Introduction

Substituted polyacetylenes exhibit interesting properties and functions, including photoconductivity, electroluminescence, stimuli responsiveness, gas permeability and helical conformation,1, 2, 3, 4, 5 which are mainly based on their conjugated and stiff main-chain structure. A great deal of effort has been made to develop new active catalysts that are effective for a wide range of acetylenic monomers and in precisely controlled polymerization. In the last decade, rhodium (Rh) catalysts have been commonly used for the synthesis of stereoregular polymers from monosubstituted acetylenes such as phenylacetylene (1a),6, 7 propiolic esters8 and N-propargylamides.9 Some of the Rh catalysts have been found to achieve living polymerization of monosubstituted acetylenes. The first example of Rh-catalyzed living polymerization of 1a was reported by Kishimoto et al. using two catalyst systems, namely, [(nbd)Rh(C≡CPh)(PPh3)2]/DMAP (nbd: 2,5-norbornadiene, DMAP: 4-dimethylaminopyridine)10, 11 and [(nbd)Rh(μ-OMe)]2/PPh3/DMAP.12 Another catalyst system for living polymerization, [(nbd)Rh(μ-OMe)]2/1,4-bis(diphenylphosphino)butane, was proposed by Falcon et al.13 Our group reported a ternary catalyst, [(nbd)RhCl]2/PPh3/LiC(Ph)=CPh2.14 A presumed active initiator, [(nbd)Rh{C(Ph)=CPh2}{P(C6H4-p-F)3}], can be isolated from the ternary system and can accomplish living polymerization of 1a with virtually quantitative initiation efficiency (IE).15 The living catalyst is also applicable to the living-like polymerization of N-propargylamides.16

Thus far we have exploited a new class of active Rh catalysts bearing a highly π-acidic diene ligand, tfb.17, 18, 19, 20, 21 These tfb-bearing Rh catalysts feature higher activity in the polymerization of phenylacetylene-type monomers than do the conventional Rh catalysts bearing nbd. For instance, the polymerization of 1a catalyzed by a zwitterionic tfb-Rh complex, (tfb)Rh+[(η6-Ph)B–Ph3] (2), proceeds much faster than that with the corresponding nbd analog, (nbd)Rh+[(η6-Ph)B–Ph3].19

In the course of further study on the tfb-Rh catalysts, we have recently found that a cationic Rh complex having two triphenylphosphine ligands, [(tfb)Rh(PPh3)2]BPh4 (3a), in conjunction with amines polymerizes monomer 1a in a living manner, as briefly reported in a communication.21 The present article is a full paper dealing with detailed investigation of the living polymerization of 1a and its ring-substituted derivatives with cationic tfb-Rh catalysts, including 3a (Scheme 1).

Experimental procedure

Instruments

Monomer conversions were determined by GC (Shimadzu, Kyoto, Japan, GC-14B, capillary column (CBP10-M25-025)); column temperature 125 °C, injection temperature 250 °C, internal standard tert-butylbenzene. The number- and weight-average molecular weights (Mn and Mw, respectively) and molecular weight distribution (MWD, Mw/Mn) of polymers were measured by gel permeation chromatography (GPC) with a JASCO PU-980/RI-930 chromatograph (JASCO, Tokyo, Japan); 40 °C, eluent tetrahydrofuran (THF), columns KF-805 (Shodex, Tokyo, Japan) × 3, molecular weight limit up to 4 × 106, flow rate 1 ml min−1; calibrated with polystyrene standards. 1H (400 MHz), 13C (100 MHz), 19F (373 MHz) and 31P (160 MHz) NMR spectra were all recorded on a JEOL ECX-400 spectrometer (JEOL, Tokyo, Japan) with chemical shifts referenced to internal standards CHDCl2 (5.32 p.p.m., for 1H NMR), CD2Cl2 (53.0 p.p.m., for 13C NMR), and external standards P(OMe)3 (140 p.p.m., for 31P NMR), CFCl3 (0 p.p.m., for 19F NMR). High-resolution mass spectra (HRMS) were measured on a JEOL JMS-HX110A mass spectrometer (cationic complexes 3c–e exhibited mass peaks consistent with their cationic moieties). Elemental analyses were performed at the Microanalytical Center of Kyoto University.

Materials

Phenylacetylene (1a) was purchased (Aldrich, St Louis, MO, USA) and distilled over CaH2 under reduced pressure before use. The metal complexes of (tfb)Rh+[(η6-Ph)B–Ph3] (2),19 [(tfb)Rh(PPh3)2]BPh4 (3a),20 [(tfb)Rh{P(p-tolyl)3}2]BPh4 (3b)22 and [(nbd)Rh(PPh3)2]BPh4 (4)23 were synthesized according to the literature. NaBPh4 (Aldrich), PPh3 (Aldrich), AgBF4 (Wako, Osaka, Japan), AgOTf (Wako), 1b (Wako), 1c (NOF, Tokyo, Japan) and 1d (Wako) were purchased and used without further purification. Solvents were distilled by the standard procedures. iPrNH2 and N,N,N′,N′-tetramethylethylenediamine were purified by distillation over CaH2.

[(tfb)Rh(dppf)]BPh4 (3c)

A Schlenk tube equipped with a three-way stopcock was charged with (tfb)Rh+[(η6-Ph)B–Ph3] (50 mg, 0.077 mmol) and dppf (1,1′-bis(diphenylphosphino)ferrocene, 45 mg, 0.080 mmol) and filled with argon gas, to which CH2Cl2 (7.5 ml) was added. After the mixture was stirred at room temperature overnight, the solvent was removed under reduced pressure, and the formed orange powder was purified by recrystallization from CH2Cl2/Et2O. The obtained solid was dried in vacuo. Yield: 71 mg (77%). 1H NMR (CD2Cl2) δ: 7.72 (m, 8H), 7.59 (m, 12H), 7.32 (m, 8H), 7.01 (m, 8H), 6.86 (m, 4H), 5.65 (brs, 2H, bridgehead CH), 4.44 (br, 4H, Cp), 4.37 (brs, 4H, Cp) and 4.18 (m, 4H, sp2-CH). 13C NMR (CD2Cl2) δ: 164.1 (q, 1JC−B=49.6 Hz, ipso C of BPh4), 140.6 (dm, J=256 Hz, C-F of tfb), 139.0 (dm, J=248 Hz, C-F of tfb), 136.0 (Ph), 133.4 (Ph), 132.1 (Ph), 130.6 (Ph), 129.6 (Ph), 126.9 (m, ipso C of tfb), 125.6 (Ph), 121.7 (Ph), 83.2 (s, sp2-C of tfb), 75.6 (m, α-C of Cp), 74.4 (s, β-C of Cp), 72.2 (m, ipso C of Cp) and 42.8 (s, bridgehead of tfb). 19F NMR (CD2Cl2) δ: –145.8 (d, 2F, J=21.2 Hz C-F of tfb), –158.0 (d, 2F, J=21.6 Hz, C-F of tfb). 31P NMR (CD2Cl2) δ: 27.7 (d, JRh−P=163 Hz). HRMS fast atom bombardment (FAB) calcd. for C46H34F4FeP2Rh (m/z): 883.0476, found: 883.0474.

[(tfb)Rh(PPh3)2]BF4 (3d)

A Schlenk tube equipped with a three-way stopcock was charged with [(tfb)RhCl]2 (30 mg, 0.041 mmol), PPh3 (45 mg, 0.17 mmol) and AgBF4 (18 mg, 0.092 mmol) and filled with argon gas, to which CH2Cl2 (4.0 ml) was added. After the mixture was stirred at room temperature for 5 h, the produced silver salt was removed by filtration. The solvent was removed from the filtrate under reduced pressure to afford orange powder. The formed solid was washed with Et2O (2 ml × 3 times) and then dried in vacuo. Yield: 73.5 mg (95%). 1H NMR (CD2Cl2) δ: 7.58–7.31 (m, 30H), 5.87 (brs, 2H, bridgehead CH), 4.32 (m, 4H, sp2-CH). 13C NMR (CD2Cl2) δ: 140.5 (dm, J=241 Hz, C-F of tfb), 138.9 (dm, J=254 Hz, C-F of tfb), 133.8 [PPh3 (ortho)], 131.4 [PPh3 (para)], 129.6 [PPh3 (ipso)], 129.2 [PPh3 (meta)], 126.9 (m, bridging CAr of tfb), 83.0 (s, sp2-C of tfb) and 42.6 (s, bridgehead CH of tfb). 19F NMR (CD2Cl2) δ: –145.6 (d, 2F, J=22.7 Hz, C-F of tfb), –152.3 (s, 4F, BF4), –158.2 (d, 2F, J=22.7 Hz, C-F of tfb). 31P NMR (CD2Cl2) δ: 28.7 (d, JRh−P=156 Hz). HRMS (FAB) calcd. for C48H36F4P2Rh (m/z): 853.1283, found: 853.1270.

[(tfb)Rh(PPh3)2]OTf (3e)

Catalyst 3e was prepared by the same method as that for 3d by using AgOTf (23 mg, 0.090 mmol) instead of AgBF4. Yield: 80 mg (97%). 1H NMR (CD2Cl2) δ: 7.43–7.31 (m, 30H), 5.86 (brs, 2H, bridgehead CH), 4.31 (brs, 4H, sp2-CH). 13C NMR (CD2Cl2) δ: 140.5 (dm, J=245 Hz, C-F of tfb), 138.9 (dm, J=257 Hz, C-F of tfb), 133.8 [PPh3 (ortho)], 131.4 [PPh3 (para)], 129.4 [PPh3 (ipso)], 129.2 [PPh3 (meta)], 127.0 (m, bridging CAr of tfb), 121.2 (q, 1JC−F=321.2 Hz, CF3), 83.0 (s, sp2-C of tfb), 42.6 (s, bridgehead CH of tfb). 19F NMR (CD2Cl2) δ: 78.3 (s, 3F, OTf), –145.6 (d, 2F, J=22.7 Hz, C-F of tfb), –158.2 (d, 2F, J=22.7 Hz, C-F of tfb). 31P NMR (CD2Cl2) δ: 28.6 (d, JRh−P=157 Hz). HRMS (FAB) calcd. for C48H36F4P2Rh (m/z): 853.1283, found: 853.1274. Anal. calcd. for C49H36F7O3P2RhS: C, 58.69%; H, 3.62%, found: C, 58.43%; H, 4.03%.

Polymerization

All the polymerizations of 1a–d were carried out under an argon atmosphere in a Schlenk tube equipped with a three-way stopcock (for polymerization conditions, see, for example, footnote a in Table 1). A THF solution of 1a–d was added to an Rh catalyst solution of the same volume in the Schlenk tube. The polymerization solution was stirred at 30 °C for 24 h, and the formed polymer was isolated by precipitation in a large amount of methanol, including a drop of acetic acid, filtered with a PTFE membrane and dried under vacuum to constant weight.

Table 1 Polymerization of 1a with a series of cationic Rh catalysts 3a–e and 4a
Full size table

Multistage polymerization

Three-stage polymerization of 1a with catalyst 3a and iPrNH2 was performed by the following experiment. The first-stage polymerization was started by the addition of a 1a (0.22 ml, 2.0 mmol) solution in THF (2.0 ml) to a solution of Rh catalyst 3a (9.4 mg, 8.0 μmol) with iPrNH2 (6.9 μl, 80 μmol) in THF (2.0 ml), which was kept at 30 °C during polymerization. After confirming the completion of the first stage by GC, a part (1.0 ml) of the polymerization solution was taken out and poured into methanol to precipitate the polymer as an orange solid. Then the second feed of 1a (0.16 ml, 1.5 mmol) was added to the remaining polymerization solution. After complete consumption of 1a in the second stage, an aliquot of the polymerization solution (1.0 ml) was taken out and poured into a large amount of methanol to isolate the polymer. To the remaining polymerization solution, the third feed of 1a (0.11 ml, 1.0 mmol) was added and monitored by GC until the end of polymerization. Then the polymerization solution was poured into methanol to isolate the polymer formed at the last stage. Each of the obtained poly(1a)s were subjected to GPC measurement to determine their Mn and Mw/Mn.

Block copolymerization

Monomer 1a was firstly polymerized in THF for 60 min using catalyst 3a and iPrNH2. After the completion of monomer consumption was confirmed by GC, an aliquot was taken out to isolate the formed poly(1a) from excess MeOH, which was subjected to GPC measurement. Then, an equimolar amount of monomer 1b or 1c to monomer 1a was added to the remaining polymerization solution, and the resulting solution was further stirred for 60 min (polymerization conditions: in THF, 30 °C; [monomer]0=[monomer]added=0.50 M, [monomer]0/[Rh]=250, [iPrNH2]/[Rh]=10). The formed block copolymers were isolated in a manner similar to the homopolymer of 1a.

Results and Discussion

Our recent paper demonstrated that a zwitterionic Rh catalyst 2 is very active in polymerization of 1a to yield poly(1a) with high molecular weight (Mn 180 000) (conditions: in THF, 30 °C, 24 h, [1a]0=0.50 M, [1a]0/[2]=500).19 However, the obtained polymer showed rather broad MWD (Mw/Mn 1.80), suggesting the non-living character of the catalytic system. Relatively well-controlled polymerization was achieved by a cationic complex 3a, derived from 2 (Scheme 2), which afforded poly(1a) with lower molecular weight (Mn 54 000) and somewhat narrower MWD (Mw/Mn 1.40) under the same polymerization conditions (Table 1, entry 1). Besides, the addition of amines such as Et3N, Et2NH, nBuNH2, iPrNH2 and pyridine to the polymerization system with 3a was found to reduce the MWD of the formed poly(1a) (Mw/Mn<1.23). The narrowest MWD (Mw/Mn 1.09) with Mn of 100 000 of poly(1a) was accomplished by the combination of 3a and 10 equivalents of iPrNH2 (Table 1, entry 2).21 The 1H NMR confirmed that the formed poly(1a) with Mw/Mn of 1.09 possessed highly stereoregular cis-transoidal structure (cis content=97%) in the main chain as in the case of other Rh catalysts.7

The monomer consumption rate in the polymerization of 1a with catalyst 3a and iPrNH2 was monitored (Figure 1). The polymerization proceeded in first order with respect to monomer concentration, indicating a bimolecular reaction between monomer 1a and the propagating end. The line was found not to pass the origin, implying that the formation of active species at an early stage of the polymerization requires a part of monomer 1a and its rate is rather higher than that of propagation.

Figure 1
figure 1

(a) Time-conversion curves and (b) first-order plots for the polymerization of 1a with Rh catalyst 3a (♦) or 4 (▪) in conjunction with iPrNH2 (conditions: in THF, 30 °C, [1a]0=0.50 M, [1a]0/[Rh]=500, [iPrNH2]/[Rh]=10).

Full size image

The living nature of the polymerization of 1a by the current binary catalyst, 3a/iPrNH2, was also confirmed by the multistage polymerization. In the employed three-stage polymerization of 1a (conditions, in THF, 30 °C, [iPrNH2]/[Rh]=10, [1a]0/[Rh]=250 and [1a]0=[1a]add=0.50 M), monomer 1a was completely consumed at the end of each stage, confirmed by GC analyses. The second and third stages were then started by supplying 1a to the polymerization solution. The Mn of the produced polymers increased in proportion to the consumption of monomer (Mn=55 000 (first stage), 119 000 (second) and 168 000 (third)), whereas MWDs remained in a range of small value (Mw/Mn=1.23 (first stage), 1.13 (second) and 1.17 (third)).

It is of interest to compare the catalytic activity of 3a with that of an analogous cationic complex bearing nbd, 4 (Scheme 1). It has been reported that the highly π-acidic tfb ligand is preferable for high turnover in the propagation reaction of 1a17, 19 and even for the synthesis of high-molecular-weight living poly(1a).18 Results for the polymerization of 1a with 3a and 4 are displayed in Figure 1 and Table 1, entries 1–4. Both 3a and 4 were active for the polymerization of 1a irrespective of the presence or absence of iPrNH2. Interestingly, the rate of monomer consumption with nbd-Rh catalyst 4 was faster than that with tfb-Rh catalyst 3a. It is the opposite tendency to the results obtained using the other type of catalysts, such as [(diene) RhCl]2 and zwitterionic Rh complexes.17, 19 The MWDs of the polymers obtained with 3a were narrower than those with 4 regardless of the presence or absence of iPrNH2.

A series of cationic Rh catalysts 3b–e (Scheme 1) were investigated for the polymerization of 1a (entries 5–12, Table 1). For observing the effect of phosphine ligand, the Rh catalysts, 3b and 3c, bearing more electron-donative bulky monophosphine, P(p-tolyl)3, and bidentate phosphine, dppf, were examined. Both with and without addition of iPrNH2, catalyst 3b gave similar results to those of 3a with respect to the MWD of the formed polymers, although the molecular weights appreciably increased (entries 5 and 6). It suggests that the number of active species derived from 3b is smaller than that derived from 3a. It is supported by the polymerization results using catalyst 3c bearing a dppf ligand. The polymerization with catalyst 3c resulted in even higher molecular weights of poly(1a)s regardless of the addition of iPrNH2 (entries 7 and 8), probably because slow dissociation of the bidentate dppf led to the formation of small number of active propagation species. With addition of iPrNH2 to catalyst 3c, no effect of the obtained poly(1a) on the MWD could be observed, suggesting that iPrNH2 did not accelerate the dissociation of dppf from an Rh center by ligand exchange. These tendencies are consistent with the results of the calculated initiation efficiencies described in Table 1.

The effect of the counter anion of the cationic Rh catalyst was investigated with catalysts 3d and 3e, having tetrafluoroborate and triflate as counter anion, respectively. Both catalysts 3d and 3e without amine afforded poly(1a)s with much higher molecular weights and broad MWDs, more than 2.0 (entries 9 and 11, respectively). In case of both 3d and 3e, addition of iPrNH2 remarkably decreased the MWDs of the produced polymers (entries 10 and 12, respectively). However, the molecular weights of the polymers did not decrease as much as those of the 3a-based polymer, probably because of the insufficient formation of active species from 3d and 3e. Thus, it can be said that the catalyst composed of 3a and iPrNH2 is the most appropriate for precise control of the polymerization among the catalysts examined.

The effect of monomer/Rh ratio ([1a]0/[Rh]) on the polymerization was investigated by changing the catalyst concentration and keeping the monomer concentration at 0.25 M (Table 2). A monomer concentration of 0.25 M instead of 0.50 M was adopted to avoid the formation of insoluble high-molecular-weight polymers. It is noteworthy that the polymer's molecular weight increased with higher monomer/Rh ratio, whereas the monomer conversion was invariably quantitative. The MWD remained lesser than 1.2 in the range of [1a]0/[Rh]=250–2000, supporting the living nature of polymerization. Apparent catalyst initiation efficiencies were calculated on the basis of the molecular weights (Mn), monomer/Rh ratios and yields of the formed polymers, which are displayed in Table 2. Admittedly, the IE tends to be improved with higher monomer/Rh ratio: the highest IE of 82% was achieved in the case of [1a]0/[Rh]=4000, although MWD went broad to 1.47. It suggests that the formation of active initiator requires incorporation of 1a and the rate of the initiation reaction is slower than that of propagation.

Table 2 Effect of monomer/Rh and amine/Rh ratios on the polymerization of 1a with 3a/iPrNH2 catalysta
Full size table

The effect of amine/Rh ratio was also examined by changing the [iPrNH2]/[Rh] ratio from 1 to 500. As seen in Table 2, the Mw/Mn remained below 1.20 in a wide range of amine/Rh (=1–250). It is noted that even 1 equivalent of iPrNH2 to the Rh catalyst is effective to some extent. On the other hand, a large excess of iPrNH2, such as 500 equivalents, showed a slight adverse effect. Thus, it is concluded that 10 equivalents of amine is optimal to suppress the polydispersity.

In the specific polymerizations in entries 4 and 5 of Table 2, the monomer conversion was again monitored by GC (Figure 2). It was confirmed that the polymerizations proceeded in first order with respect to monomer concentration at [1a]0/[Rh] ratios of 1000 or 2000. Furthermore, these first-order plots exactly passed the origin unlike the cases of Figure 1. This is probably because the amount of monomer 1a required for the formation of initiating species is negligible at such high monomer/Rh ratios. Reducing the catalyst ratio to half (from 0.25 to 0.125 mM of [Rh], Figures 2a and b, respectively) did not result in the half rate of monomer consumption: only a slightly lower rate was observed with the half of the catalyst concentration of 0.125 mM. It indicates that the IE increases with lower catalyst concentration, consistent with the results in Table 2, entries 4 and 5.

Figure 2
figure 2

First-order plots for the polymerization of 1a with 3a/iPrNH2 catalyst (conditions: (a) [1a]0=0.25 M, [1a]0/[Rh]=1000, [iPrNH2]/[Rh]=10. (b) [1a]0=0.25 M, [1a]0/[Rh]=2000, [iPrNH2]/[Rh]=10).

Full size image

The present catalyst consisting of 3a and iPrNH2 also showed activity in other polymerization solvents than THF. For example, when the polymerization of 1a was carried out in toluene under the same conditions as those of entry 4, Table 2, practically the same results were obtained (in toluene: polymer yield 100%, Mn 170 000, Mw/Mn 1.10). On the other hand, CH2Cl2 affected the polymerization in a different way; that is, although the polymerization proceeded quantitatively, the polymer showed a higher molecular weight (Mn 300 000) and a larger MWD (Mw/Mn 1.32), suggesting a lower IE.

The following NMR experiments elucidated a part of the roles of iPrNH2 in the current living polymerization. In 31P NMR, complex 3a showed a doublet at 29.6 p.p.m. in CD2Cl2 (Figure 3a). This doublet kept its narrow shape even in the presence of PPh3 (1.1 equivalents) (Figure 3b), indicating slow ligand exchange between the coordinating and free PPh3 in NMR time scale. With addition of iPrNH2 (three equivalents), however, both of the signals apparently became broad (Figure 3c). With 10 equivalents of iPrNH2, the two peaks almost coalesced (Figure 3d). These results show that iPrNH2 accelerates the ligands exchange between the coordinating and free PPh3. A mixture of 3a and iPrNH2 (three equivalents) with no addition of PPh3 gave one broad peak at ca. 26 p.p.m. and did not show any signal assignable to free PPh3 (Figure 3e). This is explained by the fast exchange between the major coordinating PPh3 of 3a and minor dissociated PPh3. Another supportive experimental result was obtained by mixing 3a with a bidentate amine ligand, N,N,N′,N′-tetramethylethylenediamine, which shows a broad peak at ca. –4 p.p.m. assignable to free PPh3 (Figure 3f).

Figure 3
figure 3

31P NMR spectra of (a) 3a, (b) 3a with PPh3 (1.1 equiv), (c) 3a with PPh3 (1.1 equiv) and iPrNH2 (3 equiv), (d) 3a with PPh3 (1.1 equiv) and iPrNH2 (10 equiv), (e) 3a with iPrNH2 (3 equiv) and (f) 3a with TMEDA (2 equiv) in CD2Cl2.

Full size image

Unfortunately, it was unable to identify the newly formed complex in the reaction of 3a with iPrNH2. The 1H and 31P NMR of the mixture of 3a and iPrNH2 in CD2Cl2 even at –85 °C did not give clear results to prove the formation of new Rh species, because the observed spectra at the low temperature remained almost unchanged from the one at room temperature. From the mixture of 3a and iPrNH2 in CD2Cl2, evaporation of all volatiles perfectly reproduced 3a. A CD2Cl2 solution of the residue showed no other signal assignable to new species.

The results shown above indicate that iPrNH2 promotes the dissociation of PPh3 from 3a with its weak coordination. Thus, a certain amine complex like [(tfb)Rh(PPh3)(H2NiPr)]+ may presumably be formed for a short time in the mixture of 3a and iPrNH2. Jimenéz et al.24 recently demonstrated that a cationic Rh complex bearing a phosphine–amine bidentate ligand, namely, [Rh(cod){Ph2P(CH2)3NMe2}][BF4], is active for polymerization of 1a. This complex forms a phenylethynyl-Rh initiating species by reaction with 1a along with the formation of an ammonium moiety on the bidentate ligand. In the present living polymerization, the same experiment could not provide sufficient evidence for the formation of a similar alkynyl–Rh complex because of the instability of the new amine complex. Assuming the formation of the analogous amine–phosphine complex, [(tfb)Rh(PPh3) (H2NiPr)]+, a possible mechanism of the polymerization is depicted in Scheme 3.

To survey further application of 3a/iPrNH2 system, polymerizations of monomer 1b–d (Scheme 1) were experimented and the polymerization results are summarized in Table 3. In the polymerization of 1b bearing a tert-butyl group, the catalyst 3a with iPrNH2 showed excellent activity to afford the corresponding polymer in quantitative yield with a high molecular weight of 190 000, but it exhibited relatively broad MWD (Mw/Mn 1.40) (entry 2). Monomer 1c bearing a trimethylsilyl group also polymerized with the current living catalyst to give poly(1c) in excellent yield with a high molecular weight of 200 000 (entry 3). It is noteworthy that the produced poly(1c) showed even smaller MWD (Mw/Mn 1.13) than that of the poly(1b). It implies that even a slight difference of electron donativity between tert-butyl and trimethylsilyl can affect the control of polymerization with the current catalytic system and the trimethylsilyl-substituted 1c is rather favored for its well-controlling the MWD. In the polymerization of 1d having an electron-withdrawing fluoro group, the monomer could not be sufficiently consumed in the given condition to afford only 37% yield of the corresponding polymer with broad MWD (Mw/Mn 1.70) (entry 4). It appears that the electron-withdrawing substituent lowers the monomer reactivity and renders living polymerization difficult.

Table 3 Polymerization of monomers 1a–d with 3a/iPrNH2 catalysta
Full size table

Block copolymer synthesis

The present catalyst, 3a/iPrNH2, was applied to the block copolymer synthesis by the sequential polymerization of 1a and its derivative 1b or 1c (Scheme 4), which yielded the corresponding block copolymers almost quantitatively. Figure 4a shows the GPC profiles of the first-stage homopolymer, poly(1a), and the second-stage block copolymer, poly(1a)-block-poly(1b). The first-stage polymer showed a unimodal GPC curve (Mn 69 000, Mw/Mn 1.10). The second-stage polymer also exhibited a unimodal peak in a higher-molecular-weight region (Mn 161 000, Mw/Mn 1.18), which is indicative of selective formation of the block copolymer. Poly(1a)-block-poly(1c) showed a GPC peak similar to that of poly(1a)-block-poly(1b), although a small shoulder was observed at the position of the first-stage poly(1a) (Figure 4b). The ratios of the monomer units in the block copolymers were calculated from the peak area ratios of aromatic and either tert-butyl or trimethylsilyl protons in their 1H NMR spectra to be 1:1.14 (1a:1b) and 1:1.07 (1a:1c), respectively. These ratios are close to the supplied monomer ratio of 1:1.

Figure 4
figure 4

GPC charts: (a) poly(1a) and poly(1a)-block-poly(1b) and (b) poly(1a) and poly(1a)-block-poly(1c).

Full size image

Conclusions

A new binary catalyst consisting of [(tfb)Rh(PPh3)2]BPh4 (3a) and iPrNH2 proved to achieve living polymerization of phenylacetylene-type monomers. The present living polymerization is applicable to the syntheses of block copolymers composed of two types of phenylacetylene monomers. Further studies of the present Rh catalyst and living polymerization, especially the polymerization mechanism, are now in progress.

scheme 1

Polymerization of 1a–d with Rh catalysts/iPrNH2.

scheme 2

Synthesis of the cationic Rh catalyst 3a by the reaction of 2 with PPh3.

scheme 3

Possible mechanism of the polymerization of 1a with 3a/iPrNH2. The counter anions of the cationic Rh complexes were omitted for clarity.

scheme 4

Synthesis of block copolymers with 3a/iPrNH2 catalyst.