Tert-Butyl Hydroperoxide and Tetrabutylammonium Iodide-Promoted Free Radical Cyclization of α-Imino-N-arylamides and α-Azido-N-arylamides

Article abstract of DOI:10.1002/adsc.201500305

The oxidizing system of tert-butyl hydroperoxide (TBHP) and tetrabutylammonium iodide (TBAI) is capable of generating α-(arylaminocarbonyl)iminyl radicals from ethyl 2-(N-arylcarbamoyl)-2-iminoacetates. These iminyl radicals preferably undergo intramolecular ipso attack on the benzene ring to give azaspirocyclohexadienyl radicals, which are readily captured by molecular oxygen under an oxygen atmosphere to yield azaspirocyclohexadienones. In the absence of oxygen, the reaction affords quinoxalin-2-one products. This oxidizing system is also effective to convert α-aryl-α-azido-N-arylamides to the corresponding iminyl radicals under basic conditions (sodium tert-butoxide, t-BuONa), and the subsequent cyclization of these iminyl radicals results in the formation of azaspirocyclohexadienone products in high yields under an oxygen atmosphere. Plausible mechanisms are proposed to rationalize the experimental results, and factors influencing the reactions are discussed.

Full text of DOI:10.1002/adsc.201500305

FULL PAPERS
DOI: 10.1002/adsc.201500305
tert-Butyl Hydroperoxide and Tetrabutylammonium Iodide-Pro-
moted Free Radical Cyclization of a-Imino-N-arylamides and a-
Azido-N-arylamides
Dianjun Li,a Tonghao Yang,a Hailin Su,a and Wei Yua,*
a
State Key Laboratory of Applied Organic Chemistry, College of Chemistry and Chemical Engineering, Lanzhou
University, Lanzhou 730000, Peopleꢀs Republic of China
Fax: +86-931-891-2582; e-mail: yuwei@lzu.edu.cn
Received: March 27, 2015; Revised: April 25, 2015; Published online: July 29, 2015
Abstract: The oxidizing system of tert-butyl hydro- under basic conditions (sodium tert-butoxide, t-
peroxide (TBHP) and tetrabutylammonium iodide BuONa), and the subsequent cyclization of these
(TBAI) is capable of generating a-(arylaminocarbo- iminyl radicals results in the formation of azaspirocy-
nyl)iminyl radicals from ethyl 2-(N-arylcarbamoyl)- clohexadienone products in high yields under an
2-iminoacetates. These iminyl radicals preferably un- oxygen atmosphere. Plausible mechanisms are pro-
dergo intramolecular ipso attack on the benzene ring posed to rationalize the experimental results, and
to give azaspirocyclohexadienyl radicals, which are factors influencing the reactions are discussed.
readily captured by molecular oxygen under an
oxygen atmosphere to yield azaspirocyclohexadi-
enones. In the absence of oxygen, the reaction af- Keywords: (N-arylcarbamyl)-2-iminoacetates; a-
fords quinoxalin-2-one products. This oxidizing azido-N-arylacetamide; tert-butyl hydroperoxide;
system is also effective to convert a-aryl-a-azido-N- iminyl radicals; oxidative cyclization; tetrabutylam-
arylamides to the corresponding iminyl radicals monium iodide
Introduction
carbon to the phenyl radical (formed under standard
tin hydride conditions) and dinitrogen expulsion. The
The radical chemistry of organic azides has proved to authors found that these a-(aminocarbonyl)iminyl
be very useful in organic synthesis.[1] Apart from the radicals exhibit a strong tendency to fragment to ni-
radical azidation methods,[2,3] organic azides can un- triles, and meanwhile minor amounts of quinoxali-
dergo radical reactions of other patterns: by reaction none products were generated as a result of iminyl
with carbon[4] or hetero-centered radicals[5] or reduc- radical cyclization. Later on, Chiba et al. reported an
tion by low valent-metal reductants,[1b] they can also elegant copper-catalyzed aerobic oxidative cyclization
be converted to aminyl radicals; the a-azidyl carbon of a-azido-N-arylamides, which produced azaspirocy-
radicals, on the other hand, can readily extrude a ni- clohexadienones in good yields.[10] In this reaction, the
trogen molecule to afford iminyl radicals.[6–8] Thus, or- copper-mediated iminyl radical is firstly genera-
ganic azides provide a convenient source for other ni- ted,[9f,l1] which then undergoes intramolecular ipso
À
trogen-centered radicals. Nonetheless, despite the im- attack on the benzene ring to form the C N bond.
portance of these nitrogen-centered radicals in the Recently, Zhang et al. reported an N-bromosuccini-
synthesis of nitrogen heterocycles,[9] these significant mide (NBS)-mediated cyclization of 2-azido-N-aryl-
features of azides have not been fully explored.
acetamides.[12] This process also entails the a-(amino-
This is especially the case when one considers the carbonyl)iminyl radical as a key intermediate; its sub-
potential usefulness of azides as the iminyl radical sequent cyclization yields both the quinoxalin-2-one
source. Only recently has it begun to receive attention and spirocyclic products, the ratio of which is largely
from the synthetic point of view. Studies toward this dependent on the electronic nature of substituents at
end were firstly reported by Spagnolo et al.,[6] who in- the benzene ring.
vestigated the reactions of a-(aminocarbonyl)iminyl
It can be seen from these published results that
radicals derived from a-azido-o-iodoanilides.[6b] In while organic azides have proved to be valuable
that study, the generation of the iminyl radical was re- iminyl radical precursors, the related studies are still
alized through 1,5-hydrogen transfer from the a- limited. As far as a-(aminocarbonyl)iminyl radicals
Adv. Synth. Catal. 2015, 357, 2529 – 2539
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2529

FULL PAPERS
Dianjun Li et al.
are concerned, there are several issues that need to
be addressed. Firstly, their reactivity, seemingly vary-
ing from case to case according to the available exam-
ples, should be further explored. As demonstrated by
the aforementioned recent studies, the intramolecular
attack of an iminyl radical center on the N-phenyl
ring can take place at both ipso and ortho positions,
Scheme 2. Formation of 2-(N-arylcarbamoyl)-2-iminoace-
tates
but factors influencing the selectivity have not been which were assigned as a-ethoxycarbonyl-a-azido-N-
clarified. Secondly, the role played by oxygen needs phenylacetamides in this paper, are actually ethyl 2-
to be addressed. It is possible that molecular oxygen (N-arylcarbamoyl)-2-iminoacetates that resulted from
has a big influence on the reaction outcome. Thirdly, the denitrogenation of the former during the prepara-
considering the synthetic usefulness of iminyl radicals, tion (Scheme 2, see the Supporting Information). The
new protocols are highly desirable to enhance the imino compounds are the proposed intermediates to-
general applicability of the azide-based/iminyl radical- wards the cyclization products. Due to these mis-
mediated methodology.
takes,[15] we retracted this paper at the agreement of
During our previous investigations on the oxidative the Editorial Office, and reworked it extensively.
coupling involving 1,3-dicarbonyl compounds, we Herein we wish to present the corrected version of
found that the reagent combination of tert-butyl hy- this report.
droperoxide (TBHP) and tetrabutylammonium iodide
À
(TBAI) could effect the C N oxidative coupling of 2-
aminopyridines with b-keto esters, and the reaction Results and Discussion
afforded imidazo[1,2-a]pyridines in moderate to good
yields.[13] Following this work, we reported that a-sub- At the initial stage of this investigation, we chose
stituted b-acetylamides could be converted to a-keto compound
2-(N-benzyl-N-phenylcarbamoyl)-2-imi-
À
amides via oxidative C C bond cleavage by treatment noacetate (1a) as the model compound, and subjected
with CuCl2/BF3·OEt2/TBHP under an oxygen atmos- it to TBHP under a variety of conditions. The results
phere.[14] On the basis of these results, we hoped that are summarized in Table 1, from which it can be seen
the TBHP-based oxidizing systems might be used to that the reagent combination of TBHP and TBAI is
turn 2-azido-N-phenylacetamides into a-(aminocarbo- capable of oxidizing 1a, whereas using TBHP alone is
nyl)iminyl radicals under metal-free conditions. In ineffective (Table 1, entry 1). When 1a was treated
this way, the iminyl radicals would be generated with 1.4 equiv. of TBAI and 1.9 equiv. of TBHP
under both oxygen and argon atmospheres, and thus (either in water or in decane) in acetonitrile under an
the effect of oxygen could be evaluated.
air atmosphere, 2a and 3a were formed as the major
Our subsequent study demonstrated that the de- products, the ratio of which was dependent on the re-
sired transformations can be really achieved by using action temperature (entries 2–4, and 7). The yield of
the oxidizing system of TBHP/TBAI, and the results 2a could be increased when the reaction was carried
were revealed in a recent paper (Scheme 1).[15] How- out under an O2 atmosphere (O2 balloon) (entries 6, 9
ever, we soon found that one group of the reactants, and 10). On the other hand, when the reaction was
carried out under an argon atmosphere, only a tiny
amount of 2a was obtained, but the yield of 3a was
still moderate (entries 5 and 8). The reaction was very
slow when only a catalytic amount of TBAI was used
(entry 11), and it did not happen in the absence of
TBHP (entry 15). Using H2O2 as oxidant in place of
TBHP failed to effect the reaction (entry 12). On the
other hand, when TBAI was replaced by molecular
iodine, a complex mixture was formed (entry 13). The
reaction exhibited a large solvent effect: it did not
take place in toluene (entry 14), but the yield of 2a
was remarkably improved in 1,2-dichloroethane
(DCE) under an O2 atmosphere (entries 19 and 20).
It is interesting to see that with DCE as the solvent,
2a was formed in roughly the same yield as 3a even
under an argon atmosphere (entry 16).
These results can be interpreted with the mecha-
nism shown in Scheme 3. The reaction is initiated by
the oxidation of 1a to give iminyl radical A. The
Scheme 1. Results revealed in ref.[15]
2530
asc.wiley-vch.de
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 2529 – 2539

FULL PAPERS
tert-Butyl Hydroperoxide and Tetrabutylammonium Iodide-Promoted Free Radical
Table 1. Screening of reaction conditions for the reaction of 1.[a]
Entry Oxidant (equiv.) Additive (equiv.) Atmosphere Solvent (Temp.) Reaction time [h] Product (Yield [%])[b]
1
2
3
4
5
6
7
8
TBHP[c] (1.9)
TBHP[c] (1.9)
TBHP[c] (1.9)
TBHP[c] (1.9)
TBHP[c] (1.9)
TBHP[c] (1.9)
TBHP[d] (1.9)
TBHP[d] (1.9)
TBHP[d] (1.9)
TBHP[d] (1.9)
TBHP[d] (1.9)
none
air
air
air
air
Ar
O2
air
Ar
O2
O2
O2
O2
O2
O2
air
Ar
air
air
O2
O2
O2
CH3CN (1008C) 2.5
CH3CN (1008C) 2.5
CH3CN (808C) 2.5
N.R.[e]
TBAI (1.4)
TBAI (1.4)
TBAI (1.4)
TBAI (1.4)
TBAI (1.4)
TBAI (1.4)
TBAI (1.4)
TBAI (1.4)
TBAI (0.9)
TBAI (0.2)
2a (28), 3a (33)
2a (41), 3a (25)
2a (14), 3a (40)
2a (3), 3a (46)+mixture
2a (56), 3a (20)
2a (47), 3a (21)
2a (trace), 3a (38)+mixture
2a (66), 3a (15)
2a (65), 3a (13)
2a (14), 3a (5)[f]
N.R.[e]
CH3CN (508C)
4
CH3CN (1008C) 3
CH3CN (808C) 2.5
CH3CN (808C)
CH3CN (808C)
3
3
9
CH3CN (808C) 2.5
CH3CN (808C) 2.5
10
11
12
13
14
15
16
17
18
19
20
21
CH3CN (808C)
CH3CN (808C)
CH3CN (808C)
toluene (758C)
CH3CN (808C)
DCE (808C)
DCE (r.t.)
DCE (808C)
DCE (808C)
DCE (808C)
DCE (808C)
3
4
4
4
H2O2 (30%) (1.9) TBAI (1.4)
TBHP[d] (1.9)
TBHP[d] (1.9)
none
I2 (1.4)
mixture
TBAI (1.4)
TBAI (1.4)
TBAI (1.4)
TBAI (1.4)
TBAI (1.4)
TBAI (1.2)
TBAI (0.9)
TBAI (1.2)
N.R.[e]
6
N.R.[e]
TBHP[d] (1.9)
TBHP[d] (1.9)
TBHP[d] (1.9)
TBHP[d] (1.9)
TBHP[d] (1.9)
TBHP[d] (1.1)
2.5
24
2.5
1.5
1.5
2
2a (42), 3a (40)
2a (11), 3a (trace)[g]
2a (52), 3a (32)
2a (85), 3a (trace)
2a (83), 3a (trace)
2a (67), 3a (trace)
[a]
[b]
[c]
[d]
[e]
[f]
The reaction was performed in a solution of 0.1M 1a.
Isolated yield.
70% in water.
Ca. 5.5M in decane.
No reaction.
70% starting material recovered.
75% starting material recovered.
[g]
latter undergoes facile ipso-5-exo cyclization to afford yield of quinoxalin-2-one product is quite low in their
radical B. Under an oxygen atmosphere, B is captured case. This tendency of fragmentation has also been
by oxygen to yield peroxy radical C, from which 2a is demonstrated by Chibaꢀs work on the copper-cata-
generated. This reaction pattern is consistent with the lyzed reaction of a-azido carbonyl compounds.[16] In
previously reported studies.[10,11c,12] When the reaction the present case, the fragmentation of radical A is
proceeds under an argon atmosphere, radical B un- much less favored compared with the intermolecular
dergoes 1,2-migration of the iminyl moiety to yield ipso attack, and thus A was transformed to B prefer-
radical D, or ring opening to give back iminyl radical entially. In an environment saturated with oxygen, B
A, and from D quinoxalin-2-one 3a is eventually would be captured rapidly by oxygen to give eventu-
formed.
ally 2a in high yield. However, in the absence of
Two possible pathways are available for the forma- oxygen, the transformation from B to A becomes re-
tion of radical D: the first is the intramolecular ortho- versible, and the fragmentation will have a chance to
attacking cyclization of iminyl radical A (path a, compete with other pathways. Therefore, when the re-
Scheme 3); the second involves the 1,2-migration of action was carried out under argon atmosphere, the
the iminyl group from B (path b, Scheme 3). Under yield of 3a was considerably lower than that of 2a in
both circumstances, B is formed in the first place. the presence of oxygen.
That 2a was formed predominantly under an oxygen
The transformation from A to 3a through D is a typ-
ical homolytic aromatic substitution (HAS) reac-
atmosphere clearly proves this point.
Previous work by Spagnolo et al. demonstrates that tion.[17] For the formation of 3a, a hydrogen atom
a-(phenylaminocarbonyl)iminyl radicals like A are in- must be removed from D, a process that might be re-
clined to fragment to nitriles.[6b] That explains why the alized via oxidation followed by deprotonation (path
Adv. Synth. Catal. 2015, 357, 2529 – 2539
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2531

FULL PAPERS
Dianjun Li et al.
The oxidizing system of TBHP and TBAI has re-
cently been extensively applied to various types of ox-
idative transformations.[19] From the mechanistic point
of view, its oxidizing capacities generally fall into two
categories: the first involves the in situ generated tert-
butoxy radical; the second involves the hypervalent
iodine
sÀpe[2c0i]es
[(n-Bu)4N]+IOÀ
and
[(n-
Bu)4N]+IO2 . In the present cases, we assume that
these hypervalent iodines are less likely to be the
active oxidants since the combination of H2O2 (30%)
and TBAI, which is also capable of generating hyper-
valent iodine species, is ineffective here (Table 1,
entry 12).
Consequently, a mechanism is proposed to account
for the formation of iminyl radical A from 1a
(Scheme 4). We believe that the tert-butoxy radical is
the active oxidant in the present reaction system. Be-
sides the tert-butoxy radical, the tert-butyl peroxyl
radical probably also acts as the oxidant since it can
be generated from tert-butoxy radical and TBHP.[21]
That 2a was generated together with 3a in DCE even
under an argon atmosphere supports this point: in
situ generation of O2 probably occurs via dimerization
À
of the tert-butyl peroxyl radical followed by O O
cleavage. The oxidation takes place probably via
a single electron transfer between 1a and tert-butoxyl
radical (or tert-butyl peroxyl radical).
Scheme 3. Proposed mechanism for the TBHP/TBAI-pro-
moted reaction of 1a.
To explore the scope of the reaction with regard to
c, Scheme 3). Very recently, Studer and Curran pro- the substituent effect, the optimal conditions for the
posed a different scenario to rationalize the homolytic formation of 2a (Table 1, entry 19) were applied to
aromatic substitution under basic conditions. Their a variety of substituted a-imino-N-arylacetamides 1,
study indicates that the cyclohexadienyl radicals like and the results are illustrated in Scheme 5, Scheme 6
D formed in the HAS reactions are quite acidic, and and Scheme 7. For the reactions of ortho- and meta-
can be readily removed with a base.[18] According to substituted substrates 1, azaspirocyclohexadienones 2
this base-promoted homolytic aromatic substitution were generated predominantly, the yields of which
(BHAS) mechanism, it is possible that the conversion varied according to the substituents. Besides the sub-
of D to 3a firstly involves a deprotonation step to stituent at the phenyl ring, that at the amido nitrogen
afford radical anion E; the latter is then oxidized to atom also influenced the yield of 2. As such, the yield
3a (path d, Scheme 3). As can be seen in Scheme 4, of 2l was considerably lower than that for its N-
the reaction of TBHP and iodide ion produces hy- benzyl
or
N-phenyl
substituted
counterpart
droxide ion, which can promote the deprotonation (Scheme 5).
step.
When substrates incorporating a substituent at the
para position were subjected to the current condi-
tions, mixed results were obtained after reaction: for
the reaction of 1m–1o, the major product was 2a, ac-
companied by a minor amount of 3m–3o (Scheme 6).
In the cases of 1m and 1n, a tiny amount of di-tert-bu-
tylperoxy-substituted azaspirocyclohexadienyl product
4 was also obtained. On the other hand, when the
substrate was 1p or 1q, the reaction afforded a mixture
of
mono-tert-butylperoxy-attached
azaspirocyclic
product 4’ and quinoxalin-2-one 3 (Scheme 7) (the
relative configuration of 4’ was undetermined).
The formation of compound 4 and 4’ can be ac-
counted for with the mechanisms shown in Scheme 8:
In path (a), the azaspirocyclic radical B derived from
the corresponding substrate is transformed to perox-
Scheme 4. Proposed mechanism for the TBHP/TBAI-medi-
ated oxidation of 1a to iminyl radical A.
2532
asc.wiley-vch.de
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 2529 – 2539

FULL PAPERS
tert-Butyl Hydroperoxide and Tetrabutylammonium Iodide-Promoted Free Radical
Scheme 6. Reactions of para-substituted 1m–1o.
Scheme 7. Reactions of para-substituted 1p and 1q.
capable of capturing intermediate B (persistent radi-
cal effect) to give 4’ [mechanism (b)]. Similar cou-
pling reactions as steered by the persistent radical
effect have been well documented in the literature.[23]
When compound 1r was treated with TBHP and
TBAI under an oxygen atmosphere, 3r was generated
in 76% yield (Scheme 9). This result is consistent with
the anticipation that the intramolecular ipso-attack in
Scheme 5. TBHP and TBAI-mediated oxidative cyclization
of 1 under oxygen atmosphere.
ide F under an oxygen atmosphere, which subsequent- this particular case was difficult due to the unfavor-
ly undergoes elimination of HOOÀ to afford carbocat- able conformational effect posed by the tetrahydro-
ion G. The latter is captured by t-BuOOÀ to yield 4’. quinoline ring, while the formation of a D-like radical
When R is a halogen atom, a competitive elimination became favorable.
from peroxide F and 4’ takes place, and thus 2a is gen-
As indicated in Table 1, apart from 2a, quinoxalin-
erated. In the cases of 1m and 1n, a tiny amount of 4’ 2-one 3a could also form from the oxidation of 1a,
undergoes nucleophilic substitution with TBHP to and compound 3a was obtained almost exclusively
afford compound 4. Similar elimination products have when the reaction was carried out in CH3CN under
been reported by Zhang et al.[12] As an alternative to an argon atmosphere (Table 1, entry 5). To further ex-
this mechanism, 4’ can result from the trapping of B plore the influence of the substituent, we next applied
by tert-butyl peroxyl radical. The tert-butyl peroxyl these conditions to several substituted substrates 1,
radical has a rather long lifetime,[22] and therefore is and the results are illustrated in Table 2. Here again
Adv. Synth. Catal. 2015, 357, 2529 – 2539
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2533

FULL PAPERS
Dianjun Li et al.
Scheme 8. Proposed mechanism for the formation of 2a, 4 and 4’ in the reactions of 1m–1q.
Scheme 9. Reaction of 1r under an oxygen atmosphere.
the position of the substituent on the N-phenyl ring
exerted a huge influence on the composition of the
products, which varied depending on the substitution
patterns. For the reactions of compound 1d and 1f,
the major products were the corresponding quinoxa-
lin-2-ones 3d and 3f as a mixture of regio-isomers,
along with a minor amount of 2d and 2f. The quinox-
alin-2-one-forming cyclization exhibited ortho selec-
tivity in both cases, which is consistent with the previ-
ous reports dealing with attack of carbon radicals on Scheme 10. Proposed mechanism for the formation of 2
the substituted benzene ring.[24]
under an argon atmosphere.
In contrast to this result, when compound 1b and 1j
were used, only 2b and 2j were obtained after reac-
tion, indicating that the ortho-attack is hampered by
the presence of an ortho-substituent. The formation
of 2b and 2j can be accounted for with the mechanism
shown in Scheme 10. On the other hand, the reaction
of para-substituted 1p and 1q delivered a mixture of 3
and 4’, while the reaction of 1m–1o produced 2a as
well as 3. All these results are in accordance with the
mechanism shown in Scheme 3 and Scheme 8.
Scheme 11. The result with 5 as the substrate.
Although this protocol is effective for the reaction
of compounds 1, it failed to produce similar results
when the ethyloxycarbonyl group was replaced by an
acetyl group (compound 5). In the latter case, the re-
We next examined the applicability of this TBHP/
action delivered compound
6
in 34% yield TBAI system to the reaction of a-aryl-a-azido-N-aryl-
(Scheme 11). The generation of 6 is consistent with amides 7. As demonstrated by Chiba et al.,[10] a-aryl-
the notion that iminyl radicals are susceptible to frag- a-azido-N-arylamides can be transformed to the cor-
mentation, and at the same time reflects the strong in- responding iminyl radicals through the ketimine inter-
fluence of the a-substituent on the reactivity of iminyl mediates under copper-mediated oxidative conditions.
radicals.
We anticipated that the current metal-free protocols
2534
asc.wiley-vch.de
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 2529 – 2539

FULL PAPERS
tert-Butyl Hydroperoxide and Tetrabutylammonium Iodide-Promoted Free Radical
Table 2. Reaction of 1 under an argon atmosphere.[a]
Adv. Synth. Catal. 2015, 357, 2529 – 2539
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2535

FULL PAPERS
Dianjun Li et al.
Table 2. (Continued)
[a]
The reaction was performed on a 0.5 mmol scale in acetonitrile under an argon atmosphere.
Isolated yield.
The relative configuration is undetermined.
[b]
[c]
Table 3. The reaction of 7 under modified conditions.[a]
Entry
7
R1
R2
R3
Product/Yield [%][b]
1
2
3
4
5
6
7
8
9
10
7a
7b
7c
7d
7e
7f
7g
7h
7i
H
Bn
Bn
Bn
Bn
Me
Bn
Bn
Bn
Bn
Bn
H
H
H
H
8a/99
8a/80
8a/96
8d/60
8e/45
8f/72
8g/67
8h/93
8i/43
8j/6[c]
p-Cl
p-MeO
o-Me
H
H
H
H
H
H
H
m-Br
p-Me
p-Cl
p-F
o-Me
7j
[a]
The reaction was performed on a 0.5 mmol scale.
Isolated yield.
With 90% 7j recovered.
[b]
[c]
would also be effective for the reaction of compounds cule almost instantaneously, and after work-up com-
7. However, our initial experiment showed that no re- pound 10 can be obtained in roughly 80% yield
action took place when 7 was subjected to the previ- (Scheme 13). Apparently, the oxidative cyclization of
ously mentioned optimal conditions. We assumed that 7 firstly involves the generation of an imine inter-
for the reaction of 7 to take place, the a-hydrogen mediate, which is then oxidized to the corresponding
had to be deprotonated first to generate the imine in- iminyl radical.
termediates. Indeed, when 3.0 equiv. of t-BuONa
It is interesting to see from Table 3 that the p-MeO
were added to the reaction mixture, compounds 7 group was as readily removed as the chlorine atom,
were converted smoothly to the azaspirocyclohexadi- and 8a was formed exclusively in both cases. This dis-
enones 8 underan oxygen atmosphere in DCE. The crepancy with the reaction outcome of 1q suggests
yields were high except for 7e, 7i and 7j (Table 3). that the a-phenyl group is detrimental to the forma-
The lower yield of 8e is consistent with the aforemen-
tioned observation that the N-methyl group is inferior
to the N-benzyl group to promote the cyclization.
These conditions were also applied to compound 9,
but the reaction did not occur (Scheme 12). Probably,
the a-H in 9 is not acidic enough for the deprotona-
tion to take place under the indicated conditions.
Our control experiment shows that when treated
with t-BuONa alone, 7a will extrude a nitrogen mole- fied conditions
Scheme 12. The result with 9 as the substrate under modi-
2536
asc.wiley-vch.de
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 2529 – 2539

FULL PAPERS
tert-Butyl Hydroperoxide and Tetrabutylammonium Iodide-Promoted Free Radical
with regard to the substituent effect and the influence
of oxygen. Our results indicate that the intramolecu-
lar ipso-attack on the N-aryl ring by the iminyl radical
is generally more favorable than the ortho-attack, and
correspondingly in most cases affords azaspirocyclo-
hexadienones as the dominant products under an
oxygen atmosphere. By contrast, when the reaction
was carried out under an argon atmosphere, the
ortho-attack products, quinoxalin-2-ones, could be
generated as the major product. The substituents
have a big influence on the reaction outcome. This
oxidizing system is applicable to the reaction of a-
phenyl a-azido-N-arylamides by adding 3.0 equiv. of t-
BuONa into the reaction mixture, but in these cases,
the formation of the quinoxalin-2-one product be-
comes very unfavorable. This study not only sheds
light on the reactivity of a-(arylaminocarbonyl)iminyl
radicals, but also provides a new metal-free protocol
for the synthesis of the azaspirocyclohexadienone de-
rivatives.
Scheme 13. Evidence for formation of the iminyl radicals
from 7 via imine intermediates.
Scheme 14. Reaction of 7k under an oxygen atmosphere.
Experimental Section
Typical Procedure for the Reaction of 1 under an O2
Atmosphere
To a 25-mL sealed tube equipped with a magnetic stirring
bar were added 1a (168 mg, 0.54 mmol), tetrabutylammoni-
um iodide (TBAI, 240 mg, 1.2 equiv.), tert-butyl hydroperox-
ide (TBHP, ~5.5M in decane) (185 mL, 1.9 equiv.) and 1,2-
dichloroethane (5.0 mL). The solution was stirred in an oil
bath at 808C under an oxygen atmosphere (with an oxygen
balloon). After the reaction was complete as indicated by
TLC (1.5 h), the reaction mixture was poured into a saturat-
ed aqueous NaHSO3 solution (15 mL), and was extracted
with EtOAc (10 mL”3). The combined organic layers were
washed with brine (30 mL) and dried with anhydrous
Na2SO4. The solvent was removed under reduced pressure,
and the residue was treated by silica gel chromatography to
give product 2a; yield: 149 mg (85%).
Scheme 15. Reaction of 7a in CH3CN under an argon atmos-
phere.
tion of the quinoxalin-2-one ring, and meanwhile the
strong basic conditions in the present cases are favor-
able for the removal of the p-MeO group. This hy-
pothesis is supported by the fact that the cyclization
of compound 7k gave quinoxalin-2-one 11k in only
22% yield (Scheme 14), which is much lower than
that of the structurally analogous 1r under similar
conditions (Scheme 9). Moreover, when compound 7a
was subjected to TBHP/TBAI in acetonitrile under an
argon atmosphere, no cyclization product was ob-
tained; the reaction only resulted in the decomposi-
tion of 7a (Scheme 15).
General Procedure for the Reaction of 1 under an
Argon Atmosphere
To a 25-mL sealed tube equipped with a magnetic stirring
bar were added 1 (0.5 mmol, 1.0 equiv.), TBAI (277 mg,
1.5 equiv.), TBHP (70% in water) (140 mL, 2.0 equiv.) and
5.0 mL of acetonitrile (bubbled with argon for 10 min.
before use). The solution was stirred in an oil bath at 1008C
under an argon atmosphere. After the reaction was com-
plete as indicated by TLC (generally 2.5–3 h), the reaction
mixture was poured into a saturated aqueous NaHSO3 solu-
tion (15 mL), and was extracted with EtOAc (10 mL”3).
The combined organic layers were washed with brine
(30 mL) and dried with anhydrous Na2SO4. The solvent was
removed under reduced pressure, and the residual was treat-
Conclusions
In summary, we have demonstrated that the reagent
combination of TBHP/TBAI is capable of oxidizing
(N-aryl-carbamoyl)-2-iminoacetates to the corre-
sponding a-(arylaminocarbonyl)iminyl radicals. With
the help of this new method, the cyclization of a-
(arylaminocarbonyl)iminyl radicals was investigated ed byh silica gel chromatography to give the products.
Adv. Synth. Catal. 2015, 357, 2529 – 2539
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2537

FULL PAPERS
Dianjun Li et al.
[5] a) L. Benati, G. Bencivenni, R. Leardini, M. Minozzi,
D. Nanni, R. Scialpi, P. Spagnolo, G. Zanardi, Org.
Lett. 2006, 8, 2499–2502; b) L. Benati, G. Bencivenni,
R. Leardini, M. Minozzi, D. Nanni, R. Scialpi, P. Spag-
nolo, G. Zanardi, J. Org. Chem. 2006, 71, 5822–5825;
c) H. Zhai, M. Zlotorzynska, G. Sammis, Chem.
Commun. 2009, 45, 5716–5718; d) G. Bencivenni, T.
Lanza, M. Minozzi, D. Nanni, P. Spagnolo, G. Zanardi,
Org. Biomol. Chem. 2010, 8, 3444–3450.
[6] a) P. C. Montevecchi, M. L. Navacchia, P. Spagnolo, J.
Org. Chem. 1997, 62, 5846–5848; b) G. Bencivenni, T.
Lanza, R. Leardini, M. Minozzi, D. Nanni, P. Spagnolo,
G. Zanardi, J. Org. Chem. 2008, 73, 4721–4724; c) T.
Lanza, R. Leardini, M. Minozzi, D. Nanni, P. Spagnolo,
G. Zanardi, Angew. Chem. 2008, 120, 9581–9584;
Angew. Chem. Int. Ed. 2008, 47, 9439–9442; d) T.
Lanza, M. Minozzi, A. Monesi, D. Nanni, P. Spagnolo,
G. Zanardi, Adv. Synth. Catal. 2010, 352, 2275–2280.
[7] S. Muthukrishnan, J. Sankaranarayanan, R. F. Klima,
T. C. S. Pace, C. Bohne, A. D. Gudmundsdottir, Org.
Lett. 2009, 11, 2345–2348.
General Procedure for the Reaction of 7 under an O2
Atmosphere
To a 25-mL sealed tube equipped with a magnetic stirring
bar were added 7 (0.5 mmol, 1.0 equiv.), tetrabutylammoni-
um iodide (TBAI, 277 mg, 1.5 equiv.), tert-butyl hydroperox-
ide (TBHP, ~5.5M in decane) (185 mL, 2.0 equiv.), sodium
tert-butoxide (t-BuONa, 145 mg, 3.0 equiv.) and DCE
(5.0 mL). The solution was stirred in an oil bath at 808C
under an oxygen atmosphere. After the reaction was com-
plete as indicated by TLC (generally 1–1.5 h), the reaction
mixture was poured into a saturated aqueous NaHSO3 solu-
tion (15 mL), and was extracted with EtOAc (10 mL”3).
The combined organic layers were washed with brine
(30 mL) and dried with anhydrous Na2SO4. The solvent was
removed under reduced pressure, and the residual was treat-
ed by silica gel chromatography to give products 8.
Caution: tert-butyl hydroperoxide is hazardous and flam-
mable. When used in the presence of oxygen it should be
handled with care and proper protections.
[8] Y.-F. Wang, K. K. Toh, E. P. J. Ng, S. Chiba, J. Am.
Chem. Soc. 2011, 133, 6411–6421.
Acknowledgements
[9] a) A. G. Fallis, I. M. Brinza, Tetrahedron 1997, 53,
17543–17594; b) L. Stella, Nitrogen-centered radicals,
in: Radicals in Organic Synthesis, (Eds.: P. Renaud,
M. P. Sibi), John Wiley & Sons, Chichester, 2001,
pp 407–426; c) S. Z. Zard, Chem. Soc. Rev. 2008, 37,
1603–1618; d) M. Kitamura, K. Narasaka, Bull. Chem.
Soc. Jpn. 2008, 81, 539–547; e) X. Xu, X. Wan, Y.
Geng, J. Zhang, H. Xu, Chin. J. Org. Chem. 2011, 81,
453–465; f) S. Chiba, Bull. Chem. Soc. Jpn. 2013, 86,
1400–1411.
The authors thank the National Natural Science Foundation
of China (No. 21372108) for financial support.
References
[1] a) M. Minozzi, D. Nanni, P. Spagnolo, Chem. Eur. J.
2009, 15, 7830–7840; b) C. Jimeno, P. Renaud, Radical
Chemistry with Azides, in: Organic Azides: Syntheses
and Applications, (Eds.: S. Bräse, K. Banert) John
Wiley & Sons, Chichester, 2010, pp 239–267; c) D.
Nanni, P. Spagnolo, Unusual Radical Acceptors, in: En-
cyclopedia of Radicals in Chemistry, Biology and Mate-
rials, (Eds.: C. Chatgilialoglu, A. Studer), John Wiley &
Sons, Chichester, 2012, Vol. 2, pp 1019–1058.
[2] For free radical azide transfer reactions, see: a) V. V.
Zhdankin, A. P. Krasutsky, C. J. Kuehl, A. J. Simonsen,
J. K. Woodward, B. Mismash, J. T. Bolz, J. Am. Chem.
Soc. 1996, 118, 5192–5197; b) P. Panchaud, L. Chabaud,
Y. Landais, C. Ollivier, P. Renaud, S. Zigmantas, Chem.
Eur. J. 2004, 10, 3606–3614; c) G. Lapointe, A. Kapat,
K. Weidner, P. Renaud, Pure Appl. Chem. 2012, 84,
1633–1641.
[3] For recent examples dealing with azidyl radicals, see:
a) K. Matcha, R. Narayan, A. P. Antonchick, Angew.
Chem. 2013, 125, 8143–8147; Angew. Chem. Int. Ed.
2013, 52, 7985–7989; b) X.-H. Wei, Y.-M. Li, A.-X.
Zhou, T.-T. Yang, S.-D. Yang, Org. Lett. 2013, 15, 4158–
4161; c) Y. Yuan, T. Shen, K. Wang, N. Jiao, Chem.
Asian J. 2013, 8, 2932–2935; d) B. Zhang, A. Studer,
Org. Lett. 2013, 15, 4548–4551.
[4] a) S. Kim, G. H. Joe, J. Y. Do, J. Am. Chem. Soc. 1994,
116, 5521–5522; b) B. Patro, J. A. Murphy, Org. Lett.
2000, 2, 3599–3601; c) D. Lizos, R. Tripoli, J. A.
Murphy, Chem. Commun. 2001, 37, 2732–2733; d) S.
Zhou, S. Bommezijn, J. A. Murphy, Org. Lett. 2002, 4,
443–445.
[10] S. Chiba, L. Zhang, J.-Y. Lee, J. Am. Chem. Soc. 2010,
132, 7266–7267.
[11] For recent synthetic studies in Chibaꢀ group concerning
the Cu-iminyl radicals, see: a) L. Zhang, G. Y. Ang, S.
Chiba, Org. Lett. 2010, 12, 3682–3685; b) L. Zhang,
G. Y. Ang, S. Chiba, Org. Lett. 2011, 13, 1622–1625;
c) Y. L. Tnay, C. Chen, Y. Y. Chua, L. Zhang, S. Chiba,
Org. Lett. 2012, 14, 3550–3553.
[12] Z.-S. Li, W.-X. Wang, J.-D. Yang, Y.-W. Wu, W. Zhang,
Org. Lett. 2013, 15, 3820–3823.
[13] L. Ma, X. Wang, W. Yu, B. Han, Chem. Commun.
2011, 47, 11333–11335.
[14] D. Li, W. Yu, Adv. Synth. Catal. 2013, 355, 3708–3714.
[15] D. Li, T. Yang, H. Su, W. Yu, Adv. Synth. Catal. 2014,
356, 3148–3156. This paper has been retracted at the
agreement of the Editorial Office and John Wiley &
Sons, Ltd., see: Adv. Synth. Catal. 2015, 357, 601. The
main reason why we wrongly assigned the structures of
1 was that, in MS measurement of the first batch of
compounds, compound 1p showed a molecular weight
of 370.1863 (calcd. for C19H20N4O3 +NH4: 370.1874), in-
dicating the existence of N3 group. Later on, this mea-
surement was proved to be wrong. We let the sample
of the same molecule be measured two more times,
and both datasets showed the loss of N2. The convinc-
ing evidence for the structural difference between
1 and 7 came from the FT-IR data (see the Supporting
Information). In the FT-IR spectra of 7, a strong sharp
peak appears at about 2100 cmÀ1, which is caused by
2538
asc.wiley-vch.de
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2015, 357, 2529 – 2539

FULL PAPERS
tert-Butyl Hydroperoxide and Tetrabutylammonium Iodide-Promoted Free Radical
the absorption of the N3 group, whereas in the FT-IR
12709; Angew. Chem. Int. Ed. 2012, 51, 12538–12541;
j) L.-T. Li, J. Huang, H.-Y. Li, L.-J. Wen, P. Wang, B.
Wang, Chem. Commun. 2012, 48, 5187–5189; k) Z.-Q.
Lao, W.-H. Zhong, Q.-H. Lou, Z.-J. Li, X.-B. Meng,
Org. Biomol. Chem. 2012, 10, 7869–7871; l) X. Li, X.
Xu, C. Zhou, Chem. Commun. 2012, 48, 12240–12242;
m) Q. Xue, J. Xie, H. Li, Y. Cheng, C. Zhu, Chem.
Commun. 2013, 49, 3700–3702; n) X. Li, X. Xu, Y.
Tang, Org. Biomol. Chem. 2013, 11, 1739–1742; o) Q.
Xue, J. Xie, P. Xu, K. Hu, Y. Cheng, C. Zhu, ACS
Catal. 2013, 3, 1365–1368; p) J. Zhang, J. Jiang, Y. Li,
X. Wan, J. Org. Chem. 2013, 78, 11366–11372; q) D.
Zhao, T. Wang, Q. Shen, J.-X. Li, Chem. Commun.
2014, 50, 4302–4304; r) X.-F. Wu, J.-L. Gong, X. Qi,
Org. Biomol. Chem. 2014, 12, 5807–5817.
spectra of 1, this peak is absent. In the latter cases, an
À
absorption corresponding to an N H stretch vibration
appears at about 3200 cmÀ1. We apologize for these ter-
rible mistakes in the previous paper caused by our neg-
ligence.
[16] S. Chiba, L. Zhang, G. Y. Ang, B. W.-Q. Hui, Org. Lett.
2010, 12, 2052–2055.
[17] W. R. Bowman, J. M. D. Storey, Chem. Soc. Rev. 2007,
36, 1803–1822.
[18] a) A. Studer, D. P. Curran, Angew. Chem. 2011, 123,
5122–5127; Angew. Chem. Int. Ed. 2011, 50, 5018–5022;
b) A. Studer, D. P. Curran, Nat. Chem. 2014, 6, 765–
773.
[19] For representative examples, see: a) M. Uyanik, H.
Okamoto, T. Yasui, K. Ishihara, Science 2010, 328,
1376–1379; b) M. Uyanik, D. Suzuki, T. Yasui, K. Ishi-
hara, Angew. Chem. 2011, 123, 5443–5446; Angew.
Chem. Int. Ed. 2011, 50, 5331–5334; c) L. Chen, E. Shi,
Z. Liu, S. Chen, W. Wei, H. Li, K. Xu, X. Wan, Chem.
Eur. J. 2011, 17, 4085–4089; d) T. Froehr, C. P. Sindlin-
ger, U. Kloeckner, P. Finkbeiner, B. J. Nachtsheim, Org.
Lett. 2011, 13, 3754–3757; e) J. Xie, H. Jiang, Y. Cheng,
C. Zhu, Chem. Commun. 2012, 48, 979–981; f) Z. Liu, J.
Zhang, S. Chen, E. Shi, Y. Xu, X. Wan, Angew. Chem.
2012, 124, 3285–3289; Angew. Chem. Int. Ed. 2012, 51,
3231–3235; g) E. Shi, Y. Shao, S. Chen, H. Hu, Z. Liu,
J. Zhang, X. Wan, Org. Lett. 2012, 14, 3384–3387; h) W.
Wei, Y. Shao, H. Hu, F. Zhang, C. Zhang, Y. Xu, X.
Wan, J. Org. Chem. 2012, 77, 7157–7165; i) B. Tan, N.
Toda, C. F. Barbas III, Angew. Chem. 2012, 124, 12706–
[20] M. Uyanik, K. Ishihara, ChemCatChem 2012, 4, 177–
185.
[21] C. Walling, L. Heaton, J. Am. Chem. Soc. 1965, 87, 38–
47.
[22] a) A. Bravo, H.-R. Bjørsvik, F. Fontana, L. Liguori, F.
Minisci, J. Org. Chem. 1997, 62, 3849–3857; b) R.
Shchepin, M. N. Mçller, H. H. Kim, D. M. Hatch, S.
Bartesaghi, B. Kalyanaraman, R. Radi, N. A. Porter, J.
Am. Chem. Soc. 2010, 132, 17490–17500.
[23] a) A. Studer, Chem. Eur. J. 2001, 7, 1159–1164; b) A.
Studer, Chem. Soc. Rev. 2004, 33, 267–273.
[24] For recent examples, see: a) X. Ju, Y. Liang, P. Jia, W.
Li, W. Yu, Org. Biomol. Chem. 2012, 10, 498–501;
b) L.-P. B. Beaulieu, D. S. Roman, F. VallØe, A. B. Char-
ette, Chem. Commun. 2012, 48, 8249–8251.
Adv. Synth. Catal. 2015, 357, 2529 – 2539
ꢁ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2539