Article pubs.acs.org/Macromolecules
Self-Assembly of Carbohydrate-block-Poly(3-hexylthiophene) Diblock Copolymers into Sub-10 nm Scale Lamellar Structures Yoko Sakai-Otsuka,†,§ Soraia Zaioncz,†,‡,§,∥,⊥ Issei Otsuka,†,§ Sami Halila,†,§ Patrice Rannou,‡,∥,⊥ and Redouane Borsali*,†,§ †
CERMAV and ‡INAC-SPrAM, Univ. Grenoble Alpes, F-38000 Grenoble, France CERMAV and ∥INAC-SPrAM, CNRS, F-38000 Grenoble, France ⊥ INAC-SPrAM, CEA, F-38000 Grenoble, France §
S Supporting Information *
ABSTRACT: We here report the synthesis of a new class of carbohydrate-based block copolymers, poly(3-hexylthiophene)-block-peracetylated maltoheptaose (P3HT-b-AcMal7) and poly(3-hexylthiophene)-block-maltoheptaose (P3HT-bMal 7 ), and their bulk and self-assembled thin films morphological characterizations by atomic force microscopy, transmission electron microscopy, and small-angle X-ray scattering. The block copolymers were synthesized via copper(I)-catalyzed 1,3-dipolar azide−alkyne cycloaddition of azido-functionalized AcMal7 and end-functionalized P3HT with alkyne group prepared by modified Grignard metathesis polymerization, followed by deacetylation of the AcMal7 block. The half-pitch of sub-10 nm scale lamellar structures, one of the smallest domain sizes of microphase separated block copolymers reported to date, was self-organized in the bulk and thin films of P3HT-b-AcMal7 by thermal annealing above the melting temperature of the P3HT segment. Meanwhile, thermodynamic microphase separation of P3HT-b-Mal7 was restricted due to strong inter- and intrachain hydrogen bonding among the hydroxyl groups of the Mal7 block, which was confirmed by an in situ stepwise heating and cooling Fourier transform infrared spectroscopy study.
■
INTRODUCTION Organic photovoltaics (OPVs) have emerged as a challenging new research field during the past two decades. Compared with classical inorganic systems, OPVs offer many potential advantages, such as mechanical flexibility, applicability for “roll-to-roll” processing, low-cost fabrication, mass production, short energy payback time (EPBT), etc.1−5 However, the demonstrated external power conversion efficiencies (EPCEs) of single junction donor/acceptor (D/A) bulk-heterojunction (BHJ) solar cells are limited to ca. 5−6%6 and 12.2%7 for first (e.g., P3HT) and second (e.g. push−pull alternated copolymers such as PTB7) generations materials, respectively. Beyond device architecture issues (direct vs inverted), this is mainly due on the material side to a rather limited exciton diffusion length (ca. 5−15 nm), unfavorable recombination processes, and poor charge carrier (holes and electrons) mobility displayed by organic semiconductors. These shortcomings are strongly influenced by the optoelectronically active D/A BHJ layer morphology and the crystallinity of the π-conjugated materials used.8,9 One of the promising approaches to improve EPCE is to design and control the active layer morphology between the donor and acceptor materials on a 5−15 nm scale that corresponds to the exciton diffusion length.10 In particular, a periodic lamellar structure in which the electron donor and acceptor compounds are mutually interpenetrating is consid© 2017 American Chemical Society
ered to be an ideal structure because this structure makes it possible to (i) suppress the exciton decay, (ii) provide the shortest charge transportation pathway, (iii) reduce charge trap sites, and (iv) promote charge separation of excitons by enlarging the interface contact area.11,12 The self-assembly of π-conjugated diblock copolymers has drawn considerable attention as a promising bottom-up strategy to achieve an ideal structure of the active layer in organic electronics.13−15 Among a variety of candidates for the electronically active polymer blocks, regioregular head-to-tail (HT) coupled poly(3-hexylthiophene) (P3HT) is one of the most important representative π-conjugated polymers owing to its excellent characteristics. These include high charge carrier mobility, solution processability, thermal and environmental stability, and synthetic versatility, etc. Hence, a number of P3HT-based diblock copolymers have been synthesized, and their self-assembly behaviors have been intensively studied.16−24 However, in most cases, a strong π−π interaction of P3HT hindered microphase separation of the diblock copolymers, resulting in nanofibril structures of crystallized P3HT blocks.16−18 In the diblock copolymer systems having Received: January 19, 2017 Revised: March 13, 2017 Published: April 5, 2017 3365
DOI: 10.1021/acs.macromol.7b00118 Macromolecules 2017, 50, 3365−3376
Article
Macromolecules
are studied in detail by means of atomic force microscopy (AFM), transmission electron microscopy (TEM), and smallangle X-ray scattering (SAXS) taking into consideration the thermal properties of the P3HT and oligosaccharidic blocks as investigated by differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA).
crystalline segments like P3HT, the self-assembly process is mainly governed by crystallization and microphase separation. To direct the counterbalance between crystallization and microphase separation toward the later, and to achieve the ideal lamellar structure with the domain feature comparable to the exciton diffusion length (5−15 nm), the diblock copolymer should have sufficiently high segregation strength between the two blocks and a small total degree of polymerization. Thermodynamic theories indicate that the phase behavior of A−B type diblock copolymers is regulated by the following three factors: (i) the volume fractions of each block, fA and f B (fA + f B = 1), (ii) the total degree of polymerization, N (N = NA + NB), and (iii) the Flory−Huggins interaction parameter, χ, which quantify the thermodynamic driving force for microphase separation. The strength of segregation of the two blocks is proportional to χN; that is, block copolymers are predicted to order into diverse morphologies as a function of f when χN is larger than a critical value (10.5 when fA = f B = 0.5).25 Therefore, block copolymers having high χ and low N values, so-called “high χ−low N block polymers”,26 have attracted significant interest to realize phase separation with much smaller features than one could ever achieve using “low χ” block copolymers such as polystyrene-block-poly(methyl methacrylate).27,28 For such purpose, we have reported a variety of carbohydrate-based block copolymer systems that self-organize into sub-10 nm scale periodic structures such as spheres, cylinders, and lamellae.29−34 The strong incompatibility, i.e. high χ, between carbohydrate and synthetic polymer blocks that arises from their hydrophilicity−hydrophobicity imbalance that is further enhanced sterically by the “rod”-like structure of the carbohydrate blocks is the key to achieve such small domain features. In this context, we here propose a new type of block copolymer consisting of P3HT and carbohydrate with the aim to self-organize them into lamellar structures with sub-10 nm domain features as a template for the ideal active layer in OPV. The carbohydrate blocks are designed for not only high segregation from the P3HT block but also for replacement with electron acceptors by selective decomposition of the glycosidic linkages upon chemical etching with an acidic medium followed by filling the resulting cavity with appropriate electronaccepting compounds. Indeed, Hillmyer et al.18 reported that the microphase-separated poly(3-alkylthiophene)-b-polylactide (P3AT-b-PLA) thin film gives nanoporous structures made of the P3AT segment after selective chemical etching of the PLA segment by exposure of the film to an alkaline solution. Later on, Botiz and Darling15 reported filling the cavity, which was prepared by selective etching of the poly(L-lactide) (PLLA) block from the microphase-separated P3HT-b-PLLA thin film in a similar manner as given in the report of Hillmyer et al., with fullerene C60 hydroxide as an electron acceptor material for photovoltaic applications. In this study, we report the synthesis of diblock copolymers consisting of P3HT and maltoheptaose (Mal7) or acetylated maltoheptaose (AcMal7) via the following three steps: (i) a modified Grignard metathesis (GRIM) polymerization of 3hexylthiophene to give the regioregular end-functionalized P3HT with alkyne group, (ii) preparation of azido-functionalized AcMal7, and (iii) coupling reaction of the P3HT and AcMal7 blocks via copper(I)-catalyzed 1,3-dipolar azide−alkyne cycloaddition (CuAAC) “click” reaction followed by deacetylation of the AcMal7 block. The self-organization of these block copolymers in bulk and thin film states via thermal annealing
■
EXPERIMENTAL SECTION
Materials. Maltoheptaose was purchased from NAGASE & CO., Ltd., Japan, and used as received. All other reagents were purchased from Sigma-Aldrich Chemicals Co., St. Louis, MO. tert-Butylmagnesium chloride (2.0 M in diethyl ether), [1,2-bis(diphenylphosphino)propane] dichloronickel(II) (Ni(dppp)Cl2), ethynylmagnesium bromide (0.5 M in THF), sodium azide (NaN3), 2-chloro-1,3-dimethylimidazolinium chloride (DMC), N,N-diisopropylethylamine (DIPEA), 4-(dimethylamino)pyridine (DMAP), acetic anhydride, pyridine, N,N,N′,N″,N″-pentamethyldiethylenetriamine (PMDETA), and copper(I) bromide (CuBr) were used as received. Tetrahydrofuran (THF) was refluxed and distilled from sodium benzophenone ketyl prior to use. Lithium chloride (LiCl) was dried in vacuo at 140 °C before use. 2,5-Dibromo-3-hexylthiophene and 1-hexene were stirred overnight over CaH2 and distilled prior to use. Instruments. 1H (400 MHz) and 13C (100 MHz) NMR spectra were recorded using a Bruker Avance DRX 400 MHz. Solid-state 13C NMR experiments were performed with a Bruker Avance DSX 400 MHz spectrometer operating at 100.6 MHz, using the combination of cross-polarization (CP), high-power proton decoupling, and magic angle spinning (MAS) method. The spinning speed was set at 12 kHz, sweep width 29761 Hz, and recycle delay 2 s. The spectra were averaged over 23 000 scans. The 13C chemical shifts were calibrated with glycine carboxyl group (176.03 ppm). FT-IR spectra were recorded using a PerkinElmer Spectrum RXI FT-IR spectrometer. Matrix-assisted laser desorption ionization mass spectrometry (MALDI-TOF MS) measurements were performed on a Bruker Daltonics Autoflex apparatus using dithranol as the matrix. Size exclusion chromatography (SEC) was performed at 30 °C using a Agilent 390-MDS system (290-LC pump injector, ProStar 510 column oven, 390-MDS refractive index detector) equipped with Knauer Smartline UV detector 2500 and two Agilent PLgel 5 μm MIXED-D 300 × 7.5 mm columns (Part No. PL1110-6504) in THF at the flow rate of 1.0 mL/min. TGA was performed using a SETARAM 92-12 TGA apparatus at a heating rate of 5 °C/min up to 800 °C under a nitrogen atmosphere. DSC analysis was performed using a TA Instruments DSC Q200 equipped with RCS 90 cooling unit. The samples (ca. 5 mg) were first dissolved in THF and deposited on the aluminum DSC pan followed by slow evaporation of the solvent. These DSC samples were stored at 100 °C under vacuum for several hours prior to analysis. The measurements were carried out under a nitrogen atmosphere at a scan rate of 10 °C/min. AFM phase images were realized in tapping mode by using a Picoplus microscope (Molecular Imaging, Inc., Tempe, AZ). TEM observations were carried out using a CM200 Philips microscope (Hillsboro, OR) operating at 80 kV. SAXS experiments were performed on the BM02 beamline of the European Synchrotron Radiation Facility (ESRF, Grenoble, France). Synthesis of Alkyne-Terminated P3HT. Alkyne-terminated poly(3-hexylthiophene) (alkyne-P3HT) was prepared by a modified GRIM polymerization method described in the literature35 under an argon atmosphere using oven-dried glassware. In a 100 mL Schlenk flask, LiCl (0.83 g, 19.4 mmol, 26 equiv) was placed and dissolved in freshly distilled dry THF (37 mL). The solution was stirred under argon at room temperature until LiCl was completely dissolved. Then, 2,5-dibromo-3-hexylthiophene (4.0 mL, 19 mmol, 25 equiv) and tertbutylmagnesium chloride (9.0 mL, 18 mmol, 24 equiv) were transferred to the flask via syringe under an argon atmosphere. The whole mixture was stirred at room temperature overnight in order to complete the metal−halogen exchange reaction. In another 500 mL Schlenk flask, Ni(dppp)Cl2 (0.39 g, 0.75 mmol, 1 equiv) was added and flame-dried, followed by the addition of 145 mL of dry THF. The 3366
DOI: 10.1021/acs.macromol.7b00118 Macromolecules 2017, 50, 3365−3376
Article
Macromolecules
Scheme 1. Strategy Used for the Synthesis of P3HT-b-AcMal7 and P3HT-b-Mal7 Diblock Copolymers: (a) Synthesis of AlkyneTerminated P3HT (Alkyne-P3HT) by Modified Grignard Metathesis Polymerization, (b) Synthesis of Azido EndFunctionalized Peracetylated Maltoheptaose (AcMal7-N3), and (c) “Click” Reaction of Alkyne-P3HT with AcMal7-N3 and Following Deacetylation of the Oligosaccharidic Block
z): calcd for [M + Na]+: 1200.38, found: 1200.37. FT-IR (CaF2): 3660−3020 cm−1 (OH, sugars), 2120 cm−1 (N3). Synthesis of Peracetylated Azido-Functionalized Maltoheptaose (AcMal7-N3). Mal7-N3 (1.00 g, 83.3 mmol) was peracetylated with acetic anhydride in pyridine (acetic anhydride:pyridine = 1:2 (v/ v), 30 mL) in the presence of a catalytic amount of DMAP. The mixture was stirred overnight at room temperature until the TLC (eluent: ethyl acetate/ether = 7/3) showed complete disappearance of the starting material. The solvent was then evaporated and redissolved in ethyl acetate. The solution was washed with 1 M HCl, saturated NaHCO3, and water. Then, the organic layer was dried over anhydrous Na2SO4 and concentrated by evaporation to afford peracetylated azido-functionalized maltoheptaose (AcMal7-N3) as a yellow solid (1.52 g, 86%). 1H NMR (400 MHz, CDCl3): δ (ppm) 6.18−3.75 (m, 49H) and 2.20−1.98 (m, 66H, CH3 acetyl group). MALDI-TOF MS (m/z): calcd for [M + Na]+: 2124.61, found: 2124.67. FT-IR (CaF2): 2120 cm−1 (N3), 1750 cm−1 (CO, esters). Synthesis of Poly(3-hexylthiophene)-b-Peracetylated Maltoheptaose (P3HT-b-AcMal7). In a two-neck round-bottom flask (flask 1), AcMal7-N3 (1.05 g, 0.50 mmol, 1.8 equiv) and PMDETA (88 μL, 0.42 mmol, 1.5 equiv) were dissolved in THF (10 mL), degassed by bubbling with argon for 5 min, and stirred at room temperature. In another two-neck round-bottom flask (flask 2), alkyne-P3HT (1.01 g, 0.28 mmol, 1 equiv) and CuBr (59 mg, 0.41 mmol, 1.5 equiv) were dissolved in THF (90 mL), degassed by bubbling with argon for 5 min, and stirred at room temperature. The content of flask 1 was transferred to flask 2 via cannula under an argon atmosphere. The reaction mixture was stirred at room temperature for 4 h until the SEC traces showed complete disappearance of the elution peak corresponding to the P3HT. The solution was passed through a neutral alumina column in order to remove copper salt. After concentration, the product was recovered by precipitation in cold acetone and dried under reduced pressure to yield the P3HT-b-AcMal7 block copolymer as a purple solid (1.05 g, 67%). 1H NMR (400 MHz, CDCl3): δ (ppm) 7.77 (s, triazole), 7.06−6.80 (m, aromatic), 5.96−
activated Grignard monomer solution was transferred into the 500 mL Schlenk flask using a cannula, stirred for 30 min at room temperature, and finally cooled to 0 °C. 1-Hexene (94.0 mL, 752 mmol, 1008 equiv) was added to the reaction mixture before the termination step according to a previous report36 in order to obtain the monofunctional polymer. The termination reaction and functionalization were carried out by the one-shot addition of ethynylmagnesium bromide (15.0 mL, 7.50 mmol, 10 equiv). The mixture was stirred for 3 min and then poured into cold MeOH for quenching. The precipitate was redissolved in THF, and the insoluble solid was removed by filter paper. The filtrate was recovered from successive precipitations in cold methanol, and the precipitate was dried under reduced pressure to afford alkyne-P3HT as a purple solid (2.11 g, 68%). 1H NMR (400 MHz, CDCl3): δ (ppm) 7.06−6.80 (m, aromatic), 3.53 (s, −CC− H), 2.90−2.48 (m, α-CH2), 1.71 (m, β-CH2), 1.54−1.15 (m, −CH2− CH2−CH2−CH3), 0.92 (br, −CH3). 13C NMR (100 MHz, CDCl3): δ (ppm) 140.1, 133.9, 130.7, 128.8, 31.9, 30.7, 29.6, 29.4, 22.8, 14.3. SEC: Mw = 5330 g/mol, Mn = 3490 g/mol, Mw/Mn = 1.35. FT-IR (CaF2): 3312 cm−1 (H−CC) and 2096 cm−1 (CC). Synthesis of Azido-Functionalized Maltoheptaose. Azidofunctionalized maltoheptaose (Mal7-N3) was synthesized by using a direct anomeric azidation method according to the previous reported procedure37 with minor modification. DMC (7.32 g, 43.3 mmol, 25 equiv) and DIPEA (16.8 g, 130 mmol, 75 equiv) were added to a solution of maltoheptaose (2.00 g, 1.74 mmol, 1 equiv) and NaN3 (28.1 g, 433 mmol, 250 equiv) in water (200 mL). The mixture was stirred at room temperature until the TLC (eluent: acetonitrile/water = 7/3) showed complete disappearance of the starting material. The mixture was first dialyzed against water using a dialysis membrane (Spectra/Por, Biotech CE tubing, MWCO: 100−500 D) and then freeze-dried to afford Mal7-N3 as a white solid (1.64 g, 80%). 1H NMR (400 MHz, D2O): δ (ppm) 5.48−5.41 (m, 6H), 4.70 (d, 1H), 4.07− 3.59 (m, 40H), 3.46 (t, 1H) and 3.34 (t, 1H). 13C NMR (100 MHz, D2O): δ (ppm) 100.1, 99.9, 90.3, 77.1, 77.0, 76.8, 76.6, 73.7, 73.2, 73.1, 72.1, 71.9, 71.8, 71.5, 69.7, 60.8, and 60.8. MALDI-TOF MS (m/ 3367
DOI: 10.1021/acs.macromol.7b00118 Macromolecules 2017, 50, 3365−3376
Article
Macromolecules
Figure 1. (a) MALDI-TOF MS spectrum of the obtained P3HT and (b) magnified part in the range of 21 repeating units. band at 2922 cm−1 corresponding to the CH2 asymmetric stretching vibration of the hexyl side chain, vas(CH2).
3.87 (m, sugar), 2.90−2.48 (m, α-CH2), 2.20−1.98 (m, CH3 acetyl group), 1.71 (m, β-CH2), 1.54−1.15 (m, −CH2−CH2−CH2−CH3), 0.92 (br, −CH3). SEC: Mw = 12 200 g/mol, Mn = 9480 g/mol, Mw/Mn = 1.29. FT-IR (CaF2): 1750 cm−1 (CO, esters). Deacetylation: Synthesis of Poly(3-hexylthiophene)-b-Maltoheptaose (P3HT-b-Mal7). In a round-bottom flask, P3HT-bAcMal7 (0.95 g, 1 equiv) was dissolved in THF (188 mL) and stirred at room temperature. A solution of 1 M sodium methoxide in methanol (7.4 mL, 2 equiv with respect to acetyl groups) was dissolved in THF (33 mL) and added dropwise to the stirred solution of the copolymer. The mixture was stirred overnight at room temperature until the FT-IR spectrum showed complete disappearance of the signal due to the acetyl groups (1750 cm−1). Then, the mixture was evaporated, and the residue was dissolved in THF. The product was recovered by successive precipitations in a mixture of methanol and water (9:1). The precipitate was filtrated and dried under reduced pressure to afford the P3HT-b-Mal7 block copolymer as a purple solid (0.72 g, 90%). SEC: Mw = 9530 g/mol, Mn = 6850 g/mol, Mw/Mn = 1.39. FT-IR (CaF2): 3660−3110 cm−1 (OH, sugars). Ultrathin Section Preparation for TEM. An about 10 μm thick film was prepered by drop-casting from a concentrated THF solution of P3HT-b-AcMal7 onto a PTFE substrate. After complete evaporation of the solvent, the thick film obtained was annealed at 220 °C for 10 min under vacuum to induce the microphase separtion. The thick film was then soaked in absolute ethanol overnight to dehydrate and then embedded in an epoxy resin mixture (Embed-812). The hardened resin block was set on an ultramicrotome with the film plane being orthogonal to the sectioning direction. Ultrathin sections with a thickness of less than 100 nm were obtained using a 35° diamond knife with a clearance angle of 6° and mounted onto carbon-coated copper grid. Stepwise Heating and Cooling FT-IR Measurements. The sample for an in situ FT-IR study on the structural evolution was prepared by drop-casting of the polymers onto a ZnSe window and set on a Pike Technologies heated solid transmission accessory. The sample was heated from room temperature to 210 °C and held at the target temperature for 3 min. The spectra were then recorded at a 2 cm−1 resolution with 4 scans. All processes were performed under a nitrogen flow. The obtained spectra were normalized to the absorption
■
RESULTS AND DISCUSSION Synthesis and Structural Characterization of P3HT-bCarbohydrate Diblock Copolymers. Our synthetic path toward diblock copolymers composed of P3HT and oligosaccharides is depicted in Scheme 1. First, alkynefunctionalized P3HT (alkyne-P3HT) was prepared via modified GRIM polymerization followed by the addition of 1-hexene before the end-capping reaction of the active chain end (Scheme 1a). According to Pickel et al.,36 during the endcapping process, the additives containing unsaturated groups such as allyl, vinyl, or alkynyl groups could prevent further oxidative addition of Ni0 to the α-chain end, which causes undesired end-functionalization, including bis-end-functional products, by forming a stable π-complex between Ni0 and the unsaturated group. Thus, in this study, we used 1-hexene as an unsaturated additive to increase the yield of monofunctional P3HT by controlling the end-capping process. The molar ratio of the monomer [M] and Ni catalyst [Ni] was fixed as [M]/ [Ni] = 25/1 to provide P3HT with a number-average molecular weight Mn of ca. 4000 g/mol. This target molecular weight was estimated based on our previous studies on the selfassembly, in which we revealed that oligosaccharide-containing high χ block copolymer systems self-organize into sub-10 nm scale structures, such as spheres, cylinders, and lamellae, when using synthetic blocks with molecular weights in the 2000− 5000 g/mol range.29−32,34 The obtained polymer was purified by precipitation in cold methanol and characterized by MALDITOF MS, 1H NMR, and SEC. The end-group composition of P3HT was determined by MALDI-TOF MS analysis (Figure 1). As expected, the major products were identified as monofunctional P3HT terminated with an alkyne group at the ω-end and bromine at the α-end 3368
DOI: 10.1021/acs.macromol.7b00118 Macromolecules 2017, 50, 3365−3376
Article
Macromolecules
Figure 2. 1H NMR spectrum of (a) alkyne-P3HT, (b) AcMal7-N3, (c) P3HT-b-AcMal7, and (d) P3HT-b-Mal7 in CDCl3.
Table 1. Characterization Details of the Synthesized P3HT-b-Oligosaccharide Block Copolymers sample
Mn,SEC,P3HTa [noncorrected] (g/mol)
Mn,SEC,P3HTb [corrected] (g/mol)
Mp,MALDI,P3HTc (g/mol)
Msugard (g/mol)
Mw/Mn,totala
ϕP3HTe
alkyne-P3HT P3HT-b-Mal7 P3HT-b-AcMal7
5330 5330 5330
3490 3490 3490
3596 3596 3596
1136 2060
1.35 1.39 1.29
1 0.79 0.64
a
Number-average molecular weight determined by SEC in THF based on PS standards. bCorrected SEC value using the Mark−Houwink−Sakurada constants: aP3HT and KP3HT are 0.96 and 2.28 × 10−3 cm3/g, respectively. cMolecular weight at the peak maximum determined by MALDI-TOF MS. d Molecular weight of sugars. eVolume fraction of P3HT in copolymers calculated by using density values 1.16 g/cm3 for P3HT, 1.36 g/cm3 for Mal7, and 1.20 g/cm3 for AcMal7.
5330 g/mol and 1.35, respectively, by SEC (Table 1 and Figure 3) based on polystyrene standards. With this analysis, the molecular weight of P3HT is overestimated compared to the real value because of the stiffness of the π-conjugated polymer backbones and the large difference in the hydrodynamic volume between P3HT and polystyrene standards in the THF eluent. Therefore, the SEC data were corrected by using the Mark−Houwink−Sakurada constants, KP3HT = 2.28 × 10−3 cm3/g and aP3HT = 0.96 in THF at 25 °C, according to a previous report.38 The corrected Mn value was calculated to be 3490 g/mol, which is in good agreement with the maximum value (3596 g/mol) of the weight distribution in the MALTDITOF-MS spectrum. In parallel, reducing-end azido-functionalized maltoheptaose (Mal7-N3) was prepared by direct anomeric azidation according to Shoda’s methodology.37 The azido functionality on the maltoheptaose was clearly detected by the IR absorption at 2100 cm−1 as shown in Figure 4b. Attempts to couple the obtained P3HT and maltoheptaose molecular bricks to prepare the block copolymer were then
(88%). The other small amounts of byproducts were assigned to be bis-end-functionalized P3HT having alkyne/alkyne groups (7%), monofunctionalized P3HT having alkyne/H groups (3%), and nonfunctionalized P3HT having H/Br (2%) end groups. Hence, the in situ end-functionalization of P3HT via the GRIM method in the presence of 1-hexene led to monofunctionalized alkyne-P3HT with 91% purity. Further structural characterization of alkyne-P3HT was carried out by 1H NMR spectroscopy as shown Figure 2a. The proton resonance signal at 3.53 ppm (signal x in Figure 2a) was assigned to the alkyne proton (−CCH), indicating successful introduction of alkyne functional group to the polymer chain. The HT-HT regioregularity was estimated to be 95% by comparing integral ratio of the signals corresponding to α-methylene protons, that is, the signal at 2.80 ppm due to HT linkage and the signals around 2.48−2.75 ppm due to head-tohead (HH) linkages (signal f in Figure 2a). The number-average molecular weight (Mn,SEC) and the dispersity (Mw/Mn) of alkyne-P3HT were determined to be 3369
DOI: 10.1021/acs.macromol.7b00118 Macromolecules 2017, 50, 3365−3376
Article
Macromolecules
P3HT and AcMal7-N3 was performed in THF at room temperature (Scheme 1c). The resulting product was characterized by SEC, 1H NMR, and IR analyses. The SEC trace of the product (red line in Figure 3) indicated a clear shift toward the higher molecular weight region compared to that of alkyne-P3HT (black line in Figure 3). It should be noted that the shoulder peaks observed in the SEC traces of the product and alkyne-P3HT should be dimers of P3HTs as it is generally observed in the one-pot synthesis of alkyne-functionalized P3HTs by GRIM method.22,39,40 The 1H NMR spectrum of the product showed characteristic signals assignable to the protons of P3HT and AcMal7 as well as the signal corresponding to the proton on the triazole ring at 7.77 ppm (signal y in Figure 2c). The complete disappearance of the signal at 3.53 ppm corresponding to an alkyne proton (−CCH) indicated that all functional alkyne groups were consumed during the reaction. In addition, diffusion-ordered spectroscopy (DOSY) clearly supported the covalent attachment of the AcMal7 to the P3HT by showing only a single diffusion coefficient with D = 3.8 × 10−10 m2/s for all proton signals related to the P3HT and AcMal7 segments as well as triazole ring (Figure S1). In the IR spectrum of resulting product (Figure 4d), the peak at 2120 cm−1 due to the azido group of AcMal7-N3 (Figure 4c) completely disappeared whereas the peak at 1750 cm−1 corresponding to the CO stretching vibration of the acetyl group in AcMal7 were clearly observed. From these results, the obtained product was identified as the targeted block copolymer P3HT-b-AcMal7. Finally, P3HT-b-AcMal7 was deacetylated with sodium methoxide in dry THF to give the block copolymer P3HT-b-Mal7. Figures 4d and 4e show the IR spectra of P3HT-b-AcMal7 and its deacetylated product, respectively. The disappearance of the sharp absorption at 1750 cm−1 of acetyl groups and the appearance of the broad band at 3100−3600 cm−1 due to hydroxyl groups clearly indicate the deacetylation of P3HT-b-AcMal7. Figure 2d shows 1 H NMR spectra of the product in CDCl3. The characteristic signals assignable to the protons of P3HT were observed whereas the signals due to the protons of Mal7 and the triazole ring were not observed. This is probably due to reverse micellar aggregation of the block copolymer in CDCl3 owing to its amphiphilic feature where the core of the aggregates consists of hydrophilic Mal7 segments and the shell consists of hydrophobic P3HT segments. This aggregation strongly restricts the intramolecular mobility of Mal7 and the triazole ring segments in the core, and thus, the detection of the resonance from the protons of Mal7 and the triazole ring was prevented. To verify this hypothesis and confirm the product structure, solid-state 13 C cross-polarization magic angle spinning (13C CP/MAS) NMR analysis of the product was performed. The spectrum is shown in Figure 5d together with the solution state 13C NMR spectra. Similarly as the 1H NMR spectra of the product in CDCl3 (Figure 2d), the solution state 13C NMR spectrum (Figure 5c) displays signals assignable to the 13C resonance of P3HT segments whereas no signal due to Mal7 and triazole segments was observed. On the other hand, the 13C CPMAS measurement allowed detecting the 13C resonance of both P3HT (signals a−g) and Mal7 segments (55−110 ppm) as shown in Figure 5d. Therefore, the obtained product was identified as P3HT-b-Mal7. The characteristics of the polymers and the volume fraction of P3HT (f P3HT) for obtained block copolymers are summarized in Table 1. Thermal Properties. The thermal stability of block copolymers, P3HT-b-AcMal7 and P3HT-b-Mal7, and P3HT
Figure 3. SEC traces of alkyne-P3HT (black line) and P3HT-bAcMal7 (red line).
Figure 4. FT-IR spectra of (a) alkyne-P3HT, (b) Mal7-N3, (c) AcMal7-N3, (d) P3HT-b-AcMal7, and (e) P3HT-b-Mal7.
carried out. First, a direct coupling reaction of alkyne-P3HT and Mal7-N3 was performed by copper-catalyzed azide−alkyne cycloaddition, the so-called CuAAC “click” reaction, using CuBr and PMDETA as a catalytic system. A major concern for this reaction was the choice of solvent. Since the solubility of the two segments, hydrophobic P3HT and the hydrophilic maltoheptaose moiety, are very different, we could not find a common solvent for both moieties, even by mixing a good solvent for P3HT with a good solvent for maltoheptaose. As a result, the coupling reaction did not proceed in this heterogeneous medium. For this reason, we had decided to acetylate the hydroxyl groups of Mal7-N3 with the aim to find a common good solvent to perform the CuAAC click reaction in a homogeneous state. The acetylation of the hydroxyl group of Mal7-N3 was performed with an acetic anhydride/pyridine mixture in the presence of DMAP as a catalyst to give peracetylated Mal7-N3 (AcMal7-N3). The coupling of alkyne3370
DOI: 10.1021/acs.macromol.7b00118 Macromolecules 2017, 50, 3365−3376
Article
Macromolecules
Figure 5. 13C NMR spectra of (a) alkyne-P3HT in CDCl3, (b) Mal7-N3 in D2O, and (c) P3HT-b-Mal7 in CDCl3, and (d) 13C CP/MAS NMR spectrum of P3HT-b-Mal7.
homopolymer was investigated by TGA as shown in Figure 6. The results are summarized in Table 2. The TGA curve of
P3HT-b-AcMal7 (red line) showed a two-stage weight loss behavior, corresponding to the thermal decomposition of the AcMal7 block (slope i) and the thermal decomposition of the P3HT block (slope ii). The TGA curve of P3HT-b-Mal7 (blue line) shows multiple weight loss stages, corresponding to the loss of adsorbed water in the hydroxyl groups of Mal7 (slope a), the thermal decomposition of the Mal7 block (slope b), and the thermal decomposition of the P3HT block (slope c). The thermal decomposition temperatures for a 5% weight loss (Td,5%) of P3HT, P3HT-b-AcMal7, and P3HT-b-Mal7 were 422, 291, and 187 °C, respectively. Both block copolymers had lower Td,5% values than that of the P3HT homopolymer, probably due to the low thermal stability of oligosaccharidic blocks (see Supporting Information, Figure S2). To investigate thermal properties and crystallinity of P3HT, DSC measurements were performed on the P3HT homopol-
Figure 6. TGA curves of alkyne-P3HT (black line), P3HT-b-AcMal7 (red line), and P3HT-b-Mal7 (blue line).
Table 2. Thermal Properties of P3HT Homopolymer, P3HT-b-AcMal7, and P3HT-b-Mal7 TGA
DSC ΔHm (J/g)
temp range of inclinationa
sample alkyne-P3HT P3HT-b-AcMal7 P3HT-b-Mal7
(i) (ii) (a) (b) (c)
400−530 250−380 400−500 95−160 200−310 390−490
factor decomposition decomposition decomposition loss of water decomposition decomposition
of P3HT of AcMal7 of P3HT
Td,5% (°C)
Tm (°C)
Tc (°C)
ΔHm,totalb
ΔHm,P3HTc
422 291
178 189
145 166
7.8 6.6
7.8 10.4
187
189
165
7.3
9.6
of Mal7 of P3HT
a
Temperature range of inclination of each slope shown in Figure 6. bThe experimental values of total melting enthalpy. cCalculated values based on the P3HT weight fraction. 3371
DOI: 10.1021/acs.macromol.7b00118 Macromolecules 2017, 50, 3365−3376
Article
Macromolecules
was not clearly observed while the Tg of Mal7 was reported to be around 150−160 °C.41 Self-Assembly Behaviors and Morphological Study. The self-assembly behavior of the P3HT-b-AcMal7 and P3HTb-Mal7 was characterized by AFM, TEM, and SAXS and was compared with that of the P3HT homopolymer. First, we investigated the morphology of the as-cast film samples. The AFM phase images of the as-cast films of P3HT-b-Mal7 and P3HT homopolymer showed a fibril-like structure with micrometer-scale lengths and nanometer-scale widths (Figures 8a and 8c). Such a fibril-like structure is generally observed for other P3HT-based block copolymer thin films.16,42,43 Indeed, it is well-known that P3HT forms so-called “nanofibril” structure in the bulk and thin film states because of the crystallization based-on strong π−π interactions between π-conjugated thiophene backbones. The SAXS profiles of the bulk P3HTb-Mal7 and the P3HT homopolymer exhibited a broad primary scattering peak (q*) corresponding roughly to the firstneighbor distance of randomly oriented nanofibrils (black lines in Figures 8g and 8i). Therefore, it is assumed that the self-organization of the P3HT-b-Mal7 during film casting was dictated by the crystallization of the P3HT block, which is kinetically fast enough to form nanofibril structures. On the other hand, the as-cast film of P3HT-b-AcMal7 showed a substantially different morphology. The AFM phase image of the as-cast P3HT-b-AcMal7 film showed globular aggregates over the whole film surface while the nanofibril structure of the P3HT block was not observed (Figure 8b). This difference in morphology between the as-cast films of P3HT-b-Mal7 and P3HT-b-AcMal7 can be explained by the lower volume fractions of the P3HT block in P3HT-b-AcMal7 (f P3HT = 0.64) than that in the P3HT-b-Mal7 (f P3HT = 0.79). Because of the relatively lower influence of the P3HT crystallinity on the self-organization of the P3HT-b-AcMal7, the formation of P3HT nanofibril structures was hindered by the higher contribution of the microphase separation behavior of the block copolymer. In addition, steric hindrance due to the slightly bulkier AcMal7 group than Mal7 group may also contribute to inhibiting the formation of P3HT nanofibril structures. These as-cast film samples were then thermally annealed at 220 °C for 10 min under vacuum. This temperature was chosen based on the melting temperature of the P3HT-b-AcMal7 and P3HT-b-Mal7 (Tm,P3HT‑b‑AcMal7 = Tm,P3HT‑b‑Mal7 = 189 °C) to melt the P3HT segment and trigger the microphase separation of the block copolymers. The thermally annealed P3HT homopolymer film showed short-range nanofibril-like structures due to recrystallization of P3HT from the molten state (Figure 8f). Similar structure due to the recrystallized P3HT nanofibrils was observed in the thermal annealed P3HT-b-Mal7 block copolymer film as shown in Figure 8d. The SAXS profile of the thermal annealed P3HT-b-Mal7 bulk sample showed only a broad primary peak q* arising from nanofibril-like P3HT crystalline structures (red line in Figure 8g). These results indicated that microphase separation of P3HT-b-Mal7 block copolymer was not induced by thermal annealing. On the contrary, P3HT-b-AcMal7 showed clear microphase separation after thermal annealing. The AFM image of thermally annealed P3HT-b-AcMal7 thin films showed microphase-separated periodic stripe pattern (Figure 8e). The SAXS profile of thermally annealed bulk P3HT-b-AcMal7 clearly supports this microphase separation (red line in Figure 8h); i.e., sharp peaks in the SAXS profile located at q*, 2q*, and 3q* positions
ymer and the diblock copolymers. The DSC thermograms recorded during the first cooling and the second heating cycles with the heating/cooling rates of 10 °C/min are shown in Figure 7. The detailed data are summarized in Table 2. For the
Figure 7. DSC thermograms of alkyne-P3HT (black line), P3HT-bAcMal7 (red line), and P3HT-b-Mal7 (blue line) obtained during first cooling run and second heating run with a heating and cooling rate of 10 °C/min.
P3HT homopolymer (black line), an endothermic peak at 178 °C related to the P3HT melting temperature (Tm) was observed in the heating cycle. The DSC curves for P3HT-bAcMal7 (red line) and P3HT-b-Mal7 (blue line) showed similar endothermic peaks at 189 °C that are higher than the Tm of the P3HT homopolymer. In the cooling cycle, a similar tendency was observed in the crystallization temperature (Tc); i.e., the Tc value of the P3HT homopolymer (145 °C) was lower than those of P3HT-b-AcMal7 (166 °C) and P3HT-b-Mal7 (165 °C). These higher Tm and Tc values of the block copolymers than those of the P3HT homopolymer suggest that the crystal state of the P3HT segment confined in the block copolymers is more ordered or more stable than that of the P3HT homopolymer. The melting enthalpy obtained by the integration of melting peak area normalized by the total sample weight (ΔHm,total) for the P3HT homopolymer, P3HT-bAcMal7, and P3HT-b-Mal7 were 7.8, 6.6, and 7.3 J/g, respectively. However, since the melting enthalpy is directly associated with the required heat energy for breaking down the crystal structure, the weight fraction of P3HT in the block copolymer should be taken into account in order to evaluate the crystallinity of P3HT in an adequate manner. Hence, the melting enthalpies were corrected on the basis of the weight fraction of P3HT to give the amount of heat per P3HT unit weight. The corrected melting enthalpy (ΔHm,P3HT) of P3HTb-AcMal7 and P3HT-b-Mal7 was calculated to be 10.4 and 9.6 J/g, respectively. These values were higher than that of P3HT homopolymer (7.8 J/g), showing a good correlation with the higher Tm and Tc of the block copolymers than those of P3HT homopolymer. This result also suggests that the crystal structure of the P3HT segment in the block copolymers was more stable than that of the P3HT homopolymer. Meanwhile, the glass transition temperature (Tg) of P3HT-b-AcMal7 was not clearly observed although the Tg of AcMal7 was determined to be around 115−120 °C in the DSC thermogram of free AcMal7 (Figure S3), indicating that the AcMal7 segment covalently linked with P3HT chain exhibits different thermal behavior from the free AcMal7. Similarly, Tg of P3HT-b-Mal7 3372
DOI: 10.1021/acs.macromol.7b00118 Macromolecules 2017, 50, 3365−3376
Article
Macromolecules
Figure 8. AFM phase images of as-cast films for (a) P3HT-b-Mal7, (b) P3HT-b-AcMal7, and (c) alkyne-P3HT and the films thermally annealed at 220 °C for 10 min for (d) P3HT-b-Mal7, (e) P3HT-b-AcMal7, and (f) alkyne-P3HT. Scale bars indicate 200 nm. SAXS profiles of as-prepared samples in bulk (black line) and the bulk samples thermally annealed at 220 °C for 10 min (red line) for (g) P3HT-b-Mal7, (h) P3HT-b-AcMal7, and (i) alkyne-P3HT.
Figure 9. TEM images of (a) as-cast and (b) thermally annealed thin films of P3HT-b-AcMal7 and (c) cross section of thermally annealed thick film of P3HT-b-AcMal (annealing condition: 220 °C for 10 min under vacuum). Insets are fast Fourier transform (FFT) images.
confirms the lamellar structure. Whereas the as-cast P3HT-bAcMal7 thin film showed a heterogeneous morphology consisting of dot- and nanofibril-like structures in Figure 9a, the thermally annealed P3HT-b-AcMal7 thin film displayed a uniform lamellar structure. The clear lamellar structure observed in the cross-sectional TEM image of the thermally annealed P3HT-b-AcMal7 thick film (Figure 9c) indicates
indicate a microphase-separated lamellar structure with interdomain spacing (d) of 15.4 nm (half-pitch
