Rate constants for 1,2-hydrogen migration in cyclohexylidene and in substituted cyclohexylidenes

Article abstract of DOI:10.1021/jo990171q

Laser flash photolysis (UV-LFP, 308 nm) of suitably substituted oxadiazolines leads to cyclohexylidene (14a), 4-tert-butyl-cyclohexylidene (14b), 2-trifluoromethylcyclohexylidene (14d), 8-aza-8-methyl[3.2.1]oct-3- ylidene (14e), diethylcarbene (14f), and ethyl(methyl)carbene (14h). Carbene intermediates were inferred from the products of steady state photolyses, and their pyridinium ylides were inferred from transient absorption spectra observed when pyridine was present. Yields of the pyridinium ylides 15a-h as a function of pyridine concentration gave the lifetimes (τ) for carbenes 14a-h in cyclohexane, cyclohexane-d₁₂, and benzene solutions, at 22 °C. The intermediacy of cyclohexylidene (14a) was inferred from the observation of cyclohexene formed in both the LFP and steady state (SS) experiments. The major products from dual wavelength irradiation of the oxadiazolines (at 254 and 300 nm) were those of 1,2-migration of hydrogen (1,2-H) in the corresponding carbenes. 2-Trifluoromethylcyclohexylidene gave 3- trifluoromethylcyclohexene and 1-trifluoromethylcyclohexene in a 9.8:1 ratio. The kinetic data support the conclusion that 1,2-H in the cyclohexylidenes is accelerated, relative to 1,2-H in dimethylcarbene. A 4-tert-butyl substituent has a negligible effect on the rate constant for 1,2-H, but the CF₃ group at the α-position decelerates 1,2-H by roughly 10-fold, as inferred from the distribution of products.

Full text of DOI:10.1021/jo990171q

4456
J . Org. Chem. 1999, 64, 4456-4464
Ra te Con sta n ts for 1,2-Hyd r ogen Migr a tion in Cycloh exylid en e
a n d in Su bstitu ted Cycloh exylid en es1
J ohn Paul Pezacki,†,‡ Philippe Couture, J ames A. Dunn,† and J ohn Warkentin*
Department of Chemistry, McMaster University, 1080 Main Street West,
Hamilton, Ontario, Canada L8S 4M1
Paul D. Wood and J anusz Lusztyk*
The Steacie Institute for Molecular Sciences, National Research Council of Canada, 100 Sussex Drive,
Ottawa, Ontario, Canada K1A 0R6
Francis Ford and Matthew S. Platz
Newman and Wolfrom Laboratory of Chemistry, The Ohio State University, 120 West 18th Avenue,
Columbus, Ohio 43210
Received J anuary 29, 1999
Laser flash photolysis (UV-LFP, 308 nm) of suitably substituted oxadiazolines leads to cyclohexy-
lidene (14a ), 4-tert-butyl-cyclohexylidene (14b), 2-trifluoromethylcyclohexylidene (14d ), 8-aza-8-
methyl[3.2.1]oct-3-ylidene (14e), diethylcarbene (14f), and ethyl(methyl)carbene (14h ). Carbene
intermediates were inferred from the products of steady state photolyses, and their pyridinium
ylides were inferred from transient absorption spectra observed when pyridine was present. Yields
of the pyridinium ylides 15a -h as a function of pyridine concentration gave the lifetimes (τ) for
carbenes 14a -h in cyclohexane, cyclohexane-d12, and benzene solutions, at 22 °C. The intermediacy
of cyclohexylidene (14a ) was inferred from the observation of cyclohexene formed in both the LFP
and steady state (SS) experiments. The major products from dual wavelength irradiation of the
oxadiazolines (at 254 and 300 nm) were those of 1,2-migration of hydrogen (1,2-H) in the
corresponding carbenes. 2-Trifluoromethylcyclohexylidene gave 3-trifluoromethylcyclohexene and
1-trifluoromethylcyclohexene in a 9.8:1 ratio. The kinetic data support the conclusion that 1,2-H
in the cyclohexylidenes is accelerated, relative to 1,2-H in dimethylcarbene. A 4-tert-butyl substituent
has a negligible effect on the rate constant for 1,2-H, but the CF3 group at the R-position decelerates
1,2-H by roughly 10-fold, as inferred from the distribution of products.
In tr od u ction
(methyl)carbene7 must rearrange very slowly to methyl
vinyl ether by 1,2-H at 25 °C, as that rearrangement
barely competes with attack on precursor diazirine (azine
formation) in dilute solutions of hydrocarbons.
The intra- and intermolecular reactivities of alkyl- and
dialkylcarbenes have been actively studied in recent
years.2,3 For singlet dialkylcarbenes, 1,2-migration of
hydrogen (1,2-H) is well-known2,4 and is modulated by
substitution at the carbene carbon.2a For example, dim-
ethylcarbene has a lifetime (τ) of ∼7 ns4 in perfluorohex-
ane at ambient temperature which corresponds to k1,2-H
= 1.4 × 108 s-1, while in chloro(methyl)carbene5 and
bromo(methyl)carbene,6 k1,2-H ) (1-3) × 106 s-1 and
(5-6) × 106 s-1, respectively. By comparison, methoxy-
Presumably modulation of rate constants for 1,2-H
occurs through substituent effects on the stabilities of
both the ground and transition states for intramolecular
rearrangement. In some cases, 1,2-H in carbenes appears
to be accelerated by polar solvents.8 In other cases, the
formation of π-complexes with arenes may affect the rate
constants.9
The degree of overlap between the C-H bond orbital
of the migrating hydrogen and the virtual p orbital of
the carbene, in cases where conformational interconver-
sion is restricted, is believed to affect the absolute rate
constant for migration.10 In conformationally restricted
cyclohexylidenes, 1,2-H migrations could be separated
into contributions from axial and equatorial hydrogens.10-14
The ratio of rate constants at 120-160 °C (kax:keq) in
† NSERC Scholar 1996-1998.
‡ NRCC Visiting Scientist, summers, 1996-1998.
NRCC Research Associate, 1996-1998.
(1) Issued as NRCC No. 42177.
(2) For reviews, see: (a) Moss, R. A. In Advances in Carbene
Chemistry; Brinker, U., Ed.; J AI Press: Greenwich, 1994; Vol. 1, p
59. (b) J ackson, J . E.; Platz, M. S. In Advances in Carbene Chemistry;
Brinker, U., Ed.; J AI Press: Greenwich, 1994; Vol. 1, p 89.
(3) (a) J ackson, J . E.; Soundararajan, N.; Platz, M. S.; Doyle, M. P.;
Liu, M. T. H. Tetrahedron Lett. 1989, 30, 1335. (b) J ackson, J . E.;
Soundararajan, N.; Platz, M. S.; Liu, M. T. H. J . Am. Chem. Soc. 1988,
110, 5595. (c) Platz, M. S.; Modarelli, D.; Morgan, S.; White, W. R.;
Mullins, M.; Celebi, S.; Toscano, J . P. Progress of Reaction Kinetics;
Rodgers, M. A., Ed.; Elsevier: 1994; Vol. 19, p 93.
(4) (a) Modarelli, D. A.; Platz, M. S. J . Am. Chem. Soc. 1991, 113,
8985. (b) Morgan, S.; J ackson, J . E.; Platz, M. S. J . Am. Chem. Soc.
1991, 113, 2782. (c) Modarelli, D. A.; Morgan, S.; Platz, M. S. J . Am.
Chem. Soc. 1992, 114, 7034. (d) Ford, F.; Yuzawa, T.; Platz, M. S.;
Matzinger, S.; Fu¨lscher, M. J . Am. Chem. Soc. 1998, 120, 4430.
(5) (a) Bonneau, R.; Liu, M. T. H.; Rayez, M. T. J . Am. Chem. Soc.
1989, 111, 5973. (b) Liu, M. T. H.; Bonneau, R. J . Am. Chem. Soc.
1989, 111, 6873. (c) LaVilla, J . A.; Goodman, J . L. J . Am. Chem. Soc.
1989, 111, 6877.
(6) LaVilla, J . A.; Goodman, J . L. Tetrahedron Lett. 1990, 31, 5109.
(7) Sheridan, R. S.; Moss, R. A.; Wilk, B. K.; Shen, S.; Włostowski,
M.; Kesselmayer, M. A.; Subramanian, R.; Kmiecik-Lawrynowicz, G.;
Krogh-J espersen, K. J . Am. Chem. Soc. 1988, 110, 7563.
(8) Sugiyama, M. H.; Celebi, S.; Platz, M. S. J . Am. Chem. Soc. 1992,
114, 966.
(9) Moss, R. A.; Yan, S.; Krogh-J espersen, K. J . Am. Chem. Soc.
1998, 120, 1088.
(10) (a) Kenar, J . A.; Nickon, A. Tetrahedron 1997, 53, 14871. (b)
Nickon, A. Acc. Chem. Res. 1993, 26, 84. (c) Kenar, J . A.; Nickon, A.
Tetrahedron Lett. 1994, 35, 9657. (d) Evanseck, J . D.; Houk, K. N. J .
Am. Chem. Soc. 1990, 112, 9148.
10.1021/jo990171q CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/27/1999

1,2-Hydrogen Migration in Cyclohexylidenes
J . Org. Chem., Vol. 64, No. 12, 1999 4457
homobrexylidene (1)10a,11 is 2.2:1 and in 2,2-dimethyl-4-
tert-butylcyclohexylidene (2)10a,12 it is 1.9:1 at 155 °C.
Substituents such as axial or equatorial 2-phenyl in
5-tert-butylcyclohexylidene (3)13 or equatorial 2-methoxy
in 4-tert-butylcyclohexylidene (6)14 favor 1,2-H from the
2-position, leading preferentially to 4 and 7 (160 °C, eq
1; 190 °C, eq 2).10b Equatorial 2-methyl-4-tert-butylcyclo-
hexylidene affords 79.7% 5-tert-butyl-1-methylcyclohex-
ene and 20.3% 5-tert-butyl-3-methylcyclohexene.10c
F igu r e 1.
2-trifluoromethylcyclohexylidene (14d ), 8-aza-8-methyl-
[3.2.1]oct-3-ylidene (14e), diethylcarbene (14f), and eth-
ylmethylcarbene (14h ), Table 1. Oxadiazolines 12a -
c,e-h were prepared by oxidative cyclization (eq 3) of
Despite extensive studies of cyclohexylidenes, gener-
ated from precursors such as 1,2-diazaspiro[2.5]oct-1-ene
(9) by laser flash photolysis (LFP), all attempts15 to
estimate rate constants for 1,2-H by means of the
pyridine ylide trapping method3,4 failed. Excited state
reactions that mimic carbene reactions by forming the
same products (not necessarily in the same ratios) have
been found,16 and it is possible that excited state chem-
istry minimizes carbene formation from precursors such
as 9. Having used oxadiazoline precursors successfully
to study dialkylcarbenes,17 we turned to them again to
generate four different cyclohexylidenes. The photoprod-
ucts were determined from dual wavelength (254 and 300
nm) steady state (SS) photolysis experiments. Rate
constants for rearrangement of the carbenes were esti-
mated by intercepting them with pyridine in LFP experi-
ments.
the corresponding acyl hydrazones with lead tetraacetate
as described previously,18 and 12d was prepared by
alkylation of the pyrrolidine enamine of cyclohexanone
with CF3I to give 2-trifluoromethylcyclohexanone (10d ),19
which was converted to 11d and then to a mixture of
diastereomeric oxadiazolines 12d , Scheme 1.
Steady state and laser flash photolysis of oxadiazolines
analogous to 12 gives diazoalkanes,20 in contrast to the
thermal reactions of ∆3-1,3,4-oxadiazolines which lead to
dialkoxycarbenes,21 and dialkylcarbenes.17,22 Diazoal-
(18) El-Saidi, M.; Kassam, K.; Pole, D. L.; Tadey, T.; Warkentin, J .
J . Am. Chem. Soc. 1992, 114, 8751 and references therein.
(19) Cantacuze`ne, D.; Wakselman, C.; Dorme, R. J . Chem. Soc.,
Perkin Trans. 1 1977, 1365.
(20) (a) Pezacki, J . P.; Wagner, B. D.; Lew, C. S. Q.; Warkentin, J .;
Lusztyk, J . J . Am. Chem. Soc. 1997, 119, 1789. (b) Majchrzak, M. W.;
Be´khazi, M.; Tse-Sheepy, I.; Warkentin, J . J . Org. Chem. 1989, 54,
1842. (c) Majchrzak, M. W.; J efferson, E. A.; Warkentin, J . J . Am.
Chem. Soc. 1990, 112, 1842. (d) J efferson, E. A.; Warkentin, J . J . Org.
Chem. 1994, 59, 455. (e) Adam, W.; Finzel, R. Tetrahedron Lett. 1990,
31, 863.
(21) (a) Kassam, K.; Warkentin, J . Can. J . Chem. 1997, 75, 120. (b)
Win, W. W.; Kao, M.; Eiermann, M.; McNamara, J . J .; Wudl, F.; Pole,
D. L.; Kassam, K.; Warkentin, J . J . Org. Chem. 1994, 59, 5871. (c)
Kassam, K.; Warkentin, J . J . Org. Chem. 1994, 59, 5071-5. (d) Isaacs,
L.; Diederich, F. Helv. Chim. Acta 1993, 76, 2454-64. (e) El-Saidi, M.;
Kassam, K.; Pole, D. L.; Tadey, T.; Warkentin, J . J . Am. Chem. Soc.
1992, 114, 8751-2. (f) Rigby, J . H.; Cavezza, A.; Ahmed, G. J . Am.
Chem. Soc. 1996, 118, 12848. (g) Rigby, J . H.; Cavezza, A.; Heeg, M.
J . J . Am. Chem. Soc. 1998, 120, 3664.
Meth od s, Resu lts, a n d Discu ssion
Oxadiazolines 12a -h (Figure 1) served as precursors
of cyclohexylidene (14a), 4-tert-butylcyclohexylidene (14b),
(11) (a) Nickon, A.; Stern, A. G.; Ilao, M. C. Tetrahedron Lett. 1993,
34, 1391. (b) Stern, A. G.; Ilao, M. C.; Nickon, A. Tetrahedron 1993,
49, 8107.
(12) Kyba, E. P.; J ohn, A. M. J . Am. Chem. Soc. 1977, 99, 8329.
(13) Seghers, L.; Schechter, H. Tetrahedron Lett. 1976, 1943.
(14) Press, L. S.; Schechter, H. J . Am. Chem. Soc. 1979, 101, 509.
(15) Unpublished results by the Platz group.
(16) For discussions and leading references, see: (a) Moss, R. A. In
Advances in Carbene Chemistry; Brinker, U., Ed.; J AI Press: Green-
wich, 1994; Vol. 1, p 59. (b) Moss, R. A. Pure Appl. Chem. 1995, 67,
741. (c) Platz, M. S.; White, W. R., III; Modarelli, D. A.; Celebi, S. Res.
Chem. Intermed. 1994, 20, 175.
(17) Pezacki, J . P.; Pole, D. L.; Warkentin, J .; Chen, T.; Ford, F.;
Toscano, J .; Fell, J .; Platz, M. S. J . Am. Chem. Soc. 1997, 119, 3191.
(22) (a) Pezacki, J . P.; Wood, P. D.; Gadosy, T. A.; Lusztyk, J .;
Warkentin, J . J . Am. Chem. Soc. 1998, 120, 8681. (b) Pezacki, J . P.;
Warkentin, J .; Wood, P. D.; Lusztyk, J .; Yuzawa, T.; Gudmundsdottir,
A. D.; Morgan, S.; Platz, M. S. J . Photochem. Photobiol. A Chem. 1998,
116, 1.

4458 J . Org. Chem., Vol. 64, No. 12, 1999
Pezacki et al.
Ta ble 1. Lifetim es of Ca r ben es fr om Oxa d ia zolin es, Ded u ced w ith th e P yr id in e Ylid e Tech n iqu e
Sch em e 1
presence of pyridine, gives pyridinium ylides, and those
ylides arise from the trapping of the corresponding
dialkylcarbenes.17 Indirect evidence for the formation of
dialkylcarbenes from oxadiazolines was provided by
comparison of carbene lifetimes and pyridinium ylide
spectra derived from both an oxadiazoline and the
diazirine precursor of dimethylcarbene.17 However, LFP
(308 nm) of 5,5-dialkyl-2-methoxy-2-methyl-∆3-1,3,4-oxa-
diazolines in the presence of pyridine affords two pyri-
dinium ylides, one resulting from trapping of the dialky-
lcarbene (Scheme 2, upper path) and one resulting from
trapping of methoxy(methyl)carbene (Scheme 2, R3 ) Me,
lower path).17 For kinetic work it is therefore better to
use 2,2-dimethoxy-5,5-dialkyl-∆3-1,3,4-oxadiazolines, such
as 12a ,d ,e,h , because the reaction of dimethoxycarbene
with pyridine is too slow to be observed by ns LFP.23
La ser F la sh P h otolysis of Oxa d ia zolin es. LFP of
oxadiazolines 12a -h at 308 nm in the presence of
pyridine24 gave transient absorptions, all with λmax ∼360
nm. Examples of time-resolved pyridinium ylide spectra
from carbenes 14a -h are in Figure 2. The lifetimes of
those carbenes were determined by Stern-Volmer analy-
kanes are the primary photoproducts in steady state
photolyses with 300 nm light, and they may be photo-
lyzed further with 254 nm light to generate dialkylcar-
benes in steady state experiments. Such dual wavelength
experiments were done here to determine product dis-
tributions. LFP of oxadiazolines at 308 nm, in the
(23) Moss, R. A.; Włostowski, M.; Shen, S.; Krogh-J espersen, K.;
Matro, A. J . Am. Chem. Soc. 1988, 110, 4443.

1,2-Hydrogen Migration in Cyclohexylidenes
J . Org. Chem., Vol. 64, No. 12, 1999 4459
F igu r e 3. Absorbance of the pyridinium ylide of cyclohexy-
lidene vs pyridine concentration in cyclohexane at 22 °C. Inset
shows the double reciprocal plot of 1/absorbance of the pyri-
dinium ylide of cyclohexylidene vs 1/pyridine concentration in
cyclohexane at 22 °C.
of the pyridinium ylide at infinite [pyridine] and constant
laser intensity, ko is the sum of all the rate constants
leading to the disappearance of the transient carbene, τ
) 1/ko is the lifetime of the transient carbene, and kpyr is
the bimolecular rate constant for the reaction of the
carbene with pyridine. Fitting the data to a curve allows
one to solve for the product kpyrτ without using double
reciprocal plots.25
F igu r e 2. (a) Time-resolved UV-vis spectrum observed 380
ns after 308 nm LFP of 3,4-diaza-2,2-dimethoxy-1-oxa[4.3]-
spirodec-3-ene in cyclohexane containing 2.0 M pyridine at 22
°C. (b) Time-resolved UV-vis spectrum observed following 308
nm LFP of 8-tert-butyl-3,4-diaza-2-methoxy-2-methyl-1-oxa-
[4.3]spirodec-3-ene in cyclohexane containing 1.0 M pyridine
at 22 °C. The data were collected at intervals of 460 ns (O)
and 32.6 µs (9) after the laser pulse.
kpyr[pyridine]
AYLIDE ) A∞
)
YLIDEko + kpyr[pyridine]
Sch em e 2
kpyrτ[pyridine]
YLIDE1 + kpyrτ[pyridine]
A∞
(4)
Results of lifetime measurements in cyclohexane, cy-
clohexane-d12, and in benzene are summarized in Table
1. The lifetimes of carbenes 14a ,b,d ,e do not depend
significantly on the isotope (H or D) in the cyclohexane
solvent, and the slightly longer lifetimes in solutions of
benzene may be the result of the formation of π-complex-
es with these carbenes or to a slightly smaller value of
kpyr in benzene. Such π-complexes have been implicated
as the cause of extended carbene lifetimes in benzene vs
nonaromatic hydrocarbon solvents by as much as an
order of magnitude.9
Cyclohexylidenes are likely to have singlet ground
states with low-lying triplet states. Dimethylcarbene is
believed to have a singlet ground state,4 with ∆ES-T ∼1.6
kcal mol-1 26
Adamantylidene, with an angle constraint
.
similar to that expected in cyclohexylidene, is also a
sis of the maximum amplitudes of the ylide absorptions
as a function of [pyridine], Figure 3. The data were
analyzed by linear least-squares fitting of the curve to
eq 4 and by double reciprocal treatment of the data.3c In
eq 4, AYLIDE is the amplitude of absorbance of the
pyridinium ylide in the presence of various amounts of
singlet in the ground state, with ∆ES-T ∼3 kcal mol-1 27
.
In cyclohexylidenes, 1,2-H migration from the singlet
dominates over triplet state chemistry, presumably be-
cause the barriers for triplet state reactions are higher.28
As expected, 12b and 12f also afforded the pyridinium
ylide from methoxy(methyl)carbene in the LFP experi-
ments. Kinetic traces obtained following LFP of 12b at
added pyridine, A∞
is the amplitude of absorbance
YLIDE
(24) Oxadiazoline precursors give the best signals for pyridinium
ylides of dialkyl- and cycloalkylcarbenes when irradiated with 308 nm
by LFP. For these experiments pyridine purity was extremely impor-
tant as impurities (presumably the N-oxide) absorb at the excitation
wavelength. In some cases it was necessary to purify pyridine by ZnCl2
complexation and release to obtain solutions of pyridine which did not
absorb light at 308 nm, see: Heap, J .; J ones, W. J .; Speakman, J . B.
J . Am. Chem. Soc. 1921, 43, 1936. Very pure pyridine turns yellow
over a period of 10-30 min at room temperature when exposed to
oxygen.
(25) The major problem with the linearized functions is that they
give excessive weighting to points which may be poorly defined.
(26) (a) Matzinger, S.; Fu¨lscher, M. P. J . Phys. Chem. 1995, 99,
10747. (b) Richards, C. A., J r.; Kim, S.-J .; Yamaguchi, Y.; Schaefer,
H. F., III. J . Am. Chem. Soc. 1995, 117, 10104. (c) Sulzbach, H. M.;
Bolton, E.; Lenoir, D.; Schleyer, P. v. R.; Schaefer, H. F., III. J . Am.
Chem. Soc. 1996, 118, 9908.
(27) (a) Morgan, S.; J ackson, J . E.; Platz, M. S. J . Am. Chem. Soc.
1991, 113, 2782. (b) Bally, T.; Matzinger, S.; Truttman, C.; Platz, M.
S.; Morgan, S. Angew. Chem., Int. Ed. Engl. 1994, 33, 1964.

4460 J . Org. Chem., Vol. 64, No. 12, 1999
Pezacki et al.
substituent should raise the energies of twist boat
transition states for the H-migrations. The fact that the
tert-butyl group has an effect that is too small to be
detected by the method could mean that both 14b and
14a prefer 1,2-H from the chair conformation (19)30 or
that their lifetimes are too short for conformational
change to occur. A model for cyclohexylidene is cyclohex-
anone, which undergoes conformational change31 over a
barrier of <2.3 kcal/mol. At 25 °C, the rate constant must
be >109 s-1 and it is likely then that cyclohexylidenes do
undergo conformational change during their lifetimes and
that the method used here is not sensitive enough to
detect small substituent effects. The 1,2-H shift reaction
is very exothermic. Thus it will have a very early TS and
may not feel small substituent and conformational ef-
fects. Also the alkyl substituent may weaken the migrat-
ing CH bond and thus accelerate the rearrangement.
The lifetimes of carbenes 14a ,b,e in benzene, relative
to that of dimethylcarbene, are between (∼0.1-0.3):1
although they should be somewhat longer if statistics
were dominant. Presumably the increase in rates of 1,2-H
in the former carbenes is a “bystander effect” 10b resulting
mainly from increased substitution at the R-carbon,
which would stabilize positive charge there in the transi-
tion state (20). Analysis of the products of photolysis of
12a (hexadecane, 308 nm LFP, and 254 + 300 nm SS)
by GC-MS showed that cyclohexene is the major product
in both experiments. Analyses (1H NMR, 500 MHz) of
the products of photolysis of 12a -c, 12f, and 12h
(cyclohexane-d12, 254 + 300 nm SS) showed that 1,2-H
migration is also the major reaction pathway of carbenes
14a , 14b, 14f, and 14h . From diethylcarbene (14f), the
three major products were (E)-2-pentene, (Z)-2-pentene,
and 1,2-dimethylcyclopropane (∼15:10:1 (58%, 38%, 3.8%),
while ethyl(methyl)carbene (14h ) gave (E)-2-butene, (Z)-
2-butene, 1-butene, and methylcyclopropane (∼45:34:20:
1, respectively). Similar ratios of products have been
reported from ethyl(methyl)carbene generated from other
nitrogenous precursors.32 The barriers for migrations in
ethyl(methyl)carbene leading to (E)- and (Z)-2-butene,
1-butene, and methylcyclopropane have recently been
computed to be 5.2, 5.9, 8.5, and 8.3 kcal/mol, respec-
tively.33 It is likely that excited state migrations in the
diazo precursors of carbenes 14a -h also yield the alkene
products derived from 1,2-H in the carbenes.16,17 There-
fore, the ratios of products from dual wavelength irradia-
tions most likely reflect a composite of carbene and
excited state rearrangements. Contributions of each
process to the overall yield from a given carbene can
sometimes be estimated by increasing the concentration
of a suitable trap. When all the carbene can be trapped,
the yields of alkenes represent the contribution of the
excited state rearrangements. Attempts to reach satura-
F igu r e 4. (a) Trace obtained upon 308 nm LFP of 8-tert-butyl-
3,4-diaza-2-methoxy-2-methyl-1-oxa[4.3]spirodec-3-ene (12b)
in the presence of 0.4 M pyridine in cyclohexane at 22 °C. (b)
Trace obtained upon 308 nm LFP of 8-tert-butyl-3,4-diaza-2,2-
dimethoxy-1-oxa[4.3]spirodec-3-ene (12c) in the presence of 0.4
M pyridine in cyclohexane at 22 °C.
308 nm in the presence of pyridine, observed at 360 nm,
showed “instantaneous” growths which varied in ampli-
tude as a function of pyridine concentration (Aylide1, Figure
4) and a slow growth the kinetics of which also varied
with pyridine concentration (Aylide2, Figure 4). For meth-
29
oxy(methyl)carbene kpyr ) 5.5 × 105 M-1 s-1 and the
slow growth could easily be distinguished from the
“instantaneous” growths of signals for the pyridinium
ylides 15b and 15e. Linear least-squares fitting of the
rate constants for formation of the pyridinium ylide of
methoxy(methyl)carbene as a function of pyridine con-
centration gave kpyr ) 6.4 ( 0.9 × 105 M-1 s-1, in reason-
able agreement with the literature value.29 LFP (308 nm)
of oxadiazolines 12c and 12g in the presence of pyridine
led to only one observable ylide in each case; the ylides
of 14b (4-tert-butylcyclohexylidene) and 14f (diethylcar-
bene), respectively. The lifetimes of carbenes 14b and 14f,
derived from Stern-Volmer analyses of data obtained
from dimethoxyoxadiazolines 12c and 12g, were the
same as those from monomethoxy analogues 12b and 12f,
but analyses of the data from the dimethoxy precursors
were more straightforward (Figure 4).
1,2-Hyd r ogen Migr a tion in Cycloh exylid en e a n d
4-ter t-Bu tylcycloh exylid en e. The lifetimes of 14a and
14b, in both cyclohexane and benzene solutions, were the
same within experimental error, although the 4-tert-butyl
(28) It is most likely that vertical transitions lead to the formation
of singlet carbenes 114a -h in photolyses of oxadiazolines 12a -h . The
rate constants for reactions proceeding through the triplet state
manifold are then the products of the equilibrium constant for inter-
system crossing (ISC), KISC, and the rate constants for the reactions
of triplet carbenes 314a -h , kt (assuming that the preequilibrium
approximation applies) according to the scheme below.
(30) We prefer not to draw a virtual (unreal) p orbital, to which
migration formally occurs, at the carbene carbon.
(31) Oki, M. Applications of Dynamic NMR Spectroscopy to Organic
Chemistry. In Methods of Stereochemical Analysis; Marchand, A. P.,
Ed.; Vol. 4, VCH: Deerfield Beach, FL, 1985; pp 297-298.
(32) Mansoor, A. M.; Stevens, I. D. R. Tetrahedron Lett. 1966, 1733.
(33) Sulzbach, H. M.; Platz, M. S.; Schaefer, H. F., III; Hadad, C.
M. J . Am. Chem. Soc. 1997, 119, 5682.
KISC
kt
hν
12a -h 98 114a -h y z 314a -h
98 triplet derived products
(29) Ge, C. S.; J ang, E. G.; J efferson, E. A.; Liu, W.; Moss, R. A.;
Wostowska, J .; Xue, S. J . Chem. Soc., Chem. Commun. 1994, 1479.

1,2-Hydrogen Migration in Cyclohexylidenes
J . Org. Chem., Vol. 64, No. 12, 1999 4461
tion with tetramethylethylene as the trap for 14a were
not successful, and excited state rearrangements could
not be separated from carbene rearrangements.
ever observation of retardation is in accord with the view
that 1,2-H in singlet carbenes involves a transition state
in which the migrating H is hydride-like and, conse-
quently, the migration origin is cation-like. The analysis
is, of course, predicated upon the assumption that the
1,2-H from the excited state of 13d causes only minor
skewing of the product distribution.
Mod u la tion of Rea ctivity by a n R-Tr iflu or om -
eth yl Su bstitu en t. Previous studies of substituted cy-
clohexylidenes have shown that electron-donating sub-
stituents at the R-position favor 1,2-H from that site, as
inferred from product ratios.10c,34,35 2-Trifluoromethylcy-
clohexylidene (14d ) was therefore expected to undergo
1,2-H preferentially from C6. In practice, photolysis (254
+ 300 nm SS) of 12d in benzene and in hexadecane gave
3-trifluoromethyl- (21) and 1-trifluoromethylcyclohexene
(22) in a 9.8:1 ratio, according to 1H NMR, 19F NMR, and
GC-MS.36 Although rearrangement of diazoalkane 13d
in its excited state may have contributed to the formation
of 21 and 22, it is unlikely that hydrogen migration would
be more selective in the excited state than in the
carbene.16,17 Thus, the CF3 substituent in 14d favors
migration from the C6 position. The result could be a
composite of electronic and stereochemical effects, since
the substituent at C2 of cyclohexylidenes can affect
product ratios.10b,c,13,14 If the CF3 group were in the axial
position, then the geminal equatorial H could migrate
more slowly than an axially oriented counterpart. For-
tunately, it was possible to assign the conformation of
diazo compound 13d and, by extrapolation, that of
carbene 14d as described below.
Although the CF3 substituent affects the distribution
of products from carbene 14d substantially, the lifetime
of 14d is not significantly longer than that of the parent
cyclohexylidene, because the dominant reaction of 14d
is migration from C6, which is presumably not strongly
affected by the CF3 group at C2. Carbene 14d did not
show a significant solvent kinetic isotope effect (C6H12
vs C6D12), and it is therefore assumed that it is a ground
state singlet and that its chemistry in solution is pre-
dominantly that of the singlet state.
Da tive Sta biliza tion in 8-Aza -8-m eth ylbicyclo-
[3.2.1]oct-3-ylid en e (14e). Photolysis of 12e (LFP, 308
nm) in the presence of pyridine also gave a pyridinium
ylide. The lifetimes of 14e, which were determined as
described above, are given in Table 1. Although dative
stabilization of the carbenic center of 14e by the neigh-
boring amino group, in a pseudo-boat conformation (24),
might have enhanced its lifetime, the effect, if any, is
small. Moreover, the enhanced lifetime may not be real,
because kpyr may be reduced by steric hindrance, thereby
giving the appearance of a longer lived carbene. The same
qualifier applies to the other cyclohexylidenes that have
some hindrance to the approach of pyridine to the
carbene site.
Although 12d was obtained as a mixture of diastere-
omers, diazoalkane 13d was stable enough to be analyzed
by spectroscopy. The infrared spectrum recorded follow-
ing the photolysis of 12d in a SS experiment with 300
nm light (30 min) showed a diazo stretching band at 2052
cm-1 corresponding to 13d , a broad C-F stretching band
centered at 1282 cm-1, and a carbonyl band at 1756 cm-1
assigned to dimethyl carbonate. The COSY spectrum of
the photolysis mixture indicated that the two major
products were dimethyl carbonate (1H NMR 500 MHz,
C6D6, δ 3.38 ppm) and diazoalkane 13d , as a single
conformer. Cross-peaks in the COSY spectrum showed
that all peaks assigned to 13d are correlated with each
other and therefore originate from the same molecule.
Coupling constants for the CF3CH group of 13d (δ 2.37)
were 10.0 Hz, due to coupling to F of the CF3 group, as
well as 7.1 and 3.3 Hz from axial/axial and axial/
equatorial coupling to protons at C3. It follows that C2-H
and CF3 in 13d must be axial and equatorial, respec-
tively. Since 2-trifluoromethyldiazocyclohexane is gener-
ated initially in the SS experiment and is subsequently
converted to the carbene, the product ratio (∼9.8:1)
presumably reflects reactions of the carbene with the CF3
group in the equatorial position. Thus, migration of H
from C2 is that of axially oriented H (normally favored)
and its retardation, relative to migration from C6, must
reflect an electronic effect of the CF3 group. This first
Azin e F or m a tion . Steady state irradiations of oxa-
diazolines 12a -h with 300 nm light led to diazoalkanes
(as evidenced by the pink color of the resulting solutions
and the characteristic CNN band in IR spectra) in high
yield as determined spectroscopically (see Experimental
Section). The slow thermal decomposition of these diazo
compounds at room temperature gave the corresponding
azines with little (0-2%) or no products of rearrangement
of the corresponding carbene, even at concentrations of
10-3 M. This suggests that a mechanism other than the
reaction of carbenes 14a -h with their corresponding
diazoalkanes 13a -h (eq 5) is responsible for the forma-
tion of azines. For example, if cyclohexylidene has a
lifetime of ∼1 ns in a hydrocarbon solvent at room
temperature, then k1,2-H is ca. 1 × 109 s-1. Assuming that
cyclohexylidene traps diazocyclohexane, as in eq 5, with
(34) Seghers, L.; Shechter, H. Tetrahedron Lett. 1976, 1943.
(35) Press, L. S.; Shechter, H. J . Am. Chem. Soc. 1979, 101, 509.
(36) (a) Haas, A.; Plu¨mer, R.; Schiller, A. Chem. Ber. 1985, 118, 3004.
(b) Bouillon, J .-P.; Maliverny, C.; Mere`nyi, R.; Viehe, H. G. J . Chem.
Soc., Perkin Trans. 1 1991, 2147.
a rate constant of 5 × 109 M-1 s-1, then, at concentrations

4462 J . Org. Chem., Vol. 64, No. 12, 1999
Pezacki et al.
focusing mass spectrometer. Infrared (FT) spectra are from
samples in KBr windows.
Sch em e 3
Ma ter ia ls. Pyridine (Aldrich) was distilled from either
calcium hydride or barium oxide and stored under nitrogen
over potassium hydroxide. In some cases it was necessary to
purify pyridine by Zn complexation.24 Cyclohexane and ben-
zene (BDH Omnisolv) were distilled from sodium prior to use.
All other solvents and reagents were of the highest purity
commercially available and were used as received.
∆3-1,3,4-Oxa d ia zolin es (12). Oxadiazolines were prepared
by oxidative cyclization of the appropriate hydrazones using
lead tetraacetate as described previously.18 The products were
purified by radial chromatography on silica gel with hexane/
ethyl acetate (9:1) as the eluent. Yields were determined from
hydrazones. In a number of cases oxadiazoline structures were
confirmed using COSY, HSQC, and HMBC 2-dimensional
NMR techniques.
Ca r bom eth oxyh yd r a zon e of cycloh exa n on e (11a ):38
yield 91%; 1H NMR (200 MHz, CDCl3) δ 1.5-1.8 (m, 6H), 2.20
(dd, J ) 5.4, 6.4 Hz, 2H), 2.34 (dd, J ) 5.5, 6.3 Hz, 2H), 3.80
(s, 3H), 7.5-7.7 (br s, 1H).
3,4-D ia za -2,2-d im e t h o x y -1-o x a [4.3]s p ir o d e c -3-e n e
(12a ): yield 66%; clear oil; 1H NMR (500 MHz, CDCl3) δ 1.47-
2.01 (m, 10H), 3.42 (s, 6H); 13C NMR (50.3 MHz, CDCl3) δ 22.7
(+), 24.6 (+), 33.5 (+), 51.8 (-), 121.0 (+), 136.1 (+); IR (neat,
KBr) 3003, 2945, 2862, 1576, 1447, 1226 cm-1; UV λmax ) 328
nm (ꢀ ) 400); MS (EI) m/z (molecular ion not obsd), 169 [M -
OMe]+, 113, 91, 90, 81, 67, 59 (100%); MS (CI, NH3) m/z 218
[M + NH4]+, 201 [M + H]+. Anal. Calcd for C9H16N2O3: C
53.99, H 8.05, N 13.99. Found: C 53.75, H 8.18, N 13.40.
Acetylh yd r a zon e of 4-ter t-bu tylcycloh exa n on e (11b):
yield 95%; 1H NMR (200 MHz, CDCl3) δ 0.86 (s, 9H), 1.5-1.8
(m, 6H), 2.02 (dd, J ) 5.5, 6.4 Hz, 2H), 2.23 (dd, J ) 5.5, 6.4
Hz, 2H), 2.15 (s, 3H), 8.4-8.6 (br s, 1H).
8-ter t-Bu t yl-3,4-d ia za -2-m et h oxy-2-m et h yl-1-oxa [4.3]-
sp ir od ec-3-en e (12b): yield 71%; 4 diastereomers (10:10:1:
1, estimated by integration of the minor (mi.) and major (ma.)
methoxy signals); white solid, mp 15-20 °C; 1H NMR (500
MHz, CDCl3) δ 0.83 (s, mi.), 0.87 (s, mi.), 0.89 (s, ma.), 0.92 (s,
ma.), 1.21-2.25 (m) overlapping with 1.60 (s) and 1.62 (s), 3.08
(s, mi.), 3.10 (s, mi.), 3.12 (s, ma.), 3.17 (s, ma.); 13C NMR (50.3
MHz, CDCl3) δ 23.1, 23.3, 23.6, 23.8, 24.0, 25.1, 27.5, 27.7,
32.4, 32.8, 33.1, 34.2, 35.2, 36.9, 46.8, 47.3, 50.3, 50.5, 120.1,
133.8; UV λmax ) 328 nm (ꢀ ) 300); MS (EI) m/z (molecular
ion not obsd), 209 [M - OMe]+, 167, 155, 139, 123, 75, 59, 43
(100%); MS (CI, NH3) m/z 258 [M + NH4]+, 241 [M + H]+.
Anal. Calcd for C13H24N2O2: C 64.97, H 10.06, N 11.66.
Found: C 64.72, H 9.76, N 11.46.
of 10-3 M diazoalkane, intramolecular rearrangement
rather than intermolecular capture of cyclohexylidene
should dominate. Therefore, a bimolecular reaction of
diazoalkanes 13a -h , such as that in Scheme 3, must be
responsible for the high yields of azines. The lack of
significant yields of azines from dual wavelength irradia-
tions, where dialkyl- and cycloalkylcarbenes are gener-
ated photochemically, supports this conclusion. Dimer-
ization of diazo compounds has been proposed previously
to explain stereoselective azine formation in the decom-
position of phenyldiazomethanes.37
Con clu sion s
Three substituted cyclohexylidenes and the parent,
generated photochemically from oxadiazolines by LFP,
could be trapped as the corresponding pyridinium ylides
to yield the first absolute rate constants for 1,2-H
migrations in cyclohexylidenes. Those rate constants,
which are about 10-fold larger than those of acyclic
analogues, show that the effect of a 4-tert-butyl substitu-
ent is small and that an amino group, which could
stabilize the carbene intermediate by intramolecular
electron donation, does not have a large effect. 2-Trif-
luoromethyl is the first substituent shown to decrease
the rate constant for 1,2-H migration from the 2-position,
by roughly 10-fold, as inferred from the product distribu-
tion.
Diethylcarbene and ethyl(methyl)carbene can also be
generated by means of the oxadiazoline approach. Those
carbenes undergo 1,2-H migrations about 8-fold more
slowly than cyclohexylidene, ca. 1.7 × 108 s-1 vs ca. 1.4
× 109 s-1 in hydrocarbon solvent, assuming that kpyr is
invariant at 1 × 109 M-1 s-1 . Previous failures to trap
cyclohexylidenes with pyridine can probably be attributed
to low yields of the carbenes (e.g., from diazirines), rather
than to extraordinarily fast 1,2-H migrations.
Car bom eth oxyh ydr azon e of 4-ter t-bu tylcycloh exan on e
(11c):38 yield 81%; 1H NMR (200 MHz, CDCl3) δ 0.89 (s, 9H),
1.5-1.8 (m, 6H), 2.19 (dd, J ) 5.5, 6.3 Hz, 2H), 2.35 (dd, J )
5.5, 6.3 Hz, 2H), 3.80 (s, 3H), 7.5-7.7 (br s, 1H).
8-ter t-Bu tyl-3,4-diaza-2,2-dim eth oxy-1-oxa[4.3]spir odec-
3-en e (12c): yield 77%, 2 diastereomers (1.3:1 estimated by
1
integration of the methoxy signals); viscous liquid; H NMR
(500 MHz, CDCl3, integrations normalized for minor (mi.) and
major (ma.) isomers) δ 0.86 (s, mi., 9 H), 0.91 (s, ma., 9 H),
1.12-2.15 (m), 3.44 (s, mi., 6 H); 3.46 (s, ma., 6 H); 13C NMR
(50.3 MHz, CDCl3) δ 23.3, 24.8, 27.5, 27.6, 32.4, 33.2, 35.2,
46.5, 47.1, 51.9, 120.9, 136.4; UV λmax ) 328 nm (ꢀ ) 300); MS
(EI) m/z (molecular ion not obsd), 225 [M - OMe]+, 204, 167,
109, 101, 75, 57, 43 (100%); MS (CI, NH3) m/z 257 [M + H]+.
Anal. Calcd for C13H24N2O3: C 60.91, H 9.44, N 10.93. Found:
C 60.56, H 9.66, N 10.32.
Steady state photolysis of the oxadiazolines leads to
the same carbenes, via diazoalkane intermediates. With
a suitable choice of wavelength, the diazoalkanes can be
accumulated and characterized. The steady state pho-
tolysis makes it possible to characterize carbene-derived
products under conditions that roughly simulate LFP
conditions.
6-Tr iflu or om eth yl-3,4-d ia za -2,2-d im eth oxy-1-oxa [4.3]-
sp ir od ec-3-en e (12d ). A two-necked, 500 mL round-bottom
flask was fitted with an acetone-dry ice condenser, and CF3I,
ca. 6 g, was transferred into the flask. Pentane (75 mL) was
added, followed by 9.0 g (60 mmol, 0.5 equiv) of the pyrrolidine
enamine of cyclohexanone (18) in 75 mL of pentane. The
solution was stirred at room temperature for 3 h. The
precipitate was then filtered off and washed with pentane, and
Exp er im en ta l Section
Gen er a l In for m a tion . GC-MS analyses were carried out
with a gas chromatograph equipped with a mass selective
detector and a DB-1 capillary column (12 m × 0.2 mm). UV-
visible spectra were measured with a double-beam spectro-
photometer, and mass spectra were obtained with a double-
(37) Abelt, C. J .; Pleier, J . M. J . Am. Chem. Soc. 1989, 111, 1795.
(38) Okimoto, M.; Chiba, T. J . Org. Chem. 1990, 55, 1070.

1,2-Hydrogen Migration in Cyclohexylidenes
the solvent was removed with rotary evaporator. The
J . Org. Chem., Vol. 64, No. 12, 1999 4463
a
Acetylh yd r a zon e of 3-p en ta n on e (11g):20b yield 93%; 1H
NMR (200 MHz, CDCl3) δ 1.02 (t, J ) 7.5 Hz, 6H), 1.95 (q,
4H), 2.14 (s, 3H) 8.2-8.4 (br s, 1H).
resulting oil was acidified with 5 M H2SO4 and stirred at room
temperature for another 3 h. Extraction with ether (3 × 50
mL) was followed by washing of the organic layer with water
(20 mL), 5% sodium bicarbonate (3 × 25 mL), and brine (25
mL). After the solution had been dried over magnesium sulfate
it was filtered, and the solvent was removed by rotary
evaporation. The product, 2-trifluoromethylcyclohexanone (10d),
was purified by fractional distillation at ca. 25 Torr. The
fraction collected at 85-87 °C (2.0 g) consisted of 85% 10d
and 15% cyclohexanone as determined by GC; overall yield
33%. A solution of benzene (50 mL) containing 2-trifluorom-
ethylcyclohexanone (10d , 1.77 g, 9 mmol, weight adjusted for
cyclohexanone impurity), methyl hydrazinocarboxylate (1.24
g, 13.8 mmol), and acetic acid (2 drops) was refluxed for 24 h
in a Dean-Stark apparatus. The crude product containing 11d
was isolated as an oil, and attempts to crystallize 11d from it
(various solvent mixtures) failed: 1H NMR (500 MHz, CDCl3)
δ 1.2-2.1 (m, 8H), 2.1-2.3 (m, 3H), 3.80 (s, 3H), 7.6-7.8 (br
s, 1H). Hydrazone 11d (2.92 g, 1.23 mmol, weight adjusted
for carbomethoxyhydrazone of cyclohexanone) dissolved in 20
mL of methanol under N2 was added from a dropping funnel
during 30 min to an ice-cold solution of lead tetraacetate (LTA,
5.55 g, 12.5 mmol) in 25 mL of methanol. The solution was
kept at ca. -10 °C for 5 days with occasional stirring before
the solvent was removed by rotary evaporation and methylene
chloride (50 mL) was added to the residue. The solution was
filtered to remove inorganic salts, and it was washed with 5%
sodium bicarbonate (4 × 25 mL) before the organic layer was
dried over magnesium sulfate. Removal of the solvent left
oxadiazoline 12d which was purified from minor amounts of
oxadiazoline 12a by repeated radial chromatography on silica,
eluting with 5-20% ethyl acetate in hexane. Yield of 12d , 12%,
a 2:1 mixture of diastereomers; clear liquid: 1H NMR (500
MHz, CDCl3, integrations normalized for minor (mi.) and
major (ma.) isomers) δ 1.42-2.03 (m), 2.05-2.15 (m), 2.25-
2.40 (m), 3.45 (s, ma., 3H), 3.52 (s, mi., 3H), 3.54 (s, mi., 3H),
3.55 (s, ma., 3H); 13C NMR (125 MHz, CDCl3) δ 22.8, 22.9,
23.3, 23.4, 24.1, 24.8, 33.7, 36.1, 46.0 (q, J ) 26.3 Hz), 48.2 (q,
J ) 26.0 Hz), 51.9, 52.0, 52.1, 118.1, 118.6, 120.3 (q, J ) 299
Hz), 125.4 (q, 280 Hz), 136.2, 137.4; 19F NMR (470 MHz, CDCl3
referenced to CFCl3) δ -64.9 (ma., d, J ) 7.5 Hz), -66.3 (mi.,
d, J ) 7.5 Hz); UV λmax ) 328 nm (ꢀ ) 300); MS (EI) m/z
(molecular ion not obsd), 237 [M - OMe]+, 219, 169, 150, 131,
119, 100, 69 (100%). Anal. Calcd for C10H15F3N2O3: C 44.78,
H 5.64, N 10.44. Found: C 44.67, H 5.53, N 10.05.
5,5-Die t h yl-2-m e t h oxy-2-m e t h yl-∆3-1,3,4-oxa d ia zo-
lin e (12g): yield 68%; clear oil; 1H NMR (500 MHz, CDCl3) δ
0.86 (t, J ) 7.5 Hz, 6H), 0.95 (t, J ) 7.5 Hz, 6H), 1.59 (s, 3H),
1.65-2.05 (m, 4H), 3.18 (s, 3H); 13C NMR (125 MHz, CDCl3) δ
8.1, 14.4, 22.1, 29.4, 29.6, 50.3, 122.8, 133.1; UV λmax ) 328
nm (ꢀ ) 500); MS (EI) m/z (molecular ion not obsd), 141 [M -
OMe]+, 113, 91, 75, 59 (100%), 43; MS (CI, NH3) m/z 190 [M
+ NH4]+,173 [M + H]+.
Ca r bom eth oxyh yd r a zon e of 2-bu ta n on e (12h ): yield
1
88%; H NMR (200 MHz, CDCl3) δ 1.08 (t, J ) 7.5 Hz, 3H),
1.78 (s, 3H), 2.31 (q, J ) 7.5 Hz, 2H), 3.80 (s, 3H), 7.5-7.7 (br
s, 1H).
5-E t h yl-2,2-d im e t h oxy-5-m e t h yl-∆3-1,3,4-oxa d ia zo-
lin e (12h ): yield 70%; clear oil; 1H NMR (500 MHz, CDCl3) δ
0.81 (dd, J ) 7.6 Hz, 3H), 1.36 (s, 3H), 1.69 (ABX3, J ) 7.6,
-14.3 Hz, 1H), 1.79 (ABX3, J ) 7.6, -14.3 Hz, 1H), 3.44 (s,
3H), 3.49 (s, 3H); 13C NMR (125 MHz, CDCl3) δ 7.4, 21.4, 30.3,
51.4, 51.6, 121.8, 136.7; IR (neat, KBr) 2982, 2949, 2887, 2847,
1577, 1459, 1448, 1376, 1137, 1078, 930, 862 cm-1; UV λmax
)
328 nm (ꢀ ) 500); MS (EI) m/z (molecular ion not obsd), 143
[M - OMe]+, 91, 74, 59 (100%); MS (CI, NH3) m/z 192 [M +
NH4]+, 175 [M + H]+. Anal. Calcd for C7H14N2O3: C 48.26, H
8.10, N 16.08. Found: C 48.20, H 8.14, N 15.79.
La ser F la sh P h otolysis. The nanosecond laser flash
photolysis system at NRC has been described.39 A Lumonics
EX-530 excimer laser (XeCl, 308 nm, 6 ns pulses, g40 mJ /
pulse) was used for photolyses of oxadiazoline precursors.
Sample solutions with absorbances (A) of 0.3-0.6 at the
excitation wavelength (308 nm) were prepared as described
below. These solutions were irradiated in 7 × 7 mm quartz
cells. The nanosecond laser flash photolysis system at Ohio
State has been described.40 An excimer laser (Lambda Physic
LPX-100) was used for photolysis of an oxadiazoline precursor.
Stock solutions of oxadiazolines 12a -h were prepared in
freshly distilled benzene or cyclohexane containing various
concentrations of pyridine (typically between 0.1 and 5.0 M)
so that the values of A308 were approximately 0.4-0.6. Typical
concentrations of oxadiazolines required to accomplish this
were ∼(2-3) × 10-3 M. For each oxadiazoline, solutions
without pyridine were degassed with N2, and the time-resolved
UV-vis spectra for each were acquired (300-700 nm) after
308 nm LFP to ensure that signals assigned to pyridinium
ylides did not arise from intermediates other than pyridinium
ylides. Stern-Volmer kinetics were obtained from values of
A360 of solutions of oxadiazolines 12a -h containing various
amounts of pyridine, after 308 nm LFP.
Ca r bom eth oxyh yd r a zon e of tr op in on e (11e): yield
1
91%; H NMR (200 MHz, CDCl3) δ 1.4-1.6 (m, 2H), 1.9-2.1
(m, 4H), 2.40 (s, 3H), 2.62 (dd, J ) 4.4, -14 Hz, 4H), 3.3-3.5
(m, 2H), 3.80 (s, 3H), 7.5-7.7 (br s, 1H).
For each pyridine quenching experiment, UV-visible spec-
tra were taken of each solution prior to LFP, to ensure that
the absorbances of the solutions at the excitation wavelength
remained constant. Residual absorption of impurities in py-
ridine at the excitation wavelength can lead to errors in Stern-
Volmer quenching, and the stock solutions of pyridine and LFP
solutions were carefully monitored to eliminate this possibility.
St ea d y St a t e P h ot olysis. Steady state photolyses were
carried out in a Rayonet photochemical reactor equipped with
a “merry-go-round” apparatus. Photolytic conversions of oxa-
diazolines to diazo compounds were performed by irradiating
0.001 M solutions of oxadiazolines 12a -h in deoxygenated (N2
or argon) benzene-d6 in Pyrex reaction vessels (NMR tubes in
some cases) with 10-14 300 nm lamps. Irradiation for 15-45
min was sufficient to convert each oxadiazoline completely to
its corresponding diazoalkane. Diazoalkanes 13a -h were
stable for several hours at room temperature in dilute solution,
but thermal azine formation occurred at room temperature
over a period of 4-5 days, after which the solutions were
analyzed by 1H NMR and GC-MS. Conversions of oxadiazolines
12a -h to cyclohexylidene (14a ), substituted cyclohexylidenes
(1r,3â,5r)-5′,5′-Dim eth oxy-8-m eth ylsp ir o[8-a za bicyclo-
[3.2.1]octa n e]-3,2′-[∆3-1,3,4-oxa d ia zolin e] (12e): yield 32%,
one diastereomer, presumably because of participation of the
amino group during oxidative cyclization, clear liquid; 1H NMR
(500 MHz, CDCl3) δ 1.23 (d, J ) 13.8 Hz, 1H), 1.35 (d, J )
13.9 Hz, 1H), 1.76-1.91 (m, 4H), 2.08 (s, 3H), 2.09 (s, 3H),
2.10-2.17 (m, 1H), 2.65 (dd, J ) 13.9, 3.5 Hz, 1H), 2.82-2.89
(m, 2H) 3.24 (s, 6H); 13C NMR (125 MHz, CDCl3) δ 25.5, 25.8,
26.9, 37.7, 38.8, 39.3, 39.7, 51.3, 51.5, 59.4, 59.9, 118.8, 138.9;
UV λmax ) 328 nm (ꢀ ) 400); MS (EI) m/z (molecular ion not
obsd), 210 [M - OMe]+, 198, 162, 155, 138, 108, 95, 82 (100%);
MS (CI, NH3) m/z 259 [M + NH4]+, 242 [M + H]+. Anal. Calcd
for C11H19N3O3: C 54.76, H 7.94, N 17.41. Found: C 54.38, H
8.23, N 17.20.
Ca r bom eth oxyh yd r a zon e of 3-p en ta n on e (11f):38 yield
1
82%; H NMR (200 MHz, CDCl3) δ 1.06 (t, J ) 7.5 Hz, 6H),
2.32 (q, J ) 7.5 Hz, 4H), 3.80 (s, 3H), 7.5-7.7 (br s, 1 H).
5,5-Dieth yl-2,2-d im eth oxy-∆3-1,3,4-oxa d ia zolin e (12f):
yield 66%; clear oil; 1H NMR (500 MHz, CDCl3) δ 0.88 (t, J )
7.5 Hz, 6H), 1.86 (q, J ) 7.5 Hz, 4H), 3.53 (s, 6H); 13C NMR
(125 MHz, CDCl3) δ 7.3, 28.3, 51.6, 124.9, 136.6; UV λmax
)
328 nm (ꢀ ) 500); MS (EI) m/z (molecular ion not obsd), 157
[M - OMe]+, 129, 91, 75, 59 (100%), 43; MS (CI, NH3) m/z
206 [M + NH4]+, 189 [M + H]+. Anal. Calcd for C8H16N2O3: C
51.05, H 8.57, N 14.88. Found: C 50.96, H 8.27, N 14.80.
(39) Kazanis, S.; Azarani, A.; J ohnston, L. J . J . Phys. Chem. 1991,
95, 4430.
(40) Gritsan, N. P.; Zhai, H. B.; Yuzawa, T.; Brooke, J .; Platz, M. S.
J . Phys. Chem. A 1997, 101, 2833.

4464 J . Org. Chem., Vol. 64, No. 12, 1999
Pezacki et al.
14b,d ,e, and dialkylcarbenes 14f,h were done by irradiating
0.01 M solutions of oxadiazolines 12a -h in deoxygenated (N2
or argon) cyclohexane-d12 in quartz reaction vessels (NMR
tubes in some cases) with 8-12 254 nm lamps and 1-2 300
nm lamps. Dual wavelength irradiation was required to
convert each oxadiazoline to the corresponding carbene ef-
ficiently, and typical reaction times for complete conversion
of the starting materials were 2-4 h. The reaction times were
shortest when 12 254 nm lamps were used in conjunction with
2 300 nm lamps. Short reaction times were preferable to
minimize azine formation. Dual wavelength irradiations of
solutions of oxadiazolines in hexadecane allowed for GC and
GC-MS analyses of reaction mixtures. Compounds 12a ,c-
e,g,h afforded dimethyl carbonate as a coproduct, whereas
compounds 12b,f afforded methyl acetate as a coproduct. The
steady state photolysis results have been deposited as Sup-
porting Information.
McMaster) for help with NMR experiments, Dr. W. J .
Leigh (McMaster) for helpful discussions during the
course of this work, Dr. D. Shukla (NRC) for help with
pyridine purification, and the Natural Sciences and
Engineering Research Council (NSERC) of Canada for
support through a Grant-in-Aid. J .P.P. and J .A.D.
gratefully acknowledge NSERC of Canada for graduate
scholarships (1996-1998). M.S.P. gratefully acknowl-
edges the support of the NSF (CHE-8814950). We thank
Dr. Kurt Loening for his gracious assistance with some
of the nomenclature.
Su p p or tin g In for m a tion Ava ila ble: Experimental data
pertaining to steady state photolyses, infrared and COSY
spectra. This material is available free of charge via the
Internet at http://pubs.acs.org.
Ack n ow led gm en t. J .P.P, P.C., J .A.D., and J .W.
thank Dr. Don Hughes and Mr. Brian Sayer (both at
J O990171Q