Assignment and conformational investigation of asymmetric phenylindenylidene ruthenium complexes bearing N, O-bidentate ligands

Article abstract of DOI:10.1002/mrc.2599

The NMR conformational study of three asymmetric phenylindenylidene ruthenium complexes 4.1-4.3, is presented. Complete ¹H and ¹³C assignments could be obtained for 4.1-4.3 in benzene solution from multiple 2D homonuclear and heteron

Full text of DOI:10.1002/mrc.2599

Research Article
Received: 22 January 2010
Revised: 23 February 2010
Accepted: 1 March 2010
Published online in Wiley Interscience: 29 March 2010
(www.interscience.com) DOI 10.1002/mrc.2599
Assignment and conformational investigation
of asymmetric phenylindenylidene ruthenium
complexes bearing N,O-bidentate ligands
P. M. S. Hendrickx,a∗ R. Drozdzak,b F. Verpoortb and J. C. Martinsa
The NMR conformational study of three asymmetric phenylindenylidene ruthenium complexes 4.1–4.3, is presented. Complete
1H and 13C assignments could be obtained for 4.1–4.3 in benzene solution from multiple 2D homonuclear and heteronuclear
NMR techniques. Our NMR analysis shows that each complex exists as a 55 : 45 mixt◦ure of two rotational isomers in slow
exchange on the NMR chemical shift timescale. They are shown to be related by a 180 flip of the indenylidene ligand along
the Ru CR bond. Both rotational isomers can be discriminated by means of NOEs contacts between the various ligands
coordinating to the Ru. By matching these stereospecific assignments to the chemical shift, a chemical shift based fingerprint
of the isomers that may allow straightforward assignment of future asymmetric phenylindenylidene ruthenium complexes is
c
proposed. Copyright ꢀ 2010 John Wiley & Sons, Ltd.
Supporting information may be found in the online version of this article.
Keywords: NMR; 1H; 13C; ruthenium; indenylidene; complex; olefin; metathesis; chemical exchange; rotational isomerism
Introduction
from analysis of the NMR spectroscopic data. Our results for
4.1–4.3 presented here include complete 1H and 13C resonance
assignments for all complexes and a qualitative description of
chemical exchange processes from exchange spectroscopy (EXSY)
spectra. As each complex was present in two rotameric forms,
additional analysis of NOE data was performed to unambiguously
assign both conformations. Furthermore, chemical shift reference
datawereobtainedfromliteraturethatcouldproveofinterestfora
more straightforward assignment of asymmetric Ru-indenylidene
catalysts.
Arapidlyincreasinginterestinolefinmetathesishasbeenobserved
in recent decades as it constitutes a useful tool for the synthesis
of an impressive range of molecules that required a significantly
longer or tedious route before. Ruthenium catalysts that combine
high activity and good stability have revolutionized the field,
and the development of more efficient and selective ruthenium
catalysts is a key to spread the application of olefin metathesis in
organic chemistry.[1]
Many efforts in the design of ruthenium metathesis catalysts
have focused on improving both activity and latency of the
Grubbs’ catalyst 1, by modifying its neutral donor ligands resulting
in various ruthenium catalysts.[2,3] This study is focused on
catalysts containing a Schiff-base indenylidene ligand, as these
have been found to overcome stability problems common to
other catalysts.[4] Furthermore, Dunbar[5] recently developed a
reversible inhibition/activation protocol using N-donor ligands
and H3PO4 for controlling ring opening metathesis polymerization
(ROMP) reactions, making Schiff-base phenylindenylidene ligands
promising for developing latent catalysts.[6] Finally, these catalysts
can be prepared safely on a large scale using only commercial
reagents[7–9] in contrast to complexes based on the expensive
Grubbs’ precursor 1.
Results and Discussion
Synthesis of Ru-indenylidene complexes bearing salicy-
laldimine ligands 4.1–4.3
Treatment of stoichiometric amounts of a toluene solution
of dichlorido(pyridine)(3-phenyl-1H-inden-1-ylidene)(1,3-dime-
sitylimidazolidin-2-ylidene)ruthenium(II) (2.1),[2,8] which is
robust and versatile precursor, with thallium salts of 2-[(2,6-
dimethyl-phenylimino)-methyl]-4-nitro-phenolate (3.1), 2-[(2,6-
dimethyl-phenylimino)-methyl]-phenolate (3.2) or 2-[(2,6-dime-
thyl-phenylimino)-methyl]-4-methoxy-phenolate (3.3) effected
a
In this study, we describe the NMR characterization of
three novel ruthenium Schiff-base phenylindenylidene catalysts
4.1–4.3 obtained via straightforward synthesis and purification
methods. This was motivated by the fact that in spite of
their importance, little information is available on the solution
behavior of such complexes.[10] In addition, the chemical shift
is very sensitive to both conformational and subtle stereo
electronic effects. Therefore, provided such NMR spectroscopic
data are obtained from a sufficiently large collection of complexes,
correlations with the catalytic activity may ultimately be inferred
∗
Correspondenceto:P. M. S. Hendrickx,DepartmentofOrganicChemistry,Ghent
University, Krijgslaan 281 (S4), B-9000 Ghent, Belgium.
E-mail: Pieter.Hendrickx@UGent.Be
a
DepartmentofOrganicChemistry,GhentUniversity,Krijgslaan281(S4),B-9000
Ghent, Belgium
b
Department of Inorganic and Physical Chemistry, Ghent University, Krijgslaan
281 (S3), B-9000 Ghent, Belgium
c
Magn. Reson. Chem. 2010, 48, 443–449
Copyright ꢀ 2010 John Wiley & Sons, Ltd.

P. M. S. Hendrickx et al.
OTl
R
N
PCy3
Ru
N
N
N
N
Cl
Cl
Cl
Ru
N
Ru
N
- TlCl
Cl
O
Cl
PCy3
R
1
2.1
3.1 ; R = NO2
3.2 ; R = H
3.3 ; R = OMe
4.1 ; R = NO2
4.2 ; R = H
4.3 ; R = OMe
Scheme 1. Left: Grubbs’ catalyst 1; Right: Synthesis of phenylindenylidene-salicylaldimine complexes 4.1–4.3.
All salicylaldimine complexes 4.1–4.3 reveal an impressive
stability to air and heat. They tolerate storage for months as
solids in ambient conditions, including contact with air, without
suffering any degradation or change in isomer ratio. In solution, no
decomposition was observed and the isomer ratio was preserved
over a period of months. Unfortunately, all attempts to obtain
crystals of 4.1–4.3 of sufficient quality or to separate the isomers
failed, further highlighting the importance of conformational
analysis by NMR.
7
1
8
18
10 11
5
3
6
2
17 16
12
13 14
9
4
N
N
15 20
33
21
32
34
33
19
23
31
Cl
24
25
32
Ru
22
Structure validation and conformation of 4.1–4.3
O
30
26
27
The complete 1H and 13C resonance assignment for each
compound is collected in Table 1. It was achieved using a
48 49
46 45
29
N
28
1
combination of 2D 1H-{ H} COSY, TOCSY, ROESY and 1H-{13C}
47
42
35 36
44 43
41
HSQC, HSQC–TOCSY and HMBC spectra, complemented with 1D
1H, 13C and DEPT spectra.[11] In the following, only the spectral
assignment of 4.1a and 4.1b is described in detail, as that of 4.2
and4.3washighlysimilar.ThetermNOEwillbeusedfordescribing
correlations from direct dipole couplings observed in the ROESY
spectra. The presence of the diverse aromatic functionalities in
addition to the use of benzene as solvent was the key in facilitating
the assignment process by dispersing the 1H resonances. For
instance, of the 16 methyl resonances contributed by the SIMes
(1,3-dimesitylimidazolidin-2-ylidene) and salicylaldimine ligand
in both rotational isomers of 4.1, only two were found to be
overlapping.Thearbitrarynumberingscheme,depictedinScheme
2, is used throughout.
40
37
R
39
38
Scheme 2. Arbitrary numbering scheme of compounds 4.1–4.3 used in
the NMR analysis (Table 1).
completeconversiontoSchiffbasecontainingrutheniumindenyli-
dene complexes chlorido(3-phenyl-1H-inden-1-ylidene)(1,3-
dimesitylimidazolidin-2-ylidene)(2-[(2,6-dimethyl-phenylimino-
κN)-methyl]-4-nitro/H/methoxy-phenoxido)ruthenium(II)
The assignment approach initially focused on grouping the 1H
and 13C resonances belonging to the same ligand (SIMes, salicy-
laldimine and indenylidene). Next, these were assembled into the
rotational isomers 4.1a and 4.1b using the 1H resonance intensity
ratio (55 : 45) and inter-ligand NOEs within each rotational isomer
as a guide. The DEPT135 spectrum provides an unambiguous
starting point for the analysis as the four methylene carbon res-
onances can only be contributed by the SIMes bridgehead (C10,
C11) of each isomer. The associated protons are identified from
the HSQC spectrum, which in turn can be linked to their respec-
tive C21 carbene atoms at 216.4 (4.1a) and 216.8 (4.1b) ppm
(4.1–4.3) after 15 min at room temperature, as judged by
1H NMR analysis. Removal of the TlCl by-product was effected by
filtering the reaction mixture through Celit. After evaporation of
the solvent and precipitation with cold pentane, complexes were
isolated as red-brown powders in high yield (98%) (Scheme 1).
1
The initial analysis of the H NMR spectra of 4.1–4.3 reveals
two distinct sets of signals. Most notably, the resonance of H29
(Scheme 2) in the indenylidene moiety that is generally well
recognizable as it gives rise to the only singlet devoid of small
long-range couplings appears twice and with 55 : 45 intensity ratio
ineachcompoundinvestigated.Inthe13CNMRspectrum,lowfield
quaternary resonances associated with the carbene atoms of the
indenylidene(Ru CR)andN-heterocycliccarbene(NHC)moieties,
as well as the imine and C–O of the salicylaldimine ligand appear
in pairs. (Fig. 1a). As will be described in more detail below, further
NMR analysis revealed that compounds 4.1–4.3 are mixtures of
two rotational isomers resulting from a high barrier to the rotation
of the indenylidene group around the ruthenium–carbon bond
(Scheme 3).
3
via JHC correlations in the HMBC. The assignment of the SIMes
ligand is performed by analyzing long-range correlations from
the methyl 1H resonances to the SIMes bridgehead and carbene
carbon atoms. It is noteworthy that such contacts, while weak, can
even be seen to the methyls in para-position. Thus the methyl
signals originating from the SIMes and the Schiff base can be dis-
tinguished. NOEs linking the mesitylene ortho-methyl groups with
the bridgehead methylenes allow discriminating these from the
c
www.interscience.wiley.com/journal/mrc
Copyright ꢀ 2010 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2010, 48, 443–449

Investigation of asymmetric phenylindenylidene ruthenium complexes
Figure 1. (a) 13C NMR spectrum of catalyst 4.1a in C6D6 at 22 ◦C. Only the low field part of the spectrum is shown. (b) Extract from the 1H-{13C} HMBC
spectrum of compound 4.1. Correlations to the C22 carbene atoms in both isomers are indicated. (c) Aliphatic section of the 300 ms 1H off-resonance
ROESY spectrum of 4.1. Cross-peaks arising from exchange phenomena (blue) are annotated above the diagonal, those arising from NOE interactions
(black) are indicated below the diagonal.
H29
H23
H23
Cl
Ru
Cl
Ru
Cl
Cl
Ru
O
O
H
H
H29
4.1a - 4.3a
2.1 - 2.6
4.1b - 4.3b
Scheme 3. Representation of both rotational isomers of complexes 4.1–4.3 (left) and complexes 2.1–2.6 (right) used as chemical shift references.
para methyl groups. Furthermore, it can be deduced from a model
of SIMes that an NOE contact mutually involving one ortho-methyl
group from opposing mesitylene rings can only be expected when
these are not related by the local C2 symmetry, i.e. only NOE con-
tacts between H7–H18 and H8–H19 can be observed. Using this
as a guide, the different ortho-methyl groups of each ring can
be discriminated. The complete assignment of all SIMes methyls
finally allows distinguishing the bridgehead methylene functions
higher rotational energy barrier than for 4.1a (see also Supporting
Information). Two other hindered rotations around N–C1 and
N–C12 could not be observed in the ROESY spectrum as exchange
peaks between Me7–Me8 and Me18–Me19.
The assignment of the Schiff base ortho-methyl-substituted
phenyl can be performed by noticing the three spin TOCSY pattern
involving its aromatic protons (6.7–7.0 ppm) for each isomer. By
analyzing the ROESY spectrum for contacts involving each of the
ortho-methyls, the meta and para positions can be distinguished.
Theidentityoftheattachedcarbonatomswerederivedeitherfrom
the HSQC or HSQC–TOCSY spectra. No exchange cross-peaks are
apparent between the ortho-methyls within the Schiff-base ligand
(Me41–Me42) indicating a considerable rotational energy barrier
aroundtheN–C35bond. Long-rangecorrelationsoriginatingfrom
the ortho-methyl protons to quaternary carbon atoms were used
to identify the meta-positioned carbon atoms (129–131 ppm) and
the ipso-atoms (150.4 ppm, 151.2 ppm). Correlations linking a CH-
type (from DEPT135) carbon atom at 167.2 ppm with two resolved
singlet 1H resonance in the HSQC allow unambiguous assignment
to the isochronous imine carbons (C43) of the Schiff base of
both isomers. Protons on the nitro-containing aromatic ring can
be assigned from ROESY, HMBC and HSQC-TOCSY spectra. NOE
contactscanbefoundbetweenH45andbothmethylgroupsofthe
3
via the JHC correlations mentioned above at the outset of the
analysis. The assignment of the remaining aromatic SIMes protons
1
was tackled using an integrated analysis of the TOCSY, H-{13C}
HSQC and HMBC spectra (Table 1).
IntheROESYspectrum, theC2 relatedmethylgroupsincomplex
4.1 are connected by exchange cross-peaks (Fig. 1c). These are
easily recognized as they show opposite sign with respect to
the abovementioned NOEs (Fig. 1c). Their observation provides
independent validation of the methyl group assignment. Indeed,
such exchange peaks can only arise from a hindered rotation
around the Ru–C21 bond (which coincides with the C2 symmetry
axis of the SIMes ligand) which is slow on the NMR timescale.
From the weak exchange cross-peak intensity, the lifetime of
the rotamers can be estimated to be on the order of seconds.
The weaker intensity of such exchange peaks in 4.1b indicates a
c
Magn. Reson. Chem. 2010, 48, 443–449
Copyright ꢀ 2010 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/mrc

P. M. S. Hendrickx et al.
Table 1. 1H and 13C assignments of the rotational isomers of Ru-indenylidene complexes 4.1–4.3 bearing salicylaldimine ligands
4.1a 4.1b 4.2a 4.2b 4.3a
4.3b
Atom label
1H
13C
1H
13C
1H
13C
1H
13C
1H
13C
1H
13C
1
–
133.5
138.4
128.9
138.9
130.2
139.9
20.2
–
133.0
138.4
128.9
138.9
130.3
139.9
20.2
–
134.0
138.8
129.0
138.2
130.0
139.9
20.2
–
133.4
138.8
128.8
138.2
130.1
139.9
20.2
–
134.0
138.6
129.0
138.2
130.0
139.9
20.2
–
133.6
138.6
128.8
138.2
130.1
139.9
20.2
2
–
–
–
–
–
–
3
6.51
–
6.43
–
6.54
–
6.49
–
6.58
–
6.53
–
4
5
6.87
–
6.83
–
6.93
–
6.89
–
6.93
–
6.90
–
6
Me7
Me8
Me9
10
2.60
2.34
2.04
2.60
2.35
1.98
2.67
2.51
2.09
3.40
3.25
–
2.66
2.51
2.04
3.40
3.25
–
2.67
2.51
2.10
3.39
3.26
–
2.66
2.52
2.04
3.39
3.26
–
18.2
18.4
18.2
18.2
18.2
18.2
20.6
20.6
20.6
20.5
20.6
20.5
3.40/3.20
3.40/3.20
–
50.4
3.40/3.20
3.40/3.20
–
50.4
50.5
50.5
50.6
50.6
11
51.5
51.5
51.3
51.3
51.2
51.2
12
136.2
135.7
129.0
137.1
129.2
137.1
18.4
136.2
136.0
128.5
137.0
129.3
137.0
18.4
136.6
135.9
128.7
136.8
129.0
136.9
18.3
136.5
136.0
128.4
136.8
129.1
136.8
18.3
136.5
136.1
128.9
136.8
128.9
137.0
18.3
136.6
136.2
128.4
136.9
129.1
137.1
18.3
13
–
–
–
–
–
–
14
6.46
–
6.06
–
6.45
–
6.09
–
6.46
–
6.10
–
15
16
5.99
–
6.39
–
6.03
–
6.42
–
6.02
–
6.42
–
17
Me18
Me19
Me20
21
1.97
2.32
1.72
–
2.18
2.19
1.76
–
2.02
2.39
1.77
–
2.22
2.31
1.78
–
2.03
2.41
1.76
–
2.23
2.32
1.78
–
18.0
18.4
17.8
18.2
17.8
18.2
20.6
20.6
20.7
20.7
20.6
20.6
216.4
296.8
133.3
136.6
145.1
115.7
127.7
128.7
129.9
139.3
137.8
126.2
128.9
127.9
150.4
130.1
127.6
125.5
126.9
131.7
19.4
216.8
295.3
139.4
136.5
143.2
116.3
127.6
128.2
125.5
139.9
135.8
126.3
129.0
127.7
151.2
129.3
127.8
125.7
126.9
132.9
19.5
217.7
294.0
133.5
136.9
145.9
115.5
127.1
128.4
129.8
140.0
136.5
126.1
128.7
127.2
151.6
130.6
127.8
125.0
126.8
132.1
19.3
218.1
291.7
139.0
137.2
143.3
116.0
126.8
128.3
125.8
140.0
135.2
126.3
128.9
127.2
152.0
129.9
127.9
125.1
126.9
132.5
19.4
217.9
293.3
133.4
137.3
146.3
115.6
127.1
128.4
129.9
139.9
136.4
126.1
128.7
127.1
152.0
130.8
127.7
125.0
126.8
132.3
19.2
218.2
291.2
138.9
137.3
143.4
116.0
126.8
128.2
125.7
139.9
135.4
126.4
128.9
127.2
152.7
129.9
127.9
125.1
126.9
132.7
19.3
22
–
–
–
–
–
–
23
6.51
–
7.71
–
6.62
–
7.90
–
6.61
–
7.95
–
24
25
–
–
–
–
–
–
26
6.97
7.02
6.92
8.87
–
7.01
7.01
7.01
7.94
–
7.02
7.04
6.98
9.10
–
7.03
6.99
6.98
8.15
–
7.03
7.04
7.00
9.15
–
7.03
6.96
6.98
8.11
–
27
28
29
30
31
–
–
–
–
–
–
32
7.66
7.11
7.19
–
7.69
7.15
7.26
–
7.63
7.08
7.18
–
7.77
7.15
7.26
–
7.64
7.06
7.16
–
7.77
7.16
7.25
–
33
34
35
36
–
–
–
–
–
–
37
6.71
6.90
6.98
–
6.71
6.87
6.97
–
6.74
6.89
7.00
–
6.74
6.86
7.00
–
6.79
6.92
7.04
–
6.79
6.89
7.03
–
38
39
40
Me41
Me42
43
1.82
1.09
6.78
–
1.84
1.33
6.76
–
1.83
1.23
7.32
–
1.87
1.41
7.36
–
1.85
1.27
7.23
–
1.88
1.46
7.25
–
16.8
17.3
16.9
17.2
17.1
17.3
167.2
116.9
135.0
135.6
–
167.2
116.8
135.0
135.6
–
167.2
118.6
137.0
122.7
–
167.3
118.0
137.1
122.7
–
166.2
116.6
115.0
148.4
54.9
166.1
116.0
115.1
148.4
54.9
44
45
7.56
–
7.54
–
6.60
7.19
–
6.60
7.19
–
5.88
–
5.87
–
46
Me46
47
–
–
3.22
7.15
7.12
–
3.25
7.18
7.14
–
8.00
6.71
–
129.0
122.7
173.7
8.07
6.77
–
128.8
122.6
173.4
7.20
6.35
–
133.9
113.4
170.0
7.23
6.38
–
133.9
113.4
169.8
125.8
123.8
166.0
125.6
123.7
165.6
48
49
Arbitrary numbering scheme is depicted in Scheme 2. Chemical shifts are quoted in parts per million with respect to residual C6D5H as secondary
internal reference.
c
www.interscience.wiley.com/journal/mrc
Copyright ꢀ 2010 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2010, 48, 443–449

Investigation of asymmetric phenylindenylidene ruthenium complexes
and close to the nitro-containing salicylaldimine phenyl moiety
experiences similar magnetic environments in both isomers. On
theotherhand, largedifferencesinchemicalshiftsareobservedfor
the other SIMes mesitylene ring, the indenylidene and the ortho-
methyl phenyl of the Schiff base. These observations suggest the
phenylindenylidene ligand is the source of the isomerism.
N
N
N
N
Cl
Cl
Ru
Ru
Further proof is obtained by thorough analysis of the NOE
connectivities.For4.1aand4.1btogether,156resolvedNOEcross-
peaks were assigned resulting in 78 nontrivial spatial connections
of which 39 are of inter-ligand nature. For both isomers, the
presence of Me8-H43, Me8-H45, Me8-H47, Me9-H43 and Me9-H45
NOE contacts in addition to the absence of Me7-H43, Me7-H45
and Me7-H47 contacts indicates that the relative position of the
SIMes phenyl ring to the Cl–Ru–O plane is such that Me7 is
positioned over the chlorine ligand, while Me8 is located over
the Schiff-base ligand (see Supporting Information for a detailed
list). NOE contacts linking Me18 and Me19 to the indenylidene
ligand can be used for analysis of the indenylidene conformation.
In 4.1a, intense H23-Me19, H32-Me19, H32-H14 and H29-Me18
NOE contacts are observed. This proves the indenylidene ligand
is orientated such that H23 is positioned under Me19, and H29
under Me18 as schematically shown in Scheme 4. A different set
of NOE contacts is observed for the second isomer, including H23-
Me18, H32-Me18, H32-H16 and H29-Me19. These correlations are
clearly related to isomer 4.1b obtained by a 180◦ rotation of the
indenylidene ligand with respect to 4.1a.
Although a thorough investigation of NOE correlation data
provides the information necessary to distinguish between both
the rotameric isomers present in solution, such an analysis is quite
tedious. To facilitate the assignment process of future asymmetric
ruthenium phenylindenylidene catalysts, the potential to use 1H
and 13C chemical shifts alone to assign the different rotamers
was investigated. From Table 1, only H23, H29, C23 and C29 show
significant chemical shift differences between both isomers to suit
this purpose. This is even more apparent when these chemical
shifts are presented in an HSQC plot (Fig. 2) and compared to
chemical shifts of six symmetric reference complexes 2.1–2.6
obtained from the literature.[10] In these complexes, both H23 and
H29 oppose a chlorine ligand (Scheme 3), whereas in 4.1a–4.3a
only H29 opposes the chlorine and H23, the Schiff-base ligand.
The inverse holds for 4.1b–4.3b. This conformational difference is
clearlyreflectedinFig. 2,astheH23–C23resonancesof4.1b–4.3b
and the H29–C29 resonances of 4.1a–4.3a coincide with those
of 2.1–2.6. Thus, the analysis of these resonances allows for a
straightforward assignment of the different rotameric isomers.
Furthermore, no exchange cross-peaks could be identified
between indenylidene resonances of the different rotamers a
and b using ROESY mixing times up to 400 ms. Thus, the lifetime of
the rotational isomers must exceed the second timescale. Similar
arguments regarding the conformation and the identity of the
rotational isomerism hold for the complexes 4.2 and 4.3.
O
O
N
N
O2N
O2N
4.1a
4.1b
Scheme 4. Schematic representations of the most important inter-ligand
spatial interactions.
Schiff-base ligand as well as with the imine proton. Furthermore,
H45 can be correlated with H47, which in turn can be linked
to H48 via strong COSY/TOCSY cross-peaks, thus validating the
assignment. For each isomer, long-range correlations from these
protons were found to quaternary carbon atoms (∼173.4 ppm,
∼135.6 ppm, ∼116.8 ppm) but no unambiguous assignment
could be made based on the NMR data alone. Using chemical shift
calculations (ChemDraw 7.0, Cambridge Soft), these resonances
could be assigned to the ortho-, meta- and ipso-carbon atoms with
respect to the imine substituent, respectively.
For the analysis of the indenylidene ligand, the quaternary
carbene atom of each isomer, located at 296.8 (4.1a) and 295.3
(4.1b) ppm, respectively, was chosen as a starting point given
the conspicuous 13C frequency. Each carbene features two long-
range correlations (Fig. 1b) which can only arise from protons H23
(6.51 ppm, 7.71 ppm) and H29 (8.87 ppm, 7.94 ppm). As all four
resonances are well resolved, these could readily be distinguished
by their multiplet fine structure. Further assignments of the
H26–H29 spin system can be easily performed by analyzing the
HSQC–TOCSY and ROESY spectra. Long-range correlations were
then used to assign the quaternary carbon atoms in the main
aromatic ring system. The attached phenyl ring was characterized
using ROESY spectroscopy to show NOE contacts between H26
and H32 for both isomers. Further analysis of the phenyl spin
system was again performed by HSQC–TOCSY. As both ortho- and
meta-protons are isochronous, it can be concluded that the ring
flip of the phenyl ring is fast on the NMR timescale.
To complete the assignment, the ligands belonging to the same
isomer of 4.1 were identified from inter-ligand NOE contacts.
As all methyl groups are resolved, possible contacts of these
methyls, involving two of the ligands, with the H23 and H29 of the
indenylidene ligand were investigated. In each isomer, numerous
contacts can be found that allow placing one mesityl moiety of
the SIMes ligand over the indenylidene ligand (Scheme 4). This
necessarily places the other mesityl moiety in proximity to the
nitro-substituted phenyl ring of the salicylaldimine ligand. This is
validated by clear NOEs involving Me8 of the SIMes and H47 and
Conclusions
1
H48 of the salicylaldimine. The complete H assignment of 4.1a
In this work, new second-generation Ru-indenylidene complexes
4.1, 4.2 and 4.3 have been synthesized and characterized. They
possess air and moisture stability allowing a shelf lifetime of
months without any decomposition. Using NMR, the presence of
a 55 : 45 mixture of two rotational isomers a and b, related via
a 180◦ flip along the Ru C indenylidene bond and separated
by a high activation barrier could be established. Rotational
and4.1btogetherwithasubsequentmorein-depthanalysisofthe
respective inter-ligand NOE contacts shows that both isomers are
indeed related by a 180◦ flip of the indenylidene ligand around the
Ru C bond (Scheme 3). First, the analysis of 1H and 13C chemical
shift differences between both isomers shows that the SIMes
mesitylene ring orientated away from the indenylidene ligand
c
Magn. Reson. Chem. 2010, 48, 443–449
Copyright ꢀ 2010 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/mrc

P. M. S. Hendrickx et al.
Figure 2. Schematic representation of the H23 and H29 HSQC resonances of 4.1–4.3 in comparison with those of 2.1–2.6.
exchange kinetics on a second timescale around the local C2
axis of the Ru–C21 bond as well as a fast exchange process due to
rotation around the C24–C31 bond could also be established. No
exchange peaks corresponding to rotations around N–C1, N–C12
or N–C35 could be observed (see also Supporting Information).
The unambiguous full resonance assignment and NOE based
structural information of each isomer could provide an avenue to
futurecalculationsoftheelectronicpropertiesofthesecomplexes.
Additionally, a straightforward method based on chemical shift
references was found proficient for discriminating between the
rotamerspresentandmayprovideasuitableandmoreconvenient
avenue for the assignment of asymmetric indenylidene ruthenium
complexes in the future.
temperature. Filtration of the solid under a nitrogen or argon
atmosphere yielded the thallium salts 3.1–3.3 in quantitative
amounts. These salts were used in the next step without further
purification.
Preparation of Schiff-base-substituted Ru complexes 4.1–4.3
To a solution of Ru complex 2.1 in THF (5 ml) was added a solution
of thallium salts 3.1–3.3. The reaction mixture was stirred at room
temperature for 1 h. After evaporation of the solvent, the residue
was dissolved in a minimal amount of toluene and cooled to 0 ◦C.
Thethalliumchloride(theby-productofthereaction)wasremoved
via filtration. The filter residue was then washed with cold toluene
andthefiltratewasevaporated.Thesolidresiduewasrecrystallized
from CH2Cl2/pentane (−70 ◦C) to give the Schiff-base-substituted
Ru complexes in good yields (>98%) as brown-orange or red
solids.
Experimental
General remarks
NMR-investigation of catalyst 4.1–4.3
All synthetic manipulations were performed under argon (oxy-
gen free) using the Schlenk technique. Argon was dried
by passage through drierite. Tetrahydrofuran (THF), toluene,
dichloromethane, benzene-d6 and chloroform-d, dried by
Samplesarepreparedusing ascrew-cappedNMRtube, underinert
Aratmosphereusingasolutionofapproximately35 mM of4.1–4.3
in benzene-d6. All experiments were performed on a Bruker DRX
spectrometer equipped with a 5-mm TXI-Z probe and operating
at a 1H and 13C frequency of 500.13 and 125.77 MHz, respectively.
The temperature used was 298 K throughout. All 1D and 2D
experiments were performed using pulse sequences available
from the standard Bruker library.[11] Gradient enhanced sequences
were used for the COSY and heteronuclear 2D experiments. TOCSY
mixingtimesinthe40–100 msrangewereused.TheHMBCspectra
standard methods, were degassed by
a standard three
freeze–pump–thaw cycles. Catalyst 2.1 was prepared accord-
ing to the reported methods.[7,10] 1H NMR and 13C spectra were
recorded on a Bruker Avance 500 MHz spectrometer. Chemical
shifts (δ) are given in parts per million and were referenced accord-
ing to the C6D5H resonance. Kinetic experiments were conducted
on a Varian Unity 300 MHz NMR spectrometer. Elemental analyses
were performed at Chevron Technology, Ghent.
n
were optimized for 3 and 8 Hz long-range JCH coupling. Off-
resonance ROESY spectra with a tilt angle of 60◦ and 200–400 ms
mixing time were recorded[12] to distinguish NOE and exchange
correlations. Typically, 2D spectra consisted of 512 t1 increments
of 8–32 scans each, sampled with 4 K points. Prior to Fourier
transformation, the data were zero-filled along t1 and suitably
apodized using a squared cosine bell in both dimensions, except
Preparation of Schiff-base Ligands
The condensation of an appropriate derivative of salicylaldehydes
with 2,6-dimethylamine was carried out with stirring in isopropyl
alcohol at 80 ◦C for 24 h. Upon cooling to 0 ◦C, a yellow solid
precipitated from the reaction mixture. The solid was filtered,
washed with cold ethyl alcohol, and then dried in vacuo to afford
the desired salicylaldimine ligand in excellent yields.
1
for the COSY and HMBC where a squared sine bell was used. H
chemical shifts were obtained using 1D proton NMR if resolved
or deduced from 2D NMR cross-peaks otherwise. 13C chemical
shifts were obtained from the 1D 13C spectrum. All processing and
analysis were performed using Bruker’s TopSpin 1.3 software suite.
Preparation of thallium salts 3.1–3.3
A solution of thallium ethoxide in benzene or THF (5 ml) was
added drop-wise at room temperature to a solution of Schiff bases
in benzene, toluene or THF (10 ml). (Note: the solution of thallium
ethoxideinTHFwasfilteredthroughaplugofglasswooltoremove
any impurities.) Immediately after the addition, a pale yellow solid
is formed and the reaction mixture was stirred for 2 h at room
Acknowledgements
P. M. S. H., J. C. M. and F. V. thank the FWO for financial support and
several NMR equipment grants (G.0036.00N, G.0365.03). R. D and
F. V. gratefully acknowledge the Ghent University Research Fund
(BOF) for the financial support.
c
www.interscience.wiley.com/journal/mrc
Copyright ꢀ 2010 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2010, 48, 443–449

Investigation of asymmetric phenylindenylidene ruthenium complexes
[4] A. Lozano Vila, S. Monseart, R. Drozdzak, S. Wolowiec, F. Verpoort,
Adv. Syn. Catal. 2009, 351, 2689.
[5] M. A. Dunbar,S. L. Balof,L. J. LeBeaud,B. Yu,A. B. Lowe,E. J. Valente,
H-J. Schanz, Chem. Eur. J. 2009, 15, 12435.
[6] N. Ledoux, B. Allaert, D. Schaubroeck, S. Monsaert, R. Drozdzak,
P. Van Der Voort, F. Verpoort, J. Organomet. Chem. 2006, 691, 5482.
[7] S. Monsaert, R. Drozdzak, V. Dragutan, I. Dragutan, F. Verpoort, Eur.
J. Inorg. Chem. 2008, 3, 432.
Supporting information
Supporting information may be found in the online version of this
article.
[8] A. Furstner, O. Guth, A. Duffels, G. Seidel, M. Liebl, B. Gabor,
References
R. Mynott, Chem. Eur. J. 2001, 7, 4811.
[9] (a)
H.-J. Schanz,
L. Jafarpour,
E. D. Stevens,
S. P. Nolan,
[1] (a) R. H. Grubbs, HandbookofMetathesis, vol. 1–3, Wiley-VCH Verlag
GmbH & Co. KGaA: Weinheim, 2003; (b) R. H. Grubbs, Angew. Chem.
Int. Ed. 2006, 45, 3760; (c) T. M. Trnka, R. H. Grubbs, Acc. Chem. Res.
2001, 34, 18; (d) A. Fu¨rstner, Angew. Chem. Int. Ed. 2000, 39, 3012;
(e) R. R. Schrock, Tetrahedon 1999, 55, 8141; (f) M. R. Buchmeiser,
Chem. Rev. 2000, 100, 1565; (g) B. C. G. Soderberg, Coord. Chem. Rev.
2006, 250, 2411.
Organometallics 1999, 18, 5187; (b) L. Jafarpour, H. -J. Schanz,
E. D. Stevens, S. P. Nolan, Organometallics 1999, 18, 5416.
[10] S. Monsaert, E. De Canck, R. Drozdzak, P. Van Der Voort, F. Verpoort,
J. C. Martins, P. M. S. Hendrickx, Eur. J. Org.Chem. 2009, 655.
[11] S. Berger,S. Braun,200andMoreNMRExperiments,Wiley-VCHVerlag
GmbH & Co. KGaA: Weinheim, 2004, and references cited herein.
[12] H. Desvaux, M. Goldman, J. Magn. Reson. 1996, 110, 198.
[2] S. Monseart, A. Lozano Vila, R. Drozdzak, P. Van Der Voort,
F. Verpoort, Chem. Soc. Rev. 2009, 38, 3360.
[3] C. E. Diesendruck, E. Tzur, N. G. Lemcoff, Eur. J. Inorg. Chem. 2009,
4185.
c
Magn. Reson. Chem. 2010, 48, 443–449
Copyright ꢀ 2010 John Wiley & Sons, Ltd.
www.interscience.wiley.com/journal/mrc