Carbonylation of styrenes catalyzed by bioxazoline Pd(II) complexes: Mechanism of enantioselectivity

Article abstract of DOI:10.1039/c1dt10101c

Preliminary studies of the elementary steps involved in the reaction of a chiral methyl carbonyl bioxazoline Pd(II) complex with aromatic olefins and CO have allowed development of a new enantioselective catalytic carbonylation process, leading to γ-ketoester derivatives with high yield and good enantiomeric excess. The intermediate palladacycle complexes have been isolated and characterized by NMR spectroscopy and X-ray diffraction. Factors that govern the stereoselectivity of the olefin carbonylation process are discussed.

Full text of DOI:10.1039/c1dt10101c

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 6792
www.rsc.org/dalton
PAPER
Carbonylation of styrenes catalyzed by bioxazoline Pd(II) complexes:
mechanism of enantioselectivity†
Carla Carfagna,*a Giuseppe Gatti,a Luca Mosca,a Paolo Natanti,a Paola Paoli,b Patrizia Rossi,b
Bartolo Gabrielec and Giuseppe Salernod
Received 18th January 2011, Accepted 31st March 2011
DOI: 10.1039/c1dt10101c
Preliminary studies of the elementary steps involved in the reaction of a chiral methyl carbonyl
bioxazoline Pd(II) complex with aromatic olefins and CO have allowed development of a new
enantioselective catalytic carbonylation process, leading to g-ketoester derivatives with high yield and
good enantiomeric excess. The intermediate palladacycle complexes have been isolated and
characterized by NMR spectroscopy and X-ray diffraction. Factors that govern the stereoselectivity of
the olefin carbonylation process are discussed.
insertion of the comonomers in methyl carbonyl palladium
Introduction
complexes, which proved to be the real catalytic species,9,10b,11a
Insertions of unsaturated molecules into palladium-carbon bonds
have been isolated and characterized. Analogous studies carried
represent the fundamental steps in many catalytic processes.1 In
particular, in the last ten years insertion reactions of olefins into
Pd-acyl bonds and of carbon monoxide into Pd-alkyl bonds have
out with 1,2-substituted olefins and with Pd complexes bearing
achiral a-diimine ligands have allowed some important aspects
concerning the regio- and stereochemistry of the insertion process
to be highlighted.12 With the aim of creating a new way to
selectively synthesize highly substituted chiral carbonyl com-
pounds, we have now turned our attention to the study of the
reactivity of an optically active bioxazoline palladium complex
been extensively studied, since they correspond to the elementary
steps involved in several metal-catalyzed processes including
alkoxycarbonylation1 of alkenes and synthesis of polyketones.2
The starting compound is generally represented by a monocationic
palladium(II) complex containing a chelating ligand and an acyl
moiety.3 Concerning the olefin, in many cases ethylene4 has been
used and, among a-olefins, styrene5 and propylene6 insertions have
been reported. Moreover, while several examples with cyclic olefin
have been described,7 insertions of acyclic 1,2-substituted olefins
are rare.6,8 In this field, our research group has studied copoly-
merization reactions of a-olefins with carbon monoxide and
identified the elementary process steps. In particular, using precat-
-
[Pd(Me)(CO)(BIOX)][BAr¢4 ] 1 (Scheme 1) with aromatic olefins
(namely cis-b-methylstyrene, trans-b-methylstyrene, styrene) and
carbon monoxide. The resulting intermediate complexes have
been shown to be interesting model compounds for the design
of catalytic olefin carbonylation reactions. These investigations
have been undertaken also to evaluate how steric and electronic
properties of the alkenes and the geometry of the ligand can
influence the stereochemistry of the synthesized products.
-
alysts of general formula [Pd(Me)(MeCN)(N-N)][X] (X = PF6 ,
-
B[(CF3)2C6H3]4 ), we have developed catalytic systems for the syn-
thesis, under mild conditions, of copolymers vinylarene/CO hav-
ing a syndiotactic microstructure when N-N = diazabutadienes,9
isotactic if N-N = bioxazolines10 and stereoblock-isotactic when N-
N = aryl a-diimine ortho-disubstituted ligands.11 In all cases, the
first organometallic intermediates deriving from stoichiometric
Results and discussion
Olefin insertion reactions and synthesis of the palladacycle
complexes 2–4
Insertion of the 1,2-substituted olefins into the acylpalladium
bond was achieved by reaction of the methyl carbonyl Pd complex
1 with a 10 : 1 excess of trans-b-methylstyrene (tbms) or cis-b-
methylstyrene (cbms) in dichloromethane at 0 ◦C.
Using an analogous methodology, the styrene insertion was
previously described by us as an intermediate of the copolymer-
ization process.10b The reactions were monitored by H NMR,
following the disappearance of the Pd-CH3 signal of the starting
compound together with the increase of an acetyl peak at around
2.4 ppm. After complete consumption of complex 1, the solvent
was evaporated, and the resulting solids were washed with hexane
aDipartimento di Scienze Biomolecolari, Universita` di Urbino, Piazza
Rinascimento, 6–61029, Urbino, Italy. E-mail: carla.carfagna@uniurb.it;
Fax: (+39) 0722 303311; Tel: (+39) 0722 303312
bDipartimento di Energetica “S. Stecco”, Universita` di Firenze, Via S. Marta,
3–50139, Firenze, Italy
cDipartimento di Scienze Farmaceutiche, Universita` della Calabria, 87036,
Arcavacata di Rende, Cosenza, Italy
1
dDipartimento di Chimica, Universita` della Calabria, 87036, Arcavacata di
Rende, Cosenza, Italy
† CCDC reference numbers 735169 & 735170. For crystallographic data
in CIF or other electronic format see DOI: 10.1039/c1dt10101c
6792 | Dalton Trans., 2011, 40, 6792–6801
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
the cis or trans relation between the phenyl group and the adjacent
methyl is the same as the starting olefin. The observation of NOE
contacts, in 2a and 3a, between the CH(Ph) resonances and the
CH(CH3)2 signals of the bioxazoline isopropyl group reveals that
these two protons are on the same side with respect to the Pd
coordination plane, resulting in an R configuration of the CH(Ph)
carbons. Considering the similarity of the 1H NMR patterns of 2b
with 2a and the values of the vicinal coupling constant CH(Ph)-
CH(CH3), we can reasonably assume that compound 2b is the
trans SS diastereoisomer (Scheme 1). Regarding the insertion
of styrene in complex 1, NOE experiments, carried out on the
resulting cyclopalladate product 4, indicated that also in this case
the regiochemistry is of 2,1 type and that one diastereoisomeric
species 4a, with an R configuration of the new –CHPh– stereogenic
center, is generated.10b However a closer investigation has revealed
also the presence of a small amount (4%) of the diastereoisomer 4b,
having an S configuration, as shown in Scheme 2, this percentage
was determined from the intensity of the signal of CHPh at d 3.85.
Scheme 1
Crystal structure of compounds 2a and 3a
obtaining in high yield products 2–4, which revealed a structure
containing a 5-membered palladacycle (Scheme 1 and 2).
Isomorphous crystals of compounds 2a and 3a are triclinic, space
group P1 (Z = 1). Also, the two metal complexes are roughly
superimposable (Fig. 1) except for the relative arrangement of the
hydrogen and the methyl group bound to C(15), which makes
the absolute configuration of this stereocenter opposite for the
complexes 2a and 3a, being R and S respectively.
Scheme 2 Proposed reaction path for olefin insertion.
The individuation of the spin system CH(Ph)-CH2 or CH(Ph)-
1
CH(CH3) and of an acetyl fragment in the H and 13C-NMR
spectra confirms the olefin insertion. The coordination of the
acetyl oxygen atom to the metal was proved by the presence of
a very deshielded carbonyl signal (at more than 238 ppm) together
with an IR stretching band at around 1600 cm-1.3 Interestingly,
almost exclusive formation of the five-membered palladacycle
complex 3a was observed with cis-b-methylstyrene, while a 4 : 1
mixture of two diastereomeric complexes 2a and 2b was obtained
with trans-b-methylstyrene (Scheme 1 and 2).
Fig. 1 ORTEP3 views of the complex cation of 2a (above) and 3a¢ (below).
The 3a¢¢ drawing has not been shown due to the close similarity with 3a¢.
All atoms are drawn at 50% probability, except hydrogens to which have
been assigned arbitrary thermal parameters.
The regio- and stereochemistry of complexes 2 and 3 were
determined by means of NOE measurements: the experimental
data prove that the process is regioselective and in the palladacycles
As a consequence of the disorder affecting the atoms O(3),
C(13), C(14), C(15), and C(16) in 3a, two cationic species (indi-
cated with ¢ and ¢¢ in the following) are present in the crystal lattice
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6792–6801 | 6793
©

View Article Online
◦
Table 2 Free energy differences (kcal mol-1) and molar fractions of
reaction species
˚
Table 1 Selected bond lengths (A) and angles ( ) for complex cations 2a
and 3a
2a
3a
Molar fractions xa
Species
G◦ - G◦
Calc.
Obs.
G◦b- G◦
¯
G‡ - G‡
a
a
b
b
a
a¯
b
a
Pd(1)–N(1)
Pd(1)–N(2)
Pd(1)–O(3)
2.05(1)
2.19(1)
2.069(9)
Pd(1)–N(1)
2.06(1)
2.17(1)
1.95(2)
2.15(2)
2.05(2)
77.2(5)
168.9(8)
170.9(7)
99.9(7)
102.0(7)
97.2(7)
176.9(8)
80.6(8)
85.8(8)
Pd(1)–N(2)
Pd(1)–O(3¢)a
2
3
4
1.81
1.24
2.10
0.95
0.89
0.97
0.80
0.993
0.96
-0.74
-0.32
-0.33
0.82
2.93
1.88
Pd(1)–O(3¢¢)
Pd(1)–C(17)
N(1)–Pd(1)–N(2)
N(1)–Pd(1)–O(3)
2.04(1)
78.0(5)
175.9(6)
Pd(1)–C(17)
N(1)–Pd(1)–N(2)
N(1)–Pd(1)–O(3¢)
N(1)–Pd(1)–O(3¢¢)
N(1)–Pd(1)–C(17)
N(2)–Pd(1)–O(3¢)
N(2)–Pd(1)–O(3¢¢)
N(2)–Pd(1)–C(17)
O(3¢)–Pd(1)–C(17)
O(3¢¢)–Pd(1)–C(17)
a Calculated from DFT electronic energies including the Gibbs energy cor-
rection at 298 K. LANL2DZ basis set for the palladium ion. b Calculated
with eqn (1), using the observed molar fractions.
N(1)–Pd(1)–C(17)
N(2)–Pd(1)–O(3)
99.6(5)
99.8(4)
(where iPr2dab = 1,4-diisopropyl-1,4-diazabuta-1,3-diene and
Ar¢ = 3,5-(CF3)2C6H3), which features the same palladacycle, the
value of t3 corresponds to a -anti-periplanar conformation, while
in the (cis-[Pd(CH(Ph)CH(Me)C(O)Me(iPr2dab)]+[BAr¢4]- com-
plex a -syn-clinal conformation was found.12 More comparable
t1 and t2 values are observed within each trans and cis couple of
metal complexes. Finally, there are no significant intermolecular
interactions in the crystal lattice.
N(2)–Pd(1)–C(17)
O(3)–Pd(1)–C(17)
175.2(5)
82.8(5)
a Due to disorder in the crystal lattice of 3a, two positions, labelled with ¢
and ¢¢, were assigned to atoms O(3), C(13), C(14), C(15) and C(16).
of this compound, which only slightly differ in the conformation
of the acetyl palladacycle ring (vide infra). However, it must be
mentioned here that despite this disorder the S stereochemistry
of the carbon atom C(15) has been undoubtedly established (see
Experimental section for further details).
Given that the bioxazoline (BIOX) ligand has a SS configu-
ration and the stereochemistry of C(17) is R in both 2a and
3a complexes, the latter are diastereoisomers, with the adjacent
methyl and phenyl groups trans disposed in 2a and cis oriented
in 3a. Bond distances and angles of BIOX, which lies almost in a
plane except for the iso-propyl groups (the highest deviation being
the one of N(1) and C(5) in 2a and 3a, respectively), are within the
expected range. In both the complex cations, the conformation of
the iso-propyl groupings of the BIOX ligand with respect to the
C(3)–C(7) and C(6)–C(10) bonds is gauche (ª-60◦) with the less
bulky iPr hydrogen atom facing in all cases the 5-membered acetyl
palladacycle.
Selectivity of the olefin insertion
Concerning substituents configuration of the obtained pallada-
cycles and the regio- and stereoselectivity of their formation, the
above reported experimental finding can be rationalized as follows.
i. The conservation of the trans or cis geometry of substituents in
compounds 2 and 3 can be explained with a concerted syn addition
of the Pd-acetyl fragment to the olefin double bond, through a
four-membered transition state (Scheme 2), in agreement with a
generally accepted mechanism.12
ii. The high regioselectively observed in the insertion of styrene,
tbms and cbms into the chiral bioxazoline complex 1 can be
attributed to the steric effect of the phenyl group, which directs the
acetyl group to bind to the less hindered carbon of the alkene.5a,6,15
iii. Concerning stereoselectivity, it is evident that in each case the
molar fractions of isomers of a type (80 to 99%) are much larger
than those of isomers b (Scheme 2). This fact could be attributed in
first order approximation to the larger thermodynamic stability of
palladacycles a over b. Actually, free energy differences G◦ - G◦
Concerning the metal coordination environment, in both com-
plexes, the palladium ion is tetra-coordinated in a square planar
geometry, the greatest displacement of the metal ion from the mean
˚
plane defined by the four donors is 0.1013(4) A in complex 3a¢. The
b
a
palladium-donor atom distances (Table 1) are in agreement with
those found in analogous complexes retrieved in the Cambridge
Structural Database (CSD), V. 5.31,13 with the Pd(1)–N(1) distance
shorter than the Pd(1)–N(2) one in both complexes as a result of
the larger trans effect due to the (Ph)C moiety compared to the O-
donor group. The moiety O(3)–C(13)–C(15)–C(17) is not planar
as proved by the torsion about C(13)–C(15) (t1 = -25(2), 13(5) and
-33(4)◦ for 2a, 3a¢ and 3a¢¢, respectively).
The orientation of the phenyl ring as defined by the value of
the dihedral angle t2 (t2 = C(15)–C(17)–C(18)–C(19)) is in all
cases comparable within 3s (t2 = -65(2)◦in 2a, t2 = -80(3)◦ in
3a¢ and t2 = -62(3)◦ in 3a¢¢). Moreover the relative orientation
of the phenyl and methyl groups, trans in 2a and cis in 3a,
estimated by density functional calculations (see computational
details in the Experimental section) and reported in Table 2, are
qualitatively in agreement with this suggestion. It appears that
these values are largely due to the unfavourable steric interaction
between phenyl and isopropyl groups in isomers b, while in isomers
a these two groups are always far apart. Nevertheless, small but
significant discrepancies between thermodynamically calculated
and observed molar fractions (see Table 2) still remain to be
explained, so that a consideration of the reaction path illustrated
in Scheme 2 is necessary.
A simplistic hypothesis could be that stereoselectivity is con-
¯
trolled by the relative stability of the intermediates a¯ and b in
which the Re or Si face of the olefin is coordinated respectively.
can be quantified by the value of the dihedral angle t3 (t3
=
However, a DFT estimate of the differences in the free energy of
C(16)–C(15)–C(17)–C(18)): -78(2)◦ in 2a, -19(4) and 22(4)◦
in 3a¢ and 3a¢¢, respectively. Thus, while in 2a the conforma-
tion about t3 can be classified as -syn-clinal, in 3a is syn-
periplanar.14 In this respect, it is noteworthy that in the analogous
complex (trans-[Pd(CH(Ph)CH(Me)C(O)Me(iPr2dab)]+[BAr¢4]-
these intermediates G◦b- G◦ revealed that they do not correlate at
¯
a¯
all with the observed concentrations of the final reaction products:
¯
for example, intermediate 2a¯ is less stable than 2b, while leading
to predominant product 2a. Apparently, in any of the examined
cases the less stable intermediates are more reactive, and this
6794 | Dalton Trans., 2011, 40, 6792–6801
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
suggests that the transition states for the conversion of a¯ forms
Similar conclusions were reached recently to rationalize the
different reactivity observed in the polymerization of (E)- and
(Z)-2-butene isomers with (a-diimine)Ni(II) complexes, known as
“Brookhart catalysts”.16
must have a considerably lower energy than the transition states
¯
for the conversion of forms b. The difference between these two
energy values could be simply estimated in the following way. The
reaction mechanism is assumed to be a rapid interconversion of
¯
intermediates a¯ and b followed by their change to products a and
Reaction of complexes 2 and 3 with CO and methanolysis of the
resulting products
b via the respective transition states a‡ and b‡ (Scheme 3).
With the aim of checking whether the 2 and 3 complexes can
be intermediates for the synthesis of highly substituted chiral
carbonyl compounds, their reactivity with carbon monoxide was
investigated. Bubbling CO in a CDCl3 solution of 3a or 2 resulted
in the formation of complex 5 or 7, respectively; the open chain
structure of the latter is evidenced by three CO chemical shifts
observed in the 13C NMR spectra: one at around 170 ppm, due
to the CO coordinated to Pd, and the other two at more than
200 ppm due to the acyl CO of the chain. This assignment is
further confirmed by the presence in the IR spectrum of three
bands, one of which (2130 cm-1) is typical of a CO ligand bonded
to a metal center. With regard to the stereochemistry, the number
of signals observed in the 13C NMR spectrum of compound
5 indicates the presence of just one diastereoisomer and the
successive methanolysis provides the 2,3-disubstituted g-ketoester
6 in only one diastereomeric form; for the two chiral centres
an RR configuration is predicted, corresponding to a retention
of configuration with respect to complex 3a and 5. Instead, in
the case of the mixture of palladacycle complexes 2a and 2b,
after reaction with CO, compound 7 is obtained as a mixture
of two diastereoisomers (7a,7b) (Scheme 1). Unfortunately, due to
Scheme 3
According to this Curtin–Hammett-type scheme, the ratio
between molar fractions xb and xa is given by [eqn (1)]:
xb
G‡b −G‡ = −RT ln
(1)
a
1
xa
free carbon monoxide in solution, the H NMR signals of 7 are
broad and the less abundant diastereoisomer 7b (corresponding
to CO insertion in 2b) is not clearly detectable; the presence
of a shoulder in the terminal –COCH3 signal is however an
evidence of the formation of this diastereoisomer. Methanolysis
of complex 7 leads mainly to the formation of g-ketoester 8 (60%)
together with the two epimeric lactones 9 and 10 (40%). The
concomitant formation of compounds 9 and 10, probably due
to steric reasons, should take place through a non-stereoselective
nucleophilic attack of methanol on the terminal acetyl carbon
of the acylpalladium complex 7, with consequent cyclization
and formation of highly substituted chiral lactones.17 Also, the
palladacycle deriving from styrene reacts with CO to give the
open chain structure complex having, nearly exclusively, an SSR
configuration as its precursor 4.10b
This relationship shows that the relative amounts of products
a and b do not depend on the energy difference between the
intermediates ground states G◦
¯
- G◦ , but only on the free
b
a¯
energy difference between the transition states G‡ - G‡ . It is
b
a
just this energy difference that governs the stereoselectivity of the
process. The calculated values, reported in Table 2, show that
the energy of the transition state for the production of b-type
stereoisomers is in any case higher than for a-type isomers, in a
range of about 1–3 kcal mol-1, the upper value corresponding to
high stereoselectivity. Now, there is one last question concerning
the structure-selectivity relationship. Apparently, the free energy
differences of the transition states depend mainly on the presence
i
of steric interactions between the Ph of the olefin and the Pr
group of the bioxazoline ligand and the methyl of olefin and
the formed acetyl group. Indeed, considering a TS as depicted
in Scheme 2, in the case of the mono-substituted olefin styrene,
the Ph-iPr interaction destabilizes species 4b‡ relative to 4a‡
by 1.88 kcal mol-1 (Table 2). On the other hand, for the di-
substituted olefins, it turns out that the steric effects depend on
configuration of substituents. In the case of the trans isomer,
the Ph-iPr interaction destabilizes the species 2b‡, while the Me-
Acetyl interaction destabilizes the species 2a‡, resulting in a net
difference of 1.88–1.05 = 0.82 kcal mol-1, corresponding to a
modest stereoselectivity. In the case of the cis isomer the sum of the
two interactions Ph-iPr + Me-Acyl = 1.88 + 1.05 = 2.93 kcal mol-1
destabilizes the species 3b‡, resulting in very high selectivity.
In summary, the above proposed mechanism suggests that the
ultimate source of stereoselectivity is the steric interaction of the
catalytic site with the olefin substituents in the transition states.
Catalytic synthesis of c-ketoesters 6,8,11
On the basis of the acquired knowledge on the stoichiometric
reactive sequences above described, we have devised a new
enantioselective synthesis of g-ketoesters, according to eqn (2):
(2)
For the realization of catalytic alkoxycarbonylation reactions
of the aromatic olefins, we used enantiopure Pd(II) complex
[Pd(OTf)2(BIOX)] as catalyst, methanol as reagent and solvent,
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6792–6801 | 6795
©

View Article Online
Table 3 Catalytic synthesis of g-ketoestersa
Entry
Catalyst
Olefin
T/◦C
Yieldb (%)
TON
eed (%)
1
2
3
4
5
6
7
[Pd(OTf)2(BIOX)]
[Pd(OTf)2(BIOX)]
[Pd(OTf)2(BzBOX)]
[Pd(Me)(OTf)(BIOX)]
[Pd(OTf)2(BIOX)]
[Pd(OTf)2(BIOX)]
[Pd(OTf)2(BIOX)]
cbms
18
0
18
20
18
50
18
87
74
94
72
88
55
68c
174
148
188
144
176
110
136
85
83
87
85
90
10
—
cbms
cbms
cbms
styrene
styrene
tbms
a Reaction conditions: n Pd = 0.045 mmol, olefin = 9 mmol, (olefin/Pd cat = 200/1), MeOTf V = 0.38 mL (9 mmol); solvent MeOH V = 2 mL; TFE V =
30 mL (0.3 mmol); PCO = 3 atm. b Isolated yield, based on starting olefin. c Total yield. d Evaluated by 1H NMR with shift reagent.
and MeOTf as methylation agent. Small amounts of TFE (TFE =
2,2,2-trifluoroethanol) were added to stabilize the active species.18
The reaction was conducted in autoclave for 5 days under very
mild conditions (T = 18 ◦C, PCO = 3 atm); products were easily
purified from the reaction mixture since no by-products were
formed.
The nature and the composition of obtained compounds
reflected closely what was observed for the stoichiometric model
reactions. In fact, with cis-b-methylstyrene and with styrene the
process resulted to be perfectly chemo- and regioselective leading
to the exclusive formation of the disubstituted b-ketoesters 6
and 11 respectively, in good yield; while, in the case of trans-b-
methylstyrene the reaction was less selective, producing the linear
g-ketoester 8 (41%) together with the two epimeric lactones 9 and
10 (27% total yield) (Scheme 4, Table 3).
Accordingly, products deriving from b-hydrogen elimination were
detected in the stoichiometric reactions run at 40–50 ◦C.
In order to evaluate whether a different steric arrangement on
the metal centre could have an effect, we performed a reaction
with cis-b-methylstyrene using the catalyst [Pd(OTf)2(BzBOX)]
were BzBOX = (S)-4,4¢-bis(phenylmethyl)-4,4¢,5,5¢-tetrahydro-
2,2¢-bioxazole and a slight increase in yield and ee was ob-
served (entry 3 vs. 1). The use of the preformed complex
[Pd(Me)(OTf)(BIOX)] resulted in a decrease in productivity (entry
4) while the ee (85%) was the same as found in entry 1 with catalyst
[Pd(OTf)2(BIOX)].
Regarding the catalytic cycle, we presume that, starting from
a catalyst of the type [Pd(Me)(OTf)(BIOX)], the olefin and CO
insertions should proceed easily, similarly to the stoichiometric
reactions, producing the palladacycle C and then the open-chain
acyl palladium complex complex D (Scheme 5). Alcoholysis of
the latter is expected to generate, besides the g-ketoester, also a
palladium hydride species that evolves to give a Pd(0) complex,
which will be reoxidized by methyl triflate restoring catalyst A.
Scheme 4
Regarding the enantioselectivity, with styrene an ee of 90% was
achieved (entry 5). This value is in agreement with the de (92%)
found for the intermediate palladacycle complex 4 (Scheme 2).
Instead in the case of cis-b-methylstyrene the ee value (85%),
obtained in the catalytic reaction (entry 1, 3, 4), turned out to be
lower than the de value (98.6%) of the intermediate palladacycle
complex 3, isolated in the stoichiometric reaction. This discrep-
ancy suggests that a counterion effect is present; evidently the
olefin enantioface discrimination is partially influenced also by the
nature of the counterion and hence by the interactions between
the anion and the catalyst metal centre.19
Scheme 5
Although 1,4-dicarbonyl compounds and in particular g-
ketoesters are useful building blocks in organic synthesis and in the
production of pharmaceutical compounds,20 to our knowledge a
direct catalytic synthesis of chiral g-ketoesters, promoted by metal
complexes, is unprecedented. Indeed, so far the synthesis of g-
ketoesters was achieved only through several steps, using a variety
of different reaction pathways5a,21 and only in very few particular
cases the obtained products are optically active.22
◦
The best temperature for the process resulted to be ca. 20 C;
when running the reaction at 0 ◦C a slight decrease in productivity
was observed (entry 2 vs. 1). Moreover, at a temperature of 50 ◦C a
sharp decline in yields and a complete loss of enantioselectivity was
registered (entry 6 vs. 5), probably due to a quicker decomposition
of the catalyst and/or to a fast process of b-hydrogen elimination
involving the intermediate five-membered palladacycle complex.
Experimental
General
All manipulations were carried out under a nitrogen atmo-
sphere by using Schlenk techniques. Solvents were dried by
6796 | Dalton Trans., 2011, 40, 6792–6801
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
standard methods and freshly distilled under nitrogen. cis-
b-Methylstyrene (TCI Europe), styrene (Aldrich) and trans-
b-methylstyrene (Aldrich) were distilled before use. Carbon
monoxide (Cp grade 99.99%) was supplied by Air Liquide. CP
grade chemicals were used as received unless otherwise stated.
Chloroform-d and methylene chloride-d2 were degassed and
was washed with hexane (2 ¥ 3 ml) and dried under vacuum to
yield almost exclusively the palladacycle 3a(RS) together a trace
(0.7%) of the diastereoisomer 3b(SR) (Scheme 2) as yellow powder
(114 mg, 0.0844 mmol, 94%).
IR (liquid, CDCl3): 1618 (CO-CH3), 1640, 1612 (C N) cm-1
3a: 1H NMR (CDCl3, 293 K): d 7.33 (3H, m, H10), 7.19 (3H, m,
H11) 7.03–6.99 (2H, m, H9), 4.71–4.42 (4H, m, H1 and H1¢), 4.21–
4.13, 3.77–3.69 (1H each, m, H2 and H2¢), 4.16 (1H, d, J = 6.8 Hz,
H7), 3.15 (1H, dq, J = 6.8 and 7.2 Hz, H12), 2.41 (3H, s, H15),
2.20–2.04 (2H, m, H4 and H4¢), 1.12 (3H, d, J = 7.2 Hz, H13),
0.96, 0.94, 0.92, 0.85 (3H each, d, J = 7.0, H5–H6 and H5¢-H6¢).
˚
stored over 3 A molecular sieves.
Complex [Pd(CH3)(CO)(BIOX)]+[BAr¢4]- (1) where BIOX =
(4S,4¢S)-(-)-4,4¢,5,5¢-tetrahydro-4,4¢-bis(1-methylethyl)-2,2¢-bio-
xazole was synthesized as previously reported by us.10b
Catalysts [Pd(OTf)2(BIOX)] and [Pd(OTf)2(BzBOX)] were
1
BzBOX
=
(S)-4,4¢-Bis(phenylmethyl)-4,4¢,5,5¢-tetrahydro-2,2¢-
13C{ H}NMR (CDCl3, 253 K): d 238.6 (s, C14), 159.7, 157.5 (C3
bioxazole were prepared by a two step procedure, based on
the synthesis of the corresponding neutral dichloride derivatives
followed by the usual dehalogenation using silver trifluoromethane
sulfonate.19,23 Compounds [PdCl2(BIOX)],24 [PdCl2(BzBOX)]21
and NaBAr¢4 (Ar¢ = 3,5-(CF3)2C6H3)25 were synthesized according
to the literature. Complex [Pd(Me)(OTf)(BIOX)] was obtained
from [Pd(Me)(Cl)(BIOX)]10b by adding silver trifluoromethane
sulfonate.
and C3¢), 141.2 (C8), 129.9 (C9), 126.9 (C10), 126.9 (C11), 74.5,
72.9 (C1 and C1¢), 69.6, 66.2 (C2 and C2¢), 59.4 (C12), 47.0 (C7),
30.9, 30.1 (C4 and C4¢), 28.3 (C15), 18.5, 18.3, 16.4, 13.6 (C5–C6
and C5¢-C6¢), 14.6 (C13).
3b compound is present only in trace 0,7%, this percentage was
determined from the intensity of the signal of H12 at d 2.95.
Anal. calcd for C55H45F24N2O3BPd (1355.16): C, 48.75; H, 3.35;
N, 2.07%. Found: C, 48.45; H, 2.90; N, 2.18%.
Elemental analyses (C, H, N) were carried out with a Fisons
Instruments 1108 CHNS-O Elemental Analyser. Infrared spectra
were measured in the range 4000–600 cm-1 on a Nicolet FT-
IR Avatar 360 spectrometer. NMR spectra were measured in
CD2Cl2 or CDCl3 on a Bruker Advance 200 spectrometer with
a multinuclear 5 mm probehead. 1H and 13C NMR chemical shifts
are relative to TMS and were measured using the residual proton
or carbon resonance of the deuterated solvents. The assignment
(R,R)-[Pd(C(O)CH(Ph)CH(CH3)C(O)CH3)(CO)(BIOX)]+
[BAr¢4]- (5)
1
1
of the H and 13C resonances was performed by using the H-
COSY, 13C-DEPT, 1H13C-HSQC, 1H13C-HMBC and 1H-NOESY
experiments.
Bubbling of CO at -30 ◦C for 3 min into a CDCl3 solution (0.6 ml)
of 3 (50 mg, 0.037 mmol) resulted in the formation of the open
chain complexes 5. This complex was stable only in solution and
for a few hours, which precludes elemental analysis.
IR (liquid, CDCl3): 2127 (C O), 1750, 1713 (C O), 1636,
1611 (C N) cm-1.
Synthesis and characterization of complexes 2, 3, 5, 7
In compounds 2, 3, 5, 7 the counterion [BAr¢4]- tetrakis[3,5-
bis(trifluoromethyl)phenyl]borate(1-) gives a pattern of NMR
signals with the following typical chemical shifts:
1H NMR (CDCl3, 293 K): d 7.71 (8H, s, Ar¢-Ho), 7.54 (4H, s,
Ar¢-Hp). 13C NMR (CDCl3, 293 K): d 161.7 (q, 1J(C,B) = 49.3 Hz,
1H NMR (CDCl3, 293 K): d 7.45–7.38 (3H, m, H10 and H11),
7.27–7.19 (2H, m, H9), 4.69–4.43 (4H, m br, H1 and H1¢), 4.47
(1H, d, J = 6.9 Hz, H7), 4.28–4.13 (2H, m br, H2 and H2¢), 3.25
(1H, dq, J = 6.9 and 7.1 Hz, H12), 2.07 (3H, s, H15), 1.94–1.78 (2
H, m br, H4 and H4¢), 1.31 (3H, d, J = 7.1 Hz, H13), 0.85, 0.78
2
Ar¢-Ci), 134.8 (s, Ar¢-Co), 128.8 (q, J(C,F) = 31.2 Hz; Ar¢-Cm),
124.6 (q, 1J(C,F) = 270.8 Hz, CF3), 117.5 (s, Ar¢-Cp).
cis-(R,S)-[Pd(CH(Ph)CH(CH3)C(O)CH3)(BIOX)]+[BAr¢4]- (3a)
1
(6H each, d, J = 6.8 Hz, H5–H6 and H5¢–H6¢). 13C{ H} NMR
(CDCl3, 253 K): d 210.3 (C16), 207.0 (C14), 171.3 (C17), 161.5,
158.7 (C3 and C3¢), 131.5 (C8), 130.5, 130.2, 130.1 (C9–C10 and
C11), 74.4, 73.7 (C1 and C1¢), 69.1, 68.1 (C2 and C2¢), 68.9 (C7),
50.0 (C12), 30.9, 29.3 (C4 and C4¢), 30.3 (C15), 18.3 15.7, 13.4
(C5–C6 and C5¢-C6¢), 15.4 (s, C13).
Stoichiometric methoxycarbonylation of complex 3a leading to the
synthesis of compound 6
cis-b-Methylstyrene (125.83 mL, 0.97 mmol) was added to a
dichloromethane solution (3 mL) of compound 1 (120 mg,
0.097 mmol) cooled to -25 ◦C. The solution was warmed to 0 ◦C
and allowed reaction of until the starting Pd complex completely
disappeared (1 h). Then the solution was filtered through Celite
and the solvent was evaporated in vacuum. The resulting solid
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6792–6801 | 6797
©

View Article Online
Complex 3a (90 mg, 0.066 mmol) was dissolved in a mixture
of dichloromethane (4 mL) and methanol (1 ml) and CO was
bubbled, at -30 ◦C for 5 min. The reaction was stirred for
Elemental analysis of the 2a and 2b mixture: anal. calcd for
C55H45F24N2O3BPd (1355.16): C, 48.75; H, 3.35; N, 2.07%. Found:
C, 48.48; H, 2.91; N, 2.13%.
◦
24 h at 4 C, then the solvents were evaporated in vacuum. The
(R,S)(S,R)-[Pd(C(O)CH(Ph)CH(CH3)C(O)CH3)(CO)(BIOX)]+
[BAr¢4]- (7)
residue was purified by column chromatography on silica gel with
hexane/ethyl acetate = 6/4 as eluent, giving compound 6 as yellow
oil (15.01 mg, 0.064 mmol, 97%).
(2R,3R)-Methyl 3-methyl-4-oxo-2-phenylpentanoate (6). 1H
NMR (CDCl3, 293 K): d 7.46–7.27 (5H, m, H5–H6 and H7),
3.87 (1H, d, J = 10.9, H3), 3.68 (3H, s, H1), 3.34 (1H, dq, J = 6.8
and 10.9 Hz, H8), 2.02 (3H, s, H11), 1.29 (3H, d, J = 6.8 Hz, H9).
1
13C{ H} NMR (CDCl3, 293 K): d 210.5 (s, C10), 173.1 (s, C2),
136.9 (s, C4), 128.7, 128.3 (s, C5 and C6), 127.7 (s, C7), 54.1 (s,
C1), 52.1 (s, C3), 50.0 (s, C8), 30.0 (s, C11), 15.8 (s, C9).
Anal. calcd for C13H16O3 (220.11): C, 70.89; H, 7.32%. Found:
C, 70.88; H, 7.31%.
Bubbling of CO at -30 ◦C for 3 min into a CDCl3 solution (0.6 ml)
of 2a and 2b (50 mg, 0.037 mmol) resulted in the formation of
the open chain complex 7 as a mixture of two diastereoisomers
(7a, 7b). These complexes are stable only in solution and for a
few hours, which precludes elemental analysis.
trans-(R,R)-[Pd(CH(Ph)CH(CH3)C(O)CH3)(BIOX)]+[BAr¢4]-
(2a) and trans-(S,S)-[Pd(CH(Ph)CH(CH3)C(O)CH3)(BIOX)]+
[BAr¢4]- (2b)
IR (liquid, CDCl3): 2130 (C O), 1757, 1716 (C O), 1639,
1610 (C N) cm-1.
1H NMR (CDCl3, 293 K): d 7.48–7.25 (5H, m br, H9–H10 and
H11), 4.78–4.10 (6H, m br, H1–H1¢ and H2–H2¢), 4.47 (1H, d,
J = 10.5 Hz, H7), 3,44 (1H, dq, J = 10.5 and 7.2 Hz, H12), 2.28
(3H, s, H15), 1.86–1.68 (2H, m br, H4 and H4¢), 0.90 (3H, d, J =
7.2 Hz, H13), 0.86, 0.77 (6H each, 2d, J = 6.7 Hz, H5–H6 and
1
H5¢–H6¢). 13C{ H} NMR (CDCl3, 253 K): d 212.4, (C16), 209.1
(C14), 171.2 (C17), 161.2, 159.0 (C3 and C3¢), 131.2 (C8), 130.6
(C9), 130.2 (C10), 129.9 (C11), 74.3, 74.1 (C1 and C1¢), 70.0 (C7),
69.1, 68.0 (C2 and C2¢), 49.4 (C12), 30.9, 29.1 (C4 and C4¢), 29.5
(C15), 18.2 15.8, 13.5 (C5–C6 and C5¢-C6¢), 15.6 (C13).
trans-b-Methylstyrene (125.83 mL, 0.97 mmol) was added to
a dichloromethane solution (3 mL) of compound 1 (120 mg,
0.097 mmol) cooled to -25 ◦C. The solution was warmed to 0 ◦C
and allowed reaction until the starting Pd complex completely
disappeared (3 h). Then the solution was filtered through Celite
and the solvent was evaporated in vacuum. The resulting solid
was washed with hexane (2 ¥ 3 ml) and dried under vacuum. A
120 mg (0.088 mmol, 91%) sample of compound 2 was collected as
a yellow solid which was shown to be a diastereoisomeric mixture
of 2a/2b (4/1 ratio).
Stoichiometric methoxycarbonylation of complexes 2a/2b leding
to the synthesis of compounds 8, 9 and 10
IR (liquid, CDCl3): 1612 (CO–CH3), 1636, 1612 (C N) cm-1.
2a: 1H NMR (CDCl3, 293 K): d 7.45–7.42 (2H, m, H9), 7.33–
7.29 (1H, m, H11), 7.24–7.20 (2H, m, H10), 4.68 (1H, dd, J = 5.6
and J = 6.4 Hz, H1¢), 4.53 (1H, dd, J = 4.8 and J = 5.6 Hz, H1¢),
4.44 (1H, dd, J = 4.8 and J = 9.6 Hz, H1), 4.21 (1H, dd, J = 10.0
and J = 9.6 Hz, H1), 4.28–4.23 (1H, m, H2¢) 2.15–2.10 (1H, m,
H2), 3.68 (1H, d, J = 7.6 Hz, H7), 2.99 (1H, dq, J = 7.6 and 7.2 Hz,
H12), 2.36 (3H, s, H15), 2.10–2.02 (1H, m, H4), 1.69–1.61 (1H,
m, H4¢), 1.17 (1H, d, J = 7.2 Hz, H13), 0.71, 0.62 (3H each, d, J =
7.0 Hz, H5 and H6), 0.94, 0.93, (3H each, d, J = 7.0 Hz, H5¢ and
The 2a/2b mixture (100 mg, 0.073 mmol) was dissolved in a
solution of dichloromethane (4 mL) and methanol (0.5 ml) and
CO was bubbled, at -20 ◦C for 5 min. The reaction was stirred for
24 h at 4◦ C, formation of a black Pd precipitate was observed then
solvents were evaporated in vacuum. The residue was purified by
column chromatography on silica gel with hexane/ethyl acetate =
6/4 as eluent, giving a yellow oil containing a mixture of 8 (60%),
9 and 10 (40%) (16.99 mg, 0.072 mmol, 98%).
1
H6¢). 13C{ H}NMR (CDCl3, 253 K): d 238.1 (C14), 157.92 (C3),
159.2 (C3¢), 145.1 (C8), 129.2, (C10), 127.9 (C9), 127.3 (C11), 74.3,
72.3 (C1 and C1¢), 69.9, 64.8 (C2 and C2¢), 62.9 (C12), 48.7 (C7),
30.6, 29.4 (C4 and C4¢), 27.5 (C15), 18.0, 17.5, 16.2, 13.3 (C5–C6
and C5¢-C6¢), 14.4 (C13).
(2RS,3SR) Methyl 3-methyl-4-oxo-2-phenylpentanoate (8). 1H
NMR (CDCl3, 293 K): d 7.39–7.18 (5H, m, H5–H6 and H7), 3.80
(1H, d, J = 11.2 Hz, H3), 3.61 (3H, s, H1), 3.29 (1H, dq, J = 7.4
and 11.2 Hz, H8), 2.30 (3H, s, H11), 0.88 (3H, d, J = 7.4 Hz, H9).
2b: 1H-NMR (CDCl3, 293 K): d 4.20 (1H, d, J = 7.2 Hz, H7),
2.88 (1H, dq, J = 7.2 Hz, H12), 2.34 (3H, s, H15), 1.22 (1H, d, J =
7.2 Hz, H13).
13C{ H} NMR (CDCl3, 293 K): d 210.5 (C10), 173.1 (C2), 136.9
1
(C4), 128.7, 128.3 (C5 and C6), 127.7 (C7), 54.1 (C1), 52.1 (C3),
50.0 (C8), 30.0 (C11), 15.8 (C9).
6798 | Dalton Trans., 2011, 40, 6792–6801
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
5-Methoxy-4,5-dimethyl-3-phenyldihydrofuran-2(3H)-one (9).
1H NMR (CDCl3, 293 K): d 7.39–7.18 (5H, m, H4–H5 and H6),
3.67 (1H, d, J = 12.1 Hz, H2), 3.44 (3H, s, H10), 2.43 (1H, dq, J =
6.8 and 12.1 Hz, H7), 1.64 (3H, s, H11), 1.07 (3H, d, J = 6.8 Hz,
H8).
Synthesis of the catalyst [Pd(Me)(OTf)(BIOX)]
1
13C{ H} NMR (CDCl3, 293 K): d 174.4 (C1), 135.2 (C3), 128.5,
128.1 (C4 and C5), 127.8 (C6), 118.12 (C9), 54.6 (C2), 49.8 (C10),
37.9 (C7), 20.1 (C11), 14.3 (C8).
5-Methoxy-4,5-dimethyl-3-phenyldihydrofuran-2(3H)-one (10).
1H NMR (CDCl3, 293 K): d 7.39–7.18 (5H, m, H4–H5 and H6),
3.51 (3H, s, H10), 3.44 (1H, d, J = 11.8 Hz, H2), 2.64 (1H, dq, J =
7.0 and 11.8 Hz, H7), 1.55 (3H, s, H11), 1.10 (3H, d, J = 7.0 Hz,
H8).
To a flame-dried 10 mL round bottom flask equipped with a stir
bar under a nitrogen atmosphere, were added [Pd(Me)(Cl)(BIOX)]
(100 mg, 0.25 mmol) and 5 mL of CH2Cl2. The orange solution was
stirred for 5 min and 13 mg of silver trifluoromethane sulfonate
(0.50 mmol, 2.00 equiv.) was added in the exclusion of light. The
cloudy orange solution was stirred for 4 h and the precipitate AgCl
was removed by filtering the solution through Celite rinsing with
CH2Cl2 (2 ¥ 5 mL). The solvent was removed in vacuo to yield an
yellow–orange solid in 96% yield (150.96 mg, 0.24 mmol).
1
13C{ H} NMR (CDCl3, 293 K): d 175.2 (C1), 136.2 (C3), 128.3,
128.0 (C4 and C5), 127.6 (C6), 117.64 (C9), 54.3 (C2), 49.6 (C10),
37.7 (C7), 20.4 (C11), 15.1 (C8).
Elemental analysis of the 8, 9 and 10 mixture: anal. calcd
for C13H16O3 (220.11): C, 70.89; H, 7.32%. Found: C, 70.85; H,
7.34%.
1H NMR (200 MHz, CD2Cl2, 293 K): d (ppm) 4.63–4.40 (4H,
m, H1 and H1¢), 4.18–4.12, 3.81–3.64 (1H each, m, H2 and H2¢),
2.42 (2H, m, H4 and H4¢), 1.12 (3H, s, Pd-Me), 0.98, 0.96, 0.94,
0.88 (3H each, J = 7.2, d, H5–H6 and H5¢-H6¢). 13C{1H}NMR
(CD2Cl2, 293 K): d (ppm) 157.2 (s, C3 and C3¢), 74.3, 72.6 (s, C1
and C1¢), 69.3, 66.7 (s, C2 and C2¢), 30.5, 29.9 (s, C4 and C4¢),
(2R)-Methyl-4-oxo-2-phenylpentanoate (11)
18.5, 18.3, 16.4, 13.6 (s, C5–C6 and C5¢–C6¢), 8.2 (s, Pd-Me). 19
NMR (94.14 MHz, CD2Cl2): d (ppm) = -78.35.
F
The experimental data of this compound turned out to be the same
as those reported in the literature.5a
Anal. calcd for C14H23F3N2O5PdS (494.83): C 33.98, H 4.68, N
5.66%. Found: C 33.53, H 4.65, N 5.22%.
Synthesis of the catalyst [Pd(OTf)2(BIOX)]
Catalytic carbonylation procedure
Typically, a MeOH solution (2 mL) of the selected olefin (cis-
b-methylstyrene, styrene or trans-b-methylstyrene) (9 mmol) was
introduced by suction into an autoclave, previously evacuated by
a vacuum pump, containing 0.045 mmol of the chosen catalyst,
MeOTf (9 mmol, 1.47 g, 1.01 mL) and TFE (30 mL, 0.3 mmol).
The autoclave was pressurized with CO to 3 atm. After 5 days,
the reaction was stopped and the pressure was released. The
formation of a black precipitate of palladium metal was observed,
diethyl ether was added and the mixture was filtered off through
Celite. The solution was evaporated and the residue was purified by
column chromatography (SiO2, hexane/ethyl acetate = 9/1, v/v)
to give the corresponding g-ketoesters. Isolated yields are given in
Table 1.
To a flame dried 10 mL round bottom flask equipped with a
stir bar under a nitrogen atmosphere, were added [PdCl2(BIOX)]
(100 mg, 0.25 mmol) and 5 mL of CH2Cl2. The orange solution
was stirred for 5 min and 13 mg of silver trifluoromethane
sulfonate (0.50 mmol, 2.00 equiv.) was added in the exclusion
of light. The cloudy orange solution was stirred for 4 h and the
precipitate AgCl was removed by filtering the solution through
Celite rinsing with CH2Cl2 (2 ¥ 5 mL). The solvent was removed
in vacuo to yield an yellow–orange solid in 96% yield (150.96 mg,
0.24 mmol).
The optical purity of g-ketoesters 6 and 11, resulting from cbms
or styrene respectively, was determined by 1H NMR using as shift
reagent europium tris[3-(heptafluoropropylhydroxymethylene)-
(+)-camphorate]. An excess of the latter (three equivalents) was
added to a chloroform-d1 solution of compound 6 or 11 producing
a splitting of the –COMe signals.
1H NMR (200 MHz, CD2Cl2, 293 K): d (ppm) 4.65–4.42 (4H,
m, H1 and H1¢), 4.11–4.08, 3.72–3.58 (1H each, m, H2 and H2¢),
2.34 (2H, m, H4 and H4¢), 0.96, 0.94, 0.92, 0.85 (3H each, J =
◦
1H NMR (200 MHz, CDCl3, 25 C, TMS): d 2.22 (92.5%), d
2.02 (7.5%) (s, 3H) in the case of compound 6. 1H NMR (200 MHz,
CDCl3, 25 ◦C, TMS): d 2.22 (95%), d 2.02 (5%) (s, 3H) in the case
of compound 11.
1
7.2, d, H5–H6 and H5¢-H6¢). 13C{ H}NMR (CD2Cl2, 293 K): d
(ppm)157.2 (s, C3 and C3¢), 74.3, 72.6 (s, C1 and C1¢), 69.3, 66.7
(s, C2 and C2¢), 30.5, 29.9 (s, C4 and C4¢), 18.5, 18.3, 16.4, 13.6 (s,
C5–C6 and C5–C6¢).
Crystallographic data collection and refinement
19F NMR (94.14 MHz, CD2Cl2): d (ppm) = -78.46.
Anal. calcd for C14H20F6N2O8PdS2 (628.86): C 26.74, H 3.21, N
4.45%. Found: C 26.42, H 2.87, N 4.82%.
Crystals of 2a and 3a were obtained by slow diffusion of hexane
into a dichloromethane solution of compounds 2 and 3.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6792–6801 | 6799
©

View Article Online
Crystallographic data for 2a and 3a were collected on a Siemens
SMART diffractometer equipped with a rotating anode and
controlled with the SMART software.26 The radiation used was
Computational details
The Gaussian 03 (Revision C.02)36 package was used. All the
species were fully optimized by using the density functional
method (DFT) by means of Becke’s three-parameter hybrid
method using the LYP correlation functional.37 Two sets of
calculations were performed on each modelled species by using
two different effective core potentials for the palladium ion: the
Hay and Wadt38 potential and the Stuttgart/Dresden ECPs.39
Neither the relative energy trends or the geometrical parameters
were significantly affected by the two different potential schemes,
as a consequence only data from the Hay and Wadt ECP were
reported. The 6-31G* basis set40 was used for all the other atomic
species. In all cases the reliability of the stationary points (minima
on the potential energy hypersurface) was assessed by evaluating
the vibrational frequencies. Starting geometries for the 2a¯ - 4a¯
˚
Cu-Ka (l = 1.5418 A) and intensity data were acquired at
200 K. Five settings of w were used and narrow data “frames”
were collected for 0.3◦ increments in w. A total of 3000 frames
of data were stored providing a sphere of data.27 Data reduction
was performed with the SAINT 4.0 program.28 Structures were
then solved using the SIR97 program,29 and then refined using
the SHELX97 program.30 Absorption corrections to all data were
applied (SADABS).31 In both cases the correct enantiomer was
identified by means of the Flack parameter.32 In this respect
it must be noticed that in spite of the disorder affecting the
atoms O(3), C(13), C(15) as well as the methyl carbon atoms
C(14) and C(16) in 3a, the stereochemistry of the carbon atom
C(15) has been clearly established. For each disordered atom two
positions were introduced, each one with an occupancy factor
of 0.5. As often found for this species,33 in both structures the
fluorine atoms of the counter ion are disordered. During the
refinement it was decided for every CF3 group (with just one
exception), to introduce and refine anisotropically the three highest
density peaks. In the last CF3 group a two sets of fluorine atoms
were introduced with an occupancy factor of 0.5, these atoms
were isotropically refined. Anisotropic thermal parameters were
used for all the atoms but carbon and hydrogen atoms because
of the otherwise poor observed reflections/parameters ratio. All
the hydrogen atoms were introduced in calculated position and
refined with a temperature factor depending on the one of the
atom to which the hydrogen atom is bound. In complex 3a, the
hydrogen atoms bound to C(14), C(15), C(16) and C(17) were not
introduced because of the above-mentioned disorder. Geometrical
calculations were performed by PARST9734 and molecular plots
were produced by the ORTEP3 program.35 Crystal data and
structure refinement parameters are listed in Table 4.
¯
¯
and 2b - 4b olefin adducts, as well as of their corresponding
insertion products, were based on the available X-ray diffraction
data of metal complexes featuring analogous molecular fragments
(Cambridge Structural Database, version 5.31, November 2009).13
Conclusions
In short, the result of the present investigation is twofold. A
carbonylation process of styrenes, catalyzed by an optically active
bioxazoline palladium complex has been studied by isolating the
reaction intermediates and determining their structure in solution
by NMR spectroscopy and, for two cases, in the solid state by
X-ray crystallography. The stereoselectivity of the reaction has
been rationalized on the basis of a simple model of Curtin–
Hammett type. According to this model, the degree of stereoselec-
tivity appears to be proportional to the difference in the free energy
of the two possible transition states G‡b and G‡ , which correspond
a
to the insertion of either the Si or the Re olefin face respectively.
In conclusion, the ultimate source of stereoselectivity is the steric
interaction of the catalytic site with the olefin substituents in the
transition states.
Table 4 Crystal data and structure refinement details for 2a and 3a
Moreover, on the practical application side, the investigated
reaction scheme allowed a new simple way to selectively synthesize,
in one pot and in mild conditions, highly substituted chiral g-
ketoesters with a good enantiomeric excess.
The optimization of the process, the extension to variously
substituted olefin as well as the use of other nucleophiles instead
of the simple alcohol are currently in progress.
2a
3a
Empirical formula
Mr/g mol-1
T/K
C55H45BF24N2O3Pd
1355.14
200
1.54184
Triclinic
P1
10.231(1)
11.168(1)
13.729(1)
72.340(2)
75.249(2)
80.614(2)
1439.2(2)
1
1.564
3.712
5458
4376
0.0364
4376/3/500
0.0734
0.1912
1.034
C55H45BF24N2O3Pd
1355.14
200
1.54184
Triclinic
P1
10.193(1)
11.051(1)
13.743(1)
72.314(3)
78.093(3)
81.707(3)
1437.6(2)
1
1.565
3.716
6533
4821
0.0386
4821/3/512
0.0829
0.2172
0.0829
˚
l/A
Crystal system
Space group
˚
a/A
˚
b/A
Acknowledgements
˚
c/A
a (◦)
b (◦)
g (◦)
This Work was supported by the Ministero dell’Istruzione,
dell’Universita` e della Ricerca (MIUR, PRIN N. 2008A7P7YJ).
3
˚
V/A
Z
Dc/g cm-3
Notes and references
m(Cu-K a)/mm-1
Reflections collected
Unique reflections
Rint
1 E. Negishi, and A. De Meijere, Handbook of Organopalladium Chem-
istry for Organic Synthesis, Wiley, New York, 2002.
2 (a) E. Drent and P. H. M. Budzelaar, Chem. Rev., 1996, 96, 663;
(b) K. Nozaki and T. J. Hiyama, J. Organomet. Chem., 1999, 576,
248; (c) C. Bianchini and A. Meli, Coord. Chem. Rev., 2002, 225, 35;
(d) K. Nakano, N. Kosaka, T. Hiyama and K. Nozaki, Dalton Trans.,
2003, 4039; (e) E. J. G. Sua´rez, C. Godard, A. Ruiz and C. Claver,
Eur. J. Inorg. Chem., 2007, 2582; (f) J. Durand and B. Milani, Coord.
Chem. Rev., 2006, 250, 542.
Data/restraints/param
R1 (I > 2s(I))
wR2 (all data)
GOF on F2
Flack’s parameter
0.05(1)
0.06(1)
6800 | Dalton Trans., 2011, 40, 6792–6801
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
3 K. R. Reddy, K. Surekha, G. H. Lee, S. M. Peng, J. T. Chen and S. T.
Liu, Organometallics, 2001, 20, 1292, and references cited therein.
4 (a) F. C. Rix, M. Brookhart and P. S. White, J. Am. Chem. Soc., 1996,
118, 4746; (b) S. Stoccoro, G. Minghetti, M. A. Cinellu, A. Zucca and
M. Manassero, Organometallics, 2001, 20, 4111.
5 (a) K. Nozaki, H. Komaki, Y. Kawashima, T. Hiyama and T.
Matsubara, J. Am. Chem. Soc., 2001, 123, 534; (b) M. Brookhart, F. C.
Rix, J. M. DeSimone and J. C. Barborak, J. Am. Chem. Soc., 1992,
114, 5894; (c) A. Gsponer, T. M. Schmid and G. Consiglio, Helv. Chim.
Acta, 2001, 84, 2986; (d) F. C. Rix, M. J. Rachita, M. I. Wagner, M.
Brookhart, B. Milani and J. C. Barborak, Dalton Trans., 2009, 8977.
6 K. Nozaki, N. Sato, Y. Tonomura, M. Yasutomi, H. Takaya, T. Hiyama,
T. Matsubara and N. Koga, J. Am. Chem. Soc., 1997, 119, 12779.
7 R. Van Asselt, E. E. C. G. Gielsens, R. E. Ru¨lke, K. Vrieze and C. J.
Elsevier, J. Am. Chem. Soc., 1994, 116, 977, and references cited therein.
8 G. P. C. M. Dekker, C. J. Elsevier, K. Vrieze, P. W. N. M. Van, Leeuwen
and C. F. Roobeek, J. Organomet. Chem., 1992, 430, 357.
Ronsheim and C. K. Zercher, J. Org. Chem., 2003, 68, 4535; (d) W. J.
Zhao, M. Yan, D. Huang and S. J. Ji, Tetrahedron, 2005, 61, 5585; (e) M.
Miyashita, R. Yamaguchi and A. Yoshikoshi, J. Org. Chem., 1984, 49,
2857.
22 (a) T. Hayashi, A. Yamamoto and Y. Ito, Chem. Lett., 1987, 177;
(b) D. A. Evans, C. E. Masse and J. Wu, Org. Lett., 2002, 4, 3375.
23 F. Montserrat, A. Gutie´rrez, M. Mounir, O. Rossell, E. Ruiz, A. Rang
and M. Engeser, Inorg. Chem., 2007, 46, 3395.
24 S. E. Denmark, R. A. Stavenger, A. M. Faucher and J. P. Edwards,
J. Org. Chem., 1997, 62, 3375.
25 (a) S. R. Bahr and P. Boudjouk, J. Org. Chem., 1992, 57, 5545; (b) M.
Brookhart, B. Grant and A. Volpe, Organometallics, 1992, 11, 3920.
26 SMART: Area-Detector Integration SoftwareSiemens Industrial Au-
tomation, Inc., Madison, WI, 1995.
27 Due to the not very good quality of crystals of 2a and 3a a lot of them
were tested on the Siemens SMART diffractometer equipped with a
rotating anode. Data reported in the present work refer to the best
crystal selected for each compound.
9 (a) C. Carfagna, G. Gatti, D. Martini and C. Pettinari, Organometallics,
2001, 20, 2175; (b) C. Carfagna, M. Formica, G. Gatti, A. Musco and
A. Pierleoni, Chem. Commun., 1998, 1113.
28 SAINT Version 4.0; Siemens Industrial Automation, Inc., Madison,
WI, 1995.
10 (a) A. Scarel, J. Durand, D. Franchi, E. Zangrado, G. Mestroni, C.
Carfagna, L. Mosca, R. Seraglia, G. Consiglio and B. Milani, Chem.–
Eur. J., 2005, 11, 6014; (b) B. Binotti, C. Carfagna, G. Gatti, D. Martini,
L. Mosca and C. Pettinari, Organometallics, 2003, 22, 1115; (c) S.
Bartolini, C. Carfagna and A. Musco, Macromol. Rapid Commun.,
1995, 16, 9.
11 (a) C. Carfagna, G. Gatti, L. Mosca, A. Passeri, P. Paoli and A.
Guerri, Chem. Commun., 2007, 4540; (b) B. Binotti, G. Bellachioma,
G. Cardaci, C. Carfagna, C. Zuccaccia and A. Macchioni, Chem.–
Eur. J., 2007, 13, 1570; (c) B. Binotti, C. Carfagna, C. Zuccaccia and
A. Macchioni, Chem. Commun., 2005, 92.
12 C. Carfagna, G. Gatti, L. Mosca, P. Paoli and A. Guerri, Helv. Chim.
Acta, 2006, 89, 1660.
13 F. H. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58, 380.
14 W. Klyne and V. Prelog, Experientia, 1960, 16, 521.
15 G. Consiglio, Chimia, 2001, 55, 809.
16 G. Milano, G. Guerra, M. Mazzeo, C. Pellecchia and L. Cavallo,
Macromolecules, 2005, 38, 2072.
17 A similar cyclisation, leading to unsaturated alkoxy-g-lactones, has
been observed in the PdI2-catalysed oxidative carbonylation of alkynes:
(a) B. Gabriele, M. Costa, G. Salerno and G. P. Chiusoli, J. Chem. Soc.,
Perkin Trans. 1, 1994, 83; (b) B. Gabriele, G. Salerno, M. Costa and
G. P. Chiusoli, Curr. Org. Chem., 2004, 8, 919.
18 A. Scarel, J. Durand, D. Franchi, E. Zangrando, G. Mestroni, B.
Milani, S. Gladiali, C. Carfagna, B. Binotti, S. Bronco and T. Gragnoli,
J. Organomet. Chem., 2005, 690, 2106.
19 D. Sirbu, G. Consiglio, B. Milani, P. G. A. Kumar, P. S. Pregosin and
S. Gischig, J. Organomet. Chem., 2005, 690, 2254.
29 A. Altomare, G. L. Cascarano, C. Giacovazzo, A. Guagliardi, M. C.
Burla, G. Polidori and M. Camalli, J. Appl. Crystallogr., 1999, 32, 115.
30 G. M. Sheldrick, SHELX 97, University of Go¨ttingen, Germany, 1997.
31 G. M. Sheldrick, SADABS, University of Go¨ttingen, Germany, 1996.
32 H. D. Flack, Acta Crystallogr., Sect. A: Found. Crystallogr., 1983, 39,
876–881.
33 G. Gatti, J. A. Lopez, C. Mealli and A. Musco, J. Organomet. Chem.,
1994, 483, 77.
34 M. Nardelli, Comput. Chem., 1983, 7, 95.
35 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
36 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb,
J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven, K. N. Kudin, J. C.
Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone, B. Mennucci,
M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E.
Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo,
R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi,
C. Pomelli, J. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P.
Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D.
Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K.
Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S.
Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz,
I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y.
Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. G. Johnson,
W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, GAUSSIAN 03
(Revision C.02), Gaussian, Inc., Wallingford, CT, 2004.
37 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
20 W. Wang, B. Xu and G. B. Hammond, J. Org. Chem., 2009, 74, 1640.
21 (a) E. Starodubtseva, O. Turova, M. Vinogradov, A. Ferapontov and
I. Struchkova, Tetrahedron, 2008, 64, 11713; (b) M. Yan, W. J. Zhao,
D. Huang and S. J. Ji, Tetrahedron Lett., 2004, 45, 6365; (c) M. D.
38 P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 299.
39 X. Y. Cao and M. J. Dolg, THEOCHEM, 2002, 581, 139.
40 G. A. Petersson, A. Bennett, T. G. Tensfeldt, M. A. Al-Laham, W. A.
Shiirlay and J. Mantzaris, J. Chem. Phys., 1988, 89, 2193.
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 6792–6801 | 6801
©