Palladium complexes of phosphinamine ligands in the intramolecular asymmetric Heck reaction

Article abstract of DOI:10.1016/j.jorganchem.2003.09.045

The synthesis of two novel cyclisation substrates for the asymmetric intramolecular Heck reaction is reported. Their cyclisation, in addition to a known substrate for cis-decalin formation, were tested with palladium complexes of BINAP and heterobidentate oxazoline-containing ligands. In general BINAP provides a more active catalyst system for the range of substrates tested although excellent enantioselectivities of up to 85% were obtained with the P,N ligands studied. A trend was noted whereby the t-leucine-derived oxazoline ligands were more reactive and enantioselective than the valine-derived analogues. Similarly, the diphenylphosphinoferrocenyloxazoline ligands were more reactive and selective than the corresponding diphenylphosphinophenyloxazoline ligands.

Full text of DOI:10.1016/j.jorganchem.2003.09.045

Journal of Organometallic Chemistry 687 (2003) 545Á561
/
www.elsevier.com/locate/jorganchem
Palladium complexes of phosphinamine ligands in the intramolecular
asymmetric Heck reaction
Denis Kiely, Patrick J. Guiry *
Centre for Synthesis and Chemical Biology, Conway Institute of Biomolecular and Biomedical Research, Department of Chemistry, University College
Dublin, Belfield, Dublin 4, Ireland
Received 8 September 2003; accepted 16 September 2003
Dedicated with respect to Jean-Pierre Genet * ‘Allez les bleus’
/
Abstract
The synthesis of two novel cyclisation substrates for the asymmetric intramolecular Heck reaction is reported. Their cyclisation, in
addition to a known substrate for cis-decalin formation, were tested with palladium complexes of BINAP and heterobidentate
oxazoline-containing ligands. In general BINAP provides a more active catalyst system for the range of substrates tested although
excellent enantioselectivities of up to 85% were obtained with the P,N ligands studied. A trend was noted whereby the t-leucine-
derived oxazoline ligands were more reactive and enantioselective than the valine-derived analogues. Similarly, the diphenylpho-
sphinoferrocenyloxazoline ligands were more reactive and selective than the corresponding diphenylphosphinophenyloxazoline
ligands.
# 2003 Elsevier B.V. All rights reserved.
Keywords: Palladium; Asymmetric catalysis; Intramolecular Heck reaction; Phosphinamine ligands
1. Introduction
of fused ring systems such as decalins 1 [19], hydrindans
2 [20], indolizidines 3 [21] and diquinanes 4Á5 [22] (Fig.
1).
(R)-BINAP (6) proved to be the ligand of choice in
the majority of the intramolecular Heck reactions
studied although the related diphosphines BINAs (7)
[23], BITIANP (8), TMBTP (9) [24] and BPPFOH (10)
[21] have compared favourably to BINAP in selected
cyclisations (Fig. 2).
/
The Heck reaction is a versatile and useful palladium-
catalysed elaboration of substituted alkenes by direct
CÃ
/
C bond formation at a vinylic carbon centre [1Á4].
/
Its potential has been exploited in the key steps of many
total syntheses [5] and a better understanding of the
reaction mechanism continues to emerge [6Á9]. Both
intramolecular and intermolecular asymmetric variants
have been extensively studied and have been the subject
/
Pfaltz and co-workers described the application of
palladium complexes of diphenylphosphinooxazoline
of numerous reviews [10Á17]. The asymmetric intramo-
/
lecular Heck reaction was reported initially in 1989
when Overman and Shibasaki independently described
enantioselective cyclisations [18,19]. The asymmetric
induction obtained was due to palladium complexes of
the diphosphine ligands, DIOP and (R)-BINAP, respec-
tively. Since those initial reports the methodology has
been applied to the enantioselective synthesis of a range
ligands 11Á
with 2,3-dihydrofuran [25,26]. Hallberg subsequently
employed ligands 11Á12 in his enamide synthesis
/
12 to the intermolecular Heck reaction
/
obtaining both excellent enantioselectivities and good
regioselectivities [27]. More recently, Busacca and co-
workers have applied related phosphinoimadazoline
ligands to the preparation of spirooxindoles with
enantioselectivities of up to 88% [28]. Metal complexes
of ferrocene-based planar chiral ligands have proven to
be highly effective for a range of catalytic asymmetric
transformations [29]. Diphenylphosphinoferrocenyloxa-
* Corresponding author. Tel.: ꢀ
2127.
E-mail address: p.guiry@ucd.ie (P.J. Guiry).
/
353-1-716-2309; fax: ꢀ353-1-716-
/
0022-328X/03/$ - see front matter # 2003 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2003.09.045

546
D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á561
/
Fig. 1. Products of intramolecular asymmetric Heck reactions.
zoline ligands 13Á
/
14 were first reported by the groups of
2. Results and discussion
Uemura and Richards [30,31]. We had independently
synthesised these ligands and observed enantioselectiv-
ities of up to 92% using ligand 14 in the palladium-
catalysed allylic alkylation of 1,3-diphenyl-propenylace-
tate [32] and up to 72% ee in the allylic amination of
ethyl (2E)-1,3-diphenylprop-2-enyl carbonate [33].
More recently, we have demonstrated the efficacy of
We intended to test the efficiency of palladium
complexes of ligands 11Á14 with both novel and known
cyclisation substrates (15, 17Á18) for comparison pur-
poses with related ligands and to expand the range of
substrates studied (Fig. 3).
Specifically, we prepared the novel triflate 15 as the
potentially complicated double bond isomerisation of
the Heck reaction product is avoided with this substrate.
This cyclisation precursor is therefore an excellent
substrate for the direct screening and comparison of a
/
/
the heterobidentate PÁ
molecular Heck reaction with both 2,3-dihydrofuran
and 2,2-diaikyl-2,3-dihydrofurans as substrates [34Á37].
This led us to believe that P,N ligands 11Á14 would have
/
N systems 11Á/14 in the inter-
/
/
range of ligandÁpalladium complexes. Triflate 15 was
/
potential in the Pd-catalysed asymmetric intramolecular
Heck reaction and we now wish to report in full our
results on their application to this transformation
[38,39].
chosen as during the catalytic cycle the triflate anion
spontaneously dissociates and acts as a counterion, thus
stabilising the proposed cationic palladium intermediate
Fig. 2. Ligands applied in the intramolecular asymmetric Heck reaction.

D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á
/
561
547
Fig. 3. Substrates for the intramolecular asymmetric Heck reaction.
and allowing the ligand to remain bidentate [40,41]. This
cationic pathway proposal by Cabri [40] to effectively
induce enantioselectivity in the Heck reaction was
subsequently challenged by Overman where an appar-
ently neutral pathway afforded high levels of enantios-
electivities [42]. The proposed cationic square planar
palladium intermediate may also be accessed from the
corresponding iodide 16, but generally requires the
addition of silver salts to remove the halide ligand
from the palladium coordination sphere. The novel
triflate 17 was prepared in order to investigate the
relative regioselectivity and stereoselectivity of cyclisa-
tion induced by palladium complexes of bisphosphine or
phosphinamine ligands respectively, since with this
substrate double bond isomerisation of the Heck reac-
tion product is possible. Finally the alkenyl triflate 18
was screened using palladium complexes of phosphina-
mine ligands in order to further investigate the scope of
heterobidentate ligands in asymmetric Heck reactions.
This substrate had previously been extensively studied
by the group of Shibasaki [20,43] using BINAP as the
source of chiral induction, which would provide a useful
comparative source for our study with phosphinamine
ligands.
triflic anhydride and dimethylaminopyridine in dichlor-
omethane, a protocol which we have used in our
synthesis of axially chiral P,N ligands [46]. The required
novel cyclic substrate 15 was thus prepared in five steps
from carboxylic acid 19 in an overall yield of 35%.
2.2. Intramolecular Heck studies of substrate 15
Prior to the commencement of the study of the
asymmetric cyclisation of substrate 15 it was necessary
to prepare the cyclisation product 24 as a racemate in
order to establish conditions for the separation of the
enantiomers (Scheme 2).
The cyclised product was formed in a 90% isolated
yield, although the reaction employed a high catalyst
loading of 20 mol%. With this racemic mixture in hand
separation of the enantiomers was now possible. Se-
paration by chiral HPLC was initially attempted,
however despite trying three different chiral columns
(Daicel, Chiralpak OJ, OD, AD), sufficient baseline
separation was not achieved. Resolution of the enantio-
1
mers could however be achieved by H-NMR spectro-
scopy in the presence of the chiral shift reagent tris[3-
(heptafluoropropylhydroxymethylene)-(ꢀ)-camphora-
/
to]europium(III). The N-methyl singlets of the two
diastereomers formed appeared at different chemical
2.1. Synthesis of substrate 15
shifts moving from 3.21 for the enantiomers to 3.6Á
/4.0
2-Trifluoromethanesulfonyloxy-(1,4-dioxaspir-
for the diastereomers.
The results of our investigations on the asymmetric
cyclisation of substrate 15 with BINAP 6 and the P,N
o[4.5]dec-7-en-8-ylcarbonyl)-N-methylaniline (15) was
prepared in five steps from the known carboxylic acid,
1,4-dioxaspiro[4.5]dec-7-ene-8-carboxylic acid (19) [42]
(Scheme 1). Coupling via the acid chloride with 2-
aminophenol gave the corresponding enamide 20 in 60%
yield. The remaining functional group manipulations
required were amide N-methylation and conversion of
the phenol to an aryl triflate. Methylation without
protection of the alcohol resulted in significant compe-
titive O-methylation. Following the general procedure
of McKillop the phenol was protected as an allyl-ether
21 [44] and subsequent N-methylation yielded the
required amide 22 in 85% yield. Following the procedure
of Deziel, the allyl-ether was deprotected using catalytic
quantities of tetrakis(triphenylphosphine)palladium(0)
generating 23 in 84% yield [45]. Conversion of the
phenol to the triflate 15 was achieved in 82% yield using
ligands 11Á14 (Scheme 3), are given in Table 1.
/
BINAP 6 was determined by Overman to be the
ligand of choice in the cyclisation of the analogous aryl
iodide giving an optimal enantioselectivity of 71% in
81% yield [43]. Following a similar procedure in the
present study, cyclisations of 15 were carried out using 5
mol% of palladium precursor (Pd2(dba)3), 10 mol% of
(R)-BINAP 6 and three equivalents of base (Scheme 3).
Palladium complexes of ligand 6 were generated in situ
prior to addition of solvent and base and the formation
of the catalyst was assumed to be complete when an
orange homogenous solution had formed (typically this
took 1 h). The Heck reaction time was 2 days (to allow
direct comparison with the iodide analogue) and the

548
D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á
/
561
Scheme 1. Synthesis of substrate 15.
reactions were observed to be proceeding when the
precipitation of a salt (base.HOTf) was observed.
sations of the iodide analogue with an optimal ee of 71%
compared to 45% for the triflate variant.
Heck reactions catalysed by palladium complexes of
BINAP 6, using toluene as solvent and PMP as base
produced cyclised product in 23% yield and 35% ee
(entry 1). A switch of solvent to benzene resulted in an
increase in enantioselectivity to 45% with a similar
reaction yield of 24% (entry 2). A combination of
diisopropylethylamine and benzene resulted in an im-
proved yield of 40% but again reduced enantioselectivity
of 35% (entry 3). A combination of N,N-dimethylace-
tamide (DMA) and proton sponge promoted cyclisation
in a reasonable yield of 38%, but with a poor 7% ee
(entry 4). In Overman’s study of the silver salt promoted
Cyclisations promoted by palladium complexes of the
diphenylphosphinooxazoline ligand 11 were initially
screened with PMP as base and toluene as solvent
affording spirooxindole product in a poor 9% yield and
31% ee (entry 5). A slight increase in yield to 14% was
observed for the corresponding reaction in benzene but
also a slight decrease in ee to 27% (entry 6). Employing
DMA as solvent gave an increased yield of 30% but
again with a poor enantioselectivity of 25% (entry 7).
Using toluene as solvent and triethylamine as base the
enantioselectivity was improved to 46%, however with
low catalyst activity (13% yield, entry 8). The corre-
sponding palladium complexes of the i-propyl-substi-
tuted ligand 12 were less reactive and enantioselective
than their t-butyl analogues with poor yields and
enantioselectivities being observed under all of the
conditions screened with our optimal result being
obtained in toluene with PMP as base (entry 9). Similar
cyclisation of the analogous aryl iodide, yields of ꢀ80%
/
were routinely obtained compared to a maximum yield
of 40% in the present investigation [43]. However, our
results are not completely unexpected as aryl triflates are
generally less reactive than the corresponding aryl
iodides. Enantioselectivities were also superior in cycli-
Scheme 2. Preparation of product 24 as a racemate.

D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á
/
561
549
Scheme 3. Asymmetric cyclisation of substrate 15.
lowering of reactivity and asymmetric induction in going
from t-butyl- to i-propyl-substituted analogues was
previously noted by Pfaltz in their study of the inter-
molecular asymmetric Heck reaction. Although yields
obtained in reactions catalysed by palladium complexes
benzene resulted in slightly lower yields and enantios-
electivities compared to the analogous reactions in
toluene. When Pd(OAc)2 was used as the catalyst
precursor a lower yield and enantioselectivity were
observed (compare entries 10 and 15). A pronounced
difference in reactivity and enantioselectivity was again
observed between palladium complexes of 13 and 14,
with the more sterically demanding t-butyl-substituted
ligand affording a significantly more reactive and also
more selective catalyst. The palladium complexes of 14
with toluene and PMP afforded cyclised product in 14%
yield and 39% ee (entry 16) and other combinations of
solvent and base tested gave yields of less than 5%.
However, despite one example (entry 12) the yields
obtained with this novel substrate were poor to moder-
ate and it may be that the ketal ring, which acts as a
blocking group to double bond isomerisation, may have
of ligands 11Á12 were poor in all cases, enantioselec-
/
tivities were competitive when compared with BINAP
(optimal ee of 46%, 45% for BINAP).
Employing palladium complexes of the t-butyl-sub-
stituted ferrocenyloxazoline ligand 13 in toluene as
solvent with DIPEA as base produced cyclised product
in 25% yield and 40% ee (entry 10). Changing the base to
triethylamine resulted in an improved yield of 35% and
increased enantioselectivity of 46%. With the more
sterically demanding PMP as base, a vastly improved
yield of 63% was observed, with an ee of 47%, our
optimal result for substrate 15. Switching the solvent to
Table 1
Heck cyclisations of aryl triflate 15
a
Entry
Ligand
Solvent
Base
Time (h)
Temp. (8C)
% Yield b
% Ee (config.) c
1
2
6
6
Toluene
Benzene
Benzene
DMA
PMP
PMP
48
48
80
80
80
110
80
80
80
80
80
80
80
80
80
80
80
80
23
24
40
38
9
35 (S)
45 (S)
35 (S)
7 (S)
3
6
i-Pr2NEt
PS
PMP
48
48
4
6
5
11
11
11
11
12
13
13
13
13
13
13
14
Toluene
Benzene
DMA
168
168
168
168
168
168
168
168
168
168
168
168
31 (R)
27 (R)
25 (R)
46 (R)
12 (R)
40 (R)
46 (R)
47 (R)
28 (R)
37 (R)
35 (R)
39 (R)
6
PMP
PMP
14
30
13
14
25
35
63
22
17
8
7
8
Toluene
Toluene
Toluene
Toluene
Toluene
Benzene
Benzene
Toluene
Toluene
Et3N
PMP
9
10
11
12
13
14
15 d
16
i-Pr2NEt
Et3N
PMP
PMP
i-Pr2NEt
i-Pr2NEt
PMP
14
a
PMPꢁ
/
1,2,2,6,6-pentamethylpiperidine, PSꢁproton sponge (1,8-bis(dimethylamino)naphthalene).
/
b
c
Determined by 1H-NMR.
Absolute configuration inferred by comparing the optical rotation to literature values [42].
Pd(OAc)2 was used as the catalyst precursor.
d

550
D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á561
/
an adverse effect on the reactivity of this substrate.
There may be a steric clash between it and one of the
phosphine phenyl groups of the ligand leading to a
subsequent retardation of the migratory insertion step,
and hence the relatively low yields observed in this
study.
dichloromethane. Thus, the required novel cyclic sub-
strate 2-trifluoromethanesulfonyloxy-[4.5]-dec-7-en-8-
yl-carbonyl-N-methylaniline was prepared in five steps
in an overall yield of 37% from 1-cyclohexene-1-
carboxylic acid 25.
2.4. Intramolecular Heck studies of substrate 17
2.3. Synthesis of substrate 17
The synthesis of the racemic Heck products from the
cyclisation of substrate 17 was carried out using
palladium complexes of racemic BINAP as ligand
(Scheme 5).
A second novel substrate for the intramolecular
asymmetric Heck reaction, trifluoromethanesulfony-
loxy-[4.5]-dec-7-en-8-yl-carbonyl-N-methylaniline (17)
was prepared in order to investigate the relative regios-
electivity and enantioselectivity of cyclisation promoted
by palladium complexes of bisphosphine or phosphina-
mine ligands. We proposed to prepare this cyclic
precursor via a similar route to that described in Scheme
1 for the preparation of triflate 15. Therefore, 1-
cyclohexene-1-carboxylic acid 25 was coupled via the
acid chloride with 2-aminophenol yielding the corre-
sponding amide 26 in 60% overall yield (Scheme 4). The
remaining synthetic transformations required were
amide N-methylation and conversion of the phenol to
an aryl triflate. In order to achieve clean methylation it
was again necessary to protect the phenol 26 as an
allylether 27, which was then treated with iodomethane
yielding the tertiary amide 28 in 78% overall yield.
Deprotection of the ether as before using catalytic
quantities of tetrakis(triphenylphosphine)palladium(0)
gave the phenol 29 in 93% yield. Conversion of the
phenol to the aryl triflate 17 was realised in 85% yield
using triflic anhydride and dimethylaminopyridine in
As expected, a mixture of double bond isomers 30 and
31 was obtained. Unfortunately, despite numerous
attempts, this mixture proved inseparable by column
1
chromatography. The 270 MHz H-NMR spectrum of
this mixture shows characteristic alkene peaks for the
alkene isomer 30 at 5.12 (H2) and 6.13 (H3) and for the
alkene isomer 31 at 5.91 (H3,4). However both sets of
enantiomers could be resolved by chiral HPLC on a
Daicel Chiralpak OJ column and with this information
in hand asymmetric Heck reactions were then carried
out using palladium complexes of (R)-BINAP 6 and
phosphinamine ligands 11Á14 (Scheme 6, Table 2).
/
Using palladium complexes generated in situ from
(R)-BINAP 6 and Pd2(dba)3, cyclisation proceeded in a
good yield of 77% employing dimethylacetamide
(DMA) as solvent and PMP as base (entry 1). However,
as expected, a mixture of spirooxindole double bond
isomers 30 and 31 was obtained in a ratio of 23:77 in
which the enantiomeric excess for 30 was 74%,whereas
31 was formed with a much decreased stereoselectivity
Scheme 4. Synthesis of substrate 17.

D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á
/
561
551
Scheme 5. Preparation of racemate product 30/31.
of 31% ee. A switch of solvent to toluene caused the
regioselectivity for isomer 30/31 to be reversed, which
suggests a significant role for coordinating versus non-
coordinating solvents in the cationic pathway (entry 2).
Increased reactivity was also observed on switching to
the non-polar solvent (compare entries 1 and 2). When
employing dimethylformamide as solvent, excellent
conversions were observed although the enantioselec-
tivity was significantly lower, just 38% ee for isomer 30
and racemic for isomer 31. In addition, an alternative
inorganic base was utilised in toluene but in this case
enantioselectivity was again moderate (47% ee for
isomer 30) and just 5% for isomer 31 (entry 4). The
regioselectivity in this case was 91:9 in favour of isomer
31, which highlights the significant role of the base
(compare entries 2 and 4).
additive in the work of Overman, the results obtained
with the diphenylphosphinoaryloxazolines in the present
study were not encouraging. Therefore a variety of
alternative bases were studied with these ligands in an
attempt to improve reactivity and stereoselectivity.
Using dimethylacetamide as solvent (entry 8), palla-
dium complexes of 11 afforded cyclised product in a
yield of 20%, with a good enantioselectivity of 76% and
with an excellent regioselectivity of ꢀ
was again poor (10%) and the enantioselectivity mod-
erate (54%) albeit with excellent regioselectivity (ꢀ99:1)
/
99:1. The yield
/
using palladium complexes of ligand 12 (entry 9).
Improved yields were observed in Heck reactions
promoted by DIPEA compared to the other amine
bases screened. Enantioselectivities were moderate how-
ever, ranging from 12 to 39% with an optimal result of
39% ee in an excellent 96% yield for the t-Bu-substituted
aryloxazoline ligand 11 with benzene as solvent (entry
11).
The cyclisation of initiate 16 was also investigated
using palladium complexes of the diphenylphosphino-
ferrocenyloxazoline-containing ligands 13 and 14
(Scheme 6, Table 3). Initially PMP was used as base as
this proved to be the optimal additive in Overman’s
investigation, and also in the present study using (R)-
BINAP as ligand.
Palladium complexes of the diphenylphosphinoary-
loxazoline ligands 11Á12 were initially screened in PMP-
/
promoted asymmetric Heck cyclisations of triflate 17.
The intramolecular Heck reaction of 17 in dimethylace-
tamide using palladium complexes of the t-butyl-sub-
stituted aryloxazoline ligand 11 proceeded in poor yield
(20%), but with reasonable enantioselectivity (57%)
(entry 5). Of particular note was the almost complete
regioselectivity observed for the alkene double bond
isomer 30. Catalysts derived from palladium complexes
of the i-propyl-substituted diphenylphosphinoaryloxa-
zoline 12 proved even less reactive than their t-butyl
analogues, affording just 7% of cyclised product (entry
7). Although PMP had proven to be an excellent
Using DMA as solvent at 110 8C, palladium com-
plexes of the t-butyl-substituted diphenylphosphinofer-
rocenyloxazoline ligand 13 induced cyclisation with
almost complete regioselectivity for alkene isomer 30,
Scheme 6. Asymmetric cyclisation of substrate 17.

552
D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á561
/
Table 2
Heck cyclisations of aryl triflate 17
Entry
Ligand
Solvent
Base
Time (h)
Temp. (8C)
% Yield a
30:31 a
% Ee 30:31 (config.) b
1
2
6
6
DMA
Toluene
DMF
PMP
PMP
PMP
48
48
110
110
110
80
77
90
90
70
20
20
7
23:77
75:25
15.85
9:91
74:31 (S)
71:27 (S)
38.0 (S) c
47:5 (S)
57 (R) c
19 (R) c
53 (R) c
76 (R) c
54 (R) c
36 (R) c
39 (R) c
12 (R) c
3
6
48
48
4
6
Toluene
DMA
K2CO3
PMP
PMP
5
11
11
12
11
12
11
11
12
168
168
168
168
168
168
168
168
110
80
ꢀ99:1
/
6
Toluene
DMA
DMA
90:10
98:2
7
PMP
PS
PS
110
110
110
80
8
20
10
35
96
20
ꢀ
ꢀ/
/
99:1
99:1
92:8
98:2
94:6
9
DMA
10
11
12
Toluene
Benzene
Toluene
i-Pr2NEt
i-Pr2NEt
i-Pr2NEt
80
80
Determined by 1H-NMR.
a
b
Enantioselectivities were determined by chiral HPLC analysis using a Daicel Chiralpak OJ column (0.4625 cm), hexaneÁ
(tRꢁ41.5 (S)-30 and 57.8 (R)-30; tRꢁ28.1 (S)-31 and 31.4 (R)-31 min.
Minor isomer formed as a racemate.
/2-propanol 99.5:0.5
/
/
c
with a good enantioselectivity of 64% (entry 1). A drop
in reactivity (13% yield), regioselectivity (71:29) and
enantioselectivity (37%), was observed under similar
reaction conditions, but employing palladium catalysts
of the i-propyl-substituted ferrocenyl ligand (14, entry
2). Using 13-derived catalysts and increasing the reac-
tion temperature did not provide any significant accom-
panying increase in reactivity and was detrimental to
regioselectivity and enantioselectivity (entry 3). Under
these reaction conditions a significant deposition of
palladium black was observed. Using the t-butyl-
derived ferroceneoxazoline ligand 13 but switching the
solvent to toluene, previously found by Shibasaki to
lead to improved reactivity for initiate substrates,
improved the yield to 70% albeit with slightly poorer
regioselectivity and enantioselectivity (compare entries 1
and 4). Lowering the reaction temperature from reflux
to 80 8C led to cyclised product in excellent regioselec-
tivity (ꢀ99:1) and slightly improved enantioselectivity
/
(compare entries 4 and 5). The pattern of lower
reactivity and selectivity of the i-propyl-substituted
ferrocenyloxazoline observed when using DMA was
again repeated when toluene was used as solvent (entry
6). Using palladium complexes of 13 in refluxing
benzene (entry 7), a slight increase in enantioselectivity
was seen in comparison to the corresponding reaction in
toluene (53 vs. 57%) with a slight reduction in yield (55
vs. 71%). Using proton sponge as base afforded a 30%
yield in DMA at 110 8C with an excellent enantioselec-
tivity of 85% using palladium complexes of ligand 13
(entry 9). Changing the solvent to toluene and reducing
the temperature to 80 8C greatly improved the reactivity
Table 3
Heck cyclisations of aryl triflate 17
Entry
Ligand
Solvent
Base
Time (h)
Temp. (8C)
% Yield a
30:31 a
99:1
% Ee 30:31 (config.) b
1
2
13
14
13
13
13
14
13
14
13
13
14
13
13
14
DMA
DMA
DMA
PMP
PMP
168
168
168
168
168
168
168
168
168
168
168
168
168
168
110
110
145
110
80
36
13
34
70
71
20
55
20
30
71
21
55
61
25
ꢀ/
64 (R)
37 (R)
13 (R)
71:29
76.24
94:6
3
PMP
PMP
4
Toluene
Toluene
Toluene
Benzene
Benzene
DMA
51 (R)
5
PMP
PMP
ꢀ/99:1
53 (R) c
19 (R) c
57 (R) c
47 (R) c
85 (R) c
82 (R) c
66 (R) c
57 (R) c
47 (R) c
15 (R) c
6
110
80
90:10
98:2
76:24
7
PMP
PMP
8
80
9
Proton sponge
Proton sponge
Proton sponge
i-Pr2NEt
i-Pr2NEt
i-Pr2NEt
110
80
ꢀ
ꢀ/
/
99:1
99:1
94:6
98:2
98:2
95:5
10
11
12
13
14
Toluene
DMA
110
80
Toluene
Benzene
Benzene
80
80
Determined by 1H-NMR.
a
b
Enantioselectivities were determined by chiral HPLC analysis using a Daicel Chiralpak OJ column (0.4625 cm), hexaneÁ
(tRꢁ41.5 (S)-30 and 57.8 (R)-30; tRꢁ28.1 (S)-31 and 31.4 (R)-31 min.
Minor isomer formed as a racemate.
/2-propanol 99.5:0.5
/
/
c

D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á
/
561
553
of the catalyst giving 30 in 71% yield with a regioselec-
tivity of ꢀ99:1, and with a high enantioselectivity of
/
82% (entry 10). Palladium complexes of 13 in toluene
afforded cyclised product in good yield and good
regioselectivity (98:2), with a reasonable enantioselec-
tivity of 57% (entry 12) when employing Hunig’s base.
¨
Scheme 8. Asymmetric cyclisation of substrate 18.
Switching to benzene resulted in a slight increase in yield
but also to a decrease in enantioselectivity (entry 13).
Palladium complexes of the i-propyl-substituted ferro-
cenyloxazoline (14) again afforded cyclic product in low
silver phosphate as the optimal additive [47]. In a
subsequent publication from the same group [20] the
cyclisation of the analogous alkenyl initiates 18 was
described (Scheme 8).
yield (typically in the range 5Á
/
25%) and enantioselec-
tivity (up to 15%) (entry 14).
It was assumed that silver additives would not be
required in the case of the triflate, as the triflate anion
spontaneously dissociates, thus generating the required
cationic 16 electron palladium species. These reaction
conditions gave 91% ee as compared to the 46% ee from
the corresponding iodide. Palladium complexes of
BINAP, and its chiral arsine analogue BINAs 7 are
the only catalyst systems reported to date which have
proven successful for this system [23].
This investigation of the novel substrate 17 shows that
palladium catalysts of phosphinamine ligands provided
products of good stereo- and excellent regioselectivity in
its asymmetric Heck reaction. Interestingly, the product
distribution obtained from cyclisations promoted by
palladium catalysts of phosphinamine ligands differs
appreciably from that obtained using the corresponding
BINAP catalysts. For example, employing BINAP in
DMA as solvent predominantly double bond isomer 31
was formed, whereas ligand 12 under analogous condi-
tions yielded almost exclusively double bond isomer 30.
Thus with certain substrates such as the aryl initiate 17,
it may be possible to produce different isomers selec-
tively, depending on the exact reaction conditions
employed. It is also worth noting that palladium
complexes of BINAP proved considerably more reactive
than their corresponding phosphinamine analogues.
Conversions of up to 90% were obtained after 48 h
using BINAP-derived catalysts, whereas using the
phosphinamine systems reaction times of up to 7 days
were required to afford reasonable yields of product.
It is therefore of interest to test phosphinamine
ligands of type 11Á14 in this cyclisation and compare
/
their reactivity and stereoselectivity to BINAP (Scheme
8). The results of our investigations are detailed in Table
4.
Using palladium complexes generated in situ from
Pd2(dba)3 and (R)-BINAP 6, cyclisation proceeded in
reasonable yield (50%), and with good enantioselectivity
(82% ee, entry 1). A similar yield (54%), albeit with
higher enantioselectivity (91%), was obtained in Shiba-
saki’s study which employed Pd(OAc)2 as the precursor.
Increasing the reaction temperature to 110 8C resulted in
an improved yield (65%) without having an adverse
effect on the stereoselectivity of the reaction (entry 2).
Using N-methylpyrrolidine as solvent resulted in in-
creased activity of the catalyst (80% yield) but also
lowered enantioselectivity (46% ee, entry 3). Using
palladium complexes of the t-butyl-substituted ferroce-
nyloxazoline ligand 13 cyclisation proceeded in lower
yield (30%) but with an improved enantioselectivity of
85% (entry 4). In an attempt to increase yields, higher
reaction temperatures of 80 8C and 110 8C were applied
but this led to decreased yields (24% and 23%, respec-
tively) and also decreased enantioselectivities (67% and
2.5. Intramolecular Heck studies of substrate 18 *
/
the
formation of cis-decalins
The original report on the asymmetric intramolecular
Heck reaction from Shibasaki [19] involved the synthesis
of cis-decalins 32 from the corresponding alkenyl
iodides 33 (Scheme 7). The basic strategy involved
enantioselective ring closure of the prochiral substrate
resulting in the formation of a tertiary chiral centre.
When alkenyl iodides were used as substrates it was
necessary to add silver salts in order to achieve good
enantioselectivity. Screening of a variety of salts yielded
42% ee, respectively, entries 5Á6). In both cases the mass
/
balance was recovered unreacted starting initiate 18.
This, and the precipitation of palladium black, suggests
that higher reaction temperatures lead to catalyst
degradation. Alternative bases to potassium carbonate
were screened using palladium complexes of ligand 13
(entries 7Á
/
9) but low yields of 10Á11% were observed.
/
However, there were marked differences in the enantios-
electivities obtained as both sodium carbonate and
1,2,2,6,6-pentamethylpiperidine (PMP) gave good en-
antioselectivities (67% and 74%, respectively), whereas
Scheme 7. Formation of cis-decalin 32.

554
D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á561
/
Table 4
Heck cyclisations of alkenyl triflate 18
Entry
Ligand
Solvent
Base
T (8C)
Yield (%) a
ee b,c
1
2
6
6
Toluene
Toluene
NMP
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
Na2CO3
PMP
60
110
60
50
65
80
30
24
23
11
10
10
82
83
46
85
67
42
67
74
28
3
6
4
13
13
13
13
13
13
14
11
11
13
13
13
11
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Toluene
Benzene
DMA
60
80
5
6
110
60
7
8
60
60
9
Proton sponge
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
10
11
12
13
14
15
16
60
60
B/5
Á/
20
17
11
47
42
62
110
60
60
60
B/5
Á/
58
43
NMP
NMP
45
47
60
a
Yields were determined by 1H-NMR analysis of the reaction mixture.
b
Enantioselectivities were determined by chiral HPLC analysis using a Daicel Chiralpak OJ column (0.4625 cm), hexane: 100% (tRꢁ
and 93.7 (ꢀ) min).
The specific rotation of the major enantiomer in entries 1Á
by comparison with Shibasaki’s original work [19].
/
84.2 (ꢂ)
/
/
c
/
3 was (ꢀ
/
) and in entries 4Á
/
16 was (ꢂ) and the absolute configuration was assigned
/
proton sponge gave a poor ee of 28%. A similar trend
was observed by Shibasaki when he screened a range of
bases in the BINAP-promoted cyclisations of 33 and
found that the use of tertiary amines as base led to less
reactive systems [42]. Using palladium complexes of the
i-propyl-substituted ferrocenyloxazoline ligand 14 af-
ligand 13 afforded our optimal enantioselectivity of 85%
for the range of phosphinamine ligands tested. Variation
of solvent and base appears to indicate that this system
is very sensitive to catalyst deactivation and that a
combination of either toluene or NMP with potassium
carbonate as base was required for acceptable catalytic
activity. However, the excellent enantioselectivities ob-
tained (up to 85%) using phosphinamine ligands sug-
gests that there is potential for the application of their
palladium complexes in this reaction.
forded only trace amounts (B/5%) of cyclised product
(entry 10).
Palladium complexes of the t-butyl-substituted ary-
loxazoline ligand 11 proved less reactive and enantiose-
lective than the ferrocene analogue affording a yield of
20% and an enantioselectivity of 47% (compare entries 4
and 11). Increasing the reaction temperature to 110 8C
again led to both a decrease in yield and enantioselec-
tivity (entry 12), also attributable to catalyst inactivation
at elevated temperatures. Using palladium complexes of
ligand 13 but switching the solvent to benzene resulted
in a significant attenuation of yield to 11%, with an
enantioselectivity of 62% (entry 13). Similarly poor
yields were observed in DMA (entry 14). Reactivity
was improved using N-methylpyrrolidine (NMP) as
solvent, as outlined by a yield of 45% (with palladium
complexes of 13), with a good enantioselectivity of 58%
(entry 15). Improved activity was also seen with
catalysts derived from the t-butyl-substituted aryloxazo-
line ligand 11 using NMP as solvent, but only moderate
enantioselectivity (43%) was observed (entry 16).
3. Mechanistic considerations
For the Heck cyclisation of the aryltriflates 15 and 17
the putative intermediates prior to migratory insertion
in palladium complexes of the tert-butyl-substituted
diphenylphosphinoferrocenyloxazoline 13 can have the
alkene bound to palladium by either of its faces and can
bind in a trans-fashion to either phosphorus (Scheme 9),
or nitrogen (Scheme 10).
When the alkene approach is trans to phosphorus
both of the possible intermediates (34, 35) suffer from
steric repulsion. When the alkene approach is trans to
nitrogen as in intermediate 36, which would lead to (S)
configured product, there is steric repulsion between the
cyclohexyl ring and one of the phenyl groups on the
phosphine. Intermediate 37, which would lead to (R)
configured product is sterically unencumbered. Thus,
intermediate 37 appears to be the most likely route to
migratory insertion as this is the palladium complex of
the lowest relative energy. This route should lead to
In conclusion, BINAP and a range of phosphinamine
ligands were screened in asymmetric Heck reactions of
initiate 18. Palladium complexes formed from Pd2(dba)3
and BINAP gave cyclised product in up to 65% yield
and 85% ee. The t-butyl-substituted ferrocenyloxazoline

D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á
/
561
555
Scheme 9. Alkene approach trans to phosphorus.
cyclised product of (R) configuration with high en-
antioselectivity and this is what is seen experimentally.
reported by Pfaltz in the asymmetric intermolecular
Heck reaction.
Mixtures of double bond isomers (75:25) were ob-
tained in reasonable enantioselectivity (up to 74%) when
using BINAP as ligand for the cyclisation of initiate 17.
Palladium complexes of the diphenylphosphinooxazo-
lines afforded cyclised product with good enantioselec-
4. Conclusions
We have prepared two novel substrates for the
intramolecular asymmetric Heck reaction and have
tested a range of palladium complexes of phosphina-
mine ligands and BINAP in their asymmetric intramo-
lecular Heck reaction in addition to testing a known
substrate for cis-decalin formation. A number of inter-
esting trends appeared for the different ligand types
during the course of this study.
Cyclisations of initiate 15 proceeded in poor to
moderate yields for all of the catalyst systems screened.
The optimal enantioselectivity (47%) was afforded with
the t-butyl-substituted ferrocenyloxazoline ligand 13.
Reactions in nonpolar solvents (particularly toluene)
proceeded with better yields than the corresponding
reactions with polar solvents, a trend that was apparent
for all of the triflate substrates screened. This would
suggest that coordinating solvents can have a significant
effect in the cationic Heck pathway. Another curious
phenomenon is the higher reactivity of the t-butyl-
substituted over the i-propyl-substituted oxazoline
based ligands with all the substrates screened. The
reason for this is unclear although similar trends were
tivity (up to 85%) and excellent regioselectivity (ꢀ99:1)
/
in the cyclisation of this substrate. The t-butyl-substi-
tuted ferrocenyloxazoline ligand again proved to be the
most enantioselective of all the ligands screened.
These results suggest that palladium complexes of
phosphinamine ligands may be applied successfully in
enantioselective Heck reactions, via the cationic Heck
pathway, which is accessible from aryl or alkenyl
initiates.
5. Experimental
1H and 13C spectra were recorded at 270 (67.5) or 500
(125) MHz at ambient temperature on JEOL JNM-
PMX-270 MHz or Varian-Unity 500 MHz spectro-
meters with Me4Si as the internal standard. Peak
assignments were aided by HÁ1
1
/
H correlation experi-
ments. Coupling constants are given as absolute values.
Low resolution electron-impact MS spectra were mea-
sured on a VG Analytical spectrometer with attached
Scheme 10. Alkene approach trans to nitrogen.

556
D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á561
/
INCOS 2400 data system at an ionisation potential of 70
eV. Isomers are assumed to have the same response
factors. Elemental analyses were performed by Ms Anne
Connolly, Department of Chemistry, University College
H11a,b); 13C-NMR (75.4 MHz, CDCl3): dꢁ
166.5
/
(C(O)), 149.1 (C2), 133.4 (C1), 132.1 (C7), 126.9 (C10),
125.7 (C4), 123.4 (C5), 119.8 (C6), 116.5 (C3), 107.2 (C8),
64.5 (O(CH2)2), 36.1 (C12), 31.0 (C9), 24.3 (C10); MS (EI,
70 eV): m/z (%): 275 [Mꢀ] (28%), 167 (89), 99 (58), 81
(100); Found: C, 64.8; H, 6.2; N, 5.1. C15H17O4N
requires C, 65.4; H, 6.2; N, 5.0%.
Dublin. Infra-red spectra were recorded on a PerkinÁ
Elmer Paragon 1000 Infra-red FT spectrometer. Optical
rotation values were measured on a PerkinÁElmer 241
/
/
polarimeter. Melting points were determined in open
capillary tubes in a Gallenkamp melting point apparatus
and are uncorrected. Thin layer chromatography (TLC)
was carried out on aluminium sheets pre-coated with
silica gel 60 F 254 (0.25 mm, Macherey-Nagel). Column
chromatography separations were performed using
Merck Kieselgel 60 (Art. 7734), Merck Alumina (Art.
1097) or Merck Alumina (Art. 1104) as stated. Solvents
were dried immediately prior to use by distillation from
standard drying agents. 1,4-Dioxaspiro[4.5]dec-7-ene-8-
carboxylic acid (19) was prepared according to literature
method [48]. Methyl 1-[4-[[(trifluoromethyl)sulfonyl]-
oxy]-3(Z)-butenyl-2,5-cyclohexadiene-1-carboxylate
(18) was prepared according to the method of Shibasaki
[43].
5.2. Preparation of 2-allyloxy-(1,4-dioxaspiro[4.5]dec-
7-en-8-ylcarbonyl)-aniline (21)
Following the general method of McKillop, the amide
20 (0.85 g, 3.2 mmol), allyl bromide (0.8 ml, 8.0 mmol),
benzyltriethylammonium chloride (0.075 g, 0.32 mmol),
1.0 M aqueous sodium hydroxide (3.5 ml), CH2Cl2 (20
ml) and water (7 ml) were placed in a 100 ml Schlenk
flask. The flask was wrapped with aluminium foil to
prevent exposure to light and the orange mixture was
stirred vigorously for 14 h at r.t. The reaction was
allowed to stand for 15 min and the layers were
separated. The aqueous layer was extracted with
CH2Cl2 (220 ml) and the combined organic layers
were washed with saturated aqueous citric acid solution
(100 ml), brine (100 ml), dried over magnesium sulfate
and concentrated in vacuo. The crude residue thus
obtained was carried through to the next reaction
without further purification, IR (KBr disc cmꢂ1):
5.1. Preparation of 2-hydroxy-(1,4-dioxaspiro[4.5]dec-
7-en-8-ylcarbonyl)-aniline (20)
1,4-Dioxaspiro[4.5]dec-7-ene-8-carboxylic acid (3.4 g,
18.4 mmol) in THF (30 ml) was added dropwise to a
stirred suspension of sodium hydride (1.0 g, 25.0 mmol),
in THF (30 ml) at 0 8C yielding a milky white suspen-
sion. Stirring was continued for a further 2 h at ambient
temperature. The reaction was recooled to 0 8C and
thionyl chloride (1.6 ml, 21.0 mmol), was added drop-
wise over approximately 5 min giving a light yellow
almost transparent solution. This was stirred for a
further 2.5 h at room temperature (r.t.). 2-Aminophenol
(4.00 g, 18.4 mmol) was added as a solution in THF (30
ml), followed by triethylamine (2.7 ml, 19.0 mmol).
Evolution of HCl was apparent at this stage and the
light yellow solution became a deep mustard coloured
suspension. The reaction mixture was then refluxed for a
further 5 h. After cooling the deep yellow suspension
was diluted with ether (300 ml) and washed consecu-
tively with water (250 ml), saturated sodium bicarbonate
(250 ml) and brine (250 ml). The organic layer was
concentrated in vacuo yielding a dark brown solid which
was purified by column chromatography giving 2-
hydroxy-(1,4-dioxaspiro[4.5]dec-7-en-8-ylcarbonyl)-ani-
3423, 1673, 1634; 1H-NMR (300 MHz, CDCl3): dꢁ
8.42 (dd, 1H, Jꢁ7.3, 2.3 Hz, H6), 8.20 (br. s, 1H,
NH), 6.98 (m, 3H, H3,4,5), 6.62 (m, 1H, H8), 6.00 (m, 1H,
H2?), 5.40 (dd, 1H, Jꢁ15.8, 1.5 Hz, H3?a), 5.30 (dd, 1H,
Jꢁ9.1, 1.5 Hz, H3?b), 4.61 (brd, 2H, Jꢁ2.0 Hz, H1?a,b),
4.01 (s, 4H, (OCH2)2), 2.64 (app t, 2H, Jꢁ1.5 Hz,
12a,b), 2.48 (m, 2H, H9a,b), 1.87 (t, 1H, Jꢁ6.4 Hz,
H10a,b 165.6 (C(O)),
13C-NMR (75 MHz, CDCl3): dꢁ
147.3 (C2), 134.1 (C1), 133.1 (Ar-CH), 131.4 (Ar-CH),
128.3 (C10), 123.8 (C7), 121.7 (Ar-CH), 120.2 (Ar-CH),
118.3 (C3?), 111.6 (C2?), 107.5 (C8), 69.8 (C1?), 64.81
(2CH2), 36.2 (C12), 31.0 (C9), 23.9 (C11); MS (EI, 70 eV):
m/z (%): 315 [Mꢀ] (26%), 229 (17), 188 (18), 167 (70),
135 (37), 99 (48), 81 (100); HRMS for C18H21O4N:
Requires 351.3690; found 351.3715.
/
/
/
/
/
/
H
/
)
/
5.3. Preparation of 2-allyloxy-(1,4-dioxaspiro[4.5]dec-
7-en-8-ylcarbonyl)-N-methylaniline (22)
A mixture of the allyl ether 21 (0.89 g, 2.7 mmol),
sodium hydride (60% paraffin dispersion, 200 mg, 4.9
mmol) and dry THF (20 ml) were stirred for 10 min at
0 8C under a nitrogen atmosphere and then for a further
3 h at r.t. Iodomethane (0.5 ml, 8.0 mmol) was then
added in one portion and the mixture was stirred at r.t.
overnight. Saturated sodium bicarbonate solution (50
ml) was then added followed by 20 ml of CH2Cl2. The
layers were separated and the aqueous layer was
extracted with ether (220 ml) and EtOAc (20 ml). The
line, as a white solid (3.0 g, 60%), m.p. 158Á
/
160 8C, TLC
(100% ethyl acetate) Rfꢁ
/
0.3; IR (KBr disc cmꢂ1): 3410,
1
2961, 1685; H-NMR (300 MHz, CDCl3): dꢁ
1H, OH), 7.83 (s, 1H, NH), 7.17 (dd, 1H, Jꢁ
Hz, H6), 7.08 (td, 1H, Jꢁ
1H, Jꢁ6.7, 1.4 Hz, H4), 6.85 (td, 1H, Jꢁ
H3), 6.71 (m, 1H, H8), 4.01 (s, 4H, (OCH2)2), 2.62 (m,
2H, H12a,b), 2.48 (m, 2H, H9a,b), 1.87 (t, 1H, Jꢁ6.4 Hz,
/9.75 (s,
/
7.9, 1.5
/0.9 Hz, 6.5 Hz, H5), 6.98 (dd,
/
/7.9, 1.5 Hz,
/

D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á
/
561
557
combined organic layers were washed with brine (50 ml),
dried over magnesium sulfate and concentrated in
vacuo. The resultant brown residue was purified by
column chromatography giving 2-allyloxy-(1,4-dioxas-
C, 66.7; H, 6.6; N, 4.5. C16H19O4N requires C, 66.4; H,
6.6; N, 4.8%.
5.5. Preparation of 2-trifluoromethanesulfonyloxy-(1,4-
dioxaspiro[4.5]dec-7-en-8-ylcarbonyl)-N-methylaniline
(15)
piro[4.5]dec-7-en-8-ylcarbonyl)-N-methylaniline as
light yellow oil (0.89 g, 96%), TLC: (petroleum ether
40Á60 8CÁEtOAcꢁ2:1) Rfꢁ
0.05; IR (neat, cmꢂ1):
2956, 1643; H-NMR (300 MHz, CDCl3): dꢁ7.20 (dt,
1H, Jꢁ7.0, 1.7 Hz, H5), 7.10 (brd, 1H, Jꢁ7.6 Hz, H6),
6.90 (m, 2H, Ar-H), 6.02 (m, 1H, H2?), 5.64 (m, 1H, H8),
5.43 (dd, 1H, Jꢁ18.1, 1.5 Hz, H3?a), 5.30 (dd, 1H, Jꢁ
a
/
/
/
/
1
/
2-Hydroxy-(1,4-dioxaspiro[4.5]dec-7-en-8-ylcarbo-
nyl)-N-methylaniline (0.501 g, 2.00 mmol), dimethyla-
minopyridine (1.83 g, 15.0 mmol) and CH2Cl2 (5 ml)
were added to a 10 ml Schlenk flask under a nitrogen
atmosphere giving a light orange solution. The solution
was cooled to 0 8C and triflic anhydride (0.85 ml, 5.0
mmol) was added dropwise yielding a green mixture
with some precipitation. The reaction was allowed to stir
at r.t. for a further 14 h, recooled to 0 8C and quenched
with a saturated citric acid solution (10 ml). The
contents of the Schlenk flask were transferred to a 100
ml separatory funnel. The organic layer was removed
and the aqueous layer was extracted with 20 ml of
CH2Cl2. The organic fractions were combined, dried
over magnesium sulfate and concentrated in vacuo. The
resulting light orange almost solid residue was purified
by column chromatography yielding 2-trifluorometha-
nesulfonyloxy-[4.5]-dec-7-en-8-yl-carbonyl-N-methyla-
niline (0.67 g, 82%) as a clear oil, TLC: (100% ethyl
/
/
/
/
10.5, 1.5 Hz, H3?b), 4.54 (m, 2H, H1?), 3.85 (s, 4H,
(OCH2)2), 3.24 (s, 3H), 2.28 (m, 2H, H12a,b), 2.04 (m,
2H, H9a,b), 1.56 (t, 2H, Jꢁ
/
6.3 Hz, H11a,b); 13C-NMR
(75 MHz, CDCl3): dꢁ172.3 (C(O)), 153.5 (C2), 134.0
/
(C1), 133.0 (CH), 128.9 (C10), 128.5 (CH), 128.2 (CH),
121.1 (C3?), 117.7 (C2?), 113.0 (C7), 107.3 (C8), 69.0
(CH), 64.4 (2CH2), 35.6 (C12), 30.9 (C9), 25.2 (C11); MS
(EI, 70 eV): m/z (%): 329 [Mꢀ] (27%), 202 (23), 186 (36),
167 (49), 122 (57), 99 (77), 81(96), 53 (100); Found C,
68.7; H, 6.91; N, 4.15. C19H23O4N requires C, 69.3; H,
7.0; N, 4.3%.
5.4. Preparation of 2-hydroxy-(1,4-dioxaspiro[4.5]dec-
7-en-8-ylcarbonyl)-N-methylaniline (23)
acetate) Rfꢁ
/
0.7; IR (neat, cmꢂ1): 2930, 1660, 1425,
Following a modification of the procedure of Deziel,
a mixture of 2-allyloxy-(1,4-dioxaspiro[4.5]-dec-7-en-8-
yl-carbonyl)-N-methylaniline (1.03 g, 3.4 mmol),
Pd(PPh3)4 (0.14 g, 0.12 mmol), triphenylphosphine
(0.04 g, 0.16 mmol) and dry MeCN (15 ml) were stirred
for 5 min at 0 8C. A solution of pyrrolidine (1.2 ml, 14.0
mmol), in MeCN (10 ml) was then added and the
solution was stirred for a further 10 min at 0 8C. The
solution was heated at 40 8C for 2 h. After cooling to r.t.
the reaction mixture was diluted with saturated citric
acid solution (20 ml) and the layers were separated. The
organic layer was washed with saturated citric acid
solution (20 ml), brine (20 ml), dried over magnesium
sulfate and concentrated in vacuo. The resulting crude
orange residue was purified by column chromatography
to give 2-hydroxy-(1,4-dioxaspiro[4.5]dec-7-en-8-ylcar-
bonyl)-N-methylaniline as a white solid (0.82 g, 84%),
1214, 1141; 1H-NMR (300 MHz, CDCl3): dꢁ
/
7.30Á7.43
/
(m, 4H, Ar-H), 5.61 (m, 1H, H8), 3.90 (s, 4H, (OCH2)2),
3.35 (s, 3H, NMe), 2.42 (m, 2H, H12a,b), 2.13 (m, 2H,
H
9a,b), 1.67 (app. t, 2H, Jꢁ
/
6.3 Hz, H11a,b); 13C-NMR
(75 MHz, CDCl3): dꢁ172.6 (C(O)), 141.3 (C2), 133.4
/
(C1), 129.4 (C7), 129.3 (C3), 128.8 (C10), 122.7 (C5),
121.4 (C6), 116.6 (C3), 107.2 (C8), 64.5 (2CH2), 35.8
(C12), 30.9 (C9), 25.2 (C11); MS (EI, 70 eV): m/z (%): 421
[Mꢀ] (9%), 288 (32), 167 (44), 99 (78), 86 (100); Found:
C, 48.0; H, 5.2; N, 2.9. C17H18O6NSF3 requires C, 48.4;
H, 4.3; N, 3.3%.
5.6. Synthesis of 1?,2?-dihydro-1?-methyl-2?-oxospiro[1,4-
dioxaspiro[4.5]dec-6-ene-8,3?-3?H-indole (24)
Pd(OAc)2 (10.5 mg, 0.048 mmol), (ꢀ
/
/ꢂ)-B1NAP (40
/
mg, 0.064 mmol) and dimethylacetamide (1 ml) were
added to a 10 ml Schlenk flask and stirred for 1 h. A
dimethylacetamide (1 ml) solution of the triflate (130
mg, 0.24 mmol) was added along with proton sponge
(0.2 g, 1.0 mmol) and the reaction was allowed to stir for
m.p. 125Á
EtOAcꢁ1:1) Rfꢁ
1612, 1459; H-NMR (300 MHz, CDCl3): dꢁ
1H, m), 7.15 (dt, 1H, Jꢁ7.0, 1.5 Hz, H6), 7.00 (m, 2H,
Ar-H), 6.85 (app. t, 1H, Jꢁ7.3 Hz, H3), 5.81 (m, 1H,
H8) 3.88 (s, 4H, (OCH2)2), 3.29 (s, 3H), 2.33 (m, 2H,
12a,b), 2.18 (m, 2H, H9a,b), 1.61 (m, 2H, H11a,b); 13C-
NMR (75 MHz, CDCl3): dꢁ173.3 (C(O)), 152.2 (C2),
/
127 8C, TLC: (petroleum ether 40Á
0.2; IR (neat, cmꢂ1): 2927, 2362,
7.52 (s,
/
60 8CÁ
/
/
/
1
/
/
/
5 min at r.t. The reaction mixture was degassed (ꢃ
/3) by
the pumpÁfreezeÁthaw method. The Schlenk flask was
/
/
H
sealed and heated at 110 8C for 48 h. The deep red
solution was quenched via addition of saturated sodium
bicarbonate solution (5 ml). The contents of the flask
were transferred to a separatory funnel, water (10 ml)
and EtOAc (10 ml) were added and the layers were
separated. The aqueous layer was extracted with EtOAc
/
133.7 (C1), 132.3 (C7), 131.9 (Ar-CH), 129.1 (Ar-CH),
129.0 (Ar-CH), 128.8 (C10), 120.6 (Ar-CH), 107.3 (C8),
64.5 (2CH2), 35.7 (C12), 30.9 (C9), 25.3 (C11); MS (EI, 70
eV): m/z (%): 289 [Mꢀ] (18%), 167 (62), 81 (100); Found

558
D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á561
/
(210 ml). The organic extracts were combined, washed
with brine, dried over magnesium sulfate and concen-
trated in vacuo. The crude product thus obtained was
further purified by preparative chromatography, TLC:
5.9. Preparation of 2-hydroxy-[4.5]-dec-7-en-8-yl-
carbonyl-aniline (26)
1-Cyclohexene-1-carboxylic acid (5.0 g, 40.0 mmol) in
THF (30 ml) was added dropwise to a stirred suspension
of sodium hydride (1.76 g, 44.0 mmol), in THF (50 ml)
at 0 8C yielding a milky white suspension Stirring was
continued for a further 2 h at ambient temperature. The
reaction was recooled to 0 8C and thionyl chloride (3.2
ml, 42.0 mmol), was added dropwise over approximately
5 min giving a light yellow almost transparent suspen-
sion. This was stirred for a further 2.5 h at r.t. 2-
Aminophenol (4.36 g, 40.0 mmol), was added as a
solution in THF (30 ml), followed by Et3N (6.4 ml, 43.0
mmol). Evolution of HCL was apparent at this stage
and the light yellow solution became a deep mustard-
coloured suspension. The reaction mixture was refluxed
for a further 5 h. After cooling the reaction mixture was
diluted with ether (300 ml) and washed consecutively
with water (250 ml), saturated sodium bicarbonate (250
ml) and brine (250 ml). The organic layer was concen-
trated in vacuo yielding a dark brown solid. This was
washed consecutively with petroleum spirits (50 ml), and
then ice-cold EtOAc (20 ml), followed by recrystallisa-
(petroleum ether 40Á
(KBr disc cmꢂ1): 1719, 1613; 1H-NMR (300 MHz,
CDCl3): dꢁ7.29 (td, 1H, Jꢁ7.7, 1.2 Hz, Ar-H), 7.25
(d, 1H, 1ꢁ7.6 Hz, Ar-H), 7.05 (td, 1H, Jꢁ7.2, 0.9 Hz,
Ar-H), 6.84 (d, 1H, Jꢁ7.9 Hz, Ar-H), 5.96 (d, 1H, Jꢁ
9.9 Hz, H9), 5.48 (d, 1H, Jꢁ9.9 Hz, H8), 3.98Á4.06 (m,
4H), 3.21 (s, 3H), 2.45 (ddd, 1H, Jꢁ2.6, 9.5, 13.1 Hz,
2.26 (m, 1H, H12a), 1.96Á2.06 (m, 2H,
11a,b); 13C-NMR (75.4 MHz, CDCl3): dꢁ
178.0
/
60 8C/EtOAcꢁ
/
3:1) Rfꢁ/0.35; IR
/
/
/
/
/
/
/
/
/
H
H
12b), 2.16Á
/
/
/
(C(O)), 143.1 (C1), 132.8 (CH), 131.1 (C10), 129.9
(CH), 128.3 (CH), 123.8 (CH), 122.6 (CH), 108.0 (C6),
104.3 (CH), 64.7 (OCH2), 64.6 (OCH2), 49.2 (C7), 30.6
(NMe), 29.7 (C12), 26.4 (C11); Found: C, 70.6; H, 6.3; N,
5.1. Calc. for C16H17NO3: C, 70.8; H, 6.3; N, 5.1%.
5.7. General procedure for the asymmetric cyclisation of
triflate 15
Pd2(dba)3 (1.8 mg, 0.002 mmol), ligand (0.008 mmol)
and benzene (0.4 ml) were charged to a 10 ml Schlenk
flask and stirred for 1 h. A benzene (0.5 ml) solution of
the triflate (30 mg, 0.08 mmol) and PMP (70 ml, 0.41
mmol) was added and the reaction was stirred for 5 min.
The reaction mixture was degassed, sealed and heated
for the appropriate reaction time at either 80 8C or
110 8C. Work-up was equivalent to the racemic variant
and for the optimal result with (R)-BINAP, 1?,2?-
dihydro-1?-methyl-2?-oxospiro[1,4-dioxaspiro[4.5]dec-6-
ene-8,3?-3?H-indole was isolated in 24% yield (5.2 mg)
tion from MeOHÁ
7-en-8-yl-carbonyl-aniline (3.4 g, 39.17%) as an off-
white solid, m.p. 156Á
158 8C, IR (KBr cmꢂ1) 3409,
3100, 2957, 1661; H-NMR (300 MHz, Me2SO): dꢁ
9.77 (s, 1H, OH), 8.76 (s, 1H, NH), 7.80 (dd, 1H, Jꢁ
6.57, 1.54 Hz, H6), 7.00 (dt, 1H, Jꢁ1.5 Hz, 5.6 Hz, H5),
6.95Á6.77 (m, 2H, H4, H3), 6.74 (m, 1H, H8), 2.16Á2.32
(m, 4H, H12a,b, H9a,b), 1.61 (m, 4H, H11a,b, H10a,b); 13C-
NMR (75.4 MHz Me2SO): dꢁ167.0 (C(O)), 148.7 (C2),
134.6 (Ar-CH), 133.9 (C1), 127.1 (C7), 125.4 (Ar-CH),
122.9 (Ar-CH), 119.8 (C8), 116.4 (C3), 25.6 (C9), 24.6
(C12), 22.4 (C11), 21.8 (C10); MS (EI, 70 eV): m/z (%):
217 [Mꢀ] (13%), 109 (100), 81(56); Found: C, 71.7; H,
6.9; N, 6.4. C13H15O2N requires C, 71.9; H, 6.9; N,
6.4%.
/
hexanes yielding 2-hydroxy-[4.5]-dec-
/
1
/
/
/
/
/
/
and 45% ee (S) with an optical rotation of [a]1D8ꢁ
(cꢁ0.25 MeOH) and otherwise equivalent in all re-
spects to a previously prepared sample.
/
ꢀ2.48
/
/
5.8. Determination of enantiomeric excess of 1?,2?-
dihydro-1?-methyl-2?-oxospiro[1,4-dioxaspiro[4.5]dec-6-
ene-8,3?-3?H-indole (24)
5.10. Preparation of 2-allyloxy-[4.5]-dec-7-en-8-yl-
carbonyl-N-methylaniline (27)
To a well stirred solution of the 2-hydroxy-[4.5]-dec-7-
en-8-yl-carbonyl-aniline (2.4 g, 10.1 mmol) in CH2Cl2
(60 ml) in a 250 ml Schlenk flask were added water (35
ml), 1 M NaOH (15 ml) and benzyltriethylammonium
chloride (227 mg, 1.0 mmol). The flask was then
wrapped in aluminium foil. Allyl bromide (1.9 ml, 20.2
mmol) was added and the reaction mixture was stirred
vigorously for a further 16 h. The reaction mixture was
then transferred to a 250 ml separatory funnel and the
layers were separated. The aqueous layer was further
extracted with CH2Cl2 (250 ml). The organic extracts
were combined and then washed consecutively with
saturated sodium bicarbonate (50 ml) and brine (50 ml),
A 0.1 M solution of Eu(hfc)3 was prepared by
dissolving 0.126 g of the reagent in 1.0 ml of CDCl3.
An NMR sample of 24 was prepared using 0.005 g of
sample in 1.0 ml of CDCl3. Approximately 30 ml of the
Eu(hfc)3 solution was added to the NMR tube and the
N-methyl singlet at 3.21 ppm split into two singlets
(3.6Á3.8 ppm) which could then be integrated. If
/
complete baseline separation is not apparent at this
stage the Eu(hfc)3 solution can be added dropwise, as
required, although addition of an excess of shift reagent
caused the N-methyl singlets to overlap with signals of
the ethylene ketal.

D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á
/
561
559
concentrated in vacuo and then dried over magnesium
sulfate. The crude 2-allyloxy-[4.5]-dec-7-en-8-yl-carbon-
yl-aniline (2.65 g, 97%) thus obtained was carried
through to the next reaction without further purifica-
tion. A solution of the allyl-amide (2.65 g, 9.8 mmol) in
THF (10 ml) was added dropwise to a suspension of
sodium hydride (0.82 g, 20.4 mmol) in THF (30 ml) at
0 8C under a nitrogen atmosphere. The resultant light
yellow mixture was stirred for a further 10 min at 0 8C
and then 3 h at r.t. Iodomethane (1.6 ml, 25.5 mmol)
was added in one portion and the reaction was allowed
to stir at r.t. for 14 h. Saturated sodium bicarbonate (25
ml) was added and the resultant mixture was extracted
with ether (230 ml) and EtOAc (30 ml). The organic
extracts were combined, washed with brine (50 ml),
dried with magnesium sulfate and concentrated in
vacuo. The crude product was purified by column
chromatography on silica gel yielding 2-allyloxy-[4.5]-
dec-7-en-8-yl-carbonyl-N-methylaniline (1.73 g, 70%) as
(ddd, 1H, Jꢁ
/
5.7, 1.6, 0.9 Hz, H6), 7.01 (dd, 2H, Jꢁ
/
6.8,
1.3 Hz, H4,5), 6.84 (m, 1H, H3), 5.93 (m, 1H, H8), 3.26 (s,
3H, NMe), 2.04 (m, 2H, H12a,b), 1.89 (m, 2H, H9a,b),
1.44 (m, 4H, H10a,b, H11a,b); 13C-NMR (75.4 MHz,
CDCl3): dꢁ174.6 (C(O)), 152.8 (C2), 134.3 (C1), 132.0,
/
131.4, 129.1 (C7), 120.7, 25.9 (C12), 25.0 (C9), 22.3 (C11),
21.7 (C10); MS (EI, 70 eV): m/z (%): 231 (Mꢀ, 7%) 123
(66), 109 (90), 81 (100); Found: C, 72.4; H, 7.4; N, 5.9.
C14H17O2N requires C, 72.7; H, 7.4; N, 6.0%.
5.12. Preparation of 2-trifluoromethanesulfonyloxy-
[4.5]-dec-7-en-8-yl-carbonyl-N-methylaniline (17)
2-Hydroxy-[4.5]-dec-7-en-8-yl-carbonyl-N-methylani-
line (0.1 g, 0.4 mmol), dimethylaminopyridine (0.36 g,
3.0 mmol) and CH2Cl2 (5 ml) were added to a 10 ml
Schlenk flask under a nitrogen atmosphere giving a light
orange solution. The solution was cooled to 0 8C and
triflic anhydride (0.17 ml, 1.0 mmol) was added drop-
wise yielding a green mixture with some precipitation.
The reaction was allowed to stir at r.t. for a further hour
then recooled to 0 8C and quenched with a saturated
citric acid solution (3 ml). The contents of the Schlenk
flask were transferred to a 100 ml separatory funnel. The
organic layer was removed and the aqueous layer was
extracted with a further 20 ml of CH2Cl2. The organic
fractions were combined, dried over magnesium sulfate
and concentrated in vacuo. The resulting brown residue
was purified by column chromatography yielding 2-
trifluoromethanesulfonyloxy-[4.5]-dec-7-en-8-yl-carbon-
yl-N-methylaniline (0.12 g, 83%) as a clear oil, TLC
a light yellow oil, TLC (petroleum ether 40:60 8CÁ
EtOAcꢁ1:1) Rfꢁ
0.25; IR (neat, cmꢂ1): 2950, 1645;
1H-NMR (300 MHz, CDCl3): dꢁ
7.20 (td, 1H, Jꢁ1.3,
1.7, 7.9 Hz, H6), 7.07 (dd, 1H, Jꢁ7.5, 1.6 Hz, H5),
6.94Á6.86 (m, 2H, H4,3), 6.02 (m, 1H, H8), 5.76 (m, 1H),
5.40 (ddd, 1H, Jꢁ17.4, 1.7, 1.5 Hz), 5.28 (ddd, 1H, Jꢁ
10.4, 1.5, 1.5 Hz), 4.56 (app. d, 2H, Jꢁ4.1 Hz), 3.22 (s,
/
/
/
/
/
/
/
/
/
/
3H, NMe), 2.04 (m, 2H, H12a,b), 1.81 (m, 2H, H9a,b),
1.40 (m, 4H, H10a,b, H11a,b); 13H-NMR (75 MHz,
CDCl3) dꢁ172.2 (C(O)), 154.6 (C2), 134.8 (C1), 133.0,
/
130.8, 129.3, 128.5 (C7), 121.0, 117.7 (C3), 112.9, 69.0
(allyl-CH2), 35.8 (C12), 25.0 (C9), 22.3 (C11), 21.8 (C10);
Found: C, 74.7; H, 7.7; N, 5.1. C17H21O2N requires C,
75.2; H, 7.8; N, 5.2%.
(EtOAc) Rfꢁ
1213, 1139; 1H-NMR (300 MHz, CDCl3): dꢁ
(m, 4H, Ar-H), 5.72 (m, 1H, H8), 3.34 (s, 3H, NMe),
2.16 (m, 2H, H12a,b), 1.87 (m, 2H, H9a,b), 1.45Á1.64 (m,
4H, H10a,b, H11a,b
13C-NMR (75.4 MHz, CDCl3) dꢁ
/
0.7; IR (neat, cmꢂ1): 2932, 1661, 1422,
7.29Á7.43
/
/
5.11. Preparation of 2-hydroxy-[4.5]-dec-7-en-8-yl-
/
carbonyl-N-methylaniline (28)
)
/
172.0 (C(O)), 146.3 (C2), 133.8 (C1), 129.4 (CH), 129.3
(CH), 128.7 (C7), 122.6 (C8), 117.3 (C3), 98.3, 87.9, 25.5
(C12), 25.1 (C9), 22.2 (C11), 21.6 (C10); MS (EI, 70 eV):
m/z (%): 363 (Mꢀ, 28%), 214 (51), 109 (78), 81 (100);
Found: C, 49.5; H, 4.5; N, 3.7. C15H16O4NSF3 requires
C, 49.6; H, 4.4; N, 3.8%.
Following a modification of the procedure of Deziel,
a mixture of 2-allyloxy-[4.5]-dec-7-en-8-yl-carbonyl-N-
methylaniline (2.67 g, 9.8 mmol), Pd(PPh3)4 (0.3 g, 3.0
mmol), triphenylphosphine (0.09 g, 0.3 mmol) and dry
MeCN (15 ml) were stirred for 5 min at 0 8C. A solution
of pyrrolidine (3.3 ml, 40.0 mmol), in MeCN (10 ml) was
then added and the solution was stirred for a further 10
min at 0 8C. The solution was heated at 4 8C for 2 h.
After cooling to r.t. the reaction mixture was diluted
with saturated citric acid solution (20 ml) and the layers
were separated. The organic layer was washed with
saturated citric acid solution (20 ml), brine (20 ml), dried
with magnesium sulfate and concentrated in vacuo. The
resulting crude orange residue was purified by column
chromatography to give 2-hydroxy-[4.5]-dec-7-en-8-yl-
carbonyl-N-methylaniline as a white solid (2.1 g, 93%),
5.13. Preparation of 1?,2?-dihydro-1?-methyl-2?-oxospiro-
(cyclohex-2-ene-1,3?-3?-indoles) (30 and 31)
An oven-dried 10 ml Schlenk flask was charged with
Pd2(dba)3 (11.5 mg, 0.013 mmol), racemic BINAP (16.8
mg, 0.027 mmol) and purged with nitrogen for 5 min.
Dimethylacetamide (1 ml) was added giving a purple
suspension which was allowed to stir for approximately
2 h until an orange solution had formed. A dimethyla-
cetamide (1 ml) solution of the triflate 17 (100 mg, 0.275
mmol) was added followed by PMP (0.17 ml, 1.1 mmol).
The resulting solution was frozen with liquid nitrogen,
m.p. 164Á
Rfꢁ
0.17; IR (KBr disc, cmꢂ1): 3121, 2931, 1656, 1614;
1H-NMR (300 MHz, CDCl3): dꢁ
7.51 (s, 1H), 7.14
/
166 8C, TLC (2:1 pet spirits 40:60ÁEtOAc)
/
/
/
evacuated (0.1Á0.01 mm) and allowed to warm to r.t.
/

560
D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á561
/
This freezeÁ
/
pumpÁ
/
thaw cycle was repeated three times.
(R) solely as the double bond isomer 30, 1H-NMR (300
MHz, CDCl3): dꢁ7.30Á7.20 (m, 2H, H5?,6?), 7.0 (m, 1H,
H4?), 6.83 (d, 1H, 17.6 Hz, H3?), 6.13 (dt, 1H, Jꢁ9.8, 4.0
Hz, H2), 5.28 (d, 1H, Jꢁ9.9 Hz, H3), 3.21 (s, 3H, NMe),
2.25Á2.20 (m, 2H, H4a,b), 2.05Á2.00 (m, 2H, H6a,b),
1.65Á 2.48 (cꢁ0.25
1.95 (m, 2H, H5a,b); [a]1D8ꢁ
MeOH).
The Schlenk flask was sealed and heated to 110 8C for
36 h. The reaction was quenched with saturated aqueous
sodium bicarbonate (5 ml), diluted with water (5 ml) and
EtOAc (5 ml). The layers were separated and the
aqueous layer was extracted with EtOAc (25 ml). The
combined organic extracts were washed with brine (20
ml), dried over magnesium sulfate and concentrated in
vacuo. The crude product was further purified by
column chromatography to give 1?,2?-dihydro-1?-meth-
yl-2?-oxospiro-(cyclohex-2-ene-1,3?-3?-indole), (0.045 g,
78%), as a white solid (mixture (23:72) of 30 and 31
/
/
/
/
/
/
/
/
ꢂ
/
/
5.15. Preparation of methyl 3,8a-dihydro-4H-
naphthalene-4a-carboxylate (32)
An oven-dried 10 ml Schlenk flask was flushed with
nitrogen for 5 min and then charged with racemic
BINAP (9.2 mg, 0.015 mmol), Pd2(dba)3 (6.7 mg,
0.0075 mmol) and toluene (0.5 ml). The purple suspen-
sion thus obtained was stirred for 1 h under a nitrogen
atmosphere and had then become an orange solution.
Methyl 1-[4-[[(trifluoromethyl)sulfonyl]oxy]-3(Z)-buten-
yl-2,5-cyclohexadiene-1-carboxylate (50 mg, 0.15 mmol)
was then added in toluene (1.0 ml) followed by
potassium carbonate (41.4 mg, 0.3 mmol). The reaction
double bond isomers), TLC (petroleum ether 40:60Á
ethyl acetateꢁ5:1) Rfꢁ
0.31; IR (KBr disc, cmꢂ1):
2961, 1705, 1610, 1465; 31: 1H-NMR (300 MHz,
CDCl3): dꢁ7.31 (t, 1H, Jꢁ7.3 Hz, H3) 7.27Á7.25 (m,
1H, H6), 7.04 (t, 1H, Jꢁ7.5 Hz, H5), 6.86 (d, 1H, Jꢁ
7.7 Hz, H4), 5.84Á5.95 (m, 2H, H9,10), 3.23 (s, 3H,
NMe), 2.61Á2.70 (m, 1H, H12a), 2.30Á2.36 (m, 2H, H7),
1.91Á2.00 (m, 2H, H11), 1.49Á
1.58 (m, 1H, H12b); 13C-
NMR (125 MHz, CDCl3) dꢁ175.9 (C(O)), 135.0 (C1),
/
/
/
/
/
/
/
/
/
/
/
/
/
/
127.9, 127.0, 125.1 (C8), 124.2 (C9), 122.6, 108.1, 101.4,
46.3 (C7), 31.9 (C10), 29.2 (C12), 22.1 (C11); MS (EI, 70
eV): m/z (EIMS 70 eV): 213 [Mꢀ] (53%), 184 (31%), 159
(100%); Found: C, 78.3; H, 7.2; N, 6.0. Calc. for
C14H15NO: C, 78.8; H, 7.1; N, 6.6%.
mixture was degassed by the freezeÁ
/
thawÁ
/
pump
method (ꢃ3), sealed and heated at 60 8C for 72 h.
/
The reaction mixture was now allowed to cool to r.t.,
diluted with ether (5 ml), washed with brine (5 ml), dried
over magnesium sulfate and concentrated in vacuo. The
crude residue was then further purified by column
chromatography on silica gel to give methyl 3,8a-
dihydro-4a-4H-naphthalenecarboxylate as a colourless
5.14. General procedure for the asymmetric cyclisation of
triflate 17
oil (14.0 mg, 51%), TLC (pet spirits 40:60Á
acetateꢁ10:1) Rfꢁ
0.2; IR (neat, cmꢂ1): 1730; 1H-
NMR (300 MHz, CDCl3): dꢁ5.94 (ddd, 1H, Jꢁ9.5,
5.1, 0.7 Hz, H5), 5.79 (m, 1H, H4), 5.60Á5.75 (m, 3H,
H4?, H3, H6), 5.56 (d, 1H, 9.7 Hz, H3?), 3.72 (s, 3H,
OMe), 3.61 (m, 1H, H2), 1.93Á2.10 (m, 2H, H2?a,b),
1.85Á
1.90 (m, 2H, H1?a,b); 13C-NMR (75.4 MHz,
CDCl3): dꢁ176.1 (C(O)), 129.7 (CH), 128.9 (CH),
126.7 (CH), 125.7 (CH), 124.0 (CH), 120.6 (CH), 52.3
(OMe), 46.0 (C1), 36.1 (C2), 27.4 (C2), 21.7 (C2?); MS
(EI, 70 eV): m/z (%): 190 [Mꢀ] (35%), 131 (100%), 115
(23%).
/
ethyl
5.14.1. Using BINAP 6 as ligand
/
/
Pd2(dba)3 (1.8 mg, 0.002 mmol), (R)-BINAP 6 (5.1
mg, 0.008 mmol) and toluene (0.4 ml) were charged to a
10 ml Schlenk flask and stirred for 1 h. A toluene (0.5
ml) solution of the triflate (30 mg, 0.08 mmol) and PMP
(70 ml, 0.41 mmol) was added and the reaction was
stirred for 5 min. The reaction mixture was degassed,
sealed and heated for 48 h at 110 8C. Work-up was
equivalent to the racemic variant and 1?,2?-dihydro-1?-
methyl-2?-oxospiro-(cyclohex-2-ene-1,3?-3?-indole) was
isolated in 90% yield (16.8 mg) as a white solid with
an enantioselectivity of 71% (S) and an optical rotation
/
/
/
/
/
/
of [a]1D8ꢁ
/
ꢀ
/
2.38 (cꢁ/0.25 MeOH) and otherwise equiva-
5.16. General procedure for the asymmetric cyclisation of
triflate 18
lent in all respects to a previously prepared sample.
5.14.2. Using ligand 13
An oven-dried 10 ml Schlenk flask was flushed with
nitrogen for 5 min and then charged with (R)-BINAP
(4.6 mg, 0.008 mmol), Pd2(dba)3 (1.8 mg, 0.002 mmol)
and toluene (0.5 ml). The purple suspension thus
obtained was stirred for 1 h under a nitrogen atmo-
sphere and had then become an orange solution. Methyl
1-[4-[[(trifluoromethyl)sulfonyl]oxy]-3(Z)-butenyl-2,5-
cyclohexadiene-1-carboxylate (26 mg, 0.08 mmol) was
then added in toluene (0.5 ml) followed by potassium
carbonate (20.4 mg, 0.15 mmol). The reaction mixture
Pd2(dba)3 (1.8 mg, 0.002 mmol), 13 (4.0 mg, 0.008
mmol) and toluene (0.4 ml) were charged to a 10 ml
Schlenk flask and stirred for 1 h. A toluene (0.5 ml)
solution of the triflate 17 (27 mg, 0.08 mmol) and proton
sponge (88.0 mg, 0.4 mmol) was added and the reaction
was stirred for 5 min. The reaction mixture was
degassed, sealed and heated for 48 h at 110 8C. Work-
up was equivalent to the racemic variant and 1?,2?-
dihydro-1?-methyl-2?-oxospiro-(cyclohex-2-ene-1,3?-3?-
indole) was isolated in 71% yield (13.2 mg) and 82% ee
was degassed by the freezeÁ
/
thawÁ
/
pump method (ꢃ3),
/

D. Kiely, P.J. Guiry / Journal of Organometallic Chemistry 687 (2003) 545Á
/
561
561
[11] Y. Donde, L.E. Overman, in: I. Ojima (Ed.), Catalytic Asym-
metric Synthesis (Chapter 8g), Wiley, New York, 2000 (Chapter
8g).
sealed and heated at 60 8C for 72 h. Work-up was as for
the racemic variant and methyl 3,8a-dihydro-4H-naph-
thalene-4a-carboxylate 32 was isolated as a viscous
[12] M. Shibasaki, C.D.J. Boden, A. Kojima, Tetrahedron 53 (1997)
7371.
colourless oil (7.4 mg, 50%), [a]1D8ꢁ
MeOH) (82% ee).
/
ꢀ
/
39.78 (cꢁ0.3
/
[13] P.J. Guiry, A.J. Hennessy, J.P. Cahill, Topics Catal. 4 (1997) 311.
[14] S.E. Gibson, R.J. Middleton, Contemp. Org. Synth. 3 (1996) 447.
[15] M. Shibasaki, E.M. Vogl, J. Organomet. Chem. 576 (1999) 1.
[16] M. Ogasawara, K. Yoshida, T. Hayashi, Heterocycles 52 (2000)
195.
5.17. Determination of enantiomeric excess of methyl
3,8a-dihydro-4H-naphthalene-4a-carboxylate (32)
[17] D. Kiely, P.J. Guiry, Curr. Org. Chem., in press.
[18] N.E. Carpenter, D.J. Kucera, L.E. Overman, J. Org. Chem. 54
(1989) 5846.
Enantioselectivities were determined by chiral HPLC
analysis using a Daicel Chiralpak OJ column (0.46 cm25
[19] Y. Sato, M. Sodeoka, M. Shibasaki, J. Org. Chem. 54 (1989)
4738.
cm), 0.3 ml minꢂ1, hexane: 100% (tRꢁ
/
84.2 (ꢂ) and
/
[20] Y. Sato, T. Honda, M. Shibasaki, Tetrahedron Lett. 33 (1992)
2593.
93.7 (ꢀ) min).
/
[21] Y. Sato, S. Nukui, M. Sodeoka, M. Shibasaki, Tetrahedron Lett.
50 (1994) 371.
[22] K. Kagechika, M. Shibasaki, J. Org. Chem. 56 (1991) 4093.
[23] A. Kojima, C.D.J. Boden, M. Shibasaki, Tetrahedron Lett. 38
(1997) 3459.
Acknowledgements
The asymmetric Heck project and D.K. have been
supported by an Enterprise Ireland Basic Research
Award (BR/98/186). The award of a UCD Open
Postgraduate Scholarship is gratefully acknowledged
by D.K. We thank Merck Sharpe and Dohme (Ireland)
Ltd. for their support of and interest in our Heck
research and also for financial support of D.K., and
Johnson Matthey for a loan of Pd salts. We thank Dr
Tim O’Sullivan, Ms Helen McManus and Mr Kevin
Murtagh for a critical reading of this manuscript.
[24] L.F. Tietze, K. Thede, R. Schimpf, F. Sannicolo, Chem. Com-
mun. (2000) 583.
[25] O. Loiseleur, P. Meier, A. Pfaltz, Angew. Chem. Int. Ed. Engl. 35
(1996) 200.
[26] O. Loiseleur, M. Hayashi, M. Keenan, N. Schmees, A. Pfaltz, J.
Organomet. Chem. 576 (1999) 16.
[27] L. Ripa, A. Hallberg, J. Org. Chem. 62 (1997) 595.
[28] C.A. Busacca, D. Grossbach, R.C. So, E.M. O’Brien, E.M.
Spinelli, Org. Lett. 5 (2003) 595.
[29] A. Togni, T. Hayashi (Eds.), Ferrocenes, Vheim, 1995, and
references therein.
[30] Y. Nishibayashi, S. Uemura, Synlett (1995) 79.
[31] C.J. Richards, T. Damalidis, D.E. Hibbs, M.B. Hursthouse,
Synlett (1995) 74.
[32] Y. Malone, PhD. Thesis, National University of Ireland, 1998.
[33] Y.M. Malone, P.J. Guiry, J. Organomet. Chem. 603 (2000) 110.
[34] A.J. Hennessy, Y.M. Malone, P.J. Guiry, Tetrahedron Lett. 40
(1999) 9163.
References
[1] A. de Meijere, F.E. Meyer, Angew. Chem. Int. Ed. Engl. 33 (1994)
2379.
[35] A.J. Hennessy, Y.M. Malone, P.J. Guiry, Tetrahedron Lett. 41
(2000) 2261.
[2] I.P. Beletskaya, A.V. Cheprakov, Chem. Rev. 100 (2000) 3009.
[3] T. Jeffery, in: L.S. Liebeskind (Ed.), Advances in Metal-Organic
[36] A.J. Hennessy, D.J. Connolly, Y.M. Malone, P.J. Guiry, Tetra-
hedron Lett. 41 (2000) 7757.
Chemistry, vol. 5, JAI Press, Greenwich, 1996, pp. 153Á260.
/
[4] M. Beller, T.H. Riermeir, G. Stark, in: M. Beller, C. Bohm (Eds.),
Transition Metals for Organic Synthesis, Wiley-VCH, Weinheim,
1998, p. 208.
[37] T.G. Kilroy, A.J. Hennessy, D.J. Connolly, Y.M. Malone, A.
Farrell, P.J. Guiry, J. Mol. Cat. A: Chem. 196 (2003) 65.
[38] D. Kiely, P.J. Guiry, Tetrahedron Lett. 43 (2002) 9545.
[39] D. Kiely, P.J. Guiry, Tetrahedron Lett. 44 (2003) 7377.
[40] W. Cabri, L. Candiani, S. DeBernardinis, F. Francalanci, S.
Penco, R. Santi, J. Org. Chem. 56 (1991) 5796.
[41] F. Ozawa, A. Kubo, T. Hayashi, J. Am. Chem. Soc. 113 (1991)
1417.
[5] J.T. Link, L.E. Overman, in: F. Diederich, P.J. Stang (Eds.),
Metal-Catalysed Cross-coupling Reactions (Chapter 6), Wiley-
VCH, Weinheim, 1998, p. 231 (Chapter 6).
[6] C. Amatore, G. Broeker, A. Jutand, F. Khalil, J. Am. Chem. Soc.
119 (1997) 5176.
[7] K.K. Hii, T.D.W. Claridge, J.M. Brown, Angew. Chem. Int. Ed.
Engl. 109 (1997) 1033.
[42] A. Ashimori, B. Bachand, L.E. Overman, D.J. Poon, J. Am.
Chem. Soc. 120 (1998) 6477.
˚
[8] M. Ludwig, S. Stromberg, M. Svensson, B. .Akermark, Organo-
[43] K. Ohrai, K. Kondo, M. Sodeoka, M. Shibasaki, J. Am. Chem.
Soc. 116 (1994) 11737.
¨
metallics 18 (1999) 970.
[9] K.K. Hii, T.D.W. Claridge, J.M. Brown, A. Smith, R.J. Deeth,
Helv. Chim. Acta 84 (2001) 3043.
[44] A. McKillop, J.C. Fiaud, R.P. Hug, Tetrahedron 30 (1974) 1379.
[45] R. Deziel, Tetrahedron Lett. 28 (1978) 4371.
[46] M. McCarthy, P.J. Guiry, Tetrahedron 55 (1999) 3061.
[47] Y. Sato, M. Sodeoka, M. Shibasaki, Chem. Lett. (1990) 1953.
[48] H. Iio, M. Isobe, T. Kawai, T. Goto, Tetrahedron 35 (1979) 941.
[10] M. Shibasaki, F. Miyazaki, in: E. Negishi (Ed.), Handbook of
Organopalladium Chemistry for Organic Synthesis, vol. 1, Wiley,
Hoboken, 2002, p. 1283.