Azobenzene-Modified Poly(l-glutamic acid) (AZOPLGA): Its

Zahid Mahimwalla , Kevin G. Yager , Jun-ichi Mamiya , Atsushi Shishido , Arri Priimagi , Christopher J. Barrett. Polymer Bulletin 2012 69 (8), 967-100...
0 downloads 0 Views 184KB Size
Download PDF
Biomacromolecules 2003, 4, 366-371

366

Azobenzene-Modified Poly(L-glutamic acid) (AZOPLGA): Its Conformational and Photodynamic Properties† Suizhou Yang, Lian Li, Ashok L. Cholli, Jayant Kumar,* and Sukant K. Tripathy‡ Center for Advanced Materials, Departments of Chemistry and Physics, University of Massachusetts Lowell, Lowell, Massachusetts 01854 Received September 24, 2002

Azobenzene-modified poly(L-glutamic acid) (AZOPLGA) polymers with 22 and 35 mol % of azo chromophores in the side chains have been synthesized by condensing 4-methoxy-4′-aminoazobenzene and poly(L-glutamic acid). These polymers have been characterized by NMR, FT-IR, and UV-visible spectroscopic techniques. The conformational features of the polymer backbone chains in the films that were cast from the polymer solutions prepared in different solvents have been investigated by circular dichroism spectroscopy. Experimental data suggested that the thermal cis-trans relaxation and photoinduced birefringence, which are related to the azo chromophores in the side chains of polymer, are not affected by the conformations of polymer backbones. However, the modulations of the surface relief gratings, the result of photoinduced mass transport process, recorded on these polymers are sensitive to polymer main chain conformation, as well as the degree of functionalization. Introduction It is a well-known phenomenon that azobenzene chromophores upon irradiation with polarized light will undergo photoinduced reorientation through trans-cis-trans isomerization. This conformational cycling can induce the polymer main chain to move and result in the formation of surface relief gratings (SRG) on azobenzene-functionalized polymers at temperatures well below the glass transition temperature (Tg). This phenomenon has been observed on the surfaces of a number of azobenzene polymers, such as epoxy polymers, polyacrylates, polyesters, conjugated polymers, polyazophenol, and azocellulose polymers.1-8 Poly(L-glutamic acid) (PLGA) is a polypeptide and may exist in random coil, R-helix, or β-sheet conformations depending on the environmental conditions. In the random coil state, the polymer main chains have no preferred orientation. On the other hand, the polymer main chains are tightly coiled through intramolecular hydrogen bonding in the R-helix conformation, a rodlike structure. 9 In the case of β-sheet, the polymer main chains are almost fully extended and arrange into pleated structure by intermolecular hydrogen bonding.10 In R-helix and β-sheet conformations, the polymer backbones are tightly hydrogen-bonded and the motions of polymer main chain segments are severely restricted. In contrast, the main chain segments in the random coil conformation are relatively free to move. Thus, conformations of polymer main chains are expected to have a role in the photoinduced mass transport process. In this paper, azobenzene-modified poly(L-glutamic acid) (AZOPLGA) polymers were synthesized by linking 4-methoxy-4′-aminoazobenzene to poly(L-glutamic acid) (PLGA). The struc†

Dedicated in memory of Professor S. K. Tripathy. * To whom corresponding should be addressed. ‡ Deceased.

tural characterization was carried out using FT-IR, UVvis, and NMR spectroscopic techniques. The backbone conformations of AZOPLGA in the thin films that were cast from the polymer solutions prepared in different solvents were determined by circular dichroism (CD) spectroscopy. The role of conformation of polymer backbone in the films on the efficiency of inscribing SRGs was investigated for the first time. Experimental Section Materials. p-Anisidine, sodium nitrite, methanol, trifluoroacetic acid (TFA), pyridine, sodium formaldehyde bisulfite, propionic acid, dicyclohexylcarbodiimide (DCCI), and Nhydroxybenzotriazole (HOBT) were purchased from Aldrich Chemical Co. and were used as received. Anhydrous N,Ndimethylformamide (DMF) was purchased from Fisher Scientific Co. and kept over 3 Å molecular sieves and anhydrous magnesium sulfide. Poly(L-glutamic acid) sodium salt (MW ) 58 100, Mw/Mn ) 1.26) was purchased from Sigma company. It was dissolved in water and protonated with 1 M HCl. The precipitated poly(L-glutamic acid) (PLGA) was then centrifuged, separated, and vacuum-dried. Synthesis of 4-Methoxy-4′-aminoazobenzene. A 40 mL aqueous solution containing sodium nitrite (4.14 g, 0.06 mol) was added dropwise to a suspension of p-anisidine (7.40 g, 0.06 mol) in 10 mL of concentrated HCl and 40 mL of water, which was cooled in an ice bath. The reaction mixture was stirred for 1 h below 5 °C and then treated with sulfamic acid to eliminate any excess nitrite ions. Aniline (22.0 g, 0.23 mol) was added to a solution of sodium formaldehyde bisulfite (33.2 g, 0.23 mol) in 100 mL of water with stirring. The mixture was allowed to stand at 65 °C for 2 h. White needles of omega acid or sodium (anilinomethylene) sulfonate, which separated out upon

10.1021/bm025687p CCC: $25.00 © 2003 American Chemical Society Published on Web 01/03/2003

Biomacromolecules, Vol. 4, No. 2, 2003 367

Azobenzene-Modified Poly(L-glutamic acid)

cooling, were collected, washed with cold water and ether, successively, and dried under vacuum. Yield was 32.0 g (66.5%). The diazonium salt solution was added to an aqueous solution of sodium acetate (0.90 M, 150 mL) containing omega acid (16.2 g, 0.074 mol) over a period of 30 min under cooling in an ice bath. The reaction mixture was allowed to stand overnight below 4 °C. The orange precipitate was collected, washed with 5% NaCl aqueous solution, and dried in a vacuum. The dry orange powder was added into toluene (75 mL) containing NaOH (4.0 g). The mixture was heated to reflux for 4 h and filtered. The filtrate was heated to remove solvent by rotary evaporator. Then the residue was dissolved into acetone and passed through an active alumina column using toluene as eluent. The orange product with a melting point of 135-137 °C was obtained by evaporating toluene. Yield was 5.50 g (40%). 1H NMR (DMSO-d6): 3.83 (s, 3H), 6.00 (s, 2H), 6.66 (m, 2H), 7.06 (m, 2H), 7.62 (m, 2H), and 7.74 ppm (m, 2H). 13C NMR (DMSO-d6): 55.36, 113.28, 114.24, 123.26, 124.53, 142.67, 146.44, 152.07, and 160.36 ppm. Synthesis of Model Compound 4-Propionamido-4′methoxyazobenzene. 4-Methoxy-4′-aminoazobenzene (1.78 g, 7.83 mmol) and propionic acid (8.5 mL) were dissolved in anhydrous dimethylformide (DMF) (50 mL). Then dicyclohexylcarbodiimide (DCCI) (1.63 g) and N-hydroxybenzotriazole (HOBT) (1.08 g) were added into the above solution, which was cooled in ice bath. The mixture was stirred at room temperature for 3 days under nitrogen atmosphere. The precipitated N,N′-dicyclohexylurea was filtered from the reaction mixture, and the filtrate was added into 300 mL of 1 N HCl aqueous solution and then filtered. The solid was then dissolved into 30 mL of acetone and added into 300 mL of 1 N NaOH aqueous solution and filtered. The crude product was dried in a vacuum at 60 °C and purified by passing through an active alumina column using toluene as eluent. The final product is a brownishyellow crystalline powder with a melting point of 179-181 °C. Yield was 1.77 g (80%). 1H NMR (DMSO-d6): 1.10 (m, 3H), 2.37 (m, 2H), 3.86 (s, 3H), 7.12 (m, 2H), 7.81 (m, 4H), 7.85 (m, 2H), and 10.17 ppm (s, 1H). 13C NMR (DMSO-d6): 10.42, 30.51, 56.49, 115.23, 120.01, 124.11, 125.10, 142.74, 147.13, 148.29, 162.48, and 173.27 ppm. The molar extinction coefficient (max) of the trans form of this model compound was obtained by measuring the absorbance of solutions with varying concentration in DMF at 365 nm (λmax). The max is 2.23 × 104, which was used to determine the azobenzene chromophore content in the polymers. Synthesis of Azobenzene-Modified Poly(L-glutamic acid) (AZOPLGA). A typical synthesis procedure for AZOPLGA is described here for the experimental conditions set for sample 22% AZOPLGA in Table 1. 4-Methoxy-4′-aminoazobenzene (0.202 g, 0.89 mmol) and PLGA (0.152 g, 1.18 mmol γ-COOH groups) were dissolved in 50 mL of anhydrous DMF in a 250 mL flask cooled in ace bath. DCCI (0.247 g) and HOBT (0.163 g) were added into this solution. The mixture was stirred at room temperature for 1 week under nitrogen atmosphere. The precipitated N,N′-dicyclo-

Table 1. AZOPLGA Polymers Obtained by Modification of PLGA with 4-Methoxy-4′-aminoazobenzene sample

PLGA (g)

dye (g)

Ra

DCCI (g)

HOBT (g)

22% AZOPLGA 35% AZOPLGA

0.152 0.116

0.202 0.253

0.75 1.24

0.247 0.310

0.163 0.205

a

Azo dye/γ-COOH ratio.

hexylurea was filtered from the reaction mixture. The filtrate was then added to ethyl ether (150 mL), and the polymer was precipitated and isolated by centrifuge procedure. Any unreacted azo reagent was removed by repeated dissolution and precipitation using DMF as solvent and ether as nonsolvent. The polymer was finally dissolved in alkaline water and precipitated in acid water and isolated. The final azobenzene polymer product was obtained as a yellow powder after washing with distilled water three times and freeze-drying. Characterization. 1H and 13C NMR spectra were acquired on a Bruker ARX-500 MHz FT-NMR spectrometer. NMR samples were prepared in deuterated dimethyl sulfoxide (DMSO-d6). Infrared spectra were recorded on a PerkinElmer 1720 FT-IR spectrometer. The sample of 4-Methoxy4′-aminoazobenzene was compacted into KBr pellets for recording FT-IR data. In the case of polymers, a thin film was cast on KBr disks from a polymer solution in DMF. Azobenzene-modified PLGA was dissolved separately in trifluoroacetic acid and pyridine. The solutions were cast on glass substrates in a vacuum oven at room temperature, and vacuum was immediately applied for 20 min. Then the films were dried at 85 °C in a vacuum for 12 h. The thickness of the films were measured using a Sloan Dektak IIA profiler and were controlled between 0.5 and 2.0 µm by adjusting the solution concentration. The UV-visible absorption spectra were recorded using a Perkin-Elmer Lambda 9 spectrophotometer. Circular Dichroism Measurements. The circular dichroism (CD) spectra of azobenzene-modified PLGA thin films on fused quartz glass substrates were recorded by JASCO J-810 spectropolarimeter. Typical experimental parameters for collecting CD data were as follows: sensitivity ) 100 mdeg; scan speed ) 50 nm/min; response time ) 8 s; bandwidth ) 2 nm; scan range ) 190-400 nm. Photoinduced Birefringence. The optically induced birefringence in polymer films was measured as described in ref 4. The film on the glass substrate was placed between two crossed polarizers in the path of a low-power unpolarized He-Ne laser beam at 633 nm. The birefringence was induced in the polymer film by a linearly polarized beam at 488 nm from an Ar+ laser with an intensity of 100 mW/cm2. The polarization angle of the Ar+ laser beam was set at 45° with respect to the orientation of the polarizers. An increase in transmission of the He-Ne laser beam resulting from induced birefringence in the polymer film was measured. The birefringence was then calculated from the intensity of the transmission.11 SRG Fabrication. SRGs were written on the polymer film using a simple interferometric setup4 at 488 nm light with an intensity of 200 mW/cm2 and an incidence angle of 14°. The polarization angle of the laser beam was set at 45° with

368

Biomacromolecules, Vol. 4, No. 2, 2003

Yang et al.

Scheme 1. Synthesis of Azobenzene-Modified Poly(L-glutamic acid)

Figure 2. 13C NMR spectra of 4-methoxy-4′-aminoazobenzene and azobenzene chromophore in 22% AZOPLGA. Figure 1.

13C

NMR spectra of PLGA and 22% AZOPLGA.

respect to the s-polarization direction. The SRGs were investigated by atomic force microscopy (AFM). The surface topology and profiles provided the spacing and the surface modulation depth of the surface gratings. Results and Discussions Azobenzene-modified PLGA (AZOPLGA) polymers were synthesized by linking 4-methoxy-4′-aminoazobenzene to poly(L-glutamic acid) (PLGA) in the presence of DCCI and HOBT in DMF (Scheme 1). Table 1 shows the experimental conditions for the synthesis of azobenzene-modified poly(L-glutamic acid). Two samples were obtained with azobenzene chromophore content of 22% (22% AZOPLGA) and 35.0% (35% AZOPLGA). Figure 1 shows the 13C NMR spectra of PLGA and 22% AZOPLGA polymer as a result of reaction (Scheme 1). The assignments of NMR resonance peaks for PLGA were based on the work of Lader et al.12 The most upfield resonance peak at 26.22 ppm is assigned to carbon c (the β carbon of PLGA). The resonance peak for carbon d (the γ carbon of PLGA) appears at 31.32 ppm. The resonance peak at 56.68 ppm is attributed to the carbon b (the R carbon of PLGA).

Carbon a, the carbon of backbone carbonyl group, appears at 174.32 ppm. The most downfield peak at 176.22 is associated with carbon e, the carbon of side chain carboxyl group. In the spectrum 22% AZOPLGA, the peaks a′, b′, c′, d′, and e′ appear as the carbons for the PLGA repeat units, which are linked with azobenzene chromophores. The resonance peaks in the range of 55-165 ppm are attributed to azobenzene chromophores. These peaks are narrow compared with the peaks of PLGA because the azo chromophores are in the side chain and more free to move so the relaxation times are shorter compared to backbone carbons of the polymer. Figure 2 shows the 13C NMR spectra of 4-methoxy-4′aminoazobenzene and the azobenzene chromophore of 22% AZOPLGA. The significant change of the resonance peaks for the carbons in azobenzene chromophores before and after they are linked to PLGA can be identified. There is a 10 ppm difference between the chemical shifts of C9 and C9′. The resonance peaks for C8′ and C6′ are shifted 4 ppm downfield when compared to the resonance peaks for C8 and C6. On the other hand, the resonance peaks of C4′ and C7′ are shifted 0.2 and 0.5 ppm upfield from C4 and C7, respectively. C3′, C5′, and C7′ are shifted 1 ppm downfield from C3, C5, and C7. All of these chemical shift changes

Azobenzene-Modified Poly(L-glutamic acid)

Biomacromolecules, Vol. 4, No. 2, 2003 369

Figure 3. FTIR spectra of PLGA, 22% AZOPLGA, 35% AZOPLGA, and 4-methoxy-4′-aminoazobenzene.

indicate that the azobenzene chromophores are linked to PLGA. The FTIR spectra of PLGA, 22% AZOPLGA, 35% AZOPLGA, and 4-methoxy-4′-aminoazobenzene are shown in Figure 3. As a result of reaction, the characteristic peaks arising from the AZOPLGA polymers can be easily identified in their spectra if the spectrum for AZOPLGA is compared with spectra of PLGA and azo dye. The intensity of the vibration band at 1720 cm-1, which is attributed to the CdO stretching band of carboxylic acid, is decreased when compared with the band in the spectrum of PLGA. The double bands for N-H stretching of primary amino groups13 in the azo dye spectrum at 3454 and 3325 cm-1 changed into a single band in the spectra of AZOPLGA polymers. In addition, the band at 3218 cm-1 in the spectrum of azo dye is attributed to the overtone of the NH bending13 and was significantly decreased in the spectra of AZOPLGA polymers. These observations suggest the formation of amide bond between the carboxylic acid group of side chain of PLGA and the amine group of azo dye. The characteristic peaks for azobenzene chromophores, C-O-C stretching band for alkyl aryl ether at 1250 and 1022 cm-1, aromatic CdC stretching at 1595 and 1499 cm-1, C-N stretching at 1142 cm-1, and the out-of-plane C-H bending of 1,4disubstituted benzene rings at 840 cm-1, also appear in the spectra in the AZOPLGA polymers. The intensities of these peaks increase with azobenzene chromophore content from 22% to 35%. These results further suggest the coupling of azobenzene chromophores to the PLGA chains. Because the polypeptide chains give rise an optical rotation, circular dichroism (CD) has been a sensitive assay for the type of secondary structures in a polypeptide. In the range 205-230 nm of CD spectra, R-helix makes a dominant contribution with two negative bands, while β-sheet has a single negative band. A random coil, by contrast, has a negative band centered at less than 205 nm.14 Figures 4 and 5 show the CD spectra of films of 22% AZOPLGA and 35% AZOPLGA, respectively. The backbone conformations of AZOPLGA in the films can be easily identified by carefully comparing the spectra for AZOPLGA polymers with spectra

Figure 4. Circular dichroism spectra of 22% AZOPLGA films: (solid line) cast from pyridine solution; (dotted line) cast from TFA solution.

Figure 5. Circular dichroism spectra of 35% AZOPLGA films: (solid line) cast from pyridine solution: (dotted line) cast from TFA solution.

of polypeptide in different conformations as shown in ref 14. The spectral features of 22% AZOPLGA film cast from TFA in Figure 4 (dotted line) and 35% AZOPLGA film cast from pyridine (solid line) in Figure 5 are same as R-helix conformation. While the spectrum of 22% AZOPLGA film cast from pyridine in Figure 4 (solid line) is similar to that of the β-form peptide spectrum, the spectrum of 35% AZOPLGA film cast from TFA in Figure 5 (dotted line) shows the random coil conformations.

370

Biomacromolecules, Vol. 4, No. 2, 2003

Yang et al. Table 2. Cis-Trans Thermal Relaxation Constants for AZOPLGA Polymer in Films Cast from Different Solution sample

solvent

conformation λmax (nm) k × 103 (s-1)

22% AZOPLGA 22% AZOPLGA 35% AZOPLGA 35% AZOPLGA

pyridine TFA pyridine TFA

β-sheet R-helix R-helix random coil

355 354 349 348

1.59 1.08 1.03 1.65

Table 3. Photoinduced Birefringence and SRGs Modulation Depth of AZOPLGA Polymer Films sample Figure 6. Cis-trans thermal relaxation spectra of the azo chromophores in the film cast from 22% AZOPLGA pyridine solution. The spectra were recorded every 20 min after irradiation using UV light at 360 nm for 5 min. The top spectrum was recorded before photoexcitation.

Figure 7. ln(A∞ - At) vs t for the film cast from 22% AZOPLGA pyridine solution. A∞ and At are the absorbance (λ ) 355 nm) at infinite time (10 days) and at time t.

When irradiated with UV light of appropriate wavelength, azobenzene chromophores undergo a reversible trans w cis photoisomeriaztion, cis w trans photoreconversion, and cis w trans thermal reconversion.15 The thermal reconversion process is much slower than cis w trans photoreconversion and is often the rate-determining process of repeated trans w cis w trans isomerization cycles.16 Figure 6 shows the typical cis w trans thermal relaxation spectra of the azo chromophores in the film cast from pyridine solution for the 22% AZOPLGA polymer. The intense absorption at 355 nm due to π-π* transition of the trans isomer increases, while the absorption maximum at 470 nm attributed to n-π* transition of cis isomer15 decreases in the thermal reconversion or thermal relaxation process. In the dark, the azobenzene chromophores in the polymer will undergo cis-trans isomerization eventually returning back to the stable trans state. Figure 7 shows a plot of ln(A∞ - At) vs time obtained from a series of thermal relaxation spectra shown in Figure 6 for the 355 nm peak. The plot shows a linear behavior 40 min after the photoexcitation. The fast recovery component in the initial stage is attributed to the relaxation of strained conformations of azobenzene chromophores that are trapped in the rigid glassy matrix.15,16 The cis-trans thermal relaxations of 22% AZOPLGA for the film cast from TFA solution, as well as those of 35% AZOPLGA, show similar features to those presented in Figures 6 and 7. However, the

22% AZOPLGA 22% AZOPLGA 35% AZOPLGA 35% AZOPLGA

solvent conformation pyridine TFA pyridine TFA

β-sheet R-helix R-helix random coil

birefringence modulation (∆n) depth (nm) 0.038 0.039 0.071 0.076

35 13 220 400

maximum wavelengths (λmax) for 35% AZOPLGA have a 4 nm blue shift when compared with those of 22% AZOPLGA. The rate constants for thermal relaxation of these two polymers are shown in Table 2. These rate constants are of the same order (10-5 s-1) and suggested that the cis-trans thermal relaxations of azo chromophores in the side chains are not affected by the polymer main chain conformation in the solid films. The birefringence was optically induced in the AZOPLGA polymer films by exposure to a linear polarized Ar+ laser beam. The polarized writing beam induced a preferred orientation of the azobenzene chromophores in the direction perpendicular to the polarization of the writing beam through trans-cis-trans isomerization,11 thus resulting in birefringence in the polymer films. The photoinduced birefringence (∆n) for the AZOPLGA polymers is shown in Table 3. The results indicate that photoinduced birefringence resulting from azobenzene group in the polymer side chain increased with the content of azobenzene chromophores, and it appears that the effect of the conformations of main chains is not significant. The grating recording was performed using linearly polarized beams with polarization (45° with respect to the s-polarization direction. Figure 8 presents the typical AFM three-dimensional images of the SRGs with a grating spacing of 1 µm on the 35% AZOPLGA films cast from polymer pyridine solution (a) and TFA solution (b). The surface modulation depths of the SRGs are shown in Table 3. SRGs on all samples were recorded at an intensity of 200 mW/ cm2, and the writing beams were turned off after the diffraction efficiency saturated. The amplitude of the surface modulation on the 35% AZOPLGA film cast from pyridine is 220 nm, whereas the value for 35% AZOPLGA film cast from TFA is 400 nm. In addition, the gratings in image b have better surface smoothness than those in image a. This result is probably due to the conformation difference of the polymer main chains in the films. For 35% AZOPLGA film cast from TFA solution, the polymer main chains are mainly in random coils, so the main chains can easily undergo the mass transport induced by the isomerization of azobenzene chromophores. However in 35% AZOPLGA film cast from pyridine solution, the main chains of the polymer are predominately in R helix, a rodlike structure. The tightly

Azobenzene-Modified Poly(L-glutamic acid)

Biomacromolecules, Vol. 4, No. 2, 2003 371

and photoinduced birefringence is not significant. The backbones of 35% AZOPLGA are mainly in random coil conformational state in the film cast from TFA solution. However, for the film cast from pyridine solution, the polymer chains of 35% AZOPLGA are in R-helix conformation. The modulation depth of surface relief gratings on the film in which the polymer chains are in random coil conformation is significantly higher than that on the films that have polymer chains predominately in R-helix conformation. Inscribing surface relief grating on the films of 22% AZOPLGA is not efficient because of the lower azobenzene chromophore content, as well as the tightly hydrogen-bonded conformations in R-helix and β-sheet of the polymer main chains. Acknowledgment. Financial support from the NSF-DMR and ONR is gratefully acknowledged. We thank Prof. David Kaplan and the Tufts Biotechnology Center for use of circular dichroism spectrometer. References and Notes

Figure 8. AFM images showing three-dimensional views of the SRGs with a grating spacing of 1 µm on the 35% AZOPLGA films cast from (a) polymer pyridine solution and (b) polymer TFA solution.

coiled polypeptide main chains are less free to move and consequently mass transport is less efficient. In the case of 22% AZOPLGA, the amplitudes of the surface modulation are rather small. For the same reason, the motion of the main chains of 22% AZOPLGA, which are predominantly either in R helix or in β sheet depending on the solvent used for casting the film, is restricted because of strong intra- or intermolecular hydrogen bonding. In addition, the lower content of azobenzene chromophore is also a result in the small modulation of gratings.17 It appears at the functional level of 22% that the polypeptide chains have relatively high levels of intra- and intermolecular hydrogen bonding, while at 35% functionality, the hydrogen bonding is significantly reduced, especially for the random coil conformation. Conclusions Azobenzene-modified poly(L-glutamic acid) polymers were synthesized through linking azobenzene chromophores to the side chain of PLGA. The effect of AZOPLGA polymer main chain conformation on thermal cis-trans isomerization

(1) Kim, D. Y.; Tripathy, S. K.; Li, L.; Kumar J. Appl. Phys. Lett. 1995, 66, 1166. (2) Rochon, P.; Batalla, E.; Natansohn, A. Appl. Phys. Lett. 1995, 66, 136. (3) Ramanujam, P. S.; Holme, N. C. R.; Hvilsted, S. Appl. Phys. Lett. 1996, 68, 1329. (4) Kim, D. Y.; Li, L.; Jiang, X. L.; Shivshankar, V.; Kumar. J.; Tripathy, S. K. Macromolecules 1995, 28, 8835. (5) Barrett, C. J.; Natansohn, A. L.; Rochon, P. L. J. Phys. Chem. 1996, 100, 8836. (6) Holme, N. C. R.; Nikolova, L.; Ramanujam, P. S.; Hvilsted, S. Appl. Phys. Lett. 1997, 70, 1518. (7) Bian, S.; Liu, W.; Williams, J.; Samuelson, L.; Kumar, J.; Tripathy, S. Chem. Mater. 2000, 12, 1585. (8) Yang, S.; Jacob M. M.; Li, L.; Cholli, A. L.; Kumar, J.; Tripathy, S. K. Macromolecules 2001, 34, 9193. (9) Pauling, L.; Corey, R. B.; Branson, H. R. Proc. Natl. Acad. Sci. U.S.A. 1951, 37, 205. (10) Pauling, L.; Corey, R. B. Proc. Natl. Acad. Sci. U.S.A. 1951, 37, 729. (11) Todorov, T.; Nikolova, L.; Tomova, N. Appl. Opt. 1984, 23, 4309. (12) Lader, H. J.; Komoroski, R. A.; Mandelkern, L. Biopolymers 1977, 16, 895. (13) Silverstein, R. M.; Webster, F. X. In Spectrometric Identification of Organic Compounds, 6th ed.; John Wiley & Sons: New York, 1998; p 102. (14) D’Albis, A.; Gratzer, W. B. In Companion to Biochemistry; Bull, A. T., Lagmado, J. R., Thomas, J. O., Tipton, K. F., Eds.; Longman: London, 1970; p 190. (15) Kumar, G. S.; Neckers, D. C. Chem. ReV. 1989, 89, 1915. (16) Barrett, C.; Natansohn, A.; Rochon, P. Chem. Mater. 1995, 7 (5), 899. (17) Fukuda, T.; Matsuda, H.; Shiraga, T.; Kimura, T.; Kato M.; Viswanathan, N. K.; Kumar, J.; Tripathy, S. K. Macromolecules 2000, 33, 4220.

BM025687P