Takuya
Kochi
,
Kenji
Yoshimura
and
Kyoko
Nozaki
*
Department of Chemistry and Biotechnology, Graduate School of Engineering, The University of Tokyo, Tokyo 113-8656, Japan. E-mail: nozaki@chembio.t.u-tokyo.ac.jp; Fax: +81 3 5341 7261; Tel: +81 3 5841 7261
First published on 14th November 2005
Reaction of diarylphosphinobenzene-2-sulfonic acids with tertially amines, followed by addition of [PdMeCl(cod)], provided anionic methylpalladium(II) complexes with bidentate phosphine–sulfonate ligands, which show high activity for copolymerization of ethylene with methyl acrylate.
![]() | ||
Scheme 1 Generation of a vacant site on monomethyl palladium complexes with bidentate ligands. |
In the past decade, significant efforts have been devoted to developing late transition-metal catalysts capable of producing highly functionalized polymers by copolymerization of non-polar olefins with readily-available polar olefins. For example, Brookhart's diimine palladium complexes can copolymerize ethylene or propylene with polar vinyl monomers such as methyl acrylate.3c,d The copolymers produced were highly branched and methyl acrylate was incorporated only at the chain end.
In 2002, Pugh and coworkers have developed palladium catalysts which can produce unique polymers such as a highly linear copolymer of ethylene and alkyl acrylates5a and a non-perfectly alternating copolymer of ethylene and CO.5b In their reports, the palladium catalysts were formed in situ by combination of diarylphosphinobenzene-2-sulfonic acid with Pd(dba)2 or Pd(OAc)2. Very recently, Rieger et al. reported the first examples of structurally-characterized palladium complexes with bidentate phosphine–sulfonate ligands, which gave higher catalytic activity and a higher number of extra ethylene insertions in the non-perfectly alternating copolymerization of ethylene and CO.6 To improve the catalyst system, our group has concurrently attempted to synthesize palladium complexes with diarylphosphinobenzene-2-sulfonic acids. Herein we report the first synthesis of methylpalladium(II) complexes with bidentate phosphine–sulfonate ligands. Preliminary results for the copolymerization of ethylene and methyl acrylate are also described.
Treatment of diphenylphosphinobenzene-2-sulfonic acid 1a with diisopropylethylamine in dichloromethane, followed by addition of PdMeCl(cod), afforded the anionic methylpalladium complex 2a in 94% yield (Scheme 2).7 Similarly, the reaction of o-methoxyphenyl substituted ligand 1b gave methylpalladium complex 2b in 92% yield. When triethylamine was used instead of diisopropylethylamine for the reaction of 1a, the corresponding methylpalladium complex 2a′ was also obtained in 89% yield.
![]() | ||
Scheme 2 Preparation of anionic methylpalladium(II) complexes with phosphine–sulfonate ligands. |
The structure of 2a′ was determined by X-ray crystallography‡ to reveal that the phosphine–sulfonate ligand coordinates to palladium in a bidentate fashion, similar to Rieger's dicyclopentadienylethoxy palladium complex (Fig. 1). The palladium complex adopts a square planar geometry and the methyl group is located at the trans-position to the oxygen atom of the sulfonate. It is interesting that a chloride ion is bound to the palladium to form an anionic methylpalladium complex, considering simple loss of the chloride ion would give a neutral complex. The synthesis of 2a can also be performed in acetonitrile, meaning that acetonitrile did not replace the chloride ion on the palladium. Anionic methylpalladium(II) complexes are relatively rare, and as far as we know, all of the previously reported examples possess a dianionic ligand and a methyl group on palladium.8 The nitrogen atom of the ammonium ion is in close contact with the chloride (3.27 Å) and the ammonium hydrogen should be located at a reasonable position for a hydrogen bond with the chloride ligand.
![]() | ||
Fig. 1 Structure of complex 2a′. Hydrogen atoms are omitted for clarity. Thermal ellipsoids are shown at the 50% probability level. Selected bond lengths (Å) and angles (Å): Pd(1)–C(1) 2.051(4); Pd(1)–O(1) 2.184(3); Pd(1)–P(1) 2.2195(12); Pd(1)–Cl(1) 2.3997(12); Cl(1)⋯N(1) 3.268(5). |
Notable temperature dependence was observed in the 1H NMR spectra of complex 2b. At room temperature the two methoxy groups of 2b gave one sharp singlet at 3.58 ppm, which split into two broad signals at 3.49 and 3.66 ppm on cooling at −60 °C. This behavior can be accounted for by considering the conformational interchange of the aryl groups which are arranged in an unsymmetric structure as disclosed by the crystallographic study on complex 2a′ (Scheme 3).
![]() | ||
Scheme 3 Conformational interchange of the aryl groups of 2b. |
Copolymerization of ethylene and methyl acrylate was then investigated with catalyst 2 (Table 1). When the copolymerization was carried out at 80 °C with NaBArf4, 2b showed higher activity than 2a, as is consistent with Pugh's result (entry 2, 3).5a,9,10 Screening of additives such as NaBArf4 and AgOTf was performed to examine the effectiveness in removing the chloride ion from the palladium metal of 2b. Effect of the additives on the copolymerization was quite notable. For example, the catalytic activity of the copolymerization with NaBArf4 at 80 °C (entry 2) was twice as high as that reported by Pugh with Pd(dba)2 and 1b in the same reaction conditions,5a while the reactions with AgOTf or no additive showed much lower activities (entry 6, 8). However, the highest comonomer incorporation was observed with no additive, and particularly, the reaction at 80 °C provided the copolymer with 16% of methyl acrylate incorporated in the polymer chain (entry 8). An increase in reaction temperature generally leads to an enhancement in catalytic activity and a reduction of polymer molecular weight, but no gain or a decrease in the activity was observed at high temperature possibly due to decomposition of the catalyst (entry 1, 6, 8). The highest molecular weight (12500) was obtained in the reaction with no additive at 50 °C for 72 h. Addition of galvinoxyl had little influence on the activity of the reaction with no additive at 80 °C, while significantly reduced activity (4 g mmol−1 h−1) was observed for the reaction with NaBArf4. Espinet and co-workers have reported that galvinoxyl not only serves as a radical trap but can react with a Pd–H bond.11 The observed reduction of activity with NaBArf4 can be attributed to the frequent chain transfer reaction via
β-hydride elimination and generation of a Pd–H bond, relative to those in the reaction with no additive.
Entry | Catalyst | Additive | Temp/°C | Yield/g | Activity/g mmol−1 h−1 | TOF/h−1b | Methyl acrylate incorporation (mol%)c | M n | M w/Mn |
---|---|---|---|---|---|---|---|---|---|
a Conditions: 0.01 mmol of catalyst 2; ethylene pressure, 3.0 MPa; 2.5 mL of methyl acrylate; solvent, 2.5 mL of toluene; reaction time, 15 h. b TOF (turnover frequency): mol olefin/(mol Pd·h). c Determined by 1H NMR spectroscopy. d Copolymerization was carried out for 72 h. | |||||||||
1 | 2b | NaBArf4 (0.01 mmol) | 90 | 2.59 | 17 | 490 | 13 | 3200 | 1.64 |
2 | 2b | NaBArf4 (0.01 mmol) | 80 | 2.61 | 17 | 490 | 13 | 4100 | 1.87 |
3 | 2a | NaBArf4 (0.01 mmol) | 80 | 0.94 | 6.3 | 200 | 7 | 2700 | 1.46 |
4 | 2b | NaBArf4 (0.01 mmol) | 70 | 2.00 | 13 | 400 | 9 | 6200 | 1.54 |
5 | 2b | NaBArf4 (0.01 mmol) | 60 | 1.40 | 9.3 | 290 | 8 | 6100 | 2.44 |
6 | 2b | AgOTf (0.02 mmol) | 80 | 0.49 | 3.3 | 90 | 14 | 2400 | 1.54 |
7 | 2b | AgOTf (0.02 mmol) | 60 | 0.96 | 6.4 | 180 | 14 | 4500 | 1.61 |
8 | 2b | None | 80 | 0.24 | 1.6 | 43 | 16 | 4500 | 1.61 |
9 | 2b | None | 70 | 0.30 | 2.0 | 55 | 15 | 6800 | 1.61 |
10 | 2b | None | 50 | 0.19 | 1.3 | 35 | 14 | 8500 | 2.11 |
11 | 2b d | None | 50 | 0.57 | 0.8 | 22 | 13 | 12![]() |
1.79 |
In conclusion, the first syntheses of methylpalladium(II) complexes with phosphine–sulfonate bidentate ligands were achieved. An X-ray structure of 2a′ revealed that the complexes were rare examples of anionic methylpalladium(II) complexes. We further examined the copolymerization of ethylene and methyl acrylate with the catalysts and found interesting effects of additives on the catalytic activity, the molecular weight and the degree of methyl acrylate incorporation. These preliminary copolymerization results indicate that further examination of additives, ligands, bases used to prepare the catalysts, and reaction conditions would lead to an improved catalyst system for production of the unique highly linear copolymer of ethylene and methyl acrylate.
Footnotes |
† Electronic supplementary information (ESI) available: Synthetic procedures and characterization data for 2a, 2a′ and 2b; X-Ray data for 2a′. See DOI: 10.1039/b512452m |
‡ Crystal data for 2a′: C25H33ClNO3PPdS, M = 600.40, monoclinic, a = 14.186(6) Å, b = 10.679(5) Å, c = 17.678(8) Å, β = 92.549(2)°, V = 2675.5(19) Å3, T = 120(2) K, space group P21/c, Z = 4, µ(Mo-Kα) = 0.958 mm−1, 14896 reflections measured, 4364 unique (Rint = 0.0255) which were used in all calculations. The final wR(F2) was 0.1208 (all data). CCDC reference number 283231. For crystallographic data in CIF or other electronic format see DOI: 10.1039/b512452m |
This journal is © The Royal Society of Chemistry 2006 |