Claimed 2,1-benzisoxazoles are indazalones

Article abstract of DOI:10.1021/jo048153i

(Chemical Equation Presented) Claims, by two groups, to have prepared 2,1-benzisoxazole derivatives are corrected to show that the products are indazalones (5). In addition, a simple preparation of 3-oxysubstituted 2H-indazole, by an unrecognized method i

Full text of DOI:10.1021/jo048153i

Claimed 2,1-Benzisoxazoles Are
Indazalones
SCHEME 1
Mark J. Kurth,*,† Marilyn M. Olmstead, and
†
Makhluf J. Haddadin§
Department of Chemistry, University of California, Davis,
One Shields Avenue, Davis, California 95616, and
Department of Chemistry, American University of Beirut,
Beirut, Lebanon
mjkurth@ucdavis.edu
Received October 19, 2004
Claims, by two groups, to have prepared 2,1-benzisoxazole
derivatives are corrected to show that the products are
indazalones (5). In addition, a simple preparation of 3-oxy-
substituted 2H-indazole, by an unrecognized method in the
literature, is reported.
In relation to another project, we needed to prepare
o-nitrosobenzaldehyde (1) and thus became aware of the
SCHEME 2
1
relatively recent report of Chen and Burka which, in
turn, led us to the work of Boduszek, Halama, and Zon.2
The Chen group claimed that o-nitrosobenzaldehyde (1)
3
reacts with benzylamine (2) in CHCl at room tempera-
ture to yield 3-benzylamino-2,1-benzisoxazole (anthranil
4
or its tautomer 4′) and proposed path a in Scheme 1 to
explain their assignment.
In support of their claim, these researchers cited the
2
finding of the Boduszek group, who reported the syn-
thesis of “4” by two methods (Scheme 2), the second of
which gives an additional product to which they specu-
latively assigned structure 9. Given that secondary
amines are more nucleophilic than secondary alcohols,
we suspected that the product obtained by the Chen
1
group should be the indazalone derivative 5 (-NH
attacking the -NdO, as depicted in Scheme 1/path b)
rather than the reported 2,1-benzisoxazole derivative 4
or 4′ (-OH attacking the -NdO).
We therefore repeated the work of both groups1 with
,2
regard to their common claim of having synthesized
NMR, 13C NMR) with a sample prepared by the method
of Diez-Barra and co-workers (Scheme 3).
3
-benzylamino-2,1-benzisoxazole (4, 4′). It is established
3
below that indeed the methods used by either group lead
to 2-N-benzylindazalone (5; Scheme 1). The structure of
Interestingly, the neat IR of 5 shows extended hydro-
-
1
2
-N-benzylindazalone, obtained by the Chen or the
gen bonding (3100-2500 cm ) and a rather low fre-
1
-1
Boduszek methods, was confirmed by comparison (IR, H
quency, though strong, carbonyl absorption (1619 cm ;
conjugation). Finally, we confirmed the structure of 5
1
*
Address correspondence to this author.
University of California, Davis.
American University of Beirut.
(prepared by the Chen method) by X-ray crystallography.
†
The X-ray crystallographic structure in Figure 1 shows
§
(
1) Chen, L.-J.; Burka, L. T. Tetrahedron Lett. 1998, 39, 5351-5354.
2) Boduszek, B.; Halama, A.; Z o´ n, J. Tetrahedron 1997, 53, 11399-
(
(3) Aran, V. J.; Diez-Barra, E.; de la Hoz, A.; Sanchez-Verdu, P.
Tetrahedron 1997, 45, 129-136. Although there are other methods in
the literature for the preparation of 5, we chose this method because
1
1410. This reference includes more examples of alledged 2,1-ben-
zisoxazoles prepared by the methods cited here in Scheme 2.
1
3
of its simplicity and the C NMR data of the desired product.
10.1021/jo048153i CCC: $30.25 © 2005 American Chemical Society
1060
J. Org. Chem. 2005, 70, 1060-1062
Published on Web 12/31/2004

FIGURE 2.
SCHEME 5
FIGURE 1. X-ray crystallographic structure of 5.
SCHEME 3
SCHEME 4
methanolic KOH should give 10. Indeed, this reaction
proceeded smoothly to afford 10 in 61% yield. While a
carbanion mechanism is proposed in Scheme 4, a radical-
6
based alternative can be envisaged as well. The general-
ity of this simple and unrecognized method for the
clearly the extended H-bonding observed in the IR of 5
preparation of 3-oxy-substituted 2H-indazoles is being
(note: the numbering of the X-ray structure is not related
7
investigated further. In Scheme 1, we propose how 5 is
to the proper name of this compound).
formed from 1 plus 2 (path b), and in Scheme 5, we
suggest what we believe are likely mechanistic pathways
to explain the conversion of 72 to 5 and 10. This
postulated mechanism applies also to the conversion of
Having established 5 as the true structure of “4, 4′”,
we turned our attention to the additional product “9”
2
(
Scheme 2) obtained by the Boduszek group. We pre-
dicted and verified that “9” is actually 3-methoxy-2-
6
(Scheme 2) to 5.
2
benzyl-2H-indazole (10). The Boduszek group attributed
The fact that 10 is formed in low yield (the Boduszek
1
the H NMR methyl signal at δ 4.21 to the NCH
3
protons
2
group reported that “9” “never exceeding 20% even when
the reaction was carried out in pure methanol”) suggests
that it could arise from a minor intermediate such as 12
of “9”, whereas it is more befitting of a methoxy group.
1
3
Indeed, the C NMR of the sample we prepared accord-
2
3
ing to the prescribed procedure showed two sp hybrid-
ized carbons at δ 52.4 and 60.7, which is consistent with
a methoxy carbon and an N-methylene benzylic carbon,
respectively, with nitrogen being part of the 2H-indazole
heteroring. Moreover, brief heating of 10 with 48%
aqueous HBr converted it into 5 in 95% yield (Scheme
(
(
Scheme 5). Finally, it is evident that structure “8”
Scheme 2, 2nd method) assigned to the product of the
2 5
reaction of “4” with monoethyl phosphonic acid (C H -
OPO2
3 2
H ) is structure 13 (Figure 2). The cited spectral
data and the ease of its cleavage to 5 are consistent with
this structural assignment.
4
).
By analogy with our earlier unpublished results with
(
5) Ishikawa, F.; Watanabe, Y.; Saegusa, J. Chem. Pharm. Bull.
4
N-substituted o-nitrobenzylamines, we envisaged that
1980, 28, 1357-1364.
heating 2-nitro(N-benzyl)benzylamine (11)5 with 5%
(6) Bjφrsvik, H,-R.; Gonz a´ lez, R. R.; Liguori, L. J. Org. Chem. 2004,
6
9, 7720-7727.
(
7) For a comprehensive review of the chemistry of indazoles, see:
(
4) Haddadin, M. J.; Nazer, M. Z.; Olmstead, M. M.; Kurth, M. J.
Stadlbauer, W. Product class 2: 1H- and 2H-indazoles. Sci. Synth.
2002, 12, 227-324.
Unpublished results.
J. Org. Chem, Vol. 70, No. 3, 2005 1061

It is evident that we (and others, though unknow-
ingly ) demonstrate here a novel method for preparing
-substituted indazalones. In addition, this method con-
stitutes an easy method for constructing a N-N single
bond.
showed almost complete conversion to the product. Methanol
was evaporated, water was added, and the turbid solution was
extracted with DCM (three 20-mL portions). The DCM extract
was washed with 1 N HCl and washed again with water. The
dried solution was evaporated to yield a yellow-colored oil that
was purified by column chromatography (silica gel, 25 g; elution
with n-hexane/5% ethyl acetate, 200 mL, 10%, 100 mL, 15%,
1
,2
2
2
00 mL). Pale yellow oil, 0.3 g, 61%. The elemental analyses
Experimental Section
2
1
are reported. The reported H NMR and that measured by this
work are identical. IR (neat) 1624 (s), 1526 (vs), 1510 (s), 1406
We confirm the experimental, physical, and spectral data of
the isolated intermediates and products reported by both
-
1 13
(
s), 1385 (vs), 1113 (vs), 737 (s), 701 (s) cm
.
C NMR (CDCl
3
)
1
,2
δ 52.14 (OCH ), 60.53 (NCH C H ), 107.00, 117.89, 119.58,
3 2 6 5
groups leading to 2-benzylindazalone (5). o-Nitrosobenzalde-
8
119.80, 126.26, 127.71, 127.93, 128.80, 136.48, 146.72, 147.48.
hyde was prepared by the Bamberger method, using O
degassed methanol as solvent and sunlight, followed by evapo-
ration of the solvent and chromatography with Al , n-hexane,
and n-hexane/5% ethyl acetate. o-Nitro(N-benzyl)benzylamine
2
-
Conversion of 10 to the 2-Benzylindazalone 5: Compound
1
0 (11 mg, 0.05 mmol) was heated with aqueous 48% HBr (2
2
O
3
mL), with stirring for 1 min, and the cooled solution was diluted
with water (10 mL). A white solid appeared and was collected,
washed with water, and dried (9.8 mg, 95% yield). This com-
pound was identical in all respects with 5.
5
(
11) was prepared by either the literature method or the
following procedure: o-nitrobenzyl bromide (1,4 g, 6.5 mmol) was
alkylated at room temperature with benzylamine (1.5 g, 14
mmol) in THF (10 mL) to give a white salt that was removed by
filtration, the filtrate was evaporated, and the residue was
Acknowledgment. We thank Dr. Yat Fan Suen for
the NMR data and M.J.H. is grateful to the American
University of Beirut for a research leave at the Univer-
sity of California, Davis. We thank the National Science
Foundation for support of this work. The 400- and 300-
MHz NMR spectrometers used in this study were
funded in part by a grant from NSF (CHE-9808183).
chromatographed (neutral Al
% and 5% ethyl acetate) to give 11 (0.5 g, 32% yield).
-Benzylindazalone (5): The procedure described by Chen
2 3
O , 50 g; elution with n-hexane/
3
2
1
and Burka for the reaction of o-nitrosobenzaldehyde (1) and
benzylamine (2) gave 5, not the claimed 3-(N-benzylamino)-2,1-
benzisoxazole 4 (or its tautomer 4′; see Scheme 1).
3
-Methoxy-2-benzyl-2H-indazole (10): o-Nitro(N-benzyl)-
Supporting Information Available: X-ray crystal struc-
tures for 5. This material is available free of charge via the
Internet at http://pubs.acs.org.
benzylamine (11, 0.5 g, 2.1 mmol) was dissolved in 5% metha-
nolic KOH (50 mL). The solution was refluxed for 5 h. TLC
(8) Bamberger, E. Chem. Ber. 1918, 51, 606-12.
JO048153I
1062 J. Org. Chem., Vol. 70, No. 3, 2005