Retro-Ene Reactions in Acylallene Derivatives

Article abstract of DOI:10.1021/jo972137m

Allenic esters and amides 4 undergo a retro-ene reaction to vinylketene (6) and an aldehyde or imine (5) under the conditions of flash vacuum thermolysis (FVT). The same products are obtained by FVT of cyclobutenones 7 via electrocyclic ring opening to alkoxy- or aminovinylketenes 3 and 1,3-rearrangement of ketenes 3 to allenes 4. All the intermediates and products were characterized by matrix isolation IR spectroscopy, and in the case of 4c the reaction was also monitored by online mass spectrometry. A lower temperature for the retro-ene reaction of 4c, eliminating an imine, than for 4a, eliminating formaldehyde, is in agreement with a lower calculated activation barrier (167 and 181 kJ mol^-1, respectively, at the G2(MP2,SVP) level of theory). The allenic amide 11 undergoes an analogous retro-ene reaction to the (unobserved) vinylketene 13, the latter isomerizing to cyclohexenylacrolein 16 in a 1,5-H shift (calculated barrier 125 kJ mol^-1; G2 (MP2, SVP)).

Full text of DOI:10.1021/jo972137m

J . Org. Chem. 1998, 63, 2619-2626
2619
Retr o-En e Rea ction s in Acyla llen e Der iva tives
Herve´ Bibas, Rainer Koch,† and Curt Wentrup*
Department of Chemistry, The University of Queensland, Brisbane, QLD 4072, Australia
Received November 21, 1997
Allenic esters and amides 4 undergo a retro-ene reaction to vinylketene (6) and an aldehyde or
imine (5) under the conditions of flash vacuum thermolysis (FVT). The same products are obtained
by FVT of cyclobutenones 7 via electrocyclic ring opening to alkoxy- or aminovinylketenes 3 and
1,3-rearrangement of ketenes 3 to allenes 4. All the intermediates and products were characterized
by matrix isolation IR spectroscopy, and in the case of 4c the reaction was also monitored by online
mass spectrometry. A lower temperature for the retro-ene reaction of 4c, eliminating an imine,
than for 4a , eliminating formaldehyde, is in agreement with a lower calculated activation barrier
(167 and 181 kJ mol-1, respectively, at the G2(MP2,SVP) level of theory). The allenic amide 11
undergoes an analogous retro-ene reaction to the (unobserved) vinylketene 13, the latter isomerizing
to cyclohexenylacrolein 16 in a 1,5-H shift (calculated barrier 125 kJ mol-1; G2 (MP2, SVP)).
Sch em e 1
In tr od u ction
In recent papers we have described the thermal 1,3-
migrations of groups R in the interconversions of R-ox-
oketenes (1a f 2a ),1 imidoylketenes and R-oxoketen-
imines (1b f 2b),2 and vinylketenes and acylallenes (1c
f 2c) (eq 1).3 As observed experimentally, these reac-
tions are highly facilitated by electron rich migrating
groups R, especially those possessing lone pairs, due to
a favorable interactions between a high-lying orbital of
the migrating group and the low-lying ketene-type LUMO,
which has a large coefficient at the central cumulenic
carbon atom in the plane of the molecule.3a,4 The
calculated migratory aptitudes are in the order NMe2 >
SMe > SH > Cl > NH2 > OMe > OH > F > H > Ph >
Me (G2(MP2,SVP) theory).3a,4 The activation barriers for
the 1,3-shifts are, therefore, related to the LUMO ener-
gies and to the exothermicities or endothermicities of the
reactions.3a,5 The calculated activation barriers for the
1,3-H shift increase in the series R-oxoketene < R-ox-
oketenimine < imidoylketene < acylallene < vinylketene
(this ordering can be influenced by substituents). There-
fore, in the vinylketene/acylallene system (1c/2c), one can
only expect to observe this migration for the most
electron-donating substituents, i.e., dialkylamino, alky-
lthio, chloro, and alkoxy. Indeed, these are the reactions
that have been documented.3,6 They take place under
flash vacuum thermolysis (FVT) conditions in the tem-
perature range 370-630 °C. The calculated activation
barriers, starting from vinylketenes 1c, are 129 (NMe2),
147 (SMe), 157 (Cl), and 169 (OMe) kJ mol-1, respec-
tively.3 The facile 1,3-chlorine migration (1c/2c; R ) Cl;
R′ ) H) takes place above 370 °C prior to loss of HCl to
give butatrienone, H2CdCdCdCdO.3a In the case of
ethoxyvinylketene/ethyl allenecarboxylate (1c/2c; R )
OEt), the 1,3-ethoxy shift takes place prior to elimination
of ethene to give acetylketene.3
We now report conclusive evidence for a retro-ene
reaction operating in this system, which at higher FVT
temperatures causes loss of formaldehyde (1c/2c; R )
OMe) or N-methylmethanimine (R ) NMe2) with forma-
tion of the unsubstituted vinylketene (6; Scheme 1). The
ene- and retro-ene reactions are well-known in
hydrocarbon chemistry.7 For example, the retro-ene
reaction of pentene produces propene and ethene. The
experimental7c and calculated7d,e (MP2/6-31G*) activation
energies are of the order 35-38 kcal mol-1 (140-160 kJ
† Present address: Fachbereich Chemie der Universita¨t, D-26111
Oldenburg, Germany.
(1) (a) Wentrup, C.; Netsch, K.-P. Angew. Chem., Int. Ed. Engl. 1984,
23, 802. (b) Wentrup, C.; Winter, H.-W.; Gross, G.; Netsch, K.-P.;
Kollenz, G.; Ott, W.; Biedermann, G. Angew. Chem., Int. Ed. Engl.
1984, 23, 800.
(2) (a) Fulloon, B.; Wentrup, C. J . Org. Chem. 1996, 61, 1363. (b)
Fullon, B.; El-Nabi, H. A. A.; Kollenz, G.; Wentrup, C. Tetrahedron
Lett. 1995, 36, 6547. (c) Kappe, C. O.; Kollenz, G.; Leung-Toung, R.;
Wentrup, C. J . Chem. Soc., Chem. Commun. 1992, 487. (d) Ben
Cheikh, A.; Chuche, J .; Manisse, N.; Pommelet, J . C.; Netsch, K.- P.;
Lorencak, P.; Wentrup, C. J . Org. Chem. 1991, 56, 970. (e) Bibas, H.;
Fulloon, B. E.; Moloney, D. W. J .; Wong, M. W.; Wentrup, C. Pure Appl.
Chem. 1996, 68, 891.
(3) (a) Bibas, H.; Wong, M. W.; Wentrup, C. Chem. Eur. J . 1997, 3,
237. (b) Preliminary communication: Bibas, H.; Wong, M. W.; Wen-
trup, C. J . Am. Chem. Soc. 1995, 117, 9582.
(4) (a) Wong, M. W.; Wentrup, C. J . Org. Chem. 1994, 59, 5279. (b)
Koch, R.; Wong, M. W.; Wentrup, C. J . Org. Chem. 1996, 61, 6809. (c)
Finnerty, J .; Andraos, J .; Yamamoto, Y.; Wong, M. W.; Wentrup, C. J .
Am. Chem. Soc. 1998, 120, 1701.
(6) (a) Himbert, G.; Fink, D. Z. Naturforsch. 1994, 49b, 542. (b)
Himbert, G.; Fink, D. J . Prakt. Chem. 1994, 336, 654.
(5) Cf. Birney, D. M. J . Org. Chem. 1996, 61, 243.
S0022-3263(97)02137-3 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/21/1998

2620 J . Org. Chem., Vol. 63, No. 8, 1998
Bibas et al.
Sch em e 2
F igu r e 1. Partial Ar matrix FTIR spectra (12 K) of the results
of FVT of (a) 4a at 1000 °C, (b) 4b at 850 °C, (c) 4c at 750 °C,
and (d) 10 at 850 °C. The bands at 2130 and 2122 cm-1 belong
to 6. Those at 2142 and 2138 cm-1 are due to s-trans-3a (see
ref 3a). The peaks at 2143 and 2133 cm-1 are assigned to s-cis-
and s-trans-acetylketene (see refs 3 and 10). The variable
peaks at 2148 and/or 2138 cm-1 are due (in part in Figure 1a)
to CO.39
unsubstituted vinylketene 6 was again observed with the
emergence of bands at 2130 and 2122 cm-1
.
mol-1) for the ene reaction and ∼50 kcal mol-1 (200 kJ
mol-1) for the retro-ene reaction.7d There has been a
single prior report of a retro-ene reaction taking place in
N,N,N,N-tetraethylallene-1,3-dicarboxamide on heating
to 180-200 °C during distillation (0.1 mbar), which
caused formation of an R-pyrone via a hypothetical ketene
intermediate.8
Further investigation was performed on the ethyl
analogues 4b and 7b. FVT/matrix isolation of ethyl 2,3-
butadienoate (4b) between 570 and 800 °C caused partial
conversion to the vinylketene 3b.3 At 850 °C, the
unsubstituted vinylketene 6 (2130, 2122 cm-1) as well
as acetylketene3,10 were observed (Figure 1b). On neat
deposition at 77 K, two bands at 2137 and 2118 cm-1
were observed due to acetylketene and unsubstituted
vinylketene (6), respectively. The value of 2118 cm-1 for
neat 6 is in agreement with a previous report.11 The
formation of acetylketene is due to elimination of ethene
from 3b as previously elucidated.3
FVT of 3-ethoxycyclobutenone12 (7b) at 400-700 °C
gave the observable ketene 4b.3 At 800 °C, weak bands
due to vinylketene 6 were observed at 2130 and 2122
cm-1 (Ar, 12 K). The weakness of the bands due to 6 in
the ethoxy case contrasts the strong signals in the
methoxy case (4a and 7a ) and is due to competing
reactions leading to the formation of acetylketene, pro-
pyne, and CO2 from the ethoxy derivatives 4b and 7b at
high FVT temperatures.
Resu lts a n d Discu ssion
1. F VT E xp er im en t s. In the system alkoxyvi-
nylketene (3a ,b), -acylallene (4a ,b), -cyclobutenone (7a ,b)
(Scheme 1) the interconversions 7 W 3 and 3 W 4 have
been established by matrix isolation FTIR spectroscopy.3
The interconversion of 3a and 4a takes place at ca. 650
°C under FVT conditions. FVT of allene 4a above 780
°C caused formation of unsubstituted vinylketene (6) as
evidenced by Ar matrix isolation of the ketene at 12 K.
The characteristic bands of 6 are at 2130 and 2122 cm-1
(Figure 1a) as further elaborated below. It appears that
the band at 2122 cm-1 is due to either a site or another
conformer of 6 as the ratio between these two bands
remained nearly constant, even when other precursors
were employed (Figure 1). This 2130/2122 cm-1 species
became dominant above 900 °C. A set of bands due to
formaldehyde9 (5a ) appeared concomitantly with those
of 6 (2995, 2816, 2864, 2798, 1742, 1738, 1499, 1174
cm-1).
To confirm the generality of the reaction and to give
further evidence for the formation of vinylketene 6, the
dimethylamino derivatives 7c and 4c were investigated.
Here, although the aminovinylketenes 3c are not directly
observable under FVT conditions, 7c and 4c were shown
to interconvert, starting from both sides, at ca. 470 °C.3
FVT/matrix isolation of allenic amides 4c gave rise to
bands due to vinylketene 6 starting already at 500 °C,
becoming very strong at 700 °C (Figure 1c).
These results can be understood in terms of a retro-
ene reaction of the allenic ester 4a (Scheme 1). In further
support of this, FVT of 3-methoxycyclobutenone (7a ) was
performed. At 400 °C, this causes ring opening to the
observable vinylketene 3a .3 From 800 °C onward, the
In addition, a new set of bands was observed in the
range 2800-3100 cm-1, increasing in intensity in concert
with those due to ketene 6. These characteristic bands
are due to N-methylmethanimine (5c) (Scheme 2). Proof
of the presence of 5c was secured by comparing the Ar
(7) (a) Hoffmann, H. M. R. Angew. Chem., Int. Ed. Engl. 1969, 8,
556-577. (b) Oppolzer, W.; Snieckus, V. Angew. Chem., Int. Ed. Engl.
1978, 17, 476-486. (c) Gajewski, J . J . Hydrocarbon Thermal Isomer-
izations; Academic: New York, 1981. (d) Loncharich, R. J .; Houk, K.
N. J . Am. Chem. Soc. 1987, 109, 6947. (e) Houk, K. N.; Li, Y.;
Evanseck, J . D. Angew Chem., Int. Ed. Engl. 1992, 31, 682-708.
(8) Ficini, J .; Pouliquen, J .; Paulme, J .-P. Tetrahedron Lett. 1971,
2483.
(10) s-cis-Acetylketene: 3095, 2148, 2143, 1681, 1378, 1345 and 1168
cm-1
cm-1
.
.
s-trans-Acetylketene: 3083, 2137, 2133, 1699, 1343, and 1221
(11) Wentrup, C.; Lorencak, P. J . Am. Chem. Soc. 1988, 110, 1880.
(12) Wasserman, H. H.; Piper, J . U.; Dehmlow, E. V. G. J . Org.
Chem. 1973, 38, 1451.
(9) (a) Harvey, K. B.; Ogilvie, J . F. Can. J . Chem. 1962, 40, 85. (b)
Khoshkhoo, H.; Nixon, E. R. Spectrochim. Acta 1971, 27A, 603.

Retro-Ene Reactions in Acylallene Derivatives
J . Org. Chem., Vol. 63, No. 8, 1998 2621
F igu r e 2. FTIR spectra (Ar, 12 K) of 6 and 5c. (Top) Spectrum of 6 and 5c (positive peaks) obtained after subtracting the
spectrum resulting from FVT of the allenic amide 4c at 750 °C from the spectrum of 4c itself (negative peaks); #: absorptions due
to 3-(N,N-dimethylamino)cyclobutenone (7c). The peaks in the 2100 cm-1 region are not shown with their correct intensities.
For details of the ketene region see Figure 1c. (Middle) Ar matrix spectrum (12 K) of 6 obtained after FVT of 10 at 850 °C; *:
bands due to 10. (Bottom) Ar matrix spectrum (12 K) of 5c obtained after FVT of 9 at 850 °C.
matrix IR spectrum with literature values,13 and by
generating an authentic sample of 5c by thermolysis of
1,3,5-trimethyl-1,3,5-triazine13,14 (9) above 800 °C with
matrix isolation of the product at 12 K. Very good
agreement is observed (Figure 2). Likewise, direct
comparison between the thermolysis spectrum from 4c
and an authentic sample of 6 (obtained by thermolysis
of (E)-2-butenoic acid15 (10)) also shows excellent agree-
ment (Figures 1c,d, 2).
The thermolysis of 4c was also monitored by mass
spectrometry (FVT-MS) between 475 and 850 °C (Figure
3). From 525 °C to 750 °C the peak at m/z 68 increased
in intensity due to the formation of 6, whereas the peak
at m/z 72 (M+ - C3H3), characteristic of 4c, decreased. A
new peak at m/z 42 due to imine 5c developed concomi-
tantly ([M - H]+; R-cleavage; not shown).
The (dimethylamino)cyclobutenone 7c was thermo-
lyzed analogously, and again the bands due to 6 and 5c
were observed already at an FVT temperature of 550 °C
(Ar, 12 K). An explanation for the early appearance of 6
in comparison with the use of alkoxy precursors (7a ,b)
is a lower barrier for the retro-ene reaction, as confirmed
by theoretical calculations (section 2).
F igu r e 3. Results of the FVT-MS of 4c. Relative intensities
of ions are shown. b: m/z 72 (characteristic peak of 4c (R-
cleavage; M+ - C3H3). 9: m/z 68 (vinylketene (6)).
Although vinylketene 6 was observed already at ca. 550
°C on Ar matrix isolation (12 K), it was not detectable
when either allene 4c or cyclobutenone 7c were thermo-
lyzed between 500 and 800 °C with product isolation at
77 K. This is explained by a very high reactivity of
ketene 6 toward the imine 5c, even at 77 K.
subsequent warm-up, the ketene band at 2118 cm-1 (the
value for neat ketene 6, also obtained in the alkoxy cases)
had completely disappeared at 75 K. A facile reaction of
the imine 5c with the ketene moiety of 6 (Staudinger
reaction16) is in agreement with previous observations
from this laboratory, where it was demonstrated that
nucleophiles such as pyridine react well below 77 K with
ketenes to give ketene-pyridine zwitterion inter-
To confirm this, 4c and 7c were separately thermo-
lyzed at 600 °C with neat deposition at 12 K. On
(13) Stolkin, I.; Ha, T.-K.; Gu¨nthard, Hs. H. Chem. Phys. 1977, 21,
327.
(14) (a) Graymore, J . J . Chem. Soc. 1924, 2283. (b) Kahovec, L. Z.
Phys. Chem. 1939, 43, 364.
(15) (a) Mohmand, S.; Hirabayashi, T.; Bock, H. Chem. Ber. 1981,
114, 2609. (b) Holmes, J . L.; Terlouw, J . K.; Vijfhuizen, P. C.; A′Campo,
C. Org. Mass Spectrom. 1979, 14, 204.
(16) (a) Staudinger, H. Liebigs Ann. Chem. 1907, 356, 51. (b)
Manhas, M. S.; Ghosh, M.; Bose, A. K. J . Org. Chem. 1990, 55, 575.
(c) Bose, A. K.; Manhas, M. S.; Ghosh, M.; Shah, M.; Raju, V. G.; Bari,
S. S.; Newaz, S. N.; Banik, B. K.; Chaudhary, A. G.; Barakat, K. J . J .
Org. Chem. 1991, 56, 6968. (d) Van der Steen, F. H.; Van Koten, G.
Tetrahedron 1991, 47, 7503.

2622 J . Org. Chem., Vol. 63, No. 8, 1998
Bibas et al.
Sch em e 3
Sch em e 4
mediates.17a,b Moreover, vinylketene 6 itself has been
reported to be stable at 77 K.18 We confirmed this by
generation of 6 by FVT of 10 with neat isolation at 12 K.
In the absence of the imine, ketene 6 remained stable
above 80 K.19
To obtain further evidence on the reaction between 5c
and 6, these were generated by thermolysis (850 °C) of 9
and 10, respectively, deposited simultaneously on the cold
window (neat, 12 K), observed by their characteristic IR
spectra, and subsequently warmed. Again, at 75 K both
ketene and imine bands vanished, but further studies
would be required for an identification of the products
formed.20 Reactions between ketene 6 and imine 5c could
be expected to lead to zwitterionic intermediates which
subsequently cyclize to â or δ-lactam, as has been found
in many related reactions.16 However, no direct IR
spectroscopic evidence for the formation of such a zwit-
terion (17, see below) or δ-lactam was obtainable due to
the strong absorptions of 4c, 5c and 7c in the 1650-1690
cm-1 region. Also no evidence for the formation of a
â-lactam was found (expected absorptions16b,c ca. 1760
cm-1).
It was a priori possible that the vinylketenes 3 could
also undergo a retro-ene reaction, as illustrated in
Scheme 3. This would generate the same aldehyde,
ketone, and imine 5, together with 2,3-butadienal (8).
However, no new bands were observed in the allenic
region or the 1700-1600 cm-1 region that could be
ascribed to 2,3-butadienal in any of these experiments
on FVT of acylallenes 4 or cyclobutenones 7. It is also
inconceivable that an initially formed formylallene 8
would easily isomerize to 6. Allene 8 has a higher
calculated energy than the isomeric vinylketene (6), but
the 1,3-H shift required in the process 8 f 6 has a very
high computed activation energy (272.5 kJ mol-1 at the
G2(MP2, SVP) level of theory).3a Thus, there is no
evidence for the reactions of Scheme 3 occurring.
When 2-cyclohexylidene-N,N-diethylethenecarboxam-
ide21 (11) was thermolyzed between 400 and 760 °C with
product isolation in Ar at 12 K, a band ascribed to the
vinylketene 12 was observed at 2120 cm-1 above 450 °C
(Scheme 4). The formation of 12 is ascribed to the facile
1,3-shift of the diethylamino group in the allenic amide.
An electrocyclization of this ketene to give a spiro
compound 14 was not observed, as testified by the
absence of bands in the 1700-1800 cm-1 region which
could correspond to a conjugated carbonyl group in this
strained compound (the carbonyl band for 3-(dimethy-
lamino)cyclobutenone (7c) is at 1772 cm-1 in Ar matrix3).
The absence of 14 can be ascribed to increased ring
strain, making this compound slightly less stable than
ketene 12. An analogous case is 3-chlorovinylketene,
which on warmup fails to cyclize to 3-chlorocyclo-
butenone.3a
(17) (a) Qiao, G. G.; Andraos, J .; Wentrup, C. J . Am. Chem. Soc.
1996, 118, 5634. (b) Visser, P.; Zuhse, R.; Wong, M. W.; Wentrup, C.
J . Am. Chem. Soc. 1996, 118, 12598. (c) Pacansky, J .; Chang, J . S.;
Brown, D. W.; Schwarz, W. J . Org. Chem. 1982, 47, 2233. (d)
D′Annibale, A.; Pesce, A.; Resta, S.; Trogolo, C. Tetrahedron Lett. 1996,
37, 7429. (e) Cf. Gompper, R.; Wolf, U. Liebigs Ann. Chem. 1979, 1388.
(f) Moore, H. W.; Hernandez, L.; Chambers, R. J . Am. Chem. Soc. 1978,
100, 2245. (g) Duran, F.; Ghosez, L. Tetrahedron Lett. 1970, 245. (h)
Huisgen, R.; Davis, B. A.; Morikawa, M. Angew. Chem., Int. Ed. Engl.
1968, 7, 826. (i) Kagan, H. B.; Luche, J . L. Tetrahedron Lett. 1968,
3093.
Above 700 °C, the band at 2120 cm-1 decreased in
intensity and a new set of bands in the 2800 cm-1 region
and at 1679 and 1630 cm-1 increased. These are char-
acteristic of an aldehyde, e.g. 16 (Scheme 4).
Chemical evidence for the structure 16 was obtained
by generation at 700 °C with product isolation on a
coldfinger at -196 °C, both with and without trapping
agent (methanol). After warming to room temperature,
(18) Trahanovsky, W. S.; Surber, B. W.; Wilkes, M. C.; Preckel, M.
M. J . Am. Chem. Soc. 1982, 104, 6779.
1
16 was isolated in 70% yield, and the H NMR spectrum
(19) (a) When vinylketene (6) is generated by FVT of butenoic acid
10, it is isolated in the presence of water ice at 77 K. We have shown
elsewhere that ketenes can be stable at -40 °C (and above) in the
presence of water-ice,19b whereas the same ketenes react at -100 to
-90 °C with MeOH,19b and other ketenes react at temperatures as
low as 15-75 K (-258 to -198 °C) with amines (pyridine).17a,b Birney
et al. have also reported a high selectivity for reaction of acetylketene
with an amine vis-a`-vis alcohols.19c (b) Gross, G.; Wentrup, C. J . Chem.
Soc., Chem. Commun. 1982, 360. (c) Birney, D. M.; Xu, X.; Ham, S.;
Huang, X. J . Org. Chem. 1997, 62, 7114.
(20) In the low-pressure gas-phase reaction, the vinylketene 6 and
the imine 5c will be prevented from reacting with each other and
cyclizing to lactams. â-Lactams are the products expected in solution
phase reactions.16b,c
was in excellent agreement with literature values.22 At
FVT temperatures below 700 °C, only starting material
11 was recovered. The formation of 16 is understood in
terms of formation of an unobserved ketene 13 via the
retro-ene pathway. A subsequent 1,5-H shift causes
(21) (a) Parker, K. A.; Petraitis, J . J .; Kosley, R. W.; Buchwald, J r.,
S. L. J . Org. Chem. 1982, 47, 389. (b) Parker, K. A.; Petraitis J . J .
Tetrahedron Lett. 1977, 52, 4561.
(22) Zimmermann, B.; Lerche, H.; Severin, Th. Chem. Ber. 1986,
119, 2848.

Retro-Ene Reactions in Acylallene Derivatives
J . Org. Chem., Vol. 63, No. 8, 1998 2623
Sch em e 5
isomerization to the more stable aldehyde 16 (Scheme
4; cf. Section 2). Even though this known type of
sigmatropic shift23 is reversible, in this particular case
the equilibrium is completely displaced toward the alde-
hyde 16 due to its higher thermodynamic stability.
Indeed, semiempirical calculations predict the aldehyde
16 to be more stable than the ketene sZ-13 by 30 (PM3),
or 26 (AM1) kJ mol-1
.
In attempted trapping reactions with MeOH at 77 K,
mainly starting material 11 was recovered at FVT
temperatures below 700 °C. The ketene 12 was not
trappable. One explanation for this could be the high
reactivity (below 77 K) of the amine moieties in 11 or 12
toward the ketene, resulting in the formation of ketene/
amine zwitterions, possibly followed by polymerization.19
F igu r e 4. Calculated structures of TS1c (upper) and TS2
(lower) at the MP2/6-31G* level.
Similarly, when the allene carboxamide 11 was ther-
molyzed between 500 and 800 °C with product isolation
at 77 K, no trace of ketene 12 was observed, probably
due to the formation of ketene/nucleophile zwitterions
and subsequent reactions well below 77 K. The starting
material 11 remained the only identifiable species, until
580 °C, when 16 started to appear.
In this reaction too, there was no evidence for an
alternative retro-ene reaction of ketene 12, which in
analogy with Scheme 3 would have yielded an allenic
aldehyde, 2-cyclohexylideneacrolein.
2. Th eor y. 2.1. Retr o-En e Rea ction . To gain more
insight into the abovementioned retro-ene reactions,
G2(MP2,SVP) calculations were performed for allenes 4a
and 4c, reacting to vinylketene 6 and formaldehyde or
methanimine 5 via transition states TS1a and TS1c,
respectively (Scheme 5). In both cases, TS1 features the
transfer of a hydrogen atom from a methyl group to the
central carbon atom of the allene moiety (see Figure 4)
together with some lengthening of the CO-X bond
compared to the starting material (1.536 Å in TS1a
(1.359 Å in 4a ); 1.425 Å in TS1c (1.370 Å in 4c). TS1c
is shown in Figure 4, and that of TS1a in Figure S1 in
the Supporting Information, where the full computed
structural data is also presented. The structure of TS1
is further discussed in section 2.3. The calculated
activation energies, presented in Table 1, give at the
highest level, G2(MP2,SVP) theory, a barrier for TS1a
of 181 kJ mol-1, and for TS1c of 167 kJ mol-1 above the
allene starting material, in agreement with the experi-
mental observation that the retro-ene reaction takes
place significantly more easily for 4c than for 4a .
F igu r e 5. Hypothetical zwitterion 17, and calculated ketene-
imine van der Waals complex 17′ (MP2/6-31G*).
Ta ble 1. Ca lcu la ted Rela tive En er gies (k J m ol-1) for th e
Retr o-En e Rea ction s of 4A a n d 4c (Sch em e 5)
R ) methoxy
R ) dimethylamino
level of calcn
HF/6-31G*
4a TS1a 6 + 5a 4c TS1c 6 + 5c
0
0
0
0
0
300.8
205.5
186.8
211.2
181.4
40.1
48.1
47.9
51.6
35.6
0
0
0
0
0
285.7
177.5
157.2 100.8
197.2
167.1
86.1
99.0
MP2(fc)/6-31G*
MP2/6-311+G(3df,2p)
QCISD(T)/6-31G*
G2 (MP 2,SVP )
93.3
80.6
involved in the addition of an imine to a ketene. It is
possible, therefore, that this intermediate could also exist
on the retro-ene energy surface. At the HF/6-31G* level
(24) Calculations on the ketene-imine Staudinger reaction taking
dielectric solvent effects into account indicate that zwitterionic inter-
mediates may be favored in solution, whereas they may be concerted
in the gas phase: (a) Sordo, J . A.; Gonza`lez, J .; Sordo, T. L. J . Am.
Chem. Soc. 1992, 114, 6249. (b) Assfeld, X.; Sordo, J . A.; Gonza`lez, J .;
Ruiz-Lopez, M. F.; Sordo, T. L. J . Mol. Struct. (THEOCHEM) 1993,
287, 193. (c) Assfeld, X.; Ruiz-Lopez, M. F.; Gonza`lez, J .; Lopez, R.;
Sordo, J . A.; Sordo, T. L. J . Comput. Chem. 1994, 17, 479. (d) Lopez,
R.; Ruiz-Lopez, M. F.; Rinaldi, D.; Sordo, J . A.; Sordo, T. L. J . Phys.
Chem. 1996, 100, 10600. (e) Lecea, B.; Arrastia, I.; Arrieta, A.; Roa,
G.; Lopez, X.; Arriortua, M. I.; Ugalde, J . M.; Cossio, F. P. J . Org. Chem.
1996, 61, 3070.
On the basis of previous work, both experimental16,17
and theoretical,24,25 a zwitterion 17 (Figure 5) may be
(23) (a) Schiess, P.; Radimerski, P. Helv. Chim. Acta 1974, 57, 2583.
(b) Schiess, P.; Radimerski, P. Angew. Chem., Int. Ed. Engl. 1972, 11,
288. (c) Schiess, P.; Wisson, M. Tetrahedron Lett. 1971, 26, 2389. (d)
Pirkle, W. H.; Seto, H.; Turner, W. V. J . Am. Chem. Soc. 1970, 92,
6984.
(25) Concerted Staudinger reactions have also been predicted
computationally: (a) Yamabe, S.; Minato, T.; Osamuro, Y. J . Chem.
Soc., Chem. Commun. 1993, 450. (b) Fang, D. C.; Fu, X. Y. Int. J .
Quantum Chem. 1992, 43, 669.

2624 J . Org. Chem., Vol. 63, No. 8, 1998
Bibas et al.
Ta ble 2. Ca lcu la ted Rela tive En er gies (k J m ol-1) for th e
Sch em e 6
1,5-H Sh ift Rea ction (18 f 19) (Sch em e 6)
level of calcn
(s-Z,Z)-18 TS2 (s-E,Z,s-Z)-19 (s-E,E,s-E)-19
HF/6-31G*
MP2(fc)/6-31G
MP2/6-311+G(3df,2p)
QCISD(T)/6-31G
G2(MP 2,SVP )
0
0
0
0
0
215.9
130.7
119.2
143.9
125.0
-2.2
12.8
17.5
1.8
-26.7
-8.7
-5.0
-19.4
-13.5
10.4
of theory, such an intermediate is indeed found, in a very
shallow minimum, 140.4 kJ mol-1 above 4c. At this level
of theory, TS1c is 286 kJ mol-1 above 4c, but there is
only a 0.4 kJ mol-1 barrier for the intermediate to
dissociate to 6 + 5c (transition structure 140.8 kJ mol-1
above 4c). The intermediate becomes already less favor-
able than its dissociation transition state TS1c when
entropy is taken into account, calculating ∆G. There is
no corresponding intermediate for the retro-ene reaction
of 4a at the HF/6-31G* level. The abovementioned
intermediate 17 in the reaction of 4c disappears at
correlated levels (MP2 and G2), where the retro-ene
reaction is concerted in both methoxy (4a ) and dimethy-
lamino (4c) cases. When lengthening the N-CO distance
in TS1c, no stationary point can be found corresponding
to an intermediate zwitterion, but only a loose van der
Waals complex 17′ with a very long N‚‚‚CO contact (2.98
Å). At the Becke3LYP level it is even longer, about 3.2
Å. The MP2/6-31G* structure is shown in Figure 5. This
species has still an almost linear CdCdO moiety (176.5°)
and is best described as a ketene complex rather than a
zwitterion. Very recently, calculations on a cyanoketene-
formaldimine van der Waals complex have been pub-
lished.26 Complex 17′ exists in a shallow minimum, 9
kJ mol-1 below the separated products, 5 + 6c. For
chemical purposes, the reactions can be regarded as
concerted. A more detailed investigation of the potential
energy surface, including â- and δ-lactam formation, at
ab initio and density functional levels of theory, and
including a dielectric field, is in progress and will be
published elsewhere.
2.2. 1,5-H Sh ift. In the case of allene 11, the
vinylketene 13 resulting from the retro-ene rearrange-
ment could not be observed (Scheme 4). Instead, the
aldehyde 16, formed from 13 by a 1,5-hydrogen shift, was
isolated. We therefore investigated a model system, the
rearrangement of methylvinylketene 18 to penta-2,4-
diene-1-al 19 (Scheme 6).
Starting from the required s-Z,Z conformer of 18 (6 kJ
mol-1 above the corresponding s-E,Z conformer), the
transition state TS2 for the 1,5-H shift lies 125 kJ mol-1
above s-Z,Z-18. The structure of TS2 is similar to, but
flatter than that of TS1, with the hydrogen atom again
being transferred nearly in-plane toward the central
carbon p orbital of the ketene (Figure 4). For further
discussion see section 2.3. The migration initially leads
to the aldehyde (s-E,Z,s-E)-19 (+ 9.4 kJ mol-1) which then
isomerizes to the more stable conformation (s-E,E,s-Z)-
19: -13.5 kJ mol-1; Table 2). These data clearly il-
lustrate why the vinylketene 13 cannot be observed. Once
formed at temperatures which overcome the barrier for
the retro-ene reaction (167-184 kJ mol-1), the ketene can
easily rearrange to the aldehyde 16 via the more facile
(125 kJ mol-1) 1,5-hydrogen shift.
TS1a and TS1c, are shown in Figure 4 and in the
Supporting Information. In TS1c the hydrogen atom
moves ca. 17° out of the plane of the four carbon atoms.
The terminal CH2 group of the allene moiety has twisted
to a large extent, so that an essentially planar vinyl group
can form when the H atom is fully transferred. This is
even more pronounced in TS1a , which is more advanced
(i.e. “later”), with a forming bond length C3-H ) 1.173
Å. The rough planarity of these transition structures
could suggest a pseudopericyclic nature of the reactions,27
but further computational studies of model compounds
will be required in order to determine the exact orbital
interactions. 1,5- and 1,7-H shifts in conjugated allenes
have been investigated by Okamura et al.,28 and the
“allene effect” (a lowering of the activation barrier
compared to the nonallenic 1,3 pentadiene and 1,3,5-
heptatriene systems) has been explained by J ensen in
terms of extra conjugation in the TS, which can be
formally regarded as biradicaloid.29
The nearly-planar TS1 with the CH2 end groups
twisted into the plane are in accord with that interpreta-
tion, but again model compound calculations are required
in order to determine the magnitude of the effect.
TS2 is much flatter than corresponding transition
structures for 1,5-H shifts in 1,3-pentadiene or cyclopen-
tadiene, where the migrating hydrogen atom moves as
much as 79° out of plane.7e,30 The near-planarity of TS2
(Figure 4) suggests that the H atom is transferring
toward the in-plane ketene p orbital at C3, which is
orthogonal to the remaining π backbone. This process
therefore has the characteristics of a pseudopericyclic
reaction.27 Closely related, nearly planar transition
structures have been computed for the 1,5-H shift in
4-formylvinylketene (HCO-CHdCH-CHdCdO; calc. Ea
∼ 99 kJ mol-1),5 and for the 1,5-Cl shift in 4-(chlorocar-
bonyl)vinylketene (ClCO-CHdCH-CHdCdO; calc. Ea
∼ 32 kJ mol-1).4b
The calculated barrier separating 18 and TS2 (125 kJ
mol-1) is not particularly low. This is in line with results
for related compounds containing a CH2 group rather
than
a CdO group, e.g. chloromethyl isocyanate
(ClCH2NCO) compared with chlorocarbonyl isocyanate
(ClCONCO); the calculated barrier for the 1,3-Cl shift is
191 kJ mol-1 in the former,4b and only 85 kJ mol-1 in
(27) (a) Pseudopericyclic reactions were defined by Lemal as those
having an orbital disconnection.27b This disconnection or orthogonality27c
renders the Woodward-Hoffmann rules inapplicable, and all reactions
are allowed, regardless of the number of electrons. The reactions have
been investigated extensively by Birney in recent years.27d,e (b) Ross,
J . A.; Seiders, R. P.; Lemal, D. M. J . Am. Chem. Soc. 1976, 98, 4325.
(c) Wentrup, C.; Netsch, K.-P. Angew. Chem., Int. Ed. Engl. 1984, 23,
802. (d) Birney, D. M.; Ham, S.; Unruh, G. R. J . Am Chem. Soc. 1997,
119, 4509 and references therein. (e) See also Luo, L.; Bartberger, M.
D.; Dolbier, W. R. J . Am. Chem. Soc. 1997, 119, 12366 and references
therein.
(28) (a) Shen, G.-Y.; Tapia, R.; Okamura, W. H. J . Am. Chem. Soc.
1987, 109, 7499. (b) Okamura, W. H.; Peter, R.; Reischl, W. J . Am.
Chem. Soc. 1985, 107, 1034. (c) Elnagar, H. Y.; Okamura, W. H. J .
Org. Chem. 1988, 53, 3060.
2.3. Th e Na tu r e of th e Tr a n sition Sta tes. The
transition structures calculated for the retro-ene reaction,
(29) J ensen, F. J . Am. Chem. Soc. 1995, 117, 7487.
(30) J ensen, F.; Houk, K. N. J . Am. Chem. Soc. 1987, 109, 3139.
(26) Suarez, D.; Sordo, T. L. J . Am. Chem. Soc. 1997, 119, 10291.

Retro-Ene Reactions in Acylallene Derivatives
J . Org. Chem., Vol. 63, No. 8, 1998 2625
of G2(MP2)34 theory but computationally more efficient. The
frozen-core approximation was employed for all correlated
calculations. Zero-point vibrational energy calculations were
performed at the HF/6-31G* level.
the latter.4c While all these reactions can be termed
pseudopericyclic, the tremendous effects of ketenes (com-
pared with allenes) and halogens as well as other
electron-donating lone-pair groups on lowering the acti-
vation barriers are due to better HOMO-LUMO interac-
tion terms,4b which can also be expressed in terms of the
nucleophilicity and electrophilicity of the reacting cen-
ters.5
Other things being equal, we then expect lower activa-
tion barriers for ketenes than for allenes and much lower
barriers for the transfer of lone-pair groups than hydro-
gen atoms.
Gen er a l E xp er im en t a l Met h od s. Preparative flash
vacuum thermolysis (FVT) was carried out in electrically
heated quartz tubes, 40 cm long, of 2 cm diameter. Samples
were vaporized or sublimed into the pyrolysis tube using a
Bu¨chi sublimation oven. The system was evacuated to ca. 10-5
mbar and continuously pumped during FVT using a Leybold-
Heraeus turbomolecular pump PT150. The thermolysates
were cocondensed with or without MeOH vapor on the 77 K
coldfinger, whereby methanol was introduced between the exit
of the pyrolysis oven and the coldfinger. Further details of
the FVT apparatus have been published.35
Con clu sion
Matrix isolation was carried out using 10 cm long, 0.8 cm
internal diameter quartz tubes in an oven directly attached
to the vacuum shroud of a Leybold-Heraeus closed cycle liquid
He cryostat.3b,35 Ar was used as matrix host, which was passed
over the sample while it was subliming and cocondensed at
ca. 12 K on a BaF2 window for IR spectroscopy. Neat isolation
at 77 K was carried out in a similar apparatus using a liquid
N2 cryostat.35
The allenic esters 4a and 4b undergo a retro-ene
reaction on FVT above 800 °C to give unsubstituted
vinylketene (6) together with formaldehyde (5a ) or ac-
etaldehyde (5b).
The thermolysis of N,N-dimethylallenecarboxamide
(4c) likewise gives vinylketene 6 and N-methylmethyl-
eneimine (5c), starting at lower FVT temperatures (>500
°C).
The FVT-MS combination has been described.36
Ma ter ia ls. 3-Methoxycyclobutenone,37 3-ethoxycyclobuten-
one,12 3-(N,N-dimethylamino)cyclobutenone,3a 4a ,b,38 and 4c3a
were synthesized according to the literature. IR matrix
isolation data for these compounds have been described.3a 10
was obtained commercially (Aldrich).
These observations are in very good agreement with
ab initio calculations which predict activation energies
for the retro-ene reaction of 181 (R ) OMe) or 167 (R )
NMe2) kJ mol-1
.
(2E)-Bu ten oic a cid (10): IR (Ar matrix, 12 K): 3585 (w),
3579 (w), 3570 (sh), 3567 (m), 3562 (w), 3551 (w), 2992 (w),
2963 (w), 2953 (w), 2924 (w), 1755 (s), 1738 (m), 1729 (m), 1710
(m), 1707 (w), 1664 (m), 1449 (m), 1436 (w), 1379 (w), 1366
(w), 1360 (w), 1356 (w), 1346 (w), 1320 (b), 1298 (b), 1294 (w),
1273 (b), 1229 (w), 1201 (w), 1181 (b), 1164 (m), 1160 (w), 1144
(m), 1102 (w), 1093 (m), 1090 (m), 979 (w), 973 (w), 963 (w),
852 (w), 869 (w), 866 (w), 849 (w), 846 (w), 842 (w), 840 (w)
The same acylallenes 4 are formed on FVT of cy-
clobutenones 7 via vinylketenes 3.3 At higher FVT
temperatures, the retro-ene reactions again lead to the
formation of 6.
In the case of the 2-cyclohexylidene-N,N-diethylethen-
ecarboxamide (11), Ar matrix isolation indicates the
formation of the vinylketene 12, formed by a 1,3-NEt2
shift, but the vinylketene 13 expected from the retro-ene
reaction is not directly observable. Instead, formation
of a conjugated aldehyde 16 occurs by a 1,5-H shift in
13. Ab initio calculations carried out on 4-methylvi-
nylketene 18 confirm that the 1,5-H shift has a much
lower activation barrier (ca. 125 kJ mol-1) than the retro-
ene reaction.
cm-1
.
1,3,5-Tr im eth yl-1,3,5-tr ia zin e (9). 9 was synthesized
1
according to Graymore.14a Yield 30-50%; bp 160-172 °C. H
NMR (60 MHz, CDCl3): δ 2.30 (s, 9 H), 3.21 (s, 6 H); IR (neat,
-196 °C): 2966 (s), 2942 (s), 2892 (m), 2843 (w), 2809 (sh),
2787 (s), 2773 (sh), 2726 (m), 2664 (m), 2649 (m), 2630 (w),
2596 (m), 2572 (w), 2503 (w), 2453 (w), 1471 (s), 1444 (s), 1428
(s), 1386 (m), 1373 (m), 1346 (w), 1305 (w), 1273 (sh), 1262
(s), 1234 (m), 1156 (m), 1116 (s), 1050 (w), 1025 (m), 1003 (m),
981 (w), 915 (s), 898 (w), 860 (m), 836 (w), 627 (w) cm-1; (Ar,
12 K): 2981 (m), 2952 (m), 2901 (w), 2847 (w), 2815 (w), 2793
(m), 2730 (w), 2670 (w), 2636 (w), 2605 (w), 2576 (w), 1475
(m), 1446 (m), 1430 (w), 1425 (w), 1387 (m), 1375 (m), 1277
(m), 1266 (m), 1237 (m), 1163 (m), 1126 (m), 1119 (s), 1052
Exp er im en ta l a n d Com p u ta tion a l Section
Com p u ta tion a l Meth od s. Standard ab initio molecular
orbital calculations31 were carried out using the GAUSSIAN
94 system of programs.32 The structures and energies of all
model compounds were examined at the G2(MP2,SVP) level
of theory.33 This corresponds effectively to QCISD(T)/6-
311+G(3df,2p)//MP2/6-31G* energies together with zero-point
vibrational and isogyric corrections. In the G2(MP2,SVP)
theory, the basis-set extension energy contribution is calcu-
lated at the MP2 level, and the QCISD(T) energy is evaluated
using the 6-31G* basis set. It has been shown that the
accuracy of the G2(MP2, SVP) method is comparable to that
(w), 1005 (m), 917 (m), 915 (m), 862 (w), 629 (w), 496 cm-1
.
Ma tr ix Isola tion of N-Meth ylm eth a n im in e (5c). The
preparation of a matrix of 5c was slightly different from the
one described by Stolkin and Gu¨nthard.13 Instead of using a
low FVT temperature and an alumina-silica catalyst (90:10),
9 was detrimerized by FVT at 850 °C. The data for 5c are as
follows: IR (neat, 12 K): 3013 (m), 2943 (s), 2896 (s), 2850
(s), 2776 (m), 1653 (s), 1473 (s), 1437 (m), 1405 (w), 1384 (w),
1262 (w), 1232 (m), 1160 (w), 1112 (m), 1046 (m), 1002 (w),
949 (m), 917 (w), 861 (w), 630 (vw), 498 (w) cm-1; IR (Ar, 12
K): 3014 (m), 2964 (m), 2955 (m), 2940 (m), 2902 (m), 2886
(31) Hehre, W. J .; Radom, L.; Schleyer, P. v. R.; Pople, J . A. Ab Initio
Molecular Orbital Theory; Wiley: New York, 1986.
(32) Frisch, M. J .; Trucks, G. W.; Schlegel, H. B.; Gill, P. M. W.;
J ohnson, B. G.; Robb, M. A.; Cheeseman, J . R.; Keith, T. A.; Petersson,
G. A.; Montgomery, J . A.; Raghavachari, K.; Al-Laham, M. A.;
Zakrzewski, V. G.; Ortiz, J . V.; Foresman, J . B.; Cioslowski, J .;
Stefanov, B. B.; Nanayakkara, A.; Challacombe, M.; Peng, C. Y.; Ayala,
P. Y.; Chen, W.; Wong, M. W.; Andres, J . L.; Replogle, E. S.; Gomperts,
R.; Martin, R. L.; Fox, D. J .; Binkley, J . S.; DeFrees, D. J .; Baker, J .;
Stewart, J . J . P.; Head-Gordon, M.; Gonzalez, C.; Pople, J . A. Gaussian
94; Gaussian Inc., Pittsburgh, PA, 1995.
(34) Curtiss, L. A.; Raghavachari, K.; Pople, J . A. J . Chem. Phys.
1993, 98, 1293.
(35) Wentrup, C.; Blanch, R.; Briehl, H.; Gross, G. J . Am. Chem.
Soc. 1988, 110, 1874.
(36) Wentrup, C.; Kambouris, P.; Evans, R. A.; Owen, D.; Macfar-
lane, G.; Chuche, J .; Pommelet, J . C.; Ben Cheikh, A.; Plisnier, M.;
Flammang, R. J . Am. Chem. Soc. 1991, 113, 3130.
(37) Stone, G. B.; Liebeskind, L. S. J . Org. Chem. 1990, 55, 4614.
(38) Hamlet, Z.; Barker, W. D. Synthesis 1970, 543.
(39) Pimentel, G. C.; Charles, S. W. Pure Appl. Chem. 1963, 7, 111.
(33) (a) Smith, B. J .; Radom, L. J . Phys. Chem. 1995, 99, 6468. (b)
Curtiss, L. A.; Redfern, P. C.; Smith, B. J .; Radom, L. Chem. Phys.
1996, 104, 5148.

2626 J . Org. Chem., Vol. 63, No. 8, 1998
Bibas et al.
(w), 2878 (w), 2856 (m), 2850 (m), 2839 (w), 2776 (w), 1657
(m), 1470 (vs), 1441 (m), 1222 (m), 1027 (m), 950 (m), 663 (w),
479 (w) cm-1. The Ar matrix IR data are in good agreement
with the literature.13
Purity by NMR was higher than 90%. 13C NMR (200 MHz,
CDCl3): δ 194.4, 156.3, 141.3, 135.5, 125.8, 26.7, 24.2, 21.9,
21.8.
Ack n ow led gm en t. We thank the Australian Re-
search Council for financial support (to C.W.), The
University of Queensland for generous allocations of
computer time, and Mr. G. Macfarlane for the FVP-MS
experiments. R.K. thanks the Deutsche Forschungsge-
meinschaft for a Postdoctoral Fellowship.
2-Cycloh exyliden e-N,N-dieth yleth en ecar boxam ide (11).
The synthesis of 11 was described by Parker et Petraitis.21
The 1H NMR data were in good agreement with the literature.21a
Purity according to NMR was higher than 98%. 13C NMR (400
MHz, CDCl3): δ 204.3, 165.2, 105.4, 85.9, 42.6, 40.8, 30.3, 26.8,
25.8, 14.6, 13.0; IR (Ar, 12 K): 2981 (m), 2971 (sh), 2941 (m),
2896 (w), 2862 (m), 2848 (w), 1973 (m), 1642 (s), 1481 (m),
1473 (m), 1465 (m), 1460 (m), 1451 (m), 1441 (m), 1431 (m),
1402 (w), 1380 (m), 1364 (w), 1349.5 (w), 1346 (w), 1323 (w),
1315 (w), 1280 (m), 1251 (w), 1237 (m), 1226 (m), 1182 (w),
1155 (w), 1135 (m), 1099 (w), 1090 (w), 1085 (w), 1073 (w),
1027 (w), 1022 (w), 977 (w), 947 (w), 933 (w), 922 (w), 911 (w),
898 (w), 853 (w), 828 (w), 821 (w), 811 (w), 788 (w), 781 (w)
Su p p or tin g In for m a tion Ava ila ble: Structure TS1a
calculated at the MP2/6-31G* level of theory (Figure S1),
calculated structural parameters for 4a , 4c, TS1a , TS1c, 17′,
and 6 (MP2/6-31G*) (Table S1), Cartesian coordinates for the
calculated structures (Table S2), and IR and GC-MS data for
compound 16 (5 pages). This material is contained in libraries
on microfiche, immediately follows this article in the microfilm
version of the journal, and can be ordered from the ACS; see
any current masthead page for ordering information.
cm-1
.
F VT of 11: Isola tion of (E)-3-(1-Cyloh exen -1-yl)p r op e-
n a l (16). FVT of 11 was carried out at 700 °C and gave, after
column chromatography (SiO2/CHCl3), 16 (yield: 70%). 1H
NMR data are in excellent agreement with the literature.22
J O972137M