Concentration Mediated Structural Transition of Triblock Copolymer

Apr 30, 2014 - A drastic change in the film coverage of L2 is observed in the case of 4 g/L film as shown in Figure 6a,b. The height profile shows tha...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Langmuir

Concentration Mediated Structural Transition of Triblock Copolymer Ultrathin Films Jayanta K. Bal,*,† Manabendra Mukherjee,⊥ Nicolas Delorme,† Milan K. Sanyal,⊥ and Alain Gibaud† †

LUNAM Université, IMMM, Faculté de Sciences, Université du Maine, UMR 6283 CNRS, Le Mans Cedex 9, 72000, France Saha Institute of Nuclear Physics, 1/AF Bidhannagar, Kolkata 700064, India

⊥

S Supporting Information *

ABSTRACT: X-ray reflectivity, atomic force microscopy, X-ray photoelectron spectroscopy, and contact angle measurement techniques are used to study the structural changeover as a function of concentration of poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) (PEO-PPO-PEO) triblock copolymer diluted in toluene spin-coated as ultrathin films on hydrophilic Si substrate. A lamellar structure made of three alternating incomplete bilayers is observed until the concentration of copolymer solution attains a threshold value of about 3.6−4 g/L. Around this concentration and beyond, the entanglement of polymer chains takes place during drying and the growth of a homogeneous film made of complete bilayers on Si substrate is observed. The strong hydrophilic nature of the Si substrate dictates the growth of this amphiphilic copolymer. We evidence that the lower part of the films is made of hydrophilic PEO blocks attached to the substrate while the hydrophobic PPO blocks are directed toward air.

■

in drug delivery,15,16 biomembrane activities,17 and in material science (adhesive properties, lubricants, membranes, and coatings),18,19 lithography20 and electronic devices (light emitting diode, photodiodes, and transistors, etc.).21 Furthermore, such copolymers are well-known as reducing agents for different types of nanoparticle formation.22−30 A detailed study of this polymer solution has been made with or without nanoparticles. Thus, at first the growth of pure copolymer is essential prior to further understanding. The adsorption and desorption behaviors of pure symmetric triblock copolymers from aqueous solutions onto solid surfaces were studied via ellipsometry,31 quartz crystal microgravimetry (QCM),32 total internal fluorescence spectroscopy (TIRF),33 and surface plasmon resonance spectroscopy (SPR).34 In this regard, nondestructive X-ray reflectivity (XRR) technique35,36 would be more advantageous as it simultaneously provides the information about thickness (resolution ∼ angstrom) and electron density, which is proportional to the mass density or coverage of the grown film on solid surfaces. To the best of our knowledge, no such studies have been made so far revealing the structure of ultrathin films of asymmetric triblock copolymer prepared by spin coating. In this work, we demonstrate that the structure of ultrathin films made of highly asymmetric PEO-PPO-PEO copolymer on Si surface can be tuned by varying the concentration of polymer solutions. Vertical stacking of three alternating bilayers for all the films with increasing density of the upper two bilayers with

INTRODUCTION Amphiphilic block copolymers have emerged as smart materials for combining antagonist properties of different molecules by building a macromolecule in which the antagonist polymer blocks are attached by a covalent bond. The ability of these block copolymers to undergo microphase separation has generated significant interest in their application for forming ordered morphologies.1−10 The conformation of any macromolecule in solution is governed by the equilibrium between the interaction strengths of the polymer segments among themselves and with the molecules of the solvent.11,12 This equilibrium is broadly referred to as the “solvent quality” and is described by a variety of parameters such as the Hildebrand solubility13 and the Flory parameters.14 According to the Hildebrand solubility theory, solubility parameter (describes as the square root of the cohesive energy density of a material) of a good solvent is closer to that of the macromolecule, whereas Flory theory suggests that the smaller (or more negative) the Flory parameter, the more favorable the monomer−solvent interaction. For a block copolymer like PEO-PPO-PEO, this balance becomes more complex due to the contributions from the interactions of two different building blocks, namely, PEO block, which is hydrophilic, and PPO, hydrophobic. Therefore, an unusual circumstance appears when these interactions involve a solvent which is good for one of the units but bad for the other(s). This is called a selective solvent, being selective to the block whose solvation is stronger. Thus, it leads to an amphiphilic behavior of the polymer, and as a consequence, it has a tendency to selfassemble according to the nature of surfactant solutions. The self-assembly of block copolymers into well-defined structure has opened up numerous possibilities for applications © 2014 American Chemical Society

Received: January 21, 2014 Revised: April 16, 2014 Published: April 30, 2014 5808

dx.doi.org/10.1021/la5002607 | Langmuir 2014, 30, 5808−5816

Langmuir

Article

Final WCA value was determined from an average of the different measurements.

increase in polymer concentration is observed. A transition from incomplete to complete bilayer structure growth, which appears as a homogeneous monolayer film in terms of electron density, is encountered with increasing concentrations. Ordering of these stacked bilayers is initiated by the nature of Si surfaces.

■

■

RESULTS AND DISCUSSION Dynamic Light Scattering and Chain Conformation in Solution. In order to understand the chain conformation in toluene solution and their connection with film structure, we performed hydrodynamic size measurements by DLS technique. DLS measurements were carried out with different solutions having the concentrations ranging from 1 g/L to 6 g/L. The hydrodynamic size of polymer chains as a function of concentration is plotted in Figure 1. As the copolymer molecules

EXPERIMENTAL SECTION

PEO208-PPO69-PEO2 polymer (Polysciences Inc., USA, Mw = 13 300) thin films of different concentrations (varied from 1 g/L to 4.5 g/L) were prepared by adjusting the concentration of polymer−toluene solutions. Films were deposited by spin coating (Karl Suss) these solutions at 1700 rpm (acceleration 4000 rpm) for 1 min onto RCA cleaned Si(100) substrates. In RCA cleaning, the Si surfaces (of size ∼15 × 15 mm2) were made hydrophilic by introducing hydroxyl group (−OH) after boiling them in a mixture of ammonium hydroxide (NH4OH, Sigma-Aldrich, 25%), hydrogen peroxide (H2O2, Acros Organics, 39%), and Milli-Q water (H2O:NH4OH:H2O2 = 2:1:1, by volume) for 10 min at 100 °C. This solution is designated to remove organic contaminants by both solvating action of the NH4OH and the oxidizing action of the H2O2. The oxidizing agent H2O2 forms a continuous thin silicon oxide layer (thickness ∼1.5−2 nm) on the substrate surface. Furthermore, the metal surface contaminants are also oxidized by H2O2 and dissolved by the complexing effectiveness of the NH4OH. Contact angle for RCA cleaned Si surfaces of ∼14−17°37was observed which corresponds to hydrophilic nature. Then the substrates were dried prior to spin coating. XRR technique was used to characterize the thickness and quality of the spin coated films. The films were annealed at 45 °C for 24 h. Dynamic light scattering (DLS) measurements (Zetasizer Nano-S, Malvern Instruments) performed on different concentrations of solutions prior to film preparation. XRR measurements were carried out using a versatile X-ray diffractometer (XRD) to investigate the structure of these films. The diffractometer (Empyrean Panalytical) was equipped with a Cu source (sealed tube) followed by a W/C mirror to select and enhance Cu Kα radiation (λ = 1.54 Å). All measurements were carried out in θ−θ geometry for which the sample was kept fixed during the measurements. The intensity was measured with a Pixel 3D detector using a fixed aperture of 3 channels (0.165°) in the 2θ direction. Under such conditions, at a given angle of incidence θ, a nonvanishing wave vector component, qz, is given by (4π/λ)sin θ with resolution 0.0014 Å−1. XRR technique essentially provides an electron density profile (EDP), i.e., average in-plane (x−y) electron density (ρ) as a function of depth (z). From EDP it is possible to estimate film thickness, electron density, and interfacial roughness. Analysis of XRR data has been carried out using the matrix technique.35,36 For the analysis, the film is divided into a number of layers including roughness at each interface. Surface morphology of the films was collected through Agilent 5500 AFM in intermittent-contact mode in air (Nanosensor PPP-NCHR-W tip) and scans were performed over several portions of the films for different scan areas after completion of XRR measurements. WSxM software was used for image processing and analysis.38 XPS core-level spectra were taken with an Omicron Multiprobe (Omicron NanoTechnology GmbH, UK) spectrometer fitted with an EA125 hemispherical analyzer. A monochromated Al Kα X-ray source operated at 150 W was used for the experiments. The analyzer pass energy was kept fixed at 30 eV for all the scans with analyzer angular acceptance of ±1°. The background correction of the data was done by straight line method and the curve fitting of the core XPS lines was carried out using Gaussian−Lorentzian sum functions in PeakFit software. Water contact angle (WCA) measurements of triblock ultrathin films were carried out using CA goniometer (Raḿ e-Hart) to investigate the hydrophilicity/hydrophobicity of the ultrathin films. For the measurement, a drop (4 μL) of water (Milli-Q) was placed on the sample surface. The image of the water droplet was captured at a fixed interval of time from the placement of the droplet. Such measurements were done at five or six different places over the sample surface in each case. WCA values were calculated after analyzing those images using Scion image software.

Figure 1. Hydrodynamic size distribution observed from DLS for triblock copolymer conformation in toluene solvent as a function of copolymer concentration.

contain hydrophobic as well as hydrophilic blocks, it is expected that they would form core−shell like structures with the hydrophilic block as core and the hydrophobic block forming the shell facing toluene. We have observed previously39 that in aqueous solution the size of unimer and micellar structures are on the order of 5 and 20 nm, respectively. On the contrary, in toluene solution in the present case the size of the polymer chains/coils is found to be even larger than the completely elongated chain (∼118 nm). This indicates that the amphiphilic molecules do not form a core−shell-like structure in toluene solution unlike that in aqueous solution. The observed size of the chains is also much larger that the calculated radius of gyration of the chains (∼15 nm). Furthermore, the shape of polymer objects in solution can also be estimated from so-called Perrin or shape factor F which provides insightful information regarding shape of molecules. This factor is defined by the ratio between hydrodynamic diameters, obtained from DLS experiments, and theoretical calculations from molecular mass. If F is close to 1, this means the object is nearly spherical; when it deviates from 1 that means the object is elongated. In our case, F is turing out to be very large, i.e., between 7 and 10. Therefore, these observations imply that the chains are in a somewhat open structure in the solution rather than forming a core−shell or a Gaussian coil-type conformation. This may be explained in terms of the fact that the hydrophobic PPO block in the chains is shorter in length to eclipse the much longer hydrophilic PPO block. It may be possible that few open chains together form an energetically convenient structure in the toluene solution. The large fluctuation of size and the statistical error with change of concentration observed from the data also supports that the chain conformation is not very stable, but rather it has a tendency to fluctuate. 5809

dx.doi.org/10.1021/la5002607 | Langmuir 2014, 30, 5808−5816

Langmuir

Article

X-ray Reflectivity and Electron Density Profile. Solutions of this block copolymer were used to fabricate ultrathin films on hydrophilic RCA cleaned Si substrates. Observed and calculated XRR data of different films grown from different solutions are shown in Figure 2. It is clear from this figure that the

marked by down-headed arrows in the XRR profiles of Figure 2, is not observable in more concentrated samples starting from 3.6 g/L (shown in Figure S1) to 4.5 g/L films. The films then appear to be homogeneous as evidenced by a single period in the Kiessig fringes. The analysis of the Kiessig fringes and of their modulation allows evidencing the changes occurring in the films when increasing the concentration of the copolymer. To obtain EDPs, which contain detailed quantitative information about the film structure, all the reflectivity curves were analyzed by the matrix technique. As shown in the inset of Figure 2, it is interesting to note that the overall electron density profile (L) of the copolymer films can be divided into two main parts: the first part denoted L1 corresponds to a highly dense thin layer near the Si substrate; the second part denoted L2, on the top of L1, is less dense and thicker. The electron density and the thickness of layers L1 and L2 are labeled as ρ1, ρ2 and d1, d2 respectively. Theses parameters along with surface electron densities (ρS1, ρS2) are calculated from EDPs and plotted in Figure 3 as a function of concentration for clarity. The electron density of L2 is not uniform throughout the layer, but rather a mild variation along depth (z) is visible at low concentration. Hence, ρ2 is considered as the average value of nonuniform L2 layers. The thickness of both layers is found to increase with the concentration of copolymer until it reaches 3 g/L (Figure 3a). The electron density of L1 remains nearly constant while that of L2 increases with concentration to actually merge with the layer density of L1 (Figure 3a). In the concentration regime 1 g/L to 3.6 g/L, ρ2 increases rapidly in a nonlinear way, and beyond this range, it increases slowly in a linear fashion. Note that the average density of L1 (0.324 e/Å3) is fairly close to that found for the PPO (0.377 e/Å3) and PEO (0.369 e/Å3) blocks in the micellar phase of P123.40 The value of electron density of PEO is a bit controversial and was also found to be 0.411 e/Å3 in another study carried out on Brij58 surfactant.41 This is not so surprising since the electron density of such blocks is quite difficult to access

Figure 2. Normalized XRR data (different symbols) and analyzed curves (solid line) of different film concentrations (curves are shifted vertically for clarity). Two arrows in each curve indicate the periodicity of modulation present in overall film oscillations. Insets: corresponding EDPs. Reflectivity of 3.4 g/L and 3.6 g/L are shown in Figure S1.

films exhibit Kiessig fringes with a modulated amplitude especially in the case of the 1 g/L, 2 g/L, 3 g/L, and 3.4 g/L (shown in Figure S1) films. This modulation, whose period is

Figure 3. Evolution of (a) electron density (open symbols), thickness (solid symbols), and (b) surface electron density of L1, L2, and L layers of grown films as a function of copolymer concentration. 5810

dx.doi.org/10.1021/la5002607 | Langmuir 2014, 30, 5808−5816

Langmuir

Article

Figure 4. AFM images showing the topography of 1 g/L PEO-PPO-PEO films (a) size 3 × 3 μm2 and (c) size 2 × 1.2 μm2 on Si substrate. (b) and (d) are their corresponding phase images, respectively. hm indicates maximum height variation. Inset of (c) shows the line profile on the film surface. A, B, and C represents 3 locations along the line profile.

of the bulk film is the same with L1 and L2, which might not be the case. This can be verified from AFM topography information which provides true coverage in terms of surface area. Atomic Force Microscopy and Surface Topography. Figure 4a,b shows topographic and phase-contrast AFM images, respectively, recorded for 1 g/L film. The topographic image shows a top layer partially covering a complete layer underneath which can be primarily thought to be the Si substrate. In order to confirm this statement, the lower layer was scratched with the AFM tip42 revealing that a film was fully covering the underlying Si substrate. The topography is thus in agreement with the EDP which indicates the presence of a denser film L1 at the bottom and a less dense film L2 on top. However, this is not the regular morphology over the entire surface, but rather in some areas the L1 layer is more uncovered, which results in lower electron density of the L2 layer as observed in EDP. AFM phase imaging is known to be sensitive to adhesion contrast43 and thus to hydrophilic/hydrophobic contrast. Phase image in Figure 2b shows that there is no difference in phase between the top parts of layers L1 and L2 except at their boundaries (characterized by darker regions). High resolution topographic image of the boundaries indicates that apart from L1 (marked as A) and L2 (marked as B), an additional intermediate layer (marked as C) is clearly present. The presence of this additional layer C was also realized from the nonuniformity or gradual electron density variation of L2 layer as observed in EDP (inset of Figure 1), where the bottom part of L2 is denser than the top part. Interestingly, the phase-contrast image, shown in Figure 4d, shows two distinct regions with clear and well-defined boundaries. Region C shows completely different contrast in comparison to A and B. It may be noted that PEO-PPO-PEO is composed of hydrophilic PEO and hydrophobic PPO blocks. As the third PEO block is very small (2 EO monomers) compared to other two blocks it can be ignored for simplicity. Hence, it is very tempting to explain the phase contrast by the difference in hydrophilicity and hydrophobicity of the two major PEO and

in absolute, as (1) it is strongly affected by the amount of solvent remaining in the blocks and (2) it depends on the mass density that is taken for the blocks. Nevertheless, it can be said that the contrast of electron density between the two blocks is rather weak so that even a continuous film made of these blocks having a lamellar ordering will appear uniform. With increase in concentration a drastic difference in the electron density is encountered (Figure 3a). The contrast of electron density between layers L1 and L2 decreases substantially until an almost homogeneous layer is formed at a concentration of 4.5 g/L. In this transition regime where an incomplete film transforms into an apparent monolayer (in terms of density) in the concentration ranging from 3 g/L to 3.6 g/L, a decrement in the thickness of layer L2 as well as in the overall film thickness is observed. Once it forms almost a continuous layer at 4 g/L, a further increase in concentration is needed to produce a conventional increment in thickness of the film. Furthermore, the surface electron densities of L1, L2, and L, labeled as ρS1, ρS2, and ρS, respectively, which accounts for the number of electrons per unit area, are strongly evolving with concentration as shown in Figure 3b. Notably, surface electron density of a layer is calculated by multiplying the thickness and electron density of that particular layer. ρS1 changes a little and finally saturates at around 10 e/Å2, but ρS2 shows monotonic increase with concentration compared to ρS. This effect is a consequence of the increase of both the thickness of layer L2 and its electron density. It is evident from the figure that above 3.6 g/L, since where the modulation in reflectivity (Figure S1) is no longer visible due to the small electron density contrast between L1 and L2 layer, ρS2 and ρS increase steadily. Assuming bulk density 0.37 e/Å3 of PEO or PPO block as 100% coverage, the estimated coverage values of L2 (from the density point of view) for 1 g/L, 2 g/L, 3 g/L, 3.4 g/L, 3.6 g/L, 4 g/L, and 4.5 g/L films are ∼16%, 24%, 35%, 50%, 63%, 67%, and 71%, respectively, whereas the coverage of L1 varies from 87% to 75%. Here one assumption is made that the compactness 5811

dx.doi.org/10.1021/la5002607 | Langmuir 2014, 30, 5808−5816

Langmuir

Article

Figure 5. AFM images showing the topography of 3g/L PEO-PPO-PEO films (a) size 3 × 3 μm2 and (c) size 2 × 1.2 μm2 on Si substrate. (b) and (d) are their corresponding phase images, respectively. hm indicates maximum height variation. Inset of (c): Line profile on the film surface. A, B, and C represent 3 locations along the line profile.

Figure 6d which suggests homogeneous properties of the whole top surface. The coverages of L1 and L2 obtained from AFM and XRR analysis for different films are depicted in Figure 7. It is evident from the figure that the coverage obtained from AFM is much greater compared to that obtained from XRR. The bared L1 layer, marked as A, in AFM images is helpful to estimate its coverage, which is nearly 100% in all the films, as we could not see the underlying substrate. In contrast, XRR shows less coverage, considering that the bulk density of this polymer is ∼0.37 e/Å3. These results suggest that the polymer within L2 is less compact than in L1. Interestingly their compactness approaches that of each other with higher concentration than that governing the homogeneous film growth. Even the compactness of the highly dense L1 layer remains less than their bulk compactness. Notably a large error (∼10%) is possible when estimating the coverage from AFM images especially for 1 g/L film, as it is inhomogeneous in large scale. X-ray Photoelectron Spectroscopy and Chemical Structure. As neither XRR nor AFM provides any information about the chemical nature of the layers we have employed XPS for this purpose. Angle resolved XPS was used to identify materials present in each individual layer. In this technique, attenuation of the photoemitted electrons within the sample has been utilized. As the mean free path of electrons does not change with the angle of emission, the electrons emitted normal to the surface are emitted from a larger depth of the sample compared to those emitted at grazing angle. In other words, electrons emitted at grazing angle are more surface sensitive compared to the ones emitted normal to the surface. Using this principle we have identified the chemical nature of the films as a function of their depth. It is worth noting that carbon atoms in PEO and PPO blocks are in a different environment and hence have different binding energy (BE). The peak at higher BE (at 286.6 eV) corresponds to C−O for both PPO and PEO, whereas the one at lower BE (at 285.2 eV) corresponds to C−C of PPO only.

PPO blocks, respectively. However, AFM images in phase contrast do not allow identifying which one is the PEO or the PPO block. A similar three-layer structure, labeled A, B, and C, is observed in the topography of the 3 g/L film and is depicted in Figure 5. Analogously to the previous case, it appears at first glance that A is the substrate and B is the copolymer film grown on it, but the height of B with respect to A (obtained from line profile) of ∼12 nm does not match the overall film thickness obtained by XRR (∼16.5 nm, Figure 2). It rather corresponds to layer L2 only, whose thickness obtained by XRR is ∼12.7 nm. Therefore, it is further clear that the substrate is fully covered by layer L1 which is again partially covered by layer L2. An intermediate layer corresponding to a different phase previously labeled as C is clearly observed near the edges of region B in this case also. Its existence was also envisioned from EDP analogously to 1 g/L film. A drastic change in the film coverage of L2 is observed in the case of 4 g/L film as shown in Figure 6a,b. The height profile shows that the height of region B with respect to A is ∼10 nm, which is close to the thickness of layer L2 (∼11.3 nm) obtained from XRR. The electron density of layer L2 is found to be close to that of L1 which indicates that the coverage of layer L2 is very high. It should be noted that the AFM picture is not a true representative of the average film morphology. Regions with defects are searched out to gain information on the buried layers. However, there is no appreciable difference in phase image (Figure 6b). Furthermore, intermediate layer C is absent here which is also expected from the uniform density of layer L2 as observed in EDP. This layer is probably completely covered due to better filling of layer L2 with increasing concentration. At a concentration of 4.5 g/L, the films are very uniform and practically defect-free as shown in Figure 6c. XRR also shows almost uniform average electron density throughout the film indicating the absence of any layered structure in terms of density. This is confirmed by the phase contrast image shown in 5812

dx.doi.org/10.1021/la5002607 | Langmuir 2014, 30, 5808−5816

Langmuir

Article

Figure 6. AFM images showing the topography of (a) 4 g/L (scan size = 10 × 10 μm2) and (c) 4.5 g/L (scan size = 5 × 5 μm2) PEO-PPO-PEO films on Si substrate. (b) and (d) are their corresponding phase images. Inset of (a): Line profile on the film surface. A and B represents 2 locations along the line profile. Region A disappears in (c) and (d). hm indicates maximum height variation.

compared to 90°. This indicates that PPO molecules are located in the upper part of the films. This feature is valid for all the films. However, a mild signal of toluene is observed in all the films. Water Contact Angle and Hydrophobicity/Hydrophilicity. In order to investigate the hydrophobicity and hydrophilicity of film surfaces we have performed WCA measurements of different films. Results show that there is large variation in WCA value for the incomplete films (where the L2 layer is incomplete), i.e., 1.5 g/L, 2 g/L, and 3 g/L films. They yield WCA that varies from ∼65° to ∼86° from one place to another place on the film surfaces which indicates the inhomogeneous hydrophobic-like nature of the surface. This is also in agreement with phase contrast images obtained by AFM. In very few regions hydrophilic PEO blocks are exposed along with hydrophobic PPO blocks which are probably reducing the WCA. On the other hand, WCA values are relatively stable between ∼85° and ∼92° for 3.6 g/L, 4 g/L, and 4.5 g/L films which strongly suggests that these film surfaces are hydrophobic. Hence it indicates the presence of PPO blocks on the top part of all films. This is in agreement with the angle dependent XPS findings. Furthermore, it also predicts that the surfaces of lowconcentration films (≤3 g/L) are chemically inhomogeneous unlike the high-concentration ones (≥3.6 g/L). Mechanism of Structural Transition. Figure 9 summarizes the structural transition of PEO-PPO-PEO films on hydrophilic Si substrates as a function of polymer concentration. As the third block contains only two EO monomers, the polymer can be considered as a diblock copolymer. It is known that asymmetric amphiphilic copolymers usually tend to form core−shell-like structures in selective solvent where the outer part likes the solvent. In the nonpolar toluene solution used here, the

Figure 7. Coverage of L1 and L2 layers obtained from XRR (solid symbol) and AFM (open symbol) analysis. In XRR, the coverage is calculated with respect to (w.r.t.) the bulk electron density (∼0.37 e/Å3) of PEO or PPO.

Figure 8 shows the plot of C 1s XPS spectra measured for the 3 and 4 g/L samples with electron emission angles of 20° and 90°. The fitting of all the C 1s spectra requires two components corresponding to the two different carbon environments. Assuming inelastic mean free path of electron to be ∼3 nm here,44 the probing depth for XPS is ∼9 nm at 90° and ∼3 nm at 20°. It can be observed from the figure that for both samples the C−C peak intensity corresponding to PPO increases at 20° 5813

dx.doi.org/10.1021/la5002607 | Langmuir 2014, 30, 5808−5816

Langmuir

Article

Figure 8. C 1s XPS spectra of 3 g/L (left panel) and 4 g/L (right panel) PEO-PPO-PEO films for electron emission angles 20° and 90°.

easily attached to a hydrophilic surface in contact. Therefore, when such a solution is spin-coated the attachment of this triblock copolymer is principally governed by interactions of PEO block with hydrophilic Si by building hydrogen bond between substrate surface silanol groups with the highly electronegative oxygen atom of PEO [e.g., −SiOH···(−OCH2CH−)]. Owing to strong hydrophilic−hydrophilic interactions in accordance with toluene evaporation at the time of spin coating, the aggregated and extended polymer chains try to cover up the underlying hydrophilic Si surface. Hence the film initiates with a highly dense ultrathin layer (shown as L1 in the schematic) of thickness ∼2 nm in the vicinity of the substrate. The first layer is thus probably formed by the PEO block in contact with the substrate and the PPO block toward air saturated by the solvent. Hence, the L1 layer is formed by one PEO-PPO bilayer. Once this layer is formed the strong hydrophilic interactions of substrate is suppressed and further growth is governed by the subsequent attachment of PPO to PPO and PEO to PEO in order to form pancake-like layered structures (shown as L2 in the schematic).47,48 Thus, the films are likely to be always terminated by the PPO block. This is confirmed by angle resolved XPS analysis and by WCA measurement. Therefore, both layers L1 and L2 are terminated by PPO block; consequently, no appreciable phase contrast is observed between A and B (shown in AFM images and schematic). Hence C is likely representing the hydrophilic PEO block. However, the coverage of the top layer is strongly dependent on the concentration of the mother solution. At low concentration it is only partial, while at higher concentration it becomes complete as shown in the schematic model of Figure 9. During block-by-block stacking it is highly possible that some of the PEO blocks remain exposed to air and give rise to the hydrophilic C regions shown in the AFM images and in Figure 9. Due to the opposite polarity of the exposed layer a clear phase contrast in AFM images is observed. In addition, the presence of one intermediate layer in every film which has nonuniform electron density of layer L2 entices us to believe that layer L2 is made of two bilayers, i.e., PPO-PEO (bottom) and PEO-PPO (top). Notably the thicknesses of bilayers constituting L1 and L2 are considered to be largely

Figure 9. Schematic presentation of possible structures of PEO-PPOPEO films (a) 1 g/L, (b) 2 g/L or 3g/L or 3.4 g/L, (c) 3.6 or 4 g/L, and (d) 4.5 g/L on Si surface.

hydrophobic PPO block is expected to protect the PEO block from contact with the solvent. The DLS study clearly shows that a core−shell structure is not achieved in this case. It is rather observed that big aggregates of elongated chains are formed instead. Such structures are well apart in dilute solution. It is wellknown that the PEO segments have strong affinity for hydrophilic substrate.45,46 As these chains are uncoiled, they would be 5814

dx.doi.org/10.1021/la5002607 | Langmuir 2014, 30, 5808−5816

Langmuir

Article

PPO-PEO triblock copolymer spin-coated ultrathin films with polymer concentration on the hydrophilic Si surface. At lower concentration (≤3.4 g/L) an incompletely covered layer is formed by the attachment of extended polymer chain layered-like structure on a fully covered ultrathin layer giving rise to a film with an overall two-layered structure from the coverage or density point of view. With further rise in concentration, the twolayered structure almost disappears and eventually one-layered films are formed. Each layer consists of one or two bilayers of PEO-PPO blocks comprising variable thickness and compactness through alternating stacking. Angle dependent XPS measurement, which reveals the out-of-plane chemical composition in the film, and WCA study confirm the presence of hydrophobic PPO and hydrophilic PEO blocks in the top and bottom parts of the films, respectively. The results demonstrate that the large PEO block of this kind of copolymer determines the adsorption behavior that prefers to reside on the hydrophilic Si surface, whereas small PPO blocks dislike the surface and as a result they are dangled in the air.

different in order to explain the fact that the L2 layer is more than twice as thick as L1. As previously mentioned, strong hydrophilic−hydrophilic interactions of PEO and the underlying silanol group of RCA cleaned Si are probably responsible for lowering the thickness of bilayer in L1 layer compared to that in L2 layer. The deformation of the polymer spherical micellar structure, which results in bigger and thinner hemimicelles, is also evidenced by Li et al. due to the presence of hydrophobic− hydrophobic interaction between substrate surface and polymer shell.46 In addition, owing to much lower compactness of bilayers in L2 compared to that in L1, it is plausible to believe that the bilayers are even more than twice as thick in the former layer than in the latter. However, an increase in concentration facilitates the in-plane growth of layer L2 that attains its maximum density value at a concentration of 4.5 g/L. At low concentration, the density and thickness of layer L2 are very low due to a very low coverage. Both these parameters increase when increasing the concentration of the polymer solution. This trend is maintained until the concentration reaches 3 g/L. Beyond this concentration, the thickness of the overall film decreases slightly and the contrast between L1 and L2 becomes very small. This is a clear indication that beyond this concentration the film becomes almost homogeneous. It also means that the coverage of layer L2 is close to that of L1 (shown in Figure 7). As there is no appreciable density contrast between PEO and PPO blocks, it is normal that the electron density of L2 should approach that of L1 assuming that the two layers have the same compactness. This is not exactly the case, as it is very likely that polymer chains have different packing in different bilayers due to the gradation in out-of-plane interactions. An unusual behavior, i.e., reduction of d2, is observed when the concentration increases from 3 g/L to 3.6 g/L. At low concentration when L2 is incomplete, moisture in the environment can be absorbed at L2 through the exposed edges of a large number of hydrophilic PEO blocks, though most of the top parts are covered by hydrophobic PPO blocks (Figure 9b). Hence, L2 remains in a swollen state with enhanced thickness, but with the increase of concentration the coverage of L2 becomes almost complete (Figure 9c), which in turn inhibits the moisture absorption or swelling by minimizing the exposed edges of PEO blocks, and accordingly a lower d2 is obtained in 3.6 g/L film. Once the film becomes nearly homogeneous at 3.6 g/L, a further increase of concentration is responsible for the conventional rise of d2. Furthermore, the entanglement of polymer chains in the L2 layer increases with concentration. This can explain the sudden reduction of thickness of layer L2 after the concentration of 3 g/L was reached. It should be noted here that no clear signal of entanglement is observed from the DLS study. We believe that the entanglement does not happen in the mother solution but may occur at increased concentration during the drying stage of the films. At even higher concentration the entanglement may become stronger during drying as expected, and therefore the polymer chains effectively form a compact homogeneous film with an appropriate uniform density. However, as the substrate is hydrophilic the PEO blocks of the chains prefer to be attached to the Si substrate, and as a result, the PPO blocks are closer to the top surface of the films. This layered-structure model is also supported by the XPS and WCA observations of PPO blocks toward the upper surface for all the films.

■

ASSOCIATED CONTENT

■

AUTHOR INFORMATION

S Supporting Information *

Normalized XRR data of 3.4 g/L and 3.6 g/L films. This material is available free of charge via the Internet at http://pubs.acs.org. Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was carried out within the framework of the CEFIPRA program for which we would like to acknowledge the financial support. Authors also acknowledge Goutam Sarkar for his help in acquiring XPS data.

■

REFERENCES

(1) Bita, I.; Yang, J. K. W.; Jung, Y. S.; Ross, C. A.; Thomas, E. L.; Berggren, K. K. Graphoepitaxy of Self-Assembled Block Copolymers on Two-Dimensional Periodic Patterned Templates. Science 2008, 361, 943−943. (2) Ruiz, R.; Kang, H.; Detcheverry, F. A.; Dobisz, E.; Kercher, D. S.; Albrecht, T. R.; de Pablo, J. J.; Nealey, P. F. Density Multiplication and Improved Lithography by Directed Block Copolymer Assembly. Science 2008, 361, 940−943. (3) Park, S. H.; Shin, D. O.; Kim, B. H.; Yoon, D. K.; Kim, K.; Lee, S. Y.; Oh, S.-H.; Choi, S.-W.; Jeon, S. C.; Kim, S. O. Block Copolymer Multiple Patterning Integrated with Conventional ArF Lithography. Soft Matter 2010, 6, 120−125. (4) Jeong, S.-J.; Moon, H.-S.; Shin, J.; Kim, B. H.; Shin, D. O.; Kim, J. Y.; Lee, Y.-H.; Kim, J. U.; Kim, S. O. One-Dimensional Metal Nanowire Assembly via Block Copolymer Soft Graphoepitaxy. Nano Lett. 2010, 10, 3900−3905. (5) Bang, J.; Jeong, U.; Ryu, D. Y.; Russell, T. P.; Hawker, C. J. Block Copolymer Nanolithography: Translation of Molecular Level Control to Nanoscale Patterns. Adv. Mater. 2009, 21, 4769−4792. (6) Park, S.-M.; Ravindran, P.; La, Y.-H.; Craig, G. S. W.; Ferrier, N. J.; Nealey, P. F. Combinatorial Generation and Replication-Directed Assembly of Complex and Varied Geometries with Thin Films of Diblock Copolymers. Langmuir 2007, 23, 9041−9045. (7) Wang, Q.; Nealey, P. F.; de Pablo, J. J. Simulations of the Morphology of Cylinder-Forming Asymmetric Diblock Copolymer Thin Films on Nanopatterned Substrates. Macromolecules 2003, 40, 1735−1744.

■

CONCLUSION We have performed DLS, XRR, AFM, XPS, and WCA measurements to investigate the structural transition of PEO5815

dx.doi.org/10.1021/la5002607 | Langmuir 2014, 30, 5808−5816

Langmuir

Article

(8) Detcheverry, F. A.; Nealey, P. F.; de Pablo, J. J. Directed Assembly of a Cylinder-Forming Diblock Copolymer: Topographic and Chemical Patterns. Macromolecules 2010, 43, 6495−6504. (9) Pavan, M. J.; Shenhar, R. Two-Dimensional Nanoparticle Organization Using Block Copolymer Thin Films As Templates. J. Mater. Chem. 2011, 21, 2028−2044. (10) Son, J. G.; Hannon, A. F.; Gotrik, K. W.; Alexander-Katz, A.; Ross, C. A. Hierarchical Nanostructures by Sequential Self-Assembly of Styrene-Dimethylsiloxane Block Copolymers of Different Periods. Adv. Mater. 2011, 23, 638−643. (11) Bergeron, V. Forces and Structure in Thin Liquid Soap Films. J. Phys.: Condens. Matter 1999, 11, R215−R242. (12) Israelachvili, J. N. Intermolecular and Surface Forces; Academic Press: London, 1992. (13) Barton, A. F. M. Handbook of Solubility Parameters and Other Cohesion Parameters; CRC Press: Boca Raton, FL, 1988. (14) Flory, P. J. Principles of Polymer Chemistry; Cornell University Press: Ithaca, NY, 1953. (15) Schmolka, I. R. In Polymer for Controlled Drug Delivery; Tarcha, P. J., Ed.; CRC: Boston, 1992. (16) Yu, S.; He, C.; Ding, J.; Cheng, Y.; Song, W.; Zhuanga, X.; Chena, X. pH and Reduction Dual Responsive Polyurethane Triblock Copolymers for Efficient Intracellular Drug Delivery. Soft Matter 2013, 9, 2641−2645. (17) Cheng, C.-Y.; Wang, J.-Y.; Kausik, R.; Lee, K. Y. C.; Han, S. Nature of Interactions between PEO-PPO-PEO Triblock Copolymers and Lipid Membranes: (II) Role of Hydration Dynamics Revealed by Dynamic Nuclear Polarization. Biomacromolecules 2012, 13, 2624− 2637. (18) Ghandi, S. K. VLSI Fabrication Principles, Silicon and Gallium Arsenide; John Wiley and Sons: New York, 1983. (19) Li, Y.; Rojas, O. J.; Hinestroza, J. P. Boundary Lubrication of PEOPPO-PEO Triblock Copolymer Physisorbed on Polypropylene, Polyethylene, and Cellulose Surfaces. Ind. Eng. Chem. Res. 2012, 51, 2935− 2944. (20) Tang, C.; Lennon, E. M.; Fredrickson, G. H.; Kramer, E. J.; Hawker, C. J. Evolution of Block Copolymer Lithography to Highly Ordered Square Arrays. Science 2008, 362, 429−436. (21) Kang, N.-G.; Cho, B.; Kang, B.-G.; Song, S.; Lee, T.; Lee, J.-S. Structural and Electrical Characterization of a Block Copolymer-Based Unipolar Nonvolatile Memory Device. Adv. Mater. 2012, 24, 425. (22) Longenberger, L.; Mills, G. Formation of Metal Particles in Aqueous Solutions by Reactions of Metal Complexes with Polymers. J. Phys. Chem. 1995, 99, 475−478. (23) Warshawsky, A.; Kalir, R.; Deshe, A.; Berkovitz, H.; Patchornik, A. Pseudocrown Ethers. 1. Synthesis and Complexation with Transition Metal Anions. J. Am. Chem. Soc. 1979, 101, 4249−4258. (24) Sakai, T.; Alexandridis, P. Single-Step Synthesis and Stabilization of Metal Nanoparticles in Aqueous Pluronic Block Copolymer Solutions at Ambient Temperature. Langmuir 2004, 20, 8426−8430. (25) Sakai, T.; Alexandridis, P. Mechanism of Gold Metal Ion Reduction, Nanoparticle Growth and Size Control in Aqueous Amphiphilic Block Copolymer Solutions at Ambient Conditions. J. Phys. Chem. B 2005, 109, 7766−7777. (26) Sakai, T.; Alexandridis, P. Spontaneous Formation of Gold Nanoparticles in Poly(ethylene oxide)−Poly(propylene oxide) Solutions: Solvent Quality and Polymer Structure Effects. Langmuir 2005, 21, 8019−8025. (27) Sakai, T.; Alexandridis, P. Size- and Shape-Controlled Synthesis of Colloidal Gold through Autoreduction of the Auric Cation by Poly(Ethylene Oxide)−poly(Propylene Oxide) Block Copolymers in Aqueous Solutions at Ambient Conditions. Nanotechnology 2005, 16, S384−S393. (28) Wang, L.; Chen, X.; Zhao, J.; Sui, Z.; Zhuang, W.; Xu, L.; Yang, C. Synthesis and Association of Ag(0) Nanoparticles in Aqueous Pluronic F127 Triblock Copolymer Solutions. Colloids Surf., A 2005, 257−258, 235−239.

(29) Alexandridis, P.; Tsianou, M. Block Copolymer-Directed Metal Nanoparticle Morphogenesis and Organization. Eur. Polym. J. 2011, 47, 569−583. (30) Maidul Islam, A. K. M.; Mukherjee, S.; Nannarone, S.; Mukherjee, M. Morphology and Chemical Properties of Silver-Triblock Copolymer Nanocomposite Thin Films. Mater. Chem. Phys. 2013, 144, 284−293. (31) Malmsten, M.; Tiberg, F. Adsorption of Ethyl(Hydroxyethyl) Cellulose at Polystyrene. Langmuir 1993, 9, 1098−1103. (32) Liu, X.; Wu, D.; Turgman-Cohen, S.; Genzer, J.; Theyson, T. W.; Rojas, O. J. Adsorption of a Nonionic Symmetric Triblock Copolymer on Surfaces with Different Hydrophobicity. Langmuir 2010, 26, 9565− 9574. (33) Fu, Z.; Santore, M. M. Kinetics of Competitive Adsorption of PEO Chains with Different Molecular Weights. Macromolecules 1998, 35, 7014−7022. (34) Brandani, P.; Stroeve, P. Adsorption and Desorption of PEO− PPO−PEO Triblock Copolymers on a Self-Assembled Hydrophobic Surface. Macromolecules 2003, 36, 9492−9501. (35) Daillant, J.; Gibaud, A. X-ray and Neutron Reflectivity: Principles and Applications; Springer: Berlin, 2009. (36) Gibaud, A.; Chebil, M. S.; Beuvier, T. Surface Science Techniques; Bracco, G., Holst, B., Eds.; Springer Verlag: Berlin, 2013; Chapter 7. (37) Bal, J. K.; Kundu, S.; Hazra, S. Hydrophobic to Hydrophilic Transition of HF-Treated Si Surface during Langmuir−blodgett Film Deposition. Chem. Phys. Lett. 2010, 500, 90−95. (38) Horcas, I.; Fernndez, R.; Gwez-Rodrguez, J. M.; Colchero, J.; Gomez-Herrero, J.; Baro, A. M. WSXM: a Software for Scanning Probe Microscopy and a Tool for Nanotechnology. Rev. Sci. Instrum. 2007, 78, 014105−1. (39) Maidul Islam, A. K. M.; Mukherjee, M. Effect of Temperature in Synthesis of Silver Nanoparticles in Triblock Copolymer Micellar Solution. J. Exp. Nanosci. 2011, 5, 596−611. (40) Soni, S. S.; Brotons, G.; Bellour, M.; Narayanan, T.; Gibaud, A. Quantitative SAXS Analysis of the P123/Water/Ethanol Ternary Phase Diagram. J. Phys. Chem. B 2006, 110, 15157−15165. (41) Fall, S.; Pattier, B.; Benyayia, L.; Gibaud, A. Binary Phase Diagram of Water/Brij58 Studied by SAXS. Acta Phys. Polym., A 2012, 121, 428− 436. (42) Esnault, C.; Chenais, B.; Casse, N.; Delorme, N.; Louarn, G.; Pilard, J. F. Electrochemically Modified Carbon and Chromium Surfaces for AFM Imaging of Double-Strand DNA Interaction with Transposase Protein. ChemPhysChem 2013, 14, 338−345. (43) Noy, A.; Sanders, C. H.; Vezenov, D. V.; Wong, S. S.; Lieber, C. M. Chemically-Sensitive Imaging in Tapping Mode by Chemical Force Microscopy: Relationship between Phase Lag and Adhesion. Langmuir 1998, 14, 1508−1511. (44) Roberts, R. F.; Allara, D. L.; Pryde, C. A.; Buchanan, D. N. E.; Hobbins, N. D. Mean Free Path for Inelastic Scattering of 1.2 keV Electrons in Thin Poly(methylmethacrylate) Films. Surf. Interface Anal. 1979, 2, 5−10. (45) Elisseeva, O. V.; Besseling, N. A. M.; Koopal, L. K.; Cohen Stuart, M. A. Influence of NaCl on the Behavior of PEO−PPO−PEO Triblock Copolymers in Solution, at Interfaces, and in Asymmetric Liquid Films. Langmuir 2005, 21, 4954−4963. (46) Li, Y.; Liu, H.; Song, J.; Rojas, O. J.; Hinestroza, J. P. Adsorption and Association of a Symmetric PEO-PPO-PEO Triblock Copolymer on Polypropylene, Polyethylene, and Cellulose Surfaces. ACS Appl. Mater. Interfaces 2011, 3, 2349−2357. (47) Schroën, C. G. P. H.; Cohen Stuart, M. A.; van der Voort Maarschalk, K.; van der Padt, A.; Van’t Riet, K. Influence of Preadsorbed Block Copolymers on Protein Adsorption: Surface Properties, Layer Thickness, and Surface Coverage. Langmuir 1995, 11, 3068−3074. (48) Schar, J. A.; Obey, T. M.; Cosgrove, T. Adsorption Studies of Polyethers: Part II: Adsorption onto Hydrophilic Surfaces. Colloids Surf., A: Physicochem. Eng. Aspects 1999, 150, 15−23.

5816

dx.doi.org/10.1021/la5002607 | Langmuir 2014, 30, 5808−5816