Deciphering the mechanistic dichotomy in the cyclization of 1-(2-ethynylphenyl)-3,3-dialkyltriazenes: Competition between pericyclic and pseudocoarctate pathways

Article abstract of DOI:10.1021/ja027809r

The mechanistic aspects of the cyclization of (2-ethynylphenyl)triazenes under both thermal and copper-mediated conditions are reported. For cyclization to an isoindazole, a carbene mechanistic pathway is proposed. The carbene intermediate can react with oxygen, dimerize to give an alkene, or be trapped either intermolecularly (using 2,3-dimethyl-2-butene to generate a cyclopropane) or intramolecularly (using a biphenyl moiety at the terminus of the acetylene to form a fluorene). Density-functional theory (DFT) calculations support a pseudocoarctate pathway for this type of cyclization. Thermal cyclization to give a cinnoline from (2-ethynylphenyl)triazenes is proposed to occur through a pericyclic pathway. DFT calculations predict a zwitterionic dehydrocinnolinium intermediate that is supported by deuterium trapping studies as well as cyclizations performed using a 2,2,6,6-tetramethylpiperidine moiety at the 3-position of the triazene.

Full text of DOI:10.1021/ja027809r

Published on Web 10/19/2002
Deciphering the Mechanistic Dichotomy in the Cyclization of
1-(2-Ethynylphenyl)-3,3-dialkyltriazenes: Competition between
Pericyclic and Pseudocoarctate Pathways
David B. Kimball,† Timothy J. R. Weakley,† Rainer Herges,*,‡ and
Michael M. Haley*,†
Contribution from the Department of Chemistry, UniVersity of Oregon,
Eugene, Oregon 97403-1253, and Institut fu¨r Organische Chemie, UniVersita¨t Kiel,
24098 Kiel, Germany
Received July 22, 2002
Abstract: The mechanistic aspects of the cyclization of (2-ethynylphenyl)triazenes under both thermal and
copper-mediated conditions are reported. For cyclization to an isoindazole, a carbene mechanistic pathway
is proposed. The carbene intermediate can react with oxygen, dimerize to give an alkene, or be trapped
either intermolecularly (using 2,3-dimethyl-2-butene to generate a cyclopropane) or intramolecularly (using
a biphenyl moiety at the terminus of the acetylene to form a fluorene). Density-functional theory (DFT)
calculations support a pseudocoarctate pathway for this type of cyclization. Thermal cyclization to give a
cinnoline from (2-ethynylphenyl)triazenes is proposed to occur through a pericyclic pathway. DFT calculations
predict a zwitterionic dehydrocinnolinium intermediate that is supported by deuterium trapping studies as
well as cyclizations performed using a 2,2,6,6-tetramethylpiperidine moiety at the 3-position of the triazene.
Scheme 1
Introduction
The chemistry of triazenes has received considerable attention
in the recent literature.1 These compounds are useful as linkers
to solid supports,2 as protecting groups for amines,3 and as
precursors to haloarenes.4 Among these and numerous other
uses,5 our group recently reported a new method for the syn-
thesis of heterocycles via the cyclization of 1-(2-ethynylphenyl)-
3,3-dialkyltriazenes (e.g., 1, Scheme 1).6 Heating 1 to 170 °C
in o-dichlorobenzene (ODCB) gave both an isoindazole (2) and
a cinnoline (3), both of which could be generated independently
in good to excellent yields by either adding CuCl or heating 1
to 200 °C, respectively.6c These reactions are tolerant of a wide
variety of functional groups.
* Address correspondence to these authors. E-mail: haley@oregon.
uoregon.edu; rherges@oc.uni-kiel.de.
† University of Oregon.
‡ Universita¨t Kiel.
(1) Kimball, D. B.; Haley, M. M. Angew. Chem., Int. Ed. 2002, 41, 3338-
3351.
The mechanistic aspects of these unusual cyclizations are
atypical of more traditional syntheses of isoindazoles and
cinnolines. The concurrent production of both heterocycles from
the same triazene in the same pot is unique. Neutral conditions
and halocarbon solvents do not favor the vast majority of
isoindazole syntheses, which usually require mineral acids and
activated hydrazines.7 These types of reactions can be explained
through ionic transitions states as well as electrophile-nucleo-
phile interactions. No examples of cyclization between two fairly
unreactive functional groups (a triazene and a phenylacetylene)
without preactivation8 (conversion to a diazonium, substitution
on the acetylene) have been reported. Cinnoline formation also
usually requires a diazonium species ortho to an activated
methylene.9 In cases where an acetylene is used,10 a halogen or
(2) (a) Bra¨se, S.; Dahmen, S.; Pfefferkorn, M. J. Comb. Chem. 2000, 2, 710-
715. (b) Rademann, J.; Smerdka, J.; Jung, G.; Grosche, P.; Schmid, D.
Angew. Chem., Int. Ed. 2001, 40, 381-385. (c) Dahmen, S.; Bra¨se, S. Org.
Lett. 2000, 2, 3563-3565. (d) Schunk, S.; Enders, D. Org. Lett. 2000, 2,
907-910.
(3) (a) Gross, M. L.; Blank, D. H.; Welch, W. M. J. Org. Chem. 1993, 58,
2104-2109. (b) Bra¨se, S.; Ko¨bberling, J.; Enders, D.; Lazny, R.; Wang,
M.; Bandtner, S. Tetrahedron Lett. 1999, 40, 2105-2108. (c) Lazny, R.;
Sienkiewicz, M.; Bra¨se, S. Tetrahedron 2001, 57, 5825-5832. (d)
Bursavich, M. G.; Rich, D. H. Org. Lett. 2001, 3, 2625-2628.
(4) Inter alia: (a) Zhu, Z.; Moore, J. S. J. Org. Chem. 2000, 65, 116-123. (b)
Tour, J. M. Acc. Chem. Res. 2000, 33, 791-804. (c) Wautelet, P.; Moroni,
M.; Oswald, L.; Le Moigne, J.; Pham, A.; Bigot, J.-Y. Macromolecules
1996, 29, 446-455. (d) Khalaj, A.; Beiki, D.; Rafiee, H.; Najafi, R. J.
Labelled Compd. Radiopharm. 2001, 44, 235-240.
(5) Inter alia: (a) Rouzer, C. A.; Sabourin, M.; Skinner, T. L.; Thompson, E.
J.; Wood, T. O.; Chmurny, G. N.; Klose, J. R.; Roman, J. M.; Smith, R.
H., Jr.; Michejda, C. J. Chem. Res. Toxicol. 1996, 9, 172-178. (b) Nicolaou,
K. C.; Boddy, C. N. C.; Li, H.; Koumbis, A. E.; Hughes, R.; Natarajan, S.;
Jain, N. F.; Ramanjulu, J. M.; Bra¨se, S.; Solomon, M. E. Chem.sEur. J.
1999, 5, 2602-2621. (c) Nicolaou, K. C.; Boddy, C. N. C.; Natarajan, S.;
Yue, T.-Y.; Li, H.; Bra¨se, S.; Ramanjulu, J. M. J. Am. Chem. Soc. 1997,
119, 3421-3422. (d) Wirshun, W.; Winkler, M.; Lutz, K.; Jochims, J. C.
J. Chem. Soc., Perkin Trans. 2 1998, 1755-1762. (e) Connelly, N. G.;
Davis, P. R. G.; Harry, E. E.; Klangsinsirikul, P.; Venter, M. J. Chem.
Soc., Dalton Trans. 2000, 2273-2277.
(6) (a) Kimball, D. B.; Hayes, A. G.; Haley, M. M. Org. Lett. 2000, 2, 3825-
3827. (b) Kimball, D. B.; Herges, R.; Haley, M. M. J. Am. Chem. Soc.
2002, 124, 1572-1573. (c) Kimball, D. B.; Weakley, T. J. R.; Haley, M.
M. J. Org. Chem. 2002, 67, 6395-6405.
9
10.1021/ja027809r CCC: $22.00 © 2002 American Chemical Society
J. AM. CHEM. SOC. 2002, 124, 13463-13473
13463

A R T I C L E S
Scheme 2 a
Kimball et al.
Figure 1. Molecular structure of trimer 6; ellipsoids are drawn at the 30%
probability level.
(e.g., 2). Copper salts have been widely known to stabilize
carbene intermediates and to help generate products at reason-
able temperatures.12 Our first attempts at cyclization in the
presence of copper salts gave higher yields of isoindazole but
with several new byproducts. These reactions were done at 110
°C, using CuI, triazene 1a (R ) CN), and ODCB as the solvent
(Scheme 2). The major byproduct was isolated, and a crystal
structure was obtained (Figure 1; see Supporting Information
for structure details). This new compound (6) contained a central
tricyclic structure that was formed by the cyclization of three
ethynylphenyltriazene units. Again, the connectivity that resulted
could not readily be explained without involving a carbene
intermediate(s).
When the temperature was lowered to 50 °C or to ambient
temperature, cyclization in the presence of CuI resulted in much
higher yields of isoindazole. Although a number of Cu salts
(as well as Zn and Rh salts) were examined, optimal conditions
were found to be 1,2-dichloroethane as the solvent, CuCl as
the carbene stabilizer, and a temperature of 50 °C. These
conditions gave excellent yields of isoindazole for every
compound studied.6c Although using 2-3 equiv of CuCl results
in complete, exclusive production of isoindazole, the reaction
can be done in the presence of catalytic amounts of CuCl (10
mol %) using a longer reaction time (ca. 3-4 days).
With ideal conditions for the exclusive, high yielding produc-
tion of isoindazoles, we turned our attention to the use of
trapping agents to provide additional evidence of a carbene
intermediate. Cyclization of triazene 1b (R ) t-Bu) in the
presence of 2,3-dimethyl-2-butene and CuCl gave isoindazole
cyclopropane 7 in 65% yield (Scheme 3). This product undoubt-
edly resulted from carbene (carbenoid) formation and intermo-
lecular trapping by the electron-rich alkene.
To lend further evidence to the carbene intermediate, we
attempted cyclization to the isoindazole with a starting com-
pound that could trap the carbene intramolecularly. Saito and
co-workers13 have shown that the use of a biphenyl moiety
a
Reagents and conditions: (a) MeI, 145 °C; (b) CuI, ODCB, 110 °C;
(c) CuCl, DCE, rt.
hydroxide ion is needed to cyclize in typical Richter fashion11
(i.e., simultaneous attack of the nucleophile on the alkyne carbon
proximal to the phenyl ring and cyclization of the distal alkyne
carbon with the diazonium group in a pseudo-Michael-type
fashion). To study the generality of our new cyclizations for
the practical synthesis of these and other heterocycles, we
needed to determine the mechanisms for both transformations.
The following represents the full experimental and theoretical
details of our efforts to delineate the mechanistic pathways
responsible for the formation of isoindazoles and cinnolines from
1-(2-ethynylphenyl)-3,3-dialkyltriazenes.
Results and Discussion
Isoindazole Production. Of the several possible types of
mechanisms responsible for the formation of an isoindazole from
the cyclization of 1-(2-alkynylphenyl)-3,3-dialkyltriazenes, the
one that produced a carbene intermediate (4, Scheme 2) was
most supported by the products obtained. Carbenes are known
to both dimerize to alkenes (e.g., 5) and trap molecular oxygen
(7) (a) Elderfield, R. C. In Heterocyclic Compounds; Elderfield, R. C., Ed.;
John Wiley & Sons: New York, 1957; Vol. 5, pp 162-193. (b) Behr, L.
C.; Fusco, R.; Jarboe, C. H. In Pyrazoles, Pyrazolines, Pyrazolidines,
Indazoles and Condensed Rings; Wiley, R. H., Ed.; International Wiley
Science: New York, 1967; pp 289-382. For more recent examples,
see: (c) Sun, J. H.; Teleha, C. A.; Yan, J. S.; Rodgers, J. D.; Nugiel, D. A.
J. Org. Chem. 1997, 62, 5627-5629. (d) Baraldi, P. G.; Cacciari, B.;
Spalluto, G.; Romagnoli, R.; Braccioli, G.; Zaid, A. N.; de las Infantes,
M. J. P. Synthesis 1997, 1140-1142. (e) Molina, P.; Arques, A.;
Vinader, M. V. J. Org. Chem. 1990, 55, 4724-4731. (f) Dell’Erba, C.;
Novi, M.; Petrillo, G.; Tavani, C. Tetrahedron 1994, 50, 3529-3536. (g)
Kim, J. I.; Kim, B. C.; Moon, S. W.; Jahng, Y. Heterocycles 1995, 41,
1471-1478.
(8) (a) Prikhodko, T. A.; Vasilevsky, S. F. MendeleeV Commun. 1996, 98-
99. (b) Prikhodko, T. A.; Vasilevsky, S. F. MendeleeV Commun. 1998,
149-150. (c) Prikhodko, T. A.; Vasilevsky, S. F. Russ. Chem. Bull. Int.
Ed. 2001, 50, 1268-1273.
(9) (a) Simpson, J. C. E. In The Chemistry of Heterocyclic Compounds.
Condensed Pyridazine and Pyrazine Rings (Cinnolines, Phthalazines, and
Quinoxalines); Weisberger, A., Ed.; Interscience Publishers: New York,
1953; pp 3-65. (b) van der Plas, H. C.; Buurman, D. J.; Vos, C. M. Recl.
TraV. Chim. Pays-Bas 1978, 97, 50-51.
(12) (a) Stechl, H. Chem. Ber. 1964, 97, 2681-2688. (b) Nozaki, H.; Moriuti,
S.; Takaya, H.; Noyori, R. Tetrahedron Lett. 1966, 5239-5244.
(13) (a) Nakatani, K.; Adachi, K.; Tanabe, K.; Saito, I. J. Am. Chem. Soc. 1999,
121, 8221-8228. (b) See also: Frey, L. F.; Tillyer, R. D.; Ouellet, S. G.;
Reamer, R. A.; Grabowski, E. J. J.; Reider, P. J. J. Am. Chem. Soc. 2000,
122, 1215-1216.
(10) (a) Vasilevsky, S. F.; Tretyakov, E. V. Liebigs Ann. Chem. 1995, 775-
779. (b) Tretyakov, E. V.; Knight, D. W.; Vasilevsky, S. F. J. Chem. Soc.,
Perkin Trans. 1 1999, 24, 3721-3726. (c) Bra¨se, S.; Dahmen, S.; Heuts,
J. Tetrahedron Lett. 1999, 40, 6201-6203.
(11) von Richter, V. Chem. Ber. 1883, 16, 677-683.
9
13464 J. AM. CHEM. SOC. VOL. 124, NO. 45, 2002

Cyclization of 1-(2-Ethynylphenyl)-3,3-dialkyltriazenes
A R T I C L E S
Scheme 3
Scheme 4 a
a Reagents and conditions: (a) 2-iodobiphenyl, PdCl2(PPh3)2, CuI, Et3N,
50 °C; (b) CuCl, DCE, 90 °C.
Figure 2. Molecular structure of fluorene 9; ellipsoids are drawn at the
30% probability level.
allows carbenes to be trapped as fluorenes. Ethynylphenyltria-
zene 1c (R ) Me) was coupled to 2-iodobiphenyl14 under
Sonogashira conditions to give 8 in 75% yield (Scheme 4).
Cyclization of 8 in the presence of CuCl produced fluorene 9
in 55% yield. The structure of 9 was confirmed by X-ray
crystallography (Figure 2). These results strongly support a
carbene intermediate in the formation of the isoindazole.
Experimental and theoretical results indicate that the forma-
tion of the carbene intermediate in the cyclization of ethy-
nylphenyltriazenes (1) proceeds via a concerted pseudocoarctate
pathway and that the cinnoline cyclization follows a pericyclic
mechanism. Since there is some confusion in the literature,
the terms pericyclic, pseudopericyclic, coarctate, and pseudoco-
arctate should be clarified at this point. Pericyclic reactions,
by definition, exhibit a cyclic (aromatic) transition state of
delocalized electrons.17-19 Bond making and bond breaking
occur simultaneously in a cyclic array.20 Bond making and
breaking in coarctate reactions21 do not follow a cyclic path.
As opposed to pericyclic reactions, there is at least one atom at
which two bonds are made and two bonds are broken simul-
taneously. For the five-membered ring cyclization (Scheme 6),
the coarctate atom is the carbon neighboring the carbene center.
At this carbon, two bonds are broken (triple bondfsingle bond)
and two bonds are made (new CsN bond and CdC double
bond in five-membered ring) simultaneously; thus, there is an
exocyclic part in the bond formation process. The reaction is
therefore not electrocyclic since this term is reserved only for
pericyclic reactions.
As Lemal et al. stated as early as 1976,22 within pericyclic
reactions there are special cases. Even though a reaction (e.g.,
Diels-Alder) from a formal point of view may look like a
pericyclic reaction and follow a concerted mechanism, it does
not necessarily have to exhibit a cyclic, aromatic transition state.
Lemal termed these reactions pseudopericyclic. Numerous
experimental and theoretical investigations have been performed
on such systems,16,23 some of them controversial.24 In pseudo-
pericyclic reactions, the cyclic delocalization of electrons in the
To better understand the reaction energetics and mechanisms,
the stationary points (reactants, transition states, and products)
for the cyclization of 1-(2-ethynylphenyl)-3,3-dialkyltriazenes
to isoindazoles and cinnolines were calculated using DFT
(Figure 3). In the calculated model system, the NEt2 group was
replaced by NH2 to save computational costs and to reduce the
complexity of the energy hypersurface. The energy barrier for
isoindazole formation was calculated to be ca. 2.0 kcal mol-1
lower than that for the cinnoline. The cinnoline intermediate,
however, was predicted to be much lower in energy than the
isoindazole carbene. This agreed well with experimental re-
sults: cyclizations at 170 °C gave a prototypical kinetic-
thermodynamic product distribution of isoindazole and cinno-
line. This also explains why the cinnoline product could be
obtained exclusively in high yield when the cyclization was done
at higher temperatures. An alternative pathway leading to the
cinnoline involving an ethyne-vinylcarbene rearrangement was
calculated to have a much higher barrier (46.1 kcal mol-1; see
Supporting Information).
Key to providing evidence for our proposed pathways,
however, was the demonstration of reversibility. If the isoin-
dazole carbene intermediate was in equilibrium with the starting
material, then independent generation of this carbene at 200
°C should provide the corresponding cinnoline species. The
sodium salts of tosylhydrazones are known to eliminate N2
readily to generate carbenes at temperatures close to 200 °C.15
We thus prepared the tosylhydrazone of isoindazole 2d (R )
Cl, Scheme 5). Heating the salt resulting from treatment of
hydrazone 10 with NaH to 200 °C in ODCB resulted in good
conversion to the corresponding cinnoline 3d (R ) Cl),
presumably through regeneration of the ethynylphenyltriazene
from the carbene.
(16) Birney, D. M. J. Am. Chem. Soc. 2000, 122, 10917-10925.
(17) Herges, R.; Jiao, H.; Schleyer, P. v. R. Angew. Chem., Int. Ed. Engl. 1994,
33, 1376-1378.
(18) (a) Evans, M. G.; Warhurst, E. Trans. Faraday Soc. 1938, 34, 614-624.
(b) Evans, M. G. Trans. Faraday Soc. 1939, 35, 824-834. (c) Evans, M.
G.; Polanyi, M. Trans. Faraday Soc. 1938, 34, 11-29.
(19) (a) Dewar, M. J. S. Angew. Chem., Int. Ed. Engl. 1971, 10, 761-776. (b)
Zimmerman, H. E. Acc. Chem. Res. 1971, 4, 272-280.
(20) Woodward, R. B.; Hoffmann, R. Angew. Chem., Int. Ed. Engl. 1969, 8,
781-853.
(21) (a) Review on coarctate reactions: Herges, R. Angew. Chem., Int. Ed. Engl.
1994, 33, 255-276. (b) Coarctate is derived from the Latin coarctatus,
past participle of coarctare: to press together, to constrict. (c) Herges, R.;
Geuenich, D.; Bucher, G.; To¨nshoff, C. Chem.sEur. J. 2000, 6, 1224-
1228. (d) Herges, R. J. Comput. Chem. 1994, 43, 91-106.
(22) Ross, J. A.; Seiders, R. P.; Lemal, D. M. J. Am. Chem. Soc. 1976, 98,
4325-4327.
(14) John, J. A.; Tour, J. M. Tetrahedron 1997, 53, 15515-15534.
(15) Cummins, J. A.; Porter, T. A.; Jones, M., Jr. J. Am. Chem. Soc. 1998, 120,
6473-6476.
9
J. AM. CHEM. SOC. VOL. 124, NO. 45, 2002 13465

A R T I C L E S
Kimball et al.
Figure 3. DFT (B3LYP/6-31G* + ZPE) calculated relative energies of reactants, transition states, and products of the cyclization to the isoindazole carbene
and the cinnolinium zwitterion.
Scheme 5
disconnections can also occur in coarctate reactions. In analogy
to the pseudopericyclic reactions, we term these reactions
pseudocoarctate.
Our ACID method (anisotropy of induced current density)25
is an excellent tool to distinguish between pericyclic/pseudo-
pericyclic and coarctate/pseudocoarctate reactivity. The ACID
scalar field is interpreted as the density of delocalized electrons.
A cyclic topology in an ACID plot indicates a pericyclic
reaction, and a noncyclic but contiguous delocalized system
(constricted or coarctate cycle) indicates a coarctate system.
Disconnections that are characteristic for pseudopericyclic and
pseudocoarctate systems are immediately visible by a discon-
nection in the (otherwise) contiguous system of the ACID
boundary surface.
Scheme 6. Theoretically (B3LYP/6-31G*, ACID) Investigated
Cyclization Reactions
Figure 4a presents the ACID isosurface of the five-membered
ring cyclization of the parent ethynylphenyltriazene. For com-
parison, the ACID plot of the all-carbon analogue of the
transition state of the five-ring cyclization is shown in Figure
4b. The delocalized system of electrons in the transition state
of the isoindazole cyclization (Figure 4a) is not pericyclic
because it involves the carbene center that is exocyclic to the
five-membered ring. The coarctate topology of the transition
state, however, is disconnected between the N- and the C-atom
at which the new bond is formed. It is worth noting that the
current density vectors do not describe a circle (ring current) in
the forming heterocyclic ring. The π system of the NdN and
the π orbital of the triple bond that is involved in the bond
formation are orthogonal with respect to each other; thus, the
reaction is pseudocoarctate. Interestingly, similar to many
pseudopericyclic systems, the all-carbon analogue (Figure 4b)
transition state is interrupted (disconnected) because the orbitals
involved in the delocalized system are orthogonal at these points.
Consequently, these reactions do not follow the Woodward-
Hoffmann rules and in some cases nevertheless have surprisingly
low activation barriers. Birney16 recently discovered that such
(23) (a) Wagenseller, P. E.; Birney, D. M.; Roy, D. J. Org. Chem. 1995, 60,
2853-2859. (b) Birney, D. M. J. Org. Chem. 1996, 61, 243-251. (c)
Fabian, W. M. F.; Bakulev, V. A.; Kappe, C. O. J. Org. Chem. 1998, 63,
5801-5805. (d) Birney, D. M.; Ham, S.; Unruh, G. R. J. Am. Chem. Soc.
1997, 119, 4509-4517. (e) Birney, D. M.; Wagenseller, P. E. J. Am. Chem.
Soc. 1994, 116, 6262-6270. (f) Birney, D. M.; Xu, X.; Ham, S. Angew.
Chem., Int. Ed. 1999, 38, 189-193. (g) Zipse, H. J. Chem. Soc., Perkin
Trans. 2 1997, 1797-1800. (h) Shumway, W.; Ham, S.; Whittlesey, B.
R.; Birney, D. M. J. Org. Chem. 2000, 65, 7731-7739.
(24) (a) de Lera, A. R.; Alvarez, R.; Lecea, B.; Torrado, A.; Cossio, F. P. Angew.
Chem., Int. Ed. 2000, 40, 557-561. (b) Rodriguez-Otero, J.; Cabaleiro-
Lago, E. M. Angew. Chem., Int. Ed. 2002, 41, 1147-1150. (c) de Lera, A.
R.; Cossio, F. P. Angew. Chem., Int. Ed. 2002, 41, 1150-1152.
(25) (a) Herges, R.; Geuenich, D. J. Phys. Chem. A 2001, 105, 3214-3220. (b)
Herges, R.; Papafilippopoulos, A. Angew. Chem., Int. Ed. 2001, 40, 4671-
4674.
9
13466 J. AM. CHEM. SOC. VOL. 124, NO. 45, 2002

Cyclization of 1-(2-Ethynylphenyl)-3,3-dialkyltriazenes
A R T I C L E S
Figure 4. (a) ACID plot of the transition state of the cyclization of (2-ethynylphenyl)triazene to 2-amino-3-methylideneisoindazole. The topology of delocalized
electrons exhibits a disconnection between the C and N atoms at which the new C-N bond is formed. The current density vectors (green arrows with red
tips) do not exhibit a closed circle in the five-membered ring. Therefore, the transition state is pseudocoarctate in contrast to the all-carbon analogue in panel
b. (b) ACID plot of the transition state of the cyclization of 2-ethynylstyrene to 1-methylidene-2H-indene, the parent all-carbon analogue of the title isoindazole
cyclization. The current density vectors, which are plotted onto the ACID isosurface, indicate the coarctate nature of the transition state. The ring current
forms an exocyclic loop involving the carbon atom that becomes the carbene center in the product; thus, the topology of the delocalized system of electrons
corresponds to a constricted or coarctate cycle. The CIV of 0.054 between the two carbon atoms at which the new bond is formed indicates that there is no
disconnection; hence, the reaction is genuine coarctate in contrast to the title isoindazole cyclization (panel a) that is pseudocoarctate.
of the isoindazole cyclization does not exhibit a disconnection
in the transition state and thus is a true coarctate reaction. In
contrast to the nitrogen system (Figure 4a), the transition state
(Figure 4b) is nonplanar, and the orbitals involved are not
orthogonal. To gain further insight into the electronic structure,
we plotted the current density vectors onto the ACID isosurface.
If one follows the current density vectors, the coarctate nature
of the delocalized system in Figure 4b becomes immediately
evident. The diatropic ring current forms a hairpin loop
involving the C-atom that becomes the carbene center in the
product. The delocalized system corresponds to a constricted
cycle, hence the name coarctate.21b
3,3-dialkyltriazenes readily suggested a mechanism responsible
for the generation of a cinnoline species. The most common
methods for cinnoline synthesis give 4-substituted products as
the result of a Michael-type nucleophilic attack of an alkyne
ortho to a diazonium functionality.9 No such products were
observed even when a hydroxide source was present. No source
of H+ or H- was present to induce cyclization nor to aid in the
elimination of Et2NH. Simply heating the starting alkynylphen-
yltriazenes in ODCB to 200 °C in a sealed glass pressure tube
gave the cinnoline products in excellent yield.6c NMR scale
experiments confirmed that proton or hydride abstraction was
not obtained from the solvent.
Our first inclination was that cinnoline production occurred
by means of a Bergman-type26 reaction to give a 1,4-diradical
species. Bergman has shown that these systems can scavenge
H• from the glass reaction flasks. Reactions performed in the
presence of radical donors such as dihydroanthracene-d427 and
The ACID plot of the transition state of the six-membered
ring cyclization (Figure 5a) exhibits a cyclic topology of
delocalized electrons and thus is a pericyclic system. The
connection between the two bond-forming centers, however, is
rather weak. To quantify the degree of conjugation, we defined
the critical isosurface value (CIV). In the case of the six-ring
cyclization, the CIV value in forming the C-N bond is 0.045.
Compared to the corresponding CIV value of the parent Diels-
Alder reaction (0.069), this indicates that the aromaticity in the
transition state is rather weak and that the reaction is a borderline
case between pericyclic and pseudopericyclic. Similar to the
five-ring cyclization, the all-carbon analogue (Figure 5b) exhibits
a strong cyclic, aromatic, and thus pericyclic transition state
(CIV ) 0.072). The current density vectors plotted onto the
ACID isosurface reveal a strong diatropic ring current as
expected for an aromatic system.
26
1,4-cyclohexadiene-d4 gave neither deuterated products nor
materials that incorporated the dihydroanthracene or cyclohexa-
diene structure. We also synthesized several triazenes with R
groups at N3 capable of intramolecularly trapping a radical at
N1 of the cinnoline. Triazene 11 provides a representative
example (Scheme 7). The diazonium salt of iodoaniline was
treated with amine 1228 and K2CO3 to generate triazene 13 in
91% yield. Reaction of 13 with TMSA under Sonogashira
(26) (a) Jones, R. R.; Bergman, R. G. J. Am. Chem. Soc. 1972, 94, 660-661.
(b) Johnson, G. C.; Stofko, J. J., Jr.; Lockhart, T. P.; Brown, D. W.;
Bergman, R. G. J. Org. Chem. 1979, 44, 4215-4218.
(27) Gleiter, R.; Weigl, H.; Haberhauer, G. Eur. J. Org. Chem. 1998, 1447-
1453.
(28) Lewis, F. D.; Reddy, G. D.; Schneider, S.; Gahr, M. J. Am. Chem. Soc.
1991, 113, 3498-3506.
Cinnoline Production. Unlike the formation of the isoinda-
zole, no products from the cyclization of 1-(2-alkynylphenyl)-
9
J. AM. CHEM. SOC. VOL. 124, NO. 45, 2002 13467

A R T I C L E S
Kimball et al.
Figure 5. (a) ACID plot of the transition state of the cyclization of (2-ethynylphenyl)triazene to the cinnolinium zwitterion. The topology of delocalized
electrons corresponds to a closed cycle and thus is pericyclic. The CIV value of 0.045 between the two carbon atoms at which the new bond is formed,
however, is rather weak; thus, the reaction can be viewed as a borderline case between pericyclic and pseudopericyclic. (b) ACID plot of the transition state
of the cyclization of 2-ethynylstyrene to isonaphthalene, the parent all-carbon analogue of the title cinnoline cyclization. Current density vectors that are
plotted onto the ACID isosurface indicate a strong diatropic ring current and thus an aromatic, pericyclic transition state. The rather high CIV value of 0.072
between the two carbon atoms at which the new bond is formed confirms that the transition state is pericyclic.
Scheme 7 a
Scheme 8
Table 1. Yields of 3e, 3e′, and 3e′′ from 1e in the Presence of
Deuterated Benzhydrol
R1
R2
3e (%)
3e
′
(%)
3e′′ (%)
D
H
D
D
D
H
0
0
97
58
59
0
38
35
0
a Reagents and conditions: (a) (i) HCl, NaNO2, MeCN, H2O, -5 °C;
(ii) 12, K2CO3; (b) Me3SiCtCH, PdCl2(PPh3)2, CuI, Et3N; (c) K2CO3,
MeOH, THF; (d) ODCB, 200 °C.
tions were performed in the presence of benzhydrol species that
were selectively deuterated at either the alcohol or the methylene
position, the cinnoline species showed deuterium incorporation
only when the alcohol was deuterated.
conditions, followed by cleavage of the trimethylsilyl group with
K2CO3 in MeOH/THF, gave 11 in 96% yield. Heating 11 in
ODCB to 200 °C gave, after workup, cinnoline (3, R ) H) in
62% yield along with a complex mixture of byproducts. No
products corresponding to the structures of 14 or 15 were
isolated however.
Results obtained using benzhydrol-d2 as the deuterium source
suggested a different mechanism. Heating ethynylphenyltriazene
1e (R ) Br) in the presence of an excess of benzhydrol-d2
(Scheme 8) generated the cinnoline compound that had incor-
porated deuterium predominantly at C4 (3e′, Table 1). Although
benzophenone was generated in identical quantities to the
resulting cinnoline, a radical mechanism was unlikely since no
incorporation of benzhydrol was observed. Also, when cycliza-
These results showed that C4 in the intermediate in cinnoline
formation had a propensity for proton/deuterium abstraction
from the strong but rather acidic O-H bond. Natural bond order
(NBO) calculations corroborated this type of reactivity. As the
ethynylphenyltriazene cyclizes to the cinnoline intermediate,
NBO analysis shows a partial negative charge developing at
C4 (Figure 6). The calculated singlet/triplet gap, which can be
interpreted as a measure of the diradical character, is 13.6 kcal
mol-1. This indicated that the intermediate in the formation of
the cinnoline compound was likely a zwitterionic cinnolinium
with rather low diradical character. Such a species could abstract
9
13468 J. AM. CHEM. SOC. VOL. 124, NO. 45, 2002

Cyclization of 1-(2-Ethynylphenyl)-3,3-dialkyltriazenes
A R T I C L E S
Scheme 10 a
Figure 6. Bond lengths (Å, top) and NBO charges (bottom) of the reactant
(left), transition state (middle), and product (right) of the pericyclic cinnoline
cyclization reaction.
Scheme 9
a hydrogen/deuterium from an -OH source after cyclization
produced an anion at C4. Further electrocyclic rearrangement
through a six-membered transition state would be consistent with
a pericyclic pathway. This reactivity would give the zwitterion
directly rather than a Hopf-type allene29 that was not found to
be a minimum on the energy hypersurface; however, the cyclic
allene structure is the predominant valence bond configuration
in the all-carbon analogue (Scheme 6). Shevlin and co-workers30
have recently reported similar zwitterionic systems produced
at low temperatures in inert matrixes.
a Reagents and conditions: (a) BTEA‚ICl2, CaCO3, MeOH, CHCl3; (b)
(i) HCl, NaNO2, MeCN, H2O, -5 °C; (ii) piperidine or 2,2,6,6-tetrameth-
ylpiperidine, K2CO3; (c) Me3SiCtCH, PdCl2(PPh3)2, CuI, Et3N, 50 °C;
(d) K2CO3, MeOH, THF; (e) ODCB, 200 °C.
Scheme 11 a
Calculations for the cyclization of ethynylphenyltriazenes to
cinnolines using a pericyclic pathway are consistent with our
observed results. The subsequent mechanism of Et2N- loss,
however, remained in question. NMR scale experiments sug-
gested that Et2NH was generated during the reaction, and in
one example this nucleophile reacted with the cinnoline species
during the cyclization of a fluoroarene (Scheme 9). Presumably,
6-fluorocinnoline (3f) reacted in a nucleophilic aromatic sub-
stitution to generate the 6-diethylaminocinnoline (3g).
Of several possibilities, loss of the diethylamine moiety as
either the nitrogen cation or as an imine generated by proton
abstraction was the most reasonable. One solution to this
question was to install methyl groups on the methylene carbons
attached to the terminal nitrogen (N3) of the triazene. By doing
so, cyclization to the cinnoline should be blocked if the
mechanism occurs through an imine species, while the genera-
tion of a nitrenium ion should be enhanced by sterics and
inductive stabilization. Toward this end, 4-chloroaniline was
iodinated using BTEA‚ICl2, diazotized using HCl and NaNO2,
and reacted with 2,2,6,6-tetramethylpiperidine to give triazene
16a (Scheme 10). Treatment of 16a with trimethylsilylacetylene
under Pd-mediated conditions, followed by desilylation with
K2CO3 in MeOH/THF, provided ethynylphenyltriazene 17a in
25% overall yield. Cyclization at 200 °C in ODCB provided
only products resulting from isoindazole formation, isoindazole
aldehyde 18 and dimer 19 in 98% combined yield. No cinnoline
formation was observed. As a control, ethynylphenyltriazene
a Reagents and conditions: (a) BTEA‚ICl2, CaCO3, MeOH, CHCl3; (b)
i. HCl, NaNO2, MeCN, H2O, -5 °C; ii. (CD3)2NH‚HCl, K2CO3; (c)
Me3SiCtCH, PdCl2(PPh3)2, CuI, Et3N, 50 °C; (d) K2CO3, MeOH, THF;
(e) ODCB, 200 °C.
17b was synthesized for comparison with tetramethylated
analogue 17a. Compound 17b was prepared similarly in 93%
overall yield from 4-chloroaniline. Unlike 17a, when 17b was
heated to 200 °C in ODCB, cinnoline 3d (R ) Cl) was observed
in 96% yield. No isoindazole products were observed when the
reaction was monitored by NMR.
These results encouraged us to confirm the generation of an
imine species as the mechanism of diethylamine elimination.
This could be done using a triazene with deuterated methylene
groups on the 3-nitrogen. If a zwitterionic cinnolinium is gene-
rated through a pericyclic reaction, then this species could re-
move a proton from the triazene methylene carbon to generate
the imine. Using a deuterated dimethyltriazene should then give
a deuterated cinnoline as the product of thermal cyclization.
Treating the diazonium salt of 4-chloro-2-iodoaniline with com-
mercially available Me2NH-d6 gave the triazene 20 with >98%
deuteration of the methyl groups (Scheme 11). Reaction of 20
with trimethylsilylacetylene in the usual manner, followed by
desilylation, gave ethynylphenyltriazene 21 in 78% overall yield.
When 21 was heated to 200 °C in ODCB, cinnoline 3d′ (R )
Cl) was formed in 86% yield, with deuteration predominantly
at the 4-position with some deuteration at the 3-position (Figure
(29) Zimmermann, G. Eur. J. Org. Chem. 2001, 457-471.
(30) (a) Pan, W.; Shevlin, P. B. J. Am. Chem. Soc. 1997, 119, 5091-5094. (b)
Emanuel, C. J.; Shevlin, P. B. J. Am. Chem. Soc. 1994, 116, 5991-5992.
9
J. AM. CHEM. SOC. VOL. 124, NO. 45, 2002 13469

A R T I C L E S
Kimball et al.
Computational Methods
All theoretical calculations have been performed using the Gaussian
98 suite of programs31 at the B3LYP/6-31G* level32 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. Natural charges were
calculated using Weinberger’s NBO method33 implemented in Gaussian.
ACID scalar fields were computed using our own program.25 In all
ACID plots the standard isosurface value of 0.05 was used. Current
density vectors were calculated using the CSGT method of Keith and
Bader.34 To save computational costs and to reduce the complexity of
the energy hypersurface, in all calculations of the cyclization reactions
the two ethyl groups at the terminal NEt2 are replaced by hydrogen
(NH2). The stationary points of the intramolecular proton-transfer
reactions (Scheme 12) were checked at the higher B3LYP/6-31+G**//
B3LYP/6-31G* level of theory. The calculated activation barriers (53.6
and 25.5 kcal mol-1) (52.5 and 23.1 kcal mol-1 at B3LYP/6-31G*)
reveal a weak influence of diffuse and polarization functions, the anionic
character of the species notwithstanding.
Experimental Section
1
General. H and 13C NMR spectra were recorded using a Varian
Inova 300 NMR (1H: 299.94 MHz; 13C: 75.43 MHz) spectrometer.
Chemical shifts (δ) are expressed in ppm downfield from tetrameth-
ylsilane using the residual solvent as internal standard (CDCl3 1H: 7.26
ppm; 13C: 77.0 ppm). Coupling constants are expressed in hertz. IR
spectra were recorded using a Nicolet Magna-FTIR 550 spectrometer.
Melting points were determined on a Meltemp II apparatus and are
uncorrected. Mass spectra were recorded using either an Agilent 1100
LC/MSD (ESI) or Kratos MS50 (HRMS) spectrometer. Et3N and
CH2Cl2 were distilled from CaH2 under an N2 atmosphere prior to
use. THF was distilled from Na and benzophenone under an N2
atmosphere prior to use. All other chemicals were of reagent quality
and used as obtained from the manufacturers. Column chromatography
was performed on Whatman reagent grade silica gel (230-400 mesh).
Sorbent Technologies precoated silica gel plates were used for
preparative (200 × 200 × 1 mm) thin-layer chromatography. Reactions
were carried out in an inert atmosphere (dry nitrogen or argon) when
necessary.
Trimer 6. To a solution of 1a6c (R ) CN, 350 mg, 1.5 mmol) in
ODCB (20 mL) was added CuI (29 mg, 0.15 mmol). The mixture was
brought to 110 °C and stirred for 16 h. After being cooled, the mixture
was diluted with hexanes and vacuum filtered through silica, first
washing with hexanes and then eluting with 1:1 CH2Cl2/EtOAc.
Purification by column chromatography (3:1:1 hexanes/CH2Cl2/EtOAc)
gave 2a6c (R ) CN, 215 mg, 62%) as the first component. Isolation of
the second band furnished 6 (84 mg, 28%) as a bright red crystalline
solid. Recrystallization from EtOH provided red needles suitable for
X-ray diffraction. mp 270.6-271.9 °C. 1H NMR (CDCl3) δ 9.23-
9.22 (m, 1H), 8.21 (d, J ) 1.8 Hz, 1H), 7.95 (s, 1H), 7.93 (dd, J )
8.5, 1.8 Hz, 1H), 7.86 (d, J ) 1.5 Hz, 1H), 7.81 (d, J ) 7.9 Hz, 1H),
7.80 (d, J ) 8.8 Hz, 1H), 7.54 (dd, J ) 8.8, 1.5 Hz, 1H), 7.20 (d, J )
0.6 Hz, 1H), 7.17 (dd, J ) 9.1, 1.8 Hz, 1H), 6.78 (d, J ) 9.1 Hz, 1H),
2
Figure 7. 1H (a) and H (b) NMR spectra of 3d′/3d′′ showing deuterium
incorporation predominantly at C4 with minor deuteration at C3.
1
7). H NMR data indicated a ratio of deuteration of C4 versus
C3 of ca. 9:1. The reaction produced only trace quantities (by
NMR) of material where there was no deuterium incorporation.
Although this strongly suggested that loss of the diethylamino
group occurred through an imine, deuterium incorporation at
C3 was unexpected. Shevlin and Pan reported a similar
rearrangement to give a more stable 1,2-zwitterionic pyridinium
species prepared at low temperature.30a This might allow
hydrogen or deuterium abstraction through an intramolecular,
five-membered transition state. Cyclizations done under higher
dilutions resulted in deuteration at C4 and C3 in ratios similar
to those previously obtained (ca. 9:1). Although this might
support an intramolecular pathway, proton abstraction from the
carbon R to the nitrogen by the anion at C4 has a DFT calculated
barrier of 52.5 kcal mol-1. We therefore suggest that loss of
the dialkylamino group as the imine and rearrangement to the
1,2-zwitterion (22) occur through intermolecular pathways, both
of which originate from 1,3-zwitterion 23 (Scheme 12). In the
latter case, rearrangement is less favored than imine loss.
Conclusions
Our results strongly support a carbene intermediate in the
cyclization of ethynylphenyltriazenes to give isoindazoles
through a pseudocoarctate rearrangement. Further trapping
studies show that a pericyclic pathway is favored for the thermal
cyclization of ethynylphenyltriazenes to cinnolines under neutral
conditions. Proton abstraction by a partially charged zwitterionic
cinnolinium species occurs through an intermolecular process
that eliminates the dialkylamino moiety of the cinnolinium as
an imine. Efforts are underway to determine the generality of
this transformation with regard to other triazene and ethyne
analogues as well as with heteroarene derivatives.
(31) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; 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.;
Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.6; Gaussian,
Inc.: Pittsburgh, PA, 1998.
(32) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(33) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. ReV. 1988, 88, 899-926.
(34) Keith, B. A.; Bader, R. F. W. J. Chem. Phys. 1993, 99, 3669-3682.
9
13470 J. AM. CHEM. SOC. VOL. 124, NO. 45, 2002

Cyclization of 1-(2-Ethynylphenyl)-3,3-dialkyltriazenes
A R T I C L E S
a
Scheme 12. Possible Mechanisms Leading to Cinnolines 3′ and 3′′
a Relative energies Erel and activation barriers ∆H‡ are calculated at the B3LYP/6-31G* level of DFT theory where R ) Me and R′ ) H in 1. The
energies Erel are relative to the 1,2-zwitterion, and the activation enthalpies ∆H‡ are calculated with respect to the corresponding reactants (1,2- and 1,3-
zwitterion).
3.58-3.48 (m, 4H), 3.34 (br s, 2H), 3.06-2.99 (m, 1H), 2.74-2.66
(m, 1H), 1.10 (t, J ) 7.4 Hz, 3H), 0.89 (br s, 6H), 0.19 (t, J ) 7.1 Hz,
3H). 13C NMR (CDCl3) δ 153.22, 147.72, 146.82, 145.66, 144.79,
135.59, 133.75, 133.59, 132.21, 130.26, 129.46, 128.75, 127.96, 126.44,
122.97, 120.38, 120.10, 119.80, 119.23, 118.83, 118.16, 115.54, 108.63,
107.52, 106.76, 106.36, 94.70, 53.00, 50.29, 42.56, 14.26, 12.23, 9.99.
IR (KBr) 3153, 3116, 3064, 2978, 2223, 1593 cm-1. HRMS: calcd
for C35H31N11, 605.2764; found, 605.2652. Anal. Calcd for C35H31N11:
C, 69.40; H, 5.16; N, 25.44. Found: C, 69.69; H, 5.19; N, 25.11.
Cyclopropane 7. A solution of 1b6c (R ) t-Bu, 54 mg, 0.21 mmol)
and 2,3-dimethyl-2-butene (1.0 mL, 8.4 mmol) in dry CH2Cl2 (20 mL)
was deoxygenated via bubbling with N2 that had passed through Fieser’s
solution. After 30 min, CuCl (84 mg, 0.85 mmol) was added, and the
mixture was stirred for 16 h. After vacuum filtration through silica
and evaporation of solvent, purification by preparative TLC (9:1
hexanes/EtOAc) gave 7 (47 mg, 65%) as a yellow powder. mp 108.9-
Column chromatography (4:1 to 2:1 hexanes/CH2Cl2 gradient) gave 8
(1.81 g, 75%) as a tan oil. H NMR (CDCl3) δ 7.74-7.71 (m, 2H),
1
7.62 (dd, J ) 7.5, 1.2 Hz, 1H), 7.47-7.41 (m, 3H), 7.40-7.27 (m,
4H), 7.06-7.02 (m, 2H), 3.75 (q, J ) 7.4 Hz, 4H), 2.27 (s, 3H), 1.28
(t, J ) 7.4 Hz, 6H). 13C NMR (CDCl3) δ 150.01, 143.37, 140.66,
134.13, 132.94, 129.74, 129.42, 129.39, 128.04, 127.83, 127.80, 127.24,
126.87, 122.31, 117.96, 116.72, 91.34, 48.5 (br), 41.7 (br), 20.67, 14.1
(br), 11.6 (br). IR (neat) 3060, 3022, 2973, 2872, 2207 cm-1. MS (ESI)
m/z (%): 368.2 (100, M+ + H), 295.1 (60, M+ - C4H10N), 267.1 (70,
M+ - C4H10N3), 252.1 (100, M+ - C5H13N3).
Fluorene 9. A solution of 8 (51 mg, 0.14 mmol) in DCE (30 mL)
was deoxygenated via bubbling with N2. After 15 min, CuCl (100 mg,
1.0 mmol) was added, and the mixture was heated to 90 °C for 16 h.
After being cooled, the mixture was vacuum filtered through silica using
CH2Cl2 as eluent. The solvent was evaporated, and purification by
preparative TLC (3:1 hexanes/EtOAc) provided 9 (28 mg, 55%) as a
yellow solid. Recrystallization from EtOH gave crystals suitable for
X-ray diffraction. mp 203.1-204.0 °C. 1H NMR (CDCl3) δ 7.89 (d, J
) 7.8 Hz, 2H), 7.50 (d, J ) 8.7 Hz, 1H), 7.45-7.40 (m, 2H), 7.24-
7.21 (m, 4H), 6.95 (d, J ) 9.0 Hz, 1H), 6.12 (s, 1H), 5.93 (s, 1H),
3.58-3.42 (m, 4H), 2.01 (s, 3H), 1.23 (t, J ) 7.2 Hz, 6H). 13C NMR
(CDCl3) δ 145.75, 145.04, 140.88, 133.20, 129.75, 129.06, 127.59,
127.50, 124.92, 120.06, 118.23, 116.92, 116.84, 52.39, 44.58, 21.55,
13.03. IR (KBr) 3062, 3015, 2975, 1447 cm-1. MS (ESI) m/z (%):
368.2 (100, M+ + H), 295.1 (10, M+ - C4H10N), 165.1 (20, C13H9+).
Anal. Calcd for C25H25N3 (367.49): C, 81.71; H, 6.86; N, 11.43.
Found: C, 81.46; H, 6.69; N, 11.51.
1
110.2 °C. H NMR (CDCl3) δ 7.58 (d, J ) 9.0 Hz, 1H), 7.48 (d, J )
1.9 Hz, 1H), 7.37 (dd, J ) 9.0, 1.9 Hz, 1H), 3.35 (br s, 2H), 3.18 (br
s, 2H), 1.48 (s, 1H), 1.38 (s, 6H), 1.37 (s, 9H), 1.06 (s, 6H), 1.04 (t, J
) 7.2 Hz, 6H). 13C NMR (CDCl3) δ 144.75, 142.37, 133.58, 126.83,
125.24, 119.24, 116.73, 116.02, 51.51, 34.64, 31.27, 30.21, 24.28, 23.58,
19.88, 12.56. IR (KBr) 3085, 3040, 2964, 2867, 2735, 1464 cm-1. MS
(ESI) m/z (%): 342.3 (100, M+ + H), 271.2 (10, M+ - C4H10N). Anal.
Calcd for C22H35N3 (341.53): C, 77.37; H, 10.33; N, 12.30. Found:
C, 77.19; H, 10.23; N, 11.99.
1-[2-([1,1′-Biphenyl]-2-ylethynyl)-4-methylphenyl]-3,3-diethyl-1-
triazene (8). Triazene 1c (R ) Me, 1.41 g, 6.5 mmol), 2-iodobiphenyl14
(1.81 g, 6.5 mmol), PdCl2(PPh3)2 (182 mg, 0.26 mmol), and CuI (74
mg, 0.39 mmol) were combined in Et3N (40 mL). The mixture was
degassed by three successive freeze-pump-thaw cycles. The reaction
was heated to 50 °C and stirred under N2 overnight. After being cooled,
the solvent was evaporated, and the crude product was redissolved (9:1
hexanes/CH2Cl2) and vacuum filtered through a pad of silica gel.
Tosylhydrazone 10. Isoindazole 2d6c (R ) Cl, 192 mg, 0.81 mmol)
was combined with p-toluenesulfonylhydrazide (166 mg, 0.89 mmol)
and EtOH (4 mL) and refluxed for 16 h. The solvent was evaporated,
after which column chromatography (3:1 EtOAc/hexanes) gave 10 (255
1
mg, 75%) as a clear oil. H NMR (CDCl3) δ 8.62 (br s, 1H), 8.30 (s,
1H), 7.97 (d, J ) 8.4 Hz, 2H), 7.92 (d, J ) 2.1 Hz, 1H), 7.57 (d, J )
9
J. AM. CHEM. SOC. VOL. 124, NO. 45, 2002 13471

A R T I C L E S
Kimball et al.
9.1 Hz, 1H), 7.39 (d, J ) 8.4 Hz, 2H), 7.27 (dd, J ) 9.1, 2.1 Hz, 1H),
3.35 (br s, 2H), 3.10 (br s, 2H), 2.43 (s, 3H), 0.73 (t, J ) 7.2 Hz, 6H).
13C NMR (CDCl3) δ 144.78, 144.50, 138.17, 134.68, 129.87, 129.43,
128.24, 121.00, 119.05, 118.16, 52.37, 21.63, 11.87. IR (neat) 3197,
3069, 2925, 1167 cm-1. MS (ESI) m/z (%): 420.1 (100, M+ + H),
319.2 (30, M+ - C4H10N3). Anal. Calcd for C19H22ClN5O2S (419.93):
C, 54.34; H, 5.28; N, 16.68. Found: C, 54.62; H, 5.39; N, 16.57.
6-Chlorocinnoline (3d) from Tosylhydrazone 10. The following
procedure for the formation of the sodium salt of 10 was modified
from the literature.15 To a mixture of NaH (4 mg, 0.17 mmol) in THF
(3 mL) was added via syringe a solution of 10 (64 mg, 0.15 mmol) in
THF (3 mL). After being stirred for 15 min, the solvent was evaporated.
The solid obtained was suspended in ODCB (6 mL) and heated
overnight at 200 °C. After being cooled, the solvent was removed, and
the crude product was purified by preparative TLC (2:1:1 hexanes/
CH2Cl2/EtOAc) to give 3d (R ) Cl, 13 mg, 51%) as a white solid.
Spectral data were identical to those reported previously.6c
d2 (10 equiv) in ODCB (6 mL) and heated to 200 °C for 24 h. The
mixture was cooled and diluted with hexanes. The mixture was vacuum
filtered through silica, washing with 1:1 hexanes/CH2Cl2 to remove
less polar compounds and then eluting the cinnoline products with
1:1 CH2Cl2/EtOAc. Purification by preparative TLC (3:1:1 hexanes/
CH2Cl2/EtOAc) gave 3e/e′/e′′ as given in Table 1. Ratios were
determined by NMR spectroscopy.
6-Diethylaminocinnoline (3g). To a sealable glass pressure tube
was added 1f (R ) F, 106 mg, 0.48 mmol) and ODCB (10 mL). The
tube was sealed and heated to 170 °C with stirring overnight. After
being cooled, the solvent was evaporated, and the crude product was
purified by preparative TLC (2:1:1 hexanes/CH2Cl2/EtOAc) to provide
6-diethylaminocinnoline (3g) (32 mg, 33%) as a yellow oil, in addition
to previously reported cinnoline 3f6c (25 mg, 35%) and isoindazole
2f6c (28 mg, 25%). 3g: 1H NMR (CDCl3) δ 8.90 (d, J ) 5.9 Hz, 1H),
8.24 (d, J ) 9.7 Hz, 1H), 7.48 (d, J ) 6.1 Hz, 1H), 7.37 (dd, J ) 9.7,
2.9 Hz, 1H), 6.52 (d, J ) 2.6 Hz, 1H), 3.51 (q, J ) 7.1 Hz, 4H), 1.26
(t, J ) 7.1 Hz, 6H). 13C NMR (CDCl3) δ 148.57, 146.40, 144.81,
131.10, 129.07, 120.16, 119.76, 99.34, 44.74, 12.56. IR (neat) 3065,
2973, 2932, 2900, 2872, 1683, 1611 cm-1. HRMS calcd for C12H16N3:
202.1344. Found: 202.1342.
Triazene 13. 2-Iodoaniline (219 mg, 1.0 mmol) was dissolved in a
minimal amount of MeCN (3 mL), after which concentrated HCl (0.67
mL, 8.0 mmol) and ice (∼1 g) were added. The suspension was cooled
to -5 °C, and a solution of NaNO2 (173 mg, 2.5 mmol) in water (3
mL) and MeCN (1 mL) was added slowly such that the temperature
remained between -5 and -2 °C. Once the addition was complete,
the solution was stirred at -5 °C for 30 min, after which it was
transferred slowly via cannula to a quench solution of amine 1228 (1.4
g, 10 mmol), K2CO3 (661 mg, 5 mmol), water (30 mL), and MeCN
(10 mL) cooled to 0 °C. Once the transfer was complete, the mixture
was allowed to gradually warm to room temperature overnight. The
mixture was diluted with water and extracted with Et2O. The com-
bined organics were dried (MgSO4), filtered, and concentrated. Puri-
fication by column chromatography (1:1 hexanes/CH2Cl2) provided 13
Triazene 16a. 4-Chloroaniline (750 mg, 5.9 mmol) was combined
with BTEA‚ICl2 (2.5 g, 6.4 mmol) and CaCO3 (650 mg, 6.5 mmol) in
MeOH (5 mL) and CHCl3 (40 mL) and stirred at room temperature
for 24 h. The resulting mixture was filtered, and the solvent was
evaporated. The crude product was redissolved in Et2O and washed
with 5% NaHSO3 solution and then water. The combined organics were
dried (MgSO4), filtered, and concentrated. The resultant solid was used
without additional purification.
The iodinated product (500 mg, 2.0 mmol) was dissolved in MeCN
(5 mL) and reacted with concentrated HCl (1.3 mL, 15.8 mmol), NaNO2
(301 mg, 4.4 mmol), 2,2,6,6-tetramethylpiperidine (3.3 mL, 19.8 mmol),
and K2CO3 (1.32 g, 10 mmol) under conditions analogous to triazene
13. Purification by column chromatography (50:1 hexanes/CH2Cl2)
1
(343 mg, 91%) as a light yellow oil. H NMR (CDCl3) δ 7.95 (dd, J
) 7.8, 1.2 Hz, 1H), 7.53 (d, J ) 7.8 Hz, 1H), 7.48-7.33 (m, 6H), 6.93
(td, J ) 7.8, 1.5 Hz, 1H), 6.67 (d, J ) 15.6 Hz, 1H), 6.40-6.26 (m,
1H), 4.59 (dd, J ) 6.6, 1.5 Hz, 2H), 3.33 (br s, 3H). 13C NMR (CDCl3)
δ 149.81, 138.98, 136.10, 133.58, 128.59, 128.51, 127.83, 126.84,
126.39, 124.25, 117.59, 96.67, 57.96, 34.97. IR (neat) 3058, 3026, 2919,
1449, 1345 cm-1. HRMS calcd for C16H17N3I: 378.0467. Found:
378.0460.
1
provided 16a (260 mg, 32%) as a light yellow oil. H NMR (CDCl3)
δ 7.82 (d, J ) 2.1 Hz, 1H), 7.24 (dd, J ) 8.8, 2.1 Hz, 1H), 7.18 (d, J
) 8.8 Hz, 1H), 1.75-1.70 (m, 6H), 1.56 (s, 12H). 13C NMR (CDCl3)
δ 150.33, 138.11, 130.43, 128.60, 117.60, 96.14, 61.38, 41.53 (br),
28.88 (br), 16.54. IR (neat) 3067, 3015, 2933, 1423, 563 cm-1. MS
(ESI) m/z (%): 406.0 (100, M+ + H), 368.2 (50, M+ - HCl).
Alkyne 17a. Triazene 16a (260 mg, 0.64 mmol) was reacted with
TMSA (0.11 mL, 0.78 mmol), PdCl2(PPh3)2 (21 mg, 0.03 mmol), CuI
(11 mg, 0.06 mmol), and Et3N (20 mL) under conditions analogous to
8. The crude product was concentrated and treated with K2CO3 (790
mg, 6 mmol) under conditions analogous to 11. Purification by column
chromatography (9:1 hexanes/CH2Cl2) provided 17a (150 mg, 77%)
Alkyne 11. Triazene 13 (325 mg, 0.86 mmol) was reacted with
trimethylsilylacetylene (TMSA, 0.18 mL, 1.3 mmol), PdCl2(PPh3)2 (18
mg, 0.026 mmol), CuI (10 mg, 0.052 mmol), and Et3N (20 mL) under
conditions analogous to 8. After concentration, the crude product was
redissolved in MeOH (5 mL) and THF (25 mL), and K2CO3 (570 mg,
4.3 mmol) was added. After being stirred at room temperature for 24
h, the reaction was diluted with Et2O and washed with concentrated
NH4Cl solution. The organic layer was dried (MgSO4), filtered, and
concentrated. Purification by column chromatography (1:1 hexanes/
1
as a white semisolid. H NMR (CDCl3) δ 7.45 (d, J ) 2.3 Hz, 1H),
7.29 (d, J ) 8.5 Hz, 1H), 7.21 (dd, J ) 8.5, 2.3 Hz, 1H), 3.26 (s, 1H),
1.71 (br s, 6H), 1.54 (s, 12H). 13C NMR (CDCl3) δ 152.56, 132.68,
129.43, 129.38, 118.79, 117.19, 81.63, 81.50, 61.15, 41.46 (br), 28.47
(br), 16.65. IR (neat) 3304, 3019, 2939, 2108 cm-1. MS (ESI) m/z
(%): 304.1 (100, M+ + H), 142.1 (50, C9H19N).
1
CH2Cl2) gave 11 (233 mg, 96%) as a yellow oil. H NMR (CDCl3) δ
7.53 (dd, J ) 7.6, 1.5 Hz, 1H), 7.48 (d, J ) 8.2 Hz, 1H), 7.42-7.24
(m, 6H), 7.11 (td, J ) 7.3, 1.2 Hz, 1H), 6.63 (d, J ) 15.8 Hz, 1H),
6.35-6.26 (m, 1H), 4.56 (dd, J ) 6.7, 1.5 Hz, 2H), 3.29 (s, 4H). 13C
NMR (CDCl3) δ 152.47, 136.34, 133.67, 129.38, 128.62, 127.91,
126.48, 125.03, 124.49, 117.22, 117.11, 81.88, 81.04, 58.1 (br), 34.1
(br). IR (neat) 3293, 3060, 3027, 2910, 2582, 2104 cm-1. HRMS calcd
for C18H18N3: 276.1497. Found: 276.1497.
Isoindazole Aldehyde 18 and Dimer 19. To a sealable glass
pressure tube was added triazene 17a (40 mg, 0.13 mmol) and ODCB
(4 mL). The tube was sealed and heated overnight at 200 °C. After
being cooled, the solvent was evaporated, and the crude product was
purified by preparative TLC (6:1:1 hexanes/CH2Cl2/EtOAc) to provide
18 (10 mg, 23%) as a yellow oil and dimer 19 (30 mg, 75%) as a
Thermal Cyclization of 11. Alkyne 11 (70 mg, 0.25 mmol) was
dissolved in ODCB (7 mL) and heated to 200 °C for 24 h. The mixture
was cooled, and the solvent was evaporated. Successive purification
by column chromatography (40-60% CH2Cl2 in hexanes) and prepara-
tive TLC (4:1:1 hexanes/CH2Cl2/EtOAc) gave cinnoline 3 (20 mg, 62%)
1
yellow powder. 18: H NMR (CDCl3) δ 10.34 (s, 1H), 8.23 (d, J )
2.2 Hz, 1H), 7.77 (d, J ) 9.1 Hz, 1H), 7.35 (dd, J ) 9.1, 2.2 Hz, 1H),
1.76-1.72 (m, 6H), 1.59 (s, 6H), 0.74 (s, 6H). 13C NMR (CDCl3) δ
182.54, 143.35, 132.72, 129.06, 128.64, 120.50, 120.30, 119.98, 60.17,
41.10, 31.30, 23.94, 17.71. IR (neat) 3074, 2930, 1728, 1664, 1453,
1
as a light oil along with several other byproducts. H and 13C NMR
analyses and MS spectra of the separated byproducts gave no indication
of structures corresponding to 14 or 15.
Cyclization of 1e in the Presence of Benzhydrol-d1/d2. Alkyne 1e
(50 mg, 0.18 mmol) was combined with benzhydrol-d1 or benzhydrol-
790 cm-1. MS (ESI) m/z (%): 320.1 (100, M+ + H), 282.3 (20, M+
-
HCl), 149.0 (100, M+ - C9H18N3), 142.0 (20, C9H18N). 19: mp 323.5-
325.0 °C. 1H NMR (CDCl3) δ 8.05 (d, J ) 1.5 Hz, 2H), 7.80 (s, 2H),
9
13472 J. AM. CHEM. SOC. VOL. 124, NO. 45, 2002

Cyclization of 1-(2-Ethynylphenyl)-3,3-dialkyltriazenes
A R T I C L E S
7.72 (d, J ) 9.1 Hz, 2H), 7.31 (dd, J ) 9.1, 1.5 Hz, 2H), 1.90-1.74
(m, 6H), 1.62 (s, 6H), 0.76 (s, 6H). 13C NMR (CDCl3) δ 143.63, 135.19,
128.24, 127.16, 120.27, 119.52, 118.28, 116.73, 60.00, 41.33, 30.70,
24.12, 17.77. IR (KBr) 3080, 3009, 2931, 1476, 805 cm-1. MS (ESI)
m/z (%): 607.3 (100, M+ + H). Anal. Calcd for C34H44Cl2N6
(607.66): C, 67.20; H, 7.30; N, 13.86. Found: C, 66.95; H, 7.24; N,
13.81.
by column chromatography (20:1 hexanes/CH2Cl2) provided 20 (520
1
mg, 82%) as a yellow oil. H NMR (CDCl3) δ 7.82 (d, J ) 2.1 Hz,
1H), 7.31 (d, J ) 8.8 Hz, 1H), 7.24 (dd, J ) 8.8, 2.1 Hz, 1H). 13C
NMR (CDCl3) δ 148.77, 138.11, 130.77, 128.76, 117.80, 96.26. IR
(neat) 3077, 2961, 2685, 2215, 2065, 1552, 1417 cm-1. MS (ESI) m/z
(%): 316.0 (100, M+ + H).
1-(4-Chloro-2-ethynylphenyl)-3,3-dimethyl-d6-triazene (21). Tri-
azene 20 (520 mg, 1.7 mmol) was reacted with TMSA (0.33 mL, 4.8
mmol), PdCl2(PPh3)2 (47 mg, 0.07 mmol), CuI (22 mg, 0.11 mmol),
and Et3N (25 mL) under conditions analogous to 8. The solvent was
evaporated, and the crude product was treated with K2CO3 (1.32 g, 10
mmol) under conditions analogous to 11. Purification by column
chromatography (8:1 hexanes/CH2Cl2) provided 21 (335 mg, 95%) as
a yellow oil. 1H NMR (CDCl3) δ 7.47 (d, J ) 2.3 Hz, 1H), 7.38 (d, J
) 8.8 Hz, 1H), 7.23 (dd, J ) 8.8, 2.3 Hz, 1H), 3.30 (s, 1H). 13C NMR
(CDCl3) δ 151.15, 132.92, 132.88, 129.72, 129.55, 118.24, 82.02, 80.54.
IR (neat) 3298, 3067, 2217, 2105, 2067 cm-1. MS (ESI) m/z (%): 214.1
(100, M+ + H).
Triazene 16b. 4-Chloroaniline (750 mg, 5.9 mmol), BTEA‚ICl2 (2.5
g, 6.4 mmol), and CaCO3 (650 mg, 6.5 mmol) were reacted under
analogous conditions to 16a. After workup, the iodinated product (438
mg, 1.7 mmol) was dissolved in MeCN (8 mL) and reacted with
concentrated HCl (3.5 mL, 42 mmol), NaNO2 (300 mg, 4.3 mmol),
piperidine (4.5 mL, 45 mmol), and K2CO3 (1.32 g, 10 mmol) under
conditions analogous to triazene 13. Purification by column chroma-
tography (20:1 hexanes/CH2Cl2) provided 16b (583 mg, 96%) as a
yellow oil. 1H NMR (CDCl3) δ 7.82 (d, J ) 2.1 Hz, 1H), 7.31 (d, J )
8.6 Hz, 1H), 7.24 (dd, J ) 8.6, 2.1 Hz, 1H), 3.84 (br s, 4H), 1.72 (br
s, 6H). 13C NMR (CDCl3) δ 148.73, 138.11, 130.94, 128.78, 117.73,
96.45, 52.9 (br), 44.4 (br), 26.5 (br), 24.3 (br), 24.24. IR (neat) 3078,
2939, 2855, 1552 cm-1. MS (ESI) m/z (%): 350.0 (100, M+ + H).
Alkyne 17b. Triazene 16b (583 mg, 1.7 mmol) was reacted with
TMSA (0.31 mL, 4.5 mmol), PdCl2(PPh3)2 (47 mg, 0.07 mmol), CuI
(22 mg, 0.11 mmol), and Et3N (25 mL) under conditions analogous to
8. The crude product was concentrated and treated with K2CO3 (1.32
g, 10 mmol) under conditions analogous to 11. Purification by column
chromatography (8:1 hexanes/CH2Cl2) provided 17b (400 mg, 97%)
6-Chlorocinnoline-d1 (3d′) from Alkyne 21. Compound 21 (24 mg,
0.11 mmol) was dissolved in ODCB (2 mL) in a glass pressure tube
and heated to 200 °C for 16 h. The mixture was cooled, diluted with
hexanes, and vacuum filtered through silica. After evaporation, purifica-
tion by preparative TLC (2:1:1 hexanes/CH2Cl2/EtOAc) gave 6-chlo-
rocinnoline-d1 (3d′) (16 mg, 86%) as a light yellow powder. Partial 1H
2
and H NMR spectra are given in Figure 7.
1
as an orange oil. H NMR (CDCl3) δ 7.47 (d, J ) 2.4 Hz, 1H), 7.41
Acknowledgment. We thank the National Science Founda-
tion; the Petroleum Research Fund, administered by the
American Chemical Society; the Fonds der Chemischen Indus-
trie; and the Alexander von Humboldt Foundation for financial
support. We thank Profs. David Birney and Bruce Branchaud
for helpful comments and suggestions.
(d, J ) 8.8 Hz, 1H), 7.23 (dd, J ) 8.8, 2.4 Hz, 1H), 3.84 (br s, 4H),
3.30 (s, 1H), 1.72 (br s, 6H). 13C NMR (CDCl3) δ 151.09, 132.93,
129.87, 129.58, 118.35, 118.00, 82.12, 80.53, 52.5 (br), 25.0 (br), 24.25.
IR (neat) 3298, 3066, 2940, 2856, 2107 cm-1. MS (ESI) m/z (%): 248.1
(100, M+ + H).
6-Chlorocinnoline (3d) from Alkyne 17b. Compound 17b (20 mg,
0.08 mmol) was dissolved in ODCB (2 mL) in a glass pressure tube
and heated to 200 °C for 16 h. The mixture was cooled, diluted with
hexanes, and vacuum filtered through silica. After evaporation, purifica-
tion by preparative TLC (2:1:1 hexanes/CH2Cl2/EtOAc) gave 3d (13
mg, 96%) as a white solid. Spectral data were identical to those reported
previously.6c
1-(4-Chloro-2-iodophenyl)-3,3-dimethyl-d6-triazene (20). 4-Chlo-
roaniline (750 mg, 5.9 mmol), BTEA‚ICl2 (2.5 g, 6.4 mmol), and CaCO3
(650 mg, 6.5 mmol) were reacted under conditions analogous to 16a.
After workup, the iodinated product (500 mg, 2.0 mmol) was dissolved
in MeCN (8 mL) and reacted with concentrated HCl (1.3 mL, 16 mmol),
NaNO2 (300 mg, 4.3 mmol), (CD3)2NH‚HCl (1.75 g, 20 mmol), and
K2CO3 (3.9 g, 29 mmol) under conditions analogous to 13. Purification
Supporting Information Available: Gaussian archive files
for all structures in Figure 3 (cyclization of 1); calculated
energies of stationary points of the ethyne-vinylidene cycliza-
tion pathway (Supplemental Figure 1) and -NEt2 elimination
pathways (Supplemental Figures 2 and 3); X-ray structures of
6 and 9, structure refinement details, tables of atomic coordi-
nates, thermal parameters, bond lengths, bond angles, torsion
angles, and mean planes. This material is available free of charge
via the Internet at http://pubs.acs.org.
JA027809R
9
J. AM. CHEM. SOC. VOL. 124, NO. 45, 2002 13473