Palladium catalyzed oxidative carbonylation of alcohols: Effects of diphosphine ligands

Article abstract of DOI:10.1039/c4cy01588f

The catalytic activity of a series of palladium diphosphine complexes of the type [PdX₂(P∩P)] has been studied in the oxidative carbonylation of i-PrOH with p-benzoquinone as an oxidant. Diphosphine ligands have been chosen in order to cover a wide range of bite angles and electronic and steric parameters. Their properties have been correlated with the catalytic activity and selectivity of the reaction. The best catalytic performance has been achieved with weakly coordinating anions as well as non-bulky and electron-donating P∩P ligands with a relatively wide bite angle yet capable of maintaining a cis-coordination, such as cis-[Pd(OTs)₂(pMeO-dppf)]. These results and those on the reactivity of dicarboalkoxy species of the type cis-[Pd(COOMe)₂(P∩P)] toward reductive elimination, which is a crucial step in oxalate formation, suggest that the slow step of the catalysis depends on the nature of the P∩P ligand.

Full text of DOI:10.1039/c4cy01588f

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Palladium catalyzed oxidative carbonylation of
alcohols: effects of diphosphine ligands†
Cite this: DOI: 10.1039/c4cy01588f
ab
bcd
b
Emanuele Amadio,‡* Zoraida Freixa,
Piet W. N. M. van Leeuwen
and Luigi Toniolo*a
The catalytic activity of a series of palladium diphosphine complexes of the type ijPdX
2
IJP∩P)] has been
studied in the oxidative carbonylation of i-PrOH with p-benzoquinone as an oxidant. Diphosphine ligands
have been chosen in order to cover a wide range of bite angles and electronic and steric parameters. Their
properties have been correlated with the catalytic activity and selectivity of the reaction. The best catalytic
performance has been achieved with weakly coordinating anions as well as non-bulky and electron-
donating P∩P ligands with a relatively wide bite angle yet capable of maintaining a cis-coordination, such
Received 2nd December 2014,
Accepted 12th March 2015
as cis-ijPdIJOTs)
type cis-ijPdIJCOOMe)
gest that the slow step of the catalysis depends on the nature of the P∩P ligand.
2
IJpMeO-dppf)]. These results and those on the reactivity of dicarboalkoxy species of the
DOI: 10.1039/c4cy01588f
2
IJP∩P)] toward reductive elimination, which is a crucial step in oxalate formation, sug-
www.rsc.org/catalysis
possible to reduce the propagation chain rate to the extreme
point that methyl propanoate is formed.
1
Introduction
6
As far as catalysis is concerned, palladium is one of the most
versatile elements of the transition metal series. In particular,
it catalyses a wide spectrum of carbonylation reactions, such
as the Reppe reaction with alkenes, alkynes, conjugated
Whereas carbonylation of alkenes has been thoroughly
investigated both in industry and academia, particularly for
6
,7
the synthesis of PKs and methyl propanoate, which is an
intermediate to methyl methacrylate (Lucite process), oxida-
tive carbonylation of alkanols has received less attention, in
spite of the interest in synthesising organic carbonates via a
1
dienes, and aryl-substituted alkenes, copolymerization of CO
and alkenes to alternating and nonalternating polyketones
2
8
(
PKs), oxidative carbonylation of alkenes, alkanols and
one step environmentally friendly, phosgene-free technology.
3
4
alkynes, and carbonylative cross-coupling, just to mention
some representative examples.
Another product of oxidative carbonylation of alkanols is the
corresponding oxalate, which is prepared in industry using a
9
The nature of the ligand plays a role of paramount impor-
Pd/C catalyst.
5
tance in governing the catalytic activity and selectivity, as
Even though the use of palladium catalysts for the synthe-
1
0
exemplified by carbonylation of ethene. In fact, by carefully
increasing the steric hindrance around the metal center of
catalysts suitable for the synthesis of PKs, it has been
sis of organic oxalates has been known since the seventies,
a
Department of Molecular Sciences and Nanosystems, Ca' Foscari University,
Dorsoduro 2137, 30123, Venice, Italy. E-mail: emanuele.amadio@unipd.it,
toniolo@unive.it
b
Institute of Chemical Research of Catalonia (ICIQ), Avda Països Catalans 16,
4
3007, Tarragona, Spain
c
Faculty of Chemistry, University of the Basque Country (UPV–EHU),
San Sebastián, Spain
d
IKERBASQUE, Basque Foundation for Science, Bilbao, Spain
†
Electronic supplementary information (ESI) available: Effect of bite angle on
activity or selectivity, experimental details and spectroscopic characterization of
P∩P, ijPdX IJP∩P)] and ijPdIJCOOMe) IJP∩P)], together with the carbonylation
procedures. See DOI: 10.1039/c4cy01588f
Current address: Department of Chemical Sciences, University of Padua, via
2
2
‡
Scheme 1 Carboalkoxy palladiumIJII) intermediates involved in the
Marzolo 1, 35131, Padua, Italy.
product forming steps.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
1
1
only a few examples have been reported later. More 2 Results and discussion
recently, the interest in this chemistry has received further
impulse using well defined PdIJII)-PPh systems in combina-
3
2
.1 Influence of the anion of the catalyst precursor
ijPdX IJdppf)] (dppf = 1,1′-bisIJdiphenylphosphino)ferrocene)
2
tion with a base, NEt , and 1,4-benzoquinone (BQ) as a stoi-
3
Chart 1 shows the results using the precursors ijPdX
2
IJdppf)]
chiometric oxidant. The main product, oxalate, is formed
together with carbonate, through the intermediacy of mono-
(
X or X = TsO,§ SO , AcO, C O , Cl). The highest TOF to O
2
4
2 4
1
1g,12
was achieved with the most weakly coordinating TsO anion.
This result is not surprising since strongly coordinating
anions compete with CO and alcohol for coordination, a pre-
requisite for the formation of the products. This behavior is
opposite to that observed with monophosphine-based precur-
sors ijPdX IJPPh ) ], in which the highest catalytic activity was
and di-carboalkoxy palladiumIJII) species (Scheme 1).
It was established that both BQ and the base play a role of
paramount importance in controlling the product distribu-
tion. Both favor the formation of a PdIJCOOR)2 species and
BQ, in addition to being an oxidant, promotes the reductive
2
3 2
elimination step of
oxalate.
a
dicarboalkoxy-PdIJII) species to
1
2i
1
2h
achieved with X = Br and the lowest with TsO.
−
With a strongly coordinating anion such as Cl both the
activity and selectivity toward the formation of O were signifi-
cantly suppressed, whereas the TOFs for the formation of C
and A were less influenced. Reasonable explanations are as
follows. The formation of the products occurs through a com-
mon Pd–OR intermediate (a) (Scheme 3). β-hydride elimina-
While oxidative carbonylation of alcohols has been
explored using PdIJII)-monophosphine complexes, only little
attention has been given to the use of diphosphine
1
2g
ligands, though their potentiality in governing the catalyst
5
,13
performance is well recognized.
This lack of knowledge
prompted us to study, for the first time in an exhaustive man-
ner, the performance of PdIJII)-diphosphine catalysts with the
aim of correlating their properties with the activity and selec-
tivity of the reaction. A large variety of diphosphine ligands,
chosen to cover a wide range of bite angles and electronic
and steric properties, has been used in order to establish a
correlation between these properties and the catalyst's activ-
ity and selectivity. The effect of the nature of the counter
anion and CO pressure on the catalytic outcome has also
been studied. Most of the catalytic experiments have been
carried out in isopropanol (iPrOH). Diisopropyl oxalate (O) is
formed as the major product, together with diisopropyl car-
bonate (C). Also, formation of acetone (A) from iPrOH occurs
as a side reaction (Scheme 2).
1
4
tion of Pd–OR gives the side product A and a Pd–H species,
which is oxidized by BQ, thus restoring the catalyst, or
deprotonated to Pd(0). If a migratory insertion of coordinated
CO into the Pd–OR bond takes place before β-hydride elimi-
nation, a monocarboalkoxy intermediate Pd–COOR (b) is
formed. Addition of the OR nucleophile to palladium
followed by reductive elimination rather than a direct nucleo-
philic abstraction process gives C and Pd(0), which is oxi-
dized by BQ to give the initial PdIJII). For the formation of O,
it is necessary that a dicarboalkoxy species (d) is formed
before the formation of C occurs. Moving from TsO to Cl
the activity toward C and A was little influenced, while a dra-
matic decrease of the TOF toward O was detected. This fact
suggests that once a Pd–COOR intermediate is formed, Cl
competes more strongly than TsO for the coordination of
1
5
−
−
−
The synthesis and reactivity of PdIJII)-dicarboalkoxy com-
plexes have been also investigated in order to determine the
relationship between the bite angle of the ligand and the
reactivity toward reductive elimination to oxalate and “PdIJ0)-
IJP∩P)”, which is a crucial step in the “oxalate catalytic cycle”.
−
another molecule of ROH or CO necessary to give the
PdIJCOOR)2 species leading to O. This might be due to the
electron withdrawing properties of the COOR ligand, which
−
makes the displacement of Cl more difficult. In Scheme 3,
Pd(0) may be coordinated by diphosphine and BQ, similar to
what was found using PdIJII)-monophosphine catalysts, for
which a mechanism for the reoxidation of Pd(0) was
1
2h,i
proposed.
2.2 Effect of the pressure of carbon monoxide
Chart 2 shows that upon increasing the pressure of CO, the
−
1
activity and selectivity toward O increase (TOF = 245 h and
3% selectivity), with a concomitant decrease in the forma-
9
tion of A. Instead, the formation of carbonate is little
influenced and remains low (maximum selectivity 3%).
These trends suggest that the formation of O is connected
to that of A and that they hardly interfere with the formation
of C. As already mentioned, acetone is formed through
β-hydride elimination of the Pd–OR species (a) (Scheme 3).
Alternatively, this species undergoes CO insertion giving
monocarboalkoxy intermediates (Pd–COOR) b and b′. Since
species b is the most abundant, its concentration is not
Scheme 2 Products formed in palladium-catalyzed oxidative carbon-
ylation of iPrOH.
§
2
This precursor is the cationic aquo complex cis-ijPdIJOTs)IJH O)IJdppf)]IJTsO).
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
−
4
−1
Chart 1 Effect of the anion on the catalytic activity using cis-ijPdX
atm, T = 80 °C, 1 h, 5 mL of anhydrous iPrOH.
2
IJdppf)]. Conditions: [Pd] = 2 × 10 mol L , Pd/BQ/NEt
3
= 1/700/2, PCO = 80
Scheme 3 Proposed catalytic cycles for the formation of O, C and A (R = iPr).
much affected by CO pressure. b interacts with ROH giving
the intermediate PdIJCOOR)IJOR) (c), which forms C and Pd(0)
in a relatively slow step. Species b′, in which CO is already
coordinated, interacts with ROH giving intermediate
PdIJCOOR)2 (d) which yields O and Pd(0). The CO pressure
influences significantly the equilibrium between a and b′, so
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Chart 2 Effect of PCO on the activity of the oxidative carbonylation
−4
reaction using cis-ijPdIJOH
2
)IJOTs)IJdppf)]IJTsO). Conditions: [Pd] = 2 × 10
−1
mol L , Pd/BQ/NEt
3
= 1/700/2, 80 °C, 1 h, 5 mL of anhydrous iPrOH.
that, upon increasing the CO pressure, the selectivity toward
O increases at the expense of A.
2
.3 Influence of the diphosphine ligand
As already mentioned, the main goal of this investigation was
to correlate the catalytic performance with ligand features. To
this end, several diphosphine ligands P∩P were selected, cov-
ering a wide range of bite angles and electronic and steric
properties. The selected P∩P ligands span from typically
cis-chelating dppe to trans-chelating SPANPhos (Fig. 1).
When chelating diphosphine-based catalysts are used, the
bite angle is a key parameter since it may influence signifi-
cantly the electronic and steric properties and therefore the
Fig. 1 Diphosphine ligands used in this study. The natural bite angle
(β ) is taken from ref. 16. Extracted from the X-ray structure of trans-
n
17
2
ijPdCl IJSPANphos)].
5
,13
reactivity of the metal center.
In more detail, an increment
of the bite angle can give two distinct effects: i) it increases
the effective steric bulk and ii) it electronically favours or
disfavours certain geometries especially the ones involved in
1
8
reductive elimination. For instance, in the case of palla-
dium complexes, when ligands with small bite angles (dppe,
dppp and dppb) are used, the complexes typically show a sta-
ble cis coordination to PdIJII), while wider bite angles lead to
1
9
rapid reductive elimination. Also, for larger bite angle
diphosphines, such as DPEphos or Xantphos, the cis species
might eventually get involved in a cis–trans isomerisation
equilibrium promoted by the formation of a palladium–
1
5,20
oxygen bond.
By contrast, an extreme case is observed with SPANphos,
typically considered trans diphosphine, which forms
cis-complexes only when cis-enforcing conditions are used.
a
2
1
These different ligand coordination features should therefore
influence the reactivity of the resulting palladium complexes.
In Chart 3 (Table S1, ESI†), the dependence of the catalytic
performance on the ligand bite angle is presented.
Both the activity and selectivity toward oxalate are strongly
dependent on the bite angle. dppf presents the optimal value,
Chart 3 Effect of bite angle on the catalytic activity. Conditions: [Pd] =
−4
−1
2
× 10 mol L , Pd/BQ/NEt
3
= 1/700/2, PCO = 80 atm, T = 80 °C, 1 h,
5 mL of anhydrous iPrOH.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
whereas narrower or wider bite angles dramatically decrease
the catalytic performance. The trend observed is consistent
with the hypothesis that oxalate formation is favoured by
wide bite ligands, but it requires a stable cis geometry of the
intermediates involved in the reaction. Bite angles of around
ferrocenyl ligands (dippf, dcypf, dtbpf; entries 4–6 in Table 1).
However, it has been already proven that these strong Lewis
base ferrocenyl ligands may coordinate to a palladium centre
2
3
through both cis k -P,P and trans k -P,P,Fe bonding modes
and that this equilibrium is favoured when the electron-
withdrawing COOMe moiety is present in the complex as
1
00° appear to be the best compromise. Similar bite angle
2
5
effects, though only partially explored, were reported in the
literature when oxidative carbonylation of MeOH to dimethyl
shown in Scheme 4.
Since Pd–COOR type intermediates are involved in the
reaction mechanism, it is reasonable to presume that the
observed modest catalytic performance is mainly due to the
in situ formation of these more stable but less active trans
isomers.
1
2g
carbonate/oxalate with nitrobenzene as an oxidant
and cross-coupling and C–C reductive elimination reactions
1
3,16b,19
using palladium complexes were studied,
although for
these reactions slightly wider bite angles were optimal, e.g.
DPEphos or Xantphos.
Despite the low activity, it is interesting to observe that,
within the dippf, dcypf, and dtbpf series (entries 4–6 in
Table 1), the increase in the steric hindrance favours the for-
mation of acetone, via β-hydride elimination of a Pd–OiPr
moiety, and therefore disfavours the formation of carbonate
and oxalate which requires the formation of bulkier Pd–COOiPr
moieties as intermediates. It can be suggested that in this
case, β-hydride elimination diminishes steric strain in the
complex.
2
.4 Electronic and steric effects of the diphosphine ligand
As already mentioned, in order to evaluate the role of the cat-
alyst's properties in controlling the catalytic activity and
selectivity, the electronic and steric properties should also be
taken into account. Insight into these effects has been
obtained from catalytic experiments employing a variety of
palladium complexes with ferrocenyl ligands bearing differ-
ent alkyl or aryl substituents, as shown in Fig. 2.
2.5 Mechanistic investigation of the oxalate forming step
The results are reported in Table 1, together with those
relevant to the dppf-based catalyst for comparative purposes.
The notable differences observed in activity and selectivity
when aryl-substituted ferrocenyl phosphines were employed
With the aim of rationalizing the large bite angle effect and
gaining more insight into the role of the dicarboalkoxy
species in oxalate formation, we attempted their synthesis
under conditions close to those used in catalysis, i.e. from
(all with identical steric properties) suggest that electronic
ijPdIJOTs)
2
IJP∩P)] dissolved in iPrOH, under carbon monoxide,
rather than steric effects play a dominant role in controlling
oxalate formation.
either in the presence or absence of NEt
absence of BQ (reaction 1).
3
, but always in the
The aryl ferrocenyl ligand with electron-withdrawing sub-
stituents ( pCF ) strongly inhibits the catalysis, whereas the
3
ligand with electron-releasing groups ( pMeO) presents a
slightly accelerating effect (entries 1–3, Table 1).
(ð1Þ)
In order to rationalize the data from a mechanistic point
of view, it should be noted that electron acceptor groups
2
4
accelerate reductive elimination,
while electron donors
favour reoxidation. Thus, for the aryl ferrocenyl ligands,
reductive elimination to oxalate might not be the rate deter-
mining step because a higher activity would be expected
In spite of the wide spectrum of conditions explored,
when iPrOH was used as a solvent, all synthesis attempts
gave unsatisfactory results. In contrast, with MeOH, it was
when pCF -dppf is used, which is not the case. On the con-
trary, the rate limiting step might be the reoxidation of Pd(0)
that is formed in the product forming step.
All the explanations given above seem to be in contrast
with the results obtained using the more basic alkyl
possible to isolate two complexes, cis-ijPdIJCOOMe) IJP∩P)]
3
2
(P∩P = dppe and dppp) (ESI†). Therefore, we decided to use
the “carbomethoxy species” as models for mechanistic
investigations.
The reactivity of preformed cis-ijPdIJCOOMe)
2
IJP∩P)] (P∩P =
dppe, dppp) was studied by NMR spectroscopy. The dppe-based
complex in CD Cl at 25 °C was stable for days, whereas the
2
2
dppp one gave 80% DMO (dimethyl oxalate) and unidentified
Pd complexes after 24 hours. The cis-ijPdIJCOOMe) IJdppp)]
2
complex was dissolved in CD Cl (0.01 mmol in 1 mL) at
2
2
−
78 °C and decomposition was observed in time upon
1
31
1
increasing the temperature. The H (Fig. 3) and P{ H}
Fig. S1, ESI†) NMR spectra showed a slight shift of both
(
3
1
1
P{ H} and methoxy “Pd-COOMe” signals with increasing
temperature but no decomposition products were observed
Fig. 2 Ferrocenyl-diphosphine ligands.
below 25 °C. However, after 24 h the intensity of the
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 1 Electronic and steric effects on the oxidative carbonylation reaction
−1
TOF [mol (mol h)
]
Selectivity [%]
Entry
P∩P
θa
χa
O
C
A
O
C
A
1
2
3
4
5
6
pCF
dppf
pMeO-dppf
dippf
dcypf
3
-dppf
145
145
145
160
170
182
20.5
13.2
10.5
3.4
1.4
0.0
4
8
6
7
1.0
2.2
1.3
10
15
5
1.1
4.3
3.7
18
92
96
62
21
16
36
2
2
18
27
21
45
6
6
20
52
62
245
291
3.5
1.7
0.9
dtbpf
−4
−1
Conditions: ijPdIJOH
2
)
n
IJOTs)2−n(P∩P)](TsO)
anhydrous iPrOH. Tolman's cone angle θ (steric parameter) and the χ values (electronic parameter) are taken from ref. 22 and 23, respectively,
considering the corresponding PAr
n 3
(n = 0, 1) = 2 × 10 mol L , Pd/BQ/NEt = 1/700/2, PCO = 80 atm, T = 80 °C, t = 1 h, 5 mL of
a
3
.
The same results were observed when a 2,2′-dipyridyl
2
6
dicarbomethoxy complex (0.01 mmol), prepared separately,
was reacted with one equivalent of either dppe or dppp
Scheme 5) in CD Cl . In both cases, the formation of the cor-
(
2
2
responding cis-ijPdIJCOOMe)
plete. In more detail, upon the addition of dppe or dppp an
upper shift of the methoxy signals occurred due to the forma-
2
IJP∩P)] was immediate and com-
3
1
tion of cis-ijPdIJCOOMe)
nitely confirmed their formation (see Fig. 4, S2, and S3,
2
IJP∩P)]. The P-NMR analysis defi-
ESI†). As was already observed when starting from preformed
2
cis-ijPdIJCOOMe) IJdppp)], after 24 hours at room temperature,
the in situ formed complex also decomposed giving ca. 80%
DMO.
2
3
Scheme
4
Equilibrium between cis k -P,P and trans k -P,P,Fe
bonding modes (R = tBu, iPr, Cy; R′ = alcohol residue).
In view of the same reactivity observed for both pre- or in
situ-formed dppe and dppp complexes we decided to extend
this simple synthetic approach to other P∩P ligands (the pro-
cedure used is described in the ESI†). The results are
reported in Table 2.
The reactivity of these complexes provides valuable infor-
mation. Compared to the others, the dppe- and dppp-
dicarbomethoxy complexes are relatively stable. These results
and the poor performance of the dppe- and dppp-based cata-
lysts (Table 1) indicate that when these diphosphines are
used the reductive elimination of oxalate might be the slow
step of the catalysis.
When wider bite angle diphosphines are used, the
dicarboalkoxy complexes are unstable and they decompose
immediately, with dppf even at −78 °C and in the presence of
50 atm CO, giving oxalate as a product (except for SPANphos
in which the disappearance of Pd–COMe signals occurs with-
out the formation of defined products). All attempts to define
the nature of the in situ formed Pd(0) through ESI and MALDI
1
analysis gave unsatisfactory results. In Fig. 5 the H-NMR spec-
tra registered after the addition of P∩P to the 2,2′-dipyridyl
dicarbomethoxy palladium complex precursor are shown.
This is in line with the above reported hypothesis that, at
least for the ferrocenyl ligands, the slow step in the catalysis is
not reductive elimination but might be the reoxidation step.
In order to gain some insight into the reoxidation step,
the productivity of the dppf-based catalyst was studied using
different BQ/Pd ratios and different reaction times (Table 3).
The results show that the productivity toward O does not
depend on the BQ/Pd initial ratio (entries 1–6) and exhibits a
1
Fig.
3
H
NMR spectra relevant to the stability of cis-
IJdppp)] in CDCl
ijPdIJCOOMe)
2
2
.
“
Pd-carbomethoxy” signal decreased with
appearance of a new singlet at 3.92 ppm which is assigned to
DMO. The P{ H} spectra revealed the formation of ill-defined
compounds.
a concomitant
3
1
1
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
Scheme
5
Synthesis of ijPdIJCOOMe)
2
IJP∩P)] complexes and their
reactivity toward the reductive elimination step.
1
Fig. 5 H NMR spectra relevant to the in situ formation and stability of
ijPdIJCOOMe) .* = Ligand signals.
IJP∩P)] complexes in CDCl
2
2
linear trend with time (entries 4, 7 and 1, 8, 9). Even though
these data do not suggest that the reoxidation step is the
slow one, this cannot be excluded because it might be that
either i) uncoordinated BQ is not involved or more likely ii)
palladium is present as a species in which Pd(0) is coordi-
nated to BQ, for instance ijPdIJBQ)IJP∩P)]. Indeed, the PdIJ0)–BQ
27
complexes with mono- or bidentate ligands are well known
and recently it has been demonstrated that they are able to
1
2f,28
promote oxidative carbonylation reactions.
Other hypoth-
eses can be suggested, but in the absence of more significant
data, this discussion would be speculative.
Going back to the reactivity of the in situ-formed Pd–
carbomethoxy species, only with SPANphos, the formation of
oxalate was not observed, not even after 24 hours. This fact
was at first attributed to the formation of a trans-palladium
1
Fig.
4
H
NMR spectra relevant to the reaction between cis-
IJdipy)] and dppp in CDCl
ijPdIJCOOMe)
2
2
.
2
compound such as trans-ijPdIJCOOMe) IJSPANphos)] which is
unable to undergo elimination of oxalate. However, the
nature of the complexes formed in situ has not been deter-
mined. In order to gain more insight, the reactivity of trans-
Table 2 Reactivity of in situ formed ijPdIJCOOMe)
2
IJP∩P)]
a
Entry
P∩P
β
n
[°]
Stability
Time
Yield DMO [%]
1
2
3
4
5
6
7
dppe
dppp
dppb
dppf
DPEphos
Xantphos
SPANphos
78.1
86.2
98.7
99.1
102.9
110.0c
171.9
Stable
24 h
24 h
0
80
100
Moderately stable
Unstableb
Unstable
Unstable
Unstable
Stable
Immediatelyb
Immediately
Immediately
Immediately
24 h
b
100
100
100
0
−
2
−1
a
b
Conditions: [Pd] = 1 × 10 mol L , T = 25 °C, 1 mL of anhydrous CD
2
Cl
2
.
The natural bite angles (β
n
) are taken from ref. 16. Even at −78 °C
c
17
2
and at 50 atm CO. Extracted from the X-ray structure of trans-ijPdCl IJSPANphos)].
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 3 Influence of BQ/Pd ratio and reaction time on the oxidative car-
ijPdIJCOOMe)IJOTs)IJSPANphos)] is
a stable complex, thus
a
bonylation reaction
preventing further transformation (i.e. its conversion to
dicarboalkoxy species and oxalate or formation of carbonate
by alcoholysis).
−1
BQ/Pd
TON [mol mol
]
Selectivity [%]
[
mol
Time
[h]
−
1
Entry
mol
]
O
C
A
O
C
A
The fact that reductive elimination takes place immedi-
ately with Xantphos or DPEphos seems in contrast with the
poor catalytic performance obtained with these wide bite
1
2
3
4
5
6
7
8
9
1400
1000
900
700
600
500
1400
700
1
1
1
1
1
1
2
2
3
253
253
248
245
250
247
537
543
846
6
6
5
6
5
4
11
9
16
13
12
10
15
16
15
25
31
49
93
93
94
92
92
93
94
93
93
2
2
2
2
2
2
2
2
2
5
4
4
6
6
6
4
5
5
19
angle bidentates. This observation suggests that in these
cases the slow step in catalysis might be the formation of a
dicarboalkoxy species or any species that leads to its forma-
tion, and as expected from previous work, reductive elimina-
tion is fast for Xantphos and DPEphos complexes.
1400
a
−4
−1
Conditions: ijcis-ijPdIJOH
2
)IJOTs)IJdppf)]IJTsO)] = 2 × 10 mol L , Pd/
Conclusions
3
NEt = 1/2, PCO = 80 atm, T = 80 °C, t = 1 h, 5 mL of anhydrous iPrOH.
2
The oxidative carbonylation of iPrOH catalysed by ijPdX -
IJP∩P)] gives the corresponding oxalate as the major product
and carbonate. Acetone is also formed as a minor by-product.
The catalytic performance is strongly influenced by the prop-
erties of the diphosphine ligand (bite angle and electronic
and steric parameters), the nature of the counter anion and
the pressure of carbon monoxide. Specifically, high activity
and selectivity toward oxalate are achieved i) with weakly
coordinating anions, ii) with electron-donating and non
bulky P∩P ligands with a relatively wide bite angle capable of
maintaining a cis geometry, and iii) under relatively high CO
pressure. The best results were obtained with cis-ijPdIJOTs)2-
IJpMeO-dppf)]. These results and those on the reactivity of the
3
1
1
1
ijPdIJOTs)
NMR spectroscopy under conditions closer to those of cataly-
sis IJMeOH/CD Cl
= 1/10, v/v, PCO = 5 atm). At −78 °C the
monocarbomethoxy complex trans-ijPdIJCOOMe)IJOTs)-
IJSPANphos)] is formed (ESI†), which is stable even in the
presence of NEt up to 60 °C. Above this temperature decom-
2
IJSPANphos)] was studied by VT P{ H} and
H
2
2
3
position takes place, without the formation of DMO, dimethyl
carbonate, or formaldehyde. No dicarboalkoxy Pd-complexes
were observed either. It is supposed that the formation of
trans-monocarboalkoxy species occurs through
coordinated intermediate in which CO and MeO are in the
a
five-
dicarboalkoxy species of the type cis-ijPdIJCOOMe)
2
IJP∩P)] sug-
cis position so they can react to give a Pd–COOMe moiety
2
9
gest that the slow step of the catalysis is related to the nature
of P∩P. For dppe and dppp, reductive elimination to oxalate
in the product forming step may be rate limiting, while for
dppb, dppf and substituted dppf the limiting step might be
reoxidation. For wider bite angle ligands such as DPEphos,
Xantphos and SPANphos the difficult step may be the forma-
tion of a dicarboalkoxy species or any species leading to its
formation.
(Scheme
6).
Probably,
the
square-planar
trans-
The importance of cis-chelation has been shown by study-
ing the reactivity of the SPANphos-based catalyst. Although in
this case the monocarbomethoxy complex trans-ijPdIJCOOMe)-
IJOTs)IJSPANphos)] can be isolated, catalysis does not occur to
a significant extent, probably because the trans geometry
does not favour the steps required for the advancement of
the catalysis.
Acknowledgements
Financial support from MIUR (Rome) and ICIQ support units
are gratefully acknowledged.
Notes and references
1
2
G. Kiss, Chem. Rev., 2001, 101, 3435.
(a) E. Drent and P. H. M. Budzelaar, Chem. Rev., 1996, 96,
6
63; (b) E. Drent, R. van Dijk, R. van Ginkel, B. van Ort and
R. I. Pugh, Chem. Commun., 2002, 964; (c) A. Nakamura,
Scheme
6 Putative mechanism of the formation of trans-
ijPdIJCOOR)IJOTs)IJSPANphos)] (R = Me).
T. M. J. Anselment, J. Claverie, B. Goodall, R. J. Jordan, S.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
Mecking, B. Rieger, A. Sen, P. W. N. M. van Leeuwen and K.
Nozaki, Acc. Chem. Res., 2013, 46, 1438.
(a) A. Klausener and J. D. Jentsch, Oxidative Carbonylation,
in Applied Homogeneous Catalysis with Organometallic
Compounds, ed. B. Cornils and W. A. Herrmann, Wiley–VCH,
14 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.
15 P. W. N. M. van Leeuwen, M. A. Zuideveld, B. H. G.
Swennenhuis, Z. Freixa, P. C. J. Kamer, K. Goubitz, J.
Fraanje, M. Luitz and A. L. Spek, J. Am. Chem. Soc.,
2003, 125, 5523.
16 (a) M. Kranenburg, P. C. J. Kamer, P. W. N. M. van Leeuwen,
D. Vogt and W. Keim, J. Chem. Soc., Chem. Commun.,
1995, 2177; (b) M. Kranenburg, J. P. C. Kamer and
P. W. N. M. van Leeuwen, Eur. J. Inorg. Chem., 1998, 155.
17 Z. Freixa, M. S. Beentjes, G. D. Batema, C. B. Dieleman,
G. P. F. van Strijdonck, J. N. H. Reek, P. C. J. Kamer, J.
Fraanje, K. Goubitz and P. W. N. M. van Leeuwen, Angew.
Chem., Int. Ed., 2003, 42, 1284.
18 Z. Freixa and P. W. N. M. van Leeuwen, Dalton Trans.,
2003, 1980.
19 J. E. Marcone and K. J. Moloy, J. Am. Chem. Soc., 1998, 120,
8527.
20 M. Zuideveld, B. H. G. Swennenhuis, M. D. K. Boele, Y.
Guari, G. P. F. van Strijdonck, J. N. H. Reek, P. C. J. Kamer,
K. Goubitz, J. Fraanje, M. Lutz, L. Spek and P. W. N. M. van
Leeuwen, J. Chem. Soc., Dalton Trans., 2002, 2308.
21 C. Jimenez-Rodriguez, F. X. Roca, C. Bo, J. Benet-Buchholz,
E. C. Escudero-Adan, Z. Freixa and P. W. N. M. van Leeuwen,
Dalton Trans., 2006, 268.
3
1
996; (b) V. R. Khabibulin, A. V. Kulik, I. V. Oshanina, L. G.
Bruk, O. N. Temkin, V. M. Nosova, Y. A. Ustynyuk, V. K.
Belskii, A. I. Stash, K. A. Lysenko and M. Y. Antipin, Kinet.
Catal., 2007, 48, 228.
4
5
S. D. Friis, T. Skrydstrup and S. L. Buchwald, Org. Lett.,
2
014, 16, 4296.
(a) J. A. Gillespie, L. D. Debora and P. C. J. Kamer, Dalton
Trans., 2010, 39, 2751; (b) J. A. Gillespie, E. Zuidema,
P. W. N. M. van Leeuwen and P. C. J. Kamer, Phosphorus
Ligand Effects in Homogeneous Catalysis and Rational
Catalyst Design, in PhosphorusIJIII) Ligands in Homogeneous
Catalysis: Design and Synthesis, ed. P. C. J. Kamer and P. W.
N. M. van Leeuwen, John Wiley & Sons, 2012.
G. Cavinato and L. Toniolo, Molecules, 2014, 19, 15116.
A. Sen, Catalytic Synthesis of Alkene–Carbon Monoxide
Copolymers and Cooligomers, ed. A. Sen, Kluwer Academic
Publishers, Dordrecht, 2003.
(a) Y. Ono, Pure Appl. Chem., 1996, 68, 367; (b) D. Delledonne,
F. Rivetti and U. Romano, Appl. Catal., A, 2001, 221, 241; (c) P.
Tundo and M. Selva, Acc. Chem. Res., 2002, 35, 706; (d) J. L.
Gong, X. B. Ma and S. P. Wang, Appl. Catal., A, 2007, 316, 1.
S. Uchiumi, K. Ataka and T. Matsuzaki, J. Organomet. Chem.,
6
7
8
9
22 C. A. Tolman, Chem. Rev., 1977, 77, 313.
1
999, 576, 279.
23 T. Bartik, T. Himmler, H. Schulte and K. J. Seevogel,
J. Organomet. Chem., 1984, 272, 29.
24 E. Negishi, T. Takahashi and K. Akiyoshi, J. Organomet.
Chem., 1987, 334, 181.
1
1
0 D. M. Fenton and P. J. Steinwand, J. Org. Chem., 1974, 39, 701.
1 (a) S. P. Current, J. Org. Chem., 1983, 48, 1779; (b) L. N. Zhir-
Lebed and O. N. Temkin, Kinet. Katal., 1984, vol. 25, p. 255;
(
c) L. N. Zhir-Lebed and O. N. Temkin, Kinet. Katal., 1984,
25 (a) C. Bianchini, A. Meli, W. Oberhauser, S. Parisel, E.
Passaglia, F. Ciardelli, O. V. Gusev, A. M. Kal'si and N.
Vologdin, Organometallics, 2005, 24, 1018; (b) M. A.
Zuideveld, B. H. G. Swennenhuis, P. C. J. Kamer and
P. W. N. M. van Leeuwen, J. Organomet. Chem., 2001, 637–
639, 805.
26 (a) R. Santi, A. M. Romano, R. Garrone and R. Millini,
J. Organomet. Chem., 1998, 566, 37; (b) P. Giannoccaro, N.
Ravasio and M. Aresta, J. Organomet. Chem., 1993, 451, 243;
(c) G. D. Smith, B. E. Hanson, J. S. Merola and F. Waller,
Organometallics, 1993, 12, 568.
27 (a) T. Ukai, H. Kawazura, Y. Ishii, J. J. Bonnet and J. A. Ibers,
J. Organomet. Chem., 1974, 65, 253; (b) M. Hiramatsu, K.
Shiozaki, T. Fujinami and S. Sakai, J. Organomet. Chem.,
1983, 246, 203.
28 (a) A. V. Kulik, L. G. Bruk, O. N. Temkin, V. R. Kabibulin,
V. K. Belsky and V. E. Zavodnik, Mendeleev Commun.,
2002, 12, 47; (b) O. N. Temkin and L. G. Bruk, Kinet. Catal.,
2003, 44, 601.
vol. 25, p. 263; (d) R. Bertani, G. Cavinato, L. Toniolo and G.
Vasapollo, J. Mol. Catal., 1993, 84, 165; (e) D. Delledonne, F.
Rivetti and U. Romano, J. Organomet. Chem., 1995, 488, C15;
(
f ) P. Giannoccaro, J. Organomet. Chem., 1994, 470, 249; (g)
S. Uchiumi, K. Ataka and T. Matsuzaki, J. Organomet. Chem.,
999, 576, 279; (h) C. S. Chin, D. Shin, G. Won, J. Ryu, H. S.
1
Kim and B. G. Lee, J. Mol. Catal. A: Chem., 2000, 160, 315.
1
2 (a) F. Rivetti and U. Romano, J. Organomet. Chem., 1978, 154,
3
1
1
23; (b) F. Rivetti and U. Romano, J. Organomet. Chem.,
979, 174, 221; (c) F. Rivetti and U. Romano, Chim. Ind.,
980, 62, 7; (d) L. Toniolo, J. Organomet. Chem., 1993, 444,
C65; (e) E. Amadio, G. Cavinato, A. Dolmella and L. Toniolo,
Inorg. Chem., 2010, 49, 3721; ( f ) G. Cavinato, S. Facchetti and
L. Toniolo, J. Mol. Catal. A: Chem., 2012, 352, 63; (g) J. M.
Tiddo, E. Bouwman and E. Drent, Eur. J. Inorg. Chem.,
2
2
012, 1403; (h) E. Amadio and L. Toniolo, J. Organomet. Chem.,
014, 750, 74; (i) E. Amadio and L. Toniolo, J. Organomet.
Chem., 2014, 767, 72.
1
3 M. N. Birkholz, Z. Freixa and P. W. N. M. van Leeuwen,
Chem. Soc. Rev., 2009, 38, 1099.
29 E. Zuidema, P. W. N. M. van Leeuwen and C. Bo,
Organometallics, 2005, 24, 3703.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.