Irreversible cleavage of a carbene-rhodium bond in Rh-N-heterocyclic carbene complexes: Implications for catalysis

Article abstract of DOI:10.1016/j.jorganchem.2004.07.021

Despite the generally accepted belief that carbene-metal bonds are strong and do not dissociate, the reaction of Rh-N-heterocyclic carbene complexes with triphenylphosphine in dichloroethane was determined to take place via cleavage of the Rh-carbene bond. The products of this reaction are Wilkinson's catalyst and a bisimidazolium salt derived from reaction between dichloroethane and two equivalents of the carbene. The implications of this reaction for catalysis are significant since the carbene complex shows lower activity than Wilkinson's catalyst in hydrogenation reactions. In non-halogenated solvents, the catalyst shows higher stability, such that the rate of exchange with free phosphine could be measured, and was determined to be ca. 10 times slower than in Wilkinson's catalyst.

Full text of DOI:10.1016/j.jorganchem.2004.07.021

Journal of Organometallic Chemistry 689 (2004) 3203–3209
www.elsevier.com/locate/jorganchem
Irreversible cleavage of a carbene–rhodium bond in
Rh-N-heterocyclic carbene complexes: implications for catalysis
a
a,
b
a,1
Daryl P. Allen , Cathleen M. Crudden *, Larry A. Calhoun , Ruiyao Wang
a
Department of Chemistry, Queenꢀs University, Kingston, Ont., Canada K7L 3N6
Department of Chemistry, University of New Brunswick, Fredericton, New Brunswick, Canada E3B 6E2
b
Received 12 May 2004; accepted 12 July 2004
Available online 19 August 2004
Abstract
Despite the generally accepted belief that carbene–metal bonds are strong and do not dissociate, the reaction of Rh-N-heterocy-
clic carbene complexes with triphenylphosphine in dichloroethane was determined to take place via cleavage of the Rh–carbene
bond. The products of this reaction are Wilkinsonꢀs catalyst and a bisimidazolium salt derived from reaction between dichloroethane
and two equivalents of the carbene. The implications of this reaction for catalysis are significant since the carbene complex shows
lower activity than Wilkinsonꢀs catalyst in hydrogenation reactions. In non-halogenated solvents, the catalyst shows higher stability,
such that the rate of exchange with free phosphine could be measured, and was determined to be ca. 10 times slower than in Wil-
kinsonꢀs catalyst.
ꢀ 2004 Elsevier B.V. All rights reserved.
Keywords: N-heterocyclic carbene; Catalysis; Hydrogenation; Rhodium complex
1. Introduction
and co-workers [4] have clearly demonstrated that
carbene ligands can suffer decomposition by reductive
elimination and both C–H and C–C activation of the
2,4,6-trimethylphenyl substituents have been reported
[5]. Two reports of equilibration of the carbene–metal
complexes with free carbene have recently appeared
[6], which, along with early results from Lappertꢀs lab
[7], challenge the commonly held view that carbenes
are substitutionally inert [8]. We report herein the irre-
versible displacement of a carbene ligand from rhodium
complexes which leads to the formation of a completely
different species, with higher catalytic activity. These
findings have significant implications for catalysis, since
the transformation of a carbene complex into a simple
phosphine-containing species can lead to incorrect
assumptions regarding the activity of the original car-
bene-ligated complex. At the same time, our studies shed
light on a debate in the literature over the reactivity of
N-heterocyclic carbenes with halogenated hydrocarbons
such as dichloroethane and dichloromethane [9].
The use of N-heterocyclic carbenes (NHCs) as
replacements for phosphines has provided some signifi-
cant improvements over traditional phosphine-based
ligands [1]. The most commonly employed N-heterocy-
clic carbene, 1,3-bis (2,4,6-trimethylphenyl)-imidazol-2-
ylidene (IMes), and its congeners are considered to be
equivalent to strongly electron rich phosphines such as
PCy3, but with the added benefits of increased catalyst
stability [2] and decreased ligand lability [1]. Significant
studies of the stoichiometric reactivity of isolated car-
bene complexes are now beginning to accompany the
numerous catalytic studies of these species [3]. Cavell
*
Corresponding author. Tel.: +613 5336755; fax: +613 5336669.
E-mail address: cruddenc@chem.queensu.ca (C.M. Crudden).
1
Author to whom correspondence concerning X-ray crystallogra-
phy should be addressed.
0022-328X/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.07.021

3204
D.P. Allen et al. / Journal of Organometallic Chemistry 689 (2004) 3203–3209
Table 1
Phosphine exchange in complex 2ba
2. Results and discussion
Entry Equivalents of PR3 Temperature (ꢁC) Exchange rate (sꢀ1
)
2.1. NMR studies of phosphine dissociation
1
2
5.6
5.5
50
60
60
60
60
60
60
60
60
70
0.35
0.90
0.90
1.00
0.85
0.95
0.89
0.95
0.87
2.40
Building on the seminal work of Lappert and co-
workers [10] and Herrmann and Kocher [11], we re-
cently introduced an analog of Wilkinsonꢀs catalyst (1)
in which one of the phosphines was replaced by an N-
heterocyclic carbene ligand (2) [12]. Carbonyl complex
3 was prepared by treating 2 with carbon monoxide
(Scheme 1). The resulting CO complexes can be handled
in air and they are robust enough to be purified by chro-
matography on silica gel. Carbonyl complex (3) was
found to be more active and more selective than
[ClRh(CO)(PPh3)2] in the hydroformylation of vinyl
arenes [12]. Lebel and co-workers [13a] examined the
activity of 2a in several catalytic reactions, and found
that in methenylation reactions [13b] the carbene com-
plex was significantly less active and also displayed an
induction period.
Unlike the large majority of [ClRh(PAr3)2L] type
complexes, rhodium complex 2 has the phosphines ar-
ranged in a cis-fashion in both solution and solid states
[14,15]. Since the carbene is believed to be a strong r-do-
nor, dissociation of the trans-phosphine, which is impor-
tant for the catalytic activity of such complexes, should
be facilitated. Contrary to these expectations, when we
examined the exchange of free phosphine with bound
phosphine in a selective inversion transfer experiment,
we found that the dissociation of the trans-phosphine in
2a and 2b is slower than in 1a/b by at least an order of
magnitude [16].
3
50.4
23.3
6.4
4
5
6
3.8
3.5
7
8
7.5
9
10
19.3
3.6
a
Exchange experiments carried out at the indicated temperature in
dry distilled d8-toluene, [2b] = 0.04 using delay time at
M
a
least = 5 · T1. Solutions were prepared in a glove box under argon and
are in sealed tubes or NMR tubes fitted with J–Young valves. See
supplementary material for full experimental details.
sꢀ1 (average of eight runs). With a constant concentra-
tion of 2b, increasing the amount of free P(p-tolyl)3 to
50 equivalents gave an exchange rate of 0.90 sꢀ1, indi-
cating that the exchange is likely dissociative (compare
entries 2 and 3).
These results are in line with the recent reports of
Grubbs et al. [19] who have shown that in Ru NHC
complexes, dissociation of phosphine ligands is signifi-
cantly decreased relative to the all phosphine systems.
To the best of our knowledge, our results represent the
only other study of the effect of N-heterocyclic carbene
ligands on phosphine dissociation. Interestingly, despite
the significant differences between the two systems, the
effect of the NHC on the trans-phosphine is roughly
equivalent.
The results obtained in our inversion transfer experi-
ments are described in Table 1. At room temperature,
carbene complexes 2a, 2b and 3 showed no evidence of
exchange even though Wilkinsonꢀs complex (1a) and
the tolyl version (1b), were found to exchange with free
phosphine at a rate of 0.3 sꢀ1 at 25 ꢁC [17]. Upon heat-
ing to 50 ꢁC or greater, catalysts 2a and 2b did undergo
slow exchange, but 3 did not exchange with free phos-
phine even at 80 ꢁC. At 50 ꢁC, 2b undergoes exchange
with free phosphine (3 eq., 0.13 M) at a rate of 0.35
2.2. Stability of the Rh–NHC bond
During this study, the limited solubility of 2a in aro-
matic solvents led us to examine several other solvents.
Although dichloroethane appeared to be a good solvent
at room temperature, upon heating to 60 ꢁC to measure
the rate of exchange, we observed the disappearance of
signals attributed to 2a, and the appearance of signals
consistent with the formation of Wilkinsonꢀs catalyst
(1) (Eq. (1))! Similar reactivity was observed for complex
2b. Both complexes 2a and 2b were significantly more
s
ꢀ1. Only the upfield signal assigned to the phosphine
trans to the carbene was affected, indicating that ex-
change occurred exclusively into this position [18]. At
60 ꢁC, the rate of exchange was determined to be 0.91
IMes
Cl
Ar3P
Ar3P
Rh
PAr3
N
N
N
N
1a (Ar = Ph)
1b (Ar = p-Tolyl)
Ar3P Rh Cl
OC Rh Cl
PAr3
CO
IMes
PAr3
[Rh(H2C=CH2)2Cl]2
2a,b
2 PAr3
3a,b
Scheme 1. Preparation of carbene complexes 2 and 3.

D.P. Allen et al. / Journal of Organometallic Chemistry 689 (2004) 3203–3209
3205
Table 2
Crystal data and structure refinement for dication 6
stable in aromatic solvents such as toluene, although
Wilkinsonꢀs complex does form after extended reaction
times (t1/2 = 3 weeks at 80 ꢁC) [20]. Carbonyl complex
3 displayed enhanced stability relative to the more
crowded and electron rich bisphosphine complexes 2,
with no evidence of decomposition or loss of the carbene
ligand even after heating to 80 ꢁC for 100 hrs in
dichloroethane.
6
Empirical formula
Formula weight
Crystal system
Space group
C52H68Cl10N4 (C44H52Cl2N4 Æ 2(CH2)2Cl2)
1103.60
Triclinic
ꢀ
P1
9.266(4)
˚
a (A)
˚
b (A)
11.294(5)
14.482(6)
˚
c (A)
Cl
Cl
a (ꢁ)
b (ꢁ)
c (ꢁ)
100.908(7)
105.297(7)
dichloroethane
Ar3P Rh IMes
+
PAr3
Ar3P Rh PAr3
ð1Þ
ð2Þ
60 ˚C
Ar3P
Ar3P
94.929(6)
1420.6(10)
2a/b
1a/b
3
˚
V (A )
Z
1
1.290
Cl
Ar3P Rh IMes
CO
Dcalc (Mg/m3)
dichloroethane
60 ˚C
+
PAr3
no reaction
Reflections collected
Independent reflections
Refinement method
9519
6243 (Rint = 0.341)
Full-matrix least-squares on F2
2a/b
Data/restraints/parameters 6243/0/416
Although the formation of Wilkinsonꢀs complex ap-
peared to be relatively clean at 60 ꢁC, there were no sig-
Goodness-of-fit on F2
Final R indices [I > 2r(I)]
R indices (all data)
0.921
R1 = 0.0689, wR2 = 0.1675
R1 = 0.1274, wR2 = 0.1849
1
nals apparent in the H or 13C NMR spectra of these
reactions indicating the fate of the carbene. Visual
inspection of the reactions showed that a fine precipitate
had formed. NMR and mass spectroscopy of the solid
material indicated that it was derived from IMes, and
was dimeric in nature. Considering the possibility that
this compound resulted from liberation of free carbene
(5) and its subsequent reaction with solvent, we exam-
ined the reaction between free carbene and dichloro-
ethane. Dissolving carbene IMes (5) in neat
dichloroethane at 60 ꢁC gave the same product in 77%
isolated yield, the structure of which was confirmed by
X-ray crystallographic analysis (Fig. 1, Eq. (3)). Dimer
6 likely results from a double SN2 reaction between
dichloroethane and two molecules of the carbene [21].
The remaining material was IMesHCl (18% yield, 95%
mass balance). Remarkably, the reaction of free IMes
and the decomposition of the Rh complex gave the same
ratio of dimer to IMesHCl within experimental error.
Selected crystallographic data and bond lengths for
compound 6 are given in Tables 2 and 3.
Table 3
˚
Selected bond lengths (A) and angles (ꢁ) for dimer 6
6
Bond lengths
C(12)–C(13)
C(13)–C(13A)
N(1)–C(12)
N(1)–C(10)
N(2)–C(12)
N(2)–C(11)
C(10)–C(11)
1.479(5)
1.541(7)
1.348(4)
1.388(4)
1.340(4)
1.401(4)
1.331(5)
Bond angles
C(12)–C(13)–C(13A)
N(2)–C(12)–C(13)
N(1)–C(12)–C(13)
N(2)–C(12)–N(1)
C(12)–N(1)–C(10)
C(12)–N(2)–C(11)
C(11)–C(10)–N(1)
C(10)–C(11)–N(2)
C(12)–N(2)–C(14)
C(12)–N(1)–C(1)
111.0(3)
127.4(3)
125.5(3)
107.1(3)
108.8(3)
109.4(3)
108.2(3)
106.5(3)
126.1(3)
127.1(3)
2Cl–
N
N
Mes
Mes
Mes
Mes
dichloroethane
60 ˚C
+ IMesHCl
(18%)
N
N
Mes
Mes
5
N
N
6
77% yield
ð3Þ
Cl
dichloroethane
Ar3P Rh IMes
+
PAr3
+
6
80%
+
IMesHCl
20%
1a/b
60 ˚C
Ar3P
2a/b
Fig. 1. X-ray structure of [IMesCH2CH2IMes]2+ 2Clꢀ (6).
ð4Þ

3206
D.P. Allen et al. / Journal of Organometallic Chemistry 689 (2004) 3203–3209
The stability of carbene complex 2b was also examined
added dichloroethane (compare entries 4 and 7) [25].
This clearly illustrates the difficulty interpreting catalytic
activity in solvents where catalyst stability is unknown.
in dichloromethane at 45 ꢁC under inert atmosphere,
since this is a more common solvent. Decomposition of
complex 2b and production of Wilkinsonꢀs catalyst was
also observed. Interestingly, Lebel and co-workers [13a]
noted that 2a was active for the methenylation of alde-
hydes in THF, but completely inactive when methylene
chloride was employed as the solvent. This result may
be ascribed to decomposition of the carbene complex.
3. Conclusions
In conclusion, we have demonstrated that the IMes
ligand in rhodium complexes 2a and b is completely re-
placed by phosphine upon treatment with PAr3 in
dichloroethane at slightly elevated temperatures. The
displaced carbene reacts with dichloroethane leading
to alkylation of the solvent. The product distribution
obtained from the reaction of free IMes with dichloro-
ethane is identical to that obtained with the Rh complex,
suggesting that the carbene is being displaced from the
coordination sphere of rhodium prior to reaction. This
reaction has significant implications for reactions
involving carbene complexes, since loss of the carbene
and formation of new species with greater catalytic
activity may hamper the interpretation of results. Thus
halogenated solvents such as dichloromethane and
dichloroethane pose a threat to the integrity of certain
carbene complexes as decomposition of the carbene lig-
and may occur to a significant extent by removing it
from equilibration with the complexed form.
2.3. Catalytic studies
The hydrogenation of isosafrole (7) was examined
employing carbene catalyst 2a and Wilkinsonꢀs catalyst
1a in THF (Eq. 5) [22,23]. At room temperature, catalyst
2a gave a paltry 2 turnover per hour compared with 40
hꢀ1 for Wilkinsonꢀs catalyst (1a) (Table 4, entries 1 and
2). At higher temperatures, the rate of both reactions in-
creased, but the carbene catalyst was still inferior to Wil-
kinsonꢀs catalyst (50 TO/h for 2a compared to 250 TO/h
for 1a at 60 ꢁC), presumably because of the decreased
rate of phosphine dissociation in the former catalyst,
which is essential for catalyst activation in hydrogena-
tions with 1a [24]. The addition of CuCl, a known phos-
phine sponge, increases the activity significantly,
indicating that phosphine dissociation is critical for the
activity of the NHC complexes as well. The addition
of one equiv. of CuCl per Rh, increased the activity of
2a by a factor of 45 (compare entries 2 and 6). CuCl also
increases the rate of reaction at 60 ꢁC, but the most
notable feature of Table 4 is the effect of dichloroethane
at 60 ꢁC. As shown in entry 7, the addition of dichloro-
ethane to the reaction has an accelerating effect presum-
ably by promoting the formation of Wilkinsonꢀs catalyst
via Eq. (4). Thus the rate of reaction with 2a increases
from 50 TO/h at 60 ꢁC in THF to 110 TO/h in THF with
4. Experimental
4.1. General considerations
All the reactions were carried out in a dry Argon or
nitrogen atmosphere using standard Schlenk techniques
or in an Mbraun glovebox containing less than 1 ppm of
oxygen and water. NMR spectra were recorded using
Bruker 300, 400 or 500 MHz spectrometers. All reported
1
TOFs were determined by H NMR using hexamethyl-
benzene as an internal standard.
Table 4
Hydrogenation activity of catalysts 1 and 2a
CH3
CH3
O
O
O
O
4.2. Reagents
1 or 2
200 psi H2
THF
ð5Þ
All solvents were purified by standard procedure and
degassed prior to use by 3 freeze/pump/thaw cycles. Iso-
safrole was purchased from commercial suppliers and
distilled and degassed (3 · fpt) and stored cold in the
glovebox. It was passed through a plug of activated ba-
sic alumina before each hydrogenation experiment. Wil-
kinsonꢀs catalysts was purchased commercially and used
as received. IMesHCl [26], IMes [26], RhCl((P-p-tolyl)3)3
[27] were prepared according to literature procedures.
7
8
Entry Catalyst Temperature (ꢁC) Additive
TOFb (hꢀ1
)
1
2
3
4
5
6
7
8
9
1
25
25
60
60
25
25
60
60
60
–
40
2
2a
1
–
–
250
50
50
90
2a
1
–
CuCl
CuCl
2a
2a
1
ClCH2CH2Cl 110
CuCl
CuCl
300
430
2a
4.3. Preparation of [(IMes)Rh(PPh3)2Cl] (2a)
a
Reactions were carried out in an autoclave, under 200 psi of H2,
with 0.75% catalyst, [substrate]init = 0.1 M in dry, deoxygenated, dis-
tilled THF. Solutions were prepared in a glove box under argon.
In a 100 ml schlenk flask, IMes (5) (185.0 mg, 0.61
mmol) and Wilkinsonꢀs catalyst (1a) (522.0 mg, 0.56
b
Turnover frequency per hour, measured after 30 min.

D.P. Allen et al. / Journal of Organometallic Chemistry 689 (2004) 3203–3209
3207
mmol) were suspended in deoxygenated toluene (50 ml).
The resulting purple suspension was stirred at ambient
temperature for 16 h. The resulting brown solution
was concentrated to ꢁ10 ml in total volume on the sch-
lenk line and precipitation of a bright yellow powder
could be affected by the addition of 40 ml of deoxygen-
ated hexane. The solid was isolated via schlenk filtration
and washed with deoxygenated hexane (3 · 15 ml). After
drying under vacuum for 5 h, 472.5 mg (87%) of com-
plex 2a was isolated.
from bright orange, the light orange to a golden brown.
The white crystals were collected by schlenk filtration
and washed with deoxygenated hexane (3 · 10 ml). After
drying under vacuum for 5 h, 46.5 mg (78%) of the dica-
tion was isolated as X-ray quality crystals.
1H NMR (500 MHz, CD2Cl2) d 1.83 (s, 24H), 2.49 (s,
12H), 2.57 (s, 4H), 6.99 (s, 8H), 8.21 (s, 4H).
13C NMR (500 MHz, CD2Cl2) d 17.82 (s), 20.74 (s),
21.77 (s), 127.30 (s), 129.52 (s), 131.06 (s), 134.54 (s),
141.62 (s), 142.96 (s).
1H NMR (300 MHz, C6D6) d 1.76 (s, 6H), 2.43 (s,
6H), 2.82 (s, 6H), 6.32 (s, 2H), 6.86 (m, 22 H), 7.41
(m,12H).
4.6. Preparation of dication 6 from Rh complex 2b
13C NMR (500 MHz, d8-thf) d 20.45, 21.57, 22.38,
124.20, 126.97 (d, J = 9 Hz), 127.54 (d, J = 9 Hz),
128.06, 128.89, 129.33, 130.37, 135.87 (br), 136.56, 136.78
(br), 138.18, 138.61, 138.90, 139.26, 139.59, 190.50 (ddd,
JC–Rh = 115 Hz, JC–P(trans) = 49 Hz, JC–P(cis) = 14 Hz).
31P NMR (300 MHz, C6D6) d 35.70 (dd, JP–Rh = 120
In a schlenk NMR tube, Rh-complex 2b (20 mg, 0.02
mmol), PPh3 (16 mg, 0.06 mmol) and deoxygenated d4-
1,2-dichloroethane were mixed. The NMR tube was
deoxygenated and held under vacuum and then flame
sealed. The tube was then heated at 60 ꢁC until no fur-
ther signals were seen for the 2b. At this time, the tube
was taken into the glovebox and cracked. A small
amount of white solid was isolated by filtration and
washed with deoxygenated toluene (3 · 1 ml). This was
pumped on under vacuum for 12 h. The yield of the
dication was 11.4 mg (81%).
Hz, JP–P = 40 Hz), 48.78 (dd, JP–Rh = 207 Hz, JP–P
40 Hz).
=
4.4. Preparation of [(IMes)Rh(P-p-tolyl3)2Cl] (2b)
In a 50 ml schlenk flask, IMes (5) (98 mg, 0.32 mmol)
and RhCl((P-p-tolyl)3)3 (1b) (275.0 mg, 0.26 mmol) were
dissolved in deoxygenated toluene (25 ml). The resulting
orange solution was stirred at ambient temperature for
16 h. The volatiles were removed on the vacuum line
from the resulting yellow solution. Then deoxygenated
hexane (30 ml) was added and the residue was triturated
against a counter flow of argon. This resulted in a fine
yellow precipitate, which was isolated by schlenk filtra-
tion and washed with deoxygenated cold hexane
(3 · 10 ml) to obtain 214 mg (78%) of 2b.
4.7. Typical NMR experiment for determination of
exchange rate
NMR samples were prepared inside an Argon or
Nitrogen filled glove box using tubes fitted with J–
Young valves, or in flame sealed tubes. Rh complex 2b
was added into the NMR tube (31.5 mg, 0.03 mmol)
along with excess phosphine (3–50 equivalents), and
0.75 ml of solvent. The T1 of free phosphine and the
phosphines in Rh complex 2 were measured at room
temperature using the standard inversion recovery meth-
od, and a relaxation delay time of 5 · T1 for the largest
T1 was employed for the inversion transfer experiments.
The temperature was raised to the desired point, and al-
lowed to equilibrate with the sample in the probe for 15
min before beginning the experiment.
1H NMR (500 MHz, C6D6) d 1.87 (s, 6H), 2.01 (s,
9H), 2.06 (s, 9H), 2.50 (s, 6H), 2.89 (s, 6H), 6.40 (s,
2H), 6.74 (m, 12H), 6.93 (s, 2H), 7.20 (s, 2H), 7.41 (br,
12H).
13C NMR (600 MHz, C6D6) d 20.52, 21.44, 21.62,
21.78, 22.76, 123.25, 127.77 (d, J = 9 Hz), 128.01 (br),
128.96, 130.31, 135.66, 135.70, 136.64 (br), 137.40,
The 31P inversion transfer experiments were per-
formed as follows: a 180ꢁ selective pulse was applied
to the signal for free phosphine with the transmitter
on resonance. Delay times of 0 s to 5 · T1 were applied
in increments followed by a 90ꢁ read pulse. The data
were acquired successively in blocks of four transients
at each delay time. The free-induction decays were
zero-filled prior to Fourier transformation and a base-
line correction routine was applied to the resulting spec-
tra prior to integration of the phosphine signals. The
exchange rate of bound and free phosphine was ex-
tracted by fitting the integration of these signals at in-
creasing delay times to a two-site exchange model using
either Bainꢀs CIFIT program [28], or a programmable
137.69, 138.10, 139.03, 139.70, 191.66 (ddd, JC–Rh
116 Hz, JC–P(trans) = 49 Hz, JC–P(cis) = 14 Hz).
=
31P NMR (500 MHz, C6D6) d 34.26 (dd, JP–Rh = 120
Hz, JP–P = 40 Hz), 46.69 (dd, JP–Rh = 207 Hz, JP–P = 40
Hz).
4.5. Preparation of dication 6 from free IMes (5)
In a 35 ml pressure tube, IMes (5) (52 mg, 0.17 mmol)
and deoxygenated 1,2-dichloroethane were mixed. The
tube was sealed and the resulting orange solution was
heated at 60 ꢁC for 5 days. During this time, white nee-
dle like crystals formed while the solution color faded

3208
D.P. Allen et al. / Journal of Organometallic Chemistry 689 (2004) 3203–3209
[4] (a) D.S. McGuinness, W. Mueller, P. Wasserscheid, K.J. Cavell,
B.W. Skelton, A.H. White, U. Englert, Organometallics 21
(2002) 175 (references therein);
spreadsheet. Error analysis of the calculated exchange
rates was performed using the method of Bain and Cra-
mer [28] to yield typical 95% confidence limits of
ꢁ 15%.
(b) D.S. McGuinness, N. Saendig, B.F. Yates, K.J. Cavell, J.
Am. Chem. Soc. 123 (2001) 4029;
(c) D.S. McGuinness, M.J. Green, K.J. Cavell, B.W. Skelton,
A.H. White, J. Organomet. Chem. 565 (1998) 165.
[5] (a) J. Huang, E.D. Stevens, S. Nolan, Organometallics 19 (2000)
1194;
4.8. Typical hydrogenation reaction
The Rh catalyst (1 or 2a) (0.01 mmol) and isosafrole
(1.3 mmol) were loaded into a glassliner and dissolved in
7 ml of THF. This was then placed in a high pressure
autoclave and assembled. The autoclave was filled and
flushed three times with H2 before it was finally pressu-
rized to 200 psi. The autoclave was then allowed to stir
for 30 min (either at rt or 60 ꢁC). At this time the pres-
sure was released and the autoclave was disassembled.
All the contents were then transferred to a 50 ml rbf con-
taining a known amount of hexamethylbenzene as an
internal standard. The volatiles were removed under re-
duced pressure and the residue was analyzed by 1H
NMR to determine the % conversion and subsequent
TOF.
(b) R.F.R. Jazzar, S.A. Macgregor, M.F. Mahon, S.P.
Richards, M.K. Whittlesey, J. Am. Chem. Soc. 124 (2002)
4944;
(c) M.J. Chilvers, R.F.R. Jazzar, M.F. Mahon, M.K. Whit-
tlesey, Adv. Synth. Catal. 345 (2003) 1111;
(d) R. Dorta, E.D. Stevens, S.P. Nolan, J. Am. Chem. Soc.
126 (2004) 5054.
[6] (a) A.K. de K. Lewis, S. Caddick, F.G.N. Cloke, N.C. Billingham,
P.B. Hitchcock, J. Leonard, J. Am. Chem. Soc. 125 (2003) 10066;
(b) S. Caddick, F.G.N. Cloke, P.B. Hitchcock, J. Leonard, A.K.
de K. Lewis, D. McKerrecher, L.R. Titcomb, Organometallics 21
(2002) 4318;
(c) R.W. Simms, M.J. Drewitt, M.C. Baird, Organometallics 21
(2002) 2958;
(d) R. Dorta, E.D. Stevens, C.D. Hoff, S.P. Nolan, J. Am. Chem.
Soc. 125 (2003) 10490.
[7] P.B. Hitchcock, M.F. Lappert, P.L. Pye, J. Chem. Soc., Dalton
Trans. (1978) 826.
[8] A.A. Danopoulos, N. Tsoureas, J.C. Green, M.B. Hursthouse,
Chem. Commun. (2003) 756.
5. Supplementary material
[9] (a) N. Kuhn, J. Fahl, R. Fawzi, C. Maichle-Mossmer, M.
Steimann, Z. Naturforsch. B 53 (1998) 720;
Crystallographic data for the structural analysis have
been deposited with the Cambridge Crystallographic
Data Centre, CCDC No. 246782 for compound 5. Cop-
ies of this information may be obtained free of charge
from The Director, CCDC, 12 Union Road, Cambridge,
CB2 1EZ UK [fax: (int code) +44(1223)336-033, or E-
mail: deposit@ccdc.cam.ac.uk or www: http://www.
ccdc.cam.ac.uk.
(b) M.L. Cole, C. Jones, P.C. Junk, New J. Chem. 262 (2002)
1296.
[10] D.J. Cardin, B. Cetinkaya, M.F. Lappert, Chem. Rev. 72 (1972)
545.
[11] (a) C. Kocher, W.A. Herrmann, J. Organomet. Chem. 532 (1997)
261;
(b) W.A. Herrmann, L.J. Goossen, C. Kocher, G.R. Artus,
Organometallics 6 (1997) 2472;
(c) W.A. Herrmann, L.J. Goossen, C. Kocher, G.R. Artus,
Angew. Chem., Int. Ed. Engl. 35 (1996) 2805.
[12] A.C. Chen, L. Ren, A. Decken, C.M. Crudden, Organometallics
19 (2000) 3459.
[13] (a) G.A. Grasa, Z. Moore, K.L. Martin, E.D. Stevens, S.P.
Nolan, V. Paquet, H. Lebel, J. Organomet. Chem. 658 (2002) 126;
(b) H. Lebel, V. Paquet, C. Proulx, Angew. Chem., Int. Ed. Engl.
40 (2001) 2887.
[14] The 31P NMR spectrum of 2b is composed of two dd at 46.53 ppm
(JP–P = 39.8 Hz, JP–Rh = 206.4 Hz) and 33.95 ppm (JP–P = 39.8
Hz, JP–Rh = 119.8 Hz) indicating that the two phosphines are cis
to each other [12]. X-ray crystallographic analysis confirms the cis
orientation in the solid state [13].
Acknowledgements
We acknowledge financial support for this research
from the Natural Sciences and Engineering Research
Council of Canada (NSERC) in the form of research
grants to CMC and scholarships to D.P.A., and thank
Professor Ilhyong Ryu of Osaka Prefecture University
and Professor Jason Clyburne of Simon Fraser Univer-
sity for valuable comments on the manuscript.
[15] With less bulky carbenes, a trans arrangement of phosphines is
observed: B. Cetinkaya, I. Ozdemir, P.H. Dixneuf, J. Organomet.
Chem. 534 (1997) 153.
[16] Note that these experiments were carried out in toluene under
conditions where the formation of Wilkinsonꢀs catalysts was not
taking place (vide infra).
References
[17] J.M. Brown, P.L. Evans, A.R. Lucy, J. Chem. Soc., Perkin Trans.
II (1987) 1589.
[1] (a) W.A. Herrmann, Angew. Chem., Int. Ed. 41 (2002) 1290;
(b) W.A. Herrmann, C. Kocher, Angew. Chem., Int. Ed. 36
(1997) 2162.
[18] This is in stark contrast to ClRh(PPh3)3, in which the two
phosphine sites exchange by an intramolecular mechanism more
rapidly than they exchange with external phosphine (rate of
intramolecular exchange = 22 sꢀ1 at 25 ꢁC compared to 0.3 sꢀ1 for
intermolecular exchange) [17].
[2] (a) W.A. Herrmann, M. Elison, J. Fischer, C. Ko¨cher, G.R.J.
Artus, Angew. Chem., Int. Ed. Engl. 34 (1995) 2371;
(b) E. Peris, J.A. Loch, J. Mata, R.H. Crabtree, Chem. Commun.
(2001) 201.
[19] (a) R.H. Grubbs, M. Sanford, J.L. Love, J. Am. Chem. Soc. 123
(2001) 6543;
[3] C.M. Crudden, D.A. Allen, Coord. Chem. Rev. (in press).

D.P. Allen et al. / Journal of Organometallic Chemistry 689 (2004) 3203–3209
3209
(b) J.A. Love, M.S. Sanford, M.W. Day, R.H. Grubbs, J. Am.
Chem. Soc. 125 (2003) 10103.
[22] (a) For hydrogenations with Ru and Ir NHCs: M.T. Powell, D.R.
Hou, M.C. Perry, X. Cui, K. Burgess, J. Am. Chem. Soc. 123
(2001) 8878;
[20] Trace amounts of water accelerate production of Wilkinsonꢀs
catalyst dramatically. All of the experiments described herein are
carried out in flame sealed NMR tubes prepared using standard
Schlenk techniques and rigorously deoxygenated samples.
[21] Note that the course of this reaction has been debated in the
literature. Kuhn has reported that tetramethylimidazol-2-ylidene
reacts with ClCH2CH2Cl to give 2-chloroimidazolium chloride
[9a]. Junk and Jones observed a 50/50 mixture of IMesHBr and 2-
bromoimidazolium bromide upon treatment of IMes with
BrCH2CH2Br [9b]. Since the C–Cl bond is stronger than the C–
Br bond, they proposed that the abstraction of chlorine as
reported by Kuhn was unreasonable, although it should be noted
that different carbenes were employed in the two studies. In order
to rule out the presence of 2-chloroimidazolium chloride resulting
from Cl abstraction in our reaction, we prepared this compound
by reaction of IMes with Cl6C2 [9b]. The positions of several of the
protons are diagnostic and show definitively that this species is not
present in the 60 ꢁC reaction or in the 25 ꢁC reaction. Since there is
a considerable difference between the tetramethyl carbene Kuhn
employed and IMes which we have employed, a difference in
reactivity may be reasonable.
(b) H.M. Lee, T. Jiang, E.D. Stevens, S.P. Nolan, Organometallics
20 (2001) 1255;
(c) M. Albrecht, J.R. Miecznikowski, A. Samuel, J.M. Faller,
R.H. Crabtree, Organometallics 21 (2002) 3596;
´
(d) L.D. Vazquesz-Serrano, B.T. Owens, J.M. Buriak, Chem.
Commun. (2002) 2518 See also reference [13].
[23] Note that carbene complex 2 is stable in THF at 25 and 60 ꢁC.
[24] (a) J. Halpern, T. Okamoto, A. Zakhariev, J. Mol. Catal. 2 (1976)
65;
(b) J. Halpern, Inorg. Chim. Acta 50 (1981) 11.
[25] The rate does not increase to the same level as Wilkinsonꢀs
itself (entry 3) because the short time frame of the experiment
(30 min) allows only partial formation of Wilkinsonꢀs catalyst
from 2a.
[26] (a) H.-W. Wanzlick, Angew. Chem., Int. Ed. Engl. 1 (1962) 75;
(b) A.J. Arduengo III, H.V.R. Dias, R.L. Harlow, M. Kline, J.
Am. Chem. Soc. 114 (1992) 5530.
[27] C.A. Tolman, P.Z. Meakin, D.L. Lindner, J.P. Jesson, J. Am.
Chem. Soc. 96 (1974) 2762.
[28] A.D. Bain, J.A. Cramer, J. Magn. Reson. 118A (1996) 21.