Methyl and phenyl substituent effects on the catalytic behavior of NHC ruthenium complexes

Article abstract of DOI:10.1039/c6ra20608e

New second-generation ruthenium benzylidene and isopropoxybenzylidene catalysts bearing N-heterocyclic carbene (NHC) ligands with o-biphenyl groups at the N-atoms and syn methyl or phenyl groups on the backbone were obtained and their catalytic behaviors were compared to those of analogous N-o-tolyl catalysts in standard ring-closing metathesis (RCM) reactions. A pronounced difference in catalyst efficiency was observed depending on the nature of ortho-N-aryl substituents (methyl or phenyl). Notably, very impressive catalytic performances were exhibited by N-o-biphenyl complexes with a syn dimethyl backbone in the formation of di- and trisubstituted cycloalkenes. To rationalize catalytic results, methyl and phenyl substituent effects on the steric and electronic properties of NHC ligands were assessed through experimental and theoretical investigations involving ruthenium complexes as well as newly developed rhodium derivatives. Despite the different electron donor capacities of the examined carbenes, the steric differences shown by N-o-biphenyl and N-o-tolyl NHCs, although subtle, were found to be the key factor in addressing catalyst behavior.

Full text of DOI:10.1039/c6ra20608e

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Methyl and phenyl substituent effects on the
catalytic behavior of NHC ruthenium complexes†
Cite this: RSC Adv., 2016, 6, 95793
Alessandra Perfetto,a Valerio Bertolasi,b Chiara Costabile,a Veronica Paradiso,a
a
a
a
*
Tonino Caruso, Pasquale Longo and Fabia Grisi
New second-generation ruthenium benzylidene and isopropoxybenzylidene catalysts bearing
N-heterocyclic carbene (NHC) ligands with o-biphenyl groups at the N-atoms and syn methyl or phenyl
groups on the backbone were obtained and their catalytic behaviors were compared to those of
analogous N-o-tolyl catalysts in standard ring-closing metathesis (RCM) reactions. A pronounced
difference in catalyst efficiency was observed depending on the nature of ortho-N-aryl substituents
(methyl or phenyl). Notably, very impressive catalytic performances were exhibited by N-o-biphenyl
complexes with a syn dimethyl backbone in the formation of di- and trisubstituted cycloalkenes. To
rationalize catalytic results, methyl and phenyl substituent effects on the steric and electronic properties
of NHC ligands were assessed through experimental and theoretical investigations involving ruthenium
complexes as well as newly developed rhodium derivatives. Despite the different electron donor
capacities of the examined carbenes, the steric differences shown by N-o-biphenyl and N-o-tolyl NHCs,
although subtle, were found to be the key factor in addressing catalyst behavior.
Received 15th August 2016
Accepted 1st October 2016
DOI: 10.1039/c6ra20608e
www.rsc.org/advances
the presence of bulky substituents on the backbone was found
to improve catalyst stability3c limiting decomposition pathways
due to C–H bond activation of N-aryl rings.4 Furthermore, the
symmetry of the NHC backbone associated with mono-ortho-
substituted N-aryl groups has been recognized as a key param-
eter for successful ring-closing metathesis (RCM) reactions and,
more interestingly, complexes with syn phenyl substituents on
the backbone and N-o-tolyl rings has been identied as privi-
leged catalysts in the most representative olen metathesis
Introduction
N-heterocyclic carbenes (NHCs) are nowadays ubiquitous
ligands in organometallic and coordination chemistry. Their
unique steric and electronic features have made possible the
development of effective metal catalysts for various applica-
tions, the most successful examples being in the eld of
ruthenium-catalyzed olen metathesis.1 Ruthenium complexes
bearing NHC ligands, commonly referred to as second-
generation catalysts, have demonstrated their superiority over
rst generation catalysts containing classical phosphine ligands
exhibiting higher thermal stability, activity and selectivity
(Chart 1).2
One of the major attractiveness of this class of complexes lies
in the possibility of ne-tuning their catalytic properties by
modifying the NHC stereoelectronics. Variation of the degree of
backbone and/or N-aryl substitution of the NHC ligand has led
to ruthenium catalysts especially competent for metathesis
transformations involving hindered substrates.3 In particular,
a
`
Dipartimento di Chimica e Biologia “Adolfo Zambelli”, Universita di Salerno, Via
Giovanni Paolo II 132, I-84084 Fisciano, Salerno, Italy. E-mail: fgrisi@unisa.it
bDipartimento di Scienze Chimiche e Farmaceutiche and Centro di Strutturistica
`
Diffrattometrica, Universita di Ferrara, Via L. Borsari 46, I-44121 Ferrara, Italy
† Electronic supplementary information (ESI) available: Experimental procedures,
gures giving NMR spectra of the new complexes 3, 4,
7 and 8, cyclic
voltammograms of 1–8, tables and CIF les giving experimental details for
complexes 7, 21–24, and computational details. CCDC 1483950–1483954. For
ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/c6ra20608e
Chart 1 Grubbs' and Hoveyda–Grubbs' catalysts.
RSC Adv., 2016, 6, 95793–95804 | 95793
This journal is © The Royal Society of Chemistry 2016

View Article Online
RSC Advances
Paper
Chart 2 syn-NHC backbone substituted Ru catalysts.
processes (1 and 5, Chart 2).3g,h Indeed, the steric pressure of catalysts 4 and 8, Pd-catalyzed diarylation of commercially
encumbered syn phenyl groups on the NHC backbone has been available diamine D with 2-bromobiphenyl allowed to obtain E,
found to be an essential requirement to froze the most suitable that was then converted in the NHC ligand precursor F by
NHC conformation (that is the one with syn oriented N-tolyl reacting with triethyl orthoformate in the presence of ammo-
rings) for efficiently accomplishing metathesis reactions (2 and nium tetrauoroborate.6 In situ deprotonation of F by potas-
6, Chart 2).3g,h
sium hexauoro-tert-butoxide [(CF3)2CH3COK] produced the
To gain a better understanding of the role played by the desired free NHC that was treated with GI to afford complex 4
nature of syn NHC backbone substituents and of mono-ortho- (66% yield). Phosphine-containing complex 4 was quantitatively
substituted N-aryl groups in the achievement of highly per- transformed in the corresponding phosphine-free catalyst 8 by
forming catalysts, we thought to further investigate this type of a cross-metathesis reaction with 2-isopropoxystyrene in the
substitution pattern employing different combinations of presence of CuCl as a phosphine scavenger. All the complexes
methyl and phenyl groups on the NHC ligand and evaluating were obtained as air- and moisture-stable solids aer ash
their impact on the catalytic properties of the resulting cata- column chromatography and were fully characterized by 1D and
lysts. Therefore, in this study, we present the synthesis and 2D NMR spectroscopy.7b
characterization of new second generation Grubbs-type
As previously observed for analogous complexes,3e–g the
complexes 3, 4 and Hoveyda–Grubbs-type complexes 7, 8 preferred NHC conformation for phosphine-containing
(Chart 2), bearing methyl or phenyl substituents at the ring complexes 3 and 4 was assigned to be that with syn-oriented
backbone associated with o-biphenyl groups at the nitrogen N-aryl groups pointing on the opposite sides of the syn NHC
atoms, and we compare their behaviors in standard ring-closing backbone substituents. For both complexes, variable tempera-
metathesis reactions to those of already known complexes 1, 2 ture (VT) 1H and 31P NMR experiments revealed the presence of
and 5, 6 having the same NHC backbone substitution and o-
tolyl groups at the N atoms.3e–h The steric and electronic prop-
erties of the examined NHC ligands were assessed using both
ruthenium and rhodium complexes and their possible correla-
tions to observed catalyst reactivities are discussed, providing
further advances in the understanding of which NHC charac-
teristics are important for the rational design of RCM catalysts.
Results and discussion
Synthesis and characterization of Ru-based catalysts
Ru-catalysts 1, 2, 5 and 6 were prepared as earlier described.3e–g
The synthesis of new complexes 3, 4, 7 and 8 were carried out as
shown in Scheme 1. With regard to the synthesis of catalysts 3
and 7, bearing syn methyl substituents on the NHC backbone,
the chosen synthetic route involved the preparation of diamine
A,5 that was subsequently coupled with 2-bromobiphenyl by Pd-
catalyzed reaction to give diarylated diamine B. Dihy-
droimidazolium salt C was obtained by treatment of B with HCl
and then condensation with triethyl orthoformate. The corre-
sponding free carbene generated in situ by treatment of C with
hexamethyldisilazide (KHMDS) was reacted with GI to afford
monophosphine complex 3 (33% yield)7a or, alternatively, with
HGI to give phosphine-free complex 7 (72% yield). As for Scheme 1 Synthesis of new ruthenium complexes 3, 4, 7 and 8.
95794 | RSC Adv., 2016, 6, 95793–95804
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
two rotational isomers at low temperature, resulting from malonate (9, 13, 17) and tosylamine (11, 15, 19) derivatives with
rotation around the benzylidene C–Ru bond or rotation about increasing steric hindrance, and compared to those of previ-
the Ru–NHC bond (see ESI†). Solution-state structures of ously reported complexes 1, 2, 5 and 6.3e–h All the reactions were
phosphine-free complexes 7 and 8, determined via NMR anal- conducted at 30 ꢀC in CD2Cl2 with phosphine-containing
ysis, revealed the presence of a single isomer, corresponding to complexes 1–4, and at 60 ꢀC in C6D6 with slow-initiating
the species with the NHC ligand locked in the same confor- phosphine-free catalysts 5–8. The conversion of each substrate
1
mation observed for 3 and 4. The solid-state structure of 7 was to product was monitored over time by H NMR and the cor-
unambiguously established by X-ray diffraction (Fig. 1). In this responding kinetic proles are sketched in Fig. 2–4. Tables S1–
complex, the Ru center is penta-coordinated and adopts a dis- S3 in the ESI† provide additional details on the examined RCM
torted square pyramidal coordination geometry. The Cl atoms reactions.
are trans oriented in the basal plane and the carbene C1 atom is
As clearly shown by the kinetic proles of the ring-closures of
in trans position with respect to the O1 oxygen of 2-iPrO diethyl diallylmalonate (9) and diallyltosylamine (11) promoted
substituent at the benzylidene ligand, which is almost coplanar by phosphine-containing complexes 1–4 (Fig. 2A and C), cata-
with the NHC ring, being rotated by only 18.37(5)ꢀ.
lysts 3 and 4, characterized by o-biphenyl N-substituents, dis-
This compound crystallizes in the centro-symmetric P21/n played signicantly higher activities than 1 and 2 bearing o-tolyl
space group with the NHC methyl groups in cis position with N-groups, rivaling those of the most efficient systems known up
respect to C2–C3 bond. Accordingly the crystal contains to now.3f,9 Catalyst 3 with methyl groups on the backbone fur-
a racemic mixture of both the enantiomers having opposite nished quantitatively cycloalkenes 10 in 5 min and 12 in 3 min,
congurations (RS or SR) at the C2 and C3 asymmetric carbon performing slightly better than 4, that needed 8 and 6 minutes
atoms. The phenyl rings C14/C19 bonded to N1 and C26/C31 to give the corresponding cyclic products in 95% and >98%
bonded to N2 are rotated by 53.68(6)ꢀ and 74.48(5)ꢀ, respec- conversion, respectively. Catalyst 2 was found to be slightly
tively, with respect to NHC ring. These conformations are more active than 1, nearly completing cyclization of 9 (95%
mainly determined by short intramolecular interactions: H19/ conversion) and 11 (>98% conversion) in 18 and 22 minutes,
˚
˚
Ru1 ¼ 2.71 A and H4/C26 ¼ 2.49 A. The other structural respectively.
parameters are very similar with respect to those observed in
Differences in reactivity are much more difficult to notice for
many other similar Hoveyda–Grubbs' second generation cata- the same RCM reactions carried out in the presence of the
lysts that differ only in the NHC substituents. Unfortunately, all corresponding phosphine-free catalysts 5–8 10 (Fig. 2B and D),
the efforts of growing single crystals of 3, 4 and 8 suitable for X- very likely as a consequence of the higher temperature
ray crystallographic analyses were unproductive.
employed to improve performances of this family of catalysts.3f
However, also in this case the catalyst presenting methyls on the
NHC backbone and N-o-biphenyl groups (7) outperformed the
RCM catalytic behavior
The catalytic behaviors of new complexes 3, 4, 7 and 8 were
investigated in standard RCM transformations, involving
Fig. 1 ORTEP8 view of 7 showing the thermal ellipsoids at 30%
probability level. The H atoms, excluded those linked to C2 and C3
carbons of the NHC ring, have been omitted for sake of clarity.
Fig. 2 Kinetic plots of the RCM of 9 and 11.
RSC Adv., 2016, 6, 95793–95804 | 95795
This journal is © The Royal Society of Chemistry 2016

View Article Online
RSC Advances
Paper
Fig. 3 Kinetic plots of the RCM of 13 and 15.
other ones providing quantitatively the desired cyclic products complexes 1 and 2. As depicted in Fig. 4A, in the RCM of mal-
10 and 12 in 2 and 4 min, respectively. onate derivative 17 complexes 1–4 exhibited nearly identical
The catalytic behavior of complexes 1–8 was then investigated proles at the beginning of the reaction. Their behaviors
in the RCM of more encumbered diolens 13 and 15 (Fig. 3). As differentiated aer roughly the rst ve minutes, when for
for phosphine-containing catalysts 1–4, once again catalyst 3 catalysts 3 and 4 a strong decrease of reaction rate was observed,
exhibited the highest efficiency in forming cycloolens 14 and 16 and plateau conversion values of 61% and 45% were respec-
(>99% conversion in 9 and 4 min, respectively), emerging as the tively reached. In the easier RCM of tosyl derivative 19, catalysts
best performing system in this class of reaction up to now.3f,9 On 3 and 4 gave better performances than in the previous cycliza-
the other hand, catalyst 4, despite the high initiation rate, gave tion, leading to 82% and 61% conversion respectively (Fig. 4C).
nal conversions slightly slower than those of catalysts 1 and 2. Once again they proved to be less efficient than catalyst 2, and to
The minor efficiency observed for 4 is likely due to a higher a lesser extent than catalyst 1.
decomposition rate of the catalyst, as suggested by the analysis of
the slope of the corresponding curves in Fig. 3A and C.
Notably, for what concerns phosphine-free catalysts 5–8,
differently from the RCM of 9 and 11 where all the catalysts gave
The ring-closures of 13 and 15 promoted by phosphine-free very high reaction rates as well as fast initiation, in the RCM of
Ru-catalysts 5–8 displayed the same reactivity trend observed 13 N-o-tolyl catalysts 5 and 6 retained fast initiation rates,
in the previous RCM, with only marginal deviations. Again, whereas 7 and 8 initiated slowly (Fig. 4B). Therefore, while 5 and
catalyst 7, possessing N-o-biphenyl groups and syn dimethyl 6 converted almost quantitatively 17 within 60 min, 7 and 8
backbone, turned out to be the most active catalyst in both the reached 86% and 78% conversion, respectively, completing
transformations.
cyclization within 7 h. The slowest initiation makes 8 the less
In the challenging RCM of sterically encumbered diolens performing among the herein reported catalysts in the RCM of
17 and 19 (Fig. 4), more marked reactivity differences among hindered olens. The reactivity trend observed in the RCM of
catalysts 1–4 emerged and, in contrast to the previous results, malonate derivative 17 was not respected in the less difficult
the catalytic performances of complexes 3 and 4 with bulkier ring-closure of tosyl derivative 19, where all catalysts gave nearly
N-o-biphenyl substituents were inverted with respect to those of full conversions in 30 min (Fig. 4D).
95796 | RSC Adv., 2016, 6, 95793–95804
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Fig. 4 Kinetic plots of the RCM of 17 and 19.
21–24 and bis(carbonyl) derivatives 25–28 incorporating the NHCs
of interest, were synthesized (Scheme 2).
Steric and electronic properties of NHCs
In order to nd correlations between the structural features of
NHCs L1–L4 and catalytic behaviors of related ruthenium
complexes 1–8, the steric and electronic properties of the carbenes
were assessed by using various metrics.11 To quantify experi-
mentally the steric and electronic parameters of NHC ligands,
rhodium(I) derivatives of general formula [RhCl(COD)(NHC)]
(COD ¼ 1,5-cyclooctadiene) and [RhCl(CO)2(NHC)] are commonly
employed as valuable probes, thanks to the their ease of prepa-
ration and high stability.3i,12 For these purposes, two series of new
rhodium complexes, consisting of cyclooctadiene derivatives
Reaction of dihydroimidazolium salts of NHCs L1–L4 with
KHMDS in toluene at room temperature and subsequent
treatment with [RhCl(COD)]2 afforded, aer purication by
column chromatography, the desired Rh complexes 21–24 as
yellow, air stable solids in good yields (67–83%). Rhodium
carbonyl complexes 25–28 were then obtained in quantitative
yields by bubbling CO in CH2Cl2 solutions of 21–24 (Scheme 2).
All complexes were fully characterized by one- and two
dimensional 1H and 13C NMR techniques, and, for rhodium
Scheme 2 Synthesis of Rh complexes 21–24 and 25–28.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 95793–95804 | 95797

View Article Online
RSC Advances
Paper
Fig. 5 ORTEP8 view of complexes 21 (A), 22 (B), 23 (C) and 24 (D) with the thermal ellipsoids at 30% probability. The H atoms, excluded those
linked to C2 and C3 carbons of the NHC ring, have been omitted for sake of clarity.
˚
cyclooctadiene derivatives 21–24, single crystals suitable for NHC [2.10 A, on average] than for those trans to Cl1 atom [1.98
X-ray structure analysis were independently obtained.13
The resulting molecular structures are depicted in Fig. 5.
In all the four 21–24 compounds the Rh center adopts
A, on average]. This can be ascribed to the greater trans inu-
ence of NHC moiety.
˚
Furthermore, the square coordination planes are almost
a square planar coordination geometry. The Cl1 and C1(NHC) perpendicular to NHC rings making angles of 87.61(8), 88.13(8)
atoms are trans oriented to the C]C p systems of the COD and 89.6(1)ꢀ for complexes 21–23, respectively, and 87.8(1) and
alkene molecule.
85.6(1)ꢀ for molecules A and B of compound 24. The phenyl
In all complexes the distances of the Rh center to the substituted rings linked to the NHC nitrogens, N1 and N2, are
centroids at the COD alkene bonds are longer for those trans to twisted with respect to the NHC ring. These conformations are
Table 1 Selected data for the characterization of NHCs L1–L4 in metal complexes used herein
L1
L2
L3
L4
% VBur from 21–24a
% VBur from 5–8c
1H NMR chemical shi in 21–24d
nav(CO) for 25–28e
TEP from 25–28f
33.4
32.3
4.38 ppm
2037.5 cmꢂ1
2050.2 cmꢂ1
0.467 V
34.0
33.1
4.70 ppm
2039.5 cmꢂ1
2051.8 cmꢂ1
0.583 V
34.8
33.9
4.87 ppm
2039.0 cmꢂ1
2051.4 cmꢂ1
0.574 V
34.8 ꢁ 0.1b
34.7
5.06 ppm
2042.0 cmꢂ1
2053.8 cmꢂ1
0.671 V
DE1/2 for 1–4g
DE1/2 for 5–8g
0.897 V
0.956 V
0.967 V
1.05 V
BDE for 25–28h
55.2 kcal molꢂ1
54.8 kcal molꢂ1
52.8 kcal molꢂ1
52.2 kcal molꢂ1
Obtained from crystallographic data. b Average of the two independent structures. c Obtained from DFT optimized structures. d d (]CH) of the
a
e
olen trans to the NHC registered in CD2Cl2. Carbonyl stretching frequencies determined using IR spectroscopy for complexes in CH2Cl2
g
solutions. f Calculated using the equation TEP ¼ 0.8001nav(CO) + 420.0 cmꢂ1
.
Redox potentials determined using cyclic voltammetry in CH2Cl2
h
under nitrogen; 0.1 M NH4PF6 as supporting electrolyte; internal reference octamethylferrocene; scan rate 100 mV sꢂ1
calculations.
.
Determined by DFT
95798 | RSC Adv., 2016, 6, 95793–95804
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
probably related to short H/C(phenyl ring) interactions, of the growth of the ortho substituents. The nature of the distri-
˚
about 2.70 A such as: H4/C12 and H5/C19, in compounds 21 bution of the NHC ligands L1–L4 around the metal center in the
and 22, as well asH4/C12 and H5/24 in compounds 23 and topographic steric maps of 5–8 with analogous ligands (Fig. 6,
24. All the other structural parameters are in good agreement right) reects the adaptability of NHCs to the changed geometry
with those observed for other similar complexes.
of Ru complexes with respect to Rh compounds. Indeed,
Evaluation of the steric bulkiness of the NHCs L1–L4 of differently from NHC–Rh complexes, an additional asymmetry
complexes 21–24 was made by calculating the percentage of has to be considered for 5–8. NHC–Ru carbene bonds are not
buried volume (% VBur) of each NHC ligand from the crystal collinear with z-axis depicted in the topographic steric maps, as
data of the corresponding Rh complex via the computational already shown by Cavallo et al.,14,16 leading to an increase of
tool SambVca2 developed by Cavallo.14 The % VBur values are ligand steric hindrance illustrated by orange contour line on the
reported in Table 1. Although very close, these values well right low quadrant.
correlate with the size of the substituents within the NHC
To assess the electronic properties of NHCs L1–L4, in analogy
framework. Moving from methyl to phenyl substituents, steric with the approach presented by Bielawski,17 we investigated
bulk is higher when the more encumbered phenyl groups are rhodium complexes 21–28 using spectroscopic techniques. The
located on the ortho positions of the N-phenyl rings rather than series of rhodium derivatives presenting a cyclooctadiene ligand
1
on the backbone. The most signicant steric variation in the in trans position to the NHCs L1–L4 (21–24) was studied by H
examined NHCs is observed changing from all methyl (L1) to all NMR spectroscopy. Dependently from electron density on the
phenyl substituents (L4). The same trend was found for the % metal, coordinated trans double bond of the olen can act as
VBur values extracted from the Density Functional Theory (DFT) a classical two-electron donor maintaining a signicant degree of
optimized geometries of ruthenium complexes 5–8.15
double bond, or adopt metallocyclopropane character as
The steric impact of the different NHCs was investigated at a consequence of metal-to-olen p-backbonding. The capacity of
the molecular level analyzing the topographic steric maps NHCs to participate in p-backbonding interactions can be thus
calculated for both Rh complexes 21–24 and Ru complexes 5–8 estimated by the chemical shis of the olen protons trans to the
(Fig. 6). The steric contour maps for rhodium complexes re- NHC ligand.18 The values observed for the diagnostic olenic
ported in Fig. 5 (le) show that the steric hindrance of the NHC protons of 21–24 are summarized in Table 1. The nature of the
ligands is located in the lower part of the map (on the le and substitution pattern of L1–L4 seems to exert a profound effect on
on the right) where the more intense yellow area for 23 and 24 the electron donor ability of the NHC ligand. The most signi-
corresponds to an increased bent of the N-phenyl groups due to cant difference (0.68 ppm) was observed for complexes 21 and 24
Fig. 6 Topographic steric maps of 21–24 (left) and 5–8 (right). The iso-contour curves are in A˚.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 95793–95804 | 95799

View Article Online
RSC Advances
Paper
bearing all methyl and all phenyl substituents, respectively, and the metal center, as underlined by cathodic shis of the redox
could be interpreted as a consequence of the different group potentials (83–97 mV). Very small change in redox potential (9–
electronegativities of methyl and phenyl groups.19 A much less 11 mV) of complexes characterized by NHCs L2 and L3 with
marked difference for complexes 22 and 23 characterized by mixed methyl and phenyl substituents, alternatively on the
mixed methyl–phenyl substituents on the NHC ring (0.17 ppm) backbone or at the ortho positions of N-phenyl groups, con-
was registered, suggesting a negligible effect of the relative rming that the relative disposition of these substituents on the
disposition of methyl and phenyl groups within the NHC ring on NHC ring is not relevant in terms of resulting NHC electronic
the resulting NHC electron donicity.
properties. This nding provides a further evidence of the
As for rhodium derivatives 25–28 containing a CO ligand in combined effect of backbone substitution and N-aryl substitu-
trans to the NHC, p-back-bonding contribution were assessed tion in determining electronic properties of the NHC ligand. In
by IR spectroscopy. IR spectra for 25–28 were recorded in particular, with regard to the N-aryl substituents, we can
CH2Cl2 and the average stretching frequencies of the carbonyl consider that interaction between electron active ortho-groups
ligands nav(CO) were listed in Table 1.
of N-aryl substituents and metal center operates “through
The Tolman electronic parameters (TEPs)20 of the NHCs L1– space” via donation from the Cipso atom of the N-substituent,
L4 were then estimated from the average stretching vibration as described by Plenio and Cavallo for NHC ruthenium
wavenumbers (Table 1) using the linear regression proposed by complexes characterized by para-groups on the N-aryl substit-
21
Droge and Glorius. The values obtained suggest a more uents.24 It is worthy to underline, moreover, that the presence of
¨
electron-donating nature for the NHC L1, bearing methyl ortho-substituents inuences at the same time both steric and
groups as substituents on both the backbone and the N-phenyl electronic properties of the system.
rings, with respect to the analogue NHC L4 with all phenyl
As a further characterization, bond-dissociation energies
substituents (2050.2 cmꢂ1 for L1 vs. 2053.8 for L4). This nding (BDE) for NHCs L1–L4 from 25–28 were evaluated by DFT
is consistent with previous results and can be explained calculations (see ESI for computational details†) and were re-
considering the methyl substituents as better donors than ported in Table 1.25
phenyl groups. The hybrid substitution patterns represented by
The BDE are mainly affected by two factors: the electron
NHCs L2 and L3 showed TEP values nearly identical, suggesting donation of the backbone substituents and the different
that electronic contribution of backbone substituents is in turn dimension of ortho N-aryl substituents. Electron donation of the
compensated by an almost equal and opposite contribution of backbone substituents seems to play a minor role: the BDE
substituents on the N-aryl groups.
differences between L1 and L2 and between L3 and L4 are 0.4
To gain more information on the electronic situation at the and 0.6 kcal molꢂ1, respectively, with backbone methyl
metal center, we decided to supplement measured TEP data for substituted L1 and L3 presenting higher BDE with respect to the
L1–L4 with electrochemical studies22 of Ru complexes 1–8. corresponding phenyl substituted L2 and L4. By modeling
Indeed, while TEP is a measurement of the ability of a metal simplied complexes with NHCs bearing N-methyl substitu-
center to donate electron density into the p* orbital of a CO ents, similar BDE energy gap (0.8 kcal molꢂ1) was also found
ligand, the redox potential of a metal center should provide between methyl and phenyl backbone substituted NHCs.
a measure of the electronic density at the metal center. The Higher energy gaps can be observed comparing NHCs with
Ru(II)/Ru(III) redox potentials for complexes 1–8 were estab- different ortho N-aryl substituents (2.4 kcal molꢂ1 between L1
lished by cyclic voltammetry and presented in Table 1. A very and L3, 2.6 kcal molꢂ1 between L2 and L4). We believe that
similar trend was observed for both the series of Grubbs' and these differences can be mainly ascribed to the steric hindrance
Hoveyda–Grubbs' second generation catalysts, with an anodic of the ortho N-aryl substituents, that involves a major repulsion
shi of about +0.400 V changing from phosphine-containing with the other ligands and that causes the main energy gaps
catalysts 1–4 to the corresponding oxygen-chelated complexes among these complexes. As a consequence, together with the %
5–8, as already reported by Plenio in an analogous comparative VBur and the steric maps, also the BDE can be read as another
study.23
The electron donating properties of the NHCs L1–L4, deter-
measure of the NHC steric hindrance.
In light of the above results, catalytic outcomes of previously
mined studying the electrochemistry of the related ruthenium described RCM reactions promoted by ruthenium complexes 1–
complexes, were found to be in very good agreement with those 8 turned out to be principally dominated by steric factors. As for
determined through IR spectroscopy of the corresponding phosphine-containing catalysts 1–4, the best performances in
rhodium carbonyl complexes. The difference between the most the easy ring closures of substrates 9 and 11 were registered
and the least electron-donating NHC, L1 and L4, respectively, is with 3 and 4, presenting the most encumbered NHCs (L3 and
0.204 V for the series of Grubbs' second generation catalysts, L4), whereas no correlation between the electron donor abilities
and 0.153 V for the series of Hoveyda' second generation cata- of the carbenes and the observed reactivity trend was found.
lysts. Within the two classes of catalysts, the replacement of Moreover, catalysts 1 and 2, possessing very similar steric
a methyl group of L1 by a phenyl group to give L2 or L3 leads to bulkiness, behave likewise although electronic properties of L1
anodic shis of the Ru(II)/Ru(III) redox potentials (59–116 mV), and L2 are quite different. Therefore, the size of N-substituents
indicating a lower electron density at the metal center; on the of the NHC ligands (o-biphenyl vs. o-tolyl) seems to be the key
other hand, the replacement of a phenyl group of L4 by a methyl factor in determining catalyst efficiency. It is worth to underline
group to provide L2 or L3 implies a higher electron density at that the steric bulk of N-substituents of the NHC ligands in
95800 | RSC Adv., 2016, 6, 95793–95804
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
monophosphine ruthenium complexes has been reported to
have a great inuence on the catalytic activities of the corre-
sponding complexes, favoring the dissociation of the phos-
phine ligand and facilitating the formation of the 14-electron
active species.26 However, faster initiation does not improve
reaction rates but, on the contrary, is related to a faster
decomposition, caused by attack of free phosphine to propa-
gating Ru–methylidene species.4 This drawback becomes more
evident with the increase of steric hindrance of RCM substrates.
In fact, in the RCM of 11 and 17, 3 is conrmed as the best
performing catalyst, whereas the efficiency of 4 worsens
signicantly, and in the most sterically demanding RCM of 13
and 19, both 3 and 4 were surpassed by less encumbered cata-
lysts 1 and 2. Between 3 and 4 possessing NHCs with roughly the
same steric hindrance the highest conversions were constantly
reached with catalyst 3. This nding could be a consequence of
the more electron donating character of NHC L3, which could
better stabilize the 14-electron active species and lead to better
catalytic efficiency. Nevertheless, steric factors affecting the
propagation step of the reaction cannot be excluded.
Scheme 3 Hoveyda-type catalysts and TS previous MCB.
In the light of computational studies conducted by Hillier
and Percy31 where, for substituted olens as propene, highest
energy barrier during the initiation was identied as the tran-
sition state (TS) leading to the formation of metallacyclobutane
(MCB), we performed DFT calculations on the barriers for the
formation of metallacyclobutane in case of substrates 9 (or 13)32
and 17 33 with catalysts 5–8, by locating the geometries and
energies of starting Hoveyda-type catalysts and the TS previous
MCB (Scheme 3).
Due to ligand symmetry, four possible TS can be located for
each catalyst depending on the orientation of the substrate and
of the benzylidene alkoxy moiety. In Fig. 7, TS free energy
barriers, calculated as energy difference between the TS and the
corresponding Hoveyda catalyst (Scheme 3), for low hindered
substrate 9 (or 13) in the presence of o-methyl N-aryl catalysts 5
and 6 are reported. The TS species are shown in quadrant
The slightly different behavior of 1 and 2 appreciable in the
RCM of more hindered diolens 13 and 19 may be explained
considering again steric factors. Indeed, as already reported,3e
catalyst 1 exists as a mixture of conformers with syn and anti
oriented N-o-tolyl groups, the latter offering a slightly more
crowded reactive pocket to accommodate encumbered
substrates.
The above considerations about correlations between steric
and electronic properties of NHCs L1–L4 and catalyst behavior
cannot be easily extended to the series of phosphine free cata-
lysts 5–8. In this case, the absence of decomposition pathways
of the active species ensures high efficiency for all the catalysts
and the most signicant result is related to the slow initiation
rate observed for catalysts 7 and 8 in the RCM of 13. This nding
contrasts sharply with the behavior shown by the analogous
phosphine-containing catalysts 3 and 4 in the same RCM
transformation, where actually they displayed high initiation
rates, even if followed by a rapid decomposition, and may be
ascribed to the different initiation mechanism for the two
families of catalysts. The different behavior of 5–6 and 7–8
registered only in the RCM of the most challenging malonate
derivative 13 would suggest that the nature of the substrate
plays an important role during the initial step of the reaction.
DFT calculations
To clarify this issue, theoretical studies were undertaken. It
should be rst noted that the initiation mechanism in the
presence of Hoveyda-type catalysts is still under discussion.27
Grubbs showed that entropy of activation is negative, implying
an interchange or an associative mechanism,28 whereas later
studies by Plenio suggested a simultaneous competition
between a dissociative and an interchange.29 Computational
work conducted by Solans-Monfort concluded that, while
associative mechanism appeared clearly unfavored, neither the
dissociative nor the interchange mechanism can be excluded as
possible initiation mechanisms.30
Fig. 7 Transition state free energies (TS in Scheme 3) involving low
hindered substrate 9 (or 13)32 in the presence Hoveyda-type catalysts 5
and 6. The TS species are shown in quadrant representation, viewed
along the NHC–Ru bond. Yellow quadrants represent hindered space
due to the bent-down N-phenyl groups. Free energies are in kcal
molꢂ1
.
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 95793–95804 | 95801

View Article Online
RSC Advances
Paper
representation, viewed along the NHC–Ru bond. Yellow quad-
rants represent hindered space due to the bent-down N-phenyl
groups.
Indeed, is not surprising that the lowest energy TS geome-
tries involve substrate and alkoxy benzylidene moiety oriented
toward the less encumbered catalyst side for both 5 and 6.
Lowest TS free energy barriers are also very similar for the two
catalysts, being 21.8 and 21.6 kcal molꢂ1 for 5 and 6,
respectively.
As for o-phenyl N-aryl catalysts 7 and 8 in the RCM of 9 and
13, the space occupancy of the ligand is quite different, as
already highlight with steric maps in previous section. Corre-
sponding TS free energies are reported in Fig. 8 and TS species
Fig. 9 Lowest transition state free energies (TS in Scheme 3) involving
highly hindered substrate 17 33 in the presence Hoveyda-type catalysts
are shown in quadrant representation as well. Yellow quadrants 5–8. The TS species are shown in quadrant representation, viewed
along the NHC–Ru bond. Yellow quadrants represent hindered space
due to the bent-down N-phenyl groups, whereas orange quadrants
always represent hindered space due to the bent-down N-phenyl
groups, whereas orange quadrants represent hindered space
represent hindered space due to the o-phenyl substituents. Free
due to the o-phenyl N-aryl groups. The o-phenyl substituents
balance the hindrance on the two sides of the catalysts
producing as primary effect the increasing of energy barriers,
which are, as secondary effect, attened with respect to the
possible substrate and alkoxy benzylidene moiety orientations.
Lowest barriers where located at 26.9 and 26.6 kcal molꢂ1 for 7
and 8, respectively.
Finally, TS free energy barriers were located for all catalysts
5–8 in case of challenging substrate 17. Only lowest barriers
have been reported for 5–8 in Fig. 9, in quadrants representa-
tion. Corresponding geometries are shown in Fig. 10 (TS-5, TS-6,
energies are in kcal molꢂ1
.
TS-7, TS-8). TS energies climb up to 30–35 kcal molꢂ1, due to
substrate hindrance. Catalyst 6 gives the lowest barrier (29.9
kcal molꢂ1), whereas N-o-biphenyl substituted catalysts 7–8
showed the highest barriers (35.8 and 36.1 kcal molꢂ1
respectively).
Computational results are able to rationalize the experi-
mental kinetic data reported above for catalysts 5–8. Indeed, the
induction period observed for catalysts 7 and 8 in the RCM of
hindered substrates would be generated by high TS barriers for
the formation of the MCB in the initiation step. The same
induction period is not visible for catalyst 5 and 6 that give
signicantly lower barriers, neither for catalysts 7 and 8 in the
RCM of substrates 9 and 13 for the same reason.
Fig. 8 Transition state free energies (TS in Scheme 3) involving low
hindered substrate 9 (or 13)32 in the presence Hoveyda-type catalysts 7
and 8. The TS species are shown in quadrant representation, viewed
along the NHC–Ru bond. Yellow quadrants represent hindered space
due to the bent-down N-phenyl groups, whereas orange quadrants Fig. 10 Lowest free energy transition state geometries (also reported
represent hindered space due to the o-phenyl substituents. Free in quadrant representation in Fig. 9) involving highly hindered substrate
energies are in kcal molꢂ1
.
1733 in the presence Hoveyda catalysts 5–8.
95802 | RSC Adv., 2016, 6, 95793–95804
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
environment provided by the NHC around the metal seems to
cause a signicantly different catalytic behavior. The predomi-
nance of steric on electronic effects was also found to be
responsible for the induction period observed for catalysts 7
and 8 in the RCM of hindered substrates, as suggested by DFT
calculations. The search for the right balance between steric
hindrance of NHC backbone substituents and N-aryl groups is
therefore crucial in the development of tailor-made RCM cata-
lysts. More in general, fully understanding of sterics and elec-
tronics of NHCs L1–L4 represents an important advancement
for their application not only in organometallic catalysis, but
also in organocatalysis.
Conclusions
In the present study, we have reported the synthesis and char-
acterization of two new Grubbs and Hoveyda–Grubbs type
complexes bearing NHCs with N-o-biphenyl substituents and,
alternately, syn methyl (3, 7) or syn phenyl groups (4, 8) on the
backbone. Their catalytic behaviors were investigated in model
RCM reactions and compared to those of already known related
complexes possessing N-o-tolyl groups. The introduction of more
encumbered ortho phenyl instead of o-methyl groups led to quite
conicting results. Indeed, while with catalyst 3, incorporating
syn methyl groups on the backbone, an outstanding improve-
ment of reaction rates was observed in the easiest RCM reactions
forming a di- or trisubstituted cycloolen, with catalyst 4, having
phenyl groups on the backbone, the same behaviour was found
only in the formation of disubstituted cycloolens. It is worth to
underline that catalyst 3 has been identied as the most efficient
system known to date in this class of RCM reactions. With the
increase of the steric hindrance of the RCM substrates, the
presence of N-o-biphenyl substituents revealed to be not bene-
cial for the successful accomplishment of the reactions. As for the
series of Hoveyda-type complexes, a similar trend in catalytic
behavior may be noticed, even if the reactivity differences are
denitely attenuated. Interestingly, the RCM of malonate deriv-
ative 17 promoted by N-o-biphenyl catalysts 7 and 8 displayed
a marked induction period than in the presence of corresponding
N-o-tolyl catalysts 5 and 6. To establish correlation between
catalytic outcomes and structural features of NHCs involved (L1–
L4), a detailed characterization of the steric and electronic
parameters of these carbenes was performed using different
metrics, based also on the employment of new rhodium
derivatives 21–28. The steric demand of L1–L4 was quantied
as % VBur using X-ray analysis of rhodium cyclooctadiene
complexes 21–24 and DFT optimized geometries of ruthenium
isopropoxybenzylidene complexes 5–8. Also the BDE evaluated
from rhodium carbonyl complexes 25–28 were employed as
a measure of the steric hindrance of L1–L4. The most signicant
steric difference in the examined NHCs is between L1 and L4,
containing all methyl and all phenyl substituents, respectively.
The electronic properties of L1–L4 were quantied employing the
carbonyl IR stretching frequencies of rhodium complexes 25–28
(TEP method) and through electrochemical studies on ruthe-
nium complexes 1–8. L1 and L4 were found to be the most and
the less electron donating NHC, respectively, while L3 and L4,
characterized by opposite arrangement of methyl and phenyl
substituents within the NHC ligand, showed nearly the same
donor ability, indicating a negligible role of their relative
disposition.
Acknowledgements
The authors thank Dr Patrizia Iannece, Dr Ivano Immediata,
Patrizia Oliva for technical assistance. Financial support from
`
the Ministero dell'Universita e della Ricerca Scientica e Tec-
nologica is gratefully acknowledged.
References
1 (a) R. H. Grubbs, Handbook of Metathesis, Wiley-VCH,
Weinheim, Germany, 2003, vol. 1–3; (b) S. J. Connon and
S. Blechert, in Ruthenium Catalysts and Fine Chemistry, ed.
P. H. Dixneufand C. Bruneau, Springer, Heidelberg,
¨
Germany, 2004, vol. 11, p. 93; (c) A. Furstner, Angew. Chem.,
2000, 112, 3140; Angew. Chem., Int. Ed., 2000, 39, 3012; (d)
A. H. Hoveyda and A. R. Zhugralin, Nature, 2007, 450, 243;
(e) P. H. Deshmuhk and S. Blechert, Dalton Trans., 2007,
2479; (f) D. Astruc, New J. Chem., 2005, 29, 42; (g)
A. M. Lozano-Vila, S. Monsaert, A. Bajek and F. Verpoort,
Chem. Rev., 2010, 110, 4865; (h) Olen Metathesis Theory
and Practice, ed. K. Grela, J. Wiley & Sons, Hoboken, NJ, 2014.
2 (a) G. C. Vougioukalakis and R. H. Grubbs, Chem. Rev., 2010,
110, 1746; (b) C. Samojlowicz, M. Bieniek and K. Grela, Chem.
Rev., 2009, 109, 3708; (c) M. Bieniek, A. Michrowska,
D. L. Usanov and K. Grela, Chem.–Eur. J., 2008, 14, 806; (d)
V. Paradiso, C. Costabile and F. Grisi, Molecules, 2016, 21,
117.
3 (a) I. C. Stewart, T. Ung, A. A. Pletnev, J. M. Berlin, R. H. Grubbs
and Y. Schrodi, Org. Lett., 2007, 9, 1589; (b) J. M. Berlin,
K. Campbell, T. Ritter, T. W. Funk, A. Chlenov and
R. H. Grubbs, Org. Lett., 2007, 9, 1339; (c) C. K. Chung and
R. H. Grubbs, Org. Lett., 2008, 10, 2693; (d) K. M. Kuhn,
J.-B. Bourg, C. K. Chung, S. C. Virgil and R. H. Grubbs, J. Am.
Chem. Soc., 2009, 131, 5313; (e) F. Grisi, A. Mariconda,
C. Costabile, V. Bertolasi and P. Longo, Organometallics,
2009, 28, 4988; (f) C. Costabile, A. Mariconda, L. Cavallo,
P. Longo, V. Bertolasi, F. Ragone and F. Grisi, Chem.–Eur. J.,
2011, 17, 8618; (g) A. Perfetto, C. Costabile, P. Longo,
V. Bertolasi and F. Grisi, Chem.–Eur. J., 2013, 19, 10492; (h)
A. Perfetto, C. Costabile, P. Longo and F. Grisi,
Organometallics, 2014, 33, 2747; (i) Y. Borguet, G. Zaragoza,
A. Demonceau and L. Delaude, Dalton Trans., 2015, 44, 9744.
4 S. H. Hong, A. Chlenov, M. W. Day and R. H. Grubbs, Angew.
Chem., Int. Ed., 2007, 46, 5148.
Therefore, to summarize, we have illustrated that the
systematic variation of substituents on the NHC backbone and
on the ortho positions of the N-aryl substituents (methyl and/or
phenyl) allowed the ne-tuning of the stereoelectronic proper-
ties of the corresponding ligands. Although electronic effects
deriving from the different substitution within the NHC ligand
proved to be important, no correlations with catalytic results
emerged. On the other hand, small differences in the steric
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 95793–95804 | 95803

View Article Online
RSC Advances
Paper
5 S. Y. M. Chooi, P.-h. Leung, S.-c. Ng, G. H. Quek and 17 (a) D. M. Khramov, V. M. Lynch and C. W. Bielawski,
K. Y. Sim, Tetrahedron: Asymmetry, 1991, 2, 981.
6 The synthesis of the analogous chloride salt was also
attempted. No formation of the desired product was
observed.
Organometallics, 2007, 26, 6042; (b) D. M. Khramov,
E. L. Rosen, J. A. V. Er, P. D. Vu, V. M. Lynch and
C. W. Bielawski, Tetrahedron, 2008, 64, 6853.
18 Only few data on 1H NMR chemical shi measurements
used to evaluate electronic properties of NHCs in
[RhCl(COD)(NHC)] complexes are available in the literature
(see ref. 17), thus conclusions drawn from this method
could be questionable and need to be conrmed by other
metrics.
7 (a) The low yield obtained for complex 3 was ascribed to
a partial decomposition occurring during the step of
purication, very likely related to the rather low stability of
the complex in solution; (b) Complex 3 decompose in both
CD2Cl2 and C6D6 solution within 24 h, therefore its NMR
characterization was limited to 1H and 31P NMR 19 Group electronegativities: CH3 2.3, Ph 3.0, see: (a)
experiments.
P. R. Wells, Prog. Phys. Org. Chem., 1968, 6, 111; (b)
E. V. Anslyn and D. A. Dougherty, Modern Physical Organic
Chemistry, University Science Books, Sausalito, CA, 2006, p.
16.
8 M. N. Burnett and C. K. Johnson, ORTEP III, Report ORNL-
6895, Oak Ridge National Laboratory, Oak Ridge, TN, 1996.
9 (a) M. B. Dinger and J. C. Mol, Adv. Synth. Catal., 2002, 344,
671; (b) T. Ritter, A. Hejl, A. G. Wenzel, T. W. Funk and 20 C. A. Tolman, Chem. Rev., 1977, 77, 313.
¨
R. H. Grubbs, Organometallics, 2006, 25, 5740.
21 T. Droge and F. Glorius, Angew. Chem., Int. Ed., 2010, 49,
10 It should be noted that the very fast RCM reactions of N-tosyl
6940.
derivatives 11, 15 and 19 promoted by catalysts 5–8 were 22 (a) A. B. P. Lever, Inorg. Chem., 1990, 29, 1271; (b)
performed at lower catalyst loading than the A. B. P. Lever, Inorg. Chem., 1991, 30, 1980.
corresponding with malonate derivatives to better 23 M. Sußner and H. Plenio, Chem. Commun., 2005, 5417.
a
¨
¨
appreciate differences in catalyst behaviour.
24 (a) S. Leuthaußer, V. Schmidts, C. M. Thiele and H. Plenio,
11 (a) L. Falivene, A. Poater and L. Cavallo, in N-Heterocyclic
Carbenes: Effective Tools for Organometallic Synthesis, ed. S.
Chem.–Eur. J., 2008, 14, 5465; (b) R. Credendino,
L. Falivene and L. Cavallo, J. Am. Chem. Soc., 2012, 134, 8127.
P. Nolan, Wiley-VCH, 2014; (b) D. J. Nelson and 25 L. Cavallo, A. Correa, C. Costabile and H. Jacobsen, J.
S. P. Nolan, Chem. Soc. Rev., 2013, 42, 6723. Organomet. Chem., 2005, 690, 5407.
12 Selected examples: (a) S. Wolf and H. Plenio, J. Organomet. 26 (a) F. C Courchay, J. C. Sworen and K. B. Wagener,
Chem., 2009, 694, 1487; (b) M. Iglesias, D. J. Beetstra,
B. Kariuki, K. J. Cavell, A. Dervisi and I. A. Fallis, Eur. J.
Inorg. Chem., 2009, 13, 1913; (c) R. Savka and H. Plenio,
Eur. J. Inorg. Chem., 2014, 6246; (d) C. Mejuto, G. Guisado-
Barrios and E. Peris, Organometallics, 2014, 33, 3205; (e)
Macromolecules, 2003, 36, 8231; (b) T. M. Trnka and
¨
R. H. Grubbs, Acc. Chem. Res., 2001, 34, 18; (c) A. Furstner,
L. Ackermann, B. Gabor, R. Goddard, C. W. Lehann,
R. Mynott, F. Stelzer and O. R. Thiel, Chem.–Eur. J., 2001,
7, 3236.
V. Paradiso, V. Bertolasi, C. Costabile and F. Grisi, Dalton 27 D. J. Nelson, S. Manzini, C. A. Urbina-Blanco and S. P. Nolan,
Trans., 2016, 45, 561; (f) R. Savka, M. Bergmann, Y. Kanai,
S. Foro and H. Plenio, Chem.–Eur. J., 2016, 22, 9667.
13 Single crystals of 21–24 were obtained by slow evaporation of
a CH2Cl2/pentane 2 : 1 mixture. Any attempts to obtain
crystals for rhodium carbonyl complexes 25–28 failed.
14 L. Falivene, R. Credendino, A. Poater, A. Petta, L. Serra,
Chem. Commun., 2014, 50, 10355.
28 (a) G. C. Vougioukalakis and R. H. Grubbs, Chem.–Eur. J.,
2008, 14, 7545; (b) T. Vorfalt, K.-J. Wannowius and
H. Plenio, Angew. Chem., Int. Ed., 2010, 49, 5533.
29 V. Thiel, M. Hendann, K.-J. Wannowius and H. Plenio, J. Am.
Chem. Soc., 2012, 134, 1104.
~
´
R. Oliva, V. Scarano and L. Cavallo, Organometallics, 2016, 30 F. Nunez-Zarur, X. Solans-Monfort, L. Rodrıguez-Santiago
35, 2286. The default processing parameters of SambVca
2.0 web application were kept unchanged.
15 Since no X-ray structure was available for complex 8, DFT
and M. Sodupe, Organometallics, 2012, 31, 4203.
31 I. W. Ashworth, I. H. Hillier, D. J. Nelson, J. M. Percy and
M. A. Vincent, ACS Catal., 2013, 3, 1929.
optimized structures of 5–8 were used for the % VBur 32 Substrate 9 and 13 were simulated by 4,4-dimethylpent-1-
calculation % VBur values determined by DFT are in good
agreement with % VBur from X-ray structures of 5–7.
16 F. Ragone, A. Poater and L. Cavallo, J. Am. Chem. Soc., 2010,
132, 4249.
ene.
33 Substrate 17 was simulated by 2,4,4-trimethylpent-1-ene.
95804 | RSC Adv., 2016, 6, 95793–95804
This journal is © The Royal Society of Chemistry 2016