Tandem palladium-catalyzed cyclocarbonylation of isolimonene: A mechanistic investigation and theoretical calculations on the fully diastereoselective step

Article abstract of DOI:10.1002/ejic.200300540

Cyclocarbonylation of isolimonene catalyzed by complexes of the type [HPd(SnCl₃)L₂] provides two cyclopentanone isomers (2a and 2b) containing two new stereogenic centers with a good diastereomeric excess (up to 69% with L₂ = dppf). These results show that chiral phosphane ligands are not necessary to ensure asymmetric induction. The diastereomeric excess is due, in fact, to the substrate itself, assisted by the steric hindrance of the diphosphanes. A full characterization of the two isomers by ¹H, ¹³C and DPFGSE NOE NMR spectroscopy gives all the signal assignations and all the positions of the substituents of interest on the chiral carbon atoms. In order to have a better understanding of the catalytic process, we also present a density functional study of the crucial intermediate species 4 involved in the proposed catalytic cycle. Our calculations show that there is no coordination of the C=C bond in an exo mode. Conversely, we have found a pentacoordinate species with a trigonal bipyramidal geometry in which the C=C bond is coordinated in an endo mode. Selected bond lengths and bond orders are reported. The calculated net charge distribution supports the cyclization process, which proceeds through a C-C(O) and a C-Pd coupling. A β-hydride elimination reaction of 4 provides the two 2a and 2b isomers. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejic.200300540

FULL PAPER
Tandem Palladium-Catalyzed Cyclocarbonylation of Isolimonene:
A Mechanistic Investigation and Theoretical Calculations on the Fully
Diastereoselective Step
[a]
[b]
[a]
[b]
´
¨
Geraldine Lenoble, Corinne Lacaze-Dufaure, Martine Urrutigoıty, Claude Mijoule,
and Philippe Kalck*[a]
Keywords: Asymmetric induction / Cyclocarbonylation / Density functional calculations / Homogeneous catalysis / NMR
spectroscopy / Palladium
Cyclocarbonylation of isolimonene catalyzed by complexes
of the type [HPd(SnCl3)L2] provides two cyclopentanone iso-
mers (2a and 2b) containing two new stereogenic centers
the crucial intermediate species 4 involved in the proposed
catalytic cycle. Our calculations show that there is no coor-
dination of the C=C bond in an exo mode. Conversely, we
have found a pentacoordinate species with a trigonal bipyr-
with a good diastereomeric excess (up to 69 % with L2
=
dppf). These results show that chiral phosphane ligands are amidal geometry in which the C=C bond is coordinated in
not necessary to ensure asymmetric induction. The diastere-
omeric excess is due, in fact, to the substrate itself, assisted
an endo mode. Selected bond lengths and bond orders are
reported. The calculated net charge distribution supports the
by the steric hindrance of the diphosphanes. A full character- cyclization process, which proceeds through a C−C(O) and a
ization of the two isomers by 1H, 13C and DPFGSE NOE NMR
spectroscopy gives all the signal assignations and all the po-
sitions of the substituents of interest on the chiral carbon
C−Pd coupling. A β-hydride elimination reaction of 4 pro-
vides the two 2a and 2b isomers.
atoms. In order to have a better understanding of the cata- ( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
lytic process, we also present a density functional study of
Germany, 2004)
Introduction
alkoxycarbonylation of alkenes.[5] During the nineties, great
effort went into developing enantio- or diastereoselective
Carbon monoxide is a very powerful building block in catalysts by introduction of chiral ligands into the coordi-
organic synthesis, provided it can be inserted into a frame- nation sphere of a metal.[6] Our studies on the car-
work with high selectivity. For several years we have used bonylation of (1R,2S,5R)-isopulegol, to provide the corre-
this strategy to functionalize monoterpenes.[1] Particularly sponding lactone, and our preliminary results on the car-
interesting are the tandem reactions in which functionality bonylation of (1R,4R)-isolimonene [(3R,6R)-3-isopropenyl-
already present on the terpene reacts with an active inter- 6-methylcyclohexene] have shown that the introduction of
mediate resulting from the insertion of carbon monoxide.[2] a chiral diphosphane ligand does not influence the dia-
Recently, we have specially focused our attention on the stereoselectivity of the two cyclocarbonylation reactions of
cyclocarbonylation of isolimonene, where the endocyclic interest.[2,3] Thus, the enantiofacial selectivity should be in-
carbon-carbon double bond is involved in the process; pre- duced by the chiral substrate itself. The situation of isoli-
liminary results have been reported.[3] Rhodium and pal- monene appeared to us particularly interesting because, of
ladium play a central role in such carbonylation reactions the three chiral centers present in the resulting product, two
and many papers have reported the catalytic activity of of them (C1 and C5 in Scheme 1) are obtained with exclus-
complexes such as [HRh(CO)L3] or [PdCl2L2].[4] Palladium ively one configuration (1R and 5S). Thus, the dia-
complexes, often with SnCl2 as co-catalyst, are active in the stereomeric excess results in fact from an enantiomeric ex-
cess at C4 to provide 2a and 2b of Scheme 1.
[a]
`
Laboratoire de Catalyse, Chimie Fine et Polymeres, Ecole
In order to understand why the cyclization reaction cata-
lyzed by palladium occurs exclusively at the C2 atom of
isolimonene (which becomes C1 in the resulting product,
see Scheme 1) and provides only the R configuration, we
have performed quantum chemical calculations using the
density functional theory (DFT) on the main intermediate
which would give the enantioselective cyclization step.
´
´
Nationale Superieure des Ingenieurs en Arts Chimiques et
Technologiques
118 route de Narbonne, 31077 Toulouse Cedex 4, France
Fax: (internat.) ϩ 33-5-62885600
E-mail: Philippe.Kalck@ensiacet.fr
[b]
´
CIRIMAT, Laboratoire des Interfaces et Materiaux, Ecole
Nationale Superieure des Ingenieurs en Arts Chimiques et
´
´
Technologiques
118 route de Narbonne, 31077 Toulouse Cedex 4, France
Eur. J. Inorg. Chem. 2004, 791Ϫ797
DOI: 10.1002/ejic.200300540
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
791

G. Lenoble, C. Lacaze-Dufaure, M. Urrutigo¨ıty, C. Mijoule, P. Kalck
FULL PAPER
tional approach, which is well adapted to this type of prob-
lem.[8,9]
All structures presented here were fully optimized using
conjugate gradient methods without constraints at the den-
sity functional theory (DFT) level with the Gaussian 98
software package release A.3.[10] The method has been de-
scribed extensively elsewhere.[11Ϫ13] We used the B3LYP hy-
brid functional.[14,15] For all optimizations a valence basis-
set of double-ξ quality was employed and relativistic effects
were addressed implicitly by the use of relativistic effective
core potentials for Pd, P and Cl.[16Ϫ18] For H, C and O the
standard DunningϪHay D95V basis was used.[19] A Mul-
liken population analysis, together with significant Mayer
bond orders between palladium and its neighbors were thus
determined. Finally, the charge-transfer effects arising be-
tween the metal atom and the ligands were examined.
Scheme 1. Cyclocarbonylation of (1R,4R)-isolimonene (1) cata-
lyzed by palladium
This paper deals with the complete identification of the
two diastereomers 2a and 2b resulting from the car-
bonylation of (1R,4R)-isolimonene (1) based on DPFGSE
NOE experiments[7] as well as the configuration at all the
chiral carbon atoms. The analysis of the structure of these
two isomers leads us to confirm that the active species con-
tains a palladium-hydride bond. Calculations have been
carried out on the two isomers 4a and 4b, and the two other
possible diastereomers. DFT computations have been done
on the crucial intermediate species, which contains an acyl
group, and the fate of the endocyclic carbon-carbon double
bond and the palladium center either in an endo or in an
exo mode (Scheme 2); such calculations show that an en-
antioselective C1ϪC2 coupling in the endo mode is the only
one possible. Finally, the full mechanism of the tandem car-
bonylation reaction is considered.
Results and Discussion
Catalytic Experiments
Carbonylation of (1R,4R)-isolimonene (1; Scheme 1) at
40 bar (4 MPa) catalyzed by [PdCl2(PPh3)2] in the presence
of SnCl2·2H2O and a slight excess of PPh3, leads to a mix-
ture of the two cyclopentanones 2a and 2b with good selec-
tivity. These two products have been shown by mass spec-
trometry to contain a C11 framework and a ketone func-
tion; further strong evidence for the presence of a CO group
results from the νCO band at 1740 cmϪ1 in the IR spectrum
of both products. The mixture was enriched in one product
to reach a 2a/2b ratio of 90:10 by column chromatography
on silica. 1H and 13C NMR characterizations confirmed
that 2a and 2b are two diastereomers, and a complete as-
signment of the various configurations at the carbon atoms
was obtained by NOE experiments, as discussed later.
Scheme 3. Isomers of 1
From Table 1 it can be seen that a 95 % conversion of
the starting material can be obtained in 18 hours, giving a
selectivity in 2aϩ2b of 84 %, operating with a substrate-to-
catalyst ratio of 50. The other products, which account for
16 %, were identified by GC-MS as isomers of 1, the most
Scheme 2. Representation of two conceivable intermediates, 2a re-
sulting from an endo coordination mode and 2b from an exo mode
Method of Calculation
As the complex [PdCl2(PPh3)2] reacts to give the same abundant of which are shown in Scheme 3.
products in catalysis as [PdCl(SnCl3)(PPh3)2], although
Addition of more substrate, as in runs 2 and 3, decreases
somewhat more slowly, we chose to model the species the conversion but does not affect significantly the selec-
[PdCl2(PH3)2] in which PPh3 has been replaced by the PH3 tivity in 2. The 2a/2b ratio gives rise to a diastereomeric
ligand, as is normal in these types of calculations. The sys- excess of 12 % in the three experiments.
tem contains a palladium atom, which is a transition metal
We then introduced the classical chiral ligand (ϩ)-DIOP
and thus has strongly correlated d electrons. Consequently, into the coordination sphere of palladium and prepared the
we chose to carry out our study by using the density func- catalyst precursor [PdCl2{(ϩ)-DIOP}] to carry out run 4.
792
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 791Ϫ797

Palladium-Catalyzed Cyclocarbonylation of Isolimonene
FULL PAPER
Table 1. Cyclocarbonylation reactions of (1R,4R)-isolimonene (1)[a]
Catalytic precursor
Substrate/catalyst
Conversion (%)
Selectivity (%)
2a/2b[b]
de[b] (%)
1
2
3
4
5
6
7
[PdCl2(PPh3)2]
[PdCl2(PPh3)2]
[PdCl2(PPh3)2]
[PdCl2{(ϩ)-DIOP}]
[PdCl2{(Ϫ)-DIOP}]
[PdCl2(dppb)][c]
[PdCl2(dppf)][c]
50
80
100
100
100
100
100
95
78
62
75
75
75
71
84
80
82
52
52
65
68
56:44
55:45
56:44
85:15
85:15
82:18
84.5:15.5
12
10
12
70
70
64
69
[a]
General conditions: Catalytic precursor ϭ 1 mmol, SnCl2·2H2O ϭ 2.5 mmol, P/Pd ϭ 4, toluene ϭ 25 mL, temp. ϭ 70 °C, pCO ϭ 40
[b]
[c]
bar, t ϭ 18 h. Determined by GC. Temp. ϭ 97 °C.
Under the same catalytic conditions as in run 3, a 75 % activity.[20] It can therefore be concluded that simple di-
conversion of 1 and a 52 % selectivity in 2 were observed. phosphane ligands can play an important role, and thus
However, a diastereomeric excess of 70 % was calculated assist the asymmetric catalysis, presumably through their
from the relative amounts of 2a and 2b. A similar experi- electronic effects during the various steps of the catalytic
ment carried out with [PdCl2{(Ϫ)-DIOP}] led to the same cycle, but also thanks to their bulkiness during the asym-
results, especially the 2a/2b ratio, which is still 85:15 and metric induction.
not the reverse. Therefore, whatever the configurations of
the two chiral carbons of the diphosphane ligand, they have
no influence on the diastereoselectivity and we suspected
Stereochemistry Determination of 2a and 2b by NMR
Spectroscopy
that the reaction could be governed by the chirality of the
substrate itself.
In previous studies, due to the isolation in the solid state
of one isomer of the lactone obtained by cyclocar-
Various experiments, not reported here, showed that a bonylation of (1R,2S,5R)-isopulegol and solved by an X-
change in the temperature, pressure, or L*/Pd ratio did not ray crystal structure, we have been able to determine all the
affect significantly these performances, except maybe a configurations of the chiral centers present in the molecule
higher pressure, which improved the conversion somewhat; and its diastereoisomers by NMR spectroscopy.[2d,3] With
for instance, at 80 bar, the yield is 82 %.
this knowledge, we have used NOE experiments to assign
For this reason we have prepared two catalyst precursors the positions of the various substituents of interest in 2a
containing the classical diphosphane ligands bis(diphenyl- and 2b. NMR spectroscopic data for the two diastereomers
phosphanyl)butane (dppb) and bis(diphenylphosphanyl)fer- 2a and 2b are displayed in Table 2, with the numbering sys-
rocene (dppf). Runs 6 and 7 show that the performances of tem shown in Scheme 4, which differs from normal terpene
1
these two non-chiral catalysts are essentially the same as nomenclature. These data were obtained from 1D H and
1
those of the DIOP-containing catalysts: the two dia- 13C NMR spectra and COSY (1H,1H), HMQC (1H,13C) J
stereomeric excess values are 64 % and 69 % respectively. In (see Figure 1) as well as long-range experiments. Due to the
experiments not described here, we also checked that presence of a five-membered cycle and an endocyclic car-
shorter carbon chains in the ligands, like bis(diphenylphos- bon-carbon double bond, most of the substituents are lab-
phanyl)ethane and -propane, did not lead to any catalytic eled as pseudo-equatorial in Schemes 5 and 6.
Table 2. NMR spectroscopic data for diastereomers 2a and 2b in CDCl3
Atom
2a
2b
[b]
δ(1H)[a]
2JH,H
3JH,H
δ(13C)
δ(1H)[a]
2JH,H
3JH,H
δ(13C)
[b]
[b]
[b]
1
2
3pa
3pe
4
2.45 (dd)
49.3
217.7
45.0
2.35 (dd)
52.0
218.0
41.9
2.04 (dd)
1.52 (dd)
1.70 (m)
1.70 (m)
1.54 (m)
1.35 (m)
1.54 (m)
1.35 (m)
18.3 (3pe)
18.3 (3pa)
6.8 (4)
8.1 (4)
1.92 (dd)
1.52 (dd)
2.00 (m)
1.90 (m)
1.55 (m)
1.35 (m)
1.55 (m)
1.35 (m)
18 (3pe)
6 (4)
31.0
41.1
21.8
31.4
38.2
19.8
5
6pa
6pe
7pa
7pe
8
9
10
11
26.9
29.0
135.6
116.1
18.7
135.7
116.4
15.1
4.90 (d)
0.76 (d)
1.30 (s)
1.4 (1)
6.4 (4)
5.16 (d)
0.76 (d)
1.30 (s)
1.5 (1)
6.4 (4)
23.6
23.2
[a]
[b]
Resonance multiplicities in parentheses. Coupled partners in parentheses.
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 791Ϫ797
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
793

G. Lenoble, C. Lacaze-Dufaure, M. Urrutigo¨ıty, C. Mijoule, P. Kalck
FULL PAPER
overlap of several signals. As the spatial arrangement at C5
does not change during the reaction [C1 (R) in isolimonene
becomes C5 (S) in cyclopentanone], irradiation of H1 on
C1 in 2a (Figure 2) allowed us to detect interactions with
H9, H7 in a pseudo-equatorial position (pe), H5pe, H10pe,
H11pe and H3pe (Scheme 5) indicating that all these atoms
or groups are in the up position. Thus, C1 presents an R
configuration, and C4 an R one.
Scheme 4. Compound 2 (with IUPAC nomenclature)
Figure 2. 1D NOESY-DPFGSE spectrum (400 MHz; irradiation
of H1 on C1 in 2a) of mixture of diastereomers 2a and 2b at 298 K
in CDCl3; top: full spectrum; middle: mixing time 800 ms; bottom:
mixing time 400 ms
Figure 1. 2D COSY.DQF.GE (1H,1H) spectrum (400 MHz) of mix-
ture of diastereomers 2a and 2b at 298 K in CDCl3
In the same 90:10 mixture of 2a/2b, irradiation of H1 in
2b (Figure 3) led us to assign the configurations C1 (R) and
C4 (S; Scheme 6).
Scheme 5. Interactions observed for H1 in 2a
Scheme 6. Interactions observed for H1 in 2b
Figure 3. 1D NOESY-DPFGSE spectrum (400 MHz; irradiation
of H1 on C1 in 2b) of mixture of diastereomers 2a and 2b at 298 K
in CDCl3; top: full spectrum; middle: mixing time 800 ms; bottom:
The measurement of NOE effects by a 1D DPFGSE
method appears to be much more selective for resonances
that are not well resolved, especially when this is due to the mixing time 400 ms
794  2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.eurjic.org
Eur. J. Inorg. Chem. 2004, 791Ϫ797

Palladium-Catalyzed Cyclocarbonylation of Isolimonene
FULL PAPER
This NMR study shows that in the two products ob-
tained in the cyclocarbonylation reaction, the stereochem-
istry of C1 is unchanged in both isomers. The dia-
stereomeric excess of 70 % results in fact from the two (S
and R) configurations at the C4 atom (R/S ϭ 85:15).
DFT Calculations
Scheme 8. Coordination of the endocyclic CϭC bond of 1 to the
The shift of the endocyclic CϭC bond from the substrate
1 to the resulting cyclopentanone 2 (Scheme 7) is indicative
of a β-hydride elimination in an intermediate which affords
a palladium hydride species. Thus, we have supposed the
following sequence for the early steps of the catalytic cycle:
formation of an alkylpalladium species by hydride transfer
to the exocyclic CϭC bond, and CO migratory insertion to
afford the acyl intermediate 3. During the cyclization pro-
cess, which involves the carbonϪcarbon coupling of the two
C1 and C2 atoms, we have considered that the stereoselec-
tivity results from one privileged approach of the palladium
center toward the endocyclic CϭC bond in species 4
(Scheme 8). As the full system involves 39 atoms the poten-
tial hypersurface is rather complex, therefore we calculated
only certain regions of this hypersurface, taking into ac-
count the classical geometries for d8 complexes.
palladium center
nate species with a trigonal bipyramidal (TBP) geometry
in which the CϭC bond is coordinated in an endo mode
(Scheme 9). Indeed, the two palladiumϪcarbon bond
˚
lengths are 2.31 and 2.35 A, corresponding to Mayer bond-
orders of 0.36 and 0.31, respectively; in addition, the CϪC
˚
bond length is 1.40 A and its bond order is 1.53 (see
Table 3). Among the various axial and/or equatorial posi-
tions for the ligands, the calculations seem to slightly favor
(1.3 kcal·molϪ1) the Cl and CO-acyl ligands in the two ax-
ial positions.
Concerning the Mulliken population analysis, our results
show clearly that there are no significant differences be-
tween the net charges on each atom in both species.
Proposed Catalytic Cycle
Our calculations give strong evidence that the TBP inter-
mediate in which the (endo)CϭC double bond and the CO-
acyl are in close vicinity allows the next step in the catalytic
˚
cycle. Indeed, the C2ϪC1 distance is 2.97 A, whereas it is
˚
3.58 and 4.41 A in the less- and more-stable square-planar
palladium species. The resulting C1 configuration is thus
100 % R.
A further analysis of the mechanism shows that the new
intermediate 5, in which the cyclopentanone has been
formed and the palladium atom is still bonded to C9, pre-
sents the two palladium and H8 hydrogen atoms in mutual
cis positions. Thus, a β-hydride elimination reaction, which
re-forms the active palladium hydride species, is possible. In
addition, the H1 atom is trans to Pd and cannot provide
the β-H elimination required for the further step. Moreover,
it is worth mentioning that a possible intermediate resulting
from the coordination of the CϭC bond in the exo mode
would provide the two H1 and H8 hydrogen atoms in trans
positions to the palladium atom; such a situation, in which
no β-H elimination can occur, precludes any catalytic reac-
tion. We did not detect these two isomers. Scheme 7 shows
the catalytic cycle that we propose.
Scheme 7. Catalytic cycle of cyclocarbonylation of 1
Conclusion
Our calculations show that there is no coordination of
the CϭC bond in an exo mode and we found two species
The main point of this study is related to an asymmetric
with the CϭC bond not coordinated to the metal center, induction that is caused by the substrate itself so that chiral
which adopts a square-planar environment, at all (dPdϪC
ϭ
ancillary phosphane ligands are not necessary in the cata-
˚
˚
4.75 and 5.63 A for the more stable and 3.69 and 4.66 A lytic reaction. Moreover, the steric hindrance of the diphos-
for the second one, which is 3 kcal·molϪ1 higher in energy, phane ligands plays an essential role in obtaining a large
as shown in Table 3). Moreover, we found a pentacoordi- diastereomeric excess.
Eur. J. Inorg. Chem. 2004, 791Ϫ797
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
795

G. Lenoble, C. Lacaze-Dufaure, M. Urrutigo¨ıty, C. Mijoule, P. Kalck
FULL PAPER
Table 3. Absolute energies, representative bond lengths, selected Mayer bond orders and Mulliken net charges distribution in the inter-
mediate species 4
˚
Complexes
Absolute energy (Hartree)
Distances (A)
Bond order
Net charges
CϭC uncoordinated species
Ϫ662.862124
PdϪC9: 5.63
PdϪC1: 4.75
C1ϪC9: 1.35
PdϪC9: 4.66
PdϪC1: 3.69
C1ϪC9: 1.35
PdϪC9: 0.00
PdϪC1: 0.00
C1ϪC9: 1.95
PdϪC9: 0.00
PdϪC1: 0.00
C1ϪC9: 1.90
Pd: Ϫ0.17
C1: Ϫ0.25
C9: Ϫ0.23
Pd: Ϫ0.17
C1: Ϫ0.26
C9: Ϫ0.18
Ϫ662.857868
Ϫ662.856565
CϭC coordinated species
PdϪC9: 2.35
PdϪC1: 2.31
C1ϪC9: 1.40
PdϪC9: 0.31
PdϪC1: 0.36
C1ϪC9: 1.53
Pd: Ϫ0.14
C1: Ϫ0.23
C9: Ϫ0.20
duced into a 250 mL stainless steel autoclave with mechanical stir-
ring. nitrogen-saturated mixture of isolimonene (13.624 g,
A
100 mmol) in toluene (25 mL) was introduced into the evacuated
autoclave by aspiration. It was heated to 70 °C under 40 bar press-
ure of carbon monoxide at constant pressure. After 18 h, the auto-
clave was cooled and then slowly depressurized. The yellow-orange
reaction mixture was analyzed by gas chromatography. The same
procedure was followed for other ligands maintaining the P/Pd ra-
tio at 4.
After the catalytic reaction, the organometallic compounds and the
excess phosphane were separated from the crude solution by wash-
ing with CCl4. The deeply colored oily layer was decanted. After
concentration by rotary evaporation, the oily residue was purified
by column chromatography on silica gel (petroleum ether/dichloro-
methane/ethyl acetate, 70:24:6).
Scheme 9. Compound 4 (CϭC endo coordination mode)
Experimental Section
Complexes: [PdCl2(PPh3)2] was prepared according to the literature
procedure.[21] Yield 95 %. 31P{1H} NMR (250 MHz, CDCl3, 25
°C): δ ϭ 23.47 (s) ppm. IR (KBr): ν˜ ϭ 1481, 1436, 1099, 746,
693 cmϪ1. Complexes containing diphosphanes were prepared by
addition of the diphosphane to [PdCl2(PhCN)2][22] in hot acetone.
[PdCl2(dppb)]: Yield 90 %. 31P{1H} NMR (250 MHz, CDCl3, 25
°C): δ ϭ 63.75 (s) ppm.
Materials: Solvents with high purity, as well as (1R,4R)-isoli-
´
´
´
monene, were used as received from Derives Resiniques et Ter-
´
peniques (DRT SA). SnCl2·2H2O (Prolabo 98 %), triphenylphos-
phane (Aldrich 99 %), 1,4-bis(diphenylphosphanyl)butane (dppb)
(Acros 98 %), 1,1Ј-bis(diphenylphosphanyl)ferrocene (dppf) (Ald-
rich 97 %), (Ϫ)-2,3-O-isopropylidene-2,3-dihydroxy-1,4-bis(diphos-
phanyl)butane [(Ϫ)-DIOP] and (ϩ)-2,3-O-isopropylidene-2,3-di-
hydroxy-1,4-bis(diphosphanyl)butane [(ϩ)-DIOP] (Janssen 98 %).
Column chromatography was performed with SiO2 (Merck).
[PdCl2(dppf)]: Yield 88 %. 31P{1H} NMR (250 MHz, CDCl3, 25
°C): δ ϭ 36.14 (s) ppm.
[PdCl2[(ϩ)-DIOP)] and [PdCl2[(Ϫ)-DIOP)]: Yield 85 %. 31P{1H}
NMR (250 MHz, CDCl3, 25 °C): δ ϭ 16.47 (s) ppm.
1
Instrumentation: The H, 13C and 31P NMR spectra were recorded
(1R,4R,5S)-4,8-Dimethylbicyclo[4.3.0]non-8-en-2-one
NMR (400 MHz, CDCl3, 25 °C, TMS): δ ϭ 4.90 (d, JH,H
1.4 Hz, 1 H, H9), 2.45 (dd, n.d., 1 H, H1), 2.04 (dd, JH,H ϭ 18.3,
(2a):
1H
ϭ
with CDCl3 solutions containing TMS as internal standard on a
Bruker AMX 400 spectrometer operating at 400.13 MHz (1H) and
100.62 MHz (13C), and a Bruker AM 250 spectrometer operating
at 250.13 MHz, 62.90 MHz and 101.26 MHz, respectively. Data are
reported as follows: chemical shift in ppm (δ), multiplicity (s ϭ
singlet, d ϭ doublet, t ϭ triplet, q ϭ quadruplet and m ϭ mul-
tiplet), coupling constant (J) in Hz (n.d. ϭ no determined), inte-
gration, and assignment (pa ϭ pseudo-axial and pe ϭ pseudo-
equatorial). IR spectra were obtained on a PerkinϪElmer 1710
spectrometer; absorptions are reported in cmϪ1. Analytical GC was
carried out on a Carlo Erba MFC 500 apparatus equipped with a
Econo-Cap FFAP (30 m; 0.53 mm; 1.2 µm) capillary column and
a flame ionization detector. Products were identified by GC/MS on
a PerkinϪElmer QMass 910 Mass Spectrometer with ionization
voltage of 70 eV, with a Crompack CP WAX 52 CB (50 m;
0.32 mm; 0.2 µm) polar column.
3
3
6.8 Hz, 1 H, H3pa), 1.70 (m, 2 H, H4 and H5), 1.54 (m, 2 H, H6pa
3
and H7pa), 1.52 (dd, JH,H ϭ 8.1, 18.3 Hz, 1 H, H3pe), 1.35 (m, 2
H, H6pe and H7pe), 1.30 (s, 3 H, CH3 11), 0.76 (d, 3JH,H ϭ 6.4 Hz,
3 H, CH3 10) ppm 13C{1H} NMR (400 MHz, CDCl3, 25 °C): δ ϭ
217.7 (C2), 135.6 (C8), 116.1 (C9), 49.3 (C1), 45.0 (C3), 41.1 (C5),
31.0 (C4), 26.9 (C7), 23.6 (C11), 21.8 (C6), 18.7 (C10) ppm. IR
(KBr): ν˜ ϭ 1740 cmϪ1 (CϭO). MS (70 eV, EI): m/z (%) ϭ 79 (100)
[C6H7]ϩ, 94 (100) [C7H10], 164 (61) [M], 165 (9) [M Ϫ H]ϩ.
(1R,4S,5S)-4,8-Dimethylbicyclo[4.3.0]non-8-en-2-one (2b): 1H NMR
3
(400 MHz, CDCl3, 25 °C, TMS): δ ϭ 5.16 (d, JH,H ϭ 1.5 Hz, 1
H, H9), 2.35 (dd, n.d., 1 H, H1), 2.00 (m, 1 H, H4), 1.92 (dd,
3JH,H ϭ 6 and 18 Hz, 1 H, H3pa), 1.90 (m, 1 H, H5), 1.55 (m, 2
H, H6pa and H7pa), 1.52 (dd, n.d., 1 H, H3pe), 1.35 (m, 2 H,
3
General Procedure: A mixture of dichlorobis(triphenylphosphane)-
H6pe and H7pa), 1.30 (s, 3 H, CH3 11), 0.76 (d, JH,H ϭ 6.4 Hz,
palladium() (0.702 g, 1 mmol), hydrated tin() chloride (0.474 g, 3 H, CH3 10) ppm. 13C{1H} NMR (400 MHz, CDCl3, 25 °C): δ ϭ
2.5 mmol) and triphenylphosphane (0.524 g, 2 mmol) was intro-
796
218.0 (C2), 135.7 (C8), 116.4 (C9), 52.0 (C1), 41.9 (C3), 38.2 (C5),
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 791Ϫ797

Palladium-Catalyzed Cyclocarbonylation of Isolimonene
FULL PAPER
E. V. Gusevskaya, J. Mol. Catal. A: Chem. 1998, 132, 213Ϫ221.
Catalytic Asymmetric Synthesis (Ed.: K. Nozaki, I. Ojima),
Wiley-VCH, Weinheim, 2000, pp.429Ϫ464.
31.4 (C4), 29.0 (C7), 23.2 (C11), 19.8 (C6), 15.1 (C10) ppm. IR
(KBr): ν˜ ϭ 1740 cmϪ1 (CϭO). MS (70 eV, EI): m/z (%) ϭ 94 (100)
[C7H10], 79 (92) [C6H7]ϩ, 164 (61) [M], 165 (9) [M Ϫ H]ϩ.
[6]
[7] [7a]
Q. N. Van, A. J. Shaka, J. Magn. Reson. 1996, 119,
[7b]
295Ϫ302.
K. Stott, J. Keeler, Q. N. Van, A. J. Shaka, J.
Magn. Reson. 1997, 125, 302Ϫ309.
Acknowledgments
[8]
[9]
I. Papai, D. R. Sahalub, C. Mijoule, Surf. Sci. 1990, 236, 241.
J. Roques, C. Lacaze-Dufaure, C. Mijoule, Surf. Sci. 2001,
479, 231.
`
We thank the Ministere de l’Education Nationale de la Recherche
et de la Technologie for a Research Grant (L.G.). Acknowledg-
ments are also due to Engelhard-CLAL for a generous loan of
palladium, and DRT SA for the gift of (1R,4R)-isolimonene. We
are indebted to CALMIP for computer resources with the GAUS-
SIAN 98 software package.
[10]
M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M.
A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A. Montgom-
ery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M. Mil-
lam, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J.
Tomasi, V. Barone, M. Cossi, R. Cammi, B. Mennucci, C. Po-
melli, C. Adamo, S. Clifford, J. Ochterski, G. A. Petersson, P.
Y. Ayala, Q. Cui, K. Morokuma, D. K. Malick, A. D. Rabuck,
K. Raghavachari, J. B. Foresman, J. Cioslowski, J. V. Ortiz, B.
B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.
Gomperts, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L. Andres,
C. Gonzalez, M. Head-Gordon, E. S. Replogle, J. A. Pople,
Gaussian 98, Revision A.3, Gaussian, Inc., Pittsburgh PA,
1998.
[1] [1a]
`
I. Cipres, J. Jenck, Ph. Kalck, J. Mol. Catal. 1990, 58,
387Ϫ392. [1b] I. Cipres, Ph. Kalck, D. C. Park, F. Serein-Spirau,
`
[1c]
`
T. Chenal, I. Cipres, J.
J. Mol. Catal. 1991, 6, 399Ϫ407.
Jenck, Ph. Kalck, J. Mol. Catal. 1993, 78, 351Ϫ366.
[1d]
R.
`
Naigre, T. Chenal, I. Cipres, Ph. Kalck, J-C. Daran, J. Vaisser-
mann, J. Organomet. Chem. 1994, 480, 91Ϫ102. [1e] S. Sirol, Ph.
Kalck, New. J. Chem. 1997, 21, 1129Ϫ1137.
[2] [2a]
`
I. Cipres, Ph. Kalck, D. C. Park, J. Mol. Catal. 1991, 66,
[2b]
`
T. Chenal, R. Naigre, I. Cipres, Ph. Kalck, J. C.
399Ϫ407.
[11]
[12]
[13]
W. Kohn, L. J. Sham, Phys. Rev. 1965, 140, A1133.
P. Hohenberg, W. Kohn, Phys. Rev. 1964, 136, B864.
R. G. Parr, W. Yang, Density functional Theory of Atoms and
Molecules, Oxford Press, 1989.
Daran, J. Vaissermann, J. Chem. Soc., Chem. Commun. 1993,
[2c]
747Ϫ748.
S. Sirol, J. P. Gorrichon, Ph. Kalck, P. M. Nieto,
[2d]
G. Commenges, Magn. Reson. Chem. 1999, 37, 127Ϫ132.
G. Lenoble, R. Naigre, T. Chenal, M. Urrutigo¨ıty, J. C. Daran,
Ph. Kalck, Tetrahedron: Asymmetry 1999, 10, 929Ϫ936.
C. Lacaze-Dufaure, G. Lenoble, M. Urrutigo¨ıty, J. P. Gor-
richon, C. Mijoule, Ph. Kalck, Tetrahedron: Asymmetry 2001,
12, 185Ϫ187.
[14]
[15]
[16]
[17]
[18]
[19]
A. D. Becke, J. Chem. Phys. 1993, 88, 5648.
[3]
J. P. Perdew, W. R. Wang, Phys. Rev. B 1992, 45, 13244.
P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 270.
W. R. Wadt, P. J. Hay, J. Chem. Phys. 1985, 82, 284.
P. J. Hay, W. R. Wadt, J. Chem. Phys. 1985, 82, 299.
T. H. Dunning Jr, P. J. Hay, in Modern Theoretical Chemistry
(Ed.: H. F. Schaefer, III) Plenum, New York, 1976.
R. Naigre, PhD thesis, University of Toulouse (France), 1994.
D. R. Coulson, L. C. Satek, S. O. Orim, Inorg. Synth. 1972,
13, 121Ϫ124.
[4] [4a]
J. Tsuji, Palladium Reagents and Catalysts, John Wiley &
[4b]
Sons Ltd., Chichester, 1995.
sis with Organometallic Compounds (Eds.: B. Cornils, W. A.
Herrmann), VCH, Weinheim, 1996.
Applied Homogeneous Cataly-
[20]
[21]
[4c]
Transition Metals for
Organic Synthesis (Eds. M. Beller, C. Bolm), Wiley-VCH,
Weinheim, 1998.
[5] [5a]
[5b]
[22]
J. F. Knifton, J. Org. Chem. 1976, 41, 2885Ϫ2894.
G.
I.
J. M. Jenckins, J. C. Verkade, Inorg. Synth. 1968, 11, 108Ϫ111.
[5c]
Cavinato, L. Toniolo, J. Mol. Catal. 1981, 10, 161Ϫ170.
Received August 6, 2003
Cipres, J. Jenck, Ph. Kalck, J. Mol. Catal. 1990, 58, 387Ϫ392.
Early View Article
[5d]
L. L. da Rocha, A. O. Dias, E. N. dos Santos, R. Augusti,
Published Online January 2, 2004
Eur. J. Inorg. Chem. 2004, 791Ϫ797
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
797