Electrochemically induced cascade reaction for the assembly of libraries of biologically relevant 1,4-benzoxazine derivatives

Article abstract of DOI:10.1021/jo060452f

The scope and mechanism of an electrochemically induced cascade reaction, which leads to highly substituted 1,4-benzoxazine derivatives, have been explored through the variation of the structure of the o-azaquinone mediator. This reaction sequence, wherein both cycloaddition partners are generated in situ, at room temperature, under metal-free conditions, allows the regiospecific inverse-electron-demand Diels-Alder (IEDDA) reaction of an o-azaquinone heterodiene and a secondary alkylenamine dienophile, two chemically nonaccessible unstable entities. The cascade reaction was found to be general with electron-poor o-azaquinone entities generated from substituted 2-aminoresorcinol substrates. In the case of o-aminophenol derivatives which lack the 2-hydroxyl group, the generated o-azaquinone species failed to catalyze the oxidation of the amine to the corresponding imine, precursor of the enamine dienophile, because the absence of an intramolecular hydrogen bond at the origin of a highly reactive Schiff base cyclic transition state. To overcome this problem, a tandem oxidation-IEDDA reaction, in which the o-azaquinone is generated in the presence of a preformed enamine, has been developed as an alternative. These one-pot methodologies, which offer the opportunity to introduce variations in both cycloaddition partners, should be particularly useful for the development of libraries of biologically relevant 1,4-benzoxazine derivatives.

Full text of DOI:10.1021/jo060452f

Electrochemically Induced Cascade Reaction for the Assembly of
Libraries of Biologically Relevant 1,4-Benzoxazine Derivatives
Daiwang Xu,† Ange`le Chiaroni,‡ Maurice-Bernard Fleury,† and Martine Largeron*,†
UMR 8638 CNRS-UniVersite´ Paris 5, Synthe`se et Structure de Mole´cules d’Inte´reˆt Pharmacologique,
Faculte´ des Sciences Pharmaceutiques et Biologiques, 4 AVenue de l’ObserVatoire,
75270 Paris Cedex 06, France, and Institut de Chimie des Substances Naturelles, Centre National de la
Recherche Scientifique, 1 AVenue de la Terrasse, 91198 Gif-sur-YVette Cedex, France
martine.largeron@uniV-paris5.fr.
ReceiVed March 2, 2006
The scope and mechanism of an electrochemically induced cascade reaction, which leads to highly
substituted 1,4-benzoxazine derivatives, have been explored through the variation of the structure of the
o-azaquinone mediator. This reaction sequence, wherein both cycloaddition partners are generated in
situ, at room temperature, under metal-free conditions, allows the regiospecific inverse-electron-demand
Diels-Alder (IEDDA) reaction of an o-azaquinone heterodiene and a secondary alkylenamine dienophile,
two chemically nonaccessible unstable entities. The cascade reaction was found to be general with electron-
poor o-azaquinone entities generated from substituted 2-aminoresorcinol substrates. In the case of
o-aminophenol derivatives which lack the 2-hydroxyl group, the generated o-azaquinone species failed
to catalyze the oxidation of the amine to the corresponding imine, precursor of the enamine dienophile,
because the absence of an intramolecular hydrogen bond at the origin of a highly reactive Schiff base
cyclic transition state. To overcome this problem, a tandem oxidation-IEDDA reaction, in which the
o-azaquinone is generated in the presence of a preformed enamine, has been developed as an alternative.
These one-pot methodologies, which offer the opportunity to introduce variations in both cycloaddition
partners, should be particularly useful for the development of libraries of biologically relevant
1,4-benzoxazine derivatives.
Introduction
such as solid-state synthesis, use of palladium catalysis, or
microwave irradiation, which allow one-pot multiple function-
1,4-Benzoxazine derivatives have received great attention in
chemical and medicinal research because of their natural
occurrence and important biological activities. Some of them
are central nervous system depressants, antipsychotic agents,
calcium antagonists, and antibacterial agents, while others are
potential drugs for treating neurodegenerative, inflammatory,
autoimmune, cardiovascular, and diabetic disorders.1
Past syntheses of 1,4-benzoxazine derivatives have relied
predominantly upon multistep procedures, using 2-nitrophenols
or 2-aminophenols as the starting materials. However, these
methods suffered some limitations in the generation of molecular
diversity.2 The recent development of new synthetic strategies
(1) For a recent review on biologically active 1,4-benzoxazine derivatives,
see the following: (a) Achari, B.; Mandal, S. B.; Dutta, P. K.; Chowdhury,
C. Synlett 2004, 2449 and references therein. For some recent examples,
see also the following: (b) Rybczynski, P. J.; Zeck, R. E.; Dudash, J.;
Combs, D. W.; Burris, T. P.; Yang, M.; Osborne, M. C.; Chen, X.; Demarest,
K. T. J. Med. Chem. 2004, 47, 196. (c) Yang, W.; Wang, Y.; Ma, Z.; Golla,
R.; Stouch, T.; Seethala, R.; Johnson, S.; Zhou, R.; Gu¨ngo¨r, T.; Feyen, J.
H. M.; Dickson, J. K. Bioorg. Med. Chem. Lett. 2004, 14, 2327. (d) Thomas,
A.; Ross, R. A.; Saha, B.; Mahadevan, A.; Razdan, R. K.; Pertwee, R. G.
Eur. J. Pharmacol. 2004, 487, 213. (e) Caliendo, G.; Perissutti, E.;
Santagada, V.; Fiorino, F.; Severino, B.; Cirillo, D.; d’Emmanuele di Villa
Bianca, R.; Lippolis, L.; Pinto, A.; Sorrentino, R. Eur. J. Med. Chem. 2004,
39, 815. (f) Dougherty K. J.; Bannatyne, B. A.; Jankowska, E.; Krutki, P.;
Maxwell, D. J. J. Neurosci. 2005, 25, 584. (g) Stefanic Anderluh, P.;
Anderluh, M.; Ilas, J.; Mravljak, J.; Sollner Dolenc, M.; Stegnar, M.; Kikelj,
D. J. Med. Chem. 2005, 48, 3110. (h) Lee, H. J.; Ban, J. Y.; Seong, Y. H.
Life Sci. 2005, 78, 294.
† UMR 8638 CNRS-Universite´ Paris 5.
‡ Centre National de la Recherche Scientifique.
10.1021/jo060452f CCC: $33.50 © 2006 American Chemical Society
Published on Web 07/15/2006
6374
J. Org. Chem. 2006, 71, 6374-6381

Electrochemically Induced Cascade Reaction
SCHEME 1. 1ox-Mediated Cascade Reaction Leading To Highly Functionalized 1,4-Benzoxazine Derivatives
alization, has therefore reinforced the interest in the 1,4-
benzoxazine core.1a,2a Because of its efficiency for introducing
a high degree of diversity through the variation of the structure
of both cycloaddition partners, the Diels-Alder reaction may
also constitute an attractive strategy for the synthesis of the 1,4-
benzoxazine scaffold.3 Despite this potential, the synthetic scope
of this reaction is limited by the requirement of o-azaquinone
heterodienes, which are not readily accessible stable compounds.
Most of the work has been mainly restricted to o-quinone
monooximes4 and o-quinonemonoimides.5 To the best of our
knowledge, only a few reports described the synthesis of 1,4-
benzoxazine derivatives from in situ chemically generated
o-azaquinone heterodienes.5j,l
A few years ago, we showed that electrogenerated 3,4-
azaquinone 1ox (Scheme 1) behaved as an effective organic
catalyst for the autorecycling oxidation of aliphatic primary
amines. The catalytic cycle produced the reduced catalyst 1red
and an imine as the product of amine oxidation.6 However, in
the case of R1R2CHCH2NH2 amines, the catalytic process ceased
after a few turnovers, as the catalyst 1ox was trapped through
inverse-electron-demand Diels-Alder (IEDDA) reaction with
the simultaneously electrogenerated tautomeric enamine form
of the alkylimine extruded during the catalytic process. This
cascade sequence, wherein both cycloaddition partners were
generated in situ, at room temperature, under metal-free condi-
tions, allowed the rapid synthesis of polyfunctionalized 1,4-
benzoxazine derivatives.7a
Although similar reactions with in situ chemically generated
tertiary enamine are known,8 the electrochemically induced
enamine cycloaddition reaction represented uncharted terrain.
Consequently, in a preceding work, we examined the scope of
this cascade reaction, first with respect to the secondary enamine
dienophile.7b Especially, to increase the molecular diversity, we
generated enamines in which the substituents on the amino group
were different from those linked to the double bond. For this
purpose, the amine R1R2CHCH2NH2 was catalytically oxidized
by o-azaquinone 1ox in the presence of a second primary
aliphatic amine R3NH2 (Scheme 1). Then, the 1ox-mediated
cascade reaction led to the construction of complex 1,4-
benzoxazine derivatives in a regiospecific manner and allowing
diastereospecific annellation.7b Interestingly, some of these
compounds were found to be significant targets as neuropro-
tective agents both in vitro and in vivo.7c, d
At this stage, a sole heterodiene (1ox) was engaged in the
cascade reaction with diverse enamine dienophiles. For a more
comprehensive investigation of the scope of the cascade reaction,
the variation of the heterodiene part proved to be essential. In
this prospect, there were several challenges inherent to this
reaction sequence which needed to be addressed. As a continu-
ation of our preceding work,7b we present hereafter the results
of new experiments aimed at exploring further the synthetic
potential of this cascade reaction using a variety of electro-
generated o-azaquinone entities, with the ultimate goal of
developing libraries of biologically relevant 1,4-benzoxazine
derivatives.
(2) For a recent review on the synthesis of 1,4-benzoxazine derivatives,
see ref 1a; for a recent report, see (a) Ilas, J.; Stefanic Anderluh, P.; Sollner
Dolenc, M.; Kikelj, D. Tetrahedron 2005, 61, 7325. See also (b) Teller, J.
In Houben-Weyl Methods of Organic Chemistry; Schaumann, E.; Ed.; Georg
Thieme: Stuttgart, Germany, 1997; Vol. E 9a, p 141. (c) Sainsbury, M. In
ComprehensiVe Heterocyclic Chemistry; Katritzky, A. R.; Rees, C. W. Eds.;
Oxford, 1984; Vol. 3, p 995.
(3) For reviews on hetero Diels-Alder reactions see (a) Boger, D. L.
Tetrahedron 1983, 39, 2869. (b) Boger, D. L. Chem. ReV. 1986, 86, 781.
(c) Boger, D. L.; Weinreb, S. M. In Hetero Diels-Alder Methodology in
Organic Synthesis; Academic Press: San Diego, CA, 1987.
(4) (a) McKillop, A.; Sayer, T. S. B. J. Org. Chem. 1976, 41, 1079. (b)
Nicolaides, D. N.; Awad, R. W.; Varella, E. A. J. Heterocyclic Chem. 1996,
33, 633. (c) Nicolaides, D. N.; Bezergiannidou-Balouctsi, C.; Awad, R.
W.; Litinas, K. E.; Malamidou-Xenikaki, E.; Terzis, A.; Raptopoulou, C.
P. J. Org. Chem. 1997, 62, 499.
Results and Discussion
(5) (a) Heine, H. W.; Barchiesi, B. J.; Williams, E. A. J. Org. Chem.
1984, 49, 2560. (b) Black, D. S. C.; Craig, D. C.; Heine, H. W.; Kumar,
N.; Williams, E. A. Tetrahedron Lett. 1987, 28, 6691. (c) Heine, H. W.;
Olsson, C.; Bergin, J. D.; Foresman, J. B.; Williams, E. A. J. Org. Chem.
1987, 52, 97. (d) Heine, H. W.; Schairer, W. C.; Suriano, J. A.; Williams,
E. A. Tetrahedron 1988, 44, 3181. (e) Heine, H. W.; Suriano, J. A.; Winkel,
C.; Burik, A.; Taylor, C. M.; Williams, E. A. J. Org. Chem. 1989, 54,
5926. (f) Desimoni, G.; Faita, G.; Righeti, P. P. Tetrahedron 1991, 47,
5857. (g) Heine, H. W.; Williams, D. K.; Rutherford, J. L.; Ramphal, J.;
Williams, E. A. Heterocycles 1993, 35, 1125. (h) Heine, H. W.; La Porte,
M. G.; Overbaugh, R. H.; Williams, E. A. Heterocycles 1995, 40, 743. (i)
Nicolaou, K. C.; Zhong, Y. L.; Baran, P. S. Angew. Chem., Int. Ed. 2000,
39, 622. (j) Nicolaou, K. C.; Baran, P. S.; Kranich, R.; Zhong, Y. L.; Sugita,
K.; Zou, N. Angew. Chem., Int. Ed. 2001, 40, 202. (k) Nicolaou, K. C.;
Sugita, K.; Baran, P. S.; Zhong, Y. L. Angew. Chem., Int. Ed. 2001, 40,
207. (l) Nicolaou, K. C.; Baran, P. S.; Zhong, Y. L.; Sugita, K. J. Am.
Chem. Soc. 2002, 124, 2212. (m) Nicolaou, K. C.; Sugita, K.; Baran, P. S.;
Zhong, Y. L. J. Am. Chem. Soc. 2002, 124, 2221.
The scope of the cascade reaction was investigated from
newly synthesized o-aminophenol derivatives 2 (Table 1), under
the optimum conditions previously reported7b which required a
mercury anode, methanol as the solvent, tetraethylammonium
(7) (a) Largeron, M.; Neudo¨rffer, A.; Vuilhorgne, M.; Blattes, E.; Fleury,
M.-B. Angew. Chem., Int. Ed. 2002, 41, 824. (b) Blattes, E.; Fleury, M.-
B.; Largeron, M. J. Org. Chem. 2004, 69, 882. (c) Blattes, E.; Fleury, M.-
B.; Largeron, M. Electrochim. Acta 2005, 50, 4902. (d) Blattes, E.; Lockhart,
B.; Lestage, P.; Schwendimann, L.; Gressens, P.; Fleury, M.-B.; Largeron,
M. J. Med. Chem. 2005, 48, 1282.
(8) For some examples of in situ chemically generated tertiary enamines,
see: (a) Juhl, K.; Jorgensen, K. A. Angew. Chem., Int. Ed. 2003, 42, 1498.
(b) Wabnitz, T. C.; Saaby, S.; Jorgensen, K. A. Org. Biomol. Chem. 2004,
2, 828. (c) Raw, S. A.; Taylor, R. J. K. Chem. Commun. 2004, 508. (d)
Raw, S. A.; Taylor, R. J. K. J. Am. Chem. Soc. 2004, 126, 12260. (e)
Ramachary, D. B.; Barbas III, C. F. Chem. Eur. J. 2004, 10, 5323. (f) Sainz,
Y. F.; Raw, S. A.; Taylor, R. J. K. J. Org. Chem. 2005, 70, 10086.
(6) (a) Largeron, M.; Fleury, M.-B. J. Org. Chem. 2000, 65, 8874. (b)
Largeron, M.; Neudo¨rffer, A.; Fleury, M.-B. Angew. Chem., Int. Ed. 2003,
42, 1026.
J. Org. Chem, Vol. 71, No. 17, 2006 6375

Xu et al.
TABLE 1. 2ox-Mediated Cascade Reaction Affording 2-Alkylamino-1,4-benzoxazines 3a-ia
a Reagents and conditions for 2a-i: 2 mM, (isobutylamine) ) 40 mM, MeOH, room temperature, Hg anode (E ) +50 mV vs SCE), 5-8 h. b Total
number of electrons transferred per molecule of mediator. c Yields refer to chromatographically pure isolated products.
hexafluorophosphate (TEAHFP) as the supporting electrolyte,
and 20.0 equiv of amine substrate per molecule of o-aminophe-
nol derivative 2. Under these conditions, the cyclic voltammo-
gram of compound 2ared (2 mM), in deaerated MeOH containing
isobutylamine (40 mM) chosen as an example of amine
substrate, showed, at a dropping mercury electrode, an oxidation
peak Pa at -50 mV versus SCE, due to a two-electron process,
the sweep rate being 0.5 V s-1. As can be seen in Figure 1, a
cathodic peak Pc appeared on the reverse sweep at -160 mV
versus SCE, illustrating the quasireversibility of the two-electron
transfer that could be assigned to the o-aminophenol 2ared/o-
iminoquinone 2aox redox couple.
When the controlled potential of the mercury pool was fixed
at +50 mV versus SCE, which is at a potential for which 2ared
could be oxidized to the iminoquinone form 2aox, the system
of peaks (Pa, Pc) remained unchanged for a certain time,
FIGURE 1. Voltammetric changes accompanying the electrochemical
consistent with steady-state catalytic behavior (Figure 1).
Accordingly, the anodic current remained constant over a certain
time, while a value of 12 was found for the total number of
electrons (n) transferred per molecule of 2ared in the catalytic
oxidation of 2a (2 mM) at a mercury anode (E ) +50 mV vs SCE),
in deaerated MeOH containing tetraethylammonium hexafluorophos-
phate (20 mM) and isobutylamine (40 mM): continuous line, before
electrolysis; dashed line, (a) 2-6, (b) 8, (c) 10, and (d) 12 mol of
electrons. Arrowheads indicate the direction of the potential sweep; V
) 0.5 V s-1. The vertical arrow indicates the initial potential point.
process (six turnovers). These results indicated that the 2ared
/
2aox system behaved as a redox mediator for the electrochemical
oxidation of isobutylamine to the corresponding alkylimine
(Scheme 1), according to the transamination mechanism previ-
ously reported.6b Further, the catalytic process ceased, and close
inspection of the exhaustively oxidized solution revealed that
6376 J. Org. Chem., Vol. 71, No. 17, 2006

Electrochemically Induced Cascade Reaction
rationalized from X-ray crystallographic analyses of benzox-
azines 3f,h, which showed the presence of a hydrogen bond
between the 2-hydroxyl substituent and the carbonyl group of
the benzophenone skeleton (Figure 3). As a consequence, the
substituted phenyl group twisted out of the plane,10 then
affecting the transmission of the substituent effects. Similarly,
a hydrogen bond could be expected on the 3,4-azaquinone
heterodiene which would produce the same effects.
A few other attempts to accomplish the cascade reaction
which are not listed in Table 1 deserve special note. The
electrochemical oxidation of 3,4-aminophenol derivatives 4a,b
(Table 2), which lack the 2-hydroxyl substituent, was performed
under the experimental conditions reported above. However,
as these substrates were oxidizable at a higher potential, the
mercury anode, which has a low anodic decomposition potential,
was replaced by a platinum anode. Then, the cyclic voltammo-
gram of compound 4ared (2 mM) showed an oxidation peak Pa
due to a two-electron process at + 200 mV versus SCE, the
sweep rate being 0.1 V s-1 (Figure 4). On the reverse sweep,
the lack of peak Pc illustrated the irreversibility of the overall
two-electron oxidation process, indicating that the produced 3,4-
azaquinone 4aox was reactive enough to be engaged in a fast
follow-up chemical reaction.
FIGURE 2. Spectrophotometric changes accompanying the electro-
chemical oxidation of 2a (2 mM) at a mercury anode (E ) +50 mV
vs SCE), in deaerated MeOH containing tetraethylammonium hexafluo-
rophosphate (20 mM) and isobutylamine (40 mM): (a) 0 (before
electrolysis), (b) 8, (c) 9, (d) 10, (e) 11, and (f) 12 mol of electrons.
Cell thickness is 0.1 cm.
electrogenerated o-azaquinone 2aox was trapped with the
tautomeric enamine form of the alkylimine, to give the
substituted 2-alkylamino-1,4-benzoxazine 3a in 71% yield
(Table 1, entry 1).
When the controlled potential of the Pt anode was fixed at
+ 350 mV versus SCE, which is at a potential for which 4ared
could be oxidized to the 3,4-azaquinone form 4aox, a decrease
in both anodic electrolysis current and Pa height was observed
(Figure 4). At the same time, the solution became dark, while
a coulometric value of 4.0 was found for n. These results
indicated that the produced 3,4-azaquinone 4aox was probably
converted into a Michael adduct which, after a subsequent two-
electron oxidation reaction, spontaneously decomposed to
melanin-like polymers. Accordingly, no stable product could
be detected at the end of the electrolysis. To evaluate the ability
of the 2-substituent to prevent the competing formation of
Michael adduct, 3,4-aminophenol 4b (Table 2), bearing a
2-methyl group, was used as the substrate and treated under
the same experimental conditions. Similarly, no stable product
could be isolated at the end of the electrolysis.
The progress of the electrolysis was simultaneously followed
by monitoring the UV-vis absorption spectrum. After applica-
tion of the anodic potential (E ) +50 mV vs SCE), no spectral
changes were observed for a time, consistent with steady-state
catalytic behavior (Figure 2). After the consumption of 7 F
mol-1, a decrease in the UV-vis absorption band shown by
the monoanionic form of 2red at 334 nm (ꢀ/mol-1 L cm-1
)
16500) was observed, while new bands at 302 and 254 nm
developed. Spectral changes showed two isosbestic points at
313 and 268 nm, indicating that a simple equilibrium between
two species was shifted. The new bands at 302 and 254 nm
could be assigned to the 2-alkylamino-1,4-benzoxazine 3a, as
corroborated after recording the UV-vis absorption spectrum
of the isolated product.
At this point, it could be concluded that the presence of the
2-hydroxyl substituent was of overriding importance for the
success of the cascade reaction. On the basis of the low value
found for n using 4a or 4b as the substrate (4 instead of 10-12
in the case of examples collected in Table 1), we suspected
that the requirement for a 2-hydroxyl group was mainly inherent
to the catalytic process. So, to circumvent the limitation related
to the catalytic process, we envisioned a variant of our initial
procedure in which the o-azaquinone entity was electrogenerated
in the presence of a preformed enamine. In other words, we
sought to design an alternative procedure which could be
generally applied to construct polyfunctionalized 1,4-benzox-
azine derivatives, without the need to employ a 2-aminoresor-
cinol derivative as the starting material (Table 1).
With the reliable set of conditions in hand, the electrochemi-
cally induced cascade reaction was investigated using different
o-aminophenol substrates 2a-i. Table 1 shows some examples
of the molecular diversity that is accessible through this reaction
sequence, which encompasses an IEDDA reaction between the
electron-poor o-azaquinone heterodiene and the electron-rich
secondary alkylenamine dienophile. This cycloaddition reaction
occurred at room temperature, within 5-8 h, with complete
regioselectivity. The more electron-rich carbon atom of the
enamine dienophile added to the nitrogen atom of the o-
azaquinone heterodiene. In all cases (Table 1, entries 1-9), the
methodology was successful, producing the 1,4-benzoxazine
derivatives in good yields. Interestingly, bromosubstituted 3,4-
aminophenol derivative 2d reacted effectively to furnish the
desired cycloadduct 3d in 80% yield (Table 1, entry 4).9 In the
specific case of 3,4-aminophenol derivatives bearing a ben-
zophenone framework, the scope of the cascade reaction could
be also extended by the attachment of substituents on the
benzoyl moiety (Table 1, entries 5-8). Surprisingly, the yield
of the cycloadduct was good regardless of the electronic and/
or steric effects of the substituents. This outcome could be
We first performed the anodic controlled-potential electrolysis
of 3,4-aminophenol 4a (Table 2 and Scheme 2), in the presence
of 4-cyclohexylidenemethylmorpholine. The optimized condi-
tions required 5 equiv of enamine per molecule of 4a. However,
as the basicity of the alkylenamine was not sufficient to produce
the phenolate anion of 4ared, which is the sole form that can be
oxidized to 3,4-azaquinone 4aox, 1 equiv of morpholine was
added to the bulk solution as the ancillary base. Then, the
J. Org. Chem, Vol. 71, No. 17, 2006 6377

Xu et al.
FIGURE 3. Ortep view of 3f and 3h. Displacement ellipsoids are drawn at the 30% probability level.
TABLE 2. Anodic Oxidation of Various o-Aminophenol Derivatives 4a-h and Trapping of the Electrogenerated o-Azaquinone Heterodiene
by 4-Cyclohexylidenemethylmorpholinea
a Reagents and conditions for 4a-h: 2 mM, (enamine) ) 10 mM, MeOH, room temperature, Pt anode (E ) + 450 mV vs SCE), 4 h; 1equiv of
morpholine was added to the bulk solution for producing the phenolate anion of 4a-h. b Yields refer to chromatographically pure isolated products.
reaction could go to completion by acid-base equilibrium
by the major neutral form of 4ared at 343 and 244 nm was
observed, while two new bands at 360 and 254 nm developed
(Figure 5). Spectral changes showed two isosbestic points at
355 and 310 nm, indicating that a simple equilibrium between
two species was shifted.
displacement.
When E was fixed at +450 mV versus SCE, that is, at a
potential for which 4ared could be oxidized to the 3,4-azaquinone
form 4aox (Figure 4), a coulometric value of 2.2 was found for
n, consistent with a two-electron transfer. After applying the
potential, a decrease in the UV-vis absorption bands shown
After 4 h, treatment of the exhaustively electrolyzed solution
afforded the desired 2-alkylamino-1,4-benzoxazine derivative
6378 J. Org. Chem., Vol. 71, No. 17, 2006

Electrochemically Induced Cascade Reaction
FIGURE 4. Voltammetric changes accompanying the electrochemical
oxidation of 4a (2 mM) at a platinum anode (E ) +450 mV vs SCE),
in deaerated MeOH containing tetraethylammonium hexafluorophos-
phate (20 mM) and isobutylamine (40 mM): (a) 0 (before electrolysis),
(b) 1, (c) 2, (d) 3, and (e) 4; dashed line, 4.1 mol of electrons.
Arrowheads indicate the direction of the potential sweep; V ) 0.1 V
s-1. The vertical arrow indicates the initial potential point.
FIGURE 5. Spectrophotometric changes accompanying the electro-
chemical oxidation of 4a (2 mM) at a platinum anode (E ) +450 mV
vs SCE), in deaerated MeOH containing tetraethylammonium hexafluo-
rophosphate (20 mM), 4-cyclohexylidenemethylmorpholine (10 mM),
and morpholine (2 mM): (a) 0 (before electrolysis), (b) 1, and (c) 2.2
mol of electrons. Cell thickness is 0.1 cm.
SCHEME 2. Tandem Oxidation-IEDDA Reaction Leading
to 2-Alkylamino-1,4-benzoxazine Derivatives 5a-h
SCHEME 3. Tandem Oxidation-IEDDA Reaction using 2i
as the Starting Material
5a, in 56% yield, as a single regioisomer (Table 2, entry 1).
Conclusive evidence of the structure of the regioisomer was
provided by X-ray crystallographic analyses of the benzoxazine
derivatives 5e,f (Figure 6).11
The scope of the tandem oxidation-IEDDA reaction was then
investigated using a series of o-aminophenol derivatives 4a-
h. As illustrated in Table 2, the method was sufficiently general
to accommodate a variety of substituents.
As predicted for an IEDDA reaction, electrogenerated o-
azaquinone heterodienes, with a pronounced electron-poor
character, produced the desired cycloadduct in good yield (Table
2, entries 1-6), whereas introduction of an amide substituent
resulted in lower yield (Table 2, entry 7). Anodic oxidation of
2-aminophenol generated a poor reactive o-iminoquinone het-
erodiene, which mainly decomposed to melanin-like polymers,
giving only 10% of the desired cycloadduct (Table 2, entry 8).
In general, the tandem oxidation-IEDDA reaction produced
slightly lower yields of cycloadducts than those yielded by the
aforementioned cascade reaction, presumably because of the
replacement of the Hg anode by a Pt anode rather than because
of the lack of the 2-hydroxyl substituent. To ensure that, we
carried out the anodic controlled-potential electrolysis of the
2-aminoresorcinol derivative 2i, in the presence of 4-cyclo-
hexylidenemethylmorpholine using the two anodes, alternatively
(Scheme 3). Accordingly, we found that replacing the Hg anode
by a Pt anode lowered the yield of cycloadduct 6 from 70 to
58%.
Note that o-azaquinone heterodienes displayed an interesting
reactivity in the sense that the cycloaddition reactions were
regiospecific, irrespective of the nature of the 2-substituent, thus
ensuring their synthetic utility. These results contrast with our
preceding observations concerning related o-quinone hetero-
dienes, generated from pyrogallol derivatives, for which the
2-hydroxyl group is required for inducing the regiospecificity.12
Although the o-quinone entities have been the most widely
utilized for IEDDA reactions,13 the o-iminoquinone species
proved to be more versatile heterodienes which would merit
further developments.
(9) Bromosubstituted aromatic rings are particularly attractive in diversity-
oriented synthesis because they can be easily transformed into differently
substituted aromatics by cross-coupling reactions.
(10) As shown in Figure 3, in both compounds, a strong intramolecular
hydrogen bond is established between the hydroxyl groups O9-H and the
oxygen atoms O10 (for 3f, distances O9‚‚‚O10 ) 2.547(3) Å, O9-H )
1.27, HO9‚‚‚O10 ) 1.47 Å, angle O-H‚‚‚O ) 136.2°; for 3h, distances
O9‚‚‚O10 ) 2.603(3) Å, O9-H ) 1.16, HO9‚‚‚O10 ) 1.59 Å, angle O-
H‚‚‚O ) 141.2°), leading to respective shorter bonds C5-OH of 1.360(3)
and 1.354(3) Å, and longer bonds C11-O10 of 1.247(3) and 1.245(3) Å.
Torsion angle values and mean plane calculations show that the oxazine
rings are in a half-chair conformation with atoms C2 and C3 deviated by,
respectively, -0.349(3) and 0.435(2) Å in 3f and by -0.373(2) and 0.446-
(2) Å in 3h, from the mean plane of the other four atoms. The mean planes
of the benzoxazine and the phenyl ring at C11 are tilted by 132.5° in 3f
and 101.5° in 3h.
(11) For the two enantiomers, torsion angle and mean plane calculations
show that the oxazine rings exhibit a nearly C3 envelope conformation,
with atom C3 deviated by -0.629(1) Å in 5e and -0.690(2) Å in 5f from
the mean plane of the other five atoms. In 5f, the nitro group lies in the
mean plane of the benzoxazine (torsion angles C5-C6-N9-O10 ) 5.7-
(3)°, C7-C6-N9-O11 ) 4.6(3)°). However, differences appear at the C2
atom where the morpholine ring is fixed in an axial position (H2 atom
being equatorial) in 5e, while it is the opposite in 5f. Moreover, the
cyclohexyl group at C3 is orientated differently.
(12) Xu, D.; Chiaroni, A.; Largeron, M. Org. Lett. 2005, 7, 5273.
(13) For a review on cycloaddition reactions of o-quinone heterodienes
see Nair, V.; Kumar, S. Synlett 1996, 1143 and references therein.
J. Org. Chem, Vol. 71, No. 17, 2006 6379

Xu et al.
FIGURE 6. ORTEP view of 5e and 5f. Displacement ellipsoids are drawn at the 30% probability level.
SCHEME 4. Effect of 2-Hydroxyl Group on The Reaction
Pathway
partners are generated in situ, under mild conditions, allows
the IEDDA reaction of an o-azaquinone heterodiene and a
secondary alkylenamine dienophile, two chemically nonacces-
sible unstable materials. The cycloaddition reaction proceeds
with complete regioselectivity, ensuring its synthetic utility. The
cascade reaction was found to be general with electron-poor
o-azaquinone entities generated from substituted 2-aminoresor-
cinol substrates. Using o-aminophenol derivatives which lack
the 2-hydroxyl group, the generated o-azaquinone species was
devoid of catalytic efficiency toward the amine oxidation, then
preventing the in situ formation of the enamine dienophile. To
circumvent this limitation, a variant, in which the o-azaquinone
heterodiene is generated in the presence of a preformed enamine,
has been achieved. This tandem oxidation-IEDDA reaction,
which tolerates diverse functional groups, completes the initial
cascade sequence in terms of increasing the molecular diversity.
These one-pot methodologies, which allow variations in both
cycloaddition partners, should be particularly useful for the
assembly of libraries of biologically relevant 1,4-benzoxazine
derivatives. In this respect, the preparation of the new series of
1,4-benzoxazine derivatives 3 and 5 offers the opportunity to
explore further the structural requirements for efficient neuro-
protective activity, with the objective to identify the most
promising targets.
All these results indicated that, in the course of the cascade
reaction, the presence of the 2-hydroxyl substituent constituted
a prerequisite only to the development of the catalytic process.
Then, we questioned about the exact role of the 2-hydroxyl
substituent to convert a catalytic inert o-azaquinone species into
a highly effective organic catalyst. For this purpose, we
thoroughly reexamined the ionic transamination mechanism we
have previously reported for the autorecycling oxidation of
aliphatic amines.6b This process involves R-proton abstraction
and the subsequent electron flow from the R-carbon to the
o-azaquinone moiety, which aromatizes to the Schiff base 2′ox
(Scheme 4). At this stage, the activation of the imine function
for further nucleophilic attack of amine would be provided by
an intramolecular hydrogen bond between the 2-hydroxyl group
and the imine nitrogen, generating the highly reactive cyclic
transition state 2′ox. This activated nucleophilic attack of amine
would constitute the driving force of the overall transamination
mechanism, then preventing any competitive Michael addition
reaction. Similar effects of 2-phenolic hydroxyl group on the
reactivity of ketimine derivatives have been recently reported
in the literature.14
Experimental Section
All apparatus, cells and electrodes were identical with those
described previously.15 Chemicals were commercial products of the
highest available purity and were used as supplied. 2-Amino-4-
nitrophenol 4f and 2-aminophenol 4h were commercially available
compounds. Substituted o-aminophenol starting materials 2i, 4a,
4c, and 4e were synthesized through previously reported proce-
dures.16 The syntheses of new 2-aminoresorcinol derivatives 2a-h
together with o-aminophenol substrates 4b, 4d, and 4g, are reported
in the Supporting Information.
2ox-Mediated Cascade Reaction Leading to Benzoxazine
Derivatives 3a-i. General Procedure A. o-Aminophenol substrate
2 (0.5 mmol) and excess isobutylamine (10 mmol) were dissolved
in methanol (250 mL), which contained tetraethylammonium
hexafluorophosphate (TEAHFP) as the supporting electrolyte (5
mmol). The resulting solution was then oxidized under nitrogen,
at room temperature, at a mercury pool whose potential was fixed
Conclusion
New insights into the scope and mechanism of an electro-
chemically induced cascade reaction, which affords highly
functionalized 1,4-benzoxazine derivatives, have been delineated
through the variation of the structure of the o-azaquinone
mediator. This cascade sequence, wherein both cycloaddition
(15) Largeron, M.; Neudo¨rffer, A.; Fleury, M.-B. J. Chem. Soc., Perkin
Trans. 2 1998, 2721.
(16) (a) Lindner H. Chem. Ber. 1923, 56, 1870. (b) Tseitlin, G. M.;
Tokarev, B. V.; Kulagin, V. N. J. Org. Chem. USSR (Engl. Trans.) 1982,
18, 931. (c) Edkins, L. J. Pharm. Pharmacol. 1936, 75, 102. (d) Caliendo,
G.; Perissutti, E.; Santagada, V.; Fiorino, F.; Severino, B.; D’Emmanuele
di Villa Bianca, R.; Lippolis, L.; Pinto, A.; Sorrentino, R. Bioorg. Med.
Chem. 2002, 10, 2663.
(14) (a) Miyabe, H.; Yamaoka, Y.; Takemoto, Y. Synlett 2004, 2597.
(b) Miyabe, H.; Yamaoka, Y.; Takemoto, Y. J. Org. Chem. 2006, 71, 2099.
6380 J. Org. Chem., Vol. 71, No. 17, 2006

Electrochemically Induced Cascade Reaction
at +50 mV versus SCE (initial current 50 mA). After exhaustive
electrolysis (5-8 h, n ) 10-12), that is, when a negligible current
was recorded (1 mA), the solution was neutralized with dry ice,
and the solvent was removed under reduced pressure. The brown
oil residue was then poured into diethyl ether (20 mL). Insoluble
TEAHFP was filtered off, and the filtrate was evaporated under
reduced pressure, at 30 °C. Flash chromatography of the residue
on silica gel afforded the expected 1,4-benzoxazine derivatives.
[(R,S)-3,3-Dimethyl-5-hydroxy-2-isobutylamino-3,4-dihydro-
2H-1,4-benzoxazin-6-yl](4-fluorophenyl) Methanone 3f. Orange
crystal: mp 134-136 °C (from diethyl ether and petroleum ether).
1H NMR (300 MHz, CDCl3): δ 12.63 (s, 1 H), 7.80-7.60 (m, 2
H), 7.30-7.10 (m, 2 H), 6.96 (d, J ) 9.0 Hz, 1 H), 6.41 (d, J )
9.0 Hz, 1 H), 4.65 (s, 1 H), 4.02 (s, 1 H), 2.90-2.70 (m, 1 H),
2.60-2.40 (m, 1 H), 1.91 (bs, 1H), 1.80-1.60 (m, 1 H), 1.31 (s, 3
H), 1.23 (s, 3 H), 0.91 (d, J ) 6.0 Hz, 3 H), 0.88 (d, J ) 9.0 Hz,
3 H). 13C NMR (75 MHz, CDCl3): δ 199.0, 166.3, 162.9, 152.1,
147.6, 134.7, 131.5, 131.4, 123.7, 121.2, 115.5, 115.2, 112.5, 108.7,
93.0, 53.2, 50.4, 28.9, 26.4, 24.6, 20.4. MS (ES+): m/z 373 [M +
H]+, m/z 395 [M + Na]+. X-ray analysis: A little plate of 0.35
mm × 0.30 mm × 0.10 mm was used. Empirical formula C21H25-
FN2O3, Mr ) 372.43, T ) 293 K. Monoclinic system, space group
P21/c, Z ) 4, a ) 14.022(13), b ) 10.920(9), c ) 12.843(12) Å,
â ) 92.64(5)°, V ) 1964 Å3, dcalcd ) 1.259 g cm-3, F(000) ) 792,
µ ) 0.091 mm-1, λ(Mo KR) ) 0.71073 Å. A total of 9182 intensity
data were measured with a Nonius Kappa-CCD diffractometer
giving 7486 monoclinic reflections, of which 3929 unique. Refine-
ment of 249 parameters on F2 led to R1(F) ) 0.0658 calculated
with 1836 observed reflections as I g 2σ(I) and wR2(F2) ) 0.2129
considering all the 3929 data. Goodness of fit ) 1.022. CCDC
deposition number: 298621.
[(R,S)-3,3-dimethyl-5-hydroxy-2-isobutylamino-3,4-dihydro-
2H-1,4-benzoxazin-6-yl](2,6-dimethylphenyl) Methanone 3h.
Yellow crystal: mp 154-156 °C (from petroleum ether). 1H NMR
(300 MHz, CDCl3): δ 12.82 (s, 1 H), 7.24 (t, J ) 9.0 Hz, 1 H),
7.08 (d, J ) 9.0 Hz, 2 H), 6.54 (d, J ) 9.0 Hz, 1 H), 6.28 (d, J )
9.0 Hz, 1 H), 4.63 (s, 1 H), 3.98 (s, 1 H), 2.90-2.70 (m, 1 H),
2.60-2.40 (m, 1 H), 2.19 (s, 6 H), 1.88 (bs, 1H), 1.80-1.60 (m, 1
H), 1.31 (s, 3 H), 1.27 (s, 3 H), 0.91 (d, J ) 6.0 Hz, 3 H), 0.88 (d,
J ) 6.0 Hz, 3 H). 13C NMR (75 MHz, CDCl3): δ 204.6, 151.7,
148.0, 138.7, 134.1, 128.8, 127.4, 122.8, 120.8, 113.9, 109.3, 93.0,
53.3, 50.4, 28.9, 26.5, 24.5, 20.4, 19.4. MS (ES+): m/z 383 [M +
H]+, m/z 405 [M + Na]+. X-ray analysis: A bright yellow needle
of 0.40 mm × 0.25 mm × 0.125 mm was used. Empirical formula
C23H30N2O3, Mr ) 382.49, T ) 293 K. Monoclinic system, space
group C2/c, Z ) 8, a ) 31.003(13), b ) 7.393(4), c ) 20.004(8)
Å, â ) 112.19(3)°, V ) 4245 Å3, dcalcd ) 1.197 g cm-3, F(000)
)1648, µ ) 0.079 mm-1, λ(Mo KR) ) 0.71073 Å. A total of 17704
data was measured with a Nonius Kappa-CCD diffractometer
reduced to 7921 monoclinic reflections, of which 4764 unique.
Refinement of 260 parameters on F2 led to R1(F) ) 0.0623
calculated with the 2470 observed reflections having I g 2σ(I) and
wR2(F2) ) 0.1944 considering all the 4764 data. Goodness of fit
) 1.030. CCDC deposition number: 298622.
was oxidized at a platinum anode whose potential was fixed at +450
mV versus SCE, under nitrogen, at room temperature. After
exhaustive electrolysis (4 h, n ) 2), the solution was treated as
above (general procedure A) to give the 1,4-benzoxazine derivatives
5a-h.
[(R,S)-2-Morpholino-3-spiro-1′-cyclohexyl-3,4-dihydro-2H-
1,4-benzoxazin-6-yl] Nitrile 5e. Colorless crystal: mp 174-176
°C (from diethyl ether). 1H NMR (300 MHz, CDCl3): δ 6.98 (d,
J ) 9.0 Hz, 1 H), 6.83 (s, 1 H), 6.81 (d, J ) 9.0 Hz, 1 H), 4.59 (s,
1 H), 4.05 (s, 1 H), 3.75-3.55 (m, 4 H), 3.40-3.25 (m, 2 H), 2.65-
2.50 (m, 2 H), 1.80-1.00 (m, 10 H). 13C NMR (75 MHz, CDCl3):
δ 148.3, 132.4, 123.7, 119.8, 117.9, 115.8, 103.5, 95.6, 67.1, 52.5,
48.9, 35.1, 34.0, 25.2, 21.4, 21.2. MS (ES+): m/z 314 [M + H]+,
m/z 336 [M + Na]+. X-ray analysis: A little plate of 0.70 mm ×
0.40 mm × 0.10 mm was used. Empirical formula C18H23N3O2,
Mr ) 313.39, T ) 293 K. Monoclinic system, space group P21/c,
Z ) 4, a ) 10.734(4), b ) 10.465(4), c ) 15.663(6) Å, â ) 107.34
(2)°, V ) 1679 Å3, dcalcd ) 1.239 g cm-3, F(000) ) 672, µ )
0.082 mm-1, λ(Mo KR) ) 0.71073 Å. A total of 18442 intensity
data were measured with a Nonius Kappa-CCD diffractometer
giving 8335 monoclinic reflections, of which 4489 unique. Refine-
ment of 208 parameters on F2 led to R1(F) ) 0.0483 calculated
with 2923 observed reflections as I g 2σ(I) and wR2(F2) ) 0.1286
considering all the 4489 data. Goodness of fit ) 1.031. CCDC
deposition number: 298623
(R,S)-2-Morpholino-3-spiro-1′-cyclohexyl-6-nitro-3,4-dihydro-
2H-1,4-benzoxazine 5f. Yellow crystal: mp 210-212 °C (from
diethyl ether and chloroform). 1H NMR (300 MHz, CDCl3): δ 7.62
(dd, J ) 9.0 and 3.0 Hz, 1 H), 7.50 (d, J ) 3.0 Hz, 1 H), 6.82 (d,
J ) 9.0 Hz, 1 H), 4.63 (s, 1 H), 4.18 (s, 1 H), 3.80-3.60 (m, 4 H),
3.40-3.20 (m, 2 H), 2.70-2.40 (m, 2 H), 1.80-1.00 (m, 10 H).
13C NMR (75 MHz, CDCl3) δ 150.3, 141.5, 131.8, 115.6, 114.9,
110.1, 96.0, 67.1, 52.5, 48.9, 35.1, 33.9, 25.2, 21.4, 21.2; MS
(ES+): m/z 334 [M + H]+, m/z 356 [M + Na]+. X-ray analysis:
A little prismatic crystal of 0.40 mm × 0.25 mm × 0.125 mm was
used. Empirical formula C17H23N3O4, Mr ) 333.38, T ) 293 K.
Triclinic system, space group P1h, Z ) 2, a ) 9.373(6), b ) 9.802-
(5), c ) 10.391(8) Å R ) 71.79(4), â ) 70.18(4), γ ) 72.29(4)°,
V ) 832 Å3, dcalcd ) 1.331 g cm-3, F(000) ) 356, µ ) 0.096
mm-1, λ(Mo KR) ) 0.71073 Å. A total of 4474 reflections was
measured with a Nonius Kappa-CCD diffractometer of which 3327
unique. Refinement of 220 parameters on F2 led to R1(F) ) 0.0504
calculated with the 2629 observed reflections having I g 2σ(I) and
wR2(F2) ) 0.1377 considering all the 3327 data. Goodness of fit
) 1.022. CCDC deposition number: 298624.
Acknowledgment. This work is dedicated to the memory
of Professor Pierre Potier. D.X. thanks Paris 5 University and
the MRT for a postdoctoral grant.
Supporting Information Available: Typical experimental
procedures and spectroscopic data for new starting materials 2a-
h, 4b,d,g; analytical data for compounds 3a-e,g,i, 5a-d,g,h and
6, including 1H and 13C NMR spectra of all 1,4-benzoxazine
derivatives 3a-i, 5a-h and 6 together with crystallographic files
(CIF) for compounds 3f,h and 5e,f. This material is available free
of charge via the Internet at http://pubs.acs.org.
Tandem Oxidation-IEDDA Reaction Leading to Benzoxazine
Derivatives 5a-h. General Procedure B. Freshly distilled 4-cy-
clohexylidenemethylmorpholine (2.5 mmol) and morpholine (0.5
mmol) were added to a 0.02 M solution of TEAHFP in MeOH.
Then, o-aminophenol 4 (0.5 mmol) was added to the solution which
JO060452F
J. Org. Chem, Vol. 71, No. 17, 2006 6381