9790
Langmuir 2008, 24, 9790-9794
Polar and Azimuthal Alignment of a Nematic Liquid Crystal by Alkylsilane Self-Assembled Monolayers: Effects of Chain-Length and Mechanical Rubbing Stephanie M. Malone and Daniel K. Schwartz* Department of Chemical and Biological Engineering, UniVersity of Colorado, Boulder, Colorado 80309-0424 ReceiVed April 28, 2008. ReVised Manuscript ReceiVed June 27, 2008 Alkylsilane self-assembled monolayers (SAMs) on oxide substrates are commonly used as liquid crystal (LC) alignment layers. We have studied the effects of alkyl chain length, photolytic degradation, and mechanical rubbing on polar and azimuthal LC anchoring. Both gradient surfaces (fabricated using photolytic degradation of C18 SAMs) and unirradiated SAMs composed of short alkyl chains show abrupt transitions from homeotropic to tilted alignment as a function of degradation or chain length. In both cases, the transition from homeotropic to tilted anchoring corresponds to increasing wettability of the SAM surfaces. However, there is an offset in the critical contact angle for the transition on gradient vs unirradiated SAMs, suggesting that layer thickness is more relevant than wettability for LC alignment. Mechanical rubbing can induce azimuthal alignment along the rubbing direction for alignment layers sufficiently near the homeotropic-to-planar transition. Notably, mechanical rubbing causes a small but significant shift in the homeotropic-to-tilted transition, e.g., unrubbed C5 SAMs induce homeotropic anchoring, but the same surface after rubbing induces LC pretilt.
Introduction Chemical and physical surface properties define the preferred alignment of an adjacent liquid crystal (LC) phase, a phenomenon known as anchoring. For LC applications, substrates are generally pretreated to ensure a predetermined anchoring; a number of appropriate treatments have been developed empirically. From a fundamental perspective, the anchoring and orientation of LCs can be regarded as a sensitive probe of surface properties. Liquid crystals are capable of detecting surface features that are invisible to AFM, ellipsometry, and FTIR.1–3 Previous studies have used various surface modification approaches to create gradients,1,4 patterns,5–8 and blended substrates9,10 in order to probe the details of the LC/substrate interaction. Both polar and azimuthal anchoring are of significant interest; changes in either type of alignment can provide information on surface properties. Polar anchoring has been shown to be affected by monolayer density,9,11–13 substrate blending,14 irradiation,15,16 * To whom correspondence should be addressed. E-mail: daniel.schwartz@ colorado.edu. Phone: 303-735-0240. (1) Price, A. D.; Schwartz, D. K. Langmuir 2006, 22, 9753–9759. (2) Moon, D. W.; Kurokawa, A.; Ichimura, S.; Lee, H. W.; Jeon, I. C. J. Vac. Sci. Technol., A 1999, 17, 150–154. (3) Ye, T.; Wynn, D.; Dudek, R.; Borguet, E. Langmuir 2001, 17, 4497–4500. (4) Clare, B. H.; Efimenko, K.; Fischer, D. A.; Genzer, J.; Abbott, N. L. Chem. Mater. 2006, 18, 2357–2363. (5) Wan, J. T. K.; Tsui, O. K. C.; Kwok, H.-S.; Sheng, P. Phys. ReV. E 2005, 72, 1–4. (6) Behdani, M.; Keshmiri, S. H.; Soria, S.; Bader, M. A.; Ihlemann, J.; Marowsky, G.; Rasing, T. Appl. Phys. Lett. 2003, 82, 2553–2555. (7) Behdani, M.; Rastegar, A.; Keshmiri, S. H.; Missat, S. I.; Vlieg, E.; Rasing, T. Appl. Phys. Lett. 2002, 80, 4635–4637. (8) Lee, B.-W.; Clark, N. A. Science 2001, 291, 2576–2580. (9) Ogawa, K.; Mino, N.; Nakajima, K. Jpn. J. Appl. Phys. 1990, 29, L1689– L1692. (10) Vaughn, K. E.; Sousa, M.; Kang, D.; Rosenblatt, C. Appl. Phys. Lett. 2007, 90, 194102. (11) Price, A. D.; Schwartz, D. K. J. Phys. Chem. B 2007, 111, 1007–1015. (12) Brake, J. M.; Mezera, A. D.; Abbott, N. L. Langmuir 2003, 19, 8629– 8637. (13) Lockwood, N. A.; dePablo, J. J.; Abbott, N. L. Langmuir 2005, 21, 6805– 6814. (14) Filas, R. W.; Patel, J. S. Appl. Phys. Lett. 1987, 50, 1426–1428.
and topographical corrugation of the substrate.17 Subtle changes in the molecular composition of surface layers can have significant and complex effects. For example, a change in the length of polymeric side chains affected the LC tilt angle.18 Drawhorn and Abbott19 found that alkanethiol SAMs on gold of all chain lengths induced tilted or planar alignment; however, appropriate chain length mixtures caused homeotropic alignment. Crawford et al.20 found that in alumina membrane pores modified with aliphatic acids, the chain length directly affected polar anchoring, with n < 7 causing tilted anchoring, and longer chain lengths inducing homeotropic anchoring. Azimuthal alignment is also affected by SAM chain length, e.g. Gupta and Abbot21 found that odd numbered alkyl chains gave alignment orthogonal to that caused by even numbered alkyl chains in their work on alkanethiol SAMs on textured gold. In previous work1 we prepared surfaces with a continuous gradient of wettability by photolytic degradation of an alkylsilane SAM using spatially varying UV irradiation. We observed an abrupt (discontinuous) transition from homeotropic to planar anchoring as a function of SAM degradation, and we correlated the transition to the water contact angle of the surface. This agrees with the work of Alkharialla et al.,22 who also found that increasing hydrophobicity correlated with homeotropic alignment. These gradient surfaces are experimentally useful because one can conceptually treat the degree of degradation as a continuous variable in order to probe the detailed nature of a transition. (15) Seo, D.-S.; Han, J.-M. Liq. Cryst. 1999, 26, 959–964. (16) Chaudhari, P.; Lacey, J.; Doyle, J.; Galligan, E.; Lien, S.-C. A.; Callegari, A.; Hougham, G.; Lang, N. D.; Andry, P. S.; John, R.; Yang, K.-H.; Lu, M.; Cai, C.; Speidell, J.; Purushothaman, S.; Ritsko, J.; Samant, M.; Stohr, J.; Nakagawa, Y.; Katoh, Y.; Saitoh, Y.; Sakai, K.; Satoh, H.; Odahara, S.; Nakano, H.; Nakagaki, J.; Shiota, Y. 2001, 411, 56–59. (17) Sinha, G. P.; Wen, B.; Rosenblatt, C. Appl. Phys. Lett. 2001, 79, 2543– 2545. (18) Ban, B. S.; Kim, Y. B. J. Phys. Chem. B 1999, 103, 3869–3871. (19) Drawhorn, R. A.; Abbott, N. L. J. Phys. Chem. 1995, 99, 16511–16515. (20) Crawford, G. P.; Ondris-Crawford, R. J.; Doane, J. W. Phys. ReV. E 1996, 53, 3647–3662. (21) Gupta, V. K.; Abbott, N. L. Phys. ReV. E 1996, 54, R4540–R4543. (22) Alkhairalla, B.; Allinson, H.; Boden, N.; Evans, S. D.; Henderson, J. R. Phys. ReV. E 1999, 59, 3033–3039.
10.1021/la801322x CCC: $40.75 2008 American Chemical Society Published on Web 08/08/2008
Polar and Azimuthal Alignment of a Nematic Liquid Crystal
Figure 1. Polarized microscope images of HAN cells; the analyzer and polarizer are oriented as shown in the annotation. The surfaces of interest are: (a) photolytically degraded OTES (C18) SAM with a continuous gradient of surface properties (least degraded at left, most degraded at right), (b) C6 SAM, (c) C5 SAM, (d) C4 SAM, (e) C3 SAM, (f) C2 SAM, (g) C1 SAM, (h) glass.
However, because the degradation is extremely sensitive to the details of the experimental conditions, it is not a practical way to reproducibly create homogeneous alignment layers with predetermined properties. Furthermore, while it is known that photolytic degradation results in gradual shortening of the alkyl chains in SAMs,2,3 the laterally random cleavage of terminal C-C bonds results in increasing chain-length dispersion as the degradation proceeds. Therefore, it is unclear which continuously varying parameter causes the anchoring transition observed on gradient surfaces, i.e. wettability, average chain length, chain length dispersion, etc. The comparisons made, in this manuscript, with unirradiated uniform SAMs prepared from short chain alkylsilanes provide insight into these issues. Mechanical rubbing of organic surface films (e.g., polymer layers or SAMs) is a common way to produce surface anisotropy and therefore azimuthal alignment, as is required for display applications. There is evidence that rubbing can cause nanogrooves in a polymer substrate,23 or that it can actually physically align polymer backbones on some substrates.24 The effects of rubbing strength,25 multidirectional rubbing,26 and nanogrooves27 have been explored on various surfaces, all of which affect azimuthal anchoring energy. Walba et al.28 showed that rubbed OTES SAMs induced azimuthal alignment for an LC in the smectic C phase which has inherent tilt, but not the untilted nematic or Smectic A phases. In the current work, degraded gradient and unirradiated SAMs are used as model systems to explore this phenomenon. Specifically, we seek to determine whether rubbing simply adds anisotropy to a surface which already induces tilted anchoring, or whether rubbing can alter the surface anchoring in a more significant way. We find that mechanical rubbing causes an observable shift in the homeotropic-to-planar anchoring transition, suggesting that the rubbing does not simply provide anisotropy but also causes a change in surface properties that affects the polar anchoring.
Experimental Methods Alignment Layer Preparation. Soda-lime glass slides (Fisher Scientific) were cleaned with piranha solution (30% aqueous H2O2 and concentrated H2SO4, 1:3 v/v) at 100 °C. (Warning: piranha solution reacts strongly with organic compounds and should be handled with extreme caution; do not store solution in closed container). Octadecyltriethoxysilane (OTES, C18) self-assembled monolayers (SAMs) were prepared using the amine-catalyzed (23) Berreman, D. W. Phys. ReV. Lett. 1972, 28, 1686–1686. (24) Ishihara, S.; Wakemoto, H.; Nakazima, K.; Matsuo, Y. Liq. Cryst. 1989, 4, 669–675. (25) Oka, S.; Mitsumoto, T.; Kimura, M.; Akahane, T. Phys. ReV. E 2004, 69. (26) Kim, Y. J.; Zhuang, Z.; Patel, J. S. Appl. Phys. Lett. 2000, 77, 513–515. (27) Faetti, S. Phys. ReV. A 1987, 36, 408–410. (28) Walba, D. M.; Liberko, C. A.; Shao, R.; Clark, N. A. Liq. Cryst. 2002, 29, 1015–1024.
Langmuir, Vol. 24, No. 17, 2008 9791 approach as described by Walba et al.29 For discrete alkyl chain length SAMs, decyltriethoxysilane (C10), hexyltriethoxysilane (C6), pentyltriethoxysilane (C5), butyltriethoxysilane (C4), propyltriethoxysilane (C3), ethyltriethoxysilane (C2), and methyltriethoxysilane (C1) were substituted for octadecyltriethoxysilane. All alkyltriethoxysilane compounds were obtained from Gelest and used as received. Trimethysilane (TMS) monolayers were prepared by placing clean slides into a glass staining dish, where slides were exposed to hexamethyldisilazane vapor (99.8% purity, Acros Organics, NY) for ∼12 h. The substrate was held approximately ∼2 cm above the liquid. Substrates with a continuous gradient of surface properties were prepared as described previously.1 Briefly, C18 SAM slides were cut in half and degraded by placing the edge of the slide in direct contact with a mercury pen lamp (UVP, 254 nm) for 7 min. The intensity was 3.6 mW/cm2 at the point of the sample in contact with the lamp, and fell to 0.3 mW/cm2 at the far end of the sample, 4 cm away. This range of intensity is similar to that used in other reports of silane degradation.3,30,31 The gradual decrease of the irradiation intensity as a function of distance from the lamp resulted in a continuous gradient of surface properties. There was variability from sample to sample in the details of this gradient, thus we characterized each substrate independently by measuring contact angles along the gradient direction.32 Shorter chain SAMs (C1-C10, TMS) were not irradiated. Mechanical rubbing was performed using a custom-built rotating brush apparatus (Winston and Newton Series 240 brush). Each sample was brushed 100 times for uniformity. Samples were then placed in an ultrasonic cleaner in a solution of micro-90 and water for 10 min to remove any surface contamination. For some substrates, half of the SAM was brushed, while half of the sample was shielded from brushing by covering with a piranha-cleaned glass slide to act as an internal control. Averages and standard deviations were calculated from six independent short chain SAMs and three irradiated gradient SAMs. LC Cell Preparation. Hybrid aligned nematic (HAN) cells were made by cementing two SAMs with 10 µm glass fibers blended into the epoxy. The cells were then clamped while the epoxy cured, while the glass fibers in the epoxy maintained uniform spacing. One bounding surface of each HAN cell was always an unirradiated OTES layer, producing uniform homeotropic anchoring at that surface. The other side of the HAN cell was the substrate of interest in each experiment: either a degraded OTES gradient SAM or an unirradiated shorter chain length SAM. Cell thickness was measured using interference fringes from monochromatic light transmitted through air. The HAN cells were filled with 4-n-pentyl-4′cyanobiphenyl (5CB, Alfa Aesar, N-I transition 35.5 °C) by capillary action while heating 5CB above its N-I transition to eliminate any flow alignment. Polarized Light Microscopy. After sample preparation, the HAN cells were viewed between crossed polarizers using a Lightbox light table. Quantitative effective birefringence values were calculated from retardation measurements with an Olympus UCT-B Berek compensator using monochromatic light. Each data point reflects the mean and standard deviation of six individual measurements. For a HAN cell, one boundary condition is known, therefore the effective birefringence, nj ) nje - no, where no ) n⊥, and
n¯e )
1 θ2 - θ1
∫θθ
n||n⊥dθ
2
1
(n||2cos2
θ + n⊥2 sin2 θ)1 ⁄ 2
where n⊥ is the index of refraction perpendicular to the optical axis, n|| is the index of refraction for 5CB parallel to the optical axis.33 (29) Walba, D. M.; Liberko, C. A.; Korblova, E.; Farrow, M.; Furtak, T. E.; Chow, B. C.; Schwartz, D. K.; Freeman, A. S.; Douglas, K.; Williams, S. D.; Klittnick, A. F.; Clark, N. A. Liq. Cryst. 2004, 31, 481–489. (30) Ye, T.; McArthur, E. A.; Borguet, E. J. Phys. Chem. B 2005, 109, 9927– 9938. (31) Hong, L.; Sugimura, H.; Furukawa, T.; Takai, O. Langmuir 2003, 19, 1966–1969. (32) Uyama, Y.; Inoue, H.; Ito, K.; Kishida, A.; Ikada, Y. J. Colloid Interface Sci. 1991, 141, 275–279.
9792 Langmuir, Vol. 24, No. 17, 2008
Malone and Schwartz
Figure 2. Cosine of the contact angle of water measured on (a) a photolytically degraded gradient surface as a function of distance from the UV source, and (b) unirradiated SAMs as a function of alkyl chain length.
Refractive indices used were 1.5326 and 1.729 for n⊥and n||, respectively.34 This expression was integrated numerically with Mathematica. Contact Angle Goniometry. Contact angle measurements were made using a custom built goniometer, using static drops32 of Millipore water (18.2 MΩ). For gradient substrates, measurements were taken at 3 mm intervals along the gradient direction. Contact angle measurements were made on a minimum of three independent SAMs with ten samples at random positions on each SAM.
Results Polar Anchoring. Figure 1 shows polarized microscope images of HAN cells prepared using a photolytically degraded C18 SAM gradient (Figure 1a), HAN cells prepared using unirradiated shorter SAMs of varying chain lengths (Figure 1b-g), and clean glass (Figure 1h). As reported previously in greater detail,1 a relatively abrupt transition from homeotropic anchoring to nearly planar polar anchoring is observed with increasing degradation along the gradient surface. This appears in Figure 1a as a transition from optical extinction on the left to colors associated with large birefringence on the right. Figure 1b-g demonstrates that related behavior is observed as a function of chain length for shorter unirradiated SAMs. For chain lengths gC5 (Figure 1b,c) extinction is observed as a result of homeotropic anchoring, while for shorter chains (Figure 1d-g), the birefringence is consistent with tilted or planar anchoring. Consistent with this trend, TMS monolayers also induced tilted anchoring (not shown). These observations of changes in LC alignment suggest that the interfacial free energies associated with homeotropic and tilted/planar anchoring evolve systematically with UV exposure and/or chain length. It is therefore interesting to compare the surface anchoring with other measures of interfacial free energy, such as wettability, and Figure 2 shows the correlation between contact angle and degradation (Figure 2a) or chain length (Figure 2b) respectively. The cosine of the contact angle decreases systematically with distance from the UV source, i.e. irradiation intensity, or increasing chain length for alkyltriethoxysilanes, suggesting at least a qualitative correlation between wettability and LC anchoring. Although they were only one methyl group thick, TMS monolayers had a contact angle of φ ) 97° ( 2° (cos φ ) -0.12). Thus TMS monolayers were much more hydrophobic than C1 SAMs, which have a similar thickness. This is presumably due to the increased lateral density of methyl groups in the TMS monolayers. In order to compare the gradient SAM and unirradiated SAMs directly, one can collapse the measured birefringence data by plotting it against cos φ (Figure 3). As expected from the qualitative observations, both types of alignment layers show an
abrupt transition of birefringence (from homeotropic to tilted) vs wettability. However, because of the significant offset between the critical values of φ (85° for unirradiated monoalkylsilanes, ∼55° for irradiated monolayer), this suggests that water and the LC phase probe somewhat different chemical or structural surface properties. Furthermore, LC cells prepared from TMS alignment layers were birefringent, despite the TMS contact angle of 97°. These observations indicate that contact angle alone is not the controlling factor for LC alignment. The C4 SAM is particularly interesting because it falls directly within the transitional region between homeotropic and tilted anchoring. In Figure 3, the uncertainty of the birefringence for C4 samples is very large; this uncertainly is calculated as the standard deviation of values measured for six independent samples. The broad distribution for the C4 samples supports the notion that the transition from homeotropic to tilted anchoring is extremely sensitive to sample variation. Azimuthal Anchoring Due to Mechanical Rubbing. The effect of mechanical rubbing is also directly correlated to degree of degradation or alkyl chain length. For gradient surfaces, azimuthal alignment occurs only in the region of rapidly increasing birefringence (Figure 4a,b). Far to the left of the transition region, where the degradation is smaller, the homeotropic anchoring is not affected by rubbing, and optical extinction is observed regardless of the sample orientation. Similarly, far to the right of the transition region, rubbing has no effect, presumably because the surface is virtually bare. In this region, the hydrophilicity is similar to that of piranha-cleaned glass substrates, i.e., the contact angle is too low to be measured. Within the transition region, however, there is strong azimuthal alignment, and one observes extinction in the rubbed region when the rubbing direction is aligned with the polarizer. Furthermore, there is a subtle shift in the location of the transition toward the left, which is visible in Figure 4b, indicating that near the transition point, rubbing can convert homeotropic anchoring to tilted anchoring. This effect is more easily visualized in the unirradiated SAMs described below. As with the gradient surface, when the chain length of an unirradiated SAM is large compared to the transition chain length, rubbing has no effect. For example, C6 SAMs induce uniformly homeotropic anchoring regardless of rubbing (Figure 4c,d). However, for C5 SAMs, while unrubbed regions induce homeotropic anchoring, rubbing results in weak birefringence. C4 through C1 SAMs are all aligned by rubbing, while no effect is observed on the control substrate, glass. The lack of alignment on rubbed glass demonstrates that azimuthal alignment on SAMs is directly related to the effect of rubbing on a SAM, not an
Polar and Azimuthal Alignment of a Nematic Liquid Crystal
Langmuir, Vol. 24, No. 17, 2008 9793
Figure 3. Direct comparison of birefringence vs wettability for degraded gradient and unirradiated SAMs. The optical birefringence is directly related to the anchoring tilt angle.
artifact (e.g., due to contamination from the brush). Thus, mechanical rubbing affects the magnitude of the optical birefringence (i.e., the polar tilt angle) and shifts the homeotropicto-tilted transition toward less-degraded or longer chain length.
Discussion Thickness Dependence of Polar Anchoring. UV irradiation in the presence of oxygen gas generates short-lived oxygencontaining radicals that degrade OTES SAMs by cleaving carbon-carbon bonds.1–3,35 In particular, previous work suggests that the outermost C-C bond is preferentially cleaved,30,31 resulting in sequential shortening of each chain. We previously showed using AFM that the degraded gradient SAMs were laterally homogeneous over length scales probed by the AFM tip, i.e., g5 nm. Thus, we hypothesized that the gradient surfaces described above are characterized by a gradual decrease in average chain length, and that the polar anchoring transition is directly related to film thickness. The anchoring behavior as a function of chain length for unirradiated SAMs is qualitatively consistent with this view, in particular the abrupt change from homeotropic to tilted anchoring for ∼C4 SAMs. The tilted anchoring on TMS layers also confirms this hypothesis. We note that this is in contrast with alkanethiol SAMs on gold, which induce tilted or planar alignment regardless of chain length.19,21 For both gradient and unirradiated SAMs, the anchoring transition is also correlated with a change in the wettability of water. However, the specific contact angles associated with the anchoring transition differ significantly: ∼55° for gradient SAMs and ∼85° for unirradiated monoalkylsilane SAMs. We note that the transitional contact angle on gradient SAMs is smaller than the contact angle even on unirradiated C1 SAMs, so this is indeed a significant shift. Thus while substrate wettability may provide qualitative information about LC anchoring, and may illustrate appropriate trends for a homologous series of substrates, it is apparently not directly correlated to anchoring. This shift in critical contact angle may be due to differences in surface topography and/or chemistry between irradiated and (33) Van Doorn, C. Z.; Gerritsma, C. J.; deKlerk, J. J. M. J. Influence of the DeVice Parameters on the Performance of Twisted-Nematic Liquid Crystal Matrix Displays; Plenium Press: New York, 1980. (34) Dunmur, D.; Fukuda, A.; Luckhurst, G. R., Physical Properties of Liquid Crystals: Nematics, INSPEC Institution of Electrical Engineers: London, 2001. (35) Ito, Y.; Heydari, M.; Hashimoto, A.; Konno, T.; Hirasawa, A.; Hori, S.; Kurita, K.; Nakajima, A. Langmuir 2007, 23, 1845–1850.
Figure 4. Polarized microscope images (between crossed polarizers indicated by A and P) for HAN cells created using various surfaces of interest. The bottom half of each substrate was mechanically rubbed in the direction of the arrows. The top half of each substrate was not rubbed to serve as an internal control. Each row of the figure represents a different sample. In the images on the left, the rubbing direction is parallel to the polarizer. In the images on the right, the rubbing direction is offset from the angle of the polarizer. (a, b) Degraded OTES SAM (gradient surface). (c, d) C6 SAM, (e, f) C5 SAM, (g, h) C4 SAM, (i, j) C3 SAM, (k, l) C2 SAM, (m, n) C1 SAM, (o, p) glass.
unirradiated SAMs. In particular, the dispersion in chain length is expected to be significant for the degraded SAMs. Assuming random photolytic cleavage of chains (Poisson statistics), the reduction of a C18 SAM by an average of 14 methylene units would result in a dispersion of (3.7 units giving chain lengths of 4 ( 4. We speculate that this large variation in chain length at the molecular level leads to a significant reduction of the contact angle relative to the uniform discrete SAMs. Wenzel’s equation relating surface roughness to wettability states that when φ < 90o, wettability increases, while it increases when φ > 90o.36–38 If this work on textured surfaces extrapolates to roughness at the molecular level, the inhomogeneity caused by (36) Adamson, A. W.; Gast, A. P., Physical Chemistry of Surfaces, 6th ed.; Jon Wiley and Sons, Inc.: New York, 1997. (37) Wenzel, R. N. Ind. Eng. Chem. 1936, 28, 988–994. (38) Bico, J.; Thiele, U.; QuE`E`, D. Colloids Surf., A 2002, 206, 41–46.
9794 Langmuir, Vol. 24, No. 17, 2008
irradiation could be responsible for the increased wetting of the irradiated SAMs. It has also been reported that some chain ends are oxidized to -COOH during the degradation process;31 this would also result in increased wettability. Since degradation is likely to result in a mixture of chain lengths, it is relevant to compare our observations with studies that deliberately prepared alignment layers using mixed chain lengths. Drawhorn et al.19 observed that certain mixtures of C16 with either C5 or C10 alkanethiols induced homeotropic alignment, while SAMs prepared of any single chain length resulted in tilted or planar alignment. Similarly Fazio and coworkers39 prepared Langmuir-Blodgett film alignment layers from mixtures of C18 and C22 fatty acids. They also observed that certain mixtures resulted in improved homeotropic alignment. Taken very broadly, these results suggest that mixed chain lengths may bias an LC toward homeotropic alignment, which could shift the homeotropic-to-planar transition toward alignment layers with a smaller average thickness. However, we feel that it is unlikely that this effect could be responsible for the magnitude of the shift that we observe, given the extraordinarily low value of the critical contact angle on irradiated SAMs. For unirradiated SAM substrates, the anchoring behavior as a function of chain length suggests that the influence of the underlying substrate on LC phase becomes significant when the barrier SAM is four methylene/methyl units thick. This particular thickness corresponds to the results of studies on the depth sensitivity of wetting, where Bain and Whitesides40 examined the contact angle of water on mercapto-ether monolayers on gold. They found that if an oxygen atom was embedded more than four carbons deep in an alkyl chain, it no longer affected the contact angle at the surface of the monolayer and the contact angle approached that of alkanethiols on gold. Because clean glass induces approximately planar alignment, this is consistent with the observation of tilted alignment observed in short chain unirradiated SAMs and suggests that four carbon groups (∼0.5 nm) is a critical alignment layer thickness on dielectric substrates. As mentioned previously, even much thicker SAMs on gold substrates do not induce homeotropic anchoring. (39) Fazio, V. S. U.; Komitov, L.; Lagerwall, S. T. Thin Solid Films 1998, 329, 681–685. (40) Bain, C. D.; Whitesides, G. M. J. Am. Chem. Soc. 1988, 110, 5897–5898.
Malone and Schwartz
Azimuthal Alignment. In situations where unrubbed SAM substrates induced tilted or planar anchoring, rubbing was found to cause strong azimuthal alignment along the rubbing direction. However, rubbing did not induce azimuthal alignment on clean glass surfaces. For SAMs that were slightly longer (or less degraded), rubbing caused a transformation from homeotropic to aligned tilted anchoring. This effect was observed only close to the transition. For significantly longer or less degraded SAMs, rubbing caused no noticeable change in the polar anchoring angle. These observations suggest that rubbing does not merely choose a particular azimuthal orientation on surfaces which, when unrubbed, induce tilted anchoring. Instead, the rubbing itself is capable of inducing tilt.
Conclusions Both discrete and gradient monolayers exhibited an abrupt transition from homeotropic to tilted alignment suggesting that the transition may correspond to a critical monolayer thickness. Both types of monolayer showed a direct correlation between wettability and chain length/degradation; however, there was an offset in the critical contact angles for the transition between the two types of substrate. This suggested that while wettability can provide a qualitative guide to the surface energy associated with LC anchoring, the LC responds to a different surface property than is measured by contact angle. Mechanical rubbing was shown to induce azimuthal alignment only in the transition region of degraded SAMs and in chain lengths eC5. In most cases, therefore, azimuthal alignment occurred only on substrates where tilt was observed in the absence of rubbing. However, for alignment layers close to the transition, rubbing was found to induce polar tilt; the C5 SAM was a notable example. Acknowledgments This work was supported by the Liquid Crystal Materials Research Center (NSF MRSEC, Award No. DMR-0213918). S.M.M. acknowledges support from a Department of Education GAANN fellowship. The authors are indebted to Joe Maclennan, Renfan Shao, and Art Klittnick for many helpful discussions. LA801322X
