Synthesis of 1-aryl-1H-indazoles via palladium-catalyzed intramolecular amination of aryl halides

Article abstract of DOI:10.1021/jo048671t

(Chemical Equation Presented) Palladium-catalyzed cyclization of arylhydrazones of 2-bromoaldehydes and 2-bromoacetophenones to give 1-aryl-1H-indazoles has been studied in detail. The cyclization of arylhydrazone of 2-bromobenzaldehydes can be performed with good to high yields using Pd(dba)₂ and chelating phosphines, of which the most effective are rac-BINAP, DPEphos, and dppf, in the presence of Cs₂CO₃ or K₃PO₄ as a base. Electron-rich, bulky ligands commonly employed for intermolecular amination such as P^tBu₃ and o-PhC₆H₄P^tBu₂ were shown to be ineffective for cyclization and to lead instead to extensive oligomerization and tarring. The method developed is applicable for preparation of a wide scope of indazoles bearing electron-donating or electron-withdrawing substituents, among them, unprotected carboxyl, as well as various indazole heteroanalogues. The cyclization of arylhydrazones of less reactive halides such as 2-chlorobenzaldehyde, as well as 2-bromoacetophenone and bromotetralone, has been achieved. The purity of the starting hydrazone has been shown to be a critical parameter, as various impurities inhibit the cyclization.

Full text of DOI:10.1021/jo048671t

Synthesis of 1-Aryl-1H-indazoles via Palladium-Catalyzed
Intramolecular Amination of Aryl Halides
Artyom Y. Lebedev, Anton S. Khartulyari, and Alexander Z. Voskoboynikov*
Department of Chemistry, Moscow State University, V-234, Moscow, GSP, 119899, Russian Federation
voskoboy@org.chem.msu.ru
Received July 31, 2004
Palladium-catalyzed cyclization of arylhydrazones of 2-bromoaldehydes and 2-bromoacetophenones
to give 1-aryl-1H-indazoles has been studied in detail. The cyclization of arylhydrazone of
2-bromobenzaldehydes can be performed with good to high yields using Pd(dba)2 and chelating
phosphines, of which the most effective are rac-BINAP, DPEphos, and dppf, in the presence of
Cs2CO3 or K3PO4 as a base. Electron-rich, bulky ligands commonly employed for intermolecular
amination such as PtBu3 and o-PhC6H4PtBu2 were shown to be ineffective for cyclization and to
lead instead to extensive oligomerization and tarring. The method developed is applicable for
preparation of a wide scope of indazoles bearing electron-donating or electron-withdrawing
substituents, among them, unprotected carboxyl, as well as various indazole heteroanalogues. The
cyclization of arylhydrazones of less reactive halides such as 2-chlorobenzaldehyde, as well as
2-bromoacetophenone and bromotetralone, has been achieved. The purity of the starting hydrazone
has been shown to be a critical parameter, as various impurities inhibit the cyclization.
Introduction
unactivated arylhydrazones are not nucleophilic enough
to undergo noncatalytic cyclization. Meanwhile, the pal-
ladium-catalyzed intermolecular arylation of hydrazones
has recently been described.7 Thus, it seemed reasonable
to try the intramolecular catalyzed version.
It should be noted that examples of intramolecular Pd-
catalyzed C-N bond formation leading to heteroaromatic
fragments are still quite rare in the literature. For ex-
ample, the synthesis of carbazole from 2-iodo-2′-amino-
biphenyl has been described.8 Other examples are
Pd-catalyzed cyclization of N-aryl-N-(2-bromobenzyl)-
The 1-aryl-1H-indazole fragment can be recognized in
a number of biologically active molecules,1 as well as
known pharmaceuticals, among them antidepressants2
and contraceptives.3 Nevertheless, the published methods
of synthesis of this system are either limited in scope4 or
poorly elaborated;5 therefore, an effective approach to
obtaining compounds of this class is still desirable. One
of the most attractive ways to such products is the
catalytic cyclization of hydrazones of 2-haloaldehydes or
2-haloacetophenones. The cyclization of more nucleophilic
alkylhydrazones or activated arylhydrazones takes place
readily in the absence of a catalyst.6 However, common
(5) Frasca, A. R. Tetrahedron Lett. 1962, 24, 1115. Dennler, E. B.;
Frasca, A. R. Tetrahedron, 1966, 22, 3131. Gladstone, W. A. F.;
Norman, R. O. C. J. Chem. Soc. C 1966, 1527. Fedorov, A. Y.; Finet,
J.-P. Tetrahedron Lett. 1999, 40, 2747. Lam, P. Y. S.; Clark, C. G.;
Saubern, S.; Adams, J.; Averill, K. A.; Chan, D. M. T.; Combs, A. Synlett
2000, 674. Collot, V.; Bovy, P. R.; Rault, S. Tetrahedron Lett. 2000,
41, 9053. Krapars, A.; Antilla, J. C.; Huang, X.; Buchwald, S. L. J.
Am. Chem. Soc. 2001, 123, 7727. Chan, D. M. T.; Monaco, K. L.; Li,
R.; Bonne, D.; Crark, C. G.; Lam, P. Y. S. Tetrahedron Lett. 2003, 44,
3863. Antilla, J. C.; Baskin, J. M.; Barder, T. E.; Buchwald, S. L. J.
Org. Chem. 2004, 69, 5578-5587.
(6) Borsche, W.; Diacont, K. Ann. Chem. 1934, 510, 287. Borsche,
W.; Buetschli, L. Ann. Chem. 1936, 522, 285. Borsche, W.; Scriba, W.
Ann. Chem. 1939, 541, 283. Showalter, H. D.; Hollis, J. L.; Hoftiezer,
J. M. J. Heterocycl. Chem 1986, 23, 1491. Capps, D. B.; Dunbar, J.;
Kesten, S. R.; Shillis, J.; Werbel, L. M. J. Med. Chem. 1992, 35, 4770.
(7) Wagaw, S.; Yang, B. H.; Buchwald, S. L. J. Am. Chem. Soc. 1998,
120 (26), 6621. Jiang, L.; Job, G. E.; Klapars, A.; Buchwald S. L. Org.
Lett. 2003, 5 (20), 3667.
(1) Bistochi, G. A.; De Meo, G.; Pedini, M.; Ricci, A.; Brouilhet, H.;
Bucherie, S.; Rabaud, M.; Jacquignon P. Farmaco Ed. Sci. 1981, 36,
315.
(2) Gschwend, H. W. (Ciba-Geigy). U.S. Patent 3,681,382, 1972.
(3) Corsi, G.; Palazzo, G.; Germani, C.; Barcellona, P. S.; Silvestrini
B. J. Med. Chem. 1976, 19, 778.
(4) Ruggli, P.; Straub, O. Helv. Chim. Acta 1938, 16, 1084. Theilack-
er, W.; Raabe, C. Tetrahedron Lett. 1966, 1, 91. Kuehne, M.; Weaver,
S. J.; Franz, P. J. Org. Chem. 1963, 29, 1582. Chinn, L. J. J. Org.
Chem. 1971, 36, 12. Tsuge, O.; Samura, H. J. Het. Chem. 1971, 8, 707.
Croce, P. D.; La Rosa, C. Synthesis 1984, 982. Mercedes, M.-S.; Jose
Alvarez de, L. M.; Sequlveda-Arques, J. J. Chem. Soc., Perkin Trans.
1 1990, 10, 2749. Lyubchanskaya, V. M.; Alekseeva, L. M.; Granik, V.
G. Tetrahedron 1997, 53, 15005. Olivera, R.; SanMartin, R.; Domingues,
E. J. Org. Chem. 2000, 65, 7010.
10.1021/jo048671t CCC: $30.25 © 2005 American Chemical Society
Published on Web 12/22/2004
596
J. Org. Chem. 2005, 70, 596-602

Synthesis of 1-Aryl-1H-indazoles
hydrazines9a and N-aryl-N′-(2-bromobenzyl)hydrazines9b
followed by spontaneous aromatization of aryl-substitut-
ed 2,3-dihydro-1H-indazoles formed. Wider application
of this promising synthetic strategy is handicapped by
the poor availability of cyclizable substrates and side
reactions that complicate the intramolecular cross-
coupling. The interaction of cyclizable substrates with
catalysts often leads to the formation of stable metalla-
cycles, thus leading to catalyst inactivation. Therefore,
those palladium10 and copper11 catalysts that have proven
to be highly useful in intermolecular C-N cross-coupling
reactions may fail in cyclizations. Thus, the complexes
of Pd with highly basic bulky phosphines, which are
among the best catalysts for Pd-catalyzed intermolecular
amination, behave poorly in cyclization reactions due to
a high preference for intermolecular coupling, leading to
oligomerization and tarring. The same is true for effective
copper catalysts of intermolecular amination. On the
other hand, palladium complexes with chelating diphos-
phines, which are known to be less effective catalysts for
intermolecular amination,10a are preferable for cyclization
reactions, possibly because the chelating ligand does not
allow inactive palladacycle to be formed.12 As soon as
catalyst deactivation is suppressed, the use of a less
active catalyst is justified; otherwise, the intramolecular
process is more favorable than the intermolecular one.
This work is devoted to the development of effective
intramolecular amination in palladium-catalyzed cycliza-
tion of arylhydrazones of 2-bromoaldehydes or 2-bro-
moacetophenones (eq 1).
the key to better yields, milder reaction conditions, and
extension of scope to heterocyclic substrates and 2-chlo-
robenzaldehyde. Moreover, we succeeded in achieving the
cyclization of arylhydrazones of 2-bromoacetophenone
and 8-bromo-R-tetralone. Though the yields of the respec-
tive indazoles in the latter case are only moderate, the
synthesis of these molecules under the conditions re-
ported by Cho et al.13 is impossible, as shown below.
Results and Discussion
In initial experiments, we have shown that the reaction
of 2-bromobenzaldehyde with phenylhydrazine can be
performed in toluene in the presence of palladium
catalyst and anhydrous K3PO4 and yields 65% 1-phenyl-
1H-indazole (1b) after 12 h at 110 °C (eq 2). We believe
that the reaction involves intermolecular catalytic ary-
lation of NH fragment of arylhydrazine.
To test this hypothesis, we studied the Pd-catalyzed
cyclization of 2-bromobenzaldehyde phenylhydrazone
(1a), which was prepared in an analytically pure form
by condensation of 2-bromobenzaldehyde with phenyl-
hydrazine. Catalytic cyclization of this compound turned
out to give 83% yield of indazole 1b after 12 h at 110 °C
with quantitative conversion of starting materials (eq 3),
which is a much better result than that achieved in the
published method using in situ-generated hydrazone.13
Recently, Cho et al.13 have shown that various 1-aryl-
1H-indazoles can be obtained in modest to good yields
through the Pd-catalyzed reaction of 2-bromobenzalde-
hydes and arylhydrazines. However, the reported method
suffers from essential limitations of scope. Simulta-
neously with these researchers, we accomplished a
similar study on the Pd-catalyzed synthesis of arylinda-
zoles; we found that the use of preformed hydrazones is
In the absence of Pd catalyst, no cyclization occurred.
The significant improvement of yield in the reaction with
pure hydrazone as compared to the in situ reaction is
likely to be associated with the inhibition of catalyst by
condensation step byproducts or possibly by starting
materials themselves. The negative influence of water
liberated during hydrazone formation should also be
considered. To estimate the influence of water, we
performed a separate test. After a toluene solution of
2-bromobenzaldehyde and phenylhydrazine was stirred
for 30 min, water was removed by 4 Å molecular sieves.
The solution was charged by a Pd catalyst, and after 12
h at 110 °C the yield was only 65%. Thus, we can
conclude that the influence of water is negligible com-
pared to inhibition by byproducts or starting materials.
To check the latter hypothesis, we studied the cyclization
of 1a in the presence of a 10% additive of either
phenylhydrazine or 2-bromobenzaldehyde. The reaction
with phenylhydrazine gave essentially the same results
as a stoichiometric reaction. On the contrary, the addition
of 2-bromobenzaldehyde led to dramatic inhibition of
cyclization. The yield of indazole 1b decreased to 40%
versus 83% for the reaction with no additive under
otherwise identical conditions (12 h at 110 °C), though
(8) Driver, M. S.; Hartwig, J. F. J. Am. Chem. Soc. 1995, 117, 4708.
Wood, J.-L.; Stoltz, B. M.; Dietrich, H.-J.; Pflum, D. A.; Petsch, D. T.
J. Am. Chem. Soc. 1997, 119, 9641.
(9) (a) Song, J. J.; Yee, N. K. Org. Lett. 2000, 2, 519. (b) Song, J. J.;
Yee, N. K. Tetrahedron Lett. 2001, 42, 2937.
(10) (a) Wagaw, S.; Marcoux, J.-F.; Buchwald, S. L. Acc. Chem. Res.
1998, 31, 805. Hartwig, J. F. Angew. Chem., Int. Ed. 1998, 37, 2046.
Wolfe, J. P.; Yang, B. H.; Buchwald, S. L. J. Organomet. Chem. 1999,
576, 125. (b) Alcazar-Roman, L. M.; Hartwig, J. F.; Rheingold, A. L.;
Liable-Sandc, L. M.; Guzei, I. A. J. Am. Chem. Soc. 2000, 122, 4618.
Alcazar-Roman, L. M.; Hartwig, J. F. J. Am. Chem. Soc. 2001, 123,
1295. Viciu, M. S.; Germaneau, R. F.; Nolan, S. P. Org. Lett. 2002, 4,
4053. Strieter, E. R.; Blackmond, D. G.; Buchwald, S. L. J. Am. Chem.
Soc. 2003, 125, 13978. Zim, D.; Buchwald, S. L. Org. Lett. 2003, 5,
2413.
(11) Antilla, J. C.; Klapars, A.; Buchwald, S. L. J. Am. Chem. Soc.
2002, 124, 11684. Kwong, F. Y.; Buchwald, S. L. Org. Lett. 2003, 5,
793. Klapars, A.; Parris, S.; Anderson, K. W.; Buchwald, S. L. J. Am.
Chem. Soc. 2004, 126, 3529.
(12) Paul, F.; Patt, J.; Hartwig, J. F. Organometallics 1995, 14, 3030.
Wagaw, S.; Buchwald, S. L. J. Org. Chem. 1996, 61, 7240.
(13) Cho, C. S.; Lim, D. K.; Heo, N. H.; Kim, T.-J.; Shim, S. C. Chem.
Commun. 2004, 104.
J. Org. Chem, Vol. 70, No. 2, 2005 597

Lebedev et al.
TABLE 1. Cyclization of 1a to 1b Catalyzed by
Pd(dba)2/Phosphinea
TABLE 2. Cyclization of 1a to 1b Catalyzed by
Pd(dba)2/DPEphos in the Presence of Various Basesa
yield (%)
conversion (%)
entry
base
yield (%) of 1b
conversion (%) of 1a
entry
phosphineb
time, h
of 1b
of 1a
1
2
3
4
5
6
7
Cs2CO3
K3PO4
Na3PO4
K2CO3
tBuOLi
tBuONa
tBuOK
68
58
4
78
60
16
37
39
24
28
1
2
DPEphos
DPEphos
DPEphos
DPEphos
rac-BINAP
dppf
Xantphos
Xantphos
dppp
0.5
1
2
12
1
1
1
12
1
15
12
12
12
43
58
44
60
97
>99
85
65
30
3
94
28
31
9
4
83
5
85
6
57
0
7
12
a Conditions: 1a (1.0 mmol), Pd(dba)2 (0.02 mmol), DPEphos
(0.02 mmol), base (2.5 mmol), toluene (6 mL), 1 h, 110 °C.
8
69
>99
9
9
1
10
11
12
13
dppp
10
35
c
PtBu3
<0.1
<0.1
11
40
that of the BINAP complex. As DPEphos is much cheaper
and more readily available than BINAP, it should be
preferred for practical reasons as the catalyst for the
preparative synthesis of 1-aryl-1H-indazoles.
2
3
40
12
a Conditions: 1a (1.0 mmol), Pd(dba)2 (0.02 mmol), phosphine
(0.02 mmol), K3PO4 (2.5 mmol), toluene (6 mL), 110 °C. b Phos-
phines used were bis[2-(diphenylphosphino)phenyl] ether
(DPEphos), 4,5-bis(diphenylphosphino)-9,9-dimethyl-9H-xanthene
(Xantphos), 1,3-bis(diphenylphosphino)propane (dppp), rac-2,2′-
bis(diphenyl-phos-phino)-1,1′-binaphthyl (rac-BINAP), 1,1′-bis-
(diphenylphosphino)ferrocene (dppf), 2-N,N-dimethyl-amino-2′-(di-
tert-butylphosphino)-1,1′-biphenyl (2), and dicyclohexyl[2-(9-
phenanthryl)phenyl]phosphine (3). c Pd/P ) 1/1 was used; Pd(PtBu3)2
was found to have the same activity.
The xantphos ligand gave a catalyst with a much lower
activity (12% yield/30% conversion after 1 h). After
prolonged heating for 12 h, the yield increases to 69%
with quantitative conversion of hydrazone. Such ligands
as PtBu3 or 3 altogether failed to give the desired product,
though the conversion was high. Thus, the complexes of
palladium with highly basic bulky phosphine ligands,
which are among the best catalysts for intermolecular
amination,15 behave poorly in the cyclization reaction due
to a preference for intermolecular coupling, leading to
oligomerization and tarring.
Further studies on the effect of the base, solvent, and
substituents in the substrate were carried out with the
DPEphos system, having an optimal combination of
activity, selectivity, price, and availability. First, fast
cyclization of 1a was shown to occur in dioxane to give
1b in 67% yield over 2 h at 90 °C. Under similar
conditions in DMF or toluene solution, the desired
product was formed in 7 or 23% yield, respectively.
However, in further studies, we used toluene as a solvent,
which allowed the reaction to be carried out faster at
higher temperatures. For instance, indazole 1b is formed
in as high as 94% yield in toluene over 2 h at reflux. The
variation of the base gave the results shown in Table 2.
Cs2CO3 is the best base under the conditions studied.
However, K3PO4 was only marginally inferior, and, given
the immensely lower price of the latter base, the choice
is obvious. Potassium carbonate and lithium tert-butoxide
both gave much poorer yields furnishing only a modest
yield after 1 h. Sodium phosphate and sodium and
potassium tert-butoxides are even less useful, causing the
decomposition of hydrazone. Thus, in further studies, we
used potassium phosphate as an optimal base.
the conversion in both cases was nearly quantitative.
This effect can be explained by the initiation of a side
reaction, intermolecular C-N coupling, in the presence
of 2-bromobenzaldehyde (eq 4). Therefore, all further
experiments on the catalytic formation of indazoles were
carried out with preformed and carefully purified hydra-
zones.
Next, we investigated the influence of phosphine ligand
and base on the cyclization of 1a. These data are shown
in Table 1. It should be noted that indazole 1b undergoes
a slow spontaneous decomposition when heated in tolu-
ene solution for a long time. Indeed, while the yields of
1b were nearly quantitative after 2 h (Table 1, entries
1-3), further refluxing resulted in a decrease of the yield
by ca. 12% (entry 4) due to formation of unidentified
products. Other indazoles show similar behavior, which
should be taken into consideration to avoid losses of
product.
A comparison of ligands can best be drawn from the
results of incomplete conversion reactions quenched after
1 h (entries 2, 5-7, 9, 13). We tested a wide selection of
ligands typically employed in Pd-catalyzed cross-coupling
reactions,14 particularly, the arylation of amines. The best
results were obtained with rac-BINAP, which afforded
an 85% yield after 1 h at reflux. Moreover, in this case,
all reacted hydrazone transformed into the product
without losses. All other ligands invariably gave yields
lower than conversion. The catalysts containing DPEphos
and dppf showed similar activities marginally inferior to
The study of substituent effects on Pd-catalyzed cy-
clization was performed with arylhydrazones prepared
from the respective 2-bromobenzaldehydes and arylhy-
drazines in an analytically pure state. The results
obtained for the cyclization of these hydrazones in the
presence of Pd(dba)2/DPEphos/K3PO4 are given in Table
3. With all hydrazones, the cyclization occurred only at
elevated temperatures. Substituents in arylhydrazine
had no significant influence on yields. Similar results
were obtained for hydrazones bearing electron-withdraw-
ing (indazoles 5b, 8b, 11b) and electron-donating (inda-
(14) (a) Tsuji, J. Palladium Reagents and Catalysts, Innovations in
Organic Synthesis; Wiley-VCH: New York, 1995. (b) Metal-Catalyzed
Cross-Coupling Reactions; Diederich, F., Stang, P. J., Eds.; Wiley-
VCH: New York, 1997; p 2.
(15) Old, D. V.; Wolfe, J. P.; Buchwald S. L. J. Am. Chem. Soc. 1998,
120, 9722. Hartwig, J. F.; Kawatsura, M.; Hauck, S. I.; Shaudhnessy,
K. H.; Alcazar-Roman, L. M. J. Org. Chem. 1999, 64, 5575.
598 J. Org. Chem., Vol. 70, No. 2, 2005

Synthesis of 1-Aryl-1H-indazoles
TABLE 3. Cyclization of Hydrazones of 2-Bromobenzaldehydes Catalyzed by Pd(dba)2/DPEphosa
a Conditions: aldehyde hydrazone (1.0 mmol), Pd(dba)2 (0.02 mmol), DPEphos (0.02 mmol), base (2.5 mmol), toluene (6 mL), 12 h,110
°C. b At 43% conversion of the starting arylhydrazone. c Performed with 3.0 mmol of tBuOLi as a base. d Byproduct observed is indazole
16b (12%) e Time ) 120 h.
J. Org. Chem, Vol. 70, No. 2, 2005 599

Lebedev et al.
zole 7b) substituents in the benzene ring of arylhydra-
zine. Ortho substituents in arylhydrazine were also well
tolerated (indazoles 4b, 6b). In all cases, the respective
indazoles were formed in high yield with quantitative
conversion of the starting hydrazones.
Lithium tert-butoxide was used as the base for the
synthesis of 4-(1H-indazol-1-yl)benzoic acid (11b). In this
case, K3PO4 could not be used, as potassium salt of the
respective hydrazone is insoluble in toluene. With lithium
tert-butoxide, indazole is formed in 56% yield and can
be readily isolated upon acidification of the reaction
mixture.
after the oxidative addition stage may explain the
observed low reactivity of R-bromopyridine hydrazone
28a. Interestingly, the closest relative of this compound,
phenylhydrazone of 3-bromoisonicotinic aldehyde (26a),
is smoothly transformed into the corresponding 5-azain-
dazole 26b in a near-quantitative yield after 12 h at
110 °C.
The cyclization of 4-bromophenylhydrazone of 2-bro-
mobenzaldehyde was much slower than the reactions of
all other hydrazones. Indazole 9b is formed in only 40%
yield (with 43% conversion) after 12 h reflux. Longer
heating (130 h) gives 26% yield and quantitative conver-
sion. Apparently, this indazole is unstable toward heating
in the presence of Pd catalyst, as the extra Br atom can
be involved in followup reactions leading to formation of
unidentified oligomeric products. Similarly, the reaction
of phenylhydrazone of 2,5-dibromo-4-methylbenzalde-
hyde gives only 32% yield of target indazole 19b, though
in this case we could isolate 12% yield of byproduct
indazole 16b (eq 5). Debromination is a well-known side
reaction in the Pd-catalyzed amination of aryl bromides.16
Here again the conversion was quantitative, revealing
that the substrate is consumed mostly due to intermo-
lecular reaction.
Next, we attempted to extend the method to chloro
derivatives. Phenylhydrazone of 2-chlorobenzaldehyde
gave no cyclization product in the system Pd(dba)2/
DPEphos (2 mol %), K3PO4 after 120 h at 110 °C. With
PtBu3, DPEphos, or P-N ligand, the conversion did not
exceed 1%. However, activated substrates such as aryl-
hydrazones of 5-nitro-2-chlorobenzaldehyde 29a-31a
were found to give readily the respective indazoles
29b-31b (eq 6). It should be noted that, while 29a (R )
H), involving unsubstituted phenyl, gave 29b in as high
as 70% yield, the analogous substituted substrates 30a
(R ) OMe) and 31a (R ) CF3) gave the respective
cyclization products in lower yields, i.e., 25 and 43%,
respectively. These reactions take place only in the
presence of a Pd catalyst. We ascertained in a blank run
without Pd catalyst that uncatalyzed nucleophilic sub-
stitution18 does not take place.
Further, we investigated the cyclization of arylhydra-
zones of heterocyclic analogues of 2-bromobenzaldehyde.
Pd-catalyzed cyclization of arylhydrazones of 3-bromo-
2-thiophencarbaldehyde 21a-23a afforded new 1-aryl-
thieno[3,2]pyrazoles 21b-23b. These products were,
however, obtained in lower yields than in the case of the
analogous indazoles, though the conversion was always
quantitative. An attempt to carry out the reaction at
lower temperatures (80 °C) was unsuccessful, as no
product was observed in the reaction mixture. A shorter
reaction time (5 h at 110 °C) could not solve the prob-
lem: 43% yield of 19b with 65% conversion was formed.
Therefore, as prolonged heating at 110 °C is likely to
result in considerable destruction of the product, care-
ful optimization is required if higher yields are to be
obtained.
The behavior of hydrazones containing a pyridine ring
depends strongly on their structure. Hydrazones 27a and
28a showed no traces of cyclization in the presence of
Pd(dba)2/DPEphos/K3PO4 after 120 h at 110 °C (starting
materials were not consumed). For hydrazine 27a, this
can be easily explained by inhibition due to chelating of
Pd by hydrazone. For hydrazone 28a, the cause of the
inhibition is not obvious, though possible formation of a
stable dimer A involving a six-membered palladacycle17
Next, we turned our attention to the synthesis of
2-substituted 1-aryl-1H-indazoles from the respective
2-bromoacetophenones. The main difficulty here is to
employ a mild and effective method for the preparation
of hydrazones. The formation of hydrazones from alkyl-
aromatic ketones is much slower than from benzalde-
hydes.19 Moreover, there is a risk of initiating the Fischer
indole synthesis20 if the reaction conditions are not mild
enough. After optimization studies, we found that the
best and safest method is to perform hydrazone formation
by prolonged (50-170 h) reflux of an equimolar mixture
of a ketone (2-bromoacetophenone or 8-bromo-R-tetra-
lone) and arylhydrazine (eqs 7, 8) in low-boiling diethyl
ether. It should be noted that attempts to increase the
reaction rate by increasing the temperature led to the
formation of a complex mixture of products. At this point,
(18) Beller, M.; Briendl, C.; Riermeier, T. H.; Tillack, A. J. Org.
Chem. 2001, 66, 1403.
(16) Hartwig, J. F.; Richards, S.; Baran, D.; Paul, F. J. Am. Chem.
Soc. 1996, 118, 3626.
(17) Benetollo, F.; Bombieri, G.; DiBianca, F.; Crociani, B. Inorg.
Chim. Acta 1991, 188, 51.
(19) Smith, M. B.; March, J. Advanced Organic Chemistry, 5th ed.;
Wiley-VCH: New York, 2001; p 1186.
(20) Sundberg, R. J. Indoles; Academic Press: New York, 1996.
600 J. Org. Chem., Vol. 70, No. 2, 2005

Synthesis of 1-Aryl-1H-indazoles
SCHEME 1. Catalytic Cycle for Cyclization of Arylhydrazones of 2-Bromobenzaldehydes
the inapplicability of the previously reported technique13
became obvious, as it requires prolonged heating.
are formed with Buchwald’s phenanthrene-based phos-
phine 3, which is known to form a chelate complex with
palladium involving Pd-P and Pd-arene interactions.21
Thus, indazole 35b was formed in 45% yield after 24 h
at 120 °C with 20 mol % Pd catalyst, and indazole 36b
was formed in 40% yield in the presence of 10 mol %
catalyst (eq 11). Lower loadings of catalyst gave unsat-
isfactory results due to unfavorable competition of slower
cyclization and decomposition of indazole product.
8-Bromo-R-tetralone for this synthesis was obtained by
oxidation of 8-bromo-R-tetralole, which itself was syn-
thesized via ortho lithiation of R-tetralole followed by
treatment with 1,2-dibromoethane (eq 9).
The tentative mechanism of the described cyclization
reaction should take into consideration the published
data on palladacycles formed from phenylhydrazones of
aromatic aldehydes and ketones.22,23
The oxidative addition of Pd(0) to the carbon-halogen
bond of hydrazone I should give the five-membered
palladacycle III, possibly in equilibrium with open struc-
ture II (Scheme 1). The cyclization should involve a
different six-membered palladacycle, which can be formed
from II by deprotonation of NH by base and the subse-
quent metallotropic shift of Pd to form palladacycle IV.
The shift should first involve the dissociation of the Pd-N
bond; otherwise, cis/trans isomerization of the hydrazone
fragment cannot take place. The actual order of these
Hydrazones 32a and 33a were found to undergo
cyclization in the presence of 7 mol % Pd(dba)2/DPEphos
and K3PO4 to form indazoles 32b and 33b in 47% and
48% yields, respectively, after 160 h at 120° (eq 10).
(21) Yin, J.; Rainka, M. P.; Zhang, X.-X.; Buchwald, S. L. J. Am.
Chem. Soc. 2002, 124, 1162.
(22) Espinet, P.; Garcia-Herbosa, G.; Herrero, F. J.; Jeannin, Y.;
Philoche-Levisalles, M. Inorg. Chem. 1989, 28, 4207.
(23) Espinet, P.; Alonso, M. Y.; Garcia-Herbosa, G.; Ramos, J. M.;
Jeannin, Y.; Philoche-Levisalles, M. Inorg. Chem. 1992, 31, 2501.
R-Tetralone hydrazones 35a and 36a in this system
were reluctant to react. The screening of various ligands
has shown that the best, though still rather low, yields
J. Org. Chem, Vol. 70, No. 2, 2005 601

Lebedev et al.
events (Pd-N dissociation, deprotonation of NH, isomer-
ization, and reclosure of the palladacycle) cannot be
unambiguously established. The transformation of the
five-membered palladacycle into indazole by a strong
base was described earlier in a study devoted to the
chemistry of the palladacyclic carbonyl complex (eq 12),
which is transformed into 1-phenyl-3-methylindazole.24
suppress the formation of stable inactive palladacycles
such as III or similar dimers as shown in eq 13. The
method developed is applicable for preparation of a wide
variety of indazoles bearing electron-donating or electron-
withdrawing substituents, among them, unprotected
carboxyl, as well as various heteroanalogues of inda-
zoles. The cyclization of arylhydrazones of less reactive
halides such as 2-chlorobenzaldehyde, as well as 2-bromo-
acetophenone and bromotetralone, has been achieved.
The purity of the starting hydrazone has been shown to
be a critical parameter, as various impurities inhibit the
cyclization.
Experimental Section
One of the key roles of the chelating phosphine ligand
in the proposed mechanism should be to promote rupture
of the Pd-N bond in palladacycle III, which otherwise
would be stable enough to terminate the catalytic cycle.
In the absence of a good chelating phosphine ligand, these
five-membered palladacycles can even be further trans-
formed by deprotonation into very stable dimeric com-
plexes, as has been shown by Espinet et al.,22 thus
resulting in the further deactivation of catalyst (eq 13).
Reaction of arylhydrazines with aldehydes (general proce-
dure A). Hot solutions of 3 mmol of arylhydrazine in a minimal
volume of methanol and of 3 mmol of arylaldehyde in 5 mL of
methanol were mixed and stirred with refluxing for 30 min.
After that, the reaction mixture was cooled to -30 °C and kept
at this temperature for 12 h. The precipitates were collected
by filtration, washed twice with a small portion of cold
methanol, and dried in a vacuum.
Reaction of arylhydrazines with acetophenones (general
procedure B). A 24 mL vial was charged with 5 mmol of the
corresponding phenylhydrazine, 5 mmol of 2-bromoacetophe-
none, and 20 mL of dry ether. The vial was flushed with argon,
sealed, and thermostated for 140 h at 40 °C. The product was
isolated by evaporation of ether, washing of the residue with
two 10 mL portions of hot hexanes, and drying in a vacuum.
Synthesis of 1-aryl-1H-indazoles (general procedure C). An
8 mL vial was charged with 1 mmol of arylhydrazone, 530 mg
(2.5 mmol) of K3PO4, 10.8 mg (0.02 mmol) of phosphine ligand,
11.5 mg (0.02 mmol) of Pd(dba)2, and 6 mL of toluene. This
mixture was stirred for 8 h (if not otherwise specified; see Table
3) at 110 °C. Next, the resulting mixture was cooled to room
temperature and passed through a short column with Silica
Gel 60 (40-63 µm, d ) 50 mm, l ) 30 mm). The column was
additionally washed with 50 mL of THF. The eluate was
evaporated to dryness, and the product was purified using
semipreparative HPLC.
Conclusion
In conclusion, we have shown that the cyclization of
arylhydrazone of 2-bromobenzaldehydes can be per-
formed in good to high yields using Pd(dba)2 and chelat-
ing phosphines, of which the most effective are rac-
BINAP, DPEphos, and dppf, in the presence of Cs2CO3
or K3PO4 as a base. Electron-rich, bulky ligands com-
monly employed for intermolecular amination such as
PtBu3 and 2 were shown to be ineffective for cyclization
and to lead instead to extensive oligomerization and
tarring. We believe that the most important role of
chelating ligands in the intramolecular reaction is to
Acknowledgment. Financial support from the Rus-
sian Government is gratefully acknowledged. We also
thank Prof. Irina P. Beletskaya.
Supporting Information Available: Experimental pro-
cedures, characterization of new compounds, and references
to known compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
(24) Carbayo, A.; Cuevas, J. V.; Garcia-Herbosa, G. J. Organomet.
Chem. 2002, 658, 15.
JO048671T
602 J. Org. Chem., Vol. 70, No. 2, 2005