End-group stannylation of regioregular poly(3-hexylthiophene)s

Introduction

Polythiophenes have been intensively studied as semiconducting polymers for various applications such as in organic solar cells (OSCs)¹ and organic field-effect transistors.² To modify the intrinsic properties of polythiophenes, such as crystallinity and optoelectronic properties, and to improve their device performance, much work has been performed on the modifications of their side chains³ and end groups.^{4, 5} Concerning these end groups, it has been reported that hydrogen-terminated regioregular poly(3-hexylthiophene) (P3HT) shows a slightly higher absorption coefficient, stronger photoluminescence, and more highly ordered interchain packing than bromine-terminated P3HT, resulting in better performance in OSCs. Cho et al.⁴ reported that the surface energy of P3HT can be varied by the introduction of different end groups and that an improved phase separation of the active layer can be realized with a matched surface energy between P3HT and [6,6]-phenyl-C₆₁ butyric acid methyl ester, leading to enhanced photovoltaic efficiency.⁵ Besides these simple modifications, end-group functionalization has been used for the initiation of polymerization, leading to the synthesis of various P3HT-based block copolymers.^{6, 7}

There are three possible strategies for functionalizing the end groups of P3HT: the quenching of polymerization with a functionalized quencher,⁸ the initiation of polymerization with a functionalized initiator,⁹ and the post-functionalization of purified polymer. The first two strategies rely on the chain-growth nature of Ni-catalyzed Grignard metathesis (GRIM) polymerization of P3HT, in which the Ni complex serves as the initiator and the growing P3HT chains have living ends during the reaction process. Although both the initiator and quencher methods are effective for introducing functionality into one of the ends, a prerequisite of both approaches is that target end groups should be chemically tolerant to Grignard reagents, thus limiting their applicability to a certain extent. On the other hand, even though the post-functionalization of purified P3HT has certain drawbacks, such as lower introduction rate and selectivity, it is free from the limitations of functionalities and can be used to functionalize both ends of chains.^{10, 11, 12} Taking advantage of this, reactive groups such as carboxylic acids^{10, 11} and aldehydes¹² have been introduced into the ends of P3HT by post-functionalization. However, this approach has been relatively less studied.

In this work, we studied the stannylation of the chain ends of P3HT by post-functionalization. As trialkylstannyl groups are key functionalities for Stille coupling with aryl bromides, the synthesis of stannylated P3HTs enables us to connect various π-conjugated moieties to chain ends by direct conjugation with P3HT.¹³ This type of modification is particularly important for modifying the optoelectronic properties of P3HT, as end groups can have direct electronic coupling with P3HT through the conjugation. We found that both the end groups of P3HT can be successfully lithiated with s-butyllithium (s-BuLi) and stannylated with trimethyltin chloride (Me₃SnCl); however, only mono-stannylated P3HT was formed through reactions with a Grignard reagent and Me₃SnCl.

Experimental procedure

Synthesis

All the chemicals used were purchased from Aldrich, Wako Chemicals or Tokyo Chemical Industry and used without further purification, unless otherwise stated. All the glassware used was dried by heating before use. 2-Bromo-3-hexyl-5-iodothiophene was synthesized by following a reported procedure.¹⁴

Poly(3-hexylthiophene) (Br-P3HT)

A typical procedure reported in the literature¹⁵ was followed. To a 200-ml flask charged with 2-bromo-3-hexyl-5-iodothiophene (2.5 g, 6.7 mmol) and dry THF (67 ml), a solution of i-PrMgCl in THF (2 mol l⁻¹, 3.35 ml) was added dropwise at 0 °C, and the resultant mixture was stirred for 30 min. After removing the mixture from the bath, a designated amount of Ni(dppp)Cl₂ catalyst (0.025–0.1 equiv. to monomer) was immediately added to the solution. After 30 min, the reaction was quenched with 5 mol l⁻¹ HCl aqueous solution. The organic layer was then extracted with CHCl₃ and washed three times with NaHCO₃ aq and water, and then dried with MgSO₄. After removing the solvent by rotary evaporation, Br-P3HT was collected by sequential washing with MeOH and hexane (64%). ¹H NMR (500 MHz, CDCl₃): δ 6.98 (s, 1H), 2.80 (t, 2H), 1.71 (quint, 2H), 1.44 (m, 2H), 1.35 (t, 4H) and 0.92 (t, 3H).

Bis-trimethylstannylated poly(3-hexylthiophene) (Sn₂-P3HT)

To a solution of Br-P3HT (30.3 mg, molecular weight (M_n)=5050, polydispersity index (PDI)=1.11) in dry THF (3 g l⁻¹), N,N,N′,N′-tetramethylethylenediamine (TMEDA) (120 equiv. to Br-P3HT, according to the M_n of Br-P3HT) and s-BuLi (1.05 mol l⁻¹ in cyclohexane/n-hexane, 100 equiv.) were added dropwise via a syringe at −78 °C under N₂. After 30 min, Me₃SnCl (1.0 mol l⁻¹ in THF, 200 equiv.) was injected in one portion. After stirring for 15 min, the reaction mixture was warmed to room temperature. After reaction for another 4 h, the mixture was poured into MeOH. The precipitate was collected by filtration and dried. Yield: 92%. ¹H NMR (500 MHz, CDCl₃): δ 6.98 (s, 1H), 2.80 (t, 2H), 1.71 (quint, 2H), 1.44 (m, 2H), 1.35 (t, 4H), 0.92 (t, 3H) and 0.40 (m, 18H).

Mono-trimethylstannylated poly(3-hexylthiophene) (Sn-P3HT)

To a solution of Br-P3HT (48.3 mg, M_n=5900, PDI=1.09) in dry THF (16 ml) was added dropwise a solution of i-PrMgCl in THF (2 mol l⁻¹, 0.1 ml, 25 equiv.) at room temperature under N₂. After 1 h, the solution was cooled to 0 °C and Me₃SnCl solution (1.0 mol l⁻¹ in THF, 0.82 ml, 100 equiv.) was added dropwise. After stirring for 10 min, the reaction mixture was warmed to room temperature and stirred continuously overnight. The reaction was quenched by pouring the mixture into MeOH under sonication. Yield: 99%. ¹H NMR (500 MHz, CDCl₃): δ 6.98 (s, 1H), 2.80 (t, 2H), 1.71 (m, 2H), 1.44 (m, 2H), 1.35 (t, 4H), 0.92 (t, 3H) and 0.40 (m, 9H).

Material characterization

¹H NMR spectra in CDCl₃ were measured on a JEOL Alpha FT-NMR spectrometer equipped with an Oxford superconducting magnet system (500 MHz). Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) was carried out in the reflection ion mode on an Applied Biosystems BioSpectrometry Workstation model Voyager DE-STR spectrometer using dithranol as the matrix and CHCl₃ as the solvent. Gel permeation chromatography (GPC) was performed at 40 °C using a Shimadzu Prominence system equipped with a UV detector and CHCl₃ as the eluent. The sample solution in CHCl₃ was passed through a polytetrafluoroethylene filter (pore size: 0.2 μm) before injection.

Results and discussion

Regioregular P3HT was synthesized according to the literature through the GRIM polymerization method.¹⁵ Ni(dppp)Cl₂ catalyst was added in one portion, and the reaction mixture was quenched with 5 M HCl aq in order to attain a small PDI and to avoid the dimerization of living P3HT chains,¹⁶ respectively. The number averaged M_n of P3HT determined by GPC was 5600 with a PDI of 1.08 when the amount of Ni(dppp)Cl₂ was 0.05 equiv. to the monomer. The MALDI-TOF-MS profile of the product is shown in Figure 1a, in which most of the end groups can be assigned to P3HT with Br–/H– end groups, and only a small fraction with the H–/H– end groups was observed (ca. 24% by peak intensity). Thus, the product is denoted Br-P3HT hereafter.

Figure 1

MALDI-TOF-MS profiles of (a) Br-P3HT (M_n=5600 and PDI=1.08, by GPC) and (b) stannylated P3HT. Insets: magnified images of largest-fraction region. Observed mass of Br-P3HT: (Br–/H–) 3572, 3738, 3905; (H–/H–) 3657, 3824, 3991. Calculated mass: (Br–/H–) 3572.8, 3739.1, 3905.3; (H–/H–): 3660.2, 3826.5, 3992.7.

The synthetic route of bis-stannylated P3HT (Sn₂-P3HT) is shown in Scheme 1. The end groups of Br-P3HT were successively lithiated with s-BuLi and stannylated with Me₃SnCl. First, we followed a report¹⁷ in which the end groups of P3HT were converted to carboxyl groups by the lithiation of P3HT in the absence of TMEDA, the subsequent deactivation of residual s-BuLi at an elevated temperature and treatment with CO₂. However, this reaction procedure yielded neither mono-stannylated P3HT (Sn-P3HT) nor Sn₂-P3HT as confirmed from the ¹H NMR spectra, even when a large excess (300 equiv.) of Me₃SnCl was added. Note that the reaction mixture exhibited a purple color at −78 °C, indicating that most of the P3HT aggregated at such a low temperature (a THF solution of P3HT at room temperature is orange). This could decrease the reactivity of the end groups. TMEDA has been reported to enhance the polarity of alkyllithium species or to break them into smaller clusters, leading to higher reactivity and faster metalation.^{18, 19} Thus, we have conducted the lithiation of P3HT in the presence of TMEDA. The ¹H NMR spectrum recorded after stannylation shows that the addition of TMEDA to the lithiation process resulted in the formation of stannylated P3HT (see Figure 2 for a typical ¹H NMR spectrum), as indicated by the presence of a characteristic –CH₃ peak at δ=0.40 in the Me₃Sn– group. It should be emphasized here that this is the first synthesis of P3HT-based stannylated compounds.

Figure 2

¹H NMR spectrum of Sn₂-P3HT (M_n=5050, conversion of end groups: 81.4%).

In the MALDI-TOF-MS profile (Figure 1b), there are two series of peaks of which the minor series (ca. 12% by peak intensity) could be assigned to the residual starting P3HT with the Br–/H– end group. The major peak series (ca. 88%) could be derived from stannylated P3HT, as shown in the ¹H NMR spectrum. Compared with the spectrum of the starting material (Figure 1a), the unimodal peak distribution of the major peak series shifted to higher molecular weight as a whole, which suggests the attachment of stannyl groups to the ends. However, it is difficult to distinguish Sn₂-P3HT from Sn-P3HT and P3HT with both ends protonated (denoted as H₂-P3HT) in the MALDI-TOF-MS profile, as their molecular weights are too close. For example, the calculated mass of Sn₂-P3HT with n=23 is 4152, whereas that of Sn-P3HT with n=24 is 4156 and that of H₂-P3HT with n=25 is 4159.

To evaluate the conversion of Me₃Sn– groups quantitatively, the percentage of Me₃Sn- groups in the final product is estimated by calculating the integrated areas of the peaks at δ=2.80 (–CH₂– next to the thiophene ring in the main chain), 2.60–2.56 (–CH₂– next to the thiophene ring at the chain ends) and 0.40 (–CH₃ in Me₃Sn–) in ¹H NMR spectra.²⁰ The conversion from either Br– or H– ends to Me₃Sn– was 59% as estimated from the ¹H NMR spectrum, and the amounts of H– and Br– end groups in the product were about 35 and 6%, respectively, as estimated from MALDI-TOF-MS profile based on the assumption that there is no degradation of C–Sn and C–Br bonds during the MALDI-TOF-MS measurements. This estimation suggests that the major peaks of MALDI-TOF-MS in Figure 1b are from the mixture of Sn₂-P3HT, Sn-P3HT and H₂-P3HT. When the M_n of P3HT was changed from 2500 to 9200, the conversion of the chain ends to Me₃Sn– after stannylation varied between 59 and 72%, suggesting that M_n of P3HT has weak influence on the conversion.

The reaction conditions were analyzed to achieve a maximum conversion of Me₃Sn– groups. First, the amounts of TMEDA and s-BuLi were optimized in the ranges of 10–120 and 50–100 equiv., respectively, to increase the conversion of the reaction. We found that the conversions were in the range of 57–73% and not very sensitive to the amount of TMEDA or s-BuLi. Next, the effect of reaction temperature was investigated. Although it was presumed that TMEDA functions to enhance the activity of BuLi, it has also been reported that TMEDA can stabilize lithiated products.²¹ To confirm this, the lithiation duration was elongated from 1 h to 3 h with 120 equiv. of TMEDA and 100 equiv. of s-BuLi. In this case, no formation of Me₃Sn- groups was observed in the final product, suggesting that the lithiated product is of low stability even at −78 °C. In view of this instability of lithiated P3HT, the lithiation duration was shortened from 1 h to 30 min. As a result, a much improved conversion of 81% was obtained, as calculated from the peak intensities of the ¹H NMR spectrum in Figure 2, suggesting that the presence of TMEDA in the reaction can stabilize lithiated P3HT, thus improving stannylation conversion.

The use of a Grignard reagent instead of s-BuLi can lead to the functionalization of only one end of P3HT (Scheme 1) owing to the selectivity of the thiophene to be brominated at the 2-position. A similar reaction was reported for the synthesis of monocarboxylated P3HT.¹¹ The MALDI-TOF-MS profiles of Br-P3HT and the product (Figure 3) show that Br– groups were completely removed and almost only one series of peaks could be observed after the mono-stannylation. The NMR spectrum of the product (Figure 4) clearly demonstrates that Me₃Sn– groups were successfully attached to the chain ends. The conversion of Br– ends to Me₃Sn– groups in the product is estimated to be 74%. The rest of the polymer should be H₂-P3HT formed by the quenching of the unreacted Grignard end of P3HT.

Figure 3

MALDI-TOF-MS profiles of (a) Br-P3HT (M_n=5900 and PDI=1.09 by GPC) and (b) mono-stannylated P3HT. Insets: magnified images of largest-fraction region. Observed mass after mono-stannylation: (Me₃Sn–/H–) 3490, 3656, 3822. Calculated mass: (Me₃Sn–/H–) 3490.4, 3656.7, 3823.0.

Figure 4

¹H NMR spectrum of Sn-P3HT (M_n=5900, conversion of end groups: 74%).

Conclusion

By choosing the reaction conditions, the end groups of regioregular P3HT were either bis- or mono-stannylated by post-functionalization. Both stannylated P3HTs are expected to be useful in Stille coupling for the synthesis of P3HT-based materials. For example, electron or energy acceptor groups could be attached to either one or both ends of P3HT through π-conjugation.¹² This could lead to novel optoelectronic polymer materials with precisely controlled structures.

Synthetic schemes for Sn₂-P3HT and Sn-P3HT . rt, room temperature.