Orientational Behavior of Thermotropic Liquid ... - ACS Publications

In this paper, we report a study of the interactions of nematic liquid crystals with vesicular stomatitis virus (VSV), a member of the Rhabdoviridae f...
0 downloads 0 Views 1MB Size
Download PDF
Langmuir 2004, 20, 2375-2385

2375

Orientational Behavior of Thermotropic Liquid Crystals on Surfaces Presenting Electrostatically Bound Vesicular Stomatitis Virus Luis A. Tercero Espinoza, Kate R. Schumann,† Yan-Yeung Luk, Barbara A. Israel,† and Nicholas L. Abbott* Department of Chemical and Biological Engineering, University of WisconsinsMadison, 1415 Engineering Drive, Madison, Wisconsin 53706 Received September 22, 2003. In Final Form: December 23, 2003 We report the orientational behavior of nematic phases of 4-cyano-4′-pentylbiphenyl (5CB) on cationic, anionic, and nonionic surfaces before and after contact of these surfaces with solutions containing the negatively charged vesicular stomatitis virus (VSV). The surfaces were prepared on evaporated films of gold by either adsorption of poly-L-lysine (cationic) or formation of self-assembled monolayers (SAMs) from HS(CH2)2SO3- (anionic) or HS(CH2)11(OCH2CH2)4OH (nonionic). Prior to treatment with virus, we measured the initial orientation of 5CB (∆ ) | - ⊥ > 0) to be parallel to the cationic surfaces (planar anchoring) but perpendicular (homeotropic) after equilibration for 5 days. A similar transition from planar to homeotropic orientation of 5CB was observed on the anionic surfaces. Only planar orientations of 5CB were observed on the nonionic surfaces. Because N-(4-methoxybenzylidene)-4-butylaniline (MBBA, ∆ ) | - ⊥ < 0) exhibited planar alignment on all surfaces, the time-dependent alignment of 5CB on the ionic surfaces is consistent with a dipolar coupling between the 5CB and electrical double layers formed at the ionic interfaces. Treatment of poly-L-lysine-coated gold films (cationic) with purified solutions of VSV containing 108-1010 plaque-forming units per milliliter (pfu/mL) led to the homeotropic alignment of 5CB immediately after contact of 5CB with the surface. In contrast, treatment of anionic surfaces and nonionic surfaces with solutions of VSV containing ∼1010 pfu/mL did not cause immediate homeotropic alignment of 5CB. These results and others suggest that homeotropic alignment of 5CB on cationic surfaces treated with VSV of titer g108 pfu/mL reflects the presence of virus electrostatically bound to these surfaces.

Introduction Several recent studies have exploited the delicate balance of intermolecular forces underlying the anchoring of thermotropic liquid crystals (e.g., 4-cyano-4′-pentylbiphenyl (5CB)) on surfaces to amplify and report a range of molecular interactions, including the presence of proteins specifically bound to functionalized surfaces of gold.1-3 The objective of the study reported herein was to investigate the response of thermotropic liquid crystals to viral particles bound to solid surfaces and to determine if there exists an orientational response of the liquid crystal that permits the presence of bound viral particles to be distinguished from other biological species (e.g., proteins). We hypothesized that the anchoring of liquid crystals on surfaces presenting proteins or viruses might reflect the structural differences between these species. Virus particles have sizes ranging from tens to hundreds of nanometers and are composed of assemblies of proteins and nucleic acids (either RNA or DNA). Hence, viruses are considerably larger and more complex than individual proteins.4,5 Also, in the case of enveloped viruses, the surfaces of the viral particles comprise phospholipid bilayers studded with viral membrane proteins.4 Whereas * To whom correspondence should be addressed. Phone: (608) 265-5278. Fax: (608) 262-5434. E-mail: [email protected] † Department of Pathobiological Science, School of Veterinary Medicine. (1) Gupta, V. K.; Skaife, J. J.; Dubrovsky, T. B.; Abbott, N. L. Science 1998, 279, 2077. (2) Skaife, J. J.; Brake, J. M.; Abbott, N. L. Langmuir 2001, 17, 5448. (3) Luk, Y.-Y.; Tingey, M. L.; Hall, D. J.; Israel, B. A.; Murphy, C. J.; Bertics, P. J.; Abbott, N. L. Langmuir 2003, 19, 1671. (4) Flint, S. J.; Enquist, L. W.; Krug, R. M.; Racaniello, V. R.; Skalka, A. M. Principles of Virology: Molecular Biology, Pathogenesis, and Control; ASM Press: Washington, DC, 2000.

thermotropic liquid crystals on gold and glass surfaces decorated with proteins have been reported to align the liquid crystals parallel to the supporting surfaces,1-3,6-8 surfaces presenting monolayers of phospholipids have been reported to induce planar, tilted, or homeotropic alignment of thermotropic liquid crystals depending on the surface pressure, organization, and tilt of the phospholipids comprising the monolayer.9,10 We note also that studies of the anchoring of 5CB on cells have found homeotropic or near-planar alignment on the surfaces of adhered cells, depending on the type of cell.11 In this paper, we report a study of the interactions of nematic liquid crystals with vesicular stomatitis virus (VSV), a member of the Rhabdoviridae family. VSV is a widely studied virus that has a well-defined structure. The virus causes periodic outbreaks of disease in cattle, horses, and swine in Central and North America.12-14 Its main structural features have been established by electron microscopy.12,15-17 The infectious VS virion is bullet-shaped (5) Lodish, H.; Berk, A.; Zipursky, S. L.; Matsudaira, P.; Baltimore, D.; Darnell, J. Molecular Cell Biology; W. H. Freeman: New York, 2000. (6) Kim, S.-R.; Abbott, N. L. Adv. Mater. 2001, 13, 1445. (7) Kim, S.-R.; Abbott, N. L. Langmuir 2002, 18, 5269. (8) Tingey, M. L.; Luk, Y.-Y.; Abbott, N. L. Adv. Mater. 2002, 14, 1224. Skaife, J. J.; Abbott, N. L. Langmuir 2001, 17, 5595. (9) Fang, J.; Gehlert, U.; Shashidhar, R.; Knobler, C. M. Langmuir 1999, 15, 297. (10) Hiltrop, K.; Stegemeyer, H. Liq. Cryst. Ordered Fluids 1984, 4, 515. (11) Fang, J.; Ma, W.; Selinger, J. V.; Shashidhar, R. Langmuir 2003, 19, 2865. (12) Howatson, A. F. Adv. Virus Res. 1970, 16, 195. (13) de Mattos, C. A.; de Mattos, C. C.; Rupprecht, C. E. Rhabdoviruses. In Fields Virology; Knipe, D. M., Howley, P. M., Eds.; Lippincott Williams & Wilkins: Philadelphia, PA, 2001. (14) Rose, J. K.; Whitt, M. A. Rhabdoviridae: The Viruses and Their Replication. In Fields Virology; Knipe, D. M., Howley, P. M., Eds.; Lippincott Williams & Wilkins: Philadelphia, PA, 2001.

10.1021/la035774i CCC: $27.50 © 2004 American Chemical Society Published on Web 02/17/2004

2376

Langmuir, Vol. 20, No. 6, 2004

Tercero Espinoza et al.

Figure 1. Schematic representation of vesicular stomatitis virus (VSV). The structural components are the following: G, viral glycoprotein (∼1200 per virion, variable); M, matrix protein (∼1800 per virion); P, polymerase subunit (∼466 per virion), N, nucleocapsid protein (∼1250 per virion); L, polymerase subunit (∼50 per virion); a phospholipid bilayer membrane; negative sense RNA genome. P and L proteins, complexed, form the viral RNA-dependent RNA polymerase. Adapted from Flint4 and Rose.14

and consists of an internal helical structure (RNA and nucleocapsid protein) surrounded by a protein matrix and a phospholipid bilayer studded with fine projections (VSV-G protein).12,17 The virion is approximately 173 nm (range 125-205 nm) in length and 75 nm (range 45-85 nm) in diameter.15,17 A schematic illustration of the infectious VS virion is shown in Figure 1. Electrokinetic studies have determined that VSV has an isoelectric range from pH 4.7 to pH 7.0.18 In the study reported in this paper, we immobilized VSV on surfaces by taking advantage of the net negative surface charge of VSV at slightly basic solution pHs. We prepared cationic surfaces by adsorption of poly-L-lysine from aqueous solutions onto the surfaces of gold films. We exploited electrostatic interactions to capture the virus particles onto these surfaces. We further examined the interactions of virus particles with anionic and neutral surfaces. The anionic surfaces comprised SAMs formed from 2-mercaptoethanesulfonic acid (HS(CH2)2SO3-Na+), whereas the neutral surfaces comprised SAMs formed from 1-mercaptoundec-11-yltetra(ethylene glycol) (HS(CH2)11(OCH2CH2)4OH). Although electrostatic capture of VSV lacks the specificity of antibody-decorated surfaces, such specificity was not required in this study because the principal goal of the study was to understand how nematic liquid crystals orient in response to virions bound to surfaces and to compare this response to that observed for proteins. We did not seek to selectively capture viruses from mixtures onto surfaces. We also note that cationic surfaces have been used to concentrate and purify viruses engineered to express vesicular stomatitis virus glycoprotein (VSV-G) on their surfaces.19 Materials and Methods Materials. Aluminum oxide and glass microscope slides (Fisher’s Finest, premium grade) were purchased from Fisher Scientific (Pittsburgh, PA). The nematic liquid crystal 4-cyano4′-pentylbiphenyl (5CB), manufactured by BDH, was purchased from EM Industries (Hawthorne, NY). Ethanol was purchased from AAPER Alcohol and Chemical Co. (Shelbyville, KY). Antibiotic-antimycotic solution, minimum essential medium Eagle (MEM) (with Earle’s salts and L-glutamine), nonessential amino acid solution, and Hank’s balanced salt solution (HBSS) were purchased from Mediatech, Inc. (Herndon, VA). Heat-inactivated fetal bovine serum was purchased from (15) Bradish, C. J.; Kirkham, J. B. J. Gen. Microbiol. 1966, 44, 359. (16) (a) Chow, T. L.; Chow, F. H.; Hanson, R. P. J. Bacteriol. 1954, 68, 724. (b) Howatson, A. F.; Whitmore, G. F. Virology 1962, 16, 466. (c) Reczko, E. Arch. Gesamte Virusforsch. 1961, 10, 588. (17) Nakai, T.; Howatson, A. F. Virology 1968, 35, 268. (18) Miki, T. Microbiol. Immunol. 1981, 25, 585. (19) (a) Scherr, M.; Battmer, K.; Blomer, U.; Schiedlmeier, B.; Ganser, A.; Grez, M.; Eder, M. Blood 2002, 99, 709. (b) Yamada, K.; McCarty, D. M.; Madden, V. J.; Walsh, C. E. BioTechniques 2003, 34, 1074.

GibcoBRL (Carlsbad, CA). Poly-L-lysine (0.1% w/v, aqueous) was purchased from Ted Pella, Inc. (Redding, CA). Phosphate-buffered saline (PBS), crystal violet, and formaldehyde were purchased from Sigma-Aldrich Co. (St. Louis, MO). Octyltrichlorosilane, n-heptane, 2-mercaptoethanesulfonic acid (sodium salt) (HS(CH2)2SO3-Na+), decanethiol, hexadecanethiol, and the nematic liquid crystal N-(4-methoxybenzylidene)-4-butylaniline (MBBA) were purchased from Aldrich Chemicals (Milwaukee, WI) and used as received. HS(CH2)11(OCH2CH2)4OH was synthesized using procedures reported by Whitesides and coworkers.20 Cell Culture. Vero cells (CCL-81, ATCC, Manassas, VA) were grown using minimum essential medium Eagle (MEM), with Earle’s salts and L-glutamine, supplemented with heat-inactivated fetal bovine serum, antibiotic-antimycotic solution, and a solution containing nonessential amino acids for MEM. Cell culture was carried out using polystyrene flasks with a growth area of 150 cm2 to allow for the cultivation of ∼2 × 107 cells per flask. The cells were infected with VSV Indiana strain (VR-1238, ATCC, Manassas, VA) using a proportion of approximately one infectious virus particle per 100 cells. After incubation for ∼48 h at 37 °C (100% relative humidity, 5.5% CO2), the supernatant was collected for purification. Purification of Virus. The supernatant from infected Vero cells was collected and spun at 10 500 rpm in a Beckman J2-21M centrifuge (Beckman, Fullerton, CA) fitted with a JA 20 rotor for 30 min at 4 °C. The supernatant was collected and spun again at 27 500 rpm for 1 h at 4 °C onto a 30% “sucrose cushion” using a Beckman L8-70M centrifuge fitted with an SW 28 rotor. The “sucrose cushion” (2-3 mL of 30% w/v solution of sucrose in Tris buffer saline (TBS), 10 mM Tris-NaCl, 0.1 M NaCl, 1 mM EDTA, pH 8.0) helped to separate the viral particles, which pellet through the cushion, from cellular fragments that stay in solution after the first centrifugation step and that do not readily penetrate the cushion. After the sample was spun, the supernatant was discarded and the remaining pellet was resuspended in TBS buffer. This solution was then stored at -80 °C in working aliquots until needed. We prepared control solutions using the procedure described above except that supernatant from uninfected Vero cells was used as the starting material. Approximately equal numbers of cells were used to produce the virus and control solutions. Determination of Virus Titer. The titer or concentration of infectious virus particles in solution was determined by means of a plaque assay, as described elsewhere.21 Briefly, Vero cells were grown in six-well plates (Corning, Acton, MA) until a monolayer of cells was formed. The cells were rinsed twice with HBSS, and then 100 µL of serial dilutions of the virus solution was added to each well and incubated for ∼1 h at room temperature in a rocker. The virus solutions were then removed, and the cells were covered with a 1:1 mixture of 1.5% agar and MEM. After ∼5 min, the mixture solidified and the samples were placed in an incubator (37 °C, 100% relative humidity, 5.5% CO2) (20) Pale-Grosdemange, C.; Simon, E. S.; Prime, K. L.; Whitesides, G. M. J. Am. Chem. Soc. 1991, 113, 12. (21) Burleson, F. G.; Chambers, T. M.; Wiedbrauk, D. L. Virology: A Laboratory Manual; Academic Press: San Diego, CA, 1992.

Behavior of Liquid Crystals on Surfaces

Langmuir, Vol. 20, No. 6, 2004 2377

for 4 days. After the 4-day incubation, ∼ 1 mL of crystal violet stain was added to each well and incubated overnight at room temperature. After incubation, the plugs of agar were removed with a stream of water and each well was drained of any residual water. Since the agar overlay restricts the spread of progeny virus to surrounding cells, localized areas of cell death are formed in the monolayer. These areas are called “plaques” and are not stained by crystal violet, allowing for direct visualization of the sites of infection. The plaques were then counted, and the titer of the solution was calculated according to the relation

titer )

Nplaques (d)(V)

(1)

where the titer is given in plaque-forming units per milliliter (pfu/mL), Nplaques refers to the number of plaques in an individual well, d is the dilution of the virus solution used to inoculate the well (e.g., 10-6), and V is the volume of solution used for inoculation in milliliters. For statistical reasons, dilutions that yielded 20-100 plaques per well were used for calculation of the titer. Glass Cleaning. Glass slides were immersed in piranha solution (70:30 v/v % H2SO4/H2O2) at 60-80 °C for 45 min to 1 h. Warning: piranha solution reacts strongly with organic compounds and should be handled with extreme caution; do not store the solution in closed containers. The slides were then rinsed at least 10 times with 18.2 MΩ‚cm water (Milli-Qplus, Millipore, Bedford, MA) and immersed in a base bath (70:30 v/v % 3 M KOH/H2O2) at 60-80 °C for approximately 30 min. The slides were then rinsed sequentially with 18.2 MΩ‚cm water, ethanol (200 proof), and methanol. The slides were thoroughly dried under a stream of nitrogen and placed overnight in an oven at 110 °C. Electron Beam Deposition of Metals. Thin films of gold were deposited onto clean glass slides using a thin layer (∼80 Å) of titanium to promote adhesion between the gold and glass (metal purity 99.999%, International Advanced Materials, New York, NY). The process was carried out at a maximum pressure of 4 × 10-6 Torr in a VES-3000C electron beam evaporator manufactured by Tek-Vac Industries (Brentwood, NY). The rate of deposition was 0.02 nm/s and was controlled by a quartz crystal microbalance (QCM). “Uniformly deposited” films of gold were prepared by rotating the slide holders in an epicyclical manner, thereby removing all preference in the direction of deposition of the metals. “Obliquely deposited” gold films were prepared by using a fixed angle of incidence of the metals and stationary substrates. The angle of incidence used in this study was 60°, measured from the normal of the substrate. “Obliquely deposited” gold films with thicknesses of ∼200 Å were used as substrates for optical cells, while “uniformly deposited” gold films with thicknesses of ∼500 Å were used as a substrate for ellipsometric measurements. Optical Cells Containing Liquid Crystal. The orientational behavior of liquid crystals on surfaces supporting bound VSV was determined by preparing optical cells that were filled with liquid crystal (LC cells). The LC cells were formed by spacing apart two solid substrates using Saran wrap (thickness of ∼13 µm). One of the substrates was a glass slide treated with octyltrichlorosilane to give homeotropic alignment of the liquid crystal. The other substrate was the surface of interest. The LC cell was placed in an oven set to 55-60 °C for ∼5 min; 5CB heated above its clearing temperature (∼35 °C) was then introduced into the LC cell and allowed to cool to room temperature. Alternatively, MBBA, heated above its clearing temperature (∼43 °C), was used instead of 5CB. Optical Examination of Liquid Crystals. The orientations of 5CB were examined by using plane-polarized light in transmission mode on an Olympus BX60 microscope with crossed polarizers. The LC cells were placed on a rotating stage located between the polarizers. In-plane birefringence was determined by rotating the stage by 45° and observing the extent of modulation in the intensity of transmitted light. Homeotropic alignment was determined by first observing no transmission of light during a 360° rotation of the stage. Conoscopy was used to confirm the homeotropic alignment of the liquid crystal. Conoscopy was performed by the insertion of a condenser below the

sample stage and a Bertrand lens above the stage. The condenser produced convergent incident light, and the Bertrand lens permitted imaging of the back focal plane of the microscope. When a liquid crystal was homeotropically aligned, an interference pattern consisting of two crossed isogyres was observed in the back focal plane.22 The isogyres were parallel to the crossed polars. The reader is referred to ref 22 for a detailed discussion of conoscopy. Optical images were captured using a digital camera (Olympus C2020 Zoom) mounted on the microscope. Treatment of Glass Microscope Slides with Octyltrichlorosilane. A solution of n-heptane was passed through a column of aluminum oxide to remove residual water from n-heptane. Glass slides cleaned with piranha solution were immersed in a 10 mM solution of octyltrichlorosilane in n-heptane for ∼30 min at room temperature, rinsed with methylene chloride, and dried under a stream of nitrogen. We confirmed that each glass slide treated with octyltrichlorosilane caused homeotropic orientation of 5CB. Any slide not causing homeotropic orientation of 5CB was discarded. Deposition of Poly-L-lysine. Poly-L-lysine was deposited onto the gold surfaces from aqueous drops of 0.1% poly-L-lysine. The drops were incubated in a water-saturated environment for ∼30 min. The surfaces were then rinsed with water and dried under a stream of nitrogen. Alternatively, poly-L-lysine-treated gold films were rinsed with water and submerged in PBS buffer (Sigma, St. Louis, MO) for storage. Prior to use, the stored gold films were rinsed again with water. Self-Assembled Monolayers (SAMs). SAMs formed from HS(CH2)2SO3Na, HS(CH2)11(OCH2CH2)4OH, HS(CH2)9CH3, or HS(CH2)15CH3 were prepared by soaking gold substrates in ethanolic solutions containing the corresponding alkanethiols (1 mM) overnight. The SAMs were then rinsed with ethanol and dried under a stream of nitrogen prior use. Deposition of Vesicular Stomatitis Virus (VSV). PolyL-lysine-treated gold substrates were contacted with solutions of VSV for about 4-6 h at room temperature by placement of a 10 µL drop of virus solution on the surface. Incubation took place in a water-saturated environment to prevent evaporation. For these experiments, stock virus solutions with titers of 2.5 × 109 and 2 × 1010 pfu/mL were used, as determined by plaque titration (see above); control solutions were prepared as described above. Ellipsometry. Ellipsometry was performed using uniformly deposited films of gold with thicknesses of 500 Å because these gold films are optically reflective. Ellipsometric constants were measured at three locations on each sample using a Rudolph Research Auto EL II ellipsometer (wavelength 632 nm, angle of incidence 70°, Rudolph Technologies, Flanders, NJ). The ellipsometric constants for the gold surfaces were determined before each experiment by using decanethiol and hexadecanethiol selfassembled monolayers (SAMs) as standards of known thickness (∼15 and ∼23 Å, respectively).23 A simple slab model was used to interpret the ellipsometric constants of the SAMs and polyL-lysine films formed on the gold surfaces. The slab was assumed to have an index of refraction of 1.46.

Results and Discussion Characterization of Capture Surfaces. First, we sought to fabricate cationic, anionic, and nonionic surfaces for the capture of VSV. For this purpose, we chose to use poly-L-lysine (cationic), a sulfonate-terminated alkanethiol (anionic), and a tetra(ethylene glycol)-terminated alkanethiol (nonionic) to functionalize the gold surfaces. For the last two surfaces, we used procedures that have been reported previously to form SAMs of these compounds on the surfaces of films of gold.3 Here, we report the characterization of gold surfaces that were incubated with solutions of poly-L-lysine. Past studies have reported that proteins and polypeptides adsorb onto the surfaces of gold films.24,25 (22) Hartshorne, N. H.; Stuart, A. Crystals and the Polarizing Microscope; Edward Arnold & Co.: New York, 1970. (23) Miller, W.; Abbott, N. L. Langmuir 1997, 13, 7106. (24) Du, Y. J.; Cornelius, R. M.; Brash, J. L. Colloids Surf., B 2000, 17, 59.

2378

Langmuir, Vol. 20, No. 6, 2004

Tercero Espinoza et al.

Figure 2. Geometry of the asymmetric liquid crystal cell comprising octyltrichlorosilane-treated glass and poly-L-lysinetreated gold film. The director field of 5CB is shown schematically between the surfaces.

Visual inspection of the behavior of water on gold surfaces treated with poly-L-lysine revealed the treated surfaces to be hydrophilic. In contrast, untreated slides were hydrophobic (when measured 1 h or more after removal from the evaporator) because adventitious adsorbates collect on the gold (see below). We used ellipsometry to determine the optical thickness of poly-L-lysine adsorbed on the gold films. Because of the presence of the adventitious adsorbates on the gold surfaces, our measurements of optical thicknesses reflect both the adsorbed contaminants and adsorbed poly-L-lysine. We measured the ellipsometric thickness of the poly-L-lysine layer (plus adventitious adsorbates) to be ∼1.7 nm. Prior to adsorption of poly-L-lysine, the ellipsometric thickness of the adventitious adsorbates was ∼1.0 nm. Thus, we conclude that the ellipsometric thickness of the layer of poly-L-lysine is ∼0.7 nm and that poly-L-lysine adsorbs onto the surface of gold from aqueous solutions of poly-L-lysine. The ellipsometric results described above were obtained when using gold films with ages (measured from the time of removal from the evaporator) that were between 1 h and 2 weeks. Anchoring of Thermotropic Liquid Crystals on Cationic Surfaces at Short Times. Because the nonionic surfaces we used are known to induce a planar alignment of 5CB,3 we explored the alignment of 5CB on the cationic and anionic surfaces prepared on obliquely deposited gold films. First, we prepared cationic surfaces by adsorption of poly-L-lysine to gold surfaces as described above. We prepared optical cells by pairing the gold surfaces treated with poly-L-lysine with glass slides treated with octyltrichlorosilane. The glass slides treated with octyltrichlorosilane induce homeotropic alignment of thermotropic liquid crystals. The surfaces were spaced apart by using thin films of Saran wrap (thickness of ∼13 µm) and secured together with binder clips. Hereafter, we refer to an optical cell prepared by pairing the surface of interest with a slide of octyltrichlorosilane-treated glass as an “asymmetric optical cell” (Figure 2). Next, we introduced 5CB above its clearing temperature into the optical cell and imaged the optical texture of the liquid crystal in transmission through crossed polarizers after it cooled into its nematic phase. Our initial observations were made within 10 min of filling the optical cells with liquid crystal. We observed a uniformly dark optical texture, except for line defects, when the sample was oriented with the direction of gold deposition parallel to either of the polarizers (Figure 3a). Upon rotation of the sample by 45°, the optical texture appeared bright, and we observed domains of two distinct colors (Figure 3b). The domains (25) Holmstrom, N.; Askendal, A.; Tengvall, P. Colloids Surf., B 1998, 11, 265.

Figure 3. Optical textures of 5CB on poly-L-lysine-treated gold films (obliquely deposited) (a, b) and on poly-L-lysinetreated gold films incubated in TBS buffer for >3 h (c-f). The images (cross polars) were obtained with the direction of deposition of the gold film oriented at an angle of 0° (a, c, e) or 45° (b, d, f) from one of the polarizers. Scale bars ) 500 µm.

were encircled by the line defects. These defects were also visible when the second polarizer (analyzer) was removed from the light path in the microscope, indicating that the defects correspond to localized regions where the nematic order was lost, causing scattering of light. The uniformly dark texture observed when the sample was oriented parallel to one of the polarizers, when combined with the observation of strong modulation of light upon rotation of the sample between crossed polarizers, indicates that the nematic director points in a uniform azimuthal direction throughout the sample and that the alignment is either parallel to the surface (planar alignment) or tilted from it. We attributed the appearance of domains of two distinct colors to two distinct tilt angles of the nematic director on the surface of the treated gold. These colors are close to each other in a Michel-Le´vy color chart. On the basis of calculations discussed below, we concluded that the tilt angles differ from each other by less than 5°. Because the capture of viruses on these surfaces required the contact of the surfaces with solutions of VSV in TBS buffer for several hours, we investigated the effect that incubation of the poly-L-lysine-coated gold surfaces in TBS buffer had on the alignment of 5CB on those surfaces. We prepared cationic surfaces as described above and immersed them in TBS buffer for ∼3 h at room

Behavior of Liquid Crystals on Surfaces

temperature. We rinsed the surfaces with water, dried them under a stream of nitrogen, and then prepared asymmetric optical cells as described above. We imaged the optical cells under crossed polarizers shortly after contacting the cationic surfaces with 5CB. When a sample was oriented with the direction of gold deposition parallel to one of the polarizers, we observed a uniformly dark texture with the exception of line defects (Figure 3c). Most, but not all, defects formed closed loops of irregular shape and varying size. Upon rotation of the sample by 45° under crossed polarizers, we observed a bright texture consisting of two colors separated almost always by line defects (Figure 3d). The colors lie next to each other in a MichelLe´vy chart and indicate that the domains possessed tilt angles that differed by less than 5° (see below for details). These optical textures are very similar to those observed for surfaces of gold treated with poly-L-lysine that had not been contacted with TBS buffer (Figure 3a,b). We note that we observed some variability in the shape and abundance of line defects on surfaces of gold treated with poly-L-lysine, independent of whether they had or had not been contacted with TBS buffer. For example, the defects observed in Figure 3e did not form closed loops. When the sample was rotated by 45° under crossed polarizers, we observed a bright texture of a single color across the entire sample (Figure 3f). In contrast to the textures shown in parts b and d of Figure 3, the color was the same for the inside and outside areas bounded by line defects. We also note that the interference colors in Figure 3 depend on the thickness of the film of liquid crystal in each optical cell. Thus, it is not possible to directly compare interference colors between samples. However, colors can be compared within the same optical cell. Abrupt changes in color, like those seen in parts b and d of Figure 3, indicate two different tilt angles of the nematic director, but a gradual change in color across an optical cell most likely reflects small changes in the thickness of the liquid crystal film across the sample. To determine if 5CB assumes planar alignment on at least some regions of the surfaces of gold treated with poly-L-lysine, we prepared asymmetric optical cells with a continuously varying thickness of the film of liquid crystal across the length of the sample (wedge cells, Figure 4a). As before, the top surface of the optical cell was prepared from octyltrichlorosilane-treated glass. The bottom surface of the optical cell was a gold surface patterned along the length of the sample (parts b and c of Figure 4). The pattern was made by first submerging half of the gold slide in an ethanolic solution of nhexadecanethiol for ∼2 h. The sample was then rinsed with ethanol and dried under a stream of nitrogen. SAMs formed from n-hexadecanethiol on the surfaces of gold films are known to cause planar alignment of 5CB.26 Next, we incubated the other half of the gold film with polyL-lysine solution for ∼30 min; the sample was then rinsed with water and dried under a stream of nitrogen. The poly-L-lysine solution did not contact the SAM formed from hexadecanethiol. We compared the progression of the interference colors generated by 5CB across the two areas of the optical cell. By patterning poly-L-lysine and hexadecanethiol on the same substrate, we ensured identical thicknesses of the film of liquid crystal in the two regions of surface (Figure 4). Thus, the same progression of interference colors across the sample would imply the same anchoring of liquid crystals on the two areas of the surface. Because the length (26) (a) Drawhorn, R. A.; Abbott, N. L. J. Phys. Chem. 1995, 99, 16511. (b) Gupta, V. K.; Abbott, N. L. Langmuir 1996, 12, 2587.

Langmuir, Vol. 20, No. 6, 2004 2379

Figure 4. (a) Side view of the patterned wedge cell and top view of the gold film patterned with areas of n-hexadecanethiol and poly-L-lysine. Shown are optical textures of 5CB in the wedge cell in regions (b) treated with n-hexadecanethiol and (c) poly-L-lysine. The images of the liquid crystal cells were obtained within 10 min of filling the cell with 5CB. Scale bar ) 1 mm.

of the sample was greater than the field of view of our microscope, we obtained a series of overlapping images along the length of the sample and assembled them into a composite image. The images shown in Figure 4b,c were obtained from areas fully contained within each patterned area. Comparison of the progression of interference colors along the length of the slide revealed that they were nearly identical. Because SAMs formed from hexadecanethiol cause planar anchoring of 5CB, this result leads us to conclude that gold films treated with poly-L-lysine cause planar anchoring of 5CB immediately after contact of the surface and the liquid crystal. Anchoring of Thermotropic Liquid Crystals on Cationic Surfaces at Long Times. Whereas the results above demonstrate that gold surfaces treated with polyL-lysine cause planar or near-planar anchoring of 5CB shortly after contact of the liquid crystal with the surface, we observed homeotropic alignment of 5CB when the samples were examined a week later. Thus, below we characterize this time-dependent behavior of 5CB on gold surfaces treated with poly-L-lysine. Figure 5a shows a series of images of a sample (asymmetric optical cell) oriented 45° from either polarizer (crossed polarizers). During the first day of observation, the sample exhibited a pale-orange to pale-green color. These colors changed very little during the first 2 days. By the third day, however, the colors had changed to lightblue and yellow. By the fourth day, the color of the sample had turned gray. On the fifth day, the sample was dark and exhibited no modulation of light upon rotation under crossed polarizers. By using conoscopy, we observed an interference pattern consisting of two crossed isogyres, which confirmed the alignment to be homeotropic. We interpreted the progression of interference colors to indicate a continuous transition from planar to homeotropic anchoring on the surface of gold treated with polyL-lysine. Past studies have demonstrated that it is possible to quantitatively relate the interference colors generated by the liquid crystal to the tilt angle on one of the surfaces comprising the LC cell, provided the anchoring at the second surface and the thickness of the film of liquid crystal are known.27 Briefly, starting from the description of

2380

Langmuir, Vol. 20, No. 6, 2004

Tercero Espinoza et al.

Figure 5. Temporal evolution of the optical textures of (a) 5CB and (c) MBBA on poly-L-lysine-treated gold films. (b) Progression of tilt angle corresponding to textures in (a). Scale bars ) 250 µm.

effective birefringence, ∆neff, of a tilted material,27-29

∆neff )

n|n⊥

xn|2 cos2 θ + n⊥2 sin2 θ

- n⊥

(2)

where n| is the refractive index parallel to the director, n⊥ is the refractive index perpendicular to the director, and θ is the tilt of the liquid crystal, and by noting that minimization of the elastic energy stored in the splay and bend distortion of the liquid crystal (Figure 2) leads to a linear variation of the tilt angle across the thickness of the sample,27

x θ(x) ) θ2 ) x˜ θ2 h

(3)

we estimate the effective birefringence of the sample as

∆neff ≈

∫01

(x

n|n⊥

n⊥2 sin2(x˜ θ2) + n|2 cos2(x˜ θ2)

)

- n⊥ dx˜ (4)

where θ2 is the tilt angle of the liquid crystal at the polylysine-treated surface. From knowledge of the optical properties of 5CB27 and the geometry of the sample, we (27) Brake, J. M.; Mezera, A. D.; Abbott, N. L. Langmuir 2003, 19, 8629. (28) Van Doorn, C. Z.; Gerritsma, C. J.; de Klerk, J. J. M. J. Influence of the Device Parameters on the Performance of Twisted-Nematic LiquidCrystal Matrix Displays. In The Physics and Chemistry of Liquid Crystal Devices; Sprokel, G. J., Ed.; Plenum Press: New York, 1980. (29) Bloss, F. D. An Introduction to the Methods of Optical Crystallography; Holt, Rinehart and Winston: New York, 1961.

estimated the tilt angle of 5CB at the surface of gold treated with poly-L-lysine by determining the birefringence corresponding to the colors in Figure 5 (using a MichelLe´vy chart) and then solving for θ2 using eq 3. The results are shown in Figure 5b. The results reported above reveal that poly-L-lysine causes homeotropic alignment of 5CB after 5 days of contact of the liquid crystal with the surface. A past study using gold surfaces supporting sodium carboxylate salts has demonstrated that the electric field of electrical double layers formed by dissociation of surface-immobilized salts can influence the orientations of supported nematic liquid crystals.30 In that study, the authors reported planar alignment of 5CB (∆ ) | - ⊥ > 0, where | is the dielectric constant parallel to the director and ⊥ is the dielectric constant perpendicular to the director) on surfaces presenting low densities of sodium carboxylate salts but homeotropic alignment on surfaces presenting high densities of sodium carboxylate groups. This observation and other past studies31 are consistent with reorientation of the liquid crystals under the influence of an electric field normal to the surface. Planar alignment of MBBA (∆ ) | - ⊥ < 0) on gold surfaces presenting low and high densities of sodium carboxylate groups was also observed, consistent with the larger  (⊥) aligning parallel to the electric field. We hypothesized that the time-dependent behavior of the orientation of 5CB on surfaces of gold decorated with poly-L-lysine may also arise from the formation of an electrical double layer. To test this proposition, we prepared asymmetric optical cells as described above, (30) Shah, R. R.; Abbott, N. L. J. Phys. Chem. B 2001, 105, 4936. (31) Luk, Y.-Y.; Abbott, N. L. Science 2003, 301, 623.

Behavior of Liquid Crystals on Surfaces

Figure 6. Optical textures of nematic liquid crystals with positive (5CB) and negative (MBBA) dielectric anisotropy supported on SAMs formed from HS(CH2)2SO3-Na+ (anionic surfaces). Shown are images of 5CB obtained shortly (a) and 6 h (b) after assembly of optical cells and images of MBBA obtained shortly (c) and 6 h (b) after assembly of optical cells. All images (cross polars) were obtained with the direction of deposition of the gold film oriented at an angle of 45° from one of the polarizers. Scale bars ) 500 µm.

substituting MBBA for 5CB, and observed the evolution of the optical textures of MBBA over a period of 5 days (Figure 5c). Initially (on day 1), we observed a uniformly dark texture virtually free of line defects when the sample was oriented parallel to one of the polarizers (crossed polarizers). As in the case of 5CB, we observed strong modulation of light upon rotation of the sample under crossed polarizers. We measured the interference colors to remain virtually unchanged over the course of 5 days, indicating that the anchoring of MBBA remained planar. This observation is consistent with the formation of an electrical double layer on the surface of gold treated with poly-L-lysine because the presence of an electric field perpendicular to the surface would favor planar alignment of MBBA, in which the larger  (⊥) is aligned parallel to the electric field in the electrical double layer. Finally, we note that Fang and co-workers have reported homeotropic anchoring of 5CB on glass substrates coated with polyD-lysine, although the time dependence of the anchoring was not discussed.11 Anchoring of Thermotropic Liquid Crystals on Anionic Surfaces. Next, we characterized the anchoring of 5CB on the anionic surfaces prior to treatment of VSV. Anionic surfaces were formed by soaking gold substrates in a 2 mM ethanolic solution of HS(CH2)2SO3-Na+. We then assembled asymmetric optical cells with the anionic surfaces and examined them under crossed polarizers immediately (