Continuous flow homogeneous alkene metathesis with built-in catalyst separation

Article abstract of DOI:10.1039/c1gc15048k

Continuous flow homogeneous alkene metathesis using a supported ionic liquid phase (SILP) catalyst with CO₂ as a transport vector allows the self-metathesis of methyl oleate with only a slight loss of activity for at least 10 h; cross-metathesis of dimethyl maleate with methyl oleate ceases after 3 h, but the catalyst remains active for methyl oleate metathesis. The reasons for this unusual behaviour are explored and a practical system for the cross-metathesis of methyl oleate with dimethyl maleate, under batch conditions, is described.

Full text of DOI:10.1039/c1gc15048k

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Green Chemistry
Cite this: Green Chem., 2011, 13, 1187
www.rsc.org/greenchem
PAPER
Continuous flow homogeneous alkene metathesis with built-in catalyst
separation†
a
a
a
b
a
b
¨
Rube´n Duque, Eva Ochsner, Herve´ Clavier,‡ Fre´deric Caijo,§ Steven P. Nolan,* Marc Mauduit* and
David J. Cole-Hamilton*a
Received 13th January 2011, Accepted 8th February 2011
DOI: 10.1039/c1gc15048k
Continuous flow homogeneous alkene metathesis using a supported ionic liquid phase (SILP)
catalyst with CO2 as a transport vector allows the self-metathesis of methyl oleate with only a
slight loss of activity for at least 10 h; cross-metathesis of dimethyl maleate with methyl oleate
ceases after 3 h, but the catalyst remains active for methyl oleate metathesis. The reasons for this
unusual behaviour are explored and a practical system for the cross-metathesis of methyl oleate
with dimethyl maleate, under batch conditions, is described.
Solutions to circumvent this problem of product/catalyst
Introduction
separation have been addressed,7 in particular by carrying out
Alkene metathesis is a reaction in which two alkenes undergo
reactions in an ionic liquid.8 In this medium, ionic liquid-
scrambling of the chains attached to the double bonds and has
immobilized catalysts are soluble and importantly are not
become very important in the construction of a wide array of
removed during product isolation by simple liquid/liquid ex-
unsaturated molecules.1 Applications include the construction
traction using an appropriate organic solvent, which allows for
of cyclic compounds by ring-closing metathesis (RCM),2 ring-
catalyst recycling at the end of the reaction. As an example, with
opening metathesis polymerisation (ROMP),3 allowing the
formation of linear polymers from cyclic starting materials, the
complex 3, used in the present study, it is possible to perform
eight consecutive RCM reactions of substrate 1 with quantitative
adjustment of alkene chain length as in the Shell higher olefins
yields using a 2.5 mol% catalyst loading (Scheme 1).9 More
process4 and the formation of polymer composites for use in,
importantly, product contamination in ruthenium was found to
for example, materials fabrication. Since they are more tolerant
be low (1–22 ppm). These latter experiments were conducted
to functional groups than molybdenum complexes, ruthenium-
in batch mode, which is not time or energy efficient. So, for
based catalysts have been extensively studied.5 Nevertheless,
with the exception of polymerization processes, rather few
commercial applications have ensued.6 Besides the high catalyst
cost, the main reason for this industrial underdevelopment,
particularly in the pharmaceutical area, may be associated
with issues concerning the separation of metal waste from the
product, which is time-consuming and expensive.7
example, eight consecutive 3 h batch reactions produced a total
-1
turnover number of 386 molproduct molcatalyst
.
Scheme 1 RCM using an ionic liquid-tagged boomerang catalyst.
aEaStCHEM, School of Chemistry, University of St. Andrews, St.
Andrews, Fife, Scotland, UK KY16 9ST. E-mail: djc@st-and.ac.uk;
Fax: +44 1334-463808; Tel: +44 1334 463805
Having recently developed a system permitting continuous
flow catalysis,10 we now report the first robust examples of
continuous flow homogeneous olefin metathesis. For this, a
supported ionic liquid phase (SILP) catalyst consisting of 3
immobilised in a thin film of imidazolium ionic liquid within
the pores of silica, was used with compressed CO2 as the
flowing medium (SILP-SCF). Such systems have been used
before for hydroformylation10 and asymmetric hydrogenation,11
and they offer advantages over systems where liquids flow
through SILP catalysts12 because the solubilities of the ionic
liquid and catalyst are greatly reduced by the presence of CO2.
bUMR CNRS 6226 "Sciences Chimiques de Rennes", Ecole Nationale
Supe´rieure de Chimie de Rennes, Avenue du Ge´ne´ral Leclerc – CS 50837,
35700, Rennes cedex 7, France. E-mail: marc.mauduit@ensc-rennes.fr;
Fax: +33 2 23 23 81 08; Tel: +33 2 23 23 81 12
† Electronic supplementary information (ESI) available: Experi-
mental detail and graph for ring closing metathesis. See DOI:
10.1039/c1gc15048k
‡ Current address UMR CNRS 6180 “Chirotechnologies”, Laboratoire
de synthe`se asyme´trique, Universite´ Paul Ce´zanne, Avenue Escadrille
Normandie Niemen, 13397, Marseille Cedex 20, France.
§ Current address: OmegaCatSystemꢀR -Ecole Nationale Supe´rieure de
Chimie de Rennes, Av. du Ge´ne´ral Leclerc, 35700, Rennes, France.
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1187–1195 | 1187
©

View Article Online
They also offer advantages over all gas phase systems using
SILP catalysts13 because the flow rate of less volatile substrates
and products can be much higher. SILP-SCF systems also
offer potential advantages over systems where the catalyst is
immobilised by dissolving it in a bulk ionic liquid as substrates
and products are transported through this solution dissolved in
scCO2 (SCF-IL).14 SILP-SCF systems can be used below the
critical point of the flowing medium (i.e. using gas expanded
liquids), allowing lower pressure operation and better contact
between the substrate and the catalyst. Much less ionic liquid is
required in the SILP-SCF system than in the SCF-IL system.
Significant progress has been made in the preparation of
metathesis catalysts covalently attached to solid supports,15
but there are rather few examples of their use in con-
tinuous flow processing where the substrates and/or prod-
ucts are of relatively low volatility. Flow metathesis has
been reported using a similar catalyst to ours supported on
Raschig rings. Leaching and deactivation in this case are
severe.15e A related metathesis catalyst has been immobilised
within a PEEK column on a monolith prepared by the
ROMP of norbornene (NBE) and (NBE-CH2O)3SiCH3 using
[Ru( CHPh)Cl2(PCy3)2] in the presence of suitable porogens.
Further metathesis of the monolith, still containing the catalyst,
with norborn-5-en-2-ylmethylhexafluoroglutonate gave an acid
resin that was converted to its silver salt before reaction with
[Ru( CH(2-C6H4OPri))Cl2(IMesH2)] (IMesH2 = 1,3-dimesityl-
4,5-dihydroimidazol-2-ylidene) to give a supported metathesis
catalyst bound to the resin through a Ru-carboxylate linkage.15f
This catalyst was used under continuous flow conditions for
the RCM of diethyl 2,2-diallylmalonate (as in Scheme 1) in
dichloromethane. The catalyst remained active for 140 min,
although the activity dropped regularly with time (initial yield
70% dropping to < 10% at the end of the reaction). Even
so, 480 turnovers were obtained within the 140 min period.
Ruthenium (<0.2%) and ligand leaching were very low. Using
a size expanded (through substituents on the N-heterocyclic
carbene N atoms) catalyst, continuous flow metathesis on
the same substrate has also been carried out in a membrane
reactor.15g The conversion rose to ca. 35% after 100 min but then
dropped steadily to < 10% after 500 min. Overall, turnover was
866, and 97.6% of the ruthenium was retained in the reactor.
The drop in reactivity with time in these cases suggests that
catalyst deactivation may be occurring, possibly because of a
build up of ethene in the system.15g Perhaps the most successful
flow system reported15h to date involves depositing a Grubbs–
Hoveyda catalyst on silica by the evaporation of a toluene
solution. This has been successfully used as a low leaching
supported catalyst for the metathesis of a variety of alkenes
and gives up to 4000 catalyst turnovers in 2 h in the continuous
flow ring-opening metathesis of cyclooctene, with ruthenium
leaching as low as 7 ppb. Non-polar organic solvents were used
as the transport vector but the system is not really suitable for
polar substrates because the catalyst is washed away.
diallylmalonate was purchased from Aldrich and used without
further treatment. 1-Octene, 2-octene, methyl oleate, dimethyl
maleate, diethyl maleate and diethyl fumarate were purchased
from Aldrich, distilled and passed through a silica plug before
use. Silica (Sigma-Aldrich, Merck silica gel 100, grade 10184, 70–
230 mesh, 10 nm pore diameter) was purchased from Aldrich
and heated to 500 ◦C for several hours before use.
Grubbs–Hoveyda 2nd generation catalyst was purchased
from Aldrich. Catalyst 3,9a BMIM·NTf2 and OMIM·NTf2
14c
were prepared by literature methods.
1H, 13C and 19F NMR spectra were recorded using a Bruker
400 spectrometer. Gas chromatography was performed using a
Hewlett-Packard 6890 series gas chromatograph equipped with
an Agilent 6890 series injector. Analyses were performed with
a Hewlett-Packard 5973 series mass selective detector and the
peak areas integrated using a Hewlett-Packard Chemstation.
Compound 4 was synthesised by a method adapted from the
literature.9d
3-(4-Isopropoxy-3-(prop-1-en-1-yl)phenyl)propan-1-ol
A Schlenk flask was charged with dry toluene (30 mL) and
tetrakis(triphenylphosphine)palladium (370 mg, 0.321 mmol,
0.07 equiv.). The mixture was de-gassed and 3-(3-bromo-4-
isoproxyphenyl)propan-1-ol (1.25, 4.58 mmol) diluted in dry
toluene (5 mL) was added dropwise through a syringe. The
resulting mixture was stirred for 15 min before adding tributyl-
propenylstannane (2.1 mL, 6.86 mmol, 1.5 equiv.). The flask
was then heated at 110 ◦C overnight. After cooling to room
R
temperature, the mixture was filtered on a plug of Celiteꢀ and
the cake washed with diethylether. The solvent was evaporated
under vacuum and purification by silica gel using pentane–ethyl
acetate (8/2) as the eluent afforded the desired product as a
colourless oil (878 mg, 3.75 mmol, 82%).
1H NMR (400 MHz, CDCl3) d 7.25–7.05 (m, 1H), 6.98 (ddd,
J = 18.0, 8.4, 2.2 Hz, 1H), 6.82 (dd, J = 12.5, 8.4 Hz, 1H), 6.75–
6.44 (m, 1H), 6.02 (ddd, J = 18.0, 12.5, 6.8 Hz, 1H), 4.48 (dd,
J = 12.3, 6.2 Hz, 1H), 4.30–4.19 (m, 2H), 3.00 (d, J = 1.9 Hz,
2H), 2.69 (dd, J = 15.0, 7.6 Hz, 2H), 1.88 (dd, J = 6.9, 1.8 Hz,
3H), 1.34 (dd, J = 8.0, 6.1 Hz, 6H).
4-(3-Bromopropyl)-1-isopropoxy-2-(prop-1-en-1-yl)benzene
To a solution of 3-(4-isopropoxy-3-(prop-1-en-1-yl)phenyl)-
propan-1-ol (9.817 g, 41.89 mmol) and triethylamine (8.8 mL,
62,84 mmol, 1.5 equiv.) in dry dichloromethane (200 mL) was
added methanesulfonyl chloride (4.86 mL, 62.84 mmol, 1.5
equiv.) at 0 ◦C. The reaction mixture was stirred for 3.5 h at
room temperature before being diluted with dichloromethane.
The organic phase was washed four times with a 5% citric acid
solution, dried over sodium sulfate and concentrated to a yellow
oil. The product was used without further purification in the
following reaction.
To
a crude mixture of 3-(4-isopropoxy-3-(prop-1-en-1-
yl)phenyl)propyl methanesulfonate (13,09 g, 41.89 mmol) in
tetrahydrofuran (200 mL) and dimethylformamide (78 mL) was
added lithium bromide (7.28 g, 83.78 mmol, 2 equiv.) in one
portion. The mixture was stirred overnight. After evaporation
of the solvent, the residue was diluted in ethyl acetate. The
organic layer was washed three times with a saturated sodium
Experimental
All manipulations were carried out using standard Schlenk
techniques and oven dried glassware under a nitrogen atmo-
sphere. CO2 (99.9995%) was purchased from BOC gases. Diethyl
1188 | Green Chem., 2011, 13, 1187–1195
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
hydrogencarbonate solution, then with brine, and dried over
magnesium sulfate, filtered and concentrated. Purification by
silica gel chromatography using pentane–ethyl acetate (98/2) as
the eluent afforded the desired product as a colourless oil (9.34 g,
31.42 mmol, 75% for two steps).
1H NMR (400 MHz, CDCl3) d 7.18 (dd, J = 50.0, 2.2 Hz, 1H),
7.04–6.91 (m, 1H), 6.81 (dd, J = 13.0, 8.3 Hz, 1H), 6.74–6.48 (m,
1H), 6.29–5.71 (m, 1H), 4.48 (dd, J = 12.3, 6.1 Hz, 1H), 3.41 (td,
J = 6.6, 4.5 Hz, 2H), 2.72 (dd, J = 15.6, 7.7 Hz, 3H), 2.20–2.11
(m, 2H), 1.88 (ddd, J = 21.8, 6.9, 1.8 Hz, 3H), 1.34 (dd, J = 8.0,
6.0 Hz, 6H).
a vertical position and the flow being from bottom to top.
The tubular reactor previously loaded with the SILP catalyst
was fixed onto the rig after purging it with a flow of CO2.
The rig was pressurised to the desired pressure and the reactor
heater and pre-heater raised to the desired temperature. After
equilibration of the system, the substrate was pumped in with
an HPLC pump, and liquid CO2 was pumped through a Pickel
pump to maintain the total pressure. The flow stream was
decompressed through a decompression valve to atmospheric
pressure into a collection vessel, from which the gas was passed
through a flow meter, where the total gas flow was monitored
at atmospheric pressure. The liquid in the collection vessel was
removed periodically, weighed to check the mass balance, and
analysed by GC-MS (organic products), NMR (IL leaching)
and ICP-OES (Ru leaching).
3-(3-(4-Isopropoxy-3-(prop-1-en-1-yl)phenyl)propyl)-1-methyl-
1H-imidazol-3-ium hexafluorophosphate (4)
A 100 mL round-bottomed flask equipped with a condenser was
charged with 4-(3-bromopropyl)-1-isopropoxy-2-(prop-1-en-1-
yl)benzene (4.15 g, 13.96 mmol), methylimidazole (2.22 mL,
27.92 mmol, 2 equiv.) and dry toluene (28 mL). The mixture
was stirred overnight at 100 ◦C and the solvent evaporated.
The residue was dissolved in distilled water (200 mL) and
hexafluorophosphoric acid (5.14 g, 27.92 mmol, 2 equiv.) was
slowly added dropwise. After 10 min stirring, dichloromethane
and brine were added. The organic phase was washed with
brine until a neutral pH was reached, dried over magnesium
sulfate, filtered and concentrated. Purification by silica gel
chromatography using dichloromethane/acetone (3/1) as the
eluent afforded the desired product as a colourless oil (5.548 g,
12.48 mmol, 89% for two steps).
1H NMR (400 MHz, MeOD) d 8.77 (d, J = 17.1 Hz, 1H), 7.53
(ddt, J = 33.0, 10.8, 1.8 Hz, 2H), 7.13–6.94 (m, 2H), 6.87 (dd, J =
15.1, 8.4 Hz, 1H), 6.70–6.38 (m, 1H), 6.33–5.64 (m, 1H), 4.51
(ddd, J = 12.1, 7.8, 5.8 Hz, 1H), 4.32–4.08 (m, 2H), 3.32 (dt,
J = 3.3, 1.6 Hz, 3H), 2.64 (td, J = 7.3, 4.6 Hz, 2H), 2.20 (ddd,
J = 15.4, 8.9, 5.3 Hz, 2H), 1.83 (ddd, J = 31.4, 6.8, 1.8 Hz, 3H),
1.43–1.20 (m, 6H).
31P NMR (162 MHz, MeOD) d -144.56 (hept, J = 708.1 Hz).
19F NMR (376 MHz, MeOD) d -73.94 (dd, J = 363.4, 356.0
Hz).
Substrate ratios and flow rates are described in tables later in
the article.
Batch experiments
In a typical batch reaction, methyl oleate (5 g, 16.8 mmol) was
mixed with dimethyl maleate (9.7 g, 67.3 mmol) in a ratio of
1 : 4. Catalyst 3 (0.0308 g, 0.034 mmol, 0.2 mol% MO) was
dissolved in OMIM·NTf2 (5 mL) and heated at 50 ◦C. The
substrate mixture was added to the catalyst solution. Samples
were taken periodically and quenched with ethyl vinyl ether prior
to GC-MS analysis.
Results and discussion
A SILP catalyst was used for the continuous flow SILP-SCF
experiments. 3, prepared as described previously,9a was dissolved
in dichloromethane containing 1-butyl-3-methylimidazolium
triflamide (BMIM·NTf2) and silica (Sigma-Aldrich, Merck silica
gel 100, grade 10184, 70–230 mesh, 10 nm pore diameter),
previously heated to 500 ◦C to remove surface hydroxides, was
added. After evaporation of the dichloromethane, the catalyst, a
free-flowing pale green powder, was loaded into a tube in a glove
box and fixed onto the previously described reactor10b under a
CO2 flow. The CO2 was passed through the reactor while the tube
was heated to the desired reaction temperature, and the substrate
flow initiated once the temperature had stabilized (10 min). The
effluent exiting the tube was passed through a decompression
valve and the products condensed into a collection vessel. The
collected material was removed from the vessel every hour and
analysed by GC for organic products, by inductively coupled
plasma optical emission spectroscopy (ICPOES) to determine
the ruthenium content, and by 1H and 19F NMR spectroscopy to
analyze the organic products and track possible traces of leached
BMIM·NTf2. In most cases, mass balance between substrate
and product was found to be close to 100%, assuming, where
relevant, that the ethene and but-2-ene were lost, as is confirmed
by GC analysis.
13C NMR (101 MHz, MeOD) d (two isomers) 155.33, 133.65,
133.08, 131.24, 129.50, 129.16, 128.78, 128.66, 127.25, 127.12,
127.00, 126.95, 126.67, 124.87, 123.61, 115.94, 115.88, 72.10,
36.39, 32.79, 32.71, 32.43, 22.50, 22.48, 19.05, 14.99.
Catalyst preparation
A typical preparation of the SILP catalyst was performed
by dissolving catalyst 3 (0.0072 g, 0.02 wt% of silica),
BMIM·NTf2 (1.276 g, 29 wt% of silica) and silica gel (4.4 g)
in purified dichloromethane (20 mL). After 2 h stirring, the
dichloromethane was removed at 35 ◦C under vacuum to recover
the SILP catalyst, which, in a glove box, was charged into a
15 cm3 tubular reactor fitted with glass wool at each end to
avoid it being flushed away during the reaction. The catalyst was
stored in a glove box until its use.
Our initial experiments focused on the RCM of diethyl diallyl-
malonate (1; run 1, Table 1). Fig. 1 highlights the good activity
of the ruthenium complex during the first hour; nevertheless,
after this period, the catalyst performance decreased, possibly
due to its degradation. Nonetheless, considering the low catalyst
loading and the fast flow rate, high turnovers were obtained
Continuous flow experiments
The catalytic reactions were carried out in a reactor, the design
of which has previously been described,10b with the reactor in
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1187–1195 | 1189
©

View Article Online
Table 1 Conditions for continuous flow ring-closing or self-metathesis using 3 under SILP conditionsa
Catalystb loading
(Ru% w/w)
Substrate flow
rate/cm3 min-1
Conversion after 3 h
on stream (%)
Run
Substrate
T/◦C
Time on stream/h
TONd
1
2
3
4
5
6
7
1
50
23
50
50
50
50
23
0.18
0.1
0.2
0.2
0.2
0.1
0.05
0.05
6
7
6
6
5
10
9
7.2
2.4
0.6
0
40
66.9
57.2
242
564
550
28
3371
4247
10148
1
0.048
0.022
0.022
0.024
0.020c
0.007c
1
1-Octene
2-Octene
MO
MO
a BMIM·NTf2 (29% w/w on silica; mass of silica = 3.4 g), CO2 flow, pressure = 100 bar, total flow = 645 cm3 min-1 at NTP, MO = methyl oleate. b g (g
-1
silica-1). c SiO2 = 4.4 g. d Total molproduct molcatalyst at the end of the reaction.
Fig. 1 Continuous flow RCM of 1 catalysed by 3 in a SILP system (for
conditions, see Experimental section and Table 1).
Fig. 2 Continuous flow self-metathesis of 2-octene and methyl oleate
catalyzed by 3 in a SILP system (for conditions, see Table 1).
in a short time period. By reducing the catalyst loading and
increasing the flow rate of the substrate (runs 2 and 3, Table 1
and Fig. 1), it was possible to obtain a turnover number of 564
in 7 h, but the conversions were found to be low and dropped
significantly with time. This behaviour is similar to that in other
flow metathesis systems. Although more turnover was obtained
than in repetitive batch reactions in a shorter time, the instability
of the catalyst makes this system unacceptable for use with this
type of substrate.
It is known that ruthenium methylidenes, which represent the
true active species when terminal alkenes are used as substrates,
can be degraded by dimerization, and hence their stability is a
crucial issue for olefin metathesis.16 In order to test whether this
was in fact the problem, we studied the metathesis of 1- and
2-octenes using the same catalytic system.
and 9-octadecene. We have shown that internal alkenes of this
kind can be selectively methoxycarbonylated at the terminal
position by a tandem isomerisation—methoxycarbonylation
process (Scheme 2b).18 The products may be important in
detergents. We therefore examined the metathesis of methyl
oleate in the flow system conceptualised in Fig. 3 using catalyst
3 (runs 6 and 7, Table 1 and Fig. 2). Using a flow rate of
0.05 cm3 min-1 at 50 ◦C, up to 64% conversion was obtained
over the first 6 h. At equilibrium, the conversion would be
66.7%. From the 7th hour, the conversion dropped steadily
to 44% after 10 h, showing some catalyst instability. The total
number of catalyst turnovers in this period was 4247. Even more
impressively, on decreasing the temperature to 23 ◦C and the
For 1-octene (run 4, Table 1), only 0.9% conversion was
observed in the sample collected after 1 h, representing 28
turnovers, but subsequent samples contained no metathesis
products. These observations confirm that the system is not
stable for the metathesis of terminal alkenes.
We therefore turned our attention to internal alkenes and used
the self-metathesis of 2-octene to 3-hexene and 2-butene as a
benchmark substrate (run 5, Table 1, Scheme 2a and Fig. 2). For
this transformation, catalyst 3 exhibited an enhanced stability,
providing a turnover frequency of 813 h-1 at the end of the
second hour on stream, with this only dropping to 707 h-1 after
5 h. The total turnover in this case was 3371 after 5 h.
The self-metathesis of oleochemicals is of considerable interest
for the conversion of low value feedstocks into useful chemicals
in petrochemistry and polymerization.17 The self-metathesis
of methyl oleate (Scheme 2b) yields dimethyl 1,18-octadec-9-
enedioate, which may be valuable for materials applications,
Fig. 3 Conceptual visualisation of the SILP process for the self-
metathesis of methyl oleate catalysed by 3.
1190 | Green Chem., 2011, 13, 1187–1195
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Scheme 2 (a) Self-metathesis of 2-octene; (b) self-metathesis of methyl oleate, showing a possible downstream reaction of 9-nonadecene to a
detergent ester; (c) cross-metathesis of methyl oleate with dimethyl maleate, showing a possible route to dimethyl 1,12-dodecanedioate and dimethyl
1,11-undecendioate.
catalyst loading to 77 ppm of Ru on SiO2 (run 6, Table 1, and
Fig. 2), high conversions were still obtained (58% at 2 h and 41%
at 9 h), with a total turnover number of >10 000 after 9 h. Much
higher turnover numbers (up to 470 000) have been obtained in
the cross-metathesis of methyl oleate with 1-butene by using very
low catalyst loadings in batch reactors, but the catalyst, albeit in
small amounts, remained in the reaction products.19 Although
the catalyst in our system exhibited a relatively good stability,
all the reactions presented some activity loss in time (downward
curvature from linear in Fig. 2).
Because of our interest in a,w-difunctionalised materials, we
Fig. 4 Conversions in the continuous flow cross-metathesis of methyl
were interested in the potential of the metathesis reaction to
form such compounds by cross-metathesis. Therefore, the cross-
metathesis of methyl oleate with dimethyl maleate. (Scheme 2c)
was investigated. This reaction should produce two C11 products,
one an unsaturated a,w-diester and the other an a,b-unsaturated
terminal ester, which in turn might be carbonylated to dimethyl
1,12-dodecandioate using chemistry we have previously de-
scribed (Fig. 2c).20 These shorter chain products are useful as
bio-derived feedstocks for applications such as the synthesis of
nylon 11,11 or 12,12.19,21 This cross-metathesis reaction between
methyl oleate and dimethyl maleate has been suggested before
in the literature but has not apparently been carried out.17b
We performed catalysis using a dimethyl maleate/methyl oleate
molar ratio of either 4 : 1 (run 1, Table 2 and Fig. 4) or 8 : 1
(run 2, Table 2 and ESI Fig. S1†) to favour the formation of
cross-metathesis products. In both cases, the results were found
to be similar. At the early stage of the reaction, cross-metathesis
mainly occurred, but rapidly homo-metathesis became more
competitive and then the major process. After 7 h, cross-
oleate with dimethyl maleate catalysed by 3 in a SILP system (run 1,
Table 2), and comparison to the self-metathesis of methyl oleate (run 6,
Table 1). For conditions, see the tables.
metathesis had completely stopped, but the self-metathesis of
methyl oleate continued until the reaction was stopped after 8 h,
although the conversion dropped after about 6 h. The conversion
of methyl oleate to self-metathesis products was still occurring
with a conversion of 38% in the 8th hour.
These rather surprising results suggest that the nature of
the catalysts evolve during the course of the reaction. Cross-
metathesis is observed during the early part of the reaction
but self-metathesis dominates later on. Turnover frequencies
started at 664 molproduct (molcatalyst h)-1 to a mixture of self- and
cross-metathesis products but decreased during the reaction to
391 h-1 exclusively to methyl oleate self-metathesis products,
giving a total turnover number of 3962 molproduct molcatalyst-1 after
8 h of reaction. Leaching from both the IL and the catalyst
were low, the ICP-OES analysis of the products showing a
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1187–1195 | 1191
©

View Article Online
Table 2 Methyl oleate and dimethyl maleate cross-metathesis parameters using 3 under SILP conditionsa
Dimethyl maleate :
methyl oleate
Temperature
(◦C)
Substrate
Catalyst loading
(wt% Ru/SiO2)b
Cross TON
after 5 h
Total TON
after 5 hc
Run
Pressure/bar
flow/mL min-1
1
2
3
4
4
8
8
4
4
4
4
8
8
4
8
8
50
50
r.t.
r.t.
70
50
50
50
50
50
50
50
100
100
30
0.15
0.15
0.15
0.15
0.15
0.1
0.1
0.1
0.15
0.1
0.15
0.15
0.023
0.024
0.021
0.020
0.020
0.021
0.022
0.019
0.019
0.020
0.019
0.018
502
360
223
56
359
1226
265
287
15
2607
1409
981
100
100
100
100
100
100
100
100
100
1911
1641
1813
1701
1397
1780
1705
1727
1447
5
6d
7e
8e
9f
10g
11h
12i
445
847
3
a BMIM·NTf2 (29% w/w on silica; mass of silica = 4.4 g), CO2 flow, pressure = 100 bar, total flow = 645 cm3 min-1 at NTP. b g (g silica-1). c Total
-1
molproduct molcatalyst after 5 h. d 2-Octene in place of dimethyl maleate. e 2-Octene in place of methyl oleate. f 4 (5 mol (mol 3)-1) added. g The dimethyl
maleate flow was started 1 h before the methyl oleate flow. h Diethyl maleate in place of dimethyl maleate and OMIM·NTf2 in place of BMIM·NTF2.
i Diethyl fumarate in place of dimethyl maleate and OMIM·NTf2 in place of BMIM·NTF2.
maximum ruthenium concentration of 8 ppm. The higher excess
of dimethyl maleate (8-fold; run 2, Table 2 and ESI Fig. S1†) did
not produce a higher conversion to cross-metathesis products
but a lower one, the total turnover number for the cross-
-1
metathesis products being 360 molproduct molcatalyst as opposed
-1
to 502 molproduct molcatalyst after 5 h in the previous run with a
4-fold excess of dimethyl maleate. Lowering the temperature to
room temperature or raising it to 70 ◦C (runs 4 and 5, Table 2,
and ESI Fig. S2 and Fig. S3†) did not improve the conversion to
cross-metathesis products. Indeed, the deactivation was faster at
the higher temperature and there was almost no cross-metathesis
at room temperature.
In seeking to understand the results obtained during cross-
metathesis reactions, we carried out a number of additional
experiments.
Fig. 5 Cross-metathesis of methyl oleate with 2-octene. Conditions:
50 ◦C, 100 bar, 0.1 mL min-1 substrate flow, 0.021 wt% Ru/SiO2 and a
4-fold excess of 2-octene.
In order to check that there was no intrinsic problem with
cross-metathesis in the flow system, a reaction was carried
out between 2-octene and methyl oleate. This reaction was
expected to give a large array of products from the desired
cross-metathesis and from the self-metathesis of the two starting
alkenes.
but decreased gradually to zero over 6.5 h. Self-metathesis was
constant at around 50% for 4 h and then dropped slowly.
Because of the dramatic change in selectivity observed in
the cross-metathesis of methyl oleate with dimethylmaleate, we
wondered whether some change in the catalytically active species
might be occurring during the reaction. Boomerang catalysts
are generally believed to work by decoordination of the ether to
create a site at which the alkene coordinates. The alkene then
undergoes metathesis with the carbene from the boomerang
ligand to introduce a carbene derived from the reacting alkene.
The modified boomerang ligand is then believed to remain free
in solution until the end of the reaction, when it undergoes
metathesis with a metal-bound carbene (formed from a reacting
alkene) and recoordinates the ether oxygen.22 Recent studies
involving fluorescence (fluorescent aryl ether) or 19F NMR
studies (fluorinated aryl ether) have, however, cast doubt on
this interpretation.23 We thought it possible that the boomerang
ligand might be required for the activation of dimethyl maleate
but not for methyl oleate. If this were the case, cross-metathesis
would be observed as long as the boomerang ligand were capable
of reacting back with the catalyst, but not if it diffused away from
the active site. We therefore carried out a reaction in which we
added excess of boomerang ligand precursor 4.
Of the 8 possible products, only 7 were recovered because 2-
butene, which has low boiling point, was lost through venting to
the atmosphere. The conversion towards cross-metathesis prod-
ucts was higher (over 60% of the total conversion) throughout
the reaction than that towards self-metathesis products (run 6,
Table 2 and Fig. 5). The conversion was constant throughout
the reaction, suggesting that dimethyl maleate was in some
way responsible for the catalyst deactivation towards cross-
metathesis observed in the reaction between methyl oleate and
dimethyl maleate. This was further confirmed by carrying out
the cross-metathesis of dimethyl maleate with 2-octene. Using a
4-fold excess of dimethylmaleate over 2-octene, the reaction was
reasonably stable (run 7, Table 2 and ESI Fig. S4†), but cross-
metathesis only accounted for 5–11% of the products, whilst
self-metathesis accounted for >40%. Using a larger excess of
dimethyl maleate (8-fold; run 8, Table 2 and ESI Fig. S5†),
cross-metathesis increased to 30% at the start of the reaction,
1192 | Green Chem., 2011, 13, 1187–1195
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
despite the much higher catalyst loading than in the flow system.
What these results demonstrate is that the rate of self-metathesis
of methyl oleate is very much faster than the rate of cross-
metathesis with dimethyl maleate, despite the excess of dimethyl
maleate being used and that the formation of cross-metathesis
products is predominantly a secondary reaction of the methyl
oleate self-metathesis products.
Given the results obtained in batch systems, it is surprising
that we saw any cross-metathesis at all in the flow system,
since the residence time in the reactor under our normal flow
conditions was only 30–130 min and the catalyst loading was
very low. However, what we should find is that cross-metathesis
is increased if the residence time in the reactor is increased. We
have achieved this by reducing the flow rate of the substrate.
With a flow rate of 0.05 cm3 min-1 (residence time = 130 min)
as opposed to 0.15 cm3 min-1 (residence time = 42 min), in most
reactions, the initial conversion to the cross-metathesis products
was over 70% and, although the conversion dropped with time,
it was still >25% after 6 h and remained higher than that to the
self-metathesis products for 5 h (Fig. 7).
Using a 5-fold excess of this compound over ruthenium and
an 8-fold excess of dimethyl maleate over methyl oleate, almost
no cross-metathesis products were observed, only the self-
metathesis of methyl oleate at about 60% conversion dropping
to 52 over 6.5 h (run 9, Table 2 and ESI Fig S6†). We also
found that there was no advantage to be gained from pre-treating
the catalyst with dimethyl maleate under metathesis conditions,
either in the presence or absence (run 10, Table 2) of added 4.
Since dimethyl maleate undergoes self-metathesis to give the E
isomer, dimethyl fumarate, which is very insoluble in the reaction
medium, we wondered whether dimethyl fumarate might be
crystallising in the silica pores, thus lowering the accessibility
of the catalyst available for the reaction. In order to prevent this,
we carried out reactions using diethyl maleate (run 11, Table 2
and ESI Fig. S7†) and diethyl fumarate (run 12, Table 2 and
ESI Fig. S8†), both of which are liquids, in place of dimethyl
maleate. In these reactions, we also replaced BMIM·NTf2 by
OMIM·NTf2. Diethyl maleate gave results very similar to those
obtained with dimethyl maleate, whilst diethyl fumarate only
gave the self-metathesis of methyl oleate (60% throughout the
reaction). Interestingly, using a mixture of diethyl maleate and
diethyl fumarate for the cross-metathesis with methyl oleate gave
results in between those obtained using the two diesters on their
own (ESI Fig. S9†), showing that the fumarate does not actually
act as an inhibitor of the reaction.
Fig. 7 Cross-metathesis of dimethyl maleate with methyl oleate at lower
substrate flow.
Since none of these proposed solutions improved the cross-
metathesis of dimethyl maleate with methyl oleate, we studied
the cross-metathesis under batch conditions in the ionic liquid
OMIM·NTf2. Using a catalyst loading of 1 mol% and a 4-fold
excess of dimethyl maleate, all the methyl oleate was converted
to the desired cross-metathesis products in <10 min. We believe
that this is the first time that this particular cross-metathesis
reaction has been successfully carried out. Lowering the catalyst
loading to 0.2 mol%, the results were as shown in Fig. 6. Initially,
the only products formed were those of the self-metathesis
of methyl oleate. As the products built up, they started to
undergo metathesis with dimethyl maleate so that after 2 h, a
full conversion to the cross-metathesis products was observed. It
should be noted that significant conversion to cross-metathesis
products was not observed for the first 5 min of the reaction,
We propose that the reason for the slightly unusual results
obtained during the cross-metathesis reactions is that at the
start of the reaction, all the catalyst is active and it converts
the methyl oleate to self-metathesis products, which are in
contact with enough catalyst and dimethyl maleate to give
significant amounts of cross-metathesis products. The catalyst,
which is very much more active for self-metathesis than for cross-
metathesis, gradually deactivates to the point where the cross-
metathesis becomes vanishingly small, but the self-metathesis
still reaches an equilibrium position and so appears to be stable.
The catalyst continues to degrade so that, after some time, there
is insufficient active catalyst left to reach an equilibrium position
in the self-metathesis reaction and so the conversion to the
self-metathesis products gradually drops. The reason for the
degradation of the catalyst is not clear. It does not, however,
appear to be associated with the tag on the boomerang ligand,
since using the unmodified Grubbs–Hoveyda catalyst gave very
similar results to those obtained with 3 (ESI, Fig. S10†).
Effect of the flowing medium
Phase behaviour observations have been made on the SILP-SCF
system for 1-octene hydroformylation, and have been correlated
with the reaction rate and extent of IL and Rh leaching.10b At
low pressures or in the absence of CO2, the reaction rate was
Fig. 6 Cross-metathesis of methyl oleate with dimethyl maleate in bulk
OMIM MTf2 (for conditions, see the Experimental section).
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1187–1195 | 1193
©

View Article Online
low and leaching was high because the reaction was carried
out in the liquid phase, in which the ionic liquid has significant
solubility. On increasing the pressure, the rate increased and the
amount of leaching of ionic liquid and catalyst dropped as long
as the reaction was carried out in an expanded liquid. Once
the critical pressure was reached, i.e. once the flowing medium
was a single phase, the rate decreased because the substrate
partitioned less and less well into the ionic liquid film, where
the catalyst resides. For that system, 1-octene, CO, H2 and CO2
at 100 ◦C, the critical pressure was measured to be 106 bar.
For the metathesis reactions, we have typically worked at lower
temperatures, sometimes below the critical temperature of CO2
(31.1 ◦C) but about the same pressure (100 bar), and have used
the much less volatile and soluble methyl oleate as one of the
substrates. This ensures that the reactions are carried out in
liquid methyl oleate expanded by CO2. We have visually observed
that a condensed phase is present under the same flow conditions
as for the cross-metathesis experiments.
Phase behaviour studies have been carried out on ethyl oleate
in CO2 at 50 ◦C between 0 and 120 bar.24 They show that, as the
pressure of CO2 is increased, the reaction rate and equilibrium
position for ethenolysis are both increased because: a) the
viscosity is reduced by the substantial mole fractions of CO2
dissolved, and b) the products are much more soluble in the
vapour phase than in the starting material, hence allowing the
products selectively to be extracted and the equilibrium position
to lie in favour of the products. Above 120 bar, the rate drops
dramatically because the substrates are all in the supercritical
phase and the catalyst remains as a solid at the bottom of the
reactor.
Fig. 8 Samples recovered after 3, 4, 5 and 6 h from the cross-metathesis
of methyl oleate with dimethyl maleate when using N2 as the carrier gas
(i.e. liquid phase flow). Conditions as for entry 2, Table 2, but with N2
(30 bar) in place of CO2. The two phases occur because dimethyl maleate
does not mix with methyl oleate or its cross-metathesis products.
10 h, a slight fall off in activity with time on stream suggesting
some catalyst instability. Total turnover numbers >10 000 over
9 h are possible, with the products containing low ruthenium
and ionic liquid contents. In the most favourable cases, 6 g of
substrate per hour can be converted to an equilibrium mixture of
products in a 9 cm3 reactor. Overall, we have developed a system
for alkene metathesis that allows the products to be obtained
with high yields and with very low metal contamination.
No solvent is present in the collected product and the only
purification needed is fractional distillation of the equilibrium
mixture of products.
The cross-metathesis of methyl oleate with diethyl maleate can
be carried out in the ionic liquid OMIM·NTf2 with a catalyst
loading of 0.2% using batch conditions, but in a flow system,
an initially reasonable activity towards the cross-metathesis
products drops rapidly, whilst the self-metathesis of methyl
oleate continues. Studies using a variety of different conditions
and substrates suggest that the problem is a low residence time
within the reactor, coupled with some catalyst instability. An
improved flow performance was achieved by lowering the flow
rate of the substrates.
We have analysed reaction products using ICP-OES analysis
and observed only very low levels of residual Ru: 10–15 ppm
in the self-metathesis of methyl oleate, 5–8 ppm in the cross-
metathesis of methyl oleate with dimethyl maleate and only
0.5–0.8 ppm for 2-octene. These values for Ru loss are in the
same range as those observed when catalyst 3 was used in batch
mode.9a,b 19F NMR analysis of the recovered products from
-
reactions involving CO2 showed only traces of fluorine (NTf2 ),
1
but the ionic liquid is not detectible by H NMR spectroscopy
(<50 mol ppm)
The apparent correlation of substrate polarity (all reactions
were carried out at 50 ◦C and 100 bar pressure) with the
loss of ruthenium and IL suggests that the catalyst is slightly
soluble in the flowing medium. Confirmation that this is likely
to be a problem, but one that is greatly reduced by the use of
CO2, came from a cross-metathesis reaction of methyl oleate
with dimethyl maleate carried out by replacing CO2 with N2
at 30 bar in a classical liquid flow reaction (ESI, Fig. S11†).
The solutions emanating from the reactor were highly coloured
(brown, Fig. 8) and the reaction stopped after 3 h, suggesting
substantial leaching of the catalyst and ionic liquid.
Acknowledgements
We thank the EC for financial support under the NANOHOST
programme no. PITN-GA-2008-215193 (R. D.) under the FP7
grant EUMET (CP-FP 211468-2) (to S. P. N. and M. M.) and
¨
the DAAD for a scholarship (E. O). S. P. N. is a Royal Society
Wolfson Research Award holder.
Notes and references
Conclusions
1 For reviews on applications in metathesis, see: (a) S. T. Diver and
A. J. Giessert, Chem. Rev., 2004, 104, 1317; (b) I. Nakamura and Y.
Yamamoto, Chem. Rev., 2004, 104, 2127; (c) A. Dieters and S. F.
Martin, Chem. Rev., 2004, 104, 2199; (d) M. D. McReynolds, J. M.
Dougherty and P. R. Hanson, Chem. Rev., 2004, 104, 2239; (e) K. C.
Nicolau, P. G. Bulger and D. Sarlah, Angew. Chem., 2005, 117, 4564,
(Angew. Chem., Int. Ed., 2005, 44, 4490); (f) T. J. Donohoe, A. J. Orr
and M. Bingham, Angew. Chem., 2006, 118, 2730, (Angew. Chem.,
In conclusion, we have described the first catalytic system that
allows continuous flow olefin metathesis using a homogeneous
catalyst. It involves a boomerang catalyst with an imidazolium
tag on the reactive carbene. As long as the double bonds in the
alkenes are not terminal, high reactivity is observed over at least
1194 | Green Chem., 2011, 13, 1187–1195
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Int. Ed., 2006, 45, 2664); (g) W. A. L. van Otterlo and C. B. de Koning,
Chem. Rev., 2009, 109, 3743.
2 (a) S. T. Diver, Coord. Chem. Rev., 2007, 251, 671; (b) M. Mori, Adv.
Synth. Catal., 2007, 349, 121.
3 C. W. Bielawski and R. H. Grubbs, Prog. Polym. Sci., 2007, 32, 1.
4 W. Keim, M. Berger, A. Eisenbeis, J. Kadelka and I. Schlupp, J. Mol.
Catal., 1981, 13, 95.
5 For reviews on Ru-based metathesis catalysts, see: (a) T. M. Trnka
and R. H. Grubbs, Acc. Chem. Res., 2001, 34, 18; (b) R. H. Grubbs,
Handbook of Metathesis, Wiley-VCH, Weinheim, Germany, 2003;
(c) S. J. Connon and S. Blechert, Angew. Chem., 2003, 115, 1944,
(Angew. Chem., Int. Ed., 2003, 42, 1900); (d) C. Samojłowicz, M.
Bieniek and K. Grela, Chem. Rev., 2009, 109, 3708.
6 For reports on the pharmaceutical application of RCM in multi-
kilogram scale, see: (a) T. Nicola, M. Brenner, K. Donsbach and P.
Kreye, Org. Process Res. Dev., 2005, 9, 513; (b) N. K. Yee, V. Farina,
I. N. Houpis, N. Haddad, R. P. Frutos, F. Gallou, X.-J. Wang, X.
Wei, R. D. Simpson, X. Feng, V. Fuchs, Y. Xu, J. Tan, L. Zhang, J.
Xu, L. L. Smith-Keenan, J. Vitous, M. D. Ridges, E. M. Spinelli, P.
Johnson, K. Donsbach, T. Nicola, M. Brenner, E. Winter, P. Kreye
and W. Samstag, J. Org. Chem., 2006, 71, 7133; (c) Y. S. Tsantrizos,
J.-M. Ferland, A. McClory, M. Poirier, V. Farina, N. K. Yee, X.-J.
Wang and L. Hadda, J. Organomet. Chem., 2006, 691, 5163.
7 H. Clavier, K. Grela, A. Kirschning, M. Mauduit and S. P. Nolan,
Angew. Chem., 2007, 119, 6906, (Angew. Chem., Int. Ed., 2007, 46,
6786).
8 P. Sledz, M. Mauduit and K. Grela, Chem. Soc. Rev., 2008, 37, 2433.
9 (a) H. Clavier, N. Audic, M. Mauduit and J.-C. Guillemin, Chem.
Commun., 2004, 2282; (b) H. Clavier, N. Audic, M. Mauduit and J.-
C. Guillemin, J. Organomet. Chem., 2005, 690, 3585; (c) H. Clavier,
S. P. Nolan and M. Mauduit, Organometallics, 2008, 27, 2287; (d) N.
Audic, H. Clavier, M. Mauduit and J.-C. Guillemin, J. Am. Chem.
Soc., 2003, 125, 9248.
R. W. Fischer, M. Haumann and P. Wasserscheid, ChemCatChem,
2010, 2, 1399.
14 (a) M. F. Sellin, P. B. Webb and D. J. Cole-Hamilton, Chem.
Commun., 2001, 781; (b) B. Webb, T. E. Kunene and D. J. Cole-
Hamilton, Green Chem., 2005, 7, 373; (c) P. B. Webb, M. F. Sellin,
T. E. Kunene, S. Williamson, A. M. Z. Slawin and D. J. Cole-
Hamilton, J. Am. Chem. Soc., 2003, 125, 15577; (d) D. J. Cole-
Hamilton, T. E. Kunene, P. B. Webb, in Multiphase Homogeneous
Catalysis, ed. B. Cornils, Wiley VCH, Weinheim, 2005, pp. 688–
716.
15 (a) M. R. Buchmeiser, Chem. Rev., 2009, 109, 303; (b) M. Mayr, B.
Mayr and M. R. Buchmeiser, Angew. Chem., Int. Ed., 2001, 40, 3839;
(c) G. Jas and A. Kirschning, Chem.–Eur. J., 2003, 9, 5708; (d) A.
Kirschning, W. Solodenko and K. Mennecke, Chem.–Eur. J., 2006,
12, 5972; (e) A. Michrowska, K. Mennecke, U. Kunz, A. Kirschning
and K. Grela, J. Am. Chem. Soc., 2006, 128, 13261; (f) J. O. Krause,
S. H. Lubbad, O. Nuyken and M. R. Buchmeiser, Macromol. Rapid
Commun., 2003, 24, 875; (g) D. Schoeps, K. Buhr, M. Dijkstra, K.
Ebert and H. Plenio, Chem.–Eur. J., 2009, 15, 2960; (h) B. Van Berlo,
K. Houthoofd, B. F. Sels and P. A. Jacobs, Adv. Synth. Catal., 2008,
350, 1949.
16 (a) S. H. Hong, A. G. Wenzel, T. T. Salguero, M. W. Day and R. H.
Grubbs, J. Am. Chem. Soc., 2007, 129, 7961; (b) J. C. Conrad and
D. E. Fogg, Curr. Org. Chem., 2006, 10, 185.
17 (a) J. C. Mol, Top. Catal., 2004, 27, 97; (b) A. Behr, A. Westfechtel
and J. P. Gomes, Chem. Eng. Technol., 2008, 31, 700; (c) A. Rybak,
P. A. Fokou and M. A. R. Meier, Eur. J. Lipid Sci. Technol., 2008,
110, 797.
18 C. J. Rodriguez, D. F. Foster, G. R. Eastham and D. J. Cole-Hamilton,
Chem. Commun., 2004, 1720.
19 J. Patel, S. Mujcinovic, W. R. Jackson, A. J. Robinson, A. K. Serelis
and C. Such, Green Chem., 2006, 8, 450.
20 C. Jime´nez-Rodriguez, G. R. Eastham and D. J. Cole-Hamilton,
Inorg. Chem. Commun., 2005, 8, 878.
21 (a) J. Patel, J. Elaridi, W. R. Jackson, A. J. Robinson, A. K. Serelis and
C. Such, Chem. Commun., 2005, 5546; (b) A. J. Robinson, J. Elaridi,
J. Patel and W. R. Jackson, Chem. Commun., 2005, 5544; (c) Y. L.
Zhu, J. Patel, S. Mujcinovic, W. R. Jackson and A. J. Robinson, Green
Chem., 2006, 8, 746.
10 (a) U. Hintermair, G. Y. Zhao, C. C. Santini, M. J. Muldoon and D. J.
Cole-Hamilton, Chem. Commun., 2007, 1462; (b) U. Hintermair, Z.
Gong, A. Serbanovic, M. J. Muldoon, C. C. Santini and D. J. Cole-
Hamilton, Dalton Trans., 2010, 39, 8501.
11 U. Hintermair, T. Hofener, T. Pullmann, G. Francio and W. Leitner,
ChemCatChem, 2010, 2, 150.
12 (a) C. P. Mehnert, Chem.–Eur. J., 2005, 11, 50; (b) C. P. Mehnert,
R. A. Cook, N. C. Dispenziere and M. Afeworki, J. Am. Chem. Soc.,
2002, 124, 12932; (c) C. P. Mehnert, E. J. Mozeleski and R. A. Cook,
Chem. Commun., 2002, 3010.
13 (a) M. Haumann, K. Dentler, J. Joni, A. Riisager and P. Wasserscheid,
Adv. Synth. Catal., 2007, 349, 425; (b) S. Werner, N. Szesni, M. Kaiser,
22 J. S. Kingsbury, J. P. A. Harrity, P. J. Bonitatebus and A. H. Hoveyda,
J. Am. Chem. Soc., 1999, 121, 791.
23 T. Vorfalt, K. J. Wannowius, V. Thiel and H. Plenio, Chem.–Eur. J.,
2010, 16, 12312.
24 J. Y. Song, M. Q. Hou, G. Liu, J. L. Zhang, B. X. Han and G. Y.
Yang, J. Phys. Chem. B, 2009, 113, 2810.
This journal is
The Royal Society of Chemistry 2011
Green Chem., 2011, 13, 1187–1195 | 1195
©