pubs.acs.org/Langmuir © 2009 American Chemical Society
Adsorption of the Cationic Surfactant Cetyltrimethylammonium Bromide to Silica in the Presence of Sodium Salicylate: Surface Excess and Kinetics Shaun C. Howard and Vincent S. J. Craig* Department of Applied Mathematics Research School of Physics and Engineering Australian National University, Canberra, ACT 0200, Australia Received May 27, 2009. Revised Manuscript Received September 13, 2009 The adsorption of cetyltrimethylammonium bromide (CTAB) to silica in the presence of sodium salicylate has been investigated using atomic force microscopy, optical reflectometry, and a quartz crystal microbalance. Salicylate is found to have a dramatic influence on surface adsorption in terms of the kinetics, surface excess, structure of adsorbed aggregates and the mechanical rigidity of the adsorbed film. This is consistent with the bulk solution behavior of more concentrated CTAB-salicylate solutions and reflects the higher local concentration induced by adsorption to the silica surface. Slow adsorption kinetics are found over a wide range of concentrations below the critical micelle concentration.
Introduction The adsorption of surfactants to solid surfaces is critical in detergency,1 froth flotation,2 wetting and surface passivation.3 The morphology of the adsorbed surfactant layer is strongly influenced by the substrate. When the surface is initially hydrophobic, surfactant molecules adsorb tail first in order to minimize the exposure of hydrophobic groups to the aqueous phase, and at maximum coverage a single layer of surfactant is formed.4-6 For hydrophilic surfaces, the tendency to minimize exposure of hydrophobic groups to the aqueous phase gives rise to structures that are fundamentally different, favoring bilayered arrangements at high coverages. The bilayered structures that are found on the surface are, to differing extents, influenced by the substrate. For cationic surfactants, a highly negatively charged crystalline surface such as mica acts as a template and strongly influences the arrangement of the aggregates,7 whereas silica with an amorphous moderately charged surface has a weak templating effect.8 On silica, the type of aggregates observed often parallel those found in solution at slightly higher concentrations. That is, the surface structures of spheres, rods, and bilayers can be seen as analogous to the spheres, rods, and vesicles that form in solution at higher concentrations,9 and the degree of counterion binding is similar between the surface and the bulk.10 Most single-chain cationic surfactants form micellar-like aggregates in the bulk at concentrations up to several times the critical micelle concentration (cmc), whereas the addition of low concentrations of salicylate (o-hydroxybenzoate) results in the formation of wormlike micelles (wlm’s) in the bulk, hence similar aggregate structures might be expected at the silica-solution interface. Consistent with *Corresponding author. E-mail:
[email protected]. (1) Paria, S.; Khilar, K. C. Adv. Colloid Interface Sci. 2004, 110(3), 75–95. (2) Pugh, R. J. Colloids Surf. 1986, 18(1), 19–41. (3) Muster, T. H.; Neufeld, A. K.; Cole, I. S. Corros. Sci. 2004, 46(9), 2337–2354. (4) Wanless, E. J.; Ducker, W. A. J. Phys. Chem. 1996, 100(8), 3207–3214. (5) Atkin, R.; Warr, G. G. J. Am. Chem. Soc. 2005, 127(34), 11940–11941. (6) Kiraly, Z.; Findenegg, G. H. J. Phys. Chem. B 1998, 102(7), 1203–1211. (7) Ducker, W. A.; Wanless, E. J. Langmuir 1999, 15(1), 160–168. (8) Tyrode, E.; Rutland, M. W.; Bain, C. D. J. Am. Chem. Soc. 2008, 130(51), 17434–17445. (9) Mitchell, D. J.; Ninham, B. W. J. Chem. Soc., Faraday Trans 2 1981, 77, 601– 629. (10) Hayes, P. L.; Chen, E. H.; Achtyl, J. L.; Geiger, F. M. J. Phys. Chem. A 2009, 113(16), 4269–4280.
Langmuir 2009, 25(22), 13015–13024
this is the report that the addition of salicylate to a cetylpyridinium chloride (CPC) solution results in a greater level of surfactant adsorption.11,12 The addition of electrolyte,13 specifically the salicylate anion, reduces the surfactant headgroup area by screening the repulsive charge, leading to an increase in the critical packing parameter14 that favors less highly curved aggregates. An additional property of the salicylate ion is that it orients such that the aromatic carbon atoms are within the hydrocarbon core of the aggregate14,15 thus increasing the effective volume of the hydrocarbon chains and the critical packing parameter.14,16 Hence the addition of a small amount of salicylate has a strong influence on surfactant structure16,17 and rheology.17-19 Notably, other additives of similar structure such as p-hydroxybenzoate17,20 are not as effective, as these ions do not enter the hydrocarbon core of the aggregates.14 Recently it has been revealed that cation-π interactions between the charged surfactant head groups and the benzene ring of the salicylate molecule play a significant role.21-23 The benzene ring is a quadrupole, and the faces of the ring carry a partial negative charge. This negative charge is able to interact with cations, and this is seen in many biological systems.24 As the electron density is increased on either side of the planar benzene, there is a possibility for it to interact with two cetyltrimethylammonium (CTAþ) head groups that sit on either side, effectively pulling the surfactant head groups together. This effect is also favored by the neutralizing charge on the carboxyl group, but this (11) Favoriti, P.; Mannebach, M. H.; Treiner, C. Langmuir 1996, 12(20), 4691– 4696. (12) Favoriti, P.; Treiner, C. Langmuir 1998, 14(26), 7493–7502. (13) Israelachvili, J. N.; Mitchell, D. J.; Ninham, B. W. J. Chem. Soc., Faraday Trans. 2: Mol. Chem. Phys. 1976, 72, 1525–1568. (14) Penfold, J.; Tucker, I.; Staples, E.; Thomas, R. K. Langmuir 2004, 20(19), 8054–8061. (15) Srinivasan, V.; Blankschtein, D. Langmuir 2003, 19(23), 9946–9961. (16) Rao, U. R. K.; Manohar, C.; Valaulikar, B. S.; Iyer, R. M. J. Phys. Chem. 1987, 91(12), 3286–3291. (17) Gravsholt, S. J. Colloid Interface Sci. 1976, 57(3), 575–577. (18) Rehage, H.; Hoffmann, H. J. Phys. Chem. 1988, 92(16), 4712–4719. (19) Imae, T. J. Phys. Chem. 1990, 94(15), 5953–5959. (20) Lin, Z.; Cai, J. J.; Scriven, L. E.; Davis, H. T. J. Phys. Chem. 1994, 98(23), 5984–5993. (21) Frounfelker, B. D.; Kalur, G. C.; Cipriano, B. H.; Danino, D.; Raghavan, S. R. Langmuir 2009, 25(1), 167–172. (22) Wu, H.; Kawaguchi, S.; Ito, K. Colloid Polym. Sci. 2005, 283(6), 636–645. (23) Nakamura, K.; Shikata, T. Macromolecules 2004, 37(22), 8381–8388. (24) Dougherty, D. A. Science 1996, 271(5246), 163–168.
Published on Web 10/05/2009
DOI: 10.1021/la901889m
13015
Article
Howard and Craig
effect alone cannot explain the charge reversal seen in these systems.18 We note that the strength of the quadrupole is strongly affected by substituents on the ring. Some groups strongly activate the ring (i.e., increase the electron density in the ring), while others deactivate the ring (i.e., withdraw density). Further, for induction and resonance reasons, any substituent in the ortho position of benzoic acid makes the carboxyl group more acidic. This is especially true for substituents that are also strongly activating, such as hydroxyl groups, hence the low pKa (∼3) of salicylate. In the meta position, the hydroxyl group is only weakly activating, but in the para position it is weakly deactivating. So, in other words, the greatest electron density in the ring occurs for the o-hydroxy isomer (i.e., salicylate). This explains why the isomers of salicylate are less active. Chloride substitution is an exception, in that, it is deactivating in the ortho position but weakly activating for the meta and para positions (viscoelasticy produced in the order ortho < meta < para). This also corresponds to the observed behavior of the chlorobenzoates.17 In this manuscript we are interested in evaluating how the addition of salicylate alters both the surface aggregate structures and the surface excess of cetyltrimethylammonium bromide (CTAB) at the silica surface and whether the adsorption and desorption kinetics are influenced by the presence of salicylate. In previous studies of the adsorption of cationic surfactants to silica, it was demonstrated that, below the cmc, adsorption can be very slow (taking many minutes to hours), and it was argued that this slow adsorption is linked to the surface aggregates that are present.25-27 This has now been refined, and the proposal is that equilibration is slow because the surface aggregates are laterally immobile and hence unable to collide. This removes aggregate fusion as a means of equilibration leaving monomer diffusion to drive equilibration, which is slow near equilibrium.28 In the presence of salicylate, we expect that the morphology of these aggregates will be much altered, and this may have implications for the adsorption kinetics.
Materials and Methods CTAB (purity greater than 99%) was obtained from Aldrich, recrystallized twice from an acetone/ethanol mixture, and freezedried prior to use. Sodium bromide (NaBr) and sodium salicylate (NaSal) were obtained from Aldrich and used as received (Analytical grade, 99.9þ%). All water used was filtered and passed through a Millipore Gradient filtration unit before use. The surface tension of the water at room temperature was measured periodically by the pendant drop technique (KSV Instruments, Ltd.) and found to be 73.0 ( 2 mN m-1 at approximately 20 °C. The cmc for CTAB in the absence of salt (0.89 mM) was determined from the break in the surface tension plotted against the log of concentration using data obtained via the pendant drop method (KSV CAM-300, Finland). No minima were found in the surface tension data around the cmc, indicating no gross contamination of the surfactant. Attempts to determine the cmc from surface tension measurements for solutions containing 1 mM salicylate were not successful, as it was found that the surface tension of these solutions did not equilibrate on the scale of minutes. Surface tension measurements for the 20:1 CTAB/ NaSal system showed evidence of a cmc between 0.5 and 0.6 mM CTAB; however, the minima that results from the partitioning of the salicylate into micelles precluded more accurate determination of the cmc by this method. Instead we have determined the cmc by the break from linearity in the electrical conductivity as a function (25) Atkin, R.; Craig, V. S. J.; Biggs, S. Langmuir 2000, 16(24), 9374–9380. (26) Atkin, R.; Craig, V. S. J.; Biggs, S. Langmuir 2001, 17(20), 6155–6163. (27) Atkin, R.; Craig, V. S. J.; Wanless, E. J.; Biggs, S. Adv. Colloid Interface Sci. 2003, 103(3), 219–304. (28) Howard, S. C.; Craig, V. S. J. Soft Matter 2009, 5, 3061.
13016 DOI: 10.1021/la901889m
of concentration to be 0.51 mM for the 20:1 CTAB/NaSal system and 0.45 mM for the 10:1 CTAB/NaSal system. This data has been included as Supporting Information. Pieces of silicon wafer (50 mm 10 mm, root-mean-square (rms) roughness 0.6 nm over 10 μm 10 μm) with a well-defined madeto-order 319 nm oxide layer were used in all studies (Silicon Valley Microelectronics, CA). The thickness of the oxide layer present on the silicon wafer was confirmed ellipsometrically to be 319 ( 2 nm (Beaglehole instruments). The oxide layer is grown at high temperature (>500 °C); this produces surfaces with low hydroxyl group density due to condensation reactions at the silica surface that result in the formation of siloxane bonds.29 The remaining hydroxyl groups are isolated and therefore less likely to participate in hydrogen bonded stabilization of hydronium ions at the surface.29 In solution, these hydroxyl groups are therefore more acidic, and the silica surface will be more highly charged. Silica of this type is known as pyrogenic. Pyrogenic silica will slowly rehydroxylate when immersed in water resulting in hydroxylated silica. In this work, the initially pyrogenic silica was cleaned with Piranha solution (H2O2 80% w/w; 20% w/w HNO3) for
