Sonoka
Yamamoto
a,
Ryutaro
Yamashita
a,
Chihiro
Kubota
a,
Kentaro
Okano
a,
Masatoshi
Kitamura
a,
Masahiro
Funahashi
b,
Syu-Cheng
Ye
c,
Yung-Tin
Pan
c,
Masaki
Horie
c,
Takuji
Shintani
d,
Hironori
Murata
a,
Hideto
Matsuyama
ad and
Atsunori
Mori
*ad
aGraduate School of Engineering, Kobe University, 1-1 Rokkodai, Nada, Kobe 657-8501, Japan. E-mail: amori@kobe-u.ac.jp
bDepartment of Advanced Materials Science, Kagawa University, 2217-20 Hayashi-cho, Takamatsu, Kagawa 761-0396, Japan
cDepartment of Chemical Engineering, National Tsing Hua University, 101, Sec. 2, Kuang-Fu Road, Hsinchu 30013, Taiwan
dResearch Center for Membrane and Film Technology, Kobe University, 1-1 Rokkodai, Nada 657-8501, Japan
First published on 23rd January 2023
A thiophene–thiophene block copolymer composed of hydrophilic and hydrophobic side chain functionalities was designed and synthesized. The deprotonative metalation nickel-catalyzed polymerization protocol successfully afforded the block copolymer, in which the side chains are derived from alkyl and benzenesulfonic acid ester groups. The benzene sulfonate moiety of the block copolymer in the film state was shown to be transformed into hydrophilic sulfonic acid upon thermal treatment at ca. 200 °C without the addition of an external additive. Thus, the formed block copolymer thin film exhibited cylindrical microphase separation and the hydrophilic domain was revealed to penetrate the film perpendicular to the substrate. The measurement of electric conductivity suggested that the block copolymer thin film was conductive when placed parallel to the substrate, while the film was insulative when placed perpendicular to the substrate. Electrochemical analyses revealed that lithium ions transfer through the cylindrical domain composed of a benzene sulfonic acid side chain, which is perpendicular to the substrate. These results represent dual and orthogonal conductivities of electrons and ions in the block copolymer thin film.
We have previously reported the deprotonative synthesis of polythiophene that involves the reaction of a 2-halo-3-substituted thiophene with magnesium amide to afford the corresponding organometallic monomer,7–11 which may then be subjected to cross-coupling polymerization followed by the addition of a nickel catalyst to afford a regioregular HT-type polythiophene with a well-controlled molecular weight as shown in Fig. 1(a).
![]() | ||
| Fig. 1 Synthesis of head-to-tail (HT) type polythiophene and side chain functionality of polythiophenes. | ||
Furthermore, the rational design and introduction of side-chain moieties can be employed to confer additional functionality to polythiophenes.12–14 For instance, we have previously reported that (i) the introduction of an oligosiloxane side-chain moiety improves the solubility of polythiophene in organic solvents, particularly hydrocarbons such as hexanes,15 and (ii) polythiophenes bearing benzenesulfonic acid ester groups undergo thermally induced doping of the polymer main chain upon thermal decomposition of neopentyl benzene sulfonate to the corresponding sulfonic acid. This transformation at the side chain allows doping of the polythiophene main chain while simultaneously conferring water-solubility to the polymer (Fig. 1(b)).
Our extensive research on the structural design of polythiophenes has focused on the introduction of composite functionality, which can be achieved by the synthesis of polythiophene copolymers comprising monomers with different functional groups on their side chains.16,17 For instance, we have recently reported that introducing a small amount of a thiophene comonomer with thermally induced-doping functionality to poly(3-alkylthiophene) imparts self-doping ability to the resultant random copolymer.18
Our further concerns are related to the synthesis of copolymers composed of different polythiophene moieties, which represents a breakthrough in terms of functionality design in polythiophene synthesis.19 Because the formation of a block copolymer induces microphase separation in its corresponding polymer film, such films have application potential in areas such as selective gas permeation, ion transfer, etc.20 It has been shown that a block copolymer with a certain ratio of each copolymer exhibits a microdomain structure in the form of a cylinder or a lamella. Therefore, such a microphase separation would lead to the formation of a penetrating substructure in the thin film state on a substrate.21–23 Specifically, a copolymer containing some polythiophene units with benzenesulfonic acid side-chain functionality could potentially form micro-sized domains capable of ion conduction24,25 along with the electron-conductive characteristics of the polythiophene itself. The ion conductivity imparted by discrete sulfonate microdomains would reflect in the thin film formed on the substrate.26,27 However, such characteristics are rarely found in other electronic materials.
Herein, we report the block copolymerization of different types of thiophene monomers with either alkyl or benzenesulfonate side chains. A thin film formed from this copolymer was demonstrated to exhibit cylindrical microphase separation that penetrates throughout the film providing ion conductivity as well as the electrically conductive characteristics of regioregular poly(3-substituted thiophene). Both conductivities of the film are demonstrated in orthogonal directions perpendicular and parallel to the substrate, respectively (Fig. 1(c)).
:
0.9 based on the integral values of the methylene signal for the hexylthiophene moiety derived from 1 (2H)n and the neopentyl methylene signal from 2 (2H)m.
![]() | ||
| Scheme 1 Synthesis of a head-to-tail (HT) type polythiophene and side chain functionality of polythiophenes. | ||
Block copolymerization of monomer precursors 1 and 2 at different ratios was also performed. Table 1 shows the details of these products. The reactions provided the corresponding products in good to excellent yields with reasonable average molecular weights (Mn) and relatively narrow molecular weight distributions (Mw/Mn). The ratios of the components in block copolymers 5b–g were estimated from their 1H NMR spectra and are presented as n
:
m. This ratio was found to correspond reasonably well with the feed ratio of the monomer precursors 1 and 2. These results show that polymerization of metalated 2 proceeds at the living end of the polymerized alkylthiophene 1 smoothly and revealed that incorporation of both metalated monomers of 1 and 2 took place at room temperature within 24 h.
| 5 | 1/2b | Yield (%) | M n (Mw/Mn) |
m : nd |
|---|---|---|---|---|
a The first polymerization was carried out by the reaction of chlorothiophene 1 (0.3 mmol), TMPMgCl·LiCl (3, 0.33 mmol) and nickel catalyst 4 in THF. Polymerization was carried out at room temperature for 3 h. The monomer derived from 2 (0.3 mmol) was added to the reaction mixture for the first polymerization to initiate the second propagation and the resulting thiophene–thiophene block copolymer 5 was obtained as a dark purple solid.
b The feed ratio of 1 and 2 (mol mol−1). In parenthesis, the mass ratio of the benzenesulfonate moiety is shown.
c
M
n and Mw/Mn were determined by SEC analysis based on polystyrene standards.
d The ratio of m : n was determined by 1H NMR analysis.
|
||||
| 5a | 0.5 : 0.5 |
74 | 24 300 (1.47) |
0.53 : 0.47 (62%) |
| 5b | 0.8 : 0.2 |
71 | 16 000 (1.73) |
0.83 : 0.16 (28%) |
| 5c | 0.7 : 0.3 |
87 | 13 000 (1.75) |
0.76 : 0.24 (37%) |
| 5d | 0.6 : 0.4 |
79 | 17 700 (1.62) |
0.67 : 0.33(48%) |
| 5e | 0.4 : 0.6 |
83 | 24 000 (1.36) |
0.43 : 0.57(71%) |
| 5f | 0.3 : 0.7 |
71 | 21 800 (1.32) |
0.29 : 0.71 (82%) |
| 5g | 0.2 : 0.80 |
48 | 14 000 (1.32) |
0.21 : 0.79 (87%) |
The obtained copolymer 5a was found to be soluble in organic solvents such as chloroform and THF. However, it was found to be insoluble in both chloroform and water after being heated at ca. 200 °C for 10 min. This result indicates the formation of an amphiphilic block copolymer 6a bearing hydrophobic alkyl groups and hydrophilic sulfonic acid (–SO3H) pendant groups. It is very different from the result for the corresponding benzenesulfonate-bearing homopolymer, which is soluble in water after heating. We also prepared and characterized two related copolymers, both with a similar n
:
m ratio to that of 5a. These were the random 5a′ prepared by the copolymerization of 1 and 218 and the alternating copolymer 5a′′, prepared by the polymerization of a bithiophene dimer bearing both the substituents present on 1 and 2.16
The TG-DTA trace for 5a (m
:
n = ca. 0.5
:
0.5) shows a weight loss of ca. 15% at 200 °C. (Fig. 2) This result was found to agree reasonably with the calculated 1H NMR result (m
:
n = 0.53
:
0.47) and the molecular formulas of 2 –(C10H14S)m–(C15H16S2O3)n–, indicating a 14% loss with the elimination of C5H10 from the side-chain neopentyl sulfonate. Based on the results from TG-DTA, spectroscopic analyses of block copolymer 5a in the film state were performed on a quartz substrate. Measurements after thermal treatment were performed by heating the substrate at 200 °C for 20 min. As shown in Fig. 3(a), the UV-vis absorption spectrum of 5a in the film state shows a λmax value of ca. 560 nm, while an emergent absorption at >700 nm along with a decrease in absorbance at λmax is observed in 6a upon thermal treatment, suggesting that thermally induced self-doping of the polythiophene main chain via the formation of SO3H on the side chains occurs.18 We next performed XRD with the thin film of block copolymer 6a after thermal treatment. The measurement of the out-of-plane profile revealed a characteristic peak at 2θ = 5.4°, which was assigned as the (100) reflection suggesting that the layer distance derived from the polythiophene side chain (16.4 Å). In contrast, a peak at 2θ = 15–20° that was characterized as the layer distance between conjugated polymers with π–π stacking (5.7 Å) was broad and relatively smaller toward the (100) reflection (Fig. 3(b)). These results in the out-of-plane measurement strongly suggest edge-on preference of the polythiophene film on the substrate, whereas attempted in-plane measurement was unsuccessful.
![]() | ||
| Fig. 3 (a) UV-vis absorption spectra of block copolymer 5a before and after thermal treatment. (b) X-ray diffraction pattern (out-of-plane) of 5a after thermal treatment on a glass substrate. | ||
These results suggest that microphase separation appears in 6a after thermal treatment at 200 °C for 20 min owing to the presence of the two different side chains in a specific arrangement. Although the dotted pattern is not observed for 5a before heating, this does not necessarily show that phase separation proceeds upon heating of the as-deposited film, because it becomes much more apparent in AFM observation after the decrease in volume caused by the elimination of the neopentyl group (as C5H10). Considering that the related dotted pattern was not observed in the related random (6a′) or alternating (6a′′) copolymer (Fig. 4(c) and (d), respectively), these results also support that the observed dots are the result of microphase separation and the formation of a gas bubble by the evolving C5H10 may also be ruled out (see also Fig. S1 and S2, ESI†).
Similar to the results for AFM, the TEM image of 6a (Fig. 4(e)) was measured at 200 kV and shows a dotted pattern, albeit weak contrast because of the structural similarity of the main chain in each domain, as that observed by AFM,23,30–32 confirming that the thin film of 5a features cylindrical phase separation that penetrates the film perpendicular to the substrate. Measurement of small-angle X-ray scattering (SAXS) of 6a also supported the formation of microphase separation. As shown in Fig. 4(f), the SAXS curve showed a primary peak along with a hump. The interdomain distance D determined from the peak position was 28.46 nm, which agreed with that estimated by AFM and TEM analyses. It should be noted that such a peak was only observed in the thermally treated block copolymer, where no peak was found in 5a before thermal treatment as well as the corresponding random copolymer 5a′ (see also the ESI†).
We also subjected 5b–5g to AMF analysis, as shown in Fig. 4(g). Similar dotted patterns are observed for 6b, 6c, 6d, and 6e but not for 6f and 6g. This suggests that phase separation occurs after heating when the alkylthiophene moiety content is high, while a higher benzenesulfonate content appears to disrupt the phase separation.
The conductivity in the perpendicular direction of the substrate was measured using a gold electrode at a lower voltage (0–10 V) as shown in Fig. 5(b), with a gap length of ca. 80 μm and a film thickness of ca. 200 nm, where the film was cast on the Au electrode followed by the additional deposition of the Au electrode on the polythiophene film. The related gold electrode for the measurement in the parallel direction under conditions similar to those shown in Fig. 5(b) (0–10 V) was also applied as shown in Fig. 5(c), where widths of the Au electrodes were set as 1.4 mm and 100 μm, respectively, and the film thickness was 200 nm.
The conductivities of the block copolymers at an applied voltage of 100 V before and after thermal treatment are summarized in Table 2. It was found that the conductivity of 5a is 8.92 × 10−5 S cm−1 before thermal treatment and that of 6a is 9.37 × 10−3 S cm−1 (100 V) after thermal treatment. This suggests that the conductivity is improved ca. 100-fold by thermal treatment. A smaller improvement in conductivity is observed for copolymer 6b because of its lower benzenesulfonate content (<20%). Improved conductivity after thermal treatment is observed for 6g owing to its higher sulfonate content, albeit with little microphase separation. The conductivity after thermal treatment is observed in the parallel direction to the substrate. The results suggest that the polythiophene block copolymer 5 in the film state on the substrate is cast as edge-on orientation as depicted in Fig. 1(c) (right).
| 1/2b (mol mol−1) | Applied voltate,c V | Conductivity before thermal treatment, S cm−1 | Conductivity after thermal treatment, S cm−1 | |
|---|---|---|---|---|
| a Measurements were carried out using devices as depicted in Fig. 5(a)–(c). b The ratio derived from 1 and 2 in the block copolymer of 5. Based on the result of 1H NMR measurement. c In parenthesis, the measured device is shown in Fig. 5. | ||||
| 5a | 0.53/0.47 | 100 (a) | 8.92 × 10−5 | 9.37 × 10−3 |
| 5b | 0.83/0.16 | 100 (a) | 1.10 × 10−5 | 2.51 × 10−4 |
| 5g | 0.21/0.79 | 100 (a) | 8.12 × 10−2 | |
| 5a | 0.53/0.47 | 10 (b) | 6.06 × 10−10 | |
| 5a | 0.53/0.47 | 10 (c) | 1.75 × 10−4 | 1.58 × 10−3 |
It should be pointed out that the conductivity using the thin film shown in Fig. 5(b) at the applied voltage of 0–10 V is remarkably low even after thermal treatment showing insulative characteristics in the perpendicular direction. On the other hand, the related Au electrode (Fig. 5(c)) showed a much superior conductivity of 6a (1.58 × 10−3 S cm−1 at 10 V) in the parallel direction of the film similar to the case of the ITO electrode. These results support the edge-on orientation of the thiophene–thiophene block copolymer 5a and that the thermally induced self-doping of the polythiophene moiety bearing benzenesulfonic acid improved the conductivity in the poly(alkylthiophene) moiety.34,35
The Nyquist plot acquired for the thermally treated film of 6a at 0.6 V vs. the reversible hydrogen electrode (RHE) features an additional semicircle at the high-frequency region, as shown in Fig. 6(b) (red plot), while that of 6a′ does not (black plot). The plots for 6a and 6a′ can be fitted with an equivalent circuit composed of one resistor (Rs) in series with two resistor–capacitor (R1-CPE1 and R2-CPE2) units (Fig. 6(b), solid lines). The impedances and capacitances of the systems fabricated with 6a and 6a′ were estimated as summarized in Table 3, demonstrating the remarkably low R1 and R2 values for 6a compared with those for 6a′. The obtained values for resistance for 6a, both R1 and R2, were 2.06 × 102 W cm2 and 8.06 × 104 W cm2, respectively. These values were significantly lower than those obtained for 6a′, which were 3.48 × 104 W cm2 and 2.66 × 105 W cm2, respectively (Table 3).
| R 1 (Ω cm2) | CPEb (F cm−2) | R 2 (Ω cm2) | CPE2-Tc (F cm−2) | |
|---|---|---|---|---|
| a The measurement was performed as illustrated in Fig. 6(a). b Impedance and capacitance of copolymers before thermal treatment. c Impedance and capacitance of copolymers after thermal treatment. | ||||
| 5a | 2.06 × 102 | 9.67 × 10−5 | 8.06 × 104 | 7.38 × 10−5 |
| 5a′ | 3.48 × 104 | 9.04 × 10−5 | 2.66 × 105 | 2.03 × 1015 |
Cyclic voltammetry (CV) analysis of 5a and 6a (Fig. 6(c)) supports the EIS results. A cyclic voltammogram is only clearly observed for the thermally treated film of 6a, and the capacitance was calculated to be 0.431 mF cm−2 at a scan rate of 200 mV s−1, as compared with 0.08 mF cm−2 before thermal treatment. In addition, the excellent retention of capacitance compared with that at the slower scan rate of 20 mV s−1 (21% loss) reflects the superior ionic conductivity of thermally treated 6a. The cyclic voltammograms for 5a′ and 6a′ also show differences before and after thermal treatment, but the capacitance of 5a′ is much lower (Fig. 6(d)). Furthermore, even after thermal treatment, 6a′ loses 71% of its capacitance when the scan rate is increased from 20 to 200 mV s−1.
These results show that lithium-ion conductivity is observed for the thermally treated copolymers 6a and 6a′, while the as-cast films are much less ion conductive. Because the thermal treatment of the polythiophene film converts benzenesulfonate (–SO2OR) into the corresponding sulfonic acid (–SO3H), lithium ions can interact with the SO3− moiety, resulting in ion conductivity.
Compared with the corresponding values for 6a′, the lower resistance of 6a by EIS measurements as well as its higher capacitance and a smaller decrease upon increasing the CV scan rate (200 vs. 20 mV s−1) indicate that 6a does indeed feature channel domains that transport hydrated lithium ions, as depicted in Fig. 6(e). These results also correspond with the AFM and TEM results shown in Fig. 4, which support the formation of cylinder-like microphase separation regions penetrating the film perpendicular to the substrate.
Combined with the electric conductivity measurements shown in Fig. 5, the thin film of 6a exhibits dual orthogonal conductivities for electrons (parallel to the substrate) and lithium ions (perpendicular to the substrate). To the best of our knowledge, these unique characteristics have not been observed in any previous polythiophene materials.
000). High resolution mass spectra (HRMS) were recorded using a JEOL JMS-T100LP AccuTOF LC-Plus (ESI) with a JEOL MS-5414DART attachment. Elemental analyses were carried out at the Department of Instrumental Analysis & Cryogenics Division of Instrumental Analysis, Okayama University on a PerkinElmer 2400II Elemental analyzer supported by the Inter-University Network for Common Utilization of Research Equipments. IR spectra were recorded on a Bruker Alpha with an ATR attachment (Ge). UV-vis spectra (as a thin film) were recorded using a Shimadzu UV-3101PC. Thermal analyses were carried out with a RIGAKU Thermo plus EVO2 TG-DTA 8121. XRD analysis was carried out with a Rigaku SmartLab with the power of 40 kV, 30 mA, and the wavelength of 1.5418 Å (CuKα). The incidence angle α of the X-ray beam was 0.20°. The XRD measurements were performed with the out-of-plane geometry. Atomic Force Microscopy (AFM) measurements of the thin film were performed in a tapping mode using a Shimadzu UV-3101PC. TEM measurements were performed using a HITACHI-SU8010 (200 kV). The formation of polymer thin films was performed using spin coater SWINCO and the thickness of the film was measured with stylus profiler ULVAC DEKTAK 8. SAXS measurements (transmission mode) were conducted at beamline TPS13A of the National Synchrotron Radiation Research Center (NSRRC), Hsinchu, Taiwan. The energy of the employed monochromatic radiation was 15 keV, affording the X-ray wavelength λ of 0.8266 Å. The 2D scattering patterns were collected using an In-vacuum Eiger X 9M (SAXS) detector. The conductivity of polymer films was measured using a digital electrometer ADCMT8340A and an Agilent B1500 semiconductor parameter analyzer. Cyclic voltammetry measurement and electrochemical impedance spectroscopy (EIS) analysis were carried out using a Biologic SP-200 potentiostat. For thin layer chromatography (TLC) analyses throughout the work, Merck-precoated TLC plates (silica gel 60 F254) were used. Purification by HPLC with a preparative SEC column (JAI-GEL-1H, JAI-GEL-2H) was performed using a JAI LC-9201.
:
n = ca. 1
:
1) was prepared in a manner as we reported previously.18 THF (anhydrous grade) was purchased from Kanto Chemical Co. Ltd and passed through alumina and copper columns (Nikko Hansen & Co. Ltd) or distilled from the sodium dispersion in a mineral oil/benzophenone ketyl prior to use.39
:
n = 1.0
:
0.9). 1H NMR (400 MHz, CDCl3): δ 7.89 (d, 2H, J = 8.2 Hz), 7.55 (d, 2H, J = 8.2 Hz), 6.98 (s, 1H), 6.93 (s, 1H), 3.71 (s, 2H), 2.80 (t, 2H, J = 7.8 Hz), 1.74–1.67 (br, 2H), 1.44–1.42 (br, 2H), 1.37–1.34 (br, 4H), 0.93–0.84(br, 12H). 13C{1H} NMR (100 MHz, CDCl3) δ 140.0, 137.9, 135.6, 134.1, 133.8, 130.1, 128.7, 128.4, 80.0, 31.8, 30.6, 29.5, 29.4, 26.1, 22.7 14.3. Several signals were found to be not observed because of their broadening. IR (ATR) 3760, 3483, 3409, 2971, 2958, 1739, 1366, 1228, 1217, 1188 cm−1.
The related block copolymers 5b–5g composed of different ratios of m
:
n were prepared in a similar manner shown above and the results are summarized in Table 1.
Footnote |
| † Electronic supplementary information (ESI) available: Experimental details, further details on microscopic analyses, SAXS analyses, electric conductivity measurements, and electrochemical analyses. See DOI: https://doi.org/10.1039/d2tc05454j |
| This journal is © The Royal Society of Chemistry 2023 |