Adsorption of Nonionic Surfactants on Cellulose Surfaces: Adsorbed

compounds by microchip capillary electrophoresis with pulsed amperometric detection. Yongsheng Ding , Maria F. Mora , Grant N. Merrill , Carlos D...
8 downloads 0 Views 141KB Size
7768

Langmuir 2005, 21, 7768-7775

Adsorption of Nonionic Surfactants on Cellulose Surfaces: Adsorbed Amounts and Kinetics Lambertus H. Torn, Luuk K. Koopal,* Arie de Keizer, and Johannes Lyklema Laboratory of Physical Chemistry and Colloid Science, Wageningen University, Dreijenplein 6, 6703 HB Wageningen, The Netherlands Received April 25, 2005. In Final Form: June 9, 2005 Kinetic and equilibrium aspects of three different poly(ethylene oxide) alkylethers (C12E5, C12E7, C14E7) near a flat cellulose surface are studied. The equilibrium adsorption isotherms look very similar for these surfactants, each showing three different regions with increasing surfactant concentration. At low surfactant content both the headgroup and the tail contribute to the adsorption. At higher surface concentrations, lateral attraction becomes prominent and leads to the formation of aggregates on the surface. The general shape of the isotherms and the magnitude of the adsorption resemble mostly those for hydrophilic surfaces, but both the ethylene oxide and the aliphatic segments determine affinity for the surface. The adsorption and desorption kinetics are strongly dependent on surfactant composition. At bulk concentrations below the CMC, the initial adsorption rate is attachment-controlled. Above the CMC, the micellar diffusion coefficient and the micellar dissociation rate play a crucial role. For the most hydrophilic surfactant, C12E7, both parameters are relatively large. In this case, the initial adsorption rate increases with increasing surfactant concentration, also above the CMC. For C12E5 and C14E7 there is no micellar contribution to the initial adsorption rate. The initial desorption kinetics are governed by monomer detachment from the surface aggregates. The desorption rate constants scale with the CMC, indicating an analogy between the surface aggregates and those formed in solution.

1. Introduction The adsorption of nonionic surfactants at solid/liquid interfaces is of great practical importance, and it has therefore been extensively studied.1-15 Not in the least due to the availability of homodisperse poly(ethylene oxide) alkylethers, usually denoted as CnEm, their adsorption behavior is well understood nowadays. Some of the studies have concerned hydrophilic6-11 surfaces, others hydrophobic12-17 surfaces. * Corresponding author. Mailing address: Laboratory of Physical Chemistry and Colloid Science, Wageningen University, P.O. Box. 8038, 6700 EK Wageningen, The Netherlands. E-mail: [email protected]. (1) Clunie, J. S.; Ingram, B. T. In Adsorption from Solution at the Solid/Liquid Interface; Parfitt, G. D., Rochester, C. H., Eds.; Academic Press: New York, 1983; p 105. (2) Von Rybinsky, W.; Schwuger, M. J. In Non-ionic Surfactants; Surfactant Science Series, Vol. 23;Schick, M. J., Ed.; Marcel Dekker: New York, 1987. (3) Lyklema, J. Fundamentals of Interface and Colloid Science, Vol. II: Solid-Liquid Interfaces; Academic Press: London, 1995; Chapter 2.75. (4) Tiberg, F. J. Chem. Soc., Faraday Trans. 1996, 92, 531. (5) Jo¨nsson, B.; Lindmann, B.; Holmberg, K.; Kronberg, B. Surfactants and Polymers in Aqueous Solution; John Wiley & Sons: Chichester, 1998. (6) Partyka, S.; Zaini, S.; Lindheimer, M.; Brun, B. Colloids Surf. 1984, 12, 255. (7) Levitz, P.; van Damme, H.; Keravis, D. J. Phys. Chem. 1984, 88, 2228. (8) Gellan, A.; Rochester, C. H. J. Chem. Soc., Faraday Trans. 1 1985, 81, 2235. (9) Bo¨hmer, M. R.; Koopal, L. K.; Janssen, R.; Lee, E. M.; Thomas, R. K.; Rennie, A. R. Langmuir 1992, 8, 2228. (10) Tiberg, F.; Jo¨nsson, B.; Tang, J.; Lindman, B. Langmuir 1994, 10, 2294. (11) Kira´ly, Z.; Bo¨rner, R. H. K.; Findenegg, G. H. Langmuir 1997, 13, 3308. (12) Corkill, J. M.; Goodman, J. F.; Tate, J. R. Trans. Faraday Soc. 1966, 62, 750. (13) Kronberg, B. J. Colloid Interface Sci. 1983, 96, 55. (14) Gellan, A.; Rochester, C. H. J. Chem. Soc., Faraday Trans. 1 1985, 81, 1503. (15) Douillard, J. M.; Pougnet, S.; Faucompre, B.; Partyka, S. J. Colloid Interface Sci. 1992, 154, 113.

Silica is the most studied hydrophilic adsorbent. At low surfactant concentrations, CnEm adsorption is driven by attraction between the headgroups and the surface and the adsorbed amount remains rather low. This implies that the interactions are weak. For bulk concentrations close to the critical micelle concentration (CMC), the adsorption increases strongly toward a plateau value.9,10 This increase in adsorption is due to hydrophobic attraction between the hydrocarbon moieties of the adsorbed surfactant molecules. For small headgroups this increase is stepwise, while it is more gradual for surfactants with longer headgroups. The concentration at which the adsorption strongly increases indicates the onset of aggregate formation at the surface. The amount adsorbed in the plateau and the structure of the adsorbed layer both depend on the relative sizes of headgroup and tailgroup. This can be understood using the critical packing parameter concept introduced by Israelachvili.18,19 As a rule of thumb, the plateau adsorption increases with decreasing size of the headgroup, and increasing length of the alkyl chain.6,8,10 For the composition of the adsorbed layer, the trend is that parallel to the surface extended cylindrical structures are formed by surfactants with a short headgroup, whereas smaller surface aggregates are formed when the headgroup is larger.4,9,10 These aggregate shapes resemble those formed in solution. Since the attraction between headgroups and the surface is weak, the whole process may be viewed as a surface-induced self-assembly. Carbon black surfaces are studied among the hydrophobic adsorbents. Measured isotherms mostly show a Langmuir-type shape, although it is clear that the (16) Kronberg, B.; Stenius, P.; Igeborn, G. J. Colloid Interface Sci. 1984, 102, 418. (17) Kumar, N.; Tilton, R. D. Langmuir 2004, 20, 4452. (18) Israelachvili, J. N.; Mitchell, D. J.; Ninham, B. W. J. Chem. Soc., Faraday Trans. 2 1976, 72, 1525. (19) Israelachvili, J. N. Intermolecular and Surface Forces, 2nd ed.; Academic Press: San Diego, 1991.

10.1021/la051102b CCC: $30.25 © 2005 American Chemical Society Published on Web 07/08/2005

Adsorption of Nonionic Surfactants on Cellulose

conditions for ideal Langmuir behavior are not met.13 It has been suggested that surfactant molecules lie flat on the surface at low concentrations whereas at higher concentrations surface aggregates are formed,14,15 in which the alkyl chains are oriented toward the surface and headgroups toward the solution. The influence of the composition of the surfactants on the adsorbed amount is equal to that for hydrophilic surfaces, i.e. the adsorption increases with decreasing size of the headgroup and increasing size of the tail.12,14-16 Calculations based on a self-consistent field lattice theory for surfactant adsorption show that only at relatively high concentrations are the isotherms Langmuir-like.20 Cooperative adsorption already takes place at very low bulk concentrations. This might be the reason that in experimental studies this phenomenon has been largely overlooked. In addition to knowing the adsorbed amount at equilibrium, information about the kinetics is very useful to obtain further insight into the aggregation behavior of surfactants at solid interfaces. Klimenko et al.21 focused on the adsorption of a poly(ethylene oxide) alkylphenolether, C8φE10, and C12E23 onto a silica gel and did not find any effect of micelles on the kinetics. More recent work on the adsorption kinetics of surfactants22-24 and diblock copolymers25-27 showed that monomers and micelles might both contribute to the initial rate. The micellar contribution was interpreted either as direct adsorption or as suppliers of monomers by breaking up near the surface. Tiberg et al.4,22,28 carried out an extensive study on the adsorption kinetics of surfactants at bare and hydrophobic silica. These authors identified five regimes in their adsorption-desorption curves, each with its own time characteristic. Initially both the adsorption and the desorption rate are limited by diffusion of monomers and micelles. At intermediate adsorption values the rate of adsorption decreases because the driving force goes down: there are fewer open sites on the surface and, as equilibrium is approached, the concentration gradient over the stagnant layer also decreases.22 Recently Avena and Koopal29 have proposed a model for the adsorption kinetics of charged species on a charged surface. They showed that, in the case of electrostatic attraction between adsorbate and surface, the rate of adsorption is likely determined by transport of molecules to the surface, whereas in the case of electrostatic repulsion adsorption is likely governed by the attachment/detachment of monomers from the surface. The present paper focuses on nonionic surfactant adsorption on a flat cellulose surface. Despite its practical relevance e.g. in detergency and papermaking, no systematic study of this type of system could be found in the literature. A reason for this lack might be that the preparation of well-defined smooth flat cellulose surfaces that are suited for a systematic study has been developed only recently (see e.g. refs 30-35). Torn32 describes a (20) Bo¨hmer, M. R.; Koopal, L. K. Langmuir 1990, 6, 1478. (21) Klimenko, N. A.; Permilovskaya, A.; Tryasokurova, A. A.; Koganovski, A. M. Kolloidn. Zh. 1975, 37, 972. (22) Tiberg, F.;. Jo¨nsson, B.; Lindman, B. Langmuir 1994, 10, 3714. (23) Pagac, E. S.; Prieve, D. C.; Tilton, R. D. Langmuir 1998, 14, 2333. (24) Atkin, R.; Craig, V. S. J.; Biggs, S. Langmuir 2000, 16, 9374. (25) Tassin, J. F.; Siemens, R. L.; Tang, W. T.; Hadziioannou, G.; Swalen, J. D.; Smith, B. A. J. Phys. Chem. 1989, 93, 2106. (26) Munch, M. R.; Gast, A. P. Macromolecules 1990, 23, 2313. (27) Bijsterbosch, H. D.; Cohen Stuart, M. A.; Fleer, G. J. Macromolecules 1998, 31, 9281. (28) Brinck, J.; Jo¨nsson, B.; Tiberg, F. Langmuir 1998, 14, 1058. (29) Koopal, L. K.; Avena, M. J. Colloids Surf. A 2001, 192, 93. (30) Fa¨lt, S.; Wagberg, L.; Vesterlind, E. L.; Larsson, P. T. Cellulose 2004, 11, 151.

Langmuir, Vol. 21, No. 17, 2005 7769 Table 1. Critical Micelle Concentration (CMC),36 Refractive Index (n),36 and Refractive Index Increment (dn/dc)37 of the Nonionic Surfactants surfactant

CMC/mol‚m-3

n/-

(dn/dc)/dm3‚kg-1

C12E5 C12E7 C14E7

6.5 × 10-2 8.0 × 10-2 1.0 × 10-2

1.443 1.446 1.447

0.131 0.138 0.139

method that has been developed to coat wafers with a relatively thick cellulose film. The availability of such surfaces allows us to study both the kinetics and the equilibrium adsorption by stagnation point flow reflectometry. In order to study the effects of the size of the headgroup and tailgroup on both kinetic and equilibrium aspects, we investigated surfactants of different compositions. 2. Experimental Section 2.1. Materials. Homodisperse poly(ethylene oxide) alkylethers, C12E5, C12E7, and C14E7, were purchased from NIKKO Chemicals (Japan) and used as received. The CMCs and optical properties of these surfactants are shown in Table 1. HCl and NaCl were of analytical grade. Tap water was purified by being passed through two mixed-bed ion exchangers, a carbon column, and a microfilter (resistivity18 MΩ‚cm). Silicon wafers with a native oxide layer of 2-3 nm were purchased from Aurel (Germany). They were rinsed with water and ethanol, dried, and cleaned in a plasma cleaner (Harrick Scientific Corp., model PDC32G) for 45 s just before use. The preparation method of cellulose layers on the wafers is similar to that of Wegner et al.33,34 and Geffroy et al.,35 and it is described in detail in ref 31. The cellulose layer is firmly anchored to the wafer surface by using an interlayer of a block copolymer, polystyrene-b-poly(4-vinylpyridine) or PS-b-P4VP, obtained from Polymer Scource, Inc. (Canada). The PS and PVP parts had a number average molar mass of 21 400 and 20 700 g/mol, respectively. The cleaned wafers were exposed to a 100 mg‚dm-3 PS-b-P4VP solution in chloroform for 30 min, rinsed with chloroform, and dried with a stream of nitrogen gas. The smooth PS-b-P4VP layer is adsorbed with the PVP segments to the silica, and the PS blocks act as an anchor for the cellulose. The starting material for the cellulose layer was microcrystalline cellulose from Sigma (Sigmacell type 20). Trimethylsilyl cellulose (TMSC) is obtained from Sigmacell using the procedure of Stein.33 A solution of TMSC dissolved in chloroform (20 g‚dm-3) is spincoated (20 s at 2500 rpm) on the wafers coated with PS-b-P4VP. Regeneration of the TMSC film in cellulose is achieved by exposing the film to a gaseous atmosphere of 10% HCl solution for 15 min. The dried cellulose layer has a thickness of about 100 nm. When the layer is exposed to 0.01 mol‚dm-3 NaCl, the film swells (water content 39% v/v) to a thickness of 120 nm with a root-meansquare roughness of 2.5 nm. These thicknesses yield a good sensitivity for the output signal of the reflectometry. The contact angle of the cellulose layer against water is 25°. 2.2. Methods. Surfactant adsorption is studied by stagnation point flow reflectometry. The reflectivity of a flat surface changes due to adsorption. For details about this setup we refer to Dijt et al.38,39 Incoming polarized light from a He-Ne laser is reflected (31) Kontturi, E.; Thune, P. C.; Niemandsverdriet, J. W. Langmuir 2003, 19, 5735. (32) Torn, L. H. Polymers and surfactants in solution and at interfaces. Ph.D. Thesis, Wageningen University, Wageningen, The Netherlands, 2000; Chapter 5. (33) Schaub, M.; Wenz, G.; Wegner, G.; Stein, A.; Klemm, D. Adv. Mater. 1993, 5, 919. (34) Buchholz, V.; Wegner, G.; Stemme, S.; O ¨ dberg, L. Adv. Mater. 1996, 8, 399. (35) Geffroy, C.; Labeau, M. P.; Wong, K.; Cabane, B.; Cohen Stuart, M. A. Colloids Surf. A. 2000, 172, 47. (36) Van Os, N. M.; Haak, J. R.; Rupert, L. A. M. Physico-chemical Properties of Selected Anionic, Cationic and Non-ionic Surfactants; Elsevier: Amsterdam, 1993. (37) Chiu, Y. C.; Chen, L. J. Colloid Surf. 1989, 41, 239. (38) Dijt, J. C.; Cohen Stuart, M. A.; Hofman, J. E.; Fleer, G. J. Colloids Surf. 1990, 51, 141.

7770

Langmuir, Vol. 21, No. 17, 2005

Torn et al.

Table 2. Thickness (d) and Refractive Index (n) of the Layers on the Silicon Wafer layer silicon silica PS-b-P4VP cellulose (dry) cellulose (wet: 39% water) water

d/nm

n

2 4 100 120

3.85 1.45 1.45 1.53 1.45 1.333

by a surface at the Brewster angle in a hydrodynamically welldefined position (stagnation point) of the incoming fluid. The reflected beam is split into its parallel (p) and perpendicular (s) components, and the ratio S ) Ip/Is of the respective intensities is continuously recorded. Adsorption results in a change ∆S of the output signal. Under appropriate conditions, the relation between ∆S and the adsorbed amount Γ is to a very good approximation linear:39

Γ)

∆S 1 S0 MAs

[mol‚m-2]

(1)

where S0 is the initial ratio prior to adsorption, M is the molar mass of the surfactant, and As is a sensitivity factor determined by the optical properties of the surface layers. The parameter As can be calculated by an optical model in which every layer i is assumed to be homogeneous and characterized by a thickness di and refractive index ni following the method of Hansen40 which is based on the exact matrix method of Abeles. In our case a five-layer model (silicon, PS-b-P4VP, cellulose, nonionic surfactant, and water) is used to represent the system. Actual values of thicknesses di and refractive indices ni were measured by ellipsometry or taken from the literature. They are summarized in Table 2. It should be noted that reflectometry only measures the adsorbed amount and not the layer thickness. The implications are that conformational transitions in the adsorbate layer, that take place at given Γ, are not observable, whereas changes in Γ resulting from such transitions are visible. All experiments are carried out at a flow rate of 0.001 dm-3‚min-1, pH ) 5.0, an electrolyte concentration of 0.01 mol‚dm-3 NaCl, and a temperature of 22 °C. Under these conditions the cellulose surface charge is slightly negative.32,35

3. Results and Discussion 3.1. Adsorption Isotherms. 3.1.1. General Features. The surfactant adsorption isotherms after an equilibration time of twelve minutes are shown in Figure 1. At this time the adsorption rate has become negligibly small. The isotherms have usual shapes, and adsorption is reversible. To illustrate clearly the different aspects of the isotherms, three representations of the concentration scale are used. In Figure 1a the concentration is in mol‚dm-3, and in Figure 1b and Figure 1c the concentration axis is scaled to the CMC. Figure 1a shows the specific differences between the surfactants most clearly. The lower the CMC, the lower the concentration at which the adsorption starts. Figure 1b shows that the surfactants C12E5 and C14E7 reach about the same saturation adsorption, whereas that for C12E7 is slightly lower. Figure 1b and Figure 1c show that the differences in affinity for the surface largely disappear if the concentration is scaled with respect to the CMC. The semilogarithmic plots show three distinct regions. At low concentrations ( CMC. It is recalled that at bulk concentrations above the CMC the initial adsorption rates continued to increase for C12E7 whereas they remain constant for C12E5 and C14E7 (Figure 5). In solution, micelles and monomers are in dynamic equilibrium. The micelle concentration increases with surfactant concentration, while the monomer concentration is roughly constant. Micelles can contribute to the adsorption kinetics in two (50) Aniansson, E. A. G.; Wall, S. N.; Almgren, M.; Hoffman, H.; Kielmann, I.; Ulbright, W.; Zana, R.; Lang, J.; Tondre, C. J. Phys. Chem. 1976, 80, 905. (51) Creutz, S.; van Stam, J.; de Schrijver, F. C.; Jerome, R. Macromolecules 1998, 31, 681.

Adsorption of Nonionic Surfactants on Cellulose

ways: (1) by diffusing through the stagnant layer and subsequent adsorption, or (2) by diffusing through this layer till cb < CMC, after which they dissociate thereby acting as a source of monomers, causing an additional flux of surfactant toward the surface. Bijsterbosch et al.27 solved the corresponding transport equations for relatively slow and rapid dissociation of the micelles, showing that the contribution of micelles to the adsorption kinetics depends on their dissociation rate. If this dissociation is relatively slow, adsorption kinetics is simply determined by the adsorption rate of monomers. If it is rapid, a gradual increase of the adsorption rate as a function of the surfactant concentration is observed above the CMC. Since the affinity of the headgroups for the surface is low, it is not likely that micelles directly adsorb onto the surface. By excluding direct adsorption, three processes remain that may contribute to the adsorption rate: monomer and micellar diffusion, and micellar dissociation. The magnitudes of the monomer diffusion coefficients will be fairly similar for the three surfactants. Therefore, in order to relate the observations of Figure 5 above the CMC to the composition of the surfactants, the magnitudes of the micellar diffusion coefficients and micellar dissociation rates should be considered. The micellar size and shape18,19 determine the micellar diffusion coefficients. It is known that C12E7 forms spherical micelles, whereas C12E5 micelles have rather a prolate shape.41,42 Based on geometric packing constraints,18 the size and shape of C14E7 will most probably be in between those of C12E7 and C12E5. For particles of similar shape, the Stokes-Einstein equation states that the hydrodynamic radius is inversely proportional to the diffusion coefficient. If this is the case, for the relative magnitude of the micellar diffusion coefficients, the > DCmic > DCmic . The following order should hold: DCmic 12E7 14E7 12E5 dissociation rate constant of micelles, k-, is mainly determined by the length of the aliphatic chain52 and increases with decreasing hydrophobicity.50,53 According to the above-mentioned model of Aniansson k- should be related to the CMC, and we expect the following order: kC-12E7 > kC-12E5 > kC-14E7 (compare also with kd in Table 3). (52) Clint, J. H. Surfactant Aggregation; Blackie & Son: Glasgow, 1992. (53) Kato, S.; Nomura, H.; Honda, H.; Zielinski, R.; Ikeda, S. J. Phys. Chem. 1988, 92, 2305.

Langmuir, Vol. 21, No. 17, 2005 7775

Summing up, it follows that for the most hydrophilic surfactant, C12E7, the micellar diffusion coefficient and certainly the micellar dissociation rate constant are relatively large. In this case, micelles play a crucial role in the adsorption kinetics and the initial adsorption rate increases with increasing surfactant concentration, also above the CMC. The low micellar dissociation rate constant of C14E7, the most hydrophobic surfactant, is the likely reason for the observation that C14E7 micelles do not contribute to the initial adsorption rate. Surfactant C12E5 has a moderate dissociation rate, but a relatively low micellar diffusion coefficient. Apparently, neither do in this case micelles significantly contribute to the adsorption rate. 4. Conclusions Nonionic surfactants readily adsorb onto cellulose, thereby showing three distinct regions, which are most visible if their concentration is plotted on a logarithmic scale. The adsorption of the nonionic surfactants on cellulose shows features that resemble surfactant adsorption on hydrophilic surfaces, but both the aliphatic and the ethylene oxide segments contribute to the adsorption affinity. Most likely, at low surfactant concentrations the molecules are somewhat embedded in the swollen cellulose surface. At increased concentrations, lateral attraction between surfactant molecules becomes prominent. Above the CMC, the adsorption does not further increase. The adsorption level at the CMC is higher than that on silica surfaces. The adsorption and desorption kinetics sensitively depend on surfactant composition and suggest that micellar surface aggregates are formed that resemble the micellar aggregates in solution. Below the CMC, the initial adsorption rate is attachment-controlled. The desorption kinetics are governed by the dissociation rate of surface aggregates. Above the CMC, the magnitudes of the micellar diffusion and micellar dissociation determine whether micelles play a role or not. Micelles contribute as monomer suppliers to the adsorption kinetics if micellar diffusion and dissociation are sufficiently large. This is the case for C12E7, the most hydrophilic surfactant. LA051102B