Flash Vacuum Pyrolysis of Pyrazoles
J . Org. Chem., Vol. 63, No. 23, 1998 8191
formed from a direct [1,2] hydrogen shift to a sp2 carbon
or through a [1,3] hydrogen shift in the undetected allene.
On the other hand, 3-(trifluoromethyl)butyne-1 (25) and
3-(trifluoromethyl)-4,4,4-trifluorobutyne-1 (26) are un-
doubtedly formed from the undetected allene through a
[1,3] CF3 shift.
It is known that methyl groups have poor migration
ability,12 and probably [1,3] shifts of trifluoromethyl
groups are comparable to [1,3] hydrogen shifts, both
being of lower energy than those of the methyl groups.
This trifluoromethyl group shift was also detected in
some thermal reactions of isoxazoles.19
Nitrogen extrusion reactions from 1-benzoylpyrazole
(27) should also be considered. FVP reactions of 27 show
three competitive ways, all of them taking place at the
same reaction temperature, indicating that they have
similar energies: isomerization, radical fission, and
nitrogen extrusion.4b Of these reactions, the one impor-
tant for our present purpose is nitrogen extrusion,
affording 2-phenylfuran (28). This is an interesting
result because it indicates that addition to a carbonyl
bond is exclusive over H shift and insertion into Csp2-H
bond.
Oxygen free dry nitrogen or a nitrogen/toluene mixture was
used as carrier gas. Products were trapped at the liquid air
temperature, extracted with the appropriate solvent, and
submitted to the different analyses or separation techniques.
Samples to be pyrolyzed were 30-50 mg. Contact times were
10-2 s, and pressures were 0.2 to 0.1 Torr. Column chroma-
tography and TLC were performed on silica gel.
F VP of 1. Compound 1 was prepared as described in the
literature.20 After the reaction was finished, reaction products
were separated by column chromatography (petroleum ether)
and then submitted to spectroscopy, affording the following
data. Compound 3: 1H NMR (Cl4C), δ (ppm) 3.50 (2H, d, J )
2 Hz), 6.65 (1H, t, J ) 2 Hz), 7.50 (9H, m). MS m/z (relative
intensity) 192 (M+, 100), 165 (18), 115 (8). These results
agreed with those previously reported for 3.21 Compound 4:
1H NMR (Cl4C), δ (ppm) 3.75 (2H, s), 6.90-7.60 (10H, m).22
MS m/z (relative intensity) 192 (M+, 100), 165 (13), 115 (5).
Kinetic measurements were done with 1H NMR. After the
reaction was finished, the entire crude reaction mixture was
extracted with CCl4 and suitably diluted for 1H NMR mea-
surement. After choosing a proton signal for 1, 3, and 4, the
total amount was taken as Co (100%), and then the percentage
1
of 1 was calculated. These H NMR experiments were carried
out in CCl4 solution with a capillary tube filled with acetone-
d6 inside the NMR tube.
F VP of 2. Compound 2 was synthesized as described in
the literature.20 The FVP reaction crude was submitted to
column chromatography (petroleum ether and petroleum
ether:chloroform, 60:40), and the different isolated products
afforded the following spectral data. Compound 5: 1H NMR
(DMSO-d6) δ (ppm) 5.70 (2H, m), 7.10 (3H, m), 7.30-7.90 (5H,
m). These results agree with previous ones.23 Compound 6
was identified by comparison with an authentic sample. 1H
NMR (DMSO-d6), δ (ppm) 7.65 (4H, m), 8.05 (4H, m). Com-
pound 7: 1H NMR (DMSO-d6), δ (ppm) 1.27 (3H, t, J ) 8 Hz),
2.45 (2H, q, J ) 8 Hz), 7.10-7.80 (5H, m).24 IR (NaBr) 2259
cm-1 (CtC). Compound 8: 1H NMR (DMSO-d6), δ (ppm) 1.33
(3H, d, J ) 7 Hz), 3.15 (1H, q, J ) 7 Hz), 6.15 (2H, dd, J ) 2
and 6 Hz), 7.10 (4H, m).25 MS m/z (relative intensity) 130 (M+,
100). Compound 9: 1H NMR (DMSO-d6) δ (ppm) 2.25 (3H,
s), 3.35 (2H, m), 6.38 (1H, m), 7.20-7.50 (4H, m);25 MS m/z
(relative intensity) 130 (M+, 100). Compound 10: 1H NMR
(DMSO-d6) δ (ppm) 2.18 (3H, s), 3.30 (2H, s), 6.50 (1H, s),
7.10-7.40 (4H, m);25 MS (relative intensity) 130 (M+, 100).
Con clu sion
The fact that FVP reactions of pyrazoles may be used
to study vinylcarbenes arising from vinyldiazo com-
pounds should be remarked. This alternative opens a
new way to study these intermediates, considering that
synthesis of NH pyrazoles is easier than that of the
corresponding vinyldiazo isomer and that no competitive
reactions are present in thermal studies.
Some conclusions on the thermal reactions of vinyl-
carbenes can also be drawn:
(1) With at least two aliphatic substituents, the lower
energy reaction is a hydrogen shift. (2) With at least two
aromatic substituents, insertion into an aryl C-H bond
is the lower energy reaction. (3) With asymmetric
substituents (aliphatic and aromatic), a hydrogen shift
as well as insertion into an aryl C-H bond are competi-
tive reactions, which may be evidence that at least two
substituents of the same kind should be present to get a
selective reaction. (4) In fluorinated vinylcarbenes, a
hydrogen shift is of similar energy as fluorine and CF3
shifts. (5) In vinylcarbenes with a carbonyl bond, if the
geometry is optimum, the only reaction is addition to the
carbonyl bond; neither insertion nor hydrogen shifts are
competitive reactions.
Ack n ow led gm en t. This work was supported by
CONICET, CONICOR, SECYT-UNC, and FUNDA-
CION ANTORCHAS to whom we are indebted.
J O980670B
(20) Elguero, J ., J acquier, R. Bull. Soc. Chim. Fr. 1966, 90.
(21) Friedrich, E.; Taggart, D. J . Org. Chem. 1975, 40, 723.
(22) Galton, S. A., Kalafer, M., Beringer, F. M. J . Org. Chem. 1970,
35, 1.
(23) Padwa, A.; Carusso, T., Nahm S.; Rodriguez, A. J . Am. Chem.
Soc. 1982, 104, 2869.
Exp er im en ta l Section
Gen er a l. Reactions were carried out in a Vycor glass flash
vacuum thermolysis equipment, using a GAYNOR PRDH
temperature controller and a Thermolyne 21100 furnace.
(24) Aldrich NMR FT Catalog 1, 1993, 3, 532A.
(25) Bosch, A., Brown, R. K. Can. J . Chem. 1975, 40, 1718.
(19) Wunderlin, D. A. Personal communication.