Ruthenium-Catalyzed Hydroamination of Unactivated Terminal Alkenes with Stoichiometric Amounts of Alkene and an Ammonia Surrogate by Sequential Oxidation and Reduction

Article abstract of DOI:10.1021/jacs.0c11043

Hydroamination of alkenes catalyzed by transition-metal complexes is an atom-economical method for the synthesis of amines, but reactions of unactivated alkenes remain inefficient. Additions of N-H bonds to such alkenes catalyzed by iridium, gold, and lanthanide catalysts are known, but they have required a large excess of the alkene. New mechanisms for such processes involving metals rarely used previously for hydroamination could enable these reactions to occur with greater efficiency. We report ruthenium-catalyzed intermolecular hydroaminations of a variety of unactivated terminal alkenes without the need for an excess of alkene and with 2-aminopyridine as an ammonia surrogate to give the Markovnikov addition product. Ruthenium complexes have rarely been used for hydroaminations and have not previously catalyzed such reactions with unactivated alkenes. Identification of the catalyst resting state, kinetic measurements, deuterium labeling studies, and DFT computations were conducted and, together, strongly suggest that this process occurs by a new mechanism for hydroamination occurring by oxidative amination in concert with reduction of the resulting imine.

Full text of DOI:10.1021/jacs.0c11043

pubs.acs.org/JACS
Article
Ruthenium-Catalyzed Hydroamination of Unactivated Terminal
Alkenes with Stoichiometric Amounts of Alkene and an Ammonia
Surrogate by Sequential Oxidation and Reduction
Senjie Ma, Christopher K. Hill, Casey L. Olen, and John F. Hartwig*
Cite This: https://dx.doi.org/10.1021/jacs.0c11043
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Hydroamination of alkenes catalyzed by transition-
metal complexes is an atom-economical method for the synthesis
of amines, but reactions of unactivated alkenes remain inefficient.
Additions of N−H bonds to such alkenes catalyzed by iridium,
gold, and lanthanide catalysts are known, but they have required a
large excess of the alkene. New mechanisms for such processes
involving metals rarely used previously for hydroamination could
enable these reactions to occur with greater efficiency. We report
ruthenium-catalyzed intermolecular hydroaminations of a variety of
unactivated terminal alkenes without the need for an excess of
alkene and with 2-aminopyridine as an ammonia surrogate to give
the Markovnikov addition product. Ruthenium complexes have
rarely been used for hydroaminations and have not previously catalyzed such reactions with unactivated alkenes. Identification of the
catalyst resting state, kinetic measurements, deuterium labeling studies, and DFT computations were conducted and, together,
strongly suggest that this process occurs by a new mechanism for hydroamination occurring by oxidative amination in concert with
reduction of the resulting imine.
the development of a method that involves the direct addition
of the N−H bond of an amine to unactivated alkenes is needed
and would address these drawbacks.
INTRODUCTION
■
Amines and their derivatives are important as both
pharmaceuticals and agrochemicals.1−3 Traditional methods
to synthesize amines include nucleophilic substitution of
organic halides,4 reductive amination of carbonyl compounds,5
and reduction of amides, nitriles, and azides.6 The hydro-
amination of alkenes catalyzed by transition-metal complexes is
an attractive alternative to these methods because it occurs
directly with alkenes and could be applied to the functionaliza-
tion of both simple alkenes and complex molecules containing
alkene units.7−9 Despite the potential utility of hydro-
amination, examples of intermolecular hydroaminations are
often limited to conjugated and strained alkenes, such as
dienes,10,11 vinylarenes,12,13 norbornenes,14,15 and cyclopro-
penes.16,17 Hydroaminations of unactivated alkenes are rare
and generally require a large excess of alkene (Scheme
1a).18−26
Two main strategies have been followed to enable
hydroamination to occur without an excess amount of alkene.
The first involves catalytic reactions of alkenes possessing a
directing group (Scheme 1b);27,28 the second involves a formal
catalytic hydroamination achieved by combining a silane
reducing agent and a nitrogen-based electrophile (Scheme
1c).29,30 These strategies require special alkenes or generate
stoichiometric amounts of waste from the silane and aminating
reagents and require synthesis of the aminating reagent. Thus,
Two general classes of mechanisms have been followed by
most late, transition-metal catalysts for the hydroamination of
alkenes. The first mechanism involves nucleophilic attack of a
nitrogen nucleophile on a coordinated alkene, followed by
protonation of the resulting amino-alkyl intermediate. The
second involves oxidative addition of an N−H bond, followed
by migratory insertion of the alkene into the metal−nitrogen
bond and reductive elimination to form the C−H bond.
Cationic rhodium31 and gold22 systems react by the first
pathway, whereas neutral iridium complexes19 react by the
second pathway. Ruthenium complexes, the subject of this
work, catalyze the hydroamination of terminal alkynes32 and
vinylarenes,13 but they have not been shown to catalyze the
hydroamination of unconjugated alkenes. We considered that
such complexes and a carefully designed amine could catalyze
the hydroaminations of alkenes and do so by mechanisms that
Received: October 20, 2020
https://dx.doi.org/10.1021/jacs.0c11043
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
© XXXX American Chemical Society
A

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Scheme 1. Catalytic Hydroamination of Unactivated
Terminal Alkenes
Table 1. Evaluation of the Conditions for the
Hydroamination of 1-Dodecene with 1a
a
b
entry
conditions
standard
yield (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
67%
43%
18%
<1%
<1%
<1%
62%
<1%
<1%
29%
48%
<1%
56%
61%
45%
<1%
<1%
Ru-2 as catalyst
Ru-3 as catalyst
Ru-4 as catalyst
Ru-5 as catalyst
Ru-6 as catalyst
Ru-7 as catalyst
Ru-8 as catalyst
1,2-DCE as solvent
toluene as solvent
PhCl as solvent
CH3CN as solvent
THF as solvent
dioxane as solvent
Ru-1 (2 mol %)
no Ru-1
are distinct from those followed by complexes of other
transition metals.
Multidentate coordination of an amine that is tethered to a
Lewis basic group, rather than an alkene tethered to such a
group, can accelerate hydroaminations due to a series of
effects. First, the Lewis basic group can stabilize the
intermediates resulting from oxidative addition of the N−H
bond and products from insertion. Second, such an amine can
serve as an ammonia equivalent by removal of the Lewis basic
group. Third, an appropriate heteroaryl substituent can alter
the thermodynamics of this nearly thermoneutral process33 to
favor addition by rendering the N−H bond of the ammonia
surrogate weaker than the N−H bond of ammonia.
We report ruthenium-catalyzed Markovnikov hydroamina-
tions of unactivated, terminal alkenes with 2-aminopyridine as
an ammonia surrogate (Scheme 1d). Terminal alkenes
containing a series of functional groups undergo hydro-
amination in the presence of this catalyst to afford the
corresponding amine products without excess alkene. Detailed
experimental and computational mechanistic studies provide
strong evidence that this reaction occurs by a new pathway for
hydroamination that comprises oxidative amination of the
alkene and reduction of the corresponding imine intermediate.
HNTf2 (5 mol %)
a
Standard condition: 1-dodecene (0.2 mmol), Ru-1 (0.01 mmol), 1a
b
1
(0.2 mmol), 1,2-DCB (50 μL), 80 °C, 48 h. Determined by H
NMR spectroscopy of the crude reaction mixture with 1,3,5-
trimethoxybenzene as the internal standard.
to afford the amine product 2a in 67% yield (Table 1, entry 1).
Entries 2−6 show the effect of the phosphine ligand.
Ruthenium complexes that contain more sterically demanding
phosphine ligands (entry 2) and less electron-rich phosphine
ligands (entry 3) catalyzed the reactions to form the addition
product in lower yields. Complexes of ruthenium ligated by
trimethylphosphine, 1,2-bis(diethylphosphino)-butane, and a
tripodal phosphine ligand, N(CH2PEt2)3, did not catalyze the
hydroamination reaction (entries 4−6). Entries 7 and 8 show
the effect of the anionic ligand. Hydroamination conducted
with the PEt3-ligated ruthenium complex containing triflate
counteranion (Ru-7) afforded the product in a slightly lower
62% yield (entry 7) than that with Ru-1, and the PEt3-ligated
ruthenium complex containing chlorides (Ru-8) did not
catalyze the hydroamination reaction (entry 8). This hydro-
amination proceeded in a variety of polar, noncoordinating
solvents, and the reaction in 1,2-DCB (entry 1) occurred in a
higher yield than did reactions in other solvents of this class
(entries 9−14). The reaction performed with 2 mol % of Ru-1
instead of 5 mol % gave a lower yield of product (entry 15).
Hydroaminations of alkenes catalyzed by electrophilic metal
complexes, in some cases, have been suspected to occur by
acid generated from the amine and the complex.35 Although
the basicity of the 2-aminopyridine would render such a
pathway unlikely, we conducted control experiments to test
this potential. To assess the role of the ruthenium complex in
this reaction, we conducted the reaction in the absence of Ru-
1. Without Ru-1, no detectable product formed, indicating that
RESULTS AND DISCUSSION
■
Reaction Development. Our studies on ruthenium-
catalyzed hydroamination of unactivated terminal alkenes
began by examining the reaction between 1-dodecene (1.0
equiv) and 2-aminopyridine 1a (1.0 equiv) in the presence of a
variety of electrophilic ruthenium complexes, a class of catalyst
previously reported by our group to catalyze the oxidation of
alcohols34 (Table 1). 2-Aminopyridine was investigated as the
amine component because it could coordinate to the cationic
ruthenium complex in a bidentate fashion to form a four-
membered ruthenacycle that would retain reactivity at the Ru−
N bond, due to the ring strain.
A survey of a series of phosphine-ligated ruthenium
triflimide and triflate complexes showed that the triethylphos-
phine-ligated ruthenium complex Ru-1 containing triflimide as
the anionic ligand catalyzes the hydroamination of 1-dodecene
B
https://dx.doi.org/10.1021/jacs.0c11043
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
the ruthenium complex is necessary for this transformation
(entry 16). The hydroamination of 1-dodecene with
substoichiometric amounts of strong acid also did not form
detectable amounts of product (entry 17), providing evidence
against a pathway for hydroamination with 2-aminopyridine
catalyzed by a Brønsted acid.
In addition to unconjugated, terminal alkenes, vinylarenes
underwent this hydroamination reaction (Table 3). Vinyl-
Table 3. Scope of Vinylarenes for Catalytic
a b
,
Hydroamination
Scope of Terminal Alkenes and 2-Aminopyridines for
Catalytic Hydroamination. Studies on the scope of the
ruthenium-catalyzed hydroamination under the developed
conditions are summarized in Table 2. The hydroamination
Table 2. Scope of Unconjugated Alkenes and 2-
a b c d
, , ,
Aminopyridines for Catalytic Hydroamination
a
b
c
Isolated yields. 100 °C. 2 equiv of alkene.
arenes bearing a variety of electron-donating and electron-
withdrawing para-substituents underwent hydroamination to
afford products in good yields (18a−24a). However, higher
temperatures were required for the hydroaminations of
electron-deficient vinylarenes catalyzed by Ru-1 (20a, 22−
24a). A lower yield was observed from the reaction of a
vinylarene bearing an ortho-substituent (25a), indicating that
the reaction is sensitive to steric hindrance. The hydro-
amination of vinylnaphthalene and vinylferrocene also
proceeded to afford hydroamination products (27a−28a).
The major side reaction observed during the hydroamination
was the isomerization of the terminal alkene to form internal
alkenes, which did not undergo hydroamination under our
reaction conditions. High yields of the amine products 2a, 8a,
12a, 16a, and 28a were obtained for reactions with just 2 equiv
of alkene with respect to the aminopyridine.
Removal of Pyridyl Group from the Hydroamination
Product. The product from the hydroamination reaction was
converted to a primary amine by a two-step sequence (Scheme
2).36−38 Compound 2a was first converted to an amidine, and
a
b
c
d
e
Isolated yields. 100 °C. Ru-1 10 mol %. Ru-1 15 mol %. 2 equiv
f
Scheme 2. Removal of the Pyridyl Group from 2a
of alkene. 72 h.
of terminal alkenes with alkyl and aryl substituents occurred to
afford products in good yields (2a−4a). The hydroamination
of a series of alkenes bearing ether, fluoride, amide, and tertiary
alcohol functional groups also formed the hydroamination
products in good to moderate yields (5a−8a). Alkenes
containing multiple double bonds underwent hydroamination
exclusively at the terminal alkene (9a−11a). Even the
hydroamination of boldenone undecylenate, an anabolic
steroid, occurred, showing the potential of this methodology
for late-stage functionalization of architecturally complex
molecules containing potential competing functionality
(12a). An investigation of the scope of 2-aminopyridines for
this reaction showed that a variety of 2-aminopyridines
containing alkyl, ether, sulfide, and methylpiperazine sub-
stituents reacted to give the hydroamination products (13a−
17a).
the amidine was cleaved by NaBH4 to form primary amine 2b.
For convenience on the laboratory scale, the Boc-protected
amine 2c was isolated instead of the free amine 2b.
Mechanistic Studies of Alkene Hydroamination.
Because side reactions, such as alkene isomerization, alkene
oligomerization, and oxidative amination, compete with alkene
hydroamination,9,19,39 literature protocols for the hydro-
amination of unactivated alkenes require either the use of
alkenes containing a directing group or excess alkene. Our
hydroamination protocol represents a rare example of catalytic
hydroamination of unactivated alkenes with a stoichiometric
amount of alkene. Therefore, a detailed investigation of the
C
https://dx.doi.org/10.1021/jacs.0c11043
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
mechanism of this ruthenium-catalyzed hydroamination
reaction was conducted to provide insight into the features
that promote the hydroamination reaction over other side
reactions.
An independent synthesis of the resting state was pursued
(Figure 2) because we were unable to isolate it in pure form
Determination of the Catalyst Resting State. Our
mechanistic investigation began with identifying the catalyst
resting state. To obtain preliminary information about this
species, we monitored the reaction of 2-amino-4-methylpyr-
idine (1b) and 1-dodecene catalyzed by Ru-1 by 31P NMR
spectroscopy at 80 °C (spectrum b of Figure 1). Amino-
Figure 2. (a) Synthesis of complex 29. (b) Synthesis of complex 30.
(c) Solid-state structure of complex 29 with ellipsoids set at 30% and
selected hydrogen atoms and free triflimide anion omitted for clarity.
(d) Solid-state structure of complex 30 with ellipsoids set at 30% and
selected hydrogen atoms and free triflimide anion omitted for clarity.
directly from the mixture of Ru-1 and excess 1b. Treatment of
Ru-1 with 1 equiv of 1b afforded the aminopyridine-
coordinated ruthenium complex 29 in 82% yield after
recrystallization (Figure 2a). The 31P NMR spectrum of
complex 29 contained a single 31P resonance at 38 ppm, and
the structure of 29 was determined by single-crystal X-ray
diffraction. The Ru center in 29 is ligated by three PEt3 ligands,
one O-bound triflimide, and one κ2-bound aminopyridine in an
octahedral geometry (Figure 2c). The 31P NMR chemical shift
of 38 ppm for complex 29 is distinct from that of the catalyst
resting state (50 ppm).
Figure 1. (a) 31P NMR spectrum Ru-1. (b) 31P NMR spectrum of the
catalytic hydroamination of 1-dodecene with Ru-1 as the catalyst. (c)
31P NMR spectrum of the mixture of Ru-1 and 1-dodecene (20
equiv). (d) 31P NMR spectrum of the mixture of Ru-1 and 1b (20
equiv). (e) 31P NMR spectrum of complex 30. (f) 31P NMR spectrum
of the catalytic hydroamination of 1-dodecene with complex 30 as the
catalyst. (g) 31P NMR spectrum of the mixture of complexes 30 and
1b (20 equiv). All of the above spectra were acquired at 80 °C.
Thus, we postulated that coordination of 1b to the cationic
ruthenium center sufficiently acidifies the N−H bonds to cause
deprotonation of 29 by another equivalent of aminopyridine.
To test this hypothesis, we treated complex 29 with an
insoluble base (K3PO4). This reaction led to the isolation of
the Ru−amido complex 30 in 85% yield after recrystallization
(Figure 2b). The 31P NMR spectrum of complex 30 consisted
of a single sharp resonance at 51 ppm (e of Figure 1). The
solid-state structure of complex 30 was determined by X-ray
diffraction (Figure 2d). This complex adopts a square
pyramidal geometry with an empty coordination site trans to
a PEt3 ligand.
The hydroamination of 1-dodecene by aminopyridine 1b
catalyzed by 5 mol % of the isolated complex 30 afforded the
product 13a in 65% yield in 48 h. This result indicates that
complex 30 is kinetically competent to be an intermediate in
this hydroamination reaction. Furthermore, a broad resonance
at 50 ppm, which matched that observed for the catalytic
reaction initiated with Ru-1, was observed in the 31P NMR
spectrum of the hydroamination reaction initiated with
complex 30 as catalyst (f of Figure 1).
pyridine 1b was used for these studies, instead of 1a, because
ruthenium complexes containing 1b were more crystalline than
those containing 1a (vide infra). A broad resonance at 50 ppm,
corresponding to the resting state of the catalyst, was observed
in the 31P NMR spectrum of the reaction mixture.
Further information on the identity of the resting state was
obtained by allowing Ru-1 to react with 1-dodecene and 1b
separately and monitoring the reaction mixtures by 31P NMR
spectroscopy. Treatment of Ru-1 with excess 1-dodecene led
to no new Ru complexes, as determined by 31P NMR
spectroscopy at 80 °C (a and c of Figure 1). However,
treatment of Ru-1 with excess aminopyridine 1b led to the
observation of a broad resonance at 50 ppm in the 31P NMR
spectrum of the reaction mixture at 80 °C, and this resonance
was identical to that of the resting state of the Ru catalyst (d of
Figure 1). Therefore, the major ruthenium complex in the
reaction system forms from the combination of Ru-1 with one
or more equiv of aminopyridine 1b.
D
https://dx.doi.org/10.1021/jacs.0c11043
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
These data suggest that complex 30 could be the resting
state of the catalyst, but the resonance corresponding to
isolated 30 is much sharper and slightly more downfield (0.9
ppm) than that of the major ruthenium complex in the
catalytic system. We determined that this difference in line
shape and chemical shift depends on the presence or absence
of an excess of the aminopyridine; the 31P NMR signal of
isolated 30 broadened and shifted slightly upfield in the
presence of 20 equiv of aminopyridine 1b at 80 °C (g of Figure
1).
To investigate the origin of this difference in line shape and
chemical shift further, we conducted variable-temperature
NMR spectroscopy with a 20:1 mixture of aminopyridine 1b
and complex 30 (Figure 3). At −40 °C, the 31P NMR
chemical shift of this resonance lies further upfield in solutions
containing higher concentrations of aminopyridine, implying
that the resonance results from an equilibrium between 30 and
an adduct formed between 30 and the aminopyridine that is
distinct from complex 31a. At the same time, the similarity in
chemical shift between pure complex 30 and the resonance in
the presence of aminopyridine at the concentration and 80 °C
temperature of the reaction implies that the major component
in the catalytic reaction is complex 30.
Kinetic Studies on Catalytic Hydroamination. Kinetic
experiments were conducted to gain further information on the
mechanism of the hydroamination. Kinetic experiments with 1-
dodecene as the substrate were complicated by the observation
of alkene isomerization during the reaction. Therefore, we
conducted kinetic studies with vinylcyclohexane, which did not
undergo competing isomerization.
Initial rates of the hydroamination reaction were measured
at a series of concentrations of vinylcyclohexane, amino-
pyridine 1b, and catalyst Ru-1 (see the Supporting Information
for details). Plots of initial rates against the concentration of
the vinylcyclohexane, 1b, and Ru-1 are shown in Figure 4a−c.
We found that the hydroamination reaction is first order in the
concentration of vinylcyclohexane, zero order in the
concentration of 1b, and first order in the concentration of
Ru-1. The orders in alkene and catalyst imply that these
species react in the turnover-limiting step. The zero-order
dependence of the reaction on the concentration of amino-
pyridine 1b suggests that the hydroamination does not proceed
by turnover-limiting nucleophilic attack of 1b on the alkene in
a Ru complex like 32 (Scheme 3, pathway a). Instead, the
kinetic data are consistent with a migratory insertion of the
alkene into the Ru−N bond of the Ru−alkene complex 32 to
generate a Ru−alkyl complex 33 (Scheme 3, pathway b).
The effect of the concentration of PEt3 on the rate of the
catalytic hydroamination was also studied. Added PEt3
inhibited the reaction, and a plot of 1/initial rate against the
concentration of PEt3 was linear (Figure 4d). To account for
this observation, we considered two hypotheses. First, the
added PEt3 could inhibit the reaction by reversibly
coordinating to complex 30 to generate complex 34 (Scheme
4a). Second, reversible dissociation of PEt3 from complex 30
could generate the catalytically active complex 35 (Scheme
4b). To test these hypotheses, variable-temperature NMR
spectroscopy analysis was conducted with a 5:1 mixture of
PEt3 and complex 30. Full conversion of 30 to complex 34 was
observed by 31P NMR spectroscopy of this mixture at −20 °C
(see the Supporting Information for details). In addition, the
association of PEt3 to complex 30 was computed by DFT to be
exergonic by 0.3 kcal/mol at −20 °C, whereas the dissociation
of PEt3 from complex 30 was calculated to be endergonic by an
energy (31.5 kcal/mol at 80 °C) that exceeds the entire 30.5
kcal/mol barrier for the reaction determined from the
experimental rates. Therefore, we conclude that the origin of
the inhibition of the hydroamination by added PEt3 results
from the reversible coordination of PEt3 to complex 30 to form
the inactive tetraphosphine complex 34 (Scheme 4a).
Figure 3. 31P NMR spectra of the mixture of complexes 30 and 1b
(20 equiv) at different temperatures and possible structures for
complexes 31a and 31b.
spectrum of the mixture contained three triplet resonances. A
potential structure for the complex corresponding to the three
triplet resonances is 31a shown at the bottom of Figure 3. This
structure contains an additional 2-aminopyridine occupying
the sixth coordination site and a hydrogen bond to the amido
ligand. The hydrogen-bonding interaction in 31a is evidenced
by the observation of a broad downfield proton resonance (14
ppm) in the 1H NMR spectrum of this mixture. The
integration of the three triplet resonances in the 31P NMR
spectrum decreased at higher temperatures, and we were
unable to detect the triple resonances when the temperature of
the mixture reached 40 °C, indicating that complex 31a is a
minor species in the catalytic reaction at 80 °C.
Studies on the Pathway for the Formation of the
Product from Ru−Alkyl Complex 33. To elucidate the
pathway by which the product of this hydroamination reaction
was formed from the Ru−alkyl complex 33, the hydro-
amination of vinylcyclohexane was performed with 1b-d2 under
the standard catalytic conditions (Table 4). Deuterium
incorporation was observed at four different positions in the
The chemical shift of the broad resonance corresponding to
the major ruthenium complex at 80 °C migrated significantly
upfield at lower temperature. The origin of this line shape and
change in chemical shift is not clear, but could result from
hydrogen bonding between the amide ligand NH and the
aminopyridine without coordination of the pyridine nitrogen
to the metal as we suggest is present in complex 31b. The
E
https://dx.doi.org/10.1021/jacs.0c11043
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Scheme 3. Possible Pathways for the Formation of the C−N
Bond
Scheme 4. Two Hypotheses to Explain the Rate Inhibition
by Additional PEt3
a
Table 4. Hydroamination of Vinylcyclohexane with 1b-d2
a
a
reaction time (h)
%D at the Ha position
%D at the Hb position
24
36
48
29%
31%
29%
29%
32%
28%
a
%D incorporation = moles of D atoms/moles of product.
Scheme 5. Proposed Pathway for the Formation of
a
Hydroamination Product from 33
Figure 4. (a) Initial rates of product formation as a function of [1b].
(b) Initial rates of product formation as a function of [vinyl-
cyclohexane]. (c) Initial rates of product formation as a function of
[Ru-1]. (d) 1/initial rates of product formation as a function of
[PEt3].
product 17a-dn. The ratios of the percentage of deuterium
incorporation at the Ha and Hb positions were 1:1 at various
time points during the reaction (Table 4), and the
accumulation of deuterium into the starting vinylcyclohexane
was not observed.
a
The PEt3 ligands are omitted for clarity.
The results of this labeling study are consistent with the
pathway in Scheme 5. By this pathway, coordination and
migratory insertion of the alkene occur to form Ru−alkyl
complex 33-d from the catalyst resting state 30-d, after which
complex 33-d undergoes a subsequent β-hydride elimination to
generate the ruthenium−hydride species 44 ([Ru−H])
F
https://dx.doi.org/10.1021/jacs.0c11043
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(Figure 5a) and enamine 36-d. The enamine 36-d would then
undergo tautomerization catalyzed by protonated amino-
Scheme 6. Hydroamination of 1-Dodecene in the Presence
of Acetone
Figure 5. (a) Synthesis of the [Ru−H] intermediate 44. (b) Solid-
state structure of complex 44 with ellipsoids set at 30% and selected
hydrogen atoms and free triflimide anion omitted for clarity. (c)
Catalytic imine hydrogenation with complex 30 in the presence of H2.
To elucidate the identity of the proposed [Ru−H]
intermediate that would result from β-H elimination of the
aminoalkyl intermediate, we exposed complex 30 to 1 atm of
H2 (Figure 5a). Complex 43, which resulted from the overall
addition of H2 across the Ru−N bond of complex 30 and
coordination of H2, was observed. The structure of this
complex was deduced by the observation of two distinct
pyridine, which was formed during the generation of the
catalyst resting state 30-d. This tautomerization forms the β-
deuterated imine 37-d. Intramolecular H−D exchange between
the Ru−H position and bound aminopyridine in complex 44
would generate a ruthenium−deuteride. Subsequent reduction
of imine 37-d by complex 44 containing a Ru−D bond would
generate the final hydroamination product 17a-dn, containing
an enhanced level of deuterium at the Hα and Hβ positions.
Because the tautomerization step and H/D exchange of the
hydride position of complex 44 occur by reversible proton
transfers involving the amino group in 1b, the amounts of
deuterium in the Ha and Hb positions of the product are equal
to each other. Furthermore, the observation of the
incorporation of deuterium at the α position of the product
17a-dn provides evidence against a nucleophilic attack of 1b on
the metal-bound alkene complex, like 32 (Scheme 3, pathway
a), because the coordinative saturation of the complex formed
after a nucleophilic attack of the aminopyridine would disfavor
β-hydrogen elimination.
To investigate the pathway for the formation of the
hydroamination product further, we conducted the hydro-
amination of 1-dodecene with 1b in the presence of 1 equiv of
acetone. 2-Dodecanone 38 and the product from reductive
amination of acetone 39 were observed, along with the
hydroamination product 13a (Scheme 6). The formation of 38
and 39 implies that the reaction occurs through an imine
intermediate. Dodecyl enamine 40 is generated by migratory
insertion of the alkene into the amidopyridine complex 30 and
β-hydride elimination from metallacycle 33. A subsequent
tautomerization of the enamine 40 forms dodecyl imine 42.
Isopropyl imine 41 and water are generated by reversible
condensation of acetone with aminopyridine 1b. The imines
41 and 42 undergo competitive reduction to yield amines 13a
and 39, respectively. The portion of imine 42 that is not
reduced is hydrolyzed to give ketone 38, which is detected by
1H NMR spectroscopy.
1
ruthenium hydride signals in the H NMR spectrum. The H2
ligand in 43 was slowly replaced by N2 in a nitrogen
atmosphere to afford complex 44, which was characterized
by single-crystal X-ray diffraction (Figure 5a,b).
Under an atmosphere of H2, the imine generated in situ
from ketone 38 and aminopyridine 1b underwent hydro-
genation to form amine 13a catalyzed by the amido complex
30 in 95% yield (Figure 5c). This result indicates that the
hydridoruthenium complex 43 that would form in situ during
the catalytic process from complex 30 and H2 catalyzes the
reduction of imines, such as 41 that would be formed in the
catalytic process, to generate the final amine 13a.
DFT Computational Studies of the Proposed Cata-
lytic Cycle for the Ruthenium-Catalyzed Hydroamina-
tion of Terminal Alkenes. Our mechanistic experiments
imply that this Ru-catalyzed hydroamination of terminal
alkenes with 2-aminopyridine occurs by coordination of alkene
to the catalyst resting state 30, followed by migratory insertion
of the alkene into the Ru−N bond, β-hydride elimination to
generate an enamine, tautomerization of the enamine to an
imine, and reduction of the imine to afford to amine product.
Our kinetic studies indicate that the alkene and the ruthenium
catalyst are involved in the turnover-limiting step. Hydro-
genation of the imine cannot be turnover limiting because this
step occurs faster than the overall hydroamination of the
alkene (Figure 5c). To determine whether migratory insertion
or β-hydride elimination following a reversible insertion
process is turnover-limiting, DFT computations were con-
ducted on the oxidative amination portion of the mechanism.
This mechanism was calculated for the hydroamination of
propene, as a model alkene, with aminopyridine 1b. An energy
G
https://dx.doi.org/10.1021/jacs.0c11043
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
diagram for this process is shown in Figure 6. The
coordination of propene to the catalyst resting state 30 was
Figure 7. Proposed catalytic cycle for the ruthenium-catalyzed
hydroamination of terminal alkenes.
the Ru−alkyl intermediate 33, which undergoes β-hydride
elimination to generate the hydridoruthenium enamine
intermediate 45. Subsequent exchange of the enamine for
aminopyridine 1b in intermediate 45 forms the hydrido-
ruthenium aminopyridine intermediate 46 and free enamine
36. Eliminated enamine 36 undergoes tautomerization to
generate imine 37, which is computed to be more stable
thermodynamically than enamine 36 by 3.3 kcal/mol and
would be expected to react with amine-ligated ruthenium
hydride complexes by a metal−ligand bifunctional mecha-
nism.40 Thus, reduction of imine 37 by intermediate 46 affords
the final hydroamination product.
The proposed catalytic cycle is supported by a large body of
experimental data. The catalyst resting state 30 was identified
by 31P NMR spectroscopy and an independent synthesis. The
kinetic measurements support a mechanism by which
migratory insertion of the alkene into the Ru−N bond occurs,
rather than a potential nucleophilic attack on a coordinated
alkene by the amine. Deuterium labeling experiments indicate
that the Ru−alkyl complex 33 undergoes β-hydride elimi-
nation, rather than a potential direct protonation by 1b, and
hydroamination reactions conducted with added acetone imply
the formation of an imine intermediate. Independent studies of
the reactivity of the ruthenium−hydride complex arising from
β-hydride elimination with the imine resulting from tautome-
rization of the initially formed enamine show that the
reduction of the imine is much faster than the overall catalytic
cycle. Finally, DFT calculations indicate that the turnover-
limiting step of the reaction is migratory insertion of the alkene
into a ruthenium amido complex, rather than β-hydride
elimination from a reversibly formed insertion product.
Figure 6. DFT computational studies on the ruthenium-catalyzed
hydroamination of terminal alkenes. Free energies in kcal/mol at 80
°C are provided in parentheses.
calculated to be endergonic at 16.2 kcal/mol. The free-energy
barrier to the elementary migratory insertion of propene into
the Ru−N bond was computed to be 12.5 kcal/mol (32 to
TS1). These energies lead to an overall barrier for
coordination and insertion of 28.7 kcal/mol. The metallacycle
33 formed from insertion was computed to lie 8.1 kcal/mol
uphill of the starting complex and free propene, and the barrier
for β-hydride elimination from complex 33 was computed to
be 11.2 kcal/mol (33 to TS2). Thus, the transition state for
the β-hydride elimination lies 19.3 kcal/mol uphill of the
starting species and nearly 10 kcal/mol below the transition
state for migratory insertion. The oxidative amination portion
of the hydroamination reaction was calculated to be uphill by
5.9 kcal/mol, and the overall hydroamination was computed to
be exergonic by 3.8 kcal/mol. These calculations strongly
suggest that the combination of coordination of alkene and
migratory insertion of the alkene into the Ru−N bond is
irreversible and the turnover-limiting portion of this hydro-
amination reaction.
The results of our mechanistic investigation are summarized
in Figure 7. By this mechanism, the major component of the
catalyst resting state, complex 30, is formed by coordination of
the aminopyridine 1b to the ruthenium precatalyst Ru-1,
followed by deprotonation of the aminopyridine ligand by
another molecule of 1b. The first step of the catalytic cycle,
then, involves reversible coordination of the alkene to 30 to
form the alkene complex 32. Turnover-limiting migratory
insertion of the alkene into the Ru−N bond of 32 then affords
CONCLUSION
Ruthenium-catalyzed Markovnikov hydroamination of both
unactivated and activated terminal alkenes occurs with 2-
■
H
https://dx.doi.org/10.1021/jacs.0c11043
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Notes
aminopyridine as a surrogate for ammonia with a stoichio-
metric amount of alkene by an unusual pathway for
hydroamination. This process constitutes a rare example of
hydroamination of alkenes with ruthenium, and it is enabled by
a combination of a cationic metal center and a carefully
designed aminopyridine as an ammonia surrogate. This
combination facilitates the deprotonation of the aminopyridine
coordinated to an electron-deficient ruthenium center, the
migratory insertion of the alkene into the strained four-
member ruthenacycle, and the cooperative reduction of the
imine intermediate generated from β-hydrogen elimination to
lead to an overall redox-neutral addition process. This reaction
proceeds with a variety of terminal alkenes to afford the amine
products under conditions with the alkene in stoichiometric
quantities. A combination of experimental and computational
mechanistic studies reveals that this hydroamination reaction
occurs by turnover-limiting migratory insertion of the alkene
into the Ru−N bond, followed by β-hydride elimination to
generate an enamine, tautomerization of the enamine to an
imine, and reduction of the imine by the hydridoruthenium
aminopyridine complex to generate the amine product. This
pathway implies that an enantioselective process could be
developed if the step involving reduction of the imine
intermediate can be rendered enantioselective. Studies to
achieve such a process by this mechanism are ongoing.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the NIH (R35 GM130387), the
Molecular Graphics and Computation Facility at UC Berkeley
(NSF CHE-0840505) for DFT calculations, and the College of
Chemistry’s NMR facility at UC Berkeley (NIH
S10OD024998) for NMR experiments. We gratefully acknowl-
edge Dr. Hasan Celik for assistance with NMR experiments,
Dr. Nicholas Settineri for X-ray crystallography (NIH S10-
RR027172), Dr. David Small and Yehao Qiu for assistance
with DFT calculations, and Dr. Yumeng Xi for helpful
discussions.
REFERENCES
■
(1) Bagley, M. C.; Dale, J. W.; Merritt, E. A.; Xiong, X. Thiopeptide
Antibiotics. Chem. Rev. 2005, 105, 685−714.
(2) Nugent, T. C. Chiral Amine Synthesis: Methods, Developments and
Applications; Wiley-VCH: Weinheim, 2010.
(3) Wang, C.; Bai, X.; Wang, R.; Zheng, X.; Ma, X.; Chen, H.; Ai, Y.;
Bai, Y.; Liu, Y. Synthesis of Imatinib by C-N Coupling Reaction of
Primary Amide and Bromo-Substituted Pyrimidine Amine. Org.
Process Res. Dev. 2019, 23, 1918−1925.
(4) Shen, Q.; Hartwig, J. F. [(CyPF-tBu)PdCl2]: An Air-Stable, One-
Component, Highly Efficient Catalyst for Amination of Heteroaryl
and Aryl Halides. Org. Lett. 2008, 10, 4109−4112.
ASSOCIATED CONTENT
■
sı
* Supporting Information
(5) Afanasyev, O. I.; Kuchuk, E.; Usanov, D. L.; Chusov, D.
Reductive Amination in the Synthesis of Pharmaceuticals. Chem. Rev.
2019, 119, 11857−11911.
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.0c11043.
(6) Tripathi, R. P.; Verma, S. S.; Pandey, J.; Tiwari, V. K. Recent
Development on Catalytic Reductive Amination and Applications.
Curr. Org. Chem. 2008, 12, 1093−1115.
Experimental procedures, characterization of new
compounds, information about computational studies,
and crystallographic information (PDF)
(7) Mu
and Alkynes. Chem. Rev. 1998, 98, 675−704.
(8) Muller, T. E.; Hultzsch, K. C.; Yus, M.; Foubelo, F.; Tada, M.
̈
ller, T. E.; Beller, M. Metal-Initiated Amination of Alkenes
X-ray crystallographic data for Ru(PEt3)3(NTf2)2 (CIF)
̈
X-ray crystallographic data for Ru(PEt3)3(amine)-
(NTf2)2 (29) (CIF)
Hydroamination: Direct Addition of Amines to Alkenes and Alkynes.
Chem. Rev. 2008, 108, 3795−3892.
(9) Huang, L.; Arndt, M.; Gooßen, K.; Heydt, H.; Gooßen, L. J. Late
Transition Metal-Catalyzed Hydroamination and Hydroamidation.
Chem. Rev. 2015, 115, 2596−2697.
X-ray crystallographic data for Ru(PEt3)3(amido)(NTf2)
(30) (CIF)
X-ray crystallographic data for Ru(H)(PEt3)3(amine)-
(NTf2)2 (44) (CIF)
(10) Pawlas, J.; Nakao, Y.; Kawatsura, M.; Hartwig, J. F. A General
Nickel-Catalyzed Hydroamination of 1,3-Dienes by Alkylamines:
Catalyst Selection, Scope, and Mechanism. J. Am. Chem. Soc. 2002,
124, 3669−3679.
AUTHOR INFORMATION
(11) Yang, X.-H.; Lu, A.; Dong, V. M. Intermolecular Hydro-
amination of 1,3-Dienes to Generate Homoallylic Amines. J. Am.
Chem. Soc. 2017, 139, 14049−14052.
■
Corresponding Author
John F. Hartwig − Department of Chemistry, University of
California, Berkeley, California 94720, United States;
orcid.org/0000-0002-4157-468X; Email: jhartwig@
berkeley.edu
(12) Utsunomiya, M.; Hartwig, J. F. Intermolecular, Markovnikov
Hydroamination of Vinylarenes with Alkylamines. J. Am. Chem. Soc.
2003, 125, 14286−14287.
(13) Utsunomiya, M.; Hartwig, J. F. Ruthenium-Catalyzed Anti-
Markovnikov Hydroamination of Vinylarenes. J. Am. Chem. Soc. 2004,
126, 2702−2703.
Authors
Senjie Ma − Department of Chemistry, University of
California, Berkeley, California 94720, United States;
orcid.org/0000-0002-7673-1390
Christopher K. Hill − Department of Chemistry, University of
California, Berkeley, California 94720, United States;
orcid.org/0000-0002-4575-9403
Casey L. Olen − Department of Chemistry, University of
California, Berkeley, California 94720, United States;
orcid.org/0000-0002-8621-2973
(14) Zhou, J.; Hartwig, J. F. Intermolecular, Catalytic Asymmetric
Hydroamination of Bicyclic Alkenes and Dienes in High Yield and
Enantioselectivity. J. Am. Chem. Soc. 2008, 130, 12220−12221.
(15) Dorta, R.; Egli, P.; Zurcher, F.; Togni, A. The [IrCl-
̈
(Diphosphine)]2/Fluoride System. Developing Catalytic Asymmetric
Olefin Hydroamination. J. Am. Chem. Soc. 1997, 119, 10857−10858.
(16) Teng, H.-L.; Luo, Y.; Nishiura, M.; Hou, Z. Diastereodivergent
Asymmetric Carboamination/Annulation of Cyclopropenes with
Aminoalkenes by Chiral Lanthanum Catalysts. J. Am. Chem. Soc.
2017, 139, 16506−16509.
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.0c11043
(17) Teng, H.-L.; Luo, Y.; Wang, B.; Zhang, L.; Nishiura, M.; Hou,
Z. Synthesis of Chiral Aminocyclopropanes by Rare-Earth-Metal-
I
https://dx.doi.org/10.1021/jacs.0c11043
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Catalyzed Cyclopropene Hydroamination. Angew. Chem., Int. Ed.
2016, 55, 15406−15410.
(18) Sevov, C. S.; Zhou, J.; Hartwig, J. F. Iridium-Catalyzed
Intermolecular Hydroamination of Unactivated Aliphatic Alkenes
with Amides and Sulfonamides. J. Am. Chem. Soc. 2012, 134, 11960−
11963.
Group from α-Substituted N-(Pyridin-2-yl)piperidines Obtained via
Directed Ru-Catalyzed sp3 C-H Functionalization. J. Org. Chem.
2013, 78, 9803−9814.
(37) Ronson, T. O.; Renders, E.; Van Steijvoort, B. F.; Wang, X.;
́
Wybon, C. C. D.; Prokopcova, H.; Meerpoel, L.; Maes, B. U. W.
Ruthenium-Catalyzed Reductive Arylation of N-(2-Pyridinyl)amides
with Isopropanol and Arylboronate Esters. Angew. Chem., Int. Ed.
2019, 58, 482−487.
(19) Sevov, C. S.; Zhou, J.; Hartwig, J. F. Iridium-Catalyzed,
Intermolecular Hydroamination of Unactivated Alkenes with Indoles.
J. Am. Chem. Soc. 2014, 136, 3200−3207.
(38) Pan, S.; Endo, K.; Shibata, T. Ir(I)-Catalyzed Enantioselective
Secondary sp3 C-H Bond Activation of 2-(Alkylamino)pyridines with
Alkenes. Org. Lett. 2011, 13, 4692−4695.
̈
(20) Nguyen, H. N.; Lee, H.; Audorsch, S.; Reznichenko, A. L.;
Nawara-Hultzsch, A. J.; Schmidt, B.; Hultzsch, K. C. Asymmetric
Intra- and Intermolecular Hydroamination Catalyzed by 3,3′-Bis-
(trisarylsilyl)- and 3,3′-Bis(arylalkylsilyl)-Substituted Binaphtholate
Rare-Earth-Metal Complexes. Organometallics 2018, 37, 4358−4379.
(21) Miller, D. C.; Ganley, J. M.; Musacchio, A. J.; Sherwood, T. C.;
Ewing, W. R.; Knowles, R. R. Anti-Markovnikov Hydroamination of
Unactivated Alkenes with Primary Alkyl Amines. J. Am. Chem. Soc.
2019, 141, 16590−16594.
(39) Bruneau, C.; Dixneuf, P. H. Metal Vinylidenes and
Allenylidenes in Catalysis: Applications in Anti-Markovnikov
Additions to Terminal Alkynes and Alkene Metathesis. Angew.
Chem., Int. Ed. 2006, 45, 2176−2203.
(40) Cobley, C. J.; Henschke, J. P. Enantioselective Hydrogenation
of Imines Using a Diverse Library of Ruthenium Dichloride-
(diphosphine)(diamine) Precatalysts. Adv. Synth. Catal. 2003, 345,
195−201.
(22) Zhang, Z.; Lee, S. D.; Widenhoefer, R. A. Intermolecular
Hydroamination of Ethylene and 1-Alkenes with Cyclic Ureas
Catalyzed by Achiral and Chiral Gold(I) Complexes. J. Am. Chem.
Soc. 2009, 131, 5372−5373.
(23) Musacchio, A. J.; Lainhart, B. C.; Zhang, X.; Naguib, S. G.;
Sherwood, T. C.; Knowles, R. R. Catalytic intermolecular hydro-
aminations of unactivated olefins with secondary alkyl amines. Science
2017, 355, 727.
(24) Xi, Y.; Ma, S.; Hartwig, J. F. Catalytic asymmetric addition of an
amine N-H bond across internal alkenes. Nature 2020, 588, 254.
(25) Zhang, J.; Yang, C.; He, C. Gold(I)-Catalyzed Intra- and
Intermolecular Hydroamination of Unactivated Olefins. J. Am. Chem.
Soc. 2006, 128, 1798−1799.
(26) Reznichenko, A. L.; Nguyen, H. N.; Hultzsch, K. C.
Asymmetric Intermolecular Hydroamination of Unactivated Alkenes
with Simple Amines. Angew. Chem., Int. Ed. 2010, 49, 8984−8987.
(27) Gurak, J. A.; Yang, K. S.; Liu, Z.; Engle, K. M. Directed,
Regiocontrolled Hydroamination of Unactivated Alkenes via Proto-
depalladation. J. Am. Chem. Soc. 2016, 138, 5805−5808.
(28) Vanable, E. P.; Kennemur, J. L.; Joyce, L. A.; Ruck, R. T.;
Schultz, D. M.; Hull, K. L. Rhodium-Catalyzed Asymmetric
Hydroamination of Allyl Amines. J. Am. Chem. Soc. 2019, 141,
739−742.
(29) Zhu, S.; Buchwald, S. L. Enantioselective CuH-Catalyzed Anti-
Markovnikov Hydroamination of 1,1-Disubstituted Alkenes. J. Am.
Chem. Soc. 2014, 136, 15913−15916.
(30) Guo, S.; Yang, J. C.; Buchwald, S. L. A Practical Electrophilic
Nitrogen Source for the Synthesis of Chiral Primary Amines by
Copper-Catalyzed Hydroamination. J. Am. Chem. Soc. 2018, 140,
15976−15984.
(31) Liu, Z.; Yamamichi, H.; Madrahimov, S. T.; Hartwig, J. F.
Rhodium Phosphine-π-Arene Intermediates in the Hydroamination of
Alkenes. J. Am. Chem. Soc. 2011, 133, 2772−2782.
(32) Cheung, H. W.; So, C. M.; Pun, K. H.; Zhou, Z.; Lau, C. P.
Hydro(trispyrazolyl)borato-Ruthenium(II) Diphosphinoamino Com-
plex-Catalyzed Addition of β-Diketones to 1-Alkynes and Anti-
Markovnikov Addition of Secondary Amines to Aromatic 1-Alkynes.
Adv. Synth. Catal. 2011, 353, 411−425.
(33) Johns, A. M.; Sakai, N.; Ridder, A.; Hartwig, J. F. Direct
Measurement of the Thermodynamics of Vinylarene Hydroamination.
J. Am. Chem. Soc. 2006, 128, 9306−9307.
(34) Hill, C. K.; Hartwig, J. F. Site-selective oxidation, amination and
epimerization reactions of complex polyols enabled by transfer
hydrogenation. Nat. Chem. 2017, 9, 1213−1221.
(35) McBee, J. L.; Bell, A. T.; Tilley, T. D. Mechanistic Studies of
the Hydroamination of Norbornene with Electrophilic Platinum
Complexes: The Role of Proton Transfer. J. Am. Chem. Soc. 2008,
130, 16562−16571.
(36) Smout, V.; Peschiulli, A.; Verbeeck, S.; Mitchell, E. A.;
Herrebout, W.; Bultinck, P.; Vande Velde, C. M. L.; Berthelot, D.;
Meerpoel, L.; Maes, B. U. W. Removal of the Pyridine Directing
J
https://dx.doi.org/10.1021/jacs.0c11043
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX