Notable effect of an electron-withdrawing group at C3 on the selective formation of alkylidenecyclobutanes in the thermal denitrogenation of 4-spirocyclopropane-1-pyrazolines. Nonstatistical dynamics effects in the denitrogenation reactions

Article abstract of DOI:10.1021/ja068513e

A detailed study of the thermal denitrogenation of 3-carbomethoxy- substituted 4-spirocyclopropane-1-pyrazolines 6 was conducted. Alkylidenecyclobutane derivatives 7 were selectively formed in a stereospecific manner. Unrestricted density functional calculations for a 1-pyrazoline 10a indicated that the concerted cleavage of two C-N bonds is the energetically favored process for the denitrogenation reaction to give the 2-spirocyclopropyl 1,3-diyl, followed by a conrotatory ring-closure process, which was calculated to be the energy minimum pathway, to afford a spiropentane derivative. The calculated energy minimum pathway is largely inconsistent with the experimental results observed for the denitrogenation of 6 and 10a. The contradiction between the experimental and standard computational results was solved by considering nonstatistical dynamics effects in the concerted denitrogenation reactions. Although the energy minimum pathway from the transition states of the concerted denitrogenation of the 3-carboalkoxy-substituted 1-pyrazolines involves generation of the corresponding 1,3-diradicals, many trajectory calculations using the Bohn-Oppenheimer molecular dynamics model from the transition state for the concerted denitrogenation led directly to the formation of alkylidenecyclobutanes at the UB3LYP/6-31G(d) level of theory.

Full text of DOI:10.1021/ja068513e

Published on Web 10/09/2007
Notable Effect of an Electron-Withdrawing Group at C3 on the
Selective Formation of Alkylidenecyclobutanes in the Thermal
Denitrogenation of 4-Spirocyclopropane-1-pyrazolines.
Nonstatistical Dynamics Effects in the Denitrogenation
Reactions
Masashi Hamaguchi,*,† Masahiro Nakaishi,† Toshikazu Nagai,†
Takeshi Nakamura,‡ and Manabu Abe*,†,‡
Contribution from the Department of Applied Chemistry, Graduate School of Engineering,
Osaka UniVersity (HANDAI), Suita, Osaka 565-0871, Japan, and the Department of Chemistry,
Graduate School of Science, Hiroshima UniVersity (HIRODAI), 1-3-1 Kagamiyama,
Higashi-Hiroshima, Hiroshima 739-8526, Japan
Received November 27, 2006; Revised Manuscript Received June 7, 2007; E-mail: hamaro@chem.eng.osaka-u.ac.jp; mabe@hiroshima-u.ac.jp
Abstract: A detailed study of the thermal denitrogenation of 3-carbomethoxy-substituted 4-spirocyclopro-
pane-1-pyrazolines 6 was conducted. Alkylidenecyclobutane derivatives 7 were selectively formed in a
stereospecific manner. Unrestricted density functional calculations for a 1-pyrazoline 10a indicated that
the concerted cleavage of two C-N bonds is the energetically favored process for the denitrogenation
reaction to give the 2-spirocyclopropyl 1,3-diyl, followed by a conrotatory ring-closure process, which was
calculated to be the energy minimum pathway, to afford a spiropentane derivative. The calculated energy
minimum pathway is largely inconsistent with the experimental results observed for the denitrogenation of
6 and 10a. The contradiction between the experimental and standard computational results was solved by
considering nonstatistical dynamics effects in the concerted denitrogenation reactions. Although the energy
minimum pathway from the transition states of the concerted denitrogenation of the 3-carboalkoxy-substituted
1-pyrazolines involves generation of the corresponding 1,3-diradicals, many trajectory calculations using
the Bohn-Oppenheimer molecular dynamics model from the transition state for the concerted denitroge-
nation led directly to the formation of alkylidenecyclobutanes at the UB3LYP/6-31G(d) level of theory.
In 1977, Bergman and co-worker4 reported on the selective
Introduction
formation of spiropentane (2) (2/3 ) 99/1) in the thermolysis
of 4-spirocyclopropane-1-pyrazoline (1) (Scheme 1). In contrast,
the exclusive formation of methylenecyclobutane derivatives 5
The denitrogenation of azo compounds (RNdNR′) is an
important reaction for the clean generation of radical species,
which has been utilized for synthetically useful radical-chain
was reported by Tokuda and co-workers in the thermal deni-
reactions as well as the preparation of strained molecules.1 The
trogenation of 3-carboalkoxy-substituted 4-spirocyclopropane-
focus of recent research interest in denitrogenation reactions is
1-pyrazolines 4.5 The 2-spirocyclopropane-1,3-diyl derivatives
on the mechanism of the C-N bond cleavage step, that is,
DR have been proposed as plausible intermediates. The dramatic
stepwise C-N bond cleavage versus a concerted two C-N bond
cleavage.2,3 The studies clarify that the product distributions,
including stereoselectivity, are largely dependent on the pre-
ferred mechanism.
substituent effects of an electron-withdrawing group at C3 on
product selectivity prompted us to examine the role of the
carboalkoxy group in the thermal decomposition of the 4-spi-
† Osaka University.
(3) For photochemical denitrogenation reactions, see: (a) Chang, M. H.;
Dougherty, D. A. J. Am. Chem. Soc. 1982, 104, 2333-2334. (b) Adam,
W.; Fragale, G.; Klapstein, D.; Nau, W. M.; Wirz, J. J. Am. Chem. Soc.
1995, 117, 12578-12592. (c) Yamamoto, N.; Olivucci, M.; Celani, P.;
Bernardi, F.; Robb, M. A. J. Am. Chem. Soc. 1998, 120, 2391-2407. (d)
Diau, E. W.-G.; Abou-Zied, O. K.; Scala, A. A.; Zeweil, A. H. J. Am.
Chem. Soc. 1998, 120, 3245-3246. (e) Abe, M.; Adam, W.; Ino, Y.;
Nojima, M. J. Am. Chem. Soc. 2000, 122, 6508-6509. (f) Cattaneo, P.;
Persico, M. J. Am. Chem. Soc. 2001, 123, 7638-7645. (g) Sinicropi, A.;
Page, C. S.; Adam, W.; Olivucci, M. J. Am. Chem. Soc. 2003, 125, 10947-
10959. (h) Adam, W.; Trofimov, A. V. Acc. Chem. Res. 2003, 36, 571-
579. (i) Adam, W.; Diedering, M.; Trofimov, A. J. Phys. Org. Chem. 2004,
17, 643-655. (j) Chen, H.; Li, S. J. Org. Chem. 2006, 71, 9013-9022.
(4) Shen, K. K.-W.; Bergman, R. G. J. Am. Chem. Soc. 1977, 99, 1655-1657.
(5) (a) Danion-Bougot, R.; Danion, D.; Carrie, R. Tetrahedron 1985, 41, 1953-
1958. (b) Chowdury, M. A.; Senboku, H.; Tokuda, M. Tetrahedron Lett.
2003, 44, 3329-3332.
‡ Hiroshima University.
(1) (a) Gassman, P. G.; Mansfield, K. T. Org. Synth. 1969, 49, 1-5. (b) Berson,
J. A. Acc. Chem. Res. 1978, 11, 446-453. (c) Engel, P. S. Chem. ReV.
1980, 80, 99-150. (d) Berson, J. A. Acc. Chem. Res. 1991, 24, 215-222.
(e) Little, D. R. Chem. ReV. 1996, 96, 93-114. (f) Motherwell, W. B.;
Crich, D. In Free Radical Chain Reactions in Orgnic Synthesis; Academic
Press: London, 1992.
(2) For thermal denitrogenation reactions, see; (a) Lyons, B. A.; Pfeifer, J.;
Carpenter, B. K. J. Am. Chem. Soc. 1991, 113, 9006-9007. (b) Lyons, B.
A.; Pfeifer, J.; Peterson, T. H.; Carpenter, B. K. J. Am. Chem. Soc. 1993,
115, 2427-2437. (c) Reyes, M. B.; Carpenter, B. K. J. Am. Chem. Soc.
2000, 122, 10163-10176. (d) Yamabe, S.; Minato, T. J. Phys. Chem. A
2001, 105, 7281-7286. (e) Khuong, K. S.; Houk, K. N. J. Am. Chem.
Soc. 2003, 125, 14867-14883. (f) Breton, G. W.; Shugart, J. H. J. Org.
Chem. 2003, 68, 8643-8649. (g) Abe, M.; Ishihara, C.; Kawanami, S.;
Masuyama, A. J. Am. Chem. Soc. 2005, 127, 10-11.
9
10.1021/ja068513e CCC: $37.00 © 2007 American Chemical Society
J. AM. CHEM. SOC. 2007, 129, 12981-12988
12981

A R T I C L E S
Hamaguchi et al.
Scheme 1. Denitrogenation of 4-Spirocyclopropane-1-pyrazolines
Table 1. Thermal Denitrogenation of 1-Pyrazolines 6 at 80 °C in
Benzene
Scheme 2. Possible Mechanism for the Thermal Denitrogenation
of 4-Spirocyclopropane-1-pyrazoline
product (%)a
entry
6
R1
R2
R3
R4
time /h
7
8b
1
2
3
4
5
a
b
c
d
e
H
Me
H
H
H
H
H
Me
H
H
H
H
H
Me
H
H
H
H
H
16
2
20
3
7a (90)
7b (86)
7c (93)
7d (>95)
7d (50)
7e (50)
8a (10)
8b (14)
8c (7)
Me
24
a Product ratios were determined by 1H NMR, mass balance >95%. b The
configuration of 8 was not assigned because of the low yield of 8.
Scheme 3. Summary for the Computational Results of the
Denitrogenation of 1
rocyclopropane-1-pyrazolines more closely. On the basis of
recent studies of the thermal denitrogenation mechanism of
azoalkanes,2 possible mechanisms for the denitrogenation reac-
tion of 4-spirocyclopropane-1-pyrazolines are summarized in
Scheme 2. The spiropentane can be formed via 2-spirocyclo-
propyl-1,3-diyl DR which may be generated by the concerted6
or stepwise denitrogenation of the 1-pyrazoline. An SH2 process7
involving the diazenyl diradical DZ would be also possible for
the formation of the spiropentane. Three pathways are possible
for the formation of the alkylidenecyclobutane derivative:
(1) ring-opening of the cyclopropane in the 1,3-diradical DR;
(2) ring-opening and concomitant elimination of nitrogen via
the diazenyl diradical DZ; (3) a concerted [2 + 2 + 2] cyclo-
reversion2e reaction from the 1-pyrazoline, in which the ring-
enlargement of the cyclopropane participates in the concerted
denitrogenation.
In the present study, the 1-pyrazoline derivative 6a and
stereochemically labeled 6b-e, which have no extra-ring on
the cyclopropane moiety, were prepared and the product
distributions in the thermal denitrogenation reactions were
investigated, to better understand the effects of an electron-
withdrawing group at C3 on the denitrogenation mechanism
(Table 1). The selective formation of alkylidenecyclobutanes 7
was observed (Table 1), which clearly indicates that the electron-
withdrawing group at C3 plays an important role in controlling
product distribution, the spiropentane versus alkylidenecyclobu-
tane. Computational studies provide a reasonable explanation
for the notable effects of substituents on the denitrogenation
mechanism, in which nonstatistical dynamics8 plays an important
role in controlling the product selectivity in the denitrogenation
reactions.
Results and Discussion
Thermolysis of 1-Pyrazolines 6. 1-Pyrazolines 6a-e were
synthesized by the 1,3-dipolar cycloaddition of diazomethane
or diazoethane with the corresponding methylenecyclopropanes9
(Supporting Information). The structures of 6a-e were un-
equivocally confirmed by spectroscopic analyses including NOE
measurements. When pyrazoline 6a was heated at 80 °C in
benzene for 16 h, methylenecyclobutane 7a was formed in 90%
yield, along with a small amount of the spiropentane derivative
8a (10%), mass balance >95% (entry 1 in Table 1). Prolonging
the thermolysis at 80 °C resulted in no isomerization between
7a and 8a, indicating that they are primary products in the
thermal decomposition of 6a.
The stereochemically labeled trans-configured 6b and cis-
configured 6c underwent decomposition with the stereospecific
ring-enlargement of the cyclopropane ring to give the cis-
configured methylenecyclobutanes 7b (86%) and the trans-
configured 7c (93%), respectively (entries 2-3). The configu-
rations of the methylenecyclobutanes were unequivocally
confirmed by 1H NMR NOE measurements (Supporting Infor-
mation). In addition, spiropentanes 8b and 8c were also formed
as minor products. Configurational determination was, however,
difficult because of the low yields of 8. It should be noted that
the decomposition of the cis-isomer 6c required 20 h for
complete decomposition to occur, while the corresponding trans-
(6) Allred, E. L.; Smith, R. L. J. Am. Chem. Soc. 1967, 89, 7133-7134.
(7) (a) Roth, W.; Martin, M. Tetrahedron Lett. 1967, 4695-4698. (b) Simpson,
C. J. S. M.; Wilson, G. J.; Adam, W. J. Am. Chem. Soc. 1991, 113, 4728-
4732.
(8) (a) Sun, L.; Song, K.; Hase, W. L. Science 2002, 296, 875-878. (b)
Carpenter, B. K. In ReactiVe Intermediate Chemistry; Moss, R. A.; Platz,
M. S.; Jones, M., Jr. Eds.; Wiley Intersience: Hoboken, NJ, 2004; Chapter
21, 925-960.
(9) (a) Hamaguchi, M.; Nagai, T. Tetrahedreon Lett. 1992, 33, 1321-1324.
(b) Hamaguchi, M.; Nakaishi, M.; Nagai, T.; Tamura, H. J. Org. Chem.
2003, 68, 9711-9722.
9
12982 J. AM. CHEM. SOC. VOL. 129, NO. 43, 2007

Notable Effect of an Electron-Withdrawing Group
A R T I C L E S
Scheme 4. Summary for the Computational Results of the Denitrogenation of 10a
isomer 6b decomposed completely within 2 h (entries 2-3).
Prolonging the thermolysis did not change the product ratios
of 7 and 8, and no configurational changes were observed in
the products.
judged by the comparison of the calculated results with the
experimental facts,4 for example, product distribution and
activation energy that were determined by Bergman and co-
worker. High-level ab initio calculations with configuration
interaction, that is, CASPT2 method, were also reported for the
reactivity of the diradical DR1 by Borden and co-workers.12
Thus, the DFT predictions for such an open-shell molecule can
be compared with predictions computed by the CASPT2 level
of theory.
Finally, the thermal decomposition of 1-pyrazolines 6d,e
bearing a methyl group on the spirocyclopropane ring were
performed, in an attempt to determine the regioselectivity of
the migrating carbon of the cyclopropane ring in the formation
of methylenecyclobutanes (entries 4-5). The decomposition of
6d gave methylenecyclobutane 7d exclusively (entry 4). The
regioselective formation of 7d clearly indicates that the C1′
carbon of the cyclopropane ring selectively migrates to the C5
position of 6d. In contrast, the decomposition of 6e gave a 1:1
mixture of the methylene-cyclobutanes 7d and 7e under similar
conditions (mass balance >95%, entry 5). The formation of
regioisomers indicates that carbons C1′ and C2′ both migrate
equally to the C5 position. The decomposition of 6e required
24 h for complete decomposition, whereas pyrazoline 6d was
completely denitrogenated within 3 h at 80 °C (entries 4-5).
All of the products were stable under the conditions used for
the thermolysis. The experimental results for the thermal
denitrogenation of 3-methoxycalbonyl-substituted 1-pyrazolines
6 are summarized as follows: (1) The selective formation of
methylenecyclobutanes 7 was observed. (2) The stereospecific
formation of cis-configured 7b and trans-configured 7c was
observed in the denitrogenation of trans-configured 6b and cis-
configured 6c. (3) The regioselective formation of 7d was
observed in the denitrogenation of 6d, while a mixture of 7d
and 7e was produced in the reaction of 6e. (4) The thermal
decomposition of 6b and 6d was faster than that of 6c and 6e,
respectively.
To understand the notable substituent effects on product
selectivity and the rate of the thermal denitrogenation of
1-pyrazolines 6, computational studies on the denitrogenation
of 1 and model pyrazolines 10 were performed with the
Gaussian 0310 suite of programs (Schemes 3, 4; Supporting
Information).
Molecular Orbital Calculations for the Denitrogenation
Reaction of 1. First of all, the denitrogenation reaction of the
parent 1-pyrazoline 1 and the reactivity of the diradical DR1
were calculated by using density functional theory (DFT) at
the (U)B3LYP/6-31G(d)11 level (Scheme 3). The adequacy of
the DFT calculations for such denitrogenation reactions can be
The transition state TSa (S2 ) 0.40) for the concerted
denitrogenation, producing singlet propane-1,3-diyl DR1 (path
a), was found at the UB3LYP/6-31G(d) level of theory. The
energy barrier of the two-bond cleavage was calculated to be
∆E+ZPE ) 40.2 kcal/mol. The calculated value is in very good
agreement with the experimental value4 of Ea ) 39.9 ( 1.5
kcal/mol reported by Bergman for the denitrogenation of
1-pyrazoline 1, to produce spiropentane (2); see, Scheme 1.
Although the diazenyl diradical 9 was found as an equilibrated
structure, the energy was calculated to be 2.1 kcal/mol higher
than the energy of the transition state TSa at the same level of
theory. The computational calculations clearly suggest that the
pathway involving diazenyl diradical 9 can be excluded for the
formation of spiropentane. The favored two C-N bond cleavage
is consistent with our previous findings on the effect of an
electron-donating substituent at C4, for example, cyclopropane,
in the denitrogenation mechanism.2g Thus, the concerted two
C-N bond cleavage is a symmetry-allowed process, since the
phases of the diyl DR1 LUMO (ψS) and the N2 HOMO (π)
match (Scheme 3).2g The energy barrier for the formation of
spiropentane from DR1 was computed to be a very small, that
is, ∆E ) 0.6 kcal/mol, at the same level of theory, which is
consistent with Borden’s CASPT2 value12 of 0.8 kcal/mol. The
calculated barrier is much lower than the ∆E value of 7.0 kcal/
mol for the 1,2-migration, to produce methylenecyclobutane 3.
Thus, the UB3LYP calculations were strongly supportive of the
experimental results for the selective formation of spiropentane
(2) from 1 (Scheme 1).13,14 The potential energy surface of the
three-bond pericyclic denitrogenation, that is, a [2 + 2 + 2]
(11) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652. (b) Lee, C.; Yang,
W.; Parr, R. G. Phys. ReV. B 1988, 37, 785-789. (c) Foresman, J. B.;
Head-Gordon, M.; Pople, J. A.; Frisch, M. J. J. Phys. Chem. 1992, 96,
135-149. (d) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28,
213-222.
(12) Johnson, W. T. G.; Hrovat, D. A.; Borden. W. T. J. Am. Chem. Soc. 1999,
121, 7766-7772.
(13) For the accuracy of DFT calculations for other open-shell molecules, see
ref 2e and: Goldstein, E.; Beno, B.; Houk, K. N. J. Am. Chem. Soc. 1996,
118, 6036-6043.
(10) Frisch, M. J.; et al. Gaussian 03, revision C. 02; Gaussian, Inc.:
Wallingford, CT, 2004.
9
J. AM. CHEM. SOC. VOL. 129, NO. 43, 2007 12983

A R T I C L E S
Hamaguchi et al.
cycloreversion2e (path b) via TSb, was not found even at the
RB3LYP/6-31G(d) level of theory. Although the transition state
(∆E+ZPE ) 41.1 kcal/mol) of the two bond cleavage producing
the diyl DR1 was located at the restricted DFT level of theory,
the wavefunction was calculated to be unstable.
Molecular Orbital Calculations and Experimental Studies
on the Denitrogenation of 10a. To obtain information concern-
ing the role of the carbomethoxy group on the selective
formation of alkylidenecyclobutane derivatives (Table 1), the
denitrogenation of a 1-pyrazoline 10a and the reactivity of the
1,3-diyl DR10a were examined at the UB3LYP/6-31G(d) level
of theory (Scheme 4). As found in pyrazoline 1 (Scheme 3),
the concerted denitrogenation (path a) was calculated to be the
energetically more favored pathway than the stepwise denitro-
genation for producing diazenyl diradical 13 (∆E+ZPE ) 29.3
kcal/mol). The transition state TSa110a, in which the car-
bomethoxy group is located in a pseudoaxial position, was lower
in energy by 2.0 kcal/mol than the transition state TSa210a, in
which the methyl group is located in a pseudoaxial position.
The five-membered pyrazoline ring of the precursor 10a was
optimized to be in an almost planar conformation. Intrinsic
reaction coordinate (IRC) calculations for TSa110a revealed that
the energy minimum pathway for the denitrogenation of 10a
involves the generation of the singlet state of 1,3-diyl DR10a
via TSa110a. The [2 + 2 + 2] cycloreversion via TSb10a (path
b) was not found at either the RB3LYP/6-31G(d) or the RMP2/
6-31G(d) level of theory. As found for the reactivity of the
parent diradical DR1, the energy barrier (∆Eq ) 1.0 kcal/mol)
for the conrotatory ring closure of the diradical DR10a was
found to be lower than that for the ring-enlargement of the
cyclopropane ring in the diradical DR10a (∆Eq ) 6.0 kcal/
mol). Thus, the computational results, again, clearly indicate
that the energy minimum pathway of the denitrogenation of 10a
is the generation of the diradical DR10a by concerted denitro-
genation, followed by ring-closure occurs to selectively give
the spiropentane derivative 11a. The results of the statistical
calculations are apparently inconsistent with the experimental
results, that is, the selective formation of alkylidenecyclobutanes
7 in the denitrogenation of 6 (Table 1). To exclude an effect of
p-nitrophenyl group, which is necessary for the separation of
stereoisomers in 6b-e by recrystallization, on the product
selectivity in the denitrogenation of 6a, 1-pyrazoline 10a was
prepared (see, Supporting Information) and the product analysis
and the determination of the activation parameters (Ea and ln A)
was performed for the denitrogenation reaction in toluene
solution (Figure 1). Thus, one can compare directly the DFT
calculations (Scheme 4) with the experimental results. As shown
in Figure 1, the activation energy (Ea) for the denitrogenation
was determined to be 26.8 ( 0.5 kcal/mol according to the
Figure 1. Arrhenius plot for the denitrogenation of 1-pyrazoline 10a in
toluene. The first-order rate constants (k) were determined by the decay
traces of the band of the azo-chromophore at 327 nm.
Figure 2. A schematic potential energy (PE) surface for the denitrogenation
reaction of 1-pyrazolines.
60 °C to 82 °C. Prolonging the thermolysis resulted in no
isomerization between 11a and 12a, indicating that they are
primary products in the thermal decomposition of 10a. Although
DFT is known to handle properly open-shell structures and their
reactivities,13 the MO calculations did not reproduce the product
selectivity in the denitrogenation of 1-pyrazolines 6 and 10a.
On the basis of the computational results for the pyrazolines
1 and 10a, the potential energy (PE) surface in the denitroge-
nation reaction can be drawn as shown in Figure 2. A large
activation energy is required for the first concerted-denitroge-
nation step to generate the unstable diradical with kinetic energy,
which affords the spiropentane derivative with a small activation
energy. The alkylidenecyclobutane may be formed from the
diradical intermediate with a small activation energy. Therefore,
it would be possible that the trajectory passing through the
denitrogenation transition-state could lead directly to the alky-
lidenecyclobutane via the valley-ridge inflection point (VRI).15
Such a reaction has a high probability of exhibiting nonstatistical
dynamic effects.8,16
Arrhenius plot, which is very closed to the value (∆Eq
)
+ZPE
26.9 kcal/mol) obtained by the DFT calculations (Scheme 4).
However, the statistical calculations did not accurately predict
the product selectivity. Thus, the theory predicts that the
exclusive formation of spiropentane 11a (Scheme 4), although
the slightly favored formation of methylenecyclobutane 12a was
experimentally observed in the denitrogenation of 10a (Figure
1). The product ratio of 11a and 12a were virtually temperature-
independent at least in the reaction-temperature range from
(14) An unreliability of the DFT energies for strained molecules was reported
by Schreiner and coworkers, see: Schreiner, P. R.; Fokin, A. A.; Pascal,
R. A., Jr.; de Meijere, A. Org. Lett. 2006, 8, 3635-3638.
9
12984 J. AM. CHEM. SOC. VOL. 129, NO. 43, 2007

Notable Effect of an Electron-Withdrawing Group
A R T I C L E S
Scheme 5. Bohn-Oppenheimer Molecular Dynamics Simulations
in the Denitrogenation of 1-Pyrazolines 1, 10a, and 10b
1-pyrazoline (1) after 120 fs; 8 in 10 trajectories produced
diradical DR1 after 100 fs. One trajectory gave spiropentane
(2) directly with conrotatory ring-closure from the transition
state TSa (Scheme 6a). Thus, the trajectory calculation results
are in good agreement with Bergman’s experiment, that is, the
selective formation of spiropentane (2) from 1-pyrazoline (1)
(see Scheme 2). In the trajectory calculations from the transition
state TSa110a, which contains a carbomethoxy group, the results
were striking (Scheme 5). Five out of 31 trajectories produced
alkylidenecyclobutane 12a. Twenty-five trajectories went back
to the pyrazoline 10a. One trajectory gave the diradical DR10a,
which is the precursor of the spiropentane 11a. Thus, the results
of the trajectory calculations are in good agreement with the
selective formation of an alkylidenecyclobutane derivative in
the denitrogenation of 3-carboalkoxy-substituted 1-pyrazolines
4,6, and 10a. It should be noted that the C5 carbon selectively
migrates to the C2 carbon from the backside of the departing
nitrogen atom, as shown in Scheme 6b. Thus, a formal [2 + 2
+ 2] cycloreversion reaction is proposed for the selective
formation of alkylidenecyclobutane derivatives in the denitro-
genation of the 3-carbomethoxy-substituted 1-pyrazoline. To
examine the generality of the selective formation of alky-
lidenecyclobutane derivatives, trajectory calculations were
performed at the same level of theory for the transition state
TSa110b. Eleven of 36 trajectories produced the 1,3-diradical
DR10b after 120 fs. One of the trajectories afforded spiropen-
tane derivative 11a after 350 fs (Figure S1 in Supporting
Information). The remaining 25 trajectories afforded the ex-
pected alkylidenecyclobutane 12b. Thus, an electron-withdraw-
ing group at C3 in the 1-pyrazolines plays an crucial role in
the selective production of alkylidenecyclubutanes, although the
energy minimum path affords the corresponding 1,3-diradical.
Nonstatistical Dynamics Effects in the Denitrogenation of
1-Pyrazolines. To study the question of dynamics effects, the
transition-states of the concerted denitrogenation, TSa, TSa110a,
TSa110b, were used as starting points for trajectory calculations
using a Bohn-Oppenheimer molecular dynamics (BOMD)
model17 at the UB3LYP/6-31G(d) level of theory, 0.2-fs steps
(Scheme 5). The direct dynamics trajectories were initiated with
conditions chosen from a 353 K Boltzmann distribution for the
reaction coordinate translation.18 Although the energy minimum
pathway from the denitrogenation transition-state involves
generating the 1,3-diradical, that is, the precursor of the
spiropentane derivative, the chemical dynamics calculations
showed that two products, 1,3-diradical and alkylidenecyclobu-
tane, were formed from the transition state. From the transition
state TSa, 1 out of 10 trajectories afforded the starting
Effects of an Electron-Withdrawing Group at C3 on the
Selective Formation of Alkylidenecyclobutane Derivatives
in the Denitrogenation of 1-Pyrazoline. To obtain information
on the role of the electron-withdrawing substituent at C3 on
changing the product distribution in the denitrogenation reac-
tions, spiropentane versus alkylidenecyclobutane, the singlet-
triplet energy gaps (∆EST in kcal/mol ) ES - ET) were
calculated for the 2-spirocyclopropane-substituted propane-1,3-
diyls DR1 (Z ) H) and DR10b (Z ) CO2Me) at the UB3LYP/
6-31G(d) level of theory, in which the magnitude of the
hyperconjugative interaction19 between the symmetric nonbond-
ing molecular orbital (ψS) of the diyl and the C-C σ orbital of
the cyclopropane ring can be evaluated precisely (Scheme 7).
(15) Metiu, H.; Ross, J.; Silbey, R.; Grorge, T. F. J. Chem. Phys. 1974, 61,
3200-3209. Shaik, S.; Danovich, D.; Sastry, G. N.; Ayala, P. Y.; Schlegel,
H. B. J. Am. Chem. Soc. 1997, 9237-9245. Sastry, G. N.; Shaik, S. J. Am.
Chem. Soc. 1998, 120, 2131-2147.
(16) (a) Carpenter, B. K. J. Am. Chem. Soc. 1995, 117, 6336-6344. (b)
Carpenter, B. K. J. Am. Chem. Soc. 1996, 118, 10329-10330. (c) Levine,
R. D. Pure and Appl. Chem. 1997, 69, 83-90. (d) Doubleday, C., Jr.;
Bolton, K.; Hase, W. L. J. Am. Chem. Soc. 1997, 119, 5251-5252. (e)
Carpenter, B. K. Angew. Chem., Int. Ed. 1998, 37, 3340-3350. (f)
Doubleday, C., Jr.; Bolton, K.; Hase, W. L. J. Phys. Chem. A 1998, 102,
3648-3658. (g) Doubleday, C. J. Phys. Chem. A 2001, 105, 6333-6341.
(h) Sun, L.; Hase, W. L.; Song, K. J. Am. Chem. Soc. 2001, 123, 5753-
5756. (i) Mann, D. J.; Hase, W. L. J. Am. Chem. Soc. 2002, 124, 3208-
3209. (j) Debbert, S. L.; Carpenter, B. K.; Hrovat, D. A.; Borden, W. T. J.
Am. Chem. Soc. 2002, 124, 7896-7897. (k) Ammal, S. C.; Yamataka, H.;
Aida, M.; Dupuis, M. Science 2003, 299, 1555-1557. (l) Singelton, D.
A.; Hang, C.; Szymanski, M. J.; Greenwald, E. E. J. Am. Chem. Soc. 2003,
125, 1176-1177. (m) Bekele, T.; Christian, C. F.; Lipton, M. A.; Singleton,
D. A. J. Am. Chem. Soc. 2005, 127, 9216-9223. (n) Suhrada, C. P.; Selcuki,
C.; Nendel, M.; Cannizzaro, C.; Houk, K. N.; Rissing, P.-J.; Baumann, D.;
Hasselmann, D. Angew. Chem., Int. Ed. 2005, 44, 3548-3552. (o) Northrop,
B. H.; Houk, K. N. J. Org. Chem. 2006, 71, 3-13. (p) Polo, V.; Domingo,
L. R.; Andres, J. J. Org. Chem. 2006, 71, 754-762. (q) Doubleday, C.;
Suhrada, C. P.; Houk, K. N. J. Am. Chem. Soc. 2006, 128, 90-94. (r)
Ussing, B. R.; Hang, C.; Singleton, D. A. J. Am. Chem. Soc. 2006, 128,
7594-7607. (s) Ammal, S. C.; Yamataka, H. Eur. J. Org. Chem. 2006,
4327-4330. (t) Carpenter, B. K. Chem. Soc. ReV. 2006, 35, 736-747. (u)
Vayner, G.; Addepalli, S. V.; Song, K.; Hase, W. L. J. Chem. Phys. 2006,
125, 014317. (u) Yamataka, H.; Aida, M.; Dupuis, M. Chem. Phys. Lett.
1999, 300, 583-587. (v) Yamataka, H.; Aida, M.; Dupuis, M. Chem. Phys.
Lett. 2002, 353, 310-316.
The parent 1,3-diradical DR1 was calculated to be the triplet
ground-state molecule, ∆EST ) +1.5 kcal/mol. The value for
the preference for the triplet state is close to the CASPT2 value12
of +2.0 kcal/mol.20 In contrast, the singlet ground state was
calculated for the carbomethoxy-substituted diradical DR10b,
(19) The hyperconjugation effects on the singlet-triplet energy gap in propane-
1,3-diyls, see: (a) Getty, S. J.; Hrovat, D. A.; Borden, W. T. J. Am. Chem.
Soc. 1994, 116, 1521-1527. (b) Borden, W. T. Chem. Commun. 1998,
1919-1925. (c) Shelton, G. R.; Hrovat, D. A.; Borden, W. T. J. Org. Chem.
2006, 71, 2982-2986. (d) Abe, M.; Adam, W.; Hara, M.; Hattori, M.;
Majima, T.; Nojima, M.; Tachibana, K.; Tojo, S. J. Am. Chem. Soc. 2002,
124, 6540-6541. (e) Abe, M.; Hattori, M.; Takegami, A.; Masuyama, A.;
Hayashi, T.; Seki, S.; Tagawa, S. J. Am. Chem. Soc. 2006, 128, 8008-
8014. (f) Abe, M.; Kubo, E.; Nozaki, K. Angew. Chem., Int. Ed., 2006, 45,
7828-7831.
(17) (a) Helgaker, T.; Uggerud, E.; Jensen, H. J. A. Chem. Phys. Lett. 1990,
173, 145. (b) Uggerud, E.; Helgaker, T. J. Am. Chem. Soc. 1992, 114,
4265. (c) Tuckerman, M. E.; Ungar, P. J.; von Rosenvinge, T.; Klein, M.
L. J. Phys. Chem. 1996, 100, 12878. (d) Millam, J. M.; Bakken, V.; Chen,
W.; Hase, W. L.; Schlegel, H. B. J. Chem. Phys. 1999, 111, 3800. (e) Li,
X.; Millam, J. M.; Schlegel, H. B. J. Chem. Phys. 2000, 113, 10062.
(18) Peslherbe, G. H.; Wang, H.; Hase, W. L. AdV. Chem. Phys. 1999, 105,
171-201.
(20) Jain, R.; McElwee-White, L.; Dougherty, D. A. J. Am. Chem. Soc. 1988,
110, 552-560.
9
J. AM. CHEM. SOC. VOL. 129, NO. 43, 2007 12985

A R T I C L E S
Hamaguchi et al.
Scheme 6. Selected Structures in the Bohn-Oppenheimer Molecular Dynamics Simulations in the Denitogenation of (a) 1-Pyrazolines 1
and (b) 3-Carbomethoxy-substituted 1-Pyrazoline 10a
Scheme 7. Z-group Effect on the Singlet-Triplet Energy Gap in
2-Spirocyclopropane-1,3-diyls
affords the alkylidenecyclubutane derivative. In fact, the energy
barrier for the ring-cleavage in DR10a was calculated to be
lower than that for the diradical DR1 (see Schemes 3 and 4).
Thus, cyclopropane σ bond-breaking would be prone to
participate in the concerted denitrogenation of the 3-car-
bomethoxy-1-pyrazolines.
Stereospecific Formation of Alkylidenecyclobutanes in the
Denitrogenation of 3-Carbomethoxy-Substituted 1-Pyrazo-
lines. As shown in Table 1, the thermal denitrogenation of the
trans-configured 6b and the cis-configured 6c gave stereospe-
cifically the cis-configured 7b and the trans-configured 7c,
respectively (entries 2, 3 in Table 1). The regioselective
formation of 7d was observed in the denitrogenation of 6d, while
the regiorandom migration of the carbon in the cyclopropane
ring was verified by the formation of a 1:1 mixture of 7d and
7e (entries 4, 5 in Table 1). To clarify the origin of the
stereospecific and regioselective formation of the alkylidenecy-
clobutanes 7, the structures of the transition states for the
concerted denitrogenation of the model 1-pyrazolines 10c-f,
which correspond to 6b-e (Table 1), were optimized at the
UB3LYP/6-31G(d) level of theory (Scheme 8).
∆EST ) -2.0 kcal/mol, after spin-correction.21,22 The notable
substituent effect on the ground state spin-multiplicity can be
rationalized by the larger hyperconjugative stabilization of the
singlet state in DR10b, compared to that for DR1. Since the
energy level of the orbital ψS of DR10b (Z ) CO2Me) is lower
than that of DR1 (Z ) H), a stronger hyperconjugative
interaction with the C-C σ orbital of the electron-donating
cyclopropane would be expected for DR10b. In fact, the C2-
C1′ bond length in DR10b (C2-C1′ ) 1.596 Å) was calculated
to be significantly longer than that in DR1 (C2-C1′ ) 1.562
Å). Such a hyperconjugative interaction should also be operative
in the ring-enlargement reaction of the cyclopropane ring, which
For the concerted denitrogenation of the trans-configured 10c,
the transition-state structure TSa110c, which would be the
precursor of the cis-configured alkylidenecyclobutane 12c, was
successfully optimized at the UB3LYP/6-31G(d) level of theory.
However, the transition state TSa210c, in which the two methyl
groups are located in the pseudoaxial positions, could not be
found as a transition state for the denitrogenation at the same
level of theory. The transition state TSa110c was finally obtained
during the optimization of the transition state TSa210c. The
energetically disfavored structure of TSa210c can be reasonably
explained by the severe 1,3-diaxial repulsions between the two
(21) Yamaguchi, K.; Jensen, F.; Dotigo, A.; Houk, K. N. Chem. Phys. Lett.
1988, 149, 537-542.
(22) For other singlet ground state of localized diradicals, see; (a) Niecke, E.;
Fuchs, A.; Baumeister, F.; Martin, Nieger, Schorller, W. W. Angew. Chem.,
Int. Ed. 1995, 34, 555-557. (b) Scheschkewitz, D.; Amii, H.; Gornitzka,
H.; Schoeller, W. W.; Bourissou, D.; Bertrand, G. Science 2002, 295, 1880-
1881. (c) Sugiyama, H.; Ito, S.; Yoshifuji, M. Angew. Chem., Int. Ed. 2003,
32, 3802-3804. (d) Cox, H.; Hitchcock, P. B.; Lappert, M. F.; Pierssens,
L. J.-M. Angew. Chem., Int. Ed. 2004, 43, 4500-4504. (e) Ma, J.; Ding,
Y.; Hattori, K.; Inagaki, S. J. Org. Chem. 2004, 69, 4245-4255. (f) Cui,
C.; Brynda, M.; Olmstead, M. M.; Power, P. P. J. Am. Chem. Soc. 2004,
126, 6510-6511.
9
12986 J. AM. CHEM. SOC. VOL. 129, NO. 43, 2007

Notable Effect of an Electron-Withdrawing Group
A R T I C L E S
Scheme 8. Summary of Computational Studies for the Denitrogenation of Model 1-Pyrazolines 10c-f at the UB3LYP/6-31G(d) Level of
Theory
*Although the location of such a structure TSa210c was attempted, the transition state TSa110c was finally obtained after the optimization.
Summary
methyl groups. The energetic preference of the transition state
TSa110c is in very good agreement with experimental results
for the selective formation of the cis-configured 7b from the
trans-configured 1-pyrazoline 6b (see entry 2 in Table 1). In
the case of the denitrogenation of the cis-configured 10d, the
transition state TSa210d, which would be the precursor of the
trans-configured 12d, was calculated to be energetically more
stable than the transition state TSa110d by 1.5 kcal/mol including
zero-point energies. Thus, the theoretical prediction reproduced
the selective formation of the trans-configured 7c in the
denitrogenation of the cis-configured 6c very well (entry 3 in
Table 1). The energy barrier for the trans- configured 10c was
calculated to be smaller than that of the cis-configured 10d,
compare the relative energy of TSa110c and TSa210d. In fact,
the denitrogenation of the trans-configured 6b was much faster
than that of the cis-configured 6c (entries 2,3 in Table 1).
The theoretical calculations for a model 10e, which possesses
a methyl group at the spirocyclopropane ring, predicted the
selective formation of 12e, since the energy of TSa210e was
calculated to be more stable than the transition state TSa110e
by 3.3 kcal/mol (Scheme 8). The significant substituent effect
is rationalized by the severe steric repulsion of the substituents
between the two methyl groups in TSa110e. In fact, the selective
formation of 7d was observed in the denitrogenation of 6d (entry
4 in Table 1). In the case of the denitrogenation of 10f, the
energy difference between the transition states TSa110f and
TSa210f was calculated to be relatively small, ∆∆E+ZPE ) 0.9
kcal/mol (Scheme 8). As predicted by the computational re-
sults, a mixture of 7d and 7e was obtained in the denitrogenation
of 6e (entry 5 in Table 1). The energy barrier for 10e was
calculated to be smaller than that for 10f; compare the relative
energy of TSa210e and TSa210f (Scheme 8). In fact, the
denitrogenation of 6d was much faster than that of 6e (entries
4,5 in Table 1).
The mechanism of the denitrogenation of 4-spirocycloproane-
1-pyrazolines was investigated in detail in combined experi-
mental and computational studies. For the reaction of the parent
1-pyrazoline 1, the selective formation of spiropentane (2) was
rationalized by the generation of the singlet 1,3-diyl DR1 via
the concerted denitrogenation transition-state TSa, followed by
the fast radical-coupling reaction of the diyl. The mechanism
was strongly supported by standard computational calculations
at the UB3LYP/6-31G(d) level of theory. The stepwise forma-
tion of spiropentanes was also calculated to be the energy
minimum pathway for the denitrogenation of 3-carboalkoxy-
substituted 1-pyrazolines, although the selective formation of
alkylidenecyclobutanes was observed in experiments. The
contradiction between the experimental and standard compu-
tational results was solved by considering nonstatistical dynam-
ics effects in the denitrogenation reactions. In classical trajectory
calculations using the Bohn-Oppenheimer molecular dynamics
model, the product distribution varies greatly. Although the
energy minimum path from the transition states of the concerted
denitrogenation of the 3-carboalkoxy-substituted 1-pyrazolines
involves generating the corresponding 1,3-diradicals, many
trajectories led directly to the alkylidenecyclobutane, in which
the migration of the carbon occurs from the backside of the
departing C-N bond. Thus, the formal [2 + 2 + 2] cyclor-
eversion reaction mechanism adequately explains the selective
formation of alkylidenecyclobutanes in the thermal denitroge-
nation of the carboalkoxy-substituted 1-pyrazolines (Scheme 9).
The stereospecific formation of alkylidenecyclobutanes 6b,c
supports the formal [2 + 2 + 2] cycloreversion reaction. The
notable electron-withdrawing group effects on product distribu-
tion can be rationalized by the large contribution of the
hyperconjugative resonance structure in the singlet state of the
9
J. AM. CHEM. SOC. VOL. 129, NO. 43, 2007 12987

A R T I C L E S
Hamaguchi et al.
Scheme 9. Z-group Effect on the Denitrogenation Mechanism
Acknowledgment. This work was partially supported by the
Ministry of Education, Science, Sports, and Culture, Japan,
Grant-in-Aid for Scientific Research (B), No. 17350019 and
No. 19350021, Priority Area No. 19027033 and No. 1902034,
and Mitsubishi Chemical Corporation Fund. This paper is
dedicated to Professor Herbert Mayr (LMU Muenchen) on the
occasion of his 60th birthday.
Supporting Information Available: Complete ref 10, experi-
mental, and computational details. This material is available free
of charge via the Internet at http://pubs.acs.org.
1,3-diyl (Scheme 7). The importance of dynamics effects should
stimulate future calculations and experiments on the mechanisti-
cally and synthetically fascinating denitrogenation reactions of
azoalkanes.
JA068513E
9
12988 J. AM. CHEM. SOC. VOL. 129, NO. 43, 2007