Retention of regiochemistry of allylic esters in palladium-catalyzed allylic alkylation in the presence of a MOP ligand

Article abstract of DOI:10.1021/ja973150r

In the palladium-catalyzed allylic alkylation of (E)-3-substituted-2-propenyl acetates (1), 1-substituted-2-propenyl acetates (2), and 1- or 3-deuterio-2-cyclohexenyl acetate (5), which proceeds through 1,3-unsymmetrically substituted π-allylpalladium intermediates, selective substitution at the position originally substituted with acetate was observed by use of a sterically bulky monodentate phosphine ligand, 2-(diphenylphosphino)-2'-methoxy-1,1'-binaphthyl (MeO-MOP). Studies of the structure of π-allylpalladium complexes generated by mixing [PdCl(π-cyclohexenyl)]₂ with 1 or 2 equiv of MeO-MOP (L*) revealed that cationic bisphosphine complex [Pd(L*)₂(π-cyclohexenyl)]⁺Cl^- is not formed even in the presence of excess ligand but neutral monophosphine complex PdCl(L*)(π-cyclohexenyl) (11) is formed, leaving excess ligand free, and that the exchange of the coordination site of Cl and L* in 11 is much slower than that in triphenylphosphine complex PdCl(PPh₃)(π-cyclohexenyl) (13). The slow exchange can rationalize the retention of regiochemistry in the allylic alkylation catalyzed by palladium/MeO-MOP complex.

Full text of DOI:10.1021/ja973150r

J. Am. Chem. Soc. 1998, 120, 1681-1687
1681
Retention of Regiochemistry of Allylic Esters in Palladium-Catalyzed
Allylic Alkylation in the Presence of a MOP Ligand
Tamio Hayashi,* Motoi Kawatsura, and Yasuhiro Uozumi
Contribution from the Department of Chemistry, Faculty of Science, Kyoto UniVersity, Sakyo,
Kyoto 606-01, Japan
ReceiVed September 8, 1997
Abstract: In the palladium-catalyzed allylic alkylation of (E)-3-substituted-2-propenyl acetates (1), 1-substituted-
2-propenyl acetates (2), and 1- or 3-deuterio-2-cyclohexenyl acetate (5), which proceeds through 1,3-
unsymmetrically substituted π-allylpalladium intermediates, selective substitution at the position originally
substituted with acetate was observed by use of a sterically bulky monodentate phosphine ligand,
2-(diphenylphosphino)-2′-methoxy-1,1′-binaphthyl (MeO-MOP). Studies of the structure of π-allylpalladium
complexes generated by mixing [PdCl(π-cyclohexenyl)]2 with 1 or 2 equiv of MeO-MOP (L*) revealed that
cationic bisphosphine complex [Pd(L*)2(π-cyclohexenyl)]+Cl- is not formed even in the presence of excess
ligand but neutral monophosphine complex PdCl(L*)(π-cyclohexenyl) (11) is formed, leaving excess ligand
free, and that the exchange of the coordination site of Cl and L* in 11 is much slower than that in
triphenylphosphine complex PdCl(PPh3)(π-cyclohexenyl) (13). The slow exchange can rationalize the retention
of regiochemistry in the allylic alkylation catalyzed by palladium/MeO-MOP complex.
Introduction
Scheme 1
Palladium-catalyzed allylic substitution reactions including
catalytic asymmetric reactions have attracted considerable
attention owing to their synthetic utility and mechanistic
interest.1 The allylic substitutions proceed by way of a
π-allylpalladium(II) intermediate which is formed by oxidative
addition of an allylic ester to a palladium(0) species. It has
been generally accepted that, in the allylic substitution which
proceeds through 1,3-unsymmetrically substituted π-allyl-
palladium(II) intermediate, the regiochemistry of starting allylic
ester is lost at the formation of the π-allylpalladium(II)
intermediate and the regiochemistry in the substitution product
is determined at the attack of nucleophile on the π-allylpalladium
(Scheme 1). Similarly, in the asymmetric allylic substitution
which proceeds through a meso type π-allylpalladium(II)
intermediate, the enantioselectivity is usually determined at the
nucleophilic attack on either of diastereotopic π-allyl carbons
on the π-allylpalladium(II) intermediate which is not concerned
with the absolute configuration of starting allylic ester any more.
Recently, Trost found2 an interesting phenomenon that the
absolute configuration of a starting allyl ester has an effect on
the structure of π-allylpalladium intermediate in his asymmetric
alkylation, though the effect is modest. We report here a new
type of palladium-catalyzed allylic alkylation where the regio-
chemistry of the starting allyl esters is retained in the alkylation
products (Scheme 2). The retention of the regiochemistry is
Scheme 2
observed with a sterically bulky monodentate phosphine ligand,
2-(diphenylphosphino)-2′-methoxy-1,1′-binaphthyl (MeO-MOP).3
Results
(1) For a review on catalytic allylic substitutions, see: Tsuji, J. Palladium
Reagents and Catalysts; John Wiley: Chichester, 1995. For reviews on
catalytic asymmetric allylic substitutions, see: (a) Trost, B. M.; Van
Vranken, D. L. Chem. ReV. 1996, 96, 395. (b) Hayashi, T. In Catalytic
Asymmetric Synthesis; Ojima, I., Ed.; VCH: New York, 1993; p 325. (c)
Frost, C. G.; Howarth, J.; Williams, J. M. J. Tetrahedron: Asymmetry 1992,
3, 1089. (d) Consiglio, G.; Waymouth, M. Chem. ReV. 1989, 89, 257.
(2) (a) Trost, B. M.; Bunt, R. C. J. Am. Chem. Soc. 1996, 118, 235. See
also: (b) Fiaud, J. C.; Malleron, J. L. Tetrahedron Lett. 1981, 22, 1399. (c)
Trost, B. M.; Schmuff, N. R. Tetrahedron Lett. 1981, 22, 2999.
Palladium-Catalyzed Allylic Alkylation. In the presence
of 2 mol % palladium catalyst generated in situ by mixing
[PdCl(π-C3H5)]2 with 2-(diphenylphosphino)-2′-methoxy-1,1′-
binaphthyl (MeO-MOP)3, γ-substituted allyl acetates 1 and
(3) (a) Uozumi, Y.; Hayashi, T. J. Am. Chem. Soc. 1991, 113, 9887. (b)
Uozumi, Y.; Tanahashi, A.; Lee, S.-Y.; Hayashi, T. J. Org. Chem. 1993,
58, 1945. (c) Uozumi, Y.; Suzuki, N.; Ogiwara, A.; Hayashi, T. Tetrahedron
1994, 50, 4293.
S0002-7863(97)03150-8 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/04/1998

1682 J. Am. Chem. Soc., Vol. 120, No. 8, 1998
Hayashi et al.
Scheme 3
phenyl-2-propenyl)methylmalonate (4a), was 79/21 (entry 1).
On the other hand, the alkylation of 1-phenyl-2-propenyl acetate
(2a), which is a regioisomeric allyl ester of 1a, catalyzed by
the palladium/MeO-MOP complex under the same reaction
conditions, gave branch isomer 4a preferentially, the ratio of
3a to 4a being 23/77 (entry 2). The same catalytic activity and
regioselectivity were observed in the reaction of 2a catalyzed
by a palladium catalyst consisting of MeO-MOP and palladium
in a ratio of 1/1 (entry 3). The results obtained above for the
reaction of 1a and 2a clearly show that the alkylation took place
preferentially at the position originally substituted with the
leaving acetate. The regiochemistry of starting allyl ester is
retained in the product in the palladium-catalyzed alkylation in
the presence of MeO-MOP ligand.
The retention of regiochemistry observed here is quite unusual
in the palladium-catalyzed allylic substitution reactions. The
allylic substitution of γ-substituted allyl acetates 1 and R-sub-
stituted allyl acetates 2 usually gives products consisting of the
regioisomers in the same ratio, because the π-allylpalladium
intermediate formed by the oxidative addition of 1 and 2 is
considered to be the same. Actually, the alkylation of 1a and
2a in the presence of palladium catalyst coordinated with 1,2-
bis(diphenylphosphino)ethane (dppe) or triphenylphosphine (P/
Pd ) 2/1) gave the alkylation products in the same ratio
irrespective of the regiochemistry of the starting allylic esters
(entries 4-7). It is noteworthy that the reaction catalyzed by
palladium/PPh3 requires 2 equiv (to Pd) of phosphine ligand.
With 1 equiv of phosphine ligand, a deposit of palladium black
was observed and the reaction stops at 66% conversion (entry
8), though the regioselectivity is the same. Similar regiochemi-
cal results were obtained in the allylic alkylation of allyl acetates
substituted at the R or γ position with aryl groups, 4-methoxy-
phenyl (1b and 2b) and 4-chlorophenyl (1c and 2c) (entries
9-16). The palladium/MeO-MOP catalyst gave the alkylation
products in a different ratio of regioisomers while palladium/
PPh3 catalyst gave the products of the same ratio starting from
a pair of regioisomeric allyl acetates, irrespective of the electron-
donating or -withdrawing characters of the aryl substituents.
The retention of regiochemistry was also observed in the
alkylation of methyl-substituted allyl esters 1d and 2d in the
presence of palladium/MeO-MOP catalyst, linear ester 1d giving
linear product 3d with high selectivity and branch ester 2d
giving branch product 4d as a major product (entries 17 and
18). Here again, the regiochemistry was lost in the reaction
catalyzed by palladium complexes of dppe and triphenylphos-
phine (entries 19-22).
The substitution at the carbon originally substituted with
acetate was also observed in the reaction of specifically
deuterated cyclohexenyl acetates 5 (Scheme 3). Thus, the
alkylation of 3-deuterated acetate 5-3-d1 and 1-deuterated acetate
5-1-d1 with sodium salt of dimethyl methylmalonate in the
presence of palladium/MeO-MOP catalyst took place with high
selectivity (83% at 20 °C) at the position originally substituted
with acetate, giving 6a and 7a, respectively (entries 1 and 2 in
Table 2). Deuterium isotope effects were not observed in the
present alkylation. The reaction carried out at 0 °C in the
presence of 0.5 equiv of lithium chloride increased the regio-
selectivity up to 88% (entries 3 and 4). Use of dppe or
triphenylphosphine ligand in place of MeO-MOP gave a 1/1
mixture of 6a and 7a, starting with either 5-3-d1 or 5-1-d1
(entries 5-8), indicating that the regiochemical integrity of
cyclohexenyl acetates 5-3-d1 and 5-1-d1 is lost before the
nucleophilic attack in the case of dppe or PPh3 as a ligand. In
the allylic alkylation with dimethyl malonate catalyzed by
Table 1. Allylic Alkylation of Acetates 1 and 2 with
NaCMe(COOMe)2 Catalyzed by Palladium-Phosphine Complexesa
yieldb (%) ratioc
ligand (ratio P/Pd) of 3 and 4 3/4
entry
substrate
1a (R ) Ph)
1
2
3
4
5
6
7
8
9
(R)-MeO-MOP (2/1)
(R)-MeO-MOP (2/1)
(R)-MeO-MOP (1/1)
dppe (2/1)
dppe (2/1)
PPh3 (2/1)
99
96
95
99
94
99
99
66
99
93
99
99
89
99
88
93
92
94
92
94
91
93
79/21
23/77
23/77
89/11
89/11
91/9
2a (R ) Ph)
2a (R ) Ph)
1a (R ) Ph)
2a (R ) Ph)
1a (R ) Ph)
2a (R ) Ph)
2a (R ) Ph)
PPh3 (2/1)
PPh3 (1/1)
92/8
91/9
1b (R ) 4-MeOC6H4) (R)-MeO-MOP (2/1)
71/29
16/84
75/25
76/24
81/11
35/65
93/7
94/6
95/5
38/62
78/22
76/24
81/19
82/18
10 2b (R ) 4-MeOC6H4) (R)-MeO-MOP (2/1)
11 1b (R ) 4-MeOC6H4) PPh3 (2/1)
12 2b (R ) 4-MeOC6H4) PPh3 (2/1)
13 1c (R ) 4-ClC6H4)
14 2c (R ) 4-ClC6H4)
15 1c (R ) 4-ClC6H4)
16 2c (R ) 4-ClC6H4)
17 1d (R ) Me)
(R)-MeO-MOP (2/1)
(R)-MeO-MOP (2/1)
PPh3 (2/1)
PPh3 (2/1)
(R)-MeO-MOP (2/1)
(R)-MeO-MOP (2/1)
dppe (2/1)
dppe (2/1)
PPh3 (2/1)
18 2d (R ) Me)
19 1d (R ) Me)
20 2d (R ) Me)
21 1d (R ) Me)
22 2d (R ) Me)
PPh3 (2/1)
a All reactions were carried out in THF at 20 °C for 12 h under
nitrogen: THF (1.0 mL), allylic acetate (0.20 mmol), NaCMe(COOMe)2
(0.40 mmol), [PdCl(π-C3H5)]2 (0.002 mmol), and phosphine ligand.
b Isolated yield by silica gel column chromatography. c The ratio was
1
determined by H NMR analysis of the products.
R-substituted allyl acetates 2 were allowed to react with sodium
salt of dimethyl methylmalonate in THF at 20 °C for 20 h
(Scheme 3). The results are summarized in Table 1, which also
contains the data obtained with some other phosphine ligands
for comparison. An interesting new regiochemistry, which has
not been observed before in allylic substitution reactions by way
of 1,3-unsymmetrically substituted π-allylpalladium intermedi-
ates, was found in the allylic alkylation catalyzed by a palladium
complex coordinated with MeO-MOP. Thus, the alkylation of
(E)-3-phenyl-2-propenyl acetate (1a) in the presence of pal-
ladium/MeO-MOP catalyst (P/Pd ) 2/1) gave linear product,
dimethyl ((E)-3-phenyl-2-propenyl)methylmalonate (3a), as a
major product. The ratio of 3a to branch isomer, dimethyl (1-

Regiochemistry of Allylic Esters
J. Am. Chem. Soc., Vol. 120, No. 8, 1998 1683
Table 2. Allylic Alkylation of 2-Cyclohexenyl Acetates 5 with
Nucleophiles, NaCMeE2 and NaCHE2 (E ) COOMe), Catalyzed by
Palladium-Phosphine Complexesa
Scheme 5
yieldb (%) ratioc
entry acetate
Nu
ligand (ratio P/Pd)
of 6 and 7
6/7
1
2
5-3-d1 CMeE2 (R)-MeO-MOP (2/1)
5-1-d1 CMeE2 (R)-MeO-MOP (2/1)
95
85
90
93
80
94
91
91
87
89
83/17
17/83
88/12
12/88
49/51
53/47
51/49
50/50
18/82
45/55
3d 5-3-d1 CMeE2 (R)-MeO-MOP (2/1)
4d 5-1-d1 CMeE2 (R)-MeO-MOP (2/1)
5
6
7
8
5-3-d1 CMeE2 dppe (2/1)
5-1-d1 CMeE2 dppe (2/1)
5-3-d1 CMeE2 PPh3 (2/1)
5-1-d1 CMeE2 PPh3 (2/1)
9
10
5-1-d1 CHE2
5-1-d1 CHE2
(R)-MeO-MOP (2/1)
PPh3 (2/1)
a All reactions were carried out in THF at 20 °C for 12 h under
nitrogen unless otherwise noted: THF (1.0 mL), allylic acetate (0.20
mmol), NaCMe(COOMe)2 or NaCH(COOMe)2 (0.40 mmol), [PdCl(π-
C3H5)]2 (0.002 mmol), and phosphine ligand. b Isolated yield by silica
addition of either allylic ester 1 or 2. It is likely that the
π-allylpalladium intermediate 10 does not have the original
regiochemical characteristics of allylic esters any more. Hence,
it is reasonable that allyl esters 1 and 2 gave the alkylation
products 3 and 4 with the same regioselectivity (see Scheme
1).
1
gel column chromatography. c The ratio was determined by H NMR
analysis of the products. d Carried out at 0 °C in the presence of 0.5
equiv of LiCl.
Scheme 4
On the other hand, the reaction of 8 with 2 equiv (to Pd) of
(R)-MeO-MOP gave neutral monophosphine complex [PdCl-
(π-cyclohexenyl)(MeO-MOP)] (11), leaving one molecule of
MeO-MOP ligand free from palladium (Scheme 5). The same
neutral monophosphine complex 11 was formed by mixing 8
with 1 equiv (to Pd) of MeO-MOP. With MeO-MOP ligand,
the π-allylpalladium cannot accommodate two molecules of
phosphine ligand because of the steric bulkiness of the MeO-
MOP ligand.6 The complex 11 consists of a pair of diastereo-
isomers 11a and 11b in a ratio of 6/1 at 20 °C in CDCl3.
π-Allyl protons were fully assigned by 31P,1H-correlation
spectroscopy (Supporting Information). The protons which have
correlation peaks with 31P signals are assigned as those on the
π-allyl carbons trans to phosphorus (H1 proton of the major
isomer and H3 proton in the minor isomer). The major isomer
is tentatively assigned to be 11a, which has axial chirality R
around the π-allyl-palladium bond axis,7 on the basis of our
structural studies of related PdCl(π-allyl)((R)-MeO-MOP) com-
plexes.8 The unusual high field shift (δ 3.15 ppm) of the H2
proton of the major isomer supports the structure 11a.8 In the
1H 2-D NOESY spectrum of 11 (Figure 1) obtained by mixing
8 with 1 equiv (to Pd) of MeO-MOP in CDCl3 at 20 °C, cross-
peaks arising from exchange were observed between isomers
11a and 11b for allylic protons, H1, H2, and H3, and methoxy
groups on the MeO-MOP ligand. The allylic proton trans to
phosphorus in the major isomer (H1 in 11a: δ 5.02) found a
cross-peak for the allylic proton cis to phosphorus in the minor
isomer (H1 in 11b: δ 4.26), and that cis to phosphorus in the
major isomer (H3 in 11a: δ 3.98) found a cross-peak for that
trans to phosphorus in the minor isomer (H3 in 11b: δ 5.61).
Thus, the isomerization between 11a and 11b is formally
recognized to be trans-cis isomerization, exchange of the
coordination site of the phosphine ligand and chloride. The
rate of isomerization between 11a and 11b was measured by a
palladium/MeO-MOP, the alkylation also took place at the
carbon substituted with acetate (entry 9).
Structure and Isomerization of π-Allylpalladium-Phos-
phine Complexes. It has been reported that π-allylpalladium-
(II) complexes adopt cationic square planar structure on
coordination with a chelating bisphosphine ligand or two
molecules of monophosphine ligand.4 We have examined the
coordination number of the phosphine ligand for the (π-
cyclohexenyl)palladium system (Scheme 4). Addition of 1
equiv of dppe (P/Pd ) 2/1) to [PdCl(π-cyclohexenyl)]2 (8) in
THF or CDCl3 gave cationic bisphosphine complex [Pd(π-
cyclohexenyl)(dppe)]+Cl- (9a) in a quantitative yield. Simi-
larly, a cationic bisphosphine complex, [Pd(π-cyclohexenyl)-
(PPh3)2]+Cl- (9b), was formed selectively on addition of 2 equiv
of PPh3 (P/Pd ) 2/1) to 8. The formation of the cationic
1
bisphosphine complexes was readily assigned by H and 31P
NMR spectroscopic studies.5 1H NMR for the π-cyclohexenyl
moiety showed a symmetric structure in both 9a and 9b (see
the Experimental Section). The palladium-catalyzed allylic
substitution in the presence of dppe or 2 equiv of triphenylphos-
phine should contain cationic π-allylpalladium complex 10
which is a common intermediate formed by the oxidative
(6) The cone angle of the MeO-MOP ligand is calculated to be larger
than 200°. The structure of MeO-MOP on coordination to palladium has
been studied; see: Uozumi, Y.; Kitayama, K.; Hayashi, T.; Yanagi, K.;
Fukuyo, E. Bull. Chem. Soc. Jpn. 1995, 68, 713. See also ref 8.
(7) We propose this nomenclature of absolute configuration for square
planar π-allylmetal complexes containing meso type π-allyl and two
different ligands.
(8) Hayashi, T.; Iwamura, H.; Naito, M.; Matsumoto, Y.; Uozumi, Y.;
Miki, M.; Yanagi, K. J. Am. Chem. Soc. 1994, 116, 775. (b) Hayashi, T.;
Iwamura, H.; Uozumi, Y.; Matsumoto, Y.; Ozawa, F. Synthesis 1994, 526.
(4) For a review on π-allylpalladium complexes, see: Davies, J. A. In
ComprehensiVe Organometallic Chemistry II; Abel, E. W., Stone, F. G.
A., Wilkinson, G., Eds.; Pergamon: Oxford, 1995; Vol. 9, Chapter 6.
(5) For recent relevant papers on NMR studies of π-allylpalladium
complexes, see: (a) Pregosin, P. S.; Salzmann, R.; Togni, A. Organome-
tallics 1995, 14, 842. (b) Trabesinger, G.; Albinati, A.; Feiken, N.; Kunz,
R. W.; Pregosin, P. S.; Tschoerner, M. J. Am. Chem. Soc. 1997, 119, 6315.

1684 J. Am. Chem. Soc., Vol. 120, No. 8, 1998
Hayashi et al.
Figure 1. 2-D NOESY spectrum of PdCl(π-cyclohexenyl)((R)-MeO-
MOP) (11) in CDCl3 at 20 °C showing the π-allyl region. Strong
correlation peaks are observed between HM1 and Hm1, between HM2
and Hm2, and between HM3 and Hm3 (M ) major isomer 11a; (m )
minor isomer 11b).
Figure 2. 1H NMR spectra for PdCl(π-cyclohexenyl)(PPh3) (13)
generated by mixing [PdCl(π-cyclohexyenyl)]2 (8) with triphenylphos-
phine in CDCl3. The ratios of PPh3 to Pd are 0.9/1 in (a) and 1.1/1 in
(b), (c), and (d). Peaks indicated by ∆ are for π-allyl protons of 8, and
those indicated by × are for π-allyl protons of cationic bisphosphine
complex [Pd(π-cyclohexenyl)(PPh3)2]+Cl- (9b).
1
magnetization saturation transfer technique in H NMR.9 The
rate constant, k(11af11b), obtained by saturation of the H1 proton
in 11b at 20 °C in CDCl3 was 0.5 s-1 and the rate constant,
k
(11bf11a), obtained by saturation of the H1 proton in 11a was
3.2 s-1. The isomerization rate in THF-d8 was a little slower
than that in CDCl3, k(11bf11a) being 2.8 s-1. At -15 °C, the
rate constant k(11af11b) was decreased to 0.08 s-1. The isomer-
ization rate was not affected by addition of an excess of MOP
ligand, with MOP/Pd ) 2/1, k(11af11b) being 0.6 s-1 at 20 °C in
CDCl3, which indicates that the isomerization is taking place
intramolecularly, probably by way of σ-allylpalladium inter-
mediates 12 which can undergo the trans-cis isomerization by
bond rotation around the palladium carbon bond axis.
Scheme 6
Addition of 0.9 equiv (to Pd) of triphenylphosphine to [PdCl-
(π-cyclohexenyl)]2 (8) gave monophosphine complex [PdCl-
(π-cyclohexenyl)(PPh3)] (13) and a small amount (0.1 equiv)
of starting 8. The isomerization of 13 is as slow as that of
MeO-MOP analogue 11 in the absence of an excess of PPh3
(Scheme 6). The rate constants for the isomerization, which
were measured by the saturation transfer technique using
exchange between H1 (δ 5.81 in 13a) and H3 (δ 4.14 in 13a)
protons, were 1.4 and 0.08 s-1 at 20 and -15 °C, respectively.
The isomerization was found to be greatly accelerated by
addition of an excess of triphenylphosphine. In the presence
of 0.1 equiv excess of triphenylphosphine, the rate of isomer-
ization was 3.4 s-1 at -15 °C, 40 times faster than that in the
absence of excess triphenylphosphine. At 20 °C, the isomer-
ization is so fast on the NMR time scale that nonequivalent
allylic protons H1 and H3 in 13 appear as very broad signals
(Figure 2). The signal broadening was also observed for the
protons on C4 and C6 carbons of complex 13. At lower
(9) For examples, see: (a) Cho, H.; Iwashita, T.; Ueda, M.; Mizuno,
A.; Mizukawa, K.; Hamaguchi, M. J. Am. Chem. Soc. 1988, 110, 4832. (b)
Breutel, C.; Pregosin, P. S.; Salzmann, R.; Togni, A. J. Am. Chem. Soc.
1994, 116, 4067.

Regiochemistry of Allylic Esters
J. Am. Chem. Soc., Vol. 120, No. 8, 1998 1685
Scheme 7
that the palladium(0) complex coordinated with one molecule
of MeO-MOP, which was generated by addition of sodium
dimethyl malonate to a mixture of [PdCl(π-cyclohexenyl)]2 (8)
and MeO-MOP (P/Pd ) 1/1) in THF, is stable in solution for
days at room temperature. The coordination of a naphthyl group
was found in the palladium(0) complex coordinated with MeO-
MOP which was demonstrated by low field shifts of the protons
at the 7′ and 8′ positions of the MeO-MOP ligand in 1H NMR.11
The catalytic allylic alkylation in the presence of triphenylphos-
phine ligand should contain palladium(0) species coordinated
with two molecules of the phosphine ligand even if the initial
ratio at the generation of the catalyst was P/Pd ) 1/1. A deposit
of palladium black which was observed in the catalytic reaction
(entry 8 in Table 1) increases the ratio of P/Pd to more than 1.
Thus, the catalytic cycle of the allylic alkylation catalyzed by
the palladium/triphenylphosphine system involves a cationic
bisphosphine intermediate [Pd(π-allyl)(PPh3)2]+ (16) or com-
bination of a neutral monophosphine intermediate [PdX(π-allyl)-
(PPh3)] (15) and an excess of the phosphine. The bisphosphine
complex 16 does not have the regiochemical characteristics of
the starting allylic esters, and the monophosphine complex 15
undergoes the fast isomerization in the presence of an excess
of the phosphine which will lose the original regiochemistry.
On the other hand, the palladium/MeO-MOP system involves
only monophosphine intermediate [PdX(π-allyl)(MeO-MOP)]
(14) which does not undergo the fast isomerization even in the
presence of excess ligand.12
Experimental Section
General Procedures. All manipulations were carried out under a
nitrogen atmosphere. Nitrogen gas was dried by passage through P2O5
(Merck, SICAPENT). NMR spectra were recorded on a JEOL JNM-
temperature, the isomerization is slower, and the 1H NMR
spectrum in the presence of 10% excess triphenylphosphine
ligand showed formation of monophosphine complex 13 and
0.1 equiv of cationic bisphosphine complex 9b. The great
acceleration of the isomerization by the addition of triphenyl-
phosphine demonstrates that the fast isomerization of PPh3
complex 13 takes place by an associative mechanism via cationic
bisphosphine complex 9b or a five-coordinated species. Thus,
the difference between MeO-MOP complex 11 and triphenyl-
phosphine complex 13 is that the trans-cis isomerization of
11 is much slower than that of 13 in the presence of an excess
of the phosphine ligand.
1
EX270 spectrometer (270 MHz for H and 109 MHz for 31P), JEOL
1
JNM-AL400 spectrometer (400 MHz for H NMR), or JEOL JNM
LA500 spectrometer (500 MHz for 1H, 125 MHz for 13C, and 202 MHz
for 31P). Chemical shifts are reported in δ (ppm) referenced to an
1
internal SiMe4 standard for H NMR, and to an external 85% H3PO4
standard for 31P NMR. Residual chloroform (δ 77.0 for 13C) was used
as the internal reference for 13C NMR. 1H, 13C, and 31P NMR spectra
were recorded in CDCl3 at 25 °C unless otherwise noted. Preparative
medium-pressure liquid chromatography (MPLC) was performed with
a silica gel prepacked column Si-10 (Kusano).
Materials. THF was dried over sodium-benzophenone ketyl and
distilled prior to use. [PdCl(π-C3H5)]2,13 [PdCl(π-cyclohexenyl)]2,14
(R)-MeO-MOP,3 1-aryl-2-propenyl acetates,15,16 (E)-3-aryl-2-propenyl
Discussion
(11) 1H NMR for palladium(0)/MeO-MOP (THF-d8): δ 6.00 (t, J ) 7.6
Hz, 1H, H7′), 6.22 (d, J ) 7.6 Hz, 1H, H8′), 6.65 (t, J ) 7.6 Hz, 1H, H6′).
By this second coordination the unusual stability of the palladium(0)/MeO-
MOP complex is rationalized. This type of coordination was not observed
in the π-allyl(MeO-MOP)palladium(II) complexes, formed either by the
oxidative addition of allylic esters to palladium(0) or by addition of MOP
ligand to [PdX(π-allyl)]2. The coordination of a biaryl double bond has
been reported in a rhuthenium complex coordinated with MeO-BIPHEP:
Feiken, N.; Pregosin, P. S.; Trabesinger, G.; Scalone, M. Organometallics
1997, 16, 537. The details of the palladium(0)/MeO-MOP complex will be
described elsewhere.
(12) Another mechanism, for example, involving palladium-carbon bond
formation by reductive elimination of σ-allyl and malonate bonded to
palladium, is excluded by the net retention of stereochemistry in the catalytic
allylic alkylation catalyzed by Pd/MeO-MOP as well as other palladium
catalysts bearing phosphine ligands. Thus, the reaction of cis-3-acetoxy-
5-carbomethoxy-1-cyclohexene with the sodium salt of dimethyl malonate
in the presence of the palladium/MeO-MOP catalyst proceeded with net
retention of configuration to give dimethyl cis-(5-carbomethoxy-
1-cyclohexen-3-yl)malonate: Trost, B. M.; Verhoeven, T. R. J. Am. Chem.
Soc. 1980, 102, 4730.
The retention of the regiochemistry in the catalytic alkylation
of cyclohexenyl acetates 5 in the presence of MeO-MOP ligand
(Scheme 5) must be related to the slow isomerization of the
π-allylpalladium intermediates 14 coordinated with MeO-MOP
ligand (Scheme 7). It is reasonable that the nucleophilic attack
takes place selectively on either of the π-allyl carbons C1 and
C3, most probably on the carbon trans to the phosphine ligand
because of its stronger trans influence than acetate or chloride.10
Provided that the oxidative addition of 5-3-d1 and 5-1-d1 to a
palladium(0) species coordinated with MeO-MOP takes place
selectively in forming cis-14 and trans-14, respectively, the slow
isomerization between cis-14 and trans-14 can rationalize the
retention of the regiochemistry observed in the catalytic
alkylation. Attempts to isolate and characterize π-allylpalladium
intermediates 14 formed by oxidative addition of 5-3-d1 or 5-1-
d1 to a Pd(0)/MeO-MOP species before trans-cis isomerization
are not successful because the isomerization is not so slow that
they are characterized before the isomerization. It is noteworthy
(13) Auburn, P. R.; Mackenzie, P. B.; Bosnich, B. J. Am. Chem. Soc.
1985, 107, 2033.
(14) Trost, B. M.; Strege, P. E.; Weber, L.; Fullerton, T. J.; Dietsche, T.
J. J. Am. Chem. Soc. 1978, 100, 3407.
(10) Hayashi, T.; Kawatsura, M.; Uozumi, Y. Chem. Commun. 1997,
561.
(15) Lehmann, J.; Lloyd-Jones, G. C. Tetrahedron 1995, 52, 8863.

1686 J. Am. Chem. Soc., Vol. 120, No. 8, 1998
Hayashi et al.
acetates,15,17 and 3-deuterio-2-cyclohexen-1-ol18 were prepared accord-
ing to the reported procedures.
Hz, 1H), 7.19-7.34 (m, 5H). Anal. Calcd for C15H18O4: C, 68.69;
H, 6.92. Found: C, 68.55; H, 7.01.
1
Preparation of Allylic Acetates 5-1-d1 and 5-3-d1. 1-Deuterio-
2-cyclohexenyl Acetate 5-1-d1. To a solution of LiAlD4 (1.12 g, 26.7
mmol) in ether (60 mL) was slowly added a solution of 2-cyclohexen-
1-one (5.03 g, 52.3 mmol) in ether (20 mL) at 0 °C. The reaction
mixture was stirred at room temperature for 2 h and diluted with ether.
Addition of NaSO4‚10H2O followed by filtration through a pad of Celite
and evaporation of the solvent gave a quantitative yield of crude
1-deuterio-2-cyclohexen-1-ol. To a solution of this crude alcohol,
pyridine (8.43 mL, 104 mmol), and a catalytic amount of 4-(dimethyl-
amino)pyridine in ether (50 mL) was added acetic anhydride (12.3 mL,
130 mmol). The mixture was stirred at room temperature for 12 h,
quenched with water, and extracted with ether. The organic phase was
washed with 10% CuSO4 solution, water, and brine, dried over
anhydrous magnesium sulfate, and evaporated. The residue was
chromatographed on silica gel (hexane/EtOAc ) 5/1) to give 5.81 g
(79%) of 1-deuterio-2-cyclohexenyl acetate (5-1-d1): 1H NMR (CDCl3,
500 MHz) δ 1.62-2.03 (m, 6H), 2.05 (s, 3H), 5.23-5.27 (m, 1H),
5.70 (m, 1H); MS m/z 141 (M+, 10), 99 (81), 79 (100).
3-Deuterio-2-cyclohexenyl Acetate (5-3-d1). To a solution of
3-deuterio-2-cyclohexen-1-ol18 (3.13 g, 31.9 mmol), pyridine (3.7 mL,
46 mmol), and a catalytic amount of 4-(dimethylamino)pyridine in ether
(30 mL) was added acetic anhydride (5.8 mL, 61 mmol). The mixture
was stirred at room temperature for 12 h, quenched with water, and
extracted with ether. The organic phase was washed with 10% CuSO4
solution, water, and brine, dried over anhydrous magnesium sulfate,
and evaporated. The crude product was purified by silica gel column
chromatography (hexane/EtOAc ) 5/1) to give 4.46 g (99%) of
3-deuterio-2-cyclohexenyl acetate (5-3-d1): 1H NMR (CDCl3, 500 MHz)
δ 1.62-2.03 (m, 6H), 2.05 (s, 3H), 5.69-5.72 (m, 1H), 5.94-5.98
(m, 1H); MS m/z 141 (M+, 5), 99 (51), 80 (100).
Dimethyl (1-Phenyl-2-propenyl)methylmalonate (4a): H NMR
(CDCl3, 270 MHz) δ 1.43 (s, 3H), 3.62 (s, 3H), 3.72 (s, 3H), 4.10 (d,
J ) 8.6 Hz, 1H), 5.11 (d, J ) 16.8 Hz, 1H), 5.12 (d, J ) 10.0, 1H),
6.32 (ddd, J ) 8.6, 10.0, 16.8 Hz, 1H), 7.18-7.34 (m, 5H). Anal.
Calcd for C15H18O4: C, 68.69; H, 6.92. Found: C, 68.70; H, 6.95.
Dimethyl ((E)-3-(4-Chlorophenyl)-2-propenyl)methylmalonate
(3c): 1H NMR (CDCl3, 500 MHz) δ 1.45 (s, 3H), 2.75 (d, J ) 7.3 Hz,
2H), 3.73 (s, 6H), 6.07 (dt, J ) 15.6, 7.3 Hz, 1H), 6.39 (d, J ) 15.6
Hz, 1H), 7.25 (s, 4H). Anal. Calcd for C15H17O4Cl: C, 60.71; H,
5.77. Found: C, 60.57; H, 5.93.
Dimethyl (1-(4-Chlorophenyl)-2-propenyl)methylmalonate (4c):
1H NMR (CDCl3, 500 MHz) δ 1.42 (s, 3H), 3.63 (s, 3H), 3.70 (s, 3H),
4.11 (d, J ) 8.8 Hz, 1H), 5.09 (d, J ) 17.1 Hz, 1H), 5.15 (d, J ) 10.3,
1H), 6.26 (ddd, J ) 8.8, 10.3, 17.1 Hz, 1H), 7.22 (d, J ) 8.7 Hz, 2H),
7.25 (d, J ) 8.7 Hz, 2H). Anal. Calcd for C15H17O4Cl: C, 60.71; H,
5.77. Found: C, 60.57; H, 5.89.
Dimethyl ((E)-2-Butenyl)methylmalonate (3d): 1H NMR (CDCl3,
500 MHz) δ 1.38 (s, 3H), 1.65 (dd, J ) 1.0, 6.9 Hz, 3H), 2.54 (d, J )
7.3 Hz, 2H), 3.71 (s, 6H), 5.29 (qdt, J ) 1.0, 15.1, 7.3 Hz, 1H), 5.51
(dq, J ) 15.1, 6.9 Hz, 1H). Anal. Calcd for C10H16O4: C, 59.98; H,
8.05. Found: C, 59.80; H, 8.33.
1
Dimethyl (1-Methyl-2-propenyl)methylmalonate (4d): H NMR
(CDCl3, 500 MHz) δ 0.98 (d, J ) 6.8 Hz, 3H), 1.29 (s, 3H), 2.93 (dq,
J ) 8.3, 6.8 Hz, 1H), 3.63 (s, 3H), 3.64 (s, 3H), 4.93 (d, J ) 10.7 Hz,
1H), 5.00 (d, J ) 18.1 Hz, 1H), 5.70 (ddd, J ) 8.3, 10.7, 18.1 Hz,
1H). Anal. Calcd for C10H16O4: C, 59.98; H, 8.05. Found: C, 59.71;
H, 8.00.
Palladium-Catalyzed Allylic Alkylation of 2-Cyclohexenyl
Acetates 5-1-d1 and 5-3-d1. The reaction conditions and results are
shown in Table 2. The ratio of regioisomers 6 and 7 was determined
Palladium-Catalyzed Allylic Alkylation of 1 and 2. The reaction
conditions and results are shown in Table 1. A typical procedure is
given for the reaction of 1-(4-methoxyphenyl)-2-propenyl acetate (2b)-
. To a solution of [PdCl(π-C3H5)]2 (0.90 mg, 0.0025 mmol) and (R)-
MeO-MOP (5.1 mg, 0.011 mmol) in THF (0.1 mL) was added a
solution of the sodium salt of dimethyl methylmalonate prepared from
dimethyl methylmalonate (73 mg, 0.50 mmol) and sodium hydride in
THF (1.0 mL). Allyl acetate 2b (56 mg, 0.27 mmol) was added, and
the mixture was stirred at 20 °C for 12 h. The catalyst was removed
by filtration through a short silica gel pad (ether). The filtrate was
evaporated, and the residue was chromatographed on silica gel (hexane/
EtOAc ) 6/1) to give 74 mg (99%) of a mixture of dimethyl ((E)-3-
(4-methoxyphenyl)-2-propenyl)methylmalonate (3b) and dimethyl (1-
(4-methoxyphenyl)-2-propenyl)methylmalonate (4b). The ratio of 3b
to 4b was determined to be 16/84 by 1H NMR. Analytically pure
samples of 3b and 4b were obtained by MPLC (hexane/EtOAc ) 6/1).
Dimethyl ((E)-3-(4-Methoxyphenyl)-2-propenyl)methylmalonate
1
by H NMR studies of the mixture of 6 and 7.
Dimethyl (3-Deuterio-2-cyclohexenyl)methylmalonate (6a): 1H
NMR (CDCl3, 500 MHz) δ 1.26-1.32 (m, 1H), 1.34 (s, 3H), 1.51-
1.62 (m, 2H), 1.78-1.81 (m, 1H), 1.95-1.98 (m, 2H), 3.04 (m, 1H),
3.71 (s, 3H), 3.72 (s, 3H), 5.48 (br s, 1H).
Dimethyl (1-Duterio-2-cyclohexenyl)methylmalonate (7a): 1H
NMR (CDCl3, 500 MHz) δ 1.26-1.31 (m, 1H), 1.34 (s, 3H), 1.51-
1.62 (m, 2H), 1.77-1.81 (m, 1H), 1.95-1.98 (m, 2H), 3.71 (s, 3H),
3.72 (s, 3H), 5.48 (m, 1H), 5.78 (m, 1H).
Dimethyl (3-Deuterio-2-cyclohexenyl)malonate (6b): 1H NMR
(CDCl3, 500 MHz) δ 1.34-1.43 (m, 1H), 1.54-1.61 (m, 1H), 1.70-
1.80 (m, 2H), 1.97-2.04 (m, 2H), 2.88-2.94 (m, 1H), 3.29 (m, 1H),
3.74 (s, 6H), 5.53 (br s, 1H).
Dimethyl (1-Deuterio-2-cyclohexenyl)malonate (7b): 1H NMR
(CDCl3, 500 MHz) δ 1.34-1.43 (m, 1H), 1.54-1.61 (m, 1H), 1.70-
1.80 (m, 2H), 1.97-2.04 (m, 2H), 3.73 (s, 3H), 3.74 (s, 3H), 5.12 (m,
1H), 5.78 (m, 1H).
1
(3b): H NMR (CDCl3, 500 MHz) δ 1.59 (s, 3H), 2.74 (dd, J ) 7.8,
NMR Study of [Pd(π-cyclohexenyl)(dppe)]+Cl- (9a). In an NMR
sample tube were placed dppe (11.3 mg, 0.028 mmol) and [PdCl(π-
cyclohexenyl)]2 (6.3 mg, 0.014 mmol). The tube was filled with
nitrogen, and CDCl3 (0.5 mL) was added. 1H NMR and 31P NMR
spectra were measured at 25 °C: 1H NMR (CDCl3, 500 MHz, 25 °C)
δ 1.06-1.08 (m, 1H), 1.22-1.31 (m, 3H), 2.15-2.20 (m, 2H), 2.57-
2.64 (m, 2H), 3.02-3.06 (m, 2H), 5.81 (t, J ) 6.8 Hz, 1H), 5.97 (br d,
JH-P ) 4.9 Hz, 2H), 7.29-7.59 (m, 20H); 31P{1H} NMR (CDCl3, 202
MHz, 25 °C) δ 46.7.
NMR Study of [Pd(π-cyclohexenyl)(PPh3)2]+Cl- (9b). In an NMR
sample tube were placed PPh3 (11.8 mg, 0.045 mmol) and [PdCl(π-
cyclohexenyl)]2 (4.9 mg, 0.011 mmol). The tube was filled with
nitrogen, and CDCl3 (0.5 mL) was added. 1H NMR and 31P NMR
spectra were measured at -30 °C: 1H NMR (CDCl3, 500 MHz, -30
°C) δ 0.89-1.58 (m, 6H), 5.15 (br s, 2H), 6.09 (br s, 1H), 7.24-7.54
(m, 30H); 31P{1H} NMR (CDCl3, 202 MHz, -30 °C) δ 22.2.
Isolation of [PdCl(π-cyclohexenyl)(MeO-MOP)] (11a). A solution
of (R)-MeO-MOP (21.2 mg, 0.045 mmol) and [PdCl(π-cyclohexenyl)]2
(10.1 mg, 0.023 mmol) in benzene (0.8 mL) was placed in a small
open bottle (5 mL), and the bottle was placed in a reagent bottle (25
mL) which contained ether (3 mL). After 1 day, yellow crystals (13.5
1.0 Hz, 2H), 3.73 (s, 6H), 3.80 (s, 3H), 5.93 (dt, J ) 15.6, 7.8 Hz,
1H), 6.38 (d, J ) 15.6 Hz, 1H), 6.82 (d, J ) 6.4 Hz, 2H), 7.25 (d, J
) 6.4 Hz, 2H). Anal. Calcd for C16H20O5: C, 65.74; H, 6.90.
Found: C, 65.80; H, 7.11.
Dimethyl (1-(4-Methoxyphenyl)-2-propenyl)methylmalonate (4b):
1H NMR (CDCl3, 500 MHz) δ 1.51 (s, 3H), 3.71 (s, 3H), 3.79 (s, 3H),
3.87 (s, 3H), 4.19 (d, J ) 8.3 Hz, 1H), 5.09 (d, J ) 16.8 Hz, 1H), 5.14
(d, J ) 11.3, 1H), 6.32 (ddd, J ) 8.3, 11.3, 16.8 Hz, 1H), 6.90 (d, J
) 8.3 Hz, 2H), 7.25 (d, J ) 8.3 Hz, 2H). Anal. Calcd for C16H20O5:
C, 65.74; H, 6.90. Found: C, 66.00; H, 6.85.
1H NMR and analytical data for other allylic alkylation products 3
and 4 are shown below.
Dimethyl ((E)-3-Phenyl-2-propenyl)methylmalonate (3a): 1H
NMR (CDCl3, 270 MHz) δ 1.46 (s, 3H), 2.77 (dd, J ) 7.8, 1.0 Hz,
2H), 3.73 (s, 6H), 6.08 (dt, J ) 15.6, 7.8 Hz, 1H), 6.45 (d, J ) 15.6
(16) Ward, Y. D.; Villanueva, L. A.; Allred, G. D.; Payne, S. C.;
Semones, M. A.; Liebeskind, L. S. Organometallics 1995, 14, 4132.
(17) Malet, R.; Moreno-Man˜as, M.; Parella, T.; Pleixats, R. Organo-
metallics 1995, 14, 2463.
(18) Kawasaki, M.; Suzuki, Y.; Terashima, S. Chem. Pharm. Bull. 1985,
33, 52.

Regiochemistry of Allylic Esters
J. Am. Chem. Soc., Vol. 120, No. 8, 1998 1687
mg, 43%) were formed owing to dispersion of the solvents. Anal. Calcd
for C39H34OClPPd: C, 67.73; H, 5.86. Found: C, 68.01; H, 5.15.
NMR Study of [PdCl(π-cyclohexenyl)(MeO-MOP)] (11a + 11b).
(R)-MeO-MOP (21.2 mg, 0.045 mmol) and [PdCl(π-cyclohexenyl)]2
(10.0 mg, 0.023 mmol) were placed in an NMR sample tube. The
tube was filled with nitrogen, and CDCl3 (0.5 mL) was added. 1H
NMR, 13C NMR, and 31P NMR spectra were measured at 20 °C. The
ratio of major isomer 11a to minor isomer 11b was 6/1. The rate of
isomerization was measured by a saturation transfer experiment in 1H
NMR.
are shown below. The rate of isomerization was measured by a
saturation transfer experiment in H NMR. The tube was placed at
1
-20 °C. After 7 days, yellow crystals (6.3 mg, 58%) were formed.
1H NMR (CDCl3, 500 MHz, 20 °C) δ 1.00-1.08 (m, 1H), 1.25-1.32
(m, 1H), 1.50-1.57 (m, 1H), 1.68-1.76 (m, 1H), 2.07-2.15 (m, 1H),
2.18-2.33 (m, 1H), 4.14 (m, 1H), 5.64 (br t, J ) 6.8 Hz, 1H), 5.81 (br
d, J
) 5.4 Hz, 1H), 7.27-7.69 (m, 15H); 13C{1H} NMR (CDCl3,
H-P
125 MHz, 20 °C) δ 77.4 (C3), 96.2 (JC-P ) 29.9 Hz, C1), 108.4 (JC-P
) 5.3 Hz, C2); 31P{1H} NMR (CDCl3, 202 MHz, 20 °C) δ 23.9. Anal.
Calcd for C24H24ClPPd: C, 59.40; H, 4.99. Found: C, 59.45; H, 5.28.
Major Isomer (11a): 1H NMR (CDCl3, 500 MHz, 20 °C) δ 0.20-
0.25 (m, 1H), 0.63-0.71 (m, 1H), 0.87-0.94 (m, 1H), 1.28-1.33 (m,
1H), 1.57-1.63 (m, 1H), 1.75-1.81 (m, 1H), 3.15 (br t, J ) 6.4 Hz,
1H), 3.33 (s, 3H), 3.98 (br s, 1H), 5.02 (br d, JH-P ) 5.4 Hz, 1H),
6.86-8.00 (m, 22H); 13C{1H} NMR (CDCl3, 125 MHz, 20 °C) δ 54.9
(CH3), 75.4 (C3), 94.4 (d, JC-P ) 29.0 Hz, C1), 107.6 (d, JC-P ) 5.2
Hz, C2); 31P{1H} NMR (CDCl3, 202 MHz, 25 °C) δ 20.8.
Acknowledgment. We thank Professor Paul S. Pregosin,
ETH Zu¨rich, for information on the measurement of the
isomerization rate by a magnetization saturation transfer tech-
nique. This work was supported by the “Research for the Future”
Program, the Japan Society for the Promotion of Science and a
Grant-in-Aid for Scientific Research, the Ministry of Education,
Japan.
1
Minor Isomer (11b): H NMR (CDCl3, 500 MHz, 20 °C) 0.86-
2.03 (m, 6H), 3.52 (s, 3H), 4.26 (br s, 1H), 5.37 (br t, J ) 6.8 Hz, 1H),
5.61 (br d, JH-P ) 5.8 Hz, 1H), 6.86-8.00 (m, 22H); 31P{1H} NMR
(CDCl3, 202 MHz, 20 °C) δ 28.7.
Supporting Information Available: 1H NMR spectrum for
complex 11 and 31P,1H-correlation spectrum for complex 11 (2
pages). See any current masthead page for ordering information
and Web access instructions.
NMR Study and Isolation of [PdCl(π-cyclohexenyl)(PPh3)] (13).
PPh3 (5.3 mg, 0.020 mmol) and [PdCl(π-cyclohexenyl)]2 (5.0 mg, 0.011
mmol) were placed in an NMR sample tube. The tube was filled with
nitrogen, and CDCl3 (0.5 mL) was added. 1H NMR, 13C NMR, and
31P NMR spectra were measured at 20 °C. The spectral data at 20 °C
JA973150R