Catalytic Oxygenative Allylic Transposition of Alkenes into Enones with an Azaadamantane-Type Oxoammonium Salt Catalyst

Article abstract of DOI:10.1002/chem.201702512

The first catalytic oxygenative allylic transposition of unactivated alkenes into enones has been developed using an oxoammonium salt as the catalyst. This reaction converts various tri- and trans-disubstituted alkenes into their corresponding enones with transposition of their double bonds at ambient temperature in good yields. The use of a less-hindered azaadamantane-type oxoammonium salt as the catalyst and a combination of two distinct stoichiometric oxidants, namely, iodobenzene diacetate and magnesium monoperoxyphthalate hexahydrate (MMPP?6 H₂O) are essential to facilitate the enone formation efficiently.

Full text of DOI:10.1002/chem.201702512

DOI: 10.1002/chem.201702512
Communication
&
Organocatalysis |Hot Paper|
Catalytic Oxygenative Allylic Transposition of Alkenes into Enones
with an Azaadamantane-Type Oxoammonium Salt Catalyst
Shota Nagasawa, Yusuke Sasano, and Yoshiharu Iwabuchi*[a]
Abstract: The first catalytic oxygenative allylic transposi-
tion of unactivated alkenes into enones has been devel-
oped using an oxoammonium salt as the catalyst. This re-
action converts various tri- and trans-disubstituted alkenes
into their corresponding enones with transposition of
their double bonds at ambient temperature in good
yields. The use of a less-hindered azaadamantane-type
oxoammonium salt as the catalyst and a combination of
two distinct stoichiometric oxidants, namely, iodobenzene
diacetate and magnesium monoperoxyphthalate hexahy-
drate (MMPP·6H2O) are essential to facilitate the enone
formation efficiently.
Alkene is a ubiquitous and readily-accessible structural motif in
organic chemistry, and selective modifications of alkenes have
been extensively investigated to develop various methods.[1]
Those methods are applied to the synthesis of valuable chemi-
cal products, including pharmaceuticals, agrochemicals, and
polymer materials. The selective oxygenation of alkenes at the
periphery of the double bond poses attractive opportunities to
access useful molecules, such as allylic alcohols and enones,
spurring chemists to achieve these transformations efficiently.[2]
Schematically, two possible ways can be drawn for oxygena-
tion of an alkene to give an enone (Scheme 1). One is a direct
oxygenation at the allylic position of alkene substrate 1 to give
enone 2’ (“normal” allylic oxygenation, Scheme 1a). This type
of reaction has been investigated for a long time to establish
numbers of reliable methods to synthesize complex mole-
cules.[2d] The other is an oxygenative allylic transposition that
involves a migration of the alkene functionality of 1 to give re-
gioisomeric enone 2 (Scheme 1b). To the best of our knowl-
edge, a sequence of an ene reaction of an alkene and singlet
oxygen (Schenck ene reaction) and subsequent acetylation–de-
hydroacetoxylation of the resulting allyl hydroperoxide[3] is the
only known single-step method for this transformation.[4] Al-
though some successful applications of this method to the
synthesis of complex molecules have been demonstrated,[5]
this method has some major drawbacks: it is potentially dan-
Scheme 1. Enone synthesis from alkenes.
gerous due to the explosive nature of the hydroperoxide inter-
mediate, it gives modest yields of enones from trisubstituted
alkenes due to poor regioselectivity of the singlet oxygen ene
reaction, and it is generally not applicable to trans-disubstitut-
ed alkenes due to lack of the reactivity of singlet oxygen
toward them.[6] Here, we describe the first catalytic oxygena-
tive allylic transposition of alkenes into enones, featuring the
use of a less-hindered azaadamantane-type oxoammonium salt
as a catalyst (Scheme 1c). This novel method is applicable to
not only trisubstituted alkenes but also trans-disubstituted al-
kenes.
The reaction design is shown in Scheme 2. We set a member
of 2-azaadamantane N-oxyl (AZADO)-derived oxoammonium
salts as the key catalyst for the catalytic oxygenative allylic
transposition, which we have investigated as catalysts for oxi-
dation of alcohols[7] and silyl enol ethers.[8] Based on some pre-
liminary results in our pioneering study of the reactivity of the
azaadamantane-type oxoammonium salts toward alkenes, we
envisioned that a trisubstituted or acyclic alkene reacts with an
azaadamantane-type oxoammonium salt through an O-prefer-
ential ene-like addition to afford an allylic alkoxyamine (the
first reaction).[9] We envisaged that connecting this alkoxy-
amine formation with two transformations, namely, alkoxy-
amine oxidation-oxy-Cope elimination giving a carbonyl com-
pound and a hydroxylamine (the second reaction),[10] and oxi-
[a] Dr. S. Nagasawa, Dr. Y. Sasano, Prof. Dr. Y. Iwabuchi
Department of Organic Chemistry
Graduate School of Pharmaceutical Sciences, Tohoku University
6-3 Aoba, Aramaki, Aoba-ku, Sendai 980-8578 (Japan)
E-mail: y-iwabuchi@m.tohoku.ac.jp
Supporting information and the ORCID number for the author of this article
can be found under https://doi.org/10.1002/chem.201702512.
Chem. Eur. J. 2017, 23, 10276 –10279
10276
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
at room temperature to give alkoxyamines 3 and 4 in a high
combined yield (89%) with good regioselectivity (5.4:1). In
contrast, the ene reaction of singlet oxygen and 1a gave poor
regioselectivity (1:1).[12] The oxidative conversion of alkoxy-
amine 3 into enone 2a was examined using various oxidants.
The screening of oxidants indicated that only organic peracids
promoted this reaction in high yield, and identified magnesi-
um monoperoxyphthalate hexahydrate (MMPP·6H2O) as the
most promising oxidant for the catalytic reaction. MMPP·6H2O
converted 3 into 2a in 83% yield, which proceeded selectively
even in the presence of an equimolar amount of alkene 1a.[12]
Encouraged by the aforementioned promising results, we
proceeded to investigate the catalytic conditions (Table 1). The
À
conditions consisting of AZADO+BF4 (20 mol%), MMPP·6H2O
Scheme 2. Reaction design.
dation of a hydroxylamine into an oxoammonium ion (the
third reaction), would form a catalytic system to convert an
alkene into an enone with the transposition of the alkene func-
tionality. Although a 2,2,6,6-tetramethylpiperidine-type oxoam-
monium salt is also known to lead the first ene-like reaction,
which was reported by Bobbitt and co-workers,[11] we anticipat-
ed that this highly hindered oxoammonium salt is much less
suitable for the catalyst than the azaadamantane-type oxoam-
monium salts. To develop the catalytic system, competing
direct oxidation of the alkene substrate by a stoichiometric oxi-
dant (e.g., epoxidation) is anticipated as a problem. We expect-
ed that the less-hindered azaadamantane-type oxoammonium
salts and alkoxyamine/hydroxylamine intermediates should
react more readily with alkene substrates and stoichiometric
oxidants, respectively, than tetramethylpiperidine-type ones.
Therefore, the use of the azaadamantane-type oxoammonium
salts would increase the efficiency of the catalytic cycle to
eventually suppress the problematic side reaction. Further-
more, the less-hindered azaadamantane-type oxoammonium
salts would exhibit broader substrate scope, whereas the tetra-
methylpiperidine-type ones would react with only electron-rich
trisubstituted alkenes.
Table 1. Initial investigation of the catalytic conditions[a]
Entry
1
Conditions
1H NMR yield [%][b]
5
2a
6
1a
AZADO+BF4À (20 mol%)
MMPP·6H2O (1.5 equiv)
LiBF4 (2 equiv)
43
17
11
N.D.
2
AZADO+BF4À (20 mol%)
PhI(OAc)2 (1.0 equiv)
MMPP·6H2O (0.55 equiv)
LiBF4 (1 equiv)
64
N.D.
13
N.D.
powdered MS3A (200 mg)
[a] All of the reactions were conducted with slow addition of the solution
of alkene 1a into the mixture of the other reagents in solvent over 2 h
using a syringe pump. [b] Yields were determined by 1H NMR of crude
products using 1,3,5-trimethoxybenzene as an internal standard. N.D.=
not detected.
To evaluate the feasibility of the reaction design, we exam-
ined the first and second reactions separately using stoichio-
metric amounts of reagents (Scheme 3). The ene-like addition
of 2-azaadamantane N-oxoammonium tetrafluoroborate
(AZADO+BF4À) to trisubstituted alkene 1a proceeded smoothly
(1.5 equiv), and LiBF4 (2 equiv) in acetonitrile, which was estab-
lished for a catalytic oxygenation of silyl enol ethers into 1,2-di-
ketones,[8] afforded the expected enone 2a, but its yield (43%)
was unsatisfying (entry 1). The close analysis of the reaction
identified epoxide 5 and tertiary alcohol 6 as byproducts,
which would be generated by the direct oxidation of alkene
1a with MMPP·6H2O and by the oxidative elimination of the
azaadamantane moiety from alkoxyamine 4, respectively. We
considered that minimization of the amount of MMPP·6H2O by
using another oxidant that selectively promotes the third reac-
tion in Scheme 2 (a hydroxylamine to an oxoammonium ion)
would suppress the generation of epoxide 5 to improve the
yield of 2a. Gratifyingly, we found that the use of 0.55 equiv of
MMPP·6H2O with 1.0 equiv of PhI(OAc)2 suppressed the gener-
ation of 5 below our detectable limit and increased the yield
of 2a to 64% (entry 2). Note that molecular sieves were neces-
sary to ensure good reproducibility.
À
Scheme 3. Stoichiometric reaction of AZADO+BF4 and alkene 1a (top); oxi-
dative conversion of alkoxyamine 3 (bottom).
Chem. Eur. J. 2017, 23, 10276 –10279
www.chemeurj.org
10277
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Table 2. Further optimization of the reaction conditions[a]
Entry
Change(s) from the
initial conditions
1H NMR yield [%][b]
2a
5
6
1a
1
2
none
64
26
N.D.
28
13
N.D.
N.D.
11
À
4-NHAc-TEMPO+BF4
instead of AZADO+BF4
MeCN-PhMe (1:4)
instead of MeCN
AZADOL
instead of AZADO+BF4
AZADO+BF4À (10 mol%)
LiBF4 (0 equiv)
À
À
3
71
N.D.
N.D.
10
7
N.D.
25
4[c,d]
52
5[d]
6[d]
62
17
12
11
8
N.D.
N.D.
55
[a] All of the reactions were conducted with slow addition of the solution
of alkene 1a into the mixture of the other reagents in solvent over 2 h
using a syringe pump. [b] Yields were determined by 1H NMR of crude
products using 1,3,5-trimethoxybenzene as an internal standard.
[c] 1.2 equiv of PhI(OAc)2 and LiBF4 were used. [d] MeCN-PhMe (1:4) was
used as solvent. N.D.=not detected.
On the basis of these key findings, other reaction parame-
ters were optimized (Table 2).[12] The screening of oxoammoni-
um salts with various steric and electronic properties indicated
À
that AZADO+BF4 showed the best catalytic efficiency. It is
noteworthy that 4-acetamido-2,2,6,6-tetramethylpiperidine-1-
À
oxoammonium tetrafluoroborate (4-NHAc-TEMPO+BF4 ), which
is known to react with trisubstituted alkenes to give alkoxy-
amines,[11] hardly catalyzed the reaction (entry 2). This result in-
dicates that the less-hindered structure of an oxoammonium
salt is essential to realize the catalytic reaction. The screening
of solvents revealed that the mixed solvent of acetonitrile and
toluene (1:4 v/v) gave the best yield of 2a (71%; entry 3). Note
that this mixed solvent promoted the reaction much more effi-
ciently than acetonitrile alone, consuming the alkene substrate
Scheme 4. Substrate scope. All of reactions were conducted with slow addi-
tion of the solution of alkene 1 into the mixture of other reagents in solvent
over 2 h using a syringe pump except for 1t. Yields are of isolated product.
[a] 30 mol% of AZADO+BF4À, 1.5 equiv of PhI(OAc)2, and 0.75 equiv of
MMPP·6H2O were used. [b] 30 mol% of AZADO+BF4À was used. [c] 30 mol%
of 4-Cl-AZADO+BF4À, 1.5 equiv of PhI(OAc)2, and 0.75 equiv of MMPP·6H2O
were used. [d] Reaction conditions: 4-Cl-AZADO+BF4À (30 mol%), PhI(OAc)2
(1.5 equiv), MMPP·6H2O (0.75 equiv), LiNTf2 (1.25 equiv), MS3A (200 mg),
MeCN-PhMe (1:4), RT. 1t was added to a suspension of 4-Cl-AZADO+BF4
À
within two hours. The conditions that generate AZADO+BF4
À
,
in situ from commercially available hydroxylamine AZADOL did
PhI(OAc)2, MMPP·6H2O, LiNTf2 and MS3A in MeCN-PhMe in one portion.
not achieve the full conversion of alkene 1a. (entry 4). Reduc-
À
ing the amount of AZADO+BF4 to 10 mol% afforded enone
2a in good yield (62%), and a detectable amount of undesired
epoxide 5 (Table 1, entry 5). A significant decrease of the reac-
tion efficiency was observed under the conditions without
LiBF4 (entry 6). Eventually, we identified the conditions shown
in entry 3 as the optimal conditions.
The optimal conditions tolerated various functional groups in-
cluding esters (2a, 2 f, and 2i), a phthalimide (2b), electron-
rich and -deficient aryl groups (2c–e), a cyclopropane (2 f),
a silyl ether (2g), an amide (2h), and a carbamate (2j). Note
that a terminal alkene was intact in this reaction condition (2i).
The optimal conditions allowed for the oxidation of other
types of trisubstituted alkenes, namely, cyclohexylidenes (2k
and 2l), a 3-pentylidene (2m), and others (2n and 2o) to give
With the optimized conditions in hand, we evaluated their
substrate scope (Scheme 4). The functional group tolerance
was evaluated by using isopropylidene-type substrates 1a–j.
Chem. Eur. J. 2017, 23, 10276 –10279
www.chemeurj.org
10278
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Living Polymerizations: From Mechanisms to Applications, Wiley-VCH,
Weinheim, 2009; f) P. G. Andersson, I. J. Munslow, Modern Reduction
Method, Wiley-VCH, Weinheim, 2008.
their corresponding enones in good to high yields. Further-
more, a-pinene and a pregnenolone derivative were also con-
verted into enones 2p and 2q, respectively. Unfortunately,
a simple cyclohexene-type substrate gave not its correspond-
ing enone 2r, but a complex mixture. Less electron-rich trans-
disubstituted alkenes that are difficult to be oxidized by the
conventional method using singlet oxygen were also exam-
ined. Although the optimal conditions for trisubstituted alkene
1a did not convert the trans-disubstituted alkenes into enones
2s and 2t, the use of a more electrophilic oxoammonium salt,
namely, 4-chloro-2-azaadamantane N-oxoammnium tetrafluo-
roborate (4-Cl-AZADO+BF4À),[9] effectively converted these al-
kenes into enones 2s and 2t in 61 and 64% yields, respective-
ly. Note that LiNTf2, instead of LiBF4, is especially effective for
synthesis of 2t.
In summary, we have developed the first catalytic oxygena-
tive allylic transposition of alkenes using an oxoammonium
salt as a catalyst, which poses an alternative synthetic method
to obtain enones from alkenes. This reaction tolerates and
works efficiently with various functional groups and structural-
ly diverse alkenes including trans-disubstituted ones. The use
of the less-hindered azaadamantane-type oxoammonium salt
as the catalyst and the combination of the two distinct stoi-
chiometric oxidants play key roles for facilitating the catalytic
reaction efficiently. This is a rare example of oxoammonium
salt-catalyzed oxidative transformation of alkenes.[13] We be-
lieve that the reaction developed here will contribute to the
synthesis of new enones and facilitate the research of applied
chemistry focusing on unique the reactivity of oxoammonium
salts.
[2] For recent reviews, see: a) A. L. García-Cabeza, F. J. Moreno-Dorado,
M. J. Ortega, F. M. Guerra, Synthesis 2016, 48, 2323–2342; b) S. E. Mann,
L. Benhamou, T. D. Sheppard, Synthesis 2015, 47, 3079–3117; c) K. X.
Chen, P. F. Zhang, Y. Wang, H. R. Li, Green Chem. 2014, 16, 2344–2374;
d) V. Weidmann, W. Maison, Synthesis 2013, 45, 2201–2221; e) A. Naka-
mura, M. Nakada, Synthesis 2013, 45, 1421–1451; f) T. Punniyamurthy, S.
Velusamy, J. Iqbal, Chem. Rev. 2005, 105, 2329–2363. g) For selected
recent examples, see: E. J. Horn, B. R. Rosen, Y. Chen, J. Z. Tang, K. Chen,
M. D. Eastgate, P. S. Baran, Nature 2016, 533, 77–81; h) S. E. Ammann,
W. Liu, M. C. White, Angew. Chem. Int. Ed. 2016, 55, 9571–9575; Angew.
Chem. 2016, 128, 9723–9727; i) P. S. Wang, P. Liu, Y. J. Zhai, H. C. Lin,
Z. Y. Han, L. Z. Gong, J. Am. Chem. Soc. 2015, 137, 12732–12735; j) Y.
Wang, Y. Kuang, Y. H. Wang, Chem. Commun. 2015, 51, 5852–5855;
k) W. J. Ang, Y. Lam, Org. Biomol. Chem. 2015, 13, 1048–1052.
[3] E. D. Mihelich, D. J. Eickhoff, J. Org. Chem. 1983, 48, 4135–4137.
[4] Some conditions for conversion of terminal alkenes into enals with mi-
gration of the double bonds using PdII catalysis have been reported.
See: X. Y. Xing, N. R. O’Connor, B. M. Stoltz, Angew. Chem. Int. Ed. 2015,
54, 11186–11190; Angew. Chem. 2015, 127, 11338–11342.
[5] a) I. Hayakawa, K. Niida, H. Kigoshi, Chem. Commun. 2015, 51, 11568–
11571; b) J. B. Farcet, M. Himmelbauer, J. Mulzer, Eur. J. Org. Chem.
2013, 4379–4398; c) F. Lçbermann, L. Weisheit, D. Trauner, Org. Lett.
2013, 15, 4324–4326; d) J. B. Farcet, M. Himmelbauer, J. Mulzer, Org.
Lett. 2012, 14, 2195–2197; e) H. J. Liu, Y. L. Ho, J. D. Wu, K. S. Shia, Syn-
lett 2001, 1805–1807; f) H. J. Liu, K. S. Shia, Tetrahedron 1998, 54,
13449–13458; g) H. J. Liu, K. S. Shia, Y. X. Han, D. Q. Sun, Y. Wang, Can.
J. Chem. 1997, 75, 646–655.
[6] Substrate scope and regioselectivity of the singlet oxygen ene reaction
have been studied well. For selected reviews, see: a) E. L. Clennan, Tet-
rahedron 2000, 56, 9151–9179; b) M. Stratakis, M. Orfanopoulos, Tetra-
hedron 2000, 56, 1595–1615.
[7] a) M. Shibuya, M. Tomizawa, I. Suzuki, Y. Iwabuchi, J. Am. Chem. Soc.
2006, 128, 8412–8413; b) Y. Iwabuchi, Chem. Pharm. Bull. 2013, 61,
1197–1213.
[8] M. Hayashi, M. Shibuya, Y. Iwabuchi, Synlett 2012, 1025–1030.
[9] The siteselectivity of the ene-like addition of the azaadamantane-type
oxoammonium species to the alkenes (N- or O-preferred) is dependent
on the structure of the alkene substrates. Empirically, cyclic disubstitut-
ed alkenes are forecasted to undergo the N-preferential addition. The
computational study into the site-selectivity is underway. see: S. Naga-
sawa, Y. Sasano, Y. Iwabuchi, Angew. Chem. Int. Ed. 2016, 55, 13189–
13194; Angew. Chem. 2016, 128, 13383–13388 and its Supporting Infor-
mation.
[10] For some examples of alkoxyamine oxidation-oxy-Cope elimination, see
: a) B. A. Chalmers, J. C. Morris, K. E. Fairfull-Smith, R. S. Grainger, S. E.
Bottle, Chem. Commun. 2013, 49, 10382–10384; b) B. Zhang, A. Studer,
Org. Lett. 2013, 15, 4548–4551; c) M. Hartmann, Y. Li, A. Studer, J. Am.
Chem. Soc. 2012, 134, 16516–16519; d) T. Inokuchi, H. Kawafuchi, Tetra-
hedron 2004, 60, 11969–11975.
Acknowledgements
This research was partly supported by JSPS KAKENHI Grant
Numbers 23105011 in “Advanced Molecular Transformations
by Organocatalysis” and JP16H00998 in “Precisely Designed
Catalysts with Customized Scaffolding”, and by a Grant-in-Aid
for JSPS fellows (No. 266364 to S.N.). We thank Dr. Shuhei
Manabe (Tohoku University) for useful discussion.
Conflict of interest
[11] P. P. Pradhan, J. M. Bobbitt, W. F. Bailey, Org. Lett. 2006, 8, 5485–5487.
[12] See the Supporting Information for details.
The authors declare no conflict of interest.
[13] Only one report on catalytic allylic oxidation of alkenes employing elec-
trochemically-generated oxoammonium cation has been published, in
which a completely different reaction mechanism from that which we
rely on was proposed. The substrates applied to the oxygenative allylic
transposition were limited to highly activated skipped dienes and the
products were obtained in poor yield or as a mixture of isomers. see: T.
Breton, D. Liaigre, E. Belgsir, Tetrahedron Lett. 2005, 46, 2487–2490.
Keywords: alkenes · enones · organocatalysis · oxygenation ·
synthetic methods
[1] a) R. H. Grubbs, A. G. Wenzel, D. J. O’Leary, E. Khosravi, Handbook of
Metathesis, 2nd ed., Wiley-VCH, Weinheim, 2015; b) N. Nishiwaki, Meth-
ods and Applications of Cycloaddition Reactions in Organic Synthesis,
Wiley, Hoboken, 2014; c) A. de Meijere, S. Bräse, M. Oestreich, Metal-
Catalyzed Cross-Coupling Reactions and More, Wiley-VCH, Weinheim,
2014; d) J.-E. Bäckvall, Modern Oxidation Method, 2nd ed., Wiley-VCH,
Weinheim, 2011; e) A. H. E. Müller, K. Matyjaszewski, Controlled and
Manuscript received: June 1, 2017
Accepted manuscript online: June 3, 2017
Version of record online: July 12, 2017
Chem. Eur. J. 2017, 23, 10276 –10279
www.chemeurj.org
10279
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim