Room temperature asymmetric Pd-catalyzed methoxycarbonylation of norbornene: Highly selective catalysis and HP-NMR studies

Article abstract of DOI:10.1039/c2dt30267e

Palladium complexes bearing monodentate and bidentate phosphine ligands (1-7) were synthesised and used as catalyst precursors in the methoxycarbonylation of norbornene. The catalytic systems bearing ligands 1, 3 and 4 afforded excellent conversions (>99%) and selectivity of the ester (>99%). NMR investigations showed that using complex 1a as the precursor resulted in the protonated phosphine, 1-H⁺, being formed under catalytic conditions and thus the addition of acid is not required for the activation of this system since the reaction involving the precursor with methanol under CO pressure produces 2 equivalents of HCl and leads to the formation of the active species. The protonation of ligand 4 under methoxycarbonylation conditions was also observed and the diprotonated diphosphine was isolated and characterised. This compound was tested as a ligand and acid source in a catalysis and provided excellent conversion and high selectivity to the ester.

Full text of DOI:10.1039/c2dt30267e

View Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 6980
www.rsc.org/dalton
PAPER
Room temperature asymmetric Pd-catalyzed methoxycarbonylation of
norbornene: highly selective catalysis and HP-NMR studies†
Carolina Blanco,a Cyril Godard,*a Ennio Zangrando,b Aurora Ruiza and Carmen Clavera
Received 6th February 2012, Accepted 21st March 2012
DOI: 10.1039/c2dt30267e
Palladium complexes bearing monodentate and bidentate phosphine ligands (1–7) were synthesised and
used as catalyst precursors in the methoxycarbonylation of norbornene. The catalytic systems bearing
ligands 1, 3 and 4 afforded excellent conversions (>99%) and selectivity of the ester (>99%). NMR
investigations showed that using complex 1a as the precursor resulted in the protonated phosphine, 1-H+,
being formed under catalytic conditions and thus the addition of acid is not required for the activation of
this system since the reaction involving the precursor with methanol under CO pressure produces 2
equivalents of HCl and leads to the formation of the active species. The protonation of ligand 4 under
methoxycarbonylation conditions was also observed and the diprotonated diphosphine was isolated and
characterised. This compound was tested as a ligand and acid source in a catalysis and provided excellent
conversion and high selectivity to the ester.
Introduction
Carbonylation reactions are among the most important industrial
processes in homogeneous transition metal catalysis.1 In these
reactions, a low cost substrate is transformed in the presence of a
metal catalyst into valuable compounds and/or intermediates for
the fine chemical industry. The Pd-catalysed alkoxycarbonylation
of alkenes is a process of interest for both academic and indus-
trial researchers, which produces esters from simple alkenes.2 In
this reaction, excellent results can be achieved with substrates,
such as ethene3 and vinylarenes.4 However, minor changes in
the catalytic system and/or the reaction conditions were reported
to greatly affect the selectivity of the catalytic process, usually
leading to the formation of a variety of by-products, such as
cooligomers, copolymers and ethers.5 While the formation of the
ether is explained by the addition of the alcohol across the
alkene under acidic conditions, the copolymerization presents
itself as a competing Pd-catalysed reaction that usually takes
place under similar reaction conditions to those used in
alkoxycarbonylations.
In the Pd-catalysed methoxycarbonylation of alkenes, two
mechanisms have been proposed: the Pd–hydride and the Pd–
carbomethoxy mechanisms (Scheme 1).1d,6
Scheme 1 The proposed mechanisms for the Pd-catalyzed alkoxycar-
bonylation of alkenes.
aDepartament de Química Física i Inorgànica, Universitat Rovira i
Virgili, Marcel.li Domingo s/n, 43007, Tarragona, Spain. E-mail:
The first mechanism starts with the formation of a palladium
hydride complex. Subsequent coordination of the substrate, fol-
lowed by insertion into the Pd–H bond affords a Pd–alkyl
complex which is transformed into an acyl complex by the
migratory insertion of CO. Finally, methanolysis of the acyl
intermediate yields the ester product(s) and regenerates the initial
cyril.godard@urv.cat; Fax: +34 977 559563; Tel: +34 977 559574
bDipartimento di Scienze Chimiche e Farmaceutiche,
Università di Trieste, via Licio Giorgieri 1, 34127 Trieste, Italy.
E-mail: ezangrando@units.it
†Electronic supplementary information (ESI) available: tables of X-ray
crystallographic data in CIF format for compounds 4a–7a (CCDC
866062 (4a) and 866063 (7a)). See DOI: 10.1039/c2dt30267e
6980 | Dalton Trans., 2012, 41, 6980–6991
This journal is © The Royal Society of Chemistry 2012

View Online
hydride species. The carbomethoxy mechanism starts with a
Pd–OMe complex, which, after coordination and migratory
insertion of CO, yields the corresponding carbomethoxy
complex, Pd–COOMe. Coordination and insertion of the alkene
followed by methanolysis yields the product and regenerates the
initial catalyst. The production of the Pd–H species from the
complexes formed in the carbomethoxy cycle has also been
demonstrated to occur through the β-elimination of an unsatu-
rated ester after alkene insertion. It is noteworthy that, in the case
of cyclic substrates such as norbornene, this β-elimination
process does not occur. Experimental evidence for the hydride
mechanism have been reported for systems modified with mono-
dentate and bidentate ligands.7–10 Zacchini et al. observed that
oxidative addition of strong acids, such as HCl to Pd(0) com-
plexes, including [Pd(CO)(PPh3)3] or [Pd(PPh3)4].18 Related
hydride complexes have also been reported with other monoden-
8
tate phosphines, such as PCy3 and PBu3.19 From an industrial
point of view, the use of acid is often an issue due to the cor-
rosion of the reaction vessels. In this context, an interesting cata-
lytic system, namely PdCl2/NMDPP, was reported to afford high
activity and excellent selectivities in the methoxycarbonylation
of styrene in the absence of acid (NMDPP = neo-menthyl
diphenyl phosphine).11,20 To the best of our knowledge, the
reason for which this catalytic system operates in the absence of
acid has not been reported.
Recently, we briefly described the catalytic performance of a
series of palladium systems bearing monodentate and bidentate
phosphine ligands in the methoxycarbonylation of norbornene
(Fig. 1).15a The effect of the acid was studied in detail and high
activity and selectivity were achieved. Herein, we report further
investigations into these catalytic systems, including some
insights into the reaction mechanism using conventional and
high pressure NMR methods.
[PdH(L–L)(MeOH)X)] complexes (where L–L
= 1,2-bis-
(CH2PBut2)2C6H4 and X = BF4 or CF3SO3−) were formed by
the reaction of [Pd(L–L)(dba)] with methanol and HBF4 or
CF3SO3H in the presence of oxygen or benzoquinone. The
hydride complex [PdH(L–L)(MeOH)]+ was shown to be a key
intermediate in the catalytic methoxycarbonylation of ethene.9
However, no mechanistic studies on the methoxycarbonylation
of norbornene has been reported to date.
−
The asymmetric Pd-catalysed methoxycarbonylation of sub-
strates, such as 2-vinyl-6-methoxynaphthalene,11 4-methoxystyr-
ene,12 α-methylstyrene13 and acenaphthylene,14 have been
extensively studied. However, the alkoxycarbonylation of nor-
bornene has been less well studied15 (Scheme 2), although its
functionalisation is particularly relevant since the product pre-
sents unique structural and chemical features and can produce
valuable compounds with potential applications in the fragrance
industry, such as in Dupical® and in sandalwood.16 The functio-
nalisation of this olefin generates three chiral carbon centres
upon one C–C bond formation. In this process, chemo-, stereo-
(exo/endo) and enantio-selectivity are important issues.15a,c,d In
1996, Zhou et al. first reported the asymmetric alkoxycarbonyla-
tion of norbornene using Pd(OAc)2/DDPPI/p-TsOH as a cataly-
tic system and achieved 72% chemoselectivity toward esters and
92% ee under 50 atm of CO at 120 °C (DDPPI = 1,4:3,6-dian-
hydro-2,5-dideoxy-2,5-bis(diphenylphosphino)-L-iditol).15c,d In
1997, Inoue et al. carried out the alkoxycarbonylation of various
olefins using the cationic palladium complexes [Pd-
(MeCN)2(PPh3)2](BF4)2 as catalysts. When norbornene was used
as a substrate, up to 74% chemoselectivity toward esters was
achieved.15b
Results and discussion
Synthesis of the palladium complexes
The palladium complexes, [PdCl2(1)2] (1a),21 [PdCl2(3)] (3a),6c
and [PdCl2(5)] (5a),22 were synthesised according to previously
reported methods. The complex [PdCl2(4)] (4a) was prepared by
reaction of [PdCl2(COD)] in the presence of 1.1 equiv. of biden-
tate ligand 4 in dichloromethane at room temperature. The
complex was isolated after precipitation with diethyl ether as an
orange powder that recrystallizes in a mixture of THF–ether as
red crystals. Single crystals suitable for X-ray diffraction were
obtained from this mixture of solvents. The molecular structure
of 4a is shown in Fig. 2. Selected bond lengths and angles are
listed in Table 1. The crystal structure, containing a lattice mol-
ecule of THF, shows the Pd atom adopting a distorted square
planar coordination geometry. The Pd–Cl(1) and Pd–Cl(2) bond
lengths were measured to be 2.3481(17) and 2.3529(17) Å,
respectively, while the Pd–P bonds were also found to be com-
parable in length (2.3331(17) and 2.3206(16) Å for Pd–P(1) and
Pd–P(2), respectively). These values are very similar to those
previously reported for related Pd(II) dichloride complexes con-
taining other phosphine ligands.12
In the alkoxycarbonylation reaction, the presence of acid is
often required to promote catalytic activity via the formation of
Pd–hydride species.17 Under such conditions, neutral palladium
complexes of the type [PdHCl(PPh3)2] can be formed by
Scheme 2 Methoxycarbonylation of norbornene catalyzed by
palladium.
Fig. 1 The ligands used in this study.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6980–6991 | 6981

View Online
Table 2 Methoxycarbonylation of norbornene using monodentate
phosphine ligands 1 and 2a
T
Cb,c
Acid (°C) (%)
Chemob
(%)
% exo
(ee)
Entry Catalyst
1
2
3
4
5
6
7
8
9
PdCl2/1
PdCl2/1
[PdCl2(1)2] (1a)
[PdCl2(1)2] (1a)
[PdCl2(1)2] (1a)
PdCl2/2
PdCl2/2
[PdCl2(PhCN)2]/2 YES RT
PdCl2/2 NO 70
NO RT
NO 70
NO RT
YES RT
NO 70
YES RT
YES 70
>99
>99
>99
89
>99
25
99
14
53
>99
>99
>99
90
>99
>99
90
>99(3)
>99(9)
>99(10)
>99(10)
>99(11)
>99(33)
>99(40)
>99(46)
—
>99
25
a Reaction conditions: Pd (0.021 mmol), TFA (0.210 mmol), norbornene
(1.05 mmol), MeOH–THF (1 : 1), 30 bars of CO, 24 h. b All
conversions and chemoselectivities toward the ester were determined
by GC and GC-MS (chemo
= chemoselectivity to the ester).
c C = conversion.
remained unaltered (Table 2, entry 2). When the isolated
[PdCl2(1)2] (1a) system was probed as a catalytic precursor, both
room temperature and 70 °C reactions yielded excellent conver-
sions (>99%) (Table 2, entries 3 and 5) together with excellent
chemo- (>99%) and exo-selectivities (>99%). However, the ee
remained low. When this catalytic system was used under acidic
conditions at room temperature, both the conversion and che-
moselectivity decreased (89% and 90%, respectively); however,
the exo-selectivity remained excellent (Table 2, entry 4). The
addition of acid was thus detrimental and the methanol addition
product, 2-methoxybicyclo[2.2.1]heptane, was identified as the
by-product under these conditions.
Fig. 2 An ORTEP diagram of complex 4a (30% probability ellipsoids;
solvent molecule omitted for clarity).
Table 1 Selected bond lengths (Å, top) and angles (°, bottom) for 4a
Pd–P(1)
Pd–P(2)
2.3331(17)
2.3206(16)
Pd–Cl(1)
Pd–Cl(2)
2.3481(17)
2.3529(17)
P(1)–Pd–P(2)
P(1)–Pd–Cl(1)
P(1)–Pd–Cl(2)
103.95(6)
163.23(6)
88.24(6)
P(2)–Pd–Cl(1)
P(2)–Pd–Cl(2)
Cl(1)–Pd–Cl(2)
87.08(6)
162.48(6)
84.12(6)
When catalytic systems containing the (S,S)-ferrocenyl-
phosphine (2)23 ligand were used at room temperature in the
presence of trifluoroacetic acid, excellent chemo- and exo-selec-
tivities were achieved (Table 2, entries 6 and 8). It is noteworthy
that when [PdCl2(PhCN)2] was used as a precursor at room
temperature, the enantioselectivity increased to 46% (Table 2,
entry 8). However, the conversions were low under these con-
ditions (25 and 14%). A significant increase in conversion (up to
99%) was observed when the temperature was increased to
70 °C. At this temperature, the exo-selectivity remained
unchanged although the chemoselectivity and ee decreased to
90% and 40%, respectively (Table 2, entry 7). In the absence of
any acid, both the conversion and the chemoselectivity of the
reaction were drastically affected (Table 2, entry 9).
In view of the results obtained with [PdCl2(1)2] (1a), in terms
of the conversion and selectivity, larger substrate concentrations
(from 50 to 1000 equiv./Pd) were used. The results are summar-
ized in Table 3. Excellent conversion (>99%), chemo- (>99%)
and exo-selectivities (>99%) were observed in the presence of 50
and 100 equiv. of norbornene (Table 3, entries 1 and 2).
However, when the ratio of NBN : Pd was increased to 200, a
decrease in chemoselectivity (71%) was observed, although both
the conversion (>99%) and exo-selectivity remained excellent
(>99%) (Table 3, entry 3). When the ratio of NBN : Pd was
increased from 200 to 500 and 1000, a significant decrease in
both conversion and chemoselectivity was observed (Table 3,
entries 4 and 5). In all cases, co-oligomers were detected as the
The bite angle, P(1)–Pd–P(2), of diphosphine 4 when coordi-
nated at the palladium center was 103.95(6)°, while the
Cl(1)–Pd–Cl(2) bond angle was found to be narrower at
84.12(6)°, which is likely due to steric constraints.
Catalytic results
Palladium catalyzed methoxycarbonylation of norbornene
using the monodentate ligands 1 and 2. Our group recently
reported that using the phosphine ligands 1 and 2 (Fig. 1) in the
Pd-catalyzed methoxycarbonylation of norbornene resulted in
excellent conversions and selectivities at 70 °C. Furthermore,
when using ligand 1, the presence of acid was not required.15a
Promising enantiomeric excess values and high conversions and
selectivities were obtained using ligand 2 in the presence of trifl-
uoroacetic acid. After optimisation of the reaction conditions,
these catalytic systems were found to be highly active even at
room temperature (Table 2).
When PdCl2/(S)-NMDPP (1) was used as a catalytic system at
room temperature (RT) without additional acid, total conversion
(>99%) was achieved together with an excellent chemo- (>99%)
and exo- selectivity (>99%) (Table 2, entry 1). However, the
enantiomeric excess was low. When the reaction temperature was
increased from room temperature to 70 °C, the catalytic results
6982 | Dalton Trans., 2012, 41, 6980–6991
This journal is © The Royal Society of Chemistry 2012

View Online
Table 3 The effect of the norbornene concentration using complex 1a
Table 4 Palladium complexes bearing bidentate ligands (3–7) as
in the palladium-catalyzed methoxycarbonylation of norbornenea
catalyst precursors in the methoxycarbonylation of norbornenea
% exo
Entry Precursor
T (°C) Cb, c (%) Chemob (%) % exo
Entry Catalyst
NBN (equiv.) Cb,c (%) Chemob,c (%) (ee)
1
2
3
4
5
6
7
8
PdCl2/3
PdCl2/3
[PdCl2(3)] (3a) RT
[PdCl2(3)] (3a) 70
PdCl2/4
PdCl2/4
[PdCl2(4)] (4a) RT
[PdCl2(4)] (4a) 70
PdCl2/5
[PdCl2(5)]
PdCl2/6
PdCl2/7
RT
70
19
91
42
99
75
0
99
99
14
56
<5
<5
>99
>99
>99
>99
>99
—
>99
>99
89
>99
>99
>99
>99
>99
—
>99
>99
>99
>99
—
1
[PdCl2(1)2](1a) 50
>99
>99
>99
(10)
>99(6)
>99(7)
>99(8)
>99(11)
2
3
4
5
[PdCl2(1)2](1a) 100
[PdCl2(1)2](1a) 200
[PdCl2(1)2](1a) 500
[PdCl2(1)2](1a) 1000
>99
>99
42
>99
71
40
70
RT
23
26
a Reaction conditions: Pd (0.021 mmol), MeOH–THF (1 : 1), 30 bars of
CO, RT, 24 h. b All conversions and chemoselectivities toward the ester
were determined by GC and GC-MS. c C = conversion; Chemo =
selectivity to the ester.
9
70
70
70
70
10
11
12
69
<5
<5
—
a Reaction conditions: Pd (0.021 mmol), TFA (0.210 mmol), norbornene
(1.05 mmol), MeOH–THF(1 : 1), 30 bars of CO, 24 h. b All conversions
and chemoselectivities towards the ester were determined by GC and
GC-MS. c C = conversion, Chemo = selectivity to the ester.
by-products of the reaction. Although the decrease in the conver-
sion as a function of the norbornene concentration was predict-
able, the lower chemoselectivity observed in these experiments
was unexpected. These results could be explained by an increase
in the rate of substrate insertion into the Pd–C bond of the
expected Pd–acyl intermediate when compared to the rate of the
methanolysis step of the same species, thus forming oligomerisa-
tion products in preference to the esters.
(<5%) was observed in this reaction using the in situ systems
PdCl2/6–7 at 70 °C (Table 4, entries 11–12). To check whether
the formation of the catalysts could explain these results,
attempts to isolate the palladium dichloride complex, [PdCl2(7)],
by reaction of [PdCl2(COD)] or [PdCl2(PhCN)2] in the presence
of the ligand were completed but the expected complex was not
obtained. Instead, the formation of the palladium(II) pincer
complex, 7a, was observed and this species was characterized by
multinuclear NMR spectroscopy, elemental analysis and X-ray
diffraction.
Surprisingly, the molecular structure of complex 7a showed
the coordination of a bromide ligand to the palladium centre. To
understand the origin of the presence of this ligand, several
experiments were completed and electrospray mass spectroscopic
analysis revealed that ligand 7, which was purchased from a
commercial source, was contaminated with one equivalent of
HBr per ligand. It was thus concluded that the contamination of
this reagent with HBr was responsible for the coordination of a
bromide ligand in the palladium complex, 7a. The molecular
structure of 7a is shown in Fig. 3 and a selection of bond lengths
and angles are shown in Table 5.
The coordination sphere around the palladium atom was
found to correspond to a distorted square-planar geometry, with
P(1)–Pd–P(2) and C(12)–Pd–Br bond angles of 166.84(4)
and 178.00(12)°, respectively, and cis-angles in the range of
83.33(12) to 97.06(3)°. The Pd–Br and Pd–C were found to be
2.5152(7) and 2.034(4) Å, respectively, whereas Pd–P bond
lengths were closely comparable (mean 2.282 Å) and similar to
those observed in the analogous [PdBr(C6H3-2,6-CH2PCy2)2]24
and [PdBr(C6H3-2,6-CH2PPh2)2] complexes.25 All these values
agree with those reported for similar palladium(II) pincer
complexes.24,25
When complex 7a was tested in the methoxycarbonylation
reaction, no catalytic activity was observed. It was therefore con-
cluded that the formation of this inactive species was occurring
when the in situ system was used, thus explaining the lack of cat-
alytic activity when systems containing ligands 6 and 7 were uti-
lized. To ensure that the HBr impurity did not have an influence
on the catalytic results, ligand 7 was purified by treatment with
Pd-catalyzed methoxycarbonylation of norbornene using the
bidentate ligands 3–7. The PdCl2/3–7 in situ catalytic systems
were tested in the palladium-catalyzed methoxycarbonylation of
norbornene. The catalytic reactions were optimised and the
results compared to those of the corresponding isolated com-
plexes. The reactions were carried out at room temperature and
70 °C in the presence of acid and under 30 bar of CO during a
period of 24 hours. The results are summarised in Table 4.
When the in situ PdCl2/3 system (Table 4, entry 1) was tested
at room temperature, 19% conversion was obtained together with
excellent chemo- (>99%) and stereoselectivities (>99%).
However, when the reaction temperature was increased to 70 °C,
a higher conversion was obtained (91%) (Table 4, entry 2). The
chemo- and stereoselectivities were not affected by the reaction
temperature. Similar behaviour was observed when the isolated
complex, [PdCl2(3)] (3a), was used as a precursor (Table 4,
entries 3 and 4). When the in situ PdCl2/4 system was used at
70 °C, high conversion (75%) with excellent chemo- (>99%)
and exo-selectivities (>99%) were obtained (Table 4, entry 5).
However, when the reaction was performed at room temperature
no conversion was observed (Table 4, entry 6). In contrast, when
the isolated complex, [PdCl2(4)] (4a), was used as a precursor,
total conversion (99%) was achieved at both temperatures
(Table 4, entries 7 and 8). This indicated that, under in situ con-
ditions at room temperature, no active species were formed,
while such species were efficiently formed from [PdCl2(4)]
(Table 4, entries 6 vs. 7).
When the in situ system containing ligand 5 was tested at
70 °C, the conversion was very poor (14%) and the chemoselec-
tivity was only moderate (Table 4, entry 9). When the corre-
sponding isolated palladium complex was tested, a higher
conversion was obtained although the chemoselectivity to the
ester slightly decreased (Table 4, entry 10). No catalytic activity
was observed at room temperature with systems bearing this
ligand. We previously reported that very low catalytic activity
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6980–6991 | 6983

View Online
Table 6 The effect of the norbornene concentration using complex 4a
in the Pd-catalyzed methoxycarbonylation reactiona
Entry
NBN (equiv.)
Cb,c (%)
Chemob,c (%)
% exo(ee)
1
2
3
4
50
99
99
23
<5
>99
>99
>99
nd
>99/—
>99/—
>99/—
nd
100
200
500
a Reaction conditions: [PdCl2(4)] (0.021 mmol), TFA (0.210 mmol),
MeOH–THF 5 ml (1 : 1), 30 bars of CO, RT. nd = not determined. b All
conversions and chemoselectivities towards the ester were determined
by GC and GC-MS. c C = conversion; Chemo = selectivity to the ester.
Fig. 3 An ORTEP diagram (30% probability of ellipsoids) of complex
7a.
Table 5 Selected bond lengths (Å, top) and angles (°, bottom) for
complex 7a
Pd–P(1)
Pd–P(2)
2.2828(12)
2.2823(12)
Pd–C(12)
Pd–Br
2.034(4)
2.5152(7)
P(1)–Pd–P(2)
P(1)–Pd–C(12)
P(2)–Pd–C(12)
166.84(4)
83.33(12)
83.97(13)
C(12)–Pd–Br
P(1)–Pd–Br
P(2)–Pd–Br
178.00(12)
95.74(3)
97.06(3)
Fig. 4 The reaction sequence by 31P{1H} NMR spectroscopy at RT.
(a) 1a in CD3OD–THF, (b) 1a + 10 equiv. NBN + 30 atm CO in
CD3OD–THF after 5 h shaking and (c) after 4 days (d) after one week
and (e) after releasing the CO pressure.
NEt3 and used in the methoxycarbonylation of norbornene under
the conditions previously described (Table 4). However, no con-
version was again observed, showing that the presence of HBr
did not influence the catalytic behaviour of this system under
these conditions. When the synthesis of PdCl2(7) was attempted
using the purified ligand 7, the 31P{1H} spectrum of the result-
ing species showed a singlet signal at δ 52.7 ppm, which was
very similar to the signal observed for the pincer complex, 7a,
(δ 52.1). Very similar NMR features were previously reported for
analogous Pd bromide and chloride pincer complexes.26 It was
therefore concluded that the analogue pincer complex containing
a chloride ligand was formed.
High pressure NMR (HP-NMR) studies
Reactivity of complex 1a under methoxycarbonylation con-
ditions. To gain an insight into the activity of the palladium cat-
alytic system containing monophosphine 1 in the absence of
additional acid, the reactivity of this system in the presence of
MeOH, CO and norbornene was studied by HP-NMR
spectroscopy.
Next, the variation of the substrate concentration was studied
in the palladium-catalyzed methoxycarbonylation of norbornene
with complex [PdCl2(4)] (4a) as a catalyst precursor. The results
are listed in Table 6.
First, complex 1a was reacted in the presence of all the
reagents: norbornene (10 equiv.) and carbon monoxide (30 atm)
in a solvent mixture of CD3OD–THF. A 10 mm HP-NMR sap-
phire tube was therefore charged with a solution of complex
[PdCl2(1)2] (1a) in a mixture of CD3OD–THF (ratio 1 : 1 by
volume). An initial 31P{1H} NMR spectrum was acquired at
room temperature before pressurizing with CO. In the spectrum,
a signal with singlet multiplicity at δ 22.7 corresponding to the
starting material, [PdCl2(1)2] 1a, was readily detected (Fig. 4a).
When norbornene (10 equiv.) was added and the tube was
charged with 30 atm of CO, the 31P{1H} NMR spectrum
acquired at room temperature showed the presence of two new
signals: a singlet resonance at δ 22.1 and a broad singlet at δ
0.79 (Fig. 4b). The starting material, 1a, was also detected. At
longer reaction times, the disappearance of the signal corre-
sponding to the starting material, 1a, and an increase in the
intensity of the new signals at δ 22.1 and 0.79 was observed
When the catalytic reaction was carried out in the presence of
50 equiv. norbornene, an excellent conversion (99%) and
chemo- (>99%) and stereoselectivities (>99% exo) were obtained
(Table 6, entry 1). Identical results were achieved when the ratio
of NBN : Pd was increased from 50 to 100 (Table 6, entry 2).
However, when the ratio of NBN : Pd was increased from 100 to
200, the conversion drastically decreased (23%), although both
the chemo- and stereoselectivity remained excellent (Table 6,
entry 3). When the ratio of NBN : Pd was 500, practically no cat-
alytic activity was observed (<5%) (Table 6, entry 4). Such a
striking drop in the conversion was unexpected and could be due
to the formation of inactive Pd species in the presence of a large
excess of NBN, thus deactivating the catalyst.
6984 | Dalton Trans., 2012, 41, 6980–6991
This journal is © The Royal Society of Chemistry 2012

View Online
Table 7 The effect of the palladium precursor using ligand 1 in the
palladium-catalyzed methoxycarbonylation of norbornenea
c
Entry Catalyst
HCl (1 equiv.) Cb, (%) Chemob (%) % exo(ee)
1
2
3
4
5
[PdCl2)]/1
NO
98
0
65
0
>99
0
>99
0
>99(10)
0
>99(7)
0
[Pd2(dba)3]/1 NO
[Pd2(dba)3]/1 YES
Pd(OAc)2/1 NO
Pd(OAc)2/1 YES
>99
>99
>99(9)
a Reaction conditions: Pd (0.021 mmol), L/Pd: 2, MeOH–THF (1 : 1),
30 bars of CO, 70 °C, 24 h. b All conversions and chemoselectivities
towards the ester were determined by GC and GC-MS. c C = conversion,
Chemo = selectivity to the ester.
Fig. 5 The reaction sequence by 31P{1H} NMR spectroscopy of 1a in
CD3OD–THF + 30 atm CO at variable temperatures: (a) at RT, (b) after
45 min at 343 K, (c) after 2 h at 373 K and (d) at 183 K.
However, when [Pd2(dba)3] (entry 2) and [Pd(OAc)2] (entry 4)
were tested in the presence of ligand 1 without any acid, no con-
version was observed. These results suggests that palladium
dichloride plays a crucial role in the activity and selectivity of
the reaction. To corroborate this observation, we carried out the
catalytic reaction using a [Pd2(dba)3]/1 system in the presence of
HCl (1 equiv.). A moderate conversion (65%) together with
excellent chemo- and exo-selectivities (>99%) were achieved
(Table 7, entry 3). Similarly, an excellent conversion (>99%)
together with excellent chemo- (>99%) and exo-selectivities
(>99%) were also obtained when Pd(OAc)2/1/HCl (1 equiv.)
(Table 7, entry 5) was used as a catalytic system.
The striking similarity of the results obtained using a Pd pre-
cursor containing chloride ligands without additional acid and
those achieved using a chloride-free precursor in the presence of
1 equiv. of HCl clearly indicated that this acid could be gener-
ated in situ during catalysis.
(Fig. 4c and d). When the CO pressure was released and a new
31P{1H} NMR spectrum was acquired at room temperature, the
signal corresponding to 1a was again detected (Fig. 4e), indicat-
ing a reversible process involving CO.
When the experiment was repeated using 13CO, two signals
were detected at δ 177 and 185 in the corresponding 13C{1H}
NMR spectra at room temperature. These signals were readily
assigned to the ester product and free CO, respectively. No other
signals could be detected in the carbonyl region of the spectrum
at that stage.
When complex 1a was placed under 30 atm of carbon monox-
ide in CD3OD–THF at room temperature, the same three signals
were detected by 31P NMR spectroscopy (Fig. 5a), demonstrat-
ing that the formation of the corresponding species did not
involve norbornene. After heating the sample to 343 K over
45 min, the signal corresponding to 1a was not detected, indicat-
ing total conversion of the starting complex (Fig. 5b). When the
sample was heated at 373 K, only the broad singlet at δ 0.79 was
detected at this temperature (Fig. 5c). However, when this reac-
tion mixture was cooled to room temperature, the signal at δ
0.79 resolved into two distinct resonances at δ 2.9 and −16.5
(Fig. 5d). The latter signal was readily assigned to the free
ligand, 1. The former signal resolved as a pseudo triplet
(1 : 1 : 1) with J = 72 Hz, suggesting a direct P–D coupling.
To check whether this signal arose from the deuterated phos-
phine 1-D+, the stability of the free ligand, 1, in MeOD was
investigated. However, no reaction was observed even when the
mixture was heated to 343 K over 24 h.
Previous reports have described the protonation of phosphine
ligands under methoxycarbonylation conditions.6,10 However,
the protonation of the phosphine ligands usually occurs by reac-
tion of the ligand with an excess of acid used to produce the
active species from the Pd precursors. In this system, no
additional acid was used. The detection of such a species there-
fore implied the in situ generation of acid from the catalytic
mixture. To investigate whether the acid produced involved the
metallic precursor, various palladium complexes were investi-
gated as precursors (Table 7).
In order to corroborate this hypothesis, the reactivity of ligand
1
in the presence of HCl was investigated by NMR
spectroscopy.
A sample of ligand 1 was thus dissolved in a mixture MeOH–
THF and the solution charged into a 5 mm NMR tube fitted with
a Young’s tap. In the corresponding 31P{1H} NMR spectrum,
the expected singlet signal at δ −16.5 was readily detected
(Fig. 6a). At this point, 1 equiv. of HCl was then added and a
new 31P{1H} NMR spectrum was acquired. The signal corre-
sponding to the free ligand was no longer present in the spec-
trum and only the previously detected signal at δ 0.79 was
observed (Fig. 6b). At 183 K, however, the signal for the free
ligand was again detected and the two signals at δ 2.9 and −16.5
(Fig. 6c) were observed as broad resonances.
1
When a H coupled 31P NMR spectrum was acquired at this
temperature, the 31P signal resonating at δ 0.79 exhibited a
doublet multiplicity (Fig. 6d) characteristic of a direct P–H coup-
ling (1JP–H = 506 Hz).27,28
Attempts to isolate the protonated phosphine by evaporation
of the solvent under reduced pressure failed to produce the proto-
nated phosphine, as ligand 1 was the only product obtained each
time. This indicated that an equilibrium between ligand 1 and its
protonated form, 1-H+, was taking place.
The species detected under CO pressure was therefore ident-
ified as the protonated phosphine, 1-H+, in equilibrium with
monophosphine 1 (Scheme 3). The detection of species 1-H+
under methoxycarbonylation conditions can be explained by the
As previously mentioned, when PdCl2/1 was used as a cataly-
tic system, excellent conversions and chemo- and exo-selectiv-
ities were obtained in the absence of any acid (Table 7, entry 1).
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6980–6991 | 6985

View Online
Scheme 4 Proposed structure for the complex [PdCl2(1)(CO)] 1b.
Fig. 6 The reaction sequence by 31P{1H} NMR spectroscopy of 1 in
MeOH–THF: (a) at RT, (b) + 1 equiv. HCl at RT, (c) + 1 equiv. HCl at
183 K and (d) 1H-coupled 31P NMR spectrum of 1 + 1 equiv. HCl at
183 K.
Scheme 5 The proposed mechanism for the formation of 1-H+.
Scheme 3 The reaction of ligand 1 in the presence of HCl (1 equiv.) in
MeOH–THF.
When the CO pressure was released from the NMR tube and a
new spectrum was acquired. The signal at δ 30.7 was not
detected, suggesting that the reaction of 1a in the presence of
CO to form the new product was reversible. This species was
identified as the complex [PdCl2(1)(CO)] 1b (Scheme 4).
Interestingly, the protonation of the phosphine was not
observed when the reaction was carried out in the absence of
methanol. The in situ generation of HCl was previously reported
by the reaction of a Pd complex containing chloride ligands in
methanol.29 We concluded that similar reactivity was taking
place in the presence of MeOH from the previously identified
species [PdCl2(1)(CO)] 1b, which is generated from complex 1a
under CO. A mechanism matching our observations in this study
is proposed in Scheme 5.
in situ production of HCl from the PdCl2 precursor, which reacts
with the free phosphine present in solution.
As previously mentioned, when the reaction of complex 1a in
the presence of norbornene and carbon monoxide (30 atm) was
performed in CD3OD–THF, no carbonyl containing species
could be detected. The rapid reaction of such species in the pres-
ence of methanol could explain this result. To confirm this
hypothesis, the reaction was repeated in neat THF-d8.
A solution of precursor 1a in a mixture of THF-d8 was pre-
pared and placed in a 10 mm high pressure (HP)-NMR tube.
After pressurizing the tube with 30 atm of CO enriched with
13CO, a 31P{1H} NMR spectrum was acquired at room tempera-
ture and the signal at δ 22.7 corresponding to the starting
material, 1a, was readily detected. Under these conditions, no
other signals were observed. However, at longer reaction times, a
new singlet signal was detected at δ 30.7. The detection of this
signal indicates that 1a was reacting under these conditions.
When the temperature was decreased to 193 K, the same signals
were observed, together with a broad resonance at δ −16.5,
which was readily assigned to the free ligand, 1. The detection
of free ligand 1 in solution suggested that a substitution reaction
had taken place. In the corresponding 13C{1H} NMR spectrum
acquired at room temperature, two signals were detected in the
carbonyl region of the spectrum at δ 185 and 183. The former
singlet signal was readily assigned to free CO. The latter signal
was detected as a broad singlet and was attributed to a carbonyl
ligand coordinated to palladium. The presence of the new 31P
signals at δ 30.7 suggested that the Pd–CO species contained a
phosphorus ligand. Furthermore, the broad singlet multiplicity of
the 13C signal suggested the cis arrangement of the CO and
phosphine ligand.
Such a formation of HCl from a Pd chloride species and an
alcohol has previously been described during the Wacker
process.30
Reactivity of complex 4a under methoxycarbonylation con-
ditions. A 10 mm HP-NMR sapphire tube was charged with a
solution of complex 4a in CD3OD–THF and a 31P{1H} spec-
trum was acquired (Fig. 7a). The signal corresponding to the
starting material, 4a, at δ 44.0 was readily detected. Two broad
signals with low intensity at δ 19.1 and 37.7 were also detected
at this stage. The corresponding species could not be identified
unambiguously. At this point, 10 equiv. of TFA, 50 equiv. of
norbornene and 30 atm of CO were added and the reaction was
monitored by 31P NMR spectroscopy. In the first 31P{1H} NMR
spectrum at room temperature (Fig. 7b), two signals at δ 36.9
and 38.1 with complex multiplicity suggesting the formation of
species containing
a direct P–D coupling were detected
(Fig. 7b). A broad resonance at 46.1 and a singlet resonance at δ
53.1 (major species) were also observed.
6986 | Dalton Trans., 2012, 41, 6980–6991
This journal is © The Royal Society of Chemistry 2012

View Online
Fig. 8 The reaction sequence by 31P {1H} NMR spectroscopy. (a)
Ligand 4 + 10 equiv. TFA in CD3OD–THF at RT, (b) 4 + 10 equiv. TFA
in CD3OD–THF at 213 K, (c) 4 + 10 equiv. TFA in CD2Cl2 at RT and
Fig. 7 The reaction sequence by 31P{1H} NMR spectroscopy. (a) 4a in
CD3OD–THF at RT, (b) 4a + 10 equiv. TFA + 50 equiv. NBN + 30 atm
of CO in CD3OD–THF at RT.
1
(d) an H-coupled 31P NMR spectrum of 4 + 10 equiv. TFA in CD2Cl2
at 193 K.
When the experiment was repeated using 13CO, in the corre-
sponding 13C{1H} NMR spectrum at room temperature, two
singlet resonances were detected at δ 185 and 177. The former
signal was readily assigned to free CO. This signal appeared as a
broad singlet suggesting some fluxional behaviour. The latter
signal corresponds to the ester product indicating that the meth-
oxycarbonylation reaction was taking place under these con-
ditions. No other carbonyl species were detected at this stage.
In order to identify the corresponding signals observed during
this experiment, the reactivity of complex 4a was investigated in
a step-wise manner with regards to TFA, CO and norbornene.
A solution of complex 4a in CD3OD–THF was prepared and
transferred to a 5 mm NMR tube and TFA (10 equiv.) was
added. A 31P{1H} NMR spectrum was acquired and the signal at
δ 44.0, corresponding to the starting material 4a, was readily
detected. Under these conditions, the previously detected singlet
resonance at δ 36.9 was observed with a high intensity, indicat-
ing that this signal arose from the reaction of complex 4a or
ligand 4 in the presence of TFA.
Scheme 6 The reactivity of ligand 4 in the presence of TFA in
CD3OD–THF.
These experiments showed that the previously observed reson-
ance at δ 36.9 was arising from the reaction product of ligand 4
and TFA, namely the diprotonated species, 4-(H+)2.
When complex 4a was placed under 30 atm of CO in the pres-
ence of TFA (10 equiv.) in CD3OD–THF at room temperature,
the previously observed signals at δ 46.1 (a broad multiplet),
44.0 (complex 4a), 38.1 (a broad multiplet) and 36.9 (diproto-
nated phosphine, 4-(H+)2) were detected in the corresponding
31P{1H}NMR spectrum. The resonance at δ 53.1 (s) was also
observed at this point, suggesting that the corresponding species
was formed in the presence of CO. When the CO pressure was
removed, the signal at δ 53.1 (s) was not observed anymore.
This indicated that the formation of this species involved a
reversible reaction with CO. When the experiment was repeated
using 13C-enriched CO, in the corresponding 13C{1H}NMR
spectrum recorded at 213 K, only the singlet resonance for free
CO was detected at δ 185. The broadness of the signal suggested
a fluxional behaviour that could be explained by the reversible
formation of species 4b containing a CO ligand rapidly exchan-
ging with free CO (Scheme 7).
As previously mentioned, when the reactivity of complex 4a
under methoxycarbonylation conditions was studied in CD3OD–
THF, the formation of the diprotonated phosphine, 4-(H+)2, was
evidenced. Interestingly, it was observed that the intensity of the
signal corresponding to this species decreasing during the exper-
iment, which suggested that this species may react during the
catalytic reaction. To confirm this hypothesis, species 4-(H+)2
was tested as an acid and ligand source in the Pd-methoxycarbo-
nylation of norbornene using [PdCl2(COD)] and [Pd(OAc)2] as
palladium precursors with 30 bars of CO pressure at room temp-
erature over 24 hours (Scheme 8). The catalytic results suggested
To corroborate this hypothesis, a sample of ligand 4 was dis-
solved in a CD3OD–THF mixture (ratio 1 : 1 by volume) and
charged into a 5 mm NMR tube fitted with a Young’s tap. When
a
31P{1H} NMR spectrum was acquired, one singlet resonance
at δ 22.8, which can be readily assigned to free ligand 4, was
detected. When TFA (10 equiv.) was added, the singlet signal at
δ 36.9 was observed as the only signal at room temperature
(Fig. 8a). When a 31P{1H} NMR spectrum was acquired at
193 K, this signal appeared as a pseudo triplet with a 1 : 1 : 1
intensity pattern (Fig. 8b). The multiplicity of this signal indi-
cates the formation of a species containing deuterium with a
P–D bond (JP–D = 69 Hz), suggesting that H/D exchange had
taken place.
The experiment was then repeated in a CH3OH–THF mixture
and, after evaporation of the solvent, the residue was re-dissolved
in CD2Cl2. In the 31P{1H} NMR spectrum at 193 K, the signal
at δ 36.9 was detected as a single resonance (Fig. 8c). When the
1H-coupled 31P NMR spectrum was acquired at 193 K, this peak
exhibited a doublet multiplicity JP–H = 460 Hz (Fig. 8d), which
is characteristic of a compound containing a P–H moiety. The
species corresponding to this signal was therefore identified as
the
diprotonated
diphosphine
[Fe(η5-C5H5)-(η5-C5H3)-
(+H(tBu2)2P)2] 4-(H+)2 (Scheme 6).
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6980–6991 | 6987

View Online
Scheme 7 The proposed structure for species 4b.
Fig. 9 The reaction sequence by 31P{1H}NMR spectroscopy. The reac-
tivity of diprotonated diphosphine 4-(H+)2 and [PdCl2(COD)] in
CH3OH–THF-d8. (a) The initial spectrum at 213 K, (b) at 233 K after a
few minutes, (c) at 253 K after a few minutes, and (d) the spectrum after
increasing the temperature to RT.
Scheme 8 Pd-catalysed methoxycarbonylation of norbornene using
the [PdCl2(COD)]/4-(H+)2 system as a catalytic precursor.
total conversion (>99%) and high chemoselectivity (85%) when
[PdCl2(COD)] was used as a precursor. Interestingly, when [Pd-
(OAc)2] was used as a Pd precursor, no conversion was obtained,
indicating that the presence of chloride in the solution is crucial
to obtain catalytic activity with this system.
When the catalytic reaction was performed using complex 4a
in the presence of only 2 equiv. of TFA, identical results for the
conversion (>99%) and similar results in terms of the chemos-
electivity (>99%) to those achieved with the [PdCl2(COD)]/
diprotonated diphosphine catalytic system were obtained. In
view of these results, the reactivity of the Pd precursor in the
presence of 4-(H+)2 was investigated by NMR spectroscopy.
First, the stability of the diprotonated diphosphine, 4-(H+)2, in
CH3OH–THF-d8 was examined. A solution of the diprotonated
diphosphine, 4-(H+)2, in CH3OH–THF-d8 was charged into a
5 mm NMR tube fitted with a Young’s tap. The broad singlet at
δ 36.9 corresponding to the starting material, 4-(H+)2, was
readily detected in the corresponding 1H and 31P{1H}NMR
spectra acquired at room temperature. After several hours at this
temperature, no new signals were observed, indicating that com-
pound 4-(H+)2 was stable under these conditions.
Then, in a 5 mm NMR tube fitted with a Young’s tap, a sol-
ution of the diprotonated diphosphine, 4-(H+)2, and
[PdCl2(COD)] (1 equiv.) in CH3OH–THF-d8 was prepared at
213 K and the reaction was monitored by 1H and 31P NMR spec-
troscopy in a pre-cooled spectrometer at 213 K. In the first 31P
{1H} NMR spectrum, the resonance at δ 36.9, which corre-
sponds to the diprotonated diphosphine 4-(H+)2 (Fig. 9a), was
the only signal detected at this stage. When the temperature of
the sample was increased to room temperature, the solution
rapidly darkened and several resonances attributed to decompo-
sition products were observed by NMR spectroscopy.
1
Fig. 10 The hydride region of the H NMR spectrum of a solution of
diprotonated diphosphine 4-(H+)2 and [PdCl2(COD)] in CH3OH–THF-
d8 at 233K.
Fig. 11 The proposed structure of species 4c.
phosphorus centres. The values of these coupling constants are
b
a
characteristic of trans (2JP -H = 193 Hz) and cis (2JP -H = 19 Hz)
couplings. The corresponding species was therefore identified as
the diphosphine Pd hydride species 4c (Fig. 11).
No information on the nature of the remaining X ligand could
be obtained by NMR techniques at this point. Interestingly, in
the H spectrum, a singlet resonance at δ 4.5 was also observed
1
and attributed to H2.
The sample was then slowly heated to 253 K and a new
31P{1H} NMR spectrum was acquired. At this temperature, the
2
signals at δ 34.0 and 41.2 (dd, JP–P = 18 Hz) were detected
together with a singlet signal at δ 45.6. This signal was attributed
to complex 4a, although the formation of an intermediate
species can not be discarded. All the signals remained
unchanged at this temperature (Fig. 9c). However, the signal of
the diprotonated phosphine, 4-(H+)2, was not detected at this
point. When the sample was slowly heated to room temperature,
the signals for the species, 4c, were not detected, which is in
agreement with the previous results. At this temperature, signals
for 4a and 4-(H+)2 were readily detected (Fig. 9d).
The experiment was therefore repeated at a low temperature.
At 233 K, two new signals were observed at δ 34.0 and 41.2 as
2
two mutually coupled doublets (dd, JP–P = 18 Hz) (Fig. 9b). In
the corresponding 1H NMR spectrum, a hydride signal was
2
a
detected as a doublet of doublets at δ −11.0 (dd, JP -H = 19 Hz
b
and 2JP -H = 193 Hz) (Fig. 10).
The coupling pattern of this signal clearly indicates that the
corresponding hydride complex contained two inequivalent
6988 | Dalton Trans., 2012, 41, 6980–6991
This journal is © The Royal Society of Chemistry 2012

View Online
In order to gain more information about the other unknown X
ligand in complex 4c, the hydride species [PdH(MeOH)(4)]-
[TfO] was synthesised according to a previously reported proce-
dure6b and its NMR features were compared to those of 4c. The
31P{1H} NMR spectrum of [PdH(MeOH)(4)][TfO] presented
two mutually coupled 31P signals at δ 30.1 (d, JP–P = 21.7 Hz)
and 76.1 (d, JP–P = 21.7 Hz), which disagrees with the signals
observed for species 4c. It was therefore concluded that species
4c does not contain a MeOH ligand and the nature of the
unknown ligand is proposed to be a chloride, which remains
from the Pd dichloride precursor (Fig. 11).
Substances Naturelles C.N.R.S., and Dr Graham Eastham from
Lucite International, respectively. PdCl2 was purchased from
Johnson Matthey Inc. and used without further purification. All
other reagents were used as received from commercial suppliers.
Deuterated solvents used for routine NMR measurements were
dried over molecular sieves. 1H, 13C{1H} and 31P{1H} NMR
spectra were recorded on a Varian Mercury 400 spectrometer
(400.14, 100.63 and 161.98 MHz, respectively). Chemical shifts
were referenced to either TMS as an internal standard (1H, 13C-
{1H} NMR spectra) or 85% H3PO4 as an external standard (31P-
{1H}NMR spectra). Gas chromatography analyses were per-
formed using a Hewlett–Packard 5890 series II chromatograph
with a flame ionization detector and Ultra-2 (5% diphenylsili-
cone, 95% dimethylsilicone) (25 m × 0.2 mm Ø) capillary
column. Enantiomeric excesses were determined by GC analysis
(Chiraldex-GTA capillar column 30 m × 0.25 mm × 0.12 μm
film thickness).
Conclusions
A series of palladium complexes bearing monodentate and
bidentate phosphine ligands were synthesised and used as cata-
lyst precursors for the methoxycarbonylation of norbornene.
The new complexes, [PdCl2(4)] 4a and [PdBr(7)] 7a, were
characterised by multinuclear NMR spectroscopy and X-ray
crystallography.
HP-NMR measurements
In the methoxycarbonylation of norbornene, ligands 1, 3 and
4 afforded excellent conversions (>99%) and selectivities toward
the ester formation (>99%).
In a typical experiment, the NMR tube was charged under N2
with
a
solution containing the palladium precursor
(0.021 mmol), TFA (when needed, 0.210 mmol) and norbornene
(1.05 mmol) in a mixture of CD3OD–THF (ratio 1 : 1 by
volume) (2 ml) as the solvent. The tube was then pressurized
with CO to the desired pressure.
Excellent conversions and chemo- and exo-selectivities were
achieved without any additional acid at room temperature using
[PdCl2(1)2] 1a as a catalytic precursor. NMR investigations
showed that the protonated phosphine, 1-H+, is formed under
catalytic conditions by the reaction of free ligand 1 and in situ
generated HCl. The addition of acid is thus not required with this
system since the reaction of the precursor with methanol under
CO pressure produces 2 equiv. of HCl and leads to the formation
of the active species.
The catalytic system containing ligand 4 [PdCl2(4)] afforded
an excellent conversion and chemo- and exo-selectivities at room
temperature under acidic conditions. The protonation of ligand 4
under methoxycarbonylation conditions was observed and the
isolation and characterization of the diprotonated diphosphine
was carried out. This diprotonated compound was tested under
Crystal data for structures [PdCl2(4)] (4a·THF) and [PdBr(7)]
(7a). Data collection for the crystal structures of compounds
4a·THF and 7a was carried out at 293(2) K on an Enraf Nonius
CAD4 single crystal diffractometer (Mo-Kα radiation, λ =
0.71073 Å). Cell refinement, indexing and scaling of all the data
set were performed using the Denzo and Scalepack programs.33
The structure was solved by direct methods and subsequent
Fourier analyses31 and refined by the full-matrix least-squares
method based on F2 with all observed reflections.34 The THF
molecule appears slightly disordered in 4a leading to unreliable
C–C distances. The large thermal motion (or disorder) in the
THF lattice molecule of 4a and in one cyclopentyl ring of 7a led
to short C(sp3)–C(sp3) distances (in the range of 1.34–1.41 Å).
All the calculations were performed using the WinGX System,
Version 1.80.05.35
catalytic conditions as
a ligand and acid source with
[PdCl2(COD)] as a precursor and an excellent conversion and
high selectivity toward the ester were achieved. Once more, the
presence of chloride ligands in the precursor was revealed to be
crucial to achieve a high level of activity and selectivity. The
reaction of the diprotonated ligand with [PdCl2(COD)] evi-
denced the formation of the Pd hydride species, [PdCl(H)(4)].
4a·THF: C28H48Cl2FeP2Pd·C4H8O, M = 751.86, monoclinic,
space group P21/n, a = 11.558(3), b = 14.646(3), c = 20.721(4)
Å, β = 94.29(3)°, V = 3497.8(13) Å3, Z = 4, final R1 = 0.0510,
wR2 = 0.1159 for the observed data I > 2σ(I), R1 = 0.1008, wR2
= 0.1354 (all data), S = 0.894 for 352 parameters and 6015
reflections, residuals in ΔF map 0.668, −0.406 e Å−3
.
Experimental
7a: C28H43BrP2Pd, M = 627.87, orthorhombic, space group
Pbca, a = 15.151(3), b = 17.180(4), c = 21.361(4) Å, V = 5560
(2) Å3, Z = 8, final R1 = 0.0395, wR2 = 0.0959 for the observed
data I > 2σ(I), R1 = 0.0852, wR2 = 0.1079 (all data), S = 0.860
for 289 parameters and 5864 reflections, residuals in ΔF map
General methods
All palladium complexes were synthesized using standard
Schlenk techniques under a nitrogen atmosphere. Diethyl ether
and THF were distilled over sodium benzophenone and dichloro-
methane was distilled over P2O5. All solvents were deoxyge-
nated before use. Ligand 322 and the palladium complexes
0.382, −0.697 e Å−3
.
[PdCl2(NCPh)2],31
[PdCl2(COD)],32
[PdCl2(1)2]
(1a),21
MS analysis
[PdCl2(3)](3a)6c and [PdCl2(5)] (5a),22 were prepared according
to the literature methods. Ligands 2 and 4 were generously sup-
plied by Dr Angela Marinetti from the Institut de Chimie des
The MS spectrum was obtained by direct injection on an 6210
LC/TOF mass spectrometer (Agilent Technologies, Palo Alto,
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6980–6991 | 6989

View Online
USA), equipped with an electrospray ionisation source. MS
source parameters were set with a gas temperature of 250 °C, a
drying gas flow of 5 l min−1, a nebulizer gas pressure of 12 psi
and a capillary voltage of 3500 V. Data acquisition was realised
both in the negative and positive centroid acquisition mode, with
a skimmer voltage of 65 V, a fragmentor voltage of 150 Vand an
acquisition range from 50 to 1500 m/z, at a rate of 1.02 spectra
per sec. The instrument was calibrated in the positive and nega-
tive ion modes using a tuning mixture and an internal reference
standard was used to calibrate the exact mass during the analysis
(both from Agilent Technologies, Palo Alto and USA). The
average of the scans obtained were, in the positive mode,
the protonated ion (m/z 443.3010) and, in the negative mode, the
bromide ion with m/z values of 78.9200 and 80.9178.
to a solution of [PdCl2(COD)] (146.00 mg, 0.51 mmol) in
dichloromethane (5 ml). The resulting red solution was stirred
for 1 h and concentrated under reduced pressure. Addition of
diethyl ether led to the precipitation of an orange solid, which
was filtered off, washed with diethyl ether and dried under
vacuum. The orange solid was recrystallized from CH2Cl2–ether
to obtain complex 4a as red crystals. Yield: 300.5 mg (82%).
1H NMR (CD2Cl2, 400.14 MHz, ppm): δ 1.48 (d, 18H, JHP
=
13.6 Hz, t-Bu), 1.51 (d, 18H, JHP = 13.6 Hz, t-Bu), 3.16 (dd,
2H, JHH = 4.8 Hz, JHP = 14.4 Hz, RCH2P), 3.44 (m, 2H,
RCH2P), 4.10 (s, 5H, Cp-ring), 4.18 (m, 3H, C5H3-ring), 4.50
(d, 1H, JHH = 4.0 Hz, C5H3-ring). 13C{1H} NMR (CD2Cl2,
100.63 MHz, ppm): δ 25.0 (m, CH2), 30.0 (m, CH3), 30.8 (m,
CH3), 31.8 (s, tert-C), 32.3 (s, tert-C), 66.3 (s, CH), 70.7 (s,
Cp), 71.5 (s, Cp subst), 82.3 (s, C quat.). 31P{1H} NMR
(CD2Cl2, 161.98 MHz, ppm): δ 44.0 (s). Anal. Calc. for
C28H48FeP2PdCl2 (679.79 g mol−1): Calc.: C, 49.42; H, 7.06%.
Found: C, 48.31 H, 7.00%.
Catalysis
High-pressure experiments were carried out in a Berghof auto-
clave and the reaction mixtures were magnetically stirred and
electrically heated. In a typical experiment, a solution of the pal-
ladium precursor (0.021 mmol), TFA (0.210 mmol) and norbor-
nene (1.05 mmol) in 5 ml of a mixture of THF–CH3OH (1 : 1)
were introduced into the evacuated autoclave. Carbon monoxide
was introduced and the system was then heated. When a thermal
equilibrium was reached, stirring was initiated. After reaction,
the autoclave was cooled to room temperature and depressurized.
The product was filtered in a short column of celite and the
solvent was removed under vacuum. The conversions and
chemo- and stereo-selectivities were determined by GC, GC-MS
and NMR analyses.
Synthesis of diprotonated phosphine 4-(H+)2. Bidentate ligand
4 (65.1 mg, 0.129 mmol) was dissolved in a mixture MeOH–
THF (ratio 1 : 1 by volume) (2 ml) and TFA (100 μl, 1.29 mmol)
was added to the solution. The reaction was stirred for a few
minutes at room temperature. Then, the yellow solution was con-
centrated under vacuum and diethyl ether was added to precipi-
tate compound 4-(H+)2 as a yellow powder. Yield: 62.5 mg
(96%).
1H NMR (THF-d8, 400.14 MHz, ppm): δ 1.3 (d, 18H, JHP
=
13.2 Hz, CH3), 1.4 (d, 18H, JHP = 13.2 Hz, CH3), 3.3 (d, 4H,
JHH = 7.2 Hz, CH2), 4.0 (t, 1H, JHH = 2.8 Hz, C5H3-ring), 4.1
(s, 5H, C5H3-ring), 4.5 (d, 2H, JHH = 2.4 Hz, C5H3-ring). 13C
{1H} NMR (THF-d8, 100.63 MHz, ppm): δ 29.4 (d, JPC = 9.1
Hz, CH3), 20.0 (brs, CH2), 33.4 (s, tert-C), 33.70 (s, tert-C),
66.6 (s, CH, Cp), 71.3 (s, CH, Cp), 71.8 (m, CH, Cp), 83.9 (m,
C, Cp). 31P{1H} NMR (THF-d8, 161.98 MHz, ppm): δ 36.9 (s).
Calc. for C28H50FeP2 (504.49 g mol−1): Calc.: C, 66.66; H,
9.99%. Found: C, 67.53; H, 10.83%.
Synthesis of [PdCl2(1)2] (1a). The synthesis of this compound
was carried out according to a previous report21 and the NMR
characterization was completed.
A solution of ligand 1
(173.48 mg, 0.53 mmol) in dichloromethane (5 ml) was added
to a solution of [PdCl2(PhCN)2] (100.00 mg, 0.26 mmol) in
dichloromethane (3 ml) at room temperature. The resulting
yellow solution was stirred for 1 h and concentrated under
vacuum. The addition of diethyl ether yielded the precipitation
of a pale yellow solid, which was filtered and washed with
diethyl ether before being dried under vacuum. Yield: 166.3 mg
(75%).
Synthesis of [PdBr(7)] (7a). A solid sample of
[PdCl2(PhCN)2] (141.47 mg, 0.37 mmol) was added to a sol-
ution of ligand 7 (217.51 mg, 0.49 mmol) in 2-methoxyethanol
(10 ml). The yellow solution was refluxed for 30 min, then
allowed to cool and the solvent was removed to leave a yellow
powder. Extraction with hot ethanol and cooling followed by
reduction of the volume gave a white solid. The solid was recrys-
tallized from a mixture of CH2Cl2–pentane to obtain complex 7a
as colourless crystals. Yield: 185.3 mg (60%).
1H NMR (CD2Cl2, 400.14 MHz, ppm): δ 0.65 (d, JHH = 6.8
Hz, CH3), 1.02 (d, JHH = 6.0 Hz, CH3), 1.06 (d, JHH = 7.2 Hz,
CH3), 1.15 (m, CH2), 1.45 (m, CH2), 1.55 (m, CH2), 1.63 (m,
CH2), 1.69 (m, CH2), 1.85 (s br. CH), 2.55 (m, CH), 2.61 (m,
CH), 3.70 (m, P–CH), 7.31–7.44 (m, Ar), 7.62–7.69 (m, Ar).
13C{1H} NMR (CD2Cl2, 100.63 MHz, ppm): δ 18.76 (s, CH3),
21.32 (brs, CH2), 21.43 (s, CH3), 25.17 (s, CH3), 28.6 (t, JPC
=
1H NMR (CD2Cl2, 400.14 MHz, ppm): δ 1.58 (m, 8H,
–CH2–), 1.72 (m, 8H, –CH2–), 1,85 (m, 8H, P–CH2), 2,06 (m,
8H, P–CH2), 2.47 (m, 4H, P–CH), 3.23 (t, J = 11.6 Hz, 4H,
P–CH2Ar), 6.90–7.30 (m, 3H, Ar). 13C{1H} NMR (CD2Cl2,
100.63 MHz, ppm): δ 26.20 (m, CH2), 26.54 (m, CH2), 28.90 (s,
CH2), 29.34, (s, CH2), 35.50 (t, J(C–P) = 9.0 Hz, P–CH2Ar),
35.58 (d, J(C–P) = 3.0 Hz, P–CH), 122.41, 122.51, 122.62,
124.91, 150,41, 16 102. 31P{1H} NMR (CD2Cl2, 161.98 MHz,
ppm): δ 52.1 (s). Anal. Calc. for C28H43P2PdBr (627.91 g
mol−1): Calc.: C, 53.56; H, 6.90%. Found: C, 54.16; H, 6.72%.
8.0 Hz, CH), 29.73 (brs, CH2), 31.66 (t, JPC = 6.8 Hz, CH), 32.8
(m, CH), 32.9 (s, CH2), 33.08 (s, CH2), 40.37 (s, CH), 127.6 (t,
JPC = 10.1 Hz, Ar), 128.6 (t, JPC = 8.5 Hz, Ar), 13 009 (s, Ar),
13 066 (s, Ar), 133.41 (t, JPC = 10.4 Hz, Ar), 137.03 (t, JPC
=
11.1 Hz, Ar). 31P{1H} NMR (CD2Cl2, 161.98 MHz, ppm): δ
22.7 (s). Anal. Calc. for C44H58P2PdCl2 (826.20 g mol−1):
Calc.: C, 63.96; H, 7.08%. Found: C, 62.53; H, 6.92%.
Synthesis of [PdCl2(4)] (4a). A solution of ligand
4
(269.70 mg, 0.54 mmol) in dichloromethane (10 ml) was added
6990 | Dalton Trans., 2012, 41, 6980–6991
This journal is © The Royal Society of Chemistry 2012

View Online
12 B. K. Muñoz, C. Godard, A. Marinetti, A. Ruiz, J. Benet-Buchholz and
C. Claver, Dalton Trans., 2007, 5524.
Acknowledgements
13 (a) C. Botteghi, G. Consiglio and P. Pino, Chimia, 1973, 27, 477;
(b) T. Hayashi, M. Tanaka and I. Ogata, Tetrahedron Lett., 1978, 19,
3925; (c) T. Hayashi, M. Tanaka and I. Ogata, J. Mol. Catal., 1984, 26,
17.
We are grateful to the Spanish Ministerio de Educación (Ramón
y
Cajal to C.G., CTQ2010-15835, Consolider Ingenio
CSD2006-0003) and the Universitat Rovira i Virgili (fellowship
14 J. Gironés, J. Duran, A. Polo and J. Real, Chem. Commun., 2003, 1776.
15 (a) C. Blanco, A. Ruiz, C. Godard, A. Marinetti, N. Fleury-Brégeot and
C. Claver, Adv. Synth. Catal., 2009, 351, 1813; (b) S. Oi, M. Nomura,
T. Aiko and Y. Inoue, J. Mol. Catal. A: Chem., 1997, 115, 289;
(c) H. Zhou, S. Lu, J. Hou, H. Chen, H. Fu and H. Wang, Chem. Lett.,
1996, 339; (d) J. Hou, H. Zhou, L. Chen and Q. Ou, Anal. Lett., 1996, 29
(15), 2755.
16 (a) G. Fráter, J. A. Bajgrowicz and P. Kraft, Tetrahedron, 1998, 54, 7633;
(b) G. Buchbauer, I. Stappen, C. Pretterklieber and P. Wolschann,
Eur. J. Med. Chem., 2004, 39, 1039.
17 W. Clegg, G. R. Eastham, M. R. J. Elsegood, R. P. Tooze, X. L. Wang
and K. Whiston, Chem. Commun., 1999, 1877.
18 K. Kudo, M. Hidal, T. Murayama and Y. Uchida, J. Chem. Soc. D, 1970,
1701b–1702.
19 J. Grévin and P. Kalck, J. Organomet. Chem., 1994, 476, c23.
20 K. Nozaki, M. L. Kantam, T. Horiuchi and H. Takaya, J. Mol. Catal. A:
Chem., 1997, 118, 247.
21 W. de Graaf, J. Boersma, G. van Koten and C. J. Elsevier, J. Organomet.
Chem., 1989, 378, 115.
22 B. Sesto and G. Consiglio, Chem. Commun., 2000, 1011.
23 A. Voituriez, A. Panossian, N. Fleury-Brégeot, P. Retailleau and
A. Marinetti, Adv. Synth. Catal., 2009, 351, 1968.
24 R. J. Cross, A. R. Kennedy and K. W. Muir, J. Organomet. Chem., 1995,
487, 227.
25 F. Bachechi, Struct. Chem., 2003, 14, 263.
26 R. J. Cross, A. R. Kennedy and K. W. Muir, J. Organomet. Chem., 1995,
487, 227.
27 (a) C. W. McFarlane and R. F. M. White, J. Chem. Soc. D, 1969, 744;
(b) G. A. Olah and C. W. McFarlane, J. Org. Chem., 1969, 34, 1832.
28 P. Leoni, M. Sommovigo, M. Pasquali, S. Midollini, D. Braga and
P. Sabatino, Organometallics, 1991, 10, 1038.
29 M. Portnoy and D. Milstein, Organometallics, 1994, 13, 600.
30 R. F. Heck, Palladium Reagents in Organic Syntheses, Academic Press,
New York, 1985, p. 59.
to C.B.) for financial support.
References
1 (a) R. I. Pugh and E. Drent, Adv. Synth. Catal., 2002, 344, 837;
(b) E. Drent and P. H. M. Budzelaar, Chem. Rev., 1996, 96, 663;
(c) C. Bianchini, A. Meli and W. Oberhauser, Dalton Trans., 2003, 2627;
(d) I. del Río, C. Claver and P. W. M. N. van Leeuwen, Eur. J. Inorg.
Chem., 2001, 2719; (e) A. Sen, in Catalytic Synthesis of Alkene–Carbon
Monoxide Copolymers and Cooligomers, ed. A. Sen, Kluwer, Dordrecht,
vol. 27, 2003.
2 K. Gabor, Chem. Rev., 2001, 101, 3435.
3 G. R. Eastham, R. P. Tooze, X. L. Wang and K. Whiston, Int. Pat.,
9619434, 1996.
4 (a) C. Godard, A. Ruiz, M. Diéguez, O. Pàmies and C. Claver, in Cataly-
tic Asymmetric Synthesis, ed. I. Ojima, Wiley, 3rd edn, 2010, p. 799;
(b) E. Guiu, M. Caporali, B. Muñoz, C. Müller, M. Lutz, A. L. Spek,
C. Claver and P. W. N. M. van Leeuwen, Organometallics, 2006, 25,
3102.
5 (a) C. Bianchini and A. Meli, Coord. Chem. Rev., 2002, 225, 35;
(b) C. Bianchini, H. M. Lee, G. Mantovani, A. Meli and W. Oberhauser,
New J. Chem., 2002, 26, 387; (c) C. Bianchini, G. Mantovani, A. Meli,
W. Oberhauser, P. Brugeller and T. Stampfl, J. Chem. Soc., Dalton Trans.,
2001, 690.
6 (a) J. K. Liu, B. T. Heaton, J. A. Iggo and R. Whyman, Angew. Chem.,
Int. Ed., 2004, 43, 90; (b) W. Clegg, G. R. Eastham, M. R. J. Elsegood,
B. T. Heaton, J. A. Iggo, R. P. Tooze, R. Whyman and S. Zacchini, J.
Chem. Soc., Dalton Trans., 2002, 3300; (c) W. Clegg, G. R. Eastham,
M. R. J. Elsegood, B. T. Heaton, J. A. Iggo, R. P. Tooze, R. Whyman and
S. Zacchini, Organometallics, 2002, 21, 1832; (d) B. K. Muñoz,
E. Santos, C. Godard, E. Zangrando, C. Bo, A. Ruiz and C. Claver,
Eur. J. Inorg. Chem., 2008, 4625.
7 G. R. Eastham, R. P. Tooze, M. Kilner, D. F. Foster and D. J. Cole-Hamil-
ton, J. Chem. Soc., Dalton Trans., 2002, 1613.
8 P. W. N. M. van Leeuwen, M. A. Zuideveld, B. H. G. Swennenhuis,
Z. Freixa, P. C. J. Kamer, K. Goubitz, J. Fraanje, M. Lutz and
A. N. Spek, J. Am. Chem. Soc., 2003, 125, 5523.
9 G. R. Eastham, B. T. Heaton, J. A. Iggo, R. P. Tooze, R. Whyman and
S. Zacchini, Chem. Commun., 2000, 609.
10 V. de la Fuente, M. Waugh, G. R. Eastham, J. A. Iggo, S. Castillón and
C. Claver, Chem.–Eur. J., 2010, 16(23), 6919.
31 M. S. Kharasch, R. C. Seyler and F. R. Mayo, J. Am. Chem. Soc., 1938,
60, 882.
32 D. Drew and J. R. Doyle, Inorg. Synth., 1972, 13, 47.
33 Z. Otwinowski and W. Minor, Processing of X-ray Diffraction Data Col-
lected in Oscillation Mode, in Methods in Enzymology, ed. C. W. Carter
Jr. and R. M. Sweet, Academic Press, New York, 1997, vol. 276, pp.
307–326.
34 G. M. Sheldrick, SHELX97 Programs for Crystal Structure Analysis
(Release 97-2), University of Göttingen, Germany, 1998.
35 L. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
11 Y. Kawashima, K. Okano, K. Nozaki and T. Hiyama, Bull. Chem. Soc.
Jpn., 2004, 77, 347.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6980–6991 | 6991