Two-Dimensional Phase Separation of a Poly(methyl methacrylate

Jul 9, 2013 - Decomposition Mechanism. Go Sato,. †. Shotaro Nishitsuji, and Jiro Kumaki*. Department of Polymer Science and Engineering, Graduate ...
6 downloads 0 Views 5MB Size
Download PDF
Article pubs.acs.org/JPCB

Two-Dimensional Phase Separation of a Poly(methyl methacrylate)/ Poly(L‑lactide) Mixed Langmuir Monolayer via a Spinodal Decomposition Mechanism Go Sato,† Shotaro Nishitsuji, and Jiro Kumaki* Department of Polymer Science and Engineering, Graduate School of Science and Engineering, Yamagata University, Yonezawa, Yamagata 992-8510, Japan ABSTRACT: We have found the first evidence that a polymer blend Langmuir monolayer can phase-separate via spinodal decomposition (SD) mechanism. The system was a poly(methyl methacrylate)/poly(L-lactide) mixture. It phaseseparated immediately after compression on a water surface and formed a spinodal-like morphology, as observed by atomic force microscopy (AFM). The fast Fourier transform of the AFM images showed a clear spinodal ring with a maximum intensity at a wavenumber of qm. At the small quench depth at a surface pressure of 2 mN/m, qm did not change, but the concentration difference between the domains (ΔΦ) grew with time, corresponding to the early-stage SD. At larger quench depths at 4 and 5 mN/m, qm significantly decreased, but ΔΦ was constant with time; this behavior corresponded well to the latestage SD. Thus, the 2D phase separation in the Langmuir monolayer was basically explained by the SD mechanism well-known in 3D systems. In the late stage SD of the monolayer, qm scaled with time much faster than that expected by theories for the 2D state. Phase separation via a SD mechanism is a promising new way to control the lateral morphology of Langmuir monolayers, one of the main issues in nanotechnology that remains difficult to attain even today.

1. INTRODUCTION As has been reported by many authors over the past decades, the Langmuir−Blodgett (LB) technique provides a very powerful tool to construct well-ordered ultrathin films.1,2 Because of the amphiphilic nature of the compounds usually used in LB films, the structure perpendicular to the water surface can be strictly designed, while the structure parallel to the water surface is still difficult to control. One of the attempts to control the lateral structures of the polymer monolayer was to use block copolymers. Lennox and Eisenberg first demonstrated that various regular morphologies − sphere, rod, and planar disk − were obtained in the spread monolayers using block copolymers.3−9 Although they used various block copolymers, they all had polystyrene as an essential block component and other hydrophilic groups, such as poly(vinylpyridinum salt), poly(butyl methacrylate), and poly(dimethylsiloxane), as another block. Polystyrene does not contain hydrophilic groups and does not spread on the water surface as a monolayer. Thus, although regular morphologies were observed, these were not formed through a true 2D phase separation. Kumaki et al. first attained a true 2D phase separation by using a poly(methyl methacrylate)-b-poly(octadecyl methacrylate) diblock copolymer, both blocks of which spread on the water surface as a monolayer.10 Aoki and Ito further studied chain packing in 2D phase separation of block copolymers and their mixtures with homopolymers by fluorescent label technique using scanning near-field optical microscopy (SNOM).11−13 © 2013 American Chemical Society

Another way to control morphology is to use polymer blend systems that possess a phase diagram. If the polymer blend system is quenched from a miscible state to an immiscible state, then it phase-separates via nucleation and growth (NG) or spinodal decomposition (SD) mechanisms, depending on whether the blend is quenched in a thermodaynamically metastable or instable region. Phase separations of 3D blend systems, especially via SD, have been extensively studied to control and design the morphology and physical properties of the blends;14−17 however, to the best of our knowledge, SD phase separation in polymer blend monolayers has not yet been reported. Some of the reports on the morphologies of polymer blend monolayers observed by AFM are summarized in refs 18−23. In this study, we found the first evidence that a poly(methyl methacrylate) (PMMA)/poly(L-lactide) (PLLA) mixed monolayer phase-separated via SD mechanism. The phase separation proceeded in a quite similar manner to well-known early-stage and late-stage SD processes depending on the quench depths where the monolayer phase-separated. The power law of domain growth at late stage SD is compared with those expected from theory. Received: March 31, 2013 Revised: June 25, 2013 Published: July 9, 2013 9067

dx.doi.org/10.1021/jp403195g | J. Phys. Chem. B 2013, 117, 9067−9072

The Journal of Physical Chemistry B

Article

2. EXPERIMENTAL SECTION 2.1. Materials. A PMMA with a number-average molecular weight (Mn) of 6.00 × 103 and a molecular weight distribution (Mw/Mn) of 1.08 was purchased from Showa Denko (Tokyo, Japan). A PLLA with Mn of 6.00 × 103 and Mw/Mn of 1.18 was purchased from Polymer Source (Montreal, Canada). Highly purified chloroform (Infinity Pure, Wako Chemicals, Osaka, Japan) was used as the solvent for the spreading solutions without further purification. Water was purified by a Milli-Q system and used as the subphase for the Langmuir−Blodgett (LB) investigations. 2.2. Surface Pressure−Area (π−A) Isotherm Measurements and Langmuir−Blodgett Film Preparations for AFM. The π−A isotherms of a PMMA, PLLA, and their mixtures (PMMA/PLLA = 25/75, 50/50, and 75/25 in weight fraction) were measured as follows. A PMMA, PLA, and their mixture solution in chloroform having a total polymer concentration from 7.4 × 10−5 to 1.19 × 10−4 g/mL were spread on a water surface at 23 °C in a commercial LB trough with an area of 37.5 × 10 cm2 and an effective moving barrier length of 10 cm (3-22YG3, USI, Fukuoka, Japan). The π−A isotherms were measured at a constant compression rate with a moving barrier speed of 0.5 mm/s using filter paper as the Wilhelmy plate. A monolayer was deposited onto a piece of freshly cleaved mica by pulling it out of the water at a rate of 4.2 mm/min while compressing the monolayer at a constant pressure (the vertical dipping method) using a larger LB trough with an area of 60 × 15 cm2 and an effective moving barrier length of 15 cm (EDS-300AS, USI). The monolayers were compressed at a particular surface pressure and deposited onto a piece of freshly cleaved mica once per hour for 2 and 4 mN/ m and every 30 min for 5 mN/m to follow the kinetics of phase separation of the mixed monolayers at constant surface pressures of 2, 4, and 5 mN/m. 2.3. AFM Observations. After the deposited monolayers were dried in vacuo, they were observed by a commercial AFM (NanoScope IIIa or IIId/multimode AFM unit, Bruker AXS, Santa Barbara, CA) with standard silicon cantilevers (NCH, Bruker AXS) in air in the tapping mode. The typical settings of the AFM observations were as follows: a drive amplitude of 1.0 to 1.5 V, a set point of 0.70 to 0.95 V, and a scan rate of 1.2 to 2.5 Hz. The AFM images obtained are presented without any image processing except flattening. The 2D fast Fourier transform (FFT) of the AFM images and the circular average of the resultant spinodal rings to obtain a power spectrum were done by image analysis software (Scion Image, Scion Corporation, Frederick, MD).

Figure 1. π−A isotherms of monolayers of PMMA, PLLA, and their mixtures (PMMA/PLLA = 75/25, 50/50, and 25/75 by weight fraction) on a water surface.

Figure 2. Plots of the limiting area (π = 0 mN/m) and areas at 8 and 18 mN/m derived from the π−A isotherms in Figure 1 as a function of the PMMA weight content of the monolayers. The areas obey an additive rule (dotted lines), thus indicating both the PMMA and PLLA spread on the water surface as a monolayer.

transition derived from the π−A isotherms in Figure 1 are plotted as a function of the PMMA weight content. At all surface pressures, the areas of the PMMA/PLLA mixtures linearly varied with the PMMA content and agreed with an additive rule of the areas of the pure PMMA and PLLA monolayers (dotted lines), indicating that both the PMMA and PLLA spread on the water surface as a monolayer at all surface pressures studied. Figure 3 shows tapping-mode AFM height images of the PMMA/PLLA mixed monolayers (25/75, 50/50, and 75/25 wt/wt) deposited on mica at 1, 5, 10, and 12 mN/m after being compressed at a rate of 0.5 mm/s. The PMMA/PLLA = 25/75 mixture did not show a clear phase separation at 1 mN/m, but at 5 mN/m, isolated island-like domains with a diverse diameter distribution appeared in the surrounding matrix phase. The concentration dependence of the phase-separation structures indicated that the higher (brighter) domains correspond to a PMMA-rich phase, whereas the lower (darker) domains correspond to a PLLA-rich phase. At a surface pressure higher than the onset of the crystallization transition of the PLLA at 9

3. RESULTS AND DISCUSSION 3.1. π−A Isotherms and AFM of Monolayers of PMMA/PLLA Mixtures. Figure 1 shows π−A isotherms of PMMA, PLLA, and their mixtures (PMMA/PLLA = 75/25, 50/50, and 25/75 by weight fraction). The PMMA monolayer (blue line) shows a condensed-type π−A isotherm with a small unidentified shoulder at ∼15 mN/m. The PLLA monolayer (red line) shows an expanded-type π−A isotherm with a clear transition at ∼9 mN/m, which corresponds to its crystallization, as previously reported.24 All PMMA/PLLA mixtures showed intermediate π−A isotherms between those of the PMMA and PLLA, depending on the compositions. In Figure 2, the limiting areas (extrapolated to π = 0 mN/m) and the areas before (π = 8 mN/m) and after (π = 18 mN/m) the 9068

dx.doi.org/10.1021/jp403195g | J. Phys. Chem. B 2013, 117, 9067−9072

The Journal of Physical Chemistry B

Article

the PLLA component to crystallize as lamella (10 mN/m), which grew to be bundles of the lamella at 12 mN/m. The PMMA-rich phases had some residual phase separation at 10 mN/m, but after most of the PLLA crystallized and had been removed from the phases the residual PMMA became a homogeneous monolayer without a specific phase separation at 12 mN/m. On the basis of similar experiments, we determined a phase diagram of the PMMA/PLLA mixed monolayer, as shown in Figure 4. The y axis indicates the surface pressure. The points

Figure 3. Tapping-mode AFM height images of the monolayers of PMMA/PLLA mixtures (25/75, 50/50, and 75/25 in weight fraction) deposited on mica at a surface pressure of 1, 5, 10, and 12 mN/m. The inserts are FFT of the corresponding AFM images, which show a typical spinodal ring. Z range: 2 nm. The fibrils shown at the surface pressure higher than 10 mN/m are crystallized PLLA lamella. Figure 4. Phase diagram of the PMMA/PLLA mixed monolayer determined from the π−A isotherms and AFM images of the PMMA/ PLLA blend monolayers. The phase boundaries are shown simply as a guide to the eye, and were not precisely determined due to the limited number of data points.

mN/m, the PLLA started to crystallize to form lamella crystals that looked like long fibrils, as shown at 10 mN/m; further compression to 12 mN/m resulted in a development of the lamella crystals to be bundles of the fibrils. After the removal of the crystalline PLLA, the residual amorphous PMMA phase became a continuous film without a specific phase separation. The PMMA/PLLA = 50/50 and 75/25 mixtures started a vague phase separation at 1 mN/m and developed contrast at 5 mN/m. In contrast with the irregular sea-island structure of the PMMA/PLLA = 25/75 mixture, the 50/50 and 75/25 mixtures had homogeneous, somewhat interconnected, structures with a definite size. The former (PMMA/PNA = 25/75) and latter (50/50 and 75/25) structures are reminiscent of the two major phaseseparation mechanisms well-known in 3D blend systems, namely, the NG and the SD mechanisms. In a 3D polymer blend system quenched from a miscible state into a metastable immiscible region, a phase separation proceeds via the NG mechanism. Being in the metastable region, a small composition fluctuation diminishes, but if a nuclei with a sufficiently large size forms, then it grows. The composition of the nuclei is known to be close to the equilibrium composition of the blend. Thus, the resultant morphology formed by the NG mechanism is a sea-island structure with a diverse size distribution. If a blend is quenched into an unstable regime, then the blend phase-separates via a SD mechanism. In the unstable state, concentration fluctuations with a wavelength larger than a critical value grow spontaneously, with a maximum grow rate at a certain wavelength. As the result, the phaseseparated structure via SD mechanism shows specific interconnected structure with a predominant wavelength. The FFT of the morphology shows a doughnut-like pattern known as “a spinodal ring”. In fact, FFTs of the AFM images of the 50/ 50 and 75/25 mixtures shown as inserts in Figure 3 showed a clear spinodal ring, indicating that the phase-separation structures possessed a dominant concentration fluctuation with a specific wavelength. Further compression of the 50/50 and 75/25 mixtures beyond the crystallization transition caused

for the NG, SD, and PLA crystallization were determined based on the morphology of the PMMA/PLLA mixtures. The PMMA/PLLA = 25/75 mixture at 1 mN/m in Figure 3 did not show a phase-separation structure, but we confirmed that it phase-separated after being compressed at the surface pressure for longer times. Phase separation was confirmed for all compositions studied at 1 mN/m after compression for a long time. As shown in Figure 4, the phase diagram represented by the surface pressure versus composition resembled a normal phase diagram with a lower critical solution temperature (LCST) for 3D blends represented by the temperature versus the composition. The phase separations occurred by SD for the central compositions, PMMA/PLLA = 50/50 and 75/25, while NG was observed at the off-critical compositions, 10/90, 25/75, and 90/10 mixtures. The PLLA crystallization was observed at 10 mN/m for all compositions except the PLLA minor composition of PMMA/PLLA = 90/10. 3.2. Time Lapse of Phase-Separation Structures of a PMMA/PLLA = 50/50 Mixed Monolayers. Figure 5 shows AFM height images of monolayers of the PMMA/PLLA = 50/ 50 mixtures deposited on mica after compression at 2, 4, and 5 mN/m for the various times indicated in the images. At a small quench depth at 2 mN/m, a specific spinodal-like phase separation with a homogeneous size appeared at 0 h, which increased the height contrast, but the domain size did not grow with time. FFT patterns of the AFM images showed a typical spinodal ring that did not change in size but whose contrast increased with time. The height profile shows that the height difference of the two domains gradually increased from 0.28 nm at 0 h to 0.49 nm at 4 h. The height of the monolayer depends on the composition of the domain; therefore, the increase in the height difference corresponds to the increase in the composition difference between the two domains. In contrast, 9069

dx.doi.org/10.1021/jp403195g | J. Phys. Chem. B 2013, 117, 9067−9072

The Journal of Physical Chemistry B

Article

Figure 5. Time lapses of AFM images of a PMMA/PLLA = 50/50 blend monolayer during phase separation at constant surface pressures. The monolayers were deposited on mica after being compressed at 2, 4, and 5 mN/m for a specific time indicated in the images. FFT of the corresponding AFM images are shown as inserts. Height profiles along the yellow lines in the images are also shown. Z range: 2 nm.

at larger quench depths at 4 and 5 mN/m, typical spinodal-like morphologies appeared at 0 mN/m, the sizes of which grew with time; however, the height differences were almost constant. Thus at the larger quench depths, the domains grow with time, but the compositions of the domains are constant, that is, close to the equilibrium compositions at the surface pressures. The FFT patterns of the AFM images showed again a typical spinodal ring, which shifted to smaller angle and increased in contrast with time. Figure 6 shows the power spectrum obtained by a circular average of the FFT AFM images in Figure 5. The arrows in the spectrum indicate qm, the wavenumber at the maximum intensity. At the small quench depth at 2 mN/m, qm did not shift, but the intensity at the maximum slightly increased. At the larger quench depths at 4 and 5 mN/m, qm shifted to a smaller angle and also the intensity at the maximum increased. Figure 7 shows the height difference between the domains in Figure 5 as a function of time. As previously mentioned, the height different increased at the small quench depth (2 mN/m), but it did not change at the larger quench depths (4 and 5 mN/m). On the basis of the previous studies in 3D polymer blend systems, the SD is known to be classified into three stages, that is, the early, intermediate, and late stages.15 In the early-stage SD, the wavelength λ (= 2π/qm) of the dominant mode of the phase-separating structure is essentially independent of time t, whereas the concentration fluctuations, Δϕ(t), grow with time. In the intermediate stage, both λ and Δϕ(t) grow with time. In

Figure 6. Power spectrum obtained by circular average of FFT AFM images in Figure 5. Arrows indicate qm, the wavenumber at which the scattering intensity shows a maximum.

the late stage, λ increases with time, while Δϕ(t) already saturates to its equilibrium value Δϕe, which is determined by the coexistence curve of the phase diagram and thus does not change any more. The 2D phase-separation behavior of the PMMA/PLLA monolayer observed here closely resembles the well-known 3D SD mechanism. The phase separations at the small quench depth (2 mN/m) and at the large quench depths (4 and 5 mN/m) are assigned to the early- and late-stage SD, respectively. The present 2D phase separation can be explained 9070

dx.doi.org/10.1021/jp403195g | J. Phys. Chem. B 2013, 117, 9067−9072

The Journal of Physical Chemistry B

Article

not precise; however, we could conclude that it was significantly faster than those predicted by the theory for 2D states. Therefore, an additional driving force need to be considered to explain the fast growth rate in the monolayer system. Wiltzius and Cumming investigated spinodal decomposition of a polyisoprene and poly(ethylene-co-propylene) blend sandwiched between two quartz plates by light scattering and optical microscopy28 and found the growth law near the quartz plate (qm ∼ t−3/2) was much faster than that in the bulk (qm ∼ t−1). They explained that the faster phase separation at the quartz interface was due to long-range van der Waals force caused by the quartz surface on the interfaces of the coexisting phases, which lead to wetting and acted as an additional driving force to hasten the phase separation. In our case, the PMMA/ PLLA blend monolayer phase-separated on the water surface. We also need to consider the strong interaction between the phase-separating domains and the water surface. As mentioned, the PMMA forms a condensed-type monolayer, while the PLLA forms an expanded-type monolayer on the water surface (Figure 1). Thus, when the monolayer phase-separated into a PMMA-rich and a PLLA-rich phase, the former should condense but the latter should expand on the water surface. This condensation and expansion effect should act as an additional driving force and accelerate the phase-separation rate. This may explain the large power law observed in the PMMA/PLLA monolayer system.

Figure 7. Time dependence of the height difference of the domains in the PMMA/PLLA = 50/50 blend monolayer in Figure 5.

within the scope of the well-known SD mechanism. To the best of our knowledge, this is the first observation of 2D SD in a polymer blend monolayer. We want to emphasize that this is a promising new way to control the lateral morphology of the Langmuir monolayer, something that is still very difficult to attain using a simple blend system. 3.3. Time Scaling of qm. Figure 8 shows the time dependence of qm at 2, 4, and 5 mN/m. As previously

4. CONCLUDING REMARKS We studied a PMMA/PLLA mixed monolayer by AFM and π− A isotherm measurement and found that it phase-separated by the SD mechanism. The monolayer started to phase-separate immediately after compression on the water surface. The phaseseparation structure was similar to a 2D intersection of a typical interconnected SD structure with a homogeneous size. In fact, the FFT of the AFM images clearly showed a typical spinodal ring with a maximum intensity at a wavenumber, qm. For the PMMA/PLLA = 50/50 monolayer, at a small quench depth at a surface pressure of 2 mN/m, the phase separation proceeded with a constant qm but the composition difference between the domains grew with time. The behavior is similar to the early stage SD. At larger quench depth at 4 and 5 mN/m, the qm increased, but the composition difference was constant during the phase separation; this behavior resembled the late stage SD mechanism. The phase-separation behavior of the PMMA/ PLLA monolayer systems can be understood within the framework of the well-known SD mechanism. To the best of our knowledge, this is the first observation of SD in polymer blend monolayer systems. In the late stage, the qm scaled with time much faster than that expected for 2D systems by theories. We propose that an additional driving force due to a difference of the spreading behavior of the PMMA (condense-type monolayer) and PLLA (expanded-type monolayer) may explain the acceleration of the phase-separation rate. The control of the lateral morphology of Langmuir monolayers is one of the main issues of nanotechnology. We emphasize that the use of SD is a promising new way to control the lateral morphology to build up sophisticated LB films for use in nanotechnology.

Figure 8. Time dependence of qm, derived from Figure 6.

mentioned, qm did not change at 2 mN/m, corresponding to the early stage SD. qm decreased significantly at 4 mN/m and much faster at the larger quench depth at 5 mN/m, both corresponding to the late stage SD. The initial qm should depend on the quench depth, and the larger qm (smaller λ) is expected for the larger quench depth. The initial qm at 2 mN/m was smallest, but those at 4 and 5 mN/m were almost the same, but from the rapid decrease in the qm, we expect the largest initial qm (smallest λ) at 5 mN/m, which could not be detected under our experimental conditions. For the late stage SD, qm decreased with the power law dependence, qm ∼ t−1.6 at 4 mN/ m and qm ∼ t−1.2 at 5 mN/m after a long time, respectively. On the basis of the previous studies of SD in polymer blend systems, in the late stage SD, qm obeys the power law: qm ∼ tα. In 3D states, α = −1/3 was predicted for fluid mixtures based on the diffusion and coalescence of clusters.25 In the viscous hydrodynamic regime, Siggia predicted α = −1 taking into account the hydrodynamic effect.26 San Miguel et al. studied spinodal decomposition in 2D states, then proposed α = −1/3 for a diffusion-dominant process and α = −1/2 for a hydrodynamics-dominated process.27 Because of the limited number of the experimental points available in this experiment, the derived power law for the PMMA/PLLA monolayer was

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. 9071

dx.doi.org/10.1021/jp403195g | J. Phys. Chem. B 2013, 117, 9067−9072

The Journal of Physical Chemistry B

Article

Present Address

(19) Yamamoto, S.; Tsujii, Y.; Fukuda, T. Characteristic PhaseSeparated Monolayer Structure Observed for Blends of Rodlike and Flexible Polymers. Polymer 2001, 42, 2007−2013. (20) Ohkita, M.; Higuchi, M.; Kawaguchi, M. AFM Studies on LB Films of Poly(n-hexyl isocyanate), Poly(vinyl acetate), and Their Binary Mixtures. J. Colloid Interface Sci. 2005, 292, 300−303. (21) Kumaki, J.; Kawauchi, T.; Okoshi, K.; Kusanagi, H.; Yashima, E. Supramolecular Helical Structure of the Sterocomplex Composed of Complementary Isotactic and Syndiotactic Poly(methl methacrylates)s as Reviealed by Atomic Force Microscopy. Angew. Chem., Int. Ed. 2007, 46, 5341−5351. (22) Aiba, N.; Sasaki, Y.; Kumaki, J. Strong Compression Rate Dependence of Phase Separation and Stereocomplexation between Isotactic and Syndiotactic Poly(methyl methacrylates)s in a Langmuir Monolayer Observed by Atomic Force Microscopy. Langmuir 2010, 26, 12703−12708. (23) Sasaki, Y.; Aiba, N.; Hashimoto, H.; Kumaki, J. Reversible Hierarchical Phase Separation of a Poly(methyl methacrylate) and Poly(n-nonyl acrylate) Blend in a Langmuir Monolayer. Macromolecules 2010, 43, 9077−9086 References on the miscibility of polymer blend monolayers are listed in this reference. (24) Klass, J. M.; Lennex, R. B.; Brown, G. R. Enantiomeric Polylactides at the Air−Water Interface: π−A Isotherms and PMIRRAS Studies of Enantiomers and Their Blend. Langmuir 2003, 19, 333−340. (25) Binder, K.; Stauffer, D. Theory for the Slowing Down of the Relaxation and Spinodal Decomposition of Binary Mixtures. Phys. Rev. Lett. 1974, 33, 1006−1009. (26) Siggia, E. D. Late Stages of Spinodal Decomposition in Binary Mixtures. Phys. Rev. A 1979, 20, 595−605. (27) San Miguel, M.; Grant, M.; Gunton, J. D. Phase Separation in Two-Dimensional Binary Fluids. Phys. Rev. A 1985, 31, 1001−1005. (28) Wiltzius, P.; Cumming, A. Domain Growth and Wetting in Polymer Mixtures. Phys. Rev. Lett. 1991, 66, 3000−3003.

†

Go Sato: Toray Battery Separator Film Co., Ltd., 1190−13 Iguchi, Nasu-Shiobara, Tochigi 329−2763, Japan.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by a Grant-in-Aid for Scientific Research on Innovative Areas “Molecular Soft-Interface Science” (20106009), Scientific Research (B) (21350059, 24350113), and Challenging Exploratory Research (23655208, 24655091), Grant-in-Aid for Young Scientists (B) (25810069) from the Ministry of Education, Culture, Sports, Science, and Technology, Japan.

■

REFERENCES

(1) Gaines, G. L. Insoluble Monolayers at Liquid-Gas Interfaces; Interscience: New York, 1966. (2) Ulman, A. An Introduction to Ultrathin Organic Films from Langmuir-Blodgett to Self-Assembly; Academic Press: New York, 1991. (3) Zhu, J.; Eisenberg, A.; Lennox, R. B. Interfacial Behavior of Block Polyelectrolytes. 1. Evidence for Novel Surface Micelle Formation. J. Am. Chem. Soc. 1991, 113, 5583−5588. (4) Zhu, J.; Lennox, R. B.; Eisenberg, A. Interfacial Behavior of Block Polyelectrolytes. 2. Aggregation Numbers of Surface Micelles. Langmuir 1991, 7, 1579−1584. (5) Zhu, J.; Lennox, R. B.; Eisenberg, A. Interfacial Behavior of Block Polyelectrolytes. 4. Polymorphism of (Quasi) 2-Dimensional Micelles. J. Phys. Chem. 1992, 96, 4727−4730. (6) Zhu, J.; Eisenberg, A.; Lennox, R. B. Interfacial Behavior of Block Polyelectrolytes. 6. Properties of Surface Micelles as a Function of R and X in P(S260-b-VP240/RX). Macromolecules 1992, 25, 6556−6562. (7) Zhu, J.; Eisenberg, A.; Lennox, R. B. Interfacial Behavior of Block Polyelectrolytes. 5. Effect of Varying Block Lengths on the Properties of Surface Micelles. Macromolecules 1992, 25, 6547−6555. (8) Li, S.; Hanley, S.; Khan, I.; Varshney, S. K.; Eisenberg, A.; Lennox, R. B. Surface Micelle Formation at the Air-Water-Interface from Nonionic Diblock Copolymers. Langmuir 1993, 9, 2243−2246. (9) Cox, J. K.; Eisenberg, A.; Lennox, R. B. Patterned Surfaces via Self-Assembly. Curr. Opin. Colloid Interface Sci. 1999, 4, 52−59. (10) Kumaki, J.; Hashimoto, T. Two-Dimensional Microphase Separation of a Block Copolymer in a Langmuir-Blodgett Film. J. Am. Chem. Soc. 1998, 120, 423−424. (11) Aoki, H.; Kunai, Y.; Ito, S.; Yamada, H.; Matsushige, K. TwoDimensional Phase Separation of Block Copolymer and Homopolymer Blend Studied by Scanning Near-Field Optical Microscopy. Appl. Surf. Sci. 2002, 188, 534−538. (12) Sekine, R.; Aoki, H.; Ito, S. Conformation of Single Block Copolymer Chain in Two-Dimensional Microphase-Separated Structure Studied by Scanning Near-Field Optical Microscopy. J. Phys. Chem. B 2009, 113, 7095−7100. (13) Tamai, Y.; Sekine, R.; Aoki, H.; Ito, S. Conformation of Single Homopolymer Chain in Microphase-Separated Block Copolymer Monolayer Studied by Scanning Near-Field Optical Microscopy. Macromolecules 2009, 42, 4224−4229. (14) Furukawa, H. A Dynamic Scaling Assumption for Phase Separation. Adv. Phys. 1985, 34, 703−750. (15) Hashimoto, T. Dynamics in Spinodal Decomposition of Polymer Mixtures. Phase Transitions 1988, 12, 47−119. (16) Bray, A. J. Theory of Phase-Ordering Kinetics. Adv. Phys. 1994, 43, 357−459. (17) Binder, K. Phase Transitions in Polymer Blends and Block Copolymer Melts: Some Recent Developments. Adv. Polym. Sci. 1994, 112, 181−299. (18) Kawaguchi, M.; Suzuki, S.; Imae, T.; Kato, T. Surface Morphology of Langmuir−Blodgett Blend Films of Poly(vinyl acetate)−Poly(methyl acrylate). Langmuir 1997, 13, 3794−3799. 9072

dx.doi.org/10.1021/jp403195g | J. Phys. Chem. B 2013, 117, 9067−9072