Article pubs.acs.org/Langmuir
Asymmetrically Tilted Alignment of Rigid-Rod Helical Polysilanes on a Rubbed Polyimide Surface Kento Okoshi,*,† Michiya Fujiki,‡ and Junji Watanabe§ †
Department of Bio- and Material Photonics, Chitose Institute of Science and Technology, 758-65 Bibi, Chitose, Hokkaido 066-8655, Japan ‡ Graduate School of Materials Science, Nara Institute of Science and Technology, 8916-5 Takayama, Ikoma, Nara 630-0101, Japan § Department of Polymer Chemistry, Tokyo Institute of Technology, Ookayama, Meguro-ku, Tokyo 152-8552, Japan ABSTRACT: The homogeneous alignments of helical rod-like polysilanes on a rubbed polyimide alignment layer were investigated by polarized optical microscopy (POM) and atomic force microscopy (AFM) analyses. The POM and AFM observations determined that polysilanes with a series of aliphatic side chains helically arranged around the main chains were tilted to the right and left by 33° from the rubbing direction when the handedness of the side-chain helical array is left and right, respectively. It is interesting to note that the side-chain arrays run perpendicular to the rubbing direction on the polyimide surface, which is different from intuitive “knob and hole” packing of the extended polyimide chain and the helical grooves between the side-chain arrays surrounding the polysilane backbone. More surprisingly, both right- and left-tilting smectic domains were simultaneously observed with an equal probability for an achiral polysilane, which apparently has the interconverting right- and left-handed helical segments separated by helical reversals. This might be the first observation of the chiral segregation of dynamic helical polymers.
■
INTRODUCTION The liquid crystal (LC) alignment induced by the mechanically rubbed thin polymer layer (alignment layer) on a substrate has been empirically established for display applications.1 The most widely accepted explanation for the alignment is that the rubbing-induced orientation of the polymer chains produces the LC orientation, although it still lacks direct evidence for the correlation between the LC and polymer orientations.1b,2 These LC molecules align with the plane containing the substrate normal and the rubbing direction for a wide variety of combinations with the polymer and the LC, even if the LC is chiral. When the chiral LC molecules are used, there is an expected chiral interaction between the LC and alignment layer; however, no such interfacial chirality has been reported to date. In contrast, however, the alignment of achiral LC molecules on films of aligned double-stranded DNA has been recently explored,3 and such an interfacial chirality as tilting of the LC molecules from the alignment direction of DNA has been reported, indicating the coupling between the helical chirality of the DNA and LC molecules. The direction of the observed tilting was strongly dependent upon the molecular species at play, which was reported to be attributable to the electric couplings between the localized dipole of the LC molecules and the helically arranged phosphate groups of DNA and the segregation of the nonpolar groups of the LC molecules into the hydrophobic grooves of DNA. We have recently reported the thermotropic smectic LC phases formed by the systems of the carefully prepared polysilanes with helical rod-like structures and narrow molecular-weight distributions, which show the theoretically predicted liquid crystal © 2012 American Chemical Society
sequence for rod-like particles of the nematic−smectic−columnar phases.4 The rod-like nature of polysilane comes from the predominantly one-handed helical structure of the main chain for chiral polysilanes. Even the achiral polysilanes retain a certain level of stiffness, although there should be a number of helical reversals between the interconverting right- and left-handed helical segments, which may destabilize the rigidity of the polymer backbone.5 The nonpolar nature of these polymers is also obvious because these polymers consist of only silicon, carbon, and hydrogen atoms. This feature is advantageous for the preparation of a sample with a narrow molecular-weight distribution by a simple fractional precipitation method because no specific intermolecular interactions take place in the solution. In this study, we prepared a series of polysilanes with the following formula:
Received: December 5, 2011 Revised: February 13, 2012 Published: February 17, 2012 4811
dx.doi.org/10.1021/la204789g | Langmuir 2012, 28, 4811−4814
Langmuir
Article
Figure 1. POM observations of rod-like helical polysilane films on a rubbed polyimide surface under crossed polarizers. (a−c) Polarized optical micrographs of the poly-S-1 (Mw = 21.8 × 103, with Mw/Mn = 1.16) film and (d−f) those with a 530 nm sensitive color plate. P and A represent the transmission axes of the polarizer and analyzer, respectively. X′ and Z′ represent the fast and slow axes of the sensitive color plate, respectively. The film was tilted by −33° (a, counterclockwise), 0° (b, parallel), and +12° (c, clockwise) to the transmission axis of the analyzer. The white arrows indicate the rubbing direction. (g and h) Polar plots of the transmission against polarization angles with respect to the rubbing direction for poly-S-1 in panel d and poly-S-2 (blue line) and poly-R-2 (red line) in panel e. and Mw = 111 × 103, with Mw/Mn = 1.25 (for AFM), for poly-S-1; Mw = 59.6 × 103, with Mw/Mn = 2.77, for poly-S-2; Mw = 80.9 × 103, with Mw/Mn = 2.34, for poly-R-2; and Mw = 119 × 103, with Mw/Mn = 1.34, for poly-3, by gel permeation chromatography (GPC). Coating on the Rubbed Polyimide Alignment Layer. The chloroform solutions of each polysilane (8.0 mg/mL) were spincoated at a speed of 200 rpm for 10 s and then at 1000 rpm for 60 s on a glass substrate (10 × 10 mm) coated with the polyimide film, RN1199A (Nissan Chemical, Tokyo, Japan) and rendered anisotropic by rubbing, followed by annealing at 120 °C for 5 h before the observations. The polyimide film did not have any significant birefringence. The thickness of polysilane was estimated to be 400 nm by AFM measurements.9 Instruments. The GPC measurements were performed using a LC-2000plus HPLC system (Jasco, Tokyo, Japan) equipped with a GPC K-805 L column (Showa Denko, Tokyo, Japan). Chloroform was used as the eluent at 40 °C at the flow rate of 0.5 mL/min. The molecular weights of the samples were estimated by calibration with standard polystyrenes (Showa Denko). Polarizing optical microscopic observations were carried out using an Olympus BH-2 polarizing microscope (Olympus, Tokyo, Japan) equipped with a 530 nm sensitive color plate. The images were captured by a digital camera coupled to the microscope. The AFM measurements were carried out using a MAC mode III 5500 (Agilent Technology, Inc., Santa Clara, CA) in the AC mode at room temperature under ambient conditions. Imaging was conducted in the tapping mode using a silicon cantilever with a resonance frequency of 260−320 kHz. The typical settings of AFM for observations were as follows: set point, 1−2 V; scan rate, 0.5−1.0 line/s; and scan size, 1 × 1 μm2.
having a 7-residue−3-turn (73) helical main-chain structure6 with a diameter of 1.49 nm7 for all of them and narrow molecular-weight distributions, with which they form the smectic A phase,4a for poly-S-1 and poly-3. We examined how these polymers align on a rubbed polyimide surface to determine the interfacial chiral interactions using the polarized optical microscopy (POM) and the atomic force microscopy (AFM) observations of the smectic layers. Although the interfacial chiral interaction of the helical polymer is substantially the same as in the case with a rubbed DNA surface and LC molecules, it should be of considerable interest because polysilanes are totally apolar, so that no specific interfacial interaction can be expected, such as those in the DNA system.
■
MATERIALS AND METHODS
Materials. Chloroform was used as the eluent for the size-exclusion chromatography (SEC) measurements and was of high-performance liquid chromatography (HPLC) grade (Wako, Osaka, Japan). Anhydrous chloroform (water content < 30 ppm) was purchased from Wako and used for the preparation of the samples for the POM and AFM observations. Sample Preparation. Polysilanes were synthesized using the corresponding dichlorosilane monomers for each polymer by the Wurtz-type condensation in toluene at 120 °C based on previously reported methods.8 The resulting polymer in toluene was repeatedly fractionated by a fractional precipitation method using 2-propanol, ethanol, and methanol as the precipitants to yield samples with narrow molecular-weight distributions. The molecular weights of the samples were determined to be Mw = 21.8 × 103, with Mw/Mn = 1.16 (for POM), 4812
dx.doi.org/10.1021/la204789g | Langmuir 2012, 28, 4811−4814
Langmuir
Article
Figure 2. AFM images of the smectic A layers in the poly-S-1 (Mw = 111 × 103, with Mw/Mn = 1.25) spin-cast film on (a and b) rubbed polyimide surface and (c) corresponding Fourier transform. Scale = 2 × 2 μm for panels a and b.
■
RESULTS AND DISCUSSION The birefringence of the poly-S-1 film could be clearly seen by POM (panels a−c of Figure 1). The bright states of the uniform texture were clearly observed when the rubbing direction of the film was rotated −78° (Figure 1a, counterclockwise) and +12° (Figure 1c, clockwise) from the transmission axis of the analyzer around the normal to the surface, while it became completely dark when rotated −33° (Figure 1b, counterclockwise). The alignment of poly-S-1 was also probed by the transmission under the crossed polarizers plotted as a function of the polarization angle with respect to the rubbing direction (Figure 1g). A good quality extinction was observed at +33° (clockwise) and −57° (counterclockwise), indicating that the orientation direction of poly-S-1 is apparently tilted either +33° or −57° from the rubbing direction. The POM observations of the film with a 530 nm sensitive color plate successfully identified the orientation direction of poly-S-1 (panels d−f of Figure 1). The transmitted light at a sample rotation angle of −78° turned orange (Figure 1d) from the original color of pink in the dark state (Figure 1e), while it turned bluish pink at +12° (Figure 1f). These results clearly indicate that the orientation direction of poly-S-1 lies in the direction tilted +33° from the rubbing direction because poly-S-1 has a positive birefringence, as elucidated from the color of its sheared film under POM with a sensitive color plate. We further investigated the orientation direction of the optical isomers, poly-S-2 and poly-R-2, which have 73 helical main-chain structures with an opposite helical handedness, and found that these two polymers on the rubbed polyimide film interestingly align in the directions symmetrically tilted +33° and −33° from the rubbing direction, respectively, which was elucidated by the plots of the transmission under the crossed polarizers versus the polarization angles with respect to the rubbing direction in Figure 1h.10 The symmetrical tilting of the orientation directions for the optical isomers clearly substantiates the coupling between the helical chirality of the polymers and the alignment layer. The identification of the orientation direction of poly-S-1 was further confirmed by AFM observations, which have been reported to be an excellent way to collect information on the smectic layer structure and order.11 Figure 2 shows the AFM images of the spin-cast film of poly-S-1 with a narrow molecular-weight distribution on the rubbed polyimide surface. Poly-S-1 forms a smectic phase, as evidenced by its clear smectic layers, which are uniformly aligned across the entire field of view. The 2D Fourier transform of the AFM image (Figure 2c) shows the repeat distance of 68.9 nm, which is approximately equal to the molecular length (72.2 nm) of polyS-1 calculated using Mn and 0.196 nm as the translational length per residue.12 It also shows that the smectic layer normal, i.e., the orientation direction of poly-S-1, is tilted +33°
from the rubbing direction, clearly reproducing the orientation direction determined in the POM observations. Therefore, it is reasonable to conclude that poly-S-1 is asymmetrically tilted from the rubbing direction, which proves the interfacial chiral interaction between the helical structure of polysilane and the rubbed polyimide surface. To the best of our knowledge, such a clear identification of the orientation has never been reported with helical polymers cast on an alignment layer, as schematically illustrated in Figure 3a.
Figure 3. (a) Schematic illustration of tilted alignment of poly-S-1 on a rubbed polyimide surface. The long side chains are replaced with ethyl groups and colored in purple and blue for each side-chain helical array for clarity. (b) Radial projection of the alkyl side-chain helical arrays of poly-S-1.
It was naturally anticipated that the extended polyimide chains created by the rubbing fit in the helical grooves between the side-chain arrays surrounding the polymer backbone13 like the “knob into hole” helix−helix packing.14 However, the actual results turned out to be very different. Figure 3b represents a radial projection of side-chain helical arrays of poly-S-1, in which the aliphatic side chains are represented as filled circles consecutively numbered from 0 to 14 along the main chain. The tilt angle of the side-chain array from the perimeter was calculated to be 33° based on the diameter of the polymer rod and the fiber period of poly-S-1, indicating that the aliphatic side-chain helical arrays run perpendicular to the rubbing direction on the polyimide surface, which is counterintuitively different from the “knob into hole” helix−helix packing.15 Although a similar perpendicular alignment of achiral LC molecules to the grooves of double-stranded DNA has been reported to be attributed to the electric and hydrophobic interactions between the components, the exact mechanism of the alignment remains to be clarified because polysilanes are apolar enough to expect no specific interaction with the rubbed polyimide. On the basis of these results, one might anticipate that achiral polysilanes align in the rubbing direction because of its achiral 4813
dx.doi.org/10.1021/la204789g | Langmuir 2012, 28, 4811−4814
Langmuir
Article
Figure 4. AFM images of the smectic A layers observed on the poly-3 (Mw = 119 × 103, with Mw/Mn = 1.34) film on (a and b) rubbed polyimide surface and (c) corresponding Fourier transforms. Scale = 5 × 5 μm for panels a and b.
Creative Scientific Research and Scientific Research (C), and Japan Science and Technology, Research Seeds Program.
nature. Surprisingly, however, both the right- and left-tilting smectic domains were observed in poly-3 with a narrow molecular-weight distribution cast on a rubbed polyimide surface by AFM (Figure 4). Changing the scan direction or speed of AFM did not affect these images, thereby precluding the possibility of some similar imaging artifact. Both domains were observed with an almost equal probability. However, the tilting in each domain was not exactly ±33° like those in poly-S-1, which might be possibly due to the semi-rigid dynamically interconverting helical main-chain structure. Even though the segregation is not perfect, this might be the first observation of the optical resolution of right- and left-handed helical polymers.
■
(1) (a) Cognard, J. Alignment of Nematic Liquid Crystals and Their Mixtures; Gordon and Breach Science Publishers: New York, 1982. (b) Geary, J. M.; Goodby, J. W.; Kmetz, A. R.; Patel, J. S. J. Appl. Phys. 1987, 62, 4100−4108. (2) (a) Castellano, J. A. Mol. Cryst. Liq. Cryst. 1983, 94, 33−41. (b) Barmentlo, M.; van Aerle, N. A. J. M.; Hollering, R. W. J.; Damen, J. P. M. J. Appl. Phys. Rev. 1992, 71, 4799−4804. (3) Nakata, M.; Zanchetta, G.; Buscaglia, M.; Bellini, T.; Clark, N. A. Langmuir 2008, 24, 10390−10394. (4) (a) Okoshi, K.; Kamee, H.; Suzaki, G.; Tokita, M.; Fujiki, M.; Watanabe, J. Macromolecules 2002, 35, 4556−4559. (b) Okoshi, K.; Saxena, A.; Fujiki, M.; Suzaki, G.; Watanabe, J.; Tokita, M. Mol. Cryst. Liq. Cryst. 2004, 419, 57−68. (c) Okoshi, K.; Saxena, A.; Naito, M.; Suzaki, G.; Tokita, M.; Watanabe, J.; Fujiki, M. Liq. Cryst. 2004, 31, 279−283. (d) Okoshi, K.; Suzuki, A.; Tokita, M.; Fujiki, M.; Watanabe, J. Macromolecules 2009, 42, 3443−3447. (e) Okoshi, K.; Watanabe, J. Macromolecules 2010, 43, 5177−5179. (5) Fujiki, M.; Koe, J. R.; Terao, K.; Sato, T.; Teramoto, A.; Watanabe, J. Polym. J. 2003, 35, 297−344. (6) Watanabe, J.; Kamee, H.; Fujiki, M. Polym. J. 2001, 33, 495−497. (7) The diameter of the helical main chain was evaluated by AFM measurements of an isolated polymer on a HOPG substrate. (8) Fujiki, M. J. Am. Chem. Soc. 1996, 118, 7424−7425. (9) The orientations of polysilanes were not successfully observed with the film thickness over 1 μm. (10) The orientations of poly-2s were not as good as poly-S-1, and the reason is unclear. (11) (a) Oka, H.; Suzaki, G.; Edo, S.; Suzuki, A.; Tokita, M.; Watanabe, J. Macromolecules 2008, 41, 7783−7786. (b) Onouchi, H.; Okoshi, K.; Kajitani, T.; Sakurai, S.-i.; Nagai, K.; Kumaki, J.; Onitsuka, K.; Yashima, E. J. Am. Chem. Soc. 2008, 130, 229−236. (c) Wu, Z.-Q.; Nagai, K.; Banno, M.; Okoshi, K.; Onitsuka, K.; Yashima, E. J. Am. Chem. Soc. 2009, 131, 6708−6718. (12) We fully understand that the molecular weight estimated by GPC using polystyrene standards may be insufficient and overestimated for polymers with rigid main chains. (13) The handedness of the main-chain helical structure of poly-S-1 has been successfully determined to be right-handed by a spectroscopic method.5 The aliphatic side chains arranged around the rod-like main chain form an opposite left-handed 14-residue− 1-turn (141) double-helical array. (14) Crick, F. H. C. Acta Crystallogr. 1953, 6, 685−689. (15) It might be worthwhile to note that the separation between polysilanes along the rubbing direction [2.8 nm; =1.5 nm (lateral distance between polysilanes4a)/sin 33°] is substantially larger than the repeat unit length of commonly used polyimide (1.6−1.7 nm).
■
CONCLUSION In conclusion, we observed the interfacial chiral interaction between the helical rod-like polysilanes and a rubbed polyimide surface. The POM and AFM observations determined that the rod-like poly-S-1 cast on a rubbed polyimide surface was tilted to the right by 33° from the rubbing direction. This asymmetrical tilting is apparently attributed to the chirality of the left-handed helical arrays of the alkyl side chains because the optical isomers with the left- and right-handed helical arrays of the side chains were found to be symmetrically tilted to the right and left by 33° from the rubbing direction, respectively. To our surprise, these right- and left-tilting domains were simultaneously observed for an achiral polysilane, which consists of the interconverting right- and left-handed helical segments. Furthermore, we found that these helical arrays of the side chains intertwining along the polymer main chain run perpendicular to the rubbing direction on the polyimide surface, although the exact mechanism of this counterintuitive alignment behavior still remains unclear. The present results demonstrated not only how to manipulate the alignment of the helical rod-like polymers on a surface, which could be crucial for use in the areas of optical, electronic, and sensing devices, but also a new strategy to provide a better understanding of the supramolecular assembly of macromolecules. Some of the projects outlined above will be the objective of future investigations.
■
REFERENCES
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This research was partially supported by the Ministry of Education, Science, and Culture in Japan, Grant-in-Aid for 4814
dx.doi.org/10.1021/la204789g | Langmuir 2012, 28, 4811−4814
