Catalytic activity and selectivity of a range of ruthenium complexes tested in the styrene/EDA reaction system

Article abstract of DOI:10.1016/j.molcata.2014.02.009

The complex ensemble of competing chemical processes (cyclopropanation, metathesis, dimerisation) involved in the reaction of ethyl diazoacetate with styrene is examined in the presence of a panel of ten ruthenium complexes. Our results, focusing on the catalysts' activity and selectivity, showcased the new NHC-containing complex 10 and the Fischer carbene 7 as leading to best chemoselectivities for cyclopropanation while the bidentate Schiff-base complexes 3 and 4 provided highest stereoselectivity. The traditionally metathesis-active Grubbs I catalyst (5) could be manipulated, by working under high dilution, to display moderate activity in cyclopropanation whereas the Grubbs II catalyst (6) totally promoted metathesis. Data obtained with the above set of Ru complexes strongly support the premise that ligand structure and configuration in the Ru coordination sphere are essential factors in controlling the reaction pathways.

Full text of DOI:10.1016/j.molcata.2014.02.009

Journal of Molecular Catalysis A: Chemical 386 (2014) 86–94
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Catalytic activity and selectivity of a range of ruthenium complexes
tested in the styrene/EDA reaction system
Fu Dinga, Ya-guang Suna, Francis Verpoortb,c,∗∗, Valerian Dragutand,∗∗
,
Ileana Dragutand,∗
a Laboratory of Coordination Chemistry, Shenyang University of Chemical Technology, Shenyang 110142, PR China
b State Key Laboratory of Advanced Technology for Materials Synthesis and Processing, Center for Chemical and Material Engineering, Department of
Chemistry, Faculty of Sciences, Wuhan University of Technology, Wuhan 430070, PR China
c Department of Inorganic and Physical Chemistry, Ghent University, Krijgslaan 281 (S3), 9000 Ghent, Belgium
d Institute of Organic Chemistry of the Romanian Academy, 060023 Bucharest, Romania
a r t i c l e i n f o
a b s t r a c t
Article history:
The complex ensemble of competing chemical processes (cyclopropanation, metathesis, dimerisation)
involved in the reaction of ethyl diazoacetate with styrene is examined in the presence of a panel of ten
ruthenium complexes. Our results, focusing on the catalysts’ activity and selectivity, showcased the new
NHC-containing complex 10 and the Fischer carbene 7 as leading to best chemoselectivities for cyclo-
propanation while the bidentate Schiff-base complexes 3 and 4 provided highest stereoselectivity. The
traditionally metathesis-active Grubbs I catalyst (5) could be manipulated, by working under high dilu-
tion, to display moderate activity in cyclopropanation whereas the Grubbs II catalyst (6) totally promoted
metathesis. Data obtained with the above set of Ru complexes strongly support the premise that ligand
structure and configuration in the Ru coordination sphere are essential factors in controlling the reaction
pathways.
Received 18 October 2013
Received in revised form 1 February 2014
Accepted 8 February 2014
Available online 17 February 2014
Keywords:
Cyclopropanation
N-heterocyclic carbenes
Schiff bases
Ruthenium complexes
C
C coupling
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
excellent reviews on different aspects of cyclopropanation reac-
tions have been published [4].
Many natural or unnatural compounds incorporate cyclo-
propane as a basic structural motif that often confers them
biological activity, frequently also therapeutic properties [1]. In
addition, the propensity of cyclopropane to induce conformation
constraints promotes the use of this highly strained carbocy-
cle as a synthetic building block in more complex structures
demanding a particular stereoconfiguration. Starting from these
premises, an enabling construction of the useful cyclopropane
entity by metal-catalyzed [2] or organocatalyzed [3] intermolec-
ular cyclopropanation has always been an attractive research area
for synthetic organic chemists. To date, a substantial number of
Of the plethora of methods published for stereoselective
cyclopropanation starting from alkenes [4b], the transition-metal
mediated transfer of carbene (in situ delivered through decompo-
sition of aliphatic diazo compounds) is the most developed and
used protocol [4e,h]. Several traditional transition metal catalysts
for cyclopropanation are presently enjoying a revived interest,
e.g. those based on copper [5a], gold [5a,b], rhodium [5c,d] pal-
ladium [5e,f], cobalt [5g,h] and ruthenium [2a,b, 4a–d, 6a–h].
Owing to the various oxidation states that ruthenium can assume,
with easy redox transitions between them during catalytic cycles,
as well as to the flexible coordination abilities of ruthenium,
large libraries of mononuclear and dinuclear ruthenium com-
plexes have been created for various applications. In contrast
to rhodium- and copper-based catalysts leading to either cis- or
trans-substituted cyclopropanes, depending on the structure of
the catalyst employed, ruthenium catalysts provide mostly trans-
cyclopropanated products [4]. Currently cyclopropanation employs
an array of ruthenium catalysts based on arene and/or NHC [3a–c,
7] or more complicated ligands such as pybox, salen, porphyrin
[4b,c,h,i]. Previous research unequivocally evidenced that some
catalysts with established activity for alkene metathesis could also
Abbreviations:
diazoacetate; Grubbs
ATRP, atom transfer radical polymerization; EDA, ethyl
catalyst, Grubbs first generation catalyst; Grubbs II
I
catalyst, Grubbs second generation catalyst; NHC, N-heterocyclic carbene;
Pybox, pyridinebis(oxazoline); rt, room temperature; Ru, ruthenium; Salen, 2,2’-
ethylenebis(nitrilomethylidene)diphenol,N,N’-ethylenebis(salicylimine).
∗
Corresponding author. Tel.: +40 21 3167900; fax: +40 21 3121601.
Corresponding authors.
E-mail addresses: francis.verpoort@ugent.be (F. Verpoort),
∗∗
vdragutan@yahoo.com (V. Dragutan), idragutan@yahoo.com (I. Dragutan).
http://dx.doi.org/10.1016/j.molcata.2014.02.009
1381-1169/© 2014 Elsevier B.V. All rights reserved.

F. Ding et al. / Journal of Molecular Catalysis A: Chemical 386 (2014) 86–94
87
be valorized in alkene cyclopropanation [8]. However, rigorous con-
trol of the reaction conditions can favour either reaction pathway.
As part of our ongoing programme to extend the utility of
ruthenium catalytic systems [9], we have previously reported on
cyclopropanation catalyzed by bidentate N,O-Schiff base ligated
precursors [10]. This class of catalysts also proved good activity
in metathesis, ATRP, enol-ester synthesis, vinylation [11]. In olefin
cyclopropanation with diazo compounds, Ru(II)–arene catalysts
bearing Schiff base ligands were found to be moderately active
producing, as expected, higher proportions of cyclopropanated
products with activated olefins than with non-activated counter-
parts [10]. On optimizing reaction conditions, cyclopropanation
could overcome side reactions (dimerisation and metathesis). It
was concluded that the electronic properties of the Schiff base
ligand influence the decomposition rate of the diazo compound,
hence generation of the carbene. With more donating Schiff bases
higher yields (max. 76%) and trans/cis ratios (dr: 2.8/1) in cyclo-
propanated products were recorded. For practical purposes such
yields are, however, not satisfactory.
We were prompted therefore to investigate a whole range of
structurally diverse Ru precursors and determine their perfor-
mance, for the first time under uniform conditions and in the
same reaction, of styrene with ethyl diazoacetate (EDA), generally
accepted as standard. The primary goal of this work is to extend
the scope of several commercial and known or new Ru catalysts
to prevailingly cyclopropanation. A secondary focus has been on
rationalizing the effect that ligands within the Ru coordination
sphere play on the selectivity in cyclopropanation, relative to that
in concurrent metathesis or dimerisation. These new findings could
provide fresh insights into this intricate transformation.
Fig. 1. EDA decomposition induced by Ru catalysts followed by FTIR spectroscopy
in the presence of catalyst 7 (at 40 ◦C).
By modulating the reaction conditions a certain control on unde-
sired reactions can be exerted. To enhance cyclopropanation we
decided to minimize dimerisation by slowly adding a diluted solu-
tion of the diazoacetate to the mixture of styrene and catalyst and
also by using a large excess of styrene. Working under high sub-
strate:catalyst loadings we expected a subdued metathesis, thus
enhancing cyclopropanation.
Progress of the reaction of styrene with EDA as proceeding under
promotion of our Ru catalysts (Scheme 2) was examined using FTIR
and GC methodologies.
2. Results and discussion
Due to release of nitrogen gas, EDA decomposition in the pres-
ence of Ru catalysts is an irreversible process. To determine the
influence of EDA decomposition on the overall catalytic process,
advantage was taken of the characteristic stretching vibration of the
Intermolecular transition metal-catalyzed carbenoid cyclo-
propanation of alkenes with diazo compounds has been extensively
studied, in particular the reaction between styrene and ethyl dia-
zoacetate (EDA) [12]. Ethyl diazoacetate generates in situ a carbene
species which undergoes chemoselective insertion yielding cis and
trans cyclopropanes (Scheme 1, Eq. (1)). However, as mentioned
above, the overall process also commonly involves different side-
reactions resulting in products observed for all types of transition
metal catalysts (Scheme 1): metathesis (Eq. (2)), cross-metathesis
styrene-diazoester (Eq. (3)), homocoupling (dimerisation) of EDA to
a mixture of diethyl maleate and fumarate, usually strongly favou-
ring the former (Eq. (4)).
N
N double bond (2110 cm−1) from EDA whose decrease in time
at the reaction temperature was followed by FTIR spectroscopy, in
the presence of each catalyst; a typical example is shown in Fig. 1,
for the catalyst 7 (at 40 ◦C). Since above 80 ◦C a very rapid thermal
decomposition of EDA in toluene could be observed, our catalytic
tests were run at lower temperatures (30–60 ◦C).
In order to evaluate the effect of the ligands in the four- and
five-coordinate complexes we have synthesized and applied in
cyclopropanation quite structurally different 16- or 18-electron
ruthenium complexes selected so as to bear diverse ligands
COOEt
Ph
[Ru]
Ph
COOEt
N2
N2
Ph
COOEt
cis
trans
(Eq. 1)
(Eq. 2)
(Eq. 3)
(Eq. 4)
Ph
[Ru]
Ph
Ph
C2H4
2
Ph
Ph
COOEt
COOEt
EtOOC
[Ru]
COOEt
Ph
N2
N2
Ph
Ph
COOEt
EtOOC
COOEt
[Ru]
N
2
2
N2
2
COOEt
Scheme 1. Concurrent reactions in cyclopropanation of styrene with diazoesters (Eqs. (1)–(4)).

88
F. Ding et al. / Journal of Molecular Catalysis A: Chemical 386 (2014) 86–94
Scheme 2. Ruthenium catalysts employed in reaction of styrene with EDA.
attached to the metal core (Scheme 2) including the, to our knowl-
edge, not yet reported Ru complexes (4, 10), several known (3,
7–9) and some commercial (1, 2, 5, 6) catalysts. Synthesis of the
non-commercial complexes was carried out using established pro-
cedures as indicated in Scheme 3 (see also Section 3).
Previous investigations in our group have already shown that
catalysts 2 and 3 are adequate promoters for styrene cyclopropa-
nation with ethyl diazoacetate as the carbenoid source [10]. It is
worth mentioning that in the present work catalysts 4, 7, 8 and 10
are being used for the first time in cyclopropanation via the carbene
Scheme 3. Syntheses of new and known catalysts used in cyclopropanation: (a) Ref. [10]; (b) Ref. [2b,c]; (c) This work; (d) Ref. [13].

F. Ding et al. / Journal of Molecular Catalysis A: Chemical 386 (2014) 86–94
89
Table 1
Product distribution in the reaction of styrene with EDA (in toluene, at 60 ◦C, 4 h).a,b
Reaction selectivity
Yieldc in cyclopropa-nated
products (%)
Entry
Catalyst
Cyclopropanation (%)
trans/cis
Metathesis (%)
Dimerisation (%)
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
77
58
73
72
35
–
84
50
44
85
1.81
1.49
2.3
2.7
1.02
–
0.98
1.38
1.21
1.47
12
13
8
21
59
100d
8
33
13
5
11
29
19
7
6
–
87.5
66.6
79.3
91.1
85.4
–
91.3
74.6
50.6
89.5
8
17
43
10
10
10
11
12
5e
7e
82
75
2.0
–
9.5
5
8.5
20
90.6
78.9
a
Based on ethyl diazoacetate. Determined by GC by comparison with authentic samples (internal standard diethyl adipate). Ratio of the dimerisation products: diethyl-
maleate/diethylfumarate ∼2/1.
b
Molar ratios: catalyst/EDA/styrene = 1/125/2500; toluene/styrene = 10.
Based on EDA. EDA was totally consumed.
c
d
Determined by GC by comparison with authentic samples.
e
Molar ratios: catalyst/EDA/styrene = 1/400/8000. Reaction temperature: 50 ◦C. Reaction time: 4 h.
transfer approach. Results are compiled in Table 1 and graphically
highlighted in Fig. 2.
the bidentate Schiff-base Ru complexes 3 and 4 all other catalysts
investigated here directed more towards the trans diastereomer,
with the exception of 7 (Table 1, Entry 7). These data on diastereos-
electivity comply with literature reports on a variety of ruthenium
catalysts that also exert trans control. The same is true for the
trans/cis stereoselectivities of the metathesis products found to be
roughly 2/1. The divergent behaviour of catalysts 1–10 in the reac-
tion of styrene with diazoester is more suggestively illustrated in
Fig. 2.
If we consider just the two reactions where EDA is consumed
(cyclopropanation and dimerisation) the yields vs. EDA (Table 1,
last column, entries 1–10) are, as should be expected, higher than
the selectivity in the cyclopropane-containing product. These cal-
culated yields are a better indication of the efficiency of the different
catalysts in cyclopropanation since the highest values are attained
for runs where the catalyst is disfavouring dimerisation, even
though the selectivity is not as high as is the case for 1, 4 and 5.
Reactions carried out with the Grubbs I catalyst (5) using
styrene as the solvent (Table 1, entry 11) gave high selectivity in
cyclopropanation (82%), at 50 ◦C and at a lesser catalyst loading
(catalyst/EDA/styrene = 1/400/8000). In further experiments con-
ducted under the latter conditions the influence of temperature on
cyclopropanation and its competing processes was also followed.
Within the temperature range 30–50 ◦C we found that selectivity
in the desired cyclopropanated product is increasing from 62.5%
to 82.0% with dimerisation and styrene metathesis diminishing
accordingly (Table 2).
Of the catalysts examined for reactions performed under
high dilution in toluene, at 60 ◦C and molar ratios cata-
lyst/EDA/styrene = 1/125/2500, the best chemoselectivity for asym-
metric intermolecular cyclopropanation yielding 1-carbethoxy-
2-phenylcyclopropane has been observed for the IPr-containing
promoter 10 (85%) and the ethoxy Ru-carbene 7 (84%) (Table 1).
Besides, cyclopropanation is still the main reaction pathway in the
case of catalysts 1–4 and 8. Under these conditions the Grubbs
I catalyst (5) displays just a moderate selectivity for cyclopropa-
nation with metathesis prevailing. In contrast to its congener, the
Grubbs II catalyst (6) gives neither cyclopropanation nor dimerisa-
tion, while metathesis reaction is entirely favoured. Regarding now
the competing processes, dimerisation and metathesis, results evi-
dence them as balanced in the case of catalysts 1 and 7, whereas
nearly a twice as much dimerisation is recorded with catalysts 2, 3
and 10. An exceptionally sharp increase of dimerisation is found for
catalyst 9 (43%), quite close to the extent to which cyclopropanation
occurs (44%). Though pretty good promoters for cyclopropanation,
the ethoxy Ru-carbene 8 and our new Schiff-base bidentate-Ru
catalyst 4 favour metathesis over dimerisation. Remarkably, while
best diastereoselectivities in cyclopropanation were attained with
As seen below, analysis of data acquired in this research helps
illuminate some of the underpinnings of this complex chemical
transformation and provides fresh understanding of the nature of
the active catalytic species and the mechanistic pathways. Our tests
with a larger array of ruthenium catalysts enable a deeper con-
sideration of the concerted influence of the catalyst and reaction
conditions (molar ratios, concentration, solvent, temperature) on
yields and selectivity in the targeted cyclopropanated product.
A
good accord could be established with available liter-
ature information on previously reported cyclopropanation
catalysts (1–3). For example, catalyst employed in cyclo-
1
propanation of styrene with EDA (1/EDA/styrene = 1/200/5740;
addition of EDA over 4 h) was reported [14] to give 93% yield
of 1-carbethoxy-2-phenylcyclopropane at 60 ◦C, vs. our result
of 87.5%, as expected slightly lower because of working at
the same temperature but under high dilution in toluene
Fig. 2. Conversion (%; based on EDA) of styrene into the cyclopropanation product
(1-carbethoxy-2-phenyl cyclopropane), in the reaction with EDA (60 ◦C, 4 h). Condi-
tions for experiments with catalysts 5 and 7 (last two columns; in red): molar ratios
catalyst/EDA/styrene = 1/400/8000; 50 ◦C, 4 h. (For interpretation of the references
to color in this text, the reader is referred to the web version of the article.)

90
F. Ding et al. / Journal of Molecular Catalysis A: Chemical 386 (2014) 86–94
Table 2
Influence of temperature on cyclopropanation, dimerisation and metathesis in the reaction of styrene with EDA (in excess styrene).a
Entry
Temp. (◦C)
Cyclopropanation (%)
trans/cis
Dimerisation (%)
Metathesis (%)
Yield (%)
1
2
3
4
5
30
35
40
45
50
62.5
71.0
73.0
77.0
82.0
2.2
1.9
1.8
1.9
2.0
23.5
18.0
18.0
14.5
8.5
14.0
11.0
9.0
8.5
9.5
72.7
79.8
80.2
84.2
90.6
a
Molar ratios: catalyst/EDA/styrene = 1/400/8000.
(catalyst/EDA/styrene = 1/125/2500); toluene/styrene = 10; Table 1,
entry 1). This comparison is supported by EDA dosage during the
same time interval (4 h) which is a crucial factor for manipulating
competition by dimerisation. It was found, indeed, that the same
catalyst can provide 70% dimerisation, at a similar catalyst loading
(1/EDA = 1/100; room temperature, in THF) but by adding together
the catalyst and EDA at the onset of the experiment [15].
It should be mentioned that results from runs with the Ru dimer
2 and the Schiff-base Ru arene complex 3, affording only moder-
ate yields and selectivities for cyclopropanation (Table 1, entries
2 and 3) parallel those earlier communicated by our group [10].
This resemblance suggests a similar catalyst activation pathway
i.e. decoordination of the same arene ligand (p-cymene). The bet-
ter selectivity for 3 vs. 2 is to be assigned to the stabilizing effect
of the Schiff-base ligand on the Ru active species resulting from
3.
metathesis, while high temperatures act in the opposite direc-
tion (Table 2). Under the indicated reaction conditions, within the
temperature range 30–50 ◦C, it was found that selectivity in 1-
carbethoxy-2-phenylcyclopropane increases from 62.5% to 82.0%
while dimerisation and metathesis diminish; of the latter two reac-
tions, dimerisation is more sensitive to temperature changes. In
addition, variation of the reaction temperature in this range has no
significant effect on the trans/cis ratio of the cyclopropanated prod-
ucts that oscillates around the value of 2 (Table 2), in agreement
with experiments of Snapper who found that the stereoselectivity
of the cyclopropanation with the Grubbs I catalyst is moderate at
best (E/Z = 1/1–3/1) [6b].
A decrease of cyclopropanation at lower temperatures, associ-
ated with a practically constant stereoselectivity, as found for the
catalyst 5, has also been previously communicated for the catalyst
1 (43% yield at 30 ◦C vs. 93% at 60 ◦C) [14].
The new catalyst 4 bearing an O,N-bidentate Schiff-base and
two PPh3 ligands (Table 1, entry 4) proved to be a remarkable
cyclopropanation promoter (yield 91.1%). Under the applied reac-
tion conditions it disfavoured dimerisation, though an increase in
metathesis could not be prevented in this case (as compared to 1).
This is probably due to a longer life-time of the active species aris-
ing from 4, stabilized by the Schiff-base, relative to that from 1, in
spite that both involve the same phosphine (PPh3) decoordination.
Noteworthy, the Grubbs I and II catalysts (5 and 6) appear
as particular cases in cyclopropanation. Though these metathesis
catalysts have been largely employed in tandem metathe-
sis/cyclopropanation [4c,16], it is well-documented that in the
presence of diazoesters catalyst 6 promotes overwhelmingly
metathesis with only trace amounts of cyclopropanation prod-
ucts being formed [6b]. On the other hand, the Grubbs catalysts
proved also entirely suitable for homocoupling of diazoesters to
give maleate and fumarate esters [17a]. Nevertheless, with a1:1
molar ratio EDA/styrene and 0.01% mol 6, especially under more
elevated temperature (50 ◦C), styrene metathesis to stilbene par-
tially occurred [17]. In case of catalyst 6 all seem to depend,
however, on molar ratios and reaction conditions as reported in
a recent investigation where by using 6/EDA/styrene = 1/200/5200
in chlorobenzene (C6H5Cl/styrene = 0.4), at 60 ◦C for 24 h, the
cyclopropanated product (1-carbethoxy-2-phenylcyclopropane)
formed predominantly (80% yield, based on EDA) and just below
1% metathesis (calculated vs. styrene) was observed [2c].
Aiming at a better understanding of the role played by EDA in
the cyclopropanation reaction, we also performed a detailed inves-
tigation on the EDA decomposition induced by the catalysts 5 and
7 (Fig. 3).
Thus, we could easily observe that even in the presence of
the ruthenium catalyst alone, EDA undergoes rapid decomposi-
tion with a rate depending on the catalyst kind. Quite remarkably,
catalyst 7 leads to a higher decomposition rate than 5 (Fig. 3). In
the absence of the olefin, the decomposition outcome is dimeri-
sation to EtOOC-CH CH-COOEt, generally accepted as proceeding
through the intermediacy of the unstable metalcarbene species
[Ru] = CH-COOEt. However, when an olefinic substrate is added,
the decomposition rate follows a different course being acceler-
ated when the olefin is styrene and decelerated in the presence of
cis-stilbene.
It is intriguing that information regarding applications of Ru
complexes 5 and 6 as promoters of styrene cyclopropanation
with diazo compounds are quite scarce. Therefore, we undertook
a detailed study of this process under these conditions. Note-
worthy, when the reaction was carried out with the Grubbs I
catalyst (5) using styrene as the solvent (Table 1, entry 11), a
high selectivity in cyclopropanation (82%) could be achieved, even
at 50 ◦C and at a lesser catalyst loading than in the run at 60 ◦C
(Table 1, entry 5). Furthermore, while keeping the same solvent,
concentration and catalyst loading, the effect of reaction temper-
ature on cyclopropanation in the presence of Grubbs I catalyst is
totally different vs. that on dimerisation and metathesis processes:
low temperatures restrain cyclopropanation vs. dimerisation and
Fig. 3. Decomposition of EDA vs. time in the presence of catalyst 5 (♦) and 7 (ꢀ), in
the absence of an olefin substrate, at 40 ◦C; 5 or 7/EDA = 1/400.

F. Ding et al. / Journal of Molecular Catalysis A: Chemical 386 (2014) 86–94
91
Scheme 4. Carbenoid and coordination mechansims for cyclopropanation, dimerisation and metathesis.
Data illustrated in Fig. 3 support results recorded with catalysts
5 and 7 (Table 1, entries 11 and 12): the more rapid decomposi-
tion of EDA by the latter catalyst explains the larger proportion
of dimerisation, at the detriment of cyclopropanation. Both cat-
alysts lead to a feable metathesis yield, as anticipated lower in
the case of 7 reputed as a quite poor metathesis catalyst. The
EDA decomposition in the presence of stilbene, found to be slower
than in the absence of any olefinic substrate, seems to indicate
that stilbene does not compete with styrene for reacting with the
[Ru] = CH-COOEt species. The possible rationalization could be the
unfavourable steric factors, also accounting for the very reduced
percentage (if any) of the putative product of stilbene cyclopropa-
nation, 1,2-diphenylcyclopropane.
in cyclopropanation. By just replacing a phosphine ligand in 1
by an NHC ligand, C C forming through cyclopropanation was
enhanced, at the expense of metathesis pathway.
Further evaluation of our results in cyclopropanation of
EDA/styrene induced by catalysts 1–10 led us to suggest a com-
plex mechanistic scheme that involves separate pathways for the
three competing C C coupling reactions, as generally accepted in
research on alkene cyclopropanation [4a,c,i,6d,7,8,18]. To account
for the particular behaviour we observed for our catalysts, it is rea-
sonable to assume the intervention of distinct Ru active species, of
varying stability, for cyclopropanation, dimerisation and metathe-
sis (Scheme 4).
In support of this point of view we decided to look at NMR signals
assignable to carbene species that arise at the onset of EDA decom-
position in the presence of the catalyst 5 monitoring the evolution
of the 1H NMR spectrum of a CDCl3 solution containing the catalyst
and one equivalent of EDA. Already at the start, two different
carbene signals could be detected: the usual benzylidene carbene
deriving from the Grubbs I catalyst and a new signal at 17.128 ppm
(Fig. 4). Also, it was observed that the reaction proceeded much
slower, as compared to the FT-IR kinetic measurements.
At the same time, it should be outlined that 2,6-
isopropylphenyl-imidazolinylidene arene-Ru catalyst
9 [2b,c],
displayed, under similar reaction conditions, only a moderate
cyclopropanation selectivity while also leading to some metathe-
sis and the largest dimerisation production of all catalysts tested in
this research (Table 1, entry 9). This behaviour strongly contrasted
with chemoselectivity found for the same catalyst but working
under a larger excess of styrene and EDA, for a longer duration
and in the presence of a more polar solvent (ClC6H5) [2c] where
cyclopropanation prevailed and metathesis was almost completely
suppressed, possibly due to the much higher concentration and to
a solvent effect [15].
When fresh amounts of EDA were added to the sample, both
signals disappeared in time. What is more, when 50 equiv. of EDA
To our satisfaction, the new catalyst 10, prepared from
Cl2Ru(PPh3)3 and the same N-heterocyclic carbene as for
9 (Scheme 2), provided the highest selectivity (85%) in this
cyclopropanation reaction, along with a reduced proportion of
dimerisation and metathesis products. Accordingly, catalyst 10
falls in the range of promoters which activate through decoordi-
nation of a triphenylphosphine (1, 4, 10), which are all efficient
Fig. 4. 1H NMR of EDA decomposition induced by catalyst 5.

92
F. Ding et al. / Journal of Molecular Catalysis A: Chemical 386 (2014) 86–94
(instead of 1 equiv. as in Fig. 4) were initially introduced no carbene
signals were noticed, at any time. Possibly, several carbene species
are implied: at low EDA concentration the benzylidene carbene is
gradually replaced by a second carbene fragment (ı = 17.128 ppm),
whereas at high EDA concentrations a distinct, very reactive car-
benoid species is formed. Therefore, deviations observed in FTIR
measurements at initial stages of EDA decomposition (Fig. 1) could
be explained by a change in the active species when a large excess
of EDA is applied.
We assume that the catalytic cycle involves first the decompo-
sition of the diazoester, with extrusion of nitrogen gas, to afford the
reactive [Ru] = CHCOOEt intermediate which is playing a role in the
various channels for C C coupling (Scheme 4). Though not detected
by NMR in many investigations on cyclopropanation, intervention
of this ß-carbonyl-carbene entity is widely accepted. This species
is known to be quite unstable, even in solution at low tempera-
ture, and to decompose rapidly as compared with the Ru alkylidene
complexes of unfunctionalised olefins, yet to be a highly active
metathesis promoter [19].
The selectivity in cyclopropanation products found in this
study could be accounted for by two main mechanistic path-
ways [4c,6b]. They imply essentially the carbene transfer from the
[Ru] = CHCOOEt carbene to styrene via either a carbenoid pathway
(A) or/and a coordination pathway (B) (Scheme 4).
As inferred from previous data [4c], in the carbenoid pathway
formation of the cyclopropane moiety involves a late, unsymmetri-
cal transition state A-I with build-up of a positive charge at the most
distant carbon atom of styrene. Taking into account the 16-electron
configuration of our phosphine Ru complexes the carbenoid mech-
anism, triggered immediately after dissociation of one phosphine,
seems to be the favoured pathway. Obviously, for sterical rea-
sons, C C coupling in the transition state A-I occurs predominantly
towards the trans diastereoisomer.
the new catalysts, 4 and 10, suggest generation of this unstable
metalcarbene in both the carbenoid and the coordination mecha-
nisms.
3. Experimental
3.1. Synthesis of catalysts 4 and 10
Catalysts 1, 2, 5, 6 are commercially available and were used
without further purification. The ruthenium complexes 3 [10], 7
[13], 8 [13] and 9 (in situ) [2c] were synthesized according to liter-
ature procedures as shown in Scheme 3.
Synthesis of the new ruthenium promoters 4 and 10 was carried
out in this work. For the synthesis of complex 4 a THF solution of
the thalium salt of the Schiff base ligand [20] was added the equiv-
alent amount of Cl2Ru(PPh3)3 and the mixture stirred overnight
at room temperature. After work-up [11b,20] the solid residue was
dissolved in a minimal amount of toluene, reprecipitated with pen-
tane, filtered off, briefly washed on the funnel with pentane and
dried in vacuo to afford an orange-brown powder (78% yield) which
was stored under inert atmosphere. 1H NMR (300 MHz, CDCl3): ı
2.36 [s, 3H, CH3]; 7.10–7.80 [m, 34H, aryl-CH]; 9.95 ppm (s, 1H,
aldimine ligand).
Catalyst 10 was obtained by adding to1,3-bis(2,6-
diisopropylphenyl)imidazolinylidene (in situ prepared in THF)
[3c] one equivalent of Cl2Ru(PPh3)3. The reaction mixture was
refluxed in THF with stirring for 1 h, then cooled down and the
solid materials filtered off. The filtrate was concentrated in vacuo,
the residue solved in a minimal amount of toluene, reprecipitated
with pentane and then handled as above for complex 4 to give
a dark-brown powder. Isolated yield: 81%. 1H NMR (300 MHz,
CDCl3): ı 1,25 [d, 24H, CH(CH3)2]; 2,93 [m, 4H, CH(CH3)2], 3.65 ppm
(m, 4H, CH2).
On the other hand, in the coordination mechanism (path B),
both the olefin and the CHCOOEt carbene coordinate at the metal.
Therefore, the catalyst must provide two coordination sites, easily
accessible through ligand dissociation within our 18- or 16-electron
Ru complexes which contain either two displaceable phosphines (1,
4, 5, 10) or a p-cymene (2, 3, 9). Thus, highly reactive, coordinatively
unsaturated 14- or 12-electron Ru species may arise. However,
the steric configuration imposed by the Schiff-base and NHC lig-
ands in our Ru complexes might sometimes impede simultaneous
coordination of both the olefin and the carbene moiety favouring
instead the alternative carbenoid pathway (A), hence cyclopropa-
nation. As soon as the complex B-I is formed, it rearranges to the
ruthenacyclobutane complex B-II which, by reductive elimination
of the Ru fragment provides substituted cyclopropanes (trans- and
cis-1-carbethoxy-2-phenylcyclopropane) or, by [2 + 2] cyclorever-
sion gives the metathesis products (trans- and cis-ethylcinnamate).
Because in our experiments the cis- and trans-cinnamic esters were
detected by GC only in trace amounts, pathway B-b2 seems dis-
favoured. Indirect support for formation of distinct carbene species
in the routes directing to cyclopropanation (A, B-b1) comes from
a study with the Grubbs I catalyzed tandem enyne metathesis-
cyclopropanation where the Ru-complex 5 is modified in situ by
the diazoester to form a cyclopropanation active catalyst that no
longer promotes metathesis [6b]. Competing in the overall mecha-
nistic scheme, the carbenoid pathway (C, involving a highly reactive
transition state C-I and the intermediate complex C-II) and the
coordination pathway (D) can respectively explain the forma-
tion of C C coupling products (maleic and fumaric esters) and
metathesis products (cis- and trans-stilbene). The commonality
in this ensemble is the intermediacy of the ß-carbonyl-carbene
species [Ru] = CHCOOEt. Although also our 1H NMR experiments
(Fig. 4) could not detect the real occurrence of the carbene
Ru = CHCOOEt, the high cyclopropanation yields obtained with
4. FTIR measurements
Kinetic measurements on the decomposition of diazoacetate
in the presence of the Ru catalysts 1–10 were performed using a
Brucker FT-Raman/FT-IR spectrometer with a Nernst glowing ele-
ment as the IR source and a DTGS detector. The flow-through IR-cell
consists of KBr windows with a 2 mm inner separation. The fol-
lowing operations were carried out in a typical FT-IR experiment.
From a preliminary prepared solution, with precise concentration,
of the catalyst in freshly distilled toluene, 3 ␮mol of catalyst were
transferred under continuous Ar-flow to an empty 15 ml vessel. The
solvent was evaporated under vacuum giving a small amount of
solid catalyst. Then 12 ml of extra dry toluene was added to adjust
concentration at a predetermined value. For experiments also using
an olefin substrate, 4 mmol of olefin were usually added. The ves-
sel was sealed with a septum and stored at −18 ◦C. Just before the
experiment the vessel was introduced into a heating block to reach
the desired temperature and 400 ␮mol EDA were added all at once
by a syringe. A needle was inserted through the septum to prevent
pressure building. The reaction mixture was circulated through the
IR cell in a closed circuit by use of a peristaltic pump. All measure-
ments were performed by following the (N N) stretch vibration
at 2110 cm−1. Each experiment was repeated at least 5 times. All
spectra were normalized to the solvent, integrated and correlated
to the concentration by means of a calibration curve.
5. GC measurements
To determine the yield and chemoselectivity in cyclopropana-
tion, capillary GC measurements were performed using a Varian
CDS 401 with a Supelco 5DB-24030 capillary column and a FID
detection system.

F. Ding et al. / Journal of Molecular Catalysis A: Chemical 386 (2014) 86–94
93
In a typical GC experiment the following operations were under-
taken: from a preliminary prepared solution of catalyst in freshly
distilled toluene, with known concentration, 2.5 ␮mol of catalyst
were transferred under Ar-flow to an empty 15 ml vessel. The sol-
vent was evaporated in vacuo affording a small amount of solid
catalyst. Styrene (10 mmol) was next added. EDA (1 mmol) was dis-
solved in styrene (10 mmol), cooled to 0 ◦C and the solution added
very slowly, at 0 ◦C (over 4 h, using a peristaltic pump) to the above
reaction mixture. After addition of a few drops of the EDA solution,
the mixture was heated to the reaction temperature and kept at this
temperature overnight. Before GC-analysis, the reaction mixture
was passed through a celite filter in order to remove the catalyst.
Celite was washed with 20 ml toluene/EtOAc (1/1). The composi-
tion of the reaction mixture was determined by GC using authentic
samples and diethyl adipate as internal standard.
(b) V. Dragutan, I. Dragutan, L. Delaude, A. Demonceau, Coord. Chem. Rev. 251
(2007) 765–794;
(c) L. Delaude, A. Demonceau, A.F. Noels, Curr. Org. Chem. 10 (2006) 203–215;
(d) A. Tudose, A. Demonceau, L. Delaude, J. Organomet. Chem. 691 (2006)
5356–5365;
(e) L. Beaufort, A. Demonceau, A.F. Noels, Tetrahedron 61 (2005) 9025–9030;
(f) C. Bianchini, H.M. Lee, Organometallics 19 (2000) 1833–1840.
[3] (a) For organocatalyzed cyclopropanation see: V. Terrasson, A. van der Lee, R.
Marcia de Figueiredo, J.M. Campagne, Chem. Eur. J. 16 (2010) 7875–7880, and
references therein;
(b) S.R. Goudreau, A.B. Charette, Angew. Chem. Int. Ed. 49 (2010) 486–488.
[4] For excellent reviews see:
(a) A.F. Noels, A. Demonceau, in: B. Cornils, W.A. Herrmann (Eds.), A Compre-
hensive Handbook in Three Volumes, vol. 2, 2nd ed., Wiley-VCH, Weinheim,
2002, pp. 793–808;
(b) H. Lebel, J.-F. Marcoux, C. Molinaro, A.B. Charette, Chem. Rev. 103 (2003)
977–1050;
(c) G. Maas, Chem. Soc. Rev. 33 (2004) 183–190;
(d) H. Nishiyama, in: C. Bruneau, P.H. Dixneuf (Eds.), Ruthenium Catalysts and
Fine Chemistry, Springer, Berlin/Heidelberg, 2004, pp. 81–92;
(e) A.J. Delmonte, E.D. Dowdy, D.J. Watson, Top. Organomet. Chem. 6 (2004)
97–122;
6. Conclusions
(f) H. Pellissier, Tetrahedron 64 (2008) 7041–7095;
(g) I. Nicolas, P. Le Maux, G. Simonneaux, Coord. Chem. Rev. 252 (2008)
727–735;
(h) A. Caballero, A. Prieto, M. Mar Díaz-Requejo, P.J. Pérez, Eur. J. Inorg. Chem.
(2009) 1137–1144;
The contrasting outcome observed in intermolecular carbenoid
transfer to styrene, catalyzed by different Ru complexes, demon-
strated that chemoselectivity in concomitant cyclopropanation,
metathesis and dimerisation depends on the Ru catalyst, the kind
of its ligands and also on the reaction conditions. The new NHC-Ru
complex 10 and the Fischer carbene 7 were proved as most suitable
for cyclopropanation (84–85% selectivity), under low catalyst load-
ings and a large excess of substrate. The Schiff-base ligated complex
4 gave an excellent yield but unimpressive selectivity for cyclo-
propanation whereas an equal balance between cyclopropanation
and dimerisation was found for catalysts 2 and 9. Therefore, in spite
of the competing side-reactions, the right choice of the Ru catalyst
can direct mainly to the desired process.
(i) C.-Y. Zhou, J.-S. Huang, C.-M. Che, Synlett (2010) 2681–2700.
[5] (a) A. Corma, M. Iglesias, F.X. Labres i Xamena, F. Sanchez, Chem. Eur. J. 16
(2010) 9789–9795;
(b) M.R. Fructos, T.R. Belderrain, P. de Fremont, N.M. Scott, S.P. Nolan, M.M.
Diaz-Requejo, P.J. Perez, Angew. Chem. Int. Ed. 44 (2005) 5284–5288;
(c) E. Gross, J.H.-C. Liu, F.D. Toste, G.A. Somorjai, Nat. Chem. 4 (2012) 947–952;
(d) C.-T. Yeung, W.-S. Lee, C.-S. Tsang, S.-M. Yiu, W.-T. Wong, W.-Y. Wong, H.-L.
Kwong, Polyhedron 29 (2010) 1497–1507;
(e) L. Palacios, X. Miao, A. Di Giuseppe, S. Pascal, C. Cunchillos, R. Castarlenas,
J.J. Perez-Torrente, F.J. Lahoz, P.H. Dixneuf, L.A. Oro, Organometallics 30 (2011)
5208–5213;
(f) D. Kalia, G. Merey, M. Guo, H.O. Sintim, J. Org. Chem. 78 (2013) 6131–6142;
(g) J. Tsuji, Palladium Reagents and Catalysts – New Perspectives for the 21-st
Century, John Wiley & Sons, London, 2004;
An unexpected behaviour was evidenced for the Grubbs I
metathesis catalyst manipulated towards prevailingly cyclopropa-
nation (82%) by using still higher substrate ratios, at a lower
temperature. However, for its second generation counterpart (6)
only metathesis was observed. Data obtained with our Ru promo-
ters could be rationalized assuming a complex mechanistic scheme,
implying carbenoid and/or coordinative pathways and distinct cat-
alytically active species in the competing C C coupling channels
considered.
(h) S. Chen, J. Ma, J. Wang, Tetrahedron Lett. 49 (2008) 6781–6783;
(i) T. Uchida, T. Katsuki, Synthesis (2006) 1715–1723;
(j) M.P. Doyle, Angew. Chem. Int. Ed. 48 (2009) 850–852.
[6] (a) K. Thommes, G. Kiefer, R. Scopelliti, K. Severin, Angew. Chem. Int. Ed. 48
(2009) 8115–8119;
(b) B.G. Kim, M.L. Snapper, J. Am. Chem. Soc. 128 (2006) 52–53;
(c) V. Dragutan, I. Dragutan, J. Organomet. Chem. 691 (2006) 5129–5147;
(d) B.M. Trost, F. Dean Toste, A.B. Pinkerton, Chem. Rev. 101 (2001) 2067–2096;
(e) J. Ito, S. Ujiie, H. Nishiyama, Chem. Eur. J. 16 (2010) 4986–4990;
(f) C.W. Jones, Top. Catal. 53 (2010) 942–952.
[7] (a) A. Demonceau, F. Simal, A.F. Noels, C. Vinas, R. Nunez, F. Teixidor, Tetrahe-
dron Lett. 38 (1997) 4079–4082;
(b) F. Simal, J. Dominique, A. Demonceau, A.F. Noels, Tetrahedron Lett. 40 (1999)
1653–1656;
(c) A.F. Noels, A. Demonceau, D. Jan, Russ. Chem. Bull. 48 (1999) 1206–1211;
(d) O. Tutusaus, S. Delfosse, A. Demonceau, A.F. Noels, R. Nunez, C. Vinas, F.
Teixidor, Tetrahedron Lett. 43 (2002) 983–987.
Acknowledgements
F.D. and Y.S. acknowledge the support of the Shenyang Institute
of Chemical Technology, Shenyang, PR China. F.V. acknowledges the
Chinese Central Government for an “Expert of the State” position
in the program of “Thousand talents” and is indebted to FWO-
Flanders (project grant Number 3G022912), Ministry of Science and
Technology of PR China (Project No 41-24) and the Wuhan Univer-
sity of Technology for financial support during elaboration of this
work. I.D. and V.D. thank FWO-Flanders for financial support. Assis-
tance from the Institute of Organic Chemistry, C.D. Nenitzescu” of
the Romanian Academy and the Ministry of National Education –
UEFISCDI (Contract No 727/2013) are also gratefully appreciated.
[8] A.F. Noels, A. Demonceau, J. Phys. Org. Chem. 11 (1998) 602–609.
[9] (a) N. Ledoux, B. Allaert, S. Pattyn, H. Vander Mierde, C. Vercaemst, F. Verpoort,
Chem. Eur. J. 12 (2006) 4654–4661;
(b) I. Dragutan, V. Dragutan, Platinum Metals Rev. 50 (2006) 81–94;
(c) B. Allaert, N. Dieltiens, C. Stevens, R. Drozdzak, I. Dragutan, V. Dragutan,
F. Verpoort, in: Y. Imamoglu, V. Dragutan (Eds.), Metathesis Chemistry: From
Nanostructure Design to Synthesis of Advanced Materials, Springer, Dordrecht,
2007, pp. 39–78;
(d) I. Dragutan, V. Dragutan, L. Delaude, A. Demonceau, A.F. Noels, Rev.
Roumaine Chim. 52 (2007) 1013–1025;
(e) S. Monsaert, R. Drozdzak, V. Dragutan, I. Dragutan, F. Verpoort, Eur. J. Inorg.
Chem. (2008) 432–440;
(f) V. Dragutan, I. Dragutan, H. Fischer, J. Inorg. Organomet. Polym. Mater. 18
(2008) 18–31;
(g) V. Dragutan, I. Dragutan, H. Fischer, J. Inorg, Organomet. Polym. Mater. 18
(2008) 311–324;
(h) A.M. Lozano Vila, S. Monsaert, R. Drozdzak, S. Wolowiec, F. Verpoort, Adv.
Synth. Catal. 351 (2009) 2689–2701;
(i) S. Monsaert, A.M. Lozano Vila, R. Drozdzak, P. Van Der Voort, F. Verpoort,
Chem. Soc. Rev. 38 (2009) 3360–3372;
(j) F. Ding, B. Yu, S. Monsaert, Y.-g. Sun, E. Gao, I. Dragutan, V. Dragutan,
F. Verpoort, Spectrochim. Acta A: Mol. Biomol. Spectrosc. 77 (2010)
170–174;
References
[1] (a) O. Robles, S.O. Serna-Saldívar, J.A. Gutiérrez-Uribe, D. Romo, Org. Lett. 14
(2012) 1394–1397;
(b) F. Brackmann, A. de Meijere, Chem. Rev. 107 (2007) 4493–4537;
(c) D.Y.-K. Chen, R.H. Pouwer, J.-A. Richard, Chem. Soc. Rev. 41 (2012)
4631–4642;
(d) P. Tang, Y. Qin, Synthesis 44 (2012) 2969–2984;
(e) H.-U. Reissig, R. Zimmer, Chem. Rev. 103 (2003) 1151–1196;
(f) J. Petruszka, Chem. Rev. 103 (2003) 1051–1070;
(g) W.A. Donaldson, Tetrahedron 57 (2001) 8589–8627.
[2] (a) S. Chanthamath, D.T. Nguyen, K. Shibatomi, S. Iwasa, Org. Lett. 15 (2013)
772–775;
(k) I. Dragutan, V. Dragutan, C. Mitan, H.C.M. Vosloo, L. Delaude, A. Demonceau,
Beilstein J. Org. Chem. 7 (2011) 699–716;
(l) I. Dragutan, V. Dragutan, A. Demonceau, RSC Adv. 2 (2012) 719–736.
[10] J. Van Craenenbroeck, K. Van Isterdael, C. Vercaemst, F. Verpoort, New J. Chem.
19 (2005) 942–947.

94
F. Ding et al. / Journal of Molecular Catalysis A: Chemical 386 (2014) 86–94
[11] (a) R. Drozdzak, B. Allaert, N. Ledoux, I. Dragutan, V. Dragutan, F. Verpoort,
Coord. Chem. Rev. 249 (2005) 3055–3074;
(b) A. Demonceau, C.A. Lemoine, A.F. Noels, L.T. Chizhevsky, P.V. Sorokin, Tetra-
hedron Lett. 36 (1995) 8419–8422.
(b) R. Drozdzak, B. Allaert, N. Ledoux, I. Dragutan, V. Dragutan, F. Verpoort, Adv.
Synth. Catal. 347 (2005) 1721–1743;
(c) R. Drozdzak, N. Ledoux, B. Allaert, I. Dragutan, V. Dragutan, F. Verpoort, Cent.
Eur. J. Chem. 3 (2005) 404–416;
(d) F. Ding, Y. Sun, S. Monsaert, R. Drozdzak, I. Dragutan, V. Dragutan, F. Ver-
poort, Curr. Org. Synth. 5 (2008) 291–304;
(e) S. Monsaert, N. Ledoux, R. Drozdzak, F. Verpoort, J. Polym. Sci. A: Polym.
Chem. 48 (2010) 302–310;
[15] E. Graban, F.R. Lemke, Organometallics 21 (21) (2002) 3823–3826.
[16] (a) B.P. Peppers, S.T. Diver, J. Am. Chem. Soc. 126 (2004) 9524–9525;
(b) B. Alcaide, P. Almendros, A. Luna, Chem. Rev. 109 (2009) 3817–3858.
[17] (a) D.M. Hodgson, D. Angrish, Chem. Eur. J. 13 (2007) 3470–3479;
(b) D.M. Hodgson, D. Angrish, Chem. Commun. (2005) 4902–4904.
[18] (a) W. Baratta, W.A. Herrmann, R.M. Kratzer, P. Rigo, Organometallics 19 (2000)
3664–3669;
(b) M. Basato, C. Tubaro, A. Biffis, M. Bonato, G. Buscemi, F. Lighezzolo, P. Lunardi,
C. Vianini, F. Benetollo, A. Del Zotto, Chem. Eur. J. 15 (2009) 1516–1526.
[19] (a) M. Ulman, T.R. Belderrain, R.H. Grubbs, Tetrahedron Lett. 41 (2000)
4689–4693;
(f) M. Dimonie, S. Coca, V. Dragutan, J. Mol. Catal. A: Chem. 76 (1992) 79–91.
[12] H.M.L. Davies, E.G. Antoulinakis, Org. React. 57 (2001) 1–326.
[13] J. Louie, R.H. Grubbs, Organometallics 21 (2002) 2153–2164.
[14] (a) A. Demonceau, E. Abreu Dias, C.A. Lemoine, A.W. Stumpf, A.F. Noels, C.
Pietraszuk, J. Gulinski, B. Marciniec, Tetrahedron Lett. 36 (1995) 3519–3522;
(b) G.S. Forman, R.P. Tooze, J. Organomet. Chem. 690 (2005) 5863–5866.
[20] B. De Clercq, F. Verpoort, J. Mol. Catal. A: Chem. 180 (2002) 67–76.