Electronically Mismatched Cycloaddition Reactions via First-Row Transition Metal, Iron(III)-Polypyridyl Complex

Article abstract of DOI:10.1021/acs.orglett.8b02541

The iron(III)-polypyridyl complex and its derivatives showed sufficient oxidizing potential to act as a one-electron oxidant, producing radical cations from olefins and promoting the efficient radical cation [2 + 2] and [2 + 4] cycloaddition reactions. Subsequent chain propagation afforded trisubstituted cyclobutane or cyclohexene derivatives, and this facile route enables the replacement of rare metals with sustainable, green, and inexpensive iron in radical cation cycloadditions.

Full text of DOI:10.1021/acs.orglett.8b02541

Letter
pubs.acs.org/OrgLett
Cite This: Org. Lett. XXXX, XXX, XXX−XXX
Electronically Mismatched Cycloaddition Reactions via First-Row
Transition Metal, Iron(III)−Polypyridyl Complex
Jung Ha Shin,† Eun Young Seong,† Hyeon Jin Mun, Yu Jeong Jang, and Eun Joo Kang*
Department of Applied Chemistry, Kyung Hee University, Yongin 17104, Korea
S
* Supporting Information
ABSTRACT: The iron(III)−polypyridyl complex and its derivatives showed sufficient oxidizing potential to act as a one-
electron oxidant, producing radical cations from olefins and promoting the efficient radical cation [2 + 2] and [2 + 4]
cycloaddition reactions. Subsequent chain propagation afforded trisubstituted cyclobutane or cyclohexene derivatives, and this
facile route enables the replacement of rare metals with sustainable, green, and inexpensive iron in radical cation cycloadditions.
ycloaddition, forming a cyclic adduct from two π-
electron systems, is the most commonly utilized class of
utilized photocatalysts are Ru(II) and Ir(I) polypyridyl
complexes.6 The redox potentials of Ru(II) complexes can
be tuned to initiate alkene oxidation and, at the same time,
prevent cycloreversion by utilizing diverse polypyridyl ligands.
The Ir(I) polypyridyl catalyst can promote the energy
transfer from the excited state of the Ir(I) complex that
has a higher lowest-lying triplet state energy level than that of
the organic substrates, and this triplet sensitization was
unrestricted by the redox property of the interactive alkene
substrate and promoted the single-electron transfer. Efforts to
explore other transition metals as photocatalysts have been
established, and Cr(III) polypyridyl complexes were success-
fully employed in radical cation mediated Diels−Alder
reactions between two electron-rich components as well as
two electron-deficient components. It is worth noting that
chromium, like iron, is an earth abundant, first-row transition
metal and that chromium complexes show higher excited
state reduction potentials (from +1.40 V to +1.84 V vs SCE)
which enabled the oxidation of stilbene.7
C
pericyclic reactions. Under general pericyclic cycloaddition
reactions, most C−C bond formations take place between the
electron-rich component and the electron-deficient reaction
partner, among the inherently different electronic nature of
the π-systems. Electronically mismatched cycloaddition
reactions between two electron-rich or two electron-deficient
components are considered to be challenging and require
high pressure, high energy, or electrochemical modification of
the reactants. As a solution to this problem, intermediary
radical cationic or anionic approaches have been proposed
employing single-electron oxidation or reduction systems.
While studies on the radical anion cycloadditions have been
relatively limited to enone substrates due to their nature of
generating stable anions,1 radical cation cycloadditions may
allow for a broader scope and more general application.
The first radical cation cyclodimerization of N-vinyl-
carbazole (NVC) was reported by Ledwith in 1969, via an
enamine radical cation using an Fe(III) or Ce(IV) salt as a
single-electron oxidant.2 Bauld developed various radical
cation catalyzed [2 + 4] and [2 + 2] cycloaddition reactions
using triaryl aminium (amine radical cation) antimonate to
oxidize 1,3-cyclohexadiene or anethole.3 More recently,
hypervalent iodine reagents (PhI(OAc)2 and DMP) were
used for styrene dimerization by Donohoe in 2016,4 and
since then, research on photocatalysis has expanded rapidly.
While photocatalytic systems utilizing the triaryloxopyrylium
salt, p-OMePTP,5 have been developed, the most commonly
Iron(III) perchlorate has been employed as a chemical
oxidant in the homodimerization of N-vinylcarbazole2 or
trans-anethole. Itoh discovered alumina-supported iron
catalysis in the early days, and Zhong recently expanded
this methodology in crossed intermolecular [2 + 2] and [2 +
4] cycloadditions.8 Arguably, the most straightforward
method for the preparation of cyclobutane rings is the [2 +
Received: August 8, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.8b02541
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
2] photocycloaddition reaction, and this prototypical photo-
chemical reaction has been demonstrated in numerous
synthetic applications. However, in-depth research is still
needed to adapt benign and sustainable iron catalysts for use
in these cycloadditions to address the cost, scalability, and
environmental impact of first-row transition metal catalysts
and to remove the need for high-energy irradiation
processes.9 Based on our research interests10 investigating
the competency of polypyridine ligands in controlling the
oxidation potential of Fe(III) catalysts, herein, we report a
radical cation catalyzed electron-mismatched heterocyclodi-
merization and Diels−Alder reaction to effect a highly
stereoconvergent outcome with E/Z mixtures of anethole
derivatives.
Table 1. [2 + 2] Cycloaddition Condition Screening
b
entry
Fe(III) complex (x)
solvent
yield (%)
1
2
3
4
Fe(phen)3(PF6)3 (15)
Fe(phen)3(PF6)3 (25)
Fe(phen)3(PF6)3 (10)
Fe(phen)3(PF6)3 (10)
Fe(phen)3(PF6)3 (10)
Fe(phen)3(PF6)3 (3)
Fe(phen)3(NTf2)3 (3)
Fe(Ph2phen)3(PF6)3 (3)
Fe(Me2phen)3(PF6)3 (3)
Fe(Me4phen)3(PF6)3 (3)
Fe(bpy)3(PF6)3 (3)
DCE
DCE
65
82
76
87
71
85
69
63
80
86
44
57
66
DCE/TFE (9:1)
DCE/TFE (4:1)
DCE/TFE (4:1)
DCE/TFE (4:1)
DCE/TFE (4:1)
DCE/TFE (4:1)
DCE/TFE (4:1)
DCE/TFE (4:1)
DCE/TFE (4:1)
DCE/TFE (4:1)
DCE/HFIP (9:1)
c
5
6
7
d
We initially focused our studies on the [2 + 2]
heterocyclodimerization of trans-anethole and styrene using
the simple iron complex Fe(phen)3(PF6)3. To our delight,
the trisubstituted cyclobutane 3aa was produced in 65% yield
for the chemoselective crossed cycloaddition without any
homodimerization cycloadduct formation, indicating that the
Fe(III) complex acts as a single-electron oxidant and can
selectively oxidize the trans-anethole to a radical cation
species. While increasing the amount of the iron complex up
to 25 mol % could induce a full conversion of the trans-
anethole, the cooperative use of trifluoroethanol enabled
sufficient reactivity even with a reduced amount (10 mol %)
of the iron complex, due to the solvent’s ability to stabilize
charged intermediates. Various fluorinated solvents and ratios
were utilized to find the optimum solvent ratio (Table 1,
entries 3−4, 13).11 Reducing the amount of styrene (5 equiv)
decreased the isolated yield of 3aa, but there were no
significant differences between using 10 and 3 mol % of the
Fe(III) complex (entries 5−6). Changing the anion to
bis(trifluoromethylsulfonyl)amide (NTf2) to improve solubil-
ity resulted in decomposition even at reduced temperature
(entry 7). Furthermore, various polypyridyl ligands were
utilized to control the oxidation potential of the Fe(III)
complexes (entries 8−12). As we expected, varying the
oxidation potential of the Fe(III) complex effected the yield:
for catalysts with phenanthroline ligands, the catalyst with the
lowest oxidation potential (E1/2 = +0.91 V) afforded the
highest yield,12 but the efficiency decreased somewhat when
the oxidation potential passed +1.15 V. Also, while the
addition of oxygen gas is known to benefit this reaction,6b,7a
using additional oxygen was not necessary in the general
optimization (Table S3), but was found to be beneficial for
cases where the electronic or steric demand decreased the
rate of the reaction (vide infra). Ultimately, the [Fe-
(Me4phen)3] (PF6)3 catalyst with a lower oxidizing capability
(+0.91 V) was determined to be the optimal Fe(III) complex
to effect this selective and mild oxidation (trans-anethole, E1/2
= +1.11 V).
8
9
10
11
12
Fe(tBu2bpy)3(PF6)3 (3)
Fe(Me4phen)3(PF6)3 (3)
d
13
a
Reaction conditions: 1a (0.1 mmol), 2a (1.0 mmol), solvent (0.1
b
c
d
M), rt, 3 h. Isolated yield. Reaction with 0.5 mmol of 2a. Reaction
at 0 °C. Ph2phen = 4,7-diphenyl-1,10-phenanthroline, Me2phen = 2,9-
dimethyl-1,10-phenanthroline, Me4phen = 3,4,7,8-tetramethyl-1,10-
phenanthroline, bpy = 2,2′-bipyridine, Bu2bpy = 4,4′-di-tert-butyl-
2,2′-dipyr-idine, TFE = 2,2,2-trifluoroethanol, HFIP = 1,1,1,3,3,3-
hexafluoro-2-propanol.
t
with recovery of trans-anethole (32%). The methoxy
substituent on benzene in the anethole radical cation could
be replaced with an allyloxy or benzyloxy substituent (3ba−
3ca), implying the necessity of an electron-donating group on
benzene, which is similar to other results involving a radical
cation intermediate.4−8 The dimethoxy-substituted anetholes,
1d and 1e, showed a slight reduction in reactivity,
presumably because the radical cation intermediate formed
is more stable and, thus, less reactive.
Treatment of reactive styrene 2b with a balloon of oxygen
afforded the production of 3db and 3eb in moderate
yield.7a,13 Longer carbon chains on the terminus of the
anethole olefin were introduced, and these reactants also
afforded the product in good yield with a balloon of oxygen.
The yields of adducts 3gb−3hb are noteworthy in that both
ethers and free hydroxy groups are tolerated in this
reaction.14 Additionally, the E/Z ratio of the anethole (1f,
2:1; 1g, 1:1; 1h, 3:1) had no impact on the outcome of the
cycloaddition; only trans stereochemical arrangements (anti-
products) were observed exclusively. Aliphatic olefins are not
suitable reaction partners, such as ethyl vinyl ketone,
dihydropyran, and dihydrofuran, and the observed difference
of alkene reactivities (1 vs 2) appears to correlate well with
their respective reduction potentials.
Under the optimized conditions, both electron-rich and
electron-poor styrenes reacted smoothly with the generated
radical cation of anethole 1a (Scheme 1). Reaction with 5
equiv of the electron-rich 4-methylstyrene proceeded well to
afford cyclobutane 3ab. 4-Chlorostyrene and 3- or 2-
bromostyrene required an additional amount of the Fe(III)
complex to complete the heterodimerization reaction,
suggesting that the reactivity of the styrene depended on
the degree of electron withdrawal and steric hindrance (3ad−
3ag). The reaction with 4-acetoxy styrene did not afford full
conversion, and only 43% of the product (3ah) was obtained
This radical cation cycloaddition strategy could be
expanded to the [2 + 4] Diels−Alder reaction described in
B
DOI: 10.1021/acs.orglett.8b02541
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
a
complex under more hydrophilic conditions to afford 5ca.
Additionally, using a mixture of the E and Z isomers of the
anethole derivatives (1f, 2:1; 1i, 1:1.5) yielded one
diastereomeric product, 5fa or 5ia respectively, and a longer
alkyl substituent or an ester functional group had little effect
on their reactivity.
Scheme 1. Radical Cation [2 + 2] Cycloaddition
To better understand the mechanism of these cyclo-
addition reactions, we first set out to identify the radical
cation intermediate and verify the role of the Fe(III)
polypyridyl complex. Treatment of 1a with Fe-
(Me4phen)3(PF6)3 and TEMPO afforded the corresponding
TEMPO adduct 7 (Scheme S1), identifying the position of
the carbon radical as well as the carbocation center. Next, the
stereochemical outcome of the [2 + 2] cycloaddition was
examined. In most of the reactions utilizing stereoisomeric
mixtures (Scheme 1, 1f−j), the stereochemical E/Z integrity
is lost, and only one diastereomer is produced. Further
experimentation showed that the E substrate was consumed
faster than the Z substrate. Typically, the radical cation
cycloaddition reaction can be explained using an ‘asynchro-
nous concerted pathway’ or ‘stepwise pathway.’ However,
several previous studies demonstrated that the distonic (long-
bond) radical cation intermediate formed with the first bond
is highly reactive, so any rearrangement including a single-
bond rotation is not thought likely to occur.3b,16 Thus, the
stereoconvergent phenomenon observed here can arise from
two possibilities: C−C rotation of cis- to trans-radical cation
anethole intermediate or cycloreversion during the reaction.
When 1f and 1i were subjected to the Fe(III) reaction
conditions without styrene, the isomerization reaction
proceeded, increasing the ratio of the E-isomer (Scheme
3A). At the same time, when cycloadduct 3aa was
resubjected to the reaction conditions, the monomeric
anethole 1a did not appear in the reaction medium. Attempts
to use isoprene to trap the putative radical cation
intermediate, which would have been present if cyclo-
reversion was possible under these conditions, did not afford
any Diels−Alder cycloadduct (Scheme 3B).6c Thus, the
stereoconvergent nature of the radical cation cycloaddition
under these reaction conditions arises from C−C rotation to
the stable trans intermediate rather than a cycloreversion
process. In a competitive cycloaddition experiment, the
reactivity of the diene toward the radical cation intermediate
was higher than the reactivity with the alkene, forming a 2:1
ratio of the [2 + 4] and [2 + 2] cycloadducts (Scheme 3C).
The addition of oxygen was also required for reactivity when
sterics and electronics hindered the reaction progress.
Though the role of oxygen in facilitating this specific process
is not certain, it should not be directly involved in enabling
iron(II) catalyst oxidation.17 Instead, oxygen may be
stabilizing the radical cation intermediates on the reaction
pathway, supported by the oxidative cleavage of the radical
cation-oxygen intermediate.18 More studies are necessary to
further establish the feasibility of the potential mechanistic
pathways and the critical role of oxygen.
a
Reaction conditions: 1 (0.1 mmol), 2 (1.0 mmol), DCE/TFE (4:1,
b
c
d
0.1 M), rt. Reaction with 1 mmol of 1a. Reaction at 0 °C. Reaction
with 5 equiv of 2b. DCE/TFE (9:1). Reaction under O2 balloon.
e
f
Scheme 2.15 Treatment of trans-anethole with Fe-
(Me4phen)3(PF6)3 in a DCE/TFE (9:1) solvent system
a
Scheme 2. Radical Cation [2 + 4] Cycloaddition
a
Reaction conditions: 1 (0.1 mmol), 4 (1.0 mmol), DCE/TFE (9:1,
b
c
d
0.1 M), rt. Reaction at 0 °C. Reaction with 5 equiv of 4d. DCE/
TFE (4:1). Reaction under O2 balloon.
5
Our mechanism for this radical cation cycloaddition is
proposed in Figure 1. Single-electron transfer from the
anethole to the Fe(III) complex generates the radical cation
(1a•+). At this point, the Z-configured anethole derivatives
can transform to the stable trans-radical cation intermediate
that can be stabilized by oxygen via equilibrium with a
dioxetane radical cation. Next, the olefin radical cation
undergoes asynchronous cycloaddition to the cyclized radical
afforded the cycloaddition with isoprene (4a) or 2,3-
dimethyl-1,3-butadiene (4b) to the corresponding adducts
(5aa and 5ab) in excellent yield. Endo stereoselectivity was
exclusively achieved with 1-substituted dienes (5ac, 5ad),
while 1,4-substituted dienes were unreactive. The allyloxy
substituted 1b showed excellent reactivity, while the
benzyloxy substituted 1c required 10 mol % of the Fe(III)
C
DOI: 10.1021/acs.orglett.8b02541
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 3. Mechanistic Investigation
Figure 1. Plausible mechanism for radical cation cycloaddition in Fe(III)−polypyridyl system.
cation intermediate. Electron transfer with another equivalent
of the anethole affords the product and propagates the
following process by generating another equivalent of the
radical cation (1a•+).
ASSOCIATED CONTENT
■
S
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.8b02541.
This work represents a detailed report of Fe(III)−
polypyridyl complexes applied to radical cation cycloaddition
reactions. Not only does this effort demonstrate the viability
of designing catalysts utilizing first-row transition metals, but
it also validates the notion that sensitive reactivity can be
discovered in these redox reaction systems. The current
efforts are dedicated toward the development of additional
catalyst systems based on earth-abundant Fe metal and
further synthetic applications, and these will be reported in
due course.
Experimental procedures and characterization data for
new compounds (PDF)
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: ejkang24@khu.ac.kr.
D
DOI: 10.1021/acs.orglett.8b02541
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
ORCID
Letter
(14) In the Cr photocatalysis (ref 7a), the positioning of the
pendant oxygen atom is critical, as the ether gave diminished yield
and the alcohol afforded no product whatsoever. These observations
were explained by a competitive intramolecular donation of the
oxygen electrons into the putative radical cation intermediate.
(15) See Table S1 for [2 + 4] condition screening experiments.
(16) Marquez, C. A.; Wang, H.; Fabbretti, F.; Metzger, J. O. J. Am.
Chem. Soc. 2008, 130, 17208.
(17) Mendez, M. A.; Partovi-Nia, R.; Hatay, I.; Su, B.; Ge, P. Y.;
Olaya, A.; Younan, N.; Hojeij, M.; Girault, H. H. Phys. Chem. Chem.
Phys. 2010, 12, 15163.
(18) (a) Chen, C.-C.; Fox, M. A. J. Comput. Chem. 1983, 4, 488.
(b) Additional experiments on the oxygen effect were described in
the Supporting Information (Table S3 and Scheme S2).
Eun Joo Kang: 0000-0002-7600-3609
Author Contributions
†J.H.S. and E.Y.S. contributed equally to this work.
Notes
The authors declare no competing financial interest.
́
ACKNOWLEDGMENTS
■
This study was supported by the Kyung Hee University
Research Fund (KHU-20160603).
REFERENCES
■
(1) (a) Baik, T. G.; Luis, A. L.; Wang, L. C.; Krische, M. J. J. Am.
Chem. Soc. 2001, 123, 6716. (b) Roh, Y.; Jang, H. Y.; Lynch, V.;
Bauld, N. L.; Krische, M. J. Org. Lett. 2002, 4, 611. (c) Yang, J.;
Felton, G. A. N.; Bauld, N. L.; Krische, M. J. J. Am. Chem. Soc.
2004, 126, 1634. (d) Ischay, M. A.; Anzovino, M. E.; Du, J.; Yoon,
T. P. J. Am. Chem. Soc. 2008, 130, 12886. (e) Du, J.; Yoon, T. P. J.
Am. Chem. Soc. 2009, 131, 14604. (f) Hurtley, A. E.; Cismesia, M.
A.; Ischay, M. A.; Yoon, T. P. Tetrahedron 2011, 67, 4442. (g) Du,
J.; Skubi, K. L.; Schultz, D. M.; Yoon, T. P. Science 2014, 344, 392.
(2) Bell, F. A.; Crellin, R. A.; Fujii, H.; Ledwith, A. J. Chem. Soc. D
1969, 251.
(3) (a) Bellville, D. J.; Wirth, D. D.; Bauld, N. L. J. Am. Chem. Soc.
1981, 103, 718. (b) Bauld, N. L.; Pabon, R. J. Am. Chem. Soc. 1983,
105, 633. (c) Bauld, N. L. Tetrahedron 1989, 45, 5307. (d) Bauld,
N. L.; Yang, J.; Gao, D. J. Chem. Soc., Perkin Trans. 2000, 2, 207.
(e) Gao, D.; Bauld, N. L. Tetrahedron Lett. 2000, 41, 5997.
(4) Colomer, I.; Barcelos, R. C.; Donohoe, T. Angew. Chem., Int.
Ed. 2016, 55, 4748.
(5) Riener, M.; Nicewicz, D. A. Chem. Sci. 2013, 4, 2625.
(6) (a) Ischay, M. A.; Lu, Z.; Yoon, T. P. J. Am. Chem. Soc. 2010,
132, 8572. (b) Lin, S.; Ischay, M. A.; Fry, C. G.; Yoon, T. P. J. Am.
Chem. Soc. 2011, 133, 19350. (c) Ischay, M. A.; Ament, M. S.;
Yoon, T. P. Chem. Sci. 2012, 3, 2807. (d) Lu, Z.; Yoon, T. P. Angew.
Chem., Int. Ed. 2012, 51, 10329. (e) Hurtley, A. E.; Lu, Z.; Yoon, T.
P. Angew. Chem., Int. Ed. 2014, 53, 8991.
(7) (a) Stevenson, S. M.; Shores, M. P.; Ferreira, E. M. Angew.
Chem., Int. Ed. 2015, 54, 6506. (b) Stevenson, S. M.; Higgins, R. F.;
Shores, M. P.; Ferreira, E. M. Chem. Sci. 2017, 8, 654. (c) Sarabia,
F. J.; Ferreira, E. M. Org. Lett. 2017, 19, 2865.
(8) (a) Ohara, H.; Itoh, T.; Nakamura, M.; Nakamura, E. Chem.
Lett. 2001, 30, 624. (b) Ohara, H.; Kiyokane, H.; Itoh, T.
Tetrahedron Lett. 2002, 43, 3041. (c) Yu, Y.; Fu, Y.; Zhong, F. Green
Chem. 2018, 20, 1743.
(9) (a) Hoyt, J. M.; Schmidt, V. A.; Tondreau, A. M.; Chirik, P. J.
Science 2015, 349, 960. (b) Hu, L.; Chen, H. J. Am. Chem. Soc.
2017, 139, 15564.
(10) Hwang, J. Y.; Baek, J. H.; Shin, T. I.; Shin, J. H.; Oh, J. W.;
Kim, K. P.; You, Y.; Kang, E. Org. Lett. 2016, 18, 4900.
(11) (a) Tian, Y.; Xu, X.; Zhang, L.; Qu, J. Org. Lett. 2016, 18,
268. (b) DiPoto, M. C.; Hughes, R. P.; Wu, J. J. Am. Chem. Soc.
2015, 137, 14861. (c) Colomer, I.; Batchelor-McAuley, C.; Odell,
B.; Donohoe, T. J.; Compton, R. G. J. Am. Chem. Soc. 2016, 138,
8855. (d) Solvent effect on redox potentials of anethole and Fe
catalysts are evaluated in Table S2.
(12) Although Fe(phen)3(PF6)3 gave a similar result with
Fe(Me4phen)4(PF6)3, a small amount of unknown byproduct was
observed in the former system on the TLC analysis. For treatment
under more effective reaction conditions, entry 10 in Table 1 was
chosen for the optimized reaction conditions.
(13) (a) Higgins, R. F.; Fatur, S. M.; Shepard, S. G.; Stevenson, S.
́
M.; Boston, D. J.; Ferreira, E. M.; Damrauer, N. H.; Rappe, A. K.;
Shores, M. P. J. Am. Chem. Soc. 2016, 138, 5451. (b) Yang, Y.; Liu,
Q.; Zhang, L.; Yu, H.; Dang, Z. Organometallics 2017, 36, 687.
E
DOI: 10.1021/acs.orglett.8b02541
Org. Lett. XXXX, XXX, XXX−XXX