Azobenzene-containing linear–dendritic block copolymers prepared by sequential ATRP and click chemistry

Polymer 53 (2012) 4604e4613

Contents lists available at SciVerse ScienceDirect

Polymer journal homepage: www.elsevier.com/locate/polymer

Azobenzene-containing linearedendritic block copolymers prepared by sequential ATRP and click chemistry Eva Blasco a, Jesús del Barrio a, Milagros Piñol a, *, Luis Oriol a, *, Cristina Berges b, Carlos Sánchez b, Rafael Alcalá b a b

Instituto de Ciencia de Materiales de Aragón (ICMA), Universidad de Zaragoza-CSIC, Departamento de Química Orgánica, Facultad de Ciencias, 50009 Zaragoza, Spain Instituto de Ciencia de Materiales de Aragón (ICMA), Universidad de Zaragoza-CSIC, Departamento de Física de la Materia Condensada, Facultad de Ciencias, 50009 Zaragoza, Spain

a r t i c l e i n f o

a b s t r a c t

Article history: Received 25 June 2012 Received in revised form 24 July 2012 Accepted 9 August 2012 Available online 16 August 2012

A series of light responsive linearedendritic BCs consisting of a liquid crystalline aliphatic polyester dendron functionalized with sixteen 4-cyanoazobenzene moieties linked to a linear block of poly(methyl methacrylate), poly(ethyl methacrylate) or poly(styrene) have been investigated. The linear block was synthesis by atom transfer radical polymerization using an alkyne terminated initiator and was coupled to dendron with an azido group at the focal point by a Huisgen’s 1,3-dipolar cycloaddition. The effectiveness of the coupling reaction and purity of the block copolymers was asserted by chromatographic and spectroscopic techniques as well as their thermal behaviour was studied by thermogravimetric analysis, differential scanning calorimetry and optical microscopy. Morphology, optical properties and photoinduced response of these block copolymers have also been evaluated. Ó 2012 Elsevier Ltd. All rights reserved.

Keywords: Block copolymers Azobenzene ATRP

1. Introduccion Photoresponsive polymers having azobenzene moieties in the side chain have been widely studied in the field of optical applications mainly because linearly polarized light (LPL) provokes reorientation of the azobenzene molecules giving as a result photoinduced anisotropy [1e7]. This behaviour is related to the trans-cis isomerization of the azobenzene moiety which is a clean and reversible light induced process activated upon irradiation with particular wavelengths of the ultravioletevisible region [8]. In this way, a large and stable optical anisotropy can be induced in azopolymer films with the long axis of trans-azobenzene units preferentially oriented perpendicular to the light polarization direction. To fully exploit the advantages of some applications such as volume holographic optical storage [9e11], the azobenzene content has to be diluted with a non-absorbing material to reduce the otherwise large optical absorption of azopolymers at the recording wavelength (usually 488 nm). For this reason, block copolymers (BCs) containing side chain liquid crystalline (LC) azobenzene blocks have received an increased attention over the last years [12e16]. These BCs can exhibit a photo-responsive behaviour similar to that

* Corresponding authors. Tel.: þ34 976 76 22 76. E-mail addresses: [email protected] (M. Piñol), [email protected] (L. Oriol). 0032-3861/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.polymer.2012.08.022

of azo homopolymers but at lower contents of azobenzene and better than random copolymers of similar composition. The reason lies on the BCs segregation ability at the nanoscale level giving place to well defined structures where the confinement of the azobenzene moieties into nanodomains might preserve the azobenzenee azobenzene interactions [17e19]. For instance, it has been shown that some methacrylic diblock copolymers with azo units in the side chain can provide a dilution of the azo content up to 5 times (20 wt-% azo content) while keeping a photoinduced anisotropy similar to that of the corresponding azo homopolymer [18]. These copolymers have been used to store holographic polarization gratings, obtaining stable efficiency values [19,20]. Most of the reported azobenzene BCs are diblock linearelinear copolymers where the non-absorbing block is either poly(methyl methacrylate) (PMMA) [14,20] or poly(styrene) (PS) [21,22]. Usually, controlled radical polymerizations including atom transfer radical polymerization (ATRP), ring opening metathesis polymerization (ROMP) and reversible addition fragmentation chain transfer polymerization (RAFT) have been employed to prepare such linearelinear azobenzene BCs. However, radical polymerization of azobenzene monomers is not always straightforward. In this context, recent developments in ‘Click Chemistry’ have emerged to facilitate the access to complex and novel polymeric architectures with well-defined structures as well as to incorporate functionality into polymeric systems [23e27]. An example of these novel architectures are the linearedendritic BCs that were introduced by

E. Blasco et al. / Polymer 53 (2012) 4604e4613

Gitsov and Fréchet and comprise hybrid macromolecules containing linear segments linked to dendrons [28e30]. The great advantage of appending azobenzene moieties at the periphery of a dendron instead of using a linear block is that a very precise control can be exerted on the number of the photoresponsive units introduced in the macromolecule at the same time that the polymerization of azobenzene monomers is circumvented. There are three basic strategies which can be used for the synthesis of these linearedendritic BCs: ‘chain-first’ route, ‘dendron-first’ route and the coupling of the preformed blocks. The ‘chain-first’ route consists in the synthesis of a terminally functional polymer followed by a divergent dendron construction [31,32]. The ‘dendron-first’ route, developed by Matyjaszewski and, Hawker and Fréchet, implies the preparation of a dendron functionalized at the focal point which is used as a macroinitiator for the polymerization of the linear block [33,34]. Finally, the coupling strategy requires the previous synthesis of the linear and dendron segments, followed by their coupling through complementary functional groups at their end-chain and the focal point, respectively. Although this is the most versatile synthetic approach for the preparation of lineare dendritic BCs, it relies on the availability of highly efficient and selective chemistry under mild conditions, which are the main features of the ‘click chemistry’ reactions [35e40]. Recently, we have synthesized a series of photoresponsive linearedendritic BCs with azobenzene units by coupling the first four generations of dendritic aliphatic polyesters based on 2,2di(hydroxymethyl)propionic acid (bis-MPA) functionalized at the periphery with 4-cyanoazobenzene moieties to either poly(ethylene glycol) (PEG) or PMMA linear segments [38,41,42]. The BCs were obtained by a Huisgen’s 1,3-dipolar cycloaddition between the azodendron bearing an azido group at the focal point and the alkyne-terminated linear block catalysed by copper salts proving the effectiveness of the approach. In these studies, it has been found that the incorporation of PEG is appropriate for the preparation of supramolecular aggregates in water but not as much for optical applications due to its high crystallinity and low glass transition, below room temperature (r.t.). Even so, stable photoinduced anisotropy (with an order parameter above 0.50) was achieved by irradiating (at 488 nm) the higher dendron generation, with sixteen azobenzene units. For optical applications it is preferable the use of amorphous linear blocks and PMMA has been used which confers processability and transparency to the linear-dendritic BCs although the achieved photoinduced anisotropy is lower than in PEGeazodendron BCs. To extend our study about the influence of the linear block on the properties of linearedendritic BCs, we report in this paper the synthesis and thermal characterization of linear-dendritic BCs containing a poly(ethyl methacrylate) (PEMA) or a PS linear block linked to a liquid crystalline aliphatic polyester dendron functionalized with sixteen 4-cyanoazobenzene moieties, as well as the analogous with PMMA in order to establish comparisons. PS, which is easily synthesized by ATRP, has lower polarity than PMMA and is aromatic in character as the azobenzene. PEMA has a chemical structure that is closely related to that of PMMA but has a lower Tg. Morphology, optical properties and photoinduced response of these BCs have also been evaluated. 2. Experimental section 2.1. Chemicals Methyl methacrylate (Aldrich, 99%), ethyl methacrylate (Aldrich, 99%) and styrene (Aldrich, 99%) were passed through a basic alumina column, stored over CaH2, and vacuum-distilled before use. 1,1,4,7,7-Pentamethyldiethylenetriamine (PMDETA) was used

4605

as received. Copper(I) bromide was used as received and handled in a dry box. Tetrahydrofurane (THF) was freshly distilled under argon atmosphere. All other commercially available reagents were purchased from Aldrich and used as received without further purification. The ATRP inititator 3-(trimethylsilyl)prop-2-ynyl 2bromo-2-methylpropanoate, the azide-functionalized PS resin and 11-[4-(40 -cyanophenylazo)phenyloxy]undecanoic acid were prepared according to literature procedures [43,44]. Experimental details for the synthesis of the precursor dendron d16OH are given in the supporting information. 2.2. Synthetic procedures and characterization data 2.2.1. Synthesis of the azodendron d16AZO Compound d16OH (0.75 g, 0.41 mmol), 11-[4-(40 -cyanophenylazo)phenyloxy]undecanoic acid (3.11 g, 7.62 mmol) and 4-(dimethylamino)pyridinium 4-toluenesulfonate (DPTS) (1.87 g, 6.42 mmol), were dissolved in a mixture of dichloromethane (40 mL) and N,N-dimethylformamide (DMF) (15 mL). The reaction flask was flushed with argon, and N,N’-dicyclohexylcarbodiimide (DCC) (1.73 g, 8.39 mmol) was added. The mixture was stirred at r.t. for 48 h under argon atmosphere. The white precipitate formed was filtered off, and the solvent was evaporated. The crude product was purified by flash column chromatography on silica gel and eluted with dichloromethane, gradually increasing the polarity to 1:10 ethyl acetate/dichloromethane. The target azodendron was obtained as a red powdery solid. Yield: 55%. IR (KBr), n (cm1): 2227 (CN), 2096 (N3), 1741 (C]O), 1600, 1582, 1501 (Ar), 1257 (CeO), 859 (Ar). 1H NMR (CDCl3, 400 MHz) d (ppm): 7.93e7.91 (m, 64H), 7.79e 7.77 (m, 32H), 7.00e6.98 (m, 32H), 4.36e4.11 (m, 62H), 4.02 (t, 32H, J ¼ 6.5 Hz), 3.29 (t, 2H, J ¼ 6.7 Hz), 2.31 (t, 32H, 7.5 Hz), 1.81e1.78 (m, 32H), 1.64e1.56 (m, 32H), 1.50e1.40 (m, 36H), 1.39e1.24 (m, 209H). 13 C-RMN (CDCl3, 100 MHz) d (ppm): 173.1, 172.0, 162.6, 154.7, 146.6, 133.1, 125.4, 123.0, 118.6, 114.8, 113.2, 68.4, 64.7, 46.3, 34.0, 29.6, 29.5, 29.4, 29.3, 29.2, 26.0, 24.8, 17.8. EM (MALDIþ, dithranol) m/z: 8116.9 [M]þ. Anal. Calcd for C465H565N51O78: C, 68.82; H, 6.97; N, 8.81. Found: C, 68.34; H, 7.22; N, 8.64. 2.2.2. Synthesis of the linear blocks PMMA, PEMA and PS were synthesized by ATRP using 3-(trimethylsilyl)prop-2-ynyl 2-bromo-2-methylpropanoate, an initiator with a protected alkyne function that was subsequently deprotected. 2.2.2.1. PMMA polymerization. Methyl methacrylate (18.7 g, 0.19 mol), PMDETA (200 mL, 0.94 mmol), CuBr (134.3 mg, 0.94 mmol) and the initiator (260.2 mg, 0.94 mmol) were added to a Schlenk tube. The reaction mixture was degassed by three freezee pumpethaw cycles and flushed with argon. The polymerization was carried out in a thermostated oil bath at 90  C. After 5 min for PMMA1eTMS or 10 min for PMMA2eTMS the polymerization mixture was diluted with THF, passed through a column of neutral alumina to remove the catalyst and precipitated into methanol. The polymer was dried in a vacuum oven at 40  C for 24 h. 2.2.2.2. PEMA polymerization. Ethyl methacrylate (13.79 g, 0.12 mol), PMDETA (144.1 mL, 0.69 mmol), CuBr (99.1 mg, 0.69 mmol) and the initiator (190.2 mg, 0.69 mmol) were added to a Schlenk tube. The reaction mixture was degassed by three freezeepumpethaw cycles and flushed with argon. The polymerization was carried out in a thermostated oil bath at 90  C. After 3 min for PEMA1eTMS or 8 min for PEMA2eTMS the polymerization mixture was diluted with THF, passed through a column of neutral alumina to remove the catalyst and precipitated into hexane. The polymer was dried in a vacuum oven at 40  C for 24 h.

4606

E. Blasco et al. / Polymer 53 (2012) 4604e4613

2.2.2.3. PS polymerization. Styrene (13.59 g, 0.13 mol), PMDETA (and 21.3 mL, 0.11 mmol), CuBr (14.6 mg, 0.11 mmol) and the initiator (28.3 mg, 0.11 mmol) were added to a Schlenk tube. The reaction mixture was degassed by three freezeepumpethaw cycles and flushed with argon. The polymerization was carried out in a thermostated oil bath at 110  C. After 45 min for PS1-TMS or 4 h for PS2-TMS the polymerization mixture was diluted with THF, passed through a column of neutral alumina to remove the catalyst and precipitated into methanol. The polymer was dried in a vacuum oven at 40  C for 24 h. 2.2.2.4. General alkyne deprotection procedure. A 0.01 M solution of the protected alkyne-terminated polymer in THF was prepared and a five-fold excess of 1.0 M solution of tetra-n-butylammonium fluoride (TBAF) in THF with respect to trimethylsilyl (TMS) group was added dropwise. The reaction mixture was stirred overnight at r.t. and the product was precipitated into cold methanol. The alkyne-ended linear polymer was dried at 40  C under vacuum for 48 h. 2.2.2.5. Characterization data for PMMA1. IR (KBr), n (cm1): 1728 (C]O), 1240, 1150. 1H NMR (CDCl3, 400 MHz) d (ppm): 4.68e4.58 (m), 3.59 (s), 2.06e1.75 (m), 1.48e1.38 (m), 1.26e1.13 (m), 1.10e 0.80 (m). Anal. Calcd: C, 59.98%; H, 8.05%. Found: C, 60.30%; H, 7.89. SEC: Mn ¼ 12,100, Mw/Mn ¼ 1.04 (PMMA standards). 2.2.2.6. Characterization data for PMMA2. IR (KBr), n (cm1): 1729 (C]O), 1242, 1147. 1H NMR (CDCl3, 400 MHz) d (ppm): 4.65e4.52 (m), 3.60 (s), 2.07e1.72 (m), 1.50e1.35 (m), 1.26e1.13 (m), 1.09e 0.70 (m). Anal. Calcd: C, 59.98%; H, 8.05%. Found: C, 59.20%; H, 7.80%. SEC: Mn ¼ 20,200, Mw/Mn ¼ 1.04 (PMMA standards). 2.2.2.7. Characterization data for PEMA1. IR (KBr), n (cm1): 1728 (C]O), 1269, 1146. 1H NMR (CDCl3, 400 MHz) d (ppm): 4.60e4.36 (m), 4.02 (c, J ¼ 6.7 Hz), 2.10e1.70 (m), 1.34e1.10 (m), 1.10e0.80 (m). Anal. Calcd: C, 63.14%; H, 8.83%. Found: C, 63.34%; H, 9.13%. SEC: Mn ¼ 11,800, Mw/Mn ¼ 1.22 (PMMA standards). 2.2.2.8. Characterization data for PEMA2. IR (KBr), n (cm1): 1729 (C]O), 1272, 1147. 1H NMR (CDCl3, 400 MHz) d (ppm): 4.60e4.36 (m), 4.02 (c, J ¼ 6.7 Hz), 2.10e1.70 (m), 1.24e1.10 (m), 1.10e0.80 (m). Anal. Calcd: C, 63.14%; H, 8.83%. Found: C, 63.42%; H, 9.08%. SEC: Mn ¼ 22,800, Mw/Mn ¼ 1.20 (PMMA standards). 2.2.2.9. Characterization data for PS1. IR (KBr), n (cm1): 1601, 1493, 756, 698 (Ar). 1H NMR (CDCl3, 400 MHz) d (ppm): 7.36e6.89 (m), 6.85e6.30 (m), 4.60e4.36 (m), 4.08e4.01 (m), 2.29 (s), 2.05e1.65 (m) 1.62e0.85 (m). Anal. Calcd: C, 92.26%; H, 7.74%. Found: C, 91.99%; H, 7.89%. SEC: Mn ¼ 11,100, Mw/Mn ¼ 1.04 (PS standards). 2.2.2.10. Characterization data for PS2. IR (KBr), n (cm1): 1601, 1492, 756, 697 (Ar). 1H NMR (CDCl3, 400 MHz) d (ppm): 7.36e6.89 (m), 6.85e6.30 (m), 4.60e4.36 (m), 4.08e3.98 (m), 2.29 (s), 2.03e 1.65 (m) 1.63e0.85 (m). Anal. Calcd: C, 92.26%; H, 7.74%. Found: C, 91.81%; H, 7.98%. SEC: Mn ¼ 19,500, Mw/Mn ¼ 1.05 (PS standards). 2.2.3. Coupling of the preformed blocks Azodendron d16AZO, 1.2 fold excess of alkyne-functionalized polymer and two-fold excess of CuBr were placed into a Schlenk tube. Two-fold excess of PMDETA and deoxygenated DMF (around 1 mL per 100 mg of polymer) were added with an argon-purged syringe, and the flask was further degassed by three freezee pumpethaw cycles and flushed with argon. The reaction mixture was stirred at 40  C for 72 h. Then, an azido-functionalized resin was added under argon flow to remove alkyne-functionalized

polymer excess and the reaction mixture was stirred for further 24 h. The resin was filtered off, the mixture diluted with THF and then passed through a short column of neutral alumina. The solvent was partially evaporated and the resulting polymer solution carefully precipitated into cold methanol. 2.2.3.1. Characterization data for PMMA1-b-d16AZO. IR (KBr), n (cm1): 2228 (CN), 1727 (C]O), 1601, 1582, 1500 (Ar), 1256, 1141 (CeO), 849 (Ar). 1H NMR (CDCl3, 400 MHz) d (ppm): 7.91e7.89 (m), 7.76e7.64 (m), 7.56 (s), 6.98e6.95 (m), 5.15e5.11 (m), 4.71e4.68 (m), 4.36e4.11 (m), 4.01 (t), 3.59 (s) 2.28 (t), 2.10e1.70 (m), 1.65e 1.48 (m), 1.47e1.34 (m), 1.34e1.10 (m), 1.10e0.80 (m). Anal. Calcd: C, 62.47%; H, 7.69%; N, 3.88%. Found: C, 62.71%; H, 7.38%; N, 3.53%. 2.2.3.2. Characterization data for PMMA2-b-d16AZO. IR (KBr), n (cm1): 2228 (CN), 1728 (C]O), 1600, 1582, 1499 (Ar), 1265, 1146 (CeO), 852 (Ar). 1H NMR (CDCl3, 400 MHz) d (ppm): 7.91e7.89 (m), 7.76e7.64 (m), 7.55 (s), 6.98e6.95 (m), 5.14e5.11 (m), 4.36e4.11 (m), 3.99 (t), 3.60 (s), 2.28 (t), 2.10e1.70 (m), 1.65e1.48 (m), 1.47e 1.34 (m), 1.34e1.10 (m), 1.10e0.80 (m). Anal. Calcd: C, 62.34%; H, 7.68%; N, 2.47%. Found: C, 62.63%; H, 7.41%; N, 2.11%. 2.2.3.3. Characterization data for PEMA1-b-d16AZO. IR (KBr), n (cm1): 2227 (CN), 1727 (C]O), 1600, 1582, 1501 (Ar), 1257, 1141 (CeO), 851 (Ar). 1H NMR (CDCl3, 400 MHz) d (ppm): 7.91e7.89 (m), 7.76e7.64 (m), 7.56 (s), 6.98e6.95 (m), 5.15e5.11 (m), 4.71e4.68 (m), 4.36e4.11 (m), 4.09e3.91 (m), 2.28 (t), 2.10e1.70 (m), 1.65e 1.48 (m), 1.47e1.34 (m), 1.34e1.10 (m), 1.10e0.80 (m). Anal. Calcd: C, 64.49%; H, 8.39%; N, 2.10%. Found: C, 64.92%; H, 7.95%; N, 2.67%. 2.2.3.4. Characterization data for PEMA2-b-d16AZO. IR (KBr), n (cm1): 2228 (CN), 1728 (C]O), 1601, 1583, 1501 (Ar), 1265, 1145 (CeO), 858 (Ar). 1H NMR (CDCl3, 400 MHz) d (ppm): 7.91e7.89 (m), 7.76e7.64 (m), 7.55 (s), 6.98e6.95 (m), 5.14e5.11 (m), 4.36e4.11 (m), 4.09e3.91 (m), 2.28 (t), 2.10e1.70 (m), 1.65e1.48 (m), 1.47e 1.34 (m), 1.34e1.10 (m), 1.10e0.80 (m). Anal. Calcd: C, 64.51%; H, 8.33%; N, 2.29%. Found: C, 64.06%; H, 7.97%; N, 1.99%. 2.2.3.5. Characterization data for PS1-b-d16AZO. IR (KBr), n (cm1): 2227 (CN), 1741 (C]O), 1600, 1582, 1493 (Ar), 1255 (CeO), 848, 757, 698 (Ar). 1H NMR (CDCl3, 400 MHz) d (ppm): 7.91e7.89 (m), 7.77e 7.64 (m), 7.51e6.89 (m), 6.85e6.30 (m), 4.36e4.11 (m), 4.00 (t), 2.28 (t), 1.97e1.63 (m), 1.63e0.89 (m). Anal. Calcd: C, 82.30%; H, 7.42%; N, 3.74%. Found: C, 82.68%; H, 7.43%; N, 3.59%. 2.2.3.6. Characterization data for PS2-b-d16AZO. IR (KBr), n (cm1): 2227 (CN), 1741 (C]O), 1600, 1582, 1493 (Ar), 1256 (CeO), 848, 756, 698 (Ar). 1H NMR (CDCl3, 400 MHz) d (ppm): 7.91e7.89 (m), 7.77e 7.64 (m), 7.51e6.89 (m), 6.85e6.30 (m), 4.36e4.11 (m), 4.01 (t), 2.28 (t), 1.97e1.63 (m), 1.63e0.89 (m). Anal. Calcd: C, 85.37%; H, 7.52%; N, 2.59%. Found: C, 85.68%; H, 7.90%; N, 2.07%. 2.3. Characterization techniques FT-IR spectra were obtained on a Nicolet Avatar 360-FT-IR espectrometer using KBr pellets. 1H and 13C NMR spectra were recorded on a Bruker AV-400 spectrometer operating at 400 MHz and 100 MHz, respectively. Elemental analyses were performed using a PerkineElmer 2400 microanalyzer. MALDI-TOF MS was performed on an Autoflex mass spectrometer (Bruker Daltonics). UVeVis spectra were determined on an ATI-Unicam UV4-200 espectrophotometer. Size Exclusion Chromatography (SEC) was carried out on a Waters e2695 Alliance liquid chromatography system equipped with a Waters 2424 evaporation light scattering detector and a Waters 2998 PDA detector using two UltrastyragelÒ

E. Blasco et al. / Polymer 53 (2012) 4604e4613

A pore size. columns, HR 4 and HR 2 from Waters, of 500 and 104  Measurements were performed in THF with a flow of 1 mL/min using poly(methyl methacrylate) (PMMA) or polystyrene (PS) narrow molecular weight standards. Preparative Size Exclusion Chromatography (SEC) was carried out on a Waters 600 pump and a Waters 2998 PDA detector using two UltrastyragelÔ columns, A pore size. Measurements were 19  300 mm, of 500 and 104  performed in THF with a flow of 6 mL/min. Thermogravimetric analysis (TGA) was performed using a Q5000IR from TA Instruments under nitrogen atmosphere using 5e10 mg of the sample. Thermal transitions were determined by Differential Scanning Calorimetry (DSC) using a DSC Q2000 from TA Instruments with powdered samples (2e5 mg) sealed in aluminium pans. Glass transition temperatures were determined at the midpoint of the baseline jump and the isotropic temperatures were determined at the maximum of the corresponding peaks. Mesomorphic behaviour was also evaluated by polarizing optical microscopy (POM) using an Olympus BH-2 polarizing microscope fitted with a Linkam THMS600 hot stage. Morphology of the block copolymers has been studied by TEM in a JEOL-2000 FXIII electron microscope operating at 200 kV. Film thickness was measured using a DEKTAK profilometer. A varian Cary 500 UVeViseIR spectrophotometer was used for optical absorption and dichroism studies. Birefringence (Dn) measurements were performed using the standard setup previously reported [14]. 3. Results and discussion 3.1. Synthesis and characterization of the linearedendritic BCs The copper(I)-catalysed azide-alkyne [3 þ 2] cycloaddition was chosen as the suitable coupling method for the synthesis of target linearedendritic BCs containing a PMMA, PEMA or PS linear block linked to an aliphatic polyester dendron (Fig. 1). For the azobenzene block, the fourth generation of a polyester dendron based on the bis-MPA acid bearing sixteen 4cyanoazobenzene moieties linked through a decamethylenic spacer and with an azido functional group at the focal point was used. The dendron was synthesized using previously reported procedures [38]; however, a hexamethylenic spacer was used this time to incorporate the azido group at the focal point merely

4607

because it implies the manipulation of 6-azido-1-hexanol instead of the more dangerous 2-azido-1-ethanol used in [38]. The linear blocks were prepared by ATRP techniques using an initiator containing a trimethylsilyl protected-alkyne group (Fig. 2). Linear blocks of PMMA, PEMA and PS with two different average molecular weights of approx. 1  104 and 2  104 were prepared. The polymerizations were performed in bulk at 90  C (for PMMA and PEMA) or 110  C (for PS) according to procedures reported in the literature [41,45]. Polymerization times were adjusted to obtain different molecular weights (see Experimental Part). Average molar masses of the linear blocks were determined by end-group analysis of the TMS-ended polymers by 1H NMR using the relative integral of the eSi(CH3)3 and eCOOCH3 of PMMA eCOOCH2e of PEMA or aromatic protons of PS. Molar masses were also determined by SEC, data are gathered in Table 1. SEC traces of linear blocks show monomodal molar mass distributions. Low polydispersities (PI < 1.1) were determined for PMMA and PS homopolymers and slightly higher (PI z 1.2) for PEMA ones. This increment in the molecular weight distribution could arise from termination processes [46]. The trimethylsilyl protected-alkyne functionalized linear blocks were deprotected with TBAF and coupled with the azodendron in DMF using copper (I) as catalyst. A slight excess of the alkyne functionalized linear block was used and then it was removed onto an azido-functionalized polystyrene resin. The efficiency of the coupling was asserted from SEC traces by the shift of the molar mass distribution peak towards lower retention times that indicates linear-dendritic BC formation (Fig. 3). For PMMA and PEMAcontaining linearedendritic BCs, evidence of residual azodendron was not observed in SEC traces. This was not the case of PS lineare dendritic BCs, where a small peak corresponding to residual azodendron was detected in the SEC curve (Fig. S2 in Supporting information). Therefore, preparative SEC was used in order to purify completely the linearedendritic BC. Further evidence for the formation of the BCs was gained from the IR spectra, where the band at 2100 cm1 due to the azide functionality of the azodendron completely disappeared (Fig. S3 in Supporting information). The 1H NMR spectra of the BCs also confirm the coupling. Relative integration of azobenzene aromatic protons signals and the corresponding ones to the linear block protons, eCOOCH3 of PMMA at 3.60 ppm, eCOOCH2e of the PEMA at 4.02 ppm or aromatic protons of the PS at 6.50 ppm, is in good

Fig. 1. Coupling strategy and chemical structure of the investigated linearedendritic block copolymers.

4608

E. Blasco et al. / Polymer 53 (2012) 4604e4613

Fig. 2. Synthesis of the alkyne-terminated linear blocks (see Table 1 for the corresponding average molar mass).

agreement with the linear-dendritic BCs structure. Besides the signals due to the dendron and the linear block, new peaks related to the triazol ring formation appeared at 7.56 ppm (triazol ring proton), and at 5.15 and 4.70 ppm (methylenic protons linked to it) (Fig. 4).

3.2. Thermal characterization and morphological study Thermal stability of the linearedendritic BCs as well as of the isolated blocks was studied by TGA using powdered samples. Weight losses associated to the presence of residual solvents or water were not detected. From the TGA curves, significant differences were observed for PMMA, PEMA and PS linearedendritic BCs (Table S1 in Supporting information). PMMA and PEMA lineare dendritic BCs showed major weight losses associated to sample decomposition above 340  C and 315  C, respectively. PS imparts superior thermal stability with major weight losses associated to sample decomposition above 390  C. Thermal transitions were studied by DSC and POM. Relevant data are collected in Table 2. The DSC curve of the azodendron d16AZO presents a glass transition at 22  C and a peak at 141  C corresponding to mesophase-to-isotropic transition (Fig. 5). POM

Table 1 Molecular weight and composition of the synthesised polymers. Polymer

Mn (NMR)

Mn (GPC)c

PDIc

PMMA1-TMS PMMA1 PMMA2-TMS PMMA2 PEMA1-TMS PEMA1 PEMA2-TMS PEMA2 PS1-TMS PS1 PS2-TMS PS2 PMMA1-b-d16AZO PMMA2-b-d16AZO PEMA1-b-d16AZO PEMA2-b-d16AZO PS1-b-d16AZO PS2-b-d16AZO

10,300a e 19,800a e 9120a e 18,240a e 10,500a e 20,490a e 18,420b 27,920b 17,240b 26,360b 18,620b 28,700b

11,800 12,100 19,100 20,200 11,500 11,800 23,600 22,800 10,900 11,100 19,100 19,500 16,200 21,600 20,900 32,400 16,700 23,500

1.05 1.04 1.05 1.04 1.26 1.22 1.16 1.20 1.04 1.04 1.05 1.05 1.08 1.09 1.14 1.19 1.08 1.08

a Mn calculated from 1H NMR by comparison of the integration of the methyl signal (0.18 ppm) corresponding to the TMS group with signals of the repetitive units of the polymer (see text). b Calculated by the sum of the linear block Mn calculated by NMR plus the molecular weight of d16AZO. c Values for PMMA and PEMA polymers and their corresponding BCs were determined by SEC using PMMA standards. Mn and PDI of PS polymers and their corresponding BCs were determined by SEC using PS standards.

images of the azodendron showed fan-shaped textures characteristic of a smectic A mesophase. On the contrary, the linear blocks are essentially amorphous materials. DSC curves showed a clear baseline jump corresponding to the glass transition with Tg values of around 115  C for PMMA, 70  C for PEMA and 100  C for PS. As it was expected, Tg of PEMA homopolymers is significant lower in comparison to PMMA one. The investigated BCs exhibited DSC curves where two Tg are evident coincident with those of the dendritic and linear blocks. The lowest Tg, at 33e34  C, corresponds to the glass transition of the azodendron block even if the estimated values are slightly higher (about 10  C) than that of d16AZO. The highest Tg corresponds to the linear block, and values are also slightly higher than those of the corresponding homopolymers. All the lineardendritic BCs show a peak corresponding to the mesophase-toisotropic liquid transition. A comparison between DSC curves of the azodendron, a linear block (PS2) and the corresponding linearedendritic BC (PS2-b-d16AZO) is shown in Fig. 5. For PMMA linear-dendritic BCs, the Tg (around 115  C) overlaps the mesophase-to-isotropic liquid transition but not for PEMA and PS BCs due to the lower Tg values (around 70  C and 100  C) of the linear block (Fig. S4 in Supporting information). All these lineare dendritic BCs show liquid crystalline behaviour (Fig. S5 in Supporting information). Nanostructure of the BCs has been studied by TEM. Small pellets of the linear-dendritic BCs were prepared by heating the powder samples at 180  C (for about 2 min) and subsequent fast cooling to

Fig. 3. SEC traces of PS2-b-d16AZO and their precursors d16AZO and PS2.

E. Blasco et al. / Polymer 53 (2012) 4604e4613

4609

Fig. 5. DSC curves of PS2-b-d16AZO and the corresponding precursors d16AZO and PS2.

Fig. 4. 1H NMR spectrum of PEMA1-b-d16AZO in CDCl3 at 400 MHz.

r.t. Pellets were annealed for 1 h at 140  C and fast cooled again to r.t. It was checked that longer annealing times at 140  C do not introduce any significant change neither in the nanostructure nor in the photoresponse. Then, thin slices (of about 100 nm thickness) were cut from the pellets, put on copper grids and stained with RuO4. TEM images show a lamellar nanostructure for all the compounds. As an example, TEM images showing the nanostructure corresponding to the three BCs with higher molecular weights are presented in Fig. 6. 3.3. Optical properties and photoinduced response Optical absorption measurements for BCs were recorded in dichloromethane solutions and in films prepared by casting. Since the recorded spectra were very similar for all the BCs, only those corresponding to PEMA2-b-d16AZO are given in Fig. 7 as an example. Solution spectrum shows the typical bands of this type of azobenzene units, a main band at around 365 nm and a weak band at around 450 nm that correspond to the pep* and the nep* Table 2 Transition temperatures determined by DSC and mesophase type determined by POM of the linearedendritic BCs and their building blocksa. Polymer

Tg1

Tg2

Ti

D Hi

Mesophase

d16AZO PMMA1 PMMA2 PEMA1 PEMA2 PS1 PS2 PMMA1-b-d16AZO PMMA2-b-d16AZO PEMA1-b-d16AZO PEMA2-b-d16AZO PS1-b-d16AZO PS2-bd16AZO

22 e e e e e e 32 36 33 33 34 34

e 115 113 69 66 97 98 116b 116b 70 76 102 102

141 e e e e e e 135b 134b 134 133 134 141

78.8 e e e e e e 63.4b 67.4b 64.9 69.4 34.6 39.7

SmA e e e e e e LC LC LC LC LC LC

a Transition temperatures and enthalpies were determined by DSC from the second heating scan (10  C/min. Temperatures are given in  C, enthalpies given in kJ per mole. Tg ¼ glass transition, Ti ¼ isotropization (M  I), DHi ¼ enthalpy associated to isotropization, SmA ¼ smectic A, LC ¼ liquid crystal phase. b Data can not be calculated accurately because Ti overlaps with Tg2.

electronic transitions of the azobenzene trans isomer, respectively. Besides these two bands, the ‘as casted’ films show another band at around 330 nm that is associated with H-aggregates of azobenzenes. The absorption spectrum of the films after thermal treatment (2 min at 180  C and 1 h at 140  C) is given in Fig. 8. The spectra of the all the BCs but the one corresponding to PS2-bd16AZO are similar to that of the ‘as casted’ films, showing the presence of azobenzene aggregates. Therefore, only the spectrum corresponding to PEMA2-b-d16AZO is shown. On the other hand PS2-b-d16AZO copolymer shows a spectrum close to the one found in solution. This indicates that for this last compound the thermal treatment is able to break azobenzene aggregates. Photoinduced anisotropy in the thermally treated films has been studied by birefringence and dichroism measurements. In the setup for birefringence measurements the film is placed in a HeeNe laser beam (633 nm, 30 mW/cm2) between two crossed polarizers with the polarization directions at þ45 and 45 with the vertical direction. It has been checked that this red light does not induce any effect on the optical properties of the samples. The red light intensity transmitted during irradiation with vertically polarized 488 nm light (irradiation temperature 25  C and exciting light power 300 mW/cm2) was measured as a function of the time. The transmitted intensity (I) is given by the equation:

I ¼ I0 sin2 ðpjDnjd=lÞ where I0 is the intensity transmitted by the unirradiated films with parallel polarizers, jDnj the modulus of birefringence, d the film thickness, and l the wavelength of the measuring light (633 nm). The jDnj values as a function of the time are plotted in Fig. 9. These jDnj values have been normalized (jDnjN) dividing them by the azo concentration (jDnjN ¼ 100 jDnj/x, where x is the %azo content) of the different samples. The evolution of jDnjN for PS1-b-d16AZO and PS2-b-d16AZO is similar: (jDnjN) increases and reaches an almost stationary value in less than 1 h. When the 488 nm light is switched off, jDnjN shows a strong decrease reaching a final value in about 1h. It has been checked that increasing irradiation time and/or the time after switching off 488 nm light (up to 5 h) the final jDnjN values do not show significant changes. The final jDnjN values are similar for the two compounds (around 0.01). For PMMA and PEMA BCs, birefringence keeps growing for times longer than 1 h. In fact, saturation has only been achieved after 5 h irradiation in our experimental conditions. The decrease of jDnjN when blue light

4610

E. Blasco et al. / Polymer 53 (2012) 4604e4613

Fig. 6. TEM bright field micrographs of (a) PEMA2-b-d16AZO, (b) PS2-b-d16AZO and (c) PMMA2-b-d16AZO BCs. The length of the white bar corresponds to 200 nm.

irradiation is switched off is still high for PEMA2-b-d16AZO but smaller for PEMA1-b-d16AZO and the two PMMA BCs. The final value of jDnjN after 5 h irradiation and 5 h with the 488 nm light off is around 0.04 for PEMA2-b-d16AZO and between 0.1 and 0.13 for the other three BCs.

The photoinduced orientation of azo units has also been checked by dichroism measurements which were performed in films of the different materials after irradiation with linearly polarized 488 nm light (irradiation temperature 25  C and exciting power 300 mW/cm2). Films were irradiated for 1 h for PS BCs and

Fig. 7. Optical absorption spectrum of PEMA2-b-d16AZO in dichloromethane solution (dashed line) and in “as casted film” (solid line). All the BCs show similar spectra.

Fig. 8. Optical absorption spectra of PEMA2-b-d16AZO (solid line) and PS2-b-d16AZO (dashed line) after thermal treatment (2 min at 180  C and 1 h at 140  C).

E. Blasco et al. / Polymer 53 (2012) 4604e4613

4611

Fig. 9. Evolution of jDnjN (photoinduced birefringence normalized to the azobenzene content) of the BCs films. Irradiated 1 h with linearly polarized 488 nm light (25  C and 300 mW/cm2). Irradiation light has been switched off after 1 h.

5 h for the other BCs, and kept with the 488 nm light off for 1 h and 5 h respectively. The absorption spectrum was then measured with light polarized perpendicular (At) and parallel (Ak) to the polarization of the 488 nm light. The results are shown in Fig. 10. We define an in-plane order parameter h as:




h ¼ At  Ak

.
 At þ Ak

where Ak and At are the optical absorptions in the maximum of the main band. The h value is about 0.06 for the two PS BCs, while a h value around 0.3 has been obtained for PEMA1-b-d16AZO and the two PMMA containing BCs. PEMA2-b-d16 gives an h value around 0.16. Although all the BCs show a lamellar structure, they present very different photoinduced anisotropy. A low response has been obtained for PS-containing BCs while the PMMA BCs show a much higher response. In the case of PEMA BCs, different behaviour has been observed in PEMA1-b-d16AZO and in PEMA2-b-d16AZO, which only differs in the length of the linear block. Therefore, an unequivocal dependence of photoresponse either with the chemical structure or the molecular weight of the linear block is not

evident, although the methacrylic linear blocks give rise to better properties. A clear relationship with glass transition temperature is not observed. Although the Tg of PS and PMMA BCs are similar their photoinduced response is very different while the two PEMA containing BC have similar Tg value but different photoinduced anisotropy. Photoresponse can not either be related to the presence of azobenzene aggregates because the photoinduced anisotropy of PEMA and PMMA containing BCs and PS1-b-d16AZO is very different although all of them show similar aggregation, as can be seen in their optical absorption spectra. On the other hand, the photoresponse of PS2-b-d16AZO with very few aggregates is similar to that of PS1-b-d16AZO, which show a strong band associated to H aggregates. 4. Conclusions The synthesis of BCs having a linearedendritic architecture has been successfully approached by ‘click’ coupling of preformed blocks with azido and alkyne funtionalities. PMMA, PEMA and PS

4612

E. Blasco et al. / Polymer 53 (2012) 4604e4613

Fig. 10. Photoinduced dichroism in the BCs. PS1-b-d16AZO and PS2-b-16AZO were irradiated for 1 h with linearly polarized 488 nm light and kept 1 h with the exciting light off. PEMA1-b-d16AZO, PEMA2-b-d16AZO, PMMA1-b-d16AZO and PMMA2-b-d16AZO were irradiated for 5 h and kept 5 h with light off (see text).

have been synthesized by ATRP for the linear blocks containing and ethynyl end group. A dendron having sixteen peripheral 4cyanoazobenzene photoaddressable units and a 6-azidohexyl chain at the focal point was coupled to the linear segment by Cu(I)-catalysed dipolar 1,3-cycloaddition. This coupling was particularly efficient in the case of the polymethacrylic derivatives. All BCs exhibited liquid crystalline properties. Concerning photoinduced anisotropy the studied lineare dendritic BCs show different response, going from low values for those containing PS through an intermediate one for PEMA2-bd16AZO to a higher response in PEMA1-b-d16AZO and the two PMMA containing BCs. The observed differences can not be clearly correlated either with changes in microphase segregation, nor with composition, molecular weight or with the presence of azobenzene aggregates.

Acknowledgements This work was supported by the MINECO, Spain, under the projects MAT2011-27978-C02-01 and 02, Fondo Europeo de Desarrollo Regional (FEDER), Fondo Social Europeo and Gobierno de Aragon. E. Blasco acknowledges the CSIC JAE-Pre contract and C. Berges the FPI grant founding for their PhD. The authors acknowledge Dr Ma Ángeles Laguna of the Servicio de Microscopía Electrónica-Universidad de Zaragoza.

Appendix A. Supplementary material Supplementary data related to this article can be found online at http://dx.doi.org/10.1016/j.polymer.2012.08.022.

E. Blasco et al. / Polymer 53 (2012) 4604e4613

References [1] Ivanov S, Yakovlev I, Kostromin S, Shibaev V, Lasker L, Stumpe J, et al. Makromolekulare Chemie-Rapid Communications 1991;12(12):709e15. [2] Ichimura K. Chemical Reviews 2000;100(5):1847e73. [3] Hagen R, Bieringer T. Advanced Materials 2001;13(23):1805e10. [4] Natansohn A, Rochon P. Chemical Reviews 2002;102(11):4139e75. [5] Zhao Y, Ikeda T. Smart light-responsive materials: azobenzene-containing polymers and liquid crystals. Hoboken, New Jersey: Wiley-Interscience; 2009. [6] Ikeda T, Ube T. Materials Today 2011;14(10):480e7. [7] Yu H, Ikeda T. Advanced Materials 2011;23(19):2149e80. [8] Bandara HMD, Burdette SC. Chemical Society Reviews 2012;41(5):1809e25. [9] Hvilsted S, Sánchez C, Alcalá R. Journal of Materials Chemistry 2009;19(37): 6641e8. [10] Shishido A. Polymer Journal 2010;42(7):525e33. [11] Bruder F-K, Hagen R, Roelle T, Weiser M-S, Faecke T. Angewandte ChemieInternational Edition 2011;50(20):4552e73. [12] Hackel M, Kador L, Kropp D, Frenz C, Schmidt HW. Advanced Functional Materials 2005;15(10):1722e7. [13] Haeckel M, Kador L, Kropp D, Schmidt H-W. Advanced Materials 2007;19(2): 227. [14] Forcén P, Oriol L, Sánchez C, Alcalá R, Hvilsted S, Jankova K, et al. Journal of Polymer Science Part A-Polymer Chemistry 2007;45(10):1899e910. [15] Audorff H, Kreger K, Walker R, Haarer D, Kador L, Schmidt H-W. Advances in Polymer Science (Complex Macromolecular Systems II) 2010;228:59e121. [16] Yu H, Kobayashi T. Molecules 2010;15(1):570e603. [17] Kim JY, Fukuda T. Molecular Crystals and Liquid Crystals 2007;463:365e73. [18] Forcén P, Oriol L, Sánchez C, Rodríguez FJ, Alcalá R, Hvilsted S, et al. European Polymer Journal 2008;44(1):72e8. [19] Forcén P, Oriol L, Sánchez C, Rodríguez FJ, Alcalá R, Hvilsted S, et al. European Polymer Journal 2007;43(8):3292e300. [20] Gimeno S, Forcén P, Oriol L, Piñol M, Sánchez C, Rodríguez FJ, et al. European Polymer Journal 2009;45(1):262e71. [21] Chen W, Wei X, Balazs AC, Matyjaszewski K, Russell TP. Macromolecules 2011;44(5):1125e31. [22] Huang C-F, Chen W, Russell TP, Balazs AC, Chang F-C, Matyjaszewski K. Macromolecular Chemistry and Physics 2009;210(18):1484e92. [23] Kolb HC, Finn MG, Sharpless KB. Angewandte Chemie-International Edition 2001;40(11):2004.

4613

[24] Binder WH, Sachsenhofer R. Macromolecular Rapid Communications 2007; 28(1):15e54. [25] Fournier D, Hoogenboom R, Schubert US. Chemical Society Reviews 2007; 36(8):1369e80. [26] Kempe K, Krieg A, Becer CR, Schubert US. Chemical Society Reviews 2012; 41(1):176e91. [27] Durmaz H, Sanyal A, Hizal G, Tunca U. Polymer Chemistry 2012;3(4):825e35. [28] Gitsov I, Wooley KL, Frechet JMJ. Angewandte Chemie-International Edition in English 1992;31(9):1200e2. [29] Gitsov I. Journal of Polymer Science Part A-Polymer Chemistry 2008;46(16): 5295e314. [30] Wurm F, Frey H. Progress in Polymer Science 2011;36(1):1e52. [31] Iyer J, Fleming K, Hammond PT. Macromolecules 1998;31(25):8757e65. [32] vanHest JCM, Delnoye DAP, Baars M, ElissenRoman C, vanGenderen MHP, Meijer EW. Chemistry-A European Journal 1996;2(12):1616e26. [33] Matyjaszewski K, Shigemoto T, Frechet JMJ, Leduc M. Macromolecules 1996; 29(12):4167e71. [34] Leduc MR, Hawker CJ, Dao J, Frechet JMJ. Journal of the American Chemical Society 1996;118(45):11111e8. [35] Gitsov I, Wooley KL, Hawker CJ, Ivanova PT, Frechet JMJ. Macromolecules 1993;26(21):5621e7. [36] Malenfant PRL, Groenendaal L, Frechet JMJ. Journal of the American Chemical Society 1998;120(42):10990e1. [37] Hua C, Peng S-M, Dong C-M. Macromolecules 2008;41(18):6686e95. [38] del Barrio J, Oriol L, Alcalá R, Sánchez C. Macromolecules 2009;42(15):5752e60. [39] Tornoe CW, Christensen C, Meldal M. The Journal of Organic Chemistry 2002; 67(9):3057e64. [40] Javakhishvili I, Binder WH, Tanner S, Hvilsted S. Polymer Chemistry 2010; 1(4):506e13. [41] Del Barrio J, Oriol L, Alcalá R, Sánchez C. Journal of Polymer Science Part APolymer Chemistry 2010;48(7):1538e50. [42] del Barrio J, Oriol L, Sánchez C, Serrano JL, Di Cicco A, Keller P, et al. Journal of the American Chemical Society 2010;132(11):3762e9. [43] Opsteen JA, van Hest JCM. Chemical Communications 2005;(1):57e9. [44] Lober S, Rodriguez-Loaiza P, Gmeiner P. Organic Letters 2003;5(10):1753e5. [45] Durmaz H, Dag A, Erdogan E, Demirel AL, Hizal G, Tunca U. Journal of Polymer Science Part A-Polymer Chemistry 2010;48(1):99e108. [46] Maria Munoz-Molina J, Belderrain TR, Perez PJ. Macromolecules 2010;43(7): 3221e7.