Langmuir 2006, 22, 6189-6194
6189
Molecular Dynamics Simulation Study of the Structure of Poly(ethylene oxide) Brushes on Nonpolar Surfaces in Aqueous Solution Dmitry Bedrov* and Grant D. Smith Department of Material Science & Engineering and Department of Chemical Engineering, UniVersity of Utah, 122 S. Central Campus Dr., Room 304, Salt Lake City, Utah 84112 ReceiVed February 24, 2006. In Final Form: May 1, 2006 The structure of poly(ethylene oxide) (PEO, Mw ) 526) brushes of various grafting density (σ) on nonpolar graphite and hydrophobic (oily) surfaces in aqueous solution has been studied using atomistic molecular dynamics simulations. Additionally, the influence of PEO-surface interactions on the brush structure was investigated by systematically reducing the strength of the (dispersion) attraction between PEO and the surfaces. PEO chains were found to adsorb strongly to the graphite surface due primarily to the relative strength of dispersion interactions between PEO and the atomically dense graphite compared to those between water and graphite. For the oily surface, PEO-surface and water-surface dispersion interactions are much weaker, greatly reducing the energetic driving force for PEO adsorption. This reduction is mediated to some extent by a hydrophobic driving force for PEO adsorption on the oily surface. Reduction in the strength of PEO-surface attraction results in reduced adsorption of PEO for both surfaces, with the effect being much greater for the graphite surface where the strong PEO-surface dispersion interactions dominate. At high grafting density (σ ≈ 1/Rg2), the PEO density profiles exhibited classical brush behavior and were largely independent of the strength of the PEO-surface interaction. With decreasing grafting density (σ < 1/Rg2), coverage of the surface by PEO requires an increasingly large fraction of PEO segments resulting in a strong dependence of the PEO density profile on the nature of the PEO-surface interaction.
I. Introduction Poly(ethylene oxide) (PEO) is used extensively to alter the interfacial properties of surfaces in aqueous solutions in applications ranging from controlling particle aggregation in solutions1 to improving biocompatibility by preventing protein and microbial adsorption on organic and inorganic surfaces.2 It has been observed that the ability of PEO (adsorbed or chemically grafted to a substrate) to exert a repulsive force on another surface, particle, or biomolecule depends on surface coverage and the nature of the interactions between PEO and the substrate. It has been suggested that at relatively low surface coverages, corresponding to σ < 1/Rg2, where σ is the grafting density (polymer chains/unit area) and Rg is the root-mean-square radius of gyration of the chain, conventional ideal brush theories are no longer valid and the interaction of polymer chains with the surface determines the brush structure and hence defines the repulsive property of the brush.3 Particularly puzzling is the interaction of PEO chains with nonpolar surfaces in aqueous solution. Systematic study of the interaction between PEO brushes and self-assembled alkanethiol monolayers has revealed an increased adhesion of PEO chains to the self-assembled monolayers with increasing fraction of terminal hydrophobic methyl groups, indicating attraction between PEO chains and the nonpolar interface,4 whereas other studies reported repulsive interactions * To whom correspondence should be addressed. (1) Harris, J. M. In Poly(Ethylene Glycol) Chemistry: Biotechnical and Biomedical Applications; Harris, J. M., Ed.; Plenum Press: New York, 1992. Andrade, J. D.; Hlady, V.; Jeon, S. I. Polym. Mater.: Sci. Eng. 1993, 69, 60. Abo-El-Enein, S. A.; Hanafi, S.; El-Hosiny, F. I.; El-Mosallamy, El-Said H. M.; Amin, M. S. Adsorption Sci. Technol. 2005, 23, 245. (2) Efremova, N. V.; Sheth, S. R.; Leckband, D. E. Langmuir 2001, 17, 7628. McPherson, T.; Kidane, A.; Szleifer, I.; Park, K. Langmuir 1998, 14, 176. Roosjen, A.; van der Mei, H. C.; Busscher, H. J.; Norde, W. Langmuir 2004, 20, 10949. (3) McPherson, T.; Kidane, A.; Szleifer, I.; Park, K. Langmuir 1998, 14, 176. (4) Sheth, S. R.; Efremova, N.; Leckband, D. E. J. Phys. Chem. B 2000, 104, 7652.
between PEO brushes and an alkane-modified AFM tip.5 PEO has been also reported to adsorb on polystyrene latexes6 and hydrophobic (methylated) silica,7 and there are also reports indicating that some proteins have a tendency to adsorb on PEO brushes supposedly due to attractive interaction between PEO chains and hydrophobic patches of proteins.8 Clearly, an improved understanding of PEO interaction with nonpolar surfaces in aqueous media is important for a wide variety of PEO brush applications.
II. Simulation Details A. System Description. In this work, we utilize atomistic molecular dynamics (MD) simulations to better understand the structure of PEO brushes on solid nonpolar surfaces in aqueous solution. In all systems, methyl terminated PEO chains (Mw ) 526, 12 repeat units) were grafted to a single surface (described below). A fully atomistic, quantum chemistry based force field for inter- and intramolecular interactions, PEO and PEO-water interactions,9 and the TIP4P water model10 were used. These force fields have been extensively validated in our previous studies of PEO melts and PEO in aqueous solution.11 All systems consisted of 3000 water molecules and 9, 25, or 64 PEO chains (5) Feldman, K.; Ha¨hner, G.; Spencer, N. D.; Harder, P.; Grunze, M. J. Am. Chem. Soc. 1999, 121, 10134. (6) Mears, S.; Cosgrove, T.; Obey, T.; Thompson, L.; Howell, I. Langmuir 1998, 14, 4997. (7) Wind, B.; Killmann, E. Colloid Polym. Sci. 1998, 276, 903. (8) Sheth, S. R.; Leckband, D. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 8399. (9) Smith, G. D.; Borodin, O.; Bedrov, D. J. Comput. Chem. 2002, 23, 1480. (10) Jorgensen, W. L.; Chandrasekhar, J.; Madura, J. D.; Impey, R. W.; Klein, M. L. J. Chem. Phys. 1983, 79, 926. (11) Smith, G. D.; Bedrov, D.; Borodin, O. J. Am. Chem. Soc. 2000, 122, 9548. Smith, G. D.; Bedrov, D.; Borodin, O. Phys. ReV. Lett. 2000, 85, 5583. Smith, G. D.; Bedrov, D. Macromolecules 2002, 35, 5712. Smith, G. D.; Bedrov D. J. Phys. Chem. B 2003, 107, 3095. Borodin, O.; Bedrov, D.; Smith, G. D. Macromolecules 2002, 35, 2410. Borodin, O.; Bedrov, D.; Smith, G. D. J. Phys. Chem. B 2002, 106, 5194. Borodin, O.; Trow, F.; Bedrov, D.; Smith, G. D. J. Phys. Chem. B 2002, 106, 5184.
10.1021/la060535r CCC: $33.50 © 2006 American Chemical Society Published on Web 06/09/2006
6190 Langmuir, Vol. 22, No. 14, 2006
BedroV and Smith
Figure 1. Snapshot of a simulation cell with model PEO brush on a graphite-like surface in aqueous solution. Hydrogen atoms are not shown.
attached to the surface (σ ) 0.35, 0.96, and 2.46 PEO chains/ nm2) in a square array. The chains were attached only to one side of the surface, whereas the other side was exposed to bulk water as illustrated in Figure 1. The Rg of a 12-repeat unit PEO chain in dilute aqueous solution is about 6.4 Å, yielding σ* ) 1/Rg2 ) 2.5/nm2. The graphite surface consisted of three stacked 52 Å × 50 Å graphene sheets as shown in Figure 1. Nonbonded interactions between graphite carbons, labeled Cg, were describe using the Lennard-Jones (LJ) potential (eq 1a) extensively used for simulation of carbon nanoparticles12 and previously employed by us in studies of graphene sheets in water.13 Parameters for intramolecular (bond, bend, torsion, and out-of-plane bending) and LJ interactions for graphite are given in Table 1. Simulations with these parameters provide spacing between graphene sheets of 3.4 Å which is in a good agreement with experiment.14 Nonbonded interactions of graphite atoms with water oxygens (Ow) and PEO atoms (C,H,O) were described using the LennardJones (LJ) (eq 1a) and Buckingham (eq 1b) potentials, respectively.
[( ) ( ) ]
UCg-Ow(r) ) 4Cg-Ow
σCg-Ow r
-
UCg-(C,H,O)(r) )
[
σCg-Ow
12
4 ACg-(C,H,O) exp(-BCg-(C,H,O)r) - f
6
(1a)
r
(
)]
σCg-(C,H,O) r
6
(1b)
The first distance dependent term in both potentials describes repulsive (steric) interaction, whereas the second term represents attractive, dispersion interaction. For graphite-water interactions, we used the LJ potential parametrized by Werder et al.15 to reproduce the measured macroscopic contact angle of 86° for a water droplet on a graphite sheet.16 This potential has been used by us in studies of C60 fullerenes, single-walled carbon nanotubes, and graphene sheets in water.13 For graphite-PEO interactions, a geometric combining rule for all parameters of the Buckingham potential has been used. To do this, we first represented the LJ form of the graphite-graphite interaction with a Buckingham (12) Cheng, A.; Klein, M. L. Phys. ReV. Lett. 1993, 71, 1200. Hasegawa, M.; Ohno, K. J. Chem. Phys. 1999, 111, 5955. (13) Li, L.; Bedrov, D.; Smith, G. D. J. Phys. Chem. B in press. (14) Nicklow, R.; Wakabayashi, N.; Smith, H. G. Phys. ReV. B 1972, 5, 4951. (15) Werder, T.; Walther, J. H.; Jaffe, R. L.; Halicioglu, T.; Koumoutsakos, P. J. Phys. Chem. B 2003, 107, 1345. (16) Fowkes, F. M.; Harkins, W. D. J. Am. Chem. Soc. 1940, 2, 3377.
form of the potential. Interaction parameters for water-graphite and PEO-graphite are summarized in Table 2 and are also shown in Figure 2. We also studied PEO chains grafted on a representative hydrophobic, or “oily”, surface. For this surface, the structure and interactions within the surface were kept the same as for the graphite surface, whereas surface interactions with PEO and water have been altered. Specifically, in eqs 1a and 1b was reduced by a factor of 3.7 relative to the value used for graphite-water and graphite-PEO interactions, yielding the parameters for the water-oily surface and PEO-oily surface given in Table 2 and pair potentials illustrated in Figure 2. This reduction in results in a dramatic decrease in the strength of the water-surface and PEO-surface interactions. The water-oily surface potential was parametrized so that a C60 fullerene interacting with water using this potential has the same interaction as a droplet of C13H28 of the same diameter.17 Reduction of by a factor of 3.7 reflects the significantly lower density of dispersion sites (carbon atoms) in the alkane compared to the fullerene. We also investigated the role of PEO-surface interactions on the structure of the grafted PEO chains by reducing the strength of the PEO-surface dispersion interactions for both surfaces. In one case, for PEO-surface interactions was reduced by a factor of 1.9, resulting in “graphite-reduced” and “oily reduced” systems. Because PEO-surface dispersion interactions are stronger than water-surface dispersion interactions on a unit volume (of water or PEO) basis, this reduction results in PEO-surface dispersion interactions that are equivalent to the weaker water-surface interactions. Additionally, we investigated systems where PEO has only soft repulsive interactions with the graphite surface by setting f ) 0 in eq 1b, yielding the “graphite-repulsive” system. Parameters for the reduced interactions are also summarized in Table 2, whereas the distance dependence of these interactions is shown in Figure 2. Note that only PEO-surface interactions have been altered for these systems. B. Simulation Methodology. MD simulations were carried out using the simulation package Lucretius described elsewhere.18 The SHAKE algorithm19 was employed to constrain the bond lengths for water and PEO, whereas bonds between graphene carbons were unconstrained. All nonbonded interactions were truncated at 10 Å, and the particle mesh Ewald algorithm20 was used to treat the long-range water-water, water-PEO, and PEOPEO Coulomb interactions (graphene carbons had no partial charges). All simulations were performed in periodic orthorhombic cells. After equilibration at atmospheric pressure and 298 K using an NPT ensemble, sampling was carried out in the NVT ensemble with a trajectory length of 10 ns employing a multiple time step reversible reference system propagator algorithm.21,22 The long range corrections to the pressure, forces and potential energy due to anisotropic density profiles normal to the graphite surface were applied.
III. Results and Discussion A. Water Density Profiles. In Figure 3, the density profiles of water in the z′ direction (see Figure 1) near the bare graphite surface and bare oily surface are shown. The dense arrangement (17) Li, L.; Bedrov, D.; Smith, G. D. Phys. ReV. E 2005, 71, 011502 (1-4). (18) Lucretius: http://www.che.utah.edu/∼gdsmith/mdcode/main.html. (19) Ryckaert, J. P.; Ciccotti, G.; Berendsen, H. J. C. J. Comput. Phys. 1977, 23, 327. (20) Essmann, U.; Perera, L.; Berkowitz, M. L.; Darden, T.; Lee, H.; Pedersen, L. G. J. Chem. Phys. 1995, 103, 8577. (21) Tuckerman, M.; Berne, B. J.; Martyna, G. J. J. Chem. Phys. 1992, 97, 1990. (22) Tuckerman, M.; Berne, B. J.; Martyna, G. J. J. Chem. Phys. 1991, 94, 6811.
Structure of Poly(ethylene oxide) Brushes
Langmuir, Vol. 22, No. 14, 2006 6191 Table 1. Description of Graphite Force Field
nonbonded interaction
UCg(r) ) 4Cg Cg (kcal/mol) 0.066
Cg-Cg bonds
Kbond (kcal/mol/Å2) 618
Cg-Cg bends Cg-Cg-Cg torsions Cg-Cg-Cg-Cg
Cg-Cg-Cg-Cg
Kbend (kcal/mol/rad2) 100.0
[( ) ( ) ] σCg r
12
-
σCg
6
r
σCg (Å) 3.469
Ubond(rij) ) 0.5Kbond(rij - rij0)2 (rij is the distance between bonded atoms i and j)
rij0 (Å) 1.42
Ubend(φijk) ) 0.5Kbend(φijk - φijk0)2 (φijk is the bend angle formed by i-j-k atoms) Utort(θijkl) ) 0.5∑n -Kdihedral(n)cos(nθijkl) (θijkl is the dihedral angle formed by i-j-k-l atoms)
Kdihedral(1) (kcal/mol) 0.00
Uout-of-plane() ) 0.5Kout-of-planeγijk*l2 (γijk*l is the angle between the j-l bond and the i-j-k plane) Kout-of-plane (kcal/mol/rad2) 80.0
φijk0 (deg) 120.0 Kdihedral(2) (kcal/mol) 24.0
Table 2. Description of Water-Surface and PEO-Surface Interactions system graphite
water-surface (eq 1a) Cg-Ow ) 0.0936 kcal/mol σCg-Ow ) 3.19 Å
PEO-surface interaction (eq 1b) Graphite Surface ACg-C ) 29864.9 BCg-C ) 3.333 (Å-1) ACg-O ) 67208.7 BCg-O ) 3.822 (Å-1) ACg-H ) 12561.8, BCg-H ) 3.667 (Å-1)
σCg-C ) 2.825 Å σCg-O ) 2.715 Å σCg-H ) 2.172 Å
graphite-reduced graphite-repulsive oily
Cg-Ow ) 0.0231 kcal/mol σCg-Ow ) 3.241 Å
Oily Surface ACg-C ) 29864.9 BCg-C ) 3.333 (Å-1) ACg-O ) 67208.7, BCg-O ) 3.822 (Å-1) ACg-H ) 12561.8, BCg-H ) 3.667 (Å-1)
oily reduced
of carbon atoms in graphite results in very strong dispersion attraction with water leading to densification (relative to bulk water) of water in the first hydration layer. Although graphite surface is nonpolar the extent of water densification in the first hydration layer is comparable to that observed near polar hydrophilic surfaces.23-25 As we discussed in our recent study,13 although water molecules in the first hydration shell of graphite have a noticeable reduction in the number of water-water hydrogen bonds compared to bulk water, the gain in dispersion attraction with graphite carbon atoms is more favorable resulting in densification of water near graphite. In this regard, the nonpolar graphite surface cannot be considered as a classical hydrophobic surface where a reduction of water density relative to the bulk is expected near the surface.26 In contrast, the water density profile for the oily surface exhibits behavior expected for conventional hydrophobic surfaces. B. Brush Structure on the Graphite Surface. In Figure 4, we show representative snapshots of PEO grafted to graphite for the three grafting densities studied, whereas in Figure 5a, the corresponding density profiles of PEO along the z direction (perpendicular to the surface) are shown. The brush heights, (23) Uchida H.; Takiyama, H.; Matsuoka, M. Cryst. Growth Des. 2003, 3, 209. (24) Cicero, G.; Grossman, J. C.; Catellani, A.; Galli, G. J. Am. Chem. Soc. 2005, 127, 6830. (25) Grigera, J. R.; Kalko, S. G.; Fischbarg, J. Langmuir 1996, 12, 154. (26) Doshi, D. A.; Watkins, E. B.; Israelachvili, J. N.; Majewski, J. Proc. Natl. Acad. Sci. 2005, 102, 9458.
σCg-C ) 2.825 Å σCg-O ) 2.715 Å σCg-H ) 2.172 Å
) 0.25 kcal/mol
f)1
) 0.134 kcal/mol ) 0.25 kcal/mol
f)1 f)0
) 0.0676 kcal/mol
f)1
) 0.0363 kcal/mol
f)1
given as twice the first moment of the PEO density profile, are indicated with arrows in Figure 5a and are summarized for all systems in Table 3. We find that at the highest grafting density (σ ) 2.46 chains/nm2) investigated the PEO chains form a wellformed, extended brush of height h ) 29.2 Å which is in perfect agreement with the value predicted by the ideal brush theory using Flory version of the Alexander model for this system.27 However, at the lowest surface coverage (σ ) 0.35 chains/nm2) investigated, significant adsorption of PEO chains on the graphite surface is observed. Despite the hydrophilic nature of PEO (consolubility with water for all molecular weights), the strong dispersion interaction of PEO chains with the graphite surface results in a strong affinity of PEO to the surface. In Figure 5b, the water density profiles as a function of separation from the PEO-modified graphite surface are shown. Despite the strong affinity of PEO for the graphite surface, there are not enough PEO monomers at low grafting density (σ ) 0.35 chains/nm2) to completely cover the surface, and therefore a strong peak in water density profile near the surface (first solvation layer within 0.5 nm from the surface) can be observed. For systems with higher grafting density (σ ) 0.96 and 2.46 chains/nm2), there are enough PEO units to completely cover the available graphite surface resulting in almost no water in the first solvation layer. However, beyond the first solvation layer, PEO chains are well hydrated for all grafting densities. Finally, examination of density (27) Roosjen, A.; van der Mei, H. C.; Busscher, H. J.; Norde, W. Langmuir 2004, 20, 10949.
6192 Langmuir, Vol. 22, No. 14, 2006
Figure 2. Nonbonded pair interaction potentials between surface atoms and (a) water oxygens or (b) PEO carbons for all cases investigated. Parameters for these potentials are reported in Table 2.
Figure 3. Water density profiles near bare graphite and oily surfaces in the z′ direction (see Figure 1) perpendicular to the surface.
profiles for PEO oxygen and carbon atoms in the first solvation layer indicates that the most probable position for carbon atoms is about 0.1 nm closer to the surface than for oxygen atoms, consistent with the relatively stronger dispersion interaction per CH2 group (than per PEO oxygen atom) and a relatively hydrophilic character (ability to form a hydrogen bond with water) of the PEO oxygen atoms. C. PEO-Surface Interactions and Grafting Density. In Figure 6, the influence of the strength of the PEO-surface dispersion interaction on the PEO density profile is shown for the graphite and graphite-repulsive systems for the three grafting densities investigated. For the lowest grafting density (σ ) 0.35 chains/nm2) elimination of the PEO-surface dispersion attraction results in desorption of the PEO chains and an increase in brush height by a factor of more than 3. However, as the grafting
BedroV and Smith
Figure 4. Representative snapshots of PEO brushes on graphite surface in aqueous solution for three different PEO grafting densities. Water molecules and hydrogen atoms are not shown and only one graphene layer is shown.
density of PEO chains increases, the difference in density profiles and brush height between the graphite and graphite-repulsive systems becomes smaller (σ ) 0.96 chains/nm2) and almost completely disappears for the densest brush (σ ) 2.46 chains/ nm2). This behavior can be understood in terms of the maximum fraction of PEO monomers that can be adsorbed, i.e., corresponds to complete coverage of the surface, as seen for the graphite surface for σ ) 0.96 and 2.46 chains/nm2. Here, there are approximately 5 PEO monomers/nm2 within the first solvation layer. For σ ) 2.46 chains/nm2, there are 12 × 2.46 ≈ 30 monomers/nm2 available for adsorption, and therefore, the saturation of the surface corresponds to adsorption of only around 17% of the PEO monomers and 1/2 of these are covalently bonded to the surface. Hence, desorption of PEO due elimination of PEO-surface dispersion interaction (graphite-repulsive) has little effect on the brush profile. For σ ) 0.96 chains/nm2, there are only 12 × 0.96 ≈ 12 monomers/nm2 available; hence, the saturation corresponds to adsorption of about 40% of the PEO monomers. As a consequence, elimination of PEO-surface dispersion interaction has a more dramatic effect on the PEO density profile. Finally, for σ ) 0.35 chains/nm2, only 4 monomers/nm2 are available, insufficient to cover the surface. For the polymer molecular weight that we investigated, the insensitivity of the PEO brush to the surface-polymer interaction occurs for grafting densities σ ≈ σ*. However, as the polymer molecular weight increases, this regime may move to grafting densities less than σ*. In good solvent, Rg ∼ N0.6 (N is number of monomers) and hence σ* ∼ N-1.2. The number of monomers available for adsorption/unit area at σ* scales therefore as Nσ* ∼ N-0.2. D. Comparison of Brush Structure on the Graphite and Oily Surfaces. Taking into account that systems with high grafting
Structure of Poly(ethylene oxide) Brushes
Langmuir, Vol. 22, No. 14, 2006 6193
Figure 6. Average density profiles of PEO in the z direction on graphite (thin lines) and graphite-repulsive (bold lines) surfaces in aqueous solution for three PEO grafting densities.
Figure 5. The average density profiles of (a) PEO and (b) water in the z direction (see Figure 1) on graphite surface in aqueous solution. Arrows in panel (a) indicate the corresponding brush height reported in Table 3. Table 3. Average Brush Height system
h, Å
graphite (σ ) 2.46 chains/nm2) graphite (σ ) 0.96 chains/nm2) graphite (σ ) 0.35 chains/nm2) graphite-repulsive (σ ) 2.46 chains/nm2) graphite-repulsive (σ ) 0.96 chains/nm2) graphite-repulsive (σ ) 0.35 chains/nm2) graphite-reduced (σ ) 0.35 chains/nm2) oily (σ ) 0.35 chains/nm2) oily-reduced (σ ) 0.35 chains/nm2)
29.2 15.8 8.4 33.1 28.8 28.0 21.0 14.9 18.0
density are not sensitive to changes in the polymer-surface interaction, our further discussion of the influence of interactions of PEO with the surface will be limited to the lowest grafting density (σ ) 0.35 chains/nm2). In Figure 7, panels a and b, the density profiles of PEO for the lowest grafting density are shown for the graphite and oily (Figure 7a) and graphite-reduced and oily-reduced (Figure 7b) surfaces. Figure 7a reveals that, although significant adsorption of PEO occurs for the oily surface, it is reduced dramatically from that observed for the graphite surface. The strong adsorption of PEO chains to the graphite surface is due primarily to the relative strength of dispersion interactions between PEO and the atomically dense graphite compared to those between water and graphite. For the oily surface, PEOsurface and water-surface dispersion interactions are much weaker, greatly reducing the energetic driving force for and hence the extent of PEO adsorption. The role of dispersion interaction in the adsorption of PEO on graphite is clearly seen when Figure 7b is compared with Figure 7a. Here, it can be seen that reducing the strength of the PEO-surface dispersion interaction to match the water-surface dispersion interaction results in significant reduction in the extent of PEO adsorption on the graphite surface.
Figure 7. Average density profiles of PEO in the z direction on (a) graphite and oily surfaces and on (b) graphite-reduced and oily reduced surfaces for PEO brush σ ) 0.35 chains/nm2 in aqueous solution. Arrows indicate the corresponding brush height reported in Table 3.
The effect of reducing the strength of the PEO-surface dispersion interaction to match the water-surface dispersion interaction is much less dramatic for the oily surface. Surprisingly, Figure 7b reveals that PEO adsorption for the oily reduced system is actually greater than that seen for the graphite-reduced system, despite the much greater adsorption observed on graphite compared to the oily surface (Figure 7a). This is due to the fact that hydrophobic forces play an important role in the adsorption of PEO on the oily (hydrophobic) surface. Since the hydrophobic driving force favoring replacement of water with PEO on the oily surface is not influenced by the strength of the PEO-surface interaction, it is equally operative for both the oily and oily-reduced surfaces. Finally, examination of Figure 7b reveals that even when PEOsurface and water-surface dispersion interactions are equal the
6194 Langmuir, Vol. 22, No. 14, 2006
PEO chains have some tendency to adsorb (indicated by a moderate first peak in the density profiles near the surface) on both the graphite and oily surfaces. For oily surfaces this is due, at least in part, to hydrophobic forces as discussed above. The reason for weak adsorption seen for the graphite-reduced system is unclear. It may be that there are still some hydrophobic effects present for the graphite surface (despite the preference of water for the graphite interface as illustrated in Figure 3), or perhaps due to weak PEO-PEO attraction observed in our simulations of PEO solutions.28
IV. Conclusions Our MD simulations lead us to conclude that adsorption of PEO should be anticipated for PEO brushes on nonpolar surfaces in aqueous solution. For low grafting densities (σ < σ*) for the
BedroV and Smith
molecular weight examined, adsorption can lead to significant perturbation of the brush structure. The extent of adsorption for low grafting densities increases dramatically with increasing atomic density of the surface. For conventional hydrophobic surfaces (e.g., dense self-assembled alkyl monolayers, polystyrene colloids) we expect moderate adsorption driven by a combination of dispersion and hydrophobic interactions. For atomically dense nonpolar surfaces (e.g., graphite, carbon nanotubes, fullerenes) strong adsorption driven by dispersion interactions is observed. Acknowledgment. The authors acknowledge the support of the National Science Foundation through Collaborative Research in Chemistry Grant CHE-0304807. LA060535R (28) Bedrov, D.; Li, L.; Smith, G. D. Langmuir 2005, 21, 5251.
