substrate 1a under relatively mild conditions followed by
the development of a method for the quantitative synthesis
of dihydroindazoloisoquinolines 3. We envisaged that both
the products 2a and 3a were the outcome of a single
intermediate with the amine 2a having synthetic preference
over that of the cyclized product 3a when exposed to SnCl2‚
2H2O conditions. Reduction of the nitro group is generally
well-known to proceed through nitroso and hydroxylamine
intermediates.12 Although, Pd-catalyzed reduction (Pd/C) also
proceeds via the same intermediate, it is possible that
conversion of amines may be occurring on the catalyst’s
surface without the concomitant release of intermediates at
any stage,12 thus furnishing amine 2a without the cyclized
product 3a. To comprehend the mechanistic course for the
formation of dihydroindazoloisoquinoline 3a, the reduction
of the nitro group was abated at the hydroxylamine stage
with the view to examine its ability to undergo N-N
cyclization. Examples of cyclization involving the intramo-
lecular capture of an intermediate hydroxylamine are well-
documented in the literature.13 However, the role of hydroxy-
lamine as an intermediate in the formation of a N-N bond
has not yet been reported.
dual role by hindering the complete reduction of the nitro
group to amine 2a and by promoting the cyclization to
furnish product 3a, thus kinetically favoring formation of
3a over 2a. A closer look on the course of the cyclization
via hydroxylamine 11a to the cyclized product 3a revealed
that the cyclization may have occurred by the loss of a water
molecule, which in turn may have formed by the combination
of a proton obtained from the protonated base and the
hydroxyl anion departing from the hydroxylamine in 11a.
This gets support from the fact that the formation of cyclized
product 3a did occur, albeit in low yield (29%), because the
hydroxyl group is known to be a poor leaving group. Hence,
we decided to improve the leaving group tendency of the
hydroxyl group by derivatizing it with a tosyl group.15
Accordingly, we treated 11a with tosyl chloride, and after
15 min, we observed complete conversion to indazole 3a
(route II, Scheme 3). Next, the one-pot conversion of 1a to
3a was achieved by sequentially treating 1a with SnCl2‚
2H2O, PhSH, and Et3N for 15 min and then with tosyl
chloride for 15 min (route III, Scheme 3). The crude product
so obtained was purified by successively treating with water,
hexane, and methanol to yield the final product with >99%
purity.16 However, we further established the purity of 3a
by determining its melting point and HRMS, which matched
with the material obtained after passing through silica gel
column chromatography. A plausible mechanism for the
intramolecular cyclization of arylhydroxylamine 11a to
dihydroindazoloisoquinoline 3a could be via an electron-
deficient nitrene intermediate, as depicted in Scheme 4.
Recently, Lebel et al. demonstrated nitrene formation starting
from tosyloxy carbamates in the presence of potassium
carbonate as base.17 The scope and limitation of the strategy
was established by synthesizing three more congeners based
on 3 (route III, Scheme 3) by varying the phenethylamines
and o-nitro benzoic acids. The dihydroisoquinoline deriva-
tives 1a-d were synthesized by literature procedure using
For our studies, reduction of the nitro group to hydroxy-
lamine was achieved by treating substrate 1a with SnCl2‚
2H2O in the presence of PhSH and Et3N (route I, Scheme
3) as per literature procedure.14 The progress of the reduction
Scheme 3
(15) Stanetty, P.; Bahardoust, M. H.; Mihovilovic, M. D.; Mereiter, K.
Monatsh. Chem. 1999, 130, 1257-1268.
(16) General experimental procedure: Triethylamine (0.45 g, 3.20
mmol) was added dropwise to a stirred solution of SnCl2‚2H2O (0.22 g,
0.96 mmol) and PhSH (0.32 g, 2.90 mmol) in acetonitrile (5 mL) at room
temperature to generate yellow precipitate over a period of 5 min. Then,
1a (0.180 g, 0.64 mmol) was transferred to the suspension and stirred at
room temperature for 5 more min. After complete consumption of the nitro
substrate on TLC, TsCl (0.15 g, 0.83 mmol) was added to this solution and
monitoring of the reaction was done on TLC and HPLC. After completion
of the reaction, the solvent was evaporated and the resulting residue was
triturated with water (10 mL). The solid material was filtered and washed
extensively with water. It was dried in a desiccator in vacuo and then again
triturated with hexane. The solid was again filtered and washed extensively
with hexane and then dried in vacuo. The solid obtained was then dissolved
in methanol, and the insoluble material separated was removed by filtration.
The filtrate was finally evaporated, and the residue was crystallized from
EtOAc-hexane to give 2,3-dimethoxy-5,6-dihydroindazolo[3,2-a]isoquino-
line [3a]: white solid; yield ) 0.16 g (89%); mp 180-182 °C; tR ) 16.9
# Conversion was monitored by HPLC (%). $Isolated yield of 3
from 1.
was monitored by HPLC, and within 15 min, we were
pleased to observe the complete disappearance of 1a and
the formation of 3a (29%) along with the intermediate 11a
(71%). It appears that the basic condition probably played a
1
min; Rf ) 0.63 (2:3 EtOAc/hexane); H NMR (CDCl3, 300 MHz) δ 7.96
(d, 1H, J ) 8.4 Hz, ArH), 7.73 (d, 1H, J ) 8.7 Hz, ArH), 7.48 (s, 1H,
ArH), 7.32 (t, 1H, J ) 7.2 Hz, ArH), 7.16 (t, 1H, J ) 7.5 Hz, ArH), 6.85
(s, 1H, ArH), 4.62 (t, 2H, J ) 6.9 Hz, CH2), 4.02 (s, 3H, CH3), 3.94 (s,
(12) Enwistle, I. D.; Gilkerson, T. Tetrahedron 1978, 34, 213-215 and
references therein.
13
3H, CH3), 3.21 (t, 2H, J ) 6.9 Hz, CH2); C NMR (CDCl3, 50 MHz) δ
(13) (a) Yang, D.; Fokas, D.; Li, J.; Yu, L.; Baldino, C. M. Synthesis
2005, 47-56. (b) Bates, D. K.; Li, K. J. Org. Chem. 2002, 67, 8662-
8665. (c) Sykes, B. M.; Atwell, G. J.; Denny, W. A.; O’Connor, C. J. J.
Phys. Org. Chem. 1995, 8, 587-596.
(14) Bartra, M.; Romea, P.; Urpi, F.; Vilarrasa, J. Tetrahedron 1990,
46, 587-594.
149.1, 148.8, 131.1, 126.4, 125.3, 122.3, 121.1, 120.5, 118.2, 117.9, 111.9,
107.7, 56.6, 56.5, 48.3, 29.1; IR (KBr) νmax 1603, 1696, 1217; MS (FAB)
281 for [M + 1]+; MS (HR EI) m/z calcd for [M]+ 280.12118, found
280.12300.
(17) Lebel, H.; Huard, S.; Lectard, S. J. Am. Chem. Soc. 2005, 127,
14198-14199.
Org. Lett., Vol. 8, No. 8, 2006
1527