Langmuir 2006, 22, 1493-1499
1493
Interaction between Silicates and Ionic Surfactants in Dilute Solution Wiliana Tjandra,† Jia Yao,‡ and Kam C. Tam*,† Singapore-MIT Alliance, School of Mechanical and Aerospace Engineering, Nanyang Technological UniVersity, 50 Nanyang AVenue, Singapore 639798, and Department of Chemistry, Zhejiang UniVersity, Hangzhou, Zhejiang ProVince, P R China 310027 ReceiVed August 3, 2005. In Final Form: NoVember 25, 2005 Understanding the interaction between silicate ions and surfactants is critical for the design and development of mesoporous siliceous materials. We examined the interaction between sodium silicate ions and three different cationic surfactants [namely, cetyltrimethylammonium bromide (CTAB), tetradecyltrimethylammonium bromide (TTAB), and dodecyltrimethylammonium bromide (DTAB)] and an anionic surfactant [sodium dodecyl sulfate (SDS)] in dilute solution at room temperature. From the combination of several techniques, such as conductometric and potentiometric titrations, dynamic light scattering, and isothermal titration calorimetry, the phase behavior of the sodium silicate and CTAB system was determined. We observed that the aggregation behavior of the silicate-CTAB system is similar to that of a polymer-surfactant system. The formation of the silicate-CTAB complex is induced by the adsorption of SiOH and SiO- groups, aided by CTAB unimers. The electrostatic attraction and hydrophobic interaction are the dominant forces controlling the formation of silicate-CTAB complexes. When these complexes are saturated with CTAB unimers, free CTAB micelles are then produced. TEM micrographs revealed that a stable Si-O-Si network is absent within the silicate-CTAB complexes, and surprisingly, stable silicate-CTAB complexes with ordered structure were observed. The present finding is important for understanding the interaction between silicate and surfactant in the synthesis of mesoporous structure in the dilute solution regime.
Introduction Understanding the interaction between silicate ions and surfactants is critical for the design and development of supramolecular siliceous mesoporous materials, for e.g. the M41S family of materials. In 1992, scientists at Mobil Oil Company first synthesized organized siliceous mesoporous MCM-41, which possesses a hexagonal arrangement of uniform mesopores with tunable pore sizes of 15 to 100 Å, high surface area (above 700 m2/g), and high hydrocarbon sorption capacity (above 0.7 cm3/ g).1 The earliest approach in the development of mesoporous materials is based on the liquid crystal templating (LCT) process, where the surfactants with condensed silicate ions self-assembled into hexagonal arrays.2 Several mechanisms based on the promotion of the hexagonal LC phase by silicate ions at surfactant concentrations lower than the cmc have been proposed. Some of the most common mechanisms are based on ion-exchange between surfactant counterions and silicate ions, where the silicate ions condensed on preformed micelles, as confirmed by in-situ nuclear magnetic resonance (NMR) coupled with neutron scattering measurements.3 Progress made in the field of MCM41 chemistry and proposed formation mechanisms, such as the LCT templating mechanism, the transformation mechanism from lamellar to hexagonal phase, and the “folded sheet” mechanism (based on intercalation of surfactant to the kanemite process), were reviewed by Zhao et al. in 1996.4 Other in-situ methods, * To whom correspondence should be addressed. Fax: (65) 6791 1859. E-mail:
[email protected]. † Nanyang Technological University. ‡ Zhejiang University. (1) Beck, J. S.; Vartuli, J. C.; Roth, W. J.; Leonowicz, M. E.; Kresge, C. T.; Schmitt, K. D.; Chu, C.-T-W.; Olson, D. H.; Sheppard, E. W.; McCullen, S. B.; Higgins, J. B.; Schlenker, J. L. J. Am. Chem. Soc. 1992, 114, 10834-19843. (2) Ying, J. Y.; Mehnert, C. P.; Wong, M. S. Angew. Chem., Int. Ed. 1999, 38, 56-77. (3) Firouzi, A.; Kumar, D.; Bull, L. M.; Besier, T.; Sieger, P.; Huo, Q.; Walker, S. A.; Zasadzinski, J. A.; Glinka, C.; Nicol, J.; Margolese, D.; Stucky, G. D.; Chmelka, B. F. Science 1995, 267, 1138-1143. (4) Zhao, X. S.; Lu, G. Q.; Millar, G. J. Ind. Eng. Chem. Res. 1996, 35, 2075-2090.
such as electron paramagnetic resonance (EPR) and small angle X-ray scattering (SAXS), have been used to provide additional information to support the proposed mechanisms. Analysis of the EPR spectra showed a decrease in the mobility of surfactant micelles throughout the reaction, which is attributed to the formation of an inorganic lattice at the micellar surface. From XRD results, the reduction in the mobility of micelles corresponds to the transformation of an amorphous solid into an ordered hexagonal micelle-templated material.5 Investigation using a spin probe that detects the diffusional changes of surfactant molecules during the formation of silica based MCM-41 was also performed. During the course of the reaction, two clear stages were observed. In the first stage of the reaction, silicate oligomers are rapidly produced in aqueous solution, and they interact with the micellar interface, which then transforms the micelles into a hexagonal array of partially condensed silicates. The latter undergoes further adsorption and dehydration during the second slow stage of the reaction.6 In-situ SAXS experiments confirmed the formation of a bicontinuous cubic phase, MCM-48. Changes in the surfactant packing parameter during the course of reaction were observed, which resulted in a hexagonal-to-lamellar-to-hexagonal-to-cubic phase sequence. The cubic phase resulting from the collapse of the hexagonal phase was only produced after sintering.7 In 2000, an in-situ study using fluorescence spectroscopy suggested that ion-exchange is not an important mechanism for the formation of the mesoporous structure. Therefore, a new mechanism based on polyelectrolyte/oppositely charged surfactant systems was proposed, where the formation of siliceous pre-polymer is the controlling step. As the pre-polymer grows, it interacts with a larger number of surfactant molecules, in a cooperative manner to produce silicate-surfactant hybrid micellar aggregates. Further (5) Ottaviani, M. F.; Galarnneau, A.; Desplantier-Giscard, D.; Di Renzo, F.; Fajula, F. Microporous Mesoporous Mater. 2001, 44-45, 1-8. (6) Zhang, J.; Goldfarb, D. Microporous Mesoporous Mater. 2001, 48, 143149. (7) Pevzner, S.; Regev, O. Microporous Mesoporous Mater. 2000, 38, 413421.
10.1021/la0521185 CCC: $33.50 © 2006 American Chemical Society Published on Web 01/17/2006
1494 Langmuir, Vol. 22, No. 4, 2006
polymerization and organization of the hybrid complexes takes place during the precipitation-aging step, to yield the mesoporous silica.8 A simulation using the off-lattice Monte Carlo method in two dimensions was recently performed to examine the selfassembly of ionic surfactants. This was extended to the micellization of ionic surfactants in the presence of other neutral host particles. The morphologies of the surfactant-host composites were studied as a function of host particle densities, sizes, and their interaction strengths. The host was added to two extreme initial surfactant conditions: the first to preformed micellar structures, and the second to a random distribution of surfactants and host particles. Formation of ordered mesoporous material for both cases was demonstrated, confirming this to be the true equilibrium state.9 These proposed mechanisms were supported by experimental results; however, a definite and conclusive mechanism is still being debated. It is important to point out that previous investigations were conducted under alkaline conditions, where the silica is anionic in character. In diluted silicate solution, where the dilution of a concentrated alkali silicate solution reduces the concentration of hydroxyl ions and the pH, these changes cause the polymeric species to dissociate into silicate anions.10 Studies are needed to support and identify the interaction mechanism between the ionic species present in the reaction mixtures. To further elucidate the formation mechanism at high pH, we focused on the interaction between sodium silicate ions and various cationic surfactants, i.e., cetyltrimethylammonium bromide (CTAB), tetradecyltrimethylammonium bromide (TTAB), and dodecyltrimethylammonium bromide (DTAB), and the anionic surfactant sodium dodecyl sulfate (SDS) in the dilute solution regime and under room-temperature conditions. There are some published studies conducted in the dilute solution region, such as the formation of rodlike silica using CTAB and TEOS,11 adsorption microcalorimetry to study interfacial aggregation of quaternary ammonium surfactants on powdered silica supports,12 and the kinetic study of silica oligomerization and nanocolloid formation as a function of pH and ionic strength.13 Surprisingly, we found that the silicate-surfactant systems possess similar features to polymer-surfactant systems, and the interaction is dominated by the adsorption of surfactants to the silica surface.14-16 In the present study, the chemistry of silica in an alkaline medium will be examined,17 where a new mechanism based on the cooperative binding between surfactant unimers and sodium silicate is proposed. Experimental Section Materials. CTAB, TTAB, DTAB, SDS, and sodium silicate were used as supplied (Sigma-Aldrich). The sodium silicate consists of 27% SiO2 and 10% NaOH by weight % (molar ratio 1.8:1). Sample solutions were prepared by dilution in deionized water to the desired concentration. Synthesis of Mesoporous Sodium Silicate. Sodium silicate (2.5 g, 50 mM) was prepared in 100 g of deionized water, and the pH (8) Frasch, J.; Lebeau, B.; Soulard, M.; Patarin, L.; Zana, R. Langmuir 2000, 16, 9049-9057. (9) Bhattacharya, A.; Mahanti, S. D. J. Phys. Condens. Matter 2001, 13, L861L869. (10) Boschel, D.; Janich, M.; Roggendorf, H. J. Colloid Interface Sci. 2003, 267, 360-368. (11) Cai, Q.; Luo, Z. S.; Pang, W. Q. Chem Mater. 2001, 13, 258-268. (12) Zajac, J. Colloids Surf., A: Physicochem. Eng. Aspects. 2000, 167, 3-19. (13) Icopini, G. A.; Brantley, S. A.; Heaney, P. J. Geochim. Cosmochim. Acta 2005, 69, 293-303. (14) Subramanian, V.; Ducker, W. J. Phys. Chem. B 2001, 105, 1389-1402. (15) Golub, T. P.; Koopal, L. P.; Sidorova, M. P. Colloid J. 2004, 66, 38-43. (16) Zajac, J.; Trompette, J. L.; Partyka, S. Langmuir 1996, 12, 1357-1367. (17) Iler, R. K. The Chemistry of Silica: Solubility, Polymerization, Colloid and Surface Properties and Biochemistry; John Wiley & Sons: New York, 1979.
Tjandra et al. was adjusted to ∼11 by adding NaOH. Cetyltrimethylammonium bromide (CTAB, CH3(CH2)15(CH3)3NBr, 0.9 g, 12.5 mM) was prepared in 100 g of water and added dropwise to the sodium silicate solution under constant stirring. The pH was controlled during mixing by the addition of NaOH. This combined mixture was then heated at 100 °C with stirring for 24 h. The resulting white solid products were recovered and calcinated in air at 500 °C for 6 h using a furnace. Conductometric and Potentiometric Titrations. The conductivity and pH of the solutions were measured using a Radiometer Copenhagen ABU93 Triburette Titration System. All the titrations were performed at 25 °C with a 1 min waiting period between each titration. Dynamic Light Scattering (DLS). The Brookhaven laser light scattering system was used. The equipment consists of a BI200SM goniometer, a BI-9000AT digital correlator, and other supporting data acquisition and analysis software and accessories. A 200-mW, argon-ion, vertically polarized, 488-nm laser was used as the light source. The G2(t) functions obtained from DLS were analyzed using the inverse Laplace transformation technique (REPES in this case) to produce the distribution function of decay times. τ was obtained from the mean of each peak on the distribution function. All the measurements were carried out at 45, 60, 75, and 90° for a given sample to ensure the presence of true particles. The scattering vector q was calculated using q ) [4πn0 sin(θ/2)]/λ where λ is the wavelength in the medium. The relaxation rate Γ was determined from the inverse of τ and plotted against q2 for the four angles to yield a slope corresponding to the diffusion coefficient D. Rh was calculated using the Stokes-Einstein equation, where Rh ) kT/6πηD for samples taken during the titration experiments and measured 24 h after sampling. Isothermal Titration Calorimetry (ITC). The calorimetric data were obtained using a Microcal isothermal titration calorimeter (ITC). It has a reference and a sample cell of approximately 1.35 mL, and the cells are both insulated by an adiabatic shield. The titration was carried out at 25 °C. The syringe is tailor-made such that the tip acts as a blade-type stirrer to ensure an optimum mixing efficiency at 400 rpm. An injection schedule was automatically carried out using interactive software after setting up the number of injections, the volume of each injection, and the time between each injection. Morphology Analysis Using Transmission Electron Microscopy (TEM). The morphology of the samples was examined using TEM (with a JEOL 2010 instrument, 200 kV). The samples were prepared by evaporating a drop of the sample solution on a carbon coated copper grid under vacuum. Samples were taken 24 h after mixing and dried thoroughly in a desiccator for 24 h after preparation. The sample in region E was prepared by evaporating a drop of dispersed, ground dried powder in ethanol. It was then dried thoroughly in a desiccator for 24 h. Small Angle X-ray Diffraction. Small-angle X-ray diffraction measurements were conducted using a STD-XRD, Standard attachment XRD 6000, Shimadzu, Cu-Ka tube operated at 50 kV and 20 mA. The sample was a calcinated dried powder in region E. Nuclear Magnetic Resonance. Pure sodium silicate, stage A and stage B NMR studies were conducted using liquid state 29Si (Bruker 400 MHz), with a 1D sequence with power gated decoupling as the pulse program and 8000 as the number of scans with a run time of about 13 h. Unheated sample at stage E was studied by solid state 29Si (Bruker DRX400 MHz) with 400 scans, a single pulse program, a spinning rate of 10 000, and a relaxation delay of 20 s.
Results and Discussion Polymer-surfactant interaction is known to induce micellization of surfactant by screening the charge repulsion between surfactant head groups.18 The interaction between surfactants and polymers can be described by two critical concentrations: The first is the critical aggregation concentration (cac), which corresponds to the formation of polymer-surfactant aggregation (18) Wang, C.; Tam, K. C. Langmuir 2002, 18, 6484-6490.
Silicates and Ionic Surfactants in Dilute Solution
Langmuir, Vol. 22, No. 4, 2006 1495
complexes, where surfactant micelles are induced on the polymeric chains. The second critical concentration, Cs, is commonly used to represent the surfactant concentration when the polymer becomes saturated with surfactant aggregates. In addition, another critical concentration, cmc, represents the formation of free surfactant micelles in the polymer solution. In some systems, free surfactant micelles start to form after Cs, and under these conditions, cmc is analogous to Cs. However, for other systems, the formation of free surfactant micelles precedes Cs, where cmc is less than Cs; thus, there is a competition between the formation of free surfactant micelles and polymer-surfactant aggregation complexes. When cationic surfactants were titrated into dilute silicate solution (at high pH conditions), two transition concentrations, i.e., cac and Cs, were observed, which is consistent with the case for polymer-surfactant systems. Such behavior may be related to the chemical equilibria (eqs 1-4) in the silicate system under alkali conditions, which control the formation of silicate-surfactant complexes.
Si2O52- + H2O h 2HSiO3-
(1)
HSiO3- + 2H2O h Si(OH)4 + OH-
(2)
2Si(OH)4 h (HO)3SiOSi(OH)3 + H2O
(3)
SiOH + SiO- h Si-O-Si + OH-
(4)
Initially, the sodium silicates are mainly in the form of polymeric ions, and by direct dilution at pH greater than 7, the polymeric ions are hydrolyzed via eqs 1 and 2.17 Si(OH)4 polymerizes by self-condensation according to eq 3, where the dimers grow in a very short time to a 3-dimensional structure of diameter 1-2 nm via the adsorption of SiOH groups (eq 4) and OH- ions are released. The 3D structure becomes a “nucleus” for the formation of colloidal silica with diameter smaller than 5 nm comprised of a SiO2 core stabilized at the surface by SiOH. Partial ionization of OH groups to SiO- makes the colloidal silica anionic;17 hence, cationic surfactant molecules are adsorbed on the silica surface via the double layer phenomena. We hypothesized that SiO- groups are drawn to the cationic surfactant unimers8 via long-range electrostatic forces, resulting in a higher local concentration of silicate ions around the surfactant head groups, which induces the adsorption of silicate ions. The onset for such interaction is defined by the critical adsorption concentration, C1. The hydrophobic interaction between the alkyl chains on the surfactant produces silicate-surfactant complexes, defined by the critical aggregation concentration, cac, until the surfactant concentration reaches the critical micelle concentration (cmc). Then the surfactant molecules within the complex would have completely reorganized to form micelles. The cmc in the silicate-surfactant system corresponds to Cs in the polymersurfactant system. The formation of discrete micelles occurs just after this critical point (cmc g Cs). Conductometric and Potentiometric Titrations. The changes in the pH during the titration indicate that the presence of surfactant molecules enhances the interaction of SiOH and SiO- groups, which promotes the hydrolysis of silicate ions. The hydrolysis of silicate ions in the absence of CTAB only produces minute quantities of 1-5 nm particles that cannot be detected by DLS measurements. Figure 1 shows changes in [OH-] concentration during the titration of 50 mM CTAB into a 3 mM sodium silicate solution. We observed that the [OH-] concentration increases from ∼0.07 to 0.21 mM when the CTAB concentration was increased from
Figure 1. Changes in OH- concentration when CTAB was titrated into 3 mM sodium silicate solution.
Figure 2. Changes in conductivity when CTAB was titrated into sodium silicate solutions: (b) 0 mM; (]) 1 mM; (2) 3 mM; (0) 5 mM.
0 to 1.0 mM. Such an increase suggests that silicate ions undergo additional adsorption aided by CTAB according to eq 4. As the silica surface contains negative charges, cationic surfactants are adsorbed on the surface by electrostatic attraction to form a hydrophobic layer, which induces the formation of aggregates. Further addition of surfactant molecules promotes the adsorption of more CTAB molecules, resulting in an increase in the size of the complex, and the aggregates are flocculated and bridged to produce hemi-micelles induced by hydrophobic forces.18,19 When the CTAB concentration was increase to 0.8 mM (corresponding to the cmc of CTAB), the concentration of OH- ions remained constant, since “free” CTAB micelles, which do not induce large complex formation, are produced at cmc. We deduced from Figure 1 that two critical concentrations are present, namely, the critical adsorption concentration (C1) of 0.1 mM, where CTAB molecules are adsorbed on the silicate pre-polymer, and the cmc of 0.8 mM, signifying the formation of “free” CTAB micelles on the silicate-CTAB complexes. Changes in the conductivity were measured during the titration of CTAB into different concentration (0-5 mM) sodium silicate solutions at 25 °C, and the results are shown in Figure 2. As a comparison, the cmc of CTAB without silicate at 25 °C is 1 mM.20 We observed an obvious inflection (indicated by the arrows for 0 and 5 mM sodium silicate solutions) in each conductivity curve, which corresponds to the change in the concentration of mobile surfactant molecules. Each of the inflection point marks the cmc of CTAB at varying amounts of sodium silicate. In the silicate-CTAB system, a silicate-CTAB complex was produced (19) Hunter, R. J. Zeta Potential in Colloid Science: Principles and Applications; Academic Press: New York, 1981. (20) Myers, D. Surfactant Science and Technology; VCH: Republic of Argentina, 1988.
1496 Langmuir, Vol. 22, No. 4, 2006
Tjandra et al.
Figure 3. Changes in OH- concentration when surfactants were titrated into sodium silicate: (0) DTAB to 3 mM sodium silicate; (9) DTAB to 1 mM sodium silicate; (]) SDS to 2 mM sodium silicate; ([) SDS to 1 mM sodium silicate; (b) TTAB to 1 mM sodium silicate; (2) CTAB to 1 mM sodium silicate. Table 1. Particle Sizes for Different Concentrations of DTAB and TTAB in 2 mM Sodium Silicate Solution surfactant conc (mM) in 2 mM sodium silicate 1 5 20 50
DTAB intensity at 90° Rh (nm) 8 10 25 20
undetected undetected 292 328
TTAB intensity at 90°
Rh (nm)
4 17 41 24
undetected 282 291 350
Figure 4. Relaxation time distribution functions of (a) 0.5 mM CTAB in 2 mM sodium silicate and (b) 5 mM CTAB in 2 mM sodium silicate. Table 2. Intensity Counts for Different Concentrations of CTAB in 2 mM Sodium Silicate Solution CTAB conc (mM) in 2 mM sodium silicate
prior to the formation of “free” CTAB micelle; hence, the inflection point corresponds to formation of CTAB micelles which do not induce the formation of a large complex on existing silicate-CTAB complexes. The critical CTAB concentration represented by the inflection point decreases with increasing silicate content, and this is attributed to the screening of surface charges of CTA+ by silicate ions. At 5 mM sodium silicate, precipitation of the complexes prevents further binding of CTAB onto the silicate-CTAB complexes; hence, a slightly larger slope after the inflection point was observed. Further evidence on the hydrophobic binding between surfactant molecules and silica is shown in Figure 3, where the release of OH- ions was observed for TTAB (alkyl chains with 14 carbon atoms) and DTAB (alkyl chains with 12 carbon atoms) surfactant systems. At a fixed concentration of 1 mM sodium silicate, the silicate/DTAB complex exhibits the lowest hydrolysis compared to silicate/TTAB and silicate/CTAB complexes since the interaction is the weakest. Beyond the cmc, we observed a drop in the OH- concentration, caused by the dilution effect. We note that, below the cmc, the profiles for the DTAB and TTAB systems exhibit two slopes (as indicated by the dotted lines). The first slope represents the dominant electrostatic interaction, which enhances hydrolysis of SiOH, as reflected by the gradient of the slope, resulting in the release of a larger fraction of OH- ions at a fixed surfactant concentration. The second slope represents the hydrophobic attraction between the CTA+ ions, which leads to larger aggregates, whose size depends on the magnitude of hydrophobic forces as shown in Tables 1 and 2. Aggregates of silicates produced with DTAB and TTAB are less compact and larger in size compared to the case of the stronger interaction of CTAB, which produces more compact aggregates, as reflected by the scattering intensities. The profile for the CTAB system only exhibits one slope, as CTAB is the most hydrophobic surfactant where the hydrophobic and electrostatic interaction regimes overlap with each other. Since SDS is an anionic
0.1 0.5 1.0 5.0 20.0
intensity (kcps) at the following scattering angles 45° 60° 75° 90° 48 269 262 118 40
23 206 166 87 28
18 165 139 73 23
15 150 123 66 20
Rh (nm) undetected 43 45 42 46
surfactant, no interaction with silica is expected, as both systems possess similar charges. Therefore, negligible release of OHions was observed. Dynamic Light Scattering (DLS). Samples for 0.1, 0.5, 1, 5, 20, and 50 mM CTAB in 1, 2, 3, 4, and 5 mM sodium silicate titrations were measured using DLS. Beyond 0.5 mM CTAB, the scattering intensity decreases with increasing CTAB concentrations, as summarized in Table 2. However, the particle size of the complex remained constant; hence, the probable reason for the reduction in the scattering intensities without any observable change in the particle size is a lower refractive index, as complexes with higher CTAB content are less compact. Parts a and b of Figure 4 show the distribution functions at various scattering angles for mixtures of 0.5 and 5 mM CTAB with 2 mM sodium silicate solutions, respectively. In the 0.5 mM CTAB-2 mM sodium silicate system, only one peak corresponding to a silicate-CTAB complex with a hydrodynamic radius, Rh, of 43 nm was observed. For 5 mM CTAB in 2 mM sodium silicate solution, two peaks were observed corresponding to an Rh of 1.2 nm (CTAB micelle) and one for the silicateCTAB complex of 42 nm. The literature Rh of CTAB at 25 °C is 1.5 nm.20 Isothermal Titration Calorimetry (ITC). ITC was conducted by titrating 50 mM CTAB into a 2 mM sodium silicate solution. The enthalpies for the titration of CTAB into water and CTAB into sodium silicate solution are shown in Figure 5. In the low concentration region (
