Catalytic asymmetric synthesis of quaternary carbon centers. Exploratory studies of intramolecular Heck reactions of (Z)-α,β-unsaturated anilides and mechanistic investigations of asymmetric Heck reactions proceeding via neutral intermediates

Article abstract of DOI:10.1021/ja980787h

Intramolecular Heck reactions of seven (Z)-α,β-unsaturated 2-iodoanilides catalyzed by Pd-BINAP were surveyed using two reaction conditions: (1) silver-promoted cyclizations in the presence of 2 equiv of Ag₃PO₄ and (2)base-promoted cyclizations in the presence of 4 equiv of 1,2,2,6,6-pentamethylpiperidine (PMP). A comparison of these results with the outcome of the corresponding intramolecular Heck reactions of the E stereoisomers leads to the following conclusions: (1) (E)- and (Z)-α,β-unsaturated 2-iodoanilides give opposite enantiomers of the Heck product when Ag₃PO₄ is the HI acceptor (cationic pathway); the absolute configuration of the product is independent of alkene geometry when the HI acceptor is PMP (neutral pathway). (2) When the 2-substituent is Me or prim-alkyl, stereoinduction is optimal in PMP-promoted insertions of Z substrates which occur with ee's as high as 97%. (3) When the 2-substituent is large, stereoinduction is optimal in insertions of E substrates carried out in the presence of Ag₃PO₄. (4) Contributions from the β-alkene substituent are minor. The neutral Heck reaction manifold can be entered from triflate substrates by carrying out the cyclization in the presence of added halide salts. The ability to vary both the alkene geometry and the Heck reaction pathway allows chiral 3,3-disubstituted-2-oxindoles having a wide range of substituents at the quaternary carbon (Me, prim-alkyl, tert-alkyl, and aryl) to be prepared with useful levels of enantioselection (72-97% ee). A number of studies were carried out aimed at clarifying the mechanism of the neutral Heck reaction pathway. Key results are the following: (1) Chiral amplification studies show that the catalyst is monomeric Pd-BINAP. (2) Investigations of monophosphine analogues of BINAP, which were designed to mimic a partially dissociated BINAP chelate, support the conclusion that BINAP is chelated during the enantioselective step. (3) Enantioselectivity is insensitive to solvent polarity. From these data, we propose that the stereochemistry determining step of the neutral pathway occurs during the process in which iodide is displaced by the tethered alkene (Figure 2, 41 → 47 → 45). In light of the variety of pentacoordinate intermediates that could be involved, it is premature to advance a model to rationalize stereoinduction in asymmetric Heck reactions proceeding by the neutral pathway. Nonetheless, the finding that the enantioselective step of asymmetric Heck reactions taking place by the neutral pathway involves five-coordinate intermediates significantly broadens the vista for design of asymmetric ligands for this and related reactions.

Full text of DOI:10.1021/ja980787h

6488
J. Am. Chem. Soc. 1998, 120, 6488-6499
Catalytic Asymmetric Synthesis of Quaternary Carbon Centers.
Exploratory Studies of Intramolecular Heck Reactions of
(Z)-R,â-Unsaturated Anilides and Mechanistic Investigations of
Asymmetric Heck Reactions Proceeding via Neutral Intermediates
Atsuyuki Ashimori,1a Benoit Bachand,1b Michael A. Calter,1c Steven P. Govek,
Larry E. Overman,* and Daniel J. Poon1d
Contribution from the Department of Chemistry, 516 Physical Sciences 1, UniVersity of California,
IrVine, California 92697-2025
ReceiVed March 9, 1998
Abstract: Intramolecular Heck reactions of seven (Z)-R,â-unsaturated 2-iodoanilides catalyzed by Pd-BINAP
were surveyed using two reaction conditions: (1) silver-promoted cyclizations in the presence of 2 equiv of
Ag3PO4 and (2) base-promoted cyclizations in the presence of 4 equiv of 1,2,2,6,6-pentamethylpiperidine
(PMP). A comparison of these results with the outcome of the corresponding intramolecular Heck reactions
of the E stereoisomers leads to the following conclusions: (1) (E)- and (Z)-R,â-unsaturated 2-iodoanilides
give opposite enantiomers of the Heck product when Ag3PO4 is the HI acceptor (cationic pathway); the absolute
configuration of the product is independent of alkene geometry when the HI acceptor is PMP (neutral pathway).
(2) When the 2-substituent is Me or prim-alkyl, stereoinduction is optimal in PMP-promoted insertions of Z
substrates which occur with ee’s as high as 97%. (3) When the 2-substituent is large, stereoinduction is optimal
in insertions of E substrates carried out in the presence of Ag3PO4. (4) Contributions from the â-alkene
substituent are minor. The neutral Heck reaction manifold can be entered from triflate substrates by carrying
out the cyclization in the presence of added halide salts. The ability to vary both the alkene geometry and the
Heck reaction pathway allows chiral 3,3-disubstituted-2-oxindoles having a wide range of substituents at the
quaternary carbon (Me, prim-alkyl, tert-alkyl, and aryl) to be prepared with useful levels of enantioselection
(72-97% ee). A number of studies were carried out aimed at clarifying the mechanism of the neutral Heck
reaction pathway. Key results are the following: (1) Chiral amplification studies show that the catalyst is
monomeric Pd-BINAP. (2) Investigations of monophosphine analogues of BINAP, which were designed to
mimic a partially dissociated BINAP chelate, support the conclusion that BINAP is chelated during the
enantioselective step. (3) Enantioselectivity is insensitive to solvent polarity. From these data, we propose
that the stereochemistry determining step of the neutral pathway occurs during the process in which iodide is
displaced by the tethered alkene (Figure 2, 41 f 47 f 45). In light of the variety of pentacoordinate
intermediates that could be involved, it is premature to advance a model to rationalize stereoinduction in
asymmetric Heck reactions proceeding by the neutral pathway. Nonetheless, the finding that the enantioselective
step of asymmetric Heck reactions taking place by the neutral pathway involves five-coordinate intermediates
significantly broadens the vista for design of asymmetric ligands for this and related reactions.
Introduction
unsaturated acids were summarized.4 These studies identified
BINAP as a particularly efficacious ligand and showed that for
many (E)-R,â-unsaturated 2-iodoanilide substrates either enan-
tiomer of the Heck product could be formed with good
enantioselectivity using a single enantiomer of a chiral diphos-
phine ligand (Figure 1). For halide substrates, this latter
discovery led to the identification of two synthetically useful
asymmetric Heck reaction manifolds: cyclization in the presence
of a halide scavenger such as a silver salt (cationic pathway) or
cyclization in the presence of a tertiary amine (neutral path-
way).3,5
The asymmetric intramolecular Heck reaction, first described
in 1989,2 has emerged as one of the most important methods
for enantioselective synthesis of C-C bonds.3 In the preceding
paper, our initial systematic studies of asymmetric Heck
cyclizations of N-alkyl-2′-iodoanilide derivatives of (E)-R,â-
(1) Current addresses: (a) The Green Cross Corporation, Research
Division, 2-25-1, Shodai-Ohtani, Hirakata, Osaka 573-1153, Japan. (b)
BioChem Therapeutique Inc., 275, Boul, Armand-Frappier, Laval, Quebec,
H7V 4A7, Canada. (c) Chemistry Department, Virginia Polytechnic Institute
and State University, 107 Davidson Hall, Blacksburg, Virginia 24061-0212.
(d) Department of Chemistry, Colorado State University, Fort Collins,
Colorado 80523-1872.
(2) (a) Sato, Y.; Sodeoka, M.; Shibasaki, M. J. Org. Chem. 1989, 54,
4738. (b) Carpenter, N. E.; Kucera, D. J.; Overman, L. E. J. Org. Chem.
1989, 54, 5846.
In this paper we describe our complementary investigations
of asymmetric Heck cyclizations of (Z)-R,â-unsaturated 2-io-
(3) For recent reviews, see: (a) Shibasaki, M. In AdVances in Metal-
Organic Chemistry; Liebeskind, L. S., Ed.; JAI: Greenwich, 1996; pp 119-
151. (b) Shibasaki, M.; Boden, C. D. J.; Kojima, A. Tetrahedron 1997, 53,
7371.
(4) Ashimori, A.; Bachand, B.; Overman, L. E.; Poon, D. J. J. Am. Chem.
Soc. 1998, 120, 6477-6487.
(5) For our initial report, see: Ashimori, A.; Overman, L. E. J. Org.
Chem. 1992, 57, 4571.
S0002-7863(98)00787-2 CCC: $15.00 © 1998 American Chemical Society
Published on Web 06/20/1998

Heck Reactions of (Z)-R,â-Unsaturated Anilides
J. Am. Chem. Soc., Vol. 120, No. 26, 1998 6489
Scheme 2
Figure 1.
Scheme 1
I2 afforded (Z)-3-iodo-2-butenol,9 which was protected to
provide 1-3. Palladium mediated carbonylation of 1 and 2 in
the presence of MeOH10 furnished (Z)-R,â-unsaturated esters 4
and 5. To reduce the volatility of the corresponding ester, iodide
3 was carbonylated in the presence of butanol to provide 6.
Condensation of these esters with 2-iodoaniline under Weinreb
conditions11 yielded the corresponding anilides, which were
methylated to afford cyclization substrates 7-9. The corre-
sponding anilide 10 lacking the γ-siloxy group was prepared
in short order from the acid chloride of angelic acid (Scheme
1).
The 2-methyl-(Z)-2-butenanilide triflate 15 was prepared as
outlined in Scheme 2. Condensation of R,â-unsaturated ester
5 with the dimethylaluminum derivative of triethylsilyl (TES)-
protected 2-aminophenol (11), followed by exposure of this
product to NaH in THF to affect cleavage of the TES group
gave anilide 12 in 78% yield. The phenol was then protected
as an allyl ether,12 and the secondary amide was N-methylated
to furnish anilide 13. Cleavage of the allyl group13 and exposure
of the resulting phenolic anilide 14 to Tf2O and pyridine gave
triflate 15 in good overall yield. Our inability to selectively
allylate the oxygen of o-aminophenol and the sensitivity of the
triethylsilyl group to the N-methylation conditions necessitated
this somewhat cumbersome protection strategy.
Preparation of the 2-phenyl-, 2-tert-butyl-, and 2-(2,2-
dimethoxyethyl)-(Z)-2-butenanilides 20-22 began with reduc-
tive iodination of the corresponding propargylic alcohols
(Scheme 3). Conversion of these products to the desired anilide
products was simplified by the discovery that lithium reagents
derived from these iodides could be directly condensed with
2-iodophenyl isocyanate (19).14 Thus, reaction of vinyl iodides
16-18 with 2 equiv of t-BuLi at -78 °C in THF,15 followed
by trapping with 19 and subsequent methylation of the anilide
nitrogen, furnished the respective (Z)-2-butenanilides 20-22 in
21-74% overall yield.
doanilides and related triflate substrates.6 In many cases these
cyclizations proceed with higher enantioselection than those of
the corresponding E stereoisomers, with enantioselectivities as
high as 97% being obtained in cyclizations conducted in the
presence of amines. We report also that the neutral reaction
manifold can be entered from triflate substrates if the Heck
cyclization is carried out in the presence of halide additives and
that both phosphorus atoms of BINAP and one halide are
coordinated to palladium during the enantioselective step7 of
the neutral pathway.8
Results
Scope of Asymmetric Heck Cyclizations. Intramolecular
Heck reactions of seven (Z)-R,â-unsaturated 2-iodoanilide
A. Asymmetric Heck Cyclizations of (Z)-r,â-Unsaturated
2-Iodoanilides. Preparation of Cyclization Substrates. (Z)-
4-Siloxy(or alkoxy)-2-butenanilides containing a 2-methyl group
were prepared from 2-butyn-1-ol as summarized in Scheme 1.
Reduction with Red-Al and quenching the resulting alane with
(9) (a) Corey, E. J.; Katzenellenbogen, J. A.; Posner, G. H. J. Am. Chem.
Soc. 1967, 89, 4245. (b) Blanchette, M. A.; Malmas, M. S.; Nantz, M. H.;
Roberts, J. C.; Somjai, P.; Whritenour, D. C.; Masamune, S.; Kageyama,
M.; Tamura, T. J. Org. Chem. 1989, 54, 2817.
(10) Schoenberg, A.; Bartoletti, I.; Heck R. F. J. Org. Chem. 1974, 39,
3318.
(6) For a preliminary report, see: Ashimori, A.; Matsuura, T.; Overman,
L. E.; Poon, D. J. J. Org. Chem. 1993, 58, 6949.
(7) The first irreversible enantiodifferentiating step would determine
asymmetric induction, see: C. R. Landis, J. Halpern, J. Am. Chem. Soc.
1987, 109, 425.
(8) For a preliminary report, see: Overman, L. E.; Poon, D. J. Angew.
Chem., Int. Ed. Engl. 1997, 36, 518.
(11) (a) Lipton, M. F.; Bisha, A.; Weinreb, S. M. Org. Synth. 1978, 59,
49. (b) Greenlee, W. J.; Hangauer, D. G. Tetrahedron Lett. 1983, 24, 4559.
(12) McKillop, A.; Fiaud, J. C.; Hug, R. P. Tetrahedron 1974, 30, 1379.
(13) Deziel, R. Tetrahedron Lett. 1987, 28, 4371.
(14) Cizmarik, J.; Novosedlikova, D.; Racanska, E. Pharmazie 1990, 45,
64.
(15) Neumann, H.; Seebach, D. Chem. Ber. 1978, 111, 2785.

6490 J. Am. Chem. Soc., Vol. 120, No. 26, 1998
Ashimori et al.
Scheme 3
Separation of these products by preparative TLC, hydrolysis to
oxindole aldehyde 25, and reduction provided alcohol 29 in 92%
ee from the (E)-enoxysilane intermediate and 73% ee from the
corresponding Z stereoisomer, corresponding to an overall ee
of 90%. Asymmetric Heck cyclization of 7 in the presence of
Ag3PO4 yielded a 4:1:3 mixture of (E),(Z)-enoxysilanes 23 (R1
) Me, R2 ) OTIPS) and aldehyde 25. Isolation of these
components by flash and preparative thin-layer chromatography
and conversion of each to oxindole alcohol 29 produced material
of 87, 84, and 70% ee, respectively, corresponding to an overall
ee of 80%. Similar results were observed in cyclizations of
(Z)-2-butenanilides 8 and 9 and are discussed elsewhere.17
Assignment of Absolute Configuration. The absolute
configuration of oxindole aldehyde 25 was secured by chemical
correlation with (+)-esermethole (34), whose absolute config-
uration is well-established (Scheme 5).18 (-)-Oxindole alde-
hyde 25, produced from Pd-(R)-BINAP catalyzed cyclization
of 2-methyl-(Z)-2-butenanilide 8 in the presence of PMP, was
reductively aminated to yield dextrorotatory pyrroloindoline 32.
This intermediate was then converted to (+)-esermethole
following the procedure of Fuji and co-workers.19 The absolute
configurations of oxindole aldehydes 27 and 28 were established
by chemical correlation with 25 as documented in the preceding
paper in this issue. The 3-vinyloxindole 24 was correlated with
29 by hydroboration-oxidation.4 The absolute configuration of
the R-phenyl oxindole aldehyde 26 was not established.
B. Mechanistic Investigations. Nature and Stability of
the Catalyst. Evidence suggesting that monomeric Pd-BINAP
was the active catalyst was secured through a chiral amplifica-
tion study.20 Asymmetric Heck cyclizations of the iodide 8 were
performed under both PMP and Ag3PO4 conditions using
BINAP of varying enantiopurities (Table 2). A linear relation-
ship between enantioselection and BINAP enantiopurity was
found with both HI scavengers.21
To ascertain whether enantioselectivity was affected by
adventitious water, asymmetric Heck cyclizations of 8 were
conducted in DMA containing varying concentrations of water.
As is apparent in the data summarized in Table 3 (entries 1-5),
addition of up to 5% (v/v) water had no detectable impact on
enantioselectivity and only a minor effect on enoxysilane ratios
and reaction rate. Measurable but minor erosion of yield and
ee was observed when 10% water was added. Although the
presence of 5% water had no effect, changing the solvent to
THF increased enantioselectivity to 97% (entry 7).
Adventitious oxygen can oxidize phosphine ligands.22 Oxi-
dation of a chiral bidentate ligand such as BINAP could lead
to ligand dissociation and loss of stereoinduction. An investiga-
tion into the integrity of Pd-BINAP at prolonged reaction times
was undertaken. The catalyst, generated in situ by mixing Pd2-
(dba)3‚CHCl3 and (R)-BINAP in DMA, was preheated to 100
°C for varying lengths of time. Iodide 8 and PMP were then
added, and the enantioselectivity of the reaction was determined
(Table 4).
substrates (Table 1 and Scheme 4) were surveyed using the two
reaction conditions developed during our earlier investigations:4
(1) silver-promoted cyclizations were conducted with 5 mol %
Pd2(dba)3‚CHCl3 and 12 mol % (R)-BINAP in the presence of
2 equiv of Ag3PO4 at 100 °C in N,N-dimethylacetamide (DMA,
0.07-0.1 M), and (2) base-promoted cyclizations were carried
out identically in the presence of 4 equiv of PMP (no Ag3PO4).
Results of these investigations are summarized in Table 1. The
enantiomeric purity of the oxindole products was determined
by HPLC analysis using a chiral stationary phase; with the
exception of oxindole aldehyde 28, this assay was carried out
on the derived oxindole alcohol.
Pd-BINAP catalyzed asymmetric Heck cyclizations of
2-methyl-4-siloxy (or methoxy)-(Z)-2-butenanilides 7-9 in the
presence of either Ag3PO4 or 1,2,2,6,6-pentamethylpiperidine
(PMP) furnished, after hydrolysis of the intermediate (E)- and
(Z)-enol ethers 23, (R)-oxindole aldehyde 25. Enantioselectivity
was highest (89-92% ee) when PMP was employed as the HI
scavenger and varied little with the nature of the oxygen
protecting group (entries 1-6). 2-Butenanilide 10 having no
substituent on the 4-carbon cyclized to (R)-oxindole 24 in similar
fashion with only slightly reduced enantioselectivity (entries 7
and 8). As was seen in the (E)-2-butenanilide series, dramatic
changes in enantioselectivity resulted from variations in the
R-substituent. Cyclization of the 2-Me and 2-(2,2-dimethoxy-
ethyl)-2-butenanilides (7 and 22) in the presence of PMP
proceeded with excellent enantioselectivities (entries 1 and 13);
the opposite stereochemical descriptor for oxindole aldehyde
28 arises from a change in Cahn-Ingold-Prelog priority
assignment. Attempted cyclization of 22 in the presence of Ag3-
PO4 resulted in decomposition. The 2-Ph substrate 20 cyclized
with poor enantioselectivity (19% ee) in the presence of PMP
but underwent intramolecular Heck insertion in 65% ee in the
presence of Ag3PO4; an analogous trend was also observed in
the (E)-2-butenanilide series.4 Reaction of the 2-t-Bu analogue
21 under either cyclization condition gave the corresponding
oxindole aldehyde 27 in low enantiomeric purity (entries 11
and 12).
The experiments reported in entries 1-4 were conducted in
a 10 mL recovery flask sealed with a rubber septum under a
Kinetic Resolution in the â-Hydride Elimination Step.
Cyclizations of the 2-methyl-(Z)-2-buten-2′-iodoanilides were
studied in greatest detail; in each case the aldehyde and
stereoisomeric enol ethers generated in the initial Heck cycliza-
tion step were separated and individually converted to oxindole
alcohol 29 for enantiomeric purity assessment. These analyses
reveal that internal kinetic resolution in the â-hydride elimination
step is operative.16 For example, Pd-BINAP catalyzed cy-
clization of 7 in the presence of PMP afforded a 32:1 mixture
of the (E)- and (Z)-enoxysilanes 23 (R1 ) Me, R2 ) OTIPS).
(16) Ozawa, F.; Kubo, A.; Matsumoto, Y.; Hayashi, T. Organometallics
1993, 12, 4188.
(17) Poon, D. J. Ph.D. Dissertation, University of California, Irvine, 1996.
(18) Takano, S.; Ogasawara, K. In The Alkaloids; Brossi, A., Ed.;
Academic: New York, 1989; Vol. 36, pp 224-251.
(19) Node, M.; Itoh, A.; Masaki, Y.; Fuji, K. Heterocycles 1991, 32,
1705.
(20) Puchot, C.; Samuel, O.; Dun˜ach, E.; Zhao, S.; Agami, C.; Kagan,
H. B. J. Am. Chem. Soc. 1986, 108, 2353.
(21) Plots of these data are provided in the Supporting Information.
(22) Wilke, G.; Schott, H.; Hemibach, P. Angew. Chem. 1967, 79, 62.

Heck Reactions of (Z)-R,â-Unsaturated Anilides
J. Am. Chem. Soc., Vol. 120, No. 26, 1998 6491
Table 1. Cyclizations of 2′-Iodo-N-methyl-(Z)-2-substituted-2-butenanilides with Pd-(R)-BINAPa
entry
substrate
R1
R2
additive
E:Z:aldehydeb
overall ee,c %
yield,d %
1
2
3
4
5
6
7
8
9
10
11
12
13
7
Me
Me
Me
Me
Ph
OTIPS
PMP
Ag3PO4
PMP
Ag3PO4
PMP
Ag3PO4
PMP
Ag3PO4
PMP
Ag3PO4
PMP
Ag3PO4
PMP
32:1:0
4:1:3
24:1:0
1:0:1
9:1:0
4:1:
90 (R)
80 (R)
92 (R)
78 (R)
89 (R)
80 (R)
85 (R)
69 (R)
19e
87
73
80
53
76
72
89
85
ndf
86
76
55
93
8
9
OTBDMS
OMe
10
20
21
22
H
OTIPS
OTIPS
OTIPS
9:1:0
12:1:9
19:1:0
6:1:0
65e
t-Bu
0
33 (S)
91 (S)
CH2CH(OMe)2
49:1:0
a Conditions: 5 mol % Pd2(dba)3‚CHCl3, 12 mol % (R)-BINAP, 4 equiv of PMP or 2 of equiv Ag3PO4, DMA (0.07-0.1 M), 100 °C. b Ratios
1
of (E)-enol ether:(Z)-enol ether:aldehyde were determined by capillary GC or by H NMR analysis. c Determined by HPLC analysis of 28-31
using a Chiralcel OB-H, OD, or OJ column. d Combined yield of oxindole 28-31. e The PMP- and Ag3PO4-promoted cyclizations produce the
same enantiomersd of unknown absolute configuration. The major enantiomer elutes first on a Chiracel OJ column (85:15 hexane-propanol). f Not
determined.
Table 2. Cyclization of Iodide 8 To Form Oxindole Alcohol 29
Using (R)-BINAP of Varying Enantiopuritya
Scheme 4
ee of
(R)-BINAP, %
PMP-promoted
ee of 29,b %
Ag3PO4-promoted
ee of 29,b %
25
50
75
23
44
69
92
21
43
60
78
>99
a Conditions: 5 mol % Pd2(dba)3‚CHCl3, 12 mol % (R)-BINAP, 4
of equiv PMP or 2 of equiv Ag3PO4, DMA (0.1 M); reactions
employing PMP were conducted at 100 °C, those employing Ag3PO4
at 80 °C. All reactions proceeded to completion and produced (R)-29.
b Determined by HPLC analysis of 29 using a CHIRALCEL OB-H
column.
Table 3. Pd-(R)-BINAP-Catalyzed Cyclization of Iodide 8 To
Ultimately Form Oxindole Alcohol 29 in DMA Containing Varying
Amounts of Water and in THFa
(E)/(Z) ratiob yield,c % ee,d %
Scheme 5
entry
solvent
1
2
3
4
5
6
7
DMA
22:1
22:1
23:1
22:1
28:1
28:1
e
73
76
e
91
92
91
90
90
87
97
0.1% H2O-DMA (v/v)
0.5% H2O-DMA (v/v)
1.0% H2O-DMA (v/v)
5.0% H2O-DMA (v/v)
10.0% H2O-DMA (v/v)
THF
e
75
68
73
a Conditions: 5 mol % Pd2(dba)3‚CHCl3, 12 mol % (R)-BINAP, 4
equiv of PMP, DMA (0.1 M), 100 °C. All reactions proceeded to
completion. b Enoxysilane stereoisomer ratios were determined by GLC
analysis. c Overall yield alcohol 29. d Enantiomeric excess ((2% ee)
of (R)-29 was measured by HPLC using a Chiralcel OB-H or AS
column. e Not determined.
Effects of Added Salts on Asymmetric Heck Cyclizations
of 2-Methyl-(Z)-2-butenanilide Iodide 8 and Triflate 15.
There are many reports that added salts can enhance reactivity
and selectivity of palladium mediated processes.23,24 The effect
of added halide salts on Pd-BINAP catalyzed cyclizations of
iodide 8 carried out in the presence of PMP are summarized in
Table 5. Addition of 1-2 equiv of hydrohalide salts of PMP
had little effect on enantioselection (93 ( 2% ee); however,
nitrogen atmosphere. Under these conditions, significant erosion
of enantioselectivity was observed when the precatalysts were
heated at 100 °C for 12 h prior to adding the substrate and PMP.
This apparent loss of catalyst integrity could be avoided using
more rigorous Schlenk techniques (entry 5). In this case, the
catalyst was rigorously deoxygenated by the freeze-pump-
thaw method and purged with argon. A DMA solution of iodide
8 and PMP was deoxygenated in the same way prior to being
added to a preheated DMA solution of Pd-BINAP. Under
these conditions, no erosion of enantioselectivity was observed
when the catalyst was incubated for 18 h at 100 °C prior to
adding 8 and PMP. As a result of these findings, Schlenk
techniques were employed in all of our subsequent mechanistic
studies (vide infra).
(23) For a recent review, see: Jeffery, T. In AdVances in Metal Organic
Chemistry; Liebeskind, L. S., Ed.; JAI: Greenwich, 1996; Vol. 5, pp 153-
260.
(24) Selected examples include: (a) Sonesson, C.; Larhead, M.; Nyqvist,
C.; Hallberg, A. J. Org. Chem. 1996, 61, 4756. (b) Jeffery, T. Tetrahedron
1996, 52, 10113. (c) Amatore, A.; Jutand, A.; Suarez, A. J. Am. Chem.
Soc. 1993, 115, 9531. (d) Merlic, C. A.; Semmelhack, M. F. J. Organomet.
Chem. 1990, 391, C23. (e) Scott, W. J.; Stille, J. K. J. Am. Chem. Soc.
1986, 108, 3033.

6492 J. Am. Chem. Soc., Vol. 120, No. 26, 1998
Ashimori et al.
Table 4. Cyclization of Iodide 8 To Ultimately Form Oxindole
Alcohol 29 Using Pd-(R)-BINAP Preformed in Different Waysa
Table 6. Cyclization of Triflate 15 with Pd-(R)-BINAP To
Ultimately Form Oxindole 29 in the Presence of PMP and Added
Saltsa
catalyst formation
time,b h
enoxysilane
entry
(E)/(Z)c
29 ee,d %
entry
additiveb
solvent
DMA
5% H2O-DMA
10% H2O-DMA
THF
DMA
DMA
DMA
DMA
DMA
DMA
DMA
5% H2O-DMA
10% H2O-DMA
THF
ee,c %
yield,d %
1
2
3
1
3
6
12
18
23:1
23:1
20:1
4:1
89
91
90
54
91
1
2
3
4
5
6
7
8
9
10
11
12
13
14
43-48
49
47
49
42
90
93
93
91
92
88
90
88
72
78
72
81
70e
62
59
52
40
62
60
90
91
89
4
5e
21:1
n-Bu4NOTf
n-Bu4NI
a Conditions: 5 mol % Pd2(dba)3‚CHCl3, 12 mol % (R)-BINAP, 4
equiv of PMP, DMA (0.1 M) at 100 °C. Reactions were conducted
under nitrogen in recovery flasks sealed with a septa unless noted
otherwise. All reactions proceeded to completion. b The precatalysts
were heated for the specified time at 100 °C prior to adding 8 and
PMP. c Enoxysilane ratios were determined by GLC analysis. d Enan-
tiomeric excess ((2%) of (R)-29 was measured by HPLC using a
Chiralcel OB-H column. e Reaction was conducted in Schlenk equip-
ment under Ar and degassed by the freeze-pump-thaw method (see
text).
n-Bu4NBr
n-Bu4NCl
PMP‚HIf
PMP‚HBr
PMP‚HCl
n-Bu4NI
n-Bu4NI
n-Bu4NI
93
a Conditions: 5 mol % Pd2(dba)3‚CHCl3, 12 mol % (R)-BINAP, 4
equiv of PMP, DMA (0.1 M), 100 °C. All reactions proceeded to
completion. b One equiv with respect to substrate. c Enantiomeric excess
((2% ee) of (R)-29 was measured by HPLC using a Chiralcel OB-H
or AS column. d Overall yield of alcohol 29. e Overall yield of aldehyde
25. f Reaction proceeded to 66%.
Table 5. Cyclization of Iodide 8 with Pd(R)-BINAP To
Ultimately Form Oxindole 29 in the Presence of PMP and Added
Saltsa
entry
additiveb
solvent
ee,c %
yield,d %
1
2
3
4
5
6
none
DMA
DMA
DMA
DMA
DMA
DMA
91
91
92
95
94
43
76
62
39
45
75
f
Scheme 6
PMP‚HI
PMP‚HIe
PMP‚HBr
PMP‚HCl
AgOTf
a Conditions: 5 mol % Pd2(dba)3‚CHCl3, 12 mol % (R)-BINAP, 4
equiv of PMP, DMA (0.1 M), 100 °C. All reactions proceeded to
completion. b One equiv with respect to substrate. c Enantiomeric excess
((2% ee) of (R)-29 were measured by HPLC using a Chiralcel OB-H
or AS column. d Overall yield of alcohol 29. e Two equiv with respect
to substrate. f Not determined.
the yield of oxindole alcohol 29 was somewhat reduced. As
expected from the results summarized in Table 1, addition of
AgOTf decreased enantioselection. The reduction was consid-
erably greater with AgOTf (to 43% ee) than with Ag3PO4
(compare entry 6 of Table 5 with entry 4 of Table 1).
unsaturated 2-iodoanilide substrates, we examined the cycliza-
tion of iodide 8 using monophosphine analogues of (R)-BINAP.
Ligands 35 and 3626 were prepared as described by Hayashi,26d
while 38 was synthesized from the known (R)-phosphine ester
37 (Scheme 6).27 Condensation of 37 with excess PhLi followed
by in situ protection of the phosphine moiety as a borane
complex28 and reduction of the triaryl carbinol with NaBH4 and
trifluoroacetic acid29 furnished ligand 38 in 50% overall yield.
Asymmetric Heck cyclizations of iodide 8 in the presence of
5% Pd2(dba)3‚CHCl3 and 11% (R)-monophosphines 35, 36, or
38 at 100 °C in DMA in the presence of 4 equiv of PMP
provided (S)-oxindole alcohol 29 (the opposite enantiomer to
that produced using (R)-BINAP in low ee (Table 7). Increasing
the ratio of 36 to Pd did not significantly affect enantioselectivity
(entries 3-6). A striking feature of the monophosphine-
mediated cyclizations is their high reaction rate. Heck cycliza-
tion of 8 mediated by Pd(R)-BINAP requires 20-60 min at 100
°C to proceed to completion, while this reaction using mono-
Asymmetric Heck cyclization of 2-butenanilide triflate 15
proceeded with an enantioselectivity of 45 ( 3% in DMA,
within experimental error to the enantioselection realized in
cyclization of iodide 8 in the presence of AgOTf (compare entry
6 of Tables 5 and entry 1 of Table 6). Within experimental
uncertainty, enantioselectivity was unaffected by the addition
of 5-10% water (Table 6, entries 2 and 3) or if THF was
employed as the reaction solvent (entry 4). Addition of excess
triflate also had no discernible effect (entry 1 vs 5). However,
addition of alkylammonium halides to cyclizations of triflate
15 in DMA dramatically raised enantioselectivity to 90 ( 3%
ee (Table 6, entries 1, 6-8). The high enantiomeric excesses
obtained in these reactions are identical, within experimental
uncertainty, to those realized in the cyclization of iodide 8 (Table
5). Qualitatively similar results were observed upon adding
hydrohalic salts of PMP (entries 9-11). The marked effect of
halide salts was not dependent on the polarity of the reaction
solvent, since nearly identical enhancements upon addition of
n-Bu4NI were observed in DMA, DMA containing 5-10%
water, and THF (entries 1 and 12-14).
(25) (a) Cabri, W.; Candiani, I. Acc. Chem. Res. 1995, 28, 2.
(26) (a) Uozumi, Y.; Tanahashi, A.; Lee, S. Y.; Hayashi, T. J. Org. Chem.
1993, 58, 1945. (b) Hayashi, T. J. Synth. Org. Chem. Jpn. 1994, 52, 900.
(c) Uozumi, Y.; Hayashi, T. J. Am. Chem. Soc. 1991, 113, 9887.
(27) Uozumi, Y.; Suzuki, N.; Ogiwara, A.; Hayashi, T. Tetrahedron 1994,
50, 4293.
(28) (a) Frisch, M. A.; Heal, H. G.; Mackle, H.; Maden, I. O. J. Chem.
Soc. 1965, 899. (b) Gough, S. T. D.; Triplett, S. J. Chem. Soc. 1961, 4264.
(c) Bailey, W. J.; Buckler, S. A. J. Am. Chem. Soc. 1957, 79, 3567.
(29) Frisch, M. A.; Heal, H. G.; Mackle, H.; Maden, I. O. J. Chem. Soc.
1965, 899.
Studies with Chiral Monodentate Phosphine Ligands.
Phosphine dissociation has been postulated to be the cause of
the low stereoinduction often seen in asymmetric Heck reactions
of aryl and vinyl halides conducted in the absence of halide
scavengers.3,16,25 To explore the possibility of phosphine
dissociation in the PMP-promoted Heck reactions of R,â-

Heck Reactions of (Z)-R,â-Unsaturated Anilides
J. Am. Chem. Soc., Vol. 120, No. 26, 1998 6493
cationic and neutral.3,25,30c-g,34 These are depicted schematically
in Figure 2 for intramolecular insertion of an unsaturated
aromatic substrate brought about by a (bisphosphine)Pd(0)
catalyst.
Table 7. Cyclization of Iodide 8 To Ultimately Form Oxindole
Alcohol 29 Using Palladium-Monophosphine Catalysts and
Pd-(R)-BINAP
enoxysilane
ee of
config
of 29
entry
ligand
ligand:Pdb
ratio (E:Z)c
29,d %
First introduced independently by Cabri34a and Hayashi,34b
the cationic reaction manifold has been uniformly invoked to
describe asymmetric Heck reactions of unsaturated triflates or
halides when carried out in the presence of Ag(I) or Tl(I) salts.3
Triflate dissociation, or halide abstraction by Ag(I) or Tl(I) salts,
vacates a coordination site on the Pd(II) complex, facilitating
olefin complexation (Figure 2, 40 f 42 or 39 f 41 f 42). A
convincing demonstration of the stability of (bisphosphine)-
(aryl)palladium(II) halide complexes such as 41 and their
activation for migratory insertion by halide removal is provided
by Brown’s recent stoichiometric simulation studies.35 As
illustrated in Scheme 7, reaction of 48 with [1,1′-bis(diphen-
ylphosphino)ferrocene](2η-cyclooctatetraene)palladium gave the
stable, crystallographically characterized adduct 49 having no
direct interaction of the alkene and palladium. However,
exposing this arylpalladium iodide complex to AgOTf at -78
°C resulted in immediate precipitation of AgI and rapid reaction
(at -40 °C) to form ultimately 50. Since both phosphines of a
chiral bisphosphine ligand are readily accommodated during the
migratory insertion step (42 f 43), the cationic mechanism
provides a simple rationale for the higher enantioselectivity often
seen in asymmetric Heck reactions of halides when carried out
in the presence of Ag(I) or Tl(I) salts.
1
2
3
4
5
6
7
35
36
36
36
36
38
BINAP
1.1:1
1.1:1
1.6:1
2.1:1
2.9:1
1.2:1
1.1:1
1.4:1
1.2:1
1.4:1
1.5:1
1.5:1
1.3:1
22:1
27
23
21
21
24
19
91
S
S
S
S
S
S
R
a Conditions: 5 mol % Pd2(dba)3‚CHCl3, 12 mol % ligand, 4 equiv
of PMP, DMA (0.1 M), 100 °C. All ligands are of the R configuration
and all reactions proceeded to completion. b Equiv. c Determined by
GLC analysis. d Enantiomeric excess ((2% ee) of 29 was measured
by HPLC using a Chiralcel OB-H or AS column.
phosphine 36 as ligand is complete in 10 min at ambient
temperature. Reactions carried out at 54 °C for 5 min using 5
mol % Pd2(dba)3‚CHCl3 without added ligand or 5 mol % Pd2-
(dba)3‚CHCl3 in the presence of 11 mol % of BINAP or 36
give a rough estimate of relative rates. Under these conditions,
the Pd-(R)-BINAP reaction did not proceed to a detectable
extent, the Pd2(dba)3‚CHCl3 catalyzed reaction proceeded to
∼50% completion, and the Pd-36 catalyzed reaction was
complete in 5 min.
Discussion
Heck reactions of aryl or vinyl halides with palladium
catalysts having bidentate ligands in the absence of Ag(I) or
Tl(I) salts are notorious for their sluggishness.30 The intransi-
gence of these reactions has been widely attributed to the
reluctance of a chelating bisphosphine ligand to dissociate from
(bisphosphine)(aryl)Pd(halide) complexes. In asymmetric Heck
reactions, partial dissociation of a chiral bisphosphine ligand
would result in a loss of ligand rigidity which could erode
enantioselectivity.3 In contrast to this conventional wisdom,
we first reported in 1992 that Pd-BINAP catalyzed intramo-
lecular Heck reactions of some aryl halide substrates take place
at 80-100 °C with high enantioselectivity in solvents such as
DMA in the absence of halide scaVengers.5 Current under-
standing of the neutral Heck reaction manifold is obviously
incomplete, and this pathway warrants closer scrutiny, an issue
we will return to shortly.
Synthetic Applications of Intramolecular Heck Cycliza-
tions. The studies reported in this and the preceding paper4
demonstrate that enantioenriched 3,3-disubstituted oxindoles,
indolines, and dihydrobenzofurans can be prepared by Pd-
BINAP-catalyzed intramolecular Heck reactions of o-iodoarenes
tethered to (E)- or (Z)-trisubstituted alkenes. Enantioselectivities
range from moderate to high (up to 97% ee). These investiga-
tions moreover convincingly demonstrate that synthetically
useful enantioselectivities can be realized in intramolecular Heck
insertions of iodide substrates proceeding by either the cationic
(HI acceptor is Ag3PO4) or neutral (HI acceptor is PMP)
pathways.4
Introductory Observations. The general outlines of the
mechanism of the Heck reactionsoxidative addition, coordina-
tion of the alkene, insertion, and â-hydride eliminationshave
been appreciated since the 1970s and have been discussed in
numerous reviews.30 During the past few years, individual steps
of the Heck reaction have been subjected to increasing scrutiny.
The extensive investigations of Amatore, Jutand, and co-workers
have shed much light on the nature of the active catalyst and
its participation in the oxidative addition step.31 Recently, for
example, they demonstrated that the predominant species in
mixtures of Pd(dba)2 and BINAP is the monomeric Pd(dba)-
BINAP complex which oxidatively adds to PhI to form
(BINAP)Pd(Ph)I.31d Other recent studies have shown that
(Ph3P)2(aryl)PdCl complexes do not readily undergo chloride
dissociation in DMF,32 whereas the corresponding triflate
complexes are completely ionized.32,33 That two different
oxidative addition products can be formed has led to general
acceptance of two reaction manifolds for Heck reactions:
(30) Selected reviews include: (a) Heck, R. F. Org. React. 1982, 27,
345. (b) Heck, R. F. In ComprehensiVe Organic Synthesis; Trost, B. M.,
Ed.; Pergamon: London, 1991; Vol. 4, pp 833-863. (c) Davies; G. D.;
Hallberg, A. Chem. ReV. 1989, 89, 1433. (d) de Meijere, A.; Meyer, F. E.
Angew. Chem., Int. Ed. Engl. 1994, 33, 2379. (e) Gibson, S. E.; Middleton,
R. J. Contemp. Org. Synth. 1996, 3, 447. (f) Bra¨se, S.; de Meijere, A. In
Metal-catalyzed Cross-coupling Reactions; Diederick, F., Stang, P. J., Eds.;
Wiley-VCH: Weinheim, 1998; Chapter 3. (g) Link, J. T.; Overman, L. E.
In Metal-catalyzed Cross-coupling Reactions; Diederick, F., Stang, P. J.,
Eds.; Wiley-VCH: Weinheim, 1998; Chapter 6.
(31) (a) Amatore, C.; Azzabi, M.; Jutand, A. J. Am. Chem. Soc. 1991,
113, 8375. (b) Amatore, C.; Jutand, A.; M’Barki, M. A. Organometallics
1995, 14, 1818. (c) Amatore, C. Jutand, A.; Suarez, J. Am. Chem. Soc.
1993, 115, 9531. (d) Amatore, C.; Broeker, G.; Jutand, A.; Khalil, F. J.
Am. Chem. Soc. 1997, 119, 5176.
(32) Jutand, A.; Mosleh, A. Organometallics 1995, 14, 1810.
(33) (a) Stang, P. J.; Kowalski, M. H.; Schiavilli, M. D.; Longford, D.
J. Am. Chem. Soc. 1989, 111, 3347. (b) Stang, P. J.; Kowalski, M. H. J.
Am. Chem. Soc. 1989, 111, 3356. (c) Hinkle, R. J.; Stang, P. J.; Kowalski,
M. H. J. Org. Chem. 1990, 55, 5033. (d) Brown, J. M.; King, K. Angew.
Chem., Int. Ed. Eng. 1996, 35, 657.
Our investigations of experimental procedures for conducting
asymmetric Heck reactions provided a useful caveat: To obtain
optimum enantioselectivities in asymmetric Heck reactions that
proceed slowly at elevated temperatures, the reaction should
be conducted with rigorous exclusion of oxygen. We also
demonstrated that when the reaction is carried out in a dipolar
aprotic solvent such as DMA, small amounts of water (up to
5%) do not affect enantioselection.
(34) Cabri, W.; Candiani, I.; DeBernardis, S.; Francalanci, F.; Penco, S.
J. Org. Chem. 1991, 56, 5796.
(35) Brown, J. M.; Pe´rez-Torrente, J. J.; Alcock, N. W.; Clase, H. J.
Organometallics 1995, 14, 207.

6494 J. Am. Chem. Soc., Vol. 120, No. 26, 1998
Ashimori et al.
Figure 2. Cationic and neutral Heck reaction manifolds.
Scheme 7
PMP-promoted insertions of Z substrates and take place with
ee’s as high as 90-97%. (3) When the 2-substituent is large,
stereoinduction is optimal in insertions of E substrates carried
out in the presence of Ag3PO4. (4) Contributions from the â
alkene substituent are minor.
The ability to vary both the alkene geometry and the Heck
reaction manifold allows chiral 3,3-disubstituted-2-oxindoles
having a wide range of substituents at the quaternary carbon
(Me, prim-alkyl, tert-alkyl, and aryl) to be prepared with ee’s
ranging from 72 to 97%. Enantioenriched oxindoles of this type
are certain to be valuable intermediates for asymmetric synthesis
of a variety of indole alkaloids. Our initial use of these
intermediates to prepare both enantiomers of physostigmine and
other Calabar alkaloids is detailed in the following paper.37
It has often been observed that stereoisomeric alkenes give
opposite enantiomers in asymmetric Heck cyclizations proceed-
ing by the cationic pathway.3 The sense of stereoinduction
observed in Pd-BINAP catalyzed cyclizations of the R,â-
unsaturated 2-iodoanilides examined in this and the companion
study4 can be rationalized by the insertion proceeding prefer-
entially38,39 via the most stable cationic square planar alkene
complex having an approximate coplanar orientation of the
alkene π and aryl-C σ bonds.40 Models of these complexes
are shown in Figure 3 for intermediates resulting from the
reaction of Pd-(R)-BINAP with iodide 10 and its E stereoiso-
mer.41,42 Two destabilizing interactions are discernible: (a)
interaction between the â-methyl group of the alkene and the
While the cationic manifold of the Heck reaction can be
entered from an aryl halide precursor in the presence of silver
or thallium salts, the converse, accessing the neutral pathway
from a triflate substrate, had not been explored in depth.24b,36
The identical ee’s obtained in PMP-promoted Heck cyclizations
of (Z)-2-butenanilide triflate 10 in the presence of added iodide
and (Z)-2-butenanilide iodide 8 alone demonstrate the feasibility
of entering the neutral reaction manifold from triflate precursors.
Since the iodide and triflate functional groups display quite
different reactivity with many common reagents, the flexibility
of being able to enter the cationic or neutral asymmetric Heck
reaction manifolds from either precursor allows for important
latitude in the preparation of cyclization substrates.
A comparison of the results presented in Table 1 for
intramolecular Heck insertions of (Z)-R,â-unsaturated 2-iodo-
anilide substrates with the outcome of corresponding insertions
of (E)-2-butenanilides4 leads to the following conclusions: (1)
(E)- and (Z)-R,â-unsaturated 2-iodoanilides give opposite enan-
tiomers of the Heck product when Ag3PO4 is the HI acceptor,
while the sense of stereoinduction is independent of alkene
geometry when the HI acceptor is PMP. (2) When the
2-substituent is Me or prim-alkyl, stereoinduction is optimal in
(37) Matsuura, T.; Overman, L. E.; Poon, D. J. Am. Chem. Soc. 1998,
120, 6500-6503.
(38) Thorn, D. L.; Hoffmann, R. J. Am. Chem. Soc. 1978, 100, 2079.
(39) Either formation or insertion of this intermediate could be the
enantioselective step.
(40) Abelman, M. M.; Overman, L. E.; Tran, V. D. J. Am. Chem. Soc.
1990, 112, 6959.
(41) The models depicted in Figure 3 represent local minima and were
built using the Tripos 5.7 force field in the molecular mechanics module
of Spartan 2.0. The X-ray coordinates of PdCl2[(R)-BINAP] served as the
starting point.42
(36) Maddaford, S. P.; Anderson, W. G.; Cristofoli, W. A.; Keay, B. A.
J. Am. Chem. Soc. 1996, 118, 7108.
(42) Toshiro, K.; Ito, T.; Takaya, H.; Souchi, S.; Noyori, R. Acta
Crystallogr. 1982 B38, 807.

Heck Reactions of (Z)-R,â-Unsaturated Anilides
J. Am. Chem. Soc., Vol. 120, No. 26, 1998 6495
Figure 3. Models of cationic square planar complexes formed from Pd-(R)-BINAP and 2′-iodo-N-methyl-(E)- or -(Z)-2-methyl-2-butenanilide
having the alkene π and aryl C-Pd σ bonds nearly coplanar. The naphthalene rings of BINAP have been removed for clarity.
proximal pseudo-equatorial phenyl group of BINAP43 and (b)
contact between the twisted aryl ring of the substrate and the
other pseudo-equatorial phenyl group of BINAP. For the Z
substrate, the closest contact between the â-methyl carbon and
the nearest carbon of the pseudo-equatorial phenyl is 3.2 Å in
complex A and 2.9 Å in diastereomeric complex B. Considering
the aryl-phenyl interaction, the closest C-C contact is 3.2 Å
in complex A and 2.8 Å in diastereomeric complex B. The
situation is less clear with the E substrate, since destabilizing
interactions with the â-methyl group are present in D and with
the aryl group in C.44 Preferential formation of (S)-oxindole
products with substrates in the E series (via intermediate C)
suggests that the former interaction is more severe. That higher
enantioselectivity is typically observed in the Z series would
be consistent with reinforcing destabilizing interactions in this
series.
pursued this issue through chiral amplification studies22 and
found a linear relationship between the enantiopurity of the
BINAP and enantioselectivity in insertions taking place by both
the cationic and neutral pathways (Table 2). This relationship
indicates that there is no detectable preequilibrium self-
association of the catalyst prior to entry into the catalytic cycle,
a result consistent with Amatore’s recent findings.31d,45,46
Two possibilities for the enantioselective step can be easily
dismissed. Our observation that added halide directs reactions
of triflate substrates down the neutral pathway, undoubtedly by
forming 41,24c proVides strong eVidence that oxidatiVe addition
is not reVersible. If it were, when the added halide was Cl- or
Br-, the aryl bromide or aryl chloride analogue of 39 would
have accumulated in experiments with triflate 15. Oxidative
addition, however, cannot be the enantioselective step,47 since
iodides give products of quite different enantiopurity when silver
(or thallium) salts are present. An alternate scenario in which
migratory insertion is reversible and enantioselectivity is dictated
in the penultimate â-hydride elimination step can also be ruled
out since it would be inconsistent with the dependence of
enantioselection on the geometry of the alkene substrate.
Within the neutral manifold, C-C bond formation could
occur through at least three pathways as depicted in Figure 2:
(1) Insertion from a neutral four coordinate complex 44 resulting
Enantioselective Step of the Neutral Pathway. As noted
previously, details on how a (bisphosphine)(aryl)palladium
halide complex undergoes migratory insertion in the absence
of halide scavengers are virtually nonexistent. Since the results
we initially obtained in 1992 were unexpected,5 we decided to
initially verify that monomeric Pd-BINAP was involved. We
(43) This interaction has been commented on in general terms by previous
authors.3
(44) That the phenyl groups, particularly the quasi equatorial phenyl
groups, of BINAP and related ligands play a critical role in enantiodis-
crimination has been pointed out by many authors See, inter alia.: (a)
Fryzuk, M. D.; Bosnich, B. J. Am. Chem. Soc. 1997, 99, 66262. (b) Pavlov,
V. A.; Klabunovskii, E. I.; Struchkov, T.; Voloboev, A. A.; Yanovsky, A.
I. J. Mol. Catal. 1988, 44, 217. (c) Kagan, H. B. Compr. Organomet. Chem.
1982, 8, 463. (d) Seebach, D.; Plattner, D. A.; Beck, A. K.; Wang, Y. M.;
Hunziker, D. HelV. Chem. Acta 1992, 75, 2171.
(45) Buchwald has recently isolated and characterized Pd[(R)-
BINAP](dba): Wolfe, J. P.; Wagaw, S.; Buchwald, S. L. J. Am. Chem.
Soc. 1996, 118, 7215.
(46) Formation of µ-bridged dimers with the bridging ligand being dba
or BINAP, although unlikely, is nonetheless conceivable.
(47) Oxidative addition could be the enantioselective step if the
intermediates undergoing oxidative addition were the diastereomeric Pd-
BINAP alkene complexes.

6496 J. Am. Chem. Soc., Vol. 120, No. 26, 1998
Ashimori et al.
from partial BINAP dissociation. (2) Insertion via a cationic
four coordinate complex 45 resulting from halide ionization.
(3) Direct insertion of a neutral pentacoordinate complex
represented in generic fashion by 47. Enantioselection in the
neutral pathway must occur during olefin coordination to form
intermediates 44, 45, or 47 or in the migratory insertion of these
species.
Most previous authors have assumed, or specifically sug-
gested, that the neutral Heck reaction manifold proceeds by
phosphine dissociation: the sequence 41 f 44 f 46.3,25,30 Our
studies with monophosphine analogues of BINAP were designed
to examine the validity of this supposition by mimicking a
partially dissociated BINAP chelate. Should a bisphosphine be
monocoordinated during the enantioselective step, enantiose-
lectivities obtained with monodentate models of BINAP should
be similar to those obtained with BINAP. Just the opposite
was observed. Asymmetric Heck cyclization of iodide 8 with
the monophosphine BINAP analogues 35, 36, and most
importantly 38, proceeded in low ee to produce the enantiomeric
oxindole to that produced as the major isomer using BINAP.
This result supports the conclusion that in the corresponding
reactions of BINAP both phosphines are bound to Pd during
the enantioselective step. The 41 f 44 f 46 sequence can be
ruled out.
Substitution chemistry of square planar Pd(II) complexes is
dominated by associative processes.48,49 In the neutral Heck
reaction manifold, associative olefin complexation would occur
by initial axial coordination of the alkene, as depicted in 47.48,50
At this juncture, migratory insertion could proceed in two ways,
via halide expulsion (47 f 45 f 46) or directly from a five
coordinate species (47 f 46). That a pentacoordinate inter-
mediate is conceivable gains some support from the number of
pentacoordinate Pd(II) alkene complexes that have been recently
isolated and characterized.51
Theoretical calculations of Thorn and Hoffmann predict that
migratory insertion from five coordinate Pt(II) complexes will
have higher activation barriers than from four coordinate
complexes.38 This prediction is supported by the kinetic
investigation of Samsel and Norton52 who established that
insertion of a tethered alkyne into a (Ph3P)2(aryl)Pd(II) chloride
complex proceeds by phosphine expulsion. In light of the
literature precedence available at this time, direct insertion
through a five coordinate Pd(II) intermediate appears unlikely.
Ar-Pd(PPh3)2OTf is fully dissociated in N,N-dimethylforma-
mide (DMF)32 strongly suggesting similar behavior for 42 in
DMA.53 Second, our experimental results provide no evidence
for ion pairing in the neutral pathway. For example, enanti-
oselectivity in the neutral pathway is unchanged in going from
dry DMA to DMA containing 5% water (Table 3). An identical
lack of solvent effect was seen when the neutral pathway was
entered from a triflate precursor (Table 6). We suggest that
the stereochemistry determining step of the neutral pathway
occurs during the process where halide is displaced by the
tethered alkene (41 f 47 f 45).54 This proposal is consistent
with the following experimental observations: (1) BINAP is
coordinated in bidentate fashion during the enantioselective step.
(2) Enantioselectivity is insensitive to solvent polarity. (3)
Addition of iodide (or other halides) to Heck reactions of a
triflate substrate leads to the same stereoinduction realized with
the corresponding iodide precursor in the absence of additives.
While it can be deduced that enantioselection in asymmetric
Heck reactions proceeding via the neutral pathway resides in
the process of olefin substitution, it cannot at present be surmised
which step of this undoubtedly complex pathway is stereoregu-
lating. The sequence 41 f 47 f 45 as written in Figure 2 is
a significant oversimplification of a much more complicated
process. A number of distinct pentacoordinate intermediates
undoubtedly intervene between 41 and 45.48,55 At a minimum,
an initially formed square-pyramidal complex must evolve to a
square-pyramidal isomer having X apical, the most plausible
immediate precursor of 45. In principle, the enantioselective
step could be the formation or breakdown of any of these
intermediates. A corollary to this conclusion is that it is vastly
premature to advance a three-dimensional model to rationalize
stereoinduction in asymmetric Heck reactions that proceed by
the neutral pathway.
Conclusion
This and the companion studies in this issue4,37 detail two
useful extensions of intramolecular asymmetric Heck chemis-
try: (a) use of basic amines as the HI scavenger in reactions of
iodide substrates and (b) use of halide additives to direct
reactions of triflate substrates to the neutral pathway. For
example for the first time we have shown that enantioselection
in asymmetric Heck insertions of some triflates can be dramati-
cally enhanced by the addition of halide salts. One now has
the ability to vary three important reaction parameters: stere-
ochemistry of the alkene, nature of the group undergoing
oxidative addition, and the Heck reaction manifold (cationic or
neutral). Tuning these variables will allow a wide variety of
enantioenriched products to be prepared in high enantiopurity
by intramolecular asymmetric Heck reactions. The studies
described here illustrate this versatility through the synthesis
of chiral 3,3-disubstituted-2-oxindoles having a wide range of
substituents at the quaternary carbonsMe, prim-alkyl, tert-alkyl,
and arylsin useful levels of enantiocontrol (72-97%).
Through the process of elimination, the neutral manifold is
reduced to the sequence 41 f 47 f 45 f 46. Stereoinduction
must then reside either in the preferential formation of one
diastereomer of complex 47 or 45 or in the 45 f 46 insertion
step. If insertion were the enantioselective step in both the
cationic and neutral pathways, cationic alkene complex 45
having a halide counterion would have to behave quite differ-
ently from the similar triflate complex 42. Disparate behavior,
however, would be impossible if complexes 42 and 45 were
dissociated. The evidence to date supports just such a conclu-
sion. First, conductivity measurements demonstrate that trans-
In 1992, we first reported that Pd-BINAP-catalyzed intramo-
lecular Heck reactions of some aryl halide substrates take place
(48) For reviews, see: (a) Cross, R. J. AdV. Inorg. Chem. 1989, 34, 219.
(b) Tobe, M. L. In ComprehensiVe Coordination Chemistry; Wilkinson,
G., Gillard, R. D., McCleverty, J. A., Eds.; Pergamon: Oxford, 1987; Vol.
1, pp 281-329.
(49) In certain circumstances dissociative substitution is observed. For
a review, see: Romero, R. Comments Inorg. Chem. 1990, 11, 21.
(50) The set of all five-coordinate intermediates is represented in Figure
2 by 47.
(53) Carlson, R.; Lundstedt, T.; Albano, C. Acta Chem. Scand. 1985,
39, 79.
(54) There is some precedent for olefin displacement of halide in
(bisphosphine)(aryl) Pd(halide) complexes: Portnoy, M.; Ben-David, Y.;
Rousso, I.; Milstein, D. Organometallics 1994, 13, 3465.
(55) Relevant issues are discussed in a recent study of the insertion of
norbornene into palladium-carbon bonds of complexes having bidentate
nitrogen ligands: Groen, J. H.; Delis, J. G. P.; van Leeuwen, P. W. N. M.;
Vrieze, K. Organometallics 1997, 16, 68.
(51) (a) Albano, V. G.; Castellari, C.; Cucciolito, M. E.; Panunzi, A.;
Vitagliano, A. Organometallics 1990, 9, 1269. (b) Albano, V. G.; Natile,
G.; Panunzi, A. Coord. Chem. ReV. 1994, 133, 67.
(56) General experimental details have been described.4 Conventional
silica gel flash chromatography is abbreviated as sgc.
(52) Samsel, E. G.; Norton, J. R. J. Am. Chem. Soc. 1984, 106, 5505.

Heck Reactions of (Z)-R,â-Unsaturated Anilides
J. Am. Chem. Soc., Vol. 120, No. 26, 1998 6497
2′-Iodo-N-methyl-(Z)-4-(tert-butyldimethylsiloxy)-2-methyl-2-bu-
tenanilide (8). Following a modification of the procedure of Weinreb,11
Me3Al (2.0 M in hexane, 2.9 mL, 5.8 mmol) was added dropwise to a
solution of 2-iodoaniline (1.04 g, 4.74 mmol) and dry CH2Cl2 (9.5 mL)
at 0 °C. Hydrogen evolution ensued, and the reaction was maintained
at 0 °C for 10 min and then allowed to warm to room temperature.
After 45 min, a solution of 5 (549 mg, 2.25 mmol) and dry CH2Cl2 (9
mL) was added dropwise, and the reaction was maintained at room
temperature overnight. The reaction was quenched with 1 M Rochelle’s
salt solution (10 mL) at 0 °C, and the resulting mixture was stirred for
10 min at room temperature, poured into water (50 mL), and extracted
with CH2Cl2 (3 × 50 mL). The combined organic layers were then
washed with brine (150 mL), dried (MgSO4), filtered, and concentrated.
The residue was dissolved in THF (20 mL) and added dropwise to a
stirring suspension of NaH (60% oil dispersion, 220 mg, 5.5 mmol) in
THF (20 mL) at 0 °C. After 15 min, the reaction was allowed to warm
to room temperature, and MeI (0.33 mL, 5.30 mmol) was added. The
reaction was maintained overnight at room temperature and then poured
into saturated aqueous NaHCO3 solution (50 mL), and the resulting
mixture was extracted with Et2O (2 × 30 mL) and EtOAc (30 mL).
The combined organic layers were washed with brine (100 mL), dried
(MgSO4), filtered, and concentrated. Purification of the residue by sgc
(5:1 hexanes-EtOAc) afforded 738 mg (73%) of 8 as an orange oil.
NMR analysis showed that this compound is a 2.4:1 mixture of amide
rotamers. 1H NMR (300 MHz, CDCl3) major rotamer: δ 7.88 (app d,
J ) 7.4 Hz, 1H), 7.31 (m, 2H), 7.05-7.02 (m, 1H), 5.34 (m, 1H),
4.34-4.27 (m, 2H), 3.25 (s, 3H), 1.56 (s, 3H), 0.92 (s, 9H), 0.10 (s,
6H); 13C NMR (125 MHz, CDCl3) major rotamer: δ 170.1, 145.2,
140.0, 131.8, 130.5, 129.5, 129.2, 129.1, 98.3, 61.2, 36.5, 25.9, 20.1,
-5.2; IR (film) 2952, 2927, 2855, 1649, 1578, 1470, 1381, 1360, 1252,
1102, 1058 cm-1; MS (EI) m/z 445.0933 (M, 445.0934 calcd for C18H28-
INO2Si). Anal. Calcd for C18H28INO2Si: C 48.54; H 6.34; N 3.14.
Found: C 48.57; H 6.34; N 3.20.
at 80-100 °C with high enantioselectivity in the absence of
halide scaVengers.5 We disclosed herein the first investigations
of the mechanism of this poorly understood neutral Heck
reaction manifold. Key results were the following: (1) Chiral
amplification studies show that the catalyst is monomeric Pd-
BINAP. (2) Investigations of monophosphine analogues of
BINAP, which were designed to mimic a partially dissociated
BINAP chelate, support the conclusion that BINAP is chelated
during the enantioselective step. (3) Enantioselectivity is
insensitive to solvent polarity. (4) Oxidative addition is
irreversible. From these data, we suggest that the stereochem-
istry determining step of the neutral pathway occurs during the
process in which iodide is displaced by the tethered alkene
(Figure 2, 41 f 47 f 45). In light of the variety of
pentacoordinate intermediates that could be involved, it is
premature to advance a model to rationalize stereoinduction in
asymmetric Heck reactions proceeding by the neutral pathway.
Nonetheless, the finding that asymmetric Heck reactions that
take place by the neutral pathway involve five-coordinate
intermediates broadens the vista for the design of asymmetric
ligands for this and related reactions.
Despite these advances, many aspects of the neutral Heck
reaction manifold remain poorly understood. For example,
several authors have described Heck reactions of halide
substrates that proceed by the neutral pathway with low ee.3
Critical is learning what are the structural requirements for
obtaining high enantioselectivity in the less-studied neutral
pathway of intramolecular Heck reactions. Our current inves-
tigations are directed at this and related issues.
Experimental Section56
2′-Hydroxy-(Z)-4-(tert-butyldimethylsiloxy)-2-methyl-2-butena-
nilide (12). A solution of triethylsilyl chloride (2.4 mL, 14.3 mmol),
2-aminophenol (1.03 g, 9.44 mmol), imidazole (2.51 g, 36.87 mmol),
and dry CH2Cl2 (9 mL) was stirred at room temperature for 3 h and
poured into water (20 mL). The layers were separated, and the aqueous
layer was extracted with Et2O (20 mL). The organic extracts were
combined, dried (Na2SO4), and concentrated. The resulting crude
residue was purified by sgc (9:1 to 4:1 hexanes-EtOAc) to give 1.79
g (85%) of 11 as a brown oil: 1H NMR (500 MHz, CDCl3) δ 6.81 (m,
3 H), 6.61 (app td, J ) 7.6, 1.8 Hz, 1H), 3.72 (s, 2H, NH), 1.01 (t, J
) 8.0 Hz, 9H), 0.76 (q, J ) 8.0 Hz, 6H); 13C NMR (125 MHz, CDCl3)
δ 143.0, 138.1, 121.9, 118.4, 118.3, 118.2, 115.6, 6.8, 5.3; IR (CCl4)
3478, 3388, 3032, 2960, 2877, 1742, 1613, 1592, 1503, 1414, 1277,
1218 cm-1. Anal. Calcd for C12H21NOSi: C 64.52; H 9.48; N 6.27.
Found: C 64.34; H 9.59; N 6.24.
Methyl (Z)-4-(tert-Butyldimethylsiloxy)-2-methyl-2-butenoate (5).
A solution of (Z)-3-iodo-2-buten-1-ol9 (6.4 g, 32.5 mmol) and dry CH2-
Cl2 (100 mL) was treated dropwise sequentially with 2,6-lutidine (11
mL, 94 mmol) and tert-butyldimethylsilyl triflate (11 mL, 46 mmol)
at -78 °C. The reaction was stirred at -78 °C for 3 h and poured
into saturated aqueous NH4Cl solution (80 mL). The layers were
separated, and the aqueous layer was extracted with Et2O (3 × 100
mL). The combined organic layers were washed with brine (100 mL),
dried (Na2SO4), and filtered, and the filtrates were concentrated.
Purification of the pale violet residue by sgc (49:1 hexanes-EtOAc)
gave 8.06 g (77%) of 2 as a pale orange oil: 1H NMR (300 MHz,
CDCl3) δ 5.68 (td, J ) 3.8 Hz, 1.5, 1H), 4.17 (dd, J ) 2.1, 1.5 Hz,
2H), 2.50 (d, J ) 1.4 Hz, 3H), 0.91 (s, 9H), 0.08 (s, 6H); 13C NMR
(125 MHz, CDCl3) δ 135.3, 99.1, 68.5, 33.5, 25.9, 18.3, -5.1; IR (film)
2959, 2925, 2875, 2851, 1466, 1459, 1378 cm-1
.
A hexane solution of Me3Al (2.0 M, 14 mL) was added dropwise to
a cooled solution of 11 (6.67 g, 30 mmol) and dry toluene (30 mL) at
0 °C. The resulting dark green solution was maintained at 0 °C for 10
min and then allowed to warm to room temperature. After 45 min, a
solution of methyl ester 5 (7.35 g, 30 mmol) and dry toluene (15 mL)
was added dropwise. The reaction was maintained at room temperature
for 4 h, warmed to 60 °C for 2 h, cooled to 0 °C, and quenched by
slow addition of 1 M aqueous Rochelle’s salt solution (100 mL). The
resulting mixture was stirred overnight at room temperature and then
poured into EtOAc (50 mL), and the layers were separated. The
aqueous layer was extracted with Et2O (100 mL), and the organic layers
were combined, washed with brine (200 mL), dried (MgSO4), filtered,
and concentrated. A solution of this crude anilide and dry THF (10
mL) was cooled to 0 °C and added to a mixture of NaH (60% oil
dispersion, 1.6 g, 42 mmol, prewashed with hexane) and THF (10 mL)
at 0 °C, and the resulting mixture was stirred for 20 min and then
allowed to warm to room temperature. After 3 h the reaction was
quenched with ice and poured into saturated aqueous NH4Cl solution
(50 mL), and the layers were separated. The aqueous layer was
extracted with EtOAc (50 mL), and Et2O (50 mL) and the combined
organic layers were washed with brine (100 mL), dried (Na2SO4),
filtered, and concentrated. The resulting brown oil was purified by
sgc (9:1 to 4:1 hexanes-EtOAc) to give 7.0 g (72% from 5) of 12 as
A solution of N,N-diisopropylethylamine (10 mL, 57 mmol) and
methanol (24 mL, 0.6 mmol) was purged with CO (3×) at room
temperature. A solution of 2 (3.72 g, 11.9 mmol) and dry DMF (30
mL) was added, the resulting solution was purged again with CO (3×),
and Pd(PPh3)4 (1.38 g, 1.2 mmol) was added in one portion followed
by further purging with CO (3×). After 1 h, the reaction became a
red homogeneous solution which was maintained overnight at room
temperature, whereupon it became a thick yellow slurry. The reaction
was treated with saturated aqueous NaHCO3 (20 mL), the resulting
solids were removed by filtration, the filtrate was extracted with hexanes
(3 × 50 mL), and the organic layers were combined and concentrated.
The residue was dissolved in THF (30 mL), and a 3:1 water-bleach
solution (60 mL) was added. The reaction was maintained for 30 min
and then extracted with hexanes (3 × 50 mL). The combined organic
layers were washed with 1:1 brine-water solution (2 × 100 mL), dried
(MgSO4), filtered, and concentrated. Purification of the residue by sgc
(12:1 hexanes-EtOAc) gave 1.94 g (67%) of 5 as a pale yellow oil:
1H NMR (300 MHz, CDCl3) δ 6.17-6.10 (m, 1H), 4.59 (app d, J )
17 Hz, 2H), 3.71 (s, 3H), 2.21 (s, 3H), 0.89 (s, 9H), 0.06 (s, 6H); 13C
NMR (125 MHz, CDCl3) δ 167.7, 146.4, 125.4, 62.1, 51.4, 25.9, 19.7,
18.3; IR (film) 2956, 2931, 2887, 1720, 1483, 1367, 1333, 1254, 1228,
1141, 1104 cm-1
.

6498 J. Am. Chem. Soc., Vol. 120, No. 26, 1998
Ashimori et al.
an amber oil which solidified upon standing: mp 69-71 °C; 1H NMR
(300 MHz, CDCl3) δ 9.21 (br s, 1H), 9.09 (s, 1H), 7.14 (app td, J )
7.6, 1.3 Hz, 1H) 7.04 (dd, J ) 7.6, 0.9 Hz, 1H), 6.96 (dd, J ) 7.9, 1.2
Hz, 1H), 6.85 (app td, J ) 7.9, 1.2 Hz, 1H), 6.10 (app td, J ) 7.5, 1.2
Hz, 1H), 4.31 (d, J ) 7.5 Hz, 2H), 2.06 (s, 3H), 0.90 (s, 9H), 0.12 (s,
6H); 13C NMR (125 MHz, CDCl3) δ 168.0, 149.0, 134.4, 127.1, 125.8,
122.6, 120.3, 119.6, 59.6, 25.9, 20.5, 18.3, -5.0; IR (film) 3276, 2954,
2929, 2857, 1664, 1633, 1598, 1533, 1454, 1256, 1059 cm-1. Anal.
Calcd for C17H27NO3Si: C 63.51; H 8.47; N, 4.36. Found: C 63.43;
H 8.48; N 4.30.
to room temperature over 30 min, then recooled to 0 °C, and quenched
with saturated aqueous citric acid solution (50 mL). The layers were
separated, and the organic layer was dried (MgSO4) and concentrated
to afford a dark green residue which was purified by sgc (6:1 to 4:1,
hexanes-EtOAc) to give 6.40 g (82%) of 15 as a yellow oil; NMR
analysis of this compound showed it to be a 1.2:1 mixture of amide
rotamers. Triflate 15 was stored in a refrigerator, since it slowly
decomposed when stored at room temperature in the light. 1H NMR
(300 MHz, CDCl3) major rotamer: δ 7.48-7.33 (m, 4H), 5.34 (app t,
J ) 6.0 Hz, 1H), 4.13 (app d, J ) 6.1 Hz, 2H), 3.32 (s, 3H), 1.58 (s,
3H), 0.80 (s, 9H), 0.06 (s, 6H); 13C NMR (125 MHz, CDCl3) major
rotamer: δ 169.9, 143.8, 135.0, 131.5, 131.3, 128.2, 128.0, 121.4, 60.0,
36.0, 25.0, 19.0, 17.3, -6.3; IR (film) 2955, 2930, 2858, 1657, 1605,
1582, 1423, 1249, 1140, 1082 cm-1; MS (EI) m/z 468.1486 (MH,
468.1488 calcd for C19H29NO5 F3SSi). Anal. Calcd for C19H28NO5F3-
SSi: C 48.81; H 6.04; N 3.00. Found: C 48.81; H 6.00; N 2.99.
2′-Iodo-N-methyl-(Z)-4-(triisopropylsiloxy)-2-phenyl-2-butenanil-
ide (20). A pentane solution of t-BuLi (1.6 M, 24 mL, 39 mmol) was
added to a solution of crude 16 (12 g of a 2:1 mixture of 16 and the
corresponding TIPS-protected propargyl ether) and dry Et2O (100 mL)
at -78 °C. The reaction was maintained at -78 °C for 1 h, and then
a solution of 2-iodophenylisocyanate (9.0 g, 40 mmol)14 and Et2O (45
mL) was added rapidly. The reaction was maintained at -78 °C for
10 min, allowed to warm to room temperature over 3 h, and then
quenched with water (100 mL). The layers were separated, and the
aqueous layer was extracted with EtOAc (2 × 100 mL). The combined
organic layers were washed with brine (200 mL), dried (MgSO4), and
concentrated. The crude secondary amide was purified by sgc (19:1
hexanes-EtOAc) to give 8.8 g (86%, corrected for the purity of 16)
of an opaque oil.
A solution of the above product (1.92 g, 3.59 mmol) and dry THF
(20 mL) was added to a stirring suspension of NaH (60%, 250 mg, 6.2
mmol) and THF (20 mL) over 30 min at 0 °C. The resulting suspension
was stirred at 0 °C for 20 min, MeI (1.5 mL, 21 mmol) was added in
one portion, and the cooling bath was removed. The reaction was
stirred for an additional 2 h at room temperature and quenched with
saturated aqueous NH4Cl solution (25 mL). The layers were then
separated, the aqueous layer was extracted with EtOAc (50 mL), and
the combined organic layers were washed with brine (100 mL), dried
(MgSO4), and concentrated. Purification of the residue by sgc (19:1
hexanes-EtOAc) afforded 1.54 g (78%) of 20 as an amber oil; NMR
analysis indicates that this compound is a 2.4:1 mixture of rotamers.
1H NMR (300 MHz, CDCl3) major rotamer: δ 7.59 (dd, J ) 1.2, 8.9
Hz, 1H), 7.44-7.34 (m, 2H), 7.25-7.00 (m, 4H), 6.90 (dd, J ) 1.3,
7.3, Hz, 1H), 6.85 (dd, J ) 1.8, 10.4 Hz, 1H), 5.76 (t, J ) 6.3 Hz,
1H), 4.61 (d, J ) 6.3 Hz, 2H), 3.30 (s, 3H), 1.11 (m, 21H); 13C NMR
(125 MHz, CDCl3) δ 169.4, 145.5, 139.9, 137.4, 134.8, 131.3, 130.9,
129.3, 129.1, 128.7, 128.4, 126.8, 98.6, 62.0, 39.1, 18.4, 12.3; IR (film)
3477, 3056, 2939, 2859, 2714, 1639, 1574, 1465, 1370, 1247, 1072,
1014, 876 cm-1. Anal. Calcd for C26H36INO2Si: C 56.82; H 6.60; N
2.25. Found: C 56.59; H 6.53; N 2.51.
2′-(Allyloxy)-N-methyl-(Z)-4-(tert-butyldimethylsiloxy)-2-methyl-
2-butenanilide (13). Following the general method of McKillop,12
a
mixture of anilide 12 (1.42 g, 4.42 mmol), allyl bromide (0.9 mL, 10
mmol), tetrabutylammonium bromide (150 mg, 0.47 mmol), 1 M
aqueous NaOH (6 mL), CH2Cl2 (20 mL), and water (14 mL) was
wrapped with aluminum foil to prevent exposure to light and the
resulting yellow-orange mixture was stirred vigorously for 12 h at room
temperature. The reaction was allowed to stand, and the layers were
separated. The aqueous layer was extracted with CH2Cl2 (2 × 20 mL),
and the combined organic layers were washed with saturated aqueous
citric acid solution (100 mL) and brine (100 mL), dried (Na2SO4),
filtered, and concentrated.
A mixture of this crude residue (1.37 g, 3.79 mmol), NaH (60% oil
dispersion, 250 mg, 6.2 mmol), and dry THF (16 mL) was stirred at 0
°C for 10 min and allowed to warm to room temperature over 3 h.
Methyl iodide (0.7 mL, 1.6 g, 11 mmol) was then added in one portion,
and the mixture was stirred overnight at room temperature. Workup
as described for the preparation of 8 and purification by sgc (7:1 to
4:1 to 2:1 hexanes-Et2O) gave 1.05 g (74%) of 13 as a golden oil;
NMR analysis indicates that this compound is a 10:1 mixture of amide
rotamers. 1H NMR (500 MHz, CDCl3) major rotamer: δ 7.24 (app
td, J ) 7.9, 1.6 Hz, 1H), 7.16 (dd, J ) 7.6, 1.5 Hz, 1H), 6.88 (app t,
J ) 6.3 Hz, 1H), 6.86 (m, 1H), 6.02 (m, 2H), 5.43 (dd, J ) 17.2, 1.3
Hz, 1H), 5.28 (dd, J ) 10.4, 1.2 Hz, 1H), 4.56 (d, J ) 1.0 Hz, 2H),
4.29-4.08 (m, 2H), 3.23 (s, 3H), 1.59 (s, 3 H), 0.9 (s, 9H), 0.09 (s,
6H); 13C NMR (125 MHz, CDCl3) major rotamer: δ 170.5, 152.9,
132.3, 131.6, 130.9, 129.4, 128.8, 128.1, 119.7, 116.7, 116.6, 67.8,
60.2, 34.7, 25.0, 19.3, -6.2; IR (film) 2954, 2928, 2856, 1643, 1596,
1501, 1453, 1423, 1348, 1275, 1255, 1096, 1060 cm-1. Anal. Calcd
for C21H33NO3Si: C 66.44; H 8.64; N 3.87. Found: C 66.67; H 8.59;
N 3.98.
2′-(Hydroxy)-N-methyl-(Z)-4-(tert-butyldimethylsiloxy)-2-methyl-
2-butenanilide (14). Following a modification of the procedure of
Deziel,13 a mixture of anilide 13 (1.5 g, 4.1 mmol), Pd(PPh3)4 (140
mg, 0.12 mmol), Ph3P (42 mg, 0.16 mmol), and dry MeCN (16 mL)
was stirred at 0 °C for 5 min and a solution of pyrrolidine (1.4 mL, 17
mmol) and dry MeCN (16 mL) was added dropwise. The resulting
mixture was stirred at 0 °C for 10 min and then at 40 °C for 2 h. After
cooling to room temperature, the reaction was poured into a saturated
aqueous citric acid solution (20 mL), and the layers were separated.
The organic layer was washed with saturated aqueous citric acid solution
(20 mL) and brine (20 mL), dried (Na2SO4), and concentrated. The
resulting crude residue was purified by sgc (7:1 to 4:1 to 2:1 hexanes-
EtOAc) to give 1.15 g (83%) of 14 as a viscous yellow oil; NMR
analysis indicates that this compound is a 9:1 mixture of amide rotamers.
1H NMR (300 MHz, CDCl3) major rotamer: δ 8.20 (br s, 1 H), 7.21
(m, 1H), 7.08 (dd, J ) 7.8, 1.6 Hz, 1H), 6.96 (dd, J ) 8.2, 1.4 Hz,
1H), 6.87 (app td, J ) 7.6, 1.4 Hz, 1H), 5.12 (m, 1H), 4.29 (m, 2H),
3.21 (s, 3H), 1.66 (d, J ) 1.7 Hz, 1H), 0.95 (s, 9H), 0.19 (s, 6H); 13C
NMR (125 MHz, CDCl3) major rotamer: δ 171.5, 152.1, 131.8, 129.3,
128.6, 127.9, 126.7, 119.2, 117.1, 61.1, 35.0, 25.1, 20.2, 17.7, -6.2;
IR (film) 3172, 2886, 2857, 1741, 1620, 1588, 1513, 1461, 1392, 1291,
1256, 1094 cm-1. Anal. Calcd for C18H29NO3Si: C 64.44; H 8.71; N
4.17. Found: C 64.33; H 8.76; N 4.12.
Heck Cyclizations. Cyclizations reported in Tables 1-4 were
carried out using the standard conditions described in the preceding
paper.4 Experiments reported in Tables 5-7 were conducted using
more rigorous Schlenk techniques. The following procedure is
representative.
General Procedure for Pd-BINAP Catalyzed Cyclizations of
Iodide 8 and Triflate 15 in the Presence of Additives. A base-washed
and flame-dried 10 mL Schlenk flask was fitted with a three-way
stopcock. The flask was charged with Pd2(dba)3‚CHCl3 (8.1 mg, 0.008
mmol) and (R)-BINAP (11.2 mg, 0.018 mmol) and purged under an
Ar flow for 10 min. Dry DMA (0.8 mL) was added, and the resulting
purple-brown suspension was stirred for 2 h to give a bright orange
solution. A solution of triflate 15 (74.6 mg, 0.16 mmol), PMP (120
µL, 100 mg, 0.66 mmol), and dry DMA (0.4 mL) was added followed
by tetrabutylammonium chloride monohydrate (53 mg, 0.18 mmol) and
additional DMA (0.4 mL). The resulting suspension was stirred for
15 min, the resulting red-orange solution was frozen at -78 °C, and
the apparatus was evacuated (0.1-0.2 mm), sealed, and allowed to
warm to room temperature. This freeze-pump-thaw cycle was
repeated four times. The reaction then was heated at 100 °C for 23 h
2′-(Trifluoromethanesulfonyloxy)-N-methyl-(Z)-4-(tert-butyldi-
methylsiloxy)-2-methyl-2-butenanilide (15). A flame dried, 250 mL
round-bottom flask was wrapped in aluminum foil and charged with
phenol 14 (5.6 g, 17 mmol), pyridine (8 mL), and dry CH2Cl2 (65 mL).
The resulting yellow solution was cooled to 0 °C, and freshly distilled
triflic anhydride (5.0 mL, 29 mmol) was added dropwise (the reaction
turned dark green). After 30 min, the reaction was allowed to warm

Heck Reactions of (Z)-R,â-Unsaturated Anilides
J. Am. Chem. Soc., Vol. 120, No. 26, 1998 6499
under an argon atmosphere. After cooling to room temperature, the
reaction was quenched with saturated aqueous NaHCO3 solution (5
mL) and diluted with water (10 mL) and EtOAc (10 mL). The layers
were separated, and the aqueous layer was extracted with EtOAc (2 ×
20 mL). The combined organic extracts were washed with brine (50
mL), dried (Na2SO4), filtered, and concentrated. The resulting dark
brown residue was partially purified by sgc (19:1 to 4:1 hexanes-
EtOAc) to give a bright yellow oil that was a mixture of enoxysilane
stereoisomers contaminated with residual dba. Capillary GLC analysis
indicated a 20:1 ratio of E to Z enoxysilane stereoisomers.4
This mixture was dissolved in THF (2 mL), 3 M HCl (2 mL) was
added, and the reaction was maintained overnight at room temperature
and then quenched with saturated NaHCO3 aqueous solution (5 mL).
The layers were separated, the aqueous layer was extracted with EtOAc
(5 mL), and the combined organic layers were dried (Na2SO4), filtered,
and concentrated. The resulting bright yellow oil was purified by sgc
(4:1 to 2:1 hexanes-EtOAc) to give 21 mg (66%) of oxindole aldehyde
25.
mmol), MgSO4 (900 mg), and THF (25 mL). After 17 h, LiAlH4 (465
mg, 12.3 mmol) was added, and the resulting mixture was refluxed for
90 min. Excess LiAlH4 was decomposed by dropwise addition of
EtOAc (30 mL) followed by saturated aqueous NaHCO3 (30 mL). The
phases were separated, the aqueous layer was extracted with EtOAc (2
× 50 mL), and the combined organic extracts were washed with brine
(80 mL), dried (Na2SO4), concentrated, and the residue was purified
by sgc (99:1 to 20:1 CHCl3-MeOH) to give 152 mg (61%) of
pyrroloindoline 32.
Following the procedure of Fuji and co-workers,19 a solution of
N-bromosuccinimide (100 mg, 0.56 mmol) and DMF (2 mL) was added
to a solution of 32 (113 mg, 0.56 mmol) and DMF (2 mL) at room
temperature. The reaction was stirred at room temperature for 27 h,
poured into water (175 mL), and extracted with CH2Cl2 (30 mL, 5×).
The combined organic extracts were washed with brine (80 mL), dried
(Na2SO4), and concentrated, and the residue was purified by sgc (99:1
to 20:1 CHCl3-MeOH) to give 137 mg (87%) of bromopyrroloindoline
33.19
(R)-2-Diphenylmethane-2′-diphenylphosphino-1,1′-binaphthyl (38).
A solution of phenyllithium (1.4 M in Et2O-cyclohexane, 7.5 mL, 10
mmol) was added dropwise using a syringe pump to a cooled solution
of ester 3727 (1.02 g, 2.05 mmol) and dry THF (10 mL) at 0 °C. The
reaction was maintained at 0 °C for 20 min and then allowed to warm
to room temperature. After 40 min, the reaction was poured into a
mixture of ice and 1 M HCl (20 mL). The layers were separated, and
the aqueous layer was extracted with Et2O (15 mL) and EtOAc (15
mL). The organic layers were combined and washed with 1 M HCl
(20 mL) and brine (20 mL). The dried (MgSO4) organic layer was
concentrated, and the resulting residue was adsorbed onto silica gel
and chromatographed (19:1 to 9:1 cyclohexanes-Et2O) to give 791
mg (62%) of the corresponding carbinol as a pale yellow solid.
Diborane was generated by the dropwise addition of BF3‚OEt2 (1.6
mL, 13 mmol) to a solution of NaBH4 (0.35 g, 9.3 mmol) and dry
diglyme. The resulting diborane gas was bubbled into a cooled solution
of the carbinol (104 mg, 0.168 mmol) and dry CH2Cl2 (1 mL) at 0 °C,
and the reaction was maintained at 0 °C for 1 h. The resulting colorless
solution was added to a freshly prepared suspension of trifluoroacetic
acid (1.8 mL, 24 mmol), NaBH4 (0.30 g, 7.9 mmol), and dry CH2Cl2
(5 mL) at 0 °C. After 5 min at 0 °C, the reaction was quenched with
water (10 mL) and poured into a mixture of ice and 1 M NaOH (20
mL). The layers were separated, the aqueous layer was extracted with
Et2O (10 mL), and the combined organic layers were washed with brine
(20 mL), dried (MgSO4), filtered, and concentrated. The crude residue
was purified by sgc (hexanes to 19:1 hexanes-Et2O) to give 81 mg
(80%) of 36 as a colorless solid: 1H NMR (500 MHz, CDCl3) δ 7.92
(d, J ) 8.6 Hz, 1H), 7.87 (d, J ) 8.5 Hz, 1H), 7.80 (app t, J ) 7.7 Hz,
2H), 7.47-7.44 (m, 2H), 7.31 (t, J ) 7.4 Hz, 1H), 7.23 (app d, J )
7.8 Hz, 1H), 7.21-7.08 (m, 18 H), 6.82 (app t, J ) 7.5 Hz, 1H), 6.74
(app t, J ) 7.2 Hz, 1H), 6.67 (m, 3H), 6.60 (app d, J ) 8.4 Hz, 1H),
5.41 (s, 1H); MS (CI) m/z 604.2345 (M, 604.2320 calcd for C45H33P).
(+)-Esermethole (34). Triethylamine (1.70 mL, 12.2 mmol) was
added dropwise at room temperature to a stirring mixture of (R)-
aldehyde 25 (249 mg, 1.23 mmol; 92% ee by Chiral HPLC analysis of
derived primary alcohol), methylamine hydrochloride (827 mg, 12.2
Chips of sodium metal (150 mg, 6.52 mmol) were cautiously added
to methanol (1.5 mL) to generate NaOMe. A solution of bromopyr-
roloindoline 33 (165 mg, 0.59 mmol) and DMF (2 mL) was then added
followed immediately by cuprous iodide (220 mg, 1.2 mmol), and the
resulting mixture was heated at 120 °C for 15 h. After cooling to room
temperature, the reaction mixture was filtered through Celite with DMF
and concentrated. The concentrate was partitioned between CH2Cl2
(15 mL) and 2% NaOH (15 mL), and the aqueous layer was washed
with CH2Cl2 (2 × 15 mL). The combined organic extracts were washed
with brine (20 mL), dried (Na2SO4), and concentrated, and the residue
was purified by sgc (99:1 to 20:1 CHCl3-MeOH) to give 100 mg (74%)
25
of (+)-esermethole (34): [R]D +117°, [R]450 +284°, [R]435 +239°,
[R]546 +139°, [R]577 +122° (c 0.14, benzene).
Acknowledgment. This research was supported NIH Grant
GM-30859 and the Green Cross Corporation. Additional support
from the National Science and Engineering Research Council
of Canada through a Postdoctoral Fellowship to B.B., from the
American Cancer Society through a Postdoctoral Fellowship
to M.C., from the U.S. DOEd through a GAANN Predoctoral
Fellowship to D.J.P., and from Merck, Pfizer, Roche Bio-
sciences, and SmithKline Beecham and are gratefully acknowl-
edged. NMR and mass spectra were determined at Irvine using
instruments acquired with the assistance of NSF and NIH Shared
Instrumentation Programs. We also wish to thank Professor T.
Keith Hollis (UCR) for his insightful discussion.
Supporting Information Available: Experimental proce-
dures and characterization data for new compounds not de-
scribed in the Experimental Section (8 pages, print/PDF). See
any current masthead page for ordering information and Web
access instructions.
JA980787H