Mechanistic aspects of the alternating copolymerization of propene with carbon monoxide catalyzed by Pd(II) complexes of unsymmetrical phosphine - Phosphite ligands

Article abstract of DOI:10.1021/ja973199x

The reaction steps responsible for the highly enantioselective asymmetric copolymerization of propene with carbon monoxide catalyzed by a cationic Pd(II) complex bearing an unsymmetrical chiral bidentate phosphine- phosphite, (R,S)-BINAPHOS [(R,S)-2-(diphenylphosphino)-1,1'-binaphthalen-2'- yl 1,1'-binaphthalene-2,2'-diyl phosphite = L¹], have been studied. Stepwise identification and characterization were carded out for catalyst precursors (SP-4-2)- and (SP-4-3)-Pd(CH₃)Cl(L¹) (1a and 1b) and (SP-4-3)- [Pd(CH₃)(CH₃CN)(L¹)]·X¹ (X¹ = B{3,5-(CF₃)₂C₆H₃}₄) (2), and complexes related to the reaction steps, (SP-4-3)- [Pd(COCH₃)(CH₃CN)(L¹)]·X¹ (3), (SP-4-3)- and (SP-4-4)- [Pd{CH₂CH(CH₃)COCH₃}(L¹)]·X¹ (4a and 4b), (SP-4-3)- [Pd{COCH₂CH(CH₃)COCH₃}(CH₃CN)(L¹)]·X¹ (5), and (SP-4-3)- [Pd{CH₂CH(CH₃)COCH₂CH(CH₃)COCH₃}(L¹]·X¹ (6). An X-ray structure of alkyl complex 4a has been obtained. Studies on [Pt(CH₃)₂(L¹)] (8) reveal that the methyl group is more stabilized at a position trans to the phosphine than at the cis position. This is consistent with the structures of 1-6 in which all carbon substituents are trans to the phosphine moiety in their major forms. On the basis of analogous studies using platinum complexes, an isomerization from (SP-4-3)-[Pd(CH₃)(CO)(L¹)]·X¹ (13a) to the (SP-4-4) isomer (13b) is suggested to occur for the CO-insertion process 2 → 3, which results in the activation of the methyl group for the migration to the coordinated CO. Rapid equilibrium was observed between the two isomers 4a and 4b during the CO insertion process to give 5. Theoretical studies have been carded out on the transformation of 3 to 4a and 4b. The B3LYP and MPn calculations indicated that the alkene insertion into the Pd-acyl bond trans to a phosphine is more favorable than that into the Pd-acyl bond trans to a phosphite. The MM3 calculations demonstrated that one specific transition structure is more favorable than the other possible transition structures for the transformation of (SP-4-4)-[Pd(COCH₃)(propene)(L¹)]·X¹ (14b) to 4b. The difference originates from the steric effects of the BINAPHOS ligand, and the results account for high enantio- and regioselectivities experimentally observed. The two key steps, propene insertion into 3 and CO insertion into 4, were monitored by ¹H NMR spectroscopy. The activation energies for these two steps were estimated to be 19.0-19.6 kcal/mol at -20 to 0 °C, their difference being insignificant. The living nature of the copolymerization was proved. Some related chiral ligands were examined for the copolymerization. Copolymerization of other olefins with CO was also investigated.

Full text of DOI:10.1021/ja973199x

J. Am. Chem. Soc. 1997, 119, 12779-12795
12779
Mechanistic Aspects of the Alternating Copolymerization of
Propene with Carbon Monoxide Catalyzed by Pd(II) Complexes
of Unsymmetrical Phosphine-Phosphite Ligands
Kyoko Nozaki,*,† Naomasa Sato,† Yoichi Tonomura,† Masako Yasutomi,†
Hidemasa Takaya,†,‡ Tamejiro Hiyama,† Toshiaki Matsubara,*,§ and
Nobuaki Koga*,§
Contribution from the Department of Material Chemistry, Graduate School of Engineering,
Kyoto UniVersity, Yoshida, Sakyo-ku, Kyoto 606-01, Japan, and School of Informatics and
Sciences and Graduate School of Human Informatics, Nagoya UniVersity, Nagoya 464-01, Japan
ReceiVed September 11, 1997X
Abstract: The reaction steps responsible for the highly enantioselective asymmetric copolymerization of propene
with carbon monoxide catalyzed by a cationic Pd(II) complex bearing an unsymmetrical chiral bidentate phosphine-
phosphite, (R,S)-BINAPHOS [(R,S)-2-(diphenylphosphino)-1,1′-binaphthalen-2′-yl 1,1′-binaphthalene-2,2′-diyl phos-
phite ) L1], have been studied. Stepwise identification and characterization were carried out for catalyst precursors
(SP-4-2)- and (SP-4-3)-Pd(CH3)Cl(L1) (1a and 1b) and (SP-4-3)-[Pd(CH3)(CH3CN)(L1)]‚X1 (X1 ) B{3,5-
(CF3)2C6H3}4) (2), and complexes related to the reaction steps, (SP-4-3)-[Pd(COCH3)(CH3CN)(L1)]‚X1 (3), (SP-4-
3)- and (SP-4-4)-[Pd{CH2CH(CH3)COCH3}(L1)]‚X1 (4a and 4b), (SP-4-3)-[Pd{COCH2CH(CH3)COCH3}(CH3CN)-
(L1)]‚X1 (5), and (SP-4-3)-[Pd{CH2CH(CH3)COCH2CH(CH3)COCH3}(L1)]‚X1 (6). An X-ray structure of alkyl
complex 4a has been obtained. Studies on [Pt(CH3)2(L1)] (8) reveal that the methyl group is more stabilized at a
position trans to the phosphine than at the cis position. This is consistent with the structures of 1-6 in which all
carbon substituents are trans to the phosphine moiety in their major forms. On the basis of analogous studies using
platinum complexes, an isomerization from (SP-4-3)-[Pd(CH3)(CO)(L1)]‚X1 (13a) to the (SP-4-4) isomer (13b) is
suggested to occur for the CO-insertion process 2 f 3, which results in the activation of the methyl group for the
migration to the coordinated CO. Rapid equilibrium was observed between the two isomers 4a and 4b during the
CO insertion process to give 5. Theoretical studies have been carried out on the transformation of 3 to 4a and 4b.
The B3LYP and MPn calculations indicated that the alkene insertion into the Pd-acyl bond trans to a phosphine is
more favorable than that into the Pd-acyl bond trans to a phosphite. The MM3 calculations demonstrated that one
specific transition structure is more favorable than the other possible transition structures for the transformation of
(SP-4-4)-[Pd(COCH3)(propene)(L1)]‚X1 (14b) to 4b. The difference originates from the steric effects of the
BINAPHOS ligand, and the results account for high enantio- and regioselectivities experimentally observed. The
1
two key steps, propene insertion into 3 and CO insertion into 4, were monitored by H NMR spectroscopy. The
activation energies for these two steps were estimated to be 19.0-19.6 kcal/mol at -20 to 0 °C, their difference
being insignificant. The living nature of the copolymerization was proved. Some related chiral ligands were examined
for the copolymerization. Copolymerization of other olefins with CO was also investigated.
Introduction
One of the few examples was reported by Vrieze and his co-
workers in 1992. They investigated the CO insertion into a
methylpalladium complex possessing a phosphorus-nitrogen
ligand, 1-(dimethylamino)-8-(diphenylphosphino)naphthalene
(PAN).3b It has been revealed that both the methyl group in
Pd(CH3)(CF3SO3)(PAN) and the acetyl group in Pd(COCH3)(CF3-
SO3)(PAN) are located trans to the nitrogen site. Two years
later, van Leeuwen and his co-workers prepared an unsym-
metrical bisphosphine ligand which consists of a diarylmonoalk-
enylphosphine site and a trialkylphosphine site.3c With Pt(II)
and Pd(II) complexes of this unsymmetrical bidentate ligand,
they investigated the insertion of CO into phenylplatinum and
methylpalladium complexes. They reported a possible migration
of the phenyl and methyl groups from trans to the trialkylphos-
phine site to cis and the higher activity of the former complex
than the latter in migratory insertion of CO.
The alternating copolymerization of olefins with carbon
monoxide is of great interest due to the potential use of the
resulting polymer as a new material.1 The process includes two
of the representative reactions of a palladium complex: (1) CO
insertion into an alkyl-Pd bond and (2) an olefin insertion to
an acyl-Pd bond.
Insertion of CO into a transition-metal-carbon σ-bond is one
of the most important steps in homogeneous catalytic processes
utilizing CO, and a wide range of studies have been devoted to
this subject.2a In contrast to the intensive studies with mono-
dentate and C2 symmetrical cis bidentate ligands,2 less work
has been reported on unsymmetrical bidentate ligand systems.3
† Kyoto University.
‡ Deceased October 4, 1995.
§ Nagoya University.
X Abstract published in AdVance ACS Abstracts, December 15, 1997.
(1) (a) Drent, E.; Budzelaar, P. H. M. Chem. ReV. 1996, 96, 663. (b)
Drent, E.; van Broekhoven, J. A. M.; Budzelaar, P. H. M. Recl. TraV. Chim.
Pays-Bas 1996, 115, 263. (c) Sen, A. Acc. Chem. Res. 1993, 26, 303. (d)
Drent, E.; van Broekhoven, J. A. M.; Doyle, M. J. J. Organomet. Chem.
1991, 417, 235. (e) Drent, E. Pure Appl. Chem. 1990, 62, 661. (f) Sen, A.
AdV. Polym. Sci. 1986, 73/74, 125. (g) Sen, A. Chemtech 1986, 48.
Compared to the CO insertion, fewer reports have appeared
on the olefin insertion into an acylpalladium.4 Direct observa-
tion of this process has been reported very recently by Rix and
Brookhart using cationic 1,10-phenanthroline-Pd(II) complexes.4f
Important reports have appeared in the last five years describing
S0002-7863(97)03199-5 CCC: $14.00 © 1997 American Chemical Society

12780 J. Am. Chem. Soc., Vol. 119, No. 52, 1997
Nozaki et al.
mechanistic studies on the alternating copolymerization of
olefins with CO.5 Brookhart and his co-workers have investi-
gated the microscopic steps responsible for the alternating
copolymerization of ethene with CO using the same 1,10-
phenanthroline system.5ab On the basis of the kinetic and
thermodynamic data, they proposed an accurate model for the
polymer chain growth. In two other examples, stepwise
isolation of the intermediates was accomplished by employing
norbornene as the substrate.5c-f Symmetrical bidentate nitrogen
ligands were used in both studies.
Chart 1
When R-olefins such as propene and styrene are used in place
of ethene or norbornene for this copolymerization, regio- and
enantioselectivities of the olefin insertion arise and the control
of these becomes a problematic issue for obtaining stereoregular
polyketones. In the head-to-tail copolymer, a chirotopic center
exists per monomer unit. If the same enantioface of each
R-olefin is selected by a catalyst, the resulting copolymer is
isotactic in which all the chirotopic carbons in a polymer
backbone possess the same absolute configuration. Thus,
asymmetric copolymerization using chiral catalysts is now
attracting much attention. Most of the successful reports on
this subject deal with C2 symmetrical bidentate ligands.6 For
example, Consiglio6a and Sen6b used Pd(II) complexes of C2
symmetrical bisphosphines BICHEP and DUPHOS, respec-
tively, for the asymmetric copolymerization of propene with
CO. Enantioselective copolymerization of (4-tert-butylphenyl)-
ethene with CO has been reported by Brookhart using a Pd-
(II)-chiral bisoxazoline complex as a catalyst.6g Two excep-
tions, phosphine-oxazoline7a and phosphine-phosphinite7b cis
bidentate ligands, have been disclosed in very recent publications
by Consiglio and by Keim, independently. Using the phos-
phine-oxazoline system, Consiglio improved the catalytic
activity in comparison to his previous studies with bis-
phosphines.7a,8 In 1995, we have demonstrated that a cationic
Pd(II) complex of unsymmetrical chiral bidentate phosphine-
phosphite, (R,S)-BINAPHOS [(R,S)-2-(diphenylphosphino)-1,1′-
binaphthalen-2′-yl 1,1′-binaphthalene-2,2′-diyl phosphite] serves
as an efficient catalyst for the asymmetric copolymerization of
propene with carbon monoxide.9 Highly isotactic polyketone
was obtained in high enantioselectivity. The simplicity of the
13C{1H} NMR spectrum of the polyketone demonstrates that
the tacticity control is almost complete (Chart 1). Moreover,
the polymer showed the highest molar optical rotation ever
reported. Our interest in the characteristic unsymmetrical feature
of this bidentate ligand prompted us to explore further the
mechanistic aspects of this catalyst system. In this paper we
describe (1) the stepwise identification of the complexes related
to the reaction steps, (2) the ligand effect of (R,S)-BINAPHOS
on Pd(II) and Pt(II), (3) direct observation of the reaction process
and discussion on the unseen pathways, and (4) the reaction
with the related ligands and (5) with other olefins. Throughout
the studies, the reaction pathway of the alternating copoly-
merization is demonstrated in relation to the characteristic
features of the unsymmetrical bidentate ligand.
(2) For recent reviews on CO insertion with monodentate and C2
symmetrical cis bidentate ligands see: (a) Colquhoun, H. M.; Thompson,
D. J.; Twigg, M. V. Carbonylation. Direct Synthesis of Carbonyl Com-
pounds; Plenum Press: New York, 1991. (b) Yamamoto, A.; Ozawa, F.;
Osakada, K.; Huang, L.; Son, T.-I.; Kawasaki, N.; Doh, M.-K. Pure Appl.
Chem. 1991, 63, 687. (c) Yamamoto, A. J. Organomet. Chem. 1995, 500,
337. For other related publications, see for example: (d) Dekker, G. P. C.
M.; Elsevier, C. J.; Vrieze, K.; van Leeuwen, P. W. N. M. Organometallics
1992, 11, 1598. (e) Grushin, V. V.; Alper, H. Organometallics 1993, 12,
1890. (f) To´th, I.; Elsevier, C. J. J. Chem. Soc., Chem. Commun. 1993,
529. (g) To´th, I.; Elsevier, C. J. J. Am. Chem. Soc. 1993, 115, 10388. (h)
Kapteijn, G. M.; Dervisi, A.; Verhoef, M. J.; van den Broek, M. A. F. H.;
Grove, D. M.; van Koten, G. J. Organomet. Chem. 1996, 517, 123. (i)
Goren, J. H.; Elsevier, C. J.; Vrieze, K.; Smeets, W. J. J.; Spek, A. L.
Organometallics 1996, 15, 3445. (j) Kayaki, Y.; Shimizu, I.; Yamamoto,
A. Bull Chem. Soc. Jpn 1997, 70, 917, 1135, 1141.
(3) CO insertion with C1 symmetrical cis bidentate ligands: (a) Anderson,
G. K.; Lumetta, G. J. Organometallics 1985, 4, 1542. (b) Dekker, G. P. C.
M.; Buijs, A.; Elsevier, C. J.; Vrieze, K.; van Leeuwen, P. W. N. M.; Smeets,
W. J. J.; Spek, A. L.; Wang, Y. F.; Stam, C. H. Organometallics 1992, 11,
1937. (c) van Leewen, P. W. N. M.; Roobek, C. F.; van der Heijden, H. J.
J. Am. Chem. Soc. 1994, 116, 12117. (d) Ru¨lke, R. E.; Kaasjager, V. E.;
Wehman, P.; Elsevier, C. J.; van Leeuwen, P. W. N. M.; Vrieze, K.; Fraanje,
J.; Goubitz, K.; Spek, A. L. Organometallics 1996, 15, 3022. (e) Ankersmit,
H. A.; Veldman, N.; Spek, A. L.; Eriksen, K.; Goubitz, K.; Vrieze, K.; van
Koten, G. Inorg. Chim. Acta 1996, 252, 203. (f) Cavell, K. J.; Jin, H.;
Skelton, B. W.; White, A. H. J. Chem. Soc., Dalton Trans. 1992, 2923. (g)
Cavell, K. J.; Jin, H.; Skelton, B. W.; White, A. H. J. Chem. Soc., Dalton
Trans. 1993, 1973.
(4) Olefin insertion: (a) Samsel, E. G.; Norton, J. R. J. Am. Chem. Soc.
1984, 106, 5505. (b) Brumbaugh, J.; Whittle, R. R.; Parvez, M.; Sen, A.
Organometallics 1990, 9, 1735. (c) Ozawa, F.; Hayashi, T.; Koide, H.;
Yamamoto, A. J. Chem. Soc., Chem. Commun. 1991, 1469. (d) Markies,
B. A.; Rietveld, M. H. P.; Boersma, J.; Spek, A. L.; van Koten, G. J.
Organomet. Chem. 1992, 424, C12. (e) Dekker, G. P. C. M.; Elsevier, C.
J.; Vrieze, K.; van Leeuwen, P. W. N. M.; Roobeek, C. F. J. Organomet.
Chem. 1992, 430, 357. (f) Rix, F. C.; Brookhart, M. J. Am. Chem. Soc.
1995, 117, 1137. (g) Green, M. J.; Britovsek, G. J. P.; Cavell, K. J.; Skelton,
B. W.; White, A. H. Chem. Commun. 1996, 1563.
(5) Stepwise observation: (a) Brookhart, M.; Rix, F. C.; DeSimone, J.
M. J. Am. Chem. Soc. 1992, 114, 5894. (b) Rix, F. C.; Brookhart, M.; White,
P. S. J. Am. Chem. Soc. 1996, 118, 4746. (c) van Asselt, R.; Gielens, E. E.
C. G.; Ru¨lke, R. E.; Elsevier, C. J. J. Chem. Soc., Chem. Commun. 1993,
1203. (d) van Asselt, R.; Gielens, E. E. C. G.; Ru¨lke, R. E.; Vries, K.;
Elsevier, C. J. J. Am. Chem. Soc. 1994, 116, 977. (e) Markies, B. A.;
Verkerk, K. A. N.; Rietveld, M. H. P.; Boersma, J.; Kooijman, H.; Spek,
A. L.; van Koten, G. J. Chem. Soc., Chem. Commun. 1993, 1317. (f) Markis,
B. A.; Kruis, D.; Rietveld, M. H. P.; Verkerk, K. A. V.; Boersma, J.;
Kooijman, H.; Lakin, M. T.; Spek, A. L.; van Koten, G. J. Am. Chem. Soc.
1995, 117, 5263. For a theoretical study, (g) Svensson, M.; Matsubara, T.;
Morokuma, K. Organometallics 1996, 15, 5568. (h) Margl, P.; Ziegler, T.
J. Am. Chem. Soc. 1996, 118, 7737. (i) Margl, P.; Ziegler, T. Organome-
tallics 1996, 15, 5519.
(6) Asymmetric synthesis: (a) Barsacchi, M.; Batistini, A.; Consiglio,
G.; Suter, U. W. Macromolecules 1992, 25, 3604. (b) Jiang, Z.; Adams, S.
E.; Sen, A. Macromolecules 1994, 27, 2694. (c) Bronco, S.; Consiglio, G.;
Hutter, R.; Batistini, A.; Suter, U. W. Macromolecules 1994, 27, 4436. (d)
Jiang, Z.; Sen, A. J. Am. Chem. Soc. 1995, 117, 4455. (e) Sperrle, M.;
Consiglio, G. J. Am. Chem. Soc. 1995, 117, 12130. (f) Kacker, S.; Jiang,
Z.; Sen, A. Macromolecules 1996, 29, 5852. (g) Brookhart, M.; Wagner,
M. I.; Balavoine, G. G. A.; Haddou, H. A. J. Am. Chem. Soc. 1994, 116,
3641. (h) Bartolini, S.; Carfagna, C.; Musco, A. Macromol. Rapid Commun.
1995, 16, 9. (i) Bronco, S.; Consiglio, G. Macromol. Chem. Phys. 1996,
197, 355.
(7) (a) Sperrle, M.; Aeby, A.; Consiglio, G.; Pfaltz, A. HelV. Chim. Acta
1996, 79, 1387. (b) Keim, W.; Maas, H. J. Organomet. Chem. 1996, 514,
271.
(8) For papers on other improved activity with achiral catalyst systems,
see for example: (a) Abu-Surrah, A. S.; Wursche, R.; Rieger, B.; Eckert,
G.; Pechhold, W. Macromolecules 1996, 29, 4806. (b) Miliani, B.; Vicentini,
L.; Sommazzi, A.; Garbassi, F.; Chiarparin, E.; Zangrando, E.; Mestroni,
G. J. Chem. Soc., Dalton Trans. 1996, 3139.
(9) Nozaki, K.; Sato, N.; Takaya, H. J. Am. Chem. Soc. 1995, 117, 9911.

Copolymerization of Propene with Carbon Monoxide
J. Am. Chem. Soc., Vol. 119, No. 52, 1997 12781
Results and Discussion
1. Structure Determination of the Complexes Related to
the Reaction Steps. First, we followed the reaction steps of
the copolymerization by adding one monomer, either propene
or CO, to the catalyst in the absence of the other monomer. In
other words, alkylpalladium species were treated with carbon
monoxide without propene, and acylpalladium complexes were
exposed to propene in the absence of carbon monoxide. Unlike
the complexes with C2 symmetrical bidentate ligands, (R,S)-
BINAPHOS produces cis and trans stereoisomers of square
planar PdX1X2{(R,S)-BINAPHOS} complexes when the ligand
X1 is not equal to X2. The structure of the observed complex
for each step has been determined as follows.
that the ratio of 5:1 for 1a and 1b is that controlled thermody-
namically rather than kinetically (Vide supra).
1.a. Methylpalladium Chloride Complex 1. Neutral Pd-
(II) complex Pd(CH3)Cl{(R,S)-BINAPHOS} (1) was used as a
catalyst precursor for the present copolymerization. The
compound 1 was obtained as a 5:1 mixture of two diastereomeric
isomers by admixture of Pd(CH3)Cl(cod)10 and (R,S)-BINA-
PHOS. The structure of each diastereomer has been determined
by 31P and 13C NMR spectroscopy of carbon-13-enriched
1-13CH3 which was prepared by disproportionation of Pt(13CH3)2-
(cod) and PdCl2(PhCN)2 in the presence of (R,S)-BINAPHOS
(eq 1).11 The major isomer was assigned as 1a-13CH3 whose
1.c. Acyl Complex 3. The structure of the acylpalladium
species 3 was determined using 13CO. In order to detect the
methyl proton in the acetyl group by H NMR, we used CD3-
CN in place of CH3CN. When a cationic complex, [Pd-
(CH3)(CD3CN){(R,S)-BINAPHOS}]‚[BAr4] (2-D), was treated
with 1 atm of 13CO, the corresponding acyl complex, [Pd-
(13COCH3)(CD3CN){(R,S)-BINAPHOS}]‚[BAr4] (3-D-13CO),
was obtained as a single species (eq 4). The 2J{C-P} coupling
1
constants for the acetyl carbon were 94 Hz with the phosphine
and 20 Hz with the phosphite, respectively. Those values are
in good agreement with the data reported by To´th and Elsevier
for Pd(13COCH3)(13CO){(2S,4S)-2,4-bis(diphenylphosphino)-
methyl group is located at a position trans to the phosphine
2
site, because J{P-C} in this complex was larger (92 Hz) for
2
pentane} in which J{C-P} for the acetyl carbon was 79 Hz
the methyl carbon and the phosphine than that (5 Hz) for the
2
methyl and the phosphite.12 The opposite order in J{P-C},
with its trans phosphorus nucleus and 19.8 Hz with its cis
phosphorus nucleus.2g,13 Thus, the structure of 3-D-13CO has
been assigned wherein the acetyl group is placed trans to the
phosphine. It is noteworthy that, in 3-D-13CO, acetonitrile, but
not CO, coordinates to Pd under 1 atm of CO. This has been
150 Hz for the phosphite and less than 1 Hz for the phosphine,
was observed in the minor complex, which was assigned as
1b-13CH3.
The 5:1 ratio of these diastereomers seems to reflect the
relative thermodynamic stability of 1a and 1b rather than their
kinetic selectivity, which will be discussed in section 1.b. In
the reaction of eq 1, a platinum complex, (SP-4-2)-Pt(CH3)Cl-
{(R,S)-BINAPHOS}, an analog of 1a, was formed as a single
species, simultaneously.
1.b. Cationic Methylpalladium Complex 2. A diastereo-
meric mixture of neutral complexes 1a-13CH3 and 1b-13CH3
was treated with NaBAr4 (Ar ) 3,5-(CF3)2C6H3) to give a
cationic complex, [Pd(13CH3)(CH3CN){(R,S)-BINAPHOS}]‚
[BAr4] (2-13CH3), as shown in eq 2. A single species was
observed by 31P and 13C NMR spectrometries. A large coupling
constant of 2J{P-C} ) 78 Hz was observed between the methyl
group and the phosphine, indicating that those two were trans
to each other.12 The coupling between the methyl and the
phosphite was less than 1 Hz.
2
proved by the fact that the J{C-P} was not observed for the
coordinated CO. This is in contrast to the observation by To´th
and Elsevier in which the 2J{C-P} was detected for the
phosphorus ligand and the coordinated CO in [Pd(13CO-
CH3)(13CO){2,4-bis(diphenylphosphino)pentane}]‚[BF4].2g In
the present phosphine-phosphite system, the π-acidic nature
of the phosphite may prefer the coordination of acetonitrile at
its trans phosphorus nucleus because the π-acidity of acetonitrile
is lower than that of CO.
1.d. Alkyl Complex 4. Olefin insertion into an acyl complex
is reported to give an alkyl complex in which the ketonic oxygen
coordinates to the metal to form a five-membered ring.4,5 In
(12) (a) Pidcock, A. In Transition Metal Complexes of P-, As-, and Sb-
Ligands; McAuliffe, C. A., Ed.; Macmillan: New York, 1973; pp 1-31.
(b) Hitchock, P. B.; Jacobson, B.; Pidcock, A. J. Chem. Soc., Dalton. Trans.
1977, 2038 and 2043. (c) Pidcock, A.; Richards, R. E.; Venanzi, L. M. J.
Chem. Soc. A 1966, 1707. (d) Allen, F. H.; Pidcock, A. J. Chem. Soc. A
1968, 2700.
Treatment of 2 with tetraethylammonium chloride gave an
8:1 mixture of 1a and 1b after 90 min and ended up in a ratio
of 5:1 after 24 h at 20 °C (eq 3). This experiment suggests
(13) Chen reported similar 2J{C-P} values for cis- and trans-
Pd(13COOCH3)(CH3)(PPh3)2. The values are 163 Hz for the P-C trans to
each other and 12 Hz for the those cis to each other. Chen, T. W.; Yeh, Y.
S.; Yang, C. S.; Tsai, F. Y.; Huang, G. L.; Shu, B. C.; Huang, T. M.; Chen,
Y. S.; Lee, G. H.; Cheng, M. C.; Wang, C. C.; Wang, Y. Organometallics
1994, 13, 4804.
(10) Ru¨lke, R. E.; Ernsting, J. M.; Spek, A. L.; Elsevier, C. J.; van
Leeuwen, P. W. N. M.; Vrieze, K. Inorg. Chem. 1993, 32, 5769.
(11) Visser, J. P.; Jager, W. W.; Masters, C. Recl. TraV. Chim. Pays-
Bas 1975, 94, 70.

12782 J. Am. Chem. Soc., Vol. 119, No. 52, 1997
Nozaki et al.
this study, such alkyl complexes were obtained as a 4:1 mixture
of two diastereomers. Complexes 4a and 4b were obtained
when the acyl complex 3 was treated with propene (eq 5). At
first, the structure determination of the diastereomers was carried
out using carbon-13-labeled dodecene, C10H21CHd13CH2, in
place of propene due to the uneasy treatment of CH3CHd13CH2.
Addition of C10H21CHd13CH2 to acyl complex 3 afforded two
alkyl complexes, 7a and 7b, as a 4:1 mixture (eq 6). In the
Figure 1. An ORTEP drawing for [Pd{CH2CH(CH3)C(dO)CH3}-
{(R,S)-BINAPHOS}]‚[B{3,5-(CF3)2C6H3}4]. The counteranion and all
hydrogen atoms are omitted. Representative bond lengths (Å) and angles
(deg): Pd(1)-P(2) 2.181(2), Pd(1)-P(3) 2.381(2), Pd(1)-C(68)
2.067(9), Pd(1)-O(6) 2.098(9), C(41)-O(6) 1.221(11), C(41)-C(58),
1.477(13), C(58)-C(68) 1.479(19), P(2)-Pd(1)-P(3) 96.1(1), P(2)-
Pd(1)-C(68) 92.7(3), P(3)-Pd(1)-O(6) 91.8(6), O(6)-Pd(1)-C(68)
79.5(4), P(3)-Pd(1)-C(68) 171.3(3), Pd(1)-O(6)-C(41) 115.3(6),
Pd(1)-C(68)-C(58) 108.9(7), O(6)-C(41)-C(82) 121.3(8), C(41)-
C(58)-C(68) 111.4(9), P(2)-Pd(1)-O(6)-C(41) 5.5(11), P(2)-Pd(1)-
C(68)-C(58) 160.3(7), P(3)-Pd(1)-O(6)-C(41) -170.2(7), P(3)-
Pd(1)-C(68)-C(58) -19.8(14), C(68)-Pd(1)-O(6)-C(41) 9.9(7),
O(6)-Pd(1)-C(68)-C(58) -19.1(7), Pd(1)-O(6)-C(41)-C(58) 2.6(7),
O(6)-C(41)-C(58)-C(68) -19.6(9).
copolymerization using the present catalyst system, because it
discloses that the SSSS... isomer is produced in more than 95%
ee out of the two possible enantiomers RRRR... and SSSS....
The relation between the absolute configuration of the chirotopic
carbon and the optical rotation of the polymer, that is, the SSSS...
isomer shows positive optical rotation in (CF3)2CHOH, is
consistent with the very recent studies by Bronco and Consiglio
based on the sign of the CD band associated with the n f π*
transition and the application of the octant rule to the carbonyl
chromophore.6i
major compound, the alkyl group was determined to be trans
to the phosphine by 2J{P-C} in 31P and 13C NMR. The
similarity in 31P and 13C NMR charts for 4 and 7 revealed that
4 is also a mixture of the diastereomers analogous to 7a and
7b (eq 5; see the Experimental Section for the NMR data). Rapid
equilibrium seems to exist between the two diastereomers 4a
and 4b, which will be discussed in section 3.
The 4:1 mixture of 4 was treated with CO in methanol to
give the corresponding keto ester methyl (S)-3-methyl-4-oxo-
pentanoate in 95% ee (eq 7).9 This result is very important for
The major complex 4a was recrystallized from CH2Cl2 and
diethyl ether and analyzed by X-ray crystallography. Its ORTEP
drawing is shown in Figure 1. Two crystal structures of the
five-membered alkylpalladium complexes have been reported
using norbornene as the olefinic precursor.4b,5f Very recently,
one structure of PdCH2CH2COCH3(ligand) has been reported
by Cavell and his co-workers.4g Accordingly, the structure of
4a is the second example of X-ray analysis using an olefinic
precursor in which â-hydride elimination is possible. The four
ligands and the Pd center are almost in plane. The bite angle
of the ligand P-Pd-P is 96.0°, slightly larger than the ideal
angle 90° for the square planar coordination. The phosphine-
Pd bond length of 2.38 Å is much longer than the phosphite-
Pd bond of 2.18 Å, due to the strong trans influence of the
alkyl group. Such a difference in the P-Pd bond length is
consistent with the reported data with a bis(triphenylphosphine)
complex4b and a bipyridine complex.5f In our previous com-
munication, we assigned the coordinating oxygen in 4a to be
3
trans to the phosphine on the basis of the larger J{P-C}
coupling in the phosphine than in the phosphite of 4-13CO (Chart
3
2).9,14 However, the above study has revealed that the J{P-
the discussion of the enantioselectivity in the alternating
(14) For a similar assignment, see ref 4b.

Copolymerization of Propene with Carbon Monoxide
J. Am. Chem. Soc., Vol. 119, No. 52, 1997 12783
Chart 2
C} coupling through 13CdO-Pd-P is smaller for the trans
isomer than the cis isomer in this case.
1.e. Following Complexes 5 and 6. The following acyl
complex 5 and alkyl complex 6 exhibited 31P NMR spectra
superimposable on those of the first generation acyl complex 3
and alkyl complex 4, respectively (eq 8). Thus, 5 and 6 were
Figure 2. (a) 13C NMR of Pd[COCH2CHCH3COCH3]‚[(R,S)-BINA-
PHOS] in the absence of CD3CN. The 13C-labeled δ-carbonyl carbon
is observed as a broad peak at δ 219-228. The R-carbonyl which is
not 13C-labeled is observed at δ 240 as a doublet of doublets. (b) The
same in the presence of CD3CN (5-D). The δ-carbonyl carbon (not
labeled in this case) is sharpened at δ 210. The R-carbonyl is observed
at δ 223 as a doublet of doublets.
Single-crystal X-ray analysis of 8 shows its structure in detail.
As shown in Figure 3, the C-Pt distance trans to the phosphine
is 2.102(10) Å which is possibly shorter than that trans to the
phosphite, 2.135(12) Å {t ) (2.135 - 2.102) × (0.0102 +
0.0122)-1/2 ) 2.1}. This implies that the methyl group trans
to the phosphine site is stabilized compared to that at the cis
position in the (R,S)-BINAPHOS complexes. The following
NMR studies support this conclusion.
We labeled the two methyl groups of 8 with carbon-13, and
compared the 2J{P-C} to find whether each methyl group was
either cis or trans to the phosphine. The representative δ and
J values are summarized near formula 8-13CH3 in Chart 3.
Generally, the coupling constant 1J{Pt-C} is mostly consistent
with the bond strength when two similar bonds are compared.12
Here, the Pt-C bond is compared. Accordingly, the stronger
Pt-C bond trans to the phosphine has been suggested by the
assigned to have the same cis/trans configuration as 3 and 4.
In the absence of acetonitrile, the δ-carbonyl in 5 weakly
coordinates to the Pd center to form a six-membered ring.5 This
was observed by 13C NMR as the peak broadening of the sp2
carbon of the ketone group in the 13CO-labeled 5-13CO (Figure
2a). By the addition of 1 equiv of CD3CN, the peak was
sharpened with an upfield shift, showing the formation of 5-D
(Figure 2b). The R-carbonyl, attached directly to the palladium,
also upfield shifted. It is noteworthy that, among the two
possible alkyl complexes 6a and 6b, 6a which corresponds to
4a was observed as a single species. The diastereomeric
structure 6b corresponding to 4b was not detected. Due to the
longer chain in 6 than in 4, the steric bulk around the
coordinating carbonyl oxygen increases in 6. This may be the
reason for the larger equilibrium constant for [6a]/[6b] than for
[4a]/[4b] (the existence of the equilibrium will be discussed in
section 3).
The results obtained in section 1 are summarized in Scheme
1. All the major isomers observed have either alkyl or acyl
groups trans to the phosphine site.
2. The Stabilization and Activation Influence of (R,S)-
BINAPHOS. In this section, we describe the higher thermo-
dynamic stability for a methyl group trans to the phosphine
site compared to trans to the phosphite site in a square planar
(R,S)-BINAPHOS complex, Pt(CH3)2{(R,S)-BINAPHOS} (8).
1
larger J (588 Hz) for Pt-C trans to the phosphine than that
trans to the phosphite (481 Hz) (Chart 3). This tendency
contradicts the generally accepted “trans influence order” in
which phosphine is ranked at either a higher or at least even
level compared to phosphite.15
In addition to 8-13CH3, the corresponding dimethylplatinum
complexes 9-13CH3 and 10-13CH3 have been prepared from (R)-
BINAP {(R)-2,2′-(diphenylphosphino)-1,1′-binaphthyl} and (R)-
BISPHOSPHITE {(R)-1,1′-binaphthyl-2,2′-diyl bis(diphenyl
1
phosphite)}, respectively. When the J{Pt-C} for the methyl
trans to the phosphine is compared, the value in 8-13CH3 (588
Hz) is much larger than that in 9-13CH3 (493 Hz). At the same
1
time, the J for Pt-C trans to the phosphite in 8-13CH3 (481
Hz) is much smaller that in 10-13CH3 (574 Hz) (Chart 3).
Rather than the trans relation, therefore, comparison of the cis
relation shows better agreement, that is, between 481 Hz (8-
13CH3) and 493 Hz (9-13CH3) for cis to the phosphines and
between 588 Hz (8-13CH3) and 574 Hz (10-13CH3) for cis to
(15) (a) Allen, F. H.; Pidcock, A.; Waterhouse, C. R. J. Chem. Soc. A
1970, 2087. (b) Appleton, T. G.; Bennett, M. A. Inorg. Chem. 1978, 17,
738.

12784 J. Am. Chem. Soc., Vol. 119, No. 52, 1997
Nozaki et al.
Scheme 1
Chart 3
BINAP complexes, and (2) the Pt-P bond for the phosphite is
stronger in BINAPHOS complexes than in BISPHOSPHITE
complexes.
Figure 3. An ORTEP drawing for Pt(CH3)2{(R,S)-BINAPHOS} (8).
All hydrogen atoms are omitted. Representative bond lengths (Å) and
angles (deg): Pt(1)-C(11) 2.102(10), Pt(1)-C(37) 2.135(12), Pt(1)-
P(2) 2.315(2), Pt(1)-P(3) 2.206(2), P(2)-Pt(1)-P(3) 95.2(1), P(2)-
Pt(1)-C(11) 167.9(4), P(2)-Pt(1)-C(37) 86.8(4), P(3)-Pt(1)-C(11)
94.7(4), P(3)-Pt(1)-C(37) 176.4(4), C(11)-Pt(1)-C(37) 83.0(5).
In summarizing this section, X-ray data combined with the
NMR study have revealed that a carbon substituent, a methyl
group, is stabilized trans to the phosphine site in (R,S)-
BINAPHOS complexes. This result is consistent with the above
observation in section 1 that both alkyl and acyl groups are
placed trans to the phosphine site in all the major isomers of
1-6 and 7.
3. A Proposed Pathway for the Reaction Process. In
section 1, we have identified the stable complexes related to
the copolymerization process. In section 3, efforts are devoted
to see or predict the actual reaction pathways. The results are
summarized in Scheme 2.
the phosphites. The coupling between Pt and the protons of
the methyl group (2J{Pt-CH3}) shows the same tendency.16
Thus, apparently, the “cis influence” seems more significant
than the trans influence in this system.17 Steric factors may be
another possibility for such a tendency.
Meanwhile, comparison of the 1J{Pt-P} values discloses that
1
(1) the J{Pt-P} of the phosphine in 8 is 1602 Hz, smaller
1
3.a. The First CO Insertion, 2 f 3. The insertion of CO
into a carbon-metal bond in bivalent group 10 metal complexes
is known to proceed through a migration. Recently, van
Leeuwen and his co-workers reported an elegant example using
a platinum complex of an unsymmetrical bidentate bisphosphine
ligand, L-L′ ) 1-(diphenylphosphino)-2-tert-butyl-3-(dicyclo-
hexylphosphino)-1-propene.3c Migration of the phenyl group
in [Pt(Ph)(solvent)(L-L′)]‚[OTf] takes place in the absence of
than 1874 Hz in 9, and (2) the J{Pt-P} of the phosphite in 8
is 3392 Hz which is larger than 3043 Hz in 10 (Chart 4). These
tendencies were also observed in the corresponding PtCl2(L)
and Pt(CH3)(Cl)(L) complexes. Thus (1) the Pt-P bond for
the phosphine seems weaker in BINAPHOS complexes than in
(16) Appleton, T. G.; Clark, H. C.; Manzer, L. E. Coord. Chem. ReV.
1973, 10, 335.
(17) Allen, F. H.; Sze, S. N. J. Chem. Soc. A 1971, 2054.

Copolymerization of Propene with Carbon Monoxide
J. Am. Chem. Soc., Vol. 119, No. 52, 1997 12785
Chart 4
[Pd(CH3)(CO){(R,S)-BINAPHOS}]‚[BAr4] migrates from trans
to the phosphine to cis to the phosphine prior to the migration
to the coordinated CO as shown in 13a f 13b in eq 10. This
corresponds to the observation in the Pt complexes, 11 f 12a
f 12b (eq 9). With this isomerization, the methyl group in 2
may become more actiVated trans to the phosphite than at the
original position (see discussions in section 2). In this way, its
migration to the coordinated CO should be accelerated to give
the acyl complex 3.5g,18 A similar isomerization of an alkyl-
platinum complex prior to the migration of the alkyl group to
the coordinated CO is well-known in monodentate phosphine
complexes.19
3.b. The Subsequent CO Insertion into the Five-
Membered Alkyl Complex 4. Rapid equilibrium is suggested
to take place between the alkyl complexes 4a and 4b, on the
basis of the following observations. When a 4:1 mixture of 4a
and 4b was treated with 1 atm of CO in the presence of CH3-
CN, the corresponding acyl complex 5 was formed as a single
CO. Similarly, we have prepared a cationic Pt complex, [Pt-
(CH3)(CH3CN){(R,S)-BINAPHOS}]‚[B{3,5-(CF3)2C6H3}4] (11),
and observed a migration of the methyl group in the presence
of CO. As was so for its Pd analog 2, the methyl group was
proven to be trans to the phosphine by carbon-13 labeling on
the methyl group. Next, the cationic Pt complex 11 was treated
with CO (eq 9). At 1 atm of CO, only the replacement of CH3-
product. The ratio of 4a and 4b was determined by 1H and 31
P
NMR to be constant during the CO insertion reaction; in other
words, the 4:1 ratio did not change during the CO insertion.
Thus, it seems reasonable to explain this result by the rapid
equilibrium between 4a and 4b rather than by the exactly same
reactivity of 4a and 4b against the CO insertion. At this
moment, no direct proof has yet been obtained regarding which,
4a or 4b, is actually involved in the catalytic process. Never-
theless, we may predict that the alkyl group could be placed
trans to the phosphite (4b) prior to the CO insertion into 4 as
shown in eq 11 in a manner similar to that which we proposed
CN to CO occurred and the carbonyl complex 12a was observed
as a single species. At elevated CO pressure (20 atm)
isomerization of 12a to 12b took place to give a mixture of
12a/12b ) 1/8. Since the higher pressure was required for this
isomerization, the coordination of CO to the platinum at its
apical site as the fifth ligand seems to be essential for this
process. After the CO pressure was released, the ratio of 12a
to 12b was not changed. Accordingly, 12b is the thermody-
namically more stable form of 12 in which the methyl is trans
to the phosphite. This can be attributed to the strong π-acidic
nature of CO which prefers phosphine rather than phosphite at
its trans position. In fact, the infrared absorption of the carbonyl
in 12b (νCO ) 2128 cm-1) suggests a slightly stronger back-
donation from the platinum metal than in 12a (2132 cm-1).
For the palladium species, van Leeuwen reported that both
[Pd(CH3)(solvent)(L-L′)]‚(OTf) and [Pd(COCH3)(solvent)(L-
L′)]‚[OTf] are under rapid equilibrium between their cis/trans
isomers.3c In our Pd-(R,S)-BINAPHOS system, similar cis/
trans isomerization should have also taken place either before
or after the migratory CO insertion, because both the alkyl group
in 2 and the acyl group in 3 are placed trans to the phosphine,
exclusively. Although no direct evidence is obtained in the
palladium system, here we presume that the methyl group in
for the first CO insertion into 2 in eq 10. In this way, CO
insertion proceeds from 4b to 5 Via 15. For 6, 6a was observed
as a single species and 6b which corresponds to 4b was not
detected in the absence of CO. However, with our assumption
that 4b is more suitable for CO insertion than 4a, the invisible
6b might be responsible for the subsequent CO insertion.
(18) Migration of the methyl group is also suggested by theoretical
studies. (a) Koga, N.; Morokuma, K. J. Am. Chem. Soc. 1986, 108, 6136.
(b) Koga, N.; Morokuma, K. Chem. ReV. 1991, 91, 823.
(19) (a) Anderson, G. K.; Cross, R. J. Acc. Chem. Res. 1984, 17, 67. (b)
Anderson, G. K.; Cross, R. J. Chem. Soc. ReV. 1980, 9, 185. (c) Anderson,
G. K.; Clark, H. C.; Davies, J. A. Organometallics 1982, 1, 64.

12786 J. Am. Chem. Soc., Vol. 119, No. 52, 1997
Nozaki et al.
Scheme 2
3.c. Theoretical Studies for the Olefin Insertion to Acyl
Complex 3. Olefin insertion into C-M σ-bonds where M is a
d8 metal such as Ni(II), Pd(II), and Pt(II) has been of much
interest not only for the olefin-CO copolymerization but also
for other catalytic reactions, for example, polymerization of
olefins. Intensive theoretical studies have been performed on
this subject in the last decade, to show that the transition state
is a four-centered metallacycle.5g-i,20 Simply applying the
theoretical results to the present experimental system, it is
expected as shown in eq 12 that CH3CN of 3 was displaced by
modeling, we could shed light on the intrinsic electronic effects
of atoms or groups such as Pd, phosphine, and phosphite on
the reaction. Steric effects of the bulky (R,S)-BINAPHOS
ligand will be examined later in this section. We determined
the structures of the reactants, products, and transition states
(TSs) for the model reactions, as shown in Figure 4, using the
density functional B3LYP method with the smaller basis set,
called the B3LYP/I method. Furthermore, for the structures
thus determined, the more reliable energies were calculated using
the B3LYP and Møller-Plesset perturbation method with better
basis set II. The energies relative to 14b′ are summarized in
Table 1.21
Reactant 14b′ with an acyl group trans to the phosphine was
calculated at all the levels with better basis set II to be more
stable than another reactant, 14a′, with the acyl group trans to
the phosphite, which agrees with the experimental structure of
acyl complex 3 with the acyl group trans to the phosphine.
Product 4a′ with an alkyl ligand trans to the phosphine is more
stable than 4b′, also consistent with the experimental finding
of the 4:1 mixture of 4a and 4b under thermodynamic
equilibrium. The alkyl ligand as well as the acyl ligand prefers
the site trans to phosphine, in agreement with the “stabilization
influence order” experimentally found in section 2. The Pd-P
bond distances in 4a′ and 4b′ show the strong trans influence
of the alkyl ligand; the Pd-P bond trans to the alkyl ligand is
about 0.2 Å longer, in agreement with the trend in 4a (Figure
1).22
Reactions 14b′ to 4b′ and 14a′ to 4a′ pass through the tight
four-centered TSs, ts(14b′-4b′) and ts(14a′-4a′), respectively,
being accompanied by the rotation of ethene from perpendicular
to in-plane coordination. The calculations with basis set II show
that both reactions require a comparable activation energy; the
activation energies for 14b′ to 4b′ and for 14a′ to 4a′ at the
B3LYP/II level are 12.8 and 12.7 kcal/mol, respectively, and
those at the MP4SDQ/II level are 14.7 and 14.4 kcal/mol,
respectively. Accordingly, one can conclude that the reaction
from more stable and thus more populated 14b′ is more
favorable than the reaction from less stable 14a′. Product 4b′
of the more favorable reaction is less stable than 4a′, the product
propene to give π-olefin complex 14b or 14a and that, passing
through transition state ts(14b-4b) or ts(14a-4a), the corre-
sponding alkyl complexes 4b and 4a would be produced. In
this study we have carried out theoretical calculations for the
two processes 14b f ts(14b-4b) f 4b (path B) and 14a f
ts(14a-4a) f 4a (path A) to investigate the mechanism of
propene insertion into a Pd-acyl bond and that of enantiofacial
selection.
First we investigated the mechanism of model reactions in
which propene was modeled by ethene and (R,S)-BINAPHOS
by PH3 and P(OH)3 using density functional theory and ab initio
molecular orbital methods. The model reactants corresponding
to 14b and 14a are designated as 14b′ and 14a′, respectively,
and similarly the model products as 4b′ and 4a′. With this
(21) See the Experimental Section for computational details.
(22) The calculations overestimate the Pd-P distances. It is known that
the metal-P distances tend to be overestimated without phosphorous d
polarization function.23
(20) Ethene insertion into a CH3-Ni bond: (a) Musaev, D. G.; Froese,
R. D. J.; Svensson, M.; Morokuma, K. J. Am. Chem. Soc. 1997, 119, 367.
A carbon-carbon triple bond insertion into an acylpalladium bond: (b)
Samsel, E. G.; Norton, J. R. J. Am. Chem. Soc. 1984, 106, 5505.
(23) Low, J. J.; Goddard, W. A., III; J. Am. Chem. Soc. 1984, 106, 6928.

Copolymerization of Propene with Carbon Monoxide
J. Am. Chem. Soc., Vol. 119, No. 52, 1997 12787
Figure 4. B3LYP/I optimized structure of model reactants 14b′ and 14a′, transition states ts(14b′-4b′) and ts(14a′-4a′), and products 4b′ and 4a′
in the propene insertion into the Pd-acyl bond. The phosphine and phosphite sites of (R,S)-BINAPHOS are displaced by PH3 and P(OH)3, respectively,
and propene is displaced by ethene.
Table 1. Calculated Potential Energies for 14′ f 4′ Relative to
14b′ (kcal/mol) at Various Levels
BINAPHOS ligand. Upon this substitution, four possible
transition state structures, I-IV, are generated from each
transition state. Thus, we compared their steric energies by the
combined MO + MM method to show which insertion mode
is the most favorable.21 The MM3 optimized structures and
relative steric energies of the four insertion modes for both paths
are shown in Figure 5.
14b′ 14a′ ts(14b′-4b′) ts(14a′-4a′) 4b′
4a′
B3LYP/I
B3LYP/II
MP2/II
0.0 -0.9
11.5
12.8
10.3
12.9
14.7
16.0
16.2
14.5
15.5
17.9
-22.0 -20.9
-18.9 -21.5
-12.9 -16.3
-22.7 -25.6
-17.8 -20.8
0.0
0.0
0.0
3.5
3.5
4.0
3.5
MP3/II
MP4SDQ/II 0.0
The steric energy for Ib was found to be the smallest among
four TSs for path B. Since TS Ib for path B leads to 4b, this
result really accounts for the experimental result. TS IIIb which
would result in the opposite enantioselection or IIb and IVb
which would cause the opposite regioselection are less stable
than Ib by more than 3.2 kcal/mol. As shown in Figure 5, the
methyl group of propene in IIb and IVb and that in IIIb have
closer contact with the phenyl group and the naphthalene group
of the (R,S)-BINAPHOS ligand, respectively, than in Ib.
For path A we found that TS Ia which leads to 4a is the
most stable as in the case of path B. However, the other
transition states, especially IVa, are similarly stable to Ia,
different from the case of path B.
We also made the combined MO + MM calculations
similarly for another four possible TSs of I-IV in which the
enantiomers of the optimized structures by the B3LYP/I method
are used and PH3 and P(OH)3 are displaced by (R,S)-BINA-
PHOS and ethene by propene. As shown in Figure 6, there is
no explicit change in the trends of the relative steric energy of
the TSs for both path A and path B. The high enantioselectivity
in the propene insertion such as experimentally observed (95%
ee, eq 7) will not appear if these TSs for path A are involved
in the process. This also supports the above result that path B
is more favorable than path A.
Table 2. Calculated Potential Energy (∆E), Enthalpy (∆H),
Entropy (∆S), and Free Energy (∆G) for 14′ f 4′ Relative to 14b′
(kcal/mol) at the B3LYP/II Level
14b′ 14a′ ts(14b′-4b′) ts(14a′-4a′)
4b′
4a′
∆E
0.0
0.0
0.0
0.0
3.5
3.6
1.3
2.3
12.8
12.3
-3.3
15.6
16.2
15.6
-2.4
18.0
-18.9 -21.5
-17.7 -20.4
∆Ha
T∆Sa
∆Ga
-3.2
-3.3
-14.5 -17.1
a ∆H, T∆S, and ∆G ) ∆H - T∆S calculated at 298.15 K.
of the less favorable reaction. This theoretical finding and the
experimental fact that the product is a 1:4 mixture of 4b and
4a, which corresponds to 4b′ and 4a′, respectively, suggest that
there exists an isomerization pathway between them.
The enthalpy change calculated from the zero-point energy
and thermal correction to a potential energy does not modify
the trend as shown in Table 2. On the other hand, an entropy
effect slightly favors 14a′ and ts(14a′-4a′) more than 14b′ and
ts(14b′-4b′) to a similar extent. As a result, the free energies
of activation for these paths remain to be comparable.
In the second part, the selectivity experimentally observed
in propene insertion is discussed using molecular mechanics
(MM) calculations in which we replaced PH3 and P(OH)3 by
(R,S)-BINAPHOS and ethene by propene in the B3LYP/I
optimized transition state structures. Thus, this combined MO
+ MM method incorporates the steric effects of the (R,S)-
In summary, the B3LYP calculations and MPn calculations
show that alkene insertion into the Pd-acyl bond trans to
phosphine is more favorable than that into the Pd-acyl bond

12788 J. Am. Chem. Soc., Vol. 119, No. 52, 1997
Nozaki et al.
Figure 5. MM3 optimized structures and steric energies relative to I of the four possible isomers of TS I-IV for path B (top) and for path A
(bottom) (see the text for details). All hydrogen atoms are omitted for clarity.
Figure 6. MM3 optimized structures and steric energies relative to I′ of the four possible isomers of TS I′-IV′ for path B (top) and for path A
(bottom) (see the text for details). All hydrogen atoms are omitted for clarity.
1
trans to phosphite, and the MM3 calculations demonstrate that
one specific transition state among four possible transition states
for propene insertion with the (R,S)-BINAPHOS ligand is the
most favorable. These results unequivocally demonstrate that
the most probable path for eq 12 is 3 f 14b f ts(14b-4b) f
4b f (4b+4a).
3.d. Kinetic Studies. A kinetic study was carried out on
the two representative processes of (1) propene insertion into 3
to give 4 and (2) CO insertion into 4 to give 5. The reaction of
3 with propene was monitored by 1H NMR at -10 and 0 °C. A
CDCl3 solution of 3 was saturated with propene (1 atm) in an
NMR tube, and the conversion was monitored by H NMR.
The process was first-order in palladium complex concentration,
and the observed rate constant kobs was 5.3 ( 2.0 × 10-4 (∆Gq
) 19.3 ( 0.2 kcal/mol) at -10 °C as the average of five
independently repeated experiments. At 0 °C, kobs was 11.8 ×
10-4 (∆Gq ) 19.6 kcal/mol). An analogous experiment with
the 1,10-phenanthroline-Pd system has been reported by
Brookhart and his co-workers.5b They determined ∆Gq ) 16.6
kcal/mol at -46 °C for the ethene insertion into Pd(COCH3)-
(C2H4)(phen). The larger ∆Gq in our system should largely
depend on the difference of the employed olefins. The reaction

Copolymerization of Propene with Carbon Monoxide
J. Am. Chem. Soc., Vol. 119, No. 52, 1997 12789
31P NMR data are listed in Table 3. By using these cationic
complexes as catalyst precursors, asymmetric copolymerization
of propene with CO was examined. The catalytic activity was
compared by quenching the reaction after 6 h. The yield and
the molecular weight (Mn and Mw/Mn) of the resulting polyke-
tones are also summarized in Table 3.
When a symmetrical bis(triarylphosphine), (R)-BINAP, was
used as the ligand, both the neutral and the cationic complexes
were obtained quantitatively. Nevertheless, no polymeric or
oligomeric products were produced. We investigated the CO
and propene insertion in a stepwise manner (eq 13). When the
Figure 7. Mn plotted against the conversion of propene into the
copolymer catalyzed by Pd(II)-(R,S)-BINAPHOS.
Chart 5
cationic complex [Pd(CH3)(CH3CN)(R)-BINAP]‚[B{3,5-
(CF3)2C6H3}4] was treated with CO (1 atm), two acyl complexes,
[Pd(COCH3)(L)(R)-BINAP]‚[B{3,5-(CF3)2C6H3}4] in which L
is either CH3CN or CO, arose in an about 1:1 ratio. The reaction
rate was comparable to that of the corresponding (R,S)-
BINAPHOS complex, that is, 80% conversion with (R)-BINAP
and 100% with (R,S)-BINAPHOS after 1 h at 20 °C. In
constrast, the subsequent reaction with propene was much slower
with (R)-BINAP than with (R,S)-BINAPHOS, that is, less than
10% conversion with (R)-BINAP after 20 h at 20 °C and 100%
with (R,S)-BINAPHOS after 3 h at 0 °C.
of 4 with CO was carried out in CDCl3 under a CO pressure of
20 atm in an autoclave. The conversion was measured by
1
release-repressurize of CO to measure H NMR under 1 atm
On the other hand, the corresponding cationic complex was
not formed when a symmetrical bis(triaryl phosphite), (R)-
BISPHOSPHITE, was the ligand. Addition of NaB{3,5-
(CF3)2C6H3}4 to Pd(CH3)(Cl){(R)-BISPHOSPHITE} at 20 °C
afforded a complex mixture with which no polymerization
proceeded. Recently, we reported that a phosphine-phosphinite
ligand, (R)-BIPNITE (Chart 5), is as efficient as (R,S)-
BINAPHOS in Rh(I)-catalyzed asymmetric hydroformylation
of 4-vinyl-â-lactams.24 In the present copolymerization, how-
ever, much lower activity was observed with (R)-BIPNITE.
(R,R)-BINAPHOS, a diastereomer of (R,S)-BINAPHOS, did not
form a stable cationic complex, [Pd(CH3)(CH3CN){(R,R)-
BINAPHOS}]‚{B{3,5-(CF3)2C6H3}4]. This may be attributed
to the less stable coordination of the (R,R) isomer to the Pd
center. In our previous study,25 (R,R)-BINAPHOS did not form
a stable apical-equatorial coordination to trigonal bipyamidal
Rh(I) while (R,S)-BINAPHOS did. We explained this tendency
that (R,R)-BINAPHOS requires a P-Rh-P angle much larger
than 90° due to the steric repulsion of the two binaphthyl groups.
This also accounts for the present result.
of CO at -23 °C. The process was also first-order in the
concentration of the palladium complex, and kobs was 1.20 ×
10-4 (∆Gq ) 19.0 kcal/mol) at -23 °C and 1.16 × 10-4 (∆Gq
) 18.9 kcal/mol) at -12 °C. For the phenanthroline system,
∆Gq ) 14.9 kcal/mol (at -66°C) is reported for the CO insertion
of Pd(CH2CH2COCH3)(CO)(phen).5b Because the two steps 2
f 4 and 4 f 5 show such close ∆Gq, we cannot determine the
rate-determining step from those results. Nevertheless, the
present study provides an estimation of the activation energy
(∆Gq) for the two steps. The characteristic feature of the present
phosphine-phosphite system is the retarded CO insertion in
comparison to Brookhart’s 1,10-phenanthroline system.
3.e. The Chain-Terminating Step. The chain-terminating
step is supposed to be the â-hydride elimination from an
alkylpalladium species such as 4 and 6.1 The Pd-H bond
formation trans to the phosphite may be disfavored since the
σ-bond at this position has been suggested to be less stable.
This may explain the highest molecular weight obtained with
our catalyst although details are not clear yet. The living nature
of the present copolymerization has been verified by demon-
strating a linear relationship between Mn and the degree of
monomer conversion (Figure 7). The results obtained in
sections 1-3 are summarized in Scheme 2.
5. Asymmetric Copolymerization of Other Olefins with
CO. Other olefins were subjected to the present copolymeri-
zation with CO.26 (4-tert-Butylphenyl)ethene gave the corre-
sponding polymer in almost complete isotacticity although both
the Mn and [Φ]D values, 5600 and -429, respectively, were
lower than those reported by Brookhart6g (eq 14).
4. Asymmetric Copolymerization of Propene with CO
Catalyzed by Pd(II) Complexes Bearing Other Ligands. In
addition to (R,S)-BINAPHOS, (R)-BINAP, (R)-BISPHOS-
PHITE, and other related ligands illustrated in Chart 5 were
examined for the asymmetric copolymerization. Neutral com-
plexes of type Pd(CH3)(Cl)(ligand) and cationic complexes of
the [Pd(CH3)(CH3CN)(ligand)]‚[B{3,5-(CF3)2C6H3}4] were pre-
pared in a manner similar to that for (R,S)-BINAPHOS. Their
(24) Nozaki, K.; Wen-ge, Li.; Horiuchi, T.; Takaya, H.; Saito, T.;
Yoshida, A.; Matsumura, K.; Kato, Y.; Imai, T.; Miura, T.; Kumobayashi,
H. J. Org. Chem. 1996, 61, 7658.
(25) Nozaki, K.; Sakai, N.; Nanno, T.; Higashijima, T.; Mano, S.;
Horiuchi, T.; Takaya, H. J. Am. Chem. Soc. 1997, 119, 4413.
(26) Kacker, S.; Jiang, Z.; Sen, A. Macromolecules 1996, 29, 5852. See
also ref 6d.

12790 J. Am. Chem. Soc., Vol. 119, No. 52, 1997
Nozaki et al.
Table 3. Neutral- and Cationic-Complex Formation with Other Ligands and Their Use as Catalysts for Copolymerization of Propylene and
COa
31P NMR
[Pd(CH3)(CH3CN)(L)]‚
L
Pd(CH3)(Cl)(L)
[B{3,5-(CF3)2C6H3}4]
yield (%)
14.6
Mn
Mw/Mn
(R,S)-BINAPHOS
δ 149.4, 12.3 (JP-P ) 64.0, major)
δ 154.2, 33.2 (JP-P ) 56.0, minor)
δ 39.9, 14.6 (JP-P ) 39.7)
δ 119.3, 114.7 (JP-P ) 91.6)
δ 136.7, 14.7 (JP-P ) 35.0)
δ 148.8, 18.4 (JP-P ) 62.6, major) and several
unidentified complexes
δ 142.9, 13.0 (JP-P ) 63.0)
16 400
1.07
(R)-BINAP
(R)-BISPHOSPHITE
(R)-BIPNITE
δ 40.4, 14.0 (JP-P ) 43.5)
complex mixtureb
0
0
d
d
d
d
1.40
d
δ 135.9, 16.4 (JP-P ) 32.8)
5.5c
0
10 200
(R,R)-BINAPHOS
complex mixtureb
d
a The reaction was stopped after 6 h to compare the catalytic activity. b The cationic complex was not obtained as a single species. c The employed
reaction time was 48 h. d Not determined.
of these two diasteromers are close to that of 4a, we tentatively
assign the two diastereomers 16a and 16b as the cis and trans
(or trans and cis) isomers of the two methyl groups on the
palladacycle, respectively. When the mixture was treated with
20 atm of CO pressure for 3.5 h, no reaction was observed on
the alkyl complexes 16a and 16b. Thus, it has been shown
that the CO insertion is the key step which retarded the
copolymerization of internal olefins. The methyl group on the
R-carbon of the palladium center hinders the approach of CO
to the metal center, and accordingly, the double-bond migration
shown in eq 15 is essential for the copolymerization of 2-butene
with CO to take place. In 16a and 16b, the absence of the
Cyclocopolymerization of R,ω-dienes with CO catalyzed by
(R,S)-BINAPHOS-Pd(II) has been reported elsewhere.27 In-
Pd-H elimination-addition process may be the reason for the
ternal olefins such as (E)- and (Z)-2-butene were inert under
poor reactivity with CO. â-Elimination of the polymer chain
the same reaction conditions. This is in sharp contrast to the
is suggested to be the chain-termination step. The absence of
â-elimination in 16a or 16b is consistent with the highest level
of molecular weight achieved by this catalyst system.
report by Sen that 1,3-insertion of (Z)-2-butene was observed
with the Me-DUPHOS-Pd(II) system.6d They reported that the
1,3-insertion proceeds via the Pd-H elimination-addition
process shown in eq 15.
Conclusion
The mechanism of the asymmetric copolymerization of
propene with carbon monoxide catalyzed by Pd(II)-(R,S)-
BINAPHOS has been disclosed as shown in Scheme 2. In the
stepwise observation, all the complexes 1-6 have the carbon
substituents trans to the phosphine site in their major isomers.
This can be explained by the “stabilization influence” at this
position in (R,S)-BINAPHOS complexes. This stabilization
influence trans to the phosphine is characteristic for (R,S)-
BINAPHOS which contradicts the conventional order of the
“trans influence”. Such an influence has been further studied
in comparison with C2 symmetrical bidentate ligands using their
platinum complexes 8-10. In the actual reaction condition,
however, the major complexes of 1-6 may not be responsible
for the copolymerization. We propose that the alkyl group in
2 is activated in the form of 13b prior to its migration to the
coordinated CO. Density functional and ab initio MO calcula-
tions for the model olefin insertion suggested the olefin insertion
via 14b f ts(14b-4b) f 4b (path B). The calculations also
gave the geometrical and energetic evidence of the above-
mentioned stabilization influence, which makes another isomer
of the reactant, 14a, less stable than 14b and thus the reaction
from 14a less favorable. The high regio- and enantioselectivities
are clearly explained by the steric effect of the BINAPHOS
ligand which restricts a transition structure to one specific
isomeric form as shown in the combined MO + MM calcula-
tions. Some of the product 4b isomerizes into the more stable
isomer 4a to give a 1:4 mixture of 4b and 4a. The equilibrium
is so rapid that the 1:4 ratio does not change during the CO
insertion. For the subsequent CO insertion from 4 to 5, a
speculation analogous to that for 2 to 3 is applied. In this
manner, the route 4b f 15 f 5 is suggested. For the
subsequent CO insertion, 6b rather than 6a might be responsible.
In order to clarify the sluggish step of our system, the reaction
was followed stepwise. With (E)- and (Z)-2-butene, the
corresponding alkylpalladium complexes were prepared by the
reaction of the olefins with the acyl complex 3 (eq 16). The
result reveals that internal olefins do insert into 3. It is of interest
that the same 1:1 diastereomeric mixture was obtained from
both (E)- and (Z)-2-butene. Because both of the chemical shifts
(27) (a) Nozaki, K.; Sato, N.; Nakamoto, K.; Takaya, H. Bull. Chem.
Soc. Jpn. 1997, 70, 659. (b) Borkowsky, S. L.; Waymouth, R. M.
Macromolecules 1996, 29, 6377.

Copolymerization of Propene with Carbon Monoxide
J. Am. Chem. Soc., Vol. 119, No. 52, 1997 12791
Preparation of Carbon-13-Labeled Cationic Complex (SP-4-3)-
[Pd(13CH3)(CH3CN){(R,S)-BINAPHOS}]‚[B{3,5-(CF3)2C6H3}4] (2-
13CH3). A 5:1 mixture of 1a-13CH3 and 1b-13CH3 was similarly
converted to the corresponding cationic complex 2-13CH3 in the
presence of (SP-4-2)-[Pt(13CH3)(Cl){(R,S)-BINAPHOS}. In CH2Cl2-
CH3CN, the cationic Pt complex (SP-4-3)-[Pt(13CH3)(CH3CN){(R,S)-
BINAPHOS}]‚[B{3,5-(CF3)2C6H3}4] (11-13CH3) gradually precipitated
as a white solid while 2-13CH3 remained in the solution. Filtration
through a pad of Celite and removal of the solvent gave 2-13CH3: JC-P
) 78 Hz (phosphine), ∼0 Hz (phosphite). The larger JC-P value
observed in the phosphine than in the phosphite disclosed that the
phosphine is trans to the methyl group.
Related ligands such as chiral bisphosphine, bisphosphite,
phosphine-phosphinite, and phosphine-phosphite were exam-
ined, but none of these showed higher activity than (R,S)-
BINAPHOS. The copolymerization took place with 4-tert-
butylstyrene but not with 2-butene. This suggests that the
isomerization of 2-butene into a terminal olefin via â-elimination
did not proceed in the alkyl complex 16. This shows good
agreement with the highest level of molecular weight achieved
by this catalyst system.
Throughout the studies, we provide an example of how one
unsymmetrical ligand behaves in a catalytic cycle.
Addition of Chloride Anion to Cationic Complex 2. Tetraethy-
lammonium chloride (5.0 mg, 0.0302 mmol) was added to a solution
of (SP-4-3)-Pd(CH3)(CD3CN){(R,S)-BINAPHOS}‚[B{3,5-(CF3)2C6H3}4]
(2) (0.0145 mmol) in CDCl3 (0.5 mL). The reaction mixture was
analyzed by NMR. After 90 min, 2 was completely converted to 1.
The ratio of 1a and 1b was 8:1 after 90 min, and 5:1 after 24 h.
CO Insertion To Give Acylpalladium Complex (SP-4-3)-[Pd-
(COCH3)(CD3CN){(R,S)-BINAPHOS}]‚[B{3,5-(CF3)2C6H3}4] (3-D).
A solution of 2-D in CDCl3 (0.5 mL) was treated with CO (1 atm) at
20 °C to give 3-D: 1H NMR (CDCl3, 20 °C) δ 2.35 (s, 3H, CH3), 4.94
(d, J ) 8.90 Hz, 1H), 6.08 (d, J ) 8.58 Hz, 1H), 6.71-8.29 (m, 44H);
13C NMR (CDCl3, 20 °C) δ 42.1 (dd, JP-C ) 41, 48 Hz), 117.4-135.9
(m), 139.1 (d, J ) 17 Hz), 145.6, 146.2 (d, J ) 17 Hz), 146.2 (d, J )
12 Hz), 161.7 (q, J ) 50 Hz), 222.4 (dd, JP-C ) 21, 94 Hz); 31P NMR
(CDCl3, 20 °C) δ 10.3 (d, JP-P ) 113 Hz), 133.8 (d, JP-P ) 113 Hz);
19F NMR (CDCl3, 20 °C) δ -5.91; IR (CDCl3) 2254, 1724 cm-1 (νCO).
(SP-4-3)-[Pd(13COCH3)(CD3CN){(R,S)-BINAPHOS}]‚B{3,5-
(CF3)2C6H3}4 (3-D-13CO) was similarly prepared from 2-D and 13CO.
Data for 3-D-13C: JP-C ) 94 Hz (phosphine), 20 Hz (phosphite).
Propene Insertion To Afford Alkyl Complex [Pd(CH2CH(CH3)-
COCH3){(R,S)-BINAPHOS}]‚B(3,5-(CF3)2C6H3)4 (4). A solution of
3-D in CDCl3 (0.5 mL) was cooled to -25 °C and treated with propene
(1 atm) for 14 h. The solution was concentrated at -25 °C to give a
4:1 mixture of (SP-4-3)-4 (4a) and (SP-4-4)-4 (4b). Data for 4a: 1H
NMR (CDCl3, -25 °C) δ 1.41 (d, J ) 7.26 Hz, 3H, CHCH3), 2.10 (s,
3H, C(O)CH3), 2.34-2.45 (br, 1H, CHH), 2.75-2.89 (br, 1H, CHH),
2.98-3.15 (br, 1H, CHCH3), 4.77 (d, J ) 8.91 Hz, 1H), 6.02 (d, J )
8.57 Hz, 1H), 6.67-8.30 (m, 44H); 13C NMR (CDCl3, -25 °C) δ 20.3
(d, JP-C ) 4 Hz) (CH3), 27.1 (CH3), 39.6 (dd, JP-C ) 6, 79 Hz) (CH2),
55.8 (d, JP-C ) 5 Hz) (CH), 117.4-134.9 (m), 139.1 (d, J ) 18 Hz),
Experimental Section
Preparation of Neutral Complex Pd(CH3)(Cl){(R,S)-BINA-
PHOS} (1). A benzene (10.0 mL) solution of (R,S)-BINAPHOS (0.360
g, 0.468 mmol) was added to a solution of Pd(CH3)(Cl)(COD)10 (0.124
g, 0.468 mmol) in benzene (10.0 mL). The solution was stirred at 20
°C for 2 h, and the solution was concentrated. The crude product was
reprecipitated from CH2Cl2 by addition of ether to give a 5:1 mixture
of (SP-4-2)-1 (1a) and (SP-4-3)-1 (1b) (0.371 g, 86% yield). Data for
1a: 1H NMR (CDCl3, 20 °C) δ 1.50 (dd, JP-H ) 2.6, 8.3 Hz, 3H,
CH3), 4.72 (d, J ) 8.9 Hz, 1H of the ligand), 5.88 (d, J ) 8.6 Hz, 1H
of the ligand), 6.55-8.20 (m, 32H); 13C NMR (CDCl3, 20 °C) δ 9.30
(dd, J ) 5, 92 Hz), 120.5-139.0 (m), 146.9-147.7 (m); 31P NMR
(CDCl3, 20 °C) δ 12.3 (phosphine, d, JP-P ) 64 Hz), 149.4 (phosphite,
d, JP-P ) 64 Hz); with partial decoupling of the aromatic protons, 3JP-H
) 8 Hz (phosphine), ∼0 Hz (phosphite). Data for 1b: 1H NMR
(CDCl3, 20 °C) δ 0.51 (dd, JP-H ) 4.0, 11.2 Hz, CH3), 4.68 (d, J )
11.2 Hz, 1H of the ligand), 5.84 (d, J ) 9.9 Hz, 1H of the ligand),
6.55-8.20 (m, 32H); 31P NMR (CDCl3, 20 °C) δ 33.2 (phosphine, d,
JP-P ) 56 Hz), 154.2 (phosphite, d, JP-P ) 56 Hz); with partial
3
decoupling of aromatic protons, JP-H ) ∼0 Hz (phosphine), 11 Hz
3
(phosphite). Comparison of the JP-H values observed for CH3 and
the phosphorus atoms of (R,S)-BINAPHOS suggests that the methyl
group is placed trans to the phosphine site in 1a and trans to the
phosphite in 1b. Anal. Calcd for C53H37ClO3P2Pd: C, 68.77; H, 4.03.
Found: C, 68.56; H, 3.96.
Preparation of Carbon-13-Labeled Neutral Complex Pd(13CH3)-
(Cl){(R,S)-BINAPHOS} (1-13CH3). To a mixture of PdCl2(PhCN)2
(20.0 mg, 0.0520 mmol) and Pt(13CH3)2(COD) (17.6 mg, 0.0520 mmol)
was added dichloromethane (2 mL), and the resulting mixture was
stirred at room temperature for 15 min.11 Addition of (R,S)-BINAPHOS
(81.1 mg, 0.105 mmol) followed by filtration and removal of the solvent
in Vacuo afforded a 5:1 mixture of 1a-13CH3 and 1b-13CH3 (120.1 mg).
Data for 1a-13CH3: JC-P ) 92 Hz (phosphine), 5 Hz (phosphite). Data
for 1b-13CH3: JC-P ) ∼0 Hz (phosphine), 150 Hz (phosphite).
Comparison of the 2JC-P values also suggests that the methyl group is
placed trans to the phosphine site in 1a and trans to the phosphite in
1b. A platinum complex, (SP-4-2)-Pt(13CH3)(Cl){(R,S)-BINAPHOS},
was produced simultaneously: 1H NMR (CDCl3, 20 °C) δ 1.30 (tdd,
JPt-H ) 60.0 Hz, JP-H ) 7.8, 3.4 Hz, CH3), 4.66 (d, J ) 8.9 Hz, 1H of
the ligand), 5.86 (d, J ) 8.6 Hz, 1H of the ligand), 6.55-8.20 (m,
146.1, 146.6 (d, J ) 11 Hz), 161.7 (q, J ) 50 Hz), 240.3 (dd, JP-C
)
2, 10 Hz); 31P NMR (CDCl3, -25 °C) δ 13.0 (d, JP-P ) 66 Hz), 142.4
(d, JP-P ) 66 Hz). Data for 4b: 31P NMR (CDCl3, -25 °C) δ 30.3
(d, JP-P ) 66 Hz), 150.3 (d, JP-P ) 66 Hz). Data for the mixture: 19F
NMR (CDCl3, -25 °C) δ -5.99; IR (CDCl3) 1710 (CO) cm-1. The
carbon-13-labeled [Pd((R,S)-BINAPHOS)(CH2CH(CH3)13COCH3)]‚
[B(3,5-(CF3)2C6H3)4] (4-13CO) was similarly prepared from 13CO to
give a 4:1 mixture of 4a-13CO and 4b-13CO. Data for 4a-13CO: 1H
NMR (CDCl3, -25 °C) δ 1.35-1.50 (m, 3H, CHCH3), 2.10 (d, J )
5.61 Hz, 3H, C(O)CH3), 2.34-2.45 (br, 1H, CHH), 2.75-2.89 (br,
1H, CHH), 2.98-3.15 (br, 1H, CHCH3), 4.77 (d, J ) 8.91 Hz, 1H),
6.02 (d, J ) 8.57 Hz, 1H), 6.67-8.30 (m, 44H); 13C NMR (CDCl3,
-25 °C) δ 117.4-134.9 (m), 139.1 (d, J ) 18 Hz), 146.1, 146.6 (d, J
) 11 Hz), 161.7 (q, J ) 50 Hz), 240.2 (dd, JP-C ) 2, 10 Hz); 31P
NMR (CDCl3, -25 °C) δ 13.0 (dd, JP-C ) 10, JP-P ) 66 Hz), 142.4
(d, JP-P ) 66 Hz); 19F NMR (CDCl3, -25 °C) δ -5.99. Data for
4b-13CO: 31P NMR (CDCl3, -25 °C) δ 30.3 (d, JP-P ) 66 Hz), 150.3
(dd, JP-C ) 11 JP-P ) 66 Hz).
32H); 13C NMR (CDCl3, 20 °C) δ 3.6 (tdd, JPt-C ) 459 Hz, JC-P
)
89, 6 Hz), 120.5-139.0 (m), 146.9-147.7 (m); 31P NMR (CDCl3, 20
°C) δ 23.0 (phosphine, tdd, JPt-P ) 1402 Hz, JP-P ) 26 Hz, JC-P ) 89
Hz), 115.2 (phosphite, tdd, JPt-P ) 7562 Hz, JP-P ) 26 Hz, JC-P ) 6
2
Hz). Comparison of the JC-P values suggests that the methyl group
is placed trans to the phosphine site in this complex.
Preparation of Cationic Complex (SP-4-3)-[Pd(CH3)(CD3CN)-
{(R,S)-BINAPHOS}]‚[B{3,5-(CF3)2C6H3}4] (2-D). To a 5:1 mixture
of 1a and 1b (50.8 mg, 0.0549 mmol) in CH2Cl2 (3.0 mL) was added
a solution of NaB(3,5-(CF3)2C6H3)4 (48.6 mg, 0.0549 mmol) in CD3-
CN (2.0 mL). The solution was stirred at 20 °C for 2 h, and
Preparation of 1-Dodecene-1-13C. First, 13CH3PPh3I was prepared.
To the solution of triphenylphosphine (1.68 g, 6.41 mmol) in toluene
(20 mL) was added iodomethane-13C (0.400 mL, 6.38 mmol). White
precipitate formed, and the suspension was stirred at room temperature
for 3 h. The white precipitates were collected by filtration, washed
with hexane, and dried to give 1-dodecene-1-13C in 51% yield (1.31
g): 1H NMR (CDCl3) δ 3.25 (dd, JC-H ) 135 Hz, JP-H ) 13.2 Hz,
3H), 7.6-7.9 (m, 15H). 31P NMR (CDCl3) δ 22.4 (d, JC-P ) 58 Hz).
To a suspension of 13CH3PPh3I (1.29 g, 3.18 mmol) in diethyl ether
(10 mL) was added potassium tert-butoxide (425.6 mg, 3.79 mmol) at
0 °C. The suspension turned yellow and was stirred for 10 min.
Undecanal (0.7 mL, 3.39 mmol) was added, and the whole was stirred
concentrated to give 2-D: 1H NMR (CDCl3, 20 °C) δ 1.29 (d, JP-H
)
7.6 Hz, CH3Pd), 4.74 (d, J ) 8.9 Hz, 1H of the ligand), 5.98 (d, J )
8.3 Hz, 1H of the ligand), 6.66-8.27 (m, 44H); 13C NMR (CDCl3, 20
°C) δ 6.27 (dd, JP-C ) 5, 78 Hz), 117.4-135.1 (m), 139.0 (d, J ) 16
Hz), 146.1, 146.7 (d, J ) 28 Hz), 146.8 (d, J ) 24 Hz), 161.7 (q, J )
50 Hz); 31P NMR (CDCl3, 20 °C) δ 13.1 (d, JP-P ) 64 Hz), 143.1 (d,
JP-P ) 64 Hz).

12792 J. Am. Chem. Soc., Vol. 119, No. 52, 1997
Nozaki et al.
at room temperature for 20 h. Water was added to the reaction mixture,
and the aqueous layer was extracted with diethyl ether. The combined
organic layer was dried with magnesium sulfate and concentrated under
reduced pressure. The crude product was purified by silica-gel column
chromatography (hexane) and distilled under reduced pressure to give
a colorless liquid (427.0 mg, 79%): 13C NMR (CDCl3) δ 114.1.
Preparation of Alkylpalladium Complex 7 from 1-Dodecene-1-
13C. Neutral complex 1 (19.3 mg, 0.0208 mmol) and Na[B(C6H3-3,5-
(CF3)2)4] (19.2 mg, 0.0217 mmol) were mixed in dichloromethane (2
mL) and acetonitrile (0.2 mL) at room temperature for 1 h. After
removal of the solvent, the residue was dissolved in dichloromethane
(2 mL), and to this was introduced carbon monoxide (1 atm). After 2
h, 1-dodecene-1-13C (50 µL, 0.22 mmol) was added to the reaction
mixture and stirred at room temperature for 2 h. The resulting solution
was then concentrated and dried in Vacuo to give a 4:1 mixture of 7a
and 7b as a red solid. Data for 7a: 31P NMR (CDCl3) δ 13.7 (dd,
JC-P ) 80 Hz, JP-P ) 67 Hz), 141.1 (dd, JC-P ) 8 Hz, JP-P ) 67 Hz);
13C NMR (CDCl3) δ 37.0 (dd, JP-C ) 80, 8 Hz). Data for 7b: 31P
NMR (CDCl3) δ 29.5 (dd, JC-P ) 5 Hz, JP-P ) 64 Hz), 150.0 (dd,
117.0-135.1 (m), 138.7 (d, J ) 17 Hz), 145.7-146.5 (m), 161.5 (q,
J ) 50 Hz), 210.3, 223.3 (dd, JP-C ) 21, 100 Hz); 31P NMR (CDCl3,
-25 °C) δ 9.7 (d, JP-P ) 108 Hz), 137.0 (d, JP-P ) 108 Hz); 19F
NMR (CDCl3, -25 °C) δ -5.71; IR (CDCl3) 1716, 1612 (νCO) cm-1
.
Propene Insertion To Give the Second-Generation Alkylpalla-
dium Complex [Pd(CH2CH(CH3)COCH2CH(CH3)COCH3){(R,S)-
BINAPHOS}]‚[B{3,5-(CF3)2C6H3}4] (6). A solution of 5-D in CDCl3
(0.5 mL) and CD3CN (0.03 mL) was treated with propene (1 atm) at
-25 °C for 48 h. The solution was concentrated at -25 °C to give
(SP-4-3)-6 (6a) as a single product. Data for 6a: 1H NMR (CDCl3,
-25 °C) δ 0.85 (d, J ) 7.26 Hz, 3H, CHCH3COCH3), 1.39 (s, 3H,
C(O)CH3), 1.58 (d, J ) 7.26 Hz, 3H, CHCH3COCH2), 2.07-2.32 (m,
2H, CHCH3COCH3 and COCHH), 2.35-2.50 (m, 1H, PdCHH), 2.69-
2.80 (m, 1H, PdCHH), 3.06-3.30 (m, 2H, COCHH and CHCH3-
COCH2), 4.65 (d, J ) 8.90 Hz, 1H), 5.99 (d, J ) 8.58 Hz, 1H), 6.68-
8.29 (m, 44H); 13C NMR (CDCl3, -25 °C) δ 15.9, 21.5, 27.1, 40.5
(dd, JP-C ) 7, 93 Hz), 41.3, 41.3, 55.4 (d, JP-C ) 6 Hz), 117.0-136.2
(m), 138.7 (d, J ) 16 Hz), 146.0, 146.3 (d, J ) 11 Hz), 161.5 (q, J )
50 Hz), 209.0, 240.4 (dd, JP-C ) 4, 9 Hz); 31P NMR (CDCl3, -25 °C)
δ 13.1 (d, JP-P ) 67 Hz), 142.0 (d, JP-P ) 67 Hz); 19F NMR (CDCl3,
JC-P ) 128 Hz, JP-P ) 66 Hz); 13C NMR (CDCl3) δ 46.8 (dd, JP-C
)
-25 °C) δ -5.74; IR (CDCl3) 1712, 1641 (νCO) cm-1
.
128, 5 Hz). Comparison of JC-P values disclosed the structures of 7a
and 7b.
Preparation of Pt(13CH3)2(COD). At first, 13CH3MgI was prepared.
To a mixture of magnesium turnings (103.3 mg, 4.25 mmol) in ether
(5 mL) was added dropwise 13CH3I in ether (0.70 M solution, 5 mL,
3.5 mmol). The solution turned brown. After the solution was stirred
for 2 h, the supernatant was transferred to a Schlenk tube and stored
under argon. The corresponding Grignard reagent 13CH3MgI (0.17 M)
was obtained. To a yellow suspension of PtI2(COD) (147.5 mg, 0.26
mmol) in ether (5 mL) was added 13CH3MgI in ether (0.17 M, 3 mL,
0.51 mmol) dropwise at 0 °C. The suspension turned to a clear solution.
After being stirred at 0 °C for 1.5 h, the solution was quenched with
saturated NH4Cl(aq) and then extracted with ether (3 × 10 mL). The
ether layer was collected and evaporated under reduced pressure. The
crude product was purified by silica-gel colunm chromatography (CH2-
Cl2/hexane ) 1/2) to give a white solid (125.5 mg, >99% yield): 1H
NMR (CDCl3) δ 0.74 (JPt-H ) 82.2 Hz, JC-H ) 128, 1.7 Hz), 2.2-2.4
(m), 4.7-4.9 (m).
Carbomethoxylation of the Alkylpalladium Complex 4. To a 5:1
mixture of 1a and 1b (81.2 mg, 0.0877 mmol) in CH2Cl2 (2.5 mL)
was added a solution of NaB(3,5-(CF3)2C6H3)4 (77.7 mg, 0.0877 mmol)
in CH3CN (0.5 mL). The solution was stirred at 20 °C for 1 h, and
the solution was concentrated. The resulting crude 2 was then dissolved
with CHCl3, and the solution was degassed by three cycles of freeze-
thaw. After the solution was treated with CO (1 atm) for 0.5 h, the
solution was cooled to -25 °C and treated with propene (1 atm) for
21 h. The solvents were evaporated at -25 °C to give 4 in situ. A
solution of this crude 4 in CHCl3 (2.0 mL) and CH3OH (1.0 mL) was
treated with CO (20 atm) at -25 °C for 4 days. After release of CO
pressure, the solvent was evaporated and the residue was purified by
distillation (80 °C, 1.0 Torr) to give methyl (S)-2-methyl-4-oxopen-
tanoate in 68% total yield from 1. The enantiomeric excess of the
product (95% ee) was determined by GLC with a chiral column
(Chrompack CP-cyclodextrin-â-236M-19, 100 °C). Data for methyl
(S)-2-methyl-4-oxopentanoate: 1H NMR (CDCl3) δ 1.43 (d, J ) 7.26
Hz, 3H, CHCH3), 2.22 (s, 3H, C(O)CH3), 2.30 (dd, J ) 16.83, 5.61
Hz, 1H, CHH), 2.77 (dd, J ) 16.83, 8.58 Hz, 1H, CHH), 2.95-3.06
(m, 1H, CHCH3), 3.66 (s, 3H, COOCH3). The absolute configuration
was determined to be S by comparison of the optical rotation value
Preparation of Pt(CH3)2{(R,S)-BINAPHOS} (8 and 8-13CH3). To
a mixture of Pt(CH3)2(COD) (52.2 mg, 0.16 mmol) and (R,S)-
BINAPHOS (119.8 mg, 0.16 mmol) was added dichloromethane (5
mL), and the solution was stirred at room temperature for 1.5 h. The
reaction mixture was concentrated and dried in Vacuo. Purification
by silica-gel column chromatography (CH2Cl2/hexane ) 1/1) gave the
product (89.3 mg, 57% yield). The carbon-13-labeled 8-13CH3 was
similarly prepared from Pt(13CH3)2(COD). Data for 8: 1H NMR
(CDCl3) δ 0.00 (JPt-H ) 68.3 Hz, CH3 trans to phosphite), 1.00 (JPt-H
) 74.9 Hz, CH3 trans to phosphine), 4.70 (d, J ) 8.9 Hz, 1H of
aromatic), 5.88 (d, J ) 8.3 Hz, 1H of aromatic), 6.5-8.2 (m, aromatic);
31P NMR (CDCl3) δ 21.4 (JPt-P ) 1602 Hz, JP-P ) 18 Hz, phosphine),
159.0 (JPt-P ) 3392 Hz, JP-P ) 18 Hz, phosphite). Anal. Calcd for
C54H40O3P2Pt: C, 65.25; H, 4.06. Found: C, 65.14; H, 3.85. From
8-13CH3, the following data were given: 31P NMR JC-P ) 90, 8 Hz
with the literature data.28 [R]25 ) -54° (c 0.42, THF).
D
Crystal Data for 4a. A single crystal of 4a (C89H65BF24O4P2Pd,
M ) 1833.70) was obtained from hexane/dichloromethane solution.
Data were collected on a Mac Science MXC18 diffractometer (Cu KR,
λ ) 1.541 78 Å), 2θmax ) 60°. Three standard reflections were
measured every 100 reflections at 220 K. Of the 9248 reflections which
were collected, 8068 were unique (Rint ) 0.031). The crystal structure
was solved by direct methods (SIR92). The non-hydrogen atoms were
refined anisotropically. Hydrogen atoms were placed in fixed calculated
positions. The final cycle of full-matrix least-squares refinement was
based on 6986 observed reflections (I > 3.00σ) and 1144 variable
parameters and converged with unweighted and weighted agreement
factors of R ) ∑||Fo| - |Fc||/∑|Fo| ) 0.0901 and Rw ) (∑w(|Fo| -
(phosphine), 155, 9 Hz (phosphite); 13C NMR (CDCl3) δ 0.1 (JPt-C
)
588 Hz, JP-C ) 90 Hz, JC-C ) 8.6 Hz, trans to phosphine), 6.7 (JPt-C
) 481 Hz, JP-C ) 155 Hz, JC-C ) 8.6 Hz, trans to phosphite).
Assignment of the carbon and proton peaks to carbons and protons
either cis or trans to the phosphine or the phosphite was carried out by
the following procedure. For the phosphorus nucleus at δ 21.4
(phosphine) by 31P NMR, the larger JC-P (90 Hz) should arise from
the coupling with the carbon at its trans nucleus. This JC-P value is
seen in the 13C NMR peak at δ 0.1. Similarly, JC-P of 155 Hz is
common for the phosphite peak at δ 159.0 (31P NMR) and the 13C
NMR peak at δ 6.7. Measurement of 13C NMR with decoupling of
individual proton peaks revealed that the proton at δ 0.00 (1H NMR)
is attached to the carbon at δ 6.7 (13C NMR) and that the proton at δ
1.00 (1H NMR) is attached to the carbon at δ 0.1 (13C NMR).
Crystal Data for 8. A single crystal of 8 (C54H40O3P2Pt, M )
993.94) was obtained from hexane/ethyl acetate solution. Data were
collected on a Mac Science MXC18 diffractometer (Cu KR, λ )
1.541 78 Å), 2θmax ) 60°. Three standard reflections were measured
every 100 reflections at 297 K. The unique 5167 reflections with I >
3σ(I) were considered as observed. The crystal structure was solved
2
|Fc|)2/∑wFo )0.5 ) 0.0964. Orthorhombic, space group P212121, a )
20.9034(4) Å, b ) 29.9506(7) Å, c ) 12.7725(9) Å, V ) 7796.45(6)
Å3, dcalcd ) 1.50 g/cm3, Z ) 4. Details are given in the Supporting
Information.
CO Insertion To Give the Second-Generation Acylpalladium
Complex (SP-4-3)-[Pd(COCH2CH(CH3)COCH3)(CD3CN){(R,S)-
BINAPHOS}]‚[B{3,5-(CF3)2C6H3}4] (5-D). A 4:1 mixture of 4a and
4b in CDCl3 (0.5 mL) and CD3CN (0.03 mL) was treated with CO (20
atm) at -25 °C to give 5-D: 1H NMR (CDCl3, -25 °C) δ 1.19 (d, J
) 6.93 Hz, 3H, CHCH3), 2.12 (s, 3H, COCH3), 2.94 (br d, J ) 17.81
Hz, 1H), 3.19-3.38 (m, 1H), 3.41-3.64 (m, 1H), 4.58 (d, J ) 9.24
Hz, 1H), 5.87 (d, J ) 8.57 Hz, 1H), 6.58-8.21 (m, 44H); 13C NMR
(CDCl3, -25 °C) δ 16.3, 28.5, 42.8, 56.6 (dd, JP-C ) 36, 74 Hz),
(28) (a) Blanco, L.; Rousseau, G.; Barnier, J.-P.; Guibe´-Jampel, E.
Tetrahedron: Asymmetry 1993, 4, 783. (b) Enders, D.; Gerdes, P.;
Kipphardt, H. Angew. Chem., Int. Ed. Engl. 1990, 29, 179.

Copolymerization of Propene with Carbon Monoxide
J. Am. Chem. Soc., Vol. 119, No. 52, 1997 12793
by the direct method (SIR 92). The non-hydrogen atoms were refined
anisotropically. Hydrogen atoms were placed in fixed calculated
positions. A series of standard full-matrix least-squares refinement and
Fourier synthesis revealed the remaining atoms. All calculations were
performed using the Crystan crystallographic software package. Current
residual values for 603 parameters are R ) 0.0594 and Rw ) 0.0712.
Orthorhombic, space group P212121, FW ) 993.94 a ) 17.2570 (25)
Å, b ) 28.9158 (45) Å, c ) 9.9413 (22) Å, V ) 4960.68 (152) Å3,
dcalcd ) 1.663 g/cm3, Z ) 5. Details are given in the Supporting
Information.
Preparation of PtCl2{(R)-BISPHOSPHITE}. A mixture of (R)-
BISPHOSPHITE (110.3 mg, 0.15 mmol) and PtCl2(COD) (56.3 mg,
0.15 mmol) was stirred in dichloromethane (5 mL) at room temperature
for 18 h, and then the resulting clear solution was concentrated and
dried in Vacuo to give a colorless solid (147.1 mg, 99% yield): 1H
NMR (CDCl3) δ 6.28 (d, J ) 7.9 Hz), 6.7-8.0 (m); 31P NMR (CDCl3)
δ 61.9 (JPt-P ) 5672 Hz). Anal. Calcd for C44H32Cl2O6P2Pt: C, 53.67;
H, 3.28. Found: C, 53.51; H, 3.30.
Preparation of Cationic Complex (SP-4-3)-[Pt(CH3)(CH3CN)-
{(R,S)-BINAPHOS}]‚[B{3,5-(CF3)2C6H3}4] (11). A mixture of Pt-
(CH3)Cl((R,S)-BINAPHOS) (53.6 mg, 0.053 mmol) and NaB[3,5-
(CF3)2C6H3]4 (46.6 mg, 0.053 mmol) was dissolved in dichloromethane
(10 mL) and acetonitrile (0.2 mL, 3.8 mmol) and stirred at room
temperature for 5 h. The reaction mixture was concentrated and dried
in Vacuo to give 11 as a colorless solid (99.7 mg, 99% yield): 1H
NMR (CDCl3) 1.12 (br t, JPt-H ) 61.0 Hz, CH3Pt), 1.72 (s, CH3CN);
31P NMR (CDCl3) δ 22.9 (phosphine, JPt-P ) 1476 Hz, JP-P ) 29
Hz), 103.5 (phosphite, JPt-P ) 7935 Hz, JP-P ) 29 Hz).
Replacement of CH3CN of 11 by CO To Give (SP-4-3)-[Pt(CH3)-
(CO){(R,S)-BINAPHOS}]‚[B{3,5-(CF3)2C6H3}4] (12a). To a mixture
of Pt(CH3)Cl{(R,S)-BINAPHOS} (60.4 mg, 0.0596 mmol) and Na-
[B{3,5-(CF3)2C6H3}4] (53.9 mg, 0.0608 mmol) was added dichlo-
romethane saturated with CO (5 mL). After being stirred at room
temperature for 4 h, the reaction mixture was concentrated and dried
in Vacuo to give a white solid. Data for 12a: 31P NMR (CDCl3) δ
15.4 (JPt-P ) 1567 Hz, JP-P ) 37 Hz), 121.8 (JPt-P ) 6023 Hz, JP-P
) 37 Hz); 1H NMR (CDCl3) δ 1.45 (JPt-H ) 55.0 Hz, JP-H ) 6.3 Hz),
4.74 (d, J ) 8.9 Hz), 5.89 (d, J ) 8.6 Hz), 6.6-8.4 (m); IR (CHCl3)
2132 cm-1 (νCO).
Preparation of Pt(CH3)2{(R)-BINAP} (9 and 9-13CH3). To a
mixture of Pt(CH3)2(COD) (25.6 mg, 0.077 mmol) and (R)-BINAP (49.3
mg, 0.077 mmol) was added chloroform (5 mL), and the solution was
stirred at 40 °C for 20 h. The reaction mixture was concentrated and
dried in Vacuo. Purification by silica-gel column chromatography (CH2-
Cl2/hexane ) 1/2) gave the product (47.9 mg, 73% yield): 1H NMR
(CDCl3) δ 0.34 (JPt-H ) 69.3 Hz), 6.6-7.7 (m); 31P NMR (CDCl3) δ
24.1 (JPt-P ) 1874 Hz). Anal. Calcd for C46H38P2Pt: C, 65.17; H,
4.52. Found: C, 64.31; H, 4.29. Data for 9-13CH3: 31P NMR (CDCl3)
δ 24.3 (J
) 1755 Hz, J
) 96, 17 Hz). 13C NMR (CDCl3) δ
Pt-P
C-P
11.7 (JPt-C ) 493 Hz, JP-C ) 95, 5 Hz).
Preparation of Pt(CH3)2{(R)-BISPHOSPHITE} (10 and 10-
13CH3). To a solution of (R)-BISPHOSPHITE (94.9 mg, 0.13 mmol)
in dichloromethane (5 mL) was added Pt(CH3)2(COD) (42.5 mg, 0.13
mmol), and the solution was stirred at room temperature for 2.5 h.
The clear solution was concentrated and dried in Vacuo to give a
colorless solid (130.1 mg, 99% yield): 1H NMR (CDCl3) δ 0.41 (JPt-H
) 70.6 Hz), 6.26 (d, J ) 8.3 Hz), 6.8-7.9 (m); 31P NMR (CDCl3) δ
125.4 (JPt-P ) 3043 Hz). Anal. Calcd for C46H38O6P2Pt: C, 58.54;
H, 4.06. Found: C, 58.48; H, 4.08. Data for 10-13CH3: 31P NMR
(CDCl3) δ 125.5 (JPt-P ) 3041 Hz, JC-P ) 154, 24 Hz); 13C NMR
(CDCl3) δ 2.6 (JPt-C ) 574 Hz, JP-C ) 153, 24 Hz).
Isomerization of Cationic Carbonyl Complex 12a to 12b.
A
solution of 12a (37.5 mg) in dichloromethane (10 mL) was placed in
an autoclave. Carbon monoxide was introduced (20 atm), and the
mixture was stirred at room temperature for 20 h. After the pressure
was released, the reaction mixture was transferred to a Schlenk tube
and concentrated and dried in Vacuo. A white solid was obtained. Data
for 12b: 31P NMR (CDCl3) δ 14.9 (JPt-P ) 3265 Hz, JP-P ) 40 Hz),
148.3 (JPt-P ) 3029 Hz, JP-P ) 40 Hz); 1H NMR (CDCl3) δ 0.53 (JPt-H
) 54.8 Hz, JP-H ) 8.2, 6.6 Hz), 4.80 (d J ) 8.9 Hz), 5.94 (d J ) 8.6
Hz), 6.6-8.4 (m); IR (CHCl3) 2128 cm-1 (νCO). CO insertion to give
the corresponding acylplatinum did not proceed.
Observation of the Rapid Equilibrium between 4a and 4b. A
4:1 mixture of 4a and 4b in CDCl3 (0.5 mL) and CD3CN (0.1 mL)
was treated with CO (1 atm) at 20 °C, and the reaction was followed
by 1H and 31P NMR. After 2 h, 30% of 4 was converted to 5-D, without
changing the isomeric ratio of 4a and 4b.
Computational Details for the Process 3 f 14 f 4. The
geometries of transition states, ts(14a′-4a′) and ts(14b′-4b′) as well as
reactants 14a′ and 14b′ and products 4a′ and 4b′ were fully optimized
using the hybrid density functional method B3LYP with the LANL2DZ
basis set, called basis set I in this paper, which includes for the Pd
atom an effective core potential replacing core electrons up to 3d and
double ú valence basis functions (8s6p4d)/[3s3p2d] by Hay and Wadt29
and for the ligand atoms the Huzinaga-Dunning valence double ú basis
functions.30 In order to obtain more reliable energetics, we carried out
energy calculations by the B3LYP method and by the Møller-Plesset
perturbation theory at the second (MP2) up to the fourth order without
triple substitution (MP4SDQ). In these energy calculations we used
better basis set, called basis set II, in which the triple ú contraction of
Pd d orbitals was adopted and the single set of f orbitals with an
exponent of 1.472 was augmented.31 For the ligand atoms were used
6-31G(d,p) basis functions.32 In the calculation of the zero-point energy
and thermal correction to a potential energy, the vibrational frequencies
were scaled with a factor of 0.9614.33 All the ab initio molecular orbital
Preparation of Pt(CH3)Cl{(R,S)-BINAPHOS}. Pt(CH3)Cl(COD)
(197.7 mg, 0.56 mmol) and (R,S)-BINAPHOS (431.2 mg, 0.56 mmol)
were dissolved in dichloromethane (20 mL) by stirring at room
temperature for 1.5 h. The resulting clear solution was concentrated
and dried in Vacuo. Recrystallization of the product from CH2Cl2/
Et2O gave a colorless solid (415.9 mg, 73% yield): 1H NMR (CDCl3)
δ 1.30 (JPt-H ) 60.0 Hz, JP-H ) 7.8, 3.4 Hz), 4.66 (d, J ) 8.9 Hz),
5.86 (d, J ) 8.6 Hz), 6.6-8.2 (m); 31P NMR (CDCl3) δ 23.0 (JPt-P
)
1402 Hz, JP-P ) 26 Hz), 115.2 (JPt-P ) 7562 Hz, JP-P ) 26 Hz).
Anal. Calcd for C53H37ClO3P2Pt: C, 62.76; H, 3.68. Found: C, 62.47;
H, 3.68.
Preparation of Pt(CH3)Cl{(R)-BINAP}. Pt(CH3)Cl(COD) (45.1
mg, 0.13 mmol) and (R)-BINAP (82.1 mg, 0.13 mmol) were dissolved
in dichloromethane (1 mL) by stirring at room temperature for 2 h.
The reaction mixture was concentrated and dried in Vacuo. Purification
by silica-gel column chromatography (CH2Cl2) gave the product (108.8
mg, 96% yield): 1H NMR (CDCl3) δ 0.54 (JPt-H ) 54.8 Hz, JP-H
)
7.6, 4.3 Hz), 6.6-7.9 (m); 31P NMR (CDCl3) δ 19.0 (JPt-P ) 4446 Hz,
JP-P ) 15 Hz), 24.4 (JPt-P ) 1753 Hz, JP-P ) 15 Hz). Anal. Calcd
for C45H35ClP2Pt: C, 62.25; H, 4.06. Found: C, 62.38; H, 4.31.
Preparation of Pt(CH3)Cl{(R)-BISPHOSPHITE}. To a solution
of (R)-BISPHOSPHITE (44.7 mg, 0.062 mmol) in dichloromethane
(5 mL) was added Pt(CH3)Cl(COD) (22.2 mg, 0.063 mmol), and the
solution was stirred at room temperature for 1 h. The reaction mixture
was concentrated and dried in Vacuo. Purification by silica gel column
chromatography (CH2Cl2) gave the product (54.9 mg, 92% yield): 1H
NMR (CDCl3) δ 0.78 (JPt-H ) 54.8 Hz, JP-H ) 10.6, 3.3 Hz), 6.17 (d,
J ) 8.3 Hz), 6.40 (d, J ) 8.2 Hz), 6.7-8.0 (m); 31P NMR (CDCl3) δ
81.0 (JPt-P ) 7047 Hz, JP-P ) 37 Hz), 125.0 (JPt-P ) 2682 Hz, JP-P
) 37 Hz). Anal. Calcd for C45H35ClO6P2Pt: C, 56.05; H, 3.66.
Found: C, 56.12; H, 3.75.
Preparation of PtCl2{(R,S)-BINAPHOS}. After a mixture of
PtCl2(COD) (28.7 mg, 0.077 mmol) and (R,S)-BINAPHOS (60.6 mg,
0.079 mmol) in dichloromethane (2 mL) was stirred at room temperature
for 30 min, the resulting clear solution was concentrated and dried in
Vacuo to give a white solid (73.0 mg, 92% yield): 1H NMR (CDCl3)
δ 4.54 (d, J ) 7.9 Hz), 5.83 (d, J ) 8.3 Hz), 6.6-8.6 (m); 31P NMR
(CDCl3) δ 8.2 (JPt-P ) 3243 Hz, JP-P ) 18 Hz), 103.8 (JPt-P ) 6270
Hz, JP-P ) 18 Hz). Anal. Calcd for C52H34Cl2O3P2Pt: C, 60.36; H,
3.31. Found: C, 59.79; H, 3.13.
(29) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
(30) Dunning, T. H. J. Chem. Phys. 1971, 55, 716.
(31) Ehlers, A. W.; Bo¨hme, M.; Dapprich, S.; Gobbi, A.; Ho¨llwarth,
A.; Jonas, V.; Ko¨hler, K. F.; Stegmann, R.; Veldkamp, A.; Frenking G.
Chem. Phys. Lett. 1993, 208, 111.
(32) (a) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972,
56, 2257. (b) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28,
213.
(33) Scott, A. P.; Radom, L. J. Phys Chem. 1996, 100, 16502.

12794 J. Am. Chem. Soc., Vol. 119, No. 52, 1997
Nozaki et al.
and density functional calculations were carried out using the GAUSS-
IAN94 program.34
and hydroxyl groups of P(OH)3, were frozen, which is called the
combined MO + MM method.36 We used the MM3 program in the
calculations.37 The standard MM3 parameters are used when available.
The van der Waals parameters reported by Rappe´ et al. are used for
the Pd atom.38
Kinetic Experiments. (1) Propene Insertion into 3 To Give 4.
The acylpalladium 3 (12.0 mg, 6.8 µmol) was dissolved in CDCl3 (0.6
mL) under Ar in an NMR tube. The solution was degassed by the
freeze-thaw method and then saturated with 1 atm of propene at -78
°C. When the solution became homogeneous, the mixture was warmed
to -10 °C. The appearence of the doublet due to the methyl protons
of 4, originating from propene, was monitored in relation to the
unchanging aromatic protons to determine the first-order rate constant
for the reaction. Spectra were recorded every 15 min in the first 1 h
While 14b′ is more stable than 14a′ at all the levels using basis set
II, the B3LYP/I calculation gave the reverse result that 14a′ is more
stable than 14b′. In addition, the Pd-acyl bond in 14a′ is shorter than
that in 14b′, different from the trend in 8 where the Pt-methyl bond
trans to phosphine is shorter than that trans to phosphite. In order to
evaluate the reliability of the B3LYP/I method, we calculated the proton
affinity of PH3 and P(OH)3, in which the structures were determined
at the B3LYP/I level, and furthermore the energies were calculated at
the B3LYP/II level. The calculations showed that at the B3LYP/I level
the proton affinity of PH3 is 19.8 kcal/mol larger than that of P(OH)3,
whereas at the B3LYP/II//B3LYP/I level it is smaller by 16.3 kcal/
mol. The B3LYP/II geometry determinations did not change the trend.
The larger trans influence of phosphite seems to correlate with this
larger proton affinity of P(OH)3. Use of polarization functions is
indispensable to reproduce the difference in electronic effects between
phosphine and phosphite. Although the B3LYP/I calculations did not
reproduce the difference in bond distances, the calculations of proton
affinity clearly demonstrate that the energy calculations with basis set
II at the B3LYP/I structures are reliable. Similarly at the B3LYP/I
level 4b′ is more stable than 4a′, and the calculations with basis set II
gave the correct results.
The structures of the transition states have several points to be
remarked. At ts(14a′-4a′), the newly forming CC bond (1.862 Å) is
0.06 Å shorter than that of ts(14b′-4b′) (1.922 Å). The Pd-acyl bond
at ts(14a′-4a′) (2.336 Å) is stretched by 0.31 Å, whereas that at ts(14b′-
4b′) is stretched by only 0.23 Å. These results show that ts(14b′-4b′)
is located earlier, consistent with the larger exothermicity of the reaction
of 14b′ to 4b′ at the B3LYP/I level. However, the calculations with
basis set II showed that the reaction of 14a′ to 4a′ is more exothermic,
and thus ts(14a′-4a′) would be located earlier at a higher level of
calculations.
Second, one can notice that one of the hydroxy hydrogens of P(OH)3
at ts(14b′-4b′) is close to the acyl oxygen, its distance of 2.433 Å
suggesting weak hydrogen bonding between them. However, this
hydrogen bonding does not exist in a real complex with the BINAPHOS
ligand, since BINAPHOS does not have hydroxy groups. In order to
assess the stabilization of ts(14b′-4b′) by this hydrogen bonding, we
determined the structure of ts(14b′-4b′), fixing the H-O-P-Pd
dihedral angle to be 178.4° and P(phosphine)-Pd-P-O to be 150.0°,
which are those determined by the integrated molecular orbital plus
molecular mechanics method35 for I. These restrictions destabilized
ts(14b′-4b′) by 2.4 kcal/mol at both B3LYP/II and MP4SDQ/II levels.
The reactant 14b′ is also destabilized by the same restrictions by 0.9
and 0.7 kcal/mol at the B3LYP and MP4SDQ/II levels, respectively,
and consequently the hydrogen bonding decreases the activation energy
for path B by only 1.5-1.7 kcal/mol. Without hydrogen bonding the
activation energy for path B is slightly larger than that for path A.
However, such a small effect would be negligible compared with errors
due to modeling or approximation in computational methods. At the
present level of calculations it is safer to say that path B passing through
4b′ and ts(14b′-4b′) is more favorable because of the larger stability
of 4b′ compared with 4a′. Note that ts(14b′-4b′) is always more stable
than ts(14a′-4a′).
and after that every hour for 4 h at -10 °C. The rate constant (kobs
)
(5.3 ( 2.0) × 10-4, ∆Gq ) 19.3 ( 0.2 kcal/mol) was determined from
the slope of the line from ln [4-CH3/aromatic H] versus time. The
same experiment was repeated five times to estimate the error. The
rate was measured at 0 °C in the same manner (kobs ) 11.8 × 10-4
,
∆Gq ) 19.6 kcal/mol, R2 ) 0.931). Both experiments were duplicated.
(2) CO Insertion into 4 To Afford 5. The alkylpalladium 4 (15.0
mg, 8.2 µmol) was dissolved in CDCl3 (0.6 mL) under Ar in an NMR
tube. One equivalent of CH3CN (0.40 µL, 7.7 µmol) was added, and
then the solution was saturated by CO under 1 atm by the freeze-
thaw method. After the solids were dissolved at -78 °C, the NMR
tube was put into a cooled autoclave at -10.0 °C. Under a CO pressure
of 20 atm, the reaction was checked after 71, 151, and 319 min by
1
release-repressurize of CO to measure H NMR under 1 atm of CO
at -23 °C. The reaction did not proceed under 1 atm of CO at this
temperature, and thus the time for recording the spectra was omitted
from the reaction time. The decay of the CH3 signal of 4 relative to
the unchanging aromatic protons was used to determine the first-order
rate constant for the reaction. The rate constant (kobs ) 1.20 × 10-4
,
∆Gq ) 19.0 kcal/mol, R2 ) 0.993) was determined from the slope of
the line from -ln [4-CH3/aromatic H] versus time. The rate was
measured at -12 °C in the same manner (kobs ) 1.16 × 10-4, ∆Gq )
18.9 kcal/mol, R2 ) 0.991).
Asymmetric Copolymerization of Propene with CO Catalyzed
by [Pd(CH3)(CH3CN){(R,S)-BINAPHOS}]‚[B{3,5-(CF3)2C6H3}4] (2).
A solution of (R,S)-BINAPHOS (8.7 mg, 0.011 mmol) in benzene (0.5
mL) was added to a solution of Pd(CH3)(Cl)(COD) (3.0 mg, 0.011
mmol) in benzene (1.0 mL). The solution was stirred at 20 °C for 1
h and concentrated in Vacuo. The product was dissolved with CH2Cl2
(1.0 mL), and then to the solution was added a solution of NaB(3,5-
(CF3)2C6H3)4 (10.0 mg, 0.011 mmol) in CH3CN (1.0 mL). After the
solution was stirred at 20 °C for 1 h, the solvents were removed. The
residue was dissolved in CH2Cl2 (2 mL) and was degassed by three
cycles of freeze-thaw. After being treated with CO (1 atm), the
solution was charged with propene (3 atm, 50 mL). The mixture was
stirred under CO (20 atm) for 4 days at 20 °C. After the pressure was
released, 0.2 mL of CH3OH was added, and the mixture repressurized
with CO in order to cleave the Pd-C bond by carbomethoxylation.
After 1 h, the reaction mixture was poured into CH3OH (100 mL) under
air to precipitate poly(propene-alt-CO) (0.333 g, 76% yield): Tm
)
1
164 °C; Tg ) 8 °C; Mw ) 104 400; Mn ) 65 300; Mw/Mn ) 1.6; H
NMR ((CF3)2CDOD) δ 1.03 (d, J ) 6.93 Hz, 3H, CH3), 2.52 (dd, J )
16.83, 1.98 Hz, 1H, CHH), 2.88-3.09 (m, 2H, CHH and CHCH3);
13C NMR ((CF3)2CDOD) δ 15.8, 41.5, 44.9, 218.0; IR (Nujol) 1705
In molecular mechanics (MM) calculations, PH3 and P(OH)3 were
replaced by (R,S)-BINAPHOS, and one of the hydrogen atoms of ethene
was replaced by a methyl group. The Cartesian coordinates of the
propene methyl group and all the atoms but phosphorous atoms of the
BINAPHOS ligand were optimized, whereas the B3LYP optimized
geometries, excluding the replaced hydrogen atoms of PH3 and ethene
(CO) cm-1; [φ]24 ) +40° (c 0.51, (CF3)2CHOH).
D
Asymmetric Copolymerization of (4-tert-Butylphenyl)ethene with
CO Catalyzed by [Pd(CH3)(CH3CN){(R,S)-BINAPHOS}]‚[B{3,5-
(CF3)2-C6H3}4] (2). A solution of 2 (0.011 mmol) in CH2Cl2 (1.0 mL)
was degassed by three cycles of freeze-thaw. After the solution was
treated with CO (1 atm), (4-tert-butylphenyl)ethene (1.77 g, 11.0 mmol)
was added. The mixture was stirred under CO (20 atm) for 4 days at
(34) Gaussian 94: M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W.
Gill, B. G. Johnson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A. Petersson,
J. A. Montgomery, K. Raghavachari, M. A. Al-Laham, V. G. Zakrzewski,
J. V. Ortiz, J. B. Foresman, J. Cioslowski, B. B. Stefanov, A. Nanayakkara,
M. Challacombe, C. Y. Peng, P. Y. Ayala, W. Chen, M. W. Wong, J. L.
Andres, E. S. Replogle, R. Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley,
D. J. Defrees, J. Baker, J. P. Stewart, M. Head-Gordon, C. Gonzalez, and
J. A. Pople, Gaussian, Inc., Pittsburgh, PA, 1995.
(35) (a) Maseras, F. Morokuma, K. J. Comput. Chem. 1995, 16, 1170.
(b) Matsubara, T.; Sieber, S. Morokuma, K. Int. J. Quantum Chem. 1996,
60, 1101. (c) Matsubara, T.; Maseras, F.; Koga, N.; Morokuma, K. J. Phys.
Chem. 1996, 100, 2573.
(36) (a) Kawamura-Kuribayashi, H.; Koga, N.; Morokuma, K. J. Am.
Chem. Soc. 1992, 114, 8687. (b) Maseras, F.; Koga, N.; Morokuma, K.
Organometallics 1994, 13, 4008. (c) Yoshida, T.; Koga, N.; Morokuma,
K. Organometallics 1996, 15, 766.
(37) Allinger, N. L. mm3(92); QCPE: Bloomington, IN, 1992.
(38) Rappe´, A. K.; Casewit, C. J.; Colwell, K. S. Goddard, W. A., III;
Skiff, W. M. J. Am. Chem. Soc. 1992, 114, 10024.

Copolymerization of Propene with Carbon Monoxide
J. Am. Chem. Soc., Vol. 119, No. 52, 1997 12795
20 °C and then treated with CH3OH (100 mL) to precipitate poly((4-
tert-butylphenyl)ethene-alt-CO) (0.224 g, 11% yield): Tm ) 158 °C;
Tg ) 122 °C; Mw ) 5600; Mn ) 4300; Mw/Mn ) 1.3; 1H NMR (CDCl3)
δ 1.24 (s, 9H, C(CH3)3), 2.59-2.74 (m, 1H, CHH), 3.10-3.20 (m,
1H, CHH), 3.89-4.10 (m, 1H, CH), 6.85 (d, J ) 8.25 Hz, 2H, Ar),
7.12 (d, J ) 8.25 Hz, 2H, Ar); 13C NMR (CDCl3) δ 31.3, 34.4, 45.6,
CDCl3 (0.5 mL). The solution was degassed by three freeze-thaw
cycles and treated with (E)-2-butene (1 atm) for 14 h to give the
corresponding alkylpalladium complexes 16a and 16b as an almost
1:1 mixture. Data for 16a: 31P NMR (CDCl3) δ 143.0 (d, JP-P ) 64
Hz), 13.0 (d, JP-P ) 64 Hz). Data for 16b: δ 141.1 (d, JP-P ) 67
Hz), 13.6 (d, JP-P ) 67 Hz). (Z)-2-Butene was treated similarly to
give the identical 31P NMR chart.
51.7, 125.6, 127.8, 134.4, 149.9, 207.0; IR (Nujol) 1711 (CO) cm-1
;
[φ]23 -492° (c 0.50, CH2Cl2).
D
Attempted CO and Propene Insertion into [Pd(COCH3)(CH3CN)-
{(R)-BINAP}]‚[B{3,5-(CF3)2C6H3}4]. A solution of [Pd(CH3)(CH3-
CN){(R)-BINAP}]‚[B{3,5-(CF3)2C6H3}4] (6.0 mg, 6.5 × 10-3 mmol)
in CDCl3 (0.5 mL) was treated with CO (1 atm) at room temperature
for 1 h to give [Pd(COCH3)(S){(R)-BINAP}]‚[B{3,5-(CF3)2C6H3}4].
Three species have been characterized by 31P NMR (the ratio was about
1:1:1). From 1H NMR, two of them were assigned to be acyl
complexes: 1H NMR (CDCl3) δ 2.37 (CH3CN), 2.17 (d, J ) 2.0 Hz,
CH3COPd), 2.04 (d, J ) 7.9 Hz, CH3COPd); 31P NMR (CDCl3) δ 21.3
(d, JP-P ) 55 Hz), 23.7 (d, JP-P ) 55 Hz); 20.6 (d, JP-P ) 54 Hz),
23.9 (d, JP-P ) 54 Hz); 12.6 (d, JP-P ) 66 Hz), 24.9 (d, JP-P ) 66
Hz). A solution of the above mixture (6.0 mg, 0.0065 mmol) in CDCl3
(0.5 mL) was treated with propene (1 atm) at room temperature for 20
h. A new pair of peaks were observed by 31P NMR, but the conversion
was lower than 10%: 31P NMR (CDCl3) δ 38.0 (d, JP-P ) 44 Hz), δ
16.0 (d, JP-P ) 44 Hz).
Acknowledgment. K.N. thanks Professor M. A. Bennett
(Australian National University) for helpful discussions. K.N.
and N.K. are grateful for Grants-in-Aid for Scientific Research
on Priority Area “New Polymers and Their Nano-Organized
Systems” (No. 08246231) and for Scientific Research (C) (No.
09650901), respectively, from the Ministry of Education,
Science, Sports, and Culture, Japan. Part of the calculations were
carried out at the Computer Center of the Institute for Molecular
Science, Japan.
Supporting Information Available: X-ray data of [Pd{CH2-
CH(CH3)CO}{(R,S)-BINAPHOS}]‚[B{3,5-(CF3)2C6H3}4] and
Pt(CH3)2{(R,S)-BINAPHOS} (29 pages). See any current
masthead page for ordering and Internet access instructions.
(E)-2-Butene Insertion into 3. The acylpalladium complex 3 (6.0
mg, 0.0065 mmol) was prepared as described above and dissolved in
JA973199X