9932
Langmuir 2005, 21, 9932-9937
Unsteady Motion of Receding Contact Lines of Surfactant Solutions: The Role of Surfactant Re-Self-Assembly† K. S. Varanasi* and S. Garoff Department of Physics and Center for Complex Fluids Engineering, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213 Received June 13, 2005 Re-self-assembly of surfactant molecules must occur at moving contact lines of soluble surfactant solutions. Molecules are transported into and out of the contact line region from four sources: the three interfaces meeting at the contact line and the fluid confined between the solid-liquid and liquid-vapor interfaces. As molecules move among these sources at the contact line, they must rearrange. The dynamics of this re-self-assembly has been shown to have a dominating effect on the structure of advancing contact lines, causing unsteady motion and complex structure of the contact line. It might be assumed that the reself-assembly for receding contact lines leads to more steady contact line movement. However, in this article we show that for a wide variety of systems this is not true. Quasi-static distortions of the contact line occur as it retreats because of the inability of the surfactant to completely re-self-assemble at localized positions along the contact line.
1. Introduction Self-assembly of surfactant molecules at interfaces has a major impact on the wetting of solid surfaces by their aqueous solutions. Their adsorption to the solid-liquid and liquid-vapor interfaces alters the surface tensions, γSL and γLV, respectively. This contributes to the force balance determining the static contact angle of the solution
γLV cos θ ) γSV - γSL
(1)
where θ is the static contact angle.1 We expect surfactant molecules to be deposited on the solid-vapor interface behind a receding contact line. This sets the final interfacial tension γSV and the receding static contact angle. However, even when the static situation follows the advance of a contact line, surfactant molecules are found on the solid surface ahead of the contact line in a region where the solution has never touched the solid surface. This assembly of molecules has been seen on highenergy2,3 and low-energy surfaces.4 Thus, the self-assembly of surfactant at all three interfaces near a static contact line controls the static contact angle of surfactant solutions. At a receding contact line of a solution of soluble surfactantssthe case reported in this articlesthe transport to and from the contact line region is more complex than for solutions of insoluble surfactants. For insoluble surfactants, the deposition of surfactant onto the emerging solid-vapor interface is only from the liquid-vapor interface. For soluble surfactants, molecules are transported to and from the contact line region not only along the liquid-vapor and solid-vapor interfaces but also from the solid-liquid interface and the solution.5 In all cases, surfactant is transported to the receding contact line region †
Part of the Bob Rowell Festschrift special issue. * Corresponding author. E-mail:
[email protected].
(1) Adamson, A. W.; Gast, A. P. Physical Chemistry of Surfaces; John Wiley & Sons: New York, 1997. (2) Frank, B.; Garoff, S.Langmuir 1995, 11, 4333. (3) Qu, D.; Suter, R.; Garoff, S. Langmuir 2002, 18, 1649. (4) Kumar, N.; Varanasi, K.; Tilton, R.; Garoff, S. Langmuir 2003, 19, 5366. (5) Loukkala, B.; Garoff, S.; Tilton, R.; Suter, R. Langmuir 2001, 17, 5917.
Figure 1. Surfactant movement to and from the contact line region. The direction of surfactant transport on the liquidvapor interface and in the bulk fluid varies from system to system.
along the solid-liquid interface and away from the region along the solid-vapor interface. Transport at the liquidvapor interface and in the bulk may be either into or away from the contact line.6 However, in the contact line region there must be a mass balance of surfactant entering and leaving the region. This transport process is shown schematically in Figure 1. During this transport process, the molecules must reself-assemble as they are deposited onto the emerging solid-vapor surface. Some specific examples indicate how complex this process can be. For solutions of cetyltrimethylammonium bromide (CTAB) at the critical micelle concentration (cmc), admicelles appear at the solid-liquid interface far from the contact line,7,8 yet a hydrophobic monolayer appears on a silica surface withdrawn from solution.5,9,10 Similarly, for the nonionic surfactants poly(6) Loukkala, B. Private communication. (7) Velegol, S. B. The Effect of Counterion and Silimarly-Charged Polyelectrolytes on Interfacial and Bulk Self-Assembly of Ionic Surfactants. Thesis, Carnegie Mellon University, Pittsburgh, PA, 2000. (8) Manne, S.; Cleveland, J. P.; Gaub, H. E.; Stucky, G. D.; Hansma, P. K. Langmuir 1994, 10, 4409.
10.1021/la051576a CCC: $30.25 © 2005 American Chemical Society Published on Web 08/30/2005
Unsteady Motion of Receding Contact Lines
(ethylene glycol) polyethers, CiEj, the self-assembly at the solid-liquid interface is consistent with the adsorption of heterogeneous discrete aggregates greater than a monolayer coverage,8,11,13 but these aggregates do not appear at the solid-vapor interface after withdrawal from solution.5 These cases indicate that the emerging surfactant structure is not simply formed by drawing an unperturbed surfactant layer from the solid-liquid interface through the contact line. Preliminary studies of SiO2 surfaces withdrawn from a 0.4 cmc CTAB solution below film entrainment velocities show that 1.2 mg/m2 CTAB emerges from the contact line onto the solid-vapor surface. The amount of surfactant approaching the contact line from the solid-liquid and liquid-vapor interfaces is 0.4 and 0.9 mg/m2, respectively. In this case, the liquid-vapor and solid-liquid interfaces supply the surfactant deposition onto the emerging solid surface while also rejecting surfactant back into the solution. During re-self-assembly, the surfactant deposition on the solid-vapor interface alters the surface energy and influences the now dynamic force balance at the contact line. Re-self-assembly occurs in some microscopic region of length Lr near the contact line. The re-self-assembly process occurs when the solid-liquid and liquid-vapor interfaces begin to interact as they approach each other near the contact line. For systems similar to those we study on SiO2 in this article, Ducker14 et al. observed surfactant re-self-assembly when two solid surfaces approach at distances of ∼10 nm. This length scale is typical of disjoining pressure effects. The systems with SiO2 as a substrate have a small local contact angle, ∼3°. Thus, a driven re-self-assembly occurs at a distance Lr ≈ 200 nm from the contact line. In such confined geometries, short-range surfactant-surfactant and surfactantsurface interactions become significant, providing the driving mechanism for re-self-assembly. At these interface separations, diffusive transport is not a limiting mechanism as it might be near bulk interfaces where depletion lengths are on the order of 104 nm.15 The time imposed on re-self-assembly near a threephase contact line is Lr/U, where 1 µm/s < U < 100 µm/s is the range of receding velocities used in the experiments reported in this article. Using the above estimate of Lr, the imposed re-self-assembly time ranges from 0.002 to 0.2 s. As will be shown below, successful re-self-assembly at the contact line resulting in a relaxed structure occurs in most instances at these speeds for the systems we examine. However, these imposed time scales are an order of magnitude shorter16,17 than those for self-assembly18-20 onto bulk interfaces for similar systems. Thus, the reself-assembly at the contact line is not driven by the same processes that drive the equilibration of interfaces with bulk solutions. (9) Birch, W. R.; Knewston, M. A.; Garoff, S.; Suter, R. M.; Satija, S. Colloids Surf. 1994, 89, 145. (10) Eskilsson, K.; Yaminsky, V. V. Langmuir 1998, 14, 2444. (11) Rutland, M. W.; Senden, T. J. Langmuir 1993, 9, 412. (12) Tiberg F.; Jonsson, B.; Tang, J.-A.; Lindman, B. Langmuir 1994, 10, 2294. (13) Thirtle, P. N.; Li, X. Z.; Thomas, R. K. Langmuir 1997, 13, 5451. (14) Subramanian, V.; Ducker, W. J. Phys. Chem. B 2001, 105, 1389. (15) Ferri, J. K.; Stebe, K. J. Adv. Colloid Interface Sci 2000, 85, 61. (16) Kumar, N.; Couzis, A.; Maldarelli, C. J. Colloid Interface Sci. 2003, 267, 272. (17) Valkovska, D. S.; Shearman, G. C.; Bain, C. D.; Darton, R. C.; Eastoe, J. Langmuir 2004, 20, 4436. (18) Hsu, C.-T.; Shao, M.-J.; Lin, S.-Y. Langmuir 2000, 16, 3187. (19) Lin, S.-Y.; Tsay, R.-Y.; Lin, L.-W.; Chen, S.-I. Langmuir 1996, 12, 6530. (20) Hutchison, J.; Klenerman, D.; Manning-Benson, S.; Bain, C. Langmuir 1999, 15, 7530.
Langmuir, Vol. 21, No. 22, 2005 9933
Figure 2. Typical distortion on a receding contact line of 0.8cmc tetraethylene glycol monododecyl ether (C12E4) on SiO2 at a receding velocity of 20 µm/s.
The re-self-assembly near advancing contact lines has been shown to have a dramatic impact on the dynamic wetting of soluble surfactant solutions. Some systems exhibit autophobing where surfactant deposited ahead of an advancing contact line drives the contact line into retreat.3 When solutions of ionic surfactants are forced across a surface of opposite charge, the contact line advances by a stick-jump motion.2 When solutions of nonionic surfactants, CiEj, advance across a negatively charged surface, the re-self-assembly at the contact line causes a complex advance of the contact line that is a combination of a stick-jump motion and dendritic spreading. As the interaction of the surfactant with the surface becomes weaker (altering the re-self-assembly), the characteristics of the spreading change systematically from stick-jump to dendritic.21,22 In these cases, the deposition on the solid-vapor interface ahead of the contact line is not uniform across the contact line. This causes the dendritic structure of the contact line. Although one might assume that re-self-assembly at a receding contact line is a more uniform process, producing smooth motion of the contact line, we show in this article that for a wide variety of systems this is not true. The majority of the contact line is smooth, but there are occasional localized micrometer-scale distortions. Spatial distortions of the contact line occur because the re-selfassembly process in localized regions cannot be completed as the contact line moves across the surface. A part of the contact line (∼10 µm laterally) is pinned to the substrate, stretches for about 10 µm, and is then released to rejoin the flat part of the contact line. A typical fluctuation is shown in Figure 2 at its maximum height. 2. Materials and Methods 2.1. Sample Preparation. We used [100]-oriented silicon wafers (International Wafer Service Inc.) with their native oxide layer as our high-energy substrates. The substrates are acid cleaned such that there is a 0° contact angle against water and condensed water vapor forms a uniform film.21,23 All of the experiments were done right after the cleaning process, so there is minimal dust accumulation. AFM images of bare SiO2 wafers were taken under the same conditions as the experiments. These images show a uniform surface with too low a frequency of features of height J1 nm to account for the contact line distortions we discuss here. Low-energy surfaces were silicon wafers coated with a monolayer of octadecyltrichlorosilane (OTS). The preparations used produced a uniform, stable, methylated substrate4 with an advancing static angle of 112 ( 4° and a 95 ( 5° receding (21) Frank, B.; Garoff, S. Langmuir 1995, 11, 87. (22) Frank, B.; Garoff, S. Colloids Surf., A 1996, 116, 31. (23) Iler, R. K. The Chemistry of Silica: Solubility, Polymerization, Colloid and Surface Properties and Biochemistry of Silica; John Wiley & Sons: New York, 1979.
9934
Langmuir, Vol. 21, No. 22, 2005
Varanasi and Garoff
Table 1. Surfactant-Surface Systems Studied Surfactant (cmc)
Surface
CTAB (1 × 10-3 M)2
SiO2 OTS
CTAB + 10 mM KBr (0.12 × 10-3 M)7
SiO2
C12E8 (9.2 × 10-5 M)25
SiO2 OTS
C12E4 (6.9 × 10-5 M)25
SiO2
contact angle with a hysteresis energy of 20 dynes/cm for pure water. These preparations end with a final rinse and sonication of the surface under chloroform and produce complete, uniform OTS monolayers.4,24 The characteristics of the OTS surfaces used in this study are entirely consistent with those previously reported.4 AFM studies of similarly prepared OTS surfaces show that the monolayer is uniform.24 We also performed AFM of our OTS surfaces under the same conditions as in our experiments. We do not see defects in the monolayer. On this monolayer, we do see features (∼1 µm), but the spatial frequency indicates that they are not the cause of contact line distortions. We explored the receding contact line behavior of solutions of cationic surfactant cetyltrimethylammonium bromide (CH3(CH2)15N(CH3)3Br; CTAB); CTAB with 10 mM potassium bromide; and two ethylene oxide surfactants, octaethylene glycol monododecyl ether (CH3(CH2)11(OCH2CH2)8OH; C12E8) and tetraethylene glycol monododecyl ether (CH3(CH2)11(OCH2CH2)4OH; C12E4). All of the surfactants were used as received from Fluka Chemical Corp. (Buchs, Switzerland). CTAB is assayed at >99% purity; KBr, at >99.5%; and the CiEj’s, at >98%. We made aqueous solutions using RO pure water (from a Barnstead Nanopure II water system with posttreatment to remove organic impurities) with a resistivity >17.5 MΩ‚cm. Unless noted otherwise, all of the solution concentrations were 0.8cmc. Table 1 lists the surfactant-surface systems studied. 2.2. Experimental Geometry and Optical Microscopy. The substrates were suspended vertically in a Teflon beaker overfilled with the solution (Figure 3.) The substrates were advanced and receded from the fluid bath at controlled speeds by a motorized stage (Newport Motion Controller, Newport Klinger Corporation, CA). First, the substrates were advanced into the solution at speeds ranging from 500 to 2000 µm/s. They were left soaking in the solution for about an hour or more so that, in all instances, the solid-liquid interface would come to equilibrium with the solution.4,6,7,25,26 They were then withdrawn at speeds ranging from 2 to 30 µm/s. The speed of the motor was calibrated by tracking the motion of features on a plate moved by the stage. The experiments were carried out in ambient laboratory conditions where the temperature ranged from 20 to 25 °C and the relative humidity ranged from less than 10 to 60%. These variations did not have any detectable effect on the results reported here. The receding contact line was imaged using video reflectance microscopy.2,21,27 Figure 4 gives a schematic of the experimental setup. Using a telemicroscope (Bausch and Lomb, model 31-2978), microscopic structures at the contact line were examined at a magnifications of 21×. The sample was illuminated with coaxial lighting. The flat substrate ahead of the receding contact line reflects light into the microscope and appears bright in the image. The curved meniscus reflects light out of the camera aperture and appears dark. The images were recorded on videotape and then sampled either at video frame rates or slower for image analysis. Data on contact line distortions at each withdrawal speed were taken at several places on a single substrate and on multiple substrates. The variation in the frequency of distortions is the same for measurements at several points on a single substrate as for measurements on multiple substrates. (24) Kumar, N.; Maldarelli, C.; Steiner, C.; Couzis, A. Langmuir 2001, 17, 7789. (25) Loukkala, B. Interfacial Structure and Rearrangement of Nonionic Surfactants near a Receding Contact Line. Thesis, Carnegie Mellon University, Pittsburgh, PA, 2001. (26) Velegol, S. B.; Fleming, B. D.; Biggs, S.; Wanless, E. J.; Tilton, R. D. Langmuir 2000, 16, 2548. (27) Nadkarni, G. D.; Garoff, S. Langmuir 1994, 10, 1618.
Figure 3. Sample (seen edge-on) suspended above a Teflon beaker of fluid.
Figure 4. Schematic diagram of optics for reflectance microscopy: S, sample; F, fluid; H, half-silvered mirror; M, microscope; C, camera. Dashed arrows are incident light rays. Solid arrows are reflected light rays. 2.3. Atomic Force Microscopy. To characterize the surface topography of substrates withdrawn from surfactant solutions, atomic force microscopy in contact mode11,28,29 was carried out at 23 ( 3 °C using a Multimode Nanoscope IIIa (Digital Instruments, Inc.) with a 150 µm scanner (serial no. 6219). Pyramidal AFM probes of silicon nitride, featuring cantilevers with spring constants ranging from 0.03 to 0.5 N/m, were purchased from Thermomicroscopes, Sunnyvale, CA (model no. MSCT-AUHN). Multiple scans at rates of 1-5 Hz with a deflection set point between 1 and 1.5 V were made on a surface both in the friction and height mode. If a feature appears repeatedly in the same region in both modes, we assume that it is real and not an artifact of the tip altering the sample. Changing the scanning angle does not deform these features. Some plowing of the surfactant-covered surface by the tip is unavoidable with the high set point. We process the final images to remove noise that might prevent us from recognizing real features.
3. Results and Discussion 3.1. Characteristics of Contact Line Distortions. Figure 5 illustrates a typical formation and collapse of a distortion on a receding contact line. As the distortion begins, a slight deformation forms on an otherwise smooth contact line. As time progresses, this deformation grows both laterally and vertically. When the distortion is about 10 µm in extent laterally and vertically, it drops back to reform a smooth part of the contact line. Figure 6 shows plots of two typical evolutions of the height of the distortion. In Figure 6a, we see that the contact line is pinned to a fixed position on the substrate and the distortion stretches the contact line. When the contact line reaches a maximum height, it drops down but at a relatively slow speed (∼50 µm/s). Even during this drop of the contact line, fluid movement is characterized by very small capillary and Reynolds numbers, 10-6 and
