Selectivity controlled by ligand tuning in the palladium-catalysed cyclocarbonylation: Synthesis of new γ and δ lactones from a natural sesquiterpene

Article abstract of DOI:10.1016/j.molcata.2015.02.010

The cyclocarbonylation of homoallylic alcohols, derived from α-atlantone, catalysed by [Pd(Cl₂)L₂]/SnCl₂] system, afforded new γ and δ lactones as the main products with high selectivities. The regiochemical control depends on the nature of phosphines involved. Catalysts based on monophosphines with large cone angles such as tricyclohexyl phosphine produce the six-membered ring lactones with selectivities up to 80%, while diphosphines with wide bite angle such as xantphos are found to favor the five-membered ring lactones with up to 99% selectivity. Monocrystals of the γ and δ lactones suitable for X-ray diffraction analysis have been obtained and the stereochemistry of the lactones 5a,a′ and 6 have been elucidated. This study is the first example of an attractive catalytic transformation of one single natural sesquiterpene easily available from bio-renewable resources to obtain either γ- or δ-lactones.

Full text of DOI:10.1016/j.molcata.2015.02.010

Accepted Manuscript
Title: Selectivity controlled by ligand tuning in the
palladium-catalysed cyclocarbonylation: synthesis of new ␥
and ␦ lactones from a natural sesquiterpene
Author: Jamal El Karroumi Abderrahim El Haib Eric
Manoury Ahmed Benharref Jean Claude Daran Maryse
Gouygou Martine Urrutigo ¨ı ty
PII:
S1381-1169(15)00058-8
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2015.02.010
MOLCAA 9426
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
5-12-2014
8-2-2015
10-2-2015
Please cite this article as: Jamal El Karroumi, Abderrahim El Haib, Eric Manoury,
¨
Ahmed Benharref, Jean Claude Daran, Maryse Gouygou, Martine Urrutigoity,
Selectivity controlled by ligand tuning in the palladium-catalysed cyclocarbonylation:
synthesis of new rmgamma and rmdelta lactones from a natural sesquiterpene, Journal
of Molecular Catalysis A: Chemical http://dx.doi.org/10.1016/j.molcata.2015.02.010
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Selectivity controlled by ligand tuning in the palladium-catalysed
cyclocarbonylation : synthesis of new  and  lactones from a natural
sesquiterpene
a,b
a,b
a*
b
Jamal El Karroumi , Abderrahim El Haib , Eric Manoury , Ahmed Benharref ,
a
a*
a*
Jean Claude Daran , Maryse Gouygou and Martine Urrutigoïty
a
CNRS ; LCC (Laboratoire de Chimie de Coordination) ; 205, route de Narbonne, F-31077
Toulouse, France
Université de Toulouse ; UPS, INPT ; LCC ; F-31077 Toulouse, France
b
,
Laboratoire de Chimie des substances naturelles Faculté des Sciences Semlalia,
Université Cadi Ayyad, B.P. 2390, Marrakech, Morroco
Corresponding authors e-mail: martine.urrutigoity@ensiacet.fr, gouygou@lcc-toulouse.fr,
*
manoury@lcc-toulouse.fr
Graphical abstract
Highlights

Palladium-catalyzed cyclocarbonylation of a natural sesquiterpene affords new
gamma- and delta-lactones


High conversions and good selectivities in lactones are obtained
Regioselectivity can be controlled by the nature of the ligand and a control by the
substrate itself is also observed

It’s an attractive catalytic transformation of a sesquiterpene easily available form bio-
renewableressources
Abstract
The cyclocarbonylation of homoallylic alcohols, derived from -atlantone, catalysed by
[
Pd(Cl
2
)L
2
]/SnCl ] system, afforded new  and  lactones as the main products with high
2
selectivities. The regiochemical control depends on the nature of phosphines involved.
Catalysts based on monophosphines with large cone angles such as tricyclohexyl phosphine
produce the six-membered ring lactones with selectivities up to 80%, while diphosphines with
wide bite angle such as xantphos are found to favour the five-membered ring lactones with up
to 99% selectivity. Monocrystals of the  and  lactones suitable for X-ray diffraction analysis
have been obtained and the stereochemistry of the lactones 5a,a’ and 6 have been elucidated.
This study is the first example of an attractive catalytic transformation of one single natural
sesquiterpene easily available from bio-renewable resources to obtain either - or -lactones.

Keywords: homogeneous catalysis, cyclocarbonylation, palladium, terpenes, lactones
1
. Introduction
Lactones are a broad family of compounds which can be very often found in nature [1]
and have been largely studied because of their biological properties and as building blocks
2], like, for instance, Corey’s lactone [3]. Because of their broad range of applications, many
[
synthetic efforts have been devoted to the synthesis of lactones [4], in particular in
enantiomerically enriched forms [5]. Out of the classical Baeyer-Villiger oxidation of cyclic
ketones [6,7], palladium metal-catalysed cyclocarbonylation of unsaturated alcohols
represents one of the most efficient and atom-economy approaches since it can provide access
[
8] to unsaturated [9] or saturated lactones [10,11,12] which are often difficult to selectively
synthesise. Using Pd(II) catalytic system, 5-membered ring saturated lactones can be obtained
from allylic alcohols [10] or homoallylic alcohols [11], while 6-membered ring saturated
lactones are obtained from alkenol with 3 carbons between the alcohol and the alkene
function (Scheme 1) [11a]. In addition, 2-allylphenols yield a mixture of 5-, 6- and 7-
membered ring lactones [12]
As a possible way to use biomass as a source of valuable chemicals, various
representative C10 monoterpenes have been also used as starting materials in the
cyclocarbonylation reactions by us [13] and others authors [10c] yielding lactones. The
palladium-based catalytic tool used permits to obtain these heterocycles with different ring
sizes. Indeed from geraniol a 6-membered lactone is obtained while perillyl alcohol leads to a
bicyclic 5-membered lactone and isopulegol to a bicyclic 6-membered one (Scheme 2).
More recently, as part of an ongoing program on the use of Moroccan essential oils as
feedstocks for fine chemistry [14], we have been interested by the essential oil of the Atlas
Cedar (Cedrus Atlantica), an interesting raw material for the perfume industry [15]. The main
components of this essential oil are the sesquiterpenic bicyclic hydrocarbons, -, - and -

himachalene, and the oxygenated part, -atlantone. We already worked on the catalytic
transformation of the himachalenes by Lewis acid catalysed rearrangement of their
monoepoxides [14].
In this paper, we will now focus on the oxygenated fraction which essentially contains the
sesquiterpenic ketone -atlantone easily isolated by fractional distillation. In the present
contribution, we present an efficient way to obtain new chiral - or -lactones from this
purified natural molecule based on palladium-catalysed cyclocarbonylation reaction as a key
step.
2
. Experimental
2
.1. Materials and general procedure
All commercially available reagents were used as received without purification. All reactions
were run under Argon using Schlenk techniques. Toluene, dichloromethane, pentane, THF
were dried under N using a solvent purification system (SPS).
2
Z, E α-atlantones were extracted according to the literature [16].
All palladium complexes were prepared according an already described method [17].
All NMR spectral data were recorded at 25 °C on a DPX300 or AV 500 spectrometers with
1
13
31
TMS as internal reference for H and C, 85% H
3
PO as external reference for P. Spectral
4
assignments were made by means of routine one and two dimensional NMR experiments
where appropriate, the data are reported as s (singlet), d (doublet), t (triplet), m (multiplet), dd
(
doublet of doublet) High-resolution mass spectra (HRMS) were recorded on GC/TOF using
chemical ionisation by desorption in methane. IR analysis was performed using a FT-IR
R
spectrometer. Analytical GC was carried out on a chromatograph with Stabilwax -AD column
and flame ionisation detector. HPLC analysis was done using a prep 4000 Waters. Single
crystal X-ray structures were determined on Gemini and Xcalibur diffractometers.
2
.2 Synthesis of 6-methyl-2-(p-tolyl)-hepta-4-one, 2

In a 2-necked flask equipped with a magnetic stirrer, a thermometer and a condenser was
charged a mixture of the two Z and E isomers of α-atlantone 1 (5 g, 23 mmol), and palladium
on activated charcoal (10%) (0.1 eq, 250 mg). The reaction mixture was heated at 160°C for
1
2h. The reaction was then cooled. 20 mL of dichloromethane was added and the solution was
filtered to remove palladium on activated charcoal. Then the solvent was removed and the
residue was separated by column chromatography with hexane/ethyl acetate (v/v 99/1); the
product was isolated in 80% yield as a yellow oil as a racemic mixture.
1
NMR H (298K, CDCl
3
, 500.33MHz): 7.12 (s, 2H, C11-H; C13-H); 7.12 (s, 2H, C10-H, C14-
H); 3.32 (m, 1H, C2-H); 2.71 (d, 1H, J=5Hz, C3-H); 2,61 (d, 1H, J=11Hz, C3-H); 2.33 (s,
3
0
H, C15-H); 2.22 (d, 2H, J=5Hz, C5-H); 2.11 (m, 1H, C6-H); 1.27 (d, 3H, J=5Hz, C1-H);
1
3
1
.89 (s, 3H, C8-H); 0.88 (s, 3H, C7-H). NMR C{ H} (298K, CDCl , 100.613 MHz) : 209.8
3
(
C4); 143.3 (C9); 135.4 (C12); 129.0 (C11 , C13); 126.9 (C10,C14); 52.7 (C3); 51.4 (C5);
-1
3
5.5 (C2); 25.6 (C15); 22.8 (C1); 23.4 (C6); 24.5 (C7); 24.1 (C8). IR  (C=O): 1710 cm .
+
+
HRMS (DCI-CH ) (m/z) : calcd for [C15H22O +H] [M+H] 219.1749 found 219.1745
4
2
.3 Synthesis of 6-p-tolyl-4-isobutylhept-1-en-4-ol, 4a and 4b
Under dry N atmosphere in a schlenk tube, compound 1 (1g, 4.58mmol) was dissolved in
2
THF (10ml). The mixture was cooled to 0°C in an ice bath, and a solution of allylmagnesium
bromide in THF 1M (1.1 eq, 5.04 mL) was added dropwise. The solution was warmed to
room temperature and stirred for 2h. The reaction mixture was then quenched by addition of
saturated aqueous NH
4
Cl solution. Extraction with diethyl ether was done three times and the
, then filtered and the solvent
organic layers were collected and dried on anhydrous MgSO
4
was removed. The residue was separated by column chromatography with hexane first and
then a hexane/ ethyl acetate mixture (95/5, v/v). The product was isolated in 92% yield as a
yellow oil. The partial separation of the two diastereoisomers was carried out by HPLC using

a Sunfire Silica chiral column type with a mixture petroleum ether/ diethylether ether (98:2)
as eluent (4b, retention time=13.15 min; 4a, retention time=13.20 min).
1
4
a: NMR H (298K, CDCl
3
, 500.33MHz): 7.12 (s, 2H, C13-H, C17-H); 7.16 (s, 2H, C14-H,
C16-H); 5.76 (m, 1H, C2-H); 5.16 (dd, 2H, C1-H, J = 5Hz, J = 11Hz); 2.99 (m, 1H, C6-H);
2
1
.34 (s, 3H, C18-H); 2.23 (m, 2H, C3-H); 1.97 (m, 1H, C9-H); 1.80 (d, 1H, C5-H, J=11Hz);
.85 (d,, 1H, C5-H, J=5Hz); 1.37 (m, 2H,C8-H); 1.29 (d, 3H, C7-H, J=5Hz); 0.97 (d, 3H,
1
3
1
C11-H, 5Hz); 0.94 (d, 3H, C10-H, J=5Hz). NMR C{ H} (298K, CDCl
44.8(C12); 135.7(C15); 134.4 (C2); 129.4 (C14,C16) ; 127.1 (C13, C17); 118.4 (C1); 75.1
C4); 48.24 (C8); 47.5 (C5); 44.8 (C3); 35.8 (C6); 25.2 (C7); 24.8 (C19); 24.7 (C10); 23.8
3
, 100.613 MHz):
1
(
-1
-1
(
C11) 21.0 (C18). IR  (O-H) 3571 cm , (C=C) 1640 cm . HRMS (DCI-CH ) (m/z) : calcd
4
+
+
for [C18H28O] [M] 260.2140 found 260.2138
2
.4. Catalytic tests
Under dry N atmosphere in a tube schlenk a mixture of PdCl
0.05 mmol), and compound 4a-b (1mmol) was added in 10mL of toluene/dichloromethane
1:1), then it was transferred in 100mL stainless steel autoclave with magnetic stirring. The
2
2
2
L (0.02 mmol), tin chloride
(
(
autoclave was purged three times with carbon monoxyde, then stirred and heated in an oil
bath at 80°C and pressurized under 40 bar of CO for 16h. After 16h, the autoclave was cooled
to room temperature and then slowly depressurized. The crude residue was filtered through a
pad of silica and then analysed by gas chromatography. The lactones were purified by column
chromatography using hexane-ethyl acetate as eluent (v/v, 95/5). Lactones 6, as mixture of the
two diastereomers 6a and 6b, and 5, as a mixture of the four diastereomers, 5a, 5a’, 5b and
5
5
b’, were separated.
1
a, 5a': NMR H (298K, CDCl
3
, 500.33MHz): 7.09 (s, 2H, C12-H, C14-H); 7.10 (s, 2H, C11-
H, C15-H); 2.96 (s, 1H, C8-H); 2.73(s, 1H, C3-H); 2.35 (s, 3H, C16-H); 2.25 (m, 1H, C18-
H);1.95 (d, 1H, C4-H, J=11Hz); 2.01 (d, 1H, C4-H, J=6.5Hz); 1,55(d, 2H, C7-H, J=5Hz);

1
.44 (d, 2H, C17-H, J=5Hz); 1.29 (d, 3H, C9-H, J = 5Hz); 1.15 (d, 3H, C6-H, J=5Hz); 0.93
1
3
1
(
d, 3H, C19-H, J=5Hz); 0.92(d, 3H, C20-H, J=5Hz). NMR C{ H} (298K, CDCl
3
, 100.613
MHz): 179.3(C2); 144.8(C10); 135.9(C13); 129.4 (C12, C14); 126.8 (C11, C15); 87.0 (C5);
4
7.6 (C17); 45.9 (C7); 42.1 (C4); 36.0 (C8); 35 (C3); 29.7 (C18); 24.7 (C9); 24.2 (C20); 24.1
1
(
C19); 21.0 (C16); 15.8 (C6). NMR H (298K, CDCl
3
, 500.33MHz): 7.12 (s, 2H,C12-H, C14-
H); 7.14 (s, 2H, C11-H, C15-H); 2.96 (m, 1H,C8-H); 2.73 (m, 1H,C3-H); 2.35 (s, 3H, C16-
H); 2.25 (m, 1H, C18-H);1.95 (d, 1H, C4-H, J=11Hz); 2.01 (d, 1H, C4-H, J=6.5Hz); 1.55(d,
2
H, C7-H, J=5Hz); 1.44(d, 2H, C17-H, J=5Hz); 1.33 (d, 3H, C9-H, J = 5Hz); 1.18 (d, 3H,
1
3
1
C6-H, J=5Hz); 0.98 (d, 3H, C19-H, J=5Hz); 0.95 (d, 3H, C20-H, J=5Hz). NMR C{ H}
298K, CDCl , 100.613 MHz): 179.2 (C2); 143.81 (C10); 135.8 (C13) ; 129.2 (C12, C14) ;
26.6 (C11, C15); 86.9 (C5); 47.3 (C17); 45.5 (C.7); 41.0 (C4); 35.7 (C8); 34.9 (C3); 29.6
C18); 24.6 (C9); 24.2 (C20); 24.0 (C19); 21.0 (C16); 15.5 (C6). HRMS (DCI-CH ) (m/z) :
(
3
1
(
4
+
+
calcd for [C19
H
28
O + H] [M+H] 289.2168 found 289.2171.
2
1
6
a: NMR H (298K, CDCl
3
, 500.33MHz): 7.11 (s, 2H,C12-H, C14-H); 7.11 (s, 2H, C11-H,
C15-H); 2.85 (m, 1H,C8-H); 2.41 (m, 2H, C3-H); 2.34 (s, 3H, C16-H); 2.06 (d, 2H, C7-H,
J=5.19Hz); 1.80 (m, 1H, C18-H); 1.67 (m, 2H, C4-H); 1.55 (m, 1H, C5-H); 1.45 (m, 1H, C5-
H); 1.35 (d, 3H, C9-H, 5Hz); 0.98 (d, 3H, C19-H, J=5Hz); 0.96 (d, 3H, C20-H, J=5Hz). NMR
1
3
1
C{ H} (298K, CDCl
3
, 100.613 MHz): 171.1 (C2); 144.4 (C10); 135.7 (C13); 129.3 (C12,
C14); 126.8 (C11, C15); 87.2 (C6); 47.1 (C17); 46.9 (C7); 35.6 (C8); 30.9 (C5); 29.2 (C18);
-1
2
5.1 (C9); 24.6 (C20); 23.9 (C20,C19); 20.0 (C16); 16.3 (C4) . IR  (C=O): 1776 cm .
+
+
HRMS (DCI-CH
4
) (m/z) : calcd for [C19
H
28
O + H] [M+H] 289.2168 found 289.2158.
2
1
6
b: NMR H (298K, CDCl
3
, 500.33MHz): 7.11 (s, 2H,C12-H, C14-H); 7.11 (s, 2H, C11-H,
C15-H); 2.85 (m, 1H, C8-H); 2.41 (m, 2H, C3-H); 2.34 (s, 3H, C16-H); 2.06 (d, 2H, C7-H,
J=5.19Hz); 1.80 (m, 1H, C18-H); 1.67 (m, 2H, C4-H); 1.65 (d, 2H, C17-H); 1.55 (m, 1H, C5-

H); 1.45 (m, 2H, C5-H); 1.35 (d, 3H, C9-H, J=5Hz); 0.98 (d, 3H, C19-H, J=5Hz); 0.96 (d,
1
3
1
3
1
7
H, C20-H, J=5Hz). NMR C{ H} (298K, CDCl , 100.613 MHz): 171.3 (C-2); 144.8 (C-
3
0); 135.6 (C-13); 129.2 (C-12, C-14); 126.8 (C-11, C-15); 87.0 (C-6); 47.6 (C-17); 46.9 (C-
); 35.2 (C-8); 30.9 (C-5); 29.7 (C18); 29.3 (C-3); 25.2 (C-9); 25.0 (C-20); 24.4 (C-19); 21.0
+
+
(
C-16); 16.5 (C-4). HRMS (DCI-CH
4
) m/z : calcd for [C19
H
28
O + H] [M+H] 289.2168
2
found 289.2158.
3
. Results and discussion
.1. Chemical modification of α-atlantone 1
To obtain lactones via a cyclocarbonylation catalytic route, α-atlantone 1 was
3
chemically modified in order to synthesise a suitable homallylic alcohol. Heating the Z, E
mixture of the two isomers of α-atlantone 1 to 160 °C in the presence of 5% palladium on
activated carbon (Pd/C) quantitatively produces racemic ketone 2 which was transformed into
alcohol 4 in the presence of allylmagnesium bromide 3 (Scheme 3).
4
is obtained in excellent yield (90%) as two diastereoisomers, 4a and 4b, in a 62:38 ratio
determined by GC) non-separable in the usual conditions of chromatography on silica gel.
.2. Catalytic cyclocarbonylation
Previous studies in our group have shown that [Pd(Cl .2H O catalytic system can
(
3
2
)L
2
]/SnCl
2
2
be efficient in the cyclocarbonylation reaction on monoterpenes, such as isopulegol and
isolimonene [13]. In a first step of experiments, the carbonylation reaction of this new
substrate using [Pd(Cl
80°C, 40 bar of CO, 16h, toluene/DCM (dichloromethane) (1:1, v/v), P/Pd ratio of 2] affords
a mixture of lactones such as the five-membered lactone 5 and the six-membered lactone 6
Scheme 4). The six-membered lactone 6 is obtained as a mixture of two diastereoisomers 6a
2
)(PPh
3
)
2
]/SnCl
2
.2H
2
O and conducted under optimised conditions18
[
(
and 6b while the five-membered lactone 5 is formed as four diastereoisomers 5a,5a’,5b,5b’
due to the creation of a new asymmetric carbon.

It should be noted that high conversion (95%) and selectivity (80%) were obtained. We only
observed a side reaction of dehydration giving a mixture of isomers [19].
As the steric and electronic properties of the phosphine ligands could play an important role in
the selectivity of the cyclocarbonylation reactions [13a,Error! Bookmark not defined.] we
have then investigated the influence of various monophosphine and diphosphine ligands
(
Figure 1).
Cyclocarbonylation using isolated palladium complexes of different monophosphines,
including phosphole ligands, has been first studied. The results are compiled in Table 1.
The comparison of results show that the activities of these catalytic systems are strongly
affected by the nature of the phosphine ligands as the conversion of the substrate ranges
between 0 and 90%. The catalyst activity can be roughly correlated to the value of the
phosphine cone angle. Indeed, ligands with cone angles between 145 and 212° (entries E11-
E17) lead to active catalysts while ligands with smaller (entries E9-E10) and larger (entry
E18) cone angles give less active or inactive catalysts. However, the electronic properties of
the ligands also seem to influence the catalyst activity as phosphines having similar cone
angles but different donating abilities [20,21] (entries E13-E14 and E15-E16) yield catalytic
systems of different efficiencies.
However, all active catalytic systems are chemoselective producing a mixture of lactones 5
and 6 with high selectivities (80-85%) and regioselectivities in favor of lactone 6 with good
diastereoselectivies (6a/6b ratio up to 82/18). It is interesting to note that P(Cy)
3
(entry E16)
or P(Mes) (entry E17) give the best catalysts in terms of chemo-, regio- and
3
diastereoselectivities (up to a 6a/6b ratio of 82/18 starting from a 4a/4b mixture with a 62/38
ratio). It is worth to point out that diastereoisomer 4b reacts with a lower rate than 4a and
accumulates in the remaining alcohol (Table 1).

Table 2 presents the results obtained with isolated palladium complexes of various
diphosphine ligands. In the series of dppe, dppp, dppb ligands, which differ only by the
number of methylene group in the linker between the two PPh group (Figure 1b),
2
conversions increased with the size of the linker from 10% to 49% (entries E19-E21). Similar
trends with dppe and dppp were observed before in the cyclocarbonylation reaction [12a,22].
Interestingly, using dcpb (entry E22), with steric properties similar to the ones of dppb but
different electronic properties, the conversions could be increased to 100% with even better
chemoselectivity in favor of lactones (from 70% to 82%). The regioselectivity with these
diphosphines is, contrary to the case of monophosphine ligand, now in favor of the five-
membered lactone 5. These results prompted us to try other wide bite angle diphosphines,
namely dppf and xantphos. With dppf, the conversion could be increased but not the chemo-
and regioselectivities (entry E23). However, it is interesting to note that the ratio of the two
diastereoisomers 6a and 6b (65/35) is similar to that of the starting alcohol 4a, 4b (62/38).
With xantphos, excellent conversion of 100%, high chemoselectivity (82% for lactones) and
excellent regioselectivity up to 99% in favor of the 5-membered ring lactones 5 is obtained
(
entry E24). We therefore find an excellent catalytic system to efficiently obtain the 5-
membered ring lactone 5.
As mentioned before, the 5-membered lactones 5 are obtained as a mixture of four
diastereomers. The mixture resulting from the reaction with xantphos (entry 24) was analysed
by GC and HPLC but unfortunately the ratio between the different isomers could not be
precisely determined. In addition, the separation of the different products of the reaction
proved to be difficult. Only the five-membered lactones could be separated from the six-
membered lactones by column chromatography.
To get more insight into this cyclocarbonylation reaction, we have investigated the separation
of the two diastereoisomers of alcohol 4. These two diastereoisomers, 4a and 4b, are not

separable in the usual conditions of silica gel chromatography and difficult to separate by
HPLC. By preparative HPLC, a pure sample of 4a can be obtained as a racemic mixture but
4
b was always obtained with significant amounts of 4a (>10%). The cyclocarbonylation
reaction was then carried out with diastereoisomerically pure 4a with the best catalytic
systems, ie those with PCy and xantphos (Tables 1 and 2, E16 and E24).
With PCy catalytic system, alcohol 4a can be converted to a mixture of only 3 lactones: 5a,
a’ and 6a (Table 3, entry E25) when 6 different lactones (5a, 5a’, 5b, 5b’,6a and 6b) are
obtained starting from the 4a/4b mixture . Similarly with xantphos, the reaction yields lactone
as a mixture of only 5a and 5a’, while 5b and 5b’ are also obtained from the 4a/4b mixture.
The catalytic tests have not been performed on alcohol 4b but it must lead to 5b, 5b'and/or
b. This result shows that the racemic diastereoisomer 4a yields only one single 6-membered
3
3
5
5
6
lactone, 6a. During this transformation, no new asymmetric carbon center is formed. We can
suppose that the asymmetric carbons already present in 4a are preserved during the catalytic
reaction yielding only the lactones 6a. Similarly, we can suppose that during the formation of
lactones 5 the asymmetric carbons already present in 4a are left untouched, producing only
two lactones 5a and 5a’ due to the creation of a new asymmetric carbon.
Compound 6a could be isolated from the mixture obtained in E25 conditions but all attempts
to separate 5a and 5a’ failed, even from the mixture obtained in E26. In addition, the ratio of
these 2 diastereoisomers could not be determined either by GC or HPLC in spite of several
attempts.
3
.3. Structural determination
Monocrystals of the six-membered ring compound 6a suitable for X-ray diffraction analysis
were obtained. Similarly, monocrystals were grown from a mixture of 5-membered ring
lactones 5a and 5a’. The two compounds crystallised in a P21/n centrosymmetric space

group, consistent with their racemic nature. A view of the two compounds is shown on
Figures 2 and 3. Bond distances and angles are reported in Supporting Information. The
difference between the two structures is the occurrence of pyran-2-one ring in compound 6a
and furan-2-one ring in compounds 5a/5a’. The geometry within the isobutyl and 2-p-
tolylpropyl substituents is the same within experimental error as reported in Supporting
Information.
The puckering parameters [23,24], θ= 134° and φ= 318.4°, observed for the pyran-2-one ring
in 6a are related to a half-chair conformation. In compound 6a, there is a weak C-H π
interaction between the C4 atom and the symmetry related tolyl ring: C-H = 0.96 Å; H---Cg =
2
.83 Å; C4-H4---Cg = 163°(Cg is the centroid of the tolyl ring). From the molecular structure,
depicted in Figure 2, the configurations could be determined. 6a is the racemic mixture of
6S,8R)-6a and (6R,8S)-6a.
(
Owing to the disorder observed within the furan-2-one ring for compound 5, the
stereochemistry for the atom C5 could be (R) or (S): the two diastereomers 5a and 5a’ are
present in the unit cell. In addition, the two diastereoisomers cocrystallise in a 66/34 ratio, the
major diastereoisomer, called 5a, being the racemic mixture (3R,5S,8R)/(3S,5R,8S). Indeed,
the minor diasteroisomer 5a’ is the (3R,5R,8S)/(3S,5S,8R) mixture. Whatever the disorder,
the conformation of this ring is half-chair with puckering parameter φ= 89.8(5)°. Similar
conformation has been also observed in related compounds [25].
Because the synthesis of 6a from 4a does not change the configuration of the asymmetric
carbons (see above), we can conclude that the homoallylic alcohol 4a is a racemic mixture of
(
1R,6R)-4a and (1S,6S)-4a. Therefore, 4b is a racemic mixture of (1R,6S) and (1S,6R)
isomers (see figure 4). We can also conclude that 6b, the diastereoisomer of 6a, is the racemic
mixture of (6S,8S) and (6R,8R) isomers.

In the synthesis of five-membered lactones 5, an additional asymmetric carbon is created
yielding two different stereoisomers from 4a (5a and 5a’) or 4b (5b and 5b’). As shown
before, both 5a and 5a’ cocrystallised. Indeed, the molecular structure confirms the
configurations of the asymmetric carbons 5 and 8 already present in 4a but the configuration
of new asymmetric carbon 3 could not be determined (see Figure 4).
3
.4 Influence of ligands and substrate on selectivities
Finally, the question arises on the control of the regioselectivity and the stereoselectivity for
the formation of lactones which suggests that different intermediates might be involved in the
catalytic cycle. From the commonly accepted catalytic cycle involving the formation of a Pd-
H key intermediate [13d,26] (Figure 5), the regioselective determining step (step B) occurs
during the insertion of the C-C double bond in the [Pd]-H bond and leads to the lactone 6
from the linear intermediate and lactone 5 from the branched intermediate. Our results show
that the ligand is able to control this regioselectivity since diphosphine ligand such as
xantphos gives mainly lactones 5 while monophosphine ligand such as PCy promotes
3
essentially lactone 6 formation. In addition, it is worth to point out that a control by the
substrate itself is observed since the regioselectivities observed with diasteroisomers 4a and
4
b are significantly different. Indeed lactone 6 is mainly derived from alcohol 4a while the
lactone 5 is mainly derived from alcohol 4b (see table 4).
The substrates 4 could induce asymmetric induction but as the ratio of the four
diastereoisomers of 5 could not be determined at this stage, we cannot validate this
hypothesis.
4
. Conclusions
In this article, we described efficient palladium-based catalytic systems for the
cyclocarbonylation of homoallylic alcohols 4 obtained by straightforward chemical
transformation of natural atlantone extracted from Moroccan Atlas Cedar. High catalytic

activities as well as high chemoselectivities in favour of lactones could be achieved. The
regioselectivity can be easily switched by a judicious choice of ligands in favour of 5
membered ring lactone 5 or 6-membered ring lactone 6. When monophosphines, such as PCy
3
ligand, are used, the major product is the lactone 6. In contrast the use of diphosphines as
auxiliary ligands affords the lactone 5, with almost total regioselectivity when the ligand
xantphos was used. The reversal of regioselectivity by changing monophosphines by
diphosphines could be related to the geometry of the corresponding palladium complexes:
trans relative geometry of the two phosphines in the square-planar complexes for
monophosphines and cis geometry for diphosphines. To the best of our knowledge, this work
is the second example of efficient synthesis and modulated regioselectivity into -lactone or

-lactone starting from the same homoallylic alcohol [27]. Further studies including
theoretical calculations are currently in progress to get more insights in the effect of both
ligands and substrates on the outcome of the catalytic reaction.
Moreover in the general context of the use of biomass, obtention of enantiomerically pure
lactones 5 and 6 is of high interest because they could be valuable intermediates in
agrochemical and pharmaceutical industries. Two approaches involving either the resolution
of racemic alcohols 4a and 4b or asymmetric catalytic carbonylation with chiral phosphines
will be explored.
Acknowledegments
We are grateful to the Institut de Chimie of the CNRS (Centre National de la Recherche
Scientifique) for a Ph.D. fellowship to J. El Karroumi and to Chantal Zedde and Isabelle
Fabing (Service HPLC de l’ICT (FR2599)/ SPCMIB (UMR5068) for HPLC experiments
Supplementary data
1
13
H and C NMR, 2D NMR data for new compounds. Including also X-ray data for
compounds 5a, 5a’ and 6b.
References

1
a) H. Guth, Helv. Chim. Acta. 79 (1996) 1559-1571; b) J. Giacco,. D. L. Capone,. A. E.
Hakansson, H. E. Smyth, G. M. Rilsey, M. A. Sefton, D. K. Taylor, J. Agr. Food Chem. 59
2001) 660-664; c) E. N. Pitsinos, N. Athinaios, V. P. Vidali, Org. Lett. 14 (2012) 4666-4669;
(
d) P. Polaskova, J. Herszage, S. E. Ebeler, Chem. Soc Rev. 37 (2008), 2478-2489; e) F.
Bjeldanes, I. S. Kim, J. Org. Chem. 42 (1977) 2333-2335; f) Y. Kakigi, N. Mochizuki, T.
Icho, T. Hakamatsuka, Y. Goda, Biosci. Biotechnol. Biochem. 74 (2010) 590-594; g) E.
Sarrazin, E. Fréroy, A. Bagnoud, K. Aeberhardt, M. Rubin, J. Agric. Food Chem. 59 (2011)
6
657-6666.
2
For reviews, see: a) G. D. Pandey, K. P. Tiwari, Heterocycles 16 (1981) 449-485; b) C.
Schneider, Angew. Chem. Int. Ed. 41 (2002) 744-746; c) G. Bringmann, S. Tasler, R. M.
Pfeifer, J. Organomet. Chem. 661 (2002) 49-65.
3
4
G. Mehta, S. Lakshminath, Tetrahedron. Lett. 47 (2006) 327-330.
a) L. Wang, K. Thai, M. Gravel, Org. Lett. 4 (2009) 801-893; b) W. Zhao, Z. Li, J. Sun, J.
Am. Chem. Soc. 135 (2013) 4680-4683; c) H. Rudler, C. Alvarez, A. Parlier, E. Perez, B.
Denise, Y. Xu, J. Vaissermann, Tetrahedron. Lett. 45 (2004) 2409-2411; d) H. Rudler, B.
Denise, Y. Xu, A. Parlier, J. Vaissermann, Eur. J. Org. Chem. 17 (2005) 3724-3744; e) E.
Aldeco-Perez, H. Rudler, A. Parlier, C. Alvarez, P. Herson, A. Toscano, Tetrahedron. Lett. 47
(
2006) 9053-9056; f) H. Rudler, A. Parlier, C. Sandoval Chavez, P. Herson, J. C. Daran,
Tetrahedron. Lett. 49 (2008) 5843-5846; g) B. Mao, K. Geurts, M. Fananas-Masrral, A. W.
van Zijl, S. P. Fletcher, B. L. Feringa, Org. Lett. 13 (2011) 948-951; h) A. Parenty, X.
Moreau, G. Niel, J.-M.Campagne, Chem. Rev. 1 (2013) PR1–PR40; i) F. Felluga, C. Forzato,
F. Ghelfi, P. Nitti, G. Pitacco, U. M. Pagnoni, F. Roncaglia, Tetrahedron Asym. 18 (2007)
5
27-536; j) H. Konishi, T. Ueda, K. Manabe, Org. Lett. 18 (2012) 4722-4725; k) I. Shiina,
Chem. Rev. 107 (2007) 239-273; l) K.-N. T.Tseng, J. W. Kampf, N. K. Szymczak,
Organometallics 32 (2013) 2046-2049; m) J. Hierold, D. W. Lupton, Org. Lett. 14 (2012)
3
412-3415.
5
a) C. Zhang, N. Hosoda, M. Asami, Tetrahedron Asymm. 18 (2007) 2185-2189; b) Z. Shen,
H. A. Khan, V. M. Dong, J. Am. Chem. Soc. 130 (2008) 2916-2917; c) N. Riache, A. Blond,
B. Nay, Tetrahedron 64 (2008) 10853-10859; d) S. Xu, Z. Wang, X. Zhang, K. Ding, Eur. J.
Org. Chem. 1, (201) 110-116; e) D. A. Devalankar, P. V. Chouthawaile, A. Sudalai,
Tetrahedron Asym. 23 (2012) 240-244; f) T.L. Liu, W. Li, H. Geng, C.-J. Wang, X. Zhang,
Org. Lett. 15 (2013) 1740-1743; g) S. Murphy, V. M. Dong, J. Am. Chem. Soc. 135 (2013)
5
553-5556.
6
For recent reviews on Baeyer Villiger reaction, see: a) M. Uyanik, K. Ishihara, ACS
Catalysis 3 (2013) 513-520; b) K. Balke, M. Kadow, H. Mallin, S. Saß, U. T. Bornscheuer,
Org. Biomol. Chem. 10 (2012) 6249-6265; c) Z. Zhang, L. P. Parra, M. T. Reetz, Chem. Eur.
J. 18 (2012) 10160-10172.
7
a) A. Chrobok, Tetrahedron 66 (2010) 2940-2943; b) S. Xu, Z. Wang, X. Zhang, K. Ding,
Eur. J. Org. Chem. 1 (2011) 110-116.
8
For reviews, see: B. El Ali, H. Alper, Synlett. 2 (2000) 161-171; B. Gabriele, R. Mancuso,
G. Salerno, Eur. J. Org. Chem. (2012) 6825-6839; X.F. Wu, H. Neumann, M. Beller,
ChemSusChem 6 (2013) 229-249

9
2
5
a) T. Joh, K. Doyama, K. Onitsuka, T. Shiohara, S. Takahashi, Organometallics 10 (1991)
493-2498; b) K. Matsuhita, T. Komori, S. Oi, Y. Inoue, Tetrahedron Lett. 35 (1994) 5889-
890; c) W.-Y. Yu, H. Alper, J. Org. Chem. 62 (1997) 5684-5687; d); B. Gabriele, G.
Salerno, F. De Pascali, M. Costa, G.P. Chiusoli J. Chem. Soc.,Perkin trans 1 (1997)147-149;
e) B. Gabriele, G. Salerno, P. Plastina, M. Costa, A. Crispini, Adv.Synth. Catal. 346 (2004)
3
7
51-358; f) B. Gabriele, R. Mancuso, G. Salerno, P. Plastina, J. Org. Chem. 73 (2008) 756-
59; g) J. Ferguson, F. Zeng, H. Alper, Org. Lett. 14 (2012) 5602-5605.
1
0 a) H. Alper, N. Hamel, J. Chem. Soc., Chem. Commun. 2 (1990) 135-136; b) W.-Y. Yu,
C. Bensimon, H. Alper, Chem. Eur. J. 3 (1997) 417-423; c) M. Brunner, H. Alper, J. Org.
Chem. 62 (1997) 7565-7568; d) P. Cao, X. Zhang, J. Am. Chem. Soc. 121 (1999) 7708-7709.
1
1 (a) H. Alper, D. Leonard, J. Chem. Soc., Chem. Commun. (1985) 511-512; b) B. El Ali, H.
Alper, J. Org. Chem. 56 (1991) 5357-5360.
1
2 a) B. El Ali, K. Okuro, G. Vasapollo, H. Alper, J. Am. Chem. Soc. 118 (1996) 4264-4270;
b) C. Dong, H. Alper, J. Org. Chem. 69 (2004) 5011-5014; c) R. Chanthateyanonth, H. Alper,
Adv. Synth. Catal. 346 (2004) 1375-1384; d) F. Ye, H. Alper, Adv. Synth. Catal. 348 (2006)
1
855-1861.
1
3 a) G. Lenoble, R. Naigre, T. Chenal, M. Urrutigoïty, J.-C. Daran, P. Kalck, Tetrahedron
Asymm. 10 (1999) 929-936; b) G. Lenoble, M. Urrutigoïty, P. Kalck, J. Organomet. Chem.
6
3
43-644 (2002) 12-18; c) G. Lenoble, M. Urrutigoïty, Ph. Kalck, Tetrahedron Lett. 42 (2001)
697-3700; d) D. H. Nguyen, F. Hebrard, J. Duran, A. Polo, M. Urrutigoïty, P. Kalck, Appl.
Organomet. Chem. 19 (2005) 30-34.
1
4 a) A. El Haib, A. Benharref, S. Parrès-Maynadié, E. Manoury, J.-C. Daran, M. Urrutigoïty,
M. Gouygou, Tetrahedron Asymm. 21 (2010) 1272-1277; b) A. El Haib, A. Benharref, E.
Manoury, J.-C. Daran, M. Urrutigoïty, M. Gouygou, Tetrahedron Asymm. 22 (2011) 101-
1
08.
1
5 a) J. G. Smith, Synthesis 8 (1984) 629-656; b) A. S. Rao, S.K. Paknikar, J. G. Kirtane,
Tetrahedron 39 (1983) 2323-2367
1
1
1
6 M. Plattier, P. Teisseire, Recherches (1974) 131-144.
7 J. M. Jenkins, J. C. Verkade, Inorg. Synth. 11 (1968) 108-111.
8 Catalytic parameters, such as CO pressure, temperature, reaction time, solvent, P/Pd ratio,
have been optimised. For lower CO or higher temperature, the selectivity in lactones
decreases due to increased side reactions (dehydratation).
1
9 The dehydrated isomers have been isolated by column chromatography (hexane as eluent)
1
and have been analysed by IR (no OH, no CO bands), H NMR (presence of vinylic protons)
and Mass Spectroscopy.
2
0 For the donating abilities (including both -donor ability and -acceptor strength) see: a)
C. A. Tolman, Chem. Rev. 77 (1977) 313-348 for classical phosphine ligands; b) K. Fourmy,
S. Mallet-Ladeira, O. Dechy-Cabaret, M. Gouygou, Organometallics 32 (2013) 1571−1574
for phosphole ligands.

2
1DBP, PPh
P(p-FC
donating abilities but different cone angles.
3
, TPP and P(p-FC
6
H
4
)
3
have the same cone angle but the donating abilities of
6
H
4
)
3
and PPh are lower than those of TPP et DBP. P(Cy)
3
3
et P(t-Bu) have the same
3
2
2 a) H. Alper, M. Saldana-Maldonado, Organometallics 8 (1989) 1124-1125; b) B. El Ali, H.
Alper J. Org. Chem. 56 (1991) 5357-5360.
2
2
2
3 J. C. A. Boeyens, J. Cryst. Mol. Struct. 8 (1978) 317-320
4 D. Cremer, Acta Cryst. B40 (1984) 498-500.
5 a) M. H. Xu, W. Wang, G. Q. Lin, Org. Lett. 2 (2000) 2229-2232; b) M. H. Xu, W. Wang,
G.-Q. Lin, J. Org. Chem. 66 (2001) 3953-3962; c) W. Wang, Y. Zhong, G.-Q. Lin,
Terahedron Lett. 44 (2003) 4613-4616.
2
4
6 a) D.H. Nguyen, G. Laurenczy, M. Urrutigoïty, Ph. Kalck, Eur. J. Inorg. Chem. 20 (2005)
215-4225; b) D.H. Nguyen, Y. Coppel, M. Urrutigoïty, Ph. Kalck, J. Organomet. Chem. 690
(
2005) 2947-2951; c) R. Naigre, T. Chenal, I. Cipres, Ph. Kalck, J.-C. Daran, J. Vaissermann,
J. Organomet. Chem. 480 (1994) 91-102.
2
7 C. Granito, L. Troisi, L. Ronzini, Heterocycles 63 (2004). 1027-1044.
Figures And Tables:
Scheme 1 Selective cyclocarbonylation of various alkenols.
Scheme 2 Regioselectivity of cyclocarbonylation of various allylic and homoallylic terpenols
Scheme 3 Synthesis of homoallylic alcohol 4.
Scheme 4 Cyclocarbonylation reaction of compound 4. Figure 1 Phosphine ligands used in
this study. Figure 2 Molecular structure of 6a. Only one enantiomer is represented Figure 3
Molecular structure of 5a. Only one enantiomer is represented. Figure 4 Stereochemistry of
compounds 4a, 4b, 5a, 5a’, 5b-b’, 6a, 6b (only one enantiomer is represented for each
compound).
Figure 5. Proposed catalytic cycle.
Table 1: Cyclocarbonylation of 4a, 4b catalysed by [PdCl
2
L
2
/SnCl
2
,2H O],
2
L=monophosphine
Entries Ligand L
a
Cone
Angle [1]
Conversion Selectivity
b
b
(%)
in lactones 5/6 ratio
6a/6b ratio
a,b,c
%
E9
P(n-Bu)
PPh Me
PPh
P(p-FC
DBP
TMP
TPP
3
130
136
145
145
147
150
167
0
/
/
/
nd
E10
E11
E12
E13
E14
E15
2
15
95
91
60
73
61
80
80
85
80
81
82
33/67
33/67
34/66
32/68
31/69
34/66
3
67/33
68/32
68/32
68/32
67/33
6
H
)
4 3

c
c
E16
E17
E18
P(Cy)
P(Mes)
JohnPhos
3
170
212
246
90
85
0
85
82
/
20/80
20/80
/
82/18
75/25
/
3
Conditions : S/C =50, SnCl /C= 2.5, PCO=40 bar, T(°C)=80, t(h)=16, 1 mmol 4a,4b in 10mL
2
a
b
(
toluene/DCM:(1/1)). selectivity: moles of lactones 5 and 6/moles of converted substrate.
determined by GC. the ratio of the four diastereoisomers of the minor lactone 5 could not be
determined due to a difficult separation. in the remaining alcohol, the 4a/4b ratio is 30/70 for
c
c
E16 and 33/67 for E17 starting from an initial 4a/4b ratio of 62/38.
Table 2: Cyclocarbonylation of 4a, 4b catalysed by [PdCl
diphosphine
2
(L
2
)/SnCl
2
,2H
2
O], L
2
=
b
Entry
Ligand L
Bite angle
1]
Conversion (%) Selectivity in
5/6 ratio
a
[
lactones (%)
E19
E20
E21
E22
E23
E24
dppe
dppp
dppb
78
86
98
-
99
110
10
15
49
100
81
-
-
70
82
69
82
-
-
71/29
68/32
62/38
>99/1
c
dcpb
d
dppf
xantphos
100
Conditions: S/C =50, acid/C= 2.5, PCO=40 bar, T=80°C, t(h)=16 Solvent 10mL
a
b
(
toluene/DCM:(1/1). selectivity: moles of lactones 5 and 6/moles of converted substrate
c
d
determined by GC. bite angle not determined. for E23, the 4a/4b ratio is 60/40 in the
remaining alcohol, and the 6a/6b ratio is 65/35.
Table 3 : Cyclocarbonylation of 4a catalysed by [PdCl
2
L
2
/ SnCl
2
,2H
2
O] system
b
a,
Entry Ligand L Conversion
%)
92
xantphos 100
Selectivity in lactones
(%)
85
84
b
b
5/6 ratio
(
E25
PCy
3
19(5a+5a’)/81(6a)
>99(5a+5a’)/1(6a)
E26
Conditions : S/C =50, SnCl
(
2
/C= 2.5, PCO=40 bar, T(°C)=80, t(h)=16 Solvent 10mL
toluene/DCM:(1/1)). selectivity: moles of lactones 5 and 6/moles of converted substrate
determined by GC
a
b
Table 4: Products distribution in the reactions E16 and E17
Products distribution
homoallylic remaining lactones 6
a
Lactones 5
byproducts
alcohol
4a 62%
4b 38%
alcohol
4a 3%
4b 7%
10%
E16 conversion
6a 55,8%
6b 12,2%
68%
5a 0-3,2%
5b 13,8-18,8%
17%
0-5%
0-5%
5%
9
8
0% lactones
5% (5/6=20/80) 100%
(
6a/6b=82/18)
E17 conversion
4a 62%
4b 38%
4a 4,95
4b 10,05
15%
6a 49,2%
6b 16,4%
65,6%
5a 4,85-7,85%
5b 8,55-11,55% 0-3%
16,4% 3%
0-3%
8
8
5% lactones
2% (5/6=20/80) 100
(
6a/6b=75/25)
dehydrated products from 4a or 4b.
a

fx1 .

gr1 .

gr2 .

gr3 .

gr4 .

gr5 .
Scheme 1

OH
O
OH
O
isopulegol
perillyl alcohol
O
[
Pd] / CO
O
OH
geraniol
O
O
Scheme 2
Scheme 3
Scheme 4