Catalyst life in imidazolium-based ionic liquids for palladium-catalysed asymmetric allylic alkylation

Article abstract of DOI:10.1039/c4dt00169a

A Pd/(S)-BINAP system was successfully applied to the asymmetric allylic alkylation of rac-1,3-diphenyl-3-acetoxyprop-1-ene (I) using imidazolium-based ionic liquids (ILs) attaining up to 225 h^-1 TOF and 88% ee of the (R)-product. Although the system was barely active in the recycling experiments, the catalyst life was confirmed after recharging the system with substrate/reactants resulting in an alkylated product. In the latter case, the conversion rates and enantiomeric excesses were similar or lower compared to those in the first cycle. In order to explain the observed catalyst performance in the recycling as well as in the recharging experiments, we investigated the reactivity between the catalyst precursors, substrate and reactants in ILs. We were able to identify the species involved in the catalytic reactions under various conditions by means of ³¹P NMR analyses. Allylpalladium intermediates (3) were found to be the active and selective species at a high substrate concentration. When the substrate was consumed, competing reactions took place leading to different palladium complexes. [PdCl(NHC ^Bu,Me)((S)-BINAP)]Cl (4), together with [Pd((S)-BINAP)₂] (5), were recognised as the species responsible for the loss of activity, meanwhile, the decrease in enantioselectivity was accounted for by the formation of mixed (NHC)(monophosphine)-palladium species. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4dt00169a

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Catalyst life in imidazolium-based ionic liquids for
palladium-catalysed asymmetric allylic alkylation†
Cite this: Dalton Trans., 2014, 43,
7533
I. Guerrero-Ríos and E. Martin*
A Pd/(S)-BINAP system was successfully applied to the asymmetric allylic alkylation of rac-1,3-diphenyl-
3-acetoxyprop-1-ene (I) using imidazolium-based ionic liquids (ILs) attaining up to 225 h−1 TOF and 88%
ee of the (R)-product. Although the system was barely active in the recycling experiments, the catalyst life
was confirmed after recharging the system with substrate/reactants resulting in an alkylated product. In
the latter case, the conversion rates and enantiomeric excesses were similar or lower compared to those
in the first cycle. In order to explain the observed catalyst performance in the recycling as well as in the
recharging experiments, we investigated the reactivity between the catalyst precursors, substrate and
reactants in ILs. We were able to identify the species involved in the catalytic reactions under various con-
ditions by means of 31P NMR analyses. Allylpalladium intermediates (3) were found to be the active and
selective species at a high substrate concentration. When the substrate was consumed, competing reac-
tions took place leading to different palladium complexes. [PdCl(NHCBu,Me)((S)-BINAP)]Cl (4), together
with [Pd((S)-BINAP)2] (5), were recognised as the species responsible for the loss of activity, meanwhile,
the decrease in enantioselectivity was accounted for by the formation of mixed (NHC)(monophosphine)-
palladium species.
Received 16th January 2014,
Accepted 17th February 2014
DOI: 10.1039/c4dt00169a
www.rsc.org/dalton
successfully applied in diverse catalytic reactions (i.e. cross
coupling, allylic alkylation, polymerization, carbonylation,
Introduction
The use of ionic liquids (ILs) in metal-catalysed reactions has C–H activation, oxidation, reduction, addition, telomerization
become a viable alternative to organic solvents in many pro- reactions, etc.),6 its application to allylic alkylation has been
cesses due to the feasibility of product separation and reuse of limited to the non-asymmetric version,7 where the presence of
the IL-catalytic phase.1 The field of asymmetric catalysis phosphine ligands is required to achieve high conversions. In
profits from IL technologies, as demonstrated by research order to avoid the formation of NHC-palladium species in imi-
groups in academia.2 However, the use of imidazolium-based dazolium-based ILs and therefore produce recyclable systems,
ILs in asymmetric allylic alkylation (AAA) reactions has failed pyrrolidonium-based ILs have been used to reproduce ees and
to reproduce the activity and enantioselectivity after recycling yields in asymmetric allylic amination and phosphination
of the IL-catalytic system.3 The latter has been attributed to the reactions.8 Recently, imidazolium-based ILs were successfully
possible formation of inactive species during the catalytic applied in palladium catalysed AAA under MW irradiation to
process. It is generally agreed that imidazolium-based ILs react increase the reaction rate, thus avoiding the formation of in-
with catalytic palladium species via oxidative addition to gen- active and non-selective carbenic species, which are otherwise
erate stable N-heterocyclic-carbene (NHC) palladium complexes, observed.9
or by deprotonation of the C(2)–H bond of the imidazolium
In order to gain insight into the role of palladium species
cation in the presence of a base.4 Additionally, protons at the involved in catalysed AAA in imidazolium-based ILs, we
C(4) and C(5) positions could react to afford “abnormal” carbe- assessed the abilities of the catalytic system Pd/(S)-BINAP to
nic species.5 Although NHC-palladium precursors have been perform AAA in a variety of imidazolium-based ILs (Scheme 1).
Furthermore, we investigated the species responsible for the
activity and selectivity observed.
Departamento de Química Inorgánica, Facultad de Química, Universidad Nacional
Autónoma de México, Av. Universidad 3000, 04510 D.F., México.
E-mail: erikam@unam.mx; Fax: +525556223720
†Electronic supplementary information (ESI) available: Description of pro-
cedures for recycling experiments (Table S1); 1H and 31P NMR spectra of AAA
Results and discussion
We studied the palladium-catalysed alkylation of rac-1,3-diphe-
nyl-3-acetoxyprop-1-ene (I) using dimethylmalonate (II) as the
stoichiometric reaction studies (Fig. S1–S3, S9–S11) NMR, MS-FAB and HRMS
(ESI-TOF) spectra of complex 4 (Fig. S4–S8). See DOI: 10.1039/c4dt00169a
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 7533–7539 | 7533

View Article Online
Paper
Dalton Transactions
Catalytic reactions using the preformed catalyst [Pd(η3-
C3H5)((S)-BINAP)]BF4 (1)11 in ILs A and D showed comparable
activities and enantioselectivities to those exhibited by analo-
gous systems (in situ formed catalyst “[Pd(η3-C3H5)((S)-BINAP)]-
Cl”) (Table 1, entries 5 and 6 vs. 1 and 4, 1st cycle column).
These results contrast with previous observations about the
presence of coordinating anions as chlorides and their effect
on reaction selectivity. Lloyd-Jones and coworkers found that
chlorides have a detrimental effect on the reaction enantio-
selectivity,15 furthermore, Norrby and coworkers proposed that
chlorides can exert positive memory effects but also favour the
isomerisation of allylpalladium intermediates affecting the
reaction stereocontrol.16
Attempts to reuse the IL-catalytic system, after product
extraction with hexane, resulted in poor conversion for all ILs
used (<10%, Table S1 in ESI†), which is in agreement with the
pioneering work of Toma and coworkers.3 Since the catalyst
activity drop in recycling systems is commonly attributed to
catalyst loss or poisoning during product extraction, we
assessed the catalyst life upon recharging the system with the
substrate and reactants consumed during the first cycle
(Table 1, 2nd cycle column). The systems were still active in the
second cycle but failed to reproduce the enantioselectivities and
conversions of the first cycle. Interestingly, when the preformed
catalyst 1 was used, the activity was maintained although the
enantioselectivity was reduced (Table 1, entries 5 and 6, 2nd
cycle column). Similar behaviour was displayed by the in situ
generated catalyst in A, which achieved 90% conversion and
36% ee of (R)-III in the second cycle (Table 1, entry 1, 2nd cycle
column).
Scheme 1 Asymmetric allylic alkylation in imidazolium-based ILs.
Table 1 Pd/(S)-BINAP catalysed asymmetric allylic alkylation of I in ILs
1st cyclea
2nd cycleb
Conv.c (%)
Entry
IL
L*/Pd
Conv.c (%)
eec (%)
eec (%)
1
2
3
4
5
6
7
A
1.25
1.25
1.25
1.25
1e
97
97
100
100
98
90
100
72
80
84
88
77
86
87
90
53
54
69
98
93
26
36
38
56
76
68
56
75
Bd
C
D
A
D
D
1e
2.5
a 20 °C; 120 min; 1% [Pd(η3-C3H5)Cl]2; I (0.5 mmol); I/II/BSA 1/3/3 (BSA
= N,O-bis(trimethylsilyl)acetamide); KOAc (0.005 mmol); IL (1 mL).
b 2nd cycle recharging the system with I (0.5 mmol); I/II/BSA 1/1/1.
c Determined by HPLC; ee values for (R)-configuration. d Performed at
50 °C. e 2% [Pd(η3-C3H5)-((S)-BINAP)]BF4 (1).
nucleophile under basic Trost conditions (Scheme 1).10 Cataly-
tic species were formed either in situ by reacting [Pd(η3-C3H5)-
Cl]2 with the appropriate amount of (S)-BINAP (Pd : L ratio
of 1.25 or 2.5), or by addition of the preformed complex
[Pd(η3-C3H5)((S)-BINAP)]BF4 (1).11 Regarding ILs, we focused
our attention on anion effects and cation C(2)-substitutions
(A–D in Scheme 1).
Table 1 summarises the catalytic results we obtained. Com-
plete conversions and good enantioselectivities (up to 88%)
were reached within two hours at 20 °C in all the ionic liquids
tested (Table 1, 1st cycle column). It is worth mentioning that
the catalytic behaviour was independent of the catalyst gene-
ration method (Table 1, entries 1 vs. 5 and 4 vs. 6, 1st cycle
column).
With the aim of identifying species involved in the catalytic
cycle that could explain the failure of the recycling experi-
ments, as well as the complex catalytic behaviour observed
1
under recharging conditions, we performed H and 31P NMR
experiments on the Pd/(S)-BINAP catalytic system in the pres-
ence of a substrate, nucleophile and IL D (Fig. 1 and S1†).
Initially, complex [Pd(η3-C3H5)((S)-BINAP)]Cl (2) was formed
When the reactions were carried out in [BMIM][BF4] (A) and
[HDMIM][BF4] (B), the enantioselectivities proved to be lower
than those in the reactions performed in [BMIM][PF6] (C) and
[BMIM][NTf2] (D) (Table 1, entries 1–2 vs. 3–4, 1st cycle
column). Since ILs A12 and B13 are more viscous than C12 and
D,12 the observed catalytic behaviour can be rationalised in
terms of the reactants’ diffusion, especially the nucleophile
diffusion. The limited mobility of the reactants reduced the
rate of nucleophilic attack and, consequently, the isomerisa-
tion reactions of the allylic entities gave rise to diverse allyl-
palladium isomers.14 Furthermore, when the nucleophilic
attack takes place at the terminal allylic carbons of the allylpal-
ladium isomers, a drop in ee occurs as we observed for the
more viscous media A and B.
Fig. 1 31P NMR (121 Hz) spectra of the stoichiometric AAA reaction
using 2 ((S)-BINAP/Pd = 1.25) in CD2Cl2: (a) 2 in the presence of D; (b)
addition of I, II and BSA in a 1 : 1 : 1 ratio with respect to Pd; (c) addition
of II and BSA in a 1 : 1 ratio with respect to Pd. (*) Unknown species.
7534 | Dalton Trans., 2014, 43, 7533–7539
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
Scheme 2 Isomeric species of 3.
from [Pd(η3-C3H5)Cl]2 and (S)-BINAP (the (S)-BINAP : Pd ratio
was 1.25) in the presence of D and CD2Cl2. In the 31P NMR
spectrum (Fig. 1a), the signal centred at 22.5 ppm was
assigned to 2, and the broadening peak and chemical shift is a
consequence of interactions between 2 and D. It is important
to note that in the absence of D, a sharp signal at 19.1 ppm
was observed for complex 2. After the addition of equimolar
Fig. 2 Formation of NHC-species 4 and its corresponding 31P NMR
spectrum displaying a 1 : 1 diastereoisomeric mixture.
amounts of I, II and BSA, species 2 turned, upon nucleophilic isomer was completely characterised by NMR spectroscopy
attack, into the species [Pd(η3-Ph2C3H3)((S)-BINAP)]Cl (3) (Fig. S4–S7 in ESI†). It is worth mentioning that complex 4 was
together with oxidised diphosphine ligand (S)-BINAPO17 not active in the asymmetric allylic alkylation, reaching less
(Fig. 1b). Species 3 was identified as a mixture of confor- than 8% conversion in three days in dichloromethane with
mational isomers,18 assigned as syn–syn and syn–anti iso- 66% ee of the (R)-product. In addition, saturated species 5 was
mers,18b in a 4 : 1 ratio (Scheme 2). Upon the addition of a not active in AAA until dissociation of a (S)-BINAP ligand
second equivalent of II and BSA (Fig. 1c) a new species was occurred, thus forming intermediate 3.
formed, and we were able to assign it as the NHC-carbenic
The different selectivity attained on recharging the IL-
species [PdCl(NHCBu,Me)((S)-BINAP)]Cl (4) since we carried out catalytic systems indicates that different palladium species
its independent synthesis (below). The formation of carbenic were formed during the catalytic reactions affecting the activi-
species 4 should be a consequence of the oxidative addition of ties and selectivities in all the ILs tested in subsequent cycles.
an imidazolium cation to the Pd(0) species generated in situ,4b,c According to the 31P NMR monitoring experiments and the
although C(2) deprotonation of the imidazolium ring by base catalytic behaviour of 4 (above), it is possible to propose that
and subsequent attack to the metal centre cannot be dis- the catalytic performance in IL D, in the first cycle, is due to
regarded.19 Additionally, [Pd((S)-BINAP)2] (5)20 was identified species 3 (Path i in Scheme 3) and that the drop in activity in
together with a smaller amount of an unknown species. subsequent cycles is due to the formation of species 4–5
Similar experiments performed in the absence of IL did not (Table 1, entry 4). In a similar way, it is possible to propose
generate the NHC-carbenic species 4; addition of an IL, just that inactive species 4–5 are responsible for the failure in re-
after the second nucleophilic attack, resulted in the generation cycling experiments, since their formation was favoured when
of NHC species 4 (Fig. S2–S3 in ESI†).
the substrate was consumed and the Pd(0) species reacted with
To further investigate the viability of carbenic species 4, we IL or (S)-BINAP (Path ii, Scheme 3).
looked at its independent synthesis to assess its stability and
When the system was recharged with substrate and reac-
catalytic activity in AAA. Species 4 was obtained by the reaction tants, moderate enantioselectivities were attained in the
of [PdCl2((S)-BINAP)] with AgCl(NHCBu,Me
) in good yield second cycle, suggesting that the observed behaviour is a con-
(Fig. 2). The 31P NMR spectrum of complex 4 corresponds to sequence of the ratio of species 3 (Path i in Scheme 3) and
four set of doublets, two in the region of 17–18 ppm and two possible mixed (NHC)(monophosphine)Pd(II) species. Analo-
in the 28–30 ppm region. 2D 31P–31P experiments indicated a gous systems, containing triphenylphosphine and NHC-
correlation between the doublets at 29.8 ppm and 18.0 ppm ligands, have been previously reported as very active catalysts
and between the doublets at 28.6 and 17.6 ppm. Therefore, in allylic alkylation reactions.7 The differences in catalytic per-
complex 4 consists of a mixture of two diastereoisomeric formance in the second cycle for all the ILs tested, could be
species (4a and 4b, Fig. 2). The formation of similar diastereo- accounted for in terms of the relative ratios of 3, 4, 5 and
isomeric species has been observed in platinum complexes mixed (NHC)(monophosphine)Pd(II) species. The three latter
bearing a chiral ligand (S,S)-ChiraPhos and a dissymmetrical species are formed when I is totally consumed (Scheme 3).
NHC-carbene.21 The existence of both isomers, 4a and 4b, Attempts to isolate the mixed Pd(II) complex containing (S)-
stems from the dissymmetry of the carbene ligand together BINAP were unsuccessful presumably because bulkier substitu-
with the axial chirality of (S)-BINAP, resulting in a palladium ents on the imidazole ring are required.22
complex exhibiting C1 symmetry caused by the hindered
In an analogous stoichiometric experiment using pre-
rotation of the carbene ligand. We were not able to separate formed complex 1 and IL A, the initial palladium complex
the diastereoisomeric species by crystallisation, instead, we turned, upon a second nucleophilic attack, into species 3 and
attained mixtures enriched in 4a (4a/4b = 1.5), whereby each a new species, with a chemical shift at 25.3 ppm, which could
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 7533–7539 | 7535

View Article Online
Paper
Dalton Transactions
Scheme 3 Palladium species involved in AAA in the presence of ILs.
be attributed to the mixed (NHC)(monophosphine)Pd(II)
species (Fig. S8–S9 in ESI†).7 Surprisingly, species 4 and 5 were
not formed in this case. The presence of 3 and the mixed
(NHC)(monophosphine)Pd(II) complex is accountable for the
observed catalytic performance. During the first cycle, species
3 is responsible for the high activity and enantioselectivity
obtained. The activity is maintained in the second cycle since
both species, 3 and the (NHC)(monophosphine)Pd(II), are
active, and furthermore, the enantioselectivity is reduced by
the presence of the latter complex (Table 1, entries 5–6).7
In order to prevent the formation of inactive NHC species 4
during the catalytic reaction in ILs, we envisaged that the
addition of trapping ligands could stabilise the Pd(0) active
species [Pd((S)-BINAP)(Solvent)] and a further trapping ligand
de-coordination would regenerate the catalyst in subsequent
cycles. Therefore, we explored (S)-BINAP as a stabilising ligand
(in a (S)-BINAP/Pd = 2.5 ratio) in similar stoichiometric experi-
ments as previously described (Fig. 3 and S10†). 31P NMR
showed that initially formed species 2 in the presence of D
(Fig. 3a) reacted with I, II and BSA resulting in the formation
of species 3, (S)-BINPO,23 (S)-BINAPO17 and a small amount of
4 (Fig. 3b). The subsequent addition of II and BSA led to the
formation of [Pd((S)-BINAP)2] (5),20 a small amount of 3,
(S)-BINPO, and (S)-BINAPO (Fig. 3c). When a (S)-BINAP/Pd
ratio of 2.5 was tested, under catalytic conditions in D, similar
activities were reached in the first cycle compared to the
system with a lower (S)-BINAP/Pd ratio (Table 1, entry 4 vs. 7,
1st cycle column), which is in agreement with the results of
the monitoring experiments (Fig. 3b). The system failed to
maintain the conversion rates in the second cycle due to the
formation of species 4 and 5 (Fig. 3c), meanwhile the ee values
fell by only ca. 10%. Attempts to recycle the catalytic system
Fig. 3 31P NMR (121 Hz) spectra for the stoichiometric AAA reaction
using 2 ((S)-BINAP/Pd = 2.5) in CD2Cl2: (a) 2 in the presence of D; (b)
addition of I, II and BSA in a 1 : 1 : 1 ratio with respect to [Pd]; (c) addition
of II and BSA in a 1 : 1 ratio with respect to [Pd]. (*) Unknown species.
with a higher (S)-BINAP/Pd ratio resulted in a loss of both
activity and enantioselectivity after product extraction with
hexane, which was related to the formation of the inactive
species 4 and 5 (Table S1 in ESI†). Formation of the latter
species also explains the previous results by Toma3a in AAA in
IL C, since a fourfold excess of the chiral ligand was used,
where the catalyst showed high activity and ee in the first cycle
but the recycled system underwent a dramatic drop in both
activity and enantioselectivity.
Additionally, we tested different trapping ligands such as
cyclooctadiene and norbornadiene, or coordinating solvents
such as acetonitrile in the AAA in IL A. After extraction of the
7536 | Dalton Trans., 2014, 43, 7533–7539
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
activity and enantioselectivity, respectively. The use of trapping
agents for palladium(0) species were inadequate to prevent the
formation of undesired (NHC)-palladium compounds. When
carbene species formation is suppressed by the use of alterna-
tive ionic liquids such as [HDBU][OAc], it is possible to suc-
cessfully recycle the IL-catalytic system.
Experimental
General materials, methods and instruments
All manipulations of air- and moisture sensitive compounds
were performed under a nitrogen atmosphere using standard
Schlenk and vacuum line techniques. Reagents were pur-
chased from commercial suppliers and used without further
purification. Hexane and CH2Cl2 were dried over CaH2 and dis-
tilled under nitrogen prior to use. [Pd(η3-C3H5)((S)-BINAP)]-
[BF4],11 [PdCl2((S)-BINAP)],24 [BMIM][NTf2],25 [HDMIM][BF4],25
[BMIM][Cl]26 and [HDBU][OAc]27 were prepared following pro-
cedures reported in the literature. NMR spectra were recorded
on Varian Inova 300 and 400 instruments. Chemical shifts are
given in ppm referenced to the solvent (1H and 13C) or the
external reference of 85% aqueous solution of H3PO4 (31P),
and the coupling constants are given in hertz (Hz). FAB+ mass
spectra were acquired using a Jeol SX102A inverse geometry
spectrometer using a 3-nitrobenzylalcohol matrix. High-resolu-
tion mass spectra were obtained using an Agilent G1969A elec-
trospray-ionization time-of-flight mass spectrometer. HPLC
analyses of the catalytic reaction mixtures were performed on
an Alliance-Waters apparatus, equipped with a photodiode
array detector, using Diacel OJH as the column and 10% 2-pro-
panol–hexane, with a flow rate of 1 mL min−1, as the eluent
(λ = 254 nm, guard cartridge 4 × 3 mm). The absolute configur-
ation of the chiral products was assigned by comparing their
retention time with that of optically pure compounds.
Fig. 4 Pd/(S)-BINAP catalysed asymmetric allylic alkylation of
[HDBU][OAc].
I in
product with hexane, the IL-catalytic systems failed to main-
tain the catalytic performance upon recycling (Table S1 in
ESI†). This suggests a possible replacement of the trapping
ligands by (NHC) ligands derived from the IL, generating the
inactive species 4 and non-enantioselective species such as
(NHC)(monophosphine)-Pd(II).
Given that the use of imidazolium-based ILs did not allow
recycling of the catalytic system when base or low-valent cataly-
tic species were present (as discussed above), we explored a
less reactive IL, [HDBU][OAc], in the AAA of I. The preliminary
results demonstrated the feasibility of recycling in this IL
using the same catalytic system. High conversions with good
enantioselectivities are reached along three cycles and the
activity only decreased in a forth cycle (Fig. 4). A profound cata-
lytic study using this promising system is in progress.
Conclusions
Synthetic procedures
[PdCl(NHCBu,Me)((S)-BINAP)]Cl (4). A solution of [BMIM]Cl
High activities and enantioselectivities in imidazolium-based
ionic liquids were attained. Recycling of the IL-catalytic system
failed due to the formation of (NHC)-palladium complexes, (12.2 mg, 0.07 mmol) in 2 mL of CH2Cl2 was treated with Ag2O
even in the case of the 2-methyl substituted IL B, suggesting (9 mg, 0.04 mmol) overnight in the dark. A solution of AgCl-
abnormal (NHC)-palladium complex formation.
(NHCBu,Me) was formed, then filtered through celite and the
A careful AAA stoichiometric study demonstrated that the filtrate was received on a Schlenk containing a solution of
substrate and the IL compete for the palladium(0) species to [PdCl2((S)-BINAP)] (47.9 mg, 0.06 mmol) in CH2Cl2 (2 mL). The
form AAA active species and inactive (NHC)-palladium com- reaction mixture was stirred in the dark for 24 h. The solution
plexes, respectively. The observed catalytic performance is a was filtered through celite and the solvent was removed under
result of the relationship between species 3, 4, 5 and mixed reduced pressure. The product was recrystallised from a
(NHC)(monophosphine)Pd(II). Additionally, it was observed mixture of CH2Cl2 and cyclohexane to afford the desired
that the use of complex 1 prevents the formation of inactive product (42 mg, 77% yield). Spectral data for 4a (from the
species (i.e. 4). Active carbene species such as mixed (NHC)- mixture): 31P NMR (CD2Cl2, 121 MHz) δ 29.8 (d, JP–P = 23.3 Hz),
1
(monophosphine)Pd(II) and 3 are responsible for the high 18.0 (d, JP–P = 23.3 Hz); H NMR (CD2Cl2, 400 MHz) δ 8.2–6.7
activities however, the former causes a drop in enantio- (32H, Ar) 7.15 (1H, NCH), 6.65 (d, J = 8.5 Hz, 1H, NCH), 4.27
selectivity in subsequent cycles.
(m, 1H, NCHH), 4.15 (3H, NMe), 3.70 (m, 1H, NCHH), 1.95 (m,
Our results prove that imidazolium-based ionic liquids, 1H, NCH2CHH), 1.64 (m, 1H, NCH2CHH), 1.53 (2H,
employed in AAA under basic conditions, generate inactive N(CH2)2CH2), 1.05 (3H, N(CH2)3CH3); 13C NMR (CD2Cl2,
(NHC)-palladium complexes as well as active mixed (NHC)- 76.4 MHz) δ 136–120 (Ar), 127.8 (NCH), 127.3 (NCH), 51.2
(monophosphine)Pd(II) complexes, which reduce the catalytic (NCH2), 38.9 (NMe), 32.3 (NCH2CH2), 20.5 (N(CH2)2CH2), 13.9
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 7533–7539 | 7537

View Article Online
Paper
Dalton Transactions
(N(CH2)3CH3) ppm. Spectral data for 4b (from the mixture): Stoichiometric experiments of catalysed allylic alkylation of
31P NMR (CD2Cl2, 121 MHz) δ = 28.6 (d, JP–P = 23.2 Hz), 17.6 (d, rac-1,3-diphenyl-3-acetoxyprop-1-ene (I) in the presence of
JP–P = 23.2 Hz); 1H NMR (CD2Cl2, 400 MHz) δ 8.2–6.7 (32H, Ar), an IL
7.11 (1H, NCH), 6.58 (d, J = 8.5 Hz, 1H, NCH), 4.39 (m, 1H,
An NMR tube was charged with [PdCl(η3-C3H5)]2 (4.5 mg,
NCHH), 4.22 (m, 1H, NCHH), 3.84 (3H, NMe), 2.05 (m, 1H,
NCH2CHH), 1.97 (m, 1H, NCH2CHH), 1.53 (2H, N(CH2)2CH2),
0.012 mmol) and (S)-BINAP in the corresponding ratio (L/Pd =
1.25 and 2.5) or with preformed catalyst
1 (10.2 mg,
1.03 (3H, N(CH2)3CH3); 13C NMR (CD2Cl2, 76.4 MHz) δ
136–120 (Ar), 127.6 (NCH), 127.2 (NCH), 51.8 (NCH2), 38.3
(NMe), 32.7 (NCH2CH2), 20.6 (N(CH2)2CH2), 14.0 (N(CH2)3CH3)
ppm. MS (FAB): m/z = 903 for [C52H46ClN2P2Pd]+. HRMS
(ESI-TOF+): m/z calcd for [C52H46ClN2P2Pd]+: 901.1854 [M +
H]+; found: 901.1855. E.A. calcd for C52H46Cl2N2P2Pd C, 65.79;
H, 5.11; N, 3.08%; found: C, 66.57; H, 4.94; N, 2.99.
0.012 mmol) and CD2Cl2 (0.5 mL). After the addition of the IL
(0.123 mmol), the solution was stirred for 0.5 h. Subsequently,
I (6.2 mg, 0.024 mmol), dimethyl malonate (II) (3.2 mg,
0.024 mmol, 2.8 μL), N,O-bis(trimethylsilyl)acetamide (BSA)
(24.8 mg, 0.024 mmol, 6.0 μL) and solid KOAc (1 mg) were
added and stirred manually for 10 min, and then analysed by
1H and 31P NMR. The same operation was repeated after
addition of II (3.2 mg, 0.024 mmol, 2.8 μL) and BSA (24.8 mg,
0.024 mmol, 6.0 μL).
Catalysed allylic alkylation of rac-1,3-diphenyl-3-acetoxyprop-
1-ene (I)
General procedure using imidazolium-based IL’s. A solution
of [PdCl(η3-C3H5)]2 (1.8 mg, 0.005 mmol) and (S)-BINAP in the
corresponding ratio (L/Pd = 1.25 and 2.5) in CH2Cl2 (1 mL) was
stirred for 0.5 h. Subsequently, the IL was added (1 mL) and
CH2Cl2 was removed under reduced pressure. Using a micro-
pipette, I (126 mg, 0.5 mmol, 117 μL), dimethyl malonate (II)
(198 mg, 1.5 mmol, 126 μL), N,O-bis(trimethylsilyl)acetamide
(BSA) (305 mg, 1.5 mmol, 366 μL) and solid KOAc (2 mg) were
added to start the catalytic reaction. Aliquots were taken from
the reaction mixture at certain time intervals, diluted with
diethyl ether, washed with saturated aqueous ammonium
chloride solution, filtered over silica using diethyl ether as the
eluent and analysed by HPLC. When catalyst 1 was employed,
it was dissolved in the IL (1 mL) overnight at 20 °C, and the
catalytic reaction started with the addition of I, II, BSA and
KOAc in the appropriate amounts (see above).
General procedure for recycling experiments. At the end of
the reaction, the product (III) was extracted with dry hexane
(8 × 3 mL) and the ionic liquid was dried for 3 h at 60 °C and
stirred in order to remove any trace solvent. The IL-catalytic
system was reused for another catalytic reaction by simply
adding I, II, BSA and KOAc in the appropriate amounts
(see above).
Acknowledgements
Authors thank the DGAPA PAPIIT-IN231211 project and Pos-
doctoral Fellowship Program of UNAM.
Notes and references
1 (a) Ionic liquids in synthesis, ed. P. Wasserscheid and
T. Welton, Wiley-VCH, Wienheim, 2nd edn, 2008, vol. 2;
(b) T. Welton, Coord. Chem. Rev., 2004, 248, 2459;
(c) H. Olivier-Bourbigou, L. Magna and D. Morvan, Appl.
Catal., A, 2010, 373, 1.
2 (a) L. Xu and J. Xiao, in Recoverable and Recyclable Catalysts,
ed. M. Benaglia, John Wiley & Sons, Ltd, Chichester, 2009,
p. 259; (b) A. F. Trindade, P. M. P. Gois and
C. A. M. Afonso, Chem. Rev., 2009, 109, 418.
3 (a) I. Kmentová, B. Gotov, E. Solcániováb and S. Toma,
Green Chem., 2002, 4, 103; (b) S. Toma, B. Gotov,
I. Kmentová and E. Solcániová, Green Chem., 2000, 2, 149.
4 (a) J. Dupont and J. Spencer, Angew. Chem., Int. Ed., 2004,
43, 5296; (b) N. D. Clement, K. J. Cavell, C. Jones and
C. J. Elsevier, Angew. Chem., Int. Ed., 2004, 43, 1277;
(c) J. D. Scholten, G. Ebeling and J. Dupont, Dalton Trans.,
2007, 5554.
5 (a) H. Lebel, M. K. Janes, A. B. Charette and S. P. Nolan,
J. Am. Chem. Soc., 2004, 126, 5046; (b) A. R. Chianese,
A. Kovacevic, B. M. Zeglis, J. W. Faller and R. H. Crabtree,
Organometallics, 2004, 23, 2461.
6 (a) E. A. B. Kantchev, C. J. O’Brien and M. J. Organ, Angew.
Chem., Int. Ed., 2007, 46, 2768; (b) S. Diez-González,
N. Marion and S. P. Nolan, Chem. Rev., 2009, 109, 3612.
7 (a) A. Flahaut, S. Roland and P. Mangeney, J. Organomet.
Chem., 2007, 692, 5754; (b) S. Roland, W. Cotet and
P. Mangeney, Eur. J. Inorg. Chem., 2009, 1796; (c) N. Toselli,
D. Martin and G. Buono, Org. Lett., 2008, 10, 1453.
8 I. Favier, A. Balanta-Castillo, C. Godard, S. Castillón,
C. Claver, M. Gómez and E. Teuma, Chem. Commun., 2011,
47, 7869.
General procedure for recharging experiments. At the end
of the reaction, the IL-catalytic system was recharged for
another catalytic reaction by simply adding I (0.5 mmol), II
and BSA in a I/II/BSA 1 : 1 : 1 ratio.
General procedure using [HDBU][OAc]. A solution of [PdCl-
(η3-C3H5)]2 (1.8 mg, 0.005 mmol) and (S)-BINAP in the corres-
ponding ratio (L/Pd = 1.25) in CH2Cl2 (1 mL) was stirred for
0.5 h. Subsequently, [HDBU][OAc] was added (1 mL) and
CH2Cl2 was removed under reduced pressure. Using a micro-
pipette, I (126 mg, 0.5 mmol, 117 μL), II (99 mg, 0.75 mmol,
86 μL) and 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) (76 mg,
0.5 mmol, 75 μL) were added to start the catalytic reaction. The
reaction was stirred at 50 °C for 2 h. At the end of the reaction,
product III was extracted with hexanes (8 × 3 mL), the IL was
dried under reduced pressure at 60 °C for 3 h with stirring.
The system was charged with I, II and DBU in the quantities
described above. The recycling was repeated for 4 cycles.
7538 | Dalton Trans., 2014, 43, 7533–7539
This journal is © The Royal Society of Chemistry 2014

View Article Online
Dalton Transactions
Paper
9 V. De la Fuente, N. Fleury-Brégeot, S. Castillón and 19 (a) L. J. Xu, W. P. Chen and J. L. Xiao, Organometallics,
C. Claver, Green Chem., 2012, 14, 2715.
2000, 19, 1123; (b) C. J. Mathews, P. J. Smith, T. Welton,
A. J. P. White and D. J. Williams, Organometallics, 2001, 20,
3848.
10 B. M. Trost and D. J. Murry, Organometallics, 1985, 4, 1143.
11 A. M. Johns, M. Utsunomiya, C. D. Incarvito and
J. F. Hartwig, J. Am. Chem. Soc., 2006, 128, 1828.
12 G. Yu, D. Zhao, L. Wen, S. Yang and X. Chen, AIChE J.,
2012, 58, 2885.
20 Complete characterization of [Pd((S)-BINAP)2], reported in:
F. Ozawa, A. Kubo, Y. Matsumoto and T. Hayashi, Organo-
metallics, 1993, 12, 4188.
13 Even when there is no reported viscosity data for [HDMIM]- 21 D. Brissy, M. Skander, P. Retailleau, G. Frison and
[BF4] (B), it was evidently more viscous than the rest of the A. Marinetti, Organometallics, 2009, 28, 140.
ILs used, therefore catalytic reactions were performed at 22 (a) A. T. Normand, A. Stasch, L.-L. Ooi and K. J. Cavell,
50 °C.
Organometallics, 2008, 27, 6507; (b) S. Fantasia and
S. P. Nolan, Chem.–Eur. J., 2008, 14, 6987; (c) S. Zheng,
Y. Wang, C. Zhang, J. Liu and C. Xia, Appl. Organomet.
Chem., 2014, 28, 48.
14 (a) B. M. Trost and D. L. van Vranken, Chem. Rev., 1996, 96,
395; (b) G. Helmchen, U. Kazmaier and S. Förster, in Cata-
lytic asymmetric synthesis, ed. I. Ojima, Wiley-VCH,
New York, 2010, p. 497.
15 L. A. Evans, N. Fey, J. N. Harvey, D. Hose, G. C. Lloyd-Jones,
P. Murray, G. Orpen, R. Osborne, G. J. J. Owen-Smith and
M. Purdie, J. Am. Chem. Soc., 2008, 130, 14471.
16 P. Fristrup, T. Jensen, J. Hoppe and P.-O. Norrby, Chem.–
Eur. J., 2006, 12, 5352.
23 S. Gladiali, R. Taras, R. M. Ceder, M. Rocamora, G. Muller,
X. Solans and M. Fort-Bardia, Tetrahedron: Asymmetry,
2004, 15, 1477.
24 Similar synthesis followed for [PdCl2((R)-BINAP)] reported
in: G. Celentano, T. Benincori, S. Radaelli, M. Sada and
F. Sannicolò, J. Organomet. Chem., 2002, 643–644, 424.
17 (a) H. Takaya, K. Mashima, K. Koyano, M. Yagi, 25 C. Cassol, G. Ebeling, B. Ferrera and J. Dupont, Adv. Synth.
H. Kumobayashi, T. Taketomi, S. Akutagawa and R. Noyori,
J. Org. Chem., 1986, 151, 629; (b) C. Petit, A. Favre-Reguil-
lon, B. Albela, L. Bonneviot, G. Mignani and M. Lemaire,
Organometallics, 2009, 28, 6379.
Catal., 2006, 348, 243. A similar procedure as described in
this reference was followed for the synthesis of [HDMIM]-
[BF4] (see ESI†).
26 I. Hemeon and R. D. Singer, J. Mol. Catal. A: Chem., 2004,
214, 33.
18 (a) M. Yamaguchi, M. Yakubi, T. Yamagishi, K. Sakai and
T. Tsubomura, Chem. Lett., 1996, 241; (b) A. Rosas, MSc, 27 A.-G. Ying, L. Liu, G.-F. Wua, G. Chen, X.-Z. Chen and
UNAM, 2012.
W.-D. Ye, Tetrahedron Lett., 2009, 50, 1653.
This journal is © The Royal Society of Chemistry 2014
Dalton Trans., 2014, 43, 7533–7539 | 7539