Pd-catalyzed copolymerization of methyl acrylate with carbon monoxide: Structures, properties and mechanistic aspects toward ligand design

ARTICLE

pubs.acs.org/JACS

Pd-Catalyzed Copolymerization of Methyl Acrylate with Carbon

Monoxide: Structures, Properties and Mechanistic Aspects toward

Ligand Design

Akifumi Nakamura,^†Kagehiro Munakata,^†Shingo Ito,^†Takuya Kochi,^†,‡Lung Wa Chung,^§

Keiji Morokuma,*^,§and Kyoko Nozaki*^,†

^†Department of Chemistry and Biotechnology, Graduate School of Engineering, The University of Tokyo, 7-3-1 Hongo,

Bunkyo-ku, Tokyo 113-8656, Japan

^§Fukui Institute for Fundamental Chemistry, Kyoto University, Takano-Nishishiraki-cho, 34-4, Sakyo-ku, Kyoto 606-8103, Japan

S Supporting Information

b

ABSTRACT: Full details are provided for the alternating copolymerization of

acrylic esters with carbon monoxide (CO) catalyzed by palladium species

bearing a phosphineꢀsulfonate bidentate ligand. The copolymer of methyl

acrylate (MA) and CO had complete regioregularity with stereocenters that

slowly epimerize in the presence of methanol. In the presence of ethylene,

terpolymers of MA/ethylene/CO were also prepared. The glass transition

temperatures of the co- and terpolymers were higher than that of the ethylene/

CO copolymer. Both experimental and theoretical investigations were per-

formed to clarify the superior nature of the palladium phosphineꢀsulfonate system compared to an unsuccessful conventional

palladium diphosphine system: (i) The reversible insertion of CO was directly observed with the isolated alkylpalladium complexes,

[{o-((o-MeOC₆H₄)₂P)C₆H₄SO₃}PdCH(CO₂Me)CH₂COMe], whereas it was not observed with the corresponding complex

bearing 1,2-bis(diphenylphosphino)ethane (DPPE). (ii) The transition state of the subsequent MA insertion, the rate-determining

step of the catalytic cycle, was lower in energy in the phosphineꢀsulfonate system than in the DPPE system. This stabilization could

be attributed to the less hindered sulfonate moiety as well as the stronger back-donation from palladium to the electron-deﬁcient

oleﬁn, which is located trans to the sulfonate.

’ INTRODUCTION

inexpensive and readily available. The resulting copolymers

would possess functional groups directly attached to the polymer

main chain and their properties are expected to be quite diﬀerent

from ethylene/CO copolymers.

γ-Polyketones obtained by metal-catalyzed alternating copo-

lymerization of oleﬁns with carbon monoxide (CO) have

attracted much attention from academic and industrial

communities.^1,2This is because CO is an inexpensive resource

and the resulting copolymers exhibit unique properties, such as

high crystallinity and photodegradability. Potential industrial

applications of the polyketones include ﬁbers, ﬁlm coatings,

adhesives, membranes, and packaging materials.³Furthermore,

the reactive carbonyl groups in the polymer backbone can be

chemically modiﬁed and, thus, the polyketones serve as excellent

starting materials for other types of functionalized polymers.⁴

The properties of the copolymers can be tuned by changing

the oleﬁnic comonomers. Thus far, monosubstituted alkenes

such as propylene and styrene derivatives have been well

documented as comonomers for copolymerization with CO.¹

A variety of functionalized oleﬁns whose functional groups are

separated from the oleﬁn moiety by at least one methylene spacer

are also available for copolymerization (Scheme 1).^5,6In sharp

contrast, the copolymerization of fundamental polar monomers

with carbon monoxide has been a long-standing challenge in

polymer chemistry.^7,8The fundamental polar vinyl monomers⁹

such as methyl acrylate (MA), vinyl acetate, and vinyl chloride are

Especially in the case of MA, the possibility of copolymeriza-

tion with CO has been well studied for complexes ligated by

several types of ligands shown in Scheme 2.⁷It had been

postulated that no reaction occurred following the formation

of III (or further reaction was not described), generated by the

insertion of CO into the PdꢀMe bond of I and the subsequent

insertion of MA into the PdꢀCOMe bond of II. The resulting

ﬁve-membered chelate complexes (III) were found to be inactive

toward further insertion of CO and/or MA.

In 2007 and 2008, we found that vinyl acetate/CO¹⁰and MA/

CO¹¹can be copolymerized when a mixture of phosphoniumꢀ

sulfonate salt 1 and a Pd(0) source was employed as a catalyst. In

the case of MA, it was shown that the isolated Pd(II) complex 3a

bearing a phosphineꢀsulfonate ligand, corresponding to III in

Scheme 2, also successfully initiates the copolymerization. The

series of Pd phosphineꢀsulfonate complexes, the catalysts we

Received: January 13, 2011

Published: April 11, 2011

r

2011 American Chemical Society

6761

dx.doi.org/10.1021/ja2003268 J. Am. Chem. Soc. 2011, 133, 6761–6779

|

Journal of the American Chemical Society

ARTICLE

Scheme 1. Reported Functionalized γ-Polyketones

(Section 1) and the structural analyses (Section 2) of acrylates/

CO copolymers are disclosed. Terpolymers of MA/ethylene/

CO were also obtained and characterized (Section 3). In addi-

tion, the thermal properties of these co- and terpolymers were

investigated (Section 4). Subsequently, the mechanism of the

MA/CO copolymerization was investigated experimentally

(Section 5) and theoretically (Section 6) to clarify why the Pd

phosphineꢀsulfonate complex catalyzed the copolymerization

while the other catalysts could not (Section 7).

’ RESULTS AND DISCUSSIONS

1. Copolymerization of Acrylates with Carbon Monoxide.

The alternating copolymers of MA with CO were obtained using

catalysts generated in situ from Pd(dba)₂and phosphoniumꢀ

sulfonate 1.¹¹Treatment of MA with 6.0 MPa of CO at 70 °C for

20 h in the presence of Pd(dba)₂/1a afforded the copolymer with

M_nof 30 000 (white solid after reprecipitation, see Experimental

section, entry 1, Table 1).^17,18o-Methoxyphenyl substituted

ligand 1a offers both higher activity and molecular weights of

the copolymers than the phenyl substituted 1b (entries 1 and 2).

It should be noted that the reaction using cyclohexyl substituted

ligand (Cy₂PH)C₆H₄SO₃^12hor 2-(diphenylphosphino)benzoic

acid did not afford the copolymer. When 2 equiv of ligand 1a

were used, the activity was decreased (entry 3). This tendency

was also observed in the copolymerization of ethylene with

CO,^15bsuggesting that the initiation with in situ formed [PꢀO]-

Pd[PꢀO] complexes is sluggish.

Scheme 2. Reported Inactive Pd Complexes for Copolym-

erization of Methyl Acrylate with Carbon Monoxide

The activity of the copolymerization was dependent on CO

pressure (entries 1, 4 and 5). An increase in CO pressure (up to

8.0 MPa) led to an enhancement in catalytic activity. According

to the mechanistic studies in Sections 5 and 6 (vide infra), CO

coordinationꢀinsertion is a reversible, pre-equilibrium step.

Thus, the increased activity at higher CO pressure can be inter-

preted as a result of increased concentration of acylpalladium

species, which is advantageous for subsequent MA insertion.

Increasing the reaction temperature from 45 °C (entry 6) to

70 °C (entry 1) substantially enhanced the activity and molecular

weight. However, the molecular weight of the copolymer pro-

duced at 100 °C (entry 7) decreased while the catalytic activity

was similar to that produced at 70 °C. The number of the

polymer chains produced per catalyst (P/C) at 100 °C was 2.6,

which indicates that the chain transfer reaction (see Section 2)

was accelerated as well as the polymerization reaction at high

temperature.

To obtain the copolymer with higher molecular weight, a

longer reaction time was employed (entry 8). After 144 h, the

copolymer with higher molecular weight (M_n40 000), broader

M_w/M_nand low TOF was obtained. In this case, high viscosity of

the reaction mixture seems to be the major obstacle to the rapid

propagation reaction.

The isolated Pd phosphineꢀsulfonate complexes were also

investigated as catalysts. In the case of 2,6-lutidine-bound com-

plex 2a, only a trace amount of the copolymer was obtained

(entry 9). This result suggests that the strong binding aﬃnity of

2,6-lutidine inhibits the coordination of electron-deﬁcient MA.¹⁹

On the other hand, nitrogen-free complex 3a, which was pre-

pared by the insertion of one equivalent of MA to acylpalladium

complex (4a) (Scheme 3), initiated and catalyzed the copoly-

merization of MA/CO eﬀectively (entry 10).¹¹

used, have been recently recognized as potent catalysts for

coordination copolymerization of ethylene with polar vinyl

monomers since their ﬁrst academic report by Drent in

2002.^1h,9c,12,13These polyethylenes possessed highly linear

structures¹⁴with functional groups directly attached to the linear

polyethylene chain, which has not been obtained by any other

catalysts. Furthermore, Pd phosphineꢀsulfonate catalysts can

produce nonalternating ethylene/CO copolymers, which are not

accessible by Pd catalysts with conventional diphosphine or

diamine ligands.^15,16For the past decade, the Pd phosphineꢀ

sulfonate complexes have conquered a considerable number of

challenging problems in polymer chemistry. Nevertheless, the

reasons for the success are still unclear. Speciﬁcally, a reasonable

understanding of how the unique design of the ligand contributes

to the successful and unprecedented copolymerization of polar

monomers is needed.

This article presents the details of this unprecedented alter-

nating copolymerization of MA with CO catalyzed by Pd

phosphineꢀsulfonate catalysts. The main aim of this study is

to reveal the role of the sulfonate moiety in the catalyst through

comparison with conventional ligands. The contents of this

report are in the following order. The details of the syntheses

The copolymerization of a variety of monomers has been

examined in addition to MA. t-Butyl acrylate can be copolymerized

6762

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

Table 1. Copolymerization of Acrylates with Carbon Monoxide^a

entry

catalyst

R

P_CO(MPa)

temp (°C)

time (h)

yield (mg/%)^b

TOF (h^ꢀ1

)

M_n^c(ꢁ 10³)

M_w/M_n

P/C^d

1

1a/Pd(dba)₂

1b/Pd(dba)₂

1a^e/Pd(dba)₂

1a/Pd(dba)₂

2a

Me

t-Bu

6.0

3.0

8.0

6.0

70

45

100

70

20

144

20

490/15

140/4.3

260/12

370/12

540/17

150/4.8

500/16

1500/48

43/1.3

21

6.0

11

16

24

6.7

22

9.2

1.8

25

2.9

30

26

6.0

21

17

16

19

40

2.1

24

7.9

1.6

1.2

1.3

1.6

1.7

1.3

1.7

2.0

1.1

1.6

1.5

1.7

0.53

4.3

1.8

3.4

1.0

2.6

3.8

2.0

2.4

1.1

2

3

4

5

6

7

8

9

10

11

3a

580/18

90/3.3

1a/Pd(dba)₂

^aUnless otherwise noted, reaction was performed with 0.012 mmol of ligand 1, 0.010 mmol of Pd source, and 2.5 mL of methyl acrylate without

additional solvent. ^bYield of the copolymer was determined by subtraction of the weight of catalyst from the amount of solid product obtained and

calculated based on the acrylate used. ^cMolecular weights were determined using narrow polystyrene standards. ^dNumber of polymer chains produced

per catalyst based on palladium. ^eTwo equivalents of ligand 1a (0.020 mmol) was used.

Scheme 3. Synthesis of Five-Membered Chelate Complex 3a

Bearing a PhosphineꢀSulfonate Ligand¹¹

with CO (entry 11) and the regioregularity of the resulting copol-

ymer was not controlled while the copolymerization of methyl

acrylate/CO aﬀorded regiocontrolled product (see Section 2).

Aside from the acrylates and vinyl acetate,¹⁰none of the attempted

Figure 1. Microstructures in γ-polyketones. FG = ꢀCH₃for poly-

polar vinyl monomers whose oleﬁn moieties are directly func-

tionalized^1hparticipated in copolymerization under the condi-

tions of entry 1. Attempted copolymerization of methyl metha-

crylate with CO resulted in homopolymerization of methyl

methacrylate probably via radical pathway.²⁰Methyl vinyl ke-

tone, N-isopropylacrylamide, acrylonitrile, vinyl chloride did not

aﬀord any polymeric products. When we used N-vinyl pyrroli-

done, only head-to-tail dimer was obtained.²¹A trace amount of

cooligomer of n-butyl vinyl ether/CO was detected by MALDI-

TOF MS spectrum, but the only major product was poly(butyl

vinyl ether)²²as conﬁrmed by NMR.

2. Structural Analyses of the Copolymers of Acrylates with

Carbon Monoxide. The microstructure of γ-polyketones is

one of the most important factors that affect their phys-

ical properties.^1b,eThe regio-, stereo-, and enantiocontrol of

γ-polyketones have, thus far, been well investigated for

(propylene-alt-CO) and ꢀCO₂Me for poly(methyl acrylate-alt-CO).

poly(propylene-alt-CO) and poly(styrene-alt-CO) (Figure 1).

In the previous communication, we confirmed the regiocon-

trolled, alternating structure of poly(methyl acrylate-alt-CO) by

NMR and MALDI-TOF mass spectroscopy.^11,23The exquisite

control of head-to-tail architecture was inferred from the ob-

servation of only a single singlet resonance in the ketone region

in ¹³C NMR spectrum (Figure 2B).

Although a single resonance was observed in the ketone region

at room temperature, it was found to split into multiple peaks

at ꢀ60 °C (Figure 2C). In addition, the signal is wider than that

of regio-, stereo- and enantiocontrolled iso-poly(propylene-

alt-CO) in Figure 2D.²⁴Therefore, it was assumed that

epimerization at the asymmetric carbon center might occur,

6763

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

Figure 2. ¹³C NMR spectra of (A) poly(methyl acrylate-alt-CO), (B) the ketone carbonyl region of poly(methyl acrylate-alt-CO) at ambient temperature,

(C) the ketone carbonyl region of poly(methyl acrylate-alt-CO) at ꢀ60 °C, and (D) the ketone carbonyl region of isotactic poly(propylene-alt-CO).²⁴

Scheme 4. H/D Exchange of Methine Protons of Poly-

(methyl acrylate-alt-CO)

Scheme 5. Initiation Chain End Analysis of the Copolymers

Initiated by (a) the Mixture of 1 and Pd(dba)₂and (b)

Isolated Complex 3a¹¹

since poly(methyl acrylate-alt-CO) is a β-ketoester whose pK_a

(≈10) is much lower than that of other γ-polyketones.²⁵

However, the epimerization rate was found to be too slow to

be observed in NMR time scale (10^ꢀ1ꢀ10^ꢀ6s). When the

copolymer was treated with 15 equiv of MeOD (based on the

number of repetitive units) in CD₂Cl₂, only 35% of the methine

protons were exchanged with deuterium even after 4 h at room

temperature (Scheme 4).²⁶Thus, the epimerization can occur

gradually in the presence of methanol but it does not account for

the split of NMR signals in Figure 2C.

According to these results, there are two probable explanations

regarding the stereochemistry of poly(methyl acrylate-alt-CO). The

ﬁrst possibility is that the stereochemistry is not controlled and the

relatively wide NMR signal in Figure 2B is due to the presence of

many signals overlapped. In this case, the change of the signal shape

at low temperature could be attributed to the restriction of the

ﬂuctuation of a high-order structure, for example, helixꢀcoil transi-

tion. The second is that the stereochemistry is controlled but the

ﬂuctuation of a higher-order structure occurs both at room tem-

perature and at ꢀ60 °C.

CDCl₃) of isobutyric acid t-butyl ester (i-PrCO₂t-Bu, 176.7 ppm)

and isobutyric acid (i-PrCO₂H, 184.0 ppm) suggest that the t-Bu

group of the obtained copolymer is intact.

Initiation chain ends of the poly(MA-alt-CO) were character-

ized by ¹H NMR spectroscopy and the structures were conﬁrmed

as shown in Scheme 5.¹¹When a mixture of 1 and Pd(dba)₂was

used, a 1-methoxycarbonylethyl group was detected as an initiat-

ing chain end, which suggests that the reaction was initiated

by the formation of a PdꢀH bond via protonation of Pd(0)

species,^1h,12afollowed by subsequent insertion of MA and CO. In

contrast, copolymers formed with 3a showed a signal corre-

sponding to the acetyl initiating end group, which originated

from the complex 3a (Scheme 5b).

When t-butyl acrylate was used as a monomer (Table 1, entry

11), the signals in the ketone carbonyl region in the ¹³C NMR

1

spectrum were more complex. The methine protons in the H

NMR spectrum showed at least three separate signals which indicate

the lack of regioselectivity (see Supporting Information).²⁷The

multiple signals for the ester region (-CO₂t-Bu) are within a quite

narrow region (168.1ꢀ168.4 ppm). The chemical shifts (¹³C NMR,

In the cases where the P/C values in Table 1 are larger than 1,

chain-transfer reactions are suggested to take place during the

polymerization, but the structures of terminating chain ends for

these polymers could not be identiﬁed.¹¹

6764

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

Table 2. Terpolymerization of Methyl Acrylate/Ethylene/CO^a

entry

P_ethylene(MPa)

P_total(MPa)

yield (mg)^b

activity (g mmol^ꢀ1h^ꢀ1

)

x:y

M_n(ꢁ 10³)

M_w/M_n

3

1^c

2^d

2.0

0.5

5.5

6.0

80

45

2.4

0.68

11:89

45:55

n.d.^b

1.9

3.1

^aReactions were performed with 0.012 mmol of ligand 1, 0.010 mmol of Pd source, and 2.5 mL of methyl acrylate without additional solvent. The yield

(mg) of the terpolymer was determined by subtraction of the weight of catalyst from the amount of solid product obtained. ^bMolecular weight was not

determined due to the low solubility of the ethylene-rich terpolymer in common solvents for SEC analyses. Note that the NMR spectra of this terpolymer

were collected in a mixture of CDC1₃/1,1,1,3,3,3-hexaﬂuoro-2-propanol (1:2). ^cRatio of x:y was based on ¹H NMR analysis. ^dRatio of x:y was based on

the carbonyl signals in the ¹³C NMR spectrum.

4. Thermal Properties of Co- and Terpolymers. Thermal

analyses were carried out for the obtained co- and terpolymers

(Figure 3). According to the differential scanning calorimetry

(DSC) analyses, T_gof the copolymer of entry 1 in Table 1

(poly(MA-alt-CO)) and the terpolymer of entry 2 in Table 2

(poly(CO-alt-(ethylene; MA)) were found around 50 °C^11,29

and 22 °C, respectively. These values are higher than that

of poly(ethylene-alt-CO) (5ꢀ20 °C).^1bThis is in a sharp contrast

to the n-alkyl substituted γ-polyketones, which exhibit lower

T_gthan that of poly(ethylene-alt-CO). The low T_gof the n-alkyl

substituted γ-polyketones was attributed to the reduction

of dipoleꢀdipole interactions between the ketones in the

molecular chains.^1b,30The uniquely high T_gof ester-containing

γ-polyketones could be attributed to the stronger dipoleꢀdi-

pole interactions between molecular chains induced by the

ester groups in addition to the ketone moieties. As for melt-

ing temperatures of poly(MA-alt-CO) and poly(CO-alt-

(ethylene; MA)), no apparent signals were observed in DSC

analyses.³¹

Figure 3. Comparison of the glass transition temperatures (T_g) of the

γ-polyketone. The values in parentheses are melting temperatures (T_m).

n.d. = not detected.

Thermogravimetry (TG) analyses (Figure 4) showed thermal

decomposition starts at ca. 200 °C for both co- and terpolymers.

In the case of the terpolymer, additional decomposition starts at

300 °C, which corresponds to the decomposition temperature of

3. Syntheses and Characterization of Terpolymers of

Methyl Acrylate/Ethylene/CO. A mixture of ligand 1a and

poly(ethylene-alt-CO).^1b

Pd(dba)₂catalyzes terpolymerization of MA, ethylene and CO

(Table 2).^28,29The incorporation ratio of the resulting terpo-

lymers depends upon the ethylene pressure. When 2.5 MPa of

ethylene was introduced, ethylene-enriched terpolymer was

obtained with the ratio of MA and ethylene (x:y) to be ca.

11:89 in a mixture of CDCl₃/1,1,1,3,3,3-hexafluoro-2-propanol

(1:2) (entry 1, Table 2). In the obtained terpolymers, olefins

and CO are aligned in an alternating fashion without any

successive olefin moieties.^15,16In addition, no successive

“methyl acrylate-CO” units described as “x” in Table 2 was

detected. This is based on the fact that the signals around

201 ppm and 168 ppm, which are the signals for ketone and

ester carbonyls in the copolymer, were not detected and other

detectable signals in the carbonyl region were all assigned (see

Supporting Information).

5. Experimental Mechanistic Studies. In contrast to the

catalysts described in Scheme 2, the Pd phosphineꢀsulfonate

system could uniquely copolymerize methyl acrylate with carbon

monoxide. This suggested that the Pd phosphineꢀsulfonate

system allowed furtherinsertion ofmonomers into five-membered

chelate complex 3a (Scheme 5),¹¹while other Pd complexes

resisted further reaction to III (Scheme 2).⁷In this section, the

structure and the reactivity of these key five-membered chelate

intermediates, [LꢀL]PdCH(CO₂Me)CH₂COMe (3a, 3cꢀe),

are compared.

First, we synthesized an analogous complex (3c) ligated by

1,2-bis(diphenylphosphino)ethane (DPPE) according to a mod-

iﬁed procedure reported by van Leeuwen^7b(see Experimental

section) because complexes ligated by DPPE are some of the

most frequently employed for the copolymerization of aliphat-

ic oleﬁns with CO.^1a,c,32The copolymerization of MA with CO by

3c did not proceed,^7b,33but rather the formation of poly(methyl

acrylate) was observed. Formation of the homopolymer could be

initiated by adventitious radical species or homolytic cleavage of

PdꢀCH(CO₂Me)R bond.³⁴

By decreasing ethylene pressure to 0.5 MPa, the ratio changed

to approximately x:y = 45:55 (entry 2, Table 2). However, a

detailed assignment of the polymer microstructure was diﬃcult

due to its highly complex NMR signals, suggesting a random

architecture of the copolymer.

6765

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

Figure 4. Thermolysis curves of (a) poly(MA-alt-CO) and (b) poly(CO-alt-(ethylene; MA)), entry 2 in Table 2.

Table 3. Comparison of the Experimental PdꢀC and PdꢀO Lengths (Å) of the Five-Membered Chelate Complexes

Next, the structures of the key intermediates were compared.

Experimentally determined PdꢀC and PdꢀO bond lengths for a

series of ﬁve-membered chelate intermediates are listed in

Table 3. Although one might expect that the ﬁve-membered

chelate in the active Pd phosphineꢀsulfonate complex 3a would

be more weakly³⁵than the carbonyl oxygen of the inactive com-

plexes (3cꢀe), its PdꢀC (2.042(6) Å) and PdꢀO (2.137(4) Å)

distances are not signiﬁcantly diﬀerent than those in 3cꢀe. The

PdꢀC bond length of 3a is almost the same as those of 3d and 3e,

while that of 3c is slightly longer because of the strong trans

inﬂuence of the phosphorus moiety located trans to the alkyl

group. In this sense, the trans inﬂuence of a sulfonate group can

be recognized as being as weak as an imine group.^36,37In terms of

PdꢀO bond lengths, 3a is within the range of 3dꢀe as listed in

Table 3.

Finally, the reactivities of these ﬁve-membered chelate com-

plexes (3a and 3c) with CO were investigated using a high

pressure NMR apparatus.³⁸Exposure of ﬁve-membered chelate

complex 3a to 6 MPa of CO at ambient temperature aﬀorded a

mixture of novel complexes along with the starting complex 3a

(Figure 5). In the ¹³C NMR spectrum in CD₂Cl₂, new signals in

the carbonyl region were observed at 181.3, 202.4, 205.3 ppm in

addition to the signal of free CO at 183.8 ppm. Considering that

the R-position of CO₂Me group shifted from 34.6 ppm to 63.7

ppm, the new complex observed was assigned to PdCOCH-

(CO₂Me)CH₂COCH₃and the signal at 202.4 ppm (d, J = 16.1

Hz) was assigned to be that of the Pdꢀacyl carbonyl moiety

located cis to the phosphine (C4 in Figure 5). Furthermore, a

ketone functionality coordinated to the Pd center was not

present, typically observed around 220ꢀ240 ppm.^7cInstead,

the ꢀCH₂COCH₃carbonyl signal was observed at 205.3 ppm,

which indicates that the ketone carbonyl oxygen does not

coordinate to the Pd center (C3).^7cThe fourth coordination

site of Pd is occupied by the carbonyl oxygen of the ester group,

which is observed at 181.3 ppm (br) (C5).^12f

Therefore, the new signals were assigned as the complex

[PꢀSO₃]Pd(CO)COCH(CO₂Me)CH₂COCH₃(5a), formed

by associative CO coordination (see Section 6), migratory inser-

tion of CO, and cis/trans isomerization reaction (Scheme 6).

Detailed assignments of the complex 5a are shown in the Experi-

mental section. Most of 5a underwent decarbonylation after

releasing the CO pressure to regenerate the ﬁve-membered

complex 3a.³⁹This result suggests that the CO insertion to the

complex 3a is reversible even at ambient temperature.

In contrast, no new signals were detected in the carbonyl

region after exposing the ﬁve-membered chelate complex 3c,

bearing DPPE ligand, to CO (6 MPa) (Scheme 6, see Supporting

Information). This inert nature of 3c toward CO insertion can be

understood either by kinetic and/or thermodynamic reasoning,

which will be discussed in Section 7.

Consequently, the comparison of the key ﬁve-membered

chelate complexes bearing a phosphineꢀsulfonate (3a) and

DPPE (3c) can be summarized as follows: (i) the structures

of the complexes are similar, (ii) 3a undergoes reversible inser-

tion of CO to aﬀord acylpalladium complex 5a, (iii) the reaction

of 3c with CO does not aﬀord an acylpalladium complex.

6. Mechanistic Studies by Theoretical Calculations. To

study the detailed mechanism of the overall catalytic cycle, we

conducted theoretical calculations. Our aim was to understand

why the Pd phosphineꢀsulfonate system accomplished the

6766

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

Figure 5. ¹³C NMR (CD₂Cl₂) of complex 3a (top) and 3a þ 6 MPa of CO using a high-pressure NMR tube (bottom). Representative signals are

assigned to 3a (blue) and 5a (red). Ar = o-CH₃OC₆H₄, [PꢀO] = {o-((o-MeOC₆H₄)₂P)C₆H₄SO₃}.

scrutinize various possible pathways in some complexes while

avoiding optimization of highly complicated structures having

many possible conformations.

Scheme 6. Reaction of Key Five-Membered Chelate Com-

plexes 3a and 3c with 6 MPa of CO

For all the stationary points, the relative energies with zero

point energy correction (EþZPC) and free energies (G, at

298.15 K and 1 atm) are listed in front of and behind the forward

slash (/), respectively. The energies are expressed relative to the

complexes A_cisand A_PPwhich are analogous to 3a and 3c,

respectively. The optimized structures of A_cisand A_PPwere in

reasonable agreement with the X-ray crystal structures of 3a and

3c (see Supporting Information).⁴²

The processes we studied include relative stabilities of cis/trans

isomers of palladium phosphineꢀsulfonate complexes and their

isomerization reactions (Section 6.1), CO coordination and

insertion (Section 6.2), MA insertion (Section 6.3), and possible

side reactions (Section 6.4). After the calculation of these essential

reactions, we will discuss the overall catalytic cycle and compare

the phosphineꢀsulfonate ligand with DPPE in Section 7.

6.1. Relative Stabilities of cis/trans Isomers and Their Iso-

merization Reactions. Due to the unsymmetric nature of the

phosphineꢀsulfonate ligands, their square-planar palladium(II)

complexes have both cis/trans isomers.⁴³Duringourinvestigations

on the stability of these isomers, as well as isomerization transi-

tion states in ethylene homopolymerization catalyzed by Pd

phosphineꢀsulfonate complexes,¹⁴we found that the relative

location of an alkyl group in all the intermediates was preferen-

tially cis to the phosphine moiety (Figure 7). This preference can

be attributed to the strong trans influence of the alkyl group and

the phosphine moiety, which prefer not being trans to each other.

This tendency was in agreement with the experimental fact that

an alkyl group is positioned cis to the phosphine moiety in the

X-ray structure of 3a.¹¹It is confirmed by DFT calculation that

complex A_cis, the analogue of 3a, is more stable than its cis/trans

isomer A_transby 5.4/4.6 kcal/mol.

copolymerization of MA with CO while the Pd dppe complex,

one of the most commonly used catalysts in the copolymeriza-

tion of aliphatic olefins with CO,^1a,ccould not. In our theoretical

studies, density functional theory method (B3LYP⁴⁰/6-31G* and

Lanl2dz⁴¹) was employed.⁴²As ligands, diphenylphosphinoben-

zenesulfonate (deprotonated form of 1b, see entry 2 in Table 1)

and DPPE were used. Since both five-membered chelate com-

plexes bearing phosphineꢀsulfonate (3a) and DPPE (3c)

ligands were isolable and supposed to be stable intermediates

in the catalytic cycles, we adopted these complexes as starting

materials for the calculations.

Hereafter, the isomer having an X ligand (X = alkyl or acyl)

and phosphorus atom in cis positions to each other will be de-

scribed as “cis” for the united deﬁnition throughout the manu-

script. In other words, an L ligand (L = carbonyl oxygen, CO,

oleﬁn) and phosphorus atom are cis to each other in the “trans”

isomer (Figure 6). For example, A_cisis the cis/trans isomer of

A_trans. Additionally, we used an asterisk (*) to indicate the simpli-

ﬁed chain replacement of the ꢀCH(CO₂CH₃)CH₂COCH₃

group by a ꢀCH₃group. For instance, *D_cis([PꢀO]PdCOCH₃-

(MA)) is the simpliﬁed structure of D_cis([PꢀO]PdCOCH-

(CO₂CH₃)CH₂CH₃(MA)). The simpliﬁed structures are used to

The above-mentioned preference is generally applicable to

most of the intermediates in this article. However, carbonyl

complexes were the only exceptions to this rule. The carbonyl

complex B_transis lower in energy than complex B_cisby 2.1/3.3

kcal/mol although the alkyl group in B_transis trans to the

6767

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

Figure 6. Deﬁnitions for the basic structures used in theoretical part.

that phosphines are eﬃcient π-acceptors due to their PꢀC σ*

orbital overlapping with the d_πorbital of a metal center.⁴⁶For

this reason, the portion of π-back-donation from Pd to CO is

decreased when CO is trans to the phosphine. On the other hand,

the sulfonate group does not have an appropriate vacant orbital

which can withdraw the electrons eﬃciently from a ﬁlled d orbital

of a Pd center.⁴⁷Thus, the sulfonate group can be considered as a

weak π-acceptor. Furthermore, lone pairs on the oxygen atom of

the sulfonate group can act as weak π-donors. The back-donation

to CO should be further enhanced in order to avoid the repulsion

between the ﬁlled d_πorbitals of Pd and ﬁlled nonbonding orbitals

of the oxygen. These two characteristics (weak π-acceptor and

weak π-donor) signiﬁcantly contribute to the stronger back-

donation from Pd to CO when it is located trans to the sulfonate

(Figure 8).

Figure 7. Standard cis/trans preference of Pdꢀalkyl complex A_cisand

A_trans

.

The above-discussed diﬀerence in back-donation can also be

seen in other intermediates (Figure 6). Aside from the carbonyl

complexes, acylpalladium species and MA-coordinated com-

plexes also exhibited stronger back-donation when these moi-

eties are trans to the SO₃group. Acyl groups in C_cis, and D_cis

exhibited a longer CdO bond than in C_transand D_trans

,

respectively. This is because π* orbitals of CdO bond overlaps

with ﬁlled d orbitals of Pd center. In addition, the distance

between Pd and the oleﬁn moiety of MA is shorter and CdC

bond is longer in D_transcompared to those in D_cis, suggesting

that the back-donation from Pd to the electron deﬁcient CdC is

eﬀective when the sulfonate is trans to MA. Although the acyl

moiety and MA can accept the back-donation from the Pd center,

the extent to which it occurs is weaker than CO. In fact, the

stability of the series of C_transand D_transfollows the standard cis/

trans preference shown in Figure 7.

The back-donating ability of the cationic Pd dppe complexes

are also compared and shown in Table 4. In the case of the Pd

dppe complexes, CO, acyl moiety, and MA are always trans to the

phosphorus atom. The bond lengths of a moiety in the Pd dppe

complexes are more similar to the corresponding bond lengths in

the isomer of the Pd phosphineꢀsulfonate complex where the

moiety in question is trans to the phosphorus. For example, the

PdꢀC and CꢀO bond lengths of CO in B_PP(1.995 and 1.137 Å)

Figure 8. Conceptual pictures for the diﬀerence in back-donation

between carbonyl complexes B_cisand B_trans

.

phosphine (Figure 8).⁴⁴The reverse preference can be attributed

to the strong trans-directing nature of carbon monoxide,⁴⁵which

avoids being trans to the phosphine. In addition, the back-

donation from the ﬁlled d_πorbitals of the Pd center to the empty

π* orbitals of CO play a key role in the stabilization.^45bIn fact, the

calculated distances between Pd and the carbon of CO (2.001 Å

for B_cisvs 1.888 Å for B_trans) and the bond lengths of CO (1.137

Å for B_cisvs 1.144 Å for B_trans) suggested that the back-donation

in B_transis stronger than that in B_cis(Table 4). Thus, the back-

donation to CO is enhanced when CO is trans to the sulfonate

group (B_trans) compared to its isomer (B_cis). It is well-known

6768

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

Table 4. Calculated Bond Lengths (Å) for the Selected Intermediates Bearing a PhosphineꢀSulfonate Ligand and a DPPE Ligand

ARTICLE

phosphineꢀsulfonate

DPPE

trans to SO₃

trans to P

PdꢀCO

PdꢀC

CꢀO

PdꢀC

CꢀO

PdꢀC

CꢀO

PdꢀC

CdC

1.888 (B_trans

1.144 (B_trans

)

<

>

<

>

<

>

<

>

2.001 (B_cis

)

1.995 (B_PP

1.137 (B_PP

)

1.137 (B_cis

PdꢀC(O)Me

1.983 (C_cis

1.209 (C_cis

)

1.995 (C_trans

1.199 (C_trans

)

2.047 (C_PP

1.201 (C_PP

)

1.988 (D_cis

1.204 (D_cis

)

2.027 (D_trans

)

2.062 (D_PP

1.201 (D_PP

)

1.196 (D_trans

PdꢀMA

2.242ꢀ2.265 (D_trans

)

2.289ꢀ2.365 (D_cis

)

2.329ꢀ2.391 (D_PP

1.374 (D_PP

1.382 (D_trans

1.379 (D_cis

Figure 9. Charge distributions of A_cisand A_PPestimated by NBO population analyses (B3LYP/6-31G*, Lanl2dz).

phosphineꢀsulfonate complexes is that the pentacoordinate TS

is lower in energy than that of the corresponding simple rotation

TS. We also investigated transition states of cis/trans isomeriza-

tion required for the copolymerization of MA with CO and

showed that TS(C_cisꢀC_trans) and TS(D_cisꢀD_trans) represent

pentacoordinated Berry’s pseudorotation transition states, which

need only 7.2/7.0 (from C_trans), and 7.3/8.0 (from D_trans) kcal/

mol, respectively (in Scheme 7 and Scheme 10). Between A_cis

and A_transand between B_cisand B_trans, pentacoordinate struc-

tures Amed and Bmed were found as intermediates instead of TS

(Scheme 7).⁵⁰

6.2. CO Coordination and Insertion. The alternating copo-

lymerization is initiated by CO coordination to five-membered

chelate complex A_cis(Scheme 7). For the 16e d⁸square planar Pd

complex, the associative ligand substitution is widely accepted,

particularly with the strong π-acceptor CO.⁵¹As expected,

chelate opening by CO can take place via trigonal-bipyramidal

transition states either from A_cisor A_transwith barriers of 5.0/

15.2 and 9.3/18.4 kcal/mol from A_cis, respectively.⁵²Through

TS(A_cisꢀB_cis) and TS(A_transꢀB_trans), the CO molecule moves

to the equatorial plane and the chelating carbonyl group moves

away from the equatorial plane.⁴⁴

Figure 10. Cis/trans isomerization reaction via Berry’s pseudorotation-

like transition state.¹⁴

are quite close to those of B_cis(2.001 and 1.137 Å) rather than

those of B_trans(1.888 and 1.144 Å).⁴⁸Therefore, the back-

donation in these complexes appears to be inﬂuenced primarily

by the ligand located trans to CO. On the other hand, the

contribution of the net charge of these complexes (i.e., neutral Pd

phosphineꢀsulfonate and cationic Pd dppe) to the back-donat-

ing ability seems to be minor. In fact, NBO population analyses⁴⁹

showed that the negative charge is mainly located on the oxygen

atoms of the sulfonate group for the Pd phosphineꢀsulfonate

complexes, and the charge on Pd is almost the same as that for

the cationic Pd dppe complex (For the case of A_cisand A_PP, see

Figure 9). Thus, the ability of back-donation depends not on the

net charge of the complexes but mainly on the stereoelectronic

eﬀect of the trans ligands and the palladium center.

The transition states for the subsequent CO insertion into the

Previously, we proposed that cis/trans isomerization of Pd

phosphineꢀsulfonate proceeds via so-called Berry’s pseudorota-

tion of pentacoordinated complexes involving the associ-

ative exchange of the oxygen atoms in the sulfonate group

(Figure 10).¹⁴One of the most unique characteristics of Pd

PdꢀC bond were successfully located both from B_cisand B_trans

.

TS(B_transꢀC_cis) is lower in energy than TS(B_cisꢀC_trans) by

11.4/10.8 kcal/mol mainly because the alkyl group in B_transis

more activated for the migratory insertion by the stronger trans

eﬀect of the phosphine moiety than the sulfonate oxgen.^14,16,43

6769

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

Scheme 7. CO Coordination and Insertion to Five-Membered Chelate Complex Bearing PhosphineꢀSulfonate Ligand (EþZPE/

G, kcal/mol). The red arrow exhibits the most preferable pathway

Scheme 8. CO Coordination and Insertion to Five-Membered Chelate Complex Bearing DPPE Ligand (EþZPE/G, kcal/mol)

In fact, the PdꢀC (alkyl chain) bond length in B_trans(2.152 Å) is

longer than that in B_cis(2.109 Å). Accordingly, CO insertion

takes place from B_transafter CO coordination to the ﬁve-

membered chelate complex A_cis, accompanied by cis/trans iso-

merization via Bmed.⁵³After CO insertion, the internal ester

carbonyl oxygen coordinates to the Pd center to aﬀord a ﬁve-

membered chelate complex C_cisas observed in experiment (5a,

Figure 5).⁵⁴

For comparison, we calculated the reaction pathways of CO

coordination and insertion toward A_PPbearing a DPPE ligand

(Scheme 8). Associative CO coordination via TS(A_PPꢀB_PP)

requires a barrier of 3.6/13.0 kcal/mol from A_PP, which is

similar to the case of the Pd phosphineꢀsulfonate complex

(TS(A_cisꢀB_cis), 5.0/15.2 kcal/mol). This result indicates that a

high barrier for CO coordination is not the reason why Pd dppe

complexes cannot copolymerize MA and CO (Scheme 2).^7bSub-

sequently, CO insertion can take place via TS(B_PP-C_PP) with a

barrier of 13.2/21.0 kcal/mol.⁵⁵The interpretation of these

calculations and experiments in Section 5 will be discussed in

Section 7.

acylpalladium species to complete the catalytic cycle. Since it

was difficult to calculate all conformations of the highly complex

intermediates, we first studied MA insertion into a simple

Pdꢀacetyl (PdCOMe) bond as a model instead of the PdꢀCH-

(CO₂CH₃)CH₂COCH₃group (i.e., we first used *D_cisand

*D_transinstead of D_cisand D_trans). While the MA insertion into

Pdꢀalkyl bonds has been widely studied by theoretical

calculations,^12f,56,57this is the first theoretical study dealing with

MA insertion into a Pdꢀacetyl bond to the best of our

knowledge.^1h

From cis/trans isomers of PdCOMe(MA) complex *D_cisand

*D_trans, two types of MA insertion pathways, i.e., 2,1- and 1,2-

insertion, could proceed (Scheme 9). Both 2,1- and 1,2-insertion

transition states from *D_cisand *D_transwere successfully located. It

was found that 2,1-insertion predominates over 1,2-insertion (i.e.,

TS(*D_transꢀA_cis) and TS(*D_cisꢀA_trans) lower in energy than TS-

(*D_transꢀE_cis) and TS(*D_cisꢀE_trans), respectively). This is consis-

tent with the experimental result that 3a was obtained as the sole

product when MA was added to the solution of 4a (Scheme 3). The

strong preference toward 2,1-insertion results in highly regiocon-

trolled architecture of the copolymer, as conﬁrmed in Section 2.⁵⁸

The preference toward 2,1-insertion could be attributed to three

6.3. MA Insertion. After insertion of CO into five-membered

chelate complexes (Section 6.2), MA should react with the

6770

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

Scheme 9. MA Insertion into the PdꢀAcetyl Bond by Pd PhosphineꢀSulfonate Complexes (EþZPE/G, kcal/mol)^a

^aThe long alkyl chain (PdꢀCOCH(CO₂Me)CH₂COCH₃) was simpliﬁed to PdꢀCOCH₃. The red arrow shows the lowest energy pathway.

factors: (i) steric repulsion between the migrating acetyl group

and the methoxycarbonyl group should disfavor 1,2-insertion,^56a

(ii) electron-withdrawing nature of the methoxycarbonyl group

makes a diﬀerence in LUMO orbital (CdC, 2p_z) coeﬃcients to

some extent, which strengthen the 2,1-insertion preference,^56a

and (iii) the energies required for the distortion of MA through

2,1-insertion from MA-coordinated species are smaller compared

to those of 1,2-insertion.^56b,57aIn fact, a smaller distortion energy

for the 2,1-insertion was also found inthe caseof MA insertinginto

the Pdꢀacetyl bond (see Supporting Information). In contrast, t-

butyl acrylate underwent both 2,1- and 1,2-insertion (Section 2)

because the steric repulsion between the ligand and the t-butox-

ycarbonyl group was large enough to compete with the factors

above (iꢀiii).

generally reﬂects the order of the stability of the products (E_trans

E_cis> A_trans> A_cis(the most stable)).

>

According to the results obtained in the simpliﬁed model, 2,1-

insertion of MA from D is suggested to be the most preferable

pathway. Thus, we calculated the route D_cisfTS(D_cisꢀD_trans)f

D

_transfTS(D_transꢀA_cisnext) f A_cisnext using full-size PdCOCH-

(CO₂Me)CH₂COCH₃complexes (Scheme 10). It should be

noted that the transition state for MA coordination to complexes

C_cisor C_transcould not be located possibly because of nearly ﬂat

potential energy surface.⁵⁹We assume that the barrier of the

coordination is negligible compared to the subsequent insertion

barrier.^12f,14,57Since the insertion of MA, TS(D_transꢀA_cisnext),

needs 3.9/27.2 kcal/mol from complex A_cis, this acrylate inser-

tion has the highest barrier through the catalytic cycle from A_cis

to A_cisnext, and makes it the rate-determining step.

It was also found that the insertion from *D_transis more

favorable than that from *D_cis(i.e., TS(*D_transꢀA_cis) and TS-

(*D_transꢀE_cis) are lower in energy than TS(*D_cisꢀA_trans) and

TS(*D_cisꢀE_trans), respectively). The preference is due to the

stronger trans eﬀect of the phosphine moiety than the sulfonate

group; the acetyl group in *D_transis more activated toward

migratory insertion.^14,16,43In fact, the PdꢀC bond length in

*D_trans(2.028 Å) is longer than that in *D_cis(1.993 Å). This

tendency suggests that the MA insertion into the Pdꢀacyl bond

can be recognized as acyl anion migration to MA. Moreover, it was

found that the order of the relative energy of these transition states

Similarly, 2,1-insertion of MA into a Pdꢀacyl bond of Pd

dppe complexes was also examined (Scheme 11). By using the

full-size PdCOCH(CO₂Me)CH₂COMe complex, 2,1-insertion

of MA could proceed with a barrier of 11.3/33.6 kcal/mol from

A_PPvia D_PPfTS(D_PP-A_PPnext)fA_PPnext. The MA insertion

is, again, a rate-determining step because this has the highest

energy in the catalytic cycle from A_PPto A_PPnext. In addi-

tion, the MA insertion TS for the Pd dppe complex is less

favorable than that for the Pd phosphineꢀsulfonate complexes

by 7.4/6.4 kcal/mol.

6771

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

Scheme 10. MA Insertion into PdꢀAcyl Bond of Pd PhosphineꢀSulfonate Complexes (EþZPE/G, kcal/mol)^a

^aR = CH(CO₂Me)CH₂COMe.

Scheme 11. MA Insertion into PdꢀAcyl Bond of Pd dppe

Complexes (EþZPE/G, kcal/mol)^a

7. Discussion on the Mechanism: Why Phosphineꢀ

Sulfonate? According to the discussion above, the most prob-

able pathway for the full catalytic cycle of MA/CO copolymer-

ization by the Pd phosphineꢀsulfonate complex is summarized

in Figure 11 (bold line).⁶²First, CO coordinates to the five-

membered chelate complex A_cisto form carbonyl complex B_cis

and, after cis/trans isomerization, CO insertion takes place to

generate C_ciswhich was observed in high pressure NMR as 5a in

Section 5. After the coordination of MA to C_cisand cis/trans

isomerization, exclusive 2,1-insertion takes place from complex

D_transto regenerate the five-membered chelate complex A_cisnext.

The rate-determining step of the catalytic cycle is the MA

insertion step via TS(D_transꢀA_cisnext), which is 27.2 kcal/mol

higher in energy than A_cis. The double insertion of either MA or

CO⁶¹was energetically unfavorable.

The dotted line in Figure 11 shows the catalytic cycle for

the Pd dppe system. Similarly, CO coordination and insertion

takes place from the ﬁve-membered chelate complex A_PPto aﬀord

acylpalladium complex C_PP. Then, MA coordination and insertion

through the highest barrier via TS(D_PPꢀA_PPnext) (33.6 kcal/mol

from A_PP) complete the catalytic cycle.

^aR = CH(CO₂Me)CH₂COMe.

6.4. Possible Side Reactions.³⁴In the presence of CO, double

olefin insertion is generally disfavored because coordination of CO

and subsequent insertion after olefin insertion is much faster than

the second olefin insertion.^{1,7c,52,60,61}In our experimental studies,

the reaction of MA with CO catalyzed by the Pd phosphineꢀ

sulfonate complexes afforded completely alternating copolymers

without any signal corresponding to double insertion of MA by

NMR analysis, as discussed in Section 2. However, the Pd phos-

phineꢀsulfonate catalyst is a potent catalyst capable of producing

nonalternating ethylene/CO copolymers with excess ethylene con-

tents by the coordinationꢀinsertion mechanism.^15,16Recently,

Caporaso and Mecking experimentally and theoretically demon-

strated that double (or more) insertion of MA can take place in the

absence of CO.^12d,fThus, it is important to understand why double

MA insertion does not occur when Pd phosphineꢀsulfonate

complexes are employed in the presence of CO.

With the full catalytic cycle in hand (Figure 11), the experi-

mental results were interpreted in detail. The obvious diﬀerence

between the phosphineꢀsulfonate system and dppe system is

recognized in the diﬀerence of the highest barriers, TS-

(D_transꢀA_cisnext) vs TS(D_PPꢀA_PPnext). This is consistent with

the experimental fact that the Pd phosphineꢀsulfonate complex

3a copolymerized MA with CO while the Pd dppe complex 3c

did not.

From complex A_cis(after insertion of one MA), another MA

can coordinate cis to the phosphine (F_trans), which is endother-

mic/endergonic by 15.7/27.2 kcal/mol. Migratory insertion of the

alkyl groups trans to the phosphine (TS(F_transꢀG_cis)) requires an

overall barrier of 25.5/38.3 kcal/mol (Scheme 12). This is

consistent with the overall barrier for multiple coordinationꢀ

insertion of MA reported by Caporaso and Mecking (ca. 25 kcal/

mol in ΔE).^12fThis route is thought to be the most probable

pathway according to their detailed calculations. The barrier

relative to A_cisis higher in energy than that of the alternating

copolymerization cycle which requires an overall barrier of

3.9/27.2 kcal/mol (TS(D_transꢀA_cisnext), Scheme 10). Therefore,

MA double insertion did not occur because it is kinetically

disfavored in the presence of CO. Concerning the nonalternating

copolymerization of ethylene with CO, it was reported that

ethylene coordination into the ﬁve-membered chelate complex

[PꢀO]PdCH₂CH₂COEt is exothermic (ꢀ7.5 kcal/mol in ΔH)

and successive insertion can take place with a barrier of 25.0 kcal/

mol (in ΔH) from the chelate complex.^16aThus, diﬀerence in the

ability of double insertion in the presence of CO can be attributed

to the lower binding aﬃnity of the electron-deﬁcient MA com-

pared to that of ethylene.^28,56

Concerning the high-pressure NMR experiments in Section 5,

CO insertion observed from the Pd phosphineꢀsulfonate complex

3a could be attributed to the fact that the barriers for CO

coordination (TS(A_cisꢀB_cis)) and insertion (TS(B_transꢀC_cis))

are reasonably low to be overcome at ambient temperature and

that the resulting acylpalladium complexes are stable enough to be

observed.⁵⁴In contrast, when Pd dppe complex 3c was exposed to 6

MPa of CO at ambient temperature,³⁸no new signals were detected.

This result could be explained as follows (i) acylpalladium com-

plexes were not detected because the barrier for CO insertion

(TS(B_PPꢀC_PP)) cannot be overcome at ambient temperature

(kinetic argument), and (ii) carbonyl complexes were also not

observed since B_PPis less stable than A_PPby 5.6 kcal/mol (which

corresponds to K < 0.1), although the barrier for the coordination,

TS(A_PPꢀB_PP), is easy to overcome (thermodynamic argument).

Before our ﬁrst report on the copolymerization of MA with

CO by the Pd phosphineꢀsulfonate system,¹¹several attempts

on this copolymerization had been reported (Scheme 2). It was

proposed that the one possible reason for the inert nature of

the ﬁve-membered chelate complexes^1his the lower nucleo-

philicity of the carbon R to the Pd center as a result of the

electron-withdrawing group.^7b,8dIn fact, the CO insertion from

6772

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

Scheme 12. Double MA Insertion into PdꢀAlkyl Bond of Pd PhosphineꢀSulfonate Complexes (EþZPE/G, kcal/mol)

Figure 11. Comparison of the energy proﬁles for the copolymerization of MA with CO catalyzed by Pd phospineꢀsulfonate complex (Bold black) and

Pd dppe complex (dotted line). The Gibbs free energies (in kcal/mol) relative to A_cisor A_PPare given. R = CH(CO₂Me)CH₂COMe.⁶²

[PꢀO]Pd(CO)CH₂CH₂COCH₃(without ꢀCO₂Me group)

requires only a barrier of 11.5 kcal/mol,^16awhile the CO

insertion from B_transrequires a barrier of 15.6 kcal/mol

(Figure 11). Thus, it is the case that an electron-withdrawing

group on the R-carbon slows down the CO insertion rate and,

thus, the acylpalladium species would not be observed at ambient

temperature (Scheme 6). However, Figure 11 suggests that

coordination and insertion of CO are possible if higher tempera-

ture is provided because of their moderate barriers (13.0 and 21.0

kcal/mol, respectively).³⁸Instead, the subsequent coordina-

tionꢀinsertion of MA requires a rather high barrier (via TS-

(D_PPꢀA_PPnext), 33.6 kcal/mol from A_PP), which would be the

main obstacle for the copolymerization by the Pd dppe complex.

All of these observations raise an important question: why do

the Pd phosphineꢀsulfonate complexes make the copolymeri-

zation possible? The relative energy of the rate-determining MA

insertion is lower for this Pd phosphineꢀsulfonate system

compared to that of the Pd dppe system (Figure 11). This could

6773

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

Figure 12. Calculated structures of the rate-determining step: TS(D_transꢀA_cisnext) and TS(D_PPꢀA_PPnext).

be attributed to the steric repulsion between the bulky Ph groups

on the phosphorus atoms of DPPE and the polymer chain in the

transition state (Figure 12). In contrast, the sulfonate group is

not bulky and, therefore, TS(D_transꢀA_cisnext) is suﬃciently low

in energy to be overcome.

sulfonate system. The regiochemistry of poly(methyl acrylate-

alt-CO) was excellently controlled, even though its methine

carbons were gradually epimerizing in the presence of MeOH.

Terpolymers of MA, ethylene, and CO were also synthesized by

the same catalyst and its ester content could be controlled by

changing the ethylene pressure. The incorporation of the ester

group into γ-polyketones resulted in an increase in glass-transi-

tion temperature.

The second reason for the lower-energy transition state of MA

insertion when using the Pd phosphineꢀsulfonate system is

electronic in nature. Because of the early nature of the MA

insertion TS,⁶³they should have similar structures to precursor

complexes, D_transand D_PP. It was found by comparing the

structure of D_transand D_PPthat the CdC bond of MA in D_trans

is longer than in D_PPand also that the Pdꢀoleﬁn distance is

shorter in D_transthan D_PP(Table 4). The results indicate that the

back-donation from Pd to the electron-deﬁcient oleﬁn moiety is

more eﬃcient in the Pd phosphineꢀsulfonate than in the Pd

dppe. The stronger back-donation should contribute to the

higher stability of D_trans. The lower-energy TS(D_transꢀAnext)

can also be explained by the fact that the oleﬁn moiety of MA is

more activated by the stronger back-donation from Pd to

π*(CdC) of MA. Thus, the longer carbonꢀcarbon bond length

of the bound MA facilitates the conversion from sp²to sp³

hybridization.

It should be noted that most of the intermediates in Figure 11

(except A_cis) and transition states are lower in energy in the Pd

phosphineꢀsulfonate system compared to those of the Pd dppe

system. This is because the back-donation from Pd to CO, acyl

groups, or MA in these intermediates is facilitated when the

sulfonate moiety is located trans to these substituents. As we

discussed in Section 6.1, the main contribution to this stronger

back-donation is not the neutral character of these complexes,

but rather a stereoelectronic eﬀect of the sulfonate moiety. In

other words, the SO₃group does not withdraw the π-electrons

from Pd but, instead, repulsion between lone pairs on the ligated

oxygen and π-electrons of Pd facilitates the back-donation to

CO, acyl group or MA, which is located trans to the sulfonate

moiety. Only the intermediate A_ciscannot be stabilized by back-

donation because of a lack of appropriate low-lying vacant

orbitals of the alkyl group.

The mechanism for the alternating copolymerization of MA

with CO catalyzed by the Pd phosphineꢀsulfonate complexes is

summarized in Figure 13. Reversible CO insertion to isolated

ﬁve-membered chelate complex 3a was observed by high-pres-

sure NMR study. Next, exclusive 2,1-insertion of MA takes place

to regenerate the ﬁve-membered chelate complex, resulting in

the regiocontrolled architectures of the copolymer. Subsequent

MA insertion is not preferable because of its high energy barrier

and the weak binding ability of MA. DFT calculation showed that

the rate-determining step is MA insertion into the Pd-acyl bond

activated by the trans phosphine ligand. It was found by

comparison with the conventional Pd dppe system that the

barrier of MA insertion for the Pd phosphineꢀsulfonate system

was decreased for the following two reasons: (i) the sulfonate

group is less hindered than the phosphine group and (ii) a

coordinating MA is more activated due to the stronger back-

donation from a Pd center. The enhanced back-donation can be

mainly attributed to the weak π-acceptor character of the

sulfonate group and the electronic repulsion between lone pairs

of the sulfonate oxygens and π-electrons of a Pd center.

Throughout this study, the role of this simple and elegantly

designed unsymmetrical ligand has been revealed by an under-

standing of its molecular orbitals and their eﬀect on the Pd

center.^64,65We believe that these detailed studies will allow the

further design of novel catalysts.

’ EXPERIMENTAL SECTION

General Methods. All manipulations were carried out using

standard glovebox or Schlenk techniques under argon purified by

passing through a hot column packed with BASF catalyst R3ꢀ11.

NMR spectra were recorded on JEOL JNM-ECP500 (¹H: 500 MHz,

²H: 77 MHz, ¹³C: 126 MHz, ³¹P: 202 MHz with digital resolution of

0.239, 0.141 Hz, 0.960, 4.33 Hz, respectively) or JEOL JNM-ECS400

(¹H: 400 MHz, ¹³C: 101 MHz, ³¹P: 162 MHz with digital resolution of

’ CONCLUSIONS

In this article we provided the full details for the copolymer-

ization of acrylates with CO catalyzed by a Pd phosphineꢀ

6774

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

Figure 13. Summary of the mechanism of the copolymerization of methyl acrylate with CO catalyzed by a Pd phosphineꢀsulfonate complex and the

role of sulfonate moiety.

0.09125, 0.767, 3.46 Hz, respectively) NMR spectrometers. Size exclu-

sion chromatography (SEC) analyses were carried out at 40 °C with

a GL Sciences instrument (model PU 610 high-performance-liquid-

chromatography pump, CO 631A liquid chromatography column oven,

and RI 713 refractive-index detector) equipped with two columns

(Shodex KF-804 L) by eluting the columns with chloroform at 1 mL/min.

Molecular weights were determined using narrow polystyrene standards.

Fast atom bombardment mass spectrometry (FAB-MS) was carried out

on a JEOL JMS-700 spectrometer using PEG calibration and NBA

matrix solvent. Differential scanning calorimetry (DSC) analysis was

performed on a SII EXTAR DSC-7020, with a heating rate of 10 K/min

from ꢀ50 to 160 °C. Thermogravimetric analysis (TGA) of the

polymers was measured on a SII EXTAR6000 TG/DTA-6200,

with heating rate of 10 K/min form 30 to 500 °C. X-ray crystallograph-

ic analyses were performed on a Rigaku Mercury CCD diffractometer.

Materials Information. Anhydrous solvents (dichloromethane,

diethyl ether, hexane, and tetrahydrofuran) were purchased from Kanto

Chemical Co. Inc. and purified by the method of Pangborn et al.⁶⁶

Carbon monoxide (>99.95 vol%) was obtained from Takachiho Che-

mical Industrial Co. Ethylene (>99.9%) was purchased from Takachiho

Chemical Industrial Co., Ltd., dried, and deoxygenated by passing

through columns. Methyl acrylate and t-butyl acrylate were purchased

from Kanto Chemical Co. Inc. and purified by vacuum-transfer over

CaH₂and stored under dark, argon, at ꢀ20 °C. 1,1,1,3,3,3-hexafluoro-2-

propanol (Kanto) and methan-(ol-d) (Aldrich, 99 atom %-D) were

purchased and used as received.

transferred to a round-bottom flask with CH₂Cl₂and volatile materials

were removed by a rotary evaporator. The remaining solid was dried

under vacuum to afford the copolymer. Yield of the copolymer were

determined by subtraction of the weight of catalyst (10.7 mg) from the

amount of solid product obtained. Molecular weight of the copolymer

was determined after filtration of insoluble materials in CHCl₃. Repre-

cipitation and filtration of the crude materials twice from CH₂Cl₂/

MeOH or once from CH₂Cl₂/diethyl ether afforded dba-free alternating

copolymer.

¹H NMR (CDCl₃, 500 MHz, Figure S1, top): δ 3.04ꢀ3.54 (ꢀCH₂ꢀ),

3.60ꢀ3.95 (ꢀCO₂CH₃), 3.96ꢀ4.11 (ꢀCH(CO₂Me)ꢀ).

¹³C NMR (CDCl₃, 126 MHz, Figure S2, top): δ 40.2ꢀ41.1 (ꢀCH₂ꢀ),

52.4ꢀ52.8 (ꢀCO₂CH₃), 52.9 (ꢀCH(CO₂Me)ꢀ), 168.1ꢀ168.4

(ꢀCO₂CH₃), 201.0ꢀ201.4 (ꢀCOꢀ).

Procedures of Copolymerization of t-Butyl Acrylate with

Carbon Monoxide (Table 1, entry 11). Pd(dba)₂and ligand 1a

(1:1.2 mol ratio) were mixed in a mortar with a pestle. The mixed

catalyst (10.7 mg, 0.010 mmol of Pd) were dried in vacuo in a 50 mL-

autoclave more than 1.5 h before introducing 1 atm of carbon monoxide.

t-Butyl acrylate (2.5 mL) was then added to the autoclave and the

mixture was stirred at 0 °C for 5 min and charged with carbon monoxide

at 6.0 MPa. The autoclave was heated to 70 °C and the mixture was

stirred. After 20 h, the cooled contents of the autoclave were transferred

to a round-bottom flask with CH₂Cl₂and volatile materials were

removed by a rotary evaporator. The remaining solid was dried under

vacuum to afford the copolymer. Yield of the copolymer were deter-

mined by subtraction of the weight of catalyst (10.7 mg) from the

amount of solid product obtained. Molecular weight of the copolymer

was determined after filtration of insoluble materials in CHCl₃. Removal

of ligand and dba by reprecipitation failed because of the solubility of the

copolymer in MeOH and in diethyl ether.

The following compounds were prepared according to literature

procedures: Pd(dba)₂,¹⁷Pd₂(dba)₃CHCl₃,¹⁷2-{di(2-methoxyphe-

3

nyl)phosphino}benzenesulfonic acid (1a),^15a2-(diphenylphosphino)-

benzenesulfonic acid (1b),^13b[{o-(o-MeOC₆H₄)₂P}C₆H₄SO₃]PdMe-

(2,6-Me₂C₅H₃N) (2a),^12I[{o-((o-MeOC₆H₄)₂P)C₆H₄SO₃}PdCH-

(CO₂Me)CH₂COMe] (3a),¹¹[(dppe)PdMe(NCMe)][OTf].⁶⁷

General Procedures of Copolymerization of Methyl Acry-

late with Carbon Monoxide (Table 1, entry 1). Pd(dba)₂and

ligand 1a (1:1.2 mol ratio) were mixed in a mortar with a pestle. The

mixed catalyst (10.7 mg, 0.010 mmol of Pd) were dried in vacuo in a

50 mL-autoclave more than 1.5 h before introducing 1 atm of carbon

monoxide. Methyl acrylate (2.5 mL) were then added to the autoclave

and the mixture was stirred at 0 °C for 5 min and charged with carbon

monoxide at 6.0 MPa. The autoclave was heated to 70 °C and the

mixture was stirred. After 20 h, the cooled contents of the autoclave were

¹H NMR (CDCl₃, 400 MHz, Figure S5): δ 1.37ꢀ1.55 (ꢀC(CH₃)₃),

2.65ꢀ3.52 (ꢀCH₂ꢀ), 3.69ꢀ4.39 (ꢀCH(CO₂t-Bu)ꢀ).

¹³C NMR (CDCl₃, 101 MHz, Figure S6): δ 27.5ꢀ28.4 (ꢀC(CH₃)₃),

39.2ꢀ41.8 (ꢀCH₂ꢀ), 53.0ꢀ55.8 (ꢀCH(CO₂t-Bu)ꢀ), 81.7ꢀ83.2

(ꢀC(CH₃)₃), 166.3ꢀ167.2 (ꢀCH(CO₂t-Bu)ꢀ), 168.1ꢀ168.4

(ꢀCO₂C(CH₃)₃), 200.8ꢀ202.6 (ꢀCOꢀ).

General Procedures of Terpolymerization of Methyl Acry-

late, Ethylene and Carbon Monoxide Catalyzed by Pd(dba)₂

and PhosphoniumꢀSulfonate (Table 2). Pd(dba)₂and ligand 1a

(1:1.2 mol ratio) were mixed in a mortar with a pestle. The mixed

6775

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

catalyst (10.7 mg, 0.010 mmol of Pd) were dried in vacuo in a 50 mL-

autoclave more than 1.5 h. To the autoclave were added 2.5 mL of

methyl acrylate and then was charged with ethylene quickly at described

pressure. Carbon monoxide was quickly charged to the autoclave up to

the described total pressure. The autoclave was heated to 70 °C and the

mixture was stirred. After 1 h, the cooled contents of the autoclave were

transferred to a round-bottom flask with CH₂Cl₂and volatile materials

were removed by a rotary evaporator. The remaining solid was dried

under vacuum to afford the terpolymer. Yield of the copolymer were

determined by subtraction of the weight of catalyst (10.7 mg) from the

amount of solid product obtained. For the reprecipitaion of the

terpolymer in entry 1 in Table 2, the obtained materials were dissolved

in ca. 1 mL of a mixture of CDCl₃/1,1,1,3,3,3-hexafluoro-2-propanol

(1:2) was poured into stirring diethyl ether through filtration. The

reprecipitated white solid was used for NMR (Figure S10ꢀ14, Support-

ing Information), TG (Figure S21, Supporting Information) and DCS

(Figure S22, Supporting Information) analyses. For the terpolymer in

entry 2 in Table 2 (NMR: Figure S15, 16, Supporting Information),

molecular weight of the copolymer was determined after filtration of

insoluble materials in CHCl₃.

Preparation of [(dppe)PdCH(CO₂Me)CH₂COMe][OTf] (3c).

Under ambient CO pressure, 2.15 g of [(dppe)PdMe(NCMe)][OTf]

(3.03 mmol) was dissolved in 30 mL of dichloromethane, and the

reaction mixture was stirred for 1 h at rt. To the resulting mixture,

0.80 mL (8.88 mmol, 2.9 equiv) of methyl acrylate in 10 mL of

dichloromethane was added and stirred for 5 h at rt. The resulting

mixture was filtered through Celite, and the filter cake was washed with

CH₂Cl₂. The condensed solution (ca. 40 mL) was poured into diethyl

ether (ꢀ25 °C) to afford yellow oil. (993 mg, 1.27 mmol, 41.9% yield)

The white crystal suitable for X-ray analysis was obtained from THF.;

mp: 96ꢀ99 °C dec; IR (KBr) cm^ꢀ1: 1674, 1622, 1437, 1263, 1157,

1039, 637; ¹H NMR (CD₂Cl₂, 400 MHz, Figure S19, Supporting

Information): δ1.85ꢀ2.00 (m, 1H, PdCH(CO₂CH₃)(CHHC(O)CH₃)),

2.56ꢀ2.87 (m, 4H, PdCH, PdCH(CO₂CH₃)(CHHC(O)CH₃),

PCH₂CH₂P), 2.63 (s, 3H, PdCH(CO₂CH₃)(CH₂C(O)CH₃)), 3.07

(dd, J = 18.3, 6.8 Hz, 1H, PCH₂CHHP), 3.28 (dd, J = 18.7, 14.1 Hz,

1H, PCH₂CHHP), 3.14 (s, 3H, CO₂CH₃), 7.49ꢀ7.81 (m, 20H); ¹³C

NMR (CD₂Cl₂, 101 MHz, Figure S20, top, Supporting Information): δ

20.8 (dd, ³J_PC= 30, 5 Hz, PdCH(CO₂CH₃)(CH₂C(O)CH₃)), 28.4 (d,

⁴J_PC= 2 Hz, PdCH(CO₂CH₃)(CH₂C(O)CH₃)), 31.0 (dd, ¹J_PC= 37 Hz,

²J_PC= 19 Hz, PCH₂CH₂P), 48.1 (d, ¹J_PC= 77 Hz, PdCH), 50.1 (d, ¹J_PC= 6

Hz, PCH₂CH₂P), 50.3 (s, CO₂CH₃), 126.1 (d, ¹J_PC= 58 Hz, 4°), 127.4 (d,

¹J_PC= 58 Hz, 4°), 128.2 (d, ¹J_PC= 40 Hz, 4°), 128.7ꢀ129.6 (m, CH and

4°), 131.8ꢀ132.4 (m, CH), 132.7 (d, J_PC= 2 Hz, CH), 134.0 (d, J_PC= 13

H/D Exchange of Copoly(Methyl Acrylate-alt-CO) (Figure

S3, Supporting Information). In a NMR sample tube, 16 mg of

copoly(methyl acrylate-alt-CO) (0.14 mmol unit, a sample of M_n=

53,000, M_w/M_n= 1.6) was dried under vacuum. After introducing argon,

freshly distilled CD₂Cl₂(0.6 mL) was added to the sample tube and the

polymer was dissolved. To a solution of copolymer was added 2.1 mmol

of methan-(ol-d). After 4 h at ambient temperature, ¹H NMR was

3

Hz, CH), 176.7 (d, J_PC= 6 Hz, PdCH(CO₂CH₃)(CH₂C(O)CH₃)),

3

238.7 (d, J_PC= 9 Hz, PdCH(CO₂CH₃)(CH₂C(O)CH₃)); ³¹P NMR

(CD₂Cl₂, 162 MHz) δ 45.4 (d, ²J_PP= 26 Hz), 59.2 (d, ²J_PP= 26 Hz); ¹⁹

F

2

collected. H NMR of the resulting copolymer was also collected in

NMR (CD₂Cl₂, 376 MHz) δ ꢀ78.8; HRMS-FAB (m/z): [M ꢀ

CF₃SO₃

]

^{ꢀ þ}calcd for C₃₂H₃₃O₃P₂Pd, 633.0940; found, 633.0916.

CH₂Cl₂after removing CD₂Cl₂and methan-(ol-d) under vaccuo.

Reaction of 3a under High Pressure of CO: Spectropcopic

Characterization of (SP-4-3)-[{o-((o-MeOC₆H₄)₂P)C₆H₄SO₃}-

Pd{COCH(CO₂Me)CH₂COMe}] (5a).⁶⁸In a globe box, a high

pressure-NMR tube was charged with a filtered solution of 3a (0.25

mmol) in CD₂Cl₂(0.6 mL). The tube was charged with CO to 6.0 MPa

and was kept at room temperature for 3.5 h. Then the NMR data were

collected at room temperature. The 3:2 mixture of 5a and 3a were

observed. (Figure S17, Supporting Information).

Reaction of 3c under High Pressure of CO. In a globe box, a

high pressure-NMR tube was charged with a filtered solution of 3c (0.14

mmol) in CD₂Cl₂(0.6 mL). The tube was charged with CO to 6.0 MPa

and was kept at room temperature for 3.5 h. Then the NMR data were

collected at room temperature (Figure S20, bottom, Supporting In-

formation). Other than free CO, any change could not be confirmed, at

least in carbonyl region.

Theoretical Methods. All geometries of the reactants, intermediates,

transition states, and products of the reactions were fully optimized by using

Gaussian 03 program,⁶⁹and B3LYP⁴⁰method with combined basis sets

(Lanl2dz basis sets and effective core potential for palladium atom⁴¹and

6-31G(d) basis sets for the other atoms). Harmonic vibration frequency

calculations at the same level were performed to verify all stationary points

as local minimum (with no imaginary frequency) or transition state (with

one imaginary frequency). IRC calculations⁷⁰were also performed to

confirm the connectivity between the transition state and the reactant/

product. Zero-point energy, enthalpy, and Gibbs free energy at 298.15 K

and 1 atm were calculated from the gas-phase harmonic frequencies.

NMR data (other than aromatic region) of 3a¹¹: ¹H NMR (CD₂Cl₂):

δ 1.66ꢀ1.79 (m, 1H, PdCH), 2.47 (s, 3H, PdCH(CO₂CH₃)(CH₂C-

(O)CH₃)), 2.59 (d, J = 18.3 Hz, 1H, PdCH(CO₂CH₃)(CH₂C(O)CH₃)),

2.78 (dd, J = 18.3, 6.0 Hz, 1H, PdCH(CO₂CH₃)(CH₂C(O)CH₃)), 3.26

(s, 3H, PdCH(CO₂CH₃)(CH₂C(O)CH₃)), 3.50 (s, 3H, CH₃OC₆H₄),

3.71 (s, 3H, CH₃OC₆H₄), 6.89 (dd, J = 8.3, 4.6 Hz, 1H), 6.98ꢀ7.07 (m,

2H), 7.14 (dddd, J = 7.6, 7.6, 1.8, 0.9 Hz, 1H), 7.24ꢀ7.37 (m, 3H),

7.41ꢀ7.49 (m, 1H), 7.52ꢀ7.62 (m, 2H), 7.94ꢀ8.07 (m, 1H), 8.27 (dd,

J = 16.3, 7.3 Hz, 1H); ¹³C NMR (CD₂Cl₂): δ 27.8 (s, PdCH(CO₂CH₃)-

(CH₂C(O)CH₃)), 34.6 (s, PdCH), 50.4 (s, PdCH(CO₂CH₃)(CH₂C-

(O)CH₃)), 50.5 (s, PdCH(CO₂CH₃)(CH₂C(O)CH₃)), 54.5 (s,

CH₃OC₆H₄), 55.3 (s, CH₃OC₆H₄), 176.9 (s, PdCH(CO₂CH₃)(CH₂C-

(O)CH₃)), 230.9 (s, PdCH(CO₂CH₃)(CH₂C(O)CH₃)).; ³¹P NMR

(CD₂Cl₂): δ 19.7.

’ ASSOCIATED CONTENT

S

Supporting Information. Copies of all spectral data, full

b

1

5a: H NMR (CD₂Cl₂, Figure S17, 18, Supporting Information):

characterization, and details of theoretical part including

Cartesian coordinates and complete citation of ref 69. This material

is available free of charge via the Internet at http://pubs.acs.org.

δ 2.10 (s, 3H, PdCOCH(CO₂CH₃)(CH₂C(O)CH₃)), 2.65ꢀ2.74 (m,

1H, PdCOCH(CO₂CH₃)(CH₂C(O)CH₃)), 2.96ꢀ3.06 (m, 1H,

PdCOCH(CO₂CH₃)(CH₂C(O)CH₃)), 3.88ꢀ3.92 (m, 1H, PdCOCH-

(CO₂CH₃)(CH₂C(O)CH₃)), 3.97 (s, 3H, PdCOCH(CO₂CH₃)(CH₂C-

(O)CH₃)), 3.60 (s, 3H, CH₃OC₆H₄), 3.64 (s, 3H, CH₃OC₆H₄).; ¹³C

NMR (CD₂Cl₂): δ 28.7 (s, PdCOCH(CO₂CH₃)(CH₂C(O)CH₃)), 40.6

(s, PdCOCH(CO₂CH₃)(CH₂C(O)CH₃)), 54.8 (s, CH₃OC₆H₄), 55.1 (s,

CH₃OC₆H₄), 55.6 (br, PdCOCH(CO₂CH₃)(CH₂C(O)CH₃)), 63.7 (d,

³J_PC= 15.0 Hz, PdCOCH(CO₂CH₃)(CH₂C(O)CH₃)), 181.3 (br,

’ AUTHOR INFORMATION

Corresponding Author

nozaki@chembio.t.u-tokyo.ac.jp

Present Addresses

^‡Department of Chemistry, Faculty of Science and Technology,

Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama, Kanagawa

223-8522, Japan.

2

PdCOCH(CO₂CH₃)(CH₂C(O)CH₃), 202.4 (d, J_PC= 16.1 Hz,

PdCOCH(CO₂CH₃)(CH₂C(O)CH₃)), 205.3 (s, PdCOCH(CO₂CH₃)-

(CH₂C(O)CH₃)).; ³¹P NMR (CD₂Cl₂): δ 19.9 (br).

6776

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

’ ACKNOWLEDGMENT

(e) Braunstein, P.; Frison, C.; Morise, X. Angew. Chem., Int. Ed. 2000,

39, 2867–2870. (f) Braunstein, P.; Durand, J.; Knorr, M.; Strohmann, C.

Chem. Commun. 2001, 211–212. (g) Reddy, K. R.; Surekha, K.; Lee,

G. H.; Peng, S. M.; Chen, J. T.; Liu, S. T. Organometallics 2001,

20, 1292–1299. (h) Agostinho, M.; Braunstein, P. Chem. Commun.

2007, 58–60. (i) Agostinho, M.; Braunstein, P. C. R. Chimie 2007,

10, 666–676. (j) Hamada, A.; Braunstein, P. Organometallics 2009,

28, 1688–1696.

(8) Reactions of vinyl acetate with CO: (a) Williams, B. S.; Leatherman,

M. D.; White, P. S.; Brookhart, M. J. Am. Chem. Soc. 2005, 127, 5132–5146.

See also refs 7d and 7g. Reaction of vinyl chloride with CO:(b) Shen, H.;

Jordan, R. F. Organometallics 2003, 22, 1878–1887. Reaction of acryloni-

trile with CO:(c) Wu, F.; Foley, S. R.; Burns, C. T.; Jordan, R. F. J. Am.

Chem. Soc. 2005, 127, 1841–1853. Reaction of vinyl ethers with CO:(d)

Chen, C. L.; Chen, Y. C.; Liu, Y. H.; Peng, S. M.; Liu, S. T. Organometallics

2002, 21, 5382–5385 .

(9) Coordination copolymerizations of these common polar vinyl

monomers are one of the most challenging issues in polymer chem-

istry, see: (a) Boﬀa, L. S.; Novak, B. M. Chem. Rev. 2000, 100, 1479–

1493. (b) Sen, A.; Borkar, S. J. Organomet. Chem. 2007, 692, 3291–3299.

(c) Berkefeld, A.; Mecking, S. Angew. Chem., Int. Ed. 2008, 47, 2538–

2542. (d) Chen, E. Y. X. Chem. Rev. 2009, 109, 5157–5214. (e) Ito, S.;

Nozaki, K. Chem. Rec. 2010, 10, 315–325. See also ref 1h.

We are grateful to Assoc. Prof. Y. Nishibayashi and Dr. Y.

Miyake at the University of Tokyo for high-resolution FAB-MS

analysis of 3c. We are also grateful to Prof. T. Kato and Dr. T.

Yasuda at the University of Tokyo for preliminary XRD analyses.

We thank Dr. Zhuofeng Ke at Kyoto University for fruitful

discussion on theoretical part. Computer resources for theore-

tical calculation were mainly provided by the Research Center for

Computational Science in the National Institutes of Natural

Sciences. This work was supported by the Global COE Program

for Chemistry Innovation. A.N. is grateful to the Japan Society for

the Promotion of Science (JSPS) for a Research Fellowship for

Young Scientists. L.W.C. acknowledges the Fukui Institute

Fellowship. This work is in part supported by Japan Science

and Technology Agency (JST) with a Core Research for Evolu-

tional Science and Technology (CREST) grant in the Area of

High Performance Computing for Multiscale and Multiphysics

Phenomena. Partial support by Funding Program for Next

Generation World-Leading Researchers, Green Innovation,

JSPS, is also acknowledged.

(10) Kochi, T.; Nakamura, A.; Ida, H.; Nozaki, K. J. Am. Chem. Soc.

2007, 129, 7770–7771.

’ REFERENCES

(11) Nakamura, A.; Munakata, K.; Kochi, T.; Nozaki, K. J. Am. Chem.

Soc. 2008, 130, 8128–8129.

(1) (a) Drent, E.; Budzelaar, P. H. M. Chem. Rev. 1996, 96, 663–681.

(b) Sommazzi, A.; Garbassi, F. Prog. Polym. Sci. 1997, 22, 1547–1605.

(c) Bianchini, C.; Meli, A. Coord. Chem. Rev. 2002, 225, 35–66.

(d) Bianchini, C.; Meli, A.; Oberhauser, W. Dalton Trans. 2003, 2627–

2635. (e) Nakano, K.; Kosaka, N.; Hiyama, T.; Nozaki, K. Dalton Trans.

2003, 4039–4050. (f) Durand, J.; Milani, B. Coord. Chem. Rev. 2006,

250, 542–560. (g) García Suꢀarez, E. J.; Godard, C.; Ruiz, A.; Claver, C.

Eur. J. Inorg. Chem. 2007, 2582–2593. (h) Nakamura, A.; Ito, S.; Nozaki,

K. Chem. Rev. 2009, 109, 5215–5244.

(2) Catalytic Synthesis of Alkene-Carbon Monoxide Copolymers and

Cooligomers; Sen, A., Ed.; Kluwer Academic Publishers: Dordrecht,

2003.

(3) For examples, see: (a) European Plastics News; 1995, Oct, 57.

(b) Yanai, T.; Okajima, S.; Miyaji, K. Jpn. Kokai Tokkyo Koho

2008,075,219, April 3, 2008.

(4) For examples, see: (a) Milani, B.; Crotti, C.; Farnetti, E. Dalton

Trans. 2008, 4659–4663. (b) Kosaka, N.; Hiyama, T.; Nozaki, K.

Macromolecules 2004, 37, 4484–4487. (c) Nozaki, K.; Kosaka, N.;

Graubner, V. M.; Hiyama, T. Macromolecules 2001, 34, 6167–6168.

(d) Sen, A.; Jiang, Z. Z.; Chen, J. T. Macromolecules 1989, 22, 2012–

2014.

(5) (a) Sen, A. S.; Jiang, Z. Macromolecules 1993, 26, 911–915.

(b) Kacker, S.; Jiang, Z.; Sen, A. Macromolecules 1996, 29, 5852–5858.

(c) Klok, H. A.; Eibeck, P.; Schmid, M.; Abu-Surrah, A. S.; M€oller, M.;

Rieger, B. Macromol. Chem. Phys. 1997, 198, 2759–2768. (d) Nieuwhof,

R. P.; Marcelis, A. T. M.; Sudh€olter, E. J. R.; Wursche, R.; Rieger, B.

Macromol. Chem. Phys. 2000, 201, 2484–2492. (e) Lee, J. T.; Alper, H.

Chem. Commun. 2000, 2189–2190. (f) Moineau, C.; Mele, G.; Alper, H.

Can. J. Chem. 2001, 79, 587–592. (g) Fujita, T.; Nakano, K.; Yamashita,

M.; Nozaki, K. J. Am. Chem. Soc. 2006, 128, 1968–1975. (h) Malinova,

V.; Rieger, B. Macromol. Rapid Commun. 2005, 26, 945. (i) Malinova, V.;

Rieger, B. Biomacromolecules 2006, 7, 2931–2936.

(6) Functionalized bicyclic oleﬁns are also copolymerized with

CO, see: (a) Liaw, D. J.; Tsai, J. S.; Sang, H. C. J. Polym. Sci., Part A:

Polym. Chem. 1998, 36, 1785–1790. (b) Saﬁr, A. L.; Novak, B. M. J. Am.

Chem. Soc. 1998, 120, 643–650.

(7) Reactions of methyl acrylate with CO: (a) Ozawa, F.; Hayashi,

T.; Koide, H.; Yamamoto, A. J. Chem. Soc., Chem. Commun.

1991, 1469–1470. (b) Dekker, G. P. C. M.; Elsevier, C. J.; Vrieze, K.;

van Leeuwen, P. W. N. M.; Roobeek, C. F. J. Organomet. Chem. 1992,

430, 357–372. (c) Rix, F. C.; Brookhart, M.; White, P. S. J. Am. Chem.

Soc. 1996, 118, 4746–4764. (d) Reddy, K. R.; Chen, C. L.; Liu, Y. H.;

Peng, S. M.; Chen, J. T.; Liu, S. T. Organometallics 1999, 18, 2574–2576.

(12) Methyl acrylate: (a) Drent, E.; van Dijk, R.; van Ginkel, R.; van

Oort, B.; Pugh, R. I. Chem. Commun. 2002, 744–745. (b) Kochi, T.;

Yoshimura, K.; Nozaki, K. Dalton Trans. 2006, 25–27. (c) Skupov,

K. M.; Marella, P. R.; Simard, M.; Yap, G. P. A.; Allen, N.; Conner, D.;

Goodall, B. L.; Claverie, J. P. Macromol. Rapid Commun. 2007,

28, 2033–2038. (d) Guironnet, D.; Roesle, P.; R€unzi, T.; G€ottker-

Schnetmann, I.; Mecking, S. J. Am. Chem. Soc. 2009, 131, 422–423.

(e) Skupov, K. M.; Hobbs, J.; Marella, P.; Conner, D.; Golisz, S.;

Goodall, B. L.; Claverie, J. P. Macromolecules 2009, 42, 6953–6963.

(f) Guironnet, D.; Caporaso, L.; Neuwald, B.; G€ottker-Schnetmann, I.;

Cavallo, L.; Mecking, S. J. Am. Chem. Soc. 2010, 132, 4418–4426.

Acrylic acid:(g) R€unzi, T.; Fr€ohlich, D.; Mecking, S. J. Am. Chem. Soc.

2010, 132, 17690–17691. Vinyl acetate:(h) Ito, S.; Munakata, K.;

Nakamura, A.; Nozaki, K. J. Am. Chem. Soc. 2009, 131, 14606–14607.

Acrylonitrile:(i) Kochi, T.; Noda, S.; Yoshimura, K.; Nozaki, K. J. Am.

Chem. Soc. 2007, 129, 8948–8949. (j) Nozaki, K.; Kusumoto, S.; Noda,

S.; Kochi, T.; Chung, L. W.; Morokuma, K. J. Am. Chem. Soc. 2010,

132, 16030–16042. Vinyl Fluoride:(k) Weng, W.; Shen, Z.; Jordan, R. F.

J. Am. Chem. Soc. 2007, 129, 15450–15451. (I) Shen, Z. L.; Jordan, R. F.

Macromolecules 2010, 43, 8706–8708. Vinyl Ethers: (m) Luo, S.; Vela, J.;

Lief, G. R.; Jordan, R. F. J. Am. Chem. Soc. 2007, 129, 8946–8947. Other

vinyl polar monomers: (n) Skupov, K. M.; Piche, L.; Claverie, J. P.

Macromolecules 2008, 41, 2309–2310. (o) Borkar, S.; Newsham, D. K.;

Sen, A. Organometallics 2008, 27, 3331–3334. (p) Bouilhac, C.; Runzi,

T.; Mecking, S. Macromolecules 2010, 43, 3589–3590. See also ref 1h.

Allyl monomers: (q) Ito, S.; Kanazawa, K.; Munakata, K.; Kuroda, J.;

Okumura, Y.; Nozaki, K. J. Am. Chem. Soc. 2011, 133, 1232–1235.

(13) For some early patents, see:(a) Murray, R. E. U.S. Patent

4,689,437, August 25, 1987. (b) van Doorn, J. A.; Drent, E.; van

Leeuwen, P. W. N.; Meijboom, N.; van Oort, A. B.; Wife, R. L., Eur.

Pat. Appl. 0,280,380, August 31, 1988. (c) Bradford, A. M.; Van

Leeuwen, P. W. N.; Wullink-Schelvis, A. M. WO Patent Application

9324553, December 9, 1993. See also ref 1h.

(14) For the mechanism of the formation of the linear polyethylene

catalyzed by Pd phosphineꢀsulfonate complexes, see: (a) Noda, S.;

Nakamura, A.; Kochi, T.; Chung, L. W.; Morokuma, K.; Nozaki, K. J. Am.

Chem. Soc. 2009, 131, 14088–14100. Very recently, Jordan proposed

another possible cis/trans isomerization mechanism: (b) Conley, M. P.;

Jordan, R. F. Angew. Chem. Int. Ed. 2011, 50, 3744–3746.

(15) (a) Drent, E.; van Dijk, R.; van Ginkel, R.; van Oort, B.; Pugh,

R. I. Chem. Commun. 2002, 964–965. (b) Hearley, A. K.; Nowack,

6777

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

R. A. J.; Rieger, B. Organometallics 2005, 24, 2755–2763. (c) Newsham, D. K.;

Borkar, S.; Sen, A.; Conner, D. M.; Goodall, B. L. Organometallics 2007,

26, 3636–3638. (d) Bettucci, L.; Bianchini, C.; Claver, C.; Suarez, E. J. G.;

Ruiz, A.; Meli, A.; Oberhauser, W. Dalton Trans. 2007, 5590–5602. (e) Luo,

R.; Newsham, D. K.; Sen, A. Organometallics 2009, 28, 6994–7000.

(16) Theoretical calculations of nonalternating copolymerization

of ethylene with CO: (a) Haras, A.; Michalak, A.; Rieger, B.; Ziegler,

T. J. Am. Chem. Soc. 2005, 127, 8765–8774. (b) Haras, A.; Michalak, A.;

Rieger, B.; Ziegler, T. Organometallics 2006, 25, 946–953.

is used than that of DPPE (ref 1a). However, we chose DPPE for the

discussion because of the decreased degrees of freedom of the ligand for

the theoretical study. Note that the calculated energy required for the

rate-determining step was 38.9 kcal/mol by utilizing DPPP as a ligand,

which is even higher than that of DPPE. See section 7.

(33) Drent et. al. patented the terpolymerization of MA/ethylene/

CO by using the combination of Pd(OAc)₂and DPPE, see: Drent, E.

Eur. Pat. Appl. 0,272,727, June 29, 1988.

(34) Our calculation suggested that the relative stabilities of the

product derived from homolytic cleavage of PdꢀC bond of A_cisand A_PP

are 55.5/39.0 and 58.2/40.6 kcal/mol respectively. See also ref 9b for the

discussion of homolytic cleavage of PdꢀCH(CO₂Me)R bond.

(35) Jordan and co-workers described in ref 8b that ﬁve-membered

chelate complexes with R-Cl substituent ([NꢀN]PdCHClCH₂COCH₃)

were inert to the insertion of CO. The PdꢀC and PdꢀO bond distances,

determined by X-ray diﬀraction, are at the short end of the ranges

observed in the analogous non-halogenated ﬁve-membered chelate

complexes.

(17) When Pd₂(dba)₃CHCl₃was used, the activity and the mo-

3

lecular weight were enhanced (activity, 3.0 g mmol^ꢀ1

h

^ꢀ1; M_n, 64 000;

3

M_w/M_n, 4.3). ForthepreparationofPd(dba)₂andPd₂(dba)₃CHCl₃, see:

Ukai, T.; Kawazura, H.; Ishii, Y.; Bonnet, J. J.; Ibers, J. A. J. Organomet.

3

Chem. 1974, 65, 253–266.

(18) Slight amount of Pd black was observed after the reaction. See

also ref 39.

(19) The complex 2a can copolymerize ethylene and MA with low

MA ratio (unpublished resuls, see also references^12dfor a result by a

pyridine adduct complex). Considering the copolymerization of MA

with CO proceeds in alternating fashion, coordination of MA should

occur in each catalytic cycle. Thus, MA/CO copolymerization should be

more sluggish than ethylene/CO copolymerization with low MA ratio.

(20) Recently, the reactivity of methyl methacrylate with Pd phosphi-

neꢀsulfonate complex was studied in detail, see: Runzi, T.; Guironnet,

D.; G€ottker-Schnetmann, I.; Mecking, S. J. Am. Chem. Soc. 2010,

132, 16623–16630.

(21) Hauptman, E.; Saboetienne, S.; White, P. S.; Brookhart, M.;

Garner, J. M.; Fagan, P. J.; Calabrese, J. C. J. Am. Chem. Soc. 1994,

116, 8038–8060.

(22) Nagashima, H.; Itonaga, C.; Yasuhara, J.; Motoyama, Y.;

Matsubara, K. Organometallics 2004, 23, 5779–5786.

(36) It is widely known that OH^ꢀligand is classiﬁed as weak trans

eﬀect/inﬂuence ligand, see: (a) The Organometallic Chemistry of the

Transition Metals, 4th ed; Crabtree, R, Ed.; Wiley: New York, 2005. (b)

Organotransition Metal Chemistry From Bonding to Catalysis Hartwig, J.,

Ed.; University Science Books: Herndon, VA, 2009.

(37) The bulkier o-CH₃OC₆H₄group on phosphorus atom in 3c

makes this comparison imperfect.

(38) Because 2D NMR spectra were not applicable under the high

pressure conditions, these signals were characterized by comparisons to

the literature. Note that our high-pressure NMR apparatus is not

applicable at higher temperature.

(39) While taking ¹³C NMR, the palladium complexes gradually

decomposed to result in the formation of some Pd black.

(40) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648–5652. (b) Lee,

C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785–789.

(23) As often observed in other γ-polyketones such as poly(propylene-

alt-CO), spiroketal repeat units instead of ketone units can be also

generated. After slow evaporation for 4 month from the CDCl₃solution

of the poly(methyl acrylate-alt-CO), ¹³C NMR showed broad signal at

110.6ꢀ115.5 ppm which is typical region for spiro quarternary carbon.

(24) Nozaki, K.; Sato, N.; Takaya, H. J. Am. Chem. Soc. 1995,

117, 9911–9912.

(41) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270–283.

(42) Comparison of B3LYP/6-31G* with Lanl2dz and B3LYP/

6-311þG** with SDD þf is described in Supporting Information.

Because the copolymerization was conducted under neat condition,

PCM calculation was not performed to avoid inappropriate discussion.

Largely, B3LYP/6-31G* with Lanl2dz in gas phase suﬃciently match

with our experiments. See also ref 14.

1

(25) Enol form of the β-ketoester was not detected from the H

NMR spectrum.

(43) For example of the theoretical studies employing other unsymme-

trical bidentate ligands in coordinationꢀinsertion polymerizations, see: (a)

Nozaki, K.; Sato, N.; Tonomura, Y.; Yasutomi, M.; Takaya, H.; Hiyama,

T.; Matsubara, T.; Koga, N. J. Am. Chem. Soc. 1997, 119, 12779–12795.

(b) Nozaki, K.; Komaki, H.; Kawashima, Y.; Hiyama, T.; Matsubara, T.

J. Am. Chem. Soc. 2001, 123, 534–544. (c) Deubel, D. V.; Ziegler, T.

Organometallics 2002, 21, 4432–4441. (d) Michalak, A.; Ziegler, T.

Organometallics 2003, 22, 2069–2079. (e) Yang, S. Y.; Szabo, M. J.;

Michalak, A.; Weiss, T.; Piers, W. E.; Jordan, R. F.; Ziegler, T. Organo-

metallics 2005, 24, 1242–1251.

(26) This epimerization rate was similar to that of the corres-

ponding small molecule: when methyl-2-methyl-3-oxopentanoate

(EtCOCH(CO₂Me)CH₃) was treated with 15 equivalents of MeOD

in CD₂Cl₂, 25% of the methine protons were exchanged with deuterium

after 4 h at room temperature.

(27) If the stereochemistry of poly(methyl acrylate-alt-CO) is

thermodynamically controlled and the split signals at low temperature

are originated from the ﬂuctuation of higher-order structure, it is still

unclear whether the regiochemistry of poly(t-butyl acrylate-alt-CO) is

controlled or not.

(44) The alkyl chains of B_cisand B_transhave diﬀerent conformations

because we adopted the most stable structures for these intermediates

(See Supporting Information). For example, internal ketone carbonyl

oxygen coordinates to the Pd center at apical position in B_transwhile not

in B_cisto avoid the steric repulsion with Ph groups. It should be noted

that B_ciswas less stable in energy than B_transregardless of the ketone

coordination.

(28) For the comparison of the reactivity of MA and ethylene, see:

(a) Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart, M. J. Am. Chem.

Soc. 1998, 120, 888–899. (b) Kang, M. S.; Sen, A.; Zakharov, L.;

Rheingold, A. L. J. Am. Chem. Soc. 2002, 124, 12080–12081. (c)Popeney,

C. S.; Guan, Z. B. J. Am. Chem. Soc. 2009, 131, 12384–12393. (d) Srebro,

M.; Mitoraj, M.; Michalak, A. Can. J. Chem. 2009, 87, 1039–1054 and

references cited therein. See also refs 56 and 57.

(45) (a) Dahlenburg, L.; Vondeuten, K.; Kopf, J. J. Organomet. Chem.

1981, 216, 113–127. (b) Manojlovic-Muir., L. J.; Muir, K. W. Inorg.

Chim. Acta 1974, 10, 47–49. See also ref 16.

(46) Dias, P. B.; Depiedade, M. E. M.; Simoes, J. A. M. Coord. Chem.

Rev. 1994, 135, 737–807.

(29) Recently, we reported terpolymerization of styrene/vinylacetate/

CO and styrene/methyl acrylate/CO by using the same catalyst, see:

Kageyama, T.; Ito, S.; Nozaki, K. Chem. Asian J. 2011, 6, 690–697.

(30) AbuSurrah, A. S.; Wursche, R.; Rieger, B. Macromol. Chem.

Phys. 1997, 198, 1197–1208.

(47) Although a σ* orbital of the SꢀO single bond might be in the

suitable position to overlap the ﬁlled d_πorbitals of a Pd center, the

coeﬃcient of its vacant orbital should be mainly on the S atom. In

addition, such an interaction was not found in the molecular orbitals

from HOMO-5 to LUMOþ10 of B_trans. Instead, the repulsion between

lone pairs on the oxygen atom of a sulfonate group and d_πorbitals of a Pd

(31) However, preliminary XRD analyses showed tiny signals for

poly(MA-alt-CO) and poly(CO-alt-(ethylene; MA)). The results in-

dicate the possibility of being crystalline polymer. Further investigations

for the physical properties are in progress.

(32) It is known that the activity for the copolymerization of

ethylene with CO is higher when diphenylphosphinopropane (DPPP)

6778

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Journal of the American Chemical Society

ARTICLE

center can be conﬁrmed in several molecular orbitals. For a related study

of the orbital of H-SO₃-CCH, see: Stang, P. J.; Crittell, C. M.; Arif, A. M.;

Karni, M.; Apeloig, Y. J. Am. Chem. Soc. 1991, 113, 7461–7470.

(48) It was expected from these calculations that the IR spectrum of

these carbonyl complexes would give us important implication. Thus, we

conducted some experiments to take IR spectrum of carbonyl com-

plexes, however, carbonyl complexes were not isolable and the signals in

IR spectrum under CO atmosphere were too complicated to assign for

both Pd phosphineꢀsulfonate and Pd dppe systems.

(59) The transition state for MA coordination to complexes C_cis-

keto, C_trans-keto, H_cisor H_transcould not be located either. See ref 54.

(60) Shultz, C. S.; Ledford, J.; DeSimone, J. M.; Brookhart, M. J. Am.

Chem. Soc. 2000, 122, 6351–6356.

(61) It is also well-established that the double insertion of CO is

thermodynamically disfavorable. In the case of Pd phosphineꢀsulfonate

catalysts, double CO insertion from Pd(CO)COMe (can be expressed

as *H_trans, see ref 54) requires a barrier of 20.7/20.9 kcal/mol and the

product is 13.9/15.1 kcal/mol above the starting material. Considering

much more stable product after the sequential insertion of CO and MA

(A_cis, ꢀ7.9/ꢀ2.4 kcal/mol on the basis of Pd(CO)COMe), double CO

insertion is unlikely. See also: (a) ref 1. (b) Chen, J. T.; Sen, A. J. Am.

Chem. Soc. 1984, 106, 1506–1507.

(49) (a) Weinhold, F.; Carpenter, J. E. In The Structure of Small

Molecules and Ions; Naaman, R., Vager, Z., Eds.; Plenum: New York,

1988; pp 227ꢀ236. (b) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem.

Rev. 1988, 88, 899–926.

(50) Between B_cisand Bmed, we could ﬁnd TS(B_cisꢀBmed) and it

only requires 1.3 /1.9 kcal/mol from Bmed. We assume that the other

transition states, namely, TS(A_cisꢀAmed), TS(AmedꢀA_trans), and

TS(BmedꢀB_trans) are barrierless or have a very low barrier. In fact,

some energy (EþZPE) plots of single point calculation suggested that

(62) Although B_transand C_cisare lower in energy than A_cis, we

employed A_cisas a standard for the following discussion because the

diﬀerences are quite slight and within the error of the calculation. See

also ref 54.

(63) Hammond, G. J. Am. Chem. Soc. 1955, 77, 334.

there is no barrier between Bmed and B_trans

.

(64) Some studies of novel ligands inspired by phosphineꢀsulfonate

ligands, see: (a) Reisinger, C. M.; Nowack, R. J.; Volkmer, D.; Rieger, B.

Dalton Trans. 2007, 272–278. (b) Nagai, Y.; Kochi, T.; Nozaki, K.

Organometallics 2009, 28, 6131–6134. (c) Zhu, H. J.; Ziegler, T.

Organometallics 2009, 28, 2773–2777.

(65) Some applications of phosphineꢀsulfonate ligands to the

other organic reactions, see: (a) Schultz, T.; Pfaltz, A. Synthesis 2005,

1005–1011. (b) Garcia Suarez, E. J.; Ruiz, A.; Castillꢀon, S.; Oberhauser,

W.; Bianchini, C.; Claver, C. Dalton Trans. 2007, 2859–2861. (c)

Bettucci, L.; Bianchini, C.; Meli, A.; Oberhauser, W. J. Mol. Catal. A:

Chem. 2008, 291, 57. (d) Sundararaju, B.; Achard, M.; Demerseman, B.;

Toupet, L.; Sharma, G. V. M.; Bruneau, C. Angew. Chem., Int. Ed. 2010,

49, 2782–2785.

(66) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;

Timmers, F. J. Organometallics 1996, 15, 1518–1520.

(67) Dekker, G. P. C. M.; Elsevier, C. J.; Vrieze, K.; Vanleeuwen, P.

W. N. M. Organometallics 1992, 11, 1598–1603.

(68) “SP-4-3” represents the stereochemical information of

the square planar complex. See: IUPAC Nomenclature of Inorganic

Chemistry, Recommendations 2005; Blackwell Scientiﬁc Publications:

Oxford, 2005.

(69) Frisch, M. J. et al. Gaussian 03, Revision C02; Gaussian, Inc.:

Wallingford, CT, 2007.

(51) (a) Zuidema, E.; Bo, C.; van Leeuwen, P. W. M. N. J. Am. Chem.

Soc. 2007, 129, 3989–4000. (b) Frankcombe, K. E.; Cavell, K. J.; Yates,

B. F.; Knott, R. B. Organometallics 1997, 16, 3199–3206. (c) Frank-

combe, K. E.; Cavell, K. J.; Yates, B. F.; Knott, R. B. J. Phys. Chem. 1996,

100, 18363–18370. (d) Cross, R. J. Chem. Soc. Rev. 1985, 14, 197–223.

(e) Anderson, G. K.; Cross, R. J. Acc. Chem. Res. 1984, 17, 67–74.

(52) Similar transition states were expected in ethylene/CO

copolymerization, see: (a) Margl, P.; Ziegler, T. J. Am. Chem. Soc.

1996, 118, 7337–7344. (b) Margl, P.; Ziegler, T. Organometallics 1996,

15, 5519–5523.

(53) Other possible pathways, such as a transition state between A_cis

and A_transinvolving CO and a direct insertion of CO at apical position

could not be found.

(54) The six-membered chelate structures with ketone coordination

were also found with slight diﬀerences in energy: C_cis-keto (ꢀ8.4/0.8

kcal/mol), C_trans-keto (2.9/10.6 kcal/mol), TS(C_cis-ketoꢀC_trans

-

keto) (9.0/16.9 kcal/mol), C_PP-keto (ꢀ3.2/5.1 kcal/mol). Further

coordination of CO instead of the internal carbonyl coordination was

also considered (i.e. Pd(CO)[COCH(CO₂Me)CH₂COMe]): H_cis

(C_cisþ CO, ꢀ17.2/0.3 kcal/mol), H_trans(C_transþ CO, ꢀ19.2/ꢀ1.2

kcal/mol), and H_PP(C_PPþ CO, ꢀ12.6/4.5 kcal/mol) (for each

structure, see Supporting Information). Coordination of CO to acyl-

palladium complexes in the reaction media was also suggested in the

copolymerization of ethylene with CO (Mul, W. P.; Oosterbeek, H.;

Beitel, G. A.; Kramer, G. J.; Drent, E. Angew. Chem., Int. Ed. 2000,

39, 1848–1851). Although the calculated relative Gibbs free energies of

(70) (a) Gonzalez, C.; Schlegel, H. B. J. Chem. Phys. 1989, 90,

2154–2161. (b) Gonzalez, C.; Schlegel, H. B. J. Phys. Chem. 1990,

94, 5523–5527.

B

_trans, C_cis-keto, H_cisand H_transare comparable to A_cis(3a) and

C_cis(5a), they were not observed in the experiment in Figure 5. This

inconsistence between experiments and calculation may be attributed to

the less inaccurate estimation of entropy.

(55) The internal ketone coordination to the cationic Pd center from

the apical position can stabilize B_PPand TS(B_PPꢀC_PP). For example,

B_PPbecame unstable for 3.9 kcal/mol without ketone coordination.

This tendency is contrastive to the neutral Pd phosphineꢀsulfonate case

whose stabilities are not so inﬂuenced by the existence of internal ketone

coordination. See also ref 44.

(56) (a) von Schenck, H.; Str€omberg, S.; Zetterberg, K.; Ludwig, M.;

Åkermark, B.; Svensson, M. Organometallics 2001, 20, 2813–2819.

(b) Szabo, M. J.; Jordan, R. F.; Michalak, A.; Piers, W. E.; Weiss, T.;

Yang, S. Y.; Ziegler, T. Organometallics 2004, 23, 5565–5572.

(c) Philipp, D. M.; Muller, R. P.; Goddard, W. A.; Storer, J.; McAdon,

M.; Mullins, M. J. Am. Chem. Soc. 2002, 124, 10198–10210.

(57) (a) Michalak, A.; Ziegler, T. J. Am. Chem. Soc. 2001, 123,

12266–12278. (b) Michalak, A.; Ziegler, T. Organometallics 2003,

22, 2660–2669. (c) Haras, A.; Anderson, G. D. W.; Michalak, A.; Rieger,

B.; Ziegler, T. Organometallics 2006, 25, 4491–4497.

(58) Comparative studies by theoretical calculations of the prefer-

ence between 2,1- and 1,2-insertion with vinyl acetate/CO copolymer-

ization are now in progress. See ref 10.

6779

dx.doi.org/10.1021/ja2003268 |J. Am. Chem. Soc. 2011, 133, 6761–6779

Article Doi

Pd-catalyzed copolymerization of methyl acrylate with carbon monoxide: Structures, properties and mechanistic aspects toward ligand design

DOI: 10.1021/ja2003268

Source and publish data:

Authors:

Article abstract of DOI:10.1021/ja2003268

Full text of DOI:10.1021/ja2003268

Products guided by the article

R&D Labs maybe for 143821-70-9

Relevant to this article

Hot Product