Synthesis of α,β-unsaturated aldehydes and nitriles via cross-metathesis reactions using Grubbs' catalysts

Article abstract of DOI:10.1016/j.apcata.2014.08.032

A series of α,β-unsaturated aldehydes and nitriles of significant interest in the fragrance industry have been prepared using Grubbs' catalysts in cross-metathesis reactions of electron-deficient olefins (i.e., acrolein, crotonaldehyde, methacrolein, and acrylonitrile) with various 1-alkenes, including 1-decene, 1-octene, 1-hexene and 2-allyloxy-6-methylheptane. The latter is of particular interest, as it has not previously being used as a substrate in cross-metathesis reactions and allows access to valuable intermediates for the synthesis of new fragrances. Most reactions gave good selectivity of the desired CM product (≥90%). Detailed optimisation and mechanistic studies have been performed on the cross-metathesis of acrolein with 1-decene. Recycling of the catalyst has been attempted using ionic liquids.

Full text of DOI:10.1016/j.apcata.2014.08.032

Applied Catalysis A: General 486 (2014) 94–104
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Synthesis of ␣,␤-unsaturated aldehydes and nitriles via
cross-metathesis reactions using Grubbs’ catalysts
Sandra M. Rountreea, Sarah F.R. Taylora, Christopher Hardacrea,∗∗
,
M. Cristina Lagunasa,∗, Paul N. Daveyb
a QUILL Research Centre, School of Chemistry and Chemical Engineering, Queen’s University, Belfast BT9 5AG, Northern Ireland, UK
b Givaudan Schweiz AG, Ueberlandstrasse 138, CH-8600 Duebendorf, Switzerland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 22 May 2014
Received in revised form 18 August 2014
Accepted 25 August 2014
Available online 2 September 2014
A series of ␣,␤-unsaturated aldehydes and nitriles of significant interest in the fragrance industry have
been prepared using Grubbs’ catalysts in cross-metathesis reactions of electron-deficient olefins (i.e.,
acrolein, crotonaldehyde, methacrolein, and acrylonitrile) with various 1-alkenes, including 1-decene,
1-octene, 1-hexene and 2-allyloxy-6-methylheptane. The latter is of particular interest, as it has not
previously being used as a substrate in cross-metathesis reactions and allows access to valuable interme-
diates for the synthesis of new fragrances. Most reactions gave good selectivity of the desired CM product
(≥90%). Detailed optimisation and mechanistic studies have been performed on the cross-metathesis of
acrolein with 1-decene. Recycling of the catalyst has been attempted using ionic liquids.
© 2014 Elsevier B.V. All rights reserved.
Keywords:
Cross-metathesis
Grubbs’ catalysts
Acrolein
Fragrances
1. Introduction
It has been established that ␣,␤-unsaturated aldehydes and
nitriles do not react with Grubbs’ first generation catalyst (1,
␣,␤-Unsaturated aldehydes and nitriles are of significant inter-
est in the fragrance and flavour industries, either by themselves or
as key intermediates in the preparation of other synthetic scents
[1]. They are also important intermediates in numerous organic
processes, such as Diels Alder reactions [2], Michael additions [3]
or aldol condensations [4], to name a few. Wittig-type reactions
are commonly used to prepare ␣,␤-unsaturated aldehydes and
nitriles [5], but a variety of other methods have also been reported,
including, for example, catalytic oxidation of allylic alcohols [6],
Ru-catalysed isomerisation of propargylic alcohols [7], or synthe-
sis via Grignard reagents [8]. In addition, cross-metathesis (CM)
reactions have been applied to prepare ␣,␤-unsaturated aldehydes
and nitriles, but this route is still relatively unexplored, in particular
with substrates containing electron deficient double bonds such as
acrolein and related ␣-carbonyl compounds [9]. It should be noted
that synthetic routes involving cross-metathesis have been shown
to be advantageous over more conventional methods. For exam-
ple, they allow direct access to chiral pharmaceutical intermediates
without the need for a chiral resolution step [9c].
Chart 1) [10], but are able to react in cross-metathesis reac-
tions using ruthenium catalysts with NHC ligands (e.g., 2–3 in
Chart 1). This is attributed to the greater electron donating abil-
ity of the NHC ligand compared to tricyclohexylphosphine (PCy3)
[10a]. Thus, good yields of CM products using NHC-Ru catalysts
have been reported, for example, in reactions involving acrolein,
methacrolein, acrylic acid, crotonaldehyde or acrylates [9b–e, 9g,
11]. Cross-metathesis reactions using acrylonitrile as substrate are
more common and include processes using Schrock’s molybde-
num catalyst [12], as well as a variety of NHC-ruthenium catalysts
[9g,13]. As in the case of acrolein, acrylonitrile shows poor reac-
tivity with catalyst 1 [10c,13e]. Some of these reactions have been
reported to work in neat conditions [11].
Selective formation of the cross-metathesis product can be
achieved by combining an olefin that shows slow or no self-
metathesis (classified as Type II/III olefins) with an olefin able
of fast self-metathesis or Type I (e.g., 1-alkenes; see Scheme 1)
[14]. In these reactions, the Type I homo-dimer may form ini-
tially [Scheme 1(b)], but undergoes secondary metathesis with the
Type II/III olefin to produce the desired CM product [Scheme 1(c)].
Although the classification of a particular olefin depends on the cat-
alyst, acrolein and related compounds, including acrylonitrile, are
generally classified as Type II or Type III olefins [14].
∗
Corresponding author. Tel.: +44 028 90974436; fax: +44 028 90976524.
Corresponding author. Tel.: +44 028 90974592; fax: +44 028 90976524.
E-mail addresses: c.hardacre@qub.ac.uk (C. Hardacre), c.lagunas@qub.ac.uk
∗∗
Ruthenium-catalysed cross-coupling reactions with terminal
olefins almost always occur with a high degree of E-selectivity [15],
(M.C. Lagunas).
http://dx.doi.org/10.1016/j.apcata.2014.08.032
0926-860X/© 2014 Elsevier B.V. All rights reserved.

S.M. Rountree et al. / Applied Catalysis A: General 486 (2014) 94–104
95
acrylonitrile) with various 1-alkenes. Detailed kinetic and mech-
anistic studies have been performed on the cross-metathesis of
acrolein with 1-decene and attempts to recycle the catalyst by
using ILs are also reported. Of particular interest are the reactions
using 2-allyloxy-6-methylheptane, as this substrate has not previ-
ously being used in cross-metathesis. The product from its reaction
with acrolein, 6,10-dimethyl-5-oxoundec-2-enal, is of use in the
synthesis of a novel fragrance material [22].
2. Experimental
2.1. Catalyst preparation
Catalysts 1–3 were obtained from Aldrich and used as received.
Ionic catalyst 4 was synthesised following a literature procedure
[21e]. The ionic liquids were prepared as described elsewhere [23].
Catalysts and ionic liquids were handled and stored in a glovebox.
2.1.1. Preparation of supported catalyst 3
Buffered silica was first prepared by stirring 5 g of the cor-
responding Davicat silica 1 or 2 (see Table 4) in distilled water
(100 mL) with a BHD buffer capsule (pH 7) for 2 h at 25 ◦C. The sil-
ica was then filtered, washed with distilled water (3 mL × 20 mL)
and dried in an oven (85 ◦C) on a watch-glass overnight. Catalyst 3
(6.7 mg, 0.01 mmol) was weighed into a Schlenk flask, under nitro-
gen. To this was added dichloromethane (5 mL), [C4mim][NTf2]
(0.1 mL) and pre-dried silica (0.2 g, either untreated or buffered).
This mixture was stirred for 10–15 min and the dichloromethane
then removed under vacuum. Further drying under vacuum (4–5 h)
produced the supported catalyst as a green powder.
Chart 1. Grubbs’ first (1) and second (2) generation catalysts, and Hoveyda-Grubbs’
catalysts (3, 4).
although strategies to favour Z-selectivity have been developed
[16]. Acrylonitrile, however, generally produces cross-metathesis
products where the Z isomer is predominant, and it has been pro-
posed that this Z-stereoselectivity must be kinetically controlled
and most likely related to the small size and/or electron-
withdrawing properties of the cyano group [12,13b,13e,17]. A few
examples in which cross-metathesis reactions of acrylonitrile have
produced the E-isomer as the major component have also been
reported [18].
In order to improve recyclability of Grubbs’ catalysts, ionic
liquids (ILs) have been used as reaction media or for the immobilisa-
tion of the catalyst onto silica (e.g., using the ‘supported ionic liquid
phase’ or SILP concept) [19]. For example, Grubbs’ second genera-
tion catalyst 2 has shown enhanced activity in CM reactions in ionic
liquids, compared to dichloromethane [20], and Hoveyda-Grubbs’
type catalysts immobilised onto silica gave excellent yields in RCM
metathesis reactions [19a]. In both cases, the catalysts could be eas-
ily recycled up to four times. In addition to conventional Grubbs’
catalysts, ionic analogues have been developed (e.g., complex 4 in
Chart 1) with the aim of increasing the catalyst’s retention time
in the ionic liquid, thus improving its recyclability and reducing
leaching into the organic phase [21].
2.2. Catalyst characterisation
Catalyst 4 was characterised by NMR spectroscopy. Its 1H and
13C NMR spectra were in good agreement with literature data
[24]. NMR spectra were performed on either a Bruker Avance DRX
(300 MHz) or DPX (500 MHz) spectrometer.
ICP analyses were performed at the ASEP unit in Queen’s
University Belfast. Samples to determine the amount of cata-
lyst dissolved in various solvents were prepared by adding 9 mg
(0.014 mmol) of catalyst 3 and 10 mL of solvent (dichloromethane,
1,2-dichloroethane, ethyl acetate, methanol, tetrahydrofuran, 2-
propanol or pentane) into a Schlenk flask, under nitrogen. The
mixtures were stirred for 10 min and then passed through a sin-
tered glass funnel. Any undissolved catalyst remaining on the
funnel was washed with dichloromethane (5 mL) into a crucible.
The dichloromethane was allowed to evaporate and the remaining
residue in the crucible was submitted for ICP-MS analysis. A sample
was also analysed containing the same amount of catalyst (9 mg,
0.014 mmol) and 10 mL of dichloromethane but no filtration was
undertaken, giving a value of 250 ppm of Ru. Compared with this
standard the % of undissolved Ru in each solvent was calculated:
dichloromethane (2%), 1,2-dichloroethane (8%), ethyl acetate (14%),
methanol (17%), tetrahydrofuran (49%), 2-propanol (50%) and pen-
tane (88%).
In this paper, a series of ␣,␤-unsaturated aldehydes and nitriles
of significant interest in the fragrance industry have been prepared
using Grubbs’ catalysts in cross-metathesis reactions of electron-
deficient olefins (i.e., acrolein, crotonaldehyde, methacrolein, and
2.3. Catalytic tests
2.3.1. General
The reagents 1-decene, 1-octene, 1-hexene, 2-allyloxy-6-
methylheptane, acrolein, methacrolein, crotonaldehyde, acry-
lonitrile and decane (all obtained from Aldrich), and solvents
(dichloromethane, 1,2-dichloroethane, ethyl acetate, 2-propanol
and pentane) were dried over activated molecular sieves (85 ◦C)
and degassed with nitrogen for 1–2 h prior to use. Methanol and
tetrahydrofuran were collected from a MBRAUN manual solvent
Scheme 1. Reactions involved in the cross-metathesis of olefins of Type I (RC CH2)
and Type II/III (RꢀC CH2).

96
S.M. Rountree et al. / Applied Catalysis A: General 486 (2014) 94–104
purification system, under nitrogen, and dried over molecular
sieves (85 ◦C). All other reagents were obtained from Aldrich and
used as received. GC-FID samples were analysed using a Hewlett
Packard 6890 GC fitted with an RTX-5 column (30 m, 0.25 ␮m diam-
eter). GC–MS (Perkin Elmer Turbo Mass) was performed using a
PE5MS column (length, 30 m; thickness, 0.25 ␮m; IDO, 0.32 mm).
Decane was used as internal standard for GC analysis.
acrolein (0.07 mL, 1 mmol), and the reaction mixture was stirred,
under nitrogen, for 24 h. Most of the solvent was removed under
vacuum and the remaining mixture was then passed through a sil-
ica column (5 cm × 2.5 cm) prepared with hexane. Hexane (20 mL)
was flushed through the column followed by a 20:80 mixture of
diethyl ether:hexane (ca. 100 mL). Fractions containing the desired
product were identified by TLC, when eluting with the diethyl
ether:hexane mixture. These fractions were combined, and the
solvent removed under vacuum. The remaining residue was then
dried under vacuum overnight, yielding 132 mg (0.78 mmol; 78%)
of undec-2-enal (NMR data included in Appendix A).
2.3.2. Reaction of acrolein and 1-decene in organic solvents
2.3.2.1. General procedure. Unless otherwise stated, this procedure
was applied to all the reactions described below. The catalyst was
weighed into a Schlenk flask, in a glovebox. The flask was then
taken out of the glovebox and opened to a Schlenk line, under
nitrogen. To this, the corresponding solvent was added, followed
by decane (internal standard) and the substrates required in each
case (see below). The reaction mixture was stirred under nitro-
gen, at r.t., for 24 h. During this time, the Schlenk flask was kept
opened to the nitrogen line to allow volatile by-products (ethene
or propene) to escape. Samples (0.5 mL) for GC analysis were taken
approximately every 10 min for 1 h, and then every hour for 8 h. A
final sample was taken after 24 h. Each sample was quenched with
tert-butylhydroperoxide (TBHP, 1 drop), diluted with diethyl ether
(1.6 mL) and filtered through a silica plug prior to being submitted
for GC analysis. An example of a GC trace obtained in these reactions
is shown in Appendix A (Fig. A1).
2.3.3. Mechanistic studies
Unless otherwise indicated, samples for GC analysis were taken
and treated as described above (Section 2.3.2.1).
2.3.3.1. Reactions with added products. To a Schlenk flask con-
taining catalyst 3 (9 mg, 1.4 mol%), under nitrogen, were added
dichloromethane (20 mL), decane (0.2 mL, 1 mmol), 1-decene
(0.2 mL, 1 mmol) and acrolein (0.07 mL, 1 mmol), and the result-
ing mixture was stirred. To this, a solution of undec-2-enal (0.2 mL,
1 mmol) or octadec-9-ene (269.6 mg, 1 mmol) in dichloromethane
(5 mL), under nitrogen, was added slowly over 1 h (ca. 1 mL every
10 min).
2.3.3.2. Slow addition of catalyst. A solution of catalyst 3 (10.6 mg,
1.7 mol %) in dichloromethane (6 mL) was prepared in a Schlenk
flask. This was added slowly, either over 3 h (i.e., 1 mL every 30 min)
or over 4.5 h (ca. 0.7 mL every 30 min), to a Schlenk flask contain-
ing dichloromethane (20 mL), decane (0.2 mL, 1 mmol), 1-decene
(0.2 mL, 1 mmol) and acrolein (0.07 mL, 1 mmol). Samples (0.5 mL)
for GC analysis were taken every 30 min during addition of catalyst
and every hour after for 8 h; a final sample was taken after 24 h.
2.3.2.2. Variation of catalyst. To the corresponding catalyst 1–4
(0.1 mmol, 10 mol%), the following were added: dichloromethane
(20 mL), decane (0.2 mL, 1 mmol), 1-decene (0.2 mL, 1 mmol) and
acrolein (0.07 mL, 1 mmol).
2.3.2.3. Variation of catalyst (3) loading. To the corresponding
amount of catalyst
3 [2.2 mg (0.35 mol%), 6.7 mg (1.0 mol%),
9.0 mg (1.4 mol%), 21.2 mg (3.5 mol%), 42.5 mg (7 mol%), or 85 mg
(14 mol%)] was added dichloromethane (20 mL), decane (0.2 mL,
1 mmol), 1-decene (0.2 mL, 1 mmol) and acrolein (0.07 mL,
1 mmol).
2.3.3.3. Addition of fresh catalyst. To a Schlenk flask contain-
ing catalyst 3 (4.5 mg, 0.007 mmol), under nitrogen, were added
dichloromethane (20 mL), decane (0.2 mL, 1 mmol), 1-decene
(0.2 mL, 1 mmol) and acrolein (0.07 mL, 1 mmol). The resulting
mixture was stirred for 4 h, after which, more catalyst 3 (4.5 mg,
0.007 mmol) was added.
2.3.2.4. Variation of solvent. The same amount of catalyst
3
(21.2 mg, 3.5 mol%) was added into six Schlenk flasks. To each
flask were added 20 mL of the corresponding solvent (1,2-
dichloroethane, ethyl acetate, pentane, tetrahydrofuran, methanol
or 2-propanol) followed by decane (0.2 mL, 1 mmol), 1-decene
(0.2 mL, 1 mmol) and acrolein (0.07 mL, 1 mmol).
2.3.3.4. Delayed addition of 1-decene. The same amount of catalyst
3 (9 mg, 1.4 mol%) was added into two Schlenk flasks together with
dichloromethane (20 mL), decane (0.2 mL, 1 mmol) and acrolein
(0.07 mL, 1 mmol). The resulting mixtures were stirred and 1-
decene (0.2 mL, 1 mmol) was added after 1 h to one of the flasks,
and after 2 h to the other.
2.3.2.5. Variation of 1-decene:acrolein ratio. To
a
Schlenk
flask loaded with catalyst 3 (21.2 mg, 3.5 mol%) were added
dichloromethane (20 mL), decane (0.2 mL, 1 mmol), 1-decene
(0.2 mL, 1 mmol) and acrolein [0.14 mL (2 mmol) or 0.35 mL
(5 mmol)].
2.3.4. Reactions with various substrates
To a Schlenk flask loaded with catalyst 3 (13.4 mg, 2 mol%)
were added dichloromethane (40 mL), decane (0.2 mL, 1 mmol)
and either 1-decene, 1-octene, 1-hexene or 2-allyloxy-6-
methylheptane (1 mmol) along with acrolein, crotonaldehyde,
methacrolein or acrylonitrile (1 mmol). The reaction mixtures
were stirred, under nitrogen, for 24 h. Samples for GC analysis
were taken as described above (Section 2.3.2.1). In order to isolate
the cross-metathesis products the reactions were carried out in
the same way, in dichloromethane, but samples were not taken
for GC analysis. In each case, most of the solvent was removed
under vacuum after 24 h, and the remaining mixture passed
through a silica column (5 cm × 2.5 cm) prepared with hexane.
Hexane (20 mL) was flushed through the column followed by
diethyl ether:hexane (20:80; ca. 100 mL). Fractions containing
the desired product were identified by TLC, when eluting with
the diethyl ether:hexane mixture. These fractions were com-
bined, and the solvents removed under vacuum. The remaining
2.3.2.6. Variation of [1-decene]. The same amount of catalyst 3
(9 mg, 1.4 mol%) was added into three Schlenk flasks. To each
flask was added dichloromethane (40 mL, 5 mL or 2 mL), decane
(0.2 mL, 1 mmol), 1-decene (0.2 mL, 1 mmol) and acrolein (0.07 mL,
1 mmol).
2.3.2.7. Varying the catalyst loading in ethyl acetate. To a Schlenk
flask loaded with catalyst 3 [6.7 mg (1.0 mol%), 9.0 mg (1.4 mol%),
21.2 mg (3.5 mol%), 42.5 mg (7 mol%)] were added ethyl acetate
(20 mL), decane (0.2 mL, 1 mmol), 1-decene (0.2 mL, 1 mmol) and
acrolein (0.07 mL, 1 mmol).
2.3.2.8. Isolation of undec-2-enal. To a Schlenk flask loaded with
catalyst
3 (21.2 mg, 0.034 mmol; 3.5 mol%) dichloromethane
(20 mL) was added, along with 1-decene (0.2 mL, 1 mmol) and

S.M. Rountree et al. / Applied Catalysis A: General 486 (2014) 94–104
97
Table 1
Reaction data for the reaction between acrolein and 1-decene.a
Entry
Variant
Conversion of
1-decene (%)
Yield of
undec-2-enal (%)
Selectivity
Undec-2-enal (CM)
Octadec-9-ene (SM)
Catalyst (10 mol%)b
1a
1b
1c
1d
1
2
3
4
30
54
88
60
3
48
70
59
15
98
95
94
85
2
5
6
1e
1f
1g
1h
1i
14 mol%
7.0 mol%
3.5 mol%
1.4 mol%
1.0 mol%
0.35 mol%
88
87
88
79
74
45
80
71
69
68
63
39
95
92
95
98
98
98
5
8
5
2
2
2
Catalyst 3 loading
(dichloro-methane)b
1j
Solventb,c
1k
1l
1m
1n
1o
1p
1,2-Dichloroethane
Ethyl acetate
Pentane
Tetrahydrofuran
Methanol
87
74
69
52
37
16
67
56
40
37
5
92
93
91
96
66
94
8
7
9
4
34
6
2-Propanol
12
1-Decene: acroleind
[1-Decene]e
1q
1r
1:2
1:5
89
89
72
74
97
96
3
4
1s
1t
1u
0.025 M
0.2 M
0.5 M
87
77
64
81
69
67
99
98
99
1
2
1
1v
1x
1y
7.0 mol%
1.4 mol%
1.0 mol%
80
45
35
60
49
42
96
97
97
4
3
3
Catalyst 3 loading
(ethyl acetate)b
a
All reactions performed at r.t. in dichloromethane (20 mL), unless stated otherwise. All data obtained by GC analysis after 8 h. The estimated error for the GC yields is
2%.
b
1-Decene:acrolein = 1:1 (1.0 mmol).
Reactions in 20 mL of solvent and 3.5 mol% catalyst 3.
1.0 mmol of 1-decene, 3.5 mol% catalyst 3.
1-Decene:acrolein = 1:1 (1.0 mmol); 1.4 mol% catalyst 3 in 40 mL (1s), 5 mL (1t) or 2 mL (1u) of dichloromethane.
c
d
e
residue was then dried under vacuum overnight, yielding the
corresponding cross-metathesis product (see Table 3 for isolated
yields and Appendix A for NMR data of the products derived from
2-allyloxy-6-methylheptane).
2.3.5.3. Reactions using supported catalyst 3 in pentane. Catalyst
3 (6.7 mg, 2 mol%) was supported on Silica 1 or 2 (untreated or
buffered) using the method described above. To this was added
pentane (20 mL), decane (0.1 mL, 0.5 mmol), 1-decene (0.1 mL,
0.5 mmol) and acrolein (0.035 mL, 0.5 mmol). The reaction mixture
was stirred, at r.t., for 6 h, under nitrogen. After this, the pentane
layer was separated with a syringe and the supported catalyst
was washed with pentane (3 mL × 20 mL). The solvent from the
combined pentane extracts was removed under vacuum and the
residue analysed by GC. For this, samples were quenched with
tert-butylhydroperoxide (TBHP, 1 drop), diluted with diethyl ether
(1.6 mL) and filtered through a silica plug prior to being submitted
for GC analysis. The supported catalyst was recovered by dry-
ing under vacuum. To begin a new reaction, fresh pentane and
substrates were then added to the silica-supported catalyst. This
process was repeated twice for a total of three runs.
2.3.5. Reactions using ionic liquids
2.3.5.1. Cross-metathesis reactions in ionic liquids (ILs). To
a
Schlenk flask containing catalyst (3.4 mg, 2 mol%), under
3
nitrogen, were added 10 mL of the corresponding ionic liq-
uid ([1-butyl-3-methylimidazolium][X], X = NTf2 (Tf = CF3SO2),
PF6, OTf; or [A][NTf2], A = 1-butyl-2,3-dimethylimidazolium,
1-butyl-1-methyl-pyrrolidinium,
methyl(trioctyl)ammonium,
tetradecyl(trihexyl)phosphonium) followed by decane (0.05 mL,
0.25 mmol), 1-decene (0.05 mL, 0.25 mmol) and acrolein (0.02 mL,
0.28 mmol), and the resulting mixture was stirred at r.t. for 24 h.
After this, the IL layer was extracted with pentane (3 mL × 10 mL).
All the pentane extracts were combined and the solvent removed
under vacuum. The residue was analysed by 1H NMR spectroscopy.
To begin a new reaction fresh substrates were added to the IL.
3. Results and discussion
2.3.5.2. Cross-metathesis reactions in IL:DCM (1:1). To a Schlenk
flask containing catalyst 3 (3.4 mg, 1.4 mol%) or 4 (6.5 mg, 3 mol%)
under nitrogen, were added 5 mL of [C4mim][NTf2] and 5 mL of
dichloromethane, followed by decane (0.05 mL, 0.25 mmol), 1-
decene (0.05 mL, 0.25 mmol) and acrolein (0.02 mL, 0.28 mmol).
The resulting mixture was stirred at r.t. for 24 h. After this, the
dichloromethane was removed under vacuum and the IL layer was
extracted with pentane (3 mL × 10 mL). All the pentane extracts
combined and the solvent removed under vacuum. The residue was
analysed by 1H NMR spectroscopy. To begin a new reaction fresh
dichloromethane and substrates were added to the IL.
3.1. Cross-metathesis between acrolein and 1-decene
The cross-metathesis reaction of acrolein and 1-decene was
studied in the first instance (Scheme 2). The effect of varying the
catalyst, solvent, concentration of 1-decene and ratio of starting
materials was analysed. The results are summarised in Table 1.
In all cases, the reactions yielded the cross-metathesis (CM) prod-
uct, undec-2-enal (normally as the major product), together with
the self-metathesis (SM) product of 1-decene (octadec-9-ene). As
expected, self-metathesis of acrolein was not detected in any of the
reactions [14]. The reactions took place with high E-selectivity and

98
S.M. Rountree et al. / Applied Catalysis A: General 486 (2014) 94–104
[10a,25]. In the case of Grubbs first (1) and second (2) generation
catalysts, the initiation step involves the reversible dissociation
of phosphine to form a catalytically active 14-electron complex,
which will subsequently coordinate an olefinic substrate. Re-
coordination of free phosphine regenerates the initial 16-electron
complex resulting in loss of activity. Although phosphine dissoci-
ation in catalyst 2 has been shown to be slower than for 1, once
dissociated, coordination of the olefin is more favourable than re-
attachment of the phosphine in the case of catalyst 2. This often
results in 2 being more active than 1 in metathesis reactions [25a].
The higher conversion of 1-decene obtained with catalyst 2 com-
pared to 1 (Table 1, entries 1a,b) agrees with the above mechanism.
In addition, the better ␴-donating ability of the NHC ligand in 2
compared to PCy3 in catalyst 1 favours the trans-coordination of
acrolein (an electron-poor olefin), allowing the cross-metathesis
process to occur, whereas mainly self-metathesis of 1-decene is
observed with catalyst 1 (unreactive towards acrolein). Similar
variations in reactivity for catalysts 1 and 2 with electron-poor
olefins have been reported before [9c,10, 26].
The phosphine-free Hoveyda-Grubbs’ catalysts (such as 3 and 4)
are activated through de-coordination of the chelating oxo group of
the styrenyl ligand, which can readily detach from the metal cen-
tre thus allowing the release of the styrene and attachment of an
olefinic substrate [27]. Compared to phosphine, the styrenyl ligand
is less effective at re-binding to the metal and, therefore, does not
compete as efficiently with olefin substrates [27b]. Accordingly,
catalyst 3 has been shown to be more active with electron-deficient
olefins than second and first generation Grubbs’ catalysts [27b,28].
The higher conversion of 1-decene obtained with catalyst 3 and
4 agrees with this (Table 1, entries 1c,d). The inclusion of an
electron-withdrawing group (EWG) in the benzylidene fragment
(as in catalyst 4) often increases the catalyst’s activity, as the EWG
weakens the O Ru bond and favours the initiation of the catalytic
reaction [17b,21e]. For example, catalyst 4 has shown higher activ-
ity than 3 in ring-closing metathesis reactions in dichloromethane
as well as in more polar solvents [21e]. In the reaction of 1-decene
with acrolein, however, conversion of 1-decene was lower with
catalyst 4 than 3. This may be related to poorer solubility of 4
in dichloromethane at the conditions used, although this was not
investigated in detail.
Scheme 2. (a) Cross-metathesis (CM) and (b) self-metathesis (SM) products formed
in the reaction between acrolein and 1-decene.
only this isomeric form of undec-2-enal was isolated, as indicated
by 1H NMR spectroscopy.
3.1.1. Variation of catalyst
The reaction between acrolein and 1-decene was carried out
in the presence of ruthenium catalysts 1–4 (Chart 1, Fig. 1). All the
reactions were performed, in the first instance, in dichloromethane
at r.t. (8 h) and 10 mol% of catalyst (Table 1, entries 1a–1d). Con-
versions between 30% (1) and 88% (3) were obtained, whereas
the best yields of CM product, undec-2-enal, were achieved with
Hoveyda-Grubbs’ catalysts 3 (70%) and 4 (59%). Catalysts 2–4 all
gave a high ratio of CM:SM product (with >94% selectivity of CM
product), whereas the opposite was found for 1 (85% selectivity of
SM product, octadec-9-ene). Given the above results, catalyst 3 was
used in the optimisation studies (see below).
The mechanism for Grubbs’ catalysed metathesis reactions
involves coordination of the olefins to the metal and exchange
of alkylidene groups through a ‘ruthenacyclobutane’ intermediate
3.1.2. Catalyst loading
A series of reactions were carried out with catalyst 3, varying the
catalyst loading in the range 0.35–14 mol% (Table 1, entries 1e–1j).
As expected, increasing the catalyst loading resulted in a gradual
increase in conversion of 1-decene, which ranged between ca. 45%
(using 0.35 mol% catalyst) and ca. 88% (using 14 mol%), as well as
an increase in the yield of CM product (ranging from 39% to 80%).
In all cases, the ratio of CM:SM product was high (>92% selectivity
of CM product). However, the differences in conversion and yield
in the reactions performed with catalyst loadings between 1.4 and
14 mol% are too small to justify the use of high amounts of catalyst.
Therefore, catalyst loadings between 1.4 and 3.5 mol% were used in
all subsequent experiments, as they provided the best balance of
activity and yield of CM product per mole of catalyst.
3.1.3. Variation of solvent
Grubbs’ catalysts are known to be affected by the nature of the
solvent; for example, high coordinating ability of the solvent and
the presence of protic functionalities are known to hinder metathe-
sis reactions, whereas polar solvents often result in higher initiation
rates [25a,29]. In order to study how these various factors affect
the reaction of acrolein and 1-decene, the reaction was examined
using 3.5 mol% catalyst 3 in 1,2-dichloroethane, ethyl acetate, pen-
tane, tetrahydrofuran, methanol and 2-propanol (Table 1, 1k–1p),
and compared with the results obtained, at the same conditions,
Fig. 1. Conversion of 1-decene (top) and formation of cross-metathesis product
(bottom) in the reaction between acrolein and 1-decene in the presence of catalysts
1–4 in dichloromethane using 10 mol% of catalyst.

S.M. Rountree et al. / Applied Catalysis A: General 486 (2014) 94–104
99
in dichloromethane (entry 1g). The effect of catalyst solubility was
also considered.
attempt to improve the yield of undec-2-enal, reactions with an
excess of acrolein (i.e., 2 and 5 equiv.) and 3.5 mol% of catalyst 3
were carried out in dichloromethane (Table 1, entries 1q–1r). There
were no significant differences in the results between the 1:2 and
1:5 reactions, as both processes resulted in good conversions of
1-decene (89%) and yield of undec-2-enal (72–74%, 96–97% selec-
tivity). Interestingly, these results are also comparable with those of
the 1:1 reaction under the same conditions (88% conversion, 69%
yield; 95% selectivity; entry 1g). Therefore, the ratio of the reac-
tants does not appear to significantly affect the results and there
are not significant benefits in using excess acrolein, which is toxic
and a lachrymator. This is in contrast with previous reports on CM
reactions involving acrolein, where excess of the olefin is required
[9b,9e].
The 1:1 reaction between acrolein and 1-decene was carried out
in varying volumes of dichloromethane, while maintaining con-
stant the catalyst loading (1.4 mol%). The results are summarised
in Table 1 as a function of the concentration of 1-decene (entries
1s–1u; see entry 1 h for [1-decene] = 0.05 M). Increasing the concen-
tration of 1-decene resulted in a gradual decrease in conversion of
1-decene from 87% to 64%. The best yield of CM product (81%) is also
obtained at the lowest concentration of 1-decene (0.03 M), whereas
the reactions at higher concentrations of substrate all gave similar
yields (67–69%). The selectivity towards undec-2-enal, however,
remained unaffected and was high (98–99%) in all cases. Loss of
activity with increased substrate concentration has been previously
shown for Ru catalysts in cross-metathesis reactions. This could be
related to poisoning of the catalyst [13e].
The reactions in methanol and 2-propanol produced the lowest
conversions of 1-decene (≤37%) and yields of undec-2-enal (≤12%).
The reaction in methanol also resulted in the lowest ratio (66:34)
between the CM and SM products, with all the other solvents
giving a high selectivity, >90%, of undec-2-enal. The best conver-
sions of 1-decene (87–88%) and yields (67–69%) were obtained in
dichloromethane or 1,2-dichloroethane, followed by ethyl acetate
(74% conversion of 1-decene, 56% yield of unde-2-enal).
During the reactions, it was clearly observed that catalyst 3 did
not fully dissolve in some of the solvents. In order to analyse the
catalyst solubility in more detail, ICP studies were undertaken fol-
lowing catalyst dissolution in all solvents. These showed that ca.
88% of Ru remained undissolved in pentane, whereas 2-propanol
and tetrahydrofuran dissolved ca. 50%. Better catalyst solubility
was found in methanol and ethyl acetate (14–17% of Ru remained
undissolved); with 1,2-dichloroethane and dichloromethane giving
the highest solubility (8% and 2% of Ru undissolved, respectively).
The same trend, with comparable amounts of undissolved Ru
remaining, was also found in the case of catalyst 1 during stud-
ies on the dimerisation reaction of cyclooctene [23]. The insoluble
rest of the catalyst may play a role in the catalytic process (e.g., via a
heterogeneous reaction and/or as an additional source of catalytic
species during the reaction). However, other factors should also be
considered.
In addition to low catalyst solubility, the poor activity observed
in protic solvents methanol and 2-propanol might be attributed to
their ability to degrade the catalyst [30]. In agreement with previ-
ous work, polar solvents dichloromethane, 1,2-dichloroethane and
ethyl acetate gave the best results [23,25a]. However, the lower
conversion and yield in tetrahydrofuran, compared with the other
polar solvents, may be related to the lower catalyst solubility in
tetrahydrofuran. The coordinating nature of the solvent does not
appear to be a determining factor, since both strongly and weakly
coordinating solvents (e.g., ethyl acetate and dichloromethane,
respectively) gave good results. It should be noted, however, that
pentane showed relatively good activity despite having the low-
est catalyst solubility. This may be related to its non-coordinating
nature.
The high ratio of self-metathesis product obtained in methanol
compared with all the other solvents tested, may be due to the coor-
dinating ability of methanol, although a similar effect is not seen
in ethyl acetate, tetrahydrofuran or 2-propanol, also able to bind to
the ruthenium centre through their oxygen atoms. The smaller size
of methanol may favour its coordination to the metal, thus hinder-
ing coordination of acrolein and/or octadec-9-ene to the catalyst to
form the CM product.
3.1.5. Effect of catalyst loading in ethyl acetate
As indicated in Section 3.1.3, the best non-chlorinated solvent
for the reaction was ethyl acetate and would, therefore, be more
suitable for industrial use. In order to optimise the results in this
solvent, the 1:1 reaction between acrolein and 1-decene was stud-
ied at various catalyst loadings ranging from 1 to 7 mol%, using
catalyst 3 (Table 1, see entries 1v–1y and 1l). The trends observed
were similar to those seen in dichloromethane, discussed above.
All the reactions gave good selectivity to undec-2-enal (ca. 97%)
and the conversion and yield of undec-2-enal increased with cat-
alyst loading, as expected (i.e., conversion of 1-decene increased
from 35% to 80% and yield of CM product from 42% to 60%). The
best overall results per mol of catalyst were obtained at a loading
of 3.5 mol% (74% conversion, 56% yield, entry 1l); with highest cata-
lyst loading (7 mol%, entry 1v) leading to only a small improvement
in conversion (80%) and yield (60%).
3.1.6. Isolation of undec-2-enal
Based on the above studies and in order to isolate the CM
product, undec-2-enal, in good yield, the 1:1 reaction between 1-
decene and acrolein was carried out using 3.5 mol% of catalyst 3, in
dichloromethane (20 mL) for 24 h. Purification of the crude prod-
uct by column chromatography gave undec-2-enal with an isolated
yield of ca. 78%. The 1H NMR spectrum of the isolated product
agreed well with that reported for E-undec-2-enal [31].
Overall, the good results obtained in dichloromethane and 1,2-
dichloroethane for the cross-metathesis reaction may be attributed
to a combination of polarity, good catalyst solubility, and weakly-
coordinating properties. However, the lower toxicity of ethyl
acetate, compared with the chlorinated solvents, makes it partic-
ularly interesting for industrial applications. Herein, further work
was carried out mainly using dichloromethane, although the effect
of catalyst loading in ethyl acetate was also analysed.
3.1.7. Mechanistic studies
None of the reactions discussed above resulted in complete con-
version of 1-decene, with the highest conversion achieved being
89% (Table 1). A series of studies were carried out to determine the
cause of these incomplete conversions, as well as to shed light on
the mechanism of the reaction. In particular, the potential poison-
ing of the catalyst with products undec-2-enal and octadec-9-ene
was analysed, as well as the effect of a delayed addition of 1-decene
or of the gradual addition of catalyst to the reaction.
3.1.4. Influence of concentration of substrate and ratio of starting
materials
In order to assess if the formation of cross-metathesis vs. self-
metathesis product was dependant of the ratio of the starting
materials and/or the concentration of substrate, the reaction was
monitored at various 1-decene:acrolein ratios, and as a function of
substrate concentration, using catalyst 3 in dichloromethane.
CM reactions often need an excess of one of the reagents to shift
the equilibrium and give good CM product yields [14]. Thus, in an
Reactions between 1-decene and acrolein (1:1), with added
undec-2-enal or octadec-9-ene (i.e., 1 equiv. added over 1 h), were

100
S.M. Rountree et al. / Applied Catalysis A: General 486 (2014) 94–104
Table 2
Summary of the results for the reactions between acrolein and 1-decene with added products (entries 2a–2b), varying the catalyst addition method (entries 2c–2e), and with
delayed addition of 1-decene (entries 2f–2g).a To facilitate comparison, data taken from Table 1 for a ‘standard’ reaction performed at the same conditions are also included
(entry 2 h).a
Entry
Reaction
Conversion of 1-decene (%)
Yield of undec-2-enal (%)
Selectivity
Undec-2-enal (CM)
Octadec-9-ene (SM)
2a
2b
2c
2d
2e
2f
Added undec-2-enalb
53
72
88
84
85
86
71
79
57
39
81
76
80
82
75
68
98
56
99
99
99
99
98
98
2
44
1
1
1
1
2
2
Added octadec-9-eneb
Catalyst added gradually over 3 h
Catalyst added gradually over 4.5 h
Fresh catalyst added after 4 h
1-Decene added after 1 h
1-Decene added after 2 h
Standard reactionc
2g
2h
a
All reactions performed at r.t. for 8 h, in dichloromethane (20 mL), using 1.4–1.7 mol% catalyst 3, 1-decene:acrolein = 1:1 (1 mmol). All data obtained by GC analysis. The
estimated error for the GC yields is 2%.
b
Values show actual amount of products formed during the reaction, as the amount of products added have been subtracted from the results.
As in entry 1h in Table 1.
c
carried out in dichloromethane, using 1.4 mol% catalyst 3 (Table 2,
entries 2a–b). Compared with the same reaction without added
products (entry 2h), the addition of undec-2-enal significantly
reduced the conversion of 1-decene (from 79% to 53%) and the
yield of undec-2-enal (from 68% to 57%), while maintaining high
selectivity towards the cross-metathesis product (98%). It appears,
therefore, that undec-2-enal is able to deactivate the catalyst, thus
decreasing conversion and yield of CM product. In contrast, the
addition of octadec-9-ene had little effect on the conversion of
1-decene (72%), indicating that the catalyst is not deactivated by
the SM product. The reaction, however, resulted in formation of
significantly less undec-2-enal (39%) and more octadec-9-ene (i.e.,
CM:SM = 56:44). It appears that octadec-9-ene added to the reac-
tion disfavours pathways (a) and/or (c) (Scheme 1) and leads to the
formation of less cross-metathesis product and more homo-dimer.
Although the mechanism of this process is currently unknown,
it may involve attachment to the catalyst of octadec-9-ene when
present at high enough concentration. It should be noted that the
self-metathesis of 1-decene in analogous conditions (i.e., 1.4 mol%
of catalyst 3 in dichloromethane) resulted in only 5% conversion of
1-decene. Relatively low conversions of 1-decene have previously
been reported using the Hoveyda-Grubbs’ catalyst (i.e., 57% conver-
sion obtained in toluene at 80 ◦C, Ru/1-decene = 1:100), attributed
to gradual catalyst deactivation [32].
In order to improve conversion by reducing the possibility of
catalyst deactivation by undec-2-enal, two reactions between 1-
decene and acrolein in dichloromethane were then performed
wherein catalyst 3 (1.4 mol%) was added gradually over 3 h or 4.5 h
(Table 2, entries 2c and 2d). In a separate experiment, 0.7 mol% of 3
was added at the beginning of the reaction with a further 0.7 mol%
being added after 4 h (Table 2, entry 2e). In all three reactions, com-
parable conversions of 1-decene (85–88%), yields of undec-2-enal
(76–81%), and selectivity (99%) were obtained. Overall, these results
are slightly better than the standard procedure, where all the cat-
alyst is added at the start (entry 2h), indicating that some catalyst
decomposition takes place during the reaction.
self-metathesis of acrolein. The improved performance of cata-
lyst 3 following mixing with acrolein for 1 h suggests that acrolein
activates the catalyst by adding onto it. A longer catalyst-acrolein
reaction time results in some deactivation, probably due to decom-
position of the catalyst. It has been shown that propagating species
of the type [Ru] CH(CO)R only form in a small percentage and
rapidly decompose [9b], or are not detected at all during the process
[9d]. On this basis, [Ru] CH(CO)R intermediates are not considered
responsible for the bulk of product formation in metathesis reac-
tions [9b,9d]. Interestingly, our results indicate that formation of
an analogous catalyst-acrolein species of the type [Ru] CH(CO)H,
although short-lived, has a significant effect on the performance of
the catalyst.
3.2. Cross-metathesis reactions with various substrates
The cross-metathesis of acrolein with three other alkenes,
1-hexene, 1-octene, and 2-allyloxy-6-methylheptane was inves-
tigated using catalyst 3. In addition, other type II/III olefins [14],
including crotonaldehyde, methacrolein and acrylonitrile, were
used instead of acrolein as starting materials (Scheme 3). All
the reactions produced the expected cross-metathesis products,
although the yields varied in a wide range, from 1% to 71% (Table 3).
As observed in the acrolein:1-decene reactions discussed above, no
self-metathesis product of the ␣,␤-unsaturated aldehyde/nitrile
was detected in any of the reactions, whereas formation of the
alkenes’ self-metathesis products was seen in all cases, with
the exception of 1-hexene. The fact that the self-metathesis
product of 1-hexene is not observed is surprising, in particular
in the reaction with methacrolein which appears to favour the
self-metathesis of the other olefins. The reasons for this behaviour
are currently unclear. The homo-dimer of 1-hexene may well
form in the reactions, but the smaller size of this alkene may
favour a rapid secondary metathesis reaction to form the CM
product, according to Scheme 1c. A series of cross-metathesis
reactions between 1-hexene and methyl vinyl ketone using
various Ru-NHC catalysts have been reported to produce only
traces of the self-metathesis product of methyl vinyl ketone
together with the cross-metathesis product. The self-metathesis
of 1-hexene was only observed in a few cases where conversion
was incomplete [33]. It should be also noted that in other CM
reactions reported in the literature, the homo-dimer of 1-hexene
has been observed. For example, the 2:1 reaction between acry-
lonitrile and 1-hexene using the pyridine-containing catalyst
[RuCl( CHPh){1,3-bis(2,6-dimethylphenyl)-4,5-dihydroimidazol-
2-ylidene}(py)2] (in dichloromethane at 45 ◦C), results in 5–7%
yield of self-metathesis product. In addition, the CM product,
hept-2-enenitrile, was obtained at 37% and 51% yield using 2 mol%
Finally, the effect of adding the acrolein to the catalyst first, for
either 1 h or 2 h, before addition of 1-decene (Table 2, entries 2f
and 2g, respectively), was investigated, and results compared with
a standard reaction in the same conditions (entry 2h). After 1 h,
acrolein does not appear to have degraded the catalyst, as it gives
the best conversion of 1-decene (86%) and yield of undec-2-enal
(82%), compared with 79% conversion and 68% yield, when both
starting materials are added together at the start. Delaying the addi-
tion of 1-decene for another hour, however, did not improve the
conversion (71%) and also led to a decrease in the yield of undec-
2-enal to 75%. The selectivity towards the CM product remains
high in all cases (98–99%); and none of the reactions showed

S.M. Rountree et al. / Applied Catalysis A: General 486 (2014) 94–104
101
Scheme 3. Cross-metathesis (CM) of 1-hexene, 1-octene, 1-decene, and 2-allyloxy-6-methylheptane with various ␣,␤-unsaturated aldehydes and acrylonitrile. Note that
the reactions with acrolein and crotonaldehyde produce the same CM products but ethene or propene is formed as by-product, respectively.
2-allyloxy-6-methylheptane (Scheme 3, R4), all the products had
previously been reported, and their 1H and 13C NMR spectra
were consistent with the published data [8,17a, 34]. As expected,
only the E isomer of the CM products was observed in the reac-
tions with the ␣, ␤-unsaturated aldehydes, whereas the reactions
with acrylonitrile produced in all cases a mixture of the Z and E
isomers of the corresponding ␣,␤-unsaturated nitrile. This pre-
dominance of the Z isomer in cross-metathesis reactions involving
and 10 mol% of catalyst, respectively [17a]. These results are
comparable to that found herein (i.e., 42% yield, using 2 mol% of
catalyst 3; see below).
All the cross-metathesis products were isolated using a sim-
ilar method to that described above for undec-2-enal (Section
3.1.6), and characterised by 1H and 13C NMR spectroscopy. In
all cases, the isolated yields were comparable with those deter-
mined by GC (Table 3). With the exception of the derivatives of
Table 3
Reaction data for the reactions 1-decene, 1-octene, 1-hexene and 2-allyloxy-6-methylheptane with various substrates.a
Entry
Reaction
Conversion (%)b
Yield of CM product
[isolated yield] (%)
Selectivityc
CM product
SM product
3a
3b
3c
1-Decene
Crotonaldehyde
Acrylonitrile
Methacrolein
80
60
42
65 [71]
70 [80]
24 [30]
91
99
59
9
1
41
3d
3e
3f
1-Octene
1-Hexene
Acrolein
77
80
62
43
71 [75]
47 [51]
56 [65]
28 [35]
99
90
99
63
1
10
1
Crotonaldehyde
Acrylonitrile
Methacrolein
3g
37
3h
3i
3j
Acrolein
74
71
55
30
52 [60]
49 [55]
42 [48]
16 [25]
100
100
100
100
0
0
0
0
Crotonaldehyde
Acrylonitrile
Methacrolein
3k
3l
Acrolein
76
43
19
20
33 [33]
12 [15]
3 [5]
92
83
85
18
8
17
15
82
2-Allyloxy-6-
methylheptane
3m
3n
3o
Crotonaldehyde
Acrylonitrile
Methacrolein
1 [4]
a
All reactions performed at r.t. for 8 h, in dichloromethane (40 mL), using ca. 2 mol% of catalyst 3; 1-decene:acrolein = 1:1 (1 mmol). All data obtained by GC analysis, with
the exception of the isolated yields. The estimated error for the GC yields is 2%.
b
Conversion of 1-alkene (entries 3a–3k) or 2-allyloxy-6-methylheptane (entries 3l–3o).
See Scheme 3 for cross-metathesis (CM) and self-metathesis (SM) products of all reactions.
c

102
S.M. Rountree et al. / Applied Catalysis A: General 486 (2014) 94–104
acrylonitrile has previously been reported [12,13b, 13e, 17].
According to NMR data of the isolated products, the CM prod-
ucts of acrylonitrile with 1-decene, 1-octene or 1-hexene were all
obtained with a Z/E ratio of 3; whereas the product with 2-allyloxy-
6-methylheptane (6,10-dimethyl-5-oxodec-2-enenitrile) resulted
in Z/E = 2.
perfumers at Givaudan agreed that the samples were different in
character although they were divided in their preference.
The reactions with 2-allyloxy-6-methylheptane (Table 3, entries
3l–3o) resulted in poorer yields of CM product (1–33%) and,
generally, gave lower conversions (20–76%) and selectivities
(18–92%) than those of the linear 1-alkenes. These results may be
related to the presence of the bulkier substituent in 2-allyloxy-6-
methylheptane. Similarly to that observed with the other alkenes,
the reaction with acrolein gave the best overall results; whereas
those with acrylonitrile and methacrolein gave the worst val-
ues. The observed preference for the self-metathesis product
is probably due to poor secondary metathesis of the sterically
demanding 2-allyloxy-6-methylheptane homo-dimer [Scheme 1c].
This effect increases with the deactivating nature of the ␣,␤-
unsaturated substrate (i.e., with increasing bulkiness or electron
withdrawing ability). In addition, the lower reactivity of 2-allyloxy-
6-methylheptane compared to the other alkenes tested could be
related to the formation of a vinyl ether by-product during the pro-
cess (i.e., via a Ru-catalysed double-bond migration [38]), as vinyl
ethers are known to attach irreversibly to Ru catalysts, thus deac-
tivating them [25a,39]. Whereas vinyl ether species have not been
detected during the experiments described herein, formation of
unreactive Ru-carbenes resulting from the reaction of the catalyst
and vinyl ethers cannot be ruled out.
In the reactions with the three linear 1-alkenes (Table 3, entries
3a–3k), similar trends were observed for each 1-alkene. Thus, the
best conversions (71–80%) were achieved when either acrolein or
crotonaldehyde were used as starting materials; whereas lower
values were obtained with acrylonitrile (55–62%), followed by
methacrolein (30–43%). The best yields of CM product were
consistently obtained with acrolein (52–71%), followed by acry-
lonitrile and crotonaldehyde which gave comparable yields for
each 1-alkene (i.e., ranging between 42% and 70%). All the reac-
tions with methacrolein gave significantly lower yields (16–28%).
Methacrolein also gave the lowest selectivities of CM product
with 1-decene (59%) and 1-octene (63%), compared with the
three other substrates (90–99%). The presence of a bulky Me
substituent next to the aldehyde functionality in methacrolein
appears to hinder its reaction with catalyst 3, resulting in low
conversions and yields of CM product. Sterically demanding sub-
strates often have a deactivating effect in Grubbs’ catalysed
metathesis reactions. For example, poor reactivity of methacrolein
(i.e., compared to crotonaldehyde) has been reported in cross-
metathesis with 1,5-cyclooctediene using catalyst 1 [35]. However,
our results contrast with those of Chatterjee et al. where good
reactivity of second generation Grubbs’ catalysts was found
with geminal disubstituted olefins [9b,36]. In particular, a better
yield was reported for the cross-metathesis of 6-acetoxyhex-1-
ene with methacrolein (92%), compared to acrolein (62%), using
catalyst [RuCl2( CHCHCMe2)(1,3-dimesityl-4,5-dihydroimidazol-
2-ylidene)(PCy3)] (Cy = cyclohexyl) [9b]. Therein, it was observed
that the trans orientation was favoured in the presence of the
Me group close to the metal centre [i.e., E:Z = 1.1:1 (acrolein),
>20:1 (methacrolein)]. As discussed above, in the reactions with
acrolein and methacrolein described, herein, the E isomer is always
favoured.
The lower conversions and CM yields obtained with acrylonitrile
compared with acrolein, may be related to the higher electron-
withdrawing ability of the CN group (i.e., stronger −I effect), thus
disfavouring its addition to the catalyst. In addition, the strongest
coordinating ability of the CN group may deactivate the catalyst
by binding to the metal through the N atom [17a]. It should be
noted that the cross-metathesis of acrylonitrile with 1-octene and
1-decene has been reported using 5 mol% of Schrock’s catalyst in
dichloromethane [12]. The yields of the corresponding CM prod-
ucts were of 56% and 72% respectively, which is consistent with the
values obtained, herein, for catalyst 3. The authors report higher
overall yields in diethyl ether, where the reactions were slower,
and relatively high Z/E ratios (8.5–9). The self-metathesis products
were obtained at 4% yield. The cross-metathesis of acrylonitrile and
1-decene has also been reported using ruthenium catalyst 2 (Chart
1) in dichloromethane (20 h, 2 mol% catalyst) [13a]. Although this
study did not include details on yields or selectivity, it was noted
that the purity of the acrylonitrile used in the reactions is of utmost
importance in achieving high yields and reproducibility.
3.3. Reactions using ionic liquids
The possibility of recycling the catalyst was examined
by performing the cross-metathesis between acrolein and
1-decene in ionic liquids. Reactions using catalyst 3 were car-
ried out in a series of ionic liquids with a common cation
([C4mim][anion];
anion = NTf2
([cation][NTf2];
C4mim = 1-butyl-3-methylimidazolium;
(Tf = CF3SO2), EtSO4, PF6, OTf) or anion
cation = 1-ethyl-3-methylimidazolium, 1-
hexyl-3-methylimidazolium, 1-butyl-2,3-dimethylimidazolium,
1-butyl-1-methyl-pyrrolidinium, methyl(trioctyl)ammonium,
tetradecyl(trihexyl)phosphonium). Only partial extraction of the
products from the ILs was achieved using pentane or diethyl ether,
but preliminary analysis of the reaction mixtures using NMR
spectroscopy showed that [C4mim][NTf2] produced the best yield
of CM product, albeit very low (<5%). In order to improve the yield,
reactions were attempted in a 1:1 [C4mim][NTf2]/dichloromethane
homogeneous mixture, using catalyst 3 as well as ionic catalyst
4 (Chart 1). However, the yields of CM product still remained
low (≤5%) and no reaction was observed upon recycling of the
catalysts.
Reactions were then attempted with catalyst 3 supported on
silica using [C4mim][NTf2], according to the ‘supported ionic liq-
uid phase’ (SILP) methodology. Catalyst 3 was immobilised using
[C4mim][NTf2] and two types of Davicat silica with varying sur-
face area and pore diameter (Table 4). The silica was used either
untreated (pH 4) or buffered [pH 10 (silica 1) or 7 (silica 2)]. The
activity of the supported catalyst was then examined for the reac-
tion between 1-decene and acrolein (2 mol% catalyst in pentane,
6 h). The cross-metathesis product, undec-2-enal, was obtained in
moderate to good yields (60–72%) in all cases (Table 4). Formation
of self-metathesis product (ca. 2% selectivity) was also observed in
all the reactions. However, attempts to recycle the catalysts were
unsuccessful, with only 1–2% yield obtained on the first recycle
and no product observed on the second recycle. This indicates that
the catalyst deactivated after the first run and/or leached into the
organic phase during the recycling process.
The synthesis of undec-2-enenitrile has also been previously
reported by reaction of cyanoacetic acid with butanal, using pyri-
dine containing a few drops of piperidine as solvent and catalyst
[37]. In this case, the ␤,␥-unsaturated isomers (Z/E = 1.5) form in
addition to ␣,␤-unsaturated nitriles (Z/E = 1.7). No isomerisation
was observed during the CM reactions described, herein. The odour
character of the ␣,␤-unsaturated nitrile obtained by us, was com-
pared with that of a sample containing a mixture of the ␣,␤- and
␤,␥-unsaturated isomers at the same Z/E ratios. Interestingly, two
Examining the results after the first run, the best yield of undec-
2-enal (72%) was obtained using buffered silica 1, which had the
most alkaline pH, whereas the lowest yields were at pH 4 (60–65%
yield), although the differences are small. The surface area and pore

S.M. Rountree et al. / Applied Catalysis A: General 486 (2014) 94–104
103
Table 4
Acknowledgments
Yield of undec-2-enal obtained from the reactions of 1-decene and acrolein with
catalyst 3 supported on silica using [C4mim][NTf2]. All reactions performed at r.t. in
pentane, using 2 mol% of catalyst 3 (6 h per run).
The authors acknowledge support from QUILL. S.M.R. acknowl-
edges support from the Department of Education and Learning in
Northern Ireland and Givaudan SA. The authors thank Dominique
Lelievre and Alain Alchenberger (Givaudan SA) for odour assess-
ments.
Silica
Run
Yield of CM product (%)
Silica 1a
Untreated (pH 4)
1
2
3
1
2
3
60
1
0
72
2
0
Buffered (pH 10)
Appendix A. Supplementary data
Silica 2b
Supplementary data associated with this article can be
found, in the online version, at http://dx.doi.org/10.1016/j.apcata.
2014.08.032.
1
2
3
1
2
3
65
1
0
67
2
0
Untreated (pH 4)
Buffered (pH 7)
References
a
Surface area: 40 m2 g−1; pore diameter: 1000 A.
˚
˚
[1] (a) P. Kraft, J.A. Bajgrowicz, C. Denis, G. Fráter, Angew. Chem. Int. Ed. 39 (2000)
2980–3010;
b
Surface area: 273 m2 g−1; pore diameter: 140 A.
(b) C.S. Sell, The Chemistry of Fragrances: From Perfumer to Consumer, second
ed., The Royal Society of Chemistry, Cambridge, 2006;
(c) P.G.N. Mertens, F. Cuypers, P. Vandezande, X. Ye, F. Verpoort, I.F.J. Vankele-
com, D.E. de Vos, Appl. Catal. A: Gen. 325 (2007) 130–139;
diameter of the silica, however, did not appear to have any signif-
icant influence on the yield of cross-metathesis product. Although
the supported catalyst was not recyclable, it gave much better
yields of undec-2-enal than the IL homogeneous systems (≤5%, see
above). It also gave better results than the analogous reaction in
pentane, which produced 40% yield of CM product using 3.5 mol%
of catalyst 3 (Table 1, entry 1m).
(d) A. Stolle, T. Gallert, C. Schmoger, B. Ondruschka, RSC Adv.
2112–2153.
3 (2013)
[2] (a) H.B. Kagan, O. Riant, Chem. Rev. 92 (1992) 1007–1019;
(b) K.C. Nicolaou, S.A. Snyder, T. Montagnon, G. Vassilikogiannakis, Angew.
Chem. Int. Ed. 41 (2002) 1668–1698;
(c) Y. Hayashi, S. Samanta, H. Gotoh, H. Ishikawa, Angew. Chem. Int. Ed. 47
(2008) 6634–6637;
(d) E. Taarning, R. Madsen, Chem. Eur. J. 14 (2008) 5638–5644.
[3] (a) C. Gimbert, M. Lumbierres, C. Marchi, M. Moreno-Man˜as, R.M. Sebastián, A.
Vallribera, Tetrahedron 61 (2005) 8598–8605;
(b) M. Bell, K. Frisch, K.A. Jørgensen, J. Org. Chem. 71 (2006) 5407–5410;
(c) X. Luo, Z. Zhou, X. Li, X. Liang, J. Ye, RSC Adv. 1 (2011) 698–705;
(d) C.F. Nising, S. Brase, Chem. Soc. Rev. 41 (2012) 988–999;
(e) X. Xin, Q. Zhang, Y. Liang, R. Zhang, D. Dong, Org. Biomol. Chem. 12 (2014)
2427–2435.
4. Conclusions
The Ru-catalysed cross-metathesis reaction of acrolein and 1-
decene has been studied in detail. Overall, the best results (81%
yield of undec-2-enal, 99% selectivity, 88% conversion of 1-decene)
have been obtained in dichloromethane using catalyst 3 at rela-
tively low loading (1.4 mol%) and added gradually over a period
of 3 h (see Table 2, entry 2c). It is interesting to note that, in con-
trast with previous studies, using excess of acrolein in the reactions
did not significantly affect the outcome. Incomplete conversion
of 1-decene is partly attributed to deactivation of the catalyst by
the product undece-2-enal; hence the conversion and yield of CM
product improves by adding fresh catalyst throughout the reac-
tion. Good results (82% yield of undec-2-enal, 99% selectivity, 86%
conversion of 1-decene) were also obtained by delaying the addi-
tion of 1-decene to the catalyst-acrolein mixture (see Table 2, entry
2f), suggesting that catalyst-acrolein species may help to initiate
the process. In addition to dichloromethane, 1,2-dichloroethane
and the non-chlorinated solvent ethyl acetate also proved to be
convenient solvents for the reaction.
[4] (a) Z. Wang, Comprehensive Organic Name Reactions and Reagents: Aldol Reac-
tion and Aldol Condensation, vol. 1, John Wiley & Sons, Hoboken, 2009;
(b) S.I. Lee, J.H. Jang, G.-S. Hwang, D.H. Ryu, J. Org. Chem. 78 (2012) 770–775.
[5] (a) J.-M. Nuzillard, A. Boumendjel, G. Massiot, Tetrahedron Lett. 30 (1989)
3779–3780;
(b) M. Kajjout, M. Smietana, J. Leroy, C. Rolando, Tetrahedron Lett. 54 (2013)
1658–1660.
[6] (a) A. Abad, C. Almela, A. Corma, H. Garcia, Chem. Commun. (2006) 3178–3180;
(b) J. Liu, S. Ma, Org. Lett. 15 (2013) 5150–5153.
[7] V. Cadierno, S.E. García-Garrido, J. Gimeno, Adv. Synth. Catal. 348 (2006)
101–110.
[8] H. Ren, A. Krasovskiy, P. Knochel, Org. Lett. 6 (2004) 4215–4217.
[9] (a) R.H.A. Bosma, G.C.N. Vandenaardweg, J.C. Mol, J. Organomet. Chem. 280
(1985) 115–122;
(b) A.K. Chatterjee, J.P. Morgan, M. Scholl, R.H. Grubbs, J. Am. Chem. Soc. 122
(2000) 3783–3784;
(c) R.L. Pederson, I.M. Fellows, T.A. Ung, H. Ishihara, S.P. Hajela, Adv. Synth.
Catal. 344 (2002) 728–735;
(d) G.S. Forman, R.P. Tooze, J. Organomet. Chem. 690 (2005) 5863–5866;
(e) X. Miao, C. Fischmeister, C. Bruneau, P.H. Dixneuf, ChemSusChem 2 (2009)
542–545;
(f) S.P. Nolan, H. Clavier, Chem. Soc. Rev. 39 (2010) 3305–3316;
(g) X. Miao, C. Fischmeister, C. Bruneau, P.H. Dixneuf, J.-L. Dubois, J.-L. Couturier,
ChemSusChem 5 (2012) 1410–1414.
Based on the optimised conditions for the acrolein:1-decene
reaction,
a series of ␣,␤-unsaturated aldehydes and nitriles
[10] (a) R.H. Grubbs, Tetrahedron 60 (2004) 7117–7140;
(b) D.J. O’Leary, H.E. Blackwell, R.A. Washenfelder, K. Miura, R.H. Grubbs, Tetra-
hedron Lett. 40 (1999) 1091–1094;
(c) H.E. Blackwell, D.J. O’Leary, A.K. Chatterjee, R.A. Washenfelder, D.A. Buss-
mann, R.H. Grubbs, J. Am. Chem. Soc. 122 (1999) 58–71.
[11] (a) A. Rybak, M.A.R. Meier, Green Chem. 9 (2007) 1356–1361;
(b) A. Rybak, M.A.R. Meier, Green Chem. 10 (2008) 1099–1104;
(c) G.B. Djigoué, M.A.R. Meier, Appl. Catal. A: Gen. 368 (2009) 158–162;
(d) H. Bilel, N. Hamdi, F. Zagrouba, C. Fischmeister, C. Bruneau, RSC Adv. 2 (2012)
9584–9589.
of interest to the fragrance industry have been prepared via
cross-metathesis of acrolein, crotonaldehyde, methacrolein, or
acrylonitrile with 1-decene, 1-octene, 1-hexene and 2-allyloxy-6-
methylheptane. Most reactions gave good selectivity of the desired
CM product (≥90%), although yields varied broadly with the nature
of the substrate. The reactions with acrolein gave the best results
overall. In general, the conversion of alkene and yield of CM
product decreased with the presence of bulky substituents in the
substrates (i.e., reactions involving methacrolein and/or 2-allyloxy-
6-methylheptane gave the worst results). It is interesting to note
that isomerisation has not been observed in any of the reactions and
only the ␣,␤-unsaturated products have been detected. In the case
of the undec-2-enenitrile, the smell of the ␣,␤-isomer prepared
herein has shown to be different in character to that of the mixture
of ␣,␤- and ␤,␥-unsaturated isomers obtained by other methods.
[12] W.E. Crowe, D.R. Goldberg, J. Am. Chem. Soc. 117 (1995) 5162–5163.
[13] (a) T.M. Cameron, A.S. Gamble, K.A. Abboud, J.M. Boncella, Chem. Commun.
(2002) 1148–1149;
(b) J.A. Love, J.P. Morgan, T.M. Trnka, R.H. Grubbs, Angew. Chem. Int. Ed. 41
(2002) 4035–4037;
(c) M. Rivard, S. Blechert, Eur. J. Org. Chem. (2003) 2225–2228;
(d) N. Vinokurov, J.R. Garabatos-Perera, Z. Zhao-Karger, M. Wiebcke, H. Buten-
schoen, Organometallics 27 (2008) 1878–1886;
(e) X. Miao, P.H. Dixneuf, C. Fischmeister, C. Bruneau, Green Chem. 13 (2011)
2258–2271;

104
S.M. Rountree et al. / Applied Catalysis A: General 486 (2014) 94–104
(f) X. Miao, R. Malacea, C. Fischmeister, C. Bruneau, P.H. Dixneuf, Green Chem.
13 (2011) 2911–2919;
(e) A. Michrowska, L. Gulajski, Z. Kaczmarska, K. Mennecke, A. Kirschning, K.
Grela, Green Chem. 8 (2006) 685–688;
(g) G.-L. Wu, S.-L. Cao, J. Chen, Z. Chen, Eur. J. Org. Chem. 2012 (2012)
6777–6784;
(f) S.R. Dubberley, P.E. Romero, W.E. Piers, R. McDonald, M. Parvez, Inorg. Chim.
Acta 359 (2006) 2658–2664;
(h) X. Miao, C. Fischmeister, P.H. Dixneuf, C. Bruneau, J.L. Dubois, J.L. Couturier,
Green Chem. 14 (2012) 2179–2183;
(g) C.P. Ferraz, B. Autenrieth, W. Frey, M.R. Buchmeiser, ChemCatChem 6 (2014)
191–198.
(i) C. Bruneau, C. Fischmeister, X. Miao, R. Malacea, P.H. Dixneuf, Eur. J. Lipid
Sci. Technol. 112 (2010) 3–9.
[14] A.K. Chatterjee, T.-L. Choi, D.P. Sanders, R.H. Grubbs, J. Am. Chem. Soc. 125
(2003) 11360–11370.
[22] H.-S. Dang, S. Ellwood, A.-D. Fortineau, C. Furniss, Fragrance compositions and
compounds, Quest International Services B.V., Neth.; Givaudan Nederland Ser-
vices B.V., Patent No. WO2008092678A2 (2008).
[23] S.M. Rountree, M.C. Lagunas, C. Hardacre, P.N. Davey, Appl. Catal. A: Gen. 408
(2011) 54–62.
[15] (a) D. Lee, Nature 471 (2011) 452–453;
(b) S.J. Connon, S. Blechert, Angew. Chem. Int. Ed. 42 (2003) 1900–1923.
[16] (a) B.L. Quigley, R.H. Grubbs, Chem. Sci. 5 (2014) 501–506;
(b) K. Endo, R.H. Grubbs, J. Am. Chem. Soc. 133 (2011) 8525–8527;
(c) B.K. Keitz, K. Endo, P.R. Patel, M.B. Herbert, R.H. Grubbs, J. Am. Chem. Soc.
134 (2012) 693–699;
[24] L. Gulajski, A. Michrowska, J. Naroznik, Z. Kaczmarska, L. Rupnicki, K. Grela,
ChemSusChem 1 (2008) 103–109.
[25] (a) M.S. Sanford, J.A. Love, R.H. Grubbs, J. Am. Chem. Soc. 123 (2001)
6543–6554;
(b) T. Vorfalt, K.-J. Wannowius, H. Plenio, Angew. Chem. Int. Ed. 49 (2010)
5533–5536.
[26] J. Patel, S. Mujcinovic, W.R. Jackson, A.J. Robinson, A.K. Serelis, C. Such, Green
Chem. 8 (2006) 450–454.
(d) R.K.M. Khan, S. Torker, A.H. Hoveyda, J. Am. Chem. Soc. 135 (2013)
10258–10261;
(e) G. Occhipinti, F.R. Hansen, K.W. Tornroos, V.R. Jensen, J. Am. Chem. Soc. 135
(2013) 3331–3334;
(f) L.E. Rosebrugh, M.B. Herbert, V.M. Marx, B.K. Keitz, R.H. Grubbs, J. Am. Chem.
Soc. 135 (2013) 1276–1279;
[27] (a) S.B. Garber, J.S. Kingsbury, B.L. Gray, A.H. Hoveyda, J. Am. Chem. Soc. 122
(2000) 8168–8179;
(b) A.H. Hoveyda, D.G. Gillingham, J.J. Van Veldhuizen, O. Kataoka, S.B. Garber,
J.S. Kingsbury, J.P.A. Harrity, Org. Biomol. Chem. 2 (2004) 8–23;
(c) J.S. Kingsbury, J.P.A. Harrity, P.J. Bonitatebus, A.H. Hoveyda, J. Am. Chem. Soc.
121 (1999) 791–799.
(g) S. Shahane, C. Bruneau, C. Fischmeister, ChemCatChem 5 (2013) 3436–3459.
[17] (a) C.-X. Bai, W.-Z. Zhang, R. He, X.-B. Lu, Z.-Q. Zhang, Tetrahedron Lett. 46
(2005) 7225–7228;
(b) A. Michrowska, R. Bujok, S. Harutyunyan, V. Sashuk, G. Dolgonos, K. Grela,
J. Am. Chem. Soc. 126 (2004) 9318–9325.
[18] (a) N. Ledoux, A. Linden, B. Allaert, H.V. Mierde, F. Verpoort, Adv. Synth. Catal.
349 (2007) 1692–1700;
[28] (a) S. Randl, S. Gessler, H. Wakamatsu, S. Blechert, Synlett (2001) 430–432;
(b) S. Imhof, S. Randl, S. Blechert, Chem. Commun. (2001) 1692–1693;
(c) S. Gessler, S. Randl, S. Blechert, Tetrahedron Lett. 41 (2000) 9973–9976.
[29] C.S. Adjiman, A.J. Clarke, G. Cooper, P.C. Taylor, Chem. Commun. (2008)
2806–2808.
(b) K. Vehlow, S. Maechling, S. Blechert, Organometallics 25 (2005) 25–28;
(c) E.L. Rosen, D.H. Sung, Z. Chen, V.M. Lynch, C.W. Bielawski, Organometallics
29 (2009) 250–256.
[30] (a) D. Banti, J.C. Mol, J. Organomet. Chem. 689 (2004) 3113–3116;
(b) M.B. Dinger, J.C. Mol, Eur. J. Inorg. Chem. (2003) 2827–2833;
(c) M.B. Dinger, J.C. Mol, Organometallics 22 (2003) 1089–1095.
[31] S.-M. Seo, J. Kim, E. Kim, H.-M. Park, Y.-J. Kim, I.-K. Park, J. Agric. Food Chem. 58
(2010) 1823–1827.
[19] (a) D. Fischer, S. Blechert, Adv. Synth. Catal. 347 (2005) 1329–1332;
(b) T.S. Halbach, S. Mix, D. Fischer, S. Maechling, J.O. Krause, C. Sievers, S.
Blechert, O. Nuyken, M.R. Buchmeiser, J. Org. Chem. 70 (2005) 4687–4694;
(c) S. Doherty, P. Goodrich, C. Hardacre, V. Parvulescu, C. Paun, Adv. Synth. Catal.
350 (2008) 295–302;
ˇ
ˇ
[32] J. Pastva, J. Cejka, N. Zilková, O. Mestek, M. Rangus, H. Balcar, J. Mol. Catal. A:
Chem. 378 (2013) 184–192.
(d) B. Autenrieth, W. Frey, M.R. Buchmeiser, Chem. Eur. J. 18 (2012)
14069–14078.
[33] (a) A. Leitgeb, M. Abbas, R.C. Fischer, A. Poater, L. Cavallo, C. Slugovc, Catal. Sci.
Technol. 2 (2012) 1640–1643;
[20] X. Ding, X. Lv, B. Hui, Z. Chen, M. Xiao, B. Guo, W. Tang, Tetrahedron Lett. 47
(2006) 2921–2924.
[21] (a) S. Csihony, C. Fischmeister, C. Bruneau, I.T. Horvath, P.H. Dixneuf, New J.
Chem. 26 (2002) 1667–1670;
(b) M. Abbas, A. Leitgeb, C. Slugovc, Synlett 24 (2013) 1193–1196.
[34] (a) R. Naef, A. Jaquier, Flavour Fragr. J. 21 (2006) 768–771;
(b) W.-Z. Zhang, R. He, R. Zhang, Eur. J. Inorg. Chem. 2007 (2007)
5345–5352.
(b) H. Clavier, N. Audic, M. Mauduit, J.C. Guillemin, Chem. Commun. (2004)
2282–2283;
(c) H. Clavier, N. Audic, J.C. Guillemin, M. Mauduit, J. Organomet. Chem. 690
(2005) 3585–3599;
[35] J.P. Morgan, C. Morrill, R.H. Grubbs, Org. Lett. 4 (2001) 67–70.
[36] A.K. Chatterjee, R.H. Grubbs, Org. Lett. 1 (1999) 1751–1753.
[37] C.K. Lee, J.Y. Shim, Org. Prep. Proc. Int. 22 (1990) 94–96.
[38] N. Kuznik, S. Krompiec, Coord. Chem. Rev. 251 (2007) 222–233.
[39] R.H. Grubbs, S. Chang, Tetrahedron 54 (1998) 4413–4450.
(d) H. Clavier, S.P. Nolan, M. Mauduit, Organometallics 27 (2008) 2287–2292;