Preparation and Application of a Dual Light ... - ACS Publications

Micellization of Photo-Responsive Block Copolymers. Oliver Grimm , Felix Wendler , Felix Schacher .... Ian W. Wyman , Guojun Liu. Polymer 2013 54, 195...
0 downloads 0 Views 3MB Size
Download PDF
Article pubs.acs.org/Macromolecules

Preparation and Application of a Dual Light-Responsive Triblock Terpolymer Muhammad Rabnawaz and Guojun Liu* Department of Chemistry, Queen’s University, 90 Bader Lane, Kingston, Ontario, Canada K7L 3N6 S Supporting Information *

ABSTRACT: Reported are the preparation and application of a triblock terpolymer poly(ethylene oxide)-ONB-poly[2(perfluorooctyl)ethyl methacrylate]-block-poly(2-cinnamoyloxyethyl methacrylate) (PEO-ONB-PFOEMA-b-PCEMA). Here PEO is water-soluble, PCEMA is photo-cross-linkable, PFOEMA is of low surface tension, and ONB denotes a photocleavable o-nitrobenzyl unit at the junction of the PEO and PFOEMA blocks. To prepare the copolymer, a macroinitiator bearing an ONB unit between PEO and the initiating site was first synthesized. FOEMA and 2-trimethylsilyloxyethyl methacrylate (HEMA-TMS) were then sequentially polymerized by atom transfer radical polymerization (ATRP) to yield the triblock copolymer PEO-ONB-PFOEMA-b-P(HEMATMS). Removal of the trimethylsilyl groups from the P(HEMA-TMS) block and cinnamation of the resultant poly(2hydroxyethyl methacrylate) block with cinnamoyl chloride resulted in the targeted triblock copolymer. PEO-ONB-PFOEMA-bPCEMA formed micelles in tetrahydrofuran/water at the water volume fraction of 80%, in which only the PEO block was soluble. Upon photolysis, the micellar PCEMA cores were cross-linked and the micellar coronal PEO chains were cleft, resulting in the precipitation of cross-linked PFOEMA-b-PCEMA nanoparticles. Films of the nanoparticles were water- and oil-repellent due to the exposure of the initially hidden PFOEMA block. reported by Zhao and co-workers,30 this represents the second report on block copolymers possessing dual light responses. The targeted polymer consisted of poly(ethylene oxide)ONB-poly[2-(perfluorooctyl)ethyl methacrylate)-block-poly(2cinnamoyloxyethyl methacrylate) (PEO-ONB-PFOEMA-bPCEMA or P1). Here PEO is water-soluble, PFOEMA is of low surface tension,31 PCEMA is photo-cross-linkable,2,32 and ONB denotes a photocleavable o-nitrobenzyl unit. Also reported is the preparation of micelles from P1 in THF/ water mixtures at a water volume fraction f H2O of 80%, in which only the PEO block was soluble. Photolysis caused the PCEMA domains to cross-link and the PEO coronas to cleave, resulting in precipitation of cross-linked PFOEMA-b-PCEMA nanoparticles. Coatings made of these nanoparticles were strongly water- and oil-repellent due to the exposure of the originally masked or hidden PFOEMA block. While the copolymer composition disclosed in this paper is new, atom transfer radical polymerization (ATRP),33−35 reversible addition−fragmentation chain transfer polymerization (RAFT),36,37 and anionic polymerization38,39 have been used to prepare copolymers containing fluorinated blocks. Such diblock copolymers have included poly[oligo(ethylene oxide) monomethyl ether acrylate]-block-poly(1H,1H-perfluor-

I. INTRODUCTION Light-responsive block copolymers have various applications.1 For example, the photo-cross-linking of poly(2-cinnamoyloxyethyl methacrylate)- or PCEMA-bearing micelles2−6 has allowed their structural stabilization to yield “permanent” nanostructures including nanofibers,3,4,7 nanotubes,8−10 nanospheres,2 and hollow nanospheres.11−13 The photocleavage of a diblock copolymer at its block junction in a thin film and the subsequent removal of the minority block by solvent extraction yielded membranes with controlled and uniformly sized permeating nanochannels.14−22 Furthermore, photoinduced dissociation of block copolymer micelles has been tested in vitro for triggering reagent release from micelle carriers.23−25 Last but not the least, block copolymers bearing azobenzene units that undergo a photoinduced reversible trans−cis isomerization have been studied extensively for their potential applications in optical information storage and other areas.1 Reported in this paper is the synthesis of a triblock copolymer that bears a photo-cross-linkable block and a photocleavable block junction. Upon photolysis, the triblock copolymer undergoes not only cross-linking but also cleavage. Thus, this is a multiresponsive block copolymer that changes several of its properties or several of its substructure units when subjected to one single stimulus. While there have been many reports on multiply stimulable block copolymers that are responsive to different stimuli,26−29 reports on multiresponsive block copolymers have been rare.30 Aside from the system © 2012 American Chemical Society

Received: March 29, 2012 Revised: June 14, 2012 Published: June 22, 2012 5586

dx.doi.org/10.1021/ma3006476 | Macromolecules 2012, 45, 5586−5595

Macromolecules

Article

temperature, then passed through an alumina column using THF as eluant to remove ligated copper, and subsequently concentrated to 2.0 mL via rotary evaporation. This mixture was then added into 20 mL of diethyl ether to precipitate out the polymer. The precipitate was redissolved in 2.0 mL of THF, and the resultant solution was added into another 20 mL of diethyl ether to precipitate the polymer. This redissolution and precipitation procedure was repeated once more before the polymer was dried under vacuum for 24 h to yield 0.61 g of product in 86% yield. 1H NMR (CDCl3, 500 MHz): δ 4.4 (br, −COOCH2), δ 3.4−3.8 (br, CH2), δ 2.3 (br, CH2), and δ 0.8−1.4 (br, CH3) ppm. FOEMA Polymerization Kinetics. The procedure discussed above for the synthesis of PEO-ONB-PFOEMA was used, except that samples at ∼0.05 mL each were taken from the reaction mixture at 1.0, 2.0, and 3.0 h. The crude mixture was exposed to air in a vial placed in liquid nitrogen to stop the reaction. Subsequently 0.5 mL of CDCl3 was added for 1H NMR analysis. The decrease in the area of the FOEMA peak at 4.4 ppm was compared with that of the PEO main peak at 3.4−3.8 ppm to determine the FOEMA conversion. For SEC analysis, samples were prepared by initially passing the crude mixture over a short pad of alumina to remove residual copper and evaporating the solvent and then dissolving the crude mixture into DMF at 5 mg/mL. PEO-ONB-PFOEMA-b-PHEMA. PEO-ONB-b-PFOEMA was placed into a dialysis tube with a cutoff molecular weight of 12 000 g/mol and dialyzed against distilled THF. The solvent was changed 4 times over 36 h to remove low molecular weight impurities. The dialyzed PEO-ONB-b-PFOEMA sample was dried (0.68 g, 5.9 × 10−2 mmol), transferred into a two-neck flask, and subsequently redissolved in a solvent mixture of anisole/TFT (4.0 mL, v/v = 1/1). After 20 min of stirring at room temperature, HEMA-TMS monomer (0.80 mL, 3.8 × 10−1 mmol), bipyridine ligand (35 mg, 2.2 × 10−1 mmol), and CuBr2 (2.0 mg, 8.1 × 10−3 mmol) were added. CuBr (15.1 mg, 1.05 × 10−1 mmol) was added only after the system was purged with N2. The flask was subjected to four freeze−pump−thaw−N2 refill cycles and was immersed into a preheated oil bath at 65 °C. After 3 h, the reaction flask was quenched in liquid N2 and opened to introduce air. Ligated copper was removed by passing the crude mixture through a short pad of alumina using THF as eluant. At the end of elution, methanol and water were added to reach volume ratios of 3/0.5/0.1 for THF, methanol, and water, respectively. The TMS group was removed after the mixture was stirred overnight at room temperature. The crude polymer was dissolved into THF (4.0 mL) and precipitated from 50 mL of diethyl ether. This was repeated another two times. The product was subsequently centrifuged at 3900 rpm (2600g) for 10 min to yield a compact precipitate, which was dried under vacuum for 16 h, yielding 0.71 g of polymer in 82% yield. 1H NMR (CD3OD, 500 MHz): δ 4.2 (br, −COOCH2), δ 3.8 (br, CH2OH), δ 3.4−3.7 (br, − CH2CH2O), and δ 0.9−1.4 ppm (br, −CH3). PEO-ONB-PFOEMA-b-PCEMA. PEO-ONB-PFOEMA-b-PHEMA (0.20 g, 1.1 × 10−2 mmol containing 2.8 × 10−1 mmol of hydroxyl groups) was dissolved into 4.0 mL of dry pyridine and stirred for 30 min before cinnamoyl chloride (302 mg, 1.8 mmol, 6.5 mol equiv) was added. After stirring the mixture in the dark overnight at room temperature, the reacted mixture was centrifuged at 3900 rpm (2600g) for 10 min to settle the pyridinium salt. The supernatant was concentrated to ∼2.5 mL via rotary evaporation and added into 50 mL of diethyl ether to precipitate the polymer. The precipitate was briefly dried before it was redissolved into 3.0 mL of THF. The resultant solution was added into another 50 mL of diethyl ether to precipitate the polymer. This procedure was repeated once again. The resultant solid was dried at room temperature in a vacuum oven overnight to yield 0.20 g of polymer in 82% yield. 1H NMR (CDCl3, 500 MHz): δ 7.8−7.2 (aromatic protons), δ 4.25 (br, COOCH2CH2CF2), δ 4.2 (br, COOCH2), δ 4.05 (br, −CH2−), δ 3.5−3.6 (br, −CH2CH2O), δ 2.4 (br, CH2CF2), δ 1.8−2.2 (br, CH2), and δ 0.8−1.4 ppm (br, CH3). PEO-Br. The PEO-Br macroinitiator was prepared by reacting poly(ethylene oxide) monomethyl ether with 2-bromoisobutyryl bromide using the protocol used to prepare PEG-ONB-Br. 1H NMR

obutyl acrylate) or PEOA-b-PFBA,40 poly(butyl methacrylate)block-poly(perfluoroalkyl acrylate),41 poly(4-fluorostyrene)block-poly(methyl acrylate) and poly(1H,1H-perfluorooctyl acrylate)-block-poly(methyl methacrylate),42 PEOA-b-PFOEA [poly(perfluorooctylethyl acrylate)],43 and poly(styrene)-blockpoly(2,2,3,3,4,4,4-heptafluorobutyl methacrylate).44 Triblock copolymers reported have included poly(ethylene oxide)block-polystyrene-block-poly(perfluorohexylethyl acrylate)45 and poly[4-methyl-4-(4-vinylbenzyl)morpholin-4-ium chloride]-block-polystyrene-block-poly(pentafluorophenyl-4-vinylbenzyl ether).46 Triblock copolymers have also been prepared through different monomer addition sequences from EOA, benzyl acrylate, and FBA40 as well as from EGA, butyl or 2ethylhexyl acrylate, and FOEA.43

II. EXPERIMENTAL SECTION Materials. 2-Trimethylsiloxyethyl methacrylate (HEMA-TMS) was synthesized following a literature method47 and was distilled over calcium hydride before use. Prior to use, poly(ethylene oxide) monomethyl ether (Aldrich, Mn = 5000 g/mol) was vacuum-dried for 3 days at 55 °C, pyridine (ACS reagent, Fisher Scientific) was refluxed and distilled over CaH2 under nitrogen, and tetrahydrofuran (THF) was distilled over sodium and a small amount of benzophenone. Cinnamoyl chloride (98%, Aldrich), 5-hydroxy-2nitrobenzaldehyde (98%, Aldrich), p-toluenesulphonyl chloride (99.0%, TCI), 2-bromoisobutyryl bromide (98%, Aldrich), triethylamine (>99.5%, Sigma-Aldrich), α,α,α-trifluorotoluene or TFT (99+ %, Acros), anisole (99%, Sigma-Aldrich), CuBr (Aldrich, 99.999%), CuBr2 (Aldrich, 99.999%), bipyridine (Acros, 99+%), and poly(ethylene oxide) monomethyl ether or PEO-550 (Fluka, Mn = 550 g/ mol) were used as received. Characterization. Size exclusion chromatography (SEC) was performed at 70 °C on a Waters 515 system equipped with a Waters 2410 refractive index detector. The three columns were packed by American Polymer Standards Corporation with 5 μm AM 1000, 10 000, and 100 000 Å gels. The system was calibrated using monodisperse polystyrene (PS) standards. The eluent used was dimethylformamide (DMF) containing tetrabutylammonium bromide at 2.5 g/L with a flow rate of 0.9 mL/min. 1H NMR measurements were performed on Bruker Avance-300, Avance-400, or Avance-500 instruments using deuterated pyridine-d5, methanol-d4, or chloroformd3 as solvents and a 3 s relaxation delay. Macroinitiator. The ATRP macroinitiator PEO-ONB-Br bearing an ONB unit between PEO and the initiating site was synthesized in three steps following a literature method.18 The overall yield for the macroinitiator was 40%, and this product was characterized by 1H NMR in CDCl3: δ 8.21 (d, J = 9 Hz, 1H), δ 7.22 (s,1H), δ 6.98 (dd, 1H, J = 9.0 and 2.2 Hz), δ 5.64 (s, 2H), δ 4.24 (t, J = 5 Hz, 2H), δ 3.8−3.4 (br, −OCH2CH2), δ 3.36 (s, 3H), and δ 2.01 (s, 6H) ppm. PEO-ONB-PFOEMA. PEO-ONB-Br (0.30 g, 5.6 × 10−2 mmol) and FOEMA (0.31 mL, 9.6 × 10−1 mmol) were mixed in a two-neck flask. To this mixture were added anisole (0.7 mL), TFT (0.7 mL), bipyridine (17.6 mg, 1.13 × 10−1 mmol), and CuBr2 (1.2 mg, 5.3 × 10−3 mmol). The flask was purged with N2 before CuBr (8.14 mg, 5.6 × 10−2 mmol) was added under N2 flow. The flask was degassed by three freeze−pump−thaw−N2 refill cycles before it was immersed in a preheated oil bath at 85 °C. The polymerization was quenched after 1 h by immersing the reaction flask into a liquid nitrogen bath and the introduction of air. The crude mixture was warmed to room 5587

dx.doi.org/10.1021/ma3006476 | Macromolecules 2012, 45, 5586−5595

Macromolecules

Article

(300 MHz, CDCl3): δ 4.24 (CH2OCO, 2H), δ 3.35−3.80 (br, −CH2CH2−), δ 3.38 (3H, OCH3), δ 2.03 ppm (CH3, 6H). PEO-b-PHEMA. The polymerization procedure was similar to that used to prepare PEO-ONB-PFOEMA-b-PHEMA from PEO-ONBPFOEMA-Br, except the macroinitiator used was PEO-Br. The crude product was filtered through an aluminum pad to remove the catalyst, was dissolved into THF (2.0 mL), and then added into 40 mL of diethyl ether to precipitate the polymer. This procedure was repeated once. The polymer was centrifuged at 3900 rpm (2600g) and dried under vacuum for 24 h, producing 0.31 g of pure sample in 60% yield. 1 H NMR (CD3OD, 500 MHz): δ 4.2 (br, −COOCH2), δ 3.8−3.4 (br, −OCH2CH2), and 1.0−1.4 ppm (br, −CH3). PEO-b-PCEMA. The cinnmation reaction of PEO-b-PHEMA into PEO-b-PCEMA was carried out following the procedure described above for the cinnamation of PEO-ONB-PFOEMA-b-PHEMA. PEOb-PCEMA was obtained in 88% yield. 1H NMR (CDCl3, 500 MHz): δ 7.8−7.2 (aromatic protons), δ 4.25 (br, COOCH2), δ 4.15 (br, −CH2CO−), δ 3.5−3.6 (br, −CH2CH2O), δ 1.8−2.2 (br, CH2), and δ 0.8−1.4 ppm (br, CH3). PEO-ONB-PFOEMA-b-PCEMA Micelles. PEO-ONB-PFOEMA-bPCEMA, or P1 (2 mg), was dissolved into 2.0 mL of THF and stirred for 4 h at room temperature. Water (8.0 mL) was added at 6−7 drops/ min to the solution until f H2O reached 80%. The final solution had a concentration of ∼0.1 mg/mL and was stirred at 500 rpm at room temperature until analysis. TEM Measurements. Transmission electron microscopy (TEM) specimens were prepared by aero-spraying samples via a homemade atomizer onto cellulose-coated copper grids.48 The specimens were further dried under vacuum for 4 h before staining by OsO4 vapor for 1.5 h. The specimens were analyzed using a Hitachi H-7000 instrument operated at 75 kV. AFM Measurements. Specimens were prepared by aero-spraying samples onto freshly cleft mica surfaces. Tapping mode atomic force microscopy (AFM) was performed using a Veeco Multimode microscope equipped with a Nanoscope IIIa controller. The silicon cantilevers used had a force constant and an oscillating frequency of ∼40 N/m and ∼300 kHz, respectively. Micelle Photolysis. A P1 solution (3.0 mL at 0.06 mg/mL in THF/water at f H2O = 80%) was irradiated in a 1.00 cm thick Hellma quartz cell under stirring. The light used was a focused beam that had passed through a 300 nm cutoff filter from a 500 W mercury lamp in an Oriel 6140 lamp housing powered by an Oriel 6128 power supply. To monitor the reaction progress, UV absorption spectra were recorded at different times. For the study of reaction progress by SEC, P1 (16 mg) was dissolved into 0.40 mL of THF before water was added at a rate of 6− 7 drops/min until the total volume reached 2.0 mL. This was followed by the addition of PEO-550 (11.8 mg) as an internal standard. At predesignated times, 0.15 mL of the irradiated sample was collected, dried in a vacuum oven, and then dissolved in 0.02 mL of DMF for SEC analysis. Contact Angle Measurements. The P1 particles that were photolyzed for 3 h should have a PFOEMA corona. The photolyzed particles were settled from the solvation medium by centrifugation for 10 min at 2500 rpm (1250g). The settled particles (∼1.0 mg) were then stirred with 10 mL of methanol for 1 h before the particles were settled by centrifugation for 10 min at 1250g. The methanol rinsing step was repeated thrice, and the final particles (∼0.75 mg) were redispersed in TFT (2.0 mL). Two drops of this dispersion were then dispensed onto a glass plate to cover an area of ∼5−6 mm2. After TFT evaporation, another 2 drops were applied to the same area. This procedure was repeated once more. The particulate film was allowed to dry for 14 h before contact angle measurements using 5 μL of testing liquids.

initiating site was first synthesized.18,19 This was followed by the sequential ATRP of FOEMA and 2-trimethylsilyloxyethyl methacrylate (HEMA-TMS) to yield the triblock copolymer PEO-ONB-PFOEMA-b-P(HEMA-TMS). The targeted block copolymer was obtained after the cleavage of the trimethylsilyl group and the cinnamation of the resultant PHEMA block with cinnamoyl chloride.2 PEO-ONB-Br. PEO-ONB-Br was synthesized following a literature method using reactions depicted in Scheme 1.18 The Scheme 1. Reactions Used To Synthesize PEO-ONB-Br

commercially available 5-hydroxy-2-nitrobenzaldehyde was first reduced to yield 3-hydroxymethyl-4-nitrophenol. The latter was then reacted with α-methoxy-ω-toluenesulfonyl-PEO to yield PEO bearing a terminal hydroxyl group or PEO-ONB-OH. This hydroxyl group was further reacted with 2-bromoisobutyryl bromide to yield the targeted macroinitiator. Figure 1 shows a 1H NMR spectrum of PEO-ONB-Br in CDCl3 and its peak assignments. A 1H NMR spectrum for PEO-ONB-OH and the integrated intensities for the peaks are given in Figure S1 of the Supporting Information. The intensity ratio between the peaks at 3.5−3.8 ppm for the main chain methylene protons and the benzyl methyene was determined to be 228. Based on the molar mass of 5.0 × 103 g/mol for PEO, this ratio should be 226. The agreement between these two numbers suggested that PEO was quantitatively labeled by 3hydroxymethyl-4-nitrophenol. This was important because the incomplete end-capping of PEO by 3-hydroxymethyl-4-nitrophenol and the subsequent capping of PEO-OH by 2bromoisobutyryl bromide would yield eventually PEO-bPFOEMA-b-PCEMA. Unlike PEO-ONB-PFOEMA-b-PCEMA, PEO-b-PFOEMA-b-PCEMA would not photocleave. SEC analysis of PEO-ONB-Br was performed using DMF containing tetrabutylammonium bromide as the mobile phase, an eluant that we used routinely. Figure 2 shows the SEC trace for PEO-ONB-Br. It consisted of a main peak and a small impurity peak at 26 min. The polydispersity of the main peak was 1.04 in terms of polystyrene standards. Both of these peaks were essentially identical to those observed for the precursory PEO sample. Thus, the unknown impurity existed in the original PEO sample and was not introduced by us. PEO-ONB-PFOEMA-Br. PEO-ONB-PFOEMA-Br was synthesized by ATRP of FOEMA using PEO-ONB-Br as the macroinitiator. FOEMA was polymerized at different temperatures and using different solvents such as toluene, hexafluorobenzene, TFT, and mixtures of TFT and toluene as well as different ligands such as bipyridine and N,N,N′,N″,N″-

III. RESULTS AND DISCUSSION To prepare PEO-ONB-PFOEMA-b-PCEMA, a PEO-ONB-Br macroinitiator bearing an ONB unit between PEO and an 5588

dx.doi.org/10.1021/ma3006476 | Macromolecules 2012, 45, 5586−5595

Macromolecules

Article

Figure 1. 1H NMR spectra of PEO-ONB-Br (top) and PEO-ONB-PFOEMA (bottom) determined in CDCl3.

The prepared PEO-ONB-PFOEMA-Br was filtered through an alumina column to remove ligated copper and was purified by repeated precipitation from diethyl ether. Figure 1 also shows a 1H NMR spectrum for PEO-ONB-PFOEMA. From the comparison between the intensities of the PEO and PFOEMA peaks and based on the repeat unit number of 113 for PEO, the PFOEMA repeat unit number was calculated to be 12. Figure 2 also shows an SEC trace for PEO-ONB-PFOEMABr or PEO-ONB-PFOEMA. The polydispersity index Mw/Mn in terms of PS standards was low at 1.06. While the apparent Mn for PEO-ONB-Br in terms of PS standards was 13 300 g/ mol, the value for PEO-ONB-PFOEMA was ∼15 000 g/mol. This small increase was most likely due to the poor solubility and the compact conformation of the PFOEMA block in DMF. PEO-ONB-PFOEMA-b-PHEMA. HEMA was not directly polymerized using PEO-ONB-PFOEMA-Br because the PFOEMA block was insoluble in polar solvents, which solubilized PHEMA. Thus, a detour was taken by first polymerizing HEMA-TMS using PEO-ONB-PFOEMA-Br in TFT/anisole at v/v = 1/1, in which both the short PFOEMA and P(HEMA-TMS) blocks were soluble. PHEMA was obtained via the hydrolysis of P(HEMA-TMS) under mild conditions in the presence of water and methanol in THF. At the beginning of the HEMA-TMS polymerization, the reaction mixture was foaming, probably due to bubble stabilization by PEO-ONB-PFOEMA-Br. Foaming gradually disappeared with time. The reaction progress was again monitored by 1H NMR and SEC analysis of samples taken at different times. The optimized polymerization time was 3 h at 65 °C when bipyridine was used as the ligand. PEO-ONB-PFOEMA-b-PHEMA was freed of the ligated copper again by filtration through an alumina column and was

Figure 2. Comparison of SEC traces of PEO-ONB-Br, PEO-ONBPFOEMA, and PEO-ONB-PFOEMA-b-PCEMA.

pentamethyldiethylenetriamine. These experiments eventually established that the use of TFT/anisole at v/v = 1/1 as the solvent and bipyridine as the ligand yielded samples with the lowest polydispersity. FOEMA polymerization in TFT/anisole at v/v = 1/1 was followed by analyzing samples taken at different times by 1H NMR and SEC. Our NMR analyses indicated that the monomer conversions at the polymerization times of 1, 2, and 3 h were 67%, 90%, and 100%, respectively. Figure S2 in the Supporting Information compares the SEC traces of samples taken at these times. At 1 h, the diblock copolymer exhibited a symmetric peak and the small impurity peak at 26 min originally seen in the PEO sample. At 2 h, the peak at 26 min increased in intensity. At 3 h, an extra peak appeared at 25.2 min. The high molecular weight peaks might be due to the coupling between different chains at the latter stages of polymerization. Thus, PEO-ONB-PFOEMA was prepared using a FOEMA polymerization time of 1 h. 5589

dx.doi.org/10.1021/ma3006476 | Macromolecules 2012, 45, 5586−5595

Macromolecules

Article

Figure 3. 1H NMR spectra of PEO-ONB-PFOEMA-b-PHEMA measured in CD3OD (top) and PEO-ONB-PFOEMA-b-PCEMA measured in CDCl3 (bottom).

Table 1. Characteristics of P1 at Different Stages of Preparation

a

sample

SEC Mw (g/mol)

SEC Mw/Mn

PEO-ONB-Br PEO-OBN-PFOEMA-Br P1

14 000 15 000 38 000

1.04 1.06 1.10

NMR l/m/n

NMR Mn

l

m

n

113/12 113/12/25

5 000a 11 500 18 200

113 113 113

12 12

25

Calculated based on the supplier’s nominal molecular weight of 5000 g/mol for PEO.

purified by repeated precipitation. A 1H NMR spectrum was obtained in CD3OD, which dissolved only the PEO and PHEMA blocks and not the PFOEMA block. Figure 3 shows the 1H NMR spectrum together with peak assignments. Peaks at 4.2 and 3.8 ppm corresponded to ethylene protons of the hydroxyethyl group, confirming the production of the PHEMA block. Our quantitative analysis indicated that the EO/HEMA molar ratio was 113/25 or the repeat unit number was 25 for the PHEMA block. PEO-ONB-PFOEMA-b-PCEMA. Reacting the hydroxyl groups of PEO-ONB-PFOEMA-b-PHEMA with an excess of cinnamoyl chloride in pyridine at room temperature yielded PEO-ONB-PFOEMA-b-PCEMA.2,49,50 The resultant polymer was easily purified by repeated precipitation into diethyl ether and was analyzed by 1H NMR in CDCl3. Figure 3 also shows an NMR spectrum for PEO-ONB-PFOEMA-b-PCEMA together with peak assignments. All of the anticipated peaks for PEO, PFOEMA, and PCEMA were observed. A quantitative peak integral analysis confirmed l/m/n = 113/12/25.

The SEC trace of PEO-ONB-PFOEMA-b-PCEMA or P1 is included Figure 2. Despite its apparent width, the polydispersity of the peak including the high-molecular-weight shoulder was only 1.10. Having difficulty believing in this number, we reanalyzed this sample on another SEC system using chloroform as the eluant, and the SEC trace is shown in Figure S3. The peak was narrow, and the polydispersity based on PS standards was again low at 1.10. Thus, the apparent width of this peak might be due to the good resolution of the 1000 and 10 000 Å columns used in the molecular weight range of P1. P1 Micelles. P1 micelles were prepared in several steps. First, P1 was dissolved into THF. While THF is a good solvent for PCEMA, it does not dissolve high-molecular-weight PFOEMA or PEO. Despite this, it dissolved the PFOEMA and PEO blocks in P1 due to their low molecular weights. Second, water, a selective solvent for PEO, was added to a volume fraction f H2O of 80% to yield micelles with PEG as the corona and PFOEMA and PCEMA as the core. 5590

dx.doi.org/10.1021/ma3006476 | Macromolecules 2012, 45, 5586−5595

Macromolecules

Article

Figure 4. AFM topography images of P1 micelles (a), photolyzed P1 micelles (c), and TEM image of the P1 micelles (b). The micelles were aerosprayed from THF/water at f H2O = 80%.

Absorption Characteristics of ONB and PCEMA. To facilitate the choice of photolysis conditions that would crosslink PCEMA and cleave PEO simultaneously, the absorption properties of PEO-ONB-Br, P1, and PEG-b-PCEMA were compared. Here PEO-b-PCEMA was prepared using PEG-Br as the macroinitiator for HEMA-TMS polymerization. The TMS groups were then removed, and the resultant PHEMA block was cinnamated. The number of repeat units for the PCEMA block of PEO-b-PCEMA was also 25. Compared in Figure 5 are the UV absorption spectra of PEO-ONB-Br, PEO-b-PCEMA, and P1 in distilled THF. Since

The aqueous micellar solution was atomized or aero-sprayed using a home-built device48 onto freshly cleft mica for tappingmode AFM analysis and on a cellulose-covered grid for TEM analysis. Before TEM analysis, the PCEMA block of the sample was stained by OsO4. Aero-spraying was used because it sped up solvent evaporation from the atomized liquid droplets. Under these conditions, THF should have evaporated as it traveled from the spraying nozzle to the silicon wafer and water should have evaporated within ∼3 s after the landing of the atomized aqueous droplets. This fast solvent evaporation was to minimize the chances for a morphological transition of the micelles during specimen preparation. Figure 4a,b shows AFM height and TEM images of the aerosprayed P1 micelles. The particles seemed to have a bimodal distribution. The smaller particles had average AFM and TEM diameters of 31 ± 5 and 18 ± 4 nm, respectively, while the larger particles had AFM and TEM diameters of 47 ± 7 and 33 ± 6 nm, respectively. Our suspicion is that the smaller particles were core−shell− corona spherical micelles, where PCEMA, PFOEMA, and PEG formed the core, shell, and corona, respectively. The AFM diameter was larger than the TEM diameter because AFM probed the whole particles including PEO and PFOEMA layers, and TEM only probed the OsO4-stained PCEMA core. Also, the AFM diameter should have some contribution from the finite size of the tip used. We initially suspected that the larger particles were vesicles because the PCEMA core chains of the simple core−shell− corona micelles at an average repeat unit number of 25 could not be stretched to 33 ± 6 nm. This was, however, not supported by the TEM or the AFM results. If they were vesicles, the larger particles would have appeared in the TEM image as a circle with a dark rim and a gray center. Instead, the larger circles in Figure 4b appeared increasingly dark toward the center, suggesting that they were solid particles. Also, some vesicles would normally collapse after solvent evaporation leaving behind a Kippah-like structure.51 This Kippah-like structure was not seen by AFM in Figure 4a. While the nature of this study does not demand a detailed clarification of chain packing in the larger particles, we suspect that the larger particles were still core−shell−corona particles formed from polymer chains with a PCEMA block that was substantially longer than 25 units. Despite its apparent low polydispersity index, P1 probably possessed substantial composition heterogeneity or a fairly large distribution in FOEMA to CEMA repeat unit ratio m/n.

Figure 5. UV absorption spectra of PEO-ONB-Br (a), PEG-b-PCEMA (b), and P1 (c).

PEO should absorb negligibly at wavelengths >260 nm, the maxima at 306 and 274 nm should result from ONB and CEMA absorption, respectively. A quantitative analysis indicated that the ONB group in PEO-ONB-Br had a molar extinction coefficient ε of 8.2 × 103 M−1 cm−1 at 306 nm. The molar extinction coefficient of CEMA units at their absorption maximum has been reported to be 2.8 × 104 M−1 cm−1 by Guo et al.2 and 2.1 × 104 M−1 cm−1 by Marusich et al.52 Using 2.5 × 104 M−1 cm−1, the average for the two numbers, and the absorbance ratio of 0.237 determined from Figure 5 for PEO-bPCEMA at 306 and 274 nm, we calculated a ε value of 5.9 × 103 M−1 cm−1 for CEMA absoprtion at 306 nm. Since there were 25 CEMA units and only 1 ONB unit per P1 chain, the ONB group should contribute insignificantly to P1 light absorption at wavelengths shorter than 306 nm. A comparison of curves b and c of Figure 5 also concluded ONB would absorb light dominantly at wavelengths longer than 320 nm. Micelle Photolysis. Based on the absorption characteristics of the PCEMA and ONB, P1 micelles in THF/water at f H2O = 80% were photolyzed using light generated by a high-pressure 5591

dx.doi.org/10.1021/ma3006476 | Macromolecules 2012, 45, 5586−5595

Macromolecules

Article

Scheme 2. The ONB Rearrangement and PEO Cleavage Reaction

Figure 6. Comparison of UV absorption spectra of P1 at different photolysis times (a). Also shown are the variations in the relative absorbances at 274 and 306 nm as a function of irradiation time (b).

Figure 7. Comparison of SEC traces of P1 containing PEO-550 at photolysis times of 0, 10, 20, and 180 min (a). Also shown is the increase in the PEO-5k peak relative to that of the PEO-500 peak as a function of photolysis time (b).

mercury lamp and filtered by a 300 nm cutoff filter. Most of the light below 300 nm was removed to hopefully match the rates of ONB cleavage and PCEMA cross-linking. The micellar solution was clear before irradiation. Turbidity developed within 5 min and intensified with further irradiation. After 3 h, the particles settled from the solvation medium after stirring was stopped, suggesting the cleavage of the PEO block and photoinduced particle precipitation. Also, the settled particles were redispersed into CDCl3 to yield a cloudy solution. Since PFOEMA-b-PCEMA bearing un-cross-linked PCEMA would have dissolved in this solvent to yield a clear solution, the cloudiness suggested the retention of the micellar structure in CDCl3 and the cross-linking of the PCEMA core. The PEO chains were cleaved from the PFOEMA block because of a Norrish II rearrangement (Scheme 2) of ONB.53 PCEMA was cross-linked due to the dimerization of CEMA units of different P1 chains.2 The photolysis processes were monitored by analyzing samples collected at different irradiation times using UV and SEC. Figure 6a compares the UV absorption spectra of a P1 solution at 0.06 mg/mL in THF/water at f H2O = 80% after it was irradiated for different time periods. Plotted in Figure 6b are the changes in the relative absorbances at 274 and 306 nm

as a function of the photolysis time. The two curves almost coincided, with both curves showing rapid initial absorbance decrease that was followed by little change between 2 and 3 h. The observed absorbance decrease pattern was reasonable as the rate of a reaction depended on the amount of light absorbed by the reacting species. As the conversion increased and less reactant remained, the amount of light absorbed by the reactant per unit time decreased, and the reaction rate decreased. The almost identical rate of absorbance decrease at 274 and 306 nm should not be surprising because the PCEMA absorption dominated over ONB absorption at 306 nm as well. Thus, UV absorption analysis at these two wavelengths only allowed the monitoring of the rate of CEMA disappearance with photolysis. Meanwhile, SEC was better suited for monitoring PEG cleavage. As more sample was required for SEC analysis, the photolyzed polymer solution in this case had a higher initial concentration of 8 mg/mL. To quantify the amount of P1 lost or PEG formed, a PEO sample with Mn = 550 g/mol or PEO550 was added into the photolysis mixture as an internal standard. This oligomer was used as the internal standard because it would remain inert during photolysis, and its peak 5592

dx.doi.org/10.1021/ma3006476 | Macromolecules 2012, 45, 5586−5595

Macromolecules

Article

Figure 8. Photographs of H2O droplets impregnated with rhodamine B (a, b) and CH2I2 droplets (c, d) on films of P1 micelles (a, c) and photolyzed P1 micelles (b, d). Also shown are the variations in the contact angles of H2O and CH2I2 droplets as functions of PEG cleavage (e).

however, have been increased by using a filter with a shorter cutoff wavelength such as at 270 nm. PEO-Cleft Particles. After a P1 solution at 8 mg/mL in THF/water at f H2O = 80% was irradiated for 3 h, the solution became turbid. This solution was diluted to ∼1 mg/mL with THF/water at f H2O = 80% and then pushed through a 3.1 μm filter to remove large aggregates. The smaller particles were then aero-sprayed for AFM analysis. Such an image is shown in Figure 4c. The bimodal distribution of the particles was retained and is evident in Figure 4c. The major difference between the particles in Figures 4a and 4c was that the particles in Figure 4c had formed aggregates and those in Figure 4a were mostly individual. The particles had aggregated because they were free of the PEO corona and were not readily dispersible in THF/water. The particles after photolysis were centrifuged, separated from the supernatant, redispersed into methanol under vigorous stirring, and settled via centrifugation. This methanol rinsing step was repeated several times to remove the photocleft PEO chains. The particles were dried and then dispersed into CDCl3 to yield a turbid solution. 1H NMR analysis indicated the absence of any PEO peaks despite the solubility of PEO in CDCl3 (Figure S4). This thus suggested the complete removal of the PEO coronal chains during photolysis. No PCEMA signals were observed either because the PCEMA chains were cross-linked and not mobile. The presence of the PFOEMA block in the photolyzed sample was, however, confirmed by 19F NMR analysis (Figure S5). Visual evidence for PEO cleavage was that the H2O and CH2I2 contact angle change on films made of P1 micelles before and after their photolysis. Micellar P1 films were prepared by casting onto glass plates P1 micellar solutions at f H2O = 80% and photolyzed P1 micellar solutions dispersed in TFT, respectively. Figure 8a−d compares the shapes of H2O and CH2I2 droplets on different films. While H2O and CH2I2 contact angles on the micellar films were 52° and 32°, respectively, the values increased to 154° and 136° on films of the photolyzed samples. These large contact angles were possible only if the particle surfaces were enriched by PFOEMA and the particulate films were rough.39,54,55 Two AFM images of films cast from a TFT dispsersion of a photolyzed nanoparticle sample are shown in Figure S6. The root-mean-square roughness of the films at the probed scale of 3.1 and 10 000 μm2 from the two images were 8.5 and 154 nm, respectively. Thus, the films were indeed rough.

appeared at 32 min, a position well-resolved from that of P1 and the cleft PEO block. Solvent was removed from the photolyzed samples via rotary evaporation. They were then immediately redispersed into DMF, filtered, and analyzed by SEC. Figure 7 compares SEC traces of P1 irradiated for 0, 10, 20, and 180 min, respectively. Before irradiation, the peaks eluting at 25.5 and 32 min corresponded to P1 and PEO-550, respectively. After 10 min of irradiation, a shoulder emerged at ∼24 min on the higher molecular weight side and the intensity of the P1 peak decreased. This suggested the production of higher molecular weight species due to the photo-cross-linking of PCEMA. Although less noticeable, the shoulder on the lower-molecularweight side at 27.8 min had also grown, suggesting PEO-5k formation. At 20 min, the P1 peak had fully disappeared, leaving behind a very small peak at ∼26 min, while the PEO-5k peak eluting at 27.8 nm became distinct. Further photolysis up to 180 min did not eliminate the small peak at 26 min but increased the intensity of the peak eluting at 27.8 min. No peak was seen for the PFOEMA-b-PCEMA fragments after 20 min photolysis because this part became locked into a cross-linked micellar structure. Cross-linked particles with PFOEMA corona were not dispersed into DMF, the eluant used for SEC analysis, and were mostly removed during sample filtration. As mentioned before, the small peak at ∼26 min must have been due to an impurity present in the original PEO precursor. The Peakfit program was used to resolve the peak of the P1 sample that was irradiated for 10 min into three peaks consisting of cross-linked P1, P1, and photocleaved PEO-5k. The areas of the peak for PEO-5k and PEO-550 were calculated using the Peakfit program to yield the area ratio A5k/A550, where A5k and A550 represent the areas corresponding to PEO5k and PEO-550, respectively. Since the PEO-5k peak was free of interference from P1 at other irradiation times, the determination of A5k/A550 was straightforward. Figure 7b shows how A5k/A550 varied with photolysis time. Figure 7b clearly shows much less PEO was cleft in the first 10 min than in the next 10 min. This was probably due to the fact that PCEMA cross-linking dominated initially. It was only after the CEMA concentration and its absorption had decreased that the ONB rearrangement took off. The A5k/A550 value leveled off after 120 min, suggesting the completion of the ONB rearrangement reaction at that point. A comparison between Figures 6b and 7b further suggested that the cleavage reaction seemed to go faster than the crosslinking reaction. The PCEMA cross-linking rate could, 5593

dx.doi.org/10.1021/ma3006476 | Macromolecules 2012, 45, 5586−5595

Macromolecules

Article

While films of P1 micelles have been called micellar films, the micellar structure before photolysis was not locked and a structural rearrangement of the micelles might be possible during film formation. No other methods were attempted for preparing better micellar films because this structural rearrangement should not have changed the observed H2O and CH2I2 contact angle variation trends for films made of micelles and photolyzed micelles. Plotted in Figure 8e are the variations in the H2O and CH2I2 contact angles as functions of PEO cleavage, which was obtained from the ratio between A5k/A550 at other times to that at 120 min. Significant contact angle increases occurred only above the degree of PEG cleavage of 80%. This was reasonable because the residual PEO chains might lie flat on the PFOEMA surface and help reduce the liquid contact angles. The PFOEMA chains were exposed only after the PEO chains were almost fully removed.

(6) Wyman, I.; Njikang, G.; Liu, G. J. Prog. Polym. Sci. 2011, 36, 1152−1183. (7) Tao, J.; Stewart, S.; Liu, G. J.; Yang, M. L. Macromolecules 1997, 30, 2738−2745. (8) Stewart, S.; Liu, G. Angew. Chem., Int. Ed. 2000, 39, 340−344. (9) Yan, X. H.; Liu, G. J.; Liu, F. T.; Tang, B. Z.; Peng, H.; Pakhomov, A. B.; Wong, C. Y. Angew. Chem., Int. Ed. 2001, 40, 3593− 97. (10) Yan, X. H.; Liu, G. J.; Haeussler, M.; Tang, B. Z. Chem. Mater. 2005, 17, 6053−6059. (11) Ding, J. F.; Liu, G. J. Chem. Mater. 1998, 10, 537−542. (12) Stewart, S.; Liu, G. J. Chem. Mater. 1999, 11, 1048−1054. (13) Ding, J. F.; Liu, G. J. J. Phys. Chem. B 1998, 102, 6107−6113. (14) Goldbach, J. T.; Lavery, K. A.; Penelle, J.; Russell, T. P. Macromolecules 2004, 37, 9639−9645. (15) Hillmyer, M. A. Adv. Polym. Sci. 2005, 190, 137−181. (16) Liu, G. J.; Ding, J. F.; Guo, A.; Herfort, M.; Bazett-Jones, D. Macromolecules 1997, 30, 1851−1853. (17) Liu, G. J.; Ding, J. F.; Hashimoto, T.; Kimishima, K.; Winnik, F. M.; Nigam, S. Chem. Mater. 1999, 11, 2233−2240. (18) Kang, M.; Moon, B. Macromolecules 2009, 42, 455−458. (19) Schumers, J. M.; Gohy, J. F.; Fustin, C. A. Polym. Chem. 2010, 1, 161−163. (20) Schumers, J. M.; Vlad, A.; Huynen, I.; Gohy, J. F.; Fustin, C. A. Macromol. Rapid Commun. 2012, 33, 199−205. (21) Gamys, C. G.; Schumers, J. M.; Vlad, A.; Fustin, C. A.; Gohy, J. F. Soft Matter 2012, 8, 4486−4493. (22) Zhao, H.; Gu, W. Y.; Sterner, E.; Russell, T. P.; Coughlin, E. B.; Theato, P. Macromolecules 2011, 44, 6433−6440. (23) Zhao, Y. J. Mater. Chem. 2009, 19, 4887−4895. (24) Wang, G.; Tong, X.; Zhao, Y. Macromolecules 2004, 37, 8911− 8917. (25) Jiang, J. Q.; Tong, X.; Zhao, Y. J. Am. Chem. Soc. 2005, 127, 8290−8291. (26) Pasparakis, G.; Vamvakaki, M. Polym. Chem. 2011, 2, 1234− 1248. (27) Haag, S. F.; Fleige, E.; Chen, M.; Fahr, A.; Teutloff, C.; Bittl, R.; Lademann, J.; Schafer-Korting, M.; Haag, R.; Meinke, M. C. Int. J. Pharm. 2011, 416, 223−228. (28) Stuart, M. A. C.; Huck, W. T. S.; Genzer, J.; Muller, M.; Ober, C.; Stamm, M.; Sukhorukov, G. B.; Szleifer, I.; Tsukruk, V. V.; Urban, M.; Winnik, F.; Zauscher, S.; Luzinov, I.; Minko, S. Nat. Mater. 2010, 9, 101−113. (29) Hu, J. M.; Liu, S. Y. Macromolecules 2010, 43, 8315−8330. (30) He, J.; Tong, X.; Tremblay, L.; Zhao, Y. Macromolecules 2009, 42, 7267−7270. (31) Hirao, A.; Sugiyama, K.; Yokoyama, H. Prog. Polym. Sci. 2007, 32, 1393−1438. (32) Zhou, Z. H.; Liu, G. J.; Hong, L. Z. Biomacromolecules 2011, 12, 813−823. (33) Wang, J. S.; Matyjaszewski, K. J. Am. Chem. Soc. 1995, 117, 5614−5615. (34) Matyjaszewski, K.; Xia, J. H. Chem. Rev. 2001, 101, 2921−2990. (35) Kato, M.; Kamigaito, M.; Sawamoto, M.; Higashimura, T. Macromolecules 1995, 28, 1721−1723. (36) Chiefari, J.; Chong, Y. K.; Ercole, F.; Krstina, J.; Jeffery, J.; Le, T. P. T.; Mayadunne, R. T. A.; Meijs, G. F.; Moad, C. L.; Moad, G.; Rizzardo, E.; Thang, S. H. Macromolecules 1998, 31, 5559−5562. (37) Perrier, S.; Takolpuckdee, P. J. Polym. Sci., Part A: Polym. Chem. 2005, 43, 5347−5393. (38) Ishizone, T.; Sugiyama, K.; Sakano, Y.; Mori, H.; Hirao, A.; Nakahama, S. Polym. J. 1999, 31, 983−988. (39) Xiong, D.; Liu, G. J.; Hong, L. Z.; Duncan, E. J. S. Chem. Mater. 2011, 23, 4357−4366. (40) Marsat, J. N.; Heydenreich, M.; Kleinpeter, E.; Berlepsch, H. V.; Bottcher, C.; Laschewsky, A. Macromolecules 2011, 44, 2092−2105. (41) Li, K.; Wu, P. P.; Han, Z. W. Polymer 2002, 43, 4079−4086. (42) Xia, J. H.; Johnson, T.; Gaynor, S. G.; Matyjaszewski, K.; DeSimone, J. Macromolecules 1999, 32, 4802−4805.

IV. CONCLUSIONS ATRP has been used to produce a triblock copolymer PEOONB-PFOEMA-b-PCEMA. The polymer has been carefully characterized by 1H NMR and SEC. The number of repeat units for PEO, PFOEMA, and PCEMA were 113, 12, and 25, respectively. In addition, the sample polydispersity was low at 1.10 with respect to PS standards. The light absorption characteristics of the CEMA and ONB units were established by comparing the UV absorption spectra of PEG-b-PCEMA, PEO-ONB, and PEO-ONB-PFOEMA-b-PCEMA. In THF/ water at f H2O = 80%, P1 formed micelles. Irradiation using light with wavelengths >300 nm cross-linked the PCEMA core and cleft the PEG corona, yielding particles bearing expoesd PFOEMA chains. Casting dispersions of these particles into TFT yielded films that were both superhydrophobic and oleophobic. The water and oil repellence of films of the crosslinked micelles improved as the degree of PEO cleavage increased.

■

ASSOCIATED CONTENT

S Supporting Information *

SEC traces, AFM images, and NMR spectra. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS NSERC of Canada is thanked for financially sponsoring this research and Ian Wyman is thanked for proofreading the manuscript. G.L. acknowledges the Canada Research Chairs program for a Tier I Canada Research Chair Position in Materials Science.

■

REFERENCES

(1) Schumers, J. M.; Fustin, C. A.; Gohy, J. F. Macromol. Rapid Commun. 2010, 31, 1588−1607. (2) Guo, A.; Liu, G. J.; Tao, J. Macromolecules 1996, 29, 2487−2493. (3) Liu, G. J.; Qiao, L. J.; Guo, A. Macromolecules 1996, 29, 5508− 5510. (4) Liu, G. J. Adv. Polym. Sci. 2008, 220, 29−64. (5) Njikang, G.; Liu, G. J.; Hong, L. Z. Langmuir 2011, 27, 7176− 7184. 5594

dx.doi.org/10.1021/ma3006476 | Macromolecules 2012, 45, 5586−5595

Macromolecules

Article

(43) Skrabania, K.; von Berlepsch, H.; Bottcher, C.; Laschewsky, A. Macromolecules 2010, 43, 271−281. (44) Zhou, Y. N.; Cheng, H.; Luo, Z. H. J. Polym. Sci., Part A: Polym. Chem. 2011, 49, 3647−3657. (45) Guo, W. J.; Tang, X. D.; Xu, J.; Wang, X.; Chen, Y.; Yu, F. Q.; Pei, M. S. J. Polym. Sci., Part A: Polym. Chem. 2011, 49, 1528−1534. (46) Kubowicz, S.; Baussard, J. F.; Lutz, J. F.; Thunemann, A. F.; von Berlepsch, H.; Laschewsky, A. Angew. Chem., Int. Ed. 2005, 44, 5262− 5265. (47) Hirao, A.; Kato, H.; Yamaguchi, K.; Nakahama, S. Macromolecules 1986, 19, 1294−1299. (48) Ding, J. F.; Liu, G. J. Macromolecules 1999, 32, 8413−8420. (49) Henselwood, F.; Liu, G. J. Macromolecules 1997, 30, 488−493. (50) Wang, G. C.; Henselwood, F.; Liu, G. J. Langmuir 1998, 14, 1554−1559. (51) Azzam, T.; Eisenberg, A. Langmuir 2010, 26, 10513−10523. (52) Marusich, W. C.; Jensen, R. A.; Zamir, L. O. J. Bateriol. 1981, 146, 1013−1019. (53) Holmes, C. P. J. Org. Chem. 1997, 62, 2370−2380. (54) Xia, F.; Jiang, L. Adv. Mater. 2008, 20, 2842−2858. (55) Xiong, D. A.; Liu, G. J.; Zhang, J. G.; Duncan, S. Chem. Mater. 2011, 23, 2810−2820.

5595

dx.doi.org/10.1021/ma3006476 | Macromolecules 2012, 45, 5586−5595