The mechanism of hydroaminoalkylation catalyzed by group 5 metal binaphtholate complexes

Article abstract of DOI:10.1021/ja211945m

The intermolecular hydroaminoalkylation of unactivated alkenes and vinyl arenes with secondary amines occurs readily in the presence of tantalum and niobium binaphtholate catalysts with high regio- and enantioselectivity (up to 98% ee). Mechanistic studie

Full text of DOI:10.1021/ja211945m

Article
pubs.acs.org/JACS
The Mechanism of Hydroaminoalkylation Catalyzed by Group 5
Metal Binaphtholate Complexes
Alexander L. Reznichenko and Kai C. Hultzsch*
Department of Chemistry and Chemical Biology, Rutgers, The State University of New Jersey, 610 Taylor Road, Piscataway, New
Jersey 08854-8087, United States
S
* Supporting Information
ABSTRACT: The intermolecular hydroaminoalkylation of unac-
tivated alkenes and vinyl arenes with secondary amines occurs
readily in the presence of tantalum and niobium binaphtholate
catalysts with high regio- and enantioselectivity (up to 98% ee).
Mechanistic studies have been conducted in order to determine the
kinetic order of the reaction in all reagents and elucidate the rate-
and stereodetermining steps. The effects of substrate steric and electronic properties on the overall reaction rate have been
evaluated. The reaction is first order in amine and the catalyst, while exhibiting saturation in alkene at high alkene concentration.
Unproductive reaction events including reversible amine binding and arene C−H activation have been observed. The formation
of the metallaaziridine is a fast reversible nondissociative process and the overall reaction rate is limited either by amide exchange
or alkene insertion, as supported by reaction kinetics, kinetic isotope effects, and isotopic labeling studies. These results suggest
that the catalytic activity can be enhanced by employing a more electron-deficient ligand backbone.
Chart 1. Selected Group 4 and Group 5 Metal
Hydroaminoalkylation Catalysts
INTRODUCTION
■
The catalytic inter- and intramolecular hydroaminoalkylation is
an atom-economical process for the synthesis of industrial and
pharmaceutical valuable amines in which an α-amino C−H
moiety is added across an unsaturated carbon−carbon bond
(Scheme 1).1−3 While initial reports on catalytic hydro-
Scheme 1. Inter- and Intramolecular Hydroaminoalkylation
aminoalkylation surfaced more than 30 years ago,4 this reaction
has become a subject of intensive research efforts only
recently.1,5−8 Although few examples of main group metal-
catalyzed α-C−H addition of amines to alkenes are known,9 a
majority of hydroaminoalkylation catalysts are based on d0
group 4 and group 5 metal complexes (Chart 1).
More recently, Schafer and co-workers have developed a
family of various amidate tantalum catalysts, such as III and
similar structures.5c The observation of hydroaminoalkylation
as a side reaction of hydroamination6a has sparked the
development of group 4 metal hydroaminoalkylation catalysts.
The homoleptic tetrabenzyl titanium IVa6b,c or bis(indenyl)
In line with earlier observations,4 Herzon and Hartwig have
demonstrated that the homoleptic tantalum amide I5a and the
heteroleptic chloro bisamide II5b are active catalysts for the
intermolecular hydroaminoalkylation of unactivated olefins
with secondary amines, with the latter catalyst being more
reactive and applicable to a broader range of substrates,
including dialkylamines.
Received: December 22, 2011
Published: January 19, 2012
© 2012 American Chemical Society
3300
dx.doi.org/10.1021/ja211945m | J. Am. Chem. Soc. 2012, 134, 3300−3311

Journal of the American Chemical Society
Article
complex V6e was shown to catalyze intra- and intermolecular
hydroaminoalkylations with notable reactivity for primary
alkenyl amines in the intramolecular process. Catalysts for
asymmetric intermolecular hydroaminoalkylation have been
developed employing κ1-5c and κ1-/κ2-amidate5d,e frameworks.
In line with the selectivity trends previously reported for the
asymmetric hydroamination with related zirconium amidate
complexes,10 the sterically more hindered chelating κ2-amidates
VIII displayed higher enantioselectivities (up to 93% ee for
selected substrates).5d,e However, in contrast to the reactivity
trends observed in intramolecular hydroaminations, the steri-
cally and electronically more saturated complex VIII was
reported to be more active than the sterically more accessible
and electronically unsaturated complex VII.
aggregation and a complex kinetic behavior at higher catalyst
concentrations.6f
Recently, we have communicated8 that 3,3′-silyl-substituted
binaphtholate niobium and tantalum complexes12 (Chart 2) are
Chart 2. 3,3′-Silyl-Substituted Binaphtholate Niobium and
Tantalum Catalysts for Asymmetric Intermolecular
Hydroaminoalkylation
The general mechanism of catalytic hydroaminoalkylation
was proposed in the early studies dealing with group 5 metal
catalysts (Scheme 2).4b The key step of the catalytic cycle is
Scheme 2. Mechanism of Intermolecular
Hydroaminoalkylation
efficient catalysts for the intermolecular hydroaminoalkylation.
The lack of more detailed mechanistic and kinetic information
prompted us to study the mechanism of this transformation in
greater depth. Herein we present the results from our
investigations into the scope and limitations of catalysts 1-M
(M = Nb, Ta) in the hydroaminoalkylation reaction and a
detailed analysis of the reaction mechanism.
RESULTS AND DISCUSSION
■
Initial Catalyst Screening. The reaction of 1-octene with
N-methylaniline was chosen as a model reaction to evaluate the
different tantalum and niobium catalysts (Table 1). Complexes
1a−d showed good activity at 150 °C to form the branched
hydroaminoalkylation products exclusively. The rate of the
reaction generally increases with decreasing steric demand of
the silyl substituent of the binaphtholate ligand. The sterically
least demanding trimethylsilyl-substituted complexes 1d-Ta
and 1d-Nb displayed the highest activity, but also the lowest
selectivity in this series. Sterically more encumbered catalysts
with triphenylsilyl and benzyldiphenylsilyl substituents pro-
duced only traces of product even at 170 °C.8 Under our
standard conditions at 150 °C, the methyldiphenylsilyl-
substituted complexes 1b-Nb and 1b-Ta exhibited the highest
selectivity (Table 1, entries 3 and 5), whereas the slightly more
demanding methyldiisopropylsilyl-substituted complexes 1a-Nb
and 1a-Ta displayed slightly lower selectivities than 1b. The
niobium complexes were generally more active than their
tantalum analogues, but the enantioselectivities were quite
comparable in most cases. The discrepancy of reactivity of the
tantalum and niobium complexes is even more pronounced at
lower temperatures, as the tantalum complexes gave poor
reactivity below 140 °C, while the niobium catalyst 1b-Nb was
reactive even at 100 °C to give the highest selectivity of 81% ee
for this model reaction (Table 1, entry 7). Notably, 1b-Nb
appears to be more active at 100 °C than the chiral catalyst VII
at 130 °C.5c The relative reactivity of tantalum and niobium
complexes depends on the catalyst structure; while in some
instances niobium was found to be superior to tantalum,4b
niobium was claimed to be inactive in other.5d,e
believed to be the C−H activation of the bisamide leading to
the metallaaziridine A (Scheme 2, step 1). Subsequent alkene
insertion (Scheme 2, step 2), protonolysis of the metal−carbon
bond (Scheme 2, step 3), and amide exchange (Scheme 2, step
4) completes the catalytic cycle. The insertive reactivity of
metallaaziridines has already been used in other synthetic
applications.11 Only few mechanistic details on the reversibility
and relative feasibility of steps 1−4 are available. Herzon and
Hartwig have demonstrated that the formation of the
metallaaziridine A is reversible in the reaction catalyzed by
I,5a whereas step 1 is virtually irreversible in the case of catalyst
II,5b as suggested by isotopic labeling studies. The same studies
implicate that steps 2 and 3 should be fast, as the catalyst has no
metal−carbon bond in its resting state. Recently, Doye and co-
workers reported that the intramolecular hydroaminoalkylation
of primary aminoalkenes catalyzed by Ti(NMe2)4 (IVb)
involves a rate-limiting C−H activation step, as indicated by a
pronounced kinetic isotope effect.6f The lack of a sterically
demanding spectator ligand in IVb resulted in catalyst
The most reactive catalyst 1b-Nb was then used to evaluate
the scope of the reaction at 140 °C (optimal reaction time) and
100 °C (optimal enantioselectivity).
3301
dx.doi.org/10.1021/ja211945m | J. Am. Chem. Soc. 2012, 134, 3300−3311

Journal of the American Chemical Society
Article
Table 1. Catalytic Asymmetric Hydroaminoalkylation of 1-Octene with N-Methylaniline using Binaphtholate Tantalum and
a
Niobium Complexes
b
c
entry
cat.
T, °C; t, h
yield, %
ee, %
1
1a-Ta
1a-Nb
1b-Ta
1b-Ta
1b-Nb
1b-Nb
1b-Nb
1c-Ta
1c-Nb
1d-Ta
1d-Nb
150; 31
150; 20
150; 14
130; 65
150; 7
89
93
88
89
85
91
92
79
72
87
91
65
60
72
73
72
79
81
49
61
34
54
2
3
4
5
6
130; 58
100; 105
150; 12
150; 8
7
8
9
10
11
150; 5
150; 3
a
b
Conditions: N-methylaniline (0.2 mmol), 1-octene (0.4 mmol), cat. (0.01 mmol, 5 mol %), C6D6 (0.25 mL), Ar atm. Isolated yield of the
c
corresponding N-benzamide. Determined by chiral HPLC of the N-benzamide.
Scope in Amine. As shown in Table 2, various N-
methylanilines bearing both electron-withdrawing and electron-
donating substituents exhibit similar reactivities and selectivities
in reactions involving terminal olefins. The niobium complex
1b-Nb displayed systematically higher reactivity in comparison
to its tantalum analogue 1b-Ta. The unsymmetric dialkylamine
N-benzylmethylamine (Scheme 3) was less reactive than
arylalkylamines. Here, the difference in reactivity of niobium
and tantalum was even more pronounced, with the niobium
complex reacting at 150 °C, while tantalum required a
significantly higher reaction temperature of 180 °C. The
reactions of N-methylaniline derivatives provide enantioselec-
tivities in a relatively narrow range of 66−80% ee at 140 °C,
while the reaction of N-benzylmethylamine was significantly
less selective. None of the reactions produced any detectable
amount of the linear addition product; however, the reaction of
N-benzylmethylamine produced the regioisomer 11b originat-
ing from activation of the benzylic position as a minor
byproduct (Scheme 3). The regioselective formation of 11a is
in agreement with observations for II,5b but is in remarkable
contrast to selectivities found for the amidate III.5c
The sluggish reactivity of a secondary alkylamino group is
further manifested by the absence of product formation in the
reaction of 1-octene with N-ethylaniline, pyrrolidine, and
tetrahydroisoquinoline at 150−180 °C. It should be noted
that these and similar amines lacking an N-methyl group were
reactive with tantalum amidate catalyst III5c and Ind2TiMe2
(V).6e While multiple mechanistic scenarios may account for
the enhanced regioselectivity of catalysts 1b-Ta and 1b-Nb, it is
notable that their preference of activating the N-methyl group is
greater than that of other known group 4 and group 5 metal
catalysts.
cyclohexene, were unreactive; however, the more activated
norbornene provided the corresponding product 8, though
only in moderate yield due to formation of poly(norbornene)
as a side reaction. Methylene cyclohexane exhibited lower
reactivity because of the gem-disubstituted double bond. Other
1,1-disubstituted alkenes, such as α-methylstyrene, were
unreactive. Notably, all reactions proceeded with excellent
branched to linear regioselectivity exceeding 50:1. This also
includes the reaction of 2d with styrene at 100 °C, which
produced the branched product (S)-(−)-1014 with >30:1
regioselectivity in 71% ee when using (R)-1-Nb (Table 2, entry
21). The highest enantioselectivity of 98% ee was observed in
the reaction of N-methylaniline (2a) with trimethyl(vinyl)silane
at 100 °C (Table 2, entry 17).
Isotopic labeling studies have been used previously in
intermolecular (I5a and II5b) and intramolecular (IVb6f)
hydroaminoalkylations. We have attempted to use this
approach in order to gain further insight into the mechanism
of the reaction catalyzed by 1. Reaction of N-(methyl-d3)-
aniline with vinylcyclohexane catalyzed by 1b-Ta led to a
product with significantly depleted deuterium content at the
methylene group and significant deuterium incorporation into
the ortho-positions of the aryl ring (Scheme 4). This leakage of
deuterium from the methylene group was interpreted as an
indicator of a reversible metallaaziridine formation (Scheme 2,
step 1),5a,b whereas the incorporation of deuterium into the aryl
ring should result from the intramolecular C−H activation via
cyclometalation (Scheme 4).5a
An important feature is the low incorporation of deuterium
into the methyl group compared to that of the ortho-position of
the aryl ring. Although 1.67 equiv deuterium is released from
the original CD3 group, only 33% of the methyl groups in the
product are labeled (= 11% total deuteration of CH3),
suggesting that the protonation (Scheme 2, step 3) proceeds
predominantly with an N-proteo amine rather than an N-
deutero amine, which is in agreement with observations for I5a
and IVb.6f The reaction of N-(methyl-d3)-aniline with a
catalytic amount of 1b-Ta (1 mol %) in the absence of alkene
resulted in a similar incorporation of deuterium into the ortho-
position of the aniline (42% deuterium after 2 h at 150 °C),
indicating that the reversible cyclometalation is likely preceding
Scope in Alkene. Linear, α- and β-branched terminal
olefins exhibit comparable reactivities in hydroaminoalkylations
(Table 2, entries 1−20), which seems to be remarkable in
comparison to a significant diminished reactivity of α-branched
terminal olefins observed in the intermolecular hydroamina-
tion.13 Higher enantioselectivities were observed for sterically
more demanding alkenes (Table 2, entries 2, 3, 11, and 16) as
well at lower reaction temperatures (Table 2, entries 4, 6, 10,
12, and 17). Unactivated 1,2-disubstituted alkenes, such as
3302
dx.doi.org/10.1021/ja211945m | J. Am. Chem. Soc. 2012, 134, 3300−3311

Journal of the American Chemical Society
Article
a
Table 2. Substrate Scope in the Intermolecular Asymmetric Hydroaminoalkylation Catalyzed by 1b-M (M = Ta, Nb)
a
b
Reaction conditions: amine (0.2 mmol), alkene (0.4 mmol), 1b-M (0.01 mmol, 5 mol %), C6D6, Ar atm. Determined by chiral HPLC of the
c
d
1
corresponding N-benzamide. Determined by H NMR spectroscopy of (S)-Mosher amide. Determined by chiral HPLC.
Scheme 3. Hydroaminoalkylation of N-Benzylmethylamine
(Scheme 5). Thus, it seems that the lower reactivity of tantalum
compared to niobium can be attributed in part to the formation
of a cyclometalated species that removes the catalyst from its
active state.
The absence of ortho-metalation for the niobium catalyst
supports the hypothesis that this process is nonproductive in
character. The retention of deuterium in the methylene group
is similar to that observed with II,5b which was interpreted as a
result of irreversible C−H activation (Scheme 2, step 1)
followed by rapid olefin insertion (Scheme 2, step 2). However,
one feature of the reaction catalyzed by 1b-Nb is
unprecedented, which is almost complete (96%) deuteration
of the methyl group of the product, while the reaction with
catalyst II resulted in only 30% deuteration in our hands. Even
more strikingly, when the reaction was stopped at ∼20%
conversion and starting material and product were recovered,
high deuterium incorporation into the methyl group of the
product (91%) along with almost complete retention of
the insertion (Scheme 2, step 2) and is not necessarily part of
the catalytic cycle.
A strikingly different outcome was observed when 1b-Nb was
used instead of 1b-Ta. In this case, the reaction of N-(methyl-
d3)-aniline with vinylcyclohexane proceeded with almost
complete retention of deuterium in the methylene group and
no ortho-metalation of the aniline aromatic ring was observed
3303
dx.doi.org/10.1021/ja211945m | J. Am. Chem. Soc. 2012, 134, 3300−3311

Journal of the American Chemical Society
Article
indicates that protonolysis of the azametallacyclopentane
intermediate (Scheme 2, step 3) occurs almost exclusively
with an N-deutero amine. Thus, the deuterium from the N-
(methyl-d3)-aniline starting material is transferred with high
selectivity to the methyl group in the product. Although direct
catalytic protonolysis of the metal−alkyl bond with another C−
H bond via σ-bond metathesis is known,15 we propose a more
realistic mechanistic scenario involving the nondissociative
formation of the metallaaziridine A*PhNHMe (Scheme 6, step
1), which retains the coordinated amine molecule during the
insertion (Scheme 6, step 2). The subsequent protonolysis
(Scheme 6, step 3) proceeds in an intramolecular fashion
without participation of an external amine.
Scheme 4
Several important consequences arise from the mechanism
depicted in Scheme 6. First, the lack of deuterium scrambling
from the methylene position cannot serve as a proof5b of an
irreversible character of step 1. In fact, since a nondissociative
metallaaziridine formation can be reversed in a fully degenerate
process, reversible metallaaziridine formation will still result in
retention of deuterium. Indeed, prolonged heating of 1b-M (M
= Ta, Nb) in the presence or absence of a slight excess of 2a
(0−10 equiv) to 100−130 °C did not result in conversion to,
or even observation of, the metallaaziridine A*PhNHMe
Scheme 5
1
according to H NMR spectroscopy. However, the catalytic
reactions proceeded smoothly in the presence of an alkene,
which is another indicator that step 1 is reversible and the
equilibrium is shifted toward the bisamide species. Second, the
presence of a coordinated amine during the insertion step 2 can
result in a concerted protonolysis-assisted insertion, so that
steps 2 and 3 could merge into a single, concerted reaction step.
This process would be reminiscent to a number of early
transition metal hydroamination catalysts that have been
proposed to operate via a protonolysis-assisted alkene insertion
into a metal−nitrogen bond.16 Quenching of a reaction mixture
containing 25 mol % 1b-M (M = Ta, Nb) at ca. 40%
conversion with 12 M DCl/D2O did not result in any
noticeable deuterium incorporation into the methyl group of
either product 6a or starting material 2a, further suggesting that
the concentration of metallaaziridine and metal−alkyl species
remains low over the course of the reaction.
The incomplete (<98%) deuteration of the product methyl
group (Scheme 5) could arise from intermolecular protonation
of the metal−carbon bond by an external amine. However, no
measurable deuterium incorporation was observed in the
reaction of N-deutero-N-methyl-aniline with vinylcyclohexane
using 1 mol % 1b-Nb (Scheme 7). On the other hand, the same
reaction with N-(methyl-d3)-aniline using 7 mol % 1b-Nb led
to a diminished deuteration of the methyl group (71%) and
depletion of deuterium from the methylene group (85%).
These observations suggest that the residual protonation results
deuterium in the starting material (97%) was observed
(Scheme 5). Unfortunately, direct quantitative determination
of the extend of deuteration of the amine NH group is
complicated due to broad and overlapping signals in the
reaction mixture and loss of the amine deuterium during
workup. However, complete retention of deuterium in the
molecule of the starting material suggests that most of the N−
H groups are still undeuterated at low conversion. The high
degree of monodeuteration of the methyl group in the product
Scheme 6. Nondissociative Metallaaziridine Formation and Intramolecular Proton Delivery
3304
dx.doi.org/10.1021/ja211945m | J. Am. Chem. Soc. 2012, 134, 3300−3311

Journal of the American Chemical Society
Article
Scheme 7
Scheme 9
from C−H activation of the dimethylamido groups of the
precatalyst rather than from outer-sphere protonolysis.
As mentioned previously, a notable feature of catalysts 1 is a
pronounced preference for activation of primary over secondary
C−H bonds resulting in a lack of reactivity toward substrates
without an N-methyl group. Previously, an isotopic labeling
study with II suggested that hampered C−H activation limited
the hydroaminoalkylation of secondary C−H bonds.5b In our
case, although no hydroaminoalkylation product was formed
with N-ethyl-aniline using either 1b-Ta or 1b-Nb, N-(ethyl-d5)-
aniline displayed some deuterium scrambling after 50 h at 150
°C in the presence of 1b-Ta (Scheme 8).
expected result, the significant perturbation of enantioselectivity
in the formation of 12-D caused by the isotopic substitution is
striking. As the stereodetermining step itself (insertion, step 2)
does not directly involve the cleavage or formation of a C−H
bond as the protonolysis (step 3) does, the perturbation of
enantioselectivity suggests that step 2 is reversible21,22 and is
faster than intramolecular protonolysis (Scheme 10, a). The
Scheme 10. Scenarios for Isotopic Perturbation of
Enantioselectivity in the Hydroaminoalkylation of
Norbornene: (a) Reversible Olefin Insertion and Slow
Intramolecular Protonolysis. (b) Concerted Insertion-
Protonolysis
Scheme 8
Despite the rate of deuterium exchange being considerably
slower than in case of N-(methyl-d3)-aniline, it seems that C−H
activation of a secondary C−H bond is not prohibited. The lack
of reactivity of secondary C−H bonds should therefore result
from hampered alkene insertion (step 2) rather than hampered
C−H activation (step 1).
The stereochemistry of olefin insertion into a metal−ligand
bond can provide invaluable information on the details of the
reaction mechanism. For example, the stereochemistry of the
inter-17 and intramolecular18 olefin insertion into a palladium−
amide bond, as well as the intramolecular gold-19 and Brønsted
acid-catalyzed20 cyclization of alkenyl sulfonamides, has been
investigated in the context of the hydroamination reaction.
We have performed the reaction of norbornene with N-
(methyl-d3)-aniline in the presence of 1b-Nb (Scheme 9).
Consistent with our previous findings, high retention of
deuterium in the methylene position and high deuteration of
the cyclic methylene group were observed. The delivery of
deuterium to the ring proceeded exclusively to the exo-position,
thus yielding the syn-insertion isomer of 8 (isolated as
benzylamide 12-D). These results are in agreement with
observations of Milstein et al. on iridium-catalyzed hydro-
amination of norbornene with aniline.21 Most interestingly, the
deuterated benzylamide 12-D was formed with significant
lower enantioselectivity (30% ee) in comparison to the
undeuterated congener 12 (59% ee).
While the syn-addition to the olefin is consistent with the
inner-sphere olefin insertion mechanism and, therefore, is an
3305
dx.doi.org/10.1021/ja211945m | J. Am. Chem. Soc. 2012, 134, 3300−3311

Journal of the American Chemical Society
Article
fact that this phenomenon is observed in the case of
norbornene is likely due to the fact that the intramolecular
protonolysis (step 3) involves proton transfer to a more
sterically hindered methine carbon as opposed to the
methylene group in the case of α-olefins. Indeed, no
perturbation of enantioselectivity was observed in the case of
α-olefins: 73% ee versus 73% ee for 4a-D and 4a and 79.5% ee
versus 80% ee for 6a-D and 6a, respectively, thus strongly
suggesting an effectively irreversible insertion (step 2) and fast
intramolecular protonolysis (step 3).
An alternative explanation for the observed perturbation of
enantioselectivity would involve a concerted insertion-proto-
nolysis step with participation of a coordinated amine, which is
both stereodetermining and involves the conversion of an N−
H bond into a C−H bond (Scheme 10, b). This scenario was
used to explain the isotopic perturbation of enantioselectivity in
the zirconium-catalyzed intramolecular hydroamination of
aminoalkenes.16d,h However, this proposal does not explain
the unique behavior of norbornene with respect to the
perturbation of enantioselectivity as found in this study.
Furthermore, the concerted insertion-protonolysis sequence
recently proposed for a magnesium hydroamination catalyst
system16c was found to proceed via fast reversible insertion and
rate-determining protonolysis according to DFT calculations.23
Empirical Rate Laws. To determine the empirical rate laws,
a kinetic investigation of the hydroaminoalkylation of N-
methyl-anilines 2a−d with 1-octene and vinylcyclohexane using
1b-Nb was performed. The reaction displays a relatively well-
behaved first-order kinetic behavior in 2 over the course of 2−3
half-lives in the presence of an excess of alkene. The observed
first-order rate constant is linearly dependent on the catalyst
concentration 1b-Nb (Figure 1), therefore, indicating a first-
order dependence in 2 and 1b-Nb.
Figure 2. Dependence of pseudo-first-order rate constant on alkene
○
concentration for the reaction of 1-octene (0.26−3.75 M) ( ) and
●
vinylcyclohexane (0.29−3.75 M) ( ) with PMPNHMe (2d) (0.29 M)
in the presence of (R)-1b-Nb (0.017 M) in C6D6 at 150 °C. The solid
lines represent the least-squares fit to the data points; dotted lines are
drawn as a guide for the eye.
Essentially no rate dependence on the concentration of 1-
octene was observed, which indicates that alkene insertion
(Scheme 6, step 2) is not rate-determining and is proceeding
relatively fast within a large concentration interval of the alkene.
Saturation behavior was observed in case of the sterically more
demanding vinylcyclohexane with the rate being close to zero-
order in alkene at high alkene concentrations, while first-order
dependence was apparent at low alkene concentrations. Thus,
insertion step 2 can serve as a rate-determining step as well. On
the basis of these results, an empirical rate law can be proposed
(eq 1).
α
rate = k[1b‐Nb][2][alkene]
with 0 ≤ α ≤ 1
(1)
The insertion step (Scheme 6, step 2) is sensitive to the
steric features of the alkene and 2a−d react with vinyl-
cyclohexane and 1-octene with comparable rates in the
saturation regime. The first-order rate dependence on the
concentration of N-methyl-anilines 2a−d (Figure 3, see also
Figure 1. Dependence of observed first-order reaction constant on
catalyst concentration for the reaction of 1-octene (0.613 M) with
PMPNHMe (2d) (0.20 M) in the presence of (R)-1b-Nb (0.003−
0.022M) in C6D6 at 150 °C. The line represents the least-squares fit to
the data points.
Figure 3. First-order kinetic plots for the reaction of 1-octene (0.613
M) with 4-FC6H4NHMe (2b) (0.144−0.579 M) in the presence of
(R)-1b-Nb (0.017 M) in C6D6 at 150 °C. The solid lines represent the
least-squares fit to the data points.
Although a first-order rate dependence on catalyst
concentration is in a good agreement with a well-defined
monometallic species participating in the catalytic reaction, it
should be noted that a nonlinear saturation behavior was
recently reported for the intramolecular aminoalkene hydro-
aminoalkylation catalyzed by IVb.6f
An attempt to determine the reaction order in alkene showed
a pronounced discrepancy between nonhindered (1-octene)
and hindered (vinylcyclohexane) alkenes (Figure 2).
Supporting Information) is indicative of amide exchange
(Scheme 6, step 4) becoming rate-determining at high alkene
concentrations.
In the presence of an excess of 1-octene (zero-order in
alkene), the reaction can be adequately described as being first-
3306
dx.doi.org/10.1021/ja211945m | J. Am. Chem. Soc. 2012, 134, 3300−3311

Journal of the American Chemical Society
Article
order in 2 at all alkene concentrations. However, reactions at
high amine concentrations are significantly slower, which
indicates a pronounced amine inhibition. This inhibition may
originate from reversible metallaaziridine formation (step 1) or
from other nonproductive events which may involve 1b-Nb and
2. As clearly demonstrated above, although step 1 is reversible,
the nondissociative character of this transformation implies that
neither the elimination step (Scheme 6, step 1) nor its reversal
should depend on the concentration of free 2, and therefore,
this inhibition scenario can be ruled out. Instead, it seems more
likely that the inhibition pathway involves reversible non-
productive binding of the free amine (either 2 or 4) to the
catalytically active five-coordinate bis(anilide) A to form a
“dormant” six-coordinate species AL (Scheme 11).
Figure 4. First-order kinetic plots for the reaction of 1-octene (1.89
M) with PhNHMe (2a) and its deuterated analogues (0.311 M) in the
presence of (R)-1b-Nb (0.019 M) in C6D6 at 150 °C. The solid lines
represent the least-squares fit to the data points.
Scheme 11. Equilibrium between Catalytic Active Five-
Coordinate Species A and Its “Dormant” Amine Adduct
12. Protonolysis of precatalyst 1-Nb with an excess of substrate
S liberates two molecules of dimethylamine and yields the
Scheme 12. Proposed Mechanism for the Intermolecular
Hydroaminoalkylation
The ability of the niobium center to adopt a six-coordinate
environment is justified by the formation of the amine-
coordinated aziridine (A*PhNHMe) elucidated by isotopic
labeling. Furthermore, many of the precatalysts 1a−d include
an additional coordinated amine molecule.8,12 The inhibition
operating via nonproductive binding of an additional substrate
equivalent is also rather common in early transition metal-
catalyzed hydroamination.16a,d,e,24,25
Kinetic Isotope Effects. Determination of the kinetic
isotope effect (KIE) has proven to be a very useful mechanistic
tool, which has been applied in many studies of C−H activation
by organometallic systems.26 We performed the reaction of 2a,
2a-ND, and 2a-CD3 with 1-octene in the presence of 1b-Nb;
however, in the absence of a complete mechanistic and kinetic
model, the observed kinetic isotope effects for nonelementary
transformations should be interpreted very carefully.
As shown in Figure 4, N-deuteration of 2a resulted in a
primary KIE of 1.6 0.1, whereas deuteration of the methyl
group in 2a-CD3 did not significantly diminish the rate of the
reaction. This result suggests that the C−H activation (step 1)
is unlikely to be rate-determining under these conditions,
whereas the rate-determining amide exchange (step 4) can
account for the primary KIE as the N−H bond is being formed
and broken during this step. These findings are in remarkable
contrast to the strong primary isotope effect observed in
intramolecular hydroaminoalkylation of a C-deuterated amino-
alkene by Doye.6f More detailed studies of the KIEs will be
described below as we will proceed by introducing a
mechanistic model that will account for all mechanistic and
kinetic observations.
bisamide A, which subsequently undergoes intramolecular
rearrangement to the six-coordinate metallaaziridine
A*ArNHMe. The metallaaziridine undergoes intermolecular
alkene insertion, followed by intramolecular protonolysis to
yield the bisamide B, which may regenerate the bisamide A and
release the product P via reversible amide exchange with an
additional substrate molecule S. Reversible nonproductive
binding of S and P to bisamides A and B yields six-coordinate
bisamides AS, AP, BS, and BP.
Based on the assumptions that
1. catalyst activation is fast compared to catalytic turnover;
2. K1 < 0.05 and the concentration of the metallaaziridine
A*ArNHMe is small at all times;
3. the transformation from A to B can be described by a
Mechanistic Model for Asymmetric Hydroaminoalky-
lation. The proposed mechanistic model is depicted in Scheme
combined rate constant k12 = K1k2;
3307
dx.doi.org/10.1021/ja211945m | J. Am. Chem. Soc. 2012, 134, 3300−3311

Journal of the American Chemical Society
Article
4. all inhibition equilibrium constants have similar values
(KAS ≈ KAP ≈ KBS ≈ KBP ≡ Kdorm);
different inhibition and amide exchange rate constants used in
the derivation of eq 3 were indeed justified.
5. the bimolecular amide exchange constants are very
similar in value (k4 ≈ k−4);
This kinetic model was also applied to the nonlinear
regression analysis of the kinetic data of the reactions of 1-
octene with other N-methyl-aniline derivatives (Table 3).
Deuteration of the N-methyl group in 2a-CD3 did not result
in a measurable KIE for k4 or an equilibrium isotope effect
(EIE) for Kdorm. Because k4 describes the amide exchange
reaction in which no C−H bond is formed or broken, the
absence of a KIE matches the mechanistic expectations. In
contrast, the N-deutero substrate 2a-ND displayed a primary
KIE of 1.8 0.3, which is in agreement with the value obtained
from observed constants (Figure 4). The accuracy of the
measurements is insufficient to decide whether an EIE for the
reversible nonproductive amine binding is present or not.
The dependence of kinetic parameters on the electronic
properties of the para-substituent is nonlinear for both k4 and
Kdorm. N-Methyl-aniline (2a) is less prone to inhibition and is
reacting slower than the more electron-accepting halide-
substituted 2b and 2c, whereas the electron-donating
methoxy-substituted 2d is most reactive (Figure 6).
It can be speculated that the observed nonlinearity may in
part originate from the complex character of the interaction of
2 with A or B (Scheme 12), whether as an amide or amine.
Indeed, the more nucleophilic and more Lewis-basic substrate
2d should form stronger covalent and dative bonds to the metal
than 2a. While 2c itself is more electron-deficient and should be
less reactive, the corresponding species A contains two
electron-deficient anilide ligands and is, therefore, more
electrophilic and prone to amine inhibition. The overall
reactivity is controlled by the interplay of the electronic
features of ligand and starting material. Thus, it is expected that
catalysts with a more electron-deficient ligand backbone should
possess enhanced reactivity.
6. catalyst loadings are low ([cat]0 ≪ [S]0);
one can derive eq 2 using a steady state approach (see
Supporting Information for details).
k k [cat] [S][alkene]
12 4
1
dorm
0
rate =
k [S] + k [alkene]
1 + K
[S]
0
4
0
12
(2)
Equation 2 can be simplified for two limiting cases. (1)
Alkene insertion is faster than amide exchange (k12[alkene] ≫
k4[S]0):
rate = k [cat] [S]
(3)
obs
0
k
4
with k
=
obs
dorm
1 + K
[S]
0
In this case, the reaction will be limited by amide exchange and
will display zero-order rate dependence on alkene concen-
tration as observed for the fast insertion of 1-octene and high
concentrations of vinylcyclohexane.
(2) If alkene insertion is rate-limiting (k12[alkene] ≪ k4[S]0),
one should observe an overall third-order reaction (eq 4).
rate = k [cat] [S][alkene]
(4)
obs
0
k
12
dorm
with k
=
obs
[S] 1 + K
[S]
0
0
(
)
Certain assumptions introduced in the derivation of eq 2,
such as low catalyst loadings, restrict its broader applicability. A
numerical nonlinear regression approach allows handling of
complicated kinetic models without the aforementioned
restrictions. Thus, analysis of kinetic data for the reaction of
2b with 1-octene employing models of different complexity
using the program DynaFit27 allows evaluation of our
assumptions on the relative turnover and equilibrium constant
values (see Supporting Information for details). The most
To get further insight into the kinetics of the process and
obtain the values of the insertion constant k12, we performed a
detailed kinetic study for the reaction of 2 with vinyl-
cyclohexane at various alkene concentrations. The kinetic
data was subjected to a nonlinear least-squares fit according to
the mechanism in Scheme 12. As demonstrated above,
substrate and product inhibition is comparable to that for the
reactions with 1-octene. Therefore, the values of Kdorm obtained
for 1-octene may be used in the nonlinear fit for vinyl-
cyclohexane (Table 4). Again, the results from the numerical
fitting were in good agreement with experimental data (Figure
7).
primitive model (assumption: k4 = k−4 and KAS = KAP = KBS
=
KBP = Kdorm) showed a good fit to the experimental data
(Figure 5). Thus, the assumptions on the values for the
A majority of amide exchange rate constants k4 obtained for
the reaction of 2a−d with vinylcylohexane (Table 4) are in
good agreement to those obtained for the reaction with 1-
octene (Table 3). The observed insertion constants k12 display
nonlinear dependence on the electronic properties of the
substrate. However, since k12 is a nonelementary rate constant
derived from the equilibrium constant K1 and the “true”
insertion constant k2, the absolute value of k12 should be
interpreted carefully. Notably, a similar nonlinear correlation
was observed for the zirconaaziridine formation from a
(methyl)anilidozirconium species via methane elimination.28
An important observation is the negligible KIE for 3d: k12
(CH3)/k12(CD3) = 1.19 0.14. This suggests that no strong
primary KIE or EIE is associated with the aziridine formation,
as olefin insertion should not display any primary KIE.
Alternatively, if the insertion proceeds in a concerted fashion
with intramolecular protonolysis, this reaction can exhibit a
Figure 5. Experimental reaction progress data (dots) for the reaction
of 1-octene (0.613 M) with 2b (0.144−0.579 M) in the presence of
(R)-1b-Nb (0.017 M) in C6D6 at 150 °C vs the simulated reaction
curves (lines) for k4 = 0.58 0.09 M−1 min−1, Kdorm = 2.8 0.7 M−1.
3308
dx.doi.org/10.1021/ja211945m | J. Am. Chem. Soc. 2012, 134, 3300−3311

Journal of the American Chemical Society
Article
a
Table 3. Kinetic Parameters for the Reaction of 2a-d with 1-Octene
entry
R
k4, M−1 min−1
Kdorm, M−1
1
2
3
4
5
6
H (2a)
0.33 0.02
0.38 0.05
0.18 0.03
0.58 0.09
0.58 0.15
3.1 1.2
1.6 0.3
1.9 0.5
1.1 0.6
2.8 0.7
7.8 2.4
14.4 5.0
H (2a-CD3)
H (2a-ND)
F (2b)
Cl (2c)
MeO (2d)
a
Nonlinear fit to a fast insertion model (k12 ≫ k4 = k−4 and KAS = KAP = KBS = KBP = Kdorm), confidence intervals for 95% probability.
Figure 7. Experimental reaction progress data (dots) for the reaction
Figure 6. Hammett plot for turnover constant k4 for the reaction of
2a−d with 1-octene. Error bars indicate the error of the measurement.
The dotted line is drawn as a guide for the eye.
of vinylcylohexane (0.30−3.57 M) with PMPNHCH3 (2d) (0.270 M)
in the presence of (R)-1b-Nb (0.017 M) in C6D6 at 150 °C versus the
simulated reaction curves (lines) for k4 = (3.7 0.2) M−1 min−1, k12
=
(0.81 0.05) M−1 min−1, Kdorm = 14.4 M−1. The value of Kdorm was
primary KIE, because the hydrogen (respectively, deuterium) in
the N−H bond (respectively, N−D bond) that is being broken
originates from the methyl group of the N-methylaniline. The
negligible net KIE can originate from the superposition of the
primary KIE for insertion-protonolysis and the inverse EIE for
the aziridine formation. At this point none of these scenarios
can be ruled out with certainty. It should be noted that the
absence of isotopic perturbation of enantioselectivity for the
reaction involving 1-octene and vinylcyclohexane is a strong
argument against a concerted insertion-protonolysis mecha-
nism. The stereodetermining step would involve cleavage of the
N−H bond and one would expect some variation of
obtained from the reaction with 1-octene.
stereoselectivity, which has been frequently observed in
hydroamination reactions.16a,d,h
Although norbornene displayed a remarkable isotopic
perturbation of enantioselectivity, it is prone to polymerization
under the standard reaction conditions and reliable kinetic data
could not be obtained. Overall, the absence of a primary KIE in
the isolated reaction step directly involving the formation of the
metallaaziridine is in remarkable contrast to literature
precedence of metallaaziridine formation in titanium-catalyzed
hydroaminoalkylations,6f as well as in stoichiometric metal-
laaziridine formation via elimination of an alkane.28,29
a
Table 4. Kinetic Parameters for the Reaction of 2a−d with Vinylcyclohexane
b
entry
R
k4, M−1 min−1
k12, M−1 min−1
1
2
3
4
5
H (2a)
F (2b)
0.39 0.03
0.89 0.05
0.67 0.03
3.7 0.2
0.31 0.05
0.35 0.02
0.56 0.07
0.81 0.05
Cl (2c)
MeO (2d)
MeO (2d-CD3)
3.0 0.3
0.68 0.07
a
b
Nonlinear fit to a slow insertion model (k12 ≠ k4 = k−4 and KAS = KAP = KBS = KBP = Kdorm), confidence intervals for 95% probability. Value for
major enantiomer.
3309
dx.doi.org/10.1021/ja211945m | J. Am. Chem. Soc. 2012, 134, 3300−3311

Journal of the American Chemical Society
Article
(3) For other functionalizations of sp3-hybridized α-amino C−H
moieties see: (a) Campos, K. R. Chem. Soc. Rev. 2007, 36, 1069. (b) Li,
C.-J. Acc. Chem. Res. 2009, 42, 335. (c) DeBoef, B.; Pastine, S. J.;
Sames, D. J. Am. Chem. Soc. 2004, 126, 6556. (d) Yi, C. S.; Yun, S. Y.;
Guzei, I. A. Organometallics 2004, 23, 5392. (e) Wan, X.; Xing, D.;
Fang, Z.; Li, B.; Zhao, F.; Zhang, K.; Yang, L.; Shi, Z. J. Am. Chem. Soc.
2006, 128, 12046. (f) Pastine, S. J.; Gribkov, D. V.; Sames, D. J. Am.
Chem. Soc. 2006, 128, 14220. (g) Murahashi, S.-I.; Nakae, T.; Terai,
H.; Komiya, N. J. Am. Chem. Soc. 2008, 130, 11005.
(4) (a) Clerici, M. G.; Maspero, F. Synthesis 1980, 305. (b) Nugent,
W. A.; Ovenall, D. W.; Holmes, S. J. Organometallics 1983, 2, 161.
(5) For hydroaminoalkylations catalyzed by group 5 metal complexes
see: (a) Herzon, S. B.; Hartwig, J. F. J. Am. Chem. Soc. 2007, 129, 6690.
(b) Herzon, S. B.; Hartwig, J. F. J. Am. Chem. Soc. 2008, 130, 14940.
(c) Eisenberger, P.; Ayinla, R. O.; Lauzon, J. M. P.; Schafer, L. L.
Angew. Chem., Int. Ed. 2009, 48, 8361. (d) Zi, G.; Zhang, F.; Song, H.
Chem. Commun. 2010, 46, 6296. (e) Zhang, F.; Song, H.; Zi, G. Dalton
Trans. 2011, 40, 1547.
CONCLUSIONS
■
Group 5 metal binaphtholate complexes are highly active
catalysts in the intermolecular hydroaminoalkylation of
unactivated alkenes with secondary amines. The reaction
proceeds with high enantio- and regioselectivity (up to 98%
ee, branched products formed exclusively). The strong
preference to activate primary amine α-C−H bonds originate
from steric constrains hampering alkene insertion, as
demonstrated by isotopic labeling experiments. Mechanistic
studies suggest that the reaction proceeds via a reversible,
nondissociative formation of a six-coordinate metallaaziridine
species, which undergoes facile alkene insertion and subsequent
intramolecular protonolysis. Kinetic studies are supportive of a
monometallic catalytic pathway with either alkene insertion or
amide exchange serving as a rate-determining step. The amide
exchange step is marked by a primary KIE, whereas essentially
no KIE was observed for the aziridine formation/olefin
insertion/protonolysis sequence. The latter result is in
remarkable contrast to previous findings in metallaaziridine
chemistry and should originate from stereoelectronic effects of
the binaphtholate ligand. Simulation experiments show a good
fit of the proposed kinetic model to the original kinetic data.
Reversible nonproductive events include amine coordination to
the catalytically active species and C−H activation of the
aromatic ring for the tantalum catalysts. All turnover and
equilibrium constants displayed nonlinear dependence on the
electronic properties of the amine substrate. Evidence of an
electron-acceptor ligand favoring the amide exchange process
was obtained, which can be used as a guideline for future
catalyst developments. These studies along with work toward
the development of sterically more accessible catalysts for
hydroaminoalkylation of secondary C−H bonds will be further
pursued by our laboratory.
(6) For hydroaminoalkylations catalyzed by group 4 metal complexes
see: (a) Muller, C.; Saak, W.; Doye, S. Eur. J. Org. Chem. 2008, 2731.
̈
(b) Kubiak, R.; Prochnow, I.; Doye, S. Angew. Chem., Int. Ed. 2009, 48,
1153. (c) Prochnow, I.; Kubiak, R.; Frey, O. N.; Beckhaus, R.; Doye, S.
ChemCatChem 2009, 1, 162. (d) Bexrud, J. A.; Eisenberger, P.; Leitch,
D. C.; Payne, P. R.; Schafer, L. L. J. Am. Chem. Soc. 2009, 131, 2116.
(e) Kubiak, R.; Prochnow, I.; Doye, S. Angew. Chem., Int. Ed. 2010, 49,
2626. (f) Prochnow, I.; Zark, P.; Muller, T.; Doye, S. Angew. Chem.,
̈
Int. Ed. 2011, 50, 6401.
(7) For hydroaminoalkylations catalyzed by late transition metals see:
(a) Jun, C.-H.; Hwang, D.-C.; Na, S.-J. Chem. Commun. 1998, 1405.
(b) Chatani, N.; Asaumi, T.; Yorimitsu, S.; Ikeda, T.; Kakiuchi, F.;
Murai, S. J. Am. Chem. Soc. 2001, 123, 10935. (c) Pan, S.; Endo, K.;
Shibata, T. Org. Lett. 2011, 13, 4692.
(8) Reznichenko, A. L.; Emge, T. J.; Audorsch, S.; Klauber, E. G.;
̈
Hultzsch, K. C.; Schmidt, B. Organometallics 2011, 30, 921.
(9) Horrillo-Martínez, P.; Hultzsch, K. C.; Gil, A.; Branchadell, V.
Eur. J. Org. Chem. 2007, 3311.
(10) (a) Wood, M. C.; Leitch, D. C.; Yeung, C. S.; Kozak, J. A.;
Schafer, L. L. Angew. Chem., Int. Ed. 2007, 46, 354. (b) Gott, A. L.;
Clarke, A. J.; Clarkson, G. J.; Scott, P. Organometallics 2007, 26, 1729.
(c) Reznichenko, A. L.; Hultzsch, K. C. Organometallics 2010, 29, 24.
(d) Zi, G.; Zhang, F.; Xiang, L.; Chen, Y.; Fang, W.; Song, H. Dalton
Trans. 2010, 39, 4048. (e) Ayinla, R. O.; Gibson, T.; Schafer, L. L. J.
Organomet. Chem. 2011, 696, 50.
ASSOCIATED CONTENT
■
S
* Supporting Information
Experimental procedures and characterization data, NMR
spectra of all new ligands, complexes and products, HPLC
traces, details of kinetic experiments and simulations. This
material is available free of charge via the Internet at http://
pubs.acs.org.
(11) Cummings, S.; Tunge, J. A.; Norton, J. R. Top. Organomet.
Chem. 2005, 10, 1.
(12) Rothwell has synthesized and structurally characterized some of
the binaphtholate tantalum complexes discussed in this study;
however, no catalytic results have been reported, see: (a) Thorn, M.
G.; Moses, J. E.; Fanwick, P. E.; Rothwell, I. P. J. Chem. Soc., Dalton
Trans. 2000, 2659. (b) Ru Son, A. J.; Schweiger, S. W.; Thorn, M. G.;
Moses, J. E.; Fanwick, P. E.; Rothwell, I. P. Dalton Trans. 2003, 1620.
(13) Reznichenko, A. L.; Nguyen, H. N.; Hultzsch, K. C. Angew.
Chem., Int. Ed. 2010, 49, 8984.
AUTHOR INFORMATION
■
Corresponding Author
hultzsch@rci.rutgers.edu
Notes
The authors declare no competing financial interest.
(14) Hoffmann, S.; Nicoletti, M.; List, B. J. Am. Chem. Soc. 2006, 128,
13074.
ACKNOWLEDGMENTS
■
(15) (a) Watson, P. L.; Parshall, G. W. Acc. Chem. Res. 1985, 18, 51.
(b) Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan,
M. C.; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. J. Am. Chem.
Soc. 1987, 109, 203. (c) Fendrick, C. M.; Marks, T. J. J. Am. Chem. Soc.
1986, 108, 425. (d) Sadow, A. D.; Tilley, T. D. J. Am. Chem. Soc. 2003,
125, 7971.
This work was supported by the National Science Foundation
through a NSF CAREER Award (CHE 0956021).
REFERENCES
■
(1) (a) Roesky, P. W. Angew. Chem., Int. Ed. 2009, 48, 4892.
(b) Eisenberger, P.; Schafer, L. L. Pure Appl. Chem. 2010, 82, 1503.
(2) The hydroaminoalkylation, as discussed in this article, should be
distinguished from the hydroaminomethylation reaction that proceeds
via a tandem hydroformylation/reductive amination sequence, see:
́
(16) (a) Gagne, M. R.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc.
1992, 114, 275. (b) Allan, L. E. N.; Clarkson, G. J.; Fox, D. J.; Gott, A.
L.; Scott, P. J. Am. Chem. Soc. 2010, 132, 15308. (c) Dunne, J. F.;
Fulton, D. B.; Ellern, A.; Sadow, A. D. J. Am. Chem. Soc. 2010, 132,
17680. (d) Manna, K.; Xu, S.; Sadow, A. D. Angew. Chem., Int. Ed.
2011, 50, 1865. (e) Arrowsmith, M.; Crimmin, M. R.; Barrett, A. G.
(a) Eilbracht, P.; Barfacker, L.; Buss, C.; Hollmann, C.; Kitsos-
̈
Rzychon, B. E.; Kranemann, C. L.; Rische, T.; Roggenbuck, R.;
Schmidt, A. Chem. Rev. 1999, 99, 3329. (b) Beller, M.; Seayad, J.;
Tillack, A.; Jiao, H. Angew. Chem., Int. Ed. 2004, 43, 3368.
M.; Hill, M. S.; Kociok-Kohn, G.; Procopiou, P. A. Organometallics
̈
2011, 30, 1493. (f) Leitch, D. C.; Platel, R. H.; Schafer, L. L. J. Am.
3310
dx.doi.org/10.1021/ja211945m | J. Am. Chem. Soc. 2012, 134, 3300−3311

Journal of the American Chemical Society
Article
Chem. Soc. 2011, 133, 15453. (g) Hangaly, N. K.; Petrov, A. R.;
Rufanov, K. A.; Harms, K.; Elfferding, M.; Sundermeyer, J.
Organometallics 2011, 30, 4544. (h) Manna, K.; Kruse, M. L.;
Sadow, A. D. ACS Catal. 2011, 1, 1637. (i) Zhang, X.; Emge, T. J.;
Hultzsch, K. C. Angew. Chem., Int. Ed. 2012, 51, 394.
(17) (a) Hanley, P. S.; Markovic, D.; Hartwig, J. F. J. Am. Chem. Soc.
2010, 132, 6302. (b) Hanley, P. S.; Hartwig, J. F. J. Am. Chem. Soc.
2011, 133, 15661.
(18) (a) Neukom, J. D.; Perch, N. S.; Wolfe, J. P. J. Am. Chem. Soc.
2010, 132, 6276. (b) Neukom, J. D.; Perch, N. S.; Wolfe, J. P.
Organometallics 2011, 30, 1269. (c) White, P. B.; Stahl, S. S. J. Am.
Chem. Soc. 2011, 133, 18594.
(19) Zhang, J.; Yang, C.-G.; He, C. J. Am. Chem. Soc. 2006, 128, 1798.
(20) McKinney Brooner, R. E.; Widenhoefer, R. A. Chem.Eur. J.
2011, 17, 6170.
(21) Casalnuovo, A. L.; Calabrese, J. C.; Milstein, D. J. Am. Chem.
Soc. 1988, 110, 6738.
(22) (a) Gallazzi, M. C.; Porri, L.; Vitulli, G. J. Organomet. Chem.
́
1975, 97, 131. (b) Catellani, M.; Motti, E.; Della Ca, N. Acc. Chem.
Res. 2008, 41, 1512. (c) Martins, A.; Mariampillai, B.; Lautens, M. Top.
Curr. Chem. 2010, 292, 1.
(23) Tobisch, S. Chem.Eur. J. 2011, 17, 14974.
(24) Gribkov, D. V.; Hultzsch, K. C.; Hampel, F. J. Am. Chem. Soc.
2006, 128, 3748.
(25) Hong, S.; Kawaoka, A. M.; Marks, T. J. J. Am. Chem. Soc. 2003,
125, 15878.
(26) (a) Parkin, G. Acc. Chem. Res. 2009, 42, 315. (b) Gom
Gallego, M.; Sierra, M. A. Chem. Rev. 2011, 111, 4857.
(27) Kuzmic, P. Anal. Biochem. 1996, 237, 260.
́
ez-
̌
(28) Coles, N.; Harris, M. C. J.; Whitby, R. J.; Blagg, J.
Organometallics 1994, 13, 190.
(29) Mayer, J. M.; Curtis, C. J.; Bercaw, J. E. J. Am. Chem. Soc. 1983,
105, 2651.
3311
dx.doi.org/10.1021/ja211945m | J. Am. Chem. Soc. 2012, 134, 3300−3311