Visible-Light-Mediated Iminyl Radical Generation from Benzyl Oxime Ether: Synthesis of Pyrroline via Hydroimination Cyclization

Article abstract of DOI:10.1021/acs.orglett.8b02429

The treatment of an O-(4-methoxybenzyl) oxime ether bearing an olefin substituent and 1-chloroanthraquinone (1-Cl-AQN) catalyst in 2-butanone under visible-light irradiation affords pyrroline via an iminyl radical intramolecular hydroimination. Mechanistic studies indicate that iminyl radical generation mainly proceeds by hydrogen abstraction of the photocatalyst from the benzyl position of the oxime. Moreover, the hydrogen atom was identified in circulation from the benzylic position of the substrates between AQN and 2-butanone to quench the carbon radical without requiring any additional reagents.

Full text of DOI:10.1021/acs.orglett.8b02429

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Visible-Light-Mediated Iminyl Radical Generation from Benzyl
Oxime Ether: Synthesis of Pyrroline via Hydroimination Cyclization
Kaoru Usami, Eiji Yamaguchi, Norihiro Tada, and Akichika Itoh*
Gifu Pharmaceutical University 1-25-4, Daigaku-nishi, Gifu 501-1196, Japan
S
* Supporting Information
ABSTRACT: The treatment of an O-(4-methoxybenzyl)
oxime ether bearing an olefin substituent and 1-chloroan-
thraquinone (1-Cl-AQN) catalyst in 2-butanone under visible-
light irradiation affords pyrroline via an iminyl radical
intramolecular hydroimination. Mechanistic studies indicate
that iminyl radical generation mainly proceeds by hydrogen
abstraction of the photocatalyst from the benzyl position of
the oxime. Moreover, the hydrogen atom was identified in
circulation from the benzylic position of the substrates
between AQN and 2-butanone to quench the carbon radical
without requiring any additional reagents.
itrogen-centered radicals (NCRs) are valuable active
species for constructing nitrogen-containing heterocyclic
Scheme 1. Iminyl Radical Generation Methods via N−O
Bond Cleavage
N
skeletons, which are included in many bioactive substances.1−3
Notably, iminyl radicals are useful radical species in NCRs that
enable the embedding of the imine group for easy molecular
conversion in a cyclic skeleton.4 A typical method for
generating iminyl radicals is the homolysis of the N−O bond
of oxime. The cleavage of the N−O bond is commonly
engaged under harsh reaction conditions, such as high
temperature or UV irradiation (Scheme 1A).5 Recently, the
groups of Yu, Leonori, and Studer reported methods for
producing iminyl radicals via a photoredox reaction using a
photocatalyst (Scheme 1B).6 In these methods, the cleavage of
the oxime’s N−O bond underwent single-electron transfer
(SET) between the photoredox catalyst and the substrate to
afford iminyl radicals under moderate reaction conditions.
However, the photoredox pathway required re-oxidation
reagents or hydrogen donor reagents, such as aryl disulfide
or 1,4-cyclohexadiene, to regenerate the photocatalyst. Thus, a
novel and economical method for generating iminyl radicals
under mild conditions is required.
Hydrogen atom transfer (HAT) is a molecularly activated
mode that abstracted a hydrogen atom from a weak C−H
bond of a substrate.7 The HAT catalyst activates the reagent
and acts as a hydrogen donor if the hydrogenated catalyst
passes along a hydrogen atom to the substrate.8 Iminyl radical
generation methods via hydrogen abstraction have been
reported; however, they have been only applied to basic
experiments.9 No generation method in the HAT pathway has
been reported.
With this perspective, we have reported anthraquinone
(AQN) catalytic benzylic selective oxidation,10 wherein a
benzyl radical, generated by hydrogen abstraction from the
excited AQN, was thought to be a plausible reaction
intermediate. Thus, a novel iminyl radical evolution method
can be developed using regioselective HAT. We assumed that
an iminyl radical was produced in two steps (Scheme 1C): (1)
Received: July 31, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b02429
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a b
,
excited AQN abstracts the hydrogen atom from benzyl oxime
ether 1 and (2) generated benzyl radical 2 immediately
undergoes β-scission to produce an iminyl radical 3 and an
aldehyde. Herein, we report new synthesis pathways of
pyrrolines via the intramolecular cyclization of an iminyl
radical generated from an O-(4-methoxybenzyl) oxime ether
under visible-light irradiation.
Table 1. Optimization of the Reaction Conditions
entry
photocatalyst
additive
solvent
yield (%)
Since hydrogen abstraction with AQN from benzyl ether has
not been reported yet, the generation of a benzyl radical was
confirmed using the oxidation reaction of the preceding
example.10 After identifying hydrogen abstraction at the
benzylic position (Table S1), we attempted a pyrroline
formation reaction using an arylmethyl oxime ether possessing
an olefin moiety to confirm the iminyl radical production
(Scheme 2). At first, O-benzyl oxime ether (1c) afforded the
1
2
3
4
5
6
7
8
9
10
2-Cl-AQN
2-Cl-AQN
2-Cl-AQN
2-Cl-AQN
2-Cl-AQN
1-Cl-AQN
eosin Y
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
AcOH
2-butanone
acetone
EtOAc
MeOH
DMSO
2-butanone
2-butanone
2-butanone
2-butanone
2-butanone
2-butanone
52
7
8
0
0
68
0
0
45
58
72
1-Cl-AQN
1-Cl-AQN
1-Cl-AQN
a b
,
Scheme 2. Evaluation of Various Aryl Groups
c
11
K2CO3
a
Reaction conditions: A solution of 1d (0.15 mmol), photocatalyst
(0.1 equiv), and additive (1.0 equiv) in solvent (2 mL) was irradiated
with fluorescent lamp (23 W × 4) under an argon atmosphere and
b
c
stirred for 20 h. 1H NMR yields. 2-Butanone (5 mL).
With the optimized reaction conditions in hand, we then
studied the scope and limitation of the reaction (Scheme 3).
Oxime 1d afforded the corresponding pyrroline 5a in 78%
a b
,
Scheme 3. Synthesis of Pyrrolines
a
Reaction conditions: A solution of 1 (0.15 mmol), 2-Cl-AQN (0.1
equiv) and in 2-butanone (2 mL) was irradiated with a fluorescent
lamp (23 W × 4) under an argon atmosphere and stirred for 20 h.
b
1H NMR yields. The isolated yield is shown in parentheses.
corresponding pyrroline 5a in 13% yield in the presence of 2-
chloro-AQN in 2-butanone under visible light irradiation using
a 23 W fluorescent lamp. As hypothesized, the electron density
of the aryl group affected benzylic selective hydrogen
abstraction10 and β-scission. The oxime with an electron-
deficient aryl group did not afford the corresponding product
(1a, 1b); however, electron-rich aryl groups (1d, 1e),
particularly 4-methoxybenzyl (PMB) group 1d, afforded the
corresponding pyrroline 5a. When a more electron-rich aryl
group (1e) was used, substrate was not completely consumed.
Thus, we used O-PMB oxime ether 1d as the model
substrate and tuned the reaction conditions (Table 1). As a
photocatalyst, AQNs generally showed good reactivity. This
reaction could be performed using 2-butanone. Among our
examinations, a combination of 2-butanone and 1-Cl-AQN (1-
chloroanthraquinone) afforded a good result due to its
solubility and low nucleophilicity (entries 1−6). In contrast,
a large amount of starting material remained unreacted when
eosin Y was used as the photocatalyst (entry 7). This reaction
did not proceed without the photocatalyst (entry 8).
Furthermore, additive K2CO3 accelerated the reaction slightly
against the inert effect of an acid (entries 1, 9, and 10). After
finely tuning the reaction conditions, 72% of 5a was obtained
when 0.1 equiv of 1-Cl-AQN was used as the photosensitizer,
1.0 equiv of K2CO3 as the additive, and 5 mL of 2-butanone as
the solvent under visible-light irradiation (entry 11). Upon
diluting the substrate concentration, the permeability of the
visible light improved.
a
Reaction conditions: A solution of 1 (0.15 mmol), 1-Cl-AQN (0.1
equiv), and K2CO3 (1.0 equiv) in 2-butanone (5 mL) was irradiated
with a fluorescent lamp (23 W × 4) under an argon atmosphere and
b
c
d
stirred for 20 h. Isolated yields. 1-Cl-AQN (0.05 equiv). 1-Cl-
AQN (0.12 equiv). 1-Cl-AQN (0.15 equiv). dr = 3:1.
e
f
B
DOI: 10.1021/acs.orglett.8b02429
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
isolated yield in the presence of 0.05 equiv of the catalyst. In
general, the substituent at the aryl group on the main chain of
the oxime did not affect the reactivity of the substrate. For
example, the oximes with an aryl group possessing electron-
donating groups (1f−h) afforded the corresponding products
(5f−h) in good yields. Halogen substituents did not affect this
reactivity considerably, and the corresponding products (5i−
m) could be obtained in good yields regardless of the ortho-,
meta-, and para-positions. The cyano group, an electron-
withdrawing group, also afforded the product 5n in good
yields. Furthermore, the substrate with the pyridine moiety
could be applied to this reaction to produce the corresponding
product (5o) in 32% yield. If an oxime with an alkyl chain
instead of an aryl group 1p was used, the yield of product 5p
decreased and a nitrile byproduct 6p, which was thought to be
generated by the fragmentation of iminyl radical,11 was
detected. When tetralin oxime 1q was used, tricyclic product
5q was obtained in 81% yield. We examined substrate 1r and
expected to observe an increase in the yield due to the
Thorpe−Ingold effect; however, the yield of 5r remained
moderate. The oxime ether possessing trisubstituted internal
olefin 1s was also converted to the corresponding pyrroline 5s
in moderate yield.
reduction potentials of the substrate, it was confirmed that the
generation of the radical species via a proton concerted
electron transfer (PCET) mechanism was possible.13 Thus, the
oxidation potential of the substrate 1d was +1.28 V vs FC/
FC+, whereas the oxidation potential of 1-Cl-AQN was +1.9 V
vs FC/FC+. This indicates that the benzyl radical generation
using PCET was possible. Furthermore, the success of the
iminyl radical generation via PCET was experimentally
confirmed using 9,10-dicyanoanthracene, known as a single-
electron oxidizing agent (Table 2, entry 4). Thus, the progress
of the reaction could be confirmed; therefore, it is expected
that the benzyl radical was generated by the concerted progress
of two mechanisms, namely hydrogen abstraction and PCET.
We conducted several deuterium-labeling experiments to
identify the hydrogen donor source and found that the
deuterated substrate at the benzylic position [D2]1d rarely
afforded the deuterated product (Scheme 4A). However,
Scheme 4. Deuterium-Labeling Experiments
To further investigate the primary reaction mechanism,
several controlled experiments were conducted (Table 2).
a
Table 2. Control Experiments
deuterated 2-butanone was detected by 2H NMR spectroscopy
(Figure S1). When the reaction was performed in acetone-d6, a
deuterated product was detected, although this hydroimination
reaction progressed to low yield in acetone (Table 1, entry 2,
and Scheme 4B).
entry
altered reaction conditions
under air atmosphere
added TEMPO (1.0 equiv)
under dark
yield (%)
1
2
3
4
0
0
0
On the basis of the above results, we proposed a plausible
pathway (Scheme 5). At first, the photoexcited 1-Cl-AQN (1-
added 9,10-DCA instead of 1-Cl-AQN
33
a
1H NMR yields.
Scheme 5. Plausible Mechanism
When O-PMB oxime 1d was exposed under an atmospheric
condition, a primary alcohol 7a,6b generated by trapping
molecular oxygen with the terminal carbon radical, was
detected via 1H NMR spectroscopy without any desired
product (entry 1). Inhibition of this reaction using a radical
scavenger, such as TEMPO (1.0 equiv), suggested that the
pyrroline formation involved a radical pathway (entry 2). The
desired pyrroline product was not detected under dark reaction
conditions (entry 3). Furthermore, the time-course experiment
with intermittent light intervals showed that the cleavage of the
N−O bond only occurred upon irradiation with a fluorescent
lamp (Scheme S1). This indicates that the excited AQN by
visible-light irradiation was involved in the hydrogen abstract
reaction and that this reaction does not progress through a
radical-chain pathway where the carbon radical 5′ abstracts the
hydrogen atom from the benzylic position of the substrate
instead of excited AQN. Subsequently, we considered the
mechanism of the benzyl radical production. In addition to the
hydrogen abstraction, the photoexcited AQN acted as a single-
electron oxidizing agent.12 With this substrate, the same benzyl
radical species could be generated via single-electron oxidation
at the electron-rich aryl group and deprotonation by base at
the benzylic position. Therefore, by measuring oxidation−
Cl-AQN*), excited by visible-light irradiation, abstracted the
hydrogen atom at the benzylic position or oxidized the PMP
group via SET in conjunction with the proton transfer to afford
benzyl radical intermediate 2. We assumed that the basic
conditions promoted the PCET. Benzyl radical intermediate 2
was then immediately transformed into iminyl radical species 3
C
DOI: 10.1021/acs.orglett.8b02429
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Org. Chem. 2017, 2017, 2108. (d) Zhao, H. B.; Hou, Z. W.; Liu, Z. J.;
Zhou, Z. F.; Song, J.; Xu, H. C. Angew. Chem., Int. Ed. 2017, 56, 587.
(4) (a) Walton, J. C. Molecules 2016, 21, 660. (b) Sun, J.; He, Y.; An,
X.-D.; Zhang, X.; Yu, L.; Yu, S. Org. Chem. Front. 2018, 5, 977.
(c) Tang, J.; Sivaguru, P.; Ning, Y.; Zanoni, G.; Bi, X. Org. Lett. 2017,
19, 4026.
(5) (a) Jackman, M. M.; Im, S.; Bohman, S. R.; Lo, C. C. L.; Garrity,
A. L.; Castle, S. L. Chem. - Eur. J. 2018, 24, 594. (b) Cai, Y.; Jalan, A.;
Kubosumi, A. R.; Castle, S. L. Org. Lett. 2015, 17, 488. (c) Wang, Y.-
F.; Zhang, F.-L.; Chiba, S. Org. Lett. 2013, 15, 2842. (d) McBurney,
R. T.; Slawin, A. M. Z.; Smart, L. A.; Yu, Y.; Walton, J. C. Chem.
Commun. 2011, 47, 7974.
(6) SET reduction pathway: (a) Jiang, H.; An, X.; Tong, K.; Zheng,
T.; Zhang, Y.; Yu, S. Angew. Chem., Int. Ed. 2015, 54, 4055.
(b) Davies, J.; Booth, S. G.; Essafi, S.; Dryfe, R. A. W.; Leonori, D.
Angew. Chem., Int. Ed. 2015, 54, 14017 SET oxidation pathway:.
(c) Jiang, H.; Studer, A. Angew. Chem., Int. Ed. 2017, 56, 12273.
(d) Davies, J.; Sheikh, N. S.; Leonori, D. Angew. Chem., Int. Ed. 2017,
56, 13361.
and p-anisaldehyde 4 through N−O bond cleavage. 5-exo-trig-
Cyclization of iminyl radical 3 produced the 5′ carbon radical,
which trapped the hydrogen atom from 2-butanone to afford
the corresponding pyrroline. AQN was presumably regen-
erated by offering the hydrogen from AQH• to butanone,
thereby completing the catalytic cycle.
In conclusion, we developed a synthesis pathway of
pyrrolines using an unprecedented method of iminyl radical
generation from an O-PMB oxime ether using visible light and
a photocatalyst. We were unable to conclude that the reaction
proceeded only through the HAT pathway; however, this
approach conclusively afforded the iminyl radical generation
method under mild conditions using visible-light energy. The
previous iminyl radical method with a photoredox catalyst
required an additional hydrogen donor and oxidant to quench
the carbon radical, which was generated by the iminyl radical
cyclization. In contrast, our proposed method was highly atom
efficient as no additional reagent was used due to the
circulation of the hydrogen atom abstracted from the benzylic
position of the substrate between the AQN and 2-butanone.
Furthermore, this is the first report on a photoredox method
using a readily prepared O-PMB oxime ether and is thought to
contribute significantly to the knowledge of iminyl radical
chemistry.
̈
(7) (a) Farney, E. P.; Feng, S. S.; Schafers, F.; Reisman, S. E. J. Am.
Chem. Soc. 2018, 140, 1267. (b) Short, M. A.; Blackburn, J. M.;
Roizen, J. L. Angew. Chem., Int. Ed. 2018, 57, 296. (c) Zhou, R.; Goh,
Y. Y.; Liu, H.; Tao, H.; Li, L.; Wu, J. Angew. Chem., Int. Ed. 2017, 56,
16621. (d) Chen, D.-F.; Chu, J. C. K.; Rovis, T. J. Am. Chem. Soc.
2017, 139, 14897.
(8) (a) Fukuyama, T.; Nishikawa, T.; Yamada, K.; Ravelli, D.;
Fagnoni, M.; Ryu, I. Org. Lett. 2017, 19, 6436. (b) Zhang, Y.-Q.;
Jakoby, V.; Stainer, K.; Schmer, A.; Klare, S.; Bauer, M.; Grimme, S.;
ASSOCIATED CONTENT
* Supporting Information
■
̈
Cuerva, J. M.; Gansauer, A. Angew. Chem., Int. Ed. 2016, 55, 1523.
S
(c) Tanaka, H.; Sakai, K.; Kawamura, A.; Oisaki, K.; Kanai, M. Chem.
Commun. 2018, 54, 3215.
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b02429.
(9) McCarroll, A. J.; Walton, J. C. J. Chem. Soc., Perkin Trans. 2000,
2, 1868.
(10) Shimada, Y.; Hattori, K.; Tada, N.; Miura, T.; Itoh, A. Synthesis
2013, 45, 2684.
Experimental details and NMR spectra (PDF)
(11) (a) Yu, X.-Y.; Chen, J.-R.; Wang, P.-Z.; Yang, M.-N.; Liang, D.;
Xiao, W.-J. Angew. Chem., Int. Ed. 2018, 57, 738. (b) Ohsawa, A.;
Kawaguchi, T.; Igeta, H. Chem. Pharm. Bull. 1982, 30, 4352.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: itoha@gifu-pu.ac.jp.
ORCID
̈
(12) Gorner, H. Photochem. Photobiol. 2003, 77, 171.
(13) (a) Lennox, J. C.; Dempsey, J. L. J. Phys. Chem. B 2017, 121,
10530. (b) Kolmar, S. S.; Mayer, J. M. J. Am. Chem. Soc. 2017, 139,
10687. (c) Chen, M.; Zhao, X.; Yang, C.; Xia, W. Org. Lett. 2017, 19,
3807.
Eiji Yamaguchi: 0000-0002-1167-8832
Norihiro Tada: 0000-0003-2871-5406
Akichika Itoh: 0000-0003-3769-7406
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Dr. Bunji Uno (Gifu Pharmaceutical University) for
help with cyclic voltammetry. We thank Enago (www.enago.
jp) for English language review.
REFERENCES
■
(1) (a) Vitaku, E.; Smith, D. T.; Njardarson, J. T. J. Med. Chem.
2014, 57, 10257. (b) Taylor, R. D.; MacCoss, M.; Lawson, A. D. G. J.
Med. Chem. 2014, 57, 5845.
(2) (a) Attanasi, O. A.; Merino, P.; Spinelli, D..Targets in Heterocyclic
Systems; Chemistry and Properties; Royal Society of Chemistry, 2016.
(b) Sahn, J. J.; Granger, B. A.; Martin, S. F. Org. Biomol. Chem. 2014,
12, 7659. (c) Kaur, N.; Dwivedi, J.; Kishore, D. Synth. Commun. 2014,
44, 1671. (d) Rizzo, A.; Harutyunyan, S. R. Org. Biomol. Chem. 2014,
12, 6570. (e) Martin, S. F. Adv. Heterocycl. Chem. 2013, 110, 73.
(3) (a) Zard, S. Z. Chem. Soc. Rev. 2008, 37, 1603. (b) Crespin, L. N.
S.; Greb, A.; Blakemore, D. C.; Ley, S. V. J. Org. Chem. 2017, 82,
13093. (c) Reina, D. F.; Dauncey, E. M.; Morcillo, S. P.; Svejstrup, T.
D.; Popescu, M. V.; Douglas, J. J.; Sheikh, N. S.; Leonori, D. Eur. J.
D
DOI: 10.1021/acs.orglett.8b02429
Org. Lett. XXXX, XXX, XXX−XXX