Experimental and theoretical investigation of the coarctate cyclization of (2-ethynylphenyl)phenyldiazenes

Article abstract of DOI:10.1021/jo049011r

A new route to substituted 2-phenyl-2H-indazoles through the cyclization of (2-ethynylphenyl)-phenyldiazenes is presented. A coarctate reaction pathway forms the isoindazole carbene under neutral conditions, at moderate temperatures, and without the requirement of a carbene stabilizer. A wide variety of previously unknown diazene precursors was synthesized and cyclized. Trapping of the carbene with a silyl alcohol followed by deprotection affords the 3-hydroxymethyl-2-phenyl-2H-indazoles in good overall yield. The free carbene could also be trapped as a [2 + 1] cycloadduct with 2,3-dimethyl-2- butene.

Full text of DOI:10.1021/jo049011r

Exp er im en ta l a n d Th eor etica l In vestiga tion of th e Coa r cta te
Cycliza tion of (2-Eth yn ylp h en yl)p h en yld ia zen es‡
Laura D. Shirtcliff, Timothy J . R. Weakley,† and Michael M. Haley*
Department of Chemistry, University of Oregon, Eugene, Oregon 97403-1253
Felix Ko¨hler and Rainer Herges*
Institut fu¨r Organische Chemie, Universita¨t Kiel, 24098 Kiel, Germany
haley@oregon.uoregon.edu; rherges@oc.uni-kiel.de
Received J une 11, 2004
A new route to substituted 2-phenyl-2H-indazoles through the cyclization of (2-ethynylphenyl)-
phenyldiazenes is presented. A coarctate reaction pathway forms the isoindazole carbene under
neutral conditions, at moderate temperatures, and without the requirement of a carbene stabilizer.
A wide variety of previously unknown diazene precursors was synthesized and cyclized. Trapping
of the carbene with a silyl alcohol followed by deprotection affords the 3-hydroxymethyl-2-phenyl-
2H-indazoles in good overall yield. The free carbene could also be trapped as a [2 + 1] cycloadduct
with 2,3-dimethyl-2-butene.
In tr od u ction
Synthesis of the nitrogen-containing heterocycles men-
tioned above often involves formation of diazonium salts.
These must be generated by strong acids in polar solvents
to form a diazonium species susceptible to nucleophilic
attack by an activated o-carbon nucleophile.7 This in turn
limits the functional group tolerance of these types of
reactions. Other recent syntheses of nitrogen-containing
heterocycles often utilize a transition metal-mediated
route involving an external 1,2-disubstituted alkyne and
a substituted iodoarene or an alkynylarene and an alkyl-
or arylhalide.8 A recently reported synthesis of 2-aryl-
2H-indazoles utilizing a Pd catalyst suffers from moder-
ate cyclization yields and has limited functional group
tolerance due to the requirement of excess base for the
reaction to proceed.9 Pd-catalyzed intramolecular ami-
nations have also been utilized in heterocyclic synthesis.10
Recently, our group reported a new reaction pathway
that allows for the formation of both isoindazoles and
cinnolines from a single synthetic precursor, (2-ethy-
nylphenyl)triazenes (1), under neutral conditions (Scheme
1).11 The yields of the two different cyclization products
were very good to excellent for all functional groups
attempted. Through a combination of computational and
experimental data we were able to determine the sepa-
rate mechanisms of the two cyclizations. Cu-mediated
Heterocyclic compounds are important in organic
synthesis due to their prevalence in biologically active
compounds, such as inhibitors and antitumor agents.1
The indazole moiety constitutes the key subunit in many
drug substances with a broad range of pharmacological
activities including antitumor,2 anti-HIV,3 and antiin-
flammatory properties.4 Functionalized cinnolines also
have a broad range of pharmacological activities,5 and
2-alkylisoindazoles are currently being investigated in
part of a drug discovery program.6 Because of this, it is
desirable to synthesize these types of compounds in an
efficient and high-yielding manner, while tolerating a
wide variety of functional groups.
‡ Dedicated to Professor W. E. Billups on occasion of his 65th
birthday.
† To whom inquiries regarding the X-ray data should be directed.
(1) (a) Dell’Erba, C.; Novi, M.; Petrillo, G.; Tavani, C. Tetrahedron
1992, 48, 325-334. (b) Rodgers, J . D.; J ohnson, B. L.; Wang, H.;
Greenberg, R. A.; Erickson-Viitanen, S.; Klabe, R. M.; Cordova, B. C.;
Rayner, M. M.; Lam, G. N.; Chang, C. H. Bioorg. Med. Chem. Lett.
1996, 6, 2919-2924.
(2) (a) Baraldi, P. G.; Balboni, G.; Pavani, M. G.; Spalluto, G.;
Tabrizi, M. A.; De Clercq, E.; Balzarini, J .; Bando, T.; Sugiyama, H.;
Romagnoli, R. J . Med. Chem. 2001, 44, 2535-2742. (b) Wrobleski, S.
T.; Chen, P.; Hynes. J .; Shugun, L.; Norris, D. J .; Pandit, C. R.; Spergel,
S.; Wu, H.; Tokarski, J . S.; Chen, X.; Gillooly, K. M.; Kiener, P. A.;
McIntyre, K. W.; Patil-koota, V.; Shuster, D. J .; Turk, L. A.; Yang, G.;
Leftheris, P. A. J . Med. Chem. 2003, 46, 2110-2116.
(3) Sun, J . H.; Teleha, C. A.; Yan, J . S.; Rodgers, J . D.; Nugiel, D.
A. J . Org. Chem. 1997, 62, 5627-5629.
(4) Arciola, C. R.; Montanaro, L.; Caramazza, R.; Sassoli, V.;
Cavedagna, D. J . Bio. Mater. Res. 1998, 42, 1-5.
(7) Katritzky, A. R.; Pozharskii, A. F. Handbook of Heterocyclic
Chemistry: Elsevier Science Ltd: Oxford, 2000.
(8) (a) Vasilevsky, S. F.; Prikhodko, T. A. Russ. Chem. Bull. Int. Ed.
2001, 50, 1268-1273. (b) Larock, R. C.; Dai, G. J . Org. Chem. 2003,
68, 920-928. (c) Wu, M.; Lin, C.; Wei, L. Tetrahedron Lett. 2000, 41,
1215-1218.
(5) (a) Yu, Y.; Singh, S. K.; Liu, A.; Li, T.; Liu, L.; LaVoie, E. J .
Bioorg. Med. Chem. 2003, 11, 1475-1491. (b) Borowski, E.; Stefanska,
B.; Arciemiuk, M.; Bontemps-Gracz, M. M.; Dzieduszycka, M.; Kupiec,
A.; Martelli, S. Bioorg. Med Chem. 2003, 11, 561-572. (c) Pouyssegur,
J .; Gasmi, A.; Courcier. C.; Aloui, Z.; Sraira, N.; Marchetti, S.; Clotilde,
G.; Wedge, S. R.; Hennequin, L. J . Bio. Chem. 2002, 277, 29992-29998.
(6) Cheung, M.; Stafford, J . A.; Boloor, A. J . Org. Chem. 2003, 68,
4093-4095.
(9) Song, J . J .; Yee, N. K. Org. Lett. 2000, 2, 519-521.
(10) (a) Ibuka, T.; Takemoto, Y.; Tanaka, T.; Fuji, N.; Ohishi, S.;
Toda, A.; Anzai, M.; Ohno, H. J . Org. Chem. 2001, 66, 4904-4914. (b)
J ackson, R. F. W.; Hunter, C.; Deboves, H. J . C. J . Chem. Soc., Perkin
Trans. 1 2002, 733-736.
(11) (a) Kimball, D. B.; Weakley, T. J . R.; Herges, R.; Haley, M. M.
J . Am. Chem. Soc. 2002, 124, 13463-13473. (b) Kimball, D. B.;
Weakley T. J . R.; Haley, M. M. J . Org. Chem. 2002, 67, 6395-6405.
10.1021/jo049011r CCC: $27.50 © 2004 American Chemical Society
Published on Web 09/17/2004
J . Org. Chem. 2004, 69, 6979-6985
6979

Shirtcliff et al.
SCHEME 1
the B3LYP/6-31G* level of theory is a viable and reliable
method.11a,12,15 The calculated energy diagram of the
parent diazene 4 is shown in Figure 1. The energy
hypersurface is topologically similar to the parent
(ethynylphenyl)triazene.11a There are two modes of cy-
clization: a coarctate15 5-ring cyclization yielding a car-
bene as the primary product and a 6-ring cyclization
forming a zwitterionic intermediate. Both reactions
exhibit a barrier of activation that is 4-5 kcal mol-1 lower
than that in the previously investigated triazene deriva-
tive. Therefore, the cyclization temperature is expected
to be lower than was observed for the triazenes.
The transition structure of the coarctate cyclization
with the lowest barrier is shown in Figure 2. As expected
for an endothermic reaction the geometry is closer to
carbene product than to the reactant. With a length of
1.822 Å the C-N bond being formed during cyclization
is quite short. The transition structure is not flat and
therefore the question arises whether the cyclization is
a pseudocoarctate or a coarctate reaction. An ACID16
calculation revealed that all C-C and C-N bonds are
conjugated (Figure 3). The transition state, therefore, is
coarctate.
Following the current density vectors that are plotted
onto the ACID isosurface (Figure 3), the bond making
and breaking process can be described as a 10-electron
(4n + 2) process involving the indazole carbene system.
The diatropic ring current is constricted at the carbon
atom next to the carbene center (hence the name “coarc-
tate” reaction). At the coarctate center two orthogonal p
orbitals are involved in a delocalized system of electrons,
each contributing two electrons. At the carbene center,
the empty p orbital is also included. With 10 electrons
the reaction is thermochemically allowed. The phenyl
ring with its own diatropic ring current is in conjugation
with the delocalized system of electrons of the transition
state and probably lowers the barrier of activation
further.
In itia l Cycliza tion Stu d ies. The synthesis of sily-
lated (2-alkynylphenyl)diazenes 6-8, our starting ma-
terials in the initial cyclization studies, is outlined in
Scheme 2. Pd-catalyzed cross-coupling of iododiazene 917
with trimethylsilylacetylene (TMSA) under Sonogashira
conditions18 afforded disappointing yields of 6, with the
best result (50%) achieved with PdCl2(dppf) as catalyst.
Diazene 6 visibly degraded upon workup, which we
attribute to the lability of the TMS protecting group and
instability of the deprotected diazene (vide infra). To
circumvent this problem, the more robust triisopropylsilyl
and triethylsilyl groups were employed. Utilizing triiso-
propylsilylacetylene (TIPSA) and triethylsilylacetylene
(TESA) under standard conditions furnished the stable
diazenes 7 and 8 in 97% and 87% yield, respectively.
cyclization in 1,2-dichloroethane (DCE) to the five-
membered isoindazole (2) proceeds through a pseudoco-
arctate11a,12 pathway affording a carbene that can then
be trapped by molecular oxygen or an alkene. Cyclization
to the six-membered cinnoline (3), achieved by heating
to 200 °C in o-dichlorobenzene (ODCB), proceeds through
a cinnolinium zwitterionic intermediate that is then
trapped as the cinnoline by loss of N-ethylethanimine.11a
Density Functional Theory (DFT) calculations13 computed
at the B3LYP/6-31G* level of theory supported both the
carbene and dehydrocinnolinium mechanistic pathways.
Predicted by the relative energies, the isoindazole car-
bene is the kinetic intermediate and the dehydrocinno-
linium zwitterion is the thermodynamic intermediate.
This was proven experimentally by independently gen-
erating the isoindazole carbene and heating in ODCB to
yield the cinnoline, thus supporting the viability of DFT
calculations to assist in analyzing these cyclization
reactions.
Given the success of the triazene cyclizations, we
envisaged this type of heterocycle formation being ap-
plicable to a wide variety of compounds. Since DFT
calculations proved to be a useful tool in determining the
energy hypersurface in our initial studies, we utilized
DFT to help guide our synthetic strategy. By systemati-
cally changing the atoms in the parent triazene, a wide
variety of precursors for cyclization was elucidated. DFT
calculations were performed on this set of permutations
with the most viable precursor synthetically and com-
putationally being the phenyldiazene analogue of the
parent triazene. Although diazenes have previously been
utilized in transition metal-catalyzed cyclization reac-
tions in combination with external 1,2-disubstituted
alkynes,14 to the best of our knowledge, an intramolecular
cyclization of an o-alkynylphenyl-substituted diazenes
has never been reported. We present here the theoretical
and experimental results for the cyclization of a series
of (2-ethynylphenyl)phenyldiazenes.
Resu lts a n d Discu ssion
(15) (a) Herges, R. Angew. Chem., Int. Ed. Engl. 1994, 33, 255-
276. (b) Herges, R. J . Chem. Info. Comput. Sci. 1994, 34, 91-102. (c)
Berger, C.; Bresler, C.; Dilger, U.; Geuenich, D.; Herges, R. Angew.
Chem., Int. Ed. 1998, 37, 1850-1853. (d) Herges, R.; Geuenich, D.;
Bucher, G.; Toenshoff, C. Chem. Eur. J . 2000, 6, 1224-1228
(16) Herges, R.; Geuenich, D. J . Phys. Chem. A 2001, 105, 3214-
3220.
(17) Roling, P. V. J . Org. Chem. 1975, 40, 2421-2425.
(18) (a) Marsden J . A.; Haley, M. M. In Metal-Catalyzed Cross-
Coupling Reactions, 2nd ed.; de Meijere, A., Diederich, F., Eds.; Wiley-
VCH: Weinheim, 2004; pp 317-394. (b) Sonogashira, K. In Metal-
Catalyzed Cross-Coupling Reactions; Stang, P. J ., Diederich, F., Eds.;
VCH: Weinheim, 1997; pp 203-229.
Com p u ta tion a l Stu d ies. When performing theoreti-
cal calculations a compromise between computational
cost and accuracy (level of theory, basis set) must be met.
The results on the (2-ethynylphenyl)triazenes proved that
(12) Birney, D. M. J . Am. Chem. Soc. 2000, 122, 10917-10925.
(13) Koch, W.; Holthausen, M. C. A Chemist’s Guide to Density
Functional Theory; Wiley-VCH: Weinheim, 2000.
(14) (a) Kisch, H.; Wolski, A.; Knoch, F.; Halbritter, G. Angew.
Chem., Int. Ed. Engl. 1994, 33, 1603-1605. (b) Durr, U.; Heinemann,
F. W.; Kisch, H. J . Organomet. Chem. 1998, 558, 91-101.
6980 J . Org. Chem., Vol. 69, No. 21, 2004

Cyclization of (2-Ethynylphenyl)phenyldiazenes
F IGURE 1. DFT (B3LYP/6-31G* + ZPE) calculated relative energies of reactants, transition states, and products.
F IGURE 2. B3LYP/6-31G* optimized geometry of the transi-
tion state of the coarctate (5-ring) cyclization with the lowest
energy of activation (22.8 kcal mol-1).
Protiodesilylation of compound 7 was attempted with
Bu4NF (TBAF) in THF/MeOH. After aqueous workup and
subsequent solvent removal by heating under reduced
pressure, the solution of 4 changed from bright orange-
red to dark brown. No unreacted free alkyne was observ-
able by TLC and multiple unidentified degradation
products were present. Workup without heat, but em-
ploying filtration over a short pad of silica also led to
product degradation. Exposure of 4 to air for an extended
period of time (ca. 20-25 min) also promoted degrada-
tion. None of these problems were encountered in the
analogous studies of the parent (2-ethynylphenyl)triaz-
enes, which were stable under ambient conditions for
several months. It is possible that the degradation
products result from facile formation of the cyclized
carbene due to extended conjugation, but other degrada-
tion pathways are also possible.
Following conditions for the successful cyclization of 1
to 2, desilylation of 7 and immediate heating of crude 4
in DCE at 50 °C with 5.0 equiv of CuCl led to rapid
discoloration and formation of multiple compounds as
evident by TLC. Cooling 4 to 0 °C in DCE followed by
subsequent addition of 2.0 equiv of CuCl, however,
afforded the expected aldehyde 10 plus two other major
unidentified (and unanticipated) products in roughly
F IGURE 3. ACID plot of the coarctate transition state (see
Figure 2). Current density vectors are plotted onto the ACID
isosurface (isosurface value 0.045, magnetic field orthogonal
to the benzene ring of the indazole system and pointing toward
the reader). The “flow of electrons” is schematically depicted
in the structural formula (bottom left). The basis orbitals
involved in the 10 electron delocalized system are shown in
the orbital scheme (bottom right). The red line depicts the
topology of the orbital overlap.
equal yields (ca. 10% each, Scheme 2). Use of CuCl2 under
the same conditions furnished 10 in 39% yield and
substantially less of the other products; Cu(OAc)2 gave
very poor yields of all three compounds. Yields with
several other Cu as well as Rh salts as catalysts were
equally discouraging.
Recrystallization of the first unknown from CHCl3
afforded material suitable for X-ray diffraction (see
Supporting Information), which identified the compound
as alkyne dimer 11 (Scheme 2). This curious molecule
arises from the homodimerization of 4 followed by cy-
J . Org. Chem, Vol. 69, No. 21, 2004 6981

Shirtcliff et al.
SCHEME 2 a
SCHEME 3
SCHEME 4 a
a
Reagents and conditions: (a) BnNEt3ICl2, CaCO3, MeOH,
CH2Cl2; (b) PhNO, AcOH, 85 °C; (c) TESA, PdCl2(PPh3)2, CuI,
TEA, 50 °C.
a
Reagents and conditions: (a) trialkylsilylacetylene, PdCl2(PPh3)2
or PdCl2(dppf), CuI, TEA, 30-50 °C; (b) TBAF, MeOH, THF; (c)
“Cu” salt, DCE, 0 °C; (d) DCE, 75 °C.
result: instead of inserting into the phenyl C-H bond,
the carbene had indeed been trapped by inserting into
the O-H bond of residual TIPSOH, affording 12 in 61%
yield (Scheme 2). Although TIPSF is produced by the
desilylation reaction, the molecule is readily hydrolyzed
in the aqueous workup, giving TIPSOH. The low volatil-
ity and polarity of this latter material makes it difficult
to remove by normal vacuum or chromatographic means.
Unfortunately, tetracycle 5 was never observed in any
of our synthetic trials.
The formation of 12 in moderate yield suggested that
trapping the carbene with a silanol might be a viable
alternative to Cu-induced cyclization, which at best had
provided 10 in modest yield. For these experiments we
chose TESOH because of its greater volatility than
TIPSOH, consequently aiding in removal of the excess
silanol upon workup. Silylated precursor 8 was depro-
tected following the stringent conditions previously de-
termined, taken up in DCE, and immediately degassed.
An excess of TESOH (2.7 equiv) was added and the
reaction was allowed to stir at 75 °C for 3 h. After workup
of the reaction, the trapped isoindazole 13 was obtained
in a gratifying 77% yield (Scheme 3).
Dia zen e Syn th eses. With the success of the TESOH-
trapped isoindazole, we moved our attention to the
synthesis of a wide variety of substituted phenyl 2-(tri-
ethylsilylethynyl)phenyl diazene precursors. Compounds
14-17 were prepared in moderate to very good overall
yields for the three steps from the corresponding 4-sub-
stituted anilines (Scheme 4). Iodination of the anilines
with BnNEt3ICl2 and CaCO3 in MeOH/CH2Cl2 followed
by condensation with nitrosobenzene in AcOH at 85 °C
furnished iododiazenes 18-21.17,21 Synthesis of the more
electron-deficient diazenes 22 and 23, however, was not
possible via this route as the amine functionality was
clization of the resultant bis-diazenediyne. Similar re-
search in our laboratory has shown that homodimeriza-
tion of 1 and subsequent mild heating induces a double
cyclization to yield a bis-isoindazole dimer, with the
inclusion of CuCl not required for cyclization to occur.19
Definitive synthesis of 11 was carried out following the
method of Mori et al.20 Homodimerization of organosilane
6 in the presence of CuCl in the polar solvent DMSO at
30 °C furnished 11 in 45% yield; none of the uncyclized
homodimer was detected.
Along with searching for optimal conditions for cycliza-
tion to the 2H-indazolecarbaldehydes (10), conditions
favorable for cyclization and subsequent insertion of the
carbene into the phenyl C-H bond to furnish 5 were also
underway. To promote this insertion, we needed to heat
4 in an environment devoid of other molecules which
could function as carbene traps. Diazene 7 was depro-
tected with TBAF, subjected to aqueous workup, and
concentrated without heat. The red oil was taken up in
DCE and subjected to 4 freeze/pump/thaw cycles to
remove oxygen from the system. Heating 4 at 75 °C
overnight and subsequent workup afforded what was at
first thought to be the tetracyclic product 5, contaminated
with triisopropylsilanol (TIPSOH). The spectral data,
however, were consistent with the second of the previous
unknowns produced by the Cu-mediated cyclization. As
before, determination of the X-ray crystal structure (see
Supporting Information) provided another surprising
(19) Shirtcliff, L. D.; Hayes, A. G.; Schulze, K.; Ko¨hler, F.; Kimball,
D. B.; Herges, R.; Haley, M. M. Manuscript in preparation.
(20) (a) Ikegashira, K.; Nishihara, Y.; Hirabayashi, K.; Mori, A.;
Hiyama, T. Chem. Commun. 1997, 1039-1040. (b) Nishihara, Y.;
Ikegashira, K.; Hirabayashi, K.; Ando, J . I.; Mori, A.; Hiyama, T. J .
Org. Chem. 2000, 65, 1780-1787.
6982 J . Org. Chem., Vol. 69, No. 21, 2004

Cyclization of (2-Ethynylphenyl)phenyldiazenes
SCHEME 5 a
SCHEME 6 a
a
Reagents and conditions: (a) TBAF, THF, MeOH; (b) excess
TESOH, DCE, 75 °C.
TABLE 1. Yield s of Isoin d a zoles via Sch em e 6
s.m.
R
R′
isoindazolea
a
Reagents and conditions: (a) HCl, NaNO2, 3-iodophenol, KOH,
8
H
H
H
H
H
H
H
Br
F
36, 75%
37, 80%
38, 79%
39, 82%
40, 80%
41, 79%
42, 79%
43, 78%
44, 72%
45, 73%
46, 54%
MeCN, H2O; (b) MeI, Cs2CO3, DMF; (c) TESA, PdCl2(dppf), CuI,
DIPA, THF, 50 °C.
14
15
16
17
30
31
32
33
34
35
Me
t-Bu
Cl
CtCHb
OMe
OMe
OMe
OMe
OMe
OMe
effectively deactivated toward attack of the nitroso group.
Pd-catalyzed cross-coupling of 18-21 with TESA gave
precursors 14-17.
An alternative approach to diazene formation involved
capture of an arenediazonium electrophile with an acti-
vated arene. Our initial attempts at reacting the diazo-
nium salts of the iodinated anilines with a basic solution
of aqueous phenol failed to provide the diazenes in
satisfactory yield; however, reversing the placement of
the iodo substituent proved more successful (Scheme 5).
Thus, diazotization of several commercially available
4-substituted anilines and subsequent addition to a basic
solution of 3-iodophenol furnished the desired diazenes.
The hydroxyl groups were then converted to their methyl
ethers by addition of MeI and Cs2CO3 in DMF, affording
iododiazenes 24-29 in 45-77% yield for the two steps.
The methoxy functionality was chosen to circumvent
foreseeable problems that could be encountered if the
hydroxyl group was present during Sonogashira cross-
coupling. Also, due to the relatively high insolubility of
the polar diazenes, ease of workup and purification was
increased upon conversion to the methoxy derivatives.
Inclusion of the activating substituent on the iodoarene
ring, however, disfavored standard Sonogashira condi-
tions as cross-coupling with TESA and PdCl2(PPh3)2 as
catalyst suffered from lower than desirable yields. Utiliz-
ing the more active catalyst PdCl2(dppf) resulted in yields
that were increased typically by 15-20%, affording
ethynylated diazenes 30-35.
Op tim ized Isoin d a zole F or m a tion . With a variety
of synthesized diazenes in hand, we turned our attention
toward heterocycle formation. Deprotection, cyclization,
and subsequent removal of the silyl group to yield the
free -OH was carried out (Scheme 6) and the results are
summarized in Table 1. The yields of isoindazoles 36-
46 were very good for all functional groups attempted
except for the formation of 46, which bears the electron-
withdrawing nitro group. In this instance, isoindazole
formation was also accompanied by formation of several
low-yielding, unidentified side products. Unfortunately,
CO2Me
CN
NO2
a
b
Overall yield for the three synthetic steps. Both TES groups
are removed in step a.
SCHEME 7 a
a
Reagents and conditions: (a) TBAF, THF, MeOH; (b) excess
2,3-dimethyl-2-butene, 75 °C.
decreasing the cyclization temperature did not remedy
this problem. It should be noted that although there are
six synthetic steps from the commercially available
anilines to form the final isoindazoles, the overall yields
averaged around 40%.
Unlike the triazene cyclization reactions, which re-
quired inclusion of Cu salts to stabilize the reactive
species as the carbenoid, the diazenes underwent cy-
clization and concomitant formation of the free car-
bene, which could be intercepted with TESOH by inser-
tion into the O-H bond. To further confirm experimen-
tally formation of the free carbene, diazene 16 was
deprotected and cyclized with use of a well-known car-
bene trap, 2,3-dimethyl-2-butene, as solvent (Scheme
7). From the reaction mixture cyclopropane 47 was
obtained in 75% isolated yield, indeed confirming carbene
formation.
It is worth noting that in none of the diazene experi-
ments did we ever observe formation of the six-membered-
ring cinnoline, unlike what was observed in the triazene
(21) Kajigaeshi, S.; Kakinami, T.; Fujisaki, S.; Okamoto, T. Bull.
Chem. Soc. J pn. 1988, 61, 600-602.
J . Org. Chem, Vol. 69, No. 21, 2004 6983

Shirtcliff et al.
nitrosobenzene (1.7-2.0 equiv). The solution was heated to 85
°C for 40 h. The resulting mixture was cooled to room
temperature, diluted with EtOAc, and washed with brine (2×)
and H2O (3×). The organic layer was dried (MgSO4), filtered
over Celite, and concentrated in vacuo. The resulting oil was
dissolved in a 1:1 mixture of CH2Cl2:hexanes and filtered over
a short pad of silica. Column chromatography on silica gel gave
the desired product.
cyclizations. Heating to higher temperatures and allow-
ing for longer reaction times in a variety of solvents failed
to provide products derived from six-membered-ring
cyclization and generally resulted in complete decomposi-
tion. Whereas with a (2-ethynylphenyl)triazene intermo-
lecular deprotonation/elimination of the zwitterionic
intermediate afforded an imine and the neutral cinnoline,
the analogous reaction pathway for a (2-ethynylphenyl)-
phenyldiazene would produce a cinnoline along with
benzyne. Clearly, formation of such a high energy species
is disfavored, and thus only the coarctate pathway for
formation of the isoindazole carbene is observed.
Gen er a l Dia zen e/Met h oxy F or m a t ion P r oced u r e C.
The commercially available para-substituted aniline (1.0 equiv)
was dissolved in a minimal amount of MeCN. Concentrated
HCl (2.0 equiv) was added and a minimal amount of H2O was
added to keep the aniline in solution. The reaction mixture
was then cooled to below -5 °C in a brine/ice bath (monitored
internally). NaNO2 (1.2 equiv) was dissolved in a minimal
amount of MeCN:H2O (1:1) and slowly added to the reaction
mixture at a rate such that the temperature of the forming
diazonium solution remained below -2 °C. Upon completion,
the reaction was allowed to stir at -10 °C for 30 min. In a
separate flask, 3-iodophenol (1.1 equiv) and KOH (2.0 equiv)
were dissolved in the minimal amount of MeCN:H2O (10:1)
and cooled to below -5 °C in a brine/ice bath (monitored
internally). The cooled diazonium salt solution was then added
via cannula into the basic solution. Once addition was com-
plete, the reaction mixture was slowly allowed to warm to room
temperature over 3 h and stirring was continued at room
temperature for an additional 3 h. The reaction mixture was
then diluted with EtOAc and the aqueous layer acidified. The
organic layer was washed with 10% HCl solution (2×), brine
(2×), and H2O (1×). The organic layer was dried (MgSO4),
filtered over a short pad of silica (EtOAc), and concentrated
to give the hydroxydiazene.
Con clu sion s
A novel route for the selective synthesis of 2-phenyl-
2H-indazoles from previously unknown (2-alkynylphe-
nyl)phenyldiazenes has been developed. Unlike earlier
methods, these cyclizations are performed under neutral
conditions that allow for greater versatility and func-
tional group tolerance. Thus, selective trapping of the
isoindazole carbene with TESOH and subsequent desi-
lylation provides the 2-phenyl-2H-indazole-3-carbinols in
very good overall yield. We are currently working on
extending this new methodology to the construction of
other heterocyclic compounds.
Com p u ta tion a l Meth od s
The crude hydroxydiazene produced above was dissolved in
DMF (0.05 M) and Cs2CO3 (2.1 equiv) and MeI (1.2 equiv) were
added. The mixture was allowed to stir at ambient tempera-
ture for 40 h, diluted with 10% HCl solution, and extracted
with EtOAc (3×). The combined organics were washed with
H2O (1×), 0.5 M NaOH solution (2×), and H2O (2×). The
organic layer was dried (MgSO4), filtered over a short pad of
silica, and concentrated in vacuo. The solids were then
recrystallized from a EtOH/H2O mixture, filtered, and dried
to yield the desired product.
Gen er a l Acetylen e Cou p lin g P r oced u r e D. The (2-
iodophenyl)phenyldiazene (1 equiv), Pd(II) catalyst (0.02-0.05
equiv), CuI (0.1 equiv), and either (trimethylsilyl)acetylene,
(triisopropylsilyl)acetylene, or (triethylsilyl)acetylene (1.4 equiv)
were dissolved in an amine base (0.1 M solution based on
diazene). The mixture was immediately degassed by three
successive freeze-pump-thaw cycles, and the flask was
charged with N2. The mixture was heated to 50 °C and
stirred under N2 overnight. After cooling, the mixture was
filtered over a short pad of silica (CH2Cl2) and concentrated
in vacuo. Column chromatography on silica gel gave the
desired product.
All theoretical calculations have been performed with the
Gaussian 98 suite of programs22 at the B3LYP/6-31G* level23
of DFT. All stationary points were confirmed by harmonic
frequency analysis, and the energies of the stationary points
were determined, including zero point energies at the same
level of theory. ACID scalar fields were computed with our
own program.16 Current density vectors were calculated with
the CSGT method of Keith and Bader.24
Exp er im en ta l Section
Gen er a l. See Supporting Information.
Gen er a l Iod in a tion P r oced u r e A. The starting aniline
(1 equiv), BnNEt3ICl2 (1.15 equiv), and CaCO3 (1.4 equiv) were
dissolved in 5:1 CH2Cl2:MeOH (0.1 M) and the solution was
stirred at room temperature for 8 h. The resulting mixture
was filtered, and the solvent was evaporated. The crude
product was redissolved in Et2O and washed successively with
a NaHSO3 solution (10 wt %), brine, and H2O. The organic
layer was dried (MgSO4), filtered over a short pad of silica,
and concentrated to afford the desired product in sufficiently
pure form for further use.
Gen er a l Dep r otection /TESOH Cycliza tion /Desilyla -
t ion P r oced u r e E . The (2-triethylsilylethynylphenyl)phe-
nyldiazene was dissolved in a mixture of THF:MeOH (20:1,
0.1 M). TBAF (1 M solution in THF, 2.0 equiv) was added and
the reaction was allowed to stir at room temperature for 1-3
min. The solution was diluted with Et2O, and successively
washed with a saturated aqueous NH4Cl solution (4×), brine
(3×), and H2O (4×). The organic layer was dried (MgSO4),
filtered over Celite under a constant stream of N2, and
concentrated in vacuo with no heat. The desilylated product
was taken up in DCE (0.05 M) and immediately degassed by
two successive freeze-pump-thaw cycles. TESOH (2.7 equiv)
was added and the reaction was degassed by two additional
freeze-pump-thaw cycles. The flask was placed in a 75 °C
preheated sand bath and allowed to stir under vacuum for 3
h. After cooling to room temperature, the crude 2-phenyl-3-
triethylsilanyloxymethyl-2H-indazole was dissolved in a mix-
ture of THF:MeOH (20:1, 0.1 M). Excess TBAF (1 M solution
Gen er a l Dia zen e P r oced u r e B. The iodinated aniline (1.0
equiv) was dissolved in AcOH (0.1 M) and to this was added
(22) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J . R.; Zakrzewski, V. G.; Montgomery, J . A.,
J r.; Stratmann, R. E.; Burant, J . C.; Dapprich, S.; Millam, J . M.;
Daniels, A. D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J .;
Barone, V.; Cossi, M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo,
C.; Clifford, S.; Ochterski, J .; Petersson, G. A.; Ayala, P. Y.; Cui, Q.;
Morokuma, K.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.;
Foresman, J . B.; Cioslowski, J .; Ortiz, J . V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.; Martin, R. L.;
Fox, D. J .; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Gonzalez, C.; Challacombe, M.; Gill, P. M. W.; J ohnson, B.; Chen, W.;
Wong, M. W.; Andres, J . L.; Gonzalez, C.; Head-Gordon, M.; Replogle,
E. S.; Pople, J . A. Gaussian 98, Revision A.6; Gaussian, Inc.: Pitts-
burgh, PA, 1998.
(23) Becke, A. D. J . Chem. Phys. 1993, 98, 5648-5652.
(24) Keith, B. A.; Bader, R. F. W. J . Chem. Phys. 1993, 99, 3669-
3682.
6984 J . Org. Chem., Vol. 69, No. 21, 2004

Cyclization of (2-Ethynylphenyl)phenyldiazenes
in THF) was added and the reaction was allowed to stir at
room temperature until the complete desilylation of isoindazole
had taken place (generally 1-5 min). Progress was monitored
by TLC. Upon completion, the solution was diluted with Et2O
and washed successively with saturated aqueous NH4Cl solu-
tion (4×), brine (3×), and H2O (4×). The organic layer was
dried (MgSO4), filtered over a short pad of silica (Et2O), and
concentrated in vacuo to afford the product.
an IGERT fellowship. We thank J azmin Rivers for her
help in preparing and purifying selected compounds.
Su p p or tin g In for m a tion Ava ila ble: Synthetic details
and spectral data for 6-21 and 24-47; copies of 1H or 13C
NMR spectra for 8, 10, 11, 13, 14-17, and 30-47; X-ray
structures of 11 and 12, structure refinement details, tables
of atomic coordinates, thermal parameters, bond lengths, bond
angles, torsion angles, and mean planes, Cartesian coordinates
for all optimized structures in Figure 1. This material is
available free of charge via the Internet at http://pubs.acs.org.
Ack n ow led gm en t. We thank the National Science
Foundation and the Fonds der Chemischen Industrie
for financial support. L.D.S. acknowledges the NSF for
J O049011R
J . Org. Chem, Vol. 69, No. 21, 2004 6985