Approach to 3-aminoindolin-2-ones via oxime ether functionalized carbamoylcyclohexadienes

Article abstract of DOI:10.1021/jo049617f

O-Benzyloxime ether substituted amidocyclohexadienes were prepared in three steps in good yields from 2-aminoacetophenone. EPR spectroscopic observations and product analyses showed that peroxide-induced decompositions of model compounds led to indolin-2-ones with benzyloxyaminyl substitution at their 3-positions. The cyclization steps were very rapid and took place regioselectively at the C-atoms of the C=N bonds, by 5-exo ring closures. An O-trityloxime ether analogue was also prepared. The cyclohexadienyl intermediate smoothly yielded an alkoxylaminyl radical again by rapid 5-exo-cyclization. However, ring closure was quickly followed by another β-scission step that released the persistent trityl radical and a 3-nitrosoindolin-2-one derivative. EPR spectroscopic evidence showed that the nitroso compound trapped other transient intermediates to afford a series of nitroxides. GC-MS analyses of products formed in reactions including methyl thioglycolate indicated that 1-benzyl-3-methyl-1,3-dihydro-2H-indol-2-one was derived from the indolinone moiety.

Full text of DOI:10.1021/jo049617f

Ap p r oa ch to 3-Am in oin d olin -2-on es via Oxim e Eth er
F u n ction a lized Ca r ba m oylcycloh exa d ien es
A. Franco Bella, Alexandra M. Z. Slawin, and J ohn C. Walton*
School of Chemistry, University of St. Andrews, St. Andrews, Fife UK KY16 9ST, Scotland
jcw@st-andrews.ac.uk
Received March 9, 2004
O-Benzyloxime ether substituted amidocyclohexadienes were prepared in three steps in good yields
from 2-aminoacetophenone. EPR spectroscopic observations and product analyses showed that
peroxide-induced decompositions of model compounds led to indolin-2-ones with benzyloxyaminyl
substitution at their 3-positions. The cyclization steps were very rapid and took place regioselectively
at the C-atoms of the CdN bonds, by 5-exo ring closures. An O-trityloxime ether analogue was also
prepared. The cyclohexadienyl intermediate smoothly yielded an alkoxylaminyl radical again by
rapid 5-exo-cyclization. However, ring closure was quickly followed by another â-scission step that
released the persistent trityl radical and a 3-nitrosoindolin-2-one derivative. EPR spectroscopic
evidence showed that the nitroso compound trapped other transient intermediates to afford a series
of nitroxides. GC-MS analyses of products formed in reactions including methyl thioglycolate
indicated that 1-benzyl-3-methyl-1,3-dihydro-2H-indol-2-one was derived from the indolinone moiety.
In tr od u ction
organic halide precursor.6a,b 1-Carbamoyl-2,5-cyclohexa-
dienes release carbamoyl radicals (aminoacyl radicals),
suitably unsaturated examples of which ring close to
afford â- or γ-lactams in moderate yields.7a-c
Organotin radical methodology, with its proven reli-
ability and flexibility, is an invaluable preparative tool,
but its usefulness is somewhat impaired by neurotoxicity
problems. Despite determined efforts over the past few
years,1a-c the signs are that a single reagent, capable of
comprehensively replacing toxic organotin hydrides in
radical-mediated processes, is unlikely to be found.
Instead, diverse suites of precursors encompassing metal-
based,2a-f all-organic1,3a-e and polymer-bound types,4a,b
adapted to fulfill targeted chemical roles, are being
developed. Functionalized cyclohexadienes combine within
one compound the ability to release a desired radical and
hydrogen donor capability. We have successfully used
them in radical chain reactions,5a-e and Studer et al. have
introduced silylated 1,4-cyclohexadienes that efficiently
release silyl radicals for use in conjunction with an
The oxime ether functional group can exhibit up to 3
orders of magnitude higher radical cyclization rates than
analogous alkene acceptors.8 Moreover, a useful func-
tional group remains available for further synthetic
elaboration. Other attractive features of oxime ethers are
their stability to hydrolysis and the specificity of radical
attack at the carbon of the CdN bond. This high regio-
selectivity was incisively demonstrated by Warkentin’s
competition studies between 5-exo and 6-endo alkyl
radical ring closures. It was found that even the normally
disfavored 6-endo attack was preferred at the carbon
center over 5-exo cyclization at the nitrogen.9 Currently,
radical addition onto oxime ether derivatives is rapidly
developing and is becoming a reliable synthetic strategy
which can efficiently be applied in syntheses of complex
natural products,10a-c e.g., 1-deoxynorjirimycin,11 (+)-7-
deoxypancratistatin,12 morphine alkaloids,13 (-)-balanol
fragments,14 and pyrrolidine nucleoside analogues.15
* To whom correspondence should be addressed. Phone: 44 (0)1334
463864. Fax: 44 (0)1334 463808.
(1) (a) Baguley, P. A.; Walton, J . C. Angew. Chem., Int. Ed. 1998,
37, 3072. (b) Sugi, M.; Togo, H. Tetrahedron 2002, 58, 3171. (c) Studer,
A.; Amrein, S. Synthesis 2002, 835.
(2) (a) Chatgilialoglu, C. Acc. Chem. Res. 1992, 25, 188-194. (b)
Pike, P.; Hershberger, S.; Hershberger, J . Tetrahedron 1988, 44, 6295-
6304. (c) Pike, P.; Hershberger S.; Hershberger, J . Tetrahedron Lett.
1985, 26, 6289-6290. (d) Chatgilialoglu, C.; Ballestri, M. Organome-
tallics 1995, 14, 5017-5018. (e) Togo, H.; Yamazaki, O.; Matsubayashi,
S.; Yokoyama, M. Tetrahedron Lett. 1998, 39, 1921. (f) Chatgilialoglu,
C.; Ballestri, M.; Escudie, J .; Pailhous, I. Organometallics 1999, 18,
2395.
(3) (a) Zard, S. Z. Angew. Chem., Int. Ed. Engl. 1997, 36, 672-685.
(b) Murphy, J . A.; Khan, T. A.; Tripoli, R.; Crawford; J . J .; Martin, C.
G. Org. Lett. 2003, 5, 2971. (c) Murphy, J . A.; Graham, S. R.; Coates,
D. Tetrahedron Lett. 1999, 40, 2415. (d) Crich, D.; Yao, Q. J . Org. Chem.
1996, 61, 3566. (e) Crich, D.; Hwang, J . T.; Yuan, H. J . Org. Chem.
1996, 61, 6189.
(5) (a) Binmore, G.; Walton, J . C.; Cardellini, L. J . Chem. Soc., Chem.
Commun. 1995, 27. (b) Baguley, P. A.; Binmore, G.; Milne, A.; Walton,
J . C. Chem. Commun. 1996, 2199. (c) Baguley, P. A.; Walton, J . C. J .
Chem. Soc., Perkin Trans. 1 1998, 2073. (d) J ackson, L. V.; Walton, J .
C. Chem. Commun. 2000, 2327. (e) J ackson, L. V.; Walton, J . C. J .
Chem. Soc., Perkin Trans. 2 2001, 1758.
(6) (a) Studer, A.; Amrein, S. Angew. Chem., Int. Ed. 2000, 39, 3080.
(b) Studer, A.; Amrein, S.; Schleth, F.; Schulte, T.; Walton, J . C. J .
Am. Chem. Soc. 2003, 125, 5726-5733.
(7) (a) J ackson, L. V.; Walton, J . C. Chem. Commun. 2000, 2327-
2328. (b) Bella, A. F.; J ackson, L. V.; Walton, J . C. J . Chem. Soc., Perkin
Trans. 2 2002, 1839-1843. (c) Bella, A. F.; J ackson, L. V.; Walton, J .
C. Org. Biomol. Chem. 2004, 2, 421.
(4) (a) Gerlach, M.; J ordens, F.; Kuhn, H.; Neumann, W. P. J . Org.
Chem. 1991, 56, 5971. (b) J unggebauer, J .; Neumann, W. P. Tetrahe-
dron 1997, 53, 1301-1310.
(8) Fallis, A. G.; Brinza, I. M. Tetrahedron 1997, 53, 17543.
(9) Tomaszewski, M. J .; Warkentin, J .; Werstiuk, N. H. Aust. J .
Chem. 1995, 48, 291.
10.1021/jo049617f CCC: $27.50 © 2004 American Chemical Society
Published on Web 08/05/2004
5926
J . Org. Chem. 2004, 69, 5926-5933

Synthesis of Aminoindolinones via Carbamoylcyclohexadienes
SCHEME 1. P r ep a r a tion of O-Ben zyloxim e Eth er
The indole nucleus is a key structural feature in a large
number of alkaloids and related compounds, many of
which exhibit potent pharmacological activity. This has
stimulated interest in finding mild, free-radical-based
methods of making the basic nucleus as well as natural
products derived from it. For example, Zard has applied
xanthates as radical precursors in a synthesis of the sleep
regulatory hormone melatonin.16 Peduncularine was
made via cyclization of an amidyl radical derived from
an N-sulfinyl amide.17a-c Indolones were prepared by
diethylphosphine oxide mediated cyclizations of iodophe-
nyl amides.18 We envisaged that our amidocyclohexa-
dienes, suitably functionalized with oxime ether radical
acceptors, would be mild, nontoxic carbamoyl radical
precursors for preparations of indolones with N-function-
ality at the 3-position. We prepared several precursors
with O-benzyl oxime ether acceptors and O-trityl oxime
ether acceptors and showed by a combination of end
product analysis and EPR spectroscopy that this strategy
works well.
Su bstitu ted Am id ocycloh exa d ien esa
Resu lts a n d Discu ssion
a
P r ep a r a tion a n d Hom olytic Rea ction s of O-Ben -
zyl Oxim e E t h er Su b st it u t ed Am id ocycloh exa -
d ien es. We found that alkylation of 2′-aminoacetophen-
one oxime occurred at the amino rather than the oxime
group, and hence, the oxime was not suitable as the
starting point. The N-benzyl-2′-aminoacetophenone hy-
drochloride salt 2a was obtained in good yield and
converted to the corresponding O-benzyl oxime ether 3a
by treatment with O-benzylhydoxylamine in ethanol
catalyzed by H2SO4 (Scheme 1). We first attempted the
preparation of carbamoylcyclohexadiene 5a using our
normal procedure5e of slowly adding 1-methylcyclohexa-
2,5-diene-1-carbonyl chloride 4 to a mixture of purified
amine 3a and triethylamine in DCM containing a cata-
lytic amount of DMAP. However, no conversion was
achieved, and therefore, a pyridine solution of 3a (con-
taining DMAP) was treated with a DCM solution of 4,
refluxed, and poured into 6 M HCl finally yielding 86%
of 5a . The 4,5-dimethoxy precursor 5b was prepared in
good yield by a similar three-step sequence.
Key: (i) PhCH2Br, concd HCl, 60 °C; (ii) BnONH2, EtOH, 12
h; (iii) concd HCl; (iv) pyridine, DMAP, reflux.
SCHEME 2. P r ep a r a tion of 3-Su bstitu ted
In d olin on esa
The expected reaction sequence on treatment of 5a
with a peroxide radical initiator is shown in Scheme 2.
The bisallylic activation ensures that initial H-abstrac-
(10) (a) Naito, T.; Honda, Y.; Miyata, O.; Ninomya, I. J . Chem. Soc.,
Perkin. Trans. 1 1995, 19. (b) Miyata, O.; Nishiguchi, A.; Ninomya, I.;
Aoe, K.; Okamura, K.; Naito, T. J . Org. Chem. 2000, 65, 6922. (c)
Miyata, O.; Muroya, K.; Koide, J .; Naito, T. Synlett 1998, 271.
(11) Naito, T.; Kiguchi, T.; Tajiri, K.; Ninomiya, I.; Hiramatsu, H.
Tetrahedron Lett. 1995, 36, 253.
(12) Keck, G. E.; McHardy, S. F.; Murry, J . A. J . Am. Chem. Soc.,
1995, 117, 7289.
(13) Parker, K. A.; Spero, D. M.; VanEpp, J . J . Org. Chem. 1988,
53, 4628.
a
Key: (i) lauroyl peroxide, RSH, PhH, reflux.
(14) Naito, T.; Miyabe, H.; Torieda, M.; Tajiri, K. I. K.; Kiguchi, T.
J . Org. Chem. 1998, 63, 4397.
(15) Miyabe, H.; Kanehira, S.; Kume, K.; Kandori, H.; Naito, T.
Tetrahedron 1998, 54, 5883.
(16) Berndt, A.; Neugebauer, F. A. In Landolt-Bornstein, Magnetic
Properties of Free Radicals; Fischer, H., Ed.; Springer: Berlin, 1987;
Vol. 17c.
(17) Ingold, K. U.; Kaba, P. A. J . Am. Chem. Soc. 1976, 98, 7375.
(18) Roberts, B. P. Chem. Soc. Rev. 1999, 28, 25; Dang, H.-S.;
Elsegood, M. R. J .; Kim, K.-M.; Roberts, B. P. J . Chem. Soc., Perkin
Trans. 1 1999, 2061. Fielding, A. J .; Roberts, B. P. Tetrahedron Lett.
2001, 42, 4061.
tion will take place regiospecifically from the cyclohexa-
diene moiety to generate cyclohexadienyl radical 6. In
refluxing benzene, radical 6 will undergo rapid â-scission
to afford carbamoyl radical 7 together with toluene as a
benign and easily removed byproduct. 5-Exo-ring closure
of 7 (C5x) should produce the indolinylaminyl radical 8,
which will abstract an H-atom from more 5 and, hence,
propagate a chain reaction with 3-substituted indoline-
J . Org. Chem, Vol. 69, No. 18, 2004 5927

Bella et al.
TABLE 1. EP R P a r a m eter s (9 GHz) of Ra d ica ls Der ived
fr om Ca r ba m oylcycloh exa d ien es 5 in Nea t DTBP
induced reaction of 5a using dibenzoyl peroxide (BPO)
in refluxing benzene. However, NMR and GC-MS analy-
ses showed only products from initiator breakdown, even
when the reaction was carried out in different solvents
and at higher temperatures. We reasoned that addition
of a good H-atom donor would enable the ring closed
aminyl type radical 8 to be trapped more easily. Accord-
ingly, we carried out a reaction using dilauroyl peroxide
as initiator and including a catalytic amount of methyl
thioglycolate (RSH). After a 30 h reflux of a benzene
solution of 5a , 1-benzyl-3-[(benzyloxy)amino]-3-methyl-
1,3-dihydro-2H-indol-2-one (8a ) was isolated in 68% yield.
It is likely that the reaction is subject to polarity reversal
catalysis18 under these conditions. The electrophilic sul-
fanyl radical RS•, derived from methyl thioglycolate, will
preferentially abstract hydrogen from the bisallylic site
of 5, due to a favorable polar effect, to produce 6 and
regenerate RSH. After ring closure, alkoxyaminyl radical
8 will readily abstract hydrogen from RSH thus leading
to formation of 10 and release more RS• to continue the
chain. Photolytic radical fragmentation of 5a was also
examined using neat DTBP and UV photolysis for 3 h at
100 °C. GC-MS analysis showed that 10a was the sole
product (apart from initiator derived material). Product
isolation was more difficult with this procedure and hence
it was less convenient for preparative purposes. None of
the dihydroquinazolin-2-one derivative, that would be
formed from the 6-endo-radical 9a , was isolated or
detected by GC-MS from the thermal or photochemical
reactions of 5a or 5b. It follows that exclusive selectivity
for the 5-exo-closure of the carbamoyl radical onto the
C-atom of the CdN bond prevailed. The overall process
amounts to a clean route leading from 2′-aminoacetophen-
ones to 1,3-dihydroindol-2-ones in four steps. The prod-
ucts contain nitrogen functionality at C-3 ready for
further FG manipulations.
P r ep a r a tion a n d Hom olytic Rea ction s of O-Tr ityl
Oxim e Eth er Su bstitu ted Am id ocycloh exa d ien es.
Clive and Subedi described stannane induced radical ring
closures onto O-trityl oxime ethers.19 These cyclizations
were followed immediately by â-scission of the O-C
oxime ether bond and release of the persistent trityl
radical. The advantage of this tactic is that nitroso or
oxime functionality is thereby introduced into the N-
heterocycle, i.e., the process amounts to a partial auto
deprotection. To discover if this reaction would work in
our tin-free system we made O-tritylhydroxylamine by
the method of Lutz20 and reacted it with amino ketone
2a . The O-trityl oxime ether 11a was obtained in 84%
yield by warming an ethanol solution of 2a and O-
tritylhydroxylamine to which a drop of H2SO4 had been
added (Scheme 3). Temperature control was important
because at higher temperatures only triphenylmethane
and N-benzyl-2′-aminoacetophenone oxime were ob-
tained.21 The desired cyclohexadiene derivative 12a was
prepared in 82% yield on reaction with 1-methylcyclo-
hexa-2,5-diene-1 carbonyl chloride 4 using our standard
conditions. Attempts to prepare the 4,5-dimethoxy ana-
logue 11b were not successful. Reaction of 2b with
O-tritylhydroxylamine did not take place at T < 60 °C
radical
T/K
g-factor
hfs/G
6a
260
2.0026
a(1H) 12.5
a(2H) 9.0
a(2H) 2.6
a(1H) 12.4
a(2H) 9.0
a(2H) 2.6
a(N) 13.6
a(2H) 2.6
a(N) 13.7
a(2H) 2.5
6b
255
2.0026
8a
8b
310
310
2.0047
2.0044
2-one derivative 10 as the end product. Competition from
6-endo-cyclization (C6n ) of carbamoyl radical 7 to generate
dihydroquinazolin-2-one intermediate 9 was a potential
complication. Literature precedent suggested 5-exo mode
ring closure, regiospecific for the C-atom of the imine
bond, should predominate (see above). However, radical
9 obtained from the 6-endo mode would be tertiary
benzylic and hence would be strongly thermodynamically
stabilized.
To establish which of these reaction channels would
prevail we first examined the carbamoylcyclohexadienes
using EPR spectroscopy. When 5a (ca. 5 mg) in di-tert-
butyl peroxide (DTBP, 0.5 mL) in a quartz tube was
degassed and photolyzed with UV light from a 500 W Hg
lamp in the resonant cavity of an EPR spectrometer at
230 K a spectrum with EPR parameters appropriate for
cyclohexadienyl radical 6a was obtained (see the Sup-
porting Information and Table 1).
The observed hyperfine splittings (hfs) and g-factor
were similar to those of related cyclohexadienyl radi-
cals.7,16 When the temperature of the EPR cavity was
increased the spectrum of 6a weakened and at 310 K it
was entirely replaced by a spectrum consisting of a 1:1:1
triplet of triplets. The hfs and g-factor (Table 1) were
comparable to those of alkoxyaminyl radicals reported
in the literature (cf. PhCH2N‚OCH2Ph, g ) 2.0046, a(N)
) 14.4 G, a(OCH2) ) 2.6 G, a(CH2Ph) ) 23.8 G at 210
K)17 and we attribute this spectrum to the cyclized aminyl
radical 8a . On extinguishing the light, radical 8a per-
sisted for several minutes, as would be expected because
of its appreciable steric shielding. The carbamoyl radical
7a was not detectable at intermediate temperatures,
suggesting that the 5-exo cyclization was very rapid.
The temperature above which radical 8a first appeared
(ca. 310 K) was similar to the dissociation temperature
of other carbamoylcyclohexadienyl radicals,7b and hence,
the rate constant for release of radical 7 will also be
similar, i.e., in the range 10-100 s-1 at 300 K. No
spectroscopic evidence of the dihydroquinazolin-2-one
radical 9a was obtained at any temperature. However,
benzyl type radicals such as 9a would be difficult to
detect because of the large number of weak, narrow lines
and therefore the spectroscopic evidence did not conclu-
sively rule out minor 6-endo cyclization. A very similar
set of EPR spectra was obtained from the dimethoxy-
compound 5b and the EPR parameters of radicals 6b and
8b are also listed in Table 1.
(19) Clive, D. L. J .; Subedi, R. Chem. Commun. 2000, 237.
(20) Lutz, W. B. J . Org. Chem. 1971, 36, 3835.
(21) Probably by homolytic dissociation of the O-C bond of 11a and
subsequent H-abstraction by both fragment radicals.
The spectroscopic evidence indicated that ring closure
and indolinone formation took place rapidly. To test the
process as a synthetic method we examined the radical
5928 J . Org. Chem., Vol. 69, No. 18, 2004

Synthesis of Aminoindolinones via Carbamoylcyclohexadienes
SCHEME 3. P r ep a r a tion of O-Tr ityloxim e Eth er
SCHEME 4. Ra d ica ls Gen er a ted d u r in g In d u ced
Dissocia tion of 12a
Su bstitu ted Am id ocycloh exa d ien esa
a
Key: (i) EtOH, cat. H2SO4, 60 °C; (ii) pyridine, DMAP, reflux.
began to appear and had completely replaced that of
radical 13a by 320 K. The EPR parameters indicated this
spectrum corresponded to an alkoxyaminyl radical. Loss
of toluene from 13a will produce carbamoyl radical 14a .
However, if 14a undergoes rapid 5-exo-cyclization,
alkoxyaminyl radical 15a will be produced. The 1:1:1
triplet is evidently the spectrum of this radical. Species
15a had a long lifetime (many minutes), as would be
expected for radical 15a which is sterically shielded on
both sides of the N-radical center. The 320 K spectrum
showed an additional radical, just to high field of the
central component of the spectrum from 15a , and this
became much stronger in the 330 K spectrum. Compari-
son with simulated spectra (see Supporting Information)
showed this to be the triphenylmethyl radical [g )
2.0026, a(3H) ) 2.7 G, a(6H) ) 2.5 G, a(6H) ) 1.1 G at
330 K] released by scission of the oxime C-O bond of
15a .
On prolonged photolysis at the higher temperatures
the Ph3C• radical spectrum weakened and a new spec-
trum appeared (Figure 1, bottom) with g ) 2.0064, a(N)
) 7.4 G. These parameters are characteristic of an acyl
nitroxide [RC(O)N(O•)R′] that is most likely formed by
capture of carbamoyl radical 14a (which is being continu-
ously formed during the photolysis) by the nitroso
compound 16a, thus producing acyl nitroxide 17a (Scheme
4). After still further irradiation the spectrum became
much more complex due to many overlapping peaks from
several nitroxides (18a ) formed by nitroso-compound 16a
trapping other radicals in the system.
F IGURE 1. EPR spectra obtained on photolysis of 12a in neat
DTBP. Top spectrum at 230 K through 260 K, 320 K, and 330
K to the bottom spectrum at 310 K; after prolonged photolysis.
and at higher temperatures only thermal dissociation
products were obtained.
The crystal and molecular structures of amide 12a
were obtained by X-ray diffraction (Supporting Informa-
tion) and show that the oxime CdN bond of 12a was
formed as the E-isomer (trans), unlike most oximes which
are generally obtained as cis/ trans mixtures. The C(1)-
C(7) bond to the cyclohexadiene was comparatively long
(1.555(4) Å) as was the O(14)-C(15) bond of the oxime
ether (1.458(3) Å). These bonds undergo homolysis during
the course of the reaction so the precursor structure
shows they are well adapted for this purpose.
When a degassed solution of 12a in neat DTBP (1.7
mM) was photolyzed in the resonant cavity of the EPR
spectrometer the remarkable sequence of spectra shown
in Figure 1 was observed as the temperature was raised.
The EPR parameters (g ) 2.0026, a(1H) ) 12.3 G, a(2H)
) 9.1 G, a(2H) ) 2.7 G at 230 K) of the radical observed
at 230 K (Figure 1, top) show it to be the cyclohexadienyl
radical 13a (Scheme 4). At 260 K a new spectrum
consisting of a 1:1:1 triplet (g ) 2.0047, a(1N) ) 14.0 G)
The induced decomposition of 12a (Scheme 4) will not
be a chain reaction because the persistent trityl and
alkoxyaminyl radicals (15a ) (or nitroxides 17a , 18a ) will
J . Org. Chem, Vol. 69, No. 18, 2004 5929

Bella et al.
SCHEME 5. P r op osed Mech a n ism for F or m a tion
of In d olin on e 21a
Con clu sion s
O-Benzyloxime ether substituted amidocyclohexa-
dienes may be prepared in three steps in good yields from
2-carbonylanilines. EPR spectroscopic observations and
product analyses showed that peroxide induced decom-
positions of model compounds 5a ,b led to indolin-2-ones
10a ,b with benzyloxyaminyl substitution at their 3-posi-
tions. The cyclization steps were very rapid and took
place regioselectively at the C-atoms of the CdN bonds,
by 5-exo ring closures. No evidence for even minor 6-endo
ring closure was obtained, even though the resultant
dihydroquinazolin-2-one intermediate 9 would be strongly
resonance stabilized. This finding showed that carbamoyl
radicals selectively reacted at the iminyl C-atom and
reinforced previous research that found a strong prefer-
ence for ring closure of other C-centered radicals onto the
C-atoms of oxime ethers.8,9
An O-trityloxime ether analogue 12a was also pre-
pared. The cyclohexadienyl intermediate 13a smoothly
yielded alkoxylaminyl radical 15a again by rapid 5-exo-
cyclization. However, ring closure was quickly followed
by another â-scission step that released the persistent
trityl radical and 3-nitrosoindolinone 16a . EPR spectro-
scopic evidence showed that 16a could trap other tran-
sient intermediates to afford a series of nitroxides. Both
the trityl and alkoxyaminyl radicals were persistent so
the product mixture might have been dominated by their
coupling with transient radicals if the process were
controlled by the PRE. However, GC-MS analyses of
products formed in reactions including methyl thiogly-
colate, showed a range of products including one derived
from the indolinone moiety that had lost the N-function-
ality at the 3-position. Cross-coupling of transient inter-
mediates with trityl or radical 15a was not detected. The
lack of control by the PRE effect was probably due to the
presence of the thiol which donated H-atoms very ef-
ficiently. For example, triphenylmethane, as well as trityl
radical dimer, were the main products derived from the
trityl radical. The complex range of products meant that
this method was not useful for preparative purposes.
Note however, that the 3-methyl substituent in 16a
blocks tautomerism of the nitroso-compound to the cor-
responding oxime. For precursors lacking this methyl
group the oxime would be the main product so that
subsequent trapping reactions, as shown for 16a in
Scheme 4, would not complicate the process.
not readily abstract an H-atom from the precursor
cyclohexadiene 12a . However, by using one equivalent
of initiator, we expected to be able to drive the reaction
far enough for preparative purposes. Because trityl and
radical 15a are persistent species there was a possibility
that the system would be controlled by the persistent
radical effect (PRE);22 i.e., the main products would arise
from coupling of 15a and/or trityl with transient species
such as 13a or 14a . Attempts to induce reaction of 12a
with BPO in refluxing benzene and in refluxing toluene
were not successful. However, addition of a benzene
solution of dilauroyl peroxide and methyl thioglycolate
over 8 h to a refluxing benzene solution containing 12a ,
with continued heating for 24 h, caused essentially
complete consumption of the reactant. Attempts to isolate
products by chromatography yielded only intractable
mixtures. Analysis of the total product mixture by GC-
MS showed a range of products including one having M+
) 237, C16H15NO, that was probably 1-benzyl-3-methyl-
1,3-dihydro-2H-indol-2-one (21a ) accompanied by tri-
phenylmethane, trityl radical dimer and products derived
solely from dilauroyl peroxide and methyl thioglycolate.
Similarly, indolinone 21a , Ph3CH, and trityl dimer
were the main products from a photolytically initiated
reaction of 12a in neat DTBP. However, when the
photolysis was carried out with tert-butylbenzene as
cosolvent GC-MS showed the main product had M+
)
266, C16H14N2O2 and was probably nitroso compound 16a ,
accompanied by Ph3CH. The intriguing aspect of these
results was that product 21a , formed under thermal
conditions, had lost the N-functionality at position 3.
Scheme 5 shows a plausible mechanistic rationale of this
finding. Nitroso-compound 16a will very easily trap other
radicals to produce tertiary nitroxides 19a . EPR spec-
troscopic evidence confirmed this (vide supra). Fragmen-
tation of 19a by scission of the C-N bond to generate
radical 20a , together with a new nitroso-compound, can
readily be envisaged. Radical 20a is tertiary, benzylic and
has an R-carbonyl group so it will be strongly stabilized
by resonance delocalization and by steric shielding. These
thermodynamic factors probably provide the driving force
for this â-scission process. Radical 20a can then pick up
an H-atom from RSH or solvent, thus affording the
observed product 21a .
Exp er im en ta l Section
EP R Sp ectr a . EPR spectra were obtained with a Bruker
EMX 10/12 spectrometer operating at 9.5 GHz with 100 kHz
modulation. Solutions were made up in 4 mm o.d. quartz tubes
and deaerated by passing nitrogen for 20 min. Radicals were
generated by irradiation, directly in the EPR cavity, with
unfiltered light from a 500W super pressure Hg arc. In all
cases where spectra were obtained, hfs were assigned with the
aid of computer simulations using the Bruker Simfonia
software package. Signals were double integrated using the
Bruker WinEPR software and radical concentrations were
estimated by reference to a known concentration of DPPH.
(2-Acetylp h en yl)ben zyla m m on iu m Ch lor id e 2a . A mix-
ture of 1-(2-aminophenyl)ethanone 1a (5 g, 37 mmol) and
benzyl bromide (3.16 g, 18.5 mmol) was heated at 50 °C for 3
h to afford a dark yellow crystalline mass. The solid cake was
dissolved in hydrochloric acid (36% water solution), reprecipi-
(22) Fischer, H. Chem. Rev. 2001, 101, 3581.
5930 J . Org. Chem., Vol. 69, No. 18, 2004

Synthesis of Aminoindolinones via Carbamoylcyclohexadienes
tated into water (300 mL), and filtered. The yellow solid was
washed with water and further purified by crystallization
using a mixture of ethanol-petroleum ether to yield 2a as pale
yellow crystals (3 g, 73%): mp 72-74 °C; 1H NMR δ 2.61 (3H,
s, CH3), 4.46 (2H, s, CH2), 6.58-6.66 (2H, m, ArH), 7.25-7.34
(6H, m, ArH) 7.76-7.78 (1H, m, ArH), 9.32 (1H, br s; NH);
13C NMR δ 28.0 (CH3), 46.7 (CH2), 112.2, 114.4 (2 × CH), 117.8
(C), 126.9, 127.1, 128.6, 132.6, 135.0 (7 × CH), 138.6 (C), 150.8
For the corresponding amine: CIMS m/z 226 (M + H)+, (100);
HRMS (CI) calcd for C15H16NO (M + H)+ 226.1231, found
226.1221. Anal. Calcd for C15H15NO: C, 79.97; H, 6.71, N, 6.22.
Found: C, 79.41, H, 6.68. N, 6.16.
were those derived from the photolytic breakdown of DTBP.
When a sample containing amide 5a (0.05 g) in DTBP (0.05
mL) was heated at 100 °C and irradiated for 3 h with UV
light GC-MS analysis of the reaction mixture showed only
the presence of 10a , together with byproducts derived from
photolytic breakdown of DTBP.
1-Ben zyl-3-[(ben zyloxy)a m in o]-3-m eth yl-1,3-d ih yd r o-
2H-in d ol-2-on e 10a . Dilauroyl peroxide (0.13 g, 0.33 mmol)
was dissolved in benzene (3 mL). The solution was divided into
four portions (0.75 mL), and a first portion was refluxed for
10 min before addition of methyl thioglycolate (0.05 mL). The
remaining portions were added (1 portion/3 h) to a refluxing
benzene solution containing amide 5a (0.1 g, 0.22 mmol). After
complete addition, the solution was refluxed for 30 h and the
solvent evaporated at reduced pressure. GC-MS: peak tR 3.6
min; methyl thioglycolate, peak tR 10.3 min; undecane (from
initiator), peak tR 16.1 min; disulfide (dimer of thiol), peak tR
17.1 min; lauric acid (from initiator), peak tR 23.5 min;
docosane (from initiator), peak tR 28.3 min; 1-benzyl-3-[(ben-
zyloxy)amino]-3-methyl-1,3-dihydro-2H-indol-2-one 10a . The
total reaction product was dissolved into DCM (20 mL) and
treated with a warm 6 M solution of KOH (50 mL). The
aqueous phase was extracted with DCM (3 × 25 mL) and the
combined organic layer dried with magnesium sulfate and
evaporated at reduced pressure to leave a yellow oil which was
purified by column chromatography eluting with a mixture
30% ethyl acetate in hexane (Rf ) 0.33) to give a colorless oil
which was crystallized from ethyl acetate/hexane to afford 10a
as white crystals (0.07 g, 68%): mp 99-100 °C; 1H NMR δ
1.40 (3H, s, CH3), 4.36 and 4.48 (2H, AB, J ) 10.8 Hz, CH2),
4.70 and 5.16 (2H, AB, J ) 16.0 Hz, CH2), 6.26 (1H, br s, NH),
6.55-6.89 (2H, m, ArH), 7.03-7.24 (11H, m, ArH)), 7.44-7.46
(1H, m, ArH)); 13C NMR δ 20.2 (CH3) 41.5 (CH2), 66.0 (C), 77.7
(CH2), 109.8 (CH), 123.0 (CH), 124.0 (CH), 127.2, 127.6, 128.0,
128.5, 128.9, 129.0, 129.2, 129.4, (11 × CH), 130.8 (C), 135.6
(C), 137.2 (C), 143.5 (C), 179.1 (CdO); IR (Nujol) νmax 3229
and 3214 (NH), 1718 (CdO) cm-1; EIMS m/z 358 (M+, 15), 327
(20), 236 (53), 91 (100), 65 (7); HRMS (EI) calcd for C23H22N2O2
(M)+ 358.1681, found 358.1674. Anal. Calcd for C23H22N2O2:
C, 77.07; H, 6.19, N, 7.82. Found: C, 77.02, H, 6.02, N, 7.49.
(2-Acetyl-4,5-dim eth oxyph en yl)ben zylam m on iu m Ch lo-
r id e 2b. A stirred solution of 1-(2-amino-4,5-dimethoxy-
phenyl)ethanone 1b (5 g, 26 mmol) and benzyl bromide (2.88
g, 16.8 mmol) in DMF was heated at 60-70 °C for 3 h. The
reaction mixture was poured into water (300 mL), acidified
with HCl (36%, 10 mL), and stirred overnight. A yellow solid
separated from the aqueous phase and was filtered to give a
red wine solid which was crystallized from a mixture of ethanol
and petroleum ether to yield 2b as pale yellow crystals (3.8 g,
70%): mp 118-120 °C; 1H NMR δ 2.54 (3H, s, CH3), 3.74 (3H,
s, CH3), 3.82 (3H, s, CH3), 4.46 (2H, s, CH2), 6.08 (1H, s, ArH)
7.17 (1H, s, ArH) 7.26-7.34 (5H, m, ArH), 9.51 (1H, br, s, NH);
13C NMR δ 28.2 (CH3), 47.5 (CH2), 55.9 (CH3), 57.3 (CH3), 95.2
(CH), 110.4 (C), 115.7(CH), 127.4, 127.6, 129.1 (5 × CH), 139.1
(C), 139.3 (C), 149.1 (C), 156.3 (C), 198.6 (CdO); IR (Nujol)
νmax 3293 (NH), 1622 (CdO) cm-1. For the corresponding
amine: LRMS (EI) m/z 285 ((M + H)+, 46), 91 (100); HRMS
(EI), calcd for C17H19NO3 (M+) 285.1365, found 285.1369. Anal.
Calcd for C17H19NO3: C, 71.56; H, 6.71, N, 4.91. Found: C,
71.45, H, 6.83.N, 4.89.
(C) 201.0 (CdO); IR (Nujol) νmax 3321 (NH), 1641 (CdO) cm-1
.
1-[2-(Ben zyla m in o)p h en yl]eth a n on e O-Ben zyloxim e
3a . A solution of O-benzylhydroxylamine (0.9 g, 7.3 mmol) and
2a (1.5 g, 5.7 mmol) in ethanol (10 mL) was heated to reflux
for 30 min before a drop of sulfuric acid was added. The
solution obtained was refluxed overnight before the solvent
was evaporated under reduced pressure and the slurry residue
made basic with a 6 M NaOH and extracted with DCM (3 ×
20 mL). The combined organic layers were dried with MgSO4
and filtered, and the solvent was evaporated to give a yellow
oil. Purification was performed by column chromatography (Rf
) 0.74) (alumina, eluting with 10% ethyl acetate in hexane)
to give a colorless oil. Crystallization in hexane afforded 3a
1
as a white solid (1.5 g, 79%): mp 52-54 °C; H NMR δ 2.33
(3H, s, CH3), 4.29 (2H, 2 × s, CH2), 5.01 (2H, s, CH2), 6.56-
6.64 (2H, m, ArH) 7.20-7.38 (12H, m, ArH), 7.95 (1H, br s;
NH); 13C NMR δ 13.2 (CH3), 47.4 (CH2), 75.9 (CH2) 111.1 (CH),
115.1 (CH), 117.5 (C), 126.9-129.8, (12 × CH), 138.0 (C), 139.4
(C), 146.8 (C), 157.8 (CdN); IR (Nujol) νmax 3305 (NH), 1603
(CdN) cm-1; CIMS m/z 331 [(M + H)+, (100)]; HRMS (CI) calcd
for C22H23N2O (M + 1)+ 331.1810, found 331.1802. Anal. Calcd
for C22H22N2O: C, 79.97; H, 6.71; N, 8.48. Found: C, 79.59;
H, 7.11; N, 8.50.
N-Ben zyl-N-[2-[N-(ben zyloxy)eth a n im id oyl]p h en yl]-1-
m eth yl-2,5-cycloh exa d ien e-1-ca r boxa m id e 5a . 1-Methyl-
cyclohexa-2,5-diene-1-carbonyl chloride (0.85 g, 5.4 mmol) was
dissolved in dry DCM (5 mL) and added dropwise to a mixture
of 3a (1.2 g, 3.6 mmol) and a catalytic amount of DMAP in
dry distilled pyridine (20 mL). The resultant mixture was
refluxed overnight before addition of HCl (6 M, 50 mL). The
aqueous layer was extracted twice with DCM (2 × 100 mL),
and the combined organic layer was washed with water
containing a little NaHCO3 (20 mL) and water (20 mL) and
dried over MgSO4. The solvent was evaporated at reduced
pressure to give a brown oil, which was purified by column
chromatography, eluting with 5% ethyl acetate in hexane, to
furnish the amide 5a as a pale yellow oil which crystallized
from a mixture ethyl acetate/hexane as colorless crystals (1.4
1
g, 86%): mp 62-64 °C; H NMR δ 1.33 (3H, s, CH3), 1.92-
2.30 (2H, AB, CH2), 2.19 (3H, s, CH3), 3.73 and 5.43 (2H, AX,
J ) 14.3 Hz, CH2), 4.80-5.67 (4H, m, dCH), 5.24 (2H, s, CH2),
6.77-7.40 (14H, m, ArH); 13C NMR δ 15.0 (CH3) 25.7 (CH2),
29.5 (CH3), 45.9 (C), 55.0 (CH2), 76.2 (CH2), 119.8 (CH), 123.1,
126.8, 127.4, 127.6, 127.9, 127.9, 128.2, 128.7, 128.9, (12 × CH),
129.4, 131.65 131.8 (3 × CH), 134.8 (C), 136.6 (C), 137.8 (C),
139.7 (C), 153.4 (CdN), 174.1 (CdO); IR (Nujol) νmax: 1630
cm-1 (CdO); CIMS m/z 451 (M+ + 1, 25); HRMS (CI) calcd for
C
30H31N2O2 (M + H)+ 451.2385, found 451.2401.
P h otoch em ica lly In itia ted Rea ction of N-Ben zyl-N-[2-
1-[2-(Ben zyla m in o)-4,5-d im eth oxyp h en yl]eth a n on e O-
ben zyloxim e 3b. A solution of O-benzylhydroxylamine (0.7
g, 5.7 mmol) and 2b (1.0 g, 3.1 mmol) in ethanol (10 mL) was
heated to gentle reflux for 10 minutes before a drop of sulfuric
acid was added. The solution obtained was refluxed overnight
before the solvent was evaporated under reduced pressure and
the residue made basic by a 6M NaOH and extracted with
DCM (3 × 20 mL). The combined organic layer was dried with
MgSO4, filtered and the solvent eliminated at reduced pressure
to give a yellow oil. Purification by column chromatography
(Rf ) 0.42), (alumina, eluting with 20% ethyl acetate in
hexane) and crystallization from a mixture of pentane/ethyl
[N-(ben zyloxy)eth a n im id oyl]p h en yl]-1-m eth yl-2,5-cyclo-
h exa d ien e-1-ca r boxa m id e 5a . Amide 5a (0.05 g) was dis-
solved in neat DTBP (0.5 mL), the resultant solution was
placed in a quartz tube, and the sample was irradiated, at
room temperature, with light from a 400 W medium-pressure
Hg lamp over a 3 h period. Analysis of the reaction mixture
by GC-MS confirmed the presence of the 5-exo product
1-benzyl-3-[(benzyloxy)amino]-3-methyl-1,3-dihydro-2H-indol-
2-one 10a : peak tR 34.33 min, m/z 358 (M+, 5), 357 (40), 327
(5), 281 (20), 236 (16), 207 (43), 159 (10), 91 (100), together
with unreacted amide 5a . The only detectable impurities
J . Org. Chem, Vol. 69, No. 18, 2004 5931

Bella et al.
acetate afforded 3b as pale yellow crystals (1.0 g, 82%); mp:
82-84 °C, 1H NMR δ 2.31 (3H, s, CH3), 3.72 (3H, s, CH3), 3.79
(3H, s, CH3), 4.26 (2H, s, CH2), 5.06 (2H, s, CH2), 6.13 (1H, s,
ArH), 6.93 (1H, s, ArH), 7.22-7.28 (10H, m, ArH), 7.91 (1H,
br, s, NH); 13C NMR δ 13.7 (CH3), 48.4 (CH2), 55.9 (CH3), 57.8
(CH3), 76.2 (CH2), 96.2 (CH), 109.6 (C), 114.6 (CH), 127.4,
127.7, 128.2, 128.6, 128.8, 128.9 (10 × CH), 138.6 (C), 139.7
(C), 140.0 (C), 143.6 (C), 151.5 (C), 157.7 (CdN); IR (Nujol)
1-[2-(Ben zylam in o)ph en yl]eth an on e O-tr ityloxim e 11a.
To a stirred solution containing (2-acetyl-phenyl)benzylam-
monium chloride 2a (1.0 g, 4.4 mmol) in warm ethanol (10 mL),
was added portionwise O-tritylhydroxylamine (1.8 g, 6.5
mmol). After complete addition the mixture was heated at 60
°C and stirred for 10 min before adding a drop of sulfuric acid.
When a white solid separated from solution, the temperature
was lowered to room temperature and the solution stirred for
20 min. The reaction mixture was filtered and the solid cake
was extensively washed with ethanol and dried under vacuum,
to give white crystals of 11a (1.8 g 84%); mp: 160-162 °C; 1H
NMR δ 2.54 (3H, s, CH3), 3.93 (2H, 2 × s, CH2), 6.35 (1H, d,
ArH), 6.57 (1H, t, ArH), 6.96-7.01 (1H, m, ArH), 7.17-7.42
(21H, m, ArH)), 7.49 (1H, br s; NH); 13C NMR δ 13.8 (CH3),
46.5 (CH2), 91.2 (C-O), 111.1 (CH), 114.7 (CH), 117.8 (C),
126.4, 126.6, 127.1, 127.4, 127.7, 128.2, 128.9, 129.3, 129.7,
(22 × CH), 139.7 (C), 143.0 (C), 144.7 (C), 146.7 (C), 157.1 (Cd
N); IR (Nujol) νmax: 3305 (NH), 1603 (CdN), cm-1; LRMS (ES)
m/z 483 [(M + H)+, (16)], 243 (100), HRMS calcd for C34H31N2O
(M+H)+ 483.2436, found 483.2444. The reaction was also
carried out in refluxing ethanol for 10 h, the solvent was
evaporated under reduced pressure and the slurry residue
made basic by a 6M NaOH and extracted with DCM (3 × 20
mL). The combined organic layers were dried with MgSO4,
filtered and the solvent evaporated to give a pale yellow oil.
Separation of the products was performed by column chroma-
tography (alumina), eluting with 10% ethyl acetate in hexane
to give products deriving from thermal cleavage of 3a , i.e.,
trans-1-[2-(benzylamino) phenyl] ethanone oxime (major iso-
ν
max: 3284 (NH), 1624 (CdN) cm-1; EIMS m/z, 390 (M)+, (100)],
300 (55), 283 (87), 268 (14), 205 (7), 193 (12), 178 (15), 91 (60),
77 (6). HRMS calcd for C24H26N2O3 (M+) 390.1943, found
390.1930. Anal. Calcd for C24H26N2O3: C, 73.82; H, 6.71, N,
7.17. Found: C, 73.75, H, 6.74, N, 7.14.
N -Be n zyl-N -{2-[N -(b e n zyloxy)e t h a n im id oyl]-4,5-d i-
m eth oxyp h en yl}-1-m eth yl-2,5-cycloh exa d ien e-1-ca r box-
a m id e 5b. 1-Methylcyclohexa-2,5-diene-1-carbonyl chloride
(0.5 g, 3.2 mmol) was dissolved in dry dichloromethane (5 mL)
and added dropwise to a mixture of 3b (0.8 g, 2.0 mmol) and
a catalytic amount of DMAP in dry distilled pyridine (20 mL).
The resultant mixture was refluxed overnight before adding
HCl (6M, 50 mL). The aqueous layer was extracted with DCM
(2 × 100 mL), the combined organic layer was washed with
water containing a little NaHCO3 (20 mL), water (20 mL) and
dried over MgSO4. The solvent was evaporated at reduced
pressure to give a yellow oil, which was purified by column
chromatography, eluting with ethyl acetate in hexane (1:9f3:
7) in order to furnish amide 5b as a pale yellow oil (0.7 g, 87%);
1H NMR δ 1.25 (3H, s, CH3), 1.95-2.35 (2H, m, CH2), 2.19
(3H, s, CH3), 3.47 (3H, s, CH3), 3.69 and 5.51 (2H, AX, J )
14.0 Hz, CH2), 3.87 (3H, s, CH3), 4.84-5.71 (4H, m, CH), 5.23
(2H, s, CH2), 6.10 (1H, s, ArH), 6.75 (1H, s ArH) 7.09-7.41
(10H, m, ArH)); 13C NMR δ 15.6 (CH3) 26.1 (CH2), 30.0 (CH3),
46.4 (C), 55.2 (CH2), 55.8 (CH3), 56.5 (CH3), 76.6 (CH2), 111.4
(CH), 115.4 (CH), 119.3 (CH), 123.7 (2 × CH), 127.0 (C), 127.4-
129.8, (9 × CH), 132.58 (2 × CH), 133.2 (C), 137.8 (C), 137.9
(C), 147.6 (C), 148.0 (C), 153.8 (CdN), 174.5 (CdO); IR (Nujol)
1
mer) as pale yellow crystals: mp 52-54 °C; H NMR δ 2.33
(3H, s, CH3), 4.40 (2H, m, CH2), 6.62-6.70 (2H, m, ArH), 7.12-
7.41 (7H, ArH), 7.79 (1H, br s, NH); 13C NMR δ 12.5 (CH3),
47.5 (CH2), 111.3 (CH), 115.4 (CH), 117.9 (C), 126.9, 127.1,
128.5, 129.0, 129.9 (CH), 139.4 (C), 146.8 (C), 158.9 (CdN);
IR (Nujol) νmax 3337 (NH), 1604 (CdN), cm-1; EIMS m/z 240
[(M)+, 45), 223(100), HRMS (EI) calcd for C15H16N2O (M+)
240.1262, found 240.1269. cis-1-[2-(Benzylamino) phenyl]-
ethanone oxime as white crystals: 1H NMR δ 2.41 (3H, s, CH3),
4.21(2H, s, CH2), 7.05-7.64 (9H, m, ArH), 8.26 (1H, br s, NH);
13C NMR δ 13.9 (CH3), 47.0 (CH2), 109.3 (CH), 119.0, 121.9,
122.2, 126.1, 127.8, 128.9 (8 × CH), 135.5 (C), 135.7 (C), 142.5
(C), 151.8 (CdN). Triphenylmethane: 13C NMR δ 56.8 (CH),
126.2, 128.2, 129.4 (15 × CH), 143.8 (3 × C).
ν
max: 1638 (CdO), 1518 (CdC), 1260 (CdC), cm-1 CIMS m/z
511 ((M+H)+, 100); HRMS (CI) calcd for C32H35N2O4 (M + H)+
511.2597, found 511.2607. Anal. Calcd for C32H34N2O4: C,
75.27; H, 6.71, N, 5.49. Found: C, 74.75, H, 6.73, N, 6.35.
1-Ben zyl-3-[(ben zyloxy)a m in o]-5,6-d im eth oxy-3-m eth -
yl-1,3-d ih yd r o-2H-in d ol-2-on e 10b. To a stirred refluxing
solution of amide 5b (0.1 g, 0.25 mmol) in benzene (5 mL) was
added 0.5 mL of a solution prepared by dissolving dilauroyl
peroxide (0.12 g, 0.3 mmol) in benzene (2.5 mL). After refluxing
for 5 min methyl thioglycolate (0.05 mL) was added. The
remaining dilauroyl peroxide solution was then added over 8
h. After complete addition, the solution was refluxed for 30 h,
the solvent was evaporated at reduced pressure and the
product was analyzed by GC-MS; peak tR 3.6 min; methyl
thioglycolate, peak tR 10.3 min; undecane (from initiator), peak
tR 16.1 min; disulfide (dimer of thiol), peak tR 17.1 min; lauric
acid (from initiator), peak tR 23.5 min; docosane (from initia-
tor), peak tR 28.2 min; 10b. The total reaction product was
dissolved in DCM (20 mL) and treated with a warm 6M
solution of KOH (50 mL). The aqueous phase was extracted
with DCM (3 × 25 mL) and the combined organic layers dried
with magnesium sulfate and evaporated at reduced pressure
to leave a brown solid which was purified by column chroma-
tography eluting with a mixture 40% ethyl acetate in hexane
N-Ben zyl-1-m et h yl-N-[2-[N-(t r it yloxy)et h a n im id oyl]-
p h en yl]-2,5-cycloh exa d ien e-1-ca r b oxa m id e 12a . To a
stirred mixture containing 11a (1.4 g, 2.9 mmol) and a catalytic
amount of DMAP in dry distilled pyridine (20 mL) was slowly
added a solution prepared by dissolving 1-methylcyclohexa-
2,5-diene-1-carbonyl chloride (0.67 g, 4.4 mmol) in dry DCM
(5 mL). The resultant mixture was refluxed overnight before
pouring into aqueous HCl (6 M, 50 mL). The aqueous layer
was extracted with DCM (2 × 100 mL), and the combined
organic layers were washed with water containing a little
NaHCO3 (20 mL) and water (20 mL), and dried over MgSO4.
The solvent was evaporated at reduced pressure to leave a
reddish oil, which was purified by column chromatography,
eluting with ethyl acetate 5% in hexane (Rf ) 0.27) in order
to furnish 12a as a white solid which was crystallized from a
mixture ethyl acetate/hexane/DCM affording colorless crystals
(1.43 g, 82%): mp 164-166 °C; 1H NMR δH 1.19 (3H, s, CH3),
1.85-2.23 (2H, AB, CH2), 2.42 (3H, s, CH3), 3.40 and 5.51 (2H,
AX, J gem ) 14.3 Hz, CH2), 4.72-5.64 (4H, m, dCH), 6.53 (1H,
d, ArH), 6.78-7.37 (23H, m, ArH); 13C NMR δ 16.4 (CH3) 25.6
(CH2), 29.8 (CH3), 45.8 (C), 54.4 (CH2), 91.0 (C), 120.1 (CH),
123.1 (2 × CH), 126.8, 127.0, 127.2, 127.6, 127.8, 129.2, 129.6
129.0, (23 × CH), 131.8 (2 × CH), 135.1 (C), 137.5 (C), 138.9
(C), 144.7 (3 × C), 154.4 (CdN), 173.9 (CdO); IR (Nujol) νmax
1632 (CdO), cm-1; ESMS m/z 625 [(M + Na)+, 100); HRMS
(ES) calcd for C42H38N2O2Na (M + Na)+ 625.2831, found
625.2823. Anal. Calcd for C42H38N2O2: C, 83.69; H, 6.35, N,
4.65. Found: C, 83.75, H, 6.49, N, 4.45.
1
(Rf ) 0.28) to give 10b as a yellow pale oil (0.05 g, 63%); H
NMR δ 1.37 (3H, s, CH3), 3.74 (3H, s, CH3), 3.84 (3H, s, CH3),
4.40 and 4.50 (2H, AB, J ) 11.0 Hz, CH2), 4.70 and 5.13 (2H,
AB, J ) 15.9 Hz, CH2), 6.26 (1H, s, NH), 6.93-6.99 (1H, m,
ArH), 7.10-7.26 (11H, m, ArH), 13C NMR δ 19.9 (CH3) 43.6
(CH2), 56.0 (CH3), 56.5 (CH3) 65.6 (C), 76.4 (CH2), 95.3 (CH),
108.1 (CH), 120.8 (C), 126.6, 127.1, 127.4, 127.9, 128.2, 128.4,
(10 × CH), 135.2 (C), 136.5 (C), 136.8 (C), 144.9 (C), 149.6 (C)
178.7 (CdO); IR (neat) νmax: 3235 (NH), 1718 (CdO) cm-1
;
EIMS m/z 358 (M+, 15), 327 (20), 236 (53), 91 (100); HRMS
(EI) calcd for C25H26N2O4 418.1893, found 418.1885.
5932 J . Org. Chem., Vol. 69, No. 18, 2004

Synthesis of Aminoindolinones via Carbamoylcyclohexadienes
Cr ysta l d a ta for 12a : C42H38N2O2, M ) 602.74, colorless
plates, crystal size 0.24 × 0.12 × 0.1 mm, triclinic, space group
P-1, a ) 8.4178(15) Å, b ) 12.781(2) Å, c ) 16.416(3) Å, R )
84.841(3)°, â ) 83.997(3)°, γ ) 72.819(3)°, V ) 1674.9(5) Å3,
Dc ) 1.195 Mg/m3, T ) 125(2) K, R ) 0.0647, Rw 0.0831 for
4539 reflections with I > 2σ(I) and 418 variables. Data were
collected on a Bruker SMART diffractometer with graphite-
monochromated Mo KR radiation (R ) 0.710 73 Å). The
structure was solved by direct methods and refined using full-
matrix least-squares methods. The structure, atomic coordi-
nates, bond lengths, and coordinates are listed in the Sup-
porting Information.
DTBP -Med ia ted P h otolysis of N-Ben zyl-1-m eth yl-N-
[2-[N-(t r it yloxy)et h a n im id oyl]p h en yl]-2,5-cycloh exa d i-
en e-1-ca r boxa m id e 12a . To a solution of amide 12a (0.03 g)
in tert-butylbenzene (200 µL) in a quartz tube was added DTBP
(50 µL). The tube was capped and irradiated, at room tem-
perature, with light from a 400 W high-pressure Hg lamp for
3 h before the DTBP was evaporated to leave a yellow oil which
was analyzed by GC-MS, peak tR 21.0 min; 1-benzyl-3-methyl-
3-nitroso-1,3-dihydro-2H-indol-2-one 16a ; m/z 266 (M+, 11),
119 (100), 91 (35), peak tR 22.0 min; triphenylmethane. The
only byproducts detected were from photolytic breakdown of
DTBP.
thioglycolate (0.05 mL) in benzene (2 mL). After complete
addition, the solution was refluxed for 24 h, the solvent was
evaporated at reduced pressure to give a mixture which was
analyzed by GC-MS. Peak tR 3.6 min; methyl thioglycolate,
peak tR 10.3 min; undecane (from initiator), peak tR 16.1 min;
disulfide (dimer of thiol), peak tR 17.1 min; lauric acid (from
initiator), peak tR 23.5 min; docosane (from initiator), peak tR
21.7 min; triphenylmethane, peak tR 22.1 min; 1-benzyl-3-
methyl-1,3-dihydro-2H-indol-2-one 21a , m/z 237 (M+, 100), 208
(35), 146(35), 128 (30), 117(10), 91(100), 77 (12), 65 (10), peak
tR 23.2 min; triphenylmethanol, peak tR 27.4 min [[4-(diphen-
ylmethylene)-2,5-cyclohexadien-1-yl](diphenyl)methyl]ben-
zene (trityl dimer).
Ack n ow led gm en t. We thank the EPSRC (Grant
No. GR/N38763) for financial support.
Su p p or tin g In for m a tion Ava ila ble: General experimen-
tal methods. EPR spectra of radicals derived from 5a . Experi-
mental and simulated EPR spectra of the trityl radical and
radical 15a . Experimental details for the preparation of 1-(2-
aminophenyl)ethanone oxime, 1-(2-aminophenyl)ethanone O-
benzyloxime, 1-methyl-2,5-cyclohexadiene-1-carbonyl chloride,
N-trityloxyphthalimide hydroxyphthalimide, and trityloxyamine.
X-ray structure data for compound 12a . 13C NMR spectra of
5a , 10b, and 11a . This material is available free of charge via
the Internet at http://pubs.acs.org.
Dila u r oyl P er oxid e Med ia ted Th er m olysis of N-Ben -
zyl-1-m et h yl-N-[2-[N-(t r it yloxy)et h a n im id oyl]p h en yl]-
2,5-cycloh exa d ien e-1-ca r boxa m id e 12a . To a refluxing
benzene solution (5 mL) containing amide 12a (0.1 g, 0.16
mmol) was added over 8 h by syringe pump a solution prepared
by dissolving dilauroyl peroxide (0.12 g, 0.3 mmol) and methyl
J O049617F
J . Org. Chem, Vol. 69, No. 18, 2004 5933