanols on Mt surfaces, making the silanols less available to react with silane [21,27]. The decrease in the silanol reactivity results in a less amount of loaded silane. For nonpolar solvents, lack of the solvation ability will facilitate the silane grafting reaction and lead to more silane loaded amount. The dielectric constant of the solvent used is another important factor to affect the amount of loaded silane [27]. In the solvent with high dielectric constant, hydrogen bond will be formed between solvent molecules and the amine groups of APTES. The strength of the hydrogen bond strongly depends on the dielectric constant of the solvent used, that is, the higher the dielectric constant, the stronger the hydrogen bond. A strong hydrogen bond will decrease the mobility of silane molecules and hinder the silanes to approach to the surface silanols of Mt and grafting reaction [21,27], resulting in a less amount of silane loaded in the silylated products. This assumption was well verified by a crosscurrent between the dielectric constants of the solvents and the silane loaded amount (Table 1). The present study showed that the silane loaded amounts decreased as: cyclohexane > toluene > isopropanol > ethanol. Swelling property of the silylated products is of high importance for their industrial applications since it controls the intercalation of macromolecules and exfoliation of clay minerals layers. Hence, a series of intercalation experiments with CTAB were conducted on pristine Mt and the silylated products to investigate their swelling property. For pristine Mt, the basal spacing increased with the increase in the intercalated surfactant amount (Fig. 4a), similar to those reported in the literature [9,30]. When the CTAB concentration increased to 2.0 CEC, the maximum basal spacing reached 3.74 nm. However, a prominent difference of the basal spacings was observed between the intercalation compounds from the silylated Mts prepared in polar-protic and nonpolar solvents. For the silylated Mts prepared in ethanol and isopropanol (polarprotic solvents), a dramatic increase in the basal spacing was recorded after CTAB intercalation (Fig. 4b and c), similar to the case of intercalating pristine Mt with CTAB. The maximum basal spacings increased to 3.45 nm and 3.72 nm for silylated Mts prepared in ethanol and isopropanol, respectively, when the CTAB concentration increased to 2.0 CEC. This implies that the silylated Mts layers remain expandable like the pristine Mt, that is, the neighboring layers are not fixed by the grafted silanes or polysiloxane oligomers in the interlayer spaces. However, for the silylated products prepared in nonpolar solvents (toluene and cyclohexane) (Fig. 4d and e), both the d001 values increased to 2.16 nm after intercalating with CTAB at 0.8 CEC. But the basal spacing remained unchanged even increasing the CTAB concentration to 2.0 CEC in the preparation system, very different from the cases of pristine Mt and the silylated Mts prepared in polar-protic solvents. This implies that the neighboring montmorillonite layers were ‘‘locked’’ during silylation, and the polarity of the solvents used has a significant effect on the silylation reaction and the structure of the silylated products. In the media of polar-protic solvents, there is a competitive adsorption between silane and solvent to approach onto the Mt surfaces. Meanwhile, the hydrogen bond between the amino groups of APTES and solvents lowers the tendency to successful silane grafting. On the other hand, the hydrolysis rate of APTES is restrained when ethanol is used as grafting solvent [31,32]. These led to a low density of intercalated silane in the interlayer spaces, as indicated by XRD and TG measurements, and a low extent of condensation among silane molecules. In this case, the montmorillonite layers of silylated products would not be locked by the grafted silane and/or polysiloxane oligomers. That is to say, the silylated Mts prepared in the polar-protic solvents kept the swelling property as pristine Mt. On the contrary, the absence of hydrogen bond between silane molecules and nonpolar solvent molecules leads to
L. Su et al. / Journal of Colloid and Interface Science 391 (2013) 16–20
19
Fig. 4. XRD patterns of pristine Mt, silylated Mts and samples intercalated with CTAB at 0.8, 1.5 and 2.0 CEC, respectively. (a) Pristine Mt. (b) Silylated Mt prepared in ethanol. (c) Silylated Mt prepared in isopropanol. (d) Silylated Mt prepared in toluene. (e) Silylated Mt prepared in cyclohexane.
a higher degree of condensation among silane molecules and between silane molecules and Mt surfaces [21,33]. In this case, a bridge of polysiloxane oligomers was formed between the adjacent layers as depicted in Fig. 5. This lock effect resulted in a loss of swelling property in the silylated Mts and hindered the efficiency of subsequent CTAB intercalation (Fig. S1).
Fig. 5. Schematic illustration of Mt silylation with APTES in various solvents.
Solid-state 29Si MAS NMR spectra gave further evidence for the influence of the solvent polarity on the structure of the silylated products. Tn notations (n = 0, 1, 2, and 3, which represents the
Fig. 6. 29Si NMR spectra of pristine Mt (a) and silylated Mts prepared in ethanol (b) and cyclohexane (c).
20
L. Su et al. / Journal of Colloid and Interface Science 391 (2013) 16–20
number of bridge ‘‘O’’ atoms surrounding the silicon atom) were used to describe the different kinds of siloxane bonds formed between APTES and montmorillonite and among APTES molecules [11,34–36]. As shown in Fig. 6a, pristine Mt displayed a resonance at 93.6 ppm, corresponding to the silicon in Si–O tetrahedral sheet of montmorillonite. After silylation reaction, a new resonance was recorded at 59.6 ppm in 29Si MAS NMR spectrum of the silylated Mt prepared in ethanol, which was attributed to the hydrolyzed bidentate (T2). For the silylated Mt prepared in cyclohexane, a predominant resonance occurred at 67.6 ppm with a shoulder at 62.5 ppm, corresponding to the hydrolyzed tridentate (T3) and hydrolyzed bidentate (T2), respectively. This indicates that the nonpolar solvents with low dielectric constant facilitate hydrolysis and condensation of silane molecules. In this case, polysiloxane oligomers will be formed in the interlayer spaces of montmorillonite and result in a ‘‘lock’’ effect on the silylated products.
4. Conclusions Silylated montmorillonites were prepared in polar-protic and nonpolar solvents. The present study demonstrated that silane was readily intercalated into the interlayer spaces of montmorillonite in both polar-protic and nonpolar solvents. Nonpolar solvents with low dielectric constant facilitate intercalation, hydrolysis, and condensation of silane molecules, resulting in a high loading of silane in the silylated products and formation of polysiloxane oligomers in the interlayer spaces. The polysiloxane oligomers linked the neighboring montmorillonite layers via covalent bond, leading to a ‘‘lock’’ effect and loss of swelling ability of the silylated products. However, in the case of polar-protic solvents with high dielectric constant, the decreased hydrolysis extent and the hydrogen bond between the amino groups of APTES and solvents led to a low loading of silane and a low degree of condensation among silane molecules. The silylated products prepared in polar-protic solvents displayed swelling ability as pristine montmorillonite. The present study showed that the polarity of the solvents used has an important influence on the structure and swelling property of the silylated products. The obtained new insights advance our fundamental understanding of the mechanism underlying the silylation of clay mineral surfaces and are of importance for synthesis and application of silylated clay minerals.
Acknowledgments This is contribution No. IS-1551 from GIGCAS. This work was financially supported by the National Natural Science Foundation of China (Grant Nos. U0933003 and 41002015), the National Key Technology R&D Program (Grant No. 2011BAB03B06), and the Guangdong Natural Science Foundation (Grant No. S2011010003599).
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.2012.08.077. References [1] S.-J. Park, B.-J. Kim, D.-I. Seo, K.-Y. Rhee, Y.-Y. Lyu, Mater. Sci. Eng. A – Struct. 526 (2009) 74. [2] S.R. Ha, K.Y. Rhee, H.C. Kim, J.T. Kim, Colloid Surf. A 313–314 (2008) 112. [3] H. Sayılkan, S. Erdemog˘lu, Sß. Sßener, F. Sayılkan, M. Akarsu, M. Erdemog˘lu, J. Colloid Interface Sci. 275 (2004) 530. [4] A. de Mello Ferreira Guimarães, V.S.T. Ciminelli, W.L. Vasconcelos, Appl. Clay Sci. 42 (2009) 410. [5] I.K. Tonle, E. Ngameni, D. Njopwouo, C. Carteret, A. Walcarius, Phys. Chem. Chem. Phys. 5 (2003) 4951. [6] M.G. da Fonseca, C. Airoldi, Mater. Res. Bull. 36 (2001) 277. [7] L. Zampori, P.G. Stampino, C. Cristiani, P. Cazzola, G. Dotelli, Appl. Clay Sci. 50 (2010) 266. [8] H.P. He, R.L. Frost, T. Bostrom, P. Yuan, L. Duong, D. Yang, Y.F. Xi, J.T. Kloprogge, Appl. Clay Sci. 31 (2006) 262. [9] H.P. He, Y.H. Ma, J.X. Zhu, P. Yuan, Y.H. Qing, Appl. Clay Sci. 48 (2010) 67. [10] E. Ruiz-Hitzky, J.J. Fripiatt, Clays Clay Miner. 24 (1976) 25. [11] A.-Y. Park, H. Kwon, A.J. Woo, S.-J. Kim, Adv. Mater. 17 (2005) 106. [12] P.A. Wheeler, J.Z. Wang, J. Baker, L.J. Mathias, Chem. Mater. 17 (2005) 3012. [13] M.G. da Fonseca, C.R. Silva, C. Airoldi, Langmuir 15 (1999) 5048. [14] S. Letaief, E. Ruiz-Hitzky, Chem. Commun. (2003) 2996. [15] H.P. He, J. Duchet, J. Galy, J.-F. Gerard, J. Colloid Interface Sci. 288 (2005) 171. [16] A.M. Shanmugharaj, K.Y. Rhee, S.H. Ryu, J. Colloid Interface Sci. 298 (2006) 854. [17] N.N. Herrera, J.-M. Letoffe, J.-L. Putaux, L. David, E. Bourgeat-Lami, Langmuir 20 (2004) 1564. [18] N.N. Herrera, J.-M. Letoffe, J.-P. Reymond, E. Bourgeat-Lami, J. Mater. Chem. 15 (2005) 863. [19] M. Park, I.-K. Shim, E.-Y. Jung, J.-H. Choy, J. Phys. Chem. Solids 65 (2004) 499. [20] F. Piscitelli, P. Posocco, R. Toth, M. Fermeglia, S. Pricl, G. Mensitieri, M. Lavorgna, J. Colloid Interface Sci. 351 (2010) 108. [21] O.G. Nik, B. Nohair, S. Kaliaguine, Microporous Mesoporous Mater. 143 (2011) 221. [22] K. Song, G. Sandi, Clays Clay Miner. 49 (2001) 119. [23] K. Wang, L. Wang, J.S. Wu, L. Chen, C.B. He, Langmuir 21 (2005) 3613. [24] A. Di Gianni, E. Amerio, O. Monticelli, R. Bongiovanni, Appl. Clay Sci. 42 (2008) 116. [25] Y.H. Ma, J.X. Zhu, H.P. He, P. Yuan, W. Shen, D. Liu, Spectrochim. Acta A 76 (2010) 122. [26] W. Shen, H.P. He, J.X. Zhu, P. Yuan, R.L. Frost, J. Colloid Interface Sci. 313 (2007) 268. [27] K.K. Sharma, A. Anan, R.P. Buckley, W. Ouellette, T. Asefa, J. Am. Chem. Soc. 130 (2008) 218. [28] T.W. Bentley, J.-P. Dau-Schmidt, G. Llewellyn, H. Mayr, J. Org. Chem. 57 (1992) 2387. [29] Y. Umebayashi, N. Shigeta, Y. Zhang, M. Komiya, Y.-i. Murakami, S.-i. Ishiguro, J. Mol. Liq. 119 (2005) 177. [30] H.P. He, R.L. Frost, F. Deng, J.X. Zhu, X.Y. Wen, P. Yuan, Clays Clay Miner. 52 (2004) 350. [31] Q. Tao, J.X. Zhu, R.L. Frost, T.E. Bostrom, R.M. Wellard, J.M. Wei, P. Yuan, H.P. He, Langmuir 26 (2010) 2769. [32] M.-C.B. Salon, M. Abdelmouleh, S. Boufi, M.N. Belgacem, A. Gandini, J. Colloid Interface Sci. 289 (2005) 249. [33] K.K. Sharma, T. Asefa, Angew. Chem. Int. Ed. 46 (2007) 2879. [34] X. Feng, G.E. Fryxell, L.-Q. Wang, A.Y. Kim, J. Liu, K.M. Kemner, Science 276 (1997) 923. [35] Q. Tao, J.X. Zhu, R.M. Wellard, T.E. Bostrom, R.L. Frost, P. Yuan, H.P. He, J. Mater. Chem. 21 (2011) 10711. [36] M.-C.B. Salon, P.-A. Bayle, M. Abdelmouleh, S. Boufi, M.N. Belgacem, Colloid Surf. A 312 (2008) 83.