N-Heterocyclic carbene rhodium(i) complexes containing an axis of chirality: Dynamics and catalysis

Article abstract of DOI:10.1039/c3nj01620j

The novel rhodium(i) complexes [RhCl(NBD)(NHC)] [NBD = norbornadiene, NHC = 1-benzyl-3-R-imidazolin-2-ylidene; R = Me (3a), Bz (3b), Tr (3c), ^tBu (3d)], containing on one nitrogen the benzyl substituent and on the other increasing bulky alkyl substituents were prepared. All the complexes display restricted rotation around the metal-carbene bond and yield conformational enantiomers. The stereodynamics and racemization barriers about the Rh-carbene have been determined by means of NMR spectroscopy for 3a-c, whereas for the bulkiest 3d only the lower limit (91 kJ mol^-1) could be calculated. Whilst the racemization barriers obtained by DFT calculations for 3a,b and 3d matched the experimental values, in the case of 3c the latter (62.3 kJ mol^-1) was much smaller with respect to the calculated one (101.7 kJ mol^-1). The lower experimental barrier has been attributed to a dissociative pathway that produces a solvated ionic pair in the transition state. The catalytic activity of the neutral rhodium(i) complexes 3a and 3d in the hydrosilylation with HSiMe₂Ph of the terminal alkynes PhC≡CH, TolC≡CH, nBuC≡CH, Et₃SiC≡CH, and (CPh₂OH)C≡CH has been investigated, and compared with the amide-functionalized [RhCl(NBD){1-(2-NHBoc-ethyl)-3-Me-imidazolin-2-ylidene}] (4) and with [RhCl(NBD){1-butyl-3-Me-imidazolin-2-ylidene}] (5).

Full text of DOI:10.1039/c3nj01620j

NJC
View Article Online
View Journal | View Issue
PAPER
N-Heterocyclic carbene rhodium(I) complexes
containing an axis of chirality: dynamics and
catalysis†
Cite this: New J. Chem., 2014,
38, 1768
Maria Cristina Cassani,* Marta Anna Brucka, Cristina Femoni, Michele Mancinelli,
Andrea Mazzanti, Rita Mazzoni* and Gavino Solinas
The novel rhodium(I) complexes [RhCl(NBD)(NHC)] [NBD = norbornadiene, NHC = 1-benzyl-3-R-imidazolin-
t
2-ylidene; R = Me (3a), Bz (3b), Tr (3c), Bu (3d)], containing on one nitrogen the benzyl substituent and on
the other increasing bulky alkyl substituents were prepared. All the complexes display restricted rotation
around the metal–carbene bond and yield conformational enantiomers. The stereodynamics and racemiza-
tion barriers about the Rh–carbene have been determined by means of NMR spectroscopy for 3a–c, whereas
for the bulkiest 3d only the lower limit (91 kJ molꢀ1) could be calculated. Whilst the racemization barriers
obtained by DFT calculations for 3a,b and 3d matched the experimental values, in the case of 3c the latter
(62.3 kJ molꢀ1) was much smaller with respect to the calculated one (101.7 kJ molꢀ1). The lower experi-
mental barrier has been attributed to a dissociative pathway that produces a solvated ionic pair in the transi-
tion state. The catalytic activity of the neutral rhodium(I) complexes 3a and 3d in the hydrosilylation with
HSiMe2Ph of the terminal alkynes PhCꢁCH, TolCꢁCH, nBuCꢁCH, Et3SiCꢁCH, and (CPh2OH)CꢁCH has
been investigated, and compared with the amide-functionalized [RhCl(NBD){1-(2-NHBoc-ethyl)-3-Me-
imidazolin-2-ylidene}] (4) and with [RhCl(NBD){1-butyl-3-Me-imidazolin-2-ylidene}] (5).
Received (in Victoria, Australia)
20th December 2013,
Accepted 17th February 2014
DOI: 10.1039/c3nj01620j
www.rsc.org/njc
hindrance on the N-heterocyclic ligand and on the alkyne
substrates affects conversion and selectivity. More specifically,
1. Introduction
N-Heterocyclic carbenes (NHCs) are efficient ancillary ligands for the former the best results were achieved by employing the
because of their strong coordination ability and their tuneable less encumbered catalyst with TolCRCH, whereas by using
character which allows the control of the steric and electronic hindered alkynes such as Et3SiCRCH or (CPh2OH)CRCH the
properties on the metal centre.1
hydrosilylation selectively leads to the formation of the b-(E)-
We recently reported the structures, dynamics and catalytic vinylsilane and a-bis(silyl)alkene isomers.1d,3
activity of rhodium(I)–NHC complexes of the type [RhX(NBD)- In view of these previous results, we report here a series of
(NHC)] [NHC 1-(2-NHBoc-ethyl)-3-R-imidazolin-2-ylidene; novel [RhCl(NBD)(NHC)] rhodium(I) complexes (NHC = 1-benzyl-
=
t
R = methyl (Me), benzyl (Bz), trityl (Tr); X = Cl, I], in which 3-R-imidazolin-2-ylidene; R = Me, Bz, Tr, Bu), where a N-benzyl
the N-heterocyclic ligand was bearing a Boc-protected 1-(2- substituent was chosen as a chiral probe and the other
aminoethyl) group on one side and increasingly bulky N-alkyl N-substituent is an sp3 carbon with increasing bulkiness. In the
substituents on the other (Fig. 1).2
hydrosilylation of terminal alkynes, a comparative study between
The catalytic activity of the neutral amide-functionalized the catalytic activity of these new systems and the previously
rhodium(I) complexes in the hydrosilylation of terminal alkynes reported Rh(I) complexes, bearing amide-functionalized NHC
with HSiMe2Ph was also investigated. We found that the steric ligands, has been made. The purpose of this investigation was
to elucidate the role played by the heteroatom functionality and
the steric hindrance close to the reactive site.
Dipartimento di Chimica Industriale ‘‘Toso-Montanari’’, viale Risorgimento 4,
I-40136 Bologna, Italy. E-mail: maria.cassani@unibo.it, rita.mazzoni@unibo.it;
Fax: +39 0512093694; Tel: +39 051 2093700
† Electronic supplementary information (ESI) available: Additional experimental
2. Results and discussion
2.1. Synthetic studies
and characterization data; crystal data and experimental details for all structures,
molecular structure of 2b; computational details; and additional graphs and
experimental data for the catalysis. X-ray data in the form of CIF files for 2b, 3a
The four imidazolium salts [BzImR]X (Im = imidazole) shown in
Scheme 1 were prepared either by alkylation of 1-benzylimidazole
and 3b. CCDC 957761 (3a) and 957762 (3b). For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/c3nj01620j
1768 | New J. Chem., 2014, 38, 1768--1779
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
Fig. 1 N-Heterocylic carbene–amide rhodium(I) complexes and metal–carbene bond rotational barriers.
(1a, 1b)4,5 with the pertinent alkyl halide, or by alkylation of tert- observed for the previously reported amide-functionalized
butylimidazole with benzyl bromide (1d).6 The former synthetic rhodium(I)–NHC complexes,2 3a–d are air- and moisture-
procedure was also employed for the preparation of the novel stable in the solid state and in solution, as confirmed by
imidazolium salt 1c. The imidazolium salt 1a is a pale yellow continuous monitoring of NMR samples in CDCl3 over 2 weeks.
viscous liquid, whereas the salts 1b–d are very hygroscopic crystal- Complexes 3a–d have been fully characterized by ESI-MS mass
line solids. They are all soluble in chlorinated solvents such as spectrometry, 1H and 13C NMR using gCOSY, gHSQC, and
CH2Cl2 and CHCl3, and were fully characterized by elemental gHMBC experiments for full resonance assignments. The most
analysis, NMR spectroscopy and electrospray ionization mass significant chemical shifts and coupling constants in 1H and
spectrometry (ESI-MS). The NMR resonances of 1c were observed 13C NMR spectra of 3a–d are also summarized in Tables S1 and
at chemical shifts typical for compounds of this kind, with the S2 in the ESI.†
benzylic protons resonating at 5.97 ppm.2
The general trend which can be observed for all of the
1
Subsequent reaction of 1a–d with Ag2O in dichloromethane, compounds 3a–d is the increasing complexity of their H NMR
carried out at room temperature with the exclusion of light, spectra (acquired at room temperature) with the increasing
afforded the less bulky silver(I)–NHC complexes 2a7 and 2b8 bulkiness of the N-alkyl substituents. The bigger the steric
after 2 h, and the more encumbered 2c and 2d9 after 48 h. The hindrance, the higher the activation energy of the rotation about
complex 2c bearing a trityl group on the NHC ligand is new and the Rh–carbene bond. As an example, the signals attributed to
its formation was unambiguously confirmed by elemental the imidazole backbone protons (resonating at a slightly lower
analysis, NMR spectroscopy and ESI-MS mass spectrometry. field with respect to the corresponding imidazolium salts and
As regards the 13C NMR, it is worth mentioning that in the case silver complexes) appear as two singlets at 6.76 and 6.65 ppm for
of the hitherto not reported complex 2c, the Ag–C resonance 3a and as one singlet at 6.61 ppm for the symmetric 3b. However,
appears as a broad doublet at 185.0 ppm with a 13C–107/109Ag for the more hindered 3c,d the discussed protons appear as two
3
coupling constant of 236 Hz, in keeping with literature data.10 doublets with the JH,H E 2 Hz. At room temperature the
The rhodium complexes [RhCl(NBD)(NHC)] (NBD = 2,5- benzylic CH2 protons are found as singlets for the complexes
norbornadiene; NHC = 1-benzyl-3-R-imidazolin-2-ylidene; 3a: 3a,b (vide infra), whereas CH2 of 3c,d decoalesces into an AB
t
2
R = Me; 3b: R = Bz; 3c: R = Tr; 3d: R = Bu) were synthesized in system, showing two doublets with the JH,H E 15 Hz (Fig. S1,
high yields (70–90%) by transmetalation from the Ag(I)–NHC ESI†). The formation of the AB system is the consequence of
complexes 2a–d in dichloromethane (Scheme 1).
the C1 symmetry of the complexes in their ground state. In the
Complexes 3a–d were isolated as yellow microcrystalline NMR time scale, when the rotation about the Rh–C bond is fast
solids by gradient column chromatography under argon on (as in 3a and 3b) the benzylic CH2 is a singlet, while it splits into
anhydrous silica gel. They are completely soluble in chlorinated the AB spin system when the rotation is slow.
solvents and acetonitrile, partially soluble in diethyl ether, and
With regard to the 13C NMR spectra all 3a–d complexes show
completely insoluble in petroleum ether. Contrary to what was a doublet at ca. 180 ppm, resulting from the coupling of the
Scheme 1 Synthesis of rhodium(I)–NHC complexes.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 1768--1779 | 1769

View Article Online
NJC
Paper
dichloromethane and petroleum ether (1 : 4). The structure of
3a, which crystallized in a centrosymmetric space group (P21/c)
of the monoclinic crystal system (Z = 4), consists of separated
C
20H27ClN2Rh molecules arranged in eclipsed, alternating col-
umns of conformational enantiomers in a 1 : 1 ratio. The two
possible orientations of the benzyl and methyl substituents on
the imidazolium ring result from the axial chirality present in
the molecule of 3a. Rhodium shows a classic square-planar
coordination and is bonded to the Cl(2) atom, the carbenic C(1)
and the bidentate norbornadiene fragment. There is a stronger
trans effect observed for the carbene moiety when compared to
the chlorine ligand. This can be inferred from the difference in
bond lengths between the Rh atom and the carbon atoms of the
norbornadiene moiety situated in trans orientation with
respect to the chlorine and the carbene, respectively. As a
matter of fact, the Rh–C(16) and Rh–C(20) contacts trans to
the chlorine are 2.083(4) and 2.098(4) Å, in that order, while the
Rh–C(14) and Rh–C(15) bond lengths trans to the Cl atom are
remarkably longer: 2.194(4) and 2.213(4) Å, correspondingly.
It is worth noting that despite the presence of donor chloride
anions, there are no significant intra- or intermolecular hydrogen
bonds in the solid state.
The complex 3b crystallizes in the centrosymmetric P21/n
group of the monoclinic crystal system (Z = 4) and both
conformational enantiomers are generated by the centre of
symmetry of the unit cell, as in the previous case. Similarly,
there is also the trans effect observable in the structure of 3b,
with the Rh–Cnorbornadiene distances shorter when in trans
orientation with respect to the carbene moiety than those being
trans to the chloride ligand. More precisely, the Rh–C(20) and
Rh–C(21) contacts are 2.084(2) and 2.091(2), respectively,
whereas Rh–C(22) and Rh–C(23) distances are longer: 2.185(2)
and 2.195(2).
Fig. 2 ORTEP diagram of 3a depicted with thermal ellipsoids at 30%
probability.
carbenic carbon to the rhodium nucleus with 1J(13C–103Rh)
coupling constants of ca. 58 Hz (see also the Experimental part
and Table S2, ESI†).
It is worth mentioning that complexes [RhCl(COD)(NHC)]
having 1,5-cyclooctadiene (COD) as an ancillary ligand instead
of NBD, analogous to 3a and 3b, have been already described.
However, when R = Me11 no dynamic behaviour was reported,
whilst when R = Bz12 two separate doublets associated with
diastereotopic hydrogens of the benzyl group were observed in
the 1H NMR spectra, though the rotational barrier could not be
measured.
2.2. Crystal structure determination for 3a and 3b
The molecular structures of the rhodium(I) complexes 3a and
3b were determined via X-ray diffraction and are illustrated in
Fig. 2 and 3, while crystal data and experimental details for all
structures are collected in Table S3 (ESI†). Crystals of 3a,b
suitable for diffraction were grown from a double layer of
The bond distances and angles in complexes 3a and 3b data
are quite similar. A list of the most relevant values is reported in
Table S4 (ESI†).
2.3. Stereodynamics
The stereodynamics and the rotational barriers about the
Rh–carbene bond have been determined by means of variable-
temperature (VT) NMR spectroscopy.13 The direct consequence
of the hindered rotation about the Rh–carbene bond is the lack
of the symmetry plane in these complexes. The norbornadiene
moiety and the chlorine atom are displaced out-of-plane with
respect to the imidazole ring. The two side arms being different,
the molecule belongs to the C1 symmetry point group; therefore
a pair of conformational enantiomers is generated (Fig. 4).14
The interconversion pathway of the two enantiomeric con-
formers is shown in Scheme 2. The two enantiomeric ground
states GS and GS0 can interconvert into each other by two
possible transition states, due to the rotation around the
carbene–rhodium bond.
The first transition state (TS-1) is reached from the GS by a
901 counter clockwise rotation of the imidazole and corre-
sponds to the crossing of the chlorine atom on the benzylic
CH2 of one of the side arms of the imidazole. The second
Fig. 3 ORTEP diagram of 3b depicted with thermal ellipsoids at 30%
probability.
1770 | New J. Chem., 2014, 38, 1768--1779
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
giving a characteristic AB system.15 In addition, the CH2 signals
1
are in a free area of the H spectrum, thus allowing a reliable
simulation of the line shape. The 1H NMR spectra of the
complex 3a recorded in the ꢀ35 to 25 1C range (1,1,2,2-
tetrachloroethane-d2) are shown in Fig. 5 (left). The single
signal of the benzylic geminal protons at 5.73 ppm broadened
upon decreasing the temperature and reached the coalescence
at +10 1C. Upon further lowering the temperature, it splits into
the expected AB system, showing sharp lines at ꢀ35 1C
(5.60 and 5.85 ppm, J = 15.0 Hz).
Line shape simulation at various temperatures yielded the
rate constants for the enantiomerization process, from which
an activation energy of 56.5 kJ molꢀ1 was derived using the
Eyring equation (Table 1).
Fig. 4 Coordination plane of the rhodium(I) in the complexes 3a–d and
the two conformational enantiomers generated by the chiral axis along the
Rh–carbene bond.
In complex 3b the two benzyl groups on the N-side arms of
the imidazole caused a bigger steric hindrance with respect to
the previously discussed 3a, in that the benzylic protons
appeared anisochronous as a broad doublet at room tempera-
ture, and the coalescence was reached at +32 1C. Upon further
raising the temperature, the signal sharpened and at +75 1C it
exhibited a very sharp singlet. On the other hand, at ꢀ10 1C, the
splitting pattern characteristic of the AB system was observed
(Fig. 5, right). Although in this case the ground state has Cs
symmetry, the CH2 signals are split into an AB system because
the symmetry plane of the molecule is not coincident with
the local symmetry plane of the CH2. An energy barrier of
62.7 kJ molꢀ1 was derived by line-shape simulation (Table 1).
As usual in conformational processes, the energy barriers
measured for 3a and 3b did not change with the temperature,
thus implying a very small activation entropy. This suggests
that the observed barrier should be driven by steric effects only,
and it is consistent with the increase of the activation energy
Scheme 2 Enantiomerization pathway for compounds 3a–d; R = Me, Bz,
Tr, and Bu.
t
between the less sterically hindered 3a and 3b (DDG‡
=
transition state (TS-2) takes place when a clockwise rotation
forces the chlorine atom to cross the second alkyl side arm
group (R) on the imidazole. The steric hindrance of R obviously
influences the activation energy of the rotational barrier. DFT
calculations of the two possible transition states (Table 1),
performed at the B3LYP/LANL2DZ level, suggested that the
threshold pathway (i.e., that with the lowest transition-state
energy) corresponds in all cases to the passage of the halogen
atom on the benzyl group (TS-1), with the simultaneous cross-
ing of the norbornadiene on the second side arm (R).
6.2 kJ molꢀ1). These data very well agree with those reported
by Enders and Gielen,16 who proposed a steric origin for the
rotational barrier.
In the case of complex 3c bearing the trityl group, high
temperature VT experiments cannot be acquired because a
decomposition process took place before reaching the coales-
cence. To calculate the energy barrier, an exchange spectro-
scopy (EXSY) experiment has been employed. The 1D EXSY
experiments17 were carried out at four different temperatures
(ꢀ10, ꢀ15, ꢀ20 and ꢀ25 1C) on complex 3c. The linear func-
tions of ln([A]eq – [A]t) vs. the mixing time (25, 50, 75 and
100 ms) were plotted for each temperature (Fig. 6a), providing
the kinetic constants for the Rh–C rotation. For each tempera-
ture, the activation energy was then derived using the Eyring
The N-benzyl group on one side of the imidazole was chosen
as a probe of chirality, because its enantiotopic geminal pro-
tons become diastereotopic in the presence of a chiral axis
Table 1 Calculated and experimental energy barriers for the enantiomer- equation.
ization of 3a–d (energies in kJ molꢀ1, calculations at the B3LYP/LANL2DZ
level)
The plot of DG‡ against temperature (Fig. 6b) showed a marked
increase of the activation energy, indicating a substantial and
negative activation entropy. In particular, the EXSY experiment
showed that the activation enthalpy DH‡ (62.3 kJ molꢀ1) was
identical, within the experimental error, to the barrier derived in
the case of the dibenzyl-substituted compound 3b.
Compound
TS-1
TS-2
Experimental DG‡ (kJ molꢀ1
)
3a
3b
3c
3d
55.0
65.3
101.7
96.7
58.5
78.7
115.9
119.7
56.5
62.7
62.3a
491
The large negative activation entropy (ꢀ115 ꢂ 10 J Kꢀ1 molꢀ1
)
a
DH‡ value.
suggested that a strongly organized transition state takes place
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 1768--1779 | 1771

View Article Online
NJC
Paper
Fig. 5 Variable-temperature spectra of 3a and 3b showing the evolution of the benzylic CH2 signals (1H NMR at 600 MHz, C2D2Cl4). On the right side of
each plot the simulations with the corresponding rate constants are reported.
Fig. 6 (a) Plots of ln([A]eq – [A]t) vs. mixing time at different temperatures and (b) the dependence of the activation energy on the temperature.
in the case of compound 3c,18 a different interconversion path- molecules and hence leads to a negative entropy of activation.21
way being the only available option. The two following hypo- Since the correlated motion is predicted to have a large barrier,
theses can be formulated. For the first one the lower barrier which does not match the experimental value, in our opinion the
could be the result of dynamic gearing between the rotation of second hypothesis (two solvated ionic partners in the transition
the rhodium–carbene and the trityl group. Such a situation is state) appears to be the correct one.
well known to generate a single transition state with lower
In the case of complex 3d, the high activation energy, due to
t
activation energy with respect to the rotation of the single the presence of the sterically demanding Bu group, prevented
groups.19 For the second hypothesis a process involving a bond the observation of the coalescence of the benzylic protons,
cleavage followed by subsequent regeneration of the bond can be and both VT spectra and EXSY approaches appeared to be
conceived and, although it is known that such a process yields a unsuitable for the experimental determination of the kinetic
positive entropy,20 in the present case the trityl dissociation data. Being the highest workable temperature about +70 1C,
produces two charged sub-units. The stabilization of two only a lower limit for the barrier could be calculated. If a 0.1 sꢀ1
charged counterparts involves a much higher ordering of solvent rate constant at +70 1C is considered, the corresponding energy
1772 | New J. Chem., 2014, 38, 1768--1779
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
barrier is 91 kJ molꢀ1. This value suggests that the two
conformational enantiomers could be sufficiently stable at
room temperature to be observed with a different technique
such as dynamic HPLC.22 Unfortunately, any attempt to observe
separate or exchanging peaks on HPLC enantioselective
columns was unsuccessful because of chemical degradation
of the complex in those conditions.
2.4. Hydrosilylation of terminal alkynes
Complexes 3a and 3d, respectively, representing the less and
the more hindered rhodium species, were chosen to evaluate
the catalytic performance in the hydrosilylation of the following
terminal alkynes: phenylacetylene, tolylacetylene, 1-hexyne,
(triethylsilyl)acetylene, and 1,1-diphenyl-2-propyn-1-ol. The
catalytic reactions were carried out in CDCl3 using a slight
Fig. 7 Reaction profile of conversion vs. time for the hydrosilylation of
PhCRCH at 25 1C with 3a, 3d, 4 and 5: catalyst loading of 1%.
1
excess of HSiMe2Ph and were routinely monitored by H-NMR
chelating effect on the metal centre during the catalytic
cycle.1j,3a,25 When the methyl group on the NHC ligand of 4 is
replaced with a benzyl group (see Fig. 1, R = Bz),2 the TOF
decreases down to 40 hꢀ1 and remains significantly higher
than the one registered for the less encumbered 5. A similar
behaviour was observed for the hydrosilylation of the other
substrates (Fig. S3–S6, ESI†). As previously reported,2 for
1-hexyne and 1,1-diphenyl-2-propyn-1-ol, we again observed
longer initiation times. Such behaviours have been ascribed
to an electronic effect for the former and to the steric hindrance
for the latter substrate.3a,26,27
With regard to the selectivity, as generally reported for transition
metal catalyzed hydrosilylation of 1-alkynes,28 and for Rh(I)NHC
hydrosilylation catalysts in particular,3a the reaction is non-selective.
As a matter of fact, the NMR analyses on the reaction products
obtained by employing complexes 3a, 3d and 5 are in agreement
with literature data3a and with what was previously reported for 4.2
As shown in Scheme 3 the catalysts convert phenylacetylene and
tolylacetylene into a mixture of the three possible isomeric vinyl-
silane derivatives: b(Z) or b(E) 1-silyl-1-alkenes (anti-Markovnikov
addition), and a-2-silyl-1-alkene (Markovnikov addition). Further-
more the formation of the corresponding alkene is observed.
The b-(Z) vinylsilane is the major product until the substrate
completely disappears. Once the complete conversion is reached
b-(Z) isomerizes to b-(E) vinylsilane which, at the end of the
reaction, is always the major product. The formation of the
thermodynamically less stable b-(Z) isomer has been described
in detail by Oro et al.3a using a modified Chalk–Harrod mecha-
nism, whereas the ensuing (Z)–(E) isomerization (favoured by the
presence of a slight excess of hydrosilane) has already been widely
described in the literature.29 In Fig. 8 an example of conversion
and selectivity profiles vs. time for the reaction of PhCRCH and
dimethylphenylsilane mediated by the catalyst 3a is reported.
The bar diagrams in Fig. 9 for phenylacetylene and Fig. S7
for tolylacetylene summarize the data for all the catalysts under
study and show that the a isomer is always identified in
variable amounts, 5–16%, while the alkene formation occurs
in the range 0–12%.
spectroscopy. The results have been compared with those
previously reported for [Rh(NBD)Cl{1-(2-NHBoc-ethyl)-3-Me-
imidazolin-2-ylidene}] (4)2 and also with [Rh(NBD)Cl{1-butyl-
3-Me-imidazolin-2-ylidene}] (5), in order to shed light on the
role of the -NHBoc substituent in both the catalytic activity and
selectivity. Complex 5 was prepared following the procedure
described for the parent compound bearing a COD ancillary
ligand.23 Complexes under studies are summarized in Chart 1.
A plot of conversions versus time for the hydrosilylation
reaction of phenylacetylene performed at 25 1C with a catalyst
loading of 1% is presented in Fig. 7, which shows that all
rhodium complexes completely convert the substrate.
Although the general behaviour of the conversion profiles
confirms that the reaction rate is affected by steric encum-
brance,2 comparison of the turn over frequency (TOF) between
4 (48 hꢀ1) and 5 (27 hꢀ1 24
clearly demonstrated the beneficial
)
effect of the amide-Boc group, which provides a dynamic
The experimental data obtained for the other substrates
(reported in the ESI†) show that the a isomer is always present,
Chart 1 [RhCl(NBD)(NHC)] complexes employed in the hydrosilylation of
phenylacetylene and tolylacetylene.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 1768--1779 | 1773

View Article Online
NJC
Paper
Scheme 3 Hydrosilylation of PhCRCH and TolCRCH with Rh(I) complexes.
Fig. 8 Reaction profile of conversion and selectivity vs. time for the hydrosilylation of PhCRCH with 3a; catalyst loading 1%.
Fig. 9 Selectivity vs. conversion and time for the hydrosilylation of phenylacetylene catalyzed by 3a, 3d, 4 and 5. Catalyst loading of 1%.
while the alkene formation occurs only with phenyl and tolyl- formation of only the two isomers b-(E) and a (Scheme S2, Fig. S10
acetylene substrates.
and S11, ESI†), confirming the importance of both the steric
When BuCRCH is employed once the conversion is com- hindrance and the electronic characteristics of the precursor.
plete, the b-(Z) vinylsilane isomerizes into hex-2-enyl-dimethyl- In order to further investigate how the catalytic performance
n
phenyl-silane (Scheme S1, Fig. S8 and S9, ESI†) instead of b-(E) may be affected by the catalyst loading or by the temperature,
vinylsilane. This behaviour is in line with what was previously the catalytic activity of 3a in the hydrosilylation of phenyl-
reported by Crabtree et al. in the hydrosilylation of 1-hexyne acetylene and tolylacetylene has been examined under the
with HSiMePh2 catalyzed by [RhCl(PPh3)3].30
following conditions: 0.1%, 25 1C; 0.1%, 60 1C. At room tem-
Finally the catalyzed reaction of dimethylphenylsilane with perature the reaction still occurred with both substrates (Fig. S12
(triethylsilyl)acetylene and 1,1-diphenyl-2-propyn-1-ol leads to the for phenylacetylene; Fig. S13 for tolylacetylene, ESI†), but slower
1774 | New J. Chem., 2014, 38, 1768--1779
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
kinetics were found (phenylacetylene was identified in the reac- 7.26 ppm) and 13C NMR (CDCl3: 77.00 ppm) spectra. Spectra have
tion mixture even after 800 h); at the same time selectivity did been edited using the software MestReNova Version: 8.0.2-11021,
not show any significant improvement. Even though the reaction 2012 Mestrelab Research S.L. and unless otherwise stated, they were
1
with tolylacetylene was confirmed to be faster (the conversion recorded at 298 K for characterization purposes; full H and 13C
was complete within 660 h, Fig. S13, ESI†), in both cases a long NMR assignments were done, when necessary, by gCOSY, gHSQC,
induction time is required: after 8 hours no conversion was gHMBC, NOESY and DEPT-135 NMR experiments using standard
observed. Finally, raising the temperature to 60 1C improves the Varian pulse sequences; J. Young valve NMR tubes (Wilmad) were
reaction rates (Fig. S14 and S15, ESI†) but negatively affects, to used to carry out NMR experiments under inert conditions. ESI-MS
some extent, the reaction selectivity.
analyses were performed by direct injection of methanol solutions
of the metal complexes using a WATERS ZQ 4000 mass spectro-
meter. Elemental analyses were performed on a ThermoQuest Flash
1112 Series EA Instrument. The chemicals 1-methylimidazole,
1-benzylimidazole, benzyl bromide, tert-butyl bromide, 1-butyl-
3-methyl-imidazolium bromide, phenylacetylene, tolylacetylene,
1-hexyne, triethylsilylacetylene, 1,1-diphenyl-2-propyn-1-ol, HSi-
Me2Ph and Ag2O were used as purchased from Aldrich;
[Rh(NBD)Cl]2 was purchased from Strem and used as received.
The starting building blocks 1-tert-butylimidazole,31 trityl chloride,2
1-benzyl-3-methyl-imidazolium bromide (1a),4 1,3-dibenzyl-
imidazolium bromide (1b),5 1-benzyl-3-tert-butylimidazolium bro-
mide (1d),6 1-benzyl-3-methyl-imidazolin-2-ylidene silver bromide
(2a),7 1,3-dibenzyl-imidazolin-2-ylidene silver bromide (2b),8
1-benzyl-3-tert-butyl-imidazolin-2-ylidene silver bromide (2d),9 and
[RhCl(NBD){1-(2-NHBoc-ethyl)-3-Me-imidazolin-2-ylidene}] (4)2 were
prepared according to literature procedures and the syntheses are
also reported in the ESI.† Petroleum ether (Etp) refers to a fraction
of bp 60–80 1C. The reactions were monitored by thin-layer
chromatography (TLC) on highly purified silica gel on polyester
(w/UV indicator) and visualized using UV light (254 nm). Column
chromatography was carried out under argon on silica gel pre-
viously heated at about 200 1C while a slow stream of dry nitrogen
was passed through it;32 celite was dried in an oven at 150 1C.
3. Conclusions
We have described the synthesis of the novel rhodium(I) com-
plexes [RhCl(NBD)(NHC)] [NHC
= 1-benzyl-3-R-imidazolin-
t
2-ylidene; R = Me (3a), Bz (3b), Tr (3c), Bu (3d)], in which the
NHC ligands bear a benzyl substituent on one nitrogen and
increasing bulky alkyl substituents on the other. All complexes
display a restricted rotation about the metal–carbene bond. By
using the N-benzyl group as a chiral probe, the racemization
barriers about the Rh–carbene have been determined by
employing the following NMR spectroscopic techniques: VT for
3a,b (56.5 and 62.7 kJ molꢀ1) and EXSY for 3c (62.3 kJ molꢀ1);
for the bulkiest 3d only a lower limit for the rotational barrier
(91 kJ molꢀ1) could be determined. Whilst the calculated rotation
barriers for 3a,b and 3d matched the experimental values, in the
case of 3c the latter (62.3 kJ molꢀ1) was much smaller with respect
to the calculated one (101.7 kJ molꢀ1). This lower experimental
energy barrier, its strong dependence on the temperature, and the
large negative activation entropy derived from simulations (ꢀ115 ꢂ
10 J Kꢀ1 molꢀ1) have been attributed to a partial dissociation of
N-trityl to form a solvated ionic pair in the transition state.
The catalytic activity of the neutral rhodium(I) complexes 3a
and 3d in the hydrosilylation of terminal alkynes with HSiMe2Ph
has been investigated and compared with the amide-functionalized
[RhCl(NBD){1-(2-NHBoc-ethyl)-3-Me-imidazolin-2-ylidene}] (4) and
with [RhCl(NBD){1-butyl-3-Me-imidazolin-2-ylidene}] (5). This study
has demonstrated the beneficial influence on the catalytic activity
of [RhCl(NBD)(NHC)] complexes of an amide functionality on
the side chain, and has confirmed that steric hindrance on the
N-heterocyclic ligand and on the alkyne substrates affects
conversion and selectivity.
4.2. Syntheses
4.2.1. 1-Benzyl-3-trityl-imidazolium chloride (1c). In
a
Schlenk tube to a solution of trityl chloride (0.18 g, 0.64 mmol)
dissolved in CH2Cl2 (10 mL), 1-benzylimidazole (0.10 g,
0.65 mmol) was added. After stirring for 16 h at room tempera-
ture, the solvent was removed under vacuum and the resulting
solid was thoroughly washed with diethyl ether (3 ꢃ 10 mL) to
yield 1c as a white solid (0.25 g, 89%). 1H NMR (300.1 MHz,
CDCl3): d 9.92 (s, 1H, NCHN), 7.65 (s, 1H, CHim), 7.49–7.11
(m, 20H, Ph), 6.93 (s, 1H, CHim), 5.97 (s, 2H, CH2Ph). 13C NMR
(75.5 MHz, CDCl3):
d 139.6 (C5), 138.3 (NCHN), 133.5
4. Experimental section
4.1. Materials and procedures
(Cq, benzyl), 129.6–127.7 (Ph), 123.7 (CHim), 121.8 (CHim), 79.8
(C1), 53.9 (CH2Ph). ESI-MS (MeOH, m/z): 401 (100) [M ꢀ Cl]+; In
All reactions were carried out under argon using standard Schlenk the ESI-MS(-) spectrum, no peaks were observed. Anal. calcd (%)
techniques. Solvents were dried and distilled under nitrogen prior to
use; the deuterated solvents used after being appropriately dried
and degassed were stored in ampoules under argon on 4 Å
molecular sieves. The prepared derivatives were characterized by
elemental analysis and spectroscopic methods. The NMR spectra
were recorded using Varian Inova (1H, 300.1; 13C, 75.5 MHz), Varian
Mercury VX (1H, 399.9; 13C, 100.6 MHz), and Varian Inova
(1H, 599.7, 13C, 150.8 MHz) spectrometers; chemical shifts were
referenced internally to residual solvent peaks for 1H (CDCl3:
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 1768--1779 | 1775

View Article Online
NJC
Paper
for C29H25ClN2: C, 79.71; H, 5.77; N, 6.41. Found: C, 79.67; H, was removed under vacuum. The crude material was purified
5.71; N, 6.39. by column chromatography on silica gel treated with 5% v/v
4.2.2. 1-Benzyl-3-trityl-imidazolin-2-ylidene silver bromide triethylamine in diethyl ether, using Et2O–CH2Cl2, 1 : 1 v/v, to
(2c). To a solution of 1c (0.23 g, 0.53 mmol) in CH2Cl2 (ca. 10 mL) afford 3c as a yellow solid (0.28 g, 70%). Rf: 0.55 (Et2O–CH2Cl2,
stirred in a Schlenk tube, Ag2O (0.07 g, 0.30 mmol) was added. 50 : 50). 1H NMR (300.1 MHz, CDCl3): d 7.45–7.24 (m, 20H, Ph),
The suspension was stirred for 48 h at room temperature and 6.84 (d, 1H, 2JH,H = 14.9 Hz, CH2Ph), 6.76 (d, 1H, 3JH,H = 2.0 Hz,
3
2
then filtered on celite; the colourless filtrate was characterised by CHIm), 6.56 (d, 1H, JH,H = 2.0 Hz, CHIm), 6.17 (d, 1H, JH,H
=
NMR and then straightforwardly used for the preparation of the 14.9 Hz, CH2Ph), 4.61 (br s, 1H, NBD), 3.59 (br s, 1H, NBD), 3.44
corresponding rhodium complex 3c. 1H NMR (300.1 MHz, (br s, 1H, NBD), 3.10 (br s, 1H, NBD), 2.87 (br s, 1H, NBD), 1.94
3
CDCl3): d 7.34–7.21 (m, 20H, Ph), 7.03 (d, 1H, JH,H = 1.9 Hz, (br s, 1H, NBD), 0.87 (m, 2H, NBD). 13C NMR (75.5 MHz,
CHim), 6.91 (d, 1H, 3JH,H = 1.9 Hz, CHim), 5.25 (s, 2H, CH2Ph). 13
C
CDCl3): d 186.6 (d, JC,Rh = 55.1 Hz), 142.4 (C5), 137.5 (Cq, benzyl),
NMR (75.5 MHz, CDCl3): d 185.1 (d, JAg,C = 236.0 Hz, C–Ag), 141.9 129.1 (Ph), 128.9 (Ph), 128.3 (Ph), 128.1 (Ph), 127.9 (Ph), 127.5
(C5), 135.4 (Cq, benzyl), 129.8 (Ph), 128.9 (Ph), 128.4 (Ph), 128.2 (Ph), 127.2 (Ph), 124.3 (CHim), 120.7 (CHim), 76.6 (C1), 73.1
(Ph), 127.8 (Ph), 127.7 (Ph), 127.6 (Ph), 127.0 (Ph), 123.8 (CHim), (s, NBD), 61.8 (d, JC,Rh = 5.8 Hz, NBD), 57.2 (CH2Ph), 49.7 (NBD),
118.9 (CHim), 77.7 (C1), 56.8 (CH2Ph).
49.0 (NBD), 45.7 (d, JC,Rh = 13.0 Hz, NBD), 44.74 (d, JC,Rh =
4.2.3. [RhCl(NBD){1-benzyl-3-methyl-imidazolin-2-ylidene}] 10.8 Hz, NBD). ESI-MS (MeOH, m/z): 595 (100) [M – Cl]+. Anal.
(3a). The in situ prepared silver complex 2a was added to a calcd (%) for C36H32ClN2Rh: C, 68.52; H, 5.11; N, 4.44. Found:
solution of [Rh(NBD)Cl]2 (0.12 g, 0.26 mmol) in CH2Cl2. After C, 68.39; H, 5.21; N, 4.44.
stirring for 3 h at room temperature, the pale yellow AgBr was
4.2.6. [Rh(NBD)Cl{1-benzyl-3-tert-butyl-imidazolin-2-ylidene}]
filtered off, and the solvent was removed under vacuum. The (3d). The in situ prepared silver complex 2d was added to a solution
crude material was purified by column chromatography on silica of [Rh(NBD)Cl]2 (0.25 g, 0.54 mmol) in CH2Cl2. After stirring for 4 h
gel using first CH2Cl2 and then CH2Cl2–MeOH [100 : 3 (v/v)] as the pale yellow AgBr was filtered off, and the solvent was removed
eluent to afford 3a as a yellow solid (0.15 g, 72%). Rf: 0.33 under vacuum. The crude material was purified by column
(CH2Cl2/MeOH, 100 : 3). 1H NMR (300.1 MHz, CDCl3): d 7.35 chromatography on silica gel treated with 5% v/v triethylamine
(m, 5H, Ph), 6.76 (s, 1H, CHim), 6.65 (s, 1H, CHim), 5.72 (s, 2H, in diethyl ether, using first CH2Cl2 and then CH2Cl2–MeOH
CH2Ph), 4.84 (s, 2H, NBD), 4.08 (s, 3H, CH3), 3.72 (s, 2H, NBD), [100 : 2 (v/v)] as eluent to afford 3c as a yellow solid (0.36 g,
3.36 (br s, 2H, NBD), 1.30 (m, 2H, NBD). 13C NMR (75.5 MHz, 75%). Rf: 0.51 (CH2Cl2/MeOH, 100 : 7). 1H NMR (599.7 MHz,
CDCl3): d 184.8 (d, JC,Rh = 57.8 Hz), 136.9 (Cq, Ph), 128.8 (Ph), CDCl3): d 7.46–7.45 (m, 4H, Ph), 7.41– 7.39 (m, 1H, Ph), 7.06
128.3 (Ph), 128.1 (Ph), 122.2 (CHim), 120.7 (CHim), 79.0 (d, JC,Rh
=
(d, 1H, 3JH,H = 2.0 Hz, CHim), 6.72 (d, 1H, 3JH,H = 2.0 Hz, CHim),
6.0 Hz, NBD), 63.4 (d, JC,Rh = 5.2 Hz, NBD), 54.3 (CH2Ph), 51.0 6.41 (d, 1H, 2JH,H = 15.1 Hz, CH2Ph), 6.27 (d, 1H, 2JH,H = 15.1 Hz,
(d, JC,Rh = 2.5 Hz, NBD), 48.3 (d, JC,Rh = 12.8 Hz, NBD), 37.8 (CH3). CH2Ph), 4.84 (m, 1H, NBD), 4.70 (m, 1H, NBD), 3.81 (br s, 1H,
ESI-MS (MeOH, m/z): 367 (30) [M ꢀ Cl]+; 539 (100) [Rh(NBD)- NBD), 3.78 (br s, 1H, NBD), 3.47 (m, 2H, NBD), 2.08 (s, 9H, tBu),
(NHC)2]+. Anal. calcd (%) for C18H20ClN2Rh: C, 53.68; H, 5.01; N, 1.36 (d, 1H, JH,H = 8.0 Hz, NBD), 1.31 (d, 1H, JH,H = 8.0 Hz,
6.96. Found: C, 53.59; H, 5.08; N, 6.94.
4.2.4. [RhCl(NBD){1,3-dibenzyl-imidazolin-2-ylidene}] (3b). 136.9 (Cq, Ph), 128.7 (Ph), 128.0 (Ph), 127.8 (Ph), 119.7 (CHim),
The in situ prepared silver complex 2b was added to a solution 119.6 (CHim), 74.9 (d, JC,Rh = 6.1 Hz, NBD), 71.8 (d, JC,Rh
2
2
NBD). 13C NMR (150.8 MHz, CDCl3): d 183.0 (d, JC,Rh = 58.5 Hz),
=
t
of [Rh(NBD)Cl]2 (0.14 g, 0.30 mmol) in CH2Cl2. After stirring for 6.3 Hz, NBD), 62.7 (d, JC,Rh = 5.2 Hz, NBD), 57.9 (Cq, Bu), 56.3
3 h the pale yellow AgBr was filtered off, and the solvent was (CH2Ph), 48.2 (d, JC,Rh = 12.7 Hz, NBD), 46.3 (d, JC,Rh = 12.1 Hz,
removed under vacuum. The crude material was purified by NBD), 32.1 (tBu). ESI-MS (MeOH, m/z): 409 (100) [M ꢀ Cl]+. Anal.
column chromatography on silica gel using first CH2Cl2 and calcd (%) for C21H26ClN2Rh: C, 56.70; H, 5.89; N, 6.30. Found:
then CH2Cl2–MeOH [100 : 3 (v/v)] as eluent to afford 3b as a C, 56.51; H, 5.87; N, 6.41.
yellow solid (0.25 g, 86%). Rf: 0.45 (CH2Cl2/MeOH, 100 : 3).
4.2.7. [RhCl(NBD){1-butyl-3-methyl-imidazolin-2-ylidene}]
1H NMR (300.1 MHz, CDCl3): d 7.31 (m, 10H, Ph), 6.61 (s, 2H, (5). To a solution of 1-butyl-3-methylimidazolium bromide
CHim), 5.73 (br s, 4H, CH2Ph), 4.78 (br s, 2H, NBD), 3.60 (br s, (0.10 g, 0.46 mmol) in CH2Cl2 (10 mL), Ag2O (0.058 g, 0.25 mmol)
2H, NBD), 3.20 (br s, 2H, NBD), 1.20 (br s, 2H, NBD). 13C NMR was added. The suspension was stirred for 48 h then, solid
(75.5 MHz, CDCl3): d 185.4 (d, JC,Rh = 58.3 Hz), 136.9 (Cq, Ph), [Rh(NBD)Cl]2 (0.10 g, 0.22 mmol) was directly added to the
128.9 (Ph), 128.6 (Ph), 128.03 (Ph), 121.1 (CHim), 120.72 (CHim), reaction mixture. After stirring for further 24 h the crude
79.1 (d, JC,Rh = 5.7 Hz, NBD), 63.3 (d, JC,Rh = 5.3 Hz, NBD), 54.5 material was filtered on a celite pad, the insoluble material
(CH2Ph), 51.0 (d, JC,Rh = 2.0 Hz, NBD), 48.5 (d, JC,Rh = 17.0 Hz, was thoroughly washed with CH2Cl2 and the solvent removed
NBD). ESI-MS (MeOH, m/z): 443 (100) [M – Cl]+. Anal. calcd (%) under vacuum. The residue was purified by column chromato-
for C24H24ClN2Rh: C, 60.20; H, 5.05; N, 5.85. Found: C, 60.15; H, graphy on silica gel using Et2O : CH2Cl2 1 : 1 (v/v) to afford 5 as a
5.09; N, 5.93.
yellow solid (0.13 g, 80%). Rf: 0.68 (CH2Cl2 : MeOH 100 : 5).
3
4.2.5. [RhCl(NBD){1-benzyl-3-trityl-imidazolin-2-ylidene}] 1H NMR (300.1 MHz, CDCl3): d 6.76 (d, 1H, JH,H = 1.7 Hz,
(3c). The in situ prepared silver complex 2c was added to a CHim), 6.75 (d, 1H, 3JH,H = 1.7 Hz, CHim), 4.84 (s, 2H, NBD), 4.47
3
solution of [Rh(NBD)Cl]2 (0.15 g, 0.32 mmol) in CH2Cl2. After (t, 2H, JH,H = 7.5 Hz, NCH2), 4.07 (s, 3H, NCH3) 3.93 (br s, 2H,
stirring for 3 h the white AgCl was filtered off, and the solvent NBD), 3.47 (br s, 2H, NBD), 1.91 (m, 2H, CH2), 1.35 (m, 2H, CH2),
1776 | New J. Chem., 2014, 38, 1768--1779
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
1.20 (s, 2H, NBD), 1.05 (t, 3H, 3JH,H = 7.2 Hz, CH3). 13C NMR (75.5 activation entropy DS‡. 1D-EXSY spectra were recorded using
MHz, CDCl3): d 181.7 (d, JC–Rh = 57.5 Hz), 120.9 (CHim), 119.2 the standard DPFGSE-NOE sequence, using arrayed mixing
(CHim), 77.1 (d, JC–Rh = 6.0 Hz, NBD), 62.8 (d, JC–Rh = 5.1 Hz, times (25, 50, 75 and 100 ms).
NBD), 50.1 (NBD), 49.3 (NCH2), 36.7 (NCH3), 32.4 (CH2), 19.0
4.5. DFT calculations
(CH2), 12.8 (CH3). Anal. calcd (%) for C15H22ClN2Rh: C, 48.86; H,
6.01; N, 7.60. Found: C, 48.97; H, 5.93; N, 7.63.
Conformational searches were performed by Molecular
Mechanics (MMFF force field as implemented in Titan 1.0.5,
Wavefunction Inc.). Final geometry optimizations were carried
out at the B3LYP/LANL2DZ level36 by means of the Gaussian 09
series of programs37 (see the ESI†). The standard Berny algo-
rithm in redundant internal coordinates and default criteria of
convergence were employed in all the calculations. Harmonic
vibrational frequencies were calculated for all the stationary
points. For each optimized ground state the frequency analysis
showed the absence of imaginary frequencies, whereas each
transition state showed a single imaginary frequency. Visual
inspection of the corresponding normal mode was used to
confirm that the correct transition state had been found. All
the reported energy values represent total electronic energies.
In general, these give the best fit to experimental DNMR data.38
Therefore, the computed numbers have not been corrected for
zero-point energy contributions or other thermodynamic para-
meters. This avoids artefacts that might result from the ambig-
uous choice of the adequate reference temperature, from the
empirical scaling39 and from the treatment of low-frequency
vibration as harmonic oscillators.40
4.3. X-ray structure determination for 2b, 3a and 3b
Crystal data were collected at room temperature on a Bruker
APEX II diffractometer equipped with a CCD detector operating
at 50 kV and 30 mA, using graphite-monochromated Mo Ka
radiation (l = 0.71073 Å). An empirical absorption correction
was applied to both structures by using SADABS.33 They
were solved by direct methods and refined by full-matrix
least-squares based on all data using F2 with SHELXL97.34 All
non-hydrogen atoms were refined anisotropically, with the
exception of the hydrogen atoms, which were set geometrically
and given fixed isotropic thermal parameters. Crystal data are
collected in Table S3 (ESI†).
4.4. Variable temperature NMR
VT spectra were recorded using spectrometers at 400 and
600 MHz for 1H. The NMR tubes containing the compounds
were prepared under an argon atmosphere using a vacuum
1
line. Low-temperature H spectra were recorded without spin-
ning using a 5 mm dual direct probe with 6000 Hz (at 400 MHz)
or 9000 Hz (at 600 MHz) sweep width, 401 tip angle pulse width,
3 s acquisition time, and 1 s delay time. A shifted sine bell
weighting function equal to the acquisition time (i.e., 3 s) was
applied before the Fourier transformation. Usually 32 to 64
scans were collected. When operating the NMR apparatus at
low temperature, a stream of dry nitrogen was first passed
through a precooling unit adjusted to ꢀ50 1C. Then the gas
entered into an inox steel heat-exchanger immersed in liquid
nitrogen and connected to the NMR probe head by a vacuum-
insulated transfer line. Gas flows of 10 to 20 L minꢀ1 were
required to reach the desired temperature. Temperature cali-
brations were performed before the experiments, using a digital
thermometer and a Cu/Ni thermocouple placed in an NMR
tube filled with isopentane for the low-temperature range and
with 1,1,2,2-tetrachloroethane for the high-temperature range.
The conditions were kept as identical as possible for all sub-
sequent work. In particular, the sample was not spun and the
gas flow was the same as that used during the acquisition of the
spectra. The uncertainty in temperature measurements can be
estimated from the calibration curve as ꢂ2 1C. Line shape
simulations were performed using a PC version of the QCPE
DNMR6 program.35 Electronic superimposition of the original
spectrum and of the simulated one enabled the determination
of the most reliable rate constant. The rate constants, thus
obtained at various temperatures, afforded the free energy of
activation DG‡ by applying the Eyring equation. DH‡ and DS‡
were evaluated by linear regression of the DG‡ value vs. the
temperature. Except for 3c, in all cases investigated, the activa-
tion energy DG‡ was found to be virtually invariant in the given
temperature range, thus implying a very small or negligible
4.6. Catalysis
4.6.1. General procedure for the hydrosilylation of 1-alkynes
with HSiMe2Ph. In a typical experiment a J. Young valve NMR
tube was charged under argon with the catalyst precursor 3a, 3d
or 5 (7.7 ꢃ 10ꢀ4 mmol or 7.7 ꢃ 10ꢀ5 mmol), CDCl3 (0.6 mL), the
corresponding alkyne (0.077 mmol), and a slight excess of
HSiMe2Ph (0.085 mmol). The solution was kept at T = 25 1C or
at 60 1C and monitored by 1H-NMR spectroscopy. The products
1
were unambiguously characterized by H NMR spectroscopy by
comparison with literature spectra,2 further details on the NMR
spectra and integrations are reported in the ESI.†
Acknowledgements
The authors wish to thank the University of Bologna and the
`
Ministero dell’Universita e della Ricerca (MUR) (project: ‘‘New
strategies for the control of reactions: interactions of molecular
fragments with metallic sites in unconventional species’’, PRIN
2009) for financial support. We also thank Dr R. Cirilli (Istituto
`
Superiore di Sanita, Department of Therapeutic Research and
Medicines Evaluation, Rome) for HPLC analyses.
References
1 For reviews on N-heterocyclic carbenes see: (a) L. A. Schaper,
S. J. Hock, W. A. Herrmann and F. E. Ku¨hn, Angew. Chem.,
Int. Ed., 2013, 52, 270; (b) L. Benhamou, E. Chardon,
´
G. Lavigne, S. Bellemin-Laponnaz and V. Cesar, Chem. Rev.,
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 1768--1779 | 1777

View Article Online
NJC
2011, 111, 2705; (c) L. Mercs and M. Albrecht, Chem. Soc. Rev.,
Paper
T. Ramnial, O. C. Lightbody, C. L. B. Macdonald, J. A. C.
Clyburne, C. D. Abernethy and S. P. Nolan, Organometallics,
2005, 24, 6301.
2010, 39, 1903; (d) S. Diez-Gonzalez, N. Marion and
S. P. Nolan, Chem. Rev., 2009, 109, 3612; (e) H. Jacobsen,
A. Correa, A. Poater, C. Costabile and L. Cavallo, Coord. Chem. 11 J. Li, J. Peng, Y. Bai, G. Lai and X. Li, J. Organomet. Chem.,
Rev., 2009, 253, 687; ( f ) F. E. Hahn, ed. of themed issue: 2011, 696, 2116.
N-Heterocyclic carbenes, Dalton Trans., 2009, 35, 6873; 12 A. Bittermann, P. Harter, E. Herdtweck, S. D. Hoffmann and
(g) F. E. Hahn and M. C. Jahnke, Angew. Chem., Int. Ed., W. A. Herrmann, J. Organomet. Chem., 2008, 693, 2079.
2008, 47, 3128; (h) R. H. Crabtree, Coord. Chem. Rev., 2007, 13 For a review see: D. Casarini, L. Lunazzi and A. Mazzanti,
251, 595 ff.: a volume completely dedicated to recent devel- Eur. J. Org. Chem., 2010, 2035.
opments in the organometallic chemistry of N-heterocyclic 14 Due to the identity of the two substituents, compound 3b
carbenes; (i) C. M. Crudden and D. P. Allen, Coord. Chem.
actually belongs to the Cs symmetry.
Rev., 2004, 248, 2247; ( j) W. A. Herrmann, Angew. Chem., Int. 15 Despite the Cs symmetry of 3b, the NMR signal of CH2 is
¨
Ed., 2002, 41, 1290; (k) D. Bourissou, O. Guerret, F. P. Gabbai
and G. Bertrand, Chem. Rev., 2000, 100, 39.
2 L. Busetto, M. C. Cassani, C. Femoni, M. Mancinelli,
still a stereochemical probe because the local plane of
symmetry does not correspond to the symmetry plane of
the molecule. See W. B. Jennings, Chem. Rev., 1975, 75, 307.
A. Mazzanti, R. Mazzoni and G. Solinas, Organometallics, 16 (a) D. Enders, H. Gielen, J. Runsink, K. Breuer, S. Brode and
2011, 30, 5258.
K. Boehn, Eur. J. Inorg. Chem., 1998, 913; (b) D. Enders and
H. Gielen, J. Organomet. Chem., 2001, 617, 70.
3 For recent reports on Rh(I)(NHC) catalysed hydrosilylation
and hydrothiolation, see: (a) M. V. Jimenez, J. J. Perez- 17 The 1D EXSY experiment was acquired using the standard
`
Torrente, M. I. Bartolome, V. Gierz, F. J. Lahoz and
DPFGSE-NOE sequence with short mixing times;
(a) L. Lunazzi, M. Mancinelli, A. Mazzanti and M. Pierini,
J. Org. Chem., 2010, 75, 5927; (b) L. Lunazzi, M. Mancinelli
and A. Mazzanti, J. Org. Chem., 2012, 77, 3373.
L. A. Oro, Organometallics, 2008, 27, 224; (b) B. J. Truscott,
A. M. Z. Slawin and S. P. Nolan, Dalton Trans., 2013, 42, 270;
(c) A. Monney, F. Nastri and M. Albrecht, Dalton Trans.,
2013, 42, 5655; (d) M. Steinbeck, G. D. Frey, W. W. Schoeller 18 L. Lunazzi, G. Panciera and M. Guerra, J. Chem. Soc., Perkin
and W. A. Herrmann, J. Organomet. Chem., 2011, 696, 3945; Trans. 2, 1980, 52.
(e) J. J. Dunsford, D. S. Tromp, K. J. Cavell, C. J. Elsevier and 19 H. Iwamura and K. Mislow, Acc. Chem. Res., 1988, 21, 175.
B. M. Kariuki, Dalton Trans., 2013, 42, 7318; ( f ) 20 C. Boga, E. Del Vecchio, L. Forlani, A. Mazzanti and
´
A. Di Giuseppe, R. Castarlenas, J. J. Perez-Torrente,
M. Crucianelli, V. Polo, R. Sancho, F. J. Lahoz and 21 H. Hedayat, A. Naader, S. Mojtaba and A. Zouhair, J. Phys.
L. A. Oro, J. Am. Chem. Soc., 2012, 134, 8171.
Chem. A, 2010, 114, 7462.
4 S. Patil, A. Deally, B. Gleeson, M. Helge, F. Paradisi and 22 (a) C. Wolf, Chem. Soc. Rev., 2005, 34, 595; (b) I. D’Acquarica,
P. E. Todesco, Angew. Chem., Int. Ed., 2005, 44, 3285.
M. Tacke, Appl. Organomet. Chem., 2010, 24, 781.
5 S. Patil, J. Claffey, A. Deally, M. Hogan, B. Gleeson,
F. Gasparrini, M. Pierini, C. Villani and G. Zappia, J. Sep.
Sci., 2006, 29, 1508; (c) F. Piron, N. Vanthuyne, B. Joulin,
J.-V. Naubron, C. Cismas, A. Terec, R. A. Varga, C. Roussel,
J. Roncali and I. Grosu, J. Org. Chem., 2009, 74, 9062.
´
´
L. M. Menendez Mendez, H. Mu¨ller-Bunz, F. Paradisi and
M. Tacke, Eur. J. Inorg. Chem., 2010, 1020.
6 In the case of 1d the alkylation of 1-benzylimidazole with 23 W. Gil, T. Lis, A. M. Trzeciak and J. J. Ziołkowski, Inorg.
1-tert-butyl bromide does not afford the desired product due Chim. Acta, 2006, 359, 2835.
to the occurrence of an elimination reaction, in which tert- 24 TOF calculated for a reaction time of 2 h which corresponds
butyl bromide in the presence of 1-benzylimidazole releases to complex 5 with a conversion of 54%.
a molecule of HBr (with evolution of isobutene gas) which 25 (a) A. T. Normand and K. J. Cavell, Eur. J. Inorg. Chem., 2008,
¨hl, Chem. Soc. Rev., 2007, 36, 592; (c) D. Pugh
further reacts with benzylimidazole to form the corres-
2781; (b) O. Ku
´
ponding imidazolium bromide salt. See: R. Corberan,
M. Sanau´ and E. Peris, Organometallics, 2006, 25, 4002.
and A. A. Danopoulos, Coord. Chem. Rev., 2007, 251, 610;
(d) S. T. Liddle, I. S. Edworthy and P. L. Arnold, Chem. Soc.
Rev., 2007, 36, 1732; (e) P. L. Arnold, S. A. Mungur,
A. J. Blake and C. Wilson, Angew. Chem., Int. Ed., 2003,
¨
7 C. Hirtenlehner, C. Krims, J. Holbling, M. List, M. Zabel,
M. Fleck, R. J. F. Berger, W. Schoefberger and U. Monkowius,
Dalton Trans., 2011, 40, 9899.
¨
42, 5981; ( f ) W. A. Herrmann, C. Kocher, L. J. Gooßen and
8 The silver complex 2b has also been characterized by X-ray
diffraction (see Table S3 and Fig. S2, ESI†) and the struc-
tural data are in perfect agreement (within the experimental
G. R. J. Artus, Chem.–Eur. J., 1996, 2, 1627; (g) H. Ohara,
W. N. O. Wylie, A. J. Lough and R. H. Morris, Dalton Trans.,
2012, 41, 8797.
`
error) with those reported in C. P. Newman, G. J. Clarkson 26 (a) M. Poyatos, E. Mas-Marza, J. A. Mata, M. Sanau and E. Peris,
and J. P. Rourke, J. Organomet. Chem., 2007, 692, 4962.
9 S. Ray, R. Mohan, J. K. Singh, M. K. Samantaray, M. M.
Eur. J. Inorg. Chem., 2003, 1215; (b) I. Ojima, N. Clos, R. J.
Donovan and P. Ingallina, Organometallics, 1990, 9, 3127.
`
Shaikh, D. Panda and P. Ghosh, J. Am. Chem. Soc., 2007, 27 G. de Bo, G. Berthoz-Gelloz, B. Tinant and I. E. Marko,
129, 15042. Organometallics, 2006, 25, 1881.
10 (a) J. C. Garrison and W. J. Youngs, Chem. Rev., 2005, 28 A. K. Roy, Adv. Organomet. Chem., 2008, 55, 1 and references
´
105, 3978; (b) P. de Fremont, N. M. Scott, E. D. Stevens,
therein.
1778 | New J. Chem., 2014, 38, 1768--1779
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014

View Article Online
Paper
NJC
29 Z. T. Ball, in Comprehensive Organometallic Chemistry III,
Application II: Transition metal Compounds in Organic syn-
thesis 1, ed. R. H. Crabtree and M. Mingos, Elsevier, Oxford,
3rd edn, 2006, vol. 10, pp. 789–813.
30 C. H. Jun and R. H. Crabtree, J. Organomet. Chem., 1993,
447, 177.
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin,
V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi,
N. Rega, N. J. Millam, M. Klene, J. E. Knox, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli,
J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski,
G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D.
31 S. Sauerbrey, P. K. Majhi, J. Daniels, G. Schnakenburg,
¨
G. M. Brandle, K. Scherer and R. Streubel, Inorg. Chem.,
2011, 50, 793.
32 D. D. Perrin, W. L. F. Armarego and D. R. Perrin, Purification
of Laboratory Chemicals, Pergamon Press, New York, 2nd
edn, 1980.
33 G. M. Sheldrick, SADABS, University of Gottingen, Germany, 1996.
34 G. M. Sheldrick, SHELXL97, University of Gottingen, Germany.
35 The program is available from the authors on request.
36 For B3LYP functional: (a) A. D. Becke, J. Chem. Phys., 1993,
¨
Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski
and D. J. Fox, Gaussian 09 rev A.02, Gaussian, Inc., Wallingford
CT, 2009.
98, 5648; (b) C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: 38 P. Y. Ayala and H. B. Schlegel, J. Chem. Phys., 1998,
Condens. Matter Mater. Phys., 1988, 37, 785; (c) P. J. 108, 2314.
Stephens, F. J. Devlin, C. F. Chabalowski and M. J. Frisch, 39 C. F. Tormena, R. Rittner, R. J. Abraham, E. A. Basso and
J. Phys. Chem., 1994, 98, 11623. For the LanL2DZ basis set:
P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 299.
37 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. C. Fiorin, J. Phys. Org. Chem., 2004, 17, 42.
40 (a) M. W. Wong, Chem. Phys. Lett., 1996, 256, 391;
(b) S. E. Wheeler, A. J. McNeil, P. Mu¨ller, T. M. Swager
and K. N. Houk, J. Am. Chem. Soc., 2010, 132, 3304.
This journal is ©The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2014
New J. Chem., 2014, 38, 1768--1779 | 1779