Olefin cyclopropanation catalyzed by iridium(III) porphyrin complexes

Article abstract of DOI:10.1021/om300135f

Tetratolylporphyrinato (TTP) iridium complexes were shown to be extremely active and robust catalysts for the cyclopropanation of olefins using diazo compounds as carbene sources. Ir(TTP)CH₃ (1) catalyzed the cyclopropanation of styrene with ethyl diazoacetate (EDA) at -78 °C and achieved 4.8 × 10⁵ turnovers in three successive reagent additions with no sign of deactivation. High yields and moderate trans selectivities were attained for electron-rich and sterically unhindered substrates. A Hammett ρ⁺ value of -0.23 was determined by competition experiments with para-substituted styrenes. Furthermore, competitive cyclopropanation of styrene and styrene-d₈ with EDA and 1 demonstrated a moderate inverse secondary isotope effect of 0.86 ± 0.03. These data are consistent with a catalytic cycle that proceeds through a metalloporphyrin carbene intermediate. Carbene transfer to olefin substrates appears to be rate limiting, as indicated by kinetic studies. Hexavalent iridium halogenato tetratolylporphyrinato complexes of the form Ir(TTP)X(L), where X = Cl, Br, I, NCS and L = CO, NMe₃ (2-6), and cationic analogues, where X = BF₄ and L = CO or vacant site (7, 8), also demonstrated high catalytic cyclopropanation activity.

Full text of DOI:10.1021/om300135f

Article
pubs.acs.org/Organometallics
Olefin Cyclopropanation Catalyzed by Iridium(III) Porphyrin
Complexes
Bernie J. Anding, Arkady Ellern, and L. Keith Woo*
Department of Chemistry, Iowa State University, Ames, Iowa 50011-3111, United States
S
* Supporting Information
ABSTRACT: Tetratolylporphyrinato (TTP) iridium com-
plexes were shown to be extremely active and robust catalysts
for the cyclopropanation of olefins using diazo compounds as
carbene sources. Ir(TTP)CH3 (1) catalyzed the cyclopropana-
tion of styrene with ethyl diazoacetate (EDA) at −78 °C and
achieved 4.8 × 105 turnovers in three successive reagent
additions with no sign of deactivation. High yields and moderate
trans selectivities were attained for electron-rich and sterically unhindered substrates. A Hammett ρ+ value of −0.23 was
determined by competition experiments with para-substituted styrenes. Furthermore, competitive cyclopropanation of styrene
and styrene-d8 with EDA and 1 demonstrated a moderate inverse secondary isotope effect of 0.86
0.03. These data are
consistent with a catalytic cycle that proceeds through a metalloporphyrin carbene intermediate. Carbene transfer to olefin
substrates appears to be rate limiting, as indicated by kinetic studies. Hexavalent iridium halogenato tetratolylporphyrinato
complexes of the form Ir(TTP)X(L), where X = Cl, Br, I, NCS and L = CO, NMe3 (2−6), and cationic analogues, where X =
BF4 and L = CO or vacant site (7, 8), also demonstrated high catalytic cyclopropanation activity.
Scheme 1. Cyclopropanation Mechanism for Rh(TTP)I
INTRODUCTION
■
Over the past few decades, carbene transfer reactions used to
generate new C−C bonds have grown tremendously in
synthetic utility.1−3 Furthermore, carbene moieties generated
from diazo reagents provide attractive atom-efficient and
environmentally friendly protocols, because N2 is the only
byproduct. Although many diazo reagents react sluggishly on
their own, various catalysts efficiently assist in the formation of
carbene intermediates.4−7 Among the highly active and robust
catalysts, metalloporphyrin complexes are particularly versatile
for the design of stereoselective carbene transfers.8−10 Metal-
loporphyrins containing iron,11 ruthenium,12 osmium,13
cobalt,14,15 and rhodium15 are all active for catalytic carbene
transfer. For example, Fe and Rh complexes, exhibit a great
breadth of catalytic range, including C−H insertions,10,16 N−H
insertions,17 olefinations,18 and cyclopropanations.11,19 In
addition, iron porphyrin catalysts are robust enough to achieve
4300 turnovers,20 and rhodium adducts furnish cyclopropana-
tion and C−H insertion products with unusual selectivity for
cis-cyclopropanes and primary insertion products, respec-
tively.21,22 Despite these notable results for group 8 and
group 9 metals, the reactivity of iridium porphyrins toward
carbene transfer has not been reported.
Interestingly, the active catalyst was believed to exist primarily
as an alkylrhodium complex resulting from initial carbene
insertion into the Rh−I bond. Analogous alkyliridium
porphyrin compounds, unlike their halogenato counterparts,
are pentacoordinate species that appear to be good candidates
for catalysis. The present work explores the catalytic activity of
Examples of catalysis by halogenato iridium porphyrins are
quite rare in comparison to those for the related rhodium
porphyrins. This may be due, in part, to the presence of a
kinetically inert CO ligand bound to iridium analogues, yielding
a hexavalent complex devoid of an available vacant site.23
Indeed, the mechanism of cyclopropanation by Rh(TTP)I,
reported by Kodadek and co-workers, requires an open site for
the initial coordination of the diazo compound (Scheme 1).24
Received: February 17, 2012
Published: April 27, 2012
© 2012 American Chemical Society
3628
dx.doi.org/10.1021/om300135f | Organometallics 2012, 31, 3628−3635

Organometallics
Article
alkyliridium porphyrin complexes, as well as other iridium(III)
porphyrins, toward transformations with diazo reagents. The
work described herein serves as the first examples of carbene
transfer catalyzed by iridium porphyrin complexes.
method was also applicable to less reactive diazo reagents
such as methyl 2-phenyldiazoacetate (MPDA; eq 1).
RESULTS AND DISCUSSION
■
The catalytic cyclopropanation activity of both methyl and
halogenato iridium tetratolylporphyrinato (TTP) complexes,
Ir(TTP)CH3 (1) and Ir(TTP)Cl(CO) (2) were evaluated.
Syntheses of these complexes were reported previously.25,26
Preliminary reactivity studies showed that methyliridium
complex 1 decomposed ethyl diazoacetate (EDA) readily at
temperatures as low as −78 °C to form maleates and fumarates
(cis:trans = 6.5:1) in quantitative yield. Reaction intermediates
1
were too transient for spectroscopic observation by H NMR.
With these initial results, the cyclopropanation of various
olefins and EDA using 1 was explored using a protocol similar
to that developed for N,N′-bis(salicylidene)ethylenediiminato
iridium, Ir(salen).27 Dropwise addition of EDA to a solution of
styrene and 1 at room temperature resulted in instantaneous
gas evolution. Cyclopropanes were formed in 35% yield along
with a significant amount of dimerization of the diazo reagent
to form diethyl maleate and diethyl fumarate (Table 1). The
Reactions catalyzed by 1 displayed a moderate preference for
trans cyclopropanes. This selectivity was enhanced slightly for
reactions run in THF, albeit at higher temperatures and longer
reaction times. In comparison to similar catalysts, 1 was more
trans selective than Rh(TPP)I but less than Fe(TPP)Cl.20,21
Moreover, analysis of the diastereoselectivity for the para-
substituted styrenes revealed an influence of substrate
electronics on diastereoselectivity. Substrates with electron-
withdrawing groups at the para position displayed a larger
preference for trans cyclopropanes. Notably, cyclopropanation
with trans-β-methylstyrene formed only the isomer bearing the
phenyl group anti to both methyl and ethyl carboxylate, as
determined by NOESY.
Table 1. Cyclopropanation of EDA and Alkenes with
a
Ir(TTP)CH3
To evaluate catalyst activity further, cyclopropanation was
explored at very low catalyst loadings. Using 5.8 × 10−4 mol %
of 1 relative to EDA in the presence of excess styrene, diazo
conversion and cyclopropanation of styrene were determined
after warming the reaction mixture to room temperature from
−78 °C. For three consecutive diazo additions, EDA was
quantitatively consumed to form cyclopropanes with a small
amount of diazo dimerization. Cyclopropanation yields were
93%, 92%, and 91% for the first, second, and third additions,
respectively. Overall, 1 achieved 4.8 × 105 turnovers with
almost no indication of catalyst deactivation. This TON is
nearly 2 orders of magnitude greater than the values reported
for Ru and Rh porphyrin catalysts.28−30
Surprisingly, Ir(TTP)Cl(CO) (2) also effectively catalyzed
the cyclopropanation of olefins with EDA, despite being
coordinatively saturated (Table 2). The substrate reactivity
trends were similar for both 1 and 2. However, 2 required
higher reaction temperatures and longer reaction times.
Although 2 was less selective for trans cyclopropane products,
it was slightly more efficient for the cyclopropanation of more
hindered and electron-deficient olefins.
Various metalloporphyrin complexes, of the form [Ir(TTP)-
(L)]+ or Ir(TTP)X(L), where X is an anionic ligand and L is a
neutral ligand, were generated to compare axial ligand effects
on catalytic activity. CO ligand substitution was accomplished
by adding trimethylamine N-oxide to a CH2Cl2 solution of 2
(Scheme 2). The resulting product was established to be
Ir(TTP)Cl(NMe3) (3) on the basis of 1H NMR and IR
spectroscopy. Protons of the axially coordinated NMe3 were
strongly shifted upfield due to the porphyrin ring current effect
and appeared as a sharp singlet at −2.99 ppm. Moreover, CO
loss was verified by the disappearance of the strong Ir−CO
stretch at 2050 cm−1. The molecular structure of 3 (Figure 1)
was determined by single-crystal X-ray analysis. One molecule
of 3 occupied the asymmetric unit of the monoclinic cell
time
T
yield
b
b
substrate
styrene
solvent
(min)
(°C)
(%)
trans:cis
CH2Cl2
CH2Cl2
THF
<5
<5
30
<5
<5
<5
<5
<5
20
30
<5
<5
23
−78
0
35
85
80
86
84
91
84
90
60
62
5.5:1
5.8:1
6.3:1
3.4:1
5.8:1
7.3:1
8.0:1
2.0:1
3.3:1
3.2:1
e
styrene
styrene
4-methoxystyrene
4-methylstyrene
4-bromostyrene
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
−78
−78
−78
−78
−78
−78
−78
23
c
4-nitrostyrene
α-methylstyrene
1-hexene
indene
d
cyclohexene
14
d
f
trans-β-
methylstyrene
23
31
1
d
ethyl acrylate
CH2Cl2
<5
23
12
∼10:1
a
Conditions:
1 × M in Ir(TTP)CH3; Ir(TTP)-
10−3
b
CH3:EDA:substrate = 1:100:500. Yields and diastereomeric ratios
c
of cyclopropanes as determined by GC. 4-Nitrostyrene was not
distilled prior to use. Determined by NMR. Isomer not determined.
d
e
f
Only one isomer observed.
sum of the yields for cyclopropanation and dimerization
correspond to complete conversion of EDA. Lowering the
reaction temperature to −78 °C resulted in a significant
increase in yield for cyclopropanes (85%), with a corresponding
decrease in dimerization of the diazo reagent. This method was
cleanly extended to other terminal aryl olefins. In contrast,
electron-deficient and internal olefins (except indene) required
longer reaction times or higher temperatures. Yields for these
substrates were lower, especially for those run at room
temperature, where dimerization becomes dominant. This
3629
dx.doi.org/10.1021/om300135f | Organometallics 2012, 31, 3628−3635

Organometallics
Article
represents only the second decarbonylation procedure
published for Ir(TTP)Cl(CO).23
Table 2. Cyclopropanation of EDA and Alkenes with
Ir(TTP)Cl(CO)
a
Anionic ligand substitution products were generated by
chloride abstraction from 2 with silver(I) tetrafluoroborate,
followed by anion addition with the appropriate tetrabutylam-
monium salt. Only minor spectroscopic differences were
observed between 2 and Ir(TTP)(CO)X (4, X = Br; 5, X =
I; 6, X = SCN). Compositions and molecular structures were
confirmed by mass spectrometry and single-crystal X-ray
analysis (Figures 2 and 3). Selected metrical parameters are
time
T
yield
b
b
substrate
styrene
solvent
(min)
(°C)
(%)
trans:cis
CH2Cl2
CH2Cl2
THF
<5
20
90
20
30
60
<5
<5
23
0
92
85
75
93
59
43
4.7:1
4.9:1
4.3:1
1.8:1
3.5:1
2.5:1
d
styrene
styrene
23
0
α-methylstyrene
1-hexene
indene
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
0
0
c
cyclohexene
23
23
20
c
e
trans-β-
methylstyrene
67
1
ethyl acrylate
CH2Cl2
<5
23
24
∼10:1
a
Conditions: 1 × 10−3 M in Ir(TTP)Cl(CO); Ir(TTP)Cl-
b
(CO):EDA:substrate = 1:100:500. Yields and diastereomeric ratios
c
of cyclopropanes as determined by GC. Determined by NMR.
Figure 2. ORTEP of 5 showing one of the two configurations
disordered by a mirror plane. Thermal ellipsoids are drawn at the 50%
probability level. Selected bond distances (Å) and angles (deg) for 5:
Ir(1)−N = 2.048(7), Ir(1)−I(1) = 2.491(2), Ir(1)−C(11) = 1.852(2),
C(11)−O(11) = 1.141(2); I(1)−Ir(1)−C(11) = 180, I(1)−Ir(1)−
N4(EQP) = 90, C(11)−Ir(1)−N4(EQP) = 90. Selected bond
d
e
Isomer not determined. Only one isomer observed.
Scheme 2. Synthesis of Complexes of the form Ir(TTP)X(L)
distances (Å) and angles (deg) for isostructural 4: Ir(1)−Npyrrole
=
2.047(4), Ir(1)−Br(1) = 2.409(2), Ir(1)−C(11) = 1.855(3), C(11)−
O(11) = 1.143(3); Br(1)−Ir(1)−C(11) = 180, Br(1)−Ir(1)−
N4(EQP) = 90, C(11)−Ir(1)−N4(EQP) = 90.
Figure 3. ORTEP of 6. Thermal ellipsoids are drawn at the 50%
probability level. Selected bond distances (Å): average Ir(1)−Npyrrole
=
2.049(12), Ir(1)−N(5) = 1.961(14), N(5)−C(50) = 1.17(2), C(50)−
S(1) = 1.664(19), Ir(1)−C(49) = 1.897(14), C(49)−O(1) =
1.153(17).
together with two molecules of CH2Cl2. Presumably,
oxidization of CO to CO2 by trimethylamine N-oxide generated
a pentacoordinate chloroiridium complex, which was quickly
trapped by the in situ generated trimethylamine.31 This
given in the figure captions. Complexes 4 and 5 are
isostructural, and both crystallize in the I4/m space group
with a fourth of the molecule in the asymmetric unit and the Ir
atom in a 4/m site symmetry. The central Ir atom has typical
octahedral coordination with an ideal Ir−N4 equatorial plane
(EQP) and displays axial ligand disorder through a mirror
plane. For X = SCN, complex 6, N coordination of the
thioisocyanate ligand to the metal center through nitrogen was
confirmed by bond length analysis. Notably, the Ir−N(CS)
bond is 0.049(14) Å shorter than that for the shortest reported
Ir−N(CS) compound.33 If no anionic ligand sources were
added after chloride abstraction, an inseparable mixture of
cationic iridium(III) porphyrin complexes, Ir(TTP)(BF4) (7)
and Ir(TTP)(BF4)(CO) (8), was obtained in variable ratios.34
Attempts to cleanly isolate either of these cationic complexes
failed.
Figure 1. ORTEP32 of 3 with 50% probability thermal ellipsoids.
Selected bond distances (Å): average Ir(1)−Npyrrole = 2.035(4), Ir(1)−
Cl(1) = 2.355(1), Ir(1)−N(5) = 2.174(2), average N(5)−C =
1.479(6).
3630
dx.doi.org/10.1021/om300135f | Organometallics 2012, 31, 3628−3635

Organometallics
Article
Catalytic results, compiled in Table 3, show that the activity
of hexavalent iridium porphyrin complexes for the cyclo-
Scheme 3. Proposed General Cyclopropanation Mechanism
with Ir(TTP)CH3
Table 3. Cyclopropanation of EDA and Styrene using
a
Various Ir Porphyrin Catalysts
time
yield
(%)
cat.
Ir(TTP)Cl(CO) (2)
(min)
trans:cis
20
60
85
75
75
56
86
93
93
4.9:1
5.0:1
4.6:1
5.5:1
4.5:1
3.3:1
3.0:1
Ir(TTP)Cl(NMe3) (3)
Ir(TTP)Br(CO) (4)
60
Ir(TTP)I(CO) (5)
<5
120
<5
<5
Ir(TTP)SCN(CO) (6)
Ir(TTP)(BF4)/Ir(TTP)(BF4)(CO) (7/8)
b
2 and AgBF4
Table 4. Cyclopropanation using EDA and Ir(TTP)CH3:
Substrate Competition Reactions
a
a
Conditions: 1 × 10−3 M in catalyst; catalyst:EDA:styrene =
1:100:500; reactions run at 0 °C in CH2Cl2. Yields and diastereomeric
b
ratios were determined by GC. 2 and AgBF4 were premixed in
CH2Cl2 and stirred for 30 min.
substrate A
styrene
substrate B
ratio of A to B
propanation of styrene with EDA was relatively independent of
the type of axial ligand. The only neutral species that showed a
significant change in reactivity was Ir(TTP)I(CO). In
comparison to other halogenato iridium porphyrins, Ir(TTP)-
I(CO) was considerably more active and diastereoselective, but
cyclopropanation yields were lower, due to increased
dimerization of the diazo reagent. In contrast, the cationic
catalyst mixture (7/8) was as reactive and efficient toward
cyclopropanation, but lower diastereoselectivity was observed.
The cationic catalysts were either synthesized and isolated
before cyclopropanation or generated in situ with silver
tetrafluoroborate. Regardless of catalyst preparation, the
catalytic efficiency and selectivity were effectively unchanged.
The above results for iridium porphyrin compounds share
many similarities with group 8 metalloporphyrin catalysts. Both
types of catalysts are trans selective and react smoothly with
terminal aryl olefins. In addition, yields decrease significantly
for electron-poor or sterically hindered olefins. This reduction
in efficiency was not nearly as drastic for cobalt and rhodium
porphyrin catalysts.19,35 On the basis of these observations, a
mechanism similar to that previously proposed for group 8
metalloporphyrins seems likely for the catalytic pathway
undertaken for cyclopropanation by Ir(TTP)CH3 (Scheme
3).20,24 The initial step involves EDA coordination to form a
diazoalkyl complex. Loss of nitrogen forms an intermediate
carbene complex, which is susceptible to nucleophilic attack by
an alkene. Cyclopropane production regenerates the active
catalyst, Ir(TTP)CH3.
indene
3.9
styrene
cyclohexene
cyclohexene
styrene
only A
only A
1.36
indene
4-methylstyrene
4-bromostyrene
4-methoxystyrene
4-nitrostyrene
styrene
styrene
1.12
styrene
1.80
styrene
0.78
styrene-d8
0.85
a
Conditions:
5 × M in Ir(TTP)CH3; Ir(TTP)-
10−4
CH3:EDA:substrate A:substrate B = 1:200:1000:1000; reactions
allowed to run for 1 h. Yields were determined by GC.
compared. Significant chemoselectivity was observed between
styrene, indene, and cyclohexene. Cyclohexene was completely
unreactive in the presence of either styrene or indene. For the
competitive cyclopropanation of styrene and indene, products
derived from styrene were favored in a ratio of 3.9:1. A
comparison of para-substituted styrenes demonstrated that
more electron-rich olefins reacted preferentially. A Hammett
correlation with the σ+ parameter had good linearity with ρ+ =
−0.23 (Figure 4). This is indicative of slight positive charge
buildup in the transition state. While this ρ+ value is consistent
Kodadek and co-workers spectroscopically observed the
diazoalkyl complex formed from EDA coordination to
Rh(TTP)I at temperatures below −20 °C.36 Unlike Rh(TTP)I,
Ir(TTP)CH3 is an extremely active catalyst at temperatures as
low as −78 °C. Consequently, intermediate diazoalkyl or
carbene complexes were not observed in Ir(TTP)CH3-
catalyzed reactions. Additional studies were employed to
further explore the catalytic mechanism for Ir(TTP)CH3.
Substrate competition reactions were examined to determine
the olefin influence on the reaction rate (Table 4). Olefins with
a broad range of steric and electronic properties were
Figure 4. Hammett plot for the cyclopropanation of para-substituted
styrenes with EDA and 1.
3631
dx.doi.org/10.1021/om300135f | Organometallics 2012, 31, 3628−3635

Organometallics
Article
with the mechanism in Scheme 3, it falls below the range found
for iron and ruthenium porphyrin systems (−0.44 to
−1.29).20,30,37,38 In addition, the better correlation with σ+
instead of σ has been reported for other metalloporphyrin
cyclopropanation systems.30,37
A secondary isotope effect was observed during the
competitive cyclopropanation of styrene with styrene-d8.
Complex 1 exhibited an inverse isotope effect of 0.86 0.03,
which is comparable to that measured for iron(III) tetrakis-
(pentafluorophenyl)porphyrin chloride.20 This suggests that
some olefin rehybridization occurs before the transition state of
the carbene transfer step. Notably, no secondary isotope effect
was observed for rhodium(III) tetramesitylporphyrin chlor-
ide.39 In comparison to the transition state model proposed by
Kodadek, Woo, and co-workers,20 carbene transfers from
iridium porphyrins have a transition state later than that for
rhodium porphyrins. This implies that iridium carbene species
are more tightly bound and less electropositive, which may
rationalize the observed differences in diastereo- and chemo-
selectivity.
Figure 5. First-order integrated rate law plots at 273.0 K for different
initial concentrations of MDA with 0.577 M hexene and 5.63 × 10−7
M Ir(TTP)CH3. Data points from the first 400 s were omitted in the
linear regression to correct for temperature equilibration. The average
slope was (−2.9 0.4) × 10−4 s−1.
Despite these mechanistic insights, the rate-determining step
for cyclopropanation remained unclear. Accordingly, we sought
to explore the kinetics for cyclopropanation of styrene and
methyl diazoacetate (MDA) in the presence of 1. MDA was
used instead of EDA to simplify monitoring the reaction by 1H
NMR. Due to the reactivity of 1 as well as its propensity to
catalyze dimerization of the diazo reagent, practical reaction
conditions were not trivial to achieve. In order to monitor the
reaction by NMR for several half-lives with cyclopropanation as
the major product, kinetic experiments were carried out at
273.0 K, with ca. 6.4 × 10−4 mol % of 1 and excess styrene.
However, increasing the amount of styrene from 4.3 to 8.3
equiv relative to MDA reduced the rate of MDA consumption.
We postulated that this inhibition was the result of competitive
styrene binding to the catalyst. Indeed, on investigatation by
visible absorption spectroscopy, the Soret band of 1 was red-
shifted by ca. 7 nm in the presence of excess styrene, which is
suggestive of olefin coordination (see the Supporting
Information).
the dimerization products. Derivations of the rate law equations
as well as initial rate data are described in the Supporting
Information.
The rate dependence on the olefin concentration was further
evaluated by comparing kinetic reactions in the presence and
absence of 1-hexene. Three sets of reactions were run in
tandem to ensure nearly equivalent conditions, and the initial
rates, determined at less than 10% of reaction (before 500 s),
were compared (Figure 6). The results are compiled in Table
To circumvent this complication, the substrate was switched
to 1-hexene. Hexene binding to 1 was considerably less
substantial, as demonstrated by visible absorption spectroscopy.
Unfortunately, because hexene is also less reactive than styrene,
a significant amount of diazo dimerization was anticipated.
Nevertheless, the cyclopropanation of hexene with MDA was
studied at 273.0 K to measure the influence of catalyst, diazo
reagent, and olefin concentration on reaction rate. Reactions
were run with MDA as the limiting reagent, which allowed for
the order of MDA to be determined using integrated rate
equations. The data was plotted most suitably to the first-order
rate law equation, giving a pseudo rate constant, k′, of (2.9
0.4) × 10−4 s−1 (Figure 5). In addition, the order of 1 was
determined by plotting the natural log of k′ versus the natural
log of the concentration of 1 for a series of reactions at different
catalyst concentrations. The slope of the resulting line was 1.02,
indicating a first-order dependence on the concentration of 1.
Reactions with different concentrations of hexene demon-
strated saturation kinetics with respect to the rate of MDA
consumption. The fastest rate was observed using a 2-fold
excess of hexene (0.14 M) relative to MDA. At higher
concentrations of hexene, the rate of MDA consumption
decreased marginally, but the rate of formation of cyclo-
propanation products continued to increase at the expense of
Figure 6. Plot of MDA consumption for the reactions of MDA and 1
at 273.0 K in the presence and absence of 1-hexene. Linear regressions
were made for the data points between 0 and 500 s. The two rates
shown represent the first of three sets of reactions.
S6 (Supporting Information). For each set of reactions, the
initial rate of MDA consumption was 138−151% faster in the
presence of hexene than in the absence of hexene. Overall,
these experiments clearly demonstrate that the presence of
hexene accelerated the rate of MDA consumption.
The pertinent elementary steps for metalloporphyrin-
catalyzed transformations with diazo reagents are shown in
Figure 7. In the analogous system using Rh(TTP)I, Kodadek
and co-workers demonstrated that coordination of EDA was
rapid and carbene formation was rate-limiting (k1, k−1 > k2).36
For Ir(TTP)CH3 catalysis, the rate-limiting step must be either
carbene formation (k2) or carbene transfer (k3, k4). If k2 was
3632
dx.doi.org/10.1021/om300135f | Organometallics 2012, 31, 3628−3635

Organometallics
Article
Scheme 4. Potential Routes for Ir(TTP)Cl(CO)-Catalyzed
Cyclopropanation
105 without significant deactivation. Cyclopropanation was
explored for a variety of electronically diverse olefins. In
general, electron-rich and sterically unencumbered substrates
reacted the most efficiently. The cyclopropanation mechanism
was explored in some depth for catalyst 1. Competition studies
using para-substituted styrenes produced a Hammett correla-
tion with ρ+ = −0.23, which indicates a buildup of positive
charge in the transition state. Using styrene and styrene-d8, an
inverse secondary isotope effect of 0.86 0.03 was observed,
suggesting that moderate olefin rehybridization occurs before
the transition state of the carbene transfer step. From these
data, the catalytic cycle was proposed to follow a metal-
loporphyrin carbene pathway similar to that reported for
Rh(TTP)I. However, unlike Rh(TTP)I, the rate-limiting step
for catalyst 1 appeared to be carbene transfer, as determined by
kinetics analyses. Similar mechanistic insights for hexacoordi-
nate compounds were not explored, due to the ambiguity in
competing pathways involving the dissociation of axial ligands.
The findings presented herein indicate that significant potential
exists for catalysis with iridium porphyrin complexes.
Figure 7. Elementary steps for the competitive cyclopropanation/
dimerization of alkyl diazoacetate and 1-hexene in the presence of
Ir(TTP)CH3 or Rh(TTP)X.
rate-limiting, the presence of hexene in the reaction would not
increase the rate of MDA consumption. In fact, MDA
consumption should be inhibited, because cyclopropanation
would compete with the dimerization pathway and would
consume only 1 equiv of MDA upon cyclopropanation. This
case is not supported by the above kinetic data. Rather, the rate
of MDA consumption increases in the presence of hexene,
indicating that carbene transfer must be rate limiting. In other
words, k3 and k4 are greater than k2. Furthermore, since
dimerization generally dominates even though hexene is
present in excess, k4 is likely greater than k3.
In contrast to the case for 1, the mechanism for
cyclopropanation with the hexavalent iridium porphyrin
complex Ir(TTP)Cl(CO) likely begins with reversible ligand
dissociation to provide a vacant site for diazo coordination.
Evidence supporting the formation of a heptacoordinate
iridium porphyrin complex is limited and therefore seems
unlikely in the present system.40 Assuming only one ligand
dissociates, two catalytic pathways are possible (Scheme 4). CO
dissociation would lead to a pathway with a neutral metal-
loporphyrin, analogous to the mechanism proposed for
Ir(TTP)CH3. Alternatively, chloride dissociation to generate a
formally cationic metalloporphyrin catalyst could be occurring.
However, the quantitative recovery of Ir(TTP)Cl(CO) after
catalysis suggests that CO dissociation is unlikely. On the other
hand, comparing the cyclopropanation reactivity and selectivity
of neutral (1) and cationic (7, 8) iridium catalysts suggests that
the chloride dissociation pathway may not be dominant either.
EXPERIMENTAL SECTION
■
All manipulations were performed under a dry nitrogen atmosphere.
Ir(TTP)Cl(CO), Ir(TTP)CH3, and Ir(TTP)(BF4)/Ir(TTP)(CO)-
(BF4) were synthesized according to previously reported meth-
ods.25,26,34 Methyl phenyldiazoacetate (MPDA) and methyl diazo-
acetate were prepared by procedures adapted from the literature.41,42
All olefin substrates were dried over 4 Å molecular sieves prior to use.
The styrene derivatives, except 4-nitrostyrene, were distilled and
stored at −20 °C prior to use. Trimethylamine N-oxide was sublimed
and stored in an inert-atmosphere glovebox. Methylene chloride and
tetrahydrofuran were deoxygenated and dried by passage through
columns of reduced copper and alumina. All other chemicals were
purchased as reagent grade and used without further purification.
NMR spectra were collected using Varian VXR 300 MHz, Varian VXR
400 MHz, or Bruker DRX 400 MHz spectrometers. Kinetic
measurements were done using a Bruker DRX 400 MHz spectrometer.
1H NMR peak positions were referenced against residual proton
resonances of deuterated solvents (δ (ppm): CDCl3, 7.26; CD2Cl2,
5.33). Gas chromatography was performed using a Shimadzu GC-17a
fitted with a HP-5 column (30 m × 0.25 μm). Ratios for kinetic
isotope data were determined using a Finnegan Magnum GC-MS
fitted with a HP-5 column (30 m × 0.25 μm) and a time-of-flight mass
analyzer. Mesitylene was used as an internal standard for GC yield
determinations. Column chromatography was performed using silica
CONCLUSIONS
■
In summary, a variety of neutral and cationic iridium(III)
porphyrin complexes were shown to be exceptionally active
catalysts for the cyclopropanation of olefins with diazo
compounds. Ir(TTP)CH3 (1) was extremely robust for the
conversion of EDA, producing cyclopropane TONs of 4.8 ×
3633
dx.doi.org/10.1021/om300135f | Organometallics 2012, 31, 3628−3635

Organometallics
Article
gel (40−63 μm) purchased from Sorbent Technologies. Character-
ization data for the cyclopropanation products ethyl 2-phenyl-
cyclopropanecarboxylate,43 ethyl 2-methyl-2-phenylcyclopropanecar-
boxylate,44 ethyl 1,1a,6,6a-tetrahydrocyclopropa[a]indene-1-carboxy-
late,45 ethyl 2-butylcyclopropanecarboxylate,4 4 ethyl
bicyclo[4.1.0]heptane-7-carboxylate,45 ethyl 2-methyl-3-phenylcyclo-
propanecarboxylate,46 diethyl cyclopropane-1,2-dicarboxylate,47 and
methyl 1,2-diphenylcyclopropanecarboxylate48 were previously re-
ported. The relative stereochemistry for ethyl 2-methyl-3-phenyl-
cyclopropanecarboxylate was confirmed by NOESY spectroscopy (see
the Supporting Information).
Slow volatilization of CD2Cl2 caused the MDA stock concentration to
fluctuate over the course of a couple of weeks. Because MDA and all
the products could be monitored during the reaction, data were
normalized to the total mass balance of MDA and its products.
Ir(TTP)Cl(NMe3) (3). In a glovebox, Ir(TTP)Cl(CO) (34.3 mg,
0.0371 mmol) and trimethylamine N-oxide (15 mg, 0.20 mmol) were
dissolved in CH2Cl2 (5 mL) and stirred at room temperature for 1 day.
Volatile components were removed in vacuo. Crystals were grown by
slow evaporation from CH2Cl2/MeOH and separated from solution
by decanting the solvent to give 3 in 87% yield (30.8 mg, 0.0322
1
mmol). H NMR (CDCl3): δ 8.71 (s, 8H), 8.15 (d, J = 7.2 Hz, 4H),
8.02 (d, J = 7.2 Hz, 4H), 7.56 (t, J = 6.0 Hz, 8H), 2.74 (s, 12H). UV−
vis (CH2Cl2): λmax (log ε) 412 (5.34), 522 (4.21), 555 nm (3.47).
Ir(TTP)Br(CO) (4). All Ir(TTP)X(CO) adducts were synthesized
using the following procedure with the appropriate tetrabutylammo-
nium salts. Halide abstraction was performed similarly to a previously
reported method.34 In a glovebox, a reaction vessel was charged with
Ir(TTP)Cl(CO) (14 mg, 0.015 mmol), excess silver tetrafluoroborate
(20 mg, 0.1 mmol), and CH2Cl2 (3 mL). The vessel was wrapped in
foil and the mixture stirred for 2 days at ambient temperature. The
resulting solids were removed by filtration. Excess tetrabutylammo-
nium bromide (5−10 equiv) was added to the filtrate, and the mixture
was stirred overnight. Aqueous workup and extraction in CH2Cl2
afforded 4 as a moderately pure red solid. Purification by column
chromatography on silica gel with hexanes and CH2Cl2 (1:2) as eluent
gave 4 in 54% yield (7.8 mg, 8.1 × 10−3 mmol). X-ray-quality single
crystals were grown by slow evaporation from CH2Cl2. Anal. Calcd for
C49H36BrIrN4O: C, 60.74; H, 3.74; N, 5.78. Found: C, 61.26; H, 3.81;
General Cyclopropanation Procedure. The catalyst (2 μmol)
was weighed as a solid and transferred to a flame-dried Schlenk flask
containing a stir bar. The flask was charged with olefin substrate (1
mmol), CH2Cl2 (2.0 mL), and mesitylene (internal standard, 0.144
mmol). If the reaction was carried out at 0 or −78 °C, it was placed in
an ice bath or dry ice/acetone bath, respectively. After 15 min was
allowed for temperature equilibration, reagent grade EDA (0.255
mmol) was added neat, dropwise via syringe over the course of 30 s.
During the reaction, aliquots were quenched with pyridine and
analyzed by GC. Once the reaction was complete, volatile components
1
were removed in vacuo. H NMR (CDCl3) was used to confirm the
consumption of EDA. Cyclopropanes could be isolated by column
chromatography on silica gel using hexanes and ethyl acetate (40:1) as
the eluent system. Catalyst preparations varied slightly for in situ
generated cationic complexes. In a glovebox, Ir(TTP)Cl(CO) (2
μmol) and excess silver tetrafluoroborate (50 μmol) were dissolved in
CH2Cl2 (2 mL). The vessel was wrapped in foil and the mixture stirred
at room temperature for 30 min before continuing with the above
cyclopropanation protocol.
Experiment To Measure Catalyst TON. A dry Schlenk flask was
charged with Ir(TTP)CH3 (4.13 × 10−6 mmol) from a stock solution
(1.00 mL, 4.13 × 10−3 M) in CH2Cl2. The porphyrin solution was
taken to dryness under an N2 stream before styrene (7.6 mmol),
mesitylene (internal standard, 0.2874 mmol), and CH2Cl2 (2.0 mL)
were added. EDA (approximately 0.72 mmol) was added after the
mixture was cooled to near −78 °C in an acetone/dry ice bath. The
cold bath was removed after 10 min, and stirring was continued at
room temperature for 50 min. An aliquot was quenched in pyridine
and analyzed by GC. The reaction vessel was recooled to −78 °C, and
the EDA addition was repeated. This method was repeated over the
course of three EDA additions. One hour after the final addition, the
reaction vessel was quenched with pyridine. Turnover numbers were
measured by GC, and complete conversion of EDA was observed by
1H NMR.
1
N, 5.71. H NMR (CDCl3): δ 8.93 (s, 8H), 8.12 (t, J = 7.5 Hz, 8H),
7.56 (d, J = 6.0 Hz, 8H), 2.1 (s, 12H). UV−vis (CH2Cl2): λmax (log ε)
424 (5.59), 535 (4.39), 570 nm (3.88).
Ir(TTP)I(CO) (5). Tetrabutylammonium iodide was substituted for
tetrabutylammonium bromide in the procedure outlined for 4.
Ir(TTP)Cl(CO) (17.2 mg, 0.0186 mmol) led to 5 in 45% yield (8.5
mg) after column chromatography. 1H NMR (CDCl3): δ 8.92 (s, 8H),
8.11 (m, 8H), 7.57 (t, J = 6.6 Hz, 8H), 2.71 (s, 12H). UV−vis
(CH2Cl2): λmax (log ε) 422 (5.51), 533 (4.36), 569 nm (3.82).
Ir(TTP)(SCN)(CO) (6). Tetrabutylammonium thiocyanate was
substituted for tetrabutylammonium bromide in the procedure
outlined for 4. Ir(TTP)Cl(CO) (27.5 mg, 0.0297 mmol) led to 6 in
72% yield (20.2 mg). Anal. Calcd for C50H36IrN5OS·1/2H2O: C, 62.81;
H, 3.90; N, 7.32. Found: C, 62.26; H, 3.34; N, 7.13. 1H NMR
(CDCl3): δ 8.98 (s, 8H), 8.13 (dd, J = 15.2 Hz, 7.6 Hz, 8H), 7.60 (t, J
= 6.4 Hz, 8H), 2.73 (s, 12H). UV−vis (CH2Cl2): λmax (log ε) 420
(5.59), 531 (4.42), 566 nm (3.84).
X-ray Single-Crystal Structure Determination. The crystal
evaluation and data collection were performed at 173 K on a Bruker
APEX II CCD diffractometer using Mo Kα (λ = 0.710 73 Å). Full
sphere data with 0.3° frame width were collected until a resolution of
0.74 Å. The absorption correction was based on a fit of a spherical
harmonic function to the empirical transmission surface, as sampled by
multiple equivalent measurements.49 Structures were solved using
direct methods and were refined in full-matrix anisotropic approx-
imation for all non-hydrogen atoms. All hydrogen atoms were placed
in the structure factor calculations at idealized positions and refined
using a “riding model”. The Uiso(H) values were set at 1.5 times the
Ueq value of the carrier atom. All calculations were performed using the
APEX II software package.50,51 In complexes 4 and 5, the axial ligands
were disordered by inversion and distances in these ligands were
constrained during refinement.
General Competition Experiment. This general method was
used for all competition studies, including that with styrene-d8. A
CH2Cl2 stock solution (200 μL, 4.93 × 10−3 M) of Ir(TTP)CH3
(0.986 μmol) was transferred to a Schlenk flask and taken to dryness
under an N2 stream. Substrate A (0.987 mmol), substrate B (0.987
mmol), mesitylene (internal standard, 0.1437 mmol), and CH2Cl2 (1.0
mL) were added to the flask. The solution was allowed to equilibrate
to the desired reaction temperature for 15 min. EDA (0.2 mmol) was
added dropwise by syringe over the course of 30 s. After it was stirred
for 1 h, the reaction mixture was quenched with pyridine and analyzed
by GC.
Setup for Kinetic Measurement Experiments. An NMR tube
was charged with 1 (ranging from 1.05 × 10−4 to 4.20 × 10−4 μmol)
from a 5.25 × 10−6 M CH2Cl2 stock solution and taken to dryness
under reduced vacuum. The tube was taken into a glovebox and
loaded with 1-hexene or styrene (ranging from 27.8 to 262 μmol) and
mesitylene (2.82 μmol), and the mixture was diluted to a total volume
of 420 μL with CD2Cl2. Then, the tube was fitted with a septum,
cooled to 273 K, and taken to the NMR spectrometer. Spectrometer
settings were prepared, including temperature equilibration to 273.0 K,
prior to diazo addition. Finally, MDA (ranging from 11.6 to 56.8
umol) in a CD2Cl2 stock solution (dried over molecular sieves), which
was chilled in an ice bath, was added quickly. Data collection began ca.
1 min after addition. Adding equimolar portions of MDA was
challenging, despite storing the stock solution in a −20 °C freezer.
ASSOCIATED CONTENT
* Supporting Information
■
S
Figures, tables, and CIF files giving Hammett plot data, kinetic
isotope data, derivations of rate equations, initial rate data at
different concentrations of the reagents and catalyst, the
NOESY spectrum for ethyl 2-methyl-3-phenylcyclopropane-
carboxylate, and complete structural data for 3−6. This material
is available free of charge via the Internet at http://pubs.acs.org.
3634
dx.doi.org/10.1021/om300135f | Organometallics 2012, 31, 3628−3635

Organometallics
Article
(32) Burnett, M. N.; Johnson, C. K. ORTEP-III: Oak Ridge Thermal
Ellipsoid Plot Program for Crystal Structure Illustrations, Oak Ridge
National Laboratory Report ORNL-6895, 1996.
(33) Rohde, J. Z. Anorg. Allg. Chem. 1997, 623, 1774.
(34) Song, X.; Chan, K. S. Organometallics 2007, 26, 965.
(35) Chen, Y.; Ruppel, J. V.; Zhang, X. P. J. Am. Chem. Soc. 2007,
129, 12074.
(36) Maxwell, J.; Brown, K.; Bartley, D.; Kodadek, T. Science 1992,
256, 1544.
(37) Lai, T.-S.; Chan, F.-Y.; So, P.-K.; Ma, D.-L.; Wong, K.-Y.; Che,
C.-M. Dalton Trans. 2006, 4845.
(38) Galardon, E.; Le Maux, P.; Simonneaux, G. Tetrahedron 2000,
56, 615.
CCDC 864501−864504 also contain supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: kwoo@iastate.edu.
Notes
The authors declare no competing financial interest.
(39) Brown, K. C.; Kodadek, T. J. Am. Chem. Soc. 1992, 114, 8336.
(40) Cheung, C. W.; Chan, K. S. Organometallics 2008, 27, 3043.
(41) Zhao, W.-J.; Yan, M.; Huang, D.; Ji, S.-J. Tetrahedron 2005, 61,
5585.
(42) Searle, N. E. Organic Syntheses Wiley: New York, 1956; Collect.
Vol. 36.
ACKNOWLEDGMENTS
■
NSF Grant No. CHE-0809901 and a GAANN fellowship
(B.J.A.) provided partial financial support for this work.
(43) Chen, Y.; Fields, K. B.; Zhang, X. P. J. Am. Chem. Soc. 2004, 126,
14718.
(44) Barrett, A. G. M.; Braddock, D. C.; Lenoir, I.; Tone, H. J. Org.
Chem. 2001, 66, 8260.
(45) Rosenberg, M. L.; Krivokapic, A.; Tilset, M. Org. Lett. 2009, 11,
547.
(46) Feldman, K. S.; Simpson, R. E. J. Am. Chem. Soc. 1989, 111,
4878.
(47) Doyle, M. P.; Dorow, R. L.; Tamblyn, W. H. J. Org. Chem. 1982,
47, 4059.
(48) Tarwade, V.; Liu, X.; Yan, N.; Fox, J. M. J. Am. Chem. Soc. 2009,
131, 5382.
(49) Blessing, R. Acta Crystallogr., Sect. A 1995, 51, 33.
(50) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112.
(51) APEX2 Version 2.2; Bruker AXS Inc., Madison, WI, 2007.
REFERENCES
■
(1) Che, C.-M.; Zhou, C.-Y.; Wong, E. In Topics in Organometallic
Chemistry; Plietker, B., Ed.; Springer: Berlin/Heidelberg, Germany:
2011; Vol. 33, p 111.
(2) Lu, H.; Zhang, X. P. Chem. Soc. Rev. 2011, 40, 1899.
(3) Nicolas, I.; Le Maux, P.; Simonneaux, G. Coord. Chem. Rev. 2008,
252, 727.
(4) Davies, H. M. L.; Panaro, S. A. Tetrahedron 2000, 56, 4871.
(5) Doyle, M. P.; Duffy, R.; Ratnikov, M.; Zhou, L. Chem. Rev. 2010,
110, 704.
́
(6) Dıaz-Requejo, M. M.; Belderrain, T. R.; Nicasio, M. C.; Perez, P.
J. Dalton Trans. 2006, 5559.
́
(7) Caballero, A.; Prieto, A.; Dıaz-Requejo, M. M.; Per
Inorg. Chem. 2009, 2009, 1137.
(8) Doyle, M. P. Angew. Chem., Int. Ed. 2009, 48, 850.
(9) Ferrand, Y.; Le Maux, P.; Simonneaux, G. Org. Lett. 2004, 6,
3211.
́
ez, P. J. Eur. J.
(10) Thu, H.-Y.; Tong, G. S.-M.; Huang, J.-S.; Chan, S. L.-F.; Deng,
Q.-H.; Che, C.-M. Angew. Chem., Int. Ed. Engl. 2008, 47, 9747.
(11) Hamaker, C. G.; Mirafzal, G. A.; Woo, L. K. Organometallics
2001, 20, 5171.
(12) Zhou, C.-Y.; Che, C. Synlett 2010, 2010, 2681.
(13) Che, C.-M.; Huang, J.-S. Coord. Chem. Rev. 2002, 231, 151.
(14) Xu, X.; Lu, H.; Ruppel, J. V.; Cui, X.; Lopez de Mesa, S.; Wojtas,
L.; Zhang, X. P. J. Am. Chem. Soc. 2011, 133, 15292.
(15) Intrieri, D.; Caselli, A.; Gallo, E. Eur. J. Inorg. Chem. 2011, 2011,
5071.
(16) Mbuvi, H. M.; Woo, L. K. Organometallics 2008, 27, 637.
(17) Mbuvi, H.; Woo, L. K. J. Porphyrins Phthalocyanines 2010, 14,
284.
(18) Cheng, G.; Mirafzal, G. A.; Woo, L. K. Organometallics 2003, 22,
1468.
(19) Callot, H. Tetrahedron 1982, 38, 2365.
(20) Wolf, J. R.; Hamaker, C. G.; Djukic, J.-P.; Kodadek, T.; Woo, L.
K. J. Am. Chem. Soc. 1995, 117, 9194.
(21) Callot, H. Tetrahedron Lett. 1980, 21, 3489.
(22) Callot, H. J.; Metz, F. Tetrahedron Lett. 1982, 23, 4321.
(23) Sadasivan, N. J. Inorg. Nucl. Chem. 1968, 30, 591.
(24) Bartley, D. W.; Kodadek, T. J. Am. Chem. Soc. 1993, 115, 1656.
(25) Ogoshi, H.; Setsune, J.; Yoshida, Z. J. Organomet. Chem. 1978,
159, 317.
(26) Yeung, S. K.; Chan, K. S. Organometallics 2005, 24, 6426.
(27) Kanchiku, S.; Suematsu, H.; Matsumoto, K.; Uchida, T.;
Katsuki, T. Angew. Chem., Int. Ed. 2007, 46, 3889.
(28) Berkessel, A.; Kaiser, P.; Lex, J. Chem. Eur. J. 2003, 9, 4746.
(29) O’Malley, S.; Kodadek, T. Tetrahedron Lett. 1991, 32, 2445.
(30) Che, C.-M.; Huang, J.-S.; Lee, F.-W.; Li, Y.; Lai, T.-S.; Kwong,
H.-L.; Teng, P.-F.; Lee, W.-S.; Lo, W.-C.; Peng, S.-M.; Zhou, Z.-Y. J.
Am. Chem. Soc. 2001, 123, 4119.
(31) Shvo, Y.; Hazum, E. J. Chem. Soc., Chem. Commun. 1975, 829.
3635
dx.doi.org/10.1021/om300135f | Organometallics 2012, 31, 3628−3635