A Mild Hydroaminoalkylation of Conjugated Dienes Using a Unified Cobalt and Photoredox Catalytic System

Article abstract of DOI:10.1021/jacs.7b09252

Metallo-photoredox catalysis has redefined the available bond disconnections in the synthetic arsenal. By harnessing the one-electron chemistry of photoredox catalysis in tandem with low-valent cobalt catalysts, new methods by which functionalities may be stitched together become available. Herein we describe the coupling of photoredox-generated α-amino radical species with conjugated dienes using a unified cobalt and iridium catalytic system in order to access a variety of useful homoallylic amines from simple commercially available starting materials. We present a series of mechanistic experiments that support the intervention of Co-hydride intermediates that undergo diene insertion to generate Co-π-allyl species.

Full text of DOI:10.1021/jacs.7b09252

Article
pubs.acs.org/JACS
Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX-XXX
A Mild Hydroaminoalkylation of Conjugated Dienes Using a Unified
Cobalt and Photoredox Catalytic System
Scott M. Thullen†,‡ and Tomislav Rovis*,†,‡
†Department of Chemistry, Columbia University, New York, New York 10027, United States
‡Department of Chemistry, Colorado State University, Fort Collins, Colorado 80523, United States
S
* Supporting Information
ABSTRACT: Metallo-photoredox catalysis has redefined the available bond
disconnections in the synthetic arsenal. By harnessing the one-electron chemistry
of photoredox catalysis in tandem with low-valent cobalt catalysts, new methods
by which functionalities may be stitched together become available. Herein we
describe the coupling of photoredox-generated α-amino radical species with
conjugated dienes using a unified cobalt and iridium catalytic system in order to
access a variety of useful homoallylic amines from simple commercially available
starting materials. We present a series of mechanistic experiments that support
the intervention of Co−hydride intermediates that undergo diene insertion to generate Co−π-allyl species.
INTRODUCTION
■
Hydroaminoalkylation of double bonds is a powerful method to
access linear amine functionalities from simple starting
materials. This is traditionally achieved through hydro-
formylation/reductive amination of an alkene by a noble
metal catalyst.1 More recently, hydroaminoalkylation can be
accomplished by the direct activation of an alkylamine, most
often N-methyl, and its addition across a π unsaturation
catalyzed by early transition metal complexes (Figure 1, eq
1).2,3 Despite the power of this approach, these conditions
struggle with low functional group tolerance, forcing reaction
conditions, and substrate scope. The hydroaminoalkylation of
dienes, in particular, is problematic using established
approaches (Figure 1, eq 2).4
Our group has recently explored the coupling of photoredox
catalysis5,6 with cobalt catalysis in order to access low-valent
cobalt(I) or cobalt(0) intermediates for the construction of
arenes through [2 + 2 + 2] cycloadditions.7 These low-valent
cobalt species are traditionally accessed in situ using strong
reducing conditions such as heterogeneous metals or Grignard
reagents because of their synthetic challenge and poor bench
stability.8 The problem of limited functional group tolerance
with an in situ Grignard/reducing metal reduction is overcome
by using a tertiary amine as a reductive quench in a visible-light-
driven photoredox cycle. The α-amino radicals thus generated
are known to undergo addition to very electron-deficient olefins
such as alkylidene malonates (Figure 1, eq 3). It is noteworthy
that the use of simple acrylates under these conditions leads to
lower yields.9
We became particularly drawn to cobalt’s predilection for
unactivated dienes10 as a method to render these species
reactive toward photoredox-generated α-amino alkyl radicals.
The realization of this coupling could lead to an array of useful
building blocks, including homoallylic amines.
Figure 1. Catalyzed hydroaminoalkylation.
RESULTS AND DISCUSSION
We began our study by screening a variety of cobalt precatalysts
in the presence of dimethylaniline and isoprene (Table 1). The
■
Received: September 2, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.7b09252
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Additionally, the quantum yield of the reaction was measured
and found to be 4%, which also suggests that a radical-chain
propagation mechanism is unlikely. The initial reaction rate is
dependent on the photocatalyst employed, with a more
strongly reducing catalyst such as [Ir(ppy)2(dtbpy)]PF6
showing a higher initial rate of reaction.12 However, the
fluorinated catalyst PC was used in the optimized conditions
because of higher yields from increased longevity. The organic
photocatalyst 4Cz-IPN developed by Zhang13 also proved
competent in the reaction, albeit with lower yields.
Table 1. Reaction Optimization
With these optimized conditions in hand, we explored the
scope of this reactivity using ethyl sorbate as the diene substrate
(Figure 2). The scope of tertiary amines is quite broad,
tolerating a wide variety of functionalized anilines, including
aryl bromides and terminal alkynes (3d and 3g, respectively).
Secondary positions can also be functionalized, although with
limited control over the diastereoselectivity of the two newly
formed stereocenters (3f, 3h, and 3j, with the exception of 3h
formed in 8:1 dr). Fully aliphatic amines also participate in the
reaction, favoring primary positions for alkylation over
secondary or tertiary positions.14 This allows for the
incorporation of heterocycles such as piperidine (3l) and
morpholine (3m) as well as linear aliphatic amines (e.g., 3n). In
all cases, bis- and trisaddition of the diene contributes to the
loss in yield. However, this is largely mitigated through use of a
moderate excess of amine substrate (1.25 to 2 equiv).
The scope of the diene coupling partner was also explored
(Figure 3). Simple dienes, such as butadiene and isoprene,
participate in the reaction with lower fidelity for the (E)-olefin
and lower regioselectivity (delivering products 4a and 4b).
Ketone 4d can be accessed directly through the use of a silyl
enol ether coupling partner, which is deprotected upon workup.
Electron-poor dienes participate well in the reaction, whereas
many electron-rich dienes are typically low-yielding.15 Cyclic
dienes in particular are poor substrates, often giving less than
a
Optimizations were performed on a 0.1 mmol scale using 1a (2
equiv), 2a (1 equiv), Co source (10 mol %), ligand (10 mol %), base
(0.4 equiv), and PC (0.5 mol %) over a period of 12 h. The reaction
was run in the absence of a photocatalyst.
b
extensively used commercial [Ir(dF-CF3ppy)2dtbbpy]PF6 (PC)
was primarily used since it was deemed suitable for the
reduction of cobalt in our prior studies.7 We observed a trace
mass corresponding to our intended product in the presence of
Co(OAc)2. In the presence of a diphosphine ligand (dppp), the
yield increased to 52%. Further optimization of the reaction
conditions revealed that CsOPiv is the optimal base for this
reactivity. Control studies indicated that cobalt, light, and the
photocatalyst are all needed for reactivity, although the reaction
proceeds using several other common bidentate ligands. It
should be noted that with electron-deficient dienes, such as
ethyl sorbate, there is a background reaction in the absence of
cobalt that proceeds in ∼10% yield. For electron-neutral and
-rich dienes, there is trace or no background reaction (<1%).
The reaction shows good fidelity with respect to alternating
periods of irradiation and darkness, indicating that this reaction
is not subject to a radical-chain propagation mechanism.11
Figure 2. Amine scope. *Yield in the absence of CoBr2.
B
DOI: 10.1021/jacs.7b09252
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 3. Diene scope.
10% yield. More complicated linear dienes such as
pseudoionone, myrcene, and piperine also proved to be
competent in the reaction, delivering products 4k−m.
Importantly, these reaction conditions are tolerant of myriad
functionalities that are known to react in competing pathways,
including reducible aryl halides (en route to 3d), terminal aryl
alkynes (3g), basic amine functionality (pentamethylpiperidine,
N-methylpiperidine, N-methylmorpholine), and an enol silane
and allyl acetate (starting materials en route to 4d and 4j,
respectively). The formation of a quaternary stereocenter is also
noteworthy (4k).
Table 2. Carboxylate Base Effect
In terms of a likely mechanism, we considered that the
aminoalkyl radical may simply add to an in situ-formed low-
valent Co−diene complex. This chemistry seems to progress
solely in the presence of carboxylate bases and shows only trace
product in the presence of other common bases (entries 1 and
2 vs 3−5 in Table 2) . However, when the precatalyst salt was
changed from CoBr2 and dppp to Co(OPiv)2·dppp, some of
the reactivity with these other bases was restored (entries 8−10
in Table 2).
a
See footnote a of Table 1.
cyclability of the catalyst solely in the presence of carboxylate
bases.
The CV experiments described above implicate an in situ-
generated Co−H that acts on the diene to form a Co−π-allyl
intermediate. We considered that it should prove informative to
conduct a control experiment in which the low-valent Co was
exposed to carboxylic acid in the absence of amine. In the
event, we subjected ethyl sorbate to the reaction conditions in
the presence of acetic acid but in absence of an amine (Scheme
1c).18 The results implicate the formation of allyl species, as
dimerization of the diene occurs only in a tail-to-tail fashion
(product B) under these conditions. We often observe this
dimerization under the standard reaction conditions, especially
with more sluggish amines. While the majority of these
dimerization events appear to be photoredox-catalyzed,19 these
experiments were conducted with limiting acid, and the
observed product distribution is indicative of complete
incorporation of the available protons (Scheme 1c). Under
deuteration conditions, the isotope is found at both carbons α
to the carbonyls, which we take as strong evidence for the
We further noted that a cyclic voltammetry (CV) study16
showed an absence of a reoxidation peak of the Co catalyst in
the presence of both butadiene and pivalic acid (Scheme 1a-iv).
This potentially indicates that the Co(I) species generated by
the electrode is fully consumed in the presence of both a
carboxylic acid and a diene, presumably in an oxidative event to
generate an allyl−Co(III) intermediate (Scheme 1b). Thus, the
latter is incapable of undergoing a subsequent oxidation at these
potentials, as evidenced by the absence of a reoxidation event in
the reverse wave. Without both external components, the
reoxidation to Co(II) is still observed (Scheme 1a-ii,iii).
Additionally, the premade Co(dppp)(OPiv)2 catalyst showed
a well-behaved pseudoreversible curve within the range of the
potential of PC.17 This pseudoreversibility was not observed in
the presence of other base/ligand types such as carbonate
bases. This ligand effect of the pivalate could explain the
C
DOI: 10.1021/jacs.7b09252
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Scheme 1. Control Experiments
Scheme 2. Proposed Mechanism
(generated through deprotonation of the amine radical cation)
to generate a transient Co(III) hydride species, II.20 An
equivalent of diene can then undergo migratory insertion into
the Co−H bond to form a Co(III)−allyl species. This is
supported in part by previous low-valent Co studies in the
presence of dienes, indicating that Co(III)−allyls can be
generated with dienes and Co(III)−H.21 This Co(III)−allyl,
III, can then be reduced by the photocatalyst to generate a
Co(II)−allyl, IV. Meanwhile, an α-amino alkyl radical is
generated through deprotonation of the oxidized tertiary
amine. This α-amino alkyl radical may then attack the Co
center, V, which undergoes reductive elimination to afford the
product and Co(I) back into the catalytic cycle. Alternatively,
this radical may attack the allyl ligand directly, VI, affording the
product and a Co(I) species.
CONCLUSIONS
■
We have explored the hydroaminoalkylation of conjugated
dienes using a photoredox and Co catalytic system. This
reactivity opens new potential pathways for the coupling of π-
rich functionalities with photoredox-generated radical inter-
mediates. Ongoing studies are focused on exploring the
mechanism in more detail to shed light on other possible
transformations.
intervention of Co−allyl species. A mechanistic scenario
consistent with these observations is illustrated in Scheme 1d.
With respect to tail-to-tail dimerization of diene coupling under
these conditions, Kambe has observed that butadiene
dimerization under Co catalysis occurs by insertion of the
diene at the more substituted end of the Co−π-allyl (Scheme
1e).18a
A potential mechanism for this transformation that accounts
for all of these observations could be described by the cycle
presented in Scheme 2. A Co(II) salt is reduced to a Co(I)
species in situ by the photocatalyst. CV studies indicate that
this reduction should be easily accessible by PC17 given the
reduction potential of the precatalyst and the presumably
formed Co(dppp)(OPiv)2 (−0.66 V and −1.25 V for [CoII/
CoI] vs Ag/AgCl, respectively).20 This Co(I) species
potentially reacts with an equivalent of carboxylic acid
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.7b09252.
■
S
Experimental procedures and compound characterization
(PDF)
AUTHOR INFORMATION
Corresponding Author
*tr2504@columbia.edu
■
D
DOI: 10.1021/jacs.7b09252
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
photoredox catalysis, see: Cismesia, M. A.; Yoon, T. P. Chem. Sci.
2015, 6, 5426−5434.
ORCID
Tomislav Rovis: 0000-0001-6287-8669
(12) See the Supporting Information.
Notes
(13) Luo, J.; Zhang, J. ACS Catal. 2016, 6, 873−877.
(14) The positional selectivity for the methyl has previously been
observed for N-methylmorpholine. See: Douglas, J. J.; Cole, K. P.;
Stephenson, C. R. J. J. Org. Chem. 2014, 79, 11631−11643.
(15) For an extended scope, see the Supporting Information.
(16) Multiple cyclic voltammetry studies were performed on various
Co(II) salts based on the work outlined by Buriez et al.: Buriez, O.;
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was conducted in connection with the Catalysis
Collaboratory for Light-activated Earth Abundant Reagents (C-
CLEAR), which is supported by the National Science
Foundation and the Environmental Protection Agency through
the Network for Sustainable Molecular Design and Synthesis
Program (NSFCHE-1339674). We thank the Shores group
(CSU) for access to their chemical actinometer.
́ ́
Labbe, E.; Perichon, J. J. Electroanal. Chem. 2006, 593, 99−104. All of
the studies were run using a glassy carbon working electrode, a
platinum wire counter electrode, and a Ag/AgCl reference electrode in
acetonitrile using tetrabutylammonium hexafluorophosphate as an
electrolyte.
(17) PC [IrIII/IrII] = −1.37 V vs SCE.
(18) (a) Iwasaki, T.; Takagawa, H.; Singh, S. P.; Kuniyasu, H.;
Kambe, N. J. Am. Chem. Soc. 2013, 135, 9604−9607. (b) Iwasaki, T.;
Takagawa, H.; Okamoto, K.; Singh, S. P.; Kuniyasu, H.; Kambe, N.
Synthesis 2014, 46, 1583−1592.
(19) (a) Lin, S.; Ischay, M. A.; Fry, C. G.; Yoon, T. P. J. Am. Chem.
Soc. 2011, 133, 19350−19353. (b) Hurtley, A. E.; Lu, Z.; Yoon, T. P.
Angew. Chem., Int. Ed. 2014, 53, 8991−8994. (c) Stevenson, S. M.;
Higgins, R. F.; Shores, M. P.; Ferreira, E. M. Chem. Sci. 2017, 8, 654−
660.
REFERENCES
■
(1) (a) Campos, K. R. Chem. Soc. Rev. 2007, 36, 1069−1084.
(b) Gehrtz, P. H.; Hirschbeck, V.; Ciszek, B.; Fleischer, I. Synthesis
2016, 48, 1573−1596. (c) Crozet, D.; Urrutigoity, M.; Kalck, P.
ChemCatChem 2011, 3, 1102−1118. (d) Klinkenberg, J. L.; Hartwig, J.
F. Angew. Chem., Int. Ed. 2011, 50, 86−95.
(2) (a) Roesky, P. W. Angew. Chem., Int. Ed. 2009, 48, 4892−4894.
(b) Chong, E.; Garcia, P.; Schafer, L. L. Synthesis 2014, 46, 2884−
2896. (c) Ryken, S. A.; Schafer, L. L. Acc. Chem. Res. 2015, 48, 2576−
2586.
(20) (a) Bennett, M. A.; Nicholls, J. C.; Rahman, A. K. F.; Redhouse,
A. D.; Spencer, J. L.; Willis, A. C. J. Chem. Soc., Chem. Commun. 1989,
18, 1328−1330. (b) Elgrishi, N.; Kurtz, D. A.; Dempsey, J. L. J. Am.
Chem. Soc. 2017, 139, 239−244.
(3) (a) Herzon, S. B.; Hartwig, J. F. J. Am. Chem. Soc. 2007, 129,
6690−6691. (b) Herzon, S. B.; Hartwig, J. F. J. Am. Chem. Soc. 2008,
130, 14940−14941. (c) Garcia, P.; Lau, Y. Y.; Perry, M. R.; Schafer, L.
L. Angew. Chem., Int. Ed. 2013, 52, 9144−9148. (d) Chong, E.; Brandt,
J. W.; Schafer, L. L. J. Am. Chem. Soc. 2014, 136, 10898−10901.
(e) Nako, A. E.; Oyamada, J.; Nishiura, M.; Hou, Z. Chem. Sci. 2016, 7,
6429−6434. (f) Liu, F.; Luo, G.; Hou, Z.; Luo, Y. Organometallics
2017, 36, 1557−1565. (g) Brandt, J. W.; Chong, E.; Schafer, L. L. ACS
Catal. 2017, 7, 6323−6330.
(21) (a) Allegra, G.; Lo Giudice, F.; Natta, G.; Giannini, U.;
Fagherazzi, G.; Pino, P. Chem. Commun. 1967, 1263−1265.
(b) Bonnemann, H.; Grard, C.; Kopp, W.; Pump, W.; Tanaka, K.;
̈
Wilke, G. Angew. Chem., Int. Ed. Engl. 1973, 12, 964−975.
(c) Groβheimann, G.; Jolly, P. W. Inorg. Chim. Acta 1998, 270, 60−
67. (d) Michigami, K.; Mita, T.; Sato, Y. J. Am. Chem. Soc. 2017, 139,
6094−6097. (e) Feldman, K. S.; Grega, K. C. J. Organomet. Chem.
1990, 381, 251−260. (f) Crewdson, P.; Bryce, D. L.; Rominger, F.;
Hofmann, P. Angew. Chem., Int. Ed. 2008, 47, 3454−3457. (g) Huo,
C.-F.; Li, Y.-W.; Beller, M.; Jiao, H. Organometallics 2005, 24, 3634−
3643.
(4) (a) Dorfler, J.; Preuß, T.; Schischko, A.; Schmidtmann, M.; Doye,
̈
S. Angew. Chem., Int. Ed. 2014, 53, 7918−7922. (b) Dorfler, J.; Preuß,
̈
T.; Brahms, C.; Scheuer, D.; Doye, S. Dalton Trans. 2015, 44, 12149−
12168.
(5) For reviews of photoredox catalysis, see: (a) Tellis, J. C.; Kelly, C.
B.; Primer, D. N.; Jouffroy, M.; Patel, N. R.; Molander, G. A. Acc.
Chem. Res. 2016, 49, 1429−1439. (b) Prier, C. K.; Rankic, D. A.;
MacMillan, D. W. C. Chem. Rev. 2013, 113, 5322−5363. (c) Romero,
N. A.; Nicewicz, D. A. Chem. Rev. 2016, 116, 10075−10166. (d) Skubi,
K. L.; Blum, T. R.; Yoon, T. P. Chem. Rev. 2016, 116, 10035−10074.
(6) For the seminal reports in metallo-photoredox catalysis, see:
(a) Tellis, J. C.; Primer, D. N.; Molander, G. A. Science 2014, 345,
433−436. (b) Zuo, Z.; Ahneman, D.; Chu, L.; Terrett, J.; Doyle, A. G.;
MacMillan, D. W. C. Science 2014, 345, 437−440.
(7) Ruhl, K. E.; Rovis, T. J. Am. Chem. Soc. 2016, 138, 15527−15530.
(8) (a) Cahiez, G.; Moyeux, A. Chem. Rev. 2010, 110, 1435−1462.
(b) Gao, K.; Yoshikai, N. Acc. Chem. Res. 2014, 47, 1208−1219.
(c) Usman, M.; Ren, Z.-R.; Wang, Y.-Y.; Guan, Z.-H. Synthesis 2017,
49, 1419−1443. (d) Fiebig, L.; Kuttner, J.; Hilt, G.; Schwarzer, M. C.;
Frenking, G.; Schmalz, H.-G.; Schafer, M. J. Org. Chem. 2013, 78,
̈
10485−10493. (e) Gandeepan, P.; Cheng, C.-H. Acc. Chem. Res. 2015,
48, 1194−1206. (f) Slowinski, F.; Aubert, C.; Malacria, M. Adv. Synth.
Catal. 2001, 343, 64−67.
(9) (a) Miyake, Y.; Nakajima, K.; Nishibayashi, Y. J. Am. Chem. Soc.
2012, 134, 3338−3341. (b) Dai, X.; Cheng, D.; Guan, B.; Mao, W.;
Xu, X.; Li, X. J. Org. Chem. 2014, 79, 7212−7219. (c) Chu, L.; Ohta,
C.; Zuo, Z.; MacMillan, D. W. C. J. Am. Chem. Soc. 2014, 136, 10886−
10889.
(10) See ref 7 as well as: (a) Rose, P.; Hilt, G. Synthesis 2016, 48,
̈
463−492. (b) Chen, Q.-A.; Kim, D. K.; Dong, V. M. J. Am. Chem. Soc.
2014, 136, 3772−3775.
(11) Or perhaps more correctly, if a radical-chain mechanism is
operative, the chains are short. For a discussion of this effect in
E
DOI: 10.1021/jacs.7b09252
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX