Cyclization-endoperoxidation cascade reactions of dienes mediated by a pyrylium photoredox catalyst

Article abstract of DOI:10.3762/bjoc.10.128

Triarylpyrylium salts were employed as single electron photooxidants to catalyze a cyclization-endoperoxidation cascade of dienes. The transformation is presumed to proceed via the intermediacy of diene cation radicals. The nature of the diene component was investigated in this context to determine the structural requirements necessary for successful reactivity. Several unique endoperoxide structures were synthesized in yields up to 79%.

Full text of DOI:10.3762/bjoc.10.128

Cyclization–endoperoxidation cascade reactions of
dienes mediated by a pyrylium photoredox catalyst
Nathan J. Gesmundo and David A. Nicewicz*
Letter
Open Access
Address:
Beilstein J. Org. Chem. 2014, 10, 1272–1281.
Department of Chemistry, University of North Carolina at Chapel Hill,
Chapel Hill, NC 27599-3290, USA
doi:10.3762/bjoc.10.128
Received: 19 February 2014
Accepted: 06 May 2014
Published: 03 June 2014
Email:
David A. Nicewicz* - nicewicz@unc.edu
* Corresponding author
This article is part of the Thematic Series "Organic synthesis using
photoredox catalysis".
Keywords:
alkene; cascade; endoperoxide; oxidation; photoredox catalysis
Guest Editor: A. G. Griesbeck
© 2014 Gesmundo and Nicewicz; licensee Beilstein-Institut.
License and terms: see end of document.
Abstract
Triarylpyrylium salts were employed as single electron photooxidants to catalyze a cyclization–endoperoxidation cascade of dienes.
The transformation is presumed to proceed via the intermediacy of diene cation radicals. The nature of the diene component was
investigated in this context to determine the structural requirements necessary for successful reactivity. Several unique endoper-
oxide structures were synthesized in yields up to 79%.
Introduction
Endoperoxides are a structurally unique class of naturally- Classical approaches to the introduction of cyclic peroxides
occurring compounds that feature a reactive cyclic peroxide typically rely on cycloadditions of alkenes and dienes with
moiety of varying ring sizes (Figure 1). The lability of the endo- singlet oxygen. However, ene processes can often compete,
cyclic peroxide O–O bond engenders these compounds with a leading to complex mixtures of hydroperoxide adducts [1,6-8].
range of important biological functions, most notably, antima- More recently, cyclization reactions of hydroperoxides with
larial and antitumor activity (e.g., artemisinin, yingzhaosu A pendant alkenes or alkynes have been developed. Selected
and merulin C) [1-4]. From a synthetic standpoint, the installa- examples include Pd(II)-catalyzed hydroalkoxylation reactions
tion of the endoperoxide moiety presents a significant chal- of unsaturated hydroperoxides [9], Au(I)-catalyzed endoperoxi-
lenge due to its susceptibility to reduction and for this reason, is dation of alkynes [10] and Brønsted acid-catalyzed enantiose-
ideally introduced late-stage in target-oriented synthesis. Addi- lective acetalization/oxa-Michael addition cascade reactions of
tionally, many endoperoxide natural products possess architec- peroxyquinols [11].
turally complex frameworks (e.g., artemisinin, yingzhaosu A,
muurolan-4,7-peroxide) [5] that pose significant synthetic chal- While these extant methods are effective at installing the endo-
lenges.
peroxide functional group, our interest lay in developing strate-
1272

Beilstein J. Org. Chem. 2014, 10, 1272–1281.
Figure 1: Selected examples of endoperoxide-containing natural products.
gies that simultaneously forged both the cyclic peroxide as well (Scheme 1, reaction 2) [15]. More recently, Kamata and Kim
as the carbon framework to rapidly build molecular complexity. have employed this reaction manifold to forge endoperoxides
For this reason, we were inspired by the work of Miyashi, who, from 1,2-divinylarene precursors [16].
during the course of the investigation of the cation radical Cope
rearrangement, discovered an intriguing endoperoxide-forming We envisioned that the scope of this transformation could be
reaction (Scheme 1, reaction 1). Upon exposure of 1,5 and 1,6- extended to include non-styrenal dienes as well as alternative
dienes to catalytic quantities of 9,10-dicyanoanthracene (DCA) cyclization modes, provided that a potent single electron
under UV irradiation in the presence of oxygen, bicyclic oxidant could be identified (Ered > +1.5 V vs SCE). Given the
endoperoxides were obtained [12]. Formation of the 1,2-diox- paucity of ground-state single electron oxidants capable of this
anes was presumed to occur via single electron oxidation of the task, we elected to employ photooxidation catalysts. Addition-
diene by the excited state DCA followed by either 6-endo or ally, we sought to select visible light-activated organic single
7-endo cyclization modes to generate a fleeting distonic cation electron oxidants that do not readily sensitize singlet oxygen
radical species. Interception of the distonic cation radical by [17-19]. For these reasons, we were attracted to the use of
triplet oxygen and back electron transfer completes the catalytic triarylpyrylium salts, as they have excited state reduction poten-
cycle. While later reports expanded the scope of this transfor- tials in excess of +1.7 V vs SCE (Scheme 2) [20]. In addition,
mation modestly [13,14], we felt that this strategy had potential prior work demonstrates that these catalysts are productive in
to be a more general method. Indeed, during the course of this cation radical mediated [4 + 2], [2 + 2], oxygenation, and
work, the Yoon research group disclosed a similar strategy rearrangement chemistry [21,22]. We also sought to delineate
employing Ru(bpz)32+ as the photooxidant to effect a 5-exo the reactivity of the diene with respect to its structure to better
cyclization/endoperoxidation cascade of bis(styrene) substrates predict the mode of diene cation radical cyclization (5-exo vs
Scheme 1: Endoperoxide formation via cation radicals. In both examples, single electron oxidation is followed quickly by cyclization to form stabilized
distonic cation radical intermediates. The distonic intermediates are trapped by O2 and furnish the shown bicyclic products after reduction.
1273

Beilstein J. Org. Chem. 2014, 10, 1272–1281.
Scheme 2: Diversification strategy for endoperoxide synthesis by single electron transfer. E*red vs SCE [20].
6-endo). Herein is reported an organocatalytic photoredox- an aliphatic alkene. Using catalyst 1c in DCM at −41 °C under
mediated strategy for the endoperoxidation of 1,5-dienes using irradiation with 470 nm LEDs afforded endoperoxide 3a in 40%
3O2 to rapidly generate complex endoperoxide frameworks.
yield after 5 hours (Table 1, entry 1). The observed endoper-
oxide was attributed to a 5-exo cyclization mode of the diene
cation radical followed by capture of molecular oxygen. The
use of acetonitrile as solvent gave none of the desired adducts
Results and Discussion
Reaction optimization
We began our investigation into endoperoxidation conditions (Table 1, entry 2). Further improvement of the chemical yield of
with diene 2a as the substrate as it contained both a styene and 3a was realized by increasing the reaction concentration
Table 1: Reaction Optimization and Control Experiments.
Entry
Conditionsa
Conversion
100%
Yieldb
40%
0%
1
2
3
4
5
2 mol % 1c
0.01 M DCM, −41 °C
2 mol % 1c
0.01 M MeCN, −41 °C
100%
2 mol % 1c
0.02 M DCM, −41 °C
100%
63%
63%
70%
1 mol % 1c
0.05 M DCM, −41 °C
100%
0.7 mol % 1c
100%
0.07 M DCM, −41 °C
6c
7
Excluding O2
Excluding hν
11%
0%
0%
0%
0%
0%
0%
8
Excluding catalyst 1c
0%
9
9-Mes-10-Me-Acr-BF4 in place of 1c
Rose Bengal in place of 1c
100%
100%
10d
All reactions carried out in oxygen-saturated solvents unless otherwise noted. a−41 °C found to be the optimum temperature during initial substrate/
reaction optimization. bYields with respect to (Me3Si)2O 1H NMR internal standard. cReaction carried out under N2 atmosphere in DCM. dReaction
carried out in oxygen saturated MeOH at room temperature using a white flood lamp.
1274

Beilstein J. Org. Chem. 2014, 10, 1272–1281.
(Table 1, entries 3–5), resulting in a 70% yield (1H NMR) of the
endoperoxide (Table 1, entry 5). In all reactions, substrate
conversion was 100%, with the remainder of the mass balance
in all cases representing oxidative decomposition pathways.
When O2 was excluded from the reaction (Table 1, entry 6),
none of the endoperoxide was observed. Not surprisingly,
exclusion of light or catalyst 1c from the reaction (Table, 1
entries 7 and 8 respectively) resulted in no product formation.
Interestingly, when 9-mesityl-10-methylacridinium tetrafluoro-
borate, which has been used with success in other photooxida-
tion processes [23-26], was used in place of 1c, complete
consumption of 2a was observed, but 3a was not produced.
Once again only oxidative decomposition pathways seemed
active, this time possibly due to superoxide formation. Lastly, to
exclude the intervention of a singlet oxygen mechanism, we
Figure 2: ORTEP of 3a.
conducted the reaction in the presence of Rose Bengal. Under pyranyl radical 1•. Cyclization of diene cation radical 4 then
these conditions, we observed only 1O2 ene reactivity with the forms stabilized distonic cation radical intermediate 5, which is
isoprenyl group (65% yield of hydroperoxide), underscoring the intercepted by O2 to form 6. Single electron reduction, either
unique reactivity garnered by this catalyst system (see from 1• to regenerate active catalyst 1 or from another substrate
Supporting Information File 1 for hydroperoxide characteriza- equivalent in a chain process, forms the desired bicyclic endo-
tion). Suitable crystals of 3a provided X-ray confirmation of the peroxide.
endoperoxide structure (Figure 2).
DFT calculations suggest that the formation of the initial
We invoke a mechanism similar to that proposed by Miyashi distonic cation radical 5 is exothermic by approximately
[12] and Yoon [15] in their respective transformations 3 kcal/mol relative to cation radical 4 [27]. Superficially, this is
(Scheme 3). Following excitation of triarylpyrylium tetrafluoro- rationalized by the increased substitution on 5 relative to 4. In
borate catalyst 1, one-electron oxidation of the 1,5-diene sub- addition, the majority of the spin density on 5 is located on the
strate produces localized cation radical intermediate 4 and isoprenyl group (DFT-5, Scheme 3). This may be fortuitous as
Scheme 3: Proposed mechanism for endoperoxide synthesis from tethered dienes.
1275

Beilstein J. Org. Chem. 2014, 10, 1272–1281.
stereoselectivity in the oxygen addition step is irrelevant, 1,4-dione, presumably from double oxidative cleavage of the
whereas the opposite scenario involving a benzylic radical inter- 1,5-diene (Table 2, entry 1). Unfortunately, attempts to move
mediate would require a stereospecific addition of oxygen to the away from 2b to less electron-rich dienes such as 2c, 2d and 2e
same face of the cyclopentane system as the isoprenyl cation in (Table 1, entries 2–4), failed to produce any of the desired
order for endoperoxide formation to occur.
endoperoxide products and mainly oxidative cleavage adducts
were observed.
With optimized conditions identified, we sought to examine the
scope of the reaction with respect to the diene structure. Given the lack of reactivity of this alkene substitution pattern,
Miyashi’s 1,5-diene (2b; E1/2Ox = 1.22 V vs SCE [12]) afforded we turned our attention to the investigation of diene structures
a 50% yield of the expected endoperoxide along with ~5% of a similar to successful endoperoxidation substrate 2a. Removal of
Table 2: Diene structure investigation for endoperoxidation cascade.
Entry
1
Substrate
Expected adduct
Yieldb
50%
3b
3c
2b
2c
2
3
0%
0%
2d
2e
3d
3e
4
5
0%
66%
3a
2a
6
32%
3f
2f
1276

Beilstein J. Org. Chem. 2014, 10, 1272–1281.
Table 2: Diene structure investigation for endoperoxidation cascade. (continued)
7
9%
0%
16%
0%
0%
0%
3g
3h
3i
2g
8
2h
9
2i
10
3j
2j
11
3k
2k
12
3l
2l
All reactions carried out in oxygen-saturated solvents. Solvents examined: DCM, CHCl3, MeCN, PhMe, acetone. aCatalysts 1a, 1b, and 1c screened
for reactivity with all substrates. bAverage of two isolated yields.
the geminal dimethyl group (2f) resulted in significantly dimin- this hypothesis. While electron-rich styrenes gave modest
ished yields of the endoperoxide adduct, likely due to the lack amounts of product formation (2f, 2g and 2i; Table 2, entries 6,
of a Thorpe–Ingold effect present in 2a. These experiments also 7 and 9), styrenes with either weakly donating (4-Me; Table 2,
demonstrated that the electron-rich arene was necessary for re- entry 10) or even withdrawing (4-Cl; Table 2, entry 11) func-
activity (cf. 2e and 2f; Table 2, entries 4 and 6). The impor- tionality furnished none of the expected endoperdoxides. In
tance of the electron-rich arene may lie in the necessary distonic these cases, oxidative degradation was observed as was the case
cation radical intermediate: the electron-rich arene may provide with the highly oxidizable 2-furyl group (Table 2, entry 12).
greater stability to the distonic intermediate formed after 5-exo- Interestingly, 3,4-dimethoxystyrene-substituted diene 2h also
trig cyclization, ultimately ensuring it remains to intercept gave none of the desired adduct, which we attributed to lack of
oxygen. A further survey of the styrene component supported charge density on the alkene [28,29].
1277

Beilstein J. Org. Chem. 2014, 10, 1272–1281.
We next investigated the endoperoxidation cascade by replacing entry 1). A diene bearing a tetrasubstituted alkene (2n) was
the isoprenyl substituent with a variety of other alkenes. A reactive in this context, giving polycyclic endoperoxide 3n in
pendant styrene afforded the desired endoperoxide adduct in 64% yield (6.5:1 dr). The use of 1,2-, 1,1-dialkyl as well as
68% yield, albeit with no diastereocontrol (2m, 1:1 dr; Table 3, monoalkyl-substituted alkenes appeared to completely disfavor
Table 3: Variation of the tethered alkene component.
Entry
1b
Substrate
Observed product
Yielda
68%, 1:1 d.r.
3m
2m
2c
3d
64%, 6.5:1 d.r.
3n
2n
2o
36%
3o
3p
4d
5d
37%
27%
2p
2q
3q
6e
7e
–
0%
0%
2r
–
2s
Reactions carried out in oxygen-saturated solvents. aAverage of two isolated yields. b1:1 mixture of separable diastereomers. c6.5:1 mixture of insep-
arable diastereomers. Presumed major diastereomer shown. dDesired endoperoxide never observed. eMultiple conditions tested, no productive
chemistry observed.
1278

Beilstein J. Org. Chem. 2014, 10, 1272–1281.
the endoperoxidation pathway and resulted in the unexpected
isolation of α-allyl ketones (Table 3, entries 3–5) in modest
yields. Based on the Miyashi precedent, the formation of these
adducts can be rationalized by invoking a formal [3,3]-
rearrangement of the initial cation radical intermediate. The
competing 6-endo cyclization mode and formation of distonic
cation radical 9 ultimately provides access to the more stabi-
lized cation radical 10, which undergoes oxidative cleavage to
afford the corresponding α-allyl ketones (Scheme 4). In the
absence of the geminal dimethyl group (2r, 2s), neither the
Cope-like reactivity or the endoperoxidation was observed
(Table 3, entries 6 and 7).
Lastly, we elected to explore cyclization modes similar to the
Yoon work under the developed conditions. Diene 2t was
anticipated to undergo 6-exo-trig cyclization to form the neces-
sary distonic cation radical intermediate (Table 4, entry 1). The
expected trioxabicyclo[3.3.1]nonane product was formed, albeit
Scheme 4: Competing formal [3,3] pathway.
Table 4: Other cyclization modes and substrate designs.
Entry
1b
Substrate
Desired product
Yielda
16%
2t
3t
3u
3v
2
79%, 5.7:1 d.r.
2u
3
<5%
0%
2v
4c
2w
Reactions carried out in oxygen-saturated dichloromethane. aAverage of two isolated yields. bCarried out in 0.4 M DCE after solvent and concentra-
tion optimization. cCatalysts 1a or 1b were also tested but failed to furnish the endoperoxide.
1279

Beilstein J. Org. Chem. 2014, 10, 1272–1281.
5. Adio, A. M.; König, W. A. Phytochemistry 2005, 66, 599–609.
doi:10.1016/j.phytochem.2005.01.015
in low yields (16%), where the remainder of the mass balance
was attributed to oxidative degradation.
6. Adam, W.; Griesbeck, A. G. In CRC Handbook of Organic
Photochemistry and Photobiology; Horspool, W. M.; Song, P., Eds.;
CRC Press: Boca Raton, 1992.
Bis(styrene) 2u afforded the identical fused 1,2-dioxane
observed in Yoon’s report in 79% yield (5.7:1 dr). Tethered
trisubstituted aliphatic alkene substrate 2v, along with
bis(styrene) substrate 2w were unfortunately unsuccessful,
producing neither of the desired fused 1,2-dioxane products in
appreciable amounts. Degradation pathways were dominant for
2v and mainly unreacted starting material was observed 2w.
7. Bloodworth, A. J.; Eggelte, H. J. In Singlet Oxygen; Frimer, A. A., Ed.;
CRC Press: Boca Raton, 1985; Vol. II.
8. Iesce, M. R. In Synthetic Organic Photochemistry; Griesbeck, A. G.;
Mattay, J., Eds.; Marcel Dekker: New York; pp 299–363.
9. Harris, J. R.; Waetzig, S. R.; Woerpel, K. A. Org. Lett. 2009, 11,
3290–3293. doi:10.1021/ol901046z
10.Krabbe, S. W.; Do, D. T.; Johnson, J. S. Org. Lett. 2012, 14,
5932–5935. doi:10.1021/ol302848m
11.Rubush, D. M.; Morges, M. A.; Rose, B. J.; Thamm, D. H.; Rovis, T.
J. Am. Chem. Soc. 2012, 134, 13554–13557. doi:10.1021/ja3052427
12.Miyashi, T.; Konno, A.; Takahashi, Y. J. Am. Chem. Soc. 1988, 110,
3676–3677. doi:10.1021/ja00219a062
Conclusion
In the presence of an organic single electron photooxidant, a
variety of dienes were demonstrated to undergo a cyclization/
endoperoxidation cascade sequence to form 1,2-dioxanes.
Requirements for successful diene reactivity are the presence of
an oxidizable olefin and an alkene that can efficiently react with
the putative alkene cation radical to form a more stable distonic
cation radical. If available, a Cope-like pathway can compete
and suppress endoperoxide formation. With these parameters in
mind, this reaction could provide a platform for the discovery of
novel biologically-active endoperoxides.
13.Takahashi, Y.; Okitsu, O.; Ando, M.; Miyashi, T. Tetrahedron Lett.
1994, 35, 3953–3956. doi:10.1016/S0040-4039(00)76711-6
14.Kamata, M.; Ohta, M.; Komatsu, K.-i.; Kim, H.-S.; Wataya, Y.
Tetrahedron Lett. 2002, 43, 2063–2067.
doi:10.1016/S0040-4039(02)00166-1
15.Parrish, J. D.; Ischay, M. A.; Lu, Z.; Guo, S.; Peters, N. R.; Yoon, T. P.
Org. Lett. 2012, 14, 1640–1643. doi:10.1021/ol300428q
16.Kamata, M.; Hagiwara, J.-i.; Hokari, T.; Suzuki, C.; Fujino, R.;
Kobayashi, S.; Kim, H.-S.; Wataya, Y. Res. Chem. Intermed. 2013, 39,
127–137. doi:10.1007/s11164-012-0637-3
17.Narayanam, J. M. R.; Stephenson, C. R. J. Chem. Soc. Rev. 2011, 40,
102–113. doi:10.1039/b913880n
Supporting Information
18.Xuan, J.; Xiao, W.-J. Angew. Chem., Int. Ed. 2012, 51, 6828–6838.
doi:10.1002/anie.201200223
Supporting Information File 1
19.Prier, C. K.; Rankic, D. A.; MacMillan, D. W. C. Chem. Rev. 2013, 113,
5322–5363. doi:10.1021/cr300503r
Experimental procedures and characterization data.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-10-128-S1.pdf]
20.Martiny, M.; Steckhan, E.; Esch, T. Chem. Ber. 1993, 126, 1671–1682.
doi:10.1002/cber.19931260726
See for excited state reduction potential converted from E vs NHE to E
vs SCE.
Supporting Information File 2
X-ray data.
21.Miranda, M. A.; Garcia, H. Chem. Rev. 1994, 94, 1063–1089.
doi:10.1021/cr00028a009
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-10-128-S2.pdf]
22.Kamata, M.; Kaneko, J.-i.; Hagiwara, J.-i.; Akaba, R. Tetrahedron Lett.
2004, 45, 7423–7428. doi:10.1016/j.tetlet.2004.08.077
23.Hamilton, D. S.; Nicewicz, D. A. J. Am. Chem. Soc. 2012, 134,
18577–18580. doi:10.1021/ja309635w
Acknowledgements
24.Fukuzumi, S.; Ohkubo, K. Chem. Sci. 2013, 4, 561–574.
doi:10.1039/c2sc21449k
We gratefully acknowledge the David and Lucile Packard
Foundation for financial support. Additionally, we thank
Nathan Romero for DFT calculations and Dr. Peter White for
X-ray analysis.
25.Nguyen, T. M.; Nicewicz, D. A. J. Am. Chem. Soc. 2013, 135,
9588–9591. doi:10.1021/ja4031616
26.Perkowski, A. J.; Nicewicz, D. A. J. Am. Chem. Soc. 2013, 135,
10334–10337. doi:10.1021/ja4057294
27.Gaussian 09, Revision D.01; Gaussian, Inc.: Wallingford CT, 2013.
28.Riener, M.; Nicewicz, D. A. Chem. Sci. 2013, 4, 2625–2629.
doi:10.1039/c3sc50643f
References
1. Ando, W., Ed. Organic Peroxides; John Wiley & Sons: New York, 1992.
2. Dembitsky, V. M. Eur. J. Med. Chem. 2008, 43, 223–251.
doi:10.1016/j.ejmech.2007.04.019
29.Yamashita, T.; Yasuda, M.; Isami, T.; Tanabe, K.; Shima, K.
Tetrahedron 1994, 50, 9275–9286.
3. Li, H.; Huang, H.; Shao, C.; Huang, H.; Jiang, J.; Zhu, X.; Liu, Y.;
Liu, L.; Lu, Y.; Li, M.; Lin, Y.; She, Z. J. Nat. Prod. 2011, 74,
1230–1235. doi:10.1021/np200164k
doi:10.1016/S0040-4020(01)85505-8
4. Chokpaiboon, S.; Sommit, D.; Bunyapaiboonsri, T.; Matsubara, K.;
Pudhom, K. J. Nat. Prod. 2011, 74, 2290–2294.
doi:10.1021/np200491g
1280

Beilstein J. Org. Chem. 2014, 10, 1272–1281.
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.10.128
1281