Microscopic Wetting of Self-Assembled Monolayers with Different

Jun 3, 2011 - Quantum mechanics calculations are employed to optimize each alkyl chain of the different SAMs with one water molecule and to investigat...
0 downloads 7 Views 5MB Size
ARTICLE pubs.acs.org/Langmuir

Microscopic Wetting of Self-Assembled Monolayers with Different Surfaces: A Combined Molecular Dynamics and Quantum Mechanics Study Zhen Xu,† Ke Song,† Shi-Ling Yuan,* and Cheng-Bu Liu* Key Laboratory of Colloid and Interface Chemistry, Shandong University, Jinan 250100, China

bS Supporting Information ABSTRACT:

Molecular dynamics simulations are used to study the micronature of the organization of water molecules on the flat surface of wellordered self-assembled monolayers (SAMs) of 18-carbon alkanethiolate chains bound to a silicon (111) substrate. Six different headgroups (CH3, CdC, OCH3, CN, NH2, COOH) are used to tune the character of the surface from hydrophobic to hydrophilic, while the level of hydration is consistent on all six SAM surfaces. Quantum mechanics calculations are employed to optimize each alkyl chain of the different SAMs with one water molecule and to investigate changes in the configuration of each headgroup under hydration. We report the changes of the structure of the six SAMs with different surfaces in the presence of water, and the area of the wetted surface of each SAM, depending on the terminal group. Our results suggest that a corrugated and hydrophobic surface will be formed if the headgroups of SAM surface are not able to form H-bonds either with water molecules or between adjacent groups. In contrast, the formation of hydrogen bonds not only among polar heads but also between polar heads and water may enhance the SAM surface hydrophilicity and corrugation. We explicitly discuss the micromechanisms for the hydration of three hydrophilic SAM (CN-, NH2- and COOH-terminated) surfaces, which is helpful to superhydrophilic surface design of SAM in biomimetic materials.

1. INTRODUCTION SAM (self-assembled monolayer) techniques have attracted significant attention in the field of surface science and materials because they can be used to tailor surface properties such as wetting, micropatterning, lubrication, corrosion, and biocompatibility which inspire the design of biomimetic materials.16 The wetting characteristics of organic surfaces are fundamentally relevant to a wide and diverse range of biochemical, environmental, and technological phenomena, from molecular recognition and binding processes in biology,4 to the hygroscopicity of aerosol particles in the atmosphere,79 to biosensors in a biological environment.10 It is important to study the effect of the SAM surface wetting in a biological environment on biomaterial adsorption. Recent studies have used SAMs to evaluate the effect of surface charge, wetting ability and topography on protein adsorption, and cell r 2011 American Chemical Society

behavior using in vitro assay systems.1114 On the other hand, many groups have studied different proteins, polymers, and amino adsorption and behavior on different kinds of SAM surfaces using computational methods.1517 Since SAM surfaces with different wetting ability may contribute to biomaterial adsorption in biofluid, it is desirable to understand the wetting ability of conventional biomaterial surfaces at the microlevel in order to design new biomaterials with a better ability to support the attachment, growth, and function of biomaterials. Functionalizing the surface of SAMs with specific headgroups (hydrophilic or hydrophobic) can alter structure (packing and orientation) as well as dynamics of vicinal water molecules. The Received: September 15, 2010 Revised: June 3, 2011 Published: June 03, 2011 8611

dx.doi.org/10.1021/la201328y | Langmuir 2011, 27, 8611–8620

Langmuir behavior of water and the nature of solute water near the surface of SAMs at the molecular level have direct implications on a variety of processes ranging from the binding of proteins to surfaces to the broader colloidal and biological self-assembly in interfacial environments.1821 The wetting property of SAM surface is the main factor affecting the protein adsorption on a biomaterial.22 Although a detailed description of the organization of regions of water near flat SAM surfaces is an important element in the overall understanding of watersurface interactions, practical applications in biophysics require more convincing knowledge of the effects of heterogeneity on surface composition (i.e., different kinds of hydrophobic and hydrophilic groups) and surface roughness. The behavior of water near hydrophilic surfaces of SAMs has also been investigated experimentally2325 and computationally.2629 But water on organic surfaces is particularly not well understood, whether the interaction between water molecules and the SAM surface or the effect of hydration on the internal structure of the SAM surface. However, while extremely useful, these investigations are not able to obtain the kind of atomic-level information that is needed to understand exactly how the polar head influences the wetting ability of SAMs with different surfaces, which further influences the biomaterial adsorption behavior. Because of the substantial increase in computational power over the past few years, computer simulations such as molecular dynamics (MD)3033 have proven to be valuable tools to study the wetting ability of SAMs and can provide a detailed, atomistic level insight into the three-dimensional structure of the studied model system. These kinds of studies allow us to extract information about dynamic and structural properties at a microscopic level which is not easy to get from experiments. In this work, we focus on the microscopic wetting properties of SAMs of hydrophobic (with CH3, OCH3, and CdC terminated groups) and hydrophilic (with CN, NH2, and COOH terminated groups) surfaces. The six end groups are easy to prepare experimentally and are widely used to tailor SAM surfaces in both technological applications and fundamental research. This study presents a combined theoretical approach (both MD and quantum mechanical (QM) methods) aimed at investigating the diversification of the organization of water molecules on flat organic surfaces, specifically, SAMs bound to a silicon (111) substrate. The character of the surface is gradually tuned from hydrophobic to hydrophilic using six different neutral headgroup chemistries (CH3, OCH3, CdC, CN, NH2, COOH). An equivalent water ball was input for the same simulation time for each system to measure the wetting ability of all six SAMs. In our work, we report on the evolution of the structure of the surfaces of the six SAMs, both in the absence and presence of water, and explain the mechanism of the water surface interactions of the six SAMs.

2. SIMULATION DETAILS 2.1. Systems. Molecular dynamics simulations were performed for six SAM systems in the absence of water. The SAM was constructed with 18-carbon-long alkyl chain that was attached harmonically to a silicon (111) substrate on one end and presented a headgroup on the other end. To vary the hydrophobic/hydrophilic character of the SAM surfaces in a systematic way, six different chemistries of the headgroup (CH3, OCH3, CdC, CN, NH2, COOH)—from hydrophobic to hydrophilic—were used. Each SAM consisted of

ARTICLE

Figure 1. Substitution pattern of four alkyl monolayers in a (2  4) simulated cell on a Si (111) surface.

128 alkyl chains, arranged in a well-ordered, defect-free array of 16  8 chains with a surface occupied area of ∼25 Å2 per chain. A rhombic simulation box was built with the dimensions (x = 61.43 Å, y = 61.43 Å, and z = 103.91 Å). Two layers of silicon atoms composed the substrate, consisting of 1024 atoms totally, and the first layer was located at z = 0 Å. The z-dimension thickness of each SAM was ∼17 Å. The best substitution percentage of the Si(111) substrate by the alkyl chains is 50%, based on previous work,34 and the substitution pattern of four alkyl chains in a (2  4) simulated cell on the Si (111) surface is shown in Figure 1. All silicon atoms of the substrate were constrained by fixing their three-dimensional Cartesian positions during the simulation. Molecular dynamics simulations of the six SAMs described above were performed in the presence of water in order to characterize their wetting properties. For this object, a standard water ball, of which the radial distance is 15 Å, consisting of 509 molecules totally, was introduced into the simulation box. The simple point charge (SPC) model35 is adopted for the water molecule. The same water ball was placed on the top of each SAM in the same location, where it nearly contacted the SAM surface. 2.2. MD Simulation. MD simulation of different systems can be performed after charges and potentials are assigned to each atom. The total energy is written as a combination of valence terms including diagonal and off-diagonal cross-coupling terms and nonbond interaction terms, which include the Coulombic and Lennard-Jones functions for electrostatic and van der Waals interactions36 E ¼ Ebonds þ Eangles þ Edihedrals þ Ecross þ EVDW þ Eelec ð1Þ where EVDW37 and Eelec are given by eq 2: Enonbond ¼ EVDW þ Eelec 2 !9 !6 3 σ σ ij ij 5þ ¼ εij 42 3 rij rij



qq

∑ riij j

ð2Þ

The parameters for each like-site interaction are given by the COMPASS force field.38,39 All silicon atoms were constrained during the simulation. 8612

dx.doi.org/10.1021/la201328y |Langmuir 2011, 27, 8611–8620

Langmuir

ARTICLE

Figure 2. Atomic density profiles along the interface normal in the end-group region of hydrophobic (CH3- (a), CdC- (b), and OCH3-terminated (c)) and hydrophilic (CN- (d), NH2- (e), and COOH-terminated (f)) SAMs with different end groups in contact with a water ball.

After reaching equilibration criteria of energy minimization, 1 ns simulations of all six neat SAMs were performed in order to obtain their structural characteristics. Then the above-described water ball was introduced into the simulation box. Finally, six SAM systems with different surfaces but the same water ball were established. After removing potential overlaps between water and SAM atoms by energy minimization, the six systems were propagated for 5 ns, and atomic coordinates were saved every 5000 fs. 2.3. Quantum Chemistry Calculation. In order to further demonstrate the interactions between SAM surface and water molecules at the microlevel, we performed quantum-chemistry calculations. All the calculations were conducted using DFT and the Gaussian03 suite of programs. We choose the Becke 3-parameter LeeYangParr exchange-correlation (B3LYP) functional,40,41 and the standard 6-311þþG(d,p) basis set was used for all the elements. The geometries of every single alkyl chain of the different end groups and each of them with one single water molecule were optimized. Solvent effects of the complexation in aqueous solution are estimated using the COSMO model.42,43 In order to study the hydrogen-bonding patterns between water molecules and polar groups of neighboring chains, we also optimized the possible “bridging” structures with one or two water molecules between two hydrophilic chains. The eight carbon atoms far away to the headgroup of the two hydrophilic chains (together with H atoms) were fixed during the DFT calculation to keep them “parallel” to each other as if they were part of the SAM, and the distance is 3.8 Å (due to the SAM structure).

3. RESULTS AND DISCUSSION 3.1. Influence of Surface Wetting to the Internal Structure of SAM. In order to explore the structure changes of the SAMs

with different terminal groups on the water in their vicinity, it is interesting to examine the effect of the presence of the abovedescribed water ball on each SAM. Figure 2 shows a comparison of the density profiles in the headgroup region of every SAM in the absence of the water ball versus that of each SAM in contact with water molecules. The water molecular density forms clear peaks near the hydrophilic surfaces, which are very obvious in all

three hydrophilic SAMs, occurring at z ≈ 24.5 Å, z ≈ 25.5 Å, and z ≈ 26.5 Å. In contrast, there is no clear peak near the surface of hydrophobic SAMs with CH3-, CdC-, and OCH3-terminated alkyl chains, indicating that the hydrophobic SAMs have too little effect on the internal structure of the water ball to break H-bonds among the water molecules and cause the water molecules to spread on the SAM surface region. Meanwhile, the density profiles in the end group region of the hydrophobic SAMs remain unperturbed by the presence of water (shown in panels ac of Figure 2). There are obvious structure changes in the surface region of hydrophilic SAMs (with CN-, NH2-, and COOH-terminated alkyl chains) in the absence of water versus in contact with the water ball, shown as the density profiles in panels df of Figure 2. Panel d of Figure 2 reveals that the density profile peaks of N (CN) and C18 are shifted by ∼0.5 Å toward a larger z value, indicating more stretched CN end groups of alkyl chains compared to the neat SAM. This can be rationalized in terms of N 3 3 3 H bonds forming between water molecules and nitrogen atoms, resulting in the CN groups stretching because of the above-mentioned weak interactions between H (C17) and N (CN). From panel e of Figure 2, the density profile peak of N (NH2) is shifted by ∼0.5 Å toward a smaller z value, while the C18 peak position is with the same as that in the absence of water. H-bonds formed between H (NH2) and O (H2O) make the configuration of NH2 turn over. In panel f of Figure 2, it is notable that the C18 density profile peak of the COOH-terminated SAM in contact with water molecules is located at z ≈ 22.5 Å, consistent with the one of dO (COOH), and both peaks are shifted by ∼0.5 Å toward a smaller z value compared with those in the absence of water. Obviously, in the case of the COOHterminated SAM, water induces rather large structural changes in the surface region of the SAM. Comparison with other panels of Figure 2 reveals that the COOH-terminated surface is very sensitive to the presence of water molecules. The COOH end groups become more disordered in the presence of water, as indicated by wider and less ordered profiles of the three atoms of the COOH group when in contact with water. 8613

dx.doi.org/10.1021/la201328y |Langmuir 2011, 27, 8611–8620

Langmuir

ARTICLE

Figure 3. Surface microchanges of SAMs with COOH-, CN-, and NH2-terminated groups in the absence of water and after complete surface wetting.

Complex hydrogen bonding between adjacent COOH end groups may give a reasonable explanation for the phenomena discussed above. The hydrogen of the hydroxyl group can form a hydrogen bond both with O(H) and dO of its neighboring COOH group. Furthermore, adjacent two alkyl chains can form a structure like a dimer, as shown in Figure 3a, involving two OH 3 3 3 OdC hydrogen bonds of the two COOH end groups. After the surface wetted, H-bonds formed between water molecules and COOH groups break this kind of structure. Therefore, the density profiles of COOH group show significant changes compared to the other five terminated groups. Figure 3 summarizes the micromolecular level structural changes in the surface region of hydrophilic SAMs with CN-, NH2-, and COOH-terminated alkyl chains in the absence of water versus under surface wetting. The CN bond stretched outside along the z-axis (Figure 3b), the configuration of NH2 group turned over (Figure 3c), and dimer structure discussed above was broken, after the three hydrophilic surfaces wetted. The changes of tilt angle between silicon substrate and alkyl chains of each neat SAM, compared to that of the final configuration in the presence of water ball, are shown in Table 1. The first three columns show that almost no structural change appears in the silicon substrate region of the hydrophobic (CH3-, CdC-, and

Table 1. Title Angle (deg) between Alkyl Chains with Different Terminal Groups and Silicon Substrate system

CH3

CdC

OCH3

CN

NH2

COOH

angle (neat SAMs) angle (wetting SAMs)

50.0 49.4

53.7 53.9

55.2 55.6

45.3 48.8

48.1 51.4

56.8 64.2

OCH3-terminated) SAMs due to the formation of water droplets on the surfaces. In sharp contrast, the tilt angles in the silicon substrate region of the hydrophobic groups CN-, NH2-, and COOH-terminated SAMs change in the presence of the water ball on the surface, under the effect of hydrogen bonding. The most obvious change in tilt angle occurs at the silicon substrate region of the COOH-terminated SAM, which varies from 56.8 to 64.2, indicating that the alkyl chains gradually stretch from the bottom along the z-axis because of the water molecule interactions with the polar head. 3.2. Wetting of the SAMs. Snapshots from the simulations of six SAMs with different surfaces studied in contact with the same water ball described above are summarized in Figure 4, giving an intuitive visual of the wetting character of all the SAMs with different terminal groups (CH3, OCH3, CdC, CN, NH2, COOH) 8614

dx.doi.org/10.1021/la201328y |Langmuir 2011, 27, 8611–8620

Langmuir

ARTICLE

Figure 4. Matrix of snapshots from SAM simulations as a function of changing surface (along the y-axis) and the increase of simulation time (along the x-axis). The fraction of wetted surface area of each configuration is marked below the corresponding snapshot. Color scheme: O, red; N, blue; carbon, gray; silicon, gold.

from most hydrophobic to hydrophilic. In Figure 4, each row of the matrix of snapshots corresponds to one kind of SAM, and the end group is noted along the y-axis. From left to right, five conformation snapshots are presented for simulation times at the end of 1, 2, ..., 5 ns. The water ball containing 509 water molecules totally is the same for every SAM regardless of shape, size, or initial position. Figure 4 shows that the water ball forms distinct shapes on hydrophobic surfaces (CH3-, OCH3-, and CdC-terminated SAMs) and on hydrophilic surfaces (CN-, NH2-, and COOHterminated SAMs). The water ball on CH3-, OCH3-, and CdCterminated SAMs keeps the ball shape during the simulation because of the waterwater hydrogen-bonding effect. Minimal interaction between water molecules and the SAM surface mean that the water ball cannot spread over the hydrophobic surface. By contrast, water molecules can form N 3 3 3 H and O 3 3 3 H H-bonds with N, dO, and O(H) on the hydrophilic surfaces of CN-, NH2-, and COOH-terminated SAMs as well as with other water molecules inside the water ball. As the simulation time increases, water molecules spread over the hydrophilic surface gradually. For quantitative description of the degree of wetting of the six SAM surfaces as a function of their composition and of the water coverage during the simulation, the solvent-accessible surface area (SASA) technique that has been for the first time applied to wetted SAM surfaces by Sz€ori et al.18 was adopted to evaluate the area of the surface of each SAM that is in contact with water. In this study, the solvent-accessible surface was calculated using the Lee and Richards algorithm with a probe radius of 1.4 Å. The results of every frame of each SAM during the simulation are summarized in panel a of Figure 5. In this figure, the total area of the surface covered by water during the simulation is depicted for the six SAMs with different end groups as a function of water

coverage. For each kind of self-assembled monolayer, after 5 ns the largest area of the wetted surface is found on the COOHterminated SAM due to water ball spreading on this surface, and accordingly the smallest area of the wetted surface is on the CH3terminated SAM. The surface covered by water increases for all six SAMs, as the simulation time increases. The coverage rate of all six SAMs shows a clear difference between hydrophilic surfaces and hydrophobic surfaces. As shown in Figure 5a, the coverage rate increases with the surface of each SAM changing from hydrophobic (CH3-, CdC-, and OCH3-terminated head groups) to hydrophilic (CN-, NH2-, and COOH-terminated head groups). For the hydrophilic surface (CN-, NH2-, and COOH-terminated head groups), however, a rapid increase of the wetted surface area is observed as a result of water molecules spreading over the surface. The final coverage rate is consistent with the hydrophilicity of the polar head of the SAM, in the order COOH > NH2 > CN. By contrast, for hydrophobic surface (CH3-, CdC-, and OCH3-terminated head groups) the slope of the curves shows a very small change during simulation, indicating these three SAMs having little capacity for water loading. Since the SASA algorithm is convenient and accurate to quantify the degree of wetting of the SAM surfaces as a function of their composition and of the water load, it is very sensitive to surface corrugation of all the SAMs. So it is clear that the surface area detected by the probe (radius is 1.4 Å) is larger than that from a simple geometrical area corresponding to the diamondshaped area formula of the simulation cell (∼3268 Å2). So the SASA procedure yields a larger area compared with the bottom area of the simulation lattice. The surface area measured by the probe is of course different for all six kinds of SAMs, as the change of surface property from hydrophobic to hydrophilic is accompanied by a corresponding change in surface corrugation. Finally, the 8615

dx.doi.org/10.1021/la201328y |Langmuir 2011, 27, 8611–8620

Langmuir

Figure 5. (a) Water-covered SAM surface area as detected by the probe in contact with different SAMs as a function of simulation time. (b) Total area of the free surface (not covered by water) for the SAMs with different end groups.

SASA procedure may theoretically yield different values for each SAM surface area, depending on the effect of water molecules absorbed on the SAM surface. The interaction between water molecules in the vicinity of the surface and the SAM during the simulation may result in conformational changes of SAM chains (e.g., bending, tilting, extending, etc.), so that the SAM surface detected by the SASA procedure will be somewhat different due to changes in the surface corrugation. The “neat” SAM surface area which is in contact with water during the simulation is detected by the probe, ignoring the water molecules on the SAM surface. The total surface area of all six “neat” SAMs detected by the SASA algorithm is depicted as a function of water molecules existing on the surface and is shown in panel b of Figure 5. This analysis shows that the total surface area of all six SAMs with different terminated head groups is fairly steady with fluctuations not exceeding 50 Å2 over time. For all six kinds of SAM, the surface area detected by the probe is not sensitive to the presence of water molecules and remains constant during the entire simulation within the error bars of the calculation. Therefore, based on the above-discussion, the total surface area for all six SAM with different terminal groups was detected by the SASA algorithm as the average of the values during the simulation. These total surface area values were then used to calculate the fraction of the coverage rate for each SAM system studied during the simulation, and all the results are given in Figure 4 under the corresponding snapshots. In all six curves representing the SAM surface area values shown in Figure 5b, three lines (in red, green, and dark yellow, respectively) for hydrophobic SAMs with CH3-, CdC-, and OCH3-terminated end groups are in the upper part of the figure, which means that the hydrophobic SAM surfaces detected by the probe are more corrugated than the hydrophilic

ARTICLE

SAMs. In contrast, the surface containing COOH-terminated and NH2-terminated alkyl chains, shown in the middle part of the figure (in black and pink, respectively), are smoother than the hydrophobic surface. The CN-terminated surface of the SAM shown at the bottom of the figure in blue has the smallest surface area as calculated by the SASA algorithm, and the main reason is that every CN-terminated alkyl chain provides only one carbon atom and one nitrogen atom making up the SAM surface; therefore, a minimum number of atoms composing the CNterminated SAM surface may reduce the extent of the surface corrugation as compared to the other five SAM surfaces. Correspondingly, the area value calculated by the SASA algorithm is the smallest among the six SAMs. From the above analysis, it appears that the atom numbers in the terminal head groups forming the SAM surface are a major factor influencing the surface area value calculated by the SASA algorithm; the smaller the number of atoms the surface contains, the smaller the value detected by the probe. Therefore, alkyl chains of SAM with terminal groups containing small numbers of atoms result in decreasing surface corrugation. For SAMs with little difference in the number of end-group atoms on the surface, the surface of the hydrophilic SAMs with COOH- and NH2-terminated head groups is smoother than that of hydrophobic SAMs with CH3-, CdC-, and OCH3-terminated head groups, as clearly shown in Figure 5b. This result is mainly due to the interaction between polar heads (e.g., H-bonds) of the hydrophilic SAM surface which reduce the surface corrugation. To further verify the differences in wetting characteristics between SAMs with hydrophobic and hydrophilic surfaces, a special ordered self-assembled monolayer containing both hydrophobic (CH3-terminated alkyl chains) and hydrophilic (COOH-terminated alkyl chains) components was constructed. A ribbon of 48 COOH-terminated alkyl chains (6  8) connected to the silicon (111) substrate were placed between 80 CH3-terminated alkyl chains divided into two parts (5  8) (40 chains on each side of the COOH-terminated SAM; see the “neat” snapshot in Figure S1). A water molecule ball was placed into the simulation box, with all the characteristics the same as that described above. A 5 ns MD simulation was conducted under the same conditions as the previous simulations. The final snapshot is shown in Supporting Information Figure S1. The middle surface region in blue is completely covered by water molecules, while both sides of the surface are empty. The figure shows a clear distinction of wetting characteristics between the hydrophobic and hydrophilic surfaces of SAMs. 3.3. Spreading Process of the Water Ball on SAM Surface. In order to investigate spreading process of the water ball on the six different SAM surfaces, we have calculated the mean-square displacements (MSD) of water molecules along axis z normal the six SAM surfaces. Figure 6 shows the MSD of the water molecules from the simulations, which are calculated from the equation  N  1 2 jri ðtÞ  ri ð0Þj ð3Þ MSDðtÞ ¼ N i¼1



where N is the number of target molecules and ri(t) is the position of molecule i at time t. Diffusion coefficients (D) can then be obtained from the slope of the mean-square displacement versus time curve, using the well-known Einstein relation D ¼ lim

Δt f ¥

8616

Æjri ðtÞ  ri ð0Þjæ2 ðΔr 2 Þ ¼ lim Δt f ¥ 2dΔt 2dΔt

ð4Þ

dx.doi.org/10.1021/la201328y |Langmuir 2011, 27, 8611–8620

Langmuir

ARTICLE

Figure 6. (a) Time evolution of mean-square displacements (MSDs) of the water molecules adsorption onto the six different SAMs surfaces, along axis z. (b) MSDs of the water molecules in the upper part and in the lower half of water ball adsorbed onto the three hydrophilic SAM surface, in the xy-plane.

Figure 7. Microscopic conformational changes for one single water molecule with an alkyl chain from each SAM (CH3-, OCH3-, CdC-, CN-, NH2-, and COOH-terminated head groups). The values for single chains without water are given in parentheses.

where d is the dimensionality of the system and ri(t) and ri(0) are the center-of-mass coordinates of the ith water molecules at times t and t = 0, respectively. The data, shown in Figure 6a and Table S1 (Supporting Information), reveal that the mobility of water molecules along axis z is strongest on the COOH-terminated surface and generally restricted on the hydrophobic surfaces. The self-diffusion coefficients of water molecules are consistent with the strength of SAM surface wetting ability (from hydrophobic to hydrophilic). The adsorption rate of water molecules mainly depends on the wetting ability of SAM surface. The more hydrophilic the SAM surface, the faster water molecules are adsorbed. In order to get a deeper insight into the fluid behavior of water molecules in the water ball onto the hydrophilic SAM surface, we have calculated the MSD of water molecules on the three hydrophilic SAM surfaces parallel to the surface (i.e., the xy plane). The water molecules in the lower half of the water ball versus those in the upper part were calculated separately. The self-diffusion coefficients were also evaluated in the same way from the MSD (Figure 6b) using eq 4, and the results are listed in Table S2. It is evident that the water molecules in the lower half of the water ball exhibit a more restricted motion in the plane of hydrophilic SAM surface. It is mainly due to the polar groups (CN-, NH2-, and

COOH-terminated head groups) forms hydrogen bonding with the water molecules near the hydrophilic surface, resulting in a firmly adsorption of water molecules on the SAM surface. The calculated values of Dxy of water molecules in the upper part are much more than those of water molecules in the lower half (shown in Table S2). It indicates the effect that the three hydrophilic SAM surfaces have on the water molecules in the upper part is much less than that on the water in their vicinity. The water molecules not in the vicinity of the hydrophilic SAM surface diffuse freely parallel to the plane until reaching the right place to form H-bonds with polar groups. The significant difference of diffusion behavior on the hydrophilic SAM surface between water molecules in the upper part of the water ball and those in the vicinity of the surface finally results in the water droplet spreading gradually on the surface. 3.4. Microstructure of Water Molecules and End Groups. To further exactly study the atomic interactions between water molecules and the end groups of SAMs after complete surface wetting, quantum chemical calculations are employed. Every individual alkyl chain with its different terminal group (CH3, OCH3, CdC, CN, NH2, COOH) is optimized in the presence and absence of one water molecule, and the changes of bond lengths and angles of each terminal group are described in detail (Figure 7). For the two hydrophobic SAMs with CH38617

dx.doi.org/10.1021/la201328y |Langmuir 2011, 27, 8611–8620

Langmuir

ARTICLE

Figure 8. Hydrogen bonding patterns between only one adsorbed water molecule and the three hydrophilic surface groups (CN- (a), NH2(b), COOH- (c)). The case shown here corresponds to the binding of two headgroups with one water molecule.

and CdC-terminal groups, the distances between H (CH3) and O (H2O), H (dCH2), and O (H2O) are 2.585 and 2.510 Å, respectively (shown in Figure 7a,b), indicating that no hydrogen bond formed between the two surfaces and water molecules. The impact of the water molecule on CH3- and CdC-terminated head groups is very weak, as the nonpolar head structure does not change significantly. For the two hydrophilic SAMs with NH2- and CN-terminated head groups, an NH type hydrogen bond is formed between the water molecule and the NH2 and CN groups, respectively, as the distances are 1.688 Å for N (NH2) 3 3 3 H (H2O) (Figure 7e) and 1.893 Å for N (CN) 3 3 3 H (H2O) (Figure 7d). For the NH2 group, with the formation of the NH hydrogen bond, the NC18 bond length increases from 1.449 to 1.486 Å, which is more obvious than the change of CtN bond length (from 1.154 to 1.170 Å). Hence, the strength of the NH type hydrogen bond formed between N (NH2) and H (H2O) is stronger than that formed between N (CN) and H (H2O). The N (CN) cannot easily form an NH hydrogen bond with H (H2O) because the electron density is primarily in the triple bond CtN. This can further explain why the SAM with NH2-terminal groups has a more hydrophilic surface compared to the SAM of the CNterminal groups at the atomic scale, which is consistent with the results of the MD simulation. The greatest impact of water molecules is on the geometric structure of the COOH polar head, as shown in Figure 7f. One H-bond (distance 2.214 Å) is formed between the carbonylic O (COOH) and H (H2O); at the same time, another strong H-bond (distance 1.575 Å) is formed between the hydroxylic H (COOH) and O (H2O). Under the influence of these two OH type hydrogen bonds, an obvious structural change happens in the COOH group: the CdO bond length increases from 1.211 to 1.253 Å, the HO bond length increases from 0.971 to 1.026 Å, the CO bond length increases from 1.348 to 1.355 Å, and the COOH group opens in the direction of the water molecule, as the bond angles — O1CO2 and — HO1C change from 122.3 to 121.8 and from 108.0 to 110.6, respectively. As a result, a distorted six-membered ring is formed between the COOH group and the water molecule so that the water molecule can be firmly fixed in the region around the COOH group. So the strong hydrogen bond (H (OH) 3 3 3 O (H2O)) and relatively stable hexagonal structure are the main reasons why the SAM with the COOHterminated surface has the strongest wetting ability among the six kinds of SAMs.

For the SAM with the OCH3-terminal group, an OH type hydrogen bond is formed between O (OCH3) and H (H2O) with distance 1.734 Å (as shown in Figure 7c), but the wetting ability of the SAM is still low probably due to the steric hindrance from the methyl group connected to the O atom, which prevents water molecules from reaching the right position to form H-bonds with O (OCH3) in the film. The binding energy (Eb) for one single water molecule with an alkyl chain from each SAM represents the degree of difficulty of their combination. As shown in Figure 7, it indicates that the sequence of Eb values (from high to low) is consistent with the strength of SAM surface wetting ability (from hydrophobic to hydrophilic). The smallest value of Eb-COOH (9.48 kcal/mol) represents the situation where water molecules are most likely to combine with the polar head COOH so that the surface of the COOH-terminated SAM is hydrophilic. The largest value for EbCH3 (1.56 kcal/mol) indicates that it is very difficult to combine water molecules and the CH3 group, which causes the CH3terminated SAM to display a significantly hydrophobic surface. Therefore, the value of binding energy for one single water molecule with an alkyl chain from SAM is an important indicator of the SAM surface wetting capabilities (a large Eb value corresponding to a hydrophobic SAM surface and a small one corresponding to a hydrophilic SAM surface). The only exception is the SAM with OCH3-terminated head groups, which has a more hydrophobic surface than the SAM with CN-terminated head groups, but a smaller EbOCHH3 compared to the Eb-CN. For the actual SAM structure with the OCH3-terminated surface, the hydrophobicity is mainly due to the steric effect of CH3 groups in the SAM surface which restrain water molecules from reaching the region. Actually, the most typical situation at a very low level of hydration is that of one water molecule “bridging” between two hydrophilic group-terminated chains (i.e., making one H-bond with each of the chains). As the level of hydration increases, the predominant pattern is “bridging” two chains by water dimer. This phenomenon was first reported in Sz€ori et al.’s work44 for COOH group terminated chains; three hydrogen-bonding patterns between two adsorbed water molecules and surface COOH groups were displayed in their work through MD simulation. However, there are still two problems needed to be solved: (i) whether all three hydrophilic groups may form the “bridging” structure with only one water molecule between two chains; (ii) there are three possible hydrogen bonding patterns between two 8618

dx.doi.org/10.1021/la201328y |Langmuir 2011, 27, 8611–8620

Langmuir

ARTICLE

Figure 9. Hydrogen bonding patterns between two adsorbed water molecules and the two hydrophilic surface groups (NH2- (a), COOH- (bd)). The case shown here corresponds to the binding of two headgroups and two water molecules, which are hydrogen bonded to each other.

adsorbed water molecules and surface COOH groups; what’s the most stable structure of them? In order to solve the problem i, we calculated three possible “bridging” structures with only one water molecule between two chains for the three hydrophilic SAMs after surface wetted, through the QM method. The results are summarized in Figure 8. From Figure 8, it is indicated that one water molecule may separately form one H-bond with each of the chains containing CN or COOH groups; therefore, the “bridging” structure with only one water molecule between two chains does exist in the two hydrophilic surfaces (CN- and COOH-terminated) after surface wetted. However, for the NH2-termianted SAM, there is only one H-bond formed, (the length of H 3 3 3 N is 2.541 Å, not formed a H-bond). It indicates that the “bridging” structure with only one water molecule between two chains can hardly form in NH2-termianted surface at a very low level of hydration. To investigate hydrogen bonding patterns between two adsorbed water molecules and hydrophilic surface groups, we have calculated the possible “bridging” structures with two water molecules between two hydrophilic chains, as the level of hydration increases. We find that the structure of “bridging” two chains by water dimer only exists in NH2- and COOHterminated surfaces. The results are summarized in Figure 9. As shown in Figure 9, it is indicated that there does exist three possible hydrogen bonding patterns between two adsorbed water molecules and surface COOH groups, which is reported in Sz€ori et al.’s work. Through compared the binding energies of the three structures calculated by the DFT method, we can conclude that as the level of hydration increases, the structure (d) may most possibly appear in the COOH surface area, and the structure (b) may hardly appear in the surface.

4. CONCLUSIONS We have performed a molecular dynamics study of flat selfassembled monolayers of 18-alkanethiolates on silicon (111), the surface of which was tuned from hydrophobic to hydrophilic by

entirely replacing methyl-terminated alkyl chains with methyl ether, vinyl, cyano, amino, and carboxylic acid-terminated chains. The microscopic wetting characteristics of these six SAM surfaces have been investigated for their level of wetting ability when connected with a water ball (509 water molecules and radius of 1.5 Å) after 5 ns simulation. We have used the solvent-accessible surface area (SASA) calculation to demonstrate the degree of wetting of the six SAM surfaces with different head groups —(CH3, CdC, OCH3, CN, NH2, COOH)—from hydrophobic to hydrophilic. In the water coverage region of each SAM, the behavior of water on the surfaces from hydrophobic and hydrophilic is dramatically different. On the three hydrophobic surfaces (CH 3 , CdC, and OCH 3 terminated), water forms droplets from the very beginning of hydration, and the droplets do not fully spread for the duration of the simulation. On the other hand, for the SAMs with hydrophilic surfaces (CN, NH2, and COOH terminated), the H-bonds formed between hydrophilic head groups and water molecules have broken the H-bonds structure formed between water molecules in the water ball and finally forced the water ball to spread out over the surface gradually. The final coverage of water molecules measured by the SASA algorithm for the six SAM surfaces varies essentially consistent with the experimental conclusion on these surfaces. Through the QM calculation, our data suggest that the value of binding energy for one single water molecule with an alkyl chain from SAM is an important indicator of the SAM surface wetting capabilities (a large Eb value corresponding to a hydrophobic SAM surface and a small one corresponding to a hydrophilic SAM surface). We conclude that at a very low level of hydration the “bridging” structure with only one water molecule between two chains do exist in the two hydrophilic surfaces (CN- and COOHterminated). As the level of hydration increases, the predominant pattern is “bridging” two chains by water dimer exist in the two hydrophilic surfaces (NH2- and COOH-terminated). 8619

dx.doi.org/10.1021/la201328y |Langmuir 2011, 27, 8611–8620

Langmuir The results of this investigation provide a microscopic perspective on the wetting properties of different organic surfaces (from hydrophobic to hydrophilic) that could be helpful in developing models to describe the wetting behavior of organic materials in a biological environment. The present study provides not only molecular-scale but also atom-scale insight into the changes in wetting characteristics of flat organic surfaces with six different terminal functional groups (CH3, CdC, OCH3, CN, NH2, COOH) and further elucidates the factors affecting changes in wetting behavior that accompany the surface changes of organic materials in a biological environment. An extension of our simulations including more SAMs with complex surfaces, such as OC6H5, OCH2CF3, and ethylene glycol (EG3OH)terminated surfaces, is currently underway. The effect of pure water, peptide, and protein solutions on complex surfaces of SAMs would be of interest in future studies of realistic systems.

’ ASSOCIATED CONTENT

bS

Supporting Information. Self-diffusion coefficients of the water molecules obtained from their MSDs; microscopic wetting of mixed SAM. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected] (S.-L.Y.); [email protected] (C.-B.L.). Author Contributions †

These authors contributed equally to this work.

’ ACKNOWLEDGMENT This work was supported by the National Science Foundation (20873074 and 21043008), National Basic Research program (2009CB930104) of China, and Independent Innovation Foundation of Shandong University (2009JC018). The authors thank Dr. Pamela Holt, Shandong University, for helpful discussions and manuscript editing. ’ REFERENCES (1) Ulman, A. Chem. Rev. 1996, 96, 1533. (2) Schreiber, F. Prog. Surf. Sci. 2000, 65, 151. (3) Masuda, Y.; Sugiyama, T.; Lin, H.; Seo, W. S.; Koumoto, K. Thin Solid Films 2001, 382, 153. (4) Ball, P. Chem. Rev. 2008, 108, 74. (5) Hautman, J.; Klein, M. L. Phys. Rev. Lett. 1991, 67, 1763. (6) Winter, N.; Vieceli, J.; Benjamin, I. J. Phys. Chem. B 2008, 112, 227. (7) Ellison, G. B.; Tuck, A. F.; Vaida, V. J. Geophys. Res. 1999, 104, 11633. (8) Rudich, Y. Chem. Rev. 2003, 103, 5097. (9) Rudich, Y.; Donahue, N. M.; Mentel, T. F. Annu. Rev. Phys. Chem. 2007, 58, 321. (10) Cooper, M. A. Nat. Rev. Drug Discovery 2002, 1, 515. (11) Ito, Y. Biomaterials 1999, 20, 2333. (12) Webb, K.; Hlady, V.; Tresco, P. A. J. Biomed. Mater. Res. 2000, 49, 362. (13) Kapur, R.; Rudolph, A. S. Exp. Cell Res. 1998, 244, 275. (14) Jenney, C. R.; Defife, K. M.; Colton, E. J. Biomed. Mater. Res. 1998, 41, 171.

ARTICLE

(15) Sumanth, N.; Jamadagni, R. G.; Shekhar, G. Langmuir 2009, 25, 13092. (16) Douglas, L.; Irving; Donald, W.; Brenner. J. Phys. Chem. B 2006, 110, 15426. (17) Wang, H.; Buddy, D.; Ratner, E.; Helene., S.; Jiang, S. Y. J. Phys. Chem. C 2007, 111, 8504. (18) Sz€ori, M.; Tobias, D. J.; Roeselova, M. J. Phys. Chem. B 2009, 113, 4161. (19) Sigal, G. B.; Mrksich, M.; Whitesides, G. M. J. Am. Chem. Soc. 1998, 120, 3464. (20) Kane, R. S.; Deschatelets, P.; Whitesides, G. M. Langmuir 2003, 19, 2388. (21) Sethuraman, A.; et al. Langmuir 2004, 20, 7779. (22) Paul, R.; David, E.; Kirsty, R.; Carole., C. P. J. Mater. Sci.: Mater. Med. 2007, 18, 1263. (23) Ruan, C. Y.; Lobatsov, V. A.; Vigliotti, F.; Chen, S. Y.; Zewail, A. H. Science 2004, 304, 80. (24) Raviv, U.; Laurat, P.; Klein, J. Nature 2001, 413, 51. (25) Raviv, U.; Klein, J. Nature 2002, 297, 1540. (26) Gallo, P.; Rovere, M.; Ricci, M. A.; Hartnig, C.; Spohr, E. Europhys. Lett. 1999, 49, 183. (27) Lu, L.; Berkowitz, M. L. J. Chem. Phys. 2006, 124, 101101. (28) Leng, Y. S.; Cummings, P. T. Phys. Rev. Lett. 2005, 94, 026101. (29) Leng, Y. S.; Cummings, P. T. J. Chem. Phys. 2006, 124, 074711. (30) Moussa, S. G.; McIntire, T. M.; Sz€ori, M.; Roeselova, M. J. Phys. Chem. A 2009, 113, 2060. (31) Natalia, S.; Rahul, G.; Pawel, K.; Shekhar, G. Phys. Rev. Lett. 2009, 102, 156101. (32) Srivastava, P.; Chapman, W. G.; Laibinis, P. E. Langmuir 2009, 25, 2689. (33) Srivastava, P.; Chapman, W. G.; Laibinis, P. E. Langmuir 2005, 26, 12171. (34) Yuan, S. L.; Ma, L. X.; Zhang, X. Q.; Zheng, L. Q. Colloids Surf., A 2006, 289, 1. (35) Berendsen, H. J. C.; Grigera, J. R.; Straatsma, J. P. J. Phys. Chem. 1987, 91, 6269. (36) Sun, H.; Ren, P.; Fried, J. R. Comput. Theor. Polym. Sci. 1998, 8, 229. (37) Ewald, P. Ann. Phys. 1921, 369, 253. (38) Sun, H. J. Phys. Chem. B 1998, 102, 7338. (39) Sigal, G. B.; Mrksich, M.; Whitesides, G. M. J. Am. Chem. Soc. 1998, 14, 120. (40) Faucheux, N.; Schweiss, R.; L€utzow, K. L.; Werner, C.; Grotha, T. Biomaterials 2004, 25, 2721. (41) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian03; Gaussian, Inc.: Pittsburgh, PA, 2004. urmann, G. J. Chem. Soc., Perkin Trans. 1993, (42) Klamt, A.; Schu€ 2, 799. (43) Delley, B. Mol. Simul. 2006, 32, 117. (44) Sz€ori, M.; Jedlovszky, P.; Roeselova, M. Phys. Chem. Chem. Phys. 2010, 12, 4604.

8620

dx.doi.org/10.1021/la201328y |Langmuir 2011, 27, 8611–8620